940 files changed, 40172 insertions, 36392 deletions

diff --git a/Documentation/00-INDEX b/Documentation/00-INDEX
index 73060819ed99..438277800103 100644
--- a/Documentation/00-INDEX
+++ b/Documentation/00-INDEX
@@ -159,8 +159,6 @@ hayes-esp.txt
 	- info on using the Hayes ESP serial driver.
 highuid.txt
 	- notes on the change from 16 bit to 32 bit user/group IDs.
-hpet.txt
-	- High Precision Event Timer Driver for Linux.
 timers/
 	- info on the timer related topics
 hw_random.txt
diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 3d2d0c29f027..cc8093c15cf5 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -287,14 +287,6 @@ Who:	Glauber Costa <gcosta@redhat.com>
 
 ---------------------------
 
-What:	old style serial driver for ColdFire (CONFIG_SERIAL_COLDFIRE)
-When:	2.6.28
-Why:	This driver still uses the old interface and has been replaced
-	by CONFIG_SERIAL_MCF.
-Who:	Sebastian Siewior <sebastian@breakpoint.cc>
-
----------------------------
-
 What:	/sys/o2cb symlink
 When:	January 2010
 Why:	/sys/fs/o2cb is the proper location for this information - /sys/o2cb
diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt
index 0d5394920a31..eb154ef36c2a 100644
--- a/Documentation/filesystems/ext4.txt
+++ b/Documentation/filesystems/ext4.txt
@@ -32,9 +32,9 @@ Mailing list: linux-ext4@vger.kernel.org
     you will need to merge your changes with the version from e2fsprogs
     1.41.x.
 
-  - Create a new filesystem using the ext4dev filesystem type:
+  - Create a new filesystem using the ext4 filesystem type:
 
-    	# mke2fs -t ext4dev /dev/hda1
+    	# mke2fs -t ext4 /dev/hda1
 
     Or configure an existing ext3 filesystem to support extents and set
     the test_fs flag to indicate that it's ok for an in-development
@@ -47,13 +47,13 @@ Mailing list: linux-ext4@vger.kernel.org
 
         # tune2fs -I 256 /dev/hda1
 
-    (Note: we currently do not have tools to convert an ext4dev
+    (Note: we currently do not have tools to convert an ext4
     filesystem back to ext3; so please do not do try this on production
     filesystems.)
 
   - Mounting:
 
-	# mount -t ext4dev /dev/hda1 /wherever
+	# mount -t ext4 /dev/hda1 /wherever
 
   - When comparing performance with other filesystems, remember that
     ext3/4 by default offers higher data integrity guarantees than most.
@@ -177,6 +177,11 @@ barrier=<0|1(*)>	This enables/disables the use of write barriers in
 			your disks are battery-backed in one way or another,
 			disabling barriers may safely improve performance.
 
+inode_readahead=n	This tuning parameter controls the maximum
+			number of inode table blocks that ext4's inode
+			table readahead algorithm will pre-read into
+			the buffer cache.  The default value is 32 blocks.
+
 orlov		(*)	This enables the new Orlov block allocator. It is
 			enabled by default.
 
@@ -218,6 +223,11 @@ errors=remount-ro(*)	Remount the filesystem read-only on an error.
 errors=continue		Keep going on a filesystem error.
 errors=panic		Panic and halt the machine if an error occurs.
 
+data_err=ignore(*)	Just print an error message if an error occurs
+			in a file data buffer in ordered mode.
+data_err=abort		Abort the journal if an error occurs in a file
+			data buffer in ordered mode.
+
 grpid			Give objects the same group ID as their creator.
 bsdgroups
 
@@ -252,6 +262,7 @@ stripe=n		Number of filesystem blocks that mballoc will try
 delalloc	(*)	Deferring block allocation until write-out time.
 nodelalloc		Disable delayed allocation. Blocks are allocation
 			when data is copied from user to page cache.
+
 Data Mode
 =========
 There are 3 different data modes:
diff --git a/Documentation/filesystems/fiemap.txt b/Documentation/filesystems/fiemap.txt
new file mode 100644
index 000000000000..1e3defcfe50b
--- /dev/null
+++ b/Documentation/filesystems/fiemap.txt
@@ -0,0 +1,228 @@
+============
+Fiemap Ioctl
+============
+
+The fiemap ioctl is an efficient method for userspace to get file
+extent mappings. Instead of block-by-block mapping (such as bmap), fiemap
+returns a list of extents.
+
+
+Request Basics
+--------------
+
+A fiemap request is encoded within struct fiemap:
+
+struct fiemap {
+	__u64	fm_start;	 /* logical offset (inclusive) at
+				  * which to start mapping (in) */
+	__u64	fm_length;	 /* logical length of mapping which
+				  * userspace cares about (in) */
+	__u32	fm_flags;	 /* FIEMAP_FLAG_* flags for request (in/out) */
+	__u32	fm_mapped_extents; /* number of extents that were
+				    * mapped (out) */
+	__u32	fm_extent_count; /* size of fm_extents array (in) */
+	__u32	fm_reserved;
+	struct fiemap_extent fm_extents[0]; /* array of mapped extents (out) */
+};
+
+
+fm_start, and fm_length specify the logical range within the file
+which the process would like mappings for. Extents returned mirror
+those on disk - that is, the logical offset of the 1st returned extent
+may start before fm_start, and the range covered by the last returned
+extent may end after fm_length. All offsets and lengths are in bytes.
+
+Certain flags to modify the way in which mappings are looked up can be
+set in fm_flags. If the kernel doesn't understand some particular
+flags, it will return EBADR and the contents of fm_flags will contain
+the set of flags which caused the error. If the kernel is compatible
+with all flags passed, the contents of fm_flags will be unmodified.
+It is up to userspace to determine whether rejection of a particular
+flag is fatal to it's operation. This scheme is intended to allow the
+fiemap interface to grow in the future but without losing
+compatibility with old software.
+
+fm_extent_count specifies the number of elements in the fm_extents[] array
+that can be used to return extents.  If fm_extent_count is zero, then the
+fm_extents[] array is ignored (no extents will be returned), and the
+fm_mapped_extents count will hold the number of extents needed in
+fm_extents[] to hold the file's current mapping.  Note that there is
+nothing to prevent the file from changing between calls to FIEMAP.
+
+The following flags can be set in fm_flags:
+
+* FIEMAP_FLAG_SYNC
+If this flag is set, the kernel will sync the file before mapping extents.
+
+* FIEMAP_FLAG_XATTR
+If this flag is set, the extents returned will describe the inodes
+extended attribute lookup tree, instead of it's data tree.
+
+
+Extent Mapping
+--------------
+
+Extent information is returned within the embedded fm_extents array
+which userspace must allocate along with the fiemap structure. The
+number of elements in the fiemap_extents[] array should be passed via
+fm_extent_count. The number of extents mapped by kernel will be
+returned via fm_mapped_extents. If the number of fiemap_extents
+allocated is less than would be required to map the requested range,
+the maximum number of extents that can be mapped in the fm_extent[]
+array will be returned and fm_mapped_extents will be equal to
+fm_extent_count. In that case, the last extent in the array will not
+complete the requested range and will not have the FIEMAP_EXTENT_LAST
+flag set (see the next section on extent flags).
+
+Each extent is described by a single fiemap_extent structure as
+returned in fm_extents.
+
+struct fiemap_extent {
+	__u64	fe_logical;  /* logical offset in bytes for the start of
+			      * the extent */
+	__u64	fe_physical; /* physical offset in bytes for the start
+			      * of the extent */
+	__u64	fe_length;   /* length in bytes for the extent */
+	__u64	fe_reserved64[2];
+	__u32	fe_flags;    /* FIEMAP_EXTENT_* flags for this extent */
+	__u32	fe_reserved[3];
+};
+
+All offsets and lengths are in bytes and mirror those on disk.  It is valid
+for an extents logical offset to start before the request or it's logical
+length to extend past the request.  Unless FIEMAP_EXTENT_NOT_ALIGNED is
+returned, fe_logical, fe_physical, and fe_length will be aligned to the
+block size of the file system.  With the exception of extents flagged as
+FIEMAP_EXTENT_MERGED, adjacent extents will not be merged.
+
+The fe_flags field contains flags which describe the extent returned.
+A special flag, FIEMAP_EXTENT_LAST is always set on the last extent in
+the file so that the process making fiemap calls can determine when no
+more extents are available, without having to call the ioctl again.
+
+Some flags are intentionally vague and will always be set in the
+presence of other more specific flags. This way a program looking for
+a general property does not have to know all existing and future flags
+which imply that property.
+
+For example, if FIEMAP_EXTENT_DATA_INLINE or FIEMAP_EXTENT_DATA_TAIL
+are set, FIEMAP_EXTENT_NOT_ALIGNED will also be set. A program looking
+for inline or tail-packed data can key on the specific flag. Software
+which simply cares not to try operating on non-aligned extents
+however, can just key on FIEMAP_EXTENT_NOT_ALIGNED, and not have to
+worry about all present and future flags which might imply unaligned
+data. Note that the opposite is not true - it would be valid for
+FIEMAP_EXTENT_NOT_ALIGNED to appear alone.
+
+* FIEMAP_EXTENT_LAST
+This is the last extent in the file. A mapping attempt past this
+extent will return nothing.
+
+* FIEMAP_EXTENT_UNKNOWN
+The location of this extent is currently unknown. This may indicate
+the data is stored on an inaccessible volume or that no storage has
+been allocated for the file yet.
+
+* FIEMAP_EXTENT_DELALLOC
+  - This will also set FIEMAP_EXTENT_UNKNOWN.
+Delayed allocation - while there is data for this extent, it's
+physical location has not been allocated yet.
+
+* FIEMAP_EXTENT_ENCODED
+This extent does not consist of plain filesystem blocks but is
+encoded (e.g. encrypted or compressed).  Reading the data in this
+extent via I/O to the block device will have undefined results.
+
+Note that it is *always* undefined to try to update the data
+in-place by writing to the indicated location without the
+assistance of the filesystem, or to access the data using the
+information returned by the FIEMAP interface while the filesystem
+is mounted.  In other words, user applications may only read the
+extent data via I/O to the block device while the filesystem is
+unmounted, and then only if the FIEMAP_EXTENT_ENCODED flag is
+clear; user applications must not try reading or writing to the
+filesystem via the block device under any other circumstances.
+
+* FIEMAP_EXTENT_DATA_ENCRYPTED
+  - This will also set FIEMAP_EXTENT_ENCODED
+The data in this extent has been encrypted by the file system.
+
+* FIEMAP_EXTENT_NOT_ALIGNED
+Extent offsets and length are not guaranteed to be block aligned.
+
+* FIEMAP_EXTENT_DATA_INLINE
+  This will also set FIEMAP_EXTENT_NOT_ALIGNED
+Data is located within a meta data block.
+
+* FIEMAP_EXTENT_DATA_TAIL
+  This will also set FIEMAP_EXTENT_NOT_ALIGNED
+Data is packed into a block with data from other files.
+
+* FIEMAP_EXTENT_UNWRITTEN
+Unwritten extent - the extent is allocated but it's data has not been
+initialized.  This indicates the extent's data will be all zero if read
+through the filesystem but the contents are undefined if read directly from
+the device.
+
+* FIEMAP_EXTENT_MERGED
+This will be set when a file does not support extents, i.e., it uses a block
+based addressing scheme.  Since returning an extent for each block back to
+userspace would be highly inefficient, the kernel will try to merge most
+adjacent blocks into 'extents'.
+
+
+VFS -> File System Implementation
+---------------------------------
+
+File systems wishing to support fiemap must implement a ->fiemap callback on
+their inode_operations structure. The fs ->fiemap call is responsible for
+defining it's set of supported fiemap flags, and calling a helper function on
+each discovered extent:
+
+struct inode_operations {
+       ...
+
+       int (*fiemap)(struct inode *, struct fiemap_extent_info *, u64 start,
+                     u64 len);
+
+->fiemap is passed struct fiemap_extent_info which describes the
+fiemap request:
+
+struct fiemap_extent_info {
+	unsigned int fi_flags;		/* Flags as passed from user */
+	unsigned int fi_extents_mapped;	/* Number of mapped extents */
+	unsigned int fi_extents_max;	/* Size of fiemap_extent array */
+	struct fiemap_extent *fi_extents_start;	/* Start of fiemap_extent array */
+};
+
+It is intended that the file system should not need to access any of this
+structure directly.
+
+
+Flag checking should be done at the beginning of the ->fiemap callback via the
+fiemap_check_flags() helper:
+
+int fiemap_check_flags(struct fiemap_extent_info *fieinfo, u32 fs_flags);
+
+The struct fieinfo should be passed in as recieved from ioctl_fiemap(). The
+set of fiemap flags which the fs understands should be passed via fs_flags. If
+fiemap_check_flags finds invalid user flags, it will place the bad values in
+fieinfo->fi_flags and return -EBADR. If the file system gets -EBADR, from
+fiemap_check_flags(), it should immediately exit, returning that error back to
+ioctl_fiemap().
+
+
+For each extent in the request range, the file system should call
+the helper function, fiemap_fill_next_extent():
+
+int fiemap_fill_next_extent(struct fiemap_extent_info *info, u64 logical,
+			    u64 phys, u64 len, u32 flags, u32 dev);
+
+fiemap_fill_next_extent() will use the passed values to populate the
+next free extent in the fm_extents array. 'General' extent flags will
+automatically be set from specific flags on behalf of the calling file
+system so that the userspace API is not broken.
+
+fiemap_fill_next_extent() returns 0 on success, and 1 when the
+user-supplied fm_extents array is full. If an error is encountered
+while copying the extent to user memory, -EFAULT will be returned.
diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt
index f566ad9bcb7b..b488edad743c 100644
--- a/Documentation/filesystems/proc.txt
+++ b/Documentation/filesystems/proc.txt
@@ -923,45 +923,44 @@ CPUs.
 The   "procs_blocked" line gives  the  number of  processes currently blocked,
 waiting for I/O to complete.
 
+
 1.9 Ext4 file system parameters
 ------------------------------
-Ext4 file system have one directory per partition under /proc/fs/ext4/
-# ls /proc/fs/ext4/hdc/
-group_prealloc  max_to_scan  mb_groups  mb_history  min_to_scan  order2_req
-stats  stream_req
-
-mb_groups:
-This file gives the details of multiblock allocator buddy cache of free blocks
-
-mb_history:
-Multiblock allocation history.
-
-stats:
-This file indicate whether the multiblock allocator should start collecting
-statistics. The statistics are shown during unmount
-
-group_prealloc:
-The multiblock allocator normalize the block allocation request to
-group_prealloc filesystem blocks if we don't have strip value set.
-The stripe value can be specified at mount time or during mke2fs.
-
-max_to_scan:
-How long multiblock allocator can look for a best extent (in found extents)
 
-min_to_scan:
-How long multiblock allocator  must look for a best extent
+Information about mounted ext4 file systems can be found in
+/proc/fs/ext4.  Each mounted filesystem will have a directory in
+/proc/fs/ext4 based on its device name (i.e., /proc/fs/ext4/hdc or
+/proc/fs/ext4/dm-0).   The files in each per-device directory are shown
+in Table 1-10, below.
 
-order2_req:
-Multiblock allocator use  2^N search using buddies only for requests greater
-than or equal to order2_req. The request size is specfied in file system
-blocks. A value of 2 indicate only if the requests are greater than or equal
-to 4 blocks.
+Table 1-10: Files in /proc/fs/ext4/<devname>
+..............................................................................
+ File            Content                                        
+ mb_groups       details of multiblock allocator buddy cache of free blocks
+ mb_history      multiblock allocation history
+ stats           controls whether the multiblock allocator should start
+                 collecting statistics, which are shown during the unmount
+ group_prealloc  the multiblock allocator will round up allocation
+                 requests to a multiple of this tuning parameter if the
+                 stripe size is not set in the ext4 superblock
+ max_to_scan     The maximum number of extents the multiblock allocator
+                 will search to find the best extent
+ min_to_scan     The minimum number of extents the multiblock allocator
+                 will search to find the best extent
+ order2_req      Tuning parameter which controls the minimum size for 
+                 requests (as a power of 2) where the buddy cache is
+                 used
+ stream_req      Files which have fewer blocks than this tunable
+                 parameter will have their blocks allocated out of a
+                 block group specific preallocation pool, so that small
+                 files are packed closely together.  Each large file
+                 will have its blocks allocated out of its own unique
+                 preallocation pool.
+inode_readahead  Tuning parameter which controls the maximum number of
+                 inode table blocks that ext4's inode table readahead
+                 algorithm will pre-read into the buffer cache
+..............................................................................
 
-stream_req:
-Files smaller than stream_req are served by the stream allocator, whose
-purpose is to pack requests as close each to other as possible to
-produce smooth I/O traffic. Avalue of 16 indicate that file smaller than 16
-filesystem block size will use group based preallocation.
 
 ------------------------------------------------------------------------------
 Summary
@@ -1332,13 +1331,6 @@ determine whether or not they are still functioning properly.
 Because the NMI watchdog shares registers with oprofile, by disabling the NMI
 watchdog, oprofile may have more registers to utilize.
 
-maps_protect
-------------
-
-Enables/Disables the protection of the per-process proc entries "maps" and
-"smaps".  When enabled, the contents of these files are visible only to
-readers that are allowed to ptrace() the given process.
-
 msgmni
 ------
 
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 2ca9c8f8c8d8..2443f5bb4364 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -658,11 +658,12 @@ and is between 256 and 4096 characters. It is defined in the file
 	earlyprintk=	[X86-32,X86-64,SH,BLACKFIN]
 			earlyprintk=vga
 			earlyprintk=serial[,ttySn[,baudrate]]
+			earlyprintk=dbgp
 
 			Append ",keep" to not disable it when the real console
 			takes over.
 
-			Only vga or serial at a time, not both.
+			Only vga or serial or usb debug port at a time.
 
 			Currently only ttyS0 and ttyS1 are supported.
 
@@ -1231,6 +1232,29 @@ and is between 256 and 4096 characters. It is defined in the file
 			         or
 			         memmap=0x10000$0x18690000
 
+	memory_corruption_check=0/1 [X86]
+			Some BIOSes seem to corrupt the first 64k of
+			memory when doing things like suspend/resume.
+			Setting this option will scan the memory
+			looking for corruption.  Enabling this will
+			both detect corruption and prevent the kernel
+			from using the memory being corrupted.
+			However, its intended as a diagnostic tool; if
+			repeatable BIOS-originated corruption always
+			affects the same memory, you can use memmap=
+			to prevent the kernel from using that memory.
+
+	memory_corruption_check_size=size [X86]
+			By default it checks for corruption in the low
+			64k, making this memory unavailable for normal
+			use.  Use this parameter to scan for
+			corruption in more or less memory.
+
+	memory_corruption_check_period=seconds [X86]
+			By default it checks for corruption every 60
+			seconds.  Use this parameter to check at some
+			other rate.  0 disables periodic checking.
+
 	memtest=	[KNL,X86] Enable memtest
 			Format: <integer>
 			range: 0,4 : pattern number
@@ -1428,6 +1452,12 @@ and is between 256 and 4096 characters. It is defined in the file
 
 	nolapic_timer	[X86-32,APIC] Do not use the local APIC timer.
 
+	nox2apic	[X86-64,APIC] Do not enable x2APIC mode.
+
+	x2apic_phys	[X86-64,APIC] Use x2apic physical mode instead of
+			default x2apic cluster mode on platforms
+			supporting x2apic.
+
 	noltlbs		[PPC] Do not use large page/tlb entries for kernel
 			lowmem mapping on PPC40x.
 
diff --git a/Documentation/sound/alsa/DocBook/writing-an-alsa-driver.tmpl b/Documentation/sound/alsa/DocBook/writing-an-alsa-driver.tmpl
index b54cb5048dfa..87a7c07ab658 100644
--- a/Documentation/sound/alsa/DocBook/writing-an-alsa-driver.tmpl
+++ b/Documentation/sound/alsa/DocBook/writing-an-alsa-driver.tmpl
@@ -5073,8 +5073,7 @@ struct _snd_pcm_runtime {
       with <constant>SNDRV_DMA_TYPE_CONTINUOUS</constant> type and the
       <function>snd_dma_continuous_data(GFP_KERNEL)</function> device pointer,
       where <constant>GFP_KERNEL</constant> is the kernel allocation flag to
-      use.  For the SBUS, <constant>SNDRV_DMA_TYPE_SBUS</constant> and
-      <function>snd_dma_sbus_data(sbus_dev)</function> are used instead.
+      use.
       For the PCI scatter-gather buffers, use
       <constant>SNDRV_DMA_TYPE_DEV_SG</constant> with
       <function>snd_dma_pci_data(pci)</function>
diff --git a/Documentation/sparc/sbus_drivers.txt b/Documentation/sparc/sbus_drivers.txt
deleted file mode 100644
index eb1e28ad8822..000000000000
--- a/Documentation/sparc/sbus_drivers.txt
+++ /dev/null
@@ -1,309 +0,0 @@
-
-		Writing SBUS Drivers
-
-	    David S. Miller (davem@redhat.com)
-
-	The SBUS driver interfaces of the Linux kernel have been
-revamped completely for 2.4.x for several reasons.  Foremost were
-performance and complexity concerns.  This document details these
-new interfaces and how they are used to write an SBUS device driver.
-
-	SBUS drivers need to include <asm/sbus.h> to get access
-to functions and structures described here.
-
-		Probing and Detection
-
-	Each SBUS device inside the machine is described by a
-structure called "struct sbus_dev".  Likewise, each SBUS bus
-found in the system is described by a "struct sbus_bus".  For
-each SBUS bus, the devices underneath are hung in a tree-like
-fashion off of the bus structure.
-
-	The SBUS device structure contains enough information
-for you to implement your device probing algorithm and obtain
-the bits necessary to run your device.  The most commonly
-used members of this structure, and their typical usage,
-will be detailed below.
-
-	Here is a piece of skeleton code for performing a device
-probe in an SBUS driver under Linux:
-
-	static int __devinit mydevice_probe_one(struct sbus_dev *sdev)
-	{
-		struct mysdevice *mp = kzalloc(sizeof(*mp), GFP_KERNEL);
-
-		if (!mp)
-			return -ENODEV;
-
-		...
-		dev_set_drvdata(&sdev->ofdev.dev, mp);
-		return 0;
-		...
-	}
-
-	static int __devinit mydevice_probe(struct of_device *dev,
-				            const struct of_device_id *match)
-	{
-		struct sbus_dev *sdev = to_sbus_device(&dev->dev);
-
-		return mydevice_probe_one(sdev);
-	}
-
-	static int __devexit mydevice_remove(struct of_device *dev)
-	{
-		struct sbus_dev *sdev = to_sbus_device(&dev->dev);
-		struct mydevice *mp = dev_get_drvdata(&dev->dev);
-
-		return mydevice_remove_one(sdev, mp);
-	}
-
-	static struct of_device_id mydevice_match[] = {
-		{
-			.name = "mydevice",
-		},
-		{},
-	};
-
-	MODULE_DEVICE_TABLE(of, mydevice_match);
-
-	static struct of_platform_driver mydevice_driver = {
-		.match_table	= mydevice_match,
-		.probe		= mydevice_probe,
-		.remove		= __devexit_p(mydevice_remove),
-		.driver		= {
-			.name		= "mydevice",
-		},
-	};
-
-	static int __init mydevice_init(void)
-	{
-		return of_register_driver(&mydevice_driver, &sbus_bus_type);
-	}
-
-	static void __exit mydevice_exit(void)
-	{
-		of_unregister_driver(&mydevice_driver);
-	}
-
-	module_init(mydevice_init);
-	module_exit(mydevice_exit);
-
-	The mydevice_match table is a series of entries which
-describes what SBUS devices your driver is meant for.  In the
-simplest case you specify a string for the 'name' field.  Every
-SBUS device with a 'name' property matching your string will
-be passed one-by-one to your .probe method.
-
-	You should store away your device private state structure
-pointer in the drvdata area so that you can retrieve it later on
-in your .remove method.
-
-	Any memory allocated, registers mapped, IRQs registered,
-etc. must be undone by your .remove method so that all resources
-of your device are released by the time it returns.
-
-	You should _NOT_ use the for_each_sbus(), for_each_sbusdev(),
-and for_all_sbusdev() interfaces.  They are deprecated, will be
-removed, and no new driver should reference them ever.
-
-		Mapping and Accessing I/O Registers
-
-	Each SBUS device structure contains an array of descriptors
-which describe each register set. We abuse struct resource for that.
-They each correspond to the "reg" properties provided by the OBP firmware.
-
-	Before you can access your device's registers you must map
-them.  And later if you wish to shutdown your driver (for module
-unload or similar) you must unmap them.  You must treat them as
-a resource, which you allocate (map) before using and free up
-(unmap) when you are done with it.
-
-	The mapping information is stored in an opaque value
-typed as an "unsigned long".  This is the type of the return value
-of the mapping interface, and the arguments to the unmapping
-interface.  Let's say you want to map the first set of registers.
-Perhaps part of your driver software state structure looks like:
-
-	struct mydevice {
-		unsigned long control_regs;
-	   ...
-		struct sbus_dev *sdev;
-	   ...
-	};
-
-	At initialization time you then use the sbus_ioremap
-interface to map in your registers, like so:
-
-	static void init_one_mydevice(struct sbus_dev *sdev)
-	{
-		struct mydevice *mp;
-		...
-
-		mp->control_regs = sbus_ioremap(&sdev->resource[0], 0,
-					CONTROL_REGS_SIZE, "mydevice regs");
-		if (!mp->control_regs) {
-			/* Failure, cleanup and return. */
-		}
-	}
-
-	Second argument to sbus_ioremap is an offset for
-cranky devices with broken OBP PROM. The sbus_ioremap uses only
-a start address and flags from the resource structure.
-Therefore it is possible to use the same resource to map
-several sets of registers or even to fabricate a resource
-structure if driver gets physical address from some private place.
-This practice is discouraged though. Use whatever OBP PROM
-provided to you.
-
-	And here is how you might unmap these registers later at
-driver shutdown or module unload time, using the sbus_iounmap
-interface:
-
-	static void mydevice_unmap_regs(struct mydevice *mp)
-	{
-		sbus_iounmap(mp->control_regs, CONTROL_REGS_SIZE);
-	}
-
-	Finally, to actually access your registers there are 6
-interface routines at your disposal.  Accesses are byte (8 bit),
-word (16 bit), or longword (32 bit) sized.  Here they are:
-
-	u8 sbus_readb(unsigned long reg)		/* read byte */
-	u16 sbus_readw(unsigned long reg)		/* read word */
-	u32 sbus_readl(unsigned long reg)		/* read longword */
-	void sbus_writeb(u8 value, unsigned long reg)	/* write byte */
-	void sbus_writew(u16 value, unsigned long reg)	/* write word */
-	void sbus_writel(u32 value, unsigned long reg)	/* write longword */
-
-	So, let's say your device has a control register of some sort
-at offset zero.  The following might implement resetting your device:
-
-	#define CONTROL		0x00UL
-
-	#define CONTROL_RESET	0x00000001	/* Reset hardware */
-
-	static void mydevice_reset(struct mydevice *mp)
-	{
-		sbus_writel(CONTROL_RESET, mp->regs + CONTROL);
-	}
-
-	Or perhaps there is a data port register at an offset of
-16 bytes which allows you to read bytes from a fifo in the device:
-
-	#define DATA		0x10UL
-
-	static u8 mydevice_get_byte(struct mydevice *mp)
-	{
-		return sbus_readb(mp->regs + DATA);
-	}
-
-	It's pretty straightforward, and clueful readers may have
-noticed that these interfaces mimick the PCI interfaces of the
-Linux kernel.  This was not by accident.
-
-	WARNING:
-
-		DO NOT try to treat these opaque register mapping
-		values as a memory mapped pointer to some structure
-		which you can dereference.
-
-		It may be memory mapped, it may not be.  In fact it
-		could be a physical address, or it could be the time
-		of day xor'd with 0xdeadbeef.  :-)
-
-		Whatever it is, it's an implementation detail.  The
-		interface was done this way to shield the driver
-		author from such complexities.
-
-			Doing DVMA
-
-	SBUS devices can perform DMA transactions in a way similar
-to PCI but dissimilar to ISA, e.g. DMA masters supply address.
-In contrast to PCI, however, that address (a bus address) is
-translated by IOMMU before a memory access is performed and therefore
-it is virtual. Sun calls this procedure DVMA.
-
-	Linux supports two styles of using SBUS DVMA: "consistent memory"
-and "streaming DVMA". CPU view of consistent memory chunk is, well,
-consistent with a view of a device. Think of it as an uncached memory.
-Typically this way of doing DVMA is not very fast and drivers use it
-mostly for control blocks or queues. On some CPUs we cannot flush or
-invalidate individual pages or cache lines and doing explicit flushing
-over ever little byte in every control block would be wasteful.
-
-Streaming DVMA is a preferred way to transfer large amounts of data.
-This process works in the following way:
-1. a CPU stops accessing a certain part of memory,
-   flushes its caches covering that memory;
-2. a device does DVMA accesses, then posts an interrupt;
-3. CPU invalidates its caches and starts to access the memory.
-
-A single streaming DVMA operation can touch several discontiguous
-regions of a virtual bus address space. This is called a scatter-gather
-DVMA.
-
-[TBD: Why do not we neither Solaris attempt to map disjoint pages
-into a single virtual chunk with the help of IOMMU, so that non SG
-DVMA masters would do SG? It'd be very helpful for RAID.]
-
-	In order to perform a consistent DVMA a driver does something
-like the following:
-
-	char *mem;		/* Address in the CPU space */
-	u32 busa;		/* Address in the SBus space */
-
-	mem = (char *) sbus_alloc_consistent(sdev, MYMEMSIZE, &busa);
-
-	Then mem is used when CPU accesses this memory and u32
-is fed to the device so that it can do DVMA. This is typically
-done with an sbus_writel() into some device register.
-
-	Do not forget to free the DVMA resources once you are done:
-
-	sbus_free_consistent(sdev, MYMEMSIZE, mem, busa);
-
-	Streaming DVMA is more interesting. First you allocate some
-memory suitable for it or pin down some user pages. Then it all works
-like this:
-
-	char *mem = argumen1;
-	unsigned int size = argument2;
-	u32 busa;		/* Address in the SBus space */
-
-	*mem = 1;		/* CPU can access */
-	busa = sbus_map_single(sdev, mem, size);
-	if (busa == 0) .......
-
-	/* Tell the device to use busa here */
-	/* CPU cannot access the memory without sbus_dma_sync_single() */
-
-	sbus_unmap_single(sdev, busa, size);
-	if (*mem == 0) ....	/* CPU can access again */
-
-	It is possible to retain mappings and ask the device to
-access data again and again without calling sbus_unmap_single.
-However, CPU caches must be invalidated with sbus_dma_sync_single
-before such access.
-
-[TBD but what about writeback caches here... do we have any?]
-
-	There is an equivalent set of functions doing the same thing
-only with several memory segments at once for devices capable of
-scatter-gather transfers. Use the Source, Luke.
-
-			Examples
-
-	drivers/net/sunhme.c
-	This is a complicated driver which illustrates many concepts
-discussed above and plus it handles both PCI and SBUS boards.
-
-	drivers/scsi/esp.c
-	Check it out for scatter-gather DVMA.
-
-	drivers/sbus/char/bpp.c
-	A non-DVMA device.
-
-	drivers/net/sunlance.c
-	Lance driver abuses consistent mappings for data transfer.
-It is a nifty trick which we do not particularly recommend...
-Just check it out and know that it's legal.
diff --git a/Documentation/timers/00-INDEX b/Documentation/timers/00-INDEX
new file mode 100644
index 000000000000..397dc35e1323
--- /dev/null
+++ b/Documentation/timers/00-INDEX
@@ -0,0 +1,10 @@
+00-INDEX
+	- this file
+highres.txt
+	- High resolution timers and dynamic ticks design notes
+hpet.txt
+	- High Precision Event Timer Driver for Linux
+hrtimers.txt
+	- subsystem for high-resolution kernel timers
+timer_stats.txt
+	- timer usage statistics
diff --git a/Documentation/hpet.txt b/Documentation/timers/hpet.txt
index 6ad52d9dad6c..e7c09abcfab4 100644
--- a/Documentation/hpet.txt
+++ b/Documentation/timers/hpet.txt
@@ -1,21 +1,32 @@
 		High Precision Event Timer Driver for Linux
 
-The High Precision Event Timer (HPET) hardware is the future replacement
-for the 8254 and Real Time Clock (RTC) periodic timer functionality.
-Each HPET can have up to 32 timers.  It is possible to configure the
-first two timers as legacy replacements for 8254 and RTC periodic timers.
-A specification done by Intel and Microsoft can be found at
-<http://www.intel.com/technology/architecture/hpetspec.htm>.
+The High Precision Event Timer (HPET) hardware follows a specification
+by Intel and Microsoft which can be found at
+
+	http://www.intel.com/technology/architecture/hpetspec.htm
+
+Each HPET has one fixed-rate counter (at 10+ MHz, hence "High Precision")
+and up to 32 comparators.  Normally three or more comparators are provided,
+each of which can generate oneshot interupts and at least one of which has
+additional hardware to support periodic interrupts.  The comparators are
+also called "timers", which can be misleading since usually timers are
+independent of each other ... these share a counter, complicating resets.
+
+HPET devices can support two interrupt routing modes.  In one mode, the
+comparators are additional interrupt sources with no particular system
+role.  Many x86 BIOS writers don't route HPET interrupts at all, which
+prevents use of that mode.  They support the other "legacy replacement"
+mode where the first two comparators block interrupts from 8254 timers
+and from the RTC.
 
 The driver supports detection of HPET driver allocation and initialization
 of the HPET before the driver module_init routine is called.  This enables
 platform code which uses timer 0 or 1 as the main timer to intercept HPET
 initialization.  An example of this initialization can be found in
-arch/i386/kernel/time_hpet.c.
+arch/x86/kernel/hpet.c.
 
-The driver provides two APIs which are very similar to the API found in
-the rtc.c driver.  There is a user space API and a kernel space API.
-An example user space program is provided below.
+The driver provides a userspace API which resembles the API found in the
+RTC driver framework.  An example user space program is provided below.
 
 #include <stdio.h>
 #include <stdlib.h>
@@ -286,15 +297,3 @@ out:
 
 	return;
 }
-
-The kernel API has three interfaces exported from the driver:
-
-	hpet_register(struct hpet_task *tp, int periodic)
-	hpet_unregister(struct hpet_task *tp)
-	hpet_control(struct hpet_task *tp, unsigned int cmd, unsigned long arg)
-
-The kernel module using this interface fills in the ht_func and ht_data
-members of the hpet_task structure before calling hpet_register.
-hpet_control simply vectors to the hpet_ioctl routine and has the same
-commands and respective arguments as the user API.  hpet_unregister
-is used to terminate usage of the HPET timer reserved by hpet_register.
diff --git a/MAINTAINERS b/MAINTAINERS
index 68781ed2b734..74b808205312 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -73,6 +73,7 @@ L: Mailing list that is relevant to this area
 W: Web-page with status/info
 T: SCM tree type and location.  Type is one of: git, hg, quilt.
 S: Status, one of the following:
+F: Applicable files and/or directories
 
 	Supported:	Someone is actually paid to look after this.
 	Maintained:	Someone actually looks after it.
@@ -102,14 +103,14 @@ M:	dave@thedillows.org
 L:	netdev@vger.kernel.org
 S:	Maintained
 
-3W-XXXX ATA-RAID CONTROLLER DRIVER
+3W-9XXX SATA-RAID CONTROLLER DRIVER
 P:	Adam Radford
 M:	linuxraid@amcc.com
 L:	linux-scsi@vger.kernel.org
 W:	http://www.amcc.com
 S:	Supported
 
-3W-9XXX SATA-RAID CONTROLLER DRIVER
+3W-XXXX ATA-RAID CONTROLLER DRIVER
 P:	Adam Radford
 M:	linuxraid@amcc.com
 L:	linux-scsi@vger.kernel.org
@@ -163,16 +164,11 @@ M:	A2232@gmx.net
 L:	linux-m68k@lists.linux-m68k.org
 S:	Maintained
 
-AFS FILESYSTEM & AF_RXRPC SOCKET DOMAIN
-P:	David Howells
-M:	dhowells@redhat.com
-L:	linux-afs@lists.infradead.org
-S:	Supported
-
-AIO
-P:	Benjamin LaHaise
-M:	bcrl@kvack.org
-L:	linux-aio@kvack.org
+AACRAID SCSI RAID DRIVER
+P:	Adaptec OEM Raid Solutions
+M:	aacraid@adaptec.com
+L:	linux-scsi@vger.kernel.org
+W:	http://www.adaptec.com/
 S:	Supported
 
 ABIT UGURU 1,2 HARDWARE MONITOR DRIVER
@@ -193,27 +189,6 @@ M:	jes@trained-monkey.org
 L:	linux-acenic@sunsite.dk
 S:	Maintained
 
-IPS SCSI RAID DRIVER
-P:	Adaptec OEM Raid Solutions
-M:	aacraid@adaptec.com
-L:	linux-scsi@vger.kernel.org
-W:	http://www.adaptec.com/
-S:	Maintained
-
-DPT_I2O SCSI RAID DRIVER
-P:	Adaptec OEM Raid Solutions
-M:	aacraid@adaptec.com
-L:	linux-scsi@vger.kernel.org
-W:	http://www.adaptec.com/
-S:	Maintained
-
-AACRAID SCSI RAID DRIVER
-P:	Adaptec OEM Raid Solutions
-M:	aacraid@adaptec.com
-L:	linux-scsi@vger.kernel.org
-W:	http://www.adaptec.com/
-S:	Supported
-
 ACER WMI LAPTOP EXTRAS
 P:	Carlos Corbacho
 M:	carlos@strangeworlds.co.uk
@@ -327,6 +302,12 @@ P:	Roman Zippel
 M:	zippel@linux-m68k.org
 S:	Maintained
 
+AFS FILESYSTEM & AF_RXRPC SOCKET DOMAIN
+P:	David Howells
+M:	dhowells@redhat.com
+L:	linux-afs@lists.infradead.org
+S:	Supported
+
 AGPGART DRIVER
 P:	David Airlie
 M:	airlied@linux.ie
@@ -345,6 +326,12 @@ M:	hare@suse.de
 L:	linux-scsi@vger.kernel.org
 S:	Maintained
 
+AIO
+P:	Benjamin LaHaise
+M:	bcrl@kvack.org
+L:	linux-aio@kvack.org
+S:	Supported
+
 ALCATEL SPEEDTOUCH USB DRIVER
 P:	Duncan Sands
 M:	duncan.sands@free.fr
@@ -390,6 +377,11 @@ L:	iommu@lists.linux-foundation.org
 T:	git://git.kernel.org/pub/scm/linux/kernel/git/joro/linux-2.6-iommu.git
 S:	Supported
 
+AMD MICROCODE UPDATE SUPPORT
+P:      Peter Oruba
+M:      peter.oruba@amd.com
+S:      Supported
+
 AMS (Apple Motion Sensor) DRIVER
 P:	Stelian Pop
 M:	stelian@popies.net
@@ -732,7 +724,7 @@ L:	linux-kernel@vger.kernel.org
 W:	http://sourceforge.net/projects/xscaleiop
 S:	Supported
 
-ATA OVER ETHERNET DRIVER
+ATA OVER ETHERNET (AOE) DRIVER
 P:	Ed L. Cashin
 M:	ecashin@coraid.com
 W:	http://www.coraid.com/support/linux
@@ -865,11 +857,48 @@ L:	linux-hams@vger.kernel.org
 W:	http://www.linux-ax25.org/
 S:	Maintained
 
+B43 WIRELESS DRIVER
+P:	Michael Buesch
+M:	mb@bu3sch.de
+P:	Stefano Brivio
+M:	stefano.brivio@polimi.it
+L:	linux-wireless@vger.kernel.org
+W:	http://linuxwireless.org/en/users/Drivers/b43
+S:	Maintained
+
+B43LEGACY WIRELESS DRIVER
+P:	Larry Finger
+M:	Larry.Finger@lwfinger.net
+P:	Stefano Brivio
+M:	stefano.brivio@polimi.it
+L:	linux-wireless@vger.kernel.org
+W:	http://linuxwireless.org/en/users/Drivers/b43
+S:	Maintained
+
 BACKLIGHT CLASS/SUBSYSTEM
 P:	Richard Purdie
 M:	rpurdie@rpsys.net
 S:	Maintained
 
+BAYCOM/HDLCDRV DRIVERS FOR AX.25
+P:	Thomas Sailer
+M:	t.sailer@alumni.ethz.ch
+L:	linux-hams@vger.kernel.org
+W:	http://www.baycom.org/~tom/ham/ham.html
+S:	Maintained
+
+BEFS FILE SYSTEM
+P:	Sergey S. Kostyliov
+M:	rathamahata@php4.ru
+L:	linux-kernel@vger.kernel.org
+S:	Maintained
+
+BFS FILE SYSTEM
+P:	Tigran A. Aivazian
+M:	tigran@aivazian.fsnet.co.uk
+L:	linux-kernel@vger.kernel.org
+S:	Maintained
+
 BLACKFIN ARCHITECTURE
 P:	Bryan Wu
 M:	cooloney@kernel.org
@@ -905,43 +934,6 @@ L:	uclinux-dist-devel@blackfin.uclinux.org (subscribers-only)
 W:	http://blackfin.uclinux.org
 S:	Supported
 
-BAYCOM/HDLCDRV DRIVERS FOR AX.25
-P:	Thomas Sailer
-M:	t.sailer@alumni.ethz.ch
-L:	linux-hams@vger.kernel.org
-W:	http://www.baycom.org/~tom/ham/ham.html
-S:	Maintained
-
-B43 WIRELESS DRIVER
-P:	Michael Buesch
-M:	mb@bu3sch.de
-P:	Stefano Brivio
-M:	stefano.brivio@polimi.it
-L:	linux-wireless@vger.kernel.org
-W:	http://linuxwireless.org/en/users/Drivers/b43
-S:	Maintained
-
-B43LEGACY WIRELESS DRIVER
-P:	Larry Finger
-M:	Larry.Finger@lwfinger.net
-P:	Stefano Brivio
-M:	stefano.brivio@polimi.it
-L:	linux-wireless@vger.kernel.org
-W:	http://linuxwireless.org/en/users/Drivers/b43
-S:	Maintained
-
-BEFS FILE SYSTEM
-P:	Sergey S. Kostyliov
-M:	rathamahata@php4.ru
-L:	linux-kernel@vger.kernel.org
-S:	Maintained
-
-BFS FILE SYSTEM
-P:	Tigran A. Aivazian
-M:	tigran@aivazian.fsnet.co.uk
-L:	linux-kernel@vger.kernel.org
-S:	Maintained
-
 BLACKFIN I2C TWI DRIVER
 P:	Sonic Zhang
 M:	sonic.zhang@analog.com
@@ -1035,15 +1027,6 @@ M:	corbet@lwn.net
 L:	video4linux-list@redhat.com
 S:	Maintained
 
-CAN NETWORK LAYER
-P:	Urs Thuermann
-M:	urs.thuermann@volkswagen.de
-P:	Oliver Hartkopp
-M:	oliver.hartkopp@volkswagen.de
-L:	socketcan-core@lists.berlios.de (subscribers-only)
-W:	http://developer.berlios.de/projects/socketcan/
-S:	Maintained
-
 CALGARY x86-64 IOMMU
 P:	Muli Ben-Yehuda
 M:	muli@il.ibm.com
@@ -1053,6 +1036,15 @@ L:	linux-kernel@vger.kernel.org
 L:	discuss@x86-64.org
 S:	Maintained
 
+CAN NETWORK LAYER
+P:	Urs Thuermann
+M:	urs.thuermann@volkswagen.de
+P:	Oliver Hartkopp
+M:	oliver.hartkopp@volkswagen.de
+L:	socketcan-core@lists.berlios.de (subscribers-only)
+W:	http://developer.berlios.de/projects/socketcan/
+S:	Maintained
+
 CELL BROADBAND ENGINE ARCHITECTURE
 P:	Arnd Bergmann
 M:	arnd@arndb.de
@@ -1061,13 +1053,6 @@ L:	cbe-oss-dev@ozlabs.org
 W:	http://www.ibm.com/developerworks/power/cell/
 S:	Supported
 
-CISCO 10G ETHERNET DRIVER
-P:	Scott Feldman
-M:	scofeldm@cisco.com
-P:	Joe Eykholt
-M:	jeykholt@cisco.com
-S:	Supported
-
 CFAG12864B LCD DRIVER
 P:	Miguel Ojeda Sandonis
 M:	miguel.ojeda.sandonis@gmail.com
@@ -1099,19 +1084,11 @@ P:	Joel Schopp
 M:	jschopp@austin.ibm.com
 S:	Supported
 
-COMMON INTERNET FILE SYSTEM (CIFS)
-P:	Steve French
-M:	sfrench@samba.org
-L:	linux-cifs-client@lists.samba.org
-L:	samba-technical@lists.samba.org
-W:	http://linux-cifs.samba.org/
-T:	git kernel.org:/pub/scm/linux/kernel/git/sfrench/cifs-2.6.git
-S:	Supported
-
-CONFIGFS
-P:	Joel Becker
-M:	joel.becker@oracle.com
-L:	linux-kernel@vger.kernel.org
+CISCO 10G ETHERNET DRIVER
+P:	Scott Feldman
+M:	scofeldm@cisco.com
+P:	Joe Eykholt
+M:	jeykholt@cisco.com
 S:	Supported
 
 CIRRUS LOGIC EP93XX ETHERNET DRIVER
@@ -1147,6 +1124,15 @@ L:	codalist@coda.cs.cmu.edu
 W:	http://www.coda.cs.cmu.edu/
 S:	Maintained
 
+COMMON INTERNET FILE SYSTEM (CIFS)
+P:	Steve French
+M:	sfrench@samba.org
+L:	linux-cifs-client@lists.samba.org
+L:	samba-technical@lists.samba.org
+W:	http://linux-cifs.samba.org/
+T:	git kernel.org:/pub/scm/linux/kernel/git/sfrench/cifs-2.6.git
+S:	Supported
+
 COMPACTPCI HOTPLUG CORE
 P:	Scott Murray
 M:	scottm@somanetworks.com
@@ -1186,6 +1172,12 @@ L:	accessrunner-general@lists.sourceforge.net
 W:	http://accessrunner.sourceforge.net/
 S:	Maintained
 
+CONFIGFS
+P:	Joel Becker
+M:	joel.becker@oracle.com
+L:	linux-kernel@vger.kernel.org
+S:	Supported
+
 CONTROL GROUPS (CGROUPS)
 P:	Paul Menage
 M:	menage@google.com
@@ -1309,6 +1301,20 @@ W:	http://www.qsl.net/dl1bke/
 L:	linux-hams@vger.kernel.org
 S:	Maintained
 
+DAVICOM FAST ETHERNET (DMFE) NETWORK DRIVER
+P:	Tobias Ringstrom
+M:	tori@unhappy.mine.nu
+L:	netdev@vger.kernel.org
+S:	Maintained
+
+DC390/AM53C974 SCSI driver
+P:	Kurt Garloff
+M:	garloff@suse.de
+W:	http://www.garloff.de/kurt/linux/dc390/
+P:	Guennadi Liakhovetski
+M:	g.liakhovetski@gmx.de
+S:	Maintained
+
 DC395x SCSI driver
 P:	Oliver Neukum
 M:	oliver@neukum.name
@@ -1321,14 +1327,6 @@ L:	dc395x@twibble.org
 L:	http://lists.twibble.org/mailman/listinfo/dc395x/
 S:	Maintained
 
-DC390/AM53C974 SCSI driver
-P:	Kurt Garloff
-M:	garloff@suse.de
-W:	http://www.garloff.de/kurt/linux/dc390/
-P:	Guennadi Liakhovetski
-M:	g.liakhovetski@gmx.de
-S:	Maintained
-
 DCCP PROTOCOL
 P:	Arnaldo Carvalho de Melo
 M:	acme@ghostprotocols.net
@@ -1359,12 +1357,6 @@ P:	Doug Warzecha
 M:	Douglas_Warzecha@dell.com
 S:	Maintained
 
-DEVICE-MAPPER  (LVM)
-P:	Alasdair Kergon
-L:	dm-devel@redhat.com
-W:	http://sources.redhat.com/dm
-S:	Maintained
-
 DEVICE NUMBER REGISTRY
 P:	Torben Mathiasen
 M:	device@lanana.org
@@ -1372,6 +1364,12 @@ W:	http://lanana.org/docs/device-list/index.html
 L:	linux-kernel@vger.kernel.org
 S:	Maintained
 
+DEVICE-MAPPER  (LVM)
+P:	Alasdair Kergon
+L:	dm-devel@redhat.com
+W:	http://sources.redhat.com/dm
+S:	Maintained
+
 DIGI INTL. EPCA DRIVER
 P:	Digi International, Inc
 M:	Eng.Linux@digi.com
@@ -1379,7 +1377,7 @@ L:	Eng.Linux@digi.com
 W:	http://www.digi.com
 S:	Orphan
 
-DIRECTORY NOTIFICATION
+DIRECTORY NOTIFICATION (DNOTIFY)
 P:	Stephen Rothwell
 M:	sfr@canb.auug.org.au
 L:	linux-kernel@vger.kernel.org
@@ -1393,13 +1391,13 @@ W:	http://www.win.tue.nl/~aeb/linux/zip/zip-1.html
 W:	http://www.win.tue.nl/~aeb/partitions/partition_types-1.html
 S:	Maintained
 
-DISKQUOTA:
+DISKQUOTA
 P:	Jan Kara
 M:	jack@suse.cz
 L:	linux-kernel@vger.kernel.org
 S:	Maintained
 
-DISTRIBUTED LOCK MANAGER
+DISTRIBUTED LOCK MANAGER (DLM)
 P:	Christine Caulfield
 M:	ccaulfie@redhat.com
 P:	David Teigland
@@ -1409,12 +1407,6 @@ W:	http://sources.redhat.com/cluster/
 T:	git kernel.org:/pub/scm/linux/kernel/git/teigland/dlm.git
 S:	Supported
 
-DAVICOM FAST ETHERNET (DMFE) NETWORK DRIVER
-P:	Tobias Ringstrom
-M:	tori@unhappy.mine.nu
-L:	netdev@vger.kernel.org
-S:	Maintained
-
 DMA GENERIC OFFLOAD ENGINE SUBSYSTEM
 P:	Maciej Sosnowski
 M:	maciej.sosnowski@intel.com
@@ -1454,6 +1446,13 @@ M:	jrv@vanzandt.mv.com
 L:	blinux-list@redhat.com
 S:	Maintained
 
+DPT_I2O SCSI RAID DRIVER
+P:	Adaptec OEM Raid Solutions
+M:	aacraid@adaptec.com
+L:	linux-scsi@vger.kernel.org
+W:	http://www.adaptec.com/
+S:	Maintained
+
 DRIVER CORE, KOBJECTS, AND SYSFS
 P:	Greg Kroah-Hartman
 M:	gregkh@suse.de
@@ -1602,7 +1601,7 @@ EFS FILESYSTEM
 W:	http://aeschi.ch.eu.org/efs/
 S:	Orphan
 
-EHCA (IBM GX bus InfiniBand adapter) DRIVER:
+EHCA (IBM GX bus InfiniBand adapter) DRIVER
 P:	Hoang-Nam Nguyen
 M:	hnguyen@de.ibm.com
 P:	Christoph Raisch
@@ -1659,9 +1658,10 @@ L:	linux-ext4@vger.kernel.org
 S:	Maintained
 
 EXT4 FILE SYSTEM
-P:	Stephen Tweedie, Andrew Morton
-M:	sct@redhat.com, akpm@linux-foundation.org, adilger@sun.com
+P:	Theodore Ts'o
+M:	tytso@mit.edu, adilger@sun.com
 L:	linux-ext4@vger.kernel.org
+W:	http://ext4.wiki.kernel.org
 S:	Maintained
 
 F71805F HARDWARE MONITORING DRIVER
@@ -1681,6 +1681,42 @@ P:	Akinobu Mita
 M:	akinobu.mita@gmail.com
 S:	Supported
 
+FILE LOCKING (flock() and fcntl()/lockf())
+P:	Matthew Wilcox
+M:	matthew@wil.cx
+L:	linux-fsdevel@vger.kernel.org
+S:	Maintained
+
+FILESYSTEMS (VFS and infrastructure)
+P:	Alexander Viro
+M:	viro@zeniv.linux.org.uk
+L:	linux-fsdevel@vger.kernel.org
+S:	Maintained
+
+FIREWIRE SUBSYSTEM (drivers/firewire, <linux/firewire*.h>)
+P:	Kristian Hoegsberg, Stefan Richter
+M:	krh@redhat.com, stefanr@s5r6.in-berlin.de
+L:	linux1394-devel@lists.sourceforge.net
+W:	http://www.linux1394.org/
+T:	git kernel.org:/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6.git
+S:	Maintained
+
+FIRMWARE LOADER (request_firmware)
+L:	linux-kernel@vger.kernel.org
+S:	Orphan
+
+FPU EMULATOR
+P:	Bill Metzenthen
+M:	billm@suburbia.net
+W:	http://suburbia.net/~billm/floating-point/emulator/
+S:	Maintained
+
+FRAME RELAY DLCI/FRAD (Sangoma drivers too)
+P:	Mike McLagan
+M:	mike.mclagan@linux.org
+L:	netdev@vger.kernel.org
+S:	Maintained
+
 FRAMEBUFFER LAYER
 P:	Antonino Daplas
 M:	adaplas@gmail.com
@@ -1746,42 +1782,6 @@ L:	alsa-devel@alsa-project.org
 L:	linuxppc-dev@ozlabs.org
 S:	Supported
 
-FILE LOCKING (flock() and fcntl()/lockf())
-P:	Matthew Wilcox
-M:	matthew@wil.cx
-L:	linux-fsdevel@vger.kernel.org
-S:	Maintained
-
-FILESYSTEMS (VFS and infrastructure)
-P:	Alexander Viro
-M:	viro@zeniv.linux.org.uk
-L:	linux-fsdevel@vger.kernel.org
-S:	Maintained
-
-FIREWIRE SUBSYSTEM (drivers/firewire, <linux/firewire*.h>)
-P:	Kristian Hoegsberg, Stefan Richter
-M:	krh@redhat.com, stefanr@s5r6.in-berlin.de
-L:	linux1394-devel@lists.sourceforge.net
-W:	http://www.linux1394.org/
-T:	git kernel.org:/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6.git
-S:	Maintained
-
-FIRMWARE LOADER (request_firmware)
-L:	linux-kernel@vger.kernel.org
-S:	Orphan
-
-FPU EMULATOR
-P:	Bill Metzenthen
-M:	billm@suburbia.net
-W:	http://suburbia.net/~billm/floating-point/emulator/
-S:	Maintained
-
-FRAME RELAY DLCI/FRAD (Sangoma drivers too)
-P:	Mike McLagan
-M:	mike.mclagan@linux.org
-L:	netdev@vger.kernel.org
-S:	Maintained
-
 FREEVXFS FILESYSTEM
 P:	Christoph Hellwig
 M:	hch@infradead.org
@@ -1853,6 +1853,13 @@ L:	gigaset307x-common@lists.sourceforge.net
 W:	http://gigaset307x.sourceforge.net/
 S:	Maintained
 
+HARD DRIVE ACTIVE PROTECTION SYSTEM (HDAPS) DRIVER
+P:	Robert Love
+M:	rlove@rlove.org
+M:	linux-kernel@vger.kernel.org
+W:	http://www.kernel.org/pub/linux/kernel/people/rml/hdaps/
+S:	Maintained
+
 HARDWARE MONITORING
 L:	lm-sensors@lm-sensors.org
 W:	http://www.lm-sensors.org/
@@ -1861,13 +1868,6 @@ S:	Orphan
 HARDWARE RANDOM NUMBER GENERATOR CORE
 S:	Orphan
 
-HARD DRIVE ACTIVE PROTECTION SYSTEM (HDAPS) DRIVER
-P:	Robert Love
-M:	rlove@rlove.org
-M:	linux-kernel@vger.kernel.org
-W:	http://www.kernel.org/pub/linux/kernel/people/rml/hdaps/
-S:	Maintained
-
 HARMONY SOUND DRIVER
 P:	Kyle McMartin
 M:	kyle@mcmartin.ca
@@ -1881,6 +1881,24 @@ L:	linux-kernel@vger.kernel.org
 W:	http://www.nyx.net/~arobinso
 S:	Maintained
 
+HEWLETT-PACKARD FIBRE CHANNEL 64-bit/66MHz PCI non-intelligent HBA
+P:	Chirag Kantharia
+M:	chirag.kantharia@hp.com
+L:	iss_storagedev@hp.com
+S:	Maintained
+
+HEWLETT-PACKARD SMART2 RAID DRIVER
+P:	Chirag Kantharia
+M:	chirag.kantharia@hp.com
+L:	iss_storagedev@hp.com
+S:	Maintained
+
+HEWLETT-PACKARD SMART CISS RAID DRIVER (cciss)
+P:	Mike Miller
+M:	mike.miller@hp.com
+L:	iss_storagedev@hp.com
+S:	Supported
+
 HFS FILESYSTEM
 P:	Roman Zippel
 M:	zippel@linux-m68k.org
@@ -1894,6 +1912,14 @@ L:	linux-nvidia@lists.surfsouth.com
 W:	http://drama.obuda.kando.hu/~fero/cgi-bin/hgafb.shtml
 S:	Maintained
 
+HIBERNATION (aka Software Suspend, aka swsusp)
+P:	Pavel Machek
+M:	pavel@suse.cz
+P:	Rafael J. Wysocki
+M:	rjw@sisk.pl
+L:	linux-pm@lists.linux-foundation.org
+S:	Supported
+
 HID CORE LAYER
 P:	Jiri Kosina
 M:	jkosina@suse.cz
@@ -1926,24 +1952,6 @@ M:	jes@trained-monkey.org
 L:	linux-hippi@sunsite.dk
 S:	Maintained
 
-HEWLETT-PACKARD FIBRE CHANNEL 64-bit/66MHz PCI non-intelligent HBA
-P:	Chirag Kantharia
-M:	chirag.kantharia@hp.com
-L:	iss_storagedev@hp.com
-S:	Maintained
-
-HEWLETT-PACKARD SMART2 RAID DRIVER
-P:	Chirag Kantharia
-M:	chirag.kantharia@hp.com
-L:	iss_storagedev@hp.com
-S:	Maintained
-
-HEWLETT-PACKARD SMART CISS RAID DRIVER (cciss)
-P:	Mike Miller
-M:	mike.miller@hp.com
-L:	iss_storagedev@hp.com
-S:	Supported
-
 HOST AP DRIVER
 P:	Jouni Malinen
 M:	j@w1.fi
@@ -1952,16 +1960,16 @@ L:	linux-wireless@vger.kernel.org
 W:	http://hostap.epitest.fi/
 S:	Maintained
 
-HP100:	Driver for HP 10/100 Mbit/s Voice Grade Network Adapter Series
-P:	Jaroslav Kysela
-M:	perex@perex.cz
-S:	Maintained
-
 HP COMPAQ TC1100 TABLET WMI EXTRAS DRIVER
 P:	Carlos Corbacho
 M:	carlos@strangeworlds.co.uk
 S:	Odd Fixes
 
+HP100:	Driver for HP 10/100 Mbit/s Voice Grade Network Adapter Series
+P:	Jaroslav Kysela
+M:	perex@perex.cz
+S:	Maintained
+
 HPET:	High Precision Event Timers driver (drivers/char/hpet.c)
 P:	Clemens Ladisch
 M:	clemens@ladisch.de
@@ -2041,14 +2049,6 @@ W:	http://www.ia64-linux.org/
 T:	git kernel.org:/pub/scm/linux/kernel/git/aegl/linux-2.6.git
 S:	Maintained
 
-SN-IA64 (Itanium) SUB-PLATFORM
-P:	Jes Sorensen
-M:	jes@sgi.com
-L:	linux-altix@sgi.com
-L:	linux-ia64@vger.kernel.org
-W:	http://www.sgi.com/altix
-S:	Maintained
-
 IBM MCA SCSI SUBSYSTEM DRIVER
 P:	Michael Lang
 M:	langa2@kph.uni-mainz.de
@@ -2132,14 +2132,6 @@ W:	http://www.openib.org/
 T:	git kernel.org:/pub/scm/linux/kernel/git/roland/infiniband.git
 S:	Supported
 
-INPUT (KEYBOARD, MOUSE, JOYSTICK, TOUCHSCREEN) DRIVERS
-P:	Dmitry Torokhov
-M:	dmitry.torokhov@gmail.com
-M:	dtor@mail.ru
-L:	linux-input@vger.kernel.org
-T:	git kernel.org:/pub/scm/linux/kernel/git/dtor/input.git
-S:	Maintained
-
 INOTIFY
 P:	John McCutchan
 M:	ttb@tentacle.dhs.org
@@ -2148,6 +2140,14 @@ M:	rml@novell.com
 L:	linux-kernel@vger.kernel.org
 S:	Maintained
 
+INPUT (KEYBOARD, MOUSE, JOYSTICK, TOUCHSCREEN) DRIVERS
+P:	Dmitry Torokhov
+M:	dmitry.torokhov@gmail.com
+M:	dtor@mail.ru
+L:	linux-input@vger.kernel.org
+T:	git kernel.org:/pub/scm/linux/kernel/git/dtor/input.git
+S:	Maintained
+
 INTEL FRAMEBUFFER DRIVER (excluding 810 and 815)
 P:	Sylvain Meyer
 M:	sylvain.meyer@worldonline.fr
@@ -2252,7 +2252,7 @@ M:	pfg@sgi.com
 L:	linux-mips@linux-mips.org
 S:	Maintained
 
-IP MASQUERADING:
+IP MASQUERADING
 P:	Juanjo Ciarlante
 M:	jjciarla@raiz.uncu.edu.ar
 S:	Maintained
@@ -2267,7 +2267,7 @@ M:	jesse@icplus.com.tw
 L:	netdev@vger.kernel.org
 S:	Maintained
 
-IPATH DRIVER:
+IPATH DRIVER
 P:	Ralph Campbell
 M:	infinipath@qlogic.com
 L:	general@lists.openfabrics.org
@@ -2281,13 +2281,25 @@ L:	openipmi-developer@lists.sourceforge.net
 W:	http://openipmi.sourceforge.net/
 S:	Supported
 
-IPX NETWORK LAYER
-P:	Arnaldo Carvalho de Melo
-M:	acme@ghostprotocols.net
+IPS SCSI RAID DRIVER
+P:	Adaptec OEM Raid Solutions
+M:	aacraid@adaptec.com
+L:	linux-scsi@vger.kernel.org
+W:	http://www.adaptec.com/
+S:	Maintained
+
+IPVS
+P:	Wensong Zhang
+M:	wensong@linux-vs.org
+P:	Simon Horman
+M:	horms@verge.net.au
+P:	Julian Anastasov
+M:	ja@ssi.bg
 L:	netdev@vger.kernel.org
+L:	lvs-devel@vger.kernel.org
 S:	Maintained
 
-IPWIRELES DRIVER
+IPWIRELESS DRIVER
 P:	Jiri Kosina
 M:	jkosina@suse.cz
 P:	David Sterba
@@ -2295,6 +2307,12 @@ M:	dsterba@suse.cz
 S:	Maintained
 T:	git://git.kernel.org/pub/scm/linux/kernel/git/jikos/ipwireless_cs.git
 
+IPX NETWORK LAYER
+P:	Arnaldo Carvalho de Melo
+M:	acme@ghostprotocols.net
+L:	netdev@vger.kernel.org
+S:	Maintained
+
 IRDA SUBSYSTEM
 P:	Samuel Ortiz
 M:	samuel@sortiz.org
@@ -2302,6 +2320,11 @@ L:	irda-users@lists.sourceforge.net (subscribers-only)
 W:	http://irda.sourceforge.net/
 S:	Maintained
 
+ISAPNP
+P:	Jaroslav Kysela
+M:	perex@perex.cz
+S:	Maintained
+
 ISCSI
 P:	Mike Christie
 M:	michaelc@cs.wisc.edu
@@ -2310,11 +2333,6 @@ W:	www.open-iscsi.org
 T:	git kernel.org:/pub/scm/linux/kernel/mnc/linux-2.6-iscsi.git
 S:	Maintained
 
-ISAPNP
-P:	Jaroslav Kysela
-M:	perex@perex.cz
-S:	Maintained
-
 ISDN SUBSYSTEM
 P:	Karsten Keil
 M:	kkeil@suse.de
@@ -2339,6 +2357,14 @@ L:	video4linux-list@redhat.com
 W:	http://www.ivtvdriver.org
 S:	Maintained
 
+JFS FILESYSTEM
+P:	Dave Kleikamp
+M:	shaggy@austin.ibm.com
+L:	jfs-discussion@lists.sourceforge.net
+W:	http://jfs.sourceforge.net/
+T:	git kernel.org:/pub/scm/linux/kernel/git/shaggy/jfs-2.6.git
+S:	Supported
+
 JME NETWORK DRIVER
 P:	Guo-Fu Tseng
 M:	cooldavid@cooldavid.org
@@ -2352,24 +2378,6 @@ L:	linux-mtd@lists.infradead.org
 W:	http://www.linux-mtd.infradead.org/doc/jffs2.html
 S:	Maintained
 
-UBI FILE SYSTEM (UBIFS)
-P:	Artem Bityutskiy
-M:	dedekind@infradead.org
-P:	Adrian Hunter
-M:	ext-adrian.hunter@nokia.com
-L:	linux-mtd@lists.infradead.org
-T:	git git://git.infradead.org/~dedekind/ubifs-2.6.git
-W:	http://www.linux-mtd.infradead.org/doc/ubifs.html
-S:	Maintained
-
-JFS FILESYSTEM
-P:	Dave Kleikamp
-M:	shaggy@austin.ibm.com
-L:	jfs-discussion@lists.sourceforge.net
-W:	http://jfs.sourceforge.net/
-T:	git kernel.org:/pub/scm/linux/kernel/git/shaggy/jfs-2.6.git
-S:	Supported
-
 JOURNALLING LAYER FOR BLOCK DEVICES (JBD)
 P:	Stephen Tweedie, Andrew Morton
 M:	sct@redhat.com, akpm@linux-foundation.org
@@ -2606,11 +2614,6 @@ W:	http://www.pasemi.com/
 L:	linuxppc-dev@ozlabs.org
 S:	Supported
 
-LLC (802.2)
-P:	Arnaldo Carvalho de Melo
-M:	acme@ghostprotocols.net
-S:	Maintained
-
 LINUX SECURITY MODULE (LSM) FRAMEWORK
 P:	Chris Wright
 M:	chrisw@sous-sol.org
@@ -2618,6 +2621,11 @@ L:	linux-security-module@vger.kernel.org
 T:	git kernel.org:/pub/scm/linux/kernel/git/chrisw/lsm-2.6.git
 S:	Supported
 
+LLC (802.2)
+P:	Arnaldo Carvalho de Melo
+M:	acme@ghostprotocols.net
+S:	Maintained
+
 LM83 HARDWARE MONITOR DRIVER
 P:	Jean Delvare
 M:	khali@linux-fr.org
@@ -2718,14 +2726,6 @@ M:	kaber@trash.net
 L:	netdev@vger.kernel.org
 S:	Maintained
 
-MARVELL YUKON / SYSKONNECT DRIVER
-P:	Mirko Lindner
-M:	mlindner@syskonnect.de
-P:	Ralph Roesler
-M:	rroesler@syskonnect.de
-W:	http://www.syskonnect.com
-S:	Supported
-
 MAN-PAGES: MANUAL PAGES FOR LINUX -- Sections 2, 3, 4, 5, and 7
 P:	Michael Kerrisk
 M:	mtk.manpages@gmail.com
@@ -2745,6 +2745,14 @@ M:	buytenh@marvell.com
 L:	netdev@vger.kernel.org
 S:	Supported
 
+MARVELL YUKON / SYSKONNECT DRIVER
+P:	Mirko Lindner
+M:	mlindner@syskonnect.de
+P:	Ralph Roesler
+M:	rroesler@syskonnect.de
+W:	http://www.syskonnect.com
+S:	Supported
+
 MATROX FRAMEBUFFER DRIVER
 P:	Petr Vandrovec
 M:	vandrove@vc.cvut.cz
@@ -2781,15 +2789,6 @@ L:	linux-mm@kvack.org
 L:	linux-kernel@vger.kernel.org
 S:	Maintained
 
-MEI MN10300/AM33 PORT
-P:	David Howells
-M:	dhowells@redhat.com
-P:	Koichi Yasutake
-M:	yasutake.koichi@jp.panasonic.com
-L:	linux-am33-list@redhat.com (moderated for non-subscribers)
-W:	ftp://ftp.redhat.com/pub/redhat/gnupro/AM33/
-S:	Maintained
-
 MEMORY TECHNOLOGY DEVICES (MTD)
 P:	David Woodhouse
 M:	dwmw2@infradead.org
@@ -2798,14 +2797,6 @@ L:	linux-mtd@lists.infradead.org
 T:	git git://git.infradead.org/mtd-2.6.git
 S:	Maintained
 
-UNSORTED BLOCK IMAGES (UBI)
-P:	Artem Bityutskiy
-M:	dedekind@infradead.org
-W:	http://www.linux-mtd.infradead.org/
-L:	linux-mtd@lists.infradead.org
-T:	git git://git.infradead.org/~dedekind/ubi-2.6.git
-S:	Maintained
-
 MICROTEK X6 SCANNER
 P:	Oliver Neukum
 M:	oliver@neukum.name
@@ -3037,17 +3028,6 @@ L:	netdev@vger.kernel.org
 W:	http://www.netxen.com
 S:	Supported
 
-IPVS
-P:	Wensong Zhang
-M:	wensong@linux-vs.org
-P:	Simon Horman
-M:	horms@verge.net.au
-P:	Julian Anastasov
-M:	ja@ssi.bg
-L:	netdev@vger.kernel.org
-L:	lvs-devel@vger.kernel.org
-S:	Maintained
-
 NFS, SUNRPC, AND LOCKD CLIENTS
 P:	Trond Myklebust
 M:	Trond.Myklebust@netapp.com
@@ -3093,22 +3073,6 @@ M:	adaplas@gmail.com
 L:	linux-fbdev-devel@lists.sourceforge.net (moderated for non-subscribers)
 S:	Maintained
 
-OPENCORES I2C BUS DRIVER
-P:	Peter Korsgaard
-M:	jacmet@sunsite.dk
-L:	i2c@lm-sensors.org
-S:	Maintained
-
-ORACLE CLUSTER FILESYSTEM 2 (OCFS2)
-P:	Mark Fasheh
-M:	mfasheh@suse.com
-P:	Joel Becker
-M:	joel.becker@oracle.com
-L:	ocfs2-devel@oss.oracle.com (moderated for non-subscribers)
-W:	http://oss.oracle.com/projects/ocfs2/
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/mfasheh/ocfs2.git
-S:	Supported
-
 OMFS FILESYSTEM
 P:	Bob Copeland
 M:	me@bobcopeland.com
@@ -3144,12 +3108,28 @@ L:	osst-users@lists.sourceforge.net
 L:	linux-scsi@vger.kernel.org
 S:	Maintained
 
+OPENCORES I2C BUS DRIVER
+P:	Peter Korsgaard
+M:	jacmet@sunsite.dk
+L:	i2c@lm-sensors.org
+S:	Maintained
+
 OPROFILE
 P:	Robert Richter
 M:	robert.richter@amd.com
 L:	oprofile-list@lists.sf.net
 S:	Maintained
 
+ORACLE CLUSTER FILESYSTEM 2 (OCFS2)
+P:	Mark Fasheh
+M:	mfasheh@suse.com
+P:	Joel Becker
+M:	joel.becker@oracle.com
+L:	ocfs2-devel@oss.oracle.com (moderated for non-subscribers)
+W:	http://oss.oracle.com/projects/ocfs2/
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/mfasheh/ocfs2.git
+S:	Supported
+
 ORINOCO DRIVER
 P:	Pavel Roskin
 M:	proski@gnu.org
@@ -3161,6 +3141,14 @@ L:	orinoco-devel@lists.sourceforge.net
 W:	http://www.nongnu.org/orinoco/
 S:	Maintained
 
+P54 WIRELESS DRIVER
+P:	Michael Wu
+M:	flamingice@sourmilk.net
+L:	linux-wireless@vger.kernel.org
+W:	http://prism54.org
+T:	git kernel.org:/pub/scm/linux/kernel/git/mwu/mac80211-drivers.git
+S:	Maintained
+
 PA SEMI ETHERNET DRIVER
 P:	Olof Johansson
 M:	olof@lixom.net
@@ -3173,10 +3161,32 @@ M:	olof@lixom.net
 L:	i2c@lm-sensors.org
 S:	Maintained
 
+PANASONIC MN10300/AM33 PORT
+P:	David Howells
+M:	dhowells@redhat.com
+P:	Koichi Yasutake
+M:	yasutake.koichi@jp.panasonic.com
+L:	linux-am33-list@redhat.com (moderated for non-subscribers)
+W:	ftp://ftp.redhat.com/pub/redhat/gnupro/AM33/
+S:	Maintained
+
 PARALLEL PORT SUPPORT
 L:	linux-parport@lists.infradead.org (subscribers-only)
 S:	Orphan
 
+PARAVIRT_OPS INTERFACE
+P:	Jeremy Fitzhardinge
+M:	jeremy@xensource.com
+P:	Chris Wright
+M:	chrisw@sous-sol.org
+P:	Zachary Amsden
+M:	zach@vmware.com
+P:	Rusty Russell
+M:	rusty@rustcorp.com.au
+L:	virtualization@lists.osdl.org
+L:	linux-kernel@vger.kernel.org
+S:	Supported
+
 PARIDE DRIVERS FOR PARALLEL PORT IDE DEVICES
 P:	Tim Waugh
 M:	tim@cyberelk.net
@@ -3196,19 +3206,6 @@ W:	http://www.parisc-linux.org/
 T:	git kernel.org:/pub/scm/linux/kernel/git/kyle/parisc-2.6.git
 S:	Maintained
 
-PARAVIRT_OPS INTERFACE
-P:	Jeremy Fitzhardinge
-M:	jeremy@xensource.com
-P:	Chris Wright
-M:	chrisw@sous-sol.org
-P:	Zachary Amsden
-M:	zach@vmware.com
-P:	Rusty Russell
-M:	rusty@rustcorp.com.au
-L:	virtualization@lists.osdl.org
-L:	linux-kernel@vger.kernel.org
-S:	Supported
-
 PC87360 HARDWARE MONITORING DRIVER
 P:	Jim Cromie
 M:	jim.cromie@gmail.com
@@ -3338,14 +3335,6 @@ L:	kpreempt-tech@lists.sourceforge.net
 W:	ftp://ftp.kernel.org/pub/linux/kernel/people/rml/preempt-kernel
 S:	Supported
 
-P54 WIRELESS DRIVER
-P:	Michael Wu
-M:	flamingice@sourmilk.net
-L:	linux-wireless@vger.kernel.org
-W:	http://prism54.org
-T:	git kernel.org:/pub/scm/linux/kernel/git/mwu/mac80211-drivers.git
-S:	Maintained
-
 PRISM54 WIRELESS DRIVER
 P:	Luis R. Rodriguez
 M:	mcgrof@gmail.com
@@ -3436,13 +3425,7 @@ M:	paulus@samba.org
 L:	linux-fbdev-devel@lists.sourceforge.net (moderated for non-subscribers)
 S:	Maintained
 
-RAYLINK/WEBGEAR 802.11 WIRELESS LAN DRIVER
-P:	Corey Thomas
-M:	coreythomas@charter.net
-L:	linux-wireless@vger.kernel.org
-S:	Maintained
-
-RALINK RT2X00 WLAN DRIVER
+RALINK RT2X00 WIRELESS LAN DRIVER
 P:	rt2x00 project
 L:	linux-wireless@vger.kernel.org
 L:	rt2400-devel@lists.sourceforge.net
@@ -3467,6 +3450,18 @@ M:	mporter@kernel.crashing.org
 L:	linux-kernel@vger.kernel.org
 S:	Maintained
 
+RAYLINK/WEBGEAR 802.11 WIRELESS LAN DRIVER
+P:	Corey Thomas
+M:	coreythomas@charter.net
+L:	linux-wireless@vger.kernel.org
+S:	Maintained
+
+RCUTORTURE MODULE
+P:	Josh Triplett
+M:	josh@freedesktop.org
+L:	linux-kernel@vger.kernel.org
+S:	Maintained
+
 RDC R-321X SoC
 P:	Florian Fainelli
 M:	florian.fainelli@telecomint.eu
@@ -3486,12 +3481,6 @@ W:	http://www.rdrop.com/users/paulmck/rclock/
 L:	linux-kernel@vger.kernel.org
 S:	Supported
 
-RCUTORTURE MODULE
-P:	Josh Triplett
-M:	josh@freedesktop.org
-L:	linux-kernel@vger.kernel.org
-S:	Maintained
-
 REAL TIME CLOCK DRIVER
 P:	Paul Gortmaker
 M:	p_gortmaker@yahoo.com
@@ -3515,6 +3504,9 @@ L:	netdev@vger.kernel.org
 S:	Maintained
 F:	net/rfkill
 
+RISCOM8 DRIVER
+S:	Orphan
+
 ROCKETPORT DRIVER
 P:	Comtrol Corp.
 W:	http://www.comtrol.com
@@ -3527,9 +3519,6 @@ L:	linux-hams@vger.kernel.org
 W:	http://www.linux-ax25.org/
 S:	Maintained
 
-RISCOM8 DRIVER
-S:	Orphan
-
 RTL818X WIRELESS DRIVER
 P:	Michael Wu
 M:	flamingice@sourmilk.net
@@ -3669,6 +3658,12 @@ M:	saschasommer@freenet.de
 L:	sdricohcs-devel@lists.sourceforge.net (subscribers-only)
 S:	Maintained
 
+SECURE DIGITAL HOST CONTROLLER INTERFACE (SDHCI) DRIVER
+P:	Pierre Ossman
+M:	drzeus-sdhci@drzeus.cx
+L:	sdhci-devel@list.drzeus.cx
+S:	Maintained
+
 SECURITY CONTACT
 P:	Security Officers
 M:	security@kernel.org
@@ -3692,19 +3687,13 @@ P:	Jiri Slaby
 M:	jirislaby@gmail.com
 S:	Maintained
 
-SERIAL ATA (SATA) SUBSYSTEM:
+SERIAL ATA (SATA) SUBSYSTEM
 P:	Jeff Garzik
 M:	jgarzik@pobox.com
 L:	linux-ide@vger.kernel.org
 T:	git kernel.org:/pub/scm/linux/kernel/git/jgarzik/libata-dev.git
 S:	Supported
 
-SGI SN-IA64 (Altix) SERIAL CONSOLE DRIVER
-P:	Pat Gefre
-M:	pfg@sgi.com
-L:	linux-ia64@vger.kernel.org
-S:	Supported
-
 SFC NETWORK DRIVER
 P:	Steve Hodgson
 P:	Ben Hutchings
@@ -3712,6 +3701,17 @@ P:	Robert Stonehouse
 M:	linux-net-drivers@solarflare.com
 S:	Supported
 
+SGI GRU DRIVER
+P:	Jack Steiner
+M:	steiner@sgi.com
+S:	Maintained
+
+SGI SN-IA64 (Altix) SERIAL CONSOLE DRIVER
+P:	Pat Gefre
+M:	pfg@sgi.com
+L:	linux-ia64@vger.kernel.org
+S:	Supported
+
 SGI VISUAL WORKSTATION 320 AND 540
 P:	Andrey Panin
 M:	pazke@donpac.ru
@@ -3719,16 +3719,24 @@ L:	linux-visws-devel@lists.sf.net
 W:	http://linux-visws.sf.net
 S:	Maintained for 2.6.
 
-SGI GRU DRIVER
-P:	Jack Steiner
-M:	steiner@sgi.com
-S:	Maintained
-
 SGI XP/XPC/XPNET DRIVER
 P:	Dean Nelson
 M:	dcn@sgi.com
 S:	Maintained
 
+SHARP LH SUPPORT (LH7952X & LH7A40X)
+P:	Marc Singer
+M:	elf@buici.com
+W:	http://projects.buici.com/arm
+L:	linux-arm-kernel@lists.arm.linux.org.uk	(subscribers-only)
+S:	Maintained
+
+SHPC HOTPLUG DRIVER
+P:	Kristen Carlson Accardi
+M:	kristen.c.accardi@intel.com
+L:	linux-pci@vger.kernel.org
+S:	Supported
+
 SIMTEC EB110ATX (Chalice CATS)
 P:	Ben Dooks
 P:	Vincent Sanders
@@ -3774,6 +3782,12 @@ M:	thomas@winischhofer.net
 W:	http://www.winischhofer.at/linuxsisusbvga.shtml
 S:	Maintained
 
+SKGE, SKY2 10/100/1000 GIGABIT ETHERNET DRIVERS
+P:	Stephen Hemminger
+M:	shemminger@linux-foundation.org
+L:	netdev@vger.kernel.org
+S:	Maintained
+
 SLAB ALLOCATOR
 P:	Christoph Lameter
 M:	cl@linux-foundation.org
@@ -3800,12 +3814,25 @@ P:	Ben Nizette
 M:	bn@niasdigital.com
 S:	Maintained
 
+SN-IA64 (Itanium) SUB-PLATFORM
+P:	Jes Sorensen
+M:	jes@sgi.com
+L:	linux-altix@sgi.com
+L:	linux-ia64@vger.kernel.org
+W:	http://www.sgi.com/altix
+S:	Maintained
+
 SOC-CAMERA V4L2 SUBSYSTEM
 P:	Guennadi Liakhovetski
 M:	g.liakhovetski@gmx.de
 L:	video4linux-list@redhat.com
 S:	Maintained
 
+SOEKRIS NET48XX LED SUPPORT
+P:	Chris Boot
+M:	bootc@bootc.net
+S:	Maintained
+
 SOFTWARE RAID (Multiple Disks) SUPPORT
 P:	Ingo Molnar
 M:	mingo@redhat.com
@@ -3814,24 +3841,6 @@ M:	neilb@suse.de
 L:	linux-raid@vger.kernel.org
 S:	Supported
 
-HIBERNATION (aka Software Suspend, aka swsusp):
-P:	Pavel Machek
-M:	pavel@suse.cz
-P:	Rafael J. Wysocki
-M:	rjw@sisk.pl
-L:	linux-pm@lists.linux-foundation.org
-S:	Supported
-
-SUSPEND TO RAM:
-P:	Len Brown
-M:	len.brown@intel.com
-P:	Pavel Machek
-M:	pavel@suse.cz
-P:	Rafael J. Wysocki
-M:	rjw@sisk.pl
-L:	linux-pm@lists.linux-foundation.org
-S:	Supported
-
 SONIC NETWORK DRIVER
 P:	Thomas Bogendoerfer
 M:	tsbogend@alpha.franken.de
@@ -3876,59 +3885,7 @@ L:	alsa-devel@alsa-project.org (subscribers-only)
 W:	http://alsa-project.org/main/index.php/ASoC
 S:	Supported
 
-SPI SUBSYSTEM
-P:	David Brownell
-M:	dbrownell@users.sourceforge.net
-L:	spi-devel-general@lists.sourceforge.net
-S:	Maintained
-
-SPU FILE SYSTEM
-P:	Jeremy Kerr
-M:	jk@ozlabs.org
-L:	linuxppc-dev@ozlabs.org
-L:	cbe-oss-dev@ozlabs.org
-W:	http://www.ibm.com/developerworks/power/cell/
-S:	Supported
-
-STABLE BRANCH:
-P:	Greg Kroah-Hartman
-M:	greg@kroah.com
-P:	Chris Wright
-M:	chrisw@sous-sol.org
-L:	stable@kernel.org
-S:	Maintained
-
-SHARP LH SUPPORT (LH7952X & LH7A40X)
-P:	Marc Singer
-M:	elf@buici.com
-W:	http://projects.buici.com/arm
-L:	linux-arm-kernel@lists.arm.linux.org.uk	(subscribers-only)
-S:	Maintained
-
-SHPC HOTPLUG DRIVER
-P:	Kristen Carlson Accardi
-M:	kristen.c.accardi@intel.com
-L:	linux-pci@vger.kernel.org
-S:	Supported
-
-SECURE DIGITAL HOST CONTROLLER INTERFACE DRIVER
-P:	Pierre Ossman
-M:	drzeus-sdhci@drzeus.cx
-L:	sdhci-devel@list.drzeus.cx
-S:	Maintained
-
-SKGE, SKY2 10/100/1000 GIGABIT ETHERNET DRIVERS
-P:	Stephen Hemminger
-M:	shemminger@linux-foundation.org
-L:	netdev@vger.kernel.org
-S:	Maintained
-
-SOEKRIS NET48XX LED SUPPORT
-P:	Chris Boot
-M:	bootc@bootc.net
-S:	Maintained
-
-SPARC (sparc32):
+SPARC (sparc32)
 P:	William L. Irwin
 M:	wli@holomorphy.com
 L:	sparclinux@vger.kernel.org
@@ -3940,6 +3897,12 @@ M:	R.E.Wolff@BitWizard.nl
 L:	linux-kernel@vger.kernel.org ?
 S:	Supported
 
+SPI SUBSYSTEM
+P:	David Brownell
+M:	dbrownell@users.sourceforge.net
+L:	spi-devel-general@lists.sourceforge.net
+S:	Maintained
+
 SPIDERNET NETWORK DRIVER for CELL
 P:	Ishizaki Kou
 M:	kou.ishizaki@toshiba.co.jp
@@ -3948,12 +3911,28 @@ M:	jens@de.ibm.com
 L:	netdev@vger.kernel.org
 S:	Supported
 
+SPU FILE SYSTEM
+P:	Jeremy Kerr
+M:	jk@ozlabs.org
+L:	linuxppc-dev@ozlabs.org
+L:	cbe-oss-dev@ozlabs.org
+W:	http://www.ibm.com/developerworks/power/cell/
+S:	Supported
+
 SRM (Alpha) environment access
 P:	Jan-Benedict Glaw
 M:	jbglaw@lug-owl.de
 L:	linux-kernel@vger.kernel.org
 S:	Maintained
 
+STABLE BRANCH:
+P:	Greg Kroah-Hartman
+M:	greg@kroah.com
+P:	Chris Wright
+M:	chrisw@sous-sol.org
+L:	stable@kernel.org
+S:	Maintained
+
 STARFIRE/DURALAN NETWORK DRIVER
 P:	Ion Badulescu
 M:	ionut@cs.columbia.edu
@@ -3969,6 +3948,12 @@ M:	laredo@gnu.org
 W:	http://www.stradis.com/
 S:	Maintained
 
+SUN3/3X
+P:	Sam Creasey
+M:	sammy@sammy.net
+W:	http://sammy.net/sun3/
+S:	Maintained
+
 SUPERH
 P:	Paul Mundt
 M:	lethal@linux-sh.org
@@ -3977,11 +3962,15 @@ W:	http://www.linux-sh.org
 T:	git kernel.org:/pub/scm/linux/kernel/git/lethal/sh-2.6.git
 S:	Supported
 
-SUN3/3X
-P:	Sam Creasey
-M:	sammy@sammy.net
-W:	http://sammy.net/sun3/
-S:	Maintained
+SUSPEND TO RAM
+P:	Len Brown
+M:	len.brown@intel.com
+P:	Pavel Machek
+M:	pavel@suse.cz
+P:	Rafael J. Wysocki
+M:	rjw@sisk.pl
+L:	linux-pm@lists.linux-foundation.org
+S:	Supported
 
 SVGA HANDLING
 P:	Martin Mares
@@ -4027,7 +4016,7 @@ P:	Mark Gross
 M:	mark.gross@intel.com
 S:	Supported
 
-TENSILICA XTENSA PORT (xtensa):
+TENSILICA XTENSA PORT (xtensa)
 P:	Chris Zankel
 M:	chris@zankel.net
 S:	Maintained
@@ -4141,6 +4130,16 @@ M:	ballabio_dario@emc.com
 L:	linux-scsi@vger.kernel.org
 S:	Maintained
 
+UBI FILE SYSTEM (UBIFS)
+P:	Artem Bityutskiy
+M:	dedekind@infradead.org
+P:	Adrian Hunter
+M:	ext-adrian.hunter@nokia.com
+L:	linux-mtd@lists.infradead.org
+T:	git git://git.infradead.org/~dedekind/ubifs-2.6.git
+W:	http://www.linux-mtd.infradead.org/doc/ubifs.html
+S:	Maintained
+
 UCLINUX (AND M68KNOMMU)
 P:	Greg Ungerer
 M:	gerg@uclinux.org
@@ -4166,7 +4165,7 @@ M:	dushistov@mail.ru
 L:	linux-kernel@vger.kernel.org
 S:	Maintained
 
-UltraSPARC (sparc64):
+UltraSPARC (sparc64)
 P:	David S. Miller
 M:	davem@davemloft.net
 L:	sparclinux@vger.kernel.org
@@ -4180,6 +4179,14 @@ L:	linux-kernel@vger.kernel.org
 W:	http://www.kernel.dk
 S:	Maintained
 
+UNSORTED BLOCK IMAGES (UBI)
+P:	Artem Bityutskiy
+M:	dedekind@infradead.org
+W:	http://www.linux-mtd.infradead.org/
+L:	linux-mtd@lists.infradead.org
+T:	git git://git.infradead.org/~dedekind/ubi-2.6.git
+S:	Maintained
+
 USB ACM DRIVER
 P:	Oliver Neukum
 M:	oliver@neukum.name
@@ -4449,7 +4456,7 @@ L:	video4linux-list@redhat.com
 W:	http://royale.zerezo.com/zr364xx/
 S:	Maintained
 
-USER-MODE LINUX
+USER-MODE LINUX (UML)
 P:	Jeff Dike
 M:	jdike@addtoit.com
 L:	user-mode-linux-devel@lists.sourceforge.net
@@ -4473,7 +4480,7 @@ W:	http://kernel.org/~kzak/util-linux-ng/
 T:	git://git.kernel.org/pub/scm/utils/util-linux-ng/util-linux-ng.git
 S:	Maintained
 
-VFAT/FAT/MSDOS FILESYSTEM:
+VFAT/FAT/MSDOS FILESYSTEM
 P:	OGAWA Hirofumi
 M:	hirofumi@mail.parknet.co.jp
 L:	linux-kernel@vger.kernel.org
@@ -4496,7 +4503,7 @@ M:	romieu@fr.zoreil.com
 L:	netdev@vger.kernel.org
 S:	Maintained
 
-VIDEO FOR LINUX
+VIDEO FOR LINUX (V4L)
 P:	Mauro Carvalho Chehab
 M:	mchehab@infradead.org
 M:	v4l-dvb-maintainer@linuxtv.org
@@ -4602,6 +4609,17 @@ M:	eis@baty.hanse.de
 L:	linux-x25@vger.kernel.org
 S:	Maintained
 
+X86 ARCHITECTURE (32-BIT AND 64-BIT)
+P:	Thomas Gleixner
+M:	tglx@linutronix.de
+P:	Ingo Molnar
+M:	mingo@redhat.com
+P:	H. Peter Anvin
+M:	hpa@zytor.com
+L:	linux-kernel@vger.kernel.org
+T:	git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86.git
+S:	Maintained
+
 XEN HYPERVISOR INTERFACE
 P:	Jeremy Fitzhardinge
 M:	jeremy@xensource.com
@@ -4633,17 +4651,6 @@ M:	jacmet@sunsite.dk
 L:	linux-serial@vger.kernel.org
 S:	Maintained
 
-X86 ARCHITECTURE (32-BIT AND 64-BIT)
-P:	Thomas Gleixner
-M:	tglx@linutronix.de
-P:	Ingo Molnar
-M:	mingo@redhat.com
-P:	H. Peter Anvin
-M:	hpa@zytor.com
-L:	linux-kernel@vger.kernel.org
-T:	git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86.git
-S:	Maintained
-
 YAM DRIVER FOR AX.25
 P:	Jean-Paul Roubelat
 M:	jpr@f6fbb.org
diff --git a/arch/Kconfig b/arch/Kconfig
index 364c6dadde0a..0267babe5eb9 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -13,6 +13,20 @@ config OPROFILE
 
 	  If unsure, say N.
 
+config OPROFILE_IBS
+	bool "OProfile AMD IBS support (EXPERIMENTAL)"
+	default n
+	depends on OPROFILE && SMP && X86
+	help
+          Instruction-Based Sampling (IBS) is a new profiling
+          technique that provides rich, precise program performance
+          information. IBS is introduced by AMD Family10h processors
+          (AMD Opteron Quad-Core processor “Barcelona”) to overcome
+          the limitations of conventional performance counter
+          sampling.
+
+	  If unsure, say N.
+
 config HAVE_OPROFILE
 	def_bool n
 
diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index 1bec55d63ef6..ee35226c44e9 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -5,6 +5,7 @@
 config ALPHA
 	bool
 	default y
+	select HAVE_AOUT
 	select HAVE_IDE
 	select HAVE_OPROFILE
 	help
@@ -68,9 +69,6 @@ config AUTO_IRQ_AFFINITY
 	depends on SMP
 	default y
 
-config ARCH_SUPPORTS_AOUT
-	def_bool y
-
 source "init/Kconfig"
 
 
diff --git a/arch/alpha/include/asm/statfs.h b/arch/alpha/include/asm/statfs.h
index ad15830baefe..de35cd438a10 100644
--- a/arch/alpha/include/asm/statfs.h
+++ b/arch/alpha/include/asm/statfs.h
@@ -1,6 +1,10 @@
 #ifndef _ALPHA_STATFS_H
 #define _ALPHA_STATFS_H
 
+/* Alpha is the only 64-bit platform with 32-bit statfs. And doesn't
+   even seem to implement statfs64 */
+#define __statfs_word __u32
+
 #include <asm-generic/statfs.h>
 
 #endif
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index efeed65b4a66..4853f9df37bd 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -8,6 +8,7 @@ mainmenu "Linux Kernel Configuration"
 config ARM
 	bool
 	default y
+	select HAVE_AOUT
 	select HAVE_IDE
 	select RTC_LIB
 	select SYS_SUPPORTS_APM_EMULATION
@@ -140,9 +141,6 @@ config GENERIC_CALIBRATE_DELAY
 	bool
 	default y
 
-config ARCH_SUPPORTS_AOUT
-	def_bool y
-
 config ARCH_MAY_HAVE_PC_FDC
 	bool
 
diff --git a/arch/arm/include/asm/statfs.h b/arch/arm/include/asm/statfs.h
index a02e6a8c3d70..079447c05ba7 100644
--- a/arch/arm/include/asm/statfs.h
+++ b/arch/arm/include/asm/statfs.h
@@ -1,42 +1,12 @@
 #ifndef _ASMARM_STATFS_H
 #define _ASMARM_STATFS_H
 
-#ifndef __KERNEL_STRICT_NAMES
-# include <linux/types.h>
-typedef __kernel_fsid_t	fsid_t;
-#endif
-
-struct statfs {
-	__u32 f_type;
-	__u32 f_bsize;
-	__u32 f_blocks;
-	__u32 f_bfree;
-	__u32 f_bavail;
-	__u32 f_files;
-	__u32 f_ffree;
-	__kernel_fsid_t f_fsid;
-	__u32 f_namelen;
-	__u32 f_frsize;
-	__u32 f_spare[5];
-};
-
 /*
  * With EABI there is 4 bytes of padding added to this structure.
  * Let's pack it so the padding goes away to simplify dual ABI support.
  * Note that user space does NOT have to pack this structure.
  */
-struct statfs64 {
-	__u32 f_type;
-	__u32 f_bsize;
-	__u64 f_blocks;
-	__u64 f_bfree;
-	__u64 f_bavail;
-	__u64 f_files;
-	__u64 f_ffree;
-	__kernel_fsid_t f_fsid;
-	__u32 f_namelen;
-	__u32 f_frsize;
-	__u32 f_spare[5];
-} __attribute__ ((packed,aligned(4)));
+#define ARCH_PACK_STATFS64 __attribute__((packed,aligned(4)))
 
+#include <asm-generic/statfs.h>
 #endif
diff --git a/arch/arm/kernel/dma.c b/arch/arm/kernel/dma.c
index ba99a2035523..d006085ed7e7 100644
--- a/arch/arm/kernel/dma.c
+++ b/arch/arm/kernel/dma.c
@@ -26,23 +26,6 @@ EXPORT_SYMBOL(dma_spin_lock);
 static dma_t dma_chan[MAX_DMA_CHANNELS];
 
 /*
- * Get dma list for /proc/dma
- */
-int get_dma_list(char *buf)
-{
-	dma_t *dma;
-	char *p = buf;
-	int i;
-
-	for (i = 0, dma = dma_chan; i < MAX_DMA_CHANNELS; i++, dma++)
-		if (dma->lock)
-			p += sprintf(p, "%2d: %14s %s\n", i,
-				     dma->d_ops->type, dma->device_id);
-
-	return p - buf;
-}
-
-/*
  * Request DMA channel
  *
  * On certain platforms, we have to allocate an interrupt as well...
diff --git a/arch/arm/mach-orion5x/include/mach/orion5x.h b/arch/arm/mach-orion5x/include/mach/orion5x.h
index e67c843baa02..9f5ce1ce5840 100644
--- a/arch/arm/mach-orion5x/include/mach/orion5x.h
+++ b/arch/arm/mach-orion5x/include/mach/orion5x.h
@@ -157,9 +157,11 @@
 #define CPU_CONF		ORION5X_BRIDGE_REG(0x100)
 #define CPU_CTRL		ORION5X_BRIDGE_REG(0x104)
 #define CPU_RESET_MASK		ORION5X_BRIDGE_REG(0x108)
+#define  WDT_RESET		0x0002
 #define CPU_SOFT_RESET		ORION5X_BRIDGE_REG(0x10c)
 #define POWER_MNG_CTRL_REG	ORION5X_BRIDGE_REG(0x11C)
 #define BRIDGE_CAUSE		ORION5X_BRIDGE_REG(0x110)
+#define  WDT_INT_REQ		0x0008
 #define BRIDGE_MASK		ORION5X_BRIDGE_REG(0x114)
 #define  BRIDGE_INT_TIMER0	0x0002
 #define  BRIDGE_INT_TIMER1	0x0004
diff --git a/arch/arm/plat-omap/devices.c b/arch/arm/plat-omap/devices.c
index a716ecd1db27..97187fa0ae52 100644
--- a/arch/arm/plat-omap/devices.c
+++ b/arch/arm/plat-omap/devices.c
@@ -441,16 +441,8 @@ static inline void omap_init_uwire(void) {}
 
 #if	defined(CONFIG_OMAP_WATCHDOG) || defined(CONFIG_OMAP_WATCHDOG_MODULE)
 
-#ifdef CONFIG_ARCH_OMAP24XX
-#define	OMAP_WDT_BASE		0x48022000
-#else
-#define	OMAP_WDT_BASE		0xfffeb000
-#endif
-
 static struct resource wdt_resources[] = {
 	{
-		.start		= OMAP_WDT_BASE,
-		.end		= OMAP_WDT_BASE + 0x4f,
 		.flags		= IORESOURCE_MEM,
 	},
 };
@@ -464,6 +456,19 @@ static struct platform_device omap_wdt_device = {
 
 static void omap_init_wdt(void)
 {
+	if (cpu_is_omap16xx())
+		wdt_resources[0].start = 0xfffeb000;
+	else if (cpu_is_omap2420())
+		wdt_resources[0].start = 0x48022000; /* WDT2 */
+	else if (cpu_is_omap2430())
+		wdt_resources[0].start = 0x49016000; /* WDT2 */
+	else if (cpu_is_omap343x())
+		wdt_resources[0].start = 0x48314000; /* WDT2 */
+	else
+		return;
+
+	wdt_resources[0].end = wdt_resources[0].start + 0x4f;
+
 	(void) platform_device_register(&omap_wdt_device);
 }
 #else
diff --git a/arch/avr32/boards/atngw100/setup.c b/arch/avr32/boards/atngw100/setup.c
index b8286f1ce854..6c54580a66df 100644
--- a/arch/avr32/boards/atngw100/setup.c
+++ b/arch/avr32/boards/atngw100/setup.c
@@ -9,6 +9,7 @@
  */
 #include <linux/clk.h>
 #include <linux/etherdevice.h>
+#include <linux/gpio.h>
 #include <linux/irq.h>
 #include <linux/i2c.h>
 #include <linux/i2c-gpio.h>
@@ -53,8 +54,11 @@ static struct spi_board_info spi0_board_info[] __initdata = {
 };
 
 static struct mci_platform_data __initdata mci0_data = {
-	.detect_pin	= GPIO_PIN_PC(25),
-	.wp_pin		= GPIO_PIN_PE(0),
+	.slot[0] = {
+		.bus_width	= 4,
+		.detect_pin	= GPIO_PIN_PC(25),
+		.wp_pin		= GPIO_PIN_PE(0),
+	},
 };
 
 /*
@@ -190,7 +194,7 @@ static int __init atngw100_init(void)
 	 * PB28/EXTINT3 doesn't; it should be SMBALERT# (for PMBus),
 	 * but it's not available off-board.
 	 */
-	at32_select_periph(GPIO_PIN_PB(28), 0, AT32_GPIOF_PULLUP);
+	at32_select_periph(GPIO_PIOB_BASE, 1 << 28, 0, AT32_GPIOF_PULLUP);
 	at32_select_gpio(i2c_gpio_data.sda_pin,
 		AT32_GPIOF_MULTIDRV | AT32_GPIOF_OUTPUT | AT32_GPIOF_HIGH);
 	at32_select_gpio(i2c_gpio_data.scl_pin,
@@ -204,6 +208,15 @@ postcore_initcall(atngw100_init);
 
 static int __init atngw100_arch_init(void)
 {
+	/* PB30 is the otherwise unused jumper on the mainboard, with an
+	 * external pullup; the jumper grounds it.  Use it however you
+	 * like, including letting U-Boot or Linux tweak boot sequences.
+	 */
+	at32_select_gpio(GPIO_PIN_PB(30), 0);
+	gpio_request(GPIO_PIN_PB(30), "j15");
+	gpio_direction_input(GPIO_PIN_PB(30));
+	gpio_export(GPIO_PIN_PB(30), false);
+
 	/* set_irq_type() after the arch_initcall for EIC has run, and
 	 * before the I2C subsystem could try using this IRQ.
 	 */
diff --git a/arch/avr32/boards/atstk1000/atstk1002.c b/arch/avr32/boards/atstk1000/atstk1002.c
index dfc3443e23aa..29e5b51a7fd2 100644
--- a/arch/avr32/boards/atstk1000/atstk1002.c
+++ b/arch/avr32/boards/atstk1000/atstk1002.c
@@ -232,7 +232,7 @@ static void __init atstk1002_setup_extdac(void)
 		goto err_set_clk;
 	}
 
-	at32_select_periph(GPIO_PIN_PA(30), GPIO_PERIPH_A, 0);
+	at32_select_periph(GPIO_PIOA_BASE, (1 << 30), GPIO_PERIPH_A, 0);
 	at73c213_data.dac_clk = gclk;
 
 err_set_clk:
@@ -264,16 +264,20 @@ void __init setup_board(void)
 
 #ifndef CONFIG_BOARD_ATSTK100X_SW2_CUSTOM
 
+static struct mci_platform_data __initdata mci0_data = {
+	.slot[0] = {
+		.bus_width	= 4,
+
 /* MMC card detect requires MACB0 *NOT* be used */
 #ifdef CONFIG_BOARD_ATSTK1002_SW6_CUSTOM
-static struct mci_platform_data __initdata mci0_data = {
-	.detect_pin	= GPIO_PIN_PC(14),	/* gpio30/sdcd */
-	.wp_pin		= GPIO_PIN_PC(15),	/* gpio31/sdwp */
-};
-#define MCI_PDATA	&mci0_data
+		.detect_pin	= GPIO_PIN_PC(14), /* gpio30/sdcd */
+		.wp_pin		= GPIO_PIN_PC(15), /* gpio31/sdwp */
 #else
-#define MCI_PDATA	NULL
+		.detect_pin	= -ENODEV,
+		.wp_pin		= -ENODEV,
 #endif	/* SW6 for sd{cd,wp} routing */
+	},
+};
 
 #endif	/* SW2 for MMC signal routing */
 
@@ -326,13 +330,14 @@ static int __init atstk1002_init(void)
 	at32_add_device_spi(1, spi1_board_info, ARRAY_SIZE(spi1_board_info));
 #endif
 #ifndef CONFIG_BOARD_ATSTK100X_SW2_CUSTOM
-	at32_add_device_mci(0, MCI_PDATA);
+	at32_add_device_mci(0, &mci0_data);
 #endif
 #ifdef CONFIG_BOARD_ATSTK1002_SW5_CUSTOM
 	set_hw_addr(at32_add_device_eth(1, &eth_data[1]));
 #else
 	at32_add_device_lcdc(0, &atstk1000_lcdc_data,
-			     fbmem_start, fbmem_size, 0);
+			     fbmem_start, fbmem_size,
+			     ATMEL_LCDC_PRI_24BIT | ATMEL_LCDC_PRI_CONTROL);
 #endif
 	at32_add_device_usba(0, NULL);
 #ifndef CONFIG_BOARD_ATSTK100X_SW3_CUSTOM
diff --git a/arch/avr32/boards/atstk1000/atstk1003.c b/arch/avr32/boards/atstk1000/atstk1003.c
index 0cf664174c17..be089d7f37eb 100644
--- a/arch/avr32/boards/atstk1000/atstk1003.c
+++ b/arch/avr32/boards/atstk1000/atstk1003.c
@@ -19,6 +19,7 @@
 #include <linux/spi/spi.h>
 
 #include <asm/setup.h>
+#include <asm/atmel-mci.h>
 
 #include <mach/at32ap700x.h>
 #include <mach/board.h>
@@ -66,6 +67,16 @@ static struct spi_board_info spi1_board_info[] __initdata = { {
 } };
 #endif
 
+#ifndef CONFIG_BOARD_ATSTK100X_SW2_CUSTOM
+static struct mci_platform_data __initdata mci0_data = {
+	.slot[0] = {
+		.bus_width	= 4,
+		.detect_pin	= -ENODEV,
+		.wp_pin		= -ENODEV,
+	},
+};
+#endif
+
 #ifdef CONFIG_BOARD_ATSTK1000_EXTDAC
 static void __init atstk1003_setup_extdac(void)
 {
@@ -84,7 +95,7 @@ static void __init atstk1003_setup_extdac(void)
 		goto err_set_clk;
 	}
 
-	at32_select_periph(GPIO_PIN_PA(30), GPIO_PERIPH_A, 0);
+	at32_select_periph(GPIO_PIOA_BASE, (1 << 30), GPIO_PERIPH_A, 0);
 	at73c213_data.dac_clk = gclk;
 
 err_set_clk:
@@ -154,7 +165,7 @@ static int __init atstk1003_init(void)
 	at32_add_device_spi(1, spi1_board_info, ARRAY_SIZE(spi1_board_info));
 #endif
 #ifndef CONFIG_BOARD_ATSTK100X_SW2_CUSTOM
-	at32_add_device_mci(0, NULL);
+	at32_add_device_mci(0, &mci0_data);
 #endif
 	at32_add_device_usba(0, NULL);
 #ifndef CONFIG_BOARD_ATSTK100X_SW3_CUSTOM
diff --git a/arch/avr32/boards/atstk1000/atstk1004.c b/arch/avr32/boards/atstk1000/atstk1004.c
index 50a5273e5916..248ef237c167 100644
--- a/arch/avr32/boards/atstk1000/atstk1004.c
+++ b/arch/avr32/boards/atstk1000/atstk1004.c
@@ -21,6 +21,7 @@
 #include <video/atmel_lcdc.h>
 
 #include <asm/setup.h>
+#include <asm/atmel-mci.h>
 
 #include <mach/at32ap700x.h>
 #include <mach/board.h>
@@ -71,6 +72,16 @@ static struct spi_board_info spi1_board_info[] __initdata = { {
 } };
 #endif
 
+#ifndef CONFIG_BOARD_ATSTK100X_SW2_CUSTOM
+static struct mci_platform_data __initdata mci0_data = {
+	.slot[0] = {
+		.bus_width	= 4,
+		.detect_pin	= -ENODEV,
+		.wp_pin		= -ENODEV,
+	},
+};
+#endif
+
 #ifdef CONFIG_BOARD_ATSTK1000_EXTDAC
 static void __init atstk1004_setup_extdac(void)
 {
@@ -89,7 +100,7 @@ static void __init atstk1004_setup_extdac(void)
 		goto err_set_clk;
 	}
 
-	at32_select_periph(GPIO_PIN_PA(30), GPIO_PERIPH_A, 0);
+	at32_select_periph(GPIO_PIOA_BASE, (1 << 30), GPIO_PERIPH_A, 0);
 	at73c213_data.dac_clk = gclk;
 
 err_set_clk:
@@ -137,10 +148,11 @@ static int __init atstk1004_init(void)
 	at32_add_device_spi(1, spi1_board_info, ARRAY_SIZE(spi1_board_info));
 #endif
 #ifndef CONFIG_BOARD_ATSTK100X_SW2_CUSTOM
-	at32_add_device_mci(0, NULL);
+	at32_add_device_mci(0, &mci0_data);
 #endif
 	at32_add_device_lcdc(0, &atstk1000_lcdc_data,
-			     fbmem_start, fbmem_size, 0);
+			     fbmem_start, fbmem_size,
+			     ATMEL_LCDC_PRI_24BIT | ATMEL_LCDC_PRI_CONTROL);
 	at32_add_device_usba(0, NULL);
 #ifndef CONFIG_BOARD_ATSTK100X_SW3_CUSTOM
 	at32_add_device_ssc(0, ATMEL_SSC_TX);
diff --git a/arch/avr32/include/asm/a.out.h b/arch/avr32/include/asm/a.out.h
deleted file mode 100644
index e46375a34a72..000000000000
--- a/arch/avr32/include/asm/a.out.h
+++ /dev/null
@@ -1,20 +0,0 @@
-#ifndef __ASM_AVR32_A_OUT_H
-#define __ASM_AVR32_A_OUT_H
-
-struct exec
-{
-  unsigned long a_info;		/* Use macros N_MAGIC, etc for access */
-  unsigned a_text;		/* length of text, in bytes */
-  unsigned a_data;		/* length of data, in bytes */
-  unsigned a_bss;		/* length of uninitialized data area for file, in bytes */
-  unsigned a_syms;		/* length of symbol table data in file, in bytes */
-  unsigned a_entry;		/* start address */
-  unsigned a_trsize;		/* length of relocation info for text, in bytes */
-  unsigned a_drsize;		/* length of relocation info for data, in bytes */
-};
-
-#define N_TRSIZE(a)	((a).a_trsize)
-#define N_DRSIZE(a)	((a).a_drsize)
-#define N_SYMSIZE(a)	((a).a_syms)
-
-#endif /* __ASM_AVR32_A_OUT_H */
diff --git a/arch/avr32/include/asm/atmel-mci.h b/arch/avr32/include/asm/atmel-mci.h
index c2ea6e1c9aa1..59f3fadd0b68 100644
--- a/arch/avr32/include/asm/atmel-mci.h
+++ b/arch/avr32/include/asm/atmel-mci.h
@@ -1,9 +1,39 @@
 #ifndef __ASM_AVR32_ATMEL_MCI_H
 #define __ASM_AVR32_ATMEL_MCI_H
 
-struct mci_platform_data {
+#define ATMEL_MCI_MAX_NR_SLOTS	2
+
+struct dma_slave;
+
+/**
+ * struct mci_slot_pdata - board-specific per-slot configuration
+ * @bus_width: Number of data lines wired up the slot
+ * @detect_pin: GPIO pin wired to the card detect switch
+ * @wp_pin: GPIO pin wired to the write protect sensor
+ *
+ * If a given slot is not present on the board, @bus_width should be
+ * set to 0. The other fields are ignored in this case.
+ *
+ * Any pins that aren't available should be set to a negative value.
+ *
+ * Note that support for multiple slots is experimental -- some cards
+ * might get upset if we don't get the clock management exactly right.
+ * But in most cases, it should work just fine.
+ */
+struct mci_slot_pdata {
+	unsigned int		bus_width;
 	int			detect_pin;
 	int			wp_pin;
 };
 
+/**
+ * struct mci_platform_data - board-specific MMC/SDcard configuration
+ * @dma_slave: DMA slave interface to use in data transfers, or NULL.
+ * @slot: Per-slot configuration data.
+ */
+struct mci_platform_data {
+	struct dma_slave	*dma_slave;
+	struct mci_slot_pdata	slot[ATMEL_MCI_MAX_NR_SLOTS];
+};
+
 #endif /* __ASM_AVR32_ATMEL_MCI_H */
diff --git a/arch/avr32/include/asm/byteorder.h b/arch/avr32/include/asm/byteorder.h
index d77b48ba7338..8e3af02076dd 100644
--- a/arch/avr32/include/asm/byteorder.h
+++ b/arch/avr32/include/asm/byteorder.h
@@ -7,6 +7,9 @@
 #include <asm/types.h>
 #include <linux/compiler.h>
 
+#define __BIG_ENDIAN
+#define __SWAB_64_THRU_32__
+
 #ifdef __CHECKER__
 extern unsigned long __builtin_bswap_32(unsigned long x);
 extern unsigned short __builtin_bswap_16(unsigned short x);
@@ -17,15 +20,18 @@ extern unsigned short __builtin_bswap_16(unsigned short x);
  * the result.
  */
 #if !(__GNUC__ == 4 && __GNUC_MINOR__ < 2)
-#define __arch__swab32(x) __builtin_bswap_32(x)
-#define __arch__swab16(x) __builtin_bswap_16(x)
-#endif
+static inline __attribute_const__ __u16 __arch_swab16(__u16 val)
+{
+	return __builtin_bswap_16(val);
+}
+#define __arch_swab16 __arch_swab16
 
-#if !defined(__STRICT_ANSI__) || defined(__KERNEL__)
-# define __BYTEORDER_HAS_U64__
-# define __SWAB_64_THRU_32__
+static inline __attribute_const__ __u32 __arch_swab32(__u32 val)
+{
+	return __builtin_bswap_32(val);
+}
+#define __arch_swab32 __arch_swab32
 #endif
 
-#include <linux/byteorder/big_endian.h>
-
+#include <linux/byteorder.h>
 #endif /* __ASM_AVR32_BYTEORDER_H */
diff --git a/arch/avr32/include/asm/io.h b/arch/avr32/include/asm/io.h
index a520f77ead96..22c97ef92201 100644
--- a/arch/avr32/include/asm/io.h
+++ b/arch/avr32/include/asm/io.h
@@ -160,6 +160,14 @@ BUILDIO_IOPORT(l, u32)
 #define readw_relaxed			readw
 #define readl_relaxed			readl
 
+#define readb_be			__raw_readb
+#define readw_be			__raw_readw
+#define readl_be			__raw_readl
+
+#define writeb_be			__raw_writeb
+#define writew_be			__raw_writew
+#define writel_be			__raw_writel
+
 #define __BUILD_MEMORY_STRING(bwl, type)				\
 static inline void writes##bwl(volatile void __iomem *addr,		\
 			       const void *data, unsigned int count)	\
diff --git a/arch/avr32/kernel/process.c b/arch/avr32/kernel/process.c
index 2c08ac992ac3..134d5302b6dd 100644
--- a/arch/avr32/kernel/process.c
+++ b/arch/avr32/kernel/process.c
@@ -9,6 +9,7 @@
 #include <linux/module.h>
 #include <linux/kallsyms.h>
 #include <linux/fs.h>
+#include <linux/pm.h>
 #include <linux/ptrace.h>
 #include <linux/reboot.h>
 #include <linux/tick.h>
@@ -20,7 +21,7 @@
 
 #include <mach/pm.h>
 
-void (*pm_power_off)(void) = NULL;
+void (*pm_power_off)(void);
 EXPORT_SYMBOL(pm_power_off);
 
 /*
diff --git a/arch/avr32/kernel/setup.c b/arch/avr32/kernel/setup.c
index d8e623c426c1..5c7083916c33 100644
--- a/arch/avr32/kernel/setup.c
+++ b/arch/avr32/kernel/setup.c
@@ -283,6 +283,25 @@ static int __init early_parse_fbmem(char *p)
 }
 early_param("fbmem", early_parse_fbmem);
 
+/*
+ * Pick out the memory size.  We look for mem=size@start,
+ * where start and size are "size[KkMmGg]"
+ */
+static int __init early_mem(char *p)
+{
+	resource_size_t size, start;
+
+	start = system_ram->start;
+	size  = memparse(p, &p);
+	if (*p == '@')
+		start = memparse(p + 1, &p);
+
+	system_ram->start = start;
+	system_ram->end = system_ram->start + size - 1;
+	return 0;
+}
+early_param("mem", early_mem);
+
 static int __init parse_tag_core(struct tag *tag)
 {
 	if (tag->hdr.size > 2) {
diff --git a/arch/avr32/mach-at32ap/at32ap700x.c b/arch/avr32/mach-at32ap/at32ap700x.c
index e01dbe4ebb40..813b6844cdf6 100644
--- a/arch/avr32/mach-at32ap/at32ap700x.c
+++ b/arch/avr32/mach-at32ap/at32ap700x.c
@@ -82,8 +82,9 @@ static struct platform_device _name##_id##_device = {		\
 	.num_resources	= ARRAY_SIZE(_name##_id##_resource),	\
 }
 
-#define select_peripheral(pin, periph, flags)			\
-	at32_select_periph(GPIO_PIN_##pin, GPIO_##periph, flags)
+#define select_peripheral(port, pin_mask, periph, flags)	\
+	at32_select_periph(GPIO_##port##_BASE, pin_mask,	\
+			   GPIO_##periph, flags)
 
 #define DEV_CLK(_name, devname, bus, _index)			\
 static struct clk devname##_##_name = {				\
@@ -871,6 +872,7 @@ static struct clk atmel_psif1_pclk = {
 struct platform_device *__init at32_add_device_psif(unsigned int id)
 {
 	struct platform_device *pdev;
+	u32 pin_mask;
 
 	if (!(id == 0 || id == 1))
 		return NULL;
@@ -881,20 +883,22 @@ struct platform_device *__init at32_add_device_psif(unsigned int id)
 
 	switch (id) {
 	case 0:
+		pin_mask  = (1 << 8) | (1 << 9); /* CLOCK & DATA */
+
 		if (platform_device_add_resources(pdev, atmel_psif0_resource,
 					ARRAY_SIZE(atmel_psif0_resource)))
 			goto err_add_resources;
 		atmel_psif0_pclk.dev = &pdev->dev;
-		select_peripheral(PA(8), PERIPH_A, 0); /* CLOCK */
-		select_peripheral(PA(9), PERIPH_A, 0); /* DATA  */
+		select_peripheral(PIOA, pin_mask, PERIPH_A, 0);
 		break;
 	case 1:
+		pin_mask  = (1 << 11) | (1 << 12); /* CLOCK & DATA */
+
 		if (platform_device_add_resources(pdev, atmel_psif1_resource,
 					ARRAY_SIZE(atmel_psif1_resource)))
 			goto err_add_resources;
 		atmel_psif1_pclk.dev = &pdev->dev;
-		select_peripheral(PB(11), PERIPH_A, 0); /* CLOCK */
-		select_peripheral(PB(12), PERIPH_A, 0); /* DATA  */
+		select_peripheral(PIOB, pin_mask, PERIPH_A, 0);
 		break;
 	default:
 		return NULL;
@@ -958,26 +962,30 @@ DEV_CLK(usart, atmel_usart3, pba, 6);
 
 static inline void configure_usart0_pins(void)
 {
-	select_peripheral(PA(8),  PERIPH_B, 0);	/* RXD	*/
-	select_peripheral(PA(9),  PERIPH_B, 0);	/* TXD	*/
+	u32 pin_mask = (1 << 8) | (1 << 9); /* RXD & TXD */
+
+	select_peripheral(PIOA, pin_mask, PERIPH_B, 0);
 }
 
 static inline void configure_usart1_pins(void)
 {
-	select_peripheral(PA(17), PERIPH_A, 0);	/* RXD	*/
-	select_peripheral(PA(18), PERIPH_A, 0);	/* TXD	*/
+	u32 pin_mask = (1 << 17) | (1 << 18); /* RXD & TXD */
+
+	select_peripheral(PIOA, pin_mask, PERIPH_A, 0);
 }
 
 static inline void configure_usart2_pins(void)
 {
-	select_peripheral(PB(26), PERIPH_B, 0);	/* RXD	*/
-	select_peripheral(PB(27), PERIPH_B, 0);	/* TXD	*/
+	u32 pin_mask = (1 << 26) | (1 << 27); /* RXD & TXD */
+
+	select_peripheral(PIOB, pin_mask, PERIPH_B, 0);
 }
 
 static inline void configure_usart3_pins(void)
 {
-	select_peripheral(PB(18), PERIPH_B, 0);	/* RXD	*/
-	select_peripheral(PB(17), PERIPH_B, 0);	/* TXD	*/
+	u32 pin_mask = (1 << 18) | (1 << 17); /* RXD & TXD */
+
+	select_peripheral(PIOB, pin_mask, PERIPH_B, 0);
 }
 
 static struct platform_device *__initdata at32_usarts[4];
@@ -1057,59 +1065,69 @@ struct platform_device *__init
 at32_add_device_eth(unsigned int id, struct eth_platform_data *data)
 {
 	struct platform_device *pdev;
+	u32 pin_mask;
 
 	switch (id) {
 	case 0:
 		pdev = &macb0_device;
 
-		select_peripheral(PC(3),  PERIPH_A, 0);	/* TXD0	*/
-		select_peripheral(PC(4),  PERIPH_A, 0);	/* TXD1	*/
-		select_peripheral(PC(7),  PERIPH_A, 0);	/* TXEN	*/
-		select_peripheral(PC(8),  PERIPH_A, 0);	/* TXCK */
-		select_peripheral(PC(9),  PERIPH_A, 0);	/* RXD0	*/
-		select_peripheral(PC(10), PERIPH_A, 0);	/* RXD1	*/
-		select_peripheral(PC(13), PERIPH_A, 0);	/* RXER	*/
-		select_peripheral(PC(15), PERIPH_A, 0);	/* RXDV	*/
-		select_peripheral(PC(16), PERIPH_A, 0);	/* MDC	*/
-		select_peripheral(PC(17), PERIPH_A, 0);	/* MDIO	*/
+		pin_mask  = (1 << 3);	/* TXD0 */
+		pin_mask |= (1 << 4);	/* TXD1 */
+		pin_mask |= (1 << 7);	/* TXEN */
+		pin_mask |= (1 << 8);	/* TXCK */
+		pin_mask |= (1 << 9);	/* RXD0 */
+		pin_mask |= (1 << 10);	/* RXD1 */
+		pin_mask |= (1 << 13);	/* RXER */
+		pin_mask |= (1 << 15);	/* RXDV */
+		pin_mask |= (1 << 16);	/* MDC  */
+		pin_mask |= (1 << 17);	/* MDIO */
 
 		if (!data->is_rmii) {
-			select_peripheral(PC(0),  PERIPH_A, 0);	/* COL	*/
-			select_peripheral(PC(1),  PERIPH_A, 0);	/* CRS	*/
-			select_peripheral(PC(2),  PERIPH_A, 0);	/* TXER	*/
-			select_peripheral(PC(5),  PERIPH_A, 0);	/* TXD2	*/
-			select_peripheral(PC(6),  PERIPH_A, 0);	/* TXD3 */
-			select_peripheral(PC(11), PERIPH_A, 0);	/* RXD2	*/
-			select_peripheral(PC(12), PERIPH_A, 0);	/* RXD3	*/
-			select_peripheral(PC(14), PERIPH_A, 0);	/* RXCK	*/
-			select_peripheral(PC(18), PERIPH_A, 0);	/* SPD	*/
+			pin_mask |= (1 << 0);	/* COL  */
+			pin_mask |= (1 << 1);	/* CRS  */
+			pin_mask |= (1 << 2);	/* TXER */
+			pin_mask |= (1 << 5);	/* TXD2 */
+			pin_mask |= (1 << 6);	/* TXD3 */
+			pin_mask |= (1 << 11);	/* RXD2 */
+			pin_mask |= (1 << 12);	/* RXD3 */
+			pin_mask |= (1 << 14);	/* RXCK */
+			pin_mask |= (1 << 18);	/* SPD  */
 		}
+
+		select_peripheral(PIOC, pin_mask, PERIPH_A, 0);
+
 		break;
 
 	case 1:
 		pdev = &macb1_device;
 
-		select_peripheral(PD(13), PERIPH_B, 0);		/* TXD0	*/
-		select_peripheral(PD(14), PERIPH_B, 0);		/* TXD1	*/
-		select_peripheral(PD(11), PERIPH_B, 0);		/* TXEN	*/
-		select_peripheral(PD(12), PERIPH_B, 0);		/* TXCK */
-		select_peripheral(PD(10), PERIPH_B, 0);		/* RXD0	*/
-		select_peripheral(PD(6),  PERIPH_B, 0);		/* RXD1	*/
-		select_peripheral(PD(5),  PERIPH_B, 0);		/* RXER	*/
-		select_peripheral(PD(4),  PERIPH_B, 0);		/* RXDV	*/
-		select_peripheral(PD(3),  PERIPH_B, 0);		/* MDC	*/
-		select_peripheral(PD(2),  PERIPH_B, 0);		/* MDIO	*/
+		pin_mask  = (1 << 13);	/* TXD0 */
+		pin_mask |= (1 << 14);	/* TXD1 */
+		pin_mask |= (1 << 11);	/* TXEN */
+		pin_mask |= (1 << 12);	/* TXCK */
+		pin_mask |= (1 << 10);	/* RXD0 */
+		pin_mask |= (1 << 6);	/* RXD1 */
+		pin_mask |= (1 << 5);	/* RXER */
+		pin_mask |= (1 << 4);	/* RXDV */
+		pin_mask |= (1 << 3);	/* MDC  */
+		pin_mask |= (1 << 2);	/* MDIO */
+
+		if (!data->is_rmii)
+			pin_mask |= (1 << 15);	/* SPD  */
+
+		select_peripheral(PIOD, pin_mask, PERIPH_B, 0);
 
 		if (!data->is_rmii) {
-			select_peripheral(PC(19), PERIPH_B, 0);	/* COL	*/
-			select_peripheral(PC(23), PERIPH_B, 0);	/* CRS	*/
-			select_peripheral(PC(26), PERIPH_B, 0);	/* TXER	*/
-			select_peripheral(PC(27), PERIPH_B, 0);	/* TXD2	*/
-			select_peripheral(PC(28), PERIPH_B, 0);	/* TXD3 */
-			select_peripheral(PC(29), PERIPH_B, 0);	/* RXD2	*/
-			select_peripheral(PC(30), PERIPH_B, 0);	/* RXD3	*/
-			select_peripheral(PC(24), PERIPH_B, 0);	/* RXCK	*/
-			select_peripheral(PD(15), PERIPH_B, 0);	/* SPD	*/
+			pin_mask  = (1 << 19);	/* COL  */
+			pin_mask |= (1 << 23);	/* CRS  */
+			pin_mask |= (1 << 26);	/* TXER */
+			pin_mask |= (1 << 27);	/* TXD2 */
+			pin_mask |= (1 << 28);	/* TXD3 */
+			pin_mask |= (1 << 29);	/* RXD2 */
+			pin_mask |= (1 << 30);	/* RXD3 */
+			pin_mask |= (1 << 24);	/* RXCK */
+
+			select_peripheral(PIOC, pin_mask, PERIPH_B, 0);
 		}
 		break;
 
@@ -1177,23 +1195,28 @@ at32_add_device_spi(unsigned int id, struct spi_board_info *b, unsigned int n)
 		{ GPIO_PIN_PB(2), GPIO_PIN_PB(3),
 		  GPIO_PIN_PB(4), GPIO_PIN_PA(27), };
 	struct platform_device *pdev;
+	u32 pin_mask;
 
 	switch (id) {
 	case 0:
 		pdev = &atmel_spi0_device;
+		pin_mask  = (1 << 1) | (1 << 2);	/* MOSI & SCK */
+
 		/* pullup MISO so a level is always defined */
-		select_peripheral(PA(0),  PERIPH_A, AT32_GPIOF_PULLUP);
-		select_peripheral(PA(1),  PERIPH_A, 0);	/* MOSI	 */
-		select_peripheral(PA(2),  PERIPH_A, 0);	/* SCK	 */
+		select_peripheral(PIOA, (1 << 0), PERIPH_A, AT32_GPIOF_PULLUP);
+		select_peripheral(PIOA, pin_mask, PERIPH_A, 0);
+
 		at32_spi_setup_slaves(0, b, n, spi0_pins);
 		break;
 
 	case 1:
 		pdev = &atmel_spi1_device;
+		pin_mask  = (1 << 1) | (1 << 5);	/* MOSI */
+
 		/* pullup MISO so a level is always defined */
-		select_peripheral(PB(0),  PERIPH_B, AT32_GPIOF_PULLUP);
-		select_peripheral(PB(1),  PERIPH_B, 0);	/* MOSI  */
-		select_peripheral(PB(5),  PERIPH_B, 0);	/* SCK   */
+		select_peripheral(PIOB, (1 << 0), PERIPH_B, AT32_GPIOF_PULLUP);
+		select_peripheral(PIOB, pin_mask, PERIPH_B, 0);
+
 		at32_spi_setup_slaves(1, b, n, spi1_pins);
 		break;
 
@@ -1226,6 +1249,7 @@ struct platform_device *__init at32_add_device_twi(unsigned int id,
 						    unsigned int n)
 {
 	struct platform_device *pdev;
+	u32 pin_mask;
 
 	if (id != 0)
 		return NULL;
@@ -1238,8 +1262,9 @@ struct platform_device *__init at32_add_device_twi(unsigned int id,
 				ARRAY_SIZE(atmel_twi0_resource)))
 		goto err_add_resources;
 
-	select_peripheral(PA(6),  PERIPH_A, 0);	/* SDA	*/
-	select_peripheral(PA(7),  PERIPH_A, 0);	/* SDL	*/
+	pin_mask  = (1 << 6) | (1 << 7);	/* SDA & SDL */
+
+	select_peripheral(PIOA, pin_mask, PERIPH_A, 0);
 
 	atmel_twi0_pclk.dev = &pdev->dev;
 
@@ -1272,10 +1297,16 @@ static struct clk atmel_mci0_pclk = {
 struct platform_device *__init
 at32_add_device_mci(unsigned int id, struct mci_platform_data *data)
 {
-	struct mci_platform_data	_data;
 	struct platform_device		*pdev;
+	struct dw_dma_slave		*dws;
+	u32				pioa_mask;
+	u32				piob_mask;
 
-	if (id != 0)
+	if (id != 0 || !data)
+		return NULL;
+
+	/* Must have at least one usable slot */
+	if (!data->slot[0].bus_width && !data->slot[1].bus_width)
 		return NULL;
 
 	pdev = platform_device_alloc("atmel_mci", id);
@@ -1286,28 +1317,80 @@ at32_add_device_mci(unsigned int id, struct mci_platform_data *data)
 				ARRAY_SIZE(atmel_mci0_resource)))
 		goto fail;
 
-	if (!data) {
-		data = &_data;
-		memset(data, -1, sizeof(struct mci_platform_data));
-		data->detect_pin = GPIO_PIN_NONE;
-		data->wp_pin = GPIO_PIN_NONE;
-	}
+	if (data->dma_slave)
+		dws = kmemdup(to_dw_dma_slave(data->dma_slave),
+				sizeof(struct dw_dma_slave), GFP_KERNEL);
+	else
+		dws = kzalloc(sizeof(struct dw_dma_slave), GFP_KERNEL);
+
+	dws->slave.dev = &pdev->dev;
+	dws->slave.dma_dev = &dw_dmac0_device.dev;
+	dws->slave.reg_width = DMA_SLAVE_WIDTH_32BIT;
+	dws->cfg_hi = (DWC_CFGH_SRC_PER(0)
+				| DWC_CFGH_DST_PER(1));
+	dws->cfg_lo &= ~(DWC_CFGL_HS_DST_POL
+				| DWC_CFGL_HS_SRC_POL);
+
+	data->dma_slave = &dws->slave;
 
 	if (platform_device_add_data(pdev, data,
 				sizeof(struct mci_platform_data)))
 		goto fail;
 
-	select_peripheral(PA(10), PERIPH_A, 0);	/* CLK	 */
-	select_peripheral(PA(11), PERIPH_A, 0);	/* CMD	 */
-	select_peripheral(PA(12), PERIPH_A, 0);	/* DATA0 */
-	select_peripheral(PA(13), PERIPH_A, 0);	/* DATA1 */
-	select_peripheral(PA(14), PERIPH_A, 0);	/* DATA2 */
-	select_peripheral(PA(15), PERIPH_A, 0);	/* DATA3 */
+	/* CLK line is common to both slots */
+	pioa_mask = 1 << 10;
 
-	if (gpio_is_valid(data->detect_pin))
-		at32_select_gpio(data->detect_pin, 0);
-	if (gpio_is_valid(data->wp_pin))
-		at32_select_gpio(data->wp_pin, 0);
+	switch (data->slot[0].bus_width) {
+	case 4:
+		pioa_mask |= 1 << 13;		/* DATA1 */
+		pioa_mask |= 1 << 14;		/* DATA2 */
+		pioa_mask |= 1 << 15;		/* DATA3 */
+		/* fall through */
+	case 1:
+		pioa_mask |= 1 << 11;		/* CMD	 */
+		pioa_mask |= 1 << 12;		/* DATA0 */
+
+		if (gpio_is_valid(data->slot[0].detect_pin))
+			at32_select_gpio(data->slot[0].detect_pin, 0);
+		if (gpio_is_valid(data->slot[0].wp_pin))
+			at32_select_gpio(data->slot[0].wp_pin, 0);
+		break;
+	case 0:
+		/* Slot is unused */
+		break;
+	default:
+		goto fail;
+	}
+
+	select_peripheral(PIOA, pioa_mask, PERIPH_A, 0);
+	piob_mask = 0;
+
+	switch (data->slot[1].bus_width) {
+	case 4:
+		piob_mask |= 1 <<  8;		/* DATA1 */
+		piob_mask |= 1 <<  9;		/* DATA2 */
+		piob_mask |= 1 << 10;		/* DATA3 */
+		/* fall through */
+	case 1:
+		piob_mask |= 1 <<  6;		/* CMD	 */
+		piob_mask |= 1 <<  7;		/* DATA0 */
+		select_peripheral(PIOB, piob_mask, PERIPH_B, 0);
+
+		if (gpio_is_valid(data->slot[1].detect_pin))
+			at32_select_gpio(data->slot[1].detect_pin, 0);
+		if (gpio_is_valid(data->slot[1].wp_pin))
+			at32_select_gpio(data->slot[1].wp_pin, 0);
+		break;
+	case 0:
+		/* Slot is unused */
+		break;
+	default:
+		if (!data->slot[0].bus_width)
+			goto fail;
+
+		data->slot[1].bus_width = 0;
+		break;
+	}
 
 	atmel_mci0_pclk.dev = &pdev->dev;
 
@@ -1353,13 +1436,14 @@ static struct clk atmel_lcdfb0_pixclk = {
 struct platform_device *__init
 at32_add_device_lcdc(unsigned int id, struct atmel_lcdfb_info *data,
 		     unsigned long fbmem_start, unsigned long fbmem_len,
-		     unsigned int pin_config)
+		     u64 pin_mask)
 {
 	struct platform_device *pdev;
 	struct atmel_lcdfb_info *info;
 	struct fb_monspecs *monspecs;
 	struct fb_videomode *modedb;
 	unsigned int modedb_size;
+	u32 portc_mask, portd_mask, porte_mask;
 
 	/*
 	 * Do a deep copy of the fb data, monspecs and modedb. Make
@@ -1381,76 +1465,21 @@ at32_add_device_lcdc(unsigned int id, struct atmel_lcdfb_info *data,
 	case 0:
 		pdev = &atmel_lcdfb0_device;
 
-		switch (pin_config) {
-		case 0:
-			select_peripheral(PC(19), PERIPH_A, 0);	/* CC	  */
-			select_peripheral(PC(20), PERIPH_A, 0);	/* HSYNC  */
-			select_peripheral(PC(21), PERIPH_A, 0);	/* PCLK	  */
-			select_peripheral(PC(22), PERIPH_A, 0);	/* VSYNC  */
-			select_peripheral(PC(23), PERIPH_A, 0);	/* DVAL	  */
-			select_peripheral(PC(24), PERIPH_A, 0);	/* MODE	  */
-			select_peripheral(PC(25), PERIPH_A, 0);	/* PWR	  */
-			select_peripheral(PC(26), PERIPH_A, 0);	/* DATA0  */
-			select_peripheral(PC(27), PERIPH_A, 0);	/* DATA1  */
-			select_peripheral(PC(28), PERIPH_A, 0);	/* DATA2  */
-			select_peripheral(PC(29), PERIPH_A, 0);	/* DATA3  */
-			select_peripheral(PC(30), PERIPH_A, 0);	/* DATA4  */
-			select_peripheral(PC(31), PERIPH_A, 0);	/* DATA5  */
-			select_peripheral(PD(0),  PERIPH_A, 0);	/* DATA6  */
-			select_peripheral(PD(1),  PERIPH_A, 0);	/* DATA7  */
-			select_peripheral(PD(2),  PERIPH_A, 0);	/* DATA8  */
-			select_peripheral(PD(3),  PERIPH_A, 0);	/* DATA9  */
-			select_peripheral(PD(4),  PERIPH_A, 0);	/* DATA10 */
-			select_peripheral(PD(5),  PERIPH_A, 0);	/* DATA11 */
-			select_peripheral(PD(6),  PERIPH_A, 0);	/* DATA12 */
-			select_peripheral(PD(7),  PERIPH_A, 0);	/* DATA13 */
-			select_peripheral(PD(8),  PERIPH_A, 0);	/* DATA14 */
-			select_peripheral(PD(9),  PERIPH_A, 0);	/* DATA15 */
-			select_peripheral(PD(10), PERIPH_A, 0);	/* DATA16 */
-			select_peripheral(PD(11), PERIPH_A, 0);	/* DATA17 */
-			select_peripheral(PD(12), PERIPH_A, 0);	/* DATA18 */
-			select_peripheral(PD(13), PERIPH_A, 0);	/* DATA19 */
-			select_peripheral(PD(14), PERIPH_A, 0);	/* DATA20 */
-			select_peripheral(PD(15), PERIPH_A, 0);	/* DATA21 */
-			select_peripheral(PD(16), PERIPH_A, 0);	/* DATA22 */
-			select_peripheral(PD(17), PERIPH_A, 0);	/* DATA23 */
-			break;
-		case 1:
-			select_peripheral(PE(0),  PERIPH_B, 0);	/* CC	  */
-			select_peripheral(PC(20), PERIPH_A, 0);	/* HSYNC  */
-			select_peripheral(PC(21), PERIPH_A, 0);	/* PCLK	  */
-			select_peripheral(PC(22), PERIPH_A, 0);	/* VSYNC  */
-			select_peripheral(PE(1),  PERIPH_B, 0);	/* DVAL	  */
-			select_peripheral(PE(2),  PERIPH_B, 0);	/* MODE	  */
-			select_peripheral(PC(25), PERIPH_A, 0);	/* PWR	  */
-			select_peripheral(PE(3),  PERIPH_B, 0);	/* DATA0  */
-			select_peripheral(PE(4),  PERIPH_B, 0);	/* DATA1  */
-			select_peripheral(PE(5),  PERIPH_B, 0);	/* DATA2  */
-			select_peripheral(PE(6),  PERIPH_B, 0);	/* DATA3  */
-			select_peripheral(PE(7),  PERIPH_B, 0);	/* DATA4  */
-			select_peripheral(PC(31), PERIPH_A, 0);	/* DATA5  */
-			select_peripheral(PD(0),  PERIPH_A, 0);	/* DATA6  */
-			select_peripheral(PD(1),  PERIPH_A, 0);	/* DATA7  */
-			select_peripheral(PE(8),  PERIPH_B, 0);	/* DATA8  */
-			select_peripheral(PE(9),  PERIPH_B, 0);	/* DATA9  */
-			select_peripheral(PE(10), PERIPH_B, 0);	/* DATA10 */
-			select_peripheral(PE(11), PERIPH_B, 0);	/* DATA11 */
-			select_peripheral(PE(12), PERIPH_B, 0);	/* DATA12 */
-			select_peripheral(PD(7),  PERIPH_A, 0);	/* DATA13 */
-			select_peripheral(PD(8),  PERIPH_A, 0);	/* DATA14 */
-			select_peripheral(PD(9),  PERIPH_A, 0);	/* DATA15 */
-			select_peripheral(PE(13), PERIPH_B, 0);	/* DATA16 */
-			select_peripheral(PE(14), PERIPH_B, 0);	/* DATA17 */
-			select_peripheral(PE(15), PERIPH_B, 0);	/* DATA18 */
-			select_peripheral(PE(16), PERIPH_B, 0);	/* DATA19 */
-			select_peripheral(PE(17), PERIPH_B, 0);	/* DATA20 */
-			select_peripheral(PE(18), PERIPH_B, 0);	/* DATA21 */
-			select_peripheral(PD(16), PERIPH_A, 0);	/* DATA22 */
-			select_peripheral(PD(17), PERIPH_A, 0);	/* DATA23 */
-			break;
-		default:
-			goto err_invalid_id;
-		}
+		if (pin_mask == 0ULL)
+			/* Default to "full" lcdc control signals and 24bit */
+			pin_mask = ATMEL_LCDC_PRI_24BIT | ATMEL_LCDC_PRI_CONTROL;
+
+		/* LCDC on port C */
+		portc_mask = (pin_mask & 0xfff80000) >> 19;
+		select_peripheral(PIOC, portc_mask, PERIPH_A, 0);
+
+		/* LCDC on port D */
+		portd_mask = pin_mask & 0x0003ffff;
+		select_peripheral(PIOD, portd_mask, PERIPH_A, 0);
+
+		/* LCDC on port E */
+		porte_mask = (pin_mask >> 32) & 0x0007ffff;
+		select_peripheral(PIOE, porte_mask, PERIPH_B, 0);
 
 		clk_set_parent(&atmel_lcdfb0_pixclk, &pll0);
 		clk_set_rate(&atmel_lcdfb0_pixclk, clk_get_rate(&pll0));
@@ -1499,6 +1528,7 @@ static struct clk atmel_pwm0_mck = {
 struct platform_device *__init at32_add_device_pwm(u32 mask)
 {
 	struct platform_device *pdev;
+	u32 pin_mask;
 
 	if (!mask)
 		return NULL;
@@ -1514,14 +1544,21 @@ struct platform_device *__init at32_add_device_pwm(u32 mask)
 	if (platform_device_add_data(pdev, &mask, sizeof(mask)))
 		goto out_free_pdev;
 
+	pin_mask = 0;
 	if (mask & (1 << 0))
-		select_peripheral(PA(28), PERIPH_A, 0);
+		pin_mask |= (1 << 28);
 	if (mask & (1 << 1))
-		select_peripheral(PA(29), PERIPH_A, 0);
+		pin_mask |= (1 << 29);
+	if (pin_mask > 0)
+		select_peripheral(PIOA, pin_mask, PERIPH_A, 0);
+
+	pin_mask = 0;
 	if (mask & (1 << 2))
-		select_peripheral(PA(21), PERIPH_B, 0);
+		pin_mask |= (1 << 21);
 	if (mask & (1 << 3))
-		select_peripheral(PA(22), PERIPH_B, 0);
+		pin_mask |= (1 << 22);
+	if (pin_mask > 0)
+		select_peripheral(PIOA, pin_mask, PERIPH_B, 0);
 
 	atmel_pwm0_mck.dev = &pdev->dev;
 
@@ -1562,52 +1599,65 @@ struct platform_device *__init
 at32_add_device_ssc(unsigned int id, unsigned int flags)
 {
 	struct platform_device *pdev;
+	u32 pin_mask = 0;
 
 	switch (id) {
 	case 0:
 		pdev = &ssc0_device;
 		if (flags & ATMEL_SSC_RF)
-			select_peripheral(PA(21), PERIPH_A, 0);	/* RF */
+			pin_mask |= (1 << 21);	/* RF */
 		if (flags & ATMEL_SSC_RK)
-			select_peripheral(PA(22), PERIPH_A, 0);	/* RK */
+			pin_mask |= (1 << 22);	/* RK */
 		if (flags & ATMEL_SSC_TK)
-			select_peripheral(PA(23), PERIPH_A, 0);	/* TK */
+			pin_mask |= (1 << 23);	/* TK */
 		if (flags & ATMEL_SSC_TF)
-			select_peripheral(PA(24), PERIPH_A, 0);	/* TF */
+			pin_mask |= (1 << 24);	/* TF */
 		if (flags & ATMEL_SSC_TD)
-			select_peripheral(PA(25), PERIPH_A, 0);	/* TD */
+			pin_mask |= (1 << 25);	/* TD */
 		if (flags & ATMEL_SSC_RD)
-			select_peripheral(PA(26), PERIPH_A, 0);	/* RD */
+			pin_mask |= (1 << 26);	/* RD */
+
+		if (pin_mask > 0)
+			select_peripheral(PIOA, pin_mask, PERIPH_A, 0);
+
 		break;
 	case 1:
 		pdev = &ssc1_device;
 		if (flags & ATMEL_SSC_RF)
-			select_peripheral(PA(0), PERIPH_B, 0);	/* RF */
+			pin_mask |= (1 << 0);	/* RF */
 		if (flags & ATMEL_SSC_RK)
-			select_peripheral(PA(1), PERIPH_B, 0);	/* RK */
+			pin_mask |= (1 << 1);	/* RK */
 		if (flags & ATMEL_SSC_TK)
-			select_peripheral(PA(2), PERIPH_B, 0);	/* TK */
+			pin_mask |= (1 << 2);	/* TK */
 		if (flags & ATMEL_SSC_TF)
-			select_peripheral(PA(3), PERIPH_B, 0);	/* TF */
+			pin_mask |= (1 << 3);	/* TF */
 		if (flags & ATMEL_SSC_TD)
-			select_peripheral(PA(4), PERIPH_B, 0);	/* TD */
+			pin_mask |= (1 << 4);	/* TD */
 		if (flags & ATMEL_SSC_RD)
-			select_peripheral(PA(5), PERIPH_B, 0);	/* RD */
+			pin_mask |= (1 << 5);	/* RD */
+
+		if (pin_mask > 0)
+			select_peripheral(PIOA, pin_mask, PERIPH_B, 0);
+
 		break;
 	case 2:
 		pdev = &ssc2_device;
 		if (flags & ATMEL_SSC_TD)
-			select_peripheral(PB(13), PERIPH_A, 0);	/* TD */
+			pin_mask |= (1 << 13);	/* TD */
 		if (flags & ATMEL_SSC_RD)
-			select_peripheral(PB(14), PERIPH_A, 0);	/* RD */
+			pin_mask |= (1 << 14);	/* RD */
 		if (flags & ATMEL_SSC_TK)
-			select_peripheral(PB(15), PERIPH_A, 0);	/* TK */
+			pin_mask |= (1 << 15);	/* TK */
 		if (flags & ATMEL_SSC_TF)
-			select_peripheral(PB(16), PERIPH_A, 0);	/* TF */
+			pin_mask |= (1 << 16);	/* TF */
 		if (flags & ATMEL_SSC_RF)
-			select_peripheral(PB(17), PERIPH_A, 0);	/* RF */
+			pin_mask |= (1 << 17);	/* RF */
 		if (flags & ATMEL_SSC_RK)
-			select_peripheral(PB(18), PERIPH_A, 0);	/* RK */
+			pin_mask |= (1 << 18);	/* RK */
+
+		if (pin_mask > 0)
+			select_peripheral(PIOB, pin_mask, PERIPH_A, 0);
+
 		break;
 	default:
 		return NULL;
@@ -1745,14 +1795,15 @@ static int __init at32_init_ide_or_cf(struct platform_device *pdev,
 		unsigned int cs, unsigned int extint)
 {
 	static unsigned int extint_pin_map[4] __initdata = {
-		GPIO_PIN_PB(25),
-		GPIO_PIN_PB(26),
-		GPIO_PIN_PB(27),
-		GPIO_PIN_PB(28),
+		(1 << 25),
+		(1 << 26),
+		(1 << 27),
+		(1 << 28),
 	};
 	static bool common_pins_initialized __initdata = false;
 	unsigned int extint_pin;
 	int ret;
+	u32 pin_mask;
 
 	if (extint >= ARRAY_SIZE(extint_pin_map))
 		return -EINVAL;
@@ -1766,7 +1817,8 @@ static int __init at32_init_ide_or_cf(struct platform_device *pdev,
 		if (ret)
 			return ret;
 
-		select_peripheral(PE(21), PERIPH_A, 0); /* NCS4   -> OE_N  */
+		/* NCS4   -> OE_N  */
+		select_peripheral(PIOE, (1 << 21), PERIPH_A, 0);
 		hmatrix_sfr_set_bits(HMATRIX_SLAVE_EBI, HMATRIX_EBI_CF0_ENABLE);
 		break;
 	case 5:
@@ -1776,7 +1828,8 @@ static int __init at32_init_ide_or_cf(struct platform_device *pdev,
 		if (ret)
 			return ret;
 
-		select_peripheral(PE(22), PERIPH_A, 0); /* NCS5   -> OE_N  */
+		/* NCS5   -> OE_N  */
+		select_peripheral(PIOE, (1 << 22), PERIPH_A, 0);
 		hmatrix_sfr_set_bits(HMATRIX_SLAVE_EBI, HMATRIX_EBI_CF1_ENABLE);
 		break;
 	default:
@@ -1784,14 +1837,17 @@ static int __init at32_init_ide_or_cf(struct platform_device *pdev,
 	}
 
 	if (!common_pins_initialized) {
-		select_peripheral(PE(19), PERIPH_A, 0);	/* CFCE1  -> CS0_N */
-		select_peripheral(PE(20), PERIPH_A, 0);	/* CFCE2  -> CS1_N */
-		select_peripheral(PE(23), PERIPH_A, 0); /* CFRNW  -> DIR   */
-		select_peripheral(PE(24), PERIPH_A, 0); /* NWAIT  <- IORDY */
+		pin_mask  = (1 << 19);	/* CFCE1  -> CS0_N */
+		pin_mask |= (1 << 20);	/* CFCE2  -> CS1_N */
+		pin_mask |= (1 << 23);	/* CFRNW  -> DIR   */
+		pin_mask |= (1 << 24);	/* NWAIT  <- IORDY */
+
+		select_peripheral(PIOE, pin_mask, PERIPH_A, 0);
+
 		common_pins_initialized = true;
 	}
 
-	at32_select_periph(extint_pin, GPIO_PERIPH_A, AT32_GPIOF_DEGLITCH);
+	select_peripheral(PIOB, extint_pin, PERIPH_A, AT32_GPIOF_DEGLITCH);
 
 	pdev->resource[1].start = EIM_IRQ_BASE + extint;
 	pdev->resource[1].end = pdev->resource[1].start;
@@ -1930,6 +1986,7 @@ at32_add_device_ac97c(unsigned int id, struct ac97c_platform_data *data)
 {
 	struct platform_device *pdev;
 	struct ac97c_platform_data _data;
+	u32 pin_mask;
 
 	if (id != 0)
 		return NULL;
@@ -1956,10 +2013,10 @@ at32_add_device_ac97c(unsigned int id, struct ac97c_platform_data *data)
 				sizeof(struct ac97c_platform_data)))
 		goto fail;
 
-	select_peripheral(PB(20), PERIPH_B, 0);	/* SDO	*/
-	select_peripheral(PB(21), PERIPH_B, 0);	/* SYNC	*/
-	select_peripheral(PB(22), PERIPH_B, 0);	/* SCLK	*/
-	select_peripheral(PB(23), PERIPH_B, 0);	/* SDI	*/
+	pin_mask  = (1 << 20) | (1 << 21);	/* SDO & SYNC */
+	pin_mask |= (1 << 22) | (1 << 23);	/* SCLK & SDI */
+
+	select_peripheral(PIOB, pin_mask, PERIPH_B, 0);
 
 	/* TODO: gpio_is_valid(data->reset_pin) with kernel 2.6.26. */
 	if (data->reset_pin != GPIO_PIN_NONE)
@@ -2001,6 +2058,7 @@ static struct clk abdac0_sample_clk = {
 struct platform_device *__init at32_add_device_abdac(unsigned int id)
 {
 	struct platform_device *pdev;
+	u32 pin_mask;
 
 	if (id != 0)
 		return NULL;
@@ -2013,10 +2071,10 @@ struct platform_device *__init at32_add_device_abdac(unsigned int id)
 				ARRAY_SIZE(abdac0_resource)))
 		goto err_add_resources;
 
-	select_peripheral(PB(20), PERIPH_A, 0);	/* DATA1	*/
-	select_peripheral(PB(21), PERIPH_A, 0);	/* DATA0	*/
-	select_peripheral(PB(22), PERIPH_A, 0);	/* DATAN1	*/
-	select_peripheral(PB(23), PERIPH_A, 0);	/* DATAN0	*/
+	pin_mask  = (1 << 20) | (1 << 22);	/* DATA1 & DATAN1 */
+	pin_mask |= (1 << 21) | (1 << 23);	/* DATA0 & DATAN0 */
+
+	select_peripheral(PIOB, pin_mask, PERIPH_A, 0);
 
 	abdac0_pclk.dev = &pdev->dev;
 	abdac0_sample_clk.dev = &pdev->dev;
@@ -2073,7 +2131,7 @@ static struct clk gclk4 = {
 	.index		= 4,
 };
 
-struct clk *at32_clock_list[] = {
+static __initdata struct clk *init_clocks[] = {
 	&osc32k,
 	&osc0,
 	&osc1,
@@ -2137,7 +2195,6 @@ struct clk *at32_clock_list[] = {
 	&gclk3,
 	&gclk4,
 };
-unsigned int at32_nr_clocks = ARRAY_SIZE(at32_clock_list);
 
 void __init setup_platform(void)
 {
@@ -2168,14 +2225,19 @@ void __init setup_platform(void)
 	genclk_init_parent(&abdac0_sample_clk);
 
 	/*
-	 * Turn on all clocks that have at least one user already, and
-	 * turn off everything else. We only do this for module
-	 * clocks, and even though it isn't particularly pretty to
-	 * check the address of the mode function, it should do the
-	 * trick...
+	 * Build initial dynamic clock list by registering all clocks
+	 * from the array.
+	 * At the same time, turn on all clocks that have at least one
+	 * user already, and turn off everything else. We only do this
+	 * for module clocks, and even though it isn't particularly
+	 * pretty to  check the address of the mode function, it should
+	 * do the trick...
 	 */
-	for (i = 0; i < ARRAY_SIZE(at32_clock_list); i++) {
-		struct clk *clk = at32_clock_list[i];
+	for (i = 0; i < ARRAY_SIZE(init_clocks); i++) {
+		struct clk *clk = init_clocks[i];
+
+		/* first, register clock */
+		at32_clk_register(clk);
 
 		if (clk->users == 0)
 			continue;
diff --git a/arch/avr32/mach-at32ap/clock.c b/arch/avr32/mach-at32ap/clock.c
index 6c27ddac5adf..138a00a2a2d0 100644
--- a/arch/avr32/mach-at32ap/clock.c
+++ b/arch/avr32/mach-at32ap/clock.c
@@ -15,24 +15,40 @@
 #include <linux/err.h>
 #include <linux/device.h>
 #include <linux/string.h>
+#include <linux/list.h>
 
 #include <mach/chip.h>
 
 #include "clock.h"
 
+/* at32 clock list */
+static LIST_HEAD(at32_clock_list);
+
 static DEFINE_SPINLOCK(clk_lock);
+static DEFINE_SPINLOCK(clk_list_lock);
+
+void at32_clk_register(struct clk *clk)
+{
+	spin_lock(&clk_list_lock);
+	/* add the new item to the end of the list */
+	list_add_tail(&clk->list, &at32_clock_list);
+	spin_unlock(&clk_list_lock);
+}
 
 struct clk *clk_get(struct device *dev, const char *id)
 {
-	int i;
+	struct clk *clk;
 
-	for (i = 0; i < at32_nr_clocks; i++) {
-		struct clk *clk = at32_clock_list[i];
+	spin_lock(&clk_list_lock);
 
-		if (clk->dev == dev && strcmp(id, clk->name) == 0)
+	list_for_each_entry(clk, &at32_clock_list, list) {
+		if (clk->dev == dev && strcmp(id, clk->name) == 0) {
+			spin_unlock(&clk_list_lock);
 			return clk;
+		}
 	}
 
+	spin_unlock(&clk_list_lock);
 	return ERR_PTR(-ENOENT);
 }
 EXPORT_SYMBOL(clk_get);
@@ -203,8 +219,8 @@ dump_clock(struct clk *parent, struct clkinf *r)
 
 	/* cost of this scan is small, but not linear... */
 	r->nest = nest + NEST_DELTA;
-	for (i = 3; i < at32_nr_clocks; i++) {
-		clk = at32_clock_list[i];
+
+	list_for_each_entry(clk, &at32_clock_list, list) {
 		if (clk->parent == parent)
 			dump_clock(clk, r);
 	}
@@ -215,6 +231,7 @@ static int clk_show(struct seq_file *s, void *unused)
 {
 	struct clkinf	r;
 	int		i;
+	struct clk 	*clk;
 
 	/* show all the power manager registers */
 	seq_printf(s, "MCCTRL  = %8x\n", pm_readl(MCCTRL));
@@ -234,14 +251,25 @@ static int clk_show(struct seq_file *s, void *unused)
 
 	seq_printf(s, "\n");
 
-	/* show clock tree as derived from the three oscillators
-	 * we "know" are at the head of the list
-	 */
 	r.s = s;
 	r.nest = 0;
-	dump_clock(at32_clock_list[0], &r);
-	dump_clock(at32_clock_list[1], &r);
-	dump_clock(at32_clock_list[2], &r);
+	/* protected from changes on the list while dumping */
+	spin_lock(&clk_list_lock);
+
+	/* show clock tree as derived from the three oscillators */
+	clk = clk_get(NULL, "osc32k");
+	dump_clock(clk, &r);
+	clk_put(clk);
+
+	clk = clk_get(NULL, "osc0");
+	dump_clock(clk, &r);
+	clk_put(clk);
+
+	clk = clk_get(NULL, "osc1");
+	dump_clock(clk, &r);
+	clk_put(clk);
+
+	spin_unlock(&clk_list_lock);
 
 	return 0;
 }
diff --git a/arch/avr32/mach-at32ap/clock.h b/arch/avr32/mach-at32ap/clock.h
index bb8e1f295835..623bf0e9a1e7 100644
--- a/arch/avr32/mach-at32ap/clock.h
+++ b/arch/avr32/mach-at32ap/clock.h
@@ -12,8 +12,13 @@
  * published by the Free Software Foundation.
  */
 #include <linux/clk.h>
+#include <linux/list.h>
+
+
+void at32_clk_register(struct clk *clk);
 
 struct clk {
+	struct list_head list;		/* linking element */
 	const char	*name;		/* Clock name/function */
 	struct device	*dev;		/* Device the clock is used by */
 	struct clk	*parent;	/* Parent clock, if any */
@@ -25,6 +30,3 @@ struct clk {
 	u16		users;		/* Enabled if non-zero */
 	u16		index;		/* Sibling index */
 };
-
-extern struct clk *at32_clock_list[];
-extern unsigned int at32_nr_clocks;
diff --git a/arch/avr32/mach-at32ap/include/mach/at32ap700x.h b/arch/avr32/mach-at32ap/include/mach/at32ap700x.h
index 1e9852d65cca..a77d372f6f3e 100644
--- a/arch/avr32/mach-at32ap/include/mach/at32ap700x.h
+++ b/arch/avr32/mach-at32ap/include/mach/at32ap700x.h
@@ -83,4 +83,132 @@
 #define HMATRIX_BASE	0xfff00800
 #define SDRAMC_BASE	0xfff03800
 
+/* LCDC on port C */
+#define ATMEL_LCDC_PC_CC	(1ULL << 19)
+#define ATMEL_LCDC_PC_HSYNC	(1ULL << 20)
+#define ATMEL_LCDC_PC_PCLK	(1ULL << 21)
+#define ATMEL_LCDC_PC_VSYNC	(1ULL << 22)
+#define ATMEL_LCDC_PC_DVAL	(1ULL << 23)
+#define ATMEL_LCDC_PC_MODE	(1ULL << 24)
+#define ATMEL_LCDC_PC_PWR	(1ULL << 25)
+#define ATMEL_LCDC_PC_DATA0	(1ULL << 26)
+#define ATMEL_LCDC_PC_DATA1	(1ULL << 27)
+#define ATMEL_LCDC_PC_DATA2	(1ULL << 28)
+#define ATMEL_LCDC_PC_DATA3	(1ULL << 29)
+#define ATMEL_LCDC_PC_DATA4	(1ULL << 30)
+#define ATMEL_LCDC_PC_DATA5	(1ULL << 31)
+
+/* LCDC on port D */
+#define ATMEL_LCDC_PD_DATA6	(1ULL << 0)
+#define ATMEL_LCDC_PD_DATA7	(1ULL << 1)
+#define ATMEL_LCDC_PD_DATA8	(1ULL << 2)
+#define ATMEL_LCDC_PD_DATA9	(1ULL << 3)
+#define ATMEL_LCDC_PD_DATA10	(1ULL << 4)
+#define ATMEL_LCDC_PD_DATA11	(1ULL << 5)
+#define ATMEL_LCDC_PD_DATA12	(1ULL << 6)
+#define ATMEL_LCDC_PD_DATA13	(1ULL << 7)
+#define ATMEL_LCDC_PD_DATA14	(1ULL << 8)
+#define ATMEL_LCDC_PD_DATA15	(1ULL << 9)
+#define ATMEL_LCDC_PD_DATA16	(1ULL << 10)
+#define ATMEL_LCDC_PD_DATA17	(1ULL << 11)
+#define ATMEL_LCDC_PD_DATA18	(1ULL << 12)
+#define ATMEL_LCDC_PD_DATA19	(1ULL << 13)
+#define ATMEL_LCDC_PD_DATA20	(1ULL << 14)
+#define ATMEL_LCDC_PD_DATA21	(1ULL << 15)
+#define ATMEL_LCDC_PD_DATA22	(1ULL << 16)
+#define ATMEL_LCDC_PD_DATA23	(1ULL << 17)
+
+/* LCDC on port E */
+#define ATMEL_LCDC_PE_CC	(1ULL << (32 + 0))
+#define ATMEL_LCDC_PE_DVAL	(1ULL << (32 + 1))
+#define ATMEL_LCDC_PE_MODE	(1ULL << (32 + 2))
+#define ATMEL_LCDC_PE_DATA0	(1ULL << (32 + 3))
+#define ATMEL_LCDC_PE_DATA1	(1ULL << (32 + 4))
+#define ATMEL_LCDC_PE_DATA2	(1ULL << (32 + 5))
+#define ATMEL_LCDC_PE_DATA3	(1ULL << (32 + 6))
+#define ATMEL_LCDC_PE_DATA4	(1ULL << (32 + 7))
+#define ATMEL_LCDC_PE_DATA8	(1ULL << (32 + 8))
+#define ATMEL_LCDC_PE_DATA9	(1ULL << (32 + 9))
+#define ATMEL_LCDC_PE_DATA10	(1ULL << (32 + 10))
+#define ATMEL_LCDC_PE_DATA11	(1ULL << (32 + 11))
+#define ATMEL_LCDC_PE_DATA12	(1ULL << (32 + 12))
+#define ATMEL_LCDC_PE_DATA16	(1ULL << (32 + 13))
+#define ATMEL_LCDC_PE_DATA17	(1ULL << (32 + 14))
+#define ATMEL_LCDC_PE_DATA18	(1ULL << (32 + 15))
+#define ATMEL_LCDC_PE_DATA19	(1ULL << (32 + 16))
+#define ATMEL_LCDC_PE_DATA20	(1ULL << (32 + 17))
+#define ATMEL_LCDC_PE_DATA21	(1ULL << (32 + 18))
+
+
+#define ATMEL_LCDC(PORT, PIN)	(ATMEL_LCDC_##PORT##_##PIN)
+
+
+#define ATMEL_LCDC_PRI_24B_DATA	(					\
+		ATMEL_LCDC(PC, DATA0)  | ATMEL_LCDC(PC, DATA1)  |	\
+		ATMEL_LCDC(PC, DATA2)  | ATMEL_LCDC(PC, DATA3)  |	\
+		ATMEL_LCDC(PC, DATA4)  | ATMEL_LCDC(PC, DATA5)  |	\
+		ATMEL_LCDC(PD, DATA6)  | ATMEL_LCDC(PD, DATA7)  |	\
+		ATMEL_LCDC(PD, DATA8)  | ATMEL_LCDC(PD, DATA9)  |	\
+		ATMEL_LCDC(PD, DATA10) | ATMEL_LCDC(PD, DATA11) |	\
+		ATMEL_LCDC(PD, DATA12) | ATMEL_LCDC(PD, DATA13) |	\
+		ATMEL_LCDC(PD, DATA14) | ATMEL_LCDC(PD, DATA15) |	\
+		ATMEL_LCDC(PD, DATA16) | ATMEL_LCDC(PD, DATA17) |	\
+		ATMEL_LCDC(PD, DATA18) | ATMEL_LCDC(PD, DATA19) |	\
+		ATMEL_LCDC(PD, DATA20) | ATMEL_LCDC(PD, DATA21) |	\
+		ATMEL_LCDC(PD, DATA22) | ATMEL_LCDC(PD, DATA23))
+
+#define ATMEL_LCDC_ALT_24B_DATA (					\
+		ATMEL_LCDC(PE, DATA0)  | ATMEL_LCDC(PE, DATA1)  |	\
+		ATMEL_LCDC(PE, DATA2)  | ATMEL_LCDC(PE, DATA3)  |	\
+		ATMEL_LCDC(PE, DATA4)  | ATMEL_LCDC(PC, DATA5)  |	\
+		ATMEL_LCDC(PD, DATA6)  | ATMEL_LCDC(PD, DATA7)  |	\
+		ATMEL_LCDC(PE, DATA8)  | ATMEL_LCDC(PE, DATA9)  |	\
+		ATMEL_LCDC(PE, DATA10) | ATMEL_LCDC(PE, DATA11) |	\
+		ATMEL_LCDC(PE, DATA12) | ATMEL_LCDC(PD, DATA13) |	\
+		ATMEL_LCDC(PD, DATA14) | ATMEL_LCDC(PD, DATA15) |	\
+		ATMEL_LCDC(PE, DATA16) | ATMEL_LCDC(PE, DATA17) |	\
+		ATMEL_LCDC(PE, DATA18) | ATMEL_LCDC(PE, DATA19) |	\
+		ATMEL_LCDC(PE, DATA20) | ATMEL_LCDC(PE, DATA21) |	\
+		ATMEL_LCDC(PD, DATA22) | ATMEL_LCDC(PD, DATA23))
+
+#define ATMEL_LCDC_PRI_15B_DATA (					\
+		ATMEL_LCDC(PC, DATA0)  | ATMEL_LCDC(PC, DATA1)  |	\
+		ATMEL_LCDC(PC, DATA2)  | ATMEL_LCDC(PC, DATA3)  |	\
+		ATMEL_LCDC(PC, DATA4)  | ATMEL_LCDC(PC, DATA5)  |	\
+		ATMEL_LCDC(PD, DATA8)  | ATMEL_LCDC(PD, DATA9)  |	\
+		ATMEL_LCDC(PD, DATA10) | ATMEL_LCDC(PD, DATA11) |	\
+		ATMEL_LCDC(PD, DATA12) | ATMEL_LCDC(PD, DATA16) |	\
+		ATMEL_LCDC(PD, DATA17) | ATMEL_LCDC(PD, DATA18) |	\
+		ATMEL_LCDC(PD, DATA19) | ATMEL_LCDC(PD, DATA20))
+
+#define ATMEL_LCDC_ALT_15B_DATA	(					\
+		ATMEL_LCDC(PE, DATA0)  | ATMEL_LCDC(PE, DATA1)  |	\
+		ATMEL_LCDC(PE, DATA2)  | ATMEL_LCDC(PE, DATA3)  |	\
+		ATMEL_LCDC(PE, DATA4)  | ATMEL_LCDC(PC, DATA5)  |	\
+		ATMEL_LCDC(PE, DATA8)  | ATMEL_LCDC(PE, DATA9)  |	\
+		ATMEL_LCDC(PE, DATA10) | ATMEL_LCDC(PE, DATA11) |	\
+		ATMEL_LCDC(PE, DATA12) | ATMEL_LCDC(PE, DATA16) |	\
+		ATMEL_LCDC(PE, DATA17) | ATMEL_LCDC(PE, DATA18) |	\
+		ATMEL_LCDC(PE, DATA19) | ATMEL_LCDC(PE, DATA20))
+
+#define ATMEL_LCDC_PRI_CONTROL (					\
+		ATMEL_LCDC(PC, CC)   | ATMEL_LCDC(PC, DVAL) |		\
+		ATMEL_LCDC(PC, MODE) | ATMEL_LCDC(PC, PWR))
+
+#define ATMEL_LCDC_ALT_CONTROL (					\
+		ATMEL_LCDC(PE, CC)   | ATMEL_LCDC(PE, DVAL) |		\
+		ATMEL_LCDC(PE, MODE) | ATMEL_LCDC(PC, PWR))
+
+#define ATMEL_LCDC_CONTROL (						\
+		ATMEL_LCDC(PC, HSYNC) | ATMEL_LCDC(PC, VSYNC) |		\
+		ATMEL_LCDC(PC, PCLK))
+
+#define ATMEL_LCDC_PRI_24BIT	(ATMEL_LCDC_CONTROL | ATMEL_LCDC_PRI_24B_DATA)
+
+#define ATMEL_LCDC_ALT_24BIT	(ATMEL_LCDC_CONTROL | ATMEL_LCDC_ALT_24B_DATA)
+
+#define ATMEL_LCDC_PRI_15BIT	(ATMEL_LCDC_CONTROL | ATMEL_LCDC_PRI_15B_DATA)
+
+#define ATMEL_LCDC_ALT_15BIT	(ATMEL_LCDC_CONTROL | ATMEL_LCDC_ALT_15B_DATA)
+
 #endif /* __ASM_ARCH_AT32AP700X_H__ */
diff --git a/arch/avr32/mach-at32ap/include/mach/board.h b/arch/avr32/mach-at32ap/include/mach/board.h
index e60e9076544d..c48386d66bc3 100644
--- a/arch/avr32/mach-at32ap/include/mach/board.h
+++ b/arch/avr32/mach-at32ap/include/mach/board.h
@@ -43,7 +43,7 @@ struct atmel_lcdfb_info;
 struct platform_device *
 at32_add_device_lcdc(unsigned int id, struct atmel_lcdfb_info *data,
 		     unsigned long fbmem_start, unsigned long fbmem_len,
-		     unsigned int pin_config);
+		     u64 pin_mask);
 
 struct usba_platform_data;
 struct platform_device *
diff --git a/arch/avr32/mach-at32ap/include/mach/io.h b/arch/avr32/mach-at32ap/include/mach/io.h
index 4ec6abc68ea3..22ea79b74052 100644
--- a/arch/avr32/mach-at32ap/include/mach/io.h
+++ b/arch/avr32/mach-at32ap/include/mach/io.h
@@ -1,8 +1,7 @@
 #ifndef __ASM_AVR32_ARCH_AT32AP_IO_H
 #define __ASM_AVR32_ARCH_AT32AP_IO_H
 
-/* For "bizarre" halfword swapping */
-#include <linux/byteorder/swabb.h>
+#include <linux/swab.h>
 
 #if defined(CONFIG_AP700X_32_BIT_SMC)
 # define __swizzle_addr_b(addr)	(addr ^ 3UL)
diff --git a/arch/avr32/mach-at32ap/include/mach/portmux.h b/arch/avr32/mach-at32ap/include/mach/portmux.h
index b1abe6b4e4ef..21c79373b53f 100644
--- a/arch/avr32/mach-at32ap/include/mach/portmux.h
+++ b/arch/avr32/mach-at32ap/include/mach/portmux.h
@@ -21,9 +21,10 @@
 #define AT32_GPIOF_DEGLITCH	0x00000008	/* (IN) Filter glitches */
 #define AT32_GPIOF_MULTIDRV	0x00000010	/* Enable multidriver option */
 
-void at32_select_periph(unsigned int pin, unsigned int periph,
-			unsigned long flags);
+void at32_select_periph(unsigned int port, unsigned int pin,
+			unsigned int periph, unsigned long flags);
 void at32_select_gpio(unsigned int pin, unsigned long flags);
+void at32_deselect_pin(unsigned int pin);
 void at32_reserve_pin(unsigned int pin);
 
 #endif /* __ASM_ARCH_PORTMUX_H__ */
diff --git a/arch/avr32/mach-at32ap/pdc.c b/arch/avr32/mach-at32ap/pdc.c
index 1040bda4fda7..61ab15aae970 100644
--- a/arch/avr32/mach-at32ap/pdc.c
+++ b/arch/avr32/mach-at32ap/pdc.c
@@ -35,7 +35,6 @@ static int __init pdc_probe(struct platform_device *pdev)
 }
 
 static struct platform_driver pdc_driver = {
-	.probe		= pdc_probe,
 	.driver		= {
 		.name	= "pdc",
 	},
@@ -43,6 +42,6 @@ static struct platform_driver pdc_driver = {
 
 static int __init pdc_init(void)
 {
-	return platform_driver_register(&pdc_driver);
+	return platform_driver_probe(&pdc_driver, pdc_probe);
 }
 arch_initcall(pdc_init);
diff --git a/arch/avr32/mach-at32ap/pio.c b/arch/avr32/mach-at32ap/pio.c
index 405ee6bad4ce..ed81a8bcb22d 100644
--- a/arch/avr32/mach-at32ap/pio.c
+++ b/arch/avr32/mach-at32ap/pio.c
@@ -50,35 +50,48 @@ static struct pio_device *gpio_to_pio(unsigned int gpio)
 }
 
 /* Pin multiplexing API */
+static DEFINE_SPINLOCK(pio_lock);
 
-void __init at32_select_periph(unsigned int pin, unsigned int periph,
-			       unsigned long flags)
+void __init at32_select_periph(unsigned int port, u32 pin_mask,
+			       unsigned int periph, unsigned long flags)
 {
 	struct pio_device *pio;
-	unsigned int pin_index = pin & 0x1f;
-	u32 mask = 1 << pin_index;
 
-	pio = gpio_to_pio(pin);
+	/* assign and verify pio */
+	pio = gpio_to_pio(port);
 	if (unlikely(!pio)) {
-		printk("pio: invalid pin %u\n", pin);
+		printk(KERN_WARNING "pio: invalid port %u\n", port);
 		goto fail;
 	}
 
-	if (unlikely(test_and_set_bit(pin_index, &pio->pinmux_mask)
-			 || gpiochip_is_requested(&pio->chip, pin_index))) {
-		printk("%s: pin %u is busy\n", pio->name, pin_index);
+	/* Test if any of the requested pins is already muxed */
+	spin_lock(&pio_lock);
+	if (unlikely(pio->pinmux_mask & pin_mask)) {
+		printk(KERN_WARNING "%s: pin(s) busy (requested 0x%x, busy 0x%x)\n",
+		       pio->name, pin_mask, pio->pinmux_mask & pin_mask);
+		spin_unlock(&pio_lock);
 		goto fail;
 	}
 
-	pio_writel(pio, PUER, mask);
+	pio->pinmux_mask |= pin_mask;
+
+	/* enable pull ups */
+	pio_writel(pio, PUER, pin_mask);
+
+	/* select either peripheral A or B */
 	if (periph)
-		pio_writel(pio, BSR, mask);
+		pio_writel(pio, BSR, pin_mask);
 	else
-		pio_writel(pio, ASR, mask);
+		pio_writel(pio, ASR, pin_mask);
+
+	/* enable peripheral control */
+	pio_writel(pio, PDR, pin_mask);
 
-	pio_writel(pio, PDR, mask);
+	/* Disable pull ups if not requested. */
 	if (!(flags & AT32_GPIOF_PULLUP))
-		pio_writel(pio, PUDR, mask);
+		pio_writel(pio, PUDR, pin_mask);
+
+	spin_unlock(&pio_lock);
 
 	return;
 
@@ -134,6 +147,25 @@ fail:
 	dump_stack();
 }
 
+/*
+ * Undo a previous pin reservation. Will not affect the hardware
+ * configuration.
+ */
+void at32_deselect_pin(unsigned int pin)
+{
+	struct pio_device *pio;
+	unsigned int pin_index = pin & 0x1f;
+
+	pio = gpio_to_pio(pin);
+	if (unlikely(!pio)) {
+		printk("pio: invalid pin %u\n", pin);
+		dump_stack();
+		return;
+	}
+
+	clear_bit(pin_index, &pio->pinmux_mask);
+}
+
 /* Reserve a pin, preventing anyone else from changing its configuration. */
 void __init at32_reserve_pin(unsigned int pin)
 {
@@ -382,7 +414,6 @@ static int __init pio_probe(struct platform_device *pdev)
 }
 
 static struct platform_driver pio_driver = {
-	.probe		= pio_probe,
 	.driver		= {
 		.name		= "pio",
 	},
@@ -390,7 +421,7 @@ static struct platform_driver pio_driver = {
 
 static int __init pio_init(void)
 {
-	return platform_driver_register(&pio_driver);
+	return platform_driver_probe(&pio_driver, pio_probe);
 }
 postcore_initcall(pio_init);
 
diff --git a/arch/avr32/oprofile/Makefile b/arch/avr32/oprofile/Makefile
index 1fe81c3c1e86..e0eb520e0287 100644
--- a/arch/avr32/oprofile/Makefile
+++ b/arch/avr32/oprofile/Makefile
@@ -5,4 +5,4 @@ oprofile-y		:= $(addprefix ../../../drivers/oprofile/,	\
 				event_buffer.o oprofile_files.o		\
 				oprofilefs.o oprofile_stats.o		\
 				timer_int.o)
-oprofile-y		+= op_model_avr32.o
+oprofile-y		+= op_model_avr32.o backtrace.o
diff --git a/arch/avr32/oprofile/backtrace.c b/arch/avr32/oprofile/backtrace.c
new file mode 100644
index 000000000000..75d9ad6f99cf
--- /dev/null
+++ b/arch/avr32/oprofile/backtrace.c
@@ -0,0 +1,81 @@
+/*
+ * AVR32 specific backtracing code for oprofile
+ *
+ * Copyright 2008 Weinmann GmbH
+ *
+ * Author: Nikolaus Voss <n.voss@weinmann.de>
+ *
+ * Based on i386 oprofile backtrace code by John Levon and David Smith
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/oprofile.h>
+#include <linux/sched.h>
+#include <linux/uaccess.h>
+
+/* The first two words of each frame on the stack look like this if we have
+ * frame pointers */
+struct frame_head {
+	unsigned long lr;
+	struct frame_head *fp;
+};
+
+/* copied from arch/avr32/kernel/process.c */
+static inline int valid_stack_ptr(struct thread_info *tinfo, unsigned long p)
+{
+	return (p > (unsigned long)tinfo)
+		&& (p < (unsigned long)tinfo + THREAD_SIZE - 3);
+}
+
+/* copied from arch/x86/oprofile/backtrace.c */
+static struct frame_head *dump_user_backtrace(struct frame_head *head)
+{
+	struct frame_head bufhead[2];
+
+	/* Also check accessibility of one struct frame_head beyond */
+	if (!access_ok(VERIFY_READ, head, sizeof(bufhead)))
+		return NULL;
+	if (__copy_from_user_inatomic(bufhead, head, sizeof(bufhead)))
+		return NULL;
+
+	oprofile_add_trace(bufhead[0].lr);
+
+	/* frame pointers should strictly progress back up the stack
+	 * (towards higher addresses) */
+	if (bufhead[0].fp <= head)
+		return NULL;
+
+	return bufhead[0].fp;
+}
+
+void avr32_backtrace(struct pt_regs * const regs, unsigned int depth)
+{
+	/* Get first frame pointer */
+	struct frame_head *head = (struct frame_head *)(regs->r7);
+
+	if (!user_mode(regs)) {
+#ifdef CONFIG_FRAME_POINTER
+		/*
+		 * Traverse the kernel stack from frame to frame up to
+		 * "depth" steps.
+		 */
+		while (depth-- && valid_stack_ptr(task_thread_info(current),
+						  (unsigned long)head)) {
+			oprofile_add_trace(head->lr);
+			if (head->fp <= head)
+				break;
+			head = head->fp;
+		}
+#endif
+	} else {
+		/* Assume we have frame pointers in user mode process */
+		while (depth-- && head)
+			head = dump_user_backtrace(head);
+	}
+}
+
+
diff --git a/arch/avr32/oprofile/op_model_avr32.c b/arch/avr32/oprofile/op_model_avr32.c
index df42325c7f81..a3e9b3c4845a 100644
--- a/arch/avr32/oprofile/op_model_avr32.c
+++ b/arch/avr32/oprofile/op_model_avr32.c
@@ -22,6 +22,8 @@
 #define AVR32_PERFCTR_IRQ_GROUP	0
 #define AVR32_PERFCTR_IRQ_LINE	1
 
+void avr32_backtrace(struct pt_regs * const regs, unsigned int depth);
+
 enum { PCCNT, PCNT0, PCNT1, NR_counter };
 
 struct avr32_perf_counter {
@@ -223,6 +225,8 @@ int __init oprofile_arch_init(struct oprofile_operations *ops)
 	memcpy(ops, &avr32_perf_counter_ops,
 			sizeof(struct oprofile_operations));
 
+	ops->backtrace = avr32_backtrace;
+
 	printk(KERN_INFO "oprofile: using AVR32 performance monitoring.\n");
 
 	return 0;
diff --git a/arch/blackfin/include/asm/a.out.h b/arch/blackfin/include/asm/a.out.h
deleted file mode 100644
index 6c3d652ebd33..000000000000
--- a/arch/blackfin/include/asm/a.out.h
+++ /dev/null
@@ -1,19 +0,0 @@
-#ifndef __BFIN_A_OUT_H__
-#define __BFIN_A_OUT_H__
-
-struct exec {
-	unsigned long a_info;	/* Use macros N_MAGIC, etc for access */
-	unsigned a_text;	/* length of text, in bytes */
-	unsigned a_data;	/* length of data, in bytes */
-	unsigned a_bss;		/* length of uninitialized data area for file, in bytes */
-	unsigned a_syms;	/* length of symbol table data in file, in bytes */
-	unsigned a_entry;	/* start address */
-	unsigned a_trsize;	/* length of relocation info for text, in bytes */
-	unsigned a_drsize;	/* length of relocation info for data, in bytes */
-};
-
-#define N_TRSIZE(a)	((a).a_trsize)
-#define N_DRSIZE(a)	((a).a_drsize)
-#define N_SYMSIZE(a)	((a).a_syms)
-
-#endif				/* __BFIN_A_OUT_H__ */
diff --git a/arch/blackfin/kernel/bfin_dma_5xx.c b/arch/blackfin/kernel/bfin_dma_5xx.c
index 93229b3d6e3e..339293d677cc 100644
--- a/arch/blackfin/kernel/bfin_dma_5xx.c
+++ b/arch/blackfin/kernel/bfin_dma_5xx.c
@@ -117,15 +117,14 @@ int request_dma(unsigned int channel, char *device_id)
 
 #ifdef CONFIG_BF54x
 	if (channel >= CH_UART2_RX && channel <= CH_UART3_TX) {
-		if (strncmp(device_id, "BFIN_UART", 9) == 0) {
-			dma_ch[channel].regs->peripheral_map &= 0x0FFF;
-			dma_ch[channel].regs->peripheral_map |=
+		unsigned int per_map;
+		per_map = dma_ch[channel].regs->peripheral_map & 0xFFF;
+		if (strncmp(device_id, "BFIN_UART", 9) == 0)
+			dma_ch[channel].regs->peripheral_map = per_map |
 				((channel - CH_UART2_RX + 0xC)<<12);
-		} else {
-			dma_ch[channel].regs->peripheral_map &= 0x0FFF;
-			dma_ch[channel].regs->peripheral_map |=
+		else
+			dma_ch[channel].regs->peripheral_map = per_map |
 				((channel - CH_UART2_RX + 0x6)<<12);
-		}
 	}
 #endif
 
diff --git a/arch/blackfin/mach-bf527/include/mach/bfin_serial_5xx.h b/arch/blackfin/mach-bf527/include/mach/bfin_serial_5xx.h
index 2526b6ed6faa..75722d6008b0 100644
--- a/arch/blackfin/mach-bf527/include/mach/bfin_serial_5xx.h
+++ b/arch/blackfin/mach-bf527/include/mach/bfin_serial_5xx.h
@@ -78,6 +78,9 @@
 #  define CONFIG_UART1_RTS_PIN -1
 # endif
 #endif
+
+#define BFIN_UART_TX_FIFO_SIZE	2
+
 /*
  * The pin configuration is different from schematic
  */
@@ -119,7 +122,6 @@ static inline void UART_CLEAR_LSR(struct bfin_serial_port *uart)
 	bfin_write16(uart->port.membase + OFFSET_LSR, -1);
 }
 
-struct bfin_serial_port bfin_serial_ports[BFIN_UART_NR_PORTS];
 struct bfin_serial_res {
 	unsigned long uart_base_addr;
 	int uart_irq;
@@ -164,8 +166,6 @@ struct bfin_serial_res bfin_serial_resource[] = {
 #endif
 };
 
-int nr_ports = ARRAY_SIZE(bfin_serial_resource);
-
 #define DRIVER_NAME "bfin-uart"
 
 static void bfin_serial_hw_init(struct bfin_serial_port *uart)
diff --git a/arch/blackfin/mach-bf533/include/mach/bfin_serial_5xx.h b/arch/blackfin/mach-bf533/include/mach/bfin_serial_5xx.h
index ebf592b59aab..815bfe5dd1a9 100644
--- a/arch/blackfin/mach-bf533/include/mach/bfin_serial_5xx.h
+++ b/arch/blackfin/mach-bf533/include/mach/bfin_serial_5xx.h
@@ -69,6 +69,8 @@
 # endif
 #endif
 
+#define BFIN_UART_TX_FIFO_SIZE	2
+
 struct bfin_serial_port {
         struct uart_port        port;
         unsigned int            old_status;
@@ -111,7 +113,6 @@ static inline void UART_CLEAR_LSR(struct bfin_serial_port *uart)
 	bfin_write16(uart->port.membase + OFFSET_LSR, -1);
 }
 
-struct bfin_serial_port bfin_serial_ports[BFIN_UART_NR_PORTS];
 struct bfin_serial_res {
 	unsigned long	uart_base_addr;
 	int		uart_irq;
@@ -142,7 +143,6 @@ struct bfin_serial_res bfin_serial_resource[] = {
 
 #define DRIVER_NAME "bfin-uart"
 
-int nr_ports = BFIN_UART_NR_PORTS;
 static void bfin_serial_hw_init(struct bfin_serial_port *uart)
 {
 
diff --git a/arch/blackfin/mach-bf537/include/mach/bfin_serial_5xx.h b/arch/blackfin/mach-bf537/include/mach/bfin_serial_5xx.h
index 1bf56ffa22f9..b3f87e1d16a2 100644
--- a/arch/blackfin/mach-bf537/include/mach/bfin_serial_5xx.h
+++ b/arch/blackfin/mach-bf537/include/mach/bfin_serial_5xx.h
@@ -78,6 +78,9 @@
 #  define CONFIG_UART1_RTS_PIN -1
 # endif
 #endif
+
+#define BFIN_UART_TX_FIFO_SIZE	2
+
 /*
  * The pin configuration is different from schematic
  */
@@ -119,7 +122,6 @@ static inline void UART_CLEAR_LSR(struct bfin_serial_port *uart)
 	bfin_write16(uart->port.membase + OFFSET_LSR, -1);
 }
 
-struct bfin_serial_port bfin_serial_ports[BFIN_UART_NR_PORTS];
 struct bfin_serial_res {
 	unsigned long	uart_base_addr;
 	int		uart_irq;
@@ -164,8 +166,6 @@ struct bfin_serial_res bfin_serial_resource[] = {
 #endif
 };
 
-int nr_ports = ARRAY_SIZE(bfin_serial_resource);
-
 #define DRIVER_NAME "bfin-uart"
 
 static void bfin_serial_hw_init(struct bfin_serial_port *uart)
diff --git a/arch/blackfin/mach-bf548/include/mach/bfin_serial_5xx.h b/arch/blackfin/mach-bf548/include/mach/bfin_serial_5xx.h
index 5e29446a8e03..e4cf35e7ab9f 100644
--- a/arch/blackfin/mach-bf548/include/mach/bfin_serial_5xx.h
+++ b/arch/blackfin/mach-bf548/include/mach/bfin_serial_5xx.h
@@ -82,6 +82,9 @@
 #  define CONFIG_UART1_RTS_PIN -1
 # endif
 #endif
+
+#define BFIN_UART_TX_FIFO_SIZE	2
+
 /*
  * The pin configuration is different from schematic
  */
@@ -105,7 +108,6 @@ struct bfin_serial_port {
 #endif
 };
 
-struct bfin_serial_port bfin_serial_ports[BFIN_UART_NR_PORTS];
 struct bfin_serial_res {
 	unsigned long	uart_base_addr;
 	int		uart_irq;
@@ -170,8 +172,6 @@ struct bfin_serial_res bfin_serial_resource[] = {
 #endif
 };
 
-int nr_ports = ARRAY_SIZE(bfin_serial_resource);
-
 #define DRIVER_NAME "bfin-uart"
 
 static void bfin_serial_hw_init(struct bfin_serial_port *uart)
diff --git a/arch/blackfin/mach-bf561/include/mach/bfin_serial_5xx.h b/arch/blackfin/mach-bf561/include/mach/bfin_serial_5xx.h
index 8aa02780e642..e0ce0c1843d4 100644
--- a/arch/blackfin/mach-bf561/include/mach/bfin_serial_5xx.h
+++ b/arch/blackfin/mach-bf561/include/mach/bfin_serial_5xx.h
@@ -69,6 +69,8 @@
 # endif
 #endif
 
+#define BFIN_UART_TX_FIFO_SIZE	2
+
 struct bfin_serial_port {
         struct uart_port        port;
         unsigned int            old_status;
@@ -111,7 +113,6 @@ static inline void UART_CLEAR_LSR(struct bfin_serial_port *uart)
 	bfin_write16(uart->port.membase + OFFSET_LSR, -1);
 }
 
-struct bfin_serial_port bfin_serial_ports[BFIN_UART_NR_PORTS];
 struct bfin_serial_res {
 	unsigned long	uart_base_addr;
 	int		uart_irq;
@@ -142,7 +143,6 @@ struct bfin_serial_res bfin_serial_resource[] = {
 
 #define DRIVER_NAME "bfin-uart"
 
-int nr_ports = BFIN_UART_NR_PORTS;
 static void bfin_serial_hw_init(struct bfin_serial_port *uart)
 {
 
diff --git a/arch/cris/arch-v10/boot/tools/build.c b/arch/cris/arch-v10/boot/tools/build.c
index 2f9bbb26d603..c8adef364160 100644
--- a/arch/cris/arch-v10/boot/tools/build.c
+++ b/arch/cris/arch-v10/boot/tools/build.c
@@ -30,7 +30,6 @@
 #include <sys/sysmacros.h>
 #include <unistd.h>	/* contains read/write */
 #include <fcntl.h>
-#include <linux/a.out.h>
 #include <errno.h>
 
 #define MINIX_HEADER 32
diff --git a/arch/h8300/Kconfig b/arch/h8300/Kconfig
index 396ab059efa3..107cb5bb9f39 100644
--- a/arch/h8300/Kconfig
+++ b/arch/h8300/Kconfig
@@ -66,9 +66,6 @@ config TIME_LOW_RES
 	bool
 	default y
 
-config ARCH_SUPPORTS_AOUT
-	def_bool y
-
 config NO_IOPORT
 	def_bool y
 
diff --git a/arch/h8300/include/asm/a.out.h b/arch/h8300/include/asm/a.out.h
deleted file mode 100644
index ded780f0a492..000000000000
--- a/arch/h8300/include/asm/a.out.h
+++ /dev/null
@@ -1,20 +0,0 @@
-#ifndef __H8300_A_OUT_H__
-#define __H8300_A_OUT_H__
-
-struct exec
-{
-  unsigned long a_info;		/* Use macros N_MAGIC, etc for access */
-  unsigned a_text;		/* length of text, in bytes */
-  unsigned a_data;		/* length of data, in bytes */
-  unsigned a_bss;		/* length of uninitialized data area for file, in bytes */
-  unsigned a_syms;		/* length of symbol table data in file, in bytes */
-  unsigned a_entry;		/* start address */
-  unsigned a_trsize;		/* length of relocation info for text, in bytes */
-  unsigned a_drsize;		/* length of relocation info for data, in bytes */
-};
-
-#define N_TRSIZE(a)	((a).a_trsize)
-#define N_DRSIZE(a)	((a).a_drsize)
-#define N_SYMSIZE(a)	((a).a_syms)
-
-#endif /* __H8300_A_OUT_H__ */
diff --git a/arch/h8300/kernel/process.c b/arch/h8300/kernel/process.c
index dfbe7ab9ffe2..a8ef654a5a0b 100644
--- a/arch/h8300/kernel/process.c
+++ b/arch/h8300/kernel/process.c
@@ -34,7 +34,6 @@
 #include <linux/ptrace.h>
 #include <linux/slab.h>
 #include <linux/user.h>
-#include <linux/a.out.h>
 #include <linux/interrupt.h>
 #include <linux/reboot.h>
 #include <linux/fs.h>
diff --git a/arch/ia64/include/asm/a.out.h b/arch/ia64/include/asm/a.out.h
deleted file mode 100644
index 193dcfb67596..000000000000
--- a/arch/ia64/include/asm/a.out.h
+++ /dev/null
@@ -1,32 +0,0 @@
-#ifndef _ASM_IA64_A_OUT_H
-#define _ASM_IA64_A_OUT_H
-
-/*
- * No a.out format has been (or should be) defined so this file is
- * just a dummy that allows us to get binfmt_elf compiled.  It
- * probably would be better to clean up binfmt_elf.c so it does not
- * necessarily depend on there being a.out support.
- *
- * Modified 1998-2002
- *	David Mosberger-Tang <davidm@hpl.hp.com>, Hewlett-Packard Co.
- */
-
-#include <linux/types.h>
-
-struct exec {
-	unsigned long a_info;
-	unsigned long a_text;
-	unsigned long a_data;
-	unsigned long a_bss;
-	unsigned long a_entry;
-};
-
-#define N_TXTADDR(x)	0
-#define N_DATADDR(x)	0
-#define N_BSSADDR(x)	0
-#define N_DRSIZE(x)	0
-#define N_TRSIZE(x)	0
-#define N_SYMSIZE(x)	0
-#define N_TXTOFF(x)	0
-
-#endif /* _ASM_IA64_A_OUT_H */
diff --git a/arch/ia64/include/asm/siginfo.h b/arch/ia64/include/asm/siginfo.h
index 9294e4b0c8bc..118d42979003 100644
--- a/arch/ia64/include/asm/siginfo.h
+++ b/arch/ia64/include/asm/siginfo.h
@@ -113,11 +113,6 @@ typedef struct siginfo {
 #undef NSIGSEGV
 #define NSIGSEGV	3
 
-/*
- * SIGTRAP si_codes
- */
-#define TRAP_BRANCH	(__SI_FAULT|3)	/* process taken branch trap */
-#define TRAP_HWBKPT	(__SI_FAULT|4)	/* hardware breakpoint or watchpoint */
 #undef NSIGTRAP
 #define NSIGTRAP	4
 
diff --git a/arch/ia64/include/asm/statfs.h b/arch/ia64/include/asm/statfs.h
index 811097974f31..1e589669de56 100644
--- a/arch/ia64/include/asm/statfs.h
+++ b/arch/ia64/include/asm/statfs.h
@@ -8,55 +8,13 @@
  *	David Mosberger-Tang <davidm@hpl.hp.com>, Hewlett-Packard Co
  */
 
-#ifndef __KERNEL_STRICT_NAMES
-# include <linux/types.h>
-typedef __kernel_fsid_t	fsid_t;
-#endif
-
 /*
- * This is ugly --- we're already 64-bit, so just duplicate the definitions
+ * We need compat_statfs64 to be packed, because the i386 ABI won't
+ * add padding at the end to bring it to a multiple of 8 bytes, but
+ * the IA64 ABI will.
  */
-struct statfs {
-	long f_type;
-	long f_bsize;
-	long f_blocks;
-	long f_bfree;
-	long f_bavail;
-	long f_files;
-	long f_ffree;
-	__kernel_fsid_t f_fsid;
-	long f_namelen;
-	long f_frsize;
-	long f_spare[5];
-};
-
-
-struct statfs64 {
-	long f_type;
-	long f_bsize;
-	long f_blocks;
-	long f_bfree;
-	long f_bavail;
-	long f_files;
-	long f_ffree;
-	__kernel_fsid_t f_fsid;
-	long f_namelen;
-	long f_frsize;
-	long f_spare[5];
-};
+#define ARCH_PACK_COMPAT_STATFS64 __attribute__((packed,aligned(4)))
 
-struct compat_statfs64 {
-	__u32 f_type;
-	__u32 f_bsize;
-	__u64 f_blocks;
-	__u64 f_bfree;
-	__u64 f_bavail;
-	__u64 f_files;
-	__u64 f_ffree;
-	__kernel_fsid_t f_fsid;
-	__u32 f_namelen;
-	__u32 f_frsize;
-	__u32 f_spare[5];
-} __attribute__((packed));
+#include <asm-generic/statfs.h>
 
 #endif /* _ASM_IA64_STATFS_H */
diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c
index 200100ea7610..f482a9098e32 100644
--- a/arch/ia64/mm/init.c
+++ b/arch/ia64/mm/init.c
@@ -21,7 +21,6 @@
 #include <linux/bitops.h>
 #include <linux/kexec.h>
 
-#include <asm/a.out.h>
 #include <asm/dma.h>
 #include <asm/ia32.h>
 #include <asm/io.h>
diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig
index f57113f1f892..00289c178f89 100644
--- a/arch/m32r/Kconfig
+++ b/arch/m32r/Kconfig
@@ -36,9 +36,6 @@ config NO_IOPORT
 config NO_DMA
 	def_bool y
 
-config ARCH_SUPPORTS_AOUT
-	def_bool y
-
 config HZ
 	int
 	default 100
diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig
index 8c5e1de68fcb..41e5bf02e230 100644
--- a/arch/m68k/Kconfig
+++ b/arch/m68k/Kconfig
@@ -5,6 +5,7 @@
 config M68K
 	bool
 	default y
+	select HAVE_AOUT
 	select HAVE_IDE
 
 config MMU
@@ -53,9 +54,6 @@ config NO_IOPORT
 config NO_DMA
 	def_bool SUN3
 
-config ARCH_SUPPORTS_AOUT
-	def_bool y
-
 config HZ
 	int
 	default 100
diff --git a/arch/m68knommu/Kconfig b/arch/m68knommu/Kconfig
index 2e7515e8db98..0a8998315e5e 100644
--- a/arch/m68knommu/Kconfig
+++ b/arch/m68knommu/Kconfig
@@ -73,9 +73,6 @@ config GENERIC_CLOCKEVENTS
 config NO_IOPORT
 	def_bool y
 
-config ARCH_SUPPORTS_AOUT
-	def_bool y
-
 source "init/Kconfig"
 
 menu "Processor type and features"
diff --git a/arch/m68knommu/include/asm/a.out.h b/arch/m68knommu/include/asm/a.out.h
deleted file mode 100644
index ce18ef99de04..000000000000
--- a/arch/m68knommu/include/asm/a.out.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-m68k/a.out.h>
diff --git a/arch/m68knommu/kernel/process.c b/arch/m68knommu/kernel/process.c
index 47502d5ec19f..3f2d7745f31e 100644
--- a/arch/m68knommu/kernel/process.c
+++ b/arch/m68knommu/kernel/process.c
@@ -25,7 +25,6 @@
 #include <linux/ptrace.h>
 #include <linux/slab.h>
 #include <linux/user.h>
-#include <linux/a.out.h>
 #include <linux/interrupt.h>
 #include <linux/reboot.h>
 #include <linux/fs.h>
diff --git a/arch/m68knommu/kernel/traps.c b/arch/m68knommu/kernel/traps.c
index 46f8f9d0c408..5d5d56bcd0ef 100644
--- a/arch/m68knommu/kernel/traps.c
+++ b/arch/m68knommu/kernel/traps.c
@@ -22,7 +22,6 @@
 #include <linux/mm.h>
 #include <linux/module.h>
 #include <linux/types.h>
-#include <linux/a.out.h>
 #include <linux/user.h>
 #include <linux/string.h>
 #include <linux/linkage.h>
diff --git a/arch/mips/include/asm/a.out.h b/arch/mips/include/asm/a.out.h
deleted file mode 100644
index cad8371422ab..000000000000
--- a/arch/mips/include/asm/a.out.h
+++ /dev/null
@@ -1,35 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 1994 - 1999, 2003 by Ralf Baechle
- */
-#ifndef _ASM_A_OUT_H
-#define _ASM_A_OUT_H
-
-#ifdef __KERNEL__
-
-
-#endif
-
-struct exec
-{
-	unsigned long a_info;	/* Use macros N_MAGIC, etc for access */
-	unsigned a_text;	/* length of text, in bytes */
-	unsigned a_data;	/* length of data, in bytes */
-	unsigned a_bss;		/* length of uninitialized data area for
-				    file, in bytes */
-	unsigned a_syms;	/* length of symbol table data in file,
-				   in bytes */
-	unsigned a_entry;	/* start address */
-	unsigned a_trsize;	/* length of relocation info for text, in
-				    bytes */
-	unsigned a_drsize;	/* length of relocation info for data, in bytes */
-};
-
-#define N_TRSIZE(a)	((a).a_trsize)
-#define N_DRSIZE(a)	((a).a_drsize)
-#define N_SYMSIZE(a)	((a).a_syms)
-
-#endif /* _ASM_A_OUT_H */
diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index 22fc19bbe87f..ca2e4026ad20 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -22,7 +22,6 @@
 #include <linux/personality.h>
 #include <linux/sys.h>
 #include <linux/user.h>
-#include <linux/a.out.h>
 #include <linux/init.h>
 #include <linux/completion.h>
 #include <linux/kallsyms.h>
diff --git a/arch/mips/kernel/syscall.c b/arch/mips/kernel/syscall.c
index 343015a2f418..37970d9b2186 100644
--- a/arch/mips/kernel/syscall.c
+++ b/arch/mips/kernel/syscall.c
@@ -7,7 +7,6 @@
  * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
  * Copyright (C) 2001 MIPS Technologies, Inc.
  */
-#include <linux/a.out.h>
 #include <linux/capability.h>
 #include <linux/errno.h>
 #include <linux/linkage.h>
diff --git a/arch/mn10300/Kconfig b/arch/mn10300/Kconfig
index e856218da90d..dd557c9cf001 100644
--- a/arch/mn10300/Kconfig
+++ b/arch/mn10300/Kconfig
@@ -53,9 +53,6 @@ config QUICKLIST
 config ARCH_HAS_ILOG2_U32
 	def_bool y
 
-config ARCH_SUPPORTS_AOUT
-	def_bool n
-
 # Use the generic interrupt handling code in kernel/irq/
 config GENERIC_HARDIRQS
 	def_bool y
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index a7d4fd353c2b..8313fccced5e 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -76,9 +76,6 @@ config IRQ_PER_CPU
 	bool
 	default y
 
-config ARCH_SUPPORTS_AOUT
-	def_bool y
-
 # unless you want to implement ACPI on PA-RISC ... ;-)
 config PM
 	bool
diff --git a/arch/powerpc/include/asm/a.out.h b/arch/powerpc/include/asm/a.out.h
deleted file mode 100644
index 89cead6b176e..000000000000
--- a/arch/powerpc/include/asm/a.out.h
+++ /dev/null
@@ -1,20 +0,0 @@
-#ifndef _ASM_POWERPC_A_OUT_H
-#define _ASM_POWERPC_A_OUT_H
-
-struct exec
-{
-	unsigned long a_info;	/* Use macros N_MAGIC, etc for access */
-	unsigned a_text;	/* length of text, in bytes */
-	unsigned a_data;	/* length of data, in bytes */
-	unsigned a_bss;		/* length of uninitialized data area for file, in bytes */
-	unsigned a_syms;	/* length of symbol table data in file, in bytes */
-	unsigned a_entry;	/* start address */
-	unsigned a_trsize;	/* length of relocation info for text, in bytes */
-	unsigned a_drsize;	/* length of relocation info for data, in bytes */
-};
-
-#define N_TRSIZE(a)	((a).a_trsize)
-#define N_DRSIZE(a)	((a).a_drsize)
-#define N_SYMSIZE(a)	((a).a_syms)
-
-#endif /* _ASM_POWERPC_A_OUT_H */
diff --git a/arch/powerpc/include/asm/siginfo.h b/arch/powerpc/include/asm/siginfo.h
index 12f1bce037be..49495b0534ed 100644
--- a/arch/powerpc/include/asm/siginfo.h
+++ b/arch/powerpc/include/asm/siginfo.h
@@ -15,11 +15,6 @@
 
 #include <asm-generic/siginfo.h>
 
-/*
- * SIGTRAP si_codes
- */
-#define TRAP_BRANCH	(__SI_FAULT|3)	/* process taken branch trap */
-#define TRAP_HWBKPT	(__SI_FAULT|4)	/* hardware breakpoint or watchpoint */
 #undef NSIGTRAP
 #define NSIGTRAP	4
 
diff --git a/arch/powerpc/include/asm/statfs.h b/arch/powerpc/include/asm/statfs.h
index 67024026c10d..5244834583a4 100644
--- a/arch/powerpc/include/asm/statfs.h
+++ b/arch/powerpc/include/asm/statfs.h
@@ -1,60 +1,6 @@
 #ifndef _ASM_POWERPC_STATFS_H
 #define _ASM_POWERPC_STATFS_H
 
-/* For ppc32 we just use the generic definitions, not so simple on ppc64 */
-
-#ifndef __powerpc64__
 #include <asm-generic/statfs.h>
-#else
-
-#ifndef __KERNEL_STRICT_NAMES
-#include <linux/types.h>
-typedef __kernel_fsid_t	fsid_t;
-#endif
-
-/*
- * We're already 64-bit, so duplicate the definition
- */
-struct statfs {
-	long f_type;
-	long f_bsize;
-	long f_blocks;
-	long f_bfree;
-	long f_bavail;
-	long f_files;
-	long f_ffree;
-	__kernel_fsid_t f_fsid;
-	long f_namelen;
-	long f_frsize;
-	long f_spare[5];
-};
-
-struct statfs64 {
-	long f_type;
-	long f_bsize;
-	long f_blocks;
-	long f_bfree;
-	long f_bavail;
-	long f_files;
-	long f_ffree;
-	__kernel_fsid_t f_fsid;
-	long f_namelen;
-	long f_frsize;
-	long f_spare[5];
-};
 
-struct compat_statfs64 {
-	__u32 f_type;
-	__u32 f_bsize;
-	__u64 f_blocks;
-	__u64 f_bfree;
-	__u64 f_bavail;
-	__u64 f_files;
-	__u64 f_ffree;
-	__kernel_fsid_t f_fsid;
-	__u32 f_namelen;
-	__u32 f_frsize;
-	__u32 f_spare[5];
-};
-#endif /* ! __powerpc64__ */
 #endif
diff --git a/arch/powerpc/kernel/softemu8xx.c b/arch/powerpc/kernel/softemu8xx.c
index c906c4bf6835..23c8c5e7dc4d 100644
--- a/arch/powerpc/kernel/softemu8xx.c
+++ b/arch/powerpc/kernel/softemu8xx.c
@@ -23,7 +23,6 @@
 #include <linux/ptrace.h>
 #include <linux/slab.h>
 #include <linux/user.h>
-#include <linux/a.out.h>
 #include <linux/interrupt.h>
 
 #include <asm/pgtable.h>
diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c
index 81ccb8dd1a54..f5def6cf5cd6 100644
--- a/arch/powerpc/kernel/traps.c
+++ b/arch/powerpc/kernel/traps.c
@@ -23,7 +23,6 @@
 #include <linux/ptrace.h>
 #include <linux/slab.h>
 #include <linux/user.h>
-#include <linux/a.out.h>
 #include <linux/interrupt.h>
 #include <linux/init.h>
 #include <linux/module.h>
diff --git a/arch/powerpc/platforms/chrp/setup.c b/arch/powerpc/platforms/chrp/setup.c
index 1ba7ce5aafae..272d79a8d289 100644
--- a/arch/powerpc/platforms/chrp/setup.c
+++ b/arch/powerpc/platforms/chrp/setup.c
@@ -17,7 +17,6 @@
 #include <linux/ptrace.h>
 #include <linux/slab.h>
 #include <linux/user.h>
-#include <linux/a.out.h>
 #include <linux/tty.h>
 #include <linux/major.h>
 #include <linux/interrupt.h>
diff --git a/arch/powerpc/platforms/maple/setup.c b/arch/powerpc/platforms/maple/setup.c
index 364714757cf1..d4c61c3c9669 100644
--- a/arch/powerpc/platforms/maple/setup.c
+++ b/arch/powerpc/platforms/maple/setup.c
@@ -23,7 +23,6 @@
 #include <linux/ptrace.h>
 #include <linux/slab.h>
 #include <linux/user.h>
-#include <linux/a.out.h>
 #include <linux/tty.h>
 #include <linux/string.h>
 #include <linux/delay.h>
diff --git a/arch/powerpc/platforms/powermac/setup.c b/arch/powerpc/platforms/powermac/setup.c
index 88ccf3a08a9c..82c14d203d8b 100644
--- a/arch/powerpc/platforms/powermac/setup.c
+++ b/arch/powerpc/platforms/powermac/setup.c
@@ -33,7 +33,6 @@
 #include <linux/ptrace.h>
 #include <linux/slab.h>
 #include <linux/user.h>
-#include <linux/a.out.h>
 #include <linux/tty.h>
 #include <linux/string.h>
 #include <linux/delay.h>
diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
index 7b01d67b4e48..ec341707e41b 100644
--- a/arch/powerpc/platforms/pseries/setup.c
+++ b/arch/powerpc/platforms/pseries/setup.c
@@ -25,7 +25,6 @@
 #include <linux/unistd.h>
 #include <linux/slab.h>
 #include <linux/user.h>
-#include <linux/a.out.h>
 #include <linux/tty.h>
 #include <linux/major.h>
 #include <linux/interrupt.h>
diff --git a/arch/s390/include/asm/statfs.h b/arch/s390/include/asm/statfs.h
index 099a45579190..06cc70307ece 100644
--- a/arch/s390/include/asm/statfs.h
+++ b/arch/s390/include/asm/statfs.h
@@ -12,19 +12,16 @@
 #ifndef __s390x__
 #include <asm-generic/statfs.h>
 #else
+/*
+ * We can't use <asm-generic/statfs.h> because in 64-bit mode
+ * we mix ints of different sizes in our struct statfs.
+ */
 
 #ifndef __KERNEL_STRICT_NAMES
-
 #include <linux/types.h>
-
 typedef __kernel_fsid_t	fsid_t;
-
 #endif
 
-/*
- * This is ugly -- we're already 64-bit clean, so just duplicate the 
- * definitions.
- */
 struct statfs {
 	int  f_type;
 	int  f_bsize;
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index a214002114ed..97671dac12a6 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -20,6 +20,11 @@ config GENERIC_ISA_DMA
 	bool
 	default y
 
+config GENERIC_GPIO
+	bool
+	help
+	  Generic GPIO API support
+
 config ARCH_NO_VIRT_TO_BUS
 	def_bool y
 
@@ -69,6 +74,9 @@ config SPARC
 	select HAVE_OPROFILE
 	select HAVE_ARCH_KGDB if !SMP
 	select HAVE_ARCH_TRACEHOOK
+	select ARCH_WANT_OPTIONAL_GPIOLIB
+	select RTC_CLASS
+	select RTC_DRV_M48T59
 
 # Identify this as a Sparc32 build
 config SPARC32
@@ -204,17 +212,6 @@ config SUN_PM
 	  Enable power management and CPU standby features on supported
 	  SPARC platforms.
 
-config SUN4
-	bool "Support for SUN4 machines (disables SUN4[CDM] support)"
-	depends on !SMP
-	default n
-	help
-	  Say Y here if, and only if, your machine is a sun4. Note that
-	  a kernel compiled with this option will run only on sun4.
-	  (And the current version will probably work only on sun4/330.)
-
-if !SUN4
-
 config PCI
 	bool "Support for PCI and PS/2 keyboard/mouse"
 	help
@@ -227,11 +224,6 @@ config PCI_SYSCALL
 
 source "drivers/pci/Kconfig"
 
-endif
-
-config NO_DMA
-	def_bool !PCI
-
 config SUN_OPENPROMFS
 	tristate "Openprom tree appears in /proc/openprom"
 	help
@@ -263,9 +255,7 @@ source "net/Kconfig"
 
 source "drivers/Kconfig"
 
-if !SUN4
 source "drivers/sbus/char/Kconfig"
-endif
 
 # This one must be before the filesystem configs. -DaveM
 
diff --git a/arch/sparc/include/asm/Kbuild b/arch/sparc/include/asm/Kbuild
index a5f0ce734ff7..2d2769d766ec 100644
--- a/arch/sparc/include/asm/Kbuild
+++ b/arch/sparc/include/asm/Kbuild
@@ -15,14 +15,11 @@ header-y += signal_32.h
 header-y += signal_64.h
 header-y += stat_32.h
 header-y += stat_64.h
-header-y += statfs_32.h
-header-y += statfs_64.h
 header-y += unistd_32.h
 header-y += unistd_64.h
 
 header-y += apc.h
 header-y += asi.h
-header-y += bpp.h
 header-y += display7seg.h
 header-y += envctrl.h
 header-y += fbio.h
@@ -41,5 +38,4 @@ header-y += reg_64.h
 header-y += traps.h
 header-y += uctx.h
 header-y += utrap.h
-header-y += vfc_ioctls.h
 header-y += watchdog.h
diff --git a/arch/sparc/include/asm/asmmacro.h b/arch/sparc/include/asm/asmmacro.h
index a619a4d97aae..a995bf8aba3f 100644
--- a/arch/sparc/include/asm/asmmacro.h
+++ b/arch/sparc/include/asm/asmmacro.h
@@ -34,12 +34,7 @@
 /* sun4 probably wants half word accesses to ASI_SEGMAP, while sun4c+
    likes byte accesses. These are to avoid ifdef mania. */
 
-#ifdef CONFIG_SUN4
-#define lduXa	lduha
-#define stXa	stha
-#else
 #define lduXa	lduba
 #define stXa	stba
-#endif
 
 #endif /* !(_SPARC_ASMMACRO_H) */
diff --git a/arch/sparc/include/asm/bpp.h b/arch/sparc/include/asm/bpp.h
deleted file mode 100644
index 31f515e499a7..000000000000
--- a/arch/sparc/include/asm/bpp.h
+++ /dev/null
@@ -1,73 +0,0 @@
-#ifndef _SPARC_BPP_H
-#define _SPARC_BPP_H
-
-/*
- * Copyright (c) 1995 Picture Elements
- *	Stephen Williams
- *	Gus Baldauf
- *
- * Linux/SPARC port by Peter Zaitcev.
- * Integration into SPARC tree by Tom Dyas.
- */
-
-#include  <linux/ioctl.h>
-
-/*
- * This is a driver that supports IEEE Std 1284-1994 communications
- * with compliant or compatible devices. It will use whatever features
- * the device supports, prefering those that are typically faster.
- *
- * When the device is opened, it is left in COMPATIBILITY mode, and
- * writes work like any printer device. The driver only attempt to
- * negotiate 1284 modes when needed so that plugs can be pulled,
- * switch boxes switched, etc., without disrupting things. It will
- * also leave the device in compatibility mode when closed.
- */
-
-
-
-/*
- * This driver also supplies ioctls to manually manipulate the
- * pins. This is great for testing devices, or writing code to deal
- * with bizzarro-mode of the ACME Special TurboThingy Plus.
- *
- * NOTE: These ioctl currently do not interact well with
- * read/write. Caveat emptor.
- *
- * PUT_PINS allows us to assign the sense of all the pins, including
- * the data pins if being driven by the host. The GET_PINS returns the
- * pins that the peripheral drives, including data if appropriate.
- */
-
-# define BPP_PUT_PINS _IOW('B', 1, int)
-# define BPP_GET_PINS _IOR('B', 2, char) /* that's bogus - should've been _IO */
-# define BPP_PUT_DATA _IOW('B', 3, int)
-# define BPP_GET_DATA _IOR('B', 4, char) /* ditto */
-
-/*
- * Set the data bus to input mode. Disengage the data bin driver and
- * be prepared to read values from the peripheral. If the arg is 0,
- * then revert the bus to output mode.
- */
-# define BPP_SET_INPUT _IOW('B', 5, int)
-
-/*
- * These bits apply to the PUT operation...
- */
-# define BPP_PP_nStrobe   0x0001
-# define BPP_PP_nAutoFd   0x0002
-# define BPP_PP_nInit     0x0004
-# define BPP_PP_nSelectIn 0x0008
-
-/*
- * These apply to the GET operation, which also reads the current value
- * of the previously put values. A bit mask of these will be returned
- * as a bit mask in the return code of the ioctl().
- */
-# define BPP_GP_nAck   0x0100
-# define BPP_GP_Busy   0x0200
-# define BPP_GP_PError 0x0400
-# define BPP_GP_Select 0x0800
-# define BPP_GP_nFault 0x1000
-
-#endif
diff --git a/arch/sparc/include/asm/bugs.h b/arch/sparc/include/asm/bugs.h
index e179bc12f64a..61d86bbbe2b2 100644
--- a/arch/sparc/include/asm/bugs.h
+++ b/arch/sparc/include/asm/bugs.h
@@ -7,10 +7,6 @@
 #include <asm/cpudata.h>
 #endif
 
-#ifdef CONFIG_SPARC64
-#include <asm/sstate.h>
-#endif
-
 extern unsigned long loops_per_jiffy;
 
 static void __init check_bugs(void)
@@ -18,7 +14,4 @@ static void __init check_bugs(void)
 #if defined(CONFIG_SPARC32) && !defined(CONFIG_SMP)
 	cpu_data(0).udelay_val = loops_per_jiffy;
 #endif
-#ifdef CONFIG_SPARC64
-	sstate_running();
-#endif
 }
diff --git a/arch/sparc/include/asm/cpudata_64.h b/arch/sparc/include/asm/cpudata_64.h
index 532975ecfe10..7da7c13d23c4 100644
--- a/arch/sparc/include/asm/cpudata_64.h
+++ b/arch/sparc/include/asm/cpudata_64.h
@@ -86,7 +86,6 @@ extern struct trap_per_cpu trap_block[NR_CPUS];
 extern void init_cur_cpu_trap(struct thread_info *);
 extern void setup_tba(void);
 extern int ncpus_probed;
-extern void __init cpu_probe(void);
 extern const struct seq_operations cpuinfo_op;
 
 extern unsigned long real_hard_smp_processor_id(void);
diff --git a/arch/sparc/include/asm/dma-mapping_32.h b/arch/sparc/include/asm/dma-mapping_32.h
index f3a641e6b2c8..8a57ea0573e6 100644
--- a/arch/sparc/include/asm/dma-mapping_32.h
+++ b/arch/sparc/include/asm/dma-mapping_32.h
@@ -1,11 +1,60 @@
 #ifndef _ASM_SPARC_DMA_MAPPING_H
 #define _ASM_SPARC_DMA_MAPPING_H
 
+#include <linux/types.h>
 
-#ifdef CONFIG_PCI
-#include <asm-generic/dma-mapping.h>
-#else
-#include <asm-generic/dma-mapping-broken.h>
-#endif /* PCI */
+struct device;
+struct scatterlist;
+struct page;
+
+#define DMA_ERROR_CODE	(~(dma_addr_t)0x0)
+
+extern int dma_supported(struct device *dev, u64 mask);
+extern int dma_set_mask(struct device *dev, u64 dma_mask);
+extern void *dma_alloc_coherent(struct device *dev, size_t size,
+				dma_addr_t *dma_handle, gfp_t flag);
+extern void dma_free_coherent(struct device *dev, size_t size,
+			      void *cpu_addr, dma_addr_t dma_handle);
+extern dma_addr_t dma_map_single(struct device *dev, void *cpu_addr,
+				 size_t size,
+				 enum dma_data_direction direction);
+extern void dma_unmap_single(struct device *dev, dma_addr_t dma_addr,
+			     size_t size,
+			     enum dma_data_direction direction);
+extern dma_addr_t dma_map_page(struct device *dev, struct page *page,
+			       unsigned long offset, size_t size,
+			       enum dma_data_direction direction);
+extern void dma_unmap_page(struct device *dev, dma_addr_t dma_address,
+			   size_t size, enum dma_data_direction direction);
+extern int dma_map_sg(struct device *dev, struct scatterlist *sg,
+		      int nents, enum dma_data_direction direction);
+extern void dma_unmap_sg(struct device *dev, struct scatterlist *sg,
+			 int nents, enum dma_data_direction direction);
+extern void dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
+				    size_t size,
+				    enum dma_data_direction direction);
+extern void dma_sync_single_for_device(struct device *dev,
+				       dma_addr_t dma_handle,
+				       size_t size,
+				       enum dma_data_direction direction);
+extern void dma_sync_single_range_for_cpu(struct device *dev,
+					  dma_addr_t dma_handle,
+					  unsigned long offset,
+					  size_t size,
+					  enum dma_data_direction direction);
+extern void dma_sync_single_range_for_device(struct device *dev,
+					     dma_addr_t dma_handle,
+					     unsigned long offset, size_t size,
+					     enum dma_data_direction direction);
+extern void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
+				int nelems, enum dma_data_direction direction);
+extern void dma_sync_sg_for_device(struct device *dev,
+				   struct scatterlist *sg, int nelems,
+				   enum dma_data_direction direction);
+extern int dma_mapping_error(struct device *dev, dma_addr_t dma_addr);
+extern int dma_get_cache_alignment(void);
+
+#define dma_alloc_noncoherent	dma_alloc_coherent
+#define dma_free_noncoherent	dma_free_coherent
 
 #endif /* _ASM_SPARC_DMA_MAPPING_H */
diff --git a/arch/sparc/include/asm/dma.h b/arch/sparc/include/asm/dma.h
index aa1d90ac04c5..b554927bbaf6 100644
--- a/arch/sparc/include/asm/dma.h
+++ b/arch/sparc/include/asm/dma.h
@@ -1,8 +1,139 @@
-#ifndef ___ASM_SPARC_DMA_H
-#define ___ASM_SPARC_DMA_H
-#if defined(__sparc__) && defined(__arch64__)
-#include <asm/dma_64.h>
+#ifndef _ASM_SPARC_DMA_H
+#define _ASM_SPARC_DMA_H
+
+/* These are irrelevant for Sparc DMA, but we leave it in so that
+ * things can compile.
+ */
+#define MAX_DMA_CHANNELS 8
+#define DMA_MODE_READ    1
+#define DMA_MODE_WRITE   2
+#define MAX_DMA_ADDRESS  (~0UL)
+
+/* Useful constants */
+#define SIZE_16MB      (16*1024*1024)
+#define SIZE_64K       (64*1024)
+
+/* SBUS DMA controller reg offsets */
+#define DMA_CSR		0x00UL		/* rw  DMA control/status register    0x00   */
+#define DMA_ADDR	0x04UL		/* rw  DMA transfer address register  0x04   */
+#define DMA_COUNT	0x08UL		/* rw  DMA transfer count register    0x08   */
+#define DMA_TEST	0x0cUL		/* rw  DMA test/debug register        0x0c   */
+
+/* Fields in the cond_reg register */
+/* First, the version identification bits */
+#define DMA_DEVICE_ID    0xf0000000        /* Device identification bits */
+#define DMA_VERS0        0x00000000        /* Sunray DMA version */
+#define DMA_ESCV1        0x40000000        /* DMA ESC Version 1 */
+#define DMA_VERS1        0x80000000        /* DMA rev 1 */
+#define DMA_VERS2        0xa0000000        /* DMA rev 2 */
+#define DMA_VERHME       0xb0000000        /* DMA hme gate array */
+#define DMA_VERSPLUS     0x90000000        /* DMA rev 1 PLUS */
+
+#define DMA_HNDL_INTR    0x00000001        /* An IRQ needs to be handled */
+#define DMA_HNDL_ERROR   0x00000002        /* We need to take an error */
+#define DMA_FIFO_ISDRAIN 0x0000000c        /* The DMA FIFO is draining */
+#define DMA_INT_ENAB     0x00000010        /* Turn on interrupts */
+#define DMA_FIFO_INV     0x00000020        /* Invalidate the FIFO */
+#define DMA_ACC_SZ_ERR   0x00000040        /* The access size was bad */
+#define DMA_FIFO_STDRAIN 0x00000040        /* DMA_VERS1 Drain the FIFO */
+#define DMA_RST_SCSI     0x00000080        /* Reset the SCSI controller */
+#define DMA_RST_ENET     DMA_RST_SCSI      /* Reset the ENET controller */
+#define DMA_ST_WRITE     0x00000100        /* write from device to memory */
+#define DMA_ENABLE       0x00000200        /* Fire up DMA, handle requests */
+#define DMA_PEND_READ    0x00000400        /* DMA_VERS1/0/PLUS Pending Read */
+#define DMA_ESC_BURST    0x00000800        /* 1=16byte 0=32byte */
+#define DMA_READ_AHEAD   0x00001800        /* DMA read ahead partial longword */
+#define DMA_DSBL_RD_DRN  0x00001000        /* No EC drain on slave reads */
+#define DMA_BCNT_ENAB    0x00002000        /* If on, use the byte counter */
+#define DMA_TERM_CNTR    0x00004000        /* Terminal counter */
+#define DMA_SCSI_SBUS64  0x00008000        /* HME: Enable 64-bit SBUS mode. */
+#define DMA_CSR_DISAB    0x00010000        /* No FIFO drains during csr */
+#define DMA_SCSI_DISAB   0x00020000        /* No FIFO drains during reg */
+#define DMA_DSBL_WR_INV  0x00020000        /* No EC inval. on slave writes */
+#define DMA_ADD_ENABLE   0x00040000        /* Special ESC DVMA optimization */
+#define DMA_E_BURSTS	 0x000c0000	   /* ENET: SBUS r/w burst mask */
+#define DMA_E_BURST32	 0x00040000	   /* ENET: SBUS 32 byte r/w burst */
+#define DMA_E_BURST16	 0x00000000	   /* ENET: SBUS 16 byte r/w burst */
+#define DMA_BRST_SZ      0x000c0000        /* SCSI: SBUS r/w burst size */
+#define DMA_BRST64       0x000c0000        /* SCSI: 64byte bursts (HME on UltraSparc only) */
+#define DMA_BRST32       0x00040000        /* SCSI: 32byte bursts */
+#define DMA_BRST16       0x00000000        /* SCSI: 16byte bursts */
+#define DMA_BRST0        0x00080000        /* SCSI: no bursts (non-HME gate arrays) */
+#define DMA_ADDR_DISAB   0x00100000        /* No FIFO drains during addr */
+#define DMA_2CLKS        0x00200000        /* Each transfer = 2 clock ticks */
+#define DMA_3CLKS        0x00400000        /* Each transfer = 3 clock ticks */
+#define DMA_EN_ENETAUI   DMA_3CLKS         /* Put lance into AUI-cable mode */
+#define DMA_CNTR_DISAB   0x00800000        /* No IRQ when DMA_TERM_CNTR set */
+#define DMA_AUTO_NADDR   0x01000000        /* Use "auto nxt addr" feature */
+#define DMA_SCSI_ON      0x02000000        /* Enable SCSI dma */
+#define DMA_PARITY_OFF   0x02000000        /* HME: disable parity checking */
+#define DMA_LOADED_ADDR  0x04000000        /* Address has been loaded */
+#define DMA_LOADED_NADDR 0x08000000        /* Next address has been loaded */
+#define DMA_RESET_FAS366 0x08000000        /* HME: Assert RESET to FAS366 */
+
+/* Values describing the burst-size property from the PROM */
+#define DMA_BURST1       0x01
+#define DMA_BURST2       0x02
+#define DMA_BURST4       0x04
+#define DMA_BURST8       0x08
+#define DMA_BURST16      0x10
+#define DMA_BURST32      0x20
+#define DMA_BURST64      0x40
+#define DMA_BURSTBITS    0x7f
+
+/* From PCI */
+
+#ifdef CONFIG_PCI
+extern int isa_dma_bridge_buggy;
 #else
-#include <asm/dma_32.h>
+#define isa_dma_bridge_buggy 	(0)
 #endif
+
+#ifdef CONFIG_SPARC32
+
+/* Routines for data transfer buffers. */
+BTFIXUPDEF_CALL(char *, mmu_lockarea, char *, unsigned long)
+BTFIXUPDEF_CALL(void,   mmu_unlockarea, char *, unsigned long)
+
+#define mmu_lockarea(vaddr,len) BTFIXUP_CALL(mmu_lockarea)(vaddr,len)
+#define mmu_unlockarea(vaddr,len) BTFIXUP_CALL(mmu_unlockarea)(vaddr,len)
+
+struct page;
+struct device;
+struct scatterlist;
+
+/* These are implementations for sbus_map_sg/sbus_unmap_sg... collapse later */
+BTFIXUPDEF_CALL(__u32, mmu_get_scsi_one, struct device *, char *, unsigned long)
+BTFIXUPDEF_CALL(void,  mmu_get_scsi_sgl, struct device *, struct scatterlist *, int)
+BTFIXUPDEF_CALL(void,  mmu_release_scsi_one, struct device *, __u32, unsigned long)
+BTFIXUPDEF_CALL(void,  mmu_release_scsi_sgl, struct device *, struct scatterlist *, int)
+
+#define mmu_get_scsi_one(dev,vaddr,len) BTFIXUP_CALL(mmu_get_scsi_one)(dev,vaddr,len)
+#define mmu_get_scsi_sgl(dev,sg,sz) BTFIXUP_CALL(mmu_get_scsi_sgl)(dev,sg,sz)
+#define mmu_release_scsi_one(dev,vaddr,len) BTFIXUP_CALL(mmu_release_scsi_one)(dev,vaddr,len)
+#define mmu_release_scsi_sgl(dev,sg,sz) BTFIXUP_CALL(mmu_release_scsi_sgl)(dev,sg,sz)
+
+/*
+ * mmu_map/unmap are provided by iommu/iounit; Invalid to call on IIep.
+ *
+ * The mmu_map_dma_area establishes two mappings in one go.
+ * These mappings point to pages normally mapped at 'va' (linear address).
+ * First mapping is for CPU visible address at 'a', uncached.
+ * This is an alias, but it works because it is an uncached mapping.
+ * Second mapping is for device visible address, or "bus" address.
+ * The bus address is returned at '*pba'.
+ *
+ * These functions seem distinct, but are hard to split. On sun4c,
+ * at least for now, 'a' is equal to bus address, and retured in *pba.
+ * On sun4m, page attributes depend on the CPU type, so we have to
+ * know if we are mapping RAM or I/O, so it has to be an additional argument
+ * to a separate mapping function for CPU visible mappings.
+ */
+BTFIXUPDEF_CALL(int, mmu_map_dma_area, struct device *, dma_addr_t *, unsigned long, unsigned long, int len)
+BTFIXUPDEF_CALL(void, mmu_unmap_dma_area, struct device *, unsigned long busa, int len)
+
+#define mmu_map_dma_area(dev,pba,va,a,len) BTFIXUP_CALL(mmu_map_dma_area)(dev,pba,va,a,len)
+#define mmu_unmap_dma_area(dev,ba,len) BTFIXUP_CALL(mmu_unmap_dma_area)(dev,ba,len)
 #endif
+
+#endif /* !(_ASM_SPARC_DMA_H) */
diff --git a/arch/sparc/include/asm/dma_32.h b/arch/sparc/include/asm/dma_32.h
deleted file mode 100644
index cf7189c0079b..000000000000
--- a/arch/sparc/include/asm/dma_32.h
+++ /dev/null
@@ -1,288 +0,0 @@
-/* include/asm/dma.h
- *
- * Copyright 1995 (C) David S. Miller (davem@davemloft.net)
- */
-
-#ifndef _ASM_SPARC_DMA_H
-#define _ASM_SPARC_DMA_H
-
-#include <linux/kernel.h>
-#include <linux/types.h>
-
-#include <asm/vac-ops.h>  /* for invalidate's, etc. */
-#include <asm/sbus.h>
-#include <asm/delay.h>
-#include <asm/oplib.h>
-#include <asm/system.h>
-#include <asm/io.h>
-#include <linux/spinlock.h>
-
-struct page;
-extern spinlock_t  dma_spin_lock;
-
-static inline unsigned long claim_dma_lock(void)
-{
-	unsigned long flags;
-	spin_lock_irqsave(&dma_spin_lock, flags);
-	return flags;
-}
-
-static inline void release_dma_lock(unsigned long flags)
-{
-	spin_unlock_irqrestore(&dma_spin_lock, flags);
-}
-
-/* These are irrelevant for Sparc DMA, but we leave it in so that
- * things can compile.
- */
-#define MAX_DMA_CHANNELS 8
-#define MAX_DMA_ADDRESS  (~0UL)
-#define DMA_MODE_READ    1
-#define DMA_MODE_WRITE   2
-
-/* Useful constants */
-#define SIZE_16MB      (16*1024*1024)
-#define SIZE_64K       (64*1024)
-
-/* SBUS DMA controller reg offsets */
-#define DMA_CSR		0x00UL		/* rw  DMA control/status register    0x00   */
-#define DMA_ADDR	0x04UL		/* rw  DMA transfer address register  0x04   */
-#define DMA_COUNT	0x08UL		/* rw  DMA transfer count register    0x08   */
-#define DMA_TEST	0x0cUL		/* rw  DMA test/debug register        0x0c   */
-
-/* DVMA chip revisions */
-enum dvma_rev {
-	dvmarev0,
-	dvmaesc1,
-	dvmarev1,
-	dvmarev2,
-	dvmarev3,
-	dvmarevplus,
-	dvmahme
-};
-
-#define DMA_HASCOUNT(rev)  ((rev)==dvmaesc1)
-
-/* Linux DMA information structure, filled during probe. */
-struct sbus_dma {
-	struct sbus_dma *next;
-	struct sbus_dev *sdev;
-	void __iomem *regs;
-
-	/* Status, misc info */
-	int node;                /* Prom node for this DMA device */
-	int running;             /* Are we doing DMA now? */
-	int allocated;           /* Are we "owned" by anyone yet? */
-
-	/* Transfer information. */
-	unsigned long addr;      /* Start address of current transfer */
-	int nbytes;              /* Size of current transfer */
-	int realbytes;           /* For splitting up large transfers, etc. */
-
-	/* DMA revision */
-	enum dvma_rev revision;
-};
-
-extern struct sbus_dma *dma_chain;
-
-/* Broken hardware... */
-#ifdef CONFIG_SUN4
-/* Have to sort this out. Does rev0 work fine on sun4[cmd] without isbroken?
- * Or is rev0 present only on sun4 boxes? -jj */
-#define DMA_ISBROKEN(dma)    ((dma)->revision == dvmarev0 || (dma)->revision == dvmarev1)
-#else
-#define DMA_ISBROKEN(dma)    ((dma)->revision == dvmarev1)
-#endif
-#define DMA_ISESC1(dma)      ((dma)->revision == dvmaesc1)
-
-/* Main routines in dma.c */
-extern void dvma_init(struct sbus_bus *);
-
-/* Fields in the cond_reg register */
-/* First, the version identification bits */
-#define DMA_DEVICE_ID    0xf0000000        /* Device identification bits */
-#define DMA_VERS0        0x00000000        /* Sunray DMA version */
-#define DMA_ESCV1        0x40000000        /* DMA ESC Version 1 */
-#define DMA_VERS1        0x80000000        /* DMA rev 1 */
-#define DMA_VERS2        0xa0000000        /* DMA rev 2 */
-#define DMA_VERHME       0xb0000000        /* DMA hme gate array */
-#define DMA_VERSPLUS     0x90000000        /* DMA rev 1 PLUS */
-
-#define DMA_HNDL_INTR    0x00000001        /* An IRQ needs to be handled */
-#define DMA_HNDL_ERROR   0x00000002        /* We need to take an error */
-#define DMA_FIFO_ISDRAIN 0x0000000c        /* The DMA FIFO is draining */
-#define DMA_INT_ENAB     0x00000010        /* Turn on interrupts */
-#define DMA_FIFO_INV     0x00000020        /* Invalidate the FIFO */
-#define DMA_ACC_SZ_ERR   0x00000040        /* The access size was bad */
-#define DMA_FIFO_STDRAIN 0x00000040        /* DMA_VERS1 Drain the FIFO */
-#define DMA_RST_SCSI     0x00000080        /* Reset the SCSI controller */
-#define DMA_RST_ENET     DMA_RST_SCSI      /* Reset the ENET controller */
-#define DMA_RST_BPP      DMA_RST_SCSI      /* Reset the BPP controller */
-#define DMA_ST_WRITE     0x00000100        /* write from device to memory */
-#define DMA_ENABLE       0x00000200        /* Fire up DMA, handle requests */
-#define DMA_PEND_READ    0x00000400        /* DMA_VERS1/0/PLUS Pending Read */
-#define DMA_ESC_BURST    0x00000800        /* 1=16byte 0=32byte */
-#define DMA_READ_AHEAD   0x00001800        /* DMA read ahead partial longword */
-#define DMA_DSBL_RD_DRN  0x00001000        /* No EC drain on slave reads */
-#define DMA_BCNT_ENAB    0x00002000        /* If on, use the byte counter */
-#define DMA_TERM_CNTR    0x00004000        /* Terminal counter */
-#define DMA_SCSI_SBUS64  0x00008000        /* HME: Enable 64-bit SBUS mode. */
-#define DMA_CSR_DISAB    0x00010000        /* No FIFO drains during csr */
-#define DMA_SCSI_DISAB   0x00020000        /* No FIFO drains during reg */
-#define DMA_DSBL_WR_INV  0x00020000        /* No EC inval. on slave writes */
-#define DMA_ADD_ENABLE   0x00040000        /* Special ESC DVMA optimization */
-#define DMA_E_BURSTS	 0x000c0000	   /* ENET: SBUS r/w burst mask */
-#define DMA_E_BURST32	 0x00040000	   /* ENET: SBUS 32 byte r/w burst */
-#define DMA_E_BURST16	 0x00000000	   /* ENET: SBUS 16 byte r/w burst */
-#define DMA_BRST_SZ      0x000c0000        /* SCSI: SBUS r/w burst size */
-#define DMA_BRST64       0x00080000        /* SCSI: 64byte bursts (HME on UltraSparc only) */
-#define DMA_BRST32       0x00040000        /* SCSI/BPP: 32byte bursts */
-#define DMA_BRST16       0x00000000        /* SCSI/BPP: 16byte bursts */
-#define DMA_BRST0        0x00080000        /* SCSI: no bursts (non-HME gate arrays) */
-#define DMA_ADDR_DISAB   0x00100000        /* No FIFO drains during addr */
-#define DMA_2CLKS        0x00200000        /* Each transfer = 2 clock ticks */
-#define DMA_3CLKS        0x00400000        /* Each transfer = 3 clock ticks */
-#define DMA_EN_ENETAUI   DMA_3CLKS         /* Put lance into AUI-cable mode */
-#define DMA_CNTR_DISAB   0x00800000        /* No IRQ when DMA_TERM_CNTR set */
-#define DMA_AUTO_NADDR   0x01000000        /* Use "auto nxt addr" feature */
-#define DMA_SCSI_ON      0x02000000        /* Enable SCSI dma */
-#define DMA_BPP_ON       DMA_SCSI_ON       /* Enable BPP dma */
-#define DMA_PARITY_OFF   0x02000000        /* HME: disable parity checking */
-#define DMA_LOADED_ADDR  0x04000000        /* Address has been loaded */
-#define DMA_LOADED_NADDR 0x08000000        /* Next address has been loaded */
-#define DMA_RESET_FAS366 0x08000000        /* HME: Assert RESET to FAS366 */
-
-/* Values describing the burst-size property from the PROM */
-#define DMA_BURST1       0x01
-#define DMA_BURST2       0x02
-#define DMA_BURST4       0x04
-#define DMA_BURST8       0x08
-#define DMA_BURST16      0x10
-#define DMA_BURST32      0x20
-#define DMA_BURST64      0x40
-#define DMA_BURSTBITS    0x7f
-
-/* Determine highest possible final transfer address given a base */
-#define DMA_MAXEND(addr) (0x01000000UL-(((unsigned long)(addr))&0x00ffffffUL))
-
-/* Yes, I hack a lot of elisp in my spare time... */
-#define DMA_ERROR_P(regs)  ((((regs)->cond_reg) & DMA_HNDL_ERROR))
-#define DMA_IRQ_P(regs)    ((((regs)->cond_reg) & (DMA_HNDL_INTR | DMA_HNDL_ERROR)))
-#define DMA_WRITE_P(regs)  ((((regs)->cond_reg) & DMA_ST_WRITE))
-#define DMA_OFF(regs)      ((((regs)->cond_reg) &= (~DMA_ENABLE)))
-#define DMA_INTSOFF(regs)  ((((regs)->cond_reg) &= (~DMA_INT_ENAB)))
-#define DMA_INTSON(regs)   ((((regs)->cond_reg) |= (DMA_INT_ENAB)))
-#define DMA_PUNTFIFO(regs) ((((regs)->cond_reg) |= DMA_FIFO_INV))
-#define DMA_SETSTART(regs, addr)  ((((regs)->st_addr) = (char *) addr))
-#define DMA_BEGINDMA_W(regs) \
-        ((((regs)->cond_reg |= (DMA_ST_WRITE|DMA_ENABLE|DMA_INT_ENAB))))
-#define DMA_BEGINDMA_R(regs) \
-        ((((regs)->cond_reg |= ((DMA_ENABLE|DMA_INT_ENAB)&(~DMA_ST_WRITE)))))
-
-/* For certain DMA chips, we need to disable ints upon irq entry
- * and turn them back on when we are done.  So in any ESP interrupt
- * handler you *must* call DMA_IRQ_ENTRY upon entry and DMA_IRQ_EXIT
- * when leaving the handler.  You have been warned...
- */
-#define DMA_IRQ_ENTRY(dma, dregs) do { \
-        if(DMA_ISBROKEN(dma)) DMA_INTSOFF(dregs); \
-   } while (0)
-
-#define DMA_IRQ_EXIT(dma, dregs) do { \
-	if(DMA_ISBROKEN(dma)) DMA_INTSON(dregs); \
-   } while(0)
-
-#if 0	/* P3 this stuff is inline in ledma.c:init_restart_ledma() */
-/* Pause until counter runs out or BIT isn't set in the DMA condition
- * register.
- */
-static inline void sparc_dma_pause(struct sparc_dma_registers *regs,
-				       unsigned long bit)
-{
-	int ctr = 50000;   /* Let's find some bugs ;) */
-
-	/* Busy wait until the bit is not set any more */
-	while((regs->cond_reg&bit) && (ctr>0)) {
-		ctr--;
-		__delay(5);
-	}
-
-	/* Check for bogus outcome. */
-	if(!ctr)
-		panic("DMA timeout");
-}
-
-/* Reset the friggin' thing... */
-#define DMA_RESET(dma) do { \
-	struct sparc_dma_registers *regs = dma->regs;                      \
-	/* Let the current FIFO drain itself */                            \
-	sparc_dma_pause(regs, (DMA_FIFO_ISDRAIN));                         \
-	/* Reset the logic */                                              \
-	regs->cond_reg |= (DMA_RST_SCSI);     /* assert */                 \
-	__delay(400);                         /* let the bits set ;) */    \
-	regs->cond_reg &= ~(DMA_RST_SCSI);    /* de-assert */              \
-	sparc_dma_enable_interrupts(regs);    /* Re-enable interrupts */   \
-	/* Enable FAST transfers if available */                           \
-	if(dma->revision>dvmarev1) regs->cond_reg |= DMA_3CLKS;            \
-	dma->running = 0;                                                  \
-} while(0)
-#endif
-
-#define for_each_dvma(dma) \
-        for((dma) = dma_chain; (dma); (dma) = (dma)->next)
-
-extern int get_dma_list(char *);
-extern int request_dma(unsigned int, __const__ char *);
-extern void free_dma(unsigned int);
-
-/* From PCI */
-
-#ifdef CONFIG_PCI
-extern int isa_dma_bridge_buggy;
-#else
-#define isa_dma_bridge_buggy	(0)
-#endif
-
-/* Routines for data transfer buffers. */
-BTFIXUPDEF_CALL(char *, mmu_lockarea, char *, unsigned long)
-BTFIXUPDEF_CALL(void,   mmu_unlockarea, char *, unsigned long)
-
-#define mmu_lockarea(vaddr,len) BTFIXUP_CALL(mmu_lockarea)(vaddr,len)
-#define mmu_unlockarea(vaddr,len) BTFIXUP_CALL(mmu_unlockarea)(vaddr,len)
-
-/* These are implementations for sbus_map_sg/sbus_unmap_sg... collapse later */
-BTFIXUPDEF_CALL(__u32, mmu_get_scsi_one, char *, unsigned long, struct sbus_bus *sbus)
-BTFIXUPDEF_CALL(void,  mmu_get_scsi_sgl, struct scatterlist *, int, struct sbus_bus *sbus)
-BTFIXUPDEF_CALL(void,  mmu_release_scsi_one, __u32, unsigned long, struct sbus_bus *sbus)
-BTFIXUPDEF_CALL(void,  mmu_release_scsi_sgl, struct scatterlist *, int, struct sbus_bus *sbus)
-
-#define mmu_get_scsi_one(vaddr,len,sbus) BTFIXUP_CALL(mmu_get_scsi_one)(vaddr,len,sbus)
-#define mmu_get_scsi_sgl(sg,sz,sbus) BTFIXUP_CALL(mmu_get_scsi_sgl)(sg,sz,sbus)
-#define mmu_release_scsi_one(vaddr,len,sbus) BTFIXUP_CALL(mmu_release_scsi_one)(vaddr,len,sbus)
-#define mmu_release_scsi_sgl(sg,sz,sbus) BTFIXUP_CALL(mmu_release_scsi_sgl)(sg,sz,sbus)
-
-/*
- * mmu_map/unmap are provided by iommu/iounit; Invalid to call on IIep.
- *
- * The mmu_map_dma_area establishes two mappings in one go.
- * These mappings point to pages normally mapped at 'va' (linear address).
- * First mapping is for CPU visible address at 'a', uncached.
- * This is an alias, but it works because it is an uncached mapping.
- * Second mapping is for device visible address, or "bus" address.
- * The bus address is returned at '*pba'.
- *
- * These functions seem distinct, but are hard to split. On sun4c,
- * at least for now, 'a' is equal to bus address, and retured in *pba.
- * On sun4m, page attributes depend on the CPU type, so we have to
- * know if we are mapping RAM or I/O, so it has to be an additional argument
- * to a separate mapping function for CPU visible mappings.
- */
-BTFIXUPDEF_CALL(int,  mmu_map_dma_area, dma_addr_t *, unsigned long, unsigned long, int len)
-BTFIXUPDEF_CALL(struct page *, mmu_translate_dvma, unsigned long busa)
-BTFIXUPDEF_CALL(void,  mmu_unmap_dma_area, unsigned long busa, int len)
-
-#define mmu_map_dma_area(pba,va,a,len) BTFIXUP_CALL(mmu_map_dma_area)(pba,va,a,len)
-#define mmu_unmap_dma_area(ba,len) BTFIXUP_CALL(mmu_unmap_dma_area)(ba,len)
-#define mmu_translate_dvma(ba)     BTFIXUP_CALL(mmu_translate_dvma)(ba)
-
-#endif /* !(_ASM_SPARC_DMA_H) */
diff --git a/arch/sparc/include/asm/dma_64.h b/arch/sparc/include/asm/dma_64.h
deleted file mode 100644
index 46a8aecffc02..000000000000
--- a/arch/sparc/include/asm/dma_64.h
+++ /dev/null
@@ -1,205 +0,0 @@
-/*
- * include/asm/dma.h
- *
- * Copyright 1996 (C) David S. Miller (davem@caip.rutgers.edu)
- */
-
-#ifndef _ASM_SPARC64_DMA_H
-#define _ASM_SPARC64_DMA_H
-
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/spinlock.h>
-
-#include <asm/sbus.h>
-#include <asm/delay.h>
-#include <asm/oplib.h>
-
-/* These are irrelevant for Sparc DMA, but we leave it in so that
- * things can compile.
- */
-#define MAX_DMA_CHANNELS 8
-#define DMA_MODE_READ    1
-#define DMA_MODE_WRITE   2
-#define MAX_DMA_ADDRESS  (~0UL)
-
-/* Useful constants */
-#define SIZE_16MB      (16*1024*1024)
-#define SIZE_64K       (64*1024)
-
-/* SBUS DMA controller reg offsets */
-#define DMA_CSR		0x00UL		/* rw  DMA control/status register    0x00   */
-#define DMA_ADDR	0x04UL		/* rw  DMA transfer address register  0x04   */
-#define DMA_COUNT	0x08UL		/* rw  DMA transfer count register    0x08   */
-#define DMA_TEST	0x0cUL		/* rw  DMA test/debug register        0x0c   */
-
-/* DVMA chip revisions */
-enum dvma_rev {
-	dvmarev0,
-	dvmaesc1,
-	dvmarev1,
-	dvmarev2,
-	dvmarev3,
-	dvmarevplus,
-	dvmahme
-};
-
-#define DMA_HASCOUNT(rev)  ((rev)==dvmaesc1)
-
-/* Linux DMA information structure, filled during probe. */
-struct sbus_dma {
-	struct sbus_dma *next;
-	struct sbus_dev *sdev;
-	void __iomem *regs;
-
-	/* Status, misc info */
-	int node;                /* Prom node for this DMA device */
-	int running;             /* Are we doing DMA now? */
-	int allocated;           /* Are we "owned" by anyone yet? */
-
-	/* Transfer information. */
-	u32 addr;                /* Start address of current transfer */
-	int nbytes;              /* Size of current transfer */
-	int realbytes;           /* For splitting up large transfers, etc. */
-
-	/* DMA revision */
-	enum dvma_rev revision;
-};
-
-extern struct sbus_dma *dma_chain;
-
-/* Broken hardware... */
-#define DMA_ISBROKEN(dma)    ((dma)->revision == dvmarev1)
-#define DMA_ISESC1(dma)      ((dma)->revision == dvmaesc1)
-
-/* Main routines in dma.c */
-extern void dvma_init(struct sbus_bus *);
-
-/* Fields in the cond_reg register */
-/* First, the version identification bits */
-#define DMA_DEVICE_ID    0xf0000000        /* Device identification bits */
-#define DMA_VERS0        0x00000000        /* Sunray DMA version */
-#define DMA_ESCV1        0x40000000        /* DMA ESC Version 1 */
-#define DMA_VERS1        0x80000000        /* DMA rev 1 */
-#define DMA_VERS2        0xa0000000        /* DMA rev 2 */
-#define DMA_VERHME       0xb0000000        /* DMA hme gate array */
-#define DMA_VERSPLUS     0x90000000        /* DMA rev 1 PLUS */
-
-#define DMA_HNDL_INTR    0x00000001        /* An IRQ needs to be handled */
-#define DMA_HNDL_ERROR   0x00000002        /* We need to take an error */
-#define DMA_FIFO_ISDRAIN 0x0000000c        /* The DMA FIFO is draining */
-#define DMA_INT_ENAB     0x00000010        /* Turn on interrupts */
-#define DMA_FIFO_INV     0x00000020        /* Invalidate the FIFO */
-#define DMA_ACC_SZ_ERR   0x00000040        /* The access size was bad */
-#define DMA_FIFO_STDRAIN 0x00000040        /* DMA_VERS1 Drain the FIFO */
-#define DMA_RST_SCSI     0x00000080        /* Reset the SCSI controller */
-#define DMA_RST_ENET     DMA_RST_SCSI      /* Reset the ENET controller */
-#define DMA_ST_WRITE     0x00000100        /* write from device to memory */
-#define DMA_ENABLE       0x00000200        /* Fire up DMA, handle requests */
-#define DMA_PEND_READ    0x00000400        /* DMA_VERS1/0/PLUS Pending Read */
-#define DMA_ESC_BURST    0x00000800        /* 1=16byte 0=32byte */
-#define DMA_READ_AHEAD   0x00001800        /* DMA read ahead partial longword */
-#define DMA_DSBL_RD_DRN  0x00001000        /* No EC drain on slave reads */
-#define DMA_BCNT_ENAB    0x00002000        /* If on, use the byte counter */
-#define DMA_TERM_CNTR    0x00004000        /* Terminal counter */
-#define DMA_SCSI_SBUS64  0x00008000        /* HME: Enable 64-bit SBUS mode. */
-#define DMA_CSR_DISAB    0x00010000        /* No FIFO drains during csr */
-#define DMA_SCSI_DISAB   0x00020000        /* No FIFO drains during reg */
-#define DMA_DSBL_WR_INV  0x00020000        /* No EC inval. on slave writes */
-#define DMA_ADD_ENABLE   0x00040000        /* Special ESC DVMA optimization */
-#define DMA_E_BURSTS	 0x000c0000	   /* ENET: SBUS r/w burst mask */
-#define DMA_E_BURST32	 0x00040000	   /* ENET: SBUS 32 byte r/w burst */
-#define DMA_E_BURST16	 0x00000000	   /* ENET: SBUS 16 byte r/w burst */
-#define DMA_BRST_SZ      0x000c0000        /* SCSI: SBUS r/w burst size */
-#define DMA_BRST64       0x000c0000        /* SCSI: 64byte bursts (HME on UltraSparc only) */
-#define DMA_BRST32       0x00040000        /* SCSI: 32byte bursts */
-#define DMA_BRST16       0x00000000        /* SCSI: 16byte bursts */
-#define DMA_BRST0        0x00080000        /* SCSI: no bursts (non-HME gate arrays) */
-#define DMA_ADDR_DISAB   0x00100000        /* No FIFO drains during addr */
-#define DMA_2CLKS        0x00200000        /* Each transfer = 2 clock ticks */
-#define DMA_3CLKS        0x00400000        /* Each transfer = 3 clock ticks */
-#define DMA_EN_ENETAUI   DMA_3CLKS         /* Put lance into AUI-cable mode */
-#define DMA_CNTR_DISAB   0x00800000        /* No IRQ when DMA_TERM_CNTR set */
-#define DMA_AUTO_NADDR   0x01000000        /* Use "auto nxt addr" feature */
-#define DMA_SCSI_ON      0x02000000        /* Enable SCSI dma */
-#define DMA_PARITY_OFF   0x02000000        /* HME: disable parity checking */
-#define DMA_LOADED_ADDR  0x04000000        /* Address has been loaded */
-#define DMA_LOADED_NADDR 0x08000000        /* Next address has been loaded */
-#define DMA_RESET_FAS366 0x08000000        /* HME: Assert RESET to FAS366 */
-
-/* Values describing the burst-size property from the PROM */
-#define DMA_BURST1       0x01
-#define DMA_BURST2       0x02
-#define DMA_BURST4       0x04
-#define DMA_BURST8       0x08
-#define DMA_BURST16      0x10
-#define DMA_BURST32      0x20
-#define DMA_BURST64      0x40
-#define DMA_BURSTBITS    0x7f
-
-/* Determine highest possible final transfer address given a base */
-#define DMA_MAXEND(addr) (0x01000000UL-(((unsigned long)(addr))&0x00ffffffUL))
-
-/* Yes, I hack a lot of elisp in my spare time... */
-#define DMA_ERROR_P(regs)  ((sbus_readl((regs) + DMA_CSR) & DMA_HNDL_ERROR))
-#define DMA_IRQ_P(regs)    ((sbus_readl((regs) + DMA_CSR)) & (DMA_HNDL_INTR | DMA_HNDL_ERROR))
-#define DMA_WRITE_P(regs)  ((sbus_readl((regs) + DMA_CSR) & DMA_ST_WRITE))
-#define DMA_OFF(__regs)		\
-do {	u32 tmp = sbus_readl((__regs) + DMA_CSR); \
-	tmp &= ~DMA_ENABLE; \
-	sbus_writel(tmp, (__regs) + DMA_CSR); \
-} while(0)
-#define DMA_INTSOFF(__regs)	\
-do {	u32 tmp = sbus_readl((__regs) + DMA_CSR); \
-	tmp &= ~DMA_INT_ENAB; \
-	sbus_writel(tmp, (__regs) + DMA_CSR); \
-} while(0)
-#define DMA_INTSON(__regs)	\
-do {	u32 tmp = sbus_readl((__regs) + DMA_CSR); \
-	tmp |= DMA_INT_ENAB; \
-	sbus_writel(tmp, (__regs) + DMA_CSR); \
-} while(0)
-#define DMA_PUNTFIFO(__regs)	\
-do {	u32 tmp = sbus_readl((__regs) + DMA_CSR); \
-	tmp |= DMA_FIFO_INV; \
-	sbus_writel(tmp, (__regs) + DMA_CSR); \
-} while(0)
-#define DMA_SETSTART(__regs, __addr)	\
-	sbus_writel((u32)(__addr), (__regs) + DMA_ADDR);
-#define DMA_BEGINDMA_W(__regs)	\
-do {	u32 tmp = sbus_readl((__regs) + DMA_CSR); \
-	tmp |= (DMA_ST_WRITE|DMA_ENABLE|DMA_INT_ENAB); \
-	sbus_writel(tmp, (__regs) + DMA_CSR); \
-} while(0)
-#define DMA_BEGINDMA_R(__regs)	\
-do {	u32 tmp = sbus_readl((__regs) + DMA_CSR); \
-	tmp |= (DMA_ENABLE|DMA_INT_ENAB); \
-	tmp &= ~DMA_ST_WRITE; \
-	sbus_writel(tmp, (__regs) + DMA_CSR); \
-} while(0)
-
-/* For certain DMA chips, we need to disable ints upon irq entry
- * and turn them back on when we are done.  So in any ESP interrupt
- * handler you *must* call DMA_IRQ_ENTRY upon entry and DMA_IRQ_EXIT
- * when leaving the handler.  You have been warned...
- */
-#define DMA_IRQ_ENTRY(dma, dregs) do { \
-        if(DMA_ISBROKEN(dma)) DMA_INTSOFF(dregs); \
-   } while (0)
-
-#define DMA_IRQ_EXIT(dma, dregs) do { \
-	if(DMA_ISBROKEN(dma)) DMA_INTSON(dregs); \
-   } while(0)
-
-#define for_each_dvma(dma) \
-        for((dma) = dma_chain; (dma); (dma) = (dma)->next)
-
-/* From PCI */
-
-#ifdef CONFIG_PCI
-extern int isa_dma_bridge_buggy;
-#else
-#define isa_dma_bridge_buggy 	(0)
-#endif
-
-#endif /* !(_ASM_SPARC64_DMA_H) */
diff --git a/arch/sparc/include/asm/ebus.h b/arch/sparc/include/asm/ebus.h
deleted file mode 100644
index 83a6d16c22e6..000000000000
--- a/arch/sparc/include/asm/ebus.h
+++ /dev/null
@@ -1,8 +0,0 @@
-#ifndef ___ASM_SPARC_EBUS_H
-#define ___ASM_SPARC_EBUS_H
-#if defined(__sparc__) && defined(__arch64__)
-#include <asm/ebus_64.h>
-#else
-#include <asm/ebus_32.h>
-#endif
-#endif
diff --git a/arch/sparc/include/asm/ebus_32.h b/arch/sparc/include/asm/ebus_32.h
deleted file mode 100644
index f91f0b267ce1..000000000000
--- a/arch/sparc/include/asm/ebus_32.h
+++ /dev/null
@@ -1,99 +0,0 @@
-/*
- * ebus.h: PCI to Ebus pseudo driver software state.
- *
- * Copyright (C) 1997 Eddie C. Dost (ecd@skynet.be)
- *
- * Adopted for sparc by V. Roganov and G. Raiko.
- */
-
-#ifndef __SPARC_EBUS_H
-#define __SPARC_EBUS_H
-
-#ifndef _LINUX_IOPORT_H
-#include <linux/ioport.h>
-#endif
-#include <linux/of_device.h>
-#include <asm/oplib.h>
-#include <asm/prom.h>
-
-struct linux_ebus_child {
-	struct linux_ebus_child		*next;
-	struct linux_ebus_device	*parent;
-	struct linux_ebus		*bus;
-	struct device_node		*prom_node;
-	struct resource			 resource[PROMREG_MAX];
-	int				 num_addrs;
-	unsigned int			 irqs[PROMINTR_MAX];
-	int				 num_irqs;
-};
-
-struct linux_ebus_device {
-	struct of_device		ofdev;
-	struct linux_ebus_device	*next;
-	struct linux_ebus_child		*children;
-	struct linux_ebus		*bus;
-	struct device_node		*prom_node;
-	struct resource			 resource[PROMREG_MAX];
-	int				 num_addrs;
-	unsigned int			 irqs[PROMINTR_MAX];
-	int				 num_irqs;
-};
-#define to_ebus_device(d) container_of(d, struct linux_ebus_device, ofdev.dev)
-
-struct linux_ebus {
-	struct of_device		ofdev;
-	struct linux_ebus		*next;
-	struct linux_ebus_device	*devices;
-	struct linux_pbm_info		*parent;
-	struct pci_dev			*self;
-	struct device_node		*prom_node;
-};
-#define to_ebus(d) container_of(d, struct linux_ebus, ofdev.dev)
-
-struct linux_ebus_dma {
-	unsigned int dcsr;
-	unsigned int dacr;
-	unsigned int dbcr;
-};
-
-#define EBUS_DCSR_INT_PEND	0x00000001
-#define EBUS_DCSR_ERR_PEND	0x00000002
-#define EBUS_DCSR_DRAIN		0x00000004
-#define EBUS_DCSR_INT_EN	0x00000010
-#define EBUS_DCSR_RESET		0x00000080
-#define EBUS_DCSR_WRITE		0x00000100
-#define EBUS_DCSR_EN_DMA	0x00000200
-#define EBUS_DCSR_CYC_PEND	0x00000400
-#define EBUS_DCSR_DIAG_RD_DONE	0x00000800
-#define EBUS_DCSR_DIAG_WR_DONE	0x00001000
-#define EBUS_DCSR_EN_CNT	0x00002000
-#define EBUS_DCSR_TC		0x00004000
-#define EBUS_DCSR_DIS_CSR_DRN	0x00010000
-#define EBUS_DCSR_BURST_SZ_MASK	0x000c0000
-#define EBUS_DCSR_BURST_SZ_1	0x00080000
-#define EBUS_DCSR_BURST_SZ_4	0x00000000
-#define EBUS_DCSR_BURST_SZ_8	0x00040000
-#define EBUS_DCSR_BURST_SZ_16	0x000c0000
-#define EBUS_DCSR_DIAG_EN	0x00100000
-#define EBUS_DCSR_DIS_ERR_PEND	0x00400000
-#define EBUS_DCSR_TCI_DIS	0x00800000
-#define EBUS_DCSR_EN_NEXT	0x01000000
-#define EBUS_DCSR_DMA_ON	0x02000000
-#define EBUS_DCSR_A_LOADED	0x04000000
-#define EBUS_DCSR_NA_LOADED	0x08000000
-#define EBUS_DCSR_DEV_ID_MASK	0xf0000000
-
-extern struct linux_ebus		*ebus_chain;
-
-extern void ebus_init(void);
-
-#define for_each_ebus(bus)						\
-        for((bus) = ebus_chain; (bus); (bus) = (bus)->next)
-
-#define for_each_ebusdev(dev, bus)					\
-        for((dev) = (bus)->devices; (dev); (dev) = (dev)->next)
-
-#define for_each_edevchild(dev, child)					\
-        for((child) = (dev)->children; (child); (child) = (child)->next)
-
-#endif /* !(__SPARC_EBUS_H) */
diff --git a/arch/sparc/include/asm/ebus_64.h b/arch/sparc/include/asm/ebus_64.h
deleted file mode 100644
index 14c6a111f60c..000000000000
--- a/arch/sparc/include/asm/ebus_64.h
+++ /dev/null
@@ -1,95 +0,0 @@
-/*
- * ebus.h: PCI to Ebus pseudo driver software state.
- *
- * Copyright (C) 1997 Eddie C. Dost (ecd@skynet.be)
- * Copyright (C) 1999 David S. Miller (davem@redhat.com)
- */
-
-#ifndef __SPARC64_EBUS_H
-#define __SPARC64_EBUS_H
-
-#include <linux/of_device.h>
-
-#include <asm/oplib.h>
-#include <asm/prom.h>
-
-struct linux_ebus_child {
-	struct linux_ebus_child		*next;
-	struct linux_ebus_device	*parent;
-	struct linux_ebus		*bus;
-	struct device_node		*prom_node;
-	struct resource			 resource[PROMREG_MAX];
-	int				 num_addrs;
-	unsigned int			 irqs[PROMINTR_MAX];
-	int				 num_irqs;
-};
-
-struct linux_ebus_device {
-	struct of_device		ofdev;
-	struct linux_ebus_device	*next;
-	struct linux_ebus_child		*children;
-	struct linux_ebus		*bus;
-	struct device_node		*prom_node;
-	struct resource			 resource[PROMREG_MAX];
-	int				 num_addrs;
-	unsigned int			 irqs[PROMINTR_MAX];
-	int				 num_irqs;
-};
-#define to_ebus_device(d) container_of(d, struct linux_ebus_device, ofdev.dev)
-
-struct linux_ebus {
-	struct of_device		ofdev;
-	struct linux_ebus		*next;
-	struct linux_ebus_device	*devices;
-	struct pci_dev			*self;
-	int				 index;
-	int				 is_rio;
-	struct device_node		*prom_node;
-};
-#define to_ebus(d) container_of(d, struct linux_ebus, ofdev.dev)
-
-struct ebus_dma_info {
-	spinlock_t	lock;
-	void __iomem	*regs;
-
-	unsigned int	flags;
-#define EBUS_DMA_FLAG_USE_EBDMA_HANDLER		0x00000001
-#define EBUS_DMA_FLAG_TCI_DISABLE		0x00000002
-
-	/* These are only valid is EBUS_DMA_FLAG_USE_EBDMA_HANDLER is
-	 * set.
-	 */
-	void (*callback)(struct ebus_dma_info *p, int event, void *cookie);
-	void *client_cookie;
-	unsigned int	irq;
-#define EBUS_DMA_EVENT_ERROR	1
-#define EBUS_DMA_EVENT_DMA	2
-#define EBUS_DMA_EVENT_DEVICE	4
-
-	unsigned char	name[64];
-};
-
-extern int ebus_dma_register(struct ebus_dma_info *p);
-extern int ebus_dma_irq_enable(struct ebus_dma_info *p, int on);
-extern void ebus_dma_unregister(struct ebus_dma_info *p);
-extern int ebus_dma_request(struct ebus_dma_info *p, dma_addr_t bus_addr,
-			    size_t len);
-extern void ebus_dma_prepare(struct ebus_dma_info *p, int write);
-extern unsigned int ebus_dma_residue(struct ebus_dma_info *p);
-extern unsigned int ebus_dma_addr(struct ebus_dma_info *p);
-extern void ebus_dma_enable(struct ebus_dma_info *p, int on);
-
-extern struct linux_ebus		*ebus_chain;
-
-extern void ebus_init(void);
-
-#define for_each_ebus(bus)						\
-        for((bus) = ebus_chain; (bus); (bus) = (bus)->next)
-
-#define for_each_ebusdev(dev, bus)					\
-        for((dev) = (bus)->devices; (dev); (dev) = (dev)->next)
-
-#define for_each_edevchild(dev, child)					\
-        for((child) = (dev)->children; (child); (child) = (child)->next)
-
-#endif /* !(__SPARC64_EBUS_H) */
diff --git a/arch/sparc/include/asm/ebus_dma.h b/arch/sparc/include/asm/ebus_dma.h
new file mode 100644
index 000000000000..f07a5b541c98
--- /dev/null
+++ b/arch/sparc/include/asm/ebus_dma.h
@@ -0,0 +1,35 @@
+#ifndef __ASM_SPARC_EBUS_DMA_H
+#define __ASM_SPARC_EBUS_DMA_H
+
+struct ebus_dma_info {
+	spinlock_t	lock;
+	void __iomem	*regs;
+
+	unsigned int	flags;
+#define EBUS_DMA_FLAG_USE_EBDMA_HANDLER		0x00000001
+#define EBUS_DMA_FLAG_TCI_DISABLE		0x00000002
+
+	/* These are only valid is EBUS_DMA_FLAG_USE_EBDMA_HANDLER is
+	 * set.
+	 */
+	void (*callback)(struct ebus_dma_info *p, int event, void *cookie);
+	void *client_cookie;
+	unsigned int	irq;
+#define EBUS_DMA_EVENT_ERROR	1
+#define EBUS_DMA_EVENT_DMA	2
+#define EBUS_DMA_EVENT_DEVICE	4
+
+	unsigned char	name[64];
+};
+
+extern int ebus_dma_register(struct ebus_dma_info *p);
+extern int ebus_dma_irq_enable(struct ebus_dma_info *p, int on);
+extern void ebus_dma_unregister(struct ebus_dma_info *p);
+extern int ebus_dma_request(struct ebus_dma_info *p, dma_addr_t bus_addr,
+			    size_t len);
+extern void ebus_dma_prepare(struct ebus_dma_info *p, int write);
+extern unsigned int ebus_dma_residue(struct ebus_dma_info *p);
+extern unsigned int ebus_dma_addr(struct ebus_dma_info *p);
+extern void ebus_dma_enable(struct ebus_dma_info *p, int on);
+
+#endif /* __ASM_SPARC_EBUS_DMA_H */
diff --git a/arch/sparc/include/asm/elf_32.h b/arch/sparc/include/asm/elf_32.h
index d043f80bc2fd..b7ab60547827 100644
--- a/arch/sparc/include/asm/elf_32.h
+++ b/arch/sparc/include/asm/elf_32.h
@@ -105,11 +105,8 @@ typedef struct {
 #define ELF_DATA	ELFDATA2MSB
 
 #define USE_ELF_CORE_DUMP
-#ifndef CONFIG_SUN4
+
 #define ELF_EXEC_PAGESIZE	4096
-#else
-#define ELF_EXEC_PAGESIZE	8192
-#endif
 
 
 /* This is the location that an ET_DYN program is loaded if exec'ed.  Typical
@@ -126,7 +123,7 @@ typedef struct {
 /* Sun4c has none of the capabilities, most sun4m's have them all.
  * XXX This is gross, set some global variable at boot time. -DaveM
  */
-#define ELF_HWCAP	((ARCH_SUN4C_SUN4) ? 0 : \
+#define ELF_HWCAP	((ARCH_SUN4C) ? 0 : \
 			 (HWCAP_SPARC_FLUSH | HWCAP_SPARC_STBAR | \
 			  HWCAP_SPARC_SWAP | \
 			  ((srmmu_modtype != Cypress && \
diff --git a/arch/sparc/include/asm/fhc.h b/arch/sparc/include/asm/fhc.h
index 788cbc46a116..57f1b303ad54 100644
--- a/arch/sparc/include/asm/fhc.h
+++ b/arch/sparc/include/asm/fhc.h
@@ -1,5 +1,4 @@
-/*
- * fhc.h: Structures for central/fhc pseudo driver on Sunfire/Starfire/Wildfire.
+/* fhc.h: FHC and Clock board register definitions.
  *
  * Copyright (C) 1997, 1999 David S. Miller (davem@redhat.com)
  */
@@ -7,14 +6,6 @@
 #ifndef _SPARC64_FHC_H
 #define _SPARC64_FHC_H
 
-#include <linux/timer.h>
-
-#include <asm/oplib.h>
-#include <asm/prom.h>
-#include <asm/upa.h>
-
-struct linux_fhc;
-
 /* Clock board register offsets. */
 #define CLOCK_CTRL	0x00UL	/* Main control */
 #define CLOCK_STAT1	0x10UL	/* Status one */
@@ -29,21 +20,7 @@ struct linux_fhc;
 #define CLOCK_CTRL_MLED		0x02	/* Mid LED, 1 == on */
 #define CLOCK_CTRL_RLED		0x01	/* RIght LED, 1 == on */
 
-struct linux_central {
-	struct linux_fhc		*child;
-	unsigned long			cfreg;
-	unsigned long			clkregs;
-	unsigned long			clkver;
-	int				slots;
-	struct device_node		*prom_node;
-
-	struct linux_prom_ranges	central_ranges[PROMREG_MAX];
-	int				num_central_ranges;
-};
-
 /* Firehose controller register offsets */
-struct fhc_regs {
-	unsigned long			pregs;	/* FHC internal regs */
 #define FHC_PREGS_ID	0x00UL	/* FHC ID */
 #define  FHC_ID_VERS		0xf0000000 /* Version of this FHC		*/
 #define  FHC_ID_PARTID		0x0ffff000 /* Part ID code (0x0f9f == FHC)	*/
@@ -90,32 +67,14 @@ struct fhc_regs {
 #define  FHC_JTAG_CTRL_MENAB	0x80000000 /* Indicates this is JTAG Master	 */
 #define  FHC_JTAG_CTRL_MNONE	0x40000000 /* Indicates no JTAG Master present	 */
 #define FHC_PREGS_JCMD	0x100UL	/* FHC JTAG Command Register */
-	unsigned long			ireg;	/* FHC IGN reg */
 #define FHC_IREG_IGN	0x00UL	/* This FHC's IGN */
-	unsigned long			ffregs;	/* FHC fanfail regs */
 #define FHC_FFREGS_IMAP	0x00UL	/* FHC Fanfail IMAP */
 #define FHC_FFREGS_ICLR	0x10UL	/* FHC Fanfail ICLR */
-	unsigned long			sregs;	/* FHC system regs */
 #define FHC_SREGS_IMAP	0x00UL	/* FHC System IMAP */
 #define FHC_SREGS_ICLR	0x10UL	/* FHC System ICLR */
-	unsigned long			uregs;	/* FHC uart regs */
 #define FHC_UREGS_IMAP	0x00UL	/* FHC Uart IMAP */
 #define FHC_UREGS_ICLR	0x10UL	/* FHC Uart ICLR */
-	unsigned long			tregs;	/* FHC TOD regs */
 #define FHC_TREGS_IMAP	0x00UL	/* FHC TOD IMAP */
 #define FHC_TREGS_ICLR	0x10UL	/* FHC TOD ICLR */
-};
-
-struct linux_fhc {
-	struct linux_fhc		*next;
-	struct linux_central		*parent;	/* NULL if not central FHC */
-	struct fhc_regs			fhc_regs;
-	int				board;
-	int				jtag_master;
-	struct device_node		*prom_node;
-
-	struct linux_prom_ranges	fhc_ranges[PROMREG_MAX];
-	int				num_fhc_ranges;
-};
 
 #endif /* !(_SPARC64_FHC_H) */
diff --git a/arch/sparc/include/asm/floppy_32.h b/arch/sparc/include/asm/floppy_32.h
index ae3f00bf22ff..c792830636de 100644
--- a/arch/sparc/include/asm/floppy_32.h
+++ b/arch/sparc/include/asm/floppy_32.h
@@ -6,6 +6,9 @@
 #ifndef __ASM_SPARC_FLOPPY_H
 #define __ASM_SPARC_FLOPPY_H
 
+#include <linux/of.h>
+#include <linux/of_device.h>
+
 #include <asm/page.h>
 #include <asm/pgtable.h>
 #include <asm/system.h>
@@ -343,7 +346,7 @@ static int sun_floppy_init(void)
 	r.flags = fd_regs[0].which_io;
 	r.start = fd_regs[0].phys_addr;
 	sun_fdc = (struct sun_flpy_controller *)
-	    sbus_ioremap(&r, 0, fd_regs[0].reg_size, "floppy");
+	    of_ioremap(&r, 0, fd_regs[0].reg_size, "floppy");
 
 	/* Last minute sanity check... */
 	if(sun_fdc->status_82072 == 0xff) {
@@ -385,4 +388,15 @@ static int sparc_eject(void)
 
 #define EXTRA_FLOPPY_PARAMS
 
+static DEFINE_SPINLOCK(dma_spin_lock);
+
+#define claim_dma_lock() \
+({	unsigned long flags; \
+	spin_lock_irqsave(&dma_spin_lock, flags); \
+	flags; \
+})
+
+#define release_dma_lock(__flags) \
+	spin_unlock_irqrestore(&dma_spin_lock, __flags);
+
 #endif /* !(__ASM_SPARC_FLOPPY_H) */
diff --git a/arch/sparc/include/asm/floppy_64.h b/arch/sparc/include/asm/floppy_64.h
index c39db1060bc7..36439d67ad71 100644
--- a/arch/sparc/include/asm/floppy_64.h
+++ b/arch/sparc/include/asm/floppy_64.h
@@ -1,6 +1,6 @@
 /* floppy.h: Sparc specific parts of the Floppy driver.
  *
- * Copyright (C) 1996, 2007 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 1996, 2007, 2008 David S. Miller (davem@davemloft.net)
  * Copyright (C) 1997 Jakub Jelinek (jj@sunsite.mff.cuni.cz)
  *
  * Ultra/PCI support added: Sep 1997  Eddie C. Dost  (ecd@skynet.be)
@@ -9,18 +9,11 @@
 #ifndef __ASM_SPARC64_FLOPPY_H
 #define __ASM_SPARC64_FLOPPY_H
 
-#include <linux/init.h>
-#include <linux/pci.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/dma-mapping.h>
 
-#include <asm/page.h>
-#include <asm/pgtable.h>
-#include <asm/system.h>
-#include <asm/idprom.h>
-#include <asm/oplib.h>
 #include <asm/auxio.h>
-#include <asm/sbus.h>
-#include <asm/irq.h>
-
 
 /*
  * Define this to enable exchanging drive 0 and 1 if only drive 1 is
@@ -50,7 +43,7 @@ struct sun_flpy_controller {
 /* You'll only ever find one controller on an Ultra anyways. */
 static struct sun_flpy_controller *sun_fdc = (struct sun_flpy_controller *)-1;
 unsigned long fdc_status;
-static struct sbus_dev *floppy_sdev = NULL;
+static struct of_device *floppy_op = NULL;
 
 struct sun_floppy_ops {
 	unsigned char	(*fd_inb) (unsigned long port);
@@ -291,12 +284,11 @@ static int sun_fd_eject(int drive)
 	return 0;
 }
 
-#ifdef CONFIG_PCI
-#include <asm/ebus.h>
+#include <asm/ebus_dma.h>
 #include <asm/ns87303.h>
 
 static struct ebus_dma_info sun_pci_fd_ebus_dma;
-static struct pci_dev *sun_pci_ebus_dev;
+static struct device *sun_floppy_dev;
 static int sun_pci_broken_drive = -1;
 
 struct sun_pci_dma_op {
@@ -377,7 +369,7 @@ static void sun_pci_fd_enable_dma(void)
 	sun_pci_dma_pending.addr = -1U;
 
 	sun_pci_dma_current.addr =
-		pci_map_single(sun_pci_ebus_dev,
+		dma_map_single(sun_floppy_dev,
 			       sun_pci_dma_current.buf,
 			       sun_pci_dma_current.len,
 			       sun_pci_dma_current.direction);
@@ -394,7 +386,7 @@ static void sun_pci_fd_disable_dma(void)
 {
 	ebus_dma_enable(&sun_pci_fd_ebus_dma, 0);
 	if (sun_pci_dma_current.addr != -1U)
-		pci_unmap_single(sun_pci_ebus_dev,
+		dma_unmap_single(sun_floppy_dev,
 				 sun_pci_dma_current.addr,
 				 sun_pci_dma_current.len,
 				 sun_pci_dma_current.direction);
@@ -404,9 +396,9 @@ static void sun_pci_fd_disable_dma(void)
 static void sun_pci_fd_set_dma_mode(int mode)
 {
 	if (mode == DMA_MODE_WRITE)
-		sun_pci_dma_pending.direction = PCI_DMA_TODEVICE;
+		sun_pci_dma_pending.direction = DMA_TO_DEVICE;
 	else
-		sun_pci_dma_pending.direction = PCI_DMA_FROMDEVICE;
+		sun_pci_dma_pending.direction = DMA_FROM_DEVICE;
 
 	ebus_dma_prepare(&sun_pci_fd_ebus_dma, mode != DMA_MODE_WRITE);
 }
@@ -538,80 +530,84 @@ static int sun_pci_fd_test_drive(unsigned long port, int drive)
 #undef MSR
 #undef DOR
 
-#endif /* CONFIG_PCI */
-
-#ifdef CONFIG_PCI
-static int __init ebus_fdthree_p(struct linux_ebus_device *edev)
+static int __init ebus_fdthree_p(struct device_node *dp)
 {
-	if (!strcmp(edev->prom_node->name, "fdthree"))
+	if (!strcmp(dp->name, "fdthree"))
 		return 1;
-	if (!strcmp(edev->prom_node->name, "floppy")) {
+	if (!strcmp(dp->name, "floppy")) {
 		const char *compat;
 
-		compat = of_get_property(edev->prom_node,
-					 "compatible", NULL);
+		compat = of_get_property(dp, "compatible", NULL);
 		if (compat && !strcmp(compat, "fdthree"))
 			return 1;
 	}
 	return 0;
 }
-#endif
 
 static unsigned long __init sun_floppy_init(void)
 {
-	char state[128];
-	struct sbus_bus *bus;
-	struct sbus_dev *sdev = NULL;
 	static int initialized = 0;
+	struct device_node *dp;
+	struct of_device *op;
+	const char *prop;
+	char state[128];
 
 	if (initialized)
 		return sun_floppy_types[0];
 	initialized = 1;
 
-	for_all_sbusdev (sdev, bus) {
-		if (!strcmp(sdev->prom_name, "SUNW,fdtwo"))
+	op = NULL;
+
+	for_each_node_by_name(dp, "SUNW,fdtwo") {
+		if (strcmp(dp->parent->name, "sbus"))
+			continue;
+		op = of_find_device_by_node(dp);
+		if (op)
 			break;
 	}
-	if(sdev) {
-		floppy_sdev = sdev;
-		FLOPPY_IRQ = sdev->irqs[0];
+	if (op) {
+		floppy_op = op;
+		FLOPPY_IRQ = op->irqs[0];
 	} else {
-#ifdef CONFIG_PCI
-		struct linux_ebus *ebus;
-		struct linux_ebus_device *edev = NULL;
-		unsigned long config = 0;
+		struct device_node *ebus_dp;
 		void __iomem *auxio_reg;
 		const char *state_prop;
+		unsigned long config;
 
-		for_each_ebus(ebus) {
-			for_each_ebusdev(edev, ebus) {
-				if (ebus_fdthree_p(edev))
-					goto ebus_done;
+		dp = NULL;
+		for_each_node_by_name(ebus_dp, "ebus") {
+			for (dp = ebus_dp->child; dp; dp = dp->sibling) {
+				if (ebus_fdthree_p(dp))
+					goto found_fdthree;
 			}
 		}
-	ebus_done:
-		if (!edev)
+	found_fdthree:
+		if (!dp)
+			return 0;
+
+		op = of_find_device_by_node(dp);
+		if (!op)
 			return 0;
 
-		state_prop = of_get_property(edev->prom_node, "status", NULL);
+		state_prop = of_get_property(op->node, "status", NULL);
 		if (state_prop && !strncmp(state_prop, "disabled", 8))
 			return 0;
 
-		FLOPPY_IRQ = edev->irqs[0];
+		FLOPPY_IRQ = op->irqs[0];
 
 		/* Make sure the high density bit is set, some systems
 		 * (most notably Ultra5/Ultra10) come up with it clear.
 		 */
-		auxio_reg = (void __iomem *) edev->resource[2].start;
+		auxio_reg = (void __iomem *) op->resource[2].start;
 		writel(readl(auxio_reg)|0x2, auxio_reg);
 
-		sun_pci_ebus_dev = ebus->self;
+		sun_floppy_dev = &op->dev;
 
 		spin_lock_init(&sun_pci_fd_ebus_dma.lock);
 
 		/* XXX ioremap */
 		sun_pci_fd_ebus_dma.regs = (void __iomem *)
-			edev->resource[1].start;
+			op->resource[1].start;
 		if (!sun_pci_fd_ebus_dma.regs)
 			return 0;
 
@@ -625,7 +621,7 @@ static unsigned long __init sun_floppy_init(void)
 			return 0;
 
 		/* XXX ioremap */
-		sun_fdc = (struct sun_flpy_controller *)edev->resource[0].start;
+		sun_fdc = (struct sun_flpy_controller *) op->resource[0].start;
 
 		sun_fdops.fd_inb = sun_pci_fd_inb;
 		sun_fdops.fd_outb = sun_pci_fd_outb;
@@ -662,12 +658,15 @@ static unsigned long __init sun_floppy_init(void)
 		/*
 		 * Find NS87303 SuperIO config registers (through ecpp).
 		 */
-		for_each_ebus(ebus) {
-			for_each_ebusdev(edev, ebus) {
-				if (!strcmp(edev->prom_node->name, "ecpp")) {
-					config = edev->resource[1].start;
-					goto config_done;
-				}
+		config = 0;
+		for (dp = ebus_dp->child; dp; dp = dp->sibling) {
+			if (!strcmp(dp->name, "ecpp")) {
+				struct of_device *ecpp_op;
+
+				ecpp_op = of_find_device_by_node(dp);
+				if (ecpp_op)
+					config = ecpp_op->resource[1].start;
+				goto config_done;
 			}
 		}
 	config_done:
@@ -716,26 +715,23 @@ static unsigned long __init sun_floppy_init(void)
 #endif /* PCI_FDC_SWAP_DRIVES */
 
 		return sun_floppy_types[0];
-#else
-		return 0;
-#endif
 	}
-	prom_getproperty(sdev->prom_node, "status", state, sizeof(state));
-	if(!strncmp(state, "disabled", 8))
+	prop = of_get_property(op->node, "status", NULL);
+	if (prop && !strncmp(state, "disabled", 8))
 		return 0;
 
 	/*
-	 * We cannot do sbus_ioremap here: it does request_region,
+	 * We cannot do of_ioremap here: it does request_region,
 	 * which the generic floppy driver tries to do once again.
 	 * But we must use the sdev resource values as they have
 	 * had parent ranges applied.
 	 */
 	sun_fdc = (struct sun_flpy_controller *)
-		(sdev->resource[0].start +
-		 ((sdev->resource[0].flags & 0x1ffUL) << 32UL));
+		(op->resource[0].start +
+		 ((op->resource[0].flags & 0x1ffUL) << 32UL));
 
 	/* Last minute sanity check... */
-	if(sbus_readb(&sun_fdc->status1_82077) == 0xff) {
+	if (sbus_readb(&sun_fdc->status1_82077) == 0xff) {
 		sun_fdc = (struct sun_flpy_controller *)-1;
 		return 0;
 	}
diff --git a/arch/sparc/include/asm/gpio.h b/arch/sparc/include/asm/gpio.h
new file mode 100644
index 000000000000..a0e3ac0af599
--- /dev/null
+++ b/arch/sparc/include/asm/gpio.h
@@ -0,0 +1,36 @@
+#ifndef __ASM_SPARC_GPIO_H
+#define __ASM_SPARC_GPIO_H
+
+#include <linux/errno.h>
+#include <asm-generic/gpio.h>
+
+#ifdef CONFIG_GPIOLIB
+
+static inline int gpio_get_value(unsigned int gpio)
+{
+	return __gpio_get_value(gpio);
+}
+
+static inline void gpio_set_value(unsigned int gpio, int value)
+{
+	__gpio_set_value(gpio, value);
+}
+
+static inline int gpio_cansleep(unsigned int gpio)
+{
+	return __gpio_cansleep(gpio);
+}
+
+static inline int gpio_to_irq(unsigned int gpio)
+{
+	return -ENOSYS;
+}
+
+static inline int irq_to_gpio(unsigned int irq)
+{
+	return -EINVAL;
+}
+
+#endif /* CONFIG_GPIOLIB */
+
+#endif /* __ASM_SPARC_GPIO_H */
diff --git a/arch/sparc/include/asm/io-unit.h b/arch/sparc/include/asm/io-unit.h
index 96823b47fd45..01ab2f613e91 100644
--- a/arch/sparc/include/asm/io-unit.h
+++ b/arch/sparc/include/asm/io-unit.h
@@ -55,8 +55,4 @@ struct iounit_struct {
 #define IOUNIT_BMAPM_START	IOUNIT_BMAP2_END
 #define IOUNIT_BMAPM_END	((IOUNIT_DMA_SIZE - IOUNIT_DVMA_SIZE) >> PAGE_SHIFT)
 
-extern __u32 iounit_map_dma_init(struct sbus_bus *, int);
-#define iounit_map_dma_finish(sbus, addr, len) mmu_release_scsi_one(addr, len, sbus)
-extern __u32 iounit_map_dma_page(__u32, void *, struct sbus_bus *);
-
 #endif /* !(_SPARC_IO_UNIT_H) */
diff --git a/arch/sparc/include/asm/io_32.h b/arch/sparc/include/asm/io_32.h
index 10d7da450070..93fe21e02c86 100644
--- a/arch/sparc/include/asm/io_32.h
+++ b/arch/sparc/include/asm/io_32.h
@@ -293,14 +293,6 @@ extern void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max);
 extern void pci_iounmap(struct pci_dev *dev, void __iomem *);
 
 /*
- * Bus number may be in res->flags... somewhere.
- */
-extern void __iomem *sbus_ioremap(struct resource *res, unsigned long offset,
-    unsigned long size, char *name);
-extern void sbus_iounmap(volatile void __iomem *vaddr, unsigned long size);
-
-
-/*
  * At the moment, we do not use CMOS_READ anywhere outside of rtc.c,
  * so rtc_port is static in it. This should not change unless a new
  * hardware pops up.
@@ -308,6 +300,17 @@ extern void sbus_iounmap(volatile void __iomem *vaddr, unsigned long size);
 #define RTC_PORT(x)   (rtc_port + (x))
 #define RTC_ALWAYS_BCD  0
 
+static inline int sbus_can_dma_64bit(void)
+{
+	return 0; /* actually, sparc_cpu_model==sun4d */
+}
+static inline int sbus_can_burst64(void)
+{
+	return 0; /* actually, sparc_cpu_model==sun4d */
+}
+struct device;
+extern void sbus_set_sbus64(struct device *, int);
+
 #endif
 
 #define __ARCH_HAS_NO_PAGE_ZERO_MAPPED		1
diff --git a/arch/sparc/include/asm/io_64.h b/arch/sparc/include/asm/io_64.h
index 0bff078ffdd0..4aee21dc9c6f 100644
--- a/arch/sparc/include/asm/io_64.h
+++ b/arch/sparc/include/asm/io_64.h
@@ -482,18 +482,16 @@ struct pci_dev;
 extern void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max);
 extern void pci_iounmap(struct pci_dev *dev, void __iomem *);
 
-/* Similarly for SBUS. */
-#define sbus_ioremap(__res, __offset, __size, __name) \
-({	unsigned long __ret; \
-	__ret  = (__res)->start + (((__res)->flags & 0x1ffUL) << 32UL); \
-	__ret += (unsigned long) (__offset); \
-	if (! request_region((__ret), (__size), (__name))) \
-		__ret = 0UL; \
-	(void __iomem *) __ret; \
-})
-
-#define sbus_iounmap(__addr, __size)	\
-	release_region((unsigned long)(__addr), (__size))
+static inline int sbus_can_dma_64bit(void)
+{
+	return 1;
+}
+static inline int sbus_can_burst64(void)
+{
+	return 1;
+}
+struct device;
+extern void sbus_set_sbus64(struct device *, int);
 
 /*
  * Convert a physical pointer to a virtual kernel pointer for /dev/mem
diff --git a/arch/sparc/include/asm/iommu_64.h b/arch/sparc/include/asm/iommu_64.h
index d7b9afcba08b..caf798b56191 100644
--- a/arch/sparc/include/asm/iommu_64.h
+++ b/arch/sparc/include/asm/iommu_64.h
@@ -48,6 +48,9 @@ struct strbuf {
 	unsigned long		strbuf_control;
 	unsigned long		strbuf_pflush;
 	unsigned long		strbuf_fsync;
+	unsigned long		strbuf_err_stat;
+	unsigned long		strbuf_tag_diag;
+	unsigned long		strbuf_line_diag;
 	unsigned long		strbuf_ctxflush;
 	unsigned long		strbuf_ctxmatch_base;
 	unsigned long		strbuf_flushflag_pa;
diff --git a/arch/sparc/include/asm/irq_64.h b/arch/sparc/include/asm/irq_64.h
index e3dd9303643d..71673eca3660 100644
--- a/arch/sparc/include/asm/irq_64.h
+++ b/arch/sparc/include/asm/irq_64.h
@@ -56,7 +56,6 @@ extern unsigned int sun4u_build_msi(u32 portid, unsigned int *virt_irq_p,
 				    unsigned long imap_base,
 				    unsigned long iclr_base);
 extern void sun4u_destroy_msi(unsigned int virt_irq);
-extern unsigned int sbus_build_irq(void *sbus, unsigned int ino);
 
 extern unsigned char virt_irq_alloc(unsigned int dev_handle,
 				    unsigned int dev_ino);
diff --git a/arch/sparc/include/asm/mc146818rtc_64.h b/arch/sparc/include/asm/mc146818rtc_64.h
index e9c0fcc25c6f..7238d174e0e3 100644
--- a/arch/sparc/include/asm/mc146818rtc_64.h
+++ b/arch/sparc/include/asm/mc146818rtc_64.h
@@ -7,12 +7,8 @@
 #include <asm/io.h>
 
 #ifndef RTC_PORT
-#ifdef CONFIG_PCI
-extern unsigned long ds1287_regs;
-#else
-#define ds1287_regs (0UL)
-#endif
-#define RTC_PORT(x)	(ds1287_regs + (x))
+extern unsigned long cmos_regs;
+#define RTC_PORT(x)	(cmos_regs + (x))
 #define RTC_ALWAYS_BCD	0
 #endif
 
@@ -29,6 +25,4 @@ outb_p((addr),RTC_PORT(0)); \
 outb_p((val),RTC_PORT(1)); \
 })
 
-#define RTC_IRQ 8
-
 #endif /* __ASM_SPARC64_MC146818RTC_H */
diff --git a/arch/sparc/include/asm/memctrl.h b/arch/sparc/include/asm/memctrl.h
new file mode 100644
index 000000000000..4065c56af7b6
--- /dev/null
+++ b/arch/sparc/include/asm/memctrl.h
@@ -0,0 +1,9 @@
+#ifndef _SPARC_MEMCTRL_H
+#define _SPARC_MEMCTRL_H
+
+typedef int (*dimm_printer_t)(int synd_code, unsigned long paddr, char *buf, int buflen);
+
+int register_dimm_printer(dimm_printer_t func);
+void unregister_dimm_printer(dimm_printer_t func);
+
+#endif /* _SPARC_MEMCTRL_H */
diff --git a/arch/sparc/include/asm/mostek.h b/arch/sparc/include/asm/mostek.h
deleted file mode 100644
index 433be3e0a69b..000000000000
--- a/arch/sparc/include/asm/mostek.h
+++ /dev/null
@@ -1,8 +0,0 @@
-#ifndef ___ASM_SPARC_MOSTEK_H
-#define ___ASM_SPARC_MOSTEK_H
-#if defined(__sparc__) && defined(__arch64__)
-#include <asm/mostek_64.h>
-#else
-#include <asm/mostek_32.h>
-#endif
-#endif
diff --git a/arch/sparc/include/asm/mostek_32.h b/arch/sparc/include/asm/mostek_32.h
deleted file mode 100644
index a99590c4c507..000000000000
--- a/arch/sparc/include/asm/mostek_32.h
+++ /dev/null
@@ -1,171 +0,0 @@
-/*
- * mostek.h:  Describes the various Mostek time of day clock registers.
- *
- * Copyright (C) 1995 David S. Miller (davem@caip.rutgers.edu)
- * Copyright (C) 1996 Thomas K. Dyas (tdyas@eden.rutgers.edu)
- * Added intersil code 05/25/98 Chris Davis (cdavis@cois.on.ca)
- */
-
-#ifndef _SPARC_MOSTEK_H
-#define _SPARC_MOSTEK_H
-
-#include <asm/idprom.h>
-#include <asm/io.h>
-
-/*       M48T02 Register Map (adapted from Sun NVRAM/Hostid FAQ)
- *
- *                             Data
- * Address                                                 Function
- *        Bit 7 Bit 6 Bit 5 Bit 4Bit 3 Bit 2 Bit 1 Bit 0
- *   7ff  -     -     -     -    -     -     -     -       Year 00-99
- *   7fe  0     0     0     -    -     -     -     -      Month 01-12
- *   7fd  0     0     -     -    -     -     -     -       Date 01-31
- *   7fc  0     FT    0     0    0     -     -     -        Day 01-07
- *   7fb  KS    0     -     -    -     -     -     -      Hours 00-23
- *   7fa  0     -     -     -    -     -     -     -    Minutes 00-59
- *   7f9  ST    -     -     -    -     -     -     -    Seconds 00-59
- *   7f8  W     R     S     -    -     -     -     -    Control
- *
- *   * ST is STOP BIT
- *   * W is WRITE BIT
- *   * R is READ BIT
- *   * S is SIGN BIT
- *   * FT is FREQ TEST BIT
- *   * KS is KICK START BIT
- */
-
-/* The Mostek 48t02 real time clock and NVRAM chip. The registers
- * other than the control register are in binary coded decimal. Some
- * control bits also live outside the control register.
- */
-#define mostek_read(_addr)		readb(_addr)
-#define mostek_write(_addr,_val)	writeb(_val, _addr)
-#define MOSTEK_EEPROM		0x0000UL
-#define MOSTEK_IDPROM		0x07d8UL
-#define MOSTEK_CREG		0x07f8UL
-#define MOSTEK_SEC		0x07f9UL
-#define MOSTEK_MIN		0x07faUL
-#define MOSTEK_HOUR		0x07fbUL
-#define MOSTEK_DOW		0x07fcUL
-#define MOSTEK_DOM		0x07fdUL
-#define MOSTEK_MONTH		0x07feUL
-#define MOSTEK_YEAR		0x07ffUL
-
-struct mostek48t02 {
-	volatile char eeprom[2008];	/* This is the eeprom, don't touch! */
-	struct idprom idprom;		/* The idprom lives here. */
-	volatile unsigned char creg;	/* Control register */
-	volatile unsigned char sec;	/* Seconds (0-59) */
-	volatile unsigned char min;	/* Minutes (0-59) */
-	volatile unsigned char hour;	/* Hour (0-23) */
-	volatile unsigned char dow;	/* Day of the week (1-7) */
-	volatile unsigned char dom;	/* Day of the month (1-31) */
-	volatile unsigned char month;	/* Month of year (1-12) */
-	volatile unsigned char year;	/* Year (0-99) */
-};
-
-extern spinlock_t mostek_lock;
-extern void __iomem *mstk48t02_regs;
-
-/* Control register values. */
-#define	MSTK_CREG_WRITE	0x80	/* Must set this before placing values. */
-#define	MSTK_CREG_READ	0x40	/* Stop updates to allow a clean read. */
-#define	MSTK_CREG_SIGN	0x20	/* Slow/speed clock in calibration mode. */
-
-/* Control bits that live in the other registers. */
-#define	MSTK_STOP	0x80	/* Stop the clock oscillator. (sec) */
-#define	MSTK_KICK_START	0x80	/* Kick start the clock chip. (hour) */
-#define MSTK_FREQ_TEST	0x40	/* Frequency test mode. (day) */
-
-#define MSTK_YEAR_ZERO       1968   /* If year reg has zero, it is 1968. */
-#define MSTK_CVT_YEAR(yr)  ((yr) + MSTK_YEAR_ZERO)
-
-/* Masks that define how much space each value takes up. */
-#define	MSTK_SEC_MASK	0x7f
-#define	MSTK_MIN_MASK	0x7f
-#define	MSTK_HOUR_MASK	0x3f
-#define	MSTK_DOW_MASK	0x07
-#define	MSTK_DOM_MASK	0x3f
-#define	MSTK_MONTH_MASK	0x1f
-#define	MSTK_YEAR_MASK	0xffU
-
-/* Binary coded decimal conversion macros. */
-#define MSTK_REGVAL_TO_DECIMAL(x)  (((x) & 0x0F) + 0x0A * ((x) >> 0x04))
-#define MSTK_DECIMAL_TO_REGVAL(x)  ((((x) / 0x0A) << 0x04) + ((x) % 0x0A))
-
-/* Generic register set and get macros for internal use. */
-#define MSTK_GET(regs,var,mask) (MSTK_REGVAL_TO_DECIMAL(((struct mostek48t02 *)regs)->var & MSTK_ ## mask ## _MASK))
-#define MSTK_SET(regs,var,value,mask) do { ((struct mostek48t02 *)regs)->var &= ~(MSTK_ ## mask ## _MASK); ((struct mostek48t02 *)regs)->var |= MSTK_DECIMAL_TO_REGVAL(value) & (MSTK_ ## mask ## _MASK); } while (0)
-
-/* Macros to make register access easier on our fingers. These give you
- * the decimal value of the register requested if applicable. You pass
- * the a pointer to a 'struct mostek48t02'.
- */
-#define	MSTK_REG_CREG(regs)	(((struct mostek48t02 *)regs)->creg)
-#define	MSTK_REG_SEC(regs)	MSTK_GET(regs,sec,SEC)
-#define	MSTK_REG_MIN(regs)	MSTK_GET(regs,min,MIN)
-#define	MSTK_REG_HOUR(regs)	MSTK_GET(regs,hour,HOUR)
-#define	MSTK_REG_DOW(regs)	MSTK_GET(regs,dow,DOW)
-#define	MSTK_REG_DOM(regs)	MSTK_GET(regs,dom,DOM)
-#define	MSTK_REG_MONTH(regs)	MSTK_GET(regs,month,MONTH)
-#define	MSTK_REG_YEAR(regs)	MSTK_GET(regs,year,YEAR)
-
-#define	MSTK_SET_REG_SEC(regs,value)	MSTK_SET(regs,sec,value,SEC)
-#define	MSTK_SET_REG_MIN(regs,value)	MSTK_SET(regs,min,value,MIN)
-#define	MSTK_SET_REG_HOUR(regs,value)	MSTK_SET(regs,hour,value,HOUR)
-#define	MSTK_SET_REG_DOW(regs,value)	MSTK_SET(regs,dow,value,DOW)
-#define	MSTK_SET_REG_DOM(regs,value)	MSTK_SET(regs,dom,value,DOM)
-#define	MSTK_SET_REG_MONTH(regs,value)	MSTK_SET(regs,month,value,MONTH)
-#define	MSTK_SET_REG_YEAR(regs,value)	MSTK_SET(regs,year,value,YEAR)
-
-
-/* The Mostek 48t08 clock chip. Found on Sun4m's I think. It has the
- * same (basically) layout of the 48t02 chip except for the extra
- * NVRAM on board (8 KB against the 48t02's 2 KB).
- */
-struct mostek48t08 {
-	char offset[6*1024];         /* Magic things may be here, who knows? */
-	struct mostek48t02 regs;     /* Here is what we are interested in.   */
-};
-
-#ifdef CONFIG_SUN4
-enum sparc_clock_type {	MSTK48T02, MSTK48T08, \
-INTERSIL, MSTK_INVALID };
-#else
-enum sparc_clock_type {	MSTK48T02, MSTK48T08, \
-MSTK_INVALID };
-#endif
-
-#ifdef CONFIG_SUN4
-/* intersil on a sun 4/260 code  data from harris doc */
-struct intersil_dt {
-        volatile unsigned char int_csec;
-        volatile unsigned char int_hour;
-        volatile unsigned char int_min;
-        volatile unsigned char int_sec;
-        volatile unsigned char int_month;
-        volatile unsigned char int_day;
-        volatile unsigned char int_year;
-        volatile unsigned char int_dow;
-};
-
-struct intersil {
-	struct intersil_dt clk;
-	struct intersil_dt cmp;
-	volatile unsigned char int_intr_reg;
-	volatile unsigned char int_cmd_reg;
-};
-
-#define INTERSIL_STOP        0x0
-#define INTERSIL_START       0x8
-#define INTERSIL_INTR_DISABLE   0x0
-#define INTERSIL_INTR_ENABLE   0x10
-#define INTERSIL_32K		0x0
-#define INTERSIL_NORMAL		0x0
-#define INTERSIL_24H		0x4
-#define INTERSIL_INT_100HZ	0x2
-
-/* end of intersil info */
-#endif
-
-#endif /* !(_SPARC_MOSTEK_H) */
diff --git a/arch/sparc/include/asm/mostek_64.h b/arch/sparc/include/asm/mostek_64.h
deleted file mode 100644
index c5652de2ace2..000000000000
--- a/arch/sparc/include/asm/mostek_64.h
+++ /dev/null
@@ -1,143 +0,0 @@
-/* mostek.h:  Describes the various Mostek time of day clock registers.
- *
- * Copyright (C) 1995 David S. Miller (davem@davemloft.net)
- * Copyright (C) 1996 Thomas K. Dyas (tdyas@eden.rutgers.edu)
- */
-
-#ifndef _SPARC64_MOSTEK_H
-#define _SPARC64_MOSTEK_H
-
-#include <asm/idprom.h>
-
-/*       M48T02 Register Map (adapted from Sun NVRAM/Hostid FAQ)
- *
- *                             Data
- * Address                                                 Function
- *        Bit 7 Bit 6 Bit 5 Bit 4Bit 3 Bit 2 Bit 1 Bit 0
- *   7ff  -     -     -     -    -     -     -     -       Year 00-99
- *   7fe  0     0     0     -    -     -     -     -      Month 01-12
- *   7fd  0     0     -     -    -     -     -     -       Date 01-31
- *   7fc  0     FT    0     0    0     -     -     -        Day 01-07
- *   7fb  KS    0     -     -    -     -     -     -      Hours 00-23
- *   7fa  0     -     -     -    -     -     -     -    Minutes 00-59
- *   7f9  ST    -     -     -    -     -     -     -    Seconds 00-59
- *   7f8  W     R     S     -    -     -     -     -    Control
- *
- *   * ST is STOP BIT
- *   * W is WRITE BIT
- *   * R is READ BIT
- *   * S is SIGN BIT
- *   * FT is FREQ TEST BIT
- *   * KS is KICK START BIT
- */
-
-/* The Mostek 48t02 real time clock and NVRAM chip. The registers
- * other than the control register are in binary coded decimal. Some
- * control bits also live outside the control register.
- *
- * We now deal with physical addresses for I/O to the chip. -DaveM
- */
-static inline u8 mostek_read(void __iomem *addr)
-{
-	u8 ret;
-
-	__asm__ __volatile__("lduba	[%1] %2, %0"
-			     : "=r" (ret)
-			     : "r" (addr), "i" (ASI_PHYS_BYPASS_EC_E));
-	return ret;
-}
-
-static inline void mostek_write(void __iomem *addr, u8 val)
-{
-	__asm__ __volatile__("stba	%0, [%1] %2"
-			     : /* no outputs */
-			     : "r" (val), "r" (addr), "i" (ASI_PHYS_BYPASS_EC_E));
-}
-
-#define MOSTEK_EEPROM		0x0000UL
-#define MOSTEK_IDPROM		0x07d8UL
-#define MOSTEK_CREG		0x07f8UL
-#define MOSTEK_SEC		0x07f9UL
-#define MOSTEK_MIN		0x07faUL
-#define MOSTEK_HOUR		0x07fbUL
-#define MOSTEK_DOW		0x07fcUL
-#define MOSTEK_DOM		0x07fdUL
-#define MOSTEK_MONTH		0x07feUL
-#define MOSTEK_YEAR		0x07ffUL
-
-extern spinlock_t mostek_lock;
-extern void __iomem *mstk48t02_regs;
-
-/* Control register values. */
-#define	MSTK_CREG_WRITE	0x80	/* Must set this before placing values. */
-#define	MSTK_CREG_READ	0x40	/* Stop updates to allow a clean read. */
-#define	MSTK_CREG_SIGN	0x20	/* Slow/speed clock in calibration mode. */
-
-/* Control bits that live in the other registers. */
-#define	MSTK_STOP	0x80	/* Stop the clock oscillator. (sec) */
-#define	MSTK_KICK_START	0x80	/* Kick start the clock chip. (hour) */
-#define MSTK_FREQ_TEST	0x40	/* Frequency test mode. (day) */
-
-#define MSTK_YEAR_ZERO       1968   /* If year reg has zero, it is 1968. */
-#define MSTK_CVT_YEAR(yr)  ((yr) + MSTK_YEAR_ZERO)
-
-/* Masks that define how much space each value takes up. */
-#define	MSTK_SEC_MASK	0x7f
-#define	MSTK_MIN_MASK	0x7f
-#define	MSTK_HOUR_MASK	0x3f
-#define	MSTK_DOW_MASK	0x07
-#define	MSTK_DOM_MASK	0x3f
-#define	MSTK_MONTH_MASK	0x1f
-#define	MSTK_YEAR_MASK	0xffU
-
-/* Binary coded decimal conversion macros. */
-#define MSTK_REGVAL_TO_DECIMAL(x)  (((x) & 0x0F) + 0x0A * ((x) >> 0x04))
-#define MSTK_DECIMAL_TO_REGVAL(x)  ((((x) / 0x0A) << 0x04) + ((x) % 0x0A))
-
-/* Generic register set and get macros for internal use. */
-#define MSTK_GET(regs,name)	\
-	(MSTK_REGVAL_TO_DECIMAL(mostek_read(regs + MOSTEK_ ## name) & MSTK_ ## name ## _MASK))
-#define MSTK_SET(regs,name,value) \
-do {	u8 __val = mostek_read(regs + MOSTEK_ ## name); \
-	__val &= ~(MSTK_ ## name ## _MASK); \
-	__val |= (MSTK_DECIMAL_TO_REGVAL(value) & \
-		  (MSTK_ ## name ## _MASK)); \
-	mostek_write(regs + MOSTEK_ ## name, __val); \
-} while(0)
-
-/* Macros to make register access easier on our fingers. These give you
- * the decimal value of the register requested if applicable. You pass
- * the a pointer to a 'struct mostek48t02'.
- */
-#define	MSTK_REG_CREG(regs)	(mostek_read((regs) + MOSTEK_CREG))
-#define	MSTK_REG_SEC(regs)	MSTK_GET(regs,SEC)
-#define	MSTK_REG_MIN(regs)	MSTK_GET(regs,MIN)
-#define	MSTK_REG_HOUR(regs)	MSTK_GET(regs,HOUR)
-#define	MSTK_REG_DOW(regs)	MSTK_GET(regs,DOW)
-#define	MSTK_REG_DOM(regs)	MSTK_GET(regs,DOM)
-#define	MSTK_REG_MONTH(regs)	MSTK_GET(regs,MONTH)
-#define	MSTK_REG_YEAR(regs)	MSTK_GET(regs,YEAR)
-
-#define	MSTK_SET_REG_SEC(regs,value)	MSTK_SET(regs,SEC,value)
-#define	MSTK_SET_REG_MIN(regs,value)	MSTK_SET(regs,MIN,value)
-#define	MSTK_SET_REG_HOUR(regs,value)	MSTK_SET(regs,HOUR,value)
-#define	MSTK_SET_REG_DOW(regs,value)	MSTK_SET(regs,DOW,value)
-#define	MSTK_SET_REG_DOM(regs,value)	MSTK_SET(regs,DOM,value)
-#define	MSTK_SET_REG_MONTH(regs,value)	MSTK_SET(regs,MONTH,value)
-#define	MSTK_SET_REG_YEAR(regs,value)	MSTK_SET(regs,YEAR,value)
-
-
-/* The Mostek 48t08 clock chip. Found on Sun4m's I think. It has the
- * same (basically) layout of the 48t02 chip except for the extra
- * NVRAM on board (8 KB against the 48t02's 2 KB).
- */
-#define MOSTEK_48T08_OFFSET	0x0000UL	/* Lower NVRAM portions */
-#define MOSTEK_48T08_48T02	0x1800UL	/* Offset to 48T02 chip */
-
-/* SUN5 systems usually have 48t59 model clock chipsets.  But we keep the older
- * clock chip definitions around just in case.
- */
-#define MOSTEK_48T59_OFFSET	0x0000UL	/* Lower NVRAM portions */
-#define MOSTEK_48T59_48T02	0x1800UL	/* Offset to 48T02 chip */
-
-#endif /* !(_SPARC64_MOSTEK_H) */
diff --git a/arch/sparc/include/asm/obio.h b/arch/sparc/include/asm/obio.h
index 1a7544ceb574..4ade0c8a2c79 100644
--- a/arch/sparc/include/asm/obio.h
+++ b/arch/sparc/include/asm/obio.h
@@ -155,17 +155,6 @@ static inline void bw_set_ctrl(int cpu, unsigned ctrl)
 			      "i" (ASI_M_CTL));
 }
 
-extern unsigned char cpu_leds[32];
-
-static inline void show_leds(int cpuid)
-{
-	cpuid &= 0x1e;
-	__asm__ __volatile__ ("stba %0, [%1] %2" : :
-			      "r" ((cpu_leds[cpuid] << 4) | cpu_leds[cpuid+1]),
-			      "r" (ECSR_BASE(cpuid) | BB_LEDS),
-			      "i" (ASI_M_CTL));
-}
-
 static inline unsigned cc_get_ipen(void)
 {
 	unsigned pending;
diff --git a/arch/sparc/include/asm/of_device.h b/arch/sparc/include/asm/of_device.h
index bba777a416d3..a5d9811f9697 100644
--- a/arch/sparc/include/asm/of_device.h
+++ b/arch/sparc/include/asm/of_device.h
@@ -30,6 +30,8 @@ struct of_device
 extern void __iomem *of_ioremap(struct resource *res, unsigned long offset, unsigned long size, char *name);
 extern void of_iounmap(struct resource *res, void __iomem *base, unsigned long size);
 
+extern void of_propagate_archdata(struct of_device *bus);
+
 /* This is just here during the transition */
 #include <linux/of_platform.h>
 
diff --git a/arch/sparc/include/asm/of_platform.h b/arch/sparc/include/asm/of_platform.h
index 2348ab90a57c..90da99059f83 100644
--- a/arch/sparc/include/asm/of_platform.h
+++ b/arch/sparc/include/asm/of_platform.h
@@ -13,9 +13,6 @@
  *
  */
 
-extern struct bus_type ebus_bus_type;
-extern struct bus_type sbus_bus_type;
-
 #define of_bus_type	of_platform_bus_type	/* for compatibility */
 
 #endif
diff --git a/arch/sparc/include/asm/oplib_32.h b/arch/sparc/include/asm/oplib_32.h
index b2631da259e0..699da05235c8 100644
--- a/arch/sparc/include/asm/oplib_32.h
+++ b/arch/sparc/include/asm/oplib_32.h
@@ -21,7 +21,6 @@ enum prom_major_version {
 	PROM_V2,      /* sun4c and early sun4m V2 prom */
 	PROM_V3,      /* sun4m and later, up to sun4d/sun4e machines V3 */
 	PROM_P1275,   /* IEEE compliant ISA based Sun PROM, only sun4u */
-	PROM_SUN4,    /* Old sun4 proms are totally different, but we'll shoehorn it to make it fit */
 };
 
 extern enum prom_major_version prom_vers;
diff --git a/arch/sparc/include/asm/page_32.h b/arch/sparc/include/asm/page_32.h
index cf5fb70ca1c1..d1806edc0958 100644
--- a/arch/sparc/include/asm/page_32.h
+++ b/arch/sparc/include/asm/page_32.h
@@ -8,11 +8,8 @@
 #ifndef _SPARC_PAGE_H
 #define _SPARC_PAGE_H
 
-#ifdef CONFIG_SUN4
-#define PAGE_SHIFT   13
-#else
 #define PAGE_SHIFT   12
-#endif
+
 #ifndef __ASSEMBLY__
 /* I have my suspicions... -DaveM */
 #define PAGE_SIZE    (1UL << PAGE_SHIFT)
diff --git a/arch/sparc/include/asm/page_64.h b/arch/sparc/include/asm/page_64.h
index b579b910ef51..4274ed13ddb2 100644
--- a/arch/sparc/include/asm/page_64.h
+++ b/arch/sparc/include/asm/page_64.h
@@ -38,6 +38,8 @@
 
 #ifndef __ASSEMBLY__
 
+#define WANT_PAGE_VIRTUAL
+
 extern void _clear_page(void *page);
 #define clear_page(X)	_clear_page((void *)(X))
 struct page;
diff --git a/arch/sparc/include/asm/parport.h b/arch/sparc/include/asm/parport.h
index d9830621c906..dff3f0253aa8 100644
--- a/arch/sparc/include/asm/parport.h
+++ b/arch/sparc/include/asm/parport.h
@@ -8,7 +8,7 @@
 
 #include <linux/of_device.h>
 
-#include <asm/ebus.h>
+#include <asm/ebus_dma.h>
 #include <asm/ns87303.h>
 #include <asm/prom.h>
 
@@ -215,7 +215,7 @@ static int __devexit ecpp_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id ecpp_match[] = {
+static const struct of_device_id ecpp_match[] = {
 	{
 		.name = "ecpp",
 	},
diff --git a/arch/sparc/include/asm/pci_32.h b/arch/sparc/include/asm/pci_32.h
index 0ee949d220c0..b41c4c198159 100644
--- a/arch/sparc/include/asm/pci_32.h
+++ b/arch/sparc/include/asm/pci_32.h
@@ -3,6 +3,8 @@
 
 #ifdef __KERNEL__
 
+#include <linux/dma-mapping.h>
+
 /* Can be used to override the logic in pci_scan_bus for skipping
  * already-configured bus numbers - to be used for buggy BIOSes
  * or architectures with incomplete PCI setup by the loader.
diff --git a/arch/sparc/include/asm/pgtable_32.h b/arch/sparc/include/asm/pgtable_32.h
index 08237fda8874..e0cabe790ec1 100644
--- a/arch/sparc/include/asm/pgtable_32.h
+++ b/arch/sparc/include/asm/pgtable_32.h
@@ -14,11 +14,7 @@
 #include <linux/spinlock.h>
 #include <linux/swap.h>
 #include <asm/types.h>
-#ifdef CONFIG_SUN4
-#include <asm/pgtsun4.h>
-#else
 #include <asm/pgtsun4c.h>
-#endif
 #include <asm/pgtsrmmu.h>
 #include <asm/vac-ops.h>
 #include <asm/oplib.h>
diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h
index bb9ec2cce355..b049abf9902f 100644
--- a/arch/sparc/include/asm/pgtable_64.h
+++ b/arch/sparc/include/asm/pgtable_64.h
@@ -770,6 +770,8 @@ extern void sun4v_patch_tlb_handlers(void);
 
 extern unsigned long cmdline_memory_size;
 
+extern asmlinkage void do_sparc64_fault(struct pt_regs *regs);
+
 #endif /* !(__ASSEMBLY__) */
 
 #endif /* !(_SPARC64_PGTABLE_H) */
diff --git a/arch/sparc/include/asm/prom.h b/arch/sparc/include/asm/prom.h
index fd55522481cd..900d44714f8d 100644
--- a/arch/sparc/include/asm/prom.h
+++ b/arch/sparc/include/asm/prom.h
@@ -18,6 +18,7 @@
  */
 #include <linux/types.h>
 #include <linux/proc_fs.h>
+#include <linux/mutex.h>
 #include <asm/atomic.h>
 
 #define OF_ROOT_NODE_ADDR_CELLS_DEFAULT	2
@@ -73,6 +74,7 @@ struct of_irq_controller {
 
 extern struct device_node *of_find_node_by_cpuid(int cpuid);
 extern int of_set_property(struct device_node *node, const char *name, void *val, int len);
+extern struct mutex of_set_property_mutex;
 extern int of_getintprop_default(struct device_node *np,
 				 const char *name,
 				 int def);
@@ -94,6 +96,16 @@ static inline void of_node_put(struct device_node *node)
 {
 }
 
+/* These routines are here to provide compatibility with how powerpc
+ * handles IRQ mapping for OF device nodes.  We precompute and permanently
+ * register them in the of_device objects, whereas powerpc computes them
+ * on request.
+ */
+extern unsigned int irq_of_parse_and_map(struct device_node *node, int index);
+static inline void irq_dispose_mapping(unsigned int virq)
+{
+}
+
 /*
  * NB:  This is here while we transition from using asm/prom.h
  * to linux/of.h
diff --git a/arch/sparc/include/asm/ptrace_64.h b/arch/sparc/include/asm/ptrace_64.h
index 06e4914c13f4..3d3e9c161d8b 100644
--- a/arch/sparc/include/asm/ptrace_64.h
+++ b/arch/sparc/include/asm/ptrace_64.h
@@ -113,6 +113,8 @@ struct sparc_trapf {
 
 #ifdef __KERNEL__
 
+#include <linux/threads.h>
+
 static inline int pt_regs_trap_type(struct pt_regs *regs)
 {
 	return regs->magic & 0x1ff;
@@ -138,6 +140,7 @@ struct global_reg_snapshot {
 	struct thread_info	*thread;
 	unsigned long		pad1;
 };
+extern struct global_reg_snapshot global_reg_snapshot[NR_CPUS];
 
 #define __ARCH_WANT_COMPAT_SYS_PTRACE
 
diff --git a/arch/sparc/include/asm/reboot.h b/arch/sparc/include/asm/reboot.h
deleted file mode 100644
index 3f3f43f5be5e..000000000000
--- a/arch/sparc/include/asm/reboot.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef _SPARC64_REBOOT_H
-#define _SPARC64_REBOOT_H
-
-extern void machine_alt_power_off(void);
-
-#endif /* _SPARC64_REBOOT_H */
diff --git a/arch/sparc/include/asm/rtc.h b/arch/sparc/include/asm/rtc.h
deleted file mode 100644
index f9ecb1fe2ecd..000000000000
--- a/arch/sparc/include/asm/rtc.h
+++ /dev/null
@@ -1,26 +0,0 @@
-/*
- * rtc.h: Definitions for access to the Mostek real time clock
- *
- * Copyright (C) 1996 Thomas K. Dyas (tdyas@eden.rutgers.edu)
- */
-
-#ifndef _RTC_H
-#define _RTC_H
-
-#include <linux/ioctl.h>
-
-struct rtc_time
-{
-	int	sec;	/* Seconds (0-59) */
-	int	min;	/* Minutes (0-59) */
-	int	hour;	/* Hour (0-23) */
-	int	dow;	/* Day of the week (1-7) */
-	int	dom;	/* Day of the month (1-31) */
-	int	month;	/* Month of year (1-12) */
-	int	year;	/* Year (0-99) */
-};
-
-#define RTCGET _IOR('p', 20, struct rtc_time)
-#define RTCSET _IOW('p', 21, struct rtc_time)
-
-#endif
diff --git a/arch/sparc/include/asm/sbus.h b/arch/sparc/include/asm/sbus.h
deleted file mode 100644
index f82481ab44db..000000000000
--- a/arch/sparc/include/asm/sbus.h
+++ /dev/null
@@ -1,8 +0,0 @@
-#ifndef ___ASM_SPARC_SBUS_H
-#define ___ASM_SPARC_SBUS_H
-#if defined(__sparc__) && defined(__arch64__)
-#include <asm/sbus_64.h>
-#else
-#include <asm/sbus_32.h>
-#endif
-#endif
diff --git a/arch/sparc/include/asm/sbus_32.h b/arch/sparc/include/asm/sbus_32.h
deleted file mode 100644
index a7b4fa21931d..000000000000
--- a/arch/sparc/include/asm/sbus_32.h
+++ /dev/null
@@ -1,153 +0,0 @@
-/*
- * sbus.h:  Defines for the Sun SBus.
- *
- * Copyright (C) 1995 David S. Miller (davem@caip.rutgers.edu)
- */
-
-#ifndef _SPARC_SBUS_H
-#define _SPARC_SBUS_H
-
-#include <linux/dma-mapping.h>
-#include <linux/ioport.h>
-#include <linux/of_device.h>
-
-#include <asm/oplib.h>
-#include <asm/prom.h>
-#include <asm/scatterlist.h>
-
-/* We scan which devices are on the SBus using the PROM node device
- * tree.  SBus devices are described in two different ways.  You can
- * either get an absolute address at which to access the device, or
- * you can get a SBus 'slot' number and an offset within that slot.
- */
-
-/* The base address at which to calculate device OBIO addresses. */
-#define SUN_SBUS_BVADDR        0xf8000000
-#define SBUS_OFF_MASK          0x01ffffff
-
-/* These routines are used to calculate device address from slot
- * numbers + offsets, and vice versa.
- */
-
-static inline unsigned long sbus_devaddr(int slotnum, unsigned long offset)
-{
-  return (unsigned long) (SUN_SBUS_BVADDR+((slotnum)<<25)+(offset));
-}
-
-static inline int sbus_dev_slot(unsigned long dev_addr)
-{
-  return (int) (((dev_addr)-SUN_SBUS_BVADDR)>>25);
-}
-
-struct sbus_bus;
-
-/* Linux SBUS device tables */
-struct sbus_dev {
-	struct of_device	ofdev;
-	struct sbus_bus		*bus;
-	struct sbus_dev		*next;
-	struct sbus_dev		*child;
-	struct sbus_dev		*parent;
-	int prom_node;
-	char prom_name[64];
-	int slot;
-
-	struct resource resource[PROMREG_MAX];
-
-	struct linux_prom_registers reg_addrs[PROMREG_MAX];
-	int num_registers;
-
-	struct linux_prom_ranges device_ranges[PROMREG_MAX];
-	int num_device_ranges;
-
-	unsigned int irqs[4];
-	int num_irqs;
-};
-#define to_sbus_device(d) container_of(d, struct sbus_dev, ofdev.dev)
-
-/* This struct describes the SBus(s) found on this machine. */
-struct sbus_bus {
-	struct of_device	ofdev;
-	struct sbus_dev		*devices;	/* Link to devices on this SBus */
-	struct sbus_bus		*next;		/* next SBus, if more than one SBus */
-	int			prom_node;	/* PROM device tree node for this SBus */
-	char			prom_name[64];  /* Usually "sbus" or "sbi" */
-	int			clock_freq;
-
-	struct linux_prom_ranges sbus_ranges[PROMREG_MAX];
-	int num_sbus_ranges;
-
-	int devid;
-	int board;
-};
-#define to_sbus(d) container_of(d, struct sbus_bus, ofdev.dev)
-
-extern struct sbus_bus *sbus_root;
-
-static inline int
-sbus_is_slave(struct sbus_dev *dev)
-{
-	/* XXX Have to write this for sun4c's */
-	return 0;
-}
-
-/* Device probing routines could find these handy */
-#define for_each_sbus(bus) \
-        for((bus) = sbus_root; (bus); (bus)=(bus)->next)
-
-#define for_each_sbusdev(device, bus) \
-        for((device) = (bus)->devices; (device); (device)=(device)->next)
-
-#define for_all_sbusdev(device, bus) \
-	for ((bus) = sbus_root; (bus); (bus) = (bus)->next) \
-		for ((device) = (bus)->devices; (device); (device) = (device)->next)
-
-/* Driver DVMA interfaces. */
-#define sbus_can_dma_64bit(sdev)	(0) /* actually, sparc_cpu_model==sun4d */
-#define sbus_can_burst64(sdev)		(0) /* actually, sparc_cpu_model==sun4d */
-extern void sbus_set_sbus64(struct sbus_dev *, int);
-extern void sbus_fill_device_irq(struct sbus_dev *);
-
-/* These yield IOMMU mappings in consistent mode. */
-extern void *sbus_alloc_consistent(struct sbus_dev *, long, u32 *dma_addrp);
-extern void sbus_free_consistent(struct sbus_dev *, long, void *, u32);
-void prom_adjust_ranges(struct linux_prom_ranges *, int,
-			struct linux_prom_ranges *, int);
-
-#define SBUS_DMA_BIDIRECTIONAL	DMA_BIDIRECTIONAL
-#define SBUS_DMA_TODEVICE	DMA_TO_DEVICE
-#define SBUS_DMA_FROMDEVICE	DMA_FROM_DEVICE
-#define	SBUS_DMA_NONE		DMA_NONE
-
-/* All the rest use streaming mode mappings. */
-extern dma_addr_t sbus_map_single(struct sbus_dev *, void *, size_t, int);
-extern void sbus_unmap_single(struct sbus_dev *, dma_addr_t, size_t, int);
-extern int sbus_map_sg(struct sbus_dev *, struct scatterlist *, int, int);
-extern void sbus_unmap_sg(struct sbus_dev *, struct scatterlist *, int, int);
-
-/* Finally, allow explicit synchronization of streamable mappings. */
-extern void sbus_dma_sync_single_for_cpu(struct sbus_dev *, dma_addr_t, size_t, int);
-#define sbus_dma_sync_single sbus_dma_sync_single_for_cpu
-extern void sbus_dma_sync_single_for_device(struct sbus_dev *, dma_addr_t, size_t, int);
-extern void sbus_dma_sync_sg_for_cpu(struct sbus_dev *, struct scatterlist *, int, int);
-#define sbus_dma_sync_sg sbus_dma_sync_sg_for_cpu
-extern void sbus_dma_sync_sg_for_device(struct sbus_dev *, struct scatterlist *, int, int);
-
-/* Eric Brower (ebrower@usa.net)
- * Translate SBus interrupt levels to ino values--
- * this is used when converting sbus "interrupts" OBP
- * node values to "intr" node values, and is platform
- * dependent.  If only we could call OBP with
- * "sbus-intr>cpu (sbint -- ino)" from kernel...
- * See .../drivers/sbus/sbus.c for details.
- */
-BTFIXUPDEF_CALL(unsigned int, sbint_to_irq, struct sbus_dev *sdev, unsigned int)
-#define sbint_to_irq(sdev, sbint) BTFIXUP_CALL(sbint_to_irq)(sdev, sbint)
-
-extern void sbus_arch_bus_ranges_init(struct device_node *, struct sbus_bus *);
-extern void sbus_setup_iommu(struct sbus_bus *, struct device_node *);
-extern void sbus_setup_arch_props(struct sbus_bus *, struct device_node *);
-extern int sbus_arch_preinit(void);
-extern void sbus_arch_postinit(void);
-
-#endif /* !(_SPARC_SBUS_H) */
diff --git a/arch/sparc/include/asm/sbus_64.h b/arch/sparc/include/asm/sbus_64.h
deleted file mode 100644
index b606c14343fb..000000000000
--- a/arch/sparc/include/asm/sbus_64.h
+++ /dev/null
@@ -1,190 +0,0 @@
-/* sbus.h: Defines for the Sun SBus.
- *
- * Copyright (C) 1996, 1999, 2007 David S. Miller (davem@davemloft.net)
- */
-
-#ifndef _SPARC64_SBUS_H
-#define _SPARC64_SBUS_H
-
-#include <linux/dma-mapping.h>
-#include <linux/ioport.h>
-#include <linux/of_device.h>
-
-#include <asm/oplib.h>
-#include <asm/prom.h>
-#include <asm/iommu.h>
-#include <asm/scatterlist.h>
-
-/* We scan which devices are on the SBus using the PROM node device
- * tree.  SBus devices are described in two different ways.  You can
- * either get an absolute address at which to access the device, or
- * you can get a SBus 'slot' number and an offset within that slot.
- */
-
-/* The base address at which to calculate device OBIO addresses. */
-#define SUN_SBUS_BVADDR        0x00000000
-#define SBUS_OFF_MASK          0x0fffffff
-
-/* These routines are used to calculate device address from slot
- * numbers + offsets, and vice versa.
- */
-
-static inline unsigned long sbus_devaddr(int slotnum, unsigned long offset)
-{
-  return (unsigned long) (SUN_SBUS_BVADDR+((slotnum)<<28)+(offset));
-}
-
-static inline int sbus_dev_slot(unsigned long dev_addr)
-{
-  return (int) (((dev_addr)-SUN_SBUS_BVADDR)>>28);
-}
-
-struct sbus_bus;
-
-/* Linux SBUS device tables */
-struct sbus_dev {
-	struct of_device	ofdev;
-	struct sbus_bus		*bus;
-	struct sbus_dev		*next;
-	struct sbus_dev		*child;
-	struct sbus_dev		*parent;
-	int prom_node;
-	char prom_name[64];
-	int slot;
-
-	struct resource resource[PROMREG_MAX];
-
-	struct linux_prom_registers reg_addrs[PROMREG_MAX];
-	int num_registers;
-
-	struct linux_prom_ranges device_ranges[PROMREG_MAX];
-	int num_device_ranges;
-
-	unsigned int irqs[4];
-	int num_irqs;
-};
-#define to_sbus_device(d) container_of(d, struct sbus_dev, ofdev.dev)
-
-/* This struct describes the SBus(s) found on this machine. */
-struct sbus_bus {
-	struct of_device	ofdev;
-	struct sbus_dev		*devices;	/* Tree of SBUS devices	*/
-	struct sbus_bus		*next;		/* Next SBUS in system	*/
-	int			prom_node;      /* OBP node of SBUS	*/
-	char			prom_name[64];	/* Usually "sbus" or "sbi" */
-	int			clock_freq;
-
-	struct linux_prom_ranges sbus_ranges[PROMREG_MAX];
-	int num_sbus_ranges;
-
-	int portid;
-};
-#define to_sbus(d) container_of(d, struct sbus_bus, ofdev.dev)
-
-extern struct sbus_bus *sbus_root;
-
-/* Device probing routines could find these handy */
-#define for_each_sbus(bus) \
-        for((bus) = sbus_root; (bus); (bus)=(bus)->next)
-
-#define for_each_sbusdev(device, bus) \
-        for((device) = (bus)->devices; (device); (device)=(device)->next)
-
-#define for_all_sbusdev(device, bus) \
-	for ((bus) = sbus_root; (bus); (bus) = (bus)->next) \
-		for ((device) = (bus)->devices; (device); (device) = (device)->next)
-
-/* Driver DVMA interfaces. */
-#define sbus_can_dma_64bit(sdev)	(1)
-#define sbus_can_burst64(sdev)		(1)
-extern void sbus_set_sbus64(struct sbus_dev *, int);
-extern void sbus_fill_device_irq(struct sbus_dev *);
-
-static inline void *sbus_alloc_consistent(struct sbus_dev *sdev , size_t size,
-					  dma_addr_t *dma_handle)
-{
-	return dma_alloc_coherent(&sdev->ofdev.dev, size,
-				  dma_handle, GFP_ATOMIC);
-}
-
-static inline void sbus_free_consistent(struct sbus_dev *sdev, size_t size,
-					void *vaddr, dma_addr_t dma_handle)
-{
-	return dma_free_coherent(&sdev->ofdev.dev, size, vaddr, dma_handle);
-}
-
-#define SBUS_DMA_BIDIRECTIONAL	DMA_BIDIRECTIONAL
-#define SBUS_DMA_TODEVICE	DMA_TO_DEVICE
-#define SBUS_DMA_FROMDEVICE	DMA_FROM_DEVICE
-#define	SBUS_DMA_NONE		DMA_NONE
-
-/* All the rest use streaming mode mappings. */
-static inline dma_addr_t sbus_map_single(struct sbus_dev *sdev, void *ptr,
-					 size_t size, int direction)
-{
-	return dma_map_single(&sdev->ofdev.dev, ptr, size,
-			      (enum dma_data_direction) direction);
-}
-
-static inline void sbus_unmap_single(struct sbus_dev *sdev,
-				     dma_addr_t dma_addr, size_t size,
-				     int direction)
-{
-	dma_unmap_single(&sdev->ofdev.dev, dma_addr, size,
-			 (enum dma_data_direction) direction);
-}
-
-static inline int sbus_map_sg(struct sbus_dev *sdev, struct scatterlist *sg,
-			      int nents, int direction)
-{
-	return dma_map_sg(&sdev->ofdev.dev, sg, nents,
-			  (enum dma_data_direction) direction);
-}
-
-static inline void sbus_unmap_sg(struct sbus_dev *sdev, struct scatterlist *sg,
-				 int nents, int direction)
-{
-	dma_unmap_sg(&sdev->ofdev.dev, sg, nents,
-		     (enum dma_data_direction) direction);
-}
-
-/* Finally, allow explicit synchronization of streamable mappings. */
-static inline void sbus_dma_sync_single_for_cpu(struct sbus_dev *sdev,
-						dma_addr_t dma_handle,
-						size_t size, int direction)
-{
-	dma_sync_single_for_cpu(&sdev->ofdev.dev, dma_handle, size,
-				(enum dma_data_direction) direction);
-}
-#define sbus_dma_sync_single sbus_dma_sync_single_for_cpu
-
-static inline void sbus_dma_sync_single_for_device(struct sbus_dev *sdev,
-						   dma_addr_t dma_handle,
-						   size_t size, int direction)
-{
-	/* No flushing needed to sync cpu writes to the device.  */
-}
-
-static inline void sbus_dma_sync_sg_for_cpu(struct sbus_dev *sdev,
-					    struct scatterlist *sg,
-					    int nents, int direction)
-{
-	dma_sync_sg_for_cpu(&sdev->ofdev.dev, sg, nents,
-			    (enum dma_data_direction) direction);
-}
-#define sbus_dma_sync_sg sbus_dma_sync_sg_for_cpu
-
-static inline void sbus_dma_sync_sg_for_device(struct sbus_dev *sdev,
-					       struct scatterlist *sg,
-					       int nents, int direction)
-{
-	/* No flushing needed to sync cpu writes to the device.  */
-}
-
-extern void sbus_arch_bus_ranges_init(struct device_node *, struct sbus_bus *);
-extern void sbus_setup_iommu(struct sbus_bus *, struct device_node *);
-extern void sbus_setup_arch_props(struct sbus_bus *, struct device_node *);
-extern int sbus_arch_preinit(void);
-extern void sbus_arch_postinit(void);
-
-#endif /* !(_SPARC64_SBUS_H) */
diff --git a/arch/sparc/include/asm/serial.h b/arch/sparc/include/asm/serial.h
new file mode 100644
index 000000000000..f90d61c28059
--- /dev/null
+++ b/arch/sparc/include/asm/serial.h
@@ -0,0 +1,6 @@
+#ifndef __SPARC_SERIAL_H
+#define __SPARC_SERIAL_H
+
+#define BASE_BAUD ( 1843200 / 16 )
+
+#endif /* __SPARC_SERIAL_H */
diff --git a/arch/sparc/include/asm/spinlock_32.h b/arch/sparc/include/asm/spinlock_32.h
index de2249b267c6..bf2d532593e3 100644
--- a/arch/sparc/include/asm/spinlock_32.h
+++ b/arch/sparc/include/asm/spinlock_32.h
@@ -6,8 +6,6 @@
 #ifndef __SPARC_SPINLOCK_H
 #define __SPARC_SPINLOCK_H
 
-#include <linux/threads.h>	/* For NR_CPUS */
-
 #ifndef __ASSEMBLY__
 
 #include <asm/psr.h>
diff --git a/arch/sparc/include/asm/spinlock_64.h b/arch/sparc/include/asm/spinlock_64.h
index 0006fe9f8c7a..120cfe4577c7 100644
--- a/arch/sparc/include/asm/spinlock_64.h
+++ b/arch/sparc/include/asm/spinlock_64.h
@@ -6,8 +6,6 @@
 #ifndef __SPARC64_SPINLOCK_H
 #define __SPARC64_SPINLOCK_H
 
-#include <linux/threads.h>	/* For NR_CPUS */
-
 #ifndef __ASSEMBLY__
 
 /* To get debugging spinlocks which detect and catch
diff --git a/arch/sparc/include/asm/sstate.h b/arch/sparc/include/asm/sstate.h
deleted file mode 100644
index a7c35dbcb281..000000000000
--- a/arch/sparc/include/asm/sstate.h
+++ /dev/null
@@ -1,13 +0,0 @@
-#ifndef _SPARC64_SSTATE_H
-#define _SPARC64_SSTATE_H
-
-extern void sstate_booting(void);
-extern void sstate_running(void);
-extern void sstate_halt(void);
-extern void sstate_poweroff(void);
-extern void sstate_panic(void);
-extern void sstate_reboot(void);
-
-extern void sun4v_sstate_init(void);
-
-#endif /* _SPARC64_SSTATE_H */
diff --git a/arch/sparc/include/asm/starfire.h b/arch/sparc/include/asm/starfire.h
index 07bafd31e33c..d56ce60a5992 100644
--- a/arch/sparc/include/asm/starfire.h
+++ b/arch/sparc/include/asm/starfire.h
@@ -12,7 +12,6 @@
 extern int this_is_starfire;
 
 extern void check_if_starfire(void);
-extern void starfire_cpu_setup(void);
 extern int starfire_hard_smp_processor_id(void);
 extern void starfire_hookup(int);
 extern unsigned int starfire_translate(unsigned long imap, unsigned int upaid);
diff --git a/arch/sparc/include/asm/statfs.h b/arch/sparc/include/asm/statfs.h
index 5e937a73743d..55e607ad461d 100644
--- a/arch/sparc/include/asm/statfs.h
+++ b/arch/sparc/include/asm/statfs.h
@@ -1,8 +1,6 @@
 #ifndef ___ASM_SPARC_STATFS_H
 #define ___ASM_SPARC_STATFS_H
-#if defined(__sparc__) && defined(__arch64__)
-#include <asm/statfs_64.h>
-#else
-#include <asm/statfs_32.h>
-#endif
+
+#include <asm-generic/statfs.h>
+
 #endif
diff --git a/arch/sparc/include/asm/statfs_32.h b/arch/sparc/include/asm/statfs_32.h
deleted file mode 100644
index 304520fa8863..000000000000
--- a/arch/sparc/include/asm/statfs_32.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef _SPARC_STATFS_H
-#define _SPARC_STATFS_H
-
-#include <asm-generic/statfs.h>
-
-#endif
diff --git a/arch/sparc/include/asm/statfs_64.h b/arch/sparc/include/asm/statfs_64.h
deleted file mode 100644
index 79b3c890a5fa..000000000000
--- a/arch/sparc/include/asm/statfs_64.h
+++ /dev/null
@@ -1,54 +0,0 @@
-#ifndef _SPARC64_STATFS_H
-#define _SPARC64_STATFS_H
-
-#ifndef __KERNEL_STRICT_NAMES
-
-#include <linux/types.h>
-
-typedef __kernel_fsid_t	fsid_t;
-
-#endif
-
-struct statfs {
-	long f_type;
-	long f_bsize;
-	long f_blocks;
-	long f_bfree;
-	long f_bavail;
-	long f_files;
-	long f_ffree;
-	__kernel_fsid_t f_fsid;
-	long f_namelen;
-	long f_frsize;
-	long f_spare[5];
-};
-
-struct statfs64 {
-	long f_type;
-	long f_bsize;
-	long f_blocks;
-	long f_bfree;
-	long f_bavail;
-	long f_files;
-	long f_ffree;
-	__kernel_fsid_t f_fsid;
-	long f_namelen;
-	long f_frsize;
-	long f_spare[5];
-};
-
-struct compat_statfs64 {
-	__u32 f_type;
-	__u32 f_bsize;
-	__u64 f_blocks;
-	__u64 f_bfree;
-	__u64 f_bavail;
-	__u64 f_files;
-	__u64 f_ffree;
-	__kernel_fsid_t f_fsid;
-	__u32 f_namelen;
-	__u32 f_frsize;
-	__u32 f_spare[5];
-};
-
-#endif
diff --git a/arch/sparc/include/asm/sun4paddr.h b/arch/sparc/include/asm/sun4paddr.h
deleted file mode 100644
index d52985f19f42..000000000000
--- a/arch/sparc/include/asm/sun4paddr.h
+++ /dev/null
@@ -1,56 +0,0 @@
-/*
- * sun4paddr.h:  Various physical addresses on sun4 machines
- *
- * Copyright (C) 1997 Anton Blanchard (anton@progsoc.uts.edu.au)
- * Copyright (C) 1998 Chris Davis (cdavis@cois.on.ca)
- * 
- * Now supports more sun4's
- */
-
-#ifndef _SPARC_SUN4PADDR_H
-#define _SPARC_SUN4PADDR_H
-
-#define SUN4_IE_PHYSADDR		0xf5000000
-#define SUN4_UNUSED_PHYSADDR		0
-
-/* these work for me */
-#define SUN4_200_MEMREG_PHYSADDR	0xf4000000
-#define SUN4_200_CLOCK_PHYSADDR		0xf3000000
-#define SUN4_200_BWTWO_PHYSADDR		0xfd000000
-#define SUN4_200_ETH_PHYSADDR		0xf6000000
-#define SUN4_200_SI_PHYSADDR		0xff200000
-
-/* these were here before */
-#define SUN4_300_MEMREG_PHYSADDR	0xf4000000
-#define SUN4_300_CLOCK_PHYSADDR		0xf2000000
-#define SUN4_300_TIMER_PHYSADDR		0xef000000
-#define SUN4_300_ETH_PHYSADDR		0xf9000000
-#define SUN4_300_BWTWO_PHYSADDR		0xfb400000
-#define SUN4_300_DMA_PHYSADDR		0xfa001000
-#define SUN4_300_ESP_PHYSADDR		0xfa000000
-
-/* Are these right? */
-#define SUN4_400_MEMREG_PHYSADDR	0xf4000000
-#define SUN4_400_CLOCK_PHYSADDR		0xf2000000
-#define SUN4_400_TIMER_PHYSADDR		0xef000000
-#define SUN4_400_ETH_PHYSADDR		0xf9000000
-#define SUN4_400_BWTWO_PHYSADDR		0xfb400000
-#define SUN4_400_DMA_PHYSADDR		0xfa001000
-#define SUN4_400_ESP_PHYSADDR		0xfa000000
-
-/* 
-	these are the actual values set and used in the code. Unused items set 
-	to SUN_UNUSED_PHYSADDR 
- */
-
-extern int sun4_memreg_physaddr; /* memory register (ecc?) */
-extern int sun4_clock_physaddr;  /* system clock */
-extern int sun4_timer_physaddr;  /* timer, where applicable */
-extern int sun4_eth_physaddr;    /* onboard ethernet (ie/le) */
-extern int sun4_si_physaddr;     /* sun3 scsi adapter */
-extern int sun4_bwtwo_physaddr;  /* onboard bw2 */
-extern int sun4_dma_physaddr;    /* scsi dma */
-extern int sun4_esp_physaddr;    /* esp scsi */
-extern int sun4_ie_physaddr;     /* interrupt enable */
-
-#endif /* !(_SPARC_SUN4PADDR_H) */
diff --git a/arch/sparc/include/asm/sun4prom.h b/arch/sparc/include/asm/sun4prom.h
deleted file mode 100644
index 9c8b4cbf629a..000000000000
--- a/arch/sparc/include/asm/sun4prom.h
+++ /dev/null
@@ -1,83 +0,0 @@
-/*
- * sun4prom.h -- interface to sun4 PROM monitor.  We don't use most of this,
- *               so most of these are just placeholders.
- */
-
-#ifndef _SUN4PROM_H_
-#define _SUN4PROM_H_
-
-/*
- * Although this looks similar to an romvec for a OpenProm machine, it is 
- * actually closer to what was used in the Sun2 and Sun3.
- *
- * V2 entries exist only in version 2 PROMs and later, V3 in version 3 and later.
- * 
- * Many of the function prototypes are guesses.  Some are certainly wrong.
- * Use with care.
- */
-
-typedef struct {
-	char		*initSP;		/* Initial system stack ptr */
-	void		(*startmon)(void);	/* Initial PC for hardware */
-	int		*diagberr;		/* Bus err handler for diags */
-	struct linux_arguments_v0 **bootParam; /* Info for bootstrapped pgm */
- 	unsigned int	*memorysize;		/* Usable memory in bytes */
-	unsigned char	(*getchar)(void);	/* Get char from input device */ 
-	void		(*putchar)(char);	/* Put char to output device */
-	int		(*mayget)(void);	/* Maybe get char, or -1 */
-	int		(*mayput)(int);		/* Maybe put char, or -1 */
-	unsigned char	*echo;			/* Should getchar echo? */
-	unsigned char	*insource;		/* Input source selector */
-	unsigned char	*outsink;		/* Output sink selector */
-	int		(*getkey)(void);	/* Get next key if one exists */
-	void		(*initgetkey)(void);	/* Initialize get key */
-	unsigned int	*translation;		/* Kbd translation selector */
-	unsigned char	*keybid;		/* Keyboard ID byte */
-	int		*screen_x;		/* V2: Screen x pos (r/o) */
-	int		*screen_y;		/* V2: Screen y pos (r/o) */
-	struct keybuf	*keybuf;		/* Up/down keycode buffer */
-	char		*monid;			/* Monitor version ID */
-	void		(*fbwritechar)(char);	/* Write a character to FB */
-	int		*fbAddr;		/* Address of frame buffer */
-	char		**font;			/* Font table for FB */
-	void		(*fbwritestr)(char *);	/* Write string to FB */
-	void		(*reboot)(char *);	/* e.g. reboot("sd()vmlinux") */
-	unsigned char	*linebuf;		/* The line input buffer */
-	unsigned char	**lineptr;		/* Cur pointer into linebuf */
-	int		*linesize;		/* length of line in linebuf */
-	void		(*getline)(char *);	/* Get line from user */
-	unsigned char	(*getnextchar)(void);	/* Get next char from linebuf */
-	unsigned char	(*peeknextchar)(void);	/* Peek at next char */
-	int		*fbthere;		/* =1 if frame buffer there */
-	int		(*getnum)(void);	/* Grab hex num from line */
-	int		(*printf)(char *, ...);	/* See prom_printf() instead */ 
-	void		(*printhex)(int);	/* Format N digits in hex */
-	unsigned char	*leds;			/* RAM copy of LED register */
-	void		(*setLEDs)(unsigned char *);	/* Sets LED's and RAM copy */
-	void		(*NMIaddr)(void *);	/* Addr for level 7 vector */
-	void		(*abortentry)(void);	/* Entry for keyboard abort */
-	int		*nmiclock;		/* Counts up in msec */
-	int		*FBtype;		/* Frame buffer type */
- 	unsigned int	romvecversion;		/* Version number for this romvec */
-	struct globram  *globram;		/* monitor global variables ??? */
-	void *		kbdaddr;		/* Addr of keyboard in use */
-	int		*keyrinit;		/* ms before kbd repeat */
-	unsigned char	*keyrtick; 		/* ms between repetitions */
-	unsigned int	*memoryavail;		/* V1: Main mem usable size */
-	long		*resetaddr;		/* where to jump on a reset */
-	long		*resetmap;		/* pgmap entry for resetaddr */
-	void		(*exittomon)(void);	/* Exit from user program */
-	unsigned char	**memorybitmap;		/* V1: &{0 or &bits} */
-	void		(*setcxsegmap)(int ctxt, char *va, int pmeg);	/* Set seg in any context */
-	void		(**vector_cmd)(void *);	/* V2: Handler for 'v' cmd */
-	unsigned long	*expectedtrapsig;	/* V3: Location of the expected trap signal */
-	unsigned long	*trapvectorbasetable;	/* V3: Address of the trap vector table */
-	int		unused1;
-	int		unused2;
-	int		unused3;
-	int		unused4;
-} linux_sun4_romvec;
-
-extern linux_sun4_romvec *sun4_romvec;
-
-#endif /* _SUN4PROM_H_ */
diff --git a/arch/sparc/include/asm/system_32.h b/arch/sparc/include/asm/system_32.h
index b4b024445fc9..8623fc48fe24 100644
--- a/arch/sparc/include/asm/system_32.h
+++ b/arch/sparc/include/asm/system_32.h
@@ -34,13 +34,7 @@ enum sparc_cpu {
 
 extern enum sparc_cpu sparc_cpu_model;
 
-#ifndef CONFIG_SUN4
-#define ARCH_SUN4C_SUN4 (sparc_cpu_model==sun4c)
-#define ARCH_SUN4 0
-#else
-#define ARCH_SUN4C_SUN4 1
-#define ARCH_SUN4 1
-#endif
+#define ARCH_SUN4C (sparc_cpu_model==sun4c)
 
 #define SUN4M_NCPUS            4              /* Architectural limit of sun4m. */
 
@@ -55,6 +49,7 @@ extern unsigned long empty_zero_page;
 extern void sun_do_break(void);
 extern int serial_console;
 extern int stop_a_enabled;
+extern int scons_pwroff;
 
 static inline int con_is_present(void)
 {
diff --git a/arch/sparc/include/asm/system_64.h b/arch/sparc/include/asm/system_64.h
index db9e742a406a..8759f2a1b837 100644
--- a/arch/sparc/include/asm/system_64.h
+++ b/arch/sparc/include/asm/system_64.h
@@ -26,9 +26,8 @@ enum sparc_cpu {
 
 #define sparc_cpu_model sun4u
 
-/* This cannot ever be a sun4c nor sun4 :) That's just history. */
-#define ARCH_SUN4C_SUN4 0
-#define ARCH_SUN4 0
+/* This cannot ever be a sun4c :) That's just history. */
+#define ARCH_SUN4C 0
 
 extern char reboot_command[];
 
@@ -118,6 +117,7 @@ do {	__asm__ __volatile__("ba,pt	%%xcc, 1f\n\t" \
 
 extern void sun_do_break(void);
 extern int stop_a_enabled;
+extern int scons_pwroff;
 
 extern void fault_in_user_windows(void);
 extern void synchronize_user_stack(void);
diff --git a/arch/sparc/include/asm/thread_info_32.h b/arch/sparc/include/asm/thread_info_32.h
index cbb892d0dff0..29899fd5b1b2 100644
--- a/arch/sparc/include/asm/thread_info_32.h
+++ b/arch/sparc/include/asm/thread_info_32.h
@@ -80,11 +80,7 @@ register struct thread_info *current_thread_info_reg asm("g6");
 /*
  * thread information allocation
  */
-#if PAGE_SHIFT == 13
-#define THREAD_INFO_ORDER  0
-#else /* PAGE_SHIFT */
 #define THREAD_INFO_ORDER  1
-#endif
 
 #define __HAVE_ARCH_THREAD_INFO_ALLOCATOR
 
diff --git a/arch/sparc/include/asm/timer_32.h b/arch/sparc/include/asm/timer_32.h
index 361e53898dd7..2ec030ef3810 100644
--- a/arch/sparc/include/asm/timer_32.h
+++ b/arch/sparc/include/asm/timer_32.h
@@ -9,96 +9,9 @@
 #define _SPARC_TIMER_H
 
 #include <asm/system.h>  /* For SUN4M_NCPUS */
-#include <asm/sun4paddr.h>
 #include <asm/btfixup.h>
 
-/* Timer structures. The interrupt timer has two properties which
- * are the counter (which is handled in do_timer in sched.c) and the limit.
- * This limit is where the timer's counter 'wraps' around. Oddly enough,
- * the sun4c timer when it hits the limit wraps back to 1 and not zero
- * thus when calculating the value at which it will fire a microsecond you
- * must adjust by one.  Thanks SUN for designing such great hardware ;(
- */
-
-/* Note that I am only going to use the timer that interrupts at
- * Sparc IRQ 10.  There is another one available that can fire at
- * IRQ 14. Currently it is left untouched, we keep the PROM's limit
- * register value and let the prom take these interrupts.  This allows
- * L1-A to work.
- */
-
-struct sun4c_timer_info {
-  __volatile__ unsigned int cur_count10;
-  __volatile__ unsigned int timer_limit10;
-  __volatile__ unsigned int cur_count14;
-  __volatile__ unsigned int timer_limit14;
-};
-
-#define SUN4C_TIMER_PHYSADDR   0xf3000000
-#ifdef CONFIG_SUN4
-#define SUN_TIMER_PHYSADDR SUN4_300_TIMER_PHYSADDR
-#else
-#define SUN_TIMER_PHYSADDR SUN4C_TIMER_PHYSADDR
-#endif
-
-/* A sun4m has two blocks of registers which are probably of the same
- * structure. LSI Logic's L64851 is told to _decrement_ from the limit
- * value. Aurora behaves similarly but its limit value is compacted in
- * other fashion (it's wider). Documented fields are defined here.
- */
-
-/* As with the interrupt register, we have two classes of timer registers
- * which are per-cpu and master.  Per-cpu timers only hit that cpu and are
- * only level 14 ticks, master timer hits all cpus and is level 10.
- */
-
-#define SUN4M_PRM_CNT_L       0x80000000
-#define SUN4M_PRM_CNT_LVALUE  0x7FFFFC00
-
-struct sun4m_timer_percpu_info {
-  __volatile__ unsigned int l14_timer_limit;    /* Initial value is 0x009c4000 */
-  __volatile__ unsigned int l14_cur_count;
-
-  /* This register appears to be write only and/or inaccessible
-   * on Uni-Processor sun4m machines.
-   */
-  __volatile__ unsigned int l14_limit_noclear;  /* Data access error is here */
-
-  __volatile__ unsigned int cntrl;            /* =1 after POST on Aurora */
-  __volatile__ unsigned char space[PAGE_SIZE - 16];
-};
-
-struct sun4m_timer_regs {
-	struct sun4m_timer_percpu_info cpu_timers[SUN4M_NCPUS];
-	volatile unsigned int l10_timer_limit;
-	volatile unsigned int l10_cur_count;
-
-	/* Again, this appears to be write only and/or inaccessible
-	 * on uni-processor sun4m machines.
-	 */
-	volatile unsigned int l10_limit_noclear;
-
-	/* This register too, it must be magic. */
-	volatile unsigned int foobar;
-
-	volatile unsigned int cfg;     /* equals zero at boot time... */
-};
-
-#define SUN4D_PRM_CNT_L       0x80000000
-#define SUN4D_PRM_CNT_LVALUE  0x7FFFFC00
-
-struct sun4d_timer_regs {
-	volatile unsigned int l10_timer_limit;
-	volatile unsigned int l10_cur_countx;
-	volatile unsigned int l10_limit_noclear;
-	volatile unsigned int ctrl;
-	volatile unsigned int l10_cur_count;
-};
-
-extern struct sun4d_timer_regs *sun4d_timers;
-
 extern __volatile__ unsigned int *master_l10_counter;
-extern __volatile__ unsigned int *master_l10_limit;
 
 /* FIXME: Make do_[gs]ettimeofday btfixup calls */
 BTFIXUPDEF_CALL(int, bus_do_settimeofday, struct timespec *tv)
diff --git a/arch/sparc/include/asm/vac-ops.h b/arch/sparc/include/asm/vac-ops.h
index d10527611f11..a63e88ef0426 100644
--- a/arch/sparc/include/asm/vac-ops.h
+++ b/arch/sparc/include/asm/vac-ops.h
@@ -76,11 +76,7 @@
  * cacheable bit in the pte's of all such pages.
  */
 
-#ifdef CONFIG_SUN4
-#define S4CVAC_BADBITS     0x0001e000
-#else
 #define S4CVAC_BADBITS    0x0000f000
-#endif
 
 /* The following is true if vaddr1 and vaddr2 would cause
  * a 'bad alias'.
@@ -94,10 +90,7 @@
  */
 struct sun4c_vac_props {
 	unsigned int num_bytes;     /* Size of the cache */
-	unsigned int num_lines;     /* Number of cache lines */
 	unsigned int do_hwflushes;  /* Hardware flushing available? */
-	enum { VAC_NONE, VAC_WRITE_THROUGH,
-	    VAC_WRITE_BACK } type;  /* What type of VAC? */
 	unsigned int linesize;      /* Size of each line in bytes */
 	unsigned int log2lsize;     /* log2(linesize) */
 	unsigned int on;            /* VAC is enabled */
diff --git a/arch/sparc/include/asm/vfc_ioctls.h b/arch/sparc/include/asm/vfc_ioctls.h
deleted file mode 100644
index af8b69007b22..000000000000
--- a/arch/sparc/include/asm/vfc_ioctls.h
+++ /dev/null
@@ -1,58 +0,0 @@
-/* Copyright (c) 1996 by Manish Vachharajani */
-
-#ifndef _LINUX_VFC_IOCTLS_H_
-#define	_LINUX_VFC_IOCTLS_H_
-
-	/* IOCTLs */
-#define VFC_IOCTL(a)          (('j' << 8) | a)
-#define VFCGCTRL	(VFC_IOCTL (0))	        /* get vfc attributes */
-#define VFCSCTRL	(VFC_IOCTL (1))  	/* set vfc attributes */
-#define VFCGVID		(VFC_IOCTL (2)) 	/* get video decoder attributes */
-#define VFCSVID		(VFC_IOCTL (3))	        /* set video decoder attributes */
-#define VFCHUE		(VFC_IOCTL (4))   	/* set hue */
-#define VFCPORTCHG	(VFC_IOCTL (5))  	/* change port */
-#define VFCRDINFO	(VFC_IOCTL (6))  	/* read info */
-
-	/* Options for setting the vfc attributes and status */
-#define MEMPRST		0x1	/* reset FIFO ptr. */
-#define CAPTRCMD	0x2	/* start capture and wait */
-#define DIAGMODE	0x3	/* diag mode */
-#define NORMMODE	0x4	/* normal mode */
-#define CAPTRSTR	0x5	/* start capture */
-#define CAPTRWAIT	0x6	/* wait for capture to finish */
-
-
-	/* Options for the decoder */
-#define STD_NTSC	0x1	/* NTSC mode */
-#define STD_PAL		0x2	/* PAL mode */
-#define COLOR_ON	0x3	/* force color ON */
-#define MONO		0x4	/* force color OFF */
-
-	/* Values returned by ioctl 2 */
-
-#define NO_LOCK	        1
-#define NTSC_COLOR	2
-#define NTSC_NOCOLOR    3
-#define PAL_COLOR	4
-#define PAL_NOCOLOR	5
-
-/* Not too sure what this does yet */
-	/* Options for setting Field number */
-#define ODD_FIELD	0x1
-#define EVEN_FIELD	0x0
-#define ACTIVE_ONLY     0x2
-#define NON_ACTIVE	0x0
-
-/* Debug options */
-#define VFC_I2C_SEND 0
-#define VFC_I2C_RECV 1
-
-struct vfc_debug_inout
-{
-	unsigned long addr;
-	unsigned long ret;
-	unsigned long len;
-	unsigned char __user *buffer;
-};
-
-#endif /* _LINUX_VFC_IOCTLS_H_ */
diff --git a/arch/sparc/include/asm/visasm.h b/arch/sparc/include/asm/visasm.h
index de797b9bf552..39ca301920db 100644
--- a/arch/sparc/include/asm/visasm.h
+++ b/arch/sparc/include/asm/visasm.h
@@ -57,6 +57,7 @@ static inline void save_and_clear_fpu(void) {
 "		" : : "i" (FPRS_FEF|FPRS_DU) :
 		"o5", "g1", "g2", "g3", "g7", "cc");
 }
+extern int vis_emul(struct pt_regs *, unsigned int);
 #endif
 
 #endif /* _SPARC64_ASI_H */
diff --git a/arch/sparc/kernel/Makefile b/arch/sparc/kernel/Makefile
index 6e03a2a7863c..2d6582095099 100644
--- a/arch/sparc/kernel/Makefile
+++ b/arch/sparc/kernel/Makefile
@@ -13,15 +13,13 @@ obj-y    := entry.o wof.o wuf.o etrap.o rtrap.o traps.o $(IRQ_OBJS) \
 	    time.o windows.o cpu.o devices.o \
 	    tadpole.o tick14.o ptrace.o \
 	    unaligned.o una_asm.o muldiv.o \
-	    prom.o of_device.o devres.o
+	    prom.o of_device.o devres.o dma.o
 
 devres-y = ../../../kernel/irq/devres.o
 
 obj-$(CONFIG_PCI) += pcic.o
-obj-$(CONFIG_SUN4) += sun4setup.o
 obj-$(CONFIG_SMP) += trampoline.o smp.o sun4m_smp.o sun4d_smp.o
 obj-$(CONFIG_SUN_AUXIO) += auxio.o
-obj-$(CONFIG_PCI) += ebus.o
 obj-$(CONFIG_SUN_PM) += apc.o pmc.o
 obj-$(CONFIG_MODULES) += module.o sparc_ksyms.o
 obj-$(CONFIG_SPARC_LED) += led.o
diff --git a/arch/sparc/kernel/apc.c b/arch/sparc/kernel/apc.c
index 5267d48fb2c6..4dd1ba752ce6 100644
--- a/arch/sparc/kernel/apc.c
+++ b/arch/sparc/kernel/apc.c
@@ -12,9 +12,10 @@
 #include <linux/miscdevice.h>
 #include <linux/smp_lock.h>
 #include <linux/pm.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/io.h>
-#include <asm/sbus.h>
 #include <asm/oplib.h>
 #include <asm/uaccess.h>
 #include <asm/auxio.h>
@@ -29,11 +30,10 @@
 #define APC_OBPNAME	"power-management"
 #define APC_DEVNAME "apc"
 
-volatile static u8 __iomem *regs; 
-static int apc_regsize;
+static u8 __iomem *regs;
 static int apc_no_idle __initdata = 0;
 
-#define apc_readb(offs)			(sbus_readb(regs+offs))
+#define apc_readb(offs)		(sbus_readb(regs+offs))
 #define apc_writeb(val, offs) 	(sbus_writeb(val, regs+offs))
 
 /* Specify "apc=noidle" on the kernel command line to 
@@ -69,9 +69,9 @@ static void apc_swift_idle(void)
 #endif
 } 
 
-static inline void apc_free(void)
+static inline void apc_free(struct of_device *op)
 {
-	sbus_iounmap(regs, apc_regsize);
+	of_iounmap(&op->resource[0], regs, resource_size(&op->resource[0]));
 }
 
 static int apc_open(struct inode *inode, struct file *f)
@@ -153,52 +153,56 @@ static const struct file_operations apc_fops = {
 
 static struct miscdevice apc_miscdev = { APC_MINOR, APC_DEVNAME, &apc_fops };
 
-static int __init apc_probe(void)
+static int __devinit apc_probe(struct of_device *op,
+			       const struct of_device_id *match)
 {
-	struct sbus_bus *sbus = NULL;
-	struct sbus_dev *sdev = NULL;
-	int iTmp = 0;
-
-	for_each_sbus(sbus) {
-		for_each_sbusdev(sdev, sbus) {
-			if (!strcmp(sdev->prom_name, APC_OBPNAME)) {
-				goto sbus_done;
-			}
-		}
-	}
+	int err;
 
-sbus_done:
-	if (!sdev) {
-		return -ENODEV;
-	}
-
-	apc_regsize = sdev->reg_addrs[0].reg_size;
-	regs = sbus_ioremap(&sdev->resource[0], 0, 
-				   apc_regsize, APC_OBPNAME);
-	if(!regs) {
+	regs = of_ioremap(&op->resource[0], 0,
+			  resource_size(&op->resource[0]), APC_OBPNAME);
+	if (!regs) {
 		printk(KERN_ERR "%s: unable to map registers\n", APC_DEVNAME);
 		return -ENODEV;
 	}
 
-	iTmp = misc_register(&apc_miscdev);
-	if (iTmp != 0) {
+	err = misc_register(&apc_miscdev);
+	if (err) {
 		printk(KERN_ERR "%s: unable to register device\n", APC_DEVNAME);
-		apc_free();
+		apc_free(op);
 		return -ENODEV;
 	}
 
 	/* Assign power management IDLE handler */
-	if(!apc_no_idle)
+	if (!apc_no_idle)
 		pm_idle = apc_swift_idle;	
 
 	printk(KERN_INFO "%s: power management initialized%s\n", 
-		APC_DEVNAME, apc_no_idle ? " (CPU idle disabled)" : "");
+	       APC_DEVNAME, apc_no_idle ? " (CPU idle disabled)" : "");
+
 	return 0;
 }
 
+static struct of_device_id __initdata apc_match[] = {
+	{
+		.name = APC_OBPNAME,
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, apc_match);
+
+static struct of_platform_driver apc_driver = {
+	.name		= "apc",
+	.match_table	= apc_match,
+	.probe		= apc_probe,
+};
+
+static int __init apc_init(void)
+{
+	return of_register_driver(&apc_driver, &of_bus_type);
+}
+
 /* This driver is not critical to the boot process
  * and is easiest to ioremap when SBus is already
  * initialized, so we install ourselves thusly:
  */
-__initcall(apc_probe);
-
+__initcall(apc_init);
diff --git a/arch/sparc/kernel/auxio.c b/arch/sparc/kernel/auxio.c
index baf4ed3fb0f3..09c857215a52 100644
--- a/arch/sparc/kernel/auxio.c
+++ b/arch/sparc/kernel/auxio.c
@@ -6,6 +6,8 @@
 #include <linux/stddef.h>
 #include <linux/init.h>
 #include <linux/spinlock.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 #include <asm/oplib.h>
 #include <asm/io.h>
 #include <asm/auxio.h>
@@ -59,7 +61,7 @@ void __init auxio_probe(void)
 	r.flags = auxregs[0].which_io & 0xF;
 	r.start = auxregs[0].phys_addr;
 	r.end = auxregs[0].phys_addr + auxregs[0].reg_size - 1;
-	auxio_register = sbus_ioremap(&r, 0, auxregs[0].reg_size, "auxio");
+	auxio_register = of_ioremap(&r, 0, auxregs[0].reg_size, "auxio");
 	/* Fix the address on sun4m and sun4c. */
 	if((((unsigned long) auxregs[0].phys_addr) & 3) == 3 ||
 	   sparc_cpu_model == sun4c)
@@ -128,7 +130,7 @@ void __init auxio_power_probe(void)
 	r.flags = regs.which_io & 0xF;
 	r.start = regs.phys_addr;
 	r.end = regs.phys_addr + regs.reg_size - 1;
-	auxio_power_register = (unsigned char *) sbus_ioremap(&r, 0,
+	auxio_power_register = (unsigned char *) of_ioremap(&r, 0,
 	    regs.reg_size, "auxpower");
 
 	/* Display a quick message on the console. */
diff --git a/arch/sparc/kernel/devices.c b/arch/sparc/kernel/devices.c
index b240b8863fd0..ad656b044b8c 100644
--- a/arch/sparc/kernel/devices.c
+++ b/arch/sparc/kernel/devices.c
@@ -143,7 +143,7 @@ void __init device_scan(void)
 #endif
 	clock_stop_probe();
 
-	if (ARCH_SUN4C_SUN4)
+	if (ARCH_SUN4C)
 		sun4c_probe_memerr_reg();
 
 	return;
diff --git a/arch/sparc/kernel/dma.c b/arch/sparc/kernel/dma.c
new file mode 100644
index 000000000000..ebc8403b035e
--- /dev/null
+++ b/arch/sparc/kernel/dma.c
@@ -0,0 +1,227 @@
+/* dma.c: PCI and SBUS DMA accessors for 32-bit sparc.
+ *
+ * Copyright (C) 2008 David S. Miller <davem@davemloft.net>
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/dma-mapping.h>
+#include <linux/scatterlist.h>
+#include <linux/mm.h>
+
+#ifdef CONFIG_PCI
+#include <linux/pci.h>
+#endif
+
+#include "dma.h"
+
+int dma_supported(struct device *dev, u64 mask)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type)
+		return pci_dma_supported(to_pci_dev(dev), mask);
+#endif
+	return 0;
+}
+EXPORT_SYMBOL(dma_supported);
+
+int dma_set_mask(struct device *dev, u64 dma_mask)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type)
+		return pci_set_dma_mask(to_pci_dev(dev), dma_mask);
+#endif
+	return -EOPNOTSUPP;
+}
+EXPORT_SYMBOL(dma_set_mask);
+
+void *dma_alloc_coherent(struct device *dev, size_t size,
+			 dma_addr_t *dma_handle, gfp_t flag)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type)
+		return pci_alloc_consistent(to_pci_dev(dev), size, dma_handle);
+#endif
+	return sbus_alloc_consistent(dev, size, dma_handle);
+}
+EXPORT_SYMBOL(dma_alloc_coherent);
+
+void dma_free_coherent(struct device *dev, size_t size,
+		       void *cpu_addr, dma_addr_t dma_handle)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type) {
+		pci_free_consistent(to_pci_dev(dev), size,
+				    cpu_addr, dma_handle);
+		return;
+	}
+#endif
+	sbus_free_consistent(dev, size, cpu_addr, dma_handle);
+}
+EXPORT_SYMBOL(dma_free_coherent);
+
+dma_addr_t dma_map_single(struct device *dev, void *cpu_addr,
+			  size_t size, enum dma_data_direction direction)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type)
+		return pci_map_single(to_pci_dev(dev), cpu_addr,
+				      size, (int)direction);
+#endif
+	return sbus_map_single(dev, cpu_addr, size, (int)direction);
+}
+EXPORT_SYMBOL(dma_map_single);
+
+void dma_unmap_single(struct device *dev, dma_addr_t dma_addr,
+		      size_t size,
+		      enum dma_data_direction direction)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type) {
+		pci_unmap_single(to_pci_dev(dev), dma_addr,
+				 size, (int)direction);
+		return;
+	}
+#endif
+	sbus_unmap_single(dev, dma_addr, size, (int)direction);
+}
+EXPORT_SYMBOL(dma_unmap_single);
+
+dma_addr_t dma_map_page(struct device *dev, struct page *page,
+			unsigned long offset, size_t size,
+			enum dma_data_direction direction)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type)
+		return pci_map_page(to_pci_dev(dev), page, offset,
+				    size, (int)direction);
+#endif
+	return sbus_map_single(dev, page_address(page) + offset,
+			       size, (int)direction);
+}
+EXPORT_SYMBOL(dma_map_page);
+
+void dma_unmap_page(struct device *dev, dma_addr_t dma_address,
+		    size_t size, enum dma_data_direction direction)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type) {
+		pci_unmap_page(to_pci_dev(dev), dma_address,
+			       size, (int)direction);
+		return;
+	}
+#endif
+	sbus_unmap_single(dev, dma_address, size, (int)direction);
+}
+EXPORT_SYMBOL(dma_unmap_page);
+
+int dma_map_sg(struct device *dev, struct scatterlist *sg,
+			     int nents, enum dma_data_direction direction)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type)
+		return pci_map_sg(to_pci_dev(dev), sg, nents, (int)direction);
+#endif
+	return sbus_map_sg(dev, sg, nents, direction);
+}
+EXPORT_SYMBOL(dma_map_sg);
+
+void dma_unmap_sg(struct device *dev, struct scatterlist *sg,
+		  int nents, enum dma_data_direction direction)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type) {
+		pci_unmap_sg(to_pci_dev(dev), sg, nents, (int)direction);
+		return;
+	}
+#endif
+	sbus_unmap_sg(dev, sg, nents, (int)direction);
+}
+EXPORT_SYMBOL(dma_unmap_sg);
+
+void dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
+			     size_t size, enum dma_data_direction direction)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type) {
+		pci_dma_sync_single_for_cpu(to_pci_dev(dev), dma_handle,
+					    size, (int)direction);
+		return;
+	}
+#endif
+	sbus_dma_sync_single_for_cpu(dev, dma_handle, size, (int) direction);
+}
+EXPORT_SYMBOL(dma_sync_single_for_cpu);
+
+void dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle,
+				size_t size, enum dma_data_direction direction)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type) {
+		pci_dma_sync_single_for_device(to_pci_dev(dev), dma_handle,
+					       size, (int)direction);
+		return;
+	}
+#endif
+	sbus_dma_sync_single_for_device(dev, dma_handle, size, (int) direction);
+}
+EXPORT_SYMBOL(dma_sync_single_for_device);
+
+void dma_sync_single_range_for_cpu(struct device *dev,
+				   dma_addr_t dma_handle,
+				   unsigned long offset,
+				   size_t size,
+				   enum dma_data_direction direction)
+{
+	dma_sync_single_for_cpu(dev, dma_handle+offset, size, direction);
+}
+EXPORT_SYMBOL(dma_sync_single_range_for_cpu);
+
+void dma_sync_single_range_for_device(struct device *dev, dma_addr_t dma_handle,
+				      unsigned long offset, size_t size,
+				      enum dma_data_direction direction)
+{
+	dma_sync_single_for_device(dev, dma_handle+offset, size, direction);
+}
+EXPORT_SYMBOL(dma_sync_single_range_for_device);
+
+void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
+			 int nelems, enum dma_data_direction direction)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type) {
+		pci_dma_sync_sg_for_cpu(to_pci_dev(dev), sg,
+					nelems, (int)direction);
+		return;
+	}
+#endif
+	BUG();
+}
+EXPORT_SYMBOL(dma_sync_sg_for_cpu);
+
+void dma_sync_sg_for_device(struct device *dev,
+			    struct scatterlist *sg, int nelems,
+			    enum dma_data_direction direction)
+{
+#ifdef CONFIG_PCI
+	if (dev->bus == &pci_bus_type) {
+		pci_dma_sync_sg_for_device(to_pci_dev(dev), sg,
+					   nelems, (int)direction);
+		return;
+	}
+#endif
+	BUG();
+}
+EXPORT_SYMBOL(dma_sync_sg_for_device);
+
+int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
+{
+	return (dma_addr == DMA_ERROR_CODE);
+}
+EXPORT_SYMBOL(dma_mapping_error);
+
+int dma_get_cache_alignment(void)
+{
+	return 32;
+}
+EXPORT_SYMBOL(dma_get_cache_alignment);
diff --git a/arch/sparc/kernel/dma.h b/arch/sparc/kernel/dma.h
new file mode 100644
index 000000000000..f8d8951adb53
--- /dev/null
+++ b/arch/sparc/kernel/dma.h
@@ -0,0 +1,14 @@
+void *sbus_alloc_consistent(struct device *dev, long len, u32 *dma_addrp);
+void sbus_free_consistent(struct device *dev, long n, void *p, u32 ba);
+dma_addr_t sbus_map_single(struct device *dev, void *va,
+			   size_t len, int direction);
+void sbus_unmap_single(struct device *dev, dma_addr_t ba,
+		       size_t n, int direction);
+int sbus_map_sg(struct device *dev, struct scatterlist *sg,
+		int n, int direction);
+void sbus_unmap_sg(struct device *dev, struct scatterlist *sg,
+		   int n, int direction);
+void sbus_dma_sync_single_for_cpu(struct device *dev, dma_addr_t ba,
+				  size_t size, int direction);
+void sbus_dma_sync_single_for_device(struct device *dev, dma_addr_t ba,
+				     size_t size, int direction);
diff --git a/arch/sparc/kernel/ebus.c b/arch/sparc/kernel/ebus.c
deleted file mode 100644
index 97294232259c..000000000000
--- a/arch/sparc/kernel/ebus.c
+++ /dev/null
@@ -1,393 +0,0 @@
-/*
- * ebus.c: PCI to EBus bridge device.
- *
- * Copyright (C) 1997  Eddie C. Dost  (ecd@skynet.be)
- *
- * Adopted for sparc by V. Roganov and G. Raiko.
- * Fixes for different platforms by Pete Zaitcev.
- */
-
-#include <linux/kernel.h>
-#include <linux/types.h>
-#include <linux/init.h>
-#include <linux/slab.h>
-#include <linux/string.h>
-
-#include <asm/system.h>
-#include <asm/page.h>
-#include <asm/pbm.h>
-#include <asm/ebus.h>
-#include <asm/io.h>
-#include <asm/oplib.h>
-#include <asm/prom.h>
-#include <asm/bpp.h>
-
-struct linux_ebus *ebus_chain = NULL;
-
-/* We are together with pcic.c under CONFIG_PCI. */
-extern unsigned int pcic_pin_to_irq(unsigned int, const char *name);
-
-/*
- * IRQ Blacklist
- * Here we list PROMs and systems that are known to supply crap as IRQ numbers.
- */
-struct ebus_device_irq {
-	char *name;
-	unsigned int pin;
-};
-
-struct ebus_system_entry {
-	char *esname;
-	struct ebus_device_irq *ipt;
-};
-
-static struct ebus_device_irq je1_1[] = {
-	{ "8042",		 3 },
-	{ "SUNW,CS4231",	 0 },
-	{ "parallel",		 0 },
-	{ "se",			 2 },
-	{ NULL, 0 }
-};
-
-/*
- * Gleb's JE1 supplied reasonable pin numbers, but mine did not (OBP 2.32).
- * Blacklist the sucker... Note that Gleb's system will work.
- */
-static struct ebus_system_entry ebus_blacklist[] = {
-	{ "SUNW,JavaEngine1", je1_1 },
-	{ NULL, NULL }
-};
-
-static struct ebus_device_irq *ebus_blackp = NULL;
-
-/*
- */
-static inline unsigned long ebus_alloc(size_t size)
-{
-	return (unsigned long)kmalloc(size, GFP_ATOMIC);
-}
-
-/*
- */
-static int __init ebus_blacklist_irq(const char *name)
-{
-	struct ebus_device_irq *dp;
-
-	if ((dp = ebus_blackp) != NULL) {
-		for (; dp->name != NULL; dp++) {
-			if (strcmp(name, dp->name) == 0) {
-				return pcic_pin_to_irq(dp->pin, name);
-			}
-		}
-	}
-	return 0;
-}
-
-static void __init fill_ebus_child(struct device_node *dp,
-				   struct linux_ebus_child *dev)
-{
-	const int *regs;
-	const int *irqs;
-	int i, len;
-
-	dev->prom_node = dp;
-	regs = of_get_property(dp, "reg", &len);
-	if (!regs)
-		len = 0;
-	dev->num_addrs = len / sizeof(regs[0]);
-
-	for (i = 0; i < dev->num_addrs; i++) {
-		if (regs[i] >= dev->parent->num_addrs) {
-			prom_printf("UGH: property for %s was %d, need < %d\n",
-				    dev->prom_node->name, len,
-				    dev->parent->num_addrs);
-			panic(__func__);
-		}
-
-		/* XXX resource */
-		dev->resource[i].start =
-			dev->parent->resource[regs[i]].start;
-	}
-
-	for (i = 0; i < PROMINTR_MAX; i++)
-		dev->irqs[i] = PCI_IRQ_NONE;
-
-	if ((dev->irqs[0] = ebus_blacklist_irq(dev->prom_node->name)) != 0) {
-		dev->num_irqs = 1;
-	} else {
-		irqs = of_get_property(dp, "interrupts", &len);
-		if (!irqs) {
-			dev->num_irqs = 0;
-			dev->irqs[0] = 0;
-			if (dev->parent->num_irqs != 0) {
-				dev->num_irqs = 1;
-				dev->irqs[0] = dev->parent->irqs[0];
-			}
-		} else {
-			dev->num_irqs = len / sizeof(irqs[0]);
-			if (irqs[0] == 0 || irqs[0] >= 8) {
-				/*
-				 * XXX Zero is a valid pin number...
-				 * This works as long as Ebus is not wired
-				 * to INTA#.
-				 */
-				printk("EBUS: %s got bad irq %d from PROM\n",
-				       dev->prom_node->name, irqs[0]);
-				dev->num_irqs = 0;
-				dev->irqs[0] = 0;
-			} else {
-				dev->irqs[0] =
-					pcic_pin_to_irq(irqs[0],
-							dev->prom_node->name);
-			}
-		}
-	}
-}
-
-static void __init fill_ebus_device(struct device_node *dp,
-				    struct linux_ebus_device *dev)
-{
-	const struct linux_prom_registers *regs;
-	struct linux_ebus_child *child;
-	struct dev_archdata *sd;
-	const int *irqs;
-	int i, n, len;
-	unsigned long baseaddr;
-
-	dev->prom_node = dp;
-
-	regs = of_get_property(dp, "reg", &len);
-	if (!regs)
-		len = 0;
-	if (len % sizeof(struct linux_prom_registers)) {
-		prom_printf("UGH: proplen for %s was %d, need multiple of %d\n",
-			    dev->prom_node->name, len,
-			    (int)sizeof(struct linux_prom_registers));
-		panic(__func__);
-	}
-	dev->num_addrs = len / sizeof(struct linux_prom_registers);
-
-	for (i = 0; i < dev->num_addrs; i++) {
-		/*
-		 * XXX Collect JE-1 PROM
-		 * 
-		 * Example - JS-E with 3.11:
-		 *  /ebus
-		 *      regs 
-		 *        0x00000000, 0x0, 0x00000000, 0x0, 0x00000000,
-		 *        0x82000010, 0x0, 0xf0000000, 0x0, 0x01000000,
-		 *        0x82000014, 0x0, 0x38800000, 0x0, 0x00800000,
-		 *      ranges
-		 *        0x00, 0x00000000, 0x02000010, 0x0, 0x0, 0x01000000,
-		 *        0x01, 0x01000000, 0x02000014, 0x0, 0x0, 0x00800000,
-		 *  /ebus/8042
-		 *      regs
-		 *        0x00000001, 0x00300060, 0x00000008,
-		 *        0x00000001, 0x00300060, 0x00000008,
-		 */
-		n = regs[i].which_io;
-		if (n >= 4) {
-			/* XXX This is copied from old JE-1 by Gleb. */
-			n = (regs[i].which_io - 0x10) >> 2;
-		} else {
-			;
-		}
-
-/*
- * XXX Now as we have regions, why don't we make an on-demand allocation...
- */
-		dev->resource[i].start = 0;
-		if ((baseaddr = dev->bus->self->resource[n].start +
-		    regs[i].phys_addr) != 0) {
-			/* dev->resource[i].name = dev->prom_name; */
-			if ((baseaddr = (unsigned long) ioremap(baseaddr,
-			    regs[i].reg_size)) == 0) {
-				panic("ebus: unable to remap dev %s",
-				      dev->prom_node->name);
-			}
-		}
-		dev->resource[i].start = baseaddr;	/* XXX Unaligned */
-	}
-
-	for (i = 0; i < PROMINTR_MAX; i++)
-		dev->irqs[i] = PCI_IRQ_NONE;
-
-	if ((dev->irqs[0] = ebus_blacklist_irq(dev->prom_node->name)) != 0) {
-		dev->num_irqs = 1;
-	} else {
-		irqs = of_get_property(dp, "interrupts", &len);
-		if (!irqs) {
-			dev->num_irqs = 0;
-			if ((dev->irqs[0] = dev->bus->self->irq) != 0) {
-				dev->num_irqs = 1;
-/* P3 */ /* printk("EBUS: child %s irq %d from parent\n", dev->prom_name, dev->irqs[0]); */
-			}
-		} else {
-			dev->num_irqs = 1;  /* dev->num_irqs = len / sizeof(irqs[0]); */
-			if (irqs[0] == 0 || irqs[0] >= 8) {
-				/* See above for the parent. XXX */
-				printk("EBUS: %s got bad irq %d from PROM\n",
-				       dev->prom_node->name, irqs[0]);
-				dev->num_irqs = 0;
-				dev->irqs[0] = 0;
-			} else {
-				dev->irqs[0] =
-					pcic_pin_to_irq(irqs[0],
-							dev->prom_node->name);
-			}
-		}
-	}
-
-	sd = &dev->ofdev.dev.archdata;
-	sd->prom_node = dp;
-	sd->op = &dev->ofdev;
-	sd->iommu = dev->bus->ofdev.dev.parent->archdata.iommu;
-
-	dev->ofdev.node = dp;
-	dev->ofdev.dev.parent = &dev->bus->ofdev.dev;
-	dev->ofdev.dev.bus = &ebus_bus_type;
-	sprintf(dev->ofdev.dev.bus_id, "ebus[%08x]", dp->node);
-
-	/* Register with core */
-	if (of_device_register(&dev->ofdev) != 0)
-		printk(KERN_DEBUG "ebus: device registration error for %s!\n",
-		       dp->path_component_name);
-
-	if ((dp = dp->child) != NULL) {
-		dev->children = (struct linux_ebus_child *)
-			ebus_alloc(sizeof(struct linux_ebus_child));
-
-		child = dev->children;
-		child->next = NULL;
-		child->parent = dev;
-		child->bus = dev->bus;
-		fill_ebus_child(dp, child);
-
-		while ((dp = dp->sibling) != NULL) {
-			child->next = (struct linux_ebus_child *)
-				ebus_alloc(sizeof(struct linux_ebus_child));
-
-			child = child->next;
-			child->next = NULL;
-			child->parent = dev;
-			child->bus = dev->bus;
-			fill_ebus_child(dp, child);
-		}
-	}
-}
-
-void __init ebus_init(void)
-{
-	const struct linux_prom_pci_registers *regs;
-	struct linux_pbm_info *pbm;
-	struct linux_ebus_device *dev;
-	struct linux_ebus *ebus;
-	struct ebus_system_entry *sp;
-	struct pci_dev *pdev;
-	struct pcidev_cookie *cookie;
-	struct device_node *dp;
-	struct resource *p;
-	unsigned short pci_command;
-	int len, reg, nreg;
-	int num_ebus = 0;
-
-	dp = of_find_node_by_path("/");
-	for (sp = ebus_blacklist; sp->esname != NULL; sp++) {
-		if (strcmp(dp->name, sp->esname) == 0) {
-			ebus_blackp = sp->ipt;
-			break;
-		}
-	}
-
-	pdev = pci_get_device(PCI_VENDOR_ID_SUN, PCI_DEVICE_ID_SUN_EBUS, NULL);
-	if (!pdev)
-		return;
-
-	cookie = pdev->sysdata;
-	dp = cookie->prom_node;
-
-	ebus_chain = ebus = (struct linux_ebus *)
-			ebus_alloc(sizeof(struct linux_ebus));
-	ebus->next = NULL;
-
-	while (dp) {
-		struct device_node *nd;
-
-		ebus->prom_node = dp;
-		ebus->self = pdev;
-		ebus->parent = pbm = cookie->pbm;
-
-		/* Enable BUS Master. */
-		pci_read_config_word(pdev, PCI_COMMAND, &pci_command);
-		pci_command |= PCI_COMMAND_MASTER;
-		pci_write_config_word(pdev, PCI_COMMAND, pci_command);
-
-		regs = of_get_property(dp, "reg", &len);
-		if (!regs) {
-			prom_printf("%s: can't find reg property\n",
-				    __func__);
-			prom_halt();
-		}
-		nreg = len / sizeof(struct linux_prom_pci_registers);
-
-		p = &ebus->self->resource[0];
-		for (reg = 0; reg < nreg; reg++) {
-			if (!(regs[reg].which_io & 0x03000000))
-				continue;
-
-			(p++)->start = regs[reg].phys_lo;
-		}
-
-		ebus->ofdev.node = dp;
-		ebus->ofdev.dev.parent = &pdev->dev;
-		ebus->ofdev.dev.bus = &ebus_bus_type;
-		sprintf(ebus->ofdev.dev.bus_id, "ebus%d", num_ebus);
-
-		/* Register with core */
-		if (of_device_register(&ebus->ofdev) != 0)
-			printk(KERN_DEBUG "ebus: device registration error for %s!\n",
-			       dp->path_component_name);
-
-
-		nd = dp->child;
-		if (!nd)
-			goto next_ebus;
-
-		ebus->devices = (struct linux_ebus_device *)
-				ebus_alloc(sizeof(struct linux_ebus_device));
-
-		dev = ebus->devices;
-		dev->next = NULL;
-		dev->children = NULL;
-		dev->bus = ebus;
-		fill_ebus_device(nd, dev);
-
-		while ((nd = nd->sibling) != NULL) {
-			dev->next = (struct linux_ebus_device *)
-				ebus_alloc(sizeof(struct linux_ebus_device));
-
-			dev = dev->next;
-			dev->next = NULL;
-			dev->children = NULL;
-			dev->bus = ebus;
-			fill_ebus_device(nd, dev);
-		}
-
-	next_ebus:
-		pdev = pci_get_device(PCI_VENDOR_ID_SUN,
-				       PCI_DEVICE_ID_SUN_EBUS, pdev);
-		if (!pdev)
-			break;
-
-		cookie = pdev->sysdata;
-		dp = cookie->prom_node;
-
-		ebus->next = (struct linux_ebus *)
-			ebus_alloc(sizeof(struct linux_ebus));
-		ebus = ebus->next;
-		ebus->next = NULL;
-		++num_ebus;
-	}
-	if (pdev)
-		pci_dev_put(pdev);
-}
diff --git a/arch/sparc/kernel/entry.S b/arch/sparc/kernel/entry.S
index e8cdf715a546..faf9ccd9ef5d 100644
--- a/arch/sparc/kernel/entry.S
+++ b/arch/sparc/kernel/entry.S
@@ -20,11 +20,7 @@
 #include <asm/memreg.h>
 #include <asm/page.h>
 #include <asm/pgtable.h>
-#ifdef CONFIG_SUN4
-#include <asm/pgtsun4.h>
-#else
 #include <asm/pgtsun4c.h>
-#endif
 #include <asm/winmacro.h>
 #include <asm/signal.h>
 #include <asm/obio.h>
@@ -276,17 +272,18 @@ smp4m_ticker:
 	 */
 maybe_smp4m_msg:
 	GET_PROCESSOR4M_ID(o3)
-	set	sun4m_interrupts, %l5
-	ld	[%l5], %o5
+	sethi	%hi(sun4m_irq_percpu), %l5
+	sll	%o3, 2, %o3
+	or	%l5, %lo(sun4m_irq_percpu), %o5
 	sethi	%hi(0x40000000), %o2
-	sll	%o3, 12, %o3
 	ld	[%o5 + %o3], %o1
-	andcc	%o1, %o2, %g0
+	ld	[%o1 + 0x00], %o3	! sun4m_irq_percpu[cpu]->pending
+	andcc	%o3, %o2, %g0
 	be,a	smp4m_ticker
 	 cmp	%l7, 14
-	st	%o2, [%o5 + 0x4]
+	st	%o2, [%o1 + 0x04]	! sun4m_irq_percpu[cpu]->clear=0x40000000
 	WRITE_PAUSE
-	ld	[%o5], %g0
+	ld	[%o1 + 0x00], %g0	! sun4m_irq_percpu[cpu]->pending
 	WRITE_PAUSE
 	or	%l0, PSR_PIL, %l4
 	wr	%l4, 0x0, %psr
@@ -304,16 +301,16 @@ linux_trap_ipi15_sun4m:
 	SAVE_ALL
 	sethi	%hi(0x80000000), %o2
 	GET_PROCESSOR4M_ID(o0)
-	set	sun4m_interrupts, %l5
-	ld	[%l5], %o5
-	sll	%o0, 12, %o0
-	add	%o5, %o0, %o5
-	ld	[%o5], %o3
+	sethi	%hi(sun4m_irq_percpu), %l5
+	or	%l5, %lo(sun4m_irq_percpu), %o5
+	sll	%o0, 2, %o0
+	ld	[%o5 + %o0], %o5
+	ld	[%o5 + 0x00], %o3	! sun4m_irq_percpu[cpu]->pending
 	andcc	%o3, %o2, %g0
 	be	1f			! Must be an NMI async memory error
-	 st	%o2, [%o5 + 4]
+	 st	%o2, [%o5 + 0x04]	! sun4m_irq_percpu[cpu]->clear=0x80000000
 	WRITE_PAUSE
-	ld	[%o5], %g0
+	ld	[%o5 + 0x00], %g0	! sun4m_irq_percpu[cpu]->pending
 	WRITE_PAUSE
 	or	%l0, PSR_PIL, %l4
 	wr	%l4, 0x0, %psr
@@ -327,12 +324,11 @@ linux_trap_ipi15_sun4m:
 1:
 	/* NMI async memory error handling. */
 	sethi	%hi(0x80000000), %l4
-	sethi	%hi(0x4000), %o3
-	sub	%o5, %o0, %o5
-	add	%o5, %o3, %l5
-	st	%l4, [%l5 + 0xc]
+	sethi	%hi(sun4m_irq_global), %o5
+	ld	[%o5 + %lo(sun4m_irq_global)], %l5
+	st	%l4, [%l5 + 0x0c]	! sun4m_irq_global->mask_set=0x80000000
 	WRITE_PAUSE
-	ld	[%l5], %g0
+	ld	[%l5 + 0x00], %g0	! sun4m_irq_global->pending
 	WRITE_PAUSE
 	or	%l0, PSR_PIL, %l4
 	wr	%l4, 0x0, %psr
@@ -341,9 +337,9 @@ linux_trap_ipi15_sun4m:
 	WRITE_PAUSE
 	call	sun4m_nmi
 	 nop
-	st	%l4, [%l5 + 0x8]
+	st	%l4, [%l5 + 0x08]	! sun4m_irq_global->mask_clear=0x80000000
 	WRITE_PAUSE
-	ld	[%l5], %g0
+	ld	[%l5 + 0x00], %g0	! sun4m_irq_global->pending
 	WRITE_PAUSE
 	RESTORE_ALL
 
@@ -775,11 +771,7 @@ vac_linesize_patch_32:		subcc	%l7, 32, %l7
  * Ugly, but we cant use hardware flushing on the sun4 and we'd require
  * two instructions (Anton)
  */
-#ifdef CONFIG_SUN4
-vac_hwflush_patch1_on:		nop
-#else
 vac_hwflush_patch1_on:		addcc	%l7, -PAGE_SIZE, %l7
-#endif
 
 vac_hwflush_patch2_on:		sta	%g0, [%l3 + %l7] ASI_HWFLUSHSEG
 
@@ -798,42 +790,10 @@ vac_hwflush_patch2_on:		sta	%g0, [%l3 + %l7] ASI_HWFLUSHSEG
 ! %l7 = 1 for textfault
 ! We want error in %l5, vaddr in %l6
 sun4c_fault:
-#ifdef CONFIG_SUN4
-	sethi	%hi(sun4c_memerr_reg), %l4
-	ld	[%l4+%lo(sun4c_memerr_reg)], %l4  ! memerr ctrl reg addr
-	ld	[%l4], %l6		! memerr ctrl reg
-	ld	[%l4 + 4], %l5		! memerr vaddr reg
-	andcc	%l6, 0x80, %g0		! check for error type
-	st	%g0, [%l4 + 4]		! clear the error
-	be	0f			! normal error
-	 sethi	%hi(AC_BUS_ERROR), %l4	! bus err reg addr
-
-	call	prom_halt	! something weird happened
-					! what exactly did happen?
-					! what should we do here?
-
-0:	or	%l4, %lo(AC_BUS_ERROR), %l4	! bus err reg addr
-	lduba	[%l4] ASI_CONTROL, %l6	! bus err reg
-
-	cmp    %l7, 1			! text fault?
-	be	1f			! yes
-	 nop
-
-	ld     [%l1], %l4		! load instruction that caused fault
-	srl	%l4, 21, %l4
-	andcc	%l4, 1, %g0		! store instruction?
-
-	be	1f			! no
-	 sethi	%hi(SUN4C_SYNC_BADWRITE), %l4 ! yep
-					! %lo(SUN4C_SYNC_BADWRITE) = 0
-	or	%l4, %l6, %l6		! set write bit to emulate sun4c
-1:
-#else
 	sethi	%hi(AC_SYNC_ERR), %l4
 	add	%l4, 0x4, %l6			! AC_SYNC_VA in %l6
 	lda	[%l6] ASI_CONTROL, %l5		! Address
 	lda	[%l4] ASI_CONTROL, %l6		! Error, retained for a bit
-#endif
 
 	andn	%l5, 0xfff, %l5			! Encode all info into l7
 	srl	%l6, 14, %l4
@@ -880,12 +840,7 @@ sun4c_fault:
 	or      %l4, %lo(swapper_pg_dir), %l4
 	sll     %l6, 2, %l6
 	ld      [%l4 + %l6], %l4
-#ifdef CONFIG_SUN4
-	sethi	%hi(PAGE_MASK), %l6
-	andcc	%l4, %l6, %g0
-#else
 	andcc   %l4, PAGE_MASK, %g0
-#endif
 	be      sun4c_fault_fromuser
 	 lduXa  [%l5] ASI_SEGMAP, %l4
 
@@ -937,11 +892,7 @@ invalid_segment_patch1:
 	ld	[%l6 + 0x08], %l3	! tmp = entry->vaddr
 
 	! Flush segment from the cache.
-#ifdef CONFIG_SUN4
-	sethi	%hi((128 * 1024)), %l7
-#else
 	sethi	%hi((64 * 1024)), %l7
-#endif
 9:
 vac_hwflush_patch1:
 vac_linesize_patch:
@@ -1029,12 +980,7 @@ invalid_segment_patch2:
 	or	%l4, %lo(swapper_pg_dir), %l4
 	sll	%l3, 2, %l3
 	ld	[%l4 + %l3], %l4
-#ifndef CONFIG_SUN4
 	and	%l4, PAGE_MASK, %l4
-#else
-	sethi	%hi(PAGE_MASK), %l6
-	and	%l4, %l6, %l4
-#endif
 
 	srl	%l5, (PAGE_SHIFT - 2), %l6
 	and	%l6, ((SUN4C_PTRS_PER_PTE - 1) << 2), %l6
diff --git a/arch/sparc/kernel/head.S b/arch/sparc/kernel/head.S
index 50d9a16af795..2d325fd84579 100644
--- a/arch/sparc/kernel/head.S
+++ b/arch/sparc/kernel/head.S
@@ -63,15 +63,9 @@ cputypvar_sun4m:
 
 	.align 4
 
-#ifndef CONFIG_SUN4
 sun4_notsup:
-	.asciz	"Sparc-Linux sun4 needs a specially compiled kernel, turn CONFIG_SUN4 on.\n\n"
+	.asciz	"Sparc-Linux sun4 support does no longer exist.\n\n"
 	.align 4
-#else
-sun4cdm_notsup:
-	.asciz	"Kernel compiled with CONFIG_SUN4 cannot run on SUN4C/SUN4M/SUN4D\nTurn CONFIG_SUN4 off.\n\n"
-	.align 4
-#endif
 
 sun4e_notsup:
         .asciz  "Sparc-Linux sun4e support does not exist\n\n"
@@ -780,15 +774,6 @@ execute_in_high_mem:
 		 nop
 
 found_version:
-#ifdef CONFIG_SUN4
-/* For people who try sun4 kernels, even if Configure.help advises them. */
-		ld	[%g7 + 0x68], %o1
-		set	sun4cdm_notsup, %o0
-		call	%o1
-		 nop
-		b	halt_me
-		 nop
-#endif
 /* Get the machine type via the mysterious romvec node operations. */
 
 		add	%g7, 0x1c, %l1		
@@ -1150,15 +1135,6 @@ sun4c_continue_boot:
 		 nop
 
 sun4_init:
-#ifdef CONFIG_SUN4
-/* There, happy now Adrian? */
-		set	cputypval, %o2		! Let everyone know we
-		set	' ', %o0			! are a "sun4 " architecture
-		stb	%o0, [%o2 + 0x4]		
-
-		b got_prop 
-		 nop
-#else
 		sethi   %hi(SUN4_PROM_VECTOR+0x84), %o1
 		ld      [%o1 + %lo(SUN4_PROM_VECTOR+0x84)], %o1
 		set     sun4_notsup, %o0
@@ -1170,7 +1146,7 @@ sun4_init:
 		 nop
 1:		ba      1b                      ! Cannot exit into KMON
 		 nop
-#endif
+
 no_sun4e_here:
 		ld	[%g7 + 0x68], %o1
 		set	sun4e_notsup, %o0
diff --git a/arch/sparc/kernel/idprom.c b/arch/sparc/kernel/idprom.c
index fc511f3c4c18..223a6582e1e2 100644
--- a/arch/sparc/kernel/idprom.c
+++ b/arch/sparc/kernel/idprom.c
@@ -12,10 +12,6 @@
 #include <asm/oplib.h>
 #include <asm/idprom.h>
 #include <asm/machines.h>  /* Fun with Sun released architectures. */
-#ifdef CONFIG_SUN4
-#include <asm/sun4paddr.h>
-extern void sun4setup(void);
-#endif
 
 struct idprom *idprom;
 static struct idprom idprom_buffer;
@@ -101,7 +97,4 @@ void __init idprom_init(void)
 		    idprom->id_ethaddr[0], idprom->id_ethaddr[1],
 		    idprom->id_ethaddr[2], idprom->id_ethaddr[3],
 		    idprom->id_ethaddr[4], idprom->id_ethaddr[5]);
-#ifdef CONFIG_SUN4
-	sun4setup();
-#endif
 }
diff --git a/arch/sparc/kernel/ioport.c b/arch/sparc/kernel/ioport.c
index 2a8a847764d8..4f025b36934b 100644
--- a/arch/sparc/kernel/ioport.c
+++ b/arch/sparc/kernel/ioport.c
@@ -42,10 +42,13 @@
 #include <asm/vaddrs.h>
 #include <asm/oplib.h>
 #include <asm/prom.h>
-#include <asm/sbus.h>
 #include <asm/page.h>
 #include <asm/pgalloc.h>
 #include <asm/dma.h>
+#include <asm/iommu.h>
+#include <asm/io-unit.h>
+
+#include "dma.h"
 
 #define mmu_inval_dma_area(p, l)	/* Anton pulled it out for 2.4.0-xx */
 
@@ -139,15 +142,6 @@ void iounmap(volatile void __iomem *virtual)
 	}
 }
 
-/*
- */
-void __iomem *sbus_ioremap(struct resource *phyres, unsigned long offset,
-    unsigned long size, char *name)
-{
-	return _sparc_alloc_io(phyres->flags & 0xF,
-	    phyres->start + offset, size, name);
-}
-
 void __iomem *of_ioremap(struct resource *res, unsigned long offset,
 			 unsigned long size, char *name)
 {
@@ -164,13 +158,6 @@ void of_iounmap(struct resource *res, void __iomem *base, unsigned long size)
 EXPORT_SYMBOL(of_iounmap);
 
 /*
- */
-void sbus_iounmap(volatile void __iomem *addr, unsigned long size)
-{
-	iounmap(addr);
-}
-
-/*
  * Meat of mapping
  */
 static void __iomem *_sparc_alloc_io(unsigned int busno, unsigned long phys,
@@ -246,63 +233,19 @@ static void _sparc_free_io(struct resource *res)
 
 #ifdef CONFIG_SBUS
 
-void sbus_set_sbus64(struct sbus_dev *sdev, int x)
+void sbus_set_sbus64(struct device *dev, int x)
 {
 	printk("sbus_set_sbus64: unsupported\n");
 }
 
-extern unsigned int sun4d_build_irq(struct sbus_dev *sdev, int irq);
-void __init sbus_fill_device_irq(struct sbus_dev *sdev)
-{
-	struct linux_prom_irqs irqs[PROMINTR_MAX];
-	int len;
-
-	len = prom_getproperty(sdev->prom_node, "intr",
-			       (char *)irqs, sizeof(irqs));
-	if (len != -1) {
-		sdev->num_irqs = len / 8;
-		if (sdev->num_irqs == 0) {
-			sdev->irqs[0] = 0;
-		} else if (sparc_cpu_model == sun4d) {
-			for (len = 0; len < sdev->num_irqs; len++)
-				sdev->irqs[len] =
-					sun4d_build_irq(sdev, irqs[len].pri);
-		} else {
-			for (len = 0; len < sdev->num_irqs; len++)
-				sdev->irqs[len] = irqs[len].pri;
-		}
-	} else {
-		int interrupts[PROMINTR_MAX];
-
-		/* No "intr" node found-- check for "interrupts" node.
-		 * This node contains SBus interrupt levels, not IPLs
-		 * as in "intr", and no vector values.  We convert
-		 * SBus interrupt levels to PILs (platform specific).
-		 */
-		len = prom_getproperty(sdev->prom_node, "interrupts",
-				       (char *)interrupts, sizeof(interrupts));
-		if (len == -1) {
-			sdev->irqs[0] = 0;
-			sdev->num_irqs = 0;
-		} else {
-			sdev->num_irqs = len / sizeof(int);
-			for (len = 0; len < sdev->num_irqs; len++) {
-				sdev->irqs[len] =
-					sbint_to_irq(sdev, interrupts[len]);
-			}
-		}
-	} 
-}
-
 /*
  * Allocate a chunk of memory suitable for DMA.
  * Typically devices use them for control blocks.
  * CPU may access them without any explicit flushing.
- *
- * XXX Some clever people know that sdev is not used and supply NULL. Watch.
  */
-void *sbus_alloc_consistent(struct sbus_dev *sdev, long len, u32 *dma_addrp)
+void *sbus_alloc_consistent(struct device *dev, long len, u32 *dma_addrp)
 {
+	struct of_device *op = to_of_device(dev);
 	unsigned long len_total = (len + PAGE_SIZE-1) & PAGE_MASK;
 	unsigned long va;
 	struct resource *res;
@@ -336,13 +279,10 @@ void *sbus_alloc_consistent(struct sbus_dev *sdev, long len, u32 *dma_addrp)
 	 * XXX That's where sdev would be used. Currently we load
 	 * all iommu tables with the same translations.
 	 */
-	if (mmu_map_dma_area(dma_addrp, va, res->start, len_total) != 0)
+	if (mmu_map_dma_area(dev, dma_addrp, va, res->start, len_total) != 0)
 		goto err_noiommu;
 
-	/* Set the resource name, if known. */
-	if (sdev) {
-		res->name = sdev->prom_name;
-	}
+	res->name = op->node->name;
 
 	return (void *)(unsigned long)res->start;
 
@@ -356,7 +296,7 @@ err_nopages:
 	return NULL;
 }
 
-void sbus_free_consistent(struct sbus_dev *sdev, long n, void *p, u32 ba)
+void sbus_free_consistent(struct device *dev, long n, void *p, u32 ba)
 {
 	struct resource *res;
 	struct page *pgv;
@@ -383,8 +323,8 @@ void sbus_free_consistent(struct sbus_dev *sdev, long n, void *p, u32 ba)
 	kfree(res);
 
 	/* mmu_inval_dma_area(va, n); */ /* it's consistent, isn't it */
-	pgv = mmu_translate_dvma(ba);
-	mmu_unmap_dma_area(ba, n);
+	pgv = virt_to_page(p);
+	mmu_unmap_dma_area(dev, ba, n);
 
 	__free_pages(pgv, get_order(n));
 }
@@ -394,7 +334,7 @@ void sbus_free_consistent(struct sbus_dev *sdev, long n, void *p, u32 ba)
  * CPU view of this memory may be inconsistent with
  * a device view and explicit flushing is necessary.
  */
-dma_addr_t sbus_map_single(struct sbus_dev *sdev, void *va, size_t len, int direction)
+dma_addr_t sbus_map_single(struct device *dev, void *va, size_t len, int direction)
 {
 	/* XXX why are some lengths signed, others unsigned? */
 	if (len <= 0) {
@@ -404,17 +344,17 @@ dma_addr_t sbus_map_single(struct sbus_dev *sdev, void *va, size_t len, int dire
 	if (len > 256*1024) {			/* __get_free_pages() limit */
 		return 0;
 	}
-	return mmu_get_scsi_one(va, len, sdev->bus);
+	return mmu_get_scsi_one(dev, va, len);
 }
 
-void sbus_unmap_single(struct sbus_dev *sdev, dma_addr_t ba, size_t n, int direction)
+void sbus_unmap_single(struct device *dev, dma_addr_t ba, size_t n, int direction)
 {
-	mmu_release_scsi_one(ba, n, sdev->bus);
+	mmu_release_scsi_one(dev, ba, n);
 }
 
-int sbus_map_sg(struct sbus_dev *sdev, struct scatterlist *sg, int n, int direction)
+int sbus_map_sg(struct device *dev, struct scatterlist *sg, int n, int direction)
 {
-	mmu_get_scsi_sgl(sg, n, sdev->bus);
+	mmu_get_scsi_sgl(dev, sg, n);
 
 	/*
 	 * XXX sparc64 can return a partial length here. sun4c should do this
@@ -423,145 +363,28 @@ int sbus_map_sg(struct sbus_dev *sdev, struct scatterlist *sg, int n, int direct
 	return n;
 }
 
-void sbus_unmap_sg(struct sbus_dev *sdev, struct scatterlist *sg, int n, int direction)
-{
-	mmu_release_scsi_sgl(sg, n, sdev->bus);
-}
-
-/*
- */
-void sbus_dma_sync_single_for_cpu(struct sbus_dev *sdev, dma_addr_t ba, size_t size, int direction)
-{
-#if 0
-	unsigned long va;
-	struct resource *res;
-
-	/* We do not need the resource, just print a message if invalid. */
-	res = _sparc_find_resource(&_sparc_dvma, ba);
-	if (res == NULL)
-		panic("sbus_dma_sync_single: 0x%x\n", ba);
-
-	va = page_address(mmu_translate_dvma(ba)); /* XXX higmem */
-	/*
-	 * XXX This bogosity will be fixed with the iommu rewrite coming soon
-	 * to a kernel near you. - Anton
-	 */
-	/* mmu_inval_dma_area(va, (size + PAGE_SIZE-1) & PAGE_MASK); */
-#endif
-}
-
-void sbus_dma_sync_single_for_device(struct sbus_dev *sdev, dma_addr_t ba, size_t size, int direction)
+void sbus_unmap_sg(struct device *dev, struct scatterlist *sg, int n, int direction)
 {
-#if 0
-	unsigned long va;
-	struct resource *res;
-
-	/* We do not need the resource, just print a message if invalid. */
-	res = _sparc_find_resource(&_sparc_dvma, ba);
-	if (res == NULL)
-		panic("sbus_dma_sync_single: 0x%x\n", ba);
-
-	va = page_address(mmu_translate_dvma(ba)); /* XXX higmem */
-	/*
-	 * XXX This bogosity will be fixed with the iommu rewrite coming soon
-	 * to a kernel near you. - Anton
-	 */
-	/* mmu_inval_dma_area(va, (size + PAGE_SIZE-1) & PAGE_MASK); */
-#endif
+	mmu_release_scsi_sgl(dev, sg, n);
 }
 
-void sbus_dma_sync_sg_for_cpu(struct sbus_dev *sdev, struct scatterlist *sg, int n, int direction)
+void sbus_dma_sync_single_for_cpu(struct device *dev, dma_addr_t ba, size_t size, int direction)
 {
-	printk("sbus_dma_sync_sg_for_cpu: not implemented yet\n");
 }
 
-void sbus_dma_sync_sg_for_device(struct sbus_dev *sdev, struct scatterlist *sg, int n, int direction)
+void sbus_dma_sync_single_for_device(struct device *dev, dma_addr_t ba, size_t size, int direction)
 {
-	printk("sbus_dma_sync_sg_for_device: not implemented yet\n");
-}
-
-/* Support code for sbus_init().  */
-/*
- * XXX This functions appears to be a distorted version of
- * prom_sbus_ranges_init(), with all sun4d stuff cut away.
- * Ask DaveM what is going on here, how is sun4d supposed to work... XXX
- */
-/* added back sun4d patch from Thomas Bogendoerfer - should be OK (crn) */
-void __init sbus_arch_bus_ranges_init(struct device_node *pn, struct sbus_bus *sbus)
-{
-	int parent_node = pn->node;
-
-	if (sparc_cpu_model == sun4d) {
-		struct linux_prom_ranges iounit_ranges[PROMREG_MAX];
-		int num_iounit_ranges, len;
-
-		len = prom_getproperty(parent_node, "ranges",
-				       (char *) iounit_ranges,
-				       sizeof (iounit_ranges));
-		if (len != -1) {
-			num_iounit_ranges =
-				(len / sizeof(struct linux_prom_ranges));
-			prom_adjust_ranges(sbus->sbus_ranges,
-					   sbus->num_sbus_ranges,
-					   iounit_ranges, num_iounit_ranges);
-		}
-	}
 }
 
-void __init sbus_setup_iommu(struct sbus_bus *sbus, struct device_node *dp)
-{
-#ifndef CONFIG_SUN4
-	struct device_node *parent = dp->parent;
-
-	if (sparc_cpu_model != sun4d &&
-	    parent != NULL &&
-	    !strcmp(parent->name, "iommu")) {
-		extern void iommu_init(int iommu_node, struct sbus_bus *sbus);
-
-		iommu_init(parent->node, sbus);
-	}
-
-	if (sparc_cpu_model == sun4d) {
-		extern void iounit_init(int sbi_node, int iounit_node,
-					struct sbus_bus *sbus);
-
-		iounit_init(dp->node, parent->node, sbus);
-	}
-#endif
-}
-
-void __init sbus_setup_arch_props(struct sbus_bus *sbus, struct device_node *dp)
-{
-	if (sparc_cpu_model == sun4d) {
-		struct device_node *parent = dp->parent;
-
-		sbus->devid = of_getintprop_default(parent, "device-id", 0);
-		sbus->board = of_getintprop_default(parent, "board#", 0);
-	}
-}
-
-int __init sbus_arch_preinit(void)
+static int __init sparc_register_ioport(void)
 {
 	register_proc_sparc_ioport();
 
-#ifdef CONFIG_SUN4
-	{
-		extern void sun4_dvma_init(void);
-		sun4_dvma_init();
-	}
-	return 1;
-#else
 	return 0;
-#endif
 }
 
-void __init sbus_arch_postinit(void)
-{
-	if (sparc_cpu_model == sun4d) {
-		extern void sun4d_init_sbi_irq(void);
-		sun4d_init_sbi_irq();
-	}
-}
+arch_initcall(sparc_register_ioport);
+
 #endif /* CONFIG_SBUS */
 
 #ifdef CONFIG_PCI
diff --git a/arch/sparc/kernel/irq.h b/arch/sparc/kernel/irq.h
index 32ef3ebd0a88..db7513881530 100644
--- a/arch/sparc/kernel/irq.h
+++ b/arch/sparc/kernel/irq.h
@@ -13,7 +13,6 @@ BTFIXUPDEF_CALL(void, enable_irq, unsigned int)
 BTFIXUPDEF_CALL(void, disable_pil_irq, unsigned int)
 BTFIXUPDEF_CALL(void, enable_pil_irq, unsigned int)
 BTFIXUPDEF_CALL(void, clear_clock_irq, void)
-BTFIXUPDEF_CALL(void, clear_profile_irq, int)
 BTFIXUPDEF_CALL(void, load_profile_irq, int, unsigned int)
 
 static inline void __disable_irq(unsigned int irq)
@@ -41,11 +40,6 @@ static inline void clear_clock_irq(void)
 	BTFIXUP_CALL(clear_clock_irq)();
 }
 
-static inline void clear_profile_irq(int irq)
-{
-	BTFIXUP_CALL(clear_profile_irq)(irq);
-}
-
 static inline void load_profile_irq(int cpu, int limit)
 {
 	BTFIXUP_CALL(load_profile_irq)(cpu, limit);
diff --git a/arch/sparc/kernel/of_device.c b/arch/sparc/kernel/of_device.c
index f58c537446a8..0837bd52e28f 100644
--- a/arch/sparc/kernel/of_device.c
+++ b/arch/sparc/kernel/of_device.c
@@ -29,15 +29,38 @@ struct of_device *of_find_device_by_node(struct device_node *dp)
 }
 EXPORT_SYMBOL(of_find_device_by_node);
 
-#ifdef CONFIG_PCI
-struct bus_type ebus_bus_type;
-EXPORT_SYMBOL(ebus_bus_type);
-#endif
+unsigned int irq_of_parse_and_map(struct device_node *node, int index)
+{
+	struct of_device *op = of_find_device_by_node(node);
+
+	if (!op || index >= op->num_irqs)
+		return 0;
+
+	return op->irqs[index];
+}
+EXPORT_SYMBOL(irq_of_parse_and_map);
+
+/* Take the archdata values for IOMMU, STC, and HOSTDATA found in
+ * BUS and propagate to all child of_device objects.
+ */
+void of_propagate_archdata(struct of_device *bus)
+{
+	struct dev_archdata *bus_sd = &bus->dev.archdata;
+	struct device_node *bus_dp = bus->node;
+	struct device_node *dp;
+
+	for (dp = bus_dp->child; dp; dp = dp->sibling) {
+		struct of_device *op = of_find_device_by_node(dp);
 
-#ifdef CONFIG_SBUS
-struct bus_type sbus_bus_type;
-EXPORT_SYMBOL(sbus_bus_type);
-#endif
+		op->dev.archdata.iommu = bus_sd->iommu;
+		op->dev.archdata.stc = bus_sd->stc;
+		op->dev.archdata.host_controller = bus_sd->host_controller;
+		op->dev.archdata.numa_node = bus_sd->numa_node;
+
+		if (dp->child)
+			of_propagate_archdata(op);
+	}
+}
 
 struct bus_type of_platform_bus_type;
 EXPORT_SYMBOL(of_platform_bus_type);
@@ -327,6 +350,27 @@ static int __init build_one_resource(struct device_node *parent,
 	return 1;
 }
 
+static int __init use_1to1_mapping(struct device_node *pp)
+{
+	/* If we have a ranges property in the parent, use it.  */
+	if (of_find_property(pp, "ranges", NULL) != NULL)
+		return 0;
+
+	/* Some SBUS devices use intermediate nodes to express
+	 * hierarchy within the device itself.  These aren't
+	 * real bus nodes, and don't have a 'ranges' property.
+	 * But, we should still pass the translation work up
+	 * to the SBUS itself.
+	 */
+	if (!strcmp(pp->name, "dma") ||
+	    !strcmp(pp->name, "espdma") ||
+	    !strcmp(pp->name, "ledma") ||
+	    !strcmp(pp->name, "lebuffer"))
+		return 0;
+
+	return 1;
+}
+
 static int of_resource_verbose;
 
 static void __init build_device_resources(struct of_device *op,
@@ -373,10 +417,7 @@ static void __init build_device_resources(struct of_device *op,
 
 		flags = bus->get_flags(reg, 0);
 
-		/* If the immediate parent has no ranges property to apply,
-		 * just use a 1<->1 mapping.
-		 */
-		if (of_find_property(pp, "ranges", NULL) == NULL) {
+		if (use_1to1_mapping(pp)) {
 			result = of_read_addr(addr, na);
 			goto build_res;
 		}
@@ -565,15 +606,6 @@ static int __init of_bus_driver_init(void)
 	int err;
 
 	err = of_bus_type_init(&of_platform_bus_type, "of");
-#ifdef CONFIG_PCI
-	if (!err)
-		err = of_bus_type_init(&ebus_bus_type, "ebus");
-#endif
-#ifdef CONFIG_SBUS
-	if (!err)
-		err = of_bus_type_init(&sbus_bus_type, "sbus");
-#endif
-
 	if (!err)
 		scan_of_devices();
 
diff --git a/arch/sparc/kernel/pcic.c b/arch/sparc/kernel/pcic.c
index a6a6f9823370..462584e55fba 100644
--- a/arch/sparc/kernel/pcic.c
+++ b/arch/sparc/kernel/pcic.c
@@ -17,8 +17,6 @@
 #include <linux/slab.h>
 #include <linux/jiffies.h>
 
-#include <asm/ebus.h>
-#include <asm/sbus.h> /* for sanity check... */
 #include <asm/swift.h> /* for cache flushing. */
 #include <asm/io.h>
 
@@ -430,7 +428,6 @@ static int __init pcic_init(void)
 
 	pcic_pbm_scan_bus(pcic);
 
-	ebus_init();
 	return 0;
 }
 
@@ -493,10 +490,6 @@ static void pcic_map_pci_device(struct linux_pcic *pcic,
 				 * do ioremap() before accessing PC-style I/O,
 				 * we supply virtual, ready to access address.
 				 *
-				 * Ebus devices do not come here even if
-				 * CheerIO makes a similar conversion.
-				 * See ebus.c for details.
-				 *
 				 * Note that request_region()
 				 * works for these devices.
 				 *
@@ -677,7 +670,7 @@ void __devinit pcibios_fixup_bus(struct pci_bus *bus)
 }
 
 /*
- * pcic_pin_to_irq() is exported to ebus.c.
+ * pcic_pin_to_irq() is exported to bus probing code
  */
 unsigned int
 pcic_pin_to_irq(unsigned int pin, const char *name)
@@ -904,11 +897,6 @@ static void pcic_enable_irq(unsigned int irq_nr)
 	local_irq_restore(flags);
 }
 
-static void pcic_clear_profile_irq(int cpu)
-{
-	printk("PCIC: unimplemented code: FILE=%s LINE=%d", __FILE__, __LINE__);
-}
-
 static void pcic_load_profile_irq(int cpu, unsigned int limit)
 {
 	printk("PCIC: unimplemented code: FILE=%s LINE=%d", __FILE__, __LINE__);
@@ -934,7 +922,6 @@ void __init sun4m_pci_init_IRQ(void)
 	BTFIXUPSET_CALL(enable_pil_irq, pcic_enable_pil_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(disable_pil_irq, pcic_disable_pil_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(clear_clock_irq, pcic_clear_clock_irq, BTFIXUPCALL_NORM);
-	BTFIXUPSET_CALL(clear_profile_irq, pcic_clear_profile_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(load_profile_irq, pcic_load_profile_irq, BTFIXUPCALL_NORM);
 }
 
diff --git a/arch/sparc/kernel/pmc.c b/arch/sparc/kernel/pmc.c
index 7eca8871ff47..2afcfab4f11c 100644
--- a/arch/sparc/kernel/pmc.c
+++ b/arch/sparc/kernel/pmc.c
@@ -8,11 +8,11 @@
 #include <linux/fs.h>
 #include <linux/errno.h>
 #include <linux/init.h>
-#include <linux/miscdevice.h>
 #include <linux/pm.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/io.h>
-#include <asm/sbus.h>
 #include <asm/oplib.h>
 #include <asm/uaccess.h>
 #include <asm/auxio.h>
@@ -23,17 +23,15 @@
  * #define PMC_NO_IDLE
  */
 
-#define PMC_MINOR	MISC_DYNAMIC_MINOR
 #define PMC_OBPNAME	"SUNW,pmc"
 #define PMC_DEVNAME "pmc"
 
 #define PMC_IDLE_REG	0x00
 #define PMC_IDLE_ON		0x01
 
-volatile static u8 __iomem *regs; 
-static int pmc_regsize;
+static u8 __iomem *regs;
 
-#define pmc_readb(offs)			(sbus_readb(regs+offs))
+#define pmc_readb(offs)		(sbus_readb(regs+offs))
 #define pmc_writeb(val, offs) 	(sbus_writeb(val, regs+offs))
 
 /* 
@@ -53,31 +51,11 @@ void pmc_swift_idle(void)
 #endif
 } 
 
-static inline void pmc_free(void)
+static int __devinit pmc_probe(struct of_device *op,
+			       const struct of_device_id *match)
 {
-	sbus_iounmap(regs, pmc_regsize);
-}
-
-static int __init pmc_probe(void)
-{
-	struct sbus_bus *sbus = NULL;
-	struct sbus_dev *sdev = NULL;
-	for_each_sbus(sbus) {
-		for_each_sbusdev(sdev, sbus) {
-			if (!strcmp(sdev->prom_name, PMC_OBPNAME)) {
-				goto sbus_done;
-			}
-		}
-	}
-
-sbus_done:
-	if (!sdev) {
-		return -ENODEV;
-	}
-
-	pmc_regsize = sdev->reg_addrs[0].reg_size;
-	regs = sbus_ioremap(&sdev->resource[0], 0, 
-				   pmc_regsize, PMC_OBPNAME);
+	regs = of_ioremap(&op->resource[0], 0,
+			  resource_size(&op->resource[0]), PMC_OBPNAME);
 	if (!regs) {
 		printk(KERN_ERR "%s: unable to map registers\n", PMC_DEVNAME);
 		return -ENODEV;
@@ -92,8 +70,27 @@ sbus_done:
 	return 0;
 }
 
+static struct of_device_id __initdata pmc_match[] = {
+	{
+		.name = PMC_OBPNAME,
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, pmc_match);
+
+static struct of_platform_driver pmc_driver = {
+	.name		= "pmc",
+	.match_table	= pmc_match,
+	.probe		= pmc_probe,
+};
+
+static int __init pmc_init(void)
+{
+	return of_register_driver(&pmc_driver, &of_bus_type);
+}
+
 /* This driver is not critical to the boot process
  * and is easiest to ioremap when SBus is already
  * initialized, so we install ourselves thusly:
  */
-__initcall(pmc_probe);
+__initcall(pmc_init);
diff --git a/arch/sparc/kernel/process.c b/arch/sparc/kernel/process.c
index 4bb430940a61..e8c43ffe317e 100644
--- a/arch/sparc/kernel/process.c
+++ b/arch/sparc/kernel/process.c
@@ -75,7 +75,7 @@ void cpu_idle(void)
 {
 	/* endless idle loop with no priority at all */
 	for (;;) {
-		if (ARCH_SUN4C_SUN4) {
+		if (ARCH_SUN4C) {
 			static int count = HZ;
 			static unsigned long last_jiffies;
 			static unsigned long last_faults;
diff --git a/arch/sparc/kernel/prom.c b/arch/sparc/kernel/prom.c
index cd4fb79aa3a8..eee5efcfe50e 100644
--- a/arch/sparc/kernel/prom.c
+++ b/arch/sparc/kernel/prom.c
@@ -54,6 +54,9 @@ int of_getintprop_default(struct device_node *np, const char *name, int def)
 }
 EXPORT_SYMBOL(of_getintprop_default);
 
+DEFINE_MUTEX(of_set_property_mutex);
+EXPORT_SYMBOL(of_set_property_mutex);
+
 int of_set_property(struct device_node *dp, const char *name, void *val, int len)
 {
 	struct property **prevp;
@@ -77,7 +80,10 @@ int of_set_property(struct device_node *dp, const char *name, void *val, int len
 			void *old_val = prop->value;
 			int ret;
 
+			mutex_lock(&of_set_property_mutex);
 			ret = prom_setprop(dp->node, (char *) name, val, len);
+			mutex_unlock(&of_set_property_mutex);
+
 			err = -EINVAL;
 			if (ret >= 0) {
 				prop->value = new_val;
@@ -436,7 +442,6 @@ static void __init of_console_init(void)
 
 	switch (prom_vers) {
 	case PROM_V0:
-	case PROM_SUN4:
 		skip = 0;
 		switch (*romvec->pv_stdout) {
 		case PROMDEV_SCREEN:
diff --git a/arch/sparc/kernel/setup.c b/arch/sparc/kernel/setup.c
index 9e451b21202e..24fe3078bd4b 100644
--- a/arch/sparc/kernel/setup.c
+++ b/arch/sparc/kernel/setup.c
@@ -213,23 +213,25 @@ void __init setup_arch(char **cmdline_p)
 	/* Initialize PROM console and command line. */
 	*cmdline_p = prom_getbootargs();
 	strcpy(boot_command_line, *cmdline_p);
+	parse_early_param();
 
 	/* Set sparc_cpu_model */
 	sparc_cpu_model = sun_unknown;
-	if(!strcmp(&cputypval,"sun4 ")) { sparc_cpu_model=sun4; }
-	if(!strcmp(&cputypval,"sun4c")) { sparc_cpu_model=sun4c; }
-	if(!strcmp(&cputypval,"sun4m")) { sparc_cpu_model=sun4m; }
-	if(!strcmp(&cputypval,"sun4s")) { sparc_cpu_model=sun4m; }  /* CP-1200 with PROM 2.30 -E */
-	if(!strcmp(&cputypval,"sun4d")) { sparc_cpu_model=sun4d; }
-	if(!strcmp(&cputypval,"sun4e")) { sparc_cpu_model=sun4e; }
-	if(!strcmp(&cputypval,"sun4u")) { sparc_cpu_model=sun4u; }
-
-#ifdef CONFIG_SUN4
-	if (sparc_cpu_model != sun4) {
-		prom_printf("This kernel is for Sun4 architecture only.\n");
-		prom_halt();
-	}
-#endif
+	if (!strcmp(&cputypval,"sun4 "))
+		sparc_cpu_model = sun4;
+	if (!strcmp(&cputypval,"sun4c"))
+		sparc_cpu_model = sun4c;
+	if (!strcmp(&cputypval,"sun4m"))
+		sparc_cpu_model = sun4m;
+	if (!strcmp(&cputypval,"sun4s"))
+		sparc_cpu_model = sun4m; /* CP-1200 with PROM 2.30 -E */
+	if (!strcmp(&cputypval,"sun4d"))
+		sparc_cpu_model = sun4d;
+	if (!strcmp(&cputypval,"sun4e"))
+		sparc_cpu_model = sun4e;
+	if (!strcmp(&cputypval,"sun4u"))
+		sparc_cpu_model = sun4u;
+
 	printk("ARCH: ");
 	switch(sparc_cpu_model) {
 	case sun4:
@@ -263,7 +265,7 @@ void __init setup_arch(char **cmdline_p)
 	boot_flags_init(*cmdline_p);
 
 	idprom_init();
-	if (ARCH_SUN4C_SUN4)
+	if (ARCH_SUN4C)
 		sun4c_probe_vac();
 	load_mmu();
 
diff --git a/arch/sparc/kernel/sparc_ksyms.c b/arch/sparc/kernel/sparc_ksyms.c
index b23cea5ca5d1..b0dfff848653 100644
--- a/arch/sparc/kernel/sparc_ksyms.c
+++ b/arch/sparc/kernel/sparc_ksyms.c
@@ -38,17 +38,12 @@
 #include <asm/idprom.h>
 #include <asm/head.h>
 #include <asm/smp.h>
-#include <asm/mostek.h>
 #include <asm/ptrace.h>
 #include <asm/uaccess.h>
 #include <asm/checksum.h>
 #ifdef CONFIG_SBUS
-#include <asm/sbus.h>
 #include <asm/dma.h>
 #endif
-#ifdef CONFIG_PCI
-#include <asm/ebus.h>
-#endif
 #include <asm/io-unit.h>
 #include <asm/bug.h>
 
@@ -127,16 +122,11 @@ EXPORT_SYMBOL(phys_cpu_present_map);
 EXPORT_SYMBOL(__udelay);
 EXPORT_SYMBOL(__ndelay);
 EXPORT_SYMBOL(rtc_lock);
-EXPORT_SYMBOL(mostek_lock);
-EXPORT_SYMBOL(mstk48t02_regs);
 #ifdef CONFIG_SUN_AUXIO
 EXPORT_SYMBOL(set_auxio);
 EXPORT_SYMBOL(get_auxio);
 #endif
 EXPORT_SYMBOL(io_remap_pfn_range);
-  /* P3: iounit_xxx may be needed, sun4d users */
-/* EXPORT_SYMBOL(iounit_map_dma_init); */
-/* EXPORT_SYMBOL(iounit_map_dma_page); */
 
 #ifndef CONFIG_SMP
 EXPORT_SYMBOL(BTFIXUP_CALL(___xchg32));
@@ -153,24 +143,9 @@ EXPORT_SYMBOL(BTFIXUP_CALL(mmu_release_scsi_one));
 EXPORT_SYMBOL(BTFIXUP_CALL(pgprot_noncached));
 
 #ifdef CONFIG_SBUS
-EXPORT_SYMBOL(sbus_root);
-EXPORT_SYMBOL(dma_chain);
 EXPORT_SYMBOL(sbus_set_sbus64);
-EXPORT_SYMBOL(sbus_alloc_consistent);
-EXPORT_SYMBOL(sbus_free_consistent);
-EXPORT_SYMBOL(sbus_map_single);
-EXPORT_SYMBOL(sbus_unmap_single);
-EXPORT_SYMBOL(sbus_map_sg);
-EXPORT_SYMBOL(sbus_unmap_sg);
-EXPORT_SYMBOL(sbus_dma_sync_single_for_cpu);
-EXPORT_SYMBOL(sbus_dma_sync_single_for_device);
-EXPORT_SYMBOL(sbus_dma_sync_sg_for_cpu);
-EXPORT_SYMBOL(sbus_dma_sync_sg_for_device);
-EXPORT_SYMBOL(sbus_iounmap);
-EXPORT_SYMBOL(sbus_ioremap);
 #endif
 #ifdef CONFIG_PCI
-EXPORT_SYMBOL(ebus_chain);
 EXPORT_SYMBOL(insb);
 EXPORT_SYMBOL(outsb);
 EXPORT_SYMBOL(insw);
diff --git a/arch/sparc/kernel/sun4c_irq.c b/arch/sparc/kernel/sun4c_irq.c
index 340fc395fe2d..5dc8a5769489 100644
--- a/arch/sparc/kernel/sun4c_irq.c
+++ b/arch/sparc/kernel/sun4c_irq.c
@@ -18,6 +18,8 @@
 #include <linux/interrupt.h>
 #include <linux/slab.h>
 #include <linux/init.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 #include "irq.h"
 
 #include <asm/ptrace.h>
@@ -31,15 +33,8 @@
 #include <asm/traps.h>
 #include <asm/irq.h>
 #include <asm/io.h>
-#include <asm/sun4paddr.h>
 #include <asm/idprom.h>
 #include <asm/machines.h>
-#include <asm/sbus.h>
-
-#if 0
-static struct resource sun4c_timer_eb = { "sun4c_timer" };
-static struct resource sun4c_intr_eb = { "sun4c_intr" };
-#endif
 
 /*
  * Bit field defines for the interrupt registers on various
@@ -64,19 +59,7 @@ static struct resource sun4c_intr_eb = { "sun4c_intr" };
  *
  * so don't go making it static, like I tried. sigh.
  */
-unsigned char *interrupt_enable = NULL;
-
-static int sun4c_pil_map[] = { 0, 1, 2, 3, 5, 7, 8, 9 };
-
-static unsigned int sun4c_sbint_to_irq(struct sbus_dev *sdev,
-				       unsigned int sbint)
-{
-	if (sbint >= sizeof(sun4c_pil_map)) {
-		printk(KERN_ERR "%s: bogus SBINT %d\n", sdev->prom_name, sbint);
-		BUG();
-	}
-	return sun4c_pil_map[sbint];
-}
+unsigned char __iomem *interrupt_enable = NULL;
 
 static void sun4c_disable_irq(unsigned int irq_nr)
 {
@@ -85,7 +68,7 @@ static void sun4c_disable_irq(unsigned int irq_nr)
     
 	local_irq_save(flags);
 	irq_nr &= (NR_IRQS - 1);
-	current_mask = *interrupt_enable;
+	current_mask = sbus_readb(interrupt_enable);
 	switch(irq_nr) {
 	case 1:
 		new_mask = ((current_mask) & (~(SUN4C_INT_E1)));
@@ -103,7 +86,7 @@ static void sun4c_disable_irq(unsigned int irq_nr)
 		local_irq_restore(flags);
 		return;
 	}
-	*interrupt_enable = new_mask;
+	sbus_writeb(new_mask, interrupt_enable);
 	local_irq_restore(flags);
 }
 
@@ -114,7 +97,7 @@ static void sun4c_enable_irq(unsigned int irq_nr)
     
 	local_irq_save(flags);
 	irq_nr &= (NR_IRQS - 1);
-	current_mask = *interrupt_enable;
+	current_mask = sbus_readb(interrupt_enable);
 	switch(irq_nr) {
 	case 1:
 		new_mask = ((current_mask) | SUN4C_INT_E1);
@@ -132,37 +115,22 @@ static void sun4c_enable_irq(unsigned int irq_nr)
 		local_irq_restore(flags);
 		return;
 	}
-	*interrupt_enable = new_mask;
+	sbus_writeb(new_mask, interrupt_enable);
 	local_irq_restore(flags);
 }
 
-#define TIMER_IRQ  	10    /* Also at level 14, but we ignore that one. */
-#define PROFILE_IRQ	14    /* Level14 ticker.. used by OBP for polling */
-
-volatile struct sun4c_timer_info *sun4c_timers;
+struct sun4c_timer_info {
+	u32		l10_count;
+	u32		l10_limit;
+	u32		l14_count;
+	u32		l14_limit;
+};
 
-#ifdef CONFIG_SUN4
-/* This is an ugly hack to work around the
-   current timer code, and make it work with 
-   the sun4/260 intersil 
-   */
-volatile struct sun4c_timer_info sun4_timer;
-#endif
+static struct sun4c_timer_info __iomem *sun4c_timers;
 
 static void sun4c_clear_clock_irq(void)
 {
-	volatile unsigned int clear_intr;
-#ifdef CONFIG_SUN4
-	if (idprom->id_machtype == (SM_SUN4 | SM_4_260)) 
-	  clear_intr = sun4_timer.timer_limit10;
-	else
-#endif
-	clear_intr = sun4c_timers->timer_limit10;
-}
-
-static void sun4c_clear_profile_irq(int cpu)
-{
-	/* Errm.. not sure how to do this.. */
+	sbus_readl(&sun4c_timers->l10_limit);
 }
 
 static void sun4c_load_profile_irq(int cpu, unsigned int limit)
@@ -172,41 +140,48 @@ static void sun4c_load_profile_irq(int cpu, unsigned int limit)
 
 static void __init sun4c_init_timers(irq_handler_t counter_fn)
 {
-	int irq;
+	const struct linux_prom_irqs *irq;
+	struct device_node *dp;
+	const u32 *addr;
+	int err;
+
+	dp = of_find_node_by_name(NULL, "counter-timer");
+	if (!dp) {
+		prom_printf("sun4c_init_timers: Unable to find counter-timer\n");
+		prom_halt();
+	}
 
-	/* Map the Timer chip, this is implemented in hardware inside
-	 * the cache chip on the sun4c.
-	 */
-#ifdef CONFIG_SUN4
-	if (idprom->id_machtype == (SM_SUN4 | SM_4_260))
-		sun4c_timers = &sun4_timer;
-	else
-#endif
-	sun4c_timers = ioremap(SUN_TIMER_PHYSADDR,
-	    sizeof(struct sun4c_timer_info));
+	addr = of_get_property(dp, "address", NULL);
+	if (!addr) {
+		prom_printf("sun4c_init_timers: No address property\n");
+		prom_halt();
+	}
+
+	sun4c_timers = (void __iomem *) (unsigned long) addr[0];
+
+	irq = of_get_property(dp, "intr", NULL);
+	if (!irq) {
+		prom_printf("sun4c_init_timers: No intr property\n");
+		prom_halt();
+	}
 
 	/* Have the level 10 timer tick at 100HZ.  We don't touch the
 	 * level 14 timer limit since we are letting the prom handle
 	 * them until we have a real console driver so L1-A works.
 	 */
-	sun4c_timers->timer_limit10 = (((1000000/HZ) + 1) << 10);
-	master_l10_counter = &sun4c_timers->cur_count10;
-	master_l10_limit = &sun4c_timers->timer_limit10;
+	sbus_writel((((1000000/HZ) + 1) << 10), &sun4c_timers->l10_limit);
 
-	irq = request_irq(TIMER_IRQ,
-			  counter_fn,
+	master_l10_counter = &sun4c_timers->l10_count;
+
+	err = request_irq(irq[0].pri, counter_fn,
 			  (IRQF_DISABLED | SA_STATIC_ALLOC),
 			  "timer", NULL);
-	if (irq) {
-		prom_printf("time_init: unable to attach IRQ%d\n",TIMER_IRQ);
+	if (err) {
+		prom_printf("sun4c_init_timers: request_irq() fails with %d\n", err);
 		prom_halt();
 	}
     
-#if 0
-	/* This does not work on 4/330 */
-	sun4c_enable_irq(10);
-#endif
-	claim_ticker14(NULL, PROFILE_IRQ, 0);
+	sun4c_disable_irq(irq[1].pri);
 }
 
 #ifdef CONFIG_SMP
@@ -215,41 +190,28 @@ static void sun4c_nop(void) {}
 
 void __init sun4c_init_IRQ(void)
 {
-	struct linux_prom_registers int_regs[2];
-	int ie_node;
+	struct device_node *dp;
+	const u32 *addr;
 
-	if (ARCH_SUN4) {
-		interrupt_enable = (char *)
-		    ioremap(sun4_ie_physaddr, PAGE_SIZE);
-	} else {
-		struct resource phyres;
-
-		ie_node = prom_searchsiblings (prom_getchild(prom_root_node),
-				       	"interrupt-enable");
-		if(ie_node == 0)
-			panic("Cannot find /interrupt-enable node");
+	dp = of_find_node_by_name(NULL, "interrupt-enable");
+	if (!dp) {
+		prom_printf("sun4c_init_IRQ: Unable to find interrupt-enable\n");
+		prom_halt();
+	}
 
-		/* Depending on the "address" property is bad news... */
-		interrupt_enable = NULL;
-		if (prom_getproperty(ie_node, "reg", (char *) int_regs,
-				     sizeof(int_regs)) != -1) {
-			memset(&phyres, 0, sizeof(struct resource));
-			phyres.flags = int_regs[0].which_io;
-			phyres.start = int_regs[0].phys_addr;
-			interrupt_enable = (char *) sbus_ioremap(&phyres, 0,
-			    int_regs[0].reg_size, "sun4c_intr");
-		}
+	addr = of_get_property(dp, "address", NULL);
+	if (!addr) {
+		prom_printf("sun4c_init_IRQ: No address property\n");
+		prom_halt();
 	}
-	if (!interrupt_enable)
-		panic("Cannot map interrupt_enable");
 
-	BTFIXUPSET_CALL(sbint_to_irq, sun4c_sbint_to_irq, BTFIXUPCALL_NORM);
+	interrupt_enable = (void __iomem *) (unsigned long) addr[0];
+
 	BTFIXUPSET_CALL(enable_irq, sun4c_enable_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(disable_irq, sun4c_disable_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(enable_pil_irq, sun4c_enable_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(disable_pil_irq, sun4c_disable_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(clear_clock_irq, sun4c_clear_clock_irq, BTFIXUPCALL_NORM);
-	BTFIXUPSET_CALL(clear_profile_irq, sun4c_clear_profile_irq, BTFIXUPCALL_NOP);
 	BTFIXUPSET_CALL(load_profile_irq, sun4c_load_profile_irq, BTFIXUPCALL_NOP);
 	sparc_init_timers = sun4c_init_timers;
 #ifdef CONFIG_SMP
@@ -257,6 +219,6 @@ void __init sun4c_init_IRQ(void)
 	BTFIXUPSET_CALL(clear_cpu_int, sun4c_nop, BTFIXUPCALL_NOP);
 	BTFIXUPSET_CALL(set_irq_udt, sun4c_nop, BTFIXUPCALL_NOP);
 #endif
-	*interrupt_enable = (SUN4C_INT_ENABLE);
+	sbus_writeb(SUN4C_INT_ENABLE, interrupt_enable);
 	/* Cannot enable interrupts until OBP ticker is disabled. */
 }
diff --git a/arch/sparc/kernel/sun4d_irq.c b/arch/sparc/kernel/sun4d_irq.c
index 1290b5998f83..d3cb76ce418b 100644
--- a/arch/sparc/kernel/sun4d_irq.c
+++ b/arch/sparc/kernel/sun4d_irq.c
@@ -19,6 +19,8 @@
 #include <linux/smp.h>
 #include <linux/spinlock.h>
 #include <linux/seq_file.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/ptrace.h>
 #include <asm/processor.h>
@@ -34,7 +36,6 @@
 #include <asm/io.h>
 #include <asm/pgalloc.h>
 #include <asm/pgtable.h>
-#include <asm/sbus.h>
 #include <asm/sbi.h>
 #include <asm/cacheflush.h>
 #include <asm/irq_regs.h>
@@ -44,16 +45,22 @@
 /* If you trust current SCSI layer to handle different SCSI IRQs, enable this. I don't trust it... -jj */
 /* #define DISTRIBUTE_IRQS */
 
-struct sun4d_timer_regs *sun4d_timers;
+struct sun4d_timer_regs {
+	u32	l10_timer_limit;
+	u32	l10_cur_countx;
+	u32	l10_limit_noclear;
+	u32	ctrl;
+	u32	l10_cur_count;
+};
+
+static struct sun4d_timer_regs __iomem *sun4d_timers;
+
 #define TIMER_IRQ	10
 
 #define MAX_STATIC_ALLOC	4
 extern struct irqaction static_irqaction[MAX_STATIC_ALLOC];
 extern int static_irq_count;
-unsigned char cpu_leds[32];
-#ifdef CONFIG_SMP
 static unsigned char sbus_tid[32];
-#endif
 
 static struct irqaction *irq_action[NR_IRQS];
 extern spinlock_t irq_action_lock;
@@ -72,9 +79,9 @@ static int sbus_to_pil[] = {
 };
 
 static int nsbi;
-#ifdef CONFIG_SMP
+
+/* Exported for sun4d_smp.c */
 DEFINE_SPINLOCK(sun4d_imsk_lock);
-#endif
 
 int show_sun4d_interrupts(struct seq_file *p, void *v)
 {
@@ -257,26 +264,6 @@ void sun4d_handler_irq(int irq, struct pt_regs * regs)
 	set_irq_regs(old_regs);
 }
 
-unsigned int sun4d_build_irq(struct sbus_dev *sdev, int irq)
-{
-	int sbusl = pil_to_sbus[irq];
-
-	if (sbusl)
-		return ((sdev->bus->board + 1) << 5) + (sbusl << 2) + sdev->slot;
-	else
-		return irq;
-}
-
-static unsigned int sun4d_sbint_to_irq(struct sbus_dev *sdev,
-				       unsigned int sbint)
-{
-	if (sbint >= sizeof(sbus_to_pil)) {
-		printk(KERN_ERR "%s: bogus SBINT %d\n", sdev->prom_name, sbint);
-		BUG();
-	}
-	return sun4d_build_irq(sdev, sbus_to_pil[sbint]);
-}
-
 int sun4d_request_irq(unsigned int irq,
 		irq_handler_t handler,
 		unsigned long irqflags, const char * devname, void *dev_id)
@@ -360,36 +347,28 @@ out:
 
 static void sun4d_disable_irq(unsigned int irq)
 {
-#ifdef CONFIG_SMP
 	int tid = sbus_tid[(irq >> 5) - 1];
 	unsigned long flags;
-#endif	
 	
-	if (irq < NR_IRQS) return;
-#ifdef CONFIG_SMP
+	if (irq < NR_IRQS)
+		return;
+
 	spin_lock_irqsave(&sun4d_imsk_lock, flags);
 	cc_set_imsk_other(tid, cc_get_imsk_other(tid) | (1 << sbus_to_pil[(irq >> 2) & 7]));
 	spin_unlock_irqrestore(&sun4d_imsk_lock, flags);
-#else		
-	cc_set_imsk(cc_get_imsk() | (1 << sbus_to_pil[(irq >> 2) & 7]));
-#endif
 }
 
 static void sun4d_enable_irq(unsigned int irq)
 {
-#ifdef CONFIG_SMP
 	int tid = sbus_tid[(irq >> 5) - 1];
 	unsigned long flags;
-#endif	
 	
-	if (irq < NR_IRQS) return;
-#ifdef CONFIG_SMP
+	if (irq < NR_IRQS)
+		return;
+
 	spin_lock_irqsave(&sun4d_imsk_lock, flags);
 	cc_set_imsk_other(tid, cc_get_imsk_other(tid) & ~(1 << sbus_to_pil[(irq >> 2) & 7]));
 	spin_unlock_irqrestore(&sun4d_imsk_lock, flags);
-#else		
-	cc_set_imsk(cc_get_imsk() & ~(1 << sbus_to_pil[(irq >> 2) & 7]));
-#endif
 }
 
 #ifdef CONFIG_SMP
@@ -409,47 +388,55 @@ static void sun4d_set_udt(int cpu)
 /* Setup IRQ distribution scheme. */
 void __init sun4d_distribute_irqs(void)
 {
+	struct device_node *dp;
+
 #ifdef DISTRIBUTE_IRQS
-	struct sbus_bus *sbus;
-	unsigned long sbus_serving_map;
+	cpumask_t sbus_serving_map;
 
 	sbus_serving_map = cpu_present_map;
-	for_each_sbus(sbus) {
-		if ((sbus->board * 2) == boot_cpu_id && (cpu_present_map & (1 << (sbus->board * 2 + 1))))
-			sbus_tid[sbus->board] = (sbus->board * 2 + 1);
-		else if (cpu_present_map & (1 << (sbus->board * 2)))
-			sbus_tid[sbus->board] = (sbus->board * 2);
-		else if (cpu_present_map & (1 << (sbus->board * 2 + 1)))
-			sbus_tid[sbus->board] = (sbus->board * 2 + 1);
+	for_each_node_by_name(dp, "sbi") {
+		int board = of_getintprop_default(dp, "board#", 0);
+
+		if ((board * 2) == boot_cpu_id && cpu_isset(board * 2 + 1, cpu_present_map))
+			sbus_tid[board] = (board * 2 + 1);
+		else if (cpu_isset(board * 2, cpu_present_map))
+			sbus_tid[board] = (board * 2);
+		else if (cpu_isset(board * 2 + 1, cpu_present_map))
+			sbus_tid[board] = (board * 2 + 1);
 		else
-			sbus_tid[sbus->board] = 0xff;
-		if (sbus_tid[sbus->board] != 0xff)
-			sbus_serving_map &= ~(1 << sbus_tid[sbus->board]);
+			sbus_tid[board] = 0xff;
+		if (sbus_tid[board] != 0xff)
+			cpu_clear(sbus_tid[board], sbus_serving_map);
 	}
-	for_each_sbus(sbus)
-		if (sbus_tid[sbus->board] == 0xff) {
+	for_each_node_by_name(dp, "sbi") {
+		int board = of_getintprop_default(dp, "board#", 0);
+		if (sbus_tid[board] == 0xff) {
 			int i = 31;
 				
-			if (!sbus_serving_map)
+			if (cpus_empty(sbus_serving_map))
 				sbus_serving_map = cpu_present_map;
-			while (!(sbus_serving_map & (1 << i)))
+			while (cpu_isset(i, sbus_serving_map))
 				i--;
-			sbus_tid[sbus->board] = i;
-			sbus_serving_map &= ~(1 << i);
+			sbus_tid[board] = i;
+			cpu_clear(i, sbus_serving_map);
 		}
-	for_each_sbus(sbus) {
-		printk("sbus%d IRQs directed to CPU%d\n", sbus->board, sbus_tid[sbus->board]);
-		set_sbi_tid(sbus->devid, sbus_tid[sbus->board] << 3);
+	}
+	for_each_node_by_name(dp, "sbi") {
+		int devid = of_getintprop_default(dp, "device-id", 0);
+		int board = of_getintprop_default(dp, "board#", 0);
+		printk("sbus%d IRQs directed to CPU%d\n", board, sbus_tid[board]);
+		set_sbi_tid(devid, sbus_tid[board] << 3);
 	}
 #else
-	struct sbus_bus *sbus;
 	int cpuid = cpu_logical_map(1);
 
 	if (cpuid == -1)
 		cpuid = cpu_logical_map(0);
-	for_each_sbus(sbus) {
-		sbus_tid[sbus->board] = cpuid;
-		set_sbi_tid(sbus->devid, cpuid << 3);
+	for_each_node_by_name(dp, "sbi") {
+		int devid = of_getintprop_default(dp, "device-id", 0);
+		int board = of_getintprop_default(dp, "board#", 0);
+		sbus_tid[board] = cpuid;
+		set_sbi_tid(devid, cpuid << 3);
 	}
 	printk("All sbus IRQs directed to CPU%d\n", cpuid);
 #endif
@@ -458,13 +445,7 @@ void __init sun4d_distribute_irqs(void)
  
 static void sun4d_clear_clock_irq(void)
 {
-	volatile unsigned int clear_intr;
-	clear_intr = sun4d_timers->l10_timer_limit;
-}
-
-static void sun4d_clear_profile_irq(int cpu)
-{
-	bw_get_prof_limit(cpu);
+	sbus_readl(&sun4d_timers->l10_timer_limit);
 }
 
 static void sun4d_load_profile_irq(int cpu, unsigned int limit)
@@ -472,98 +453,121 @@ static void sun4d_load_profile_irq(int cpu, unsigned int limit)
 	bw_set_prof_limit(cpu, limit);
 }
 
-static void __init sun4d_init_timers(irq_handler_t counter_fn)
+static void __init sun4d_load_profile_irqs(void)
 {
-	int irq;
-	int cpu;
-	struct resource r;
-	int mid;
+	int cpu = 0, mid;
 
-	/* Map the User Timer registers. */
-	memset(&r, 0, sizeof(r));
+	while (!cpu_find_by_instance(cpu, NULL, &mid)) {
+		sun4d_load_profile_irq(mid >> 3, 0);
+		cpu++;
+	}
+}
+
+static void __init sun4d_fixup_trap_table(void)
+{
 #ifdef CONFIG_SMP
-	r.start = CSR_BASE(boot_cpu_id)+BW_TIMER_LIMIT;
-#else
-	r.start = CSR_BASE(0)+BW_TIMER_LIMIT;
+	unsigned long flags;
+	extern unsigned long lvl14_save[4];
+	struct tt_entry *trap_table = &sparc_ttable[SP_TRAP_IRQ1 + (14 - 1)];
+	extern unsigned int real_irq_entry[], smp4d_ticker[];
+	extern unsigned int patchme_maybe_smp_msg[];
+
+	/* Adjust so that we jump directly to smp4d_ticker */
+	lvl14_save[2] += smp4d_ticker - real_irq_entry;
+
+	/* For SMP we use the level 14 ticker, however the bootup code
+	 * has copied the firmware's level 14 vector into the boot cpu's
+	 * trap table, we must fix this now or we get squashed.
+	 */
+	local_irq_save(flags);
+	patchme_maybe_smp_msg[0] = 0x01000000; /* NOP out the branch */
+	trap_table->inst_one = lvl14_save[0];
+	trap_table->inst_two = lvl14_save[1];
+	trap_table->inst_three = lvl14_save[2];
+	trap_table->inst_four = lvl14_save[3];
+	local_flush_cache_all();
+	local_irq_restore(flags);
 #endif
-	r.flags = 0xf;
-	sun4d_timers = (struct sun4d_timer_regs *) sbus_ioremap(&r, 0,
-	    PAGE_SIZE, "user timer");
+}
 
-	sun4d_timers->l10_timer_limit =  (((1000000/HZ) + 1) << 10);
-	master_l10_counter = &sun4d_timers->l10_cur_count;
-	master_l10_limit = &sun4d_timers->l10_timer_limit;
+static void __init sun4d_init_timers(irq_handler_t counter_fn)
+{
+	struct device_node *dp;
+	struct resource res;
+	const u32 *reg;
+	int err;
+
+	dp = of_find_node_by_name(NULL, "cpu-unit");
+	if (!dp) {
+		prom_printf("sun4d_init_timers: Unable to find cpu-unit\n");
+		prom_halt();
+	}
 
-	irq = request_irq(TIMER_IRQ,
-			  counter_fn,
-			  (IRQF_DISABLED | SA_STATIC_ALLOC),
-			  "timer", NULL);
-	if (irq) {
-		prom_printf("time_init: unable to attach IRQ%d\n",TIMER_IRQ);
+	/* Which cpu-unit we use is arbitrary, we can view the bootbus timer
+	 * registers via any cpu's mapping.  The first 'reg' property is the
+	 * bootbus.
+	 */
+	reg = of_get_property(dp, "reg", NULL);
+	if (!reg) {
+		prom_printf("sun4d_init_timers: No reg property\n");
 		prom_halt();
 	}
-	
-	/* Enable user timer free run for CPU 0 in BW */
-	/* bw_set_ctrl(0, bw_get_ctrl(0) | BW_CTRL_USER_TIMER); */
 
-	cpu = 0;
-	while (!cpu_find_by_instance(cpu, NULL, &mid)) {
-		sun4d_load_profile_irq(mid >> 3, 0);
-		cpu++;
+	res.start = reg[1];
+	res.end = reg[2] - 1;
+	res.flags = reg[0] & 0xff;
+	sun4d_timers = of_ioremap(&res, BW_TIMER_LIMIT,
+				  sizeof(struct sun4d_timer_regs), "user timer");
+	if (!sun4d_timers) {
+		prom_printf("sun4d_init_timers: Can't map timer regs\n");
+		prom_halt();
 	}
-		
-#ifdef CONFIG_SMP
-	{
-		unsigned long flags;
-		extern unsigned long lvl14_save[4];
-		struct tt_entry *trap_table = &sparc_ttable[SP_TRAP_IRQ1 + (14 - 1)];
-		extern unsigned int real_irq_entry[], smp4d_ticker[];
-		extern unsigned int patchme_maybe_smp_msg[];
-
-		/* Adjust so that we jump directly to smp4d_ticker */
-		lvl14_save[2] += smp4d_ticker - real_irq_entry;
-
-		/* For SMP we use the level 14 ticker, however the bootup code
-		 * has copied the firmware's level 14 vector into the boot cpu's
-		 * trap table, we must fix this now or we get squashed.
-		 */
-		local_irq_save(flags);
-		patchme_maybe_smp_msg[0] = 0x01000000; /* NOP out the branch */
-		trap_table->inst_one = lvl14_save[0];
-		trap_table->inst_two = lvl14_save[1];
-		trap_table->inst_three = lvl14_save[2];
-		trap_table->inst_four = lvl14_save[3];
-		local_flush_cache_all();
-		local_irq_restore(flags);
+
+	sbus_writel((((1000000/HZ) + 1) << 10), &sun4d_timers->l10_timer_limit);
+
+	master_l10_counter = &sun4d_timers->l10_cur_count;
+
+	err = request_irq(TIMER_IRQ, counter_fn,
+			  (IRQF_DISABLED | SA_STATIC_ALLOC),
+			  "timer", NULL);
+	if (err) {
+		prom_printf("sun4d_init_timers: request_irq() failed with %d\n", err);
+		prom_halt();
 	}
-#endif
+	sun4d_load_profile_irqs();
+	sun4d_fixup_trap_table();
 }
 
 void __init sun4d_init_sbi_irq(void)
 {
-	struct sbus_bus *sbus;
-	unsigned mask;
+	struct device_node *dp;
+	int target_cpu = 0;
+
+#ifdef CONFIG_SMP
+	target_cpu = boot_cpu_id;
+#endif
 
 	nsbi = 0;
-	for_each_sbus(sbus)
+	for_each_node_by_name(dp, "sbi")
 		nsbi++;
 	sbus_actions = kzalloc (nsbi * 8 * 4 * sizeof(struct sbus_action), GFP_ATOMIC);
 	if (!sbus_actions) {
 		prom_printf("SUN4D: Cannot allocate sbus_actions, halting.\n");
 		prom_halt();
 	}
-	for_each_sbus(sbus) {
-#ifdef CONFIG_SMP	
-		extern unsigned char boot_cpu_id;
-		
-		set_sbi_tid(sbus->devid, boot_cpu_id << 3);
-		sbus_tid[sbus->board] = boot_cpu_id;
-#endif
+	for_each_node_by_name(dp, "sbi") {
+		int devid = of_getintprop_default(dp, "device-id", 0);
+		int board = of_getintprop_default(dp, "board#", 0);
+		unsigned int mask;
+
+		set_sbi_tid(devid, target_cpu << 3);
+		sbus_tid[board] = target_cpu;
+
 		/* Get rid of pending irqs from PROM */
-		mask = acquire_sbi(sbus->devid, 0xffffffff);
+		mask = acquire_sbi(devid, 0xffffffff);
 		if (mask) {
-			printk ("Clearing pending IRQs %08x on SBI %d\n", mask, sbus->board);
-			release_sbi(sbus->devid, mask);
+			printk ("Clearing pending IRQs %08x on SBI %d\n", mask, board);
+			release_sbi(devid, mask);
 		}
 	}
 }
@@ -572,11 +576,9 @@ void __init sun4d_init_IRQ(void)
 {
 	local_irq_disable();
 
-	BTFIXUPSET_CALL(sbint_to_irq, sun4d_sbint_to_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(enable_irq, sun4d_enable_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(disable_irq, sun4d_disable_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(clear_clock_irq, sun4d_clear_clock_irq, BTFIXUPCALL_NORM);
-	BTFIXUPSET_CALL(clear_profile_irq, sun4d_clear_profile_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(load_profile_irq, sun4d_load_profile_irq, BTFIXUPCALL_NORM);
 	sparc_init_timers = sun4d_init_timers;
 #ifdef CONFIG_SMP
diff --git a/arch/sparc/kernel/sun4d_smp.c b/arch/sparc/kernel/sun4d_smp.c
index 446767e8f569..7a6a5e795928 100644
--- a/arch/sparc/kernel/sun4d_smp.c
+++ b/arch/sparc/kernel/sun4d_smp.c
@@ -20,6 +20,7 @@
 #include <linux/swap.h>
 #include <linux/profile.h>
 #include <linux/delay.h>
+#include <linux/cpu.h>
 
 #include <asm/ptrace.h>
 #include <asm/atomic.h>
@@ -30,7 +31,6 @@
 #include <asm/pgalloc.h>
 #include <asm/pgtable.h>
 #include <asm/oplib.h>
-#include <asm/sbus.h>
 #include <asm/sbi.h>
 #include <asm/tlbflush.h>
 #include <asm/cacheflush.h>
@@ -72,6 +72,17 @@ static void smp_setup_percpu_timer(void);
 extern void cpu_probe(void);
 extern void sun4d_distribute_irqs(void);
 
+static unsigned char cpu_leds[32];
+
+static inline void show_leds(int cpuid)
+{
+	cpuid &= 0x1e;
+	__asm__ __volatile__ ("stba %0, [%1] %2" : :
+			      "r" ((cpu_leds[cpuid] << 4) | cpu_leds[cpuid+1]),
+			      "r" (ECSR_BASE(cpuid) | BB_LEDS),
+			      "i" (ASI_M_CTL));
+}
+
 void __init smp4d_callin(void)
 {
 	int cpuid = hard_smp4d_processor_id();
diff --git a/arch/sparc/kernel/sun4m_irq.c b/arch/sparc/kernel/sun4m_irq.c
index 94e02de960ea..f10317179ee6 100644
--- a/arch/sparc/kernel/sun4m_irq.c
+++ b/arch/sparc/kernel/sun4m_irq.c
@@ -20,6 +20,8 @@
 #include <linux/slab.h>
 #include <linux/init.h>
 #include <linux/ioport.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/ptrace.h>
 #include <asm/processor.h>
@@ -35,59 +37,27 @@
 #include <asm/smp.h>
 #include <asm/irq.h>
 #include <asm/io.h>
-#include <asm/sbus.h>
 #include <asm/cacheflush.h>
 
 #include "irq.h"
 
-/* On the sun4m, just like the timers, we have both per-cpu and master
- * interrupt registers.
- */
-
-/* These registers are used for sending/receiving irqs from/to
- * different cpu's.
- */
-struct sun4m_intreg_percpu {
-	unsigned int tbt;        /* Interrupts still pending for this cpu. */
-
-	/* These next two registers are WRITE-ONLY and are only
-	 * "on bit" sensitive, "off bits" written have NO affect.
-	 */
-	unsigned int clear;  /* Clear this cpus irqs here. */
-	unsigned int set;    /* Set this cpus irqs here. */
-	unsigned char space[PAGE_SIZE - 12];
+struct sun4m_irq_percpu {
+	u32		pending;
+	u32		clear;
+	u32		set;
 };
 
-/*
- * djhr
- * Actually the clear and set fields in this struct are misleading..
- * according to the SLAVIO manual (and the same applies for the SEC)
- * the clear field clears bits in the mask which will ENABLE that IRQ
- * the set field sets bits in the mask to DISABLE the IRQ.
- *
- * Also the undirected_xx address in the SLAVIO is defined as
- * RESERVED and write only..
- *
- * DAVEM_NOTE: The SLAVIO only specifies behavior on uniprocessor
- *             sun4m machines, for MP the layout makes more sense.
- */
-struct sun4m_intregs {
-	struct sun4m_intreg_percpu cpu_intregs[SUN4M_NCPUS];
-	unsigned int tbt;                /* IRQ's that are still pending. */
-	unsigned int irqs;               /* Master IRQ bits. */
-
-	/* Again, like the above, two these registers are WRITE-ONLY. */
-	unsigned int clear;              /* Clear master IRQ's by setting bits here. */
-	unsigned int set;                /* Set master IRQ's by setting bits here. */
-
-	/* This register is both READ and WRITE. */
-	unsigned int undirected_target;  /* Which cpu gets undirected irqs. */
+struct sun4m_irq_global {
+	u32		pending;
+	u32		mask;
+	u32		mask_clear;
+	u32		mask_set;
+	u32		interrupt_target;
 };
 
-static unsigned long dummy;
-
-struct sun4m_intregs *sun4m_interrupts;
-unsigned long *irq_rcvreg = &dummy;
+/* Code in entry.S needs to get at these register mappings.  */
+struct sun4m_irq_percpu __iomem *sun4m_irq_percpu[SUN4M_NCPUS];
+struct sun4m_irq_global __iomem *sun4m_irq_global;
 
 /* Dave Redman (djhr@tadpole.co.uk)
  * The sun4m interrupt registers.
@@ -101,8 +71,9 @@ unsigned long *irq_rcvreg = &dummy;
 
 #define	SUN4M_INT_MASKALL	0x80000000	  /* mask all interrupts */
 #define	SUN4M_INT_MODULE_ERR	0x40000000	  /* module error */
-#define	SUN4M_INT_M2S_WRITE	0x20000000	  /* write buffer error */
-#define	SUN4M_INT_ECC		0x10000000	  /* ecc memory error */
+#define	SUN4M_INT_M2S_WRITE_ERR	0x20000000	  /* write buffer error */
+#define	SUN4M_INT_ECC_ERR	0x10000000	  /* ecc memory error */
+#define	SUN4M_INT_VME_ERR	0x08000000	  /* vme async error */
 #define	SUN4M_INT_FLOPPY	0x00400000	  /* floppy disk */
 #define	SUN4M_INT_MODULE	0x00200000	  /* module interrupt */
 #define	SUN4M_INT_VIDEO		0x00100000	  /* onboard video */
@@ -113,75 +84,126 @@ unsigned long *irq_rcvreg = &dummy;
 #define	SUN4M_INT_SERIAL	0x00008000	  /* serial ports */
 #define	SUN4M_INT_KBDMS		0x00004000	  /* keyboard/mouse */
 #define	SUN4M_INT_SBUSBITS	0x00003F80	  /* sbus int bits */
+#define	SUN4M_INT_VMEBITS	0x0000007F	  /* vme int bits */
+
+#define	SUN4M_INT_ERROR		(SUN4M_INT_MODULE_ERR |    \
+				 SUN4M_INT_M2S_WRITE_ERR | \
+				 SUN4M_INT_ECC_ERR |       \
+				 SUN4M_INT_VME_ERR)
 
 #define SUN4M_INT_SBUS(x)	(1 << (x+7))
 #define SUN4M_INT_VME(x)	(1 << (x))
 
-/* These tables only apply for interrupts greater than 15..
- * 
- * any intr value below 0x10 is considered to be a soft-int
- * this may be useful or it may not.. but that's how I've done it.
- * and it won't clash with what OBP is telling us about devices.
+/* Interrupt levels used by OBP */
+#define	OBP_INT_LEVEL_SOFT	0x10
+#define	OBP_INT_LEVEL_ONBOARD	0x20
+#define	OBP_INT_LEVEL_SBUS	0x30
+#define	OBP_INT_LEVEL_VME	0x40
+
+/* Interrupt level assignment on sun4m:
+ *
+ *	level		source
+ * ------------------------------------------------------------
+ *        1		softint-1
+ *	  2		softint-2, VME/SBUS level 1
+ *	  3		softint-3, VME/SBUS level 2
+ *	  4		softint-4, onboard SCSI
+ *	  5		softint-5, VME/SBUS level 3
+ *	  6		softint-6, onboard ETHERNET
+ *	  7		softint-7, VME/SBUS level 4
+ *	  8		softint-8, onboard VIDEO
+ *	  9		softint-9, VME/SBUS level 5, Module Interrupt
+ *	 10		softint-10, system counter/timer
+ *	 11		softint-11, VME/SBUS level 6, Floppy
+ *	 12		softint-12, Keyboard/Mouse, Serial
+ *	 13		softint-13, VME/SBUS level 7, ISDN Audio
+ *	 14		softint-14, per-processor counter/timer
+ *	 15		softint-15, Asynchronous Errors (broadcast)
  *
- * take an encoded intr value and lookup if it's valid
- * then get the mask bits that match from irq_mask
+ * Each interrupt source is masked distinctly in the sun4m interrupt
+ * registers.  The PIL level alone is therefore ambiguous, since multiple
+ * interrupt sources map to a single PIL.
  *
- * P3: Translation from irq 0x0d to mask 0x2000 is for MrCoffee.
+ * This ambiguity is resolved in the 'intr' property for device nodes
+ * in the OF device tree.  Each 'intr' property entry is composed of
+ * two 32-bit words.  The first word is the IRQ priority value, which
+ * is what we're intersted in.  The second word is the IRQ vector, which
+ * is unused.
+ *
+ * The low 4 bits of the IRQ priority indicate the PIL, and the upper
+ * 4 bits indicate onboard vs. SBUS leveled vs. VME leveled.  0x20
+ * means onboard, 0x30 means SBUS leveled, and 0x40 means VME leveled.
+ *
+ * For example, an 'intr' IRQ priority value of 0x24 is onboard SCSI
+ * whereas a value of 0x33 is SBUS level 2.  Here are some sample
+ * 'intr' property IRQ priority values from ss4, ss5, ss10, ss20, and
+ * Tadpole S3 GX systems.
+ *
+ * esp: 	0x24	onboard ESP SCSI
+ * le:  	0x26	onboard Lance ETHERNET
+ * p9100:	0x32	SBUS level 1 P9100 video
+ * bpp:  	0x33	SBUS level 2 BPP parallel port device
+ * DBRI:	0x39	SBUS level 5 DBRI ISDN audio
+ * SUNW,leo:	0x39	SBUS level 5 LEO video
+ * pcmcia:	0x3b	SBUS level 6 PCMCIA controller
+ * uctrl:	0x3b	SBUS level 6 UCTRL device
+ * modem:	0x3d	SBUS level 7 MODEM
+ * zs:		0x2c	onboard keyboard/mouse/serial
+ * floppy:	0x2b	onboard Floppy
+ * power:	0x22	onboard power device (XXX unknown mask bit XXX)
  */
-static unsigned char irq_xlate[32] = {
-    /*  0,  1,  2,  3,  4,  5,  6,  7,  8,  9,  a,  b,  c,  d,  e,  f */
-	0,  0,  0,  0,  1,  0,  2,  0,  3,  0,  4,  5,  6, 14,  0,  7,
-	0,  0,  8,  9,  0, 10,  0, 11,  0, 12,  0, 13,  0, 14,  0,  0
-};
 
-static unsigned long irq_mask[] = {
-	0,						  /* illegal index */
-	SUN4M_INT_SCSI,				  	  /*  1 irq 4 */
-	SUN4M_INT_ETHERNET,				  /*  2 irq 6 */
-	SUN4M_INT_VIDEO,				  /*  3 irq 8 */
-	SUN4M_INT_REALTIME,				  /*  4 irq 10 */
-	SUN4M_INT_FLOPPY,				  /*  5 irq 11 */
-	(SUN4M_INT_SERIAL | SUN4M_INT_KBDMS),	  	  /*  6 irq 12 */
-	SUN4M_INT_MODULE_ERR,			  	  /*  7 irq 15 */
-	SUN4M_INT_SBUS(0),				  /*  8 irq 2 */
-	SUN4M_INT_SBUS(1),				  /*  9 irq 3 */
-	SUN4M_INT_SBUS(2),				  /* 10 irq 5 */
-	SUN4M_INT_SBUS(3),				  /* 11 irq 7 */
-	SUN4M_INT_SBUS(4),				  /* 12 irq 9 */
-	SUN4M_INT_SBUS(5),				  /* 13 irq 11 */
-	SUN4M_INT_SBUS(6)				  /* 14 irq 13 */
+static unsigned long irq_mask[0x50] = {
+	/* SMP */
+	0,  SUN4M_SOFT_INT(1),
+	SUN4M_SOFT_INT(2),  SUN4M_SOFT_INT(3),
+	SUN4M_SOFT_INT(4),  SUN4M_SOFT_INT(5),
+	SUN4M_SOFT_INT(6),  SUN4M_SOFT_INT(7),
+	SUN4M_SOFT_INT(8),  SUN4M_SOFT_INT(9),
+	SUN4M_SOFT_INT(10), SUN4M_SOFT_INT(11),
+	SUN4M_SOFT_INT(12), SUN4M_SOFT_INT(13),
+	SUN4M_SOFT_INT(14), SUN4M_SOFT_INT(15),
+	/* soft */
+	0,  SUN4M_SOFT_INT(1),
+	SUN4M_SOFT_INT(2),  SUN4M_SOFT_INT(3),
+	SUN4M_SOFT_INT(4),  SUN4M_SOFT_INT(5),
+	SUN4M_SOFT_INT(6),  SUN4M_SOFT_INT(7),
+	SUN4M_SOFT_INT(8),  SUN4M_SOFT_INT(9),
+	SUN4M_SOFT_INT(10), SUN4M_SOFT_INT(11),
+	SUN4M_SOFT_INT(12), SUN4M_SOFT_INT(13),
+	SUN4M_SOFT_INT(14), SUN4M_SOFT_INT(15),
+	/* onboard */
+	0, 0, 0, 0,
+	SUN4M_INT_SCSI,  0, SUN4M_INT_ETHERNET, 0,
+	SUN4M_INT_VIDEO, SUN4M_INT_MODULE,
+	SUN4M_INT_REALTIME, SUN4M_INT_FLOPPY,
+	(SUN4M_INT_SERIAL | SUN4M_INT_KBDMS),
+	SUN4M_INT_AUDIO, 0, SUN4M_INT_MODULE_ERR,
+	/* sbus */
+	0, 0, SUN4M_INT_SBUS(0), SUN4M_INT_SBUS(1),
+	0, SUN4M_INT_SBUS(2), 0, SUN4M_INT_SBUS(3),
+	0, SUN4M_INT_SBUS(4), 0, SUN4M_INT_SBUS(5),
+	0, SUN4M_INT_SBUS(6), 0, 0,
+	/* vme */
+	0, 0, SUN4M_INT_VME(0), SUN4M_INT_VME(1),
+	0, SUN4M_INT_VME(2), 0, SUN4M_INT_VME(3),
+	0, SUN4M_INT_VME(4), 0, SUN4M_INT_VME(5),
+	0, SUN4M_INT_VME(6), 0, 0
 };
 
-static int sun4m_pil_map[] = { 0, 2, 3, 5, 7, 9, 11, 13 };
-
-static unsigned int sun4m_sbint_to_irq(struct sbus_dev *sdev,
-				       unsigned int sbint)
-{
-	if (sbint >= sizeof(sun4m_pil_map)) {
-		printk(KERN_ERR "%s: bogus SBINT %d\n", sdev->prom_name, sbint);
-		BUG();
-	}
-	return sun4m_pil_map[sbint] | 0x30;
-}
-
 static unsigned long sun4m_get_irqmask(unsigned int irq)
 {
 	unsigned long mask;
     
-	if (irq > 0x20) {
-		/* OBIO/SBUS interrupts */
-		irq &= 0x1f;
-		mask = irq_mask[irq_xlate[irq]];
-		if (!mask)
-			printk("sun4m_get_irqmask: IRQ%d has no valid mask!\n",irq);
-	} else {
-		/* Soft Interrupts will come here.
-		 * Currently there is no way to trigger them but I'm sure
-		 * something could be cooked up.
-		 */
-		irq &= 0xf;
-		mask = SUN4M_SOFT_INT(irq);
-	}
+	if (irq < 0x50)
+		mask = irq_mask[irq];
+	else
+		mask = 0;
+
+	if (!mask)
+		printk(KERN_ERR "sun4m_get_irqmask: IRQ%d has no valid mask!\n",
+		       irq);
+
 	return mask;
 }
 
@@ -193,9 +215,9 @@ static void sun4m_disable_irq(unsigned int irq_nr)
 	mask = sun4m_get_irqmask(irq_nr);
 	local_irq_save(flags);
 	if (irq_nr > 15)
-		sun4m_interrupts->set = mask;
+		sbus_writel(mask, &sun4m_irq_global->mask_set);
 	else
-		sun4m_interrupts->cpu_intregs[cpu].set = mask;
+		sbus_writel(mask, &sun4m_irq_percpu[cpu]->set);
 	local_irq_restore(flags);    
 }
 
@@ -212,13 +234,13 @@ static void sun4m_enable_irq(unsigned int irq_nr)
 		mask = sun4m_get_irqmask(irq_nr);
 		local_irq_save(flags);
 		if (irq_nr > 15)
-			sun4m_interrupts->clear = mask;
+			sbus_writel(mask, &sun4m_irq_global->mask_clear);
 		else
-			sun4m_interrupts->cpu_intregs[cpu].clear = mask;
+			sbus_writel(mask, &sun4m_irq_percpu[cpu]->clear);
 		local_irq_restore(flags);    
 	} else {
 		local_irq_save(flags);
-		sun4m_interrupts->clear = SUN4M_INT_FLOPPY;
+		sbus_writel(SUN4M_INT_FLOPPY, &sun4m_irq_global->mask_clear);
 		local_irq_restore(flags);
 	}
 }
@@ -236,10 +258,10 @@ static unsigned long cpu_pil_to_imask[16] = {
 /*9*/	SUN4M_INT_SBUS(4) | SUN4M_INT_VME(4) | SUN4M_INT_MODULE_ERR,
 /*10*/	SUN4M_INT_REALTIME,
 /*11*/	SUN4M_INT_SBUS(5) | SUN4M_INT_VME(5) | SUN4M_INT_FLOPPY,
-/*12*/	SUN4M_INT_SERIAL | SUN4M_INT_KBDMS,
-/*13*/	SUN4M_INT_AUDIO,
+/*12*/	SUN4M_INT_SERIAL  | SUN4M_INT_KBDMS,
+/*13*/	SUN4M_INT_SBUS(6) | SUN4M_INT_VME(6) | SUN4M_INT_AUDIO,
 /*14*/	SUN4M_INT_E14,
-/*15*/	0x00000000
+/*15*/	SUN4M_INT_ERROR
 };
 
 /* We assume the caller has disabled local interrupts when these are called,
@@ -247,126 +269,141 @@ static unsigned long cpu_pil_to_imask[16] = {
  */
 static void sun4m_disable_pil_irq(unsigned int pil)
 {
-	sun4m_interrupts->set = cpu_pil_to_imask[pil];
+	sbus_writel(cpu_pil_to_imask[pil], &sun4m_irq_global->mask_set);
 }
 
 static void sun4m_enable_pil_irq(unsigned int pil)
 {
-	sun4m_interrupts->clear = cpu_pil_to_imask[pil];
+	sbus_writel(cpu_pil_to_imask[pil], &sun4m_irq_global->mask_clear);
 }
 
 #ifdef CONFIG_SMP
 static void sun4m_send_ipi(int cpu, int level)
 {
-	unsigned long mask;
-
-	mask = sun4m_get_irqmask(level);
-	sun4m_interrupts->cpu_intregs[cpu].set = mask;
+	unsigned long mask = sun4m_get_irqmask(level);
+	sbus_writel(mask, &sun4m_irq_percpu[cpu]->set);
 }
 
 static void sun4m_clear_ipi(int cpu, int level)
 {
-	unsigned long mask;
-
-	mask = sun4m_get_irqmask(level);
-	sun4m_interrupts->cpu_intregs[cpu].clear = mask;
+	unsigned long mask = sun4m_get_irqmask(level);
+	sbus_writel(mask, &sun4m_irq_percpu[cpu]->clear);
 }
 
 static void sun4m_set_udt(int cpu)
 {
-	sun4m_interrupts->undirected_target = cpu;
+	sbus_writel(cpu, &sun4m_irq_global->interrupt_target);
 }
 #endif
 
-#define OBIO_INTR	0x20
-#define TIMER_IRQ  	(OBIO_INTR | 10)
-#define PROFILE_IRQ	(OBIO_INTR | 14)
+struct sun4m_timer_percpu {
+	u32		l14_limit;
+	u32		l14_count;
+	u32		l14_limit_noclear;
+	u32		user_timer_start_stop;
+};
+
+static struct sun4m_timer_percpu __iomem *timers_percpu[SUN4M_NCPUS];
+
+struct sun4m_timer_global {
+	u32		l10_limit;
+	u32		l10_count;
+	u32		l10_limit_noclear;
+	u32		reserved;
+	u32		timer_config;
+};
+
+static struct sun4m_timer_global __iomem *timers_global;
+
+#define TIMER_IRQ  	(OBP_INT_LEVEL_ONBOARD | 10)
 
-static struct sun4m_timer_regs *sun4m_timers;
 unsigned int lvl14_resolution = (((1000000/HZ) + 1) << 10);
 
 static void sun4m_clear_clock_irq(void)
 {
-	volatile unsigned int clear_intr;
-	clear_intr = sun4m_timers->l10_timer_limit;
+	sbus_readl(&timers_global->l10_limit);
 }
 
-static void sun4m_clear_profile_irq(int cpu)
+void sun4m_nmi(struct pt_regs *regs)
 {
-	volatile unsigned int clear;
-    
-	clear = sun4m_timers->cpu_timers[cpu].l14_timer_limit;
+	unsigned long afsr, afar, si;
+
+	printk(KERN_ERR "Aieee: sun4m NMI received!\n");
+	/* XXX HyperSparc hack XXX */
+	__asm__ __volatile__("mov 0x500, %%g1\n\t"
+			     "lda [%%g1] 0x4, %0\n\t"
+			     "mov 0x600, %%g1\n\t"
+			     "lda [%%g1] 0x4, %1\n\t" :
+			     "=r" (afsr), "=r" (afar));
+	printk(KERN_ERR "afsr=%08lx afar=%08lx\n", afsr, afar);
+	si = sbus_readl(&sun4m_irq_global->pending);
+	printk(KERN_ERR "si=%08lx\n", si);
+	if (si & SUN4M_INT_MODULE_ERR)
+		printk(KERN_ERR "Module async error\n");
+	if (si & SUN4M_INT_M2S_WRITE_ERR)
+		printk(KERN_ERR "MBus/SBus async error\n");
+	if (si & SUN4M_INT_ECC_ERR)
+		printk(KERN_ERR "ECC memory error\n");
+	if (si & SUN4M_INT_VME_ERR)
+		printk(KERN_ERR "VME async error\n");
+	printk(KERN_ERR "you lose buddy boy...\n");
+	show_regs(regs);
+	prom_halt();
+}
+
+/* Exported for sun4m_smp.c */
+void sun4m_clear_profile_irq(int cpu)
+{
+	sbus_readl(&timers_percpu[cpu]->l14_limit);
 }
 
 static void sun4m_load_profile_irq(int cpu, unsigned int limit)
 {
-	sun4m_timers->cpu_timers[cpu].l14_timer_limit = limit;
+	sbus_writel(limit, &timers_percpu[cpu]->l14_limit);
 }
 
 static void __init sun4m_init_timers(irq_handler_t counter_fn)
 {
-	int reg_count, irq, cpu;
-	struct linux_prom_registers cnt_regs[PROMREG_MAX];
-	int obio_node, cnt_node;
-	struct resource r;
-
-	cnt_node = 0;
-	if((obio_node =
-	    prom_searchsiblings (prom_getchild(prom_root_node), "obio")) == 0 ||
-	   (obio_node = prom_getchild (obio_node)) == 0 ||
-	   (cnt_node = prom_searchsiblings (obio_node, "counter")) == 0) {
-		prom_printf("Cannot find /obio/counter node\n");
-		prom_halt();
+	struct device_node *dp = of_find_node_by_name(NULL, "counter");
+	int i, err, len, num_cpu_timers;
+	const u32 *addr;
+
+	if (!dp) {
+		printk(KERN_ERR "sun4m_init_timers: No 'counter' node.\n");
+		return;
 	}
-	reg_count = prom_getproperty(cnt_node, "reg",
-				     (void *) cnt_regs, sizeof(cnt_regs));
-	reg_count = (reg_count/sizeof(struct linux_prom_registers));
-    
-	/* Apply the obio ranges to the timer registers. */
-	prom_apply_obio_ranges(cnt_regs, reg_count);
-    
-	cnt_regs[4].phys_addr = cnt_regs[reg_count-1].phys_addr;
-	cnt_regs[4].reg_size = cnt_regs[reg_count-1].reg_size;
-	cnt_regs[4].which_io = cnt_regs[reg_count-1].which_io;
-	for(obio_node = 1; obio_node < 4; obio_node++) {
-		cnt_regs[obio_node].phys_addr =
-			cnt_regs[obio_node-1].phys_addr + PAGE_SIZE;
-		cnt_regs[obio_node].reg_size = cnt_regs[obio_node-1].reg_size;
-		cnt_regs[obio_node].which_io = cnt_regs[obio_node-1].which_io;
+
+	addr = of_get_property(dp, "address", &len);
+	if (!addr) {
+		printk(KERN_ERR "sun4m_init_timers: No 'address' prop.\n");
+		return;
 	}
 
-	memset((char*)&r, 0, sizeof(struct resource));
-	/* Map the per-cpu Counter registers. */
-	r.flags = cnt_regs[0].which_io;
-	r.start = cnt_regs[0].phys_addr;
-	sun4m_timers = (struct sun4m_timer_regs *) sbus_ioremap(&r, 0,
-	    PAGE_SIZE*SUN4M_NCPUS, "sun4m_cpu_cnt");
-	/* Map the system Counter register. */
-	/* XXX Here we expect consequent calls to yeld adjusent maps. */
-	r.flags = cnt_regs[4].which_io;
-	r.start = cnt_regs[4].phys_addr;
-	sbus_ioremap(&r, 0, cnt_regs[4].reg_size, "sun4m_sys_cnt");
-
-	sun4m_timers->l10_timer_limit =  (((1000000/HZ) + 1) << 10);
-	master_l10_counter = &sun4m_timers->l10_cur_count;
-	master_l10_limit = &sun4m_timers->l10_timer_limit;
-
-	irq = request_irq(TIMER_IRQ,
-			  counter_fn,
-			  (IRQF_DISABLED | SA_STATIC_ALLOC),
-			  "timer", NULL);
-	if (irq) {
-		prom_printf("time_init: unable to attach IRQ%d\n",TIMER_IRQ);
-		prom_halt();
+	num_cpu_timers = (len / sizeof(u32)) - 1;
+	for (i = 0; i < num_cpu_timers; i++) {
+		timers_percpu[i] = (void __iomem *)
+			(unsigned long) addr[i];
 	}
-   
-	if (!cpu_find_by_instance(1, NULL, NULL)) {
-		for(cpu = 0; cpu < 4; cpu++)
-			sun4m_timers->cpu_timers[cpu].l14_timer_limit = 0;
-		sun4m_interrupts->set = SUN4M_INT_E14;
-	} else {
-		sun4m_timers->cpu_timers[0].l14_timer_limit = 0;
+	timers_global = (void __iomem *)
+		(unsigned long) addr[num_cpu_timers];
+
+	sbus_writel((((1000000/HZ) + 1) << 10), &timers_global->l10_limit);
+
+	master_l10_counter = &timers_global->l10_count;
+
+	err = request_irq(TIMER_IRQ, counter_fn,
+			  (IRQF_DISABLED | SA_STATIC_ALLOC), "timer", NULL);
+	if (err) {
+		printk(KERN_ERR "sun4m_init_timers: Register IRQ error %d.\n",
+			err);
+		return;
 	}
+
+	for (i = 0; i < num_cpu_timers; i++)
+		sbus_writel(0, &timers_percpu[i]->l14_limit);
+	if (num_cpu_timers == 4)
+		sbus_writel(SUN4M_INT_E14, &sun4m_irq_global->mask_set);
+
 #ifdef CONFIG_SMP
 	{
 		unsigned long flags;
@@ -390,70 +427,43 @@ static void __init sun4m_init_timers(irq_handler_t counter_fn)
 
 void __init sun4m_init_IRQ(void)
 {
-	int ie_node,i;
-	struct linux_prom_registers int_regs[PROMREG_MAX];
-	int num_regs;
-	struct resource r;
-	int mid;
-    
-	local_irq_disable();
-	if((ie_node = prom_searchsiblings(prom_getchild(prom_root_node), "obio")) == 0 ||
-	   (ie_node = prom_getchild (ie_node)) == 0 ||
-	   (ie_node = prom_searchsiblings (ie_node, "interrupt")) == 0) {
-		prom_printf("Cannot find /obio/interrupt node\n");
-		prom_halt();
+	struct device_node *dp = of_find_node_by_name(NULL, "interrupt");
+	int len, i, mid, num_cpu_iregs;
+	const u32 *addr;
+
+	if (!dp) {
+		printk(KERN_ERR "sun4m_init_IRQ: No 'interrupt' node.\n");
+		return;
 	}
-	num_regs = prom_getproperty(ie_node, "reg", (char *) int_regs,
-				    sizeof(int_regs));
-	num_regs = (num_regs/sizeof(struct linux_prom_registers));
-    
-	/* Apply the obio ranges to these registers. */
-	prom_apply_obio_ranges(int_regs, num_regs);
-    
-	int_regs[4].phys_addr = int_regs[num_regs-1].phys_addr;
-	int_regs[4].reg_size = int_regs[num_regs-1].reg_size;
-	int_regs[4].which_io = int_regs[num_regs-1].which_io;
-	for(ie_node = 1; ie_node < 4; ie_node++) {
-		int_regs[ie_node].phys_addr = int_regs[ie_node-1].phys_addr + PAGE_SIZE;
-		int_regs[ie_node].reg_size = int_regs[ie_node-1].reg_size;
-		int_regs[ie_node].which_io = int_regs[ie_node-1].which_io;
+
+	addr = of_get_property(dp, "address", &len);
+	if (!addr) {
+		printk(KERN_ERR "sun4m_init_IRQ: No 'address' prop.\n");
+		return;
 	}
 
-	memset((char *)&r, 0, sizeof(struct resource));
-	/* Map the interrupt registers for all possible cpus. */
-	r.flags = int_regs[0].which_io;
-	r.start = int_regs[0].phys_addr;
-	sun4m_interrupts = (struct sun4m_intregs *) sbus_ioremap(&r, 0,
-	    PAGE_SIZE*SUN4M_NCPUS, "interrupts_percpu");
+	num_cpu_iregs = (len / sizeof(u32)) - 1;
+	for (i = 0; i < num_cpu_iregs; i++) {
+		sun4m_irq_percpu[i] = (void __iomem *)
+			(unsigned long) addr[i];
+	}
+	sun4m_irq_global = (void __iomem *)
+		(unsigned long) addr[num_cpu_iregs];
 
-	/* Map the system interrupt control registers. */
-	r.flags = int_regs[4].which_io;
-	r.start = int_regs[4].phys_addr;
-	sbus_ioremap(&r, 0, int_regs[4].reg_size, "interrupts_system");
+	local_irq_disable();
 
-	sun4m_interrupts->set = ~SUN4M_INT_MASKALL;
+	sbus_writel(~SUN4M_INT_MASKALL, &sun4m_irq_global->mask_set);
 	for (i = 0; !cpu_find_by_instance(i, NULL, &mid); i++)
-		sun4m_interrupts->cpu_intregs[mid].clear = ~0x17fff;
-
-	if (!cpu_find_by_instance(1, NULL, NULL)) {
-		/* system wide interrupts go to cpu 0, this should always
-		 * be safe because it is guaranteed to be fitted or OBP doesn't
-		 * come up
-		 *
-		 * Not sure, but writing here on SLAVIO systems may puke
-		 * so I don't do it unless there is more than 1 cpu.
-		 */
-		irq_rcvreg = (unsigned long *)
-				&sun4m_interrupts->undirected_target;
-		sun4m_interrupts->undirected_target = 0;
-	}
-	BTFIXUPSET_CALL(sbint_to_irq, sun4m_sbint_to_irq, BTFIXUPCALL_NORM);
+		sbus_writel(~0x17fff, &sun4m_irq_percpu[mid]->clear);
+
+	if (num_cpu_iregs == 4)
+		sbus_writel(0, &sun4m_irq_global->interrupt_target);
+
 	BTFIXUPSET_CALL(enable_irq, sun4m_enable_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(disable_irq, sun4m_disable_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(enable_pil_irq, sun4m_enable_pil_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(disable_pil_irq, sun4m_disable_pil_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(clear_clock_irq, sun4m_clear_clock_irq, BTFIXUPCALL_NORM);
-	BTFIXUPSET_CALL(clear_profile_irq, sun4m_clear_profile_irq, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(load_profile_irq, sun4m_load_profile_irq, BTFIXUPCALL_NORM);
 	sparc_init_timers = sun4m_init_timers;
 #ifdef CONFIG_SMP
@@ -461,5 +471,6 @@ void __init sun4m_init_IRQ(void)
 	BTFIXUPSET_CALL(clear_cpu_int, sun4m_clear_ipi, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(set_irq_udt, sun4m_set_udt, BTFIXUPCALL_NORM);
 #endif
+
 	/* Cannot enable interrupts until OBP ticker is disabled. */
 }
diff --git a/arch/sparc/kernel/sun4m_smp.c b/arch/sparc/kernel/sun4m_smp.c
index 9964890dc1db..5fc386d08c47 100644
--- a/arch/sparc/kernel/sun4m_smp.c
+++ b/arch/sparc/kernel/sun4m_smp.c
@@ -17,6 +17,7 @@
 #include <linux/swap.h>
 #include <linux/profile.h>
 #include <linux/delay.h>
+#include <linux/cpu.h>
 
 #include <asm/cacheflush.h>
 #include <asm/tlbflush.h>
@@ -315,6 +316,8 @@ void smp4m_cross_call_irq(void)
 	ccall_info.processors_out[i] = 1;
 }
 
+extern void sun4m_clear_profile_irq(int cpu);
+
 void smp4m_percpu_timer_interrupt(struct pt_regs *regs)
 {
 	struct pt_regs *old_regs;
@@ -322,7 +325,7 @@ void smp4m_percpu_timer_interrupt(struct pt_regs *regs)
 
 	old_regs = set_irq_regs(regs);
 
-	clear_profile_irq(cpu);
+	sun4m_clear_profile_irq(cpu);
 
 	profile_tick(CPU_PROFILING);
 
diff --git a/arch/sparc/kernel/sun4setup.c b/arch/sparc/kernel/sun4setup.c
deleted file mode 100644
index 229a52f55f16..000000000000
--- a/arch/sparc/kernel/sun4setup.c
+++ /dev/null
@@ -1,75 +0,0 @@
-/* sun4setup.c: Setup the hardware address of various items in the sun4
- * 		architecture. Called from idprom_init
- *
- * Copyright (C) 1998 Chris G. Davis (cdavis@cois.on.ca)
- */
-
-#include <asm/page.h>
-#include <asm/oplib.h>
-#include <asm/idprom.h>
-#include <asm/sun4paddr.h>
-#include <asm/machines.h>
-
-int sun4_memreg_physaddr;
-int sun4_ie_physaddr;
-int sun4_clock_physaddr;
-int sun4_timer_physaddr;
-int sun4_eth_physaddr;
-int sun4_si_physaddr;
-int sun4_bwtwo_physaddr;
-int sun4_zs0_physaddr;
-int sun4_zs1_physaddr;
-int sun4_dma_physaddr;
-int sun4_esp_physaddr;
-int sun4_ie_physaddr; 
-
-void __init sun4setup(void)
-{
-	printk("Sun4 Hardware Setup v1.0 18/May/98 Chris Davis (cdavis@cois.on.ca). ");
-	/*
-	  setup standard sun4 info
-	  */
-	sun4_ie_physaddr=SUN4_IE_PHYSADDR;
-
-	/*
-	  setup model specific info
-	  */
-	switch(idprom->id_machtype) {
-		case (SM_SUN4 | SM_4_260 ):
-			printk("Setup for a SUN4/260\n");
-			sun4_memreg_physaddr=SUN4_200_MEMREG_PHYSADDR;
-			sun4_clock_physaddr=SUN4_200_CLOCK_PHYSADDR;
-			sun4_timer_physaddr=SUN4_UNUSED_PHYSADDR;
-			sun4_eth_physaddr=SUN4_200_ETH_PHYSADDR;
-			sun4_si_physaddr=SUN4_200_SI_PHYSADDR;
-			sun4_bwtwo_physaddr=SUN4_200_BWTWO_PHYSADDR;
-			sun4_dma_physaddr=SUN4_UNUSED_PHYSADDR;
-			sun4_esp_physaddr=SUN4_UNUSED_PHYSADDR;
-			break;
-		case (SM_SUN4 | SM_4_330 ):
-			printk("Setup for a SUN4/330\n");
-			sun4_memreg_physaddr=SUN4_300_MEMREG_PHYSADDR;
-			sun4_clock_physaddr=SUN4_300_CLOCK_PHYSADDR;
-			sun4_timer_physaddr=SUN4_300_TIMER_PHYSADDR;
-			sun4_eth_physaddr=SUN4_300_ETH_PHYSADDR;
-			sun4_si_physaddr=SUN4_UNUSED_PHYSADDR;
-			sun4_bwtwo_physaddr=SUN4_300_BWTWO_PHYSADDR;
-			sun4_dma_physaddr=SUN4_300_DMA_PHYSADDR;
-			sun4_esp_physaddr=SUN4_300_ESP_PHYSADDR;
-			break;
-		case (SM_SUN4 | SM_4_470 ):
-			printk("Setup for a SUN4/470\n");
-			sun4_memreg_physaddr=SUN4_400_MEMREG_PHYSADDR;
-			sun4_clock_physaddr=SUN4_400_CLOCK_PHYSADDR;
-			sun4_timer_physaddr=SUN4_400_TIMER_PHYSADDR;
-			sun4_eth_physaddr=SUN4_400_ETH_PHYSADDR;
-			sun4_si_physaddr=SUN4_UNUSED_PHYSADDR;
-			sun4_bwtwo_physaddr=SUN4_400_BWTWO_PHYSADDR;
-			sun4_dma_physaddr=SUN4_400_DMA_PHYSADDR;
-			sun4_esp_physaddr=SUN4_400_ESP_PHYSADDR;
-			break;
-		default:
-			;
-	}
-}
-
diff --git a/arch/sparc/kernel/sys_sparc.c b/arch/sparc/kernel/sys_sparc.c
index 4d73421559c3..03035c852a43 100644
--- a/arch/sparc/kernel/sys_sparc.c
+++ b/arch/sparc/kernel/sys_sparc.c
@@ -53,7 +53,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi
 	/* See asm-sparc/uaccess.h */
 	if (len > TASK_SIZE - PAGE_SIZE)
 		return -ENOMEM;
-	if (ARCH_SUN4C_SUN4 && len > 0x20000000)
+	if (ARCH_SUN4C && len > 0x20000000)
 		return -ENOMEM;
 	if (!addr)
 		addr = TASK_UNMAPPED_BASE;
@@ -65,7 +65,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi
 
 	for (vmm = find_vma(current->mm, addr); ; vmm = vmm->vm_next) {
 		/* At this point:  (!vmm || addr < vmm->vm_end). */
-		if (ARCH_SUN4C_SUN4 && addr < 0xe0000000 && 0x20000000 - len < addr) {
+		if (ARCH_SUN4C && addr < 0xe0000000 && 0x20000000 - len < addr) {
 			addr = PAGE_OFFSET;
 			vmm = find_vma(current->mm, PAGE_OFFSET);
 		}
@@ -81,7 +81,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi
 
 asmlinkage unsigned long sparc_brk(unsigned long brk)
 {
-	if(ARCH_SUN4C_SUN4) {
+	if(ARCH_SUN4C) {
 		if ((brk & 0xe0000000) != (current->mm->brk & 0xe0000000))
 			return current->mm->brk;
 	}
@@ -221,7 +221,7 @@ out:
 
 int sparc_mmap_check(unsigned long addr, unsigned long len)
 {
-	if (ARCH_SUN4C_SUN4 &&
+	if (ARCH_SUN4C &&
 	    (len > 0x20000000 ||
 	     (addr < 0xe0000000 && addr + len > 0x20000000)))
 		return -EINVAL;
diff --git a/arch/sparc/kernel/tick14.c b/arch/sparc/kernel/tick14.c
index 707bfda86570..138bbf5f8724 100644
--- a/arch/sparc/kernel/tick14.c
+++ b/arch/sparc/kernel/tick14.c
@@ -1,31 +1,12 @@
 /* tick14.c
- * linux/arch/sparc/kernel/tick14.c
  *
  * Copyright (C) 1996 David Redman (djhr@tadpole.co.uk)
  *
  * This file handles the Sparc specific level14 ticker
  * This is really useful for profiling OBP uses it for keyboard
  * aborts and other stuff.
- *
- *
  */
-#include <linux/errno.h>
-#include <linux/sched.h>
 #include <linux/kernel.h>
-#include <linux/param.h>
-#include <linux/string.h>
-#include <linux/mm.h>
-#include <linux/timex.h>
-#include <linux/interrupt.h>
-
-#include <asm/oplib.h>
-#include <asm/timer.h>
-#include <asm/mostek.h>
-#include <asm/system.h>
-#include <asm/irq.h>
-#include <asm/io.h>
-
-#include "irq.h"
 
 extern unsigned long lvl14_save[5];
 static unsigned long *linux_lvl14 = NULL;
@@ -56,31 +37,3 @@ void install_obp_ticker(void)
 	linux_lvl14[2] =  obp_lvl14[2];
 	linux_lvl14[3] =  obp_lvl14[3]; 
 }
-
-void claim_ticker14(irq_handler_t handler,
-		    int irq_nr, unsigned int timeout )
-{
-	int cpu = smp_processor_id();
-
-	/* first we copy the obp handler instructions
-	 */
-	__disable_irq(irq_nr);
-	if (!handler)
-		return;
-    
-	linux_lvl14 = (unsigned long *)lvl14_save[4];
-	obp_lvl14[0] = linux_lvl14[0];
-	obp_lvl14[1] = linux_lvl14[1];
-	obp_lvl14[2] = linux_lvl14[2];
-	obp_lvl14[3] = linux_lvl14[3];
-
-	if (!request_irq(irq_nr,
-			 handler,
-			 (IRQF_DISABLED | SA_STATIC_ALLOC),
-			 "counter14",
-			 NULL)) {
-		install_linux_ticker();
-		load_profile_irq(cpu, timeout);
-		__enable_irq(irq_nr);
-	}
-}
diff --git a/arch/sparc/kernel/time.c b/arch/sparc/kernel/time.c
index 0762f5db1924..62c1d94cb434 100644
--- a/arch/sparc/kernel/time.c
+++ b/arch/sparc/kernel/time.c
@@ -23,22 +23,24 @@
 #include <linux/mm.h>
 #include <linux/interrupt.h>
 #include <linux/time.h>
+#include <linux/rtc.h>
+#include <linux/rtc/m48t59.h>
 #include <linux/timex.h>
 #include <linux/init.h>
 #include <linux/pci.h>
 #include <linux/ioport.h>
 #include <linux/profile.h>
+#include <linux/of.h>
 #include <linux/of_device.h>
+#include <linux/platform_device.h>
 
 #include <asm/oplib.h>
 #include <asm/timer.h>
-#include <asm/mostek.h>
 #include <asm/system.h>
 #include <asm/irq.h>
 #include <asm/io.h>
 #include <asm/idprom.h>
 #include <asm/machines.h>
-#include <asm/sun4paddr.h>
 #include <asm/page.h>
 #include <asm/pcic.h>
 #include <asm/irq_regs.h>
@@ -46,34 +48,9 @@
 #include "irq.h"
 
 DEFINE_SPINLOCK(rtc_lock);
-static enum sparc_clock_type sp_clock_typ;
-DEFINE_SPINLOCK(mostek_lock);
-void __iomem *mstk48t02_regs = NULL;
-static struct mostek48t08 __iomem *mstk48t08_regs = NULL;
 static int set_rtc_mmss(unsigned long);
 static int sbus_do_settimeofday(struct timespec *tv);
 
-#ifdef CONFIG_SUN4
-struct intersil *intersil_clock;
-#define intersil_cmd(intersil_reg, intsil_cmd) intersil_reg->int_cmd_reg = \
-	(intsil_cmd)
-
-#define intersil_intr(intersil_reg, intsil_cmd) intersil_reg->int_intr_reg = \
-	(intsil_cmd)
-
-#define intersil_start(intersil_reg) intersil_cmd(intersil_reg, \
-	( INTERSIL_START | INTERSIL_32K | INTERSIL_NORMAL | INTERSIL_24H |\
-	  INTERSIL_INTR_ENABLE))
-
-#define intersil_stop(intersil_reg) intersil_cmd(intersil_reg, \
-	( INTERSIL_STOP | INTERSIL_32K | INTERSIL_NORMAL | INTERSIL_24H |\
-	  INTERSIL_INTR_ENABLE))
-
-#define intersil_read_intr(intersil_reg, towhere) towhere = \
-	intersil_reg->int_intr_reg
-
-#endif
-
 unsigned long profile_pc(struct pt_regs *regs)
 {
 	extern char __copy_user_begin[], __copy_user_end[];
@@ -96,7 +73,6 @@ unsigned long profile_pc(struct pt_regs *regs)
 EXPORT_SYMBOL(profile_pc);
 
 __volatile__ unsigned int *master_l10_counter;
-__volatile__ unsigned int *master_l10_limit;
 
 /*
  * timer_interrupt() needs to keep up the real-time clock,
@@ -116,15 +92,7 @@ static irqreturn_t timer_interrupt(int dummy, void *dev_id)
 
 	/* Protect counter clear so that do_gettimeoffset works */
 	write_seqlock(&xtime_lock);
-#ifdef CONFIG_SUN4
-	if((idprom->id_machtype == (SM_SUN4 | SM_4_260)) ||
-	   (idprom->id_machtype == (SM_SUN4 | SM_4_110))) {
-		int temp;
-        	intersil_read_intr(intersil_clock, temp);
-		/* re-enable the irq */
-		enable_pil_irq(10);
-	}
-#endif
+
 	clear_clock_irq();
 
 	do_timer(1);
@@ -147,157 +115,56 @@ static irqreturn_t timer_interrupt(int dummy, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-/* Kick start a stopped clock (procedure from the Sun NVRAM/hostid FAQ). */
-static void __devinit kick_start_clock(void)
+static unsigned char mostek_read_byte(struct device *dev, u32 ofs)
 {
-	struct mostek48t02 *regs = (struct mostek48t02 *)mstk48t02_regs;
-	unsigned char sec;
-	int i, count;
-
-	prom_printf("CLOCK: Clock was stopped. Kick start ");
-
-	spin_lock_irq(&mostek_lock);
-
-	/* Turn on the kick start bit to start the oscillator. */
-	regs->creg |= MSTK_CREG_WRITE;
-	regs->sec &= ~MSTK_STOP;
-	regs->hour |= MSTK_KICK_START;
-	regs->creg &= ~MSTK_CREG_WRITE;
-
-	spin_unlock_irq(&mostek_lock);
-
-	/* Delay to allow the clock oscillator to start. */
-	sec = MSTK_REG_SEC(regs);
-	for (i = 0; i < 3; i++) {
-		while (sec == MSTK_REG_SEC(regs))
-			for (count = 0; count < 100000; count++)
-				/* nothing */ ;
-		prom_printf(".");
-		sec = regs->sec;
-	}
-	prom_printf("\n");
-
-	spin_lock_irq(&mostek_lock);
-
-	/* Turn off kick start and set a "valid" time and date. */
-	regs->creg |= MSTK_CREG_WRITE;
-	regs->hour &= ~MSTK_KICK_START;
-	MSTK_SET_REG_SEC(regs,0);
-	MSTK_SET_REG_MIN(regs,0);
-	MSTK_SET_REG_HOUR(regs,0);
-	MSTK_SET_REG_DOW(regs,5);
-	MSTK_SET_REG_DOM(regs,1);
-	MSTK_SET_REG_MONTH(regs,8);
-	MSTK_SET_REG_YEAR(regs,1996 - MSTK_YEAR_ZERO);
-	regs->creg &= ~MSTK_CREG_WRITE;
-
-	spin_unlock_irq(&mostek_lock);
-
-	/* Ensure the kick start bit is off. If it isn't, turn it off. */
-	while (regs->hour & MSTK_KICK_START) {
-		prom_printf("CLOCK: Kick start still on!\n");
-
-		spin_lock_irq(&mostek_lock);
-		regs->creg |= MSTK_CREG_WRITE;
-		regs->hour &= ~MSTK_KICK_START;
-		regs->creg &= ~MSTK_CREG_WRITE;
-		spin_unlock_irq(&mostek_lock);
+	struct platform_device *pdev = to_platform_device(dev);
+	struct m48t59_plat_data *pdata = pdev->dev.platform_data;
+	void __iomem *regs = pdata->ioaddr;
+	unsigned char val = readb(regs + ofs);
+
+	/* the year 0 is 1968 */
+	if (ofs == pdata->offset + M48T59_YEAR) {
+		val += 0x68;
+		if ((val & 0xf) > 9)
+			val += 6;
 	}
-
-	prom_printf("CLOCK: Kick start procedure successful.\n");
-}
-
-/* Return nonzero if the clock chip battery is low. */
-static inline int has_low_battery(void)
-{
-	struct mostek48t02 *regs = (struct mostek48t02 *)mstk48t02_regs;
-	unsigned char data1, data2;
-
-	spin_lock_irq(&mostek_lock);
-	data1 = regs->eeprom[0];	/* Read some data. */
-	regs->eeprom[0] = ~data1;	/* Write back the complement. */
-	data2 = regs->eeprom[0];	/* Read back the complement. */
-	regs->eeprom[0] = data1;	/* Restore the original value. */
-	spin_unlock_irq(&mostek_lock);
-
-	return (data1 == data2);	/* Was the write blocked? */
+	return val;
 }
 
-static void __devinit mostek_set_system_time(void)
+static void mostek_write_byte(struct device *dev, u32 ofs, u8 val)
 {
-	unsigned int year, mon, day, hour, min, sec;
-	struct mostek48t02 *mregs;
-
-	mregs = (struct mostek48t02 *)mstk48t02_regs;
-	if(!mregs) {
-		prom_printf("Something wrong, clock regs not mapped yet.\n");
-		prom_halt();
-	}		
-	spin_lock_irq(&mostek_lock);
-	mregs->creg |= MSTK_CREG_READ;
-	sec = MSTK_REG_SEC(mregs);
-	min = MSTK_REG_MIN(mregs);
-	hour = MSTK_REG_HOUR(mregs);
-	day = MSTK_REG_DOM(mregs);
-	mon = MSTK_REG_MONTH(mregs);
-	year = MSTK_CVT_YEAR( MSTK_REG_YEAR(mregs) );
-	xtime.tv_sec = mktime(year, mon, day, hour, min, sec);
-	xtime.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ);
-        set_normalized_timespec(&wall_to_monotonic,
-                                -xtime.tv_sec, -xtime.tv_nsec);
-	mregs->creg &= ~MSTK_CREG_READ;
-	spin_unlock_irq(&mostek_lock);
+	struct platform_device *pdev = to_platform_device(dev);
+	struct m48t59_plat_data *pdata = pdev->dev.platform_data;
+	void __iomem *regs = pdata->ioaddr;
+
+	if (ofs == pdata->offset + M48T59_YEAR) {
+		if (val < 0x68)
+			val += 0x32;
+		else
+			val -= 0x68;
+		if ((val & 0xf) > 9)
+			val += 6;
+		if ((val & 0xf0) > 0x9A)
+			val += 0x60;
+	}
+	writeb(val, regs + ofs);
 }
 
-/* Probe for the real time clock chip on Sun4 */
-static inline void sun4_clock_probe(void)
-{
-#ifdef CONFIG_SUN4
-	int temp;
-	struct resource r;
-
-	memset(&r, 0, sizeof(r));
-	if( idprom->id_machtype == (SM_SUN4 | SM_4_330) ) {
-		sp_clock_typ = MSTK48T02;
-		r.start = sun4_clock_physaddr;
-		mstk48t02_regs = sbus_ioremap(&r, 0,
-				       sizeof(struct mostek48t02), NULL);
-		mstk48t08_regs = NULL;  /* To catch weirdness */
-		intersil_clock = NULL;  /* just in case */
-
-		/* Kick start the clock if it is completely stopped. */
-		if (mostek_read(mstk48t02_regs + MOSTEK_SEC) & MSTK_STOP)
-			kick_start_clock();
-	} else if( idprom->id_machtype == (SM_SUN4 | SM_4_260)) {
-		/* intersil setup code */
-		printk("Clock: INTERSIL at %8x ",sun4_clock_physaddr);
-		sp_clock_typ = INTERSIL;
-		r.start = sun4_clock_physaddr;
-		intersil_clock = (struct intersil *) 
-		    sbus_ioremap(&r, 0, sizeof(*intersil_clock), "intersil");
-		mstk48t02_regs = 0;  /* just be sure */
-		mstk48t08_regs = NULL;  /* ditto */
-		/* initialise the clock */
-
-		intersil_intr(intersil_clock,INTERSIL_INT_100HZ);
-
-		intersil_start(intersil_clock);
-
-		intersil_read_intr(intersil_clock, temp);
-                while (!(temp & 0x80))
-                        intersil_read_intr(intersil_clock, temp);
-
-                intersil_read_intr(intersil_clock, temp);
-                while (!(temp & 0x80))
-                        intersil_read_intr(intersil_clock, temp);
-
-		intersil_stop(intersil_clock);
+static struct m48t59_plat_data m48t59_data = {
+	.read_byte = mostek_read_byte,
+	.write_byte = mostek_write_byte,
+};
 
-	}
-#endif
-}
+/* resource is set at runtime */
+static struct platform_device m48t59_rtc = {
+	.name		= "rtc-m48t59",
+	.id		= 0,
+	.num_resources	= 1,
+	.dev	= {
+		.platform_data = &m48t59_data,
+	},
+};
 
-#ifndef CONFIG_SUN4
 static int __devinit clock_probe(struct of_device *op, const struct of_device_id *match)
 {
 	struct device_node *dp = op->node;
@@ -306,38 +173,26 @@ static int __devinit clock_probe(struct of_device *op, const struct of_device_id
 	if (!model)
 		return -ENODEV;
 
+	m48t59_rtc.resource = &op->resource[0];
 	if (!strcmp(model, "mk48t02")) {
-		sp_clock_typ = MSTK48T02;
-
 		/* Map the clock register io area read-only */
-		mstk48t02_regs = of_ioremap(&op->resource[0], 0,
-					    sizeof(struct mostek48t02),
-					    "mk48t02");
-		mstk48t08_regs = NULL;  /* To catch weirdness */
+		m48t59_data.ioaddr = of_ioremap(&op->resource[0], 0,
+						2048, "rtc-m48t59");
+		m48t59_data.type = M48T59RTC_TYPE_M48T02;
 	} else if (!strcmp(model, "mk48t08")) {
-		sp_clock_typ = MSTK48T08;
-		mstk48t08_regs = of_ioremap(&op->resource[0], 0,
-					    sizeof(struct mostek48t08),
-					    "mk48t08");
-
-		mstk48t02_regs = &mstk48t08_regs->regs;
+		m48t59_data.ioaddr = of_ioremap(&op->resource[0], 0,
+						8192, "rtc-m48t59");
+		m48t59_data.type = M48T59RTC_TYPE_M48T08;
 	} else
 		return -ENODEV;
 
-	/* Report a low battery voltage condition. */
-	if (has_low_battery())
-		printk(KERN_CRIT "NVRAM: Low battery voltage!\n");
-
-	/* Kick start the clock if it is completely stopped. */
-	if (mostek_read(mstk48t02_regs + MOSTEK_SEC) & MSTK_STOP)
-		kick_start_clock();
-
-	mostek_set_system_time();
+	if (platform_device_register(&m48t59_rtc) < 0)
+		printk(KERN_ERR "Registering RTC device failed\n");
 
 	return 0;
 }
 
-static struct of_device_id clock_match[] = {
+static struct of_device_id __initdata clock_match[] = {
 	{
 		.name = "eeprom",
 	},
@@ -348,7 +203,7 @@ static struct of_platform_driver clock_driver = {
 	.match_table	= clock_match,
 	.probe		= clock_probe,
 	.driver		= {
-		.name	= "clock",
+		.name	= "rtc",
 	},
 };
 
@@ -364,7 +219,6 @@ static int __init clock_init(void)
  * need to see the clock registers.
  */
 fs_initcall(clock_init);
-#endif /* !CONFIG_SUN4 */
 
 static void __init sbus_time_init(void)
 {
@@ -372,51 +226,8 @@ static void __init sbus_time_init(void)
 	BTFIXUPSET_CALL(bus_do_settimeofday, sbus_do_settimeofday, BTFIXUPCALL_NORM);
 	btfixup();
 
-	if (ARCH_SUN4)
-		sun4_clock_probe();
-
 	sparc_init_timers(timer_interrupt);
 	
-#ifdef CONFIG_SUN4
-	if(idprom->id_machtype == (SM_SUN4 | SM_4_330)) {
-		mostek_set_system_time();
-	} else if(idprom->id_machtype == (SM_SUN4 | SM_4_260) ) {
-		/* initialise the intersil on sun4 */
-		unsigned int year, mon, day, hour, min, sec;
-		int temp;
-		struct intersil *iregs;
-
-		iregs=intersil_clock;
-		if(!iregs) {
-			prom_printf("Something wrong, clock regs not mapped yet.\n");
-			prom_halt();
-		}
-
-		intersil_intr(intersil_clock,INTERSIL_INT_100HZ);
-		disable_pil_irq(10);
-		intersil_stop(iregs);
-		intersil_read_intr(intersil_clock, temp);
-
-		temp = iregs->clk.int_csec;
-
-		sec = iregs->clk.int_sec;
-		min = iregs->clk.int_min;
-		hour = iregs->clk.int_hour;
-		day = iregs->clk.int_day;
-		mon = iregs->clk.int_month;
-		year = MSTK_CVT_YEAR(iregs->clk.int_year);
-
-		enable_pil_irq(10);
-		intersil_start(iregs);
-
-		xtime.tv_sec = mktime(year, mon, day, hour, min, sec);
-		xtime.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ);
-	        set_normalized_timespec(&wall_to_monotonic,
- 	                               -xtime.tv_sec, -xtime.tv_nsec);
-		printk("%u/%u/%u %u:%u:%u\n",day,mon,year,hour,min,sec);
-	}
-#endif
-
 	/* Now that OBP ticker has been silenced, it is safe to enable IRQ. */
 	local_irq_enable();
 }
@@ -522,80 +333,15 @@ static int sbus_do_settimeofday(struct timespec *tv)
 	return 0;
 }
 
-/*
- * BUG: This routine does not handle hour overflow properly; it just
- *      sets the minutes. Usually you won't notice until after reboot!
- */
-static int set_rtc_mmss(unsigned long nowtime)
+static int set_rtc_mmss(unsigned long secs)
 {
-	int real_seconds, real_minutes, mostek_minutes;
-	struct mostek48t02 *regs = (struct mostek48t02 *)mstk48t02_regs;
-	unsigned long flags;
-#ifdef CONFIG_SUN4
-	struct intersil *iregs = intersil_clock;
-	int temp;
-#endif
+	struct rtc_device *rtc = rtc_class_open("rtc0");
+	int err = -1;
 
-	/* Not having a register set can lead to trouble. */
-	if (!regs) {
-#ifdef CONFIG_SUN4
-		if(!iregs)
-		return -1;
-	 	else {
-			temp = iregs->clk.int_csec;
-
-			mostek_minutes = iregs->clk.int_min;
-
-			real_seconds = nowtime % 60;
-			real_minutes = nowtime / 60;
-			if (((abs(real_minutes - mostek_minutes) + 15)/30) & 1)
-				real_minutes += 30;	/* correct for half hour time zone */
-			real_minutes %= 60;
-
-			if (abs(real_minutes - mostek_minutes) < 30) {
-				intersil_stop(iregs);
-				iregs->clk.int_sec=real_seconds;
-				iregs->clk.int_min=real_minutes;
-				intersil_start(iregs);
-			} else {
-				printk(KERN_WARNING
-			       "set_rtc_mmss: can't update from %d to %d\n",
-				       mostek_minutes, real_minutes);
-				return -1;
-			}
-			
-			return 0;
-		}
-#endif
+	if (rtc) {
+		err = rtc_set_mmss(rtc, secs);
+		rtc_class_close(rtc);
 	}
 
-	spin_lock_irqsave(&mostek_lock, flags);
-	/* Read the current RTC minutes. */
-	regs->creg |= MSTK_CREG_READ;
-	mostek_minutes = MSTK_REG_MIN(regs);
-	regs->creg &= ~MSTK_CREG_READ;
-
-	/*
-	 * since we're only adjusting minutes and seconds,
-	 * don't interfere with hour overflow. This avoids
-	 * messing with unknown time zones but requires your
-	 * RTC not to be off by more than 15 minutes
-	 */
-	real_seconds = nowtime % 60;
-	real_minutes = nowtime / 60;
-	if (((abs(real_minutes - mostek_minutes) + 15)/30) & 1)
-		real_minutes += 30;	/* correct for half hour time zone */
-	real_minutes %= 60;
-
-	if (abs(real_minutes - mostek_minutes) < 30) {
-		regs->creg |= MSTK_CREG_WRITE;
-		MSTK_SET_REG_SEC(regs,real_seconds);
-		MSTK_SET_REG_MIN(regs,real_minutes);
-		regs->creg &= ~MSTK_CREG_WRITE;
-		spin_unlock_irqrestore(&mostek_lock, flags);
-		return 0;
-	} else {
-		spin_unlock_irqrestore(&mostek_lock, flags);
-		return -1;
-	}
+	return err;
 }
diff --git a/arch/sparc/kernel/traps.c b/arch/sparc/kernel/traps.c
index 5d45d5fd8c99..2b7d50659036 100644
--- a/arch/sparc/kernel/traps.c
+++ b/arch/sparc/kernel/traps.c
@@ -43,23 +43,6 @@ void syscall_trace_exit(struct pt_regs *regs)
 {
 }
 
-void sun4m_nmi(struct pt_regs *regs)
-{
-	unsigned long afsr, afar;
-
-	printk("Aieee: sun4m NMI received!\n");
-	/* XXX HyperSparc hack XXX */
-	__asm__ __volatile__("mov 0x500, %%g1\n\t"
-			     "lda [%%g1] 0x4, %0\n\t"
-			     "mov 0x600, %%g1\n\t"
-			     "lda [%%g1] 0x4, %1\n\t" :
-			     "=r" (afsr), "=r" (afar));
-	printk("afsr=%08lx afar=%08lx\n", afsr, afar);
-	printk("you lose buddy boy...\n");
-	show_regs(regs);
-	prom_halt();
-}
-
 void sun4d_nmi(struct pt_regs *regs)
 {
 	printk("Aieee: sun4d NMI received!\n");
diff --git a/arch/sparc/mm/Makefile b/arch/sparc/mm/Makefile
index 109c8b22cb38..ea88955d97ff 100644
--- a/arch/sparc/mm/Makefile
+++ b/arch/sparc/mm/Makefile
@@ -3,13 +3,8 @@
 
 EXTRA_AFLAGS := -ansi
 
-obj-y    := fault.o init.o loadmmu.o generic.o extable.o btfixup.o
-
-ifeq ($(CONFIG_SUN4),y)
-obj-y	 += nosrmmu.o
-else
-obj-y	 += srmmu.o iommu.o io-unit.o hypersparc.o viking.o tsunami.o swift.o
-endif
+obj-y	:= fault.o init.o loadmmu.o generic.o extable.o btfixup.o \
+	    srmmu.o iommu.o io-unit.o hypersparc.o viking.o tsunami.o swift.o
 
 ifdef CONFIG_HIGHMEM
 obj-y	+= highmem.o
diff --git a/arch/sparc/mm/btfixup.c b/arch/sparc/mm/btfixup.c
index a312d127d47a..5175ac2f4820 100644
--- a/arch/sparc/mm/btfixup.c
+++ b/arch/sparc/mm/btfixup.c
@@ -20,11 +20,7 @@
 
 extern char *srmmu_name;
 static char version[] __initdata = "Boot time fixup v1.6. 4/Mar/98 Jakub Jelinek (jj@ultra.linux.cz). Patching kernel for ";
-#ifdef CONFIG_SUN4
-static char str_sun4c[] __initdata = "sun4\n";
-#else
 static char str_sun4c[] __initdata = "sun4c\n";
-#endif
 static char str_srmmu[] __initdata = "srmmu[%s]/";
 static char str_iommu[] __initdata = "iommu\n";
 static char str_iounit[] __initdata = "io-unit\n";
@@ -86,7 +82,7 @@ void __init btfixup(void)
 	if (!visited) {
 		visited++;
 		printk(version);
-		if (ARCH_SUN4C_SUN4)
+		if (ARCH_SUN4C)
 			printk(str_sun4c);
 		else {
 			printk(str_srmmu, srmmu_name);
diff --git a/arch/sparc/mm/fault.c b/arch/sparc/mm/fault.c
index 3604c2e86709..a507e1174662 100644
--- a/arch/sparc/mm/fault.c
+++ b/arch/sparc/mm/fault.c
@@ -191,7 +191,7 @@ asmlinkage void do_sparc_fault(struct pt_regs *regs, int text_fault, int write,
 	 * only copy the information from the master page table,
 	 * nothing more.
 	 */
-	if (!ARCH_SUN4C_SUN4 && address >= TASK_SIZE)
+	if (!ARCH_SUN4C && address >= TASK_SIZE)
 		goto vmalloc_fault;
 
 	info.si_code = SEGV_MAPERR;
diff --git a/arch/sparc/mm/init.c b/arch/sparc/mm/init.c
index e103f1bb3777..677c1e187a23 100644
--- a/arch/sparc/mm/init.c
+++ b/arch/sparc/mm/init.c
@@ -23,6 +23,7 @@
 #include <linux/highmem.h>
 #include <linux/bootmem.h>
 #include <linux/pagemap.h>
+#include <linux/poison.h>
 
 #include <asm/system.h>
 #include <asm/vac-ops.h>
@@ -480,6 +481,7 @@ void free_initmem (void)
 	for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) {
 		struct page *p;
 
+		memset((void *)addr, POISON_FREE_INITMEM, PAGE_SIZE);
 		p = virt_to_page(addr);
 
 		ClearPageReserved(p);
@@ -488,20 +490,26 @@ void free_initmem (void)
 		totalram_pages++;
 		num_physpages++;
 	}
-	printk (KERN_INFO "Freeing unused kernel memory: %dk freed\n", (&__init_end - &__init_begin) >> 10);
+	printk(KERN_INFO "Freeing unused kernel memory: %dk freed\n",
+		(&__init_end - &__init_begin) >> 10);
 }
 
 #ifdef CONFIG_BLK_DEV_INITRD
 void free_initrd_mem(unsigned long start, unsigned long end)
 {
 	if (start < end)
-		printk (KERN_INFO "Freeing initrd memory: %ldk freed\n", (end - start) >> 10);
+		printk(KERN_INFO "Freeing initrd memory: %ldk freed\n",
+			(end - start) >> 10);
 	for (; start < end; start += PAGE_SIZE) {
-		struct page *p = virt_to_page(start);
+		struct page *p;
+
+		memset((void *)start, POISON_FREE_INITMEM, PAGE_SIZE);
+		p = virt_to_page(start);
 
 		ClearPageReserved(p);
 		init_page_count(p);
 		__free_page(p);
+		totalram_pages++;
 		num_physpages++;
 	}
 }
diff --git a/arch/sparc/mm/io-unit.c b/arch/sparc/mm/io-unit.c
index f167835db3df..daadf5f88050 100644
--- a/arch/sparc/mm/io-unit.c
+++ b/arch/sparc/mm/io-unit.c
@@ -12,10 +12,11 @@
 #include <linux/highmem.h>	/* pte_offset_map => kmap_atomic */
 #include <linux/bitops.h>
 #include <linux/scatterlist.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/pgalloc.h>
 #include <asm/pgtable.h>
-#include <asm/sbus.h>
 #include <asm/io.h>
 #include <asm/io-unit.h>
 #include <asm/mxcc.h>
@@ -34,13 +35,10 @@
 #define IOPERM        (IOUPTE_CACHE | IOUPTE_WRITE | IOUPTE_VALID)
 #define MKIOPTE(phys) __iopte((((phys)>>4) & IOUPTE_PAGE) | IOPERM)
 
-void __init
-iounit_init(int sbi_node, int io_node, struct sbus_bus *sbus)
+static void __init iounit_iommu_init(struct of_device *op)
 {
-	iopte_t *xpt, *xptend;
 	struct iounit_struct *iounit;
-	struct linux_prom_registers iommu_promregs[PROMREG_MAX];
-	struct resource r;
+	iopte_t *xpt, *xptend;
 
 	iounit = kzalloc(sizeof(struct iounit_struct), GFP_ATOMIC);
 	if (!iounit) {
@@ -55,18 +53,13 @@ iounit_init(int sbi_node, int io_node, struct sbus_bus *sbus)
 	iounit->rotor[1] = IOUNIT_BMAP2_START;
 	iounit->rotor[2] = IOUNIT_BMAPM_START;
 
-	xpt = NULL;
-	if(prom_getproperty(sbi_node, "reg", (void *) iommu_promregs,
-			    sizeof(iommu_promregs)) != -1) {
-		prom_apply_generic_ranges(io_node, 0, iommu_promregs, 3);
-		memset(&r, 0, sizeof(r));
-		r.flags = iommu_promregs[2].which_io;
-		r.start = iommu_promregs[2].phys_addr;
-		xpt = (iopte_t *) sbus_ioremap(&r, 0, PAGE_SIZE * 16, "XPT");
+	xpt = of_ioremap(&op->resource[2], 0, PAGE_SIZE * 16, "XPT");
+	if (!xpt) {
+		prom_printf("SUN4D: Cannot map External Page Table.");
+		prom_halt();
 	}
-	if(!xpt) panic("Cannot map External Page Table.");
 	
-	sbus->ofdev.dev.archdata.iommu = iounit;
+	op->dev.archdata.iommu = iounit;
 	iounit->page_table = xpt;
 	spin_lock_init(&iounit->lock);
 	
@@ -75,6 +68,25 @@ iounit_init(int sbi_node, int io_node, struct sbus_bus *sbus)
 	     	iopte_val(*xpt++) = 0;
 }
 
+static int __init iounit_init(void)
+{
+	extern void sun4d_init_sbi_irq(void);
+	struct device_node *dp;
+
+	for_each_node_by_name(dp, "sbi") {
+		struct of_device *op = of_find_device_by_node(dp);
+
+		iounit_iommu_init(op);
+		of_propagate_archdata(op);
+	}
+
+	sun4d_init_sbi_irq();
+
+	return 0;
+}
+
+subsys_initcall(iounit_init);
+
 /* One has to hold iounit->lock to call this */
 static unsigned long iounit_get_area(struct iounit_struct *iounit, unsigned long vaddr, int size)
 {
@@ -124,10 +136,10 @@ nexti:	scan = find_next_zero_bit(iounit->bmap, limit, scan);
 	return vaddr;
 }
 
-static __u32 iounit_get_scsi_one(char *vaddr, unsigned long len, struct sbus_bus *sbus)
+static __u32 iounit_get_scsi_one(struct device *dev, char *vaddr, unsigned long len)
 {
+	struct iounit_struct *iounit = dev->archdata.iommu;
 	unsigned long ret, flags;
-	struct iounit_struct *iounit = sbus->ofdev.dev.archdata.iommu;
 	
 	spin_lock_irqsave(&iounit->lock, flags);
 	ret = iounit_get_area(iounit, (unsigned long)vaddr, len);
@@ -135,10 +147,10 @@ static __u32 iounit_get_scsi_one(char *vaddr, unsigned long len, struct sbus_bus
 	return ret;
 }
 
-static void iounit_get_scsi_sgl(struct scatterlist *sg, int sz, struct sbus_bus *sbus)
+static void iounit_get_scsi_sgl(struct device *dev, struct scatterlist *sg, int sz)
 {
+	struct iounit_struct *iounit = dev->archdata.iommu;
 	unsigned long flags;
-	struct iounit_struct *iounit = sbus->ofdev.dev.archdata.iommu;
 
 	/* FIXME: Cache some resolved pages - often several sg entries are to the same page */
 	spin_lock_irqsave(&iounit->lock, flags);
@@ -151,10 +163,10 @@ static void iounit_get_scsi_sgl(struct scatterlist *sg, int sz, struct sbus_bus
 	spin_unlock_irqrestore(&iounit->lock, flags);
 }
 
-static void iounit_release_scsi_one(__u32 vaddr, unsigned long len, struct sbus_bus *sbus)
+static void iounit_release_scsi_one(struct device *dev, __u32 vaddr, unsigned long len)
 {
+	struct iounit_struct *iounit = dev->archdata.iommu;
 	unsigned long flags;
-	struct iounit_struct *iounit = sbus->ofdev.dev.archdata.iommu;
 	
 	spin_lock_irqsave(&iounit->lock, flags);
 	len = ((vaddr & ~PAGE_MASK) + len + (PAGE_SIZE-1)) >> PAGE_SHIFT;
@@ -165,11 +177,11 @@ static void iounit_release_scsi_one(__u32 vaddr, unsigned long len, struct sbus_
 	spin_unlock_irqrestore(&iounit->lock, flags);
 }
 
-static void iounit_release_scsi_sgl(struct scatterlist *sg, int sz, struct sbus_bus *sbus)
+static void iounit_release_scsi_sgl(struct device *dev, struct scatterlist *sg, int sz)
 {
+	struct iounit_struct *iounit = dev->archdata.iommu;
 	unsigned long flags;
 	unsigned long vaddr, len;
-	struct iounit_struct *iounit = sbus->ofdev.dev.archdata.iommu;
 
 	spin_lock_irqsave(&iounit->lock, flags);
 	while (sz != 0) {
@@ -185,12 +197,12 @@ static void iounit_release_scsi_sgl(struct scatterlist *sg, int sz, struct sbus_
 }
 
 #ifdef CONFIG_SBUS
-static int iounit_map_dma_area(dma_addr_t *pba, unsigned long va, __u32 addr, int len)
+static int iounit_map_dma_area(struct device *dev, dma_addr_t *pba, unsigned long va, __u32 addr, int len)
 {
+	struct iounit_struct *iounit = dev->archdata.iommu;
 	unsigned long page, end;
 	pgprot_t dvma_prot;
 	iopte_t *iopte;
-	struct sbus_bus *sbus;
 
 	*pba = addr;
 
@@ -212,12 +224,8 @@ static int iounit_map_dma_area(dma_addr_t *pba, unsigned long va, __u32 addr, in
 			
 			i = ((addr - IOUNIT_DMA_BASE) >> PAGE_SHIFT);
 
-			for_each_sbus(sbus) {
-				struct iounit_struct *iounit = sbus->ofdev.dev.archdata.iommu;
-
-				iopte = (iopte_t *)(iounit->page_table + i);
-				*iopte = MKIOPTE(__pa(page));
-			}
+			iopte = (iopte_t *)(iounit->page_table + i);
+			*iopte = MKIOPTE(__pa(page));
 		}
 		addr += PAGE_SIZE;
 		va += PAGE_SIZE;
@@ -228,23 +236,10 @@ static int iounit_map_dma_area(dma_addr_t *pba, unsigned long va, __u32 addr, in
 	return 0;
 }
 
-static void iounit_unmap_dma_area(unsigned long addr, int len)
+static void iounit_unmap_dma_area(struct device *dev, unsigned long addr, int len)
 {
 	/* XXX Somebody please fill this in */
 }
-
-/* XXX We do not pass sbus device here, bad. */
-static struct page *iounit_translate_dvma(unsigned long addr)
-{
-	struct sbus_bus *sbus = sbus_root;	/* They are all the same */
-	struct iounit_struct *iounit = sbus->ofdev.dev.archdata.iommu;
-	int i;
-	iopte_t *iopte;
-
-	i = ((addr - IOUNIT_DMA_BASE) >> PAGE_SHIFT);
-	iopte = (iopte_t *)(iounit->page_table + i);
-	return pfn_to_page(iopte_val(*iopte) >> (PAGE_SHIFT-4)); /* XXX sun4d guru, help */
-}
 #endif
 
 static char *iounit_lockarea(char *vaddr, unsigned long len)
@@ -271,54 +266,5 @@ void __init ld_mmu_iounit(void)
 #ifdef CONFIG_SBUS
 	BTFIXUPSET_CALL(mmu_map_dma_area, iounit_map_dma_area, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(mmu_unmap_dma_area, iounit_unmap_dma_area, BTFIXUPCALL_NORM);
-	BTFIXUPSET_CALL(mmu_translate_dvma, iounit_translate_dvma, BTFIXUPCALL_NORM);
 #endif
 }
-
-__u32 iounit_map_dma_init(struct sbus_bus *sbus, int size)
-{
-	int i, j, k, npages;
-	unsigned long rotor, scan, limit;
-	unsigned long flags;
-	__u32 ret;
-	struct iounit_struct *iounit = sbus->ofdev.dev.archdata.iommu;
-
-        npages = (size + (PAGE_SIZE-1)) >> PAGE_SHIFT;
-	i = 0x0213;
-	spin_lock_irqsave(&iounit->lock, flags);
-next:	j = (i & 15);
-	rotor = iounit->rotor[j - 1];
-	limit = iounit->limit[j];
-	scan = rotor;
-nexti:	scan = find_next_zero_bit(iounit->bmap, limit, scan);
-	if (scan + npages > limit) {
-		if (limit != rotor) {
-			limit = rotor;
-			scan = iounit->limit[j - 1];
-			goto nexti;
-		}
-		i >>= 4;
-		if (!(i & 15))
-			panic("iounit_map_dma_init: Couldn't find free iopte slots for %d bytes\n", size);
-		goto next;
-	}
-	for (k = 1, scan++; k < npages; k++)
-		if (test_bit(scan++, iounit->bmap))
-			goto nexti;
-	iounit->rotor[j - 1] = (scan < limit) ? scan : iounit->limit[j - 1];
-	scan -= npages;
-	ret = IOUNIT_DMA_BASE + (scan << PAGE_SHIFT);
-	for (k = 0; k < npages; k++, scan++)
-		set_bit(scan, iounit->bmap);
-	spin_unlock_irqrestore(&iounit->lock, flags);
-	return ret;
-}
-
-__u32 iounit_map_dma_page(__u32 vaddr, void *addr, struct sbus_bus *sbus)
-{
-	int scan = (vaddr - IOUNIT_DMA_BASE) >> PAGE_SHIFT;
-	struct iounit_struct *iounit = sbus->ofdev.dev.archdata.iommu;
-	
-	iounit->page_table[scan] = MKIOPTE(__pa(((unsigned long)addr) & PAGE_MASK));
-	return vaddr + (((unsigned long)addr) & ~PAGE_MASK);
-}
diff --git a/arch/sparc/mm/iommu.c b/arch/sparc/mm/iommu.c
index 4b934270f05e..e7a499e3aa3c 100644
--- a/arch/sparc/mm/iommu.c
+++ b/arch/sparc/mm/iommu.c
@@ -13,10 +13,11 @@
 #include <linux/slab.h>
 #include <linux/highmem.h>	/* pte_offset_map => kmap_atomic */
 #include <linux/scatterlist.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/pgalloc.h>
 #include <asm/pgtable.h>
-#include <asm/sbus.h>
 #include <asm/io.h>
 #include <asm/mxcc.h>
 #include <asm/mbus.h>
@@ -55,30 +56,21 @@ static pgprot_t dvma_prot;		/* Consistent mapping pte flags */
 #define IOPERM        (IOPTE_CACHE | IOPTE_WRITE | IOPTE_VALID)
 #define MKIOPTE(pfn, perm) (((((pfn)<<8) & IOPTE_PAGE) | (perm)) & ~IOPTE_WAZ)
 
-void __init
-iommu_init(int iommund, struct sbus_bus *sbus)
+static void __init sbus_iommu_init(struct of_device *op)
 {
-	unsigned int impl, vers;
-	unsigned long tmp;
 	struct iommu_struct *iommu;
-	struct linux_prom_registers iommu_promregs[PROMREG_MAX];
-	struct resource r;
+	unsigned int impl, vers;
 	unsigned long *bitmap;
+	unsigned long tmp;
 
 	iommu = kmalloc(sizeof(struct iommu_struct), GFP_ATOMIC);
 	if (!iommu) {
 		prom_printf("Unable to allocate iommu structure\n");
 		prom_halt();
 	}
-	iommu->regs = NULL;
-	if (prom_getproperty(iommund, "reg", (void *) iommu_promregs,
-			 sizeof(iommu_promregs)) != -1) {
-		memset(&r, 0, sizeof(r));
-		r.flags = iommu_promregs[0].which_io;
-		r.start = iommu_promregs[0].phys_addr;
-		iommu->regs = (struct iommu_regs *)
-			sbus_ioremap(&r, 0, PAGE_SIZE * 3, "iommu_regs");
-	}
+
+	iommu->regs = of_ioremap(&op->resource[0], 0, PAGE_SIZE * 3,
+				 "iommu_regs");
 	if (!iommu->regs) {
 		prom_printf("Cannot map IOMMU registers\n");
 		prom_halt();
@@ -128,13 +120,29 @@ iommu_init(int iommund, struct sbus_bus *sbus)
 	else
 		iommu->usemap.num_colors = 1;
 
-	printk("IOMMU: impl %d vers %d table 0x%p[%d B] map [%d b]\n",
-	    impl, vers, iommu->page_table,
-	    (int)(IOMMU_NPTES*sizeof(iopte_t)), (int)IOMMU_NPTES);
+	printk(KERN_INFO "IOMMU: impl %d vers %d table 0x%p[%d B] map [%d b]\n",
+	       impl, vers, iommu->page_table,
+	       (int)(IOMMU_NPTES*sizeof(iopte_t)), (int)IOMMU_NPTES);
+
+	op->dev.archdata.iommu = iommu;
+}
+
+static int __init iommu_init(void)
+{
+	struct device_node *dp;
+
+	for_each_node_by_name(dp, "iommu") {
+		struct of_device *op = of_find_device_by_node(dp);
+
+		sbus_iommu_init(op);
+		of_propagate_archdata(op);
+	}
 
-	sbus->ofdev.dev.archdata.iommu = iommu;
+	return 0;
 }
 
+subsys_initcall(iommu_init);
+
 /* This begs to be btfixup-ed by srmmu. */
 /* Flush the iotlb entries to ram. */
 /* This could be better if we didn't have to flush whole pages. */
@@ -164,9 +172,9 @@ static void iommu_flush_iotlb(iopte_t *iopte, unsigned int niopte)
 	}
 }
 
-static u32 iommu_get_one(struct page *page, int npages, struct sbus_bus *sbus)
+static u32 iommu_get_one(struct device *dev, struct page *page, int npages)
 {
-	struct iommu_struct *iommu = sbus->ofdev.dev.archdata.iommu;
+	struct iommu_struct *iommu = dev->archdata.iommu;
 	int ioptex;
 	iopte_t *iopte, *iopte0;
 	unsigned int busa, busa0;
@@ -194,8 +202,7 @@ static u32 iommu_get_one(struct page *page, int npages, struct sbus_bus *sbus)
 	return busa0;
 }
 
-static u32 iommu_get_scsi_one(char *vaddr, unsigned int len,
-    struct sbus_bus *sbus)
+static u32 iommu_get_scsi_one(struct device *dev, char *vaddr, unsigned int len)
 {
 	unsigned long off;
 	int npages;
@@ -205,22 +212,22 @@ static u32 iommu_get_scsi_one(char *vaddr, unsigned int len,
 	off = (unsigned long)vaddr & ~PAGE_MASK;
 	npages = (off + len + PAGE_SIZE-1) >> PAGE_SHIFT;
 	page = virt_to_page((unsigned long)vaddr & PAGE_MASK);
-	busa = iommu_get_one(page, npages, sbus);
+	busa = iommu_get_one(dev, page, npages);
 	return busa + off;
 }
 
-static __u32 iommu_get_scsi_one_noflush(char *vaddr, unsigned long len, struct sbus_bus *sbus)
+static __u32 iommu_get_scsi_one_noflush(struct device *dev, char *vaddr, unsigned long len)
 {
-	return iommu_get_scsi_one(vaddr, len, sbus);
+	return iommu_get_scsi_one(dev, vaddr, len);
 }
 
-static __u32 iommu_get_scsi_one_gflush(char *vaddr, unsigned long len, struct sbus_bus *sbus)
+static __u32 iommu_get_scsi_one_gflush(struct device *dev, char *vaddr, unsigned long len)
 {
 	flush_page_for_dma(0);
-	return iommu_get_scsi_one(vaddr, len, sbus);
+	return iommu_get_scsi_one(dev, vaddr, len);
 }
 
-static __u32 iommu_get_scsi_one_pflush(char *vaddr, unsigned long len, struct sbus_bus *sbus)
+static __u32 iommu_get_scsi_one_pflush(struct device *dev, char *vaddr, unsigned long len)
 {
 	unsigned long page = ((unsigned long) vaddr) & PAGE_MASK;
 
@@ -228,23 +235,23 @@ static __u32 iommu_get_scsi_one_pflush(char *vaddr, unsigned long len, struct sb
 		flush_page_for_dma(page);
 		page += PAGE_SIZE;
 	}
-	return iommu_get_scsi_one(vaddr, len, sbus);
+	return iommu_get_scsi_one(dev, vaddr, len);
 }
 
-static void iommu_get_scsi_sgl_noflush(struct scatterlist *sg, int sz, struct sbus_bus *sbus)
+static void iommu_get_scsi_sgl_noflush(struct device *dev, struct scatterlist *sg, int sz)
 {
 	int n;
 
 	while (sz != 0) {
 		--sz;
 		n = (sg->length + sg->offset + PAGE_SIZE-1) >> PAGE_SHIFT;
-		sg->dvma_address = iommu_get_one(sg_page(sg), n, sbus) + sg->offset;
+		sg->dvma_address = iommu_get_one(dev, sg_page(sg), n) + sg->offset;
 		sg->dvma_length = (__u32) sg->length;
 		sg = sg_next(sg);
 	}
 }
 
-static void iommu_get_scsi_sgl_gflush(struct scatterlist *sg, int sz, struct sbus_bus *sbus)
+static void iommu_get_scsi_sgl_gflush(struct device *dev, struct scatterlist *sg, int sz)
 {
 	int n;
 
@@ -252,13 +259,13 @@ static void iommu_get_scsi_sgl_gflush(struct scatterlist *sg, int sz, struct sbu
 	while (sz != 0) {
 		--sz;
 		n = (sg->length + sg->offset + PAGE_SIZE-1) >> PAGE_SHIFT;
-		sg->dvma_address = iommu_get_one(sg_page(sg), n, sbus) + sg->offset;
+		sg->dvma_address = iommu_get_one(dev, sg_page(sg), n) + sg->offset;
 		sg->dvma_length = (__u32) sg->length;
 		sg = sg_next(sg);
 	}
 }
 
-static void iommu_get_scsi_sgl_pflush(struct scatterlist *sg, int sz, struct sbus_bus *sbus)
+static void iommu_get_scsi_sgl_pflush(struct device *dev, struct scatterlist *sg, int sz)
 {
 	unsigned long page, oldpage = 0;
 	int n, i;
@@ -283,15 +290,15 @@ static void iommu_get_scsi_sgl_pflush(struct scatterlist *sg, int sz, struct sbu
 			}
 		}
 
-		sg->dvma_address = iommu_get_one(sg_page(sg), n, sbus) + sg->offset;
+		sg->dvma_address = iommu_get_one(dev, sg_page(sg), n) + sg->offset;
 		sg->dvma_length = (__u32) sg->length;
 		sg = sg_next(sg);
 	}
 }
 
-static void iommu_release_one(u32 busa, int npages, struct sbus_bus *sbus)
+static void iommu_release_one(struct device *dev, u32 busa, int npages)
 {
-	struct iommu_struct *iommu = sbus->ofdev.dev.archdata.iommu;
+	struct iommu_struct *iommu = dev->archdata.iommu;
 	int ioptex;
 	int i;
 
@@ -305,17 +312,17 @@ static void iommu_release_one(u32 busa, int npages, struct sbus_bus *sbus)
 	bit_map_clear(&iommu->usemap, ioptex, npages);
 }
 
-static void iommu_release_scsi_one(__u32 vaddr, unsigned long len, struct sbus_bus *sbus)
+static void iommu_release_scsi_one(struct device *dev, __u32 vaddr, unsigned long len)
 {
 	unsigned long off;
 	int npages;
 
 	off = vaddr & ~PAGE_MASK;
 	npages = (off + len + PAGE_SIZE-1) >> PAGE_SHIFT;
-	iommu_release_one(vaddr & PAGE_MASK, npages, sbus);
+	iommu_release_one(dev, vaddr & PAGE_MASK, npages);
 }
 
-static void iommu_release_scsi_sgl(struct scatterlist *sg, int sz, struct sbus_bus *sbus)
+static void iommu_release_scsi_sgl(struct device *dev, struct scatterlist *sg, int sz)
 {
 	int n;
 
@@ -323,18 +330,18 @@ static void iommu_release_scsi_sgl(struct scatterlist *sg, int sz, struct sbus_b
 		--sz;
 
 		n = (sg->length + sg->offset + PAGE_SIZE-1) >> PAGE_SHIFT;
-		iommu_release_one(sg->dvma_address & PAGE_MASK, n, sbus);
+		iommu_release_one(dev, sg->dvma_address & PAGE_MASK, n);
 		sg->dvma_address = 0x21212121;
 		sg = sg_next(sg);
 	}
 }
 
 #ifdef CONFIG_SBUS
-static int iommu_map_dma_area(dma_addr_t *pba, unsigned long va,
-    unsigned long addr, int len)
+static int iommu_map_dma_area(struct device *dev, dma_addr_t *pba, unsigned long va,
+			      unsigned long addr, int len)
 {
+	struct iommu_struct *iommu = dev->archdata.iommu;
 	unsigned long page, end;
-	struct iommu_struct *iommu = sbus_root->ofdev.dev.archdata.iommu;
 	iopte_t *iopte = iommu->page_table;
 	iopte_t *first;
 	int ioptex;
@@ -397,9 +404,9 @@ static int iommu_map_dma_area(dma_addr_t *pba, unsigned long va,
 	return 0;
 }
 
-static void iommu_unmap_dma_area(unsigned long busa, int len)
+static void iommu_unmap_dma_area(struct device *dev, unsigned long busa, int len)
 {
-	struct iommu_struct *iommu = sbus_root->ofdev.dev.archdata.iommu;
+	struct iommu_struct *iommu = dev->archdata.iommu;
 	iopte_t *iopte = iommu->page_table;
 	unsigned long end;
 	int ioptex = (busa - iommu->start) >> PAGE_SHIFT;
@@ -417,15 +424,6 @@ static void iommu_unmap_dma_area(unsigned long busa, int len)
 	iommu_invalidate(iommu->regs);
 	bit_map_clear(&iommu->usemap, ioptex, len >> PAGE_SHIFT);
 }
-
-static struct page *iommu_translate_dvma(unsigned long busa)
-{
-	struct iommu_struct *iommu = sbus_root->ofdev.dev.archdata.iommu;
-	iopte_t *iopte = iommu->page_table;
-
-	iopte += ((busa - iommu->start) >> PAGE_SHIFT);
-	return pfn_to_page((iopte_val(*iopte) & IOPTE_PAGE) >> (PAGE_SHIFT-4));
-}
 #endif
 
 static char *iommu_lockarea(char *vaddr, unsigned long len)
@@ -461,7 +459,6 @@ void __init ld_mmu_iommu(void)
 #ifdef CONFIG_SBUS
 	BTFIXUPSET_CALL(mmu_map_dma_area, iommu_map_dma_area, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(mmu_unmap_dma_area, iommu_unmap_dma_area, BTFIXUPCALL_NORM);
-	BTFIXUPSET_CALL(mmu_translate_dvma, iommu_translate_dvma, BTFIXUPCALL_NORM);
 #endif
 
 	if (viking_mxcc_present || srmmu_modtype == HyperSparc) {
diff --git a/arch/sparc/mm/nosrmmu.c b/arch/sparc/mm/nosrmmu.c
deleted file mode 100644
index 3701f70fc30a..000000000000
--- a/arch/sparc/mm/nosrmmu.c
+++ /dev/null
@@ -1,59 +0,0 @@
-/*
- * nosrmmu.c: This file is a bunch of dummies for sun4 compiles, 
- *         so that it does not need srmmu and avoid ifdefs.
- *
- * Copyright (C) 1998 Jakub Jelinek (jj@sunsite.mff.cuni.cz)
- */
-
-#include <linux/kernel.h>
-#include <linux/mm.h>
-#include <linux/init.h>
-#include <asm/mbus.h>
-#include <asm/sbus.h>
-
-static char shouldnothappen[] __initdata = "SUN4 kernel can only run on SUN4\n";
-
-enum mbus_module srmmu_modtype;
-void *srmmu_nocache_pool;
-
-int vac_cache_size = 0;
-
-static void __init should_not_happen(void)
-{
-	prom_printf(shouldnothappen);
-	prom_halt();
-}
-
-void __init srmmu_frob_mem_map(unsigned long start_mem)
-{
-	should_not_happen();
-}
-
-unsigned long __init srmmu_paging_init(unsigned long start_mem, unsigned long end_mem)
-{
-	should_not_happen();
-	return 0;
-}
-
-void __init ld_mmu_srmmu(void)
-{
-	should_not_happen();
-}
-
-void srmmu_mapioaddr(unsigned long physaddr, unsigned long virt_addr, int bus_type, int rdonly)
-{
-}
-
-void srmmu_unmapioaddr(unsigned long virt_addr)
-{
-}
-
-__u32 iounit_map_dma_init(struct sbus_bus *sbus, int size)
-{
-	return 0;
-}
-
-__u32 iounit_map_dma_page(__u32 vaddr, void *addr, struct sbus_bus *sbus)
-{
-	return 0;
-}
diff --git a/arch/sparc/mm/srmmu.c b/arch/sparc/mm/srmmu.c
index ee30462598fc..6a5d7cabc044 100644
--- a/arch/sparc/mm/srmmu.c
+++ b/arch/sparc/mm/srmmu.c
@@ -31,7 +31,6 @@
 #include <asm/mbus.h>
 #include <asm/cache.h>
 #include <asm/oplib.h>
-#include <asm/sbus.h>
 #include <asm/asi.h>
 #include <asm/msi.h>
 #include <asm/mmu_context.h>
diff --git a/arch/sparc/mm/sun4c.c b/arch/sparc/mm/sun4c.c
index d1782f6368be..fe65aeeb3947 100644
--- a/arch/sparc/mm/sun4c.c
+++ b/arch/sparc/mm/sun4c.c
@@ -31,7 +31,6 @@
 #include <asm/oplib.h>
 #include <asm/openprom.h>
 #include <asm/mmu_context.h>
-#include <asm/sun4paddr.h>
 #include <asm/highmem.h>
 #include <asm/btfixup.h>
 #include <asm/cacheflush.h>
@@ -52,15 +51,11 @@ extern int num_segmaps, num_contexts;
 
 extern unsigned long page_kernel;
 
-#ifdef CONFIG_SUN4
-#define SUN4C_VAC_SIZE sun4c_vacinfo.num_bytes
-#else
 /* That's it, we prom_halt() on sun4c if the cache size is something other than 65536.
  * So let's save some cycles and just use that everywhere except for that bootup
  * sanity check.
  */
 #define SUN4C_VAC_SIZE 65536
-#endif
 
 #define SUN4C_KERNEL_BUCKETS 32
 
@@ -285,75 +280,32 @@ void __init sun4c_probe_vac(void)
 {
 	sun4c_disable_vac();
 
-	if (ARCH_SUN4) {
-		switch (idprom->id_machtype) {
-
-		case (SM_SUN4|SM_4_110):
-			sun4c_vacinfo.type = VAC_NONE;
-			sun4c_vacinfo.num_bytes = 0;
-			sun4c_vacinfo.linesize = 0;
-			sun4c_vacinfo.do_hwflushes = 0;
-			prom_printf("No VAC. Get some bucks and buy a real computer.");
-			prom_halt();
-			break;
-
-		case (SM_SUN4|SM_4_260):
-			sun4c_vacinfo.type = VAC_WRITE_BACK;
-			sun4c_vacinfo.num_bytes = 128 * 1024;
-			sun4c_vacinfo.linesize = 16;
-			sun4c_vacinfo.do_hwflushes = 0;
-			break;
-
-		case (SM_SUN4|SM_4_330):
-			sun4c_vacinfo.type = VAC_WRITE_THROUGH;
-			sun4c_vacinfo.num_bytes = 128 * 1024;
-			sun4c_vacinfo.linesize = 16;
-			sun4c_vacinfo.do_hwflushes = 0;
-			break;
-
-		case (SM_SUN4|SM_4_470):
-			sun4c_vacinfo.type = VAC_WRITE_BACK;
-			sun4c_vacinfo.num_bytes = 128 * 1024;
-			sun4c_vacinfo.linesize = 32;
-			sun4c_vacinfo.do_hwflushes = 0;
-			break;
-
-		default:
-			prom_printf("Cannot initialize VAC - weird sun4 model idprom->id_machtype = %d", idprom->id_machtype);
-			prom_halt();
-		};
+	if ((idprom->id_machtype == (SM_SUN4C | SM_4C_SS1)) ||
+	    (idprom->id_machtype == (SM_SUN4C | SM_4C_SS1PLUS))) {
+		/* PROM on SS1 lacks this info, to be super safe we
+		 * hard code it here since this arch is cast in stone.
+		 */
+		sun4c_vacinfo.num_bytes = 65536;
+		sun4c_vacinfo.linesize = 16;
 	} else {
-		sun4c_vacinfo.type = VAC_WRITE_THROUGH;
+		sun4c_vacinfo.num_bytes =
+		 prom_getintdefault(prom_root_node, "vac-size", 65536);
+		sun4c_vacinfo.linesize =
+		 prom_getintdefault(prom_root_node, "vac-linesize", 16);
+	}
+	sun4c_vacinfo.do_hwflushes =
+	 prom_getintdefault(prom_root_node, "vac-hwflush", 0);
 
-		if ((idprom->id_machtype == (SM_SUN4C | SM_4C_SS1)) ||
-		    (idprom->id_machtype == (SM_SUN4C | SM_4C_SS1PLUS))) {
-			/* PROM on SS1 lacks this info, to be super safe we
-			 * hard code it here since this arch is cast in stone.
-			 */
-			sun4c_vacinfo.num_bytes = 65536;
-			sun4c_vacinfo.linesize = 16;
-		} else {
-			sun4c_vacinfo.num_bytes =
-			 prom_getintdefault(prom_root_node, "vac-size", 65536);
-			sun4c_vacinfo.linesize =
-			 prom_getintdefault(prom_root_node, "vac-linesize", 16);
-		}
+	if (sun4c_vacinfo.do_hwflushes == 0)
 		sun4c_vacinfo.do_hwflushes =
-		 prom_getintdefault(prom_root_node, "vac-hwflush", 0);
-
-		if (sun4c_vacinfo.do_hwflushes == 0)
-			sun4c_vacinfo.do_hwflushes =
-			 prom_getintdefault(prom_root_node, "vac_hwflush", 0);
+		 prom_getintdefault(prom_root_node, "vac_hwflush", 0);
 
-		if (sun4c_vacinfo.num_bytes != 65536) {
-			prom_printf("WEIRD Sun4C VAC cache size, "
-				    "tell sparclinux@vger.kernel.org");
-			prom_halt();
-		}
+	if (sun4c_vacinfo.num_bytes != 65536) {
+		prom_printf("WEIRD Sun4C VAC cache size, "
+			    "tell sparclinux@vger.kernel.org");
+		prom_halt();
 	}
 
-	sun4c_vacinfo.num_lines =
-		(sun4c_vacinfo.num_bytes / sun4c_vacinfo.linesize);
 	switch (sun4c_vacinfo.linesize) {
 	case 16:
 		sun4c_vacinfo.log2lsize = 4;
@@ -447,49 +399,18 @@ static void __init patch_kernel_fault_handler(void)
 
 static void __init sun4c_probe_mmu(void)
 {
-	if (ARCH_SUN4) {
-		switch (idprom->id_machtype) {
-		case (SM_SUN4|SM_4_110):
-			prom_printf("No support for 4100 yet\n");
-			prom_halt();
-			num_segmaps = 256;
-			num_contexts = 8;
-			break;
-
-		case (SM_SUN4|SM_4_260):
-			/* should be 512 segmaps. when it get fixed */
-			num_segmaps = 256;
-			num_contexts = 16;
-			break;
-
-		case (SM_SUN4|SM_4_330):
-			num_segmaps = 256;
-			num_contexts = 16;
-			break;
-
-		case (SM_SUN4|SM_4_470):
-			/* should be 1024 segmaps. when it get fixed */
-			num_segmaps = 256;
-			num_contexts = 64;
-			break;
-		default:
-			prom_printf("Invalid SUN4 model\n");
-			prom_halt();
-		};
+	if ((idprom->id_machtype == (SM_SUN4C | SM_4C_SS1)) ||
+	    (idprom->id_machtype == (SM_SUN4C | SM_4C_SS1PLUS))) {
+		/* Hardcode these just to be safe, PROM on SS1 does
+		* not have this info available in the root node.
+		*/
+		num_segmaps = 128;
+		num_contexts = 8;
 	} else {
-		if ((idprom->id_machtype == (SM_SUN4C | SM_4C_SS1)) ||
-		    (idprom->id_machtype == (SM_SUN4C | SM_4C_SS1PLUS))) {
-			/* Hardcode these just to be safe, PROM on SS1 does
-		 	* not have this info available in the root node.
-		 	*/
-			num_segmaps = 128;
-			num_contexts = 8;
-		} else {
-			num_segmaps =
-			    prom_getintdefault(prom_root_node, "mmu-npmg", 128);
-			num_contexts =
-			    prom_getintdefault(prom_root_node, "mmu-nctx", 0x8);
-		}
+		num_segmaps =
+		    prom_getintdefault(prom_root_node, "mmu-npmg", 128);
+		num_contexts =
+		    prom_getintdefault(prom_root_node, "mmu-nctx", 0x8);
 	}
 	patch_kernel_fault_handler();
 }
@@ -501,18 +422,14 @@ void __init sun4c_probe_memerr_reg(void)
 	int node;
 	struct linux_prom_registers regs[1];
 
-	if (ARCH_SUN4) {
-		sun4c_memerr_reg = ioremap(sun4_memreg_physaddr, PAGE_SIZE);
-	} else {
-		node = prom_getchild(prom_root_node);
-		node = prom_searchsiblings(prom_root_node, "memory-error");
-		if (!node)
-			return;
-		if (prom_getproperty(node, "reg", (char *)regs, sizeof(regs)) <= 0)
-			return;
-		/* hmm I think regs[0].which_io is zero here anyways */
-		sun4c_memerr_reg = ioremap(regs[0].phys_addr, regs[0].reg_size);
-	}
+	node = prom_getchild(prom_root_node);
+	node = prom_searchsiblings(prom_root_node, "memory-error");
+	if (!node)
+		return;
+	if (prom_getproperty(node, "reg", (char *)regs, sizeof(regs)) <= 0)
+		return;
+	/* hmm I think regs[0].which_io is zero here anyways */
+	sun4c_memerr_reg = ioremap(regs[0].phys_addr, regs[0].reg_size);
 }
 
 static inline void sun4c_init_ss2_cache_bug(void)
@@ -521,7 +438,6 @@ static inline void sun4c_init_ss2_cache_bug(void)
 
 	if ((idprom->id_machtype == (SM_SUN4C | SM_4C_SS2)) ||
 	    (idprom->id_machtype == (SM_SUN4C | SM_4C_IPX)) ||
-	    (idprom->id_machtype == (SM_SUN4 | SM_4_330)) ||
 	    (idprom->id_machtype == (SM_SUN4C | SM_4C_ELC))) {
 		/* Whee.. */
 		printk("SS2 cache bug detected, uncaching trap table page\n");
@@ -532,8 +448,8 @@ static inline void sun4c_init_ss2_cache_bug(void)
 }
 
 /* Addr is always aligned on a page boundary for us already. */
-static int sun4c_map_dma_area(dma_addr_t *pba, unsigned long va,
-    unsigned long addr, int len)
+static int sun4c_map_dma_area(struct device *dev, dma_addr_t *pba, unsigned long va,
+			      unsigned long addr, int len)
 {
 	unsigned long page, end;
 
@@ -555,14 +471,7 @@ static int sun4c_map_dma_area(dma_addr_t *pba, unsigned long va,
 	return 0;
 }
 
-static struct page *sun4c_translate_dvma(unsigned long busa)
-{
-	/* Fortunately for us, bus_addr == uncached_virt in sun4c. */
-	unsigned long pte = sun4c_get_pte(busa);
-	return pfn_to_page(pte & SUN4C_PFN_MASK);
-}
-
-static void sun4c_unmap_dma_area(unsigned long busa, int len)
+static void sun4c_unmap_dma_area(struct device *dev, unsigned long busa, int len)
 {
 	/* Fortunately for us, bus_addr == uncached_virt in sun4c. */
 	/* XXX Implement this */
@@ -624,11 +533,7 @@ static inline void sun4c_init_map_kernelprom(unsigned long kernel_end)
 {
 	unsigned long vaddr;
 	unsigned char pseg, ctx;
-#ifdef CONFIG_SUN4
-	/* sun4/110 and 260 have no kadb. */
-	if ((idprom->id_machtype != (SM_SUN4 | SM_4_260)) && 
-	    (idprom->id_machtype != (SM_SUN4 | SM_4_110))) {
-#endif
+
 	for (vaddr = KADB_DEBUGGER_BEGVM;
 	     vaddr < LINUX_OPPROM_ENDVM;
 	     vaddr += SUN4C_REAL_PGDIR_SIZE) {
@@ -640,9 +545,7 @@ static inline void sun4c_init_map_kernelprom(unsigned long kernel_end)
 			fix_permissions(vaddr, _SUN4C_PAGE_PRIV, 0);
 		}
 	}
-#ifdef CONFIG_SUN4
-	}
-#endif
+
 	for (vaddr = KERNBASE; vaddr < kernel_end; vaddr += SUN4C_REAL_PGDIR_SIZE) {
 		pseg = sun4c_get_segmap(vaddr);
 		mmu_entry_pool[pseg].locked = 1;
@@ -1048,14 +951,10 @@ static struct thread_info *sun4c_alloc_thread_info(void)
 	 * so we must flush the cache to guarantee consistency.
 	 */
 	sun4c_flush_page(pages);
-#ifndef CONFIG_SUN4	
 	sun4c_flush_page(pages + PAGE_SIZE);
-#endif
 
 	sun4c_put_pte(addr, BUCKET_PTE(pages));
-#ifndef CONFIG_SUN4	
 	sun4c_put_pte(addr + PAGE_SIZE, BUCKET_PTE(pages + PAGE_SIZE));
-#endif
 
 #ifdef CONFIG_DEBUG_STACK_USAGE
 	memset((void *)addr, 0, PAGE_SIZE << THREAD_INFO_ORDER);
@@ -1072,13 +971,11 @@ static void sun4c_free_thread_info(struct thread_info *ti)
 
 	/* We are deleting a mapping, so the flush here is mandatory. */
 	sun4c_flush_page(tiaddr);
-#ifndef CONFIG_SUN4	
 	sun4c_flush_page(tiaddr + PAGE_SIZE);
-#endif
+
 	sun4c_put_pte(tiaddr, 0);
-#ifndef CONFIG_SUN4	
 	sun4c_put_pte(tiaddr + PAGE_SIZE, 0);
-#endif
+
 	sun4c_bucket[entry] = BUCKET_EMPTY;
 	if (entry < sun4c_lowbucket_avail)
 		sun4c_lowbucket_avail = entry;
@@ -1211,7 +1108,7 @@ static void sun4c_unlockarea(char *vaddr, unsigned long size)
  * by implication and fool the page locking code above
  * if passed to by mistake.
  */
-static __u32 sun4c_get_scsi_one(char *bufptr, unsigned long len, struct sbus_bus *sbus)
+static __u32 sun4c_get_scsi_one(struct device *dev, char *bufptr, unsigned long len)
 {
 	unsigned long page;
 
@@ -1223,7 +1120,7 @@ static __u32 sun4c_get_scsi_one(char *bufptr, unsigned long len, struct sbus_bus
 	return (__u32)sun4c_lockarea(bufptr, len);
 }
 
-static void sun4c_get_scsi_sgl(struct scatterlist *sg, int sz, struct sbus_bus *sbus)
+static void sun4c_get_scsi_sgl(struct device *dev, struct scatterlist *sg, int sz)
 {
 	while (sz != 0) {
 		--sz;
@@ -1233,14 +1130,14 @@ static void sun4c_get_scsi_sgl(struct scatterlist *sg, int sz, struct sbus_bus *
 	}
 }
 
-static void sun4c_release_scsi_one(__u32 bufptr, unsigned long len, struct sbus_bus *sbus)
+static void sun4c_release_scsi_one(struct device *dev, __u32 bufptr, unsigned long len)
 {
 	if (bufptr < sun4c_iobuffer_start)
 		return; /* On kernel stack or similar, see above */
 	sun4c_unlockarea((char *)bufptr, len);
 }
 
-static void sun4c_release_scsi_sgl(struct scatterlist *sg, int sz, struct sbus_bus *sbus)
+static void sun4c_release_scsi_sgl(struct device *dev, struct scatterlist *sg, int sz)
 {
 	while (sz != 0) {
 		--sz;
@@ -2263,7 +2160,6 @@ void __init ld_mmu_sun4c(void)
 
 	BTFIXUPSET_CALL(mmu_map_dma_area, sun4c_map_dma_area, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(mmu_unmap_dma_area, sun4c_unmap_dma_area, BTFIXUPCALL_NORM);
-	BTFIXUPSET_CALL(mmu_translate_dvma, sun4c_translate_dvma, BTFIXUPCALL_NORM);
 
 	BTFIXUPSET_CALL(sparc_mapiorange, sun4c_mapiorange, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(sparc_unmapiorange, sun4c_unmapiorange, BTFIXUPCALL_NORM);
diff --git a/arch/sparc/prom/Makefile b/arch/sparc/prom/Makefile
index 7f5eacfcfbcf..8f7e18546c97 100644
--- a/arch/sparc/prom/Makefile
+++ b/arch/sparc/prom/Makefile
@@ -4,5 +4,3 @@
 
 lib-y := bootstr.o devmap.o devops.o init.o memory.o misc.o mp.o \
 	 palloc.o ranges.o segment.o console.o printf.o tree.o
-
-lib-$(CONFIG_SUN4) += sun4prom.o
diff --git a/arch/sparc/prom/bootstr.c b/arch/sparc/prom/bootstr.c
index 5a35c768ff7c..916831da7e67 100644
--- a/arch/sparc/prom/bootstr.c
+++ b/arch/sparc/prom/bootstr.c
@@ -6,15 +6,12 @@
 
 #include <linux/string.h>
 #include <asm/oplib.h>
-#include <asm/sun4prom.h>
 #include <linux/init.h>
 
 #define BARG_LEN  256
 static char barg_buf[BARG_LEN] = { 0 };
 static char fetched __initdata = 0;
 
-extern linux_sun4_romvec *sun4_romvec;
-
 char * __init
 prom_getbootargs(void)
 {
@@ -28,7 +25,6 @@ prom_getbootargs(void)
 
 	switch(prom_vers) {
 	case PROM_V0:
-	case PROM_SUN4:
 		cp = barg_buf;
 		/* Start from 1 and go over fd(0,0,0)kernel */
 		for(iter = 1; iter < 8; iter++) {
diff --git a/arch/sparc/prom/console.c b/arch/sparc/prom/console.c
index 790057a34616..b3075d73fc19 100644
--- a/arch/sparc/prom/console.c
+++ b/arch/sparc/prom/console.c
@@ -10,7 +10,6 @@
 #include <linux/kernel.h>
 #include <linux/sched.h>
 #include <asm/openprom.h>
-#include <asm/sun4prom.h>
 #include <asm/oplib.h>
 #include <asm/system.h>
 #include <linux/string.h>
@@ -30,7 +29,6 @@ prom_nbgetchar(void)
 	spin_lock_irqsave(&prom_lock, flags);
 	switch(prom_vers) {
 	case PROM_V0:
-	case PROM_SUN4:
 		i = (*(romvec->pv_nbgetchar))();
 		break;
 	case PROM_V2:
@@ -63,7 +61,6 @@ prom_nbputchar(char c)
 	spin_lock_irqsave(&prom_lock, flags);
 	switch(prom_vers) {
 	case PROM_V0:
-	case PROM_SUN4:
 		i = (*(romvec->pv_nbputchar))(c);
 		break;
 	case PROM_V2:
diff --git a/arch/sparc/prom/init.c b/arch/sparc/prom/init.c
index 729f87066945..873217c6d823 100644
--- a/arch/sparc/prom/init.c
+++ b/arch/sparc/prom/init.c
@@ -11,12 +11,10 @@
 
 #include <asm/openprom.h>
 #include <asm/oplib.h>
-#include <asm/sun4prom.h>
 
 struct linux_romvec *romvec;
 enum prom_major_version prom_vers;
 unsigned int prom_rev, prom_prev;
-linux_sun4_romvec *sun4_romvec;
 
 /* The root node of the prom device tree. */
 int prom_root_node;
@@ -34,10 +32,6 @@ extern void prom_ranges_init(void);
 
 void __init prom_init(struct linux_romvec *rp)
 {
-#ifdef CONFIG_SUN4
-	extern struct linux_romvec *sun4_prom_init(void);
-	rp = sun4_prom_init();
-#endif
 	romvec = rp;
 
 	switch(romvec->pv_romvers) {
@@ -50,9 +44,6 @@ void __init prom_init(struct linux_romvec *rp)
 	case 3:
 		prom_vers = PROM_V3;
 		break;
-	case 40:
-		prom_vers = PROM_SUN4;
-		break;
 	default:
 		prom_printf("PROMLIB: Bad PROM version %d\n",
 			    romvec->pv_romvers);
@@ -76,11 +67,8 @@ void __init prom_init(struct linux_romvec *rp)
 
 	prom_ranges_init();
 
-#ifndef CONFIG_SUN4
-	/* SUN4 prints this in sun4_prom_init */
 	printk("PROMLIB: Sun Boot Prom Version %d Revision %d\n",
 	       romvec->pv_romvers, prom_rev);
-#endif
 
 	/* Initialization successful. */
 	return;
diff --git a/arch/sparc/prom/memory.c b/arch/sparc/prom/memory.c
index 947f047dc95a..fac7899a29c3 100644
--- a/arch/sparc/prom/memory.c
+++ b/arch/sparc/prom/memory.c
@@ -10,7 +10,6 @@
 #include <linux/init.h>
 
 #include <asm/openprom.h>
-#include <asm/sun4prom.h>
 #include <asm/oplib.h>
 #include <asm/page.h>
 
@@ -46,15 +45,6 @@ static int __init prom_meminit_v2(void)
 	return num_ents;
 }
 
-static int __init prom_meminit_sun4(void)
-{
-#ifdef CONFIG_SUN4
-	sp_banks[0].base_addr = 0;
-	sp_banks[0].num_bytes = *(sun4_romvec->memoryavail);
-#endif
-	return 1;
-}
-
 static int sp_banks_cmp(const void *a, const void *b)
 {
 	const struct sparc_phys_banks *x = a, *y = b;
@@ -81,10 +71,6 @@ void __init prom_meminit(void)
 		num_ents = prom_meminit_v2();
 		break;
 
-	case PROM_SUN4:
-		num_ents = prom_meminit_sun4();
-		break;
-
 	default:
 		break;
 	}
diff --git a/arch/sparc/prom/ranges.c b/arch/sparc/prom/ranges.c
index f9b7def35f6e..64579a376419 100644
--- a/arch/sparc/prom/ranges.c
+++ b/arch/sparc/prom/ranges.c
@@ -9,7 +9,6 @@
 #include <asm/openprom.h>
 #include <asm/oplib.h>
 #include <asm/types.h>
-#include <asm/sbus.h>
 #include <asm/system.h>
 
 struct linux_prom_ranges promlib_obio_ranges[PROMREG_MAX];
diff --git a/arch/sparc/prom/sun4prom.c b/arch/sparc/prom/sun4prom.c
deleted file mode 100644
index 00390a2652aa..000000000000
--- a/arch/sparc/prom/sun4prom.c
+++ /dev/null
@@ -1,161 +0,0 @@
-/*
- * Copyright (C) 1996 The Australian National University.
- * Copyright (C) 1996 Fujitsu Laboratories Limited
- * Copyright (C) 1997 Michael A. Griffith (grif@acm.org)
- * Copyright (C) 1997 Sun Weenie (ko@ko.reno.nv.us)
- * Copyright (C) 1998 Jakub Jelinek (jj@sunsite.mff.cuni.cz)
- * 
- * This software may be distributed under the terms of the Gnu
- * Public License version 2 or later
- *
- * fake a really simple Sun prom for the SUN4
- */
-
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <asm/oplib.h>
-#include <asm/idprom.h> 
-#include <asm/machines.h> 
-#include <asm/sun4prom.h>
-#include <asm/asi.h>
-#include <asm/contregs.h>
-#include <linux/init.h>
-
-static struct linux_romvec sun4romvec;
-static struct idprom sun4_idprom;
-
-struct property {
-	char *name;
-	char *value;
-	int length;
-};
-
-struct node {
-	int level;
-	struct property *properties;
-};
-
-struct property null_properties = { NULL, NULL, -1 };
-
-struct property root_properties[] = {
-	{"device_type", "cpu", 4},
-	{"idprom", (char *)&sun4_idprom, sizeof(struct idprom)},
-	{NULL, NULL, -1}
-};
-
-struct node nodes[] = {
-	{ 0, &null_properties }, 
-	{ 0, root_properties },
-	{ -1,&null_properties }
-};
-
-
-static int no_nextnode(int node)
-{
-	if (nodes[node].level == nodes[node+1].level)
-		return node+1;
-	return -1;
-}
-
-static int no_child(int node)
-{
-	if (nodes[node].level == nodes[node+1].level-1)
-		return node+1;
-	return -1;
-}
-
-static struct property *find_property(int node,char *name)
-{
-	struct property *prop = &nodes[node].properties[0];
-	while (prop && prop->name) {
-		if (strcmp(prop->name,name) == 0) return prop;
-		prop++;
-	}
-	return NULL;
-}
-
-static int no_proplen(int node,char *name)
-{
-	struct property *prop = find_property(node,name);
-	if (prop) return prop->length;
-	return -1;
-}
-
-static int no_getprop(int node,char *name,char *value)
-{
-	struct property *prop = find_property(node,name);
-	if (prop) {
-		memcpy(value,prop->value,prop->length);
-		return 1;
-	}
-	return -1;
-}
-
-static int no_setprop(int node,char *name,char *value,int len)
-{
-	return -1;
-}
-
-static char *no_nextprop(int node,char *name)
-{
-	struct property *prop = find_property(node,name);
-	if (prop) return prop[1].name;
-	return NULL;
-}
-
-static struct linux_nodeops sun4_nodeops = {
-	no_nextnode,
-	no_child,
-	no_proplen,
-	no_getprop,
-	no_setprop,
-	no_nextprop
-};
-	
-static int synch_hook;
-
-struct linux_romvec * __init sun4_prom_init(void)
-{
-	int i;
-	unsigned char x;
-	char *p;
-                                
-	p = (char *)&sun4_idprom;
-	for (i = 0; i < sizeof(sun4_idprom); i++) {
-		__asm__ __volatile__ ("lduba [%1] %2, %0" : "=r" (x) :
-				      "r" (AC_IDPROM + i), "i" (ASI_CONTROL));
-		*p++ = x;
-	}
-
-	memset(&sun4romvec,0,sizeof(sun4romvec));
-
-	sun4_romvec = (linux_sun4_romvec *) SUN4_PROM_VECTOR;
-
-	sun4romvec.pv_romvers = 40;
-	sun4romvec.pv_nodeops = &sun4_nodeops;
-	sun4romvec.pv_reboot = sun4_romvec->reboot;
-	sun4romvec.pv_abort = sun4_romvec->abortentry;
-	sun4romvec.pv_halt = sun4_romvec->exittomon;
-	sun4romvec.pv_synchook = (void (**)(void))&synch_hook;
-	sun4romvec.pv_setctxt = sun4_romvec->setcxsegmap;
-	sun4romvec.pv_v0bootargs = sun4_romvec->bootParam;
-	sun4romvec.pv_nbgetchar = sun4_romvec->mayget;
-	sun4romvec.pv_nbputchar = sun4_romvec->mayput;
-	sun4romvec.pv_stdin = sun4_romvec->insource;
-	sun4romvec.pv_stdout = sun4_romvec->outsink;
-	
-	/*
-	 * We turn on the LEDs to let folks without monitors or
-	 * terminals know we booted.   Nothing too fancy now.  They
-	 * are all on, except for LED 5, which blinks.   When we
-	 * have more time, we can teach the penguin to say "By your
-	 * command" or "Activating turbo boost, Michael". :-)
-	 */
-	sun4_romvec->setLEDs(NULL);
-	
-	printk("PROMLIB: Old Sun4 boot PROM monitor %s, romvec version %d\n",
-		sun4_romvec->monid,
-		sun4_romvec->romvecversion);
-
-	return &sun4romvec;
-}
diff --git a/arch/sparc64/Kconfig b/arch/sparc64/Kconfig
index 36b4b7ab9cfb..5446e2a499b1 100644
--- a/arch/sparc64/Kconfig
+++ b/arch/sparc64/Kconfig
@@ -18,6 +18,13 @@ config SPARC64
 	select HAVE_ARCH_KGDB
 	select USE_GENERIC_SMP_HELPERS if SMP
 	select HAVE_ARCH_TRACEHOOK
+	select ARCH_WANT_OPTIONAL_GPIOLIB
+	select RTC_CLASS
+	select RTC_DRV_M48T59
+	select RTC_DRV_CMOS
+	select RTC_DRV_BQ4802
+	select RTC_DRV_SUN4V
+	select RTC_DRV_STARFIRE
 
 config GENERIC_TIME
 	bool
@@ -31,6 +38,11 @@ config GENERIC_CLOCKEVENTS
 	bool
 	default y
 
+config GENERIC_GPIO
+	bool
+	help
+	  Generic GPIO API support
+
 config 64BIT
 	def_bool y
 
@@ -185,6 +197,17 @@ config US2E_FREQ
 
 	  If in doubt, say N.
 
+config US3_MC
+	tristate "UltraSPARC-III Memory Controller driver"
+	default y
+	help
+	  This adds a driver for the UltraSPARC-III memory controller.
+	  Loading this driver allows exact mnemonic strings to be
+	  printed in the event of a memory error, so that the faulty DIMM
+	  on the motherboard can be matched to the error.
+
+	  If in doubt, say Y, as this information can be very useful.
+
 # Global things across all Sun machines.
 config GENERIC_LOCKBREAK
 	bool
diff --git a/arch/sparc64/Makefile b/arch/sparc64/Makefile
index b785a395b12f..c7214abc0d84 100644
--- a/arch/sparc64/Makefile
+++ b/arch/sparc64/Makefile
@@ -7,7 +7,7 @@
 # Copyright (C) 1998 Jakub Jelinek (jj@ultra.linux.cz)
 #
 
-CHECKFLAGS	+= -D__sparc__ -D__sparc_v9__ -m64
+CHECKFLAGS	+= -D__sparc__ -D__sparc_v9__ -D__arch64__ -m64
 
 # Undefine sparc when processing vmlinux.lds - it is used
 # And teach CPP we are doing 64 bit builds (for this case)
diff --git a/arch/sparc64/kernel/Makefile b/arch/sparc64/kernel/Makefile
index 418b5782096e..c0b8009ab196 100644
--- a/arch/sparc64/kernel/Makefile
+++ b/arch/sparc64/kernel/Makefile
@@ -7,16 +7,16 @@ EXTRA_CFLAGS := -Werror
 
 extra-y		:= head.o init_task.o vmlinux.lds
 
-obj-y		:= process.o setup.o cpu.o idprom.o \
+obj-y		:= process.o setup.o cpu.o idprom.o reboot.o \
 		   traps.o auxio.o una_asm.o sysfs.o iommu.o \
 		   irq.o ptrace.o time.o sys_sparc.o signal.o \
-		   unaligned.o central.o pci.o starfire.o \
-		   power.o sbus.o sparc64_ksyms.o chmc.o \
+		   unaligned.o central.o starfire.o \
+		   power.o sbus.o sparc64_ksyms.o ebus.o \
 		   visemul.o prom.o of_device.o hvapi.o sstate.o mdesc.o
 
 obj-$(CONFIG_DYNAMIC_FTRACE) += ftrace.o
 obj-$(CONFIG_STACKTRACE) += stacktrace.o
-obj-$(CONFIG_PCI)	 += ebus.o pci_common.o \
+obj-$(CONFIG_PCI)	 += pci.o pci_common.o psycho_common.o \
 			    pci_psycho.o pci_sabre.o pci_schizo.o \
 			    pci_sun4v.o pci_sun4v_asm.o pci_fire.o
 obj-$(CONFIG_PCI_MSI)	+= pci_msi.o
@@ -25,6 +25,7 @@ obj-$(CONFIG_COMPAT) += sys32.o sys_sparc32.o signal32.o
 obj-$(CONFIG_MODULES) += module.o
 obj-$(CONFIG_US3_FREQ) += us3_cpufreq.o
 obj-$(CONFIG_US2E_FREQ) += us2e_cpufreq.o
+obj-$(CONFIG_US3_MC) += chmc.o
 obj-$(CONFIG_KPROBES) += kprobes.o
 obj-$(CONFIG_SUN_LDOMS) += ldc.o vio.o viohs.o ds.o
 obj-$(CONFIG_AUDIT) += audit.o
diff --git a/arch/sparc64/kernel/auxio.c b/arch/sparc64/kernel/auxio.c
index dd5c7bf87619..858beda86524 100644
--- a/arch/sparc64/kernel/auxio.c
+++ b/arch/sparc64/kernel/auxio.c
@@ -109,7 +109,7 @@ void auxio_set_lte(int on)
 	}
 }
 
-static struct of_device_id auxio_match[] = {
+static struct of_device_id __initdata auxio_match[] = {
 	{
 		.name = "auxio",
 	},
diff --git a/arch/sparc64/kernel/central.c b/arch/sparc64/kernel/central.c
index f2e87d0d7e1d..05f1c916db06 100644
--- a/arch/sparc64/kernel/central.c
+++ b/arch/sparc64/kernel/central.c
@@ -1,461 +1,268 @@
 /* central.c: Central FHC driver for Sunfire/Starfire/Wildfire.
  *
- * Copyright (C) 1997, 1999 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 1997, 1999, 2008 David S. Miller (davem@davemloft.net)
  */
 
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/string.h>
-#include <linux/timer.h>
-#include <linux/sched.h>
-#include <linux/delay.h>
 #include <linux/init.h>
-#include <linux/bootmem.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
 
-#include <asm/page.h>
 #include <asm/fhc.h>
-#include <asm/starfire.h>
-
-static struct linux_central *central_bus = NULL;
-static struct linux_fhc *fhc_list = NULL;
+#include <asm/upa.h>
+
+struct clock_board {
+	void __iomem		*clock_freq_regs;
+	void __iomem		*clock_regs;
+	void __iomem		*clock_ver_reg;
+	int			num_slots;
+	struct resource		leds_resource;
+	struct platform_device	leds_pdev;
+};
+
+struct fhc {
+	void __iomem		*pregs;
+	bool			central;
+	bool			jtag_master;
+	int			board_num;
+	struct resource		leds_resource;
+	struct platform_device	leds_pdev;
+};
+
+static int __devinit clock_board_calc_nslots(struct clock_board *p)
+{
+	u8 reg = upa_readb(p->clock_regs + CLOCK_STAT1) & 0xc0;
 
-#define IS_CENTRAL_FHC(__fhc)	((__fhc) == central_bus->child)
+	switch (reg) {
+	case 0x40:
+		return 16;
 
-static void central_probe_failure(int line)
-{
-	prom_printf("CENTRAL: Critical device probe failure at central.c:%d\n",
-		    line);
-	prom_halt();
-}
+	case 0xc0:
+		return 8;
 
-static void central_ranges_init(struct linux_central *central)
-{
-	struct device_node *dp = central->prom_node;
-	const void *pval;
-	int len;
-	
-	central->num_central_ranges = 0;
-	pval = of_get_property(dp, "ranges", &len);
-	if (pval) {
-		memcpy(central->central_ranges, pval, len);
-		central->num_central_ranges =
-			(len / sizeof(struct linux_prom_ranges));
+	case 0x80:
+		reg = 0;
+		if (p->clock_ver_reg)
+			reg = upa_readb(p->clock_ver_reg);
+		if (reg) {
+			if (reg & 0x80)
+				return 4;
+			else
+				return 5;
+		}
+		/* Fallthrough */
+	default:
+		return 4;
 	}
 }
 
-static void fhc_ranges_init(struct linux_fhc *fhc)
+static int __devinit clock_board_probe(struct of_device *op,
+				       const struct of_device_id *match)
 {
-	struct device_node *dp = fhc->prom_node;
-	const void *pval;
-	int len;
-	
-	fhc->num_fhc_ranges = 0;
-	pval = of_get_property(dp, "ranges", &len);
-	if (pval) {
-		memcpy(fhc->fhc_ranges, pval, len);
-		fhc->num_fhc_ranges =
-			(len / sizeof(struct linux_prom_ranges));
-	}
-}
+	struct clock_board *p = kzalloc(sizeof(*p), GFP_KERNEL);
+	int err = -ENOMEM;
 
-/* Range application routines are exported to various drivers,
- * so do not __init this.
- */
-static void adjust_regs(struct linux_prom_registers *regp, int nregs,
-			struct linux_prom_ranges *rangep, int nranges)
-{
-	int regc, rngc;
-
-	for (regc = 0; regc < nregs; regc++) {
-		for (rngc = 0; rngc < nranges; rngc++)
-			if (regp[regc].which_io == rangep[rngc].ot_child_space)
-				break; /* Fount it */
-		if (rngc == nranges) /* oops */
-			central_probe_failure(__LINE__);
-		regp[regc].which_io = rangep[rngc].ot_parent_space;
-		regp[regc].phys_addr -= rangep[rngc].ot_child_base;
-		regp[regc].phys_addr += rangep[rngc].ot_parent_base;
+	if (!p) {
+		printk(KERN_ERR "clock_board: Cannot allocate struct clock_board\n");
+		goto out;
 	}
-}
 
-/* Apply probed fhc ranges to registers passed, if no ranges return. */
-static void apply_fhc_ranges(struct linux_fhc *fhc,
-			     struct linux_prom_registers *regs,
-			     int nregs)
-{
-	if (fhc->num_fhc_ranges)
-		adjust_regs(regs, nregs, fhc->fhc_ranges,
-			    fhc->num_fhc_ranges);
-}
+	p->clock_freq_regs = of_ioremap(&op->resource[0], 0,
+					resource_size(&op->resource[0]),
+					"clock_board_freq");
+	if (!p->clock_freq_regs) {
+		printk(KERN_ERR "clock_board: Cannot map clock_freq_regs\n");
+		goto out_free;
+	}
 
-/* Apply probed central ranges to registers passed, if no ranges return. */
-static void apply_central_ranges(struct linux_central *central,
-				 struct linux_prom_registers *regs, int nregs)
-{
-	if (central->num_central_ranges)
-		adjust_regs(regs, nregs, central->central_ranges,
-			    central->num_central_ranges);
-}
+	p->clock_regs = of_ioremap(&op->resource[1], 0,
+				   resource_size(&op->resource[1]),
+				   "clock_board_regs");
+	if (!p->clock_regs) {
+		printk(KERN_ERR "clock_board: Cannot map clock_regs\n");
+		goto out_unmap_clock_freq_regs;
+	}
 
-static void * __init central_alloc_bootmem(unsigned long size)
-{
-	void *ret;
+	if (op->resource[2].flags) {
+		p->clock_ver_reg = of_ioremap(&op->resource[2], 0,
+					      resource_size(&op->resource[2]),
+					      "clock_ver_reg");
+		if (!p->clock_ver_reg) {
+			printk(KERN_ERR "clock_board: Cannot map clock_ver_reg\n");
+			goto out_unmap_clock_regs;
+		}
+	}
 
-	ret = __alloc_bootmem(size, SMP_CACHE_BYTES, 0UL);
-	if (ret != NULL)
-		memset(ret, 0, size);
+	p->num_slots = clock_board_calc_nslots(p);
 
-	return ret;
-}
+	p->leds_resource.start = (unsigned long)
+		(p->clock_regs + CLOCK_CTRL);
+	p->leds_resource.end = p->leds_resource.end;
+	p->leds_resource.name = "leds";
 
-static unsigned long prom_reg_to_paddr(struct linux_prom_registers *r)
-{
-	unsigned long ret = ((unsigned long) r->which_io) << 32;
+	p->leds_pdev.name = "sunfire-clockboard-leds";
+	p->leds_pdev.resource = &p->leds_resource;
+	p->leds_pdev.num_resources = 1;
+	p->leds_pdev.dev.parent = &op->dev;
 
-	return ret | (unsigned long) r->phys_addr;
-}
-
-static void __init probe_other_fhcs(void)
-{
-	struct device_node *dp;
-	const struct linux_prom64_registers *fpregs;
-
-	for_each_node_by_name(dp, "fhc") {
-		struct linux_fhc *fhc;
-		int board;
-		u32 tmp;
-
-		if (dp->parent &&
-		    dp->parent->parent != NULL)
-			continue;
-
-		fhc = (struct linux_fhc *)
-			central_alloc_bootmem(sizeof(struct linux_fhc));
-		if (fhc == NULL)
-			central_probe_failure(__LINE__);
-
-		/* Link it into the FHC chain. */
-		fhc->next = fhc_list;
-		fhc_list = fhc;
-
-		/* Toplevel FHCs have no parent. */
-		fhc->parent = NULL;
-		
-		fhc->prom_node = dp;
-		fhc_ranges_init(fhc);
-
-		/* Non-central FHC's have 64-bit OBP format registers. */
-		fpregs = of_get_property(dp, "reg", NULL);
-		if (!fpregs)
-			central_probe_failure(__LINE__);
-
-		/* Only central FHC needs special ranges applied. */
-		fhc->fhc_regs.pregs = fpregs[0].phys_addr;
-		fhc->fhc_regs.ireg = fpregs[1].phys_addr;
-		fhc->fhc_regs.ffregs = fpregs[2].phys_addr;
-		fhc->fhc_regs.sregs = fpregs[3].phys_addr;
-		fhc->fhc_regs.uregs = fpregs[4].phys_addr;
-		fhc->fhc_regs.tregs = fpregs[5].phys_addr;
-
-		board = of_getintprop_default(dp, "board#", -1);
-		fhc->board = board;
-
-		tmp = upa_readl(fhc->fhc_regs.pregs + FHC_PREGS_JCTRL);
-		if ((tmp & FHC_JTAG_CTRL_MENAB) != 0)
-			fhc->jtag_master = 1;
-		else
-			fhc->jtag_master = 0;
-
-		tmp = upa_readl(fhc->fhc_regs.pregs + FHC_PREGS_ID);
-		printk("FHC(board %d): Version[%x] PartID[%x] Manuf[%x] %s\n",
-		       board,
-		       (tmp & FHC_ID_VERS) >> 28,
-		       (tmp & FHC_ID_PARTID) >> 12,
-		       (tmp & FHC_ID_MANUF) >> 1,
-		       (fhc->jtag_master ? "(JTAG Master)" : ""));
-		
-		/* This bit must be set in all non-central FHC's in
-		 * the system.  When it is clear, this identifies
-		 * the central board.
-		 */
-		tmp = upa_readl(fhc->fhc_regs.pregs + FHC_PREGS_CTRL);
-		tmp |= FHC_CONTROL_IXIST;
-		upa_writel(tmp, fhc->fhc_regs.pregs + FHC_PREGS_CTRL);
+	err = platform_device_register(&p->leds_pdev);
+	if (err) {
+		printk(KERN_ERR "clock_board: Could not register LEDS "
+		       "platform device\n");
+		goto out_unmap_clock_ver_reg;
 	}
-}
 
-static void probe_clock_board(struct linux_central *central,
-			      struct linux_fhc *fhc,
-			      struct device_node *fp)
-{
-	struct device_node *dp;
-	struct linux_prom_registers cregs[3];
-	const struct linux_prom_registers *pr;
-	int nslots, tmp, nregs;
-
-	dp = fp->child;
-	while (dp) {
-		if (!strcmp(dp->name, "clock-board"))
-			break;
-		dp = dp->sibling;
-	}
-	if (!dp)
-		central_probe_failure(__LINE__);
+	printk(KERN_INFO "clock_board: Detected %d slot Enterprise system.\n",
+	       p->num_slots);
 
-	pr = of_get_property(dp, "reg", &nregs);
-	if (!pr)
-		central_probe_failure(__LINE__);
+	err = 0;
+out:
+	return err;
 
-	memcpy(cregs, pr, nregs);
-	nregs /= sizeof(struct linux_prom_registers);
+out_unmap_clock_ver_reg:
+	if (p->clock_ver_reg)
+		of_iounmap(&op->resource[2], p->clock_ver_reg,
+			   resource_size(&op->resource[2]));
 
-	apply_fhc_ranges(fhc, &cregs[0], nregs);
-	apply_central_ranges(central, &cregs[0], nregs);
-	central->cfreg = prom_reg_to_paddr(&cregs[0]);
-	central->clkregs = prom_reg_to_paddr(&cregs[1]);
+out_unmap_clock_regs:
+	of_iounmap(&op->resource[1], p->clock_regs,
+		   resource_size(&op->resource[1]));
 
-	if (nregs == 2)
-		central->clkver = 0UL;
-	else
-		central->clkver = prom_reg_to_paddr(&cregs[2]);
+out_unmap_clock_freq_regs:
+	of_iounmap(&op->resource[0], p->clock_freq_regs,
+		   resource_size(&op->resource[0]));
 
-	tmp = upa_readb(central->clkregs + CLOCK_STAT1);
-	tmp &= 0xc0;
-	switch(tmp) {
-	case 0x40:
-		nslots = 16;
-		break;
-	case 0xc0:
-		nslots = 8;
-		break;
-	case 0x80:
-		if (central->clkver != 0UL &&
-		   upa_readb(central->clkver) != 0) {
-			if ((upa_readb(central->clkver) & 0x80) != 0)
-				nslots = 4;
-			else
-				nslots = 5;
-			break;
-		}
-	default:
-		nslots = 4;
-		break;
-	};
-	central->slots = nslots;
-	printk("CENTRAL: Detected %d slot Enterprise system. cfreg[%02x] cver[%02x]\n",
-	       central->slots, upa_readb(central->cfreg),
-	       (central->clkver ? upa_readb(central->clkver) : 0x00));
+out_free:
+	kfree(p);
+	goto out;
 }
 
-static void ZAP(unsigned long iclr, unsigned long imap)
+static struct of_device_id __initdata clock_board_match[] = {
+	{
+		.name = "clock-board",
+	},
+	{},
+};
+
+static struct of_platform_driver clock_board_driver = {
+	.match_table	= clock_board_match,
+	.probe		= clock_board_probe,
+	.driver		= {
+		.name	= "clock_board",
+	},
+};
+
+static int __devinit fhc_probe(struct of_device *op,
+			       const struct of_device_id *match)
 {
-	u32 imap_tmp;
-
-	upa_writel(0, iclr);
-	upa_readl(iclr);
-	imap_tmp = upa_readl(imap);
-	imap_tmp &= ~(0x80000000);
-	upa_writel(imap_tmp, imap);
-	upa_readl(imap);
-}
+	struct fhc *p = kzalloc(sizeof(*p), GFP_KERNEL);
+	int err = -ENOMEM;
+	u32 reg;
 
-static void init_all_fhc_hw(void)
-{
-	struct linux_fhc *fhc;
-
-	for (fhc = fhc_list; fhc != NULL; fhc = fhc->next) {
-		u32 tmp;
-
-		/* Clear all of the interrupt mapping registers
-		 * just in case OBP left them in a foul state.
-		 */
-		ZAP(fhc->fhc_regs.ffregs + FHC_FFREGS_ICLR,
-		    fhc->fhc_regs.ffregs + FHC_FFREGS_IMAP);
-		ZAP(fhc->fhc_regs.sregs + FHC_SREGS_ICLR,
-		    fhc->fhc_regs.sregs + FHC_SREGS_IMAP);
-		ZAP(fhc->fhc_regs.uregs + FHC_UREGS_ICLR,
-		    fhc->fhc_regs.uregs + FHC_UREGS_IMAP);
-		ZAP(fhc->fhc_regs.tregs + FHC_TREGS_ICLR,
-		    fhc->fhc_regs.tregs + FHC_TREGS_IMAP);
-
-		/* Setup FHC control register. */
-		tmp = upa_readl(fhc->fhc_regs.pregs + FHC_PREGS_CTRL);
-
-		/* All non-central boards have this bit set. */
-		if (! IS_CENTRAL_FHC(fhc))
-			tmp |= FHC_CONTROL_IXIST;
-
-		/* For all FHCs, clear the firmware synchronization
-		 * line and both low power mode enables.
-		 */
-		tmp &= ~(FHC_CONTROL_AOFF | FHC_CONTROL_BOFF |
-			 FHC_CONTROL_SLINE);
-
-		upa_writel(tmp, fhc->fhc_regs.pregs + FHC_PREGS_CTRL);
-		upa_readl(fhc->fhc_regs.pregs + FHC_PREGS_CTRL);
+	if (!p) {
+		printk(KERN_ERR "fhc: Cannot allocate struct fhc\n");
+		goto out;
 	}
 
-}
+	if (!strcmp(op->node->parent->name, "central"))
+		p->central = true;
 
-void __init central_probe(void)
-{
-	struct linux_prom_registers fpregs[6];
-	const struct linux_prom_registers *pr;
-	struct linux_fhc *fhc;
-	struct device_node *dp, *fp;
-	int err;
-
-	dp = of_find_node_by_name(NULL, "central");
-	if (!dp) {
-		if (this_is_starfire)
-			starfire_cpu_setup();
-		return;
+	p->pregs = of_ioremap(&op->resource[0], 0,
+			      resource_size(&op->resource[0]),
+			      "fhc_pregs");
+	if (!p->pregs) {
+		printk(KERN_ERR "fhc: Cannot map pregs\n");
+		goto out_free;
 	}
 
-	/* Ok we got one, grab some memory for software state. */
-	central_bus = (struct linux_central *)
-		central_alloc_bootmem(sizeof(struct linux_central));
-	if (central_bus == NULL)
-		central_probe_failure(__LINE__);
-
-	fhc = (struct linux_fhc *)
-		central_alloc_bootmem(sizeof(struct linux_fhc));
-	if (fhc == NULL)
-		central_probe_failure(__LINE__);
-
-	/* First init central. */
-	central_bus->child = fhc;
-	central_bus->prom_node = dp;
-	central_ranges_init(central_bus);
-
-	/* And then central's FHC. */
-	fhc->next = fhc_list;
-	fhc_list = fhc;
-
-	fhc->parent = central_bus;
-	fp = dp->child;
-	while (fp) {
-		if (!strcmp(fp->name, "fhc"))
-			break;
-		fp = fp->sibling;
+	if (p->central) {
+		reg = upa_readl(p->pregs + FHC_PREGS_BSR);
+		p->board_num = ((reg >> 16) & 1) | ((reg >> 12) & 0x0e);
+	} else {
+		p->board_num = of_getintprop_default(op->node, "board#", -1);
+		if (p->board_num == -1) {
+			printk(KERN_ERR "fhc: No board# property\n");
+			goto out_unmap_pregs;
+		}
+		if (upa_readl(p->pregs + FHC_PREGS_JCTRL) & FHC_JTAG_CTRL_MENAB)
+			p->jtag_master = true;
 	}
-	if (!fp)
-		central_probe_failure(__LINE__);
-
-	fhc->prom_node = fp;
-	fhc_ranges_init(fhc);
-
-	/* Now, map in FHC register set. */
-	pr = of_get_property(fp, "reg", NULL);
-	if (!pr)
-		central_probe_failure(__LINE__);
-	memcpy(fpregs, pr, sizeof(fpregs));
-
-	apply_central_ranges(central_bus, &fpregs[0], 6);
-	
-	fhc->fhc_regs.pregs = prom_reg_to_paddr(&fpregs[0]);
-	fhc->fhc_regs.ireg = prom_reg_to_paddr(&fpregs[1]);
-	fhc->fhc_regs.ffregs = prom_reg_to_paddr(&fpregs[2]);
-	fhc->fhc_regs.sregs = prom_reg_to_paddr(&fpregs[3]);
-	fhc->fhc_regs.uregs = prom_reg_to_paddr(&fpregs[4]);
-	fhc->fhc_regs.tregs = prom_reg_to_paddr(&fpregs[5]);
-
-	/* Obtain board number from board status register, Central's
-	 * FHC lacks "board#" property.
-	 */
-	err = upa_readl(fhc->fhc_regs.pregs + FHC_PREGS_BSR);
-	fhc->board = (((err >> 16) & 0x01) |
-		      ((err >> 12) & 0x0e));
-
-	fhc->jtag_master = 0;
-
-	/* Attach the clock board registers for CENTRAL. */
-	probe_clock_board(central_bus, fhc, fp);
-
-	err = upa_readl(fhc->fhc_regs.pregs + FHC_PREGS_ID);
-	printk("FHC(board %d): Version[%x] PartID[%x] Manuf[%x] (CENTRAL)\n",
-	       fhc->board,
-	       ((err & FHC_ID_VERS) >> 28),
-	       ((err & FHC_ID_PARTID) >> 12),
-	       ((err & FHC_ID_MANUF) >> 1));
-
-	probe_other_fhcs();
-
-	init_all_fhc_hw();
-}
 
-static inline void fhc_ledblink(struct linux_fhc *fhc, int on)
-{
-	u32 tmp;
+	if (!p->central) {
+		p->leds_resource.start = (unsigned long)
+			(p->pregs + FHC_PREGS_CTRL);
+		p->leds_resource.end = p->leds_resource.end;
+		p->leds_resource.name = "leds";
+
+		p->leds_pdev.name = "sunfire-fhc-leds";
+		p->leds_pdev.resource = &p->leds_resource;
+		p->leds_pdev.num_resources = 1;
+		p->leds_pdev.dev.parent = &op->dev;
+
+		err = platform_device_register(&p->leds_pdev);
+		if (err) {
+			printk(KERN_ERR "fhc: Could not register LEDS "
+			       "platform device\n");
+			goto out_unmap_pregs;
+		}
+	}
+	reg = upa_readl(p->pregs + FHC_PREGS_CTRL);
 
-	tmp = upa_readl(fhc->fhc_regs.pregs + FHC_PREGS_CTRL);
+	if (!p->central)
+		reg |= FHC_CONTROL_IXIST;
 
-	/* NOTE: reverse logic on this bit */
-	if (on)
-		tmp &= ~(FHC_CONTROL_RLED);
-	else
-		tmp |= FHC_CONTROL_RLED;
-	tmp &= ~(FHC_CONTROL_AOFF | FHC_CONTROL_BOFF | FHC_CONTROL_SLINE);
+	reg &= ~(FHC_CONTROL_AOFF |
+		 FHC_CONTROL_BOFF |
+		 FHC_CONTROL_SLINE);
 
-	upa_writel(tmp, fhc->fhc_regs.pregs + FHC_PREGS_CTRL);
-	upa_readl(fhc->fhc_regs.pregs + FHC_PREGS_CTRL);
-}
+	upa_writel(reg, p->pregs + FHC_PREGS_CTRL);
+	upa_readl(p->pregs + FHC_PREGS_CTRL);
 
-static inline void central_ledblink(struct linux_central *central, int on)
-{
-	u8 tmp;
-
-	tmp = upa_readb(central->clkregs + CLOCK_CTRL);
+	reg = upa_readl(p->pregs + FHC_PREGS_ID);
+	printk(KERN_INFO "fhc: Board #%d, Version[%x] PartID[%x] Manuf[%x] %s\n",
+	       p->board_num,
+	       (reg & FHC_ID_VERS) >> 28,
+	       (reg & FHC_ID_PARTID) >> 12,
+	       (reg & FHC_ID_MANUF) >> 1,
+	       (p->jtag_master ?
+		"(JTAG Master)" :
+		(p->central ? "(Central)" : "")));
 
-	/* NOTE: reverse logic on this bit */
-	if (on)
-		tmp &= ~(CLOCK_CTRL_RLED);
-	else
-		tmp |= CLOCK_CTRL_RLED;
+	err = 0;
 
-	upa_writeb(tmp, central->clkregs + CLOCK_CTRL);
-	upa_readb(central->clkregs + CLOCK_CTRL);
-}
+out:
+	return err;
 
-static struct timer_list sftimer;
-static int led_state;
+out_unmap_pregs:
+	of_iounmap(&op->resource[0], p->pregs, resource_size(&op->resource[0]));
 
-static void sunfire_timer(unsigned long __ignored)
-{
-	struct linux_fhc *fhc;
-
-	central_ledblink(central_bus, led_state);
-	for (fhc = fhc_list; fhc != NULL; fhc = fhc->next)
-		if (! IS_CENTRAL_FHC(fhc))
-			fhc_ledblink(fhc, led_state);
-	led_state = ! led_state;
-	sftimer.expires = jiffies + (HZ >> 1);
-	add_timer(&sftimer);
+out_free:
+	kfree(p);
+	goto out;
 }
 
-/* After PCI/SBUS busses have been probed, this is called to perform
- * final initialization of all FireHose Controllers in the system.
- */
-void firetruck_init(void)
+static struct of_device_id __initdata fhc_match[] = {
+	{
+		.name = "fhc",
+	},
+	{},
+};
+
+static struct of_platform_driver fhc_driver = {
+	.match_table	= fhc_match,
+	.probe		= fhc_probe,
+	.driver		= {
+		.name	= "fhc",
+	},
+};
+
+static int __init sunfire_init(void)
 {
-	struct linux_central *central = central_bus;
-	u8 ctrl;
-
-	/* No central bus, nothing to do. */
-	if (central == NULL)
-		return;
-
-	/* OBP leaves it on, turn it off so clock board timer LED
-	 * is in sync with FHC ones.
-	 */
-	ctrl = upa_readb(central->clkregs + CLOCK_CTRL);
-	ctrl &= ~(CLOCK_CTRL_RLED);
-	upa_writeb(ctrl, central->clkregs + CLOCK_CTRL);
-
-	led_state = 0;
-	init_timer(&sftimer);
-	sftimer.data = 0;
-	sftimer.function = &sunfire_timer;
-	sftimer.expires = jiffies + (HZ >> 1);
-	add_timer(&sftimer);
+	(void) of_register_driver(&fhc_driver, &of_platform_bus_type);
+	(void) of_register_driver(&clock_board_driver, &of_platform_bus_type);
+	return 0;
 }
+
+subsys_initcall(sunfire_init);
diff --git a/arch/sparc64/kernel/chmc.c b/arch/sparc64/kernel/chmc.c
index 6d4f02e8a4cf..3b9f4d6e14a9 100644
--- a/arch/sparc64/kernel/chmc.c
+++ b/arch/sparc64/kernel/chmc.c
@@ -1,6 +1,6 @@
-/* memctrlr.c: Driver for UltraSPARC-III memory controller.
+/* chmc.c: Driver for UltraSPARC-III memory controller.
  *
- * Copyright (C) 2001, 2007 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 2001, 2007, 2008 David S. Miller (davem@davemloft.net)
  */
 
 #include <linux/module.h>
@@ -13,45 +13,64 @@
 #include <linux/smp.h>
 #include <linux/errno.h>
 #include <linux/init.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 #include <asm/spitfire.h>
 #include <asm/chmctrl.h>
 #include <asm/cpudata.h>
 #include <asm/oplib.h>
 #include <asm/prom.h>
+#include <asm/head.h>
 #include <asm/io.h>
+#include <asm/memctrl.h>
+
+#define DRV_MODULE_NAME		"chmc"
+#define PFX DRV_MODULE_NAME	": "
+#define DRV_MODULE_VERSION	"0.2"
+
+MODULE_AUTHOR("David S. Miller (davem@davemloft.net)");
+MODULE_DESCRIPTION("UltraSPARC-III memory controller driver");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(DRV_MODULE_VERSION);
+
+static int mc_type;
+#define MC_TYPE_SAFARI		1
+#define MC_TYPE_JBUS		2
+
+static dimm_printer_t us3mc_dimm_printer;
 
 #define CHMCTRL_NDGRPS	2
 #define CHMCTRL_NDIMMS	4
 
-#define DIMMS_PER_MC	(CHMCTRL_NDGRPS * CHMCTRL_NDIMMS)
+#define CHMC_DIMMS_PER_MC	(CHMCTRL_NDGRPS * CHMCTRL_NDIMMS)
 
 /* OBP memory-layout property format. */
-struct obp_map {
+struct chmc_obp_map {
 	unsigned char	dimm_map[144];
 	unsigned char	pin_map[576];
 };
 
 #define DIMM_LABEL_SZ	8
 
-struct obp_mem_layout {
+struct chmc_obp_mem_layout {
 	/* One max 8-byte string label per DIMM.  Usually
 	 * this matches the label on the motherboard where
 	 * that DIMM resides.
 	 */
-	char		dimm_labels[DIMMS_PER_MC][DIMM_LABEL_SZ];
+	char			dimm_labels[CHMC_DIMMS_PER_MC][DIMM_LABEL_SZ];
 
 	/* If symmetric use map[0], else it is
 	 * asymmetric and map[1] should be used.
 	 */
-	char		symmetric;
+	char			symmetric;
 
-	struct obp_map	map[2];
+	struct chmc_obp_map	map[2];
 };
 
 #define CHMCTRL_NBANKS	4
 
-struct bank_info {
-	struct mctrl_info	*mp;
+struct chmc_bank_info {
+	struct chmc		*p;
 	int			bank_id;
 
 	u64			raw_reg;
@@ -65,28 +84,406 @@ struct bank_info {
 	unsigned long		size;
 };
 
-struct mctrl_info {
-	struct list_head	list;
-	int			portid;
+struct chmc {
+	struct list_head		list;
+	int				portid;
+
+	struct chmc_obp_mem_layout	layout_prop;
+	int				layout_size;
+
+	void __iomem			*regs;
 
-	struct obp_mem_layout	layout_prop;
-	int			layout_size;
+	u64				timing_control1;
+	u64				timing_control2;
+	u64				timing_control3;
+	u64				timing_control4;
+	u64				memaddr_control;
 
-	void __iomem		*regs;
+	struct chmc_bank_info		logical_banks[CHMCTRL_NBANKS];
+};
+
+#define JBUSMC_REGS_SIZE		8
+
+#define JB_MC_REG1_DIMM2_BANK3		0x8000000000000000UL
+#define JB_MC_REG1_DIMM1_BANK1		0x4000000000000000UL
+#define JB_MC_REG1_DIMM2_BANK2		0x2000000000000000UL
+#define JB_MC_REG1_DIMM1_BANK0		0x1000000000000000UL
+#define JB_MC_REG1_XOR			0x0000010000000000UL
+#define JB_MC_REG1_ADDR_GEN_2		0x000000e000000000UL
+#define JB_MC_REG1_ADDR_GEN_2_SHIFT	37
+#define JB_MC_REG1_ADDR_GEN_1		0x0000001c00000000UL
+#define JB_MC_REG1_ADDR_GEN_1_SHIFT	34
+#define JB_MC_REG1_INTERLEAVE		0x0000000001800000UL
+#define JB_MC_REG1_INTERLEAVE_SHIFT	23
+#define JB_MC_REG1_DIMM2_PTYPE		0x0000000000200000UL
+#define JB_MC_REG1_DIMM2_PTYPE_SHIFT	21
+#define JB_MC_REG1_DIMM1_PTYPE		0x0000000000100000UL
+#define JB_MC_REG1_DIMM1_PTYPE_SHIFT	20
+
+#define PART_TYPE_X8		0
+#define PART_TYPE_X4		1
+
+#define INTERLEAVE_NONE		0
+#define INTERLEAVE_SAME		1
+#define INTERLEAVE_INTERNAL	2
+#define INTERLEAVE_BOTH		3
+
+#define ADDR_GEN_128MB		0
+#define ADDR_GEN_256MB		1
+#define ADDR_GEN_512MB		2
+#define ADDR_GEN_1GB		3
+
+#define JB_NUM_DIMM_GROUPS	2
+#define JB_NUM_DIMMS_PER_GROUP	2
+#define JB_NUM_DIMMS		(JB_NUM_DIMM_GROUPS * JB_NUM_DIMMS_PER_GROUP)
+
+struct jbusmc_obp_map {
+	unsigned char	dimm_map[18];
+	unsigned char	pin_map[144];
+};
+
+struct jbusmc_obp_mem_layout {
+	/* One max 8-byte string label per DIMM.  Usually
+	 * this matches the label on the motherboard where
+	 * that DIMM resides.
+	 */
+	char		dimm_labels[JB_NUM_DIMMS][DIMM_LABEL_SZ];
+
+	/* If symmetric use map[0], else it is
+	 * asymmetric and map[1] should be used.
+	 */
+	char			symmetric;
+
+	struct jbusmc_obp_map	map;
+
+	char			_pad;
+};
 
-	u64			timing_control1;
-	u64			timing_control2;
-	u64			timing_control3;
-	u64			timing_control4;
-	u64			memaddr_control;
+struct jbusmc_dimm_group {
+	struct jbusmc			*controller;
+	int				index;
+	u64				base_addr;
+	u64				size;
+};
 
-	struct bank_info	logical_banks[CHMCTRL_NBANKS];
+struct jbusmc {
+	void __iomem			*regs;
+	u64				mc_reg_1;
+	u32				portid;
+	struct jbusmc_obp_mem_layout	layout;
+	int				layout_len;
+	int				num_dimm_groups;
+	struct jbusmc_dimm_group	dimm_groups[JB_NUM_DIMM_GROUPS];
+	struct list_head		list;
 };
 
+static DEFINE_SPINLOCK(mctrl_list_lock);
 static LIST_HEAD(mctrl_list);
 
+static void mc_list_add(struct list_head *list)
+{
+	spin_lock(&mctrl_list_lock);
+	list_add(list, &mctrl_list);
+	spin_unlock(&mctrl_list_lock);
+}
+
+static void mc_list_del(struct list_head *list)
+{
+	spin_lock(&mctrl_list_lock);
+	list_del_init(list);
+	spin_unlock(&mctrl_list_lock);
+}
+
+#define SYNDROME_MIN	-1
+#define SYNDROME_MAX	144
+
+/* Covert syndrome code into the way the bits are positioned
+ * on the bus.
+ */
+static int syndrome_to_qword_code(int syndrome_code)
+{
+	if (syndrome_code < 128)
+		syndrome_code += 16;
+	else if (syndrome_code < 128 + 9)
+		syndrome_code -= (128 - 7);
+	else if (syndrome_code < (128 + 9 + 3))
+		syndrome_code -= (128 + 9 - 4);
+	else
+		syndrome_code -= (128 + 9 + 3);
+	return syndrome_code;
+}
+
+/* All this magic has to do with how a cache line comes over the wire
+ * on Safari and JBUS.  A 64-bit line comes over in 1 or more quadword
+ * cycles, each of which transmit ECC/MTAG info as well as the actual
+ * data.
+ */
+#define L2_LINE_SIZE		64
+#define L2_LINE_ADDR_MSK	(L2_LINE_SIZE - 1)
+#define QW_PER_LINE		4
+#define QW_BYTES		(L2_LINE_SIZE / QW_PER_LINE)
+#define QW_BITS			144
+#define SAFARI_LAST_BIT		(576 - 1)
+#define JBUS_LAST_BIT		(144 - 1)
+
+static void get_pin_and_dimm_str(int syndrome_code, unsigned long paddr,
+				 int *pin_p, char **dimm_str_p, void *_prop,
+				 int base_dimm_offset)
+{
+	int qword_code = syndrome_to_qword_code(syndrome_code);
+	int cache_line_offset;
+	int offset_inverse;
+	int dimm_map_index;
+	int map_val;
+
+	if (mc_type == MC_TYPE_JBUS) {
+		struct jbusmc_obp_mem_layout *p = _prop;
+
+		/* JBUS */
+		cache_line_offset = qword_code;
+		offset_inverse = (JBUS_LAST_BIT - cache_line_offset);
+		dimm_map_index = offset_inverse / 8;
+		map_val = p->map.dimm_map[dimm_map_index];
+		map_val = ((map_val >> ((7 - (offset_inverse & 7)))) & 1);
+		*dimm_str_p = p->dimm_labels[base_dimm_offset + map_val];
+		*pin_p = p->map.pin_map[cache_line_offset];
+	} else {
+		struct chmc_obp_mem_layout *p = _prop;
+		struct chmc_obp_map *mp;
+		int qword;
+
+		/* Safari */
+		if (p->symmetric)
+			mp = &p->map[0];
+		else
+			mp = &p->map[1];
+
+		qword = (paddr & L2_LINE_ADDR_MSK) / QW_BYTES;
+		cache_line_offset = ((3 - qword) * QW_BITS) + qword_code;
+		offset_inverse = (SAFARI_LAST_BIT - cache_line_offset);
+		dimm_map_index = offset_inverse >> 2;
+		map_val = mp->dimm_map[dimm_map_index];
+		map_val = ((map_val >> ((3 - (offset_inverse & 3)) << 1)) & 0x3);
+		*dimm_str_p = p->dimm_labels[base_dimm_offset + map_val];
+		*pin_p = mp->pin_map[cache_line_offset];
+	}
+}
+
+static struct jbusmc_dimm_group *jbusmc_find_dimm_group(unsigned long phys_addr)
+{
+	struct jbusmc *p;
+
+	list_for_each_entry(p, &mctrl_list, list) {
+		int i;
+
+		for (i = 0; i < p->num_dimm_groups; i++) {
+			struct jbusmc_dimm_group *dp = &p->dimm_groups[i];
+
+			if (phys_addr < dp->base_addr ||
+			    (dp->base_addr + dp->size) <= phys_addr)
+				continue;
+
+			return dp;
+		}
+	}
+	return NULL;
+}
+
+static int jbusmc_print_dimm(int syndrome_code,
+			     unsigned long phys_addr,
+			     char *buf, int buflen)
+{
+	struct jbusmc_obp_mem_layout *prop;
+	struct jbusmc_dimm_group *dp;
+	struct jbusmc *p;
+	int first_dimm;
+
+	dp = jbusmc_find_dimm_group(phys_addr);
+	if (dp == NULL ||
+	    syndrome_code < SYNDROME_MIN ||
+	    syndrome_code > SYNDROME_MAX) {
+		buf[0] = '?';
+		buf[1] = '?';
+		buf[2] = '?';
+		buf[3] = '\0';
+	}
+	p = dp->controller;
+	prop = &p->layout;
+
+	first_dimm = dp->index * JB_NUM_DIMMS_PER_GROUP;
+
+	if (syndrome_code != SYNDROME_MIN) {
+		char *dimm_str;
+		int pin;
+
+		get_pin_and_dimm_str(syndrome_code, phys_addr, &pin,
+				     &dimm_str, prop, first_dimm);
+		sprintf(buf, "%s, pin %3d", dimm_str, pin);
+	} else {
+		int dimm;
+
+		/* Multi-bit error, we just dump out all the
+		 * dimm labels associated with this dimm group.
+		 */
+		for (dimm = 0; dimm < JB_NUM_DIMMS_PER_GROUP; dimm++) {
+			sprintf(buf, "%s ",
+				prop->dimm_labels[first_dimm + dimm]);
+			buf += strlen(buf);
+		}
+	}
+
+	return 0;
+}
+
+static u64 __devinit jbusmc_dimm_group_size(u64 base,
+					    const struct linux_prom64_registers *mem_regs,
+					    int num_mem_regs)
+{
+	u64 max = base + (8UL * 1024 * 1024 * 1024);
+	u64 max_seen = base;
+	int i;
+
+	for (i = 0; i < num_mem_regs; i++) {
+		const struct linux_prom64_registers *ent;
+		u64 this_base;
+		u64 this_end;
+
+		ent = &mem_regs[i];
+		this_base = ent->phys_addr;
+		this_end = this_base + ent->reg_size;
+		if (base < this_base || base >= this_end)
+			continue;
+		if (this_end > max)
+			this_end = max;
+		if (this_end > max_seen)
+			max_seen = this_end;
+	}
+
+	return max_seen - base;
+}
+
+static void __devinit jbusmc_construct_one_dimm_group(struct jbusmc *p,
+						      unsigned long index,
+						      const struct linux_prom64_registers *mem_regs,
+						      int num_mem_regs)
+{
+	struct jbusmc_dimm_group *dp = &p->dimm_groups[index];
+
+	dp->controller = p;
+	dp->index = index;
+
+	dp->base_addr  = (p->portid * (64UL * 1024 * 1024 * 1024));
+	dp->base_addr += (index * (8UL * 1024 * 1024 * 1024));
+	dp->size = jbusmc_dimm_group_size(dp->base_addr, mem_regs, num_mem_regs);
+}
+
+static void __devinit jbusmc_construct_dimm_groups(struct jbusmc *p,
+						   const struct linux_prom64_registers *mem_regs,
+						   int num_mem_regs)
+{
+	if (p->mc_reg_1 & JB_MC_REG1_DIMM1_BANK0) {
+		jbusmc_construct_one_dimm_group(p, 0, mem_regs, num_mem_regs);
+		p->num_dimm_groups++;
+	}
+	if (p->mc_reg_1 & JB_MC_REG1_DIMM2_BANK2) {
+		jbusmc_construct_one_dimm_group(p, 1, mem_regs, num_mem_regs);
+		p->num_dimm_groups++;
+	}
+}
+
+static int __devinit jbusmc_probe(struct of_device *op,
+				  const struct of_device_id *match)
+{
+	const struct linux_prom64_registers *mem_regs;
+	struct device_node *mem_node;
+	int err, len, num_mem_regs;
+	struct jbusmc *p;
+	const u32 *prop;
+	const void *ml;
+
+	err = -ENODEV;
+	mem_node = of_find_node_by_path("/memory");
+	if (!mem_node) {
+		printk(KERN_ERR PFX "Cannot find /memory node.\n");
+		goto out;
+	}
+	mem_regs = of_get_property(mem_node, "reg", &len);
+	if (!mem_regs) {
+		printk(KERN_ERR PFX "Cannot get reg property of /memory node.\n");
+		goto out;
+	}
+	num_mem_regs = len / sizeof(*mem_regs);
+
+	err = -ENOMEM;
+	p = kzalloc(sizeof(*p), GFP_KERNEL);
+	if (!p) {
+		printk(KERN_ERR PFX "Cannot allocate struct jbusmc.\n");
+		goto out;
+	}
+
+	INIT_LIST_HEAD(&p->list);
+
+	err = -ENODEV;
+	prop = of_get_property(op->node, "portid", &len);
+	if (!prop || len != 4) {
+		printk(KERN_ERR PFX "Cannot find portid.\n");
+		goto out_free;
+	}
+
+	p->portid = *prop;
+
+	prop = of_get_property(op->node, "memory-control-register-1", &len);
+	if (!prop || len != 8) {
+		printk(KERN_ERR PFX "Cannot get memory control register 1.\n");
+		goto out_free;
+	}
+
+	p->mc_reg_1 = ((u64)prop[0] << 32) | (u64) prop[1];
+
+	err = -ENOMEM;
+	p->regs = of_ioremap(&op->resource[0], 0, JBUSMC_REGS_SIZE, "jbusmc");
+	if (!p->regs) {
+		printk(KERN_ERR PFX "Cannot map jbusmc regs.\n");
+		goto out_free;
+	}
+
+	err = -ENODEV;
+	ml = of_get_property(op->node, "memory-layout", &p->layout_len);
+	if (!ml) {
+		printk(KERN_ERR PFX "Cannot get memory layout property.\n");
+		goto out_iounmap;
+	}
+	if (p->layout_len > sizeof(p->layout)) {
+		printk(KERN_ERR PFX "Unexpected memory-layout size %d\n",
+		       p->layout_len);
+		goto out_iounmap;
+	}
+	memcpy(&p->layout, ml, p->layout_len);
+
+	jbusmc_construct_dimm_groups(p, mem_regs, num_mem_regs);
+
+	mc_list_add(&p->list);
+
+	printk(KERN_INFO PFX "UltraSPARC-IIIi memory controller at %s\n",
+	       op->node->full_name);
+
+	dev_set_drvdata(&op->dev, p);
+
+	err = 0;
+
+out:
+	return err;
+
+out_iounmap:
+	of_iounmap(&op->resource[0], p->regs, JBUSMC_REGS_SIZE);
+
+out_free:
+	kfree(p);
+	goto out;
+}
+
 /* Does BANK decode PHYS_ADDR? */
-static int bank_match(struct bank_info *bp, unsigned long phys_addr)
+static int chmc_bank_match(struct chmc_bank_info *bp, unsigned long phys_addr)
 {
 	unsigned long upper_bits = (phys_addr & PA_UPPER_BITS) >> PA_UPPER_BITS_SHIFT;
 	unsigned long lower_bits = (phys_addr & PA_LOWER_BITS) >> PA_LOWER_BITS_SHIFT;
@@ -118,25 +515,18 @@ static int bank_match(struct bank_info *bp, unsigned long phys_addr)
 }
 
 /* Given PHYS_ADDR, search memory controller banks for a match. */
-static struct bank_info *find_bank(unsigned long phys_addr)
+static struct chmc_bank_info *chmc_find_bank(unsigned long phys_addr)
 {
-	struct list_head *mctrl_head = &mctrl_list;
-	struct list_head *mctrl_entry = mctrl_head->next;
+	struct chmc *p;
 
-	for (;;) {
-		struct mctrl_info *mp =
-			list_entry(mctrl_entry, struct mctrl_info, list);
+	list_for_each_entry(p, &mctrl_list, list) {
 		int bank_no;
 
-		if (mctrl_entry == mctrl_head)
-			break;
-		mctrl_entry = mctrl_entry->next;
-
 		for (bank_no = 0; bank_no < CHMCTRL_NBANKS; bank_no++) {
-			struct bank_info *bp;
+			struct chmc_bank_info *bp;
 
-			bp = &mp->logical_banks[bank_no];
-			if (bank_match(bp, phys_addr))
+			bp = &p->logical_banks[bank_no];
+			if (chmc_bank_match(bp, phys_addr))
 				return bp;
 		}
 	}
@@ -145,17 +535,15 @@ static struct bank_info *find_bank(unsigned long phys_addr)
 }
 
 /* This is the main purpose of this driver. */
-#define SYNDROME_MIN	-1
-#define SYNDROME_MAX	144
-int chmc_getunumber(int syndrome_code,
-		    unsigned long phys_addr,
-		    char *buf, int buflen)
+static int chmc_print_dimm(int syndrome_code,
+			   unsigned long phys_addr,
+			   char *buf, int buflen)
 {
-	struct bank_info *bp;
-	struct obp_mem_layout *prop;
+	struct chmc_bank_info *bp;
+	struct chmc_obp_mem_layout *prop;
 	int bank_in_controller, first_dimm;
 
-	bp = find_bank(phys_addr);
+	bp = chmc_find_bank(phys_addr);
 	if (bp == NULL ||
 	    syndrome_code < SYNDROME_MIN ||
 	    syndrome_code > SYNDROME_MAX) {
@@ -166,60 +554,18 @@ int chmc_getunumber(int syndrome_code,
 		return 0;
 	}
 
-	prop = &bp->mp->layout_prop;
+	prop = &bp->p->layout_prop;
 	bank_in_controller = bp->bank_id & (CHMCTRL_NBANKS - 1);
 	first_dimm  = (bank_in_controller & (CHMCTRL_NDGRPS - 1));
 	first_dimm *= CHMCTRL_NDIMMS;
 
 	if (syndrome_code != SYNDROME_MIN) {
-		struct obp_map *map;
-		int qword, where_in_line, where, map_index, map_offset;
-		unsigned int map_val;
+		char *dimm_str;
+		int pin;
 
-		/* Yaay, single bit error so we can figure out
-		 * the exact dimm.
-		 */
-		if (prop->symmetric)
-			map = &prop->map[0];
-		else
-			map = &prop->map[1];
-
-		/* Covert syndrome code into the way the bits are
-		 * positioned on the bus.
-		 */
-		if (syndrome_code < 144 - 16)
-			syndrome_code += 16;
-		else if (syndrome_code < 144)
-			syndrome_code -= (144 - 7);
-		else if (syndrome_code < (144 + 3))
-			syndrome_code -= (144 + 3 - 4);
-		else
-			syndrome_code -= 144 + 3;
-
-		/* All this magic has to do with how a cache line
-		 * comes over the wire on Safari.  A 64-bit line
-		 * comes over in 4 quadword cycles, each of which
-		 * transmit ECC/MTAG info as well as the actual
-		 * data.  144 bits per quadword, 576 total.
-		 */
-#define LINE_SIZE	64
-#define LINE_ADDR_MSK	(LINE_SIZE - 1)
-#define QW_PER_LINE	4
-#define QW_BYTES	(LINE_SIZE / QW_PER_LINE)
-#define QW_BITS		144
-#define LAST_BIT	(576 - 1)
-
-		qword = (phys_addr & LINE_ADDR_MSK) / QW_BYTES;
-		where_in_line = ((3 - qword) * QW_BITS) + syndrome_code;
-		where = (LAST_BIT - where_in_line);
-		map_index = where >> 2;
-		map_offset = where & 0x3;
-		map_val = map->dimm_map[map_index];
-		map_val = ((map_val >> ((3 - map_offset) << 1)) & (2 - 1));
-
-		sprintf(buf, "%s, pin %3d",
-			prop->dimm_labels[first_dimm + map_val],
-			map->pin_map[where_in_line]);
+		get_pin_and_dimm_str(syndrome_code, phys_addr, &pin,
+				     &dimm_str, prop, first_dimm);
+		sprintf(buf, "%s, pin %3d", dimm_str, pin);
 	} else {
 		int dimm;
 
@@ -240,7 +586,7 @@ int chmc_getunumber(int syndrome_code,
  * the code is executing, you must use special ASI load/store else
  * you go through the global mapping.
  */
-static u64 read_mcreg(struct mctrl_info *mp, unsigned long offset)
+static u64 chmc_read_mcreg(struct chmc *p, unsigned long offset)
 {
 	unsigned long ret, this_cpu;
 
@@ -248,14 +594,14 @@ static u64 read_mcreg(struct mctrl_info *mp, unsigned long offset)
 
 	this_cpu = real_hard_smp_processor_id();
 
-	if (mp->portid == this_cpu) {
+	if (p->portid == this_cpu) {
 		__asm__ __volatile__("ldxa	[%1] %2, %0"
 				     : "=r" (ret)
 				     : "r" (offset), "i" (ASI_MCU_CTRL_REG));
 	} else {
 		__asm__ __volatile__("ldxa	[%1] %2, %0"
 				     : "=r" (ret)
-				     : "r" (mp->regs + offset),
+				     : "r" (p->regs + offset),
 				       "i" (ASI_PHYS_BYPASS_EC_E));
 	}
 
@@ -265,178 +611,253 @@ static u64 read_mcreg(struct mctrl_info *mp, unsigned long offset)
 }
 
 #if 0 /* currently unused */
-static void write_mcreg(struct mctrl_info *mp, unsigned long offset, u64 val)
+static void chmc_write_mcreg(struct chmc *p, unsigned long offset, u64 val)
 {
-	if (mp->portid == smp_processor_id()) {
+	if (p->portid == smp_processor_id()) {
 		__asm__ __volatile__("stxa	%0, [%1] %2"
 				     : : "r" (val),
 				         "r" (offset), "i" (ASI_MCU_CTRL_REG));
 	} else {
 		__asm__ __volatile__("ldxa	%0, [%1] %2"
 				     : : "r" (val),
-				         "r" (mp->regs + offset),
+				         "r" (p->regs + offset),
 				         "i" (ASI_PHYS_BYPASS_EC_E));
 	}
 }
 #endif
 
-static void interpret_one_decode_reg(struct mctrl_info *mp, int which_bank, u64 val)
+static void chmc_interpret_one_decode_reg(struct chmc *p, int which_bank, u64 val)
 {
-	struct bank_info *p = &mp->logical_banks[which_bank];
-
-	p->mp = mp;
-	p->bank_id = (CHMCTRL_NBANKS * mp->portid) + which_bank;
-	p->raw_reg = val;
-	p->valid = (val & MEM_DECODE_VALID) >> MEM_DECODE_VALID_SHIFT;
-	p->uk = (val & MEM_DECODE_UK) >> MEM_DECODE_UK_SHIFT;
-	p->um = (val & MEM_DECODE_UM) >> MEM_DECODE_UM_SHIFT;
-	p->lk = (val & MEM_DECODE_LK) >> MEM_DECODE_LK_SHIFT;
-	p->lm = (val & MEM_DECODE_LM) >> MEM_DECODE_LM_SHIFT;
-
-	p->base  =  (p->um);
-	p->base &= ~(p->uk);
-	p->base <<= PA_UPPER_BITS_SHIFT;
-
-	switch(p->lk) {
+	struct chmc_bank_info *bp = &p->logical_banks[which_bank];
+
+	bp->p = p;
+	bp->bank_id = (CHMCTRL_NBANKS * p->portid) + which_bank;
+	bp->raw_reg = val;
+	bp->valid = (val & MEM_DECODE_VALID) >> MEM_DECODE_VALID_SHIFT;
+	bp->uk = (val & MEM_DECODE_UK) >> MEM_DECODE_UK_SHIFT;
+	bp->um = (val & MEM_DECODE_UM) >> MEM_DECODE_UM_SHIFT;
+	bp->lk = (val & MEM_DECODE_LK) >> MEM_DECODE_LK_SHIFT;
+	bp->lm = (val & MEM_DECODE_LM) >> MEM_DECODE_LM_SHIFT;
+
+	bp->base  =  (bp->um);
+	bp->base &= ~(bp->uk);
+	bp->base <<= PA_UPPER_BITS_SHIFT;
+
+	switch(bp->lk) {
 	case 0xf:
 	default:
-		p->interleave = 1;
+		bp->interleave = 1;
 		break;
 
 	case 0xe:
-		p->interleave = 2;
+		bp->interleave = 2;
 		break;
 
 	case 0xc:
-		p->interleave = 4;
+		bp->interleave = 4;
 		break;
 
 	case 0x8:
-		p->interleave = 8;
+		bp->interleave = 8;
 		break;
 
 	case 0x0:
-		p->interleave = 16;
+		bp->interleave = 16;
 		break;
 	};
 
 	/* UK[10] is reserved, and UK[11] is not set for the SDRAM
 	 * bank size definition.
 	 */
-	p->size = (((unsigned long)p->uk &
-		    ((1UL << 10UL) - 1UL)) + 1UL) << PA_UPPER_BITS_SHIFT;
-	p->size /= p->interleave;
+	bp->size = (((unsigned long)bp->uk &
+		     ((1UL << 10UL) - 1UL)) + 1UL) << PA_UPPER_BITS_SHIFT;
+	bp->size /= bp->interleave;
 }
 
-static void fetch_decode_regs(struct mctrl_info *mp)
+static void chmc_fetch_decode_regs(struct chmc *p)
 {
-	if (mp->layout_size == 0)
+	if (p->layout_size == 0)
 		return;
 
-	interpret_one_decode_reg(mp, 0,
-				 read_mcreg(mp, CHMCTRL_DECODE1));
-	interpret_one_decode_reg(mp, 1,
-				 read_mcreg(mp, CHMCTRL_DECODE2));
-	interpret_one_decode_reg(mp, 2,
-				 read_mcreg(mp, CHMCTRL_DECODE3));
-	interpret_one_decode_reg(mp, 3,
-				 read_mcreg(mp, CHMCTRL_DECODE4));
+	chmc_interpret_one_decode_reg(p, 0,
+				      chmc_read_mcreg(p, CHMCTRL_DECODE1));
+	chmc_interpret_one_decode_reg(p, 1,
+				      chmc_read_mcreg(p, CHMCTRL_DECODE2));
+	chmc_interpret_one_decode_reg(p, 2,
+				      chmc_read_mcreg(p, CHMCTRL_DECODE3));
+	chmc_interpret_one_decode_reg(p, 3,
+				      chmc_read_mcreg(p, CHMCTRL_DECODE4));
 }
 
-static int init_one_mctrl(struct device_node *dp)
+static int __devinit chmc_probe(struct of_device *op,
+				const struct of_device_id *match)
 {
-	struct mctrl_info *mp = kzalloc(sizeof(*mp), GFP_KERNEL);
-	int portid = of_getintprop_default(dp, "portid", -1);
-	const struct linux_prom64_registers *regs;
+	struct device_node *dp = op->node;
+	unsigned long ver;
 	const void *pval;
-	int len;
+	int len, portid;
+	struct chmc *p;
+	int err;
+
+	err = -ENODEV;
+	__asm__ ("rdpr %%ver, %0" : "=r" (ver));
+	if ((ver >> 32UL) == __JALAPENO_ID ||
+	    (ver >> 32UL) == __SERRANO_ID)
+		goto out;
 
-	if (!mp)
-		return -1;
+	portid = of_getintprop_default(dp, "portid", -1);
 	if (portid == -1)
-		goto fail;
+		goto out;
 
-	mp->portid = portid;
 	pval = of_get_property(dp, "memory-layout", &len);
-	mp->layout_size = len;
-	if (!pval)
-		mp->layout_size = 0;
-	else {
-		if (mp->layout_size > sizeof(mp->layout_prop))
-			goto fail;
-		memcpy(&mp->layout_prop, pval, len);
+	if (pval && len > sizeof(p->layout_prop)) {
+		printk(KERN_ERR PFX "Unexpected memory-layout property "
+		       "size %d.\n", len);
+		goto out;
 	}
 
-	regs = of_get_property(dp, "reg", NULL);
-	if (!regs || regs->reg_size != 0x48)
-		goto fail;
+	err = -ENOMEM;
+	p = kzalloc(sizeof(*p), GFP_KERNEL);
+	if (!p) {
+		printk(KERN_ERR PFX "Could not allocate struct chmc.\n");
+		goto out;
+	}
 
-	mp->regs = ioremap(regs->phys_addr, regs->reg_size);
-	if (mp->regs == NULL)
-		goto fail;
+	p->portid = portid;
+	p->layout_size = len;
+	if (!pval)
+		p->layout_size = 0;
+	else
+		memcpy(&p->layout_prop, pval, len);
+
+	p->regs = of_ioremap(&op->resource[0], 0, 0x48, "chmc");
+	if (!p->regs) {
+		printk(KERN_ERR PFX "Could not map registers.\n");
+		goto out_free;
+	}
 
-	if (mp->layout_size != 0UL) {
-		mp->timing_control1 = read_mcreg(mp, CHMCTRL_TCTRL1);
-		mp->timing_control2 = read_mcreg(mp, CHMCTRL_TCTRL2);
-		mp->timing_control3 = read_mcreg(mp, CHMCTRL_TCTRL3);
-		mp->timing_control4 = read_mcreg(mp, CHMCTRL_TCTRL4);
-		mp->memaddr_control = read_mcreg(mp, CHMCTRL_MACTRL);
+	if (p->layout_size != 0UL) {
+		p->timing_control1 = chmc_read_mcreg(p, CHMCTRL_TCTRL1);
+		p->timing_control2 = chmc_read_mcreg(p, CHMCTRL_TCTRL2);
+		p->timing_control3 = chmc_read_mcreg(p, CHMCTRL_TCTRL3);
+		p->timing_control4 = chmc_read_mcreg(p, CHMCTRL_TCTRL4);
+		p->memaddr_control = chmc_read_mcreg(p, CHMCTRL_MACTRL);
 	}
 
-	fetch_decode_regs(mp);
+	chmc_fetch_decode_regs(p);
 
-	list_add(&mp->list, &mctrl_list);
+	mc_list_add(&p->list);
 
-	/* Report the device. */
-	printk(KERN_INFO "%s: US3 memory controller at %p [%s]\n",
+	printk(KERN_INFO PFX "UltraSPARC-III memory controller at %s [%s]\n",
 	       dp->full_name,
-	       mp->regs, (mp->layout_size ? "ACTIVE" : "INACTIVE"));
+	       (p->layout_size ? "ACTIVE" : "INACTIVE"));
 
-	return 0;
+	dev_set_drvdata(&op->dev, p);
+
+	err = 0;
+
+out:
+	return err;
+
+out_free:
+	kfree(p);
+	goto out;
+}
 
-fail:
-	if (mp) {
-		if (mp->regs != NULL)
-			iounmap(mp->regs);
-		kfree(mp);
+static int __devinit us3mc_probe(struct of_device *op,
+				const struct of_device_id *match)
+{
+	if (mc_type == MC_TYPE_SAFARI)
+		return chmc_probe(op, match);
+	else if (mc_type == MC_TYPE_JBUS)
+		return jbusmc_probe(op, match);
+	return -ENODEV;
+}
+
+static void __devexit chmc_destroy(struct of_device *op, struct chmc *p)
+{
+	list_del(&p->list);
+	of_iounmap(&op->resource[0], p->regs, 0x48);
+	kfree(p);
+}
+
+static void __devexit jbusmc_destroy(struct of_device *op, struct jbusmc *p)
+{
+	mc_list_del(&p->list);
+	of_iounmap(&op->resource[0], p->regs, JBUSMC_REGS_SIZE);
+	kfree(p);
+}
+
+static int __devexit us3mc_remove(struct of_device *op)
+{
+	void *p = dev_get_drvdata(&op->dev);
+
+	if (p) {
+		if (mc_type == MC_TYPE_SAFARI)
+			chmc_destroy(op, p);
+		else if (mc_type == MC_TYPE_JBUS)
+			jbusmc_destroy(op, p);
 	}
-	return -1;
+	return 0;
+}
+
+static const struct of_device_id us3mc_match[] = {
+	{
+		.name = "memory-controller",
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, us3mc_match);
+
+static struct of_platform_driver us3mc_driver = {
+	.name		= "us3mc",
+	.match_table	= us3mc_match,
+	.probe		= us3mc_probe,
+	.remove		= __devexit_p(us3mc_remove),
+};
+
+static inline bool us3mc_platform(void)
+{
+	if (tlb_type == cheetah || tlb_type == cheetah_plus)
+		return true;
+	return false;
 }
 
-static int __init chmc_init(void)
+static int __init us3mc_init(void)
 {
-	struct device_node *dp;
+	unsigned long ver;
+	int ret;
 
-	/* This driver is only for cheetah platforms. */
-	if (tlb_type != cheetah && tlb_type != cheetah_plus)
+	if (!us3mc_platform())
 		return -ENODEV;
 
-	for_each_node_by_name(dp, "memory-controller")
-		init_one_mctrl(dp);
+	__asm__ __volatile__("rdpr %%ver, %0" : "=r" (ver));
+	if ((ver >> 32UL) == __JALAPENO_ID ||
+	    (ver >> 32UL) == __SERRANO_ID) {
+		mc_type = MC_TYPE_JBUS;
+		us3mc_dimm_printer = jbusmc_print_dimm;
+	} else {
+		mc_type = MC_TYPE_SAFARI;
+		us3mc_dimm_printer = chmc_print_dimm;
+	}
 
-	for_each_node_by_name(dp, "mc-us3")
-		init_one_mctrl(dp);
+	ret = register_dimm_printer(us3mc_dimm_printer);
 
-	return 0;
+	if (!ret) {
+		ret = of_register_driver(&us3mc_driver, &of_bus_type);
+		if (ret)
+			unregister_dimm_printer(us3mc_dimm_printer);
+	}
+	return ret;
 }
 
-static void __exit chmc_cleanup(void)
+static void __exit us3mc_cleanup(void)
 {
-	struct list_head *head = &mctrl_list;
-	struct list_head *tmp = head->next;
-
-	for (;;) {
-		struct mctrl_info *p =
-			list_entry(tmp, struct mctrl_info, list);
-		if (tmp == head)
-			break;
-		tmp = tmp->next;
-
-		list_del(&p->list);
-		iounmap(p->regs);
-		kfree(p);
+	if (us3mc_platform()) {
+		unregister_dimm_printer(us3mc_dimm_printer);
+		of_unregister_driver(&us3mc_driver);
 	}
 }
 
-module_init(chmc_init);
-module_exit(chmc_cleanup);
+module_init(us3mc_init);
+module_exit(us3mc_cleanup);
diff --git a/arch/sparc64/kernel/cpu.c b/arch/sparc64/kernel/cpu.c
index 0097c08dc600..0c9ac83ed0a8 100644
--- a/arch/sparc64/kernel/cpu.c
+++ b/arch/sparc64/kernel/cpu.c
@@ -1,7 +1,7 @@
 /* cpu.c: Dinky routines to look for the kind of Sparc cpu
  *        we are on.
  *
- * Copyright (C) 1996, 2007 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 1996, 2007, 2008 David S. Miller (davem@davemloft.net)
  */
 
 #include <linux/kernel.h>
@@ -19,53 +19,86 @@
 
 DEFINE_PER_CPU(cpuinfo_sparc, __cpu_data) = { 0 };
 
-struct cpu_iu_info {
-  short manuf;
-  short impl;
-  char* cpu_name;   /* should be enough I hope... */
+struct cpu_chip_info {
+	unsigned short	manuf;
+	unsigned short	impl;
+	const char	*cpu_name;
+	const char	*fp_name;
 };
 
-struct cpu_fp_info {
-  short manuf;
-  short impl;
-  char fpu_vers;
-  char* fp_name;
+static const struct cpu_chip_info cpu_chips[] = {
+	{
+		.manuf		= 0x17,
+		.impl		= 0x10,
+		.cpu_name	= "TI UltraSparc I   (SpitFire)",
+		.fp_name	= "UltraSparc I integrated FPU",
+	},
+	{
+		.manuf		= 0x22,
+		.impl		= 0x10,
+		.cpu_name	= "TI UltraSparc I   (SpitFire)",
+		.fp_name	= "UltraSparc I integrated FPU",
+	},
+	{
+		.manuf		= 0x17,
+		.impl		= 0x11,
+		.cpu_name	= "TI UltraSparc II  (BlackBird)",
+		.fp_name	= "UltraSparc II integrated FPU",
+	},
+	{
+		.manuf		= 0x17,
+		.impl		= 0x12,
+		.cpu_name	= "TI UltraSparc IIi (Sabre)",
+		.fp_name	= "UltraSparc IIi integrated FPU",
+	},
+	{
+		.manuf		= 0x17,
+		.impl		= 0x13,
+		.cpu_name	= "TI UltraSparc IIe (Hummingbird)",
+		.fp_name	= "UltraSparc IIe integrated FPU",
+	},
+	{
+		.manuf		= 0x3e,
+		.impl		= 0x14,
+		.cpu_name	= "TI UltraSparc III (Cheetah)",
+		.fp_name	= "UltraSparc III integrated FPU",
+	},
+	{
+		.manuf		= 0x3e,
+		.impl		= 0x15,
+		.cpu_name	= "TI UltraSparc III+ (Cheetah+)",
+		.fp_name	= "UltraSparc III+ integrated FPU",
+	},
+	{
+		.manuf		= 0x3e,
+		.impl		= 0x16,
+		.cpu_name	= "TI UltraSparc IIIi (Jalapeno)",
+		.fp_name	= "UltraSparc IIIi integrated FPU",
+	},
+	{
+		.manuf		= 0x3e,
+		.impl		= 0x18,
+		.cpu_name	= "TI UltraSparc IV (Jaguar)",
+		.fp_name	= "UltraSparc IV integrated FPU",
+	},
+	{
+		.manuf		= 0x3e,
+		.impl		= 0x19,
+		.cpu_name	= "TI UltraSparc IV+ (Panther)",
+		.fp_name	= "UltraSparc IV+ integrated FPU",
+	},
+	{
+		.manuf		= 0x3e,
+		.impl		= 0x22,
+		.cpu_name	= "TI UltraSparc IIIi+ (Serrano)",
+		.fp_name	= "UltraSparc IIIi+ integrated FPU",
+	},
 };
 
-static struct cpu_fp_info linux_sparc_fpu[] = {
-  { 0x17, 0x10, 0, "UltraSparc I integrated FPU"},
-  { 0x22, 0x10, 0, "UltraSparc I integrated FPU"},
-  { 0x17, 0x11, 0, "UltraSparc II integrated FPU"},
-  { 0x17, 0x12, 0, "UltraSparc IIi integrated FPU"},
-  { 0x17, 0x13, 0, "UltraSparc IIe integrated FPU"},
-  { 0x3e, 0x14, 0, "UltraSparc III integrated FPU"},
-  { 0x3e, 0x15, 0, "UltraSparc III+ integrated FPU"},
-  { 0x3e, 0x16, 0, "UltraSparc IIIi integrated FPU"},
-  { 0x3e, 0x18, 0, "UltraSparc IV integrated FPU"},
-  { 0x3e, 0x19, 0, "UltraSparc IV+ integrated FPU"},
-  { 0x3e, 0x22, 0, "UltraSparc IIIi+ integrated FPU"},
-};
-
-#define NSPARCFPU  ARRAY_SIZE(linux_sparc_fpu)
-
-static struct cpu_iu_info linux_sparc_chips[] = {
-  { 0x17, 0x10, "TI UltraSparc I   (SpitFire)"},
-  { 0x22, 0x10, "TI UltraSparc I   (SpitFire)"},
-  { 0x17, 0x11, "TI UltraSparc II  (BlackBird)"},
-  { 0x17, 0x12, "TI UltraSparc IIi (Sabre)"},
-  { 0x17, 0x13, "TI UltraSparc IIe (Hummingbird)"},
-  { 0x3e, 0x14, "TI UltraSparc III (Cheetah)"},
-  { 0x3e, 0x15, "TI UltraSparc III+ (Cheetah+)"},
-  { 0x3e, 0x16, "TI UltraSparc IIIi (Jalapeno)"},
-  { 0x3e, 0x18, "TI UltraSparc IV (Jaguar)"},
-  { 0x3e, 0x19, "TI UltraSparc IV+ (Panther)"},
-  { 0x3e, 0x22, "TI UltraSparc IIIi+ (Serrano)"},
-};
+#define NSPARCCHIPS ARRAY_SIZE(linux_sparc_chips)
 
-#define NSPARCCHIPS  ARRAY_SIZE(linux_sparc_chips)
-
-char *sparc_cpu_type;
-char *sparc_fpu_type;
+const char *sparc_cpu_type;
+const char *sparc_fpu_type;
 
 static void __init sun4v_cpu_probe(void)
 {
@@ -89,68 +122,45 @@ static void __init sun4v_cpu_probe(void)
 	}
 }
 
-void __init cpu_probe(void)
+static const struct cpu_chip_info * __init find_cpu_chip(unsigned short manuf,
+							 unsigned short impl)
 {
-	unsigned long ver, fpu_vers, manuf, impl, fprs;
 	int i;
-	
-	if (tlb_type == hypervisor) {
-		sun4v_cpu_probe();
-		return;
-	}
 
-	fprs = fprs_read();
-	fprs_write(FPRS_FEF);
-	__asm__ __volatile__ ("rdpr %%ver, %0; stx %%fsr, [%1]"
-			      : "=&r" (ver)
-			      : "r" (&fpu_vers));
-	fprs_write(fprs);
-	
-	manuf = ((ver >> 48) & 0xffff);
-	impl = ((ver >> 32) & 0xffff);
-
-	fpu_vers = ((fpu_vers >> 17) & 0x7);
-
-retry:
-	for (i = 0; i < NSPARCCHIPS; i++) {
-		if (linux_sparc_chips[i].manuf == manuf) {
-			if (linux_sparc_chips[i].impl == impl) {
-				sparc_cpu_type =
-					linux_sparc_chips[i].cpu_name;
-				break;
-			}
-		}
-	}
+	for (i = 0; i < ARRAY_SIZE(cpu_chips); i++) {
+		const struct cpu_chip_info *p = &cpu_chips[i];
 
-	if (i == NSPARCCHIPS) {
- 		/* Maybe it is a cheetah+ derivative, report it as cheetah+
- 		 * in that case until we learn the real names.
- 		 */
- 		if (manuf == 0x3e &&
- 		    impl > 0x15) {
- 			impl = 0x15;
- 			goto retry;
- 		} else {
- 			printk("DEBUG: manuf[%lx] impl[%lx]\n",
- 			       manuf, impl);
- 		}
-		sparc_cpu_type = "Unknown CPU";
+		if (p->manuf == manuf && p->impl == impl)
+			return p;
 	}
+	return NULL;
+}
 
-	for (i = 0; i < NSPARCFPU; i++) {
-		if (linux_sparc_fpu[i].manuf == manuf &&
-		    linux_sparc_fpu[i].impl == impl) {
-			if (linux_sparc_fpu[i].fpu_vers == fpu_vers) {
-				sparc_fpu_type =
-					linux_sparc_fpu[i].fp_name;
-				break;
-			}
+static int __init cpu_type_probe(void)
+{
+	if (tlb_type == hypervisor) {
+		sun4v_cpu_probe();
+	} else {
+		unsigned long ver, manuf, impl;
+		const struct cpu_chip_info *p;
+	
+		__asm__ __volatile__("rdpr %%ver, %0" : "=r" (ver));
+	
+		manuf = ((ver >> 48) & 0xffff);
+		impl = ((ver >> 32) & 0xffff);
+
+		p = find_cpu_chip(manuf, impl);
+		if (p) {
+			sparc_cpu_type = p->cpu_name;
+			sparc_fpu_type = p->fp_name;
+		} else {
+			printk(KERN_ERR "CPU: Unknown chip, manuf[%lx] impl[%lx]\n",
+			       manuf, impl);
+			sparc_cpu_type = "Unknown CPU";
+			sparc_fpu_type = "Unknown FPU";
 		}
 	}
-
-	if (i == NSPARCFPU) {
- 		printk("DEBUG: manuf[%lx] impl[%lx] fsr.vers[%lx]\n",
- 		       manuf, impl, fpu_vers);
-		sparc_fpu_type = "Unknown FPU";
-	}
+	return 0;
 }
+
+arch_initcall(cpu_type_probe);
diff --git a/arch/sparc64/kernel/ds.c b/arch/sparc64/kernel/ds.c
index d0fa5aa38934..f52e0534d91d 100644
--- a/arch/sparc64/kernel/ds.c
+++ b/arch/sparc64/kernel/ds.c
@@ -1,6 +1,6 @@
 /* ds.c: Domain Services driver for Logical Domains
  *
- * Copyright (C) 2007 David S. Miller <davem@davemloft.net>
+ * Copyright (C) 2007, 2008 David S. Miller <davem@davemloft.net>
  */
 
 #include <linux/kernel.h>
@@ -1217,7 +1217,7 @@ static int ds_remove(struct vio_dev *vdev)
 	return 0;
 }
 
-static struct vio_device_id ds_match[] = {
+static struct vio_device_id __initdata ds_match[] = {
 	{
 		.type = "domain-services-port",
 	},
diff --git a/arch/sparc64/kernel/ebus.c b/arch/sparc64/kernel/ebus.c
index 60d36d142559..77dbf6d45faf 100644
--- a/arch/sparc64/kernel/ebus.c
+++ b/arch/sparc64/kernel/ebus.c
@@ -1,5 +1,4 @@
-/*
- * ebus.c: PCI to EBus bridge device.
+/* ebus.c: EBUS DMA library code.
  *
  * Copyright (C) 1997  Eddie C. Dost  (ecd@skynet.be)
  * Copyright (C) 1999  David S. Miller (davem@redhat.com)
@@ -9,23 +8,11 @@
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/init.h>
-#include <linux/slab.h>
-#include <linux/string.h>
 #include <linux/interrupt.h>
 #include <linux/delay.h>
-#include <linux/pci.h>
-#include <linux/of_device.h>
-
-#include <asm/system.h>
-#include <asm/page.h>
-#include <asm/ebus.h>
-#include <asm/oplib.h>
-#include <asm/prom.h>
-#include <asm/bpp.h>
-#include <asm/irq.h>
-#include <asm/io.h>
 
-/* EBUS dma library. */
+#include <asm/ebus_dma.h>
+#include <asm/io.h>
 
 #define EBDMA_CSR	0x00UL	/* Control/Status */
 #define EBDMA_ADDR	0x04UL	/* DMA Address */
@@ -268,283 +255,3 @@ void ebus_dma_enable(struct ebus_dma_info *p, int on)
 	spin_unlock_irqrestore(&p->lock, flags);
 }
 EXPORT_SYMBOL(ebus_dma_enable);
-
-struct linux_ebus *ebus_chain = NULL;
-
-static inline void *ebus_alloc(size_t size)
-{
-	void *mem;
-
-	mem = kzalloc(size, GFP_ATOMIC);
-	if (!mem)
-		panic("ebus_alloc: out of memory");
-	return mem;
-}
-
-static void __init fill_ebus_child(struct device_node *dp,
-				   struct linux_ebus_child *dev,
-				   int non_standard_regs)
-{
-	struct of_device *op;
-	const int *regs;
-	int i, len;
-
-	dev->prom_node = dp;
-	printk(" (%s)", dp->name);
-
-	regs = of_get_property(dp, "reg", &len);
-	if (!regs)
-		dev->num_addrs = 0;
-	else
-		dev->num_addrs = len / sizeof(regs[0]);
-
-	if (non_standard_regs) {
-		/* This is to handle reg properties which are not
-		 * in the parent relative format.  One example are
-		 * children of the i2c device on CompactPCI systems.
-		 *
-		 * So, for such devices we just record the property
-		 * raw in the child resources.
-		 */
-		for (i = 0; i < dev->num_addrs; i++)
-			dev->resource[i].start = regs[i];
-	} else {
-		for (i = 0; i < dev->num_addrs; i++) {
-			int rnum = regs[i];
-			if (rnum >= dev->parent->num_addrs) {
-				prom_printf("UGH: property for %s was %d, need < %d\n",
-					    dp->name, len, dev->parent->num_addrs);
-				prom_halt();
-			}
-			dev->resource[i].start = dev->parent->resource[i].start;
-			dev->resource[i].end = dev->parent->resource[i].end;
-			dev->resource[i].flags = IORESOURCE_MEM;
-			dev->resource[i].name = dp->name;
-		}
-	}
-
-	op = of_find_device_by_node(dp);
-	if (!op) {
-		dev->num_irqs = 0;
-	} else {
-		dev->num_irqs = op->num_irqs;
-		for (i = 0; i < dev->num_irqs; i++)
-			dev->irqs[i] = op->irqs[i];
-	}
-
-	if (!dev->num_irqs) {
-		/*
-		 * Oh, well, some PROMs don't export interrupts
-		 * property to children of EBus devices...
-		 *
-		 * Be smart about PS/2 keyboard and mouse.
-		 */
-		if (!strcmp(dev->parent->prom_node->name, "8042")) {
-			if (!strcmp(dev->prom_node->name, "kb_ps2")) {
-				dev->num_irqs = 1;
-				dev->irqs[0] = dev->parent->irqs[0];
-			} else {
-				dev->num_irqs = 1;
-				dev->irqs[0] = dev->parent->irqs[1];
-			}
-		}
-	}
-}
-
-static int __init child_regs_nonstandard(struct linux_ebus_device *dev)
-{
-	if (!strcmp(dev->prom_node->name, "i2c") ||
-	    !strcmp(dev->prom_node->name, "SUNW,lombus"))
-		return 1;
-	return 0;
-}
-
-static void __init fill_ebus_device(struct device_node *dp, struct linux_ebus_device *dev)
-{
-	struct linux_ebus_child *child;
-	struct dev_archdata *sd;
-	struct of_device *op;
-	int i, len;
-
-	dev->prom_node = dp;
-
-	printk(" [%s", dp->name);
-
-	op = of_find_device_by_node(dp);
-	if (!op) {
-		dev->num_addrs = 0;
-		dev->num_irqs = 0;
-	} else {
-		const int *regs = of_get_property(dp, "reg", &len);
-
-		if (!regs)
-			len = 0;
-		dev->num_addrs = len / sizeof(struct linux_prom_registers);
-
-		for (i = 0; i < dev->num_addrs; i++)
-			memcpy(&dev->resource[i],
-			       &op->resource[i],
-			       sizeof(struct resource));
-
-		dev->num_irqs = op->num_irqs;
-		for (i = 0; i < dev->num_irqs; i++)
-			dev->irqs[i] = op->irqs[i];
-	}
-
-	sd = &dev->ofdev.dev.archdata;
-	sd->prom_node = dp;
-	sd->op = &dev->ofdev;
-	sd->iommu = dev->bus->ofdev.dev.parent->archdata.iommu;
-	sd->stc = dev->bus->ofdev.dev.parent->archdata.stc;
-	sd->numa_node = dev->bus->ofdev.dev.parent->archdata.numa_node;
-
-	dev->ofdev.node = dp;
-	dev->ofdev.dev.parent = &dev->bus->ofdev.dev;
-	dev->ofdev.dev.bus = &ebus_bus_type;
-	dev_set_name(&dev->ofdev.dev, "ebus[%08x]", dp->node);
-
-	/* Register with core */
-	if (of_device_register(&dev->ofdev) != 0)
-		printk(KERN_DEBUG "ebus: device registration error for %s!\n",
-		       dp->path_component_name);
-
-	dp = dp->child;
-	if (dp) {
-		printk(" ->");
-		dev->children = ebus_alloc(sizeof(struct linux_ebus_child));
-
-		child = dev->children;
-		child->next = NULL;
-		child->parent = dev;
-		child->bus = dev->bus;
-		fill_ebus_child(dp, child,
-				child_regs_nonstandard(dev));
-
-		while ((dp = dp->sibling) != NULL) {
-			child->next = ebus_alloc(sizeof(struct linux_ebus_child));
-
-			child = child->next;
-			child->next = NULL;
-			child->parent = dev;
-			child->bus = dev->bus;
-			fill_ebus_child(dp, child,
-					child_regs_nonstandard(dev));
-		}
-	}
-	printk("]");
-}
-
-static struct pci_dev *find_next_ebus(struct pci_dev *start, int *is_rio_p)
-{
-	struct pci_dev *pdev = start;
-
-	while ((pdev = pci_get_device(PCI_VENDOR_ID_SUN, PCI_ANY_ID, pdev)))
-		if (pdev->device == PCI_DEVICE_ID_SUN_EBUS ||
-			pdev->device == PCI_DEVICE_ID_SUN_RIO_EBUS)
-			break;
-
-	*is_rio_p = !!(pdev && (pdev->device == PCI_DEVICE_ID_SUN_RIO_EBUS));
-
-	return pdev;
-}
-
-void __init ebus_init(void)
-{
-	struct linux_ebus_device *dev;
-	struct linux_ebus *ebus;
-	struct pci_dev *pdev;
-	struct device_node *dp;
-	int is_rio;
-	int num_ebus = 0;
-
-	pdev = find_next_ebus(NULL, &is_rio);
-	if (!pdev) {
-		printk("ebus: No EBus's found.\n");
-		return;
-	}
-
-	dp = pci_device_to_OF_node(pdev);
-
-	ebus_chain = ebus = ebus_alloc(sizeof(struct linux_ebus));
-	ebus->next = NULL;
-	ebus->is_rio = is_rio;
-
-	while (dp) {
-		struct device_node *child;
-
-		/* SUNW,pci-qfe uses four empty ebuses on it.
-		   I think we should not consider them here,
-		   as they have half of the properties this
-		   code expects and once we do PCI hot-plug,
-		   we'd have to tweak with the ebus_chain
-		   in the runtime after initialization. -jj */
-		if (!dp->child) {
-			pdev = find_next_ebus(pdev, &is_rio);
-			if (!pdev) {
-				if (ebus == ebus_chain) {
-					ebus_chain = NULL;
-					printk("ebus: No EBus's found.\n");
-					return;
-				}
-				break;
-			}
-			ebus->is_rio = is_rio;
-			dp = pci_device_to_OF_node(pdev);
-			continue;
-		}
-		printk("ebus%d:", num_ebus);
-
-		ebus->index = num_ebus;
-		ebus->prom_node = dp;
-		ebus->self = pdev;
-
-		ebus->ofdev.node = dp;
-		ebus->ofdev.dev.parent = &pdev->dev;
-		ebus->ofdev.dev.bus = &ebus_bus_type;
-		dev_set_name(&ebus->ofdev.dev, "ebus%d", num_ebus);
-
-		/* Register with core */
-		if (of_device_register(&ebus->ofdev) != 0)
-			printk(KERN_DEBUG "ebus: device registration error for %s!\n",
-			       dp->path_component_name);
-
-
-		child = dp->child;
-		if (!child)
-			goto next_ebus;
-
-		ebus->devices = ebus_alloc(sizeof(struct linux_ebus_device));
-
-		dev = ebus->devices;
-		dev->next = NULL;
-		dev->children = NULL;
-		dev->bus = ebus;
-		fill_ebus_device(child, dev);
-
-		while ((child = child->sibling) != NULL) {
-			dev->next = ebus_alloc(sizeof(struct linux_ebus_device));
-
-			dev = dev->next;
-			dev->next = NULL;
-			dev->children = NULL;
-			dev->bus = ebus;
-			fill_ebus_device(child, dev);
-		}
-
-	next_ebus:
-		printk("\n");
-
-		pdev = find_next_ebus(pdev, &is_rio);
-		if (!pdev)
-			break;
-
-		dp = pci_device_to_OF_node(pdev);
-
-		ebus->next = ebus_alloc(sizeof(struct linux_ebus));
-		ebus = ebus->next;
-		ebus->next = NULL;
-		ebus->is_rio = is_rio;
-		++num_ebus;
-	}
-	pci_dev_put(pdev); /* XXX for the case, when ebusnd is 0, is it OK? */
-}
diff --git a/arch/sparc64/kernel/entry.h b/arch/sparc64/kernel/entry.h
index fc294a292899..34d7ab5e10d2 100644
--- a/arch/sparc64/kernel/entry.h
+++ b/arch/sparc64/kernel/entry.h
@@ -5,8 +5,8 @@
 #include <linux/types.h>
 #include <linux/init.h>
 
-extern char *sparc_cpu_type;
-extern char *sparc_fpu_type;
+extern const char *sparc_cpu_type;
+extern const char *sparc_fpu_type;
 
 extern void __init per_cpu_patch(void);
 extern void __init sun4v_patch(void);
@@ -22,7 +22,8 @@ extern void do_notify_resume(struct pt_regs *regs,
 			     unsigned long orig_i0,
 			     unsigned long thread_info_flags);
 
-extern asmlinkage int syscall_trace(struct pt_regs *regs, int syscall_exit_p);
+extern asmlinkage int syscall_trace_enter(struct pt_regs *regs);
+extern asmlinkage void syscall_trace_leave(struct pt_regs *regs);
 
 extern void bad_trap_tl1(struct pt_regs *regs, long lvl);
 
diff --git a/arch/sparc64/kernel/head.S b/arch/sparc64/kernel/head.S
index c9afef093d51..353226fa0239 100644
--- a/arch/sparc64/kernel/head.S
+++ b/arch/sparc64/kernel/head.S
@@ -10,6 +10,7 @@
 #include <linux/errno.h>
 #include <linux/threads.h>
 #include <linux/init.h>
+#include <linux/linkage.h>
 #include <asm/thread_info.h>
 #include <asm/asi.h>
 #include <asm/pstate.h>
diff --git a/arch/sparc64/kernel/hvapi.c b/arch/sparc64/kernel/hvapi.c
index 691760b5b012..1d272c3b5740 100644
--- a/arch/sparc64/kernel/hvapi.c
+++ b/arch/sparc64/kernel/hvapi.c
@@ -9,7 +9,6 @@
 
 #include <asm/hypervisor.h>
 #include <asm/oplib.h>
-#include <asm/sstate.h>
 
 /* If the hypervisor indicates that the API setting
  * calls are unsupported, by returning HV_EBADTRAP or
@@ -184,8 +183,6 @@ void __init sun4v_hvapi_init(void)
 	if (sun4v_hvapi_register(group, major, &minor))
 		goto bad;
 
-	sun4v_sstate_init();
-
 	return;
 
 bad:
diff --git a/arch/sparc64/kernel/hvcalls.S b/arch/sparc64/kernel/hvcalls.S
index a2810f3ac70f..e066269d1594 100644
--- a/arch/sparc64/kernel/hvcalls.S
+++ b/arch/sparc64/kernel/hvcalls.S
@@ -3,89 +3,75 @@
 	 *
 	 * returns %o0: sysino
 	 */
-	.globl	sun4v_devino_to_sysino
-	.type	sun4v_devino_to_sysino,#function
-sun4v_devino_to_sysino:
+ENTRY(sun4v_devino_to_sysino)
 	mov	HV_FAST_INTR_DEVINO2SYSINO, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 mov	%o1, %o0
-	.size	sun4v_devino_to_sysino, .-sun4v_devino_to_sysino
+ENDPROC(sun4v_devino_to_sysino)
 
 	/* %o0: sysino
 	 *
 	 * returns %o0: intr_enabled (HV_INTR_{DISABLED,ENABLED})
 	 */
-	.globl	sun4v_intr_getenabled
-	.type	sun4v_intr_getenabled,#function
-sun4v_intr_getenabled:
+ENTRY(sun4v_intr_getenabled)
 	mov	HV_FAST_INTR_GETENABLED, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 mov	%o1, %o0
-	.size	sun4v_intr_getenabled, .-sun4v_intr_getenabled
+ENDPROC(sun4v_intr_getenabled)
 
 	/* %o0: sysino
 	 * %o1: intr_enabled (HV_INTR_{DISABLED,ENABLED})
 	 */
-	.globl	sun4v_intr_setenabled
-	.type	sun4v_intr_setenabled,#function
-sun4v_intr_setenabled:
+ENTRY(sun4v_intr_setenabled)
 	mov	HV_FAST_INTR_SETENABLED, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_intr_setenabled, .-sun4v_intr_setenabled
+ENDPROC(sun4v_intr_setenabled)
 
 	/* %o0: sysino
 	 *
 	 * returns %o0: intr_state (HV_INTR_STATE_*)
 	 */
-	.globl	sun4v_intr_getstate
-	.type	sun4v_intr_getstate,#function
-sun4v_intr_getstate:
+ENTRY(sun4v_intr_getstate)
 	mov	HV_FAST_INTR_GETSTATE, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 mov	%o1, %o0
-	.size	sun4v_intr_getstate, .-sun4v_intr_getstate
+ENDPROC(sun4v_intr_getstate)
 
 	/* %o0: sysino
 	 * %o1: intr_state (HV_INTR_STATE_*)
 	 */
-	.globl	sun4v_intr_setstate
-	.type	sun4v_intr_setstate,#function
-sun4v_intr_setstate:
+ENTRY(sun4v_intr_setstate)
 	mov	HV_FAST_INTR_SETSTATE, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_intr_setstate, .-sun4v_intr_setstate
+ENDPROC(sun4v_intr_setstate)
 
 	/* %o0: sysino
 	 *
 	 * returns %o0: cpuid
 	 */
-	.globl	sun4v_intr_gettarget
-	.type	sun4v_intr_gettarget,#function
-sun4v_intr_gettarget:
+ENTRY(sun4v_intr_gettarget)
 	mov	HV_FAST_INTR_GETTARGET, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 mov	%o1, %o0
-	.size	sun4v_intr_gettarget, .-sun4v_intr_gettarget
+ENDPROC(sun4v_intr_gettarget)
 
 	/* %o0: sysino
 	 * %o1: cpuid
 	 */
-	.globl	sun4v_intr_settarget
-	.type	sun4v_intr_settarget,#function
-sun4v_intr_settarget:
+ENTRY(sun4v_intr_settarget)
 	mov	HV_FAST_INTR_SETTARGET, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_intr_settarget, .-sun4v_intr_settarget
+ENDPROC(sun4v_intr_settarget)
 
 	/* %o0:	cpuid
 	 * %o1: pc
@@ -94,37 +80,31 @@ sun4v_intr_settarget:
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_cpu_start
-	.type	sun4v_cpu_start,#function
-sun4v_cpu_start:
+ENTRY(sun4v_cpu_start)
 	mov	HV_FAST_CPU_START, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_cpu_start, .-sun4v_cpu_start
+ENDPROC(sun4v_cpu_start)
 
 	/* %o0:	cpuid
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_cpu_stop
-	.type	sun4v_cpu_stop,#function
-sun4v_cpu_stop:
+ENTRY(sun4v_cpu_stop)
 	mov	HV_FAST_CPU_STOP, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_cpu_stop, .-sun4v_cpu_stop
+ENDPROC(sun4v_cpu_stop)
 
 	/* returns %o0:	status  */
-	.globl	sun4v_cpu_yield
-	.type	sun4v_cpu_yield, #function
-sun4v_cpu_yield:
+ENTRY(sun4v_cpu_yield)
 	mov	HV_FAST_CPU_YIELD, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_cpu_yield, .-sun4v_cpu_yield
+ENDPROC(sun4v_cpu_yield)
 
 	/* %o0:	type
 	 * %o1:	queue paddr
@@ -132,14 +112,12 @@ sun4v_cpu_yield:
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_cpu_qconf
-	.type	sun4v_cpu_qconf,#function
-sun4v_cpu_qconf:
+ENTRY(sun4v_cpu_qconf)
 	mov	HV_FAST_CPU_QCONF, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_cpu_qconf, .-sun4v_cpu_qconf
+ENDPROC(sun4v_cpu_qconf)
 
 	/* %o0:	num cpus in cpu list
 	 * %o1:	cpu list paddr
@@ -147,23 +125,19 @@ sun4v_cpu_qconf:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_cpu_mondo_send
-	.type	sun4v_cpu_mondo_send,#function
-sun4v_cpu_mondo_send:
+ENTRY(sun4v_cpu_mondo_send)
 	mov	HV_FAST_CPU_MONDO_SEND, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_cpu_mondo_send, .-sun4v_cpu_mondo_send
+ENDPROC(sun4v_cpu_mondo_send)
 
 	/* %o0:	CPU ID
 	 *
 	 * returns %o0:	-status if status non-zero, else
 	 *         %o0:	cpu state as HV_CPU_STATE_*
 	 */
-	.globl	sun4v_cpu_state
-	.type	sun4v_cpu_state,#function
-sun4v_cpu_state:
+ENTRY(sun4v_cpu_state)
 	mov	HV_FAST_CPU_STATE, %o5
 	ta	HV_FAST_TRAP
 	brnz,pn	%o0, 1f
@@ -171,7 +145,7 @@ sun4v_cpu_state:
 	mov	%o1, %o0
 1:	retl
 	 nop
-	.size	sun4v_cpu_state, .-sun4v_cpu_state
+ENDPROC(sun4v_cpu_state)
 
 	/* %o0: virtual address
 	 * %o1: must be zero
@@ -180,28 +154,24 @@ sun4v_cpu_state:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_mmu_map_perm_addr
-	.type	sun4v_mmu_map_perm_addr,#function
-sun4v_mmu_map_perm_addr:
+ENTRY(sun4v_mmu_map_perm_addr)
 	mov	HV_FAST_MMU_MAP_PERM_ADDR, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_mmu_map_perm_addr, .-sun4v_mmu_map_perm_addr
+ENDPROC(sun4v_mmu_map_perm_addr)
 
 	/* %o0: number of TSB descriptions
 	 * %o1: TSB descriptions real address
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_mmu_tsb_ctx0
-	.type	sun4v_mmu_tsb_ctx0,#function
-sun4v_mmu_tsb_ctx0:
+ENTRY(sun4v_mmu_tsb_ctx0)
 	mov	HV_FAST_MMU_TSB_CTX0, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_mmu_tsb_ctx0, .-sun4v_mmu_tsb_ctx0
+ENDPROC(sun4v_mmu_tsb_ctx0)
 
 	/* %o0:	API group number
 	 * %o1: pointer to unsigned long major number storage
@@ -209,9 +179,7 @@ sun4v_mmu_tsb_ctx0:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_get_version
-	.type	sun4v_get_version,#function
-sun4v_get_version:
+ENTRY(sun4v_get_version)
 	mov	HV_CORE_GET_VER, %o5
 	mov	%o1, %o3
 	mov	%o2, %o4
@@ -219,7 +187,7 @@ sun4v_get_version:
 	stx	%o1, [%o3]
 	retl
 	 stx	%o2, [%o4]
-	.size	sun4v_get_version, .-sun4v_get_version
+ENDPROC(sun4v_get_version)
 
 	/* %o0: API group number
 	 * %o1: desired major number
@@ -228,51 +196,43 @@ sun4v_get_version:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_set_version
-	.type	sun4v_set_version,#function
-sun4v_set_version:
+ENTRY(sun4v_set_version)
 	mov	HV_CORE_SET_VER, %o5
 	mov	%o3, %o4
 	ta	HV_CORE_TRAP
 	retl
 	 stx	%o1, [%o4]
-	.size	sun4v_set_version, .-sun4v_set_version
+ENDPROC(sun4v_set_version)
 
 	/* %o0: pointer to unsigned long time
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_tod_get
-	.type	sun4v_tod_get,#function
-sun4v_tod_get:
+ENTRY(sun4v_tod_get)
 	mov	%o0, %o4
 	mov	HV_FAST_TOD_GET, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o4]
 	retl
 	 nop
-	.size	sun4v_tod_get, .-sun4v_tod_get
+ENDPROC(sun4v_tod_get)
 
 	/* %o0: time
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_tod_set
-	.type	sun4v_tod_set,#function
-sun4v_tod_set:
+ENTRY(sun4v_tod_set)
 	mov	HV_FAST_TOD_SET, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_tod_set, .-sun4v_tod_set
+ENDPROC(sun4v_tod_set)
 
 	/* %o0: pointer to unsigned long status
 	 *
 	 * returns %o0: signed character
 	 */
-	.globl	sun4v_con_getchar
-	.type	sun4v_con_getchar,#function
-sun4v_con_getchar:
+ENTRY(sun4v_con_getchar)
 	mov	%o0, %o4
 	mov	HV_FAST_CONS_GETCHAR, %o5
 	clr	%o0
@@ -281,20 +241,18 @@ sun4v_con_getchar:
 	stx	%o0, [%o4]
 	retl
 	 sra	%o1, 0, %o0
-	.size	sun4v_con_getchar, .-sun4v_con_getchar
+ENDPROC(sun4v_con_getchar)
 
 	/* %o0: signed long character
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_con_putchar
-	.type	sun4v_con_putchar,#function
-sun4v_con_putchar:
+ENTRY(sun4v_con_putchar)
 	mov	HV_FAST_CONS_PUTCHAR, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 sra	%o0, 0, %o0
-	.size	sun4v_con_putchar, .-sun4v_con_putchar
+ENDPROC(sun4v_con_putchar)
 
 	/* %o0: buffer real address
 	 * %o1: buffer size
@@ -302,9 +260,7 @@ sun4v_con_putchar:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_con_read
-	.type	sun4v_con_read,#function
-sun4v_con_read:
+ENTRY(sun4v_con_read)
 	mov	%o2, %o4
 	mov	HV_FAST_CONS_READ, %o5
 	ta	HV_FAST_TRAP
@@ -318,7 +274,7 @@ sun4v_con_read:
 	stx	%o1, [%o4]
 1:	retl
 	 nop
-	.size	sun4v_con_read, .-sun4v_con_read
+ENDPROC(sun4v_con_read)
 
 	/* %o0: buffer real address
 	 * %o1: buffer size
@@ -326,43 +282,37 @@ sun4v_con_read:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_con_write
-	.type	sun4v_con_write,#function
-sun4v_con_write:
+ENTRY(sun4v_con_write)
 	mov	%o2, %o4
 	mov	HV_FAST_CONS_WRITE, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o4]
 	retl
 	 nop
-	.size	sun4v_con_write, .-sun4v_con_write
+ENDPROC(sun4v_con_write)
 
 	/* %o0:	soft state
 	 * %o1:	address of description string
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_mach_set_soft_state
-	.type	sun4v_mach_set_soft_state,#function
-sun4v_mach_set_soft_state:
+ENTRY(sun4v_mach_set_soft_state)
 	mov	HV_FAST_MACH_SET_SOFT_STATE, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_mach_set_soft_state, .-sun4v_mach_set_soft_state
+ENDPROC(sun4v_mach_set_soft_state)
 
 	/* %o0: exit code
 	 *
 	 * Does not return.
 	 */
-	.globl	sun4v_mach_exit
-	.type	sun4v_mach_exit,#function
-sun4v_mach_exit:
+ENTRY(sun4v_mach_exit)
 	mov	HV_FAST_MACH_EXIT, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_mach_exit, .-sun4v_mach_exit
+ENDPROC(sun4v_mach_exit)
 
 	/* %o0: buffer real address
 	 * %o1: buffer length
@@ -370,44 +320,38 @@ sun4v_mach_exit:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_mach_desc
-	.type	sun4v_mach_desc,#function
-sun4v_mach_desc:
+ENTRY(sun4v_mach_desc)
 	mov	%o2, %o4
 	mov	HV_FAST_MACH_DESC, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o4]
 	retl
 	 nop
-	.size	sun4v_mach_desc, .-sun4v_mach_desc
+ENDPROC(sun4v_mach_desc)
 
 	/* %o0: new timeout in milliseconds
 	 * %o1: pointer to unsigned long orig_timeout
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_mach_set_watchdog
-	.type	sun4v_mach_set_watchdog,#function
-sun4v_mach_set_watchdog:
+ENTRY(sun4v_mach_set_watchdog)
 	mov	%o1, %o4
 	mov	HV_FAST_MACH_SET_WATCHDOG, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o4]
 	retl
 	 nop
-	.size	sun4v_mach_set_watchdog, .-sun4v_mach_set_watchdog
+ENDPROC(sun4v_mach_set_watchdog)
 
 	/* No inputs and does not return.  */
-	.globl	sun4v_mach_sir
-	.type	sun4v_mach_sir,#function
-sun4v_mach_sir:
+ENTRY(sun4v_mach_sir)
 	mov	%o1, %o4
 	mov	HV_FAST_MACH_SIR, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o4]
 	retl
 	 nop
-	.size	sun4v_mach_sir, .-sun4v_mach_sir
+ENDPROC(sun4v_mach_sir)
 
 	/* %o0: channel
 	 * %o1:	ra
@@ -415,14 +359,12 @@ sun4v_mach_sir:
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_ldc_tx_qconf
-	.type	sun4v_ldc_tx_qconf,#function
-sun4v_ldc_tx_qconf:
+ENTRY(sun4v_ldc_tx_qconf)
 	mov	HV_FAST_LDC_TX_QCONF, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_ldc_tx_qconf, .-sun4v_ldc_tx_qconf
+ENDPROC(sun4v_ldc_tx_qconf)
 
 	/* %o0: channel
 	 * %o1:	pointer to unsigned long ra
@@ -430,9 +372,7 @@ sun4v_ldc_tx_qconf:
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_ldc_tx_qinfo
-	.type	sun4v_ldc_tx_qinfo,#function
-sun4v_ldc_tx_qinfo:
+ENTRY(sun4v_ldc_tx_qinfo)
 	mov	%o1, %g1
 	mov	%o2, %g2
 	mov	HV_FAST_LDC_TX_QINFO, %o5
@@ -441,7 +381,7 @@ sun4v_ldc_tx_qinfo:
 	stx	%o2, [%g2]
 	retl
 	 nop
-	.size	sun4v_ldc_tx_qinfo, .-sun4v_ldc_tx_qinfo
+ENDPROC(sun4v_ldc_tx_qinfo)
 
 	/* %o0: channel
 	 * %o1:	pointer to unsigned long head_off
@@ -450,9 +390,7 @@ sun4v_ldc_tx_qinfo:
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_ldc_tx_get_state
-	.type	sun4v_ldc_tx_get_state,#function
-sun4v_ldc_tx_get_state:
+ENTRY(sun4v_ldc_tx_get_state)
 	mov	%o1, %g1
 	mov	%o2, %g2
 	mov	%o3, %g3
@@ -463,21 +401,19 @@ sun4v_ldc_tx_get_state:
 	stx	%o3, [%g3]
 	retl
 	 nop
-	.size	sun4v_ldc_tx_get_state, .-sun4v_ldc_tx_get_state
+ENDPROC(sun4v_ldc_tx_get_state)
 
 	/* %o0: channel
 	 * %o1:	tail_off
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_ldc_tx_set_qtail
-	.type	sun4v_ldc_tx_set_qtail,#function
-sun4v_ldc_tx_set_qtail:
+ENTRY(sun4v_ldc_tx_set_qtail)
 	mov	HV_FAST_LDC_TX_SET_QTAIL, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_ldc_tx_set_qtail, .-sun4v_ldc_tx_set_qtail
+ENDPROC(sun4v_ldc_tx_set_qtail)
 
 	/* %o0: channel
 	 * %o1:	ra
@@ -485,14 +421,12 @@ sun4v_ldc_tx_set_qtail:
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_ldc_rx_qconf
-	.type	sun4v_ldc_rx_qconf,#function
-sun4v_ldc_rx_qconf:
+ENTRY(sun4v_ldc_rx_qconf)
 	mov	HV_FAST_LDC_RX_QCONF, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_ldc_rx_qconf, .-sun4v_ldc_rx_qconf
+ENDPROC(sun4v_ldc_rx_qconf)
 
 	/* %o0: channel
 	 * %o1:	pointer to unsigned long ra
@@ -500,9 +434,7 @@ sun4v_ldc_rx_qconf:
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_ldc_rx_qinfo
-	.type	sun4v_ldc_rx_qinfo,#function
-sun4v_ldc_rx_qinfo:
+ENTRY(sun4v_ldc_rx_qinfo)
 	mov	%o1, %g1
 	mov	%o2, %g2
 	mov	HV_FAST_LDC_RX_QINFO, %o5
@@ -511,7 +443,7 @@ sun4v_ldc_rx_qinfo:
 	stx	%o2, [%g2]
 	retl
 	 nop
-	.size	sun4v_ldc_rx_qinfo, .-sun4v_ldc_rx_qinfo
+ENDPROC(sun4v_ldc_rx_qinfo)
 
 	/* %o0: channel
 	 * %o1:	pointer to unsigned long head_off
@@ -520,9 +452,7 @@ sun4v_ldc_rx_qinfo:
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_ldc_rx_get_state
-	.type	sun4v_ldc_rx_get_state,#function
-sun4v_ldc_rx_get_state:
+ENTRY(sun4v_ldc_rx_get_state)
 	mov	%o1, %g1
 	mov	%o2, %g2
 	mov	%o3, %g3
@@ -533,21 +463,19 @@ sun4v_ldc_rx_get_state:
 	stx	%o3, [%g3]
 	retl
 	 nop
-	.size	sun4v_ldc_rx_get_state, .-sun4v_ldc_rx_get_state
+ENDPROC(sun4v_ldc_rx_get_state)
 
 	/* %o0: channel
 	 * %o1:	head_off
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_ldc_rx_set_qhead
-	.type	sun4v_ldc_rx_set_qhead,#function
-sun4v_ldc_rx_set_qhead:
+ENTRY(sun4v_ldc_rx_set_qhead)
 	mov	HV_FAST_LDC_RX_SET_QHEAD, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_ldc_rx_set_qhead, .-sun4v_ldc_rx_set_qhead
+ENDPROC(sun4v_ldc_rx_set_qhead)
 
 	/* %o0: channel
 	 * %o1:	ra
@@ -555,14 +483,12 @@ sun4v_ldc_rx_set_qhead:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_ldc_set_map_table
-	.type	sun4v_ldc_set_map_table,#function
-sun4v_ldc_set_map_table:
+ENTRY(sun4v_ldc_set_map_table)
 	mov	HV_FAST_LDC_SET_MAP_TABLE, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_ldc_set_map_table, .-sun4v_ldc_set_map_table
+ENDPROC(sun4v_ldc_set_map_table)
 
 	/* %o0: channel
 	 * %o1:	pointer to unsigned long ra
@@ -570,9 +496,7 @@ sun4v_ldc_set_map_table:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_ldc_get_map_table
-	.type	sun4v_ldc_get_map_table,#function
-sun4v_ldc_get_map_table:
+ENTRY(sun4v_ldc_get_map_table)
 	mov	%o1, %g1
 	mov	%o2, %g2
 	mov	HV_FAST_LDC_GET_MAP_TABLE, %o5
@@ -581,7 +505,7 @@ sun4v_ldc_get_map_table:
 	stx	%o2, [%g2]
 	retl
 	 nop
-	.size	sun4v_ldc_get_map_table, .-sun4v_ldc_get_map_table
+ENDPROC(sun4v_ldc_get_map_table)
 
 	/* %o0:	channel
 	 * %o1:	dir_code
@@ -592,16 +516,14 @@ sun4v_ldc_get_map_table:
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_ldc_copy
-	.type	sun4v_ldc_copy,#function
-sun4v_ldc_copy:
+ENTRY(sun4v_ldc_copy)
 	mov	%o5, %g1
 	mov	HV_FAST_LDC_COPY, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%g1]
 	retl
 	 nop
-	.size	sun4v_ldc_copy, .-sun4v_ldc_copy
+ENDPROC(sun4v_ldc_copy)
 
 	/* %o0:	channel
 	 * %o1:	cookie
@@ -610,9 +532,7 @@ sun4v_ldc_copy:
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_ldc_mapin
-	.type	sun4v_ldc_mapin,#function
-sun4v_ldc_mapin:
+ENTRY(sun4v_ldc_mapin)
 	mov	%o2, %g1
 	mov	%o3, %g2
 	mov	HV_FAST_LDC_MAPIN, %o5
@@ -621,20 +541,18 @@ sun4v_ldc_mapin:
 	stx	%o2, [%g2]
 	retl
 	 nop
-	.size	sun4v_ldc_mapin, .-sun4v_ldc_mapin
+ENDPROC(sun4v_ldc_mapin)
 
 	/* %o0:	ra
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_ldc_unmap
-	.type	sun4v_ldc_unmap,#function
-sun4v_ldc_unmap:
+ENTRY(sun4v_ldc_unmap)
 	mov	HV_FAST_LDC_UNMAP, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_ldc_unmap, .-sun4v_ldc_unmap
+ENDPROC(sun4v_ldc_unmap)
 
 	/* %o0: channel
 	 * %o1:	cookie
@@ -642,14 +560,12 @@ sun4v_ldc_unmap:
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_ldc_revoke
-	.type	sun4v_ldc_revoke,#function
-sun4v_ldc_revoke:
+ENTRY(sun4v_ldc_revoke)
 	mov	HV_FAST_LDC_REVOKE, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_ldc_revoke, .-sun4v_ldc_revoke
+ENDPROC(sun4v_ldc_revoke)
 
 	/* %o0: device handle
 	 * %o1:	device INO
@@ -657,16 +573,14 @@ sun4v_ldc_revoke:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_vintr_get_cookie
-	.type	sun4v_vintr_get_cookie,#function
-sun4v_vintr_get_cookie:
+ENTRY(sun4v_vintr_get_cookie)
 	mov	%o2, %g1
 	mov	HV_FAST_VINTR_GET_COOKIE, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%g1]
 	retl
 	 nop
-	.size	sun4v_vintr_get_cookie, .-sun4v_vintr_get_cookie
+ENDPROC(sun4v_vintr_get_cookie)
 
 	/* %o0: device handle
 	 * %o1:	device INO
@@ -674,14 +588,12 @@ sun4v_vintr_get_cookie:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_vintr_set_cookie
-	.type	sun4v_vintr_set_cookie,#function
-sun4v_vintr_set_cookie:
+ENTRY(sun4v_vintr_set_cookie)
 	mov	HV_FAST_VINTR_SET_COOKIE, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_vintr_set_cookie, .-sun4v_vintr_set_cookie
+ENDPROC(sun4v_vintr_set_cookie)
 
 	/* %o0: device handle
 	 * %o1:	device INO
@@ -689,16 +601,14 @@ sun4v_vintr_set_cookie:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_vintr_get_valid
-	.type	sun4v_vintr_get_valid,#function
-sun4v_vintr_get_valid:
+ENTRY(sun4v_vintr_get_valid)
 	mov	%o2, %g1
 	mov	HV_FAST_VINTR_GET_VALID, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%g1]
 	retl
 	 nop
-	.size	sun4v_vintr_get_valid, .-sun4v_vintr_get_valid
+ENDPROC(sun4v_vintr_get_valid)
 
 	/* %o0: device handle
 	 * %o1:	device INO
@@ -706,14 +616,12 @@ sun4v_vintr_get_valid:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_vintr_set_valid
-	.type	sun4v_vintr_set_valid,#function
-sun4v_vintr_set_valid:
+ENTRY(sun4v_vintr_set_valid)
 	mov	HV_FAST_VINTR_SET_VALID, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_vintr_set_valid, .-sun4v_vintr_set_valid
+ENDPROC(sun4v_vintr_set_valid)
 
 	/* %o0: device handle
 	 * %o1:	device INO
@@ -721,16 +629,14 @@ sun4v_vintr_set_valid:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_vintr_get_state
-	.type	sun4v_vintr_get_state,#function
-sun4v_vintr_get_state:
+ENTRY(sun4v_vintr_get_state)
 	mov	%o2, %g1
 	mov	HV_FAST_VINTR_GET_STATE, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%g1]
 	retl
 	 nop
-	.size	sun4v_vintr_get_state, .-sun4v_vintr_get_state
+ENDPROC(sun4v_vintr_get_state)
 
 	/* %o0: device handle
 	 * %o1:	device INO
@@ -738,14 +644,12 @@ sun4v_vintr_get_state:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_vintr_set_state
-	.type	sun4v_vintr_set_state,#function
-sun4v_vintr_set_state:
+ENTRY(sun4v_vintr_set_state)
 	mov	HV_FAST_VINTR_SET_STATE, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_vintr_set_state, .-sun4v_vintr_set_state
+ENDPROC(sun4v_vintr_set_state)
 
 	/* %o0: device handle
 	 * %o1:	device INO
@@ -753,16 +657,14 @@ sun4v_vintr_set_state:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_vintr_get_target
-	.type	sun4v_vintr_get_target,#function
-sun4v_vintr_get_target:
+ENTRY(sun4v_vintr_get_target)
 	mov	%o2, %g1
 	mov	HV_FAST_VINTR_GET_TARGET, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%g1]
 	retl
 	 nop
-	.size	sun4v_vintr_get_target, .-sun4v_vintr_get_target
+ENDPROC(sun4v_vintr_get_target)
 
 	/* %o0: device handle
 	 * %o1:	device INO
@@ -770,14 +672,12 @@ sun4v_vintr_get_target:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	sun4v_vintr_set_target
-	.type	sun4v_vintr_set_target,#function
-sun4v_vintr_set_target:
+ENTRY(sun4v_vintr_set_target)
 	mov	HV_FAST_VINTR_SET_TARGET, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_vintr_set_target, .-sun4v_vintr_set_target
+ENDPROC(sun4v_vintr_set_target)
 
 	/* %o0: NCS sub-function
 	 * %o1:	sub-function arg real-address
@@ -785,18 +685,14 @@ sun4v_vintr_set_target:
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	sun4v_ncs_request
-	.type	sun4v_ncs_request,#function
-sun4v_ncs_request:
+ENTRY(sun4v_ncs_request)
 	mov	HV_FAST_NCS_REQUEST, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_ncs_request, .-sun4v_ncs_request
+ENDPROC(sun4v_ncs_request)
 
-	.globl	sun4v_svc_send
-	.type	sun4v_svc_send,#function
-sun4v_svc_send:
+ENTRY(sun4v_svc_send)
 	save	%sp, -192, %sp
 	mov	%i0, %o0
 	mov	%i1, %o1
@@ -806,11 +702,9 @@ sun4v_svc_send:
 	stx	%o1, [%i3]
 	ret
 	restore
-	.size	sun4v_svc_send, .-sun4v_svc_send
+ENDPROC(sun4v_svc_send)
 
-	.globl	sun4v_svc_recv
-	.type	sun4v_svc_recv,#function
-sun4v_svc_recv:
+ENTRY(sun4v_svc_recv)
 	save	%sp, -192, %sp
 	mov	%i0, %o0
 	mov	%i1, %o1
@@ -820,62 +714,50 @@ sun4v_svc_recv:
 	stx	%o1, [%i3]
 	ret
 	restore
-	.size	sun4v_svc_recv, .-sun4v_svc_recv
+ENDPROC(sun4v_svc_recv)
 
-	.globl	sun4v_svc_getstatus
-	.type	sun4v_svc_getstatus,#function
-sun4v_svc_getstatus:
+ENTRY(sun4v_svc_getstatus)
 	mov	HV_FAST_SVC_GETSTATUS, %o5
 	mov	%o1, %o4
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o4]
 	retl
 	 nop
-	.size	sun4v_svc_getstatus, .-sun4v_svc_getstatus
+ENDPROC(sun4v_svc_getstatus)
 
-	.globl	sun4v_svc_setstatus
-	.type	sun4v_svc_setstatus,#function
-sun4v_svc_setstatus:
+ENTRY(sun4v_svc_setstatus)
 	mov	HV_FAST_SVC_SETSTATUS, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_svc_setstatus, .-sun4v_svc_setstatus
+ENDPROC(sun4v_svc_setstatus)
 
-	.globl	sun4v_svc_clrstatus
-	.type	sun4v_svc_clrstatus,#function
-sun4v_svc_clrstatus:
+ENTRY(sun4v_svc_clrstatus)
 	mov	HV_FAST_SVC_CLRSTATUS, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_svc_clrstatus, .-sun4v_svc_clrstatus
+ENDPROC(sun4v_svc_clrstatus)
 
-	.globl	sun4v_mmustat_conf
-	.type	sun4v_mmustat_conf,#function
-sun4v_mmustat_conf:
+ENTRY(sun4v_mmustat_conf)
 	mov	%o1, %o4
 	mov	HV_FAST_MMUSTAT_CONF, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o4]
 	retl
 	 nop
-	.size	sun4v_mmustat_conf, .-sun4v_mmustat_conf
+ENDPROC(sun4v_mmustat_conf)
 
-	.globl	sun4v_mmustat_info
-	.type	sun4v_mmustat_info,#function
-sun4v_mmustat_info:
+ENTRY(sun4v_mmustat_info)
 	mov	%o0, %o4
 	mov	HV_FAST_MMUSTAT_INFO, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o4]
 	retl
 	 nop
-	.size	sun4v_mmustat_info, .-sun4v_mmustat_info
+ENDPROC(sun4v_mmustat_info)
 
-	.globl	sun4v_mmu_demap_all
-	.type	sun4v_mmu_demap_all,#function
-sun4v_mmu_demap_all:
+ENTRY(sun4v_mmu_demap_all)
 	clr	%o0
 	clr	%o1
 	mov	HV_MMU_ALL, %o2
@@ -883,4 +765,4 @@ sun4v_mmu_demap_all:
 	ta	HV_FAST_TRAP
 	retl
 	 nop
-	.size	sun4v_mmu_demap_all, .-sun4v_mmu_demap_all
+ENDPROC(sun4v_mmu_demap_all)
diff --git a/arch/sparc64/kernel/irq.c b/arch/sparc64/kernel/irq.c
index 7495bc774685..52fc836f464d 100644
--- a/arch/sparc64/kernel/irq.c
+++ b/arch/sparc64/kernel/irq.c
@@ -29,7 +29,6 @@
 #include <asm/system.h>
 #include <asm/irq.h>
 #include <asm/io.h>
-#include <asm/sbus.h>
 #include <asm/iommu.h>
 #include <asm/upa.h>
 #include <asm/oplib.h>
diff --git a/arch/sparc64/kernel/of_device.c b/arch/sparc64/kernel/of_device.c
index 100ebd527499..0f616ae3246c 100644
--- a/arch/sparc64/kernel/of_device.c
+++ b/arch/sparc64/kernel/of_device.c
@@ -55,15 +55,38 @@ struct of_device *of_find_device_by_node(struct device_node *dp)
 }
 EXPORT_SYMBOL(of_find_device_by_node);
 
-#ifdef CONFIG_PCI
-struct bus_type ebus_bus_type;
-EXPORT_SYMBOL(ebus_bus_type);
-#endif
+unsigned int irq_of_parse_and_map(struct device_node *node, int index)
+{
+	struct of_device *op = of_find_device_by_node(node);
+
+	if (!op || index >= op->num_irqs)
+		return 0;
+
+	return op->irqs[index];
+}
+EXPORT_SYMBOL(irq_of_parse_and_map);
+
+/* Take the archdata values for IOMMU, STC, and HOSTDATA found in
+ * BUS and propagate to all child of_device objects.
+ */
+void of_propagate_archdata(struct of_device *bus)
+{
+	struct dev_archdata *bus_sd = &bus->dev.archdata;
+	struct device_node *bus_dp = bus->node;
+	struct device_node *dp;
 
-#ifdef CONFIG_SBUS
-struct bus_type sbus_bus_type;
-EXPORT_SYMBOL(sbus_bus_type);
-#endif
+	for (dp = bus_dp->child; dp; dp = dp->sibling) {
+		struct of_device *op = of_find_device_by_node(dp);
+
+		op->dev.archdata.iommu = bus_sd->iommu;
+		op->dev.archdata.stc = bus_sd->stc;
+		op->dev.archdata.host_controller = bus_sd->host_controller;
+		op->dev.archdata.numa_node = bus_sd->numa_node;
+
+		if (dp->child)
+			of_propagate_archdata(op);
+	}
+}
 
 struct bus_type of_platform_bus_type;
 EXPORT_SYMBOL(of_platform_bus_type);
@@ -378,8 +401,7 @@ static int __init build_one_resource(struct device_node *parent,
 				     int na, int ns, int pna)
 {
 	const u32 *ranges;
-	unsigned int rlen;
-	int rone;
+	int rone, rlen;
 
 	ranges = of_get_property(parent, "ranges", &rlen);
 	if (ranges == NULL || rlen == 0) {
@@ -421,8 +443,17 @@ static int __init use_1to1_mapping(struct device_node *pp)
 
 	/* If the parent is the dma node of an ISA bus, pass
 	 * the translation up to the root.
+	 *
+	 * Some SBUS devices use intermediate nodes to express
+	 * hierarchy within the device itself.  These aren't
+	 * real bus nodes, and don't have a 'ranges' property.
+	 * But, we should still pass the translation work up
+	 * to the SBUS itself.
 	 */
-	if (!strcmp(pp->name, "dma"))
+	if (!strcmp(pp->name, "dma") ||
+	    !strcmp(pp->name, "espdma") ||
+	    !strcmp(pp->name, "ledma") ||
+	    !strcmp(pp->name, "lebuffer"))
 		return 0;
 
 	/* Similarly for all PCI bridges, if we get this far
@@ -844,15 +875,6 @@ static int __init of_bus_driver_init(void)
 	int err;
 
 	err = of_bus_type_init(&of_platform_bus_type, "of");
-#ifdef CONFIG_PCI
-	if (!err)
-		err = of_bus_type_init(&ebus_bus_type, "ebus");
-#endif
-#ifdef CONFIG_SBUS
-	if (!err)
-		err = of_bus_type_init(&sbus_bus_type, "sbus");
-#endif
-
 	if (!err)
 		scan_of_devices();
 
diff --git a/arch/sparc64/kernel/pci.c b/arch/sparc64/kernel/pci.c
index 80dad76f8b81..242ac1ccae7d 100644
--- a/arch/sparc64/kernel/pci.c
+++ b/arch/sparc64/kernel/pci.c
@@ -18,32 +18,17 @@
 #include <linux/msi.h>
 #include <linux/irq.h>
 #include <linux/init.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 #include <asm/irq.h>
-#include <asm/ebus.h>
 #include <asm/prom.h>
 #include <asm/apb.h>
 
 #include "pci_impl.h"
 
-#ifndef CONFIG_PCI
-/* A "nop" PCI implementation. */
-asmlinkage int sys_pciconfig_read(unsigned long bus, unsigned long dfn,
-				  unsigned long off, unsigned long len,
-				  unsigned char *buf)
-{
-	return 0;
-}
-asmlinkage int sys_pciconfig_write(unsigned long bus, unsigned long dfn,
-				   unsigned long off, unsigned long len,
-				   unsigned char *buf)
-{
-	return 0;
-}
-#else
-
 /* List of all PCI controllers found in the system. */
 struct pci_pbm_info *pci_pbm_root = NULL;
 
@@ -179,97 +164,6 @@ void pci_config_write32(u32 *addr, u32 val)
 	spin_unlock_irqrestore(&pci_poke_lock, flags);
 }
 
-/* Probe for all PCI controllers in the system. */
-extern void sabre_init(struct device_node *, const char *);
-extern void psycho_init(struct device_node *, const char *);
-extern void schizo_init(struct device_node *, const char *);
-extern void schizo_plus_init(struct device_node *, const char *);
-extern void tomatillo_init(struct device_node *, const char *);
-extern void sun4v_pci_init(struct device_node *, const char *);
-extern void fire_pci_init(struct device_node *, const char *);
-
-static struct {
-	char *model_name;
-	void (*init)(struct device_node *, const char *);
-} pci_controller_table[] __initdata = {
-	{ "SUNW,sabre", sabre_init },
-	{ "pci108e,a000", sabre_init },
-	{ "pci108e,a001", sabre_init },
-	{ "SUNW,psycho", psycho_init },
-	{ "pci108e,8000", psycho_init },
-	{ "SUNW,schizo", schizo_init },
-	{ "pci108e,8001", schizo_init },
-	{ "SUNW,schizo+", schizo_plus_init },
-	{ "pci108e,8002", schizo_plus_init },
-	{ "SUNW,tomatillo", tomatillo_init },
-	{ "pci108e,a801", tomatillo_init },
-	{ "SUNW,sun4v-pci", sun4v_pci_init },
-	{ "pciex108e,80f0", fire_pci_init },
-};
-#define PCI_NUM_CONTROLLER_TYPES	ARRAY_SIZE(pci_controller_table)
-
-static int __init pci_controller_init(const char *model_name, int namelen, struct device_node *dp)
-{
-	int i;
-
-	for (i = 0; i < PCI_NUM_CONTROLLER_TYPES; i++) {
-		if (!strncmp(model_name,
-			     pci_controller_table[i].model_name,
-			     namelen)) {
-			pci_controller_table[i].init(dp, model_name);
-			return 1;
-		}
-	}
-
-	return 0;
-}
-
-static int __init pci_controller_scan(int (*handler)(const char *, int, struct device_node *))
-{
-	struct device_node *dp;
-	int count = 0;
-
-	for_each_node_by_name(dp, "pci") {
-		struct property *prop;
-		int len;
-
-		prop = of_find_property(dp, "model", &len);
-		if (!prop)
-			prop = of_find_property(dp, "compatible", &len);
-
-		if (prop) {
-			const char *model = prop->value;
-			int item_len = 0;
-
-			/* Our value may be a multi-valued string in the
-			 * case of some compatible properties. For sanity,
-			 * only try the first one.
-			 */
-			while (model[item_len] && len) {
-				len--;
-				item_len++;
-			}
-
-			if (handler(model, item_len, dp))
-				count++;
-		}
-	}
-
-	return count;
-}
-
-/* Find each controller in the system, attach and initialize
- * software state structure for each and link into the
- * pci_pbm_root.  Setup the controller enough such
- * that bus scanning can be done.
- */
-static void __init pci_controller_probe(void)
-{
-	printk("PCI: Probing for controllers.\n");
-
-	pci_controller_scan(pci_controller_init);
-}
-
 static int ofpci_verbose;
 
 static int __init ofpci_debug(char *str)
@@ -348,11 +242,12 @@ static void pci_parse_of_addrs(struct of_device *op,
 	}
 }
 
-struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm,
-				  struct device_node *node,
-				  struct pci_bus *bus, int devfn)
+static struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm,
+					 struct device_node *node,
+					 struct pci_bus *bus, int devfn)
 {
 	struct dev_archdata *sd;
+	struct of_device *op;
 	struct pci_dev *dev;
 	const char *type;
 	u32 class;
@@ -366,14 +261,17 @@ struct pci_dev *of_create_pci_dev(struct pci_pbm_info *pbm,
 	sd->stc = &pbm->stc;
 	sd->host_controller = pbm;
 	sd->prom_node = node;
-	sd->op = of_find_device_by_node(node);
+	sd->op = op = of_find_device_by_node(node);
 	sd->numa_node = pbm->numa_node;
 
-	sd = &sd->op->dev.archdata;
+	sd = &op->dev.archdata;
 	sd->iommu = pbm->iommu;
 	sd->stc = &pbm->stc;
 	sd->numa_node = pbm->numa_node;
 
+	if (!strcmp(node->name, "ebus"))
+		of_propagate_archdata(op);
+
 	type = of_get_property(node, "device_type", NULL);
 	if (type == NULL)
 		type = "";
@@ -775,15 +673,15 @@ static void __devinit pci_bus_register_of_sysfs(struct pci_bus *bus)
 		pci_bus_register_of_sysfs(child_bus);
 }
 
-struct pci_bus * __devinit pci_scan_one_pbm(struct pci_pbm_info *pbm)
+struct pci_bus * __devinit pci_scan_one_pbm(struct pci_pbm_info *pbm,
+					    struct device *parent)
 {
-	struct device_node *node = pbm->prom_node;
+	struct device_node *node = pbm->op->node;
 	struct pci_bus *bus;
 
 	printk("PCI: Scanning PBM %s\n", node->full_name);
 
-	/* XXX parent device? XXX */
-	bus = pci_create_bus(NULL, pbm->pci_first_busno, pbm->pci_ops, pbm);
+	bus = pci_create_bus(parent, pbm->pci_first_busno, pbm->pci_ops, pbm);
 	if (!bus) {
 		printk(KERN_ERR "Failed to create bus for %s\n",
 		       node->full_name);
@@ -802,32 +700,6 @@ struct pci_bus * __devinit pci_scan_one_pbm(struct pci_pbm_info *pbm)
 	return bus;
 }
 
-static void __init pci_scan_each_controller_bus(void)
-{
-	struct pci_pbm_info *pbm;
-
-	for (pbm = pci_pbm_root; pbm; pbm = pbm->next)
-		pbm->scan_bus(pbm);
-}
-
-extern void power_init(void);
-
-static int __init pcibios_init(void)
-{
-	pci_controller_probe();
-	if (pci_pbm_root == NULL)
-		return 0;
-
-	pci_scan_each_controller_bus();
-
-	ebus_init();
-	power_init();
-
-	return 0;
-}
-
-subsys_initcall(pcibios_init);
-
 void __devinit pcibios_fixup_bus(struct pci_bus *pbus)
 {
 	struct pci_pbm_info *pbm = pbus->sysdata;
@@ -1105,14 +977,14 @@ int pcibus_to_node(struct pci_bus *pbus)
 EXPORT_SYMBOL(pcibus_to_node);
 #endif
 
-/* Return the domain nuber for this pci bus */
+/* Return the domain number for this pci bus */
 
 int pci_domain_nr(struct pci_bus *pbus)
 {
 	struct pci_pbm_info *pbm = pbus->sysdata;
 	int ret;
 
-	if (pbm == NULL || pbm->parent == NULL) {
+	if (!pbm) {
 		ret = -ENXIO;
 	} else {
 		ret = pbm->index;
@@ -1126,7 +998,7 @@ EXPORT_SYMBOL(pci_domain_nr);
 int arch_setup_msi_irq(struct pci_dev *pdev, struct msi_desc *desc)
 {
 	struct pci_pbm_info *pbm = pdev->dev.archdata.host_controller;
-	int virt_irq;
+	unsigned int virt_irq;
 
 	if (!pbm->setup_msi_irq)
 		return -EINVAL;
@@ -1140,10 +1012,8 @@ void arch_teardown_msi_irq(unsigned int virt_irq)
 	struct pci_dev *pdev = entry->dev;
 	struct pci_pbm_info *pbm = pdev->dev.archdata.host_controller;
 
-	if (!pbm->teardown_msi_irq)
-		return;
-
-	return pbm->teardown_msi_irq(virt_irq, pdev);
+	if (pbm->teardown_msi_irq)
+		pbm->teardown_msi_irq(virt_irq, pdev);
 }
 #endif /* !(CONFIG_PCI_MSI) */
 
@@ -1215,5 +1085,3 @@ void pci_resource_to_user(const struct pci_dev *pdev, int bar,
 	*start = rp->start - offset;
 	*end = rp->end - offset;
 }
-
-#endif /* !(CONFIG_PCI) */
diff --git a/arch/sparc64/kernel/pci_common.c b/arch/sparc64/kernel/pci_common.c
index 09a5ec200c61..23b88082d0b2 100644
--- a/arch/sparc64/kernel/pci_common.c
+++ b/arch/sparc64/kernel/pci_common.c
@@ -314,12 +314,12 @@ struct pci_ops sun4v_pci_ops = {
 
 void pci_get_pbm_props(struct pci_pbm_info *pbm)
 {
-	const u32 *val = of_get_property(pbm->prom_node, "bus-range", NULL);
+	const u32 *val = of_get_property(pbm->op->node, "bus-range", NULL);
 
 	pbm->pci_first_busno = val[0];
 	pbm->pci_last_busno = val[1];
 
-	val = of_get_property(pbm->prom_node, "ino-bitmap", NULL);
+	val = of_get_property(pbm->op->node, "ino-bitmap", NULL);
 	if (val) {
 		pbm->ino_bitmap = (((u64)val[1] << 32UL) |
 				   ((u64)val[0] <<  0UL));
@@ -365,7 +365,7 @@ static void pci_register_legacy_regions(struct resource *io_res,
 
 static void pci_register_iommu_region(struct pci_pbm_info *pbm)
 {
-	const u32 *vdma = of_get_property(pbm->prom_node, "virtual-dma", NULL);
+	const u32 *vdma = of_get_property(pbm->op->node, "virtual-dma", NULL);
 
 	if (vdma) {
 		struct resource *rp = kmalloc(sizeof(*rp), GFP_KERNEL);
@@ -389,7 +389,7 @@ void pci_determine_mem_io_space(struct pci_pbm_info *pbm)
 	int num_pbm_ranges;
 
 	saw_mem = saw_io = 0;
-	pbm_ranges = of_get_property(pbm->prom_node, "ranges", &i);
+	pbm_ranges = of_get_property(pbm->op->node, "ranges", &i);
 	if (!pbm_ranges) {
 		prom_printf("PCI: Fatal error, missing PBM ranges property "
 			    " for %s\n",
diff --git a/arch/sparc64/kernel/pci_fire.c b/arch/sparc64/kernel/pci_fire.c
index d23bb6f53cda..9462b68f4894 100644
--- a/arch/sparc64/kernel/pci_fire.c
+++ b/arch/sparc64/kernel/pci_fire.c
@@ -8,34 +8,16 @@
 #include <linux/init.h>
 #include <linux/msi.h>
 #include <linux/irq.h>
+#include <linux/of_device.h>
 
-#include <asm/oplib.h>
 #include <asm/prom.h>
 #include <asm/irq.h>
+#include <asm/upa.h>
 
 #include "pci_impl.h"
 
-#define fire_read(__reg) \
-({	u64 __ret; \
-	__asm__ __volatile__("ldxa [%1] %2, %0" \
-			     : "=r" (__ret) \
-			     : "r" (__reg), "i" (ASI_PHYS_BYPASS_EC_E) \
-			     : "memory"); \
-	__ret; \
-})
-#define fire_write(__reg, __val) \
-	__asm__ __volatile__("stxa %0, [%1] %2" \
-			     : /* no outputs */ \
-			     : "r" (__val), "r" (__reg), \
-			       "i" (ASI_PHYS_BYPASS_EC_E) \
-			     : "memory")
-
-static void __init pci_fire_scan_bus(struct pci_pbm_info *pbm)
-{
-	pbm->pci_bus = pci_scan_one_pbm(pbm);
-
-	/* XXX register error interrupt handlers XXX */
-}
+#define DRIVER_NAME	"fire"
+#define PFX		DRIVER_NAME ": "
 
 #define FIRE_IOMMU_CONTROL	0x40000UL
 #define FIRE_IOMMU_TSBBASE	0x40008UL
@@ -69,21 +51,21 @@ static int pci_fire_pbm_iommu_init(struct pci_pbm_info *pbm)
 	/*
 	 * Invalidate TLB Entries.
 	 */
-	fire_write(iommu->iommu_flushinv, ~(u64)0);
+	upa_writeq(~(u64)0, iommu->iommu_flushinv);
 
 	err = iommu_table_init(iommu, tsbsize * 8 * 1024, vdma[0], dma_mask,
 			       pbm->numa_node);
 	if (err)
 		return err;
 
-	fire_write(iommu->iommu_tsbbase, __pa(iommu->page_table) | 0x7UL);
+	upa_writeq(__pa(iommu->page_table) | 0x7UL, iommu->iommu_tsbbase);
 
-	control = fire_read(iommu->iommu_control);
+	control = upa_readq(iommu->iommu_control);
 	control |= (0x00000400 /* TSB cache snoop enable */	|
 		    0x00000300 /* Cache mode */			|
 		    0x00000002 /* Bypass enable */		|
 		    0x00000001 /* Translation enable */);
-	fire_write(iommu->iommu_control, control);
+	upa_writeq(control, iommu->iommu_control);
 
 	return 0;
 }
@@ -165,7 +147,7 @@ struct pci_msiq_entry {
 static int pci_fire_get_head(struct pci_pbm_info *pbm, unsigned long msiqid,
 			     unsigned long *head)
 {
-	*head = fire_read(pbm->pbm_regs + EVENT_QUEUE_HEAD(msiqid));
+	*head = upa_readq(pbm->pbm_regs + EVENT_QUEUE_HEAD(msiqid));
 	return 0;
 }
 
@@ -191,8 +173,7 @@ static int pci_fire_dequeue_msi(struct pci_pbm_info *pbm, unsigned long msiqid,
 	*msi = msi_num = ((ep->word0 & MSIQ_WORD0_DATA0) >>
 			  MSIQ_WORD0_DATA0_SHIFT);
 
-	fire_write(pbm->pbm_regs + MSI_CLEAR(msi_num),
-		   MSI_CLEAR_EQWR_N);
+	upa_writeq(MSI_CLEAR_EQWR_N, pbm->pbm_regs + MSI_CLEAR(msi_num));
 
 	/* Clear the entry.  */
 	ep->word0 &= ~MSIQ_WORD0_FMT_TYPE;
@@ -208,7 +189,7 @@ static int pci_fire_dequeue_msi(struct pci_pbm_info *pbm, unsigned long msiqid,
 static int pci_fire_set_head(struct pci_pbm_info *pbm, unsigned long msiqid,
 			     unsigned long head)
 {
-	fire_write(pbm->pbm_regs + EVENT_QUEUE_HEAD(msiqid), head);
+	upa_writeq(head, pbm->pbm_regs + EVENT_QUEUE_HEAD(msiqid));
 	return 0;
 }
 
@@ -217,17 +198,16 @@ static int pci_fire_msi_setup(struct pci_pbm_info *pbm, unsigned long msiqid,
 {
 	u64 val;
 
-	val = fire_read(pbm->pbm_regs + MSI_MAP(msi));
+	val = upa_readq(pbm->pbm_regs + MSI_MAP(msi));
 	val &= ~(MSI_MAP_EQNUM);
 	val |= msiqid;
-	fire_write(pbm->pbm_regs + MSI_MAP(msi), val);
+	upa_writeq(val, pbm->pbm_regs + MSI_MAP(msi));
 
-	fire_write(pbm->pbm_regs + MSI_CLEAR(msi),
-		   MSI_CLEAR_EQWR_N);
+	upa_writeq(MSI_CLEAR_EQWR_N, pbm->pbm_regs + MSI_CLEAR(msi));
 
-	val = fire_read(pbm->pbm_regs + MSI_MAP(msi));
+	val = upa_readq(pbm->pbm_regs + MSI_MAP(msi));
 	val |= MSI_MAP_VALID;
-	fire_write(pbm->pbm_regs + MSI_MAP(msi), val);
+	upa_writeq(val, pbm->pbm_regs + MSI_MAP(msi));
 
 	return 0;
 }
@@ -237,12 +217,12 @@ static int pci_fire_msi_teardown(struct pci_pbm_info *pbm, unsigned long msi)
 	unsigned long msiqid;
 	u64 val;
 
-	val = fire_read(pbm->pbm_regs + MSI_MAP(msi));
+	val = upa_readq(pbm->pbm_regs + MSI_MAP(msi));
 	msiqid = (val & MSI_MAP_EQNUM);
 
 	val &= ~MSI_MAP_VALID;
 
-	fire_write(pbm->pbm_regs + MSI_MAP(msi), val);
+	upa_writeq(val, pbm->pbm_regs + MSI_MAP(msi));
 
 	return 0;
 }
@@ -261,22 +241,19 @@ static int pci_fire_msiq_alloc(struct pci_pbm_info *pbm)
 	memset((char *)pages, 0, PAGE_SIZE << order);
 	pbm->msi_queues = (void *) pages;
 
-	fire_write(pbm->pbm_regs + EVENT_QUEUE_BASE_ADDR_REG,
-		   (EVENT_QUEUE_BASE_ADDR_ALL_ONES |
-		    __pa(pbm->msi_queues)));
+	upa_writeq((EVENT_QUEUE_BASE_ADDR_ALL_ONES |
+		    __pa(pbm->msi_queues)),
+		   pbm->pbm_regs + EVENT_QUEUE_BASE_ADDR_REG);
 
-	fire_write(pbm->pbm_regs + IMONDO_DATA0,
-		   pbm->portid << 6);
-	fire_write(pbm->pbm_regs + IMONDO_DATA1, 0);
+	upa_writeq(pbm->portid << 6, pbm->pbm_regs + IMONDO_DATA0);
+	upa_writeq(0, pbm->pbm_regs + IMONDO_DATA1);
 
-	fire_write(pbm->pbm_regs + MSI_32BIT_ADDR,
-		   pbm->msi32_start);
-	fire_write(pbm->pbm_regs + MSI_64BIT_ADDR,
-		   pbm->msi64_start);
+	upa_writeq(pbm->msi32_start, pbm->pbm_regs + MSI_32BIT_ADDR);
+	upa_writeq(pbm->msi64_start, pbm->pbm_regs + MSI_64BIT_ADDR);
 
 	for (i = 0; i < pbm->msiq_num; i++) {
-		fire_write(pbm->pbm_regs + EVENT_QUEUE_HEAD(i), 0);
-		fire_write(pbm->pbm_regs + EVENT_QUEUE_TAIL(i), 0);
+		upa_writeq(0, pbm->pbm_regs + EVENT_QUEUE_HEAD(i));
+		upa_writeq(0, pbm->pbm_regs + EVENT_QUEUE_TAIL(i));
 	}
 
 	return 0;
@@ -310,9 +287,9 @@ static int pci_fire_msiq_build_irq(struct pci_pbm_info *pbm,
 	/* XXX iterate amongst the 4 IRQ controllers XXX */
 	int_ctrlr = (1UL << 6);
 
-	val = fire_read(imap_reg);
+	val = upa_readq(imap_reg);
 	val |= (1UL << 63) | int_ctrlr;
-	fire_write(imap_reg, val);
+	upa_writeq(val, imap_reg);
 
 	fixup = ((pbm->portid << 6) | devino) - int_ctrlr;
 
@@ -320,9 +297,8 @@ static int pci_fire_msiq_build_irq(struct pci_pbm_info *pbm,
 	if (!virt_irq)
 		return -ENOMEM;
 
-	fire_write(pbm->pbm_regs +
-		   EVENT_QUEUE_CONTROL_SET(msiqid),
-		   EVENT_QUEUE_CONTROL_SET_EN);
+	upa_writeq(EVENT_QUEUE_CONTROL_SET_EN,
+		   pbm->pbm_regs + EVENT_QUEUE_CONTROL_SET(msiqid));
 
 	return virt_irq;
 }
@@ -390,77 +366,65 @@ static void pci_fire_hw_init(struct pci_pbm_info *pbm)
 {
 	u64 val;
 
-	fire_write(pbm->controller_regs + FIRE_PARITY_CONTROL,
-		   FIRE_PARITY_ENAB);
+	upa_writeq(FIRE_PARITY_ENAB,
+		   pbm->controller_regs + FIRE_PARITY_CONTROL);
 
-	fire_write(pbm->controller_regs + FIRE_FATAL_RESET_CTL,
-		   (FIRE_FATAL_RESET_SPARE |
+	upa_writeq((FIRE_FATAL_RESET_SPARE |
 		    FIRE_FATAL_RESET_MB |
 		    FIRE_FATAL_RESET_CPE |
 		    FIRE_FATAL_RESET_APE |
 		    FIRE_FATAL_RESET_PIO |
 		    FIRE_FATAL_RESET_JW |
 		    FIRE_FATAL_RESET_JI |
-		    FIRE_FATAL_RESET_JR));
+		    FIRE_FATAL_RESET_JR),
+		   pbm->controller_regs + FIRE_FATAL_RESET_CTL);
 
-	fire_write(pbm->controller_regs + FIRE_CORE_INTR_ENABLE, ~(u64)0);
+	upa_writeq(~(u64)0, pbm->controller_regs + FIRE_CORE_INTR_ENABLE);
 
-	val = fire_read(pbm->pbm_regs + FIRE_TLU_CTRL);
+	val = upa_readq(pbm->pbm_regs + FIRE_TLU_CTRL);
 	val |= (FIRE_TLU_CTRL_TIM |
 		FIRE_TLU_CTRL_QDET |
 		FIRE_TLU_CTRL_CFG);
-	fire_write(pbm->pbm_regs + FIRE_TLU_CTRL, val);
-	fire_write(pbm->pbm_regs + FIRE_TLU_DEV_CTRL, 0);
-	fire_write(pbm->pbm_regs + FIRE_TLU_LINK_CTRL,
-		   FIRE_TLU_LINK_CTRL_CLK);
-
-	fire_write(pbm->pbm_regs + FIRE_LPU_RESET, 0);
-	fire_write(pbm->pbm_regs + FIRE_LPU_LLCFG,
-		   FIRE_LPU_LLCFG_VC0);
-	fire_write(pbm->pbm_regs + FIRE_LPU_FCTRL_UCTRL,
-		   (FIRE_LPU_FCTRL_UCTRL_N |
-		    FIRE_LPU_FCTRL_UCTRL_P));
-	fire_write(pbm->pbm_regs + FIRE_LPU_TXL_FIFOP,
-		   ((0xffff << 16) | (0x0000 << 0)));
-	fire_write(pbm->pbm_regs + FIRE_LPU_LTSSM_CFG2, 3000000);
-	fire_write(pbm->pbm_regs + FIRE_LPU_LTSSM_CFG3, 500000);
-	fire_write(pbm->pbm_regs + FIRE_LPU_LTSSM_CFG4,
-		   (2 << 16) | (140 << 8));
-	fire_write(pbm->pbm_regs + FIRE_LPU_LTSSM_CFG5, 0);
-
-	fire_write(pbm->pbm_regs + FIRE_DMC_IENAB, ~(u64)0);
-	fire_write(pbm->pbm_regs + FIRE_DMC_DBG_SEL_A, 0);
-	fire_write(pbm->pbm_regs + FIRE_DMC_DBG_SEL_B, 0);
-
-	fire_write(pbm->pbm_regs + FIRE_PEC_IENAB, ~(u64)0);
+	upa_writeq(val, pbm->pbm_regs + FIRE_TLU_CTRL);
+	upa_writeq(0, pbm->pbm_regs + FIRE_TLU_DEV_CTRL);
+	upa_writeq(FIRE_TLU_LINK_CTRL_CLK,
+		   pbm->pbm_regs + FIRE_TLU_LINK_CTRL);
+
+	upa_writeq(0, pbm->pbm_regs + FIRE_LPU_RESET);
+	upa_writeq(FIRE_LPU_LLCFG_VC0, pbm->pbm_regs + FIRE_LPU_LLCFG);
+	upa_writeq((FIRE_LPU_FCTRL_UCTRL_N | FIRE_LPU_FCTRL_UCTRL_P),
+		   pbm->pbm_regs + FIRE_LPU_FCTRL_UCTRL);
+	upa_writeq(((0xffff << 16) | (0x0000 << 0)),
+		   pbm->pbm_regs + FIRE_LPU_TXL_FIFOP);
+	upa_writeq(3000000, pbm->pbm_regs + FIRE_LPU_LTSSM_CFG2);
+	upa_writeq(500000, pbm->pbm_regs + FIRE_LPU_LTSSM_CFG3);
+	upa_writeq((2 << 16) | (140 << 8),
+		   pbm->pbm_regs + FIRE_LPU_LTSSM_CFG4);
+	upa_writeq(0, pbm->pbm_regs + FIRE_LPU_LTSSM_CFG5);
+
+	upa_writeq(~(u64)0, pbm->pbm_regs + FIRE_DMC_IENAB);
+	upa_writeq(0, pbm->pbm_regs + FIRE_DMC_DBG_SEL_A);
+	upa_writeq(0, pbm->pbm_regs + FIRE_DMC_DBG_SEL_B);
+
+	upa_writeq(~(u64)0, pbm->pbm_regs + FIRE_PEC_IENAB);
 }
 
-static int __init pci_fire_pbm_init(struct pci_controller_info *p,
-				    struct device_node *dp, u32 portid)
+static int __init pci_fire_pbm_init(struct pci_pbm_info *pbm,
+				    struct of_device *op, u32 portid)
 {
 	const struct linux_prom64_registers *regs;
-	struct pci_pbm_info *pbm;
+	struct device_node *dp = op->node;
 	int err;
 
-	if ((portid & 1) == 0)
-		pbm = &p->pbm_A;
-	else
-		pbm = &p->pbm_B;
-
-	pbm->next = pci_pbm_root;
-	pci_pbm_root = pbm;
-
 	pbm->numa_node = -1;
 
-	pbm->scan_bus = pci_fire_scan_bus;
 	pbm->pci_ops = &sun4u_pci_ops;
 	pbm->config_space_reg_bits = 12;
 
 	pbm->index = pci_num_pbms++;
 
 	pbm->portid = portid;
-	pbm->parent = p;
-	pbm->prom_node = dp;
+	pbm->op = op;
 	pbm->name = dp->full_name;
 
 	regs = of_get_property(dp, "reg", NULL);
@@ -481,53 +445,77 @@ static int __init pci_fire_pbm_init(struct pci_controller_info *p,
 
 	pci_fire_msi_init(pbm);
 
-	return 0;
-}
+	pbm->pci_bus = pci_scan_one_pbm(pbm, &op->dev);
+
+	/* XXX register error interrupt handlers XXX */
+
+	pbm->next = pci_pbm_root;
+	pci_pbm_root = pbm;
 
-static inline int portid_compare(u32 x, u32 y)
-{
-	if (x == (y ^ 1))
-		return 1;
 	return 0;
 }
 
-void __init fire_pci_init(struct device_node *dp, const char *model_name)
+static int __devinit fire_probe(struct of_device *op,
+				const struct of_device_id *match)
 {
-	struct pci_controller_info *p;
-	u32 portid = of_getintprop_default(dp, "portid", 0xff);
-	struct iommu *iommu;
+	struct device_node *dp = op->node;
 	struct pci_pbm_info *pbm;
+	struct iommu *iommu;
+	u32 portid;
+	int err;
 
-	for (pbm = pci_pbm_root; pbm; pbm = pbm->next) {
-		if (portid_compare(pbm->portid, portid)) {
-			if (pci_fire_pbm_init(pbm->parent, dp, portid))
-				goto fatal_memory_error;
-			return;
-		}
+	portid = of_getintprop_default(dp, "portid", 0xff);
+
+	err = -ENOMEM;
+	pbm = kzalloc(sizeof(*pbm), GFP_KERNEL);
+	if (!pbm) {
+		printk(KERN_ERR PFX "Cannot allocate pci_pbminfo.\n");
+		goto out_err;
 	}
 
-	p = kzalloc(sizeof(struct pci_controller_info), GFP_ATOMIC);
-	if (!p)
-		goto fatal_memory_error;
+	iommu = kzalloc(sizeof(struct iommu), GFP_KERNEL);
+	if (!iommu) {
+		printk(KERN_ERR PFX "Cannot allocate PBM iommu.\n");
+		goto out_free_controller;
+	}
 
-	iommu = kzalloc(sizeof(struct iommu), GFP_ATOMIC);
-	if (!iommu)
-		goto fatal_memory_error;
+	pbm->iommu = iommu;
 
-	p->pbm_A.iommu = iommu;
+	err = pci_fire_pbm_init(pbm, op, portid);
+	if (err)
+		goto out_free_iommu;
 
-	iommu = kzalloc(sizeof(struct iommu), GFP_ATOMIC);
-	if (!iommu)
-		goto fatal_memory_error;
+	dev_set_drvdata(&op->dev, pbm);
 
-	p->pbm_B.iommu = iommu;
+	return 0;
 
-	if (pci_fire_pbm_init(p, dp, portid))
-		goto fatal_memory_error;
+out_free_iommu:
+	kfree(pbm->iommu);
+			
+out_free_controller:
+	kfree(pbm);
 
-	return;
+out_err:
+	return err;
+}
+
+static struct of_device_id __initdata fire_match[] = {
+	{
+		.name = "pci",
+		.compatible = "pciex108e,80f0",
+	},
+	{},
+};
 
-fatal_memory_error:
-	prom_printf("PCI_FIRE: Fatal memory allocation error.\n");
-	prom_halt();
+static struct of_platform_driver fire_driver = {
+	.name		= DRIVER_NAME,
+	.match_table	= fire_match,
+	.probe		= fire_probe,
+};
+
+static int __init fire_init(void)
+{
+	return of_register_driver(&fire_driver, &of_bus_type);
 }
+
+subsys_initcall(fire_init);
diff --git a/arch/sparc64/kernel/pci_impl.h b/arch/sparc64/kernel/pci_impl.h
index c385d126be11..03186824327e 100644
--- a/arch/sparc64/kernel/pci_impl.h
+++ b/arch/sparc64/kernel/pci_impl.h
@@ -10,6 +10,7 @@
 #include <linux/spinlock.h>
 #include <linux/pci.h>
 #include <linux/msi.h>
+#include <linux/of_device.h>
 #include <asm/io.h>
 #include <asm/prom.h>
 #include <asm/iommu.h>
@@ -56,15 +57,11 @@ struct sparc64_msiq_cookie {
 };
 #endif
 
-struct pci_controller_info;
-
 struct pci_pbm_info {
 	struct pci_pbm_info		*next;
+	struct pci_pbm_info		*sibling;
 	int				index;
 
-	/* PCI controller we sit under. */
-	struct pci_controller_info	*parent;
-
 	/* Physical address base of controller registers. */
 	unsigned long			controller_regs;
 
@@ -94,7 +91,7 @@ struct pci_pbm_info {
 	char				*name;
 
 	/* OBP specific information. */
-	struct device_node		*prom_node;
+	struct of_device		*op;
 	u64				ino_bitmap;
 
 	/* PBM I/O and Memory space resources. */
@@ -107,6 +104,10 @@ struct pci_pbm_info {
 	/* This will be 12 on PCI-E controllers, 8 elsewhere.  */
 	unsigned long			config_space_reg_bits;
 
+	unsigned long			pci_afsr;
+	unsigned long			pci_afar;
+	unsigned long			pci_csr;
+
 	/* State of 66MHz capabilities on this PBM. */
 	int				is_66mhz_capable;
 	int				all_devs_66mhz;
@@ -146,25 +147,19 @@ struct pci_pbm_info {
 	unsigned int			pci_first_busno;
 	unsigned int			pci_last_busno;
 	struct pci_bus			*pci_bus;
-	void (*scan_bus)(struct pci_pbm_info *);
 	struct pci_ops			*pci_ops;
 
 	int				numa_node;
 };
 
-struct pci_controller_info {
-	/* The PCI bus modules controlled by us. */
-	struct pci_pbm_info		pbm_A;
-	struct pci_pbm_info		pbm_B;
-};
-
 extern struct pci_pbm_info *pci_pbm_root;
 
 extern int pci_num_pbms;
 
 /* PCI bus scanning and fixup support. */
 extern void pci_get_pbm_props(struct pci_pbm_info *pbm);
-extern struct pci_bus *pci_scan_one_pbm(struct pci_pbm_info *pbm);
+extern struct pci_bus *pci_scan_one_pbm(struct pci_pbm_info *pbm,
+					struct device *parent);
 extern void pci_determine_mem_io_space(struct pci_pbm_info *pbm);
 
 /* Error reporting support. */
@@ -183,4 +178,8 @@ extern void pci_config_write32(u32 *addr, u32 val);
 extern struct pci_ops sun4u_pci_ops;
 extern struct pci_ops sun4v_pci_ops;
 
+extern volatile int pci_poke_in_progress;
+extern volatile int pci_poke_cpu;
+extern volatile int pci_poke_faulted;
+
 #endif /* !(PCI_IMPL_H) */
diff --git a/arch/sparc64/kernel/pci_msi.c b/arch/sparc64/kernel/pci_msi.c
index 60c71e350212..2e680f34f727 100644
--- a/arch/sparc64/kernel/pci_msi.c
+++ b/arch/sparc64/kernel/pci_msi.c
@@ -323,7 +323,7 @@ void sparc64_pbm_msi_init(struct pci_pbm_info *pbm,
 	const u32 *val;
 	int len;
 
-	val = of_get_property(pbm->prom_node, "#msi-eqs", &len);
+	val = of_get_property(pbm->op->node, "#msi-eqs", &len);
 	if (!val || len != 4)
 		goto no_msi;
 	pbm->msiq_num = *val;
@@ -346,16 +346,16 @@ void sparc64_pbm_msi_init(struct pci_pbm_info *pbm,
 			u32 msi64_len;
 		} *arng;
 
-		val = of_get_property(pbm->prom_node, "msi-eq-size", &len);
+		val = of_get_property(pbm->op->node, "msi-eq-size", &len);
 		if (!val || len != 4)
 			goto no_msi;
 
 		pbm->msiq_ent_count = *val;
 
-		mqp = of_get_property(pbm->prom_node,
+		mqp = of_get_property(pbm->op->node,
 				      "msi-eq-to-devino", &len);
 		if (!mqp)
-			mqp = of_get_property(pbm->prom_node,
+			mqp = of_get_property(pbm->op->node,
 					      "msi-eq-devino", &len);
 		if (!mqp || len != sizeof(struct msiq_prop))
 			goto no_msi;
@@ -363,27 +363,27 @@ void sparc64_pbm_msi_init(struct pci_pbm_info *pbm,
 		pbm->msiq_first = mqp->first_msiq;
 		pbm->msiq_first_devino = mqp->first_devino;
 
-		val = of_get_property(pbm->prom_node, "#msi", &len);
+		val = of_get_property(pbm->op->node, "#msi", &len);
 		if (!val || len != 4)
 			goto no_msi;
 		pbm->msi_num = *val;
 
-		mrng = of_get_property(pbm->prom_node, "msi-ranges", &len);
+		mrng = of_get_property(pbm->op->node, "msi-ranges", &len);
 		if (!mrng || len != sizeof(struct msi_range_prop))
 			goto no_msi;
 		pbm->msi_first = mrng->first_msi;
 
-		val = of_get_property(pbm->prom_node, "msi-data-mask", &len);
+		val = of_get_property(pbm->op->node, "msi-data-mask", &len);
 		if (!val || len != 4)
 			goto no_msi;
 		pbm->msi_data_mask = *val;
 
-		val = of_get_property(pbm->prom_node, "msix-data-width", &len);
+		val = of_get_property(pbm->op->node, "msix-data-width", &len);
 		if (!val || len != 4)
 			goto no_msi;
 		pbm->msix_data_width = *val;
 
-		arng = of_get_property(pbm->prom_node, "msi-address-ranges",
+		arng = of_get_property(pbm->op->node, "msi-address-ranges",
 				       &len);
 		if (!arng || len != sizeof(struct addr_range_prop))
 			goto no_msi;
diff --git a/arch/sparc64/kernel/pci_psycho.c b/arch/sparc64/kernel/pci_psycho.c
index f85b6bebb0be..dfb3ec892987 100644
--- a/arch/sparc64/kernel/pci_psycho.c
+++ b/arch/sparc64/kernel/pci_psycho.c
@@ -17,29 +17,14 @@
 #include <asm/irq.h>
 #include <asm/starfire.h>
 #include <asm/prom.h>
-#include <asm/oplib.h>
+#include <asm/upa.h>
 
 #include "pci_impl.h"
 #include "iommu_common.h"
+#include "psycho_common.h"
 
-/* All PSYCHO registers are 64-bits.  The following accessor
- * routines are how they are accessed.  The REG parameter
- * is a physical address.
- */
-#define psycho_read(__reg) \
-({	u64 __ret; \
-	__asm__ __volatile__("ldxa [%1] %2, %0" \
-			     : "=r" (__ret) \
-			     : "r" (__reg), "i" (ASI_PHYS_BYPASS_EC_E) \
-			     : "memory"); \
-	__ret; \
-})
-#define psycho_write(__reg, __val) \
-	__asm__ __volatile__("stxa %0, [%1] %2" \
-			     : /* no outputs */ \
-			     : "r" (__val), "r" (__reg), \
-			       "i" (ASI_PHYS_BYPASS_EC_E) \
-			     : "memory")
+#define DRIVER_NAME	"psycho"
+#define PFX		DRIVER_NAME ": "
 
 /* Misc. PSYCHO PCI controller register offsets and definitions. */
 #define PSYCHO_CONTROL		0x0010UL
@@ -67,37 +52,7 @@
 #define  PSYCHO_PCICTRL_RESV4	 0x00000000000000c0UL /* Reserved                     */
 #define  PSYCHO_PCICTRL_AEN	 0x000000000000003fUL /* PCI DVMA Arbitration Enable  */
 
-/* U2P Programmer's Manual, page 13-55, configuration space
- * address format:
- * 
- *  32             24 23 16 15    11 10       8 7   2  1 0
- * ---------------------------------------------------------
- * |0 0 0 0 0 0 0 0 1| bus | device | function | reg | 0 0 |
- * ---------------------------------------------------------
- */
-#define PSYCHO_CONFIG_BASE(PBM)	\
-	((PBM)->config_space | (1UL << 24))
-#define PSYCHO_CONFIG_ENCODE(BUS, DEVFN, REG)	\
-	(((unsigned long)(BUS)   << 16) |	\
-	 ((unsigned long)(DEVFN) << 8)  |	\
-	 ((unsigned long)(REG)))
-
-static void *psycho_pci_config_mkaddr(struct pci_pbm_info *pbm,
-				      unsigned char bus,
-				      unsigned int devfn,
-				      int where)
-{
-	if (!pbm)
-		return NULL;
-	return (void *)
-		(PSYCHO_CONFIG_BASE(pbm) |
-		 PSYCHO_CONFIG_ENCODE(bus, devfn, where));
-}
-
 /* PSYCHO error handling support. */
-enum psycho_error_type {
-	UE_ERR, CE_ERR, PCI_ERR
-};
 
 /* Helper function of IOMMU error checking, which checks out
  * the state of the streaming buffers.  The IOMMU lock is
@@ -122,129 +77,10 @@ enum psycho_error_type {
 #define PSYCHO_STC_DATA_B	0xc000UL
 #define PSYCHO_STC_ERR_A	0xb400UL
 #define PSYCHO_STC_ERR_B	0xc400UL
-#define  PSYCHO_STCERR_WRITE	 0x0000000000000002UL	/* Write Error */
-#define  PSYCHO_STCERR_READ	 0x0000000000000001UL	/* Read Error */
 #define PSYCHO_STC_TAG_A	0xb800UL
 #define PSYCHO_STC_TAG_B	0xc800UL
-#define  PSYCHO_STCTAG_PPN	 0x0fffffff00000000UL	/* Physical Page Number */
-#define  PSYCHO_STCTAG_VPN	 0x00000000ffffe000UL	/* Virtual Page Number */
-#define  PSYCHO_STCTAG_VALID	 0x0000000000000002UL	/* Valid */
-#define  PSYCHO_STCTAG_WRITE	 0x0000000000000001UL	/* Writable */
 #define PSYCHO_STC_LINE_A	0xb900UL
 #define PSYCHO_STC_LINE_B	0xc900UL
-#define  PSYCHO_STCLINE_LINDX	 0x0000000001e00000UL	/* LRU Index */
-#define  PSYCHO_STCLINE_SPTR	 0x00000000001f8000UL	/* Dirty Data Start Pointer */
-#define  PSYCHO_STCLINE_LADDR	 0x0000000000007f00UL	/* Line Address */
-#define  PSYCHO_STCLINE_EPTR	 0x00000000000000fcUL	/* Dirty Data End Pointer */
-#define  PSYCHO_STCLINE_VALID	 0x0000000000000002UL	/* Valid */
-#define  PSYCHO_STCLINE_FOFN	 0x0000000000000001UL	/* Fetch Outstanding / Flush Necessary */
-
-static DEFINE_SPINLOCK(stc_buf_lock);
-static unsigned long stc_error_buf[128];
-static unsigned long stc_tag_buf[16];
-static unsigned long stc_line_buf[16];
-
-static void __psycho_check_one_stc(struct pci_pbm_info *pbm,
-				   int is_pbm_a)
-{
-	struct strbuf *strbuf = &pbm->stc;
-	unsigned long regbase = pbm->controller_regs;
-	unsigned long err_base, tag_base, line_base;
-	u64 control;
-	int i;
-
-	if (is_pbm_a) {
-		err_base = regbase + PSYCHO_STC_ERR_A;
-		tag_base = regbase + PSYCHO_STC_TAG_A;
-		line_base = regbase + PSYCHO_STC_LINE_A;
-	} else {
-		err_base = regbase + PSYCHO_STC_ERR_B;
-		tag_base = regbase + PSYCHO_STC_TAG_B;
-		line_base = regbase + PSYCHO_STC_LINE_B;
-	}
-
-	spin_lock(&stc_buf_lock);
-
-	/* This is __REALLY__ dangerous.  When we put the
-	 * streaming buffer into diagnostic mode to probe
-	 * it's tags and error status, we _must_ clear all
-	 * of the line tag valid bits before re-enabling
-	 * the streaming buffer.  If any dirty data lives
-	 * in the STC when we do this, we will end up
-	 * invalidating it before it has a chance to reach
-	 * main memory.
-	 */
-	control = psycho_read(strbuf->strbuf_control);
-	psycho_write(strbuf->strbuf_control,
-		     (control | PSYCHO_STRBUF_CTRL_DENAB));
-	for (i = 0; i < 128; i++) {
-		unsigned long val;
-
-		val = psycho_read(err_base + (i * 8UL));
-		psycho_write(err_base + (i * 8UL), 0UL);
-		stc_error_buf[i] = val;
-	}
-	for (i = 0; i < 16; i++) {
-		stc_tag_buf[i] = psycho_read(tag_base + (i * 8UL));
-		stc_line_buf[i] = psycho_read(line_base + (i * 8UL));
-		psycho_write(tag_base + (i * 8UL), 0UL);
-		psycho_write(line_base + (i * 8UL), 0UL);
-	}
-
-	/* OK, state is logged, exit diagnostic mode. */
-	psycho_write(strbuf->strbuf_control, control);
-
-	for (i = 0; i < 16; i++) {
-		int j, saw_error, first, last;
-
-		saw_error = 0;
-		first = i * 8;
-		last = first + 8;
-		for (j = first; j < last; j++) {
-			unsigned long errval = stc_error_buf[j];
-			if (errval != 0) {
-				saw_error++;
-				printk("%s: STC_ERR(%d)[wr(%d)rd(%d)]\n",
-				       pbm->name,
-				       j,
-				       (errval & PSYCHO_STCERR_WRITE) ? 1 : 0,
-				       (errval & PSYCHO_STCERR_READ) ? 1 : 0);
-			}
-		}
-		if (saw_error != 0) {
-			unsigned long tagval = stc_tag_buf[i];
-			unsigned long lineval = stc_line_buf[i];
-			printk("%s: STC_TAG(%d)[PA(%016lx)VA(%08lx)V(%d)W(%d)]\n",
-			       pbm->name,
-			       i,
-			       ((tagval & PSYCHO_STCTAG_PPN) >> 19UL),
-			       (tagval & PSYCHO_STCTAG_VPN),
-			       ((tagval & PSYCHO_STCTAG_VALID) ? 1 : 0),
-			       ((tagval & PSYCHO_STCTAG_WRITE) ? 1 : 0));
-			printk("%s: STC_LINE(%d)[LIDX(%lx)SP(%lx)LADDR(%lx)EP(%lx)"
-			       "V(%d)FOFN(%d)]\n",
-			       pbm->name,
-			       i,
-			       ((lineval & PSYCHO_STCLINE_LINDX) >> 21UL),
-			       ((lineval & PSYCHO_STCLINE_SPTR) >> 15UL),
-			       ((lineval & PSYCHO_STCLINE_LADDR) >> 8UL),
-			       ((lineval & PSYCHO_STCLINE_EPTR) >> 2UL),
-			       ((lineval & PSYCHO_STCLINE_VALID) ? 1 : 0),
-			       ((lineval & PSYCHO_STCLINE_FOFN) ? 1 : 0));
-		}
-	}
-
-	spin_unlock(&stc_buf_lock);
-}
-
-static void __psycho_check_stc_error(struct pci_pbm_info *pbm,
-				     unsigned long afsr,
-				     unsigned long afar,
-				     enum psycho_error_type type)
-{
-	__psycho_check_one_stc(pbm,
-			       (pbm == &pbm->parent->pbm_A));
-}
 
 /* When an Uncorrectable Error or a PCI Error happens, we
  * interrogate the IOMMU state to see if it is the cause.
@@ -271,122 +107,7 @@ static void __psycho_check_stc_error(struct pci_pbm_info *pbm,
 #define PSYCHO_IOMMU_TSBBASE	0x0208UL
 #define PSYCHO_IOMMU_FLUSH	0x0210UL
 #define PSYCHO_IOMMU_TAG	0xa580UL
-#define  PSYCHO_IOMMU_TAG_ERRSTS (0x3UL << 23UL)
-#define  PSYCHO_IOMMU_TAG_ERR	 (0x1UL << 22UL)
-#define  PSYCHO_IOMMU_TAG_WRITE	 (0x1UL << 21UL)
-#define  PSYCHO_IOMMU_TAG_STREAM (0x1UL << 20UL)
-#define  PSYCHO_IOMMU_TAG_SIZE	 (0x1UL << 19UL)
-#define  PSYCHO_IOMMU_TAG_VPAGE	 0x7ffffUL
 #define PSYCHO_IOMMU_DATA	0xa600UL
-#define  PSYCHO_IOMMU_DATA_VALID (1UL << 30UL)
-#define  PSYCHO_IOMMU_DATA_CACHE (1UL << 28UL)
-#define  PSYCHO_IOMMU_DATA_PPAGE 0xfffffffUL
-static void psycho_check_iommu_error(struct pci_pbm_info *pbm,
-				     unsigned long afsr,
-				     unsigned long afar,
-				     enum psycho_error_type type)
-{
-	struct iommu *iommu = pbm->iommu;
-	unsigned long iommu_tag[16];
-	unsigned long iommu_data[16];
-	unsigned long flags;
-	u64 control;
-	int i;
-
-	spin_lock_irqsave(&iommu->lock, flags);
-	control = psycho_read(iommu->iommu_control);
-	if (control & PSYCHO_IOMMU_CTRL_XLTEERR) {
-		char *type_string;
-
-		/* Clear the error encountered bit. */
-		control &= ~PSYCHO_IOMMU_CTRL_XLTEERR;
-		psycho_write(iommu->iommu_control, control);
-
-		switch((control & PSYCHO_IOMMU_CTRL_XLTESTAT) >> 25UL) {
-		case 0:
-			type_string = "Protection Error";
-			break;
-		case 1:
-			type_string = "Invalid Error";
-			break;
-		case 2:
-			type_string = "TimeOut Error";
-			break;
-		case 3:
-		default:
-			type_string = "ECC Error";
-			break;
-		};
-		printk("%s: IOMMU Error, type[%s]\n",
-		       pbm->name, type_string);
-
-		/* Put the IOMMU into diagnostic mode and probe
-		 * it's TLB for entries with error status.
-		 *
-		 * It is very possible for another DVMA to occur
-		 * while we do this probe, and corrupt the system
-		 * further.  But we are so screwed at this point
-		 * that we are likely to crash hard anyways, so
-		 * get as much diagnostic information to the
-		 * console as we can.
-		 */
-		psycho_write(iommu->iommu_control,
-			     control | PSYCHO_IOMMU_CTRL_DENAB);
-		for (i = 0; i < 16; i++) {
-			unsigned long base = pbm->controller_regs;
-
-			iommu_tag[i] =
-				psycho_read(base + PSYCHO_IOMMU_TAG + (i * 8UL));
-			iommu_data[i] =
-				psycho_read(base + PSYCHO_IOMMU_DATA + (i * 8UL));
-
-			/* Now clear out the entry. */
-			psycho_write(base + PSYCHO_IOMMU_TAG + (i * 8UL), 0);
-			psycho_write(base + PSYCHO_IOMMU_DATA + (i * 8UL), 0);
-		}
-
-		/* Leave diagnostic mode. */
-		psycho_write(iommu->iommu_control, control);
-
-		for (i = 0; i < 16; i++) {
-			unsigned long tag, data;
-
-			tag = iommu_tag[i];
-			if (!(tag & PSYCHO_IOMMU_TAG_ERR))
-				continue;
-
-			data = iommu_data[i];
-			switch((tag & PSYCHO_IOMMU_TAG_ERRSTS) >> 23UL) {
-			case 0:
-				type_string = "Protection Error";
-				break;
-			case 1:
-				type_string = "Invalid Error";
-				break;
-			case 2:
-				type_string = "TimeOut Error";
-				break;
-			case 3:
-			default:
-				type_string = "ECC Error";
-				break;
-			};
-			printk("%s: IOMMU TAG(%d)[error(%s) wr(%d) str(%d) sz(%dK) vpg(%08lx)]\n",
-			       pbm->name, i, type_string,
-			       ((tag & PSYCHO_IOMMU_TAG_WRITE) ? 1 : 0),
-			       ((tag & PSYCHO_IOMMU_TAG_STREAM) ? 1 : 0),
-			       ((tag & PSYCHO_IOMMU_TAG_SIZE) ? 64 : 8),
-			       (tag & PSYCHO_IOMMU_TAG_VPAGE) << IOMMU_PAGE_SHIFT);
-			printk("%s: IOMMU DATA(%d)[valid(%d) cache(%d) ppg(%016lx)]\n",
-			       pbm->name, i,
-			       ((data & PSYCHO_IOMMU_DATA_VALID) ? 1 : 0),
-			       ((data & PSYCHO_IOMMU_DATA_CACHE) ? 1 : 0),
-			       (data & PSYCHO_IOMMU_DATA_PPAGE) << IOMMU_PAGE_SHIFT);
-		}
-	}
-	__psycho_check_stc_error(pbm, afsr, afar, type);
-	spin_unlock_irqrestore(&iommu->lock, flags);
-}
 
 /* Uncorrectable Errors.  Cause of the error and the address are
  * recorded in the UE_AFSR and UE_AFAR of PSYCHO.  They are errors
@@ -410,15 +131,14 @@ static void psycho_check_iommu_error(struct pci_pbm_info *pbm,
 static irqreturn_t psycho_ue_intr(int irq, void *dev_id)
 {
 	struct pci_pbm_info *pbm = dev_id;
-	struct pci_controller_info *p = pbm->parent;
 	unsigned long afsr_reg = pbm->controller_regs + PSYCHO_UE_AFSR;
 	unsigned long afar_reg = pbm->controller_regs + PSYCHO_UE_AFAR;
 	unsigned long afsr, afar, error_bits;
 	int reported;
 
 	/* Latch uncorrectable error status. */
-	afar = psycho_read(afar_reg);
-	afsr = psycho_read(afsr_reg);
+	afar = upa_readq(afar_reg);
+	afsr = upa_readq(afsr_reg);
 
 	/* Clear the primary/secondary error status bits. */
 	error_bits = afsr &
@@ -426,7 +146,7 @@ static irqreturn_t psycho_ue_intr(int irq, void *dev_id)
 		 PSYCHO_UEAFSR_SPIO | PSYCHO_UEAFSR_SDRD | PSYCHO_UEAFSR_SDWR);
 	if (!error_bits)
 		return IRQ_NONE;
-	psycho_write(afsr_reg, error_bits);
+	upa_writeq(error_bits, afsr_reg);
 
 	/* Log the error. */
 	printk("%s: Uncorrectable Error, primary error type[%s]\n",
@@ -463,8 +183,9 @@ static irqreturn_t psycho_ue_intr(int irq, void *dev_id)
 	printk("]\n");
 
 	/* Interrogate both IOMMUs for error status. */
-	psycho_check_iommu_error(&p->pbm_A, afsr, afar, UE_ERR);
-	psycho_check_iommu_error(&p->pbm_B, afsr, afar, UE_ERR);
+	psycho_check_iommu_error(pbm, afsr, afar, UE_ERR);
+	if (pbm->sibling)
+		psycho_check_iommu_error(pbm->sibling, afsr, afar, UE_ERR);
 
 	return IRQ_HANDLED;
 }
@@ -495,8 +216,8 @@ static irqreturn_t psycho_ce_intr(int irq, void *dev_id)
 	int reported;
 
 	/* Latch error status. */
-	afar = psycho_read(afar_reg);
-	afsr = psycho_read(afsr_reg);
+	afar = upa_readq(afar_reg);
+	afsr = upa_readq(afsr_reg);
 
 	/* Clear primary/secondary error status bits. */
 	error_bits = afsr &
@@ -504,7 +225,7 @@ static irqreturn_t psycho_ce_intr(int irq, void *dev_id)
 		 PSYCHO_CEAFSR_SPIO | PSYCHO_CEAFSR_SDRD | PSYCHO_CEAFSR_SDWR);
 	if (!error_bits)
 		return IRQ_NONE;
-	psycho_write(afsr_reg, error_bits);
+	upa_writeq(error_bits, afsr_reg);
 
 	/* Log the error. */
 	printk("%s: Correctable Error, primary error type[%s]\n",
@@ -554,164 +275,9 @@ static irqreturn_t psycho_ce_intr(int irq, void *dev_id)
  */
 #define PSYCHO_PCI_AFSR_A	0x2010UL
 #define PSYCHO_PCI_AFSR_B	0x4010UL
-#define  PSYCHO_PCIAFSR_PMA	0x8000000000000000UL /* Primary Master Abort Error   */
-#define  PSYCHO_PCIAFSR_PTA	0x4000000000000000UL /* Primary Target Abort Error   */
-#define  PSYCHO_PCIAFSR_PRTRY	0x2000000000000000UL /* Primary Excessive Retries    */
-#define  PSYCHO_PCIAFSR_PPERR	0x1000000000000000UL /* Primary Parity Error         */
-#define  PSYCHO_PCIAFSR_SMA	0x0800000000000000UL /* Secondary Master Abort Error */
-#define  PSYCHO_PCIAFSR_STA	0x0400000000000000UL /* Secondary Target Abort Error */
-#define  PSYCHO_PCIAFSR_SRTRY	0x0200000000000000UL /* Secondary Excessive Retries  */
-#define  PSYCHO_PCIAFSR_SPERR	0x0100000000000000UL /* Secondary Parity Error       */
-#define  PSYCHO_PCIAFSR_RESV1	0x00ff000000000000UL /* Reserved                     */
-#define  PSYCHO_PCIAFSR_BMSK	0x0000ffff00000000UL /* Bytemask of failed transfer  */
-#define  PSYCHO_PCIAFSR_BLK	0x0000000080000000UL /* Trans was block operation    */
-#define  PSYCHO_PCIAFSR_RESV2	0x0000000040000000UL /* Reserved                     */
-#define  PSYCHO_PCIAFSR_MID	0x000000003e000000UL /* MID causing the error        */
-#define  PSYCHO_PCIAFSR_RESV3	0x0000000001ffffffUL /* Reserved                     */
 #define PSYCHO_PCI_AFAR_A	0x2018UL
 #define PSYCHO_PCI_AFAR_B	0x4018UL
 
-static irqreturn_t psycho_pcierr_intr_other(struct pci_pbm_info *pbm, int is_pbm_a)
-{
-	unsigned long csr_reg, csr, csr_error_bits;
-	irqreturn_t ret = IRQ_NONE;
-	u16 stat, *addr;
-
-	if (is_pbm_a) {
-		csr_reg = pbm->controller_regs + PSYCHO_PCIA_CTRL;
-	} else {
-		csr_reg = pbm->controller_regs + PSYCHO_PCIB_CTRL;
-	}
-	csr = psycho_read(csr_reg);
-	csr_error_bits =
-		csr & (PSYCHO_PCICTRL_SBH_ERR | PSYCHO_PCICTRL_SERR);
-	if (csr_error_bits) {
-		/* Clear the errors.  */
-		psycho_write(csr_reg, csr);
-
-		/* Log 'em.  */
-		if (csr_error_bits & PSYCHO_PCICTRL_SBH_ERR)
-			printk("%s: PCI streaming byte hole error asserted.\n",
-			       pbm->name);
-		if (csr_error_bits & PSYCHO_PCICTRL_SERR)
-			printk("%s: PCI SERR signal asserted.\n", pbm->name);
-		ret = IRQ_HANDLED;
-	}
-	addr = psycho_pci_config_mkaddr(pbm, pbm->pci_first_busno,
-					0, PCI_STATUS);
-	pci_config_read16(addr, &stat);
-	if (stat & (PCI_STATUS_PARITY |
-		    PCI_STATUS_SIG_TARGET_ABORT |
-		    PCI_STATUS_REC_TARGET_ABORT |
-		    PCI_STATUS_REC_MASTER_ABORT |
-		    PCI_STATUS_SIG_SYSTEM_ERROR)) {
-		printk("%s: PCI bus error, PCI_STATUS[%04x]\n",
-		       pbm->name, stat);
-		pci_config_write16(addr, 0xffff);
-		ret = IRQ_HANDLED;
-	}
-	return ret;
-}
-
-static irqreturn_t psycho_pcierr_intr(int irq, void *dev_id)
-{
-	struct pci_pbm_info *pbm = dev_id;
-	struct pci_controller_info *p = pbm->parent;
-	unsigned long afsr_reg, afar_reg;
-	unsigned long afsr, afar, error_bits;
-	int is_pbm_a, reported;
-
-	is_pbm_a = (pbm == &pbm->parent->pbm_A);
-	if (is_pbm_a) {
-		afsr_reg = p->pbm_A.controller_regs + PSYCHO_PCI_AFSR_A;
-		afar_reg = p->pbm_A.controller_regs + PSYCHO_PCI_AFAR_A;
-	} else {
-		afsr_reg = p->pbm_A.controller_regs + PSYCHO_PCI_AFSR_B;
-		afar_reg = p->pbm_A.controller_regs + PSYCHO_PCI_AFAR_B;
-	}
-
-	/* Latch error status. */
-	afar = psycho_read(afar_reg);
-	afsr = psycho_read(afsr_reg);
-
-	/* Clear primary/secondary error status bits. */
-	error_bits = afsr &
-		(PSYCHO_PCIAFSR_PMA | PSYCHO_PCIAFSR_PTA |
-		 PSYCHO_PCIAFSR_PRTRY | PSYCHO_PCIAFSR_PPERR |
-		 PSYCHO_PCIAFSR_SMA | PSYCHO_PCIAFSR_STA |
-		 PSYCHO_PCIAFSR_SRTRY | PSYCHO_PCIAFSR_SPERR);
-	if (!error_bits)
-		return psycho_pcierr_intr_other(pbm, is_pbm_a);
-	psycho_write(afsr_reg, error_bits);
-
-	/* Log the error. */
-	printk("%s: PCI Error, primary error type[%s]\n",
-	       pbm->name,
-	       (((error_bits & PSYCHO_PCIAFSR_PMA) ?
-		 "Master Abort" :
-		 ((error_bits & PSYCHO_PCIAFSR_PTA) ?
-		  "Target Abort" :
-		  ((error_bits & PSYCHO_PCIAFSR_PRTRY) ?
-		   "Excessive Retries" :
-		   ((error_bits & PSYCHO_PCIAFSR_PPERR) ?
-		    "Parity Error" : "???"))))));
-	printk("%s: bytemask[%04lx] UPA_MID[%02lx] was_block(%d)\n",
-	       pbm->name,
-	       (afsr & PSYCHO_PCIAFSR_BMSK) >> 32UL,
-	       (afsr & PSYCHO_PCIAFSR_MID) >> 25UL,
-	       (afsr & PSYCHO_PCIAFSR_BLK) ? 1 : 0);
-	printk("%s: PCI AFAR [%016lx]\n", pbm->name, afar);
-	printk("%s: PCI Secondary errors [", pbm->name);
-	reported = 0;
-	if (afsr & PSYCHO_PCIAFSR_SMA) {
-		reported++;
-		printk("(Master Abort)");
-	}
-	if (afsr & PSYCHO_PCIAFSR_STA) {
-		reported++;
-		printk("(Target Abort)");
-	}
-	if (afsr & PSYCHO_PCIAFSR_SRTRY) {
-		reported++;
-		printk("(Excessive Retries)");
-	}
-	if (afsr & PSYCHO_PCIAFSR_SPERR) {
-		reported++;
-		printk("(Parity Error)");
-	}
-	if (!reported)
-		printk("(none)");
-	printk("]\n");
-
-	/* For the error types shown, scan PBM's PCI bus for devices
-	 * which have logged that error type.
-	 */
-
-	/* If we see a Target Abort, this could be the result of an
-	 * IOMMU translation error of some sort.  It is extremely
-	 * useful to log this information as usually it indicates
-	 * a bug in the IOMMU support code or a PCI device driver.
-	 */
-	if (error_bits & (PSYCHO_PCIAFSR_PTA | PSYCHO_PCIAFSR_STA)) {
-		psycho_check_iommu_error(pbm, afsr, afar, PCI_ERR);
-		pci_scan_for_target_abort(pbm, pbm->pci_bus);
-	}
-	if (error_bits & (PSYCHO_PCIAFSR_PMA | PSYCHO_PCIAFSR_SMA))
-		pci_scan_for_master_abort(pbm, pbm->pci_bus);
-
-	/* For excessive retries, PSYCHO/PBM will abort the device
-	 * and there is no way to specifically check for excessive
-	 * retries in the config space status registers.  So what
-	 * we hope is that we'll catch it via the master/target
-	 * abort events.
-	 */
-
-	if (error_bits & (PSYCHO_PCIAFSR_PPERR | PSYCHO_PCIAFSR_SPERR))
-		pci_scan_for_parity_error(pbm, pbm->pci_bus);
-
-	return IRQ_HANDLED;
-}
-
 /* XXX What about PowerFail/PowerManagement??? -DaveM */
 #define PSYCHO_ECC_CTRL		0x0020
 #define  PSYCHO_ECCCTRL_EE	 0x8000000000000000UL /* Enable ECC Checking */
@@ -719,7 +285,7 @@ static irqreturn_t psycho_pcierr_intr(int irq, void *dev_id)
 #define  PSYCHO_ECCCTRL_CE	 0x2000000000000000UL /* Enable CE INterrupts */
 static void psycho_register_error_handlers(struct pci_pbm_info *pbm)
 {
-	struct of_device *op = of_find_device_by_node(pbm->prom_node);
+	struct of_device *op = of_find_device_by_node(pbm->op->node);
 	unsigned long base = pbm->controller_regs;
 	u64 tmp;
 	int err;
@@ -762,27 +328,26 @@ static void psycho_register_error_handlers(struct pci_pbm_info *pbm)
 		       "err=%d\n", pbm->name, err);
 
 	/* Enable UE and CE interrupts for controller. */
-	psycho_write(base + PSYCHO_ECC_CTRL,
-		     (PSYCHO_ECCCTRL_EE |
-		      PSYCHO_ECCCTRL_UE |
-		      PSYCHO_ECCCTRL_CE));
+	upa_writeq((PSYCHO_ECCCTRL_EE |
+		    PSYCHO_ECCCTRL_UE |
+		    PSYCHO_ECCCTRL_CE), base + PSYCHO_ECC_CTRL);
 
 	/* Enable PCI Error interrupts and clear error
 	 * bits for each PBM.
 	 */
-	tmp = psycho_read(base + PSYCHO_PCIA_CTRL);
+	tmp = upa_readq(base + PSYCHO_PCIA_CTRL);
 	tmp |= (PSYCHO_PCICTRL_SERR |
 		PSYCHO_PCICTRL_SBH_ERR |
 		PSYCHO_PCICTRL_EEN);
 	tmp &= ~(PSYCHO_PCICTRL_SBH_INT);
-	psycho_write(base + PSYCHO_PCIA_CTRL, tmp);
+	upa_writeq(tmp, base + PSYCHO_PCIA_CTRL);
 		     
-	tmp = psycho_read(base + PSYCHO_PCIB_CTRL);
+	tmp = upa_readq(base + PSYCHO_PCIB_CTRL);
 	tmp |= (PSYCHO_PCICTRL_SERR |
 		PSYCHO_PCICTRL_SBH_ERR |
 		PSYCHO_PCICTRL_EEN);
 	tmp &= ~(PSYCHO_PCICTRL_SBH_INT);
-	psycho_write(base + PSYCHO_PCIB_CTRL, tmp);
+	upa_writeq(tmp, base + PSYCHO_PCIB_CTRL);
 }
 
 /* PSYCHO boot time probing and initialization. */
@@ -803,11 +368,12 @@ static void pbm_config_busmastering(struct pci_pbm_info *pbm)
 	pci_config_write8(addr, 64);
 }
 
-static void __init psycho_scan_bus(struct pci_pbm_info *pbm)
+static void __init psycho_scan_bus(struct pci_pbm_info *pbm,
+				   struct device *parent)
 {
 	pbm_config_busmastering(pbm);
 	pbm->is_66mhz_capable = 0;
-	pbm->pci_bus = pci_scan_one_pbm(pbm);
+	pbm->pci_bus = pci_scan_one_pbm(pbm, parent);
 
 	/* After the PCI bus scan is complete, we can register
 	 * the error interrupt handlers.
@@ -815,61 +381,6 @@ static void __init psycho_scan_bus(struct pci_pbm_info *pbm)
 	psycho_register_error_handlers(pbm);
 }
 
-static int psycho_iommu_init(struct pci_pbm_info *pbm)
-{
-	struct iommu *iommu = pbm->iommu;
-	unsigned long i;
-	u64 control;
-	int err;
-
-	/* Register addresses. */
-	iommu->iommu_control  = pbm->controller_regs + PSYCHO_IOMMU_CONTROL;
-	iommu->iommu_tsbbase  = pbm->controller_regs + PSYCHO_IOMMU_TSBBASE;
-	iommu->iommu_flush    = pbm->controller_regs + PSYCHO_IOMMU_FLUSH;
-	iommu->iommu_tags     = iommu->iommu_flush + (0xa580UL - 0x0210UL);
-
-	/* PSYCHO's IOMMU lacks ctx flushing. */
-	iommu->iommu_ctxflush = 0;
-
-	/* We use the main control register of PSYCHO as the write
-	 * completion register.
-	 */
-	iommu->write_complete_reg = pbm->controller_regs + PSYCHO_CONTROL;
-
-	/*
-	 * Invalidate TLB Entries.
-	 */
-	control = psycho_read(pbm->controller_regs + PSYCHO_IOMMU_CONTROL);
-	control |= PSYCHO_IOMMU_CTRL_DENAB;
-	psycho_write(pbm->controller_regs + PSYCHO_IOMMU_CONTROL, control);
-	for(i = 0; i < 16; i++) {
-		psycho_write(pbm->controller_regs + PSYCHO_IOMMU_TAG + (i * 8UL), 0);
-		psycho_write(pbm->controller_regs + PSYCHO_IOMMU_DATA + (i * 8UL), 0);
-	}
-
-	/* Leave diag mode enabled for full-flushing done
-	 * in pci_iommu.c
-	 */
-	err = iommu_table_init(iommu, IO_TSB_SIZE, 0xc0000000, 0xffffffff,
-			       pbm->numa_node);
-	if (err)
-		return err;
-
-	psycho_write(pbm->controller_regs + PSYCHO_IOMMU_TSBBASE,
-		     __pa(iommu->page_table));
-
-	control = psycho_read(pbm->controller_regs + PSYCHO_IOMMU_CONTROL);
-	control &= ~(PSYCHO_IOMMU_CTRL_TSBSZ | PSYCHO_IOMMU_CTRL_TBWSZ);
-	control |= (PSYCHO_IOMMU_TSBSZ_128K | PSYCHO_IOMMU_CTRL_ENAB);
-	psycho_write(pbm->controller_regs + PSYCHO_IOMMU_CONTROL, control);
-
-	/* If necessary, hook us up for starfire IRQ translations. */
-	if (this_is_starfire)
-		starfire_hookup(pbm->portid);
-
-	return 0;
-}
-
 #define PSYCHO_IRQ_RETRY	0x1a00UL
 #define PSYCHO_PCIA_DIAG	0x2020UL
 #define PSYCHO_PCIB_DIAG	0x4020UL
@@ -886,28 +397,28 @@ static void psycho_controller_hwinit(struct pci_pbm_info *pbm)
 {
 	u64 tmp;
 
-	psycho_write(pbm->controller_regs + PSYCHO_IRQ_RETRY, 5);
+	upa_writeq(5, pbm->controller_regs + PSYCHO_IRQ_RETRY);
 
 	/* Enable arbiter for all PCI slots. */
-	tmp = psycho_read(pbm->controller_regs + PSYCHO_PCIA_CTRL);
+	tmp = upa_readq(pbm->controller_regs + PSYCHO_PCIA_CTRL);
 	tmp |= PSYCHO_PCICTRL_AEN;
-	psycho_write(pbm->controller_regs + PSYCHO_PCIA_CTRL, tmp);
+	upa_writeq(tmp, pbm->controller_regs + PSYCHO_PCIA_CTRL);
 
-	tmp = psycho_read(pbm->controller_regs + PSYCHO_PCIB_CTRL);
+	tmp = upa_readq(pbm->controller_regs + PSYCHO_PCIB_CTRL);
 	tmp |= PSYCHO_PCICTRL_AEN;
-	psycho_write(pbm->controller_regs + PSYCHO_PCIB_CTRL, tmp);
+	upa_writeq(tmp, pbm->controller_regs + PSYCHO_PCIB_CTRL);
 
 	/* Disable DMA write / PIO read synchronization on
 	 * both PCI bus segments.
 	 * [ U2P Erratum 1243770, STP2223BGA data sheet ]
 	 */
-	tmp = psycho_read(pbm->controller_regs + PSYCHO_PCIA_DIAG);
+	tmp = upa_readq(pbm->controller_regs + PSYCHO_PCIA_DIAG);
 	tmp |= PSYCHO_PCIDIAG_DDWSYNC;
-	psycho_write(pbm->controller_regs + PSYCHO_PCIA_DIAG, tmp);
+	upa_writeq(tmp, pbm->controller_regs + PSYCHO_PCIA_DIAG);
 
-	tmp = psycho_read(pbm->controller_regs + PSYCHO_PCIB_DIAG);
+	tmp = upa_readq(pbm->controller_regs + PSYCHO_PCIB_DIAG);
 	tmp |= PSYCHO_PCIDIAG_DDWSYNC;
-	psycho_write(pbm->controller_regs + PSYCHO_PCIB_DIAG, tmp);
+	upa_writeq(tmp, pbm->controller_regs + PSYCHO_PCIB_DIAG);
 }
 
 static void psycho_pbm_strbuf_init(struct pci_pbm_info *pbm,
@@ -920,10 +431,16 @@ static void psycho_pbm_strbuf_init(struct pci_pbm_info *pbm,
 		pbm->stc.strbuf_control  = base + PSYCHO_STRBUF_CONTROL_A;
 		pbm->stc.strbuf_pflush   = base + PSYCHO_STRBUF_FLUSH_A;
 		pbm->stc.strbuf_fsync    = base + PSYCHO_STRBUF_FSYNC_A;
+		pbm->stc.strbuf_err_stat = base + PSYCHO_STC_ERR_A;
+		pbm->stc.strbuf_tag_diag = base + PSYCHO_STC_TAG_A;
+		pbm->stc.strbuf_line_diag= base + PSYCHO_STC_LINE_A;
 	} else {
 		pbm->stc.strbuf_control  = base + PSYCHO_STRBUF_CONTROL_B;
 		pbm->stc.strbuf_pflush   = base + PSYCHO_STRBUF_FLUSH_B;
 		pbm->stc.strbuf_fsync    = base + PSYCHO_STRBUF_FSYNC_B;
+		pbm->stc.strbuf_err_stat = base + PSYCHO_STC_ERR_B;
+		pbm->stc.strbuf_tag_diag = base + PSYCHO_STC_TAG_B;
+		pbm->stc.strbuf_line_diag= base + PSYCHO_STC_LINE_B;
 	}
 	/* PSYCHO's streaming buffer lacks ctx flushing. */
 	pbm->stc.strbuf_ctxflush      = 0;
@@ -946,7 +463,7 @@ static void psycho_pbm_strbuf_init(struct pci_pbm_info *pbm,
 	 */
 #undef PSYCHO_STRBUF_RERUN_ENABLE
 #undef PSYCHO_STRBUF_RERUN_DISABLE
-	control = psycho_read(pbm->stc.strbuf_control);
+	control = upa_readq(pbm->stc.strbuf_control);
 	control |= PSYCHO_STRBUF_CTRL_ENAB;
 	control &= ~(PSYCHO_STRBUF_CTRL_LENAB | PSYCHO_STRBUF_CTRL_LPTR);
 #ifdef PSYCHO_STRBUF_RERUN_ENABLE
@@ -956,7 +473,7 @@ static void psycho_pbm_strbuf_init(struct pci_pbm_info *pbm,
 	control |= PSYCHO_STRBUF_CTRL_RRDIS;
 #endif
 #endif
-	psycho_write(pbm->stc.strbuf_control, control);
+	upa_writeq(control, pbm->stc.strbuf_control);
 
 	pbm->stc.strbuf_enabled = 1;
 }
@@ -968,111 +485,134 @@ static void psycho_pbm_strbuf_init(struct pci_pbm_info *pbm,
 #define PSYCHO_MEMSPACE_B	0x180000000UL
 #define PSYCHO_MEMSPACE_SIZE	0x07fffffffUL
 
-static void __init psycho_pbm_init(struct pci_controller_info *p,
-			    struct device_node *dp, int is_pbm_a)
+static void __init psycho_pbm_init(struct pci_pbm_info *pbm,
+				   struct of_device *op, int is_pbm_a)
 {
-	struct property *prop;
-	struct pci_pbm_info *pbm;
-
-	if (is_pbm_a)
-		pbm = &p->pbm_A;
-	else
-		pbm = &p->pbm_B;
-
-	pbm->next = pci_pbm_root;
-	pci_pbm_root = pbm;
-
-	pbm->numa_node = -1;
-
-	pbm->scan_bus = psycho_scan_bus;
-	pbm->pci_ops = &sun4u_pci_ops;
-	pbm->config_space_reg_bits = 8;
-
-	pbm->index = pci_num_pbms++;
-
-	pbm->chip_type = PBM_CHIP_TYPE_PSYCHO;
-	pbm->chip_version = 0;
-	prop = of_find_property(dp, "version#", NULL);
-	if (prop)
-		pbm->chip_version = *(int *) prop->value;
-	pbm->chip_revision = 0;
-	prop = of_find_property(dp, "module-revision#", NULL);
-	if (prop)
-		pbm->chip_revision = *(int *) prop->value;
-
-	pbm->parent = p;
-	pbm->prom_node = dp;
-	pbm->name = dp->full_name;
-
-	printk("%s: PSYCHO PCI Bus Module ver[%x:%x]\n",
-	       pbm->name,
-	       pbm->chip_version, pbm->chip_revision);
-
-	pci_determine_mem_io_space(pbm);
+	psycho_pbm_init_common(pbm, op, "PSYCHO", PBM_CHIP_TYPE_PSYCHO);
+	psycho_pbm_strbuf_init(pbm, is_pbm_a);
+	psycho_scan_bus(pbm, &op->dev);
+}
 
-	pci_get_pbm_props(pbm);
+static struct pci_pbm_info * __devinit psycho_find_sibling(u32 upa_portid)
+{
+	struct pci_pbm_info *pbm;
 
-	psycho_pbm_strbuf_init(pbm, is_pbm_a);
+	for (pbm = pci_pbm_root; pbm; pbm = pbm->next) {
+		if (pbm->portid == upa_portid)
+			return pbm;
+	}
+	return NULL;
 }
 
 #define PSYCHO_CONFIGSPACE	0x001000000UL
 
-void __init psycho_init(struct device_node *dp, char *model_name)
+static int __devinit psycho_probe(struct of_device *op,
+				  const struct of_device_id *match)
 {
-	struct linux_prom64_registers *pr_regs;
-	struct pci_controller_info *p;
+	const struct linux_prom64_registers *pr_regs;
+	struct device_node *dp = op->node;
 	struct pci_pbm_info *pbm;
 	struct iommu *iommu;
-	struct property *prop;
+	int is_pbm_a, err;
 	u32 upa_portid;
-	int is_pbm_a;
 
-	upa_portid = 0xff;
-	prop = of_find_property(dp, "upa-portid", NULL);
-	if (prop)
-		upa_portid = *(u32 *) prop->value;
+	upa_portid = of_getintprop_default(dp, "upa-portid", 0xff);
 
-	for (pbm = pci_pbm_root; pbm; pbm = pbm->next) {
-		struct pci_controller_info *p = pbm->parent;
+	err = -ENOMEM;
+	pbm = kzalloc(sizeof(*pbm), GFP_KERNEL);
+	if (!pbm) {
+		printk(KERN_ERR PFX "Cannot allocate pci_pbm_info.\n");
+		goto out_err;
+	}
 
-		if (p->pbm_A.portid == upa_portid) {
-			is_pbm_a = (p->pbm_A.prom_node == NULL);
-			psycho_pbm_init(p, dp, is_pbm_a);
-			return;
+	pbm->sibling = psycho_find_sibling(upa_portid);
+	if (pbm->sibling) {
+		iommu = pbm->sibling->iommu;
+	} else {
+		iommu = kzalloc(sizeof(struct iommu), GFP_KERNEL);
+		if (!iommu) {
+			printk(KERN_ERR PFX "Cannot allocate PBM iommu.\n");
+			goto out_free_controller;
 		}
 	}
 
-	p = kzalloc(sizeof(struct pci_controller_info), GFP_ATOMIC);
-	if (!p)
-		goto fatal_memory_error;
-	iommu = kzalloc(sizeof(struct iommu), GFP_ATOMIC);
-	if (!iommu)
-		goto fatal_memory_error;
+	pbm->iommu = iommu;
+	pbm->portid = upa_portid;
 
-	p->pbm_A.iommu = p->pbm_B.iommu = iommu;
+	pr_regs = of_get_property(dp, "reg", NULL);
+	err = -ENODEV;
+	if (!pr_regs) {
+		printk(KERN_ERR PFX "No reg property.\n");
+		goto out_free_iommu;
+	}
 
-	p->pbm_A.portid = upa_portid;
-	p->pbm_B.portid = upa_portid;
+	is_pbm_a = ((pr_regs[0].phys_addr & 0x6000) == 0x2000);
 
-	prop = of_find_property(dp, "reg", NULL);
-	pr_regs = prop->value;
+	pbm->controller_regs = pr_regs[2].phys_addr;
+	pbm->config_space = (pr_regs[2].phys_addr + PSYCHO_CONFIGSPACE);
 
-	p->pbm_A.controller_regs = pr_regs[2].phys_addr;
-	p->pbm_B.controller_regs = pr_regs[2].phys_addr;
+	if (is_pbm_a) {
+		pbm->pci_afsr = pbm->controller_regs + PSYCHO_PCI_AFSR_A;
+		pbm->pci_afar = pbm->controller_regs + PSYCHO_PCI_AFAR_A;
+		pbm->pci_csr  = pbm->controller_regs + PSYCHO_PCIA_CTRL;
+	} else {
+		pbm->pci_afsr = pbm->controller_regs + PSYCHO_PCI_AFSR_B;
+		pbm->pci_afar = pbm->controller_regs + PSYCHO_PCI_AFAR_B;
+		pbm->pci_csr  = pbm->controller_regs + PSYCHO_PCIB_CTRL;
+	}
 
-	p->pbm_A.config_space = p->pbm_B.config_space =
-		(pr_regs[2].phys_addr + PSYCHO_CONFIGSPACE);
+	psycho_controller_hwinit(pbm);
+	if (!pbm->sibling) {
+		err = psycho_iommu_init(pbm, 128, 0xc0000000,
+					0xffffffff, PSYCHO_CONTROL);
+		if (err)
+			goto out_free_iommu;
 
-	psycho_controller_hwinit(&p->pbm_A);
+		/* If necessary, hook us up for starfire IRQ translations. */
+		if (this_is_starfire)
+			starfire_hookup(pbm->portid);
+	}
 
-	if (psycho_iommu_init(&p->pbm_A))
-		goto fatal_memory_error;
+	psycho_pbm_init(pbm, op, is_pbm_a);
 
-	is_pbm_a = ((pr_regs[0].phys_addr & 0x6000) == 0x2000);
-	psycho_pbm_init(p, dp, is_pbm_a);
-	return;
+	pbm->next = pci_pbm_root;
+	pci_pbm_root = pbm;
+
+	if (pbm->sibling)
+		pbm->sibling->sibling = pbm;
+
+	dev_set_drvdata(&op->dev, pbm);
+
+	return 0;
+
+out_free_iommu:
+	if (!pbm->sibling)
+		kfree(pbm->iommu);
 
-fatal_memory_error:
-	prom_printf("PSYCHO: Fatal memory allocation error.\n");
-	prom_halt();
+out_free_controller:
+	kfree(pbm);
+
+out_err:
+	return err;
 }
+
+static struct of_device_id __initdata psycho_match[] = {
+	{
+		.name = "pci",
+		.compatible = "pci108e,8000",
+	},
+	{},
+};
+
+static struct of_platform_driver psycho_driver = {
+	.name		= DRIVER_NAME,
+	.match_table	= psycho_match,
+	.probe		= psycho_probe,
+};
+
+static int __init psycho_init(void)
+{
+	return of_register_driver(&psycho_driver, &of_bus_type);
+}
+
+subsys_initcall(psycho_init);
diff --git a/arch/sparc64/kernel/pci_sabre.c b/arch/sparc64/kernel/pci_sabre.c
index ade5184e75d1..713257b6963c 100644
--- a/arch/sparc64/kernel/pci_sabre.c
+++ b/arch/sparc64/kernel/pci_sabre.c
@@ -16,31 +16,15 @@
 #include <asm/apb.h>
 #include <asm/iommu.h>
 #include <asm/irq.h>
-#include <asm/smp.h>
-#include <asm/oplib.h>
 #include <asm/prom.h>
+#include <asm/upa.h>
 
 #include "pci_impl.h"
 #include "iommu_common.h"
+#include "psycho_common.h"
 
-/* All SABRE registers are 64-bits.  The following accessor
- * routines are how they are accessed.  The REG parameter
- * is a physical address.
- */
-#define sabre_read(__reg) \
-({	u64 __ret; \
-	__asm__ __volatile__("ldxa [%1] %2, %0" \
-			     : "=r" (__ret) \
-			     : "r" (__reg), "i" (ASI_PHYS_BYPASS_EC_E) \
-			     : "memory"); \
-	__ret; \
-})
-#define sabre_write(__reg, __val) \
-	__asm__ __volatile__("stxa %0, [%1] %2" \
-			     : /* no outputs */ \
-			     : "r" (__val), "r" (__reg), \
-			       "i" (ASI_PHYS_BYPASS_EC_E) \
-			     : "memory")
+#define DRIVER_NAME	"sabre"
+#define PFX		DRIVER_NAME ": "
 
 /* SABRE PCI controller register offsets and definitions. */
 #define SABRE_UE_AFSR		0x0030UL
@@ -208,95 +192,6 @@
 static int hummingbird_p;
 static struct pci_bus *sabre_root_bus;
 
-/* SABRE error handling support. */
-static void sabre_check_iommu_error(struct pci_pbm_info *pbm,
-				    unsigned long afsr,
-				    unsigned long afar)
-{
-	struct iommu *iommu = pbm->iommu;
-	unsigned long iommu_tag[16];
-	unsigned long iommu_data[16];
-	unsigned long flags;
-	u64 control;
-	int i;
-
-	spin_lock_irqsave(&iommu->lock, flags);
-	control = sabre_read(iommu->iommu_control);
-	if (control & SABRE_IOMMUCTRL_ERR) {
-		char *type_string;
-
-		/* Clear the error encountered bit.
-		 * NOTE: On Sabre this is write 1 to clear,
-		 *       which is different from Psycho.
-		 */
-		sabre_write(iommu->iommu_control, control);
-		switch((control & SABRE_IOMMUCTRL_ERRSTS) >> 25UL) {
-		case 1:
-			type_string = "Invalid Error";
-			break;
-		case 3:
-			type_string = "ECC Error";
-			break;
-		default:
-			type_string = "Unknown";
-			break;
-		};
-		printk("%s: IOMMU Error, type[%s]\n",
-		       pbm->name, type_string);
-
-		/* Enter diagnostic mode and probe for error'd
-		 * entries in the IOTLB.
-		 */
-		control &= ~(SABRE_IOMMUCTRL_ERRSTS | SABRE_IOMMUCTRL_ERR);
-		sabre_write(iommu->iommu_control,
-			    (control | SABRE_IOMMUCTRL_DENAB));
-		for (i = 0; i < 16; i++) {
-			unsigned long base = pbm->controller_regs;
-
-			iommu_tag[i] =
-				sabre_read(base + SABRE_IOMMU_TAG + (i * 8UL));
-			iommu_data[i] =
-				sabre_read(base + SABRE_IOMMU_DATA + (i * 8UL));
-			sabre_write(base + SABRE_IOMMU_TAG + (i * 8UL), 0);
-			sabre_write(base + SABRE_IOMMU_DATA + (i * 8UL), 0);
-		}
-		sabre_write(iommu->iommu_control, control);
-
-		for (i = 0; i < 16; i++) {
-			unsigned long tag, data;
-
-			tag = iommu_tag[i];
-			if (!(tag & SABRE_IOMMUTAG_ERR))
-				continue;
-
-			data = iommu_data[i];
-			switch((tag & SABRE_IOMMUTAG_ERRSTS) >> 23UL) {
-			case 1:
-				type_string = "Invalid Error";
-				break;
-			case 3:
-				type_string = "ECC Error";
-				break;
-			default:
-				type_string = "Unknown";
-				break;
-			};
-			printk("%s: IOMMU TAG(%d)[RAW(%016lx)error(%s)wr(%d)sz(%dK)vpg(%08lx)]\n",
-			       pbm->name, i, tag, type_string,
-			       ((tag & SABRE_IOMMUTAG_WRITE) ? 1 : 0),
-			       ((tag & SABRE_IOMMUTAG_SIZE) ? 64 : 8),
-			       ((tag & SABRE_IOMMUTAG_VPN) << IOMMU_PAGE_SHIFT));
-			printk("%s: IOMMU DATA(%d)[RAW(%016lx)valid(%d)used(%d)cache(%d)ppg(%016lx)\n",
-			       pbm->name, i, data,
-			       ((data & SABRE_IOMMUDATA_VALID) ? 1 : 0),
-			       ((data & SABRE_IOMMUDATA_USED) ? 1 : 0),
-			       ((data & SABRE_IOMMUDATA_CACHE) ? 1 : 0),
-			       ((data & SABRE_IOMMUDATA_PPN) << IOMMU_PAGE_SHIFT));
-		}
-	}
-	spin_unlock_irqrestore(&iommu->lock, flags);
-}
-
 static irqreturn_t sabre_ue_intr(int irq, void *dev_id)
 {
 	struct pci_pbm_info *pbm = dev_id;
@@ -306,8 +201,8 @@ static irqreturn_t sabre_ue_intr(int irq, void *dev_id)
 	int reported;
 
 	/* Latch uncorrectable error status. */
-	afar = sabre_read(afar_reg);
-	afsr = sabre_read(afsr_reg);
+	afar = upa_readq(afar_reg);
+	afsr = upa_readq(afsr_reg);
 
 	/* Clear the primary/secondary error status bits. */
 	error_bits = afsr &
@@ -316,7 +211,7 @@ static irqreturn_t sabre_ue_intr(int irq, void *dev_id)
 		 SABRE_UEAFSR_SDTE | SABRE_UEAFSR_PDTE);
 	if (!error_bits)
 		return IRQ_NONE;
-	sabre_write(afsr_reg, error_bits);
+	upa_writeq(error_bits, afsr_reg);
 
 	/* Log the error. */
 	printk("%s: Uncorrectable Error, primary error type[%s%s]\n",
@@ -352,7 +247,7 @@ static irqreturn_t sabre_ue_intr(int irq, void *dev_id)
 	printk("]\n");
 
 	/* Interrogate IOMMU for error status. */
-	sabre_check_iommu_error(pbm, afsr, afar);
+	psycho_check_iommu_error(pbm, afsr, afar, UE_ERR);
 
 	return IRQ_HANDLED;
 }
@@ -366,8 +261,8 @@ static irqreturn_t sabre_ce_intr(int irq, void *dev_id)
 	int reported;
 
 	/* Latch error status. */
-	afar = sabre_read(afar_reg);
-	afsr = sabre_read(afsr_reg);
+	afar = upa_readq(afar_reg);
+	afsr = upa_readq(afsr_reg);
 
 	/* Clear primary/secondary error status bits. */
 	error_bits = afsr &
@@ -375,7 +270,7 @@ static irqreturn_t sabre_ce_intr(int irq, void *dev_id)
 		 SABRE_CEAFSR_SDRD | SABRE_CEAFSR_SDWR);
 	if (!error_bits)
 		return IRQ_NONE;
-	sabre_write(afsr_reg, error_bits);
+	upa_writeq(error_bits, afsr_reg);
 
 	/* Log the error. */
 	printk("%s: Correctable Error, primary error type[%s]\n",
@@ -413,136 +308,9 @@ static irqreturn_t sabre_ce_intr(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-static irqreturn_t sabre_pcierr_intr_other(struct pci_pbm_info *pbm)
-{
-	unsigned long csr_reg, csr, csr_error_bits;
-	irqreturn_t ret = IRQ_NONE;
-	u16 stat;
-
-	csr_reg = pbm->controller_regs + SABRE_PCICTRL;
-	csr = sabre_read(csr_reg);
-	csr_error_bits =
-		csr & SABRE_PCICTRL_SERR;
-	if (csr_error_bits) {
-		/* Clear the errors.  */
-		sabre_write(csr_reg, csr);
-
-		/* Log 'em.  */
-		if (csr_error_bits & SABRE_PCICTRL_SERR)
-			printk("%s: PCI SERR signal asserted.\n",
-			       pbm->name);
-		ret = IRQ_HANDLED;
-	}
-	pci_bus_read_config_word(sabre_root_bus, 0,
-				 PCI_STATUS, &stat);
-	if (stat & (PCI_STATUS_PARITY |
-		    PCI_STATUS_SIG_TARGET_ABORT |
-		    PCI_STATUS_REC_TARGET_ABORT |
-		    PCI_STATUS_REC_MASTER_ABORT |
-		    PCI_STATUS_SIG_SYSTEM_ERROR)) {
-		printk("%s: PCI bus error, PCI_STATUS[%04x]\n",
-		       pbm->name, stat);
-		pci_bus_write_config_word(sabre_root_bus, 0,
-					  PCI_STATUS, 0xffff);
-		ret = IRQ_HANDLED;
-	}
-	return ret;
-}
-
-static irqreturn_t sabre_pcierr_intr(int irq, void *dev_id)
-{
-	struct pci_pbm_info *pbm = dev_id;
-	unsigned long afsr_reg, afar_reg;
-	unsigned long afsr, afar, error_bits;
-	int reported;
-
-	afsr_reg = pbm->controller_regs + SABRE_PIOAFSR;
-	afar_reg = pbm->controller_regs + SABRE_PIOAFAR;
-
-	/* Latch error status. */
-	afar = sabre_read(afar_reg);
-	afsr = sabre_read(afsr_reg);
-
-	/* Clear primary/secondary error status bits. */
-	error_bits = afsr &
-		(SABRE_PIOAFSR_PMA | SABRE_PIOAFSR_PTA |
-		 SABRE_PIOAFSR_PRTRY | SABRE_PIOAFSR_PPERR |
-		 SABRE_PIOAFSR_SMA | SABRE_PIOAFSR_STA |
-		 SABRE_PIOAFSR_SRTRY | SABRE_PIOAFSR_SPERR);
-	if (!error_bits)
-		return sabre_pcierr_intr_other(pbm);
-	sabre_write(afsr_reg, error_bits);
-
-	/* Log the error. */
-	printk("%s: PCI Error, primary error type[%s]\n",
-	       pbm->name,
-	       (((error_bits & SABRE_PIOAFSR_PMA) ?
-		 "Master Abort" :
-		 ((error_bits & SABRE_PIOAFSR_PTA) ?
-		  "Target Abort" :
-		  ((error_bits & SABRE_PIOAFSR_PRTRY) ?
-		   "Excessive Retries" :
-		   ((error_bits & SABRE_PIOAFSR_PPERR) ?
-		    "Parity Error" : "???"))))));
-	printk("%s: bytemask[%04lx] was_block(%d)\n",
-	       pbm->name,
-	       (afsr & SABRE_PIOAFSR_BMSK) >> 32UL,
-	       (afsr & SABRE_PIOAFSR_BLK) ? 1 : 0);
-	printk("%s: PCI AFAR [%016lx]\n", pbm->name, afar);
-	printk("%s: PCI Secondary errors [", pbm->name);
-	reported = 0;
-	if (afsr & SABRE_PIOAFSR_SMA) {
-		reported++;
-		printk("(Master Abort)");
-	}
-	if (afsr & SABRE_PIOAFSR_STA) {
-		reported++;
-		printk("(Target Abort)");
-	}
-	if (afsr & SABRE_PIOAFSR_SRTRY) {
-		reported++;
-		printk("(Excessive Retries)");
-	}
-	if (afsr & SABRE_PIOAFSR_SPERR) {
-		reported++;
-		printk("(Parity Error)");
-	}
-	if (!reported)
-		printk("(none)");
-	printk("]\n");
-
-	/* For the error types shown, scan both PCI buses for devices
-	 * which have logged that error type.
-	 */
-
-	/* If we see a Target Abort, this could be the result of an
-	 * IOMMU translation error of some sort.  It is extremely
-	 * useful to log this information as usually it indicates
-	 * a bug in the IOMMU support code or a PCI device driver.
-	 */
-	if (error_bits & (SABRE_PIOAFSR_PTA | SABRE_PIOAFSR_STA)) {
-		sabre_check_iommu_error(pbm, afsr, afar);
-		pci_scan_for_target_abort(pbm, pbm->pci_bus);
-	}
-	if (error_bits & (SABRE_PIOAFSR_PMA | SABRE_PIOAFSR_SMA))
-		pci_scan_for_master_abort(pbm, pbm->pci_bus);
-
-	/* For excessive retries, SABRE/PBM will abort the device
-	 * and there is no way to specifically check for excessive
-	 * retries in the config space status registers.  So what
-	 * we hope is that we'll catch it via the master/target
-	 * abort events.
-	 */
-
-	if (error_bits & (SABRE_PIOAFSR_PPERR | SABRE_PIOAFSR_SPERR))
-		pci_scan_for_parity_error(pbm, pbm->pci_bus);
-
-	return IRQ_HANDLED;
-}
-
 static void sabre_register_error_handlers(struct pci_pbm_info *pbm)
 {
-	struct device_node *dp = pbm->prom_node;
+	struct device_node *dp = pbm->op->node;
 	struct of_device *op;
 	unsigned long base = pbm->controller_regs;
 	u64 tmp;
@@ -568,33 +336,34 @@ static void sabre_register_error_handlers(struct pci_pbm_info *pbm)
 	 * registering the handler so that we don't get spurious
 	 * interrupts.
 	 */
-	sabre_write(base + SABRE_UE_AFSR,
-		    (SABRE_UEAFSR_PDRD | SABRE_UEAFSR_PDWR |
-		     SABRE_UEAFSR_SDRD | SABRE_UEAFSR_SDWR |
-		     SABRE_UEAFSR_SDTE | SABRE_UEAFSR_PDTE));
+	upa_writeq((SABRE_UEAFSR_PDRD | SABRE_UEAFSR_PDWR |
+		    SABRE_UEAFSR_SDRD | SABRE_UEAFSR_SDWR |
+		    SABRE_UEAFSR_SDTE | SABRE_UEAFSR_PDTE),
+		   base + SABRE_UE_AFSR);
 
 	err = request_irq(op->irqs[1], sabre_ue_intr, 0, "SABRE_UE", pbm);
 	if (err)
 		printk(KERN_WARNING "%s: Couldn't register UE, err=%d.\n",
 		       pbm->name, err);
 
-	sabre_write(base + SABRE_CE_AFSR,
-		    (SABRE_CEAFSR_PDRD | SABRE_CEAFSR_PDWR |
-		     SABRE_CEAFSR_SDRD | SABRE_CEAFSR_SDWR));
+	upa_writeq((SABRE_CEAFSR_PDRD | SABRE_CEAFSR_PDWR |
+		    SABRE_CEAFSR_SDRD | SABRE_CEAFSR_SDWR),
+		   base + SABRE_CE_AFSR);
+
 
 	err = request_irq(op->irqs[2], sabre_ce_intr, 0, "SABRE_CE", pbm);
 	if (err)
 		printk(KERN_WARNING "%s: Couldn't register CE, err=%d.\n",
 		       pbm->name, err);
-	err = request_irq(op->irqs[0], sabre_pcierr_intr, 0,
+	err = request_irq(op->irqs[0], psycho_pcierr_intr, 0,
 			  "SABRE_PCIERR", pbm);
 	if (err)
 		printk(KERN_WARNING "%s: Couldn't register PCIERR, err=%d.\n",
 		       pbm->name, err);
 
-	tmp = sabre_read(base + SABRE_PCICTRL);
+	tmp = upa_readq(base + SABRE_PCICTRL);
 	tmp |= SABRE_PCICTRL_ERREN;
-	sabre_write(base + SABRE_PCICTRL, tmp);
+	upa_writeq(tmp, base + SABRE_PCICTRL);
 }
 
 static void apb_init(struct pci_bus *sabre_bus)
@@ -633,7 +402,8 @@ static void apb_init(struct pci_bus *sabre_bus)
 	}
 }
 
-static void __init sabre_scan_bus(struct pci_pbm_info *pbm)
+static void __init sabre_scan_bus(struct pci_pbm_info *pbm,
+				  struct device *parent)
 {
 	static int once;
 
@@ -656,12 +426,12 @@ static void __init sabre_scan_bus(struct pci_pbm_info *pbm)
 	 * to live at bus 0.
 	 */
 	if (once != 0) {
-		prom_printf("SABRE: Multiple controllers unsupported.\n");
-		prom_halt();
+		printk(KERN_ERR PFX "Multiple controllers unsupported.\n");
+		return;
 	}
 	once++;
 
-	pbm->pci_bus = pci_scan_one_pbm(pbm);
+	pbm->pci_bus = pci_scan_one_pbm(pbm, parent);
 	if (!pbm->pci_bus)
 		return;
 
@@ -672,133 +442,58 @@ static void __init sabre_scan_bus(struct pci_pbm_info *pbm)
 	sabre_register_error_handlers(pbm);
 }
 
-static int sabre_iommu_init(struct pci_pbm_info *pbm,
-			    int tsbsize, unsigned long dvma_offset,
-			    u32 dma_mask)
-{
-	struct iommu *iommu = pbm->iommu;
-	unsigned long i;
-	u64 control;
-	int err;
-
-	/* Register addresses. */
-	iommu->iommu_control  = pbm->controller_regs + SABRE_IOMMU_CONTROL;
-	iommu->iommu_tsbbase  = pbm->controller_regs + SABRE_IOMMU_TSBBASE;
-	iommu->iommu_flush    = pbm->controller_regs + SABRE_IOMMU_FLUSH;
-	iommu->iommu_tags     = iommu->iommu_flush + (0xa580UL - 0x0210UL);
-	iommu->write_complete_reg = pbm->controller_regs + SABRE_WRSYNC;
-	/* Sabre's IOMMU lacks ctx flushing. */
-	iommu->iommu_ctxflush = 0;
-                                        
-	/* Invalidate TLB Entries. */
-	control = sabre_read(pbm->controller_regs + SABRE_IOMMU_CONTROL);
-	control |= SABRE_IOMMUCTRL_DENAB;
-	sabre_write(pbm->controller_regs + SABRE_IOMMU_CONTROL, control);
-
-	for(i = 0; i < 16; i++) {
-		sabre_write(pbm->controller_regs + SABRE_IOMMU_TAG + (i * 8UL), 0);
-		sabre_write(pbm->controller_regs + SABRE_IOMMU_DATA + (i * 8UL), 0);
-	}
-
-	/* Leave diag mode enabled for full-flushing done
-	 * in pci_iommu.c
-	 */
-	err = iommu_table_init(iommu, tsbsize * 1024 * 8,
-			       dvma_offset, dma_mask, pbm->numa_node);
-	if (err)
-		return err;
-
-	sabre_write(pbm->controller_regs + SABRE_IOMMU_TSBBASE,
-		    __pa(iommu->page_table));
-
-	control = sabre_read(pbm->controller_regs + SABRE_IOMMU_CONTROL);
-	control &= ~(SABRE_IOMMUCTRL_TSBSZ | SABRE_IOMMUCTRL_TBWSZ);
-	control |= SABRE_IOMMUCTRL_ENAB;
-	switch(tsbsize) {
-	case 64:
-		control |= SABRE_IOMMU_TSBSZ_64K;
-		break;
-	case 128:
-		control |= SABRE_IOMMU_TSBSZ_128K;
-		break;
-	default:
-		prom_printf("iommu_init: Illegal TSB size %d\n", tsbsize);
-		prom_halt();
-		break;
-	}
-	sabre_write(pbm->controller_regs + SABRE_IOMMU_CONTROL, control);
-
-	return 0;
-}
-
-static void __init sabre_pbm_init(struct pci_controller_info *p,
-				  struct pci_pbm_info *pbm, struct device_node *dp)
+static void __init sabre_pbm_init(struct pci_pbm_info *pbm,
+				  struct of_device *op)
 {
-	pbm->name = dp->full_name;
-	printk("%s: SABRE PCI Bus Module\n", pbm->name);
-
-	pbm->numa_node = -1;
-
-	pbm->scan_bus = sabre_scan_bus;
-	pbm->pci_ops = &sun4u_pci_ops;
-	pbm->config_space_reg_bits = 8;
-
-	pbm->index = pci_num_pbms++;
-
-	pbm->chip_type = PBM_CHIP_TYPE_SABRE;
-	pbm->parent = p;
-	pbm->prom_node = dp;
-	pci_get_pbm_props(pbm);
-
-	pci_determine_mem_io_space(pbm);
+	psycho_pbm_init_common(pbm, op, "SABRE", PBM_CHIP_TYPE_SABRE);
+	pbm->pci_afsr = pbm->controller_regs + SABRE_PIOAFSR;
+	pbm->pci_afar = pbm->controller_regs + SABRE_PIOAFAR;
+	pbm->pci_csr = pbm->controller_regs + SABRE_PCICTRL;
+	sabre_scan_bus(pbm, &op->dev);
 }
 
-void __init sabre_init(struct device_node *dp, char *model_name)
+static int __devinit sabre_probe(struct of_device *op,
+				 const struct of_device_id *match)
 {
 	const struct linux_prom64_registers *pr_regs;
-	struct pci_controller_info *p;
+	struct device_node *dp = op->node;
 	struct pci_pbm_info *pbm;
+	u32 upa_portid, dma_mask;
 	struct iommu *iommu;
-	int tsbsize;
+	int tsbsize, err;
 	const u32 *vdma;
-	u32 upa_portid, dma_mask;
 	u64 clear_irq;
 
-	hummingbird_p = 0;
-	if (!strcmp(model_name, "pci108e,a001"))
-		hummingbird_p = 1;
-	else if (!strcmp(model_name, "SUNW,sabre")) {
-		const char *compat = of_get_property(dp, "compatible", NULL);
-		if (compat && !strcmp(compat, "pci108e,a001"))
-			hummingbird_p = 1;
-		if (!hummingbird_p) {
-			struct device_node *dp;
-
-			/* Of course, Sun has to encode things a thousand
-			 * different ways, inconsistently.
-			 */
-			for_each_node_by_type(dp, "cpu") {
-				if (!strcmp(dp->name, "SUNW,UltraSPARC-IIe"))
-					hummingbird_p = 1;
-			}
+	hummingbird_p = (match->data != NULL);
+	if (!hummingbird_p) {
+		struct device_node *cpu_dp;
+
+		/* Of course, Sun has to encode things a thousand
+		 * different ways, inconsistently.
+		 */
+		for_each_node_by_type(cpu_dp, "cpu") {
+			if (!strcmp(cpu_dp->name, "SUNW,UltraSPARC-IIe"))
+				hummingbird_p = 1;
 		}
 	}
 
-	p = kzalloc(sizeof(*p), GFP_ATOMIC);
-	if (!p)
-		goto fatal_memory_error;
+	err = -ENOMEM;
+	pbm = kzalloc(sizeof(*pbm), GFP_KERNEL);
+	if (!pbm) {
+		printk(KERN_ERR PFX "Cannot allocate pci_pbm_info.\n");
+		goto out_err;
+	}
+
+	iommu = kzalloc(sizeof(*iommu), GFP_KERNEL);
+	if (!iommu) {
+		printk(KERN_ERR PFX "Cannot allocate PBM iommu.\n");
+		goto out_free_controller;
+	}
 
-	iommu = kzalloc(sizeof(*iommu), GFP_ATOMIC);
-	if (!iommu)
-		goto fatal_memory_error;
-	pbm = &p->pbm_A;
 	pbm->iommu = iommu;
 
 	upa_portid = of_getintprop_default(dp, "upa-portid", 0xff);
 
-	pbm->next = pci_pbm_root;
-	pci_pbm_root = pbm;
-
 	pbm->portid = upa_portid;
 
 	/*
@@ -806,6 +501,11 @@ void __init sabre_init(struct device_node *dp, char *model_name)
 	 */
 	
 	pr_regs = of_get_property(dp, "reg", NULL);
+	err = -ENODEV;
+	if (!pr_regs) {
+		printk(KERN_ERR PFX "No reg property\n");
+		goto out_free_iommu;
+	}
 
 	/*
 	 * First REG in property is base of entire SABRE register space.
@@ -816,22 +516,25 @@ void __init sabre_init(struct device_node *dp, char *model_name)
 
 	/* PCI first */
 	for (clear_irq = SABRE_ICLR_A_SLOT0; clear_irq < SABRE_ICLR_B_SLOT0 + 0x80; clear_irq += 8)
-		sabre_write(pbm->controller_regs + clear_irq, 0x0UL);
+		upa_writeq(0x0UL, pbm->controller_regs + clear_irq);
 
 	/* Then OBIO */
 	for (clear_irq = SABRE_ICLR_SCSI; clear_irq < SABRE_ICLR_SCSI + 0x80; clear_irq += 8)
-		sabre_write(pbm->controller_regs + clear_irq, 0x0UL);
+		upa_writeq(0x0UL, pbm->controller_regs + clear_irq);
 
 	/* Error interrupts are enabled later after the bus scan. */
-	sabre_write(pbm->controller_regs + SABRE_PCICTRL,
-		    (SABRE_PCICTRL_MRLEN   | SABRE_PCICTRL_SERR |
-		     SABRE_PCICTRL_ARBPARK | SABRE_PCICTRL_AEN));
+	upa_writeq((SABRE_PCICTRL_MRLEN   | SABRE_PCICTRL_SERR |
+		    SABRE_PCICTRL_ARBPARK | SABRE_PCICTRL_AEN),
+		   pbm->controller_regs + SABRE_PCICTRL);
 
 	/* Now map in PCI config space for entire SABRE. */
-	pbm->config_space =
-		(pbm->controller_regs + SABRE_CONFIGSPACE);
+	pbm->config_space = pbm->controller_regs + SABRE_CONFIGSPACE;
 
 	vdma = of_get_property(dp, "virtual-dma", NULL);
+	if (!vdma) {
+		printk(KERN_ERR PFX "No virtual-dma property\n");
+		goto out_free_iommu;
+	}
 
 	dma_mask = vdma[0];
 	switch(vdma[1]) {
@@ -849,20 +552,58 @@ void __init sabre_init(struct device_node *dp, char *model_name)
 			tsbsize = 128;
 			break;
 		default:
-			prom_printf("SABRE: strange virtual-dma size.\n");
-			prom_halt();
+			printk(KERN_ERR PFX "Strange virtual-dma size.\n");
+			goto out_free_iommu;
 	}
 
-	if (sabre_iommu_init(pbm, tsbsize, vdma[0], dma_mask))
-		goto fatal_memory_error;
+	err = psycho_iommu_init(pbm, tsbsize, vdma[0], dma_mask, SABRE_WRSYNC);
+	if (err)
+		goto out_free_iommu;
 
 	/*
 	 * Look for APB underneath.
 	 */
-	sabre_pbm_init(p, pbm, dp);
-	return;
+	sabre_pbm_init(pbm, op);
 
-fatal_memory_error:
-	prom_printf("SABRE: Fatal memory allocation error.\n");
-	prom_halt();
+	pbm->next = pci_pbm_root;
+	pci_pbm_root = pbm;
+
+	dev_set_drvdata(&op->dev, pbm);
+
+	return 0;
+
+out_free_iommu:
+	kfree(pbm->iommu);
+
+out_free_controller:
+	kfree(pbm);
+
+out_err:
+	return err;
+}
+
+static struct of_device_id __initdata sabre_match[] = {
+	{
+		.name = "pci",
+		.compatible = "pci108e,a001",
+		.data = (void *) 1,
+	},
+	{
+		.name = "pci",
+		.compatible = "pci108e,a000",
+	},
+	{},
+};
+
+static struct of_platform_driver sabre_driver = {
+	.name		= DRIVER_NAME,
+	.match_table	= sabre_match,
+	.probe		= sabre_probe,
+};
+
+static int __init sabre_init(void)
+{
+	return of_register_driver(&sabre_driver, &of_bus_type);
 }
+
+subsys_initcall(sabre_init);
diff --git a/arch/sparc64/kernel/pci_schizo.c b/arch/sparc64/kernel/pci_schizo.c
index 9248c6737f0e..45d9dba1ba11 100644
--- a/arch/sparc64/kernel/pci_schizo.c
+++ b/arch/sparc64/kernel/pci_schizo.c
@@ -1,6 +1,6 @@
 /* pci_schizo.c: SCHIZO/TOMATILLO specific PCI controller support.
  *
- * Copyright (C) 2001, 2002, 2003, 2007 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 2001, 2002, 2003, 2007, 2008 David S. Miller (davem@davemloft.net)
  */
 
 #include <linux/kernel.h>
@@ -13,32 +13,15 @@
 
 #include <asm/iommu.h>
 #include <asm/irq.h>
-#include <asm/upa.h>
 #include <asm/pstate.h>
 #include <asm/prom.h>
-#include <asm/oplib.h>
+#include <asm/upa.h>
 
 #include "pci_impl.h"
 #include "iommu_common.h"
 
-/* All SCHIZO registers are 64-bits.  The following accessor
- * routines are how they are accessed.  The REG parameter
- * is a physical address.
- */
-#define schizo_read(__reg) \
-({	u64 __ret; \
-	__asm__ __volatile__("ldxa [%1] %2, %0" \
-			     : "=r" (__ret) \
-			     : "r" (__reg), "i" (ASI_PHYS_BYPASS_EC_E) \
-			     : "memory"); \
-	__ret; \
-})
-#define schizo_write(__reg, __val) \
-	__asm__ __volatile__("stxa %0, [%1] %2" \
-			     : /* no outputs */ \
-			     : "r" (__val), "r" (__reg), \
-			       "i" (ASI_PHYS_BYPASS_EC_E) \
-			     : "memory")
+#define DRIVER_NAME	"schizo"
+#define PFX		DRIVER_NAME ": "
 
 /* This is a convention that at least Excalibur and Merlin
  * follow.  I suppose the SCHIZO used in Starcat and friends
@@ -163,25 +146,25 @@ static void __schizo_check_stc_error_pbm(struct pci_pbm_info *pbm,
 	 * invalidating it before it has a chance to reach
 	 * main memory.
 	 */
-	control = schizo_read(strbuf->strbuf_control);
-	schizo_write(strbuf->strbuf_control,
-		     (control | SCHIZO_STRBUF_CTRL_DENAB));
+	control = upa_readq(strbuf->strbuf_control);
+	upa_writeq((control | SCHIZO_STRBUF_CTRL_DENAB),
+		   strbuf->strbuf_control);
 	for (i = 0; i < 128; i++) {
 		unsigned long val;
 
-		val = schizo_read(err_base + (i * 8UL));
-		schizo_write(err_base + (i * 8UL), 0UL);
+		val = upa_readq(err_base + (i * 8UL));
+		upa_writeq(0UL, err_base + (i * 8UL));
 		stc_error_buf[i] = val;
 	}
 	for (i = 0; i < 16; i++) {
-		stc_tag_buf[i] = schizo_read(tag_base + (i * 8UL));
-		stc_line_buf[i] = schizo_read(line_base + (i * 8UL));
-		schizo_write(tag_base + (i * 8UL), 0UL);
-		schizo_write(line_base + (i * 8UL), 0UL);
+		stc_tag_buf[i] = upa_readq(tag_base + (i * 8UL));
+		stc_line_buf[i] = upa_readq(line_base + (i * 8UL));
+		upa_writeq(0UL, tag_base + (i * 8UL));
+		upa_writeq(0UL, line_base + (i * 8UL));
 	}
 
 	/* OK, state is logged, exit diagnostic mode. */
-	schizo_write(strbuf->strbuf_control, control);
+	upa_writeq(control, strbuf->strbuf_control);
 
 	for (i = 0; i < 16; i++) {
 		int j, saw_error, first, last;
@@ -258,14 +241,14 @@ static void schizo_check_iommu_error_pbm(struct pci_pbm_info *pbm,
 	int i;
 
 	spin_lock_irqsave(&iommu->lock, flags);
-	control = schizo_read(iommu->iommu_control);
+	control = upa_readq(iommu->iommu_control);
 	if (control & SCHIZO_IOMMU_CTRL_XLTEERR) {
 		unsigned long base;
 		char *type_string;
 
 		/* Clear the error encountered bit. */
 		control &= ~SCHIZO_IOMMU_CTRL_XLTEERR;
-		schizo_write(iommu->iommu_control, control);
+		upa_writeq(control, iommu->iommu_control);
 
 		switch((control & SCHIZO_IOMMU_CTRL_XLTESTAT) >> 25UL) {
 		case 0:
@@ -295,24 +278,24 @@ static void schizo_check_iommu_error_pbm(struct pci_pbm_info *pbm,
 		 * get as much diagnostic information to the
 		 * console as we can.
 		 */
-		schizo_write(iommu->iommu_control,
-			     control | SCHIZO_IOMMU_CTRL_DENAB);
+		upa_writeq(control | SCHIZO_IOMMU_CTRL_DENAB,
+			   iommu->iommu_control);
 
 		base = pbm->pbm_regs;
 
 		for (i = 0; i < 16; i++) {
 			iommu_tag[i] =
-				schizo_read(base + SCHIZO_IOMMU_TAG + (i * 8UL));
+				upa_readq(base + SCHIZO_IOMMU_TAG + (i * 8UL));
 			iommu_data[i] =
-				schizo_read(base + SCHIZO_IOMMU_DATA + (i * 8UL));
+				upa_readq(base + SCHIZO_IOMMU_DATA + (i * 8UL));
 
 			/* Now clear out the entry. */
-			schizo_write(base + SCHIZO_IOMMU_TAG + (i * 8UL), 0);
-			schizo_write(base + SCHIZO_IOMMU_DATA + (i * 8UL), 0);
+			upa_writeq(0, base + SCHIZO_IOMMU_TAG + (i * 8UL));
+			upa_writeq(0, base + SCHIZO_IOMMU_DATA + (i * 8UL));
 		}
 
 		/* Leave diagnostic mode. */
-		schizo_write(iommu->iommu_control, control);
+		upa_writeq(control, iommu->iommu_control);
 
 		for (i = 0; i < 16; i++) {
 			unsigned long tag, data;
@@ -357,11 +340,12 @@ static void schizo_check_iommu_error_pbm(struct pci_pbm_info *pbm,
 	spin_unlock_irqrestore(&iommu->lock, flags);
 }
 
-static void schizo_check_iommu_error(struct pci_controller_info *p,
+static void schizo_check_iommu_error(struct pci_pbm_info *pbm,
 				     enum schizo_error_type type)
 {
-	schizo_check_iommu_error_pbm(&p->pbm_A, type);
-	schizo_check_iommu_error_pbm(&p->pbm_B, type);
+	schizo_check_iommu_error_pbm(pbm, type);
+	if (pbm->sibling)
+		schizo_check_iommu_error_pbm(pbm->sibling, type);
 }
 
 /* Uncorrectable ECC error status gathering. */
@@ -386,14 +370,13 @@ static void schizo_check_iommu_error(struct pci_controller_info *p,
 static irqreturn_t schizo_ue_intr(int irq, void *dev_id)
 {
 	struct pci_pbm_info *pbm = dev_id;
-	struct pci_controller_info *p = pbm->parent;
 	unsigned long afsr_reg = pbm->controller_regs + SCHIZO_UE_AFSR;
 	unsigned long afar_reg = pbm->controller_regs + SCHIZO_UE_AFAR;
 	unsigned long afsr, afar, error_bits;
 	int reported, limit;
 
 	/* Latch uncorrectable error status. */
-	afar = schizo_read(afar_reg);
+	afar = upa_readq(afar_reg);
 
 	/* If either of the error pending bits are set in the
 	 * AFSR, the error status is being actively updated by
@@ -401,7 +384,7 @@ static irqreturn_t schizo_ue_intr(int irq, void *dev_id)
 	 */
 	limit = 1000;
 	do {
-		afsr = schizo_read(afsr_reg);
+		afsr = upa_readq(afsr_reg);
 	} while ((afsr & SCHIZO_UEAFSR_ERRPNDG) != 0 && --limit);
 
 	/* Clear the primary/secondary error status bits. */
@@ -410,7 +393,7 @@ static irqreturn_t schizo_ue_intr(int irq, void *dev_id)
 		 SCHIZO_UEAFSR_SPIO | SCHIZO_UEAFSR_SDMA);
 	if (!error_bits)
 		return IRQ_NONE;
-	schizo_write(afsr_reg, error_bits);
+	upa_writeq(error_bits, afsr_reg);
 
 	/* Log the error. */
 	printk("%s: Uncorrectable Error, primary error type[%s]\n",
@@ -449,7 +432,7 @@ static irqreturn_t schizo_ue_intr(int irq, void *dev_id)
 	printk("]\n");
 
 	/* Interrogate IOMMU for error status. */
-	schizo_check_iommu_error(p, UE_ERR);
+	schizo_check_iommu_error(pbm, UE_ERR);
 
 	return IRQ_HANDLED;
 }
@@ -481,7 +464,7 @@ static irqreturn_t schizo_ce_intr(int irq, void *dev_id)
 	int reported, limit;
 
 	/* Latch error status. */
-	afar = schizo_read(afar_reg);
+	afar = upa_readq(afar_reg);
 
 	/* If either of the error pending bits are set in the
 	 * AFSR, the error status is being actively updated by
@@ -489,7 +472,7 @@ static irqreturn_t schizo_ce_intr(int irq, void *dev_id)
 	 */
 	limit = 1000;
 	do {
-		afsr = schizo_read(afsr_reg);
+		afsr = upa_readq(afsr_reg);
 	} while ((afsr & SCHIZO_UEAFSR_ERRPNDG) != 0 && --limit);
 
 	/* Clear primary/secondary error status bits. */
@@ -498,7 +481,7 @@ static irqreturn_t schizo_ce_intr(int irq, void *dev_id)
 		 SCHIZO_CEAFSR_SPIO | SCHIZO_CEAFSR_SDMA);
 	if (!error_bits)
 		return IRQ_NONE;
-	schizo_write(afsr_reg, error_bits);
+	upa_writeq(error_bits, afsr_reg);
 
 	/* Log the error. */
 	printk("%s: Correctable Error, primary error type[%s]\n",
@@ -600,7 +583,7 @@ static irqreturn_t schizo_pcierr_intr_other(struct pci_pbm_info *pbm)
 	u16 stat;
 
 	csr_reg = pbm->pbm_regs + SCHIZO_PCI_CTRL;
-	csr = schizo_read(csr_reg);
+	csr = upa_readq(csr_reg);
 	csr_error_bits =
 		csr & (SCHIZO_PCICTRL_BUS_UNUS |
 		       SCHIZO_PCICTRL_TTO_ERR |
@@ -610,7 +593,7 @@ static irqreturn_t schizo_pcierr_intr_other(struct pci_pbm_info *pbm)
 		       SCHIZO_PCICTRL_SERR);
 	if (csr_error_bits) {
 		/* Clear the errors.  */
-		schizo_write(csr_reg, csr);
+		upa_writeq(csr, csr_reg);
 
 		/* Log 'em.  */
 		if (csr_error_bits & SCHIZO_PCICTRL_BUS_UNUS)
@@ -650,7 +633,6 @@ static irqreturn_t schizo_pcierr_intr_other(struct pci_pbm_info *pbm)
 static irqreturn_t schizo_pcierr_intr(int irq, void *dev_id)
 {
 	struct pci_pbm_info *pbm = dev_id;
-	struct pci_controller_info *p = pbm->parent;
 	unsigned long afsr_reg, afar_reg, base;
 	unsigned long afsr, afar, error_bits;
 	int reported;
@@ -661,8 +643,8 @@ static irqreturn_t schizo_pcierr_intr(int irq, void *dev_id)
 	afar_reg = base + SCHIZO_PCI_AFAR;
 
 	/* Latch error status. */
-	afar = schizo_read(afar_reg);
-	afsr = schizo_read(afsr_reg);
+	afar = upa_readq(afar_reg);
+	afsr = upa_readq(afsr_reg);
 
 	/* Clear primary/secondary error status bits. */
 	error_bits = afsr &
@@ -674,7 +656,7 @@ static irqreturn_t schizo_pcierr_intr(int irq, void *dev_id)
 		 SCHIZO_PCIAFSR_STTO | SCHIZO_PCIAFSR_SUNUS);
 	if (!error_bits)
 		return schizo_pcierr_intr_other(pbm);
-	schizo_write(afsr_reg, error_bits);
+	upa_writeq(error_bits, afsr_reg);
 
 	/* Log the error. */
 	printk("%s: PCI Error, primary error type[%s]\n",
@@ -744,7 +726,7 @@ static irqreturn_t schizo_pcierr_intr(int irq, void *dev_id)
 	 * a bug in the IOMMU support code or a PCI device driver.
 	 */
 	if (error_bits & (SCHIZO_PCIAFSR_PTA | SCHIZO_PCIAFSR_STA)) {
-		schizo_check_iommu_error(p, PCI_ERR);
+		schizo_check_iommu_error(pbm, PCI_ERR);
 		pci_scan_for_target_abort(pbm, pbm->pci_bus);
 	}
 	if (error_bits & (SCHIZO_PCIAFSR_PMA | SCHIZO_PCIAFSR_SMA))
@@ -805,12 +787,11 @@ static irqreturn_t schizo_pcierr_intr(int irq, void *dev_id)
 static irqreturn_t schizo_safarierr_intr(int irq, void *dev_id)
 {
 	struct pci_pbm_info *pbm = dev_id;
-	struct pci_controller_info *p = pbm->parent;
 	u64 errlog;
 
-	errlog = schizo_read(pbm->controller_regs + SCHIZO_SAFARI_ERRLOG);
-	schizo_write(pbm->controller_regs + SCHIZO_SAFARI_ERRLOG,
-		     errlog & ~(SAFARI_ERRLOG_ERROUT));
+	errlog = upa_readq(pbm->controller_regs + SCHIZO_SAFARI_ERRLOG);
+	upa_writeq(errlog & ~(SAFARI_ERRLOG_ERROUT),
+		   pbm->controller_regs + SCHIZO_SAFARI_ERRLOG);
 
 	if (!(errlog & BUS_ERROR_UNMAP)) {
 		printk("%s: Unexpected Safari/JBUS error interrupt, errlog[%016lx]\n",
@@ -821,7 +802,7 @@ static irqreturn_t schizo_safarierr_intr(int irq, void *dev_id)
 
 	printk("%s: Safari/JBUS interrupt, UNMAPPED error, interrogating IOMMUs.\n",
 	       pbm->name);
-	schizo_check_iommu_error(p, SAFARI_ERR);
+	schizo_check_iommu_error(pbm, SAFARI_ERR);
 
 	return IRQ_HANDLED;
 }
@@ -863,7 +844,7 @@ static int pbm_routes_this_ino(struct pci_pbm_info *pbm, u32 ino)
  */
 static void tomatillo_register_error_handlers(struct pci_pbm_info *pbm)
 {
-	struct of_device *op = of_find_device_by_node(pbm->prom_node);
+	struct of_device *op = of_find_device_by_node(pbm->op->node);
 	u64 tmp, err_mask, err_no_mask;
 	int err;
 
@@ -910,10 +891,9 @@ static void tomatillo_register_error_handlers(struct pci_pbm_info *pbm)
 	}
 
 	/* Enable UE and CE interrupts for controller. */
-	schizo_write(pbm->controller_regs + SCHIZO_ECC_CTRL,
-		     (SCHIZO_ECCCTRL_EE |
-		      SCHIZO_ECCCTRL_UE |
-		      SCHIZO_ECCCTRL_CE));
+	upa_writeq((SCHIZO_ECCCTRL_EE |
+		    SCHIZO_ECCCTRL_UE |
+		    SCHIZO_ECCCTRL_CE), pbm->controller_regs + SCHIZO_ECC_CTRL);
 
 	/* Enable PCI Error interrupts and clear error
 	 * bits.
@@ -926,10 +906,10 @@ static void tomatillo_register_error_handlers(struct pci_pbm_info *pbm)
 
 	err_no_mask = SCHIZO_PCICTRL_DTO_ERR;
 
-	tmp = schizo_read(pbm->pbm_regs + SCHIZO_PCI_CTRL);
+	tmp = upa_readq(pbm->pbm_regs + SCHIZO_PCI_CTRL);
 	tmp |= err_mask;
 	tmp &= ~err_no_mask;
-	schizo_write(pbm->pbm_regs + SCHIZO_PCI_CTRL, tmp);
+	upa_writeq(tmp, pbm->pbm_regs + SCHIZO_PCI_CTRL);
 
 	err_mask = (SCHIZO_PCIAFSR_PMA | SCHIZO_PCIAFSR_PTA |
 		    SCHIZO_PCIAFSR_PRTRY | SCHIZO_PCIAFSR_PPERR |
@@ -938,7 +918,7 @@ static void tomatillo_register_error_handlers(struct pci_pbm_info *pbm)
 		    SCHIZO_PCIAFSR_SRTRY | SCHIZO_PCIAFSR_SPERR |
 		    SCHIZO_PCIAFSR_STTO);
 
-	schizo_write(pbm->pbm_regs + SCHIZO_PCI_AFSR, err_mask);
+	upa_writeq(err_mask, pbm->pbm_regs + SCHIZO_PCI_AFSR);
 
 	err_mask = (BUS_ERROR_BADCMD | BUS_ERROR_SNOOP_GR |
 		    BUS_ERROR_SNOOP_PCI | BUS_ERROR_SNOOP_RD |
@@ -950,16 +930,16 @@ static void tomatillo_register_error_handlers(struct pci_pbm_info *pbm)
 		    BUS_ERROR_APERR | BUS_ERROR_UNMAP |
 		    BUS_ERROR_BUSERR | BUS_ERROR_TIMEOUT);
 
-	schizo_write(pbm->controller_regs + SCHIZO_SAFARI_ERRCTRL,
-		     (SCHIZO_SAFERRCTRL_EN | err_mask));
+	upa_writeq((SCHIZO_SAFERRCTRL_EN | err_mask),
+		   pbm->controller_regs + SCHIZO_SAFARI_ERRCTRL);
 
-	schizo_write(pbm->controller_regs + SCHIZO_SAFARI_IRQCTRL,
-		     (SCHIZO_SAFIRQCTRL_EN | (BUS_ERROR_UNMAP)));
+	upa_writeq((SCHIZO_SAFIRQCTRL_EN | (BUS_ERROR_UNMAP)),
+		   pbm->controller_regs + SCHIZO_SAFARI_IRQCTRL);
 }
 
 static void schizo_register_error_handlers(struct pci_pbm_info *pbm)
 {
-	struct of_device *op = of_find_device_by_node(pbm->prom_node);
+	struct of_device *op = of_find_device_by_node(pbm->op->node);
 	u64 tmp, err_mask, err_no_mask;
 	int err;
 
@@ -1006,10 +986,9 @@ static void schizo_register_error_handlers(struct pci_pbm_info *pbm)
 	}
 
 	/* Enable UE and CE interrupts for controller. */
-	schizo_write(pbm->controller_regs + SCHIZO_ECC_CTRL,
-		     (SCHIZO_ECCCTRL_EE |
-		      SCHIZO_ECCCTRL_UE |
-		      SCHIZO_ECCCTRL_CE));
+	upa_writeq((SCHIZO_ECCCTRL_EE |
+		    SCHIZO_ECCCTRL_UE |
+		    SCHIZO_ECCCTRL_CE), pbm->controller_regs + SCHIZO_ECC_CTRL);
 
 	err_mask = (SCHIZO_PCICTRL_BUS_UNUS |
 		    SCHIZO_PCICTRL_ESLCK |
@@ -1025,18 +1004,18 @@ static void schizo_register_error_handlers(struct pci_pbm_info *pbm)
 	/* Enable PCI Error interrupts and clear error
 	 * bits for each PBM.
 	 */
-	tmp = schizo_read(pbm->pbm_regs + SCHIZO_PCI_CTRL);
+	tmp = upa_readq(pbm->pbm_regs + SCHIZO_PCI_CTRL);
 	tmp |= err_mask;
 	tmp &= ~err_no_mask;
-	schizo_write(pbm->pbm_regs + SCHIZO_PCI_CTRL, tmp);
+	upa_writeq(tmp, pbm->pbm_regs + SCHIZO_PCI_CTRL);
 
-	schizo_write(pbm->pbm_regs + SCHIZO_PCI_AFSR,
-		     (SCHIZO_PCIAFSR_PMA | SCHIZO_PCIAFSR_PTA |
-		      SCHIZO_PCIAFSR_PRTRY | SCHIZO_PCIAFSR_PPERR |
-		      SCHIZO_PCIAFSR_PTTO | SCHIZO_PCIAFSR_PUNUS |
-		      SCHIZO_PCIAFSR_SMA | SCHIZO_PCIAFSR_STA |
-		      SCHIZO_PCIAFSR_SRTRY | SCHIZO_PCIAFSR_SPERR |
-		      SCHIZO_PCIAFSR_STTO | SCHIZO_PCIAFSR_SUNUS));
+	upa_writeq((SCHIZO_PCIAFSR_PMA | SCHIZO_PCIAFSR_PTA |
+		    SCHIZO_PCIAFSR_PRTRY | SCHIZO_PCIAFSR_PPERR |
+		    SCHIZO_PCIAFSR_PTTO | SCHIZO_PCIAFSR_PUNUS |
+		    SCHIZO_PCIAFSR_SMA | SCHIZO_PCIAFSR_STA |
+		    SCHIZO_PCIAFSR_SRTRY | SCHIZO_PCIAFSR_SPERR |
+		    SCHIZO_PCIAFSR_STTO | SCHIZO_PCIAFSR_SUNUS),
+		   pbm->pbm_regs + SCHIZO_PCI_AFSR);
 
 	/* Make all Safari error conditions fatal except unmapped
 	 * errors which we make generate interrupts.
@@ -1063,8 +1042,8 @@ static void schizo_register_error_handlers(struct pci_pbm_info *pbm)
 		      BUS_ERROR_CPU0PS | BUS_ERROR_CPU0PB);
 #endif
 
-	schizo_write(pbm->controller_regs + SCHIZO_SAFARI_ERRCTRL,
-		     (SCHIZO_SAFERRCTRL_EN | err_mask));
+	upa_writeq((SCHIZO_SAFERRCTRL_EN | err_mask),
+		   pbm->controller_regs + SCHIZO_SAFARI_ERRCTRL);
 }
 
 static void pbm_config_busmastering(struct pci_pbm_info *pbm)
@@ -1084,14 +1063,15 @@ static void pbm_config_busmastering(struct pci_pbm_info *pbm)
 	pci_config_write8(addr, 64);
 }
 
-static void __init schizo_scan_bus(struct pci_pbm_info *pbm)
+static void __devinit schizo_scan_bus(struct pci_pbm_info *pbm,
+				      struct device *parent)
 {
 	pbm_config_busmastering(pbm);
 	pbm->is_66mhz_capable =
-		(of_find_property(pbm->prom_node, "66mhz-capable", NULL)
+		(of_find_property(pbm->op->node, "66mhz-capable", NULL)
 		 != NULL);
 
-	pbm->pci_bus = pci_scan_one_pbm(pbm);
+	pbm->pci_bus = pci_scan_one_pbm(pbm, parent);
 
 	if (pbm->chip_type == PBM_CHIP_TYPE_TOMATILLO)
 		tomatillo_register_error_handlers(pbm);
@@ -1133,12 +1113,12 @@ static void schizo_pbm_strbuf_init(struct pci_pbm_info *pbm)
 	 * streaming buffer and leave the rerun-disable
 	 * setting however OBP set it.
 	 */
-	control = schizo_read(pbm->stc.strbuf_control);
+	control = upa_readq(pbm->stc.strbuf_control);
 	control &= ~(SCHIZO_STRBUF_CTRL_LPTR |
 		     SCHIZO_STRBUF_CTRL_LENAB |
 		     SCHIZO_STRBUF_CTRL_DENAB);
 	control |= SCHIZO_STRBUF_CTRL_ENAB;
-	schizo_write(pbm->stc.strbuf_control, control);
+	upa_writeq(control, pbm->stc.strbuf_control);
 
 	pbm->stc.strbuf_enabled = 1;
 }
@@ -1150,24 +1130,17 @@ static void schizo_pbm_strbuf_init(struct pci_pbm_info *pbm)
 
 static int schizo_pbm_iommu_init(struct pci_pbm_info *pbm)
 {
-	struct iommu *iommu = pbm->iommu;
+	static const u32 vdma_default[] = { 0xc0000000, 0x40000000 };
 	unsigned long i, tagbase, database;
-	struct property *prop;
-	u32 vdma[2], dma_mask;
+	struct iommu *iommu = pbm->iommu;
 	int tsbsize, err;
+	const u32 *vdma;
+	u32 dma_mask;
 	u64 control;
 
-	prop = of_find_property(pbm->prom_node, "virtual-dma", NULL);
-	if (prop) {
-		u32 *val = prop->value;
-
-		vdma[0] = val[0];
-		vdma[1] = val[1];
-	} else {
-		/* No property, use default values. */
-		vdma[0] = 0xc0000000;
-		vdma[1] = 0x40000000;
-	}
+	vdma = of_get_property(pbm->op->node, "virtual-dma", NULL);
+	if (!vdma)
+		vdma = vdma_default;
 
 	dma_mask = vdma[0];
 	switch (vdma[1]) {
@@ -1187,9 +1160,9 @@ static int schizo_pbm_iommu_init(struct pci_pbm_info *pbm)
 			break;
 
 		default:
-			prom_printf("SCHIZO: strange virtual-dma size.\n");
-			prom_halt();
-	};
+			printk(KERN_ERR PFX "Strange virtual-dma size.\n");
+			return -EINVAL;
+	}
 
 	/* Register addresses, SCHIZO has iommu ctx flushing. */
 	iommu->iommu_control  = pbm->pbm_regs + SCHIZO_IOMMU_CONTROL;
@@ -1206,15 +1179,15 @@ static int schizo_pbm_iommu_init(struct pci_pbm_info *pbm)
 	/*
 	 * Invalidate TLB Entries.
 	 */
-	control = schizo_read(iommu->iommu_control);
+	control = upa_readq(iommu->iommu_control);
 	control |= SCHIZO_IOMMU_CTRL_DENAB;
-	schizo_write(iommu->iommu_control, control);
+	upa_writeq(control, iommu->iommu_control);
 
 	tagbase = SCHIZO_IOMMU_TAG, database = SCHIZO_IOMMU_DATA;
 
-	for(i = 0; i < 16; i++) {
-		schizo_write(pbm->pbm_regs + tagbase + (i * 8UL), 0);
-		schizo_write(pbm->pbm_regs + database + (i * 8UL), 0);
+	for (i = 0; i < 16; i++) {
+		upa_writeq(0, pbm->pbm_regs + tagbase + (i * 8UL));
+		upa_writeq(0, pbm->pbm_regs + database + (i * 8UL));
 	}
 
 	/* Leave diag mode enabled for full-flushing done
@@ -1222,12 +1195,14 @@ static int schizo_pbm_iommu_init(struct pci_pbm_info *pbm)
 	 */
 	err = iommu_table_init(iommu, tsbsize * 8 * 1024, vdma[0], dma_mask,
 			       pbm->numa_node);
-	if (err)
+	if (err) {
+		printk(KERN_ERR PFX "iommu_table_init() fails with %d\n", err);
 		return err;
+	}
 
-	schizo_write(iommu->iommu_tsbbase, __pa(iommu->page_table));
+	upa_writeq(__pa(iommu->page_table), iommu->iommu_tsbbase);
 
-	control = schizo_read(iommu->iommu_control);
+	control = upa_readq(iommu->iommu_control);
 	control &= ~(SCHIZO_IOMMU_CTRL_TSBSZ | SCHIZO_IOMMU_CTRL_TBWSZ);
 	switch (tsbsize) {
 	case 64:
@@ -1236,10 +1211,10 @@ static int schizo_pbm_iommu_init(struct pci_pbm_info *pbm)
 	case 128:
 		control |= SCHIZO_IOMMU_TSBSZ_128K;
 		break;
-	};
+	}
 
 	control |= SCHIZO_IOMMU_CTRL_ENAB;
-	schizo_write(iommu->iommu_control, control);
+	upa_writeq(control, iommu->iommu_control);
 
 	return 0;
 }
@@ -1280,12 +1255,11 @@ static int schizo_pbm_iommu_init(struct pci_pbm_info *pbm)
 
 static void schizo_pbm_hw_init(struct pci_pbm_info *pbm)
 {
-	struct property *prop;
 	u64 tmp;
 
-	schizo_write(pbm->pbm_regs + SCHIZO_PCI_IRQ_RETRY, 5);
+	upa_writeq(5, pbm->pbm_regs + SCHIZO_PCI_IRQ_RETRY);
 
-	tmp = schizo_read(pbm->pbm_regs + SCHIZO_PCI_CTRL);
+	tmp = upa_readq(pbm->pbm_regs + SCHIZO_PCI_CTRL);
 
 	/* Enable arbiter for all PCI slots.  */
 	tmp |= 0xff;
@@ -1294,8 +1268,7 @@ static void schizo_pbm_hw_init(struct pci_pbm_info *pbm)
 	    pbm->chip_version >= 0x2)
 		tmp |= 0x3UL << SCHIZO_PCICTRL_PTO_SHIFT;
 
-	prop = of_find_property(pbm->prom_node, "no-bus-parking", NULL);
-	if (!prop)
+	if (!of_find_property(pbm->op->node, "no-bus-parking", NULL))
 		tmp |= SCHIZO_PCICTRL_PARK;
 	else
 		tmp &= ~SCHIZO_PCICTRL_PARK;
@@ -1311,13 +1284,13 @@ static void schizo_pbm_hw_init(struct pci_pbm_info *pbm)
 			SCHIZO_PCICTRL_RDO_PREF |
 			SCHIZO_PCICTRL_RDL_PREF);
 
-	schizo_write(pbm->pbm_regs + SCHIZO_PCI_CTRL, tmp);
+	upa_writeq(tmp, pbm->pbm_regs + SCHIZO_PCI_CTRL);
 
-	tmp = schizo_read(pbm->pbm_regs + SCHIZO_PCI_DIAG);
+	tmp = upa_readq(pbm->pbm_regs + SCHIZO_PCI_DIAG);
 	tmp &= ~(SCHIZO_PCIDIAG_D_RTRYARB |
 		 SCHIZO_PCIDIAG_D_RETRY |
 		 SCHIZO_PCIDIAG_D_INTSYNC);
-	schizo_write(pbm->pbm_regs + SCHIZO_PCI_DIAG, tmp);
+	upa_writeq(tmp, pbm->pbm_regs + SCHIZO_PCI_DIAG);
 
 	if (pbm->chip_type == PBM_CHIP_TYPE_TOMATILLO) {
 		/* Clear prefetch lengths to workaround a bug in
@@ -1329,17 +1302,16 @@ static void schizo_pbm_hw_init(struct pci_pbm_info *pbm)
 		       TOMATILLO_IOC_RDONE_CPENAB |
 		       TOMATILLO_IOC_RDLINE_CPENAB);
 
-		schizo_write(pbm->pbm_regs + TOMATILLO_PCI_IOC_CSR,
-			     tmp);
+		upa_writeq(tmp, pbm->pbm_regs + TOMATILLO_PCI_IOC_CSR);
 	}
 }
 
-static int __init schizo_pbm_init(struct pci_controller_info *p,
-				  struct device_node *dp, u32 portid,
-				  int chip_type)
+static int __devinit schizo_pbm_init(struct pci_pbm_info *pbm,
+				     struct of_device *op, u32 portid,
+				     int chip_type)
 {
 	const struct linux_prom64_registers *regs;
-	struct pci_pbm_info *pbm;
+	struct device_node *dp = op->node;
 	const char *chipset_name;
 	int is_pbm_a, err;
 
@@ -1372,25 +1344,19 @@ static int __init schizo_pbm_init(struct pci_controller_info *p,
 	regs = of_get_property(dp, "reg", NULL);
 
 	is_pbm_a = ((regs[0].phys_addr & 0x00700000) == 0x00600000);
-	if (is_pbm_a)
-		pbm = &p->pbm_A;
-	else
-		pbm = &p->pbm_B;
 
 	pbm->next = pci_pbm_root;
 	pci_pbm_root = pbm;
 
 	pbm->numa_node = -1;
 
-	pbm->scan_bus = schizo_scan_bus;
 	pbm->pci_ops = &sun4u_pci_ops;
 	pbm->config_space_reg_bits = 8;
 
 	pbm->index = pci_num_pbms++;
 
 	pbm->portid = portid;
-	pbm->parent = p;
-	pbm->prom_node = dp;
+	pbm->op = op;
 
 	pbm->chip_type = chip_type;
 	pbm->chip_version = of_getintprop_default(dp, "version#", 0);
@@ -1420,6 +1386,8 @@ static int __init schizo_pbm_init(struct pci_controller_info *p,
 
 	schizo_pbm_strbuf_init(pbm);
 
+	schizo_scan_bus(pbm, &op->dev);
+
 	return 0;
 }
 
@@ -1433,62 +1401,104 @@ static inline int portid_compare(u32 x, u32 y, int chip_type)
 	return (x == y);
 }
 
-static void __init __schizo_init(struct device_node *dp, char *model_name,
-				 int chip_type)
+static struct pci_pbm_info * __devinit schizo_find_sibling(u32 portid,
+							   int chip_type)
 {
-	struct pci_controller_info *p;
+	struct pci_pbm_info *pbm;
+
+	for (pbm = pci_pbm_root; pbm; pbm = pbm->next) {
+		if (portid_compare(pbm->portid, portid, chip_type))
+			return pbm;
+	}
+	return NULL;
+}
+
+static int __devinit __schizo_init(struct of_device *op, unsigned long chip_type)
+{
+	struct device_node *dp = op->node;
 	struct pci_pbm_info *pbm;
 	struct iommu *iommu;
 	u32 portid;
+	int err;
 
 	portid = of_getintprop_default(dp, "portid", 0xff);
 
-	for (pbm = pci_pbm_root; pbm; pbm = pbm->next) {
-		if (portid_compare(pbm->portid, portid, chip_type)) {
-			if (schizo_pbm_init(pbm->parent, dp,
-					    portid, chip_type))
-				goto fatal_memory_error;
-			return;
-		}
+	err = -ENOMEM;
+	pbm = kzalloc(sizeof(*pbm), GFP_KERNEL);
+	if (!pbm) {
+		printk(KERN_ERR PFX "Cannot allocate pci_pbm_info.\n");
+		goto out_err;
+	}
+
+	pbm->sibling = schizo_find_sibling(portid, chip_type);
+
+	iommu = kzalloc(sizeof(struct iommu), GFP_KERNEL);
+	if (!iommu) {
+		printk(KERN_ERR PFX "Cannot allocate PBM A iommu.\n");
+		goto out_free_pbm;
 	}
 
-	p = kzalloc(sizeof(struct pci_controller_info), GFP_ATOMIC);
-	if (!p)
-		goto fatal_memory_error;
+	pbm->iommu = iommu;
 
-	iommu = kzalloc(sizeof(struct iommu), GFP_ATOMIC);
-	if (!iommu)
-		goto fatal_memory_error;
+	if (schizo_pbm_init(pbm, op, portid, chip_type))
+		goto out_free_iommu;
 
-	p->pbm_A.iommu = iommu;
+	if (pbm->sibling)
+		pbm->sibling->sibling = pbm;
 
-	iommu = kzalloc(sizeof(struct iommu), GFP_ATOMIC);
-	if (!iommu)
-		goto fatal_memory_error;
+	dev_set_drvdata(&op->dev, pbm);
 
-	p->pbm_B.iommu = iommu;
+	return 0;
 
-	if (schizo_pbm_init(p, dp, portid, chip_type))
-		goto fatal_memory_error;
+out_free_iommu:
+	kfree(pbm->iommu);
 
-	return;
+out_free_pbm:
+	kfree(pbm);
 
-fatal_memory_error:
-	prom_printf("SCHIZO: Fatal memory allocation error.\n");
-	prom_halt();
+out_err:
+	return err;
 }
 
-void __init schizo_init(struct device_node *dp, char *model_name)
+static int __devinit schizo_probe(struct of_device *op,
+				  const struct of_device_id *match)
 {
-	__schizo_init(dp, model_name, PBM_CHIP_TYPE_SCHIZO);
+	return __schizo_init(op, (unsigned long) match->data);
 }
 
-void __init schizo_plus_init(struct device_node *dp, char *model_name)
-{
-	__schizo_init(dp, model_name, PBM_CHIP_TYPE_SCHIZO_PLUS);
-}
+/* The ordering of this table is very important.  Some Tomatillo
+ * nodes announce that they are compatible with both pci108e,a801
+ * and pci108e,8001.  So list the chips in reverse chronological
+ * order.
+ */
+static struct of_device_id __initdata schizo_match[] = {
+	{
+		.name = "pci",
+		.compatible = "pci108e,a801",
+		.data = (void *) PBM_CHIP_TYPE_TOMATILLO,
+	},
+	{
+		.name = "pci",
+		.compatible = "pci108e,8002",
+		.data = (void *) PBM_CHIP_TYPE_SCHIZO_PLUS,
+	},
+	{
+		.name = "pci",
+		.compatible = "pci108e,8001",
+		.data = (void *) PBM_CHIP_TYPE_SCHIZO,
+	},
+	{},
+};
 
-void __init tomatillo_init(struct device_node *dp, char *model_name)
+static struct of_platform_driver schizo_driver = {
+	.name		= DRIVER_NAME,
+	.match_table	= schizo_match,
+	.probe		= schizo_probe,
+};
+
+static int __init schizo_init(void)
 {
-	__schizo_init(dp, model_name, PBM_CHIP_TYPE_TOMATILLO);
+	return of_register_driver(&schizo_driver, &of_bus_type);
 }
+
+subsys_initcall(schizo_init);
diff --git a/arch/sparc64/kernel/pci_sun4v.c b/arch/sparc64/kernel/pci_sun4v.c
index a104c80d319d..e86c73ec167b 100644
--- a/arch/sparc64/kernel/pci_sun4v.c
+++ b/arch/sparc64/kernel/pci_sun4v.c
@@ -13,12 +13,10 @@
 #include <linux/irq.h>
 #include <linux/msi.h>
 #include <linux/log2.h>
+#include <linux/of_device.h>
 
 #include <asm/iommu.h>
 #include <asm/irq.h>
-#include <asm/upa.h>
-#include <asm/pstate.h>
-#include <asm/oplib.h>
 #include <asm/hypervisor.h>
 #include <asm/prom.h>
 
@@ -27,6 +25,9 @@
 
 #include "pci_sun4v.h"
 
+#define DRIVER_NAME	"pci_sun4v"
+#define PFX		DRIVER_NAME ": "
+
 static unsigned long vpci_major = 1;
 static unsigned long vpci_minor = 1;
 
@@ -41,6 +42,7 @@ struct iommu_batch {
 };
 
 static DEFINE_PER_CPU(struct iommu_batch, iommu_batch);
+static int iommu_batch_initialized;
 
 /* Interrupts must be disabled.  */
 static inline void iommu_batch_start(struct device *dev, unsigned long prot, unsigned long entry)
@@ -542,15 +544,16 @@ static const struct dma_ops sun4v_dma_ops = {
 	.sync_sg_for_cpu		= dma_4v_sync_sg_for_cpu,
 };
 
-static void __init pci_sun4v_scan_bus(struct pci_pbm_info *pbm)
+static void __init pci_sun4v_scan_bus(struct pci_pbm_info *pbm,
+				      struct device *parent)
 {
 	struct property *prop;
 	struct device_node *dp;
 
-	dp = pbm->prom_node;
+	dp = pbm->op->node;
 	prop = of_find_property(dp, "66mhz-capable", NULL);
 	pbm->is_66mhz_capable = (prop != NULL);
-	pbm->pci_bus = pci_scan_one_pbm(pbm);
+	pbm->pci_bus = pci_scan_one_pbm(pbm, parent);
 
 	/* XXX register error interrupt handlers XXX */
 }
@@ -583,29 +586,22 @@ static unsigned long __init probe_existing_entries(struct pci_pbm_info *pbm,
 	return cnt;
 }
 
-static void __init pci_sun4v_iommu_init(struct pci_pbm_info *pbm)
+static int __init pci_sun4v_iommu_init(struct pci_pbm_info *pbm)
 {
+	static const u32 vdma_default[] = { 0x80000000, 0x80000000 };
 	struct iommu *iommu = pbm->iommu;
-	struct property *prop;
 	unsigned long num_tsb_entries, sz, tsbsize;
-	u32 vdma[2], dma_mask, dma_offset;
-
-	prop = of_find_property(pbm->prom_node, "virtual-dma", NULL);
-	if (prop) {
-		u32 *val = prop->value;
-
-		vdma[0] = val[0];
-		vdma[1] = val[1];
-	} else {
-		/* No property, use default values. */
-		vdma[0] = 0x80000000;
-		vdma[1] = 0x80000000;
-	}
+	u32 dma_mask, dma_offset;
+	const u32 *vdma;
+
+	vdma = of_get_property(pbm->op->node, "virtual-dma", NULL);
+	if (!vdma)
+		vdma = vdma_default;
 
 	if ((vdma[0] | vdma[1]) & ~IO_PAGE_MASK) {
-		prom_printf("PCI-SUN4V: strange virtual-dma[%08x:%08x].\n",
-			    vdma[0], vdma[1]);
-		prom_halt();
+		printk(KERN_ERR PFX "Strange virtual-dma[%08x:%08x].\n",
+		       vdma[0], vdma[1]);
+		return -EINVAL;
 	};
 
 	dma_mask = (roundup_pow_of_two(vdma[1]) - 1UL);
@@ -625,8 +621,8 @@ static void __init pci_sun4v_iommu_init(struct pci_pbm_info *pbm)
 	sz = (sz + 7UL) & ~7UL;
 	iommu->arena.map = kzalloc(sz, GFP_KERNEL);
 	if (!iommu->arena.map) {
-		prom_printf("PCI_IOMMU: Error, kmalloc(arena.map) failed.\n");
-		prom_halt();
+		printk(KERN_ERR PFX "Error, kmalloc(arena.map) failed.\n");
+		return -ENOMEM;
 	}
 	iommu->arena.limit = num_tsb_entries;
 
@@ -634,6 +630,8 @@ static void __init pci_sun4v_iommu_init(struct pci_pbm_info *pbm)
 	if (sz)
 		printk("%s: Imported %lu TSB entries from OBP\n",
 		       pbm->name, sz);
+
+	return 0;
 }
 
 #ifdef CONFIG_PCI_MSI
@@ -890,29 +888,20 @@ static void pci_sun4v_msi_init(struct pci_pbm_info *pbm)
 }
 #endif /* !(CONFIG_PCI_MSI) */
 
-static void __init pci_sun4v_pbm_init(struct pci_controller_info *p,
-				      struct device_node *dp, u32 devhandle)
+static int __init pci_sun4v_pbm_init(struct pci_pbm_info *pbm,
+				     struct of_device *op, u32 devhandle)
 {
-	struct pci_pbm_info *pbm;
-
-	if (devhandle & 0x40)
-		pbm = &p->pbm_B;
-	else
-		pbm = &p->pbm_A;
-
-	pbm->next = pci_pbm_root;
-	pci_pbm_root = pbm;
+	struct device_node *dp = op->node;
+	int err;
 
 	pbm->numa_node = of_node_to_nid(dp);
 
-	pbm->scan_bus = pci_sun4v_scan_bus;
 	pbm->pci_ops = &sun4v_pci_ops;
 	pbm->config_space_reg_bits = 12;
 
 	pbm->index = pci_num_pbms++;
 
-	pbm->parent = p;
-	pbm->prom_node = dp;
+	pbm->op = op;
 
 	pbm->devhandle = devhandle;
 
@@ -924,82 +913,120 @@ static void __init pci_sun4v_pbm_init(struct pci_controller_info *p,
 	pci_determine_mem_io_space(pbm);
 
 	pci_get_pbm_props(pbm);
-	pci_sun4v_iommu_init(pbm);
+
+	err = pci_sun4v_iommu_init(pbm);
+	if (err)
+		return err;
+
 	pci_sun4v_msi_init(pbm);
+
+	pci_sun4v_scan_bus(pbm, &op->dev);
+
+	pbm->next = pci_pbm_root;
+	pci_pbm_root = pbm;
+
+	return 0;
 }
 
-void __init sun4v_pci_init(struct device_node *dp, char *model_name)
+static int __devinit pci_sun4v_probe(struct of_device *op,
+				     const struct of_device_id *match)
 {
+	const struct linux_prom64_registers *regs;
 	static int hvapi_negotiated = 0;
-	struct pci_controller_info *p;
 	struct pci_pbm_info *pbm;
+	struct device_node *dp;
 	struct iommu *iommu;
-	struct property *prop;
-	struct linux_prom64_registers *regs;
 	u32 devhandle;
-	int i;
+	int i, err;
+
+	dp = op->node;
 
 	if (!hvapi_negotiated++) {
-		int err = sun4v_hvapi_register(HV_GRP_PCI,
-					       vpci_major,
-					       &vpci_minor);
+		err = sun4v_hvapi_register(HV_GRP_PCI,
+					   vpci_major,
+					   &vpci_minor);
 
 		if (err) {
-			prom_printf("SUN4V_PCI: Could not register hvapi, "
-				    "err=%d\n", err);
-			prom_halt();
+			printk(KERN_ERR PFX "Could not register hvapi, "
+			       "err=%d\n", err);
+			return err;
 		}
-		printk("SUN4V_PCI: Registered hvapi major[%lu] minor[%lu]\n",
+		printk(KERN_INFO PFX "Registered hvapi major[%lu] minor[%lu]\n",
 		       vpci_major, vpci_minor);
 
 		dma_ops = &sun4v_dma_ops;
 	}
 
-	prop = of_find_property(dp, "reg", NULL);
-	if (!prop) {
-		prom_printf("SUN4V_PCI: Could not find config registers\n");
-		prom_halt();
+	regs = of_get_property(dp, "reg", NULL);
+	err = -ENODEV;
+	if (!regs) {
+		printk(KERN_ERR PFX "Could not find config registers\n");
+		goto out_err;
 	}
-	regs = prop->value;
-
 	devhandle = (regs->phys_addr >> 32UL) & 0x0fffffff;
 
-	for (pbm = pci_pbm_root; pbm; pbm = pbm->next) {
-		if (pbm->devhandle == (devhandle ^ 0x40)) {
-			pci_sun4v_pbm_init(pbm->parent, dp, devhandle);
-			return;
+	err = -ENOMEM;
+	if (!iommu_batch_initialized) {
+		for_each_possible_cpu(i) {
+			unsigned long page = get_zeroed_page(GFP_KERNEL);
+
+			if (!page)
+				goto out_err;
+
+			per_cpu(iommu_batch, i).pglist = (u64 *) page;
 		}
+		iommu_batch_initialized = 1;
 	}
 
-	for_each_possible_cpu(i) {
-		unsigned long page = get_zeroed_page(GFP_ATOMIC);
-
-		if (!page)
-			goto fatal_memory_error;
+	pbm = kzalloc(sizeof(*pbm), GFP_KERNEL);
+	if (!pbm) {
+		printk(KERN_ERR PFX "Could not allocate pci_pbm_info\n");
+		goto out_err;
+	}
 
-		per_cpu(iommu_batch, i).pglist = (u64 *) page;
+	iommu = kzalloc(sizeof(struct iommu), GFP_KERNEL);
+	if (!iommu) {
+		printk(KERN_ERR PFX "Could not allocate pbm iommu\n");
+		goto out_free_controller;
 	}
 
-	p = kzalloc(sizeof(struct pci_controller_info), GFP_ATOMIC);
-	if (!p)
-		goto fatal_memory_error;
+	pbm->iommu = iommu;
 
-	iommu = kzalloc(sizeof(struct iommu), GFP_ATOMIC);
-	if (!iommu)
-		goto fatal_memory_error;
+	err = pci_sun4v_pbm_init(pbm, op, devhandle);
+	if (err)
+		goto out_free_iommu;
+
+	dev_set_drvdata(&op->dev, pbm);
+
+	return 0;
+
+out_free_iommu:
+	kfree(pbm->iommu);
 
-	p->pbm_A.iommu = iommu;
+out_free_controller:
+	kfree(pbm);
 
-	iommu = kzalloc(sizeof(struct iommu), GFP_ATOMIC);
-	if (!iommu)
-		goto fatal_memory_error;
+out_err:
+	return err;
+}
 
-	p->pbm_B.iommu = iommu;
+static struct of_device_id __initdata pci_sun4v_match[] = {
+	{
+		.name = "pci",
+		.compatible = "SUNW,sun4v-pci",
+	},
+	{},
+};
 
-	pci_sun4v_pbm_init(p, dp, devhandle);
-	return;
+static struct of_platform_driver pci_sun4v_driver = {
+	.name		= DRIVER_NAME,
+	.match_table	= pci_sun4v_match,
+	.probe		= pci_sun4v_probe,
+};
 
-fatal_memory_error:
-	prom_printf("SUN4V_PCI: Fatal memory allocation error.\n");
-	prom_halt();
+static int __init pci_sun4v_init(void)
+{
+	return of_register_driver(&pci_sun4v_driver, &of_bus_type);
 }
+
+subsys_initcall(pci_sun4v_init);
diff --git a/arch/sparc64/kernel/pci_sun4v_asm.S b/arch/sparc64/kernel/pci_sun4v_asm.S
index ecb81f389b06..e606d46c6815 100644
--- a/arch/sparc64/kernel/pci_sun4v_asm.S
+++ b/arch/sparc64/kernel/pci_sun4v_asm.S
@@ -1,8 +1,9 @@
 /* pci_sun4v_asm: Hypervisor calls for PCI support.
  *
- * Copyright (C) 2006 David S. Miller <davem@davemloft.net>
+ * Copyright (C) 2006, 2008 David S. Miller <davem@davemloft.net>
  */
 
+#include <linux/linkage.h>
 #include <asm/hypervisor.h>
 
 	/* %o0: devhandle
@@ -14,8 +15,7 @@
 	 * returns %o0:	-status if status was non-zero, else
 	 *         %o0:	num pages mapped
 	 */
-	.globl	pci_sun4v_iommu_map
-pci_sun4v_iommu_map:
+ENTRY(pci_sun4v_iommu_map)
 	mov	%o5, %g1
 	mov	HV_FAST_PCI_IOMMU_MAP, %o5
 	ta	HV_FAST_TRAP
@@ -24,6 +24,7 @@ pci_sun4v_iommu_map:
 	mov	%o1, %o0
 1:	retl
 	 nop
+ENDPROC(pci_sun4v_iommu_map)
 
 	/* %o0: devhandle
 	 * %o1:	tsbid
@@ -31,12 +32,12 @@ pci_sun4v_iommu_map:
 	 *
 	 * returns %o0:	num ttes demapped
 	 */
-	.globl	pci_sun4v_iommu_demap
-pci_sun4v_iommu_demap:
+ENTRY(pci_sun4v_iommu_demap)
 	mov	HV_FAST_PCI_IOMMU_DEMAP, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 mov	%o1, %o0
+ENDPROC(pci_sun4v_iommu_demap)
 
 	/* %o0: devhandle
 	 * %o1:	tsbid
@@ -45,8 +46,7 @@ pci_sun4v_iommu_demap:
 	 *
 	 * returns %o0:	status
 	 */
-	.globl	pci_sun4v_iommu_getmap
-pci_sun4v_iommu_getmap:
+ENTRY(pci_sun4v_iommu_getmap)
 	mov	%o2, %o4
 	mov	HV_FAST_PCI_IOMMU_GETMAP, %o5
 	ta	HV_FAST_TRAP
@@ -54,6 +54,7 @@ pci_sun4v_iommu_getmap:
 	stx	%o2, [%o3]
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_iommu_getmap)
 
 	/* %o0: devhandle
 	 * %o1:	pci_device
@@ -65,14 +66,14 @@ pci_sun4v_iommu_getmap:
 	 * If there is an error, the data will be returned
 	 * as all 1's.
 	 */
-	.globl	pci_sun4v_config_get
-pci_sun4v_config_get:
+ENTRY(pci_sun4v_config_get)
 	mov	HV_FAST_PCI_CONFIG_GET, %o5
 	ta	HV_FAST_TRAP
 	brnz,a,pn %o1, 1f
 	 mov	-1, %o2
 1:	retl
 	 mov	%o2, %o0
+ENDPROC(pci_sun4v_config_get)
 
 	/* %o0: devhandle
 	 * %o1:	pci_device
@@ -85,14 +86,14 @@ pci_sun4v_config_get:
 	 * status will be zero if the operation completed
 	 * successfully, else -1 if not
 	 */
-	.globl	pci_sun4v_config_put
-pci_sun4v_config_put:
+ENTRY(pci_sun4v_config_put)
 	mov	HV_FAST_PCI_CONFIG_PUT, %o5
 	ta	HV_FAST_TRAP
 	brnz,a,pn %o1, 1f
 	 mov	-1, %o1
 1:	retl
 	 mov	%o1, %o0
+ENDPROC(pci_sun4v_config_put)
 
 	/* %o0: devhandle
 	 * %o1: msiqid
@@ -104,12 +105,12 @@ pci_sun4v_config_put:
 	 * status will be zero if the operation completed
 	 * successfully, else -1 if not
 	 */
-	.globl	pci_sun4v_msiq_conf
-pci_sun4v_msiq_conf:
+ENTRY(pci_sun4v_msiq_conf)
 	mov	HV_FAST_PCI_MSIQ_CONF, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msiq_conf)
 
 	/* %o0: devhandle
 	 * %o1: msiqid
@@ -118,8 +119,7 @@ pci_sun4v_msiq_conf:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msiq_info
-pci_sun4v_msiq_info:
+ENTRY(pci_sun4v_msiq_info)
 	mov	%o2, %o4
 	mov	HV_FAST_PCI_MSIQ_INFO, %o5
 	ta	HV_FAST_TRAP
@@ -127,6 +127,7 @@ pci_sun4v_msiq_info:
 	stx	%o2, [%o3]
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msiq_info)
 
 	/* %o0: devhandle
 	 * %o1: msiqid
@@ -134,13 +135,13 @@ pci_sun4v_msiq_info:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msiq_getvalid
-pci_sun4v_msiq_getvalid:
+ENTRY(pci_sun4v_msiq_getvalid)
 	mov	HV_FAST_PCI_MSIQ_GETVALID, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o2]
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msiq_getvalid)
 
 	/* %o0: devhandle
 	 * %o1: msiqid
@@ -148,12 +149,12 @@ pci_sun4v_msiq_getvalid:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msiq_setvalid
-pci_sun4v_msiq_setvalid:
+ENTRY(pci_sun4v_msiq_setvalid)
 	mov	HV_FAST_PCI_MSIQ_SETVALID, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msiq_setvalid)
 
 	/* %o0: devhandle
 	 * %o1: msiqid
@@ -161,13 +162,13 @@ pci_sun4v_msiq_setvalid:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msiq_getstate
-pci_sun4v_msiq_getstate:
+ENTRY(pci_sun4v_msiq_getstate)
 	mov	HV_FAST_PCI_MSIQ_GETSTATE, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o2]
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msiq_getstate)
 
 	/* %o0: devhandle
 	 * %o1: msiqid
@@ -175,12 +176,12 @@ pci_sun4v_msiq_getstate:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msiq_setstate
-pci_sun4v_msiq_setstate:
+ENTRY(pci_sun4v_msiq_setstate)
 	mov	HV_FAST_PCI_MSIQ_SETSTATE, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msiq_setstate)
 
 	/* %o0: devhandle
 	 * %o1: msiqid
@@ -188,13 +189,13 @@ pci_sun4v_msiq_setstate:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msiq_gethead
-pci_sun4v_msiq_gethead:
+ENTRY(pci_sun4v_msiq_gethead)
 	mov	HV_FAST_PCI_MSIQ_GETHEAD, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o2]
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msiq_gethead)
 
 	/* %o0: devhandle
 	 * %o1: msiqid
@@ -202,12 +203,12 @@ pci_sun4v_msiq_gethead:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msiq_sethead
-pci_sun4v_msiq_sethead:
+ENTRY(pci_sun4v_msiq_sethead)
 	mov	HV_FAST_PCI_MSIQ_SETHEAD, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msiq_sethead)
 
 	/* %o0: devhandle
 	 * %o1: msiqid
@@ -215,13 +216,13 @@ pci_sun4v_msiq_sethead:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msiq_gettail
-pci_sun4v_msiq_gettail:
+ENTRY(pci_sun4v_msiq_gettail)
 	mov	HV_FAST_PCI_MSIQ_GETTAIL, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o2]
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msiq_gettail)
 
 	/* %o0: devhandle
 	 * %o1: msinum
@@ -229,13 +230,13 @@ pci_sun4v_msiq_gettail:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msi_getvalid
-pci_sun4v_msi_getvalid:
+ENTRY(pci_sun4v_msi_getvalid)
 	mov	HV_FAST_PCI_MSI_GETVALID, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o2]
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msi_getvalid)
 
 	/* %o0: devhandle
 	 * %o1: msinum
@@ -243,12 +244,12 @@ pci_sun4v_msi_getvalid:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msi_setvalid
-pci_sun4v_msi_setvalid:
+ENTRY(pci_sun4v_msi_setvalid)
 	mov	HV_FAST_PCI_MSI_SETVALID, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msi_setvalid)
 
 	/* %o0: devhandle
 	 * %o1: msinum
@@ -256,13 +257,13 @@ pci_sun4v_msi_setvalid:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msi_getmsiq
-pci_sun4v_msi_getmsiq:
+ENTRY(pci_sun4v_msi_getmsiq)
 	mov	HV_FAST_PCI_MSI_GETMSIQ, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o2]
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msi_getmsiq)
 
 	/* %o0: devhandle
 	 * %o1: msinum
@@ -271,12 +272,12 @@ pci_sun4v_msi_getmsiq:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msi_setmsiq
-pci_sun4v_msi_setmsiq:
+ENTRY(pci_sun4v_msi_setmsiq)
 	mov	HV_FAST_PCI_MSI_SETMSIQ, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msi_setmsiq)
 
 	/* %o0: devhandle
 	 * %o1: msinum
@@ -284,13 +285,13 @@ pci_sun4v_msi_setmsiq:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msi_getstate
-pci_sun4v_msi_getstate:
+ENTRY(pci_sun4v_msi_getstate)
 	mov	HV_FAST_PCI_MSI_GETSTATE, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o2]
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msi_getstate)
 
 	/* %o0: devhandle
 	 * %o1: msinum
@@ -298,12 +299,12 @@ pci_sun4v_msi_getstate:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msi_setstate
-pci_sun4v_msi_setstate:
+ENTRY(pci_sun4v_msi_setstate)
 	mov	HV_FAST_PCI_MSI_SETSTATE, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msi_setstate)
 
 	/* %o0: devhandle
 	 * %o1: msinum
@@ -311,13 +312,13 @@ pci_sun4v_msi_setstate:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msg_getmsiq
-pci_sun4v_msg_getmsiq:
+ENTRY(pci_sun4v_msg_getmsiq)
 	mov	HV_FAST_PCI_MSG_GETMSIQ, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o2]
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msg_getmsiq)
 
 	/* %o0: devhandle
 	 * %o1: msinum
@@ -325,12 +326,12 @@ pci_sun4v_msg_getmsiq:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msg_setmsiq
-pci_sun4v_msg_setmsiq:
+ENTRY(pci_sun4v_msg_setmsiq)
 	mov	HV_FAST_PCI_MSG_SETMSIQ, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msg_setmsiq)
 
 	/* %o0: devhandle
 	 * %o1: msinum
@@ -338,13 +339,13 @@ pci_sun4v_msg_setmsiq:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msg_getvalid
-pci_sun4v_msg_getvalid:
+ENTRY(pci_sun4v_msg_getvalid)
 	mov	HV_FAST_PCI_MSG_GETVALID, %o5
 	ta	HV_FAST_TRAP
 	stx	%o1, [%o2]
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msg_getvalid)
 
 	/* %o0: devhandle
 	 * %o1: msinum
@@ -352,10 +353,10 @@ pci_sun4v_msg_getvalid:
 	 *
 	 * returns %o0: status
 	 */
-	.globl	pci_sun4v_msg_setvalid
-pci_sun4v_msg_setvalid:
+ENTRY(pci_sun4v_msg_setvalid)
 	mov	HV_FAST_PCI_MSG_SETVALID, %o5
 	ta	HV_FAST_TRAP
 	retl
 	 mov	%o0, %o0
+ENDPROC(pci_sun4v_msg_setvalid)
 
diff --git a/arch/sparc64/kernel/power.c b/arch/sparc64/kernel/power.c
index 3bb987a6d03c..076cad7f9757 100644
--- a/arch/sparc64/kernel/power.c
+++ b/arch/sparc64/kernel/power.c
@@ -1,34 +1,17 @@
 /* power.c: Power management driver.
  *
- * Copyright (C) 1999, 2007 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 1999, 2007, 2008 David S. Miller (davem@davemloft.net)
  */
 
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
-#include <linux/sched.h>
-#include <linux/signal.h>
-#include <linux/delay.h>
 #include <linux/interrupt.h>
-#include <linux/pm.h>
-#include <linux/syscalls.h>
 #include <linux/reboot.h>
 #include <linux/of_device.h>
 
-#include <asm/system.h>
-#include <asm/auxio.h>
 #include <asm/prom.h>
 #include <asm/io.h>
-#include <asm/sstate.h>
-#include <asm/reboot.h>
-
-#include <linux/unistd.h>
-
-/*
- * sysctl - toggle power-off restriction for serial console 
- * systems in machine_power_off()
- */
-int scons_pwroff = 1; 
 
 static void __iomem *power_reg;
 
@@ -40,31 +23,6 @@ static irqreturn_t power_handler(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-static void (*poweroff_method)(void) = machine_alt_power_off;
-
-void machine_power_off(void)
-{
-	sstate_poweroff();
-	if (strcmp(of_console_device->type, "serial") || scons_pwroff) {
-		if (power_reg) {
-			/* Both register bits seem to have the
-			 * same effect, so until I figure out
-			 * what the difference is...
-			 */
-			writel(AUXIO_PCIO_CPWR_OFF | AUXIO_PCIO_SPWR_OFF, power_reg);
-		} else {
-			if (poweroff_method != NULL) {
-				poweroff_method();
-				/* not reached */
-			}
-		}
-	}
-	machine_halt();
-}
-
-void (*pm_power_off)(void) = machine_power_off;
-EXPORT_SYMBOL(pm_power_off);
-
 static int __init has_button_interrupt(unsigned int irq, struct device_node *dp)
 {
 	if (irq == 0xffffffff)
@@ -85,8 +43,6 @@ static int __devinit power_probe(struct of_device *op, const struct of_device_id
 	printk(KERN_INFO "%s: Control reg at %lx\n",
 	       op->node->name, res->start);
 
-	poweroff_method = machine_halt;  /* able to use the standard halt */
-
 	if (has_button_interrupt(irq, op->node)) {
 		if (request_irq(irq,
 				power_handler, 0, "power", NULL) < 0)
@@ -96,7 +52,7 @@ static int __devinit power_probe(struct of_device *op, const struct of_device_id
 	return 0;
 }
 
-static struct of_device_id power_match[] = {
+static struct of_device_id __initdata power_match[] = {
 	{
 		.name = "power",
 	},
@@ -111,8 +67,9 @@ static struct of_platform_driver power_driver = {
 	},
 };
 
-void __init power_init(void)
+static int __init power_init(void)
 {
-	of_register_driver(&power_driver, &of_platform_bus_type);
-	return;
+	return of_register_driver(&power_driver, &of_platform_bus_type);
 }
+
+device_initcall(power_init);
diff --git a/arch/sparc64/kernel/process.c b/arch/sparc64/kernel/process.c
index 15f4178592e7..d5e2acef9877 100644
--- a/arch/sparc64/kernel/process.c
+++ b/arch/sparc64/kernel/process.c
@@ -22,7 +22,6 @@
 #include <linux/ptrace.h>
 #include <linux/slab.h>
 #include <linux/user.h>
-#include <linux/reboot.h>
 #include <linux/delay.h>
 #include <linux/compat.h>
 #include <linux/tick.h>
@@ -31,7 +30,6 @@
 #include <linux/elfcore.h>
 #include <linux/sysrq.h>
 
-#include <asm/oplib.h>
 #include <asm/uaccess.h>
 #include <asm/system.h>
 #include <asm/page.h>
@@ -46,8 +44,6 @@
 #include <asm/mmu_context.h>
 #include <asm/unistd.h>
 #include <asm/hypervisor.h>
-#include <asm/sstate.h>
-#include <asm/reboot.h>
 #include <asm/syscalls.h>
 #include <asm/irq_regs.h>
 #include <asm/smp.h>
@@ -115,35 +111,6 @@ void cpu_idle(void)
 	}
 }
 
-void machine_halt(void)
-{
-	sstate_halt();
-	prom_halt();
-	panic("Halt failed!");
-}
-
-void machine_alt_power_off(void)
-{
-	sstate_poweroff();
-	prom_halt_power_off();
-	panic("Power-off failed!");
-}
-
-void machine_restart(char * cmd)
-{
-	char *p;
-	
-	sstate_reboot();
-	p = strchr (reboot_command, '\n');
-	if (p) *p = 0;
-	if (cmd)
-		prom_reboot(cmd);
-	if (*reboot_command)
-		prom_reboot(reboot_command);
-	prom_reboot("");
-	panic("Reboot failed!");
-}
-
 #ifdef CONFIG_COMPAT
 static void show_regwindow32(struct pt_regs *regs)
 {
@@ -248,7 +215,6 @@ static void __global_reg_self(struct thread_info *tp, struct pt_regs *regs,
 	global_reg_snapshot[this_cpu].o7 = regs->u_regs[UREG_I7];
 
 	if (regs->tstate & TSTATE_PRIV) {
-		struct thread_info *tp = current_thread_info();
 		struct reg_window *rw;
 
 		rw = (struct reg_window *)
@@ -304,7 +270,6 @@ void __trigger_all_cpu_backtrace(void)
 
 	for_each_online_cpu(cpu) {
 		struct global_reg_snapshot *gp = &global_reg_snapshot[cpu];
-		struct thread_info *tp;
 
 		__global_reg_poll(gp);
 
diff --git a/arch/sparc64/kernel/prom.c b/arch/sparc64/kernel/prom.c
index 7151513f156e..dbba82f9b142 100644
--- a/arch/sparc64/kernel/prom.c
+++ b/arch/sparc64/kernel/prom.c
@@ -38,7 +38,7 @@ struct device_node *of_find_node_by_phandle(phandle handle)
 {
 	struct device_node *np;
 
-	for (np = allnodes; np != 0; np = np->allnext)
+	for (np = allnodes; np; np = np->allnext)
 		if (np->node == handle)
 			break;
 
@@ -59,6 +59,9 @@ int of_getintprop_default(struct device_node *np, const char *name, int def)
 }
 EXPORT_SYMBOL(of_getintprop_default);
 
+DEFINE_MUTEX(of_set_property_mutex);
+EXPORT_SYMBOL(of_set_property_mutex);
+
 int of_set_property(struct device_node *dp, const char *name, void *val, int len)
 {
 	struct property **prevp;
@@ -82,7 +85,10 @@ int of_set_property(struct device_node *dp, const char *name, void *val, int len
 			void *old_val = prop->value;
 			int ret;
 
+			mutex_lock(&of_set_property_mutex);
 			ret = prom_setprop(dp->node, name, val, len);
+			mutex_unlock(&of_set_property_mutex);
+
 			err = -EINVAL;
 			if (ret >= 0) {
 				prop->value = new_val;
@@ -945,22 +951,30 @@ static void __init irq_trans_init(struct device_node *dp)
 		for (i = 0; i < ARRAY_SIZE(pci_irq_trans_table); i++) {
 			struct irq_trans *t = &pci_irq_trans_table[i];
 
-			if (!strcmp(model, t->name))
-				return t->init(dp);
+			if (!strcmp(model, t->name)) {
+				t->init(dp);
+				return;
+			}
 		}
 	}
 #endif
 #ifdef CONFIG_SBUS
 	if (!strcmp(dp->name, "sbus") ||
-	    !strcmp(dp->name, "sbi"))
-		return sbus_irq_trans_init(dp);
+	    !strcmp(dp->name, "sbi")) {
+		sbus_irq_trans_init(dp);
+		return;
+	}
 #endif
 	if (!strcmp(dp->name, "fhc") &&
-	    !strcmp(dp->parent->name, "central"))
-		return central_irq_trans_init(dp);
+	    !strcmp(dp->parent->name, "central")) {
+		central_irq_trans_init(dp);
+		return;
+	}
 	if (!strcmp(dp->name, "virtual-devices") ||
-	    !strcmp(dp->name, "niu"))
-		return sun4v_vdev_irq_trans_init(dp);
+	    !strcmp(dp->name, "niu")) {
+		sun4v_vdev_irq_trans_init(dp);
+		return;
+	}
 }
 
 static int is_root_node(const struct device_node *dp)
@@ -1231,32 +1245,49 @@ static void __init __build_path_component(struct device_node *dp, char *tmp_buf)
 
 	if (parent != NULL) {
 		if (!strcmp(parent->type, "pci") ||
-		    !strcmp(parent->type, "pciex"))
-			return pci_path_component(dp, tmp_buf);
-		if (!strcmp(parent->type, "sbus"))
-			return sbus_path_component(dp, tmp_buf);
-		if (!strcmp(parent->type, "upa"))
-			return upa_path_component(dp, tmp_buf);
-		if (!strcmp(parent->type, "ebus"))
-			return ebus_path_component(dp, tmp_buf);
+		    !strcmp(parent->type, "pciex")) {
+			pci_path_component(dp, tmp_buf);
+			return;
+		}
+		if (!strcmp(parent->type, "sbus")) {
+			sbus_path_component(dp, tmp_buf);
+			return;
+		}
+		if (!strcmp(parent->type, "upa")) {
+			upa_path_component(dp, tmp_buf);
+			return;
+		}
+		if (!strcmp(parent->type, "ebus")) {
+			ebus_path_component(dp, tmp_buf);
+			return;
+		}
 		if (!strcmp(parent->name, "usb") ||
-		    !strcmp(parent->name, "hub"))
-			return usb_path_component(dp, tmp_buf);
-		if (!strcmp(parent->type, "i2c"))
-			return i2c_path_component(dp, tmp_buf);
-		if (!strcmp(parent->type, "firewire"))
-			return ieee1394_path_component(dp, tmp_buf);
-		if (!strcmp(parent->type, "virtual-devices"))
-			return vdev_path_component(dp, tmp_buf);
-
+		    !strcmp(parent->name, "hub")) {
+			usb_path_component(dp, tmp_buf);
+			return;
+		}
+		if (!strcmp(parent->type, "i2c")) {
+			i2c_path_component(dp, tmp_buf);
+			return;
+		}
+		if (!strcmp(parent->type, "firewire")) {
+			ieee1394_path_component(dp, tmp_buf);
+			return;
+		}
+		if (!strcmp(parent->type, "virtual-devices")) {
+			vdev_path_component(dp, tmp_buf);
+			return;
+		}
 		/* "isa" is handled with platform naming */
 	}
 
 	/* Use platform naming convention.  */
-	if (tlb_type == hypervisor)
-		return sun4v_path_component(dp, tmp_buf);
-	else
-		return sun4u_path_component(dp, tmp_buf);
+	if (tlb_type == hypervisor) {
+		sun4v_path_component(dp, tmp_buf);
+		return;
+	} else {
+		sun4u_path_component(dp, tmp_buf);
+	}
 }
 
 static char * __init build_path_component(struct device_node *dp)
diff --git a/arch/sparc64/kernel/psycho_common.c b/arch/sparc64/kernel/psycho_common.c
new file mode 100644
index 000000000000..790996428c14
--- /dev/null
+++ b/arch/sparc64/kernel/psycho_common.c
@@ -0,0 +1,470 @@
+/* psycho_common.c: Code common to PSYCHO and derivative PCI controllers.
+ *
+ * Copyright (C) 2008 David S. Miller <davem@davemloft.net>
+ */
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+
+#include <asm/upa.h>
+
+#include "pci_impl.h"
+#include "iommu_common.h"
+#include "psycho_common.h"
+
+#define  PSYCHO_STRBUF_CTRL_DENAB	0x0000000000000002UL
+#define  PSYCHO_STCERR_WRITE		0x0000000000000002UL
+#define  PSYCHO_STCERR_READ		0x0000000000000001UL
+#define  PSYCHO_STCTAG_PPN		0x0fffffff00000000UL
+#define  PSYCHO_STCTAG_VPN		0x00000000ffffe000UL
+#define  PSYCHO_STCTAG_VALID		0x0000000000000002UL
+#define  PSYCHO_STCTAG_WRITE		0x0000000000000001UL
+#define  PSYCHO_STCLINE_LINDX		0x0000000001e00000UL
+#define  PSYCHO_STCLINE_SPTR		0x00000000001f8000UL
+#define  PSYCHO_STCLINE_LADDR		0x0000000000007f00UL
+#define  PSYCHO_STCLINE_EPTR		0x00000000000000fcUL
+#define  PSYCHO_STCLINE_VALID		0x0000000000000002UL
+#define  PSYCHO_STCLINE_FOFN		0x0000000000000001UL
+
+static DEFINE_SPINLOCK(stc_buf_lock);
+static unsigned long stc_error_buf[128];
+static unsigned long stc_tag_buf[16];
+static unsigned long stc_line_buf[16];
+
+static void psycho_check_stc_error(struct pci_pbm_info *pbm)
+{
+	unsigned long err_base, tag_base, line_base;
+	struct strbuf *strbuf = &pbm->stc;
+	u64 control;
+	int i;
+
+	if (!strbuf->strbuf_control)
+		return;
+
+	err_base = strbuf->strbuf_err_stat;
+	tag_base = strbuf->strbuf_tag_diag;
+	line_base = strbuf->strbuf_line_diag;
+
+	spin_lock(&stc_buf_lock);
+
+	/* This is __REALLY__ dangerous.  When we put the streaming
+	 * buffer into diagnostic mode to probe it's tags and error
+	 * status, we _must_ clear all of the line tag valid bits
+	 * before re-enabling the streaming buffer.  If any dirty data
+	 * lives in the STC when we do this, we will end up
+	 * invalidating it before it has a chance to reach main
+	 * memory.
+	 */
+	control = upa_readq(strbuf->strbuf_control);
+	upa_writeq(control | PSYCHO_STRBUF_CTRL_DENAB, strbuf->strbuf_control);
+	for (i = 0; i < 128; i++) {
+		u64 val;
+
+		val = upa_readq(err_base + (i * 8UL));
+		upa_writeq(0UL, err_base + (i * 8UL));
+		stc_error_buf[i] = val;
+	}
+	for (i = 0; i < 16; i++) {
+		stc_tag_buf[i] = upa_readq(tag_base + (i * 8UL));
+		stc_line_buf[i] = upa_readq(line_base + (i * 8UL));
+		upa_writeq(0UL, tag_base + (i * 8UL));
+		upa_writeq(0UL, line_base + (i * 8UL));
+	}
+
+	/* OK, state is logged, exit diagnostic mode. */
+	upa_writeq(control, strbuf->strbuf_control);
+
+	for (i = 0; i < 16; i++) {
+		int j, saw_error, first, last;
+
+		saw_error = 0;
+		first = i * 8;
+		last = first + 8;
+		for (j = first; j < last; j++) {
+			u64 errval = stc_error_buf[j];
+			if (errval != 0) {
+				saw_error++;
+				printk(KERN_ERR "%s: STC_ERR(%d)[wr(%d)"
+				       "rd(%d)]\n",
+				       pbm->name,
+				       j,
+				       (errval & PSYCHO_STCERR_WRITE) ? 1 : 0,
+				       (errval & PSYCHO_STCERR_READ) ? 1 : 0);
+			}
+		}
+		if (saw_error != 0) {
+			u64 tagval = stc_tag_buf[i];
+			u64 lineval = stc_line_buf[i];
+			printk(KERN_ERR "%s: STC_TAG(%d)[PA(%016lx)VA(%08lx)"
+			       "V(%d)W(%d)]\n",
+			       pbm->name,
+			       i,
+			       ((tagval & PSYCHO_STCTAG_PPN) >> 19UL),
+			       (tagval & PSYCHO_STCTAG_VPN),
+			       ((tagval & PSYCHO_STCTAG_VALID) ? 1 : 0),
+			       ((tagval & PSYCHO_STCTAG_WRITE) ? 1 : 0));
+			printk(KERN_ERR "%s: STC_LINE(%d)[LIDX(%lx)SP(%lx)"
+			       "LADDR(%lx)EP(%lx)V(%d)FOFN(%d)]\n",
+			       pbm->name,
+			       i,
+			       ((lineval & PSYCHO_STCLINE_LINDX) >> 21UL),
+			       ((lineval & PSYCHO_STCLINE_SPTR) >> 15UL),
+			       ((lineval & PSYCHO_STCLINE_LADDR) >> 8UL),
+			       ((lineval & PSYCHO_STCLINE_EPTR) >> 2UL),
+			       ((lineval & PSYCHO_STCLINE_VALID) ? 1 : 0),
+			       ((lineval & PSYCHO_STCLINE_FOFN) ? 1 : 0));
+		}
+	}
+
+	spin_unlock(&stc_buf_lock);
+}
+
+#define PSYCHO_IOMMU_TAG		0xa580UL
+#define PSYCHO_IOMMU_DATA		0xa600UL
+
+static void psycho_record_iommu_tags_and_data(struct pci_pbm_info *pbm,
+					      u64 *tag, u64 *data)
+{
+	int i;
+
+	for (i = 0; i < 16; i++) {
+		unsigned long base = pbm->controller_regs;
+		unsigned long off = i * 8UL;
+
+		tag[i] = upa_readq(base + PSYCHO_IOMMU_TAG+off);
+		data[i] = upa_readq(base + PSYCHO_IOMMU_DATA+off);
+
+		/* Now clear out the entry. */
+		upa_writeq(0, base + PSYCHO_IOMMU_TAG + off);
+		upa_writeq(0, base + PSYCHO_IOMMU_DATA + off);
+	}
+}
+
+#define  PSYCHO_IOMMU_TAG_ERRSTS (0x3UL << 23UL)
+#define  PSYCHO_IOMMU_TAG_ERR	 (0x1UL << 22UL)
+#define  PSYCHO_IOMMU_TAG_WRITE	 (0x1UL << 21UL)
+#define  PSYCHO_IOMMU_TAG_STREAM (0x1UL << 20UL)
+#define  PSYCHO_IOMMU_TAG_SIZE	 (0x1UL << 19UL)
+#define  PSYCHO_IOMMU_TAG_VPAGE	 0x7ffffUL
+#define  PSYCHO_IOMMU_DATA_VALID (1UL << 30UL)
+#define  PSYCHO_IOMMU_DATA_CACHE (1UL << 28UL)
+#define  PSYCHO_IOMMU_DATA_PPAGE 0xfffffffUL
+
+static void psycho_dump_iommu_tags_and_data(struct pci_pbm_info *pbm,
+					    u64 *tag, u64 *data)
+{
+	int i;
+
+	for (i = 0; i < 16; i++) {
+		u64 tag_val, data_val;
+		const char *type_str;
+		tag_val = tag[i];
+		if (!(tag_val & PSYCHO_IOMMU_TAG_ERR))
+			continue;
+
+		data_val = data[i];
+		switch((tag_val & PSYCHO_IOMMU_TAG_ERRSTS) >> 23UL) {
+		case 0:
+			type_str = "Protection Error";
+			break;
+		case 1:
+			type_str = "Invalid Error";
+			break;
+		case 2:
+			type_str = "TimeOut Error";
+			break;
+		case 3:
+		default:
+			type_str = "ECC Error";
+			break;
+		}
+
+		printk(KERN_ERR "%s: IOMMU TAG(%d)[error(%s) wr(%d) "
+		       "str(%d) sz(%dK) vpg(%08lx)]\n",
+		       pbm->name, i, type_str,
+		       ((tag_val & PSYCHO_IOMMU_TAG_WRITE) ? 1 : 0),
+		       ((tag_val & PSYCHO_IOMMU_TAG_STREAM) ? 1 : 0),
+		       ((tag_val & PSYCHO_IOMMU_TAG_SIZE) ? 64 : 8),
+		       (tag_val & PSYCHO_IOMMU_TAG_VPAGE) << IOMMU_PAGE_SHIFT);
+		printk(KERN_ERR "%s: IOMMU DATA(%d)[valid(%d) cache(%d) "
+		       "ppg(%016lx)]\n",
+		       pbm->name, i,
+		       ((data_val & PSYCHO_IOMMU_DATA_VALID) ? 1 : 0),
+		       ((data_val & PSYCHO_IOMMU_DATA_CACHE) ? 1 : 0),
+		       (data_val & PSYCHO_IOMMU_DATA_PPAGE)<<IOMMU_PAGE_SHIFT);
+	}
+}
+
+#define  PSYCHO_IOMMU_CTRL_XLTESTAT	0x0000000006000000UL
+#define  PSYCHO_IOMMU_CTRL_XLTEERR	0x0000000001000000UL
+
+void psycho_check_iommu_error(struct pci_pbm_info *pbm,
+			      unsigned long afsr,
+			      unsigned long afar,
+			      enum psycho_error_type type)
+{
+	u64 control, iommu_tag[16], iommu_data[16];
+	struct iommu *iommu = pbm->iommu;
+	unsigned long flags;
+
+	spin_lock_irqsave(&iommu->lock, flags);
+	control = upa_readq(iommu->iommu_control);
+	if (control & PSYCHO_IOMMU_CTRL_XLTEERR) {
+		const char *type_str;
+
+		control &= ~PSYCHO_IOMMU_CTRL_XLTEERR;
+		upa_writeq(control, iommu->iommu_control);
+
+		switch ((control & PSYCHO_IOMMU_CTRL_XLTESTAT) >> 25UL) {
+		case 0:
+			type_str = "Protection Error";
+			break;
+		case 1:
+			type_str = "Invalid Error";
+			break;
+		case 2:
+			type_str = "TimeOut Error";
+			break;
+		case 3:
+		default:
+			type_str = "ECC Error";
+			break;
+		};
+		printk(KERN_ERR "%s: IOMMU Error, type[%s]\n",
+		       pbm->name, type_str);
+
+		/* It is very possible for another DVMA to occur while
+		 * we do this probe, and corrupt the system further.
+		 * But we are so screwed at this point that we are
+		 * likely to crash hard anyways, so get as much
+		 * diagnostic information to the console as we can.
+		 */
+		psycho_record_iommu_tags_and_data(pbm, iommu_tag, iommu_data);
+		psycho_dump_iommu_tags_and_data(pbm, iommu_tag, iommu_data);
+	}
+	psycho_check_stc_error(pbm);
+	spin_unlock_irqrestore(&iommu->lock, flags);
+}
+
+#define  PSYCHO_PCICTRL_SBH_ERR	 0x0000000800000000UL
+#define  PSYCHO_PCICTRL_SERR	 0x0000000400000000UL
+
+static irqreturn_t psycho_pcierr_intr_other(struct pci_pbm_info *pbm)
+{
+	irqreturn_t ret = IRQ_NONE;
+	u64 csr, csr_error_bits;
+	u16 stat, *addr;
+
+	csr = upa_readq(pbm->pci_csr);
+	csr_error_bits = csr & (PSYCHO_PCICTRL_SBH_ERR | PSYCHO_PCICTRL_SERR);
+	if (csr_error_bits) {
+		/* Clear the errors.  */
+		upa_writeq(csr, pbm->pci_csr);
+
+		/* Log 'em.  */
+		if (csr_error_bits & PSYCHO_PCICTRL_SBH_ERR)
+			printk(KERN_ERR "%s: PCI streaming byte hole "
+			       "error asserted.\n", pbm->name);
+		if (csr_error_bits & PSYCHO_PCICTRL_SERR)
+			printk(KERN_ERR "%s: PCI SERR signal asserted.\n",
+			       pbm->name);
+		ret = IRQ_HANDLED;
+	}
+	addr = psycho_pci_config_mkaddr(pbm, pbm->pci_first_busno,
+					0, PCI_STATUS);
+	pci_config_read16(addr, &stat);
+	if (stat & (PCI_STATUS_PARITY |
+		    PCI_STATUS_SIG_TARGET_ABORT |
+		    PCI_STATUS_REC_TARGET_ABORT |
+		    PCI_STATUS_REC_MASTER_ABORT |
+		    PCI_STATUS_SIG_SYSTEM_ERROR)) {
+		printk(KERN_ERR "%s: PCI bus error, PCI_STATUS[%04x]\n",
+		       pbm->name, stat);
+		pci_config_write16(addr, 0xffff);
+		ret = IRQ_HANDLED;
+	}
+	return ret;
+}
+
+#define  PSYCHO_PCIAFSR_PMA	0x8000000000000000UL
+#define  PSYCHO_PCIAFSR_PTA	0x4000000000000000UL
+#define  PSYCHO_PCIAFSR_PRTRY	0x2000000000000000UL
+#define  PSYCHO_PCIAFSR_PPERR	0x1000000000000000UL
+#define  PSYCHO_PCIAFSR_SMA	0x0800000000000000UL
+#define  PSYCHO_PCIAFSR_STA	0x0400000000000000UL
+#define  PSYCHO_PCIAFSR_SRTRY	0x0200000000000000UL
+#define  PSYCHO_PCIAFSR_SPERR	0x0100000000000000UL
+#define  PSYCHO_PCIAFSR_RESV1	0x00ff000000000000UL
+#define  PSYCHO_PCIAFSR_BMSK	0x0000ffff00000000UL
+#define  PSYCHO_PCIAFSR_BLK	0x0000000080000000UL
+#define  PSYCHO_PCIAFSR_RESV2	0x0000000040000000UL
+#define  PSYCHO_PCIAFSR_MID	0x000000003e000000UL
+#define  PSYCHO_PCIAFSR_RESV3	0x0000000001ffffffUL
+
+irqreturn_t psycho_pcierr_intr(int irq, void *dev_id)
+{
+	struct pci_pbm_info *pbm = dev_id;
+	u64 afsr, afar, error_bits;
+	int reported;
+
+	afsr = upa_readq(pbm->pci_afsr);
+	afar = upa_readq(pbm->pci_afar);
+	error_bits = afsr &
+		(PSYCHO_PCIAFSR_PMA | PSYCHO_PCIAFSR_PTA |
+		 PSYCHO_PCIAFSR_PRTRY | PSYCHO_PCIAFSR_PPERR |
+		 PSYCHO_PCIAFSR_SMA | PSYCHO_PCIAFSR_STA |
+		 PSYCHO_PCIAFSR_SRTRY | PSYCHO_PCIAFSR_SPERR);
+	if (!error_bits)
+		return psycho_pcierr_intr_other(pbm);
+	upa_writeq(error_bits, pbm->pci_afsr);
+	printk(KERN_ERR "%s: PCI Error, primary error type[%s]\n",
+	       pbm->name,
+	       (((error_bits & PSYCHO_PCIAFSR_PMA) ?
+		 "Master Abort" :
+		 ((error_bits & PSYCHO_PCIAFSR_PTA) ?
+		  "Target Abort" :
+		  ((error_bits & PSYCHO_PCIAFSR_PRTRY) ?
+		   "Excessive Retries" :
+		   ((error_bits & PSYCHO_PCIAFSR_PPERR) ?
+		    "Parity Error" : "???"))))));
+	printk(KERN_ERR "%s: bytemask[%04lx] UPA_MID[%02lx] was_block(%d)\n",
+	       pbm->name,
+	       (afsr & PSYCHO_PCIAFSR_BMSK) >> 32UL,
+	       (afsr & PSYCHO_PCIAFSR_MID) >> 25UL,
+	       (afsr & PSYCHO_PCIAFSR_BLK) ? 1 : 0);
+	printk(KERN_ERR "%s: PCI AFAR [%016lx]\n", pbm->name, afar);
+	printk(KERN_ERR "%s: PCI Secondary errors [", pbm->name);
+	reported = 0;
+	if (afsr & PSYCHO_PCIAFSR_SMA) {
+		reported++;
+		printk("(Master Abort)");
+	}
+	if (afsr & PSYCHO_PCIAFSR_STA) {
+		reported++;
+		printk("(Target Abort)");
+	}
+	if (afsr & PSYCHO_PCIAFSR_SRTRY) {
+		reported++;
+		printk("(Excessive Retries)");
+	}
+	if (afsr & PSYCHO_PCIAFSR_SPERR) {
+		reported++;
+		printk("(Parity Error)");
+	}
+	if (!reported)
+		printk("(none)");
+	printk("]\n");
+
+	if (error_bits & (PSYCHO_PCIAFSR_PTA | PSYCHO_PCIAFSR_STA)) {
+		psycho_check_iommu_error(pbm, afsr, afar, PCI_ERR);
+		pci_scan_for_target_abort(pbm, pbm->pci_bus);
+	}
+	if (error_bits & (PSYCHO_PCIAFSR_PMA | PSYCHO_PCIAFSR_SMA))
+		pci_scan_for_master_abort(pbm, pbm->pci_bus);
+
+	if (error_bits & (PSYCHO_PCIAFSR_PPERR | PSYCHO_PCIAFSR_SPERR))
+		pci_scan_for_parity_error(pbm, pbm->pci_bus);
+
+	return IRQ_HANDLED;
+}
+
+static void psycho_iommu_flush(struct pci_pbm_info *pbm)
+{
+	int i;
+
+	for (i = 0; i < 16; i++) {
+		unsigned long off = i * 8;
+
+		upa_writeq(0, pbm->controller_regs + PSYCHO_IOMMU_TAG + off);
+		upa_writeq(0, pbm->controller_regs + PSYCHO_IOMMU_DATA + off);
+	}
+}
+
+#define PSYCHO_IOMMU_CONTROL		0x0200UL
+#define  PSYCHO_IOMMU_CTRL_TSBSZ	0x0000000000070000UL
+#define  PSYCHO_IOMMU_TSBSZ_1K      	0x0000000000000000UL
+#define  PSYCHO_IOMMU_TSBSZ_2K      	0x0000000000010000UL
+#define  PSYCHO_IOMMU_TSBSZ_4K      	0x0000000000020000UL
+#define  PSYCHO_IOMMU_TSBSZ_8K      	0x0000000000030000UL
+#define  PSYCHO_IOMMU_TSBSZ_16K     	0x0000000000040000UL
+#define  PSYCHO_IOMMU_TSBSZ_32K     	0x0000000000050000UL
+#define  PSYCHO_IOMMU_TSBSZ_64K     	0x0000000000060000UL
+#define  PSYCHO_IOMMU_TSBSZ_128K    	0x0000000000070000UL
+#define  PSYCHO_IOMMU_CTRL_TBWSZ    	0x0000000000000004UL
+#define  PSYCHO_IOMMU_CTRL_DENAB    	0x0000000000000002UL
+#define  PSYCHO_IOMMU_CTRL_ENAB     	0x0000000000000001UL
+#define PSYCHO_IOMMU_FLUSH		0x0210UL
+#define PSYCHO_IOMMU_TSBBASE		0x0208UL
+
+int psycho_iommu_init(struct pci_pbm_info *pbm, int tsbsize,
+		      u32 dvma_offset, u32 dma_mask,
+		      unsigned long write_complete_offset)
+{
+	struct iommu *iommu = pbm->iommu;
+	u64 control;
+	int err;
+
+	iommu->iommu_control  = pbm->controller_regs + PSYCHO_IOMMU_CONTROL;
+	iommu->iommu_tsbbase  = pbm->controller_regs + PSYCHO_IOMMU_TSBBASE;
+	iommu->iommu_flush    = pbm->controller_regs + PSYCHO_IOMMU_FLUSH;
+	iommu->iommu_tags     = pbm->controller_regs + PSYCHO_IOMMU_TAG;
+	iommu->write_complete_reg = (pbm->controller_regs +
+				     write_complete_offset);
+
+	iommu->iommu_ctxflush = 0;
+
+	control = upa_readq(iommu->iommu_control);
+	control |= PSYCHO_IOMMU_CTRL_DENAB;
+	upa_writeq(control, iommu->iommu_control);
+
+	psycho_iommu_flush(pbm);
+
+	/* Leave diag mode enabled for full-flushing done in pci_iommu.c */
+	err = iommu_table_init(iommu, tsbsize * 1024 * 8,
+			       dvma_offset, dma_mask, pbm->numa_node);
+	if (err)
+		return err;
+
+	upa_writeq(__pa(iommu->page_table), iommu->iommu_tsbbase);
+
+	control = upa_readq(iommu->iommu_control);
+	control &= ~(PSYCHO_IOMMU_CTRL_TSBSZ | PSYCHO_IOMMU_CTRL_TBWSZ);
+	control |= PSYCHO_IOMMU_CTRL_ENAB;
+
+	switch (tsbsize) {
+	case 64:
+		control |= PSYCHO_IOMMU_TSBSZ_64K;
+		break;
+	case 128:
+		control |= PSYCHO_IOMMU_TSBSZ_128K;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	upa_writeq(control, iommu->iommu_control);
+
+	return 0;
+
+}
+
+void psycho_pbm_init_common(struct pci_pbm_info *pbm, struct of_device *op,
+			    const char *chip_name, int chip_type)
+{
+	struct device_node *dp = op->node;
+
+	pbm->name = dp->full_name;
+	pbm->numa_node = -1;
+	pbm->chip_type = chip_type;
+	pbm->chip_version = of_getintprop_default(dp, "version#", 0);
+	pbm->chip_revision = of_getintprop_default(dp, "module-revision#", 0);
+	pbm->op = op;
+	pbm->pci_ops = &sun4u_pci_ops;
+	pbm->config_space_reg_bits = 8;
+	pbm->index = pci_num_pbms++;
+	pci_get_pbm_props(pbm);
+	pci_determine_mem_io_space(pbm);
+
+	printk(KERN_INFO "%s: %s PCI Bus Module ver[%x:%x]\n",
+	       pbm->name, chip_name,
+	       pbm->chip_version, pbm->chip_revision);
+}
diff --git a/arch/sparc64/kernel/psycho_common.h b/arch/sparc64/kernel/psycho_common.h
new file mode 100644
index 000000000000..092c278ef28d
--- /dev/null
+++ b/arch/sparc64/kernel/psycho_common.h
@@ -0,0 +1,48 @@
+#ifndef _PSYCHO_COMMON_H
+#define _PSYCHO_COMMON_H
+
+/* U2P Programmer's Manual, page 13-55, configuration space
+ * address format:
+ * 
+ *  32             24 23 16 15    11 10       8 7   2  1 0
+ * ---------------------------------------------------------
+ * |0 0 0 0 0 0 0 0 1| bus | device | function | reg | 0 0 |
+ * ---------------------------------------------------------
+ */
+#define PSYCHO_CONFIG_BASE(PBM)	\
+	((PBM)->config_space | (1UL << 24))
+#define PSYCHO_CONFIG_ENCODE(BUS, DEVFN, REG)	\
+	(((unsigned long)(BUS)   << 16) |	\
+	 ((unsigned long)(DEVFN) << 8)  |	\
+	 ((unsigned long)(REG)))
+
+static inline void *psycho_pci_config_mkaddr(struct pci_pbm_info *pbm,
+					     unsigned char bus,
+					     unsigned int devfn,
+					     int where)
+{
+	return (void *)
+		(PSYCHO_CONFIG_BASE(pbm) |
+		 PSYCHO_CONFIG_ENCODE(bus, devfn, where));
+}
+
+enum psycho_error_type {
+	UE_ERR, CE_ERR, PCI_ERR
+};
+
+extern void psycho_check_iommu_error(struct pci_pbm_info *pbm,
+				     unsigned long afsr,
+				     unsigned long afar,
+				     enum psycho_error_type type);
+
+extern irqreturn_t psycho_pcierr_intr(int irq, void *dev_id);
+
+extern int psycho_iommu_init(struct pci_pbm_info *pbm, int tsbsize,
+			     u32 dvma_offset, u32 dma_mask,
+			     unsigned long write_complete_offset);
+
+extern void psycho_pbm_init_common(struct pci_pbm_info *pbm,
+				   struct of_device *op,
+				   const char *chip_name, int chip_type);
+
+#endif /* _PSYCHO_COMMON_H */
diff --git a/arch/sparc64/kernel/ptrace.c b/arch/sparc64/kernel/ptrace.c
index 10306e476e38..f43adbc773ca 100644
--- a/arch/sparc64/kernel/ptrace.c
+++ b/arch/sparc64/kernel/ptrace.c
@@ -1050,31 +1050,17 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 	return ret;
 }
 
-asmlinkage int syscall_trace(struct pt_regs *regs, int syscall_exit_p)
+asmlinkage int syscall_trace_enter(struct pt_regs *regs)
 {
 	int ret = 0;
 
 	/* do the secure computing check first */
 	secure_computing(regs->u_regs[UREG_G1]);
 
-	if (unlikely(current->audit_context) && syscall_exit_p) {
-		unsigned long tstate = regs->tstate;
-		int result = AUDITSC_SUCCESS;
-
-		if (unlikely(tstate & (TSTATE_XCARRY | TSTATE_ICARRY)))
-			result = AUDITSC_FAILURE;
-
-		audit_syscall_exit(result, regs->u_regs[UREG_I0]);
-	}
-
-	if (test_thread_flag(TIF_SYSCALL_TRACE)) {
-		if (syscall_exit_p)
-			tracehook_report_syscall_exit(regs, 0);
-		else
-			ret = tracehook_report_syscall_entry(regs);
-	}
+	if (test_thread_flag(TIF_SYSCALL_TRACE))
+		ret = tracehook_report_syscall_entry(regs);
 
-	if (unlikely(current->audit_context) && !syscall_exit_p && !ret)
+	if (unlikely(current->audit_context) && !ret)
 		audit_syscall_entry((test_thread_flag(TIF_32BIT) ?
 				     AUDIT_ARCH_SPARC :
 				     AUDIT_ARCH_SPARC64),
@@ -1086,3 +1072,19 @@ asmlinkage int syscall_trace(struct pt_regs *regs, int syscall_exit_p)
 
 	return ret;
 }
+
+asmlinkage void syscall_trace_leave(struct pt_regs *regs)
+{
+	if (unlikely(current->audit_context)) {
+		unsigned long tstate = regs->tstate;
+		int result = AUDITSC_SUCCESS;
+
+		if (unlikely(tstate & (TSTATE_XCARRY | TSTATE_ICARRY)))
+			result = AUDITSC_FAILURE;
+
+		audit_syscall_exit(result, regs->u_regs[UREG_I0]);
+	}
+
+	if (test_thread_flag(TIF_SYSCALL_TRACE))
+		tracehook_report_syscall_exit(regs, 0);
+}
diff --git a/arch/sparc64/kernel/reboot.c b/arch/sparc64/kernel/reboot.c
new file mode 100644
index 000000000000..ef89d3d69748
--- /dev/null
+++ b/arch/sparc64/kernel/reboot.c
@@ -0,0 +1,53 @@
+/* reboot.c: reboot/shutdown/halt/poweroff handling
+ *
+ * Copyright (C) 2008 David S. Miller <davem@davemloft.net>
+ */
+#include <linux/kernel.h>
+#include <linux/reboot.h>
+#include <linux/module.h>
+#include <linux/pm.h>
+
+#include <asm/system.h>
+#include <asm/oplib.h>
+#include <asm/prom.h>
+
+/* sysctl - toggle power-off restriction for serial console
+ * systems in machine_power_off()
+ */
+int scons_pwroff = 1;
+
+/* This isn't actually used, it exists merely to satisfy the
+ * reference in kernel/sys.c
+ */
+void (*pm_power_off)(void) = machine_power_off;
+EXPORT_SYMBOL(pm_power_off);
+
+void machine_power_off(void)
+{
+	if (strcmp(of_console_device->type, "serial") || scons_pwroff)
+		prom_halt_power_off();
+
+	prom_halt();
+}
+
+void machine_halt(void)
+{
+	prom_halt();
+	panic("Halt failed!");
+}
+
+void machine_restart(char *cmd)
+{
+	char *p;
+
+	p = strchr(reboot_command, '\n');
+	if (p)
+		*p = 0;
+	if (cmd)
+		prom_reboot(cmd);
+	if (*reboot_command)
+		prom_reboot(reboot_command);
+	prom_reboot("");
+	panic("Reboot failed!");
+}
+
diff --git a/arch/sparc64/kernel/sbus.c b/arch/sparc64/kernel/sbus.c
index e33a8a660e9e..2ead310066d1 100644
--- a/arch/sparc64/kernel/sbus.c
+++ b/arch/sparc64/kernel/sbus.c
@@ -11,15 +11,17 @@
 #include <linux/slab.h>
 #include <linux/init.h>
 #include <linux/interrupt.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/page.h>
-#include <asm/sbus.h>
 #include <asm/io.h>
 #include <asm/upa.h>
 #include <asm/cache.h>
 #include <asm/dma.h>
 #include <asm/irq.h>
 #include <asm/prom.h>
+#include <asm/oplib.h>
 #include <asm/starfire.h>
 
 #include "iommu_common.h"
@@ -52,13 +54,23 @@
 #define STRBUF_TAG_VALID	0x02UL
 
 /* Enable 64-bit DVMA mode for the given device. */
-void sbus_set_sbus64(struct sbus_dev *sdev, int bursts)
+void sbus_set_sbus64(struct device *dev, int bursts)
 {
-	struct iommu *iommu = sdev->ofdev.dev.archdata.iommu;
-	int slot = sdev->slot;
+	struct iommu *iommu = dev->archdata.iommu;
+	struct of_device *op = to_of_device(dev);
+	const struct linux_prom_registers *regs;
 	unsigned long cfg_reg;
+	int slot;
 	u64 val;
 
+	regs = of_get_property(op->node, "reg", NULL);
+	if (!regs) {
+		printk(KERN_ERR "sbus_set_sbus64: Cannot find regs for %s\n",
+		       op->node->full_name);
+		return;
+	}
+	slot = regs->which_io;
+
 	cfg_reg = iommu->write_complete_reg;
 	switch (slot) {
 	case 0:
@@ -191,10 +203,9 @@ static unsigned long sysio_imap_to_iclr(unsigned long imap)
 	return imap + diff;
 }
 
-unsigned int sbus_build_irq(void *buscookie, unsigned int ino)
+static unsigned int sbus_build_irq(struct of_device *op, unsigned int ino)
 {
-	struct sbus_bus *sbus = (struct sbus_bus *)buscookie;
-	struct iommu *iommu = sbus->ofdev.dev.archdata.iommu;
+	struct iommu *iommu = op->dev.archdata.iommu;
 	unsigned long reg_base = iommu->write_complete_reg - 0x2000UL;
 	unsigned long imap, iclr;
 	int sbus_level = 0;
@@ -255,12 +266,12 @@ unsigned int sbus_build_irq(void *buscookie, unsigned int ino)
 #define  SYSIO_UEAFSR_RESV2 0x0000001fffffffffUL /* Reserved                  */
 static irqreturn_t sysio_ue_handler(int irq, void *dev_id)
 {
-	struct sbus_bus *sbus = dev_id;
-	struct iommu *iommu = sbus->ofdev.dev.archdata.iommu;
+	struct of_device *op = dev_id;
+	struct iommu *iommu = op->dev.archdata.iommu;
 	unsigned long reg_base = iommu->write_complete_reg - 0x2000UL;
 	unsigned long afsr_reg, afar_reg;
 	unsigned long afsr, afar, error_bits;
-	int reported;
+	int reported, portid;
 
 	afsr_reg = reg_base + SYSIO_UE_AFSR;
 	afar_reg = reg_base + SYSIO_UE_AFAR;
@@ -275,9 +286,11 @@ static irqreturn_t sysio_ue_handler(int irq, void *dev_id)
 		 SYSIO_UEAFSR_SPIO | SYSIO_UEAFSR_SDRD | SYSIO_UEAFSR_SDWR);
 	upa_writeq(error_bits, afsr_reg);
 
+	portid = of_getintprop_default(op->node, "portid", -1);
+
 	/* Log the error. */
 	printk("SYSIO[%x]: Uncorrectable ECC Error, primary error type[%s]\n",
-	       sbus->portid,
+	       portid,
 	       (((error_bits & SYSIO_UEAFSR_PPIO) ?
 		 "PIO" :
 		 ((error_bits & SYSIO_UEAFSR_PDRD) ?
@@ -285,12 +298,12 @@ static irqreturn_t sysio_ue_handler(int irq, void *dev_id)
 		  ((error_bits & SYSIO_UEAFSR_PDWR) ?
 		   "DVMA Write" : "???")))));
 	printk("SYSIO[%x]: DOFF[%lx] SIZE[%lx] MID[%lx]\n",
-	       sbus->portid,
+	       portid,
 	       (afsr & SYSIO_UEAFSR_DOFF) >> 45UL,
 	       (afsr & SYSIO_UEAFSR_SIZE) >> 42UL,
 	       (afsr & SYSIO_UEAFSR_MID) >> 37UL);
-	printk("SYSIO[%x]: AFAR[%016lx]\n", sbus->portid, afar);
-	printk("SYSIO[%x]: Secondary UE errors [", sbus->portid);
+	printk("SYSIO[%x]: AFAR[%016lx]\n", portid, afar);
+	printk("SYSIO[%x]: Secondary UE errors [", portid);
 	reported = 0;
 	if (afsr & SYSIO_UEAFSR_SPIO) {
 		reported++;
@@ -327,12 +340,12 @@ static irqreturn_t sysio_ue_handler(int irq, void *dev_id)
 #define  SYSIO_CEAFSR_RESV2 0x0000001fffffffffUL /* Reserved                  */
 static irqreturn_t sysio_ce_handler(int irq, void *dev_id)
 {
-	struct sbus_bus *sbus = dev_id;
-	struct iommu *iommu = sbus->ofdev.dev.archdata.iommu;
+	struct of_device *op = dev_id;
+	struct iommu *iommu = op->dev.archdata.iommu;
 	unsigned long reg_base = iommu->write_complete_reg - 0x2000UL;
 	unsigned long afsr_reg, afar_reg;
 	unsigned long afsr, afar, error_bits;
-	int reported;
+	int reported, portid;
 
 	afsr_reg = reg_base + SYSIO_CE_AFSR;
 	afar_reg = reg_base + SYSIO_CE_AFAR;
@@ -347,8 +360,10 @@ static irqreturn_t sysio_ce_handler(int irq, void *dev_id)
 		 SYSIO_CEAFSR_SPIO | SYSIO_CEAFSR_SDRD | SYSIO_CEAFSR_SDWR);
 	upa_writeq(error_bits, afsr_reg);
 
+	portid = of_getintprop_default(op->node, "portid", -1);
+
 	printk("SYSIO[%x]: Correctable ECC Error, primary error type[%s]\n",
-	       sbus->portid,
+	       portid,
 	       (((error_bits & SYSIO_CEAFSR_PPIO) ?
 		 "PIO" :
 		 ((error_bits & SYSIO_CEAFSR_PDRD) ?
@@ -360,14 +375,14 @@ static irqreturn_t sysio_ce_handler(int irq, void *dev_id)
 	 * XXX UDB CE trap handler does... -DaveM
 	 */
 	printk("SYSIO[%x]: DOFF[%lx] ECC Syndrome[%lx] Size[%lx] MID[%lx]\n",
-	       sbus->portid,
+	       portid,
 	       (afsr & SYSIO_CEAFSR_DOFF) >> 45UL,
 	       (afsr & SYSIO_CEAFSR_ESYND) >> 48UL,
 	       (afsr & SYSIO_CEAFSR_SIZE) >> 42UL,
 	       (afsr & SYSIO_CEAFSR_MID) >> 37UL);
-	printk("SYSIO[%x]: AFAR[%016lx]\n", sbus->portid, afar);
+	printk("SYSIO[%x]: AFAR[%016lx]\n", portid, afar);
 
-	printk("SYSIO[%x]: Secondary CE errors [", sbus->portid);
+	printk("SYSIO[%x]: Secondary CE errors [", portid);
 	reported = 0;
 	if (afsr & SYSIO_CEAFSR_SPIO) {
 		reported++;
@@ -404,11 +419,11 @@ static irqreturn_t sysio_ce_handler(int irq, void *dev_id)
 #define  SYSIO_SBAFSR_RESV3 0x0000001fffffffffUL /* Reserved                  */
 static irqreturn_t sysio_sbus_error_handler(int irq, void *dev_id)
 {
-	struct sbus_bus *sbus = dev_id;
-	struct iommu *iommu = sbus->ofdev.dev.archdata.iommu;
+	struct of_device *op = dev_id;
+	struct iommu *iommu = op->dev.archdata.iommu;
 	unsigned long afsr_reg, afar_reg, reg_base;
 	unsigned long afsr, afar, error_bits;
-	int reported;
+	int reported, portid;
 
 	reg_base = iommu->write_complete_reg - 0x2000UL;
 	afsr_reg = reg_base + SYSIO_SBUS_AFSR;
@@ -423,9 +438,11 @@ static irqreturn_t sysio_sbus_error_handler(int irq, void *dev_id)
 		 SYSIO_SBAFSR_SLE | SYSIO_SBAFSR_STO | SYSIO_SBAFSR_SBERR);
 	upa_writeq(error_bits, afsr_reg);
 
+	portid = of_getintprop_default(op->node, "portid", -1);
+
 	/* Log the error. */
 	printk("SYSIO[%x]: SBUS Error, primary error type[%s] read(%d)\n",
-	       sbus->portid,
+	       portid,
 	       (((error_bits & SYSIO_SBAFSR_PLE) ?
 		 "Late PIO Error" :
 		 ((error_bits & SYSIO_SBAFSR_PTO) ?
@@ -434,11 +451,11 @@ static irqreturn_t sysio_sbus_error_handler(int irq, void *dev_id)
 		   "Error Ack" : "???")))),
 	       (afsr & SYSIO_SBAFSR_RD) ? 1 : 0);
 	printk("SYSIO[%x]: size[%lx] MID[%lx]\n",
-	       sbus->portid,
+	       portid,
 	       (afsr & SYSIO_SBAFSR_SIZE) >> 42UL,
 	       (afsr & SYSIO_SBAFSR_MID) >> 37UL);
-	printk("SYSIO[%x]: AFAR[%016lx]\n", sbus->portid, afar);
-	printk("SYSIO[%x]: Secondary SBUS errors [", sbus->portid);
+	printk("SYSIO[%x]: AFAR[%016lx]\n", portid, afar);
+	printk("SYSIO[%x]: Secondary SBUS errors [", portid);
 	reported = 0;
 	if (afsr & SYSIO_SBAFSR_SLE) {
 		reported++;
@@ -470,34 +487,37 @@ static irqreturn_t sysio_sbus_error_handler(int irq, void *dev_id)
 #define SYSIO_CE_INO		0x35
 #define SYSIO_SBUSERR_INO	0x36
 
-static void __init sysio_register_error_handlers(struct sbus_bus *sbus)
+static void __init sysio_register_error_handlers(struct of_device *op)
 {
-	struct iommu *iommu = sbus->ofdev.dev.archdata.iommu;
+	struct iommu *iommu = op->dev.archdata.iommu;
 	unsigned long reg_base = iommu->write_complete_reg - 0x2000UL;
 	unsigned int irq;
 	u64 control;
+	int portid;
+
+	portid = of_getintprop_default(op->node, "portid", -1);
 
-	irq = sbus_build_irq(sbus, SYSIO_UE_INO);
+	irq = sbus_build_irq(op, SYSIO_UE_INO);
 	if (request_irq(irq, sysio_ue_handler, 0,
-			"SYSIO_UE", sbus) < 0) {
+			"SYSIO_UE", op) < 0) {
 		prom_printf("SYSIO[%x]: Cannot register UE interrupt.\n",
-			    sbus->portid);
+			    portid);
 		prom_halt();
 	}
 
-	irq = sbus_build_irq(sbus, SYSIO_CE_INO);
+	irq = sbus_build_irq(op, SYSIO_CE_INO);
 	if (request_irq(irq, sysio_ce_handler, 0,
-			"SYSIO_CE", sbus) < 0) {
+			"SYSIO_CE", op) < 0) {
 		prom_printf("SYSIO[%x]: Cannot register CE interrupt.\n",
-			    sbus->portid);
+			    portid);
 		prom_halt();
 	}
 
-	irq = sbus_build_irq(sbus, SYSIO_SBUSERR_INO);
+	irq = sbus_build_irq(op, SYSIO_SBUSERR_INO);
 	if (request_irq(irq, sysio_sbus_error_handler, 0,
-			"SYSIO_SBERR", sbus) < 0) {
+			"SYSIO_SBERR", op) < 0) {
 		prom_printf("SYSIO[%x]: Cannot register SBUS Error interrupt.\n",
-			    sbus->portid);
+			    portid);
 		prom_halt();
 	}
 
@@ -513,19 +533,15 @@ static void __init sysio_register_error_handlers(struct sbus_bus *sbus)
 }
 
 /* Boot time initialization. */
-static void __init sbus_iommu_init(int __node, struct sbus_bus *sbus)
+static void __init sbus_iommu_init(struct of_device *op)
 {
 	const struct linux_prom64_registers *pr;
-	struct device_node *dp;
+	struct device_node *dp = op->node;
 	struct iommu *iommu;
 	struct strbuf *strbuf;
 	unsigned long regs, reg_base;
+	int i, portid;
 	u64 control;
-	int i;
-
-	dp = of_find_node_by_phandle(__node);
-
-	sbus->portid = of_getintprop_default(dp, "upa-portid", -1);
 
 	pr = of_get_property(dp, "reg", NULL);
 	if (!pr) {
@@ -542,9 +558,9 @@ static void __init sbus_iommu_init(int __node, struct sbus_bus *sbus)
 	if (!strbuf)
 		goto fatal_memory_error;
 
-	sbus->ofdev.dev.archdata.iommu = iommu;
-	sbus->ofdev.dev.archdata.stc = strbuf;
-	sbus->ofdev.dev.archdata.numa_node = -1;
+	op->dev.archdata.iommu = iommu;
+	op->dev.archdata.stc = strbuf;
+	op->dev.archdata.numa_node = -1;
 
 	reg_base = regs + SYSIO_IOMMUREG_BASE;
 	iommu->iommu_control = reg_base + IOMMU_CONTROL;
@@ -572,8 +588,9 @@ static void __init sbus_iommu_init(int __node, struct sbus_bus *sbus)
 	 */
 	iommu->write_complete_reg = regs + 0x2000UL;
 
-	printk("SYSIO: UPA portID %x, at %016lx\n",
-	       sbus->portid, regs);
+	portid = of_getintprop_default(op->node, "portid", -1);
+	printk(KERN_INFO "SYSIO: UPA portID %x, at %016lx\n",
+	       portid, regs);
 
 	/* Setup for TSB_SIZE=7, TBW_SIZE=0, MMU_DE=1, MMU_EN=1 */
 	if (iommu_table_init(iommu, IO_TSB_SIZE, MAP_BASE, 0xffffffff, -1))
@@ -631,56 +648,27 @@ static void __init sbus_iommu_init(int __node, struct sbus_bus *sbus)
 
 	/* Now some Xfire specific grot... */
 	if (this_is_starfire)
-		starfire_hookup(sbus->portid);
+		starfire_hookup(portid);
 
-	sysio_register_error_handlers(sbus);
+	sysio_register_error_handlers(op);
 	return;
 
 fatal_memory_error:
 	prom_printf("sbus_iommu_init: Fatal memory allocation error.\n");
 }
 
-void sbus_fill_device_irq(struct sbus_dev *sdev)
+static int __init sbus_init(void)
 {
-	struct device_node *dp = of_find_node_by_phandle(sdev->prom_node);
-	const struct linux_prom_irqs *irqs;
-
-	irqs = of_get_property(dp, "interrupts", NULL);
-	if (!irqs) {
-		sdev->irqs[0] = 0;
-		sdev->num_irqs = 0;
-	} else {
-		unsigned int pri = irqs[0].pri;
+	struct device_node *dp;
 
-		sdev->num_irqs = 1;
-		if (pri < 0x20)
-			pri += sdev->slot * 8;
+	for_each_node_by_name(dp, "sbus") {
+		struct of_device *op = of_find_device_by_node(dp);
 
-		sdev->irqs[0] =	sbus_build_irq(sdev->bus, pri);
+		sbus_iommu_init(op);
+		of_propagate_archdata(op);
 	}
-}
 
-void __init sbus_arch_bus_ranges_init(struct device_node *pn, struct sbus_bus *sbus)
-{
-}
-
-void __init sbus_setup_iommu(struct sbus_bus *sbus, struct device_node *dp)
-{
-	sbus_iommu_init(dp->node, sbus);
-}
-
-void __init sbus_setup_arch_props(struct sbus_bus *sbus, struct device_node *dp)
-{
-}
-
-int __init sbus_arch_preinit(void)
-{
 	return 0;
 }
 
-void __init sbus_arch_postinit(void)
-{
-	extern void firetruck_init(void);
-
-	firetruck_init();
-}
+subsys_initcall(sbus_init);
diff --git a/arch/sparc64/kernel/smp.c b/arch/sparc64/kernel/smp.c
index 2be166c544ca..e5627118e613 100644
--- a/arch/sparc64/kernel/smp.c
+++ b/arch/sparc64/kernel/smp.c
@@ -21,6 +21,7 @@
 #include <linux/jiffies.h>
 #include <linux/profile.h>
 #include <linux/lmb.h>
+#include <linux/cpu.h>
 
 #include <asm/head.h>
 #include <asm/ptrace.h>
@@ -115,6 +116,9 @@ void __cpuinit smp_callin(void)
 	atomic_inc(&init_mm.mm_count);
 	current->active_mm = &init_mm;
 
+	/* inform the notifiers about the new cpu */
+	notify_cpu_starting(cpuid);
+
 	while (!cpu_isset(cpuid, smp_commenced_mask))
 		rmb();
 
diff --git a/arch/sparc64/kernel/sparc64_ksyms.c b/arch/sparc64/kernel/sparc64_ksyms.c
index 0804f71df6cb..30bba8b0a3b0 100644
--- a/arch/sparc64/kernel/sparc64_ksyms.c
+++ b/arch/sparc64/kernel/sparc64_ksyms.c
@@ -36,7 +36,6 @@
 #include <asm/elf.h>
 #include <asm/head.h>
 #include <asm/smp.h>
-#include <asm/mostek.h>
 #include <asm/ptrace.h>
 #include <asm/uaccess.h>
 #include <asm/checksum.h>
@@ -44,12 +43,8 @@
 #include <asm/pgalloc.h>
 #include <asm/cacheflush.h>
 #ifdef CONFIG_SBUS
-#include <asm/sbus.h>
 #include <asm/dma.h>
 #endif
-#ifdef CONFIG_PCI
-#include <asm/ebus.h>
-#endif
 #include <asm/ns87303.h>
 #include <asm/timer.h>
 #include <asm/cpudata.h>
@@ -68,7 +63,6 @@ extern void *__memscan_zero(void *, size_t);
 extern void *__memscan_generic(void *, int, size_t);
 extern int __memcmp(const void *, const void *, __kernel_size_t);
 extern __kernel_size_t strlen(const char *);
-extern void syscall_trace(struct pt_regs *, int);
 extern void sys_sigsuspend(void);
 extern int compat_sys_ioctl(unsigned int fd, unsigned int cmd, u32 arg);
 extern int (*handle_mathemu)(struct pt_regs *, struct fpustate *);
@@ -154,26 +148,12 @@ EXPORT_SYMBOL(flush_dcache_page);
 EXPORT_SYMBOL(__flush_dcache_range);
 #endif
 
-EXPORT_SYMBOL(mostek_lock);
-EXPORT_SYMBOL(mstk48t02_regs);
 #ifdef CONFIG_SUN_AUXIO
 EXPORT_SYMBOL(auxio_set_led);
 EXPORT_SYMBOL(auxio_set_lte);
 #endif
 #ifdef CONFIG_SBUS
-EXPORT_SYMBOL(sbus_root);
-EXPORT_SYMBOL(dma_chain);
 EXPORT_SYMBOL(sbus_set_sbus64);
-EXPORT_SYMBOL(sbus_alloc_consistent);
-EXPORT_SYMBOL(sbus_free_consistent);
-EXPORT_SYMBOL(sbus_map_single);
-EXPORT_SYMBOL(sbus_unmap_single);
-EXPORT_SYMBOL(sbus_map_sg);
-EXPORT_SYMBOL(sbus_unmap_sg);
-EXPORT_SYMBOL(sbus_dma_sync_single_for_cpu);
-EXPORT_SYMBOL(sbus_dma_sync_single_for_device);
-EXPORT_SYMBOL(sbus_dma_sync_sg_for_cpu);
-EXPORT_SYMBOL(sbus_dma_sync_sg_for_device);
 #endif
 EXPORT_SYMBOL(outsb);
 EXPORT_SYMBOL(outsw);
@@ -182,7 +162,6 @@ EXPORT_SYMBOL(insb);
 EXPORT_SYMBOL(insw);
 EXPORT_SYMBOL(insl);
 #ifdef CONFIG_PCI
-EXPORT_SYMBOL(ebus_chain);
 EXPORT_SYMBOL(pci_alloc_consistent);
 EXPORT_SYMBOL(pci_free_consistent);
 EXPORT_SYMBOL(pci_map_single);
@@ -300,3 +279,5 @@ EXPORT_SYMBOL(xor_niagara_2);
 EXPORT_SYMBOL(xor_niagara_3);
 EXPORT_SYMBOL(xor_niagara_4);
 EXPORT_SYMBOL(xor_niagara_5);
+
+EXPORT_SYMBOL_GPL(real_hard_smp_processor_id);
diff --git a/arch/sparc64/kernel/sstate.c b/arch/sparc64/kernel/sstate.c
index 5b6e75b7f052..8cdbe5946b43 100644
--- a/arch/sparc64/kernel/sstate.c
+++ b/arch/sparc64/kernel/sstate.c
@@ -1,14 +1,15 @@
 /* sstate.c: System soft state support.
  *
- * Copyright (C) 2007 David S. Miller <davem@davemloft.net>
+ * Copyright (C) 2007, 2008 David S. Miller <davem@davemloft.net>
  */
 
 #include <linux/kernel.h>
 #include <linux/notifier.h>
+#include <linux/reboot.h>
 #include <linux/init.h>
 
 #include <asm/hypervisor.h>
-#include <asm/sstate.h>
+#include <asm/spitfire.h>
 #include <asm/oplib.h>
 #include <asm/head.h>
 #include <asm/io.h>
@@ -50,31 +51,34 @@ static const char rebooting_msg[32] __attribute__((aligned(32))) =
 static const char panicing_msg[32] __attribute__((aligned(32))) =
 	"Linux panicing";
 
-void sstate_booting(void)
+static int sstate_reboot_call(struct notifier_block *np, unsigned long type, void *_unused)
 {
-	do_set_sstate(HV_SOFT_STATE_TRANSITION, booting_msg);
-}
+	const char *msg;
 
-void sstate_running(void)
-{
-	do_set_sstate(HV_SOFT_STATE_NORMAL, running_msg);
-}
+	switch (type) {
+	case SYS_DOWN:
+	default:
+		msg = rebooting_msg;
+		break;
 
-void sstate_halt(void)
-{
-	do_set_sstate(HV_SOFT_STATE_TRANSITION, halting_msg);
-}
+	case SYS_HALT:
+		msg = halting_msg;
+		break;
 
-void sstate_poweroff(void)
-{
-	do_set_sstate(HV_SOFT_STATE_TRANSITION, poweroff_msg);
-}
+	case SYS_POWER_OFF:
+		msg = poweroff_msg;
+		break;
+	}
 
-void sstate_reboot(void)
-{
-	do_set_sstate(HV_SOFT_STATE_TRANSITION, rebooting_msg);
+	do_set_sstate(HV_SOFT_STATE_TRANSITION, msg);
+
+	return NOTIFY_OK;
 }
 
+static struct notifier_block sstate_reboot_notifier = {
+	.notifier_call = sstate_reboot_call,
+};
+
 static int sstate_panic_event(struct notifier_block *n, unsigned long event, void *ptr)
 {
 	do_set_sstate(HV_SOFT_STATE_TRANSITION, panicing_msg);
@@ -87,18 +91,37 @@ static struct notifier_block sstate_panic_block = {
 	.priority	=	INT_MAX,
 };
 
-void __init sun4v_sstate_init(void)
+static int __init sstate_init(void)
 {
 	unsigned long major, minor;
 
+	if (tlb_type != hypervisor)
+		return 0;
+
 	major = 1;
 	minor = 0;
 	if (sun4v_hvapi_register(HV_GRP_SOFT_STATE, major, &minor))
-		return;
+		return 0;
 
 	hv_supports_soft_state = 1;
 
 	prom_sun4v_guest_soft_state();
+
+	do_set_sstate(HV_SOFT_STATE_TRANSITION, booting_msg);
+
 	atomic_notifier_chain_register(&panic_notifier_list,
 				       &sstate_panic_block);
+	register_reboot_notifier(&sstate_reboot_notifier);
+
+	return 0;
 }
+
+core_initcall(sstate_init);
+
+static int __init sstate_running(void)
+{
+	do_set_sstate(HV_SOFT_STATE_NORMAL, running_msg);
+	return 0;
+}
+
+late_initcall(sstate_running);
diff --git a/arch/sparc64/kernel/starfire.c b/arch/sparc64/kernel/starfire.c
index 7461581b3bb9..060d0f3a6151 100644
--- a/arch/sparc64/kernel/starfire.c
+++ b/arch/sparc64/kernel/starfire.c
@@ -28,11 +28,6 @@ void check_if_starfire(void)
 		this_is_starfire = 1;
 }
 
-void starfire_cpu_setup(void)
-{
-	/* Currently, nothing to do.  */
-}
-
 int starfire_hard_smp_processor_id(void)
 {
 	return upa_readl(0x1fff40000d0UL);
diff --git a/arch/sparc64/kernel/sys_sparc32.c b/arch/sparc64/kernel/sys_sparc32.c
index 3d118531baff..3320c9d0075f 100644
--- a/arch/sparc64/kernel/sys_sparc32.c
+++ b/arch/sparc64/kernel/sys_sparc32.c
@@ -575,14 +575,6 @@ asmlinkage long sys32_settimeofday(struct compat_timeval __user *tv,
 	return do_sys_settimeofday(tv ? &kts : NULL, tz ? &ktz : NULL);
 }
 
-/* These are here just in case some old sparc32 binary calls it. */
-asmlinkage long sys32_pause(void)
-{
-	current->state = TASK_INTERRUPTIBLE;
-	schedule();
-	return -ERESTARTNOHAND;
-}
-
 asmlinkage compat_ssize_t sys32_pread64(unsigned int fd,
 					char __user *ubuf,
 					compat_size_t count,
diff --git a/arch/sparc64/kernel/syscalls.S b/arch/sparc64/kernel/syscalls.S
index a2f24270ed8a..7a6786a71363 100644
--- a/arch/sparc64/kernel/syscalls.S
+++ b/arch/sparc64/kernel/syscalls.S
@@ -65,9 +65,8 @@ sys32_rt_sigreturn:
 	andcc	%l5, (_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT), %g0
 	be,pt	%icc, rtrap
 	 nop
-	add	%sp, PTREGS_OFF, %o0
-	call	syscall_trace
-	 mov	1, %o1
+	call	syscall_trace_leave
+	 add	%sp, PTREGS_OFF, %o0
 	ba,pt	%xcc, rtrap
 	 nop
 
@@ -159,9 +158,8 @@ linux_sparc_ni_syscall:
 	 or	%l7, %lo(sys_ni_syscall), %l7
 
 linux_syscall_trace32:
-	add	%sp, PTREGS_OFF, %o0
-	call	syscall_trace
-	 clr	%o1
+	call	syscall_trace_enter
+	 add	%sp, PTREGS_OFF, %o0
 	brnz,pn	%o0, 3f
 	 mov	-ENOSYS, %o0
 	srl	%i0, 0, %o0
@@ -172,9 +170,8 @@ linux_syscall_trace32:
 	 srl	%i3, 0, %o3
 
 linux_syscall_trace:
-	add	%sp, PTREGS_OFF, %o0
-	call	syscall_trace
-	 clr	%o1
+	call	syscall_trace_enter
+	 add	%sp, PTREGS_OFF, %o0
 	brnz,pn	%o0, 3f
 	 mov	-ENOSYS, %o0
 	mov	%i0, %o0
@@ -275,9 +272,8 @@ ret_sys_call:
 	b,pt	%xcc, rtrap
 	 stx	%l2, [%sp + PTREGS_OFF + PT_V9_TNPC]
 linux_syscall_trace2:
-	add	%sp, PTREGS_OFF, %o0
-	call	syscall_trace
-	 mov	1, %o1
+	call	syscall_trace_leave
+	 add	%sp, PTREGS_OFF, %o0
 	stx	%l1, [%sp + PTREGS_OFF + PT_V9_TPC]
 	ba,pt	%xcc, rtrap
 	 stx	%l2, [%sp + PTREGS_OFF + PT_V9_TNPC]
diff --git a/arch/sparc64/kernel/systbls.S b/arch/sparc64/kernel/systbls.S
index 0fdbf3ba956e..5daee4b04dd5 100644
--- a/arch/sparc64/kernel/systbls.S
+++ b/arch/sparc64/kernel/systbls.S
@@ -23,7 +23,7 @@ sys_call_table32:
 /*10*/  .word sys_unlink, sunos_execv, sys_chdir, sys_chown16, sys32_mknod
 /*15*/	.word sys_chmod, sys_lchown16, sparc_brk, sys32_perfctr, sys32_lseek
 /*20*/	.word sys_getpid, sys_capget, sys_capset, sys_setuid16, sys_getuid16
-/*25*/	.word sys32_vmsplice, compat_sys_ptrace, sys_alarm, sys32_sigaltstack, sys32_pause
+/*25*/	.word sys32_vmsplice, compat_sys_ptrace, sys_alarm, sys32_sigaltstack, sys_pause
 /*30*/	.word compat_sys_utime, sys_lchown, sys_fchown, sys32_access, sys32_nice
 	.word sys_chown, sys_sync, sys32_kill, compat_sys_newstat, sys32_sendfile
 /*40*/	.word compat_sys_newlstat, sys_dup, sys_pipe, compat_sys_times, sys_getuid
diff --git a/arch/sparc64/kernel/time.c b/arch/sparc64/kernel/time.c
index cc16fdcf98af..80d71a5ce1e3 100644
--- a/arch/sparc64/kernel/time.c
+++ b/arch/sparc64/kernel/time.c
@@ -30,13 +30,14 @@
 #include <linux/percpu.h>
 #include <linux/miscdevice.h>
 #include <linux/rtc.h>
+#include <linux/rtc/m48t59.h>
 #include <linux/kernel_stat.h>
 #include <linux/clockchips.h>
 #include <linux/clocksource.h>
 #include <linux/of_device.h>
+#include <linux/platform_device.h>
 
 #include <asm/oplib.h>
-#include <asm/mostek.h>
 #include <asm/timer.h>
 #include <asm/irq.h>
 #include <asm/io.h>
@@ -50,18 +51,7 @@
 
 #include "entry.h"
 
-DEFINE_SPINLOCK(mostek_lock);
 DEFINE_SPINLOCK(rtc_lock);
-void __iomem *mstk48t02_regs = NULL;
-#ifdef CONFIG_PCI
-unsigned long ds1287_regs = 0UL;
-static void __iomem *bq4802_regs;
-#endif
-
-static void __iomem *mstk48t08_regs;
-static void __iomem *mstk48t59_regs;
-
-static int set_rtc_mmss(unsigned long);
 
 #define TICK_PRIV_BIT	(1UL << 63)
 #define TICKCMP_IRQ_BIT	(1UL << 63)
@@ -405,313 +395,167 @@ static unsigned long timer_ticks_per_nsec_quotient __read_mostly;
 
 int update_persistent_clock(struct timespec now)
 {
-	return set_rtc_mmss(now.tv_sec);
-}
+	struct rtc_device *rtc = rtc_class_open("rtc0");
+	int err = -1;
 
-/* Kick start a stopped clock (procedure from the Sun NVRAM/hostid FAQ). */
-static void __init kick_start_clock(void)
-{
-	void __iomem *regs = mstk48t02_regs;
-	u8 sec, tmp;
-	int i, count;
-
-	prom_printf("CLOCK: Clock was stopped. Kick start ");
-
-	spin_lock_irq(&mostek_lock);
-
-	/* Turn on the kick start bit to start the oscillator. */
-	tmp = mostek_read(regs + MOSTEK_CREG);
-	tmp |= MSTK_CREG_WRITE;
-	mostek_write(regs + MOSTEK_CREG, tmp);
-	tmp = mostek_read(regs + MOSTEK_SEC);
-	tmp &= ~MSTK_STOP;
-	mostek_write(regs + MOSTEK_SEC, tmp);
-	tmp = mostek_read(regs + MOSTEK_HOUR);
-	tmp |= MSTK_KICK_START;
-	mostek_write(regs + MOSTEK_HOUR, tmp);
-	tmp = mostek_read(regs + MOSTEK_CREG);
-	tmp &= ~MSTK_CREG_WRITE;
-	mostek_write(regs + MOSTEK_CREG, tmp);
-
-	spin_unlock_irq(&mostek_lock);
-
-	/* Delay to allow the clock oscillator to start. */
-	sec = MSTK_REG_SEC(regs);
-	for (i = 0; i < 3; i++) {
-		while (sec == MSTK_REG_SEC(regs))
-			for (count = 0; count < 100000; count++)
-				/* nothing */ ;
-		prom_printf(".");
-		sec = MSTK_REG_SEC(regs);
-	}
-	prom_printf("\n");
-
-	spin_lock_irq(&mostek_lock);
-
-	/* Turn off kick start and set a "valid" time and date. */
-	tmp = mostek_read(regs + MOSTEK_CREG);
-	tmp |= MSTK_CREG_WRITE;
-	mostek_write(regs + MOSTEK_CREG, tmp);
-	tmp = mostek_read(regs + MOSTEK_HOUR);
-	tmp &= ~MSTK_KICK_START;
-	mostek_write(regs + MOSTEK_HOUR, tmp);
-	MSTK_SET_REG_SEC(regs,0);
-	MSTK_SET_REG_MIN(regs,0);
-	MSTK_SET_REG_HOUR(regs,0);
-	MSTK_SET_REG_DOW(regs,5);
-	MSTK_SET_REG_DOM(regs,1);
-	MSTK_SET_REG_MONTH(regs,8);
-	MSTK_SET_REG_YEAR(regs,1996 - MSTK_YEAR_ZERO);
-	tmp = mostek_read(regs + MOSTEK_CREG);
-	tmp &= ~MSTK_CREG_WRITE;
-	mostek_write(regs + MOSTEK_CREG, tmp);
-
-	spin_unlock_irq(&mostek_lock);
-
-	/* Ensure the kick start bit is off. If it isn't, turn it off. */
-	while (mostek_read(regs + MOSTEK_HOUR) & MSTK_KICK_START) {
-		prom_printf("CLOCK: Kick start still on!\n");
-
-		spin_lock_irq(&mostek_lock);
-
-		tmp = mostek_read(regs + MOSTEK_CREG);
-		tmp |= MSTK_CREG_WRITE;
-		mostek_write(regs + MOSTEK_CREG, tmp);
-
-		tmp = mostek_read(regs + MOSTEK_HOUR);
-		tmp &= ~MSTK_KICK_START;
-		mostek_write(regs + MOSTEK_HOUR, tmp);
-
-		tmp = mostek_read(regs + MOSTEK_CREG);
-		tmp &= ~MSTK_CREG_WRITE;
-		mostek_write(regs + MOSTEK_CREG, tmp);
-
-		spin_unlock_irq(&mostek_lock);
+	if (rtc) {
+		err = rtc_set_mmss(rtc, now.tv_sec);
+		rtc_class_close(rtc);
 	}
 
-	prom_printf("CLOCK: Kick start procedure successful.\n");
+	return err;
 }
 
-/* Return nonzero if the clock chip battery is low. */
-static int __init has_low_battery(void)
-{
-	void __iomem *regs = mstk48t02_regs;
-	u8 data1, data2;
-
-	spin_lock_irq(&mostek_lock);
+unsigned long cmos_regs;
+EXPORT_SYMBOL(cmos_regs);
 
-	data1 = mostek_read(regs + MOSTEK_EEPROM);	/* Read some data. */
-	mostek_write(regs + MOSTEK_EEPROM, ~data1);	/* Write back the complement. */
-	data2 = mostek_read(regs + MOSTEK_EEPROM);	/* Read back the complement. */
-	mostek_write(regs + MOSTEK_EEPROM, data1);	/* Restore original value. */
+static struct resource rtc_cmos_resource;
 
-	spin_unlock_irq(&mostek_lock);
-
-	return (data1 == data2);	/* Was the write blocked? */
-}
+static struct platform_device rtc_cmos_device = {
+	.name		= "rtc_cmos",
+	.id		= -1,
+	.resource	= &rtc_cmos_resource,
+	.num_resources	= 1,
+};
 
-static void __init mostek_set_system_time(void __iomem *mregs)
+static int __devinit rtc_probe(struct of_device *op, const struct of_device_id *match)
 {
-	unsigned int year, mon, day, hour, min, sec;
-	u8 tmp;
-
-	spin_lock_irq(&mostek_lock);
+	struct resource *r;
 
-	/* Traditional Mostek chip. */
-	tmp = mostek_read(mregs + MOSTEK_CREG);
-	tmp |= MSTK_CREG_READ;
-	mostek_write(mregs + MOSTEK_CREG, tmp);
+	printk(KERN_INFO "%s: RTC regs at 0x%lx\n",
+	       op->node->full_name, op->resource[0].start);
 
-	sec = MSTK_REG_SEC(mregs);
-	min = MSTK_REG_MIN(mregs);
-	hour = MSTK_REG_HOUR(mregs);
-	day = MSTK_REG_DOM(mregs);
-	mon = MSTK_REG_MONTH(mregs);
-	year = MSTK_CVT_YEAR( MSTK_REG_YEAR(mregs) );
-
-	xtime.tv_sec = mktime(year, mon, day, hour, min, sec);
-	xtime.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ);
-	set_normalized_timespec(&wall_to_monotonic,
- 	                        -xtime.tv_sec, -xtime.tv_nsec);
+	/* The CMOS RTC driver only accepts IORESOURCE_IO, so cons
+	 * up a fake resource so that the probe works for all cases.
+	 * When the RTC is behind an ISA bus it will have IORESOURCE_IO
+	 * already, whereas when it's behind EBUS is will be IORESOURCE_MEM.
+	 */
 
-	tmp = mostek_read(mregs + MOSTEK_CREG);
-	tmp &= ~MSTK_CREG_READ;
-	mostek_write(mregs + MOSTEK_CREG, tmp);
+	r = &rtc_cmos_resource;
+	r->flags = IORESOURCE_IO;
+	r->name = op->resource[0].name;
+	r->start = op->resource[0].start;
+	r->end = op->resource[0].end;
 
-	spin_unlock_irq(&mostek_lock);
+	cmos_regs = op->resource[0].start;
+	return platform_device_register(&rtc_cmos_device);
 }
 
-/* Probe for the real time clock chip. */
-static void __init set_system_time(void)
-{
-	unsigned int year, mon, day, hour, min, sec;
-	void __iomem *mregs = mstk48t02_regs;
-#ifdef CONFIG_PCI
-	unsigned long dregs = ds1287_regs;
-	void __iomem *bregs = bq4802_regs;
-#else
-	unsigned long dregs = 0UL;
-	void __iomem *bregs = 0UL;
-#endif
-
-	if (!mregs && !dregs && !bregs) {
-		prom_printf("Something wrong, clock regs not mapped yet.\n");
-		prom_halt();
-	}		
-
-	if (mregs) {
-		mostek_set_system_time(mregs);
-		return;
-	}
-
-	if (bregs) {
-		unsigned char val = readb(bregs + 0x0e);
-		unsigned int century;
+static struct of_device_id __initdata rtc_match[] = {
+	{
+		.name = "rtc",
+		.compatible = "m5819",
+	},
+	{
+		.name = "rtc",
+		.compatible = "isa-m5819p",
+	},
+	{
+		.name = "rtc",
+		.compatible = "isa-m5823p",
+	},
+	{
+		.name = "rtc",
+		.compatible = "ds1287",
+	},
+	{},
+};
 
-		/* BQ4802 RTC chip. */
+static struct of_platform_driver rtc_driver = {
+	.match_table	= rtc_match,
+	.probe		= rtc_probe,
+	.driver		= {
+		.name	= "rtc",
+	},
+};
 
-		writeb(val | 0x08, bregs + 0x0e);
+static struct platform_device rtc_bq4802_device = {
+	.name		= "rtc-bq4802",
+	.id		= -1,
+	.num_resources	= 1,
+};
 
-		sec  = readb(bregs + 0x00);
-		min  = readb(bregs + 0x02);
-		hour = readb(bregs + 0x04);
-		day  = readb(bregs + 0x06);
-		mon  = readb(bregs + 0x09);
-		year = readb(bregs + 0x0a);
-		century = readb(bregs + 0x0f);
+static int __devinit bq4802_probe(struct of_device *op, const struct of_device_id *match)
+{
 
-		writeb(val, bregs + 0x0e);
+	printk(KERN_INFO "%s: BQ4802 regs at 0x%lx\n",
+	       op->node->full_name, op->resource[0].start);
 
-		BCD_TO_BIN(sec);
-		BCD_TO_BIN(min);
-		BCD_TO_BIN(hour);
-		BCD_TO_BIN(day);
-		BCD_TO_BIN(mon);
-		BCD_TO_BIN(year);
-		BCD_TO_BIN(century);
+	rtc_bq4802_device.resource = &op->resource[0];
+	return platform_device_register(&rtc_bq4802_device);
+}
 
-		year += (century * 100);
-	} else {
-		/* Dallas 12887 RTC chip. */
-
-		do {
-			sec  = CMOS_READ(RTC_SECONDS);
-			min  = CMOS_READ(RTC_MINUTES);
-			hour = CMOS_READ(RTC_HOURS);
-			day  = CMOS_READ(RTC_DAY_OF_MONTH);
-			mon  = CMOS_READ(RTC_MONTH);
-			year = CMOS_READ(RTC_YEAR);
-		} while (sec != CMOS_READ(RTC_SECONDS));
-
-		if (!(CMOS_READ(RTC_CONTROL) & RTC_DM_BINARY) || RTC_ALWAYS_BCD) {
-			BCD_TO_BIN(sec);
-			BCD_TO_BIN(min);
-			BCD_TO_BIN(hour);
-			BCD_TO_BIN(day);
-			BCD_TO_BIN(mon);
-			BCD_TO_BIN(year);
-		}
-		if ((year += 1900) < 1970)
-			year += 100;
-	}
+static struct of_device_id __initdata bq4802_match[] = {
+	{
+		.name = "rtc",
+		.compatible = "bq4802",
+	},
+};
 
-	xtime.tv_sec = mktime(year, mon, day, hour, min, sec);
-	xtime.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ);
-	set_normalized_timespec(&wall_to_monotonic,
- 	                        -xtime.tv_sec, -xtime.tv_nsec);
-}
+static struct of_platform_driver bq4802_driver = {
+	.match_table	= bq4802_match,
+	.probe		= bq4802_probe,
+	.driver		= {
+		.name	= "bq4802",
+	},
+};
 
-/* davem suggests we keep this within the 4M locked kernel image */
-static u32 starfire_get_time(void)
+static unsigned char mostek_read_byte(struct device *dev, u32 ofs)
 {
-	static char obp_gettod[32];
-	static u32 unix_tod;
+	struct platform_device *pdev = to_platform_device(dev);
+	struct m48t59_plat_data *pdata = pdev->dev.platform_data;
+	void __iomem *regs;
+	unsigned char val;
 
-	sprintf(obp_gettod, "h# %08x unix-gettod",
-		(unsigned int) (long) &unix_tod);
-	prom_feval(obp_gettod);
+	regs = (void __iomem *) pdev->resource[0].start;
+	val = readb(regs + ofs);
 
-	return unix_tod;
+	/* the year 0 is 1968 */
+	if (ofs == pdata->offset + M48T59_YEAR) {
+		val += 0x68;
+		if ((val & 0xf) > 9)
+			val += 6;
+	}
+	return val;
 }
 
-static int starfire_set_time(u32 val)
+static void mostek_write_byte(struct device *dev, u32 ofs, u8 val)
 {
-	/* Do nothing, time is set using the service processor
-	 * console on this platform.
-	 */
-	return 0;
-}
+	struct platform_device *pdev = to_platform_device(dev);
+	struct m48t59_plat_data *pdata = pdev->dev.platform_data;
+	void __iomem *regs;
 
-static u32 hypervisor_get_time(void)
-{
-	unsigned long ret, time;
-	int retries = 10000;
-
-retry:
-	ret = sun4v_tod_get(&time);
-	if (ret == HV_EOK)
-		return time;
-	if (ret == HV_EWOULDBLOCK) {
-		if (--retries > 0) {
-			udelay(100);
-			goto retry;
-		}
-		printk(KERN_WARNING "SUN4V: tod_get() timed out.\n");
-		return 0;
+	regs = (void __iomem *) pdev->resource[0].start;
+	if (ofs == pdata->offset + M48T59_YEAR) {
+		if (val < 0x68)
+			val += 0x32;
+		else
+			val -= 0x68;
+		if ((val & 0xf) > 9)
+			val += 6;
+		if ((val & 0xf0) > 0x9A)
+			val += 0x60;
 	}
-	printk(KERN_WARNING "SUN4V: tod_get() not supported.\n");
-	return 0;
+	writeb(val, regs + ofs);
 }
 
-static int hypervisor_set_time(u32 secs)
-{
-	unsigned long ret;
-	int retries = 10000;
-
-retry:
-	ret = sun4v_tod_set(secs);
-	if (ret == HV_EOK)
-		return 0;
-	if (ret == HV_EWOULDBLOCK) {
-		if (--retries > 0) {
-			udelay(100);
-			goto retry;
-		}
-		printk(KERN_WARNING "SUN4V: tod_set() timed out.\n");
-		return -EAGAIN;
-	}
-	printk(KERN_WARNING "SUN4V: tod_set() not supported.\n");
-	return -EOPNOTSUPP;
-}
+static struct m48t59_plat_data m48t59_data = {
+	.read_byte	= mostek_read_byte,
+	.write_byte	= mostek_write_byte,
+};
 
-static int __init clock_model_matches(const char *model)
-{
-	if (strcmp(model, "mk48t02") &&
-	    strcmp(model, "mk48t08") &&
-	    strcmp(model, "mk48t59") &&
-	    strcmp(model, "m5819") &&
-	    strcmp(model, "m5819p") &&
-	    strcmp(model, "m5823") &&
-	    strcmp(model, "ds1287") &&
-	    strcmp(model, "bq4802"))
-		return 0;
-
-	return 1;
-}
+static struct platform_device m48t59_rtc = {
+	.name		= "rtc-m48t59",
+	.id		= 0,
+	.num_resources	= 1,
+	.dev	= {
+		.platform_data = &m48t59_data,
+	},
+};
 
-static int __devinit clock_probe(struct of_device *op, const struct of_device_id *match)
+static int __devinit mostek_probe(struct of_device *op, const struct of_device_id *match)
 {
 	struct device_node *dp = op->node;
-	const char *model = of_get_property(dp, "model", NULL);
-	const char *compat = of_get_property(dp, "compatible", NULL);
-	unsigned long size, flags;
-	void __iomem *regs;
-
-	if (!model)
-		model = compat;
-
-	if (!model || !clock_model_matches(model))
-		return -ENODEV;
 
 	/* On an Enterprise system there can be multiple mostek clocks.
 	 * We should only match the one that is on the central FHC bus.
@@ -720,88 +564,51 @@ static int __devinit clock_probe(struct of_device *op, const struct of_device_id
 	    strcmp(dp->parent->parent->name, "central") != 0)
 		return -ENODEV;
 
-	size = (op->resource[0].end - op->resource[0].start) + 1;
-	regs = of_ioremap(&op->resource[0], 0, size, "clock");
-	if (!regs)
-		return -ENOMEM;
-
-#ifdef CONFIG_PCI
-	if (!strcmp(model, "ds1287") ||
-	    !strcmp(model, "m5819") ||
-	    !strcmp(model, "m5819p") ||
-	    !strcmp(model, "m5823")) {
-		ds1287_regs = (unsigned long) regs;
-	} else if (!strcmp(model, "bq4802")) {
-		bq4802_regs = regs;
-	} else
-#endif
-	if (model[5] == '0' && model[6] == '2') {
-		mstk48t02_regs = regs;
-	} else if(model[5] == '0' && model[6] == '8') {
-		mstk48t08_regs = regs;
-		mstk48t02_regs = mstk48t08_regs + MOSTEK_48T08_48T02;
-	} else {
-		mstk48t59_regs = regs;
-		mstk48t02_regs = mstk48t59_regs + MOSTEK_48T59_48T02;
-	}
-
-	printk(KERN_INFO "%s: Clock regs at %p\n", dp->full_name, regs);
-
-	local_irq_save(flags);
-
-	if (mstk48t02_regs != NULL) {
-		/* Report a low battery voltage condition. */
-		if (has_low_battery())
-			prom_printf("NVRAM: Low battery voltage!\n");
-
-		/* Kick start the clock if it is completely stopped. */
-		if (mostek_read(mstk48t02_regs + MOSTEK_SEC) & MSTK_STOP)
-			kick_start_clock();
-	}
-
-	set_system_time();
-	
-	local_irq_restore(flags);
+	printk(KERN_INFO "%s: Mostek regs at 0x%lx\n",
+	       dp->full_name, op->resource[0].start);
 
-	return 0;
+	m48t59_rtc.resource = &op->resource[0];
+	return platform_device_register(&m48t59_rtc);
 }
 
-static struct of_device_id clock_match[] = {
+static struct of_device_id __initdata mostek_match[] = {
 	{
 		.name = "eeprom",
 	},
-	{
-		.name = "rtc",
-	},
 	{},
 };
 
-static struct of_platform_driver clock_driver = {
-	.match_table	= clock_match,
-	.probe		= clock_probe,
+static struct of_platform_driver mostek_driver = {
+	.match_table	= mostek_match,
+	.probe		= mostek_probe,
 	.driver		= {
-		.name	= "clock",
+		.name	= "mostek",
 	},
 };
 
+static struct platform_device rtc_sun4v_device = {
+	.name		= "rtc-sun4v",
+	.id		= -1,
+};
+
+static struct platform_device rtc_starfire_device = {
+	.name		= "rtc-starfire",
+	.id		= -1,
+};
+
 static int __init clock_init(void)
 {
-	if (this_is_starfire) {
-		xtime.tv_sec = starfire_get_time();
-		xtime.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ);
-		set_normalized_timespec(&wall_to_monotonic,
-		                        -xtime.tv_sec, -xtime.tv_nsec);
-		return 0;
-	}
-	if (tlb_type == hypervisor) {
-		xtime.tv_sec = hypervisor_get_time();
-		xtime.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ);
-		set_normalized_timespec(&wall_to_monotonic,
-		                        -xtime.tv_sec, -xtime.tv_nsec);
-		return 0;
-	}
+	if (this_is_starfire)
+		return platform_device_register(&rtc_starfire_device);
+
+	if (tlb_type == hypervisor)
+		return platform_device_register(&rtc_sun4v_device);
+
+	(void) of_register_driver(&rtc_driver, &of_platform_bus_type);
+	(void) of_register_driver(&mostek_driver, &of_platform_bus_type);
+	(void) of_register_driver(&bq4802_driver, &of_platform_bus_type);
 
-	return of_register_driver(&clock_driver, &of_platform_bus_type);
+	return 0;
 }
 
 /* Must be after subsys_initcall() so that busses are probed.  Must
@@ -814,7 +621,7 @@ fs_initcall(clock_init);
 static unsigned long sparc64_init_timers(void)
 {
 	struct device_node *dp;
-	unsigned long clock;
+	unsigned long freq;
 
 	dp = of_find_node_by_path("/");
 	if (tlb_type == spitfire) {
@@ -827,17 +634,17 @@ static unsigned long sparc64_init_timers(void)
 		if (manuf == 0x17 && impl == 0x13) {
 			/* Hummingbird, aka Ultra-IIe */
 			tick_ops = &hbtick_operations;
-			clock = of_getintprop_default(dp, "stick-frequency", 0);
+			freq = of_getintprop_default(dp, "stick-frequency", 0);
 		} else {
 			tick_ops = &tick_operations;
-			clock = local_cpu_data().clock_tick;
+			freq = local_cpu_data().clock_tick;
 		}
 	} else {
 		tick_ops = &stick_operations;
-		clock = of_getintprop_default(dp, "stick-frequency", 0);
+		freq = of_getintprop_default(dp, "stick-frequency", 0);
 	}
 
-	return clock;
+	return freq;
 }
 
 struct freq_table {
@@ -1029,16 +836,16 @@ EXPORT_SYMBOL(udelay);
 
 void __init time_init(void)
 {
-	unsigned long clock = sparc64_init_timers();
+	unsigned long freq = sparc64_init_timers();
 
-	tb_ticks_per_usec = clock / USEC_PER_SEC;
+	tb_ticks_per_usec = freq / USEC_PER_SEC;
 
 	timer_ticks_per_nsec_quotient =
-		clocksource_hz2mult(clock, SPARC64_NSEC_PER_CYC_SHIFT);
+		clocksource_hz2mult(freq, SPARC64_NSEC_PER_CYC_SHIFT);
 
 	clocksource_tick.name = tick_ops->name;
 	clocksource_tick.mult =
-		clocksource_hz2mult(clock,
+		clocksource_hz2mult(freq,
 				    clocksource_tick.shift);
 	clocksource_tick.read = tick_ops->get_tick;
 
@@ -1049,7 +856,7 @@ void __init time_init(void)
 
 	sparc64_clockevent.name = tick_ops->name;
 
-	setup_clockevent_multiplier(clock);
+	setup_clockevent_multiplier(freq);
 
 	sparc64_clockevent.max_delta_ns =
 		clockevent_delta2ns(0x7fffffffffffffffUL, &sparc64_clockevent);
@@ -1070,672 +877,8 @@ unsigned long long sched_clock(void)
 		>> SPARC64_NSEC_PER_CYC_SHIFT;
 }
 
-static int set_rtc_mmss(unsigned long nowtime)
-{
-	int real_seconds, real_minutes, chip_minutes;
-	void __iomem *mregs = mstk48t02_regs;
-#ifdef CONFIG_PCI
-	unsigned long dregs = ds1287_regs;
-	void __iomem *bregs = bq4802_regs;
-#else
-	unsigned long dregs = 0UL;
-	void __iomem *bregs = 0UL;
-#endif
-	unsigned long flags;
-	u8 tmp;
-
-	/* 
-	 * Not having a register set can lead to trouble.
-	 * Also starfire doesn't have a tod clock.
-	 */
-	if (!mregs && !dregs && !bregs)
-		return -1;
-
-	if (mregs) {
-		spin_lock_irqsave(&mostek_lock, flags);
-
-		/* Read the current RTC minutes. */
-		tmp = mostek_read(mregs + MOSTEK_CREG);
-		tmp |= MSTK_CREG_READ;
-		mostek_write(mregs + MOSTEK_CREG, tmp);
-
-		chip_minutes = MSTK_REG_MIN(mregs);
-
-		tmp = mostek_read(mregs + MOSTEK_CREG);
-		tmp &= ~MSTK_CREG_READ;
-		mostek_write(mregs + MOSTEK_CREG, tmp);
-
-		/*
-		 * since we're only adjusting minutes and seconds,
-		 * don't interfere with hour overflow. This avoids
-		 * messing with unknown time zones but requires your
-		 * RTC not to be off by more than 15 minutes
-		 */
-		real_seconds = nowtime % 60;
-		real_minutes = nowtime / 60;
-		if (((abs(real_minutes - chip_minutes) + 15)/30) & 1)
-			real_minutes += 30;	/* correct for half hour time zone */
-		real_minutes %= 60;
-
-		if (abs(real_minutes - chip_minutes) < 30) {
-			tmp = mostek_read(mregs + MOSTEK_CREG);
-			tmp |= MSTK_CREG_WRITE;
-			mostek_write(mregs + MOSTEK_CREG, tmp);
-
-			MSTK_SET_REG_SEC(mregs,real_seconds);
-			MSTK_SET_REG_MIN(mregs,real_minutes);
-
-			tmp = mostek_read(mregs + MOSTEK_CREG);
-			tmp &= ~MSTK_CREG_WRITE;
-			mostek_write(mregs + MOSTEK_CREG, tmp);
-
-			spin_unlock_irqrestore(&mostek_lock, flags);
-
-			return 0;
-		} else {
-			spin_unlock_irqrestore(&mostek_lock, flags);
-
-			return -1;
-		}
-	} else if (bregs) {
-		int retval = 0;
-		unsigned char val = readb(bregs + 0x0e);
-
-		/* BQ4802 RTC chip. */
-
-		writeb(val | 0x08, bregs + 0x0e);
-
-		chip_minutes = readb(bregs + 0x02);
-		BCD_TO_BIN(chip_minutes);
-		real_seconds = nowtime % 60;
-		real_minutes = nowtime / 60;
-		if (((abs(real_minutes - chip_minutes) + 15)/30) & 1)
-			real_minutes += 30;
-		real_minutes %= 60;
-
-		if (abs(real_minutes - chip_minutes) < 30) {
-			BIN_TO_BCD(real_seconds);
-			BIN_TO_BCD(real_minutes);
-			writeb(real_seconds, bregs + 0x00);
-			writeb(real_minutes, bregs + 0x02);
-		} else {
-			printk(KERN_WARNING
-			       "set_rtc_mmss: can't update from %d to %d\n",
-			       chip_minutes, real_minutes);
-			retval = -1;
-		}
-
-		writeb(val, bregs + 0x0e);
-
-		return retval;
-	} else {
-		int retval = 0;
-		unsigned char save_control, save_freq_select;
-
-		/* Stolen from arch/i386/kernel/time.c, see there for
-		 * credits and descriptive comments.
-		 */
-		spin_lock_irqsave(&rtc_lock, flags);
-		save_control = CMOS_READ(RTC_CONTROL); /* tell the clock it's being set */
-		CMOS_WRITE((save_control|RTC_SET), RTC_CONTROL);
-
-		save_freq_select = CMOS_READ(RTC_FREQ_SELECT); /* stop and reset prescaler */
-		CMOS_WRITE((save_freq_select|RTC_DIV_RESET2), RTC_FREQ_SELECT);
-
-		chip_minutes = CMOS_READ(RTC_MINUTES);
-		if (!(save_control & RTC_DM_BINARY) || RTC_ALWAYS_BCD)
-			BCD_TO_BIN(chip_minutes);
-		real_seconds = nowtime % 60;
-		real_minutes = nowtime / 60;
-		if (((abs(real_minutes - chip_minutes) + 15)/30) & 1)
-			real_minutes += 30;
-		real_minutes %= 60;
-
-		if (abs(real_minutes - chip_minutes) < 30) {
-			if (!(save_control & RTC_DM_BINARY) || RTC_ALWAYS_BCD) {
-				BIN_TO_BCD(real_seconds);
-				BIN_TO_BCD(real_minutes);
-			}
-			CMOS_WRITE(real_seconds,RTC_SECONDS);
-			CMOS_WRITE(real_minutes,RTC_MINUTES);
-		} else {
-			printk(KERN_WARNING
-			       "set_rtc_mmss: can't update from %d to %d\n",
-			       chip_minutes, real_minutes);
-			retval = -1;
-		}
-
-		CMOS_WRITE(save_control, RTC_CONTROL);
-		CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT);
-		spin_unlock_irqrestore(&rtc_lock, flags);
-
-		return retval;
-	}
-}
-
-#define RTC_IS_OPEN		0x01	/* means /dev/rtc is in use	*/
-static unsigned char mini_rtc_status;	/* bitmapped status byte.	*/
-
-#define FEBRUARY	2
-#define	STARTOFTIME	1970
-#define SECDAY		86400L
-#define SECYR		(SECDAY * 365)
-#define	leapyear(year)		((year) % 4 == 0 && \
-				 ((year) % 100 != 0 || (year) % 400 == 0))
-#define	days_in_year(a) 	(leapyear(a) ? 366 : 365)
-#define	days_in_month(a) 	(month_days[(a) - 1])
-
-static int month_days[12] = {
-	31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31
-};
-
-/*
- * This only works for the Gregorian calendar - i.e. after 1752 (in the UK)
- */
-static void GregorianDay(struct rtc_time * tm)
-{
-	int leapsToDate;
-	int lastYear;
-	int day;
-	int MonthOffset[] = { 0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334 };
-
-	lastYear = tm->tm_year - 1;
-
-	/*
-	 * Number of leap corrections to apply up to end of last year
-	 */
-	leapsToDate = lastYear / 4 - lastYear / 100 + lastYear / 400;
-
-	/*
-	 * This year is a leap year if it is divisible by 4 except when it is
-	 * divisible by 100 unless it is divisible by 400
-	 *
-	 * e.g. 1904 was a leap year, 1900 was not, 1996 is, and 2000 was
-	 */
-	day = tm->tm_mon > 2 && leapyear(tm->tm_year);
-
-	day += lastYear*365 + leapsToDate + MonthOffset[tm->tm_mon-1] +
-		   tm->tm_mday;
-
-	tm->tm_wday = day % 7;
-}
-
-static void to_tm(int tim, struct rtc_time *tm)
-{
-	register int    i;
-	register long   hms, day;
-
-	day = tim / SECDAY;
-	hms = tim % SECDAY;
-
-	/* Hours, minutes, seconds are easy */
-	tm->tm_hour = hms / 3600;
-	tm->tm_min = (hms % 3600) / 60;
-	tm->tm_sec = (hms % 3600) % 60;
-
-	/* Number of years in days */
-	for (i = STARTOFTIME; day >= days_in_year(i); i++)
-		day -= days_in_year(i);
-	tm->tm_year = i;
-
-	/* Number of months in days left */
-	if (leapyear(tm->tm_year))
-		days_in_month(FEBRUARY) = 29;
-	for (i = 1; day >= days_in_month(i); i++)
-		day -= days_in_month(i);
-	days_in_month(FEBRUARY) = 28;
-	tm->tm_mon = i;
-
-	/* Days are what is left over (+1) from all that. */
-	tm->tm_mday = day + 1;
-
-	/*
-	 * Determine the day of week
-	 */
-	GregorianDay(tm);
-}
-
-/* Both Starfire and SUN4V give us seconds since Jan 1st, 1970,
- * aka Unix time.  So we have to convert to/from rtc_time.
- */
-static void starfire_get_rtc_time(struct rtc_time *time)
-{
-	u32 seconds = starfire_get_time();
-
-	to_tm(seconds, time);
-	time->tm_year -= 1900;
-	time->tm_mon -= 1;
-}
-
-static int starfire_set_rtc_time(struct rtc_time *time)
-{
-	u32 seconds = mktime(time->tm_year + 1900, time->tm_mon + 1,
-			     time->tm_mday, time->tm_hour,
-			     time->tm_min, time->tm_sec);
-
-	return starfire_set_time(seconds);
-}
-
-static void hypervisor_get_rtc_time(struct rtc_time *time)
-{
-	u32 seconds = hypervisor_get_time();
-
-	to_tm(seconds, time);
-	time->tm_year -= 1900;
-	time->tm_mon -= 1;
-}
-
-static int hypervisor_set_rtc_time(struct rtc_time *time)
-{
-	u32 seconds = mktime(time->tm_year + 1900, time->tm_mon + 1,
-			     time->tm_mday, time->tm_hour,
-			     time->tm_min, time->tm_sec);
-
-	return hypervisor_set_time(seconds);
-}
-
-#ifdef CONFIG_PCI
-static void bq4802_get_rtc_time(struct rtc_time *time)
-{
-	unsigned char val = readb(bq4802_regs + 0x0e);
-	unsigned int century;
-
-	writeb(val | 0x08, bq4802_regs + 0x0e);
-
-	time->tm_sec = readb(bq4802_regs + 0x00);
-	time->tm_min = readb(bq4802_regs + 0x02);
-	time->tm_hour = readb(bq4802_regs + 0x04);
-	time->tm_mday = readb(bq4802_regs + 0x06);
-	time->tm_mon = readb(bq4802_regs + 0x09);
-	time->tm_year = readb(bq4802_regs + 0x0a);
-	time->tm_wday = readb(bq4802_regs + 0x08);
-	century = readb(bq4802_regs + 0x0f);
-
-	writeb(val, bq4802_regs + 0x0e);
-
-	BCD_TO_BIN(time->tm_sec);
-	BCD_TO_BIN(time->tm_min);
-	BCD_TO_BIN(time->tm_hour);
-	BCD_TO_BIN(time->tm_mday);
-	BCD_TO_BIN(time->tm_mon);
-	BCD_TO_BIN(time->tm_year);
-	BCD_TO_BIN(time->tm_wday);
-	BCD_TO_BIN(century);
-
-	time->tm_year += (century * 100);
-	time->tm_year -= 1900;
-
-	time->tm_mon--;
-}
-
-static int bq4802_set_rtc_time(struct rtc_time *time)
-{
-	unsigned char val = readb(bq4802_regs + 0x0e);
-	unsigned char sec, min, hrs, day, mon, yrs, century;
-	unsigned int year;
-
-	year = time->tm_year + 1900;
-	century = year / 100;
-	yrs = year % 100;
-
-	mon = time->tm_mon + 1;   /* tm_mon starts at zero */
-	day = time->tm_mday;
-	hrs = time->tm_hour;
-	min = time->tm_min;
-	sec = time->tm_sec;
-
-	BIN_TO_BCD(sec);
-	BIN_TO_BCD(min);
-	BIN_TO_BCD(hrs);
-	BIN_TO_BCD(day);
-	BIN_TO_BCD(mon);
-	BIN_TO_BCD(yrs);
-	BIN_TO_BCD(century);
-
-	writeb(val | 0x08, bq4802_regs + 0x0e);
-
-	writeb(sec, bq4802_regs + 0x00);
-	writeb(min, bq4802_regs + 0x02);
-	writeb(hrs, bq4802_regs + 0x04);
-	writeb(day, bq4802_regs + 0x06);
-	writeb(mon, bq4802_regs + 0x09);
-	writeb(yrs, bq4802_regs + 0x0a);
-	writeb(century, bq4802_regs + 0x0f);
-
-	writeb(val, bq4802_regs + 0x0e);
-
-	return 0;
-}
-
-static void cmos_get_rtc_time(struct rtc_time *rtc_tm)
-{
-	unsigned char ctrl;
-
-	rtc_tm->tm_sec = CMOS_READ(RTC_SECONDS);
-	rtc_tm->tm_min = CMOS_READ(RTC_MINUTES);
-	rtc_tm->tm_hour = CMOS_READ(RTC_HOURS);
-	rtc_tm->tm_mday = CMOS_READ(RTC_DAY_OF_MONTH);
-	rtc_tm->tm_mon = CMOS_READ(RTC_MONTH);
-	rtc_tm->tm_year = CMOS_READ(RTC_YEAR);
-	rtc_tm->tm_wday = CMOS_READ(RTC_DAY_OF_WEEK);
-
-	ctrl = CMOS_READ(RTC_CONTROL);
-	if (!(ctrl & RTC_DM_BINARY) || RTC_ALWAYS_BCD) {
-		BCD_TO_BIN(rtc_tm->tm_sec);
-		BCD_TO_BIN(rtc_tm->tm_min);
-		BCD_TO_BIN(rtc_tm->tm_hour);
-		BCD_TO_BIN(rtc_tm->tm_mday);
-		BCD_TO_BIN(rtc_tm->tm_mon);
-		BCD_TO_BIN(rtc_tm->tm_year);
-		BCD_TO_BIN(rtc_tm->tm_wday);
-	}
-
-	if (rtc_tm->tm_year <= 69)
-		rtc_tm->tm_year += 100;
-
-	rtc_tm->tm_mon--;
-}
-
-static int cmos_set_rtc_time(struct rtc_time *rtc_tm)
-{
-	unsigned char mon, day, hrs, min, sec;
-	unsigned char save_control, save_freq_select;
-	unsigned int yrs;
-
-	yrs = rtc_tm->tm_year;
-	mon = rtc_tm->tm_mon + 1;
-	day = rtc_tm->tm_mday;
-	hrs = rtc_tm->tm_hour;
-	min = rtc_tm->tm_min;
-	sec = rtc_tm->tm_sec;
-
-	if (yrs >= 100)
-		yrs -= 100;
-
-	if (!(CMOS_READ(RTC_CONTROL) & RTC_DM_BINARY) || RTC_ALWAYS_BCD) {
-		BIN_TO_BCD(sec);
-		BIN_TO_BCD(min);
-		BIN_TO_BCD(hrs);
-		BIN_TO_BCD(day);
-		BIN_TO_BCD(mon);
-		BIN_TO_BCD(yrs);
-	}
-
-	save_control = CMOS_READ(RTC_CONTROL);
-	CMOS_WRITE((save_control|RTC_SET), RTC_CONTROL);
-	save_freq_select = CMOS_READ(RTC_FREQ_SELECT);
-	CMOS_WRITE((save_freq_select|RTC_DIV_RESET2), RTC_FREQ_SELECT);
-
-	CMOS_WRITE(yrs, RTC_YEAR);
-	CMOS_WRITE(mon, RTC_MONTH);
-	CMOS_WRITE(day, RTC_DAY_OF_MONTH);
-	CMOS_WRITE(hrs, RTC_HOURS);
-	CMOS_WRITE(min, RTC_MINUTES);
-	CMOS_WRITE(sec, RTC_SECONDS);
-
-	CMOS_WRITE(save_control, RTC_CONTROL);
-	CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT);
-
-	return 0;
-}
-#endif /* CONFIG_PCI */
-
-static void mostek_get_rtc_time(struct rtc_time *rtc_tm)
-{
-	void __iomem *regs = mstk48t02_regs;
-	u8 tmp;
-
-	spin_lock_irq(&mostek_lock);
-
-	tmp = mostek_read(regs + MOSTEK_CREG);
-	tmp |= MSTK_CREG_READ;
-	mostek_write(regs + MOSTEK_CREG, tmp);
-
-	rtc_tm->tm_sec = MSTK_REG_SEC(regs);
-	rtc_tm->tm_min = MSTK_REG_MIN(regs);
-	rtc_tm->tm_hour = MSTK_REG_HOUR(regs);
-	rtc_tm->tm_mday = MSTK_REG_DOM(regs);
-	rtc_tm->tm_mon = MSTK_REG_MONTH(regs);
-	rtc_tm->tm_year = MSTK_CVT_YEAR( MSTK_REG_YEAR(regs) );
-	rtc_tm->tm_wday = MSTK_REG_DOW(regs);
-
-	tmp = mostek_read(regs + MOSTEK_CREG);
-	tmp &= ~MSTK_CREG_READ;
-	mostek_write(regs + MOSTEK_CREG, tmp);
-
-	spin_unlock_irq(&mostek_lock);
-
-	rtc_tm->tm_mon--;
-	rtc_tm->tm_wday--;
-	rtc_tm->tm_year -= 1900;
-}
-
-static int mostek_set_rtc_time(struct rtc_time *rtc_tm)
-{
-	unsigned char mon, day, hrs, min, sec, wday;
-	void __iomem *regs = mstk48t02_regs;
-	unsigned int yrs;
-	u8 tmp;
-
-	yrs = rtc_tm->tm_year + 1900;
-	mon = rtc_tm->tm_mon + 1;
-	day = rtc_tm->tm_mday;
-	wday = rtc_tm->tm_wday + 1;
-	hrs = rtc_tm->tm_hour;
-	min = rtc_tm->tm_min;
-	sec = rtc_tm->tm_sec;
-
-	spin_lock_irq(&mostek_lock);
-
-	tmp = mostek_read(regs + MOSTEK_CREG);
-	tmp |= MSTK_CREG_WRITE;
-	mostek_write(regs + MOSTEK_CREG, tmp);
-
-	MSTK_SET_REG_SEC(regs, sec);
-	MSTK_SET_REG_MIN(regs, min);
-	MSTK_SET_REG_HOUR(regs, hrs);
-	MSTK_SET_REG_DOW(regs, wday);
-	MSTK_SET_REG_DOM(regs, day);
-	MSTK_SET_REG_MONTH(regs, mon);
-	MSTK_SET_REG_YEAR(regs, yrs - MSTK_YEAR_ZERO);
-
-	tmp = mostek_read(regs + MOSTEK_CREG);
-	tmp &= ~MSTK_CREG_WRITE;
-	mostek_write(regs + MOSTEK_CREG, tmp);
-
-	spin_unlock_irq(&mostek_lock);
-
-	return 0;
-}
-
-struct mini_rtc_ops {
-	void (*get_rtc_time)(struct rtc_time *);
-	int (*set_rtc_time)(struct rtc_time *);
-};
-
-static struct mini_rtc_ops starfire_rtc_ops = {
-	.get_rtc_time = starfire_get_rtc_time,
-	.set_rtc_time = starfire_set_rtc_time,
-};
-
-static struct mini_rtc_ops hypervisor_rtc_ops = {
-	.get_rtc_time = hypervisor_get_rtc_time,
-	.set_rtc_time = hypervisor_set_rtc_time,
-};
-
-#ifdef CONFIG_PCI
-static struct mini_rtc_ops bq4802_rtc_ops = {
-	.get_rtc_time = bq4802_get_rtc_time,
-	.set_rtc_time = bq4802_set_rtc_time,
-};
-
-static struct mini_rtc_ops cmos_rtc_ops = {
-	.get_rtc_time = cmos_get_rtc_time,
-	.set_rtc_time = cmos_set_rtc_time,
-};
-#endif /* CONFIG_PCI */
-
-static struct mini_rtc_ops mostek_rtc_ops = {
-	.get_rtc_time = mostek_get_rtc_time,
-	.set_rtc_time = mostek_set_rtc_time,
-};
-
-static struct mini_rtc_ops *mini_rtc_ops;
-
-static inline void mini_get_rtc_time(struct rtc_time *time)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&rtc_lock, flags);
-	mini_rtc_ops->get_rtc_time(time);
-	spin_unlock_irqrestore(&rtc_lock, flags);
-}
-
-static inline int mini_set_rtc_time(struct rtc_time *time)
-{
-	unsigned long flags;
-	int err;
-
-	spin_lock_irqsave(&rtc_lock, flags);
-	err = mini_rtc_ops->set_rtc_time(time);
-	spin_unlock_irqrestore(&rtc_lock, flags);
-
-	return err;
-}
-
-static int mini_rtc_ioctl(struct inode *inode, struct file *file,
-			  unsigned int cmd, unsigned long arg)
-{
-	struct rtc_time wtime;
-	void __user *argp = (void __user *)arg;
-
-	switch (cmd) {
-
-	case RTC_PLL_GET:
-		return -EINVAL;
-
-	case RTC_PLL_SET:
-		return -EINVAL;
-
-	case RTC_UIE_OFF:	/* disable ints from RTC updates.	*/
-		return 0;
-
-	case RTC_UIE_ON:	/* enable ints for RTC updates.	*/
-	        return -EINVAL;
-
-	case RTC_RD_TIME:	/* Read the time/date from RTC	*/
-		/* this doesn't get week-day, who cares */
-		memset(&wtime, 0, sizeof(wtime));
-		mini_get_rtc_time(&wtime);
-
-		return copy_to_user(argp, &wtime, sizeof(wtime)) ? -EFAULT : 0;
-
-	case RTC_SET_TIME:	/* Set the RTC */
-	    {
-		int year, days;
-
-		if (!capable(CAP_SYS_TIME))
-			return -EACCES;
-
-		if (copy_from_user(&wtime, argp, sizeof(wtime)))
-			return -EFAULT;
-
-		year = wtime.tm_year + 1900;
-		days = month_days[wtime.tm_mon] +
-		       ((wtime.tm_mon == 1) && leapyear(year));
-
-		if ((wtime.tm_mon < 0 || wtime.tm_mon > 11) ||
-		    (wtime.tm_mday < 1))
-			return -EINVAL;
-
-		if (wtime.tm_mday < 0 || wtime.tm_mday > days)
-			return -EINVAL;
-
-		if (wtime.tm_hour < 0 || wtime.tm_hour >= 24 ||
-		    wtime.tm_min < 0 || wtime.tm_min >= 60 ||
-		    wtime.tm_sec < 0 || wtime.tm_sec >= 60)
-			return -EINVAL;
-
-		return mini_set_rtc_time(&wtime);
-	    }
-	}
-
-	return -EINVAL;
-}
-
-static int mini_rtc_open(struct inode *inode, struct file *file)
-{
-	lock_kernel();
-	if (mini_rtc_status & RTC_IS_OPEN) {
-		unlock_kernel();
-		return -EBUSY;
-	}
-
-	mini_rtc_status |= RTC_IS_OPEN;
-	unlock_kernel();
-
-	return 0;
-}
-
-static int mini_rtc_release(struct inode *inode, struct file *file)
-{
-	mini_rtc_status &= ~RTC_IS_OPEN;
-	return 0;
-}
-
-
-static const struct file_operations mini_rtc_fops = {
-	.owner		= THIS_MODULE,
-	.ioctl		= mini_rtc_ioctl,
-	.open		= mini_rtc_open,
-	.release	= mini_rtc_release,
-};
-
-static struct miscdevice rtc_mini_dev =
-{
-	.minor		= RTC_MINOR,
-	.name		= "rtc",
-	.fops		= &mini_rtc_fops,
-};
-
-static int __init rtc_mini_init(void)
-{
-	int retval;
-
-	if (tlb_type == hypervisor)
-		mini_rtc_ops = &hypervisor_rtc_ops;
-	else if (this_is_starfire)
-		mini_rtc_ops = &starfire_rtc_ops;
-#ifdef CONFIG_PCI
-	else if (bq4802_regs)
-		mini_rtc_ops = &bq4802_rtc_ops;
-	else if (ds1287_regs)
-		mini_rtc_ops = &cmos_rtc_ops;
-#endif /* CONFIG_PCI */
-	else if (mstk48t02_regs)
-		mini_rtc_ops = &mostek_rtc_ops;
-	else
-		return -ENODEV;
-
-	printk(KERN_INFO "Mini RTC Driver\n");
-
-	retval = misc_register(&rtc_mini_dev);
-	if (retval < 0)
-		return retval;
-
-	return 0;
-}
-
-static void __exit rtc_mini_exit(void)
-{
-	misc_deregister(&rtc_mini_dev);
-}
-
 int __devinit read_current_timer(unsigned long *timer_val)
 {
 	*timer_val = tick_ops->get_tick();
 	return 0;
 }
-
-module_init(rtc_mini_init);
-module_exit(rtc_mini_exit);
diff --git a/arch/sparc64/kernel/traps.c b/arch/sparc64/kernel/traps.c
index c824df13f589..81ccd22e78d4 100644
--- a/arch/sparc64/kernel/traps.c
+++ b/arch/sparc64/kernel/traps.c
@@ -38,6 +38,7 @@
 #include <asm/timer.h>
 #include <asm/head.h>
 #include <asm/prom.h>
+#include <asm/memctrl.h>
 
 #include "entry.h"
 #include "kstack.h"
@@ -129,6 +130,56 @@ void do_BUG(const char *file, int line)
 }
 #endif
 
+static DEFINE_SPINLOCK(dimm_handler_lock);
+static dimm_printer_t dimm_handler;
+
+static int sprintf_dimm(int synd_code, unsigned long paddr, char *buf, int buflen)
+{
+	unsigned long flags;
+	int ret = -ENODEV;
+
+	spin_lock_irqsave(&dimm_handler_lock, flags);
+	if (dimm_handler) {
+		ret = dimm_handler(synd_code, paddr, buf, buflen);
+	} else if (tlb_type == spitfire) {
+		if (prom_getunumber(synd_code, paddr, buf, buflen) == -1)
+			ret = -EINVAL;
+		else
+			ret = 0;
+	} else
+		ret = -ENODEV;
+	spin_unlock_irqrestore(&dimm_handler_lock, flags);
+
+	return ret;
+}
+
+int register_dimm_printer(dimm_printer_t func)
+{
+	unsigned long flags;
+	int ret = 0;
+
+	spin_lock_irqsave(&dimm_handler_lock, flags);
+	if (!dimm_handler)
+		dimm_handler = func;
+	else
+		ret = -EEXIST;
+	spin_unlock_irqrestore(&dimm_handler_lock, flags);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(register_dimm_printer);
+
+void unregister_dimm_printer(dimm_printer_t func)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&dimm_handler_lock, flags);
+	if (dimm_handler == func)
+		dimm_handler = NULL;
+	spin_unlock_irqrestore(&dimm_handler_lock, flags);
+}
+EXPORT_SYMBOL_GPL(unregister_dimm_printer);
+
 void spitfire_insn_access_exception(struct pt_regs *regs, unsigned long sfsr, unsigned long sfar)
 {
 	siginfo_t info;
@@ -291,10 +342,7 @@ void sun4v_data_access_exception_tl1(struct pt_regs *regs, unsigned long addr, u
 }
 
 #ifdef CONFIG_PCI
-/* This is really pathetic... */
-extern volatile int pci_poke_in_progress;
-extern volatile int pci_poke_cpu;
-extern volatile int pci_poke_faulted;
+#include "pci_impl.h"
 #endif
 
 /* When access exceptions happen, we must do this. */
@@ -376,8 +424,7 @@ static void spitfire_log_udb_syndrome(unsigned long afar, unsigned long udbh, un
 
 	if (udbl & bit) {
 		scode = ecc_syndrome_table[udbl & 0xff];
-		if (prom_getunumber(scode, afar,
-				    memmod_str, sizeof(memmod_str)) == -1)
+		if (sprintf_dimm(scode, afar, memmod_str, sizeof(memmod_str)) < 0)
 			p = syndrome_unknown;
 		else
 			p = memmod_str;
@@ -388,8 +435,7 @@ static void spitfire_log_udb_syndrome(unsigned long afar, unsigned long udbh, un
 
 	if (udbh & bit) {
 		scode = ecc_syndrome_table[udbh & 0xff];
-		if (prom_getunumber(scode, afar,
-				    memmod_str, sizeof(memmod_str)) == -1)
+		if (sprintf_dimm(scode, afar, memmod_str, sizeof(memmod_str)) < 0)
 			p = syndrome_unknown;
 		else
 			p = memmod_str;
@@ -1062,8 +1108,6 @@ static const char *cheetah_get_string(unsigned long bit)
 	return "???";
 }
 
-extern int chmc_getunumber(int, unsigned long, char *, int);
-
 static void cheetah_log_errors(struct pt_regs *regs, struct cheetah_err_info *info,
 			       unsigned long afsr, unsigned long afar, int recoverable)
 {
@@ -1105,7 +1149,7 @@ static void cheetah_log_errors(struct pt_regs *regs, struct cheetah_err_info *in
 
 		syndrome = (afsr & CHAFSR_E_SYNDROME) >> CHAFSR_E_SYNDROME_SHIFT;
 		syndrome = cheetah_ecc_syntab[syndrome];
-		ret = chmc_getunumber(syndrome, afar, unum, sizeof(unum));
+		ret = sprintf_dimm(syndrome, afar, unum, sizeof(unum));
 		if (ret != -1)
 			printk("%s" "ERROR(%d): AFAR E-syndrome [%s]\n",
 			       (recoverable ? KERN_WARNING : KERN_CRIT),
@@ -1116,7 +1160,7 @@ static void cheetah_log_errors(struct pt_regs *regs, struct cheetah_err_info *in
 
 		syndrome = (afsr & CHAFSR_M_SYNDROME) >> CHAFSR_M_SYNDROME_SHIFT;
 		syndrome = cheetah_mtag_syntab[syndrome];
-		ret = chmc_getunumber(syndrome, afar, unum, sizeof(unum));
+		ret = sprintf_dimm(syndrome, afar, unum, sizeof(unum));
 		if (ret != -1)
 			printk("%s" "ERROR(%d): AFAR M-syndrome [%s]\n",
 			       (recoverable ? KERN_WARNING : KERN_CRIT),
@@ -2224,7 +2268,6 @@ void die_if_kernel(char *str, struct pt_regs *regs)
 
 extern int handle_popc(u32 insn, struct pt_regs *regs);
 extern int handle_ldf_stq(u32 insn, struct pt_regs *regs);
-extern int vis_emul(struct pt_regs *, unsigned int);
 
 void do_illegal_instruction(struct pt_regs *regs)
 {
diff --git a/arch/sparc64/kernel/vio.c b/arch/sparc64/kernel/vio.c
index a490077891a4..92b1f8ec01de 100644
--- a/arch/sparc64/kernel/vio.c
+++ b/arch/sparc64/kernel/vio.c
@@ -152,7 +152,7 @@ show_pciobppath_attr(struct device *dev, struct device_attribute *attr,
 static DEVICE_ATTR(obppath, S_IRUSR | S_IRGRP | S_IROTH,
 		   show_pciobppath_attr, NULL);
 
-struct device_node *cdev_node;
+static struct device_node *cdev_node;
 
 static struct vio_dev *root_vdev;
 static u64 cdev_cfg_handle;
@@ -371,9 +371,9 @@ static struct mdesc_notifier_client vio_ds_notifier = {
 	.node_name	= "domain-services-port",
 };
 
-const char *channel_devices_node = "channel-devices";
-const char *channel_devices_compat = "SUNW,sun4v-channel-devices";
-const char *cfg_handle_prop = "cfg-handle";
+static const char *channel_devices_node = "channel-devices";
+static const char *channel_devices_compat = "SUNW,sun4v-channel-devices";
+static const char *cfg_handle_prop = "cfg-handle";
 
 static int __init vio_init(void)
 {
diff --git a/arch/sparc64/kernel/visemul.c b/arch/sparc64/kernel/visemul.c
index c3fd64706b53..9e05cb5cb855 100644
--- a/arch/sparc64/kernel/visemul.c
+++ b/arch/sparc64/kernel/visemul.c
@@ -243,7 +243,7 @@ static inline unsigned int *fps_regaddr(struct fpustate *f,
 struct edge_tab {
 	u16 left, right;
 };
-struct edge_tab edge8_tab[8] = {
+static struct edge_tab edge8_tab[8] = {
 	{ 0xff, 0x80 },
 	{ 0x7f, 0xc0 },
 	{ 0x3f, 0xe0 },
@@ -253,7 +253,7 @@ struct edge_tab edge8_tab[8] = {
 	{ 0x03, 0xfe },
 	{ 0x01, 0xff },
 };
-struct edge_tab edge8_tab_l[8] = {
+static struct edge_tab edge8_tab_l[8] = {
 	{ 0xff, 0x01 },
 	{ 0xfe, 0x03 },
 	{ 0xfc, 0x07 },
@@ -263,23 +263,23 @@ struct edge_tab edge8_tab_l[8] = {
 	{ 0xc0, 0x7f },
 	{ 0x80, 0xff },
 };
-struct edge_tab edge16_tab[4] = {
+static struct edge_tab edge16_tab[4] = {
 	{ 0xf, 0x8 },
 	{ 0x7, 0xc },
 	{ 0x3, 0xe },
 	{ 0x1, 0xf },
 };
-struct edge_tab edge16_tab_l[4] = {
+static struct edge_tab edge16_tab_l[4] = {
 	{ 0xf, 0x1 },
 	{ 0xe, 0x3 },
 	{ 0xc, 0x7 },
 	{ 0x8, 0xf },
 };
-struct edge_tab edge32_tab[2] = {
+static struct edge_tab edge32_tab[2] = {
 	{ 0x3, 0x2 },
 	{ 0x1, 0x3 },
 };
-struct edge_tab edge32_tab_l[2] = {
+static struct edge_tab edge32_tab_l[2] = {
 	{ 0x3, 0x1 },
 	{ 0x2, 0x3 },
 };
diff --git a/arch/sparc64/mm/fault.c b/arch/sparc64/mm/fault.c
index ea7d7ae76bc2..a9e474bf6385 100644
--- a/arch/sparc64/mm/fault.c
+++ b/arch/sparc64/mm/fault.c
@@ -51,43 +51,6 @@ static inline int notify_page_fault(struct pt_regs *regs)
 }
 #endif
 
-/*
- * To debug kernel to catch accesses to certain virtual/physical addresses.
- * Mode = 0 selects physical watchpoints, mode = 1 selects virtual watchpoints.
- * flags = VM_READ watches memread accesses, flags = VM_WRITE watches memwrite accesses.
- * Caller passes in a 64bit aligned addr, with mask set to the bytes that need to be
- * watched. This is only useful on a single cpu machine for now. After the watchpoint
- * is detected, the process causing it will be killed, thus preventing an infinite loop.
- */
-void set_brkpt(unsigned long addr, unsigned char mask, int flags, int mode)
-{
-	unsigned long lsubits;
-
-	__asm__ __volatile__("ldxa [%%g0] %1, %0"
-			     : "=r" (lsubits)
-			     : "i" (ASI_LSU_CONTROL));
-	lsubits &= ~(LSU_CONTROL_PM | LSU_CONTROL_VM |
-		     LSU_CONTROL_PR | LSU_CONTROL_VR |
-		     LSU_CONTROL_PW | LSU_CONTROL_VW);
-
-	__asm__ __volatile__("stxa	%0, [%1] %2\n\t"
-			     "membar	#Sync"
-			     : /* no outputs */
-			     : "r" (addr), "r" (mode ? VIRT_WATCHPOINT : PHYS_WATCHPOINT),
-			       "i" (ASI_DMMU));
-
-	lsubits |= ((unsigned long)mask << (mode ? 25 : 33));
-	if (flags & VM_READ)
-		lsubits |= (mode ? LSU_CONTROL_VR : LSU_CONTROL_PR);
-	if (flags & VM_WRITE)
-		lsubits |= (mode ? LSU_CONTROL_VW : LSU_CONTROL_PW);
-	__asm__ __volatile__("stxa %0, [%%g0] %1\n\t"
-			     "membar #Sync"
-			     : /* no outputs */
-			     : "r" (lsubits), "i" (ASI_LSU_CONTROL)
-			     : "memory");
-}
-
 static void __kprobes unhandled_fault(unsigned long address,
 				      struct task_struct *tsk,
 				      struct pt_regs *regs)
diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c
index a41df7bef035..3c10daf8fc01 100644
--- a/arch/sparc64/mm/init.c
+++ b/arch/sparc64/mm/init.c
@@ -46,15 +46,11 @@
 #include <asm/tsb.h>
 #include <asm/hypervisor.h>
 #include <asm/prom.h>
-#include <asm/sstate.h>
 #include <asm/mdesc.h>
 #include <asm/cpudata.h>
 #include <asm/irq.h>
 
-#define MAX_PHYS_ADDRESS	(1UL << 42UL)
-#define KPTE_BITMAP_CHUNK_SZ	(256UL * 1024UL * 1024UL)
-#define KPTE_BITMAP_BYTES	\
-	((MAX_PHYS_ADDRESS / KPTE_BITMAP_CHUNK_SZ) / 8)
+#include "init.h"
 
 unsigned long kern_linear_pte_xor[2] __read_mostly;
 
@@ -416,17 +412,9 @@ void mmu_info(struct seq_file *m)
 #endif /* CONFIG_DEBUG_DCFLUSH */
 }
 
-struct linux_prom_translation {
-	unsigned long virt;
-	unsigned long size;
-	unsigned long data;
-};
-
-/* Exported for kernel TLB miss handling in ktlb.S */
 struct linux_prom_translation prom_trans[512] __read_mostly;
 unsigned int prom_trans_ents __read_mostly;
 
-/* Exported for SMP bootup purposes. */
 unsigned long kern_locked_tte_data;
 
 /* The obp translations are saved based on 8k pagesize, since obp can
@@ -938,6 +926,10 @@ int of_node_to_nid(struct device_node *dp)
 	int count, nid;
 	u64 grp;
 
+	/* This is the right thing to do on currently supported
+	 * SUN4U NUMA platforms as well, as the PCI controller does
+	 * not sit behind any particular memory controller.
+	 */
 	if (!mlgroups)
 		return -1;
 
@@ -1206,8 +1198,44 @@ out:
 	return err;
 }
 
+static int __init numa_parse_jbus(void)
+{
+	unsigned long cpu, index;
+
+	/* NUMA node id is encoded in bits 36 and higher, and there is
+	 * a 1-to-1 mapping from CPU ID to NUMA node ID.
+	 */
+	index = 0;
+	for_each_present_cpu(cpu) {
+		numa_cpu_lookup_table[cpu] = index;
+		numa_cpumask_lookup_table[index] = cpumask_of_cpu(cpu);
+		node_masks[index].mask = ~((1UL << 36UL) - 1UL);
+		node_masks[index].val = cpu << 36UL;
+
+		index++;
+	}
+	num_node_masks = index;
+
+	add_node_ranges();
+
+	for (index = 0; index < num_node_masks; index++) {
+		allocate_node_data(index);
+		node_set_online(index);
+	}
+
+	return 0;
+}
+
 static int __init numa_parse_sun4u(void)
 {
+	if (tlb_type == cheetah || tlb_type == cheetah_plus) {
+		unsigned long ver;
+
+		__asm__ ("rdpr %%ver, %0" : "=r" (ver));
+		if ((ver >> 32UL) == __JALAPENO_ID ||
+		    (ver >> 32UL) == __SERRANO_ID)
+			return numa_parse_jbus();
+	}
 	return -1;
 }
 
@@ -1633,8 +1661,6 @@ void __cpuinit sun4v_ktsb_register(void)
 
 /* paging_init() sets up the page tables */
 
-extern void central_probe(void);
-
 static unsigned long last_valid_pfn;
 pgd_t swapper_pg_dir[2048];
 
@@ -1679,8 +1705,6 @@ void __init paging_init(void)
 	kern_base = (prom_boot_mapping_phys_low >> 22UL) << 22UL;
 	kern_size = (unsigned long)&_end - (unsigned long)KERNBASE;
 
-	sstate_booting();
-
 	/* Invalidate both kernel TSBs.  */
 	memset(swapper_tsb, 0x40, sizeof(swapper_tsb));
 #ifndef CONFIG_DEBUG_PAGEALLOC
@@ -1803,9 +1827,6 @@ void __init paging_init(void)
 	}
 
 	printk("Booting Linux...\n");
-
-	central_probe();
-	cpu_probe();
 }
 
 int __init page_in_phys_avail(unsigned long paddr)
@@ -2032,7 +2053,6 @@ pgprot_t PAGE_COPY __read_mostly;
 pgprot_t PAGE_SHARED __read_mostly;
 EXPORT_SYMBOL(PAGE_SHARED);
 
-pgprot_t PAGE_EXEC __read_mostly;
 unsigned long pg_iobits __read_mostly;
 
 unsigned long _PAGE_IE __read_mostly;
@@ -2045,14 +2065,6 @@ unsigned long _PAGE_CACHE __read_mostly;
 EXPORT_SYMBOL(_PAGE_CACHE);
 
 #ifdef CONFIG_SPARSEMEM_VMEMMAP
-
-#define VMEMMAP_CHUNK_SHIFT	22
-#define VMEMMAP_CHUNK		(1UL << VMEMMAP_CHUNK_SHIFT)
-#define VMEMMAP_CHUNK_MASK	~(VMEMMAP_CHUNK - 1UL)
-#define VMEMMAP_ALIGN(x)	(((x)+VMEMMAP_CHUNK-1UL)&VMEMMAP_CHUNK_MASK)
-
-#define VMEMMAP_SIZE	((((1UL << MAX_PHYSADDR_BITS) >> PAGE_SHIFT) * \
-			  sizeof(struct page *)) >> VMEMMAP_CHUNK_SHIFT)
 unsigned long vmemmap_table[VMEMMAP_SIZE];
 
 int __meminit vmemmap_populate(struct page *start, unsigned long nr, int node)
@@ -2136,7 +2148,6 @@ static void __init sun4u_pgprot_init(void)
 				       _PAGE_CACHE_4U | _PAGE_P_4U |
 				       __ACCESS_BITS_4U | __DIRTY_BITS_4U |
 				       _PAGE_EXEC_4U | _PAGE_L_4U);
-	PAGE_EXEC = __pgprot(_PAGE_EXEC_4U);
 
 	_PAGE_IE = _PAGE_IE_4U;
 	_PAGE_E = _PAGE_E_4U;
@@ -2147,10 +2158,10 @@ static void __init sun4u_pgprot_init(void)
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
 	kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZBITS_4U) ^
-		0xfffff80000000000;
+		0xfffff80000000000UL;
 #else
 	kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZ4MB_4U) ^
-		0xfffff80000000000;
+		0xfffff80000000000UL;
 #endif
 	kern_linear_pte_xor[0] |= (_PAGE_CP_4U | _PAGE_CV_4U |
 				   _PAGE_P_4U | _PAGE_W_4U);
@@ -2188,7 +2199,6 @@ static void __init sun4v_pgprot_init(void)
 				__ACCESS_BITS_4V | __DIRTY_BITS_4V |
 				_PAGE_EXEC_4V);
 	PAGE_KERNEL_LOCKED = PAGE_KERNEL;
-	PAGE_EXEC = __pgprot(_PAGE_EXEC_4V);
 
 	_PAGE_IE = _PAGE_IE_4V;
 	_PAGE_E = _PAGE_E_4V;
@@ -2196,20 +2206,20 @@ static void __init sun4v_pgprot_init(void)
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
 	kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZBITS_4V) ^
-		0xfffff80000000000;
+		0xfffff80000000000UL;
 #else
 	kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZ4MB_4V) ^
-		0xfffff80000000000;
+		0xfffff80000000000UL;
 #endif
 	kern_linear_pte_xor[0] |= (_PAGE_CP_4V | _PAGE_CV_4V |
 				   _PAGE_P_4V | _PAGE_W_4V);
 
 #ifdef CONFIG_DEBUG_PAGEALLOC
 	kern_linear_pte_xor[1] = (_PAGE_VALID | _PAGE_SZBITS_4V) ^
-		0xfffff80000000000;
+		0xfffff80000000000UL;
 #else
 	kern_linear_pte_xor[1] = (_PAGE_VALID | _PAGE_SZ256MB_4V) ^
-		0xfffff80000000000;
+		0xfffff80000000000UL;
 #endif
 	kern_linear_pte_xor[1] |= (_PAGE_CP_4V | _PAGE_CV_4V |
 				   _PAGE_P_4V | _PAGE_W_4V);
diff --git a/arch/sparc64/mm/init.h b/arch/sparc64/mm/init.h
new file mode 100644
index 000000000000..16063870a489
--- /dev/null
+++ b/arch/sparc64/mm/init.h
@@ -0,0 +1,49 @@
+#ifndef _SPARC64_MM_INIT_H
+#define _SPARC64_MM_INIT_H
+
+/* Most of the symbols in this file are defined in init.c and
+ * marked non-static so that assembler code can get at them.
+ */
+
+#define MAX_PHYS_ADDRESS	(1UL << 42UL)
+#define KPTE_BITMAP_CHUNK_SZ	(256UL * 1024UL * 1024UL)
+#define KPTE_BITMAP_BYTES	\
+	((MAX_PHYS_ADDRESS / KPTE_BITMAP_CHUNK_SZ) / 8)
+
+extern unsigned long kern_linear_pte_xor[2];
+extern unsigned long kpte_linear_bitmap[KPTE_BITMAP_BYTES / sizeof(unsigned long)];
+extern unsigned int sparc64_highest_unlocked_tlb_ent;
+extern unsigned long sparc64_kern_pri_context;
+extern unsigned long sparc64_kern_pri_nuc_bits;
+extern unsigned long sparc64_kern_sec_context;
+extern void mmu_info(struct seq_file *m);
+
+struct linux_prom_translation {
+	unsigned long virt;
+	unsigned long size;
+	unsigned long data;
+};
+
+/* Exported for kernel TLB miss handling in ktlb.S */
+extern struct linux_prom_translation prom_trans[512];
+extern unsigned int prom_trans_ents;
+
+/* Exported for SMP bootup purposes. */
+extern unsigned long kern_locked_tte_data;
+
+extern void prom_world(int enter);
+
+extern void free_initmem(void);
+
+#ifdef CONFIG_SPARSEMEM_VMEMMAP
+#define VMEMMAP_CHUNK_SHIFT	22
+#define VMEMMAP_CHUNK		(1UL << VMEMMAP_CHUNK_SHIFT)
+#define VMEMMAP_CHUNK_MASK	~(VMEMMAP_CHUNK - 1UL)
+#define VMEMMAP_ALIGN(x)	(((x)+VMEMMAP_CHUNK-1UL)&VMEMMAP_CHUNK_MASK)
+
+#define VMEMMAP_SIZE	((((1UL << MAX_PHYSADDR_BITS) >> PAGE_SHIFT) * \
+			  sizeof(struct page *)) >> VMEMMAP_CHUNK_SHIFT)
+extern unsigned long vmemmap_table[VMEMMAP_SIZE];
+#endif
+
+#endif /* _SPARC64_MM_INIT_H */
diff --git a/arch/sparc64/mm/tlb.c b/arch/sparc64/mm/tlb.c
index ae24919cba7c..d8f21e24a82f 100644
--- a/arch/sparc64/mm/tlb.c
+++ b/arch/sparc64/mm/tlb.c
@@ -19,7 +19,7 @@
 
 /* Heavily inspired by the ppc64 code.  */
 
-DEFINE_PER_CPU(struct mmu_gather, mmu_gathers) = { 0, };
+DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 
 void flush_tlb_pending(void)
 {
diff --git a/arch/um/Kconfig.i386 b/arch/um/Kconfig.i386
index e09edfa560da..1f57c113df6d 100644
--- a/arch/um/Kconfig.i386
+++ b/arch/um/Kconfig.i386
@@ -9,8 +9,9 @@ config UML_X86
 	default y
 
 config X86_32
-       bool
-       default y
+	bool
+	default y
+  	select HAVE_AOUT
 
 config RWSEM_XCHGADD_ALGORITHM
 	def_bool y
@@ -42,6 +43,3 @@ config ARCH_REUSE_HOST_VSYSCALL_AREA
 config GENERIC_HWEIGHT
 	bool
 	default y
-
-config ARCH_SUPPORTS_AOUT
-	def_bool y
diff --git a/arch/um/Kconfig.x86_64 b/arch/um/Kconfig.x86_64
index 5696e7b374b3..40b3407cfe16 100644
--- a/arch/um/Kconfig.x86_64
+++ b/arch/um/Kconfig.x86_64
@@ -37,6 +37,3 @@ config SMP_BROKEN
 config GENERIC_HWEIGHT
 	bool
 	default y
-
-config ARCH_SUPPORTS_AOUT
-	def_bool y
diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c
index d741f35d7b3a..14a102e877d6 100644
--- a/arch/um/drivers/line.c
+++ b/arch/um/drivers/line.c
@@ -275,6 +275,8 @@ int line_ioctl(struct tty_struct *tty, struct file * file,
 	case TIOCGLTC:
 	case TIOCSLTC:
 #endif
+	/* Note: these are out of date as we now have TCGETS2 etc but this
+	   whole lot should probably go away */
 	case TCGETS:
 	case TCSETSF:
 	case TCSETSW:
diff --git a/arch/um/sys-x86_64/syscall_table.c b/arch/um/sys-x86_64/syscall_table.c
index c128eb897008..32f5fbe2d0d2 100644
--- a/arch/um/sys-x86_64/syscall_table.c
+++ b/arch/um/sys-x86_64/syscall_table.c
@@ -41,12 +41,12 @@
 #define stub_rt_sigreturn sys_rt_sigreturn
 
 #define __SYSCALL(nr, sym) extern asmlinkage void sym(void) ;
-#undef _ASM_X86_64_UNISTD_H_
+#undef ASM_X86__UNISTD_64_H
 #include <asm-x86/unistd_64.h>
 
 #undef __SYSCALL
 #define __SYSCALL(nr, sym) [ nr ] = sym,
-#undef _ASM_X86_64_UNISTD_H_
+#undef ASM_X86__UNISTD_64_H
 
 typedef void (*sys_call_ptr_t)(void);
 
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 0d7cdbbfc1ee..f65c2744d573 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -18,6 +18,7 @@ config X86_64
 ### Arch settings
 config X86
 	def_bool y
+	select HAVE_AOUT if X86_32
 	select HAVE_UNSTABLE_SCHED_CLOCK
 	select HAVE_IDE
 	select HAVE_OPROFILE
@@ -152,9 +153,6 @@ config AUDIT_ARCH
 	bool
 	default X86_64
 
-config ARCH_SUPPORTS_AOUT
-	def_bool y
-
 config ARCH_SUPPORTS_OPTIMIZED_INLINING
 	def_bool y
 
@@ -778,23 +776,45 @@ config X86_REBOOTFIXUPS
 	  Say N otherwise.
 
 config MICROCODE
-	tristate "/dev/cpu/microcode - Intel IA32 CPU microcode support"
+	tristate "/dev/cpu/microcode - microcode support"
 	select FW_LOADER
 	---help---
 	  If you say Y here, you will be able to update the microcode on
-	  Intel processors in the IA32 family, e.g. Pentium Pro, Pentium II,
-	  Pentium III, Pentium 4, Xeon etc.  You will obviously need the
-	  actual microcode binary data itself which is not shipped with the
-	  Linux kernel.
+	  certain Intel and AMD processors. The Intel support is for the
+	  IA32 family, e.g. Pentium Pro, Pentium II, Pentium III,
+	  Pentium 4, Xeon etc. The AMD support is for family 0x10 and
+	  0x11 processors, e.g. Opteron, Phenom and Turion 64 Ultra.
+	  You will obviously need the actual microcode binary data itself
+	  which is not shipped with the Linux kernel.
 
-	  For latest news and information on obtaining all the required
-	  ingredients for this driver, check:
-	  <http://www.urbanmyth.org/microcode/>.
+	  This option selects the general module only, you need to select
+	  at least one vendor specific module as well.
 
 	  To compile this driver as a module, choose M here: the
 	  module will be called microcode.
 
-config MICROCODE_OLD_INTERFACE
+config MICROCODE_INTEL
+       bool "Intel microcode patch loading support"
+       depends on MICROCODE
+       default MICROCODE
+       select FW_LOADER
+       --help---
+         This options enables microcode patch loading support for Intel
+         processors.
+
+         For latest news and information on obtaining all the required
+         Intel ingredients for this driver, check:
+         <http://www.urbanmyth.org/microcode/>.
+
+config MICROCODE_AMD
+       bool "AMD microcode patch loading support"
+       depends on MICROCODE
+       select FW_LOADER
+       --help---
+         If you select this option, microcode patch loading support for AMD
+	 processors will be enabled.
+
+   config MICROCODE_OLD_INTERFACE
 	def_bool y
 	depends on MICROCODE
 
@@ -1061,6 +1081,56 @@ config HIGHPTE
 	  low memory.  Setting this option will put user-space page table
 	  entries in high memory.
 
+config X86_CHECK_BIOS_CORRUPTION
+        bool "Check for low memory corruption"
+	help
+	 Periodically check for memory corruption in low memory, which
+	 is suspected to be caused by BIOS.  Even when enabled in the
+	 configuration, it is disabled at runtime.  Enable it by
+	 setting "memory_corruption_check=1" on the kernel command
+	 line.  By default it scans the low 64k of memory every 60
+	 seconds; see the memory_corruption_check_size and
+	 memory_corruption_check_period parameters in
+	 Documentation/kernel-parameters.txt to adjust this.
+
+	 When enabled with the default parameters, this option has
+	 almost no overhead, as it reserves a relatively small amount
+	 of memory and scans it infrequently.  It both detects corruption
+	 and prevents it from affecting the running system.
+
+	 It is, however, intended as a diagnostic tool; if repeatable
+	 BIOS-originated corruption always affects the same memory,
+	 you can use memmap= to prevent the kernel from using that
+	 memory.
+
+config X86_BOOTPARAM_MEMORY_CORRUPTION_CHECK
+        bool "Set the default setting of memory_corruption_check"
+	depends on X86_CHECK_BIOS_CORRUPTION
+	default y
+	help
+	 Set whether the default state of memory_corruption_check is
+	 on or off.
+
+config X86_RESERVE_LOW_64K
+        bool "Reserve low 64K of RAM on AMI/Phoenix BIOSen"
+	default y
+	help
+	 Reserve the first 64K of physical RAM on BIOSes that are known
+	 to potentially corrupt that memory range. A numbers of BIOSes are
+	 known to utilize this area during suspend/resume, so it must not
+	 be used by the kernel.
+
+	 Set this to N if you are absolutely sure that you trust the BIOS
+	 to get all its memory reservations and usages right.
+
+	 If you have doubts about the BIOS (e.g. suspend/resume does not
+	 work or there's kernel crashes after certain hardware hotplug
+	 events) and it's not AMI or Phoenix, then you might want to enable
+	 X86_CHECK_BIOS_CORRUPTION=y to allow the kernel to check typical
+	 corruption patterns.
+
+	 Say Y if unsure.
+
 config MATH_EMULATION
 	bool
 	prompt "Math emulation" if X86_32
@@ -1689,6 +1759,14 @@ config DMAR_FLOPPY_WA
 	 workaround will setup a 1:1 mapping for the first
 	 16M to make floppy (an ISA device) work.
 
+config INTR_REMAP
+	bool "Support for Interrupt Remapping (EXPERIMENTAL)"
+	depends on X86_64 && X86_IO_APIC && PCI_MSI && ACPI && EXPERIMENTAL
+	help
+	 Supports Interrupt remapping for IO-APIC and MSI devices.
+	 To use x2apic mode in the CPU's which support x2APIC enhancements or
+	 to support platforms with CPU's having > 8 bit APIC ID, say Y.
+
 source "drivers/pci/pcie/Kconfig"
 
 source "drivers/pci/Kconfig"
@@ -1805,7 +1883,7 @@ config IA32_EMULATION
 
 config IA32_AOUT
        tristate "IA32 a.out support"
-       depends on IA32_EMULATION && ARCH_SUPPORTS_AOUT
+       depends on IA32_EMULATION
        help
          Support old a.out binaries in the 32bit emulation.
 
diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu
index 60a85768cfcb..0b7c4a3f0651 100644
--- a/arch/x86/Kconfig.cpu
+++ b/arch/x86/Kconfig.cpu
@@ -38,8 +38,7 @@ config M386
 	  - "Crusoe" for the Transmeta Crusoe series.
 	  - "Efficeon" for the Transmeta Efficeon series.
 	  - "Winchip-C6" for original IDT Winchip.
-	  - "Winchip-2" for IDT Winchip 2.
-	  - "Winchip-2A" for IDT Winchips with 3dNow! capabilities.
+	  - "Winchip-2" for IDT Winchips with 3dNow! capabilities.
 	  - "GeodeGX1" for Geode GX1 (Cyrix MediaGX).
 	  - "Geode GX/LX" For AMD Geode GX and LX processors.
 	  - "CyrixIII/VIA C3" for VIA Cyrix III or VIA C3.
@@ -194,19 +193,11 @@ config MWINCHIPC6
 	  treat this chip as a 586TSC with some extended instructions
 	  and alignment requirements.
 
-config MWINCHIP2
-	bool "Winchip-2"
-	depends on X86_32
-	help
-	  Select this for an IDT Winchip-2.  Linux and GCC
-	  treat this chip as a 586TSC with some extended instructions
-	  and alignment requirements.
-
 config MWINCHIP3D
-	bool "Winchip-2A/Winchip-3"
+	bool "Winchip-2/Winchip-2A/Winchip-3"
 	depends on X86_32
 	help
-	  Select this for an IDT Winchip-2A or 3.  Linux and GCC
+	  Select this for an IDT Winchip-2, 2A or 3.  Linux and GCC
 	  treat this chip as a 586TSC with some extended instructions
 	  and alignment requirements.  Also enable out of order memory
 	  stores for this CPU, which can increase performance of some
@@ -318,7 +309,7 @@ config X86_L1_CACHE_SHIFT
 	int
 	default "7" if MPENTIUM4 || X86_GENERIC || GENERIC_CPU || MPSC
 	default "4" if X86_ELAN || M486 || M386 || MGEODEGX1
-	default "5" if MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCRUSOE || MEFFICEON || MCYRIXIII || MK6 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || MVIAC3_2 || MGEODE_LX
+	default "5" if MWINCHIP3D || MWINCHIPC6 || MCRUSOE || MEFFICEON || MCYRIXIII || MK6 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || MVIAC3_2 || MGEODE_LX
 	default "6" if MK7 || MK8 || MPENTIUMM || MCORE2 || MVIAC7
 
 config X86_XADD
@@ -360,7 +351,7 @@ config X86_POPAD_OK
 
 config X86_ALIGNMENT_16
 	def_bool y
-	depends on MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1
+	depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1
 
 config X86_INTEL_USERCOPY
 	def_bool y
@@ -368,7 +359,7 @@ config X86_INTEL_USERCOPY
 
 config X86_USE_PPRO_CHECKSUM
 	def_bool y
-	depends on MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MK8 || MVIAC3_2 || MEFFICEON || MGEODE_LX || MCORE2
+	depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MK8 || MVIAC3_2 || MEFFICEON || MGEODE_LX || MCORE2
 
 config X86_USE_3DNOW
 	def_bool y
@@ -376,7 +367,7 @@ config X86_USE_3DNOW
 
 config X86_OOSTORE
 	def_bool y
-	depends on (MWINCHIP3D || MWINCHIP2 || MWINCHIPC6) && MTRR
+	depends on (MWINCHIP3D || MWINCHIPC6) && MTRR
 
 #
 # P6_NOPs are a relatively minor optimization that require a family >=
@@ -396,7 +387,7 @@ config X86_P6_NOP
 
 config X86_TSC
 	def_bool y
-	depends on ((MWINCHIP3D || MWINCHIP2 || MCRUSOE || MEFFICEON || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2 || MVIAC7 || MGEODEGX1 || MGEODE_LX || MCORE2) && !X86_NUMAQ) || X86_64
+	depends on ((MWINCHIP3D || MCRUSOE || MEFFICEON || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2 || MVIAC7 || MGEODEGX1 || MGEODE_LX || MCORE2) && !X86_NUMAQ) || X86_64
 
 config X86_CMPXCHG64
 	def_bool y
@@ -406,7 +397,7 @@ config X86_CMPXCHG64
 # generates cmov.
 config X86_CMOV
 	def_bool y
-	depends on (MK7 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MVIAC3_2 || MVIAC7 || X86_64)
+	depends on (MK8 || MK7 || MCORE2 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MVIAC3_2 || MVIAC7 || MCRUSOE || MEFFICEON || X86_64)
 
 config X86_MINIMUM_CPU_FAMILY
 	int
@@ -417,7 +408,109 @@ config X86_MINIMUM_CPU_FAMILY
 
 config X86_DEBUGCTLMSR
 	def_bool y
-	depends on !(MK6 || MWINCHIPC6 || MWINCHIP2 || MWINCHIP3D || MCYRIXIII || M586MMX || M586TSC || M586 || M486 || M386)
+	depends on !(MK6 || MWINCHIPC6 || MWINCHIP3D || MCYRIXIII || M586MMX || M586TSC || M586 || M486 || M386)
+
+menuconfig PROCESSOR_SELECT
+	bool "Supported processor vendors" if EMBEDDED
+	help
+	  This lets you choose what x86 vendor support code your kernel
+	  will include.
+
+config CPU_SUP_INTEL
+	default y
+	bool "Support Intel processors" if PROCESSOR_SELECT
+	help
+	  This enables detection, tunings and quirks for Intel processors
+
+	  You need this enabled if you want your kernel to run on an
+	  Intel CPU. Disabling this option on other types of CPUs
+	  makes the kernel a tiny bit smaller. Disabling it on an Intel
+	  CPU might render the kernel unbootable.
+
+	  If unsure, say N.
+
+config CPU_SUP_CYRIX_32
+	default y
+	bool "Support Cyrix processors" if PROCESSOR_SELECT
+	depends on !64BIT
+	help
+	  This enables detection, tunings and quirks for Cyrix processors
+
+	  You need this enabled if you want your kernel to run on a
+	  Cyrix CPU. Disabling this option on other types of CPUs
+	  makes the kernel a tiny bit smaller. Disabling it on a Cyrix
+	  CPU might render the kernel unbootable.
+
+	  If unsure, say N.
+
+config CPU_SUP_AMD
+	default y
+	bool "Support AMD processors" if PROCESSOR_SELECT
+	help
+	  This enables detection, tunings and quirks for AMD processors
+
+	  You need this enabled if you want your kernel to run on an
+	  AMD CPU. Disabling this option on other types of CPUs
+	  makes the kernel a tiny bit smaller. Disabling it on an AMD
+	  CPU might render the kernel unbootable.
+
+	  If unsure, say N.
+
+config CPU_SUP_CENTAUR_32
+	default y
+	bool "Support Centaur processors" if PROCESSOR_SELECT
+	depends on !64BIT
+	help
+	  This enables detection, tunings and quirks for Centaur processors
+
+	  You need this enabled if you want your kernel to run on a
+	  Centaur CPU. Disabling this option on other types of CPUs
+	  makes the kernel a tiny bit smaller. Disabling it on a Centaur
+	  CPU might render the kernel unbootable.
+
+	  If unsure, say N.
+
+config CPU_SUP_CENTAUR_64
+	default y
+	bool "Support Centaur processors" if PROCESSOR_SELECT
+	depends on 64BIT
+	help
+	  This enables detection, tunings and quirks for Centaur processors
+
+	  You need this enabled if you want your kernel to run on a
+	  Centaur CPU. Disabling this option on other types of CPUs
+	  makes the kernel a tiny bit smaller. Disabling it on a Centaur
+	  CPU might render the kernel unbootable.
+
+	  If unsure, say N.
+
+config CPU_SUP_TRANSMETA_32
+	default y
+	bool "Support Transmeta processors" if PROCESSOR_SELECT
+	depends on !64BIT
+	help
+	  This enables detection, tunings and quirks for Transmeta processors
+
+	  You need this enabled if you want your kernel to run on a
+	  Transmeta CPU. Disabling this option on other types of CPUs
+	  makes the kernel a tiny bit smaller. Disabling it on a Transmeta
+	  CPU might render the kernel unbootable.
+
+	  If unsure, say N.
+
+config CPU_SUP_UMC_32
+	default y
+	bool "Support UMC processors" if PROCESSOR_SELECT
+	depends on !64BIT
+	help
+	  This enables detection, tunings and quirks for UMC processors
+
+	  You need this enabled if you want your kernel to run on a
+	  UMC CPU. Disabling this option on other types of CPUs
+	  makes the kernel a tiny bit smaller. Disabling it on a UMC
+	  CPU might render the kernel unbootable.
+
+	  If unsure, say N.
 
 config X86_DS
 	bool "Debug Store support"
diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index 092f019e033a..2a3dfbd5e677 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -43,6 +43,19 @@ config EARLY_PRINTK
 	  with klogd/syslogd or the X server. You should normally N here,
 	  unless you want to debug such a crash.
 
+config EARLY_PRINTK_DBGP
+	bool "Early printk via EHCI debug port"
+	default n
+	depends on EARLY_PRINTK && PCI
+	help
+	  Write kernel log output directly into the EHCI debug port.
+
+	  This is useful for kernel debugging when your machine crashes very
+	  early before the console code is initialized. For normal operation
+	  it is not recommended because it looks ugly and doesn't cooperate
+	  with klogd/syslogd or the X server. You should normally N here,
+	  unless you want to debug such a crash. You need usb debug device.
+
 config DEBUG_STACKOVERFLOW
 	bool "Check for stack overflows"
 	depends on DEBUG_KERNEL
diff --git a/arch/x86/Makefile_32.cpu b/arch/x86/Makefile_32.cpu
index e372b584e919..80177ec052f0 100644
--- a/arch/x86/Makefile_32.cpu
+++ b/arch/x86/Makefile_32.cpu
@@ -28,7 +28,6 @@ cflags-$(CONFIG_MK8)		+= $(call cc-option,-march=k8,-march=athlon)
 cflags-$(CONFIG_MCRUSOE)	+= -march=i686 $(align)-functions=0 $(align)-jumps=0 $(align)-loops=0
 cflags-$(CONFIG_MEFFICEON)	+= -march=i686 $(call tune,pentium3) $(align)-functions=0 $(align)-jumps=0 $(align)-loops=0
 cflags-$(CONFIG_MWINCHIPC6)	+= $(call cc-option,-march=winchip-c6,-march=i586)
-cflags-$(CONFIG_MWINCHIP2)	+= $(call cc-option,-march=winchip2,-march=i586)
 cflags-$(CONFIG_MWINCHIP3D)	+= $(call cc-option,-march=winchip2,-march=i586)
 cflags-$(CONFIG_MCYRIXIII)	+= $(call cc-option,-march=c3,-march=i486) $(align)-functions=0 $(align)-jumps=0 $(align)-loops=0
 cflags-$(CONFIG_MVIAC3_2)	+= $(call cc-option,-march=c3-2,-march=i686)
@@ -45,3 +44,8 @@ cflags-$(CONFIG_MGEODEGX1)	+= -march=pentium-mmx
 # cpu entries
 cflags-$(CONFIG_X86_GENERIC) 	+= $(call tune,generic,$(call tune,i686))
 
+# Bug fix for binutils: this option is required in order to keep
+# binutils from generating NOPL instructions against our will.
+ifneq ($(CONFIG_X86_P6_NOP),y)
+cflags-y			+= $(call cc-option,-Wa$(comma)-mtune=generic32,)
+endif
diff --git a/arch/x86/boot/Makefile b/arch/x86/boot/Makefile
index 7ee102f9c4f8..cd48c7210016 100644
--- a/arch/x86/boot/Makefile
+++ b/arch/x86/boot/Makefile
@@ -72,9 +72,7 @@ KBUILD_CFLAGS	:= $(LINUXINCLUDE) -g -Os -D_SETUP -D__KERNEL__ \
 KBUILD_CFLAGS +=   $(call cc-option,-m32)
 KBUILD_AFLAGS	:= $(KBUILD_CFLAGS) -D__ASSEMBLY__
 
-$(obj)/zImage:  IMAGE_OFFSET := 0x1000
 $(obj)/zImage:  asflags-y := $(SVGA_MODE) $(RAMDISK)
-$(obj)/bzImage: IMAGE_OFFSET := 0x100000
 $(obj)/bzImage: ccflags-y := -D__BIG_KERNEL__
 $(obj)/bzImage: asflags-y := $(SVGA_MODE) $(RAMDISK) -D__BIG_KERNEL__
 $(obj)/bzImage: BUILDFLAGS   := -b
@@ -117,7 +115,7 @@ $(obj)/setup.bin: $(obj)/setup.elf FORCE
 	$(call if_changed,objcopy)
 
 $(obj)/compressed/vmlinux: FORCE
-	$(Q)$(MAKE) $(build)=$(obj)/compressed IMAGE_OFFSET=$(IMAGE_OFFSET) $@
+	$(Q)$(MAKE) $(build)=$(obj)/compressed $@
 
 # Set this if you want to pass append arguments to the zdisk/fdimage/isoimage kernel
 FDARGS =
@@ -181,6 +179,7 @@ isoimage: $(BOOTIMAGE)
 	mkisofs -J -r -o $(obj)/image.iso -b isolinux.bin -c boot.cat \
 		-no-emul-boot -boot-load-size 4 -boot-info-table \
 		$(obj)/isoimage
+	isohybrid $(obj)/image.iso 2>/dev/null || true
 	rm -rf $(obj)/isoimage
 
 zlilo: $(BOOTIMAGE)
diff --git a/arch/x86/boot/compressed/Makefile b/arch/x86/boot/compressed/Makefile
index 92fdd35bd93e..1771c804e02f 100644
--- a/arch/x86/boot/compressed/Makefile
+++ b/arch/x86/boot/compressed/Makefile
@@ -27,9 +27,8 @@ $(obj)/vmlinux.bin: vmlinux FORCE
 	$(call if_changed,objcopy)
 
 
-ifeq ($(CONFIG_X86_32),y)
-targets += vmlinux.bin.all vmlinux.relocs
-hostprogs-y := relocs
+targets += vmlinux.bin.all vmlinux.relocs relocs
+hostprogs-$(CONFIG_X86_32) += relocs
 
 quiet_cmd_relocs = RELOCS  $@
       cmd_relocs = $(obj)/relocs $< > $@;$(obj)/relocs --abs-relocs $<
@@ -43,6 +42,8 @@ quiet_cmd_relocbin = BUILD   $@
 $(obj)/vmlinux.bin.all: $(vmlinux.bin.all-y) FORCE
 	$(call if_changed,relocbin)
 
+ifeq ($(CONFIG_X86_32),y)
+
 ifdef CONFIG_RELOCATABLE
 $(obj)/vmlinux.bin.gz: $(obj)/vmlinux.bin.all FORCE
 	$(call if_changed,gzip)
@@ -59,6 +60,5 @@ $(obj)/vmlinux.bin.gz: $(obj)/vmlinux.bin FORCE
 LDFLAGS_piggy.o := -r --format binary --oformat elf64-x86-64 -T
 endif
 
-
 $(obj)/piggy.o: $(obj)/vmlinux.scr $(obj)/vmlinux.bin.gz FORCE
 	$(call if_changed,ld)
diff --git a/arch/x86/boot/cpu.c b/arch/x86/boot/cpu.c
index 75298fe2edca..6ec6bb6e9957 100644
--- a/arch/x86/boot/cpu.c
+++ b/arch/x86/boot/cpu.c
@@ -59,17 +59,18 @@ int validate_cpu(void)
 			u32 e = err_flags[i];
 
 			for (j = 0; j < 32; j++) {
-				int n = (i << 5)+j;
-				if (*msg_strs < n) {
+				if (msg_strs[0] < i ||
+				    (msg_strs[0] == i && msg_strs[1] < j)) {
 					/* Skip to the next string */
-					do {
-						msg_strs++;
-					} while (*msg_strs);
-					msg_strs++;
+					msg_strs += 2;
+					while (*msg_strs++)
+						;
 				}
 				if (e & 1) {
-					if (*msg_strs == n && msg_strs[1])
-						printf("%s ", msg_strs+1);
+					if (msg_strs[0] == i &&
+					    msg_strs[1] == j &&
+					    msg_strs[2])
+						printf("%s ", msg_strs+2);
 					else
 						printf("%d:%d ", i, j);
 				}
diff --git a/arch/x86/boot/edd.c b/arch/x86/boot/edd.c
index d93cbc6464d0..1aae8f3e5ca1 100644
--- a/arch/x86/boot/edd.c
+++ b/arch/x86/boot/edd.c
@@ -41,6 +41,7 @@ static u32 read_mbr_sig(u8 devno, struct edd_info *ei, u32 *mbrsig)
 	char *mbrbuf_ptr, *mbrbuf_end;
 	u32 buf_base, mbr_base;
 	extern char _end[];
+	u16 mbr_magic;
 
 	sector_size = ei->params.bytes_per_sector;
 	if (!sector_size)
@@ -58,11 +59,15 @@ static u32 read_mbr_sig(u8 devno, struct edd_info *ei, u32 *mbrsig)
 	if (mbrbuf_end > (char *)(size_t)boot_params.hdr.heap_end_ptr)
 		return -1;
 
+	memset(mbrbuf_ptr, 0, sector_size);
 	if (read_mbr(devno, mbrbuf_ptr))
 		return -1;
 
 	*mbrsig = *(u32 *)&mbrbuf_ptr[EDD_MBR_SIG_OFFSET];
-	return 0;
+	mbr_magic = *(u16 *)&mbrbuf_ptr[510];
+
+	/* check for valid MBR magic */
+	return mbr_magic == 0xAA55 ? 0 : -1;
 }
 
 static int get_edd_info(u8 devno, struct edd_info *ei)
diff --git a/arch/x86/boot/mkcpustr.c b/arch/x86/boot/mkcpustr.c
index bbe76953bae9..8ef60f20b371 100644
--- a/arch/x86/boot/mkcpustr.c
+++ b/arch/x86/boot/mkcpustr.c
@@ -15,33 +15,33 @@
 
 #include <stdio.h>
 
-#include "../kernel/cpu/feature_names.c"
-
-#if NCAPFLAGS > 8
-# error "Need to adjust the boot code handling of CPUID strings"
-#endif
+#include "../kernel/cpu/capflags.c"
 
 int main(void)
 {
-	int i;
+	int i, j;
 	const char *str;
 
 	printf("static const char x86_cap_strs[] = \n");
 
-	for (i = 0; i < NCAPINTS*32; i++) {
-		str = x86_cap_flags[i];
-
-		if (i == NCAPINTS*32-1) {
-			/* The last entry must be unconditional; this
-			   also consumes the compiler-added null character */
-			if (!str)
-				str = "";
-			printf("\t\"\\x%02x\"\"%s\"\n", i, str);
-		} else if (str) {
-			printf("#if REQUIRED_MASK%d & (1 << %d)\n"
-			       "\t\"\\x%02x\"\"%s\\0\"\n"
-			       "#endif\n",
-			       i >> 5, i & 31, i, str);
+	for (i = 0; i < NCAPINTS; i++) {
+		for (j = 0; j < 32; j++) {
+			str = x86_cap_flags[i*32+j];
+
+			if (i == NCAPINTS-1 && j == 31) {
+				/* The last entry must be unconditional; this
+				   also consumes the compiler-added null
+				   character */
+				if (!str)
+					str = "";
+				printf("\t\"\\x%02x\\x%02x\"\"%s\"\n",
+				       i, j, str);
+			} else if (str) {
+				printf("#if REQUIRED_MASK%d & (1 << %d)\n"
+				       "\t\"\\x%02x\\x%02x\"\"%s\\0\"\n"
+				       "#endif\n",
+				       i, j, i, j, str);
+			}
 		}
 	}
 	printf("\t;\n");
diff --git a/arch/x86/boot/video-vesa.c b/arch/x86/boot/video-vesa.c
index 401ad998ad08..1e6fe0214c85 100644
--- a/arch/x86/boot/video-vesa.c
+++ b/arch/x86/boot/video-vesa.c
@@ -224,7 +224,7 @@ static void vesa_store_pm_info(void)
 static void vesa_store_mode_params_graphics(void)
 {
 	/* Tell the kernel we're in VESA graphics mode */
-	boot_params.screen_info.orig_video_isVGA = 0x23;
+	boot_params.screen_info.orig_video_isVGA = VIDEO_TYPE_VLFB;
 
 	/* Mode parameters */
 	boot_params.screen_info.vesa_attributes = vminfo.mode_attr;
diff --git a/arch/x86/configs/i386_defconfig b/arch/x86/configs/i386_defconfig
index ef9a52005ec9..52d0359719d7 100644
--- a/arch/x86/configs/i386_defconfig
+++ b/arch/x86/configs/i386_defconfig
@@ -213,7 +213,6 @@ CONFIG_M686=y
 # CONFIG_MCRUSOE is not set
 # CONFIG_MEFFICEON is not set
 # CONFIG_MWINCHIPC6 is not set
-# CONFIG_MWINCHIP2 is not set
 # CONFIG_MWINCHIP3D is not set
 # CONFIG_MGEODEGX1 is not set
 # CONFIG_MGEODE_LX is not set
@@ -1535,7 +1534,6 @@ CONFIG_BACKLIGHT_CLASS_DEVICE=y
 CONFIG_VGA_CONSOLE=y
 CONFIG_VGACON_SOFT_SCROLLBACK=y
 CONFIG_VGACON_SOFT_SCROLLBACK_SIZE=64
-CONFIG_VIDEO_SELECT=y
 CONFIG_DUMMY_CONSOLE=y
 # CONFIG_FRAMEBUFFER_CONSOLE is not set
 CONFIG_LOGO=y
diff --git a/arch/x86/configs/x86_64_defconfig b/arch/x86/configs/x86_64_defconfig
index e620ea6e2a7a..f0a03d7a7d63 100644
--- a/arch/x86/configs/x86_64_defconfig
+++ b/arch/x86/configs/x86_64_defconfig
@@ -210,7 +210,6 @@ CONFIG_X86_PC=y
 # CONFIG_MCRUSOE is not set
 # CONFIG_MEFFICEON is not set
 # CONFIG_MWINCHIPC6 is not set
-# CONFIG_MWINCHIP2 is not set
 # CONFIG_MWINCHIP3D is not set
 # CONFIG_MGEODEGX1 is not set
 # CONFIG_MGEODE_LX is not set
@@ -1505,7 +1504,6 @@ CONFIG_BACKLIGHT_CLASS_DEVICE=y
 CONFIG_VGA_CONSOLE=y
 CONFIG_VGACON_SOFT_SCROLLBACK=y
 CONFIG_VGACON_SOFT_SCROLLBACK_SIZE=64
-CONFIG_VIDEO_SELECT=y
 CONFIG_DUMMY_CONSOLE=y
 # CONFIG_FRAMEBUFFER_CONSOLE is not set
 CONFIG_LOGO=y
diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c
index f1a2ac777faf..4bc02b23674b 100644
--- a/arch/x86/ia32/ia32_signal.c
+++ b/arch/x86/ia32/ia32_signal.c
@@ -179,9 +179,10 @@ struct sigframe
 	u32 pretcode;
 	int sig;
 	struct sigcontext_ia32 sc;
-	struct _fpstate_ia32 fpstate;
+	struct _fpstate_ia32 fpstate_unused; /* look at kernel/sigframe.h */
 	unsigned int extramask[_COMPAT_NSIG_WORDS-1];
 	char retcode[8];
+	/* fp state follows here */
 };
 
 struct rt_sigframe
@@ -192,8 +193,8 @@ struct rt_sigframe
 	u32 puc;
 	compat_siginfo_t info;
 	struct ucontext_ia32 uc;
-	struct _fpstate_ia32 fpstate;
 	char retcode[8];
+	/* fp state follows here */
 };
 
 #define COPY(x)		{ 		\
@@ -215,7 +216,7 @@ static int ia32_restore_sigcontext(struct pt_regs *regs,
 				   unsigned int *peax)
 {
 	unsigned int tmpflags, gs, oldgs, err = 0;
-	struct _fpstate_ia32 __user *buf;
+	void __user *buf;
 	u32 tmp;
 
 	/* Always make any pending restarted system calls return -EINTR */
@@ -259,26 +260,12 @@ static int ia32_restore_sigcontext(struct pt_regs *regs,
 
 	err |= __get_user(tmp, &sc->fpstate);
 	buf = compat_ptr(tmp);
-	if (buf) {
-		if (!access_ok(VERIFY_READ, buf, sizeof(*buf)))
-			goto badframe;
-		err |= restore_i387_ia32(buf);
-	} else {
-		struct task_struct *me = current;
-
-		if (used_math()) {
-			clear_fpu(me);
-			clear_used_math();
-		}
-	}
+	err |= restore_i387_xstate_ia32(buf);
 
 	err |= __get_user(tmp, &sc->ax);
 	*peax = tmp;
 
 	return err;
-
-badframe:
-	return 1;
 }
 
 asmlinkage long sys32_sigreturn(struct pt_regs *regs)
@@ -350,7 +337,7 @@ badframe:
  */
 
 static int ia32_setup_sigcontext(struct sigcontext_ia32 __user *sc,
-				 struct _fpstate_ia32 __user *fpstate,
+				 void __user *fpstate,
 				 struct pt_regs *regs, unsigned int mask)
 {
 	int tmp, err = 0;
@@ -364,31 +351,28 @@ static int ia32_setup_sigcontext(struct sigcontext_ia32 __user *sc,
 	savesegment(es, tmp);
 	err |= __put_user(tmp, (unsigned int __user *)&sc->es);
 
-	err |= __put_user((u32)regs->di, &sc->di);
-	err |= __put_user((u32)regs->si, &sc->si);
-	err |= __put_user((u32)regs->bp, &sc->bp);
-	err |= __put_user((u32)regs->sp, &sc->sp);
-	err |= __put_user((u32)regs->bx, &sc->bx);
-	err |= __put_user((u32)regs->dx, &sc->dx);
-	err |= __put_user((u32)regs->cx, &sc->cx);
-	err |= __put_user((u32)regs->ax, &sc->ax);
-	err |= __put_user((u32)regs->cs, &sc->cs);
-	err |= __put_user((u32)regs->ss, &sc->ss);
+	err |= __put_user(regs->di, &sc->di);
+	err |= __put_user(regs->si, &sc->si);
+	err |= __put_user(regs->bp, &sc->bp);
+	err |= __put_user(regs->sp, &sc->sp);
+	err |= __put_user(regs->bx, &sc->bx);
+	err |= __put_user(regs->dx, &sc->dx);
+	err |= __put_user(regs->cx, &sc->cx);
+	err |= __put_user(regs->ax, &sc->ax);
+	err |= __put_user(regs->cs, &sc->cs);
+	err |= __put_user(regs->ss, &sc->ss);
 	err |= __put_user(current->thread.trap_no, &sc->trapno);
 	err |= __put_user(current->thread.error_code, &sc->err);
-	err |= __put_user((u32)regs->ip, &sc->ip);
-	err |= __put_user((u32)regs->flags, &sc->flags);
-	err |= __put_user((u32)regs->sp, &sc->sp_at_signal);
+	err |= __put_user(regs->ip, &sc->ip);
+	err |= __put_user(regs->flags, &sc->flags);
+	err |= __put_user(regs->sp, &sc->sp_at_signal);
 
-	tmp = save_i387_ia32(fpstate);
+	tmp = save_i387_xstate_ia32(fpstate);
 	if (tmp < 0)
 		err = -EFAULT;
-	else {
-		clear_used_math();
-		stts();
+	else
 		err |= __put_user(ptr_to_compat(tmp ? fpstate : NULL),
 					&sc->fpstate);
-	}
 
 	/* non-iBCS2 extensions.. */
 	err |= __put_user(mask, &sc->oldmask);
@@ -401,7 +385,8 @@ static int ia32_setup_sigcontext(struct sigcontext_ia32 __user *sc,
  * Determine which stack to use..
  */
 static void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs *regs,
-				 size_t frame_size)
+				 size_t frame_size,
+				 void **fpstate)
 {
 	unsigned long sp;
 
@@ -420,6 +405,11 @@ static void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs *regs,
 		 ka->sa.sa_restorer)
 		sp = (unsigned long) ka->sa.sa_restorer;
 
+	if (used_math()) {
+		sp = sp - sig_xstate_ia32_size;
+		*fpstate = (struct _fpstate_ia32 *) sp;
+	}
+
 	sp -= frame_size;
 	/* Align the stack pointer according to the i386 ABI,
 	 * i.e. so that on function entry ((sp + 4) & 15) == 0. */
@@ -433,6 +423,7 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka,
 	struct sigframe __user *frame;
 	void __user *restorer;
 	int err = 0;
+	void __user *fpstate = NULL;
 
 	/* copy_to_user optimizes that into a single 8 byte store */
 	static const struct {
@@ -447,25 +438,21 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka,
 		0,
 	};
 
-	frame = get_sigframe(ka, regs, sizeof(*frame));
+	frame = get_sigframe(ka, regs, sizeof(*frame), &fpstate);
 
 	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
-		goto give_sigsegv;
+		return -EFAULT;
 
-	err |= __put_user(sig, &frame->sig);
-	if (err)
-		goto give_sigsegv;
+	if (__put_user(sig, &frame->sig))
+		return -EFAULT;
 
-	err |= ia32_setup_sigcontext(&frame->sc, &frame->fpstate, regs,
-					set->sig[0]);
-	if (err)
-		goto give_sigsegv;
+	if (ia32_setup_sigcontext(&frame->sc, fpstate, regs, set->sig[0]))
+		return -EFAULT;
 
 	if (_COMPAT_NSIG_WORDS > 1) {
-		err |= __copy_to_user(frame->extramask, &set->sig[1],
-				      sizeof(frame->extramask));
-		if (err)
-			goto give_sigsegv;
+		if (__copy_to_user(frame->extramask, &set->sig[1],
+				   sizeof(frame->extramask)))
+			return -EFAULT;
 	}
 
 	if (ka->sa.sa_flags & SA_RESTORER) {
@@ -486,7 +473,7 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka,
 	 */
 	err |= __copy_to_user(frame->retcode, &code, 8);
 	if (err)
-		goto give_sigsegv;
+		return -EFAULT;
 
 	/* Set up registers for signal handler */
 	regs->sp = (unsigned long) frame;
@@ -509,10 +496,6 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka,
 #endif
 
 	return 0;
-
-give_sigsegv:
-	force_sigsegv(sig, current);
-	return -EFAULT;
 }
 
 int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
@@ -521,6 +504,7 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	struct rt_sigframe __user *frame;
 	void __user *restorer;
 	int err = 0;
+	void __user *fpstate = NULL;
 
 	/* __copy_to_user optimizes that into a single 8 byte store */
 	static const struct {
@@ -536,30 +520,33 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		0,
 	};
 
-	frame = get_sigframe(ka, regs, sizeof(*frame));
+	frame = get_sigframe(ka, regs, sizeof(*frame), &fpstate);
 
 	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
-		goto give_sigsegv;
+		return -EFAULT;
 
 	err |= __put_user(sig, &frame->sig);
 	err |= __put_user(ptr_to_compat(&frame->info), &frame->pinfo);
 	err |= __put_user(ptr_to_compat(&frame->uc), &frame->puc);
 	err |= copy_siginfo_to_user32(&frame->info, info);
 	if (err)
-		goto give_sigsegv;
+		return -EFAULT;
 
 	/* Create the ucontext.  */
-	err |= __put_user(0, &frame->uc.uc_flags);
+	if (cpu_has_xsave)
+		err |= __put_user(UC_FP_XSTATE, &frame->uc.uc_flags);
+	else
+		err |= __put_user(0, &frame->uc.uc_flags);
 	err |= __put_user(0, &frame->uc.uc_link);
 	err |= __put_user(current->sas_ss_sp, &frame->uc.uc_stack.ss_sp);
 	err |= __put_user(sas_ss_flags(regs->sp),
 			  &frame->uc.uc_stack.ss_flags);
 	err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size);
-	err |= ia32_setup_sigcontext(&frame->uc.uc_mcontext, &frame->fpstate,
+	err |= ia32_setup_sigcontext(&frame->uc.uc_mcontext, fpstate,
 				     regs, set->sig[0]);
 	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
 	if (err)
-		goto give_sigsegv;
+		return -EFAULT;
 
 	if (ka->sa.sa_flags & SA_RESTORER)
 		restorer = ka->sa.sa_restorer;
@@ -574,7 +561,7 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	 */
 	err |= __copy_to_user(frame->retcode, &code, 8);
 	if (err)
-		goto give_sigsegv;
+		return -EFAULT;
 
 	/* Set up registers for signal handler */
 	regs->sp = (unsigned long) frame;
@@ -602,8 +589,4 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 #endif
 
 	return 0;
-
-give_sigsegv:
-	force_sigsegv(sig, current);
-	return -EFAULT;
 }
diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
index ffc1bb4fed7d..eb4314768bf7 100644
--- a/arch/x86/ia32/ia32entry.S
+++ b/arch/x86/ia32/ia32entry.S
@@ -39,11 +39,11 @@
 	.endm 
 
 	/* clobbers %eax */	
-	.macro  CLEAR_RREGS
+	.macro  CLEAR_RREGS _r9=rax
 	xorl 	%eax,%eax
 	movq	%rax,R11(%rsp)
 	movq	%rax,R10(%rsp)
-	movq	%rax,R9(%rsp)
+	movq	%\_r9,R9(%rsp)
 	movq	%rax,R8(%rsp)
 	.endm
 
@@ -52,11 +52,10 @@
 	 * We don't reload %eax because syscall_trace_enter() returned
 	 * the value it wants us to use in the table lookup.
 	 */
-	.macro LOAD_ARGS32 offset
-	movl \offset(%rsp),%r11d
-	movl \offset+8(%rsp),%r10d
+	.macro LOAD_ARGS32 offset, _r9=0
+	.if \_r9
 	movl \offset+16(%rsp),%r9d
-	movl \offset+24(%rsp),%r8d
+	.endif
 	movl \offset+40(%rsp),%ecx
 	movl \offset+48(%rsp),%edx
 	movl \offset+56(%rsp),%esi
@@ -145,7 +144,7 @@ ENTRY(ia32_sysenter_target)
 	SAVE_ARGS 0,0,1
  	/* no need to do an access_ok check here because rbp has been
  	   32bit zero extended */ 
-1:	movl	(%rbp),%r9d
+1:	movl	(%rbp),%ebp
  	.section __ex_table,"a"
  	.quad 1b,ia32_badarg
  	.previous	
@@ -157,7 +156,7 @@ ENTRY(ia32_sysenter_target)
 	cmpl	$(IA32_NR_syscalls-1),%eax
 	ja	ia32_badsys
 sysenter_do_call:
-	IA32_ARG_FIXUP 1
+	IA32_ARG_FIXUP
 sysenter_dispatch:
 	call	*ia32_sys_call_table(,%rax,8)
 	movq	%rax,RAX-ARGOFFSET(%rsp)
@@ -234,20 +233,17 @@ sysexit_audit:
 #endif
 
 sysenter_tracesys:
-	xchgl	%r9d,%ebp
 #ifdef CONFIG_AUDITSYSCALL
 	testl	$(_TIF_WORK_SYSCALL_ENTRY & ~_TIF_SYSCALL_AUDIT),TI_flags(%r10)
 	jz	sysenter_auditsys
 #endif
 	SAVE_REST
 	CLEAR_RREGS
-	movq	%r9,R9(%rsp)
 	movq	$-ENOSYS,RAX(%rsp)/* ptrace can change this for a bad syscall */
 	movq	%rsp,%rdi        /* &pt_regs -> arg1 */
 	call	syscall_trace_enter
 	LOAD_ARGS32 ARGOFFSET  /* reload args from stack in case ptrace changed it */
 	RESTORE_REST
-	xchgl	%ebp,%r9d
 	cmpl	$(IA32_NR_syscalls-1),%eax
 	ja	int_ret_from_sys_call /* sysenter_tracesys has set RAX(%rsp) */
 	jmp	sysenter_do_call
@@ -314,9 +310,9 @@ ENTRY(ia32_cstar_target)
 	testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r10)
 	CFI_REMEMBER_STATE
 	jnz   cstar_tracesys
-cstar_do_call:	
 	cmpl $IA32_NR_syscalls-1,%eax
 	ja  ia32_badsys
+cstar_do_call:
 	IA32_ARG_FIXUP 1
 cstar_dispatch:
 	call *ia32_sys_call_table(,%rax,8)
@@ -357,15 +353,13 @@ cstar_tracesys:
 #endif
 	xchgl %r9d,%ebp
 	SAVE_REST
-	CLEAR_RREGS
-	movq %r9,R9(%rsp)
+	CLEAR_RREGS r9
 	movq $-ENOSYS,RAX(%rsp)	/* ptrace can change this for a bad syscall */
 	movq %rsp,%rdi        /* &pt_regs -> arg1 */
 	call syscall_trace_enter
-	LOAD_ARGS32 ARGOFFSET  /* reload args from stack in case ptrace changed it */
+	LOAD_ARGS32 ARGOFFSET, 1  /* reload args from stack in case ptrace changed it */
 	RESTORE_REST
 	xchgl %ebp,%r9d
-	movl RSP-ARGOFFSET(%rsp), %r8d
 	cmpl $(IA32_NR_syscalls-1),%eax
 	ja int_ret_from_sys_call /* cstar_tracesys has set RAX(%rsp) */
 	jmp cstar_do_call
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 3db651fc8ec5..0d41f0343dc0 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -10,7 +10,7 @@ ifdef CONFIG_FTRACE
 # Do not profile debug and lowlevel utilities
 CFLAGS_REMOVE_tsc.o = -pg
 CFLAGS_REMOVE_rtc.o = -pg
-CFLAGS_REMOVE_paravirt.o = -pg
+CFLAGS_REMOVE_paravirt-spinlocks.o = -pg
 endif
 
 #
@@ -23,7 +23,7 @@ CFLAGS_hpet.o		:= $(nostackp)
 CFLAGS_tsc.o		:= $(nostackp)
 
 obj-y			:= process_$(BITS).o signal_$(BITS).o entry_$(BITS).o
-obj-y			+= traps_$(BITS).o irq_$(BITS).o
+obj-y			+= traps.o irq_$(BITS).o dumpstack_$(BITS).o
 obj-y			+= time_$(BITS).o ioport.o ldt.o
 obj-y			+= setup.o i8259.o irqinit_$(BITS).o setup_percpu.o
 obj-$(CONFIG_X86_VISWS)	+= visws_quirks.o
@@ -38,7 +38,7 @@ obj-y			+= tsc.o io_delay.o rtc.o
 
 obj-$(CONFIG_X86_TRAMPOLINE)	+= trampoline.o
 obj-y				+= process.o
-obj-y				+= i387.o
+obj-y				+= i387.o xsave.o
 obj-y				+= ptrace.o
 obj-y				+= ds.o
 obj-$(CONFIG_X86_32)		+= tls.o
@@ -51,7 +51,6 @@ obj-$(CONFIG_X86_BIOS_REBOOT)	+= reboot.o
 obj-$(CONFIG_MCA)		+= mca_32.o
 obj-$(CONFIG_X86_MSR)		+= msr.o
 obj-$(CONFIG_X86_CPUID)		+= cpuid.o
-obj-$(CONFIG_MICROCODE)		+= microcode.o
 obj-$(CONFIG_PCI)		+= early-quirks.o
 apm-y				:= apm_32.o
 obj-$(CONFIG_APM)		+= apm.o
@@ -69,6 +68,7 @@ obj-$(CONFIG_KEXEC)		+= machine_kexec_$(BITS).o
 obj-$(CONFIG_KEXEC)		+= relocate_kernel_$(BITS).o crash.o
 obj-$(CONFIG_CRASH_DUMP)	+= crash_dump_$(BITS).o
 obj-$(CONFIG_X86_NUMAQ)		+= numaq_32.o
+obj-$(CONFIG_X86_ES7000)	+= es7000_32.o
 obj-$(CONFIG_X86_SUMMIT_NUMA)	+= summit_32.o
 obj-y				+= vsmp_64.o
 obj-$(CONFIG_KPROBES)		+= kprobes.o
@@ -89,7 +89,7 @@ obj-$(CONFIG_DEBUG_NX_TEST)	+= test_nx.o
 obj-$(CONFIG_VMI)		+= vmi_32.o vmiclock_32.o
 obj-$(CONFIG_KVM_GUEST)		+= kvm.o
 obj-$(CONFIG_KVM_CLOCK)		+= kvmclock.o
-obj-$(CONFIG_PARAVIRT)		+= paravirt.o paravirt_patch_$(BITS).o
+obj-$(CONFIG_PARAVIRT)		+= paravirt.o paravirt_patch_$(BITS).o paravirt-spinlocks.o
 obj-$(CONFIG_PARAVIRT_CLOCK)	+= pvclock.o
 
 obj-$(CONFIG_PCSPKR_PLATFORM)	+= pcspeaker.o
@@ -99,11 +99,18 @@ scx200-y			+= scx200_32.o
 
 obj-$(CONFIG_OLPC)		+= olpc.o
 
+microcode-y				:= microcode_core.o
+microcode-$(CONFIG_MICROCODE_INTEL)	+= microcode_intel.o
+microcode-$(CONFIG_MICROCODE_AMD)	+= microcode_amd.o
+obj-$(CONFIG_MICROCODE)			+= microcode.o
+
 ###
 # 64 bit specific files
 ifeq ($(CONFIG_X86_64),y)
         obj-y				+= genapic_64.o genapic_flat_64.o genx2apic_uv_x.o tlb_uv.o
 	obj-y				+= bios_uv.o
+        obj-y				+= genx2apic_cluster.o
+        obj-y				+= genx2apic_phys.o
         obj-$(CONFIG_X86_PM_TIMER)	+= pmtimer_64.o
         obj-$(CONFIG_AUDIT)		+= audit_64.o
 
diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 7d40ef7b36e3..eb875cdc7367 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -252,10 +252,8 @@ static void __cpuinit acpi_register_lapic(int id, u8 enabled)
 		return;
 	}
 
-#ifdef CONFIG_X86_32
 	if (boot_cpu_physical_apicid != -1U)
 		ver = apic_version[boot_cpu_physical_apicid];
-#endif
 
 	generic_processor_info(id, ver);
 }
@@ -774,11 +772,9 @@ static void __init acpi_register_lapic_address(unsigned long address)
 
 	set_fixmap_nocache(FIX_APIC_BASE, address);
 	if (boot_cpu_physical_apicid == -1U) {
-		boot_cpu_physical_apicid  = GET_APIC_ID(read_apic_id());
-#ifdef CONFIG_X86_32
+		boot_cpu_physical_apicid  = read_apic_id();
 		apic_version[boot_cpu_physical_apicid] =
 			 GET_APIC_VERSION(apic_read(APIC_LVR));
-#endif
 	}
 }
 
@@ -1350,7 +1346,9 @@ static void __init acpi_process_madt(void)
 				acpi_ioapic = 1;
 
 				smp_found_config = 1;
+#ifdef CONFIG_X86_32
 				setup_apic_routing();
+#endif
 			}
 		}
 		if (error == -EINVAL) {
@@ -1420,8 +1418,16 @@ static int __init force_acpi_ht(const struct dmi_system_id *d)
  */
 static int __init dmi_ignore_irq0_timer_override(const struct dmi_system_id *d)
 {
-	pr_notice("%s detected: Ignoring BIOS IRQ0 pin2 override\n", d->ident);
-	acpi_skip_timer_override = 1;
+	/*
+	 * The ati_ixp4x0_rev() early PCI quirk should have set
+	 * the acpi_skip_timer_override flag already:
+	 */
+	if (!acpi_skip_timer_override) {
+		WARN(1, KERN_ERR "ati_ixp4x0 quirk not complete.\n");
+		pr_notice("%s detected: Ignoring BIOS IRQ0 pin2 override\n",
+			d->ident);
+		acpi_skip_timer_override = 1;
+	}
 	return 0;
 }
 
diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index fb04e49776ba..a84ac7b570e6 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -444,7 +444,7 @@ void __init alternative_instructions(void)
 					    _text, _etext);
 
 		/* Only switch to UP mode if we don't immediately boot others */
-		if (num_possible_cpus() == 1 || setup_max_cpus <= 1)
+		if (num_present_cpus() == 1 || setup_max_cpus <= 1)
 			alternatives_smp_switch(0);
 	}
 #endif
diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
index 148fcfe22f17..4cd8083c58be 100644
--- a/arch/x86/kernel/amd_iommu_init.c
+++ b/arch/x86/kernel/amd_iommu_init.c
@@ -723,9 +723,7 @@ static int __init init_iommu_one(struct amd_iommu *iommu, struct ivhd_header *h)
 	init_iommu_from_acpi(iommu, h);
 	init_iommu_devices(iommu);
 
-	pci_enable_device(iommu->dev);
-
-	return 0;
+	return pci_enable_device(iommu->dev);
 }
 
 /*
diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c
index f88bd0d982b0..21c831d96af3 100644
--- a/arch/x86/kernel/apic_32.c
+++ b/arch/x86/kernel/apic_32.c
@@ -60,10 +60,8 @@ unsigned long mp_lapic_addr;
 static int force_enable_local_apic;
 int disable_apic;
 
-/* Local APIC timer verification ok */
-static int local_apic_timer_verify_ok;
 /* Disable local APIC timer from the kernel commandline or via dmi quirk */
-static int local_apic_timer_disabled;
+static int disable_apic_timer __cpuinitdata;
 /* Local APIC timer works in C2 */
 int local_apic_timer_c2_ok;
 EXPORT_SYMBOL_GPL(local_apic_timer_c2_ok);
@@ -130,7 +128,11 @@ static inline int lapic_get_version(void)
  */
 static inline int lapic_is_integrated(void)
 {
+#ifdef CONFIG_X86_64
+	return 1;
+#else
 	return APIC_INTEGRATED(lapic_get_version());
+#endif
 }
 
 /*
@@ -145,13 +147,18 @@ static int modern_apic(void)
 	return lapic_get_version() >= 0x14;
 }
 
-void apic_wait_icr_idle(void)
+/*
+ * Paravirt kernels also might be using these below ops. So we still
+ * use generic apic_read()/apic_write(), which might be pointing to different
+ * ops in PARAVIRT case.
+ */
+void xapic_wait_icr_idle(void)
 {
 	while (apic_read(APIC_ICR) & APIC_ICR_BUSY)
 		cpu_relax();
 }
 
-u32 safe_apic_wait_icr_idle(void)
+u32 safe_xapic_wait_icr_idle(void)
 {
 	u32 send_status;
 	int timeout;
@@ -167,16 +174,48 @@ u32 safe_apic_wait_icr_idle(void)
 	return send_status;
 }
 
+void xapic_icr_write(u32 low, u32 id)
+{
+	apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(id));
+	apic_write(APIC_ICR, low);
+}
+
+u64 xapic_icr_read(void)
+{
+	u32 icr1, icr2;
+
+	icr2 = apic_read(APIC_ICR2);
+	icr1 = apic_read(APIC_ICR);
+
+	return icr1 | ((u64)icr2 << 32);
+}
+
+static struct apic_ops xapic_ops = {
+	.read = native_apic_mem_read,
+	.write = native_apic_mem_write,
+	.icr_read = xapic_icr_read,
+	.icr_write = xapic_icr_write,
+	.wait_icr_idle = xapic_wait_icr_idle,
+	.safe_wait_icr_idle = safe_xapic_wait_icr_idle,
+};
+
+struct apic_ops __read_mostly *apic_ops = &xapic_ops;
+EXPORT_SYMBOL_GPL(apic_ops);
+
 /**
  * enable_NMI_through_LVT0 - enable NMI through local vector table 0
  */
 void __cpuinit enable_NMI_through_LVT0(void)
 {
-	unsigned int v = APIC_DM_NMI;
+	unsigned int v;
 
-	/* Level triggered for 82489DX */
+	/* unmask and set to NMI */
+	v = APIC_DM_NMI;
+
+	/* Level triggered for 82489DX (32bit mode) */
 	if (!lapic_is_integrated())
 		v |= APIC_LVT_LEVEL_TRIGGER;
+
 	apic_write(APIC_LVT0, v);
 }
 
@@ -193,9 +232,13 @@ int get_physical_broadcast(void)
  */
 int lapic_get_maxlvt(void)
 {
-	unsigned int v = apic_read(APIC_LVR);
+	unsigned int v;
 
-	/* 82489DXs do not report # of LVT entries. */
+	v = apic_read(APIC_LVR);
+	/*
+	 * - we always have APIC integrated on 64bit mode
+	 * - 82489DXs do not report # of LVT entries
+	 */
 	return APIC_INTEGRATED(GET_APIC_VERSION(v)) ? GET_APIC_MAXLVT(v) : 2;
 }
 
@@ -203,8 +246,12 @@ int lapic_get_maxlvt(void)
  * Local APIC timer
  */
 
-/* Clock divisor is set to 16 */
+/* Clock divisor */
+#ifdef CONFG_X86_64
+#define APIC_DIVISOR 1
+#else
 #define APIC_DIVISOR 16
+#endif
 
 /*
  * This function sets up the local APIC timer, with a timeout of
@@ -212,6 +259,9 @@ int lapic_get_maxlvt(void)
  * this function twice on the boot CPU, once with a bogus timeout
  * value, second time for real. The other (noncalibrating) CPUs
  * call this function only once, with the real, calibrated value.
+ *
+ * We do reads before writes even if unnecessary, to get around the
+ * P5 APIC double write bug.
  */
 static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen)
 {
@@ -233,14 +283,48 @@ static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen)
 	 */
 	tmp_value = apic_read(APIC_TDCR);
 	apic_write(APIC_TDCR,
-		   (tmp_value & ~(APIC_TDR_DIV_1 | APIC_TDR_DIV_TMBASE)) |
-		   APIC_TDR_DIV_16);
+		(tmp_value & ~(APIC_TDR_DIV_1 | APIC_TDR_DIV_TMBASE)) |
+		APIC_TDR_DIV_16);
 
 	if (!oneshot)
 		apic_write(APIC_TMICT, clocks / APIC_DIVISOR);
 }
 
 /*
+ * Setup extended LVT, AMD specific (K8, family 10h)
+ *
+ * Vector mappings are hard coded. On K8 only offset 0 (APIC500) and
+ * MCE interrupts are supported. Thus MCE offset must be set to 0.
+ *
+ * If mask=1, the LVT entry does not generate interrupts while mask=0
+ * enables the vector. See also the BKDGs.
+ */
+
+#define APIC_EILVT_LVTOFF_MCE 0
+#define APIC_EILVT_LVTOFF_IBS 1
+
+static void setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask)
+{
+	unsigned long reg = (lvt_off << 4) + APIC_EILVT0;
+	unsigned int  v   = (mask << 16) | (msg_type << 8) | vector;
+
+	apic_write(reg, v);
+}
+
+u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask)
+{
+	setup_APIC_eilvt(APIC_EILVT_LVTOFF_MCE, vector, msg_type, mask);
+	return APIC_EILVT_LVTOFF_MCE;
+}
+
+u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask)
+{
+	setup_APIC_eilvt(APIC_EILVT_LVTOFF_IBS, vector, msg_type, mask);
+	return APIC_EILVT_LVTOFF_IBS;
+}
+EXPORT_SYMBOL_GPL(setup_APIC_eilvt_ibs);
+
+/*
  * Program the next event, relative to now
  */
 static int lapic_next_event(unsigned long delta,
@@ -259,8 +343,8 @@ static void lapic_timer_setup(enum clock_event_mode mode,
 	unsigned long flags;
 	unsigned int v;
 
-	/* Lapic used for broadcast ? */
-	if (!local_apic_timer_verify_ok)
+	/* Lapic used as dummy for broadcast ? */
+	if (evt->features & CLOCK_EVT_FEAT_DUMMY)
 		return;
 
 	local_irq_save(flags);
@@ -473,7 +557,7 @@ static int __init calibrate_APIC_clock(void)
 		return -1;
 	}
 
-	local_apic_timer_verify_ok = 1;
+	levt->features &= ~CLOCK_EVT_FEAT_DUMMY;
 
 	/* We trust the pm timer based calibration */
 	if (!pm_referenced) {
@@ -507,11 +591,11 @@ static int __init calibrate_APIC_clock(void)
 		if (deltaj >= LAPIC_CAL_LOOPS-2 && deltaj <= LAPIC_CAL_LOOPS+2)
 			apic_printk(APIC_VERBOSE, "... jiffies result ok\n");
 		else
-			local_apic_timer_verify_ok = 0;
+			levt->features |= CLOCK_EVT_FEAT_DUMMY;
 	} else
 		local_irq_enable();
 
-	if (!local_apic_timer_verify_ok) {
+	if (levt->features & CLOCK_EVT_FEAT_DUMMY) {
 		printk(KERN_WARNING
 		       "APIC timer disabled due to verification failure.\n");
 			return -1;
@@ -533,7 +617,8 @@ void __init setup_boot_APIC_clock(void)
 	 * timer as a dummy clock event source on SMP systems, so the
 	 * broadcast mechanism is used. On UP systems simply ignore it.
 	 */
-	if (local_apic_timer_disabled) {
+	if (disable_apic_timer) {
+		printk(KERN_INFO "Disabling APIC timer\n");
 		/* No broadcast on UP ! */
 		if (num_possible_cpus() > 1) {
 			lapic_clockevent.mult = 1;
@@ -602,7 +687,11 @@ static void local_apic_timer_interrupt(void)
 	/*
 	 * the NMI deadlock-detector uses this.
 	 */
+#ifdef CONFIG_X86_64
+	add_pda(apic_timer_irqs, 1);
+#else
 	per_cpu(irq_stat, cpu).apic_timer_irqs++;
+#endif
 
 	evt->event_handler(evt);
 }
@@ -642,35 +731,6 @@ int setup_profiling_timer(unsigned int multiplier)
 }
 
 /*
- * Setup extended LVT, AMD specific (K8, family 10h)
- *
- * Vector mappings are hard coded. On K8 only offset 0 (APIC500) and
- * MCE interrupts are supported. Thus MCE offset must be set to 0.
- */
-
-#define APIC_EILVT_LVTOFF_MCE 0
-#define APIC_EILVT_LVTOFF_IBS 1
-
-static void setup_APIC_eilvt(u8 lvt_off, u8 vector, u8 msg_type, u8 mask)
-{
-	unsigned long reg = (lvt_off << 4) + APIC_EILVT0;
-	unsigned int  v   = (mask << 16) | (msg_type << 8) | vector;
-	apic_write(reg, v);
-}
-
-u8 setup_APIC_eilvt_mce(u8 vector, u8 msg_type, u8 mask)
-{
-	setup_APIC_eilvt(APIC_EILVT_LVTOFF_MCE, vector, msg_type, mask);
-	return APIC_EILVT_LVTOFF_MCE;
-}
-
-u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask)
-{
-	setup_APIC_eilvt(APIC_EILVT_LVTOFF_IBS, vector, msg_type, mask);
-	return APIC_EILVT_LVTOFF_IBS;
-}
-
-/*
  * Local APIC start and shutdown
  */
 
@@ -715,7 +775,7 @@ void clear_local_APIC(void)
 	}
 
 	/* lets not touch this if we didn't frob it */
-#ifdef CONFIG_X86_MCE_P4THERMAL
+#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(X86_MCE_INTEL)
 	if (maxlvt >= 5) {
 		v = apic_read(APIC_LVTTHMR);
 		apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED);
@@ -732,10 +792,6 @@ void clear_local_APIC(void)
 	if (maxlvt >= 4)
 		apic_write(APIC_LVTPC, APIC_LVT_MASKED);
 
-#ifdef CONFIG_X86_MCE_P4THERMAL
-	if (maxlvt >= 5)
-		apic_write(APIC_LVTTHMR, APIC_LVT_MASKED);
-#endif
 	/* Integrated APIC (!82489DX) ? */
 	if (lapic_is_integrated()) {
 		if (maxlvt > 3)
@@ -750,7 +806,7 @@ void clear_local_APIC(void)
  */
 void disable_local_APIC(void)
 {
-	unsigned long value;
+	unsigned int value;
 
 	clear_local_APIC();
 
@@ -762,6 +818,7 @@ void disable_local_APIC(void)
 	value &= ~APIC_SPIV_APIC_ENABLED;
 	apic_write(APIC_SPIV, value);
 
+#ifdef CONFIG_X86_32
 	/*
 	 * When LAPIC was disabled by the BIOS and enabled by the kernel,
 	 * restore the disabled state.
@@ -773,6 +830,7 @@ void disable_local_APIC(void)
 		l &= ~MSR_IA32_APICBASE_ENABLE;
 		wrmsr(MSR_IA32_APICBASE, l, h);
 	}
+#endif
 }
 
 /*
@@ -789,11 +847,15 @@ void lapic_shutdown(void)
 		return;
 
 	local_irq_save(flags);
-	clear_local_APIC();
 
-	if (enabled_via_apicbase)
+#ifdef CONFIG_X86_32
+	if (!enabled_via_apicbase)
+		clear_local_APIC();
+	else
+#endif
 		disable_local_APIC();
 
+
 	local_irq_restore(flags);
 }
 
@@ -838,6 +900,12 @@ int __init verify_local_APIC(void)
 	 */
 	reg0 = apic_read(APIC_ID);
 	apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg0);
+	apic_write(APIC_ID, reg0 ^ APIC_ID_MASK);
+	reg1 = apic_read(APIC_ID);
+	apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg1);
+	apic_write(APIC_ID, reg0);
+	if (reg1 != (reg0 ^ APIC_ID_MASK))
+		return 0;
 
 	/*
 	 * The next two are just to see if we have sane values.
@@ -863,14 +931,15 @@ void __init sync_Arb_IDs(void)
 	 */
 	if (modern_apic() || boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
 		return;
+
 	/*
 	 * Wait for idle.
 	 */
 	apic_wait_icr_idle();
 
 	apic_printk(APIC_DEBUG, "Synchronizing Arb IDs.\n");
-	apic_write(APIC_ICR,
-		   APIC_DEST_ALLINC | APIC_INT_LEVELTRIG | APIC_DM_INIT);
+	apic_write(APIC_ICR, APIC_DEST_ALLINC |
+			APIC_INT_LEVELTRIG | APIC_DM_INIT);
 }
 
 /*
@@ -878,7 +947,7 @@ void __init sync_Arb_IDs(void)
  */
 void __init init_bsp_APIC(void)
 {
-	unsigned long value;
+	unsigned int value;
 
 	/*
 	 * Don't do the setup now if we have a SMP BIOS as the
@@ -899,11 +968,13 @@ void __init init_bsp_APIC(void)
 	value &= ~APIC_VECTOR_MASK;
 	value |= APIC_SPIV_APIC_ENABLED;
 
+#ifdef CONFIG_X86_32
 	/* This bit is reserved on P4/Xeon and should be cleared */
 	if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) &&
 	    (boot_cpu_data.x86 == 15))
 		value &= ~APIC_SPIV_FOCUS_DISABLED;
 	else
+#endif
 		value |= APIC_SPIV_FOCUS_DISABLED;
 	value |= SPURIOUS_APIC_VECTOR;
 	apic_write(APIC_SPIV, value);
@@ -922,6 +993,16 @@ static void __cpuinit lapic_setup_esr(void)
 {
 	unsigned long oldvalue, value, maxlvt;
 	if (lapic_is_integrated() && !esr_disable) {
+		if (esr_disable) {
+			/*
+			 * Something untraceable is creating bad interrupts on
+			 * secondary quads ... for the moment, just leave the
+			 * ESR disabled - we can't do anything useful with the
+			 * errors anyway - mbligh
+			 */
+			printk(KERN_INFO "Leaving ESR disabled.\n");
+			return;
+		}
 		/* !82489DX */
 		maxlvt = lapic_get_maxlvt();
 		if (maxlvt > 3)		/* Due to the Pentium erratum 3AP. */
@@ -942,16 +1023,7 @@ static void __cpuinit lapic_setup_esr(void)
 				"vector: 0x%08lx  after: 0x%08lx\n",
 				oldvalue, value);
 	} else {
-		if (esr_disable)
-			/*
-			 * Something untraceable is creating bad interrupts on
-			 * secondary quads ... for the moment, just leave the
-			 * ESR disabled - we can't do anything useful with the
-			 * errors anyway - mbligh
-			 */
-			printk(KERN_INFO "Leaving ESR disabled.\n");
-		else
-			printk(KERN_INFO "No ESR for 82489DX.\n");
+		printk(KERN_INFO "No ESR for 82489DX.\n");
 	}
 }
 
@@ -1089,13 +1161,17 @@ void __cpuinit setup_local_APIC(void)
 
 void __cpuinit end_local_APIC_setup(void)
 {
-	unsigned long value;
-
 	lapic_setup_esr();
-	/* Disable the local apic timer */
-	value = apic_read(APIC_LVTT);
-	value |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR);
-	apic_write(APIC_LVTT, value);
+
+#ifdef CONFIG_X86_32
+	{
+		unsigned int value;
+		/* Disable the local apic timer */
+		value = apic_read(APIC_LVTT);
+		value |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR);
+		apic_write(APIC_LVTT, value);
+	}
+#endif
 
 	setup_apic_nmi_watchdog(NULL);
 	apic_pm_activate();
@@ -1205,7 +1281,7 @@ void __init init_apic_mappings(void)
 	 * default configuration (or the MP table is broken).
 	 */
 	if (boot_cpu_physical_apicid == -1U)
-		boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id());
+		boot_cpu_physical_apicid = read_apic_id();
 
 }
 
@@ -1242,7 +1318,7 @@ int __init APIC_init_uniprocessor(void)
 	 * might be zero if read from MP tables. Get it from LAPIC.
 	 */
 #ifdef CONFIG_CRASH_DUMP
-	boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id());
+	boot_cpu_physical_apicid = read_apic_id();
 #endif
 	physid_set_mask_of_physid(boot_cpu_physical_apicid, &phys_cpu_present_map);
 
@@ -1321,59 +1397,12 @@ void smp_error_interrupt(struct pt_regs *regs)
 	irq_exit();
 }
 
-#ifdef CONFIG_SMP
-void __init smp_intr_init(void)
-{
-	/*
-	 * IRQ0 must be given a fixed assignment and initialized,
-	 * because it's used before the IO-APIC is set up.
-	 */
-	set_intr_gate(FIRST_DEVICE_VECTOR, interrupt[0]);
-
-	/*
-	 * The reschedule interrupt is a CPU-to-CPU reschedule-helper
-	 * IPI, driven by wakeup.
-	 */
-	alloc_intr_gate(RESCHEDULE_VECTOR, reschedule_interrupt);
-
-	/* IPI for invalidation */
-	alloc_intr_gate(INVALIDATE_TLB_VECTOR, invalidate_interrupt);
-
-	/* IPI for generic function call */
-	alloc_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt);
-
-	/* IPI for single call function */
-	set_intr_gate(CALL_FUNCTION_SINGLE_VECTOR,
-				call_function_single_interrupt);
-}
-#endif
-
-/*
- * Initialize APIC interrupts
- */
-void __init apic_intr_init(void)
-{
-#ifdef CONFIG_SMP
-	smp_intr_init();
-#endif
-	/* self generated IPI for local APIC timer */
-	alloc_intr_gate(LOCAL_TIMER_VECTOR, apic_timer_interrupt);
-
-	/* IPI vectors for APIC spurious and error interrupts */
-	alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt);
-	alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt);
-
-	/* thermal monitor LVT interrupt */
-#ifdef CONFIG_X86_MCE_P4THERMAL
-	alloc_intr_gate(THERMAL_APIC_VECTOR, thermal_interrupt);
-#endif
-}
-
 /**
  * connect_bsp_APIC - attach the APIC to the interrupt system
  */
 void __init connect_bsp_APIC(void)
 {
+#ifdef CONFIG_X86_32
 	if (pic_mode) {
 		/*
 		 * Do not trust the local APIC being empty at bootup.
@@ -1388,6 +1417,7 @@ void __init connect_bsp_APIC(void)
 		outb(0x70, 0x22);
 		outb(0x01, 0x23);
 	}
+#endif
 	enable_apic_mode();
 }
 
@@ -1400,6 +1430,9 @@ void __init connect_bsp_APIC(void)
  */
 void disconnect_bsp_APIC(int virt_wire_setup)
 {
+	unsigned int value;
+
+#ifdef CONFIG_X86_32
 	if (pic_mode) {
 		/*
 		 * Put the board back into PIC mode (has an effect only on
@@ -1411,54 +1444,53 @@ void disconnect_bsp_APIC(int virt_wire_setup)
 				"entering PIC mode.\n");
 		outb(0x70, 0x22);
 		outb(0x00, 0x23);
-	} else {
-		/* Go back to Virtual Wire compatibility mode */
-		unsigned long value;
+		return;
+	}
+#endif
 
-		/* For the spurious interrupt use vector F, and enable it */
-		value = apic_read(APIC_SPIV);
-		value &= ~APIC_VECTOR_MASK;
-		value |= APIC_SPIV_APIC_ENABLED;
-		value |= 0xf;
-		apic_write(APIC_SPIV, value);
+	/* Go back to Virtual Wire compatibility mode */
 
-		if (!virt_wire_setup) {
-			/*
-			 * For LVT0 make it edge triggered, active high,
-			 * external and enabled
-			 */
-			value = apic_read(APIC_LVT0);
-			value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
-				APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
-				APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED);
-			value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
-			value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_EXTINT);
-			apic_write(APIC_LVT0, value);
-		} else {
-			/* Disable LVT0 */
-			apic_write(APIC_LVT0, APIC_LVT_MASKED);
-		}
+	/* For the spurious interrupt use vector F, and enable it */
+	value = apic_read(APIC_SPIV);
+	value &= ~APIC_VECTOR_MASK;
+	value |= APIC_SPIV_APIC_ENABLED;
+	value |= 0xf;
+	apic_write(APIC_SPIV, value);
 
+	if (!virt_wire_setup) {
 		/*
-		 * For LVT1 make it edge triggered, active high, nmi and
-		 * enabled
+		 * For LVT0 make it edge triggered, active high,
+		 * external and enabled
 		 */
-		value = apic_read(APIC_LVT1);
-		value &= ~(
-			APIC_MODE_MASK | APIC_SEND_PENDING |
+		value = apic_read(APIC_LVT0);
+		value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
 			APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
 			APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED);
 		value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
-		value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_NMI);
-		apic_write(APIC_LVT1, value);
+		value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_EXTINT);
+		apic_write(APIC_LVT0, value);
+	} else {
+		/* Disable LVT0 */
+		apic_write(APIC_LVT0, APIC_LVT_MASKED);
 	}
+
+	/*
+	 * For LVT1 make it edge triggered, active high,
+	 * nmi and enabled
+	 */
+	value = apic_read(APIC_LVT1);
+	value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
+			APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
+			APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED);
+	value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
+	value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_NMI);
+	apic_write(APIC_LVT1, value);
 }
 
 void __cpuinit generic_processor_info(int apicid, int version)
 {
 	int cpu;
 	cpumask_t tmp_map;
-	physid_mask_t phys_cpu;
 
 	/*
 	 * Validate version
@@ -1471,9 +1503,6 @@ void __cpuinit generic_processor_info(int apicid, int version)
 	}
 	apic_version[apicid] = version;
 
-	phys_cpu = apicid_to_cpu_present(apicid);
-	physids_or(phys_cpu_present_map, phys_cpu_present_map, phys_cpu);
-
 	if (num_processors >= NR_CPUS) {
 		printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached."
 			"  Processor ignored.\n", NR_CPUS);
@@ -1484,17 +1513,19 @@ void __cpuinit generic_processor_info(int apicid, int version)
 	cpus_complement(tmp_map, cpu_present_map);
 	cpu = first_cpu(tmp_map);
 
-	if (apicid == boot_cpu_physical_apicid)
+	physid_set(apicid, phys_cpu_present_map);
+	if (apicid == boot_cpu_physical_apicid) {
 		/*
 		 * x86_bios_cpu_apicid is required to have processors listed
 		 * in same order as logical cpu numbers. Hence the first
 		 * entry is BSP, and so on.
 		 */
 		cpu = 0;
-
+	}
 	if (apicid > max_physical_apicid)
 		max_physical_apicid = apicid;
 
+#ifdef CONFIG_X86_32
 	/*
 	 * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y
 	 * but we need to work other dependencies like SMP_SUSPEND etc
@@ -1514,7 +1545,9 @@ void __cpuinit generic_processor_info(int apicid, int version)
 			def_to_bigsmp = 1;
 		}
 	}
-#ifdef CONFIG_SMP
+#endif
+
+#if defined(CONFIG_X86_SMP) || defined(CONFIG_X86_64)
 	/* are we being called early in kernel startup? */
 	if (early_per_cpu_ptr(x86_cpu_to_apicid)) {
 		u16 *cpu_to_apicid = early_per_cpu_ptr(x86_cpu_to_apicid);
@@ -1527,6 +1560,7 @@ void __cpuinit generic_processor_info(int apicid, int version)
 		per_cpu(x86_bios_cpu_apicid, cpu) = apicid;
 	}
 #endif
+
 	cpu_set(cpu, cpu_possible_map);
 	cpu_set(cpu, cpu_present_map);
 }
@@ -1537,6 +1571,11 @@ void __cpuinit generic_processor_info(int apicid, int version)
 #ifdef CONFIG_PM
 
 static struct {
+	/*
+	 * 'active' is true if the local APIC was enabled by us and
+	 * not the BIOS; this signifies that we are also responsible
+	 * for disabling it before entering apm/acpi suspend
+	 */
 	int active;
 	/* r/w apic fields */
 	unsigned int apic_id;
@@ -1577,7 +1616,7 @@ static int lapic_suspend(struct sys_device *dev, pm_message_t state)
 	apic_pm_state.apic_lvterr = apic_read(APIC_LVTERR);
 	apic_pm_state.apic_tmict = apic_read(APIC_TMICT);
 	apic_pm_state.apic_tdcr = apic_read(APIC_TDCR);
-#ifdef CONFIG_X86_MCE_P4THERMAL
+#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL)
 	if (maxlvt >= 5)
 		apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR);
 #endif
@@ -1601,16 +1640,23 @@ static int lapic_resume(struct sys_device *dev)
 
 	local_irq_save(flags);
 
-	/*
-	 * Make sure the APICBASE points to the right address
-	 *
-	 * FIXME! This will be wrong if we ever support suspend on
-	 * SMP! We'll need to do this as part of the CPU restore!
-	 */
-	rdmsr(MSR_IA32_APICBASE, l, h);
-	l &= ~MSR_IA32_APICBASE_BASE;
-	l |= MSR_IA32_APICBASE_ENABLE | mp_lapic_addr;
-	wrmsr(MSR_IA32_APICBASE, l, h);
+#ifdef CONFIG_X86_64
+	if (x2apic)
+		enable_x2apic();
+	else
+#endif
+	{
+		/*
+		 * Make sure the APICBASE points to the right address
+		 *
+		 * FIXME! This will be wrong if we ever support suspend on
+		 * SMP! We'll need to do this as part of the CPU restore!
+		 */
+		rdmsr(MSR_IA32_APICBASE, l, h);
+		l &= ~MSR_IA32_APICBASE_BASE;
+		l |= MSR_IA32_APICBASE_ENABLE | mp_lapic_addr;
+		wrmsr(MSR_IA32_APICBASE, l, h);
+	}
 
 	apic_write(APIC_LVTERR, ERROR_APIC_VECTOR | APIC_LVT_MASKED);
 	apic_write(APIC_ID, apic_pm_state.apic_id);
@@ -1620,7 +1666,7 @@ static int lapic_resume(struct sys_device *dev)
 	apic_write(APIC_SPIV, apic_pm_state.apic_spiv);
 	apic_write(APIC_LVT0, apic_pm_state.apic_lvt0);
 	apic_write(APIC_LVT1, apic_pm_state.apic_lvt1);
-#ifdef CONFIG_X86_MCE_P4THERMAL
+#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL)
 	if (maxlvt >= 5)
 		apic_write(APIC_LVTTHMR, apic_pm_state.apic_thmr);
 #endif
@@ -1634,7 +1680,9 @@ static int lapic_resume(struct sys_device *dev)
 	apic_write(APIC_LVTERR, apic_pm_state.apic_lvterr);
 	apic_write(APIC_ESR, 0);
 	apic_read(APIC_ESR);
+
 	local_irq_restore(flags);
+
 	return 0;
 }
 
@@ -1690,20 +1738,20 @@ static int __init parse_lapic(char *arg)
 }
 early_param("lapic", parse_lapic);
 
-static int __init parse_nolapic(char *arg)
+static int __init setup_disableapic(char *arg)
 {
 	disable_apic = 1;
 	setup_clear_cpu_cap(X86_FEATURE_APIC);
 	return 0;
 }
-early_param("nolapic", parse_nolapic);
+early_param("disableapic", setup_disableapic);
 
-static int __init parse_disable_lapic_timer(char *arg)
+/* same as disableapic, for compatibility */
+static int __init setup_nolapic(char *arg)
 {
-	local_apic_timer_disabled = 1;
-	return 0;
+	return setup_disableapic(arg);
 }
-early_param("nolapic_timer", parse_disable_lapic_timer);
+early_param("nolapic", setup_nolapic);
 
 static int __init parse_lapic_timer_c2_ok(char *arg)
 {
@@ -1712,15 +1760,40 @@ static int __init parse_lapic_timer_c2_ok(char *arg)
 }
 early_param("lapic_timer_c2_ok", parse_lapic_timer_c2_ok);
 
+static int __init parse_disable_apic_timer(char *arg)
+{
+	disable_apic_timer = 1;
+	return 0;
+}
+early_param("noapictimer", parse_disable_apic_timer);
+
+static int __init parse_nolapic_timer(char *arg)
+{
+	disable_apic_timer = 1;
+	return 0;
+}
+early_param("nolapic_timer", parse_nolapic_timer);
+
 static int __init apic_set_verbosity(char *arg)
 {
-	if (!arg)
+	if (!arg)  {
+#ifdef CONFIG_X86_64
+		skip_ioapic_setup = 0;
+		ioapic_force = 1;
+		return 0;
+#endif
 		return -EINVAL;
+	}
 
-	if (strcmp(arg, "debug") == 0)
+	if (strcmp("debug", arg) == 0)
 		apic_verbosity = APIC_DEBUG;
-	else if (strcmp(arg, "verbose") == 0)
+	else if (strcmp("verbose", arg) == 0)
 		apic_verbosity = APIC_VERBOSE;
+	else {
+		printk(KERN_WARNING "APIC Verbosity level %s not recognised"
+			" use apic=verbose or apic=debug\n", arg);
+		return -EINVAL;
+	}
 
 	return 0;
 }
diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c
index 446c062e831c..94ddb69ae15e 100644
--- a/arch/x86/kernel/apic_64.c
+++ b/arch/x86/kernel/apic_64.c
@@ -27,6 +27,7 @@
 #include <linux/clockchips.h>
 #include <linux/acpi_pmtmr.h>
 #include <linux/module.h>
+#include <linux/dmar.h>
 
 #include <asm/atomic.h>
 #include <asm/smp.h>
@@ -39,13 +40,20 @@
 #include <asm/proto.h>
 #include <asm/timex.h>
 #include <asm/apic.h>
+#include <asm/i8259.h>
 
 #include <mach_ipi.h>
 #include <mach_apic.h>
 
+/* Disable local APIC timer from the kernel commandline or via dmi quirk */
 static int disable_apic_timer __cpuinitdata;
 static int apic_calibrate_pmtmr __initdata;
 int disable_apic;
+int disable_x2apic;
+int x2apic;
+
+/* x2apic enabled before OS handover */
+int x2apic_preenabled;
 
 /* Local APIC timer works in C2 */
 int local_apic_timer_c2_ok;
@@ -73,6 +81,9 @@ static void lapic_timer_setup(enum clock_event_mode mode,
 static void lapic_timer_broadcast(cpumask_t mask);
 static void apic_pm_activate(void);
 
+/*
+ * The local apic timer can be used for any function which is CPU local.
+ */
 static struct clock_event_device lapic_clockevent = {
 	.name		= "lapic",
 	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT
@@ -99,11 +110,15 @@ static inline int lapic_get_version(void)
 }
 
 /*
- * Check, if the APIC is integrated or a seperate chip
+ * Check, if the APIC is integrated or a separate chip
  */
 static inline int lapic_is_integrated(void)
 {
+#ifdef CONFIG_X86_64
 	return 1;
+#else
+	return APIC_INTEGRATED(lapic_get_version());
+#endif
 }
 
 /*
@@ -118,13 +133,18 @@ static int modern_apic(void)
 	return lapic_get_version() >= 0x14;
 }
 
-void apic_wait_icr_idle(void)
+/*
+ * Paravirt kernels also might be using these below ops. So we still
+ * use generic apic_read()/apic_write(), which might be pointing to different
+ * ops in PARAVIRT case.
+ */
+void xapic_wait_icr_idle(void)
 {
 	while (apic_read(APIC_ICR) & APIC_ICR_BUSY)
 		cpu_relax();
 }
 
-u32 safe_apic_wait_icr_idle(void)
+u32 safe_xapic_wait_icr_idle(void)
 {
 	u32 send_status;
 	int timeout;
@@ -140,6 +160,68 @@ u32 safe_apic_wait_icr_idle(void)
 	return send_status;
 }
 
+void xapic_icr_write(u32 low, u32 id)
+{
+	apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(id));
+	apic_write(APIC_ICR, low);
+}
+
+u64 xapic_icr_read(void)
+{
+	u32 icr1, icr2;
+
+	icr2 = apic_read(APIC_ICR2);
+	icr1 = apic_read(APIC_ICR);
+
+	return icr1 | ((u64)icr2 << 32);
+}
+
+static struct apic_ops xapic_ops = {
+	.read = native_apic_mem_read,
+	.write = native_apic_mem_write,
+	.icr_read = xapic_icr_read,
+	.icr_write = xapic_icr_write,
+	.wait_icr_idle = xapic_wait_icr_idle,
+	.safe_wait_icr_idle = safe_xapic_wait_icr_idle,
+};
+
+struct apic_ops __read_mostly *apic_ops = &xapic_ops;
+EXPORT_SYMBOL_GPL(apic_ops);
+
+static void x2apic_wait_icr_idle(void)
+{
+	/* no need to wait for icr idle in x2apic */
+	return;
+}
+
+static u32 safe_x2apic_wait_icr_idle(void)
+{
+	/* no need to wait for icr idle in x2apic */
+	return 0;
+}
+
+void x2apic_icr_write(u32 low, u32 id)
+{
+	wrmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), ((__u64) id) << 32 | low);
+}
+
+u64 x2apic_icr_read(void)
+{
+	unsigned long val;
+
+	rdmsrl(APIC_BASE_MSR + (APIC_ICR >> 4), val);
+	return val;
+}
+
+static struct apic_ops x2apic_ops = {
+	.read = native_apic_msr_read,
+	.write = native_apic_msr_write,
+	.icr_read = x2apic_icr_read,
+	.icr_write = x2apic_icr_write,
+	.wait_icr_idle = x2apic_wait_icr_idle,
+	.safe_wait_icr_idle = safe_x2apic_wait_icr_idle,
+};
+
 /**
  * enable_NMI_through_LVT0 - enable NMI through local vector table 0
  */
@@ -149,6 +231,11 @@ void __cpuinit enable_NMI_through_LVT0(void)
 
 	/* unmask and set to NMI */
 	v = APIC_DM_NMI;
+
+	/* Level triggered for 82489DX (32bit mode) */
+	if (!lapic_is_integrated())
+		v |= APIC_LVT_LEVEL_TRIGGER;
+
 	apic_write(APIC_LVT0, v);
 }
 
@@ -157,14 +244,28 @@ void __cpuinit enable_NMI_through_LVT0(void)
  */
 int lapic_get_maxlvt(void)
 {
-	unsigned int v, maxlvt;
+	unsigned int v;
 
 	v = apic_read(APIC_LVR);
-	maxlvt = GET_APIC_MAXLVT(v);
-	return maxlvt;
+	/*
+	 * - we always have APIC integrated on 64bit mode
+	 * - 82489DXs do not report # of LVT entries
+	 */
+	return APIC_INTEGRATED(GET_APIC_VERSION(v)) ? GET_APIC_MAXLVT(v) : 2;
 }
 
 /*
+ * Local APIC timer
+ */
+
+/* Clock divisor */
+#ifdef CONFG_X86_64
+#define APIC_DIVISOR 1
+#else
+#define APIC_DIVISOR 16
+#endif
+
+/*
  * This function sets up the local APIC timer, with a timeout of
  * 'clocks' APIC bus clock. During calibration we actually call
  * this function twice on the boot CPU, once with a bogus timeout
@@ -174,7 +275,6 @@ int lapic_get_maxlvt(void)
  * We do reads before writes even if unnecessary, to get around the
  * P5 APIC double write bug.
  */
-
 static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen)
 {
 	unsigned int lvtt_value, tmp_value;
@@ -182,6 +282,9 @@ static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen)
 	lvtt_value = LOCAL_TIMER_VECTOR;
 	if (!oneshot)
 		lvtt_value |= APIC_LVT_TIMER_PERIODIC;
+	if (!lapic_is_integrated())
+		lvtt_value |= SET_APIC_TIMER_BASE(APIC_TIMER_BASE_DIV);
+
 	if (!irqen)
 		lvtt_value |= APIC_LVT_MASKED;
 
@@ -191,12 +294,12 @@ static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen)
 	 * Divide PICLK by 16
 	 */
 	tmp_value = apic_read(APIC_TDCR);
-	apic_write(APIC_TDCR, (tmp_value
-				& ~(APIC_TDR_DIV_1 | APIC_TDR_DIV_TMBASE))
-				| APIC_TDR_DIV_16);
+	apic_write(APIC_TDCR,
+		(tmp_value & ~(APIC_TDR_DIV_1 | APIC_TDR_DIV_TMBASE)) |
+		APIC_TDR_DIV_16);
 
 	if (!oneshot)
-		apic_write(APIC_TMICT, clocks);
+		apic_write(APIC_TMICT, clocks / APIC_DIVISOR);
 }
 
 /*
@@ -204,6 +307,9 @@ static void __setup_APIC_LVTT(unsigned int clocks, int oneshot, int irqen)
  *
  * Vector mappings are hard coded. On K8 only offset 0 (APIC500) and
  * MCE interrupts are supported. Thus MCE offset must be set to 0.
+ *
+ * If mask=1, the LVT entry does not generate interrupts while mask=0
+ * enables the vector. See also the BKDGs.
  */
 
 #define APIC_EILVT_LVTOFF_MCE 0
@@ -228,6 +334,7 @@ u8 setup_APIC_eilvt_ibs(u8 vector, u8 msg_type, u8 mask)
 	setup_APIC_eilvt(APIC_EILVT_LVTOFF_IBS, vector, msg_type, mask);
 	return APIC_EILVT_LVTOFF_IBS;
 }
+EXPORT_SYMBOL_GPL(setup_APIC_eilvt_ibs);
 
 /*
  * Program the next event, relative to now
@@ -366,7 +473,7 @@ static int __init calibrate_APIC_clock(void)
 	lapic_clockevent.min_delta_ns =
 		clockevent_delta2ns(0xF, &lapic_clockevent);
 
-	calibration_result = result / HZ;
+	calibration_result = (result * APIC_DIVISOR) / HZ;
 
 	/*
 	 * Do a sanity check on the APIC calibration result
@@ -388,10 +495,10 @@ static int __init calibrate_APIC_clock(void)
 void __init setup_boot_APIC_clock(void)
 {
 	/*
-	 * The local apic timer can be disabled via the kernel commandline.
-	 * Register the lapic timer as a dummy clock event source on SMP
-	 * systems, so the broadcast mechanism is used. On UP systems simply
-	 * ignore it.
+	 * The local apic timer can be disabled via the kernel
+	 * commandline or from the CPU detection code. Register the lapic
+	 * timer as a dummy clock event source on SMP systems, so the
+	 * broadcast mechanism is used. On UP systems simply ignore it.
 	 */
 	if (disable_apic_timer) {
 		printk(KERN_INFO "Disabling APIC timer\n");
@@ -403,7 +510,9 @@ void __init setup_boot_APIC_clock(void)
 		return;
 	}
 
-	printk(KERN_INFO "Using local APIC timer interrupts.\n");
+	apic_printk(APIC_VERBOSE, "Using local APIC timer interrupts.\n"
+		    "calibrating APIC timer ...\n");
+
 	if (calibrate_APIC_clock()) {
 		/* No broadcast on UP ! */
 		if (num_possible_cpus() > 1)
@@ -422,6 +531,7 @@ void __init setup_boot_APIC_clock(void)
 		printk(KERN_WARNING "APIC timer registered as dummy,"
 			" due to nmi_watchdog=%d!\n", nmi_watchdog);
 
+	/* Setup the lapic or request the broadcast */
 	setup_APIC_timer();
 }
 
@@ -460,7 +570,11 @@ static void local_apic_timer_interrupt(void)
 	/*
 	 * the NMI deadlock-detector uses this.
 	 */
+#ifdef CONFIG_X86_64
 	add_pda(apic_timer_irqs, 1);
+#else
+	per_cpu(irq_stat, cpu).apic_timer_irqs++;
+#endif
 
 	evt->event_handler(evt);
 }
@@ -491,6 +605,7 @@ void smp_apic_timer_interrupt(struct pt_regs *regs)
 	irq_enter();
 	local_apic_timer_interrupt();
 	irq_exit();
+
 	set_irq_regs(old_regs);
 }
 
@@ -544,6 +659,13 @@ void clear_local_APIC(void)
 		apic_write(APIC_LVTPC, v | APIC_LVT_MASKED);
 	}
 
+	/* lets not touch this if we didn't frob it */
+#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(X86_MCE_INTEL)
+	if (maxlvt >= 5) {
+		v = apic_read(APIC_LVTTHMR);
+		apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED);
+	}
+#endif
 	/*
 	 * Clean APIC state for other OSs:
 	 */
@@ -554,8 +676,14 @@ void clear_local_APIC(void)
 		apic_write(APIC_LVTERR, APIC_LVT_MASKED);
 	if (maxlvt >= 4)
 		apic_write(APIC_LVTPC, APIC_LVT_MASKED);
-	apic_write(APIC_ESR, 0);
-	apic_read(APIC_ESR);
+
+	/* Integrated APIC (!82489DX) ? */
+	if (lapic_is_integrated()) {
+		if (maxlvt > 3)
+			/* Clear ESR due to Pentium errata 3AP and 11AP */
+			apic_write(APIC_ESR, 0);
+		apic_read(APIC_ESR);
+	}
 }
 
 /**
@@ -574,8 +702,28 @@ void disable_local_APIC(void)
 	value = apic_read(APIC_SPIV);
 	value &= ~APIC_SPIV_APIC_ENABLED;
 	apic_write(APIC_SPIV, value);
+
+#ifdef CONFIG_X86_32
+	/*
+	 * When LAPIC was disabled by the BIOS and enabled by the kernel,
+	 * restore the disabled state.
+	 */
+	if (enabled_via_apicbase) {
+		unsigned int l, h;
+
+		rdmsr(MSR_IA32_APICBASE, l, h);
+		l &= ~MSR_IA32_APICBASE_ENABLE;
+		wrmsr(MSR_IA32_APICBASE, l, h);
+	}
+#endif
 }
 
+/*
+ * If Linux enabled the LAPIC against the BIOS default disable it down before
+ * re-entering the BIOS on shutdown.  Otherwise the BIOS may get confused and
+ * not power-off.  Additionally clear all LVT entries before disable_local_APIC
+ * for the case where Linux didn't enable the LAPIC.
+ */
 void lapic_shutdown(void)
 {
 	unsigned long flags;
@@ -585,7 +733,13 @@ void lapic_shutdown(void)
 
 	local_irq_save(flags);
 
-	disable_local_APIC();
+#ifdef CONFIG_X86_32
+	if (!enabled_via_apicbase)
+		clear_local_APIC();
+	else
+#endif
+		disable_local_APIC();
+
 
 	local_irq_restore(flags);
 }
@@ -629,10 +783,10 @@ int __init verify_local_APIC(void)
 	/*
 	 * The ID register is read/write in a real APIC.
 	 */
-	reg0 = read_apic_id();
+	reg0 = apic_read(APIC_ID);
 	apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg0);
 	apic_write(APIC_ID, reg0 ^ APIC_ID_MASK);
-	reg1 = read_apic_id();
+	reg1 = apic_read(APIC_ID);
 	apic_printk(APIC_DEBUG, "Getting ID: %x\n", reg1);
 	apic_write(APIC_ID, reg0);
 	if (reg1 != (reg0 ^ APIC_ID_MASK))
@@ -656,8 +810,11 @@ int __init verify_local_APIC(void)
  */
 void __init sync_Arb_IDs(void)
 {
-	/* Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1 */
-	if (modern_apic())
+	/*
+	 * Unsupported on P4 - see Intel Dev. Manual Vol. 3, Ch. 8.6.1 And not
+	 * needed on AMD.
+	 */
+	if (modern_apic() || boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
 		return;
 
 	/*
@@ -666,8 +823,8 @@ void __init sync_Arb_IDs(void)
 	apic_wait_icr_idle();
 
 	apic_printk(APIC_DEBUG, "Synchronizing Arb IDs.\n");
-	apic_write(APIC_ICR, APIC_DEST_ALLINC | APIC_INT_LEVELTRIG
-				| APIC_DM_INIT);
+	apic_write(APIC_ICR, APIC_DEST_ALLINC |
+			APIC_INT_LEVELTRIG | APIC_DM_INIT);
 }
 
 /*
@@ -684,8 +841,6 @@ void __init init_bsp_APIC(void)
 	if (smp_found_config || !cpu_has_apic)
 		return;
 
-	value = apic_read(APIC_LVR);
-
 	/*
 	 * Do not trust the local APIC being empty at bootup.
 	 */
@@ -697,7 +852,15 @@ void __init init_bsp_APIC(void)
 	value = apic_read(APIC_SPIV);
 	value &= ~APIC_VECTOR_MASK;
 	value |= APIC_SPIV_APIC_ENABLED;
-	value |= APIC_SPIV_FOCUS_DISABLED;
+
+#ifdef CONFIG_X86_32
+	/* This bit is reserved on P4/Xeon and should be cleared */
+	if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) &&
+	    (boot_cpu_data.x86 == 15))
+		value &= ~APIC_SPIV_FOCUS_DISABLED;
+	else
+#endif
+		value |= APIC_SPIV_FOCUS_DISABLED;
 	value |= SPURIOUS_APIC_VECTOR;
 	apic_write(APIC_SPIV, value);
 
@@ -706,9 +869,50 @@ void __init init_bsp_APIC(void)
 	 */
 	apic_write(APIC_LVT0, APIC_DM_EXTINT);
 	value = APIC_DM_NMI;
+	if (!lapic_is_integrated())		/* 82489DX */
+		value |= APIC_LVT_LEVEL_TRIGGER;
 	apic_write(APIC_LVT1, value);
 }
 
+static void __cpuinit lapic_setup_esr(void)
+{
+	unsigned long oldvalue, value, maxlvt;
+	if (lapic_is_integrated() && !esr_disable) {
+		if (esr_disable) {
+			/*
+			 * Something untraceable is creating bad interrupts on
+			 * secondary quads ... for the moment, just leave the
+			 * ESR disabled - we can't do anything useful with the
+			 * errors anyway - mbligh
+			 */
+			printk(KERN_INFO "Leaving ESR disabled.\n");
+			return;
+		}
+		/* !82489DX */
+		maxlvt = lapic_get_maxlvt();
+		if (maxlvt > 3)		/* Due to the Pentium erratum 3AP. */
+			apic_write(APIC_ESR, 0);
+		oldvalue = apic_read(APIC_ESR);
+
+		/* enables sending errors */
+		value = ERROR_APIC_VECTOR;
+		apic_write(APIC_LVTERR, value);
+		/*
+		 * spec says clear errors after enabling vector.
+		 */
+		if (maxlvt > 3)
+			apic_write(APIC_ESR, 0);
+		value = apic_read(APIC_ESR);
+		if (value != oldvalue)
+			apic_printk(APIC_VERBOSE, "ESR value before enabling "
+				"vector: 0x%08lx  after: 0x%08lx\n",
+				oldvalue, value);
+	} else {
+		printk(KERN_INFO "No ESR for 82489DX.\n");
+	}
+}
+
+
 /**
  * setup_local_APIC - setup the local APIC
  */
@@ -814,25 +1018,143 @@ void __cpuinit setup_local_APIC(void)
 	preempt_enable();
 }
 
-static void __cpuinit lapic_setup_esr(void)
-{
-	unsigned maxlvt = lapic_get_maxlvt();
-
-	apic_write(APIC_LVTERR, ERROR_APIC_VECTOR);
-	/*
-	 * spec says clear errors after enabling vector.
-	 */
-	if (maxlvt > 3)
-		apic_write(APIC_ESR, 0);
-}
-
 void __cpuinit end_local_APIC_setup(void)
 {
 	lapic_setup_esr();
+
+#ifdef CONFIG_X86_32
+	{
+		unsigned int value;
+		/* Disable the local apic timer */
+		value = apic_read(APIC_LVTT);
+		value |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR);
+		apic_write(APIC_LVTT, value);
+	}
+#endif
+
 	setup_apic_nmi_watchdog(NULL);
 	apic_pm_activate();
 }
 
+void check_x2apic(void)
+{
+	int msr, msr2;
+
+	rdmsr(MSR_IA32_APICBASE, msr, msr2);
+
+	if (msr & X2APIC_ENABLE) {
+		printk("x2apic enabled by BIOS, switching to x2apic ops\n");
+		x2apic_preenabled = x2apic = 1;
+		apic_ops = &x2apic_ops;
+	}
+}
+
+void enable_x2apic(void)
+{
+	int msr, msr2;
+
+	rdmsr(MSR_IA32_APICBASE, msr, msr2);
+	if (!(msr & X2APIC_ENABLE)) {
+		printk("Enabling x2apic\n");
+		wrmsr(MSR_IA32_APICBASE, msr | X2APIC_ENABLE, 0);
+	}
+}
+
+void enable_IR_x2apic(void)
+{
+#ifdef CONFIG_INTR_REMAP
+	int ret;
+	unsigned long flags;
+
+	if (!cpu_has_x2apic)
+		return;
+
+	if (!x2apic_preenabled && disable_x2apic) {
+		printk(KERN_INFO
+		       "Skipped enabling x2apic and Interrupt-remapping "
+		       "because of nox2apic\n");
+		return;
+	}
+
+	if (x2apic_preenabled && disable_x2apic)
+		panic("Bios already enabled x2apic, can't enforce nox2apic");
+
+	if (!x2apic_preenabled && skip_ioapic_setup) {
+		printk(KERN_INFO
+		       "Skipped enabling x2apic and Interrupt-remapping "
+		       "because of skipping io-apic setup\n");
+		return;
+	}
+
+	ret = dmar_table_init();
+	if (ret) {
+		printk(KERN_INFO
+		       "dmar_table_init() failed with %d:\n", ret);
+
+		if (x2apic_preenabled)
+			panic("x2apic enabled by bios. But IR enabling failed");
+		else
+			printk(KERN_INFO
+			       "Not enabling x2apic,Intr-remapping\n");
+		return;
+	}
+
+	local_irq_save(flags);
+	mask_8259A();
+	save_mask_IO_APIC_setup();
+
+	ret = enable_intr_remapping(1);
+
+	if (ret && x2apic_preenabled) {
+		local_irq_restore(flags);
+		panic("x2apic enabled by bios. But IR enabling failed");
+	}
+
+	if (ret)
+		goto end;
+
+	if (!x2apic) {
+		x2apic = 1;
+		apic_ops = &x2apic_ops;
+		enable_x2apic();
+	}
+end:
+	if (ret)
+		/*
+		 * IR enabling failed
+		 */
+		restore_IO_APIC_setup();
+	else
+		reinit_intr_remapped_IO_APIC(x2apic_preenabled);
+
+	unmask_8259A();
+	local_irq_restore(flags);
+
+	if (!ret) {
+		if (!x2apic_preenabled)
+			printk(KERN_INFO
+			       "Enabled x2apic and interrupt-remapping\n");
+		else
+			printk(KERN_INFO
+			       "Enabled Interrupt-remapping\n");
+	} else
+		printk(KERN_ERR
+		       "Failed to enable Interrupt-remapping and x2apic\n");
+#else
+	if (!cpu_has_x2apic)
+		return;
+
+	if (x2apic_preenabled)
+		panic("x2apic enabled prior OS handover,"
+		      " enable CONFIG_INTR_REMAP");
+
+	printk(KERN_INFO "Enable CONFIG_INTR_REMAP for enabling intr-remapping "
+	       " and x2apic\n");
+#endif
+
+	return;
+}
+
 /*
  * Detect and enable local APICs on non-SMP boards.
  * Original code written by Keir Fraser.
@@ -872,7 +1194,7 @@ void __init early_init_lapic_mapping(void)
 	 * Fetch the APIC ID of the BSP in case we have a
 	 * default configuration (or the MP table is broken).
 	 */
-	boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id());
+	boot_cpu_physical_apicid = read_apic_id();
 }
 
 /**
@@ -880,6 +1202,11 @@ void __init early_init_lapic_mapping(void)
  */
 void __init init_apic_mappings(void)
 {
+	if (x2apic) {
+		boot_cpu_physical_apicid = read_apic_id();
+		return;
+	}
+
 	/*
 	 * If no local APIC can be found then set up a fake all
 	 * zeroes page to simulate the local APIC and another
@@ -899,13 +1226,15 @@ void __init init_apic_mappings(void)
 	 * Fetch the APIC ID of the BSP in case we have a
 	 * default configuration (or the MP table is broken).
 	 */
-	boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id());
+	boot_cpu_physical_apicid = read_apic_id();
 }
 
 /*
  * This initializes the IO-APIC and APIC hardware if this is
  * a UP kernel.
  */
+int apic_version[MAX_APICS];
+
 int __init APIC_init_uniprocessor(void)
 {
 	if (disable_apic) {
@@ -918,6 +1247,9 @@ int __init APIC_init_uniprocessor(void)
 		return -1;
 	}
 
+	enable_IR_x2apic();
+	setup_apic_routing();
+
 	verify_local_APIC();
 
 	connect_bsp_APIC();
@@ -1004,17 +1336,57 @@ asmlinkage void smp_error_interrupt(void)
 }
 
 /**
- *  * connect_bsp_APIC - attach the APIC to the interrupt system
- *   */
+ * connect_bsp_APIC - attach the APIC to the interrupt system
+ */
 void __init connect_bsp_APIC(void)
 {
+#ifdef CONFIG_X86_32
+	if (pic_mode) {
+		/*
+		 * Do not trust the local APIC being empty at bootup.
+		 */
+		clear_local_APIC();
+		/*
+		 * PIC mode, enable APIC mode in the IMCR, i.e.  connect BSP's
+		 * local APIC to INT and NMI lines.
+		 */
+		apic_printk(APIC_VERBOSE, "leaving PIC mode, "
+				"enabling APIC mode.\n");
+		outb(0x70, 0x22);
+		outb(0x01, 0x23);
+	}
+#endif
 	enable_apic_mode();
 }
 
+/**
+ * disconnect_bsp_APIC - detach the APIC from the interrupt system
+ * @virt_wire_setup:	indicates, whether virtual wire mode is selected
+ *
+ * Virtual wire mode is necessary to deliver legacy interrupts even when the
+ * APIC is disabled.
+ */
 void disconnect_bsp_APIC(int virt_wire_setup)
 {
+	unsigned int value;
+
+#ifdef CONFIG_X86_32
+	if (pic_mode) {
+		/*
+		 * Put the board back into PIC mode (has an effect only on
+		 * certain older boards).  Note that APIC interrupts, including
+		 * IPIs, won't work beyond this point!  The only exception are
+		 * INIT IPIs.
+		 */
+		apic_printk(APIC_VERBOSE, "disabling APIC mode, "
+				"entering PIC mode.\n");
+		outb(0x70, 0x22);
+		outb(0x00, 0x23);
+		return;
+	}
+#endif
+
 	/* Go back to Virtual Wire compatibility mode */
-	unsigned long value;
 
 	/* For the spurious interrupt use vector F, and enable it */
 	value = apic_read(APIC_SPIV);
@@ -1040,7 +1412,10 @@ void disconnect_bsp_APIC(int virt_wire_setup)
 		apic_write(APIC_LVT0, APIC_LVT_MASKED);
 	}
 
-	/* For LVT1 make it edge triggered, active high, nmi and enabled */
+	/*
+	 * For LVT1 make it edge triggered, active high,
+	 * nmi and enabled
+	 */
 	value = apic_read(APIC_LVT1);
 	value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
 			APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
@@ -1055,9 +1430,20 @@ void __cpuinit generic_processor_info(int apicid, int version)
 	int cpu;
 	cpumask_t tmp_map;
 
+	/*
+	 * Validate version
+	 */
+	if (version == 0x0) {
+		printk(KERN_WARNING "BIOS bug, APIC version is 0 for CPU#%d! "
+				"fixing up to 0x10. (tell your hw vendor)\n",
+				version);
+		version = 0x10;
+	}
+	apic_version[apicid] = version;
+
 	if (num_processors >= NR_CPUS) {
 		printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached."
-		       " Processor ignored.\n", NR_CPUS);
+			"  Processor ignored.\n", NR_CPUS);
 		return;
 	}
 
@@ -1077,6 +1463,29 @@ void __cpuinit generic_processor_info(int apicid, int version)
 	if (apicid > max_physical_apicid)
 		max_physical_apicid = apicid;
 
+#ifdef CONFIG_X86_32
+	/*
+	 * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y
+	 * but we need to work other dependencies like SMP_SUSPEND etc
+	 * before this can be done without some confusion.
+	 * if (CPU_HOTPLUG_ENABLED || num_processors > 8)
+	 *       - Ashok Raj <ashok.raj@intel.com>
+	 */
+	if (max_physical_apicid >= 8) {
+		switch (boot_cpu_data.x86_vendor) {
+		case X86_VENDOR_INTEL:
+			if (!APIC_XAPIC(version)) {
+				def_to_bigsmp = 0;
+				break;
+			}
+			/* If P4 and above fall through */
+		case X86_VENDOR_AMD:
+			def_to_bigsmp = 1;
+		}
+	}
+#endif
+
+#if defined(CONFIG_X86_SMP) || defined(CONFIG_X86_64)
 	/* are we being called early in kernel startup? */
 	if (early_per_cpu_ptr(x86_cpu_to_apicid)) {
 		u16 *cpu_to_apicid = early_per_cpu_ptr(x86_cpu_to_apicid);
@@ -1088,20 +1497,28 @@ void __cpuinit generic_processor_info(int apicid, int version)
 		per_cpu(x86_cpu_to_apicid, cpu) = apicid;
 		per_cpu(x86_bios_cpu_apicid, cpu) = apicid;
 	}
+#endif
 
 	cpu_set(cpu, cpu_possible_map);
 	cpu_set(cpu, cpu_present_map);
 }
 
+int hard_smp_processor_id(void)
+{
+	return read_apic_id();
+}
+
 /*
  * Power management
  */
 #ifdef CONFIG_PM
 
 static struct {
-	/* 'active' is true if the local APIC was enabled by us and
-	   not the BIOS; this signifies that we are also responsible
-	   for disabling it before entering apm/acpi suspend */
+	/*
+	 * 'active' is true if the local APIC was enabled by us and
+	 * not the BIOS; this signifies that we are also responsible
+	 * for disabling it before entering apm/acpi suspend
+	 */
 	int active;
 	/* r/w apic fields */
 	unsigned int apic_id;
@@ -1129,7 +1546,7 @@ static int lapic_suspend(struct sys_device *dev, pm_message_t state)
 
 	maxlvt = lapic_get_maxlvt();
 
-	apic_pm_state.apic_id = read_apic_id();
+	apic_pm_state.apic_id = apic_read(APIC_ID);
 	apic_pm_state.apic_taskpri = apic_read(APIC_TASKPRI);
 	apic_pm_state.apic_ldr = apic_read(APIC_LDR);
 	apic_pm_state.apic_dfr = apic_read(APIC_DFR);
@@ -1142,10 +1559,11 @@ static int lapic_suspend(struct sys_device *dev, pm_message_t state)
 	apic_pm_state.apic_lvterr = apic_read(APIC_LVTERR);
 	apic_pm_state.apic_tmict = apic_read(APIC_TMICT);
 	apic_pm_state.apic_tdcr = apic_read(APIC_TDCR);
-#ifdef CONFIG_X86_MCE_INTEL
+#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL)
 	if (maxlvt >= 5)
 		apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR);
 #endif
+
 	local_irq_save(flags);
 	disable_local_APIC();
 	local_irq_restore(flags);
@@ -1164,10 +1582,25 @@ static int lapic_resume(struct sys_device *dev)
 	maxlvt = lapic_get_maxlvt();
 
 	local_irq_save(flags);
-	rdmsr(MSR_IA32_APICBASE, l, h);
-	l &= ~MSR_IA32_APICBASE_BASE;
-	l |= MSR_IA32_APICBASE_ENABLE | mp_lapic_addr;
-	wrmsr(MSR_IA32_APICBASE, l, h);
+
+#ifdef CONFIG_X86_64
+	if (x2apic)
+		enable_x2apic();
+	else
+#endif
+	{
+		/*
+		 * Make sure the APICBASE points to the right address
+		 *
+		 * FIXME! This will be wrong if we ever support suspend on
+		 * SMP! We'll need to do this as part of the CPU restore!
+		 */
+		rdmsr(MSR_IA32_APICBASE, l, h);
+		l &= ~MSR_IA32_APICBASE_BASE;
+		l |= MSR_IA32_APICBASE_ENABLE | mp_lapic_addr;
+		wrmsr(MSR_IA32_APICBASE, l, h);
+	}
+
 	apic_write(APIC_LVTERR, ERROR_APIC_VECTOR | APIC_LVT_MASKED);
 	apic_write(APIC_ID, apic_pm_state.apic_id);
 	apic_write(APIC_DFR, apic_pm_state.apic_dfr);
@@ -1176,7 +1609,7 @@ static int lapic_resume(struct sys_device *dev)
 	apic_write(APIC_SPIV, apic_pm_state.apic_spiv);
 	apic_write(APIC_LVT0, apic_pm_state.apic_lvt0);
 	apic_write(APIC_LVT1, apic_pm_state.apic_lvt1);
-#ifdef CONFIG_X86_MCE_INTEL
+#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL)
 	if (maxlvt >= 5)
 		apic_write(APIC_LVTTHMR, apic_pm_state.apic_thmr);
 #endif
@@ -1190,10 +1623,17 @@ static int lapic_resume(struct sys_device *dev)
 	apic_write(APIC_LVTERR, apic_pm_state.apic_lvterr);
 	apic_write(APIC_ESR, 0);
 	apic_read(APIC_ESR);
+
 	local_irq_restore(flags);
+
 	return 0;
 }
 
+/*
+ * This device has no shutdown method - fully functioning local APICs
+ * are needed on every CPU up until machine_halt/restart/poweroff.
+ */
+
 static struct sysdev_class lapic_sysclass = {
 	.name		= "lapic",
 	.resume		= lapic_resume,
@@ -1307,31 +1747,19 @@ __cpuinit int apic_is_clustered_box(void)
 	return (clusters > 2);
 }
 
-/*
- * APIC command line parameters
- */
-static int __init apic_set_verbosity(char *str)
+static __init int setup_nox2apic(char *str)
 {
-	if (str == NULL)  {
-		skip_ioapic_setup = 0;
-		ioapic_force = 1;
-		return 0;
-	}
-	if (strcmp("debug", str) == 0)
-		apic_verbosity = APIC_DEBUG;
-	else if (strcmp("verbose", str) == 0)
-		apic_verbosity = APIC_VERBOSE;
-	else {
-		printk(KERN_WARNING "APIC Verbosity level %s not recognised"
-				" use apic=verbose or apic=debug\n", str);
-		return -EINVAL;
-	}
-
+	disable_x2apic = 1;
+	clear_cpu_cap(&boot_cpu_data, X86_FEATURE_X2APIC);
 	return 0;
 }
-early_param("apic", apic_set_verbosity);
+early_param("nox2apic", setup_nox2apic);
+
 
-static __init int setup_disableapic(char *str)
+/*
+ * APIC command line parameters
+ */
+static int __init setup_disableapic(char *arg)
 {
 	disable_apic = 1;
 	setup_clear_cpu_cap(X86_FEATURE_APIC);
@@ -1340,9 +1768,9 @@ static __init int setup_disableapic(char *str)
 early_param("disableapic", setup_disableapic);
 
 /* same as disableapic, for compatibility */
-static __init int setup_nolapic(char *str)
+static int __init setup_nolapic(char *arg)
 {
-	return setup_disableapic(str);
+	return setup_disableapic(arg);
 }
 early_param("nolapic", setup_nolapic);
 
@@ -1353,14 +1781,19 @@ static int __init parse_lapic_timer_c2_ok(char *arg)
 }
 early_param("lapic_timer_c2_ok", parse_lapic_timer_c2_ok);
 
-static __init int setup_noapictimer(char *str)
+static int __init parse_disable_apic_timer(char *arg)
 {
-	if (str[0] != ' ' && str[0] != 0)
-		return 0;
 	disable_apic_timer = 1;
-	return 1;
+	return 0;
 }
-__setup("noapictimer", setup_noapictimer);
+early_param("noapictimer", parse_disable_apic_timer);
+
+static int __init parse_nolapic_timer(char *arg)
+{
+	disable_apic_timer = 1;
+	return 0;
+}
+early_param("nolapic_timer", parse_nolapic_timer);
 
 static __init int setup_apicpmtimer(char *s)
 {
@@ -1370,6 +1803,31 @@ static __init int setup_apicpmtimer(char *s)
 }
 __setup("apicpmtimer", setup_apicpmtimer);
 
+static int __init apic_set_verbosity(char *arg)
+{
+	if (!arg)  {
+#ifdef CONFIG_X86_64
+		skip_ioapic_setup = 0;
+		ioapic_force = 1;
+		return 0;
+#endif
+		return -EINVAL;
+	}
+
+	if (strcmp("debug", arg) == 0)
+		apic_verbosity = APIC_DEBUG;
+	else if (strcmp("verbose", arg) == 0)
+		apic_verbosity = APIC_VERBOSE;
+	else {
+		printk(KERN_WARNING "APIC Verbosity level %s not recognised"
+			" use apic=verbose or apic=debug\n", arg);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+early_param("apic", apic_set_verbosity);
+
 static int __init lapic_insert_resource(void)
 {
 	if (!apic_phys)
diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile
index ee76eaad3001..7f0b45a5d788 100644
--- a/arch/x86/kernel/cpu/Makefile
+++ b/arch/x86/kernel/cpu/Makefile
@@ -3,22 +3,30 @@
 #
 
 obj-y			:= intel_cacheinfo.o addon_cpuid_features.o
-obj-y			+= proc.o feature_names.o
-
-obj-$(CONFIG_X86_32)	+= common.o bugs.o
-obj-$(CONFIG_X86_64)	+= common_64.o bugs_64.o
-obj-$(CONFIG_X86_32)	+= amd.o
-obj-$(CONFIG_X86_64)	+= amd_64.o
-obj-$(CONFIG_X86_32)	+= cyrix.o
-obj-$(CONFIG_X86_32)	+= centaur.o
-obj-$(CONFIG_X86_64)	+= centaur_64.o
-obj-$(CONFIG_X86_32)	+= transmeta.o
-obj-$(CONFIG_X86_32)	+= intel.o
-obj-$(CONFIG_X86_64)	+= intel_64.o
-obj-$(CONFIG_X86_32)	+= umc.o
+obj-y			+= proc.o capflags.o powerflags.o common.o
+
+obj-$(CONFIG_X86_32)	+= bugs.o cmpxchg.o
+obj-$(CONFIG_X86_64)	+= bugs_64.o
+
+obj-$(CONFIG_CPU_SUP_INTEL)		+= intel.o
+obj-$(CONFIG_CPU_SUP_AMD)		+= amd.o
+obj-$(CONFIG_CPU_SUP_CYRIX_32)		+= cyrix.o
+obj-$(CONFIG_CPU_SUP_CENTAUR_32)	+= centaur.o
+obj-$(CONFIG_CPU_SUP_CENTAUR_64)	+= centaur_64.o
+obj-$(CONFIG_CPU_SUP_TRANSMETA_32)	+= transmeta.o
+obj-$(CONFIG_CPU_SUP_UMC_32)		+= umc.o
 
 obj-$(CONFIG_X86_MCE)	+= mcheck/
 obj-$(CONFIG_MTRR)	+= mtrr/
 obj-$(CONFIG_CPU_FREQ)	+= cpufreq/
 
 obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o
+
+quiet_cmd_mkcapflags = MKCAP   $@
+      cmd_mkcapflags = $(PERL) $(srctree)/$(src)/mkcapflags.pl $< $@
+
+cpufeature = $(src)/../../../../include/asm-x86/cpufeature.h
+
+targets += capflags.c
+$(obj)/capflags.c: $(cpufeature) $(src)/mkcapflags.pl FORCE
+	$(call if_changed,mkcapflags)
diff --git a/arch/x86/kernel/cpu/addon_cpuid_features.c b/arch/x86/kernel/cpu/addon_cpuid_features.c
index a6ef672adbba..0d9c993aa93e 100644
--- a/arch/x86/kernel/cpu/addon_cpuid_features.c
+++ b/arch/x86/kernel/cpu/addon_cpuid_features.c
@@ -7,6 +7,8 @@
 #include <asm/pat.h>
 #include <asm/processor.h>
 
+#include <mach_apic.h>
+
 struct cpuid_bit {
 	u16 feature;
 	u8 reg;
@@ -48,6 +50,92 @@ void __cpuinit init_scattered_cpuid_features(struct cpuinfo_x86 *c)
 	}
 }
 
+/* leaf 0xb SMT level */
+#define SMT_LEVEL	0
+
+/* leaf 0xb sub-leaf types */
+#define INVALID_TYPE	0
+#define SMT_TYPE	1
+#define CORE_TYPE	2
+
+#define LEAFB_SUBTYPE(ecx)		(((ecx) >> 8) & 0xff)
+#define BITS_SHIFT_NEXT_LEVEL(eax)	((eax) & 0x1f)
+#define LEVEL_MAX_SIBLINGS(ebx)		((ebx) & 0xffff)
+
+/*
+ * Check for extended topology enumeration cpuid leaf 0xb and if it
+ * exists, use it for populating initial_apicid and cpu topology
+ * detection.
+ */
+void __cpuinit detect_extended_topology(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_SMP
+	unsigned int eax, ebx, ecx, edx, sub_index;
+	unsigned int ht_mask_width, core_plus_mask_width;
+	unsigned int core_select_mask, core_level_siblings;
+
+	if (c->cpuid_level < 0xb)
+		return;
+
+	cpuid_count(0xb, SMT_LEVEL, &eax, &ebx, &ecx, &edx);
+
+	/*
+	 * check if the cpuid leaf 0xb is actually implemented.
+	 */
+	if (ebx == 0 || (LEAFB_SUBTYPE(ecx) != SMT_TYPE))
+		return;
+
+	set_cpu_cap(c, X86_FEATURE_XTOPOLOGY);
+
+	/*
+	 * initial apic id, which also represents 32-bit extended x2apic id.
+	 */
+	c->initial_apicid = edx;
+
+	/*
+	 * Populate HT related information from sub-leaf level 0.
+	 */
+	core_level_siblings = smp_num_siblings = LEVEL_MAX_SIBLINGS(ebx);
+	core_plus_mask_width = ht_mask_width = BITS_SHIFT_NEXT_LEVEL(eax);
+
+	sub_index = 1;
+	do {
+		cpuid_count(0xb, sub_index, &eax, &ebx, &ecx, &edx);
+
+		/*
+		 * Check for the Core type in the implemented sub leaves.
+		 */
+		if (LEAFB_SUBTYPE(ecx) == CORE_TYPE) {
+			core_level_siblings = LEVEL_MAX_SIBLINGS(ebx);
+			core_plus_mask_width = BITS_SHIFT_NEXT_LEVEL(eax);
+			break;
+		}
+
+		sub_index++;
+	} while (LEAFB_SUBTYPE(ecx) != INVALID_TYPE);
+
+	core_select_mask = (~(-1 << core_plus_mask_width)) >> ht_mask_width;
+
+#ifdef CONFIG_X86_32
+	c->cpu_core_id = phys_pkg_id(c->initial_apicid, ht_mask_width)
+						 & core_select_mask;
+	c->phys_proc_id = phys_pkg_id(c->initial_apicid, core_plus_mask_width);
+#else
+	c->cpu_core_id = phys_pkg_id(ht_mask_width) & core_select_mask;
+	c->phys_proc_id = phys_pkg_id(core_plus_mask_width);
+#endif
+	c->x86_max_cores = (core_level_siblings / smp_num_siblings);
+
+
+	printk(KERN_INFO  "CPU: Physical Processor ID: %d\n",
+	       c->phys_proc_id);
+	if (c->x86_max_cores > 1)
+		printk(KERN_INFO  "CPU: Processor Core ID: %d\n",
+		       c->cpu_core_id);
+	return;
+#endif
+}
+
 #ifdef CONFIG_X86_PAT
 void __cpuinit validate_pat_support(struct cpuinfo_x86 *c)
 {
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 18514ed26104..32e73520adf7 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -1,13 +1,22 @@
 #include <linux/init.h>
 #include <linux/bitops.h>
 #include <linux/mm.h>
+
 #include <asm/io.h>
 #include <asm/processor.h>
 #include <asm/apic.h>
 
+#ifdef CONFIG_X86_64
+# include <asm/numa_64.h>
+# include <asm/mmconfig.h>
+# include <asm/cacheflush.h>
+#endif
+
 #include <mach_apic.h>
+
 #include "cpu.h"
 
+#ifdef CONFIG_X86_32
 /*
  *	B step AMD K6 before B 9730xxxx have hardware bugs that can cause
  *	misexecution of code under Linux. Owners of such processors should
@@ -24,26 +33,273 @@
 extern void vide(void);
 __asm__(".align 4\nvide: ret");
 
-static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
+static void __cpuinit init_amd_k5(struct cpuinfo_x86 *c)
 {
-	if (cpuid_eax(0x80000000) >= 0x80000007) {
-		c->x86_power = cpuid_edx(0x80000007);
-		if (c->x86_power & (1<<8))
-			set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
+/*
+ * General Systems BIOSen alias the cpu frequency registers
+ * of the Elan at 0x000df000. Unfortuantly, one of the Linux
+ * drivers subsequently pokes it, and changes the CPU speed.
+ * Workaround : Remove the unneeded alias.
+ */
+#define CBAR		(0xfffc) /* Configuration Base Address  (32-bit) */
+#define CBAR_ENB	(0x80000000)
+#define CBAR_KEY	(0X000000CB)
+	if (c->x86_model == 9 || c->x86_model == 10) {
+		if (inl (CBAR) & CBAR_ENB)
+			outl (0 | CBAR_KEY, CBAR);
 	}
-
-	/*  Set MTRR capability flag if appropriate */
-	if (c->x86_model == 13 || c->x86_model == 9 ||
-	   (c->x86_model == 8 && c->x86_mask >= 8))
-		set_cpu_cap(c, X86_FEATURE_K6_MTRR);
 }
 
-static void __cpuinit init_amd(struct cpuinfo_x86 *c)
+
+static void __cpuinit init_amd_k6(struct cpuinfo_x86 *c)
 {
 	u32 l, h;
 	int mbytes = num_physpages >> (20-PAGE_SHIFT);
-	int r;
 
+	if (c->x86_model < 6) {
+		/* Based on AMD doc 20734R - June 2000 */
+		if (c->x86_model == 0) {
+			clear_cpu_cap(c, X86_FEATURE_APIC);
+			set_cpu_cap(c, X86_FEATURE_PGE);
+		}
+		return;
+	}
+
+	if (c->x86_model == 6 && c->x86_mask == 1) {
+		const int K6_BUG_LOOP = 1000000;
+		int n;
+		void (*f_vide)(void);
+		unsigned long d, d2;
+
+		printk(KERN_INFO "AMD K6 stepping B detected - ");
+
+		/*
+		 * It looks like AMD fixed the 2.6.2 bug and improved indirect
+		 * calls at the same time.
+		 */
+
+		n = K6_BUG_LOOP;
+		f_vide = vide;
+		rdtscl(d);
+		while (n--)
+			f_vide();
+		rdtscl(d2);
+		d = d2-d;
+
+		if (d > 20*K6_BUG_LOOP)
+			printk("system stability may be impaired when more than 32 MB are used.\n");
+		else
+			printk("probably OK (after B9730xxxx).\n");
+		printk(KERN_INFO "Please see http://membres.lycos.fr/poulot/k6bug.html\n");
+	}
+
+	/* K6 with old style WHCR */
+	if (c->x86_model < 8 ||
+	   (c->x86_model == 8 && c->x86_mask < 8)) {
+		/* We can only write allocate on the low 508Mb */
+		if (mbytes > 508)
+			mbytes = 508;
+
+		rdmsr(MSR_K6_WHCR, l, h);
+		if ((l&0x0000FFFF) == 0) {
+			unsigned long flags;
+			l = (1<<0)|((mbytes/4)<<1);
+			local_irq_save(flags);
+			wbinvd();
+			wrmsr(MSR_K6_WHCR, l, h);
+			local_irq_restore(flags);
+			printk(KERN_INFO "Enabling old style K6 write allocation for %d Mb\n",
+				mbytes);
+		}
+		return;
+	}
+
+	if ((c->x86_model == 8 && c->x86_mask > 7) ||
+	     c->x86_model == 9 || c->x86_model == 13) {
+		/* The more serious chips .. */
+
+		if (mbytes > 4092)
+			mbytes = 4092;
+
+		rdmsr(MSR_K6_WHCR, l, h);
+		if ((l&0xFFFF0000) == 0) {
+			unsigned long flags;
+			l = ((mbytes>>2)<<22)|(1<<16);
+			local_irq_save(flags);
+			wbinvd();
+			wrmsr(MSR_K6_WHCR, l, h);
+			local_irq_restore(flags);
+			printk(KERN_INFO "Enabling new style K6 write allocation for %d Mb\n",
+				mbytes);
+		}
+
+		return;
+	}
+
+	if (c->x86_model == 10) {
+		/* AMD Geode LX is model 10 */
+		/* placeholder for any needed mods */
+		return;
+	}
+}
+
+static void __cpuinit init_amd_k7(struct cpuinfo_x86 *c)
+{
+	u32 l, h;
+
+	/*
+	 * Bit 15 of Athlon specific MSR 15, needs to be 0
+	 * to enable SSE on Palomino/Morgan/Barton CPU's.
+	 * If the BIOS didn't enable it already, enable it here.
+	 */
+	if (c->x86_model >= 6 && c->x86_model <= 10) {
+		if (!cpu_has(c, X86_FEATURE_XMM)) {
+			printk(KERN_INFO "Enabling disabled K7/SSE Support.\n");
+			rdmsr(MSR_K7_HWCR, l, h);
+			l &= ~0x00008000;
+			wrmsr(MSR_K7_HWCR, l, h);
+			set_cpu_cap(c, X86_FEATURE_XMM);
+		}
+	}
+
+	/*
+	 * It's been determined by AMD that Athlons since model 8 stepping 1
+	 * are more robust with CLK_CTL set to 200xxxxx instead of 600xxxxx
+	 * As per AMD technical note 27212 0.2
+	 */
+	if ((c->x86_model == 8 && c->x86_mask >= 1) || (c->x86_model > 8)) {
+		rdmsr(MSR_K7_CLK_CTL, l, h);
+		if ((l & 0xfff00000) != 0x20000000) {
+			printk ("CPU: CLK_CTL MSR was %x. Reprogramming to %x\n", l,
+				((l & 0x000fffff)|0x20000000));
+			wrmsr(MSR_K7_CLK_CTL, (l & 0x000fffff)|0x20000000, h);
+		}
+	}
+
+	set_cpu_cap(c, X86_FEATURE_K7);
+}
+#endif
+
+#if defined(CONFIG_NUMA) && defined(CONFIG_X86_64)
+static int __cpuinit nearby_node(int apicid)
+{
+	int i, node;
+
+	for (i = apicid - 1; i >= 0; i--) {
+		node = apicid_to_node[i];
+		if (node != NUMA_NO_NODE && node_online(node))
+			return node;
+	}
+	for (i = apicid + 1; i < MAX_LOCAL_APIC; i++) {
+		node = apicid_to_node[i];
+		if (node != NUMA_NO_NODE && node_online(node))
+			return node;
+	}
+	return first_node(node_online_map); /* Shouldn't happen */
+}
+#endif
+
+/*
+ * On a AMD dual core setup the lower bits of the APIC id distingush the cores.
+ * Assumes number of cores is a power of two.
+ */
+static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_X86_HT
+	unsigned bits;
+
+	bits = c->x86_coreid_bits;
+
+	/* Low order bits define the core id (index of core in socket) */
+	c->cpu_core_id = c->initial_apicid & ((1 << bits)-1);
+	/* Convert the initial APIC ID into the socket ID */
+	c->phys_proc_id = c->initial_apicid >> bits;
+#endif
+}
+
+static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c)
+{
+#if defined(CONFIG_NUMA) && defined(CONFIG_X86_64)
+	int cpu = smp_processor_id();
+	int node;
+	unsigned apicid = hard_smp_processor_id();
+
+	node = c->phys_proc_id;
+	if (apicid_to_node[apicid] != NUMA_NO_NODE)
+		node = apicid_to_node[apicid];
+	if (!node_online(node)) {
+		/* Two possibilities here:
+		   - The CPU is missing memory and no node was created.
+		   In that case try picking one from a nearby CPU
+		   - The APIC IDs differ from the HyperTransport node IDs
+		   which the K8 northbridge parsing fills in.
+		   Assume they are all increased by a constant offset,
+		   but in the same order as the HT nodeids.
+		   If that doesn't result in a usable node fall back to the
+		   path for the previous case.  */
+
+		int ht_nodeid = c->initial_apicid;
+
+		if (ht_nodeid >= 0 &&
+		    apicid_to_node[ht_nodeid] != NUMA_NO_NODE)
+			node = apicid_to_node[ht_nodeid];
+		/* Pick a nearby node */
+		if (!node_online(node))
+			node = nearby_node(apicid);
+	}
+	numa_set_node(cpu, node);
+
+	printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node);
+#endif
+}
+
+static void __cpuinit early_init_amd_mc(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_X86_HT
+	unsigned bits, ecx;
+
+	/* Multi core CPU? */
+	if (c->extended_cpuid_level < 0x80000008)
+		return;
+
+	ecx = cpuid_ecx(0x80000008);
+
+	c->x86_max_cores = (ecx & 0xff) + 1;
+
+	/* CPU telling us the core id bits shift? */
+	bits = (ecx >> 12) & 0xF;
+
+	/* Otherwise recompute */
+	if (bits == 0) {
+		while ((1 << bits) < c->x86_max_cores)
+			bits++;
+	}
+
+	c->x86_coreid_bits = bits;
+#endif
+}
+
+static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
+{
+	early_init_amd_mc(c);
+
+	/* c->x86_power is 8000_0007 edx. Bit 8 is constant TSC */
+	if (c->x86_power & (1<<8))
+		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
+
+#ifdef CONFIG_X86_64
+	set_cpu_cap(c, X86_FEATURE_SYSCALL32);
+#else
+	/*  Set MTRR capability flag if appropriate */
+	if (c->x86 == 5)
+		if (c->x86_model == 13 || c->x86_model == 9 ||
+		    (c->x86_model == 8 && c->x86_mask >= 8))
+			set_cpu_cap(c, X86_FEATURE_K6_MTRR);
+#endif
+}
+
+static void __cpuinit init_amd(struct cpuinfo_x86 *c)
+{
 #ifdef CONFIG_SMP
 	unsigned long long value;
 
@@ -54,7 +310,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 	 * Errata 63 for SH-B3 steppings
 	 * Errata 122 for all steppings (F+ have it disabled by default)
 	 */
-	if (c->x86 == 15) {
+	if (c->x86 == 0xf) {
 		rdmsrl(MSR_K7_HWCR, value);
 		value |= 1 << 6;
 		wrmsrl(MSR_K7_HWCR, value);
@@ -64,209 +320,119 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 	early_init_amd(c);
 
 	/*
-	 *	FIXME: We should handle the K5 here. Set up the write
-	 *	range and also turn on MSR 83 bits 4 and 31 (write alloc,
-	 *	no bus pipeline)
-	 */
-
-	/*
 	 * Bit 31 in normal CPUID used for nonstandard 3DNow ID;
 	 * 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway
 	 */
 	clear_cpu_cap(c, 0*32+31);
 
-	r = get_model_name(c);
+#ifdef CONFIG_X86_64
+	/* On C+ stepping K8 rep microcode works well for copy/memset */
+	if (c->x86 == 0xf) {
+		u32 level;
 
-	switch (c->x86) {
-	case 4:
-		/*
-		 * General Systems BIOSen alias the cpu frequency registers
-		 * of the Elan at 0x000df000. Unfortuantly, one of the Linux
-		 * drivers subsequently pokes it, and changes the CPU speed.
-		 * Workaround : Remove the unneeded alias.
-		 */
-#define CBAR		(0xfffc) /* Configuration Base Address  (32-bit) */
-#define CBAR_ENB	(0x80000000)
-#define CBAR_KEY	(0X000000CB)
-			if (c->x86_model == 9 || c->x86_model == 10) {
-				if (inl (CBAR) & CBAR_ENB)
-					outl (0 | CBAR_KEY, CBAR);
-			}
-			break;
-	case 5:
-			if (c->x86_model < 6) {
-				/* Based on AMD doc 20734R - June 2000 */
-				if (c->x86_model == 0) {
-					clear_cpu_cap(c, X86_FEATURE_APIC);
-					set_cpu_cap(c, X86_FEATURE_PGE);
-				}
-				break;
-			}
-
-			if (c->x86_model == 6 && c->x86_mask == 1) {
-				const int K6_BUG_LOOP = 1000000;
-				int n;
-				void (*f_vide)(void);
-				unsigned long d, d2;
-
-				printk(KERN_INFO "AMD K6 stepping B detected - ");
-
-				/*
-				 * It looks like AMD fixed the 2.6.2 bug and improved indirect
-				 * calls at the same time.
-				 */
-
-				n = K6_BUG_LOOP;
-				f_vide = vide;
-				rdtscl(d);
-				while (n--)
-					f_vide();
-				rdtscl(d2);
-				d = d2-d;
-
-				if (d > 20*K6_BUG_LOOP)
-					printk("system stability may be impaired when more than 32 MB are used.\n");
-				else
-					printk("probably OK (after B9730xxxx).\n");
-				printk(KERN_INFO "Please see http://membres.lycos.fr/poulot/k6bug.html\n");
-			}
-
-			/* K6 with old style WHCR */
-			if (c->x86_model < 8 ||
-			   (c->x86_model == 8 && c->x86_mask < 8)) {
-				/* We can only write allocate on the low 508Mb */
-				if (mbytes > 508)
-					mbytes = 508;
-
-				rdmsr(MSR_K6_WHCR, l, h);
-				if ((l&0x0000FFFF) == 0) {
-					unsigned long flags;
-					l = (1<<0)|((mbytes/4)<<1);
-					local_irq_save(flags);
-					wbinvd();
-					wrmsr(MSR_K6_WHCR, l, h);
-					local_irq_restore(flags);
-					printk(KERN_INFO "Enabling old style K6 write allocation for %d Mb\n",
-						mbytes);
-				}
-				break;
-			}
-
-			if ((c->x86_model == 8 && c->x86_mask > 7) ||
-			     c->x86_model == 9 || c->x86_model == 13) {
-				/* The more serious chips .. */
-
-				if (mbytes > 4092)
-					mbytes = 4092;
-
-				rdmsr(MSR_K6_WHCR, l, h);
-				if ((l&0xFFFF0000) == 0) {
-					unsigned long flags;
-					l = ((mbytes>>2)<<22)|(1<<16);
-					local_irq_save(flags);
-					wbinvd();
-					wrmsr(MSR_K6_WHCR, l, h);
-					local_irq_restore(flags);
-					printk(KERN_INFO "Enabling new style K6 write allocation for %d Mb\n",
-						mbytes);
-				}
-
-				break;
-			}
-
-			if (c->x86_model == 10) {
-				/* AMD Geode LX is model 10 */
-				/* placeholder for any needed mods */
-				break;
-			}
-			break;
-	case 6: /* An Athlon/Duron */
-
-			/*
-			 * Bit 15 of Athlon specific MSR 15, needs to be 0
-			 * to enable SSE on Palomino/Morgan/Barton CPU's.
-			 * If the BIOS didn't enable it already, enable it here.
-			 */
-			if (c->x86_model >= 6 && c->x86_model <= 10) {
-				if (!cpu_has(c, X86_FEATURE_XMM)) {
-					printk(KERN_INFO "Enabling disabled K7/SSE Support.\n");
-					rdmsr(MSR_K7_HWCR, l, h);
-					l &= ~0x00008000;
-					wrmsr(MSR_K7_HWCR, l, h);
-					set_cpu_cap(c, X86_FEATURE_XMM);
-				}
-			}
-
-			/*
-			 * It's been determined by AMD that Athlons since model 8 stepping 1
-			 * are more robust with CLK_CTL set to 200xxxxx instead of 600xxxxx
-			 * As per AMD technical note 27212 0.2
-			 */
-			if ((c->x86_model == 8 && c->x86_mask >= 1) || (c->x86_model > 8)) {
-				rdmsr(MSR_K7_CLK_CTL, l, h);
-				if ((l & 0xfff00000) != 0x20000000) {
-					printk ("CPU: CLK_CTL MSR was %x. Reprogramming to %x\n", l,
-						((l & 0x000fffff)|0x20000000));
-					wrmsr(MSR_K7_CLK_CTL, (l & 0x000fffff)|0x20000000, h);
-				}
-			}
-			break;
+		level = cpuid_eax(1);
+		if((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58)
+			set_cpu_cap(c, X86_FEATURE_REP_GOOD);
 	}
+	if (c->x86 == 0x10 || c->x86 == 0x11)
+		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
+#else
+
+	/*
+	 *	FIXME: We should handle the K5 here. Set up the write
+	 *	range and also turn on MSR 83 bits 4 and 31 (write alloc,
+	 *	no bus pipeline)
+	 */
 
 	switch (c->x86) {
-	case 15:
-	/* Use K8 tuning for Fam10h and Fam11h */
-	case 0x10:
-	case 0x11:
-		set_cpu_cap(c, X86_FEATURE_K8);
+	case 4:
+		init_amd_k5(c);
 		break;
-	case 6:
-		set_cpu_cap(c, X86_FEATURE_K7);
+	case 5:
+		init_amd_k6(c);
+		break;
+	case 6: /* An Athlon/Duron */
+		init_amd_k7(c);
 		break;
 	}
+
+	/* K6s reports MCEs but don't actually have all the MSRs */
+	if (c->x86 < 6)
+		clear_cpu_cap(c, X86_FEATURE_MCE);
+#endif
+
+	/* Enable workaround for FXSAVE leak */
 	if (c->x86 >= 6)
 		set_cpu_cap(c, X86_FEATURE_FXSAVE_LEAK);
 
-	display_cacheinfo(c);
-
-	if (cpuid_eax(0x80000000) >= 0x80000008)
-		c->x86_max_cores = (cpuid_ecx(0x80000008) & 0xff) + 1;
+	if (!c->x86_model_id[0]) {
+		switch (c->x86) {
+		case 0xf:
+			/* Should distinguish Models here, but this is only
+			   a fallback anyways. */
+			strcpy(c->x86_model_id, "Hammer");
+			break;
+		}
+	}
 
-#ifdef CONFIG_X86_HT
-	/*
-	 * On a AMD multi core setup the lower bits of the APIC id
-	 * distinguish the cores.
-	 */
-	if (c->x86_max_cores > 1) {
-		int cpu = smp_processor_id();
-		unsigned bits = (cpuid_ecx(0x80000008) >> 12) & 0xf;
+	display_cacheinfo(c);
 
-		if (bits == 0) {
-			while ((1 << bits) < c->x86_max_cores)
-				bits++;
-		}
-		c->cpu_core_id = c->phys_proc_id & ((1<<bits)-1);
-		c->phys_proc_id >>= bits;
-		printk(KERN_INFO "CPU %d(%d) -> Core %d\n",
-		       cpu, c->x86_max_cores, c->cpu_core_id);
+	/* Multi core CPU? */
+	if (c->extended_cpuid_level >= 0x80000008) {
+		amd_detect_cmp(c);
+		srat_detect_node(c);
 	}
+
+#ifdef CONFIG_X86_32
+	detect_ht(c);
 #endif
 
-	if (cpuid_eax(0x80000000) >= 0x80000006) {
-		if ((c->x86 == 0x10) && (cpuid_edx(0x80000006) & 0xf000))
+	if (c->extended_cpuid_level >= 0x80000006) {
+		if ((c->x86 >= 0x0f) && (cpuid_edx(0x80000006) & 0xf000))
 			num_cache_leaves = 4;
 		else
 			num_cache_leaves = 3;
 	}
 
-	/* K6s reports MCEs but don't actually have all the MSRs */
-	if (c->x86 < 6)
-		clear_cpu_cap(c, X86_FEATURE_MCE);
+	if (c->x86 >= 0xf && c->x86 <= 0x11)
+		set_cpu_cap(c, X86_FEATURE_K8);
 
-	if (cpu_has_xmm2)
+	if (cpu_has_xmm2) {
+		/* MFENCE stops RDTSC speculation */
 		set_cpu_cap(c, X86_FEATURE_MFENCE_RDTSC);
+	}
+
+#ifdef CONFIG_X86_64
+	if (c->x86 == 0x10) {
+		/* do this for boot cpu */
+		if (c == &boot_cpu_data)
+			check_enable_amd_mmconf_dmi();
+
+		fam10h_check_enable_mmcfg();
+	}
+
+	if (c == &boot_cpu_data && c->x86 >= 0xf && c->x86 <= 0x11) {
+		unsigned long long tseg;
+
+		/*
+		 * Split up direct mapping around the TSEG SMM area.
+		 * Don't do it for gbpages because there seems very little
+		 * benefit in doing so.
+		 */
+		if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg)) {
+		    printk(KERN_DEBUG "tseg: %010llx\n", tseg);
+		    if ((tseg>>PMD_SHIFT) <
+				(max_low_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) ||
+			((tseg>>PMD_SHIFT) <
+				(max_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) &&
+			 (tseg>>PMD_SHIFT) >= (1ULL<<(32 - PMD_SHIFT))))
+			set_memory_4k((unsigned long)__va(tseg), 1);
+		}
+	}
+#endif
 }
 
+#ifdef CONFIG_X86_32
 static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c, unsigned int size)
 {
 	/* AMD errata T13 (order #21922) */
@@ -279,10 +445,12 @@ static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 *c, unsigned int
 	}
 	return size;
 }
+#endif
 
 static struct cpu_dev amd_cpu_dev __cpuinitdata = {
 	.c_vendor	= "AMD",
 	.c_ident	= { "AuthenticAMD" },
+#ifdef CONFIG_X86_32
 	.c_models = {
 		{ .vendor = X86_VENDOR_AMD, .family = 4, .model_names =
 		  {
@@ -295,9 +463,11 @@ static struct cpu_dev amd_cpu_dev __cpuinitdata = {
 		  }
 		},
 	},
+	.c_size_cache	= amd_size_cache,
+#endif
 	.c_early_init   = early_init_amd,
 	.c_init		= init_amd,
-	.c_size_cache	= amd_size_cache,
+	.c_x86_vendor	= X86_VENDOR_AMD,
 };
 
-cpu_vendor_dev_register(X86_VENDOR_AMD, &amd_cpu_dev);
+cpu_dev_register(amd_cpu_dev);
diff --git a/arch/x86/kernel/cpu/amd_64.c b/arch/x86/kernel/cpu/amd_64.c
deleted file mode 100644
index d1692b2a41ff..000000000000
--- a/arch/x86/kernel/cpu/amd_64.c
+++ /dev/null
@@ -1,224 +0,0 @@
-#include <linux/init.h>
-#include <linux/mm.h>
-
-#include <asm/numa_64.h>
-#include <asm/mmconfig.h>
-#include <asm/cacheflush.h>
-
-#include <mach_apic.h>
-
-#include "cpu.h"
-
-int force_mwait __cpuinitdata;
-
-#ifdef CONFIG_NUMA
-static int __cpuinit nearby_node(int apicid)
-{
-	int i, node;
-
-	for (i = apicid - 1; i >= 0; i--) {
-		node = apicid_to_node[i];
-		if (node != NUMA_NO_NODE && node_online(node))
-			return node;
-	}
-	for (i = apicid + 1; i < MAX_LOCAL_APIC; i++) {
-		node = apicid_to_node[i];
-		if (node != NUMA_NO_NODE && node_online(node))
-			return node;
-	}
-	return first_node(node_online_map); /* Shouldn't happen */
-}
-#endif
-
-/*
- * On a AMD dual core setup the lower bits of the APIC id distingush the cores.
- * Assumes number of cores is a power of two.
- */
-static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c)
-{
-#ifdef CONFIG_SMP
-	unsigned bits;
-#ifdef CONFIG_NUMA
-	int cpu = smp_processor_id();
-	int node = 0;
-	unsigned apicid = hard_smp_processor_id();
-#endif
-	bits = c->x86_coreid_bits;
-
-	/* Low order bits define the core id (index of core in socket) */
-	c->cpu_core_id = c->initial_apicid & ((1 << bits)-1);
-	/* Convert the initial APIC ID into the socket ID */
-	c->phys_proc_id = c->initial_apicid >> bits;
-
-#ifdef CONFIG_NUMA
-	node = c->phys_proc_id;
-	if (apicid_to_node[apicid] != NUMA_NO_NODE)
-		node = apicid_to_node[apicid];
-	if (!node_online(node)) {
-		/* Two possibilities here:
-		   - The CPU is missing memory and no node was created.
-		   In that case try picking one from a nearby CPU
-		   - The APIC IDs differ from the HyperTransport node IDs
-		   which the K8 northbridge parsing fills in.
-		   Assume they are all increased by a constant offset,
-		   but in the same order as the HT nodeids.
-		   If that doesn't result in a usable node fall back to the
-		   path for the previous case.  */
-
-		int ht_nodeid = c->initial_apicid;
-
-		if (ht_nodeid >= 0 &&
-		    apicid_to_node[ht_nodeid] != NUMA_NO_NODE)
-			node = apicid_to_node[ht_nodeid];
-		/* Pick a nearby node */
-		if (!node_online(node))
-			node = nearby_node(apicid);
-	}
-	numa_set_node(cpu, node);
-
-	printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node);
-#endif
-#endif
-}
-
-static void __cpuinit early_init_amd_mc(struct cpuinfo_x86 *c)
-{
-#ifdef CONFIG_SMP
-	unsigned bits, ecx;
-
-	/* Multi core CPU? */
-	if (c->extended_cpuid_level < 0x80000008)
-		return;
-
-	ecx = cpuid_ecx(0x80000008);
-
-	c->x86_max_cores = (ecx & 0xff) + 1;
-
-	/* CPU telling us the core id bits shift? */
-	bits = (ecx >> 12) & 0xF;
-
-	/* Otherwise recompute */
-	if (bits == 0) {
-		while ((1 << bits) < c->x86_max_cores)
-			bits++;
-	}
-
-	c->x86_coreid_bits = bits;
-
-#endif
-}
-
-static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
-{
-	early_init_amd_mc(c);
-
-	/* c->x86_power is 8000_0007 edx. Bit 8 is constant TSC */
-	if (c->x86_power & (1<<8))
-		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
-
-	set_cpu_cap(c, X86_FEATURE_SYSCALL32);
-}
-
-static void __cpuinit init_amd(struct cpuinfo_x86 *c)
-{
-	unsigned level;
-
-#ifdef CONFIG_SMP
-	unsigned long value;
-
-	/*
-	 * Disable TLB flush filter by setting HWCR.FFDIS on K8
-	 * bit 6 of msr C001_0015
-	 *
-	 * Errata 63 for SH-B3 steppings
-	 * Errata 122 for all steppings (F+ have it disabled by default)
-	 */
-	if (c->x86 == 0xf) {
-		rdmsrl(MSR_K8_HWCR, value);
-		value |= 1 << 6;
-		wrmsrl(MSR_K8_HWCR, value);
-	}
-#endif
-
-	/* Bit 31 in normal CPUID used for nonstandard 3DNow ID;
-	   3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */
-	clear_cpu_cap(c, 0*32+31);
-
-	/* On C+ stepping K8 rep microcode works well for copy/memset */
-	if (c->x86 == 0xf) {
-		level = cpuid_eax(1);
-		if((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58)
-			set_cpu_cap(c, X86_FEATURE_REP_GOOD);
-	}
-	if (c->x86 == 0x10 || c->x86 == 0x11)
-		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
-
-	/* Enable workaround for FXSAVE leak */
-	if (c->x86 >= 6)
-		set_cpu_cap(c, X86_FEATURE_FXSAVE_LEAK);
-
-	level = get_model_name(c);
-	if (!level) {
-		switch (c->x86) {
-		case 0xf:
-			/* Should distinguish Models here, but this is only
-			   a fallback anyways. */
-			strcpy(c->x86_model_id, "Hammer");
-			break;
-		}
-	}
-	display_cacheinfo(c);
-
-	/* Multi core CPU? */
-	if (c->extended_cpuid_level >= 0x80000008)
-		amd_detect_cmp(c);
-
-	if (c->extended_cpuid_level >= 0x80000006 &&
-		(cpuid_edx(0x80000006) & 0xf000))
-		num_cache_leaves = 4;
-	else
-		num_cache_leaves = 3;
-
-	if (c->x86 >= 0xf && c->x86 <= 0x11)
-		set_cpu_cap(c, X86_FEATURE_K8);
-
-	/* MFENCE stops RDTSC speculation */
-	set_cpu_cap(c, X86_FEATURE_MFENCE_RDTSC);
-
-	if (c->x86 == 0x10) {
-		/* do this for boot cpu */
-		if (c == &boot_cpu_data)
-			check_enable_amd_mmconf_dmi();
-
-		fam10h_check_enable_mmcfg();
-	}
-
-	if (c == &boot_cpu_data && c->x86 >= 0xf && c->x86 <= 0x11) {
-		unsigned long long tseg;
-
-		/*
-		 * Split up direct mapping around the TSEG SMM area.
-		 * Don't do it for gbpages because there seems very little
-		 * benefit in doing so.
-		 */
-		if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg)) {
-		    printk(KERN_DEBUG "tseg: %010llx\n", tseg);
-		    if ((tseg>>PMD_SHIFT) <
-				(max_low_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) ||
-			((tseg>>PMD_SHIFT) <
-				(max_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) &&
-			 (tseg>>PMD_SHIFT) >= (1ULL<<(32 - PMD_SHIFT))))
-			set_memory_4k((unsigned long)__va(tseg), 1);
-		}
-	}
-}
-
-static struct cpu_dev amd_cpu_dev __cpuinitdata = {
-	.c_vendor	= "AMD",
-	.c_ident	= { "AuthenticAMD" },
-	.c_early_init   = early_init_amd,
-	.c_init		= init_amd,
-};
-
-cpu_vendor_dev_register(X86_VENDOR_AMD, &amd_cpu_dev);
-
diff --git a/arch/x86/kernel/cpu/centaur.c b/arch/x86/kernel/cpu/centaur.c
index a0534c04d38a..89bfdd9cacc6 100644
--- a/arch/x86/kernel/cpu/centaur.c
+++ b/arch/x86/kernel/cpu/centaur.c
@@ -289,7 +289,6 @@ static void __cpuinit init_c3(struct cpuinfo_x86 *c)
 	if (c->x86_model >= 6 && c->x86_model < 9)
 		set_cpu_cap(c, X86_FEATURE_3DNOW);
 
-	get_model_name(c);
 	display_cacheinfo(c);
 }
 
@@ -475,6 +474,7 @@ static struct cpu_dev centaur_cpu_dev __cpuinitdata = {
 	.c_early_init	= early_init_centaur,
 	.c_init		= init_centaur,
 	.c_size_cache	= centaur_size_cache,
+	.c_x86_vendor	= X86_VENDOR_CENTAUR,
 };
 
-cpu_vendor_dev_register(X86_VENDOR_CENTAUR, &centaur_cpu_dev);
+cpu_dev_register(centaur_cpu_dev);
diff --git a/arch/x86/kernel/cpu/centaur_64.c b/arch/x86/kernel/cpu/centaur_64.c
index 1d181c40e2e1..a1625f5a1e78 100644
--- a/arch/x86/kernel/cpu/centaur_64.c
+++ b/arch/x86/kernel/cpu/centaur_64.c
@@ -16,9 +16,10 @@ static void __cpuinit early_init_centaur(struct cpuinfo_x86 *c)
 
 static void __cpuinit init_centaur(struct cpuinfo_x86 *c)
 {
+	early_init_centaur(c);
+
 	if (c->x86 == 0x6 && c->x86_model >= 0xf) {
 		c->x86_cache_alignment = c->x86_clflush_size * 2;
-		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
 		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
 	}
 	set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC);
@@ -29,7 +30,8 @@ static struct cpu_dev centaur_cpu_dev __cpuinitdata = {
 	.c_ident	= { "CentaurHauls" },
 	.c_early_init	= early_init_centaur,
 	.c_init		= init_centaur,
+	.c_x86_vendor	= X86_VENDOR_CENTAUR,
 };
 
-cpu_vendor_dev_register(X86_VENDOR_CENTAUR, &centaur_cpu_dev);
+cpu_dev_register(centaur_cpu_dev);
 
diff --git a/arch/x86/kernel/cpu/cmpxchg.c b/arch/x86/kernel/cpu/cmpxchg.c
new file mode 100644
index 000000000000..2056ccf572cc
--- /dev/null
+++ b/arch/x86/kernel/cpu/cmpxchg.c
@@ -0,0 +1,72 @@
+/*
+ * cmpxchg*() fallbacks for CPU not supporting these instructions
+ */
+
+#include <linux/kernel.h>
+#include <linux/smp.h>
+#include <linux/module.h>
+
+#ifndef CONFIG_X86_CMPXCHG
+unsigned long cmpxchg_386_u8(volatile void *ptr, u8 old, u8 new)
+{
+	u8 prev;
+	unsigned long flags;
+
+	/* Poor man's cmpxchg for 386. Unsuitable for SMP */
+	local_irq_save(flags);
+	prev = *(u8 *)ptr;
+	if (prev == old)
+		*(u8 *)ptr = new;
+	local_irq_restore(flags);
+	return prev;
+}
+EXPORT_SYMBOL(cmpxchg_386_u8);
+
+unsigned long cmpxchg_386_u16(volatile void *ptr, u16 old, u16 new)
+{
+	u16 prev;
+	unsigned long flags;
+
+	/* Poor man's cmpxchg for 386. Unsuitable for SMP */
+	local_irq_save(flags);
+	prev = *(u16 *)ptr;
+	if (prev == old)
+		*(u16 *)ptr = new;
+	local_irq_restore(flags);
+	return prev;
+}
+EXPORT_SYMBOL(cmpxchg_386_u16);
+
+unsigned long cmpxchg_386_u32(volatile void *ptr, u32 old, u32 new)
+{
+	u32 prev;
+	unsigned long flags;
+
+	/* Poor man's cmpxchg for 386. Unsuitable for SMP */
+	local_irq_save(flags);
+	prev = *(u32 *)ptr;
+	if (prev == old)
+		*(u32 *)ptr = new;
+	local_irq_restore(flags);
+	return prev;
+}
+EXPORT_SYMBOL(cmpxchg_386_u32);
+#endif
+
+#ifndef CONFIG_X86_CMPXCHG64
+unsigned long long cmpxchg_486_u64(volatile void *ptr, u64 old, u64 new)
+{
+	u64 prev;
+	unsigned long flags;
+
+	/* Poor man's cmpxchg8b for 386 and 486. Unsuitable for SMP */
+	local_irq_save(flags);
+	prev = *(u64 *)ptr;
+	if (prev == old)
+		*(u64 *)ptr = new;
+	local_irq_restore(flags);
+	return prev;
+}
+EXPORT_SYMBOL(cmpxchg_486_u64);
+#endif
+
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 4e456bd955bb..25581dcb280e 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -1,28 +1,62 @@
 #include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
 #include <linux/string.h>
+#include <linux/bootmem.h>
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <linux/kgdb.h>
+#include <linux/topology.h>
 #include <linux/delay.h>
 #include <linux/smp.h>
-#include <linux/module.h>
 #include <linux/percpu.h>
-#include <linux/bootmem.h>
-#include <asm/processor.h>
 #include <asm/i387.h>
 #include <asm/msr.h>
 #include <asm/io.h>
+#include <asm/linkage.h>
 #include <asm/mmu_context.h>
 #include <asm/mtrr.h>
 #include <asm/mce.h>
 #include <asm/pat.h>
 #include <asm/asm.h>
+#include <asm/numa.h>
 #ifdef CONFIG_X86_LOCAL_APIC
 #include <asm/mpspec.h>
 #include <asm/apic.h>
 #include <mach_apic.h>
+#include <asm/genapic.h>
 #endif
 
+#include <asm/pda.h>
+#include <asm/pgtable.h>
+#include <asm/processor.h>
+#include <asm/desc.h>
+#include <asm/atomic.h>
+#include <asm/proto.h>
+#include <asm/sections.h>
+#include <asm/setup.h>
+
 #include "cpu.h"
 
+static struct cpu_dev *this_cpu __cpuinitdata;
+
+#ifdef CONFIG_X86_64
+/* We need valid kernel segments for data and code in long mode too
+ * IRET will check the segment types  kkeil 2000/10/28
+ * Also sysret mandates a special GDT layout
+ */
+/* The TLS descriptors are currently at a different place compared to i386.
+   Hopefully nobody expects them at a fixed place (Wine?) */
 DEFINE_PER_CPU(struct gdt_page, gdt_page) = { .gdt = {
+	[GDT_ENTRY_KERNEL32_CS] = { { { 0x0000ffff, 0x00cf9b00 } } },
+	[GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00af9b00 } } },
+	[GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9300 } } },
+	[GDT_ENTRY_DEFAULT_USER32_CS] = { { { 0x0000ffff, 0x00cffb00 } } },
+	[GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff300 } } },
+	[GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00affb00 } } },
+} };
+#else
+DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = {
 	[GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00cf9a00 } } },
 	[GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9200 } } },
 	[GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00cffa00 } } },
@@ -56,17 +90,157 @@ DEFINE_PER_CPU(struct gdt_page, gdt_page) = { .gdt = {
 	[GDT_ENTRY_ESPFIX_SS] = { { { 0x00000000, 0x00c09200 } } },
 	[GDT_ENTRY_PERCPU] = { { { 0x00000000, 0x00000000 } } },
 } };
+#endif
 EXPORT_PER_CPU_SYMBOL_GPL(gdt_page);
 
-__u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata;
-
+#ifdef CONFIG_X86_32
 static int cachesize_override __cpuinitdata = -1;
 static int disable_x86_serial_nr __cpuinitdata = 1;
 
-struct cpu_dev *cpu_devs[X86_VENDOR_NUM] = {};
+static int __init cachesize_setup(char *str)
+{
+	get_option(&str, &cachesize_override);
+	return 1;
+}
+__setup("cachesize=", cachesize_setup);
+
+static int __init x86_fxsr_setup(char *s)
+{
+	setup_clear_cpu_cap(X86_FEATURE_FXSR);
+	setup_clear_cpu_cap(X86_FEATURE_XMM);
+	return 1;
+}
+__setup("nofxsr", x86_fxsr_setup);
+
+static int __init x86_sep_setup(char *s)
+{
+	setup_clear_cpu_cap(X86_FEATURE_SEP);
+	return 1;
+}
+__setup("nosep", x86_sep_setup);
+
+/* Standard macro to see if a specific flag is changeable */
+static inline int flag_is_changeable_p(u32 flag)
+{
+	u32 f1, f2;
+
+	/*
+	 * Cyrix and IDT cpus allow disabling of CPUID
+	 * so the code below may return different results
+	 * when it is executed before and after enabling
+	 * the CPUID. Add "volatile" to not allow gcc to
+	 * optimize the subsequent calls to this function.
+	 */
+	asm volatile ("pushfl\n\t"
+		      "pushfl\n\t"
+		      "popl %0\n\t"
+		      "movl %0,%1\n\t"
+		      "xorl %2,%0\n\t"
+		      "pushl %0\n\t"
+		      "popfl\n\t"
+		      "pushfl\n\t"
+		      "popl %0\n\t"
+		      "popfl\n\t"
+		      : "=&r" (f1), "=&r" (f2)
+		      : "ir" (flag));
+
+	return ((f1^f2) & flag) != 0;
+}
+
+/* Probe for the CPUID instruction */
+static int __cpuinit have_cpuid_p(void)
+{
+	return flag_is_changeable_p(X86_EFLAGS_ID);
+}
+
+static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
+{
+	if (cpu_has(c, X86_FEATURE_PN) && disable_x86_serial_nr) {
+		/* Disable processor serial number */
+		unsigned long lo, hi;
+		rdmsr(MSR_IA32_BBL_CR_CTL, lo, hi);
+		lo |= 0x200000;
+		wrmsr(MSR_IA32_BBL_CR_CTL, lo, hi);
+		printk(KERN_NOTICE "CPU serial number disabled.\n");
+		clear_cpu_cap(c, X86_FEATURE_PN);
+
+		/* Disabling the serial number may affect the cpuid level */
+		c->cpuid_level = cpuid_eax(0);
+	}
+}
+
+static int __init x86_serial_nr_setup(char *s)
+{
+	disable_x86_serial_nr = 0;
+	return 1;
+}
+__setup("serialnumber", x86_serial_nr_setup);
+#else
+static inline int flag_is_changeable_p(u32 flag)
+{
+	return 1;
+}
+/* Probe for the CPUID instruction */
+static inline int have_cpuid_p(void)
+{
+	return 1;
+}
+static inline void squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
+{
+}
+#endif
+
+/*
+ * Naming convention should be: <Name> [(<Codename>)]
+ * This table only is used unless init_<vendor>() below doesn't set it;
+ * in particular, if CPUID levels 0x80000002..4 are supported, this isn't used
+ *
+ */
+
+/* Look up CPU names by table lookup. */
+static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c)
+{
+	struct cpu_model_info *info;
+
+	if (c->x86_model >= 16)
+		return NULL;	/* Range check */
+
+	if (!this_cpu)
+		return NULL;
+
+	info = this_cpu->c_models;
+
+	while (info && info->family) {
+		if (info->family == c->x86)
+			return info->model_names[c->x86_model];
+		info++;
+	}
+	return NULL;		/* Not found */
+}
+
+__u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata;
+
+/* Current gdt points %fs at the "master" per-cpu area: after this,
+ * it's on the real one. */
+void switch_to_new_gdt(void)
+{
+	struct desc_ptr gdt_descr;
+
+	gdt_descr.address = (long)get_cpu_gdt_table(smp_processor_id());
+	gdt_descr.size = GDT_SIZE - 1;
+	load_gdt(&gdt_descr);
+#ifdef CONFIG_X86_32
+	asm("mov %0, %%fs" : : "r" (__KERNEL_PERCPU) : "memory");
+#endif
+}
+
+static struct cpu_dev *cpu_devs[X86_VENDOR_NUM] = {};
 
 static void __cpuinit default_init(struct cpuinfo_x86 *c)
 {
+#ifdef CONFIG_X86_64
+	display_cacheinfo(c);
+#else
 	/* Not much we can do here... */
 	/* Check if at least it has cpuid */
 	if (c->cpuid_level == -1) {
@@ -76,28 +250,22 @@ static void __cpuinit default_init(struct cpuinfo_x86 *c)
 		else if (c->x86 == 3)
 			strcpy(c->x86_model_id, "386");
 	}
+#endif
 }
 
 static struct cpu_dev __cpuinitdata default_cpu = {
 	.c_init	= default_init,
 	.c_vendor = "Unknown",
+	.c_x86_vendor = X86_VENDOR_UNKNOWN,
 };
-static struct cpu_dev *this_cpu __cpuinitdata = &default_cpu;
-
-static int __init cachesize_setup(char *str)
-{
-	get_option(&str, &cachesize_override);
-	return 1;
-}
-__setup("cachesize=", cachesize_setup);
 
-int __cpuinit get_model_name(struct cpuinfo_x86 *c)
+static void __cpuinit get_model_name(struct cpuinfo_x86 *c)
 {
 	unsigned int *v;
 	char *p, *q;
 
-	if (cpuid_eax(0x80000000) < 0x80000004)
-		return 0;
+	if (c->extended_cpuid_level < 0x80000004)
+		return;
 
 	v = (unsigned int *) c->x86_model_id;
 	cpuid(0x80000002, &v[0], &v[1], &v[2], &v[3]);
@@ -116,30 +284,34 @@ int __cpuinit get_model_name(struct cpuinfo_x86 *c)
 	     while (q <= &c->x86_model_id[48])
 		  *q++ = '\0';	/* Zero-pad the rest */
 	}
-
-	return 1;
 }
 
-
 void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
 {
-	unsigned int n, dummy, ecx, edx, l2size;
+	unsigned int n, dummy, ebx, ecx, edx, l2size;
 
-	n = cpuid_eax(0x80000000);
+	n = c->extended_cpuid_level;
 
 	if (n >= 0x80000005) {
-		cpuid(0x80000005, &dummy, &dummy, &ecx, &edx);
+		cpuid(0x80000005, &dummy, &ebx, &ecx, &edx);
 		printk(KERN_INFO "CPU: L1 I Cache: %dK (%d bytes/line), D cache %dK (%d bytes/line)\n",
-			edx>>24, edx&0xFF, ecx>>24, ecx&0xFF);
-		c->x86_cache_size = (ecx>>24)+(edx>>24);
+				edx>>24, edx&0xFF, ecx>>24, ecx&0xFF);
+		c->x86_cache_size = (ecx>>24) + (edx>>24);
+#ifdef CONFIG_X86_64
+		/* On K8 L1 TLB is inclusive, so don't count it */
+		c->x86_tlbsize = 0;
+#endif
 	}
 
 	if (n < 0x80000006)	/* Some chips just has a large L1. */
 		return;
 
-	ecx = cpuid_ecx(0x80000006);
+	cpuid(0x80000006, &dummy, &ebx, &ecx, &edx);
 	l2size = ecx >> 16;
 
+#ifdef CONFIG_X86_64
+	c->x86_tlbsize += ((ebx >> 16) & 0xfff) + (ebx & 0xfff);
+#else
 	/* do processor-specific cache resizing */
 	if (this_cpu->c_size_cache)
 		l2size = this_cpu->c_size_cache(c, l2size);
@@ -150,116 +322,106 @@ void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
 
 	if (l2size == 0)
 		return;		/* Again, no L2 cache is possible */
+#endif
 
 	c->x86_cache_size = l2size;
 
 	printk(KERN_INFO "CPU: L2 Cache: %dK (%d bytes/line)\n",
-	       l2size, ecx & 0xFF);
+			l2size, ecx & 0xFF);
 }
 
-/*
- * Naming convention should be: <Name> [(<Codename>)]
- * This table only is used unless init_<vendor>() below doesn't set it;
- * in particular, if CPUID levels 0x80000002..4 are supported, this isn't used
- *
- */
-
-/* Look up CPU names by table lookup. */
-static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c)
+void __cpuinit detect_ht(struct cpuinfo_x86 *c)
 {
-	struct cpu_model_info *info;
+#ifdef CONFIG_X86_HT
+	u32 eax, ebx, ecx, edx;
+	int index_msb, core_bits;
 
-	if (c->x86_model >= 16)
-		return NULL;	/* Range check */
+	if (!cpu_has(c, X86_FEATURE_HT))
+		return;
 
-	if (!this_cpu)
-		return NULL;
+	if (cpu_has(c, X86_FEATURE_CMP_LEGACY))
+		goto out;
 
-	info = this_cpu->c_models;
+	if (cpu_has(c, X86_FEATURE_XTOPOLOGY))
+		return;
 
-	while (info && info->family) {
-		if (info->family == c->x86)
-			return info->model_names[c->x86_model];
-		info++;
+	cpuid(1, &eax, &ebx, &ecx, &edx);
+
+	smp_num_siblings = (ebx & 0xff0000) >> 16;
+
+	if (smp_num_siblings == 1) {
+		printk(KERN_INFO  "CPU: Hyper-Threading is disabled\n");
+	} else if (smp_num_siblings > 1) {
+
+		if (smp_num_siblings > NR_CPUS) {
+			printk(KERN_WARNING "CPU: Unsupported number of siblings %d",
+					smp_num_siblings);
+			smp_num_siblings = 1;
+			return;
+		}
+
+		index_msb = get_count_order(smp_num_siblings);
+#ifdef CONFIG_X86_64
+		c->phys_proc_id = phys_pkg_id(index_msb);
+#else
+		c->phys_proc_id = phys_pkg_id(c->initial_apicid, index_msb);
+#endif
+
+		smp_num_siblings = smp_num_siblings / c->x86_max_cores;
+
+		index_msb = get_count_order(smp_num_siblings);
+
+		core_bits = get_count_order(c->x86_max_cores);
+
+#ifdef CONFIG_X86_64
+		c->cpu_core_id = phys_pkg_id(index_msb) &
+					       ((1 << core_bits) - 1);
+#else
+		c->cpu_core_id = phys_pkg_id(c->initial_apicid, index_msb) &
+					       ((1 << core_bits) - 1);
+#endif
 	}
-	return NULL;		/* Not found */
-}
 
+out:
+	if ((c->x86_max_cores * smp_num_siblings) > 1) {
+		printk(KERN_INFO  "CPU: Physical Processor ID: %d\n",
+		       c->phys_proc_id);
+		printk(KERN_INFO  "CPU: Processor Core ID: %d\n",
+		       c->cpu_core_id);
+	}
+#endif
+}
 
-static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c, int early)
+static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c)
 {
 	char *v = c->x86_vendor_id;
 	int i;
 	static int printed;
 
 	for (i = 0; i < X86_VENDOR_NUM; i++) {
-		if (cpu_devs[i]) {
-			if (!strcmp(v, cpu_devs[i]->c_ident[0]) ||
-			    (cpu_devs[i]->c_ident[1] &&
-			     !strcmp(v, cpu_devs[i]->c_ident[1]))) {
-				c->x86_vendor = i;
-				if (!early)
-					this_cpu = cpu_devs[i];
-				return;
-			}
+		if (!cpu_devs[i])
+			break;
+
+		if (!strcmp(v, cpu_devs[i]->c_ident[0]) ||
+		    (cpu_devs[i]->c_ident[1] &&
+		     !strcmp(v, cpu_devs[i]->c_ident[1]))) {
+			this_cpu = cpu_devs[i];
+			c->x86_vendor = this_cpu->c_x86_vendor;
+			return;
 		}
 	}
+
 	if (!printed) {
 		printed++;
-		printk(KERN_ERR "CPU: Vendor unknown, using generic init.\n");
+		printk(KERN_ERR "CPU: vendor_id '%s' unknown, using generic init.\n", v);
 		printk(KERN_ERR "CPU: Your system may be unstable.\n");
 	}
+
 	c->x86_vendor = X86_VENDOR_UNKNOWN;
 	this_cpu = &default_cpu;
 }
 
-
-static int __init x86_fxsr_setup(char *s)
-{
-	setup_clear_cpu_cap(X86_FEATURE_FXSR);
-	setup_clear_cpu_cap(X86_FEATURE_XMM);
-	return 1;
-}
-__setup("nofxsr", x86_fxsr_setup);
-
-
-static int __init x86_sep_setup(char *s)
-{
-	setup_clear_cpu_cap(X86_FEATURE_SEP);
-	return 1;
-}
-__setup("nosep", x86_sep_setup);
-
-
-/* Standard macro to see if a specific flag is changeable */
-static inline int flag_is_changeable_p(u32 flag)
-{
-	u32 f1, f2;
-
-	asm("pushfl\n\t"
-	    "pushfl\n\t"
-	    "popl %0\n\t"
-	    "movl %0,%1\n\t"
-	    "xorl %2,%0\n\t"
-	    "pushl %0\n\t"
-	    "popfl\n\t"
-	    "pushfl\n\t"
-	    "popl %0\n\t"
-	    "popfl\n\t"
-	    : "=&r" (f1), "=&r" (f2)
-	    : "ir" (flag));
-
-	return ((f1^f2) & flag) != 0;
-}
-
-
-/* Probe for the CPUID instruction */
-static int __cpuinit have_cpuid_p(void)
-{
-	return flag_is_changeable_p(X86_EFLAGS_ID);
-}
-
-void __init cpu_detect(struct cpuinfo_x86 *c)
+void __cpuinit cpu_detect(struct cpuinfo_x86 *c)
 {
 	/* Get vendor name */
 	cpuid(0x00000000, (unsigned int *)&c->cpuid_level,
@@ -268,50 +430,87 @@ void __init cpu_detect(struct cpuinfo_x86 *c)
 	      (unsigned int *)&c->x86_vendor_id[4]);
 
 	c->x86 = 4;
+	/* Intel-defined flags: level 0x00000001 */
 	if (c->cpuid_level >= 0x00000001) {
 		u32 junk, tfms, cap0, misc;
 		cpuid(0x00000001, &tfms, &misc, &junk, &cap0);
-		c->x86 = (tfms >> 8) & 15;
-		c->x86_model = (tfms >> 4) & 15;
+		c->x86 = (tfms >> 8) & 0xf;
+		c->x86_model = (tfms >> 4) & 0xf;
+		c->x86_mask = tfms & 0xf;
 		if (c->x86 == 0xf)
 			c->x86 += (tfms >> 20) & 0xff;
 		if (c->x86 >= 0x6)
-			c->x86_model += ((tfms >> 16) & 0xF) << 4;
-		c->x86_mask = tfms & 15;
+			c->x86_model += ((tfms >> 16) & 0xf) << 4;
 		if (cap0 & (1<<19)) {
-			c->x86_cache_alignment = ((misc >> 8) & 0xff) * 8;
 			c->x86_clflush_size = ((misc >> 8) & 0xff) * 8;
+			c->x86_cache_alignment = c->x86_clflush_size;
 		}
 	}
 }
-static void __cpuinit early_get_cap(struct cpuinfo_x86 *c)
+
+static void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c)
 {
 	u32 tfms, xlvl;
-	unsigned int ebx;
+	u32 ebx;
 
-	memset(&c->x86_capability, 0, sizeof c->x86_capability);
-	if (have_cpuid_p()) {
-		/* Intel-defined flags: level 0x00000001 */
-		if (c->cpuid_level >= 0x00000001) {
-			u32 capability, excap;
-			cpuid(0x00000001, &tfms, &ebx, &excap, &capability);
-			c->x86_capability[0] = capability;
-			c->x86_capability[4] = excap;
-		}
+	/* Intel-defined flags: level 0x00000001 */
+	if (c->cpuid_level >= 0x00000001) {
+		u32 capability, excap;
+		cpuid(0x00000001, &tfms, &ebx, &excap, &capability);
+		c->x86_capability[0] = capability;
+		c->x86_capability[4] = excap;
+	}
 
-		/* AMD-defined flags: level 0x80000001 */
-		xlvl = cpuid_eax(0x80000000);
-		if ((xlvl & 0xffff0000) == 0x80000000) {
-			if (xlvl >= 0x80000001) {
-				c->x86_capability[1] = cpuid_edx(0x80000001);
-				c->x86_capability[6] = cpuid_ecx(0x80000001);
-			}
+	/* AMD-defined flags: level 0x80000001 */
+	xlvl = cpuid_eax(0x80000000);
+	c->extended_cpuid_level = xlvl;
+	if ((xlvl & 0xffff0000) == 0x80000000) {
+		if (xlvl >= 0x80000001) {
+			c->x86_capability[1] = cpuid_edx(0x80000001);
+			c->x86_capability[6] = cpuid_ecx(0x80000001);
 		}
+	}
+
+#ifdef CONFIG_X86_64
+	if (c->extended_cpuid_level >= 0x80000008) {
+		u32 eax = cpuid_eax(0x80000008);
 
+		c->x86_virt_bits = (eax >> 8) & 0xff;
+		c->x86_phys_bits = eax & 0xff;
 	}
+#endif
+
+	if (c->extended_cpuid_level >= 0x80000007)
+		c->x86_power = cpuid_edx(0x80000007);
 
 }
 
+static void __cpuinit identify_cpu_without_cpuid(struct cpuinfo_x86 *c)
+{
+#ifdef CONFIG_X86_32
+	int i;
+
+	/*
+	 * First of all, decide if this is a 486 or higher
+	 * It's a 486 if we can modify the AC flag
+	 */
+	if (flag_is_changeable_p(X86_EFLAGS_AC))
+		c->x86 = 4;
+	else
+		c->x86 = 3;
+
+	for (i = 0; i < X86_VENDOR_NUM; i++)
+		if (cpu_devs[i] && cpu_devs[i]->c_identify) {
+			c->x86_vendor_id[0] = 0;
+			cpu_devs[i]->c_identify(c);
+			if (c->x86_vendor_id[0]) {
+				get_cpu_vendor(c);
+				break;
+			}
+		}
+#endif
+}
+
 /*
  * Do minimum CPU detection early.
  * Fields really needed: vendor, cpuid_level, family, model, mask,
@@ -321,25 +520,61 @@ static void __cpuinit early_get_cap(struct cpuinfo_x86 *c)
  * WARNING: this function is only called on the BP.  Don't add code here
  * that is supposed to run on all CPUs.
  */
-static void __init early_cpu_detect(void)
+static void __init early_identify_cpu(struct cpuinfo_x86 *c)
 {
-	struct cpuinfo_x86 *c = &boot_cpu_data;
-
-	c->x86_cache_alignment = 32;
+#ifdef CONFIG_X86_64
+	c->x86_clflush_size = 64;
+#else
 	c->x86_clflush_size = 32;
+#endif
+	c->x86_cache_alignment = c->x86_clflush_size;
+
+	memset(&c->x86_capability, 0, sizeof c->x86_capability);
+	c->extended_cpuid_level = 0;
+
+	if (!have_cpuid_p())
+		identify_cpu_without_cpuid(c);
 
+	/* cyrix could have cpuid enabled via c_identify()*/
 	if (!have_cpuid_p())
 		return;
 
 	cpu_detect(c);
 
-	get_cpu_vendor(c, 1);
+	get_cpu_vendor(c);
+
+	get_cpu_cap(c);
 
-	early_get_cap(c);
+	if (this_cpu->c_early_init)
+		this_cpu->c_early_init(c);
 
-	if (c->x86_vendor != X86_VENDOR_UNKNOWN &&
-	    cpu_devs[c->x86_vendor]->c_early_init)
-		cpu_devs[c->x86_vendor]->c_early_init(c);
+	validate_pat_support(c);
+}
+
+void __init early_cpu_init(void)
+{
+	struct cpu_dev **cdev;
+	int count = 0;
+
+	printk("KERNEL supported cpus:\n");
+	for (cdev = __x86_cpu_dev_start; cdev < __x86_cpu_dev_end; cdev++) {
+		struct cpu_dev *cpudev = *cdev;
+		unsigned int j;
+
+		if (count >= X86_VENDOR_NUM)
+			break;
+		cpu_devs[count] = cpudev;
+		count++;
+
+		for (j = 0; j < 2; j++) {
+			if (!cpudev->c_ident[j])
+				continue;
+			printk("  %s %s\n", cpudev->c_vendor,
+				cpudev->c_ident[j]);
+		}
+	}
+
+	early_identify_cpu(&boot_cpu_data);
 }
 
 /*
@@ -357,86 +592,41 @@ static void __cpuinit detect_nopl(struct cpuinfo_x86 *c)
 
 static void __cpuinit generic_identify(struct cpuinfo_x86 *c)
 {
-	u32 tfms, xlvl;
-	unsigned int ebx;
-
-	if (have_cpuid_p()) {
-		/* Get vendor name */
-		cpuid(0x00000000, (unsigned int *)&c->cpuid_level,
-		      (unsigned int *)&c->x86_vendor_id[0],
-		      (unsigned int *)&c->x86_vendor_id[8],
-		      (unsigned int *)&c->x86_vendor_id[4]);
-
-		get_cpu_vendor(c, 0);
-		/* Initialize the standard set of capabilities */
-		/* Note that the vendor-specific code below might override */
-		/* Intel-defined flags: level 0x00000001 */
-		if (c->cpuid_level >= 0x00000001) {
-			u32 capability, excap;
-			cpuid(0x00000001, &tfms, &ebx, &excap, &capability);
-			c->x86_capability[0] = capability;
-			c->x86_capability[4] = excap;
-			c->x86 = (tfms >> 8) & 15;
-			c->x86_model = (tfms >> 4) & 15;
-			if (c->x86 == 0xf)
-				c->x86 += (tfms >> 20) & 0xff;
-			if (c->x86 >= 0x6)
-				c->x86_model += ((tfms >> 16) & 0xF) << 4;
-			c->x86_mask = tfms & 15;
-			c->initial_apicid = (ebx >> 24) & 0xFF;
-#ifdef CONFIG_X86_HT
-			c->apicid = phys_pkg_id(c->initial_apicid, 0);
-			c->phys_proc_id = c->initial_apicid;
-#else
-			c->apicid = c->initial_apicid;
-#endif
-			if (test_cpu_cap(c, X86_FEATURE_CLFLSH))
-				c->x86_clflush_size = ((ebx >> 8) & 0xff) * 8;
-		} else {
-			/* Have CPUID level 0 only - unheard of */
-			c->x86 = 4;
-		}
+	c->extended_cpuid_level = 0;
 
-		/* AMD-defined flags: level 0x80000001 */
-		xlvl = cpuid_eax(0x80000000);
-		if ((xlvl & 0xffff0000) == 0x80000000) {
-			if (xlvl >= 0x80000001) {
-				c->x86_capability[1] = cpuid_edx(0x80000001);
-				c->x86_capability[6] = cpuid_ecx(0x80000001);
-			}
-			if (xlvl >= 0x80000004)
-				get_model_name(c); /* Default name */
-		}
+	if (!have_cpuid_p())
+		identify_cpu_without_cpuid(c);
 
-		init_scattered_cpuid_features(c);
-		detect_nopl(c);
-	}
-}
+	/* cyrix could have cpuid enabled via c_identify()*/
+	if (!have_cpuid_p())
+		return;
 
-static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
-{
-	if (cpu_has(c, X86_FEATURE_PN) && disable_x86_serial_nr) {
-		/* Disable processor serial number */
-		unsigned long lo, hi;
-		rdmsr(MSR_IA32_BBL_CR_CTL, lo, hi);
-		lo |= 0x200000;
-		wrmsr(MSR_IA32_BBL_CR_CTL, lo, hi);
-		printk(KERN_NOTICE "CPU serial number disabled.\n");
-		clear_cpu_cap(c, X86_FEATURE_PN);
+	cpu_detect(c);
 
-		/* Disabling the serial number may affect the cpuid level */
-		c->cpuid_level = cpuid_eax(0);
-	}
-}
+	get_cpu_vendor(c);
 
-static int __init x86_serial_nr_setup(char *s)
-{
-	disable_x86_serial_nr = 0;
-	return 1;
-}
-__setup("serialnumber", x86_serial_nr_setup);
+	get_cpu_cap(c);
+
+	if (c->cpuid_level >= 0x00000001) {
+		c->initial_apicid = (cpuid_ebx(1) >> 24) & 0xFF;
+#ifdef CONFIG_X86_32
+# ifdef CONFIG_X86_HT
+		c->apicid = phys_pkg_id(c->initial_apicid, 0);
+# else
+		c->apicid = c->initial_apicid;
+# endif
+#endif
 
+#ifdef CONFIG_X86_HT
+		c->phys_proc_id = c->initial_apicid;
+#endif
+	}
 
+	get_model_name(c); /* Default name */
+
+	init_scattered_cpuid_features(c);
+	detect_nopl(c);
+}
 
 /*
  * This does the hard work of actually picking apart the CPU stuff...
@@ -448,30 +638,29 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 	c->loops_per_jiffy = loops_per_jiffy;
 	c->x86_cache_size = -1;
 	c->x86_vendor = X86_VENDOR_UNKNOWN;
-	c->cpuid_level = -1;	/* CPUID not detected */
 	c->x86_model = c->x86_mask = 0;	/* So far unknown... */
 	c->x86_vendor_id[0] = '\0'; /* Unset */
 	c->x86_model_id[0] = '\0';  /* Unset */
 	c->x86_max_cores = 1;
+	c->x86_coreid_bits = 0;
+#ifdef CONFIG_X86_64
+	c->x86_clflush_size = 64;
+#else
+	c->cpuid_level = -1;	/* CPUID not detected */
 	c->x86_clflush_size = 32;
+#endif
+	c->x86_cache_alignment = c->x86_clflush_size;
 	memset(&c->x86_capability, 0, sizeof c->x86_capability);
 
-	if (!have_cpuid_p()) {
-		/*
-		 * First of all, decide if this is a 486 or higher
-		 * It's a 486 if we can modify the AC flag
-		 */
-		if (flag_is_changeable_p(X86_EFLAGS_AC))
-			c->x86 = 4;
-		else
-			c->x86 = 3;
-	}
-
 	generic_identify(c);
 
 	if (this_cpu->c_identify)
 		this_cpu->c_identify(c);
 
+#ifdef CONFIG_X86_64
+	c->apicid = phys_pkg_id(0);
+#endif
+
 	/*
 	 * Vendor-specific initialization.  In this section we
 	 * canonicalize the feature flags, meaning if there are
@@ -505,6 +694,10 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 				c->x86, c->x86_model);
 	}
 
+#ifdef CONFIG_X86_64
+	detect_ht(c);
+#endif
+
 	/*
 	 * On SMP, boot_cpu_data holds the common feature set between
 	 * all CPUs; so make sure that we indicate which features are
@@ -513,7 +706,7 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 	 */
 	if (c != &boot_cpu_data) {
 		/* AND the already accumulated flags with these */
-		for (i = 0 ; i < NCAPINTS ; i++)
+		for (i = 0; i < NCAPINTS; i++)
 			boot_cpu_data.x86_capability[i] &= c->x86_capability[i];
 	}
 
@@ -521,72 +714,91 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 	for (i = 0; i < NCAPINTS; i++)
 		c->x86_capability[i] &= ~cleared_cpu_caps[i];
 
+#ifdef CONFIG_X86_MCE
 	/* Init Machine Check Exception if available. */
 	mcheck_init(c);
+#endif
 
 	select_idle_routine(c);
+
+#if defined(CONFIG_NUMA) && defined(CONFIG_X86_64)
+	numa_add_cpu(smp_processor_id());
+#endif
 }
 
+#ifdef CONFIG_X86_64
+static void vgetcpu_set_mode(void)
+{
+	if (cpu_has(&boot_cpu_data, X86_FEATURE_RDTSCP))
+		vgetcpu_mode = VGETCPU_RDTSCP;
+	else
+		vgetcpu_mode = VGETCPU_LSL;
+}
+#endif
+
 void __init identify_boot_cpu(void)
 {
 	identify_cpu(&boot_cpu_data);
+#ifdef CONFIG_X86_32
 	sysenter_setup();
 	enable_sep_cpu();
+#else
+	vgetcpu_set_mode();
+#endif
 }
 
 void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c)
 {
 	BUG_ON(c == &boot_cpu_data);
 	identify_cpu(c);
+#ifdef CONFIG_X86_32
 	enable_sep_cpu();
+#endif
 	mtrr_ap_init();
 }
 
-#ifdef CONFIG_X86_HT
-void __cpuinit detect_ht(struct cpuinfo_x86 *c)
-{
-	u32 	eax, ebx, ecx, edx;
-	int 	index_msb, core_bits;
-
-	cpuid(1, &eax, &ebx, &ecx, &edx);
-
-	if (!cpu_has(c, X86_FEATURE_HT) || cpu_has(c, X86_FEATURE_CMP_LEGACY))
-		return;
-
-	smp_num_siblings = (ebx & 0xff0000) >> 16;
+struct msr_range {
+	unsigned min;
+	unsigned max;
+};
 
-	if (smp_num_siblings == 1) {
-		printk(KERN_INFO  "CPU: Hyper-Threading is disabled\n");
-	} else if (smp_num_siblings > 1) {
+static struct msr_range msr_range_array[] __cpuinitdata = {
+	{ 0x00000000, 0x00000418},
+	{ 0xc0000000, 0xc000040b},
+	{ 0xc0010000, 0xc0010142},
+	{ 0xc0011000, 0xc001103b},
+};
 
-		if (smp_num_siblings > NR_CPUS) {
-			printk(KERN_WARNING "CPU: Unsupported number of the "
-					"siblings %d", smp_num_siblings);
-			smp_num_siblings = 1;
-			return;
+static void __cpuinit print_cpu_msr(void)
+{
+	unsigned index;
+	u64 val;
+	int i;
+	unsigned index_min, index_max;
+
+	for (i = 0; i < ARRAY_SIZE(msr_range_array); i++) {
+		index_min = msr_range_array[i].min;
+		index_max = msr_range_array[i].max;
+		for (index = index_min; index < index_max; index++) {
+			if (rdmsrl_amd_safe(index, &val))
+				continue;
+			printk(KERN_INFO " MSR%08x: %016llx\n", index, val);
 		}
+	}
+}
 
-		index_msb = get_count_order(smp_num_siblings);
-		c->phys_proc_id = phys_pkg_id(c->initial_apicid, index_msb);
-
-		printk(KERN_INFO  "CPU: Physical Processor ID: %d\n",
-		       c->phys_proc_id);
-
-		smp_num_siblings = smp_num_siblings / c->x86_max_cores;
-
-		index_msb = get_count_order(smp_num_siblings) ;
-
-		core_bits = get_count_order(c->x86_max_cores);
+static int show_msr __cpuinitdata;
+static __init int setup_show_msr(char *arg)
+{
+	int num;
 
-		c->cpu_core_id = phys_pkg_id(c->initial_apicid, index_msb) &
-					       ((1 << core_bits) - 1);
+	get_option(&arg, &num);
 
-		if (c->x86_max_cores > 1)
-			printk(KERN_INFO  "CPU: Processor Core ID: %d\n",
-			       c->cpu_core_id);
-	}
+	if (num > 0)
+		show_msr = num;
+	return 1;
 }
-#endif
+__setup("show_msr=", setup_show_msr);
 
 static __init int setup_noclflush(char *arg)
 {
@@ -604,18 +816,26 @@ void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
 	else if (c->cpuid_level >= 0)
 		vendor = c->x86_vendor_id;
 
-	if (vendor && strncmp(c->x86_model_id, vendor, strlen(vendor)))
-		printk("%s ", vendor);
+	if (vendor && !strstr(c->x86_model_id, vendor))
+		printk(KERN_CONT "%s ", vendor);
 
-	if (!c->x86_model_id[0])
-		printk("%d86", c->x86);
+	if (c->x86_model_id[0])
+		printk(KERN_CONT "%s", c->x86_model_id);
 	else
-		printk("%s", c->x86_model_id);
+		printk(KERN_CONT "%d86", c->x86);
 
 	if (c->x86_mask || c->cpuid_level >= 0)
-		printk(" stepping %02x\n", c->x86_mask);
+		printk(KERN_CONT " stepping %02x\n", c->x86_mask);
 	else
-		printk("\n");
+		printk(KERN_CONT "\n");
+
+#ifdef CONFIG_SMP
+	if (c->cpu_index < show_msr)
+		print_cpu_msr();
+#else
+	if (show_msr)
+		print_cpu_msr();
+#endif
 }
 
 static __init int setup_disablecpuid(char *arg)
@@ -631,19 +851,89 @@ __setup("clearcpuid=", setup_disablecpuid);
 
 cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
 
-void __init early_cpu_init(void)
+#ifdef CONFIG_X86_64
+struct x8664_pda **_cpu_pda __read_mostly;
+EXPORT_SYMBOL(_cpu_pda);
+
+struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
+
+char boot_cpu_stack[IRQSTACKSIZE] __page_aligned_bss;
+
+void __cpuinit pda_init(int cpu)
 {
-	struct cpu_vendor_dev *cvdev;
+	struct x8664_pda *pda = cpu_pda(cpu);
+
+	/* Setup up data that may be needed in __get_free_pages early */
+	loadsegment(fs, 0);
+	loadsegment(gs, 0);
+	/* Memory clobbers used to order PDA accessed */
+	mb();
+	wrmsrl(MSR_GS_BASE, pda);
+	mb();
+
+	pda->cpunumber = cpu;
+	pda->irqcount = -1;
+	pda->kernelstack = (unsigned long)stack_thread_info() -
+				 PDA_STACKOFFSET + THREAD_SIZE;
+	pda->active_mm = &init_mm;
+	pda->mmu_state = 0;
+
+	if (cpu == 0) {
+		/* others are initialized in smpboot.c */
+		pda->pcurrent = &init_task;
+		pda->irqstackptr = boot_cpu_stack;
+		pda->irqstackptr += IRQSTACKSIZE - 64;
+	} else {
+		if (!pda->irqstackptr) {
+			pda->irqstackptr = (char *)
+				__get_free_pages(GFP_ATOMIC, IRQSTACK_ORDER);
+			if (!pda->irqstackptr)
+				panic("cannot allocate irqstack for cpu %d",
+				      cpu);
+			pda->irqstackptr += IRQSTACKSIZE - 64;
+		}
+
+		if (pda->nodenumber == 0 && cpu_to_node(cpu) != NUMA_NO_NODE)
+			pda->nodenumber = cpu_to_node(cpu);
+	}
+}
+
+char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ +
+			   DEBUG_STKSZ] __page_aligned_bss;
 
-	for (cvdev = __x86cpuvendor_start ;
-	     cvdev < __x86cpuvendor_end   ;
-	     cvdev++)
-		cpu_devs[cvdev->vendor] = cvdev->cpu_dev;
+extern asmlinkage void ignore_sysret(void);
 
-	early_cpu_detect();
-	validate_pat_support(&boot_cpu_data);
+/* May not be marked __init: used by software suspend */
+void syscall_init(void)
+{
+	/*
+	 * LSTAR and STAR live in a bit strange symbiosis.
+	 * They both write to the same internal register. STAR allows to
+	 * set CS/DS but only a 32bit target. LSTAR sets the 64bit rip.
+	 */
+	wrmsrl(MSR_STAR,  ((u64)__USER32_CS)<<48  | ((u64)__KERNEL_CS)<<32);
+	wrmsrl(MSR_LSTAR, system_call);
+	wrmsrl(MSR_CSTAR, ignore_sysret);
+
+#ifdef CONFIG_IA32_EMULATION
+	syscall32_cpu_init();
+#endif
+
+	/* Flags to clear on syscall */
+	wrmsrl(MSR_SYSCALL_MASK,
+	       X86_EFLAGS_TF|X86_EFLAGS_DF|X86_EFLAGS_IF|X86_EFLAGS_IOPL);
 }
 
+unsigned long kernel_eflags;
+
+/*
+ * Copies of the original ist values from the tss are only accessed during
+ * debugging, no special alignment required.
+ */
+DEFINE_PER_CPU(struct orig_ist, orig_ist);
+
+#else
+
 /* Make sure %fs is initialized properly in idle threads */
 struct pt_regs * __cpuinit idle_regs(struct pt_regs *regs)
 {
@@ -651,25 +941,136 @@ struct pt_regs * __cpuinit idle_regs(struct pt_regs *regs)
 	regs->fs = __KERNEL_PERCPU;
 	return regs;
 }
-
-/* Current gdt points %fs at the "master" per-cpu area: after this,
- * it's on the real one. */
-void switch_to_new_gdt(void)
-{
-	struct desc_ptr gdt_descr;
-
-	gdt_descr.address = (long)get_cpu_gdt_table(smp_processor_id());
-	gdt_descr.size = GDT_SIZE - 1;
-	load_gdt(&gdt_descr);
-	asm("mov %0, %%fs" : : "r" (__KERNEL_PERCPU) : "memory");
-}
+#endif
 
 /*
  * cpu_init() initializes state that is per-CPU. Some data is already
  * initialized (naturally) in the bootstrap process, such as the GDT
  * and IDT. We reload them nevertheless, this function acts as a
  * 'CPU state barrier', nothing should get across.
+ * A lot of state is already set up in PDA init for 64 bit
  */
+#ifdef CONFIG_X86_64
+void __cpuinit cpu_init(void)
+{
+	int cpu = stack_smp_processor_id();
+	struct tss_struct *t = &per_cpu(init_tss, cpu);
+	struct orig_ist *orig_ist = &per_cpu(orig_ist, cpu);
+	unsigned long v;
+	char *estacks = NULL;
+	struct task_struct *me;
+	int i;
+
+	/* CPU 0 is initialised in head64.c */
+	if (cpu != 0)
+		pda_init(cpu);
+	else
+		estacks = boot_exception_stacks;
+
+	me = current;
+
+	if (cpu_test_and_set(cpu, cpu_initialized))
+		panic("CPU#%d already initialized!\n", cpu);
+
+	printk(KERN_INFO "Initializing CPU#%d\n", cpu);
+
+	clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
+
+	/*
+	 * Initialize the per-CPU GDT with the boot GDT,
+	 * and set up the GDT descriptor:
+	 */
+
+	switch_to_new_gdt();
+	load_idt((const struct desc_ptr *)&idt_descr);
+
+	memset(me->thread.tls_array, 0, GDT_ENTRY_TLS_ENTRIES * 8);
+	syscall_init();
+
+	wrmsrl(MSR_FS_BASE, 0);
+	wrmsrl(MSR_KERNEL_GS_BASE, 0);
+	barrier();
+
+	check_efer();
+	if (cpu != 0 && x2apic)
+		enable_x2apic();
+
+	/*
+	 * set up and load the per-CPU TSS
+	 */
+	if (!orig_ist->ist[0]) {
+		static const unsigned int order[N_EXCEPTION_STACKS] = {
+		  [0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER,
+		  [DEBUG_STACK - 1] = DEBUG_STACK_ORDER
+		};
+		for (v = 0; v < N_EXCEPTION_STACKS; v++) {
+			if (cpu) {
+				estacks = (char *)__get_free_pages(GFP_ATOMIC, order[v]);
+				if (!estacks)
+					panic("Cannot allocate exception "
+					      "stack %ld %d\n", v, cpu);
+			}
+			estacks += PAGE_SIZE << order[v];
+			orig_ist->ist[v] = t->x86_tss.ist[v] =
+					(unsigned long)estacks;
+		}
+	}
+
+	t->x86_tss.io_bitmap_base = offsetof(struct tss_struct, io_bitmap);
+	/*
+	 * <= is required because the CPU will access up to
+	 * 8 bits beyond the end of the IO permission bitmap.
+	 */
+	for (i = 0; i <= IO_BITMAP_LONGS; i++)
+		t->io_bitmap[i] = ~0UL;
+
+	atomic_inc(&init_mm.mm_count);
+	me->active_mm = &init_mm;
+	if (me->mm)
+		BUG();
+	enter_lazy_tlb(&init_mm, me);
+
+	load_sp0(t, &current->thread);
+	set_tss_desc(cpu, t);
+	load_TR_desc();
+	load_LDT(&init_mm.context);
+
+#ifdef CONFIG_KGDB
+	/*
+	 * If the kgdb is connected no debug regs should be altered.  This
+	 * is only applicable when KGDB and a KGDB I/O module are built
+	 * into the kernel and you are using early debugging with
+	 * kgdbwait. KGDB will control the kernel HW breakpoint registers.
+	 */
+	if (kgdb_connected && arch_kgdb_ops.correct_hw_break)
+		arch_kgdb_ops.correct_hw_break();
+	else {
+#endif
+	/*
+	 * Clear all 6 debug registers:
+	 */
+
+	set_debugreg(0UL, 0);
+	set_debugreg(0UL, 1);
+	set_debugreg(0UL, 2);
+	set_debugreg(0UL, 3);
+	set_debugreg(0UL, 6);
+	set_debugreg(0UL, 7);
+#ifdef CONFIG_KGDB
+	/* If the kgdb is connected no debug regs should be altered. */
+	}
+#endif
+
+	fpu_init();
+
+	raw_local_save_flags(kernel_eflags);
+
+	if (is_uv_system())
+		uv_cpu_init();
+}
+
+#else
+
 void __cpuinit cpu_init(void)
 {
 	int cpu = smp_processor_id();
@@ -723,19 +1124,21 @@ void __cpuinit cpu_init(void)
 	/*
 	 * Force FPU initialization:
 	 */
-	current_thread_info()->status = 0;
+	if (cpu_has_xsave)
+		current_thread_info()->status = TS_XSAVE;
+	else
+		current_thread_info()->status = 0;
 	clear_used_math();
 	mxcsr_feature_mask_init();
-}
 
-#ifdef CONFIG_HOTPLUG_CPU
-void __cpuinit cpu_uninit(void)
-{
-	int cpu = raw_smp_processor_id();
-	cpu_clear(cpu, cpu_initialized);
+	/*
+	 * Boot processor to setup the FP and extended state context info.
+	 */
+	if (!smp_processor_id())
+		init_thread_xstate();
 
-	/* lazy TLB state */
-	per_cpu(cpu_tlbstate, cpu).state = 0;
-	per_cpu(cpu_tlbstate, cpu).active_mm = &init_mm;
+	xsave_init();
 }
+
+
 #endif
diff --git a/arch/x86/kernel/cpu/common_64.c b/arch/x86/kernel/cpu/common_64.c
deleted file mode 100644
index 305b465889b0..000000000000
--- a/arch/x86/kernel/cpu/common_64.c
+++ /dev/null
@@ -1,763 +0,0 @@
-#include <linux/init.h>
-#include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/string.h>
-#include <linux/bootmem.h>
-#include <linux/bitops.h>
-#include <linux/module.h>
-#include <linux/kgdb.h>
-#include <linux/topology.h>
-#include <linux/delay.h>
-#include <linux/smp.h>
-#include <linux/percpu.h>
-#include <asm/i387.h>
-#include <asm/msr.h>
-#include <asm/io.h>
-#include <asm/linkage.h>
-#include <asm/mmu_context.h>
-#include <asm/mtrr.h>
-#include <asm/mce.h>
-#include <asm/pat.h>
-#include <asm/asm.h>
-#include <asm/numa.h>
-#ifdef CONFIG_X86_LOCAL_APIC
-#include <asm/mpspec.h>
-#include <asm/apic.h>
-#include <mach_apic.h>
-#endif
-#include <asm/pda.h>
-#include <asm/pgtable.h>
-#include <asm/processor.h>
-#include <asm/desc.h>
-#include <asm/atomic.h>
-#include <asm/proto.h>
-#include <asm/sections.h>
-#include <asm/setup.h>
-#include <asm/genapic.h>
-
-#include "cpu.h"
-
-/* We need valid kernel segments for data and code in long mode too
- * IRET will check the segment types  kkeil 2000/10/28
- * Also sysret mandates a special GDT layout
- */
-/* The TLS descriptors are currently at a different place compared to i386.
-   Hopefully nobody expects them at a fixed place (Wine?) */
-DEFINE_PER_CPU(struct gdt_page, gdt_page) = { .gdt = {
-	[GDT_ENTRY_KERNEL32_CS] = { { { 0x0000ffff, 0x00cf9b00 } } },
-	[GDT_ENTRY_KERNEL_CS] = { { { 0x0000ffff, 0x00af9b00 } } },
-	[GDT_ENTRY_KERNEL_DS] = { { { 0x0000ffff, 0x00cf9300 } } },
-	[GDT_ENTRY_DEFAULT_USER32_CS] = { { { 0x0000ffff, 0x00cffb00 } } },
-	[GDT_ENTRY_DEFAULT_USER_DS] = { { { 0x0000ffff, 0x00cff300 } } },
-	[GDT_ENTRY_DEFAULT_USER_CS] = { { { 0x0000ffff, 0x00affb00 } } },
-} };
-EXPORT_PER_CPU_SYMBOL_GPL(gdt_page);
-
-__u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata;
-
-/* Current gdt points %fs at the "master" per-cpu area: after this,
- * it's on the real one. */
-void switch_to_new_gdt(void)
-{
-	struct desc_ptr gdt_descr;
-
-	gdt_descr.address = (long)get_cpu_gdt_table(smp_processor_id());
-	gdt_descr.size = GDT_SIZE - 1;
-	load_gdt(&gdt_descr);
-}
-
-struct cpu_dev *cpu_devs[X86_VENDOR_NUM] = {};
-
-static void __cpuinit default_init(struct cpuinfo_x86 *c)
-{
-	display_cacheinfo(c);
-}
-
-static struct cpu_dev __cpuinitdata default_cpu = {
-	.c_init	= default_init,
-	.c_vendor = "Unknown",
-};
-static struct cpu_dev *this_cpu __cpuinitdata = &default_cpu;
-
-int __cpuinit get_model_name(struct cpuinfo_x86 *c)
-{
-	unsigned int *v;
-
-	if (c->extended_cpuid_level < 0x80000004)
-		return 0;
-
-	v = (unsigned int *) c->x86_model_id;
-	cpuid(0x80000002, &v[0], &v[1], &v[2], &v[3]);
-	cpuid(0x80000003, &v[4], &v[5], &v[6], &v[7]);
-	cpuid(0x80000004, &v[8], &v[9], &v[10], &v[11]);
-	c->x86_model_id[48] = 0;
-	return 1;
-}
-
-
-void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
-{
-	unsigned int n, dummy, ebx, ecx, edx;
-
-	n = c->extended_cpuid_level;
-
-	if (n >= 0x80000005) {
-		cpuid(0x80000005, &dummy, &ebx, &ecx, &edx);
-		printk(KERN_INFO "CPU: L1 I Cache: %dK (%d bytes/line), "
-		       "D cache %dK (%d bytes/line)\n",
-		       edx>>24, edx&0xFF, ecx>>24, ecx&0xFF);
-		c->x86_cache_size = (ecx>>24) + (edx>>24);
-		/* On K8 L1 TLB is inclusive, so don't count it */
-		c->x86_tlbsize = 0;
-	}
-
-	if (n >= 0x80000006) {
-		cpuid(0x80000006, &dummy, &ebx, &ecx, &edx);
-		ecx = cpuid_ecx(0x80000006);
-		c->x86_cache_size = ecx >> 16;
-		c->x86_tlbsize += ((ebx >> 16) & 0xfff) + (ebx & 0xfff);
-
-		printk(KERN_INFO "CPU: L2 Cache: %dK (%d bytes/line)\n",
-		c->x86_cache_size, ecx & 0xFF);
-	}
-}
-
-void __cpuinit detect_ht(struct cpuinfo_x86 *c)
-{
-#ifdef CONFIG_SMP
-	u32 eax, ebx, ecx, edx;
-	int index_msb, core_bits;
-
-	cpuid(1, &eax, &ebx, &ecx, &edx);
-
-
-	if (!cpu_has(c, X86_FEATURE_HT))
-		return;
-	if (cpu_has(c, X86_FEATURE_CMP_LEGACY))
-		goto out;
-
-	smp_num_siblings = (ebx & 0xff0000) >> 16;
-
-	if (smp_num_siblings == 1) {
-		printk(KERN_INFO  "CPU: Hyper-Threading is disabled\n");
-	} else if (smp_num_siblings > 1) {
-
-		if (smp_num_siblings > NR_CPUS) {
-			printk(KERN_WARNING "CPU: Unsupported number of "
-			       "siblings %d", smp_num_siblings);
-			smp_num_siblings = 1;
-			return;
-		}
-
-		index_msb = get_count_order(smp_num_siblings);
-		c->phys_proc_id = phys_pkg_id(index_msb);
-
-		smp_num_siblings = smp_num_siblings / c->x86_max_cores;
-
-		index_msb = get_count_order(smp_num_siblings);
-
-		core_bits = get_count_order(c->x86_max_cores);
-
-		c->cpu_core_id = phys_pkg_id(index_msb) &
-					       ((1 << core_bits) - 1);
-	}
-out:
-	if ((c->x86_max_cores * smp_num_siblings) > 1) {
-		printk(KERN_INFO  "CPU: Physical Processor ID: %d\n",
-		       c->phys_proc_id);
-		printk(KERN_INFO  "CPU: Processor Core ID: %d\n",
-		       c->cpu_core_id);
-	}
-
-#endif
-}
-
-static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c)
-{
-	char *v = c->x86_vendor_id;
-	int i;
-	static int printed;
-
-	for (i = 0; i < X86_VENDOR_NUM; i++) {
-		if (cpu_devs[i]) {
-			if (!strcmp(v, cpu_devs[i]->c_ident[0]) ||
-			    (cpu_devs[i]->c_ident[1] &&
-			    !strcmp(v, cpu_devs[i]->c_ident[1]))) {
-				c->x86_vendor = i;
-				this_cpu = cpu_devs[i];
-				return;
-			}
-		}
-	}
-	if (!printed) {
-		printed++;
-		printk(KERN_ERR "CPU: Vendor unknown, using generic init.\n");
-		printk(KERN_ERR "CPU: Your system may be unstable.\n");
-	}
-	c->x86_vendor = X86_VENDOR_UNKNOWN;
-}
-
-static void __init early_cpu_support_print(void)
-{
-	int i,j;
-	struct cpu_dev *cpu_devx;
-
-	printk("KERNEL supported cpus:\n");
-	for (i = 0; i < X86_VENDOR_NUM; i++) {
-		cpu_devx = cpu_devs[i];
-		if (!cpu_devx)
-			continue;
-		for (j = 0; j < 2; j++) {
-			if (!cpu_devx->c_ident[j])
-				continue;
-			printk("  %s %s\n", cpu_devx->c_vendor,
-				cpu_devx->c_ident[j]);
-		}
-	}
-}
-
-/*
- * The NOPL instruction is supposed to exist on all CPUs with
- * family >= 6, unfortunately, that's not true in practice because
- * of early VIA chips and (more importantly) broken virtualizers that
- * are not easy to detect.  Hence, probe for it based on first
- * principles.
- *
- * Note: no 64-bit chip is known to lack these, but put the code here
- * for consistency with 32 bits, and to make it utterly trivial to
- * diagnose the problem should it ever surface.
- */
-static void __cpuinit detect_nopl(struct cpuinfo_x86 *c)
-{
-	const u32 nopl_signature = 0x888c53b1; /* Random number */
-	u32 has_nopl = nopl_signature;
-
-	clear_cpu_cap(c, X86_FEATURE_NOPL);
-	if (c->x86 >= 6) {
-		asm volatile("\n"
-			     "1:      .byte 0x0f,0x1f,0xc0\n" /* nopl %eax */
-			     "2:\n"
-			     "        .section .fixup,\"ax\"\n"
-			     "3:      xor %0,%0\n"
-			     "        jmp 2b\n"
-			     "        .previous\n"
-			     _ASM_EXTABLE(1b,3b)
-			     : "+a" (has_nopl));
-
-		if (has_nopl == nopl_signature)
-			set_cpu_cap(c, X86_FEATURE_NOPL);
-	}
-}
-
-static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c);
-
-void __init early_cpu_init(void)
-{
-        struct cpu_vendor_dev *cvdev;
-
-        for (cvdev = __x86cpuvendor_start ;
-             cvdev < __x86cpuvendor_end   ;
-             cvdev++)
-                cpu_devs[cvdev->vendor] = cvdev->cpu_dev;
-	early_cpu_support_print();
-	early_identify_cpu(&boot_cpu_data);
-}
-
-/* Do some early cpuid on the boot CPU to get some parameter that are
-   needed before check_bugs. Everything advanced is in identify_cpu
-   below. */
-static void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
-{
-	u32 tfms, xlvl;
-
-	c->loops_per_jiffy = loops_per_jiffy;
-	c->x86_cache_size = -1;
-	c->x86_vendor = X86_VENDOR_UNKNOWN;
-	c->x86_model = c->x86_mask = 0;	/* So far unknown... */
-	c->x86_vendor_id[0] = '\0'; /* Unset */
-	c->x86_model_id[0] = '\0';  /* Unset */
-	c->x86_clflush_size = 64;
-	c->x86_cache_alignment = c->x86_clflush_size;
-	c->x86_max_cores = 1;
-	c->x86_coreid_bits = 0;
-	c->extended_cpuid_level = 0;
-	memset(&c->x86_capability, 0, sizeof c->x86_capability);
-
-	/* Get vendor name */
-	cpuid(0x00000000, (unsigned int *)&c->cpuid_level,
-	      (unsigned int *)&c->x86_vendor_id[0],
-	      (unsigned int *)&c->x86_vendor_id[8],
-	      (unsigned int *)&c->x86_vendor_id[4]);
-
-	get_cpu_vendor(c);
-
-	/* Initialize the standard set of capabilities */
-	/* Note that the vendor-specific code below might override */
-
-	/* Intel-defined flags: level 0x00000001 */
-	if (c->cpuid_level >= 0x00000001) {
-		__u32 misc;
-		cpuid(0x00000001, &tfms, &misc, &c->x86_capability[4],
-		      &c->x86_capability[0]);
-		c->x86 = (tfms >> 8) & 0xf;
-		c->x86_model = (tfms >> 4) & 0xf;
-		c->x86_mask = tfms & 0xf;
-		if (c->x86 == 0xf)
-			c->x86 += (tfms >> 20) & 0xff;
-		if (c->x86 >= 0x6)
-			c->x86_model += ((tfms >> 16) & 0xF) << 4;
-		if (test_cpu_cap(c, X86_FEATURE_CLFLSH))
-			c->x86_clflush_size = ((misc >> 8) & 0xff) * 8;
-	} else {
-		/* Have CPUID level 0 only - unheard of */
-		c->x86 = 4;
-	}
-
-	c->initial_apicid = (cpuid_ebx(1) >> 24) & 0xff;
-#ifdef CONFIG_SMP
-	c->phys_proc_id = c->initial_apicid;
-#endif
-	/* AMD-defined flags: level 0x80000001 */
-	xlvl = cpuid_eax(0x80000000);
-	c->extended_cpuid_level = xlvl;
-	if ((xlvl & 0xffff0000) == 0x80000000) {
-		if (xlvl >= 0x80000001) {
-			c->x86_capability[1] = cpuid_edx(0x80000001);
-			c->x86_capability[6] = cpuid_ecx(0x80000001);
-		}
-		if (xlvl >= 0x80000004)
-			get_model_name(c); /* Default name */
-	}
-
-	/* Transmeta-defined flags: level 0x80860001 */
-	xlvl = cpuid_eax(0x80860000);
-	if ((xlvl & 0xffff0000) == 0x80860000) {
-		/* Don't set x86_cpuid_level here for now to not confuse. */
-		if (xlvl >= 0x80860001)
-			c->x86_capability[2] = cpuid_edx(0x80860001);
-	}
-
-	if (c->extended_cpuid_level >= 0x80000007)
-		c->x86_power = cpuid_edx(0x80000007);
-
-	if (c->extended_cpuid_level >= 0x80000008) {
-		u32 eax = cpuid_eax(0x80000008);
-
-		c->x86_virt_bits = (eax >> 8) & 0xff;
-		c->x86_phys_bits = eax & 0xff;
-	}
-
-	detect_nopl(c);
-
-	if (c->x86_vendor != X86_VENDOR_UNKNOWN &&
-	    cpu_devs[c->x86_vendor]->c_early_init)
-		cpu_devs[c->x86_vendor]->c_early_init(c);
-
-	validate_pat_support(c);
-}
-
-/*
- * This does the hard work of actually picking apart the CPU stuff...
- */
-static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
-{
-	int i;
-
-	early_identify_cpu(c);
-
-	init_scattered_cpuid_features(c);
-
-	c->apicid = phys_pkg_id(0);
-
-	/*
-	 * Vendor-specific initialization.  In this section we
-	 * canonicalize the feature flags, meaning if there are
-	 * features a certain CPU supports which CPUID doesn't
-	 * tell us, CPUID claiming incorrect flags, or other bugs,
-	 * we handle them here.
-	 *
-	 * At the end of this section, c->x86_capability better
-	 * indicate the features this CPU genuinely supports!
-	 */
-	if (this_cpu->c_init)
-		this_cpu->c_init(c);
-
-	detect_ht(c);
-
-	/*
-	 * On SMP, boot_cpu_data holds the common feature set between
-	 * all CPUs; so make sure that we indicate which features are
-	 * common between the CPUs.  The first time this routine gets
-	 * executed, c == &boot_cpu_data.
-	 */
-	if (c != &boot_cpu_data) {
-		/* AND the already accumulated flags with these */
-		for (i = 0; i < NCAPINTS; i++)
-			boot_cpu_data.x86_capability[i] &= c->x86_capability[i];
-	}
-
-	/* Clear all flags overriden by options */
-	for (i = 0; i < NCAPINTS; i++)
-		c->x86_capability[i] &= ~cleared_cpu_caps[i];
-
-#ifdef CONFIG_X86_MCE
-	mcheck_init(c);
-#endif
-	select_idle_routine(c);
-
-#ifdef CONFIG_NUMA
-	numa_add_cpu(smp_processor_id());
-#endif
-
-}
-
-void __cpuinit identify_boot_cpu(void)
-{
-	identify_cpu(&boot_cpu_data);
-}
-
-void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c)
-{
-	BUG_ON(c == &boot_cpu_data);
-	identify_cpu(c);
-	mtrr_ap_init();
-}
-
-static __init int setup_noclflush(char *arg)
-{
-	setup_clear_cpu_cap(X86_FEATURE_CLFLSH);
-	return 1;
-}
-__setup("noclflush", setup_noclflush);
-
-struct msr_range {
-	unsigned min;
-	unsigned max;
-};
-
-static struct msr_range msr_range_array[] __cpuinitdata = {
-	{ 0x00000000, 0x00000418},
-	{ 0xc0000000, 0xc000040b},
-	{ 0xc0010000, 0xc0010142},
-	{ 0xc0011000, 0xc001103b},
-};
-
-static void __cpuinit print_cpu_msr(void)
-{
-	unsigned index;
-	u64 val;
-	int i;
-	unsigned index_min, index_max;
-
-	for (i = 0; i < ARRAY_SIZE(msr_range_array); i++) {
-		index_min = msr_range_array[i].min;
-		index_max = msr_range_array[i].max;
-		for (index = index_min; index < index_max; index++) {
-			if (rdmsrl_amd_safe(index, &val))
-				continue;
-			printk(KERN_INFO " MSR%08x: %016llx\n", index, val);
-		}
-	}
-}
-
-static int show_msr __cpuinitdata;
-static __init int setup_show_msr(char *arg)
-{
-	int num;
-
-	get_option(&arg, &num);
-
-	if (num > 0)
-		show_msr = num;
-	return 1;
-}
-__setup("show_msr=", setup_show_msr);
-
-void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
-{
-	if (c->x86_model_id[0])
-		printk(KERN_CONT "%s", c->x86_model_id);
-
-	if (c->x86_mask || c->cpuid_level >= 0)
-		printk(KERN_CONT " stepping %02x\n", c->x86_mask);
-	else
-		printk(KERN_CONT "\n");
-
-#ifdef CONFIG_SMP
-	if (c->cpu_index < show_msr)
-		print_cpu_msr();
-#else
-	if (show_msr)
-		print_cpu_msr();
-#endif
-}
-
-static __init int setup_disablecpuid(char *arg)
-{
-	int bit;
-	if (get_option(&arg, &bit) && bit < NCAPINTS*32)
-		setup_clear_cpu_cap(bit);
-	else
-		return 0;
-	return 1;
-}
-__setup("clearcpuid=", setup_disablecpuid);
-
-cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
-
-struct x8664_pda **_cpu_pda __read_mostly;
-EXPORT_SYMBOL(_cpu_pda);
-
-struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
-
-char boot_cpu_stack[IRQSTACKSIZE] __page_aligned_bss;
-
-unsigned long __supported_pte_mask __read_mostly = ~0UL;
-EXPORT_SYMBOL_GPL(__supported_pte_mask);
-
-static int do_not_nx __cpuinitdata;
-
-/* noexec=on|off
-Control non executable mappings for 64bit processes.
-
-on	Enable(default)
-off	Disable
-*/
-static int __init nonx_setup(char *str)
-{
-	if (!str)
-		return -EINVAL;
-	if (!strncmp(str, "on", 2)) {
-		__supported_pte_mask |= _PAGE_NX;
-		do_not_nx = 0;
-	} else if (!strncmp(str, "off", 3)) {
-		do_not_nx = 1;
-		__supported_pte_mask &= ~_PAGE_NX;
-	}
-	return 0;
-}
-early_param("noexec", nonx_setup);
-
-int force_personality32;
-
-/* noexec32=on|off
-Control non executable heap for 32bit processes.
-To control the stack too use noexec=off
-
-on	PROT_READ does not imply PROT_EXEC for 32bit processes (default)
-off	PROT_READ implies PROT_EXEC
-*/
-static int __init nonx32_setup(char *str)
-{
-	if (!strcmp(str, "on"))
-		force_personality32 &= ~READ_IMPLIES_EXEC;
-	else if (!strcmp(str, "off"))
-		force_personality32 |= READ_IMPLIES_EXEC;
-	return 1;
-}
-__setup("noexec32=", nonx32_setup);
-
-void pda_init(int cpu)
-{
-	struct x8664_pda *pda = cpu_pda(cpu);
-
-	/* Setup up data that may be needed in __get_free_pages early */
-	loadsegment(fs, 0);
-	loadsegment(gs, 0);
-	/* Memory clobbers used to order PDA accessed */
-	mb();
-	wrmsrl(MSR_GS_BASE, pda);
-	mb();
-
-	pda->cpunumber = cpu;
-	pda->irqcount = -1;
-	pda->kernelstack = (unsigned long)stack_thread_info() -
-				 PDA_STACKOFFSET + THREAD_SIZE;
-	pda->active_mm = &init_mm;
-	pda->mmu_state = 0;
-
-	if (cpu == 0) {
-		/* others are initialized in smpboot.c */
-		pda->pcurrent = &init_task;
-		pda->irqstackptr = boot_cpu_stack;
-		pda->irqstackptr += IRQSTACKSIZE - 64;
-	} else {
-		if (!pda->irqstackptr) {
-			pda->irqstackptr = (char *)
-				__get_free_pages(GFP_ATOMIC, IRQSTACK_ORDER);
-			if (!pda->irqstackptr)
-				panic("cannot allocate irqstack for cpu %d",
-				      cpu);
-			pda->irqstackptr += IRQSTACKSIZE - 64;
-		}
-
-		if (pda->nodenumber == 0 && cpu_to_node(cpu) != NUMA_NO_NODE)
-			pda->nodenumber = cpu_to_node(cpu);
-	}
-}
-
-char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ +
-			   DEBUG_STKSZ] __page_aligned_bss;
-
-extern asmlinkage void ignore_sysret(void);
-
-/* May not be marked __init: used by software suspend */
-void syscall_init(void)
-{
-	/*
-	 * LSTAR and STAR live in a bit strange symbiosis.
-	 * They both write to the same internal register. STAR allows to
-	 * set CS/DS but only a 32bit target. LSTAR sets the 64bit rip.
-	 */
-	wrmsrl(MSR_STAR,  ((u64)__USER32_CS)<<48  | ((u64)__KERNEL_CS)<<32);
-	wrmsrl(MSR_LSTAR, system_call);
-	wrmsrl(MSR_CSTAR, ignore_sysret);
-
-#ifdef CONFIG_IA32_EMULATION
-	syscall32_cpu_init();
-#endif
-
-	/* Flags to clear on syscall */
-	wrmsrl(MSR_SYSCALL_MASK,
-	       X86_EFLAGS_TF|X86_EFLAGS_DF|X86_EFLAGS_IF|X86_EFLAGS_IOPL);
-}
-
-void __cpuinit check_efer(void)
-{
-	unsigned long efer;
-
-	rdmsrl(MSR_EFER, efer);
-	if (!(efer & EFER_NX) || do_not_nx)
-		__supported_pte_mask &= ~_PAGE_NX;
-}
-
-unsigned long kernel_eflags;
-
-/*
- * Copies of the original ist values from the tss are only accessed during
- * debugging, no special alignment required.
- */
-DEFINE_PER_CPU(struct orig_ist, orig_ist);
-
-/*
- * cpu_init() initializes state that is per-CPU. Some data is already
- * initialized (naturally) in the bootstrap process, such as the GDT
- * and IDT. We reload them nevertheless, this function acts as a
- * 'CPU state barrier', nothing should get across.
- * A lot of state is already set up in PDA init.
- */
-void __cpuinit cpu_init(void)
-{
-	int cpu = stack_smp_processor_id();
-	struct tss_struct *t = &per_cpu(init_tss, cpu);
-	struct orig_ist *orig_ist = &per_cpu(orig_ist, cpu);
-	unsigned long v;
-	char *estacks = NULL;
-	struct task_struct *me;
-	int i;
-
-	/* CPU 0 is initialised in head64.c */
-	if (cpu != 0)
-		pda_init(cpu);
-	else
-		estacks = boot_exception_stacks;
-
-	me = current;
-
-	if (cpu_test_and_set(cpu, cpu_initialized))
-		panic("CPU#%d already initialized!\n", cpu);
-
-	printk(KERN_INFO "Initializing CPU#%d\n", cpu);
-
-	clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
-
-	/*
-	 * Initialize the per-CPU GDT with the boot GDT,
-	 * and set up the GDT descriptor:
-	 */
-
-	switch_to_new_gdt();
-	load_idt((const struct desc_ptr *)&idt_descr);
-
-	memset(me->thread.tls_array, 0, GDT_ENTRY_TLS_ENTRIES * 8);
-	syscall_init();
-
-	wrmsrl(MSR_FS_BASE, 0);
-	wrmsrl(MSR_KERNEL_GS_BASE, 0);
-	barrier();
-
-	check_efer();
-
-	/*
-	 * set up and load the per-CPU TSS
-	 */
-	if (!orig_ist->ist[0]) {
-		static const unsigned int order[N_EXCEPTION_STACKS] = {
-		  [0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER,
-		  [DEBUG_STACK - 1] = DEBUG_STACK_ORDER
-		};
-		for (v = 0; v < N_EXCEPTION_STACKS; v++) {
-			if (cpu) {
-				estacks = (char *)__get_free_pages(GFP_ATOMIC, order[v]);
-				if (!estacks)
-					panic("Cannot allocate exception "
-					      "stack %ld %d\n", v, cpu);
-			}
-			estacks += PAGE_SIZE << order[v];
-			orig_ist->ist[v] = t->x86_tss.ist[v] =
-					(unsigned long)estacks;
-		}
-	}
-
-	t->x86_tss.io_bitmap_base = offsetof(struct tss_struct, io_bitmap);
-	/*
-	 * <= is required because the CPU will access up to
-	 * 8 bits beyond the end of the IO permission bitmap.
-	 */
-	for (i = 0; i <= IO_BITMAP_LONGS; i++)
-		t->io_bitmap[i] = ~0UL;
-
-	atomic_inc(&init_mm.mm_count);
-	me->active_mm = &init_mm;
-	if (me->mm)
-		BUG();
-	enter_lazy_tlb(&init_mm, me);
-
-	load_sp0(t, &current->thread);
-	set_tss_desc(cpu, t);
-	load_TR_desc();
-	load_LDT(&init_mm.context);
-
-#ifdef CONFIG_KGDB
-	/*
-	 * If the kgdb is connected no debug regs should be altered.  This
-	 * is only applicable when KGDB and a KGDB I/O module are built
-	 * into the kernel and you are using early debugging with
-	 * kgdbwait. KGDB will control the kernel HW breakpoint registers.
-	 */
-	if (kgdb_connected && arch_kgdb_ops.correct_hw_break)
-		arch_kgdb_ops.correct_hw_break();
-	else {
-#endif
-	/*
-	 * Clear all 6 debug registers:
-	 */
-
-	set_debugreg(0UL, 0);
-	set_debugreg(0UL, 1);
-	set_debugreg(0UL, 2);
-	set_debugreg(0UL, 3);
-	set_debugreg(0UL, 6);
-	set_debugreg(0UL, 7);
-#ifdef CONFIG_KGDB
-	/* If the kgdb is connected no debug regs should be altered. */
-	}
-#endif
-
-	fpu_init();
-
-	raw_local_save_flags(kernel_eflags);
-
-	if (is_uv_system())
-		uv_cpu_init();
-}
diff --git a/arch/x86/kernel/cpu/cpu.h b/arch/x86/kernel/cpu/cpu.h
index 4d894e8565fe..de4094a39210 100644
--- a/arch/x86/kernel/cpu/cpu.h
+++ b/arch/x86/kernel/cpu/cpu.h
@@ -21,23 +21,16 @@ struct cpu_dev {
 	void		(*c_init)(struct cpuinfo_x86 * c);
 	void		(*c_identify)(struct cpuinfo_x86 * c);
 	unsigned int	(*c_size_cache)(struct cpuinfo_x86 * c, unsigned int size);
+	int	c_x86_vendor;
 };
 
-extern struct cpu_dev * cpu_devs [X86_VENDOR_NUM];
+#define cpu_dev_register(cpu_devX) \
+	static struct cpu_dev *__cpu_dev_##cpu_devX __used \
+	__attribute__((__section__(".x86_cpu_dev.init"))) = \
+	&cpu_devX;
 
-struct cpu_vendor_dev {
-	int vendor;
-	struct cpu_dev *cpu_dev;
-};
-
-#define cpu_vendor_dev_register(cpu_vendor_id, cpu_dev) \
-	static struct cpu_vendor_dev __cpu_vendor_dev_##cpu_vendor_id __used \
-	__attribute__((__section__(".x86cpuvendor.init"))) = \
-	{ cpu_vendor_id, cpu_dev }
-
-extern struct cpu_vendor_dev __x86cpuvendor_start[], __x86cpuvendor_end[];
+extern struct cpu_dev *__x86_cpu_dev_start[], *__x86_cpu_dev_end[];
 
-extern int get_model_name(struct cpuinfo_x86 *c);
 extern void display_cacheinfo(struct cpuinfo_x86 *c);
 
 #endif
diff --git a/arch/x86/kernel/cpu/cyrix.c b/arch/x86/kernel/cpu/cyrix.c
index 898a5a2002ed..ffd0f5ed071a 100644
--- a/arch/x86/kernel/cpu/cyrix.c
+++ b/arch/x86/kernel/cpu/cyrix.c
@@ -121,7 +121,7 @@ static void __cpuinit set_cx86_reorder(void)
 	setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10); /* enable MAPEN */
 
 	/* Load/Store Serialize to mem access disable (=reorder it) */
-	setCx86(CX86_PCR0, getCx86(CX86_PCR0) & ~0x80);
+	setCx86_old(CX86_PCR0, getCx86_old(CX86_PCR0) & ~0x80);
 	/* set load/store serialize from 1GB to 4GB */
 	ccr3 |= 0xe0;
 	setCx86(CX86_CCR3, ccr3);
@@ -132,11 +132,11 @@ static void __cpuinit set_cx86_memwb(void)
 	printk(KERN_INFO "Enable Memory-Write-back mode on Cyrix/NSC processor.\n");
 
 	/* CCR2 bit 2: unlock NW bit */
-	setCx86(CX86_CCR2, getCx86(CX86_CCR2) & ~0x04);
+	setCx86_old(CX86_CCR2, getCx86_old(CX86_CCR2) & ~0x04);
 	/* set 'Not Write-through' */
 	write_cr0(read_cr0() | X86_CR0_NW);
 	/* CCR2 bit 2: lock NW bit and set WT1 */
-	setCx86(CX86_CCR2, getCx86(CX86_CCR2) | 0x14);
+	setCx86_old(CX86_CCR2, getCx86_old(CX86_CCR2) | 0x14);
 }
 
 /*
@@ -150,14 +150,14 @@ static void __cpuinit geode_configure(void)
 	local_irq_save(flags);
 
 	/* Suspend on halt power saving and enable #SUSP pin */
-	setCx86(CX86_CCR2, getCx86(CX86_CCR2) | 0x88);
+	setCx86_old(CX86_CCR2, getCx86_old(CX86_CCR2) | 0x88);
 
 	ccr3 = getCx86(CX86_CCR3);
 	setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10);	/* enable MAPEN */
 
 
 	/* FPU fast, DTE cache, Mem bypass */
-	setCx86(CX86_CCR4, getCx86(CX86_CCR4) | 0x38);
+	setCx86_old(CX86_CCR4, getCx86_old(CX86_CCR4) | 0x38);
 	setCx86(CX86_CCR3, ccr3);			/* disable MAPEN */
 
 	set_cx86_memwb();
@@ -291,7 +291,7 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c)
 		/* GXm supports extended cpuid levels 'ala' AMD */
 		if (c->cpuid_level == 2) {
 			/* Enable cxMMX extensions (GX1 Datasheet 54) */
-			setCx86(CX86_CCR7, getCx86(CX86_CCR7) | 1);
+			setCx86_old(CX86_CCR7, getCx86_old(CX86_CCR7) | 1);
 
 			/*
 			 * GXm : 0x30 ... 0x5f GXm  datasheet 51
@@ -301,7 +301,6 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c)
 			 */
 			if ((0x30 <= dir1 && dir1 <= 0x6f) || (0x80 <= dir1 && dir1 <= 0x8f))
 				geode_configure();
-			get_model_name(c);  /* get CPU marketing name */
 			return;
 		} else { /* MediaGX */
 			Cx86_cb[2] = (dir0_lsn & 1) ? '3' : '4';
@@ -314,7 +313,7 @@ static void __cpuinit init_cyrix(struct cpuinfo_x86 *c)
 		if (dir1 > 7) {
 			dir0_msn++;  /* M II */
 			/* Enable MMX extensions (App note 108) */
-			setCx86(CX86_CCR7, getCx86(CX86_CCR7)|1);
+			setCx86_old(CX86_CCR7, getCx86_old(CX86_CCR7)|1);
 		} else {
 			c->coma_bug = 1;      /* 6x86MX, it has the bug. */
 		}
@@ -429,7 +428,7 @@ static void __cpuinit cyrix_identify(struct cpuinfo_x86 *c)
 			local_irq_save(flags);
 			ccr3 = getCx86(CX86_CCR3);
 			setCx86(CX86_CCR3, (ccr3 & 0x0f) | 0x10);       /* enable MAPEN  */
-			setCx86(CX86_CCR4, getCx86(CX86_CCR4) | 0x80);  /* enable cpuid  */
+			setCx86_old(CX86_CCR4, getCx86_old(CX86_CCR4) | 0x80);  /* enable cpuid  */
 			setCx86(CX86_CCR3, ccr3);                       /* disable MAPEN */
 			local_irq_restore(flags);
 		}
@@ -442,14 +441,16 @@ static struct cpu_dev cyrix_cpu_dev __cpuinitdata = {
 	.c_early_init	= early_init_cyrix,
 	.c_init		= init_cyrix,
 	.c_identify	= cyrix_identify,
+	.c_x86_vendor	= X86_VENDOR_CYRIX,
 };
 
-cpu_vendor_dev_register(X86_VENDOR_CYRIX, &cyrix_cpu_dev);
+cpu_dev_register(cyrix_cpu_dev);
 
 static struct cpu_dev nsc_cpu_dev __cpuinitdata = {
 	.c_vendor	= "NSC",
 	.c_ident	= { "Geode by NSC" },
 	.c_init		= init_nsc,
+	.c_x86_vendor	= X86_VENDOR_NSC,
 };
 
-cpu_vendor_dev_register(X86_VENDOR_NSC, &nsc_cpu_dev);
+cpu_dev_register(nsc_cpu_dev);
diff --git a/arch/x86/kernel/cpu/feature_names.c b/arch/x86/kernel/cpu/feature_names.c
deleted file mode 100644
index c9017799497c..000000000000
--- a/arch/x86/kernel/cpu/feature_names.c
+++ /dev/null
@@ -1,84 +0,0 @@
-/*
- * Strings for the various x86 capability flags.
- *
- * This file must not contain any executable code.
- */
-
-#include <asm/cpufeature.h>
-
-/*
- * These flag bits must match the definitions in <asm/cpufeature.h>.
- * NULL means this bit is undefined or reserved; either way it doesn't
- * have meaning as far as Linux is concerned.  Note that it's important
- * to realize there is a difference between this table and CPUID -- if
- * applications want to get the raw CPUID data, they should access
- * /dev/cpu/<cpu_nr>/cpuid instead.
- */
-const char * const x86_cap_flags[NCAPINTS*32] = {
-	/* Intel-defined */
-	"fpu", "vme", "de", "pse", "tsc", "msr", "pae", "mce",
-	"cx8", "apic", NULL, "sep", "mtrr", "pge", "mca", "cmov",
-	"pat", "pse36", "pn", "clflush", NULL, "dts", "acpi", "mmx",
-	"fxsr", "sse", "sse2", "ss", "ht", "tm", "ia64", "pbe",
-
-	/* AMD-defined */
-	NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
-	NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL,
-	NULL, NULL, NULL, "mp", "nx", NULL, "mmxext", NULL,
-	NULL, "fxsr_opt", "pdpe1gb", "rdtscp", NULL, "lm",
-	"3dnowext", "3dnow",
-
-	/* Transmeta-defined */
-	"recovery", "longrun", NULL, "lrti", NULL, NULL, NULL, NULL,
-	NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
-	NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
-	NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
-
-	/* Other (Linux-defined) */
-	"cxmmx", "k6_mtrr", "cyrix_arr", "centaur_mcr",
-	NULL, NULL, NULL, NULL,
-	"constant_tsc", "up", NULL, "arch_perfmon",
-	"pebs", "bts", NULL, NULL,
-	"rep_good", NULL, NULL, NULL,
-	"nopl", NULL, NULL, NULL,
-	NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
-
-	/* Intel-defined (#2) */
-	"pni", NULL, NULL, "monitor", "ds_cpl", "vmx", "smx", "est",
-	"tm2", "ssse3", "cid", NULL, NULL, "cx16", "xtpr", NULL,
-	NULL, NULL, "dca", "sse4_1", "sse4_2", NULL, NULL, "popcnt",
-	NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
-
-	/* VIA/Cyrix/Centaur-defined */
-	NULL, NULL, "rng", "rng_en", NULL, NULL, "ace", "ace_en",
-	"ace2", "ace2_en", "phe", "phe_en", "pmm", "pmm_en", NULL, NULL,
-	NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
-	NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
-
-	/* AMD-defined (#2) */
-	"lahf_lm", "cmp_legacy", "svm", "extapic",
-	"cr8_legacy", "abm", "sse4a", "misalignsse",
-	"3dnowprefetch", "osvw", "ibs", "sse5",
-	"skinit", "wdt", NULL, NULL,
-	NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
-	NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
-
-	/* Auxiliary (Linux-defined) */
-	"ida", NULL, NULL, NULL, NULL, NULL, NULL, NULL,
-	NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
-	NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
-	NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
-};
-
-const char *const x86_power_flags[32] = {
-	"ts",	/* temperature sensor */
-	"fid",  /* frequency id control */
-	"vid",  /* voltage id control */
-	"ttp",  /* thermal trip */
-	"tm",
-	"stc",
-	"100mhzsteps",
-	"hwpstate",
-	"",	/* tsc invariant mapped to constant_tsc */
-		/* nothing */
-};
diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c
index f113ef4595f6..99468dbd08da 100644
--- a/arch/x86/kernel/cpu/intel.c
+++ b/arch/x86/kernel/cpu/intel.c
@@ -15,6 +15,11 @@
 #include <asm/ds.h>
 #include <asm/bugs.h>
 
+#ifdef CONFIG_X86_64
+#include <asm/topology.h>
+#include <asm/numa_64.h>
+#endif
+
 #include "cpu.h"
 
 #ifdef CONFIG_X86_LOCAL_APIC
@@ -23,23 +28,22 @@
 #include <mach_apic.h>
 #endif
 
-#ifdef CONFIG_X86_INTEL_USERCOPY
-/*
- * Alignment at which movsl is preferred for bulk memory copies.
- */
-struct movsl_mask movsl_mask __read_mostly;
-#endif
-
 static void __cpuinit early_init_intel(struct cpuinfo_x86 *c)
 {
-	/* Netburst reports 64 bytes clflush size, but does IO in 128 bytes */
-	if (c->x86 == 15 && c->x86_cache_alignment == 64)
-		c->x86_cache_alignment = 128;
 	if ((c->x86 == 0xf && c->x86_model >= 0x03) ||
 		(c->x86 == 0x6 && c->x86_model >= 0x0e))
 		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
+
+#ifdef CONFIG_X86_64
+	set_cpu_cap(c, X86_FEATURE_SYSENTER32);
+#else
+	/* Netburst reports 64 bytes clflush size, but does IO in 128 bytes */
+	if (c->x86 == 15 && c->x86_cache_alignment == 64)
+		c->x86_cache_alignment = 128;
+#endif
 }
 
+#ifdef CONFIG_X86_32
 /*
  *	Early probe support logic for ppro memory erratum #50
  *
@@ -59,15 +63,54 @@ int __cpuinit ppro_with_ram_bug(void)
 	return 0;
 }
 
+#ifdef CONFIG_X86_F00F_BUG
+static void __cpuinit trap_init_f00f_bug(void)
+{
+	__set_fixmap(FIX_F00F_IDT, __pa(&idt_table), PAGE_KERNEL_RO);
 
-/*
- * P4 Xeon errata 037 workaround.
- * Hardware prefetcher may cause stale data to be loaded into the cache.
- */
-static void __cpuinit Intel_errata_workarounds(struct cpuinfo_x86 *c)
+	/*
+	 * Update the IDT descriptor and reload the IDT so that
+	 * it uses the read-only mapped virtual address.
+	 */
+	idt_descr.address = fix_to_virt(FIX_F00F_IDT);
+	load_idt(&idt_descr);
+}
+#endif
+
+static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c)
 {
 	unsigned long lo, hi;
 
+#ifdef CONFIG_X86_F00F_BUG
+	/*
+	 * All current models of Pentium and Pentium with MMX technology CPUs
+	 * have the F0 0F bug, which lets nonprivileged users lock up the system.
+	 * Note that the workaround only should be initialized once...
+	 */
+	c->f00f_bug = 0;
+	if (!paravirt_enabled() && c->x86 == 5) {
+		static int f00f_workaround_enabled;
+
+		c->f00f_bug = 1;
+		if (!f00f_workaround_enabled) {
+			trap_init_f00f_bug();
+			printk(KERN_NOTICE "Intel Pentium with F0 0F bug - workaround enabled.\n");
+			f00f_workaround_enabled = 1;
+		}
+	}
+#endif
+
+	/*
+	 * SEP CPUID bug: Pentium Pro reports SEP but doesn't have it until
+	 * model 3 mask 3
+	 */
+	if ((c->x86<<8 | c->x86_model<<4 | c->x86_mask) < 0x633)
+		clear_cpu_cap(c, X86_FEATURE_SEP);
+
+	/*
+	 * P4 Xeon errata 037 workaround.
+	 * Hardware prefetcher may cause stale data to be loaded into the cache.
+	 */
 	if ((c->x86 == 15) && (c->x86_model == 1) && (c->x86_mask == 1)) {
 		rdmsr(MSR_IA32_MISC_ENABLE, lo, hi);
 		if ((lo & (1<<9)) == 0) {
@@ -77,13 +120,68 @@ static void __cpuinit Intel_errata_workarounds(struct cpuinfo_x86 *c)
 			wrmsr (MSR_IA32_MISC_ENABLE, lo, hi);
 		}
 	}
+
+	/*
+	 * See if we have a good local APIC by checking for buggy Pentia,
+	 * i.e. all B steppings and the C2 stepping of P54C when using their
+	 * integrated APIC (see 11AP erratum in "Pentium Processor
+	 * Specification Update").
+	 */
+	if (cpu_has_apic && (c->x86<<8 | c->x86_model<<4) == 0x520 &&
+	    (c->x86_mask < 0x6 || c->x86_mask == 0xb))
+		set_cpu_cap(c, X86_FEATURE_11AP);
+
+
+#ifdef CONFIG_X86_INTEL_USERCOPY
+	/*
+	 * Set up the preferred alignment for movsl bulk memory moves
+	 */
+	switch (c->x86) {
+	case 4:		/* 486: untested */
+		break;
+	case 5:		/* Old Pentia: untested */
+		break;
+	case 6:		/* PII/PIII only like movsl with 8-byte alignment */
+		movsl_mask.mask = 7;
+		break;
+	case 15:	/* P4 is OK down to 8-byte alignment */
+		movsl_mask.mask = 7;
+		break;
+	}
+#endif
+
+#ifdef CONFIG_X86_NUMAQ
+	numaq_tsc_disable();
+#endif
 }
+#else
+static void __cpuinit intel_workarounds(struct cpuinfo_x86 *c)
+{
+}
+#endif
 
+static void __cpuinit srat_detect_node(void)
+{
+#if defined(CONFIG_NUMA) && defined(CONFIG_X86_64)
+	unsigned node;
+	int cpu = smp_processor_id();
+	int apicid = hard_smp_processor_id();
+
+	/* Don't do the funky fallback heuristics the AMD version employs
+	   for now. */
+	node = apicid_to_node[apicid];
+	if (node == NUMA_NO_NODE || !node_online(node))
+		node = first_node(node_online_map);
+	numa_set_node(cpu, node);
+
+	printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node);
+#endif
+}
 
 /*
  * find out the number of processor cores on the die
  */
-static int __cpuinit num_cpu_cores(struct cpuinfo_x86 *c)
+static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c)
 {
 	unsigned int eax, ebx, ecx, edx;
 
@@ -98,45 +196,51 @@ static int __cpuinit num_cpu_cores(struct cpuinfo_x86 *c)
 		return 1;
 }
 
-#ifdef CONFIG_X86_F00F_BUG
-static void __cpuinit trap_init_f00f_bug(void)
+static void __cpuinit detect_vmx_virtcap(struct cpuinfo_x86 *c)
 {
-	__set_fixmap(FIX_F00F_IDT, __pa(&idt_table), PAGE_KERNEL_RO);
-
-	/*
-	 * Update the IDT descriptor and reload the IDT so that
-	 * it uses the read-only mapped virtual address.
-	 */
-	idt_descr.address = fix_to_virt(FIX_F00F_IDT);
-	load_idt(&idt_descr);
+	/* Intel VMX MSR indicated features */
+#define X86_VMX_FEATURE_PROC_CTLS_TPR_SHADOW	0x00200000
+#define X86_VMX_FEATURE_PROC_CTLS_VNMI		0x00400000
+#define X86_VMX_FEATURE_PROC_CTLS_2ND_CTLS	0x80000000
+#define X86_VMX_FEATURE_PROC_CTLS2_VIRT_APIC	0x00000001
+#define X86_VMX_FEATURE_PROC_CTLS2_EPT		0x00000002
+#define X86_VMX_FEATURE_PROC_CTLS2_VPID		0x00000020
+
+	u32 vmx_msr_low, vmx_msr_high, msr_ctl, msr_ctl2;
+
+	clear_cpu_cap(c, X86_FEATURE_TPR_SHADOW);
+	clear_cpu_cap(c, X86_FEATURE_VNMI);
+	clear_cpu_cap(c, X86_FEATURE_FLEXPRIORITY);
+	clear_cpu_cap(c, X86_FEATURE_EPT);
+	clear_cpu_cap(c, X86_FEATURE_VPID);
+
+	rdmsr(MSR_IA32_VMX_PROCBASED_CTLS, vmx_msr_low, vmx_msr_high);
+	msr_ctl = vmx_msr_high | vmx_msr_low;
+	if (msr_ctl & X86_VMX_FEATURE_PROC_CTLS_TPR_SHADOW)
+		set_cpu_cap(c, X86_FEATURE_TPR_SHADOW);
+	if (msr_ctl & X86_VMX_FEATURE_PROC_CTLS_VNMI)
+		set_cpu_cap(c, X86_FEATURE_VNMI);
+	if (msr_ctl & X86_VMX_FEATURE_PROC_CTLS_2ND_CTLS) {
+		rdmsr(MSR_IA32_VMX_PROCBASED_CTLS2,
+		      vmx_msr_low, vmx_msr_high);
+		msr_ctl2 = vmx_msr_high | vmx_msr_low;
+		if ((msr_ctl2 & X86_VMX_FEATURE_PROC_CTLS2_VIRT_APIC) &&
+		    (msr_ctl & X86_VMX_FEATURE_PROC_CTLS_TPR_SHADOW))
+			set_cpu_cap(c, X86_FEATURE_FLEXPRIORITY);
+		if (msr_ctl2 & X86_VMX_FEATURE_PROC_CTLS2_EPT)
+			set_cpu_cap(c, X86_FEATURE_EPT);
+		if (msr_ctl2 & X86_VMX_FEATURE_PROC_CTLS2_VPID)
+			set_cpu_cap(c, X86_FEATURE_VPID);
+	}
 }
-#endif
 
 static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 {
 	unsigned int l2 = 0;
-	char *p = NULL;
 
 	early_init_intel(c);
 
-#ifdef CONFIG_X86_F00F_BUG
-	/*
-	 * All current models of Pentium and Pentium with MMX technology CPUs
-	 * have the F0 0F bug, which lets nonprivileged users lock up the system.
-	 * Note that the workaround only should be initialized once...
-	 */
-	c->f00f_bug = 0;
-	if (!paravirt_enabled() && c->x86 == 5) {
-		static int f00f_workaround_enabled;
-
-		c->f00f_bug = 1;
-		if (!f00f_workaround_enabled) {
-			trap_init_f00f_bug();
-			printk(KERN_NOTICE "Intel Pentium with F0 0F bug - workaround enabled.\n");
-			f00f_workaround_enabled = 1;
-		}
-	}
-#endif
+	intel_workarounds(c);
 
 	l2 = init_intel_cacheinfo(c);
 	if (c->cpuid_level > 9) {
@@ -146,16 +250,32 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 			set_cpu_cap(c, X86_FEATURE_ARCH_PERFMON);
 	}
 
-	/* SEP CPUID bug: Pentium Pro reports SEP but doesn't have it until model 3 mask 3 */
-	if ((c->x86<<8 | c->x86_model<<4 | c->x86_mask) < 0x633)
-		clear_cpu_cap(c, X86_FEATURE_SEP);
+	if (cpu_has_xmm2)
+		set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC);
+	if (cpu_has_ds) {
+		unsigned int l1;
+		rdmsr(MSR_IA32_MISC_ENABLE, l1, l2);
+		if (!(l1 & (1<<11)))
+			set_cpu_cap(c, X86_FEATURE_BTS);
+		if (!(l1 & (1<<12)))
+			set_cpu_cap(c, X86_FEATURE_PEBS);
+		ds_init_intel(c);
+	}
 
+#ifdef CONFIG_X86_64
+	if (c->x86 == 15)
+		c->x86_cache_alignment = c->x86_clflush_size * 2;
+	if (c->x86 == 6)
+		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
+#else
 	/*
 	 * Names for the Pentium II/Celeron processors
 	 * detectable only by also checking the cache size.
 	 * Dixon is NOT a Celeron.
 	 */
 	if (c->x86 == 6) {
+		char *p = NULL;
+
 		switch (c->x86_model) {
 		case 5:
 			if (c->x86_mask == 0) {
@@ -178,71 +298,41 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 				p = "Celeron (Coppermine)";
 			break;
 		}
-	}
-
-	if (p)
-		strcpy(c->x86_model_id, p);
-
-	c->x86_max_cores = num_cpu_cores(c);
-
-	detect_ht(c);
 
-	/* Work around errata */
-	Intel_errata_workarounds(c);
-
-#ifdef CONFIG_X86_INTEL_USERCOPY
-	/*
-	 * Set up the preferred alignment for movsl bulk memory moves
-	 */
-	switch (c->x86) {
-	case 4:		/* 486: untested */
-		break;
-	case 5:		/* Old Pentia: untested */
-		break;
-	case 6:		/* PII/PIII only like movsl with 8-byte alignment */
-		movsl_mask.mask = 7;
-		break;
-	case 15:	/* P4 is OK down to 8-byte alignment */
-		movsl_mask.mask = 7;
-		break;
+		if (p)
+			strcpy(c->x86_model_id, p);
 	}
-#endif
 
-	if (cpu_has_xmm2)
-		set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC);
-	if (c->x86 == 15) {
+	if (c->x86 == 15)
 		set_cpu_cap(c, X86_FEATURE_P4);
-	}
 	if (c->x86 == 6)
 		set_cpu_cap(c, X86_FEATURE_P3);
-	if (cpu_has_ds) {
-		unsigned int l1;
-		rdmsr(MSR_IA32_MISC_ENABLE, l1, l2);
-		if (!(l1 & (1<<11)))
-			set_cpu_cap(c, X86_FEATURE_BTS);
-		if (!(l1 & (1<<12)))
-			set_cpu_cap(c, X86_FEATURE_PEBS);
-		ds_init_intel(c);
-	}
 
 	if (cpu_has_bts)
 		ptrace_bts_init_intel(c);
 
-	/*
-	 * See if we have a good local APIC by checking for buggy Pentia,
-	 * i.e. all B steppings and the C2 stepping of P54C when using their
-	 * integrated APIC (see 11AP erratum in "Pentium Processor
-	 * Specification Update").
-	 */
-	if (cpu_has_apic && (c->x86<<8 | c->x86_model<<4) == 0x520 &&
-	    (c->x86_mask < 0x6 || c->x86_mask == 0xb))
-		set_cpu_cap(c, X86_FEATURE_11AP);
+#endif
 
-#ifdef CONFIG_X86_NUMAQ
-	numaq_tsc_disable();
+	detect_extended_topology(c);
+	if (!cpu_has(c, X86_FEATURE_XTOPOLOGY)) {
+		/*
+		 * let's use the legacy cpuid vector 0x1 and 0x4 for topology
+		 * detection.
+		 */
+		c->x86_max_cores = intel_num_cpu_cores(c);
+#ifdef CONFIG_X86_32
+		detect_ht(c);
 #endif
+	}
+
+	/* Work around errata */
+	srat_detect_node();
+
+	if (cpu_has(c, X86_FEATURE_VMX))
+		detect_vmx_virtcap(c);
 }
 
+#ifdef CONFIG_X86_32
 static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 *c, unsigned int size)
 {
 	/*
@@ -255,10 +345,12 @@ static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 *c, unsigned i
 		size = 256;
 	return size;
 }
+#endif
 
 static struct cpu_dev intel_cpu_dev __cpuinitdata = {
 	.c_vendor	= "Intel",
 	.c_ident	= { "GenuineIntel" },
+#ifdef CONFIG_X86_32
 	.c_models = {
 		{ .vendor = X86_VENDOR_INTEL, .family = 4, .model_names =
 		  {
@@ -308,76 +400,12 @@ static struct cpu_dev intel_cpu_dev __cpuinitdata = {
 		  }
 		},
 	},
+	.c_size_cache	= intel_size_cache,
+#endif
 	.c_early_init   = early_init_intel,
 	.c_init		= init_intel,
-	.c_size_cache	= intel_size_cache,
+	.c_x86_vendor	= X86_VENDOR_INTEL,
 };
 
-cpu_vendor_dev_register(X86_VENDOR_INTEL, &intel_cpu_dev);
-
-#ifndef CONFIG_X86_CMPXCHG
-unsigned long cmpxchg_386_u8(volatile void *ptr, u8 old, u8 new)
-{
-	u8 prev;
-	unsigned long flags;
-
-	/* Poor man's cmpxchg for 386. Unsuitable for SMP */
-	local_irq_save(flags);
-	prev = *(u8 *)ptr;
-	if (prev == old)
-		*(u8 *)ptr = new;
-	local_irq_restore(flags);
-	return prev;
-}
-EXPORT_SYMBOL(cmpxchg_386_u8);
-
-unsigned long cmpxchg_386_u16(volatile void *ptr, u16 old, u16 new)
-{
-	u16 prev;
-	unsigned long flags;
-
-	/* Poor man's cmpxchg for 386. Unsuitable for SMP */
-	local_irq_save(flags);
-	prev = *(u16 *)ptr;
-	if (prev == old)
-		*(u16 *)ptr = new;
-	local_irq_restore(flags);
-	return prev;
-}
-EXPORT_SYMBOL(cmpxchg_386_u16);
-
-unsigned long cmpxchg_386_u32(volatile void *ptr, u32 old, u32 new)
-{
-	u32 prev;
-	unsigned long flags;
-
-	/* Poor man's cmpxchg for 386. Unsuitable for SMP */
-	local_irq_save(flags);
-	prev = *(u32 *)ptr;
-	if (prev == old)
-		*(u32 *)ptr = new;
-	local_irq_restore(flags);
-	return prev;
-}
-EXPORT_SYMBOL(cmpxchg_386_u32);
-#endif
-
-#ifndef CONFIG_X86_CMPXCHG64
-unsigned long long cmpxchg_486_u64(volatile void *ptr, u64 old, u64 new)
-{
-	u64 prev;
-	unsigned long flags;
-
-	/* Poor man's cmpxchg8b for 386 and 486. Unsuitable for SMP */
-	local_irq_save(flags);
-	prev = *(u64 *)ptr;
-	if (prev == old)
-		*(u64 *)ptr = new;
-	local_irq_restore(flags);
-	return prev;
-}
-EXPORT_SYMBOL(cmpxchg_486_u64);
-#endif
-
-/* arch_initcall(intel_cpu_init); */
+cpu_dev_register(intel_cpu_dev);
 
diff --git a/arch/x86/kernel/cpu/intel_64.c b/arch/x86/kernel/cpu/intel_64.c
deleted file mode 100644
index 1019c58d39f0..000000000000
--- a/arch/x86/kernel/cpu/intel_64.c
+++ /dev/null
@@ -1,95 +0,0 @@
-#include <linux/init.h>
-#include <linux/smp.h>
-#include <asm/processor.h>
-#include <asm/ptrace.h>
-#include <asm/topology.h>
-#include <asm/numa_64.h>
-
-#include "cpu.h"
-
-static void __cpuinit early_init_intel(struct cpuinfo_x86 *c)
-{
-	if ((c->x86 == 0xf && c->x86_model >= 0x03) ||
-	    (c->x86 == 0x6 && c->x86_model >= 0x0e))
-		set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC);
-
-	set_cpu_cap(c, X86_FEATURE_SYSENTER32);
-}
-
-/*
- * find out the number of processor cores on the die
- */
-static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c)
-{
-	unsigned int eax, t;
-
-	if (c->cpuid_level < 4)
-		return 1;
-
-	cpuid_count(4, 0, &eax, &t, &t, &t);
-
-	if (eax & 0x1f)
-		return ((eax >> 26) + 1);
-	else
-		return 1;
-}
-
-static void __cpuinit srat_detect_node(void)
-{
-#ifdef CONFIG_NUMA
-	unsigned node;
-	int cpu = smp_processor_id();
-	int apicid = hard_smp_processor_id();
-
-	/* Don't do the funky fallback heuristics the AMD version employs
-	   for now. */
-	node = apicid_to_node[apicid];
-	if (node == NUMA_NO_NODE || !node_online(node))
-		node = first_node(node_online_map);
-	numa_set_node(cpu, node);
-
-	printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node);
-#endif
-}
-
-static void __cpuinit init_intel(struct cpuinfo_x86 *c)
-{
-	init_intel_cacheinfo(c);
-	if (c->cpuid_level > 9) {
-		unsigned eax = cpuid_eax(10);
-		/* Check for version and the number of counters */
-		if ((eax & 0xff) && (((eax>>8) & 0xff) > 1))
-			set_cpu_cap(c, X86_FEATURE_ARCH_PERFMON);
-	}
-
-	if (cpu_has_ds) {
-		unsigned int l1, l2;
-		rdmsr(MSR_IA32_MISC_ENABLE, l1, l2);
-		if (!(l1 & (1<<11)))
-			set_cpu_cap(c, X86_FEATURE_BTS);
-		if (!(l1 & (1<<12)))
-			set_cpu_cap(c, X86_FEATURE_PEBS);
-	}
-
-
-	if (cpu_has_bts)
-		ds_init_intel(c);
-
-	if (c->x86 == 15)
-		c->x86_cache_alignment = c->x86_clflush_size * 2;
-	if (c->x86 == 6)
-		set_cpu_cap(c, X86_FEATURE_REP_GOOD);
-	set_cpu_cap(c, X86_FEATURE_LFENCE_RDTSC);
-	c->x86_max_cores = intel_num_cpu_cores(c);
-
-	srat_detect_node();
-}
-
-static struct cpu_dev intel_cpu_dev __cpuinitdata = {
-	.c_vendor	= "Intel",
-	.c_ident	= { "GenuineIntel" },
-	.c_early_init   = early_init_intel,
-	.c_init		= init_intel,
-};
-cpu_vendor_dev_register(X86_VENDOR_INTEL, &intel_cpu_dev);
-
diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c
index 6b0a10b002f1..3f46afbb1cf1 100644
--- a/arch/x86/kernel/cpu/intel_cacheinfo.c
+++ b/arch/x86/kernel/cpu/intel_cacheinfo.c
@@ -1,8 +1,8 @@
 /*
- *      Routines to indentify caches on Intel CPU.
+ *	Routines to indentify caches on Intel CPU.
  *
- *      Changes:
- *      Venkatesh Pallipadi	: Adding cache identification through cpuid(4)
+ *	Changes:
+ *	Venkatesh Pallipadi	: Adding cache identification through cpuid(4)
  *		Ashok Raj <ashok.raj@intel.com>: Work with CPU hotplug infrastructure.
  *	Andi Kleen / Andreas Herrmann	: CPUID4 emulation on AMD.
  */
@@ -13,6 +13,7 @@
 #include <linux/compiler.h>
 #include <linux/cpu.h>
 #include <linux/sched.h>
+#include <linux/pci.h>
 
 #include <asm/processor.h>
 #include <asm/smp.h>
@@ -130,9 +131,18 @@ struct _cpuid4_info {
 	union _cpuid4_leaf_ebx ebx;
 	union _cpuid4_leaf_ecx ecx;
 	unsigned long size;
+	unsigned long can_disable;
 	cpumask_t shared_cpu_map;	/* future?: only cpus/node is needed */
 };
 
+#ifdef CONFIG_PCI
+static struct pci_device_id k8_nb_id[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, 0x1103) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, 0x1203) },
+	{}
+};
+#endif
+
 unsigned short			num_cache_leaves;
 
 /* AMD doesn't have CPUID4. Emulate it here to report the same
@@ -182,9 +192,10 @@ static unsigned short assocs[] __cpuinitdata = {
 static unsigned char levels[] __cpuinitdata = { 1, 1, 2, 3 };
 static unsigned char types[] __cpuinitdata = { 1, 2, 3, 3 };
 
-static void __cpuinit amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax,
-		       union _cpuid4_leaf_ebx *ebx,
-		       union _cpuid4_leaf_ecx *ecx)
+static void __cpuinit
+amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax,
+		     union _cpuid4_leaf_ebx *ebx,
+		     union _cpuid4_leaf_ecx *ecx)
 {
 	unsigned dummy;
 	unsigned line_size, lines_per_tag, assoc, size_in_kb;
@@ -251,27 +262,40 @@ static void __cpuinit amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax,
 		(ebx->split.ways_of_associativity + 1) - 1;
 }
 
-static int __cpuinit cpuid4_cache_lookup(int index, struct _cpuid4_info *this_leaf)
+static void __cpuinit
+amd_check_l3_disable(int index, struct _cpuid4_info *this_leaf)
+{
+	if (index < 3)
+		return;
+	this_leaf->can_disable = 1;
+}
+
+static int
+__cpuinit cpuid4_cache_lookup(int index, struct _cpuid4_info *this_leaf)
 {
 	union _cpuid4_leaf_eax 	eax;
 	union _cpuid4_leaf_ebx 	ebx;
 	union _cpuid4_leaf_ecx 	ecx;
 	unsigned		edx;
 
-	if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
+	if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) {
 		amd_cpuid4(index, &eax, &ebx, &ecx);
-	else
-		cpuid_count(4, index, &eax.full, &ebx.full, &ecx.full,  &edx);
+		if (boot_cpu_data.x86 >= 0x10)
+			amd_check_l3_disable(index, this_leaf);
+	} else {
+		cpuid_count(4, index, &eax.full, &ebx.full, &ecx.full, &edx);
+	}
+
 	if (eax.split.type == CACHE_TYPE_NULL)
 		return -EIO; /* better error ? */
 
 	this_leaf->eax = eax;
 	this_leaf->ebx = ebx;
 	this_leaf->ecx = ecx;
-	this_leaf->size = (ecx.split.number_of_sets + 1) *
-		(ebx.split.coherency_line_size + 1) *
-		(ebx.split.physical_line_partition + 1) *
-		(ebx.split.ways_of_associativity + 1);
+	this_leaf->size = (ecx.split.number_of_sets          + 1) *
+			  (ebx.split.coherency_line_size     + 1) *
+			  (ebx.split.physical_line_partition + 1) *
+			  (ebx.split.ways_of_associativity   + 1);
 	return 0;
 }
 
@@ -453,7 +477,7 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c)
 
 /* pointer to _cpuid4_info array (for each cache leaf) */
 static DEFINE_PER_CPU(struct _cpuid4_info *, cpuid4_info);
-#define CPUID4_INFO_IDX(x, y)    (&((per_cpu(cpuid4_info, x))[y]))
+#define CPUID4_INFO_IDX(x, y)	(&((per_cpu(cpuid4_info, x))[y]))
 
 #ifdef CONFIG_SMP
 static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index)
@@ -490,7 +514,7 @@ static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index)
 
 	this_leaf = CPUID4_INFO_IDX(cpu, index);
 	for_each_cpu_mask_nr(sibling, this_leaf->shared_cpu_map) {
-		sibling_leaf = CPUID4_INFO_IDX(sibling, index);	
+		sibling_leaf = CPUID4_INFO_IDX(sibling, index);
 		cpu_clear(cpu, sibling_leaf->shared_cpu_map);
 	}
 }
@@ -572,7 +596,7 @@ struct _index_kobject {
 
 /* pointer to array of kobjects for cpuX/cache/indexY */
 static DEFINE_PER_CPU(struct _index_kobject *, index_kobject);
-#define INDEX_KOBJECT_PTR(x, y)    (&((per_cpu(index_kobject, x))[y]))
+#define INDEX_KOBJECT_PTR(x, y)		(&((per_cpu(index_kobject, x))[y]))
 
 #define show_one_plus(file_name, object, val)				\
 static ssize_t show_##file_name						\
@@ -637,6 +661,99 @@ static ssize_t show_type(struct _cpuid4_info *this_leaf, char *buf) {
 	}
 }
 
+#define to_object(k)	container_of(k, struct _index_kobject, kobj)
+#define to_attr(a)	container_of(a, struct _cache_attr, attr)
+
+#ifdef CONFIG_PCI
+static struct pci_dev *get_k8_northbridge(int node)
+{
+	struct pci_dev *dev = NULL;
+	int i;
+
+	for (i = 0; i <= node; i++) {
+		do {
+			dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev);
+			if (!dev)
+				break;
+		} while (!pci_match_id(&k8_nb_id[0], dev));
+		if (!dev)
+			break;
+	}
+	return dev;
+}
+#else
+static struct pci_dev *get_k8_northbridge(int node)
+{
+	return NULL;
+}
+#endif
+
+static ssize_t show_cache_disable(struct _cpuid4_info *this_leaf, char *buf)
+{
+	int node = cpu_to_node(first_cpu(this_leaf->shared_cpu_map));
+	struct pci_dev *dev = NULL;
+	ssize_t ret = 0;
+	int i;
+
+	if (!this_leaf->can_disable)
+		return sprintf(buf, "Feature not enabled\n");
+
+	dev = get_k8_northbridge(node);
+	if (!dev) {
+		printk(KERN_ERR "Attempting AMD northbridge operation on a system with no northbridge\n");
+		return -EINVAL;
+	}
+
+	for (i = 0; i < 2; i++) {
+		unsigned int reg;
+
+		pci_read_config_dword(dev, 0x1BC + i * 4, &reg);
+
+		ret += sprintf(buf, "%sEntry: %d\n", buf, i);
+		ret += sprintf(buf, "%sReads:  %s\tNew Entries: %s\n",  
+			buf,
+			reg & 0x80000000 ? "Disabled" : "Allowed",
+			reg & 0x40000000 ? "Disabled" : "Allowed");
+		ret += sprintf(buf, "%sSubCache: %x\tIndex: %x\n",
+			buf, (reg & 0x30000) >> 16, reg & 0xfff);
+	}
+	return ret;
+}
+
+static ssize_t
+store_cache_disable(struct _cpuid4_info *this_leaf, const char *buf,
+		    size_t count)
+{
+	int node = cpu_to_node(first_cpu(this_leaf->shared_cpu_map));
+	struct pci_dev *dev = NULL;
+	unsigned int ret, index, val;
+
+	if (!this_leaf->can_disable)
+		return 0;
+
+	if (strlen(buf) > 15)
+		return -EINVAL;
+
+	ret = sscanf(buf, "%x %x", &index, &val);
+	if (ret != 2)
+		return -EINVAL;
+	if (index > 1)
+		return -EINVAL;
+
+	val |= 0xc0000000;
+	dev = get_k8_northbridge(node);
+	if (!dev) {
+		printk(KERN_ERR "Attempting AMD northbridge operation on a system with no northbridge\n");
+		return -EINVAL;
+	}
+
+	pci_write_config_dword(dev, 0x1BC + index * 4, val & ~0x40000000);
+	wbinvd();
+	pci_write_config_dword(dev, 0x1BC + index * 4, val);
+
+	return 1;
+}
+
 struct _cache_attr {
 	struct attribute attr;
 	ssize_t (*show)(struct _cpuid4_info *, char *);
@@ -657,6 +774,8 @@ define_one_ro(size);
 define_one_ro(shared_cpu_map);
 define_one_ro(shared_cpu_list);
 
+static struct _cache_attr cache_disable = __ATTR(cache_disable, 0644, show_cache_disable, store_cache_disable);
+
 static struct attribute * default_attrs[] = {
 	&type.attr,
 	&level.attr,
@@ -667,12 +786,10 @@ static struct attribute * default_attrs[] = {
 	&size.attr,
 	&shared_cpu_map.attr,
 	&shared_cpu_list.attr,
+	&cache_disable.attr,
 	NULL
 };
 
-#define to_object(k) container_of(k, struct _index_kobject, kobj)
-#define to_attr(a) container_of(a, struct _cache_attr, attr)
-
 static ssize_t show(struct kobject * kobj, struct attribute * attr, char * buf)
 {
 	struct _cache_attr *fattr = to_attr(attr);
@@ -682,14 +799,22 @@ static ssize_t show(struct kobject * kobj, struct attribute * attr, char * buf)
 	ret = fattr->show ?
 		fattr->show(CPUID4_INFO_IDX(this_leaf->cpu, this_leaf->index),
 			buf) :
-	       	0;
+		0;
 	return ret;
 }
 
 static ssize_t store(struct kobject * kobj, struct attribute * attr,
 		     const char * buf, size_t count)
 {
-	return 0;
+	struct _cache_attr *fattr = to_attr(attr);
+	struct _index_kobject *this_leaf = to_object(kobj);
+	ssize_t ret;
+
+	ret = fattr->store ?
+		fattr->store(CPUID4_INFO_IDX(this_leaf->cpu, this_leaf->index),
+			buf, count) :
+		0;
+	return ret;
 }
 
 static struct sysfs_ops sysfs_ops = {
diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c
index 726a5fcdf341..4b031a4ac856 100644
--- a/arch/x86/kernel/cpu/mcheck/mce_64.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_64.c
@@ -860,7 +860,7 @@ error:
 	return err;
 }
 
-static void mce_remove_device(unsigned int cpu)
+static __cpuinit void mce_remove_device(unsigned int cpu)
 {
 	int i;
 
diff --git a/arch/x86/kernel/cpu/mkcapflags.pl b/arch/x86/kernel/cpu/mkcapflags.pl
new file mode 100644
index 000000000000..dfea390e1608
--- /dev/null
+++ b/arch/x86/kernel/cpu/mkcapflags.pl
@@ -0,0 +1,32 @@
+#!/usr/bin/perl
+#
+# Generate the x86_cap_flags[] array from include/asm-x86/cpufeature.h
+#
+
+($in, $out) = @ARGV;
+
+open(IN, "< $in\0")   or die "$0: cannot open: $in: $!\n";
+open(OUT, "> $out\0") or die "$0: cannot create: $out: $!\n";
+
+print OUT "#include <asm/cpufeature.h>\n\n";
+print OUT "const char * const x86_cap_flags[NCAPINTS*32] = {\n";
+
+while (defined($line = <IN>)) {
+	if ($line =~ /^\s*\#\s*define\s+(X86_FEATURE_(\S+))\s+(.*)$/) {
+		$macro = $1;
+		$feature = $2;
+		$tail = $3;
+		if ($tail =~ /\/\*\s*\"([^"]*)\".*\*\//) {
+			$feature = $1;
+		}
+
+		if ($feature ne '') {
+			printf OUT "\t%-32s = \"%s\",\n",
+				"[$macro]", "\L$feature";
+		}
+	}
+}
+print OUT "};\n";
+
+close(IN);
+close(OUT);
diff --git a/arch/x86/kernel/cpu/powerflags.c b/arch/x86/kernel/cpu/powerflags.c
new file mode 100644
index 000000000000..5abbea297e0c
--- /dev/null
+++ b/arch/x86/kernel/cpu/powerflags.c
@@ -0,0 +1,20 @@
+/*
+ * Strings for the various x86 power flags
+ *
+ * This file must not contain any executable code.
+ */
+
+#include <asm/cpufeature.h>
+
+const char *const x86_power_flags[32] = {
+	"ts",	/* temperature sensor */
+	"fid",  /* frequency id control */
+	"vid",  /* voltage id control */
+	"ttp",  /* thermal trip */
+	"tm",
+	"stc",
+	"100mhzsteps",
+	"hwpstate",
+	"",	/* tsc invariant mapped to constant_tsc */
+		/* nothing */
+};
diff --git a/arch/x86/kernel/cpu/transmeta.c b/arch/x86/kernel/cpu/transmeta.c
index b911a2c61b8f..52b3fefbd5af 100644
--- a/arch/x86/kernel/cpu/transmeta.c
+++ b/arch/x86/kernel/cpu/transmeta.c
@@ -5,6 +5,18 @@
 #include <asm/msr.h>
 #include "cpu.h"
 
+static void __cpuinit early_init_transmeta(struct cpuinfo_x86 *c)
+{
+	u32 xlvl;
+
+	/* Transmeta-defined flags: level 0x80860001 */
+	xlvl = cpuid_eax(0x80860000);
+	if ((xlvl & 0xffff0000) == 0x80860000) {
+		if (xlvl >= 0x80860001)
+			c->x86_capability[2] = cpuid_edx(0x80860001);
+	}
+}
+
 static void __cpuinit init_transmeta(struct cpuinfo_x86 *c)
 {
 	unsigned int cap_mask, uk, max, dummy;
@@ -12,7 +24,8 @@ static void __cpuinit init_transmeta(struct cpuinfo_x86 *c)
 	unsigned int cpu_rev, cpu_freq = 0, cpu_flags, new_cpu_rev;
 	char cpu_info[65];
 
-	get_model_name(c);	/* Same as AMD/Cyrix */
+	early_init_transmeta(c);
+
 	display_cacheinfo(c);
 
 	/* Print CMS and CPU revision */
@@ -85,23 +98,12 @@ static void __cpuinit init_transmeta(struct cpuinfo_x86 *c)
 #endif
 }
 
-static void __cpuinit transmeta_identify(struct cpuinfo_x86 *c)
-{
-	u32 xlvl;
-
-	/* Transmeta-defined flags: level 0x80860001 */
-	xlvl = cpuid_eax(0x80860000);
-	if ((xlvl & 0xffff0000) == 0x80860000) {
-		if (xlvl >= 0x80860001)
-			c->x86_capability[2] = cpuid_edx(0x80860001);
-	}
-}
-
 static struct cpu_dev transmeta_cpu_dev __cpuinitdata = {
 	.c_vendor	= "Transmeta",
 	.c_ident	= { "GenuineTMx86", "TransmetaCPU" },
+	.c_early_init	= early_init_transmeta,
 	.c_init		= init_transmeta,
-	.c_identify	= transmeta_identify,
+	.c_x86_vendor	= X86_VENDOR_TRANSMETA,
 };
 
-cpu_vendor_dev_register(X86_VENDOR_TRANSMETA, &transmeta_cpu_dev);
+cpu_dev_register(transmeta_cpu_dev);
diff --git a/arch/x86/kernel/cpu/umc.c b/arch/x86/kernel/cpu/umc.c
index b1fc90989d75..e777f79e0960 100644
--- a/arch/x86/kernel/cpu/umc.c
+++ b/arch/x86/kernel/cpu/umc.c
@@ -19,7 +19,8 @@ static struct cpu_dev umc_cpu_dev __cpuinitdata = {
 		  }
 		},
 	},
+	.c_x86_vendor	= X86_VENDOR_UMC,
 };
 
-cpu_vendor_dev_register(X86_VENDOR_UMC, &umc_cpu_dev);
+cpu_dev_register(umc_cpu_dev);
 
diff --git a/arch/x86/kernel/doublefault_32.c b/arch/x86/kernel/doublefault_32.c
index a47798b59f07..b4f14c6c09d9 100644
--- a/arch/x86/kernel/doublefault_32.c
+++ b/arch/x86/kernel/doublefault_32.c
@@ -66,6 +66,6 @@ struct tss_struct doublefault_tss __cacheline_aligned = {
 		.ds		= __USER_DS,
 		.fs		= __KERNEL_PERCPU,
 
-		.__cr3		= __pa(swapper_pg_dir)
+		.__cr3		= __pa_nodebug(swapper_pg_dir),
 	}
 };
diff --git a/arch/x86/kernel/dumpstack_32.c b/arch/x86/kernel/dumpstack_32.c
new file mode 100644
index 000000000000..201ee359a1a9
--- /dev/null
+++ b/arch/x86/kernel/dumpstack_32.c
@@ -0,0 +1,447 @@
+/*
+ *  Copyright (C) 1991, 1992  Linus Torvalds
+ *  Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs
+ */
+#include <linux/kallsyms.h>
+#include <linux/kprobes.h>
+#include <linux/uaccess.h>
+#include <linux/utsname.h>
+#include <linux/hardirq.h>
+#include <linux/kdebug.h>
+#include <linux/module.h>
+#include <linux/ptrace.h>
+#include <linux/kexec.h>
+#include <linux/bug.h>
+#include <linux/nmi.h>
+
+#include <asm/stacktrace.h>
+
+#define STACKSLOTS_PER_LINE 8
+#define get_bp(bp) asm("movl %%ebp, %0" : "=r" (bp) :)
+
+int panic_on_unrecovered_nmi;
+int kstack_depth_to_print = 3 * STACKSLOTS_PER_LINE;
+static unsigned int code_bytes = 64;
+static int die_counter;
+
+void printk_address(unsigned long address, int reliable)
+{
+	printk(" [<%p>] %s%pS\n", (void *) address,
+			reliable ? "" : "? ", (void *) address);
+}
+
+static inline int valid_stack_ptr(struct thread_info *tinfo,
+			void *p, unsigned int size, void *end)
+{
+	void *t = tinfo;
+	if (end) {
+		if (p < end && p >= (end-THREAD_SIZE))
+			return 1;
+		else
+			return 0;
+	}
+	return p > t && p < t + THREAD_SIZE - size;
+}
+
+/* The form of the top of the frame on the stack */
+struct stack_frame {
+	struct stack_frame *next_frame;
+	unsigned long return_address;
+};
+
+static inline unsigned long
+print_context_stack(struct thread_info *tinfo,
+		unsigned long *stack, unsigned long bp,
+		const struct stacktrace_ops *ops, void *data,
+		unsigned long *end)
+{
+	struct stack_frame *frame = (struct stack_frame *)bp;
+
+	while (valid_stack_ptr(tinfo, stack, sizeof(*stack), end)) {
+		unsigned long addr;
+
+		addr = *stack;
+		if (__kernel_text_address(addr)) {
+			if ((unsigned long) stack == bp + sizeof(long)) {
+				ops->address(data, addr, 1);
+				frame = frame->next_frame;
+				bp = (unsigned long) frame;
+			} else {
+				ops->address(data, addr, bp == 0);
+			}
+		}
+		stack++;
+	}
+	return bp;
+}
+
+void dump_trace(struct task_struct *task, struct pt_regs *regs,
+		unsigned long *stack, unsigned long bp,
+		const struct stacktrace_ops *ops, void *data)
+{
+	if (!task)
+		task = current;
+
+	if (!stack) {
+		unsigned long dummy;
+		stack = &dummy;
+		if (task && task != current)
+			stack = (unsigned long *)task->thread.sp;
+	}
+
+#ifdef CONFIG_FRAME_POINTER
+	if (!bp) {
+		if (task == current) {
+			/* Grab bp right from our regs */
+			get_bp(bp);
+		} else {
+			/* bp is the last reg pushed by switch_to */
+			bp = *(unsigned long *) task->thread.sp;
+		}
+	}
+#endif
+
+	for (;;) {
+		struct thread_info *context;
+
+		context = (struct thread_info *)
+			((unsigned long)stack & (~(THREAD_SIZE - 1)));
+		bp = print_context_stack(context, stack, bp, ops, data, NULL);
+
+		stack = (unsigned long *)context->previous_esp;
+		if (!stack)
+			break;
+		if (ops->stack(data, "IRQ") < 0)
+			break;
+		touch_nmi_watchdog();
+	}
+}
+EXPORT_SYMBOL(dump_trace);
+
+static void
+print_trace_warning_symbol(void *data, char *msg, unsigned long symbol)
+{
+	printk(data);
+	print_symbol(msg, symbol);
+	printk("\n");
+}
+
+static void print_trace_warning(void *data, char *msg)
+{
+	printk("%s%s\n", (char *)data, msg);
+}
+
+static int print_trace_stack(void *data, char *name)
+{
+	printk("%s <%s> ", (char *)data, name);
+	return 0;
+}
+
+/*
+ * Print one address/symbol entries per line.
+ */
+static void print_trace_address(void *data, unsigned long addr, int reliable)
+{
+	touch_nmi_watchdog();
+	printk(data);
+	printk_address(addr, reliable);
+}
+
+static const struct stacktrace_ops print_trace_ops = {
+	.warning = print_trace_warning,
+	.warning_symbol = print_trace_warning_symbol,
+	.stack = print_trace_stack,
+	.address = print_trace_address,
+};
+
+static void
+show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
+		unsigned long *stack, unsigned long bp, char *log_lvl)
+{
+	printk("%sCall Trace:\n", log_lvl);
+	dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl);
+}
+
+void show_trace(struct task_struct *task, struct pt_regs *regs,
+		unsigned long *stack, unsigned long bp)
+{
+	show_trace_log_lvl(task, regs, stack, bp, "");
+}
+
+static void
+show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
+		unsigned long *sp, unsigned long bp, char *log_lvl)
+{
+	unsigned long *stack;
+	int i;
+
+	if (sp == NULL) {
+		if (task)
+			sp = (unsigned long *)task->thread.sp;
+		else
+			sp = (unsigned long *)&sp;
+	}
+
+	stack = sp;
+	for (i = 0; i < kstack_depth_to_print; i++) {
+		if (kstack_end(stack))
+			break;
+		if (i && ((i % STACKSLOTS_PER_LINE) == 0))
+			printk("\n%s", log_lvl);
+		printk(" %08lx", *stack++);
+		touch_nmi_watchdog();
+	}
+	printk("\n");
+	show_trace_log_lvl(task, regs, sp, bp, log_lvl);
+}
+
+void show_stack(struct task_struct *task, unsigned long *sp)
+{
+	show_stack_log_lvl(task, NULL, sp, 0, "");
+}
+
+/*
+ * The architecture-independent dump_stack generator
+ */
+void dump_stack(void)
+{
+	unsigned long bp = 0;
+	unsigned long stack;
+
+#ifdef CONFIG_FRAME_POINTER
+	if (!bp)
+		get_bp(bp);
+#endif
+
+	printk("Pid: %d, comm: %.20s %s %s %.*s\n",
+		current->pid, current->comm, print_tainted(),
+		init_utsname()->release,
+		(int)strcspn(init_utsname()->version, " "),
+		init_utsname()->version);
+	show_trace(NULL, NULL, &stack, bp);
+}
+
+EXPORT_SYMBOL(dump_stack);
+
+void show_registers(struct pt_regs *regs)
+{
+	int i;
+
+	print_modules();
+	__show_regs(regs, 0);
+
+	printk(KERN_EMERG "Process %.*s (pid: %d, ti=%p task=%p task.ti=%p)\n",
+		TASK_COMM_LEN, current->comm, task_pid_nr(current),
+		current_thread_info(), current, task_thread_info(current));
+	/*
+	 * When in-kernel, we also print out the stack and code at the
+	 * time of the fault..
+	 */
+	if (!user_mode_vm(regs)) {
+		unsigned int code_prologue = code_bytes * 43 / 64;
+		unsigned int code_len = code_bytes;
+		unsigned char c;
+		u8 *ip;
+
+		printk(KERN_EMERG "Stack:\n");
+		show_stack_log_lvl(NULL, regs, &regs->sp,
+				0, KERN_EMERG);
+
+		printk(KERN_EMERG "Code: ");
+
+		ip = (u8 *)regs->ip - code_prologue;
+		if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) {
+			/* try starting at IP */
+			ip = (u8 *)regs->ip;
+			code_len = code_len - code_prologue + 1;
+		}
+		for (i = 0; i < code_len; i++, ip++) {
+			if (ip < (u8 *)PAGE_OFFSET ||
+					probe_kernel_address(ip, c)) {
+				printk(" Bad EIP value.");
+				break;
+			}
+			if (ip == (u8 *)regs->ip)
+				printk("<%02x> ", c);
+			else
+				printk("%02x ", c);
+		}
+	}
+	printk("\n");
+}
+
+int is_valid_bugaddr(unsigned long ip)
+{
+	unsigned short ud2;
+
+	if (ip < PAGE_OFFSET)
+		return 0;
+	if (probe_kernel_address((unsigned short *)ip, ud2))
+		return 0;
+
+	return ud2 == 0x0b0f;
+}
+
+static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED;
+static int die_owner = -1;
+static unsigned int die_nest_count;
+
+unsigned __kprobes long oops_begin(void)
+{
+	unsigned long flags;
+
+	oops_enter();
+
+	if (die_owner != raw_smp_processor_id()) {
+		console_verbose();
+		raw_local_irq_save(flags);
+		__raw_spin_lock(&die_lock);
+		die_owner = smp_processor_id();
+		die_nest_count = 0;
+		bust_spinlocks(1);
+	} else {
+		raw_local_irq_save(flags);
+	}
+	die_nest_count++;
+	return flags;
+}
+
+void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr)
+{
+	bust_spinlocks(0);
+	die_owner = -1;
+	add_taint(TAINT_DIE);
+	__raw_spin_unlock(&die_lock);
+	raw_local_irq_restore(flags);
+
+	if (!regs)
+		return;
+
+	if (kexec_should_crash(current))
+		crash_kexec(regs);
+	if (in_interrupt())
+		panic("Fatal exception in interrupt");
+	if (panic_on_oops)
+		panic("Fatal exception");
+	oops_exit();
+	do_exit(signr);
+}
+
+int __kprobes __die(const char *str, struct pt_regs *regs, long err)
+{
+	unsigned short ss;
+	unsigned long sp;
+
+	printk(KERN_EMERG "%s: %04lx [#%d] ", str, err & 0xffff, ++die_counter);
+#ifdef CONFIG_PREEMPT
+	printk("PREEMPT ");
+#endif
+#ifdef CONFIG_SMP
+	printk("SMP ");
+#endif
+#ifdef CONFIG_DEBUG_PAGEALLOC
+	printk("DEBUG_PAGEALLOC");
+#endif
+	printk("\n");
+	if (notify_die(DIE_OOPS, str, regs, err,
+			current->thread.trap_no, SIGSEGV) == NOTIFY_STOP)
+		return 1;
+
+	show_registers(regs);
+	/* Executive summary in case the oops scrolled away */
+	sp = (unsigned long) (&regs->sp);
+	savesegment(ss, ss);
+	if (user_mode(regs)) {
+		sp = regs->sp;
+		ss = regs->ss & 0xffff;
+	}
+	printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip);
+	print_symbol("%s", regs->ip);
+	printk(" SS:ESP %04x:%08lx\n", ss, sp);
+	return 0;
+}
+
+/*
+ * This is gone through when something in the kernel has done something bad
+ * and is about to be terminated:
+ */
+void die(const char *str, struct pt_regs *regs, long err)
+{
+	unsigned long flags = oops_begin();
+
+	if (die_nest_count < 3) {
+		report_bug(regs->ip, regs);
+
+		if (__die(str, regs, err))
+			regs = NULL;
+	} else {
+		printk(KERN_EMERG "Recursive die() failure, output suppressed\n");
+	}
+
+	oops_end(flags, regs, SIGSEGV);
+}
+
+static DEFINE_SPINLOCK(nmi_print_lock);
+
+void notrace __kprobes
+die_nmi(char *str, struct pt_regs *regs, int do_panic)
+{
+	if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP)
+		return;
+
+	spin_lock(&nmi_print_lock);
+	/*
+	* We are in trouble anyway, lets at least try
+	* to get a message out:
+	*/
+	bust_spinlocks(1);
+	printk(KERN_EMERG "%s", str);
+	printk(" on CPU%d, ip %08lx, registers:\n",
+		smp_processor_id(), regs->ip);
+	show_registers(regs);
+	if (do_panic)
+		panic("Non maskable interrupt");
+	console_silent();
+	spin_unlock(&nmi_print_lock);
+	bust_spinlocks(0);
+
+	/*
+	 * If we are in kernel we are probably nested up pretty bad
+	 * and might aswell get out now while we still can:
+	 */
+	if (!user_mode_vm(regs)) {
+		current->thread.trap_no = 2;
+		crash_kexec(regs);
+	}
+
+	do_exit(SIGSEGV);
+}
+
+static int __init oops_setup(char *s)
+{
+	if (!s)
+		return -EINVAL;
+	if (!strcmp(s, "panic"))
+		panic_on_oops = 1;
+	return 0;
+}
+early_param("oops", oops_setup);
+
+static int __init kstack_setup(char *s)
+{
+	if (!s)
+		return -EINVAL;
+	kstack_depth_to_print = simple_strtoul(s, NULL, 0);
+	return 0;
+}
+early_param("kstack", kstack_setup);
+
+static int __init code_bytes_setup(char *s)
+{
+	code_bytes = simple_strtoul(s, NULL, 0);
+	if (code_bytes > 8192)
+		code_bytes = 8192;
+
+	return 1;
+}
+__setup("code_bytes=", code_bytes_setup);
diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c
new file mode 100644
index 000000000000..086cc8118e39
--- /dev/null
+++ b/arch/x86/kernel/dumpstack_64.c
@@ -0,0 +1,573 @@
+/*
+ *  Copyright (C) 1991, 1992  Linus Torvalds
+ *  Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs
+ */
+#include <linux/kallsyms.h>
+#include <linux/kprobes.h>
+#include <linux/uaccess.h>
+#include <linux/utsname.h>
+#include <linux/hardirq.h>
+#include <linux/kdebug.h>
+#include <linux/module.h>
+#include <linux/ptrace.h>
+#include <linux/kexec.h>
+#include <linux/bug.h>
+#include <linux/nmi.h>
+
+#include <asm/stacktrace.h>
+
+#define STACKSLOTS_PER_LINE 4
+#define get_bp(bp) asm("movq %%rbp, %0" : "=r" (bp) :)
+
+int panic_on_unrecovered_nmi;
+int kstack_depth_to_print = 3 * STACKSLOTS_PER_LINE;
+static unsigned int code_bytes = 64;
+static int die_counter;
+
+void printk_address(unsigned long address, int reliable)
+{
+	printk(" [<%p>] %s%pS\n", (void *) address,
+			reliable ? "" : "? ", (void *) address);
+}
+
+static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
+					unsigned *usedp, char **idp)
+{
+	static char ids[][8] = {
+		[DEBUG_STACK - 1] = "#DB",
+		[NMI_STACK - 1] = "NMI",
+		[DOUBLEFAULT_STACK - 1] = "#DF",
+		[STACKFAULT_STACK - 1] = "#SS",
+		[MCE_STACK - 1] = "#MC",
+#if DEBUG_STKSZ > EXCEPTION_STKSZ
+		[N_EXCEPTION_STACKS ...
+			N_EXCEPTION_STACKS + DEBUG_STKSZ / EXCEPTION_STKSZ - 2] = "#DB[?]"
+#endif
+	};
+	unsigned k;
+
+	/*
+	 * Iterate over all exception stacks, and figure out whether
+	 * 'stack' is in one of them:
+	 */
+	for (k = 0; k < N_EXCEPTION_STACKS; k++) {
+		unsigned long end = per_cpu(orig_ist, cpu).ist[k];
+		/*
+		 * Is 'stack' above this exception frame's end?
+		 * If yes then skip to the next frame.
+		 */
+		if (stack >= end)
+			continue;
+		/*
+		 * Is 'stack' above this exception frame's start address?
+		 * If yes then we found the right frame.
+		 */
+		if (stack >= end - EXCEPTION_STKSZ) {
+			/*
+			 * Make sure we only iterate through an exception
+			 * stack once. If it comes up for the second time
+			 * then there's something wrong going on - just
+			 * break out and return NULL:
+			 */
+			if (*usedp & (1U << k))
+				break;
+			*usedp |= 1U << k;
+			*idp = ids[k];
+			return (unsigned long *)end;
+		}
+		/*
+		 * If this is a debug stack, and if it has a larger size than
+		 * the usual exception stacks, then 'stack' might still
+		 * be within the lower portion of the debug stack:
+		 */
+#if DEBUG_STKSZ > EXCEPTION_STKSZ
+		if (k == DEBUG_STACK - 1 && stack >= end - DEBUG_STKSZ) {
+			unsigned j = N_EXCEPTION_STACKS - 1;
+
+			/*
+			 * Black magic. A large debug stack is composed of
+			 * multiple exception stack entries, which we
+			 * iterate through now. Dont look:
+			 */
+			do {
+				++j;
+				end -= EXCEPTION_STKSZ;
+				ids[j][4] = '1' + (j - N_EXCEPTION_STACKS);
+			} while (stack < end - EXCEPTION_STKSZ);
+			if (*usedp & (1U << j))
+				break;
+			*usedp |= 1U << j;
+			*idp = ids[j];
+			return (unsigned long *)end;
+		}
+#endif
+	}
+	return NULL;
+}
+
+/*
+ * x86-64 can have up to three kernel stacks:
+ * process stack
+ * interrupt stack
+ * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack
+ */
+
+static inline int valid_stack_ptr(struct thread_info *tinfo,
+			void *p, unsigned int size, void *end)
+{
+	void *t = tinfo;
+	if (end) {
+		if (p < end && p >= (end-THREAD_SIZE))
+			return 1;
+		else
+			return 0;
+	}
+	return p > t && p < t + THREAD_SIZE - size;
+}
+
+/* The form of the top of the frame on the stack */
+struct stack_frame {
+	struct stack_frame *next_frame;
+	unsigned long return_address;
+};
+
+static inline unsigned long
+print_context_stack(struct thread_info *tinfo,
+		unsigned long *stack, unsigned long bp,
+		const struct stacktrace_ops *ops, void *data,
+		unsigned long *end)
+{
+	struct stack_frame *frame = (struct stack_frame *)bp;
+
+	while (valid_stack_ptr(tinfo, stack, sizeof(*stack), end)) {
+		unsigned long addr;
+
+		addr = *stack;
+		if (__kernel_text_address(addr)) {
+			if ((unsigned long) stack == bp + sizeof(long)) {
+				ops->address(data, addr, 1);
+				frame = frame->next_frame;
+				bp = (unsigned long) frame;
+			} else {
+				ops->address(data, addr, bp == 0);
+			}
+		}
+		stack++;
+	}
+	return bp;
+}
+
+void dump_trace(struct task_struct *task, struct pt_regs *regs,
+		unsigned long *stack, unsigned long bp,
+		const struct stacktrace_ops *ops, void *data)
+{
+	const unsigned cpu = get_cpu();
+	unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr;
+	unsigned used = 0;
+	struct thread_info *tinfo;
+
+	if (!task)
+		task = current;
+
+	if (!stack) {
+		unsigned long dummy;
+		stack = &dummy;
+		if (task && task != current)
+			stack = (unsigned long *)task->thread.sp;
+	}
+
+#ifdef CONFIG_FRAME_POINTER
+	if (!bp) {
+		if (task == current) {
+			/* Grab bp right from our regs */
+			get_bp(bp);
+		} else {
+			/* bp is the last reg pushed by switch_to */
+			bp = *(unsigned long *) task->thread.sp;
+		}
+	}
+#endif
+
+	/*
+	 * Print function call entries in all stacks, starting at the
+	 * current stack address. If the stacks consist of nested
+	 * exceptions
+	 */
+	tinfo = task_thread_info(task);
+	for (;;) {
+		char *id;
+		unsigned long *estack_end;
+		estack_end = in_exception_stack(cpu, (unsigned long)stack,
+						&used, &id);
+
+		if (estack_end) {
+			if (ops->stack(data, id) < 0)
+				break;
+
+			bp = print_context_stack(tinfo, stack, bp, ops,
+							data, estack_end);
+			ops->stack(data, "<EOE>");
+			/*
+			 * We link to the next stack via the
+			 * second-to-last pointer (index -2 to end) in the
+			 * exception stack:
+			 */
+			stack = (unsigned long *) estack_end[-2];
+			continue;
+		}
+		if (irqstack_end) {
+			unsigned long *irqstack;
+			irqstack = irqstack_end -
+				(IRQSTACKSIZE - 64) / sizeof(*irqstack);
+
+			if (stack >= irqstack && stack < irqstack_end) {
+				if (ops->stack(data, "IRQ") < 0)
+					break;
+				bp = print_context_stack(tinfo, stack, bp,
+						ops, data, irqstack_end);
+				/*
+				 * We link to the next stack (which would be
+				 * the process stack normally) the last
+				 * pointer (index -1 to end) in the IRQ stack:
+				 */
+				stack = (unsigned long *) (irqstack_end[-1]);
+				irqstack_end = NULL;
+				ops->stack(data, "EOI");
+				continue;
+			}
+		}
+		break;
+	}
+
+	/*
+	 * This handles the process stack:
+	 */
+	bp = print_context_stack(tinfo, stack, bp, ops, data, NULL);
+	put_cpu();
+}
+EXPORT_SYMBOL(dump_trace);
+
+static void
+print_trace_warning_symbol(void *data, char *msg, unsigned long symbol)
+{
+	printk(data);
+	print_symbol(msg, symbol);
+	printk("\n");
+}
+
+static void print_trace_warning(void *data, char *msg)
+{
+	printk("%s%s\n", (char *)data, msg);
+}
+
+static int print_trace_stack(void *data, char *name)
+{
+	printk("%s <%s> ", (char *)data, name);
+	return 0;
+}
+
+/*
+ * Print one address/symbol entries per line.
+ */
+static void print_trace_address(void *data, unsigned long addr, int reliable)
+{
+	touch_nmi_watchdog();
+	printk(data);
+	printk_address(addr, reliable);
+}
+
+static const struct stacktrace_ops print_trace_ops = {
+	.warning = print_trace_warning,
+	.warning_symbol = print_trace_warning_symbol,
+	.stack = print_trace_stack,
+	.address = print_trace_address,
+};
+
+static void
+show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
+		unsigned long *stack, unsigned long bp, char *log_lvl)
+{
+	printk("%sCall Trace:\n", log_lvl);
+	dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl);
+}
+
+void show_trace(struct task_struct *task, struct pt_regs *regs,
+		unsigned long *stack, unsigned long bp)
+{
+	show_trace_log_lvl(task, regs, stack, bp, "");
+}
+
+static void
+show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
+		unsigned long *sp, unsigned long bp, char *log_lvl)
+{
+	unsigned long *stack;
+	int i;
+	const int cpu = smp_processor_id();
+	unsigned long *irqstack_end =
+		(unsigned long *) (cpu_pda(cpu)->irqstackptr);
+	unsigned long *irqstack =
+		(unsigned long *) (cpu_pda(cpu)->irqstackptr - IRQSTACKSIZE);
+
+	/*
+	 * debugging aid: "show_stack(NULL, NULL);" prints the
+	 * back trace for this cpu.
+	 */
+
+	if (sp == NULL) {
+		if (task)
+			sp = (unsigned long *)task->thread.sp;
+		else
+			sp = (unsigned long *)&sp;
+	}
+
+	stack = sp;
+	for (i = 0; i < kstack_depth_to_print; i++) {
+		if (stack >= irqstack && stack <= irqstack_end) {
+			if (stack == irqstack_end) {
+				stack = (unsigned long *) (irqstack_end[-1]);
+				printk(" <EOI> ");
+			}
+		} else {
+		if (((long) stack & (THREAD_SIZE-1)) == 0)
+			break;
+		}
+		if (i && ((i % STACKSLOTS_PER_LINE) == 0))
+			printk("\n%s", log_lvl);
+		printk(" %016lx", *stack++);
+		touch_nmi_watchdog();
+	}
+	printk("\n");
+	show_trace_log_lvl(task, regs, sp, bp, log_lvl);
+}
+
+void show_stack(struct task_struct *task, unsigned long *sp)
+{
+	show_stack_log_lvl(task, NULL, sp, 0, "");
+}
+
+/*
+ * The architecture-independent dump_stack generator
+ */
+void dump_stack(void)
+{
+	unsigned long bp = 0;
+	unsigned long stack;
+
+#ifdef CONFIG_FRAME_POINTER
+	if (!bp)
+		get_bp(bp);
+#endif
+
+	printk("Pid: %d, comm: %.20s %s %s %.*s\n",
+		current->pid, current->comm, print_tainted(),
+		init_utsname()->release,
+		(int)strcspn(init_utsname()->version, " "),
+		init_utsname()->version);
+	show_trace(NULL, NULL, &stack, bp);
+}
+EXPORT_SYMBOL(dump_stack);
+
+void show_registers(struct pt_regs *regs)
+{
+	int i;
+	unsigned long sp;
+	const int cpu = smp_processor_id();
+	struct task_struct *cur = cpu_pda(cpu)->pcurrent;
+
+	sp = regs->sp;
+	printk("CPU %d ", cpu);
+	__show_regs(regs, 1);
+	printk("Process %s (pid: %d, threadinfo %p, task %p)\n",
+		cur->comm, cur->pid, task_thread_info(cur), cur);
+
+	/*
+	 * When in-kernel, we also print out the stack and code at the
+	 * time of the fault..
+	 */
+	if (!user_mode(regs)) {
+		unsigned int code_prologue = code_bytes * 43 / 64;
+		unsigned int code_len = code_bytes;
+		unsigned char c;
+		u8 *ip;
+
+		printk(KERN_EMERG "Stack:\n");
+		show_stack_log_lvl(NULL, regs, (unsigned long *)sp,
+				regs->bp, KERN_EMERG);
+
+		printk(KERN_EMERG "Code: ");
+
+		ip = (u8 *)regs->ip - code_prologue;
+		if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) {
+			/* try starting at IP */
+			ip = (u8 *)regs->ip;
+			code_len = code_len - code_prologue + 1;
+		}
+		for (i = 0; i < code_len; i++, ip++) {
+			if (ip < (u8 *)PAGE_OFFSET ||
+					probe_kernel_address(ip, c)) {
+				printk(" Bad RIP value.");
+				break;
+			}
+			if (ip == (u8 *)regs->ip)
+				printk("<%02x> ", c);
+			else
+				printk("%02x ", c);
+		}
+	}
+	printk("\n");
+}
+
+int is_valid_bugaddr(unsigned long ip)
+{
+	unsigned short ud2;
+
+	if (__copy_from_user(&ud2, (const void __user *) ip, sizeof(ud2)))
+		return 0;
+
+	return ud2 == 0x0b0f;
+}
+
+static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED;
+static int die_owner = -1;
+static unsigned int die_nest_count;
+
+unsigned __kprobes long oops_begin(void)
+{
+	int cpu;
+	unsigned long flags;
+
+	oops_enter();
+
+	/* racy, but better than risking deadlock. */
+	raw_local_irq_save(flags);
+	cpu = smp_processor_id();
+	if (!__raw_spin_trylock(&die_lock)) {
+		if (cpu == die_owner)
+			/* nested oops. should stop eventually */;
+		else
+			__raw_spin_lock(&die_lock);
+	}
+	die_nest_count++;
+	die_owner = cpu;
+	console_verbose();
+	bust_spinlocks(1);
+	return flags;
+}
+
+void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr)
+{
+	die_owner = -1;
+	bust_spinlocks(0);
+	die_nest_count--;
+	if (!die_nest_count)
+		/* Nest count reaches zero, release the lock. */
+		__raw_spin_unlock(&die_lock);
+	raw_local_irq_restore(flags);
+	if (!regs) {
+		oops_exit();
+		return;
+	}
+	if (in_interrupt())
+		panic("Fatal exception in interrupt");
+	if (panic_on_oops)
+		panic("Fatal exception");
+	oops_exit();
+	do_exit(signr);
+}
+
+int __kprobes __die(const char *str, struct pt_regs *regs, long err)
+{
+	printk(KERN_EMERG "%s: %04lx [#%d] ", str, err & 0xffff, ++die_counter);
+#ifdef CONFIG_PREEMPT
+	printk("PREEMPT ");
+#endif
+#ifdef CONFIG_SMP
+	printk("SMP ");
+#endif
+#ifdef CONFIG_DEBUG_PAGEALLOC
+	printk("DEBUG_PAGEALLOC");
+#endif
+	printk("\n");
+	if (notify_die(DIE_OOPS, str, regs, err,
+			current->thread.trap_no, SIGSEGV) == NOTIFY_STOP)
+		return 1;
+
+	show_registers(regs);
+	add_taint(TAINT_DIE);
+	/* Executive summary in case the oops scrolled away */
+	printk(KERN_ALERT "RIP ");
+	printk_address(regs->ip, 1);
+	printk(" RSP <%016lx>\n", regs->sp);
+	if (kexec_should_crash(current))
+		crash_kexec(regs);
+	return 0;
+}
+
+void die(const char *str, struct pt_regs *regs, long err)
+{
+	unsigned long flags = oops_begin();
+
+	if (!user_mode(regs))
+		report_bug(regs->ip, regs);
+
+	if (__die(str, regs, err))
+		regs = NULL;
+	oops_end(flags, regs, SIGSEGV);
+}
+
+notrace __kprobes void
+die_nmi(char *str, struct pt_regs *regs, int do_panic)
+{
+	unsigned long flags;
+
+	if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP)
+		return;
+
+	flags = oops_begin();
+	/*
+	 * We are in trouble anyway, lets at least try
+	 * to get a message out.
+	 */
+	printk(KERN_EMERG "%s", str);
+	printk(" on CPU%d, ip %08lx, registers:\n",
+		smp_processor_id(), regs->ip);
+	show_registers(regs);
+	if (kexec_should_crash(current))
+		crash_kexec(regs);
+	if (do_panic || panic_on_oops)
+		panic("Non maskable interrupt");
+	oops_end(flags, NULL, SIGBUS);
+	nmi_exit();
+	local_irq_enable();
+	do_exit(SIGBUS);
+}
+
+static int __init oops_setup(char *s)
+{
+	if (!s)
+		return -EINVAL;
+	if (!strcmp(s, "panic"))
+		panic_on_oops = 1;
+	return 0;
+}
+early_param("oops", oops_setup);
+
+static int __init kstack_setup(char *s)
+{
+	if (!s)
+		return -EINVAL;
+	kstack_depth_to_print = simple_strtoul(s, NULL, 0);
+	return 0;
+}
+early_param("kstack", kstack_setup);
+
+static int __init code_bytes_setup(char *s)
+{
+	code_bytes = simple_strtoul(s, NULL, 0);
+	if (code_bytes > 8192)
+		code_bytes = 8192;
+
+	return 1;
+}
+__setup("code_bytes=", code_bytes_setup);
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 66e48aa2dd1b..78e642feac30 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -148,6 +148,9 @@ void __init e820_print_map(char *who)
 		case E820_NVS:
 			printk(KERN_CONT "(ACPI NVS)\n");
 			break;
+		case E820_UNUSABLE:
+			printk("(unusable)\n");
+			break;
 		default:
 			printk(KERN_CONT "type %u\n", e820.map[i].type);
 			break;
@@ -1260,6 +1263,7 @@ static inline const char *e820_type_to_string(int e820_type)
 	case E820_RAM:	return "System RAM";
 	case E820_ACPI:	return "ACPI Tables";
 	case E820_NVS:	return "ACPI Non-volatile Storage";
+	case E820_UNUSABLE:	return "Unusable memory";
 	default:	return "reserved";
 	}
 }
@@ -1267,6 +1271,7 @@ static inline const char *e820_type_to_string(int e820_type)
 /*
  * Mark e820 reserved areas as busy for the resource manager.
  */
+static struct resource __initdata *e820_res;
 void __init e820_reserve_resources(void)
 {
 	int i;
@@ -1274,6 +1279,7 @@ void __init e820_reserve_resources(void)
 	u64 end;
 
 	res = alloc_bootmem_low(sizeof(struct resource) * e820.nr_map);
+	e820_res = res;
 	for (i = 0; i < e820.nr_map; i++) {
 		end = e820.map[i].addr + e820.map[i].size - 1;
 #ifndef CONFIG_RESOURCES_64BIT
@@ -1287,7 +1293,14 @@ void __init e820_reserve_resources(void)
 		res->end = end;
 
 		res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
-		insert_resource(&iomem_resource, res);
+
+		/*
+		 * don't register the region that could be conflicted with
+		 * pci device BAR resource and insert them later in
+		 * pcibios_resource_survey()
+		 */
+		if (e820.map[i].type != E820_RESERVED || res->start < (1ULL<<20))
+			insert_resource(&iomem_resource, res);
 		res++;
 	}
 
@@ -1299,6 +1312,19 @@ void __init e820_reserve_resources(void)
 	}
 }
 
+void __init e820_reserve_resources_late(void)
+{
+	int i;
+	struct resource *res;
+
+	res = e820_res;
+	for (i = 0; i < e820.nr_map; i++) {
+		if (!res->parent && res->end)
+			reserve_region_with_split(&iomem_resource, res->start, res->end, res->name);
+		res++;
+	}
+}
+
 char *__init default_machine_specific_memory_setup(void)
 {
 	char *who = "BIOS-e820";
diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
index 24bb5faf5efa..733c4f8d42ea 100644
--- a/arch/x86/kernel/early-quirks.c
+++ b/arch/x86/kernel/early-quirks.c
@@ -95,6 +95,52 @@ static void __init nvidia_bugs(int num, int slot, int func)
 
 }
 
+static u32 ati_ixp4x0_rev(int num, int slot, int func)
+{
+	u32 d;
+	u8  b;
+
+	b = read_pci_config_byte(num, slot, func, 0xac);
+	b &= ~(1<<5);
+	write_pci_config_byte(num, slot, func, 0xac, b);
+
+	d = read_pci_config(num, slot, func, 0x70);
+	d |= 1<<8;
+	write_pci_config(num, slot, func, 0x70, d);
+
+	d = read_pci_config(num, slot, func, 0x8);
+	d &= 0xff;
+	return d;
+}
+
+static void __init ati_bugs(int num, int slot, int func)
+{
+#if defined(CONFIG_ACPI) && defined (CONFIG_X86_IO_APIC)
+	u32 d;
+	u8  b;
+
+	if (acpi_use_timer_override)
+		return;
+
+	d = ati_ixp4x0_rev(num, slot, func);
+	if (d  < 0x82)
+		acpi_skip_timer_override = 1;
+	else {
+		/* check for IRQ0 interrupt swap */
+		outb(0x72, 0xcd6); b = inb(0xcd7);
+		if (!(b & 0x2))
+			acpi_skip_timer_override = 1;
+	}
+
+	if (acpi_skip_timer_override) {
+		printk(KERN_INFO "SB4X0 revision 0x%x\n", d);
+		printk(KERN_INFO "Ignoring ACPI timer override.\n");
+		printk(KERN_INFO "If you got timer trouble "
+		       "try acpi_use_timer_override\n");
+	}
+#endif
+}
+
 #ifdef CONFIG_DMAR
 static void __init intel_g33_dmar(int num, int slot, int func)
 {
@@ -128,6 +174,8 @@ static struct chipset early_qrk[] __initdata = {
 	  PCI_CLASS_BRIDGE_PCI, PCI_ANY_ID, QFLAG_APPLY_ONCE, via_bugs },
 	{ PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB,
 	  PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, fix_hypertransport_config },
+	{ PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP400_SMBUS,
+	  PCI_CLASS_SERIAL_SMBUS, PCI_ANY_ID, 0, ati_bugs },
 #ifdef CONFIG_DMAR
 	{ PCI_VENDOR_ID_INTEL, 0x29c0,
 	  PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, intel_g33_dmar },
diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c
index ff9e7350da54..34ad997d3834 100644
--- a/arch/x86/kernel/early_printk.c
+++ b/arch/x86/kernel/early_printk.c
@@ -3,11 +3,19 @@
 #include <linux/init.h>
 #include <linux/string.h>
 #include <linux/screen_info.h>
+#include <linux/usb/ch9.h>
+#include <linux/pci_regs.h>
+#include <linux/pci_ids.h>
+#include <linux/errno.h>
 #include <asm/io.h>
 #include <asm/processor.h>
 #include <asm/fcntl.h>
 #include <asm/setup.h>
 #include <xen/hvc-console.h>
+#include <asm/pci-direct.h>
+#include <asm/pgtable.h>
+#include <asm/fixmap.h>
+#include <linux/usb/ehci_def.h>
 
 /* Simple VGA output */
 #define VGABASE		(__ISA_IO_base + 0xb8000)
@@ -78,6 +86,7 @@ static int early_serial_base = 0x3f8;  /* ttyS0 */
 static int early_serial_putc(unsigned char ch)
 {
 	unsigned timeout = 0xffff;
+
 	while ((inb(early_serial_base + LSR) & XMTRDY) == 0 && --timeout)
 		cpu_relax();
 	outb(ch, early_serial_base + TXR);
@@ -111,7 +120,7 @@ static __init void early_serial_init(char *s)
 		if (!strncmp(s, "0x", 2)) {
 			early_serial_base = simple_strtoul(s, &e, 16);
 		} else {
-			static int bases[] = { 0x3f8, 0x2f8 };
+			static const int __initconst bases[] = { 0x3f8, 0x2f8 };
 
 			if (!strncmp(s, "ttyS", 4))
 				s += 4;
@@ -151,6 +160,721 @@ static struct console early_serial_console = {
 	.index =	-1,
 };
 
+#ifdef CONFIG_EARLY_PRINTK_DBGP
+
+static struct ehci_caps __iomem *ehci_caps;
+static struct ehci_regs __iomem *ehci_regs;
+static struct ehci_dbg_port __iomem *ehci_debug;
+static unsigned int dbgp_endpoint_out;
+
+struct ehci_dev {
+	u32 bus;
+	u32 slot;
+	u32 func;
+};
+
+static struct ehci_dev ehci_dev;
+
+#define USB_DEBUG_DEVNUM 127
+
+#define DBGP_DATA_TOGGLE	0x8800
+
+static inline u32 dbgp_pid_update(u32 x, u32 tok)
+{
+	return ((x ^ DBGP_DATA_TOGGLE) & 0xffff00) | (tok & 0xff);
+}
+
+static inline u32 dbgp_len_update(u32 x, u32 len)
+{
+	return (x & ~0x0f) | (len & 0x0f);
+}
+
+/*
+ * USB Packet IDs (PIDs)
+ */
+
+/* token */
+#define USB_PID_OUT		0xe1
+#define USB_PID_IN		0x69
+#define USB_PID_SOF		0xa5
+#define USB_PID_SETUP		0x2d
+/* handshake */
+#define USB_PID_ACK		0xd2
+#define USB_PID_NAK		0x5a
+#define USB_PID_STALL		0x1e
+#define USB_PID_NYET		0x96
+/* data */
+#define USB_PID_DATA0		0xc3
+#define USB_PID_DATA1		0x4b
+#define USB_PID_DATA2		0x87
+#define USB_PID_MDATA		0x0f
+/* Special */
+#define USB_PID_PREAMBLE	0x3c
+#define USB_PID_ERR		0x3c
+#define USB_PID_SPLIT		0x78
+#define USB_PID_PING		0xb4
+#define USB_PID_UNDEF_0		0xf0
+
+#define USB_PID_DATA_TOGGLE	0x88
+#define DBGP_CLAIM (DBGP_OWNER | DBGP_ENABLED | DBGP_INUSE)
+
+#define PCI_CAP_ID_EHCI_DEBUG	0xa
+
+#define HUB_ROOT_RESET_TIME	50	/* times are in msec */
+#define HUB_SHORT_RESET_TIME	10
+#define HUB_LONG_RESET_TIME	200
+#define HUB_RESET_TIMEOUT	500
+
+#define DBGP_MAX_PACKET		8
+
+static int dbgp_wait_until_complete(void)
+{
+	u32 ctrl;
+	int loop = 0x100000;
+
+	do {
+		ctrl = readl(&ehci_debug->control);
+		/* Stop when the transaction is finished */
+		if (ctrl & DBGP_DONE)
+			break;
+	} while (--loop > 0);
+
+	if (!loop)
+		return -1;
+
+	/*
+	 * Now that we have observed the completed transaction,
+	 * clear the done bit.
+	 */
+	writel(ctrl | DBGP_DONE, &ehci_debug->control);
+	return (ctrl & DBGP_ERROR) ? -DBGP_ERRCODE(ctrl) : DBGP_LEN(ctrl);
+}
+
+static void dbgp_mdelay(int ms)
+{
+	int i;
+
+	while (ms--) {
+		for (i = 0; i < 1000; i++)
+			outb(0x1, 0x80);
+	}
+}
+
+static void dbgp_breath(void)
+{
+	/* Sleep to give the debug port a chance to breathe */
+}
+
+static int dbgp_wait_until_done(unsigned ctrl)
+{
+	u32 pids, lpid;
+	int ret;
+	int loop = 3;
+
+retry:
+	writel(ctrl | DBGP_GO, &ehci_debug->control);
+	ret = dbgp_wait_until_complete();
+	pids = readl(&ehci_debug->pids);
+	lpid = DBGP_PID_GET(pids);
+
+	if (ret < 0)
+		return ret;
+
+	/*
+	 * If the port is getting full or it has dropped data
+	 * start pacing ourselves, not necessary but it's friendly.
+	 */
+	if ((lpid == USB_PID_NAK) || (lpid == USB_PID_NYET))
+		dbgp_breath();
+
+	/* If I get a NACK reissue the transmission */
+	if (lpid == USB_PID_NAK) {
+		if (--loop > 0)
+			goto retry;
+	}
+
+	return ret;
+}
+
+static void dbgp_set_data(const void *buf, int size)
+{
+	const unsigned char *bytes = buf;
+	u32 lo, hi;
+	int i;
+
+	lo = hi = 0;
+	for (i = 0; i < 4 && i < size; i++)
+		lo |= bytes[i] << (8*i);
+	for (; i < 8 && i < size; i++)
+		hi |= bytes[i] << (8*(i - 4));
+	writel(lo, &ehci_debug->data03);
+	writel(hi, &ehci_debug->data47);
+}
+
+static void dbgp_get_data(void *buf, int size)
+{
+	unsigned char *bytes = buf;
+	u32 lo, hi;
+	int i;
+
+	lo = readl(&ehci_debug->data03);
+	hi = readl(&ehci_debug->data47);
+	for (i = 0; i < 4 && i < size; i++)
+		bytes[i] = (lo >> (8*i)) & 0xff;
+	for (; i < 8 && i < size; i++)
+		bytes[i] = (hi >> (8*(i - 4))) & 0xff;
+}
+
+static int dbgp_bulk_write(unsigned devnum, unsigned endpoint,
+			 const char *bytes, int size)
+{
+	u32 pids, addr, ctrl;
+	int ret;
+
+	if (size > DBGP_MAX_PACKET)
+		return -1;
+
+	addr = DBGP_EPADDR(devnum, endpoint);
+
+	pids = readl(&ehci_debug->pids);
+	pids = dbgp_pid_update(pids, USB_PID_OUT);
+
+	ctrl = readl(&ehci_debug->control);
+	ctrl = dbgp_len_update(ctrl, size);
+	ctrl |= DBGP_OUT;
+	ctrl |= DBGP_GO;
+
+	dbgp_set_data(bytes, size);
+	writel(addr, &ehci_debug->address);
+	writel(pids, &ehci_debug->pids);
+
+	ret = dbgp_wait_until_done(ctrl);
+	if (ret < 0)
+		return ret;
+
+	return ret;
+}
+
+static int dbgp_bulk_read(unsigned devnum, unsigned endpoint, void *data,
+				 int size)
+{
+	u32 pids, addr, ctrl;
+	int ret;
+
+	if (size > DBGP_MAX_PACKET)
+		return -1;
+
+	addr = DBGP_EPADDR(devnum, endpoint);
+
+	pids = readl(&ehci_debug->pids);
+	pids = dbgp_pid_update(pids, USB_PID_IN);
+
+	ctrl = readl(&ehci_debug->control);
+	ctrl = dbgp_len_update(ctrl, size);
+	ctrl &= ~DBGP_OUT;
+	ctrl |= DBGP_GO;
+
+	writel(addr, &ehci_debug->address);
+	writel(pids, &ehci_debug->pids);
+	ret = dbgp_wait_until_done(ctrl);
+	if (ret < 0)
+		return ret;
+
+	if (size > ret)
+		size = ret;
+	dbgp_get_data(data, size);
+	return ret;
+}
+
+static int dbgp_control_msg(unsigned devnum, int requesttype, int request,
+	int value, int index, void *data, int size)
+{
+	u32 pids, addr, ctrl;
+	struct usb_ctrlrequest req;
+	int read;
+	int ret;
+
+	read = (requesttype & USB_DIR_IN) != 0;
+	if (size > (read ? DBGP_MAX_PACKET:0))
+		return -1;
+
+	/* Compute the control message */
+	req.bRequestType = requesttype;
+	req.bRequest = request;
+	req.wValue = cpu_to_le16(value);
+	req.wIndex = cpu_to_le16(index);
+	req.wLength = cpu_to_le16(size);
+
+	pids = DBGP_PID_SET(USB_PID_DATA0, USB_PID_SETUP);
+	addr = DBGP_EPADDR(devnum, 0);
+
+	ctrl = readl(&ehci_debug->control);
+	ctrl = dbgp_len_update(ctrl, sizeof(req));
+	ctrl |= DBGP_OUT;
+	ctrl |= DBGP_GO;
+
+	/* Send the setup message */
+	dbgp_set_data(&req, sizeof(req));
+	writel(addr, &ehci_debug->address);
+	writel(pids, &ehci_debug->pids);
+	ret = dbgp_wait_until_done(ctrl);
+	if (ret < 0)
+		return ret;
+
+	/* Read the result */
+	return dbgp_bulk_read(devnum, 0, data, size);
+}
+
+
+/* Find a PCI capability */
+static u32 __init find_cap(u32 num, u32 slot, u32 func, int cap)
+{
+	u8 pos;
+	int bytes;
+
+	if (!(read_pci_config_16(num, slot, func, PCI_STATUS) &
+		PCI_STATUS_CAP_LIST))
+		return 0;
+
+	pos = read_pci_config_byte(num, slot, func, PCI_CAPABILITY_LIST);
+	for (bytes = 0; bytes < 48 && pos >= 0x40; bytes++) {
+		u8 id;
+
+		pos &= ~3;
+		id = read_pci_config_byte(num, slot, func, pos+PCI_CAP_LIST_ID);
+		if (id == 0xff)
+			break;
+		if (id == cap)
+			return pos;
+
+		pos = read_pci_config_byte(num, slot, func,
+						 pos+PCI_CAP_LIST_NEXT);
+	}
+	return 0;
+}
+
+static u32 __init __find_dbgp(u32 bus, u32 slot, u32 func)
+{
+	u32 class;
+
+	class = read_pci_config(bus, slot, func, PCI_CLASS_REVISION);
+	if ((class >> 8) != PCI_CLASS_SERIAL_USB_EHCI)
+		return 0;
+
+	return find_cap(bus, slot, func, PCI_CAP_ID_EHCI_DEBUG);
+}
+
+static u32 __init find_dbgp(int ehci_num, u32 *rbus, u32 *rslot, u32 *rfunc)
+{
+	u32 bus, slot, func;
+
+	for (bus = 0; bus < 256; bus++) {
+		for (slot = 0; slot < 32; slot++) {
+			for (func = 0; func < 8; func++) {
+				unsigned cap;
+
+				cap = __find_dbgp(bus, slot, func);
+
+				if (!cap)
+					continue;
+				if (ehci_num-- != 0)
+					continue;
+				*rbus = bus;
+				*rslot = slot;
+				*rfunc = func;
+				return cap;
+			}
+		}
+	}
+	return 0;
+}
+
+static int ehci_reset_port(int port)
+{
+	u32 portsc;
+	u32 delay_time, delay;
+	int loop;
+
+	/* Reset the usb debug port */
+	portsc = readl(&ehci_regs->port_status[port - 1]);
+	portsc &= ~PORT_PE;
+	portsc |= PORT_RESET;
+	writel(portsc, &ehci_regs->port_status[port - 1]);
+
+	delay = HUB_ROOT_RESET_TIME;
+	for (delay_time = 0; delay_time < HUB_RESET_TIMEOUT;
+	     delay_time += delay) {
+		dbgp_mdelay(delay);
+
+		portsc = readl(&ehci_regs->port_status[port - 1]);
+		if (portsc & PORT_RESET) {
+			/* force reset to complete */
+			loop = 2;
+			writel(portsc & ~(PORT_RWC_BITS | PORT_RESET),
+				&ehci_regs->port_status[port - 1]);
+			do {
+				portsc = readl(&ehci_regs->port_status[port-1]);
+			} while ((portsc & PORT_RESET) && (--loop > 0));
+		}
+
+		/* Device went away? */
+		if (!(portsc & PORT_CONNECT))
+			return -ENOTCONN;
+
+		/* bomb out completely if something weird happend */
+		if ((portsc & PORT_CSC))
+			return -EINVAL;
+
+		/* If we've finished resetting, then break out of the loop */
+		if (!(portsc & PORT_RESET) && (portsc & PORT_PE))
+			return 0;
+	}
+	return -EBUSY;
+}
+
+static int ehci_wait_for_port(int port)
+{
+	u32 status;
+	int ret, reps;
+
+	for (reps = 0; reps < 3; reps++) {
+		dbgp_mdelay(100);
+		status = readl(&ehci_regs->status);
+		if (status & STS_PCD) {
+			ret = ehci_reset_port(port);
+			if (ret == 0)
+				return 0;
+		}
+	}
+	return -ENOTCONN;
+}
+
+#ifdef DBGP_DEBUG
+# define dbgp_printk early_printk
+#else
+static inline void dbgp_printk(const char *fmt, ...) { }
+#endif
+
+typedef void (*set_debug_port_t)(int port);
+
+static void default_set_debug_port(int port)
+{
+}
+
+static set_debug_port_t set_debug_port = default_set_debug_port;
+
+static void nvidia_set_debug_port(int port)
+{
+	u32 dword;
+	dword = read_pci_config(ehci_dev.bus, ehci_dev.slot, ehci_dev.func,
+				 0x74);
+	dword &= ~(0x0f<<12);
+	dword |= ((port & 0x0f)<<12);
+	write_pci_config(ehci_dev.bus, ehci_dev.slot, ehci_dev.func, 0x74,
+				 dword);
+	dbgp_printk("set debug port to %d\n", port);
+}
+
+static void __init detect_set_debug_port(void)
+{
+	u32 vendorid;
+
+	vendorid = read_pci_config(ehci_dev.bus, ehci_dev.slot, ehci_dev.func,
+		 0x00);
+
+	if ((vendorid & 0xffff) == 0x10de) {
+		dbgp_printk("using nvidia set_debug_port\n");
+		set_debug_port = nvidia_set_debug_port;
+	}
+}
+
+static int __init ehci_setup(void)
+{
+	struct usb_debug_descriptor dbgp_desc;
+	u32 cmd, ctrl, status, portsc, hcs_params;
+	u32 debug_port, new_debug_port = 0, n_ports;
+	u32  devnum;
+	int ret, i;
+	int loop;
+	int port_map_tried;
+	int playtimes = 3;
+
+try_next_time:
+	port_map_tried = 0;
+
+try_next_port:
+
+	hcs_params = readl(&ehci_caps->hcs_params);
+	debug_port = HCS_DEBUG_PORT(hcs_params);
+	n_ports    = HCS_N_PORTS(hcs_params);
+
+	dbgp_printk("debug_port: %d\n", debug_port);
+	dbgp_printk("n_ports:    %d\n", n_ports);
+
+	for (i = 1; i <= n_ports; i++) {
+		portsc = readl(&ehci_regs->port_status[i-1]);
+		dbgp_printk("portstatus%d: %08x\n", i, portsc);
+	}
+
+	if (port_map_tried && (new_debug_port != debug_port)) {
+		if (--playtimes) {
+			set_debug_port(new_debug_port);
+			goto try_next_time;
+		}
+		return -1;
+	}
+
+	loop = 10;
+	/* Reset the EHCI controller */
+	cmd = readl(&ehci_regs->command);
+	cmd |= CMD_RESET;
+	writel(cmd, &ehci_regs->command);
+	do {
+		cmd = readl(&ehci_regs->command);
+	} while ((cmd & CMD_RESET) && (--loop > 0));
+
+	if (!loop) {
+		dbgp_printk("can not reset ehci\n");
+		return -1;
+	}
+	dbgp_printk("ehci reset done\n");
+
+	/* Claim ownership, but do not enable yet */
+	ctrl = readl(&ehci_debug->control);
+	ctrl |= DBGP_OWNER;
+	ctrl &= ~(DBGP_ENABLED | DBGP_INUSE);
+	writel(ctrl, &ehci_debug->control);
+
+	/* Start the ehci running */
+	cmd = readl(&ehci_regs->command);
+	cmd &= ~(CMD_LRESET | CMD_IAAD | CMD_PSE | CMD_ASE | CMD_RESET);
+	cmd |= CMD_RUN;
+	writel(cmd, &ehci_regs->command);
+
+	/* Ensure everything is routed to the EHCI */
+	writel(FLAG_CF, &ehci_regs->configured_flag);
+
+	/* Wait until the controller is no longer halted */
+	loop = 10;
+	do {
+		status = readl(&ehci_regs->status);
+	} while ((status & STS_HALT) && (--loop > 0));
+
+	if (!loop) {
+		dbgp_printk("ehci can be started\n");
+		return -1;
+	}
+	dbgp_printk("ehci started\n");
+
+	/* Wait for a device to show up in the debug port */
+	ret = ehci_wait_for_port(debug_port);
+	if (ret < 0) {
+		dbgp_printk("No device found in debug port\n");
+		goto next_debug_port;
+	}
+	dbgp_printk("ehci wait for port done\n");
+
+	/* Enable the debug port */
+	ctrl = readl(&ehci_debug->control);
+	ctrl |= DBGP_CLAIM;
+	writel(ctrl, &ehci_debug->control);
+	ctrl = readl(&ehci_debug->control);
+	if ((ctrl & DBGP_CLAIM) != DBGP_CLAIM) {
+		dbgp_printk("No device in debug port\n");
+		writel(ctrl & ~DBGP_CLAIM, &ehci_debug->control);
+		goto err;
+	}
+	dbgp_printk("debug ported enabled\n");
+
+	/* Completely transfer the debug device to the debug controller */
+	portsc = readl(&ehci_regs->port_status[debug_port - 1]);
+	portsc &= ~PORT_PE;
+	writel(portsc, &ehci_regs->port_status[debug_port - 1]);
+
+	dbgp_mdelay(100);
+
+	/* Find the debug device and make it device number 127 */
+	for (devnum = 0; devnum <= 127; devnum++) {
+		ret = dbgp_control_msg(devnum,
+			USB_DIR_IN | USB_TYPE_STANDARD | USB_RECIP_DEVICE,
+			USB_REQ_GET_DESCRIPTOR, (USB_DT_DEBUG << 8), 0,
+			&dbgp_desc, sizeof(dbgp_desc));
+		if (ret > 0)
+			break;
+	}
+	if (devnum > 127) {
+		dbgp_printk("Could not find attached debug device\n");
+		goto err;
+	}
+	if (ret < 0) {
+		dbgp_printk("Attached device is not a debug device\n");
+		goto err;
+	}
+	dbgp_endpoint_out = dbgp_desc.bDebugOutEndpoint;
+
+	/* Move the device to 127 if it isn't already there */
+	if (devnum != USB_DEBUG_DEVNUM) {
+		ret = dbgp_control_msg(devnum,
+			USB_DIR_OUT | USB_TYPE_STANDARD | USB_RECIP_DEVICE,
+			USB_REQ_SET_ADDRESS, USB_DEBUG_DEVNUM, 0, NULL, 0);
+		if (ret < 0) {
+			dbgp_printk("Could not move attached device to %d\n",
+				USB_DEBUG_DEVNUM);
+			goto err;
+		}
+		devnum = USB_DEBUG_DEVNUM;
+		dbgp_printk("debug device renamed to 127\n");
+	}
+
+	/* Enable the debug interface */
+	ret = dbgp_control_msg(USB_DEBUG_DEVNUM,
+		USB_DIR_OUT | USB_TYPE_STANDARD | USB_RECIP_DEVICE,
+		USB_REQ_SET_FEATURE, USB_DEVICE_DEBUG_MODE, 0, NULL, 0);
+	if (ret < 0) {
+		dbgp_printk(" Could not enable the debug device\n");
+		goto err;
+	}
+	dbgp_printk("debug interface enabled\n");
+
+	/* Perform a small write to get the even/odd data state in sync
+	 */
+	ret = dbgp_bulk_write(USB_DEBUG_DEVNUM, dbgp_endpoint_out, " ", 1);
+	if (ret < 0) {
+		dbgp_printk("dbgp_bulk_write failed: %d\n", ret);
+		goto err;
+	}
+	dbgp_printk("small write doned\n");
+
+	return 0;
+err:
+	/* Things didn't work so remove my claim */
+	ctrl = readl(&ehci_debug->control);
+	ctrl &= ~(DBGP_CLAIM | DBGP_OUT);
+	writel(ctrl, &ehci_debug->control);
+	return -1;
+
+next_debug_port:
+	port_map_tried |= (1<<(debug_port - 1));
+	new_debug_port = ((debug_port-1+1)%n_ports) + 1;
+	if (port_map_tried != ((1<<n_ports) - 1)) {
+		set_debug_port(new_debug_port);
+		goto try_next_port;
+	}
+	if (--playtimes) {
+		set_debug_port(new_debug_port);
+		goto try_next_time;
+	}
+
+	return -1;
+}
+
+static int __init early_dbgp_init(char *s)
+{
+	u32 debug_port, bar, offset;
+	u32 bus, slot, func, cap;
+	void __iomem *ehci_bar;
+	u32 dbgp_num;
+	u32 bar_val;
+	char *e;
+	int ret;
+	u8 byte;
+
+	if (!early_pci_allowed())
+		return -1;
+
+	dbgp_num = 0;
+	if (*s)
+		dbgp_num = simple_strtoul(s, &e, 10);
+	dbgp_printk("dbgp_num: %d\n", dbgp_num);
+
+	cap = find_dbgp(dbgp_num, &bus, &slot, &func);
+	if (!cap)
+		return -1;
+
+	dbgp_printk("Found EHCI debug port on %02x:%02x.%1x\n", bus, slot,
+			 func);
+
+	debug_port = read_pci_config(bus, slot, func, cap);
+	bar = (debug_port >> 29) & 0x7;
+	bar = (bar * 4) + 0xc;
+	offset = (debug_port >> 16) & 0xfff;
+	dbgp_printk("bar: %02x offset: %03x\n", bar, offset);
+	if (bar != PCI_BASE_ADDRESS_0) {
+		dbgp_printk("only debug ports on bar 1 handled.\n");
+
+		return -1;
+	}
+
+	bar_val = read_pci_config(bus, slot, func, PCI_BASE_ADDRESS_0);
+	dbgp_printk("bar_val: %02x offset: %03x\n", bar_val, offset);
+	if (bar_val & ~PCI_BASE_ADDRESS_MEM_MASK) {
+		dbgp_printk("only simple 32bit mmio bars supported\n");
+
+		return -1;
+	}
+
+	/* double check if the mem space is enabled */
+	byte = read_pci_config_byte(bus, slot, func, 0x04);
+	if (!(byte & 0x2)) {
+		byte  |= 0x02;
+		write_pci_config_byte(bus, slot, func, 0x04, byte);
+		dbgp_printk("mmio for ehci enabled\n");
+	}
+
+	/*
+	 * FIXME I don't have the bar size so just guess PAGE_SIZE is more
+	 * than enough.  1K is the biggest I have seen.
+	 */
+	set_fixmap_nocache(FIX_DBGP_BASE, bar_val & PAGE_MASK);
+	ehci_bar = (void __iomem *)__fix_to_virt(FIX_DBGP_BASE);
+	ehci_bar += bar_val & ~PAGE_MASK;
+	dbgp_printk("ehci_bar: %p\n", ehci_bar);
+
+	ehci_caps  = ehci_bar;
+	ehci_regs  = ehci_bar + HC_LENGTH(readl(&ehci_caps->hc_capbase));
+	ehci_debug = ehci_bar + offset;
+	ehci_dev.bus = bus;
+	ehci_dev.slot = slot;
+	ehci_dev.func = func;
+
+	detect_set_debug_port();
+
+	ret = ehci_setup();
+	if (ret < 0) {
+		dbgp_printk("ehci_setup failed\n");
+		ehci_debug = NULL;
+
+		return -1;
+	}
+
+	return 0;
+}
+
+static void early_dbgp_write(struct console *con, const char *str, u32 n)
+{
+	int chunk, ret;
+
+	if (!ehci_debug)
+		return;
+	while (n > 0) {
+		chunk = n;
+		if (chunk > DBGP_MAX_PACKET)
+			chunk = DBGP_MAX_PACKET;
+		ret = dbgp_bulk_write(USB_DEBUG_DEVNUM,
+			dbgp_endpoint_out, str, chunk);
+		str += chunk;
+		n -= chunk;
+	}
+}
+
+static struct console early_dbgp_console = {
+	.name =		"earlydbg",
+	.write =	early_dbgp_write,
+	.flags =	CON_PRINTBUFFER,
+	.index =	-1,
+};
+#endif
+
 /* Console interface to a host file on AMD's SimNow! */
 
 static int simnow_fd;
@@ -165,6 +889,7 @@ enum {
 static noinline long simnow(long cmd, long a, long b, long c)
 {
 	long ret;
+
 	asm volatile("cpuid" :
 		     "=a" (ret) :
 		     "b" (a), "c" (b), "d" (c), "0" (MAGIC1), "D" (cmd + MAGIC2));
@@ -174,6 +899,7 @@ static noinline long simnow(long cmd, long a, long b, long c)
 static void __init simnow_init(char *str)
 {
 	char *fn = "klog";
+
 	if (*str == '=')
 		fn = ++str;
 	/* error ignored */
@@ -194,7 +920,7 @@ static struct console simnow_console = {
 
 /* Direct interface for emergencies */
 static struct console *early_console = &early_vga_console;
-static int early_console_initialized;
+static int __initdata early_console_initialized;
 
 asmlinkage void early_printk(const char *fmt, ...)
 {
@@ -208,10 +934,11 @@ asmlinkage void early_printk(const char *fmt, ...)
 	va_end(ap);
 }
 
-static int __initdata keep_early;
 
 static int __init setup_early_printk(char *buf)
 {
+	int keep_early;
+
 	if (!buf)
 		return 0;
 
@@ -219,8 +946,7 @@ static int __init setup_early_printk(char *buf)
 		return 0;
 	early_console_initialized = 1;
 
-	if (strstr(buf, "keep"))
-		keep_early = 1;
+	keep_early = (strstr(buf, "keep") != NULL);
 
 	if (!strncmp(buf, "serial", 6)) {
 		early_serial_init(buf + 6);
@@ -238,6 +964,17 @@ static int __init setup_early_printk(char *buf)
 		simnow_init(buf + 6);
 		early_console = &simnow_console;
 		keep_early = 1;
+#ifdef CONFIG_EARLY_PRINTK_DBGP
+	} else if (!strncmp(buf, "dbgp", 4)) {
+		if (early_dbgp_init(buf+4) < 0)
+			return 0;
+		early_console = &early_dbgp_console;
+		/*
+		 * usb subsys will reset ehci controller, so don't keep
+		 * that early console
+		 */
+		keep_early = 0;
+#endif
 #ifdef CONFIG_HVC_XEN
 	} else if (!strncmp(buf, "xen", 3)) {
 		early_console = &xenboot_console;
@@ -251,4 +988,5 @@ static int __init setup_early_printk(char *buf)
 	register_console(early_console);
 	return 0;
 }
+
 early_param("earlyprintk", setup_early_printk);
diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
index 109792bc7cfa..b21fbfaffe39 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -730,6 +730,7 @@ error_code:
 	movl $(__USER_DS), %ecx
 	movl %ecx, %ds
 	movl %ecx, %es
+	TRACE_IRQS_OFF
 	movl %esp,%eax			# pt_regs pointer
 	call *%edi
 	jmp ret_from_exception
@@ -760,20 +761,9 @@ ENTRY(device_not_available)
 	RING0_INT_FRAME
 	pushl $-1			# mark this as an int
 	CFI_ADJUST_CFA_OFFSET 4
-	SAVE_ALL
-	GET_CR0_INTO_EAX
-	testl $0x4, %eax		# EM (math emulation bit)
-	jne device_not_available_emulate
-	preempt_stop(CLBR_ANY)
-	call math_state_restore
-	jmp ret_from_exception
-device_not_available_emulate:
-	pushl $0			# temporary storage for ORIG_EIP
+	pushl $do_device_not_available
 	CFI_ADJUST_CFA_OFFSET 4
-	call math_emulate
-	addl $4, %esp
-	CFI_ADJUST_CFA_OFFSET -4
-	jmp ret_from_exception
+	jmp error_code
 	CFI_ENDPROC
 END(device_not_available)
 
@@ -814,6 +804,7 @@ debug_stack_correct:
 	pushl $-1			# mark this as an int
 	CFI_ADJUST_CFA_OFFSET 4
 	SAVE_ALL
+	TRACE_IRQS_OFF
 	xorl %edx,%edx			# error code 0
 	movl %esp,%eax			# pt_regs pointer
 	call do_debug
@@ -858,6 +849,7 @@ nmi_stack_correct:
 	pushl %eax
 	CFI_ADJUST_CFA_OFFSET 4
 	SAVE_ALL
+	TRACE_IRQS_OFF
 	xorl %edx,%edx		# zero error code
 	movl %esp,%eax		# pt_regs pointer
 	call do_nmi
@@ -898,6 +890,7 @@ nmi_espfix_stack:
 	pushl %eax
 	CFI_ADJUST_CFA_OFFSET 4
 	SAVE_ALL
+	TRACE_IRQS_OFF
 	FIXUP_ESPFIX_STACK		# %eax == %esp
 	xorl %edx,%edx			# zero error code
 	call do_nmi
@@ -928,6 +921,7 @@ KPROBE_ENTRY(int3)
 	pushl $-1			# mark this as an int
 	CFI_ADJUST_CFA_OFFSET 4
 	SAVE_ALL
+	TRACE_IRQS_OFF
 	xorl %edx,%edx		# zero error code
 	movl %esp,%eax		# pt_regs pointer
 	call do_int3
@@ -1030,7 +1024,7 @@ ENTRY(machine_check)
 	RING0_INT_FRAME
 	pushl $0
 	CFI_ADJUST_CFA_OFFSET 4
-	pushl machine_check_vector
+	pushl $do_machine_check
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index cf3a0b2d0059..1db6ce4314e1 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -667,6 +667,13 @@ END(stub_rt_sigreturn)
 	SAVE_ARGS
 	leaq -ARGOFFSET(%rsp),%rdi	# arg1 for handler
 	pushq %rbp
+	/*
+	 * Save rbp twice: One is for marking the stack frame, as usual, and the
+	 * other, to fill pt_regs properly. This is because bx comes right
+	 * before the last saved register in that structure, and not bp. If the
+	 * base pointer were in the place bx is today, this would not be needed.
+	 */
+	movq %rbp, -8(%rsp)
 	CFI_ADJUST_CFA_OFFSET	8
 	CFI_REL_OFFSET		rbp, 0
 	movq %rsp,%rbp
@@ -932,6 +939,9 @@ END(spurious_interrupt)
 	.if \ist
 	movq	%gs:pda_data_offset, %rbp
 	.endif
+	.if \irqtrace
+	TRACE_IRQS_OFF
+	.endif
 	movq %rsp,%rdi
 	movq ORIG_RAX(%rsp),%rsi
 	movq $-1,ORIG_RAX(%rsp)
@@ -1058,7 +1068,8 @@ KPROBE_ENTRY(error_entry)
 	je  error_kernelspace
 error_swapgs:	
 	SWAPGS
-error_sti:	
+error_sti:
+	TRACE_IRQS_OFF
 	movq %rdi,RDI(%rsp) 	
 	CFI_REL_OFFSET	rdi,RDI
 	movq %rsp,%rdi
@@ -1232,7 +1243,7 @@ ENTRY(simd_coprocessor_error)
 END(simd_coprocessor_error)
 
 ENTRY(device_not_available)
-	zeroentry math_state_restore
+	zeroentry do_device_not_available
 END(device_not_available)
 
 	/* runs on exception stack */
diff --git a/arch/x86/mach-es7000/es7000plat.c b/arch/x86/kernel/es7000_32.c
index 50189af14b85..f454c78fcef6 100644
--- a/arch/x86/mach-es7000/es7000plat.c
+++ b/arch/x86/kernel/es7000_32.c
@@ -39,10 +39,94 @@
 #include <asm/nmi.h>
 #include <asm/smp.h>
 #include <asm/apicdef.h>
-#include "es7000.h"
 #include <mach_mpparse.h>
 
 /*
+ * ES7000 chipsets
+ */
+
+#define NON_UNISYS		0
+#define ES7000_CLASSIC		1
+#define ES7000_ZORRO		2
+
+
+#define	MIP_REG			1
+#define	MIP_PSAI_REG		4
+
+#define	MIP_BUSY		1
+#define	MIP_SPIN		0xf0000
+#define	MIP_VALID		0x0100000000000000ULL
+#define	MIP_PORT(VALUE)	((VALUE >> 32) & 0xffff)
+
+#define	MIP_RD_LO(VALUE)	(VALUE & 0xffffffff)
+
+struct mip_reg_info {
+	unsigned long long mip_info;
+	unsigned long long delivery_info;
+	unsigned long long host_reg;
+	unsigned long long mip_reg;
+};
+
+struct part_info {
+	unsigned char type;
+	unsigned char length;
+	unsigned char part_id;
+	unsigned char apic_mode;
+	unsigned long snum;
+	char ptype[16];
+	char sname[64];
+	char pname[64];
+};
+
+struct psai {
+	unsigned long long entry_type;
+	unsigned long long addr;
+	unsigned long long bep_addr;
+};
+
+struct es7000_mem_info {
+	unsigned char type;
+	unsigned char length;
+	unsigned char resv[6];
+	unsigned long long  start;
+	unsigned long long  size;
+};
+
+struct es7000_oem_table {
+	unsigned long long hdr;
+	struct mip_reg_info mip;
+	struct part_info pif;
+	struct es7000_mem_info shm;
+	struct psai psai;
+};
+
+#ifdef CONFIG_ACPI
+
+struct oem_table {
+	struct acpi_table_header Header;
+	u32 OEMTableAddr;
+	u32 OEMTableSize;
+};
+
+extern int find_unisys_acpi_oem_table(unsigned long *oem_addr);
+extern void unmap_unisys_acpi_oem_table(unsigned long oem_addr);
+#endif
+
+struct mip_reg {
+	unsigned long long off_0;
+	unsigned long long off_8;
+	unsigned long long off_10;
+	unsigned long long off_18;
+	unsigned long long off_20;
+	unsigned long long off_28;
+	unsigned long long off_30;
+	unsigned long long off_38;
+};
+
+#define	MIP_SW_APIC		0x1020b
+#define	MIP_FUNC(VALUE)		(VALUE & 0xff)
+
+/*
  * ES7000 Globals
  */
 
@@ -72,7 +156,7 @@ es7000_rename_gsi(int ioapic, int gsi)
 			base += nr_ioapic_registers[i];
 	}
 
-	if (!ioapic && (gsi < 16)) 
+	if (!ioapic && (gsi < 16))
 		gsi += base;
 	return gsi;
 }
@@ -160,21 +244,38 @@ parse_unisys_oem (char *oemptr)
 }
 
 #ifdef CONFIG_ACPI
-int __init
-find_unisys_acpi_oem_table(unsigned long *oem_addr)
+static unsigned long oem_addrX;
+static unsigned long oem_size;
+int __init find_unisys_acpi_oem_table(unsigned long *oem_addr)
 {
 	struct acpi_table_header *header = NULL;
 	int i = 0;
-	while (ACPI_SUCCESS(acpi_get_table("OEM1", i++, &header))) {
+	acpi_size tbl_size;
+
+	while (ACPI_SUCCESS(acpi_get_table_with_size("OEM1", i++, &header, &tbl_size))) {
 		if (!memcmp((char *) &header->oem_id, "UNISYS", 6)) {
 			struct oem_table *t = (struct oem_table *)header;
-			*oem_addr = (unsigned long)__acpi_map_table(t->OEMTableAddr,
-								    t->OEMTableSize);
+
+			oem_addrX = t->OEMTableAddr;
+			oem_size = t->OEMTableSize;
+			early_acpi_os_unmap_memory(header, tbl_size);
+
+			*oem_addr = (unsigned long)__acpi_map_table(oem_addrX,
+								    oem_size);
 			return 0;
 		}
+		early_acpi_os_unmap_memory(header, tbl_size);
 	}
 	return -1;
 }
+
+void __init unmap_unisys_acpi_oem_table(unsigned long oem_addr)
+{
+	if (!oem_addr)
+		return;
+
+	__acpi_unmap_table((char *)oem_addr, oem_size);
+}
 #endif
 
 static void
diff --git a/arch/x86/kernel/genapic_64.c b/arch/x86/kernel/genapic_64.c
index eaff0bbb1444..6c9bfc9e1e95 100644
--- a/arch/x86/kernel/genapic_64.c
+++ b/arch/x86/kernel/genapic_64.c
@@ -16,87 +16,63 @@
 #include <linux/ctype.h>
 #include <linux/init.h>
 #include <linux/hardirq.h>
+#include <linux/dmar.h>
 
 #include <asm/smp.h>
 #include <asm/ipi.h>
 #include <asm/genapic.h>
 
-#ifdef CONFIG_ACPI
-#include <acpi/acpi_bus.h>
-#endif
-
-DEFINE_PER_CPU(int, x2apic_extra_bits);
+extern struct genapic apic_flat;
+extern struct genapic apic_physflat;
+extern struct genapic apic_x2xpic_uv_x;
+extern struct genapic apic_x2apic_phys;
+extern struct genapic apic_x2apic_cluster;
 
 struct genapic __read_mostly *genapic = &apic_flat;
 
-static enum uv_system_type uv_system_type;
+static struct genapic *apic_probe[] __initdata = {
+	&apic_x2apic_uv_x,
+	&apic_x2apic_phys,
+	&apic_x2apic_cluster,
+	&apic_physflat,
+	NULL,
+};
 
 /*
  * Check the APIC IDs in bios_cpu_apicid and choose the APIC mode.
  */
 void __init setup_apic_routing(void)
 {
-	if (uv_system_type == UV_NON_UNIQUE_APIC)
-		genapic = &apic_x2apic_uv_x;
-	else
-#ifdef CONFIG_ACPI
-	/*
-	 * Quirk: some x86_64 machines can only use physical APIC mode
-	 * regardless of how many processors are present (x86_64 ES7000
-	 * is an example).
-	 */
-	if (acpi_gbl_FADT.header.revision > FADT2_REVISION_ID &&
-			(acpi_gbl_FADT.flags & ACPI_FADT_APIC_PHYSICAL))
-		genapic = &apic_physflat;
-	else
-#endif
-
-	if (max_physical_apicid < 8)
-		genapic = &apic_flat;
-	else
-		genapic = &apic_physflat;
+	if (genapic == &apic_x2apic_phys || genapic == &apic_x2apic_cluster) {
+		if (!intr_remapping_enabled)
+			genapic = &apic_flat;
+	}
 
-	printk(KERN_INFO "Setting APIC routing to %s\n", genapic->name);
+	if (genapic == &apic_flat) {
+		if (max_physical_apicid >= 8)
+			genapic = &apic_physflat;
+		printk(KERN_INFO "Setting APIC routing to %s\n", genapic->name);
+	}
 }
 
 /* Same for both flat and physical. */
 
-void send_IPI_self(int vector)
+void apic_send_IPI_self(int vector)
 {
 	__send_IPI_shortcut(APIC_DEST_SELF, vector, APIC_DEST_PHYSICAL);
 }
 
 int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 {
-	if (!strcmp(oem_id, "SGI")) {
-		if (!strcmp(oem_table_id, "UVL"))
-			uv_system_type = UV_LEGACY_APIC;
-		else if (!strcmp(oem_table_id, "UVX"))
-			uv_system_type = UV_X2APIC;
-		else if (!strcmp(oem_table_id, "UVH"))
-			uv_system_type = UV_NON_UNIQUE_APIC;
+	int i;
+
+	for (i = 0; apic_probe[i]; ++i) {
+		if (apic_probe[i]->acpi_madt_oem_check(oem_id, oem_table_id)) {
+			genapic = apic_probe[i];
+			printk(KERN_INFO "Setting APIC routing to %s.\n",
+				genapic->name);
+			return 1;
+		}
 	}
 	return 0;
 }
-
-unsigned int read_apic_id(void)
-{
-	unsigned int id;
-
-	WARN_ON(preemptible() && num_online_cpus() > 1);
-	id = apic_read(APIC_ID);
-	if (uv_system_type >= UV_X2APIC)
-		id  |= __get_cpu_var(x2apic_extra_bits);
-	return id;
-}
-
-enum uv_system_type get_uv_system_type(void)
-{
-	return uv_system_type;
-}
-
-int is_uv_system(void)
-{
-	return uv_system_type != UV_NONE;
-}
-EXPORT_SYMBOL_GPL(is_uv_system);
diff --git a/arch/x86/kernel/genapic_flat_64.c b/arch/x86/kernel/genapic_flat_64.c
index 786548a62d38..9eca5ba7a6b1 100644
--- a/arch/x86/kernel/genapic_flat_64.c
+++ b/arch/x86/kernel/genapic_flat_64.c
@@ -15,9 +15,20 @@
 #include <linux/kernel.h>
 #include <linux/ctype.h>
 #include <linux/init.h>
+#include <linux/hardirq.h>
 #include <asm/smp.h>
 #include <asm/ipi.h>
 #include <asm/genapic.h>
+#include <mach_apicdef.h>
+
+#ifdef CONFIG_ACPI
+#include <acpi/acpi_bus.h>
+#endif
+
+static int __init flat_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+	return 1;
+}
 
 static cpumask_t flat_target_cpus(void)
 {
@@ -95,9 +106,33 @@ static void flat_send_IPI_all(int vector)
 		__send_IPI_shortcut(APIC_DEST_ALLINC, vector, APIC_DEST_LOGICAL);
 }
 
+static unsigned int get_apic_id(unsigned long x)
+{
+	unsigned int id;
+
+	id = (((x)>>24) & 0xFFu);
+	return id;
+}
+
+static unsigned long set_apic_id(unsigned int id)
+{
+	unsigned long x;
+
+	x = ((id & 0xFFu)<<24);
+	return x;
+}
+
+static unsigned int read_xapic_id(void)
+{
+	unsigned int id;
+
+	id = get_apic_id(apic_read(APIC_ID));
+	return id;
+}
+
 static int flat_apic_id_registered(void)
 {
-	return physid_isset(GET_APIC_ID(read_apic_id()), phys_cpu_present_map);
+	return physid_isset(read_xapic_id(), phys_cpu_present_map);
 }
 
 static unsigned int flat_cpu_mask_to_apicid(cpumask_t cpumask)
@@ -112,6 +147,7 @@ static unsigned int phys_pkg_id(int index_msb)
 
 struct genapic apic_flat =  {
 	.name = "flat",
+	.acpi_madt_oem_check = flat_acpi_madt_oem_check,
 	.int_delivery_mode = dest_LowestPrio,
 	.int_dest_mode = (APIC_DEST_LOGICAL != 0),
 	.target_cpus = flat_target_cpus,
@@ -121,8 +157,12 @@ struct genapic apic_flat =  {
 	.send_IPI_all = flat_send_IPI_all,
 	.send_IPI_allbutself = flat_send_IPI_allbutself,
 	.send_IPI_mask = flat_send_IPI_mask,
+	.send_IPI_self = apic_send_IPI_self,
 	.cpu_mask_to_apicid = flat_cpu_mask_to_apicid,
 	.phys_pkg_id = phys_pkg_id,
+	.get_apic_id = get_apic_id,
+	.set_apic_id = set_apic_id,
+	.apic_id_mask = (0xFFu<<24),
 };
 
 /*
@@ -130,6 +170,21 @@ struct genapic apic_flat =  {
  * We cannot use logical delivery in this case because the mask
  * overflows, so use physical mode.
  */
+static int __init physflat_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+#ifdef CONFIG_ACPI
+	/*
+	 * Quirk: some x86_64 machines can only use physical APIC mode
+	 * regardless of how many processors are present (x86_64 ES7000
+	 * is an example).
+	 */
+	if (acpi_gbl_FADT.header.revision > FADT2_REVISION_ID &&
+		(acpi_gbl_FADT.flags & ACPI_FADT_APIC_PHYSICAL))
+		return 1;
+#endif
+
+	return 0;
+}
 
 static cpumask_t physflat_target_cpus(void)
 {
@@ -176,6 +231,7 @@ static unsigned int physflat_cpu_mask_to_apicid(cpumask_t cpumask)
 
 struct genapic apic_physflat =  {
 	.name = "physical flat",
+	.acpi_madt_oem_check = physflat_acpi_madt_oem_check,
 	.int_delivery_mode = dest_Fixed,
 	.int_dest_mode = (APIC_DEST_PHYSICAL != 0),
 	.target_cpus = physflat_target_cpus,
@@ -185,6 +241,10 @@ struct genapic apic_physflat =  {
 	.send_IPI_all = physflat_send_IPI_all,
 	.send_IPI_allbutself = physflat_send_IPI_allbutself,
 	.send_IPI_mask = physflat_send_IPI_mask,
+	.send_IPI_self = apic_send_IPI_self,
 	.cpu_mask_to_apicid = physflat_cpu_mask_to_apicid,
 	.phys_pkg_id = phys_pkg_id,
+	.get_apic_id = get_apic_id,
+	.set_apic_id = set_apic_id,
+	.apic_id_mask = (0xFFu<<24),
 };
diff --git a/arch/x86/kernel/genx2apic_cluster.c b/arch/x86/kernel/genx2apic_cluster.c
new file mode 100644
index 000000000000..e4bf2cc0d743
--- /dev/null
+++ b/arch/x86/kernel/genx2apic_cluster.c
@@ -0,0 +1,159 @@
+#include <linux/threads.h>
+#include <linux/cpumask.h>
+#include <linux/string.h>
+#include <linux/kernel.h>
+#include <linux/ctype.h>
+#include <linux/init.h>
+#include <linux/dmar.h>
+
+#include <asm/smp.h>
+#include <asm/ipi.h>
+#include <asm/genapic.h>
+
+DEFINE_PER_CPU(u32, x86_cpu_to_logical_apicid);
+
+static int __init x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+	if (cpu_has_x2apic)
+		return 1;
+
+	return 0;
+}
+
+/* Start with all IRQs pointing to boot CPU.  IRQ balancing will shift them. */
+
+static cpumask_t x2apic_target_cpus(void)
+{
+	return cpumask_of_cpu(0);
+}
+
+/*
+ * for now each logical cpu is in its own vector allocation domain.
+ */
+static cpumask_t x2apic_vector_allocation_domain(int cpu)
+{
+	cpumask_t domain = CPU_MASK_NONE;
+	cpu_set(cpu, domain);
+	return domain;
+}
+
+static void __x2apic_send_IPI_dest(unsigned int apicid, int vector,
+				   unsigned int dest)
+{
+	unsigned long cfg;
+
+	cfg = __prepare_ICR(0, vector, dest);
+
+	/*
+	 * send the IPI.
+	 */
+	x2apic_icr_write(cfg, apicid);
+}
+
+/*
+ * for now, we send the IPI's one by one in the cpumask.
+ * TBD: Based on the cpu mask, we can send the IPI's to the cluster group
+ * at once. We have 16 cpu's in a cluster. This will minimize IPI register
+ * writes.
+ */
+static void x2apic_send_IPI_mask(cpumask_t mask, int vector)
+{
+	unsigned long flags;
+	unsigned long query_cpu;
+
+	local_irq_save(flags);
+	for_each_cpu_mask(query_cpu, mask) {
+		__x2apic_send_IPI_dest(per_cpu(x86_cpu_to_logical_apicid, query_cpu),
+				       vector, APIC_DEST_LOGICAL);
+	}
+	local_irq_restore(flags);
+}
+
+static void x2apic_send_IPI_allbutself(int vector)
+{
+	cpumask_t mask = cpu_online_map;
+
+	cpu_clear(smp_processor_id(), mask);
+
+	if (!cpus_empty(mask))
+		x2apic_send_IPI_mask(mask, vector);
+}
+
+static void x2apic_send_IPI_all(int vector)
+{
+	x2apic_send_IPI_mask(cpu_online_map, vector);
+}
+
+static int x2apic_apic_id_registered(void)
+{
+	return 1;
+}
+
+static unsigned int x2apic_cpu_mask_to_apicid(cpumask_t cpumask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * May as well be the first.
+	 */
+	cpu = first_cpu(cpumask);
+	if ((unsigned)cpu < NR_CPUS)
+		return per_cpu(x86_cpu_to_logical_apicid, cpu);
+	else
+		return BAD_APICID;
+}
+
+static unsigned int get_apic_id(unsigned long x)
+{
+	unsigned int id;
+
+	id = x;
+	return id;
+}
+
+static unsigned long set_apic_id(unsigned int id)
+{
+	unsigned long x;
+
+	x = id;
+	return x;
+}
+
+static unsigned int phys_pkg_id(int index_msb)
+{
+	return current_cpu_data.initial_apicid >> index_msb;
+}
+
+static void x2apic_send_IPI_self(int vector)
+{
+	apic_write(APIC_SELF_IPI, vector);
+}
+
+static void init_x2apic_ldr(void)
+{
+	int cpu = smp_processor_id();
+
+	per_cpu(x86_cpu_to_logical_apicid, cpu) = apic_read(APIC_LDR);
+	return;
+}
+
+struct genapic apic_x2apic_cluster = {
+	.name = "cluster x2apic",
+	.acpi_madt_oem_check = x2apic_acpi_madt_oem_check,
+	.int_delivery_mode = dest_LowestPrio,
+	.int_dest_mode = (APIC_DEST_LOGICAL != 0),
+	.target_cpus = x2apic_target_cpus,
+	.vector_allocation_domain = x2apic_vector_allocation_domain,
+	.apic_id_registered = x2apic_apic_id_registered,
+	.init_apic_ldr = init_x2apic_ldr,
+	.send_IPI_all = x2apic_send_IPI_all,
+	.send_IPI_allbutself = x2apic_send_IPI_allbutself,
+	.send_IPI_mask = x2apic_send_IPI_mask,
+	.send_IPI_self = x2apic_send_IPI_self,
+	.cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid,
+	.phys_pkg_id = phys_pkg_id,
+	.get_apic_id = get_apic_id,
+	.set_apic_id = set_apic_id,
+	.apic_id_mask = (0xFFFFFFFFu),
+};
diff --git a/arch/x86/kernel/genx2apic_phys.c b/arch/x86/kernel/genx2apic_phys.c
new file mode 100644
index 000000000000..8f1343df2627
--- /dev/null
+++ b/arch/x86/kernel/genx2apic_phys.c
@@ -0,0 +1,154 @@
+#include <linux/threads.h>
+#include <linux/cpumask.h>
+#include <linux/string.h>
+#include <linux/kernel.h>
+#include <linux/ctype.h>
+#include <linux/init.h>
+#include <linux/dmar.h>
+
+#include <asm/smp.h>
+#include <asm/ipi.h>
+#include <asm/genapic.h>
+
+static int x2apic_phys;
+
+static int set_x2apic_phys_mode(char *arg)
+{
+	x2apic_phys = 1;
+	return 0;
+}
+early_param("x2apic_phys", set_x2apic_phys_mode);
+
+static int __init x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+	if (cpu_has_x2apic && x2apic_phys)
+		return 1;
+
+	return 0;
+}
+
+/* Start with all IRQs pointing to boot CPU.  IRQ balancing will shift them. */
+
+static cpumask_t x2apic_target_cpus(void)
+{
+	return cpumask_of_cpu(0);
+}
+
+static cpumask_t x2apic_vector_allocation_domain(int cpu)
+{
+	cpumask_t domain = CPU_MASK_NONE;
+	cpu_set(cpu, domain);
+	return domain;
+}
+
+static void __x2apic_send_IPI_dest(unsigned int apicid, int vector,
+				   unsigned int dest)
+{
+	unsigned long cfg;
+
+	cfg = __prepare_ICR(0, vector, dest);
+
+	/*
+	 * send the IPI.
+	 */
+	x2apic_icr_write(cfg, apicid);
+}
+
+static void x2apic_send_IPI_mask(cpumask_t mask, int vector)
+{
+	unsigned long flags;
+	unsigned long query_cpu;
+
+	local_irq_save(flags);
+	for_each_cpu_mask(query_cpu, mask) {
+		__x2apic_send_IPI_dest(per_cpu(x86_cpu_to_apicid, query_cpu),
+				       vector, APIC_DEST_PHYSICAL);
+	}
+	local_irq_restore(flags);
+}
+
+static void x2apic_send_IPI_allbutself(int vector)
+{
+	cpumask_t mask = cpu_online_map;
+
+	cpu_clear(smp_processor_id(), mask);
+
+	if (!cpus_empty(mask))
+		x2apic_send_IPI_mask(mask, vector);
+}
+
+static void x2apic_send_IPI_all(int vector)
+{
+	x2apic_send_IPI_mask(cpu_online_map, vector);
+}
+
+static int x2apic_apic_id_registered(void)
+{
+	return 1;
+}
+
+static unsigned int x2apic_cpu_mask_to_apicid(cpumask_t cpumask)
+{
+	int cpu;
+
+	/*
+	 * We're using fixed IRQ delivery, can only return one phys APIC ID.
+	 * May as well be the first.
+	 */
+	cpu = first_cpu(cpumask);
+	if ((unsigned)cpu < NR_CPUS)
+		return per_cpu(x86_cpu_to_apicid, cpu);
+	else
+		return BAD_APICID;
+}
+
+static unsigned int get_apic_id(unsigned long x)
+{
+	unsigned int id;
+
+	id = x;
+	return id;
+}
+
+static unsigned long set_apic_id(unsigned int id)
+{
+	unsigned long x;
+
+	x = id;
+	return x;
+}
+
+static unsigned int phys_pkg_id(int index_msb)
+{
+	return current_cpu_data.initial_apicid >> index_msb;
+}
+
+void x2apic_send_IPI_self(int vector)
+{
+	apic_write(APIC_SELF_IPI, vector);
+}
+
+void init_x2apic_ldr(void)
+{
+	return;
+}
+
+struct genapic apic_x2apic_phys = {
+	.name = "physical x2apic",
+	.acpi_madt_oem_check = x2apic_acpi_madt_oem_check,
+	.int_delivery_mode = dest_Fixed,
+	.int_dest_mode = (APIC_DEST_PHYSICAL != 0),
+	.target_cpus = x2apic_target_cpus,
+	.vector_allocation_domain = x2apic_vector_allocation_domain,
+	.apic_id_registered = x2apic_apic_id_registered,
+	.init_apic_ldr = init_x2apic_ldr,
+	.send_IPI_all = x2apic_send_IPI_all,
+	.send_IPI_allbutself = x2apic_send_IPI_allbutself,
+	.send_IPI_mask = x2apic_send_IPI_mask,
+	.send_IPI_self = x2apic_send_IPI_self,
+	.cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid,
+	.phys_pkg_id = phys_pkg_id,
+	.get_apic_id = get_apic_id,
+	.set_apic_id = set_apic_id,
+	.apic_id_mask = (0xFFFFFFFFu),
+};
diff --git a/arch/x86/kernel/genx2apic_uv_x.c b/arch/x86/kernel/genx2apic_uv_x.c
index bfa837cb16be..33581d94a90e 100644
--- a/arch/x86/kernel/genx2apic_uv_x.c
+++ b/arch/x86/kernel/genx2apic_uv_x.c
@@ -12,12 +12,12 @@
 #include <linux/threads.h>
 #include <linux/cpumask.h>
 #include <linux/string.h>
-#include <linux/kernel.h>
 #include <linux/ctype.h>
 #include <linux/init.h>
 #include <linux/sched.h>
 #include <linux/bootmem.h>
 #include <linux/module.h>
+#include <linux/hardirq.h>
 #include <asm/smp.h>
 #include <asm/ipi.h>
 #include <asm/genapic.h>
@@ -26,6 +26,36 @@
 #include <asm/uv/uv_hub.h>
 #include <asm/uv/bios.h>
 
+DEFINE_PER_CPU(int, x2apic_extra_bits);
+
+static enum uv_system_type uv_system_type;
+
+static int __init uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+	if (!strcmp(oem_id, "SGI")) {
+		if (!strcmp(oem_table_id, "UVL"))
+			uv_system_type = UV_LEGACY_APIC;
+		else if (!strcmp(oem_table_id, "UVX"))
+			uv_system_type = UV_X2APIC;
+		else if (!strcmp(oem_table_id, "UVH")) {
+			uv_system_type = UV_NON_UNIQUE_APIC;
+			return 1;
+		}
+	}
+	return 0;
+}
+
+enum uv_system_type get_uv_system_type(void)
+{
+	return uv_system_type;
+}
+
+int is_uv_system(void)
+{
+	return uv_system_type != UV_NONE;
+}
+EXPORT_SYMBOL_GPL(is_uv_system);
+
 DEFINE_PER_CPU(struct uv_hub_info_s, __uv_hub_info);
 EXPORT_PER_CPU_SYMBOL_GPL(__uv_hub_info);
 
@@ -84,7 +114,7 @@ static void uv_send_IPI_one(int cpu, int vector)
 	unsigned long val, apicid, lapicid;
 	int pnode;
 
-	apicid = per_cpu(x86_cpu_to_apicid, cpu); /* ZZZ - cache node-local ? */
+	apicid = per_cpu(x86_cpu_to_apicid, cpu);
 	lapicid = apicid & 0x3f;		/* ZZZ macro needed */
 	pnode = uv_apicid_to_pnode(apicid);
 	val =
@@ -123,6 +153,10 @@ static int uv_apic_id_registered(void)
 	return 1;
 }
 
+static void uv_init_apic_ldr(void)
+{
+}
+
 static unsigned int uv_cpu_mask_to_apicid(cpumask_t cpumask)
 {
 	int cpu;
@@ -138,31 +172,59 @@ static unsigned int uv_cpu_mask_to_apicid(cpumask_t cpumask)
 		return BAD_APICID;
 }
 
+static unsigned int get_apic_id(unsigned long x)
+{
+	unsigned int id;
+
+	WARN_ON(preemptible() && num_online_cpus() > 1);
+	id = x | __get_cpu_var(x2apic_extra_bits);
+
+	return id;
+}
+
+static unsigned long set_apic_id(unsigned int id)
+{
+	unsigned long x;
+
+	/* maskout x2apic_extra_bits ? */
+	x = id;
+	return x;
+}
+
+static unsigned int uv_read_apic_id(void)
+{
+
+	return get_apic_id(apic_read(APIC_ID));
+}
+
 static unsigned int phys_pkg_id(int index_msb)
 {
-	return GET_APIC_ID(read_apic_id()) >> index_msb;
+	return uv_read_apic_id() >> index_msb;
 }
 
-#ifdef ZZZ		/* Needs x2apic patch */
 static void uv_send_IPI_self(int vector)
 {
 	apic_write(APIC_SELF_IPI, vector);
 }
-#endif
 
 struct genapic apic_x2apic_uv_x = {
 	.name = "UV large system",
+	.acpi_madt_oem_check = uv_acpi_madt_oem_check,
 	.int_delivery_mode = dest_Fixed,
 	.int_dest_mode = (APIC_DEST_PHYSICAL != 0),
 	.target_cpus = uv_target_cpus,
-	.vector_allocation_domain = uv_vector_allocation_domain,/* Fixme ZZZ */
+	.vector_allocation_domain = uv_vector_allocation_domain,
 	.apic_id_registered = uv_apic_id_registered,
+	.init_apic_ldr = uv_init_apic_ldr,
 	.send_IPI_all = uv_send_IPI_all,
 	.send_IPI_allbutself = uv_send_IPI_allbutself,
 	.send_IPI_mask = uv_send_IPI_mask,
-	/* ZZZ.send_IPI_self = uv_send_IPI_self, */
+	.send_IPI_self = uv_send_IPI_self,
 	.cpu_mask_to_apicid = uv_cpu_mask_to_apicid,
-	.phys_pkg_id = phys_pkg_id,	/* Fixme ZZZ */
+	.phys_pkg_id = phys_pkg_id,
+	.get_apic_id = get_apic_id,
+	.set_apic_id = set_apic_id,
+	.apic_id_mask = (0xFFFFFFFFu),
 };
 
 static __cpuinit void set_x2apic_extra_bits(int pnode)
@@ -222,12 +284,13 @@ static __init void map_low_mmrs(void)
 
 enum map_type {map_wb, map_uc};
 
-static __init void map_high(char *id, unsigned long base, int shift, enum map_type map_type)
+static __init void map_high(char *id, unsigned long base, int shift,
+			    int max_pnode, enum map_type map_type)
 {
 	unsigned long bytes, paddr;
 
 	paddr = base << shift;
-	bytes = (1UL << shift);
+	bytes = (1UL << shift) * (max_pnode + 1);
 	printk(KERN_INFO "UV: Map %s_HI 0x%lx - 0x%lx\n", id, paddr,
 	       					paddr + bytes);
 	if (map_type == map_uc)
@@ -243,7 +306,7 @@ static __init void map_gru_high(int max_pnode)
 
 	gru.v = uv_read_local_mmr(UVH_RH_GAM_GRU_OVERLAY_CONFIG_MMR);
 	if (gru.s.enable)
-		map_high("GRU", gru.s.base, shift, map_wb);
+		map_high("GRU", gru.s.base, shift, max_pnode, map_wb);
 }
 
 static __init void map_config_high(int max_pnode)
@@ -253,7 +316,7 @@ static __init void map_config_high(int max_pnode)
 
 	cfg.v = uv_read_local_mmr(UVH_RH_GAM_CFG_OVERLAY_CONFIG_MMR);
 	if (cfg.s.enable)
-		map_high("CONFIG", cfg.s.base, shift, map_uc);
+		map_high("CONFIG", cfg.s.base, shift, max_pnode, map_uc);
 }
 
 static __init void map_mmr_high(int max_pnode)
@@ -263,7 +326,7 @@ static __init void map_mmr_high(int max_pnode)
 
 	mmr.v = uv_read_local_mmr(UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR);
 	if (mmr.s.enable)
-		map_high("MMR", mmr.s.base, shift, map_uc);
+		map_high("MMR", mmr.s.base, shift, max_pnode, map_uc);
 }
 
 static __init void map_mmioh_high(int max_pnode)
@@ -273,7 +336,7 @@ static __init void map_mmioh_high(int max_pnode)
 
 	mmioh.v = uv_read_local_mmr(UVH_RH_GAM_MMIOH_OVERLAY_CONFIG_MMR);
 	if (mmioh.s.enable)
-		map_high("MMIOH", mmioh.s.base, shift, map_uc);
+		map_high("MMIOH", mmioh.s.base, shift, max_pnode, map_uc);
 }
 
 static __init void uv_rtc_init(void)
@@ -401,3 +464,5 @@ void __cpuinit uv_cpu_init(void)
 	if (get_uv_system_type() == UV_NON_UNIQUE_APIC)
 		set_x2apic_extra_bits(uv_hub_info->pnode);
 }
+
+
diff --git a/arch/x86/kernel/head.c b/arch/x86/kernel/head.c
index 3e66bd364a9d..1dcb0f13897e 100644
--- a/arch/x86/kernel/head.c
+++ b/arch/x86/kernel/head.c
@@ -35,6 +35,7 @@ void __init reserve_ebda_region(void)
 
 	/* start of EBDA area */
 	ebda_addr = get_bios_ebda();
+	printk(KERN_INFO "BIOS EBDA/lowmem at: %08x/%08x\n", ebda_addr, lowmem);
 
 	/* Fixup: bios puts an EBDA in the top 64K segment */
 	/* of conventional memory, but does not adjust lowmem. */
diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 73deaffadd03..acf62fc233da 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -115,13 +115,17 @@ static void hpet_reserve_platform_timers(unsigned long id)
 	hd.hd_phys_address = hpet_address;
 	hd.hd_address = hpet;
 	hd.hd_nirqs = nrtimers;
-	hd.hd_flags = HPET_DATA_PLATFORM;
 	hpet_reserve_timer(&hd, 0);
 
 #ifdef CONFIG_HPET_EMULATE_RTC
 	hpet_reserve_timer(&hd, 1);
 #endif
 
+	/*
+	 * NOTE that hd_irq[] reflects IOAPIC input pins (LEGACY_8254
+	 * is wrong for i8259!) not the output IRQ.  Many BIOS writers
+	 * don't bother configuring *any* comparator interrupts.
+	 */
 	hd.hd_irq[0] = HPET_LEGACY_8254;
 	hd.hd_irq[1] = HPET_LEGACY_RTC;
 
diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c
index eb9ddd8efb82..1f20608d4ca8 100644
--- a/arch/x86/kernel/i387.c
+++ b/arch/x86/kernel/i387.c
@@ -21,9 +21,12 @@
 # include <asm/sigcontext32.h>
 # include <asm/user32.h>
 #else
-# define save_i387_ia32		save_i387
-# define restore_i387_ia32	restore_i387
+# define save_i387_xstate_ia32		save_i387_xstate
+# define restore_i387_xstate_ia32	restore_i387_xstate
 # define _fpstate_ia32		_fpstate
+# define _xstate_ia32		_xstate
+# define sig_xstate_ia32_size   sig_xstate_size
+# define fx_sw_reserved_ia32	fx_sw_reserved
 # define user_i387_ia32_struct	user_i387_struct
 # define user32_fxsr_struct	user_fxsr_struct
 #endif
@@ -36,6 +39,7 @@
 
 static unsigned int		mxcsr_feature_mask __read_mostly = 0xffffffffu;
 unsigned int xstate_size;
+unsigned int sig_xstate_ia32_size = sizeof(struct _fpstate_ia32);
 static struct i387_fxsave_struct fx_scratch __cpuinitdata;
 
 void __cpuinit mxcsr_feature_mask_init(void)
@@ -61,6 +65,11 @@ void __init init_thread_xstate(void)
 		return;
 	}
 
+	if (cpu_has_xsave) {
+		xsave_cntxt_init();
+		return;
+	}
+
 	if (cpu_has_fxsr)
 		xstate_size = sizeof(struct i387_fxsave_struct);
 #ifdef CONFIG_X86_32
@@ -83,9 +92,19 @@ void __cpuinit fpu_init(void)
 
 	write_cr0(oldcr0 & ~(X86_CR0_TS|X86_CR0_EM)); /* clear TS and EM */
 
+	/*
+	 * Boot processor to setup the FP and extended state context info.
+	 */
+	if (!smp_processor_id())
+		init_thread_xstate();
+	xsave_init();
+
 	mxcsr_feature_mask_init();
 	/* clean state in init */
-	current_thread_info()->status = 0;
+	if (cpu_has_xsave)
+		current_thread_info()->status = TS_XSAVE;
+	else
+		current_thread_info()->status = 0;
 	clear_used_math();
 }
 #endif	/* CONFIG_X86_64 */
@@ -195,6 +214,13 @@ int xfpregs_set(struct task_struct *target, const struct user_regset *regset,
 	 */
 	target->thread.xstate->fxsave.mxcsr &= mxcsr_feature_mask;
 
+	/*
+	 * update the header bits in the xsave header, indicating the
+	 * presence of FP and SSE state.
+	 */
+	if (cpu_has_xsave)
+		target->thread.xstate->xsave.xsave_hdr.xstate_bv |= XSTATE_FPSSE;
+
 	return ret;
 }
 
@@ -395,6 +421,12 @@ int fpregs_set(struct task_struct *target, const struct user_regset *regset,
 	if (!ret)
 		convert_to_fxsr(target, &env);
 
+	/*
+	 * update the header bit in the xsave header, indicating the
+	 * presence of FP.
+	 */
+	if (cpu_has_xsave)
+		target->thread.xstate->xsave.xsave_hdr.xstate_bv |= XSTATE_FP;
 	return ret;
 }
 
@@ -407,7 +439,6 @@ static inline int save_i387_fsave(struct _fpstate_ia32 __user *buf)
 	struct task_struct *tsk = current;
 	struct i387_fsave_struct *fp = &tsk->thread.xstate->fsave;
 
-	unlazy_fpu(tsk);
 	fp->status = fp->swd;
 	if (__copy_to_user(buf, fp, sizeof(struct i387_fsave_struct)))
 		return -1;
@@ -421,8 +452,6 @@ static int save_i387_fxsave(struct _fpstate_ia32 __user *buf)
 	struct user_i387_ia32_struct env;
 	int err = 0;
 
-	unlazy_fpu(tsk);
-
 	convert_from_fxsr(&env, tsk);
 	if (__copy_to_user(buf, &env, sizeof(env)))
 		return -1;
@@ -432,16 +461,54 @@ static int save_i387_fxsave(struct _fpstate_ia32 __user *buf)
 	if (err)
 		return -1;
 
-	if (__copy_to_user(&buf->_fxsr_env[0], fx,
-			   sizeof(struct i387_fxsave_struct)))
+	if (__copy_to_user(&buf->_fxsr_env[0], fx, xstate_size))
+		return -1;
+	return 1;
+}
+
+static int save_i387_xsave(void __user *buf)
+{
+	struct task_struct *tsk = current;
+	struct _fpstate_ia32 __user *fx = buf;
+	int err = 0;
+
+	/*
+	 * For legacy compatible, we always set FP/SSE bits in the bit
+	 * vector while saving the state to the user context.
+	 * This will enable us capturing any changes(during sigreturn) to
+	 * the FP/SSE bits by the legacy applications which don't touch
+	 * xstate_bv in the xsave header.
+	 *
+	 * xsave aware applications can change the xstate_bv in the xsave
+	 * header as well as change any contents in the memory layout.
+	 * xrestore as part of sigreturn will capture all the changes.
+	 */
+	tsk->thread.xstate->xsave.xsave_hdr.xstate_bv |= XSTATE_FPSSE;
+
+	if (save_i387_fxsave(fx) < 0)
+		return -1;
+
+	err = __copy_to_user(&fx->sw_reserved, &fx_sw_reserved_ia32,
+			     sizeof(struct _fpx_sw_bytes));
+	err |= __put_user(FP_XSTATE_MAGIC2,
+			  (__u32 __user *) (buf + sig_xstate_ia32_size
+					    - FP_XSTATE_MAGIC2_SIZE));
+	if (err)
 		return -1;
+
 	return 1;
 }
 
-int save_i387_ia32(struct _fpstate_ia32 __user *buf)
+int save_i387_xstate_ia32(void __user *buf)
 {
+	struct _fpstate_ia32 __user *fp = (struct _fpstate_ia32 __user *) buf;
+	struct task_struct *tsk = current;
+
 	if (!used_math())
 		return 0;
+
+	if (!access_ok(VERIFY_WRITE, buf, sig_xstate_ia32_size))
+		return -EACCES;
 	/*
 	 * This will cause a "finit" to be triggered by the next
 	 * attempted FPU operation by the 'current' process.
@@ -451,13 +518,17 @@ int save_i387_ia32(struct _fpstate_ia32 __user *buf)
 	if (!HAVE_HWFP) {
 		return fpregs_soft_get(current, NULL,
 				       0, sizeof(struct user_i387_ia32_struct),
-				       NULL, buf) ? -1 : 1;
+				       NULL, fp) ? -1 : 1;
 	}
 
+	unlazy_fpu(tsk);
+
+	if (cpu_has_xsave)
+		return save_i387_xsave(fp);
 	if (cpu_has_fxsr)
-		return save_i387_fxsave(buf);
+		return save_i387_fxsave(fp);
 	else
-		return save_i387_fsave(buf);
+		return save_i387_fsave(fp);
 }
 
 static inline int restore_i387_fsave(struct _fpstate_ia32 __user *buf)
@@ -468,14 +539,15 @@ static inline int restore_i387_fsave(struct _fpstate_ia32 __user *buf)
 				sizeof(struct i387_fsave_struct));
 }
 
-static int restore_i387_fxsave(struct _fpstate_ia32 __user *buf)
+static int restore_i387_fxsave(struct _fpstate_ia32 __user *buf,
+			       unsigned int size)
 {
 	struct task_struct *tsk = current;
 	struct user_i387_ia32_struct env;
 	int err;
 
 	err = __copy_from_user(&tsk->thread.xstate->fxsave, &buf->_fxsr_env[0],
-			       sizeof(struct i387_fxsave_struct));
+			       size);
 	/* mxcsr reserved bits must be masked to zero for security reasons */
 	tsk->thread.xstate->fxsave.mxcsr &= mxcsr_feature_mask;
 	if (err || __copy_from_user(&env, buf, sizeof(env)))
@@ -485,14 +557,69 @@ static int restore_i387_fxsave(struct _fpstate_ia32 __user *buf)
 	return 0;
 }
 
-int restore_i387_ia32(struct _fpstate_ia32 __user *buf)
+static int restore_i387_xsave(void __user *buf)
+{
+	struct _fpx_sw_bytes fx_sw_user;
+	struct _fpstate_ia32 __user *fx_user =
+			((struct _fpstate_ia32 __user *) buf);
+	struct i387_fxsave_struct __user *fx =
+		(struct i387_fxsave_struct __user *) &fx_user->_fxsr_env[0];
+	struct xsave_hdr_struct *xsave_hdr =
+				&current->thread.xstate->xsave.xsave_hdr;
+	u64 mask;
+	int err;
+
+	if (check_for_xstate(fx, buf, &fx_sw_user))
+		goto fx_only;
+
+	mask = fx_sw_user.xstate_bv;
+
+	err = restore_i387_fxsave(buf, fx_sw_user.xstate_size);
+
+	xsave_hdr->xstate_bv &= pcntxt_mask;
+	/*
+	 * These bits must be zero.
+	 */
+	xsave_hdr->reserved1[0] = xsave_hdr->reserved1[1] = 0;
+
+	/*
+	 * Init the state that is not present in the memory layout
+	 * and enabled by the OS.
+	 */
+	mask = ~(pcntxt_mask & ~mask);
+	xsave_hdr->xstate_bv &= mask;
+
+	return err;
+fx_only:
+	/*
+	 * Couldn't find the extended state information in the memory
+	 * layout. Restore the FP/SSE and init the other extended state
+	 * enabled by the OS.
+	 */
+	xsave_hdr->xstate_bv = XSTATE_FPSSE;
+	return restore_i387_fxsave(buf, sizeof(struct i387_fxsave_struct));
+}
+
+int restore_i387_xstate_ia32(void __user *buf)
 {
 	int err;
 	struct task_struct *tsk = current;
+	struct _fpstate_ia32 __user *fp = (struct _fpstate_ia32 __user *) buf;
 
 	if (HAVE_HWFP)
 		clear_fpu(tsk);
 
+	if (!buf) {
+		if (used_math()) {
+			clear_fpu(tsk);
+			clear_used_math();
+		}
+
+		return 0;
+	} else
+		if (!access_ok(VERIFY_READ, buf, sig_xstate_ia32_size))
+			return -EACCES;
+
 	if (!used_math()) {
 		err = init_fpu(tsk);
 		if (err)
@@ -500,14 +627,17 @@ int restore_i387_ia32(struct _fpstate_ia32 __user *buf)
 	}
 
 	if (HAVE_HWFP) {
-		if (cpu_has_fxsr)
-			err = restore_i387_fxsave(buf);
+		if (cpu_has_xsave)
+			err = restore_i387_xsave(buf);
+		else if (cpu_has_fxsr)
+			err = restore_i387_fxsave(fp, sizeof(struct
+							   i387_fxsave_struct));
 		else
-			err = restore_i387_fsave(buf);
+			err = restore_i387_fsave(fp);
 	} else {
 		err = fpregs_soft_set(current, NULL,
 				      0, sizeof(struct user_i387_ia32_struct),
-				      NULL, buf) != 0;
+				      NULL, fp) != 0;
 	}
 	set_used_math();
 
diff --git a/arch/x86/kernel/i8259.c b/arch/x86/kernel/i8259.c
index dc92b49d9204..4b8a53d841f7 100644
--- a/arch/x86/kernel/i8259.c
+++ b/arch/x86/kernel/i8259.c
@@ -282,6 +282,30 @@ static int __init i8259A_init_sysfs(void)
 
 device_initcall(i8259A_init_sysfs);
 
+void mask_8259A(void)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&i8259A_lock, flags);
+
+	outb(0xff, PIC_MASTER_IMR);	/* mask all of 8259A-1 */
+	outb(0xff, PIC_SLAVE_IMR);	/* mask all of 8259A-2 */
+
+	spin_unlock_irqrestore(&i8259A_lock, flags);
+}
+
+void unmask_8259A(void)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&i8259A_lock, flags);
+
+	outb(cached_master_mask, PIC_MASTER_IMR); /* restore master IRQ mask */
+	outb(cached_slave_mask, PIC_SLAVE_IMR);	  /* restore slave IRQ mask */
+
+	spin_unlock_irqrestore(&i8259A_lock, flags);
+}
+
 void init_8259A(int auto_eoi)
 {
 	unsigned long flags;
diff --git a/arch/x86/kernel/io_apic_32.c b/arch/x86/kernel/io_apic_32.c
index 09cddb57bec4..e710289f673e 100644
--- a/arch/x86/kernel/io_apic_32.c
+++ b/arch/x86/kernel/io_apic_32.c
@@ -46,10 +46,13 @@
 #include <asm/nmi.h>
 #include <asm/msidef.h>
 #include <asm/hypertransport.h>
+#include <asm/setup.h>
 
 #include <mach_apic.h>
 #include <mach_apicdef.h>
 
+#define __apicdebuginit(type) static type __init
+
 int (*ioapic_renumber_irq)(int ioapic, int irq);
 atomic_t irq_mis_count;
 
@@ -1341,7 +1344,8 @@ static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin,
 	ioapic_write_entry(apic, pin, entry);
 }
 
-void __init print_IO_APIC(void)
+
+__apicdebuginit(void) print_IO_APIC(void)
 {
 	int apic, i;
 	union IO_APIC_reg_00 reg_00;
@@ -1456,9 +1460,7 @@ void __init print_IO_APIC(void)
 	return;
 }
 
-#if 0
-
-static void print_APIC_bitfield(int base)
+__apicdebuginit(void) print_APIC_bitfield(int base)
 {
 	unsigned int v;
 	int i, j;
@@ -1479,9 +1481,10 @@ static void print_APIC_bitfield(int base)
 	}
 }
 
-void /*__init*/ print_local_APIC(void *dummy)
+__apicdebuginit(void) print_local_APIC(void *dummy)
 {
 	unsigned int v, ver, maxlvt;
+	u64 icr;
 
 	if (apic_verbosity == APIC_QUIET)
 		return;
@@ -1490,7 +1493,7 @@ void /*__init*/ print_local_APIC(void *dummy)
 		smp_processor_id(), hard_smp_processor_id());
 	v = apic_read(APIC_ID);
 	printk(KERN_INFO "... APIC ID:      %08x (%01x)\n", v,
-			GET_APIC_ID(read_apic_id()));
+			GET_APIC_ID(v));
 	v = apic_read(APIC_LVR);
 	printk(KERN_INFO "... APIC VERSION: %08x\n", v);
 	ver = GET_APIC_VERSION(v);
@@ -1532,10 +1535,9 @@ void /*__init*/ print_local_APIC(void *dummy)
 		printk(KERN_DEBUG "... APIC ESR: %08x\n", v);
 	}
 
-	v = apic_read(APIC_ICR);
-	printk(KERN_DEBUG "... APIC ICR: %08x\n", v);
-	v = apic_read(APIC_ICR2);
-	printk(KERN_DEBUG "... APIC ICR2: %08x\n", v);
+	icr = apic_icr_read();
+	printk(KERN_DEBUG "... APIC ICR: %08x\n", icr);
+	printk(KERN_DEBUG "... APIC ICR2: %08x\n", icr >> 32);
 
 	v = apic_read(APIC_LVTT);
 	printk(KERN_DEBUG "... APIC LVTT: %08x\n", v);
@@ -1563,12 +1565,12 @@ void /*__init*/ print_local_APIC(void *dummy)
 	printk("\n");
 }
 
-void print_all_local_APICs(void)
+__apicdebuginit(void) print_all_local_APICs(void)
 {
 	on_each_cpu(print_local_APIC, NULL, 1);
 }
 
-void /*__init*/ print_PIC(void)
+__apicdebuginit(void) print_PIC(void)
 {
 	unsigned int v;
 	unsigned long flags;
@@ -1600,7 +1602,17 @@ void /*__init*/ print_PIC(void)
 	printk(KERN_DEBUG "... PIC ELCR: %04x\n", v);
 }
 
-#endif  /*  0  */
+__apicdebuginit(int) print_all_ICs(void)
+{
+	print_PIC();
+	print_all_local_APICs();
+	print_IO_APIC();
+
+	return 0;
+}
+
+fs_initcall(print_all_ICs);
+
 
 static void __init enable_IO_APIC(void)
 {
@@ -1698,8 +1710,7 @@ void disable_IO_APIC(void)
 		entry.dest_mode       = 0; /* Physical */
 		entry.delivery_mode   = dest_ExtINT; /* ExtInt */
 		entry.vector          = 0;
-		entry.dest.physical.physical_dest =
-					GET_APIC_ID(read_apic_id());
+		entry.dest.physical.physical_dest = read_apic_id();
 
 		/*
 		 * Add it to the IO-APIC irq-routing table:
@@ -1725,10 +1736,8 @@ static void __init setup_ioapic_ids_from_mpc(void)
 	unsigned char old_id;
 	unsigned long flags;
 
-#ifdef CONFIG_X86_NUMAQ
-	if (found_numaq)
+	if (x86_quirks->setup_ioapic_ids && x86_quirks->setup_ioapic_ids())
 		return;
-#endif
 
 	/*
 	 * Don't check I/O APIC IDs for xAPIC systems.  They have
@@ -2329,8 +2338,6 @@ void __init setup_IO_APIC(void)
 	setup_IO_APIC_irqs();
 	init_IO_APIC_traps();
 	check_timer();
-	if (!acpi_ioapic)
-		print_IO_APIC();
 }
 
 /*
diff --git a/arch/x86/kernel/io_apic_64.c b/arch/x86/kernel/io_apic_64.c
index 61a83b70c18f..02063ae042f7 100644
--- a/arch/x86/kernel/io_apic_64.c
+++ b/arch/x86/kernel/io_apic_64.c
@@ -37,6 +37,7 @@
 #include <acpi/acpi_bus.h>
 #endif
 #include <linux/bootmem.h>
+#include <linux/dmar.h>
 
 #include <asm/idle.h>
 #include <asm/io.h>
@@ -49,10 +50,13 @@
 #include <asm/nmi.h>
 #include <asm/msidef.h>
 #include <asm/hypertransport.h>
+#include <asm/irq_remapping.h>
 
 #include <mach_ipi.h>
 #include <mach_apic.h>
 
+#define __apicdebuginit(type) static type __init
+
 struct irq_cfg {
 	cpumask_t domain;
 	cpumask_t old_domain;
@@ -87,8 +91,6 @@ int first_system_vector = 0xfe;
 
 char system_vectors[NR_VECTORS] = { [0 ... NR_VECTORS-1] = SYS_VECTOR_FREE};
 
-#define __apicdebuginit  __init
-
 int sis_apic_bug; /* not actually supported, dummy for compile */
 
 static int no_timer_check;
@@ -108,6 +110,9 @@ static DEFINE_SPINLOCK(vector_lock);
  */
 int nr_ioapic_registers[MAX_IO_APICS];
 
+/* I/O APIC RTE contents at the OS boot up */
+struct IO_APIC_route_entry *early_ioapic_entries[MAX_IO_APICS];
+
 /* I/O APIC entries */
 struct mp_config_ioapic mp_ioapics[MAX_IO_APICS];
 int nr_ioapics;
@@ -303,7 +308,12 @@ static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, u8 vector)
 		pin = entry->pin;
 		if (pin == -1)
 			break;
-		io_apic_write(apic, 0x11 + pin*2, dest);
+		/*
+		 * With interrupt-remapping, destination information comes
+		 * from interrupt-remapping table entry.
+		 */
+		if (!irq_remapped(irq))
+			io_apic_write(apic, 0x11 + pin*2, dest);
 		reg = io_apic_read(apic, 0x10 + pin*2);
 		reg &= ~IO_APIC_REDIR_VECTOR_MASK;
 		reg |= vector;
@@ -440,6 +450,69 @@ static void clear_IO_APIC (void)
 			clear_IO_APIC_pin(apic, pin);
 }
 
+/*
+ * Saves and masks all the unmasked IO-APIC RTE's
+ */
+int save_mask_IO_APIC_setup(void)
+{
+	union IO_APIC_reg_01 reg_01;
+	unsigned long flags;
+	int apic, pin;
+
+	/*
+	 * The number of IO-APIC IRQ registers (== #pins):
+	 */
+	for (apic = 0; apic < nr_ioapics; apic++) {
+		spin_lock_irqsave(&ioapic_lock, flags);
+		reg_01.raw = io_apic_read(apic, 1);
+		spin_unlock_irqrestore(&ioapic_lock, flags);
+		nr_ioapic_registers[apic] = reg_01.bits.entries+1;
+	}
+
+	for (apic = 0; apic < nr_ioapics; apic++) {
+		early_ioapic_entries[apic] =
+			kzalloc(sizeof(struct IO_APIC_route_entry) *
+				nr_ioapic_registers[apic], GFP_KERNEL);
+		if (!early_ioapic_entries[apic])
+			return -ENOMEM;
+	}
+
+	for (apic = 0; apic < nr_ioapics; apic++)
+		for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
+			struct IO_APIC_route_entry entry;
+
+			entry = early_ioapic_entries[apic][pin] =
+				ioapic_read_entry(apic, pin);
+			if (!entry.mask) {
+				entry.mask = 1;
+				ioapic_write_entry(apic, pin, entry);
+			}
+		}
+	return 0;
+}
+
+void restore_IO_APIC_setup(void)
+{
+	int apic, pin;
+
+	for (apic = 0; apic < nr_ioapics; apic++)
+		for (pin = 0; pin < nr_ioapic_registers[apic]; pin++)
+			ioapic_write_entry(apic, pin,
+					   early_ioapic_entries[apic][pin]);
+}
+
+void reinit_intr_remapped_IO_APIC(int intr_remapping)
+{
+	/*
+	 * for now plain restore of previous settings.
+	 * TBD: In the case of OS enabling interrupt-remapping,
+	 * IO-APIC RTE's need to be setup to point to interrupt-remapping
+	 * table entries. for now, do a plain restore, and wait for
+	 * the setup_IO_APIC_irqs() to do proper initialization.
+	 */
+	restore_IO_APIC_setup();
+}
+
 int skip_ioapic_setup;
 int ioapic_force;
 
@@ -839,18 +912,98 @@ void __setup_vector_irq(int cpu)
 }
 
 static struct irq_chip ioapic_chip;
+#ifdef CONFIG_INTR_REMAP
+static struct irq_chip ir_ioapic_chip;
+#endif
 
 static void ioapic_register_intr(int irq, unsigned long trigger)
 {
-	if (trigger) {
+	if (trigger)
 		irq_desc[irq].status |= IRQ_LEVEL;
-		set_irq_chip_and_handler_name(irq, &ioapic_chip,
-					      handle_fasteoi_irq, "fasteoi");
-	} else {
+	else
 		irq_desc[irq].status &= ~IRQ_LEVEL;
+
+#ifdef CONFIG_INTR_REMAP
+	if (irq_remapped(irq)) {
+		irq_desc[irq].status |= IRQ_MOVE_PCNTXT;
+		if (trigger)
+			set_irq_chip_and_handler_name(irq, &ir_ioapic_chip,
+						      handle_fasteoi_irq,
+						     "fasteoi");
+		else
+			set_irq_chip_and_handler_name(irq, &ir_ioapic_chip,
+						      handle_edge_irq, "edge");
+		return;
+	}
+#endif
+	if (trigger)
+		set_irq_chip_and_handler_name(irq, &ioapic_chip,
+					      handle_fasteoi_irq,
+					      "fasteoi");
+	else
 		set_irq_chip_and_handler_name(irq, &ioapic_chip,
 					      handle_edge_irq, "edge");
+}
+
+static int setup_ioapic_entry(int apic, int irq,
+			      struct IO_APIC_route_entry *entry,
+			      unsigned int destination, int trigger,
+			      int polarity, int vector)
+{
+	/*
+	 * add it to the IO-APIC irq-routing table:
+	 */
+	memset(entry,0,sizeof(*entry));
+
+#ifdef CONFIG_INTR_REMAP
+	if (intr_remapping_enabled) {
+		struct intel_iommu *iommu = map_ioapic_to_ir(apic);
+		struct irte irte;
+		struct IR_IO_APIC_route_entry *ir_entry =
+			(struct IR_IO_APIC_route_entry *) entry;
+		int index;
+
+		if (!iommu)
+			panic("No mapping iommu for ioapic %d\n", apic);
+
+		index = alloc_irte(iommu, irq, 1);
+		if (index < 0)
+			panic("Failed to allocate IRTE for ioapic %d\n", apic);
+
+		memset(&irte, 0, sizeof(irte));
+
+		irte.present = 1;
+		irte.dst_mode = INT_DEST_MODE;
+		irte.trigger_mode = trigger;
+		irte.dlvry_mode = INT_DELIVERY_MODE;
+		irte.vector = vector;
+		irte.dest_id = IRTE_DEST(destination);
+
+		modify_irte(irq, &irte);
+
+		ir_entry->index2 = (index >> 15) & 0x1;
+		ir_entry->zero = 0;
+		ir_entry->format = 1;
+		ir_entry->index = (index & 0x7fff);
+	} else
+#endif
+	{
+		entry->delivery_mode = INT_DELIVERY_MODE;
+		entry->dest_mode = INT_DEST_MODE;
+		entry->dest = destination;
 	}
+
+	entry->mask = 0;				/* enable IRQ */
+	entry->trigger = trigger;
+	entry->polarity = polarity;
+	entry->vector = vector;
+
+	/* Mask level triggered irqs.
+	 * Use IRQ_DELAYED_DISABLE for edge triggered irqs.
+	 */
+	if (trigger)
+		entry->mask = 1;
+	return 0;
 }
 
 static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq,
@@ -875,24 +1028,15 @@ static void setup_IO_APIC_irq(int apic, int pin, unsigned int irq,
 		    apic, mp_ioapics[apic].mp_apicid, pin, cfg->vector,
 		    irq, trigger, polarity);
 
-	/*
-	 * add it to the IO-APIC irq-routing table:
-	 */
-	memset(&entry,0,sizeof(entry));
-
-	entry.delivery_mode = INT_DELIVERY_MODE;
-	entry.dest_mode = INT_DEST_MODE;
-	entry.dest = cpu_mask_to_apicid(mask);
-	entry.mask = 0;				/* enable IRQ */
-	entry.trigger = trigger;
-	entry.polarity = polarity;
-	entry.vector = cfg->vector;
 
-	/* Mask level triggered irqs.
-	 * Use IRQ_DELAYED_DISABLE for edge triggered irqs.
-	 */
-	if (trigger)
-		entry.mask = 1;
+	if (setup_ioapic_entry(mp_ioapics[apic].mp_apicid, irq, &entry,
+			       cpu_mask_to_apicid(mask), trigger, polarity,
+			       cfg->vector)) {
+		printk("Failed to setup ioapic entry for ioapic  %d, pin %d\n",
+		       mp_ioapics[apic].mp_apicid, pin);
+		__clear_irq_vector(irq);
+		return;
+	}
 
 	ioapic_register_intr(irq, trigger);
 	if (irq < 16)
@@ -944,6 +1088,9 @@ static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin,
 {
 	struct IO_APIC_route_entry entry;
 
+	if (intr_remapping_enabled)
+		return;
+
 	memset(&entry, 0, sizeof(entry));
 
 	/*
@@ -970,7 +1117,8 @@ static void __init setup_timer_IRQ0_pin(unsigned int apic, unsigned int pin,
 	ioapic_write_entry(apic, pin, entry);
 }
 
-void __apicdebuginit print_IO_APIC(void)
+
+__apicdebuginit(void) print_IO_APIC(void)
 {
 	int apic, i;
 	union IO_APIC_reg_00 reg_00;
@@ -1064,9 +1212,7 @@ void __apicdebuginit print_IO_APIC(void)
 	return;
 }
 
-#if 0
-
-static __apicdebuginit void print_APIC_bitfield (int base)
+__apicdebuginit(void) print_APIC_bitfield(int base)
 {
 	unsigned int v;
 	int i, j;
@@ -1087,9 +1233,10 @@ static __apicdebuginit void print_APIC_bitfield (int base)
 	}
 }
 
-void __apicdebuginit print_local_APIC(void * dummy)
+__apicdebuginit(void) print_local_APIC(void *dummy)
 {
 	unsigned int v, ver, maxlvt;
+	unsigned long icr;
 
 	if (apic_verbosity == APIC_QUIET)
 		return;
@@ -1097,7 +1244,7 @@ void __apicdebuginit print_local_APIC(void * dummy)
 	printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n",
 		smp_processor_id(), hard_smp_processor_id());
 	v = apic_read(APIC_ID);
-	printk(KERN_INFO "... APIC ID:      %08x (%01x)\n", v, GET_APIC_ID(read_apic_id()));
+	printk(KERN_INFO "... APIC ID:      %08x (%01x)\n", v, read_apic_id());
 	v = apic_read(APIC_LVR);
 	printk(KERN_INFO "... APIC VERSION: %08x\n", v);
 	ver = GET_APIC_VERSION(v);
@@ -1133,10 +1280,9 @@ void __apicdebuginit print_local_APIC(void * dummy)
 	v = apic_read(APIC_ESR);
 	printk(KERN_DEBUG "... APIC ESR: %08x\n", v);
 
-	v = apic_read(APIC_ICR);
-	printk(KERN_DEBUG "... APIC ICR: %08x\n", v);
-	v = apic_read(APIC_ICR2);
-	printk(KERN_DEBUG "... APIC ICR2: %08x\n", v);
+	icr = apic_icr_read();
+	printk(KERN_DEBUG "... APIC ICR: %08x\n", (u32)icr);
+	printk(KERN_DEBUG "... APIC ICR2: %08x\n", (u32)(icr >> 32));
 
 	v = apic_read(APIC_LVTT);
 	printk(KERN_DEBUG "... APIC LVTT: %08x\n", v);
@@ -1164,12 +1310,12 @@ void __apicdebuginit print_local_APIC(void * dummy)
 	printk("\n");
 }
 
-void print_all_local_APICs (void)
+__apicdebuginit(void) print_all_local_APICs(void)
 {
 	on_each_cpu(print_local_APIC, NULL, 1);
 }
 
-void __apicdebuginit print_PIC(void)
+__apicdebuginit(void) print_PIC(void)
 {
 	unsigned int v;
 	unsigned long flags;
@@ -1201,7 +1347,17 @@ void __apicdebuginit print_PIC(void)
 	printk(KERN_DEBUG "... PIC ELCR: %04x\n", v);
 }
 
-#endif  /*  0  */
+__apicdebuginit(int) print_all_ICs(void)
+{
+	print_PIC();
+	print_all_local_APICs();
+	print_IO_APIC();
+
+	return 0;
+}
+
+fs_initcall(print_all_ICs);
+
 
 void __init enable_IO_APIC(void)
 {
@@ -1291,7 +1447,7 @@ void disable_IO_APIC(void)
 		entry.dest_mode       = 0; /* Physical */
 		entry.delivery_mode   = dest_ExtINT; /* ExtInt */
 		entry.vector          = 0;
-		entry.dest          = GET_APIC_ID(read_apic_id());
+		entry.dest            = read_apic_id();
 
 		/*
 		 * Add it to the IO-APIC irq-routing table:
@@ -1397,6 +1553,147 @@ static int ioapic_retrigger_irq(unsigned int irq)
  */
 
 #ifdef CONFIG_SMP
+
+#ifdef CONFIG_INTR_REMAP
+static void ir_irq_migration(struct work_struct *work);
+
+static DECLARE_DELAYED_WORK(ir_migration_work, ir_irq_migration);
+
+/*
+ * Migrate the IO-APIC irq in the presence of intr-remapping.
+ *
+ * For edge triggered, irq migration is a simple atomic update(of vector
+ * and cpu destination) of IRTE and flush the hardware cache.
+ *
+ * For level triggered, we need to modify the io-apic RTE aswell with the update
+ * vector information, along with modifying IRTE with vector and destination.
+ * So irq migration for level triggered is little  bit more complex compared to
+ * edge triggered migration. But the good news is, we use the same algorithm
+ * for level triggered migration as we have today, only difference being,
+ * we now initiate the irq migration from process context instead of the
+ * interrupt context.
+ *
+ * In future, when we do a directed EOI (combined with cpu EOI broadcast
+ * suppression) to the IO-APIC, level triggered irq migration will also be
+ * as simple as edge triggered migration and we can do the irq migration
+ * with a simple atomic update to IO-APIC RTE.
+ */
+static void migrate_ioapic_irq(int irq, cpumask_t mask)
+{
+	struct irq_cfg *cfg = irq_cfg + irq;
+	struct irq_desc *desc = irq_desc + irq;
+	cpumask_t tmp, cleanup_mask;
+	struct irte irte;
+	int modify_ioapic_rte = desc->status & IRQ_LEVEL;
+	unsigned int dest;
+	unsigned long flags;
+
+	cpus_and(tmp, mask, cpu_online_map);
+	if (cpus_empty(tmp))
+		return;
+
+	if (get_irte(irq, &irte))
+		return;
+
+	if (assign_irq_vector(irq, mask))
+		return;
+
+	cpus_and(tmp, cfg->domain, mask);
+	dest = cpu_mask_to_apicid(tmp);
+
+	if (modify_ioapic_rte) {
+		spin_lock_irqsave(&ioapic_lock, flags);
+		__target_IO_APIC_irq(irq, dest, cfg->vector);
+		spin_unlock_irqrestore(&ioapic_lock, flags);
+	}
+
+	irte.vector = cfg->vector;
+	irte.dest_id = IRTE_DEST(dest);
+
+	/*
+	 * Modified the IRTE and flushes the Interrupt entry cache.
+	 */
+	modify_irte(irq, &irte);
+
+	if (cfg->move_in_progress) {
+		cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map);
+		cfg->move_cleanup_count = cpus_weight(cleanup_mask);
+		send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
+		cfg->move_in_progress = 0;
+	}
+
+	irq_desc[irq].affinity = mask;
+}
+
+static int migrate_irq_remapped_level(int irq)
+{
+	int ret = -1;
+
+	mask_IO_APIC_irq(irq);
+
+	if (io_apic_level_ack_pending(irq)) {
+		/*
+	 	 * Interrupt in progress. Migrating irq now will change the
+		 * vector information in the IO-APIC RTE and that will confuse
+		 * the EOI broadcast performed by cpu.
+		 * So, delay the irq migration to the next instance.
+		 */
+		schedule_delayed_work(&ir_migration_work, 1);
+		goto unmask;
+	}
+
+	/* everthing is clear. we have right of way */
+	migrate_ioapic_irq(irq, irq_desc[irq].pending_mask);
+
+	ret = 0;
+	irq_desc[irq].status &= ~IRQ_MOVE_PENDING;
+	cpus_clear(irq_desc[irq].pending_mask);
+
+unmask:
+	unmask_IO_APIC_irq(irq);
+	return ret;
+}
+
+static void ir_irq_migration(struct work_struct *work)
+{
+	int irq;
+
+	for (irq = 0; irq < NR_IRQS; irq++) {
+		struct irq_desc *desc = irq_desc + irq;
+		if (desc->status & IRQ_MOVE_PENDING) {
+			unsigned long flags;
+
+			spin_lock_irqsave(&desc->lock, flags);
+			if (!desc->chip->set_affinity ||
+			    !(desc->status & IRQ_MOVE_PENDING)) {
+				desc->status &= ~IRQ_MOVE_PENDING;
+				spin_unlock_irqrestore(&desc->lock, flags);
+				continue;
+			}
+
+			desc->chip->set_affinity(irq,
+					         irq_desc[irq].pending_mask);
+			spin_unlock_irqrestore(&desc->lock, flags);
+		}
+	}
+}
+
+/*
+ * Migrates the IRQ destination in the process context.
+ */
+static void set_ir_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
+{
+	if (irq_desc[irq].status & IRQ_LEVEL) {
+		irq_desc[irq].status |= IRQ_MOVE_PENDING;
+		irq_desc[irq].pending_mask = mask;
+		migrate_irq_remapped_level(irq);
+		return;
+	}
+
+	migrate_ioapic_irq(irq, mask);
+}
+#endif
+
 asmlinkage void smp_irq_move_cleanup_interrupt(void)
 {
 	unsigned vector, me;
@@ -1453,6 +1750,17 @@ static void irq_complete_move(unsigned int irq)
 #else
 static inline void irq_complete_move(unsigned int irq) {}
 #endif
+#ifdef CONFIG_INTR_REMAP
+static void ack_x2apic_level(unsigned int irq)
+{
+	ack_x2APIC_irq();
+}
+
+static void ack_x2apic_edge(unsigned int irq)
+{
+	ack_x2APIC_irq();
+}
+#endif
 
 static void ack_apic_edge(unsigned int irq)
 {
@@ -1527,6 +1835,21 @@ static struct irq_chip ioapic_chip __read_mostly = {
 	.retrigger	= ioapic_retrigger_irq,
 };
 
+#ifdef CONFIG_INTR_REMAP
+static struct irq_chip ir_ioapic_chip __read_mostly = {
+	.name 		= "IR-IO-APIC",
+	.startup 	= startup_ioapic_irq,
+	.mask	 	= mask_IO_APIC_irq,
+	.unmask	 	= unmask_IO_APIC_irq,
+	.ack 		= ack_x2apic_edge,
+	.eoi 		= ack_x2apic_level,
+#ifdef CONFIG_SMP
+	.set_affinity 	= set_ir_ioapic_affinity_irq,
+#endif
+	.retrigger	= ioapic_retrigger_irq,
+};
+#endif
+
 static inline void init_IO_APIC_traps(void)
 {
 	int irq;
@@ -1712,6 +2035,8 @@ static inline void __init check_timer(void)
 	 * 8259A.
 	 */
 	if (pin1 == -1) {
+		if (intr_remapping_enabled)
+			panic("BIOS bug: timer not connected to IO-APIC");
 		pin1 = pin2;
 		apic1 = apic2;
 		no_pin1 = 1;
@@ -1738,6 +2063,8 @@ static inline void __init check_timer(void)
 				clear_IO_APIC_pin(0, pin1);
 			goto out;
 		}
+		if (intr_remapping_enabled)
+			panic("timer doesn't work through Interrupt-remapped IO-APIC");
 		clear_IO_APIC_pin(apic1, pin1);
 		if (!no_pin1)
 			apic_printk(APIC_QUIET, KERN_ERR "..MP-BIOS bug: "
@@ -1854,8 +2181,6 @@ void __init setup_IO_APIC(void)
 	setup_IO_APIC_irqs();
 	init_IO_APIC_traps();
 	check_timer();
-	if (!acpi_ioapic)
-		print_IO_APIC();
 }
 
 struct sysfs_ioapic_data {
@@ -1977,6 +2302,9 @@ void destroy_irq(unsigned int irq)
 
 	dynamic_irq_cleanup(irq);
 
+#ifdef CONFIG_INTR_REMAP
+	free_irte(irq);
+#endif
 	spin_lock_irqsave(&vector_lock, flags);
 	__clear_irq_vector(irq);
 	spin_unlock_irqrestore(&vector_lock, flags);
@@ -1995,11 +2323,42 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms
 
 	tmp = TARGET_CPUS;
 	err = assign_irq_vector(irq, tmp);
-	if (!err) {
-		cpus_and(tmp, cfg->domain, tmp);
-		dest = cpu_mask_to_apicid(tmp);
+	if (err)
+		return err;
+
+	cpus_and(tmp, cfg->domain, tmp);
+	dest = cpu_mask_to_apicid(tmp);
+
+#ifdef CONFIG_INTR_REMAP
+	if (irq_remapped(irq)) {
+		struct irte irte;
+		int ir_index;
+		u16 sub_handle;
+
+		ir_index = map_irq_to_irte_handle(irq, &sub_handle);
+		BUG_ON(ir_index == -1);
+
+		memset (&irte, 0, sizeof(irte));
+
+		irte.present = 1;
+		irte.dst_mode = INT_DEST_MODE;
+		irte.trigger_mode = 0; /* edge */
+		irte.dlvry_mode = INT_DELIVERY_MODE;
+		irte.vector = cfg->vector;
+		irte.dest_id = IRTE_DEST(dest);
+
+		modify_irte(irq, &irte);
 
 		msg->address_hi = MSI_ADDR_BASE_HI;
+		msg->data = sub_handle;
+		msg->address_lo = MSI_ADDR_BASE_LO | MSI_ADDR_IR_EXT_INT |
+				  MSI_ADDR_IR_SHV |
+				  MSI_ADDR_IR_INDEX1(ir_index) |
+				  MSI_ADDR_IR_INDEX2(ir_index);
+	} else
+#endif
+	{
+		msg->address_hi = MSI_ADDR_BASE_HI;
 		msg->address_lo =
 			MSI_ADDR_BASE_LO |
 			((INT_DEST_MODE == 0) ?
@@ -2049,6 +2408,55 @@ static void set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
 	write_msi_msg(irq, &msg);
 	irq_desc[irq].affinity = mask;
 }
+
+#ifdef CONFIG_INTR_REMAP
+/*
+ * Migrate the MSI irq to another cpumask. This migration is
+ * done in the process context using interrupt-remapping hardware.
+ */
+static void ir_set_msi_irq_affinity(unsigned int irq, cpumask_t mask)
+{
+	struct irq_cfg *cfg = irq_cfg + irq;
+	unsigned int dest;
+	cpumask_t tmp, cleanup_mask;
+	struct irte irte;
+
+	cpus_and(tmp, mask, cpu_online_map);
+	if (cpus_empty(tmp))
+		return;
+
+	if (get_irte(irq, &irte))
+		return;
+
+	if (assign_irq_vector(irq, mask))
+		return;
+
+	cpus_and(tmp, cfg->domain, mask);
+	dest = cpu_mask_to_apicid(tmp);
+
+	irte.vector = cfg->vector;
+	irte.dest_id = IRTE_DEST(dest);
+
+	/*
+	 * atomically update the IRTE with the new destination and vector.
+	 */
+	modify_irte(irq, &irte);
+
+	/*
+	 * After this point, all the interrupts will start arriving
+	 * at the new destination. So, time to cleanup the previous
+	 * vector allocation.
+	 */
+	if (cfg->move_in_progress) {
+		cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map);
+		cfg->move_cleanup_count = cpus_weight(cleanup_mask);
+		send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
+		cfg->move_in_progress = 0;
+	}
+
+	irq_desc[irq].affinity = mask;
+}
+#endif
 #endif /* CONFIG_SMP */
 
 /*
@@ -2066,26 +2474,157 @@ static struct irq_chip msi_chip = {
 	.retrigger	= ioapic_retrigger_irq,
 };
 
-int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc)
+#ifdef CONFIG_INTR_REMAP
+static struct irq_chip msi_ir_chip = {
+	.name		= "IR-PCI-MSI",
+	.unmask		= unmask_msi_irq,
+	.mask		= mask_msi_irq,
+	.ack		= ack_x2apic_edge,
+#ifdef CONFIG_SMP
+	.set_affinity	= ir_set_msi_irq_affinity,
+#endif
+	.retrigger	= ioapic_retrigger_irq,
+};
+
+/*
+ * Map the PCI dev to the corresponding remapping hardware unit
+ * and allocate 'nvec' consecutive interrupt-remapping table entries
+ * in it.
+ */
+static int msi_alloc_irte(struct pci_dev *dev, int irq, int nvec)
 {
+	struct intel_iommu *iommu;
+	int index;
+
+	iommu = map_dev_to_ir(dev);
+	if (!iommu) {
+		printk(KERN_ERR
+		       "Unable to map PCI %s to iommu\n", pci_name(dev));
+		return -ENOENT;
+	}
+
+	index = alloc_irte(iommu, irq, nvec);
+	if (index < 0) {
+		printk(KERN_ERR
+		       "Unable to allocate %d IRTE for PCI %s\n", nvec,
+		        pci_name(dev));
+		return -ENOSPC;
+	}
+	return index;
+}
+#endif
+
+static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc, int irq)
+{
+	int ret;
 	struct msi_msg msg;
+
+	ret = msi_compose_msg(dev, irq, &msg);
+	if (ret < 0)
+		return ret;
+
+	set_irq_msi(irq, desc);
+	write_msi_msg(irq, &msg);
+
+#ifdef CONFIG_INTR_REMAP
+	if (irq_remapped(irq)) {
+		struct irq_desc *desc = irq_desc + irq;
+		/*
+		 * irq migration in process context
+		 */
+		desc->status |= IRQ_MOVE_PCNTXT;
+		set_irq_chip_and_handler_name(irq, &msi_ir_chip, handle_edge_irq, "edge");
+	} else
+#endif
+		set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge");
+
+	return 0;
+}
+
+int arch_setup_msi_irq(struct pci_dev *dev, struct msi_desc *desc)
+{
 	int irq, ret;
+
 	irq = create_irq();
 	if (irq < 0)
 		return irq;
 
-	ret = msi_compose_msg(dev, irq, &msg);
+#ifdef CONFIG_INTR_REMAP
+	if (!intr_remapping_enabled)
+		goto no_ir;
+
+	ret = msi_alloc_irte(dev, irq, 1);
+	if (ret < 0)
+		goto error;
+no_ir:
+#endif
+	ret = setup_msi_irq(dev, desc, irq);
 	if (ret < 0) {
 		destroy_irq(irq);
 		return ret;
 	}
+	return 0;
 
-	set_irq_msi(irq, desc);
-	write_msi_msg(irq, &msg);
+#ifdef CONFIG_INTR_REMAP
+error:
+	destroy_irq(irq);
+	return ret;
+#endif
+}
 
-	set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge");
+int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
+{
+	int irq, ret, sub_handle;
+	struct msi_desc *desc;
+#ifdef CONFIG_INTR_REMAP
+	struct intel_iommu *iommu = 0;
+	int index = 0;
+#endif
+
+	sub_handle = 0;
+	list_for_each_entry(desc, &dev->msi_list, list) {
+		irq = create_irq();
+		if (irq < 0)
+			return irq;
+#ifdef CONFIG_INTR_REMAP
+		if (!intr_remapping_enabled)
+			goto no_ir;
 
+		if (!sub_handle) {
+			/*
+			 * allocate the consecutive block of IRTE's
+			 * for 'nvec'
+			 */
+			index = msi_alloc_irte(dev, irq, nvec);
+			if (index < 0) {
+				ret = index;
+				goto error;
+			}
+		} else {
+			iommu = map_dev_to_ir(dev);
+			if (!iommu) {
+				ret = -ENOENT;
+				goto error;
+			}
+			/*
+			 * setup the mapping between the irq and the IRTE
+			 * base index, the sub_handle pointing to the
+			 * appropriate interrupt remap table entry.
+			 */
+			set_irte_irq(irq, iommu, index, sub_handle);
+		}
+no_ir:
+#endif
+		ret = setup_msi_irq(dev, desc, irq);
+		if (ret < 0)
+			goto error;
+		sub_handle++;
+	}
 	return 0;
+
+error:
+	destroy_irq(irq);
+	return ret;
 }
 
 void arch_teardown_msi_irq(unsigned int irq)
@@ -2333,6 +2872,10 @@ void __init setup_ioapic_dest(void)
 				setup_IO_APIC_irq(ioapic, pin, irq,
 						  irq_trigger(irq_entry),
 						  irq_polarity(irq_entry));
+#ifdef CONFIG_INTR_REMAP
+			else if (intr_remapping_enabled)
+				set_ir_ioapic_affinity_irq(irq, TARGET_CPUS);
+#endif
 			else
 				set_ioapic_affinity_irq(irq, TARGET_CPUS);
 		}
diff --git a/arch/x86/kernel/irqinit_32.c b/arch/x86/kernel/irqinit_32.c
index d66914287ee1..9200a1e2752d 100644
--- a/arch/x86/kernel/irqinit_32.c
+++ b/arch/x86/kernel/irqinit_32.c
@@ -74,6 +74,15 @@ void __init init_ISA_irqs (void)
 	}
 }
 
+/*
+ * IRQ2 is cascade interrupt to second interrupt controller
+ */
+static struct irqaction irq2 = {
+	.handler = no_action,
+	.mask = CPU_MASK_NONE,
+	.name = "cascade",
+};
+
 /* Overridden in paravirt.c */
 void init_IRQ(void) __attribute__((weak, alias("native_init_IRQ")));
 
@@ -98,6 +107,46 @@ void __init native_init_IRQ(void)
 			set_intr_gate(vector, interrupt[i]);
 	}
 
+#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_SMP)
+	/*
+	 * IRQ0 must be given a fixed assignment and initialized,
+	 * because it's used before the IO-APIC is set up.
+	 */
+	set_intr_gate(FIRST_DEVICE_VECTOR, interrupt[0]);
+
+	/*
+	 * The reschedule interrupt is a CPU-to-CPU reschedule-helper
+	 * IPI, driven by wakeup.
+	 */
+	alloc_intr_gate(RESCHEDULE_VECTOR, reschedule_interrupt);
+
+	/* IPI for invalidation */
+	alloc_intr_gate(INVALIDATE_TLB_VECTOR, invalidate_interrupt);
+
+	/* IPI for generic function call */
+	alloc_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt);
+
+	/* IPI for single call function */
+	set_intr_gate(CALL_FUNCTION_SINGLE_VECTOR, call_function_single_interrupt);
+#endif
+
+#ifdef CONFIG_X86_LOCAL_APIC
+	/* self generated IPI for local APIC timer */
+	alloc_intr_gate(LOCAL_TIMER_VECTOR, apic_timer_interrupt);
+
+	/* IPI vectors for APIC spurious and error interrupts */
+	alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt);
+	alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt);
+#endif
+
+#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86_MCE_P4THERMAL)
+	/* thermal monitor LVT interrupt */
+	alloc_intr_gate(THERMAL_APIC_VECTOR, thermal_interrupt);
+#endif
+
+	if (!acpi_ioapic)
+		setup_irq(2, &irq2);
+
 	/* setup after call gates are initialised (usually add in
 	 * the architecture specific gates)
 	 */
diff --git a/arch/x86/kernel/irqinit_64.c b/arch/x86/kernel/irqinit_64.c
index 1f26fd9ec4f4..5b5be9d43c2a 100644
--- a/arch/x86/kernel/irqinit_64.c
+++ b/arch/x86/kernel/irqinit_64.c
@@ -135,7 +135,7 @@ DEFINE_PER_CPU(vector_irq_t, vector_irq) = {
 	[IRQ15_VECTOR + 1 ... NR_VECTORS - 1] = -1
 };
 
-static void __init init_ISA_irqs (void)
+void __init init_ISA_irqs(void)
 {
 	int i;
 
@@ -164,22 +164,8 @@ static void __init init_ISA_irqs (void)
 
 void init_IRQ(void) __attribute__((weak, alias("native_init_IRQ")));
 
-void __init native_init_IRQ(void)
+static void __init smp_intr_init(void)
 {
-	int i;
-
-	init_ISA_irqs();
-	/*
-	 * Cover the whole vector space, no vector can escape
-	 * us. (some of these will be overridden and become
-	 * 'special' SMP interrupts)
-	 */
-	for (i = 0; i < (NR_VECTORS - FIRST_EXTERNAL_VECTOR); i++) {
-		int vector = FIRST_EXTERNAL_VECTOR + i;
-		if (vector != IA32_SYSCALL_VECTOR)
-			set_intr_gate(vector, interrupt[i]);
-	}
-
 #ifdef CONFIG_SMP
 	/*
 	 * The reschedule interrupt is a CPU-to-CPU reschedule-helper
@@ -207,6 +193,12 @@ void __init native_init_IRQ(void)
 	/* Low priority IPI to cleanup after moving an irq */
 	set_intr_gate(IRQ_MOVE_CLEANUP_VECTOR, irq_move_cleanup_interrupt);
 #endif
+}
+
+static void __init apic_intr_init(void)
+{
+	smp_intr_init();
+
 	alloc_intr_gate(THERMAL_APIC_VECTOR, thermal_interrupt);
 	alloc_intr_gate(THRESHOLD_APIC_VECTOR, threshold_interrupt);
 
@@ -216,6 +208,25 @@ void __init native_init_IRQ(void)
 	/* IPI vectors for APIC spurious and error interrupts */
 	alloc_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt);
 	alloc_intr_gate(ERROR_APIC_VECTOR, error_interrupt);
+}
+
+void __init native_init_IRQ(void)
+{
+	int i;
+
+	init_ISA_irqs();
+	/*
+	 * Cover the whole vector space, no vector can escape
+	 * us. (some of these will be overridden and become
+	 * 'special' SMP interrupts)
+	 */
+	for (i = 0; i < (NR_VECTORS - FIRST_EXTERNAL_VECTOR); i++) {
+		int vector = FIRST_EXTERNAL_VECTOR + i;
+		if (vector != IA32_SYSCALL_VECTOR)
+			set_intr_gate(vector, interrupt[i]);
+	}
+
+	apic_intr_init();
 
 	if (!acpi_ioapic)
 		setup_irq(2, &irq2);
diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c
index 0ed5f939b905..eee32b43fee3 100644
--- a/arch/x86/kernel/ldt.c
+++ b/arch/x86/kernel/ldt.c
@@ -52,6 +52,8 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload)
 	memset(newldt + oldsize * LDT_ENTRY_SIZE, 0,
 	       (mincount - oldsize) * LDT_ENTRY_SIZE);
 
+	paravirt_alloc_ldt(newldt, mincount);
+
 #ifdef CONFIG_X86_64
 	/* CHECKME: Do we really need this ? */
 	wmb();
@@ -74,6 +76,7 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload)
 #endif
 	}
 	if (oldsize) {
+		paravirt_free_ldt(oldldt, oldsize);
 		if (oldsize * LDT_ENTRY_SIZE > PAGE_SIZE)
 			vfree(oldldt);
 		else
@@ -85,10 +88,13 @@ static int alloc_ldt(mm_context_t *pc, int mincount, int reload)
 static inline int copy_ldt(mm_context_t *new, mm_context_t *old)
 {
 	int err = alloc_ldt(new, old->size, 0);
+	int i;
 
 	if (err < 0)
 		return err;
-	memcpy(new->ldt, old->ldt, old->size * LDT_ENTRY_SIZE);
+
+	for(i = 0; i < old->size; i++)
+		write_ldt_entry(new->ldt, i, old->ldt + i * LDT_ENTRY_SIZE);
 	return 0;
 }
 
@@ -125,6 +131,7 @@ void destroy_context(struct mm_struct *mm)
 		if (mm == current->active_mm)
 			clear_LDT();
 #endif
+		paravirt_free_ldt(mm->context.ldt, mm->context.size);
 		if (mm->context.size * LDT_ENTRY_SIZE > PAGE_SIZE)
 			vfree(mm->context.ldt);
 		else
diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c
deleted file mode 100644
index 652fa5c38ebe..000000000000
--- a/arch/x86/kernel/microcode.c
+++ /dev/null
@@ -1,853 +0,0 @@
-/*
- *	Intel CPU Microcode Update Driver for Linux
- *
- *	Copyright (C) 2000-2006 Tigran Aivazian <tigran@aivazian.fsnet.co.uk>
- *		      2006	Shaohua Li <shaohua.li@intel.com>
- *
- *	This driver allows to upgrade microcode on Intel processors
- *	belonging to IA-32 family - PentiumPro, Pentium II,
- *	Pentium III, Xeon, Pentium 4, etc.
- *
- *	Reference: Section 8.11 of Volume 3a, IA-32 Intel? Architecture
- *	Software Developer's Manual
- *	Order Number 253668 or free download from:
- *
- *	http://developer.intel.com/design/pentium4/manuals/253668.htm
- *
- *	For more information, go to http://www.urbanmyth.org/microcode
- *
- *	This program is free software; you can redistribute it and/or
- *	modify it under the terms of the GNU General Public License
- *	as published by the Free Software Foundation; either version
- *	2 of the License, or (at your option) any later version.
- *
- *	1.0	16 Feb 2000, Tigran Aivazian <tigran@sco.com>
- *		Initial release.
- *	1.01	18 Feb 2000, Tigran Aivazian <tigran@sco.com>
- *		Added read() support + cleanups.
- *	1.02	21 Feb 2000, Tigran Aivazian <tigran@sco.com>
- *		Added 'device trimming' support. open(O_WRONLY) zeroes
- *		and frees the saved copy of applied microcode.
- *	1.03	29 Feb 2000, Tigran Aivazian <tigran@sco.com>
- *		Made to use devfs (/dev/cpu/microcode) + cleanups.
- *	1.04	06 Jun 2000, Simon Trimmer <simon@veritas.com>
- *		Added misc device support (now uses both devfs and misc).
- *		Added MICROCODE_IOCFREE ioctl to clear memory.
- *	1.05	09 Jun 2000, Simon Trimmer <simon@veritas.com>
- *		Messages for error cases (non Intel & no suitable microcode).
- *	1.06	03 Aug 2000, Tigran Aivazian <tigran@veritas.com>
- *		Removed ->release(). Removed exclusive open and status bitmap.
- *		Added microcode_rwsem to serialize read()/write()/ioctl().
- *		Removed global kernel lock usage.
- *	1.07	07 Sep 2000, Tigran Aivazian <tigran@veritas.com>
- *		Write 0 to 0x8B msr and then cpuid before reading revision,
- *		so that it works even if there were no update done by the
- *		BIOS. Otherwise, reading from 0x8B gives junk (which happened
- *		to be 0 on my machine which is why it worked even when I
- *		disabled update by the BIOS)
- *		Thanks to Eric W. Biederman <ebiederman@lnxi.com> for the fix.
- *	1.08	11 Dec 2000, Richard Schaal <richard.schaal@intel.com> and
- *			     Tigran Aivazian <tigran@veritas.com>
- *		Intel Pentium 4 processor support and bugfixes.
- *	1.09	30 Oct 2001, Tigran Aivazian <tigran@veritas.com>
- *		Bugfix for HT (Hyper-Threading) enabled processors
- *		whereby processor resources are shared by all logical processors
- *		in a single CPU package.
- *	1.10	28 Feb 2002 Asit K Mallick <asit.k.mallick@intel.com> and
- *		Tigran Aivazian <tigran@veritas.com>,
- *		Serialize updates as required on HT processors due to speculative
- *		nature of implementation.
- *	1.11	22 Mar 2002 Tigran Aivazian <tigran@veritas.com>
- *		Fix the panic when writing zero-length microcode chunk.
- *	1.12	29 Sep 2003 Nitin Kamble <nitin.a.kamble@intel.com>,
- *		Jun Nakajima <jun.nakajima@intel.com>
- *		Support for the microcode updates in the new format.
- *	1.13	10 Oct 2003 Tigran Aivazian <tigran@veritas.com>
- *		Removed ->read() method and obsoleted MICROCODE_IOCFREE ioctl
- *		because we no longer hold a copy of applied microcode
- *		in kernel memory.
- *	1.14	25 Jun 2004 Tigran Aivazian <tigran@veritas.com>
- *		Fix sigmatch() macro to handle old CPUs with pf == 0.
- *		Thanks to Stuart Swales for pointing out this bug.
- */
-
-//#define DEBUG /* pr_debug */
-#include <linux/capability.h>
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/sched.h>
-#include <linux/smp_lock.h>
-#include <linux/cpumask.h>
-#include <linux/module.h>
-#include <linux/slab.h>
-#include <linux/vmalloc.h>
-#include <linux/miscdevice.h>
-#include <linux/spinlock.h>
-#include <linux/mm.h>
-#include <linux/fs.h>
-#include <linux/mutex.h>
-#include <linux/cpu.h>
-#include <linux/firmware.h>
-#include <linux/platform_device.h>
-
-#include <asm/msr.h>
-#include <asm/uaccess.h>
-#include <asm/processor.h>
-
-MODULE_DESCRIPTION("Intel CPU (IA-32) Microcode Update Driver");
-MODULE_AUTHOR("Tigran Aivazian <tigran@aivazian.fsnet.co.uk>");
-MODULE_LICENSE("GPL");
-
-#define MICROCODE_VERSION 	"1.14a"
-
-#define DEFAULT_UCODE_DATASIZE 	(2000) 	  /* 2000 bytes */
-#define MC_HEADER_SIZE		(sizeof (microcode_header_t))  	  /* 48 bytes */
-#define DEFAULT_UCODE_TOTALSIZE (DEFAULT_UCODE_DATASIZE + MC_HEADER_SIZE) /* 2048 bytes */
-#define EXT_HEADER_SIZE		(sizeof (struct extended_sigtable)) /* 20 bytes */
-#define EXT_SIGNATURE_SIZE	(sizeof (struct extended_signature)) /* 12 bytes */
-#define DWSIZE			(sizeof (u32))
-#define get_totalsize(mc) \
-	(((microcode_t *)mc)->hdr.totalsize ? \
-	 ((microcode_t *)mc)->hdr.totalsize : DEFAULT_UCODE_TOTALSIZE)
-#define get_datasize(mc) \
-	(((microcode_t *)mc)->hdr.datasize ? \
-	 ((microcode_t *)mc)->hdr.datasize : DEFAULT_UCODE_DATASIZE)
-
-#define sigmatch(s1, s2, p1, p2) \
-	(((s1) == (s2)) && (((p1) & (p2)) || (((p1) == 0) && ((p2) == 0))))
-
-#define exttable_size(et) ((et)->count * EXT_SIGNATURE_SIZE + EXT_HEADER_SIZE)
-
-/* serialize access to the physical write to MSR 0x79 */
-static DEFINE_SPINLOCK(microcode_update_lock);
-
-/* no concurrent ->write()s are allowed on /dev/cpu/microcode */
-static DEFINE_MUTEX(microcode_mutex);
-
-static struct ucode_cpu_info {
-	int valid;
-	unsigned int sig;
-	unsigned int pf;
-	unsigned int rev;
-	microcode_t *mc;
-} ucode_cpu_info[NR_CPUS];
-
-static void collect_cpu_info(int cpu_num)
-{
-	struct cpuinfo_x86 *c = &cpu_data(cpu_num);
-	struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num;
-	unsigned int val[2];
-
-	/* We should bind the task to the CPU */
-	BUG_ON(raw_smp_processor_id() != cpu_num);
-	uci->pf = uci->rev = 0;
-	uci->mc = NULL;
-	uci->valid = 1;
-
-	if (c->x86_vendor != X86_VENDOR_INTEL || c->x86 < 6 ||
-	    	cpu_has(c, X86_FEATURE_IA64)) {
-		printk(KERN_ERR "microcode: CPU%d not a capable Intel "
-			"processor\n", cpu_num);
-		uci->valid = 0;
-		return;
-	}
-
-	uci->sig = cpuid_eax(0x00000001);
-
-	if ((c->x86_model >= 5) || (c->x86 > 6)) {
-		/* get processor flags from MSR 0x17 */
-		rdmsr(MSR_IA32_PLATFORM_ID, val[0], val[1]);
-		uci->pf = 1 << ((val[1] >> 18) & 7);
-	}
-
-	wrmsr(MSR_IA32_UCODE_REV, 0, 0);
-	/* see notes above for revision 1.07.  Apparent chip bug */
-	sync_core();
-	/* get the current revision from MSR 0x8B */
-	rdmsr(MSR_IA32_UCODE_REV, val[0], uci->rev);
-	pr_debug("microcode: collect_cpu_info : sig=0x%x, pf=0x%x, rev=0x%x\n",
-			uci->sig, uci->pf, uci->rev);
-}
-
-static inline int microcode_update_match(int cpu_num,
-	microcode_header_t *mc_header, int sig, int pf)
-{
-	struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num;
-
-	if (!sigmatch(sig, uci->sig, pf, uci->pf)
-		|| mc_header->rev <= uci->rev)
-		return 0;
-	return 1;
-}
-
-static int microcode_sanity_check(void *mc)
-{
-	microcode_header_t *mc_header = mc;
-	struct extended_sigtable *ext_header = NULL;
-	struct extended_signature *ext_sig;
-	unsigned long total_size, data_size, ext_table_size;
-	int sum, orig_sum, ext_sigcount = 0, i;
-
-	total_size = get_totalsize(mc_header);
-	data_size = get_datasize(mc_header);
-	if (data_size + MC_HEADER_SIZE > total_size) {
-		printk(KERN_ERR "microcode: error! "
-			"Bad data size in microcode data file\n");
-		return -EINVAL;
-	}
-
-	if (mc_header->ldrver != 1 || mc_header->hdrver != 1) {
-		printk(KERN_ERR "microcode: error! "
-			"Unknown microcode update format\n");
-		return -EINVAL;
-	}
-	ext_table_size = total_size - (MC_HEADER_SIZE + data_size);
-	if (ext_table_size) {
-		if ((ext_table_size < EXT_HEADER_SIZE)
-		 || ((ext_table_size - EXT_HEADER_SIZE) % EXT_SIGNATURE_SIZE)) {
-			printk(KERN_ERR "microcode: error! "
-				"Small exttable size in microcode data file\n");
-			return -EINVAL;
-		}
-		ext_header = mc + MC_HEADER_SIZE + data_size;
-		if (ext_table_size != exttable_size(ext_header)) {
-			printk(KERN_ERR "microcode: error! "
-				"Bad exttable size in microcode data file\n");
-			return -EFAULT;
-		}
-		ext_sigcount = ext_header->count;
-	}
-
-	/* check extended table checksum */
-	if (ext_table_size) {
-		int ext_table_sum = 0;
-		int *ext_tablep = (int *)ext_header;
-
-		i = ext_table_size / DWSIZE;
-		while (i--)
-			ext_table_sum += ext_tablep[i];
-		if (ext_table_sum) {
-			printk(KERN_WARNING "microcode: aborting, "
-				"bad extended signature table checksum\n");
-			return -EINVAL;
-		}
-	}
-
-	/* calculate the checksum */
-	orig_sum = 0;
-	i = (MC_HEADER_SIZE + data_size) / DWSIZE;
-	while (i--)
-		orig_sum += ((int *)mc)[i];
-	if (orig_sum) {
-		printk(KERN_ERR "microcode: aborting, bad checksum\n");
-		return -EINVAL;
-	}
-	if (!ext_table_size)
-		return 0;
-	/* check extended signature checksum */
-	for (i = 0; i < ext_sigcount; i++) {
-		ext_sig = (void *)ext_header + EXT_HEADER_SIZE +
-			  EXT_SIGNATURE_SIZE * i;
-		sum = orig_sum
-			- (mc_header->sig + mc_header->pf + mc_header->cksum)
-			+ (ext_sig->sig + ext_sig->pf + ext_sig->cksum);
-		if (sum) {
-			printk(KERN_ERR "microcode: aborting, bad checksum\n");
-			return -EINVAL;
-		}
-	}
-	return 0;
-}
-
-/*
- * return 0 - no update found
- * return 1 - found update
- * return < 0 - error
- */
-static int get_maching_microcode(void *mc, int cpu)
-{
-	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
-	microcode_header_t *mc_header = mc;
-	struct extended_sigtable *ext_header;
-	unsigned long total_size = get_totalsize(mc_header);
-	int ext_sigcount, i;
-	struct extended_signature *ext_sig;
-	void *new_mc;
-
-	if (microcode_update_match(cpu, mc_header,
-			mc_header->sig, mc_header->pf))
-		goto find;
-
-	if (total_size <= get_datasize(mc_header) + MC_HEADER_SIZE)
-		return 0;
-
-	ext_header = mc + get_datasize(mc_header) + MC_HEADER_SIZE;
-	ext_sigcount = ext_header->count;
-	ext_sig = (void *)ext_header + EXT_HEADER_SIZE;
-	for (i = 0; i < ext_sigcount; i++) {
-		if (microcode_update_match(cpu, mc_header,
-				ext_sig->sig, ext_sig->pf))
-			goto find;
-		ext_sig++;
-	}
-	return 0;
-find:
-	pr_debug("microcode: CPU%d found a matching microcode update with"
-		" version 0x%x (current=0x%x)\n", cpu, mc_header->rev,uci->rev);
-	new_mc = vmalloc(total_size);
-	if (!new_mc) {
-		printk(KERN_ERR "microcode: error! Can not allocate memory\n");
-		return -ENOMEM;
-	}
-
-	/* free previous update file */
-	vfree(uci->mc);
-
-	memcpy(new_mc, mc, total_size);
-	uci->mc = new_mc;
-	return 1;
-}
-
-static void apply_microcode(int cpu)
-{
-	unsigned long flags;
-	unsigned int val[2];
-	int cpu_num = raw_smp_processor_id();
-	struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num;
-
-	/* We should bind the task to the CPU */
-	BUG_ON(cpu_num != cpu);
-
-	if (uci->mc == NULL)
-		return;
-
-	/* serialize access to the physical write to MSR 0x79 */
-	spin_lock_irqsave(&microcode_update_lock, flags);
-
-	/* write microcode via MSR 0x79 */
-	wrmsr(MSR_IA32_UCODE_WRITE,
-		(unsigned long) uci->mc->bits,
-		(unsigned long) uci->mc->bits >> 16 >> 16);
-	wrmsr(MSR_IA32_UCODE_REV, 0, 0);
-
-	/* see notes above for revision 1.07.  Apparent chip bug */
-	sync_core();
-
-	/* get the current revision from MSR 0x8B */
-	rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]);
-
-	spin_unlock_irqrestore(&microcode_update_lock, flags);
-	if (val[1] != uci->mc->hdr.rev) {
-		printk(KERN_ERR "microcode: CPU%d update from revision "
-			"0x%x to 0x%x failed\n", cpu_num, uci->rev, val[1]);
-		return;
-	}
-	printk(KERN_INFO "microcode: CPU%d updated from revision "
-	       "0x%x to 0x%x, date = %08x \n",
-	       cpu_num, uci->rev, val[1], uci->mc->hdr.date);
-	uci->rev = val[1];
-}
-
-#ifdef CONFIG_MICROCODE_OLD_INTERFACE
-static void __user *user_buffer;	/* user area microcode data buffer */
-static unsigned int user_buffer_size;	/* it's size */
-
-static long get_next_ucode(void **mc, long offset)
-{
-	microcode_header_t mc_header;
-	unsigned long total_size;
-
-	/* No more data */
-	if (offset >= user_buffer_size)
-		return 0;
-	if (copy_from_user(&mc_header, user_buffer + offset, MC_HEADER_SIZE)) {
-		printk(KERN_ERR "microcode: error! Can not read user data\n");
-		return -EFAULT;
-	}
-	total_size = get_totalsize(&mc_header);
-	if (offset + total_size > user_buffer_size) {
-		printk(KERN_ERR "microcode: error! Bad total size in microcode "
-				"data file\n");
-		return -EINVAL;
-	}
-	*mc = vmalloc(total_size);
-	if (!*mc)
-		return -ENOMEM;
-	if (copy_from_user(*mc, user_buffer + offset, total_size)) {
-		printk(KERN_ERR "microcode: error! Can not read user data\n");
-		vfree(*mc);
-		return -EFAULT;
-	}
-	return offset + total_size;
-}
-
-static int do_microcode_update (void)
-{
-	long cursor = 0;
-	int error = 0;
-	void *new_mc = NULL;
-	int cpu;
-	cpumask_t old;
-
-	old = current->cpus_allowed;
-
-	while ((cursor = get_next_ucode(&new_mc, cursor)) > 0) {
-		error = microcode_sanity_check(new_mc);
-		if (error)
-			goto out;
-		/*
-		 * It's possible the data file has multiple matching ucode,
-		 * lets keep searching till the latest version
-		 */
-		for_each_online_cpu(cpu) {
-			struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
-
-			if (!uci->valid)
-				continue;
-			set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
-			error = get_maching_microcode(new_mc, cpu);
-			if (error < 0)
-				goto out;
-			if (error == 1)
-				apply_microcode(cpu);
-		}
-		vfree(new_mc);
-	}
-out:
-	if (cursor > 0)
-		vfree(new_mc);
-	if (cursor < 0)
-		error = cursor;
-	set_cpus_allowed_ptr(current, &old);
-	return error;
-}
-
-static int microcode_open (struct inode *unused1, struct file *unused2)
-{
-	cycle_kernel_lock();
-	return capable(CAP_SYS_RAWIO) ? 0 : -EPERM;
-}
-
-static ssize_t microcode_write (struct file *file, const char __user *buf, size_t len, loff_t *ppos)
-{
-	ssize_t ret;
-
-	if ((len >> PAGE_SHIFT) > num_physpages) {
-		printk(KERN_ERR "microcode: too much data (max %ld pages)\n", num_physpages);
-		return -EINVAL;
-	}
-
-	get_online_cpus();
-	mutex_lock(&microcode_mutex);
-
-	user_buffer = (void __user *) buf;
-	user_buffer_size = (int) len;
-
-	ret = do_microcode_update();
-	if (!ret)
-		ret = (ssize_t)len;
-
-	mutex_unlock(&microcode_mutex);
-	put_online_cpus();
-
-	return ret;
-}
-
-static const struct file_operations microcode_fops = {
-	.owner		= THIS_MODULE,
-	.write		= microcode_write,
-	.open		= microcode_open,
-};
-
-static struct miscdevice microcode_dev = {
-	.minor		= MICROCODE_MINOR,
-	.name		= "microcode",
-	.fops		= &microcode_fops,
-};
-
-static int __init microcode_dev_init (void)
-{
-	int error;
-
-	error = misc_register(&microcode_dev);
-	if (error) {
-		printk(KERN_ERR
-			"microcode: can't misc_register on minor=%d\n",
-			MICROCODE_MINOR);
-		return error;
-	}
-
-	return 0;
-}
-
-static void microcode_dev_exit (void)
-{
-	misc_deregister(&microcode_dev);
-}
-
-MODULE_ALIAS_MISCDEV(MICROCODE_MINOR);
-#else
-#define microcode_dev_init() 0
-#define microcode_dev_exit() do { } while(0)
-#endif
-
-static long get_next_ucode_from_buffer(void **mc, const u8 *buf,
-	unsigned long size, long offset)
-{
-	microcode_header_t *mc_header;
-	unsigned long total_size;
-
-	/* No more data */
-	if (offset >= size)
-		return 0;
-	mc_header = (microcode_header_t *)(buf + offset);
-	total_size = get_totalsize(mc_header);
-
-	if (offset + total_size > size) {
-		printk(KERN_ERR "microcode: error! Bad data in microcode data file\n");
-		return -EINVAL;
-	}
-
-	*mc = vmalloc(total_size);
-	if (!*mc) {
-		printk(KERN_ERR "microcode: error! Can not allocate memory\n");
-		return -ENOMEM;
-	}
-	memcpy(*mc, buf + offset, total_size);
-	return offset + total_size;
-}
-
-/* fake device for request_firmware */
-static struct platform_device *microcode_pdev;
-
-static int cpu_request_microcode(int cpu)
-{
-	char name[30];
-	struct cpuinfo_x86 *c = &cpu_data(cpu);
-	const struct firmware *firmware;
-	const u8 *buf;
-	unsigned long size;
-	long offset = 0;
-	int error;
-	void *mc;
-
-	/* We should bind the task to the CPU */
-	BUG_ON(cpu != raw_smp_processor_id());
-	sprintf(name,"intel-ucode/%02x-%02x-%02x",
-		c->x86, c->x86_model, c->x86_mask);
-	error = request_firmware(&firmware, name, &microcode_pdev->dev);
-	if (error) {
-		pr_debug("microcode: data file %s load failed\n", name);
-		return error;
-	}
-	buf = firmware->data;
-	size = firmware->size;
-	while ((offset = get_next_ucode_from_buffer(&mc, buf, size, offset))
-			> 0) {
-		error = microcode_sanity_check(mc);
-		if (error)
-			break;
-		error = get_maching_microcode(mc, cpu);
-		if (error < 0)
-			break;
-		/*
-		 * It's possible the data file has multiple matching ucode,
-		 * lets keep searching till the latest version
-		 */
-		if (error == 1) {
-			apply_microcode(cpu);
-			error = 0;
-		}
-		vfree(mc);
-	}
-	if (offset > 0)
-		vfree(mc);
-	if (offset < 0)
-		error = offset;
-	release_firmware(firmware);
-
-	return error;
-}
-
-static int apply_microcode_check_cpu(int cpu)
-{
-	struct cpuinfo_x86 *c = &cpu_data(cpu);
-	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
-	cpumask_t old;
-	unsigned int val[2];
-	int err = 0;
-
-	/* Check if the microcode is available */
-	if (!uci->mc)
-		return 0;
-
-	old = current->cpus_allowed;
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
-
-	/* Check if the microcode we have in memory matches the CPU */
-	if (c->x86_vendor != X86_VENDOR_INTEL || c->x86 < 6 ||
-	    cpu_has(c, X86_FEATURE_IA64) || uci->sig != cpuid_eax(0x00000001))
-		err = -EINVAL;
-
-	if (!err && ((c->x86_model >= 5) || (c->x86 > 6))) {
-		/* get processor flags from MSR 0x17 */
-		rdmsr(MSR_IA32_PLATFORM_ID, val[0], val[1]);
-		if (uci->pf != (1 << ((val[1] >> 18) & 7)))
-			err = -EINVAL;
-	}
-
-	if (!err) {
-		wrmsr(MSR_IA32_UCODE_REV, 0, 0);
-		/* see notes above for revision 1.07.  Apparent chip bug */
-		sync_core();
-		/* get the current revision from MSR 0x8B */
-		rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]);
-		if (uci->rev != val[1])
-			err = -EINVAL;
-	}
-
-	if (!err)
-		apply_microcode(cpu);
-	else
-		printk(KERN_ERR "microcode: Could not apply microcode to CPU%d:"
-			" sig=0x%x, pf=0x%x, rev=0x%x\n",
-			cpu, uci->sig, uci->pf, uci->rev);
-
-	set_cpus_allowed_ptr(current, &old);
-	return err;
-}
-
-static void microcode_init_cpu(int cpu, int resume)
-{
-	cpumask_t old;
-	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
-
-	old = current->cpus_allowed;
-
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
-	mutex_lock(&microcode_mutex);
-	collect_cpu_info(cpu);
-	if (uci->valid && system_state == SYSTEM_RUNNING && !resume)
-		cpu_request_microcode(cpu);
-	mutex_unlock(&microcode_mutex);
-	set_cpus_allowed_ptr(current, &old);
-}
-
-static void microcode_fini_cpu(int cpu)
-{
-	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
-
-	mutex_lock(&microcode_mutex);
-	uci->valid = 0;
-	vfree(uci->mc);
-	uci->mc = NULL;
-	mutex_unlock(&microcode_mutex);
-}
-
-static ssize_t reload_store(struct sys_device *dev,
-			    struct sysdev_attribute *attr,
-			    const char *buf, size_t sz)
-{
-	struct ucode_cpu_info *uci = ucode_cpu_info + dev->id;
-	char *end;
-	unsigned long val = simple_strtoul(buf, &end, 0);
-	int err = 0;
-	int cpu = dev->id;
-
-	if (end == buf)
-		return -EINVAL;
-	if (val == 1) {
-		cpumask_t old = current->cpus_allowed;
-
-		get_online_cpus();
-		set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
-
-		mutex_lock(&microcode_mutex);
-		if (uci->valid)
-			err = cpu_request_microcode(cpu);
-		mutex_unlock(&microcode_mutex);
-		put_online_cpus();
-		set_cpus_allowed_ptr(current, &old);
-	}
-	if (err)
-		return err;
-	return sz;
-}
-
-static ssize_t version_show(struct sys_device *dev,
-			struct sysdev_attribute *attr, char *buf)
-{
-	struct ucode_cpu_info *uci = ucode_cpu_info + dev->id;
-
-	return sprintf(buf, "0x%x\n", uci->rev);
-}
-
-static ssize_t pf_show(struct sys_device *dev,
-			struct sysdev_attribute *attr, char *buf)
-{
-	struct ucode_cpu_info *uci = ucode_cpu_info + dev->id;
-
-	return sprintf(buf, "0x%x\n", uci->pf);
-}
-
-static SYSDEV_ATTR(reload, 0200, NULL, reload_store);
-static SYSDEV_ATTR(version, 0400, version_show, NULL);
-static SYSDEV_ATTR(processor_flags, 0400, pf_show, NULL);
-
-static struct attribute *mc_default_attrs[] = {
-	&attr_reload.attr,
-	&attr_version.attr,
-	&attr_processor_flags.attr,
-	NULL
-};
-
-static struct attribute_group mc_attr_group = {
-	.attrs = mc_default_attrs,
-	.name = "microcode",
-};
-
-static int __mc_sysdev_add(struct sys_device *sys_dev, int resume)
-{
-	int err, cpu = sys_dev->id;
-	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
-
-	if (!cpu_online(cpu))
-		return 0;
-
-	pr_debug("microcode: CPU%d added\n", cpu);
-	memset(uci, 0, sizeof(*uci));
-
-	err = sysfs_create_group(&sys_dev->kobj, &mc_attr_group);
-	if (err)
-		return err;
-
-	microcode_init_cpu(cpu, resume);
-
-	return 0;
-}
-
-static int mc_sysdev_add(struct sys_device *sys_dev)
-{
-	return __mc_sysdev_add(sys_dev, 0);
-}
-
-static int mc_sysdev_remove(struct sys_device *sys_dev)
-{
-	int cpu = sys_dev->id;
-
-	if (!cpu_online(cpu))
-		return 0;
-
-	pr_debug("microcode: CPU%d removed\n", cpu);
-	microcode_fini_cpu(cpu);
-	sysfs_remove_group(&sys_dev->kobj, &mc_attr_group);
-	return 0;
-}
-
-static int mc_sysdev_resume(struct sys_device *dev)
-{
-	int cpu = dev->id;
-
-	if (!cpu_online(cpu))
-		return 0;
-	pr_debug("microcode: CPU%d resumed\n", cpu);
-	/* only CPU 0 will apply ucode here */
-	apply_microcode(0);
-	return 0;
-}
-
-static struct sysdev_driver mc_sysdev_driver = {
-	.add = mc_sysdev_add,
-	.remove = mc_sysdev_remove,
-	.resume = mc_sysdev_resume,
-};
-
-static __cpuinit int
-mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu)
-{
-	unsigned int cpu = (unsigned long)hcpu;
-	struct sys_device *sys_dev;
-
-	sys_dev = get_cpu_sysdev(cpu);
-	switch (action) {
-	case CPU_UP_CANCELED_FROZEN:
-		/* The CPU refused to come up during a system resume */
-		microcode_fini_cpu(cpu);
-		break;
-	case CPU_ONLINE:
-	case CPU_DOWN_FAILED:
-		mc_sysdev_add(sys_dev);
-		break;
-	case CPU_ONLINE_FROZEN:
-		/* System-wide resume is in progress, try to apply microcode */
-		if (apply_microcode_check_cpu(cpu)) {
-			/* The application of microcode failed */
-			microcode_fini_cpu(cpu);
-			__mc_sysdev_add(sys_dev, 1);
-			break;
-		}
-	case CPU_DOWN_FAILED_FROZEN:
-		if (sysfs_create_group(&sys_dev->kobj, &mc_attr_group))
-			printk(KERN_ERR "microcode: Failed to create the sysfs "
-				"group for CPU%d\n", cpu);
-		break;
-	case CPU_DOWN_PREPARE:
-		mc_sysdev_remove(sys_dev);
-		break;
-	case CPU_DOWN_PREPARE_FROZEN:
-		/* Suspend is in progress, only remove the interface */
-		sysfs_remove_group(&sys_dev->kobj, &mc_attr_group);
-		break;
-	}
-	return NOTIFY_OK;
-}
-
-static struct notifier_block __refdata mc_cpu_notifier = {
-	.notifier_call = mc_cpu_callback,
-};
-
-static int __init microcode_init (void)
-{
-	int error;
-
-	printk(KERN_INFO
-		"IA-32 Microcode Update Driver: v" MICROCODE_VERSION " <tigran@aivazian.fsnet.co.uk>\n");
-
-	error = microcode_dev_init();
-	if (error)
-		return error;
-	microcode_pdev = platform_device_register_simple("microcode", -1,
-							 NULL, 0);
-	if (IS_ERR(microcode_pdev)) {
-		microcode_dev_exit();
-		return PTR_ERR(microcode_pdev);
-	}
-
-	get_online_cpus();
-	error = sysdev_driver_register(&cpu_sysdev_class, &mc_sysdev_driver);
-	put_online_cpus();
-	if (error) {
-		microcode_dev_exit();
-		platform_device_unregister(microcode_pdev);
-		return error;
-	}
-
-	register_hotcpu_notifier(&mc_cpu_notifier);
-	return 0;
-}
-
-static void __exit microcode_exit (void)
-{
-	microcode_dev_exit();
-
-	unregister_hotcpu_notifier(&mc_cpu_notifier);
-
-	get_online_cpus();
-	sysdev_driver_unregister(&cpu_sysdev_class, &mc_sysdev_driver);
-	put_online_cpus();
-
-	platform_device_unregister(microcode_pdev);
-}
-
-module_init(microcode_init)
-module_exit(microcode_exit)
diff --git a/arch/x86/kernel/microcode_amd.c b/arch/x86/kernel/microcode_amd.c
new file mode 100644
index 000000000000..7a1f8eeac2c7
--- /dev/null
+++ b/arch/x86/kernel/microcode_amd.c
@@ -0,0 +1,435 @@
+/*
+ *  AMD CPU Microcode Update Driver for Linux
+ *  Copyright (C) 2008 Advanced Micro Devices Inc.
+ *
+ *  Author: Peter Oruba <peter.oruba@amd.com>
+ *
+ *  Based on work by:
+ *  Tigran Aivazian <tigran@aivazian.fsnet.co.uk>
+ *
+ *  This driver allows to upgrade microcode on AMD
+ *  family 0x10 and 0x11 processors.
+ *
+ *  Licensed unter the terms of the GNU General Public
+ *  License version 2. See file COPYING for details.
+*/
+
+#include <linux/capability.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/cpumask.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/miscdevice.h>
+#include <linux/spinlock.h>
+#include <linux/mm.h>
+#include <linux/fs.h>
+#include <linux/mutex.h>
+#include <linux/cpu.h>
+#include <linux/firmware.h>
+#include <linux/platform_device.h>
+#include <linux/pci.h>
+#include <linux/pci_ids.h>
+
+#include <asm/msr.h>
+#include <asm/uaccess.h>
+#include <asm/processor.h>
+#include <asm/microcode.h>
+
+MODULE_DESCRIPTION("AMD Microcode Update Driver");
+MODULE_AUTHOR("Peter Oruba <peter.oruba@amd.com>");
+MODULE_LICENSE("GPL v2");
+
+#define UCODE_MAGIC                0x00414d44
+#define UCODE_EQUIV_CPU_TABLE_TYPE 0x00000000
+#define UCODE_UCODE_TYPE           0x00000001
+
+struct equiv_cpu_entry {
+	unsigned int installed_cpu;
+	unsigned int fixed_errata_mask;
+	unsigned int fixed_errata_compare;
+	unsigned int equiv_cpu;
+};
+
+struct microcode_header_amd {
+	unsigned int  data_code;
+	unsigned int  patch_id;
+	unsigned char mc_patch_data_id[2];
+	unsigned char mc_patch_data_len;
+	unsigned char init_flag;
+	unsigned int  mc_patch_data_checksum;
+	unsigned int  nb_dev_id;
+	unsigned int  sb_dev_id;
+	unsigned char processor_rev_id[2];
+	unsigned char nb_rev_id;
+	unsigned char sb_rev_id;
+	unsigned char bios_api_rev;
+	unsigned char reserved1[3];
+	unsigned int  match_reg[8];
+};
+
+struct microcode_amd {
+	struct microcode_header_amd hdr;
+	unsigned int mpb[0];
+};
+
+#define UCODE_MAX_SIZE          (2048)
+#define DEFAULT_UCODE_DATASIZE	(896)
+#define MC_HEADER_SIZE		(sizeof(struct microcode_header_amd))
+#define DEFAULT_UCODE_TOTALSIZE (DEFAULT_UCODE_DATASIZE + MC_HEADER_SIZE)
+#define DWSIZE			(sizeof(u32))
+/* For now we support a fixed ucode total size only */
+#define get_totalsize(mc) \
+	((((struct microcode_amd *)mc)->hdr.mc_patch_data_len * 28) \
+	 + MC_HEADER_SIZE)
+
+/* serialize access to the physical write */
+static DEFINE_SPINLOCK(microcode_update_lock);
+
+static struct equiv_cpu_entry *equiv_cpu_table;
+
+static int collect_cpu_info_amd(int cpu, struct cpu_signature *csig)
+{
+	struct cpuinfo_x86 *c = &cpu_data(cpu);
+
+	memset(csig, 0, sizeof(*csig));
+
+	if (c->x86_vendor != X86_VENDOR_AMD || c->x86 < 0x10) {
+		printk(KERN_ERR "microcode: CPU%d not a capable AMD processor\n",
+		       cpu);
+		return -1;
+	}
+
+	asm volatile("movl %1, %%ecx; rdmsr"
+		     : "=a" (csig->rev)
+		     : "i" (0x0000008B) : "ecx");
+
+	printk(KERN_INFO "microcode: collect_cpu_info_amd : patch_id=0x%x\n",
+		csig->rev);
+
+	return 0;
+}
+
+static int get_matching_microcode(int cpu, void *mc, int rev)
+{
+	struct microcode_header_amd *mc_header = mc;
+	struct pci_dev *nb_pci_dev, *sb_pci_dev;
+	unsigned int current_cpu_id;
+	unsigned int equiv_cpu_id = 0x00;
+	unsigned int i = 0;
+
+	BUG_ON(equiv_cpu_table == NULL);
+	current_cpu_id = cpuid_eax(0x00000001);
+
+	while (equiv_cpu_table[i].installed_cpu != 0) {
+		if (current_cpu_id == equiv_cpu_table[i].installed_cpu) {
+			equiv_cpu_id = equiv_cpu_table[i].equiv_cpu;
+			break;
+		}
+		i++;
+	}
+
+	if (!equiv_cpu_id) {
+		printk(KERN_ERR "microcode: CPU%d cpu_id "
+		       "not found in equivalent cpu table \n", cpu);
+		return 0;
+	}
+
+	if ((mc_header->processor_rev_id[0]) != (equiv_cpu_id & 0xff)) {
+		printk(KERN_ERR
+			"microcode: CPU%d patch does not match "
+			"(patch is %x, cpu extended is %x) \n",
+			cpu, mc_header->processor_rev_id[0],
+			(equiv_cpu_id & 0xff));
+		return 0;
+	}
+
+	if ((mc_header->processor_rev_id[1]) != ((equiv_cpu_id >> 16) & 0xff)) {
+		printk(KERN_ERR "microcode: CPU%d patch does not match "
+			"(patch is %x, cpu base id is %x) \n",
+			cpu, mc_header->processor_rev_id[1],
+			((equiv_cpu_id >> 16) & 0xff));
+
+		return 0;
+	}
+
+	/* ucode may be northbridge specific */
+	if (mc_header->nb_dev_id) {
+		nb_pci_dev = pci_get_device(PCI_VENDOR_ID_AMD,
+					    (mc_header->nb_dev_id & 0xff),
+					    NULL);
+		if ((!nb_pci_dev) ||
+		    (mc_header->nb_rev_id != nb_pci_dev->revision)) {
+			printk(KERN_ERR "microcode: CPU%d NB mismatch \n", cpu);
+			pci_dev_put(nb_pci_dev);
+			return 0;
+		}
+		pci_dev_put(nb_pci_dev);
+	}
+
+	/* ucode may be southbridge specific */
+	if (mc_header->sb_dev_id) {
+		sb_pci_dev = pci_get_device(PCI_VENDOR_ID_AMD,
+					    (mc_header->sb_dev_id & 0xff),
+					    NULL);
+		if ((!sb_pci_dev) ||
+		    (mc_header->sb_rev_id != sb_pci_dev->revision)) {
+			printk(KERN_ERR "microcode: CPU%d SB mismatch \n", cpu);
+			pci_dev_put(sb_pci_dev);
+			return 0;
+		}
+		pci_dev_put(sb_pci_dev);
+	}
+
+	if (mc_header->patch_id <= rev)
+		return 0;
+
+	return 1;
+}
+
+static void apply_microcode_amd(int cpu)
+{
+	unsigned long flags;
+	unsigned int eax, edx;
+	unsigned int rev;
+	int cpu_num = raw_smp_processor_id();
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num;
+	struct microcode_amd *mc_amd = uci->mc;
+	unsigned long addr;
+
+	/* We should bind the task to the CPU */
+	BUG_ON(cpu_num != cpu);
+
+	if (mc_amd == NULL)
+		return;
+
+	spin_lock_irqsave(&microcode_update_lock, flags);
+
+	addr = (unsigned long)&mc_amd->hdr.data_code;
+	edx = (unsigned int)(((unsigned long)upper_32_bits(addr)));
+	eax = (unsigned int)(((unsigned long)lower_32_bits(addr)));
+
+	asm volatile("movl %0, %%ecx; wrmsr" :
+		     : "i" (0xc0010020), "a" (eax), "d" (edx) : "ecx");
+
+	/* get patch id after patching */
+	asm volatile("movl %1, %%ecx; rdmsr"
+		     : "=a" (rev)
+		     : "i" (0x0000008B) : "ecx");
+
+	spin_unlock_irqrestore(&microcode_update_lock, flags);
+
+	/* check current patch id and patch's id for match */
+	if (rev != mc_amd->hdr.patch_id) {
+		printk(KERN_ERR "microcode: CPU%d update from revision "
+		       "0x%x to 0x%x failed\n", cpu_num,
+		       mc_amd->hdr.patch_id, rev);
+		return;
+	}
+
+	printk(KERN_INFO "microcode: CPU%d updated from revision "
+	       "0x%x to 0x%x \n",
+	       cpu_num, uci->cpu_sig.rev, mc_amd->hdr.patch_id);
+
+	uci->cpu_sig.rev = rev;
+}
+
+static void * get_next_ucode(u8 *buf, unsigned int size,
+			int (*get_ucode_data)(void *, const void *, size_t),
+			unsigned int *mc_size)
+{
+	unsigned int total_size;
+#define UCODE_CONTAINER_SECTION_HDR	8
+	u8 section_hdr[UCODE_CONTAINER_SECTION_HDR];
+	void *mc;
+
+	if (get_ucode_data(section_hdr, buf, UCODE_CONTAINER_SECTION_HDR))
+		return NULL;
+
+	if (section_hdr[0] != UCODE_UCODE_TYPE) {
+		printk(KERN_ERR "microcode: error! "
+		       "Wrong microcode payload type field\n");
+		return NULL;
+	}
+
+	total_size = (unsigned long) (section_hdr[4] + (section_hdr[5] << 8));
+
+	printk(KERN_INFO "microcode: size %u, total_size %u\n",
+		size, total_size);
+
+	if (total_size > size || total_size > UCODE_MAX_SIZE) {
+		printk(KERN_ERR "microcode: error! Bad data in microcode data file\n");
+		return NULL;
+	}
+
+	mc = vmalloc(UCODE_MAX_SIZE);
+	if (mc) {
+		memset(mc, 0, UCODE_MAX_SIZE);
+		if (get_ucode_data(mc, buf + UCODE_CONTAINER_SECTION_HDR, total_size)) {
+			vfree(mc);
+			mc = NULL;
+		} else
+			*mc_size = total_size + UCODE_CONTAINER_SECTION_HDR;
+	}
+#undef UCODE_CONTAINER_SECTION_HDR
+	return mc;
+}
+
+
+static int install_equiv_cpu_table(u8 *buf,
+		int (*get_ucode_data)(void *, const void *, size_t))
+{
+#define UCODE_CONTAINER_HEADER_SIZE	12
+	u8 *container_hdr[UCODE_CONTAINER_HEADER_SIZE];
+	unsigned int *buf_pos = (unsigned int *)container_hdr;
+	unsigned long size;
+
+	if (get_ucode_data(&container_hdr, buf, UCODE_CONTAINER_HEADER_SIZE))
+		return 0;
+
+	size = buf_pos[2];
+
+	if (buf_pos[1] != UCODE_EQUIV_CPU_TABLE_TYPE || !size) {
+		printk(KERN_ERR "microcode: error! "
+		       "Wrong microcode equivalnet cpu table\n");
+		return 0;
+	}
+
+	equiv_cpu_table = (struct equiv_cpu_entry *) vmalloc(size);
+	if (!equiv_cpu_table) {
+		printk(KERN_ERR "microcode: error, can't allocate memory for equiv CPU table\n");
+		return 0;
+	}
+
+	buf += UCODE_CONTAINER_HEADER_SIZE;
+	if (get_ucode_data(equiv_cpu_table, buf, size)) {
+		vfree(equiv_cpu_table);
+		return 0;
+	}
+
+	return size + UCODE_CONTAINER_HEADER_SIZE; /* add header length */
+#undef UCODE_CONTAINER_HEADER_SIZE
+}
+
+static void free_equiv_cpu_table(void)
+{
+	if (equiv_cpu_table) {
+		vfree(equiv_cpu_table);
+		equiv_cpu_table = NULL;
+	}
+}
+
+static int generic_load_microcode(int cpu, void *data, size_t size,
+		int (*get_ucode_data)(void *, const void *, size_t))
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+	u8 *ucode_ptr = data, *new_mc = NULL, *mc;
+	int new_rev = uci->cpu_sig.rev;
+	unsigned int leftover;
+	unsigned long offset;
+
+	offset = install_equiv_cpu_table(ucode_ptr, get_ucode_data);
+	if (!offset) {
+		printk(KERN_ERR "microcode: installing equivalent cpu table failed\n");
+		return -EINVAL;
+	}
+
+	ucode_ptr += offset;
+	leftover = size - offset;
+
+	while (leftover) {
+		unsigned int uninitialized_var(mc_size);
+		struct microcode_header_amd *mc_header;
+
+		mc = get_next_ucode(ucode_ptr, leftover, get_ucode_data, &mc_size);
+		if (!mc)
+			break;
+
+		mc_header = (struct microcode_header_amd *)mc;
+		if (get_matching_microcode(cpu, mc, new_rev)) {
+			if (new_mc)
+				vfree(new_mc);
+			new_rev = mc_header->patch_id;
+			new_mc  = mc;
+		} else 
+			vfree(mc);
+
+		ucode_ptr += mc_size;
+		leftover  -= mc_size;
+	}
+
+	if (new_mc) {
+		if (!leftover) {
+			if (uci->mc)
+				vfree(uci->mc);
+			uci->mc = new_mc;
+			pr_debug("microcode: CPU%d found a matching microcode update with"
+				" version 0x%x (current=0x%x)\n",
+				cpu, new_rev, uci->cpu_sig.rev);
+		} else
+			vfree(new_mc);
+	}
+
+	free_equiv_cpu_table();
+
+	return (int)leftover;
+}
+
+static int get_ucode_fw(void *to, const void *from, size_t n)
+{
+	memcpy(to, from, n);
+	return 0;
+}
+
+static int request_microcode_fw(int cpu, struct device *device)
+{
+	const char *fw_name = "amd-ucode/microcode_amd.bin";
+	const struct firmware *firmware;
+	int ret;
+
+	/* We should bind the task to the CPU */
+	BUG_ON(cpu != raw_smp_processor_id());
+
+	ret = request_firmware(&firmware, fw_name, device);
+	if (ret) {
+		printk(KERN_ERR "microcode: ucode data file %s load failed\n", fw_name);
+		return ret;
+	}
+
+	ret = generic_load_microcode(cpu, (void*)firmware->data, firmware->size,
+			&get_ucode_fw);
+
+	release_firmware(firmware);
+
+	return ret;
+}
+
+static int request_microcode_user(int cpu, const void __user *buf, size_t size)
+{
+	printk(KERN_WARNING "microcode: AMD microcode update via /dev/cpu/microcode"
+			"is not supported\n");
+	return -1;
+}
+
+static void microcode_fini_cpu_amd(int cpu)
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+	vfree(uci->mc);
+	uci->mc = NULL;
+}
+
+static struct microcode_ops microcode_amd_ops = {
+	.request_microcode_user           = request_microcode_user,
+	.request_microcode_fw             = request_microcode_fw,
+	.collect_cpu_info                 = collect_cpu_info_amd,
+	.apply_microcode                  = apply_microcode_amd,
+	.microcode_fini_cpu               = microcode_fini_cpu_amd,
+};
+
+struct microcode_ops * __init init_amd_microcode(void)
+{
+	return &microcode_amd_ops;
+}
diff --git a/arch/x86/kernel/microcode_core.c b/arch/x86/kernel/microcode_core.c
new file mode 100644
index 000000000000..936d8d55f230
--- /dev/null
+++ b/arch/x86/kernel/microcode_core.c
@@ -0,0 +1,508 @@
+/*
+ *	Intel CPU Microcode Update Driver for Linux
+ *
+ *	Copyright (C) 2000-2006 Tigran Aivazian <tigran@aivazian.fsnet.co.uk>
+ *		      2006	Shaohua Li <shaohua.li@intel.com>
+ *
+ *	This driver allows to upgrade microcode on Intel processors
+ *	belonging to IA-32 family - PentiumPro, Pentium II,
+ *	Pentium III, Xeon, Pentium 4, etc.
+ *
+ *	Reference: Section 8.11 of Volume 3a, IA-32 Intel? Architecture
+ *	Software Developer's Manual
+ *	Order Number 253668 or free download from:
+ *
+ *	http://developer.intel.com/design/pentium4/manuals/253668.htm
+ *
+ *	For more information, go to http://www.urbanmyth.org/microcode
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License
+ *	as published by the Free Software Foundation; either version
+ *	2 of the License, or (at your option) any later version.
+ *
+ *	1.0	16 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ *		Initial release.
+ *	1.01	18 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ *		Added read() support + cleanups.
+ *	1.02	21 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ *		Added 'device trimming' support. open(O_WRONLY) zeroes
+ *		and frees the saved copy of applied microcode.
+ *	1.03	29 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ *		Made to use devfs (/dev/cpu/microcode) + cleanups.
+ *	1.04	06 Jun 2000, Simon Trimmer <simon@veritas.com>
+ *		Added misc device support (now uses both devfs and misc).
+ *		Added MICROCODE_IOCFREE ioctl to clear memory.
+ *	1.05	09 Jun 2000, Simon Trimmer <simon@veritas.com>
+ *		Messages for error cases (non Intel & no suitable microcode).
+ *	1.06	03 Aug 2000, Tigran Aivazian <tigran@veritas.com>
+ *		Removed ->release(). Removed exclusive open and status bitmap.
+ *		Added microcode_rwsem to serialize read()/write()/ioctl().
+ *		Removed global kernel lock usage.
+ *	1.07	07 Sep 2000, Tigran Aivazian <tigran@veritas.com>
+ *		Write 0 to 0x8B msr and then cpuid before reading revision,
+ *		so that it works even if there were no update done by the
+ *		BIOS. Otherwise, reading from 0x8B gives junk (which happened
+ *		to be 0 on my machine which is why it worked even when I
+ *		disabled update by the BIOS)
+ *		Thanks to Eric W. Biederman <ebiederman@lnxi.com> for the fix.
+ *	1.08	11 Dec 2000, Richard Schaal <richard.schaal@intel.com> and
+ *			     Tigran Aivazian <tigran@veritas.com>
+ *		Intel Pentium 4 processor support and bugfixes.
+ *	1.09	30 Oct 2001, Tigran Aivazian <tigran@veritas.com>
+ *		Bugfix for HT (Hyper-Threading) enabled processors
+ *		whereby processor resources are shared by all logical processors
+ *		in a single CPU package.
+ *	1.10	28 Feb 2002 Asit K Mallick <asit.k.mallick@intel.com> and
+ *		Tigran Aivazian <tigran@veritas.com>,
+ *		Serialize updates as required on HT processors due to
+ *		speculative nature of implementation.
+ *	1.11	22 Mar 2002 Tigran Aivazian <tigran@veritas.com>
+ *		Fix the panic when writing zero-length microcode chunk.
+ *	1.12	29 Sep 2003 Nitin Kamble <nitin.a.kamble@intel.com>,
+ *		Jun Nakajima <jun.nakajima@intel.com>
+ *		Support for the microcode updates in the new format.
+ *	1.13	10 Oct 2003 Tigran Aivazian <tigran@veritas.com>
+ *		Removed ->read() method and obsoleted MICROCODE_IOCFREE ioctl
+ *		because we no longer hold a copy of applied microcode
+ *		in kernel memory.
+ *	1.14	25 Jun 2004 Tigran Aivazian <tigran@veritas.com>
+ *		Fix sigmatch() macro to handle old CPUs with pf == 0.
+ *		Thanks to Stuart Swales for pointing out this bug.
+ */
+#include <linux/capability.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/smp_lock.h>
+#include <linux/cpumask.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/miscdevice.h>
+#include <linux/spinlock.h>
+#include <linux/mm.h>
+#include <linux/fs.h>
+#include <linux/mutex.h>
+#include <linux/cpu.h>
+#include <linux/firmware.h>
+#include <linux/platform_device.h>
+
+#include <asm/msr.h>
+#include <asm/uaccess.h>
+#include <asm/processor.h>
+#include <asm/microcode.h>
+
+MODULE_DESCRIPTION("Microcode Update Driver");
+MODULE_AUTHOR("Tigran Aivazian <tigran@aivazian.fsnet.co.uk>");
+MODULE_LICENSE("GPL");
+
+#define MICROCODE_VERSION 	"2.00"
+
+struct microcode_ops *microcode_ops;
+
+/* no concurrent ->write()s are allowed on /dev/cpu/microcode */
+static DEFINE_MUTEX(microcode_mutex);
+
+struct ucode_cpu_info ucode_cpu_info[NR_CPUS];
+EXPORT_SYMBOL_GPL(ucode_cpu_info);
+
+#ifdef CONFIG_MICROCODE_OLD_INTERFACE
+static int do_microcode_update(const void __user *buf, size_t size)
+{
+	cpumask_t old;
+	int error = 0;
+	int cpu;
+
+	old = current->cpus_allowed;
+
+	for_each_online_cpu(cpu) {
+		struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+		if (!uci->valid)
+			continue;
+
+		set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+		error = microcode_ops->request_microcode_user(cpu, buf, size);
+		if (error < 0)
+			goto out;
+		if (!error)
+			microcode_ops->apply_microcode(cpu);
+	}
+out:
+	set_cpus_allowed_ptr(current, &old);
+	return error;
+}
+
+static int microcode_open(struct inode *unused1, struct file *unused2)
+{
+	cycle_kernel_lock();
+	return capable(CAP_SYS_RAWIO) ? 0 : -EPERM;
+}
+
+static ssize_t microcode_write(struct file *file, const char __user *buf,
+			       size_t len, loff_t *ppos)
+{
+	ssize_t ret;
+
+	if ((len >> PAGE_SHIFT) > num_physpages) {
+		printk(KERN_ERR "microcode: too much data (max %ld pages)\n",
+		       num_physpages);
+		return -EINVAL;
+	}
+
+	get_online_cpus();
+	mutex_lock(&microcode_mutex);
+
+	ret = do_microcode_update(buf, len);
+	if (!ret)
+		ret = (ssize_t)len;
+
+	mutex_unlock(&microcode_mutex);
+	put_online_cpus();
+
+	return ret;
+}
+
+static const struct file_operations microcode_fops = {
+	.owner		= THIS_MODULE,
+	.write		= microcode_write,
+	.open		= microcode_open,
+};
+
+static struct miscdevice microcode_dev = {
+	.minor		= MICROCODE_MINOR,
+	.name		= "microcode",
+	.fops		= &microcode_fops,
+};
+
+static int __init microcode_dev_init(void)
+{
+	int error;
+
+	error = misc_register(&microcode_dev);
+	if (error) {
+		printk(KERN_ERR
+			"microcode: can't misc_register on minor=%d\n",
+			MICROCODE_MINOR);
+		return error;
+	}
+
+	return 0;
+}
+
+static void microcode_dev_exit(void)
+{
+	misc_deregister(&microcode_dev);
+}
+
+MODULE_ALIAS_MISCDEV(MICROCODE_MINOR);
+#else
+#define microcode_dev_init() 0
+#define microcode_dev_exit() do { } while (0)
+#endif
+
+/* fake device for request_firmware */
+struct platform_device *microcode_pdev;
+
+static ssize_t reload_store(struct sys_device *dev,
+			    struct sysdev_attribute *attr,
+			    const char *buf, size_t sz)
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + dev->id;
+	char *end;
+	unsigned long val = simple_strtoul(buf, &end, 0);
+	int err = 0;
+	int cpu = dev->id;
+
+	if (end == buf)
+		return -EINVAL;
+	if (val == 1) {
+		cpumask_t old = current->cpus_allowed;
+
+		get_online_cpus();
+		if (cpu_online(cpu)) {
+			set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+			mutex_lock(&microcode_mutex);
+			if (uci->valid) {
+				err = microcode_ops->request_microcode_fw(cpu,
+						&microcode_pdev->dev);
+				if (!err)
+					microcode_ops->apply_microcode(cpu);
+			}
+			mutex_unlock(&microcode_mutex);
+			set_cpus_allowed_ptr(current, &old);
+		}
+		put_online_cpus();
+	}
+	if (err)
+		return err;
+	return sz;
+}
+
+static ssize_t version_show(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + dev->id;
+
+	return sprintf(buf, "0x%x\n", uci->cpu_sig.rev);
+}
+
+static ssize_t pf_show(struct sys_device *dev,
+			struct sysdev_attribute *attr, char *buf)
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + dev->id;
+
+	return sprintf(buf, "0x%x\n", uci->cpu_sig.pf);
+}
+
+static SYSDEV_ATTR(reload, 0200, NULL, reload_store);
+static SYSDEV_ATTR(version, 0400, version_show, NULL);
+static SYSDEV_ATTR(processor_flags, 0400, pf_show, NULL);
+
+static struct attribute *mc_default_attrs[] = {
+	&attr_reload.attr,
+	&attr_version.attr,
+	&attr_processor_flags.attr,
+	NULL
+};
+
+static struct attribute_group mc_attr_group = {
+	.attrs = mc_default_attrs,
+	.name = "microcode",
+};
+
+static void microcode_fini_cpu(int cpu)
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+	mutex_lock(&microcode_mutex);
+	microcode_ops->microcode_fini_cpu(cpu);
+	uci->valid = 0;
+	mutex_unlock(&microcode_mutex);
+}
+
+static void collect_cpu_info(int cpu)
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+	memset(uci, 0, sizeof(*uci));
+	if (!microcode_ops->collect_cpu_info(cpu, &uci->cpu_sig))
+		uci->valid = 1;
+}
+
+static int microcode_resume_cpu(int cpu)
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+	struct cpu_signature nsig;
+
+	pr_debug("microcode: CPU%d resumed\n", cpu);
+
+	if (!uci->mc)
+		return 1;
+
+	/*
+	 * Let's verify that the 'cached' ucode does belong
+	 * to this cpu (a bit of paranoia):
+	 */
+	if (microcode_ops->collect_cpu_info(cpu, &nsig)) {
+		microcode_fini_cpu(cpu);
+		return -1;
+	}
+
+	if (memcmp(&nsig, &uci->cpu_sig, sizeof(nsig))) {
+		microcode_fini_cpu(cpu);
+		/* Should we look for a new ucode here? */
+		return 1;
+	}
+
+	return 0;
+}
+
+void microcode_update_cpu(int cpu)
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+	int err = 0;
+
+	/*
+	 * Check if the system resume is in progress (uci->valid != NULL),
+	 * otherwise just request a firmware:
+	 */
+	if (uci->valid) {
+		err = microcode_resume_cpu(cpu);
+	} else {	
+		collect_cpu_info(cpu);
+		if (uci->valid && system_state == SYSTEM_RUNNING)
+			err = microcode_ops->request_microcode_fw(cpu,
+					&microcode_pdev->dev);
+	}
+	if (!err)
+		microcode_ops->apply_microcode(cpu);
+}
+
+static void microcode_init_cpu(int cpu)
+{
+	cpumask_t old = current->cpus_allowed;
+
+	set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
+	/* We should bind the task to the CPU */
+	BUG_ON(raw_smp_processor_id() != cpu);
+
+	mutex_lock(&microcode_mutex);
+	microcode_update_cpu(cpu);
+	mutex_unlock(&microcode_mutex);
+
+	set_cpus_allowed_ptr(current, &old);
+}
+
+static int mc_sysdev_add(struct sys_device *sys_dev)
+{
+	int err, cpu = sys_dev->id;
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+	if (!cpu_online(cpu))
+		return 0;
+
+	pr_debug("microcode: CPU%d added\n", cpu);
+	memset(uci, 0, sizeof(*uci));
+
+	err = sysfs_create_group(&sys_dev->kobj, &mc_attr_group);
+	if (err)
+		return err;
+
+	microcode_init_cpu(cpu);
+	return 0;
+}
+
+static int mc_sysdev_remove(struct sys_device *sys_dev)
+{
+	int cpu = sys_dev->id;
+
+	if (!cpu_online(cpu))
+		return 0;
+
+	pr_debug("microcode: CPU%d removed\n", cpu);
+	microcode_fini_cpu(cpu);
+	sysfs_remove_group(&sys_dev->kobj, &mc_attr_group);
+	return 0;
+}
+
+static int mc_sysdev_resume(struct sys_device *dev)
+{
+	int cpu = dev->id;
+
+	if (!cpu_online(cpu))
+		return 0;
+
+	/* only CPU 0 will apply ucode here */
+	microcode_update_cpu(0);
+	return 0;
+}
+
+static struct sysdev_driver mc_sysdev_driver = {
+	.add = mc_sysdev_add,
+	.remove = mc_sysdev_remove,
+	.resume = mc_sysdev_resume,
+};
+
+static __cpuinit int
+mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu)
+{
+	unsigned int cpu = (unsigned long)hcpu;
+	struct sys_device *sys_dev;
+
+	sys_dev = get_cpu_sysdev(cpu);
+	switch (action) {
+	case CPU_ONLINE:
+	case CPU_ONLINE_FROZEN:
+		microcode_init_cpu(cpu);
+	case CPU_DOWN_FAILED:
+	case CPU_DOWN_FAILED_FROZEN:
+		pr_debug("microcode: CPU%d added\n", cpu);
+		if (sysfs_create_group(&sys_dev->kobj, &mc_attr_group))
+			printk(KERN_ERR "microcode: Failed to create the sysfs "
+				"group for CPU%d\n", cpu);
+		break;
+	case CPU_DOWN_PREPARE:
+	case CPU_DOWN_PREPARE_FROZEN:
+		/* Suspend is in progress, only remove the interface */
+		sysfs_remove_group(&sys_dev->kobj, &mc_attr_group);
+		pr_debug("microcode: CPU%d removed\n", cpu);
+		break;
+	case CPU_DEAD:
+	case CPU_UP_CANCELED_FROZEN:
+		/* The CPU refused to come up during a system resume */
+		microcode_fini_cpu(cpu);
+		break;
+	}
+	return NOTIFY_OK;
+}
+
+static struct notifier_block __refdata mc_cpu_notifier = {
+	.notifier_call = mc_cpu_callback,
+};
+
+static int __init microcode_init(void)
+{
+	struct cpuinfo_x86 *c = &cpu_data(0);
+	int error;
+
+	if (c->x86_vendor == X86_VENDOR_INTEL)
+		microcode_ops = init_intel_microcode();
+	else if (c->x86_vendor == X86_VENDOR_AMD)
+		microcode_ops = init_amd_microcode();
+
+	if (!microcode_ops) {
+		printk(KERN_ERR "microcode: no support for this CPU vendor\n");
+		return -ENODEV;
+	}
+
+	error = microcode_dev_init();
+	if (error)
+		return error;
+	microcode_pdev = platform_device_register_simple("microcode", -1,
+							 NULL, 0);
+	if (IS_ERR(microcode_pdev)) {
+		microcode_dev_exit();
+		return PTR_ERR(microcode_pdev);
+	}
+
+	get_online_cpus();
+	error = sysdev_driver_register(&cpu_sysdev_class, &mc_sysdev_driver);
+	put_online_cpus();
+	if (error) {
+		microcode_dev_exit();
+		platform_device_unregister(microcode_pdev);
+		return error;
+	}
+
+	register_hotcpu_notifier(&mc_cpu_notifier);
+
+	printk(KERN_INFO
+	       "Microcode Update Driver: v" MICROCODE_VERSION
+	       " <tigran@aivazian.fsnet.co.uk>"
+	       " <peter.oruba@amd.com>\n");
+
+	return 0;
+}
+
+static void __exit microcode_exit(void)
+{
+	microcode_dev_exit();
+
+	unregister_hotcpu_notifier(&mc_cpu_notifier);
+
+	get_online_cpus();
+	sysdev_driver_unregister(&cpu_sysdev_class, &mc_sysdev_driver);
+	put_online_cpus();
+
+	platform_device_unregister(microcode_pdev);
+
+	microcode_ops = NULL;
+
+	printk(KERN_INFO
+	       "Microcode Update Driver: v" MICROCODE_VERSION " removed.\n");
+}
+
+module_init(microcode_init);
+module_exit(microcode_exit);
diff --git a/arch/x86/kernel/microcode_intel.c b/arch/x86/kernel/microcode_intel.c
new file mode 100644
index 000000000000..622dc4a21784
--- /dev/null
+++ b/arch/x86/kernel/microcode_intel.c
@@ -0,0 +1,480 @@
+/*
+ *	Intel CPU Microcode Update Driver for Linux
+ *
+ *	Copyright (C) 2000-2006 Tigran Aivazian <tigran@aivazian.fsnet.co.uk>
+ *		      2006	Shaohua Li <shaohua.li@intel.com>
+ *
+ *	This driver allows to upgrade microcode on Intel processors
+ *	belonging to IA-32 family - PentiumPro, Pentium II,
+ *	Pentium III, Xeon, Pentium 4, etc.
+ *
+ *	Reference: Section 8.11 of Volume 3a, IA-32 Intel? Architecture
+ *	Software Developer's Manual
+ *	Order Number 253668 or free download from:
+ *
+ *	http://developer.intel.com/design/pentium4/manuals/253668.htm
+ *
+ *	For more information, go to http://www.urbanmyth.org/microcode
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License
+ *	as published by the Free Software Foundation; either version
+ *	2 of the License, or (at your option) any later version.
+ *
+ *	1.0	16 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ *		Initial release.
+ *	1.01	18 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ *		Added read() support + cleanups.
+ *	1.02	21 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ *		Added 'device trimming' support. open(O_WRONLY) zeroes
+ *		and frees the saved copy of applied microcode.
+ *	1.03	29 Feb 2000, Tigran Aivazian <tigran@sco.com>
+ *		Made to use devfs (/dev/cpu/microcode) + cleanups.
+ *	1.04	06 Jun 2000, Simon Trimmer <simon@veritas.com>
+ *		Added misc device support (now uses both devfs and misc).
+ *		Added MICROCODE_IOCFREE ioctl to clear memory.
+ *	1.05	09 Jun 2000, Simon Trimmer <simon@veritas.com>
+ *		Messages for error cases (non Intel & no suitable microcode).
+ *	1.06	03 Aug 2000, Tigran Aivazian <tigran@veritas.com>
+ *		Removed ->release(). Removed exclusive open and status bitmap.
+ *		Added microcode_rwsem to serialize read()/write()/ioctl().
+ *		Removed global kernel lock usage.
+ *	1.07	07 Sep 2000, Tigran Aivazian <tigran@veritas.com>
+ *		Write 0 to 0x8B msr and then cpuid before reading revision,
+ *		so that it works even if there were no update done by the
+ *		BIOS. Otherwise, reading from 0x8B gives junk (which happened
+ *		to be 0 on my machine which is why it worked even when I
+ *		disabled update by the BIOS)
+ *		Thanks to Eric W. Biederman <ebiederman@lnxi.com> for the fix.
+ *	1.08	11 Dec 2000, Richard Schaal <richard.schaal@intel.com> and
+ *			     Tigran Aivazian <tigran@veritas.com>
+ *		Intel Pentium 4 processor support and bugfixes.
+ *	1.09	30 Oct 2001, Tigran Aivazian <tigran@veritas.com>
+ *		Bugfix for HT (Hyper-Threading) enabled processors
+ *		whereby processor resources are shared by all logical processors
+ *		in a single CPU package.
+ *	1.10	28 Feb 2002 Asit K Mallick <asit.k.mallick@intel.com> and
+ *		Tigran Aivazian <tigran@veritas.com>,
+ *		Serialize updates as required on HT processors due to
+ *		speculative nature of implementation.
+ *	1.11	22 Mar 2002 Tigran Aivazian <tigran@veritas.com>
+ *		Fix the panic when writing zero-length microcode chunk.
+ *	1.12	29 Sep 2003 Nitin Kamble <nitin.a.kamble@intel.com>,
+ *		Jun Nakajima <jun.nakajima@intel.com>
+ *		Support for the microcode updates in the new format.
+ *	1.13	10 Oct 2003 Tigran Aivazian <tigran@veritas.com>
+ *		Removed ->read() method and obsoleted MICROCODE_IOCFREE ioctl
+ *		because we no longer hold a copy of applied microcode
+ *		in kernel memory.
+ *	1.14	25 Jun 2004 Tigran Aivazian <tigran@veritas.com>
+ *		Fix sigmatch() macro to handle old CPUs with pf == 0.
+ *		Thanks to Stuart Swales for pointing out this bug.
+ */
+#include <linux/capability.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/smp_lock.h>
+#include <linux/cpumask.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/vmalloc.h>
+#include <linux/miscdevice.h>
+#include <linux/spinlock.h>
+#include <linux/mm.h>
+#include <linux/fs.h>
+#include <linux/mutex.h>
+#include <linux/cpu.h>
+#include <linux/firmware.h>
+#include <linux/platform_device.h>
+
+#include <asm/msr.h>
+#include <asm/uaccess.h>
+#include <asm/processor.h>
+#include <asm/microcode.h>
+
+MODULE_DESCRIPTION("Microcode Update Driver");
+MODULE_AUTHOR("Tigran Aivazian <tigran@aivazian.fsnet.co.uk>");
+MODULE_LICENSE("GPL");
+
+struct microcode_header_intel {
+	unsigned int            hdrver;
+	unsigned int            rev;
+	unsigned int            date;
+	unsigned int            sig;
+	unsigned int            cksum;
+	unsigned int            ldrver;
+	unsigned int            pf;
+	unsigned int            datasize;
+	unsigned int            totalsize;
+	unsigned int            reserved[3];
+};
+
+struct microcode_intel {
+	struct microcode_header_intel hdr;
+	unsigned int            bits[0];
+};
+
+/* microcode format is extended from prescott processors */
+struct extended_signature {
+	unsigned int            sig;
+	unsigned int            pf;
+	unsigned int            cksum;
+};
+
+struct extended_sigtable {
+	unsigned int            count;
+	unsigned int            cksum;
+	unsigned int            reserved[3];
+	struct extended_signature sigs[0];
+};
+
+#define DEFAULT_UCODE_DATASIZE 	(2000)
+#define MC_HEADER_SIZE		(sizeof(struct microcode_header_intel))
+#define DEFAULT_UCODE_TOTALSIZE (DEFAULT_UCODE_DATASIZE + MC_HEADER_SIZE)
+#define EXT_HEADER_SIZE		(sizeof(struct extended_sigtable))
+#define EXT_SIGNATURE_SIZE	(sizeof(struct extended_signature))
+#define DWSIZE			(sizeof(u32))
+#define get_totalsize(mc) \
+	(((struct microcode_intel *)mc)->hdr.totalsize ? \
+	 ((struct microcode_intel *)mc)->hdr.totalsize : \
+	 DEFAULT_UCODE_TOTALSIZE)
+
+#define get_datasize(mc) \
+	(((struct microcode_intel *)mc)->hdr.datasize ? \
+	 ((struct microcode_intel *)mc)->hdr.datasize : DEFAULT_UCODE_DATASIZE)
+
+#define sigmatch(s1, s2, p1, p2) \
+	(((s1) == (s2)) && (((p1) & (p2)) || (((p1) == 0) && ((p2) == 0))))
+
+#define exttable_size(et) ((et)->count * EXT_SIGNATURE_SIZE + EXT_HEADER_SIZE)
+
+/* serialize access to the physical write to MSR 0x79 */
+static DEFINE_SPINLOCK(microcode_update_lock);
+
+static int collect_cpu_info(int cpu_num, struct cpu_signature *csig)
+{
+	struct cpuinfo_x86 *c = &cpu_data(cpu_num);
+	unsigned int val[2];
+
+	memset(csig, 0, sizeof(*csig));
+
+	if (c->x86_vendor != X86_VENDOR_INTEL || c->x86 < 6 ||
+	    cpu_has(c, X86_FEATURE_IA64)) {
+		printk(KERN_ERR "microcode: CPU%d not a capable Intel "
+			"processor\n", cpu_num);
+		return -1;
+	}
+
+	csig->sig = cpuid_eax(0x00000001);
+
+	if ((c->x86_model >= 5) || (c->x86 > 6)) {
+		/* get processor flags from MSR 0x17 */
+		rdmsr(MSR_IA32_PLATFORM_ID, val[0], val[1]);
+		csig->pf = 1 << ((val[1] >> 18) & 7);
+	}
+
+	wrmsr(MSR_IA32_UCODE_REV, 0, 0);
+	/* see notes above for revision 1.07.  Apparent chip bug */
+	sync_core();
+	/* get the current revision from MSR 0x8B */
+	rdmsr(MSR_IA32_UCODE_REV, val[0], csig->rev);
+	pr_debug("microcode: collect_cpu_info : sig=0x%x, pf=0x%x, rev=0x%x\n",
+			csig->sig, csig->pf, csig->rev);
+
+	return 0;
+}
+
+static inline int update_match_cpu(struct cpu_signature *csig, int sig, int pf)
+{
+	return (!sigmatch(sig, csig->sig, pf, csig->pf)) ? 0 : 1;
+}
+
+static inline int 
+update_match_revision(struct microcode_header_intel *mc_header,	int rev)
+{
+	return (mc_header->rev <= rev) ? 0 : 1;
+}
+
+static int microcode_sanity_check(void *mc)
+{
+	struct microcode_header_intel *mc_header = mc;
+	struct extended_sigtable *ext_header = NULL;
+	struct extended_signature *ext_sig;
+	unsigned long total_size, data_size, ext_table_size;
+	int sum, orig_sum, ext_sigcount = 0, i;
+
+	total_size = get_totalsize(mc_header);
+	data_size = get_datasize(mc_header);
+	if (data_size + MC_HEADER_SIZE > total_size) {
+		printk(KERN_ERR "microcode: error! "
+			"Bad data size in microcode data file\n");
+		return -EINVAL;
+	}
+
+	if (mc_header->ldrver != 1 || mc_header->hdrver != 1) {
+		printk(KERN_ERR "microcode: error! "
+			"Unknown microcode update format\n");
+		return -EINVAL;
+	}
+	ext_table_size = total_size - (MC_HEADER_SIZE + data_size);
+	if (ext_table_size) {
+		if ((ext_table_size < EXT_HEADER_SIZE)
+		 || ((ext_table_size - EXT_HEADER_SIZE) % EXT_SIGNATURE_SIZE)) {
+			printk(KERN_ERR "microcode: error! "
+				"Small exttable size in microcode data file\n");
+			return -EINVAL;
+		}
+		ext_header = mc + MC_HEADER_SIZE + data_size;
+		if (ext_table_size != exttable_size(ext_header)) {
+			printk(KERN_ERR "microcode: error! "
+				"Bad exttable size in microcode data file\n");
+			return -EFAULT;
+		}
+		ext_sigcount = ext_header->count;
+	}
+
+	/* check extended table checksum */
+	if (ext_table_size) {
+		int ext_table_sum = 0;
+		int *ext_tablep = (int *)ext_header;
+
+		i = ext_table_size / DWSIZE;
+		while (i--)
+			ext_table_sum += ext_tablep[i];
+		if (ext_table_sum) {
+			printk(KERN_WARNING "microcode: aborting, "
+				"bad extended signature table checksum\n");
+			return -EINVAL;
+		}
+	}
+
+	/* calculate the checksum */
+	orig_sum = 0;
+	i = (MC_HEADER_SIZE + data_size) / DWSIZE;
+	while (i--)
+		orig_sum += ((int *)mc)[i];
+	if (orig_sum) {
+		printk(KERN_ERR "microcode: aborting, bad checksum\n");
+		return -EINVAL;
+	}
+	if (!ext_table_size)
+		return 0;
+	/* check extended signature checksum */
+	for (i = 0; i < ext_sigcount; i++) {
+		ext_sig = (void *)ext_header + EXT_HEADER_SIZE +
+			  EXT_SIGNATURE_SIZE * i;
+		sum = orig_sum
+			- (mc_header->sig + mc_header->pf + mc_header->cksum)
+			+ (ext_sig->sig + ext_sig->pf + ext_sig->cksum);
+		if (sum) {
+			printk(KERN_ERR "microcode: aborting, bad checksum\n");
+			return -EINVAL;
+		}
+	}
+	return 0;
+}
+
+/*
+ * return 0 - no update found
+ * return 1 - found update
+ */
+static int
+get_matching_microcode(struct cpu_signature *cpu_sig, void *mc, int rev)
+{
+	struct microcode_header_intel *mc_header = mc;
+	struct extended_sigtable *ext_header;
+	unsigned long total_size = get_totalsize(mc_header);
+	int ext_sigcount, i;
+	struct extended_signature *ext_sig;
+
+	if (!update_match_revision(mc_header, rev))
+		return 0;
+
+	if (update_match_cpu(cpu_sig, mc_header->sig, mc_header->pf))
+		return 1;
+
+	/* Look for ext. headers: */
+	if (total_size <= get_datasize(mc_header) + MC_HEADER_SIZE)
+		return 0;
+
+	ext_header = mc + get_datasize(mc_header) + MC_HEADER_SIZE;
+	ext_sigcount = ext_header->count;
+	ext_sig = (void *)ext_header + EXT_HEADER_SIZE;
+
+	for (i = 0; i < ext_sigcount; i++) {
+		if (update_match_cpu(cpu_sig, ext_sig->sig, ext_sig->pf))
+			return 1;
+		ext_sig++;
+	}
+	return 0;
+}
+
+static void apply_microcode(int cpu)
+{
+	unsigned long flags;
+	unsigned int val[2];
+	int cpu_num = raw_smp_processor_id();
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+	struct microcode_intel *mc_intel = uci->mc;
+
+	/* We should bind the task to the CPU */
+	BUG_ON(cpu_num != cpu);
+
+	if (mc_intel == NULL)
+		return;
+
+	/* serialize access to the physical write to MSR 0x79 */
+	spin_lock_irqsave(&microcode_update_lock, flags);
+
+	/* write microcode via MSR 0x79 */
+	wrmsr(MSR_IA32_UCODE_WRITE,
+	      (unsigned long) mc_intel->bits,
+	      (unsigned long) mc_intel->bits >> 16 >> 16);
+	wrmsr(MSR_IA32_UCODE_REV, 0, 0);
+
+	/* see notes above for revision 1.07.  Apparent chip bug */
+	sync_core();
+
+	/* get the current revision from MSR 0x8B */
+	rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]);
+
+	spin_unlock_irqrestore(&microcode_update_lock, flags);
+	if (val[1] != mc_intel->hdr.rev) {
+		printk(KERN_ERR "microcode: CPU%d update from revision "
+			"0x%x to 0x%x failed\n", cpu_num, uci->cpu_sig.rev, val[1]);
+		return;
+	}
+	printk(KERN_INFO "microcode: CPU%d updated from revision "
+	       "0x%x to 0x%x, date = %04x-%02x-%02x \n",
+		cpu_num, uci->cpu_sig.rev, val[1],
+		mc_intel->hdr.date & 0xffff,
+		mc_intel->hdr.date >> 24,
+		(mc_intel->hdr.date >> 16) & 0xff);
+	uci->cpu_sig.rev = val[1];
+}
+
+static int generic_load_microcode(int cpu, void *data, size_t size,
+		int (*get_ucode_data)(void *, const void *, size_t))
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+	u8 *ucode_ptr = data, *new_mc = NULL, *mc;
+	int new_rev = uci->cpu_sig.rev;
+	unsigned int leftover = size;
+
+	while (leftover) {
+		struct microcode_header_intel mc_header;
+		unsigned int mc_size;
+
+		if (get_ucode_data(&mc_header, ucode_ptr, sizeof(mc_header)))
+			break;
+
+		mc_size = get_totalsize(&mc_header);
+		if (!mc_size || mc_size > leftover) {
+			printk(KERN_ERR "microcode: error!"
+					"Bad data in microcode data file\n");
+			break;
+		}
+
+		mc = vmalloc(mc_size);
+		if (!mc)
+			break;
+
+		if (get_ucode_data(mc, ucode_ptr, mc_size) ||
+		    microcode_sanity_check(mc) < 0) {
+			vfree(mc);
+			break;
+		}
+
+		if (get_matching_microcode(&uci->cpu_sig, mc, new_rev)) {
+			if (new_mc)
+				vfree(new_mc);
+			new_rev = mc_header.rev;
+			new_mc  = mc;
+		} else
+			vfree(mc);
+
+		ucode_ptr += mc_size;
+		leftover  -= mc_size;
+	}
+
+	if (new_mc) {
+		if (!leftover) {
+			if (uci->mc)
+				vfree(uci->mc);
+			uci->mc = (struct microcode_intel *)new_mc;
+			pr_debug("microcode: CPU%d found a matching microcode update with"
+				 " version 0x%x (current=0x%x)\n",
+				cpu, new_rev, uci->cpu_sig.rev);
+		} else
+			vfree(new_mc);
+	}
+
+	return (int)leftover;
+}
+
+static int get_ucode_fw(void *to, const void *from, size_t n)
+{
+	memcpy(to, from, n);
+	return 0;
+}
+
+static int request_microcode_fw(int cpu, struct device *device)
+{
+	char name[30];
+	struct cpuinfo_x86 *c = &cpu_data(cpu);
+	const struct firmware *firmware;
+	int ret;
+
+	/* We should bind the task to the CPU */
+	BUG_ON(cpu != raw_smp_processor_id());
+	sprintf(name, "intel-ucode/%02x-%02x-%02x",
+		c->x86, c->x86_model, c->x86_mask);
+	ret = request_firmware(&firmware, name, device);
+	if (ret) {
+		pr_debug("microcode: data file %s load failed\n", name);
+		return ret;
+	}
+
+	ret = generic_load_microcode(cpu, (void*)firmware->data, firmware->size,
+			&get_ucode_fw);
+
+	release_firmware(firmware);
+
+	return ret;
+}
+
+static int get_ucode_user(void *to, const void *from, size_t n)
+{
+	return copy_from_user(to, from, n);
+}
+
+static int request_microcode_user(int cpu, const void __user *buf, size_t size)
+{
+	/* We should bind the task to the CPU */
+	BUG_ON(cpu != raw_smp_processor_id());
+
+	return generic_load_microcode(cpu, (void*)buf, size, &get_ucode_user);
+}
+
+static void microcode_fini_cpu(int cpu)
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+	vfree(uci->mc);
+	uci->mc = NULL;
+}
+
+struct microcode_ops microcode_intel_ops = {
+	.request_microcode_user		  = request_microcode_user,
+	.request_microcode_fw             = request_microcode_fw,
+	.collect_cpu_info                 = collect_cpu_info,
+	.apply_microcode                  = apply_microcode,
+	.microcode_fini_cpu               = microcode_fini_cpu,
+};
+
+struct microcode_ops * __init init_intel_microcode(void)
+{
+	return &microcode_intel_ops;
+}
+
diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c
index b3fb430725cb..f98f4e1dba09 100644
--- a/arch/x86/kernel/mpparse.c
+++ b/arch/x86/kernel/mpparse.c
@@ -397,7 +397,9 @@ static int __init smp_read_mpc(struct mp_config_table *mpc, unsigned early)
        generic_bigsmp_probe();
 #endif
 
+#ifdef CONFIG_X86_32
 	setup_apic_routing();
+#endif
 	if (!num_processors)
 		printk(KERN_ERR "MPTABLE: no processors registered!\n");
 	return num_processors;
diff --git a/arch/x86/kernel/numaq_32.c b/arch/x86/kernel/numaq_32.c
index eecc8c18f010..4caff39078e0 100644
--- a/arch/x86/kernel/numaq_32.c
+++ b/arch/x86/kernel/numaq_32.c
@@ -229,6 +229,12 @@ static void __init smp_read_mpc_oem(struct mp_config_oemtable *oemtable,
 	}
 }
 
+static int __init numaq_setup_ioapic_ids(void)
+{
+	/* so can skip it */
+	return 1;
+}
+
 static struct x86_quirks numaq_x86_quirks __initdata = {
 	.arch_pre_time_init	= numaq_pre_time_init,
 	.arch_time_init		= NULL,
@@ -243,6 +249,7 @@ static struct x86_quirks numaq_x86_quirks __initdata = {
 	.mpc_oem_bus_info	= mpc_oem_bus_info,
 	.mpc_oem_pci_bus	= mpc_oem_pci_bus,
 	.smp_read_mpc_oem	= smp_read_mpc_oem,
+	.setup_ioapic_ids	= numaq_setup_ioapic_ids,
 };
 
 void numaq_mps_oem_check(struct mp_config_table *mpc, char *oem,
diff --git a/arch/x86/kernel/paravirt-spinlocks.c b/arch/x86/kernel/paravirt-spinlocks.c
new file mode 100644
index 000000000000..0e9f1982b1dd
--- /dev/null
+++ b/arch/x86/kernel/paravirt-spinlocks.c
@@ -0,0 +1,37 @@
+/*
+ * Split spinlock implementation out into its own file, so it can be
+ * compiled in a FTRACE-compatible way.
+ */
+#include <linux/spinlock.h>
+#include <linux/module.h>
+
+#include <asm/paravirt.h>
+
+static void default_spin_lock_flags(struct raw_spinlock *lock, unsigned long flags)
+{
+	__raw_spin_lock(lock);
+}
+
+struct pv_lock_ops pv_lock_ops = {
+#ifdef CONFIG_SMP
+	.spin_is_locked = __ticket_spin_is_locked,
+	.spin_is_contended = __ticket_spin_is_contended,
+
+	.spin_lock = __ticket_spin_lock,
+	.spin_lock_flags = default_spin_lock_flags,
+	.spin_trylock = __ticket_spin_trylock,
+	.spin_unlock = __ticket_spin_unlock,
+#endif
+};
+EXPORT_SYMBOL(pv_lock_ops);
+
+void __init paravirt_use_bytelocks(void)
+{
+#ifdef CONFIG_SMP
+	pv_lock_ops.spin_is_locked = __byte_spin_is_locked;
+	pv_lock_ops.spin_is_contended = __byte_spin_is_contended;
+	pv_lock_ops.spin_lock = __byte_spin_lock;
+	pv_lock_ops.spin_trylock = __byte_spin_trylock;
+	pv_lock_ops.spin_unlock = __byte_spin_unlock;
+#endif
+}
diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
index e2f43768723a..e4c8fb608873 100644
--- a/arch/x86/kernel/paravirt.c
+++ b/arch/x86/kernel/paravirt.c
@@ -268,17 +268,6 @@ enum paravirt_lazy_mode paravirt_get_lazy_mode(void)
 	return __get_cpu_var(paravirt_lazy_mode);
 }
 
-void __init paravirt_use_bytelocks(void)
-{
-#ifdef CONFIG_SMP
-	pv_lock_ops.spin_is_locked = __byte_spin_is_locked;
-	pv_lock_ops.spin_is_contended = __byte_spin_is_contended;
-	pv_lock_ops.spin_lock = __byte_spin_lock;
-	pv_lock_ops.spin_trylock = __byte_spin_trylock;
-	pv_lock_ops.spin_unlock = __byte_spin_unlock;
-#endif
-}
-
 struct pv_info pv_info = {
 	.name = "bare hardware",
 	.paravirt_enabled = 0,
@@ -349,6 +338,10 @@ struct pv_cpu_ops pv_cpu_ops = {
 	.write_ldt_entry = native_write_ldt_entry,
 	.write_gdt_entry = native_write_gdt_entry,
 	.write_idt_entry = native_write_idt_entry,
+
+	.alloc_ldt = paravirt_nop,
+	.free_ldt = paravirt_nop,
+
 	.load_sp0 = native_load_sp0,
 
 #if defined(CONFIG_X86_32) || defined(CONFIG_IA32_EMULATION)
@@ -374,8 +367,6 @@ struct pv_cpu_ops pv_cpu_ops = {
 
 struct pv_apic_ops pv_apic_ops = {
 #ifdef CONFIG_X86_LOCAL_APIC
-	.apic_write = native_apic_write,
-	.apic_read = native_apic_read,
 	.setup_boot_clock = setup_boot_APIC_clock,
 	.setup_secondary_clock = setup_secondary_APIC_clock,
 	.startup_ipi_hook = paravirt_nop,
@@ -462,18 +453,6 @@ struct pv_mmu_ops pv_mmu_ops = {
 	.set_fixmap = native_set_fixmap,
 };
 
-struct pv_lock_ops pv_lock_ops = {
-#ifdef CONFIG_SMP
-	.spin_is_locked = __ticket_spin_is_locked,
-	.spin_is_contended = __ticket_spin_is_contended,
-
-	.spin_lock = __ticket_spin_lock,
-	.spin_trylock = __ticket_spin_trylock,
-	.spin_unlock = __ticket_spin_unlock,
-#endif
-};
-EXPORT_SYMBOL(pv_lock_ops);
-
 EXPORT_SYMBOL_GPL(pv_time_ops);
 EXPORT_SYMBOL    (pv_cpu_ops);
 EXPORT_SYMBOL    (pv_mmu_ops);
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index ec7a2ba9bce8..c622772744d8 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -15,7 +15,6 @@ unsigned long idle_nomwait;
 EXPORT_SYMBOL(idle_nomwait);
 
 struct kmem_cache *task_xstate_cachep;
-static int force_mwait __cpuinitdata;
 
 int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
 {
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 205188db9626..0a1302fe6d45 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -76,47 +76,12 @@ unsigned long thread_saved_pc(struct task_struct *tsk)
 	return ((unsigned long *)tsk->thread.sp)[3];
 }
 
-#ifdef CONFIG_HOTPLUG_CPU
-#include <asm/nmi.h>
-
-static void cpu_exit_clear(void)
-{
-	int cpu = raw_smp_processor_id();
-
-	idle_task_exit();
-
-	cpu_uninit();
-	irq_ctx_exit(cpu);
-
-	cpu_clear(cpu, cpu_callout_map);
-	cpu_clear(cpu, cpu_callin_map);
-
-	numa_remove_cpu(cpu);
-	c1e_remove_cpu(cpu);
-}
-
-/* We don't actually take CPU down, just spin without interrupts. */
-static inline void play_dead(void)
-{
-	/* This must be done before dead CPU ack */
-	cpu_exit_clear();
-	mb();
-	/* Ack it */
-	__get_cpu_var(cpu_state) = CPU_DEAD;
-
-	/*
-	 * With physical CPU hotplug, we should halt the cpu
-	 */
-	local_irq_disable();
-	/* mask all interrupts, flush any and all caches, and halt */
-	wbinvd_halt();
-}
-#else
+#ifndef CONFIG_SMP
 static inline void play_dead(void)
 {
 	BUG();
 }
-#endif /* CONFIG_HOTPLUG_CPU */
+#endif
 
 /*
  * The idle thread. There's no useful work to be
@@ -158,7 +123,7 @@ void cpu_idle(void)
 	}
 }
 
-void __show_registers(struct pt_regs *regs, int all)
+void __show_regs(struct pt_regs *regs, int all)
 {
 	unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L;
 	unsigned long d0, d1, d2, d3, d6, d7;
@@ -224,7 +189,7 @@ void __show_registers(struct pt_regs *regs, int all)
 
 void show_regs(struct pt_regs *regs)
 {
-	__show_registers(regs, 1);
+	__show_regs(regs, 1);
 	show_trace(NULL, regs, &regs->sp, regs->bp);
 }
 
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 2a8ccb9238b4..cd8c0ed02b7e 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -86,30 +86,12 @@ void exit_idle(void)
 	__exit_idle();
 }
 
-#ifdef CONFIG_HOTPLUG_CPU
-DECLARE_PER_CPU(int, cpu_state);
-
-#include <linux/nmi.h>
-/* We halt the CPU with physical CPU hotplug */
-static inline void play_dead(void)
-{
-	idle_task_exit();
-	c1e_remove_cpu(raw_smp_processor_id());
-
-	mb();
-	/* Ack it */
-	__get_cpu_var(cpu_state) = CPU_DEAD;
-
-	local_irq_disable();
-	/* mask all interrupts, flush any and all caches, and halt */
-	wbinvd_halt();
-}
-#else
+#ifndef CONFIG_SMP
 static inline void play_dead(void)
 {
 	BUG();
 }
-#endif /* CONFIG_HOTPLUG_CPU */
+#endif
 
 /*
  * The idle thread. There's no useful work to be
@@ -154,7 +136,7 @@ void cpu_idle(void)
 }
 
 /* Prints also some state that isn't saved in the pt_regs */
-void __show_regs(struct pt_regs *regs)
+void __show_regs(struct pt_regs *regs, int all)
 {
 	unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L, fs, gs, shadowgs;
 	unsigned long d0, d1, d2, d3, d6, d7;
@@ -193,6 +175,9 @@ void __show_regs(struct pt_regs *regs)
 	rdmsrl(MSR_GS_BASE, gs);
 	rdmsrl(MSR_KERNEL_GS_BASE, shadowgs);
 
+	if (!all)
+		return;
+
 	cr0 = read_cr0();
 	cr2 = read_cr2();
 	cr3 = read_cr3();
@@ -218,7 +203,7 @@ void __show_regs(struct pt_regs *regs)
 void show_regs(struct pt_regs *regs)
 {
 	printk(KERN_INFO "CPU %d:", smp_processor_id());
-	__show_regs(regs);
+	__show_regs(regs, 1);
 	show_trace(NULL, regs, (void *)(regs + 1), regs->bp);
 }
 
@@ -754,12 +739,12 @@ unsigned long get_wchan(struct task_struct *p)
 	if (!p || p == current || p->state == TASK_RUNNING)
 		return 0;
 	stack = (unsigned long)task_stack_page(p);
-	if (p->thread.sp < stack || p->thread.sp > stack+THREAD_SIZE)
+	if (p->thread.sp < stack || p->thread.sp >= stack+THREAD_SIZE)
 		return 0;
 	fp = *(u64 *)(p->thread.sp);
 	do {
 		if (fp < (unsigned long)stack ||
-		    fp > (unsigned long)stack+THREAD_SIZE)
+		    fp >= (unsigned long)stack+THREAD_SIZE)
 			return 0;
 		ip = *(u64 *)(fp+8);
 		if (!in_sched_functions(ip))
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index e375b658efc3..0a6d8c12e10d 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -40,7 +40,9 @@ enum x86_regset {
 	REGSET_GENERAL,
 	REGSET_FP,
 	REGSET_XFP,
+	REGSET_IOPERM64 = REGSET_XFP,
 	REGSET_TLS,
+	REGSET_IOPERM32,
 };
 
 /*
@@ -555,6 +557,29 @@ static int ptrace_set_debugreg(struct task_struct *child,
 	return 0;
 }
 
+/*
+ * These access the current or another (stopped) task's io permission
+ * bitmap for debugging or core dump.
+ */
+static int ioperm_active(struct task_struct *target,
+			 const struct user_regset *regset)
+{
+	return target->thread.io_bitmap_max / regset->size;
+}
+
+static int ioperm_get(struct task_struct *target,
+		      const struct user_regset *regset,
+		      unsigned int pos, unsigned int count,
+		      void *kbuf, void __user *ubuf)
+{
+	if (!target->thread.io_bitmap_ptr)
+		return -ENXIO;
+
+	return user_regset_copyout(&pos, &count, &kbuf, &ubuf,
+				   target->thread.io_bitmap_ptr,
+				   0, IO_BITMAP_BYTES);
+}
+
 #ifdef CONFIG_X86_PTRACE_BTS
 /*
  * The configuration for a particular BTS hardware implementation.
@@ -1385,6 +1410,12 @@ static const struct user_regset x86_64_regsets[] = {
 		.size = sizeof(long), .align = sizeof(long),
 		.active = xfpregs_active, .get = xfpregs_get, .set = xfpregs_set
 	},
+	[REGSET_IOPERM64] = {
+		.core_note_type = NT_386_IOPERM,
+		.n = IO_BITMAP_LONGS,
+		.size = sizeof(long), .align = sizeof(long),
+		.active = ioperm_active, .get = ioperm_get
+	},
 };
 
 static const struct user_regset_view user_x86_64_view = {
@@ -1431,6 +1462,12 @@ static const struct user_regset x86_32_regsets[] = {
 		.active = regset_tls_active,
 		.get = regset_tls_get, .set = regset_tls_set
 	},
+	[REGSET_IOPERM32] = {
+		.core_note_type = NT_386_IOPERM,
+		.n = IO_BITMAP_BYTES / sizeof(u32),
+		.size = sizeof(u32), .align = sizeof(u32),
+		.active = ioperm_active, .get = ioperm_get
+	},
 };
 
 static const struct user_regset_view user_x86_32_view = {
@@ -1452,7 +1489,8 @@ const struct user_regset_view *task_user_regset_view(struct task_struct *task)
 #endif
 }
 
-void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code)
+void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs,
+					 int error_code, int si_code)
 {
 	struct siginfo info;
 
@@ -1461,7 +1499,7 @@ void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code)
 
 	memset(&info, 0, sizeof(info));
 	info.si_signo = SIGTRAP;
-	info.si_code = TRAP_BRKPT;
+	info.si_code = si_code;
 
 	/* User-mode ip? */
 	info.si_addr = user_mode_vm(regs) ? (void __user *) regs->ip : NULL;
@@ -1548,5 +1586,5 @@ asmregparm void syscall_trace_leave(struct pt_regs *regs)
 	 */
 	if (test_thread_flag(TIF_SINGLESTEP) &&
 	    tracehook_consider_fatal_signal(current, SIGTRAP, SIG_DFL))
-		send_sigtrap(current, regs, 0);
+		send_sigtrap(current, regs, 0, TRAP_BRKPT);
 }
diff --git a/arch/x86/kernel/quirks.c b/arch/x86/kernel/quirks.c
index d13858818100..f6a11b9b1f98 100644
--- a/arch/x86/kernel/quirks.c
+++ b/arch/x86/kernel/quirks.c
@@ -354,9 +354,27 @@ static void ati_force_hpet_resume(void)
 	printk(KERN_DEBUG "Force enabled HPET at resume\n");
 }
 
+static u32 ati_ixp4x0_rev(struct pci_dev *dev)
+{
+	u32 d;
+	u8  b;
+
+	pci_read_config_byte(dev, 0xac, &b);
+	b &= ~(1<<5);
+	pci_write_config_byte(dev, 0xac, b);
+	pci_read_config_dword(dev, 0x70, &d);
+	d |= 1<<8;
+	pci_write_config_dword(dev, 0x70, d);
+	pci_read_config_dword(dev, 0x8, &d);
+	d &= 0xff;
+	dev_printk(KERN_DEBUG, &dev->dev, "SB4X0 revision 0x%x\n", d);
+	return d;
+}
+
 static void ati_force_enable_hpet(struct pci_dev *dev)
 {
-	u32 uninitialized_var(val);
+	u32 d, val;
+	u8  b;
 
 	if (hpet_address || force_hpet_address)
 		return;
@@ -366,14 +384,33 @@ static void ati_force_enable_hpet(struct pci_dev *dev)
 		return;
 	}
 
+	d = ati_ixp4x0_rev(dev);
+	if (d  < 0x82)
+		return;
+
+	/* base address */
 	pci_write_config_dword(dev, 0x14, 0xfed00000);
 	pci_read_config_dword(dev, 0x14, &val);
+
+	/* enable interrupt */
+	outb(0x72, 0xcd6); b = inb(0xcd7);
+	b |= 0x1;
+	outb(0x72, 0xcd6); outb(b, 0xcd7);
+	outb(0x72, 0xcd6); b = inb(0xcd7);
+	if (!(b & 0x1))
+		return;
+	pci_read_config_dword(dev, 0x64, &d);
+	d |= (1<<10);
+	pci_write_config_dword(dev, 0x64, d);
+	pci_read_config_dword(dev, 0x64, &d);
+	if (!(d & (1<<10)))
+		return;
+
 	force_hpet_address = val;
 	force_hpet_resume_type = ATI_FORCE_HPET_RESUME;
 	dev_printk(KERN_DEBUG, &dev->dev, "Force enabled HPET at 0x%lx\n",
 		   force_hpet_address);
 	cached_dev = dev;
-	return;
 }
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP400_SMBUS,
 			 ati_force_enable_hpet);
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 141efab52400..2255782e8d4b 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -302,7 +302,7 @@ static void __init relocate_initrd(void)
 		if (clen > MAX_MAP_CHUNK-slop)
 			clen = MAX_MAP_CHUNK-slop;
 		mapaddr = ramdisk_image & PAGE_MASK;
-		p = early_ioremap(mapaddr, clen+slop);
+		p = early_memremap(mapaddr, clen+slop);
 		memcpy(q, p+slop, clen);
 		early_iounmap(p, clen+slop);
 		q += clen;
@@ -379,7 +379,7 @@ static void __init parse_setup_data(void)
 		return;
 	pa_data = boot_params.hdr.setup_data;
 	while (pa_data) {
-		data = early_ioremap(pa_data, PAGE_SIZE);
+		data = early_memremap(pa_data, PAGE_SIZE);
 		switch (data->type) {
 		case SETUP_E820_EXT:
 			parse_e820_ext(data, pa_data);
@@ -402,7 +402,7 @@ static void __init e820_reserve_setup_data(void)
 		return;
 	pa_data = boot_params.hdr.setup_data;
 	while (pa_data) {
-		data = early_ioremap(pa_data, sizeof(*data));
+		data = early_memremap(pa_data, sizeof(*data));
 		e820_update_range(pa_data, sizeof(*data)+data->len,
 			 E820_RAM, E820_RESERVED_KERN);
 		found = 1;
@@ -428,7 +428,7 @@ static void __init reserve_early_setup_data(void)
 		return;
 	pa_data = boot_params.hdr.setup_data;
 	while (pa_data) {
-		data = early_ioremap(pa_data, sizeof(*data));
+		data = early_memremap(pa_data, sizeof(*data));
 		sprintf(buf, "setup data %x", data->type);
 		reserve_early(pa_data, pa_data+sizeof(*data)+data->len, buf);
 		pa_data = data->next;
@@ -582,6 +582,190 @@ static struct x86_quirks default_x86_quirks __initdata;
 struct x86_quirks *x86_quirks __initdata = &default_x86_quirks;
 
 /*
+ * Some BIOSes seem to corrupt the low 64k of memory during events
+ * like suspend/resume and unplugging an HDMI cable.  Reserve all
+ * remaining free memory in that area and fill it with a distinct
+ * pattern.
+ */
+#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION
+#define MAX_SCAN_AREAS	8
+
+static int __read_mostly memory_corruption_check = -1;
+
+static unsigned __read_mostly corruption_check_size = 64*1024;
+static unsigned __read_mostly corruption_check_period = 60; /* seconds */
+
+static struct e820entry scan_areas[MAX_SCAN_AREAS];
+static int num_scan_areas;
+
+
+static int set_corruption_check(char *arg)
+{
+	char *end;
+
+	memory_corruption_check = simple_strtol(arg, &end, 10);
+
+	return (*end == 0) ? 0 : -EINVAL;
+}
+early_param("memory_corruption_check", set_corruption_check);
+
+static int set_corruption_check_period(char *arg)
+{
+	char *end;
+
+	corruption_check_period = simple_strtoul(arg, &end, 10);
+
+	return (*end == 0) ? 0 : -EINVAL;
+}
+early_param("memory_corruption_check_period", set_corruption_check_period);
+
+static int set_corruption_check_size(char *arg)
+{
+	char *end;
+	unsigned size;
+
+	size = memparse(arg, &end);
+
+	if (*end == '\0')
+		corruption_check_size = size;
+
+	return (size == corruption_check_size) ? 0 : -EINVAL;
+}
+early_param("memory_corruption_check_size", set_corruption_check_size);
+
+
+static void __init setup_bios_corruption_check(void)
+{
+	u64 addr = PAGE_SIZE;	/* assume first page is reserved anyway */
+
+	if (memory_corruption_check == -1) {
+		memory_corruption_check =
+#ifdef CONFIG_X86_BOOTPARAM_MEMORY_CORRUPTION_CHECK
+			1
+#else
+			0
+#endif
+			;
+	}
+
+	if (corruption_check_size == 0)
+		memory_corruption_check = 0;
+
+	if (!memory_corruption_check)
+		return;
+
+	corruption_check_size = round_up(corruption_check_size, PAGE_SIZE);
+
+	while(addr < corruption_check_size && num_scan_areas < MAX_SCAN_AREAS) {
+		u64 size;
+		addr = find_e820_area_size(addr, &size, PAGE_SIZE);
+
+		if (addr == 0)
+			break;
+
+		if ((addr + size) > corruption_check_size)
+			size = corruption_check_size - addr;
+
+		if (size == 0)
+			break;
+
+		e820_update_range(addr, size, E820_RAM, E820_RESERVED);
+		scan_areas[num_scan_areas].addr = addr;
+		scan_areas[num_scan_areas].size = size;
+		num_scan_areas++;
+
+		/* Assume we've already mapped this early memory */
+		memset(__va(addr), 0, size);
+
+		addr += size;
+	}
+
+	printk(KERN_INFO "Scanning %d areas for low memory corruption\n",
+	       num_scan_areas);
+	update_e820();
+}
+
+static struct timer_list periodic_check_timer;
+
+void check_for_bios_corruption(void)
+{
+	int i;
+	int corruption = 0;
+
+	if (!memory_corruption_check)
+		return;
+
+	for(i = 0; i < num_scan_areas; i++) {
+		unsigned long *addr = __va(scan_areas[i].addr);
+		unsigned long size = scan_areas[i].size;
+
+		for(; size; addr++, size -= sizeof(unsigned long)) {
+			if (!*addr)
+				continue;
+			printk(KERN_ERR "Corrupted low memory at %p (%lx phys) = %08lx\n",
+			       addr, __pa(addr), *addr);
+			corruption = 1;
+			*addr = 0;
+		}
+	}
+
+	WARN(corruption, KERN_ERR "Memory corruption detected in low memory\n");
+}
+
+static void periodic_check_for_corruption(unsigned long data)
+{
+	check_for_bios_corruption();
+	mod_timer(&periodic_check_timer, round_jiffies(jiffies + corruption_check_period*HZ));
+}
+
+void start_periodic_check_for_corruption(void)
+{
+	if (!memory_corruption_check || corruption_check_period == 0)
+		return;
+
+	printk(KERN_INFO "Scanning for low memory corruption every %d seconds\n",
+	       corruption_check_period);
+
+	init_timer(&periodic_check_timer);
+	periodic_check_timer.function = &periodic_check_for_corruption;
+	periodic_check_for_corruption(0);
+}
+#endif
+
+static int __init dmi_low_memory_corruption(const struct dmi_system_id *d)
+{
+	printk(KERN_NOTICE
+		"%s detected: BIOS may corrupt low RAM, working it around.\n",
+		d->ident);
+
+	e820_update_range(0, 0x10000, E820_RAM, E820_RESERVED);
+	sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map);
+
+	return 0;
+}
+
+/* List of systems that have known low memory corruption BIOS problems */
+static struct dmi_system_id __initdata bad_bios_dmi_table[] = {
+#ifdef CONFIG_X86_RESERVE_LOW_64K
+	{
+		.callback = dmi_low_memory_corruption,
+		.ident = "AMI BIOS",
+		.matches = {
+			DMI_MATCH(DMI_BIOS_VENDOR, "American Megatrends Inc."),
+		},
+	},
+	{
+		.callback = dmi_low_memory_corruption,
+		.ident = "Phoenix BIOS",
+		.matches = {
+			DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies, LTD"),
+		},
+	},
+#endif
+	{}
+};
+
+/*
  * Determine if we were loaded by an EFI loader.  If so, then we have also been
  * passed the efi memmap, systab, etc., so we should use these data structures
  * for initialization.  Note, the efi init code path is determined by the
@@ -715,6 +899,10 @@ void __init setup_arch(char **cmdline_p)
 
 	finish_e820_parsing();
 
+	dmi_scan_machine();
+
+	dmi_check_system(bad_bios_dmi_table);
+
 #ifdef CONFIG_X86_32
 	probe_roms();
 #endif
@@ -758,6 +946,8 @@ void __init setup_arch(char **cmdline_p)
 #else
 	num_physpages = max_pfn;
 
+ 	if (cpu_has_x2apic)
+ 		check_x2apic();
 
 	/* How many end-of-memory variables you have, grandma! */
 	/* need this before calling reserve_initrd */
@@ -769,6 +959,10 @@ void __init setup_arch(char **cmdline_p)
 	high_memory = (void *)__va(max_pfn * PAGE_SIZE - 1) + 1;
 #endif
 
+#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION
+	setup_bios_corruption_check();
+#endif
+
 	/* max_pfn_mapped is updated here */
 	max_low_pfn_mapped = init_memory_mapping(0, max_low_pfn<<PAGE_SHIFT);
 	max_pfn_mapped = max_low_pfn_mapped;
@@ -797,8 +991,6 @@ void __init setup_arch(char **cmdline_p)
 	vsmp_init();
 #endif
 
-	dmi_scan_machine();
-
 	io_delay_init();
 
 	/*
@@ -806,6 +998,8 @@ void __init setup_arch(char **cmdline_p)
 	 */
 	acpi_boot_table_init();
 
+	early_acpi_boot_init();
+
 #ifdef CONFIG_ACPI_NUMA
 	/*
 	 * Parse SRAT to discover nodes.
@@ -901,3 +1095,5 @@ void __init setup_arch(char **cmdline_p)
 #endif
 #endif
 }
+
+
diff --git a/arch/x86/kernel/sigframe.h b/arch/x86/kernel/sigframe.h
index 8b4956e800ac..cc673aa55ce4 100644
--- a/arch/x86/kernel/sigframe.h
+++ b/arch/x86/kernel/sigframe.h
@@ -3,9 +3,18 @@ struct sigframe {
 	char __user *pretcode;
 	int sig;
 	struct sigcontext sc;
-	struct _fpstate fpstate;
+	/*
+	 * fpstate is unused. fpstate is moved/allocated after
+	 * retcode[] below. This movement allows to have the FP state and the
+	 * future state extensions (xsave) stay together.
+	 * And at the same time retaining the unused fpstate, prevents changing
+	 * the offset of extramask[] in the sigframe and thus prevent any
+	 * legacy application accessing/modifying it.
+	 */
+	struct _fpstate fpstate_unused;
 	unsigned long extramask[_NSIG_WORDS-1];
 	char retcode[8];
+	/* fp state follows here */
 };
 
 struct rt_sigframe {
@@ -15,14 +24,15 @@ struct rt_sigframe {
 	void __user *puc;
 	struct siginfo info;
 	struct ucontext uc;
-	struct _fpstate fpstate;
 	char retcode[8];
+	/* fp state follows here */
 };
 #else
 struct rt_sigframe {
 	char __user *pretcode;
 	struct ucontext uc;
 	struct siginfo info;
+	/* fp state follows here */
 };
 
 int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
diff --git a/arch/x86/kernel/signal_32.c b/arch/x86/kernel/signal_32.c
index 2a2435d3037d..d6dd057d0f22 100644
--- a/arch/x86/kernel/signal_32.c
+++ b/arch/x86/kernel/signal_32.c
@@ -27,6 +27,7 @@
 #include <asm/uaccess.h>
 #include <asm/i387.h>
 #include <asm/vdso.h>
+#include <asm/syscall.h>
 #include <asm/syscalls.h>
 
 #include "sigframe.h"
@@ -112,6 +113,27 @@ asmlinkage int sys_sigaltstack(unsigned long bx)
 	return do_sigaltstack(uss, uoss, regs->sp);
 }
 
+#define COPY(x)			{		\
+	err |= __get_user(regs->x, &sc->x);	\
+}
+
+#define COPY_SEG(seg)		{			\
+		unsigned short tmp;			\
+		err |= __get_user(tmp, &sc->seg);	\
+		regs->seg = tmp;			\
+}
+
+#define COPY_SEG_STRICT(seg)	{			\
+		unsigned short tmp;			\
+		err |= __get_user(tmp, &sc->seg);	\
+		regs->seg = tmp | 3;			\
+}
+
+#define GET_SEG(seg)		{			\
+		unsigned short tmp;			\
+		err |= __get_user(tmp, &sc->seg);	\
+		loadsegment(seg, tmp);			\
+}
 
 /*
  * Do a signal return; undo the signal stack.
@@ -120,28 +142,13 @@ static int
 restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc,
 		   unsigned long *pax)
 {
+	void __user *buf;
+	unsigned int tmpflags;
 	unsigned int err = 0;
 
 	/* Always make any pending restarted system calls return -EINTR */
 	current_thread_info()->restart_block.fn = do_no_restart_syscall;
 
-#define COPY(x)		err |= __get_user(regs->x, &sc->x)
-
-#define COPY_SEG(seg)							\
-	{ unsigned short tmp;						\
-	  err |= __get_user(tmp, &sc->seg);				\
-	  regs->seg = tmp; }
-
-#define COPY_SEG_STRICT(seg)						\
-	{ unsigned short tmp;						\
-	  err |= __get_user(tmp, &sc->seg);				\
-	  regs->seg = tmp|3; }
-
-#define GET_SEG(seg)							\
-	{ unsigned short tmp;						\
-	  err |= __get_user(tmp, &sc->seg);				\
-	  loadsegment(seg, tmp); }
-
 	GET_SEG(gs);
 	COPY_SEG(fs);
 	COPY_SEG(es);
@@ -151,38 +158,15 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc,
 	COPY_SEG_STRICT(cs);
 	COPY_SEG_STRICT(ss);
 
-	{
-		unsigned int tmpflags;
-
-		err |= __get_user(tmpflags, &sc->flags);
-		regs->flags = (regs->flags & ~FIX_EFLAGS) |
-						(tmpflags & FIX_EFLAGS);
-		regs->orig_ax = -1;		/* disable syscall checks */
-	}
-
-	{
-		struct _fpstate __user *buf;
+	err |= __get_user(tmpflags, &sc->flags);
+	regs->flags = (regs->flags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS);
+	regs->orig_ax = -1;		/* disable syscall checks */
 
-		err |= __get_user(buf, &sc->fpstate);
-		if (buf) {
-			if (!access_ok(VERIFY_READ, buf, sizeof(*buf)))
-				goto badframe;
-			err |= restore_i387(buf);
-		} else {
-			struct task_struct *me = current;
-
-			if (used_math()) {
-				clear_fpu(me);
-				clear_used_math();
-			}
-		}
-	}
+	err |= __get_user(buf, &sc->fpstate);
+	err |= restore_i387_xstate(buf);
 
 	err |= __get_user(*pax, &sc->ax);
 	return err;
-
-badframe:
-	return 1;
 }
 
 asmlinkage unsigned long sys_sigreturn(unsigned long __unused)
@@ -228,9 +212,8 @@ badframe:
 	return 0;
 }
 
-asmlinkage int sys_rt_sigreturn(unsigned long __unused)
+static long do_rt_sigreturn(struct pt_regs *regs)
 {
-	struct pt_regs *regs = (struct pt_regs *)&__unused;
 	struct rt_sigframe __user *frame;
 	unsigned long ax;
 	sigset_t set;
@@ -256,15 +239,22 @@ asmlinkage int sys_rt_sigreturn(unsigned long __unused)
 	return ax;
 
 badframe:
-	force_sig(SIGSEGV, current);
+	signal_fault(regs, frame, "rt_sigreturn");
 	return 0;
 }
 
+asmlinkage int sys_rt_sigreturn(unsigned long __unused)
+{
+	struct pt_regs *regs = (struct pt_regs *)&__unused;
+
+	return do_rt_sigreturn(regs);
+}
+
 /*
  * Set up a signal frame.
  */
 static int
-setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate,
+setup_sigcontext(struct sigcontext __user *sc, void __user *fpstate,
 		 struct pt_regs *regs, unsigned long mask)
 {
 	int tmp, err = 0;
@@ -291,7 +281,7 @@ setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate,
 	err |= __put_user(regs->sp, &sc->sp_at_signal);
 	err |= __put_user(regs->ss, (unsigned int __user *)&sc->ss);
 
-	tmp = save_i387(fpstate);
+	tmp = save_i387_xstate(fpstate);
 	if (tmp < 0)
 		err = 1;
 	else
@@ -308,7 +298,8 @@ setup_sigcontext(struct sigcontext __user *sc, struct _fpstate __user *fpstate,
  * Determine which stack to use..
  */
 static inline void __user *
-get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size)
+get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size,
+	     void **fpstate)
 {
 	unsigned long sp;
 
@@ -334,6 +325,11 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size)
 			sp = (unsigned long) ka->sa.sa_restorer;
 	}
 
+	if (used_math()) {
+		sp = sp - sig_xstate_size;
+		*fpstate = (struct _fpstate *) sp;
+	}
+
 	sp -= frame_size;
 	/*
 	 * Align the stack pointer according to the i386 ABI,
@@ -345,38 +341,29 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size)
 }
 
 static int
-setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
-	    struct pt_regs *regs)
+__setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
+	      struct pt_regs *regs)
 {
 	struct sigframe __user *frame;
 	void __user *restorer;
 	int err = 0;
-	int usig;
+	void __user *fpstate = NULL;
 
-	frame = get_sigframe(ka, regs, sizeof(*frame));
+	frame = get_sigframe(ka, regs, sizeof(*frame), &fpstate);
 
 	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
-		goto give_sigsegv;
+		return -EFAULT;
 
-	usig = current_thread_info()->exec_domain
-		&& current_thread_info()->exec_domain->signal_invmap
-		&& sig < 32
-		? current_thread_info()->exec_domain->signal_invmap[sig]
-		: sig;
+	if (__put_user(sig, &frame->sig))
+		return -EFAULT;
 
-	err = __put_user(usig, &frame->sig);
-	if (err)
-		goto give_sigsegv;
-
-	err = setup_sigcontext(&frame->sc, &frame->fpstate, regs, set->sig[0]);
-	if (err)
-		goto give_sigsegv;
+	if (setup_sigcontext(&frame->sc, fpstate, regs, set->sig[0]))
+		return -EFAULT;
 
 	if (_NSIG_WORDS > 1) {
-		err = __copy_to_user(&frame->extramask, &set->sig[1],
-				      sizeof(frame->extramask));
-		if (err)
-			goto give_sigsegv;
+		if (__copy_to_user(&frame->extramask, &set->sig[1],
+				   sizeof(frame->extramask)))
+			return -EFAULT;
 	}
 
 	if (current->mm->context.vdso)
@@ -401,7 +388,7 @@ setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
 	err |= __put_user(0x80cd, (short __user *)(frame->retcode+6));
 
 	if (err)
-		goto give_sigsegv;
+		return -EFAULT;
 
 	/* Set up registers for signal handler */
 	regs->sp = (unsigned long)frame;
@@ -416,50 +403,43 @@ setup_frame(int sig, struct k_sigaction *ka, sigset_t *set,
 	regs->cs = __USER_CS;
 
 	return 0;
-
-give_sigsegv:
-	force_sigsegv(sig, current);
-	return -EFAULT;
 }
 
-static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
-			  sigset_t *set, struct pt_regs *regs)
+static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+			    sigset_t *set, struct pt_regs *regs)
 {
 	struct rt_sigframe __user *frame;
 	void __user *restorer;
 	int err = 0;
-	int usig;
+	void __user *fpstate = NULL;
 
-	frame = get_sigframe(ka, regs, sizeof(*frame));
+	frame = get_sigframe(ka, regs, sizeof(*frame), &fpstate);
 
 	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
-		goto give_sigsegv;
+		return -EFAULT;
 
-	usig = current_thread_info()->exec_domain
-		&& current_thread_info()->exec_domain->signal_invmap
-		&& sig < 32
-		? current_thread_info()->exec_domain->signal_invmap[sig]
-		: sig;
-
-	err |= __put_user(usig, &frame->sig);
+	err |= __put_user(sig, &frame->sig);
 	err |= __put_user(&frame->info, &frame->pinfo);
 	err |= __put_user(&frame->uc, &frame->puc);
 	err |= copy_siginfo_to_user(&frame->info, info);
 	if (err)
-		goto give_sigsegv;
+		return -EFAULT;
 
 	/* Create the ucontext.  */
-	err |= __put_user(0, &frame->uc.uc_flags);
+	if (cpu_has_xsave)
+		err |= __put_user(UC_FP_XSTATE, &frame->uc.uc_flags);
+	else
+		err |= __put_user(0, &frame->uc.uc_flags);
 	err |= __put_user(0, &frame->uc.uc_link);
 	err |= __put_user(current->sas_ss_sp, &frame->uc.uc_stack.ss_sp);
 	err |= __put_user(sas_ss_flags(regs->sp),
 			  &frame->uc.uc_stack.ss_flags);
 	err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size);
-	err |= setup_sigcontext(&frame->uc.uc_mcontext, &frame->fpstate,
+	err |= setup_sigcontext(&frame->uc.uc_mcontext, fpstate,
 				regs, set->sig[0]);
 	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
 	if (err)
-		goto give_sigsegv;
+		return -EFAULT;
 
 	/* Set up to return from userspace.  */
 	restorer = VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn);
@@ -479,12 +459,12 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	err |= __put_user(0x80cd, (short __user *)(frame->retcode+5));
 
 	if (err)
-		goto give_sigsegv;
+		return -EFAULT;
 
 	/* Set up registers for signal handler */
 	regs->sp = (unsigned long)frame;
 	regs->ip = (unsigned long)ka->sa.sa_handler;
-	regs->ax = (unsigned long)usig;
+	regs->ax = (unsigned long)sig;
 	regs->dx = (unsigned long)&frame->info;
 	regs->cx = (unsigned long)&frame->uc;
 
@@ -494,15 +474,48 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	regs->cs = __USER_CS;
 
 	return 0;
-
-give_sigsegv:
-	force_sigsegv(sig, current);
-	return -EFAULT;
 }
 
 /*
  * OK, we're invoking a handler:
  */
+static int signr_convert(int sig)
+{
+	struct thread_info *info = current_thread_info();
+
+	if (info->exec_domain && info->exec_domain->signal_invmap && sig < 32)
+		return info->exec_domain->signal_invmap[sig];
+	return sig;
+}
+
+#define is_ia32	1
+#define ia32_setup_frame	__setup_frame
+#define ia32_setup_rt_frame	__setup_rt_frame
+
+static int
+setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+	       sigset_t *set, struct pt_regs *regs)
+{
+	int usig = signr_convert(sig);
+	int ret;
+
+	/* Set up the stack frame */
+	if (is_ia32) {
+		if (ka->sa.sa_flags & SA_SIGINFO)
+			ret = ia32_setup_rt_frame(usig, ka, info, set, regs);
+		else
+			ret = ia32_setup_frame(usig, ka, set, regs);
+	} else
+		ret = __setup_rt_frame(sig, ka, info, set, regs);
+
+	if (ret) {
+		force_sigsegv(sig, current);
+		return -EFAULT;
+	}
+
+	return ret;
+}
+
 static int
 handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	      sigset_t *oldset, struct pt_regs *regs)
@@ -510,9 +523,9 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	int ret;
 
 	/* Are we from a system call? */
-	if ((long)regs->orig_ax >= 0) {
+	if (syscall_get_nr(current, regs) >= 0) {
 		/* If so, check system call restarting.. */
-		switch (regs->ax) {
+		switch (syscall_get_error(current, regs)) {
 		case -ERESTART_RESTARTBLOCK:
 		case -ERESTARTNOHAND:
 			regs->ax = -EINTR;
@@ -539,15 +552,20 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	    likely(test_and_clear_thread_flag(TIF_FORCED_TF)))
 		regs->flags &= ~X86_EFLAGS_TF;
 
-	/* Set up the stack frame */
-	if (ka->sa.sa_flags & SA_SIGINFO)
-		ret = setup_rt_frame(sig, ka, info, oldset, regs);
-	else
-		ret = setup_frame(sig, ka, oldset, regs);
+	ret = setup_rt_frame(sig, ka, info, oldset, regs);
 
 	if (ret)
 		return ret;
 
+#ifdef CONFIG_X86_64
+	/*
+	 * This has nothing to do with segment registers,
+	 * despite the name.  This magic affects uaccess.h
+	 * macros' behavior.  Reset it to the normal setting.
+	 */
+	set_fs(USER_DS);
+#endif
+
 	/*
 	 * Clear the direction flag as per the ABI for function entry.
 	 */
@@ -574,6 +592,7 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	return 0;
 }
 
+#define NR_restart_syscall	__NR_restart_syscall
 /*
  * Note that 'init' is a special process: it doesn't get signals it doesn't
  * want to handle. Thus you cannot kill init even with a SIGKILL even by
@@ -626,9 +645,9 @@ static void do_signal(struct pt_regs *regs)
 	}
 
 	/* Did we come from a system call? */
-	if ((long)regs->orig_ax >= 0) {
+	if (syscall_get_nr(current, regs) >= 0) {
 		/* Restart the system call - no handlers present */
-		switch (regs->ax) {
+		switch (syscall_get_error(current, regs)) {
 		case -ERESTARTNOHAND:
 		case -ERESTARTSYS:
 		case -ERESTARTNOINTR:
@@ -637,7 +656,7 @@ static void do_signal(struct pt_regs *regs)
 			break;
 
 		case -ERESTART_RESTARTBLOCK:
-			regs->ax = __NR_restart_syscall;
+			regs->ax = NR_restart_syscall;
 			regs->ip -= 2;
 			break;
 		}
@@ -660,6 +679,12 @@ static void do_signal(struct pt_regs *regs)
 void
 do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags)
 {
+#if defined(CONFIG_X86_64) && defined(CONFIG_X86_MCE)
+	/* notify userspace of pending MCEs */
+	if (thread_info_flags & _TIF_MCE_NOTIFY)
+		mce_notify_user();
+#endif /* CONFIG_X86_64 && CONFIG_X86_MCE */
+
 	/* deal with pending signal delivery */
 	if (thread_info_flags & _TIF_SIGPENDING)
 		do_signal(regs);
@@ -669,5 +694,23 @@ do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags)
 		tracehook_notify_resume(regs);
 	}
 
+#ifdef CONFIG_X86_32
 	clear_thread_flag(TIF_IRET);
+#endif /* CONFIG_X86_32 */
+}
+
+void signal_fault(struct pt_regs *regs, void __user *frame, char *where)
+{
+	struct task_struct *me = current;
+
+	if (show_unhandled_signals && printk_ratelimit()) {
+		printk(KERN_INFO
+		       "%s[%d] bad frame in %s frame:%p ip:%lx sp:%lx orax:%lx",
+		       me->comm, me->pid, where, frame,
+		       regs->ip, regs->sp, regs->orig_ax);
+		print_vma_addr(" in ", regs->ip);
+		printk(KERN_CONT "\n");
+	}
+
+	force_sig(SIGSEGV, me);
 }
diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c
index 694aa888bb19..a5c9627f4db9 100644
--- a/arch/x86/kernel/signal_64.c
+++ b/arch/x86/kernel/signal_64.c
@@ -52,67 +52,14 @@ sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
 	return do_sigaltstack(uss, uoss, regs->sp);
 }
 
-/*
- * Signal frame handlers.
- */
-
-static inline int save_i387(struct _fpstate __user *buf)
-{
-	struct task_struct *tsk = current;
-	int err = 0;
-
-	BUILD_BUG_ON(sizeof(struct user_i387_struct) !=
-			sizeof(tsk->thread.xstate->fxsave));
-
-	if ((unsigned long)buf % 16)
-		printk("save_i387: bad fpstate %p\n", buf);
-
-	if (!used_math())
-		return 0;
-	clear_used_math(); /* trigger finit */
-	if (task_thread_info(tsk)->status & TS_USEDFPU) {
-		err = save_i387_checking((struct i387_fxsave_struct __user *)
-					 buf);
-		if (err)
-			return err;
-		task_thread_info(tsk)->status &= ~TS_USEDFPU;
-		stts();
-	} else {
-		if (__copy_to_user(buf, &tsk->thread.xstate->fxsave,
-				   sizeof(struct i387_fxsave_struct)))
-			return -1;
-	}
-	return 1;
+#define COPY(x)			{		\
+	err |= __get_user(regs->x, &sc->x);	\
 }
 
-/*
- * This restores directly out of user space. Exceptions are handled.
- */
-static inline int restore_i387(struct _fpstate __user *buf)
-{
-	struct task_struct *tsk = current;
-	int err;
-
-	if (!used_math()) {
-		err = init_fpu(tsk);
-		if (err)
-			return err;
-	}
-
-	if (!(task_thread_info(current)->status & TS_USEDFPU)) {
-		clts();
-		task_thread_info(current)->status |= TS_USEDFPU;
-	}
-	err = restore_fpu_checking((__force struct i387_fxsave_struct *)buf);
-	if (unlikely(err)) {
-		/*
-		 * Encountered an error while doing the restore from the
-		 * user buffer, clear the fpu state.
-		 */
-		clear_fpu(tsk);
-		clear_used_math();
-	}
-	return err;
+#define COPY_SEG_STRICT(seg)	{			\
+		unsigned short tmp;			\
+		err |= __get_user(tmp, &sc->seg);	\
+		regs->seg = tmp | 3;			\
 }
 
 /*
@@ -122,13 +69,13 @@ static int
 restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc,
 		   unsigned long *pax)
 {
+	void __user *buf;
+	unsigned int tmpflags;
 	unsigned int err = 0;
 
 	/* Always make any pending restarted system calls return -EINTR */
 	current_thread_info()->restart_block.fn = do_no_restart_syscall;
 
-#define COPY(x)		(err |= __get_user(regs->x, &sc->x))
-
 	COPY(di); COPY(si); COPY(bp); COPY(sp); COPY(bx);
 	COPY(dx); COPY(cx); COPY(ip);
 	COPY(r8);
@@ -143,48 +90,24 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc,
 	/* Kernel saves and restores only the CS segment register on signals,
 	 * which is the bare minimum needed to allow mixed 32/64-bit code.
 	 * App's signal handler can save/restore other segments if needed. */
-	{
-		unsigned cs;
-		err |= __get_user(cs, &sc->cs);
-		regs->cs = cs | 3;	/* Force into user mode */
-	}
+	COPY_SEG_STRICT(cs);
 
-	{
-		unsigned int tmpflags;
-		err |= __get_user(tmpflags, &sc->flags);
-		regs->flags = (regs->flags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS);
-		regs->orig_ax = -1;		/* disable syscall checks */
-	}
+	err |= __get_user(tmpflags, &sc->flags);
+	regs->flags = (regs->flags & ~FIX_EFLAGS) | (tmpflags & FIX_EFLAGS);
+	regs->orig_ax = -1;		/* disable syscall checks */
 
-	{
-		struct _fpstate __user *buf;
-		err |= __get_user(buf, &sc->fpstate);
-
-		if (buf) {
-			if (!access_ok(VERIFY_READ, buf, sizeof(*buf)))
-				goto badframe;
-			err |= restore_i387(buf);
-		} else {
-			struct task_struct *me = current;
-			if (used_math()) {
-				clear_fpu(me);
-				clear_used_math();
-			}
-		}
-	}
+	err |= __get_user(buf, &sc->fpstate);
+	err |= restore_i387_xstate(buf);
 
 	err |= __get_user(*pax, &sc->ax);
 	return err;
-
-badframe:
-	return 1;
 }
 
-asmlinkage long sys_rt_sigreturn(struct pt_regs *regs)
+static long do_rt_sigreturn(struct pt_regs *regs)
 {
 	struct rt_sigframe __user *frame;
-	sigset_t set;
 	unsigned long ax;
+	sigset_t set;
 
 	frame = (struct rt_sigframe __user *)(regs->sp - sizeof(long));
 	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
@@ -207,10 +130,15 @@ asmlinkage long sys_rt_sigreturn(struct pt_regs *regs)
 	return ax;
 
 badframe:
-	signal_fault(regs, frame, "sigreturn");
+	signal_fault(regs, frame, "rt_sigreturn");
 	return 0;
 }
 
+asmlinkage long sys_rt_sigreturn(struct pt_regs *regs)
+{
+	return do_rt_sigreturn(regs);
+}
+
 /*
  * Set up a signal frame.
  */
@@ -269,41 +197,40 @@ get_stack(struct k_sigaction *ka, struct pt_regs *regs, unsigned long size)
 			sp = current->sas_ss_sp + current->sas_ss_size;
 	}
 
-	return (void __user *)round_down(sp - size, 16);
+	return (void __user *)round_down(sp - size, 64);
 }
 
-static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
-			   sigset_t *set, struct pt_regs *regs)
+static int __setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+			    sigset_t *set, struct pt_regs *regs)
 {
 	struct rt_sigframe __user *frame;
-	struct _fpstate __user *fp = NULL;
+	void __user *fp = NULL;
 	int err = 0;
 	struct task_struct *me = current;
 
 	if (used_math()) {
-		fp = get_stack(ka, regs, sizeof(struct _fpstate));
+		fp = get_stack(ka, regs, sig_xstate_size);
 		frame = (void __user *)round_down(
 			(unsigned long)fp - sizeof(struct rt_sigframe), 16) - 8;
 
-		if (!access_ok(VERIFY_WRITE, fp, sizeof(struct _fpstate)))
-			goto give_sigsegv;
-
-		if (save_i387(fp) < 0)
-			err |= -1;
+		if (save_i387_xstate(fp) < 0)
+			return -EFAULT;
 	} else
 		frame = get_stack(ka, regs, sizeof(struct rt_sigframe)) - 8;
 
 	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
-		goto give_sigsegv;
+		return -EFAULT;
 
 	if (ka->sa.sa_flags & SA_SIGINFO) {
-		err |= copy_siginfo_to_user(&frame->info, info);
-		if (err)
-			goto give_sigsegv;
+		if (copy_siginfo_to_user(&frame->info, info))
+			return -EFAULT;
 	}
 
 	/* Create the ucontext.  */
-	err |= __put_user(0, &frame->uc.uc_flags);
+	if (cpu_has_xsave)
+		err |= __put_user(UC_FP_XSTATE, &frame->uc.uc_flags);
+	else
+		err |= __put_user(0, &frame->uc.uc_flags);
 	err |= __put_user(0, &frame->uc.uc_link);
 	err |= __put_user(me->sas_ss_sp, &frame->uc.uc_stack.ss_sp);
 	err |= __put_user(sas_ss_flags(regs->sp),
@@ -324,11 +251,11 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		err |= __put_user(ka->sa.sa_restorer, &frame->pretcode);
 	} else {
 		/* could use a vstub here */
-		goto give_sigsegv;
+		return -EFAULT;
 	}
 
 	if (err)
-		goto give_sigsegv;
+		return -EFAULT;
 
 	/* Set up registers for signal handler */
 	regs->di = sig;
@@ -348,15 +275,45 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	regs->cs = __USER_CS;
 
 	return 0;
-
-give_sigsegv:
-	force_sigsegv(sig, current);
-	return -EFAULT;
 }
 
 /*
  * OK, we're invoking a handler
  */
+static int signr_convert(int sig)
+{
+	return sig;
+}
+
+#ifdef CONFIG_IA32_EMULATION
+#define is_ia32	test_thread_flag(TIF_IA32)
+#else
+#define is_ia32	0
+#endif
+
+static int
+setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+	       sigset_t *set, struct pt_regs *regs)
+{
+	int usig = signr_convert(sig);
+	int ret;
+
+	/* Set up the stack frame */
+	if (is_ia32) {
+		if (ka->sa.sa_flags & SA_SIGINFO)
+			ret = ia32_setup_rt_frame(usig, ka, info, set, regs);
+		else
+			ret = ia32_setup_frame(usig, ka, set, regs);
+	} else
+		ret = __setup_rt_frame(sig, ka, info, set, regs);
+
+	if (ret) {
+		force_sigsegv(sig, current);
+		return -EFAULT;
+	}
+
+	return ret;
+}
 
 static int
 handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
@@ -394,51 +351,48 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 	    likely(test_and_clear_thread_flag(TIF_FORCED_TF)))
 		regs->flags &= ~X86_EFLAGS_TF;
 
-#ifdef CONFIG_IA32_EMULATION
-	if (test_thread_flag(TIF_IA32)) {
-		if (ka->sa.sa_flags & SA_SIGINFO)
-			ret = ia32_setup_rt_frame(sig, ka, info, oldset, regs);
-		else
-			ret = ia32_setup_frame(sig, ka, oldset, regs);
-	} else
-#endif
 	ret = setup_rt_frame(sig, ka, info, oldset, regs);
 
-	if (ret == 0) {
-		/*
-		 * This has nothing to do with segment registers,
-		 * despite the name.  This magic affects uaccess.h
-		 * macros' behavior.  Reset it to the normal setting.
-		 */
-		set_fs(USER_DS);
+	if (ret)
+		return ret;
 
-		/*
-		 * Clear the direction flag as per the ABI for function entry.
-		 */
-		regs->flags &= ~X86_EFLAGS_DF;
+#ifdef CONFIG_X86_64
+	/*
+	 * This has nothing to do with segment registers,
+	 * despite the name.  This magic affects uaccess.h
+	 * macros' behavior.  Reset it to the normal setting.
+	 */
+	set_fs(USER_DS);
+#endif
 
-		/*
-		 * Clear TF when entering the signal handler, but
-		 * notify any tracer that was single-stepping it.
-		 * The tracer may want to single-step inside the
-		 * handler too.
-		 */
-		regs->flags &= ~X86_EFLAGS_TF;
+	/*
+	 * Clear the direction flag as per the ABI for function entry.
+	 */
+	regs->flags &= ~X86_EFLAGS_DF;
 
-		spin_lock_irq(&current->sighand->siglock);
-		sigorsets(&current->blocked, &current->blocked, &ka->sa.sa_mask);
-		if (!(ka->sa.sa_flags & SA_NODEFER))
-			sigaddset(&current->blocked, sig);
-		recalc_sigpending();
-		spin_unlock_irq(&current->sighand->siglock);
+	/*
+	 * Clear TF when entering the signal handler, but
+	 * notify any tracer that was single-stepping it.
+	 * The tracer may want to single-step inside the
+	 * handler too.
+	 */
+	regs->flags &= ~X86_EFLAGS_TF;
 
-		tracehook_signal_handler(sig, info, ka, regs,
-					 test_thread_flag(TIF_SINGLESTEP));
-	}
+	spin_lock_irq(&current->sighand->siglock);
+	sigorsets(&current->blocked, &current->blocked, &ka->sa.sa_mask);
+	if (!(ka->sa.sa_flags & SA_NODEFER))
+		sigaddset(&current->blocked, sig);
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
 
-	return ret;
+	tracehook_signal_handler(sig, info, ka, regs,
+				 test_thread_flag(TIF_SINGLESTEP));
+
+	return 0;
 }
 
+#define NR_restart_syscall	\
+	test_thread_flag(TIF_IA32) ? __NR_ia32_restart_syscall : __NR_restart_syscall
 /*
  * Note that 'init' is a special process: it doesn't get signals it doesn't
  * want to handle. Thus you cannot kill init even with a SIGKILL even by
@@ -468,7 +422,8 @@ static void do_signal(struct pt_regs *regs)
 
 	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
 	if (signr > 0) {
-		/* Re-enable any watchpoints before delivering the
+		/*
+		 * Re-enable any watchpoints before delivering the
 		 * signal to user space. The processor register will
 		 * have been cleared if the watchpoint triggered
 		 * inside the kernel.
@@ -476,7 +431,7 @@ static void do_signal(struct pt_regs *regs)
 		if (current->thread.debugreg7)
 			set_debugreg(current->thread.debugreg7, 7);
 
-		/* Whee!  Actually deliver the signal.  */
+		/* Whee! Actually deliver the signal.  */
 		if (handle_signal(signr, &info, &ka, oldset, regs) == 0) {
 			/*
 			 * A signal was successfully delivered; the saved
@@ -499,10 +454,9 @@ static void do_signal(struct pt_regs *regs)
 			regs->ax = regs->orig_ax;
 			regs->ip -= 2;
 			break;
+
 		case -ERESTART_RESTARTBLOCK:
-			regs->ax = test_thread_flag(TIF_IA32) ?
-					__NR_ia32_restart_syscall :
-					__NR_restart_syscall;
+			regs->ax = NR_restart_syscall;
 			regs->ip -= 2;
 			break;
 		}
@@ -518,14 +472,18 @@ static void do_signal(struct pt_regs *regs)
 	}
 }
 
-void do_notify_resume(struct pt_regs *regs, void *unused,
-		      __u32 thread_info_flags)
+/*
+ * notification of userspace execution resumption
+ * - triggered by the TIF_WORK_MASK flags
+ */
+void
+do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags)
 {
-#ifdef CONFIG_X86_MCE
+#if defined(CONFIG_X86_64) && defined(CONFIG_X86_MCE)
 	/* notify userspace of pending MCEs */
 	if (thread_info_flags & _TIF_MCE_NOTIFY)
 		mce_notify_user();
-#endif /* CONFIG_X86_MCE */
+#endif /* CONFIG_X86_64 && CONFIG_X86_MCE */
 
 	/* deal with pending signal delivery */
 	if (thread_info_flags & _TIF_SIGPENDING)
@@ -535,17 +493,23 @@ void do_notify_resume(struct pt_regs *regs, void *unused,
 		clear_thread_flag(TIF_NOTIFY_RESUME);
 		tracehook_notify_resume(regs);
 	}
+
+#ifdef CONFIG_X86_32
+	clear_thread_flag(TIF_IRET);
+#endif /* CONFIG_X86_32 */
 }
 
 void signal_fault(struct pt_regs *regs, void __user *frame, char *where)
 {
 	struct task_struct *me = current;
+
 	if (show_unhandled_signals && printk_ratelimit()) {
-		printk("%s[%d] bad frame in %s frame:%p ip:%lx sp:%lx orax:%lx",
-	       me->comm, me->pid, where, frame, regs->ip,
-		   regs->sp, regs->orig_ax);
+		printk(KERN_INFO
+		       "%s[%d] bad frame in %s frame:%p ip:%lx sp:%lx orax:%lx",
+		       me->comm, me->pid, where, frame,
+		       regs->ip, regs->sp, regs->orig_ax);
 		print_vma_addr(" in ", regs->ip);
-		printk("\n");
+		printk(KERN_CONT "\n");
 	}
 
 	force_sig(SIGSEGV, me);
diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c
index 361b7a4c640c..18f9b19f5f8f 100644
--- a/arch/x86/kernel/smp.c
+++ b/arch/x86/kernel/smp.c
@@ -214,12 +214,16 @@ void smp_call_function_single_interrupt(struct pt_regs *regs)
 struct smp_ops smp_ops = {
 	.smp_prepare_boot_cpu = native_smp_prepare_boot_cpu,
 	.smp_prepare_cpus = native_smp_prepare_cpus,
-	.cpu_up = native_cpu_up,
 	.smp_cpus_done = native_smp_cpus_done,
 
 	.smp_send_stop = native_smp_send_stop,
 	.smp_send_reschedule = native_smp_send_reschedule,
 
+	.cpu_up = native_cpu_up,
+	.cpu_die = native_cpu_die,
+	.cpu_disable = native_cpu_disable,
+	.play_dead = native_play_dead,
+
 	.send_call_func_ipi = native_send_call_func_ipi,
 	.send_call_func_single_ipi = native_send_call_func_single_ipi,
 };
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index 4e7ccb0e2a9b..8c3aca7cb343 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -52,6 +52,7 @@
 #include <asm/desc.h>
 #include <asm/nmi.h>
 #include <asm/irq.h>
+#include <asm/idle.h>
 #include <asm/smp.h>
 #include <asm/trampoline.h>
 #include <asm/cpu.h>
@@ -123,7 +124,6 @@ EXPORT_PER_CPU_SYMBOL(cpu_info);
 
 static atomic_t init_deasserted;
 
-static int boot_cpu_logical_apicid;
 
 /* representing cpus for which sibling maps can be computed */
 static cpumask_t cpu_sibling_setup_map;
@@ -165,6 +165,8 @@ static void unmap_cpu_to_node(int cpu)
 #endif
 
 #ifdef CONFIG_X86_32
+static int boot_cpu_logical_apicid;
+
 u8 cpu_2_logical_apicid[NR_CPUS] __read_mostly =
 					{ [0 ... NR_CPUS-1] = BAD_APICID };
 
@@ -210,7 +212,7 @@ static void __cpuinit smp_callin(void)
 	/*
 	 * (This works even if the APIC is not enabled.)
 	 */
-	phys_id = GET_APIC_ID(read_apic_id());
+	phys_id = read_apic_id();
 	cpuid = smp_processor_id();
 	if (cpu_isset(cpuid, cpu_callin_map)) {
 		panic("%s: phys CPU#%d, CPU#%d already present??\n", __func__,
@@ -332,14 +334,17 @@ static void __cpuinit start_secondary(void *unused)
 	 * does not change while we are assigning vectors to cpus.  Holding
 	 * this lock ensures we don't half assign or remove an irq from a cpu.
 	 */
-	ipi_call_lock_irq();
+	ipi_call_lock();
 	lock_vector_lock();
 	__setup_vector_irq(smp_processor_id());
 	cpu_set(smp_processor_id(), cpu_online_map);
 	unlock_vector_lock();
-	ipi_call_unlock_irq();
+	ipi_call_unlock();
 	per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE;
 
+	/* enable local interrupts */
+	local_irq_enable();
+
 	setup_secondary_clock();
 
 	wmb();
@@ -551,8 +556,7 @@ static inline void __inquire_remote_apic(int apicid)
 			printk(KERN_CONT
 			       "a previous APIC delivery may have failed\n");
 
-		apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(apicid));
-		apic_write(APIC_ICR, APIC_DM_REMRD | regs[i]);
+		apic_icr_write(APIC_DM_REMRD | regs[i], apicid);
 
 		timeout = 0;
 		do {
@@ -584,11 +588,9 @@ wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip)
 	int maxlvt;
 
 	/* Target chip */
-	apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(logical_apicid));
-
 	/* Boot on the stack */
 	/* Kick the second */
-	apic_write(APIC_ICR, APIC_DM_NMI | APIC_DEST_LOGICAL);
+	apic_icr_write(APIC_DM_NMI | APIC_DEST_LOGICAL, logical_apicid);
 
 	pr_debug("Waiting for send to finish...\n");
 	send_status = safe_apic_wait_icr_idle();
@@ -597,10 +599,12 @@ wakeup_secondary_cpu(int logical_apicid, unsigned long start_eip)
 	 * Give the other CPU some time to accept the IPI.
 	 */
 	udelay(200);
-	maxlvt = lapic_get_maxlvt();
-	if (maxlvt > 3)			/* Due to the Pentium erratum 3AP.  */
-		apic_write(APIC_ESR, 0);
-	accept_status = (apic_read(APIC_ESR) & 0xEF);
+	if (APIC_INTEGRATED(apic_version[phys_apicid])) {
+		maxlvt = lapic_get_maxlvt();
+		if (maxlvt > 3)			/* Due to the Pentium erratum 3AP.  */
+			apic_write(APIC_ESR, 0);
+		accept_status = (apic_read(APIC_ESR) & 0xEF);
+	}
 	pr_debug("NMI sent.\n");
 
 	if (send_status)
@@ -641,13 +645,11 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip)
 	/*
 	 * Turn INIT on target chip
 	 */
-	apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));
-
 	/*
 	 * Send IPI
 	 */
-	apic_write(APIC_ICR,
-		   APIC_INT_LEVELTRIG | APIC_INT_ASSERT | APIC_DM_INIT);
+	apic_icr_write(APIC_INT_LEVELTRIG | APIC_INT_ASSERT | APIC_DM_INIT,
+		       phys_apicid);
 
 	pr_debug("Waiting for send to finish...\n");
 	send_status = safe_apic_wait_icr_idle();
@@ -657,10 +659,8 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip)
 	pr_debug("Deasserting INIT.\n");
 
 	/* Target chip */
-	apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));
-
 	/* Send IPI */
-	apic_write(APIC_ICR, APIC_INT_LEVELTRIG | APIC_DM_INIT);
+	apic_icr_write(APIC_INT_LEVELTRIG | APIC_DM_INIT, phys_apicid);
 
 	pr_debug("Waiting for send to finish...\n");
 	send_status = safe_apic_wait_icr_idle();
@@ -703,11 +703,10 @@ wakeup_secondary_cpu(int phys_apicid, unsigned long start_eip)
 		 */
 
 		/* Target chip */
-		apic_write(APIC_ICR2, SET_APIC_DEST_FIELD(phys_apicid));
-
 		/* Boot on the stack */
 		/* Kick the second */
-		apic_write(APIC_ICR, APIC_DM_STARTUP | (start_eip >> 12));
+		apic_icr_write(APIC_DM_STARTUP | (start_eip >> 12),
+			       phys_apicid);
 
 		/*
 		 * Give the other CPU some time to accept the IPI.
@@ -1176,10 +1175,17 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus)
 	 * Setup boot CPU information
 	 */
 	smp_store_cpu_info(0); /* Final full version of the data */
+#ifdef CONFIG_X86_32
 	boot_cpu_logical_apicid = logical_smp_processor_id();
+#endif
 	current_thread_info()->cpu = 0;  /* needed? */
 	set_cpu_sibling_map(0);
 
+#ifdef CONFIG_X86_64
+	enable_IR_x2apic();
+	setup_apic_routing();
+#endif
+
 	if (smp_sanity_check(max_cpus) < 0) {
 		printk(KERN_INFO "SMP disabled\n");
 		disable_smp();
@@ -1187,9 +1193,9 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus)
 	}
 
 	preempt_disable();
-	if (GET_APIC_ID(read_apic_id()) != boot_cpu_physical_apicid) {
+	if (read_apic_id() != boot_cpu_physical_apicid) {
 		panic("Boot APIC ID in local APIC unexpected (%d vs %d)",
-		     GET_APIC_ID(read_apic_id()), boot_cpu_physical_apicid);
+		     read_apic_id(), boot_cpu_physical_apicid);
 		/* Or can we switch back to PIC here? */
 	}
 	preempt_enable();
@@ -1255,39 +1261,6 @@ void __init native_smp_cpus_done(unsigned int max_cpus)
 	check_nmi_watchdog();
 }
 
-#ifdef CONFIG_HOTPLUG_CPU
-
-static void remove_siblinginfo(int cpu)
-{
-	int sibling;
-	struct cpuinfo_x86 *c = &cpu_data(cpu);
-
-	for_each_cpu_mask_nr(sibling, per_cpu(cpu_core_map, cpu)) {
-		cpu_clear(cpu, per_cpu(cpu_core_map, sibling));
-		/*/
-		 * last thread sibling in this cpu core going down
-		 */
-		if (cpus_weight(per_cpu(cpu_sibling_map, cpu)) == 1)
-			cpu_data(sibling).booted_cores--;
-	}
-
-	for_each_cpu_mask_nr(sibling, per_cpu(cpu_sibling_map, cpu))
-		cpu_clear(cpu, per_cpu(cpu_sibling_map, sibling));
-	cpus_clear(per_cpu(cpu_sibling_map, cpu));
-	cpus_clear(per_cpu(cpu_core_map, cpu));
-	c->phys_proc_id = 0;
-	c->cpu_core_id = 0;
-	cpu_clear(cpu, cpu_sibling_setup_map);
-}
-
-static int additional_cpus __initdata = -1;
-
-static __init int setup_additional_cpus(char *s)
-{
-	return s && get_option(&s, &additional_cpus) ? 0 : -EINVAL;
-}
-early_param("additional_cpus", setup_additional_cpus);
-
 /*
  * cpu_possible_map should be static, it cannot change as cpu's
  * are onlined, or offlined. The reason is per-cpu data-structures
@@ -1307,21 +1280,13 @@ early_param("additional_cpus", setup_additional_cpus);
  */
 __init void prefill_possible_map(void)
 {
-	int i;
-	int possible;
+	int i, possible;
 
 	/* no processor from mptable or madt */
 	if (!num_processors)
 		num_processors = 1;
 
-	if (additional_cpus == -1) {
-		if (disabled_cpus > 0)
-			additional_cpus = disabled_cpus;
-		else
-			additional_cpus = 0;
-	}
-
-	possible = num_processors + additional_cpus;
+	possible = num_processors + disabled_cpus;
 	if (possible > NR_CPUS)
 		possible = NR_CPUS;
 
@@ -1334,6 +1299,31 @@ __init void prefill_possible_map(void)
 	nr_cpu_ids = possible;
 }
 
+#ifdef CONFIG_HOTPLUG_CPU
+
+static void remove_siblinginfo(int cpu)
+{
+	int sibling;
+	struct cpuinfo_x86 *c = &cpu_data(cpu);
+
+	for_each_cpu_mask_nr(sibling, per_cpu(cpu_core_map, cpu)) {
+		cpu_clear(cpu, per_cpu(cpu_core_map, sibling));
+		/*/
+		 * last thread sibling in this cpu core going down
+		 */
+		if (cpus_weight(per_cpu(cpu_sibling_map, cpu)) == 1)
+			cpu_data(sibling).booted_cores--;
+	}
+
+	for_each_cpu_mask_nr(sibling, per_cpu(cpu_sibling_map, cpu))
+		cpu_clear(cpu, per_cpu(cpu_sibling_map, sibling));
+	cpus_clear(per_cpu(cpu_sibling_map, cpu));
+	cpus_clear(per_cpu(cpu_core_map, cpu));
+	c->phys_proc_id = 0;
+	c->cpu_core_id = 0;
+	cpu_clear(cpu, cpu_sibling_setup_map);
+}
+
 static void __ref remove_cpu_from_maps(int cpu)
 {
 	cpu_clear(cpu, cpu_online_map);
@@ -1344,25 +1334,9 @@ static void __ref remove_cpu_from_maps(int cpu)
 	numa_remove_cpu(cpu);
 }
 
-int __cpu_disable(void)
+void cpu_disable_common(void)
 {
 	int cpu = smp_processor_id();
-
-	/*
-	 * Perhaps use cpufreq to drop frequency, but that could go
-	 * into generic code.
-	 *
-	 * We won't take down the boot processor on i386 due to some
-	 * interrupts only being able to be serviced by the BSP.
-	 * Especially so if we're not using an IOAPIC	-zwane
-	 */
-	if (cpu == 0)
-		return -EBUSY;
-
-	if (nmi_watchdog == NMI_LOCAL_APIC)
-		stop_apic_nmi_watchdog(NULL);
-	clear_local_APIC();
-
 	/*
 	 * HACK:
 	 * Allow any queued timer interrupts to get serviced
@@ -1380,10 +1354,32 @@ int __cpu_disable(void)
 	remove_cpu_from_maps(cpu);
 	unlock_vector_lock();
 	fixup_irqs(cpu_online_map);
+}
+
+int native_cpu_disable(void)
+{
+	int cpu = smp_processor_id();
+
+	/*
+	 * Perhaps use cpufreq to drop frequency, but that could go
+	 * into generic code.
+	 *
+	 * We won't take down the boot processor on i386 due to some
+	 * interrupts only being able to be serviced by the BSP.
+	 * Especially so if we're not using an IOAPIC	-zwane
+	 */
+	if (cpu == 0)
+		return -EBUSY;
+
+	if (nmi_watchdog == NMI_LOCAL_APIC)
+		stop_apic_nmi_watchdog(NULL);
+	clear_local_APIC();
+
+	cpu_disable_common();
 	return 0;
 }
 
-void __cpu_die(unsigned int cpu)
+void native_cpu_die(unsigned int cpu)
 {
 	/* We don't do anything here: idle task is faking death itself. */
 	unsigned int i;
@@ -1400,15 +1396,45 @@ void __cpu_die(unsigned int cpu)
 	}
 	printk(KERN_ERR "CPU %u didn't die...\n", cpu);
 }
+
+void play_dead_common(void)
+{
+	idle_task_exit();
+	reset_lazy_tlbstate();
+	irq_ctx_exit(raw_smp_processor_id());
+	c1e_remove_cpu(raw_smp_processor_id());
+
+	mb();
+	/* Ack it */
+	__get_cpu_var(cpu_state) = CPU_DEAD;
+
+	/*
+	 * With physical CPU hotplug, we should halt the cpu
+	 */
+	local_irq_disable();
+}
+
+void native_play_dead(void)
+{
+	play_dead_common();
+	wbinvd_halt();
+}
+
 #else /* ... !CONFIG_HOTPLUG_CPU */
-int __cpu_disable(void)
+int native_cpu_disable(void)
 {
 	return -ENOSYS;
 }
 
-void __cpu_die(unsigned int cpu)
+void native_cpu_die(unsigned int cpu)
 {
 	/* We said "no" in __cpu_disable */
 	BUG();
 }
+
+void native_play_dead(void)
+{
+	BUG();
+}
+
 #endif
diff --git a/arch/x86/kernel/summit_32.c b/arch/x86/kernel/summit_32.c
index d67ce5f044ba..7b987852e876 100644
--- a/arch/x86/kernel/summit_32.c
+++ b/arch/x86/kernel/summit_32.c
@@ -30,7 +30,7 @@
 #include <linux/init.h>
 #include <asm/io.h>
 #include <asm/bios_ebda.h>
-#include <asm/mach-summit/mach_mpparse.h>
+#include <asm/summit/mpparse.h>
 
 static struct rio_table_hdr *rio_table_hdr __initdata;
 static struct scal_detail   *scal_devs[MAX_NUMNODES] __initdata;
diff --git a/arch/x86/kernel/time_32.c b/arch/x86/kernel/time_32.c
index bbecf8b6bf96..77b400f06ea2 100644
--- a/arch/x86/kernel/time_32.c
+++ b/arch/x86/kernel/time_32.c
@@ -47,10 +47,9 @@ unsigned long profile_pc(struct pt_regs *regs)
 	unsigned long pc = instruction_pointer(regs);
 
 #ifdef CONFIG_SMP
-	if (!v8086_mode(regs) && SEGMENT_IS_KERNEL_CODE(regs->cs) &&
-	    in_lock_functions(pc)) {
+	if (!user_mode_vm(regs) && in_lock_functions(pc)) {
 #ifdef CONFIG_FRAME_POINTER
-		return *(unsigned long *)(regs->bp + 4);
+		return *(unsigned long *)(regs->bp + sizeof(long));
 #else
 		unsigned long *sp = (unsigned long *)&regs->sp;
 
@@ -95,6 +94,7 @@ irqreturn_t timer_interrupt(int irq, void *dev_id)
 
 	do_timer_interrupt_hook();
 
+#ifdef CONFIG_MCA
 	if (MCA_bus) {
 		/* The PS/2 uses level-triggered interrupts.  You can't
 		turn them off, nor would you want to (any attempt to
@@ -108,6 +108,7 @@ irqreturn_t timer_interrupt(int irq, void *dev_id)
 		u8 irq_v = inb_p( 0x61 );	/* read the current state */
 		outb_p( irq_v|0x80, 0x61 );	/* reset the IRQ */
 	}
+#endif
 
 	return IRQ_HANDLED;
 }
diff --git a/arch/x86/kernel/time_64.c b/arch/x86/kernel/time_64.c
index e3d49c553af2..cb19d650c216 100644
--- a/arch/x86/kernel/time_64.c
+++ b/arch/x86/kernel/time_64.c
@@ -16,6 +16,7 @@
 #include <linux/interrupt.h>
 #include <linux/module.h>
 #include <linux/time.h>
+#include <linux/mca.h>
 
 #include <asm/i8253.h>
 #include <asm/hpet.h>
@@ -33,23 +34,34 @@ unsigned long profile_pc(struct pt_regs *regs)
 	/* Assume the lock function has either no stack frame or a copy
 	   of flags from PUSHF
 	   Eflags always has bits 22 and up cleared unlike kernel addresses. */
-	if (!user_mode(regs) && in_lock_functions(pc)) {
+	if (!user_mode_vm(regs) && in_lock_functions(pc)) {
+#ifdef CONFIG_FRAME_POINTER
+		return *(unsigned long *)(regs->bp + sizeof(long));
+#else
 		unsigned long *sp = (unsigned long *)regs->sp;
 		if (sp[0] >> 22)
 			return sp[0];
 		if (sp[1] >> 22)
 			return sp[1];
+#endif
 	}
 	return pc;
 }
 EXPORT_SYMBOL(profile_pc);
 
-static irqreturn_t timer_event_interrupt(int irq, void *dev_id)
+irqreturn_t timer_interrupt(int irq, void *dev_id)
 {
 	add_pda(irq0_irqs, 1);
 
 	global_clock_event->event_handler(global_clock_event);
 
+#ifdef CONFIG_MCA
+	if (MCA_bus) {
+		u8 irq_v = inb_p(0x61);       /* read the current state */
+		outb_p(irq_v|0x80, 0x61);     /* reset the IRQ */
+	}
+#endif
+
 	return IRQ_HANDLED;
 }
 
@@ -100,7 +112,7 @@ unsigned long __init calibrate_cpu(void)
 }
 
 static struct irqaction irq0 = {
-	.handler	= timer_event_interrupt,
+	.handler	= timer_interrupt,
 	.flags		= IRQF_DISABLED | IRQF_IRQPOLL | IRQF_NOBALANCING,
 	.mask		= CPU_MASK_NONE,
 	.name		= "timer"
@@ -111,16 +123,13 @@ void __init hpet_time_init(void)
 	if (!hpet_enable())
 		setup_pit_timer();
 
+	irq0.mask = cpumask_of_cpu(0);
 	setup_irq(0, &irq0);
 }
 
 void __init time_init(void)
 {
 	tsc_init();
-	if (cpu_has(&boot_cpu_data, X86_FEATURE_RDTSCP))
-		vgetcpu_mode = VGETCPU_RDTSCP;
-	else
-		vgetcpu_mode = VGETCPU_LSL;
 
 	late_time_init = choose_time_init();
 }
diff --git a/arch/x86/kernel/tlb_32.c b/arch/x86/kernel/tlb_32.c
index fec1ecedc9b7..e00534b33534 100644
--- a/arch/x86/kernel/tlb_32.c
+++ b/arch/x86/kernel/tlb_32.c
@@ -241,3 +241,11 @@ void flush_tlb_all(void)
 	on_each_cpu(do_flush_tlb_all, NULL, 1);
 }
 
+void reset_lazy_tlbstate(void)
+{
+	int cpu = raw_smp_processor_id();
+
+	per_cpu(cpu_tlbstate, cpu).state = 0;
+	per_cpu(cpu_tlbstate, cpu).active_mm = &init_mm;
+}
+
diff --git a/arch/x86/kernel/traps_32.c b/arch/x86/kernel/traps.c
index 03df8e45e5a1..e062974cce34 100644
--- a/arch/x86/kernel/traps_32.c
+++ b/arch/x86/kernel/traps.c
@@ -7,13 +7,11 @@
  */
 
 /*
- * 'Traps.c' handles hardware traps and faults after we have saved some
- * state in 'asm.s'.
+ * Handle hardware traps and faults.
  */
 #include <linux/interrupt.h>
 #include <linux/kallsyms.h>
 #include <linux/spinlock.h>
-#include <linux/highmem.h>
 #include <linux/kprobes.h>
 #include <linux/uaccess.h>
 #include <linux/utsname.h>
@@ -32,6 +30,8 @@
 #include <linux/bug.h>
 #include <linux/nmi.h>
 #include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/io.h>
 
 #ifdef CONFIG_EISA
 #include <linux/ioport.h>
@@ -46,21 +46,31 @@
 #include <linux/edac.h>
 #endif
 
-#include <asm/arch_hooks.h>
 #include <asm/stacktrace.h>
 #include <asm/processor.h>
 #include <asm/debugreg.h>
 #include <asm/atomic.h>
 #include <asm/system.h>
 #include <asm/unwind.h>
+#include <asm/traps.h>
 #include <asm/desc.h>
 #include <asm/i387.h>
+
+#include <mach_traps.h>
+
+#ifdef CONFIG_X86_64
+#include <asm/pgalloc.h>
+#include <asm/proto.h>
+#include <asm/pda.h>
+#else
+#include <asm/processor-flags.h>
+#include <asm/arch_hooks.h>
 #include <asm/nmi.h>
 #include <asm/smp.h>
 #include <asm/io.h>
 #include <asm/traps.h>
 
-#include "mach_traps.h"
+#include "cpu/mcheck/mce.h"
 
 DECLARE_BITMAP(used_vectors, NR_VECTORS);
 EXPORT_SYMBOL_GPL(used_vectors);
@@ -77,418 +87,104 @@ char ignore_fpu_irq;
  */
 gate_desc idt_table[256]
 	__attribute__((__section__(".data.idt"))) = { { { { 0, 0 } } }, };
-
-int panic_on_unrecovered_nmi;
-int kstack_depth_to_print = 24;
-static unsigned int code_bytes = 64;
-static int ignore_nmis;
-static int die_counter;
-
-void printk_address(unsigned long address, int reliable)
-{
-#ifdef CONFIG_KALLSYMS
-	unsigned long offset = 0;
-	unsigned long symsize;
-	const char *symname;
-	char *modname;
-	char *delim = ":";
-	char namebuf[KSYM_NAME_LEN];
-	char reliab[4] = "";
-
-	symname = kallsyms_lookup(address, &symsize, &offset,
-					&modname, namebuf);
-	if (!symname) {
-		printk(" [<%08lx>]\n", address);
-		return;
-	}
-	if (!reliable)
-		strcpy(reliab, "? ");
-
-	if (!modname)
-		modname = delim = "";
-	printk(" [<%08lx>] %s%s%s%s%s+0x%lx/0x%lx\n",
-		address, reliab, delim, modname, delim, symname, offset, symsize);
-#else
-	printk(" [<%08lx>]\n", address);
 #endif
-}
-
-static inline int valid_stack_ptr(struct thread_info *tinfo,
-			void *p, unsigned int size)
-{
-	void *t = tinfo;
-	return	p > t && p <= t + THREAD_SIZE - size;
-}
-
-/* The form of the top of the frame on the stack */
-struct stack_frame {
-	struct stack_frame *next_frame;
-	unsigned long return_address;
-};
-
-static inline unsigned long
-print_context_stack(struct thread_info *tinfo,
-		unsigned long *stack, unsigned long bp,
-		const struct stacktrace_ops *ops, void *data)
-{
-	struct stack_frame *frame = (struct stack_frame *)bp;
-
-	while (valid_stack_ptr(tinfo, stack, sizeof(*stack))) {
-		unsigned long addr;
-
-		addr = *stack;
-		if (__kernel_text_address(addr)) {
-			if ((unsigned long) stack == bp + 4) {
-				ops->address(data, addr, 1);
-				frame = frame->next_frame;
-				bp = (unsigned long) frame;
-			} else {
-				ops->address(data, addr, bp == 0);
-			}
-		}
-		stack++;
-	}
-	return bp;
-}
-
-void dump_trace(struct task_struct *task, struct pt_regs *regs,
-		unsigned long *stack, unsigned long bp,
-		const struct stacktrace_ops *ops, void *data)
-{
-	if (!task)
-		task = current;
-
-	if (!stack) {
-		unsigned long dummy;
-		stack = &dummy;
-		if (task != current)
-			stack = (unsigned long *)task->thread.sp;
-	}
-
-#ifdef CONFIG_FRAME_POINTER
-	if (!bp) {
-		if (task == current) {
-			/* Grab bp right from our regs */
-			asm("movl %%ebp, %0" : "=r" (bp) :);
-		} else {
-			/* bp is the last reg pushed by switch_to */
-			bp = *(unsigned long *) task->thread.sp;
-		}
-	}
-#endif
-
-	for (;;) {
-		struct thread_info *context;
-
-		context = (struct thread_info *)
-			((unsigned long)stack & (~(THREAD_SIZE - 1)));
-		bp = print_context_stack(context, stack, bp, ops, data);
-		/*
-		 * Should be after the line below, but somewhere
-		 * in early boot context comes out corrupted and we
-		 * can't reference it:
-		 */
-		if (ops->stack(data, "IRQ") < 0)
-			break;
-		stack = (unsigned long *)context->previous_esp;
-		if (!stack)
-			break;
-		touch_nmi_watchdog();
-	}
-}
-EXPORT_SYMBOL(dump_trace);
-
-static void
-print_trace_warning_symbol(void *data, char *msg, unsigned long symbol)
-{
-	printk(data);
-	print_symbol(msg, symbol);
-	printk("\n");
-}
-
-static void print_trace_warning(void *data, char *msg)
-{
-	printk("%s%s\n", (char *)data, msg);
-}
 
-static int print_trace_stack(void *data, char *name)
-{
-	return 0;
-}
-
-/*
- * Print one address/symbol entries per line.
- */
-static void print_trace_address(void *data, unsigned long addr, int reliable)
-{
-	printk("%s [<%08lx>] ", (char *)data, addr);
-	if (!reliable)
-		printk("? ");
-	print_symbol("%s\n", addr);
-	touch_nmi_watchdog();
-}
-
-static const struct stacktrace_ops print_trace_ops = {
-	.warning = print_trace_warning,
-	.warning_symbol = print_trace_warning_symbol,
-	.stack = print_trace_stack,
-	.address = print_trace_address,
-};
+static int ignore_nmis;
 
-static void
-show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
-		unsigned long *stack, unsigned long bp, char *log_lvl)
+static inline void conditional_sti(struct pt_regs *regs)
 {
-	dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl);
-	printk("%s =======================\n", log_lvl);
+	if (regs->flags & X86_EFLAGS_IF)
+		local_irq_enable();
 }
 
-void show_trace(struct task_struct *task, struct pt_regs *regs,
-		unsigned long *stack, unsigned long bp)
+static inline void preempt_conditional_sti(struct pt_regs *regs)
 {
-	show_trace_log_lvl(task, regs, stack, bp, "");
+	inc_preempt_count();
+	if (regs->flags & X86_EFLAGS_IF)
+		local_irq_enable();
 }
 
-static void
-show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
-		   unsigned long *sp, unsigned long bp, char *log_lvl)
+static inline void preempt_conditional_cli(struct pt_regs *regs)
 {
-	unsigned long *stack;
-	int i;
-
-	if (sp == NULL) {
-		if (task)
-			sp = (unsigned long *)task->thread.sp;
-		else
-			sp = (unsigned long *)&sp;
-	}
-
-	stack = sp;
-	for (i = 0; i < kstack_depth_to_print; i++) {
-		if (kstack_end(stack))
-			break;
-		if (i && ((i % 8) == 0))
-			printk("\n%s       ", log_lvl);
-		printk("%08lx ", *stack++);
-	}
-	printk("\n%sCall Trace:\n", log_lvl);
-
-	show_trace_log_lvl(task, regs, sp, bp, log_lvl);
+	if (regs->flags & X86_EFLAGS_IF)
+		local_irq_disable();
+	dec_preempt_count();
 }
 
-void show_stack(struct task_struct *task, unsigned long *sp)
+#ifdef CONFIG_X86_32
+static inline void
+die_if_kernel(const char *str, struct pt_regs *regs, long err)
 {
-	printk("       ");
-	show_stack_log_lvl(task, NULL, sp, 0, "");
+	if (!user_mode_vm(regs))
+		die(str, regs, err);
 }
 
 /*
- * The architecture-independent dump_stack generator
+ * Perform the lazy TSS's I/O bitmap copy. If the TSS has an
+ * invalid offset set (the LAZY one) and the faulting thread has
+ * a valid I/O bitmap pointer, we copy the I/O bitmap in the TSS,
+ * we set the offset field correctly and return 1.
  */
-void dump_stack(void)
+static int lazy_iobitmap_copy(void)
 {
-	unsigned long bp = 0;
-	unsigned long stack;
-
-#ifdef CONFIG_FRAME_POINTER
-	if (!bp)
-		asm("movl %%ebp, %0" : "=r" (bp):);
-#endif
-
-	printk("Pid: %d, comm: %.20s %s %s %.*s\n",
-		current->pid, current->comm, print_tainted(),
-		init_utsname()->release,
-		(int)strcspn(init_utsname()->version, " "),
-		init_utsname()->version);
-
-	show_trace(current, NULL, &stack, bp);
-}
-
-EXPORT_SYMBOL(dump_stack);
-
-void show_registers(struct pt_regs *regs)
-{
-	int i;
+	struct thread_struct *thread;
+	struct tss_struct *tss;
+	int cpu;
 
-	print_modules();
-	__show_registers(regs, 0);
+	cpu = get_cpu();
+	tss = &per_cpu(init_tss, cpu);
+	thread = &current->thread;
 
-	printk(KERN_EMERG "Process %.*s (pid: %d, ti=%p task=%p task.ti=%p)",
-		TASK_COMM_LEN, current->comm, task_pid_nr(current),
-		current_thread_info(), current, task_thread_info(current));
-	/*
-	 * When in-kernel, we also print out the stack and code at the
-	 * time of the fault..
-	 */
-	if (!user_mode_vm(regs)) {
-		unsigned int code_prologue = code_bytes * 43 / 64;
-		unsigned int code_len = code_bytes;
-		unsigned char c;
-		u8 *ip;
-
-		printk("\n" KERN_EMERG "Stack: ");
-		show_stack_log_lvl(NULL, regs, &regs->sp, 0, KERN_EMERG);
-
-		printk(KERN_EMERG "Code: ");
-
-		ip = (u8 *)regs->ip - code_prologue;
-		if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) {
-			/* try starting at EIP */
-			ip = (u8 *)regs->ip;
-			code_len = code_len - code_prologue + 1;
-		}
-		for (i = 0; i < code_len; i++, ip++) {
-			if (ip < (u8 *)PAGE_OFFSET ||
-					probe_kernel_address(ip, c)) {
-				printk(" Bad EIP value.");
-				break;
-			}
-			if (ip == (u8 *)regs->ip)
-				printk("<%02x> ", c);
-			else
-				printk("%02x ", c);
+	if (tss->x86_tss.io_bitmap_base == INVALID_IO_BITMAP_OFFSET_LAZY &&
+	    thread->io_bitmap_ptr) {
+		memcpy(tss->io_bitmap, thread->io_bitmap_ptr,
+		       thread->io_bitmap_max);
+		/*
+		 * If the previously set map was extending to higher ports
+		 * than the current one, pad extra space with 0xff (no access).
+		 */
+		if (thread->io_bitmap_max < tss->io_bitmap_max) {
+			memset((char *) tss->io_bitmap +
+				thread->io_bitmap_max, 0xff,
+				tss->io_bitmap_max - thread->io_bitmap_max);
 		}
-	}
-	printk("\n");
-}
-
-int is_valid_bugaddr(unsigned long ip)
-{
-	unsigned short ud2;
-
-	if (ip < PAGE_OFFSET)
-		return 0;
-	if (probe_kernel_address((unsigned short *)ip, ud2))
-		return 0;
-
-	return ud2 == 0x0b0f;
-}
-
-static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED;
-static int die_owner = -1;
-static unsigned int die_nest_count;
-
-unsigned __kprobes long oops_begin(void)
-{
-	unsigned long flags;
-
-	oops_enter();
-
-	if (die_owner != raw_smp_processor_id()) {
-		console_verbose();
-		raw_local_irq_save(flags);
-		__raw_spin_lock(&die_lock);
-		die_owner = smp_processor_id();
-		die_nest_count = 0;
-		bust_spinlocks(1);
-	} else {
-		raw_local_irq_save(flags);
-	}
-	die_nest_count++;
-	return flags;
-}
-
-void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr)
-{
-	bust_spinlocks(0);
-	die_owner = -1;
-	add_taint(TAINT_DIE);
-	__raw_spin_unlock(&die_lock);
-	raw_local_irq_restore(flags);
-
-	if (!regs)
-		return;
-
-	if (kexec_should_crash(current))
-		crash_kexec(regs);
-
-	if (in_interrupt())
-		panic("Fatal exception in interrupt");
-
-	if (panic_on_oops)
-		panic("Fatal exception");
-
-	oops_exit();
-	do_exit(signr);
-}
-
-int __kprobes __die(const char *str, struct pt_regs *regs, long err)
-{
-	unsigned short ss;
-	unsigned long sp;
+		tss->io_bitmap_max = thread->io_bitmap_max;
+		tss->x86_tss.io_bitmap_base = IO_BITMAP_OFFSET;
+		tss->io_bitmap_owner = thread;
+		put_cpu();
 
-	printk(KERN_EMERG "%s: %04lx [#%d] ", str, err & 0xffff, ++die_counter);
-#ifdef CONFIG_PREEMPT
-	printk("PREEMPT ");
-#endif
-#ifdef CONFIG_SMP
-	printk("SMP ");
-#endif
-#ifdef CONFIG_DEBUG_PAGEALLOC
-	printk("DEBUG_PAGEALLOC");
-#endif
-	printk("\n");
-	if (notify_die(DIE_OOPS, str, regs, err,
-			current->thread.trap_no, SIGSEGV) == NOTIFY_STOP)
 		return 1;
-
-	show_registers(regs);
-	/* Executive summary in case the oops scrolled away */
-	sp = (unsigned long) (&regs->sp);
-	savesegment(ss, ss);
-	if (user_mode(regs)) {
-		sp = regs->sp;
-		ss = regs->ss & 0xffff;
 	}
-	printk(KERN_EMERG "EIP: [<%08lx>] ", regs->ip);
-	print_symbol("%s", regs->ip);
-	printk(" SS:ESP %04x:%08lx\n", ss, sp);
-	return 0;
-}
-
-/*
- * This is gone through when something in the kernel has done something bad
- * and is about to be terminated:
- */
-void die(const char *str, struct pt_regs *regs, long err)
-{
-	unsigned long flags = oops_begin();
-
-	if (die_nest_count < 3) {
-		report_bug(regs->ip, regs);
-
-		if (__die(str, regs, err))
-			regs = NULL;
-	} else {
-		printk(KERN_EMERG "Recursive die() failure, output suppressed\n");
-	}
-
-	oops_end(flags, regs, SIGSEGV);
-}
+	put_cpu();
 
-static inline void
-die_if_kernel(const char *str, struct pt_regs *regs, long err)
-{
-	if (!user_mode_vm(regs))
-		die(str, regs, err);
+	return 0;
 }
+#endif
 
 static void __kprobes
-do_trap(int trapnr, int signr, char *str, int vm86, struct pt_regs *regs,
+do_trap(int trapnr, int signr, char *str, struct pt_regs *regs,
 	long error_code, siginfo_t *info)
 {
 	struct task_struct *tsk = current;
 
+#ifdef CONFIG_X86_32
 	if (regs->flags & X86_VM_MASK) {
-		if (vm86)
+		/*
+		 * traps 0, 1, 3, 4, and 5 should be forwarded to vm86.
+		 * On nmi (interrupt 2), do_trap should not be called.
+		 */
+		if (trapnr < 6)
 			goto vm86_trap;
 		goto trap_signal;
 	}
+#endif
 
 	if (!user_mode(regs))
 		goto kernel_trap;
 
+#ifdef CONFIG_X86_32
 trap_signal:
+#endif
 	/*
 	 * We want error_code and trap_no set for userspace faults and
 	 * kernelspace faults which result in die(), but not
@@ -501,6 +197,18 @@ trap_signal:
 	tsk->thread.error_code = error_code;
 	tsk->thread.trap_no = trapnr;
 
+#ifdef CONFIG_X86_64
+	if (show_unhandled_signals && unhandled_signal(tsk, signr) &&
+	    printk_ratelimit()) {
+		printk(KERN_INFO
+		       "%s[%d] trap %s ip:%lx sp:%lx error:%lx",
+		       tsk->comm, tsk->pid, str,
+		       regs->ip, regs->sp, error_code);
+		print_vma_addr(" in ", regs->ip);
+		printk("\n");
+	}
+#endif
+
 	if (info)
 		force_sig_info(signr, info, tsk);
 	else
@@ -515,29 +223,29 @@ kernel_trap:
 	}
 	return;
 
+#ifdef CONFIG_X86_32
 vm86_trap:
 	if (handle_vm86_trap((struct kernel_vm86_regs *) regs,
 						error_code, trapnr))
 		goto trap_signal;
 	return;
+#endif
 }
 
 #define DO_ERROR(trapnr, signr, str, name)				\
-void do_##name(struct pt_regs *regs, long error_code)			\
+dotraplinkage void do_##name(struct pt_regs *regs, long error_code)	\
 {									\
-	trace_hardirqs_fixup();						\
 	if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr)	\
 							== NOTIFY_STOP)	\
 		return;							\
-	do_trap(trapnr, signr, str, 0, regs, error_code, NULL);		\
+	conditional_sti(regs);						\
+	do_trap(trapnr, signr, str, regs, error_code, NULL);		\
 }
 
-#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr, irq)	\
-void do_##name(struct pt_regs *regs, long error_code)			\
+#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr)		\
+dotraplinkage void do_##name(struct pt_regs *regs, long error_code)	\
 {									\
 	siginfo_t info;							\
-	if (irq)							\
-		local_irq_enable();					\
 	info.si_signo = signr;						\
 	info.si_errno = 0;						\
 	info.si_code = sicode;						\
@@ -545,90 +253,68 @@ void do_##name(struct pt_regs *regs, long error_code)			\
 	if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr)	\
 							== NOTIFY_STOP)	\
 		return;							\
-	do_trap(trapnr, signr, str, 0, regs, error_code, &info);	\
+	conditional_sti(regs);						\
+	do_trap(trapnr, signr, str, regs, error_code, &info);		\
 }
 
-#define DO_VM86_ERROR(trapnr, signr, str, name)				\
-void do_##name(struct pt_regs *regs, long error_code)			\
-{									\
-	if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr)	\
-							== NOTIFY_STOP)	\
-		return;							\
-	do_trap(trapnr, signr, str, 1, regs, error_code, NULL);		\
-}
-
-#define DO_VM86_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr)	\
-void do_##name(struct pt_regs *regs, long error_code)			\
-{									\
-	siginfo_t info;							\
-	info.si_signo = signr;						\
-	info.si_errno = 0;						\
-	info.si_code = sicode;						\
-	info.si_addr = (void __user *)siaddr;				\
-	trace_hardirqs_fixup();						\
-	if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr)	\
-							== NOTIFY_STOP)	\
-		return;							\
-	do_trap(trapnr, signr, str, 1, regs, error_code, &info);	\
-}
-
-DO_VM86_ERROR_INFO(0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->ip)
-#ifndef CONFIG_KPROBES
-DO_VM86_ERROR(3, SIGTRAP, "int3", int3)
-#endif
-DO_VM86_ERROR(4, SIGSEGV, "overflow", overflow)
-DO_VM86_ERROR(5, SIGSEGV, "bounds", bounds)
-DO_ERROR_INFO(6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip, 0)
+DO_ERROR_INFO(0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->ip)
+DO_ERROR(4, SIGSEGV, "overflow", overflow)
+DO_ERROR(5, SIGSEGV, "bounds", bounds)
+DO_ERROR_INFO(6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip)
 DO_ERROR(9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun)
 DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS)
 DO_ERROR(11, SIGBUS, "segment not present", segment_not_present)
+#ifdef CONFIG_X86_32
 DO_ERROR(12, SIGBUS, "stack segment", stack_segment)
-DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0, 0)
-DO_ERROR_INFO(32, SIGILL, "iret exception", iret_error, ILL_BADSTK, 0, 1)
+#endif
+DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0)
+
+#ifdef CONFIG_X86_64
+/* Runs on IST stack */
+dotraplinkage void do_stack_segment(struct pt_regs *regs, long error_code)
+{
+	if (notify_die(DIE_TRAP, "stack segment", regs, error_code,
+			12, SIGBUS) == NOTIFY_STOP)
+		return;
+	preempt_conditional_sti(regs);
+	do_trap(12, SIGBUS, "stack segment", regs, error_code, NULL);
+	preempt_conditional_cli(regs);
+}
+
+dotraplinkage void do_double_fault(struct pt_regs *regs, long error_code)
+{
+	static const char str[] = "double fault";
+	struct task_struct *tsk = current;
+
+	/* Return not checked because double check cannot be ignored */
+	notify_die(DIE_TRAP, str, regs, error_code, 8, SIGSEGV);
 
-void __kprobes
+	tsk->thread.error_code = error_code;
+	tsk->thread.trap_no = 8;
+
+	/* This is always a kernel trap and never fixable (and thus must
+	   never return). */
+	for (;;)
+		die(str, regs, error_code);
+}
+#endif
+
+dotraplinkage void __kprobes
 do_general_protection(struct pt_regs *regs, long error_code)
 {
 	struct task_struct *tsk;
-	struct thread_struct *thread;
-	struct tss_struct *tss;
-	int cpu;
 
-	cpu = get_cpu();
-	tss = &per_cpu(init_tss, cpu);
-	thread = &current->thread;
-
-	/*
-	 * Perform the lazy TSS's I/O bitmap copy. If the TSS has an
-	 * invalid offset set (the LAZY one) and the faulting thread has
-	 * a valid I/O bitmap pointer, we copy the I/O bitmap in the TSS
-	 * and we set the offset field correctly. Then we let the CPU to
-	 * restart the faulting instruction.
-	 */
-	if (tss->x86_tss.io_bitmap_base == INVALID_IO_BITMAP_OFFSET_LAZY &&
-	    thread->io_bitmap_ptr) {
-		memcpy(tss->io_bitmap, thread->io_bitmap_ptr,
-		       thread->io_bitmap_max);
-		/*
-		 * If the previously set map was extending to higher ports
-		 * than the current one, pad extra space with 0xff (no access).
-		 */
-		if (thread->io_bitmap_max < tss->io_bitmap_max) {
-			memset((char *) tss->io_bitmap +
-				thread->io_bitmap_max, 0xff,
-				tss->io_bitmap_max - thread->io_bitmap_max);
-		}
-		tss->io_bitmap_max = thread->io_bitmap_max;
-		tss->x86_tss.io_bitmap_base = IO_BITMAP_OFFSET;
-		tss->io_bitmap_owner = thread;
-		put_cpu();
+	conditional_sti(regs);
 
+#ifdef CONFIG_X86_32
+	if (lazy_iobitmap_copy()) {
+		/* restart the faulting instruction */
 		return;
 	}
-	put_cpu();
 
 	if (regs->flags & X86_VM_MASK)
 		goto gp_in_vm86;
+#endif
 
 	tsk = current;
 	if (!user_mode(regs))
@@ -650,10 +336,12 @@ do_general_protection(struct pt_regs *regs, long error_code)
 	force_sig(SIGSEGV, tsk);
 	return;
 
+#ifdef CONFIG_X86_32
 gp_in_vm86:
 	local_irq_enable();
 	handle_vm86_fault((struct kernel_vm86_regs *) regs, error_code);
 	return;
+#endif
 
 gp_in_kernel:
 	if (fixup_exception(regs))
@@ -690,7 +378,8 @@ mem_parity_error(unsigned char reason, struct pt_regs *regs)
 	printk(KERN_EMERG "Dazed and confused, but trying to continue\n");
 
 	/* Clear and disable the memory parity error line. */
-	clear_mem_error(reason);
+	reason = (reason & 0xf) | 4;
+	outb(reason, 0x61);
 }
 
 static notrace __kprobes void
@@ -716,7 +405,8 @@ io_check_error(unsigned char reason, struct pt_regs *regs)
 static notrace __kprobes void
 unknown_nmi_error(unsigned char reason, struct pt_regs *regs)
 {
-	if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
+	if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) ==
+			NOTIFY_STOP)
 		return;
 #ifdef CONFIG_MCA
 	/*
@@ -739,41 +429,6 @@ unknown_nmi_error(unsigned char reason, struct pt_regs *regs)
 	printk(KERN_EMERG "Dazed and confused, but trying to continue\n");
 }
 
-static DEFINE_SPINLOCK(nmi_print_lock);
-
-void notrace __kprobes die_nmi(char *str, struct pt_regs *regs, int do_panic)
-{
-	if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP)
-		return;
-
-	spin_lock(&nmi_print_lock);
-	/*
-	* We are in trouble anyway, lets at least try
-	* to get a message out:
-	*/
-	bust_spinlocks(1);
-	printk(KERN_EMERG "%s", str);
-	printk(" on CPU%d, ip %08lx, registers:\n",
-		smp_processor_id(), regs->ip);
-	show_registers(regs);
-	if (do_panic)
-		panic("Non maskable interrupt");
-	console_silent();
-	spin_unlock(&nmi_print_lock);
-	bust_spinlocks(0);
-
-	/*
-	 * If we are in kernel we are probably nested up pretty bad
-	 * and might aswell get out now while we still can:
-	 */
-	if (!user_mode_vm(regs)) {
-		current->thread.trap_no = 2;
-		crash_kexec(regs);
-	}
-
-	do_exit(SIGSEGV);
-}
-
 static notrace __kprobes void default_do_nmi(struct pt_regs *regs)
 {
 	unsigned char reason = 0;
@@ -812,22 +467,25 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs)
 		mem_parity_error(reason, regs);
 	if (reason & 0x40)
 		io_check_error(reason, regs);
+#ifdef CONFIG_X86_32
 	/*
 	 * Reassert NMI in case it became active meanwhile
 	 * as it's edge-triggered:
 	 */
 	reassert_nmi();
+#endif
 }
 
-notrace __kprobes void do_nmi(struct pt_regs *regs, long error_code)
+dotraplinkage notrace __kprobes void
+do_nmi(struct pt_regs *regs, long error_code)
 {
-	int cpu;
-
 	nmi_enter();
 
-	cpu = smp_processor_id();
-
-	++nmi_count(cpu);
+#ifdef CONFIG_X86_32
+	{ int cpu; cpu = smp_processor_id(); ++nmi_count(cpu); }
+#else
+	add_pda(__nmi_count, 1);
+#endif
 
 	if (!ignore_nmis)
 		default_do_nmi(regs);
@@ -847,21 +505,44 @@ void restart_nmi(void)
 	acpi_nmi_enable();
 }
 
-#ifdef CONFIG_KPROBES
-void __kprobes do_int3(struct pt_regs *regs, long error_code)
+/* May run on IST stack. */
+dotraplinkage void __kprobes do_int3(struct pt_regs *regs, long error_code)
 {
-	trace_hardirqs_fixup();
-
+#ifdef CONFIG_KPROBES
 	if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP)
 			== NOTIFY_STOP)
 		return;
-	/*
-	 * This is an interrupt gate, because kprobes wants interrupts
-	 * disabled. Normal trap handlers don't.
-	 */
-	restore_interrupts(regs);
+#else
+	if (notify_die(DIE_TRAP, "int3", regs, error_code, 3, SIGTRAP)
+			== NOTIFY_STOP)
+		return;
+#endif
+
+	preempt_conditional_sti(regs);
+	do_trap(3, SIGTRAP, "int3", regs, error_code, NULL);
+	preempt_conditional_cli(regs);
+}
 
-	do_trap(3, SIGTRAP, "int3", 1, regs, error_code, NULL);
+#ifdef CONFIG_X86_64
+/* Help handler running on IST stack to switch back to user stack
+   for scheduling or signal handling. The actual stack switch is done in
+   entry.S */
+asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs)
+{
+	struct pt_regs *regs = eregs;
+	/* Did already sync */
+	if (eregs == (struct pt_regs *)eregs->sp)
+		;
+	/* Exception from user space */
+	else if (user_mode(eregs))
+		regs = task_pt_regs(current);
+	/* Exception from kernel and interrupts are enabled. Move to
+	   kernel process stack. */
+	else if (eregs->flags & X86_EFLAGS_IF)
+		regs = (struct pt_regs *)(eregs->sp -= sizeof(struct pt_regs));
+	if (eregs != regs)
+		*regs = *eregs;
+	return regs;
 }
 #endif
 
@@ -886,13 +567,14 @@ void __kprobes do_int3(struct pt_regs *regs, long error_code)
  * about restoring all the debug state, and ptrace doesn't have to
  * find every occurrence of the TF bit that could be saved away even
  * by user code)
+ *
+ * May run on IST stack.
  */
-void __kprobes do_debug(struct pt_regs *regs, long error_code)
+dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code)
 {
 	struct task_struct *tsk = current;
-	unsigned int condition;
-
-	trace_hardirqs_fixup();
+	unsigned long condition;
+	int si_code;
 
 	get_debugreg(condition, 6);
 
@@ -905,9 +587,9 @@ void __kprobes do_debug(struct pt_regs *regs, long error_code)
 	if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code,
 						SIGTRAP) == NOTIFY_STOP)
 		return;
+
 	/* It's safe to allow irq's after DR6 has been saved */
-	if (regs->flags & X86_EFLAGS_IF)
-		local_irq_enable();
+	preempt_conditional_sti(regs);
 
 	/* Mask out spurious debug traps due to lazy DR7 setting */
 	if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) {
@@ -915,8 +597,10 @@ void __kprobes do_debug(struct pt_regs *regs, long error_code)
 			goto clear_dr7;
 	}
 
+#ifdef CONFIG_X86_32
 	if (regs->flags & X86_VM_MASK)
 		goto debug_vm86;
+#endif
 
 	/* Save debug status register where ptrace can see it */
 	tsk->thread.debugreg6 = condition;
@@ -926,17 +610,13 @@ void __kprobes do_debug(struct pt_regs *regs, long error_code)
 	 * kernel space (but re-enable TF when returning to user mode).
 	 */
 	if (condition & DR_STEP) {
-		/*
-		 * We already checked v86 mode above, so we can
-		 * check for kernel mode by just checking the CPL
-		 * of CS.
-		 */
 		if (!user_mode(regs))
 			goto clear_TF_reenable;
 	}
 
+	si_code = get_si_code(condition);
 	/* Ok, finally something we can handle */
-	send_sigtrap(tsk, regs, error_code);
+	send_sigtrap(tsk, regs, error_code, si_code);
 
 	/*
 	 * Disable additional traps. They'll be re-enabled when
@@ -944,18 +624,37 @@ void __kprobes do_debug(struct pt_regs *regs, long error_code)
 	 */
 clear_dr7:
 	set_debugreg(0, 7);
+	preempt_conditional_cli(regs);
 	return;
 
+#ifdef CONFIG_X86_32
 debug_vm86:
 	handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, 1);
+	preempt_conditional_cli(regs);
 	return;
+#endif
 
 clear_TF_reenable:
 	set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
 	regs->flags &= ~X86_EFLAGS_TF;
+	preempt_conditional_cli(regs);
 	return;
 }
 
+#ifdef CONFIG_X86_64
+static int kernel_math_error(struct pt_regs *regs, const char *str, int trapnr)
+{
+	if (fixup_exception(regs))
+		return 1;
+
+	notify_die(DIE_GPF, str, regs, 0, trapnr, SIGFPE);
+	/* Illegal floating point operation in the kernel */
+	current->thread.trap_no = trapnr;
+	die(str, regs, 0);
+	return 0;
+}
+#endif
+
 /*
  * Note that we play around with the 'TS' bit in an attempt to get
  * the correct behaviour even in the presence of the asynchronous
@@ -992,7 +691,9 @@ void math_error(void __user *ip)
 	swd = get_fpu_swd(task);
 	switch (swd & ~cwd & 0x3f) {
 	case 0x000: /* No unmasked exception */
+#ifdef CONFIG_X86_32
 		return;
+#endif
 	default: /* Multiple exceptions */
 		break;
 	case 0x001: /* Invalid Op */
@@ -1020,9 +721,18 @@ void math_error(void __user *ip)
 	force_sig_info(SIGFPE, &info, task);
 }
 
-void do_coprocessor_error(struct pt_regs *regs, long error_code)
+dotraplinkage void do_coprocessor_error(struct pt_regs *regs, long error_code)
 {
+	conditional_sti(regs);
+
+#ifdef CONFIG_X86_32
 	ignore_fpu_irq = 1;
+#else
+	if (!user_mode(regs) &&
+	    kernel_math_error(regs, "kernel x87 math error", 16))
+		return;
+#endif
+
 	math_error((void __user *)regs->ip);
 }
 
@@ -1074,8 +784,12 @@ static void simd_math_error(void __user *ip)
 	force_sig_info(SIGFPE, &info, task);
 }
 
-void do_simd_coprocessor_error(struct pt_regs *regs, long error_code)
+dotraplinkage void
+do_simd_coprocessor_error(struct pt_regs *regs, long error_code)
 {
+	conditional_sti(regs);
+
+#ifdef CONFIG_X86_32
 	if (cpu_has_xmm) {
 		/* Handle SIMD FPU exceptions on PIII+ processors. */
 		ignore_fpu_irq = 1;
@@ -1094,16 +808,25 @@ void do_simd_coprocessor_error(struct pt_regs *regs, long error_code)
 	current->thread.error_code = error_code;
 	die_if_kernel("cache flush denied", regs, error_code);
 	force_sig(SIGSEGV, current);
+#else
+	if (!user_mode(regs) &&
+			kernel_math_error(regs, "kernel simd math error", 19))
+		return;
+	simd_math_error((void __user *)regs->ip);
+#endif
 }
 
-void do_spurious_interrupt_bug(struct pt_regs *regs, long error_code)
+dotraplinkage void
+do_spurious_interrupt_bug(struct pt_regs *regs, long error_code)
 {
+	conditional_sti(regs);
 #if 0
 	/* No need to warn about this any longer. */
 	printk(KERN_INFO "Ignoring P6 Local APIC Spurious Interrupt Bug...\n");
 #endif
 }
 
+#ifdef CONFIG_X86_32
 unsigned long patch_espfix_desc(unsigned long uesp, unsigned long kesp)
 {
 	struct desc_struct *gdt = get_cpu_gdt_table(smp_processor_id());
@@ -1122,6 +845,15 @@ unsigned long patch_espfix_desc(unsigned long uesp, unsigned long kesp)
 
 	return new_kesp;
 }
+#else
+asmlinkage void __attribute__((weak)) smp_thermal_interrupt(void)
+{
+}
+
+asmlinkage void __attribute__((weak)) mce_threshold_interrupt(void)
+{
+}
+#endif
 
 /*
  * 'math_state_restore()' saves the current math information in the
@@ -1154,14 +886,24 @@ asmlinkage void math_state_restore(void)
 	}
 
 	clts();				/* Allow maths ops (or we recurse) */
+#ifdef CONFIG_X86_32
 	restore_fpu(tsk);
+#else
+	/*
+	 * Paranoid restore. send a SIGSEGV if we fail to restore the state.
+	 */
+	if (unlikely(restore_fpu_checking(tsk))) {
+		stts();
+		force_sig(SIGSEGV, tsk);
+		return;
+	}
+#endif
 	thread->status |= TS_USEDFPU;	/* So we fnsave on switch_to() */
 	tsk->fpu_counter++;
 }
 EXPORT_SYMBOL_GPL(math_state_restore);
 
 #ifndef CONFIG_MATH_EMULATION
-
 asmlinkage void math_emulate(long arg)
 {
 	printk(KERN_EMERG
@@ -1170,12 +912,54 @@ asmlinkage void math_emulate(long arg)
 	force_sig(SIGFPE, current);
 	schedule();
 }
-
 #endif /* CONFIG_MATH_EMULATION */
 
+dotraplinkage void __kprobes
+do_device_not_available(struct pt_regs *regs, long error)
+{
+#ifdef CONFIG_X86_32
+	if (read_cr0() & X86_CR0_EM) {
+		conditional_sti(regs);
+		math_emulate(0);
+	} else {
+		math_state_restore(); /* interrupts still off */
+		conditional_sti(regs);
+	}
+#else
+	math_state_restore();
+#endif
+}
+
+#ifdef CONFIG_X86_32
+#ifdef CONFIG_X86_MCE
+dotraplinkage void __kprobes do_machine_check(struct pt_regs *regs, long error)
+{
+	conditional_sti(regs);
+	machine_check_vector(regs, error);
+}
+#endif
+
+dotraplinkage void do_iret_error(struct pt_regs *regs, long error_code)
+{
+	siginfo_t info;
+	local_irq_enable();
+
+	info.si_signo = SIGILL;
+	info.si_errno = 0;
+	info.si_code = ILL_BADSTK;
+	info.si_addr = 0;
+	if (notify_die(DIE_TRAP, "iret exception",
+			regs, error_code, 32, SIGILL) == NOTIFY_STOP)
+		return;
+	do_trap(32, SIGILL, "iret exception", regs, error_code, &info);
+}
+#endif
+
 void __init trap_init(void)
 {
+#ifdef CONFIG_X86_32
 	int i;
+#endif
 
 #ifdef CONFIG_EISA
 	void __iomem *p = early_ioremap(0x0FFFD9, 4);
@@ -1185,29 +969,40 @@ void __init trap_init(void)
 	early_iounmap(p, 4);
 #endif
 
-	set_trap_gate(0, &divide_error);
-	set_intr_gate(1, &debug);
-	set_intr_gate(2, &nmi);
-	set_system_intr_gate(3, &int3); /* int3 can be called from all */
-	set_system_gate(4, &overflow); /* int4 can be called from all */
-	set_trap_gate(5, &bounds);
-	set_trap_gate(6, &invalid_op);
-	set_trap_gate(7, &device_not_available);
+	set_intr_gate(0, &divide_error);
+	set_intr_gate_ist(1, &debug, DEBUG_STACK);
+	set_intr_gate_ist(2, &nmi, NMI_STACK);
+	/* int3 can be called from all */
+	set_system_intr_gate_ist(3, &int3, DEBUG_STACK);
+	/* int4 can be called from all */
+	set_system_intr_gate(4, &overflow);
+	set_intr_gate(5, &bounds);
+	set_intr_gate(6, &invalid_op);
+	set_intr_gate(7, &device_not_available);
+#ifdef CONFIG_X86_32
 	set_task_gate(8, GDT_ENTRY_DOUBLEFAULT_TSS);
-	set_trap_gate(9, &coprocessor_segment_overrun);
-	set_trap_gate(10, &invalid_TSS);
-	set_trap_gate(11, &segment_not_present);
-	set_trap_gate(12, &stack_segment);
-	set_trap_gate(13, &general_protection);
+#else
+	set_intr_gate_ist(8, &double_fault, DOUBLEFAULT_STACK);
+#endif
+	set_intr_gate(9, &coprocessor_segment_overrun);
+	set_intr_gate(10, &invalid_TSS);
+	set_intr_gate(11, &segment_not_present);
+	set_intr_gate_ist(12, &stack_segment, STACKFAULT_STACK);
+	set_intr_gate(13, &general_protection);
 	set_intr_gate(14, &page_fault);
-	set_trap_gate(15, &spurious_interrupt_bug);
-	set_trap_gate(16, &coprocessor_error);
-	set_trap_gate(17, &alignment_check);
+	set_intr_gate(15, &spurious_interrupt_bug);
+	set_intr_gate(16, &coprocessor_error);
+	set_intr_gate(17, &alignment_check);
 #ifdef CONFIG_X86_MCE
-	set_trap_gate(18, &machine_check);
+	set_intr_gate_ist(18, &machine_check, MCE_STACK);
 #endif
-	set_trap_gate(19, &simd_coprocessor_error);
+	set_intr_gate(19, &simd_coprocessor_error);
 
+#ifdef CONFIG_IA32_EMULATION
+	set_system_intr_gate(IA32_SYSCALL_VECTOR, ia32_syscall);
+#endif
+
+#ifdef CONFIG_X86_32
 	if (cpu_has_fxsr) {
 		printk(KERN_INFO "Enabling fast FPU save and restore... ");
 		set_in_cr4(X86_CR4_OSFXSR);
@@ -1220,37 +1015,20 @@ void __init trap_init(void)
 		printk("done.\n");
 	}
 
-	set_system_gate(SYSCALL_VECTOR, &system_call);
+	set_system_trap_gate(SYSCALL_VECTOR, &system_call);
 
 	/* Reserve all the builtin and the syscall vector: */
 	for (i = 0; i < FIRST_EXTERNAL_VECTOR; i++)
 		set_bit(i, used_vectors);
 
 	set_bit(SYSCALL_VECTOR, used_vectors);
-
-	init_thread_xstate();
+#endif
 	/*
 	 * Should be a barrier for any external CPU state:
 	 */
 	cpu_init();
 
+#ifdef CONFIG_X86_32
 	trap_init_hook();
+#endif
 }
-
-static int __init kstack_setup(char *s)
-{
-	kstack_depth_to_print = simple_strtoul(s, NULL, 0);
-
-	return 1;
-}
-__setup("kstack=", kstack_setup);
-
-static int __init code_bytes_setup(char *s)
-{
-	code_bytes = simple_strtoul(s, NULL, 0);
-	if (code_bytes > 8192)
-		code_bytes = 8192;
-
-	return 1;
-}
-__setup("code_bytes=", code_bytes_setup);
diff --git a/arch/x86/kernel/traps_64.c b/arch/x86/kernel/traps_64.c
deleted file mode 100644
index 7a31f104bef9..000000000000
--- a/arch/x86/kernel/traps_64.c
+++ /dev/null
@@ -1,1218 +0,0 @@
-/*
- *  Copyright (C) 1991, 1992  Linus Torvalds
- *  Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs
- *
- *  Pentium III FXSR, SSE support
- *	Gareth Hughes <gareth@valinux.com>, May 2000
- */
-
-/*
- * 'Traps.c' handles hardware traps and faults after we have saved some
- * state in 'entry.S'.
- */
-#include <linux/moduleparam.h>
-#include <linux/interrupt.h>
-#include <linux/kallsyms.h>
-#include <linux/spinlock.h>
-#include <linux/kprobes.h>
-#include <linux/uaccess.h>
-#include <linux/utsname.h>
-#include <linux/kdebug.h>
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/ptrace.h>
-#include <linux/string.h>
-#include <linux/unwind.h>
-#include <linux/delay.h>
-#include <linux/errno.h>
-#include <linux/kexec.h>
-#include <linux/sched.h>
-#include <linux/timer.h>
-#include <linux/init.h>
-#include <linux/bug.h>
-#include <linux/nmi.h>
-#include <linux/mm.h>
-#include <linux/smp.h>
-#include <linux/io.h>
-
-#if defined(CONFIG_EDAC)
-#include <linux/edac.h>
-#endif
-
-#include <asm/stacktrace.h>
-#include <asm/processor.h>
-#include <asm/debugreg.h>
-#include <asm/atomic.h>
-#include <asm/system.h>
-#include <asm/unwind.h>
-#include <asm/desc.h>
-#include <asm/i387.h>
-#include <asm/pgalloc.h>
-#include <asm/proto.h>
-#include <asm/pda.h>
-#include <asm/traps.h>
-
-#include <mach_traps.h>
-
-int panic_on_unrecovered_nmi;
-int kstack_depth_to_print = 12;
-static unsigned int code_bytes = 64;
-static int ignore_nmis;
-static int die_counter;
-
-static inline void conditional_sti(struct pt_regs *regs)
-{
-	if (regs->flags & X86_EFLAGS_IF)
-		local_irq_enable();
-}
-
-static inline void preempt_conditional_sti(struct pt_regs *regs)
-{
-	inc_preempt_count();
-	if (regs->flags & X86_EFLAGS_IF)
-		local_irq_enable();
-}
-
-static inline void preempt_conditional_cli(struct pt_regs *regs)
-{
-	if (regs->flags & X86_EFLAGS_IF)
-		local_irq_disable();
-	/* Make sure to not schedule here because we could be running
-	   on an exception stack. */
-	dec_preempt_count();
-}
-
-void printk_address(unsigned long address, int reliable)
-{
-	printk(" [<%016lx>] %s%pS\n",
-			address, reliable ?	"" : "? ", (void *) address);
-}
-
-static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
-					unsigned *usedp, char **idp)
-{
-	static char ids[][8] = {
-		[DEBUG_STACK - 1] = "#DB",
-		[NMI_STACK - 1] = "NMI",
-		[DOUBLEFAULT_STACK - 1] = "#DF",
-		[STACKFAULT_STACK - 1] = "#SS",
-		[MCE_STACK - 1] = "#MC",
-#if DEBUG_STKSZ > EXCEPTION_STKSZ
-		[N_EXCEPTION_STACKS ...
-			N_EXCEPTION_STACKS + DEBUG_STKSZ / EXCEPTION_STKSZ - 2] = "#DB[?]"
-#endif
-	};
-	unsigned k;
-
-	/*
-	 * Iterate over all exception stacks, and figure out whether
-	 * 'stack' is in one of them:
-	 */
-	for (k = 0; k < N_EXCEPTION_STACKS; k++) {
-		unsigned long end = per_cpu(orig_ist, cpu).ist[k];
-		/*
-		 * Is 'stack' above this exception frame's end?
-		 * If yes then skip to the next frame.
-		 */
-		if (stack >= end)
-			continue;
-		/*
-		 * Is 'stack' above this exception frame's start address?
-		 * If yes then we found the right frame.
-		 */
-		if (stack >= end - EXCEPTION_STKSZ) {
-			/*
-			 * Make sure we only iterate through an exception
-			 * stack once. If it comes up for the second time
-			 * then there's something wrong going on - just
-			 * break out and return NULL:
-			 */
-			if (*usedp & (1U << k))
-				break;
-			*usedp |= 1U << k;
-			*idp = ids[k];
-			return (unsigned long *)end;
-		}
-		/*
-		 * If this is a debug stack, and if it has a larger size than
-		 * the usual exception stacks, then 'stack' might still
-		 * be within the lower portion of the debug stack:
-		 */
-#if DEBUG_STKSZ > EXCEPTION_STKSZ
-		if (k == DEBUG_STACK - 1 && stack >= end - DEBUG_STKSZ) {
-			unsigned j = N_EXCEPTION_STACKS - 1;
-
-			/*
-			 * Black magic. A large debug stack is composed of
-			 * multiple exception stack entries, which we
-			 * iterate through now. Dont look:
-			 */
-			do {
-				++j;
-				end -= EXCEPTION_STKSZ;
-				ids[j][4] = '1' + (j - N_EXCEPTION_STACKS);
-			} while (stack < end - EXCEPTION_STKSZ);
-			if (*usedp & (1U << j))
-				break;
-			*usedp |= 1U << j;
-			*idp = ids[j];
-			return (unsigned long *)end;
-		}
-#endif
-	}
-	return NULL;
-}
-
-/*
- * x86-64 can have up to three kernel stacks:
- * process stack
- * interrupt stack
- * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack
- */
-
-static inline int valid_stack_ptr(struct thread_info *tinfo,
-			void *p, unsigned int size, void *end)
-{
-	void *t = tinfo;
-	if (end) {
-		if (p < end && p >= (end-THREAD_SIZE))
-			return 1;
-		else
-			return 0;
-	}
-	return p > t && p < t + THREAD_SIZE - size;
-}
-
-/* The form of the top of the frame on the stack */
-struct stack_frame {
-	struct stack_frame *next_frame;
-	unsigned long return_address;
-};
-
-static inline unsigned long
-print_context_stack(struct thread_info *tinfo,
-		unsigned long *stack, unsigned long bp,
-		const struct stacktrace_ops *ops, void *data,
-		unsigned long *end)
-{
-	struct stack_frame *frame = (struct stack_frame *)bp;
-
-	while (valid_stack_ptr(tinfo, stack, sizeof(*stack), end)) {
-		unsigned long addr;
-
-		addr = *stack;
-		if (__kernel_text_address(addr)) {
-			if ((unsigned long) stack == bp + 8) {
-				ops->address(data, addr, 1);
-				frame = frame->next_frame;
-				bp = (unsigned long) frame;
-			} else {
-				ops->address(data, addr, bp == 0);
-			}
-		}
-		stack++;
-	}
-	return bp;
-}
-
-void dump_trace(struct task_struct *task, struct pt_regs *regs,
-		unsigned long *stack, unsigned long bp,
-		const struct stacktrace_ops *ops, void *data)
-{
-	const unsigned cpu = get_cpu();
-	unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr;
-	unsigned used = 0;
-	struct thread_info *tinfo;
-
-	if (!task)
-		task = current;
-
-	if (!stack) {
-		unsigned long dummy;
-		stack = &dummy;
-		if (task && task != current)
-			stack = (unsigned long *)task->thread.sp;
-	}
-
-#ifdef CONFIG_FRAME_POINTER
-	if (!bp) {
-		if (task == current) {
-			/* Grab bp right from our regs */
-			asm("movq %%rbp, %0" : "=r" (bp) : );
-		} else {
-			/* bp is the last reg pushed by switch_to */
-			bp = *(unsigned long *) task->thread.sp;
-		}
-	}
-#endif
-
-	/*
-	 * Print function call entries in all stacks, starting at the
-	 * current stack address. If the stacks consist of nested
-	 * exceptions
-	 */
-	tinfo = task_thread_info(task);
-	for (;;) {
-		char *id;
-		unsigned long *estack_end;
-		estack_end = in_exception_stack(cpu, (unsigned long)stack,
-						&used, &id);
-
-		if (estack_end) {
-			if (ops->stack(data, id) < 0)
-				break;
-
-			bp = print_context_stack(tinfo, stack, bp, ops,
-							data, estack_end);
-			ops->stack(data, "<EOE>");
-			/*
-			 * We link to the next stack via the
-			 * second-to-last pointer (index -2 to end) in the
-			 * exception stack:
-			 */
-			stack = (unsigned long *) estack_end[-2];
-			continue;
-		}
-		if (irqstack_end) {
-			unsigned long *irqstack;
-			irqstack = irqstack_end -
-				(IRQSTACKSIZE - 64) / sizeof(*irqstack);
-
-			if (stack >= irqstack && stack < irqstack_end) {
-				if (ops->stack(data, "IRQ") < 0)
-					break;
-				bp = print_context_stack(tinfo, stack, bp,
-						ops, data, irqstack_end);
-				/*
-				 * We link to the next stack (which would be
-				 * the process stack normally) the last
-				 * pointer (index -1 to end) in the IRQ stack:
-				 */
-				stack = (unsigned long *) (irqstack_end[-1]);
-				irqstack_end = NULL;
-				ops->stack(data, "EOI");
-				continue;
-			}
-		}
-		break;
-	}
-
-	/*
-	 * This handles the process stack:
-	 */
-	bp = print_context_stack(tinfo, stack, bp, ops, data, NULL);
-	put_cpu();
-}
-EXPORT_SYMBOL(dump_trace);
-
-static void
-print_trace_warning_symbol(void *data, char *msg, unsigned long symbol)
-{
-	print_symbol(msg, symbol);
-	printk("\n");
-}
-
-static void print_trace_warning(void *data, char *msg)
-{
-	printk("%s\n", msg);
-}
-
-static int print_trace_stack(void *data, char *name)
-{
-	printk(" <%s> ", name);
-	return 0;
-}
-
-static void print_trace_address(void *data, unsigned long addr, int reliable)
-{
-	touch_nmi_watchdog();
-	printk_address(addr, reliable);
-}
-
-static const struct stacktrace_ops print_trace_ops = {
-	.warning = print_trace_warning,
-	.warning_symbol = print_trace_warning_symbol,
-	.stack = print_trace_stack,
-	.address = print_trace_address,
-};
-
-static void
-show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
-		unsigned long *stack, unsigned long bp, char *log_lvl)
-{
-	printk("Call Trace:\n");
-	dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl);
-}
-
-void show_trace(struct task_struct *task, struct pt_regs *regs,
-		unsigned long *stack, unsigned long bp)
-{
-	show_trace_log_lvl(task, regs, stack, bp, "");
-}
-
-static void
-show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
-		unsigned long *sp, unsigned long bp, char *log_lvl)
-{
-	unsigned long *stack;
-	int i;
-	const int cpu = smp_processor_id();
-	unsigned long *irqstack_end =
-		(unsigned long *) (cpu_pda(cpu)->irqstackptr);
-	unsigned long *irqstack =
-		(unsigned long *) (cpu_pda(cpu)->irqstackptr - IRQSTACKSIZE);
-
-	/*
-	 * debugging aid: "show_stack(NULL, NULL);" prints the
-	 * back trace for this cpu.
-	 */
-
-	if (sp == NULL) {
-		if (task)
-			sp = (unsigned long *)task->thread.sp;
-		else
-			sp = (unsigned long *)&sp;
-	}
-
-	stack = sp;
-	for (i = 0; i < kstack_depth_to_print; i++) {
-		if (stack >= irqstack && stack <= irqstack_end) {
-			if (stack == irqstack_end) {
-				stack = (unsigned long *) (irqstack_end[-1]);
-				printk(" <EOI> ");
-			}
-		} else {
-		if (((long) stack & (THREAD_SIZE-1)) == 0)
-			break;
-		}
-		if (i && ((i % 4) == 0))
-			printk("\n");
-		printk(" %016lx", *stack++);
-		touch_nmi_watchdog();
-	}
-	printk("\n");
-	show_trace_log_lvl(task, regs, sp, bp, log_lvl);
-}
-
-void show_stack(struct task_struct *task, unsigned long *sp)
-{
-	show_stack_log_lvl(task, NULL, sp, 0, "");
-}
-
-/*
- * The architecture-independent dump_stack generator
- */
-void dump_stack(void)
-{
-	unsigned long bp = 0;
-	unsigned long stack;
-
-#ifdef CONFIG_FRAME_POINTER
-	if (!bp)
-		asm("movq %%rbp, %0" : "=r" (bp) : );
-#endif
-
-	printk("Pid: %d, comm: %.20s %s %s %.*s\n",
-		current->pid, current->comm, print_tainted(),
-		init_utsname()->release,
-		(int)strcspn(init_utsname()->version, " "),
-		init_utsname()->version);
-	show_trace(NULL, NULL, &stack, bp);
-}
-EXPORT_SYMBOL(dump_stack);
-
-void show_registers(struct pt_regs *regs)
-{
-	int i;
-	unsigned long sp;
-	const int cpu = smp_processor_id();
-	struct task_struct *cur = cpu_pda(cpu)->pcurrent;
-
-	sp = regs->sp;
-	printk("CPU %d ", cpu);
-	__show_regs(regs);
-	printk("Process %s (pid: %d, threadinfo %p, task %p)\n",
-		cur->comm, cur->pid, task_thread_info(cur), cur);
-
-	/*
-	 * When in-kernel, we also print out the stack and code at the
-	 * time of the fault..
-	 */
-	if (!user_mode(regs)) {
-		unsigned int code_prologue = code_bytes * 43 / 64;
-		unsigned int code_len = code_bytes;
-		unsigned char c;
-		u8 *ip;
-
-		printk("Stack: ");
-		show_stack_log_lvl(NULL, regs, (unsigned long *)sp,
-				regs->bp, "");
-
-		printk(KERN_EMERG "Code: ");
-
-		ip = (u8 *)regs->ip - code_prologue;
-		if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) {
-			/* try starting at RIP */
-			ip = (u8 *)regs->ip;
-			code_len = code_len - code_prologue + 1;
-		}
-		for (i = 0; i < code_len; i++, ip++) {
-			if (ip < (u8 *)PAGE_OFFSET ||
-					probe_kernel_address(ip, c)) {
-				printk(" Bad RIP value.");
-				break;
-			}
-			if (ip == (u8 *)regs->ip)
-				printk("<%02x> ", c);
-			else
-				printk("%02x ", c);
-		}
-	}
-	printk("\n");
-}
-
-int is_valid_bugaddr(unsigned long ip)
-{
-	unsigned short ud2;
-
-	if (__copy_from_user(&ud2, (const void __user *) ip, sizeof(ud2)))
-		return 0;
-
-	return ud2 == 0x0b0f;
-}
-
-static raw_spinlock_t die_lock = __RAW_SPIN_LOCK_UNLOCKED;
-static int die_owner = -1;
-static unsigned int die_nest_count;
-
-unsigned __kprobes long oops_begin(void)
-{
-	int cpu;
-	unsigned long flags;
-
-	oops_enter();
-
-	/* racy, but better than risking deadlock. */
-	raw_local_irq_save(flags);
-	cpu = smp_processor_id();
-	if (!__raw_spin_trylock(&die_lock)) {
-		if (cpu == die_owner)
-			/* nested oops. should stop eventually */;
-		else
-			__raw_spin_lock(&die_lock);
-	}
-	die_nest_count++;
-	die_owner = cpu;
-	console_verbose();
-	bust_spinlocks(1);
-	return flags;
-}
-
-void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr)
-{
-	die_owner = -1;
-	bust_spinlocks(0);
-	die_nest_count--;
-	if (!die_nest_count)
-		/* Nest count reaches zero, release the lock. */
-		__raw_spin_unlock(&die_lock);
-	raw_local_irq_restore(flags);
-	if (!regs) {
-		oops_exit();
-		return;
-	}
-	if (panic_on_oops)
-		panic("Fatal exception");
-	oops_exit();
-	do_exit(signr);
-}
-
-int __kprobes __die(const char *str, struct pt_regs *regs, long err)
-{
-	printk(KERN_EMERG "%s: %04lx [%u] ", str, err & 0xffff, ++die_counter);
-#ifdef CONFIG_PREEMPT
-	printk("PREEMPT ");
-#endif
-#ifdef CONFIG_SMP
-	printk("SMP ");
-#endif
-#ifdef CONFIG_DEBUG_PAGEALLOC
-	printk("DEBUG_PAGEALLOC");
-#endif
-	printk("\n");
-	if (notify_die(DIE_OOPS, str, regs, err,
-			current->thread.trap_no, SIGSEGV) == NOTIFY_STOP)
-		return 1;
-
-	show_registers(regs);
-	add_taint(TAINT_DIE);
-	/* Executive summary in case the oops scrolled away */
-	printk(KERN_ALERT "RIP ");
-	printk_address(regs->ip, 1);
-	printk(" RSP <%016lx>\n", regs->sp);
-	if (kexec_should_crash(current))
-		crash_kexec(regs);
-	return 0;
-}
-
-void die(const char *str, struct pt_regs *regs, long err)
-{
-	unsigned long flags = oops_begin();
-
-	if (!user_mode(regs))
-		report_bug(regs->ip, regs);
-
-	if (__die(str, regs, err))
-		regs = NULL;
-	oops_end(flags, regs, SIGSEGV);
-}
-
-notrace __kprobes void
-die_nmi(char *str, struct pt_regs *regs, int do_panic)
-{
-	unsigned long flags;
-
-	if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, 2, SIGINT) == NOTIFY_STOP)
-		return;
-
-	flags = oops_begin();
-	/*
-	 * We are in trouble anyway, lets at least try
-	 * to get a message out.
-	 */
-	printk(KERN_EMERG "%s", str);
-	printk(" on CPU%d, ip %08lx, registers:\n",
-		smp_processor_id(), regs->ip);
-	show_registers(regs);
-	if (kexec_should_crash(current))
-		crash_kexec(regs);
-	if (do_panic || panic_on_oops)
-		panic("Non maskable interrupt");
-	oops_end(flags, NULL, SIGBUS);
-	nmi_exit();
-	local_irq_enable();
-	do_exit(SIGBUS);
-}
-
-static void __kprobes
-do_trap(int trapnr, int signr, char *str, struct pt_regs *regs,
-	long error_code, siginfo_t *info)
-{
-	struct task_struct *tsk = current;
-
-	if (!user_mode(regs))
-		goto kernel_trap;
-
-	/*
-	 * We want error_code and trap_no set for userspace faults and
-	 * kernelspace faults which result in die(), but not
-	 * kernelspace faults which are fixed up.  die() gives the
-	 * process no chance to handle the signal and notice the
-	 * kernel fault information, so that won't result in polluting
-	 * the information about previously queued, but not yet
-	 * delivered, faults.  See also do_general_protection below.
-	 */
-	tsk->thread.error_code = error_code;
-	tsk->thread.trap_no = trapnr;
-
-	if (show_unhandled_signals && unhandled_signal(tsk, signr) &&
-	    printk_ratelimit()) {
-		printk(KERN_INFO
-		       "%s[%d] trap %s ip:%lx sp:%lx error:%lx",
-		       tsk->comm, tsk->pid, str,
-		       regs->ip, regs->sp, error_code);
-		print_vma_addr(" in ", regs->ip);
-		printk("\n");
-	}
-
-	if (info)
-		force_sig_info(signr, info, tsk);
-	else
-		force_sig(signr, tsk);
-	return;
-
-kernel_trap:
-	if (!fixup_exception(regs)) {
-		tsk->thread.error_code = error_code;
-		tsk->thread.trap_no = trapnr;
-		die(str, regs, error_code);
-	}
-	return;
-}
-
-#define DO_ERROR(trapnr, signr, str, name) \
-asmlinkage void do_##name(struct pt_regs *regs, long error_code)	\
-{									\
-	if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr)	\
-							== NOTIFY_STOP)	\
-		return;							\
-	conditional_sti(regs);						\
-	do_trap(trapnr, signr, str, regs, error_code, NULL);		\
-}
-
-#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr)		\
-asmlinkage void do_##name(struct pt_regs *regs, long error_code)	\
-{									\
-	siginfo_t info;							\
-	info.si_signo = signr;						\
-	info.si_errno = 0;						\
-	info.si_code = sicode;						\
-	info.si_addr = (void __user *)siaddr;				\
-	trace_hardirqs_fixup();						\
-	if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr)	\
-							== NOTIFY_STOP)	\
-		return;							\
-	conditional_sti(regs);						\
-	do_trap(trapnr, signr, str, regs, error_code, &info);		\
-}
-
-DO_ERROR_INFO(0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->ip)
-DO_ERROR(4, SIGSEGV, "overflow", overflow)
-DO_ERROR(5, SIGSEGV, "bounds", bounds)
-DO_ERROR_INFO(6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip)
-DO_ERROR(9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun)
-DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS)
-DO_ERROR(11, SIGBUS, "segment not present", segment_not_present)
-DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0)
-
-/* Runs on IST stack */
-asmlinkage void do_stack_segment(struct pt_regs *regs, long error_code)
-{
-	if (notify_die(DIE_TRAP, "stack segment", regs, error_code,
-			12, SIGBUS) == NOTIFY_STOP)
-		return;
-	preempt_conditional_sti(regs);
-	do_trap(12, SIGBUS, "stack segment", regs, error_code, NULL);
-	preempt_conditional_cli(regs);
-}
-
-asmlinkage void do_double_fault(struct pt_regs *regs, long error_code)
-{
-	static const char str[] = "double fault";
-	struct task_struct *tsk = current;
-
-	/* Return not checked because double check cannot be ignored */
-	notify_die(DIE_TRAP, str, regs, error_code, 8, SIGSEGV);
-
-	tsk->thread.error_code = error_code;
-	tsk->thread.trap_no = 8;
-
-	/* This is always a kernel trap and never fixable (and thus must
-	   never return). */
-	for (;;)
-		die(str, regs, error_code);
-}
-
-asmlinkage void __kprobes
-do_general_protection(struct pt_regs *regs, long error_code)
-{
-	struct task_struct *tsk;
-
-	conditional_sti(regs);
-
-	tsk = current;
-	if (!user_mode(regs))
-		goto gp_in_kernel;
-
-	tsk->thread.error_code = error_code;
-	tsk->thread.trap_no = 13;
-
-	if (show_unhandled_signals && unhandled_signal(tsk, SIGSEGV) &&
-			printk_ratelimit()) {
-		printk(KERN_INFO
-			"%s[%d] general protection ip:%lx sp:%lx error:%lx",
-			tsk->comm, tsk->pid,
-			regs->ip, regs->sp, error_code);
-		print_vma_addr(" in ", regs->ip);
-		printk("\n");
-	}
-
-	force_sig(SIGSEGV, tsk);
-	return;
-
-gp_in_kernel:
-	if (fixup_exception(regs))
-		return;
-
-	tsk->thread.error_code = error_code;
-	tsk->thread.trap_no = 13;
-	if (notify_die(DIE_GPF, "general protection fault", regs,
-				error_code, 13, SIGSEGV) == NOTIFY_STOP)
-		return;
-	die("general protection fault", regs, error_code);
-}
-
-static notrace __kprobes void
-mem_parity_error(unsigned char reason, struct pt_regs *regs)
-{
-	printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n",
-		reason);
-	printk(KERN_EMERG "You have some hardware problem, likely on the PCI bus.\n");
-
-#if defined(CONFIG_EDAC)
-	if (edac_handler_set()) {
-		edac_atomic_assert_error();
-		return;
-	}
-#endif
-
-	if (panic_on_unrecovered_nmi)
-		panic("NMI: Not continuing");
-
-	printk(KERN_EMERG "Dazed and confused, but trying to continue\n");
-
-	/* Clear and disable the memory parity error line. */
-	reason = (reason & 0xf) | 4;
-	outb(reason, 0x61);
-}
-
-static notrace __kprobes void
-io_check_error(unsigned char reason, struct pt_regs *regs)
-{
-	printk("NMI: IOCK error (debug interrupt?)\n");
-	show_registers(regs);
-
-	/* Re-enable the IOCK line, wait for a few seconds */
-	reason = (reason & 0xf) | 8;
-	outb(reason, 0x61);
-	mdelay(2000);
-	reason &= ~8;
-	outb(reason, 0x61);
-}
-
-static notrace __kprobes void
-unknown_nmi_error(unsigned char reason, struct pt_regs *regs)
-{
-	if (notify_die(DIE_NMIUNKNOWN, "nmi", regs, reason, 2, SIGINT) ==
-			NOTIFY_STOP)
-		return;
-	printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n",
-		reason);
-	printk(KERN_EMERG "Do you have a strange power saving mode enabled?\n");
-
-	if (panic_on_unrecovered_nmi)
-		panic("NMI: Not continuing");
-
-	printk(KERN_EMERG "Dazed and confused, but trying to continue\n");
-}
-
-/* Runs on IST stack. This code must keep interrupts off all the time.
-   Nested NMIs are prevented by the CPU. */
-asmlinkage notrace __kprobes void default_do_nmi(struct pt_regs *regs)
-{
-	unsigned char reason = 0;
-	int cpu;
-
-	cpu = smp_processor_id();
-
-	/* Only the BSP gets external NMIs from the system. */
-	if (!cpu)
-		reason = get_nmi_reason();
-
-	if (!(reason & 0xc0)) {
-		if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT)
-								== NOTIFY_STOP)
-			return;
-		/*
-		 * Ok, so this is none of the documented NMI sources,
-		 * so it must be the NMI watchdog.
-		 */
-		if (nmi_watchdog_tick(regs, reason))
-			return;
-		if (!do_nmi_callback(regs, cpu))
-			unknown_nmi_error(reason, regs);
-
-		return;
-	}
-	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
-		return;
-
-	/* AK: following checks seem to be broken on modern chipsets. FIXME */
-	if (reason & 0x80)
-		mem_parity_error(reason, regs);
-	if (reason & 0x40)
-		io_check_error(reason, regs);
-}
-
-asmlinkage notrace __kprobes void
-do_nmi(struct pt_regs *regs, long error_code)
-{
-	nmi_enter();
-
-	add_pda(__nmi_count, 1);
-
-	if (!ignore_nmis)
-		default_do_nmi(regs);
-
-	nmi_exit();
-}
-
-void stop_nmi(void)
-{
-	acpi_nmi_disable();
-	ignore_nmis++;
-}
-
-void restart_nmi(void)
-{
-	ignore_nmis--;
-	acpi_nmi_enable();
-}
-
-/* runs on IST stack. */
-asmlinkage void __kprobes do_int3(struct pt_regs *regs, long error_code)
-{
-	trace_hardirqs_fixup();
-
-	if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP)
-			== NOTIFY_STOP)
-		return;
-
-	preempt_conditional_sti(regs);
-	do_trap(3, SIGTRAP, "int3", regs, error_code, NULL);
-	preempt_conditional_cli(regs);
-}
-
-/* Help handler running on IST stack to switch back to user stack
-   for scheduling or signal handling. The actual stack switch is done in
-   entry.S */
-asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs)
-{
-	struct pt_regs *regs = eregs;
-	/* Did already sync */
-	if (eregs == (struct pt_regs *)eregs->sp)
-		;
-	/* Exception from user space */
-	else if (user_mode(eregs))
-		regs = task_pt_regs(current);
-	/* Exception from kernel and interrupts are enabled. Move to
-	   kernel process stack. */
-	else if (eregs->flags & X86_EFLAGS_IF)
-		regs = (struct pt_regs *)(eregs->sp -= sizeof(struct pt_regs));
-	if (eregs != regs)
-		*regs = *eregs;
-	return regs;
-}
-
-/* runs on IST stack. */
-asmlinkage void __kprobes do_debug(struct pt_regs *regs,
-				   unsigned long error_code)
-{
-	struct task_struct *tsk = current;
-	unsigned long condition;
-	siginfo_t info;
-
-	trace_hardirqs_fixup();
-
-	get_debugreg(condition, 6);
-
-	/*
-	 * The processor cleared BTF, so don't mark that we need it set.
-	 */
-	clear_tsk_thread_flag(tsk, TIF_DEBUGCTLMSR);
-	tsk->thread.debugctlmsr = 0;
-
-	if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code,
-						SIGTRAP) == NOTIFY_STOP)
-		return;
-
-	preempt_conditional_sti(regs);
-
-	/* Mask out spurious debug traps due to lazy DR7 setting */
-	if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) {
-		if (!tsk->thread.debugreg7)
-			goto clear_dr7;
-	}
-
-	tsk->thread.debugreg6 = condition;
-
-	/*
-	 * Single-stepping through TF: make sure we ignore any events in
-	 * kernel space (but re-enable TF when returning to user mode).
-	 */
-	if (condition & DR_STEP) {
-		if (!user_mode(regs))
-			goto clear_TF_reenable;
-	}
-
-	/* Ok, finally something we can handle */
-	tsk->thread.trap_no = 1;
-	tsk->thread.error_code = error_code;
-	info.si_signo = SIGTRAP;
-	info.si_errno = 0;
-	info.si_code = TRAP_BRKPT;
-	info.si_addr = user_mode(regs) ? (void __user *)regs->ip : NULL;
-	force_sig_info(SIGTRAP, &info, tsk);
-
-clear_dr7:
-	set_debugreg(0, 7);
-	preempt_conditional_cli(regs);
-	return;
-
-clear_TF_reenable:
-	set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
-	regs->flags &= ~X86_EFLAGS_TF;
-	preempt_conditional_cli(regs);
-	return;
-}
-
-static int kernel_math_error(struct pt_regs *regs, const char *str, int trapnr)
-{
-	if (fixup_exception(regs))
-		return 1;
-
-	notify_die(DIE_GPF, str, regs, 0, trapnr, SIGFPE);
-	/* Illegal floating point operation in the kernel */
-	current->thread.trap_no = trapnr;
-	die(str, regs, 0);
-	return 0;
-}
-
-/*
- * Note that we play around with the 'TS' bit in an attempt to get
- * the correct behaviour even in the presence of the asynchronous
- * IRQ13 behaviour
- */
-asmlinkage void do_coprocessor_error(struct pt_regs *regs)
-{
-	void __user *ip = (void __user *)(regs->ip);
-	struct task_struct *task;
-	siginfo_t info;
-	unsigned short cwd, swd;
-
-	conditional_sti(regs);
-	if (!user_mode(regs) &&
-	    kernel_math_error(regs, "kernel x87 math error", 16))
-		return;
-
-	/*
-	 * Save the info for the exception handler and clear the error.
-	 */
-	task = current;
-	save_init_fpu(task);
-	task->thread.trap_no = 16;
-	task->thread.error_code = 0;
-	info.si_signo = SIGFPE;
-	info.si_errno = 0;
-	info.si_code = __SI_FAULT;
-	info.si_addr = ip;
-	/*
-	 * (~cwd & swd) will mask out exceptions that are not set to unmasked
-	 * status.  0x3f is the exception bits in these regs, 0x200 is the
-	 * C1 reg you need in case of a stack fault, 0x040 is the stack
-	 * fault bit.  We should only be taking one exception at a time,
-	 * so if this combination doesn't produce any single exception,
-	 * then we have a bad program that isn't synchronizing its FPU usage
-	 * and it will suffer the consequences since we won't be able to
-	 * fully reproduce the context of the exception
-	 */
-	cwd = get_fpu_cwd(task);
-	swd = get_fpu_swd(task);
-	switch (swd & ~cwd & 0x3f) {
-	case 0x000: /* No unmasked exception */
-	default: /* Multiple exceptions */
-		break;
-	case 0x001: /* Invalid Op */
-		/*
-		 * swd & 0x240 == 0x040: Stack Underflow
-		 * swd & 0x240 == 0x240: Stack Overflow
-		 * User must clear the SF bit (0x40) if set
-		 */
-		info.si_code = FPE_FLTINV;
-		break;
-	case 0x002: /* Denormalize */
-	case 0x010: /* Underflow */
-		info.si_code = FPE_FLTUND;
-		break;
-	case 0x004: /* Zero Divide */
-		info.si_code = FPE_FLTDIV;
-		break;
-	case 0x008: /* Overflow */
-		info.si_code = FPE_FLTOVF;
-		break;
-	case 0x020: /* Precision */
-		info.si_code = FPE_FLTRES;
-		break;
-	}
-	force_sig_info(SIGFPE, &info, task);
-}
-
-asmlinkage void bad_intr(void)
-{
-	printk("bad interrupt");
-}
-
-asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs)
-{
-	void __user *ip = (void __user *)(regs->ip);
-	struct task_struct *task;
-	siginfo_t info;
-	unsigned short mxcsr;
-
-	conditional_sti(regs);
-	if (!user_mode(regs) &&
-			kernel_math_error(regs, "kernel simd math error", 19))
-		return;
-
-	/*
-	 * Save the info for the exception handler and clear the error.
-	 */
-	task = current;
-	save_init_fpu(task);
-	task->thread.trap_no = 19;
-	task->thread.error_code = 0;
-	info.si_signo = SIGFPE;
-	info.si_errno = 0;
-	info.si_code = __SI_FAULT;
-	info.si_addr = ip;
-	/*
-	 * The SIMD FPU exceptions are handled a little differently, as there
-	 * is only a single status/control register.  Thus, to determine which
-	 * unmasked exception was caught we must mask the exception mask bits
-	 * at 0x1f80, and then use these to mask the exception bits at 0x3f.
-	 */
-	mxcsr = get_fpu_mxcsr(task);
-	switch (~((mxcsr & 0x1f80) >> 7) & (mxcsr & 0x3f)) {
-	case 0x000:
-	default:
-		break;
-	case 0x001: /* Invalid Op */
-		info.si_code = FPE_FLTINV;
-		break;
-	case 0x002: /* Denormalize */
-	case 0x010: /* Underflow */
-		info.si_code = FPE_FLTUND;
-		break;
-	case 0x004: /* Zero Divide */
-		info.si_code = FPE_FLTDIV;
-		break;
-	case 0x008: /* Overflow */
-		info.si_code = FPE_FLTOVF;
-		break;
-	case 0x020: /* Precision */
-		info.si_code = FPE_FLTRES;
-		break;
-	}
-	force_sig_info(SIGFPE, &info, task);
-}
-
-asmlinkage void do_spurious_interrupt_bug(struct pt_regs *regs)
-{
-}
-
-asmlinkage void __attribute__((weak)) smp_thermal_interrupt(void)
-{
-}
-
-asmlinkage void __attribute__((weak)) mce_threshold_interrupt(void)
-{
-}
-
-/*
- * 'math_state_restore()' saves the current math information in the
- * old math state array, and gets the new ones from the current task
- *
- * Careful.. There are problems with IBM-designed IRQ13 behaviour.
- * Don't touch unless you *really* know how it works.
- */
-asmlinkage void math_state_restore(void)
-{
-	struct task_struct *me = current;
-
-	if (!used_math()) {
-		local_irq_enable();
-		/*
-		 * does a slab alloc which can sleep
-		 */
-		if (init_fpu(me)) {
-			/*
-			 * ran out of memory!
-			 */
-			do_group_exit(SIGKILL);
-			return;
-		}
-		local_irq_disable();
-	}
-
-	clts();				/* Allow maths ops (or we recurse) */
-	/*
-	 * Paranoid restore. send a SIGSEGV if we fail to restore the state.
-	 */
-	if (unlikely(restore_fpu_checking(&me->thread.xstate->fxsave))) {
-		stts();
-		force_sig(SIGSEGV, me);
-		return;
-	}
-	task_thread_info(me)->status |= TS_USEDFPU;
-	me->fpu_counter++;
-}
-EXPORT_SYMBOL_GPL(math_state_restore);
-
-void __init trap_init(void)
-{
-	set_intr_gate(0, &divide_error);
-	set_intr_gate_ist(1, &debug, DEBUG_STACK);
-	set_intr_gate_ist(2, &nmi, NMI_STACK);
-	/* int3 can be called from all */
-	set_system_gate_ist(3, &int3, DEBUG_STACK);
-	/* int4 can be called from all */
-	set_system_gate(4, &overflow);
-	set_intr_gate(5, &bounds);
-	set_intr_gate(6, &invalid_op);
-	set_intr_gate(7, &device_not_available);
-	set_intr_gate_ist(8, &double_fault, DOUBLEFAULT_STACK);
-	set_intr_gate(9, &coprocessor_segment_overrun);
-	set_intr_gate(10, &invalid_TSS);
-	set_intr_gate(11, &segment_not_present);
-	set_intr_gate_ist(12, &stack_segment, STACKFAULT_STACK);
-	set_intr_gate(13, &general_protection);
-	set_intr_gate(14, &page_fault);
-	set_intr_gate(15, &spurious_interrupt_bug);
-	set_intr_gate(16, &coprocessor_error);
-	set_intr_gate(17, &alignment_check);
-#ifdef CONFIG_X86_MCE
-	set_intr_gate_ist(18, &machine_check, MCE_STACK);
-#endif
-	set_intr_gate(19, &simd_coprocessor_error);
-
-#ifdef CONFIG_IA32_EMULATION
-	set_system_gate(IA32_SYSCALL_VECTOR, ia32_syscall);
-#endif
-	/*
-	 * initialize the per thread extended state:
-	 */
-	init_thread_xstate();
-	/*
-	 * Should be a barrier for any external CPU state:
-	 */
-	cpu_init();
-}
-
-static int __init oops_setup(char *s)
-{
-	if (!s)
-		return -EINVAL;
-	if (!strcmp(s, "panic"))
-		panic_on_oops = 1;
-	return 0;
-}
-early_param("oops", oops_setup);
-
-static int __init kstack_setup(char *s)
-{
-	if (!s)
-		return -EINVAL;
-	kstack_depth_to_print = simple_strtoul(s, NULL, 0);
-	return 0;
-}
-early_param("kstack", kstack_setup);
-
-static int __init code_bytes_setup(char *s)
-{
-	code_bytes = simple_strtoul(s, NULL, 0);
-	if (code_bytes > 8192)
-		code_bytes = 8192;
-
-	return 1;
-}
-__setup("code_bytes=", code_bytes_setup);
diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c
index 8c9ad02af5a2..8b6c393ab9fd 100644
--- a/arch/x86/kernel/vmi_32.c
+++ b/arch/x86/kernel/vmi_32.c
@@ -905,8 +905,8 @@ static inline int __init activate_vmi(void)
 #endif
 
 #ifdef CONFIG_X86_LOCAL_APIC
-	para_fill(pv_apic_ops.apic_read, APICRead);
-	para_fill(pv_apic_ops.apic_write, APICWrite);
+       para_fill(apic_ops->read, APICRead);
+       para_fill(apic_ops->write, APICWrite);
 #endif
 
 	/*
diff --git a/arch/x86/kernel/vmlinux_32.lds.S b/arch/x86/kernel/vmlinux_32.lds.S
index af5bdad84604..a9b8560adbc2 100644
--- a/arch/x86/kernel/vmlinux_32.lds.S
+++ b/arch/x86/kernel/vmlinux_32.lds.S
@@ -140,10 +140,10 @@ SECTIONS
 	*(.con_initcall.init)
   	__con_initcall_end = .;
   }
-  .x86cpuvendor.init : AT(ADDR(.x86cpuvendor.init) - LOAD_OFFSET) {
-	__x86cpuvendor_start = .;
-	*(.x86cpuvendor.init)
-	__x86cpuvendor_end = .;
+  .x86_cpu_dev.init : AT(ADDR(.x86_cpu_dev.init) - LOAD_OFFSET) {
+	__x86_cpu_dev_start = .;
+	*(.x86_cpu_dev.init)
+	__x86_cpu_dev_end = .;
   }
   SECURITY_INIT
   . = ALIGN(4);
@@ -180,6 +180,7 @@ SECTIONS
   . = ALIGN(PAGE_SIZE);
   .data.percpu  : AT(ADDR(.data.percpu) - LOAD_OFFSET) {
 	__per_cpu_start = .;
+	*(.data.percpu.page_aligned)
 	*(.data.percpu)
 	*(.data.percpu.shared_aligned)
 	__per_cpu_end = .;
diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S
index 63e5c1a22e88..46e05447405b 100644
--- a/arch/x86/kernel/vmlinux_64.lds.S
+++ b/arch/x86/kernel/vmlinux_64.lds.S
@@ -168,12 +168,11 @@ SECTIONS
 	*(.con_initcall.init)
   }
   __con_initcall_end = .;
-  . = ALIGN(16);
-  __x86cpuvendor_start = .;
-  .x86cpuvendor.init : AT(ADDR(.x86cpuvendor.init) - LOAD_OFFSET) {
-	*(.x86cpuvendor.init)
+  __x86_cpu_dev_start = .;
+  .x86_cpu_dev.init : AT(ADDR(.x86_cpu_dev.init) - LOAD_OFFSET) {
+	*(.x86_cpu_dev.init)
   }
-  __x86cpuvendor_end = .;
+  __x86_cpu_dev_end = .;
   SECURITY_INIT
 
   . = ALIGN(8);
diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c
new file mode 100644
index 000000000000..9abac8a9d823
--- /dev/null
+++ b/arch/x86/kernel/xsave.c
@@ -0,0 +1,345 @@
+/*
+ * xsave/xrstor support.
+ *
+ * Author: Suresh Siddha <suresh.b.siddha@intel.com>
+ */
+#include <linux/bootmem.h>
+#include <linux/compat.h>
+#include <asm/i387.h>
+#ifdef CONFIG_IA32_EMULATION
+#include <asm/sigcontext32.h>
+#endif
+#include <asm/xcr.h>
+
+/*
+ * Supported feature mask by the CPU and the kernel.
+ */
+u64 pcntxt_mask;
+
+struct _fpx_sw_bytes fx_sw_reserved;
+#ifdef CONFIG_IA32_EMULATION
+struct _fpx_sw_bytes fx_sw_reserved_ia32;
+#endif
+
+/*
+ * Check for the presence of extended state information in the
+ * user fpstate pointer in the sigcontext.
+ */
+int check_for_xstate(struct i387_fxsave_struct __user *buf,
+		     void __user *fpstate,
+		     struct _fpx_sw_bytes *fx_sw_user)
+{
+	int min_xstate_size = sizeof(struct i387_fxsave_struct) +
+			      sizeof(struct xsave_hdr_struct);
+	unsigned int magic2;
+	int err;
+
+	err = __copy_from_user(fx_sw_user, &buf->sw_reserved[0],
+			       sizeof(struct _fpx_sw_bytes));
+
+	if (err)
+		return err;
+
+	/*
+	 * First Magic check failed.
+	 */
+	if (fx_sw_user->magic1 != FP_XSTATE_MAGIC1)
+		return -1;
+
+	/*
+	 * Check for error scenarios.
+	 */
+	if (fx_sw_user->xstate_size < min_xstate_size ||
+	    fx_sw_user->xstate_size > xstate_size ||
+	    fx_sw_user->xstate_size > fx_sw_user->extended_size)
+		return -1;
+
+	err = __get_user(magic2, (__u32 *) (((void *)fpstate) +
+					    fx_sw_user->extended_size -
+					    FP_XSTATE_MAGIC2_SIZE));
+	/*
+	 * Check for the presence of second magic word at the end of memory
+	 * layout. This detects the case where the user just copied the legacy
+	 * fpstate layout with out copying the extended state information
+	 * in the memory layout.
+	 */
+	if (err || magic2 != FP_XSTATE_MAGIC2)
+		return -1;
+
+	return 0;
+}
+
+#ifdef CONFIG_X86_64
+/*
+ * Signal frame handlers.
+ */
+
+int save_i387_xstate(void __user *buf)
+{
+	struct task_struct *tsk = current;
+	int err = 0;
+
+	if (!access_ok(VERIFY_WRITE, buf, sig_xstate_size))
+		return -EACCES;
+
+	BUG_ON(sig_xstate_size < xstate_size);
+
+	if ((unsigned long)buf % 64)
+		printk("save_i387_xstate: bad fpstate %p\n", buf);
+
+	if (!used_math())
+		return 0;
+	clear_used_math(); /* trigger finit */
+	if (task_thread_info(tsk)->status & TS_USEDFPU) {
+		/*
+	 	 * Start with clearing the user buffer. This will present a
+	 	 * clean context for the bytes not touched by the fxsave/xsave.
+		 */
+		err = __clear_user(buf, sig_xstate_size);
+		if (err)
+			return err;
+
+		if (task_thread_info(tsk)->status & TS_XSAVE)
+			err = xsave_user(buf);
+		else
+			err = fxsave_user(buf);
+
+		if (err)
+			return err;
+		task_thread_info(tsk)->status &= ~TS_USEDFPU;
+		stts();
+	} else {
+		if (__copy_to_user(buf, &tsk->thread.xstate->fxsave,
+				   xstate_size))
+			return -1;
+	}
+
+	if (task_thread_info(tsk)->status & TS_XSAVE) {
+		struct _fpstate __user *fx = buf;
+		struct _xstate __user *x = buf;
+		u64 xstate_bv;
+
+		err = __copy_to_user(&fx->sw_reserved, &fx_sw_reserved,
+				     sizeof(struct _fpx_sw_bytes));
+
+		err |= __put_user(FP_XSTATE_MAGIC2,
+				  (__u32 __user *) (buf + sig_xstate_size
+						    - FP_XSTATE_MAGIC2_SIZE));
+
+		/*
+		 * Read the xstate_bv which we copied (directly from the cpu or
+		 * from the state in task struct) to the user buffers and
+		 * set the FP/SSE bits.
+		 */
+		err |= __get_user(xstate_bv, &x->xstate_hdr.xstate_bv);
+
+		/*
+		 * For legacy compatible, we always set FP/SSE bits in the bit
+		 * vector while saving the state to the user context. This will
+		 * enable us capturing any changes(during sigreturn) to
+		 * the FP/SSE bits by the legacy applications which don't touch
+		 * xstate_bv in the xsave header.
+		 *
+		 * xsave aware apps can change the xstate_bv in the xsave
+		 * header as well as change any contents in the memory layout.
+		 * xrestore as part of sigreturn will capture all the changes.
+		 */
+		xstate_bv |= XSTATE_FPSSE;
+
+		err |= __put_user(xstate_bv, &x->xstate_hdr.xstate_bv);
+
+		if (err)
+			return err;
+	}
+
+	return 1;
+}
+
+/*
+ * Restore the extended state if present. Otherwise, restore the FP/SSE
+ * state.
+ */
+int restore_user_xstate(void __user *buf)
+{
+	struct _fpx_sw_bytes fx_sw_user;
+	u64 mask;
+	int err;
+
+	if (((unsigned long)buf % 64) ||
+	     check_for_xstate(buf, buf, &fx_sw_user))
+		goto fx_only;
+
+	mask = fx_sw_user.xstate_bv;
+
+	/*
+	 * restore the state passed by the user.
+	 */
+	err = xrestore_user(buf, mask);
+	if (err)
+		return err;
+
+	/*
+	 * init the state skipped by the user.
+	 */
+	mask = pcntxt_mask & ~mask;
+
+	xrstor_state(init_xstate_buf, mask);
+
+	return 0;
+
+fx_only:
+	/*
+	 * couldn't find the extended state information in the
+	 * memory layout. Restore just the FP/SSE and init all
+	 * the other extended state.
+	 */
+	xrstor_state(init_xstate_buf, pcntxt_mask & ~XSTATE_FPSSE);
+	return fxrstor_checking((__force struct i387_fxsave_struct *)buf);
+}
+
+/*
+ * This restores directly out of user space. Exceptions are handled.
+ */
+int restore_i387_xstate(void __user *buf)
+{
+	struct task_struct *tsk = current;
+	int err = 0;
+
+	if (!buf) {
+		if (used_math())
+			goto clear;
+		return 0;
+	} else
+		if (!access_ok(VERIFY_READ, buf, sig_xstate_size))
+			return -EACCES;
+
+	if (!used_math()) {
+		err = init_fpu(tsk);
+		if (err)
+			return err;
+	}
+
+	if (!(task_thread_info(current)->status & TS_USEDFPU)) {
+		clts();
+		task_thread_info(current)->status |= TS_USEDFPU;
+	}
+	if (task_thread_info(tsk)->status & TS_XSAVE)
+		err = restore_user_xstate(buf);
+	else
+		err = fxrstor_checking((__force struct i387_fxsave_struct *)
+				       buf);
+	if (unlikely(err)) {
+		/*
+		 * Encountered an error while doing the restore from the
+		 * user buffer, clear the fpu state.
+		 */
+clear:
+		clear_fpu(tsk);
+		clear_used_math();
+	}
+	return err;
+}
+#endif
+
+/*
+ * Prepare the SW reserved portion of the fxsave memory layout, indicating
+ * the presence of the extended state information in the memory layout
+ * pointed by the fpstate pointer in the sigcontext.
+ * This will be saved when ever the FP and extended state context is
+ * saved on the user stack during the signal handler delivery to the user.
+ */
+void prepare_fx_sw_frame(void)
+{
+	int size_extended = (xstate_size - sizeof(struct i387_fxsave_struct)) +
+			     FP_XSTATE_MAGIC2_SIZE;
+
+	sig_xstate_size = sizeof(struct _fpstate) + size_extended;
+
+#ifdef CONFIG_IA32_EMULATION
+	sig_xstate_ia32_size = sizeof(struct _fpstate_ia32) + size_extended;
+#endif
+
+	memset(&fx_sw_reserved, 0, sizeof(fx_sw_reserved));
+
+	fx_sw_reserved.magic1 = FP_XSTATE_MAGIC1;
+	fx_sw_reserved.extended_size = sig_xstate_size;
+	fx_sw_reserved.xstate_bv = pcntxt_mask;
+	fx_sw_reserved.xstate_size = xstate_size;
+#ifdef CONFIG_IA32_EMULATION
+	memcpy(&fx_sw_reserved_ia32, &fx_sw_reserved,
+	       sizeof(struct _fpx_sw_bytes));
+	fx_sw_reserved_ia32.extended_size = sig_xstate_ia32_size;
+#endif
+}
+
+/*
+ * Represents init state for the supported extended state.
+ */
+struct xsave_struct *init_xstate_buf;
+
+#ifdef CONFIG_X86_64
+unsigned int sig_xstate_size = sizeof(struct _fpstate);
+#endif
+
+/*
+ * Enable the extended processor state save/restore feature
+ */
+void __cpuinit xsave_init(void)
+{
+	if (!cpu_has_xsave)
+		return;
+
+	set_in_cr4(X86_CR4_OSXSAVE);
+
+	/*
+	 * Enable all the features that the HW is capable of
+	 * and the Linux kernel is aware of.
+	 */
+	xsetbv(XCR_XFEATURE_ENABLED_MASK, pcntxt_mask);
+}
+
+/*
+ * setup the xstate image representing the init state
+ */
+static void __init setup_xstate_init(void)
+{
+	init_xstate_buf = alloc_bootmem(xstate_size);
+	init_xstate_buf->i387.mxcsr = MXCSR_DEFAULT;
+}
+
+/*
+ * Enable and initialize the xsave feature.
+ */
+void __init xsave_cntxt_init(void)
+{
+	unsigned int eax, ebx, ecx, edx;
+
+	cpuid_count(0xd, 0, &eax, &ebx, &ecx, &edx);
+	pcntxt_mask = eax + ((u64)edx << 32);
+
+	if ((pcntxt_mask & XSTATE_FPSSE) != XSTATE_FPSSE) {
+		printk(KERN_ERR "FP/SSE not shown under xsave features 0x%llx\n",
+		       pcntxt_mask);
+		BUG();
+	}
+
+	/*
+	 * for now OS knows only about FP/SSE
+	 */
+	pcntxt_mask = pcntxt_mask & XCNTXT_MASK;
+	xsave_init();
+
+	/*
+	 * Recompute the context size for enabled features
+	 */
+	cpuid_count(0xd, 0, &eax, &ebx, &ecx, &edx);
+	xstate_size = ebx;
+
+	prepare_fx_sw_frame();
+
+	setup_xstate_init();
+
+	printk(KERN_INFO "xsave/xrstor: enabled xstate_bv 0x%llx, "
+	       "cntxt size 0x%x\n",
+	       pcntxt_mask, xstate_size);
+}
diff --git a/arch/x86/kvm/vmx.h b/arch/x86/kvm/vmx.h
index 23e8373507ad..17e25995b65b 100644
--- a/arch/x86/kvm/vmx.h
+++ b/arch/x86/kvm/vmx.h
@@ -331,21 +331,6 @@ enum vmcs_field {
 
 #define AR_RESERVD_MASK 0xfffe0f00
 
-#define MSR_IA32_VMX_BASIC                      0x480
-#define MSR_IA32_VMX_PINBASED_CTLS              0x481
-#define MSR_IA32_VMX_PROCBASED_CTLS             0x482
-#define MSR_IA32_VMX_EXIT_CTLS                  0x483
-#define MSR_IA32_VMX_ENTRY_CTLS                 0x484
-#define MSR_IA32_VMX_MISC                       0x485
-#define MSR_IA32_VMX_CR0_FIXED0                 0x486
-#define MSR_IA32_VMX_CR0_FIXED1                 0x487
-#define MSR_IA32_VMX_CR4_FIXED0                 0x488
-#define MSR_IA32_VMX_CR4_FIXED1                 0x489
-#define MSR_IA32_VMX_VMCS_ENUM                  0x48a
-#define MSR_IA32_VMX_PROCBASED_CTLS2            0x48b
-#define MSR_IA32_VMX_EPT_VPID_CAP               0x48c
-
-#define MSR_IA32_FEATURE_CONTROL                0x3a
 #define MSR_IA32_FEATURE_CONTROL_LOCKED         0x1
 #define MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED  0x4
 
diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c
index d9249a882aa5..65f0b8a47bed 100644
--- a/arch/x86/lguest/boot.c
+++ b/arch/x86/lguest/boot.c
@@ -55,6 +55,7 @@
 #include <linux/lguest_launcher.h>
 #include <linux/virtio_console.h>
 #include <linux/pm.h>
+#include <asm/apic.h>
 #include <asm/lguest.h>
 #include <asm/paravirt.h>
 #include <asm/param.h>
@@ -783,14 +784,44 @@ static void lguest_wbinvd(void)
  * code qualifies for Advanced.  It will also never interrupt anything.  It
  * does, however, allow us to get through the Linux boot code. */
 #ifdef CONFIG_X86_LOCAL_APIC
-static void lguest_apic_write(unsigned long reg, u32 v)
+static void lguest_apic_write(u32 reg, u32 v)
 {
 }
 
-static u32 lguest_apic_read(unsigned long reg)
+static u32 lguest_apic_read(u32 reg)
 {
 	return 0;
 }
+
+static u64 lguest_apic_icr_read(void)
+{
+	return 0;
+}
+
+static void lguest_apic_icr_write(u32 low, u32 id)
+{
+	/* Warn to see if there's any stray references */
+	WARN_ON(1);
+}
+
+static void lguest_apic_wait_icr_idle(void)
+{
+	return;
+}
+
+static u32 lguest_apic_safe_wait_icr_idle(void)
+{
+	return 0;
+}
+
+static struct apic_ops lguest_basic_apic_ops = {
+	.read = lguest_apic_read,
+	.write = lguest_apic_write,
+	.icr_read = lguest_apic_icr_read,
+	.icr_write = lguest_apic_icr_write,
+	.wait_icr_idle = lguest_apic_wait_icr_idle,
+	.safe_wait_icr_idle = lguest_apic_safe_wait_icr_idle,
+};
 #endif
 
 /* STOP!  Until an interrupt comes in. */
@@ -990,8 +1021,7 @@ __init void lguest_init(void)
 
 #ifdef CONFIG_X86_LOCAL_APIC
 	/* apic read/write intercepts */
-	pv_apic_ops.apic_write = lguest_apic_write;
-	pv_apic_ops.apic_read = lguest_apic_read;
+	apic_ops = &lguest_basic_apic_ops;
 #endif
 
 	/* time operations */
diff --git a/arch/x86/lib/Makefile b/arch/x86/lib/Makefile
index aa3fa4119424..55e11aa6d66c 100644
--- a/arch/x86/lib/Makefile
+++ b/arch/x86/lib/Makefile
@@ -17,9 +17,6 @@ ifeq ($(CONFIG_X86_32),y)
         lib-$(CONFIG_X86_USE_3DNOW) += mmx_32.o
 else
         obj-y += io_64.o iomap_copy_64.o
-
-        CFLAGS_csum-partial_64.o := -funroll-loops
-
         lib-y += csum-partial_64.o csum-copy_64.o csum-wrappers_64.o
         lib-y += thunk_64.o clear_page_64.o copy_page_64.o
         lib-y += memmove_64.o memset_64.o
diff --git a/arch/x86/lib/usercopy_32.c b/arch/x86/lib/usercopy_32.c
index 24e60944971a..9e68075544f6 100644
--- a/arch/x86/lib/usercopy_32.c
+++ b/arch/x86/lib/usercopy_32.c
@@ -14,6 +14,13 @@
 #include <asm/uaccess.h>
 #include <asm/mmx.h>
 
+#ifdef CONFIG_X86_INTEL_USERCOPY
+/*
+ * Alignment at which movsl is preferred for bulk memory copies.
+ */
+struct movsl_mask movsl_mask __read_mostly;
+#endif
+
 static inline int __movsl_is_ok(unsigned long a1, unsigned long a2, unsigned long n)
 {
 #ifdef CONFIG_X86_INTEL_USERCOPY
diff --git a/arch/x86/mach-default/setup.c b/arch/x86/mach-default/setup.c
index 3f2cf11f201a..37b9ae4d44c5 100644
--- a/arch/x86/mach-default/setup.c
+++ b/arch/x86/mach-default/setup.c
@@ -38,15 +38,6 @@ void __init pre_intr_init_hook(void)
 	init_ISA_irqs();
 }
 
-/*
- * IRQ2 is cascade interrupt to second interrupt controller
- */
-static struct irqaction irq2 = {
-	.handler = no_action,
-	.mask = CPU_MASK_NONE,
-	.name = "cascade",
-};
-
 /**
  * intr_init_hook - post gate setup interrupt initialisation
  *
@@ -62,12 +53,6 @@ void __init intr_init_hook(void)
 		if (x86_quirks->arch_intr_init())
 			return;
 	}
-#ifdef CONFIG_X86_LOCAL_APIC
-	apic_intr_init();
-#endif
-
-	if (!acpi_ioapic)
-		setup_irq(2, &irq2);
 }
 
 /**
diff --git a/arch/x86/mach-es7000/Makefile b/arch/x86/mach-es7000/Makefile
deleted file mode 100644
index 3ef8b43b62fc..000000000000
--- a/arch/x86/mach-es7000/Makefile
+++ /dev/null
@@ -1,5 +0,0 @@
-#
-# Makefile for the linux kernel.
-#
-
-obj-$(CONFIG_X86_ES7000)	:= es7000plat.o
diff --git a/arch/x86/mach-es7000/es7000.h b/arch/x86/mach-es7000/es7000.h
deleted file mode 100644
index c8d5aa132fa0..000000000000
--- a/arch/x86/mach-es7000/es7000.h
+++ /dev/null
@@ -1,114 +0,0 @@
-/*
- * Written by: Garry Forsgren, Unisys Corporation
- *             Natalie Protasevich, Unisys Corporation
- * This file contains the code to configure and interface 
- * with Unisys ES7000 series hardware system manager.
- *
- * Copyright (c) 2003 Unisys Corporation.  All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program; if not, write the Free Software Foundation, Inc., 59
- * Temple Place - Suite 330, Boston MA 02111-1307, USA.
- *
- * Contact information: Unisys Corporation, Township Line & Union Meeting 
- * Roads-A, Unisys Way, Blue Bell, Pennsylvania, 19424, or:
- *
- * http://www.unisys.com
- */
-
-/*
- * ES7000 chipsets
- */
-
-#define NON_UNISYS		0
-#define ES7000_CLASSIC		1
-#define ES7000_ZORRO		2
-
-
-#define	MIP_REG			1
-#define	MIP_PSAI_REG		4
-
-#define	MIP_BUSY		1
-#define	MIP_SPIN		0xf0000
-#define	MIP_VALID		0x0100000000000000ULL
-#define	MIP_PORT(VALUE)	((VALUE >> 32) & 0xffff)
-
-#define	MIP_RD_LO(VALUE)	(VALUE & 0xffffffff)   
-
-struct mip_reg_info {
-	unsigned long long mip_info;
-	unsigned long long delivery_info;
-	unsigned long long host_reg;
-	unsigned long long mip_reg;
-};
-
-struct part_info {
-	unsigned char type;   
-	unsigned char length;
-	unsigned char part_id;
-	unsigned char apic_mode;
-	unsigned long snum;    
-	char ptype[16];
-	char sname[64];
-	char pname[64];
-};
-
-struct psai {
-	unsigned long long entry_type;
-	unsigned long long addr;
-	unsigned long long bep_addr;
-};
-
-struct es7000_mem_info {
-	unsigned char type;   
-	unsigned char length;
-	unsigned char resv[6];
-	unsigned long long  start; 
-	unsigned long long  size; 
-};
-
-struct es7000_oem_table {
-	unsigned long long hdr;
-	struct mip_reg_info mip;
-	struct part_info pif;
-	struct es7000_mem_info shm;
-	struct psai psai;
-};
-
-#ifdef CONFIG_ACPI
-
-struct oem_table {
-	struct acpi_table_header Header;
-	u32 OEMTableAddr;
-	u32 OEMTableSize;
-};
-
-extern int find_unisys_acpi_oem_table(unsigned long *oem_addr);
-#endif
-
-struct mip_reg {
-	unsigned long long off_0;
-	unsigned long long off_8;
-	unsigned long long off_10;
-	unsigned long long off_18;
-	unsigned long long off_20;
-	unsigned long long off_28;
-	unsigned long long off_30;
-	unsigned long long off_38;
-};
-
-#define	MIP_SW_APIC		0x1020b
-#define	MIP_FUNC(VALUE) 	(VALUE & 0xff)
-
-extern int parse_unisys_oem (char *oemptr);
-extern void setup_unisys(void);
-extern int es7000_start_cpu(int cpu, unsigned long eip);
-extern void es7000_sw_apic(void);
diff --git a/arch/x86/mach-generic/Makefile b/arch/x86/mach-generic/Makefile
index 0dbd7803a1d5..6730f4e7c744 100644
--- a/arch/x86/mach-generic/Makefile
+++ b/arch/x86/mach-generic/Makefile
@@ -9,4 +9,3 @@ obj-$(CONFIG_X86_NUMAQ)		+= numaq.o
 obj-$(CONFIG_X86_SUMMIT)	+= summit.o
 obj-$(CONFIG_X86_BIGSMP)	+= bigsmp.o
 obj-$(CONFIG_X86_ES7000)	+= es7000.o
-obj-$(CONFIG_X86_ES7000)	+= ../../x86/mach-es7000/
diff --git a/arch/x86/mach-generic/bigsmp.c b/arch/x86/mach-generic/bigsmp.c
index 59d771714559..df37fc9d6a26 100644
--- a/arch/x86/mach-generic/bigsmp.c
+++ b/arch/x86/mach-generic/bigsmp.c
@@ -5,18 +5,17 @@
 #define APIC_DEFINITION 1
 #include <linux/threads.h>
 #include <linux/cpumask.h>
-#include <asm/smp.h>
 #include <asm/mpspec.h>
 #include <asm/genapic.h>
 #include <asm/fixmap.h>
 #include <asm/apicdef.h>
 #include <linux/kernel.h>
-#include <linux/smp.h>
 #include <linux/init.h>
 #include <linux/dmi.h>
-#include <asm/mach-bigsmp/mach_apic.h>
-#include <asm/mach-bigsmp/mach_apicdef.h>
-#include <asm/mach-bigsmp/mach_ipi.h>
+#include <asm/bigsmp/apicdef.h>
+#include <linux/smp.h>
+#include <asm/bigsmp/apic.h>
+#include <asm/bigsmp/ipi.h>
 #include <asm/mach-default/mach_mpparse.h>
 
 static int dmi_bigsmp; /* can be set by dmi scanners */
diff --git a/arch/x86/mach-generic/es7000.c b/arch/x86/mach-generic/es7000.c
index 4742626f08c4..6513d41ea21e 100644
--- a/arch/x86/mach-generic/es7000.c
+++ b/arch/x86/mach-generic/es7000.c
@@ -4,20 +4,19 @@
 #define APIC_DEFINITION 1
 #include <linux/threads.h>
 #include <linux/cpumask.h>
-#include <asm/smp.h>
 #include <asm/mpspec.h>
 #include <asm/genapic.h>
 #include <asm/fixmap.h>
 #include <asm/apicdef.h>
 #include <linux/kernel.h>
 #include <linux/string.h>
-#include <linux/smp.h>
 #include <linux/init.h>
-#include <asm/mach-es7000/mach_apicdef.h>
-#include <asm/mach-es7000/mach_apic.h>
-#include <asm/mach-es7000/mach_ipi.h>
-#include <asm/mach-es7000/mach_mpparse.h>
-#include <asm/mach-es7000/mach_wakecpu.h>
+#include <asm/es7000/apicdef.h>
+#include <linux/smp.h>
+#include <asm/es7000/apic.h>
+#include <asm/es7000/ipi.h>
+#include <asm/es7000/mpparse.h>
+#include <asm/es7000/wakecpu.h>
 
 static int probe_es7000(void)
 {
@@ -48,16 +47,26 @@ static __init int mps_oem_check(struct mp_config_table *mpc, char *oem,
 /* Hook from generic ACPI tables.c */
 static int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 {
-	unsigned long oem_addr;
+	unsigned long oem_addr = 0;
+	int check_dsdt;
+	int ret = 0;
+
+	/* check dsdt at first to avoid clear fix_map for oem_addr */
+	check_dsdt = es7000_check_dsdt();
+
 	if (!find_unisys_acpi_oem_table(&oem_addr)) {
-		if (es7000_check_dsdt())
-			return parse_unisys_oem((char *)oem_addr);
+		if (check_dsdt)
+			ret = parse_unisys_oem((char *)oem_addr);
 		else {
 			setup_unisys();
-			return 1;
+			ret = 1;
 		}
+		/*
+		 * we need to unmap it
+		 */
+		unmap_unisys_acpi_oem_table(oem_addr);
 	}
-	return 0;
+	return ret;
 }
 #else
 static int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id)
diff --git a/arch/x86/mach-generic/numaq.c b/arch/x86/mach-generic/numaq.c
index 8091e68764c4..8cf58394975e 100644
--- a/arch/x86/mach-generic/numaq.c
+++ b/arch/x86/mach-generic/numaq.c
@@ -4,7 +4,6 @@
 #define APIC_DEFINITION 1
 #include <linux/threads.h>
 #include <linux/cpumask.h>
-#include <linux/smp.h>
 #include <asm/mpspec.h>
 #include <asm/genapic.h>
 #include <asm/fixmap.h>
@@ -12,11 +11,12 @@
 #include <linux/kernel.h>
 #include <linux/string.h>
 #include <linux/init.h>
-#include <asm/mach-numaq/mach_apic.h>
-#include <asm/mach-numaq/mach_apicdef.h>
-#include <asm/mach-numaq/mach_ipi.h>
-#include <asm/mach-numaq/mach_mpparse.h>
-#include <asm/mach-numaq/mach_wakecpu.h>
+#include <asm/numaq/apicdef.h>
+#include <linux/smp.h>
+#include <asm/numaq/apic.h>
+#include <asm/numaq/ipi.h>
+#include <asm/numaq/mpparse.h>
+#include <asm/numaq/wakecpu.h>
 #include <asm/numaq.h>
 
 static int mps_oem_check(struct mp_config_table *mpc, char *oem,
diff --git a/arch/x86/mach-generic/summit.c b/arch/x86/mach-generic/summit.c
index a97ea0f35b1e..6ad6b67a723d 100644
--- a/arch/x86/mach-generic/summit.c
+++ b/arch/x86/mach-generic/summit.c
@@ -4,19 +4,18 @@
 #define APIC_DEFINITION 1
 #include <linux/threads.h>
 #include <linux/cpumask.h>
-#include <asm/smp.h>
 #include <asm/mpspec.h>
 #include <asm/genapic.h>
 #include <asm/fixmap.h>
 #include <asm/apicdef.h>
 #include <linux/kernel.h>
 #include <linux/string.h>
-#include <linux/smp.h>
 #include <linux/init.h>
-#include <asm/mach-summit/mach_apic.h>
-#include <asm/mach-summit/mach_apicdef.h>
-#include <asm/mach-summit/mach_ipi.h>
-#include <asm/mach-summit/mach_mpparse.h>
+#include <asm/summit/apicdef.h>
+#include <linux/smp.h>
+#include <asm/summit/apic.h>
+#include <asm/summit/ipi.h>
+#include <asm/summit/mpparse.h>
 
 static int probe_summit(void)
 {
diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile
index dfb932dcf136..59f89b434b45 100644
--- a/arch/x86/mm/Makefile
+++ b/arch/x86/mm/Makefile
@@ -13,12 +13,8 @@ obj-$(CONFIG_MMIOTRACE)		+= mmiotrace.o
 mmiotrace-y			:= pf_in.o mmio-mod.o
 obj-$(CONFIG_MMIOTRACE_TEST)	+= testmmiotrace.o
 
-ifeq ($(CONFIG_X86_32),y)
-obj-$(CONFIG_NUMA)		+= discontig_32.o
-else
-obj-$(CONFIG_NUMA)		+= numa_64.o
+obj-$(CONFIG_NUMA)		+= numa_$(BITS).o
 obj-$(CONFIG_K8_NUMA)		+= k8topology_64.o
-endif
 obj-$(CONFIG_ACPI_NUMA)		+= srat_$(BITS).o
 
 obj-$(CONFIG_MEMTEST)		+= memtest.o
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 8f92cac4e6db..3f2b8962cbd0 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -592,11 +592,6 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code)
 	unsigned long flags;
 #endif
 
-	/*
-	 * We can fault from pretty much anywhere, with unknown IRQ state.
-	 */
-	trace_hardirqs_fixup();
-
 	tsk = current;
 	mm = tsk->mm;
 	prefetchw(&mm->mmap_sem);
@@ -914,15 +909,15 @@ LIST_HEAD(pgd_list);
 
 void vmalloc_sync_all(void)
 {
-#ifdef CONFIG_X86_32
-	unsigned long start = VMALLOC_START & PGDIR_MASK;
 	unsigned long address;
 
+#ifdef CONFIG_X86_32
 	if (SHARED_KERNEL_PMD)
 		return;
 
-	BUILD_BUG_ON(TASK_SIZE & ~PGDIR_MASK);
-	for (address = start; address >= TASK_SIZE; address += PGDIR_SIZE) {
+	for (address = VMALLOC_START & PMD_MASK;
+	     address >= TASK_SIZE && address < FIXADDR_TOP;
+	     address += PMD_SIZE) {
 		unsigned long flags;
 		struct page *page;
 
@@ -935,10 +930,8 @@ void vmalloc_sync_all(void)
 		spin_unlock_irqrestore(&pgd_lock, flags);
 	}
 #else /* CONFIG_X86_64 */
-	unsigned long start = VMALLOC_START & PGDIR_MASK;
-	unsigned long address;
-
-	for (address = start; address <= VMALLOC_END; address += PGDIR_SIZE) {
+	for (address = VMALLOC_START & PGDIR_MASK; address <= VMALLOC_END;
+	     address += PGDIR_SIZE) {
 		const pgd_t *pgd_ref = pgd_offset_k(address);
 		unsigned long flags;
 		struct page *page;
diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c
index 007bb06c7504..4ba373c5b8c8 100644
--- a/arch/x86/mm/gup.c
+++ b/arch/x86/mm/gup.c
@@ -82,7 +82,7 @@ static noinline int gup_pte_range(pmd_t pmd, unsigned long addr,
 		pte_t pte = gup_get_pte(ptep);
 		struct page *page;
 
-		if ((pte_val(pte) & (mask | _PAGE_SPECIAL)) != mask) {
+		if ((pte_flags(pte) & (mask | _PAGE_SPECIAL)) != mask) {
 			pte_unmap(ptep);
 			return 0;
 		}
@@ -116,10 +116,10 @@ static noinline int gup_huge_pmd(pmd_t pmd, unsigned long addr,
 	mask = _PAGE_PRESENT|_PAGE_USER;
 	if (write)
 		mask |= _PAGE_RW;
-	if ((pte_val(pte) & mask) != mask)
+	if ((pte_flags(pte) & mask) != mask)
 		return 0;
 	/* hugepages are never "special" */
-	VM_BUG_ON(pte_val(pte) & _PAGE_SPECIAL);
+	VM_BUG_ON(pte_flags(pte) & _PAGE_SPECIAL);
 	VM_BUG_ON(!pfn_valid(pte_pfn(pte)));
 
 	refs = 0;
@@ -173,10 +173,10 @@ static noinline int gup_huge_pud(pud_t pud, unsigned long addr,
 	mask = _PAGE_PRESENT|_PAGE_USER;
 	if (write)
 		mask |= _PAGE_RW;
-	if ((pte_val(pte) & mask) != mask)
+	if ((pte_flags(pte) & mask) != mask)
 		return 0;
 	/* hugepages are never "special" */
-	VM_BUG_ON(pte_val(pte) & _PAGE_SPECIAL);
+	VM_BUG_ON(pte_flags(pte) & _PAGE_SPECIAL);
 	VM_BUG_ON(!pfn_valid(pte_pfn(pte)));
 
 	refs = 0;
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index c3789bb19308..8396868e82c5 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -31,6 +31,7 @@
 #include <linux/cpumask.h>
 
 #include <asm/asm.h>
+#include <asm/bios_ebda.h>
 #include <asm/processor.h>
 #include <asm/system.h>
 #include <asm/uaccess.h>
@@ -557,7 +558,7 @@ void zap_low_mappings(void)
 
 int nx_enabled;
 
-pteval_t __supported_pte_mask __read_mostly = ~(_PAGE_NX | _PAGE_GLOBAL);
+pteval_t __supported_pte_mask __read_mostly = ~(_PAGE_NX | _PAGE_GLOBAL | _PAGE_IOMAP);
 EXPORT_SYMBOL_GPL(__supported_pte_mask);
 
 #ifdef CONFIG_X86_PAE
@@ -969,6 +970,8 @@ void __init mem_init(void)
 	int codesize, reservedpages, datasize, initsize;
 	int tmp;
 
+	start_periodic_check_for_corruption();
+
 #ifdef CONFIG_FLATMEM
 	BUG_ON(!mem_map);
 #endif
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index fb30486c82f7..b8e461d49412 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -31,6 +31,7 @@
 #include <linux/nmi.h>
 
 #include <asm/processor.h>
+#include <asm/bios_ebda.h>
 #include <asm/system.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -88,6 +89,62 @@ early_param("gbpages", parse_direct_gbpages_on);
 
 int after_bootmem;
 
+pteval_t __supported_pte_mask __read_mostly = ~_PAGE_IOMAP;
+EXPORT_SYMBOL_GPL(__supported_pte_mask);
+
+static int do_not_nx __cpuinitdata;
+
+/*
+ * noexec=on|off
+ * Control non-executable mappings for 64-bit processes.
+ *
+ * on	Enable (default)
+ * off	Disable
+ */
+static int __init nonx_setup(char *str)
+{
+	if (!str)
+		return -EINVAL;
+	if (!strncmp(str, "on", 2)) {
+		__supported_pte_mask |= _PAGE_NX;
+		do_not_nx = 0;
+	} else if (!strncmp(str, "off", 3)) {
+		do_not_nx = 1;
+		__supported_pte_mask &= ~_PAGE_NX;
+	}
+	return 0;
+}
+early_param("noexec", nonx_setup);
+
+void __cpuinit check_efer(void)
+{
+	unsigned long efer;
+
+	rdmsrl(MSR_EFER, efer);
+	if (!(efer & EFER_NX) || do_not_nx)
+		__supported_pte_mask &= ~_PAGE_NX;
+}
+
+int force_personality32;
+
+/*
+ * noexec32=on|off
+ * Control non executable heap for 32bit processes.
+ * To control the stack too use noexec=off
+ *
+ * on	PROT_READ does not imply PROT_EXEC for 32-bit processes (default)
+ * off	PROT_READ implies PROT_EXEC
+ */
+static int __init nonx32_setup(char *str)
+{
+	if (!strcmp(str, "on"))
+		force_personality32 &= ~READ_IMPLIES_EXEC;
+	else if (!strcmp(str, "off"))
+		force_personality32 |= READ_IMPLIES_EXEC;
+	return 1;
+}
+__setup("noexec32=", nonx32_setup);
+
 /*
  * NOTE: This function is marked __ref because it calls __init function
  * (alloc_bootmem_pages). It's safe to do it ONLY when after_bootmem == 0.
@@ -139,9 +196,6 @@ set_pte_vaddr_pud(pud_t *pud_page, unsigned long vaddr, pte_t new_pte)
 	}
 
 	pte = pte_offset_kernel(pmd, vaddr);
-	if (!pte_none(*pte) && pte_val(new_pte) &&
-	    pte_val(*pte) != (pte_val(new_pte) & __supported_pte_mask))
-		pte_ERROR(*pte);
 	set_pte(pte, new_pte);
 
 	/*
@@ -256,7 +310,7 @@ static __ref void *alloc_low_page(unsigned long *phys)
 	if (pfn >= table_top)
 		panic("alloc_low_page: ran out of memory");
 
-	adr = early_ioremap(pfn * PAGE_SIZE, PAGE_SIZE);
+	adr = early_memremap(pfn * PAGE_SIZE, PAGE_SIZE);
 	memset(adr, 0, PAGE_SIZE);
 	*phys  = pfn * PAGE_SIZE;
 	return adr;
@@ -692,7 +746,7 @@ unsigned long __init_refok init_memory_mapping(unsigned long start,
 		old_start = mr[i].start;
 		memmove(&mr[i], &mr[i+1],
 			 (nr_range - 1 - i) * sizeof (struct map_range));
-		mr[i].start = old_start;
+		mr[i--].start = old_start;
 		nr_range--;
 	}
 
@@ -825,6 +879,8 @@ void __init mem_init(void)
 {
 	long codesize, reservedpages, datasize, initsize;
 
+	start_periodic_check_for_corruption();
+
 	pci_iommu_alloc();
 
 	/* clear_bss() already clear the empty_zero_page */
diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index 6ab3196d12b4..e4c43ec71b29 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -24,18 +24,47 @@
 
 #ifdef CONFIG_X86_64
 
+static inline int phys_addr_valid(unsigned long addr)
+{
+	return addr < (1UL << boot_cpu_data.x86_phys_bits);
+}
+
 unsigned long __phys_addr(unsigned long x)
 {
-	if (x >= __START_KERNEL_map)
-		return x - __START_KERNEL_map + phys_base;
-	return x - PAGE_OFFSET;
+	if (x >= __START_KERNEL_map) {
+		x -= __START_KERNEL_map;
+		VIRTUAL_BUG_ON(x >= KERNEL_IMAGE_SIZE);
+		x += phys_base;
+	} else {
+		VIRTUAL_BUG_ON(x < PAGE_OFFSET);
+		x -= PAGE_OFFSET;
+		VIRTUAL_BUG_ON(system_state == SYSTEM_BOOTING ? x > MAXMEM :
+					!phys_addr_valid(x));
+	}
+	return x;
 }
 EXPORT_SYMBOL(__phys_addr);
 
-static inline int phys_addr_valid(unsigned long addr)
+bool __virt_addr_valid(unsigned long x)
 {
-	return addr < (1UL << boot_cpu_data.x86_phys_bits);
+	if (x >= __START_KERNEL_map) {
+		x -= __START_KERNEL_map;
+		if (x >= KERNEL_IMAGE_SIZE)
+			return false;
+		x += phys_base;
+	} else {
+		if (x < PAGE_OFFSET)
+			return false;
+		x -= PAGE_OFFSET;
+		if (system_state == SYSTEM_BOOTING ?
+				x > MAXMEM : !phys_addr_valid(x)) {
+			return false;
+		}
+	}
+
+	return pfn_valid(x >> PAGE_SHIFT);
 }
+EXPORT_SYMBOL(__virt_addr_valid);
 
 #else
 
@@ -44,6 +73,28 @@ static inline int phys_addr_valid(unsigned long addr)
 	return 1;
 }
 
+#ifdef CONFIG_DEBUG_VIRTUAL
+unsigned long __phys_addr(unsigned long x)
+{
+	/* VMALLOC_* aren't constants; not available at the boot time */
+	VIRTUAL_BUG_ON(x < PAGE_OFFSET);
+	VIRTUAL_BUG_ON(system_state != SYSTEM_BOOTING &&
+		is_vmalloc_addr((void *) x));
+	return x - PAGE_OFFSET;
+}
+EXPORT_SYMBOL(__phys_addr);
+#endif
+
+bool __virt_addr_valid(unsigned long x)
+{
+	if (x < PAGE_OFFSET)
+		return false;
+	if (system_state != SYSTEM_BOOTING && is_vmalloc_addr((void *) x))
+		return false;
+	return pfn_valid((x - PAGE_OFFSET) >> PAGE_SHIFT);
+}
+EXPORT_SYMBOL(__virt_addr_valid);
+
 #endif
 
 int page_is_ram(unsigned long pagenr)
@@ -223,16 +274,16 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr,
 	switch (prot_val) {
 	case _PAGE_CACHE_UC:
 	default:
-		prot = PAGE_KERNEL_NOCACHE;
+		prot = PAGE_KERNEL_IO_NOCACHE;
 		break;
 	case _PAGE_CACHE_UC_MINUS:
-		prot = PAGE_KERNEL_UC_MINUS;
+		prot = PAGE_KERNEL_IO_UC_MINUS;
 		break;
 	case _PAGE_CACHE_WC:
-		prot = PAGE_KERNEL_WC;
+		prot = PAGE_KERNEL_IO_WC;
 		break;
 	case _PAGE_CACHE_WB:
-		prot = PAGE_KERNEL;
+		prot = PAGE_KERNEL_IO;
 		break;
 	}
 
@@ -549,12 +600,12 @@ static void __init __early_set_fixmap(enum fixed_addresses idx,
 }
 
 static inline void __init early_set_fixmap(enum fixed_addresses idx,
-					unsigned long phys)
+					   unsigned long phys, pgprot_t prot)
 {
 	if (after_paging_init)
-		set_fixmap(idx, phys);
+		__set_fixmap(idx, phys, prot);
 	else
-		__early_set_fixmap(idx, phys, PAGE_KERNEL);
+		__early_set_fixmap(idx, phys, prot);
 }
 
 static inline void __init early_clear_fixmap(enum fixed_addresses idx)
@@ -565,16 +616,22 @@ static inline void __init early_clear_fixmap(enum fixed_addresses idx)
 		__early_set_fixmap(idx, 0, __pgprot(0));
 }
 
-
-static int __initdata early_ioremap_nested;
-
+static void *prev_map[FIX_BTMAPS_SLOTS] __initdata;
+static unsigned long prev_size[FIX_BTMAPS_SLOTS] __initdata;
 static int __init check_early_ioremap_leak(void)
 {
-	if (!early_ioremap_nested)
+	int count = 0;
+	int i;
+
+	for (i = 0; i < FIX_BTMAPS_SLOTS; i++)
+		if (prev_map[i])
+			count++;
+
+	if (!count)
 		return 0;
 	WARN(1, KERN_WARNING
 	       "Debug warning: early ioremap leak of %d areas detected.\n",
-		early_ioremap_nested);
+		count);
 	printk(KERN_WARNING
 		"please boot with early_ioremap_debug and report the dmesg.\n");
 
@@ -582,18 +639,33 @@ static int __init check_early_ioremap_leak(void)
 }
 late_initcall(check_early_ioremap_leak);
 
-void __init *early_ioremap(unsigned long phys_addr, unsigned long size)
+static void __init *__early_ioremap(unsigned long phys_addr, unsigned long size, pgprot_t prot)
 {
 	unsigned long offset, last_addr;
-	unsigned int nrpages, nesting;
+	unsigned int nrpages;
 	enum fixed_addresses idx0, idx;
+	int i, slot;
 
 	WARN_ON(system_state != SYSTEM_BOOTING);
 
-	nesting = early_ioremap_nested;
+	slot = -1;
+	for (i = 0; i < FIX_BTMAPS_SLOTS; i++) {
+		if (!prev_map[i]) {
+			slot = i;
+			break;
+		}
+	}
+
+	if (slot < 0) {
+		printk(KERN_INFO "early_iomap(%08lx, %08lx) not found slot\n",
+			 phys_addr, size);
+		WARN_ON(1);
+		return NULL;
+	}
+
 	if (early_ioremap_debug) {
 		printk(KERN_INFO "early_ioremap(%08lx, %08lx) [%d] => ",
-		       phys_addr, size, nesting);
+		       phys_addr, size, slot);
 		dump_stack();
 	}
 
@@ -604,17 +676,13 @@ void __init *early_ioremap(unsigned long phys_addr, unsigned long size)
 		return NULL;
 	}
 
-	if (nesting >= FIX_BTMAPS_NESTING) {
-		WARN_ON(1);
-		return NULL;
-	}
-	early_ioremap_nested++;
+	prev_size[slot] = size;
 	/*
 	 * Mappings have to be page-aligned
 	 */
 	offset = phys_addr & ~PAGE_MASK;
 	phys_addr &= PAGE_MASK;
-	size = PAGE_ALIGN(last_addr) - phys_addr;
+	size = PAGE_ALIGN(last_addr + 1) - phys_addr;
 
 	/*
 	 * Mappings have to fit in the FIX_BTMAP area.
@@ -628,10 +696,10 @@ void __init *early_ioremap(unsigned long phys_addr, unsigned long size)
 	/*
 	 * Ok, go for it..
 	 */
-	idx0 = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*nesting;
+	idx0 = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*slot;
 	idx = idx0;
 	while (nrpages > 0) {
-		early_set_fixmap(idx, phys_addr);
+		early_set_fixmap(idx, phys_addr, prot);
 		phys_addr += PAGE_SIZE;
 		--idx;
 		--nrpages;
@@ -639,7 +707,20 @@ void __init *early_ioremap(unsigned long phys_addr, unsigned long size)
 	if (early_ioremap_debug)
 		printk(KERN_CONT "%08lx + %08lx\n", offset, fix_to_virt(idx0));
 
-	return (void *) (offset + fix_to_virt(idx0));
+	prev_map[slot] = (void *) (offset + fix_to_virt(idx0));
+	return prev_map[slot];
+}
+
+/* Remap an IO device */
+void __init *early_ioremap(unsigned long phys_addr, unsigned long size)
+{
+	return __early_ioremap(phys_addr, size, PAGE_KERNEL_IO);
+}
+
+/* Remap memory */
+void __init *early_memremap(unsigned long phys_addr, unsigned long size)
+{
+	return __early_ioremap(phys_addr, size, PAGE_KERNEL);
 }
 
 void __init early_iounmap(void *addr, unsigned long size)
@@ -648,15 +729,33 @@ void __init early_iounmap(void *addr, unsigned long size)
 	unsigned long offset;
 	unsigned int nrpages;
 	enum fixed_addresses idx;
-	int nesting;
+	int i, slot;
+
+	slot = -1;
+	for (i = 0; i < FIX_BTMAPS_SLOTS; i++) {
+		if (prev_map[i] == addr) {
+			slot = i;
+			break;
+		}
+	}
 
-	nesting = --early_ioremap_nested;
-	if (WARN_ON(nesting < 0))
+	if (slot < 0) {
+		printk(KERN_INFO "early_iounmap(%p, %08lx) not found slot\n",
+			 addr, size);
+		WARN_ON(1);
 		return;
+	}
+
+	if (prev_size[slot] != size) {
+		printk(KERN_INFO "early_iounmap(%p, %08lx) [%d] size not consistent %08lx\n",
+			 addr, size, slot, prev_size[slot]);
+		WARN_ON(1);
+		return;
+	}
 
 	if (early_ioremap_debug) {
 		printk(KERN_INFO "early_iounmap(%p, %08lx) [%d]\n", addr,
-		       size, nesting);
+		       size, slot);
 		dump_stack();
 	}
 
@@ -668,12 +767,13 @@ void __init early_iounmap(void *addr, unsigned long size)
 	offset = virt_addr & ~PAGE_MASK;
 	nrpages = PAGE_ALIGN(offset + size - 1) >> PAGE_SHIFT;
 
-	idx = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*nesting;
+	idx = FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*slot;
 	while (nrpages > 0) {
 		early_clear_fixmap(idx);
 		--idx;
 		--nrpages;
 	}
+	prev_map[slot] = 0;
 }
 
 void __this_fixmap_does_not_exist(void)
diff --git a/arch/x86/mm/discontig_32.c b/arch/x86/mm/numa_32.c
index 847c164725f4..847c164725f4 100644
--- a/arch/x86/mm/discontig_32.c
+++ b/arch/x86/mm/numa_32.c
diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c
index 1b4763e26ea9..51c0a2fc14fe 100644
--- a/arch/x86/mm/srat_64.c
+++ b/arch/x86/mm/srat_64.c
@@ -138,7 +138,7 @@ acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *pa)
 		return;
 	}
 
-	if (is_uv_system())
+	if (get_uv_system_type() >= UV_X2APIC)
 		apic_id = (pa->apic_id << 8) | pa->local_sapic_eid;
 	else
 		apic_id = pa->apic_id;
diff --git a/arch/x86/oprofile/Makefile b/arch/x86/oprofile/Makefile
index 30f3eb366667..446902b2a6b6 100644
--- a/arch/x86/oprofile/Makefile
+++ b/arch/x86/oprofile/Makefile
@@ -7,6 +7,6 @@ DRIVER_OBJS = $(addprefix ../../../drivers/oprofile/, \
 		timer_int.o )
 
 oprofile-y				:= $(DRIVER_OBJS) init.o backtrace.o
-oprofile-$(CONFIG_X86_LOCAL_APIC) 	+= nmi_int.o op_model_athlon.o \
+oprofile-$(CONFIG_X86_LOCAL_APIC) 	+= nmi_int.o op_model_amd.o \
 					   op_model_ppro.o op_model_p4.o
 oprofile-$(CONFIG_X86_IO_APIC)		+= nmi_timer_int.o
diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 8a5f1614a3d5..57f6c9088081 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -1,10 +1,11 @@
 /**
  * @file nmi_int.c
  *
- * @remark Copyright 2002 OProfile authors
+ * @remark Copyright 2002-2008 OProfile authors
  * @remark Read the file COPYING
  *
  * @author John Levon <levon@movementarian.org>
+ * @author Robert Richter <robert.richter@amd.com>
  */
 
 #include <linux/init.h>
@@ -439,6 +440,7 @@ int __init op_nmi_init(struct oprofile_operations *ops)
 	__u8 vendor = boot_cpu_data.x86_vendor;
 	__u8 family = boot_cpu_data.x86;
 	char *cpu_type;
+	int ret = 0;
 
 	if (!cpu_has_apic)
 		return -ENODEV;
@@ -451,19 +453,23 @@ int __init op_nmi_init(struct oprofile_operations *ops)
 		default:
 			return -ENODEV;
 		case 6:
-			model = &op_athlon_spec;
+			model = &op_amd_spec;
 			cpu_type = "i386/athlon";
 			break;
 		case 0xf:
-			model = &op_athlon_spec;
+			model = &op_amd_spec;
 			/* Actually it could be i386/hammer too, but give
 			 user space an consistent name. */
 			cpu_type = "x86-64/hammer";
 			break;
 		case 0x10:
-			model = &op_athlon_spec;
+			model = &op_amd_spec;
 			cpu_type = "x86-64/family10";
 			break;
+		case 0x11:
+			model = &op_amd_spec;
+			cpu_type = "x86-64/family11h";
+			break;
 		}
 		break;
 
@@ -490,17 +496,24 @@ int __init op_nmi_init(struct oprofile_operations *ops)
 		return -ENODEV;
 	}
 
-	init_sysfs();
 #ifdef CONFIG_SMP
 	register_cpu_notifier(&oprofile_cpu_nb);
 #endif
-	using_nmi = 1;
+	/* default values, can be overwritten by model */
 	ops->create_files = nmi_create_files;
 	ops->setup = nmi_setup;
 	ops->shutdown = nmi_shutdown;
 	ops->start = nmi_start;
 	ops->stop = nmi_stop;
 	ops->cpu_type = cpu_type;
+
+	if (model->init)
+		ret = model->init(ops);
+	if (ret)
+		return ret;
+
+	init_sysfs();
+	using_nmi = 1;
 	printk(KERN_INFO "oprofile: using NMI interrupt.\n");
 	return 0;
 }
@@ -513,4 +526,6 @@ void op_nmi_exit(void)
 		unregister_cpu_notifier(&oprofile_cpu_nb);
 #endif
 	}
+	if (model->exit)
+		model->exit();
 }
diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c
new file mode 100644
index 000000000000..d9faf607b3a6
--- /dev/null
+++ b/arch/x86/oprofile/op_model_amd.c
@@ -0,0 +1,543 @@
+/*
+ * @file op_model_amd.c
+ * athlon / K7 / K8 / Family 10h model-specific MSR operations
+ *
+ * @remark Copyright 2002-2008 OProfile authors
+ * @remark Read the file COPYING
+ *
+ * @author John Levon
+ * @author Philippe Elie
+ * @author Graydon Hoare
+ * @author Robert Richter <robert.richter@amd.com>
+ * @author Barry Kasindorf
+*/
+
+#include <linux/oprofile.h>
+#include <linux/device.h>
+#include <linux/pci.h>
+
+#include <asm/ptrace.h>
+#include <asm/msr.h>
+#include <asm/nmi.h>
+
+#include "op_x86_model.h"
+#include "op_counter.h"
+
+#define NUM_COUNTERS 4
+#define NUM_CONTROLS 4
+
+#define CTR_IS_RESERVED(msrs, c) (msrs->counters[(c)].addr ? 1 : 0)
+#define CTR_READ(l, h, msrs, c) do {rdmsr(msrs->counters[(c)].addr, (l), (h)); } while (0)
+#define CTR_WRITE(l, msrs, c) do {wrmsr(msrs->counters[(c)].addr, -(unsigned int)(l), -1); } while (0)
+#define CTR_OVERFLOWED(n) (!((n) & (1U<<31)))
+
+#define CTRL_IS_RESERVED(msrs, c) (msrs->controls[(c)].addr ? 1 : 0)
+#define CTRL_READ(l, h, msrs, c) do {rdmsr(msrs->controls[(c)].addr, (l), (h)); } while (0)
+#define CTRL_WRITE(l, h, msrs, c) do {wrmsr(msrs->controls[(c)].addr, (l), (h)); } while (0)
+#define CTRL_SET_ACTIVE(n) (n |= (1<<22))
+#define CTRL_SET_INACTIVE(n) (n &= ~(1<<22))
+#define CTRL_CLEAR_LO(x) (x &= (1<<21))
+#define CTRL_CLEAR_HI(x) (x &= 0xfffffcf0)
+#define CTRL_SET_ENABLE(val) (val |= 1<<20)
+#define CTRL_SET_USR(val, u) (val |= ((u & 1) << 16))
+#define CTRL_SET_KERN(val, k) (val |= ((k & 1) << 17))
+#define CTRL_SET_UM(val, m) (val |= (m << 8))
+#define CTRL_SET_EVENT_LOW(val, e) (val |= (e & 0xff))
+#define CTRL_SET_EVENT_HIGH(val, e) (val |= ((e >> 8) & 0xf))
+#define CTRL_SET_HOST_ONLY(val, h) (val |= ((h & 1) << 9))
+#define CTRL_SET_GUEST_ONLY(val, h) (val |= ((h & 1) << 8))
+
+static unsigned long reset_value[NUM_COUNTERS];
+
+#ifdef CONFIG_OPROFILE_IBS
+
+/* IbsFetchCtl bits/masks */
+#define IBS_FETCH_HIGH_VALID_BIT	(1UL << 17)	/* bit 49 */
+#define IBS_FETCH_HIGH_ENABLE		(1UL << 16)	/* bit 48 */
+#define IBS_FETCH_LOW_MAX_CNT_MASK	0x0000FFFFUL	/* MaxCnt mask */
+
+/*IbsOpCtl bits */
+#define IBS_OP_LOW_VALID_BIT		(1ULL<<18)	/* bit 18 */
+#define IBS_OP_LOW_ENABLE		(1ULL<<17)	/* bit 17 */
+
+/* Codes used in cpu_buffer.c */
+/* This produces duplicate code, need to be fixed */
+#define IBS_FETCH_BEGIN 3
+#define IBS_OP_BEGIN    4
+
+/* The function interface needs to be fixed, something like add
+   data. Should then be added to linux/oprofile.h. */
+extern void oprofile_add_ibs_sample(struct pt_regs *const regs,
+				    unsigned int * const ibs_sample, u8 code);
+
+struct ibs_fetch_sample {
+	/* MSRC001_1031 IBS Fetch Linear Address Register */
+	unsigned int ibs_fetch_lin_addr_low;
+	unsigned int ibs_fetch_lin_addr_high;
+	/* MSRC001_1030 IBS Fetch Control Register */
+	unsigned int ibs_fetch_ctl_low;
+	unsigned int ibs_fetch_ctl_high;
+	/* MSRC001_1032 IBS Fetch Physical Address Register */
+	unsigned int ibs_fetch_phys_addr_low;
+	unsigned int ibs_fetch_phys_addr_high;
+};
+
+struct ibs_op_sample {
+	/* MSRC001_1034 IBS Op Logical Address Register (IbsRIP) */
+	unsigned int ibs_op_rip_low;
+	unsigned int ibs_op_rip_high;
+	/* MSRC001_1035 IBS Op Data Register */
+	unsigned int ibs_op_data1_low;
+	unsigned int ibs_op_data1_high;
+	/* MSRC001_1036 IBS Op Data 2 Register */
+	unsigned int ibs_op_data2_low;
+	unsigned int ibs_op_data2_high;
+	/* MSRC001_1037 IBS Op Data 3 Register */
+	unsigned int ibs_op_data3_low;
+	unsigned int ibs_op_data3_high;
+	/* MSRC001_1038 IBS DC Linear Address Register (IbsDcLinAd) */
+	unsigned int ibs_dc_linear_low;
+	unsigned int ibs_dc_linear_high;
+	/* MSRC001_1039 IBS DC Physical Address Register (IbsDcPhysAd) */
+	unsigned int ibs_dc_phys_low;
+	unsigned int ibs_dc_phys_high;
+};
+
+/*
+ * unitialize the APIC for the IBS interrupts if needed on AMD Family10h+
+*/
+static void clear_ibs_nmi(void);
+
+static int ibs_allowed;	/* AMD Family10h and later */
+
+struct op_ibs_config {
+	unsigned long op_enabled;
+	unsigned long fetch_enabled;
+	unsigned long max_cnt_fetch;
+	unsigned long max_cnt_op;
+	unsigned long rand_en;
+	unsigned long dispatched_ops;
+};
+
+static struct op_ibs_config ibs_config;
+
+#endif
+
+/* functions for op_amd_spec */
+
+static void op_amd_fill_in_addresses(struct op_msrs * const msrs)
+{
+	int i;
+
+	for (i = 0; i < NUM_COUNTERS; i++) {
+		if (reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i))
+			msrs->counters[i].addr = MSR_K7_PERFCTR0 + i;
+		else
+			msrs->counters[i].addr = 0;
+	}
+
+	for (i = 0; i < NUM_CONTROLS; i++) {
+		if (reserve_evntsel_nmi(MSR_K7_EVNTSEL0 + i))
+			msrs->controls[i].addr = MSR_K7_EVNTSEL0 + i;
+		else
+			msrs->controls[i].addr = 0;
+	}
+}
+
+
+static void op_amd_setup_ctrs(struct op_msrs const * const msrs)
+{
+	unsigned int low, high;
+	int i;
+
+	/* clear all counters */
+	for (i = 0 ; i < NUM_CONTROLS; ++i) {
+		if (unlikely(!CTRL_IS_RESERVED(msrs, i)))
+			continue;
+		CTRL_READ(low, high, msrs, i);
+		CTRL_CLEAR_LO(low);
+		CTRL_CLEAR_HI(high);
+		CTRL_WRITE(low, high, msrs, i);
+	}
+
+	/* avoid a false detection of ctr overflows in NMI handler */
+	for (i = 0; i < NUM_COUNTERS; ++i) {
+		if (unlikely(!CTR_IS_RESERVED(msrs, i)))
+			continue;
+		CTR_WRITE(1, msrs, i);
+	}
+
+	/* enable active counters */
+	for (i = 0; i < NUM_COUNTERS; ++i) {
+		if ((counter_config[i].enabled) && (CTR_IS_RESERVED(msrs, i))) {
+			reset_value[i] = counter_config[i].count;
+
+			CTR_WRITE(counter_config[i].count, msrs, i);
+
+			CTRL_READ(low, high, msrs, i);
+			CTRL_CLEAR_LO(low);
+			CTRL_CLEAR_HI(high);
+			CTRL_SET_ENABLE(low);
+			CTRL_SET_USR(low, counter_config[i].user);
+			CTRL_SET_KERN(low, counter_config[i].kernel);
+			CTRL_SET_UM(low, counter_config[i].unit_mask);
+			CTRL_SET_EVENT_LOW(low, counter_config[i].event);
+			CTRL_SET_EVENT_HIGH(high, counter_config[i].event);
+			CTRL_SET_HOST_ONLY(high, 0);
+			CTRL_SET_GUEST_ONLY(high, 0);
+
+			CTRL_WRITE(low, high, msrs, i);
+		} else {
+			reset_value[i] = 0;
+		}
+	}
+}
+
+#ifdef CONFIG_OPROFILE_IBS
+
+static inline int
+op_amd_handle_ibs(struct pt_regs * const regs,
+		  struct op_msrs const * const msrs)
+{
+	unsigned int low, high;
+	struct ibs_fetch_sample ibs_fetch;
+	struct ibs_op_sample ibs_op;
+
+	if (!ibs_allowed)
+		return 1;
+
+	if (ibs_config.fetch_enabled) {
+		rdmsr(MSR_AMD64_IBSFETCHCTL, low, high);
+		if (high & IBS_FETCH_HIGH_VALID_BIT) {
+			ibs_fetch.ibs_fetch_ctl_high = high;
+			ibs_fetch.ibs_fetch_ctl_low = low;
+			rdmsr(MSR_AMD64_IBSFETCHLINAD, low, high);
+			ibs_fetch.ibs_fetch_lin_addr_high = high;
+			ibs_fetch.ibs_fetch_lin_addr_low = low;
+			rdmsr(MSR_AMD64_IBSFETCHPHYSAD, low, high);
+			ibs_fetch.ibs_fetch_phys_addr_high = high;
+			ibs_fetch.ibs_fetch_phys_addr_low = low;
+
+			oprofile_add_ibs_sample(regs,
+						(unsigned int *)&ibs_fetch,
+						IBS_FETCH_BEGIN);
+
+			/*reenable the IRQ */
+			rdmsr(MSR_AMD64_IBSFETCHCTL, low, high);
+			high &= ~IBS_FETCH_HIGH_VALID_BIT;
+			high |= IBS_FETCH_HIGH_ENABLE;
+			low &= IBS_FETCH_LOW_MAX_CNT_MASK;
+			wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
+		}
+	}
+
+	if (ibs_config.op_enabled) {
+		rdmsr(MSR_AMD64_IBSOPCTL, low, high);
+		if (low & IBS_OP_LOW_VALID_BIT) {
+			rdmsr(MSR_AMD64_IBSOPRIP, low, high);
+			ibs_op.ibs_op_rip_low = low;
+			ibs_op.ibs_op_rip_high = high;
+			rdmsr(MSR_AMD64_IBSOPDATA, low, high);
+			ibs_op.ibs_op_data1_low = low;
+			ibs_op.ibs_op_data1_high = high;
+			rdmsr(MSR_AMD64_IBSOPDATA2, low, high);
+			ibs_op.ibs_op_data2_low = low;
+			ibs_op.ibs_op_data2_high = high;
+			rdmsr(MSR_AMD64_IBSOPDATA3, low, high);
+			ibs_op.ibs_op_data3_low = low;
+			ibs_op.ibs_op_data3_high = high;
+			rdmsr(MSR_AMD64_IBSDCLINAD, low, high);
+			ibs_op.ibs_dc_linear_low = low;
+			ibs_op.ibs_dc_linear_high = high;
+			rdmsr(MSR_AMD64_IBSDCPHYSAD, low, high);
+			ibs_op.ibs_dc_phys_low = low;
+			ibs_op.ibs_dc_phys_high = high;
+
+			/* reenable the IRQ */
+			oprofile_add_ibs_sample(regs,
+						(unsigned int *)&ibs_op,
+						IBS_OP_BEGIN);
+			rdmsr(MSR_AMD64_IBSOPCTL, low, high);
+			high = 0;
+			low &= ~IBS_OP_LOW_VALID_BIT;
+			low |= IBS_OP_LOW_ENABLE;
+			wrmsr(MSR_AMD64_IBSOPCTL, low, high);
+		}
+	}
+
+	return 1;
+}
+
+#endif
+
+static int op_amd_check_ctrs(struct pt_regs * const regs,
+			     struct op_msrs const * const msrs)
+{
+	unsigned int low, high;
+	int i;
+
+	for (i = 0 ; i < NUM_COUNTERS; ++i) {
+		if (!reset_value[i])
+			continue;
+		CTR_READ(low, high, msrs, i);
+		if (CTR_OVERFLOWED(low)) {
+			oprofile_add_sample(regs, i);
+			CTR_WRITE(reset_value[i], msrs, i);
+		}
+	}
+
+#ifdef CONFIG_OPROFILE_IBS
+	op_amd_handle_ibs(regs, msrs);
+#endif
+
+	/* See op_model_ppro.c */
+	return 1;
+}
+
+static void op_amd_start(struct op_msrs const * const msrs)
+{
+	unsigned int low, high;
+	int i;
+	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
+		if (reset_value[i]) {
+			CTRL_READ(low, high, msrs, i);
+			CTRL_SET_ACTIVE(low);
+			CTRL_WRITE(low, high, msrs, i);
+		}
+	}
+
+#ifdef CONFIG_OPROFILE_IBS
+	if (ibs_allowed && ibs_config.fetch_enabled) {
+		low = (ibs_config.max_cnt_fetch >> 4) & 0xFFFF;
+		high = IBS_FETCH_HIGH_ENABLE;
+		wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
+	}
+
+	if (ibs_allowed && ibs_config.op_enabled) {
+		low = ((ibs_config.max_cnt_op >> 4) & 0xFFFF) + IBS_OP_LOW_ENABLE;
+		high = 0;
+		wrmsr(MSR_AMD64_IBSOPCTL, low, high);
+	}
+#endif
+}
+
+
+static void op_amd_stop(struct op_msrs const * const msrs)
+{
+	unsigned int low, high;
+	int i;
+
+	/* Subtle: stop on all counters to avoid race with
+	 * setting our pm callback */
+	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
+		if (!reset_value[i])
+			continue;
+		CTRL_READ(low, high, msrs, i);
+		CTRL_SET_INACTIVE(low);
+		CTRL_WRITE(low, high, msrs, i);
+	}
+
+#ifdef CONFIG_OPROFILE_IBS
+	if (ibs_allowed && ibs_config.fetch_enabled) {
+		low = 0;		/* clear max count and enable */
+		high = 0;
+		wrmsr(MSR_AMD64_IBSFETCHCTL, low, high);
+	}
+
+	if (ibs_allowed && ibs_config.op_enabled) {
+		low = 0;		/* clear max count and enable */
+		high = 0;
+		wrmsr(MSR_AMD64_IBSOPCTL, low, high);
+	}
+#endif
+}
+
+static void op_amd_shutdown(struct op_msrs const * const msrs)
+{
+	int i;
+
+	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
+		if (CTR_IS_RESERVED(msrs, i))
+			release_perfctr_nmi(MSR_K7_PERFCTR0 + i);
+	}
+	for (i = 0 ; i < NUM_CONTROLS ; ++i) {
+		if (CTRL_IS_RESERVED(msrs, i))
+			release_evntsel_nmi(MSR_K7_EVNTSEL0 + i);
+	}
+}
+
+#ifndef CONFIG_OPROFILE_IBS
+
+/* no IBS support */
+
+static int op_amd_init(struct oprofile_operations *ops)
+{
+	return 0;
+}
+
+static void op_amd_exit(void) {}
+
+#else
+
+static u8 ibs_eilvt_off;
+
+static inline void apic_init_ibs_nmi_per_cpu(void *arg)
+{
+	ibs_eilvt_off = setup_APIC_eilvt_ibs(0, APIC_EILVT_MSG_NMI, 0);
+}
+
+static inline void apic_clear_ibs_nmi_per_cpu(void *arg)
+{
+	setup_APIC_eilvt_ibs(0, APIC_EILVT_MSG_FIX, 1);
+}
+
+static int pfm_amd64_setup_eilvt(void)
+{
+#define IBSCTL_LVTOFFSETVAL		(1 << 8)
+#define IBSCTL				0x1cc
+	struct pci_dev *cpu_cfg;
+	int nodes;
+	u32 value = 0;
+
+	/* per CPU setup */
+	on_each_cpu(apic_init_ibs_nmi_per_cpu, NULL, 1);
+
+	nodes = 0;
+	cpu_cfg = NULL;
+	do {
+		cpu_cfg = pci_get_device(PCI_VENDOR_ID_AMD,
+					 PCI_DEVICE_ID_AMD_10H_NB_MISC,
+					 cpu_cfg);
+		if (!cpu_cfg)
+			break;
+		++nodes;
+		pci_write_config_dword(cpu_cfg, IBSCTL, ibs_eilvt_off
+				       | IBSCTL_LVTOFFSETVAL);
+		pci_read_config_dword(cpu_cfg, IBSCTL, &value);
+		if (value != (ibs_eilvt_off | IBSCTL_LVTOFFSETVAL)) {
+			printk(KERN_DEBUG "Failed to setup IBS LVT offset, "
+				"IBSCTL = 0x%08x", value);
+			return 1;
+		}
+	} while (1);
+
+	if (!nodes) {
+		printk(KERN_DEBUG "No CPU node configured for IBS");
+		return 1;
+	}
+
+#ifdef CONFIG_NUMA
+	/* Sanity check */
+	/* Works only for 64bit with proper numa implementation. */
+	if (nodes != num_possible_nodes()) {
+		printk(KERN_DEBUG "Failed to setup CPU node(s) for IBS, "
+			"found: %d, expected %d",
+			nodes, num_possible_nodes());
+		return 1;
+	}
+#endif
+	return 0;
+}
+
+/*
+ * initialize the APIC for the IBS interrupts
+ * if available (AMD Family10h rev B0 and later)
+ */
+static void setup_ibs(void)
+{
+	ibs_allowed = boot_cpu_has(X86_FEATURE_IBS);
+
+	if (!ibs_allowed)
+		return;
+
+	if (pfm_amd64_setup_eilvt()) {
+		ibs_allowed = 0;
+		return;
+	}
+
+	printk(KERN_INFO "oprofile: AMD IBS detected\n");
+}
+
+
+/*
+ * unitialize the APIC for the IBS interrupts if needed on AMD Family10h
+ * rev B0 and later */
+static void clear_ibs_nmi(void)
+{
+	if (ibs_allowed)
+		on_each_cpu(apic_clear_ibs_nmi_per_cpu, NULL, 1);
+}
+
+static int (*create_arch_files)(struct super_block * sb, struct dentry * root);
+
+static int setup_ibs_files(struct super_block * sb, struct dentry * root)
+{
+	char buf[12];
+	struct dentry *dir;
+	int ret = 0;
+
+	/* architecture specific files */
+	if (create_arch_files)
+		ret = create_arch_files(sb, root);
+
+	if (ret)
+		return ret;
+
+	if (!ibs_allowed)
+		return ret;
+
+	/* model specific files */
+
+	/* setup some reasonable defaults */
+	ibs_config.max_cnt_fetch = 250000;
+	ibs_config.fetch_enabled = 0;
+	ibs_config.max_cnt_op = 250000;
+	ibs_config.op_enabled = 0;
+	ibs_config.dispatched_ops = 1;
+	snprintf(buf,  sizeof(buf), "ibs_fetch");
+	dir = oprofilefs_mkdir(sb, root, buf);
+	oprofilefs_create_ulong(sb, dir, "rand_enable",
+				&ibs_config.rand_en);
+	oprofilefs_create_ulong(sb, dir, "enable",
+		&ibs_config.fetch_enabled);
+	oprofilefs_create_ulong(sb, dir, "max_count",
+		&ibs_config.max_cnt_fetch);
+	snprintf(buf,  sizeof(buf), "ibs_uops");
+	dir = oprofilefs_mkdir(sb, root, buf);
+	oprofilefs_create_ulong(sb, dir, "enable",
+		&ibs_config.op_enabled);
+	oprofilefs_create_ulong(sb, dir, "max_count",
+		&ibs_config.max_cnt_op);
+	oprofilefs_create_ulong(sb, dir, "dispatched_ops",
+		&ibs_config.dispatched_ops);
+
+	return 0;
+}
+
+static int op_amd_init(struct oprofile_operations *ops)
+{
+	setup_ibs();
+	create_arch_files = ops->create_files;
+	ops->create_files = setup_ibs_files;
+	return 0;
+}
+
+static void op_amd_exit(void)
+{
+	clear_ibs_nmi();
+}
+
+#endif
+
+struct op_x86_model_spec const op_amd_spec = {
+	.init = op_amd_init,
+	.exit = op_amd_exit,
+	.num_counters = NUM_COUNTERS,
+	.num_controls = NUM_CONTROLS,
+	.fill_in_addresses = &op_amd_fill_in_addresses,
+	.setup_ctrs = &op_amd_setup_ctrs,
+	.check_ctrs = &op_amd_check_ctrs,
+	.start = &op_amd_start,
+	.stop = &op_amd_stop,
+	.shutdown = &op_amd_shutdown
+};
diff --git a/arch/x86/oprofile/op_model_athlon.c b/arch/x86/oprofile/op_model_athlon.c
deleted file mode 100644
index 3d534879a9dc..000000000000
--- a/arch/x86/oprofile/op_model_athlon.c
+++ /dev/null
@@ -1,190 +0,0 @@
-/*
- * @file op_model_athlon.h
- * athlon / K7 / K8 / Family 10h model-specific MSR operations
- *
- * @remark Copyright 2002 OProfile authors
- * @remark Read the file COPYING
- *
- * @author John Levon
- * @author Philippe Elie
- * @author Graydon Hoare
- */
-
-#include <linux/oprofile.h>
-#include <asm/ptrace.h>
-#include <asm/msr.h>
-#include <asm/nmi.h>
-
-#include "op_x86_model.h"
-#include "op_counter.h"
-
-#define NUM_COUNTERS 4
-#define NUM_CONTROLS 4
-
-#define CTR_IS_RESERVED(msrs, c) (msrs->counters[(c)].addr ? 1 : 0)
-#define CTR_READ(l, h, msrs, c) do {rdmsr(msrs->counters[(c)].addr, (l), (h)); } while (0)
-#define CTR_WRITE(l, msrs, c) do {wrmsr(msrs->counters[(c)].addr, -(unsigned int)(l), -1); } while (0)
-#define CTR_OVERFLOWED(n) (!((n) & (1U<<31)))
-
-#define CTRL_IS_RESERVED(msrs, c) (msrs->controls[(c)].addr ? 1 : 0)
-#define CTRL_READ(l, h, msrs, c) do {rdmsr(msrs->controls[(c)].addr, (l), (h)); } while (0)
-#define CTRL_WRITE(l, h, msrs, c) do {wrmsr(msrs->controls[(c)].addr, (l), (h)); } while (0)
-#define CTRL_SET_ACTIVE(n) (n |= (1<<22))
-#define CTRL_SET_INACTIVE(n) (n &= ~(1<<22))
-#define CTRL_CLEAR_LO(x) (x &= (1<<21))
-#define CTRL_CLEAR_HI(x) (x &= 0xfffffcf0)
-#define CTRL_SET_ENABLE(val) (val |= 1<<20)
-#define CTRL_SET_USR(val, u) (val |= ((u & 1) << 16))
-#define CTRL_SET_KERN(val, k) (val |= ((k & 1) << 17))
-#define CTRL_SET_UM(val, m) (val |= (m << 8))
-#define CTRL_SET_EVENT_LOW(val, e) (val |= (e & 0xff))
-#define CTRL_SET_EVENT_HIGH(val, e) (val |= ((e >> 8) & 0xf))
-#define CTRL_SET_HOST_ONLY(val, h) (val |= ((h & 1) << 9))
-#define CTRL_SET_GUEST_ONLY(val, h) (val |= ((h & 1) << 8))
-
-static unsigned long reset_value[NUM_COUNTERS];
-
-static void athlon_fill_in_addresses(struct op_msrs * const msrs)
-{
-	int i;
-
-	for (i = 0; i < NUM_COUNTERS; i++) {
-		if (reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i))
-			msrs->counters[i].addr = MSR_K7_PERFCTR0 + i;
-		else
-			msrs->counters[i].addr = 0;
-	}
-
-	for (i = 0; i < NUM_CONTROLS; i++) {
-		if (reserve_evntsel_nmi(MSR_K7_EVNTSEL0 + i))
-			msrs->controls[i].addr = MSR_K7_EVNTSEL0 + i;
-		else
-			msrs->controls[i].addr = 0;
-	}
-}
-
-
-static void athlon_setup_ctrs(struct op_msrs const * const msrs)
-{
-	unsigned int low, high;
-	int i;
-
-	/* clear all counters */
-	for (i = 0 ; i < NUM_CONTROLS; ++i) {
-		if (unlikely(!CTRL_IS_RESERVED(msrs, i)))
-			continue;
-		CTRL_READ(low, high, msrs, i);
-		CTRL_CLEAR_LO(low);
-		CTRL_CLEAR_HI(high);
-		CTRL_WRITE(low, high, msrs, i);
-	}
-
-	/* avoid a false detection of ctr overflows in NMI handler */
-	for (i = 0; i < NUM_COUNTERS; ++i) {
-		if (unlikely(!CTR_IS_RESERVED(msrs, i)))
-			continue;
-		CTR_WRITE(1, msrs, i);
-	}
-
-	/* enable active counters */
-	for (i = 0; i < NUM_COUNTERS; ++i) {
-		if ((counter_config[i].enabled) && (CTR_IS_RESERVED(msrs, i))) {
-			reset_value[i] = counter_config[i].count;
-
-			CTR_WRITE(counter_config[i].count, msrs, i);
-
-			CTRL_READ(low, high, msrs, i);
-			CTRL_CLEAR_LO(low);
-			CTRL_CLEAR_HI(high);
-			CTRL_SET_ENABLE(low);
-			CTRL_SET_USR(low, counter_config[i].user);
-			CTRL_SET_KERN(low, counter_config[i].kernel);
-			CTRL_SET_UM(low, counter_config[i].unit_mask);
-			CTRL_SET_EVENT_LOW(low, counter_config[i].event);
-			CTRL_SET_EVENT_HIGH(high, counter_config[i].event);
-			CTRL_SET_HOST_ONLY(high, 0);
-			CTRL_SET_GUEST_ONLY(high, 0);
-
-			CTRL_WRITE(low, high, msrs, i);
-		} else {
-			reset_value[i] = 0;
-		}
-	}
-}
-
-
-static int athlon_check_ctrs(struct pt_regs * const regs,
-			     struct op_msrs const * const msrs)
-{
-	unsigned int low, high;
-	int i;
-
-	for (i = 0 ; i < NUM_COUNTERS; ++i) {
-		if (!reset_value[i])
-			continue;
-		CTR_READ(low, high, msrs, i);
-		if (CTR_OVERFLOWED(low)) {
-			oprofile_add_sample(regs, i);
-			CTR_WRITE(reset_value[i], msrs, i);
-		}
-	}
-
-	/* See op_model_ppro.c */
-	return 1;
-}
-
-
-static void athlon_start(struct op_msrs const * const msrs)
-{
-	unsigned int low, high;
-	int i;
-	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
-		if (reset_value[i]) {
-			CTRL_READ(low, high, msrs, i);
-			CTRL_SET_ACTIVE(low);
-			CTRL_WRITE(low, high, msrs, i);
-		}
-	}
-}
-
-
-static void athlon_stop(struct op_msrs const * const msrs)
-{
-	unsigned int low, high;
-	int i;
-
-	/* Subtle: stop on all counters to avoid race with
-	 * setting our pm callback */
-	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
-		if (!reset_value[i])
-			continue;
-		CTRL_READ(low, high, msrs, i);
-		CTRL_SET_INACTIVE(low);
-		CTRL_WRITE(low, high, msrs, i);
-	}
-}
-
-static void athlon_shutdown(struct op_msrs const * const msrs)
-{
-	int i;
-
-	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
-		if (CTR_IS_RESERVED(msrs, i))
-			release_perfctr_nmi(MSR_K7_PERFCTR0 + i);
-	}
-	for (i = 0 ; i < NUM_CONTROLS ; ++i) {
-		if (CTRL_IS_RESERVED(msrs, i))
-			release_evntsel_nmi(MSR_K7_EVNTSEL0 + i);
-	}
-}
-
-struct op_x86_model_spec const op_athlon_spec = {
-	.num_counters = NUM_COUNTERS,
-	.num_controls = NUM_CONTROLS,
-	.fill_in_addresses = &athlon_fill_in_addresses,
-	.setup_ctrs = &athlon_setup_ctrs,
-	.check_ctrs = &athlon_check_ctrs,
-	.start = &athlon_start,
-	.stop = &athlon_stop,
-	.shutdown = &athlon_shutdown
-};
diff --git a/arch/x86/oprofile/op_x86_model.h b/arch/x86/oprofile/op_x86_model.h
index 45b605fa71d0..05a0261ba0c3 100644
--- a/arch/x86/oprofile/op_x86_model.h
+++ b/arch/x86/oprofile/op_x86_model.h
@@ -32,6 +32,8 @@ struct pt_regs;
  * various x86 CPU models' perfctr support.
  */
 struct op_x86_model_spec {
+	int (*init)(struct oprofile_operations *ops);
+	void (*exit)(void);
 	unsigned int const num_counters;
 	unsigned int const num_controls;
 	void (*fill_in_addresses)(struct op_msrs * const msrs);
@@ -46,6 +48,6 @@ struct op_x86_model_spec {
 extern struct op_x86_model_spec const op_ppro_spec;
 extern struct op_x86_model_spec const op_p4_spec;
 extern struct op_x86_model_spec const op_p4_ht2_spec;
-extern struct op_x86_model_spec const op_athlon_spec;
+extern struct op_x86_model_spec const op_amd_spec;
 
 #endif /* OP_X86_MODEL_H */
diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c
index 19af06927fbc..1d88d2b39771 100644
--- a/arch/x86/pci/acpi.c
+++ b/arch/x86/pci/acpi.c
@@ -250,10 +250,5 @@ int __init pci_acpi_init(void)
 			acpi_pci_irq_enable(dev);
 	}
 
-#ifdef CONFIG_X86_IO_APIC
-	if (acpi_ioapic)
-		print_IO_APIC();
-#endif
-
 	return 0;
 }
diff --git a/arch/x86/pci/fixup.c b/arch/x86/pci/fixup.c
index 4bdaa590375d..3c27a809393b 100644
--- a/arch/x86/pci/fixup.c
+++ b/arch/x86/pci/fixup.c
@@ -511,3 +511,31 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1201, fam10h_pci_cfg_space_size);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1202, fam10h_pci_cfg_space_size);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1203, fam10h_pci_cfg_space_size);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1204, fam10h_pci_cfg_space_size);
+
+/*
+ * SB600: Disable BAR1 on device 14.0 to avoid HPET resources from
+ * confusing the PCI engine:
+ */
+static void sb600_disable_hpet_bar(struct pci_dev *dev)
+{
+	u8 val;
+
+	/*
+	 * The SB600 and SB700 both share the same device
+	 * ID, but the PM register 0x55 does something different
+	 * for the SB700, so make sure we are dealing with the
+	 * SB600 before touching the bit:
+	 */
+
+	pci_read_config_byte(dev, 0x08, &val);
+
+	if (val < 0x2F) {
+		outb(0x55, 0xCD6);
+		val = inb(0xCD7);
+
+		/* Set bit 7 in PM register 0x55 */
+		outb(0x55, 0xCD6);
+		outb(val | 0x80, 0xCD7);
+	}
+}
+DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_ATI, 0x4385, sb600_disable_hpet_bar);
diff --git a/arch/x86/pci/i386.c b/arch/x86/pci/i386.c
index 8791fc55e715..844df0cbbd3e 100644
--- a/arch/x86/pci/i386.c
+++ b/arch/x86/pci/i386.c
@@ -33,6 +33,7 @@
 #include <linux/bootmem.h>
 
 #include <asm/pat.h>
+#include <asm/e820.h>
 
 #include "pci.h"
 
@@ -227,6 +228,8 @@ void __init pcibios_resource_survey(void)
 	pcibios_allocate_bus_resources(&pci_root_buses);
 	pcibios_allocate_resources(0);
 	pcibios_allocate_resources(1);
+
+	e820_reserve_resources_late();
 }
 
 /**
diff --git a/arch/x86/pci/mmconfig-shared.c b/arch/x86/pci/mmconfig-shared.c
index d9635764ce3d..654a2234f8f3 100644
--- a/arch/x86/pci/mmconfig-shared.c
+++ b/arch/x86/pci/mmconfig-shared.c
@@ -209,7 +209,7 @@ static int __init pci_mmcfg_check_hostbridge(void)
 	return name != NULL;
 }
 
-static void __init pci_mmcfg_insert_resources(unsigned long resource_flags)
+static void __init pci_mmcfg_insert_resources(void)
 {
 #define PCI_MMCFG_RESOURCE_NAME_LEN 19
 	int i;
@@ -233,7 +233,7 @@ static void __init pci_mmcfg_insert_resources(unsigned long resource_flags)
 			 cfg->pci_segment);
 		res->start = cfg->address;
 		res->end = res->start + (num_buses << 20) - 1;
-		res->flags = IORESOURCE_MEM | resource_flags;
+		res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
 		insert_resource(&iomem_resource, res);
 		names += PCI_MMCFG_RESOURCE_NAME_LEN;
 	}
@@ -434,11 +434,9 @@ static void __init __pci_mmcfg_init(int early)
 	    (pci_mmcfg_config[0].address == 0))
 		return;
 
-	if (pci_mmcfg_arch_init()) {
-		if (known_bridge)
-			pci_mmcfg_insert_resources(IORESOURCE_BUSY);
+	if (pci_mmcfg_arch_init())
 		pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF;
-	} else {
+	else {
 		/*
 		 * Signal not to attempt to insert mmcfg resources because
 		 * the architecture mmcfg setup could not initialize.
@@ -475,7 +473,7 @@ static int __init pci_mmcfg_late_insert_resources(void)
 	 * marked so it won't cause request errors when __request_region is
 	 * called.
 	 */
-	pci_mmcfg_insert_resources(0);
+	pci_mmcfg_insert_resources();
 
 	return 0;
 }
diff --git a/arch/x86/power/cpu_32.c b/arch/x86/power/cpu_32.c
index d3e083dea720..274d06082f48 100644
--- a/arch/x86/power/cpu_32.c
+++ b/arch/x86/power/cpu_32.c
@@ -11,6 +11,7 @@
 #include <linux/suspend.h>
 #include <asm/mtrr.h>
 #include <asm/mce.h>
+#include <asm/xcr.h>
 
 static struct saved_context saved_context;
 
@@ -126,6 +127,12 @@ static void __restore_processor_state(struct saved_context *ctxt)
 	if (boot_cpu_has(X86_FEATURE_SEP))
 		enable_sep_cpu();
 
+	/*
+	 * restore XCR0 for xsave capable cpu's.
+	 */
+	if (cpu_has_xsave)
+		xsetbv(XCR_XFEATURE_ENABLED_MASK, pcntxt_mask);
+
 	fix_processor_context();
 	do_fpu_end();
 	mtrr_ap_init();
diff --git a/arch/x86/power/cpu_64.c b/arch/x86/power/cpu_64.c
index 66bdfb591fd8..e3b6cf70d62c 100644
--- a/arch/x86/power/cpu_64.c
+++ b/arch/x86/power/cpu_64.c
@@ -14,6 +14,7 @@
 #include <asm/page.h>
 #include <asm/pgtable.h>
 #include <asm/mtrr.h>
+#include <asm/xcr.h>
 
 static void fix_processor_context(void);
 
@@ -122,6 +123,12 @@ static void __restore_processor_state(struct saved_context *ctxt)
 	wrmsrl(MSR_GS_BASE, ctxt->gs_base);
 	wrmsrl(MSR_KERNEL_GS_BASE, ctxt->gs_kernel_base);
 
+	/*
+	 * restore XCR0 for xsave capable cpu's.
+	 */
+	if (cpu_has_xsave)
+		xsetbv(XCR_XFEATURE_ENABLED_MASK, pcntxt_mask);
+
 	fix_processor_context();
 
 	do_fpu_end();
diff --git a/arch/x86/xen/Kconfig b/arch/x86/xen/Kconfig
index 3815e425f470..87b9ab166423 100644
--- a/arch/x86/xen/Kconfig
+++ b/arch/x86/xen/Kconfig
@@ -26,5 +26,13 @@ config XEN_MAX_DOMAIN_MEMORY
 
 config XEN_SAVE_RESTORE
        bool
-       depends on PM
-       default y
\ No newline at end of file
+       depends on XEN && PM
+       default y
+
+config XEN_DEBUG_FS
+	bool "Enable Xen debug and tuning parameters in debugfs"
+	depends on XEN && DEBUG_FS
+	default n
+	help
+	  Enable statistics output and various tuning options in debugfs.
+	  Enabling this option may incur a significant performance overhead.
diff --git a/arch/x86/xen/Makefile b/arch/x86/xen/Makefile
index 59c1e539aed2..313947940a1a 100644
--- a/arch/x86/xen/Makefile
+++ b/arch/x86/xen/Makefile
@@ -1,4 +1,12 @@
-obj-y		:= enlighten.o setup.o multicalls.o mmu.o \
+ifdef CONFIG_FTRACE
+# Do not profile debug and lowlevel utilities
+CFLAGS_REMOVE_spinlock.o = -pg
+CFLAGS_REMOVE_time.o = -pg
+CFLAGS_REMOVE_irq.o = -pg
+endif
+
+obj-y		:= enlighten.o setup.o multicalls.o mmu.o irq.o \
 			time.o xen-asm_$(BITS).o grant-table.o suspend.o
 
-obj-$(CONFIG_SMP)	+= smp.o
+obj-$(CONFIG_SMP)		+= smp.o spinlock.o
+obj-$(CONFIG_XEN_DEBUG_FS)	+= debugfs.o
\ No newline at end of file
diff --git a/arch/x86/xen/debugfs.c b/arch/x86/xen/debugfs.c
new file mode 100644
index 000000000000..b53225d2cac3
--- /dev/null
+++ b/arch/x86/xen/debugfs.c
@@ -0,0 +1,123 @@
+#include <linux/init.h>
+#include <linux/debugfs.h>
+#include <linux/module.h>
+
+#include "debugfs.h"
+
+static struct dentry *d_xen_debug;
+
+struct dentry * __init xen_init_debugfs(void)
+{
+	if (!d_xen_debug) {
+		d_xen_debug = debugfs_create_dir("xen", NULL);
+
+		if (!d_xen_debug)
+			pr_warning("Could not create 'xen' debugfs directory\n");
+	}
+
+	return d_xen_debug;
+}
+
+struct array_data
+{
+	void *array;
+	unsigned elements;
+};
+
+static int u32_array_open(struct inode *inode, struct file *file)
+{
+	file->private_data = NULL;
+	return nonseekable_open(inode, file);
+}
+
+static size_t format_array(char *buf, size_t bufsize, const char *fmt,
+			   u32 *array, unsigned array_size)
+{
+	size_t ret = 0;
+	unsigned i;
+
+	for(i = 0; i < array_size; i++) {
+		size_t len;
+
+		len = snprintf(buf, bufsize, fmt, array[i]);
+		len++;	/* ' ' or '\n' */
+		ret += len;
+
+		if (buf) {
+			buf += len;
+			bufsize -= len;
+			buf[-1] = (i == array_size-1) ? '\n' : ' ';
+		}
+	}
+
+	ret++;		/* \0 */
+	if (buf)
+		*buf = '\0';
+
+	return ret;
+}
+
+static char *format_array_alloc(const char *fmt, u32 *array, unsigned array_size)
+{
+	size_t len = format_array(NULL, 0, fmt, array, array_size);
+	char *ret;
+
+	ret = kmalloc(len, GFP_KERNEL);
+	if (ret == NULL)
+		return NULL;
+
+	format_array(ret, len, fmt, array, array_size);
+	return ret;
+}
+
+static ssize_t u32_array_read(struct file *file, char __user *buf, size_t len,
+			      loff_t *ppos)
+{
+	struct inode *inode = file->f_path.dentry->d_inode;
+	struct array_data *data = inode->i_private;
+	size_t size;
+
+	if (*ppos == 0) {
+		if (file->private_data) {
+			kfree(file->private_data);
+			file->private_data = NULL;
+		}
+
+		file->private_data = format_array_alloc("%u", data->array, data->elements);
+	}
+
+	size = 0;
+	if (file->private_data)
+		size = strlen(file->private_data);
+
+	return simple_read_from_buffer(buf, len, ppos, file->private_data, size);
+}
+
+static int xen_array_release(struct inode *inode, struct file *file)
+{
+	kfree(file->private_data);
+
+	return 0;
+}
+
+static struct file_operations u32_array_fops = {
+	.owner	= THIS_MODULE,
+	.open	= u32_array_open,
+	.release= xen_array_release,
+	.read	= u32_array_read,
+};
+
+struct dentry *xen_debugfs_create_u32_array(const char *name, mode_t mode,
+					    struct dentry *parent,
+					    u32 *array, unsigned elements)
+{
+	struct array_data *data = kmalloc(sizeof(*data), GFP_KERNEL);
+
+	if (data == NULL)
+		return NULL;
+
+	data->array = array;
+	data->elements = elements;
+
+	return debugfs_create_file(name, mode, parent, data, &u32_array_fops);
+}
diff --git a/arch/x86/xen/debugfs.h b/arch/x86/xen/debugfs.h
new file mode 100644
index 000000000000..e28132084832
--- /dev/null
+++ b/arch/x86/xen/debugfs.h
@@ -0,0 +1,10 @@
+#ifndef _XEN_DEBUGFS_H
+#define _XEN_DEBUGFS_H
+
+struct dentry * __init xen_init_debugfs(void);
+
+struct dentry *xen_debugfs_create_u32_array(const char *name, mode_t mode,
+					    struct dentry *parent,
+					    u32 *array, unsigned elements);
+
+#endif /* _XEN_DEBUGFS_H */
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index 7dcd321a0508..0013a729b41d 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -30,12 +30,12 @@
 #include <xen/interface/xen.h>
 #include <xen/interface/physdev.h>
 #include <xen/interface/vcpu.h>
-#include <xen/interface/sched.h>
 #include <xen/features.h>
 #include <xen/page.h>
 #include <xen/hvc-console.h>
 
 #include <asm/paravirt.h>
+#include <asm/apic.h>
 #include <asm/page.h>
 #include <asm/xen/hypercall.h>
 #include <asm/xen/hypervisor.h>
@@ -57,6 +57,9 @@ EXPORT_SYMBOL_GPL(hypercall_page);
 DEFINE_PER_CPU(struct vcpu_info *, xen_vcpu);
 DEFINE_PER_CPU(struct vcpu_info, xen_vcpu_info);
 
+enum xen_domain_type xen_domain_type = XEN_NATIVE;
+EXPORT_SYMBOL_GPL(xen_domain_type);
+
 /*
  * Identity map, in addition to plain kernel map.  This needs to be
  * large enough to allocate page table pages to allocate the rest.
@@ -110,7 +113,14 @@ struct shared_info *HYPERVISOR_shared_info = (void *)&xen_dummy_shared_info;
  *
  * 0: not available, 1: available
  */
-static int have_vcpu_info_placement = 1;
+static int have_vcpu_info_placement =
+#ifdef CONFIG_X86_32
+	1
+#else
+	0
+#endif
+	;
+
 
 static void xen_vcpu_setup(int cpu)
 {
@@ -226,103 +236,68 @@ static unsigned long xen_get_debugreg(int reg)
 	return HYPERVISOR_get_debugreg(reg);
 }
 
-static unsigned long xen_save_fl(void)
+static void xen_leave_lazy(void)
 {
-	struct vcpu_info *vcpu;
-	unsigned long flags;
-
-	vcpu = x86_read_percpu(xen_vcpu);
-
-	/* flag has opposite sense of mask */
-	flags = !vcpu->evtchn_upcall_mask;
-
-	/* convert to IF type flag
-	   -0 -> 0x00000000
-	   -1 -> 0xffffffff
-	*/
-	return (-flags) & X86_EFLAGS_IF;
+	paravirt_leave_lazy(paravirt_get_lazy_mode());
+	xen_mc_flush();
 }
 
-static void xen_restore_fl(unsigned long flags)
+static unsigned long xen_store_tr(void)
 {
-	struct vcpu_info *vcpu;
-
-	/* convert from IF type flag */
-	flags = !(flags & X86_EFLAGS_IF);
-
-	/* There's a one instruction preempt window here.  We need to
-	   make sure we're don't switch CPUs between getting the vcpu
-	   pointer and updating the mask. */
-	preempt_disable();
-	vcpu = x86_read_percpu(xen_vcpu);
-	vcpu->evtchn_upcall_mask = flags;
-	preempt_enable_no_resched();
-
-	/* Doesn't matter if we get preempted here, because any
-	   pending event will get dealt with anyway. */
-
-	if (flags == 0) {
-		preempt_check_resched();
-		barrier(); /* unmask then check (avoid races) */
-		if (unlikely(vcpu->evtchn_upcall_pending))
-			force_evtchn_callback();
-	}
+	return 0;
 }
 
-static void xen_irq_disable(void)
+/*
+ * Set the page permissions for a particular virtual address.  If the
+ * address is a vmalloc mapping (or other non-linear mapping), then
+ * find the linear mapping of the page and also set its protections to
+ * match.
+ */
+static void set_aliased_prot(void *v, pgprot_t prot)
 {
-	/* There's a one instruction preempt window here.  We need to
-	   make sure we're don't switch CPUs between getting the vcpu
-	   pointer and updating the mask. */
-	preempt_disable();
-	x86_read_percpu(xen_vcpu)->evtchn_upcall_mask = 1;
-	preempt_enable_no_resched();
-}
+	int level;
+	pte_t *ptep;
+	pte_t pte;
+	unsigned long pfn;
+	struct page *page;
 
-static void xen_irq_enable(void)
-{
-	struct vcpu_info *vcpu;
+	ptep = lookup_address((unsigned long)v, &level);
+	BUG_ON(ptep == NULL);
 
-	/* We don't need to worry about being preempted here, since
-	   either a) interrupts are disabled, so no preemption, or b)
-	   the caller is confused and is trying to re-enable interrupts
-	   on an indeterminate processor. */
+	pfn = pte_pfn(*ptep);
+	page = pfn_to_page(pfn);
 
-	vcpu = x86_read_percpu(xen_vcpu);
-	vcpu->evtchn_upcall_mask = 0;
+	pte = pfn_pte(pfn, prot);
 
-	/* Doesn't matter if we get preempted here, because any
-	   pending event will get dealt with anyway. */
+	if (HYPERVISOR_update_va_mapping((unsigned long)v, pte, 0))
+		BUG();
 
-	barrier(); /* unmask then check (avoid races) */
-	if (unlikely(vcpu->evtchn_upcall_pending))
-		force_evtchn_callback();
-}
+	if (!PageHighMem(page)) {
+		void *av = __va(PFN_PHYS(pfn));
 
-static void xen_safe_halt(void)
-{
-	/* Blocking includes an implicit local_irq_enable(). */
-	if (HYPERVISOR_sched_op(SCHEDOP_block, NULL) != 0)
-		BUG();
+		if (av != v)
+			if (HYPERVISOR_update_va_mapping((unsigned long)av, pte, 0))
+				BUG();
+	} else
+		kmap_flush_unused();
 }
 
-static void xen_halt(void)
+static void xen_alloc_ldt(struct desc_struct *ldt, unsigned entries)
 {
-	if (irqs_disabled())
-		HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL);
-	else
-		xen_safe_halt();
-}
+	const unsigned entries_per_page = PAGE_SIZE / LDT_ENTRY_SIZE;
+	int i;
 
-static void xen_leave_lazy(void)
-{
-	paravirt_leave_lazy(paravirt_get_lazy_mode());
-	xen_mc_flush();
+	for(i = 0; i < entries; i += entries_per_page)
+		set_aliased_prot(ldt + i, PAGE_KERNEL_RO);
 }
 
-static unsigned long xen_store_tr(void)
+static void xen_free_ldt(struct desc_struct *ldt, unsigned entries)
 {
-	return 0;
+	const unsigned entries_per_page = PAGE_SIZE / LDT_ENTRY_SIZE;
+	int i;
+
+	for(i = 0; i < entries; i += entries_per_page)
+		set_aliased_prot(ldt + i, PAGE_KERNEL);
 }
 
 static void xen_set_ldt(const void *addr, unsigned entries)
@@ -425,8 +400,7 @@ static void xen_load_gs_index(unsigned int idx)
 static void xen_write_ldt_entry(struct desc_struct *dt, int entrynum,
 				const void *ptr)
 {
-	unsigned long lp = (unsigned long)&dt[entrynum];
-	xmaddr_t mach_lp = virt_to_machine(lp);
+	xmaddr_t mach_lp = arbitrary_virt_to_machine(&dt[entrynum]);
 	u64 entry = *(u64 *)ptr;
 
 	preempt_disable();
@@ -559,7 +533,7 @@ static void xen_write_gdt_entry(struct desc_struct *dt, int entry,
 }
 
 static void xen_load_sp0(struct tss_struct *tss,
-			  struct thread_struct *thread)
+			 struct thread_struct *thread)
 {
 	struct multicall_space mcs = xen_mc_entry(0);
 	MULTI_stack_switch(mcs.mc, __KERNEL_DS, thread->sp0);
@@ -580,16 +554,47 @@ static void xen_io_delay(void)
 }
 
 #ifdef CONFIG_X86_LOCAL_APIC
-static u32 xen_apic_read(unsigned long reg)
+static u32 xen_apic_read(u32 reg)
 {
 	return 0;
 }
 
-static void xen_apic_write(unsigned long reg, u32 val)
+static void xen_apic_write(u32 reg, u32 val)
 {
 	/* Warn to see if there's any stray references */
 	WARN_ON(1);
 }
+
+static u64 xen_apic_icr_read(void)
+{
+	return 0;
+}
+
+static void xen_apic_icr_write(u32 low, u32 id)
+{
+	/* Warn to see if there's any stray references */
+	WARN_ON(1);
+}
+
+static void xen_apic_wait_icr_idle(void)
+{
+        return;
+}
+
+static u32 xen_safe_apic_wait_icr_idle(void)
+{
+        return 0;
+}
+
+static struct apic_ops xen_basic_apic_ops = {
+	.read = xen_apic_read,
+	.write = xen_apic_write,
+	.icr_read = xen_apic_icr_read,
+	.icr_write = xen_apic_icr_write,
+	.wait_icr_idle = xen_apic_wait_icr_idle,
+	.safe_wait_icr_idle = xen_safe_apic_wait_icr_idle,
+};
+
 #endif
 
 static void xen_flush_tlb(void)
@@ -803,6 +808,19 @@ static int xen_write_msr_safe(unsigned int msr, unsigned low, unsigned high)
 			ret = -EFAULT;
 		break;
 #endif
+
+	case MSR_STAR:
+	case MSR_CSTAR:
+	case MSR_LSTAR:
+	case MSR_SYSCALL_MASK:
+	case MSR_IA32_SYSENTER_CS:
+	case MSR_IA32_SYSENTER_ESP:
+	case MSR_IA32_SYSENTER_EIP:
+		/* Fast syscall setup is all done in hypercalls, so
+		   these are all ignored.  Stub them out here to stop
+		   Xen console noise. */
+		break;
+
 	default:
 		ret = native_write_msr_safe(msr, low, high);
 	}
@@ -846,8 +864,8 @@ static void xen_alloc_ptpage(struct mm_struct *mm, unsigned long pfn, unsigned l
 		SetPagePinned(page);
 
 		if (!PageHighMem(page)) {
-			make_lowmem_page_readonly(__va(PFN_PHYS(pfn)));
-			if (level == PT_PTE)
+			make_lowmem_page_readonly(__va(PFN_PHYS((unsigned long)pfn)));
+			if (level == PT_PTE && USE_SPLIT_PTLOCKS)
 				pin_pagetable_pfn(MMUEXT_PIN_L1_TABLE, pfn);
 		} else
 			/* make sure there are no stray mappings of
@@ -915,7 +933,7 @@ static void xen_release_ptpage(unsigned long pfn, unsigned level)
 
 	if (PagePinned(page)) {
 		if (!PageHighMem(page)) {
-			if (level == PT_PTE)
+			if (level == PT_PTE && USE_SPLIT_PTLOCKS)
 				pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, pfn);
 			make_lowmem_page_readwrite(__va(PFN_PHYS(pfn)));
 		}
@@ -962,6 +980,7 @@ static void *xen_kmap_atomic_pte(struct page *page, enum km_type type)
 }
 #endif
 
+#ifdef CONFIG_X86_32
 static __init pte_t mask_rw_pte(pte_t *ptep, pte_t pte)
 {
 	/* If there's an existing pte, then don't allow _PAGE_RW to be set */
@@ -980,6 +999,7 @@ static __init void xen_set_pte_init(pte_t *ptep, pte_t pte)
 
 	xen_set_pte(ptep, pte);
 }
+#endif
 
 static __init void xen_pagetable_setup_start(pgd_t *base)
 {
@@ -1046,7 +1066,6 @@ void xen_setup_vcpu_info_placement(void)
 
 	/* xen_vcpu_setup managed to place the vcpu_info within the
 	   percpu area for all cpus, so make use of it */
-#ifdef CONFIG_X86_32
 	if (have_vcpu_info_placement) {
 		printk(KERN_INFO "Xen: using vcpu_info placement\n");
 
@@ -1056,7 +1075,6 @@ void xen_setup_vcpu_info_placement(void)
 		pv_irq_ops.irq_enable = xen_irq_enable_direct;
 		pv_mmu_ops.read_cr2 = xen_read_cr2_direct;
 	}
-#endif
 }
 
 static unsigned xen_patch(u8 type, u16 clobbers, void *insnbuf,
@@ -1077,12 +1095,10 @@ static unsigned xen_patch(u8 type, u16 clobbers, void *insnbuf,
 	goto patch_site
 
 	switch (type) {
-#ifdef CONFIG_X86_32
 		SITE(pv_irq_ops, irq_enable);
 		SITE(pv_irq_ops, irq_disable);
 		SITE(pv_irq_ops, save_fl);
 		SITE(pv_irq_ops, restore_fl);
-#endif /* CONFIG_X86_32 */
 #undef SITE
 
 	patch_site:
@@ -1220,6 +1236,9 @@ static const struct pv_cpu_ops xen_cpu_ops __initdata = {
 	.load_gs_index = xen_load_gs_index,
 #endif
 
+	.alloc_ldt = xen_alloc_ldt,
+	.free_ldt = xen_free_ldt,
+
 	.store_gdt = native_store_gdt,
 	.store_idt = native_store_idt,
 	.store_tr = xen_store_tr,
@@ -1241,40 +1260,8 @@ static const struct pv_cpu_ops xen_cpu_ops __initdata = {
 	},
 };
 
-static void __init __xen_init_IRQ(void)
-{
-#ifdef CONFIG_X86_64
-	int i;
-
-	/* Create identity vector->irq map */
-	for(i = 0; i < NR_VECTORS; i++) {
-		int cpu;
-
-		for_each_possible_cpu(cpu)
-			per_cpu(vector_irq, cpu)[i] = i;
-	}
-#endif	/* CONFIG_X86_64 */
-
-	xen_init_IRQ();
-}
-
-static const struct pv_irq_ops xen_irq_ops __initdata = {
-	.init_IRQ = __xen_init_IRQ,
-	.save_fl = xen_save_fl,
-	.restore_fl = xen_restore_fl,
-	.irq_disable = xen_irq_disable,
-	.irq_enable = xen_irq_enable,
-	.safe_halt = xen_safe_halt,
-	.halt = xen_halt,
-#ifdef CONFIG_X86_64
-	.adjust_exception_frame = xen_adjust_exception_frame,
-#endif
-};
-
 static const struct pv_apic_ops xen_apic_ops __initdata = {
 #ifdef CONFIG_X86_LOCAL_APIC
-	.apic_write = xen_apic_write,
-	.apic_read = xen_apic_read,
 	.setup_boot_clock = paravirt_nop,
 	.setup_secondary_clock = paravirt_nop,
 	.startup_ipi_hook = paravirt_nop,
@@ -1413,7 +1400,7 @@ static void __init xen_reserve_top(void)
 	if (HYPERVISOR_xen_version(XENVER_platform_parameters, &pp) == 0)
 		top = pp.virt_start;
 
-	reserve_top_address(-top + 2 * PAGE_SIZE);
+	reserve_top_address(-top);
 #endif	/* CONFIG_X86_32 */
 }
 
@@ -1447,48 +1434,11 @@ static void *m2v(phys_addr_t maddr)
 	return __ka(m2p(maddr));
 }
 
-#ifdef CONFIG_X86_64
-static void walk(pgd_t *pgd, unsigned long addr)
-{
-	unsigned l4idx = pgd_index(addr);
-	unsigned l3idx = pud_index(addr);
-	unsigned l2idx = pmd_index(addr);
-	unsigned l1idx = pte_index(addr);
-	pgd_t l4;
-	pud_t l3;
-	pmd_t l2;
-	pte_t l1;
-
-	xen_raw_printk("walk %p, %lx -> %d %d %d %d\n",
-		       pgd, addr, l4idx, l3idx, l2idx, l1idx);
-
-	l4 = pgd[l4idx];
-	xen_raw_printk("  l4: %016lx\n", l4.pgd);
-	xen_raw_printk("      %016lx\n", pgd_val(l4));
-
-	l3 = ((pud_t *)(m2v(l4.pgd)))[l3idx];
-	xen_raw_printk("  l3: %016lx\n", l3.pud);
-	xen_raw_printk("      %016lx\n", pud_val(l3));
-
-	l2 = ((pmd_t *)(m2v(l3.pud)))[l2idx];
-	xen_raw_printk("  l2: %016lx\n", l2.pmd);
-	xen_raw_printk("      %016lx\n", pmd_val(l2));
-
-	l1 = ((pte_t *)(m2v(l2.pmd)))[l1idx];
-	xen_raw_printk("  l1: %016lx\n", l1.pte);
-	xen_raw_printk("      %016lx\n", pte_val(l1));
-}
-#endif
-
 static void set_page_prot(void *addr, pgprot_t prot)
 {
 	unsigned long pfn = __pa(addr) >> PAGE_SHIFT;
 	pte_t pte = pfn_pte(pfn, prot);
 
-	xen_raw_printk("addr=%p pfn=%lx mfn=%lx prot=%016llx pte=%016llx\n",
-		       addr, pfn, get_phys_to_machine(pfn),
-		       pgprot_val(prot), pte.pte);
-
 	if (HYPERVISOR_update_va_mapping((unsigned long)addr, pte, 0))
 		BUG();
 }
@@ -1664,6 +1614,8 @@ asmlinkage void __init xen_start_kernel(void)
 	if (!xen_start_info)
 		return;
 
+	xen_domain_type = XEN_PV_DOMAIN;
+
 	BUG_ON(memcmp(xen_start_info->magic, "xen-3", 5) != 0);
 
 	xen_setup_features();
@@ -1673,10 +1625,18 @@ asmlinkage void __init xen_start_kernel(void)
 	pv_init_ops = xen_init_ops;
 	pv_time_ops = xen_time_ops;
 	pv_cpu_ops = xen_cpu_ops;
-	pv_irq_ops = xen_irq_ops;
 	pv_apic_ops = xen_apic_ops;
 	pv_mmu_ops = xen_mmu_ops;
 
+	xen_init_irq_ops();
+
+#ifdef CONFIG_X86_LOCAL_APIC
+	/*
+	 * set up the basic apic ops.
+	 */
+	apic_ops = &xen_basic_apic_ops;
+#endif
+
 	if (xen_feature(XENFEAT_mmu_pt_update_preserve_ad)) {
 		pv_mmu_ops.ptep_modify_prot_start = xen_ptep_modify_prot_start;
 		pv_mmu_ops.ptep_modify_prot_commit = xen_ptep_modify_prot_commit;
@@ -1700,7 +1660,7 @@ asmlinkage void __init xen_start_kernel(void)
 
 	/* Prevent unwanted bits from being set in PTEs. */
 	__supported_pte_mask &= ~_PAGE_GLOBAL;
-	if (!is_initial_xendomain())
+	if (!xen_initial_domain())
 		__supported_pte_mask &= ~(_PAGE_PWT | _PAGE_PCD);
 
 	/* Don't do the full vcpu_info placement stuff until we have a
@@ -1735,7 +1695,7 @@ asmlinkage void __init xen_start_kernel(void)
 	boot_params.hdr.ramdisk_size = xen_start_info->mod_len;
 	boot_params.hdr.cmd_line_ptr = __pa(xen_start_info->cmd_line);
 
-	if (!is_initial_xendomain()) {
+	if (!xen_initial_domain()) {
 		add_preferred_console("xenboot", 0, NULL);
 		add_preferred_console("tty", 0, NULL);
 		add_preferred_console("hvc", 0, NULL);
@@ -1743,15 +1703,6 @@ asmlinkage void __init xen_start_kernel(void)
 
 	xen_raw_console_write("about to get started...\n");
 
-#if 0
-	xen_raw_printk("&boot_params=%p __pa(&boot_params)=%lx __va(__pa(&boot_params))=%lx\n",
-		       &boot_params, __pa_symbol(&boot_params),
-		       __va(__pa_symbol(&boot_params)));
-
-	walk(pgd, &boot_params);
-	walk(pgd, __va(__pa(&boot_params)));
-#endif
-
 	/* Start the world */
 #ifdef CONFIG_X86_32
 	i386_start_kernel();
diff --git a/arch/x86/xen/irq.c b/arch/x86/xen/irq.c
new file mode 100644
index 000000000000..28b85ab8422e
--- /dev/null
+++ b/arch/x86/xen/irq.c
@@ -0,0 +1,143 @@
+#include <linux/hardirq.h>
+
+#include <xen/interface/xen.h>
+#include <xen/interface/sched.h>
+#include <xen/interface/vcpu.h>
+
+#include <asm/xen/hypercall.h>
+#include <asm/xen/hypervisor.h>
+
+#include "xen-ops.h"
+
+/*
+ * Force a proper event-channel callback from Xen after clearing the
+ * callback mask. We do this in a very simple manner, by making a call
+ * down into Xen. The pending flag will be checked by Xen on return.
+ */
+void xen_force_evtchn_callback(void)
+{
+	(void)HYPERVISOR_xen_version(0, NULL);
+}
+
+static void __init __xen_init_IRQ(void)
+{
+#ifdef CONFIG_X86_64
+	int i;
+
+	/* Create identity vector->irq map */
+	for(i = 0; i < NR_VECTORS; i++) {
+		int cpu;
+
+		for_each_possible_cpu(cpu)
+			per_cpu(vector_irq, cpu)[i] = i;
+	}
+#endif	/* CONFIG_X86_64 */
+
+	xen_init_IRQ();
+}
+
+static unsigned long xen_save_fl(void)
+{
+	struct vcpu_info *vcpu;
+	unsigned long flags;
+
+	vcpu = x86_read_percpu(xen_vcpu);
+
+	/* flag has opposite sense of mask */
+	flags = !vcpu->evtchn_upcall_mask;
+
+	/* convert to IF type flag
+	   -0 -> 0x00000000
+	   -1 -> 0xffffffff
+	*/
+	return (-flags) & X86_EFLAGS_IF;
+}
+
+static void xen_restore_fl(unsigned long flags)
+{
+	struct vcpu_info *vcpu;
+
+	/* convert from IF type flag */
+	flags = !(flags & X86_EFLAGS_IF);
+
+	/* There's a one instruction preempt window here.  We need to
+	   make sure we're don't switch CPUs between getting the vcpu
+	   pointer and updating the mask. */
+	preempt_disable();
+	vcpu = x86_read_percpu(xen_vcpu);
+	vcpu->evtchn_upcall_mask = flags;
+	preempt_enable_no_resched();
+
+	/* Doesn't matter if we get preempted here, because any
+	   pending event will get dealt with anyway. */
+
+	if (flags == 0) {
+		preempt_check_resched();
+		barrier(); /* unmask then check (avoid races) */
+		if (unlikely(vcpu->evtchn_upcall_pending))
+			xen_force_evtchn_callback();
+	}
+}
+
+static void xen_irq_disable(void)
+{
+	/* There's a one instruction preempt window here.  We need to
+	   make sure we're don't switch CPUs between getting the vcpu
+	   pointer and updating the mask. */
+	preempt_disable();
+	x86_read_percpu(xen_vcpu)->evtchn_upcall_mask = 1;
+	preempt_enable_no_resched();
+}
+
+static void xen_irq_enable(void)
+{
+	struct vcpu_info *vcpu;
+
+	/* We don't need to worry about being preempted here, since
+	   either a) interrupts are disabled, so no preemption, or b)
+	   the caller is confused and is trying to re-enable interrupts
+	   on an indeterminate processor. */
+
+	vcpu = x86_read_percpu(xen_vcpu);
+	vcpu->evtchn_upcall_mask = 0;
+
+	/* Doesn't matter if we get preempted here, because any
+	   pending event will get dealt with anyway. */
+
+	barrier(); /* unmask then check (avoid races) */
+	if (unlikely(vcpu->evtchn_upcall_pending))
+		xen_force_evtchn_callback();
+}
+
+static void xen_safe_halt(void)
+{
+	/* Blocking includes an implicit local_irq_enable(). */
+	if (HYPERVISOR_sched_op(SCHEDOP_block, NULL) != 0)
+		BUG();
+}
+
+static void xen_halt(void)
+{
+	if (irqs_disabled())
+		HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL);
+	else
+		xen_safe_halt();
+}
+
+static const struct pv_irq_ops xen_irq_ops __initdata = {
+	.init_IRQ = __xen_init_IRQ,
+	.save_fl = xen_save_fl,
+	.restore_fl = xen_restore_fl,
+	.irq_disable = xen_irq_disable,
+	.irq_enable = xen_irq_enable,
+	.safe_halt = xen_safe_halt,
+	.halt = xen_halt,
+#ifdef CONFIG_X86_64
+	.adjust_exception_frame = xen_adjust_exception_frame,
+#endif
+};
+
+void __init xen_init_irq_ops()
+{
+	pv_irq_ops = xen_irq_ops;
+}
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index aa37469da696..ae173f6edd8b 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -40,6 +40,7 @@
  */
 #include <linux/sched.h>
 #include <linux/highmem.h>
+#include <linux/debugfs.h>
 #include <linux/bug.h>
 
 #include <asm/pgtable.h>
@@ -57,6 +58,61 @@
 
 #include "multicalls.h"
 #include "mmu.h"
+#include "debugfs.h"
+
+#define MMU_UPDATE_HISTO	30
+
+#ifdef CONFIG_XEN_DEBUG_FS
+
+static struct {
+	u32 pgd_update;
+	u32 pgd_update_pinned;
+	u32 pgd_update_batched;
+
+	u32 pud_update;
+	u32 pud_update_pinned;
+	u32 pud_update_batched;
+
+	u32 pmd_update;
+	u32 pmd_update_pinned;
+	u32 pmd_update_batched;
+
+	u32 pte_update;
+	u32 pte_update_pinned;
+	u32 pte_update_batched;
+
+	u32 mmu_update;
+	u32 mmu_update_extended;
+	u32 mmu_update_histo[MMU_UPDATE_HISTO];
+
+	u32 prot_commit;
+	u32 prot_commit_batched;
+
+	u32 set_pte_at;
+	u32 set_pte_at_batched;
+	u32 set_pte_at_pinned;
+	u32 set_pte_at_current;
+	u32 set_pte_at_kernel;
+} mmu_stats;
+
+static u8 zero_stats;
+
+static inline void check_zero(void)
+{
+	if (unlikely(zero_stats)) {
+		memset(&mmu_stats, 0, sizeof(mmu_stats));
+		zero_stats = 0;
+	}
+}
+
+#define ADD_STATS(elem, val)			\
+	do { check_zero(); mmu_stats.elem += (val); } while(0)
+
+#else  /* !CONFIG_XEN_DEBUG_FS */
+
+#define ADD_STATS(elem, val)	do { (void)(val); } while(0)
+
+#endif /* CONFIG_XEN_DEBUG_FS */
 
 /*
  * Just beyond the highest usermode address.  STACK_TOP_MAX has a
@@ -229,25 +285,35 @@ void make_lowmem_page_readwrite(void *vaddr)
 }
 
 
-static bool page_pinned(void *ptr)
+static bool xen_page_pinned(void *ptr)
 {
 	struct page *page = virt_to_page(ptr);
 
 	return PagePinned(page);
 }
 
-static void extend_mmu_update(const struct mmu_update *update)
+static void xen_extend_mmu_update(const struct mmu_update *update)
 {
 	struct multicall_space mcs;
 	struct mmu_update *u;
 
 	mcs = xen_mc_extend_args(__HYPERVISOR_mmu_update, sizeof(*u));
 
-	if (mcs.mc != NULL)
+	if (mcs.mc != NULL) {
+		ADD_STATS(mmu_update_extended, 1);
+		ADD_STATS(mmu_update_histo[mcs.mc->args[1]], -1);
+
 		mcs.mc->args[1]++;
-	else {
+
+		if (mcs.mc->args[1] < MMU_UPDATE_HISTO)
+			ADD_STATS(mmu_update_histo[mcs.mc->args[1]], 1);
+		else
+			ADD_STATS(mmu_update_histo[0], 1);
+	} else {
+		ADD_STATS(mmu_update, 1);
 		mcs = __xen_mc_entry(sizeof(*u));
 		MULTI_mmu_update(mcs.mc, mcs.args, 1, NULL, DOMID_SELF);
+		ADD_STATS(mmu_update_histo[1], 1);
 	}
 
 	u = mcs.args;
@@ -265,7 +331,9 @@ void xen_set_pmd_hyper(pmd_t *ptr, pmd_t val)
 	/* ptr may be ioremapped for 64-bit pagetable setup */
 	u.ptr = arbitrary_virt_to_machine(ptr).maddr;
 	u.val = pmd_val_ma(val);
-	extend_mmu_update(&u);
+	xen_extend_mmu_update(&u);
+
+	ADD_STATS(pmd_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU);
 
 	xen_mc_issue(PARAVIRT_LAZY_MMU);
 
@@ -274,13 +342,17 @@ void xen_set_pmd_hyper(pmd_t *ptr, pmd_t val)
 
 void xen_set_pmd(pmd_t *ptr, pmd_t val)
 {
+	ADD_STATS(pmd_update, 1);
+
 	/* If page is not pinned, we can just update the entry
 	   directly */
-	if (!page_pinned(ptr)) {
+	if (!xen_page_pinned(ptr)) {
 		*ptr = val;
 		return;
 	}
 
+	ADD_STATS(pmd_update_pinned, 1);
+
 	xen_set_pmd_hyper(ptr, val);
 }
 
@@ -300,12 +372,18 @@ void xen_set_pte_at(struct mm_struct *mm, unsigned long addr,
 	if (mm == &init_mm)
 		preempt_disable();
 
+	ADD_STATS(set_pte_at, 1);
+//	ADD_STATS(set_pte_at_pinned, xen_page_pinned(ptep));
+	ADD_STATS(set_pte_at_current, mm == current->mm);
+	ADD_STATS(set_pte_at_kernel, mm == &init_mm);
+
 	if (mm == current->mm || mm == &init_mm) {
 		if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU) {
 			struct multicall_space mcs;
 			mcs = xen_mc_entry(0);
 
 			MULTI_update_va_mapping(mcs.mc, addr, pteval, 0);
+			ADD_STATS(set_pte_at_batched, 1);
 			xen_mc_issue(PARAVIRT_LAZY_MMU);
 			goto out;
 		} else
@@ -334,7 +412,10 @@ void xen_ptep_modify_prot_commit(struct mm_struct *mm, unsigned long addr,
 
 	u.ptr = virt_to_machine(ptep).maddr | MMU_PT_UPDATE_PRESERVE_AD;
 	u.val = pte_val_ma(pte);
-	extend_mmu_update(&u);
+	xen_extend_mmu_update(&u);
+
+	ADD_STATS(prot_commit, 1);
+	ADD_STATS(prot_commit_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU);
 
 	xen_mc_issue(PARAVIRT_LAZY_MMU);
 }
@@ -400,7 +481,9 @@ void xen_set_pud_hyper(pud_t *ptr, pud_t val)
 	/* ptr may be ioremapped for 64-bit pagetable setup */
 	u.ptr = arbitrary_virt_to_machine(ptr).maddr;
 	u.val = pud_val_ma(val);
-	extend_mmu_update(&u);
+	xen_extend_mmu_update(&u);
+
+	ADD_STATS(pud_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU);
 
 	xen_mc_issue(PARAVIRT_LAZY_MMU);
 
@@ -409,18 +492,26 @@ void xen_set_pud_hyper(pud_t *ptr, pud_t val)
 
 void xen_set_pud(pud_t *ptr, pud_t val)
 {
+	ADD_STATS(pud_update, 1);
+
 	/* If page is not pinned, we can just update the entry
 	   directly */
-	if (!page_pinned(ptr)) {
+	if (!xen_page_pinned(ptr)) {
 		*ptr = val;
 		return;
 	}
 
+	ADD_STATS(pud_update_pinned, 1);
+
 	xen_set_pud_hyper(ptr, val);
 }
 
 void xen_set_pte(pte_t *ptep, pte_t pte)
 {
+	ADD_STATS(pte_update, 1);
+//	ADD_STATS(pte_update_pinned, xen_page_pinned(ptep));
+	ADD_STATS(pte_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU);
+
 #ifdef CONFIG_X86_PAE
 	ptep->pte_high = pte.pte_high;
 	smp_wmb();
@@ -490,7 +581,7 @@ static void __xen_set_pgd_hyper(pgd_t *ptr, pgd_t val)
 
 	u.ptr = virt_to_machine(ptr).maddr;
 	u.val = pgd_val_ma(val);
-	extend_mmu_update(&u);
+	xen_extend_mmu_update(&u);
 }
 
 /*
@@ -517,17 +608,22 @@ void xen_set_pgd(pgd_t *ptr, pgd_t val)
 {
 	pgd_t *user_ptr = xen_get_user_pgd(ptr);
 
+	ADD_STATS(pgd_update, 1);
+
 	/* If page is not pinned, we can just update the entry
 	   directly */
-	if (!page_pinned(ptr)) {
+	if (!xen_page_pinned(ptr)) {
 		*ptr = val;
 		if (user_ptr) {
-			WARN_ON(page_pinned(user_ptr));
+			WARN_ON(xen_page_pinned(user_ptr));
 			*user_ptr = val;
 		}
 		return;
 	}
 
+	ADD_STATS(pgd_update_pinned, 1);
+	ADD_STATS(pgd_update_batched, paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU);
+
 	/* If it's pinned, then we can at least batch the kernel and
 	   user updates together. */
 	xen_mc_batch();
@@ -555,9 +651,12 @@ void xen_set_pgd(pgd_t *ptr, pgd_t val)
  * For 64-bit, we must skip the Xen hole in the middle of the address
  * space, just after the big x86-64 virtual hole.
  */
-static int pgd_walk(pgd_t *pgd, int (*func)(struct page *, enum pt_level),
-		    unsigned long limit)
+static int xen_pgd_walk(struct mm_struct *mm,
+			int (*func)(struct mm_struct *mm, struct page *,
+				    enum pt_level),
+			unsigned long limit)
 {
+	pgd_t *pgd = mm->pgd;
 	int flush = 0;
 	unsigned hole_low, hole_high;
 	unsigned pgdidx_limit, pudidx_limit, pmdidx_limit;
@@ -590,8 +689,6 @@ static int pgd_walk(pgd_t *pgd, int (*func)(struct page *, enum pt_level),
 	pmdidx_limit = 0;
 #endif
 
-	flush |= (*func)(virt_to_page(pgd), PT_PGD);
-
 	for (pgdidx = 0; pgdidx <= pgdidx_limit; pgdidx++) {
 		pud_t *pud;
 
@@ -604,7 +701,7 @@ static int pgd_walk(pgd_t *pgd, int (*func)(struct page *, enum pt_level),
 		pud = pud_offset(&pgd[pgdidx], 0);
 
 		if (PTRS_PER_PUD > 1) /* not folded */
-			flush |= (*func)(virt_to_page(pud), PT_PUD);
+			flush |= (*func)(mm, virt_to_page(pud), PT_PUD);
 
 		for (pudidx = 0; pudidx < PTRS_PER_PUD; pudidx++) {
 			pmd_t *pmd;
@@ -619,7 +716,7 @@ static int pgd_walk(pgd_t *pgd, int (*func)(struct page *, enum pt_level),
 			pmd = pmd_offset(&pud[pudidx], 0);
 
 			if (PTRS_PER_PMD > 1) /* not folded */
-				flush |= (*func)(virt_to_page(pmd), PT_PMD);
+				flush |= (*func)(mm, virt_to_page(pmd), PT_PMD);
 
 			for (pmdidx = 0; pmdidx < PTRS_PER_PMD; pmdidx++) {
 				struct page *pte;
@@ -633,28 +730,34 @@ static int pgd_walk(pgd_t *pgd, int (*func)(struct page *, enum pt_level),
 					continue;
 
 				pte = pmd_page(pmd[pmdidx]);
-				flush |= (*func)(pte, PT_PTE);
+				flush |= (*func)(mm, pte, PT_PTE);
 			}
 		}
 	}
+
 out:
+	/* Do the top level last, so that the callbacks can use it as
+	   a cue to do final things like tlb flushes. */
+	flush |= (*func)(mm, virt_to_page(pgd), PT_PGD);
 
 	return flush;
 }
 
-static spinlock_t *lock_pte(struct page *page)
+/* If we're using split pte locks, then take the page's lock and
+   return a pointer to it.  Otherwise return NULL. */
+static spinlock_t *xen_pte_lock(struct page *page, struct mm_struct *mm)
 {
 	spinlock_t *ptl = NULL;
 
-#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
+#if USE_SPLIT_PTLOCKS
 	ptl = __pte_lockptr(page);
-	spin_lock(ptl);
+	spin_lock_nest_lock(ptl, &mm->page_table_lock);
 #endif
 
 	return ptl;
 }
 
-static void do_unlock(void *v)
+static void xen_pte_unlock(void *v)
 {
 	spinlock_t *ptl = v;
 	spin_unlock(ptl);
@@ -672,7 +775,8 @@ static void xen_do_pin(unsigned level, unsigned long pfn)
 	MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
 }
 
-static int pin_page(struct page *page, enum pt_level level)
+static int xen_pin_page(struct mm_struct *mm, struct page *page,
+			enum pt_level level)
 {
 	unsigned pgfl = TestSetPagePinned(page);
 	int flush;
@@ -691,21 +795,40 @@ static int pin_page(struct page *page, enum pt_level level)
 
 		flush = 0;
 
+		/*
+		 * We need to hold the pagetable lock between the time
+		 * we make the pagetable RO and when we actually pin
+		 * it.  If we don't, then other users may come in and
+		 * attempt to update the pagetable by writing it,
+		 * which will fail because the memory is RO but not
+		 * pinned, so Xen won't do the trap'n'emulate.
+		 *
+		 * If we're using split pte locks, we can't hold the
+		 * entire pagetable's worth of locks during the
+		 * traverse, because we may wrap the preempt count (8
+		 * bits).  The solution is to mark RO and pin each PTE
+		 * page while holding the lock.  This means the number
+		 * of locks we end up holding is never more than a
+		 * batch size (~32 entries, at present).
+		 *
+		 * If we're not using split pte locks, we needn't pin
+		 * the PTE pages independently, because we're
+		 * protected by the overall pagetable lock.
+		 */
 		ptl = NULL;
 		if (level == PT_PTE)
-			ptl = lock_pte(page);
+			ptl = xen_pte_lock(page, mm);
 
 		MULTI_update_va_mapping(mcs.mc, (unsigned long)pt,
 					pfn_pte(pfn, PAGE_KERNEL_RO),
 					level == PT_PGD ? UVMF_TLB_FLUSH : 0);
 
-		if (level == PT_PTE)
+		if (ptl) {
 			xen_do_pin(MMUEXT_PIN_L1_TABLE, pfn);
 
-		if (ptl) {
 			/* Queue a deferred unlock for when this batch
 			   is completed. */
-			xen_mc_callback(do_unlock, ptl);
+			xen_mc_callback(xen_pte_unlock, ptl);
 		}
 	}
 
@@ -715,11 +838,11 @@ static int pin_page(struct page *page, enum pt_level level)
 /* This is called just after a mm has been created, but it has not
    been used yet.  We need to make sure that its pagetable is all
    read-only, and can be pinned. */
-void xen_pgd_pin(pgd_t *pgd)
+static void __xen_pgd_pin(struct mm_struct *mm, pgd_t *pgd)
 {
 	xen_mc_batch();
 
-	if (pgd_walk(pgd, pin_page, USER_LIMIT)) {
+	if (xen_pgd_walk(mm, xen_pin_page, USER_LIMIT)) {
 		/* re-enable interrupts for kmap_flush_unused */
 		xen_mc_issue(0);
 		kmap_flush_unused();
@@ -733,25 +856,35 @@ void xen_pgd_pin(pgd_t *pgd)
 		xen_do_pin(MMUEXT_PIN_L4_TABLE, PFN_DOWN(__pa(pgd)));
 
 		if (user_pgd) {
-			pin_page(virt_to_page(user_pgd), PT_PGD);
+			xen_pin_page(mm, virt_to_page(user_pgd), PT_PGD);
 			xen_do_pin(MMUEXT_PIN_L4_TABLE, PFN_DOWN(__pa(user_pgd)));
 		}
 	}
 #else /* CONFIG_X86_32 */
 #ifdef CONFIG_X86_PAE
 	/* Need to make sure unshared kernel PMD is pinnable */
-	pin_page(virt_to_page(pgd_page(pgd[pgd_index(TASK_SIZE)])), PT_PMD);
+	xen_pin_page(mm, virt_to_page(pgd_page(pgd[pgd_index(TASK_SIZE)])),
+		     PT_PMD);
 #endif
 	xen_do_pin(MMUEXT_PIN_L3_TABLE, PFN_DOWN(__pa(pgd)));
 #endif /* CONFIG_X86_64 */
 	xen_mc_issue(0);
 }
 
+static void xen_pgd_pin(struct mm_struct *mm)
+{
+	__xen_pgd_pin(mm, mm->pgd);
+}
+
 /*
  * On save, we need to pin all pagetables to make sure they get their
  * mfns turned into pfns.  Search the list for any unpinned pgds and pin
  * them (unpinned pgds are not currently in use, probably because the
  * process is under construction or destruction).
+ *
+ * Expected to be called in stop_machine() ("equivalent to taking
+ * every spinlock in the system"), so the locking doesn't really
+ * matter all that much.
  */
 void xen_mm_pin_all(void)
 {
@@ -762,7 +895,7 @@ void xen_mm_pin_all(void)
 
 	list_for_each_entry(page, &pgd_list, lru) {
 		if (!PagePinned(page)) {
-			xen_pgd_pin((pgd_t *)page_address(page));
+			__xen_pgd_pin(&init_mm, (pgd_t *)page_address(page));
 			SetPageSavePinned(page);
 		}
 	}
@@ -775,7 +908,8 @@ void xen_mm_pin_all(void)
  * that's before we have page structures to store the bits.  So do all
  * the book-keeping now.
  */
-static __init int mark_pinned(struct page *page, enum pt_level level)
+static __init int xen_mark_pinned(struct mm_struct *mm, struct page *page,
+				  enum pt_level level)
 {
 	SetPagePinned(page);
 	return 0;
@@ -783,10 +917,11 @@ static __init int mark_pinned(struct page *page, enum pt_level level)
 
 void __init xen_mark_init_mm_pinned(void)
 {
-	pgd_walk(init_mm.pgd, mark_pinned, FIXADDR_TOP);
+	xen_pgd_walk(&init_mm, xen_mark_pinned, FIXADDR_TOP);
 }
 
-static int unpin_page(struct page *page, enum pt_level level)
+static int xen_unpin_page(struct mm_struct *mm, struct page *page,
+			  enum pt_level level)
 {
 	unsigned pgfl = TestClearPagePinned(page);
 
@@ -796,10 +931,18 @@ static int unpin_page(struct page *page, enum pt_level level)
 		spinlock_t *ptl = NULL;
 		struct multicall_space mcs;
 
+		/*
+		 * Do the converse to pin_page.  If we're using split
+		 * pte locks, we must be holding the lock for while
+		 * the pte page is unpinned but still RO to prevent
+		 * concurrent updates from seeing it in this
+		 * partially-pinned state.
+		 */
 		if (level == PT_PTE) {
-			ptl = lock_pte(page);
+			ptl = xen_pte_lock(page, mm);
 
-			xen_do_pin(MMUEXT_UNPIN_TABLE, pfn);
+			if (ptl)
+				xen_do_pin(MMUEXT_UNPIN_TABLE, pfn);
 		}
 
 		mcs = __xen_mc_entry(0);
@@ -810,7 +953,7 @@ static int unpin_page(struct page *page, enum pt_level level)
 
 		if (ptl) {
 			/* unlock when batch completed */
-			xen_mc_callback(do_unlock, ptl);
+			xen_mc_callback(xen_pte_unlock, ptl);
 		}
 	}
 
@@ -818,7 +961,7 @@ static int unpin_page(struct page *page, enum pt_level level)
 }
 
 /* Release a pagetables pages back as normal RW */
-static void xen_pgd_unpin(pgd_t *pgd)
+static void __xen_pgd_unpin(struct mm_struct *mm, pgd_t *pgd)
 {
 	xen_mc_batch();
 
@@ -830,21 +973,27 @@ static void xen_pgd_unpin(pgd_t *pgd)
 
 		if (user_pgd) {
 			xen_do_pin(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(user_pgd)));
-			unpin_page(virt_to_page(user_pgd), PT_PGD);
+			xen_unpin_page(mm, virt_to_page(user_pgd), PT_PGD);
 		}
 	}
 #endif
 
 #ifdef CONFIG_X86_PAE
 	/* Need to make sure unshared kernel PMD is unpinned */
-	pin_page(virt_to_page(pgd_page(pgd[pgd_index(TASK_SIZE)])), PT_PMD);
+	xen_unpin_page(mm, virt_to_page(pgd_page(pgd[pgd_index(TASK_SIZE)])),
+		       PT_PMD);
 #endif
 
-	pgd_walk(pgd, unpin_page, USER_LIMIT);
+	xen_pgd_walk(mm, xen_unpin_page, USER_LIMIT);
 
 	xen_mc_issue(0);
 }
 
+static void xen_pgd_unpin(struct mm_struct *mm)
+{
+	__xen_pgd_unpin(mm, mm->pgd);
+}
+
 /*
  * On resume, undo any pinning done at save, so that the rest of the
  * kernel doesn't see any unexpected pinned pagetables.
@@ -859,7 +1008,7 @@ void xen_mm_unpin_all(void)
 	list_for_each_entry(page, &pgd_list, lru) {
 		if (PageSavePinned(page)) {
 			BUG_ON(!PagePinned(page));
-			xen_pgd_unpin((pgd_t *)page_address(page));
+			__xen_pgd_unpin(&init_mm, (pgd_t *)page_address(page));
 			ClearPageSavePinned(page);
 		}
 	}
@@ -870,14 +1019,14 @@ void xen_mm_unpin_all(void)
 void xen_activate_mm(struct mm_struct *prev, struct mm_struct *next)
 {
 	spin_lock(&next->page_table_lock);
-	xen_pgd_pin(next->pgd);
+	xen_pgd_pin(next);
 	spin_unlock(&next->page_table_lock);
 }
 
 void xen_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm)
 {
 	spin_lock(&mm->page_table_lock);
-	xen_pgd_pin(mm->pgd);
+	xen_pgd_pin(mm);
 	spin_unlock(&mm->page_table_lock);
 }
 
@@ -907,7 +1056,7 @@ static void drop_other_mm_ref(void *info)
 	}
 }
 
-static void drop_mm_ref(struct mm_struct *mm)
+static void xen_drop_mm_ref(struct mm_struct *mm)
 {
 	cpumask_t mask;
 	unsigned cpu;
@@ -937,7 +1086,7 @@ static void drop_mm_ref(struct mm_struct *mm)
 		smp_call_function_mask(mask, drop_other_mm_ref, mm, 1);
 }
 #else
-static void drop_mm_ref(struct mm_struct *mm)
+static void xen_drop_mm_ref(struct mm_struct *mm)
 {
 	if (current->active_mm == mm)
 		load_cr3(swapper_pg_dir);
@@ -961,14 +1110,77 @@ static void drop_mm_ref(struct mm_struct *mm)
 void xen_exit_mmap(struct mm_struct *mm)
 {
 	get_cpu();		/* make sure we don't move around */
-	drop_mm_ref(mm);
+	xen_drop_mm_ref(mm);
 	put_cpu();
 
 	spin_lock(&mm->page_table_lock);
 
 	/* pgd may not be pinned in the error exit path of execve */
-	if (page_pinned(mm->pgd))
-		xen_pgd_unpin(mm->pgd);
+	if (xen_page_pinned(mm->pgd))
+		xen_pgd_unpin(mm);
 
 	spin_unlock(&mm->page_table_lock);
 }
+
+#ifdef CONFIG_XEN_DEBUG_FS
+
+static struct dentry *d_mmu_debug;
+
+static int __init xen_mmu_debugfs(void)
+{
+	struct dentry *d_xen = xen_init_debugfs();
+
+	if (d_xen == NULL)
+		return -ENOMEM;
+
+	d_mmu_debug = debugfs_create_dir("mmu", d_xen);
+
+	debugfs_create_u8("zero_stats", 0644, d_mmu_debug, &zero_stats);
+
+	debugfs_create_u32("pgd_update", 0444, d_mmu_debug, &mmu_stats.pgd_update);
+	debugfs_create_u32("pgd_update_pinned", 0444, d_mmu_debug,
+			   &mmu_stats.pgd_update_pinned);
+	debugfs_create_u32("pgd_update_batched", 0444, d_mmu_debug,
+			   &mmu_stats.pgd_update_pinned);
+
+	debugfs_create_u32("pud_update", 0444, d_mmu_debug, &mmu_stats.pud_update);
+	debugfs_create_u32("pud_update_pinned", 0444, d_mmu_debug,
+			   &mmu_stats.pud_update_pinned);
+	debugfs_create_u32("pud_update_batched", 0444, d_mmu_debug,
+			   &mmu_stats.pud_update_pinned);
+
+	debugfs_create_u32("pmd_update", 0444, d_mmu_debug, &mmu_stats.pmd_update);
+	debugfs_create_u32("pmd_update_pinned", 0444, d_mmu_debug,
+			   &mmu_stats.pmd_update_pinned);
+	debugfs_create_u32("pmd_update_batched", 0444, d_mmu_debug,
+			   &mmu_stats.pmd_update_pinned);
+
+	debugfs_create_u32("pte_update", 0444, d_mmu_debug, &mmu_stats.pte_update);
+//	debugfs_create_u32("pte_update_pinned", 0444, d_mmu_debug,
+//			   &mmu_stats.pte_update_pinned);
+	debugfs_create_u32("pte_update_batched", 0444, d_mmu_debug,
+			   &mmu_stats.pte_update_pinned);
+
+	debugfs_create_u32("mmu_update", 0444, d_mmu_debug, &mmu_stats.mmu_update);
+	debugfs_create_u32("mmu_update_extended", 0444, d_mmu_debug,
+			   &mmu_stats.mmu_update_extended);
+	xen_debugfs_create_u32_array("mmu_update_histo", 0444, d_mmu_debug,
+				     mmu_stats.mmu_update_histo, 20);
+
+	debugfs_create_u32("set_pte_at", 0444, d_mmu_debug, &mmu_stats.set_pte_at);
+	debugfs_create_u32("set_pte_at_batched", 0444, d_mmu_debug,
+			   &mmu_stats.set_pte_at_batched);
+	debugfs_create_u32("set_pte_at_current", 0444, d_mmu_debug,
+			   &mmu_stats.set_pte_at_current);
+	debugfs_create_u32("set_pte_at_kernel", 0444, d_mmu_debug,
+			   &mmu_stats.set_pte_at_kernel);
+
+	debugfs_create_u32("prot_commit", 0444, d_mmu_debug, &mmu_stats.prot_commit);
+	debugfs_create_u32("prot_commit_batched", 0444, d_mmu_debug,
+			   &mmu_stats.prot_commit_batched);
+
+	return 0;
+}
+fs_initcall(xen_mmu_debugfs);
+
+#endif	/* CONFIG_XEN_DEBUG_FS */
diff --git a/arch/x86/xen/mmu.h b/arch/x86/xen/mmu.h
index 0f59bd03f9e3..98d71659da5a 100644
--- a/arch/x86/xen/mmu.h
+++ b/arch/x86/xen/mmu.h
@@ -18,9 +18,6 @@ void xen_activate_mm(struct mm_struct *prev, struct mm_struct *next);
 void xen_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm);
 void xen_exit_mmap(struct mm_struct *mm);
 
-void xen_pgd_pin(pgd_t *pgd);
-//void xen_pgd_unpin(pgd_t *pgd);
-
 pteval_t xen_pte_val(pte_t);
 pmdval_t xen_pmd_val(pmd_t);
 pgdval_t xen_pgd_val(pgd_t);
diff --git a/arch/x86/xen/multicalls.c b/arch/x86/xen/multicalls.c
index 9efd1c6c9776..8ea8a0d0b0de 100644
--- a/arch/x86/xen/multicalls.c
+++ b/arch/x86/xen/multicalls.c
@@ -21,16 +21,20 @@
  */
 #include <linux/percpu.h>
 #include <linux/hardirq.h>
+#include <linux/debugfs.h>
 
 #include <asm/xen/hypercall.h>
 
 #include "multicalls.h"
+#include "debugfs.h"
+
+#define MC_BATCH	32
 
 #define MC_DEBUG	1
 
-#define MC_BATCH	32
 #define MC_ARGS		(MC_BATCH * 16)
 
+
 struct mc_buffer {
 	struct multicall_entry entries[MC_BATCH];
 #if MC_DEBUG
@@ -47,6 +51,76 @@ struct mc_buffer {
 static DEFINE_PER_CPU(struct mc_buffer, mc_buffer);
 DEFINE_PER_CPU(unsigned long, xen_mc_irq_flags);
 
+/* flush reasons 0- slots, 1- args, 2- callbacks */
+enum flush_reasons
+{
+	FL_SLOTS,
+	FL_ARGS,
+	FL_CALLBACKS,
+
+	FL_N_REASONS
+};
+
+#ifdef CONFIG_XEN_DEBUG_FS
+#define NHYPERCALLS	40		/* not really */
+
+static struct {
+	unsigned histo[MC_BATCH+1];
+
+	unsigned issued;
+	unsigned arg_total;
+	unsigned hypercalls;
+	unsigned histo_hypercalls[NHYPERCALLS];
+
+	unsigned flush[FL_N_REASONS];
+} mc_stats;
+
+static u8 zero_stats;
+
+static inline void check_zero(void)
+{
+	if (unlikely(zero_stats)) {
+		memset(&mc_stats, 0, sizeof(mc_stats));
+		zero_stats = 0;
+	}
+}
+
+static void mc_add_stats(const struct mc_buffer *mc)
+{
+	int i;
+
+	check_zero();
+
+	mc_stats.issued++;
+	mc_stats.hypercalls += mc->mcidx;
+	mc_stats.arg_total += mc->argidx;
+
+	mc_stats.histo[mc->mcidx]++;
+	for(i = 0; i < mc->mcidx; i++) {
+		unsigned op = mc->entries[i].op;
+		if (op < NHYPERCALLS)
+			mc_stats.histo_hypercalls[op]++;
+	}
+}
+
+static void mc_stats_flush(enum flush_reasons idx)
+{
+	check_zero();
+
+	mc_stats.flush[idx]++;
+}
+
+#else  /* !CONFIG_XEN_DEBUG_FS */
+
+static inline void mc_add_stats(const struct mc_buffer *mc)
+{
+}
+
+static inline void mc_stats_flush(enum flush_reasons idx)
+{
+}
+#endif	/* CONFIG_XEN_DEBUG_FS */
+
 void xen_mc_flush(void)
 {
 	struct mc_buffer *b = &__get_cpu_var(mc_buffer);
@@ -60,6 +134,8 @@ void xen_mc_flush(void)
 	   something in the middle */
 	local_irq_save(flags);
 
+	mc_add_stats(b);
+
 	if (b->mcidx) {
 #if MC_DEBUG
 		memcpy(b->debug, b->entries,
@@ -115,6 +191,7 @@ struct multicall_space __xen_mc_entry(size_t args)
 
 	if (b->mcidx == MC_BATCH ||
 	    (argidx + args) > MC_ARGS) {
+		mc_stats_flush(b->mcidx == MC_BATCH ? FL_SLOTS : FL_ARGS);
 		xen_mc_flush();
 		argidx = roundup(b->argidx, sizeof(u64));
 	}
@@ -158,10 +235,44 @@ void xen_mc_callback(void (*fn)(void *), void *data)
 	struct mc_buffer *b = &__get_cpu_var(mc_buffer);
 	struct callback *cb;
 
-	if (b->cbidx == MC_BATCH)
+	if (b->cbidx == MC_BATCH) {
+		mc_stats_flush(FL_CALLBACKS);
 		xen_mc_flush();
+	}
 
 	cb = &b->callbacks[b->cbidx++];
 	cb->fn = fn;
 	cb->data = data;
 }
+
+#ifdef CONFIG_XEN_DEBUG_FS
+
+static struct dentry *d_mc_debug;
+
+static int __init xen_mc_debugfs(void)
+{
+	struct dentry *d_xen = xen_init_debugfs();
+
+	if (d_xen == NULL)
+		return -ENOMEM;
+
+	d_mc_debug = debugfs_create_dir("multicalls", d_xen);
+
+	debugfs_create_u8("zero_stats", 0644, d_mc_debug, &zero_stats);
+
+	debugfs_create_u32("batches", 0444, d_mc_debug, &mc_stats.issued);
+	debugfs_create_u32("hypercalls", 0444, d_mc_debug, &mc_stats.hypercalls);
+	debugfs_create_u32("arg_total", 0444, d_mc_debug, &mc_stats.arg_total);
+
+	xen_debugfs_create_u32_array("batch_histo", 0444, d_mc_debug,
+				     mc_stats.histo, MC_BATCH);
+	xen_debugfs_create_u32_array("hypercall_histo", 0444, d_mc_debug,
+				     mc_stats.histo_hypercalls, NHYPERCALLS);
+	xen_debugfs_create_u32_array("flush_reasons", 0444, d_mc_debug,
+				     mc_stats.flush, FL_N_REASONS);
+
+	return 0;
+}
+fs_initcall(xen_mc_debugfs);
+
+#endif	/* CONFIG_XEN_DEBUG_FS */
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index d8faf79a0a1d..d77da613b1d2 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -11,11 +11,8 @@
  * useful topology information for the kernel to make use of.  As a
  * result, all CPUs are treated as if they're single-core and
  * single-threaded.
- *
- * This does not handle HOTPLUG_CPU yet.
  */
 #include <linux/sched.h>
-#include <linux/kernel_stat.h>
 #include <linux/err.h>
 #include <linux/smp.h>
 
@@ -36,8 +33,6 @@
 #include "xen-ops.h"
 #include "mmu.h"
 
-static void __cpuinit xen_init_lock_cpu(int cpu);
-
 cpumask_t xen_cpu_initialized_map;
 
 static DEFINE_PER_CPU(int, resched_irq);
@@ -64,11 +59,12 @@ static irqreturn_t xen_reschedule_interrupt(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-static __cpuinit void cpu_bringup_and_idle(void)
+static __cpuinit void cpu_bringup(void)
 {
 	int cpu = smp_processor_id();
 
 	cpu_init();
+	touch_softlockup_watchdog();
 	preempt_disable();
 
 	xen_enable_sysenter();
@@ -89,6 +85,11 @@ static __cpuinit void cpu_bringup_and_idle(void)
 	local_irq_enable();
 
 	wmb();			/* make sure everything is out */
+}
+
+static __cpuinit void cpu_bringup_and_idle(void)
+{
+	cpu_bringup();
 	cpu_idle();
 }
 
@@ -212,8 +213,6 @@ static void __init xen_smp_prepare_cpus(unsigned int max_cpus)
 
 		cpu_set(cpu, cpu_present_map);
 	}
-
-	//init_xenbus_allowed_cpumask();
 }
 
 static __cpuinit int
@@ -281,12 +280,6 @@ static int __cpuinit xen_cpu_up(unsigned int cpu)
 	struct task_struct *idle = idle_task(cpu);
 	int rc;
 
-#if 0
-	rc = cpu_up_check(cpu);
-	if (rc)
-		return rc;
-#endif
-
 #ifdef CONFIG_X86_64
 	/* Allocate node local memory for AP pdas */
 	WARN_ON(cpu == 0);
@@ -339,6 +332,60 @@ static void xen_smp_cpus_done(unsigned int max_cpus)
 {
 }
 
+#ifdef CONFIG_HOTPLUG_CPU
+static int xen_cpu_disable(void)
+{
+	unsigned int cpu = smp_processor_id();
+	if (cpu == 0)
+		return -EBUSY;
+
+	cpu_disable_common();
+
+	load_cr3(swapper_pg_dir);
+	return 0;
+}
+
+static void xen_cpu_die(unsigned int cpu)
+{
+	while (HYPERVISOR_vcpu_op(VCPUOP_is_up, cpu, NULL)) {
+		current->state = TASK_UNINTERRUPTIBLE;
+		schedule_timeout(HZ/10);
+	}
+	unbind_from_irqhandler(per_cpu(resched_irq, cpu), NULL);
+	unbind_from_irqhandler(per_cpu(callfunc_irq, cpu), NULL);
+	unbind_from_irqhandler(per_cpu(debug_irq, cpu), NULL);
+	unbind_from_irqhandler(per_cpu(callfuncsingle_irq, cpu), NULL);
+	xen_uninit_lock_cpu(cpu);
+	xen_teardown_timer(cpu);
+
+	if (num_online_cpus() == 1)
+		alternatives_smp_switch(0);
+}
+
+static void xen_play_dead(void)
+{
+	play_dead_common();
+	HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL);
+	cpu_bringup();
+}
+
+#else /* !CONFIG_HOTPLUG_CPU */
+static int xen_cpu_disable(void)
+{
+	return -ENOSYS;
+}
+
+static void xen_cpu_die(unsigned int cpu)
+{
+	BUG();
+}
+
+static void xen_play_dead(void)
+{
+	BUG();
+}
+
+#endif
 static void stop_self(void *v)
 {
 	int cpu = smp_processor_id();
@@ -419,176 +466,16 @@ static irqreturn_t xen_call_function_single_interrupt(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-struct xen_spinlock {
-	unsigned char lock;		/* 0 -> free; 1 -> locked */
-	unsigned short spinners;	/* count of waiting cpus */
-};
-
-static int xen_spin_is_locked(struct raw_spinlock *lock)
-{
-	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
-
-	return xl->lock != 0;
-}
-
-static int xen_spin_is_contended(struct raw_spinlock *lock)
-{
-	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
-
-	/* Not strictly true; this is only the count of contended
-	   lock-takers entering the slow path. */
-	return xl->spinners != 0;
-}
-
-static int xen_spin_trylock(struct raw_spinlock *lock)
-{
-	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
-	u8 old = 1;
-
-	asm("xchgb %b0,%1"
-	    : "+q" (old), "+m" (xl->lock) : : "memory");
-
-	return old == 0;
-}
-
-static DEFINE_PER_CPU(int, lock_kicker_irq) = -1;
-static DEFINE_PER_CPU(struct xen_spinlock *, lock_spinners);
-
-static inline void spinning_lock(struct xen_spinlock *xl)
-{
-	__get_cpu_var(lock_spinners) = xl;
-	wmb();			/* set lock of interest before count */
-	asm(LOCK_PREFIX " incw %0"
-	    : "+m" (xl->spinners) : : "memory");
-}
-
-static inline void unspinning_lock(struct xen_spinlock *xl)
-{
-	asm(LOCK_PREFIX " decw %0"
-	    : "+m" (xl->spinners) : : "memory");
-	wmb();			/* decrement count before clearing lock */
-	__get_cpu_var(lock_spinners) = NULL;
-}
-
-static noinline int xen_spin_lock_slow(struct raw_spinlock *lock)
-{
-	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
-	int irq = __get_cpu_var(lock_kicker_irq);
-	int ret;
-
-	/* If kicker interrupts not initialized yet, just spin */
-	if (irq == -1)
-		return 0;
-
-	/* announce we're spinning */
-	spinning_lock(xl);
-
-	/* clear pending */
-	xen_clear_irq_pending(irq);
-
-	/* check again make sure it didn't become free while
-	   we weren't looking  */
-	ret = xen_spin_trylock(lock);
-	if (ret)
-		goto out;
-
-	/* block until irq becomes pending */
-	xen_poll_irq(irq);
-	kstat_this_cpu.irqs[irq]++;
-
-out:
-	unspinning_lock(xl);
-	return ret;
-}
-
-static void xen_spin_lock(struct raw_spinlock *lock)
-{
-	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
-	int timeout;
-	u8 oldval;
-
-	do {
-		timeout = 1 << 10;
-
-		asm("1: xchgb %1,%0\n"
-		    "   testb %1,%1\n"
-		    "   jz 3f\n"
-		    "2: rep;nop\n"
-		    "   cmpb $0,%0\n"
-		    "   je 1b\n"
-		    "   dec %2\n"
-		    "   jnz 2b\n"
-		    "3:\n"
-		    : "+m" (xl->lock), "=q" (oldval), "+r" (timeout)
-		    : "1" (1)
-		    : "memory");
-
-	} while (unlikely(oldval != 0 && !xen_spin_lock_slow(lock)));
-}
-
-static noinline void xen_spin_unlock_slow(struct xen_spinlock *xl)
-{
-	int cpu;
-
-	for_each_online_cpu(cpu) {
-		/* XXX should mix up next cpu selection */
-		if (per_cpu(lock_spinners, cpu) == xl) {
-			xen_send_IPI_one(cpu, XEN_SPIN_UNLOCK_VECTOR);
-			break;
-		}
-	}
-}
-
-static void xen_spin_unlock(struct raw_spinlock *lock)
-{
-	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
-
-	smp_wmb();		/* make sure no writes get moved after unlock */
-	xl->lock = 0;		/* release lock */
-
-	/* make sure unlock happens before kick */
-	barrier();
-
-	if (unlikely(xl->spinners))
-		xen_spin_unlock_slow(xl);
-}
-
-static __cpuinit void xen_init_lock_cpu(int cpu)
-{
-	int irq;
-	const char *name;
-
-	name = kasprintf(GFP_KERNEL, "spinlock%d", cpu);
-	irq = bind_ipi_to_irqhandler(XEN_SPIN_UNLOCK_VECTOR,
-				     cpu,
-				     xen_reschedule_interrupt,
-				     IRQF_DISABLED|IRQF_PERCPU|IRQF_NOBALANCING,
-				     name,
-				     NULL);
-
-	if (irq >= 0) {
-		disable_irq(irq); /* make sure it's never delivered */
-		per_cpu(lock_kicker_irq, cpu) = irq;
-	}
-
-	printk("cpu %d spinlock event irq %d\n", cpu, irq);
-}
-
-static void __init xen_init_spinlocks(void)
-{
-	pv_lock_ops.spin_is_locked = xen_spin_is_locked;
-	pv_lock_ops.spin_is_contended = xen_spin_is_contended;
-	pv_lock_ops.spin_lock = xen_spin_lock;
-	pv_lock_ops.spin_trylock = xen_spin_trylock;
-	pv_lock_ops.spin_unlock = xen_spin_unlock;
-}
-
 static const struct smp_ops xen_smp_ops __initdata = {
 	.smp_prepare_boot_cpu = xen_smp_prepare_boot_cpu,
 	.smp_prepare_cpus = xen_smp_prepare_cpus,
-	.cpu_up = xen_cpu_up,
 	.smp_cpus_done = xen_smp_cpus_done,
 
+	.cpu_up = xen_cpu_up,
+	.cpu_die = xen_cpu_die,
+	.cpu_disable = xen_cpu_disable,
+	.play_dead = xen_play_dead,
+
 	.smp_send_stop = xen_smp_send_stop,
 	.smp_send_reschedule = xen_smp_send_reschedule,
 
diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c
new file mode 100644
index 000000000000..dd71e3a021cd
--- /dev/null
+++ b/arch/x86/xen/spinlock.c
@@ -0,0 +1,428 @@
+/*
+ * Split spinlock implementation out into its own file, so it can be
+ * compiled in a FTRACE-compatible way.
+ */
+#include <linux/kernel_stat.h>
+#include <linux/spinlock.h>
+#include <linux/debugfs.h>
+#include <linux/log2.h>
+
+#include <asm/paravirt.h>
+
+#include <xen/interface/xen.h>
+#include <xen/events.h>
+
+#include "xen-ops.h"
+#include "debugfs.h"
+
+#ifdef CONFIG_XEN_DEBUG_FS
+static struct xen_spinlock_stats
+{
+	u64 taken;
+	u32 taken_slow;
+	u32 taken_slow_nested;
+	u32 taken_slow_pickup;
+	u32 taken_slow_spurious;
+	u32 taken_slow_irqenable;
+
+	u64 released;
+	u32 released_slow;
+	u32 released_slow_kicked;
+
+#define HISTO_BUCKETS	30
+	u32 histo_spin_total[HISTO_BUCKETS+1];
+	u32 histo_spin_spinning[HISTO_BUCKETS+1];
+	u32 histo_spin_blocked[HISTO_BUCKETS+1];
+
+	u64 time_total;
+	u64 time_spinning;
+	u64 time_blocked;
+} spinlock_stats;
+
+static u8 zero_stats;
+
+static unsigned lock_timeout = 1 << 10;
+#define TIMEOUT lock_timeout
+
+static inline void check_zero(void)
+{
+	if (unlikely(zero_stats)) {
+		memset(&spinlock_stats, 0, sizeof(spinlock_stats));
+		zero_stats = 0;
+	}
+}
+
+#define ADD_STATS(elem, val)			\
+	do { check_zero(); spinlock_stats.elem += (val); } while(0)
+
+static inline u64 spin_time_start(void)
+{
+	return xen_clocksource_read();
+}
+
+static void __spin_time_accum(u64 delta, u32 *array)
+{
+	unsigned index = ilog2(delta);
+
+	check_zero();
+
+	if (index < HISTO_BUCKETS)
+		array[index]++;
+	else
+		array[HISTO_BUCKETS]++;
+}
+
+static inline void spin_time_accum_spinning(u64 start)
+{
+	u32 delta = xen_clocksource_read() - start;
+
+	__spin_time_accum(delta, spinlock_stats.histo_spin_spinning);
+	spinlock_stats.time_spinning += delta;
+}
+
+static inline void spin_time_accum_total(u64 start)
+{
+	u32 delta = xen_clocksource_read() - start;
+
+	__spin_time_accum(delta, spinlock_stats.histo_spin_total);
+	spinlock_stats.time_total += delta;
+}
+
+static inline void spin_time_accum_blocked(u64 start)
+{
+	u32 delta = xen_clocksource_read() - start;
+
+	__spin_time_accum(delta, spinlock_stats.histo_spin_blocked);
+	spinlock_stats.time_blocked += delta;
+}
+#else  /* !CONFIG_XEN_DEBUG_FS */
+#define TIMEOUT			(1 << 10)
+#define ADD_STATS(elem, val)	do { (void)(val); } while(0)
+
+static inline u64 spin_time_start(void)
+{
+	return 0;
+}
+
+static inline void spin_time_accum_total(u64 start)
+{
+}
+static inline void spin_time_accum_spinning(u64 start)
+{
+}
+static inline void spin_time_accum_blocked(u64 start)
+{
+}
+#endif  /* CONFIG_XEN_DEBUG_FS */
+
+struct xen_spinlock {
+	unsigned char lock;		/* 0 -> free; 1 -> locked */
+	unsigned short spinners;	/* count of waiting cpus */
+};
+
+static int xen_spin_is_locked(struct raw_spinlock *lock)
+{
+	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
+
+	return xl->lock != 0;
+}
+
+static int xen_spin_is_contended(struct raw_spinlock *lock)
+{
+	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
+
+	/* Not strictly true; this is only the count of contended
+	   lock-takers entering the slow path. */
+	return xl->spinners != 0;
+}
+
+static int xen_spin_trylock(struct raw_spinlock *lock)
+{
+	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
+	u8 old = 1;
+
+	asm("xchgb %b0,%1"
+	    : "+q" (old), "+m" (xl->lock) : : "memory");
+
+	return old == 0;
+}
+
+static DEFINE_PER_CPU(int, lock_kicker_irq) = -1;
+static DEFINE_PER_CPU(struct xen_spinlock *, lock_spinners);
+
+/*
+ * Mark a cpu as interested in a lock.  Returns the CPU's previous
+ * lock of interest, in case we got preempted by an interrupt.
+ */
+static inline struct xen_spinlock *spinning_lock(struct xen_spinlock *xl)
+{
+	struct xen_spinlock *prev;
+
+	prev = __get_cpu_var(lock_spinners);
+	__get_cpu_var(lock_spinners) = xl;
+
+	wmb();			/* set lock of interest before count */
+
+	asm(LOCK_PREFIX " incw %0"
+	    : "+m" (xl->spinners) : : "memory");
+
+	return prev;
+}
+
+/*
+ * Mark a cpu as no longer interested in a lock.  Restores previous
+ * lock of interest (NULL for none).
+ */
+static inline void unspinning_lock(struct xen_spinlock *xl, struct xen_spinlock *prev)
+{
+	asm(LOCK_PREFIX " decw %0"
+	    : "+m" (xl->spinners) : : "memory");
+	wmb();			/* decrement count before restoring lock */
+	__get_cpu_var(lock_spinners) = prev;
+}
+
+static noinline int xen_spin_lock_slow(struct raw_spinlock *lock, bool irq_enable)
+{
+	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
+	struct xen_spinlock *prev;
+	int irq = __get_cpu_var(lock_kicker_irq);
+	int ret;
+	unsigned long flags;
+	u64 start;
+
+	/* If kicker interrupts not initialized yet, just spin */
+	if (irq == -1)
+		return 0;
+
+	start = spin_time_start();
+
+	/* announce we're spinning */
+	prev = spinning_lock(xl);
+
+	flags = __raw_local_save_flags();
+	if (irq_enable) {
+		ADD_STATS(taken_slow_irqenable, 1);
+		raw_local_irq_enable();
+	}
+
+	ADD_STATS(taken_slow, 1);
+	ADD_STATS(taken_slow_nested, prev != NULL);
+
+	do {
+		/* clear pending */
+		xen_clear_irq_pending(irq);
+
+		/* check again make sure it didn't become free while
+		   we weren't looking  */
+		ret = xen_spin_trylock(lock);
+		if (ret) {
+			ADD_STATS(taken_slow_pickup, 1);
+
+			/*
+			 * If we interrupted another spinlock while it
+			 * was blocking, make sure it doesn't block
+			 * without rechecking the lock.
+			 */
+			if (prev != NULL)
+				xen_set_irq_pending(irq);
+			goto out;
+		}
+
+		/*
+		 * Block until irq becomes pending.  If we're
+		 * interrupted at this point (after the trylock but
+		 * before entering the block), then the nested lock
+		 * handler guarantees that the irq will be left
+		 * pending if there's any chance the lock became free;
+		 * xen_poll_irq() returns immediately if the irq is
+		 * pending.
+		 */
+		xen_poll_irq(irq);
+		ADD_STATS(taken_slow_spurious, !xen_test_irq_pending(irq));
+	} while (!xen_test_irq_pending(irq)); /* check for spurious wakeups */
+
+	kstat_this_cpu.irqs[irq]++;
+
+out:
+	raw_local_irq_restore(flags);
+	unspinning_lock(xl, prev);
+	spin_time_accum_blocked(start);
+
+	return ret;
+}
+
+static inline void __xen_spin_lock(struct raw_spinlock *lock, bool irq_enable)
+{
+	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
+	unsigned timeout;
+	u8 oldval;
+	u64 start_spin;
+
+	ADD_STATS(taken, 1);
+
+	start_spin = spin_time_start();
+
+	do {
+		u64 start_spin_fast = spin_time_start();
+
+		timeout = TIMEOUT;
+
+		asm("1: xchgb %1,%0\n"
+		    "   testb %1,%1\n"
+		    "   jz 3f\n"
+		    "2: rep;nop\n"
+		    "   cmpb $0,%0\n"
+		    "   je 1b\n"
+		    "   dec %2\n"
+		    "   jnz 2b\n"
+		    "3:\n"
+		    : "+m" (xl->lock), "=q" (oldval), "+r" (timeout)
+		    : "1" (1)
+		    : "memory");
+
+		spin_time_accum_spinning(start_spin_fast);
+
+	} while (unlikely(oldval != 0 &&
+			  (TIMEOUT == ~0 || !xen_spin_lock_slow(lock, irq_enable))));
+
+	spin_time_accum_total(start_spin);
+}
+
+static void xen_spin_lock(struct raw_spinlock *lock)
+{
+	__xen_spin_lock(lock, false);
+}
+
+static void xen_spin_lock_flags(struct raw_spinlock *lock, unsigned long flags)
+{
+	__xen_spin_lock(lock, !raw_irqs_disabled_flags(flags));
+}
+
+static noinline void xen_spin_unlock_slow(struct xen_spinlock *xl)
+{
+	int cpu;
+
+	ADD_STATS(released_slow, 1);
+
+	for_each_online_cpu(cpu) {
+		/* XXX should mix up next cpu selection */
+		if (per_cpu(lock_spinners, cpu) == xl) {
+			ADD_STATS(released_slow_kicked, 1);
+			xen_send_IPI_one(cpu, XEN_SPIN_UNLOCK_VECTOR);
+			break;
+		}
+	}
+}
+
+static void xen_spin_unlock(struct raw_spinlock *lock)
+{
+	struct xen_spinlock *xl = (struct xen_spinlock *)lock;
+
+	ADD_STATS(released, 1);
+
+	smp_wmb();		/* make sure no writes get moved after unlock */
+	xl->lock = 0;		/* release lock */
+
+	/* make sure unlock happens before kick */
+	barrier();
+
+	if (unlikely(xl->spinners))
+		xen_spin_unlock_slow(xl);
+}
+
+static irqreturn_t dummy_handler(int irq, void *dev_id)
+{
+	BUG();
+	return IRQ_HANDLED;
+}
+
+void __cpuinit xen_init_lock_cpu(int cpu)
+{
+	int irq;
+	const char *name;
+
+	name = kasprintf(GFP_KERNEL, "spinlock%d", cpu);
+	irq = bind_ipi_to_irqhandler(XEN_SPIN_UNLOCK_VECTOR,
+				     cpu,
+				     dummy_handler,
+				     IRQF_DISABLED|IRQF_PERCPU|IRQF_NOBALANCING,
+				     name,
+				     NULL);
+
+	if (irq >= 0) {
+		disable_irq(irq); /* make sure it's never delivered */
+		per_cpu(lock_kicker_irq, cpu) = irq;
+	}
+
+	printk("cpu %d spinlock event irq %d\n", cpu, irq);
+}
+
+void xen_uninit_lock_cpu(int cpu)
+{
+	unbind_from_irqhandler(per_cpu(lock_kicker_irq, cpu), NULL);
+}
+
+void __init xen_init_spinlocks(void)
+{
+	pv_lock_ops.spin_is_locked = xen_spin_is_locked;
+	pv_lock_ops.spin_is_contended = xen_spin_is_contended;
+	pv_lock_ops.spin_lock = xen_spin_lock;
+	pv_lock_ops.spin_lock_flags = xen_spin_lock_flags;
+	pv_lock_ops.spin_trylock = xen_spin_trylock;
+	pv_lock_ops.spin_unlock = xen_spin_unlock;
+}
+
+#ifdef CONFIG_XEN_DEBUG_FS
+
+static struct dentry *d_spin_debug;
+
+static int __init xen_spinlock_debugfs(void)
+{
+	struct dentry *d_xen = xen_init_debugfs();
+
+	if (d_xen == NULL)
+		return -ENOMEM;
+
+	d_spin_debug = debugfs_create_dir("spinlocks", d_xen);
+
+	debugfs_create_u8("zero_stats", 0644, d_spin_debug, &zero_stats);
+
+	debugfs_create_u32("timeout", 0644, d_spin_debug, &lock_timeout);
+
+	debugfs_create_u64("taken", 0444, d_spin_debug, &spinlock_stats.taken);
+	debugfs_create_u32("taken_slow", 0444, d_spin_debug,
+			   &spinlock_stats.taken_slow);
+	debugfs_create_u32("taken_slow_nested", 0444, d_spin_debug,
+			   &spinlock_stats.taken_slow_nested);
+	debugfs_create_u32("taken_slow_pickup", 0444, d_spin_debug,
+			   &spinlock_stats.taken_slow_pickup);
+	debugfs_create_u32("taken_slow_spurious", 0444, d_spin_debug,
+			   &spinlock_stats.taken_slow_spurious);
+	debugfs_create_u32("taken_slow_irqenable", 0444, d_spin_debug,
+			   &spinlock_stats.taken_slow_irqenable);
+
+	debugfs_create_u64("released", 0444, d_spin_debug, &spinlock_stats.released);
+	debugfs_create_u32("released_slow", 0444, d_spin_debug,
+			   &spinlock_stats.released_slow);
+	debugfs_create_u32("released_slow_kicked", 0444, d_spin_debug,
+			   &spinlock_stats.released_slow_kicked);
+
+	debugfs_create_u64("time_spinning", 0444, d_spin_debug,
+			   &spinlock_stats.time_spinning);
+	debugfs_create_u64("time_blocked", 0444, d_spin_debug,
+			   &spinlock_stats.time_blocked);
+	debugfs_create_u64("time_total", 0444, d_spin_debug,
+			   &spinlock_stats.time_total);
+
+	xen_debugfs_create_u32_array("histo_total", 0444, d_spin_debug,
+				     spinlock_stats.histo_spin_total, HISTO_BUCKETS + 1);
+	xen_debugfs_create_u32_array("histo_spinning", 0444, d_spin_debug,
+				     spinlock_stats.histo_spin_spinning, HISTO_BUCKETS + 1);
+	xen_debugfs_create_u32_array("histo_blocked", 0444, d_spin_debug,
+				     spinlock_stats.histo_spin_blocked, HISTO_BUCKETS + 1);
+
+	return 0;
+}
+fs_initcall(xen_spinlock_debugfs);
+
+#endif	/* CONFIG_XEN_DEBUG_FS */
diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c
index 685b77470fc3..004ba86326ae 100644
--- a/arch/x86/xen/time.c
+++ b/arch/x86/xen/time.c
@@ -30,8 +30,6 @@
 #define TIMER_SLOP	100000
 #define NS_PER_TICK	(1000000000LL / HZ)
 
-static cycle_t xen_clocksource_read(void);
-
 /* runstate info updated by Xen */
 static DEFINE_PER_CPU(struct vcpu_runstate_info, runstate);
 
@@ -213,7 +211,7 @@ unsigned long xen_tsc_khz(void)
 	return xen_khz;
 }
 
-static cycle_t xen_clocksource_read(void)
+cycle_t xen_clocksource_read(void)
 {
         struct pvclock_vcpu_time_info *src;
 	cycle_t ret;
@@ -452,6 +450,14 @@ void xen_setup_timer(int cpu)
 	setup_runstate_info(cpu);
 }
 
+void xen_teardown_timer(int cpu)
+{
+	struct clock_event_device *evt;
+	BUG_ON(cpu == 0);
+	evt = &per_cpu(xen_clock_events, cpu);
+	unbind_from_irqhandler(evt->irq, NULL);
+}
+
 void xen_setup_cpu_clockevents(void)
 {
 	BUG_ON(preemptible());
diff --git a/arch/x86/xen/xen-asm_32.S b/arch/x86/xen/xen-asm_32.S
index 2497a30f41de..42786f59d9c0 100644
--- a/arch/x86/xen/xen-asm_32.S
+++ b/arch/x86/xen/xen-asm_32.S
@@ -298,7 +298,7 @@ check_events:
 	push %eax
 	push %ecx
 	push %edx
-	call force_evtchn_callback
+	call xen_force_evtchn_callback
 	pop %edx
 	pop %ecx
 	pop %eax
diff --git a/arch/x86/xen/xen-asm_64.S b/arch/x86/xen/xen-asm_64.S
index 7f58304fafb3..05794c566e87 100644
--- a/arch/x86/xen/xen-asm_64.S
+++ b/arch/x86/xen/xen-asm_64.S
@@ -26,8 +26,15 @@
 /* Pseudo-flag used for virtual NMI, which we don't implement yet */
 #define XEN_EFLAGS_NMI	0x80000000
 
-#if 0
-#include <asm/percpu.h>
+#if 1
+/*
+	x86-64 does not yet support direct access to percpu variables
+	via a segment override, so we just need to make sure this code
+	never gets used
+ */
+#define BUG			ud2a
+#define PER_CPU_VAR(var, off)	0xdeadbeef
+#endif
 
 /*
 	Enable events.  This clears the event mask and tests the pending
@@ -35,6 +42,8 @@
 	events, then enter the hypervisor to get them handled.
  */
 ENTRY(xen_irq_enable_direct)
+	BUG
+
 	/* Unmask events */
 	movb $0, PER_CPU_VAR(xen_vcpu_info, XEN_vcpu_info_mask)
 
@@ -58,6 +67,8 @@ ENDPATCH(xen_irq_enable_direct)
 	non-zero.
  */
 ENTRY(xen_irq_disable_direct)
+	BUG
+
 	movb $1, PER_CPU_VAR(xen_vcpu_info, XEN_vcpu_info_mask)
 ENDPATCH(xen_irq_disable_direct)
 	ret
@@ -74,6 +85,8 @@ ENDPATCH(xen_irq_disable_direct)
 	Xen and x86 use opposite senses (mask vs enable).
  */
 ENTRY(xen_save_fl_direct)
+	BUG
+
 	testb $0xff, PER_CPU_VAR(xen_vcpu_info, XEN_vcpu_info_mask)
 	setz %ah
 	addb %ah,%ah
@@ -91,6 +104,8 @@ ENDPATCH(xen_save_fl_direct)
 	if so.
  */
 ENTRY(xen_restore_fl_direct)
+	BUG
+
 	testb $X86_EFLAGS_IF>>8, %ah
 	setz PER_CPU_VAR(xen_vcpu_info, XEN_vcpu_info_mask)
 	/* Preempt here doesn't matter because that will deal with
@@ -122,7 +137,7 @@ check_events:
 	push %r9
 	push %r10
 	push %r11
-	call force_evtchn_callback
+	call xen_force_evtchn_callback
 	pop %r11
 	pop %r10
 	pop %r9
@@ -133,7 +148,6 @@ check_events:
 	pop %rcx
 	pop %rax
 	ret
-#endif
 
 ENTRY(xen_adjust_exception_frame)
 	mov 8+0(%rsp),%rcx
diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h
index dd3c23152a2e..d7422dc2a55c 100644
--- a/arch/x86/xen/xen-ops.h
+++ b/arch/x86/xen/xen-ops.h
@@ -2,6 +2,7 @@
 #define XEN_OPS_H
 
 #include <linux/init.h>
+#include <linux/clocksource.h>
 #include <linux/irqreturn.h>
 #include <xen/xen-ops.h>
 
@@ -31,7 +32,10 @@ void xen_vcpu_restore(void);
 
 void __init xen_build_dynamic_phys_to_machine(void);
 
+void xen_init_irq_ops(void);
 void xen_setup_timer(int cpu);
+void xen_teardown_timer(int cpu);
+cycle_t xen_clocksource_read(void);
 void xen_setup_cpu_clockevents(void);
 unsigned long xen_tsc_khz(void);
 void __init xen_time_init(void);
@@ -50,6 +54,10 @@ void __init xen_setup_vcpu_info_placement(void);
 #ifdef CONFIG_SMP
 void xen_smp_init(void);
 
+void __init xen_init_spinlocks(void);
+__cpuinit void xen_init_lock_cpu(int cpu);
+void xen_uninit_lock_cpu(int cpu);
+
 extern cpumask_t xen_cpu_initialized_map;
 #else
 static inline void xen_smp_init(void) {}
diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c
index 8dd3336efd7e..3c578ef78c48 100644
--- a/drivers/acpi/glue.c
+++ b/drivers/acpi/glue.c
@@ -369,7 +369,6 @@ static int __init acpi_rtc_init(void)
 		DBG("RTC unavailable?\n");
 	return 0;
 }
-/* do this between RTC subsys_initcall() and rtc_cmos driver_initcall() */
-fs_initcall(acpi_rtc_init);
+module_init(acpi_rtc_init);
 
 #endif
diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig
index f17cd4b572f8..78fbec8ceda0 100644
--- a/drivers/ata/Kconfig
+++ b/drivers/ata/Kconfig
@@ -7,7 +7,6 @@ menuconfig ATA
 	depends on HAS_IOMEM
 	depends on BLOCK
 	depends on !(M32R || M68K) || BROKEN
-	depends on !SUN4 || BROKEN
 	select SCSI
 	---help---
 	  If you want to use a ATA hard disk, ATA tape drive, ATA CD-ROM or
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 59fe051957ef..5d312dc9be9f 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -503,7 +503,7 @@ int ata_cmd_ioctl(struct scsi_device *scsidev, void __user *arg)
 	scsi_cmd[0] = ATA_16;
 
 	scsi_cmd[4] = args[2];
-	if (args[0] == WIN_SMART) { /* hack -- ide driver does this too... */
+	if (args[0] == ATA_CMD_SMART) { /* hack -- ide driver does this too */
 		scsi_cmd[6]  = args[3];
 		scsi_cmd[8]  = args[1];
 		scsi_cmd[10] = 0x4f;
diff --git a/drivers/atm/fore200e.c b/drivers/atm/fore200e.c
index 73338d231db9..937c9c0ef4c9 100644
--- a/drivers/atm/fore200e.c
+++ b/drivers/atm/fore200e.c
@@ -47,8 +47,9 @@
 #include <asm/atomic.h>
 
 #ifdef CONFIG_SBUS
+#include <linux/of.h>
+#include <linux/of_device.h>
 #include <asm/idprom.h>
-#include <asm/sbus.h>
 #include <asm/openprom.h>
 #include <asm/oplib.h>
 #include <asm/pgtable.h>
@@ -661,249 +662,189 @@ fore200e_pca_proc_read(struct fore200e* fore200e, char *page)
 
 #ifdef CONFIG_SBUS
 
-static u32
-fore200e_sba_read(volatile u32 __iomem *addr)
+static u32 fore200e_sba_read(volatile u32 __iomem *addr)
 {
     return sbus_readl(addr);
 }
 
-
-static void
-fore200e_sba_write(u32 val, volatile u32 __iomem *addr)
+static void fore200e_sba_write(u32 val, volatile u32 __iomem *addr)
 {
     sbus_writel(val, addr);
 }
 
-
-static u32
-fore200e_sba_dma_map(struct fore200e* fore200e, void* virt_addr, int size, int direction)
+static u32 fore200e_sba_dma_map(struct fore200e *fore200e, void* virt_addr, int size, int direction)
 {
-    u32 dma_addr = sbus_map_single((struct sbus_dev*)fore200e->bus_dev, virt_addr, size, direction);
+	struct of_device *op = fore200e->bus_dev;
+	u32 dma_addr;
 
-    DPRINTK(3, "SBUS DVMA mapping: virt_addr = 0x%p, size = %d, direction = %d --> dma_addr = 0x%08x\n",
-	    virt_addr, size, direction, dma_addr);
+	dma_addr = dma_map_single(&op->dev, virt_addr, size, direction);
+
+	DPRINTK(3, "SBUS DVMA mapping: virt_addr = 0x%p, size = %d, direction = %d --> dma_addr = 0x%08x\n",
+		virt_addr, size, direction, dma_addr);
     
-    return dma_addr;
+	return dma_addr;
 }
 
-
-static void
-fore200e_sba_dma_unmap(struct fore200e* fore200e, u32 dma_addr, int size, int direction)
+static void fore200e_sba_dma_unmap(struct fore200e *fore200e, u32 dma_addr, int size, int direction)
 {
-    DPRINTK(3, "SBUS DVMA unmapping: dma_addr = 0x%08x, size = %d, direction = %d,\n",
-	    dma_addr, size, direction);
+	struct of_device *op = fore200e->bus_dev;
 
-    sbus_unmap_single((struct sbus_dev*)fore200e->bus_dev, dma_addr, size, direction);
-}
+	DPRINTK(3, "SBUS DVMA unmapping: dma_addr = 0x%08x, size = %d, direction = %d,\n",
+		dma_addr, size, direction);
 
+	dma_unmap_single(&op->dev, dma_addr, size, direction);
+}
 
-static void
-fore200e_sba_dma_sync_for_cpu(struct fore200e* fore200e, u32 dma_addr, int size, int direction)
+static void fore200e_sba_dma_sync_for_cpu(struct fore200e *fore200e, u32 dma_addr, int size, int direction)
 {
-    DPRINTK(3, "SBUS DVMA sync: dma_addr = 0x%08x, size = %d, direction = %d\n", dma_addr, size, direction);
+	struct of_device *op = fore200e->bus_dev;
+
+	DPRINTK(3, "SBUS DVMA sync: dma_addr = 0x%08x, size = %d, direction = %d\n", dma_addr, size, direction);
     
-    sbus_dma_sync_single_for_cpu((struct sbus_dev*)fore200e->bus_dev, dma_addr, size, direction);
+	dma_sync_single_for_cpu(&op->dev, dma_addr, size, direction);
 }
 
-static void
-fore200e_sba_dma_sync_for_device(struct fore200e* fore200e, u32 dma_addr, int size, int direction)
+static void fore200e_sba_dma_sync_for_device(struct fore200e *fore200e, u32 dma_addr, int size, int direction)
 {
-    DPRINTK(3, "SBUS DVMA sync: dma_addr = 0x%08x, size = %d, direction = %d\n", dma_addr, size, direction);
-
-    sbus_dma_sync_single_for_device((struct sbus_dev*)fore200e->bus_dev, dma_addr, size, direction);
-}
+	struct of_device *op = fore200e->bus_dev;
 
+	DPRINTK(3, "SBUS DVMA sync: dma_addr = 0x%08x, size = %d, direction = %d\n", dma_addr, size, direction);
 
-/* allocate a DVMA consistent chunk of memory intended to act as a communication mechanism
-   (to hold descriptors, status, queues, etc.) shared by the driver and the adapter */
+	dma_sync_single_for_device(&op->dev, dma_addr, size, direction);
+}
 
-static int
-fore200e_sba_dma_chunk_alloc(struct fore200e* fore200e, struct chunk* chunk,
-			     int size, int nbr, int alignment)
+/* Allocate a DVMA consistent chunk of memory intended to act as a communication mechanism
+ * (to hold descriptors, status, queues, etc.) shared by the driver and the adapter.
+ */
+static int fore200e_sba_dma_chunk_alloc(struct fore200e *fore200e, struct chunk *chunk,
+					int size, int nbr, int alignment)
 {
-    chunk->alloc_size = chunk->align_size = size * nbr;
+	struct of_device *op = fore200e->bus_dev;
 
-    /* returned chunks are page-aligned */
-    chunk->alloc_addr = sbus_alloc_consistent((struct sbus_dev*)fore200e->bus_dev,
-					      chunk->alloc_size,
-					      &chunk->dma_addr);
+	chunk->alloc_size = chunk->align_size = size * nbr;
 
-    if ((chunk->alloc_addr == NULL) || (chunk->dma_addr == 0))
-	return -ENOMEM;
+	/* returned chunks are page-aligned */
+	chunk->alloc_addr = dma_alloc_coherent(&op->dev, chunk->alloc_size,
+					       &chunk->dma_addr, GFP_ATOMIC);
 
-    chunk->align_addr = chunk->alloc_addr;
+	if ((chunk->alloc_addr == NULL) || (chunk->dma_addr == 0))
+		return -ENOMEM;
+
+	chunk->align_addr = chunk->alloc_addr;
     
-    return 0;
+	return 0;
 }
 
-
 /* free a DVMA consistent chunk of memory */
-
-static void
-fore200e_sba_dma_chunk_free(struct fore200e* fore200e, struct chunk* chunk)
+static void fore200e_sba_dma_chunk_free(struct fore200e *fore200e, struct chunk *chunk)
 {
-    sbus_free_consistent((struct sbus_dev*)fore200e->bus_dev,
-			 chunk->alloc_size,
-			 chunk->alloc_addr,
-			 chunk->dma_addr);
-}
+	struct of_device *op = fore200e->bus_dev;
 
+	dma_free_coherent(&op->dev, chunk->alloc_size,
+			  chunk->alloc_addr, chunk->dma_addr);
+}
 
-static void
-fore200e_sba_irq_enable(struct fore200e* fore200e)
+static void fore200e_sba_irq_enable(struct fore200e *fore200e)
 {
-    u32 hcr = fore200e->bus->read(fore200e->regs.sba.hcr) & SBA200E_HCR_STICKY;
-    fore200e->bus->write(hcr | SBA200E_HCR_INTR_ENA, fore200e->regs.sba.hcr);
+	u32 hcr = fore200e->bus->read(fore200e->regs.sba.hcr) & SBA200E_HCR_STICKY;
+	fore200e->bus->write(hcr | SBA200E_HCR_INTR_ENA, fore200e->regs.sba.hcr);
 }
 
-
-static int
-fore200e_sba_irq_check(struct fore200e* fore200e)
+static int fore200e_sba_irq_check(struct fore200e *fore200e)
 {
-    return fore200e->bus->read(fore200e->regs.sba.hcr) & SBA200E_HCR_INTR_REQ;
+	return fore200e->bus->read(fore200e->regs.sba.hcr) & SBA200E_HCR_INTR_REQ;
 }
 
-
-static void
-fore200e_sba_irq_ack(struct fore200e* fore200e)
+static void fore200e_sba_irq_ack(struct fore200e *fore200e)
 {
-    u32 hcr = fore200e->bus->read(fore200e->regs.sba.hcr) & SBA200E_HCR_STICKY;
-    fore200e->bus->write(hcr | SBA200E_HCR_INTR_CLR, fore200e->regs.sba.hcr);
+	u32 hcr = fore200e->bus->read(fore200e->regs.sba.hcr) & SBA200E_HCR_STICKY;
+	fore200e->bus->write(hcr | SBA200E_HCR_INTR_CLR, fore200e->regs.sba.hcr);
 }
 
-
-static void
-fore200e_sba_reset(struct fore200e* fore200e)
+static void fore200e_sba_reset(struct fore200e *fore200e)
 {
-    fore200e->bus->write(SBA200E_HCR_RESET, fore200e->regs.sba.hcr);
-    fore200e_spin(10);
-    fore200e->bus->write(0, fore200e->regs.sba.hcr);
+	fore200e->bus->write(SBA200E_HCR_RESET, fore200e->regs.sba.hcr);
+	fore200e_spin(10);
+	fore200e->bus->write(0, fore200e->regs.sba.hcr);
 }
 
-
-static int __init
-fore200e_sba_map(struct fore200e* fore200e)
+static int __init fore200e_sba_map(struct fore200e *fore200e)
 {
-    struct sbus_dev* sbus_dev = (struct sbus_dev*)fore200e->bus_dev;
-    unsigned int bursts;
+	struct of_device *op = fore200e->bus_dev;
+	unsigned int bursts;
 
-    /* gain access to the SBA specific registers  */
-    fore200e->regs.sba.hcr = sbus_ioremap(&sbus_dev->resource[0], 0, SBA200E_HCR_LENGTH, "SBA HCR");
-    fore200e->regs.sba.bsr = sbus_ioremap(&sbus_dev->resource[1], 0, SBA200E_BSR_LENGTH, "SBA BSR");
-    fore200e->regs.sba.isr = sbus_ioremap(&sbus_dev->resource[2], 0, SBA200E_ISR_LENGTH, "SBA ISR");
-    fore200e->virt_base    = sbus_ioremap(&sbus_dev->resource[3], 0, SBA200E_RAM_LENGTH, "SBA RAM");
+	/* gain access to the SBA specific registers  */
+	fore200e->regs.sba.hcr = of_ioremap(&op->resource[0], 0, SBA200E_HCR_LENGTH, "SBA HCR");
+	fore200e->regs.sba.bsr = of_ioremap(&op->resource[1], 0, SBA200E_BSR_LENGTH, "SBA BSR");
+	fore200e->regs.sba.isr = of_ioremap(&op->resource[2], 0, SBA200E_ISR_LENGTH, "SBA ISR");
+	fore200e->virt_base    = of_ioremap(&op->resource[3], 0, SBA200E_RAM_LENGTH, "SBA RAM");
 
-    if (fore200e->virt_base == NULL) {
-	printk(FORE200E "unable to map RAM of device %s\n", fore200e->name);
-	return -EFAULT;
-    }
+	if (!fore200e->virt_base) {
+		printk(FORE200E "unable to map RAM of device %s\n", fore200e->name);
+		return -EFAULT;
+	}
 
-    DPRINTK(1, "device %s mapped to 0x%p\n", fore200e->name, fore200e->virt_base);
+	DPRINTK(1, "device %s mapped to 0x%p\n", fore200e->name, fore200e->virt_base);
     
-    fore200e->bus->write(0x02, fore200e->regs.sba.isr); /* XXX hardwired interrupt level */
+	fore200e->bus->write(0x02, fore200e->regs.sba.isr); /* XXX hardwired interrupt level */
 
-    /* get the supported DVMA burst sizes */
-    bursts = prom_getintdefault(sbus_dev->bus->prom_node, "burst-sizes", 0x00);
+	/* get the supported DVMA burst sizes */
+	bursts = of_getintprop_default(op->node->parent, "burst-sizes", 0x00);
 
-    if (sbus_can_dma_64bit(sbus_dev))
-	sbus_set_sbus64(sbus_dev, bursts);
+	if (sbus_can_dma_64bit())
+		sbus_set_sbus64(&op->dev, bursts);
 
-    fore200e->state = FORE200E_STATE_MAP;
-    return 0;
+	fore200e->state = FORE200E_STATE_MAP;
+	return 0;
 }
 
-
-static void
-fore200e_sba_unmap(struct fore200e* fore200e)
+static void fore200e_sba_unmap(struct fore200e *fore200e)
 {
-    sbus_iounmap(fore200e->regs.sba.hcr, SBA200E_HCR_LENGTH);
-    sbus_iounmap(fore200e->regs.sba.bsr, SBA200E_BSR_LENGTH);
-    sbus_iounmap(fore200e->regs.sba.isr, SBA200E_ISR_LENGTH);
-    sbus_iounmap(fore200e->virt_base,    SBA200E_RAM_LENGTH);
-}
+	struct of_device *op = fore200e->bus_dev;
 
+	of_iounmap(&op->resource[0], fore200e->regs.sba.hcr, SBA200E_HCR_LENGTH);
+	of_iounmap(&op->resource[1], fore200e->regs.sba.bsr, SBA200E_BSR_LENGTH);
+	of_iounmap(&op->resource[2], fore200e->regs.sba.isr, SBA200E_ISR_LENGTH);
+	of_iounmap(&op->resource[3], fore200e->virt_base,    SBA200E_RAM_LENGTH);
+}
 
-static int __init
-fore200e_sba_configure(struct fore200e* fore200e)
+static int __init fore200e_sba_configure(struct fore200e *fore200e)
 {
-    fore200e->state = FORE200E_STATE_CONFIGURE;
-    return 0;
+	fore200e->state = FORE200E_STATE_CONFIGURE;
+	return 0;
 }
 
-
-static struct fore200e* __init
-fore200e_sba_detect(const struct fore200e_bus* bus, int index)
+static int __init fore200e_sba_prom_read(struct fore200e *fore200e, struct prom_data *prom)
 {
-    struct fore200e*          fore200e;
-    struct sbus_bus* sbus_bus;
-    struct sbus_dev* sbus_dev = NULL;
-    
-    unsigned int     count = 0;
-    
-    for_each_sbus (sbus_bus) {
-	for_each_sbusdev (sbus_dev, sbus_bus) {
-	    if (strcmp(sbus_dev->prom_name, SBA200E_PROM_NAME) == 0) {
-		if (count >= index)
-		    goto found;
-		count++;
-	    }
-	}
-    }
-    return NULL;
-    
-  found:
-    if (sbus_dev->num_registers != 4) {
-	printk(FORE200E "this %s device has %d instead of 4 registers\n",
-	       bus->model_name, sbus_dev->num_registers);
-	return NULL;
-    }
-
-    fore200e = kzalloc(sizeof(struct fore200e), GFP_KERNEL);
-    if (fore200e == NULL)
-	return NULL;
+	struct of_device *op = fore200e->bus_dev;
+	const u8 *prop;
+	int len;
 
-    fore200e->bus     = bus;
-    fore200e->bus_dev = sbus_dev;
-    fore200e->irq     = sbus_dev->irqs[ 0 ];
+	prop = of_get_property(op->node, "madaddrlo2", &len);
+	if (!prop)
+		return -ENODEV;
+	memcpy(&prom->mac_addr[4], prop, 4);
 
-    fore200e->phys_base = (unsigned long)sbus_dev;
+	prop = of_get_property(op->node, "madaddrhi4", &len);
+	if (!prop)
+		return -ENODEV;
+	memcpy(&prom->mac_addr[2], prop, 4);
 
-    sprintf(fore200e->name, "%s-%d", bus->model_name, index - 1);
+	prom->serial_number = of_getintprop_default(op->node, "serialnumber", 0);
+	prom->hw_revision = of_getintprop_default(op->node, "promversion", 0);
     
-    return fore200e;
+	return 0;
 }
 
-
-static int __init
-fore200e_sba_prom_read(struct fore200e* fore200e, struct prom_data* prom)
+static int fore200e_sba_proc_read(struct fore200e *fore200e, char *page)
 {
-    struct sbus_dev* sbus_dev = (struct sbus_dev*) fore200e->bus_dev;
-    int                       len;
-
-    len = prom_getproperty(sbus_dev->prom_node, "macaddrlo2", &prom->mac_addr[ 4 ], 4);
-    if (len < 0)
-	return -EBUSY;
-
-    len = prom_getproperty(sbus_dev->prom_node, "macaddrhi4", &prom->mac_addr[ 2 ], 4);
-    if (len < 0)
-	return -EBUSY;
-    
-    prom_getproperty(sbus_dev->prom_node, "serialnumber",
-		     (char*)&prom->serial_number, sizeof(prom->serial_number));
-    
-    prom_getproperty(sbus_dev->prom_node, "promversion",
-		     (char*)&prom->hw_revision, sizeof(prom->hw_revision));
-    
-    return 0;
-}
+	struct of_device *op = fore200e->bus_dev;
+	const struct linux_prom_registers *regs;
 
+	regs = of_get_property(op->node, "reg", NULL);
 
-static int
-fore200e_sba_proc_read(struct fore200e* fore200e, char *page)
-{
-    struct sbus_dev* sbus_dev = (struct sbus_dev*)fore200e->bus_dev;
-
-    return sprintf(page, "   SBUS slot/device:\t\t%d/'%s'\n", sbus_dev->slot, sbus_dev->prom_name);
+	return sprintf(page, "   SBUS slot/device:\t\t%d/'%s'\n",
+		       (regs ? regs->which_io : 0), op->node->name);
 }
 #endif /* CONFIG_SBUS */
 
@@ -2572,7 +2513,7 @@ fore200e_load_and_start_fw(struct fore200e* fore200e)
 	device = &((struct pci_dev *) fore200e->bus_dev)->dev;
 #ifdef CONFIG_SBUS
     else if (strcmp(fore200e->bus->model_name, "SBA-200E") == 0)
-	device = &((struct sbus_dev *) fore200e->bus_dev)->ofdev.dev;
+	device = &((struct of_device *) fore200e->bus_dev)->dev;
 #endif
     else
 	return err;
@@ -2701,6 +2642,66 @@ fore200e_init(struct fore200e* fore200e)
     return 0;
 }
 
+#ifdef CONFIG_SBUS
+static int __devinit fore200e_sba_probe(struct of_device *op,
+					const struct of_device_id *match)
+{
+	const struct fore200e_bus *bus = match->data;
+	struct fore200e *fore200e;
+	static int index = 0;
+	int err;
+
+	fore200e = kzalloc(sizeof(struct fore200e), GFP_KERNEL);
+	if (!fore200e)
+		return -ENOMEM;
+
+	fore200e->bus = bus;
+	fore200e->bus_dev = op;
+	fore200e->irq = op->irqs[0];
+	fore200e->phys_base = op->resource[0].start;
+
+	sprintf(fore200e->name, "%s-%d", bus->model_name, index);
+
+	err = fore200e_init(fore200e);
+	if (err < 0) {
+		fore200e_shutdown(fore200e);
+		kfree(fore200e);
+		return err;
+	}
+
+	index++;
+	dev_set_drvdata(&op->dev, fore200e);
+
+	return 0;
+}
+
+static int __devexit fore200e_sba_remove(struct of_device *op)
+{
+	struct fore200e *fore200e = dev_get_drvdata(&op->dev);
+
+	fore200e_shutdown(fore200e);
+	kfree(fore200e);
+
+	return 0;
+}
+
+static const struct of_device_id fore200e_sba_match[] = {
+	{
+		.name = SBA200E_PROM_NAME,
+		.data = (void *) &fore200e_bus[1],
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, fore200e_sba_match);
+
+static struct of_platform_driver fore200e_sba_driver = {
+	.name		= "fore_200e",
+	.match_table	= fore200e_sba_match,
+	.probe		= fore200e_sba_probe,
+	.remove		= __devexit_p(fore200e_sba_remove),
+};
+#endif
+
 #ifdef CONFIG_PCI
 static int __devinit
 fore200e_pca_detect(struct pci_dev *pci_dev, const struct pci_device_id *pci_ent)
@@ -2784,67 +2785,40 @@ static struct pci_driver fore200e_pca_driver = {
 };
 #endif
 
-
-static int __init
-fore200e_module_init(void)
+static int __init fore200e_module_init(void)
 {
-    const struct fore200e_bus* bus;
-    struct       fore200e*     fore200e;
-    int                        index;
-
-    printk(FORE200E "FORE Systems 200E-series ATM driver - version " FORE200E_VERSION "\n");
+	int err;
 
-    /* for each configured bus interface */
-    for (bus = fore200e_bus; bus->model_name; bus++) {
+	printk(FORE200E "FORE Systems 200E-series ATM driver - version " FORE200E_VERSION "\n");
 
-	/* detect all boards present on that bus */
-	for (index = 0; bus->detect && (fore200e = bus->detect(bus, index)); index++) {
-	    
-	    printk(FORE200E "device %s found at 0x%lx, IRQ %s\n",
-		   fore200e->bus->model_name, 
-		   fore200e->phys_base, fore200e_irq_itoa(fore200e->irq));
-
-	    sprintf(fore200e->name, "%s-%d", bus->model_name, index);
-
-	    if (fore200e_init(fore200e) < 0) {
-
-		fore200e_shutdown(fore200e);
-		break;
-	    }
-
-	    list_add(&fore200e->entry, &fore200e_boards);
-	}
-    }
+#ifdef CONFIG_SBUS
+	err = of_register_driver(&fore200e_sba_driver, &of_bus_type);
+	if (err)
+		return err;
+#endif
 
 #ifdef CONFIG_PCI
-    if (!pci_register_driver(&fore200e_pca_driver))
-	return 0;
+	err = pci_register_driver(&fore200e_pca_driver);
 #endif
 
-    if (!list_empty(&fore200e_boards))
-	return 0;
+#ifdef CONFIG_SBUS
+	if (err)
+		of_unregister_driver(&fore200e_sba_driver);
+#endif
 
-    return -ENODEV;
+	return err;
 }
 
-
-static void __exit
-fore200e_module_cleanup(void)
+static void __exit fore200e_module_cleanup(void)
 {
-    struct fore200e *fore200e, *next;
-
 #ifdef CONFIG_PCI
-    pci_unregister_driver(&fore200e_pca_driver);
+	pci_unregister_driver(&fore200e_pca_driver);
+#endif
+#ifdef CONFIG_SBUS
+	of_unregister_driver(&fore200e_sba_driver);
 #endif
-
-    list_for_each_entry_safe(fore200e, next, &fore200e_boards, entry) {
-	fore200e_shutdown(fore200e);
-	kfree(fore200e);
-    }
-    DPRINTK(1, "module being removed\n");
 }
 
-
 static int
 fore200e_proc_read(struct atm_dev *dev, loff_t* pos, char* page)
 {
@@ -3163,7 +3137,6 @@ static const struct fore200e_bus fore200e_bus[] = {
       fore200e_pca_dma_sync_for_device,
       fore200e_pca_dma_chunk_alloc,
       fore200e_pca_dma_chunk_free,
-      NULL,
       fore200e_pca_configure,
       fore200e_pca_map,
       fore200e_pca_reset,
@@ -3185,7 +3158,6 @@ static const struct fore200e_bus fore200e_bus[] = {
       fore200e_sba_dma_sync_for_device,
       fore200e_sba_dma_chunk_alloc,
       fore200e_sba_dma_chunk_free,
-      fore200e_sba_detect, 
       fore200e_sba_configure,
       fore200e_sba_map,
       fore200e_sba_reset,
diff --git a/drivers/atm/fore200e.h b/drivers/atm/fore200e.h
index 5c6e7adcb19c..7f97c09aaea5 100644
--- a/drivers/atm/fore200e.h
+++ b/drivers/atm/fore200e.h
@@ -778,9 +778,9 @@ typedef struct fore200e_pca_regs {
 /* SBA-200E registers */
 
 typedef struct fore200e_sba_regs {
-    volatile u32 __iomem *hcr;    /* address of host control register              */
-    volatile u32 __iomem *bsr;    /* address of burst transfer size register       */
-    volatile u32 __iomem *isr;    /* address of interrupt level selection register */
+    u32 __iomem *hcr;    /* address of host control register              */
+    u32 __iomem *bsr;    /* address of burst transfer size register       */
+    u32 __iomem *isr;    /* address of interrupt level selection register */
 } fore200e_sba_regs_t;
 
 
@@ -810,7 +810,6 @@ typedef struct fore200e_bus {
     void                 (*dma_sync_for_device)(struct fore200e*, u32, int, int);
     int                  (*dma_chunk_alloc)(struct fore200e*, struct chunk*, int, int, int);
     void                 (*dma_chunk_free)(struct fore200e*, struct chunk*);
-    struct fore200e*     (*detect)(const struct fore200e_bus*, int);
     int                  (*configure)(struct fore200e*); 
     int                  (*map)(struct fore200e*); 
     void                 (*reset)(struct fore200e*);
diff --git a/drivers/block/hd.c b/drivers/block/hd.c
index 682243bf2e46..482c0c4b964f 100644
--- a/drivers/block/hd.c
+++ b/drivers/block/hd.c
@@ -39,6 +39,7 @@
 #include <linux/ioport.h>
 #include <linux/init.h>
 #include <linux/blkpg.h>
+#include <linux/ata.h>
 #include <linux/hdreg.h>
 
 #define REALLY_SLOW_IO
@@ -370,7 +371,7 @@ repeat:
 		struct hd_i_struct *disk = &hd_info[i];
 		disk->special_op = disk->recalibrate = 1;
 		hd_out(disk, disk->sect, disk->sect, disk->head-1,
-			disk->cyl, WIN_SPECIFY, &reset_hd);
+			disk->cyl, ATA_CMD_INIT_DEV_PARAMS, &reset_hd);
 		if (reset)
 			goto repeat;
 	} else
@@ -558,7 +559,7 @@ static int do_special_op(struct hd_i_struct *disk, struct request *req)
 {
 	if (disk->recalibrate) {
 		disk->recalibrate = 0;
-		hd_out(disk, disk->sect, 0, 0, 0, WIN_RESTORE, &recal_intr);
+		hd_out(disk, disk->sect, 0, 0, 0, ATA_CMD_RESTORE, &recal_intr);
 		return reset;
 	}
 	if (disk->head > 16) {
@@ -631,13 +632,13 @@ repeat:
 	if (blk_fs_request(req)) {
 		switch (rq_data_dir(req)) {
 		case READ:
-			hd_out(disk, nsect, sec, head, cyl, WIN_READ,
+			hd_out(disk, nsect, sec, head, cyl, ATA_CMD_PIO_READ,
 				&read_intr);
 			if (reset)
 				goto repeat;
 			break;
 		case WRITE:
-			hd_out(disk, nsect, sec, head, cyl, WIN_WRITE,
+			hd_out(disk, nsect, sec, head, cyl, ATA_CMD_PIO_WRITE,
 				&write_intr);
 			if (reset)
 				goto repeat;
diff --git a/drivers/block/sunvdc.c b/drivers/block/sunvdc.c
index a8de037ecd4a..953c0b83d758 100644
--- a/drivers/block/sunvdc.c
+++ b/drivers/block/sunvdc.c
@@ -1,6 +1,6 @@
 /* sunvdc.c: Sun LDOM Virtual Disk Client.
  *
- * Copyright (C) 2007 David S. Miller <davem@davemloft.net>
+ * Copyright (C) 2007, 2008 David S. Miller <davem@davemloft.net>
  */
 
 #include <linux/module.h>
@@ -834,7 +834,7 @@ static int vdc_port_remove(struct vio_dev *vdev)
 	return 0;
 }
 
-static struct vio_device_id vdc_port_match[] = {
+static const struct vio_device_id vdc_port_match[] = {
 	{
 		.type = "vdc-port",
 	},
diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c
index bff602ccccf3..1a50ae70f716 100644
--- a/drivers/block/xen-blkfront.c
+++ b/drivers/block/xen-blkfront.c
@@ -1066,7 +1066,7 @@ static struct xenbus_driver blkfront = {
 
 static int __init xlblk_init(void)
 {
-	if (!is_running_on_xen())
+	if (!xen_domain())
 		return -ENODEV;
 
 	if (register_blkdev(XENVBD_MAJOR, DEV_NAME)) {
diff --git a/drivers/bluetooth/hci_ldisc.c b/drivers/bluetooth/hci_ldisc.c
index 8dfcf77cb717..4426bb552bd9 100644
--- a/drivers/bluetooth/hci_ldisc.c
+++ b/drivers/bluetooth/hci_ldisc.c
@@ -484,7 +484,7 @@ static int hci_uart_tty_ioctl(struct tty_struct *tty, struct file * file,
 		return -EUNATCH;
 
 	default:
-		err = n_tty_ioctl(tty, file, cmd, arg);
+		err = n_tty_ioctl_helper(tty, file, cmd, arg);
 		break;
 	};
 
diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index caff85149b9d..700ff9679457 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -350,7 +350,7 @@ config STALDRV
 
 config STALLION
 	tristate "Stallion EasyIO or EC8/32 support"
-	depends on STALDRV && BROKEN_ON_SMP && (ISA || EISA || PCI)
+	depends on STALDRV && (ISA || EISA || PCI)
 	help
 	  If you have an EasyIO or EasyConnection 8/32 multiport Stallion
 	  card, then this is for you; say Y.  Make sure to read
@@ -361,7 +361,7 @@ config STALLION
 
 config ISTALLION
 	tristate "Stallion EC8/64, ONboard, Brumby support"
-	depends on STALDRV && BROKEN_ON_SMP && (ISA || EISA || PCI)
+	depends on STALDRV && (ISA || EISA || PCI)
 	help
 	  If you have an EasyConnection 8/64, ONboard, Brumby or Stallion
 	  serial multiport card, say Y here. Make sure to read
diff --git a/drivers/char/Makefile b/drivers/char/Makefile
index 6850f6da7576..1a4247dccac4 100644
--- a/drivers/char/Makefile
+++ b/drivers/char/Makefile
@@ -7,7 +7,7 @@
 #
 FONTMAPFILE = cp437.uni
 
-obj-y	 += mem.o random.o tty_io.o n_tty.o tty_ioctl.o tty_ldisc.o 
+obj-y	 += mem.o random.o tty_io.o n_tty.o tty_ioctl.o tty_ldisc.o tty_buffer.o tty_port.o
 
 obj-$(CONFIG_LEGACY_PTYS)	+= pty.o
 obj-$(CONFIG_UNIX98_PTYS)	+= pty.o
diff --git a/drivers/char/amiserial.c b/drivers/char/amiserial.c
index 6e763e3f5a81..98821f97583c 100644
--- a/drivers/char/amiserial.c
+++ b/drivers/char/amiserial.c
@@ -837,9 +837,6 @@ static int rs_put_char(struct tty_struct *tty, unsigned char ch)
 	struct async_struct *info;
 	unsigned long flags;
 
-	if (!tty)
-		return 0;
-
 	info = tty->driver_data;
 
 	if (serial_paranoia_check(info, tty->name, "rs_put_char"))
@@ -892,9 +889,6 @@ static int rs_write(struct tty_struct * tty, const unsigned char *buf, int count
 	struct async_struct *info;
 	unsigned long flags;
 
-	if (!tty)
-		return 0;
-
 	info = tty->driver_data;
 
 	if (serial_paranoia_check(info, tty->name, "rs_write"))
diff --git a/drivers/char/applicom.c b/drivers/char/applicom.c
index 31d08b641f5b..b899d9182c7d 100644
--- a/drivers/char/applicom.c
+++ b/drivers/char/applicom.c
@@ -712,8 +712,7 @@ static int ac_ioctl(struct inode *inode, struct file *file, unsigned int cmd, un
 	
 	IndexCard = adgl->num_card-1;
 	 
-	if(cmd != 0 && cmd != 6 &&
-	   ((IndexCard >= MAX_BOARD) || !apbs[IndexCard].RamIO)) {
+	if(cmd != 6 && ((IndexCard >= MAX_BOARD) || !apbs[IndexCard].RamIO)) {
 		static int warncount = 10;
 		if (warncount) {
 			printk( KERN_WARNING "APPLICOM driver IOCTL, bad board number %d\n",(int)IndexCard+1);
@@ -832,8 +831,7 @@ static int ac_ioctl(struct inode *inode, struct file *file, unsigned int cmd, un
 		}
 		break;
 	default:
-		printk(KERN_INFO "APPLICOM driver ioctl, unknown function code %d\n",cmd) ;
-		ret = -EINVAL;
+		ret = -ENOTTY;
 		break;
 	}
 	Dummy = readb(apbs[IndexCard].RamIO + VERS);
diff --git a/drivers/char/cyclades.c b/drivers/char/cyclades.c
index fe6d774fe2e4..5e5b1dc1a0a7 100644
--- a/drivers/char/cyclades.c
+++ b/drivers/char/cyclades.c
@@ -4993,12 +4993,14 @@ static int __devinit cy_pci_probe(struct pci_dev *pdev,
 			device_id == PCI_DEVICE_ID_CYCLOM_Y_Hi) {
 		card_name = "Cyclom-Y";
 
-		addr0 = pci_iomap(pdev, 0, CyPCI_Yctl);
+		addr0 = ioremap_nocache(pci_resource_start(pdev, 0),
+				CyPCI_Yctl);
 		if (addr0 == NULL) {
 			dev_err(&pdev->dev, "can't remap ctl region\n");
 			goto err_reg;
 		}
-		addr2 = pci_iomap(pdev, 2, CyPCI_Ywin);
+		addr2 = ioremap_nocache(pci_resource_start(pdev, 2),
+				CyPCI_Ywin);
 		if (addr2 == NULL) {
 			dev_err(&pdev->dev, "can't remap base region\n");
 			goto err_unmap;
@@ -5013,7 +5015,8 @@ static int __devinit cy_pci_probe(struct pci_dev *pdev,
 	} else if (device_id == PCI_DEVICE_ID_CYCLOM_Z_Hi) {
 		struct RUNTIME_9060 __iomem *ctl_addr;
 
-		ctl_addr = addr0 = pci_iomap(pdev, 0, CyPCI_Zctl);
+		ctl_addr = addr0 = ioremap_nocache(pci_resource_start(pdev, 0),
+				CyPCI_Zctl);
 		if (addr0 == NULL) {
 			dev_err(&pdev->dev, "can't remap ctl region\n");
 			goto err_reg;
@@ -5026,8 +5029,8 @@ static int __devinit cy_pci_probe(struct pci_dev *pdev,
 
 		mailbox = (u32)readl(&ctl_addr->mail_box_0);
 
-		addr2 = pci_iomap(pdev, 2, mailbox == ZE_V1 ?
-				CyPCI_Ze_win : CyPCI_Zwin);
+		addr2 = ioremap_nocache(pci_resource_start(pdev, 2),
+				mailbox == ZE_V1 ? CyPCI_Ze_win : CyPCI_Zwin);
 		if (addr2 == NULL) {
 			dev_err(&pdev->dev, "can't remap base region\n");
 			goto err_unmap;
@@ -5159,9 +5162,9 @@ err_null:
 	cy_card[card_no].base_addr = NULL;
 	free_irq(irq, &cy_card[card_no]);
 err_unmap:
-	pci_iounmap(pdev, addr0);
+	iounmap(addr0);
 	if (addr2)
-		pci_iounmap(pdev, addr2);
+		iounmap(addr2);
 err_reg:
 	pci_release_regions(pdev);
 err_dis:
@@ -5186,9 +5189,9 @@ static void __devexit cy_pci_remove(struct pci_dev *pdev)
 		cy_writew(cinfo->ctl_addr + 0x68,
 				readw(cinfo->ctl_addr + 0x68) & ~0x0900);
 
-	pci_iounmap(pdev, cinfo->base_addr);
+	iounmap(cinfo->base_addr);
 	if (cinfo->ctl_addr)
-		pci_iounmap(pdev, cinfo->ctl_addr);
+		iounmap(cinfo->ctl_addr);
 	if (cinfo->irq
 #ifndef CONFIG_CYZ_INTR
 		&& !IS_CYC_Z(*cinfo)
diff --git a/drivers/char/epca.c b/drivers/char/epca.c
index 456e4ede049f..4998b2761e8f 100644
--- a/drivers/char/epca.c
+++ b/drivers/char/epca.c
@@ -1376,6 +1376,7 @@ static void post_fep_init(unsigned int crd)
 		unsigned long flags;
 		u16 tseg, rseg;
 
+		tty_port_init(&ch->port);
 		ch->brdchan = bc;
 		ch->mailbox = gd;
 		INIT_WORK(&ch->tqueue, do_softint);
@@ -1510,10 +1511,6 @@ static void post_fep_init(unsigned int crd)
 		ch->fepstopca = 0;
 
 		ch->close_delay = 50;
-		ch->port.count = 0;
-		ch->port.blocked_open = 0;
-		init_waitqueue_head(&ch->port.open_wait);
-		init_waitqueue_head(&ch->port.close_wait);
 
 		spin_unlock_irqrestore(&epca_lock, flags);
 	}
diff --git a/drivers/char/generic_serial.c b/drivers/char/generic_serial.c
index 19d3afb0e50c..c6090f84a2e4 100644
--- a/drivers/char/generic_serial.c
+++ b/drivers/char/generic_serial.c
@@ -54,8 +54,6 @@ int gs_put_char(struct tty_struct * tty, unsigned char ch)
 
 	func_enter (); 
 
-	if (!tty) return 0;
-
 	port = tty->driver_data;
 
 	if (!port) return 0;
@@ -97,8 +95,6 @@ int gs_write(struct tty_struct * tty,
 
 	func_enter ();
 
-	if (!tty) return 0;
-
 	port = tty->driver_data;
 
 	if (!port) return 0;
@@ -185,7 +181,6 @@ static int gs_real_chars_in_buffer(struct tty_struct *tty)
 	struct gs_port *port;
 	func_enter ();
 
-	if (!tty) return 0;
 	port = tty->driver_data;
 
 	if (!port->rd) return 0;
@@ -274,8 +269,6 @@ void gs_flush_buffer(struct tty_struct *tty)
 
 	func_enter ();
 
-	if (!tty) return;
-
 	port = tty->driver_data;
 
 	if (!port) return;
@@ -296,8 +289,6 @@ void gs_flush_chars(struct tty_struct * tty)
 
 	func_enter ();
 
-	if (!tty) return;
-
 	port = tty->driver_data;
 
 	if (!port) return;
@@ -321,8 +312,6 @@ void gs_stop(struct tty_struct * tty)
 
 	func_enter ();
 
-	if (!tty) return;
-
 	port = tty->driver_data;
 
 	if (!port) return;
@@ -341,8 +330,6 @@ void gs_start(struct tty_struct * tty)
 {
 	struct gs_port *port;
 
-	if (!tty) return;
-
 	port = tty->driver_data;
 
 	if (!port) return;
@@ -393,8 +380,6 @@ void gs_hangup(struct tty_struct *tty)
 
 	func_enter ();
 
-	if (!tty) return;
-
 	port = tty->driver_data;
 	tty = port->port.tty;
 	if (!tty) 
@@ -426,8 +411,6 @@ int gs_block_til_ready(void *port_, struct file * filp)
 
 	tty = port->port.tty;
 
-	if (!tty) return 0;
-
 	gs_dprintk (GS_DEBUG_BTR, "Entering gs_block_till_ready.\n"); 
 	/*
 	 * If the device is in the middle of being closed, then block
@@ -523,8 +506,6 @@ void gs_close(struct tty_struct * tty, struct file * filp)
 	
 	func_enter ();
 
-	if (!tty) return;
-
 	port = (struct gs_port *) tty->driver_data;
 
 	if (!port) return;
@@ -621,8 +602,6 @@ void gs_set_termios (struct tty_struct * tty,
 
 	func_enter();
 
-	if (!tty) return;
-
 	port = tty->driver_data;
 
 	if (!port) return;
diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c
index b3f5dbc6d880..f3cfb4c76125 100644
--- a/drivers/char/hpet.c
+++ b/drivers/char/hpet.c
@@ -53,6 +53,11 @@
 
 #define HPET_RANGE_SIZE		1024	/* from HPET spec */
 
+
+/* WARNING -- don't get confused.  These macros are never used
+ * to write the (single) counter, and rarely to read it.
+ * They're badly named; to fix, someday.
+ */
 #if BITS_PER_LONG == 64
 #define	write_counter(V, MC)	writeq(V, MC)
 #define	read_counter(MC)	readq(MC)
@@ -77,7 +82,7 @@ static struct clocksource clocksource_hpet = {
         .rating         = 250,
         .read           = read_hpet,
         .mask           = CLOCKSOURCE_MASK(64),
-        .mult           = 0, /*to be caluclated*/
+	.mult		= 0, /* to be calculated */
         .shift          = 10,
         .flags          = CLOCK_SOURCE_IS_CONTINUOUS,
 };
@@ -86,8 +91,6 @@ static struct clocksource *hpet_clocksource;
 
 /* A lock for concurrent access by app and isr hpet activity. */
 static DEFINE_SPINLOCK(hpet_lock);
-/* A lock for concurrent intermodule access to hpet and isr hpet activity. */
-static DEFINE_SPINLOCK(hpet_task_lock);
 
 #define	HPET_DEV_NAME	(7)
 
@@ -99,7 +102,6 @@ struct hpet_dev {
 	unsigned long hd_irqdata;
 	wait_queue_head_t hd_waitqueue;
 	struct fasync_struct *hd_async_queue;
-	struct hpet_task *hd_task;
 	unsigned int hd_flags;
 	unsigned int hd_irq;
 	unsigned int hd_hdwirq;
@@ -173,11 +175,6 @@ static irqreturn_t hpet_interrupt(int irq, void *data)
 		writel(isr, &devp->hd_hpet->hpet_isr);
 	spin_unlock(&hpet_lock);
 
-	spin_lock(&hpet_task_lock);
-	if (devp->hd_task)
-		devp->hd_task->ht_func(devp->hd_task->ht_data);
-	spin_unlock(&hpet_task_lock);
-
 	wake_up_interruptible(&devp->hd_waitqueue);
 
 	kill_fasync(&devp->hd_async_queue, SIGIO, POLL_IN);
@@ -185,6 +182,67 @@ static irqreturn_t hpet_interrupt(int irq, void *data)
 	return IRQ_HANDLED;
 }
 
+static void hpet_timer_set_irq(struct hpet_dev *devp)
+{
+	unsigned long v;
+	int irq, gsi;
+	struct hpet_timer __iomem *timer;
+
+	spin_lock_irq(&hpet_lock);
+	if (devp->hd_hdwirq) {
+		spin_unlock_irq(&hpet_lock);
+		return;
+	}
+
+	timer = devp->hd_timer;
+
+	/* we prefer level triggered mode */
+	v = readl(&timer->hpet_config);
+	if (!(v & Tn_INT_TYPE_CNF_MASK)) {
+		v |= Tn_INT_TYPE_CNF_MASK;
+		writel(v, &timer->hpet_config);
+	}
+	spin_unlock_irq(&hpet_lock);
+
+	v = (readq(&timer->hpet_config) & Tn_INT_ROUTE_CAP_MASK) >>
+				 Tn_INT_ROUTE_CAP_SHIFT;
+
+	/*
+	 * In PIC mode, skip IRQ0-4, IRQ6-9, IRQ12-15 which is always used by
+	 * legacy device. In IO APIC mode, we skip all the legacy IRQS.
+	 */
+	if (acpi_irq_model == ACPI_IRQ_MODEL_PIC)
+		v &= ~0xf3df;
+	else
+		v &= ~0xffff;
+
+	for (irq = find_first_bit(&v, HPET_MAX_IRQ); irq < HPET_MAX_IRQ;
+		irq = find_next_bit(&v, HPET_MAX_IRQ, 1 + irq)) {
+
+		if (irq >= NR_IRQS) {
+			irq = HPET_MAX_IRQ;
+			break;
+		}
+
+		gsi = acpi_register_gsi(irq, ACPI_LEVEL_SENSITIVE,
+					ACPI_ACTIVE_LOW);
+		if (gsi > 0)
+			break;
+
+		/* FIXME: Setup interrupt source table */
+	}
+
+	if (irq < HPET_MAX_IRQ) {
+		spin_lock_irq(&hpet_lock);
+		v = readl(&timer->hpet_config);
+		v |= irq << Tn_INT_ROUTE_CNF_SHIFT;
+		writel(v, &timer->hpet_config);
+		devp->hd_hdwirq = gsi;
+		spin_unlock_irq(&hpet_lock);
+	}
+	return;
+}
+
 static int hpet_open(struct inode *inode, struct file *file)
 {
 	struct hpet_dev *devp;
@@ -199,8 +257,7 @@ static int hpet_open(struct inode *inode, struct file *file)
 
 	for (devp = NULL, hpetp = hpets; hpetp && !devp; hpetp = hpetp->hp_next)
 		for (i = 0; i < hpetp->hp_ntimer; i++)
-			if (hpetp->hp_dev[i].hd_flags & HPET_OPEN
-			    || hpetp->hp_dev[i].hd_task)
+			if (hpetp->hp_dev[i].hd_flags & HPET_OPEN)
 				continue;
 			else {
 				devp = &hpetp->hp_dev[i];
@@ -219,6 +276,8 @@ static int hpet_open(struct inode *inode, struct file *file)
 	spin_unlock_irq(&hpet_lock);
 	unlock_kernel();
 
+	hpet_timer_set_irq(devp);
+
 	return 0;
 }
 
@@ -441,7 +500,11 @@ static int hpet_ioctl_ieon(struct hpet_dev *devp)
 	devp->hd_irq = irq;
 	t = devp->hd_ireqfreq;
 	v = readq(&timer->hpet_config);
-	g = v | Tn_INT_ENB_CNF_MASK;
+
+	/* 64-bit comparators are not yet supported through the ioctls,
+	 * so force this into 32-bit mode if it supports both modes
+	 */
+	g = v | Tn_32MODE_CNF_MASK | Tn_INT_ENB_CNF_MASK;
 
 	if (devp->hd_flags & HPET_PERIODIC) {
 		write_counter(t, &timer->hpet_compare);
@@ -451,6 +514,12 @@ static int hpet_ioctl_ieon(struct hpet_dev *devp)
 		v |= Tn_VAL_SET_CNF_MASK;
 		writeq(v, &timer->hpet_config);
 		local_irq_save(flags);
+
+		/* NOTE:  what we modify here is a hidden accumulator
+		 * register supported by periodic-capable comparators.
+		 * We never want to modify the (single) counter; that
+		 * would affect all the comparators.
+		 */
 		m = read_counter(&hpet->hpet_mc);
 		write_counter(t + m + hpetp->hp_delta, &timer->hpet_compare);
 	} else {
@@ -604,57 +673,6 @@ static int hpet_is_known(struct hpet_data *hdp)
 	return 0;
 }
 
-static inline int hpet_tpcheck(struct hpet_task *tp)
-{
-	struct hpet_dev *devp;
-	struct hpets *hpetp;
-
-	devp = tp->ht_opaque;
-
-	if (!devp)
-		return -ENXIO;
-
-	for (hpetp = hpets; hpetp; hpetp = hpetp->hp_next)
-		if (devp >= hpetp->hp_dev
-		    && devp < (hpetp->hp_dev + hpetp->hp_ntimer)
-		    && devp->hd_hpet == hpetp->hp_hpet)
-			return 0;
-
-	return -ENXIO;
-}
-
-#if 0
-int hpet_unregister(struct hpet_task *tp)
-{
-	struct hpet_dev *devp;
-	struct hpet_timer __iomem *timer;
-	int err;
-
-	if ((err = hpet_tpcheck(tp)))
-		return err;
-
-	spin_lock_irq(&hpet_task_lock);
-	spin_lock(&hpet_lock);
-
-	devp = tp->ht_opaque;
-	if (devp->hd_task != tp) {
-		spin_unlock(&hpet_lock);
-		spin_unlock_irq(&hpet_task_lock);
-		return -ENXIO;
-	}
-
-	timer = devp->hd_timer;
-	writeq((readq(&timer->hpet_config) & ~Tn_INT_ENB_CNF_MASK),
-	       &timer->hpet_config);
-	devp->hd_flags &= ~(HPET_IE | HPET_PERIODIC);
-	devp->hd_task = NULL;
-	spin_unlock(&hpet_lock);
-	spin_unlock_irq(&hpet_task_lock);
-
-	return 0;
-}
-#endif  /*  0  */
-
 static ctl_table hpet_table[] = {
 	{
 	 .ctl_name = CTL_UNNUMBERED,
@@ -746,6 +764,7 @@ int hpet_alloc(struct hpet_data *hdp)
 	static struct hpets *last = NULL;
 	unsigned long period;
 	unsigned long long temp;
+	u32 remainder;
 
 	/*
 	 * hpet_alloc can be called by platform dependent code.
@@ -809,9 +828,13 @@ int hpet_alloc(struct hpet_data *hdp)
 		printk("%s %d", i > 0 ? "," : "", hdp->hd_irq[i]);
 	printk("\n");
 
-	printk(KERN_INFO "hpet%u: %u %d-bit timers, %Lu Hz\n",
-	       hpetp->hp_which, hpetp->hp_ntimer,
-	       cap & HPET_COUNTER_SIZE_MASK ? 64 : 32, hpetp->hp_tick_freq);
+	temp = hpetp->hp_tick_freq;
+	remainder = do_div(temp, 1000000);
+	printk(KERN_INFO
+		"hpet%u: %u comparators, %d-bit %u.%06u MHz counter\n",
+		hpetp->hp_which, hpetp->hp_ntimer,
+		cap & HPET_COUNTER_SIZE_MASK ? 64 : 32,
+		(unsigned) temp, remainder);
 
 	mcfg = readq(&hpet->hpet_config);
 	if ((mcfg & HPET_ENABLE_CNF_MASK) == 0) {
@@ -874,8 +897,6 @@ static acpi_status hpet_resources(struct acpi_resource *res, void *data)
 		hdp->hd_address = ioremap(addr.minimum, addr.address_length);
 
 		if (hpet_is_known(hdp)) {
-			printk(KERN_DEBUG "%s: 0x%lx is busy\n",
-				__func__, hdp->hd_phys_address);
 			iounmap(hdp->hd_address);
 			return AE_ALREADY_EXISTS;
 		}
@@ -891,8 +912,6 @@ static acpi_status hpet_resources(struct acpi_resource *res, void *data)
 						HPET_RANGE_SIZE);
 
 		if (hpet_is_known(hdp)) {
-			printk(KERN_DEBUG "%s: 0x%lx is busy\n",
-				__func__, hdp->hd_phys_address);
 			iounmap(hdp->hd_address);
 			return AE_ALREADY_EXISTS;
 		}
diff --git a/drivers/char/hvc_console.c b/drivers/char/hvc_console.c
index fd64137b1ab9..ec7aded0a2df 100644
--- a/drivers/char/hvc_console.c
+++ b/drivers/char/hvc_console.c
@@ -819,11 +819,11 @@ static int hvc_init(void)
 	hvc_driver = drv;
 	return 0;
 
-put_tty:
-	put_tty_driver(hvc_driver);
 stop_thread:
 	kthread_stop(hvc_task);
 	hvc_task = NULL;
+put_tty:
+	put_tty_driver(drv);
 out:
 	return err;
 }
diff --git a/drivers/char/hvc_xen.c b/drivers/char/hvc_xen.c
index 6b70aa66a587..538ceea5e7df 100644
--- a/drivers/char/hvc_xen.c
+++ b/drivers/char/hvc_xen.c
@@ -108,8 +108,8 @@ static int __init xen_init(void)
 {
 	struct hvc_struct *hp;
 
-	if (!is_running_on_xen() ||
-	    is_initial_xendomain() ||
+	if (!xen_pv_domain() ||
+	    xen_initial_domain() ||
 	    !xen_start_info->console.domU.evtchn)
 		return -ENODEV;
 
@@ -142,7 +142,7 @@ static void __exit xen_fini(void)
 
 static int xen_cons_init(void)
 {
-	if (!is_running_on_xen())
+	if (!xen_pv_domain())
 		return 0;
 
 	hvc_instantiate(HVC_COOKIE, 0, &hvc_ops);
diff --git a/drivers/char/hw_random/n2-drv.c b/drivers/char/hw_random/n2-drv.c
index 5220f541df25..8859aeac2d25 100644
--- a/drivers/char/hw_random/n2-drv.c
+++ b/drivers/char/hw_random/n2-drv.c
@@ -736,7 +736,7 @@ static int __devexit n2rng_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id n2rng_match[] = {
+static const struct of_device_id n2rng_match[] = {
 	{
 		.name		= "random-number-generator",
 		.compatible	= "SUNW,n2-rng",
diff --git a/drivers/char/ip2/Makefile b/drivers/char/ip2/Makefile
index 939618f62fe1..bc397d92b499 100644
--- a/drivers/char/ip2/Makefile
+++ b/drivers/char/ip2/Makefile
@@ -4,5 +4,5 @@
 
 obj-$(CONFIG_COMPUTONE)         += ip2.o
 
-ip2-objs			:= ip2base.o ip2main.o
+ip2-objs			:= ip2main.o
 
diff --git a/drivers/char/ip2/i2ellis.c b/drivers/char/ip2/i2ellis.c
index 3601017f58cf..29db44de399f 100644
--- a/drivers/char/ip2/i2ellis.c
+++ b/drivers/char/ip2/i2ellis.c
@@ -69,38 +69,6 @@ static DEFINE_RWLOCK(Dl_spinlock);
 //=======================================================
 
 //******************************************************************************
-// Function:   iiEllisInit()
-// Parameters: None
-//
-// Returns:    Nothing
-//
-// Description:
-//
-// This routine performs any required initialization of the iiEllis subsystem.
-//
-//******************************************************************************
-static void
-iiEllisInit(void)
-{
-}
-
-//******************************************************************************
-// Function:   iiEllisCleanup()
-// Parameters: None
-//
-// Returns:    Nothing
-//
-// Description:
-//
-// This routine performs any required cleanup of the iiEllis subsystem.
-//
-//******************************************************************************
-static void
-iiEllisCleanup(void)
-{
-}
-
-//******************************************************************************
 // Function:   iiSetAddress(pB, address, delay)
 // Parameters: pB      - pointer to the board structure
 //             address - the purported I/O address of the board
diff --git a/drivers/char/ip2/i2ellis.h b/drivers/char/ip2/i2ellis.h
index c88a64e527aa..fb6df2456018 100644
--- a/drivers/char/ip2/i2ellis.h
+++ b/drivers/char/ip2/i2ellis.h
@@ -511,7 +511,6 @@ typedef void (*delayFunc_t)(unsigned int);
 //
 // Initialization of a board & structure is in four (five!) parts:
 //
-// 0) iiEllisInit()  - Initialize iiEllis subsystem.
 // 1) iiSetAddress() - Define the board address & delay function for a board.
 // 2) iiReset()      - Reset the board   (provided it exists)
 //       -- Note you may do this to several boards --
@@ -523,7 +522,6 @@ typedef void (*delayFunc_t)(unsigned int);
 // loadware.  To change loadware, you must begin again with step 2, resetting
 // the board again (step 1 not needed).
 
-static void iiEllisInit(void);
 static int iiSetAddress(i2eBordStrPtr, int, delayFunc_t );
 static int iiReset(i2eBordStrPtr);
 static int iiResetDelay(i2eBordStrPtr);
diff --git a/drivers/char/ip2/ip2base.c b/drivers/char/ip2/ip2base.c
deleted file mode 100644
index 8155e247c04b..000000000000
--- a/drivers/char/ip2/ip2base.c
+++ /dev/null
@@ -1,108 +0,0 @@
-// ip2.c
-// This is a dummy module to make the firmware available when needed
-// and allows it to be unloaded when not. Rumor is the __initdata 
-// macro doesn't always works on all platforms so we use this kludge.
-// If not compiled as a module it just makes fip_firm avaliable then
-//  __initdata should work as advertized
-//
-
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/wait.h>
-
-#ifndef __init
-#define __init
-#endif
-#ifndef __initfunc
-#define __initfunc(a) a
-#endif
-#ifndef __initdata
-#define __initdata
-#endif
-
-#include "ip2types.h"		
-
-int
-ip2_loadmain(int *, int *); // ref into ip2main.c
-
-/* Note: Add compiled in defaults to these arrays, not to the structure
-	in ip2.h any longer.  That structure WILL get overridden
-	by these values, or command line values, or insmod values!!!  =mhw=
-*/
-static int io[IP2_MAX_BOARDS]= { 0, 0, 0, 0 };
-static int irq[IP2_MAX_BOARDS] = { -1, -1, -1, -1 }; 
-
-static int poll_only = 0;
-
-MODULE_AUTHOR("Doug McNash");
-MODULE_DESCRIPTION("Computone IntelliPort Plus Driver");
-module_param_array(irq, int, NULL, 0);
-MODULE_PARM_DESC(irq,"Interrupts for IntelliPort Cards");
-module_param_array(io, int, NULL, 0);
-MODULE_PARM_DESC(io,"I/O ports for IntelliPort Cards");
-module_param(poll_only, bool, 0);
-MODULE_PARM_DESC(poll_only,"Do not use card interrupts");
-
-
-static int __init ip2_init(void)
-{
-	if( poll_only ) {
-		/* Hard lock the interrupts to zero */
-		irq[0] = irq[1] = irq[2] = irq[3] = 0;
-	}
-
-	return ip2_loadmain(io, irq);
-}
-module_init(ip2_init);
-
-MODULE_LICENSE("GPL");
-
-#ifndef MODULE
-/******************************************************************************
- *	ip2_setup:
- *		str: kernel command line string
- *
- *	Can't autoprobe the boards so user must specify configuration on
- *	kernel command line.  Sane people build it modular but the others
- *	come here.
- *
- *	Alternating pairs of io,irq for up to 4 boards.
- *		ip2=io0,irq0,io1,irq1,io2,irq2,io3,irq3
- *
- *		io=0 => No board
- *		io=1 => PCI
- *		io=2 => EISA
- *		else => ISA I/O address
- *
- *		irq=0 or invalid for ISA will revert to polling mode
- *
- *		Any value = -1, do not overwrite compiled in value.
- *
- ******************************************************************************/
-static int __init ip2_setup(char *str)
-{
-	int	ints[10];	/* 4 boards, 2 parameters + 2 */
-	int	i, j;
-
-	str = get_options (str, ARRAY_SIZE(ints), ints);
-
-	for( i = 0, j = 1; i < 4; i++ ) {
-		if( j > ints[0] ) {
-			break;
-		}
-		if( ints[j] >= 0 ) {
-			io[i] = ints[j];
-		}
-		j++;
-		if( j > ints[0] ) {
-			break;
-		}
-		if( ints[j] >= 0 ) {
-			irq[i] = ints[j];
-		}
-		j++;
-	}
-	return 1;
-}
-__setup("ip2=", ip2_setup);
-#endif /* !MODULE */
diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c
index 689f9dcd3b86..6774572d3759 100644
--- a/drivers/char/ip2/ip2main.c
+++ b/drivers/char/ip2/ip2main.c
@@ -150,15 +150,12 @@ static int ip2_read_proc(char *, char **, off_t, int, int *, void * );
 /*************/
 
 /* String constants to identify ourselves */
-static char *pcName    = "Computone IntelliPort Plus multiport driver";
-static char *pcVersion = "1.2.14";
+static const char pcName[] = "Computone IntelliPort Plus multiport driver";
+static const char pcVersion[] = "1.2.14";
 
 /* String constants for port names */
-static char *pcDriver_name   = "ip2";
-static char *pcIpl    		 = "ip2ipl";
-
-// cheezy kludge or genius - you decide?
-int ip2_loadmain(int *, int *);
+static const char pcDriver_name[] = "ip2";
+static const char pcIpl[] = "ip2ipl";
 
 /***********************/
 /* Function Prototypes */
@@ -240,8 +237,8 @@ static const struct file_operations ip2_ipl = {
 	.open		= ip2_ipl_open,
 }; 
 
-static unsigned long irq_counter = 0;
-static unsigned long bh_counter = 0;
+static unsigned long irq_counter;
+static unsigned long bh_counter;
 
 // Use immediate queue to service interrupts
 #define USE_IQI
@@ -252,7 +249,6 @@ static unsigned long bh_counter = 0;
  */
 #define  POLL_TIMEOUT   (jiffies + 1)
 static DEFINE_TIMER(PollTimer, ip2_poll, 0, 0);
-static char  TimerOn;
 
 #ifdef IP2DEBUG_TRACE
 /* Trace (debug) buffer data */
@@ -268,8 +264,8 @@ static int tracewrap;
 /**********/
 
 #if defined(MODULE) && defined(IP2DEBUG_OPEN)
-#define DBG_CNT(s) printk(KERN_DEBUG "(%s): [%x] refc=%d, ttyc=%d, modc=%x -> %s\n", \
-		    tty->name,(pCh->flags),ip2_tty_driver->refcount, \
+#define DBG_CNT(s) printk(KERN_DEBUG "(%s): [%x] ttyc=%d, modc=%x -> %s\n", \
+		    tty->name,(pCh->flags), \
 		    tty->count,/*GET_USE_COUNT(module)*/0,s)
 #else
 #define DBG_CNT(s)
@@ -287,8 +283,9 @@ static int tracewrap;
 
 MODULE_AUTHOR("Doug McNash");
 MODULE_DESCRIPTION("Computone IntelliPort Plus Driver");
+MODULE_LICENSE("GPL");
 
-static int poll_only = 0;
+static int poll_only;
 
 static int Eisa_irq;
 static int Eisa_slot;
@@ -297,34 +294,46 @@ static int iindx;
 static char rirqs[IP2_MAX_BOARDS];
 static int Valid_Irqs[] = { 3, 4, 5, 7, 10, 11, 12, 15, 0};
 
+/* Note: Add compiled in defaults to these arrays, not to the structure
+	in ip2.h any longer.  That structure WILL get overridden
+	by these values, or command line values, or insmod values!!!  =mhw=
+*/
+static int io[IP2_MAX_BOARDS];
+static int irq[IP2_MAX_BOARDS] = { -1, -1, -1, -1 };
+
+MODULE_AUTHOR("Doug McNash");
+MODULE_DESCRIPTION("Computone IntelliPort Plus Driver");
+module_param_array(irq, int, NULL, 0);
+MODULE_PARM_DESC(irq, "Interrupts for IntelliPort Cards");
+module_param_array(io, int, NULL, 0);
+MODULE_PARM_DESC(io, "I/O ports for IntelliPort Cards");
+module_param(poll_only, bool, 0);
+MODULE_PARM_DESC(poll_only, "Do not use card interrupts");
+
 /* for sysfs class support */
 static struct class *ip2_class;
 
-// Some functions to keep track of what irq's we have
+/* Some functions to keep track of what irqs we have */
 
-static int
-is_valid_irq(int irq)
+static int __init is_valid_irq(int irq)
 {
 	int *i = Valid_Irqs;
 	
-	while ((*i != 0) && (*i != irq)) {
+	while (*i != 0 && *i != irq)
 		i++;
-	}
-	return (*i);
+
+	return *i;
 }
 
-static void
-mark_requested_irq( char irq )
+static void __init mark_requested_irq(char irq)
 {
 	rirqs[iindx++] = irq;
 }
 
-#ifdef MODULE
-static int
-clear_requested_irq( char irq )
+static int __exit clear_requested_irq(char irq)
 {
 	int i;
-	for ( i = 0; i < IP2_MAX_BOARDS; ++i ) {
+	for (i = 0; i < IP2_MAX_BOARDS; ++i) {
 		if (rirqs[i] == irq) {
 			rirqs[i] = 0;
 			return 1;
@@ -332,17 +341,15 @@ clear_requested_irq( char irq )
 	}
 	return 0;
 }
-#endif
 
-static int
-have_requested_irq( char irq )
+static int have_requested_irq(char irq)
 {
-	// array init to zeros so 0 irq will not be requested as a side effect
+	/* array init to zeros so 0 irq will not be requested as a side
+	 * effect */
 	int i;
-	for ( i = 0; i < IP2_MAX_BOARDS; ++i ) {
+	for (i = 0; i < IP2_MAX_BOARDS; ++i)
 		if (rirqs[i] == irq)
 			return 1;
-	}
 	return 0;
 }
 
@@ -361,53 +368,45 @@ have_requested_irq( char irq )
 /* handle subsequent installations of the driver. All memory allocated by the */
 /* driver should be returned since it may be unloaded from memory.            */
 /******************************************************************************/
-#ifdef MODULE
-void __exit
-ip2_cleanup_module(void)
+static void __exit ip2_cleanup_module(void)
 {
 	int err;
 	int i;
 
-#ifdef IP2DEBUG_INIT
-	printk (KERN_DEBUG "Unloading %s: version %s\n", pcName, pcVersion );
-#endif
-	/* Stop poll timer if we had one. */
-	if ( TimerOn ) {
-		del_timer ( &PollTimer );
-		TimerOn = 0;
-	}
+	del_timer_sync(&PollTimer);
 
 	/* Reset the boards we have. */
-	for( i = 0; i < IP2_MAX_BOARDS; ++i ) {
-		if ( i2BoardPtrTable[i] ) {
-			iiReset( i2BoardPtrTable[i] );
-		}
-	}
+	for (i = 0; i < IP2_MAX_BOARDS; i++)
+		if (i2BoardPtrTable[i])
+			iiReset(i2BoardPtrTable[i]);
 
 	/* The following is done at most once, if any boards were installed. */
-	for ( i = 0; i < IP2_MAX_BOARDS; ++i ) {
-		if ( i2BoardPtrTable[i] ) {
-			iiResetDelay( i2BoardPtrTable[i] );
+	for (i = 0; i < IP2_MAX_BOARDS; i++) {
+		if (i2BoardPtrTable[i]) {
+			iiResetDelay(i2BoardPtrTable[i]);
 			/* free io addresses and Tibet */
-			release_region( ip2config.addr[i], 8 );
+			release_region(ip2config.addr[i], 8);
 			device_destroy(ip2_class, MKDEV(IP2_IPL_MAJOR, 4 * i));
-			device_destroy(ip2_class, MKDEV(IP2_IPL_MAJOR, 4 * i + 1));
+			device_destroy(ip2_class, MKDEV(IP2_IPL_MAJOR,
+						4 * i + 1));
 		}
 		/* Disable and remove interrupt handler. */
-		if ( (ip2config.irq[i] > 0) && have_requested_irq(ip2config.irq[i]) ) {	
-			free_irq ( ip2config.irq[i], (void *)&pcName);
-			clear_requested_irq( ip2config.irq[i]);
+		if (ip2config.irq[i] > 0 &&
+				have_requested_irq(ip2config.irq[i])) {
+			free_irq(ip2config.irq[i], (void *)&pcName);
+			clear_requested_irq(ip2config.irq[i]);
 		}
 	}
 	class_destroy(ip2_class);
-	if ( ( err = tty_unregister_driver ( ip2_tty_driver ) ) ) {
-		printk(KERN_ERR "IP2: failed to unregister tty driver (%d)\n", err);
-	}
+	err = tty_unregister_driver(ip2_tty_driver);
+	if (err)
+		printk(KERN_ERR "IP2: failed to unregister tty driver (%d)\n",
+				err);
 	put_tty_driver(ip2_tty_driver);
 	unregister_chrdev(IP2_IPL_MAJOR, pcIpl);
 	remove_proc_entry("ip2mem", NULL);
 
-	// free memory
+	/* free memory */
 	for (i = 0; i < IP2_MAX_BOARDS; i++) {
 		void *pB;
 #ifdef CONFIG_PCI
@@ -417,24 +416,18 @@ ip2_cleanup_module(void)
 			ip2config.pci_dev[i] = NULL;
 		}
 #endif
-		if ((pB = i2BoardPtrTable[i]) != 0 ) {
-			kfree ( pB );
+		pB = i2BoardPtrTable[i];
+		if (pB != NULL) {
+			kfree(pB);
 			i2BoardPtrTable[i] = NULL;
 		}
-		if ((DevTableMem[i]) != NULL ) {
-			kfree ( DevTableMem[i]  );
+		if (DevTableMem[i] != NULL) {
+			kfree(DevTableMem[i]);
 			DevTableMem[i] = NULL;
 		}
 	}
-
-	/* Cleanup the iiEllis subsystem. */
-	iiEllisCleanup();
-#ifdef IP2DEBUG_INIT
-	printk (KERN_DEBUG "IP2 Unloaded\n" );
-#endif
 }
 module_exit(ip2_cleanup_module);
-#endif /* MODULE */
 
 static const struct tty_operations ip2_ops = {
 	.open            = ip2_open,
@@ -494,139 +487,168 @@ static const struct firmware *ip2_request_firmware(void)
 	return fw;
 }
 
-int
-ip2_loadmain(int *iop, int *irqp)
+#ifndef MODULE
+/******************************************************************************
+ *	ip2_setup:
+ *		str: kernel command line string
+ *
+ *	Can't autoprobe the boards so user must specify configuration on
+ *	kernel command line.  Sane people build it modular but the others
+ *	come here.
+ *
+ *	Alternating pairs of io,irq for up to 4 boards.
+ *		ip2=io0,irq0,io1,irq1,io2,irq2,io3,irq3
+ *
+ *		io=0 => No board
+ *		io=1 => PCI
+ *		io=2 => EISA
+ *		else => ISA I/O address
+ *
+ *		irq=0 or invalid for ISA will revert to polling mode
+ *
+ *		Any value = -1, do not overwrite compiled in value.
+ *
+ ******************************************************************************/
+static int __init ip2_setup(char *str)
+{
+	int j, ints[10];	/* 4 boards, 2 parameters + 2 */
+	unsigned int i;
+
+	str = get_options(str, ARRAY_SIZE(ints), ints);
+
+	for (i = 0, j = 1; i < 4; i++) {
+		if (j > ints[0])
+			break;
+		if (ints[j] >= 0)
+			io[i] = ints[j];
+		j++;
+		if (j > ints[0])
+			break;
+		if (ints[j] >= 0)
+			irq[i] = ints[j];
+		j++;
+	}
+	return 1;
+}
+__setup("ip2=", ip2_setup);
+#endif /* !MODULE */
+
+static int __init ip2_loadmain(void)
 {
 	int i, j, box;
 	int err = 0;
-	static int loaded;
 	i2eBordStrPtr pB = NULL;
 	int rc = -1;
-	static struct pci_dev *pci_dev_i = NULL;
+	struct pci_dev *pdev = NULL;
 	const struct firmware *fw = NULL;
 
-	ip2trace (ITRC_NO_PORT, ITRC_INIT, ITRC_ENTER, 0 );
+	if (poll_only) {
+		/* Hard lock the interrupts to zero */
+		irq[0] = irq[1] = irq[2] = irq[3] = poll_only = 0;
+	}
+
+	ip2trace(ITRC_NO_PORT, ITRC_INIT, ITRC_ENTER, 0);
 
 	/* process command line arguments to modprobe or
 		insmod i.e. iop & irqp */
 	/* irqp and iop should ALWAYS be specified now...  But we check
 		them individually just to be sure, anyways... */
-	for ( i = 0; i < IP2_MAX_BOARDS; ++i ) {
-		if (iop) {
-			ip2config.addr[i] = iop[i];
-			if (irqp) {
-				if( irqp[i] >= 0 ) {
-					ip2config.irq[i] = irqp[i];
-				} else {
-					ip2config.irq[i] = 0;
-				}
-	// This is a little bit of a hack.  If poll_only=1 on command
-	// line back in ip2.c OR all IRQs on all specified boards are
-	// explicitly set to 0, then drop to poll only mode and override
-	// PCI or EISA interrupts.  This superceeds the old hack of
-	// triggering if all interrupts were zero (like da default).
-	// Still a hack but less prone to random acts of terrorism.
-	//
-	// What we really should do, now that the IRQ default is set
-	// to -1, is to use 0 as a hard coded, do not probe.
-	//
-	//	/\/\|=mhw=|\/\/
-				poll_only |= irqp[i];
-			}
-		}
+	for (i = 0; i < IP2_MAX_BOARDS; ++i) {
+		ip2config.addr[i] = io[i];
+		if (irq[i] >= 0)
+			ip2config.irq[i] = irq[i];
+		else
+			ip2config.irq[i] = 0;
+	/* This is a little bit of a hack.  If poll_only=1 on command
+	   line back in ip2.c OR all IRQs on all specified boards are
+	   explicitly set to 0, then drop to poll only mode and override
+	   PCI or EISA interrupts.  This superceeds the old hack of
+	   triggering if all interrupts were zero (like da default).
+	   Still a hack but less prone to random acts of terrorism.
+
+	   What we really should do, now that the IRQ default is set
+	   to -1, is to use 0 as a hard coded, do not probe.
+
+		/\/\|=mhw=|\/\/
+	*/
+		poll_only |= irq[i];
 	}
 	poll_only = !poll_only;
 
 	/* Announce our presence */
-	printk( KERN_INFO "%s version %s\n", pcName, pcVersion );
-
-	// ip2 can be unloaded and reloaded for no good reason
-	// we can't let that happen here or bad things happen
-	// second load hoses board but not system - fixme later
-	if (loaded) {
-		printk( KERN_INFO "Still loaded\n" );
-		return 0;
-	}
-	loaded++;
+	printk(KERN_INFO "%s version %s\n", pcName, pcVersion);
 
 	ip2_tty_driver = alloc_tty_driver(IP2_MAX_PORTS);
 	if (!ip2_tty_driver)
 		return -ENOMEM;
 
-	/* Initialise the iiEllis subsystem. */
-	iiEllisInit();
-
-	/* Initialize arrays. */
-	memset( i2BoardPtrTable, 0, sizeof i2BoardPtrTable );
-	memset( DevTable, 0, sizeof DevTable );
-
 	/* Initialise all the boards we can find (up to the maximum). */
-	for ( i = 0; i < IP2_MAX_BOARDS; ++i ) {
-		switch ( ip2config.addr[i] ) { 
+	for (i = 0; i < IP2_MAX_BOARDS; ++i) {
+		switch (ip2config.addr[i]) {
 		case 0:	/* skip this slot even if card is present */
 			break;
 		default: /* ISA */
 		   /* ISA address must be specified */
-			if ( (ip2config.addr[i] < 0x100) || (ip2config.addr[i] > 0x3f8) ) {
-				printk ( KERN_ERR "IP2: Bad ISA board %d address %x\n",
-							 i, ip2config.addr[i] );
+			if (ip2config.addr[i] < 0x100 ||
+					ip2config.addr[i] > 0x3f8) {
+				printk(KERN_ERR "IP2: Bad ISA board %d "
+						"address %x\n", i,
+						ip2config.addr[i]);
 				ip2config.addr[i] = 0;
-			} else {
-				ip2config.type[i] = ISA;
-
-				/* Check for valid irq argument, set for polling if invalid */
-				if (ip2config.irq[i] && !is_valid_irq(ip2config.irq[i])) {
-					printk(KERN_ERR "IP2: Bad IRQ(%d) specified\n",ip2config.irq[i]);
-					ip2config.irq[i] = 0;// 0 is polling and is valid in that sense
-				}
+				break;
+			}
+			ip2config.type[i] = ISA;
+
+			/* Check for valid irq argument, set for polling if
+			 * invalid */
+			if (ip2config.irq[i] &&
+					!is_valid_irq(ip2config.irq[i])) {
+				printk(KERN_ERR "IP2: Bad IRQ(%d) specified\n",
+						ip2config.irq[i]);
+				/* 0 is polling and is valid in that sense */
+				ip2config.irq[i] = 0;
 			}
 			break;
 		case PCI:
 #ifdef CONFIG_PCI
-			{
-				int status;
+		{
+			u32 addr;
+			int status;
 
-				pci_dev_i = pci_get_device(PCI_VENDOR_ID_COMPUTONE,
-							  PCI_DEVICE_ID_COMPUTONE_IP2EX, pci_dev_i);
-				if (pci_dev_i != NULL) {
-					unsigned int addr;
-
-					if (pci_enable_device(pci_dev_i)) {
-						printk( KERN_ERR "IP2: can't enable PCI device at %s\n",
-							pci_name(pci_dev_i));
-						break;
-					}
-					ip2config.type[i] = PCI;
-					ip2config.pci_dev[i] = pci_dev_get(pci_dev_i);
-					status =
-					pci_read_config_dword(pci_dev_i, PCI_BASE_ADDRESS_1, &addr);
-					if ( addr & 1 ) {
-						ip2config.addr[i]=(USHORT)(addr&0xfffe);
-					} else {
-						printk( KERN_ERR "IP2: PCI I/O address error\n");
-					}
+			pdev = pci_get_device(PCI_VENDOR_ID_COMPUTONE,
+					PCI_DEVICE_ID_COMPUTONE_IP2EX, pdev);
+			if (pdev == NULL) {
+				ip2config.addr[i] = 0;
+				printk(KERN_ERR "IP2: PCI board %d not "
+						"found\n", i);
+				break;
+			}
 
-//		If the PCI BIOS assigned it, lets try and use it.  If we
-//		can't acquire it or it screws up, deal with it then.
-
-//					if (!is_valid_irq(pci_irq)) {
-//						printk( KERN_ERR "IP2: Bad PCI BIOS IRQ(%d)\n",pci_irq);
-//						pci_irq = 0;
-//					}
-					ip2config.irq[i] = pci_dev_i->irq;
-				} else {	// ann error
-					ip2config.addr[i] = 0;
-					printk(KERN_ERR "IP2: PCI board %d not found\n", i);
-				} 
+			if (pci_enable_device(pdev)) {
+				dev_err(&pdev->dev, "can't enable device\n");
+				break;
 			}
+			ip2config.type[i] = PCI;
+			ip2config.pci_dev[i] = pci_dev_get(pdev);
+			status = pci_read_config_dword(pdev, PCI_BASE_ADDRESS_1,
+					&addr);
+			if (addr & 1)
+				ip2config.addr[i] = (USHORT)(addr & 0xfffe);
+			else
+				dev_err(&pdev->dev, "I/O address error\n");
+
+			ip2config.irq[i] = pdev->irq;
+		}
 #else
-			printk( KERN_ERR "IP2: PCI card specified but PCI support not\n");
-			printk( KERN_ERR "IP2: configured in this kernel.\n");
-			printk( KERN_ERR "IP2: Recompile kernel with CONFIG_PCI defined!\n");
+			printk(KERN_ERR "IP2: PCI card specified but PCI "
+					"support not enabled.\n");
+			printk(KERN_ERR "IP2: Recompile kernel with CONFIG_PCI "
+					"defined!\n");
 #endif /* CONFIG_PCI */
 			break;
 		case EISA:
-			if ( (ip2config.addr[i] = find_eisa_board( Eisa_slot + 1 )) != 0) {
+			ip2config.addr[i] = find_eisa_board(Eisa_slot + 1);
+			if (ip2config.addr[i] != 0) {
 				/* Eisa_irq set as side effect, boo */
 				ip2config.type[i] = EISA;
 			} 
@@ -634,31 +656,32 @@ ip2_loadmain(int *iop, int *irqp)
 			break;
 		}	/* switch */
 	}	/* for */
-	if (pci_dev_i)
-		pci_dev_put(pci_dev_i);
+	pci_dev_put(pdev);
 
-	for ( i = 0; i < IP2_MAX_BOARDS; ++i ) {
-		if ( ip2config.addr[i] ) {
+	for (i = 0; i < IP2_MAX_BOARDS; ++i) {
+		if (ip2config.addr[i]) {
 			pB = kzalloc(sizeof(i2eBordStr), GFP_KERNEL);
 			if (pB) {
 				i2BoardPtrTable[i] = pB;
-				iiSetAddress( pB, ip2config.addr[i], ii2DelayTimer );
-				iiReset( pB );
-			} else {
-				printk(KERN_ERR "IP2: board memory allocation error\n");
-			}
+				iiSetAddress(pB, ip2config.addr[i],
+						ii2DelayTimer);
+				iiReset(pB);
+			} else
+				printk(KERN_ERR "IP2: board memory allocation "
+						"error\n");
 		}
 	}
-	for ( i = 0; i < IP2_MAX_BOARDS; ++i ) {
-		if ( ( pB = i2BoardPtrTable[i] ) != NULL ) {
-			iiResetDelay( pB );
+	for (i = 0; i < IP2_MAX_BOARDS; ++i) {
+		pB = i2BoardPtrTable[i];
+		if (pB != NULL) {
+			iiResetDelay(pB);
 			break;
 		}
 	}
-	for ( i = 0; i < IP2_MAX_BOARDS; ++i ) {
+	for (i = 0; i < IP2_MAX_BOARDS; ++i) {
 		/* We don't want to request the firmware unless we have at
 		   least one board */
-		if ( i2BoardPtrTable[i] != NULL ) {
+		if (i2BoardPtrTable[i] != NULL) {
 			if (!fw)
 				fw = ip2_request_firmware();
 			if (!fw)
@@ -669,7 +692,7 @@ ip2_loadmain(int *iop, int *irqp)
 	if (fw)
 		release_firmware(fw);
 
-	ip2trace (ITRC_NO_PORT, ITRC_INIT, 2, 0 );
+	ip2trace(ITRC_NO_PORT, ITRC_INIT, 2, 0);
 
 	ip2_tty_driver->owner		    = THIS_MODULE;
 	ip2_tty_driver->name                 = "ttyF";
@@ -680,20 +703,23 @@ ip2_loadmain(int *iop, int *irqp)
 	ip2_tty_driver->subtype              = SERIAL_TYPE_NORMAL;
 	ip2_tty_driver->init_termios         = tty_std_termios;
 	ip2_tty_driver->init_termios.c_cflag = B9600|CS8|CREAD|HUPCL|CLOCAL;
-	ip2_tty_driver->flags                = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV;
+	ip2_tty_driver->flags                = TTY_DRIVER_REAL_RAW |
+		TTY_DRIVER_DYNAMIC_DEV;
 	tty_set_operations(ip2_tty_driver, &ip2_ops);
 
-	ip2trace (ITRC_NO_PORT, ITRC_INIT, 3, 0 );
+	ip2trace(ITRC_NO_PORT, ITRC_INIT, 3, 0);
 
-	/* Register the tty devices. */
-	if ( ( err = tty_register_driver ( ip2_tty_driver ) ) ) {
-		printk(KERN_ERR "IP2: failed to register tty driver (%d)\n", err);
+	err = tty_register_driver(ip2_tty_driver);
+	if (err) {
+		printk(KERN_ERR "IP2: failed to register tty driver\n");
 		put_tty_driver(ip2_tty_driver);
-		return -EINVAL;
-	} else
-	/* Register the IPL driver. */
-	if ( ( err = register_chrdev ( IP2_IPL_MAJOR, pcIpl, &ip2_ipl ) ) ) {
-		printk(KERN_ERR "IP2: failed to register IPL device (%d)\n", err );
+		return err; /* leaking resources */
+	}
+
+	err = register_chrdev(IP2_IPL_MAJOR, pcIpl, &ip2_ipl);
+	if (err) {
+		printk(KERN_ERR "IP2: failed to register IPL device (%d)\n",
+				err);
 	} else {
 		/* create the sysfs class */
 		ip2_class = class_create(THIS_MODULE, "ip2");
@@ -705,84 +731,86 @@ ip2_loadmain(int *iop, int *irqp)
 	/* Register the read_procmem thing */
 	if (!proc_create("ip2mem",0,NULL,&ip2mem_proc_fops)) {
 		printk(KERN_ERR "IP2: failed to register read_procmem\n");
-	} else {
+		return -EIO; /* leaking resources */
+	}
 
-	ip2trace (ITRC_NO_PORT, ITRC_INIT, 4, 0 );
-		/* Register the interrupt handler or poll handler, depending upon the
-		 * specified interrupt.
-		 */
+	ip2trace(ITRC_NO_PORT, ITRC_INIT, 4, 0);
+	/* Register the interrupt handler or poll handler, depending upon the
+	 * specified interrupt.
+	 */
 
-		for( i = 0; i < IP2_MAX_BOARDS; ++i ) {
-			if ( 0 == ip2config.addr[i] ) {
-				continue;
-			}
+	for (i = 0; i < IP2_MAX_BOARDS; ++i) {
+		if (ip2config.addr[i] == 0)
+			continue;
 
-			if ( NULL != ( pB = i2BoardPtrTable[i] ) ) {
-				device_create_drvdata(ip2_class, NULL,
-						      MKDEV(IP2_IPL_MAJOR, 4 * i),
-						      NULL, "ipl%d", i);
-				device_create_drvdata(ip2_class, NULL,
-						      MKDEV(IP2_IPL_MAJOR, 4 * i + 1),
-						      NULL, "stat%d", i);
-
-			    for ( box = 0; box < ABS_MAX_BOXES; ++box )
-			    {
-			        for ( j = 0; j < ABS_BIGGEST_BOX; ++j )
-			        {
-				    if ( pB->i2eChannelMap[box] & (1 << j) )
-				    {
-				        tty_register_device(ip2_tty_driver,
-					    j + ABS_BIGGEST_BOX *
-						    (box+i*ABS_MAX_BOXES), NULL);
-			    	    }
-			        }
-			    }
-			}
+		pB = i2BoardPtrTable[i];
+		if (pB != NULL) {
+			device_create_drvdata(ip2_class, NULL,
+					      MKDEV(IP2_IPL_MAJOR, 4 * i),
+					      NULL, "ipl%d", i);
+			device_create_drvdata(ip2_class, NULL,
+					      MKDEV(IP2_IPL_MAJOR, 4 * i + 1),
+					      NULL, "stat%d", i);
+
+			for (box = 0; box < ABS_MAX_BOXES; box++)
+				for (j = 0; j < ABS_BIGGEST_BOX; j++)
+					if (pB->i2eChannelMap[box] & (1 << j))
+						tty_register_device(
+							ip2_tty_driver,
+							j + ABS_BIGGEST_BOX *
+							(box+i*ABS_MAX_BOXES),
+							NULL);
+		}
 
-			if (poll_only) {
-//		Poll only forces driver to only use polling and
-//		to ignore the probed PCI or EISA interrupts.
-				ip2config.irq[i] = CIR_POLL;
-			}
-			if ( ip2config.irq[i] == CIR_POLL ) {
+		if (poll_only) {
+			/* Poll only forces driver to only use polling and
+			   to ignore the probed PCI or EISA interrupts. */
+			ip2config.irq[i] = CIR_POLL;
+		}
+		if (ip2config.irq[i] == CIR_POLL) {
 retry:
-				if (!TimerOn) {
-					PollTimer.expires = POLL_TIMEOUT;
-					add_timer ( &PollTimer );
-					TimerOn = 1;
-					printk( KERN_INFO "IP2: polling\n");
-				}
-			} else {
-				if (have_requested_irq(ip2config.irq[i]))
-					continue;
-				rc = request_irq( ip2config.irq[i], ip2_interrupt,
-					IP2_SA_FLAGS | (ip2config.type[i] == PCI ? IRQF_SHARED : 0),
-					pcName, i2BoardPtrTable[i]);
-				if (rc) {
-					printk(KERN_ERR "IP2: an request_irq failed: error %d\n",rc);
-					ip2config.irq[i] = CIR_POLL;
-					printk( KERN_INFO "IP2: Polling %ld/sec.\n",
-							(POLL_TIMEOUT - jiffies));
-					goto retry;
-				} 
-				mark_requested_irq(ip2config.irq[i]);
-				/* Initialise the interrupt handler bottom half (aka slih). */
+			if (!timer_pending(&PollTimer)) {
+				mod_timer(&PollTimer, POLL_TIMEOUT);
+				printk(KERN_INFO "IP2: polling\n");
 			}
-		}
-		for( i = 0; i < IP2_MAX_BOARDS; ++i ) {
-			if ( i2BoardPtrTable[i] ) {
-				set_irq( i, ip2config.irq[i] ); /* set and enable board interrupt */
+		} else {
+			if (have_requested_irq(ip2config.irq[i]))
+				continue;
+			rc = request_irq(ip2config.irq[i], ip2_interrupt,
+				IP2_SA_FLAGS |
+				(ip2config.type[i] == PCI ? IRQF_SHARED : 0),
+				pcName, i2BoardPtrTable[i]);
+			if (rc) {
+				printk(KERN_ERR "IP2: request_irq failed: "
+						"error %d\n", rc);
+				ip2config.irq[i] = CIR_POLL;
+				printk(KERN_INFO "IP2: Polling %ld/sec.\n",
+						(POLL_TIMEOUT - jiffies));
+				goto retry;
 			}
+			mark_requested_irq(ip2config.irq[i]);
+			/* Initialise the interrupt handler bottom half
+			 * (aka slih). */
 		}
 	}
-	ip2trace (ITRC_NO_PORT, ITRC_INIT, ITRC_RETURN, 0 );
-	goto out;
+
+	for (i = 0; i < IP2_MAX_BOARDS; ++i) {
+		if (i2BoardPtrTable[i]) {
+			/* set and enable board interrupt */
+			set_irq(i, ip2config.irq[i]);
+		}
+	}
+
+	ip2trace(ITRC_NO_PORT, ITRC_INIT, ITRC_RETURN, 0);
+
+	return 0;
 
 out_chrdev:
 	unregister_chrdev(IP2_IPL_MAJOR, "ip2");
-out:
+	/* unregister and put tty here */
 	return err;
 }
+module_init(ip2_loadmain);
 
 /******************************************************************************/
 /* Function:   ip2_init_board()                                               */
@@ -1199,9 +1227,8 @@ ip2_polled_interrupt(void)
 {
 	int i;
 	i2eBordStrPtr  pB;
-	const int irq = 0;
 
-	ip2trace (ITRC_NO_PORT, ITRC_INTR, 99, 1, irq );
+	ip2trace(ITRC_NO_PORT, ITRC_INTR, 99, 1, 0);
 
 	/* Service just the boards on the list using this irq */
 	for( i = 0; i < i2nBoards; ++i ) {
@@ -1210,9 +1237,8 @@ ip2_polled_interrupt(void)
 //		Only process those boards which match our IRQ.
 //			IRQ = 0 for polled boards, we won't poll "IRQ" boards
 
-		if ( pB && (pB->i2eUsingIrq == irq) ) {
+		if (pB && pB->i2eUsingIrq == 0)
 			ip2_irq_work(pB);
-		}
 	}
 
 	++irq_counter;
@@ -1250,16 +1276,12 @@ ip2_poll(unsigned long arg)
 {
 	ip2trace (ITRC_NO_PORT, ITRC_INTR, 100, 0 );
 
-	TimerOn = 0; // it's the truth but not checked in service
-
 	// Just polled boards, IRQ = 0 will hit all non-interrupt boards.
 	// It will NOT poll boards handled by hard interrupts.
 	// The issue of queued BH interrupts is handled in ip2_interrupt().
 	ip2_polled_interrupt();
 
-	PollTimer.expires = POLL_TIMEOUT;
-	add_timer( &PollTimer );
-	TimerOn = 1;
+	mod_timer(&PollTimer, POLL_TIMEOUT);
 
 	ip2trace (ITRC_NO_PORT, ITRC_INTR, ITRC_RETURN, 0 );
 }
@@ -2871,7 +2893,7 @@ ip2_ipl_ioctl (struct file *pFile, UINT cmd, ULONG arg )
 	case 13:
 		switch ( cmd ) {
 		case 64:	/* Driver - ip2stat */
-			rc = put_user(ip2_tty_driver->refcount, pIndex++ );
+			rc = put_user(-1, pIndex++ );
 			rc = put_user(irq_counter, pIndex++  );
 			rc = put_user(bh_counter, pIndex++  );
 			break;
diff --git a/drivers/char/isicom.c b/drivers/char/isicom.c
index 8f7cc190b62d..7d30ee1d3fca 100644
--- a/drivers/char/isicom.c
+++ b/drivers/char/isicom.c
@@ -421,17 +421,16 @@ static void isicom_tx(unsigned long _data)
 	if (retries >= 100)
 		goto unlock;
 
+	tty = tty_port_tty_get(&port->port);
+	if (tty == NULL)
+		goto put_unlock;
+
 	for (; count > 0; count--, port++) {
 		/* port not active or tx disabled to force flow control */
 		if (!(port->port.flags & ASYNC_INITIALIZED) ||
 				!(port->status & ISI_TXOK))
 			continue;
 
-		tty = port->port.tty;
-
-		if (tty == NULL)
-			continue;
-
 		txcount = min_t(short, TX_SIZE, port->xmit_cnt);
 		if (txcount <= 0 || tty->stopped || tty->hw_stopped)
 			continue;
@@ -489,6 +488,8 @@ static void isicom_tx(unsigned long _data)
 			tty_wakeup(tty);
 	}
 
+put_unlock:
+	tty_kref_put(tty);
 unlock:
 	spin_unlock_irqrestore(&isi_card[card].card_lock, flags);
 	/*	schedule another tx for hopefully in about 10ms	*/
@@ -547,7 +548,7 @@ static irqreturn_t isicom_interrupt(int irq, void *dev_id)
 		return IRQ_HANDLED;
 	}
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (tty == NULL) {
 		word_count = byte_count >> 1;
 		while (byte_count > 1) {
@@ -588,7 +589,7 @@ static irqreturn_t isicom_interrupt(int irq, void *dev_id)
 			}
 
 			if (port->port.flags & ASYNC_CTS_FLOW) {
-				if (port->port.tty->hw_stopped) {
+				if (tty->hw_stopped) {
 					if (header & ISI_CTS) {
 						port->port.tty->hw_stopped = 0;
 						/* start tx ing */
@@ -597,7 +598,7 @@ static irqreturn_t isicom_interrupt(int irq, void *dev_id)
 						tty_wakeup(tty);
 					}
 				} else if (!(header & ISI_CTS)) {
-					port->port.tty->hw_stopped = 1;
+					tty->hw_stopped = 1;
 					/* stop tx ing */
 					port->status &= ~(ISI_TXOK | ISI_CTS);
 				}
@@ -660,24 +661,21 @@ static irqreturn_t isicom_interrupt(int irq, void *dev_id)
 	}
 	outw(0x0000, base+0x04); /* enable interrupts */
 	spin_unlock(&card->card_lock);
+	tty_kref_put(tty);
 
 	return IRQ_HANDLED;
 }
 
-static void isicom_config_port(struct isi_port *port)
+static void isicom_config_port(struct tty_struct *tty)
 {
+	struct isi_port *port = tty->driver_data;
 	struct isi_board *card = port->card;
-	struct tty_struct *tty;
 	unsigned long baud;
 	unsigned long base = card->base;
 	u16 channel_setup, channel = port->channel,
 		shift_count = card->shift_count;
 	unsigned char flow_ctrl;
 
-	tty = port->port.tty;
-
-	if (tty == NULL)
-		return;
 	/* FIXME: Switch to new tty baud API */
 	baud = C_BAUD(tty);
 	if (baud & CBAUDEX) {
@@ -690,7 +688,7 @@ static void isicom_config_port(struct isi_port *port)
 
 		/* 1,2,3,4 => 57.6, 115.2, 230, 460 kbps resp. */
 		if (baud < 1 || baud > 4)
-			port->port.tty->termios->c_cflag &= ~CBAUDEX;
+			tty->termios->c_cflag &= ~CBAUDEX;
 		else
 			baud += 15;
 	}
@@ -797,8 +795,9 @@ static inline void isicom_setup_board(struct isi_board *bp)
 	spin_unlock_irqrestore(&bp->card_lock, flags);
 }
 
-static int isicom_setup_port(struct isi_port *port)
+static int isicom_setup_port(struct tty_struct *tty)
 {
+	struct isi_port *port = tty->driver_data;
 	struct isi_board *card = port->card;
 	unsigned long flags;
 
@@ -808,8 +807,7 @@ static int isicom_setup_port(struct isi_port *port)
 		return -ENOMEM;
 
 	spin_lock_irqsave(&card->card_lock, flags);
-	if (port->port.tty)
-		clear_bit(TTY_IO_ERROR, &port->port.tty->flags);
+	clear_bit(TTY_IO_ERROR, &tty->flags);
 	if (port->port.count == 1)
 		card->count++;
 
@@ -823,7 +821,7 @@ static int isicom_setup_port(struct isi_port *port)
 		InterruptTheCard(card->base);
 	}
 
-	isicom_config_port(port);
+	isicom_config_port(tty);
 	port->port.flags |= ASYNC_INITIALIZED;
 	spin_unlock_irqrestore(&card->card_lock, flags);
 
@@ -934,8 +932,8 @@ static int isicom_open(struct tty_struct *tty, struct file *filp)
 
 	port->port.count++;
 	tty->driver_data = port;
-	port->port.tty = tty;
-	error = isicom_setup_port(port);
+	tty_port_tty_set(&port->port, tty);
+	error = isicom_setup_port(tty);
 	if (error == 0)
 		error = block_til_ready(tty, filp, port);
 	return error;
@@ -955,15 +953,17 @@ static void isicom_shutdown_port(struct isi_port *port)
 	struct isi_board *card = port->card;
 	struct tty_struct *tty;
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 
-	if (!(port->port.flags & ASYNC_INITIALIZED))
+	if (!(port->port.flags & ASYNC_INITIALIZED)) {
+		tty_kref_put(tty);
 		return;
+	}
 
 	tty_port_free_xmit_buf(&port->port);
 	port->port.flags &= ~ASYNC_INITIALIZED;
 	/* 3rd October 2000 : Vinayak P Risbud */
-	port->port.tty = NULL;
+	tty_port_tty_set(&port->port, NULL);
 
 	/*Fix done by Anil .S on 30-04-2001
 	remote login through isi port has dtr toggle problem
@@ -1243,9 +1243,10 @@ static int isicom_tiocmset(struct tty_struct *tty, struct file *file,
 	return 0;
 }
 
-static int isicom_set_serial_info(struct isi_port *port,
-	struct serial_struct __user *info)
+static int isicom_set_serial_info(struct tty_struct *tty,
+					struct serial_struct __user *info)
 {
+	struct isi_port *port = tty->driver_data;
 	struct serial_struct newinfo;
 	int reconfig_port;
 
@@ -1276,7 +1277,7 @@ static int isicom_set_serial_info(struct isi_port *port,
 	if (reconfig_port) {
 		unsigned long flags;
 		spin_lock_irqsave(&port->card->card_lock, flags);
-		isicom_config_port(port);
+		isicom_config_port(tty);
 		spin_unlock_irqrestore(&port->card->card_lock, flags);
 	}
 	unlock_kernel();
@@ -1318,7 +1319,7 @@ static int isicom_ioctl(struct tty_struct *tty, struct file *filp,
 		return isicom_get_serial_info(port, argp);
 
 	case TIOCSSERIAL:
-		return isicom_set_serial_info(port, argp);
+		return isicom_set_serial_info(tty, argp);
 
 	default:
 		return -ENOIOCTLCMD;
@@ -1341,7 +1342,7 @@ static void isicom_set_termios(struct tty_struct *tty,
 		return;
 
 	spin_lock_irqsave(&port->card->card_lock, flags);
-	isicom_config_port(port);
+	isicom_config_port(tty);
 	spin_unlock_irqrestore(&port->card->card_lock, flags);
 
 	if ((old_termios->c_cflag & CRTSCTS) &&
@@ -1419,7 +1420,7 @@ static void isicom_hangup(struct tty_struct *tty)
 
 	port->port.count = 0;
 	port->port.flags &= ~ASYNC_NORMAL_ACTIVE;
-	port->port.tty = NULL;
+	tty_port_tty_set(&port->port, NULL);
 	wake_up_interruptible(&port->port.open_wait);
 }
 
diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c
index 843a2afaf204..505d7a1f6b8c 100644
--- a/drivers/char/istallion.c
+++ b/drivers/char/istallion.c
@@ -623,24 +623,25 @@ static int	stli_memioctl(struct inode *ip, struct file *fp, unsigned int cmd, un
 static void	stli_brdpoll(struct stlibrd *brdp, cdkhdr_t __iomem *hdrp);
 static void	stli_poll(unsigned long arg);
 static int	stli_hostcmd(struct stlibrd *brdp, struct stliport *portp);
-static int	stli_initopen(struct stlibrd *brdp, struct stliport *portp);
+static int	stli_initopen(struct tty_struct *tty, struct stlibrd *brdp, struct stliport *portp);
 static int	stli_rawopen(struct stlibrd *brdp, struct stliport *portp, unsigned long arg, int wait);
 static int	stli_rawclose(struct stlibrd *brdp, struct stliport *portp, unsigned long arg, int wait);
-static int	stli_waitcarrier(struct stlibrd *brdp, struct stliport *portp, struct file *filp);
-static int	stli_setport(struct stliport *portp);
+static int	stli_waitcarrier(struct tty_struct *tty, struct stlibrd *brdp,
+				struct stliport *portp, struct file *filp);
+static int	stli_setport(struct tty_struct *tty);
 static int	stli_cmdwait(struct stlibrd *brdp, struct stliport *portp, unsigned long cmd, void *arg, int size, int copyback);
 static void	stli_sendcmd(struct stlibrd *brdp, struct stliport *portp, unsigned long cmd, void *arg, int size, int copyback);
 static void	__stli_sendcmd(struct stlibrd *brdp, struct stliport *portp, unsigned long cmd, void *arg, int size, int copyback);
 static void	stli_dodelaycmd(struct stliport *portp, cdkctrl_t __iomem *cp);
-static void	stli_mkasyport(struct stliport *portp, asyport_t *pp, struct ktermios *tiosp);
+static void	stli_mkasyport(struct tty_struct *tty, struct stliport *portp, asyport_t *pp, struct ktermios *tiosp);
 static void	stli_mkasysigs(asysigs_t *sp, int dtr, int rts);
 static long	stli_mktiocm(unsigned long sigvalue);
 static void	stli_read(struct stlibrd *brdp, struct stliport *portp);
 static int	stli_getserial(struct stliport *portp, struct serial_struct __user *sp);
-static int	stli_setserial(struct stliport *portp, struct serial_struct __user *sp);
+static int	stli_setserial(struct tty_struct *tty, struct serial_struct __user *sp);
 static int	stli_getbrdstats(combrd_t __user *bp);
-static int	stli_getportstats(struct stliport *portp, comstats_t __user *cp);
-static int	stli_portcmdstats(struct stliport *portp);
+static int	stli_getportstats(struct tty_struct *tty, struct stliport *portp, comstats_t __user *cp);
+static int	stli_portcmdstats(struct tty_struct *tty, struct stliport *portp);
 static int	stli_clrportstats(struct stliport *portp, comstats_t __user *cp);
 static int	stli_getportstruct(struct stliport __user *arg);
 static int	stli_getbrdstruct(struct stlibrd __user *arg);
@@ -731,12 +732,16 @@ static void stli_cleanup_ports(struct stlibrd *brdp)
 {
 	struct stliport *portp;
 	unsigned int j;
+	struct tty_struct *tty;
 
 	for (j = 0; j < STL_MAXPORTS; j++) {
 		portp = brdp->ports[j];
 		if (portp != NULL) {
-			if (portp->port.tty != NULL)
-				tty_hangup(portp->port.tty);
+			tty = tty_port_tty_get(&portp->port);
+			if (tty != NULL) {
+				tty_hangup(tty);
+				tty_kref_put(tty);
+			}
 			kfree(portp);
 		}
 	}
@@ -824,7 +829,7 @@ static int stli_open(struct tty_struct *tty, struct file *filp)
  *	requires several commands to the board we will need to wait for any
  *	other open that is already initializing the port.
  */
-	portp->port.tty = tty;
+	tty_port_tty_set(&portp->port, tty);
 	tty->driver_data = portp;
 	portp->port.count++;
 
@@ -835,7 +840,7 @@ static int stli_open(struct tty_struct *tty, struct file *filp)
 
 	if ((portp->port.flags & ASYNC_INITIALIZED) == 0) {
 		set_bit(ST_INITIALIZING, &portp->state);
-		if ((rc = stli_initopen(brdp, portp)) >= 0) {
+		if ((rc = stli_initopen(tty, brdp, portp)) >= 0) {
 			portp->port.flags |= ASYNC_INITIALIZED;
 			clear_bit(TTY_IO_ERROR, &tty->flags);
 		}
@@ -864,7 +869,7 @@ static int stli_open(struct tty_struct *tty, struct file *filp)
  *	then also we might have to wait for carrier.
  */
 	if (!(filp->f_flags & O_NONBLOCK)) {
-		if ((rc = stli_waitcarrier(brdp, portp, filp)) != 0)
+		if ((rc = stli_waitcarrier(tty, brdp, portp, filp)) != 0)
 			return rc;
 	}
 	portp->port.flags |= ASYNC_NORMAL_ACTIVE;
@@ -930,7 +935,7 @@ static void stli_close(struct tty_struct *tty, struct file *filp)
 	stli_flushbuffer(tty);
 
 	tty->closing = 0;
-	portp->port.tty = NULL;
+	tty_port_tty_set(&portp->port, NULL);
 
 	if (portp->openwaitcnt) {
 		if (portp->close_delay)
@@ -952,9 +957,9 @@ static void stli_close(struct tty_struct *tty, struct file *filp)
  *	this still all happens pretty quickly.
  */
 
-static int stli_initopen(struct stlibrd *brdp, struct stliport *portp)
+static int stli_initopen(struct tty_struct *tty,
+				struct stlibrd *brdp, struct stliport *portp)
 {
-	struct tty_struct *tty;
 	asynotify_t nt;
 	asyport_t aport;
 	int rc;
@@ -969,10 +974,7 @@ static int stli_initopen(struct stlibrd *brdp, struct stliport *portp)
 	    sizeof(asynotify_t), 0)) < 0)
 		return rc;
 
-	tty = portp->port.tty;
-	if (tty == NULL)
-		return -ENODEV;
-	stli_mkasyport(portp, &aport, tty->termios);
+	stli_mkasyport(tty, portp, &aport, tty->termios);
 	if ((rc = stli_cmdwait(brdp, portp, A_SETPORT, &aport,
 	    sizeof(asyport_t), 0)) < 0)
 		return rc;
@@ -1161,22 +1163,21 @@ static int stli_cmdwait(struct stlibrd *brdp, struct stliport *portp, unsigned l
  *	waiting for the command to complete - so must have user context.
  */
 
-static int stli_setport(struct stliport *portp)
+static int stli_setport(struct tty_struct *tty)
 {
+	struct stliport *portp = tty->driver_data;
 	struct stlibrd *brdp;
 	asyport_t aport;
 
 	if (portp == NULL)
 		return -ENODEV;
-	if (portp->port.tty == NULL)
-		return -ENODEV;
 	if (portp->brdnr >= stli_nrbrds)
 		return -ENODEV;
 	brdp = stli_brds[portp->brdnr];
 	if (brdp == NULL)
 		return -ENODEV;
 
-	stli_mkasyport(portp, &aport, portp->port.tty->termios);
+	stli_mkasyport(tty, portp, &aport, tty->termios);
 	return(stli_cmdwait(brdp, portp, A_SETPORT, &aport, sizeof(asyport_t), 0));
 }
 
@@ -1187,7 +1188,8 @@ static int stli_setport(struct stliport *portp)
  *	maybe because if we are clocal then we don't need to wait...
  */
 
-static int stli_waitcarrier(struct stlibrd *brdp, struct stliport *portp, struct file *filp)
+static int stli_waitcarrier(struct tty_struct *tty, struct stlibrd *brdp,
+				struct stliport *portp, struct file *filp)
 {
 	unsigned long flags;
 	int rc, doclocal;
@@ -1195,7 +1197,7 @@ static int stli_waitcarrier(struct stlibrd *brdp, struct stliport *portp, struct
 	rc = 0;
 	doclocal = 0;
 
-	if (portp->port.tty->termios->c_cflag & CLOCAL)
+	if (tty->termios->c_cflag & CLOCAL)
 		doclocal++;
 
 	spin_lock_irqsave(&stli_lock, flags);
@@ -1373,8 +1375,6 @@ static void stli_flushchars(struct tty_struct *tty)
 	stli_txcookrealsize = 0;
 	stli_txcooktty = NULL;
 
-	if (tty == NULL)
-		return;
 	if (cooktty == NULL)
 		return;
 	if (tty != cooktty)
@@ -1572,10 +1572,11 @@ static int stli_getserial(struct stliport *portp, struct serial_struct __user *s
  *	just quietly ignore any requests to change irq, etc.
  */
 
-static int stli_setserial(struct stliport *portp, struct serial_struct __user *sp)
+static int stli_setserial(struct tty_struct *tty, struct serial_struct __user *sp)
 {
 	struct serial_struct sio;
 	int rc;
+	struct stliport *portp = tty->driver_data;
 
 	if (copy_from_user(&sio, sp, sizeof(struct serial_struct)))
 		return -EFAULT;
@@ -1594,7 +1595,7 @@ static int stli_setserial(struct stliport *portp, struct serial_struct __user *s
 	portp->closing_wait = sio.closing_wait;
 	portp->custom_divisor = sio.custom_divisor;
 
-	if ((rc = stli_setport(portp)) < 0)
+	if ((rc = stli_setport(tty)) < 0)
 		return rc;
 	return 0;
 }
@@ -1685,17 +1686,17 @@ static int stli_ioctl(struct tty_struct *tty, struct file *file, unsigned int cm
 		rc = stli_getserial(portp, argp);
 		break;
 	case TIOCSSERIAL:
-		rc = stli_setserial(portp, argp);
+		rc = stli_setserial(tty, argp);
 		break;
 	case STL_GETPFLAG:
 		rc = put_user(portp->pflag, (unsigned __user *)argp);
 		break;
 	case STL_SETPFLAG:
 		if ((rc = get_user(portp->pflag, (unsigned __user *)argp)) == 0)
-			stli_setport(portp);
+			stli_setport(tty);
 		break;
 	case COM_GETPORTSTATS:
-		rc = stli_getportstats(portp, argp);
+		rc = stli_getportstats(tty, portp, argp);
 		break;
 	case COM_CLRPORTSTATS:
 		rc = stli_clrportstats(portp, argp);
@@ -1729,8 +1730,6 @@ static void stli_settermios(struct tty_struct *tty, struct ktermios *old)
 	struct ktermios *tiosp;
 	asyport_t aport;
 
-	if (tty == NULL)
-		return;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
@@ -1742,7 +1741,7 @@ static void stli_settermios(struct tty_struct *tty, struct ktermios *old)
 
 	tiosp = tty->termios;
 
-	stli_mkasyport(portp, &aport, tiosp);
+	stli_mkasyport(tty, portp, &aport, tiosp);
 	stli_cmdwait(brdp, portp, A_SETPORT, &aport, sizeof(asyport_t), 0);
 	stli_mkasysigs(&portp->asig, ((tiosp->c_cflag & CBAUD) ? 1 : 0), -1);
 	stli_cmdwait(brdp, portp, A_SETSIGNALS, &portp->asig,
@@ -1854,7 +1853,7 @@ static void stli_hangup(struct tty_struct *tty)
 	clear_bit(ST_TXBUSY, &portp->state);
 	clear_bit(ST_RXSTOP, &portp->state);
 	set_bit(TTY_IO_ERROR, &tty->flags);
-	portp->port.tty = NULL;
+	tty_port_tty_set(&portp->port, NULL);
 	portp->port.flags &= ~ASYNC_NORMAL_ACTIVE;
 	portp->port.count = 0;
 	spin_unlock_irqrestore(&stli_lock, flags);
@@ -1935,8 +1934,6 @@ static void stli_waituntilsent(struct tty_struct *tty, int timeout)
 	struct stliport *portp;
 	unsigned long tend;
 
-	if (tty == NULL)
-		return;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
@@ -1998,7 +1995,7 @@ static int stli_portinfo(struct stlibrd *brdp, struct stliport *portp, int portn
 	char *sp, *uart;
 	int rc, cnt;
 
-	rc = stli_portcmdstats(portp);
+	rc = stli_portcmdstats(NULL, portp);
 
 	uart = "UNKNOWN";
 	if (brdp->state & BST_STARTED) {
@@ -2188,7 +2185,7 @@ static void stli_read(struct stlibrd *brdp, struct stliport *portp)
 
 	if (test_bit(ST_RXSTOP, &portp->state))
 		return;
-	tty = portp->port.tty;
+	tty = tty_port_tty_get(&portp->port);
 	if (tty == NULL)
 		return;
 
@@ -2230,6 +2227,7 @@ static void stli_read(struct stlibrd *brdp, struct stliport *portp)
 		set_bit(ST_RXING, &portp->state);
 
 	tty_schedule_flip(tty);
+	tty_kref_put(tty);
 }
 
 /*****************************************************************************/
@@ -2362,7 +2360,7 @@ static int stli_hostcmd(struct stlibrd *brdp, struct stliport *portp)
 	if (ap->notify) {
 		nt = ap->changed;
 		ap->notify = 0;
-		tty = portp->port.tty;
+		tty = tty_port_tty_get(&portp->port);
 
 		if (nt.signal & SG_DCD) {
 			oldsigs = portp->sigs;
@@ -2399,6 +2397,7 @@ static int stli_hostcmd(struct stlibrd *brdp, struct stliport *portp)
 				tty_schedule_flip(tty);
 			}
 		}
+		tty_kref_put(tty);
 
 		if (nt.data & DT_RXBUSY) {
 			donerx++;
@@ -2535,14 +2534,15 @@ static void stli_poll(unsigned long arg)
  *	the slave.
  */
 
-static void stli_mkasyport(struct stliport *portp, asyport_t *pp, struct ktermios *tiosp)
+static void stli_mkasyport(struct tty_struct *tty, struct stliport *portp,
+				asyport_t *pp, struct ktermios *tiosp)
 {
 	memset(pp, 0, sizeof(asyport_t));
 
 /*
  *	Start of by setting the baud, char size, parity and stop bit info.
  */
-	pp->baudout = tty_get_baud_rate(portp->port.tty);
+	pp->baudout = tty_get_baud_rate(tty);
 	if ((tiosp->c_cflag & CBAUD) == B38400) {
 		if ((portp->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
 			pp->baudout = 57600;
@@ -2695,7 +2695,7 @@ static int stli_initports(struct stlibrd *brdp)
 			printk("STALLION: failed to allocate port structure\n");
 			continue;
 		}
-
+		tty_port_init(&portp->port);
 		portp->magic = STLI_PORTMAGIC;
 		portp->portnr = i;
 		portp->brdnr = brdp->brdnr;
@@ -4220,7 +4220,7 @@ static struct stliport *stli_getport(unsigned int brdnr, unsigned int panelnr,
  *	what port to get stats for (used through board control device).
  */
 
-static int stli_portcmdstats(struct stliport *portp)
+static int stli_portcmdstats(struct tty_struct *tty, struct stliport *portp)
 {
 	unsigned long	flags;
 	struct stlibrd	*brdp;
@@ -4249,15 +4249,15 @@ static int stli_portcmdstats(struct stliport *portp)
 	stli_comstats.flags = portp->port.flags;
 
 	spin_lock_irqsave(&brd_lock, flags);
-	if (portp->port.tty != NULL) {
-		if (portp->port.tty->driver_data == portp) {
-			stli_comstats.ttystate = portp->port.tty->flags;
+	if (tty != NULL) {
+		if (portp->port.tty == tty) {
+			stli_comstats.ttystate = tty->flags;
 			stli_comstats.rxbuffered = -1;
-			if (portp->port.tty->termios != NULL) {
-				stli_comstats.cflags = portp->port.tty->termios->c_cflag;
-				stli_comstats.iflags = portp->port.tty->termios->c_iflag;
-				stli_comstats.oflags = portp->port.tty->termios->c_oflag;
-				stli_comstats.lflags = portp->port.tty->termios->c_lflag;
+			if (tty->termios != NULL) {
+				stli_comstats.cflags = tty->termios->c_cflag;
+				stli_comstats.iflags = tty->termios->c_iflag;
+				stli_comstats.oflags = tty->termios->c_oflag;
+				stli_comstats.lflags = tty->termios->c_lflag;
 			}
 		}
 	}
@@ -4294,7 +4294,8 @@ static int stli_portcmdstats(struct stliport *portp)
  *	what port to get stats for (used through board control device).
  */
 
-static int stli_getportstats(struct stliport *portp, comstats_t __user *cp)
+static int stli_getportstats(struct tty_struct *tty, struct stliport *portp,
+							comstats_t __user *cp)
 {
 	struct stlibrd *brdp;
 	int rc;
@@ -4312,7 +4313,7 @@ static int stli_getportstats(struct stliport *portp, comstats_t __user *cp)
 	if (!brdp)
 		return -ENODEV;
 
-	if ((rc = stli_portcmdstats(portp)) < 0)
+	if ((rc = stli_portcmdstats(tty, portp)) < 0)
 		return rc;
 
 	return copy_to_user(cp, &stli_comstats, sizeof(comstats_t)) ?
@@ -4427,7 +4428,7 @@ static int stli_memioctl(struct inode *ip, struct file *fp, unsigned int cmd, un
 
 	switch (cmd) {
 	case COM_GETPORTSTATS:
-		rc = stli_getportstats(NULL, argp);
+		rc = stli_getportstats(NULL, NULL, argp);
 		done++;
 		break;
 	case COM_CLRPORTSTATS:
diff --git a/drivers/char/moxa.c b/drivers/char/moxa.c
index d3d7864e0c1e..5df4003ad873 100644
--- a/drivers/char/moxa.c
+++ b/drivers/char/moxa.c
@@ -205,7 +205,7 @@ static int moxa_tiocmset(struct tty_struct *tty, struct file *file,
 static void moxa_poll(unsigned long);
 static void moxa_set_tty_param(struct tty_struct *, struct ktermios *);
 static void moxa_setup_empty_event(struct tty_struct *);
-static void moxa_shut_down(struct moxa_port *);
+static void moxa_shut_down(struct tty_struct *);
 /*
  * moxa board interface functions:
  */
@@ -217,7 +217,7 @@ static void MoxaPortLineCtrl(struct moxa_port *, int, int);
 static void MoxaPortFlowCtrl(struct moxa_port *, int, int, int, int, int);
 static int MoxaPortLineStatus(struct moxa_port *);
 static void MoxaPortFlushData(struct moxa_port *, int);
-static int MoxaPortWriteData(struct moxa_port *, const unsigned char *, int);
+static int MoxaPortWriteData(struct tty_struct *, const unsigned char *, int);
 static int MoxaPortReadData(struct moxa_port *);
 static int MoxaPortTxQueue(struct moxa_port *);
 static int MoxaPortRxQueue(struct moxa_port *);
@@ -332,6 +332,7 @@ static int moxa_ioctl(struct tty_struct *tty, struct file *file,
 		for (i = 0; i < MAX_BOARDS; i++) {
 			p = moxa_boards[i].ports;
 			for (j = 0; j < MAX_PORTS_PER_BOARD; j++, p++, argm++) {
+				struct tty_struct *ttyp;
 				memset(&tmp, 0, sizeof(tmp));
 				if (!moxa_boards[i].ready)
 					goto copy;
@@ -344,10 +345,12 @@ static int moxa_ioctl(struct tty_struct *tty, struct file *file,
 				if (status & 4)
 					tmp.dcd = 1;
 
-				if (!p->port.tty || !p->port.tty->termios)
+				ttyp = tty_port_tty_get(&p->port);
+				if (!ttyp || !ttyp->termios)
 					tmp.cflag = p->cflag;
 				else
-					tmp.cflag = p->port.tty->termios->c_cflag;
+					tmp.cflag = ttyp->termios->c_cflag;
+				tty_kref_put(tty);
 copy:
 				if (copy_to_user(argm, &tmp, sizeof(tmp))) {
 					mutex_unlock(&moxa_openlock);
@@ -880,8 +883,14 @@ static void moxa_board_deinit(struct moxa_board_conf *brd)
 
 	/* pci hot-un-plug support */
 	for (a = 0; a < brd->numPorts; a++)
-		if (brd->ports[a].port.flags & ASYNC_INITIALIZED)
-			tty_hangup(brd->ports[a].port.tty);
+		if (brd->ports[a].port.flags & ASYNC_INITIALIZED) {
+			struct tty_struct *tty = tty_port_tty_get(
+						&brd->ports[a].port);
+			if (tty) {
+				tty_hangup(tty);
+				tty_kref_put(tty);
+			}
+		}
 	while (1) {
 		opened = 0;
 		for (a = 0; a < brd->numPorts; a++)
@@ -1096,13 +1105,14 @@ static void __exit moxa_exit(void)
 module_init(moxa_init);
 module_exit(moxa_exit);
 
-static void moxa_close_port(struct moxa_port *ch)
+static void moxa_close_port(struct tty_struct *tty)
 {
-	moxa_shut_down(ch);
+	struct moxa_port *ch = tty->driver_data;
+	moxa_shut_down(tty);
 	MoxaPortFlushData(ch, 2);
 	ch->port.flags &= ~ASYNC_NORMAL_ACTIVE;
-	ch->port.tty->driver_data = NULL;
-	ch->port.tty = NULL;
+	tty->driver_data = NULL;
+	tty_port_tty_set(&ch->port, NULL);
 }
 
 static int moxa_block_till_ready(struct tty_struct *tty, struct file *filp,
@@ -1161,7 +1171,7 @@ static int moxa_open(struct tty_struct *tty, struct file *filp)
 	ch = &brd->ports[port % MAX_PORTS_PER_BOARD];
 	ch->port.count++;
 	tty->driver_data = ch;
-	ch->port.tty = tty;
+	tty_port_tty_set(&ch->port, tty);
 	if (!(ch->port.flags & ASYNC_INITIALIZED)) {
 		ch->statusflags = 0;
 		moxa_set_tty_param(tty, tty->termios);
@@ -1179,7 +1189,7 @@ static int moxa_open(struct tty_struct *tty, struct file *filp)
 	if (retval) {
 		if (ch->port.count) /* 0 means already hung up... */
 			if (--ch->port.count == 0)
-				moxa_close_port(ch);
+				moxa_close_port(tty);
 	} else
 		ch->port.flags |= ASYNC_NORMAL_ACTIVE;
 	mutex_unlock(&moxa_openlock);
@@ -1219,7 +1229,7 @@ static void moxa_close(struct tty_struct *tty, struct file *filp)
 		tty_wait_until_sent(tty, 30 * HZ);	/* 30 seconds timeout */
 	}
 
-	moxa_close_port(ch);
+	moxa_close_port(tty);
 unlock:
 	mutex_unlock(&moxa_openlock);
 }
@@ -1234,7 +1244,7 @@ static int moxa_write(struct tty_struct *tty,
 		return 0;
 
 	spin_lock_bh(&moxa_lock);
-	len = MoxaPortWriteData(ch, buf, count);
+	len = MoxaPortWriteData(tty, buf, count);
 	spin_unlock_bh(&moxa_lock);
 
 	ch->statusflags |= LOWWAIT;
@@ -1409,7 +1419,7 @@ static void moxa_hangup(struct tty_struct *tty)
 		return;
 	}
 	ch->port.count = 0;
-	moxa_close_port(ch);
+	moxa_close_port(tty);
 	mutex_unlock(&moxa_openlock);
 
 	wake_up_interruptible(&ch->port.open_wait);
@@ -1417,11 +1427,14 @@ static void moxa_hangup(struct tty_struct *tty)
 
 static void moxa_new_dcdstate(struct moxa_port *p, u8 dcd)
 {
+	struct tty_struct *tty;
 	dcd = !!dcd;
 
-	if (dcd != p->DCDState && p->port.tty && C_CLOCAL(p->port.tty)) {
-		if (!dcd)
-			tty_hangup(p->port.tty);
+	if (dcd != p->DCDState) {
+		tty = tty_port_tty_get(&p->port);
+		if (tty && C_CLOCAL(tty) && !dcd)
+			tty_hangup(tty);
+		tty_kref_put(tty);
 	}
 	p->DCDState = dcd;
 }
@@ -1429,7 +1442,7 @@ static void moxa_new_dcdstate(struct moxa_port *p, u8 dcd)
 static int moxa_poll_port(struct moxa_port *p, unsigned int handle,
 		u16 __iomem *ip)
 {
-	struct tty_struct *tty = p->port.tty;
+	struct tty_struct *tty = tty_port_tty_get(&p->port);
 	void __iomem *ofsAddr;
 	unsigned int inited = p->port.flags & ASYNC_INITIALIZED;
 	u16 intr;
@@ -1476,6 +1489,7 @@ static int moxa_poll_port(struct moxa_port *p, unsigned int handle,
 		tty_insert_flip_char(tty, 0, TTY_BREAK);
 		tty_schedule_flip(tty);
 	}
+	tty_kref_put(tty);
 
 	if (intr & IntrLine)
 		moxa_new_dcdstate(p, readb(ofsAddr + FlagStat) & DCD_state);
@@ -1560,9 +1574,9 @@ static void moxa_setup_empty_event(struct tty_struct *tty)
 	spin_unlock_bh(&moxa_lock);
 }
 
-static void moxa_shut_down(struct moxa_port *ch)
+static void moxa_shut_down(struct tty_struct *tty)
 {
-	struct tty_struct *tp = ch->port.tty;
+	struct moxa_port *ch = tty->driver_data;
 
 	if (!(ch->port.flags & ASYNC_INITIALIZED))
 		return;
@@ -1572,7 +1586,7 @@ static void moxa_shut_down(struct moxa_port *ch)
 	/*
 	 * If we're a modem control device and HUPCL is on, drop RTS & DTR.
 	 */
-	if (C_HUPCL(tp))
+	if (C_HUPCL(tty))
 		MoxaPortLineCtrl(ch, 0, 0);
 
 	spin_lock_bh(&moxa_lock);
@@ -1953,9 +1967,10 @@ static int MoxaPortLineStatus(struct moxa_port *port)
 	return val;
 }
 
-static int MoxaPortWriteData(struct moxa_port *port,
+static int MoxaPortWriteData(struct tty_struct *tty,
 		const unsigned char *buffer, int len)
 {
+	struct moxa_port *port = tty->driver_data;
 	void __iomem *baseAddr, *ofsAddr, *ofs;
 	unsigned int c, total;
 	u16 head, tail, tx_mask, spage, epage;
diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index b638403e8e9c..8beef50f95a0 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -610,15 +610,13 @@ static int mxser_block_til_ready(struct tty_struct *tty, struct file *filp,
 	return 0;
 }
 
-static int mxser_set_baud(struct mxser_port *info, long newspd)
+static int mxser_set_baud(struct tty_struct *tty, long newspd)
 {
+	struct mxser_port *info = tty->driver_data;
 	int quot = 0, baud;
 	unsigned char cval;
 
-	if (!info->port.tty || !info->port.tty->termios)
-		return -1;
-
-	if (!(info->ioaddr))
+	if (!info->ioaddr)
 		return -1;
 
 	if (newspd > info->max_baud)
@@ -626,13 +624,13 @@ static int mxser_set_baud(struct mxser_port *info, long newspd)
 
 	if (newspd == 134) {
 		quot = 2 * info->baud_base / 269;
-		tty_encode_baud_rate(info->port.tty, 134, 134);
+		tty_encode_baud_rate(tty, 134, 134);
 	} else if (newspd) {
 		quot = info->baud_base / newspd;
 		if (quot == 0)
 			quot = 1;
 		baud = info->baud_base/quot;
-		tty_encode_baud_rate(info->port.tty, baud, baud);
+		tty_encode_baud_rate(tty, baud, baud);
 	} else {
 		quot = 0;
 	}
@@ -658,7 +656,7 @@ static int mxser_set_baud(struct mxser_port *info, long newspd)
 	outb(cval, info->ioaddr + UART_LCR);	/* reset DLAB */
 
 #ifdef BOTHER
-	if (C_BAUD(info->port.tty) == BOTHER) {
+	if (C_BAUD(tty) == BOTHER) {
 		quot = info->baud_base % newspd;
 		quot *= 8;
 		if (quot % newspd > newspd / 2) {
@@ -679,21 +677,20 @@ static int mxser_set_baud(struct mxser_port *info, long newspd)
  * This routine is called to set the UART divisor registers to match
  * the specified baud rate for a serial port.
  */
-static int mxser_change_speed(struct mxser_port *info,
-		struct ktermios *old_termios)
+static int mxser_change_speed(struct tty_struct *tty,
+					struct ktermios *old_termios)
 {
+	struct mxser_port *info = tty->driver_data;
 	unsigned cflag, cval, fcr;
 	int ret = 0;
 	unsigned char status;
 
-	if (!info->port.tty || !info->port.tty->termios)
-		return ret;
-	cflag = info->port.tty->termios->c_cflag;
-	if (!(info->ioaddr))
+	cflag = tty->termios->c_cflag;
+	if (!info->ioaddr)
 		return ret;
 
-	if (mxser_set_baud_method[info->port.tty->index] == 0)
-		mxser_set_baud(info, tty_get_baud_rate(info->port.tty));
+	if (mxser_set_baud_method[tty->index] == 0)
+		mxser_set_baud(tty, tty_get_baud_rate(tty));
 
 	/* byte size and parity */
 	switch (cflag & CSIZE) {
@@ -762,9 +759,9 @@ static int mxser_change_speed(struct mxser_port *info,
 			info->MCR |= UART_MCR_AFE;
 		} else {
 			status = inb(info->ioaddr + UART_MSR);
-			if (info->port.tty->hw_stopped) {
+			if (tty->hw_stopped) {
 				if (status & UART_MSR_CTS) {
-					info->port.tty->hw_stopped = 0;
+					tty->hw_stopped = 0;
 					if (info->type != PORT_16550A &&
 							!info->board->chip_flag) {
 						outb(info->IER & ~UART_IER_THRI,
@@ -774,11 +771,11 @@ static int mxser_change_speed(struct mxser_port *info,
 						outb(info->IER, info->ioaddr +
 								UART_IER);
 					}
-					tty_wakeup(info->port.tty);
+					tty_wakeup(tty);
 				}
 			} else {
 				if (!(status & UART_MSR_CTS)) {
-					info->port.tty->hw_stopped = 1;
+					tty->hw_stopped = 1;
 					if ((info->type != PORT_16550A) &&
 							(!info->board->chip_flag)) {
 						info->IER &= ~UART_IER_THRI;
@@ -804,21 +801,21 @@ static int mxser_change_speed(struct mxser_port *info,
 	 * Set up parity check flag
 	 */
 	info->read_status_mask = UART_LSR_OE | UART_LSR_THRE | UART_LSR_DR;
-	if (I_INPCK(info->port.tty))
+	if (I_INPCK(tty))
 		info->read_status_mask |= UART_LSR_FE | UART_LSR_PE;
-	if (I_BRKINT(info->port.tty) || I_PARMRK(info->port.tty))
+	if (I_BRKINT(tty) || I_PARMRK(tty))
 		info->read_status_mask |= UART_LSR_BI;
 
 	info->ignore_status_mask = 0;
 
-	if (I_IGNBRK(info->port.tty)) {
+	if (I_IGNBRK(tty)) {
 		info->ignore_status_mask |= UART_LSR_BI;
 		info->read_status_mask |= UART_LSR_BI;
 		/*
 		 * If we're ignore parity and break indicators, ignore
 		 * overruns too.  (For real raw support).
 		 */
-		if (I_IGNPAR(info->port.tty)) {
+		if (I_IGNPAR(tty)) {
 			info->ignore_status_mask |=
 						UART_LSR_OE |
 						UART_LSR_PE |
@@ -830,16 +827,16 @@ static int mxser_change_speed(struct mxser_port *info,
 		}
 	}
 	if (info->board->chip_flag) {
-		mxser_set_must_xon1_value(info->ioaddr, START_CHAR(info->port.tty));
-		mxser_set_must_xoff1_value(info->ioaddr, STOP_CHAR(info->port.tty));
-		if (I_IXON(info->port.tty)) {
+		mxser_set_must_xon1_value(info->ioaddr, START_CHAR(tty));
+		mxser_set_must_xoff1_value(info->ioaddr, STOP_CHAR(tty));
+		if (I_IXON(tty)) {
 			mxser_enable_must_rx_software_flow_control(
 					info->ioaddr);
 		} else {
 			mxser_disable_must_rx_software_flow_control(
 					info->ioaddr);
 		}
-		if (I_IXOFF(info->port.tty)) {
+		if (I_IXOFF(tty)) {
 			mxser_enable_must_tx_software_flow_control(
 					info->ioaddr);
 		} else {
@@ -855,7 +852,8 @@ static int mxser_change_speed(struct mxser_port *info,
 	return ret;
 }
 
-static void mxser_check_modem_status(struct mxser_port *port, int status)
+static void mxser_check_modem_status(struct tty_struct *tty,
+				struct mxser_port *port, int status)
 {
 	/* update input line counters */
 	if (status & UART_MSR_TERI)
@@ -874,10 +872,11 @@ static void mxser_check_modem_status(struct mxser_port *port, int status)
 			wake_up_interruptible(&port->port.open_wait);
 	}
 
+	tty = tty_port_tty_get(&port->port);
 	if (port->port.flags & ASYNC_CTS_FLOW) {
-		if (port->port.tty->hw_stopped) {
+		if (tty->hw_stopped) {
 			if (status & UART_MSR_CTS) {
-				port->port.tty->hw_stopped = 0;
+				tty->hw_stopped = 0;
 
 				if ((port->type != PORT_16550A) &&
 						(!port->board->chip_flag)) {
@@ -887,11 +886,11 @@ static void mxser_check_modem_status(struct mxser_port *port, int status)
 					outb(port->IER, port->ioaddr +
 							UART_IER);
 				}
-				tty_wakeup(port->port.tty);
+				tty_wakeup(tty);
 			}
 		} else {
 			if (!(status & UART_MSR_CTS)) {
-				port->port.tty->hw_stopped = 1;
+				tty->hw_stopped = 1;
 				if (port->type != PORT_16550A &&
 						!port->board->chip_flag) {
 					port->IER &= ~UART_IER_THRI;
@@ -903,8 +902,9 @@ static void mxser_check_modem_status(struct mxser_port *port, int status)
 	}
 }
 
-static int mxser_startup(struct mxser_port *info)
+static int mxser_startup(struct tty_struct *tty)
 {
+	struct mxser_port *info = tty->driver_data;
 	unsigned long page;
 	unsigned long flags;
 
@@ -921,8 +921,7 @@ static int mxser_startup(struct mxser_port *info)
 	}
 
 	if (!info->ioaddr || !info->type) {
-		if (info->port.tty)
-			set_bit(TTY_IO_ERROR, &info->port.tty->flags);
+		set_bit(TTY_IO_ERROR, &tty->flags);
 		free_page(page);
 		spin_unlock_irqrestore(&info->slock, flags);
 		return 0;
@@ -952,8 +951,8 @@ static int mxser_startup(struct mxser_port *info)
 	if (inb(info->ioaddr + UART_LSR) == 0xff) {
 		spin_unlock_irqrestore(&info->slock, flags);
 		if (capable(CAP_SYS_ADMIN)) {
-			if (info->port.tty)
-				set_bit(TTY_IO_ERROR, &info->port.tty->flags);
+			if (tty)
+				set_bit(TTY_IO_ERROR, &tty->flags);
 			return 0;
 		} else
 			return -ENODEV;
@@ -991,14 +990,13 @@ static int mxser_startup(struct mxser_port *info)
 	(void) inb(info->ioaddr + UART_IIR);
 	(void) inb(info->ioaddr + UART_MSR);
 
-	if (info->port.tty)
-		clear_bit(TTY_IO_ERROR, &info->port.tty->flags);
+	clear_bit(TTY_IO_ERROR, &tty->flags);
 	info->xmit_cnt = info->xmit_head = info->xmit_tail = 0;
 
 	/*
 	 * and set the speed of the serial port
 	 */
-	mxser_change_speed(info, NULL);
+	mxser_change_speed(tty, NULL);
 	info->port.flags |= ASYNC_INITIALIZED;
 	spin_unlock_irqrestore(&info->slock, flags);
 
@@ -1009,8 +1007,9 @@ static int mxser_startup(struct mxser_port *info)
  * This routine will shutdown a serial port; interrupts maybe disabled, and
  * DTR is dropped if the hangup on close termio flag is on.
  */
-static void mxser_shutdown(struct mxser_port *info)
+static void mxser_shutdown(struct tty_struct *tty)
 {
+	struct mxser_port *info = tty->driver_data;
 	unsigned long flags;
 
 	if (!(info->port.flags & ASYNC_INITIALIZED))
@@ -1035,7 +1034,7 @@ static void mxser_shutdown(struct mxser_port *info)
 	info->IER = 0;
 	outb(0x00, info->ioaddr + UART_IER);
 
-	if (!info->port.tty || (info->port.tty->termios->c_cflag & HUPCL))
+	if (tty->termios->c_cflag & HUPCL)
 		info->MCR &= ~(UART_MCR_DTR | UART_MCR_RTS);
 	outb(info->MCR, info->ioaddr + UART_MCR);
 
@@ -1051,8 +1050,7 @@ static void mxser_shutdown(struct mxser_port *info)
 	/* read data port to reset things */
 	(void) inb(info->ioaddr + UART_RX);
 
-	if (info->port.tty)
-		set_bit(TTY_IO_ERROR, &info->port.tty->flags);
+	set_bit(TTY_IO_ERROR, &tty->flags);
 
 	info->port.flags &= ~ASYNC_INITIALIZED;
 
@@ -1084,14 +1082,14 @@ static int mxser_open(struct tty_struct *tty, struct file *filp)
 		return -ENODEV;
 
 	tty->driver_data = info;
-	info->port.tty = tty;
+	tty_port_tty_set(&info->port, tty);
 	/*
 	 * Start up serial port
 	 */
 	spin_lock_irqsave(&info->slock, flags);
 	info->port.count++;
 	spin_unlock_irqrestore(&info->slock, flags);
-	retval = mxser_startup(info);
+	retval = mxser_startup(tty);
 	if (retval)
 		return retval;
 
@@ -1209,13 +1207,13 @@ static void mxser_close(struct tty_struct *tty, struct file *filp)
 				break;
 		}
 	}
-	mxser_shutdown(info);
+	mxser_shutdown(tty);
 
 	mxser_flush_buffer(tty);
 	tty_ldisc_flush(tty);
 
 	tty->closing = 0;
-	info->port.tty = NULL;
+	tty_port_tty_set(&info->port, NULL);
 	if (info->port.blocked_open) {
 		if (info->port.close_delay)
 			schedule_timeout_interruptible(info->port.close_delay);
@@ -1337,12 +1335,13 @@ static int mxser_chars_in_buffer(struct tty_struct *tty)
  * friends of mxser_ioctl()
  * ------------------------------------------------------------
  */
-static int mxser_get_serial_info(struct mxser_port *info,
+static int mxser_get_serial_info(struct tty_struct *tty,
 		struct serial_struct __user *retinfo)
 {
+	struct mxser_port *info = tty->driver_data;
 	struct serial_struct tmp = {
 		.type = info->type,
-		.line = info->port.tty->index,
+		.line = tty->index,
 		.port = info->ioaddr,
 		.irq = info->board->irq,
 		.flags = info->port.flags,
@@ -1357,9 +1356,10 @@ static int mxser_get_serial_info(struct mxser_port *info,
 	return 0;
 }
 
-static int mxser_set_serial_info(struct mxser_port *info,
+static int mxser_set_serial_info(struct tty_struct *tty,
 		struct serial_struct __user *new_info)
 {
+	struct mxser_port *info = tty->driver_data;
 	struct serial_struct new_serial;
 	speed_t baud;
 	unsigned long sl_flags;
@@ -1393,14 +1393,14 @@ static int mxser_set_serial_info(struct mxser_port *info,
 				(new_serial.flags & ASYNC_FLAGS));
 		info->port.close_delay = new_serial.close_delay * HZ / 100;
 		info->port.closing_wait = new_serial.closing_wait * HZ / 100;
-		info->port.tty->low_latency =
-				(info->port.flags & ASYNC_LOW_LATENCY) ? 1 : 0;
+		tty->low_latency = (info->port.flags & ASYNC_LOW_LATENCY)
+								? 1 : 0;
 		if ((info->port.flags & ASYNC_SPD_MASK) == ASYNC_SPD_CUST &&
 				(new_serial.baud_base != info->baud_base ||
 				new_serial.custom_divisor !=
 				info->custom_divisor)) {
 			baud = new_serial.baud_base / new_serial.custom_divisor;
-			tty_encode_baud_rate(info->port.tty, baud, baud);
+			tty_encode_baud_rate(tty, baud, baud);
 		}
 	}
 
@@ -1411,11 +1411,11 @@ static int mxser_set_serial_info(struct mxser_port *info,
 	if (info->port.flags & ASYNC_INITIALIZED) {
 		if (flags != (info->port.flags & ASYNC_SPD_MASK)) {
 			spin_lock_irqsave(&info->slock, sl_flags);
-			mxser_change_speed(info, NULL);
+			mxser_change_speed(tty, NULL);
 			spin_unlock_irqrestore(&info->slock, sl_flags);
 		}
 	} else
-		retval = mxser_startup(info);
+		retval = mxser_startup(tty);
 
 	return retval;
 }
@@ -1461,7 +1461,7 @@ static int mxser_tiocmget(struct tty_struct *tty, struct file *file)
 	spin_lock_irqsave(&info->slock, flags);
 	status = inb(info->ioaddr + UART_MSR);
 	if (status & UART_MSR_ANY_DELTA)
-		mxser_check_modem_status(info, status);
+		mxser_check_modem_status(tty, info, status);
 	spin_unlock_irqrestore(&info->slock, flags);
 	return ((control & UART_MCR_RTS) ? TIOCM_RTS : 0) |
 		    ((control & UART_MCR_DTR) ? TIOCM_DTR : 0) |
@@ -1606,6 +1606,7 @@ static int __init mxser_read_register(int port, unsigned short *regs)
 static int mxser_ioctl_special(unsigned int cmd, void __user *argp)
 {
 	struct mxser_port *port;
+	struct tty_struct *tty;
 	int result, status;
 	unsigned int i, j;
 	int ret = 0;
@@ -1643,12 +1644,14 @@ static int mxser_ioctl_special(unsigned int cmd, void __user *argp)
 
 				if (!port->ioaddr)
 					goto copy;
+				
+				tty = tty_port_tty_get(&port->port);
 
-				if (!port->port.tty || !port->port.tty->termios)
+				if (!tty || !tty->termios)
 					ms.cflag = port->normal_termios.c_cflag;
 				else
-					ms.cflag = port->port.tty->termios->c_cflag;
-
+					ms.cflag = tty->termios->c_cflag;
+				tty_kref_put(tty);
 				status = inb(port->ioaddr + UART_MSR);
 				if (status & UART_MSR_DCD)
 					ms.dcd = 1;
@@ -1704,15 +1707,18 @@ static int mxser_ioctl_special(unsigned int cmd, void __user *argp)
 				me->up_txcnt[p] = port->mon_data.up_txcnt;
 				me->modem_status[p] =
 					port->mon_data.modem_status;
-				me->baudrate[p] = tty_get_baud_rate(port->port.tty);
+				tty = tty_port_tty_get(&port->port);
 
-				if (!port->port.tty || !port->port.tty->termios) {
+				if (!tty || !tty->termios) {
 					cflag = port->normal_termios.c_cflag;
 					iflag = port->normal_termios.c_iflag;
+					me->baudrate[p] = tty_termios_baud_rate(&port->normal_termios);
 				} else {
-					cflag = port->port.tty->termios->c_cflag;
-					iflag = port->port.tty->termios->c_iflag;
+					cflag = tty->termios->c_cflag;
+					iflag = tty->termios->c_iflag;
+					me->baudrate[p] = tty_get_baud_rate(tty);
 				}
+				tty_kref_put(tty);
 
 				me->databits[p] = cflag & CSIZE;
 				me->stopbits[p] = cflag & CSTOPB;
@@ -1822,12 +1828,12 @@ static int mxser_ioctl(struct tty_struct *tty, struct file *file,
 	switch (cmd) {
 	case TIOCGSERIAL:
 		lock_kernel();
-		retval = mxser_get_serial_info(info, argp);
+		retval = mxser_get_serial_info(tty, argp);
 		unlock_kernel();
 		return retval;
 	case TIOCSSERIAL:
 		lock_kernel();
-		retval = mxser_set_serial_info(info, argp);
+		retval = mxser_set_serial_info(tty, argp);
 		unlock_kernel();
 		return retval;
 	case TIOCSERGETLSR:	/* Get line status register */
@@ -1896,7 +1902,7 @@ static int mxser_ioctl(struct tty_struct *tty, struct file *file,
 
 		lock_kernel();
 		status = mxser_get_msr(info->ioaddr, 1, tty->index);
-		mxser_check_modem_status(info, status);
+		mxser_check_modem_status(tty, info, status);
 
 		mcr = inb(info->ioaddr + UART_MCR);
 		if (mcr & MOXA_MUST_MCR_XON_FLAG)
@@ -1909,7 +1915,7 @@ static int mxser_ioctl(struct tty_struct *tty, struct file *file,
 		else
 			info->mon_data.hold_reason |= NPPI_NOTIFY_XOFFXENT;
 
-		if (info->port.tty->hw_stopped)
+		if (tty->hw_stopped)
 			info->mon_data.hold_reason |= NPPI_NOTIFY_CTSHOLD;
 		else
 			info->mon_data.hold_reason &= ~NPPI_NOTIFY_CTSHOLD;
@@ -1958,7 +1964,7 @@ static void mxser_stoprx(struct tty_struct *tty)
 		}
 	}
 
-	if (info->port.tty->termios->c_cflag & CRTSCTS) {
+	if (tty->termios->c_cflag & CRTSCTS) {
 		info->MCR &= ~UART_MCR_RTS;
 		outb(info->MCR, info->ioaddr + UART_MCR);
 	}
@@ -1995,7 +2001,7 @@ static void mxser_unthrottle(struct tty_struct *tty)
 		}
 	}
 
-	if (info->port.tty->termios->c_cflag & CRTSCTS) {
+	if (tty->termios->c_cflag & CRTSCTS) {
 		info->MCR |= UART_MCR_RTS;
 		outb(info->MCR, info->ioaddr + UART_MCR);
 	}
@@ -2040,7 +2046,7 @@ static void mxser_set_termios(struct tty_struct *tty, struct ktermios *old_termi
 	unsigned long flags;
 
 	spin_lock_irqsave(&info->slock, flags);
-	mxser_change_speed(info, old_termios);
+	mxser_change_speed(tty, old_termios);
 	spin_unlock_irqrestore(&info->slock, flags);
 
 	if ((old_termios->c_cflag & CRTSCTS) &&
@@ -2138,10 +2144,10 @@ static void mxser_hangup(struct tty_struct *tty)
 	struct mxser_port *info = tty->driver_data;
 
 	mxser_flush_buffer(tty);
-	mxser_shutdown(info);
+	mxser_shutdown(tty);
 	info->port.count = 0;
 	info->port.flags &= ~ASYNC_NORMAL_ACTIVE;
-	info->port.tty = NULL;
+	tty_port_tty_set(&info->port, NULL);
 	wake_up_interruptible(&info->port.open_wait);
 }
 
@@ -2164,9 +2170,9 @@ static int mxser_rs_break(struct tty_struct *tty, int break_state)
 	return 0;
 }
 
-static void mxser_receive_chars(struct mxser_port *port, int *status)
+static void mxser_receive_chars(struct tty_struct *tty,
+				struct mxser_port *port, int *status)
 {
-	struct tty_struct *tty = port->port.tty;
 	unsigned char ch, gdl;
 	int ignored = 0;
 	int cnt = 0;
@@ -2174,9 +2180,8 @@ static void mxser_receive_chars(struct mxser_port *port, int *status)
 	int max = 256;
 
 	recv_room = tty->receive_room;
-	if ((recv_room == 0) && (!port->ldisc_stop_rx))
+	if (recv_room == 0 && !port->ldisc_stop_rx)
 		mxser_stoprx(tty);
-
 	if (port->board->chip_flag != MOXA_OTHER_UART) {
 
 		if (*status & UART_LSR_SPECIAL)
@@ -2253,7 +2258,7 @@ intr_old:
 	} while (*status & UART_LSR_DR);
 
 end_intr:
-	mxvar_log.rxcnt[port->port.tty->index] += cnt;
+	mxvar_log.rxcnt[tty->index] += cnt;
 	port->mon_data.rxcnt += cnt;
 	port->mon_data.up_rxcnt += cnt;
 
@@ -2267,14 +2272,14 @@ end_intr:
 	spin_lock(&port->slock);
 }
 
-static void mxser_transmit_chars(struct mxser_port *port)
+static void mxser_transmit_chars(struct tty_struct *tty, struct mxser_port *port)
 {
 	int count, cnt;
 
 	if (port->x_char) {
 		outb(port->x_char, port->ioaddr + UART_TX);
 		port->x_char = 0;
-		mxvar_log.txcnt[port->port.tty->index]++;
+		mxvar_log.txcnt[tty->index]++;
 		port->mon_data.txcnt++;
 		port->mon_data.up_txcnt++;
 		port->icount.tx++;
@@ -2284,8 +2289,8 @@ static void mxser_transmit_chars(struct mxser_port *port)
 	if (port->port.xmit_buf == NULL)
 		return;
 
-	if ((port->xmit_cnt <= 0) || port->port.tty->stopped ||
-			(port->port.tty->hw_stopped &&
+	if (port->xmit_cnt <= 0 || tty->stopped ||
+			(tty->hw_stopped &&
 			(port->type != PORT_16550A) &&
 			(!port->board->chip_flag))) {
 		port->IER &= ~UART_IER_THRI;
@@ -2302,14 +2307,14 @@ static void mxser_transmit_chars(struct mxser_port *port)
 		if (--port->xmit_cnt <= 0)
 			break;
 	} while (--count > 0);
-	mxvar_log.txcnt[port->port.tty->index] += (cnt - port->xmit_cnt);
+	mxvar_log.txcnt[tty->index] += (cnt - port->xmit_cnt);
 
 	port->mon_data.txcnt += (cnt - port->xmit_cnt);
 	port->mon_data.up_txcnt += (cnt - port->xmit_cnt);
 	port->icount.tx += (cnt - port->xmit_cnt);
 
-	if (port->xmit_cnt < WAKEUP_CHARS)
-		tty_wakeup(port->port.tty);
+	if (port->xmit_cnt < WAKEUP_CHARS && tty)
+		tty_wakeup(tty);
 
 	if (port->xmit_cnt <= 0) {
 		port->IER &= ~UART_IER_THRI;
@@ -2328,6 +2333,7 @@ static irqreturn_t mxser_interrupt(int irq, void *dev_id)
 	int max, irqbits, bits, msr;
 	unsigned int int_cnt, pass_counter = 0;
 	int handled = IRQ_NONE;
+	struct tty_struct *tty;
 
 	for (i = 0; i < MXSER_BOARDS; i++)
 		if (dev_id == &mxser_boards[i]) {
@@ -2360,13 +2366,15 @@ static irqreturn_t mxser_interrupt(int irq, void *dev_id)
 				if (iir & UART_IIR_NO_INT)
 					break;
 				iir &= MOXA_MUST_IIR_MASK;
-				if (!port->port.tty ||
+				tty = tty_port_tty_get(&port->port);
+				if (!tty ||
 						(port->port.flags & ASYNC_CLOSING) ||
 						!(port->port.flags &
 							ASYNC_INITIALIZED)) {
 					status = inb(port->ioaddr + UART_LSR);
 					outb(0x27, port->ioaddr + UART_FCR);
 					inb(port->ioaddr + UART_MSR);
+					tty_kref_put(tty);
 					break;
 				}
 
@@ -2387,27 +2395,28 @@ static irqreturn_t mxser_interrupt(int irq, void *dev_id)
 					    iir == MOXA_MUST_IIR_RDA ||
 					    iir == MOXA_MUST_IIR_RTO ||
 					    iir == MOXA_MUST_IIR_LSR)
-						mxser_receive_chars(port,
+						mxser_receive_chars(tty, port,
 								&status);
 
 				} else {
 					status &= port->read_status_mask;
 					if (status & UART_LSR_DR)
-						mxser_receive_chars(port,
+						mxser_receive_chars(tty, port,
 								&status);
 				}
 				msr = inb(port->ioaddr + UART_MSR);
 				if (msr & UART_MSR_ANY_DELTA)
-					mxser_check_modem_status(port, msr);
+					mxser_check_modem_status(tty, port, msr);
 
 				if (port->board->chip_flag) {
 					if (iir == 0x02 && (status &
 								UART_LSR_THRE))
-						mxser_transmit_chars(port);
+						mxser_transmit_chars(tty, port);
 				} else {
 					if (status & UART_LSR_THRE)
-						mxser_transmit_chars(port);
+						mxser_transmit_chars(tty, port);
 				}
+				tty_kref_put(tty);
 			} while (int_cnt++ < MXSER_ISR_PASS_LIMIT);
 			spin_unlock(&port->slock);
 		}
diff --git a/drivers/char/n_hdlc.c b/drivers/char/n_hdlc.c
index 69ec6399c714..bacb3e2872ae 100644
--- a/drivers/char/n_hdlc.c
+++ b/drivers/char/n_hdlc.c
@@ -764,7 +764,7 @@ static int n_hdlc_tty_ioctl(struct tty_struct *tty, struct file *file,
 		break;
 
 	default:
-		error = n_tty_ioctl (tty, file, cmd, arg);
+		error = n_tty_ioctl_helper(tty, file, cmd, arg);
 		break;
 	}
 	return error;
diff --git a/drivers/char/n_r3964.c b/drivers/char/n_r3964.c
index ae377aa473ba..4a8215a89ad3 100644
--- a/drivers/char/n_r3964.c
+++ b/drivers/char/n_r3964.c
@@ -372,14 +372,8 @@ static void remove_from_rx_queue(struct r3964_info *pInfo,
 static void put_char(struct r3964_info *pInfo, unsigned char ch)
 {
 	struct tty_struct *tty = pInfo->tty;
-
-	if (tty == NULL)
-		return;
-
 	/* FIXME: put_char should not be called from an IRQ */
-	if (tty->ops->put_char) {
-		tty->ops->put_char(tty, ch);
-	}
+	tty_put_char(tty, ch);
 	pInfo->bcc ^= ch;
 }
 
diff --git a/drivers/char/n_tty.c b/drivers/char/n_tty.c
index 708c2b1dbe51..efbfe9612658 100644
--- a/drivers/char/n_tty.c
+++ b/drivers/char/n_tty.c
@@ -26,7 +26,7 @@
  *
  * 2002/03/18   Implemented n_tty_wakeup to send SIGIO POLL_OUTs to
  *		waiting writing processes-Sapan Bhatia <sapan@corewars.org>.
- *		Also fixed a bug in BLOCKING mode where write_chan returns
+ *		Also fixed a bug in BLOCKING mode where n_tty_write returns
  *		EAGAIN
  */
 
@@ -99,6 +99,7 @@ static inline int tty_put_user(struct tty_struct *tty, unsigned char x,
 
 static void n_tty_set_room(struct tty_struct *tty)
 {
+	/* tty->read_cnt is not read locked ? */
 	int	left = N_TTY_BUF_SIZE - tty->read_cnt - 1;
 
 	/*
@@ -121,6 +122,16 @@ static void put_tty_queue_nolock(unsigned char c, struct tty_struct *tty)
 	}
 }
 
+/**
+ *	put_tty_queue		-	add character to tty
+ *	@c: character
+ *	@tty: tty device
+ *
+ *	Add a character to the tty read_buf queue. This is done under the
+ *	read_lock to serialize character addition and also to protect us
+ *	against parallel reads or flushes
+ */
+
 static void put_tty_queue(unsigned char c, struct tty_struct *tty)
 {
 	unsigned long flags;
@@ -137,14 +148,11 @@ static void put_tty_queue(unsigned char c, struct tty_struct *tty)
  *	check_unthrottle	-	allow new receive data
  *	@tty; tty device
  *
- *	Check whether to call the driver.unthrottle function.
- *	We test the TTY_THROTTLED bit first so that it always
- *	indicates the current state. The decision about whether
- *	it is worth allowing more input has been taken by the caller.
+ *	Check whether to call the driver unthrottle functions
+ *
  *	Can sleep, may be called under the atomic_read_lock mutex but
  *	this is not guaranteed.
  */
-
 static void check_unthrottle(struct tty_struct *tty)
 {
 	if (tty->count)
@@ -158,6 +166,8 @@ static void check_unthrottle(struct tty_struct *tty)
  *	Reset the read buffer counters, clear the flags,
  *	and make sure the driver is unthrottled. Called
  *	from n_tty_open() and n_tty_flush_buffer().
+ *
+ *	Locking: tty_read_lock for read fields.
  */
 static void reset_buffer_flags(struct tty_struct *tty)
 {
@@ -181,7 +191,7 @@ static void reset_buffer_flags(struct tty_struct *tty)
  *	at hangup) or when the N_TTY line discipline internally has to
  *	clean the pending queue (for example some signals).
  *
- *	Locking: ctrl_lock
+ *	Locking: ctrl_lock, read_lock.
  */
 
 static void n_tty_flush_buffer(struct tty_struct *tty)
@@ -207,6 +217,8 @@ static void n_tty_flush_buffer(struct tty_struct *tty)
  *
  *	Report the number of characters buffered to be delivered to user
  *	at this instant in time.
+ *
+ *	Locking: read_lock
  */
 
 static ssize_t n_tty_chars_in_buffer(struct tty_struct *tty)
@@ -346,7 +358,7 @@ static int opost(unsigned char c, struct tty_struct *tty)
  *	the simple cases normally found and helps to generate blocks of
  *	symbols for the console driver and thus improve performance.
  *
- *	Called from write_chan under the tty layer write lock. Relies
+ *	Called from n_tty_write under the tty layer write lock. Relies
  *	on lock_kernel for the tty->column state.
  */
 
@@ -410,6 +422,8 @@ break_out:
  *
  *	Echo user input back onto the screen. This must be called only when
  *	L_ECHO(tty) is true. Called from the driver receive_buf path.
+ *
+ *	Relies on BKL for tty column locking
  */
 
 static void echo_char(unsigned char c, struct tty_struct *tty)
@@ -422,6 +436,12 @@ static void echo_char(unsigned char c, struct tty_struct *tty)
 		opost(c, tty);
 }
 
+/**
+ *	finsh_erasing		-	complete erase
+ *	@tty: tty doing the erase
+ *
+ *	Relies on BKL for tty column locking
+ */
 static inline void finish_erasing(struct tty_struct *tty)
 {
 	if (tty->erasing) {
@@ -439,6 +459,8 @@ static inline void finish_erasing(struct tty_struct *tty)
  *	Perform erase and necessary output when an erase character is
  *	present in the stream from the driver layer. Handles the complexities
  *	of UTF-8 multibyte symbols.
+ *
+ *	Locking: read_lock for tty buffers, BKL for column/erasing state
  */
 
 static void eraser(unsigned char c, struct tty_struct *tty)
@@ -447,6 +469,7 @@ static void eraser(unsigned char c, struct tty_struct *tty)
 	int head, seen_alnums, cnt;
 	unsigned long flags;
 
+	/* FIXME: locking needed ? */
 	if (tty->read_head == tty->canon_head) {
 		/* opost('\a', tty); */		/* what do you think? */
 		return;
@@ -481,6 +504,7 @@ static void eraser(unsigned char c, struct tty_struct *tty)
 	}
 
 	seen_alnums = 0;
+	/* FIXME: Locking ?? */
 	while (tty->read_head != tty->canon_head) {
 		head = tty->read_head;
 
@@ -583,6 +607,8 @@ static void eraser(unsigned char c, struct tty_struct *tty)
  *	may caus terminal flushing to take place according to the termios
  *	settings and character used. Called from the driver receive_buf
  *	path so serialized.
+ *
+ *	Locking: ctrl_lock, read_lock (both via flush buffer)
  */
 
 static inline void isig(int sig, struct tty_struct *tty, int flush)
@@ -1007,12 +1033,26 @@ int is_ignored(int sig)
  *	and is protected from re-entry by the tty layer. The user is
  *	guaranteed that this function will not be re-entered or in progress
  *	when the ldisc is closed.
+ *
+ *	Locking: Caller holds tty->termios_mutex
  */
 
 static void n_tty_set_termios(struct tty_struct *tty, struct ktermios *old)
 {
-	if (!tty)
-		return;
+	int canon_change = 1;
+	BUG_ON(!tty);
+
+	if (old)
+		canon_change = (old->c_lflag ^ tty->termios->c_lflag) & ICANON;
+	if (canon_change) {
+		memset(&tty->read_flags, 0, sizeof tty->read_flags);
+		tty->canon_head = tty->read_tail;
+		tty->canon_data = 0;
+		tty->erasing = 0;
+	}
+
+	if (canon_change && !L_ICANON(tty) && tty->read_cnt)
+		wake_up_interruptible(&tty->read_wait);
 
 	tty->icanon = (L_ICANON(tty) != 0);
 	if (test_bit(TTY_HW_COOK_IN, &tty->flags)) {
@@ -1143,7 +1183,7 @@ static inline int input_available_p(struct tty_struct *tty, int amt)
  *	@b: user data
  *	@nr: size of data
  *
- *	Helper function to speed up read_chan.  It is only called when
+ *	Helper function to speed up n_tty_read.  It is only called when
  *	ICANON is off; it copies characters straight from the tty queue to
  *	user space directly.  It can be profitably called twice; once to
  *	drain the space from the tail pointer to the (physical) end of the
@@ -1210,7 +1250,7 @@ static int job_control(struct tty_struct *tty, struct file *file)
 	if (file->f_op->write != redirected_tty_write &&
 	    current->signal->tty == tty) {
 		if (!tty->pgrp)
-			printk(KERN_ERR "read_chan: no tty->pgrp!\n");
+			printk(KERN_ERR "n_tty_read: no tty->pgrp!\n");
 		else if (task_pgrp(current) != tty->pgrp) {
 			if (is_ignored(SIGTTIN) ||
 			    is_current_pgrp_orphaned())
@@ -1225,7 +1265,7 @@ static int job_control(struct tty_struct *tty, struct file *file)
 
 
 /**
- *	read_chan		-	read function for tty
+ *	n_tty_read		-	read function for tty
  *	@tty: tty device
  *	@file: file object
  *	@buf: userspace buffer pointer
@@ -1239,7 +1279,7 @@ static int job_control(struct tty_struct *tty, struct file *file)
  *	This code must be sure never to sleep through a hangup.
  */
 
-static ssize_t read_chan(struct tty_struct *tty, struct file *file,
+static ssize_t n_tty_read(struct tty_struct *tty, struct file *file,
 			 unsigned char __user *buf, size_t nr)
 {
 	unsigned char __user *b = buf;
@@ -1254,10 +1294,7 @@ static ssize_t read_chan(struct tty_struct *tty, struct file *file,
 
 do_it_again:
 
-	if (!tty->read_buf) {
-		printk(KERN_ERR "n_tty_read_chan: read_buf == NULL?!?\n");
-		return -EIO;
-	}
+	BUG_ON(!tty->read_buf);
 
 	c = job_control(tty, file);
 	if (c < 0)
@@ -1444,7 +1481,7 @@ do_it_again:
 }
 
 /**
- *	write_chan		-	write function for tty
+ *	n_tty_write		-	write function for tty
  *	@tty: tty device
  *	@file: file object
  *	@buf: userspace buffer pointer
@@ -1458,7 +1495,7 @@ do_it_again:
  *	This code must be sure never to sleep through a hangup.
  */
 
-static ssize_t write_chan(struct tty_struct *tty, struct file *file,
+static ssize_t n_tty_write(struct tty_struct *tty, struct file *file,
 			  const unsigned char *buf, size_t nr)
 {
 	const unsigned char *b = buf;
@@ -1532,7 +1569,7 @@ break_out:
 }
 
 /**
- *	normal_poll		-	poll method for N_TTY
+ *	n_tty_poll		-	poll method for N_TTY
  *	@tty: terminal device
  *	@file: file accessing it
  *	@wait: poll table
@@ -1545,7 +1582,7 @@ break_out:
  *	Called without the kernel lock held - fine
  */
 
-static unsigned int normal_poll(struct tty_struct *tty, struct file *file,
+static unsigned int n_tty_poll(struct tty_struct *tty, struct file *file,
 							poll_table *wait)
 {
 	unsigned int mask = 0;
@@ -1573,6 +1610,44 @@ static unsigned int normal_poll(struct tty_struct *tty, struct file *file,
 	return mask;
 }
 
+static unsigned long inq_canon(struct tty_struct *tty)
+{
+	int nr, head, tail;
+
+	if (!tty->canon_data)
+		return 0;
+	head = tty->canon_head;
+	tail = tty->read_tail;
+	nr = (head - tail) & (N_TTY_BUF_SIZE-1);
+	/* Skip EOF-chars.. */
+	while (head != tail) {
+		if (test_bit(tail, tty->read_flags) &&
+		    tty->read_buf[tail] == __DISABLED_CHAR)
+			nr--;
+		tail = (tail+1) & (N_TTY_BUF_SIZE-1);
+	}
+	return nr;
+}
+
+static int n_tty_ioctl(struct tty_struct *tty, struct file *file,
+		       unsigned int cmd, unsigned long arg)
+{
+	int retval;
+
+	switch (cmd) {
+	case TIOCOUTQ:
+		return put_user(tty_chars_in_buffer(tty), (int __user *) arg);
+	case TIOCINQ:
+		/* FIXME: Locking */
+		retval = tty->read_cnt;
+		if (L_ICANON(tty))
+			retval = inq_canon(tty);
+		return put_user(retval, (unsigned int __user *) arg);
+	default:
+		return n_tty_ioctl_helper(tty, file, cmd, arg);
+	}
+}
+
 struct tty_ldisc_ops tty_ldisc_N_TTY = {
 	.magic           = TTY_LDISC_MAGIC,
 	.name            = "n_tty",
@@ -1580,11 +1655,11 @@ struct tty_ldisc_ops tty_ldisc_N_TTY = {
 	.close           = n_tty_close,
 	.flush_buffer    = n_tty_flush_buffer,
 	.chars_in_buffer = n_tty_chars_in_buffer,
-	.read            = read_chan,
-	.write           = write_chan,
+	.read            = n_tty_read,
+	.write           = n_tty_write,
 	.ioctl           = n_tty_ioctl,
 	.set_termios     = n_tty_set_termios,
-	.poll            = normal_poll,
+	.poll            = n_tty_poll,
 	.receive_buf     = n_tty_receive_buf,
 	.write_wakeup    = n_tty_write_wakeup
 };
diff --git a/drivers/char/nozomi.c b/drivers/char/nozomi.c
index 66a0f931c66c..9a34a1935283 100644
--- a/drivers/char/nozomi.c
+++ b/drivers/char/nozomi.c
@@ -1599,7 +1599,10 @@ static int ntty_open(struct tty_struct *tty, struct file *file)
 	return 0;
 }
 
-/* Called when the userspace process close the tty, /dev/noz*. */
+/* Called when the userspace process close the tty, /dev/noz*. Also
+   called immediately if ntty_open fails in which case tty->driver_data
+   will be NULL an we exit by the first return */
+
 static void ntty_close(struct tty_struct *tty, struct file *file)
 {
 	struct nozomi *dc = get_dc_by_tty(tty);
diff --git a/drivers/char/pcmcia/ipwireless/tty.c b/drivers/char/pcmcia/ipwireless/tty.c
index 3a23e7694d55..569f2f7743a7 100644
--- a/drivers/char/pcmcia/ipwireless/tty.c
+++ b/drivers/char/pcmcia/ipwireless/tty.c
@@ -276,6 +276,7 @@ static int ipw_write_room(struct tty_struct *linux_tty)
 	struct ipw_tty *tty = linux_tty->driver_data;
 	int room;
 
+	/* FIXME: Exactly how is the tty object locked here .. */
 	if (!tty)
 		return -ENODEV;
 
@@ -397,6 +398,7 @@ static int set_control_lines(struct ipw_tty *tty, unsigned int set,
 static int ipw_tiocmget(struct tty_struct *linux_tty, struct file *file)
 {
 	struct ipw_tty *tty = linux_tty->driver_data;
+	/* FIXME: Exactly how is the tty object locked here .. */
 
 	if (!tty)
 		return -ENODEV;
@@ -412,6 +414,7 @@ ipw_tiocmset(struct tty_struct *linux_tty, struct file *file,
 	     unsigned int set, unsigned int clear)
 {
 	struct ipw_tty *tty = linux_tty->driver_data;
+	/* FIXME: Exactly how is the tty object locked here .. */
 
 	if (!tty)
 		return -ENODEV;
@@ -433,6 +436,8 @@ static int ipw_ioctl(struct tty_struct *linux_tty, struct file *file,
 	if (!tty->open_count)
 		return -EINVAL;
 
+	/* FIXME: Exactly how is the tty object locked here .. */
+
 	switch (cmd) {
 	case TIOCGSERIAL:
 		return ipwireless_get_serial_info(tty, (void __user *) arg);
@@ -467,13 +472,6 @@ static int ipw_ioctl(struct tty_struct *linux_tty, struct file *file,
 			}
 			return 0;
 
-		case TCGETS:
-		case TCGETA:
-			return n_tty_ioctl(linux_tty, file, cmd, arg);
-
-		case TCFLSH:
-			return n_tty_ioctl(linux_tty, file, cmd, arg);
-
 		case FIONREAD:
 			{
 				int val = 0;
@@ -482,10 +480,11 @@ static int ipw_ioctl(struct tty_struct *linux_tty, struct file *file,
 					return -EFAULT;
 			}
 			return 0;
+		case TCFLSH:
+			return tty_perform_flush(linux_tty, arg);
 		}
 	}
-
-	return -ENOIOCTLCMD;
+	return tty_mode_ioctl(linux_tty, file, cmd , arg);
 }
 
 static int add_tty(dev_node_t *nodesp, int j,
@@ -588,6 +587,8 @@ void ipwireless_tty_free(struct ipw_tty *tty)
 				tty_hangup(ttyj->linux_tty);
 				/* Wait till the tty_hangup has completed */
 				flush_scheduled_work();
+				/* FIXME: Exactly how is the tty object locked here
+				   against a parallel ioctl etc */
 				mutex_lock(&ttyj->ipw_tty_mutex);
 			}
 			while (ttyj->open_count)
diff --git a/drivers/char/pty.c b/drivers/char/pty.c
index 76b27932d229..6d4582712b1f 100644
--- a/drivers/char/pty.c
+++ b/drivers/char/pty.c
@@ -8,10 +8,12 @@
  *  Added TTY_DO_WRITE_WAKEUP to enable n_tty to send POLL_OUT to
  *      waiting writers -- Sapan Bhatia <sapan@corewars.org>
  *
- *
+ *  When reading this code see also fs/devpts. In particular note that the
+ *  driver_data field is used by the devpts side as a binding to the devpts
+ *  inode.
  */
 
-#include <linux/module.h>	/* For EXPORT_SYMBOL */
+#include <linux/module.h>
 
 #include <linux/errno.h>
 #include <linux/interrupt.h>
@@ -23,26 +25,25 @@
 #include <linux/mm.h>
 #include <linux/init.h>
 #include <linux/sysctl.h>
-
-#include <asm/uaccess.h>
-#include <asm/system.h>
+#include <linux/device.h>
+#include <linux/uaccess.h>
 #include <linux/bitops.h>
 #include <linux/devpts_fs.h>
 
+#include <asm/system.h>
+
 /* These are global because they are accessed in tty_io.c */
 #ifdef CONFIG_UNIX98_PTYS
 struct tty_driver *ptm_driver;
 static struct tty_driver *pts_driver;
 #endif
 
-static void pty_close(struct tty_struct * tty, struct file * filp)
+static void pty_close(struct tty_struct *tty, struct file *filp)
 {
-	if (!tty)
-		return;
-	if (tty->driver->subtype == PTY_TYPE_MASTER) {
-		if (tty->count > 1)
-			printk("master pty_close: count = %d!!\n", tty->count);
-	} else {
+	BUG_ON(!tty);
+	if (tty->driver->subtype == PTY_TYPE_MASTER)
+		WARN_ON(tty->count > 1);
+	else {
 		if (tty->count > 2)
 			return;
 	}
@@ -59,7 +60,7 @@ static void pty_close(struct tty_struct * tty, struct file * filp)
 		set_bit(TTY_OTHER_CLOSED, &tty->flags);
 #ifdef CONFIG_UNIX98_PTYS
 		if (tty->driver == ptm_driver)
-			devpts_pty_kill(tty->index);
+			devpts_pty_kill(tty->link);
 #endif
 		tty_vhangup(tty->link);
 	}
@@ -69,13 +70,13 @@ static void pty_close(struct tty_struct * tty, struct file * filp)
  * The unthrottle routine is called by the line discipline to signal
  * that it can receive more characters.  For PTY's, the TTY_THROTTLED
  * flag is always set, to force the line discipline to always call the
- * unthrottle routine when there are fewer than TTY_THRESHOLD_UNTHROTTLE 
+ * unthrottle routine when there are fewer than TTY_THRESHOLD_UNTHROTTLE
  * characters in the queue.  This is necessary since each time this
  * happens, we need to wake up any sleeping processes that could be
  * (1) trying to send data to the pty, or (2) waiting in wait_until_sent()
  * for the pty buffer to be drained.
  */
-static void pty_unthrottle(struct tty_struct * tty)
+static void pty_unthrottle(struct tty_struct *tty)
 {
 	struct tty_struct *o_tty = tty->link;
 
@@ -87,7 +88,7 @@ static void pty_unthrottle(struct tty_struct * tty)
 }
 
 /*
- * WSH 05/24/97: modified to 
+ * WSH 05/24/97: modified to
  *   (1) use space in tty->flip instead of a shared temp buffer
  *	 The flip buffers aren't being used for a pty, so there's lots
  *	 of space available.  The buffer is protected by a per-pty
@@ -100,7 +101,8 @@ static void pty_unthrottle(struct tty_struct * tty)
  * not our partners. We can't just take the other one blindly without
  * risking deadlocks.
  */
-static int pty_write(struct tty_struct * tty, const unsigned char *buf, int count)
+static int pty_write(struct tty_struct *tty, const unsigned char *buf,
+								int count)
 {
 	struct tty_struct *to = tty->link;
 	int	c;
@@ -112,7 +114,7 @@ static int pty_write(struct tty_struct * tty, const unsigned char *buf, int coun
 	if (c > count)
 		c = count;
 	to->ldisc.ops->receive_buf(to, buf, NULL, c);
-	
+
 	return c;
 }
 
@@ -128,17 +130,17 @@ static int pty_write_room(struct tty_struct *tty)
 
 /*
  *	WSH 05/24/97:  Modified for asymmetric MASTER/SLAVE behavior
- *	The chars_in_buffer() value is used by the ldisc select() function 
+ *	The chars_in_buffer() value is used by the ldisc select() function
  *	to hold off writing when chars_in_buffer > WAKEUP_CHARS (== 256).
  *	The pty driver chars_in_buffer() Master/Slave must behave differently:
  *
  *      The Master side needs to allow typed-ahead commands to accumulate
  *      while being canonicalized, so we report "our buffer" as empty until
  *	some threshold is reached, and then report the count. (Any count >
- *	WAKEUP_CHARS is regarded by select() as "full".)  To avoid deadlock 
- *	the count returned must be 0 if no canonical data is available to be 
+ *	WAKEUP_CHARS is regarded by select() as "full".)  To avoid deadlock
+ *	the count returned must be 0 if no canonical data is available to be
  *	read. (The N_TTY ldisc.chars_in_buffer now knows this.)
- *  
+ *
  *	The Slave side passes all characters in raw mode to the Master side's
  *	buffer where they can be read immediately, so in this case we can
  *	return the true count in the buffer.
@@ -155,21 +157,22 @@ static int pty_chars_in_buffer(struct tty_struct *tty)
 	/* The ldisc must report 0 if no characters available to be read */
 	count = to->ldisc.ops->chars_in_buffer(to);
 
-	if (tty->driver->subtype == PTY_TYPE_SLAVE) return count;
+	if (tty->driver->subtype == PTY_TYPE_SLAVE)
+		return count;
 
-	/* Master side driver ... if the other side's read buffer is less than 
+	/* Master side driver ... if the other side's read buffer is less than
 	 * half full, return 0 to allow writers to proceed; otherwise return
-	 * the count.  This leaves a comfortable margin to avoid overflow, 
+	 * the count.  This leaves a comfortable margin to avoid overflow,
 	 * and still allows half a buffer's worth of typed-ahead commands.
 	 */
-	return ((count < N_TTY_BUF_SIZE/2) ? 0 : count);
+	return (count < N_TTY_BUF_SIZE/2) ? 0 : count;
 }
 
 /* Set the lock flag on a pty */
-static int pty_set_lock(struct tty_struct *tty, int __user * arg)
+static int pty_set_lock(struct tty_struct *tty, int __user *arg)
 {
 	int val;
-	if (get_user(val,arg))
+	if (get_user(val, arg))
 		return -EFAULT;
 	if (val)
 		set_bit(TTY_PTY_LOCK, &tty->flags);
@@ -182,13 +185,13 @@ static void pty_flush_buffer(struct tty_struct *tty)
 {
 	struct tty_struct *to = tty->link;
 	unsigned long flags;
-	
+
 	if (!to)
 		return;
-	
+
 	if (to->ldisc.ops->flush_buffer)
 		to->ldisc.ops->flush_buffer(to);
-	
+
 	if (to->packet) {
 		spin_lock_irqsave(&tty->ctrl_lock, flags);
 		tty->ctrl_status |= TIOCPKT_FLUSHWRITE;
@@ -197,7 +200,7 @@ static void pty_flush_buffer(struct tty_struct *tty)
 	}
 }
 
-static int pty_open(struct tty_struct *tty, struct file * filp)
+static int pty_open(struct tty_struct *tty, struct file *filp)
 {
 	int	retval = -ENODEV;
 
@@ -220,13 +223,65 @@ out:
 	return retval;
 }
 
-static void pty_set_termios(struct tty_struct *tty, struct ktermios *old_termios)
+static void pty_set_termios(struct tty_struct *tty,
+					struct ktermios *old_termios)
+{
+	tty->termios->c_cflag &= ~(CSIZE | PARENB);
+	tty->termios->c_cflag |= (CS8 | CREAD);
+}
+
+static int pty_install(struct tty_driver *driver, struct tty_struct *tty)
 {
-        tty->termios->c_cflag &= ~(CSIZE | PARENB);
-        tty->termios->c_cflag |= (CS8 | CREAD);
+	struct tty_struct *o_tty;
+	int idx = tty->index;
+	int retval;
+
+	o_tty = alloc_tty_struct();
+	if (!o_tty)
+		return -ENOMEM;
+	if (!try_module_get(driver->other->owner)) {
+		/* This cannot in fact currently happen */
+		free_tty_struct(o_tty);
+		return -ENOMEM;
+	}
+	initialize_tty_struct(o_tty, driver->other, idx);
+
+	/* We always use new tty termios data so we can do this
+	   the easy way .. */
+	retval = tty_init_termios(tty);
+	if (retval)
+		goto free_mem_out;
+
+	retval = tty_init_termios(o_tty);
+	if (retval) {
+		tty_free_termios(tty);
+		goto free_mem_out;
+	}
+
+	/*
+	 * Everything allocated ... set up the o_tty structure.
+	 */
+	driver->other->ttys[idx] = o_tty;
+	tty_driver_kref_get(driver->other);
+	if (driver->subtype == PTY_TYPE_MASTER)
+		o_tty->count++;
+	/* Establish the links in both directions */
+	tty->link   = o_tty;
+	o_tty->link = tty;
+
+	tty_driver_kref_get(driver);
+	tty->count++;
+	driver->ttys[idx] = tty;
+	return 0;
+free_mem_out:
+	module_put(o_tty->driver->owner);
+	free_tty_struct(o_tty);
+	return -ENOMEM;
 }
 
+
 static const struct tty_operations pty_ops = {
+	.install = pty_install,
 	.open = pty_open,
 	.close = pty_close,
 	.write = pty_write,
@@ -329,8 +384,11 @@ static inline void legacy_pty_init(void) { }
  * Otherwise one can eat up all kernel memory by opening /dev/ptmx repeatedly.
  */
 int pty_limit = NR_UNIX98_PTY_DEFAULT;
-static int pty_limit_min = 0;
+static int pty_limit_min;
 static int pty_limit_max = NR_UNIX98_PTY_MAX;
+static int pty_count;
+
+static struct cdev ptmx_cdev;
 
 static struct ctl_table pty_table[] = {
 	{
@@ -348,6 +406,7 @@ static struct ctl_table pty_table[] = {
 		.procname	= "nr",
 		.maxlen		= sizeof(int),
 		.mode		= 0444,
+		.data		= &pty_count,
 		.proc_handler	= &proc_dointvec,
 	}, {
 		.ctl_name	= 0
@@ -388,7 +447,127 @@ static int pty_unix98_ioctl(struct tty_struct *tty, struct file *file,
 	return -ENOIOCTLCMD;
 }
 
+/**
+ *	ptm_unix98_lookup	-	find a pty master
+ *	@driver: ptm driver
+ *	@idx: tty index
+ *
+ *	Look up a pty master device. Called under the tty_mutex for now.
+ *	This provides our locking.
+ */
+
+static struct tty_struct *ptm_unix98_lookup(struct tty_driver *driver,
+		struct inode *ptm_inode, int idx)
+{
+	struct tty_struct *tty = devpts_get_tty(ptm_inode, idx);
+	if (tty)
+		tty = tty->link;
+	return tty;
+}
+
+/**
+ *	pts_unix98_lookup	-	find a pty slave
+ *	@driver: pts driver
+ *	@idx: tty index
+ *
+ *	Look up a pty master device. Called under the tty_mutex for now.
+ *	This provides our locking.
+ */
+
+static struct tty_struct *pts_unix98_lookup(struct tty_driver *driver,
+		struct inode *pts_inode, int idx)
+{
+	struct tty_struct *tty = devpts_get_tty(pts_inode, idx);
+	/* Master must be open before slave */
+	if (!tty)
+		return ERR_PTR(-EIO);
+	return tty;
+}
+
+static void pty_unix98_shutdown(struct tty_struct *tty)
+{
+	/* We have our own method as we don't use the tty index */
+	kfree(tty->termios);
+}
+
+/* We have no need to install and remove our tty objects as devpts does all
+   the work for us */
+
+static int pty_unix98_install(struct tty_driver *driver, struct tty_struct *tty)
+{
+	struct tty_struct *o_tty;
+	int idx = tty->index;
+
+	o_tty = alloc_tty_struct();
+	if (!o_tty)
+		return -ENOMEM;
+	if (!try_module_get(driver->other->owner)) {
+		/* This cannot in fact currently happen */
+		free_tty_struct(o_tty);
+		return -ENOMEM;
+	}
+	initialize_tty_struct(o_tty, driver->other, idx);
+
+	tty->termios = kzalloc(sizeof(struct ktermios[2]), GFP_KERNEL);
+	if (tty->termios == NULL)
+		goto free_mem_out;
+	*tty->termios = driver->init_termios;
+	tty->termios_locked = tty->termios + 1;
+
+	o_tty->termios = kzalloc(sizeof(struct ktermios[2]), GFP_KERNEL);
+	if (o_tty->termios == NULL)
+		goto free_mem_out;
+	*o_tty->termios = driver->other->init_termios;
+	o_tty->termios_locked = o_tty->termios + 1;
+
+	tty_driver_kref_get(driver->other);
+	if (driver->subtype == PTY_TYPE_MASTER)
+		o_tty->count++;
+	/* Establish the links in both directions */
+	tty->link   = o_tty;
+	o_tty->link = tty;
+	/*
+	 * All structures have been allocated, so now we install them.
+	 * Failures after this point use release_tty to clean up, so
+	 * there's no need to null out the local pointers.
+	 */
+	tty_driver_kref_get(driver);
+	tty->count++;
+	pty_count++;
+	return 0;
+free_mem_out:
+	kfree(o_tty->termios);
+	module_put(o_tty->driver->owner);
+	free_tty_struct(o_tty);
+	kfree(tty->termios);
+	return -ENOMEM;
+}
+
+static void pty_unix98_remove(struct tty_driver *driver, struct tty_struct *tty)
+{
+	pty_count--;
+}
+
+static const struct tty_operations ptm_unix98_ops = {
+	.lookup = ptm_unix98_lookup,
+	.install = pty_unix98_install,
+	.remove = pty_unix98_remove,
+	.open = pty_open,
+	.close = pty_close,
+	.write = pty_write,
+	.write_room = pty_write_room,
+	.flush_buffer = pty_flush_buffer,
+	.chars_in_buffer = pty_chars_in_buffer,
+	.unthrottle = pty_unthrottle,
+	.set_termios = pty_set_termios,
+	.ioctl = pty_unix98_ioctl,
+	.shutdown = pty_unix98_shutdown
+};
+
 static const struct tty_operations pty_unix98_ops = {
+	.lookup = pts_unix98_lookup,
+	.install = pty_unix98_install,
+	.remove = pty_unix98_remove,
 	.open = pty_open,
 	.close = pty_close,
 	.write = pty_write,
@@ -397,9 +576,73 @@ static const struct tty_operations pty_unix98_ops = {
 	.chars_in_buffer = pty_chars_in_buffer,
 	.unthrottle = pty_unthrottle,
 	.set_termios = pty_set_termios,
-	.ioctl = pty_unix98_ioctl
+	.shutdown = pty_unix98_shutdown
 };
 
+/**
+ *	ptmx_open		-	open a unix 98 pty master
+ *	@inode: inode of device file
+ *	@filp: file pointer to tty
+ *
+ *	Allocate a unix98 pty master device from the ptmx driver.
+ *
+ *	Locking: tty_mutex protects the init_dev work. tty->count should
+ * 		protect the rest.
+ *		allocated_ptys_lock handles the list of free pty numbers
+ */
+
+static int __ptmx_open(struct inode *inode, struct file *filp)
+{
+	struct tty_struct *tty;
+	int retval;
+	int index;
+
+	nonseekable_open(inode, filp);
+
+	/* find a device that is not in use. */
+	index = devpts_new_index(inode);
+	if (index < 0)
+		return index;
+
+	mutex_lock(&tty_mutex);
+	tty = tty_init_dev(ptm_driver, index, 1);
+	mutex_unlock(&tty_mutex);
+
+	if (IS_ERR(tty)) {
+		retval = PTR_ERR(tty);
+		goto out;
+	}
+
+	set_bit(TTY_PTY_LOCK, &tty->flags); /* LOCK THE SLAVE */
+	filp->private_data = tty;
+	file_move(filp, &tty->tty_files);
+
+	retval = devpts_pty_new(inode, tty->link);
+	if (retval)
+		goto out1;
+
+	retval = ptm_driver->ops->open(tty, filp);
+	if (!retval)
+		return 0;
+out1:
+	tty_release_dev(filp);
+	return retval;
+out:
+	devpts_kill_index(inode, index);
+	return retval;
+}
+
+static int ptmx_open(struct inode *inode, struct file *filp)
+{
+	int ret;
+
+	lock_kernel();
+	ret = __ptmx_open(inode, filp);
+	unlock_kernel();
+	return ret;
+}
+
+static struct file_operations ptmx_fops;
 
 static void __init unix98_pty_init(void)
 {
@@ -427,7 +670,7 @@ static void __init unix98_pty_init(void)
 	ptm_driver->flags = TTY_DRIVER_RESET_TERMIOS | TTY_DRIVER_REAL_RAW |
 		TTY_DRIVER_DYNAMIC_DEV | TTY_DRIVER_DEVPTS_MEM;
 	ptm_driver->other = pts_driver;
-	tty_set_operations(ptm_driver, &pty_unix98_ops);
+	tty_set_operations(ptm_driver, &ptm_unix98_ops);
 
 	pts_driver->owner = THIS_MODULE;
 	pts_driver->driver_name = "pty_slave";
@@ -443,16 +686,26 @@ static void __init unix98_pty_init(void)
 	pts_driver->flags = TTY_DRIVER_RESET_TERMIOS | TTY_DRIVER_REAL_RAW |
 		TTY_DRIVER_DYNAMIC_DEV | TTY_DRIVER_DEVPTS_MEM;
 	pts_driver->other = ptm_driver;
-	tty_set_operations(pts_driver, &pty_ops);
-	
+	tty_set_operations(pts_driver, &pty_unix98_ops);
+
 	if (tty_register_driver(ptm_driver))
 		panic("Couldn't register Unix98 ptm driver");
 	if (tty_register_driver(pts_driver))
 		panic("Couldn't register Unix98 pts driver");
 
-	pty_table[1].data = &ptm_driver->refcount;
 	register_sysctl_table(pty_root_table);
+
+	/* Now create the /dev/ptmx special device */
+	tty_default_fops(&ptmx_fops);
+	ptmx_fops.open = ptmx_open;
+
+	cdev_init(&ptmx_cdev, &ptmx_fops);
+	if (cdev_add(&ptmx_cdev, MKDEV(TTYAUX_MAJOR, 2), 1) ||
+	    register_chrdev_region(MKDEV(TTYAUX_MAJOR, 2), 1, "/dev/ptmx") < 0)
+		panic("Couldn't register /dev/ptmx driver\n");
+	device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 2), NULL, "ptmx");
 }
+
 #else
 static inline void unix98_pty_init(void) { }
 #endif
diff --git a/drivers/char/rtc.c b/drivers/char/rtc.c
index f53d4d00faf0..b47710c17885 100644
--- a/drivers/char/rtc.c
+++ b/drivers/char/rtc.c
@@ -88,12 +88,12 @@
 #endif
 
 #ifdef CONFIG_SPARC32
-#include <linux/pci.h>
-#include <linux/jiffies.h>
-#include <asm/ebus.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <asm/io.h>
 
 static unsigned long rtc_port;
-static int rtc_irq = PCI_IRQ_NONE;
+static int rtc_irq;
 #endif
 
 #ifdef	CONFIG_HPET_RTC_IRQ
@@ -973,8 +973,8 @@ static int __init rtc_init(void)
 	char *guess = NULL;
 #endif
 #ifdef CONFIG_SPARC32
-	struct linux_ebus *ebus;
-	struct linux_ebus_device *edev;
+	struct device_node *ebus_dp;
+	struct of_device *op;
 #else
 	void *r;
 #ifdef RTC_IRQ
@@ -983,12 +983,16 @@ static int __init rtc_init(void)
 #endif
 
 #ifdef CONFIG_SPARC32
-	for_each_ebus(ebus) {
-		for_each_ebusdev(edev, ebus) {
-			if (strcmp(edev->prom_node->name, "rtc") == 0) {
-				rtc_port = edev->resource[0].start;
-				rtc_irq = edev->irqs[0];
-				goto found;
+	for_each_node_by_name(ebus_dp, "ebus") {
+		struct device_node *dp;
+		for (dp = ebus_dp; dp; dp = dp->sibling) {
+			if (!strcmp(dp->name, "rtc")) {
+				op = of_find_device_by_node(dp);
+				if (op) {
+					rtc_port = op->resource[0].start;
+					rtc_irq = op->irqs[0];
+					goto found;
+				}
 			}
 		}
 	}
@@ -997,7 +1001,7 @@ static int __init rtc_init(void)
 	return -EIO;
 
 found:
-	if (rtc_irq == PCI_IRQ_NONE) {
+	if (!rtc_irq) {
 		rtc_has_irq = 0;
 		goto no_irq;
 	}
diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c
index 19db1eb87c26..8b8f07a7f505 100644
--- a/drivers/char/stallion.c
+++ b/drivers/char/stallion.c
@@ -405,9 +405,9 @@ static unsigned int	stl_baudrates[] = {
 
 static int	stl_memioctl(struct inode *ip, struct file *fp, unsigned int cmd, unsigned long arg);
 static int	stl_brdinit(struct stlbrd *brdp);
-static int	stl_getportstats(struct stlport *portp, comstats_t __user *cp);
+static int	stl_getportstats(struct tty_struct *tty, struct stlport *portp, comstats_t __user *cp);
 static int	stl_clrportstats(struct stlport *portp, comstats_t __user *cp);
-static int	stl_waitcarrier(struct stlport *portp, struct file *filp);
+static int	stl_waitcarrier(struct tty_struct *tty, struct stlport *portp, struct file *filp);
 
 /*
  *	CD1400 uart specific handling functions.
@@ -612,8 +612,9 @@ static struct class *stallion_class;
 static void stl_cd_change(struct stlport *portp)
 {
 	unsigned int oldsigs = portp->sigs;
+	struct tty_struct *tty = tty_port_tty_get(&portp->port);
 
-	if (!portp->port.tty)
+	if (!tty)
 		return;
 
 	portp->sigs = stl_getsignals(portp);
@@ -623,7 +624,8 @@ static void stl_cd_change(struct stlport *portp)
 
 	if ((oldsigs & TIOCM_CD) && ((portp->sigs & TIOCM_CD) == 0))
 		if (portp->port.flags & ASYNC_CHECK_CD)
-			tty_hangup(portp->port.tty);
+			tty_hangup(tty);
+	tty_kref_put(tty);
 }
 
 /*
@@ -734,7 +736,7 @@ static int stl_open(struct tty_struct *tty, struct file *filp)
  *	On the first open of the device setup the port hardware, and
  *	initialize the per port data structure.
  */
-	portp->port.tty = tty;
+	tty_port_tty_set(&portp->port, tty);
 	tty->driver_data = portp;
 	portp->port.count++;
 
@@ -774,7 +776,7 @@ static int stl_open(struct tty_struct *tty, struct file *filp)
  *	then also we might have to wait for carrier.
  */
 	if (!(filp->f_flags & O_NONBLOCK))
-		if ((rc = stl_waitcarrier(portp, filp)) != 0)
+		if ((rc = stl_waitcarrier(tty, portp, filp)) != 0)
 			return rc;
 
 	portp->port.flags |= ASYNC_NORMAL_ACTIVE;
@@ -789,7 +791,8 @@ static int stl_open(struct tty_struct *tty, struct file *filp)
  *	maybe because if we are clocal then we don't need to wait...
  */
 
-static int stl_waitcarrier(struct stlport *portp, struct file *filp)
+static int stl_waitcarrier(struct tty_struct *tty, struct stlport *portp,
+							struct file *filp)
 {
 	unsigned long	flags;
 	int		rc, doclocal;
@@ -801,7 +804,7 @@ static int stl_waitcarrier(struct stlport *portp, struct file *filp)
 
 	spin_lock_irqsave(&stallion_lock, flags);
 
-	if (portp->port.tty->termios->c_cflag & CLOCAL)
+	if (tty->termios->c_cflag & CLOCAL)
 		doclocal++;
 
 	portp->openwaitcnt++;
@@ -846,8 +849,6 @@ static void stl_flushbuffer(struct tty_struct *tty)
 
 	pr_debug("stl_flushbuffer(tty=%p)\n", tty);
 
-	if (tty == NULL)
-		return;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
@@ -865,8 +866,6 @@ static void stl_waituntilsent(struct tty_struct *tty, int timeout)
 
 	pr_debug("stl_waituntilsent(tty=%p,timeout=%d)\n", tty, timeout);
 
-	if (tty == NULL)
-		return;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
@@ -949,7 +948,7 @@ static void stl_close(struct tty_struct *tty, struct file *filp)
 	tty_ldisc_flush(tty);
 
 	tty->closing = 0;
-	portp->port.tty = NULL;
+	tty_port_tty_set(&portp->port, NULL);
 
 	if (portp->openwaitcnt) {
 		if (portp->close_delay)
@@ -1033,8 +1032,6 @@ static int stl_putchar(struct tty_struct *tty, unsigned char ch)
 
 	pr_debug("stl_putchar(tty=%p,ch=%x)\n", tty, ch);
 
-	if (tty == NULL)
-		return -EINVAL;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return -EINVAL;
@@ -1070,8 +1067,6 @@ static void stl_flushchars(struct tty_struct *tty)
 
 	pr_debug("stl_flushchars(tty=%p)\n", tty);
 
-	if (tty == NULL)
-		return;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
@@ -1090,8 +1085,6 @@ static int stl_writeroom(struct tty_struct *tty)
 
 	pr_debug("stl_writeroom(tty=%p)\n", tty);
 
-	if (tty == NULL)
-		return 0;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return 0;
@@ -1122,8 +1115,6 @@ static int stl_charsinbuffer(struct tty_struct *tty)
 
 	pr_debug("stl_charsinbuffer(tty=%p)\n", tty);
 
-	if (tty == NULL)
-		return 0;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return 0;
@@ -1183,8 +1174,9 @@ static int stl_getserial(struct stlport *portp, struct serial_struct __user *sp)
  *	just quietly ignore any requests to change irq, etc.
  */
 
-static int stl_setserial(struct stlport *portp, struct serial_struct __user *sp)
+static int stl_setserial(struct tty_struct *tty, struct serial_struct __user *sp)
 {
+	struct stlport *	portp = tty->driver_data;
 	struct serial_struct	sio;
 
 	pr_debug("stl_setserial(portp=%p,sp=%p)\n", portp, sp);
@@ -1205,7 +1197,7 @@ static int stl_setserial(struct stlport *portp, struct serial_struct __user *sp)
 	portp->close_delay = sio.close_delay;
 	portp->closing_wait = sio.closing_wait;
 	portp->custom_divisor = sio.custom_divisor;
-	stl_setport(portp, portp->port.tty->termios);
+	stl_setport(portp, tty->termios);
 	return 0;
 }
 
@@ -1215,8 +1207,6 @@ static int stl_tiocmget(struct tty_struct *tty, struct file *file)
 {
 	struct stlport	*portp;
 
-	if (tty == NULL)
-		return -ENODEV;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return -ENODEV;
@@ -1232,8 +1222,6 @@ static int stl_tiocmset(struct tty_struct *tty, struct file *file,
 	struct stlport	*portp;
 	int rts = -1, dtr = -1;
 
-	if (tty == NULL)
-		return -ENODEV;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return -ENODEV;
@@ -1262,8 +1250,6 @@ static int stl_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd
 	pr_debug("stl_ioctl(tty=%p,file=%p,cmd=%x,arg=%lx)\n", tty, file, cmd,
 			arg);
 
-	if (tty == NULL)
-		return -ENODEV;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return -ENODEV;
@@ -1282,10 +1268,10 @@ static int stl_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd
 		rc = stl_getserial(portp, argp);
 		break;
 	case TIOCSSERIAL:
-		rc = stl_setserial(portp, argp);
+		rc = stl_setserial(tty, argp);
 		break;
 	case COM_GETPORTSTATS:
-		rc = stl_getportstats(portp, argp);
+		rc = stl_getportstats(tty, portp, argp);
 		break;
 	case COM_CLRPORTSTATS:
 		rc = stl_clrportstats(portp, argp);
@@ -1317,8 +1303,6 @@ static void stl_start(struct tty_struct *tty)
 
 	pr_debug("stl_start(tty=%p)\n", tty);
 
-	if (tty == NULL)
-		return;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
@@ -1334,8 +1318,6 @@ static void stl_settermios(struct tty_struct *tty, struct ktermios *old)
 
 	pr_debug("stl_settermios(tty=%p,old=%p)\n", tty, old);
 
-	if (tty == NULL)
-		return;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
@@ -1369,8 +1351,6 @@ static void stl_throttle(struct tty_struct *tty)
 
 	pr_debug("stl_throttle(tty=%p)\n", tty);
 
-	if (tty == NULL)
-		return;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
@@ -1389,8 +1369,6 @@ static void stl_unthrottle(struct tty_struct *tty)
 
 	pr_debug("stl_unthrottle(tty=%p)\n", tty);
 
-	if (tty == NULL)
-		return;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
@@ -1410,8 +1388,6 @@ static void stl_stop(struct tty_struct *tty)
 
 	pr_debug("stl_stop(tty=%p)\n", tty);
 
-	if (tty == NULL)
-		return;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
@@ -1432,8 +1408,6 @@ static void stl_hangup(struct tty_struct *tty)
 
 	pr_debug("stl_hangup(tty=%p)\n", tty);
 
-	if (tty == NULL)
-		return;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
@@ -1452,7 +1426,7 @@ static void stl_hangup(struct tty_struct *tty)
 		portp->tx.head = NULL;
 		portp->tx.tail = NULL;
 	}
-	portp->port.tty = NULL;
+	tty_port_tty_set(&portp->port, NULL);
 	portp->port.flags &= ~ASYNC_NORMAL_ACTIVE;
 	portp->port.count = 0;
 	wake_up_interruptible(&portp->port.open_wait);
@@ -1466,8 +1440,6 @@ static int stl_breakctl(struct tty_struct *tty, int state)
 
 	pr_debug("stl_breakctl(tty=%p,state=%d)\n", tty, state);
 
-	if (tty == NULL)
-		return -EINVAL;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return -EINVAL;
@@ -1484,8 +1456,6 @@ static void stl_sendxchar(struct tty_struct *tty, char ch)
 
 	pr_debug("stl_sendxchar(tty=%p,ch=%x)\n", tty, ch);
 
-	if (tty == NULL)
-		return;
 	portp = tty->driver_data;
 	if (portp == NULL)
 		return;
@@ -1805,7 +1775,7 @@ static int __devinit stl_initports(struct stlbrd *brdp, struct stlpanel *panelp)
 				"(size=%Zd)\n", sizeof(struct stlport));
 			break;
 		}
-
+		tty_port_init(&portp->port);
 		portp->magic = STL_PORTMAGIC;
 		portp->portnr = i;
 		portp->brdnr = panelp->brdnr;
@@ -1832,6 +1802,7 @@ static void stl_cleanup_panels(struct stlbrd *brdp)
 	struct stlpanel *panelp;
 	struct stlport *portp;
 	unsigned int j, k;
+	struct tty_struct *tty;
 
 	for (j = 0; j < STL_MAXPANELS; j++) {
 		panelp = brdp->panels[j];
@@ -1841,8 +1812,11 @@ static void stl_cleanup_panels(struct stlbrd *brdp)
 			portp = panelp->ports[k];
 			if (portp == NULL)
 				continue;
-			if (portp->port.tty != NULL)
-				stl_hangup(portp->port.tty);
+			tty = tty_port_tty_get(&portp->port);
+			if (tty != NULL) {
+				stl_hangup(tty);
+				tty_kref_put(tty);
+			}
 			kfree(portp->tx.buf);
 			kfree(portp);
 		}
@@ -2498,7 +2472,7 @@ static struct stlport *stl_getport(int brdnr, int panelnr, int portnr)
  *	what port to get stats for (used through board control device).
  */
 
-static int stl_getportstats(struct stlport *portp, comstats_t __user *cp)
+static int stl_getportstats(struct tty_struct *tty, struct stlport *portp, comstats_t __user *cp)
 {
 	comstats_t	stl_comstats;
 	unsigned char	*head, *tail;
@@ -2525,18 +2499,17 @@ static int stl_getportstats(struct stlport *portp, comstats_t __user *cp)
 	portp->stats.rxbuffered = 0;
 
 	spin_lock_irqsave(&stallion_lock, flags);
-	if (portp->port.tty != NULL)
-		if (portp->port.tty->driver_data == portp) {
-			portp->stats.ttystate = portp->port.tty->flags;
-			/* No longer available as a statistic */
-			portp->stats.rxbuffered = 1; /*portp->port.tty->flip.count; */
-			if (portp->port.tty->termios != NULL) {
-				portp->stats.cflags = portp->port.tty->termios->c_cflag;
-				portp->stats.iflags = portp->port.tty->termios->c_iflag;
-				portp->stats.oflags = portp->port.tty->termios->c_oflag;
-				portp->stats.lflags = portp->port.tty->termios->c_lflag;
-			}
+	if (tty != NULL && portp->port.tty == tty) {
+		portp->stats.ttystate = tty->flags;
+		/* No longer available as a statistic */
+		portp->stats.rxbuffered = 1; /*tty->flip.count; */
+		if (tty->termios != NULL) {
+			portp->stats.cflags = tty->termios->c_cflag;
+			portp->stats.iflags = tty->termios->c_iflag;
+			portp->stats.oflags = tty->termios->c_oflag;
+			portp->stats.lflags = tty->termios->c_lflag;
 		}
+	}
 	spin_unlock_irqrestore(&stallion_lock, flags);
 
 	head = portp->tx.head;
@@ -2640,7 +2613,7 @@ static int stl_memioctl(struct inode *ip, struct file *fp, unsigned int cmd, uns
 
 	switch (cmd) {
 	case COM_GETPORTSTATS:
-		rc = stl_getportstats(NULL, argp);
+		rc = stl_getportstats(NULL, NULL, argp);
 		break;
 	case COM_CLRPORTSTATS:
 		rc = stl_clrportstats(NULL, argp);
@@ -3243,7 +3216,7 @@ static void stl_cd1400flowctrl(struct stlport *portp, int state)
 
 	if (portp == NULL)
 		return;
-	tty = portp->port.tty;
+	tty = tty_port_tty_get(&portp->port);
 	if (tty == NULL)
 		return;
 
@@ -3288,6 +3261,7 @@ static void stl_cd1400flowctrl(struct stlport *portp, int state)
 
 	BRDDISABLE(portp->brdnr);
 	spin_unlock_irqrestore(&brd_lock, flags);
+	tty_kref_put(tty);
 }
 
 /*****************************************************************************/
@@ -3305,7 +3279,7 @@ static void stl_cd1400sendflow(struct stlport *portp, int state)
 
 	if (portp == NULL)
 		return;
-	tty = portp->port.tty;
+	tty = tty_port_tty_get(&portp->port);
 	if (tty == NULL)
 		return;
 
@@ -3325,6 +3299,7 @@ static void stl_cd1400sendflow(struct stlport *portp, int state)
 	}
 	BRDDISABLE(portp->brdnr);
 	spin_unlock_irqrestore(&brd_lock, flags);
+	tty_kref_put(tty);
 }
 
 /*****************************************************************************/
@@ -3478,6 +3453,7 @@ static void stl_cd1400txisr(struct stlpanel *panelp, int ioaddr)
 	int		len, stlen;
 	char		*head, *tail;
 	unsigned char	ioack, srer;
+	struct tty_struct *tty;
 
 	pr_debug("stl_cd1400txisr(panelp=%p,ioaddr=%x)\n", panelp, ioaddr);
 
@@ -3504,8 +3480,11 @@ static void stl_cd1400txisr(struct stlpanel *panelp, int ioaddr)
 	if ((len == 0) || ((len < STL_TXBUFLOW) &&
 	    (test_bit(ASYI_TXLOW, &portp->istate) == 0))) {
 		set_bit(ASYI_TXLOW, &portp->istate);
-		if (portp->port.tty)
-			tty_wakeup(portp->port.tty);
+		tty = tty_port_tty_get(&portp->port);
+		if (tty) {
+			tty_wakeup(tty);
+			tty_kref_put(tty);
+		}
 	}
 
 	if (len == 0) {
@@ -3569,7 +3548,7 @@ static void stl_cd1400rxisr(struct stlpanel *panelp, int ioaddr)
 		return;
 	}
 	portp = panelp->ports[(ioack >> 3)];
-	tty = portp->port.tty;
+	tty = tty_port_tty_get(&portp->port);
 
 	if ((ioack & ACK_TYPMASK) == ACK_TYPRXGOOD) {
 		outb((RDCR + portp->uartaddr), ioaddr);
@@ -3633,10 +3612,12 @@ static void stl_cd1400rxisr(struct stlpanel *panelp, int ioaddr)
 		}
 	} else {
 		printk("STALLION: bad RX interrupt ack value=%x\n", ioack);
+		tty_kref_put(tty);
 		return;
 	}
 
 stl_rxalldone:
+	tty_kref_put(tty);
 	outb((EOSRR + portp->uartaddr), ioaddr);
 	outb(0, (ioaddr + EREG_DATA));
 }
@@ -4175,7 +4156,7 @@ static void stl_sc26198flowctrl(struct stlport *portp, int state)
 
 	if (portp == NULL)
 		return;
-	tty = portp->port.tty;
+	tty = tty_port_tty_get(&portp->port);
 	if (tty == NULL)
 		return;
 
@@ -4226,6 +4207,7 @@ static void stl_sc26198flowctrl(struct stlport *portp, int state)
 
 	BRDDISABLE(portp->brdnr);
 	spin_unlock_irqrestore(&brd_lock, flags);
+	tty_kref_put(tty);
 }
 
 /*****************************************************************************/
@@ -4244,7 +4226,7 @@ static void stl_sc26198sendflow(struct stlport *portp, int state)
 
 	if (portp == NULL)
 		return;
-	tty = portp->port.tty;
+	tty = tty_port_tty_get(&portp->port);
 	if (tty == NULL)
 		return;
 
@@ -4269,6 +4251,7 @@ static void stl_sc26198sendflow(struct stlport *portp, int state)
 	}
 	BRDDISABLE(portp->brdnr);
 	spin_unlock_irqrestore(&brd_lock, flags);
+	tty_kref_put(tty);
 }
 
 /*****************************************************************************/
@@ -4408,6 +4391,7 @@ static void stl_sc26198intr(struct stlpanel *panelp, unsigned int iobase)
 
 static void stl_sc26198txisr(struct stlport *portp)
 {
+	struct tty_struct *tty;
 	unsigned int	ioaddr;
 	unsigned char	mr0;
 	int		len, stlen;
@@ -4422,8 +4406,11 @@ static void stl_sc26198txisr(struct stlport *portp)
 	if ((len == 0) || ((len < STL_TXBUFLOW) &&
 	    (test_bit(ASYI_TXLOW, &portp->istate) == 0))) {
 		set_bit(ASYI_TXLOW, &portp->istate);
-		if (portp->port.tty)
-			tty_wakeup(portp->port.tty);
+		tty = tty_port_tty_get(&portp->port);
+		if (tty) {
+			tty_wakeup(tty);
+			tty_kref_put(tty);
+		}
 	}
 
 	if (len == 0) {
@@ -4476,7 +4463,7 @@ static void stl_sc26198rxisr(struct stlport *portp, unsigned int iack)
 
 	pr_debug("stl_sc26198rxisr(portp=%p,iack=%x)\n", portp, iack);
 
-	tty = portp->port.tty;
+	tty = tty_port_tty_get(&portp->port);
 	ioaddr = portp->ioaddr;
 	outb(GIBCR, (ioaddr + XP_ADDR));
 	len = inb(ioaddr + XP_DATA) + 1;
@@ -4515,6 +4502,7 @@ static void stl_sc26198rxisr(struct stlport *portp, unsigned int iack)
 			stl_sc26198txunflow(portp, tty);
 		}
 	}
+	tty_kref_put(tty);
 }
 
 /*****************************************************************************/
@@ -4528,7 +4516,7 @@ static void stl_sc26198rxbadch(struct stlport *portp, unsigned char status, char
 	struct tty_struct	*tty;
 	unsigned int		ioaddr;
 
-	tty = portp->port.tty;
+	tty = tty_port_tty_get(&portp->port);
 	ioaddr = portp->ioaddr;
 
 	if (status & SR_RXPARITY)
@@ -4566,6 +4554,7 @@ static void stl_sc26198rxbadch(struct stlport *portp, unsigned char status, char
 		if (status == 0)
 			portp->stats.rxtotal++;
 	}
+	tty_kref_put(tty);
 }
 
 /*****************************************************************************/
diff --git a/drivers/char/sx.c b/drivers/char/sx.c
index c385206f9db5..5b8d7a1aa3e6 100644
--- a/drivers/char/sx.c
+++ b/drivers/char/sx.c
@@ -2504,7 +2504,7 @@ static void __devexit sx_remove_card(struct sx_board *board,
 		del_timer(&board->timer);
 		if (pdev) {
 #ifdef CONFIG_PCI
-			pci_iounmap(pdev, board->base);
+			pci_iounmap(pdev, board->base2);
 			pci_release_region(pdev, IS_CF_BOARD(board) ? 3 : 2);
 #endif
 		} else {
@@ -2703,7 +2703,7 @@ static int __devinit sx_pci_probe(struct pci_dev *pdev,
 
 	return 0;
 err_unmap:
-	pci_iounmap(pdev, board->base);
+	pci_iounmap(pdev, board->base2);
 err_reg:
 	pci_release_region(pdev, reg);
 err_flag:
diff --git a/drivers/char/tpm/tpm.c b/drivers/char/tpm/tpm.c
index ae766d868454..1fee7034a386 100644
--- a/drivers/char/tpm/tpm.c
+++ b/drivers/char/tpm/tpm.c
@@ -954,72 +954,63 @@ EXPORT_SYMBOL_GPL(tpm_store_cancel);
 
 /*
  * Device file system interface to the TPM
+ *
+ * It's assured that the chip will be opened just once,
+ * by the check of is_open variable, which is protected
+ * by driver_lock.
  */
 int tpm_open(struct inode *inode, struct file *file)
 {
-	int rc = 0, minor = iminor(inode);
+	int minor = iminor(inode);
 	struct tpm_chip *chip = NULL, *pos;
 
-	lock_kernel();
-	spin_lock(&driver_lock);
-
-	list_for_each_entry(pos, &tpm_chip_list, list) {
+	rcu_read_lock();
+	list_for_each_entry_rcu(pos, &tpm_chip_list, list) {
 		if (pos->vendor.miscdev.minor == minor) {
 			chip = pos;
+			get_device(chip->dev);
 			break;
 		}
 	}
+	rcu_read_unlock();
 
-	if (chip == NULL) {
-		rc = -ENODEV;
-		goto err_out;
-	}
+	if (!chip)
+		return -ENODEV;
 
-	if (chip->num_opens) {
+	if (test_and_set_bit(0, &chip->is_open)) {
 		dev_dbg(chip->dev, "Another process owns this TPM\n");
-		rc = -EBUSY;
-		goto err_out;
+		put_device(chip->dev);
+		return -EBUSY;
 	}
 
-	chip->num_opens++;
-	get_device(chip->dev);
-
-	spin_unlock(&driver_lock);
-
 	chip->data_buffer = kmalloc(TPM_BUFSIZE * sizeof(u8), GFP_KERNEL);
 	if (chip->data_buffer == NULL) {
-		chip->num_opens--;
+		clear_bit(0, &chip->is_open);
 		put_device(chip->dev);
-		unlock_kernel();
 		return -ENOMEM;
 	}
 
 	atomic_set(&chip->data_pending, 0);
 
 	file->private_data = chip;
-	unlock_kernel();
 	return 0;
-
-err_out:
-	spin_unlock(&driver_lock);
-	unlock_kernel();
-	return rc;
 }
 EXPORT_SYMBOL_GPL(tpm_open);
 
+/*
+ * Called on file close
+ */
 int tpm_release(struct inode *inode, struct file *file)
 {
 	struct tpm_chip *chip = file->private_data;
 
+	del_singleshot_timer_sync(&chip->user_read_timer);
 	flush_scheduled_work();
-	spin_lock(&driver_lock);
 	file->private_data = NULL;
-	del_singleshot_timer_sync(&chip->user_read_timer);
 	atomic_set(&chip->data_pending, 0);
-	chip->num_opens--;
-	put_device(chip->dev);
 	kfree(chip->data_buffer);
-	spin_unlock(&driver_lock);
+	clear_bit(0, &chip->is_open);
+	put_device(chip->dev);
 	return 0;
 }
 EXPORT_SYMBOL_GPL(tpm_release);
@@ -1093,13 +1084,11 @@ void tpm_remove_hardware(struct device *dev)
 	}
 
 	spin_lock(&driver_lock);
-
-	list_del(&chip->list);
-
+	list_del_rcu(&chip->list);
 	spin_unlock(&driver_lock);
+	synchronize_rcu();
 
 	misc_deregister(&chip->vendor.miscdev);
-
 	sysfs_remove_group(&dev->kobj, chip->vendor.attr_group);
 	tpm_bios_log_teardown(chip->bios_dir);
 
@@ -1144,25 +1133,33 @@ int tpm_pm_resume(struct device *dev)
 }
 EXPORT_SYMBOL_GPL(tpm_pm_resume);
 
+/* In case vendor provided release function, call it too.*/
+
+void tpm_dev_vendor_release(struct tpm_chip *chip)
+{
+	if (chip->vendor.release)
+		chip->vendor.release(chip->dev);
+
+	clear_bit(chip->dev_num, dev_mask);
+	kfree(chip->vendor.miscdev.name);
+}
+EXPORT_SYMBOL_GPL(tpm_dev_vendor_release);
+
+
 /*
  * Once all references to platform device are down to 0,
  * release all allocated structures.
- * In case vendor provided release function,
- * call it too.
  */
 static void tpm_dev_release(struct device *dev)
 {
 	struct tpm_chip *chip = dev_get_drvdata(dev);
 
-	if (chip->vendor.release)
-		chip->vendor.release(dev);
+	tpm_dev_vendor_release(chip);
 
 	chip->release(dev);
-
-	clear_bit(chip->dev_num, dev_mask);
-	kfree(chip->vendor.miscdev.name);
 	kfree(chip);
 }
+EXPORT_SYMBOL_GPL(tpm_dev_release);
 
 /*
  * Called from tpm_<specific>.c probe function only for devices 
@@ -1171,8 +1168,8 @@ static void tpm_dev_release(struct device *dev)
  * upon errant exit from this function specific probe function should call
  * pci_disable_device
  */
-struct tpm_chip *tpm_register_hardware(struct device *dev, const struct tpm_vendor_specific
-				       *entry)
+struct tpm_chip *tpm_register_hardware(struct device *dev,
+					const struct tpm_vendor_specific *entry)
 {
 #define DEVNAME_SIZE 7
 
@@ -1231,21 +1228,20 @@ struct tpm_chip *tpm_register_hardware(struct device *dev, const struct tpm_vend
 		return NULL;
 	}
 
-	spin_lock(&driver_lock);
-
-	list_add(&chip->list, &tpm_chip_list);
-
-	spin_unlock(&driver_lock);
-
 	if (sysfs_create_group(&dev->kobj, chip->vendor.attr_group)) {
-		list_del(&chip->list);
 		misc_deregister(&chip->vendor.miscdev);
 		put_device(chip->dev);
+
 		return NULL;
 	}
 
 	chip->bios_dir = tpm_bios_log_setup(devname);
 
+	/* Make chip available */
+	spin_lock(&driver_lock);
+	list_add_rcu(&chip->list, &tpm_chip_list);
+	spin_unlock(&driver_lock);
+
 	return chip;
 }
 EXPORT_SYMBOL_GPL(tpm_register_hardware);
diff --git a/drivers/char/tpm/tpm.h b/drivers/char/tpm/tpm.h
index e885148b4cfb..8e30df4a4388 100644
--- a/drivers/char/tpm/tpm.h
+++ b/drivers/char/tpm/tpm.h
@@ -90,7 +90,7 @@ struct tpm_chip {
 	struct device *dev;	/* Device stuff */
 
 	int dev_num;		/* /dev/tpm# */
-	int num_opens;		/* only one allowed */
+	unsigned long is_open;	/* only one allowed */
 	int time_expired;
 
 	/* Data passed to and from the tpm via the read/write calls */
@@ -132,6 +132,7 @@ extern struct tpm_chip* tpm_register_hardware(struct device *,
 				 const struct tpm_vendor_specific *);
 extern int tpm_open(struct inode *, struct file *);
 extern int tpm_release(struct inode *, struct file *);
+extern void tpm_dev_vendor_release(struct tpm_chip *);
 extern ssize_t tpm_write(struct file *, const char __user *, size_t,
 			 loff_t *);
 extern ssize_t tpm_read(struct file *, char __user *, size_t, loff_t *);
diff --git a/drivers/char/tpm/tpm_tis.c b/drivers/char/tpm/tpm_tis.c
index ed1879c0dd8d..717af7ad1bdf 100644
--- a/drivers/char/tpm/tpm_tis.c
+++ b/drivers/char/tpm/tpm_tis.c
@@ -630,12 +630,23 @@ static struct pnp_device_id tpm_pnp_tbl[] __devinitdata = {
 	{"", 0}			/* Terminator */
 };
 
+static __devexit void tpm_tis_pnp_remove(struct pnp_dev *dev)
+{
+	struct tpm_chip *chip = pnp_get_drvdata(dev);
+
+	tpm_dev_vendor_release(chip);
+
+	kfree(chip);
+}
+
+
 static struct pnp_driver tis_pnp_driver = {
 	.name = "tpm_tis",
 	.id_table = tpm_pnp_tbl,
 	.probe = tpm_tis_pnp_init,
 	.suspend = tpm_tis_pnp_suspend,
 	.resume = tpm_tis_pnp_resume,
+	.remove = tpm_tis_pnp_remove,
 };
 
 #define TIS_HID_USR_IDX sizeof(tpm_pnp_tbl)/sizeof(struct pnp_device_id) -2
@@ -683,6 +694,7 @@ static void __exit cleanup_tis(void)
 	spin_lock(&tis_lock);
 	list_for_each_entry_safe(i, j, &tis_chips, list) {
 		chip = to_tpm_chip(i);
+		tpm_remove_hardware(chip->dev);
 		iowrite32(~TPM_GLOBAL_INT_ENABLE &
 			  ioread32(chip->vendor.iobase +
 				   TPM_INT_ENABLE(chip->vendor.
@@ -694,9 +706,9 @@ static void __exit cleanup_tis(void)
 			free_irq(chip->vendor.irq, chip);
 		iounmap(i->iobase);
 		list_del(&i->list);
-		tpm_remove_hardware(chip->dev);
 	}
 	spin_unlock(&tis_lock);
+
 	if (force) {
 		platform_device_unregister(pdev);
 		driver_unregister(&tis_drv);
diff --git a/drivers/char/tty_audit.c b/drivers/char/tty_audit.c
index 3582f43345a8..5787249934c8 100644
--- a/drivers/char/tty_audit.c
+++ b/drivers/char/tty_audit.c
@@ -93,7 +93,7 @@ static void tty_audit_buf_push(struct task_struct *tsk, uid_t loginuid,
 		get_task_comm(name, tsk);
 		audit_log_untrustedstring(ab, name);
 		audit_log_format(ab, " data=");
-		audit_log_n_untrustedstring(ab, buf->data, buf->valid);
+		audit_log_n_hex(ab, buf->data, buf->valid);
 		audit_log_end(ab);
 	}
 	buf->valid = 0;
diff --git a/drivers/char/tty_buffer.c b/drivers/char/tty_buffer.c
new file mode 100644
index 000000000000..810ee25d66a4
--- /dev/null
+++ b/drivers/char/tty_buffer.c
@@ -0,0 +1,511 @@
+/*
+ * Tty buffer allocation management
+ */
+
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/tty.h>
+#include <linux/tty_driver.h>
+#include <linux/tty_flip.h>
+#include <linux/timer.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/sched.h>
+#include <linux/init.h>
+#include <linux/wait.h>
+#include <linux/bitops.h>
+#include <linux/delay.h>
+#include <linux/module.h>
+
+/**
+ *	tty_buffer_free_all		-	free buffers used by a tty
+ *	@tty: tty to free from
+ *
+ *	Remove all the buffers pending on a tty whether queued with data
+ *	or in the free ring. Must be called when the tty is no longer in use
+ *
+ *	Locking: none
+ */
+
+void tty_buffer_free_all(struct tty_struct *tty)
+{
+	struct tty_buffer *thead;
+	while ((thead = tty->buf.head) != NULL) {
+		tty->buf.head = thead->next;
+		kfree(thead);
+	}
+	while ((thead = tty->buf.free) != NULL) {
+		tty->buf.free = thead->next;
+		kfree(thead);
+	}
+	tty->buf.tail = NULL;
+	tty->buf.memory_used = 0;
+}
+
+/**
+ *	tty_buffer_alloc	-	allocate a tty buffer
+ *	@tty: tty device
+ *	@size: desired size (characters)
+ *
+ *	Allocate a new tty buffer to hold the desired number of characters.
+ *	Return NULL if out of memory or the allocation would exceed the
+ *	per device queue
+ *
+ *	Locking: Caller must hold tty->buf.lock
+ */
+
+static struct tty_buffer *tty_buffer_alloc(struct tty_struct *tty, size_t size)
+{
+	struct tty_buffer *p;
+
+	if (tty->buf.memory_used + size > 65536)
+		return NULL;
+	p = kmalloc(sizeof(struct tty_buffer) + 2 * size, GFP_ATOMIC);
+	if (p == NULL)
+		return NULL;
+	p->used = 0;
+	p->size = size;
+	p->next = NULL;
+	p->commit = 0;
+	p->read = 0;
+	p->char_buf_ptr = (char *)(p->data);
+	p->flag_buf_ptr = (unsigned char *)p->char_buf_ptr + size;
+	tty->buf.memory_used += size;
+	return p;
+}
+
+/**
+ *	tty_buffer_free		-	free a tty buffer
+ *	@tty: tty owning the buffer
+ *	@b: the buffer to free
+ *
+ *	Free a tty buffer, or add it to the free list according to our
+ *	internal strategy
+ *
+ *	Locking: Caller must hold tty->buf.lock
+ */
+
+static void tty_buffer_free(struct tty_struct *tty, struct tty_buffer *b)
+{
+	/* Dumb strategy for now - should keep some stats */
+	tty->buf.memory_used -= b->size;
+	WARN_ON(tty->buf.memory_used < 0);
+
+	if (b->size >= 512)
+		kfree(b);
+	else {
+		b->next = tty->buf.free;
+		tty->buf.free = b;
+	}
+}
+
+/**
+ *	__tty_buffer_flush		-	flush full tty buffers
+ *	@tty: tty to flush
+ *
+ *	flush all the buffers containing receive data. Caller must
+ *	hold the buffer lock and must have ensured no parallel flush to
+ *	ldisc is running.
+ *
+ *	Locking: Caller must hold tty->buf.lock
+ */
+
+static void __tty_buffer_flush(struct tty_struct *tty)
+{
+	struct tty_buffer *thead;
+
+	while ((thead = tty->buf.head) != NULL) {
+		tty->buf.head = thead->next;
+		tty_buffer_free(tty, thead);
+	}
+	tty->buf.tail = NULL;
+}
+
+/**
+ *	tty_buffer_flush		-	flush full tty buffers
+ *	@tty: tty to flush
+ *
+ *	flush all the buffers containing receive data. If the buffer is
+ *	being processed by flush_to_ldisc then we defer the processing
+ *	to that function
+ *
+ *	Locking: none
+ */
+
+void tty_buffer_flush(struct tty_struct *tty)
+{
+	unsigned long flags;
+	spin_lock_irqsave(&tty->buf.lock, flags);
+
+	/* If the data is being pushed to the tty layer then we can't
+	   process it here. Instead set a flag and the flush_to_ldisc
+	   path will process the flush request before it exits */
+	if (test_bit(TTY_FLUSHING, &tty->flags)) {
+		set_bit(TTY_FLUSHPENDING, &tty->flags);
+		spin_unlock_irqrestore(&tty->buf.lock, flags);
+		wait_event(tty->read_wait,
+				test_bit(TTY_FLUSHPENDING, &tty->flags) == 0);
+		return;
+	} else
+		__tty_buffer_flush(tty);
+	spin_unlock_irqrestore(&tty->buf.lock, flags);
+}
+
+/**
+ *	tty_buffer_find		-	find a free tty buffer
+ *	@tty: tty owning the buffer
+ *	@size: characters wanted
+ *
+ *	Locate an existing suitable tty buffer or if we are lacking one then
+ *	allocate a new one. We round our buffers off in 256 character chunks
+ *	to get better allocation behaviour.
+ *
+ *	Locking: Caller must hold tty->buf.lock
+ */
+
+static struct tty_buffer *tty_buffer_find(struct tty_struct *tty, size_t size)
+{
+	struct tty_buffer **tbh = &tty->buf.free;
+	while ((*tbh) != NULL) {
+		struct tty_buffer *t = *tbh;
+		if (t->size >= size) {
+			*tbh = t->next;
+			t->next = NULL;
+			t->used = 0;
+			t->commit = 0;
+			t->read = 0;
+			tty->buf.memory_used += t->size;
+			return t;
+		}
+		tbh = &((*tbh)->next);
+	}
+	/* Round the buffer size out */
+	size = (size + 0xFF) & ~0xFF;
+	return tty_buffer_alloc(tty, size);
+	/* Should possibly check if this fails for the largest buffer we
+	   have queued and recycle that ? */
+}
+
+/**
+ *	tty_buffer_request_room		-	grow tty buffer if needed
+ *	@tty: tty structure
+ *	@size: size desired
+ *
+ *	Make at least size bytes of linear space available for the tty
+ *	buffer. If we fail return the size we managed to find.
+ *
+ *	Locking: Takes tty->buf.lock
+ */
+int tty_buffer_request_room(struct tty_struct *tty, size_t size)
+{
+	struct tty_buffer *b, *n;
+	int left;
+	unsigned long flags;
+
+	spin_lock_irqsave(&tty->buf.lock, flags);
+
+	/* OPTIMISATION: We could keep a per tty "zero" sized buffer to
+	   remove this conditional if its worth it. This would be invisible
+	   to the callers */
+	if ((b = tty->buf.tail) != NULL)
+		left = b->size - b->used;
+	else
+		left = 0;
+
+	if (left < size) {
+		/* This is the slow path - looking for new buffers to use */
+		if ((n = tty_buffer_find(tty, size)) != NULL) {
+			if (b != NULL) {
+				b->next = n;
+				b->commit = b->used;
+			} else
+				tty->buf.head = n;
+			tty->buf.tail = n;
+		} else
+			size = left;
+	}
+
+	spin_unlock_irqrestore(&tty->buf.lock, flags);
+	return size;
+}
+EXPORT_SYMBOL_GPL(tty_buffer_request_room);
+
+/**
+ *	tty_insert_flip_string	-	Add characters to the tty buffer
+ *	@tty: tty structure
+ *	@chars: characters
+ *	@size: size
+ *
+ *	Queue a series of bytes to the tty buffering. All the characters
+ *	passed are marked as without error. Returns the number added.
+ *
+ *	Locking: Called functions may take tty->buf.lock
+ */
+
+int tty_insert_flip_string(struct tty_struct *tty, const unsigned char *chars,
+				size_t size)
+{
+	int copied = 0;
+	do {
+		int space = tty_buffer_request_room(tty, size - copied);
+		struct tty_buffer *tb = tty->buf.tail;
+		/* If there is no space then tb may be NULL */
+		if (unlikely(space == 0))
+			break;
+		memcpy(tb->char_buf_ptr + tb->used, chars, space);
+		memset(tb->flag_buf_ptr + tb->used, TTY_NORMAL, space);
+		tb->used += space;
+		copied += space;
+		chars += space;
+		/* There is a small chance that we need to split the data over
+		   several buffers. If this is the case we must loop */
+	} while (unlikely(size > copied));
+	return copied;
+}
+EXPORT_SYMBOL(tty_insert_flip_string);
+
+/**
+ *	tty_insert_flip_string_flags	-	Add characters to the tty buffer
+ *	@tty: tty structure
+ *	@chars: characters
+ *	@flags: flag bytes
+ *	@size: size
+ *
+ *	Queue a series of bytes to the tty buffering. For each character
+ *	the flags array indicates the status of the character. Returns the
+ *	number added.
+ *
+ *	Locking: Called functions may take tty->buf.lock
+ */
+
+int tty_insert_flip_string_flags(struct tty_struct *tty,
+		const unsigned char *chars, const char *flags, size_t size)
+{
+	int copied = 0;
+	do {
+		int space = tty_buffer_request_room(tty, size - copied);
+		struct tty_buffer *tb = tty->buf.tail;
+		/* If there is no space then tb may be NULL */
+		if (unlikely(space == 0))
+			break;
+		memcpy(tb->char_buf_ptr + tb->used, chars, space);
+		memcpy(tb->flag_buf_ptr + tb->used, flags, space);
+		tb->used += space;
+		copied += space;
+		chars += space;
+		flags += space;
+		/* There is a small chance that we need to split the data over
+		   several buffers. If this is the case we must loop */
+	} while (unlikely(size > copied));
+	return copied;
+}
+EXPORT_SYMBOL(tty_insert_flip_string_flags);
+
+/**
+ *	tty_schedule_flip	-	push characters to ldisc
+ *	@tty: tty to push from
+ *
+ *	Takes any pending buffers and transfers their ownership to the
+ *	ldisc side of the queue. It then schedules those characters for
+ *	processing by the line discipline.
+ *
+ *	Locking: Takes tty->buf.lock
+ */
+
+void tty_schedule_flip(struct tty_struct *tty)
+{
+	unsigned long flags;
+	spin_lock_irqsave(&tty->buf.lock, flags);
+	if (tty->buf.tail != NULL)
+		tty->buf.tail->commit = tty->buf.tail->used;
+	spin_unlock_irqrestore(&tty->buf.lock, flags);
+	schedule_delayed_work(&tty->buf.work, 1);
+}
+EXPORT_SYMBOL(tty_schedule_flip);
+
+/**
+ *	tty_prepare_flip_string		-	make room for characters
+ *	@tty: tty
+ *	@chars: return pointer for character write area
+ *	@size: desired size
+ *
+ *	Prepare a block of space in the buffer for data. Returns the length
+ *	available and buffer pointer to the space which is now allocated and
+ *	accounted for as ready for normal characters. This is used for drivers
+ *	that need their own block copy routines into the buffer. There is no
+ *	guarantee the buffer is a DMA target!
+ *
+ *	Locking: May call functions taking tty->buf.lock
+ */
+
+int tty_prepare_flip_string(struct tty_struct *tty, unsigned char **chars,
+								size_t size)
+{
+	int space = tty_buffer_request_room(tty, size);
+	if (likely(space)) {
+		struct tty_buffer *tb = tty->buf.tail;
+		*chars = tb->char_buf_ptr + tb->used;
+		memset(tb->flag_buf_ptr + tb->used, TTY_NORMAL, space);
+		tb->used += space;
+	}
+	return space;
+}
+EXPORT_SYMBOL_GPL(tty_prepare_flip_string);
+
+/**
+ *	tty_prepare_flip_string_flags	-	make room for characters
+ *	@tty: tty
+ *	@chars: return pointer for character write area
+ *	@flags: return pointer for status flag write area
+ *	@size: desired size
+ *
+ *	Prepare a block of space in the buffer for data. Returns the length
+ *	available and buffer pointer to the space which is now allocated and
+ *	accounted for as ready for characters. This is used for drivers
+ *	that need their own block copy routines into the buffer. There is no
+ *	guarantee the buffer is a DMA target!
+ *
+ *	Locking: May call functions taking tty->buf.lock
+ */
+
+int tty_prepare_flip_string_flags(struct tty_struct *tty,
+			unsigned char **chars, char **flags, size_t size)
+{
+	int space = tty_buffer_request_room(tty, size);
+	if (likely(space)) {
+		struct tty_buffer *tb = tty->buf.tail;
+		*chars = tb->char_buf_ptr + tb->used;
+		*flags = tb->flag_buf_ptr + tb->used;
+		tb->used += space;
+	}
+	return space;
+}
+EXPORT_SYMBOL_GPL(tty_prepare_flip_string_flags);
+
+
+
+/**
+ *	flush_to_ldisc
+ *	@work: tty structure passed from work queue.
+ *
+ *	This routine is called out of the software interrupt to flush data
+ *	from the buffer chain to the line discipline.
+ *
+ *	Locking: holds tty->buf.lock to guard buffer list. Drops the lock
+ *	while invoking the line discipline receive_buf method. The
+ *	receive_buf method is single threaded for each tty instance.
+ */
+
+static void flush_to_ldisc(struct work_struct *work)
+{
+	struct tty_struct *tty =
+		container_of(work, struct tty_struct, buf.work.work);
+	unsigned long 	flags;
+	struct tty_ldisc *disc;
+	struct tty_buffer *tbuf, *head;
+	char *char_buf;
+	unsigned char *flag_buf;
+
+	disc = tty_ldisc_ref(tty);
+	if (disc == NULL)	/*  !TTY_LDISC */
+		return;
+
+	spin_lock_irqsave(&tty->buf.lock, flags);
+	/* So we know a flush is running */
+	set_bit(TTY_FLUSHING, &tty->flags);
+	head = tty->buf.head;
+	if (head != NULL) {
+		tty->buf.head = NULL;
+		for (;;) {
+			int count = head->commit - head->read;
+			if (!count) {
+				if (head->next == NULL)
+					break;
+				tbuf = head;
+				head = head->next;
+				tty_buffer_free(tty, tbuf);
+				continue;
+			}
+			/* Ldisc or user is trying to flush the buffers
+			   we are feeding to the ldisc, stop feeding the
+			   line discipline as we want to empty the queue */
+			if (test_bit(TTY_FLUSHPENDING, &tty->flags))
+				break;
+			if (!tty->receive_room) {
+				schedule_delayed_work(&tty->buf.work, 1);
+				break;
+			}
+			if (count > tty->receive_room)
+				count = tty->receive_room;
+			char_buf = head->char_buf_ptr + head->read;
+			flag_buf = head->flag_buf_ptr + head->read;
+			head->read += count;
+			spin_unlock_irqrestore(&tty->buf.lock, flags);
+			disc->ops->receive_buf(tty, char_buf,
+							flag_buf, count);
+			spin_lock_irqsave(&tty->buf.lock, flags);
+		}
+		/* Restore the queue head */
+		tty->buf.head = head;
+	}
+	/* We may have a deferred request to flush the input buffer,
+	   if so pull the chain under the lock and empty the queue */
+	if (test_bit(TTY_FLUSHPENDING, &tty->flags)) {
+		__tty_buffer_flush(tty);
+		clear_bit(TTY_FLUSHPENDING, &tty->flags);
+		wake_up(&tty->read_wait);
+	}
+	clear_bit(TTY_FLUSHING, &tty->flags);
+	spin_unlock_irqrestore(&tty->buf.lock, flags);
+
+	tty_ldisc_deref(disc);
+}
+
+/**
+ *	tty_flip_buffer_push	-	terminal
+ *	@tty: tty to push
+ *
+ *	Queue a push of the terminal flip buffers to the line discipline. This
+ *	function must not be called from IRQ context if tty->low_latency is set.
+ *
+ *	In the event of the queue being busy for flipping the work will be
+ *	held off and retried later.
+ *
+ *	Locking: tty buffer lock. Driver locks in low latency mode.
+ */
+
+void tty_flip_buffer_push(struct tty_struct *tty)
+{
+	unsigned long flags;
+	spin_lock_irqsave(&tty->buf.lock, flags);
+	if (tty->buf.tail != NULL)
+		tty->buf.tail->commit = tty->buf.tail->used;
+	spin_unlock_irqrestore(&tty->buf.lock, flags);
+
+	if (tty->low_latency)
+		flush_to_ldisc(&tty->buf.work.work);
+	else
+		schedule_delayed_work(&tty->buf.work, 1);
+}
+EXPORT_SYMBOL(tty_flip_buffer_push);
+
+/**
+ *	tty_buffer_init		-	prepare a tty buffer structure
+ *	@tty: tty to initialise
+ *
+ *	Set up the initial state of the buffer management for a tty device.
+ *	Must be called before the other tty buffer functions are used.
+ *
+ *	Locking: none
+ */
+
+void tty_buffer_init(struct tty_struct *tty)
+{
+	spin_lock_init(&tty->buf.lock);
+	tty->buf.head = NULL;
+	tty->buf.tail = NULL;
+	tty->buf.free = NULL;
+	tty->buf.memory_used = 0;
+	INIT_DELAYED_WORK(&tty->buf.work, flush_to_ldisc);
+}
+
diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index e4dce8709541..7053d6333692 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -49,7 +49,7 @@
  * implement CONFIG_VT and generalize console device interface.
  *	-- Marko Kohtala <Marko.Kohtala@hut.fi>, March 97
  *
- * Rewrote init_dev and release_dev to eliminate races.
+ * Rewrote tty_init_dev and tty_release_dev to eliminate races.
  *	-- Bill Hawes <whawes@star.net>, June 97
  *
  * Added devfs support.
@@ -136,13 +136,6 @@ LIST_HEAD(tty_drivers);			/* linked list of tty drivers */
 DEFINE_MUTEX(tty_mutex);
 EXPORT_SYMBOL(tty_mutex);
 
-#ifdef CONFIG_UNIX98_PTYS
-extern struct tty_driver *ptm_driver;	/* Unix98 pty masters; for /dev/ptmx */
-static int ptmx_open(struct inode *, struct file *);
-#endif
-
-static void initialize_tty_struct(struct tty_struct *tty);
-
 static ssize_t tty_read(struct file *, char __user *, size_t, loff_t *);
 static ssize_t tty_write(struct file *, const char __user *, size_t, loff_t *);
 ssize_t redirected_tty_write(struct file *, const char __user *,
@@ -171,13 +164,11 @@ static void proc_set_tty(struct task_struct *tsk, struct tty_struct *tty);
  *	Locking: none
  */
 
-static struct tty_struct *alloc_tty_struct(void)
+struct tty_struct *alloc_tty_struct(void)
 {
 	return kzalloc(sizeof(struct tty_struct), GFP_KERNEL);
 }
 
-static void tty_buffer_free_all(struct tty_struct *);
-
 /**
  *	free_tty_struct		-	free a disused tty
  *	@tty: tty struct to free
@@ -187,7 +178,7 @@ static void tty_buffer_free_all(struct tty_struct *);
  *	Locking: none. Must be called after tty is definitely unused
  */
 
-static inline void free_tty_struct(struct tty_struct *tty)
+void free_tty_struct(struct tty_struct *tty)
 {
 	kfree(tty->write_buf);
 	tty_buffer_free_all(tty);
@@ -263,398 +254,6 @@ static int check_tty_count(struct tty_struct *tty, const char *routine)
 	return 0;
 }
 
-/*
- * Tty buffer allocation management
- */
-
-/**
- *	tty_buffer_free_all		-	free buffers used by a tty
- *	@tty: tty to free from
- *
- *	Remove all the buffers pending on a tty whether queued with data
- *	or in the free ring. Must be called when the tty is no longer in use
- *
- *	Locking: none
- */
-
-static void tty_buffer_free_all(struct tty_struct *tty)
-{
-	struct tty_buffer *thead;
-	while ((thead = tty->buf.head) != NULL) {
-		tty->buf.head = thead->next;
-		kfree(thead);
-	}
-	while ((thead = tty->buf.free) != NULL) {
-		tty->buf.free = thead->next;
-		kfree(thead);
-	}
-	tty->buf.tail = NULL;
-	tty->buf.memory_used = 0;
-}
-
-/**
- *	tty_buffer_init		-	prepare a tty buffer structure
- *	@tty: tty to initialise
- *
- *	Set up the initial state of the buffer management for a tty device.
- *	Must be called before the other tty buffer functions are used.
- *
- *	Locking: none
- */
-
-static void tty_buffer_init(struct tty_struct *tty)
-{
-	spin_lock_init(&tty->buf.lock);
-	tty->buf.head = NULL;
-	tty->buf.tail = NULL;
-	tty->buf.free = NULL;
-	tty->buf.memory_used = 0;
-}
-
-/**
- *	tty_buffer_alloc	-	allocate a tty buffer
- *	@tty: tty device
- *	@size: desired size (characters)
- *
- *	Allocate a new tty buffer to hold the desired number of characters.
- *	Return NULL if out of memory or the allocation would exceed the
- *	per device queue
- *
- *	Locking: Caller must hold tty->buf.lock
- */
-
-static struct tty_buffer *tty_buffer_alloc(struct tty_struct *tty, size_t size)
-{
-	struct tty_buffer *p;
-
-	if (tty->buf.memory_used + size > 65536)
-		return NULL;
-	p = kmalloc(sizeof(struct tty_buffer) + 2 * size, GFP_ATOMIC);
-	if (p == NULL)
-		return NULL;
-	p->used = 0;
-	p->size = size;
-	p->next = NULL;
-	p->commit = 0;
-	p->read = 0;
-	p->char_buf_ptr = (char *)(p->data);
-	p->flag_buf_ptr = (unsigned char *)p->char_buf_ptr + size;
-	tty->buf.memory_used += size;
-	return p;
-}
-
-/**
- *	tty_buffer_free		-	free a tty buffer
- *	@tty: tty owning the buffer
- *	@b: the buffer to free
- *
- *	Free a tty buffer, or add it to the free list according to our
- *	internal strategy
- *
- *	Locking: Caller must hold tty->buf.lock
- */
-
-static void tty_buffer_free(struct tty_struct *tty, struct tty_buffer *b)
-{
-	/* Dumb strategy for now - should keep some stats */
-	tty->buf.memory_used -= b->size;
-	WARN_ON(tty->buf.memory_used < 0);
-
-	if (b->size >= 512)
-		kfree(b);
-	else {
-		b->next = tty->buf.free;
-		tty->buf.free = b;
-	}
-}
-
-/**
- *	__tty_buffer_flush		-	flush full tty buffers
- *	@tty: tty to flush
- *
- *	flush all the buffers containing receive data. Caller must
- *	hold the buffer lock and must have ensured no parallel flush to
- *	ldisc is running.
- *
- *	Locking: Caller must hold tty->buf.lock
- */
-
-static void __tty_buffer_flush(struct tty_struct *tty)
-{
-	struct tty_buffer *thead;
-
-	while ((thead = tty->buf.head) != NULL) {
-		tty->buf.head = thead->next;
-		tty_buffer_free(tty, thead);
-	}
-	tty->buf.tail = NULL;
-}
-
-/**
- *	tty_buffer_flush		-	flush full tty buffers
- *	@tty: tty to flush
- *
- *	flush all the buffers containing receive data. If the buffer is
- *	being processed by flush_to_ldisc then we defer the processing
- *	to that function
- *
- *	Locking: none
- */
-
-static void tty_buffer_flush(struct tty_struct *tty)
-{
-	unsigned long flags;
-	spin_lock_irqsave(&tty->buf.lock, flags);
-
-	/* If the data is being pushed to the tty layer then we can't
-	   process it here. Instead set a flag and the flush_to_ldisc
-	   path will process the flush request before it exits */
-	if (test_bit(TTY_FLUSHING, &tty->flags)) {
-		set_bit(TTY_FLUSHPENDING, &tty->flags);
-		spin_unlock_irqrestore(&tty->buf.lock, flags);
-		wait_event(tty->read_wait,
-				test_bit(TTY_FLUSHPENDING, &tty->flags) == 0);
-		return;
-	} else
-		__tty_buffer_flush(tty);
-	spin_unlock_irqrestore(&tty->buf.lock, flags);
-}
-
-/**
- *	tty_buffer_find		-	find a free tty buffer
- *	@tty: tty owning the buffer
- *	@size: characters wanted
- *
- *	Locate an existing suitable tty buffer or if we are lacking one then
- *	allocate a new one. We round our buffers off in 256 character chunks
- *	to get better allocation behaviour.
- *
- *	Locking: Caller must hold tty->buf.lock
- */
-
-static struct tty_buffer *tty_buffer_find(struct tty_struct *tty, size_t size)
-{
-	struct tty_buffer **tbh = &tty->buf.free;
-	while ((*tbh) != NULL) {
-		struct tty_buffer *t = *tbh;
-		if (t->size >= size) {
-			*tbh = t->next;
-			t->next = NULL;
-			t->used = 0;
-			t->commit = 0;
-			t->read = 0;
-			tty->buf.memory_used += t->size;
-			return t;
-		}
-		tbh = &((*tbh)->next);
-	}
-	/* Round the buffer size out */
-	size = (size + 0xFF) & ~0xFF;
-	return tty_buffer_alloc(tty, size);
-	/* Should possibly check if this fails for the largest buffer we
-	   have queued and recycle that ? */
-}
-
-/**
- *	tty_buffer_request_room		-	grow tty buffer if needed
- *	@tty: tty structure
- *	@size: size desired
- *
- *	Make at least size bytes of linear space available for the tty
- *	buffer. If we fail return the size we managed to find.
- *
- *	Locking: Takes tty->buf.lock
- */
-int tty_buffer_request_room(struct tty_struct *tty, size_t size)
-{
-	struct tty_buffer *b, *n;
-	int left;
-	unsigned long flags;
-
-	spin_lock_irqsave(&tty->buf.lock, flags);
-
-	/* OPTIMISATION: We could keep a per tty "zero" sized buffer to
-	   remove this conditional if its worth it. This would be invisible
-	   to the callers */
-	if ((b = tty->buf.tail) != NULL)
-		left = b->size - b->used;
-	else
-		left = 0;
-
-	if (left < size) {
-		/* This is the slow path - looking for new buffers to use */
-		if ((n = tty_buffer_find(tty, size)) != NULL) {
-			if (b != NULL) {
-				b->next = n;
-				b->commit = b->used;
-			} else
-				tty->buf.head = n;
-			tty->buf.tail = n;
-		} else
-			size = left;
-	}
-
-	spin_unlock_irqrestore(&tty->buf.lock, flags);
-	return size;
-}
-EXPORT_SYMBOL_GPL(tty_buffer_request_room);
-
-/**
- *	tty_insert_flip_string	-	Add characters to the tty buffer
- *	@tty: tty structure
- *	@chars: characters
- *	@size: size
- *
- *	Queue a series of bytes to the tty buffering. All the characters
- *	passed are marked as without error. Returns the number added.
- *
- *	Locking: Called functions may take tty->buf.lock
- */
-
-int tty_insert_flip_string(struct tty_struct *tty, const unsigned char *chars,
-				size_t size)
-{
-	int copied = 0;
-	do {
-		int space = tty_buffer_request_room(tty, size - copied);
-		struct tty_buffer *tb = tty->buf.tail;
-		/* If there is no space then tb may be NULL */
-		if (unlikely(space == 0))
-			break;
-		memcpy(tb->char_buf_ptr + tb->used, chars, space);
-		memset(tb->flag_buf_ptr + tb->used, TTY_NORMAL, space);
-		tb->used += space;
-		copied += space;
-		chars += space;
-		/* There is a small chance that we need to split the data over
-		   several buffers. If this is the case we must loop */
-	} while (unlikely(size > copied));
-	return copied;
-}
-EXPORT_SYMBOL(tty_insert_flip_string);
-
-/**
- *	tty_insert_flip_string_flags	-	Add characters to the tty buffer
- *	@tty: tty structure
- *	@chars: characters
- *	@flags: flag bytes
- *	@size: size
- *
- *	Queue a series of bytes to the tty buffering. For each character
- *	the flags array indicates the status of the character. Returns the
- *	number added.
- *
- *	Locking: Called functions may take tty->buf.lock
- */
-
-int tty_insert_flip_string_flags(struct tty_struct *tty,
-		const unsigned char *chars, const char *flags, size_t size)
-{
-	int copied = 0;
-	do {
-		int space = tty_buffer_request_room(tty, size - copied);
-		struct tty_buffer *tb = tty->buf.tail;
-		/* If there is no space then tb may be NULL */
-		if (unlikely(space == 0))
-			break;
-		memcpy(tb->char_buf_ptr + tb->used, chars, space);
-		memcpy(tb->flag_buf_ptr + tb->used, flags, space);
-		tb->used += space;
-		copied += space;
-		chars += space;
-		flags += space;
-		/* There is a small chance that we need to split the data over
-		   several buffers. If this is the case we must loop */
-	} while (unlikely(size > copied));
-	return copied;
-}
-EXPORT_SYMBOL(tty_insert_flip_string_flags);
-
-/**
- *	tty_schedule_flip	-	push characters to ldisc
- *	@tty: tty to push from
- *
- *	Takes any pending buffers and transfers their ownership to the
- *	ldisc side of the queue. It then schedules those characters for
- *	processing by the line discipline.
- *
- *	Locking: Takes tty->buf.lock
- */
-
-void tty_schedule_flip(struct tty_struct *tty)
-{
-	unsigned long flags;
-	spin_lock_irqsave(&tty->buf.lock, flags);
-	if (tty->buf.tail != NULL)
-		tty->buf.tail->commit = tty->buf.tail->used;
-	spin_unlock_irqrestore(&tty->buf.lock, flags);
-	schedule_delayed_work(&tty->buf.work, 1);
-}
-EXPORT_SYMBOL(tty_schedule_flip);
-
-/**
- *	tty_prepare_flip_string		-	make room for characters
- *	@tty: tty
- *	@chars: return pointer for character write area
- *	@size: desired size
- *
- *	Prepare a block of space in the buffer for data. Returns the length
- *	available and buffer pointer to the space which is now allocated and
- *	accounted for as ready for normal characters. This is used for drivers
- *	that need their own block copy routines into the buffer. There is no
- *	guarantee the buffer is a DMA target!
- *
- *	Locking: May call functions taking tty->buf.lock
- */
-
-int tty_prepare_flip_string(struct tty_struct *tty, unsigned char **chars,
-								size_t size)
-{
-	int space = tty_buffer_request_room(tty, size);
-	if (likely(space)) {
-		struct tty_buffer *tb = tty->buf.tail;
-		*chars = tb->char_buf_ptr + tb->used;
-		memset(tb->flag_buf_ptr + tb->used, TTY_NORMAL, space);
-		tb->used += space;
-	}
-	return space;
-}
-
-EXPORT_SYMBOL_GPL(tty_prepare_flip_string);
-
-/**
- *	tty_prepare_flip_string_flags	-	make room for characters
- *	@tty: tty
- *	@chars: return pointer for character write area
- *	@flags: return pointer for status flag write area
- *	@size: desired size
- *
- *	Prepare a block of space in the buffer for data. Returns the length
- *	available and buffer pointer to the space which is now allocated and
- *	accounted for as ready for characters. This is used for drivers
- *	that need their own block copy routines into the buffer. There is no
- *	guarantee the buffer is a DMA target!
- *
- *	Locking: May call functions taking tty->buf.lock
- */
-
-int tty_prepare_flip_string_flags(struct tty_struct *tty,
-			unsigned char **chars, char **flags, size_t size)
-{
-	int space = tty_buffer_request_room(tty, size);
-	if (likely(space)) {
-		struct tty_buffer *tb = tty->buf.tail;
-		*chars = tb->char_buf_ptr + tb->used;
-		*flags = tb->flag_buf_ptr + tb->used;
-		tb->used += space;
-	}
-	return space;
-}
-
-EXPORT_SYMBOL_GPL(tty_prepare_flip_string_flags);
-
-
-
 /**
  *	get_tty_driver		-	find device of a tty
  *	@dev_t: device identifier
@@ -675,7 +274,7 @@ static struct tty_driver *get_tty_driver(dev_t device, int *index)
 		if (device < base || device >= base + p->num)
 			continue;
 		*index = device - base;
-		return p;
+		return tty_driver_kref_get(p);
 	}
 	return NULL;
 }
@@ -719,7 +318,7 @@ struct tty_driver *tty_find_polling_driver(char *name, int *line)
 
 		if (tty_line >= 0 && tty_line <= p->num && p->ops &&
 		    p->ops->poll_init && !p->ops->poll_init(p, tty_line, str)) {
-			res = p;
+			res = tty_driver_kref_get(p);
 			*line = tty_line;
 			break;
 		}
@@ -819,20 +418,6 @@ static const struct file_operations tty_fops = {
 	.fasync		= tty_fasync,
 };
 
-#ifdef CONFIG_UNIX98_PTYS
-static const struct file_operations ptmx_fops = {
-	.llseek		= no_llseek,
-	.read		= tty_read,
-	.write		= tty_write,
-	.poll		= tty_poll,
-	.unlocked_ioctl	= tty_ioctl,
-	.compat_ioctl	= tty_compat_ioctl,
-	.open		= ptmx_open,
-	.release	= tty_release,
-	.fasync		= tty_fasync,
-};
-#endif
-
 static const struct file_operations console_fops = {
 	.llseek		= no_llseek,
 	.read		= tty_read,
@@ -953,6 +538,7 @@ static void do_tty_hangup(struct work_struct *work)
 	struct tty_ldisc *ld;
 	int    closecount = 0, n;
 	unsigned long flags;
+	int refs = 0;
 
 	if (!tty)
 		return;
@@ -1019,8 +605,12 @@ static void do_tty_hangup(struct work_struct *work)
 	if (tty->session) {
 		do_each_pid_task(tty->session, PIDTYPE_SID, p) {
 			spin_lock_irq(&p->sighand->siglock);
-			if (p->signal->tty == tty)
+			if (p->signal->tty == tty) {
 				p->signal->tty = NULL;
+				/* We defer the dereferences outside fo
+				   the tasklist lock */
+				refs++;
+			}
 			if (!p->signal->leader) {
 				spin_unlock_irq(&p->sighand->siglock);
 				continue;
@@ -1046,6 +636,10 @@ static void do_tty_hangup(struct work_struct *work)
 	tty->ctrl_status = 0;
 	spin_unlock_irqrestore(&tty->ctrl_lock, flags);
 
+	/* Account for the p->signal references we killed */
+	while (refs--)
+		tty_kref_put(tty);
+
 	/*
 	 * If one of the devices matches a console pointer, we
 	 * cannot just call hangup() because that will cause
@@ -1115,6 +709,23 @@ void tty_vhangup(struct tty_struct *tty)
 EXPORT_SYMBOL(tty_vhangup);
 
 /**
+ *	tty_vhangup_self	-	process vhangup for own ctty
+ *
+ *	Perform a vhangup on the current controlling tty
+ */
+
+void tty_vhangup_self(void)
+{
+	struct tty_struct *tty;
+
+	tty = get_current_tty();
+	if (tty) {
+		tty_vhangup(tty);
+		tty_kref_put(tty);
+	}
+}
+
+/**
  *	tty_hung_up_p		-	was tty hung up
  *	@filp: file pointer of tty
  *
@@ -1167,16 +778,14 @@ void disassociate_ctty(int on_exit)
 	struct pid *tty_pgrp = NULL;
 
 
-	mutex_lock(&tty_mutex);
 	tty = get_current_tty();
 	if (tty) {
 		tty_pgrp = get_pid(tty->pgrp);
 		lock_kernel();
-		mutex_unlock(&tty_mutex);
-		/* XXX: here we race, there is nothing protecting tty */
 		if (on_exit && tty->driver->type != TTY_DRIVER_TYPE_PTY)
 			tty_vhangup(tty);
 		unlock_kernel();
+		tty_kref_put(tty);
 	} else if (on_exit) {
 		struct pid *old_pgrp;
 		spin_lock_irq(&current->sighand->siglock);
@@ -1188,7 +797,6 @@ void disassociate_ctty(int on_exit)
 			kill_pgrp(old_pgrp, SIGCONT, on_exit);
 			put_pid(old_pgrp);
 		}
-		mutex_unlock(&tty_mutex);
 		return;
 	}
 	if (tty_pgrp) {
@@ -1203,8 +811,6 @@ void disassociate_ctty(int on_exit)
 	current->signal->tty_old_pgrp = NULL;
 	spin_unlock_irq(&current->sighand->siglock);
 
-	mutex_lock(&tty_mutex);
-	/* It is possible that do_tty_hangup has free'd this tty */
 	tty = get_current_tty();
 	if (tty) {
 		unsigned long flags;
@@ -1214,13 +820,13 @@ void disassociate_ctty(int on_exit)
 		tty->session = NULL;
 		tty->pgrp = NULL;
 		spin_unlock_irqrestore(&tty->ctrl_lock, flags);
+		tty_kref_put(tty);
 	} else {
 #ifdef TTY_DEBUG_HANGUP
 		printk(KERN_DEBUG "error attempted to write to tty [0x%p]"
 		       " = NULL", tty);
 #endif
 	}
-	mutex_unlock(&tty_mutex);
 
 	/* Now clear signal->tty under the lock */
 	read_lock(&tasklist_lock);
@@ -1420,19 +1026,19 @@ static inline ssize_t do_tty_write(
 
 	/* write_buf/write_cnt is protected by the atomic_write_lock mutex */
 	if (tty->write_cnt < chunk) {
-		unsigned char *buf;
+		unsigned char *buf_chunk;
 
 		if (chunk < 1024)
 			chunk = 1024;
 
-		buf = kmalloc(chunk, GFP_KERNEL);
-		if (!buf) {
+		buf_chunk = kmalloc(chunk, GFP_KERNEL);
+		if (!buf_chunk) {
 			ret = -ENOMEM;
 			goto out;
 		}
 		kfree(tty->write_buf);
 		tty->write_cnt = chunk;
-		tty->write_buf = buf;
+		tty->write_buf = buf_chunk;
 	}
 
 	/* Do the write .. */
@@ -1466,6 +1072,31 @@ out:
 	return ret;
 }
 
+/**
+ * tty_write_message - write a message to a certain tty, not just the console.
+ * @tty: the destination tty_struct
+ * @msg: the message to write
+ *
+ * This is used for messages that need to be redirected to a specific tty.
+ * We don't put it into the syslog queue right now maybe in the future if
+ * really needed.
+ *
+ * We must still hold the BKL and test the CLOSING flag for the moment.
+ */
+
+void tty_write_message(struct tty_struct *tty, char *msg)
+{
+	lock_kernel();
+	if (tty) {
+		mutex_lock(&tty->atomic_write_lock);
+		if (tty->ops->write && !test_bit(TTY_CLOSING, &tty->flags))
+			tty->ops->write(tty, msg, strlen(msg));
+		tty_write_unlock(tty);
+	}
+	unlock_kernel();
+	return;
+}
+
 
 /**
  *	tty_write		-	write method for tty device file
@@ -1533,42 +1164,6 @@ ssize_t redirected_tty_write(struct file *file, const char __user *buf,
 	return tty_write(file, buf, count, ppos);
 }
 
-void tty_port_init(struct tty_port *port)
-{
-	memset(port, 0, sizeof(*port));
-	init_waitqueue_head(&port->open_wait);
-	init_waitqueue_head(&port->close_wait);
-	mutex_init(&port->mutex);
-	port->close_delay = (50 * HZ) / 100;
-	port->closing_wait = (3000 * HZ) / 100;
-}
-EXPORT_SYMBOL(tty_port_init);
-
-int tty_port_alloc_xmit_buf(struct tty_port *port)
-{
-	/* We may sleep in get_zeroed_page() */
-	mutex_lock(&port->mutex);
-	if (port->xmit_buf == NULL)
-		port->xmit_buf = (unsigned char *)get_zeroed_page(GFP_KERNEL);
-	mutex_unlock(&port->mutex);
-	if (port->xmit_buf == NULL)
-		return -ENOMEM;
-	return 0;
-}
-EXPORT_SYMBOL(tty_port_alloc_xmit_buf);
-
-void tty_port_free_xmit_buf(struct tty_port *port)
-{
-	mutex_lock(&port->mutex);
-	if (port->xmit_buf != NULL) {
-		free_page((unsigned long)port->xmit_buf);
-		port->xmit_buf = NULL;
-	}
-	mutex_unlock(&port->mutex);
-}
-EXPORT_SYMBOL(tty_port_free_xmit_buf);
-
-
 static char ptychar[] = "pqrstuvwxyzabcde";
 
 /**
@@ -1592,7 +1187,7 @@ static void pty_line_name(struct tty_driver *driver, int index, char *p)
 }
 
 /**
- *	pty_line_name	-	generate name for a tty
+ *	tty_line_name	-	generate name for a tty
  *	@driver: the tty driver in use
  *	@index: the minor number
  *	@p: output buffer of at least 7 bytes
@@ -1608,10 +1203,148 @@ static void tty_line_name(struct tty_driver *driver, int index, char *p)
 }
 
 /**
- *	init_dev		-	initialise a tty device
+ *	tty_driver_lookup_tty() - find an existing tty, if any
+ *	@driver: the driver for the tty
+ *	@idx:	 the minor number
+ *
+ *	Return the tty, if found or ERR_PTR() otherwise.
+ *
+ *	Locking: tty_mutex must be held. If tty is found, the mutex must
+ *	be held until the 'fast-open' is also done. Will change once we
+ *	have refcounting in the driver and per driver locking
+ */
+struct tty_struct *tty_driver_lookup_tty(struct tty_driver *driver,
+		struct inode *inode, int idx)
+{
+	struct tty_struct *tty;
+
+	if (driver->ops->lookup)
+		return driver->ops->lookup(driver, inode, idx);
+
+	tty = driver->ttys[idx];
+	return tty;
+}
+
+/**
+ *	tty_init_termios	-  helper for termios setup
+ *	@tty: the tty to set up
+ *
+ *	Initialise the termios structures for this tty. Thus runs under
+ *	the tty_mutex currently so we can be relaxed about ordering.
+ */
+
+int tty_init_termios(struct tty_struct *tty)
+{
+	struct ktermios *tp;
+	int idx = tty->index;
+
+	tp = tty->driver->termios[idx];
+	if (tp == NULL) {
+		tp = kzalloc(sizeof(struct ktermios[2]), GFP_KERNEL);
+		if (tp == NULL)
+			return -ENOMEM;
+		memcpy(tp, &tty->driver->init_termios,
+						sizeof(struct ktermios));
+		tty->driver->termios[idx] = tp;
+	}
+	tty->termios = tp;
+	tty->termios_locked = tp + 1;
+
+	/* Compatibility until drivers always set this */
+	tty->termios->c_ispeed = tty_termios_input_baud_rate(tty->termios);
+	tty->termios->c_ospeed = tty_termios_baud_rate(tty->termios);
+	return 0;
+}
+
+/**
+ *	tty_driver_install_tty() - install a tty entry in the driver
+ *	@driver: the driver for the tty
+ *	@tty: the tty
+ *
+ *	Install a tty object into the driver tables. The tty->index field
+ *	will be set by the time this is called. This method is responsible
+ *	for ensuring any need additional structures are allocated and
+ *	configured.
+ *
+ *	Locking: tty_mutex for now
+ */
+static int tty_driver_install_tty(struct tty_driver *driver,
+						struct tty_struct *tty)
+{
+	int idx = tty->index;
+
+	if (driver->ops->install)
+		return driver->ops->install(driver, tty);
+
+	if (tty_init_termios(tty) == 0) {
+		tty_driver_kref_get(driver);
+		tty->count++;
+		driver->ttys[idx] = tty;
+		return 0;
+	}
+	return -ENOMEM;
+}
+
+/**
+ *	tty_driver_remove_tty() - remove a tty from the driver tables
+ *	@driver: the driver for the tty
+ *	@idx:	 the minor number
+ *
+ *	Remvoe a tty object from the driver tables. The tty->index field
+ *	will be set by the time this is called.
+ *
+ *	Locking: tty_mutex for now
+ */
+static void tty_driver_remove_tty(struct tty_driver *driver,
+						struct tty_struct *tty)
+{
+	if (driver->ops->remove)
+		driver->ops->remove(driver, tty);
+	else
+		driver->ttys[tty->index] = NULL;
+}
+
+/*
+ * 	tty_reopen()	- fast re-open of an open tty
+ * 	@tty	- the tty to open
+ *
+ *	Return 0 on success, -errno on error.
+ *
+ *	Locking: tty_mutex must be held from the time the tty was found
+ *		 till this open completes.
+ */
+static int tty_reopen(struct tty_struct *tty)
+{
+	struct tty_driver *driver = tty->driver;
+
+	if (test_bit(TTY_CLOSING, &tty->flags))
+		return -EIO;
+
+	if (driver->type == TTY_DRIVER_TYPE_PTY &&
+	    driver->subtype == PTY_TYPE_MASTER) {
+		/*
+		 * special case for PTY masters: only one open permitted,
+		 * and the slave side open count is incremented as well.
+		 */
+		if (tty->count)
+			return -EIO;
+
+		tty->link->count++;
+	}
+	tty->count++;
+	tty->driver = driver; /* N.B. why do this every time?? */
+
+	WARN_ON(!test_bit(TTY_LDISC, &tty->flags));
+
+	return 0;
+}
+
+/**
+ *	tty_init_dev		-	initialise a tty device
  *	@driver: tty driver we are opening a device on
  *	@idx: device index
- *	@tty: returned tty structure
+ *	@ret_tty: returned tty structure
+ *	@first_ok: ok to open a new device (used by ptmx)
  *
  *	Prepare a tty device. This may not be a "new" clean device but
  *	could also be an active device. The pty drivers require special
@@ -1631,37 +1364,16 @@ static void tty_line_name(struct tty_driver *driver, int index, char *p)
  * relaxed for the (most common) case of reopening a tty.
  */
 
-static int init_dev(struct tty_driver *driver, int idx,
-	struct tty_struct **ret_tty)
+struct tty_struct *tty_init_dev(struct tty_driver *driver, int idx,
+								int first_ok)
 {
-	struct tty_struct *tty, *o_tty;
-	struct ktermios *tp, **tp_loc, *o_tp, **o_tp_loc;
-	struct ktermios *ltp, **ltp_loc, *o_ltp, **o_ltp_loc;
-	int retval = 0;
+	struct tty_struct *tty;
+	int retval;
 
-	/* check whether we're reopening an existing tty */
-	if (driver->flags & TTY_DRIVER_DEVPTS_MEM) {
-		tty = devpts_get_tty(idx);
-		/*
-		 * If we don't have a tty here on a slave open, it's because
-		 * the master already started the close process and there's
-		 * no relation between devpts file and tty anymore.
-		 */
-		if (!tty && driver->subtype == PTY_TYPE_SLAVE) {
-			retval = -EIO;
-			goto end_init;
-		}
-		/*
-		 * It's safe from now on because init_dev() is called with
-		 * tty_mutex held and release_dev() won't change tty->count
-		 * or tty->flags without having to grab tty_mutex
-		 */
-		if (tty && driver->subtype == PTY_TYPE_MASTER)
-			tty = tty->link;
-	} else {
-		tty = driver->ttys[idx];
-	}
-	if (tty) goto fast_track;
+	/* Check if pty master is being opened multiple times */
+	if (driver->subtype == PTY_TYPE_MASTER &&
+		(driver->flags & TTY_DRIVER_DEVPTS_MEM) && !first_ok)
+		return ERR_PTR(-EIO);
 
 	/*
 	 * First time open is complex, especially for PTY devices.
@@ -1671,189 +1383,69 @@ static int init_dev(struct tty_driver *driver, int idx,
 	 * and locked termios may be retained.)
 	 */
 
-	if (!try_module_get(driver->owner)) {
-		retval = -ENODEV;
-		goto end_init;
-	}
-
-	o_tty = NULL;
-	tp = o_tp = NULL;
-	ltp = o_ltp = NULL;
+	if (!try_module_get(driver->owner))
+		return ERR_PTR(-ENODEV);
 
 	tty = alloc_tty_struct();
 	if (!tty)
 		goto fail_no_mem;
-	initialize_tty_struct(tty);
-	tty->driver = driver;
-	tty->ops = driver->ops;
-	tty->index = idx;
-	tty_line_name(driver, idx, tty->name);
-
-	if (driver->flags & TTY_DRIVER_DEVPTS_MEM) {
-		tp_loc = &tty->termios;
-		ltp_loc = &tty->termios_locked;
-	} else {
-		tp_loc = &driver->termios[idx];
-		ltp_loc = &driver->termios_locked[idx];
-	}
-
-	if (!*tp_loc) {
-		tp = kmalloc(sizeof(struct ktermios), GFP_KERNEL);
-		if (!tp)
-			goto free_mem_out;
-		*tp = driver->init_termios;
-	}
-
-	if (!*ltp_loc) {
-		ltp = kzalloc(sizeof(struct ktermios), GFP_KERNEL);
-		if (!ltp)
-			goto free_mem_out;
-	}
-
-	if (driver->type == TTY_DRIVER_TYPE_PTY) {
-		o_tty = alloc_tty_struct();
-		if (!o_tty)
-			goto free_mem_out;
-		initialize_tty_struct(o_tty);
-		o_tty->driver = driver->other;
-		o_tty->ops = driver->ops;
-		o_tty->index = idx;
-		tty_line_name(driver->other, idx, o_tty->name);
+	initialize_tty_struct(tty, driver, idx);
 
-		if (driver->flags & TTY_DRIVER_DEVPTS_MEM) {
-			o_tp_loc = &o_tty->termios;
-			o_ltp_loc = &o_tty->termios_locked;
-		} else {
-			o_tp_loc = &driver->other->termios[idx];
-			o_ltp_loc = &driver->other->termios_locked[idx];
-		}
-
-		if (!*o_tp_loc) {
-			o_tp = kmalloc(sizeof(struct ktermios), GFP_KERNEL);
-			if (!o_tp)
-				goto free_mem_out;
-			*o_tp = driver->other->init_termios;
-		}
-
-		if (!*o_ltp_loc) {
-			o_ltp = kzalloc(sizeof(struct ktermios), GFP_KERNEL);
-			if (!o_ltp)
-				goto free_mem_out;
-		}
-
-		/*
-		 * Everything allocated ... set up the o_tty structure.
-		 */
-		if (!(driver->other->flags & TTY_DRIVER_DEVPTS_MEM))
-			driver->other->ttys[idx] = o_tty;
-		if (!*o_tp_loc)
-			*o_tp_loc = o_tp;
-		if (!*o_ltp_loc)
-			*o_ltp_loc = o_ltp;
-		o_tty->termios = *o_tp_loc;
-		o_tty->termios_locked = *o_ltp_loc;
-		driver->other->refcount++;
-		if (driver->subtype == PTY_TYPE_MASTER)
-			o_tty->count++;
-
-		/* Establish the links in both directions */
-		tty->link   = o_tty;
-		o_tty->link = tty;
+	retval = tty_driver_install_tty(driver, tty);
+	if (retval < 0) {
+		free_tty_struct(tty);
+		module_put(driver->owner);
+		return ERR_PTR(retval);
 	}
 
 	/*
-	 * All structures have been allocated, so now we install them.
-	 * Failures after this point use release_tty to clean up, so
-	 * there's no need to null out the local pointers.
-	 */
-	if (!(driver->flags & TTY_DRIVER_DEVPTS_MEM))
-		driver->ttys[idx] = tty;
-
-	if (!*tp_loc)
-		*tp_loc = tp;
-	if (!*ltp_loc)
-		*ltp_loc = ltp;
-	tty->termios = *tp_loc;
-	tty->termios_locked = *ltp_loc;
-	/* Compatibility until drivers always set this */
-	tty->termios->c_ispeed = tty_termios_input_baud_rate(tty->termios);
-	tty->termios->c_ospeed = tty_termios_baud_rate(tty->termios);
-	driver->refcount++;
-	tty->count++;
-
-	/*
 	 * Structures all installed ... call the ldisc open routines.
 	 * If we fail here just call release_tty to clean up.  No need
 	 * to decrement the use counts, as release_tty doesn't care.
 	 */
 
-	retval = tty_ldisc_setup(tty, o_tty);
-
+	retval = tty_ldisc_setup(tty, tty->link);
 	if (retval)
 		goto release_mem_out;
-	 goto success;
-
-	/*
-	 * This fast open can be used if the tty is already open.
-	 * No memory is allocated, and the only failures are from
-	 * attempting to open a closing tty or attempting multiple
-	 * opens on a pty master.
-	 */
-fast_track:
-	if (test_bit(TTY_CLOSING, &tty->flags)) {
-		retval = -EIO;
-		goto end_init;
-	}
-	if (driver->type == TTY_DRIVER_TYPE_PTY &&
-	    driver->subtype == PTY_TYPE_MASTER) {
-		/*
-		 * special case for PTY masters: only one open permitted,
-		 * and the slave side open count is incremented as well.
-		 */
-		if (tty->count) {
-			retval = -EIO;
-			goto end_init;
-		}
-		tty->link->count++;
-	}
-	tty->count++;
-	tty->driver = driver; /* N.B. why do this every time?? */
-
-	/* FIXME */
-	if (!test_bit(TTY_LDISC, &tty->flags))
-		printk(KERN_ERR "init_dev but no ldisc\n");
-success:
-	*ret_tty = tty;
-
-	/* All paths come through here to release the mutex */
-end_init:
-	return retval;
-
-	/* Release locally allocated memory ... nothing placed in slots */
-free_mem_out:
-	kfree(o_tp);
-	if (o_tty)
-		free_tty_struct(o_tty);
-	kfree(ltp);
-	kfree(tp);
-	free_tty_struct(tty);
+	return tty;
 
 fail_no_mem:
 	module_put(driver->owner);
-	retval = -ENOMEM;
-	goto end_init;
+	return ERR_PTR(-ENOMEM);
 
 	/* call the tty release_tty routine to clean out this slot */
 release_mem_out:
 	if (printk_ratelimit())
-		printk(KERN_INFO "init_dev: ldisc open failed, "
+		printk(KERN_INFO "tty_init_dev: ldisc open failed, "
 				 "clearing slot %d\n", idx);
 	release_tty(tty, idx);
-	goto end_init;
+	return ERR_PTR(retval);
 }
 
+void tty_free_termios(struct tty_struct *tty)
+{
+	struct ktermios *tp;
+	int idx = tty->index;
+	/* Kill this flag and push into drivers for locking etc */
+	if (tty->driver->flags & TTY_DRIVER_RESET_TERMIOS) {
+		/* FIXME: Locking on ->termios array */
+		tp = tty->termios;
+		tty->driver->termios[idx] = NULL;
+		kfree(tp);
+	}
+}
+EXPORT_SYMBOL(tty_free_termios);
+
+void tty_shutdown(struct tty_struct *tty)
+{
+	tty_driver_remove_tty(tty->driver, tty);
+	tty_free_termios(tty);
+}
+EXPORT_SYMBOL(tty_shutdown);
+
 /**
  *	release_one_tty		-	release tty structure memory
+ *	@kref: kref of tty we are obliterating
  *
  *	Releases memory associated with a tty structure, and clears out the
  *	driver table slots. This function is called when a device is no longer
@@ -1863,31 +1455,19 @@ release_mem_out:
  *		tty_mutex - sometimes only
  *		takes the file list lock internally when working on the list
  *	of ttys that the driver keeps.
- *		FIXME: should we require tty_mutex is held here ??
  */
-static void release_one_tty(struct tty_struct *tty, int idx)
+static void release_one_tty(struct kref *kref)
 {
-	int devpts = tty->driver->flags & TTY_DRIVER_DEVPTS_MEM;
-	struct ktermios *tp;
-
-	if (!devpts)
-		tty->driver->ttys[idx] = NULL;
-
-	if (tty->driver->flags & TTY_DRIVER_RESET_TERMIOS) {
-		tp = tty->termios;
-		if (!devpts)
-			tty->driver->termios[idx] = NULL;
-		kfree(tp);
-
-		tp = tty->termios_locked;
-		if (!devpts)
-			tty->driver->termios_locked[idx] = NULL;
-		kfree(tp);
-	}
-
+	struct tty_struct *tty = container_of(kref, struct tty_struct, kref);
+	struct tty_driver *driver = tty->driver;
 
+	if (tty->ops->shutdown)
+		tty->ops->shutdown(tty);
+	else
+		tty_shutdown(tty);
 	tty->magic = 0;
-	tty->driver->refcount--;
+	tty_driver_kref_put(driver);
+	module_put(driver->owner);
 
 	file_list_lock();
 	list_del_init(&tty->tty_files);
@@ -1897,6 +1477,21 @@ static void release_one_tty(struct tty_struct *tty, int idx)
 }
 
 /**
+ *	tty_kref_put		-	release a tty kref
+ *	@tty: tty device
+ *
+ *	Release a reference to a tty device and if need be let the kref
+ *	layer destruct the object for us
+ */
+
+void tty_kref_put(struct tty_struct *tty)
+{
+	if (tty)
+		kref_put(&tty->kref, release_one_tty);
+}
+EXPORT_SYMBOL(tty_kref_put);
+
+/**
  *	release_tty		-	release tty structure memory
  *
  *	Release both @tty and a possible linked partner (think pty pair),
@@ -1907,15 +1502,16 @@ static void release_one_tty(struct tty_struct *tty, int idx)
  *		takes the file list lock internally when working on the list
  *	of ttys that the driver keeps.
  *		FIXME: should we require tty_mutex is held here ??
+ *
  */
 static void release_tty(struct tty_struct *tty, int idx)
 {
-	struct tty_driver *driver = tty->driver;
+	/* This should always be true but check for the moment */
+	WARN_ON(tty->index != idx);
 
 	if (tty->link)
-		release_one_tty(tty->link, idx);
-	release_one_tty(tty, idx);
-	module_put(driver->owner);
+		tty_kref_put(tty->link);
+	tty_kref_put(tty);
 }
 
 /*
@@ -1926,20 +1522,21 @@ static void release_tty(struct tty_struct *tty, int idx)
  * WSH 09/09/97: rewritten to avoid some nasty race conditions that could
  * lead to double frees or releasing memory still in use.
  */
-static void release_dev(struct file *filp)
+void tty_release_dev(struct file *filp)
 {
 	struct tty_struct *tty, *o_tty;
 	int	pty_master, tty_closing, o_tty_closing, do_sleep;
 	int	devpts;
 	int	idx;
 	char	buf[64];
+	struct 	inode *inode;
 
+	inode = filp->f_path.dentry->d_inode;
 	tty = (struct tty_struct *)filp->private_data;
-	if (tty_paranoia_check(tty, filp->f_path.dentry->d_inode,
-							"release_dev"))
+	if (tty_paranoia_check(tty, inode, "tty_release_dev"))
 		return;
 
-	check_tty_count(tty, "release_dev");
+	check_tty_count(tty, "tty_release_dev");
 
 	tty_fasync(-1, filp, 0);
 
@@ -1951,33 +1548,27 @@ static void release_dev(struct file *filp)
 
 #ifdef TTY_PARANOIA_CHECK
 	if (idx < 0 || idx >= tty->driver->num) {
-		printk(KERN_DEBUG "release_dev: bad idx when trying to "
+		printk(KERN_DEBUG "tty_release_dev: bad idx when trying to "
 				  "free (%s)\n", tty->name);
 		return;
 	}
-	if (!(tty->driver->flags & TTY_DRIVER_DEVPTS_MEM)) {
+	if (!devpts) {
 		if (tty != tty->driver->ttys[idx]) {
-			printk(KERN_DEBUG "release_dev: driver.table[%d] not tty "
+			printk(KERN_DEBUG "tty_release_dev: driver.table[%d] not tty "
 			       "for (%s)\n", idx, tty->name);
 			return;
 		}
 		if (tty->termios != tty->driver->termios[idx]) {
-			printk(KERN_DEBUG "release_dev: driver.termios[%d] not termios "
+			printk(KERN_DEBUG "tty_release_dev: driver.termios[%d] not termios "
 			       "for (%s)\n",
 			       idx, tty->name);
 			return;
 		}
-		if (tty->termios_locked != tty->driver->termios_locked[idx]) {
-			printk(KERN_DEBUG "release_dev: driver.termios_locked[%d] not "
-			       "termios_locked for (%s)\n",
-			       idx, tty->name);
-			return;
-		}
 	}
 #endif
 
 #ifdef TTY_DEBUG_HANGUP
-	printk(KERN_DEBUG "release_dev of %s (tty count=%d)...",
+	printk(KERN_DEBUG "tty_release_dev of %s (tty count=%d)...",
 	       tty_name(tty, buf), tty->count);
 #endif
 
@@ -1985,26 +1576,19 @@ static void release_dev(struct file *filp)
 	if (tty->driver->other &&
 	     !(tty->driver->flags & TTY_DRIVER_DEVPTS_MEM)) {
 		if (o_tty != tty->driver->other->ttys[idx]) {
-			printk(KERN_DEBUG "release_dev: other->table[%d] "
+			printk(KERN_DEBUG "tty_release_dev: other->table[%d] "
 					  "not o_tty for (%s)\n",
 			       idx, tty->name);
 			return;
 		}
 		if (o_tty->termios != tty->driver->other->termios[idx]) {
-			printk(KERN_DEBUG "release_dev: other->termios[%d] "
+			printk(KERN_DEBUG "tty_release_dev: other->termios[%d] "
 					  "not o_termios for (%s)\n",
 			       idx, tty->name);
 			return;
 		}
-		if (o_tty->termios_locked !=
-		      tty->driver->other->termios_locked[idx]) {
-			printk(KERN_DEBUG "release_dev: other->termios_locked["
-					  "%d] not o_termios_locked for (%s)\n",
-			       idx, tty->name);
-			return;
-		}
 		if (o_tty->link != tty) {
-			printk(KERN_DEBUG "release_dev: bad pty pointers\n");
+			printk(KERN_DEBUG "tty_release_dev: bad pty pointers\n");
 			return;
 		}
 	}
@@ -2062,7 +1646,7 @@ static void release_dev(struct file *filp)
 		if (!do_sleep)
 			break;
 
-		printk(KERN_WARNING "release_dev: %s: read/write wait queue "
+		printk(KERN_WARNING "tty_release_dev: %s: read/write wait queue "
 				    "active!\n", tty_name(tty, buf));
 		mutex_unlock(&tty_mutex);
 		schedule();
@@ -2075,14 +1659,14 @@ static void release_dev(struct file *filp)
 	 */
 	if (pty_master) {
 		if (--o_tty->count < 0) {
-			printk(KERN_WARNING "release_dev: bad pty slave count "
+			printk(KERN_WARNING "tty_release_dev: bad pty slave count "
 					    "(%d) for %s\n",
 			       o_tty->count, tty_name(o_tty, buf));
 			o_tty->count = 0;
 		}
 	}
 	if (--tty->count < 0) {
-		printk(KERN_WARNING "release_dev: bad tty->count (%d) for %s\n",
+		printk(KERN_WARNING "tty_release_dev: bad tty->count (%d) for %s\n",
 		       tty->count, tty_name(tty, buf));
 		tty->count = 0;
 	}
@@ -2145,11 +1729,11 @@ static void release_dev(struct file *filp)
 
 	/* Make this pty number available for reallocation */
 	if (devpts)
-		devpts_kill_index(idx);
+		devpts_kill_index(inode, idx);
 }
 
 /**
- *	tty_open		-	open a tty device
+ *	__tty_open		-	open a tty device
  *	@inode: inode of device file
  *	@filp: file pointer to tty
  *
@@ -2164,14 +1748,14 @@ static void release_dev(struct file *filp)
  *	The termios state of a pty is reset on first open so that
  *	settings don't persist across reuse.
  *
- *	Locking: tty_mutex protects tty, get_tty_driver and init_dev work.
+ *	Locking: tty_mutex protects tty, get_tty_driver and tty_init_dev work.
  *		 tty->count should protect the rest.
  *		 ->siglock protects ->signal/->sighand
  */
 
 static int __tty_open(struct inode *inode, struct file *filp)
 {
-	struct tty_struct *tty;
+	struct tty_struct *tty = NULL;
 	int noctty, retval;
 	struct tty_driver *driver;
 	int index;
@@ -2193,23 +1777,25 @@ retry_open:
 			mutex_unlock(&tty_mutex);
 			return -ENXIO;
 		}
-		driver = tty->driver;
+		driver = tty_driver_kref_get(tty->driver);
 		index = tty->index;
 		filp->f_flags |= O_NONBLOCK; /* Don't let /dev/tty block */
 		/* noctty = 1; */
+		/* FIXME: Should we take a driver reference ? */
+		tty_kref_put(tty);
 		goto got_driver;
 	}
 #ifdef CONFIG_VT
 	if (device == MKDEV(TTY_MAJOR, 0)) {
 		extern struct tty_driver *console_driver;
-		driver = console_driver;
+		driver = tty_driver_kref_get(console_driver);
 		index = fg_console;
 		noctty = 1;
 		goto got_driver;
 	}
 #endif
 	if (device == MKDEV(TTYAUX_MAJOR, 1)) {
-		driver = console_device(&index);
+		driver = tty_driver_kref_get(console_device(&index));
 		if (driver) {
 			/* Don't let /dev/console block */
 			filp->f_flags |= O_NONBLOCK;
@@ -2226,10 +1812,25 @@ retry_open:
 		return -ENODEV;
 	}
 got_driver:
-	retval = init_dev(driver, index, &tty);
+	if (!tty) {
+		/* check whether we're reopening an existing tty */
+		tty = tty_driver_lookup_tty(driver, inode, index);
+
+		if (IS_ERR(tty))
+			return PTR_ERR(tty);
+	}
+
+	if (tty) {
+		retval = tty_reopen(tty);
+		if (retval)
+			tty = ERR_PTR(retval);
+	} else
+		tty = tty_init_dev(driver, index, 0);
+
 	mutex_unlock(&tty_mutex);
-	if (retval)
-		return retval;
+	tty_driver_kref_put(driver);
+	if (IS_ERR(tty))
+		return PTR_ERR(tty);
 
 	filp->private_data = tty;
 	file_move(filp, &tty->tty_files);
@@ -2257,7 +1858,7 @@ got_driver:
 		printk(KERN_DEBUG "error %d in opening %s...", retval,
 		       tty->name);
 #endif
-		release_dev(filp);
+		tty_release_dev(filp);
 		if (retval != -ERESTARTSYS)
 			return retval;
 		if (signal_pending(current))
@@ -2296,69 +1897,6 @@ static int tty_open(struct inode *inode, struct file *filp)
 
 
 
-#ifdef CONFIG_UNIX98_PTYS
-/**
- *	ptmx_open		-	open a unix 98 pty master
- *	@inode: inode of device file
- *	@filp: file pointer to tty
- *
- *	Allocate a unix98 pty master device from the ptmx driver.
- *
- *	Locking: tty_mutex protects theinit_dev work. tty->count should
- * 		protect the rest.
- *		allocated_ptys_lock handles the list of free pty numbers
- */
-
-static int __ptmx_open(struct inode *inode, struct file *filp)
-{
-	struct tty_struct *tty;
-	int retval;
-	int index;
-
-	nonseekable_open(inode, filp);
-
-	/* find a device that is not in use. */
-	index = devpts_new_index();
-	if (index < 0)
-		return index;
-
-	mutex_lock(&tty_mutex);
-	retval = init_dev(ptm_driver, index, &tty);
-	mutex_unlock(&tty_mutex);
-
-	if (retval)
-		goto out;
-
-	set_bit(TTY_PTY_LOCK, &tty->flags); /* LOCK THE SLAVE */
-	filp->private_data = tty;
-	file_move(filp, &tty->tty_files);
-
-	retval = devpts_pty_new(tty->link);
-	if (retval)
-		goto out1;
-
-	check_tty_count(tty, "ptmx_open");
-	retval = ptm_driver->ops->open(tty, filp);
-	if (!retval)
-		return 0;
-out1:
-	release_dev(filp);
-	return retval;
-out:
-	devpts_kill_index(index);
-	return retval;
-}
-
-static int ptmx_open(struct inode *inode, struct file *filp)
-{
-	int ret;
-
-	lock_kernel();
-	ret = __ptmx_open(inode, filp);
-	unlock_kernel();
-	return ret;
-}
-#endif
 
 /**
  *	tty_release		-	vfs callback for close
@@ -2369,13 +1907,13 @@ static int ptmx_open(struct inode *inode, struct file *filp)
  *	this tty. There may however be several such references.
  *
  *	Locking:
- *		Takes bkl. See release_dev
+ *		Takes bkl. See tty_release_dev
  */
 
 static int tty_release(struct inode *inode, struct file *filp)
 {
 	lock_kernel();
-	release_dev(filp);
+	tty_release_dev(filp);
 	unlock_kernel();
 	return 0;
 }
@@ -2524,7 +2062,7 @@ int tty_do_resize(struct tty_struct *tty, struct tty_struct *real_tty,
 
 	/* For a PTY we need to lock the tty side */
 	mutex_lock(&real_tty->termios_mutex);
-	if (!memcmp(ws, &tty->winsize, sizeof(*ws)))
+	if (!memcmp(ws, &real_tty->winsize, sizeof(*ws)))
 		goto done;
 	/* Get the PID values and reference them so we can
 	   avoid holding the tty ctrl lock while sending signals */
@@ -2996,7 +2534,7 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case TIOCSTI:
 		return tiocsti(tty, p);
 	case TIOCGWINSZ:
-		return tiocgwinsz(tty, p);
+		return tiocgwinsz(real_tty, p);
 	case TIOCSWINSZ:
 		return tiocswinsz(tty, real_tty, p);
 	case TIOCCONS:
@@ -3026,10 +2564,6 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 		return put_user(tty->ldisc.ops->num, (int __user *)p);
 	case TIOCSETD:
 		return tiocsetd(tty, p);
-#ifdef CONFIG_VT
-	case TIOCLINUX:
-		return tioclinux(tty, arg);
-#endif
 	/*
 	 * Break handling
 	 */
@@ -3220,113 +2754,6 @@ void do_SAK(struct tty_struct *tty)
 EXPORT_SYMBOL(do_SAK);
 
 /**
- *	flush_to_ldisc
- *	@work: tty structure passed from work queue.
- *
- *	This routine is called out of the software interrupt to flush data
- *	from the buffer chain to the line discipline.
- *
- *	Locking: holds tty->buf.lock to guard buffer list. Drops the lock
- *	while invoking the line discipline receive_buf method. The
- *	receive_buf method is single threaded for each tty instance.
- */
-
-static void flush_to_ldisc(struct work_struct *work)
-{
-	struct tty_struct *tty =
-		container_of(work, struct tty_struct, buf.work.work);
-	unsigned long 	flags;
-	struct tty_ldisc *disc;
-	struct tty_buffer *tbuf, *head;
-	char *char_buf;
-	unsigned char *flag_buf;
-
-	disc = tty_ldisc_ref(tty);
-	if (disc == NULL)	/*  !TTY_LDISC */
-		return;
-
-	spin_lock_irqsave(&tty->buf.lock, flags);
-	/* So we know a flush is running */
-	set_bit(TTY_FLUSHING, &tty->flags);
-	head = tty->buf.head;
-	if (head != NULL) {
-		tty->buf.head = NULL;
-		for (;;) {
-			int count = head->commit - head->read;
-			if (!count) {
-				if (head->next == NULL)
-					break;
-				tbuf = head;
-				head = head->next;
-				tty_buffer_free(tty, tbuf);
-				continue;
-			}
-			/* Ldisc or user is trying to flush the buffers
-			   we are feeding to the ldisc, stop feeding the
-			   line discipline as we want to empty the queue */
-			if (test_bit(TTY_FLUSHPENDING, &tty->flags))
-				break;
-			if (!tty->receive_room) {
-				schedule_delayed_work(&tty->buf.work, 1);
-				break;
-			}
-			if (count > tty->receive_room)
-				count = tty->receive_room;
-			char_buf = head->char_buf_ptr + head->read;
-			flag_buf = head->flag_buf_ptr + head->read;
-			head->read += count;
-			spin_unlock_irqrestore(&tty->buf.lock, flags);
-			disc->ops->receive_buf(tty, char_buf,
-							flag_buf, count);
-			spin_lock_irqsave(&tty->buf.lock, flags);
-		}
-		/* Restore the queue head */
-		tty->buf.head = head;
-	}
-	/* We may have a deferred request to flush the input buffer,
-	   if so pull the chain under the lock and empty the queue */
-	if (test_bit(TTY_FLUSHPENDING, &tty->flags)) {
-		__tty_buffer_flush(tty);
-		clear_bit(TTY_FLUSHPENDING, &tty->flags);
-		wake_up(&tty->read_wait);
-	}
-	clear_bit(TTY_FLUSHING, &tty->flags);
-	spin_unlock_irqrestore(&tty->buf.lock, flags);
-
-	tty_ldisc_deref(disc);
-}
-
-/**
- *	tty_flip_buffer_push	-	terminal
- *	@tty: tty to push
- *
- *	Queue a push of the terminal flip buffers to the line discipline. This
- *	function must not be called from IRQ context if tty->low_latency is set.
- *
- *	In the event of the queue being busy for flipping the work will be
- *	held off and retried later.
- *
- *	Locking: tty buffer lock. Driver locks in low latency mode.
- */
-
-void tty_flip_buffer_push(struct tty_struct *tty)
-{
-	unsigned long flags;
-	spin_lock_irqsave(&tty->buf.lock, flags);
-	if (tty->buf.tail != NULL)
-		tty->buf.tail->commit = tty->buf.tail->used;
-	spin_unlock_irqrestore(&tty->buf.lock, flags);
-
-	if (tty->low_latency)
-		flush_to_ldisc(&tty->buf.work.work);
-	else
-		schedule_delayed_work(&tty->buf.work, 1);
-}
-
-EXPORT_SYMBOL(tty_flip_buffer_push);
-
-
-/**
  *	initialize_tty_struct
  *	@tty: tty to initialize
  *
@@ -3336,9 +2763,11 @@ EXPORT_SYMBOL(tty_flip_buffer_push);
  *	Locking: none - tty in question must not be exposed at this point
  */
 
-static void initialize_tty_struct(struct tty_struct *tty)
+void initialize_tty_struct(struct tty_struct *tty,
+		struct tty_driver *driver, int idx)
 {
 	memset(tty, 0, sizeof(struct tty_struct));
+	kref_init(&tty->kref);
 	tty->magic = TTY_MAGIC;
 	tty_ldisc_init(tty);
 	tty->session = NULL;
@@ -3346,7 +2775,6 @@ static void initialize_tty_struct(struct tty_struct *tty)
 	tty->overrun_time = jiffies;
 	tty->buf.head = tty->buf.tail = NULL;
 	tty_buffer_init(tty);
-	INIT_DELAYED_WORK(&tty->buf.work, flush_to_ldisc);
 	mutex_init(&tty->termios_mutex);
 	init_waitqueue_head(&tty->write_wait);
 	init_waitqueue_head(&tty->read_wait);
@@ -3357,6 +2785,11 @@ static void initialize_tty_struct(struct tty_struct *tty)
 	spin_lock_init(&tty->ctrl_lock);
 	INIT_LIST_HEAD(&tty->tty_files);
 	INIT_WORK(&tty->SAK_work, do_SAK_work);
+
+	tty->driver = driver;
+	tty->ops = driver->ops;
+	tty->index = idx;
+	tty_line_name(driver, idx, tty->name);
 }
 
 /**
@@ -3377,10 +2810,9 @@ int tty_put_char(struct tty_struct *tty, unsigned char ch)
 		return tty->ops->put_char(tty, ch);
 	return tty->ops->write(tty, &ch, 1);
 }
-
 EXPORT_SYMBOL_GPL(tty_put_char);
 
-static struct class *tty_class;
+struct class *tty_class;
 
 /**
  *	tty_register_device - register a tty device
@@ -3420,6 +2852,7 @@ struct device *tty_register_device(struct tty_driver *driver, unsigned index,
 
 	return device_create_drvdata(tty_class, device, dev, NULL, name);
 }
+EXPORT_SYMBOL(tty_register_device);
 
 /**
  * 	tty_unregister_device - unregister a tty device
@@ -3437,8 +2870,6 @@ void tty_unregister_device(struct tty_driver *driver, unsigned index)
 	device_destroy(tty_class,
 		MKDEV(driver->major, driver->minor_start) + index);
 }
-
-EXPORT_SYMBOL(tty_register_device);
 EXPORT_SYMBOL(tty_unregister_device);
 
 struct tty_driver *alloc_tty_driver(int lines)
@@ -3447,27 +2878,65 @@ struct tty_driver *alloc_tty_driver(int lines)
 
 	driver = kzalloc(sizeof(struct tty_driver), GFP_KERNEL);
 	if (driver) {
+		kref_init(&driver->kref);
 		driver->magic = TTY_DRIVER_MAGIC;
 		driver->num = lines;
 		/* later we'll move allocation of tables here */
 	}
 	return driver;
 }
+EXPORT_SYMBOL(alloc_tty_driver);
 
-void put_tty_driver(struct tty_driver *driver)
+static void destruct_tty_driver(struct kref *kref)
 {
+	struct tty_driver *driver = container_of(kref, struct tty_driver, kref);
+	int i;
+	struct ktermios *tp;
+	void *p;
+
+	if (driver->flags & TTY_DRIVER_INSTALLED) {
+		/*
+		 * Free the termios and termios_locked structures because
+		 * we don't want to get memory leaks when modular tty
+		 * drivers are removed from the kernel.
+		 */
+		for (i = 0; i < driver->num; i++) {
+			tp = driver->termios[i];
+			if (tp) {
+				driver->termios[i] = NULL;
+				kfree(tp);
+			}
+			if (!(driver->flags & TTY_DRIVER_DYNAMIC_DEV))
+				tty_unregister_device(driver, i);
+		}
+		p = driver->ttys;
+		proc_tty_unregister_driver(driver);
+		driver->ttys = NULL;
+		driver->termios = NULL;
+		kfree(p);
+		cdev_del(&driver->cdev);
+	}
 	kfree(driver);
 }
 
+void tty_driver_kref_put(struct tty_driver *driver)
+{
+	kref_put(&driver->kref, destruct_tty_driver);
+}
+EXPORT_SYMBOL(tty_driver_kref_put);
+
 void tty_set_operations(struct tty_driver *driver,
 			const struct tty_operations *op)
 {
 	driver->ops = op;
 };
+EXPORT_SYMBOL(tty_set_operations);
 
-EXPORT_SYMBOL(alloc_tty_driver);
+void put_tty_driver(struct tty_driver *d)
+{
+	tty_driver_kref_put(d);
+}
 EXPORT_SYMBOL(put_tty_driver);
-EXPORT_SYMBOL(tty_set_operations);
 
 /*
  * Called by a tty driver to register itself.
@@ -3479,11 +2948,8 @@ int tty_register_driver(struct tty_driver *driver)
 	dev_t dev;
 	void **p = NULL;
 
-	if (driver->flags & TTY_DRIVER_INSTALLED)
-		return 0;
-
 	if (!(driver->flags & TTY_DRIVER_DEVPTS_MEM) && driver->num) {
-		p = kzalloc(driver->num * 3 * sizeof(void *), GFP_KERNEL);
+		p = kzalloc(driver->num * 2 * sizeof(void *), GFP_KERNEL);
 		if (!p)
 			return -ENOMEM;
 	}
@@ -3507,12 +2973,9 @@ int tty_register_driver(struct tty_driver *driver)
 	if (p) {
 		driver->ttys = (struct tty_struct **)p;
 		driver->termios = (struct ktermios **)(p + driver->num);
-		driver->termios_locked = (struct ktermios **)
-							(p + driver->num * 2);
 	} else {
 		driver->ttys = NULL;
 		driver->termios = NULL;
-		driver->termios_locked = NULL;
 	}
 
 	cdev_init(&driver->cdev, &tty_fops);
@@ -3521,7 +2984,7 @@ int tty_register_driver(struct tty_driver *driver)
 	if (error) {
 		unregister_chrdev_region(dev, driver->num);
 		driver->ttys = NULL;
-		driver->termios = driver->termios_locked = NULL;
+		driver->termios = NULL;
 		kfree(p);
 		return error;
 	}
@@ -3535,6 +2998,7 @@ int tty_register_driver(struct tty_driver *driver)
 		    tty_register_device(driver, i, NULL);
 	}
 	proc_tty_register_driver(driver);
+	driver->flags |= TTY_DRIVER_INSTALLED;
 	return 0;
 }
 
@@ -3545,46 +3009,19 @@ EXPORT_SYMBOL(tty_register_driver);
  */
 int tty_unregister_driver(struct tty_driver *driver)
 {
-	int i;
-	struct ktermios *tp;
-	void *p;
-
+#if 0
+	/* FIXME */
 	if (driver->refcount)
 		return -EBUSY;
-
+#endif
 	unregister_chrdev_region(MKDEV(driver->major, driver->minor_start),
 				driver->num);
 	mutex_lock(&tty_mutex);
 	list_del(&driver->tty_drivers);
 	mutex_unlock(&tty_mutex);
-
-	/*
-	 * Free the termios and termios_locked structures because
-	 * we don't want to get memory leaks when modular tty
-	 * drivers are removed from the kernel.
-	 */
-	for (i = 0; i < driver->num; i++) {
-		tp = driver->termios[i];
-		if (tp) {
-			driver->termios[i] = NULL;
-			kfree(tp);
-		}
-		tp = driver->termios_locked[i];
-		if (tp) {
-			driver->termios_locked[i] = NULL;
-			kfree(tp);
-		}
-		if (!(driver->flags & TTY_DRIVER_DYNAMIC_DEV))
-			tty_unregister_device(driver, i);
-	}
-	p = driver->ttys;
-	proc_tty_unregister_driver(driver);
-	driver->ttys = NULL;
-	driver->termios = driver->termios_locked = NULL;
-	kfree(p);
-	cdev_del(&driver->cdev);
 	return 0;
 }
+
 EXPORT_SYMBOL(tty_unregister_driver);
 
 dev_t tty_devnum(struct tty_struct *tty)
@@ -3595,9 +3032,12 @@ EXPORT_SYMBOL(tty_devnum);
 
 void proc_clear_tty(struct task_struct *p)
 {
+	struct tty_struct *tty;
 	spin_lock_irq(&p->sighand->siglock);
+	tty = p->signal->tty;
 	p->signal->tty = NULL;
 	spin_unlock_irq(&p->sighand->siglock);
+	tty_kref_put(tty);
 }
 
 /* Called under the sighand lock */
@@ -3613,9 +3053,13 @@ static void __proc_set_tty(struct task_struct *tsk, struct tty_struct *tty)
 		tty->pgrp = get_pid(task_pgrp(tsk));
 		spin_unlock_irqrestore(&tty->ctrl_lock, flags);
 		tty->session = get_pid(task_session(tsk));
+		if (tsk->signal->tty) {
+			printk(KERN_DEBUG "tty not NULL!!\n");
+			tty_kref_put(tsk->signal->tty);
+		}
 	}
 	put_pid(tsk->signal->tty_old_pgrp);
-	tsk->signal->tty = tty;
+	tsk->signal->tty = tty_kref_get(tty);
 	tsk->signal->tty_old_pgrp = NULL;
 }
 
@@ -3629,18 +3073,20 @@ static void proc_set_tty(struct task_struct *tsk, struct tty_struct *tty)
 struct tty_struct *get_current_tty(void)
 {
 	struct tty_struct *tty;
-	WARN_ON_ONCE(!mutex_is_locked(&tty_mutex));
-	tty = current->signal->tty;
-	/*
-	 * session->tty can be changed/cleared from under us, make sure we
-	 * issue the load. The obtained pointer, when not NULL, is valid as
-	 * long as we hold tty_mutex.
-	 */
-	barrier();
+	unsigned long flags;
+
+	spin_lock_irqsave(&current->sighand->siglock, flags);
+	tty = tty_kref_get(current->signal->tty);
+	spin_unlock_irqrestore(&current->sighand->siglock, flags);
 	return tty;
 }
 EXPORT_SYMBOL_GPL(get_current_tty);
 
+void tty_default_fops(struct file_operations *fops)
+{
+	*fops = tty_fops;
+}
+
 /*
  * Initialize the console device. This is called *early*, so
  * we can't necessarily depend on lots of kernel help here.
@@ -3678,12 +3124,6 @@ postcore_initcall(tty_class_init);
 /* 3/2004 jmc: why do these devices exist? */
 
 static struct cdev tty_cdev, console_cdev;
-#ifdef CONFIG_UNIX98_PTYS
-static struct cdev ptmx_cdev;
-#endif
-#ifdef CONFIG_VT
-static struct cdev vc0_cdev;
-#endif
 
 /*
  * Ok, now we can initialize the rest of the tty devices and can count
@@ -3695,32 +3135,18 @@ static int __init tty_init(void)
 	if (cdev_add(&tty_cdev, MKDEV(TTYAUX_MAJOR, 0), 1) ||
 	    register_chrdev_region(MKDEV(TTYAUX_MAJOR, 0), 1, "/dev/tty") < 0)
 		panic("Couldn't register /dev/tty driver\n");
-	device_create_drvdata(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 0), NULL,
+	device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 0), NULL,
 			      "tty");
 
 	cdev_init(&console_cdev, &console_fops);
 	if (cdev_add(&console_cdev, MKDEV(TTYAUX_MAJOR, 1), 1) ||
 	    register_chrdev_region(MKDEV(TTYAUX_MAJOR, 1), 1, "/dev/console") < 0)
 		panic("Couldn't register /dev/console driver\n");
-	device_create_drvdata(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 1), NULL,
+	device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 1), NULL,
 			      "console");
 
-#ifdef CONFIG_UNIX98_PTYS
-	cdev_init(&ptmx_cdev, &ptmx_fops);
-	if (cdev_add(&ptmx_cdev, MKDEV(TTYAUX_MAJOR, 2), 1) ||
-	    register_chrdev_region(MKDEV(TTYAUX_MAJOR, 2), 1, "/dev/ptmx") < 0)
-		panic("Couldn't register /dev/ptmx driver\n");
-	device_create_drvdata(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 2), NULL, "ptmx");
-#endif
-
 #ifdef CONFIG_VT
-	cdev_init(&vc0_cdev, &console_fops);
-	if (cdev_add(&vc0_cdev, MKDEV(TTY_MAJOR, 0), 1) ||
-	    register_chrdev_region(MKDEV(TTY_MAJOR, 0), 1, "/dev/vc/0") < 0)
-		panic("Couldn't register /dev/tty0 driver\n");
-	device_create_drvdata(tty_class, NULL, MKDEV(TTY_MAJOR, 0), NULL, "tty0");
-
-	vty_init();
+	vty_init(&console_fops);
 #endif
 	return 0;
 }
diff --git a/drivers/char/tty_ioctl.c b/drivers/char/tty_ioctl.c
index bf34e4597421..a408c8e487ec 100644
--- a/drivers/char/tty_ioctl.c
+++ b/drivers/char/tty_ioctl.c
@@ -40,6 +40,15 @@
 #define TERMIOS_OLD	8
 
 
+/**
+ *	tty_chars_in_buffer	-	characters pending
+ *	@tty: terminal
+ *
+ *	Return the number of bytes of data in the device private
+ *	output queue. If no private method is supplied there is assumed
+ *	to be no queue on the device.
+ */
+
 int tty_chars_in_buffer(struct tty_struct *tty)
 {
 	if (tty->ops->chars_in_buffer)
@@ -47,26 +56,49 @@ int tty_chars_in_buffer(struct tty_struct *tty)
 	else
 		return 0;
 }
-
 EXPORT_SYMBOL(tty_chars_in_buffer);
 
+/**
+ *	tty_write_room		-	write queue space
+ *	@tty: terminal
+ *
+ *	Return the number of bytes that can be queued to this device
+ *	at the present time. The result should be treated as a guarantee
+ *	and the driver cannot offer a value it later shrinks by more than
+ *	the number of bytes written. If no method is provided 2K is always
+ *	returned and data may be lost as there will be no flow control.
+ */
+ 
 int tty_write_room(struct tty_struct *tty)
 {
 	if (tty->ops->write_room)
 		return tty->ops->write_room(tty);
 	return 2048;
 }
-
 EXPORT_SYMBOL(tty_write_room);
 
+/**
+ *	tty_driver_flush_buffer	-	discard internal buffer
+ *	@tty: terminal
+ *
+ *	Discard the internal output buffer for this device. If no method
+ *	is provided then either the buffer cannot be hardware flushed or
+ *	there is no buffer driver side.
+ */
 void tty_driver_flush_buffer(struct tty_struct *tty)
 {
 	if (tty->ops->flush_buffer)
 		tty->ops->flush_buffer(tty);
 }
-
 EXPORT_SYMBOL(tty_driver_flush_buffer);
 
+/**
+ *	tty_throttle		-	flow control
+ *	@tty: terminal
+ *
+ *	Indicate that a tty should stop transmitting data down the stack.
+ */
+
 void tty_throttle(struct tty_struct *tty)
 {
 	/* check TTY_THROTTLED first so it indicates our state */
@@ -76,6 +108,13 @@ void tty_throttle(struct tty_struct *tty)
 }
 EXPORT_SYMBOL(tty_throttle);
 
+/**
+ *	tty_unthrottle		-	flow control
+ *	@tty: terminal
+ *
+ *	Indicate that a tty may continue transmitting data down the stack.
+ */
+
 void tty_unthrottle(struct tty_struct *tty)
 {
 	if (test_and_clear_bit(TTY_THROTTLED, &tty->flags) &&
@@ -112,6 +151,11 @@ void tty_wait_until_sent(struct tty_struct *tty, long timeout)
 }
 EXPORT_SYMBOL(tty_wait_until_sent);
 
+
+/*
+ *		Termios Helper Methods
+ */
+
 static void unset_locked_termios(struct ktermios *termios,
 				 struct ktermios *old,
 				 struct ktermios *locked)
@@ -346,6 +390,16 @@ void tty_termios_encode_baud_rate(struct ktermios *termios,
 }
 EXPORT_SYMBOL_GPL(tty_termios_encode_baud_rate);
 
+/**
+ *	tty_encode_baud_rate		-	set baud rate of the tty
+ *	@ibaud: input baud rate
+ *	@obad: output baud rate
+ *
+ *	Update the current termios data for the tty with the new speed
+ *	settings. The caller must hold the termios_mutex for the tty in
+ *	question.
+ */
+
 void tty_encode_baud_rate(struct tty_struct *tty, speed_t ibaud, speed_t obaud)
 {
 	tty_termios_encode_baud_rate(tty->termios, ibaud, obaud);
@@ -430,12 +484,11 @@ EXPORT_SYMBOL(tty_termios_hw_change);
  *	is a bit of layering violation here with n_tty in terms of the
  *	internal knowledge of this function.
  *
- *	Locking: termios_sem
+ *	Locking: termios_mutex
  */
 
 static void change_termios(struct tty_struct *tty, struct ktermios *new_termios)
 {
-	int canon_change;
 	struct ktermios old_termios;
 	struct tty_ldisc *ld;
 	unsigned long flags;
@@ -451,18 +504,6 @@ static void change_termios(struct tty_struct *tty, struct ktermios *new_termios)
 	old_termios = *tty->termios;
 	*tty->termios = *new_termios;
 	unset_locked_termios(tty->termios, &old_termios, tty->termios_locked);
-	canon_change = (old_termios.c_lflag ^ tty->termios->c_lflag) & ICANON;
-	if (canon_change) {
-		memset(&tty->read_flags, 0, sizeof tty->read_flags);
-		tty->canon_head = tty->read_tail;
-		tty->canon_data = 0;
-		tty->erasing = 0;
-	}
-
-	/* This bit should be in the ldisc code */
-	if (canon_change && !L_ICANON(tty) && tty->read_cnt)
-		/* Get characters left over from canonical mode. */
-		wake_up_interruptible(&tty->read_wait);
 
 	/* See if packet mode change of state. */
 	if (tty->link && tty->link->packet) {
@@ -508,7 +549,7 @@ static void change_termios(struct tty_struct *tty, struct ktermios *new_termios)
  *	functions before using change_termios to do the actual changes.
  *
  *	Locking:
- *		Called functions take ldisc and termios_sem locks
+ *		Called functions take ldisc and termios_mutex locks
  */
 
 static int set_termios(struct tty_struct *tty, void __user *arg, int opt)
@@ -579,25 +620,51 @@ static int get_termio(struct tty_struct *tty, struct termio __user *termio)
 	return 0;
 }
 
-static unsigned long inq_canon(struct tty_struct *tty)
+
+#ifdef TCGETX
+
+/**
+ *	set_termiox	-	set termiox fields if possible
+ *	@tty: terminal
+ *	@arg: termiox structure from user
+ *	@opt: option flags for ioctl type
+ *
+ *	Implement the device calling points for the SYS5 termiox ioctl
+ *	interface in Linux
+ */
+
+static int set_termiox(struct tty_struct *tty, void __user *arg, int opt)
 {
-	int nr, head, tail;
+	struct termiox tnew;
+	struct tty_ldisc *ld;
 
-	if (!tty->canon_data || !tty->read_buf)
-		return 0;
-	head = tty->canon_head;
-	tail = tty->read_tail;
-	nr = (head - tail) & (N_TTY_BUF_SIZE-1);
-	/* Skip EOF-chars.. */
-	while (head != tail) {
-		if (test_bit(tail, tty->read_flags) &&
-		    tty->read_buf[tail] == __DISABLED_CHAR)
-			nr--;
-		tail = (tail+1) & (N_TTY_BUF_SIZE-1);
+	if (tty->termiox == NULL)
+		return -EINVAL;
+	if (copy_from_user(&tnew, arg, sizeof(struct termiox)))
+		return -EFAULT;
+
+	ld = tty_ldisc_ref(tty);
+	if (ld != NULL) {
+		if ((opt & TERMIOS_FLUSH) && ld->ops->flush_buffer)
+			ld->ops->flush_buffer(tty);
+		tty_ldisc_deref(ld);
 	}
-	return nr;
+	if (opt & TERMIOS_WAIT) {
+		tty_wait_until_sent(tty, 0);
+		if (signal_pending(current))
+			return -EINTR;
+	}
+
+	mutex_lock(&tty->termios_mutex);
+	if (tty->ops->set_termiox)
+		tty->ops->set_termiox(tty, &tnew);
+	mutex_unlock(&tty->termios_mutex);
+	return 0;
 }
 
+#endif
+
+
 #ifdef TIOCGETP
 /*
  * These are deprecated, but there is limited support..
@@ -671,7 +738,7 @@ static void set_sgflags(struct ktermios *termios, int flags)
  *	Updates a terminal from the legacy BSD style terminal information
  *	structure.
  *
- *	Locking: termios_sem
+ *	Locking: termios_mutex
  */
 
 static int set_sgttyb(struct tty_struct *tty, struct sgttyb __user *sgttyb)
@@ -849,6 +916,7 @@ int tty_mode_ioctl(struct tty_struct *tty, struct file *file,
 {
 	struct tty_struct *real_tty;
 	void __user *p = (void __user *)arg;
+	int ret = 0;
 
 	if (tty->driver->type == TTY_DRIVER_TYPE_PTY &&
 	    tty->driver->subtype == PTY_TYPE_MASTER)
@@ -884,18 +952,24 @@ int tty_mode_ioctl(struct tty_struct *tty, struct file *file,
 		return set_termios(real_tty, p, TERMIOS_OLD);
 #ifndef TCGETS2
 	case TCGETS:
+		mutex_lock(&real_tty->termios_mutex);
 		if (kernel_termios_to_user_termios((struct termios __user *)arg, real_tty->termios))
-			return -EFAULT;
-		return 0;
+			ret = -EFAULT;
+		mutex_unlock(&real_tty->termios_mutex);
+		return ret;
 #else
 	case TCGETS:
+		mutex_lock(&real_tty->termios_mutex);
 		if (kernel_termios_to_user_termios_1((struct termios __user *)arg, real_tty->termios))
-			return -EFAULT;
-		return 0;
+			ret = -EFAULT;
+		mutex_unlock(&real_tty->termios_mutex);
+		return ret;
 	case TCGETS2:
+		mutex_lock(&real_tty->termios_mutex);
 		if (kernel_termios_to_user_termios((struct termios2 __user *)arg, real_tty->termios))
-			return -EFAULT;
-		return 0;
+			ret = -EFAULT;
+		mutex_unlock(&real_tty->termios_mutex);
+		return ret;
 	case TCSETSF2:
 		return set_termios(real_tty, p,  TERMIOS_FLUSH | TERMIOS_WAIT);
 	case TCSETSW2:
@@ -913,34 +987,59 @@ int tty_mode_ioctl(struct tty_struct *tty, struct file *file,
 		return set_termios(real_tty, p, TERMIOS_TERMIO);
 #ifndef TCGETS2
 	case TIOCGLCKTRMIOS:
+		mutex_lock(&real_tty->termios_mutex);
 		if (kernel_termios_to_user_termios((struct termios __user *)arg, real_tty->termios_locked))
-			return -EFAULT;
-		return 0;
+			ret = -EFAULT;
+		mutex_unlock(&real_tty->termios_mutex);
+		return ret;
 	case TIOCSLCKTRMIOS:
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
+		mutex_lock(&real_tty->termios_mutex);
 		if (user_termios_to_kernel_termios(real_tty->termios_locked,
 					       (struct termios __user *) arg))
-			return -EFAULT;
-		return 0;
+			ret = -EFAULT;
+		mutex_unlock(&real_tty->termios_mutex);
+		return ret;
 #else
 	case TIOCGLCKTRMIOS:
+		mutex_lock(&real_tty->termios_mutex);
 		if (kernel_termios_to_user_termios_1((struct termios __user *)arg, real_tty->termios_locked))
-			return -EFAULT;
-		return 0;
+			ret = -EFAULT;
+		mutex_unlock(&real_tty->termios_mutex);
+		return ret;
 	case TIOCSLCKTRMIOS:
 		if (!capable(CAP_SYS_ADMIN))
-			return -EPERM;
+			ret = -EPERM;
+		mutex_lock(&real_tty->termios_mutex);
 		if (user_termios_to_kernel_termios_1(real_tty->termios_locked,
 					       (struct termios __user *) arg))
-			return -EFAULT;
-			return 0;
+			ret = -EFAULT;
+		mutex_unlock(&real_tty->termios_mutex);
+		return ret;
 #endif
+#ifdef TCGETX
+	case TCGETX:
+		if (real_tty->termiox == NULL)
+			return -EINVAL;
+		mutex_lock(&real_tty->termios_mutex);
+		if (copy_to_user(p, real_tty->termiox, sizeof(struct termiox)))
+			ret = -EFAULT;
+		mutex_unlock(&real_tty->termios_mutex);
+		return ret;
+	case TCSETX:
+		return set_termiox(real_tty, p, 0);
+	case TCSETXW:
+		return set_termiox(real_tty, p, TERMIOS_WAIT);
+	case TCSETXF:
+		return set_termiox(real_tty, p, TERMIOS_FLUSH);
+#endif		
 	case TIOCGSOFTCAR:
-		/* FIXME: for correctness we may need to take the termios
-		   lock here - review */
-		return put_user(C_CLOCAL(real_tty) ? 1 : 0,
+		mutex_lock(&real_tty->termios_mutex);
+		ret = put_user(C_CLOCAL(real_tty) ? 1 : 0,
 						(int __user *)arg);
+		mutex_unlock(&real_tty->termios_mutex);
+		return ret;
 	case TIOCSSOFTCAR:
 		if (get_user(arg, (unsigned int __user *) arg))
 			return -EFAULT;
@@ -980,7 +1079,7 @@ int tty_perform_flush(struct tty_struct *tty, unsigned long arg)
 }
 EXPORT_SYMBOL_GPL(tty_perform_flush);
 
-int n_tty_ioctl(struct tty_struct *tty, struct file *file,
+int n_tty_ioctl_helper(struct tty_struct *tty, struct file *file,
 		       unsigned int cmd, unsigned long arg)
 {
 	unsigned long flags;
@@ -1018,13 +1117,6 @@ int n_tty_ioctl(struct tty_struct *tty, struct file *file,
 		return 0;
 	case TCFLSH:
 		return tty_perform_flush(tty, arg);
-	case TIOCOUTQ:
-		return put_user(tty_chars_in_buffer(tty), (int __user *) arg);
-	case TIOCINQ:
-		retval = tty->read_cnt;
-		if (L_ICANON(tty))
-			retval = inq_canon(tty);
-		return put_user(retval, (unsigned int __user *) arg);
 	case TIOCPKT:
 	{
 		int pktmode;
@@ -1050,4 +1142,4 @@ int n_tty_ioctl(struct tty_struct *tty, struct file *file,
 		return tty_mode_ioctl(tty, file, cmd, arg);
 	}
 }
-EXPORT_SYMBOL(n_tty_ioctl);
+EXPORT_SYMBOL(n_tty_ioctl_helper);
diff --git a/drivers/char/tty_port.c b/drivers/char/tty_port.c
new file mode 100644
index 000000000000..553b0e9d8d17
--- /dev/null
+++ b/drivers/char/tty_port.c
@@ -0,0 +1,96 @@
+/*
+ * Tty port functions
+ */
+
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/tty.h>
+#include <linux/tty_driver.h>
+#include <linux/tty_flip.h>
+#include <linux/timer.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/sched.h>
+#include <linux/init.h>
+#include <linux/wait.h>
+#include <linux/bitops.h>
+#include <linux/delay.h>
+#include <linux/module.h>
+
+void tty_port_init(struct tty_port *port)
+{
+	memset(port, 0, sizeof(*port));
+	init_waitqueue_head(&port->open_wait);
+	init_waitqueue_head(&port->close_wait);
+	mutex_init(&port->mutex);
+	spin_lock_init(&port->lock);
+	port->close_delay = (50 * HZ) / 100;
+	port->closing_wait = (3000 * HZ) / 100;
+}
+EXPORT_SYMBOL(tty_port_init);
+
+int tty_port_alloc_xmit_buf(struct tty_port *port)
+{
+	/* We may sleep in get_zeroed_page() */
+	mutex_lock(&port->mutex);
+	if (port->xmit_buf == NULL)
+		port->xmit_buf = (unsigned char *)get_zeroed_page(GFP_KERNEL);
+	mutex_unlock(&port->mutex);
+	if (port->xmit_buf == NULL)
+		return -ENOMEM;
+	return 0;
+}
+EXPORT_SYMBOL(tty_port_alloc_xmit_buf);
+
+void tty_port_free_xmit_buf(struct tty_port *port)
+{
+	mutex_lock(&port->mutex);
+	if (port->xmit_buf != NULL) {
+		free_page((unsigned long)port->xmit_buf);
+		port->xmit_buf = NULL;
+	}
+	mutex_unlock(&port->mutex);
+}
+EXPORT_SYMBOL(tty_port_free_xmit_buf);
+
+
+/**
+ *	tty_port_tty_get	-	get a tty reference
+ *	@port: tty port
+ *
+ *	Return a refcount protected tty instance or NULL if the port is not
+ *	associated with a tty (eg due to close or hangup)
+ */
+
+struct tty_struct *tty_port_tty_get(struct tty_port *port)
+{
+	unsigned long flags;
+	struct tty_struct *tty;
+
+	spin_lock_irqsave(&port->lock, flags);
+	tty = tty_kref_get(port->tty);
+	spin_unlock_irqrestore(&port->lock, flags);
+	return tty;
+}
+EXPORT_SYMBOL(tty_port_tty_get);
+
+/**
+ *	tty_port_tty_set	-	set the tty of a port
+ *	@port: tty port
+ *	@tty: the tty
+ *
+ *	Associate the port and tty pair. Manages any internal refcounts.
+ *	Pass NULL to deassociate a port
+ */
+
+void tty_port_tty_set(struct tty_port *port, struct tty_struct *tty)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&port->lock, flags);
+	if (port->tty)
+		tty_kref_put(port->tty);
+	port->tty = tty;
+	spin_unlock_irqrestore(&port->lock, flags);
+}
+EXPORT_SYMBOL(tty_port_tty_set);
diff --git a/drivers/char/vt.c b/drivers/char/vt.c
index 60359c360912..57029fefd64a 100644
--- a/drivers/char/vt.c
+++ b/drivers/char/vt.c
@@ -100,10 +100,10 @@
 #include <linux/font.h>
 #include <linux/bitops.h>
 #include <linux/notifier.h>
-
-#include <asm/io.h>
+#include <linux/device.h>
+#include <linux/io.h>
 #include <asm/system.h>
-#include <asm/uaccess.h>
+#include <linux/uaccess.h>
 
 #define MAX_NR_CON_DRIVER 16
 
@@ -2136,27 +2136,9 @@ static int do_con_write(struct tty_struct *tty, const unsigned char *buf, int co
 	    release_console_sem();
 	    return 0;
 	}
-	release_console_sem();
-
 	orig_buf = buf;
 	orig_count = count;
 
-	/* At this point 'buf' is guaranteed to be a kernel buffer
-	 * and therefore no access to userspace (and therefore sleeping)
-	 * will be needed.  The con_buf_mtx serializes all tty based
-	 * console rendering and vcs write/read operations.  We hold
-	 * the console spinlock during the entire write.
-	 */
-
-	acquire_console_sem();
-
-	vc = tty->driver_data;
-	if (vc == NULL) {
-		printk(KERN_ERR "vt: argh, driver_data _became_ NULL !\n");
-		release_console_sem();
-		goto out;
-	}
-
 	himask = vc->vc_hi_font_mask;
 	charmask = himask ? 0x1ff : 0xff;
 
@@ -2370,8 +2352,6 @@ rescan_last_byte:
 	FLUSH
 	console_conditional_schedule();
 	release_console_sem();
-
-out:
 	notify_update(vc);
 	return n;
 #undef FLUSH
@@ -2583,8 +2563,6 @@ int tioclinux(struct tty_struct *tty, unsigned long arg)
 	int lines;
 	int ret;
 
-	if (tty->driver->type != TTY_DRIVER_TYPE_CONSOLE)
-		return -EINVAL;
 	if (current->signal->tty != tty && !capable(CAP_SYS_ADMIN))
 		return -EPERM;
 	if (get_user(type, p))
@@ -2778,6 +2756,12 @@ static int con_open(struct tty_struct *tty, struct file *filp)
 		ret = vc_allocate(currcons);
 		if (ret == 0) {
 			struct vc_data *vc = vc_cons[currcons].d;
+
+			/* Still being freed */
+			if (vc->vc_tty) {
+				release_console_sem();
+				return -ERESTARTSYS;
+			}
 			tty->driver_data = vc;
 			vc->vc_tty = tty;
 
@@ -2798,34 +2782,20 @@ static int con_open(struct tty_struct *tty, struct file *filp)
 	return ret;
 }
 
-/*
- * We take tty_mutex in here to prevent another thread from coming in via init_dev
- * and taking a ref against the tty while we're in the process of forgetting
- * about it and cleaning things up.
- *
- * This is because vcs_remove_sysfs() can sleep and will drop the BKL.
- */
 static void con_close(struct tty_struct *tty, struct file *filp)
 {
-	mutex_lock(&tty_mutex);
-	acquire_console_sem();
-	if (tty && tty->count == 1) {
-		struct vc_data *vc = tty->driver_data;
+	/* Nothing to do - we defer to shutdown */
+}
 
-		if (vc)
-			vc->vc_tty = NULL;
-		tty->driver_data = NULL;
-		vcs_remove_sysfs(tty);
-		release_console_sem();
-		mutex_unlock(&tty_mutex);
-		/*
-		 * tty_mutex is released, but we still hold BKL, so there is
-		 * still exclusion against init_dev()
-		 */
-		return;
-	}
+static void con_shutdown(struct tty_struct *tty)
+{
+	struct vc_data *vc = tty->driver_data;
+	BUG_ON(vc == NULL);
+	acquire_console_sem();
+	vc->vc_tty = NULL;
+	vcs_remove_sysfs(tty);
 	release_console_sem();
-	mutex_unlock(&tty_mutex);
+	tty_shutdown(tty);
 }
 
 static int default_italic_color    = 2; // green (ASCII)
@@ -2950,10 +2920,19 @@ static const struct tty_operations con_ops = {
 	.throttle = con_throttle,
 	.unthrottle = con_unthrottle,
 	.resize = vt_resize,
+	.shutdown = con_shutdown
 };
 
-int __init vty_init(void)
+static struct cdev vc0_cdev;
+
+int __init vty_init(const struct file_operations *console_fops)
 {
+	cdev_init(&vc0_cdev, console_fops);
+	if (cdev_add(&vc0_cdev, MKDEV(TTY_MAJOR, 0), 1) ||
+	    register_chrdev_region(MKDEV(TTY_MAJOR, 0), 1, "/dev/vc/0") < 0)
+		panic("Couldn't register /dev/tty0 driver\n");
+	device_create(tty_class, NULL, MKDEV(TTY_MAJOR, 0), NULL, "tty0");
+
 	vcs_init();
 
 	console_driver = alloc_tty_driver(MAX_NR_CONSOLES);
@@ -2972,7 +2951,6 @@ int __init vty_init(void)
 	tty_set_operations(console_driver, &con_ops);
 	if (tty_register_driver(console_driver))
 		panic("Couldn't register console driver\n");
-
 	kbd_init();
 	console_map_init();
 #ifdef CONFIG_PROM_CONSOLE
@@ -3466,7 +3444,7 @@ int register_con_driver(const struct consw *csw, int first, int last)
 	if (retval)
 		goto err;
 
-	con_driver->dev = device_create_drvdata(vtconsole_class, NULL,
+	con_driver->dev = device_create(vtconsole_class, NULL,
 						MKDEV(0, con_driver->node),
 						NULL, "vtcon%i",
 						con_driver->node);
@@ -3577,7 +3555,7 @@ static int __init vtconsole_class_init(void)
 		struct con_driver *con = &registered_con_driver[i];
 
 		if (con->con && !con->dev) {
-			con->dev = device_create_drvdata(vtconsole_class, NULL,
+			con->dev = device_create(vtconsole_class, NULL,
 							 MKDEV(0, con->node),
 							 NULL, "vtcon%i",
 							 con->node);
diff --git a/drivers/char/vt_ioctl.c b/drivers/char/vt_ioctl.c
index c904e9ad4a71..8944ce508e2f 100644
--- a/drivers/char/vt_ioctl.c
+++ b/drivers/char/vt_ioctl.c
@@ -395,6 +395,8 @@ int vt_ioctl(struct tty_struct *tty, struct file * file,
  
 	kbd = kbd_table + console;
 	switch (cmd) {
+	case TIOCLINUX:
+		return tioclinux(tty, arg);
 	case KIOCSOUND:
 		if (!perm)
 			goto eperm;
diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig
index 3309e862f317..ebacc0af40fe 100644
--- a/drivers/hwmon/Kconfig
+++ b/drivers/hwmon/Kconfig
@@ -800,6 +800,13 @@ config SENSORS_W83627EHF
 	  This driver can also be built as a module.  If so, the module
 	  will be called w83627ehf.
 
+config SENSORS_ULTRA45
+	tristate "Sun Ultra45 PIC16F747"
+	depends on SPARC64
+	help
+	  This driver provides support for the Ultra45 workstation environmental
+	  sensors.
+
 config SENSORS_HDAPS
 	tristate "IBM Hard Drive Active Protection System (hdaps)"
 	depends on INPUT && X86
diff --git a/drivers/hwmon/Makefile b/drivers/hwmon/Makefile
index 6babc801b348..042d5a78622e 100644
--- a/drivers/hwmon/Makefile
+++ b/drivers/hwmon/Makefile
@@ -41,6 +41,7 @@ obj-$(CONFIG_SENSORS_FSCHMD)	+= fschmd.o
 obj-$(CONFIG_SENSORS_FSCPOS)	+= fscpos.o
 obj-$(CONFIG_SENSORS_GL518SM)	+= gl518sm.o
 obj-$(CONFIG_SENSORS_GL520SM)	+= gl520sm.o
+obj-$(CONFIG_SENSORS_ULTRA45)	+= ultra45_env.o
 obj-$(CONFIG_SENSORS_HDAPS)	+= hdaps.o
 obj-$(CONFIG_SENSORS_I5K_AMB)	+= i5k_amb.o
 obj-$(CONFIG_SENSORS_IBMAEM)	+= ibmaem.o
diff --git a/drivers/hwmon/ultra45_env.c b/drivers/hwmon/ultra45_env.c
new file mode 100644
index 000000000000..68e90abeba96
--- /dev/null
+++ b/drivers/hwmon/ultra45_env.c
@@ -0,0 +1,320 @@
+/* ultra45_env.c: Driver for Ultra45 PIC16F747 environmental monitor.
+ *
+ * Copyright (C) 2008 David S. Miller <davem@davemloft.net>
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/of_device.h>
+#include <linux/io.h>
+#include <linux/hwmon.h>
+#include <linux/hwmon-sysfs.h>
+
+#define DRV_MODULE_VERSION	"0.1"
+
+MODULE_AUTHOR("David S. Miller (davem@davemloft.net)");
+MODULE_DESCRIPTION("Ultra45 environmental monitor driver");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(DRV_MODULE_VERSION);
+
+/* PIC device registers */
+#define REG_CMD		0x00UL
+#define  REG_CMD_RESET	0x80
+#define  REG_CMD_ESTAR	0x01
+#define REG_STAT	0x01UL
+#define  REG_STAT_FWVER	0xf0
+#define  REG_STAT_TGOOD	0x08
+#define  REG_STAT_STALE	0x04
+#define  REG_STAT_BUSY	0x02
+#define  REG_STAT_FAULT	0x01
+#define REG_DATA	0x40UL
+#define REG_ADDR	0x41UL
+#define REG_SIZE	0x42UL
+
+/* Registers accessed indirectly via REG_DATA/REG_ADDR */
+#define IREG_FAN0		0x00
+#define IREG_FAN1		0x01
+#define IREG_FAN2		0x02
+#define IREG_FAN3		0x03
+#define IREG_FAN4		0x04
+#define IREG_FAN5		0x05
+#define IREG_LCL_TEMP		0x06
+#define IREG_RMT1_TEMP		0x07
+#define IREG_RMT2_TEMP		0x08
+#define IREG_RMT3_TEMP		0x09
+#define IREG_LM95221_TEMP	0x0a
+#define IREG_FIRE_TEMP		0x0b
+#define IREG_LSI1064_TEMP	0x0c
+#define IREG_FRONT_TEMP		0x0d
+#define IREG_FAN_STAT		0x0e
+#define IREG_VCORE0		0x0f
+#define IREG_VCORE1		0x10
+#define IREG_VMEM0		0x11
+#define IREG_VMEM1		0x12
+#define IREG_PSU_TEMP		0x13
+
+struct env {
+	void __iomem	*regs;
+	spinlock_t	lock;
+
+	struct device	*hwmon_dev;
+};
+
+static u8 env_read(struct env *p, u8 ireg)
+{
+	u8 ret;
+
+	spin_lock(&p->lock);
+	writeb(ireg, p->regs + REG_ADDR);
+	ret = readb(p->regs + REG_DATA);
+	spin_unlock(&p->lock);
+
+	return ret;
+}
+
+static void env_write(struct env *p, u8 ireg, u8 val)
+{
+	spin_lock(&p->lock);
+	writeb(ireg, p->regs + REG_ADDR);
+	writeb(val, p->regs + REG_DATA);
+	spin_unlock(&p->lock);
+}
+
+/* There seems to be a adr7462 providing these values, thus a lot
+ * of these calculations are borrowed from the adt7470 driver.
+ */
+#define FAN_PERIOD_TO_RPM(x)	((90000 * 60) / (x))
+#define FAN_RPM_TO_PERIOD	FAN_PERIOD_TO_RPM
+#define FAN_PERIOD_INVALID	(0xff << 8)
+#define FAN_DATA_VALID(x)	((x) && (x) != FAN_PERIOD_INVALID)
+
+static ssize_t show_fan_speed(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	int fan_nr = to_sensor_dev_attr(attr)->index;
+	struct env *p = dev_get_drvdata(dev);
+	int rpm, period;
+	u8 val;
+
+	val = env_read(p, IREG_FAN0 + fan_nr);
+	period = (int) val << 8;
+	if (FAN_DATA_VALID(period))
+		rpm = FAN_PERIOD_TO_RPM(period);
+	else
+		rpm = 0;
+
+	return sprintf(buf, "%d\n", rpm);
+}
+
+static ssize_t set_fan_speed(struct device *dev, struct device_attribute *attr,
+			     const char *buf, size_t count)
+{
+	int fan_nr = to_sensor_dev_attr(attr)->index;
+	int rpm = simple_strtol(buf, NULL, 10);
+	struct env *p = dev_get_drvdata(dev);
+	int period;
+	u8 val;
+
+	if (!rpm)
+		return -EINVAL;
+
+	period = FAN_RPM_TO_PERIOD(rpm);
+	val = period >> 8;
+	env_write(p, IREG_FAN0 + fan_nr, val);
+
+	return count;
+}
+
+static ssize_t show_fan_fault(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	int fan_nr = to_sensor_dev_attr(attr)->index;
+	struct env *p = dev_get_drvdata(dev);
+	u8 val = env_read(p, IREG_FAN_STAT);
+	return sprintf(buf, "%d\n", (val & (1 << fan_nr)) ? 1 : 0);
+}
+
+#define fan(index)							\
+static SENSOR_DEVICE_ATTR(fan##index##_speed, S_IRUGO | S_IWUSR,	\
+		show_fan_speed, set_fan_speed, index);			\
+static SENSOR_DEVICE_ATTR(fan##index##_fault, S_IRUGO,			\
+		show_fan_fault, NULL, index)
+
+fan(0);
+fan(1);
+fan(2);
+fan(3);
+fan(4);
+
+static SENSOR_DEVICE_ATTR(psu_fan_fault, S_IRUGO, show_fan_fault, NULL, 6);
+
+static ssize_t show_temp(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	int temp_nr = to_sensor_dev_attr(attr)->index;
+	struct env *p = dev_get_drvdata(dev);
+	s8 val;
+
+	val = env_read(p, IREG_LCL_TEMP + temp_nr);
+	return sprintf(buf, "%d\n", ((int) val) - 64);
+}
+
+static SENSOR_DEVICE_ATTR(adt7462_local_temp, S_IRUGO, show_temp, NULL, 0);
+static SENSOR_DEVICE_ATTR(cpu0_temp, S_IRUGO, show_temp, NULL, 1);
+static SENSOR_DEVICE_ATTR(cpu1_temp, S_IRUGO, show_temp, NULL, 2);
+static SENSOR_DEVICE_ATTR(motherboard_temp, S_IRUGO, show_temp, NULL, 3);
+static SENSOR_DEVICE_ATTR(lm95221_local_temp, S_IRUGO, show_temp, NULL, 4);
+static SENSOR_DEVICE_ATTR(fire_temp, S_IRUGO, show_temp, NULL, 5);
+static SENSOR_DEVICE_ATTR(lsi1064_local_temp, S_IRUGO, show_temp, NULL, 6);
+static SENSOR_DEVICE_ATTR(front_panel_temp, S_IRUGO, show_temp, NULL, 7);
+static SENSOR_DEVICE_ATTR(psu_temp, S_IRUGO, show_temp, NULL, 13);
+
+static ssize_t show_stat_bit(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	int index = to_sensor_dev_attr(attr)->index;
+	struct env *p = dev_get_drvdata(dev);
+	u8 val;
+
+	val = readb(p->regs + REG_STAT);
+	return sprintf(buf, "%d\n", (val & (1 << index)) ? 1 : 0);
+}
+
+static SENSOR_DEVICE_ATTR(fan_failure, S_IRUGO, show_stat_bit, NULL, 0);
+static SENSOR_DEVICE_ATTR(env_bus_busy, S_IRUGO, show_stat_bit, NULL, 1);
+static SENSOR_DEVICE_ATTR(env_data_stale, S_IRUGO, show_stat_bit, NULL, 2);
+static SENSOR_DEVICE_ATTR(tpm_self_test_passed, S_IRUGO, show_stat_bit, NULL, 3);
+
+static ssize_t show_fwver(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct env *p = dev_get_drvdata(dev);
+	u8 val;
+
+	val = readb(p->regs + REG_STAT);
+	return sprintf(buf, "%d\n", val >> 4);
+}
+
+static SENSOR_DEVICE_ATTR(firmware_version, S_IRUGO, show_fwver, NULL, 0);
+
+static ssize_t show_name(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	return sprintf(buf, "ultra45\n");
+}
+
+static SENSOR_DEVICE_ATTR(name, S_IRUGO, show_name, NULL, 0);
+
+static struct attribute *env_attributes[] = {
+	&sensor_dev_attr_fan0_speed.dev_attr.attr,
+	&sensor_dev_attr_fan0_fault.dev_attr.attr,
+	&sensor_dev_attr_fan1_speed.dev_attr.attr,
+	&sensor_dev_attr_fan1_fault.dev_attr.attr,
+	&sensor_dev_attr_fan2_speed.dev_attr.attr,
+	&sensor_dev_attr_fan2_fault.dev_attr.attr,
+	&sensor_dev_attr_fan3_speed.dev_attr.attr,
+	&sensor_dev_attr_fan3_fault.dev_attr.attr,
+	&sensor_dev_attr_fan4_speed.dev_attr.attr,
+	&sensor_dev_attr_fan4_fault.dev_attr.attr,
+	&sensor_dev_attr_psu_fan_fault.dev_attr.attr,
+	&sensor_dev_attr_adt7462_local_temp.dev_attr.attr,
+	&sensor_dev_attr_cpu0_temp.dev_attr.attr,
+	&sensor_dev_attr_cpu1_temp.dev_attr.attr,
+	&sensor_dev_attr_motherboard_temp.dev_attr.attr,
+	&sensor_dev_attr_lm95221_local_temp.dev_attr.attr,
+	&sensor_dev_attr_fire_temp.dev_attr.attr,
+	&sensor_dev_attr_lsi1064_local_temp.dev_attr.attr,
+	&sensor_dev_attr_front_panel_temp.dev_attr.attr,
+	&sensor_dev_attr_psu_temp.dev_attr.attr,
+	&sensor_dev_attr_fan_failure.dev_attr.attr,
+	&sensor_dev_attr_env_bus_busy.dev_attr.attr,
+	&sensor_dev_attr_env_data_stale.dev_attr.attr,
+	&sensor_dev_attr_tpm_self_test_passed.dev_attr.attr,
+	&sensor_dev_attr_firmware_version.dev_attr.attr,
+	&sensor_dev_attr_name.dev_attr.attr,
+	NULL,
+};
+
+static const struct attribute_group env_group = {
+	.attrs = env_attributes,
+};
+
+static int __devinit env_probe(struct of_device *op,
+			       const struct of_device_id *match)
+{
+	struct env *p = kzalloc(sizeof(*p), GFP_KERNEL);
+	int err = -ENOMEM;
+
+	if (!p)
+		goto out;
+
+	spin_lock_init(&p->lock);
+
+	p->regs = of_ioremap(&op->resource[0], 0, REG_SIZE, "pic16f747");
+	if (!p->regs)
+		goto out_free;
+
+	err = sysfs_create_group(&op->dev.kobj, &env_group);
+	if (err)
+		goto out_iounmap;
+
+	p->hwmon_dev = hwmon_device_register(&op->dev);
+	if (IS_ERR(p->hwmon_dev)) {
+		err = PTR_ERR(p->hwmon_dev);
+		goto out_sysfs_remove_group;
+	}
+
+	dev_set_drvdata(&op->dev, p);
+	err = 0;
+
+out:
+	return err;
+
+out_sysfs_remove_group:
+	sysfs_remove_group(&op->dev.kobj, &env_group);
+
+out_iounmap:
+	of_iounmap(&op->resource[0], p->regs, REG_SIZE);
+
+out_free:
+	kfree(p);
+	goto out;
+}
+
+static int __devexit env_remove(struct of_device *op)
+{
+	struct env *p = dev_get_drvdata(&op->dev);
+
+	if (p) {
+		sysfs_remove_group(&op->dev.kobj, &env_group);
+		hwmon_device_unregister(p->hwmon_dev);
+		of_iounmap(&op->resource[0], p->regs, REG_SIZE);
+		kfree(p);
+	}
+
+	return 0;
+}
+
+static const struct of_device_id env_match[] = {
+	{
+		.name = "env-monitor",
+		.compatible = "SUNW,ebus-pic16f747-env",
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, env_match);
+
+static struct of_platform_driver env_driver = {
+	.name		= "ultra45_env",
+	.match_table	= env_match,
+	.probe		= env_probe,
+	.remove		= __devexit_p(env_remove),
+};
+
+static int __init env_init(void)
+{
+	return of_register_driver(&env_driver, &of_bus_type);
+}
+
+static void __exit env_exit(void)
+{
+	of_unregister_driver(&env_driver);
+}
+
+module_init(env_init);
+module_exit(env_exit);
diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig
index 052879a6f853..b50b5dac95b0 100644
--- a/drivers/ide/Kconfig
+++ b/drivers/ide/Kconfig
@@ -131,29 +131,6 @@ config BLK_DEV_IDEDISK
 
 	  If unsure, say Y.
 
-config IDEDISK_MULTI_MODE
-	bool "Use multiple sector mode for Programmed Input/Output by default"
-	help
-	  This setting is irrelevant for most IDE disks, with direct memory
-	  access, to which multiple sector mode does not apply. Multiple sector
-	  mode is a feature of most modern IDE hard drives, permitting the
-	  transfer of multiple sectors per Programmed Input/Output interrupt,
-	  rather than the usual one sector per interrupt. When this feature is
-	  enabled, it can reduce operating system overhead for disk Programmed
-	  Input/Output. On some systems, it also can increase the data
-	  throughput of Programmed Input/Output. Some drives, however, seemed
-	  to run slower with multiple sector mode enabled. Some drives claimed
-	  to support multiple sector mode, but lost data at some settings.
-	  Under rare circumstances, such failures could result in massive
-	  filesystem corruption.
-
-	  If you get the following error, try to say Y here:
-
-	  hda: set_multmode: status=0x51 { DriveReady SeekComplete Error }
-	  hda: set_multmode: error=0x04 { DriveStatusError }
-
-	  If in doubt, say N.
-
 config BLK_DEV_IDECS
 	tristate "PCMCIA IDE support"
 	depends on PCMCIA
diff --git a/drivers/ide/Makefile b/drivers/ide/Makefile
index 64e0ecdc4ed5..308b8a12f314 100644
--- a/drivers/ide/Makefile
+++ b/drivers/ide/Makefile
@@ -4,8 +4,8 @@
 
 EXTRA_CFLAGS				+= -Idrivers/ide
 
-ide-core-y += ide.o ide-io.o ide-iops.o ide-lib.o ide-probe.o ide-taskfile.o \
-	      ide-pio-blacklist.o
+ide-core-y += ide.o ide-ioctls.o ide-io.o ide-iops.o ide-lib.o ide-probe.o \
+	      ide-taskfile.o ide-pio-blacklist.o
 
 # core IDE code
 ide-core-$(CONFIG_IDE_TIMINGS)		+= ide-timings.o
@@ -37,11 +37,12 @@ obj-$(CONFIG_IDE_GENERIC)		+= ide-generic.o
 obj-$(CONFIG_BLK_DEV_IDEPNP)		+= ide-pnp.o
 
 ide-cd_mod-y += ide-cd.o ide-cd_ioctl.o ide-cd_verbose.o
+ide-floppy_mod-y += ide-floppy.o ide-floppy_ioctl.o
 
 obj-$(CONFIG_BLK_DEV_IDEDISK)		+= ide-disk.o
 obj-$(CONFIG_BLK_DEV_IDECD)		+= ide-cd_mod.o
+obj-$(CONFIG_BLK_DEV_IDEFLOPPY)		+= ide-floppy_mod.o
 obj-$(CONFIG_BLK_DEV_IDETAPE)		+= ide-tape.o
-obj-$(CONFIG_BLK_DEV_IDEFLOPPY)		+= ide-floppy.o
 
 ifeq ($(CONFIG_BLK_DEV_IDECS), y)
 	ide-cs-core-y += legacy/ide-cs.o
diff --git a/drivers/ide/arm/icside.c b/drivers/ide/arm/icside.c
index df4af4083954..70f5b164828b 100644
--- a/drivers/ide/arm/icside.c
+++ b/drivers/ide/arm/icside.c
@@ -10,7 +10,6 @@
 #include <linux/slab.h>
 #include <linux/blkdev.h>
 #include <linux/errno.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/dma-mapping.h>
 #include <linux/device.h>
@@ -265,8 +264,8 @@ static void icside_set_dma_mode(ide_drive_t *drive, const u8 xfer_mode)
 	 * If we're going to be doing MW_DMA_1 or MW_DMA_2, we should
 	 * take care to note the values in the ID...
 	 */
-	if (use_dma_info && drive->id->eide_dma_time > cycle_time)
-		cycle_time = drive->id->eide_dma_time;
+	if (use_dma_info && drive->id[ATA_ID_EIDE_DMA_TIME] > cycle_time)
+		cycle_time = drive->id[ATA_ID_EIDE_DMA_TIME];
 
 	drive->drive_data = cycle_time;
 
diff --git a/drivers/ide/arm/palm_bk3710.c b/drivers/ide/arm/palm_bk3710.c
index 4fd91dcf1dc2..122ed3c072fd 100644
--- a/drivers/ide/arm/palm_bk3710.c
+++ b/drivers/ide/arm/palm_bk3710.c
@@ -27,7 +27,6 @@
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/ioport.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/delay.h>
 #include <linux/init.h>
@@ -180,7 +179,7 @@ static void palm_bk3710_setpiomode(void __iomem *base, ide_drive_t *mate,
 	val32 |= (t2i << (dev ? 8 : 0));
 	writel(val32, base + BK3710_DATRCVR);
 
-	if (mate && mate->present) {
+	if (mate) {
 		u8 mode2 = ide_get_best_pio_mode(mate, 255, 4);
 
 		if (mode2 < mode)
@@ -213,7 +212,8 @@ static void palm_bk3710_set_dma_mode(ide_drive_t *drive, u8 xferspeed)
 		palm_bk3710_setudmamode(base, is_slave,
 					xferspeed - XFER_UDMA_0);
 	} else {
-		palm_bk3710_setdmamode(base, is_slave, drive->id->eide_dma_min,
+		palm_bk3710_setdmamode(base, is_slave,
+				       drive->id[ATA_ID_EIDE_DMA_MIN],
 				       xferspeed);
 	}
 }
@@ -229,7 +229,7 @@ static void palm_bk3710_set_pio_mode(ide_drive_t *drive, u8 pio)
 	 * Obtain the drive PIO data for tuning the Palm Chip registers
 	 */
 	cycle_time = ide_pio_cycle_time(drive, pio);
-	mate = ide_get_paired_drive(drive);
+	mate = ide_get_pair_dev(drive);
 	palm_bk3710_setpiomode(base, mate, is_slave, cycle_time, pio);
 }
 
diff --git a/drivers/ide/ide-acpi.c b/drivers/ide/ide-acpi.c
index 6f704628c27d..2427c380b3dc 100644
--- a/drivers/ide/ide-acpi.c
+++ b/drivers/ide/ide-acpi.c
@@ -584,7 +584,7 @@ void ide_acpi_get_timing(ide_hwif_t *hwif)
  * This function executes the _STM ACPI method for the target channel.
  *
  * _STM requires Identify Drive data, which has to passed as an argument.
- * Unfortunately hd_driveid is a mangled version which we can't readily
+ * Unfortunately drive->id is a mangled version which we can't readily
  * use; hence we'll get the information afresh.
  */
 void ide_acpi_push_timing(ide_hwif_t *hwif)
@@ -614,10 +614,10 @@ void ide_acpi_push_timing(ide_hwif_t *hwif)
 	in_params[0].buffer.length = sizeof(struct GTM_buffer);
 	in_params[0].buffer.pointer = (u8 *)&hwif->acpidata->gtm;
 	in_params[1].type = ACPI_TYPE_BUFFER;
-	in_params[1].buffer.length = sizeof(struct hd_driveid);
+	in_params[1].buffer.length = sizeof(ATA_ID_WORDS * 2);
 	in_params[1].buffer.pointer = (u8 *)&master->idbuff;
 	in_params[2].type = ACPI_TYPE_BUFFER;
-	in_params[2].buffer.length = sizeof(struct hd_driveid);
+	in_params[2].buffer.length = sizeof(ATA_ID_WORDS * 2);
 	in_params[2].buffer.pointer = (u8 *)&slave->idbuff;
 	/* Output buffer: _STM has no output */
 
diff --git a/drivers/ide/ide-atapi.c b/drivers/ide/ide-atapi.c
index adf04f99cdeb..608c5bade929 100644
--- a/drivers/ide/ide-atapi.c
+++ b/drivers/ide/ide-atapi.c
@@ -14,12 +14,201 @@
 #define debug_log(fmt, args...) do {} while (0)
 #endif
 
+/*
+ * Check whether we can support a device,
+ * based on the ATAPI IDENTIFY command results.
+ */
+int ide_check_atapi_device(ide_drive_t *drive, const char *s)
+{
+	u16 *id = drive->id;
+	u8 gcw[2], protocol, device_type, removable, drq_type, packet_size;
+
+	*((u16 *)&gcw) = id[ATA_ID_CONFIG];
+
+	protocol    = (gcw[1] & 0xC0) >> 6;
+	device_type =  gcw[1] & 0x1F;
+	removable   = (gcw[0] & 0x80) >> 7;
+	drq_type    = (gcw[0] & 0x60) >> 5;
+	packet_size =  gcw[0] & 0x03;
+
+#ifdef CONFIG_PPC
+	/* kludge for Apple PowerBook internal zip */
+	if (drive->media == ide_floppy && device_type == 5 &&
+	    !strstr((char *)&id[ATA_ID_PROD], "CD-ROM") &&
+	    strstr((char *)&id[ATA_ID_PROD], "ZIP"))
+		device_type = 0;
+#endif
+
+	if (protocol != 2)
+		printk(KERN_ERR "%s: %s: protocol (0x%02x) is not ATAPI\n",
+			s, drive->name, protocol);
+	else if ((drive->media == ide_floppy && device_type != 0) ||
+		 (drive->media == ide_tape && device_type != 1))
+		printk(KERN_ERR "%s: %s: invalid device type (0x%02x)\n",
+			s, drive->name, device_type);
+	else if (removable == 0)
+		printk(KERN_ERR "%s: %s: the removable flag is not set\n",
+			s, drive->name);
+	else if (drive->media == ide_floppy && drq_type == 3)
+		printk(KERN_ERR "%s: %s: sorry, DRQ type (0x%02x) not "
+			"supported\n", s, drive->name, drq_type);
+	else if (packet_size != 0)
+		printk(KERN_ERR "%s: %s: packet size (0x%02x) is not 12 "
+			"bytes\n", s, drive->name, packet_size);
+	else
+		return 1;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ide_check_atapi_device);
+
+/* PIO data transfer routine using the scatter gather table. */
+int ide_io_buffers(ide_drive_t *drive, struct ide_atapi_pc *pc,
+		    unsigned int bcount, int write)
+{
+	ide_hwif_t *hwif = drive->hwif;
+	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
+	xfer_func_t *xf = write ? tp_ops->output_data : tp_ops->input_data;
+	struct scatterlist *sg = pc->sg;
+	char *buf;
+	int count, done = 0;
+
+	while (bcount) {
+		count = min(sg->length - pc->b_count, bcount);
+
+		if (PageHighMem(sg_page(sg))) {
+			unsigned long flags;
+
+			local_irq_save(flags);
+			buf = kmap_atomic(sg_page(sg), KM_IRQ0) + sg->offset;
+			xf(drive, NULL, buf + pc->b_count, count);
+			kunmap_atomic(buf - sg->offset, KM_IRQ0);
+			local_irq_restore(flags);
+		} else {
+			buf = sg_virt(sg);
+			xf(drive, NULL, buf + pc->b_count, count);
+		}
+
+		bcount -= count;
+		pc->b_count += count;
+		done += count;
+
+		if (pc->b_count == sg->length) {
+			if (!--pc->sg_cnt)
+				break;
+			pc->sg = sg = sg_next(sg);
+			pc->b_count = 0;
+		}
+	}
+
+	if (bcount) {
+		printk(KERN_ERR "%s: %d leftover bytes, %s\n", drive->name,
+			bcount, write ? "padding with zeros"
+				      : "discarding data");
+		ide_pad_transfer(drive, write, bcount);
+	}
+
+	return done;
+}
+EXPORT_SYMBOL_GPL(ide_io_buffers);
+
+void ide_init_pc(struct ide_atapi_pc *pc)
+{
+	memset(pc, 0, sizeof(*pc));
+	pc->buf = pc->pc_buf;
+	pc->buf_size = IDE_PC_BUFFER_SIZE;
+}
+EXPORT_SYMBOL_GPL(ide_init_pc);
+
+/*
+ * Generate a new packet command request in front of the request queue, before
+ * the current request, so that it will be processed immediately, on the next
+ * pass through the driver.
+ */
+void ide_queue_pc_head(ide_drive_t *drive, struct gendisk *disk,
+		       struct ide_atapi_pc *pc, struct request *rq)
+{
+	blk_rq_init(NULL, rq);
+	rq->cmd_type = REQ_TYPE_SPECIAL;
+	rq->cmd_flags |= REQ_PREEMPT;
+	rq->buffer = (char *)pc;
+	rq->rq_disk = disk;
+	memcpy(rq->cmd, pc->c, 12);
+	if (drive->media == ide_tape)
+		rq->cmd[13] = REQ_IDETAPE_PC1;
+	ide_do_drive_cmd(drive, rq);
+}
+EXPORT_SYMBOL_GPL(ide_queue_pc_head);
+
+/*
+ * Add a special packet command request to the tail of the request queue,
+ * and wait for it to be serviced.
+ */
+int ide_queue_pc_tail(ide_drive_t *drive, struct gendisk *disk,
+		      struct ide_atapi_pc *pc)
+{
+	struct request *rq;
+	int error;
+
+	rq = blk_get_request(drive->queue, READ, __GFP_WAIT);
+	rq->cmd_type = REQ_TYPE_SPECIAL;
+	rq->buffer = (char *)pc;
+	memcpy(rq->cmd, pc->c, 12);
+	if (drive->media == ide_tape)
+		rq->cmd[13] = REQ_IDETAPE_PC1;
+	error = blk_execute_rq(drive->queue, disk, rq, 0);
+	blk_put_request(rq);
+
+	return error;
+}
+EXPORT_SYMBOL_GPL(ide_queue_pc_tail);
+
+int ide_do_test_unit_ready(ide_drive_t *drive, struct gendisk *disk)
+{
+	struct ide_atapi_pc pc;
+
+	ide_init_pc(&pc);
+	pc.c[0] = TEST_UNIT_READY;
+
+	return ide_queue_pc_tail(drive, disk, &pc);
+}
+EXPORT_SYMBOL_GPL(ide_do_test_unit_ready);
+
+int ide_do_start_stop(ide_drive_t *drive, struct gendisk *disk, int start)
+{
+	struct ide_atapi_pc pc;
+
+	ide_init_pc(&pc);
+	pc.c[0] = START_STOP;
+	pc.c[4] = start;
+
+	if (drive->media == ide_tape)
+		pc.flags |= PC_FLAG_WAIT_FOR_DSC;
+
+	return ide_queue_pc_tail(drive, disk, &pc);
+}
+EXPORT_SYMBOL_GPL(ide_do_start_stop);
+
+int ide_set_media_lock(ide_drive_t *drive, struct gendisk *disk, int on)
+{
+	struct ide_atapi_pc pc;
+
+	if (drive->atapi_flags & IDE_AFLAG_NO_DOORLOCK)
+		return 0;
+
+	ide_init_pc(&pc);
+	pc.c[0] = ALLOW_MEDIUM_REMOVAL;
+	pc.c[4] = on;
+
+	return ide_queue_pc_tail(drive, disk, &pc);
+}
+EXPORT_SYMBOL_GPL(ide_set_media_lock);
+
 /* TODO: unify the code thus making some arguments go away */
 ide_startstop_t ide_pc_intr(ide_drive_t *drive, struct ide_atapi_pc *pc,
 	ide_handler_t *handler, unsigned int timeout, ide_expiry_t *expiry,
 	void (*update_buffers)(ide_drive_t *, struct ide_atapi_pc *),
 	void (*retry_pc)(ide_drive_t *), void (*dsc_handle)(ide_drive_t *),
-	void (*io_buffers)(ide_drive_t *, struct ide_atapi_pc *, unsigned, int))
+	int (*io_buffers)(ide_drive_t *, struct ide_atapi_pc *, unsigned, int))
 {
 	ide_hwif_t *hwif = drive->hwif;
 	struct request *rq = hwif->hwgroup->rq;
@@ -41,7 +230,7 @@ ide_startstop_t ide_pc_intr(ide_drive_t *drive, struct ide_atapi_pc *pc,
 
 	if (pc->flags & PC_FLAG_DMA_IN_PROGRESS) {
 		if (hwif->dma_ops->dma_end(drive) ||
-		    (drive->media == ide_tape && !scsi && (stat & ERR_STAT))) {
+		    (drive->media == ide_tape && !scsi && (stat & ATA_ERR))) {
 			if (drive->media == ide_floppy && !scsi)
 				printk(KERN_ERR "%s: DMA %s error\n",
 					drive->name, rq_data_dir(pc->rq)
@@ -56,7 +245,7 @@ ide_startstop_t ide_pc_intr(ide_drive_t *drive, struct ide_atapi_pc *pc,
 	}
 
 	/* No more interrupts */
-	if ((stat & DRQ_STAT) == 0) {
+	if ((stat & ATA_DRQ) == 0) {
 		debug_log("Packet command completed, %d bytes transferred\n",
 			  pc->xferred);
 
@@ -65,10 +254,10 @@ ide_startstop_t ide_pc_intr(ide_drive_t *drive, struct ide_atapi_pc *pc,
 		local_irq_enable_in_hardirq();
 
 		if (drive->media == ide_tape && !scsi &&
-		    (stat & ERR_STAT) && rq->cmd[0] == REQUEST_SENSE)
-			stat &= ~ERR_STAT;
+		    (stat & ATA_ERR) && rq->cmd[0] == REQUEST_SENSE)
+			stat &= ~ATA_ERR;
 
-		if ((stat & ERR_STAT) || (pc->flags & PC_FLAG_DMA_ERROR)) {
+		if ((stat & ATA_ERR) || (pc->flags & PC_FLAG_DMA_ERROR)) {
 			/* Error detected */
 			debug_log("%s: I/O error\n", drive->name);
 
@@ -95,7 +284,7 @@ ide_startstop_t ide_pc_intr(ide_drive_t *drive, struct ide_atapi_pc *pc,
 cmd_finished:
 		pc->error = 0;
 		if ((pc->flags & PC_FLAG_WAIT_FOR_DSC) &&
-		    (stat & SEEK_STAT) == 0) {
+		    (stat & ATA_DSC) == 0) {
 			dsc_handle(drive);
 			return ide_stopped;
 		}
@@ -117,17 +306,18 @@ cmd_finished:
 	/* Get the number of bytes to transfer on this interrupt. */
 	ide_read_bcount_and_ireason(drive, &bcount, &ireason);
 
-	if (ireason & CD) {
+	if (ireason & ATAPI_COD) {
 		printk(KERN_ERR "%s: CoD != 0 in %s\n", drive->name, __func__);
 		return ide_do_reset(drive);
 	}
 
-	if (((ireason & IO) == IO) == !!(pc->flags & PC_FLAG_WRITING)) {
+	if (((ireason & ATAPI_IO) == ATAPI_IO) ==
+		!!(pc->flags & PC_FLAG_WRITING)) {
 		/* Hopefully, we will never get here */
 		printk(KERN_ERR "%s: We wanted to %s, but the device wants us "
 				"to %s!\n", drive->name,
-				(ireason & IO) ? "Write" : "Read",
-				(ireason & IO) ? "Read" : "Write");
+				(ireason & ATAPI_IO) ? "Write" : "Read",
+				(ireason & ATAPI_IO) ? "Read" : "Write");
 		return ide_do_reset(drive);
 	}
 
@@ -171,9 +361,14 @@ cmd_finished:
 
 	if ((drive->media == ide_floppy && !scsi && !pc->buf) ||
 	    (drive->media == ide_tape && !scsi && pc->bh) ||
-	    (scsi && pc->sg))
-		io_buffers(drive, pc, bcount, !!(pc->flags & PC_FLAG_WRITING));
-	else
+	    (scsi && pc->sg)) {
+		int done = io_buffers(drive, pc, bcount,
+				  !!(pc->flags & PC_FLAG_WRITING));
+
+		/* FIXME: don't do partial completions */
+		if (drive->media == ide_floppy && !scsi)
+			ide_end_request(drive, 1, done >> 9);
+	} else
 		xferfunc(drive, NULL, pc->cur_pos, bcount);
 
 	/* Update the current position */
@@ -205,7 +400,8 @@ static u8 ide_wait_ireason(ide_drive_t *drive, u8 ireason)
 {
 	int retries = 100;
 
-	while (retries-- && ((ireason & CD) == 0 || (ireason & IO))) {
+	while (retries-- && ((ireason & ATAPI_COD) == 0 ||
+		(ireason & ATAPI_IO))) {
 		printk(KERN_ERR "%s: (IO,CoD != (0,1) while issuing "
 				"a packet command, retrying\n", drive->name);
 		udelay(100);
@@ -214,8 +410,8 @@ static u8 ide_wait_ireason(ide_drive_t *drive, u8 ireason)
 			printk(KERN_ERR "%s: (IO,CoD != (0,1) while issuing "
 					"a packet command, ignoring\n",
 					drive->name);
-			ireason |= CD;
-			ireason &= ~IO;
+			ireason |= ATAPI_COD;
+			ireason &= ~ATAPI_IO;
 		}
 	}
 
@@ -231,7 +427,7 @@ ide_startstop_t ide_transfer_pc(ide_drive_t *drive, struct ide_atapi_pc *pc,
 	ide_startstop_t startstop;
 	u8 ireason;
 
-	if (ide_wait_stat(&startstop, drive, DRQ_STAT, BUSY_STAT, WAIT_READY)) {
+	if (ide_wait_stat(&startstop, drive, ATA_DRQ, ATA_BUSY, WAIT_READY)) {
 		printk(KERN_ERR "%s: Strange, packet command initiated yet "
 				"DRQ isn't asserted\n", drive->name);
 		return startstop;
@@ -241,7 +437,7 @@ ide_startstop_t ide_transfer_pc(ide_drive_t *drive, struct ide_atapi_pc *pc,
 	if (drive->media == ide_tape && !drive->scsi)
 		ireason = ide_wait_ireason(drive, ireason);
 
-	if ((ireason & CD) == 0 || (ireason & IO)) {
+	if ((ireason & ATAPI_COD) == 0 || (ireason & ATAPI_IO)) {
 		printk(KERN_ERR "%s: (IO,CoD) != (0,1) while issuing "
 				"a packet command\n", drive->name);
 		return ide_do_reset(drive);
@@ -303,7 +499,7 @@ ide_startstop_t ide_issue_pc(ide_drive_t *drive, struct ide_atapi_pc *pc,
 
 	/* Issue the packet command */
 	if (drive->atapi_flags & IDE_AFLAG_DRQ_INTERRUPT) {
-		ide_execute_command(drive, WIN_PACKETCMD, handler,
+		ide_execute_command(drive, ATA_CMD_PACKET, handler,
 				    timeout, NULL);
 		return ide_started;
 	} else {
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 03c2cb6a58bc..465a92ca0179 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -436,7 +436,7 @@ static int cdrom_decode_status(ide_drive_t *drive, int good_stat, int *stat_ret)
 			ide_dump_status_no_sense(drive, "media error (blank)",
 						 stat);
 			do_end_request = 1;
-		} else if ((err & ~ABRT_ERR) != 0) {
+		} else if ((err & ~ATA_ABORTED) != 0) {
 			/* go to the default handler for other errors */
 			ide_error(drive, "cdrom_decode_status", stat);
 			return 1;
@@ -457,7 +457,7 @@ static int cdrom_decode_status(ide_drive_t *drive, int good_stat, int *stat_ret)
 		 * If we got a CHECK_CONDITION status, queue
 		 * a request sense command.
 		 */
-		if (stat & ERR_STAT)
+		if (stat & ATA_ERR)
 			cdrom_queue_request_sense(drive, NULL, NULL);
 	} else {
 		blk_dump_rq_flags(rq, "ide-cd: bad rq");
@@ -468,7 +468,7 @@ static int cdrom_decode_status(ide_drive_t *drive, int good_stat, int *stat_ret)
 	return 1;
 
 end_request:
-	if (stat & ERR_STAT) {
+	if (stat & ATA_ERR) {
 		unsigned long flags;
 
 		spin_lock_irqsave(&ide_lock, flags);
@@ -541,7 +541,7 @@ static ide_startstop_t cdrom_start_packet_command(ide_drive_t *drive,
 			drive->waiting_for_dma = 0;
 
 		/* packet command */
-		ide_execute_command(drive, WIN_PACKETCMD, handler,
+		ide_execute_command(drive, ATA_CMD_PACKET, handler,
 				    ATAPI_WAIT_PC, cdrom_timer_expiry);
 		return ide_started;
 	} else {
@@ -574,7 +574,7 @@ static ide_startstop_t cdrom_transfer_packet_command(ide_drive_t *drive,
 		 */
 
 		/* check for errors */
-		if (cdrom_decode_status(drive, DRQ_STAT, NULL))
+		if (cdrom_decode_status(drive, ATA_DRQ, NULL))
 			return ide_stopped;
 
 		/* ok, next interrupt will be DMA interrupt */
@@ -582,8 +582,8 @@ static ide_startstop_t cdrom_transfer_packet_command(ide_drive_t *drive,
 			drive->waiting_for_dma = 1;
 	} else {
 		/* otherwise, we must wait for DRQ to get set */
-		if (ide_wait_stat(&startstop, drive, DRQ_STAT,
-				BUSY_STAT, WAIT_READY))
+		if (ide_wait_stat(&startstop, drive, ATA_DRQ,
+				  ATA_BUSY, WAIT_READY))
 			return startstop;
 	}
 
@@ -938,7 +938,7 @@ static ide_startstop_t cdrom_newpc_intr(ide_drive_t *drive)
 		thislen = len;
 
 	/* If DRQ is clear, the command has completed. */
-	if ((stat & DRQ_STAT) == 0) {
+	if ((stat & ATA_DRQ) == 0) {
 		if (blk_fs_request(rq)) {
 			/*
 			 * If we're not done reading/writing, complain.
@@ -1164,13 +1164,12 @@ static void cdrom_do_block_pc(ide_drive_t *drive, struct request *rq)
 	if (rq->bio || ((rq->cmd_type == REQ_TYPE_ATA_PC) && rq->data_len)) {
 		struct request_queue *q = drive->queue;
 		unsigned int alignment;
-		unsigned long addr;
-		unsigned long stack_mask = ~(THREAD_SIZE - 1);
+		char *buf;
 
 		if (rq->bio)
-			addr = (unsigned long)bio_data(rq->bio);
+			buf = bio_data(rq->bio);
 		else
-			addr = (unsigned long)rq->data;
+			buf = rq->data;
 
 		info->dma = drive->using_dma;
 
@@ -1181,11 +1180,8 @@ static void cdrom_do_block_pc(ide_drive_t *drive, struct request *rq)
 		 * separate masks.
 		 */
 		alignment = queue_dma_alignment(q) | q->dma_pad_mask;
-		if (addr & alignment || rq->data_len & alignment)
-			info->dma = 0;
-
-		if (!((addr & stack_mask) ^
-		      ((unsigned long)current->stack & stack_mask)))
+		if ((unsigned long)buf & alignment || rq->data_len & alignment
+		    || object_is_on_stack(buf))
 			info->dma = 0;
 	}
 }
@@ -1206,7 +1202,7 @@ static ide_startstop_t ide_cd_do_request(ide_drive_t *drive, struct request *rq,
 			unsigned long elapsed = jiffies - info->start_seek;
 			int stat = hwif->tp_ops->read_status(hwif);
 
-			if ((stat & SEEK_STAT) != SEEK_STAT) {
+			if ((stat & ATA_DSC) != ATA_DSC) {
 				if (elapsed < IDECD_SEEK_TIMEOUT) {
 					ide_stall_queue(drive,
 							IDECD_SEEK_TIMER);
@@ -1813,13 +1809,12 @@ static ide_proc_entry_t idecd_proc[] = {
 	{ NULL, 0, NULL, NULL }
 };
 
-static void ide_cdrom_add_settings(ide_drive_t *drive)
-{
-	ide_add_setting(drive, "dsc_overlap", SETTING_RW, TYPE_BYTE, 0, 1, 1, 1,
-			&drive->dsc_overlap, NULL);
-}
-#else
-static inline void ide_cdrom_add_settings(ide_drive_t *drive) { ; }
+ide_devset_rw_field(dsc_overlap, dsc_overlap);
+
+static const struct ide_proc_devset idecd_settings[] = {
+	IDE_PROC_DEVSET(dsc_overlap, 0, 1),
+	{ 0 },
+};
 #endif
 
 static const struct cd_list_entry ide_cd_quirks_list[] = {
@@ -1866,14 +1861,14 @@ static const struct cd_list_entry ide_cd_quirks_list[] = {
 	{ NULL, NULL, 0 }
 };
 
-static unsigned int ide_cd_flags(struct hd_driveid *id)
+static unsigned int ide_cd_flags(u16 *id)
 {
 	const struct cd_list_entry *cle = ide_cd_quirks_list;
 
 	while (cle->id_model) {
-		if (strcmp(cle->id_model, id->model) == 0 &&
+		if (strcmp(cle->id_model, (char *)&id[ATA_ID_PROD]) == 0 &&
 		    (cle->id_firmware == NULL ||
-		     strstr(id->fw_rev, cle->id_firmware)))
+		     strstr((char *)&id[ATA_ID_FW_REV], cle->id_firmware)))
 			return cle->cd_flags;
 		cle++;
 	}
@@ -1885,7 +1880,8 @@ static int ide_cdrom_setup(ide_drive_t *drive)
 {
 	struct cdrom_info *cd = drive->driver_data;
 	struct cdrom_device_info *cdi = &cd->devinfo;
-	struct hd_driveid *id = drive->id;
+	u16 *id = drive->id;
+	char *fw_rev = (char *)&id[ATA_ID_FW_REV];
 	int nslots;
 
 	blk_queue_prep_rq(drive->queue, ide_cdrom_prep_fn);
@@ -1900,15 +1896,15 @@ static int ide_cdrom_setup(ide_drive_t *drive)
 	drive->atapi_flags = IDE_AFLAG_MEDIA_CHANGED | IDE_AFLAG_NO_EJECT |
 		       ide_cd_flags(id);
 
-	if ((id->config & 0x0060) == 0x20)
+	if ((id[ATA_ID_CONFIG] & 0x0060) == 0x20)
 		drive->atapi_flags |= IDE_AFLAG_DRQ_INTERRUPT;
 
 	if ((drive->atapi_flags & IDE_AFLAG_VERTOS_300_SSD) &&
-	    id->fw_rev[4] == '1' && id->fw_rev[6] <= '2')
+	    fw_rev[4] == '1' && fw_rev[6] <= '2')
 		drive->atapi_flags |= (IDE_AFLAG_TOCTRACKS_AS_BCD |
 				     IDE_AFLAG_TOCADDR_AS_BCD);
 	else if ((drive->atapi_flags & IDE_AFLAG_VERTOS_600_ESD) &&
-		 id->fw_rev[4] == '1' && id->fw_rev[6] <= '2')
+		 fw_rev[4] == '1' && fw_rev[6] <= '2')
 		drive->atapi_flags |= IDE_AFLAG_TOCTRACKS_AS_BCD;
 	else if (drive->atapi_flags & IDE_AFLAG_SANYO_3CD)
 		/* 3 => use CD in slot 0 */
@@ -1927,7 +1923,8 @@ static int ide_cdrom_setup(ide_drive_t *drive)
 		cd->devinfo.handle = NULL;
 		return 1;
 	}
-	ide_cdrom_add_settings(drive);
+
+	ide_proc_register_driver(drive, cd->driver);
 	return 0;
 }
 
@@ -1972,12 +1969,12 @@ static ide_driver_t ide_cdrom_driver = {
 	.remove			= ide_cd_remove,
 	.version		= IDECD_VERSION,
 	.media			= ide_cdrom,
-	.supports_dsc_overlap	= 1,
 	.do_request		= ide_cd_do_request,
 	.end_request		= ide_end_request,
 	.error			= __ide_error,
 #ifdef CONFIG_IDE_PROC_FS
 	.proc			= idecd_proc,
+	.settings		= idecd_settings,
 #endif
 };
 
@@ -2112,10 +2109,10 @@ static int ide_cd_probe(ide_drive_t *drive)
 
 	if (!strstr("ide-cdrom", drive->driver_req))
 		goto failed;
-	if (!drive->present)
-		goto failed;
+
 	if (drive->media != ide_cdrom && drive->media != ide_optical)
 		goto failed;
+
 	/* skip drives that we were told to ignore */
 	if (ignore != NULL) {
 		if (strstr(ignore, drive->name)) {
@@ -2137,8 +2134,6 @@ static int ide_cd_probe(ide_drive_t *drive)
 
 	ide_init_disk(g, drive);
 
-	ide_proc_register_driver(drive, &ide_cdrom_driver);
-
 	kref_init(&info->kref);
 
 	info->drive = drive;
@@ -2153,7 +2148,6 @@ static int ide_cd_probe(ide_drive_t *drive)
 	g->driverfs_dev = &drive->gendev;
 	g->flags = GENHD_FL_CD | GENHD_FL_REMOVABLE;
 	if (ide_cdrom_setup(drive)) {
-		ide_proc_unregister_driver(drive, &ide_cdrom_driver);
 		ide_cd_release(&info->kref);
 		goto failed;
 	}
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 33ea8c048717..01846f244b40 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -30,10 +30,8 @@
 #include <linux/delay.h>
 #include <linux/mutex.h>
 #include <linux/leds.h>
-
-#define _IDE_DISK
-
 #include <linux/ide.h>
+#include <linux/hdreg.h>
 
 #include <asm/byteorder.h>
 #include <asm/irq.h>
@@ -90,68 +88,19 @@ static void ide_disk_put(struct ide_disk_obj *idkp)
 	mutex_unlock(&idedisk_ref_mutex);
 }
 
-/*
- * lba_capacity_is_ok() performs a sanity check on the claimed "lba_capacity"
- * value for this drive (from its reported identification information).
- *
- * Returns:	1 if lba_capacity looks sensible
- *		0 otherwise
- *
- * It is called only once for each drive.
- */
-static int lba_capacity_is_ok(struct hd_driveid *id)
-{
-	unsigned long lba_sects, chs_sects, head, tail;
-
-	/* No non-LBA info .. so valid! */
-	if (id->cyls == 0)
-		return 1;
-
-	/*
-	 * The ATA spec tells large drives to return
-	 * C/H/S = 16383/16/63 independent of their size.
-	 * Some drives can be jumpered to use 15 heads instead of 16.
-	 * Some drives can be jumpered to use 4092 cyls instead of 16383.
-	 */
-	if ((id->cyls == 16383
-	     || (id->cyls == 4092 && id->cur_cyls == 16383)) &&
-	    id->sectors == 63 &&
-	    (id->heads == 15 || id->heads == 16) &&
-	    (id->lba_capacity >= 16383*63*id->heads))
-		return 1;
-
-	lba_sects   = id->lba_capacity;
-	chs_sects   = id->cyls * id->heads * id->sectors;
-
-	/* perform a rough sanity check on lba_sects:  within 10% is OK */
-	if ((lba_sects - chs_sects) < chs_sects/10)
-		return 1;
-
-	/* some drives have the word order reversed */
-	head = ((lba_sects >> 16) & 0xffff);
-	tail = (lba_sects & 0xffff);
-	lba_sects = (head | (tail << 16));
-	if ((lba_sects - chs_sects) < chs_sects/10) {
-		id->lba_capacity = lba_sects;
-		return 1;	/* lba_capacity is (now) good */
-	}
-
-	return 0;	/* lba_capacity value may be bad */
-}
-
 static const u8 ide_rw_cmds[] = {
-	WIN_MULTREAD,
-	WIN_MULTWRITE,
-	WIN_MULTREAD_EXT,
-	WIN_MULTWRITE_EXT,
-	WIN_READ,
-	WIN_WRITE,
-	WIN_READ_EXT,
-	WIN_WRITE_EXT,
-	WIN_READDMA,
-	WIN_WRITEDMA,
-	WIN_READDMA_EXT,
-	WIN_WRITEDMA_EXT,
+	ATA_CMD_READ_MULTI,
+	ATA_CMD_WRITE_MULTI,
+	ATA_CMD_READ_MULTI_EXT,
+	ATA_CMD_WRITE_MULTI_EXT,
+	ATA_CMD_PIO_READ,
+	ATA_CMD_PIO_WRITE,
+	ATA_CMD_PIO_READ_EXT,
+	ATA_CMD_PIO_WRITE_EXT,
+	ATA_CMD_READ,
+	ATA_CMD_WRITE,
+	ATA_CMD_READ_EXT,
+	ATA_CMD_WRITE_EXT,
 };
 
 static const u8 ide_data_phases[] = {
@@ -322,9 +271,9 @@ static u64 idedisk_read_native_max_address(ide_drive_t *drive, int lba48)
 	/* Create IDE/ATA command request structure */
 	memset(&args, 0, sizeof(ide_task_t));
 	if (lba48)
-		tf->command = WIN_READ_NATIVE_MAX_EXT;
+		tf->command = ATA_CMD_READ_NATIVE_MAX_EXT;
 	else
-		tf->command = WIN_READ_NATIVE_MAX;
+		tf->command = ATA_CMD_READ_NATIVE_MAX;
 	tf->device  = ATA_LBA;
 	args.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 	if (lba48)
@@ -359,10 +308,10 @@ static u64 idedisk_set_max_address(ide_drive_t *drive, u64 addr_req, int lba48)
 		tf->hob_lbal = (addr_req >>= 8) & 0xff;
 		tf->hob_lbam = (addr_req >>= 8) & 0xff;
 		tf->hob_lbah = (addr_req >>= 8) & 0xff;
-		tf->command  = WIN_SET_MAX_EXT;
+		tf->command  = ATA_CMD_SET_MAX_EXT;
 	} else {
 		tf->device   = (addr_req >>= 8) & 0x0f;
-		tf->command  = WIN_SET_MAX;
+		tf->command  = ATA_CMD_SET_MAX;
 	}
 	tf->device |= ATA_LBA;
 	args.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
@@ -385,25 +334,6 @@ static unsigned long long sectors_to_MB(unsigned long long n)
 }
 
 /*
- * Bits 10 of command_set_1 and cfs_enable_1 must be equal,
- * so on non-buggy drives we need test only one.
- * However, we should also check whether these fields are valid.
- */
-static inline int idedisk_supports_hpa(const struct hd_driveid *id)
-{
-	return (id->command_set_1 & 0x0400) && (id->cfs_enable_1 & 0x0400);
-}
-
-/*
- * The same here.
- */
-static inline int idedisk_supports_lba48(const struct hd_driveid *id)
-{
-	return (id->command_set_2 & 0x0400) && (id->cfs_enable_2 & 0x0400)
-	       && id->lba_capacity_2;
-}
-
-/*
  * Some disks report total number of sectors instead of
  * maximum sector address.  We list them here.
  */
@@ -417,7 +347,7 @@ static const struct drive_list_entry hpa_list[] = {
 static void idedisk_check_hpa(ide_drive_t *drive)
 {
 	unsigned long long capacity, set_max;
-	int lba48 = idedisk_supports_lba48(drive->id);
+	int lba48 = ata_id_lba48_enabled(drive->id);
 
 	capacity = drive->capacity64;
 
@@ -453,23 +383,23 @@ static void idedisk_check_hpa(ide_drive_t *drive)
 
 static void init_idedisk_capacity(ide_drive_t *drive)
 {
-	struct hd_driveid *id = drive->id;
+	u16 *id = drive->id;
 	/*
 	 * If this drive supports the Host Protected Area feature set,
 	 * then we may need to change our opinion about the drive's capacity.
 	 */
-	int hpa = idedisk_supports_hpa(id);
+	int hpa = ata_id_hpa_enabled(id);
 
-	if (idedisk_supports_lba48(id)) {
+	if (ata_id_lba48_enabled(id)) {
 		/* drive speaks 48-bit LBA */
 		drive->select.b.lba = 1;
-		drive->capacity64 = id->lba_capacity_2;
+		drive->capacity64 = ata_id_u64(id, ATA_ID_LBA_CAPACITY_2);
 		if (hpa)
 			idedisk_check_hpa(drive);
-	} else if ((id->capability & 2) && lba_capacity_is_ok(id)) {
+	} else if (ata_id_has_lba(id) && ata_id_is_lba_capacity_ok(id)) {
 		/* drive speaks 28-bit LBA */
 		drive->select.b.lba = 1;
-		drive->capacity64 = id->lba_capacity;
+		drive->capacity64 = ata_id_u32(id, ATA_ID_LBA_CAPACITY);
 		if (hpa)
 			idedisk_check_hpa(drive);
 	} else {
@@ -480,7 +410,7 @@ static void init_idedisk_capacity(ide_drive_t *drive)
 
 static sector_t idedisk_capacity(ide_drive_t *drive)
 {
-	return drive->capacity64 - drive->sect0;
+	return drive->capacity64;
 }
 
 #ifdef CONFIG_IDE_PROC_FS
@@ -490,10 +420,10 @@ static int smart_enable(ide_drive_t *drive)
 	struct ide_taskfile *tf = &args.tf;
 
 	memset(&args, 0, sizeof(ide_task_t));
-	tf->feature = SMART_ENABLE;
-	tf->lbam    = SMART_LCYL_PASS;
-	tf->lbah    = SMART_HCYL_PASS;
-	tf->command = WIN_SMART;
+	tf->feature = ATA_SMART_ENABLE;
+	tf->lbam    = ATA_SMART_LBAM_PASS;
+	tf->lbah    = ATA_SMART_LBAH_PASS;
+	tf->command = ATA_CMD_SMART;
 	args.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 	return ide_no_data_taskfile(drive, &args);
 }
@@ -506,9 +436,9 @@ static int get_smart_data(ide_drive_t *drive, u8 *buf, u8 sub_cmd)
 	memset(&args, 0, sizeof(ide_task_t));
 	tf->feature = sub_cmd;
 	tf->nsect   = 0x01;
-	tf->lbam    = SMART_LCYL_PASS;
-	tf->lbah    = SMART_HCYL_PASS;
-	tf->command = WIN_SMART;
+	tf->lbam    = ATA_SMART_LBAM_PASS;
+	tf->lbah    = ATA_SMART_LBAH_PASS;
+	tf->command = ATA_CMD_SMART;
 	args.tf_flags	= IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 	args.data_phase	= TASKFILE_IN;
 	(void) smart_enable(drive);
@@ -523,7 +453,7 @@ static int proc_idedisk_read_cache
 	int		len;
 
 	if (drive->id_read)
-		len = sprintf(out, "%i\n", drive->id->buf_size / 2);
+		len = sprintf(out, "%i\n", drive->id[ATA_ID_BUF_SIZE] / 2);
 	else
 		len = sprintf(out, "(none)\n");
 
@@ -549,13 +479,14 @@ static int proc_idedisk_read_smart(char *page, char **start, off_t off,
 
 	if (get_smart_data(drive, page, sub_cmd) == 0) {
 		unsigned short *val = (unsigned short *) page;
-		char *out = ((char *)val) + (SECTOR_WORDS * 4);
+		char *out = (char *)val + SECTOR_SIZE;
+
 		page = out;
 		do {
 			out += sprintf(out, "%04x%c", le16_to_cpu(*val),
 				       (++i & 7) ? ' ' : '\n');
 			val += 1;
-		} while (i < (SECTOR_WORDS * 2));
+		} while (i < SECTOR_SIZE / 2);
 		len = out - page;
 	}
 
@@ -566,14 +497,14 @@ static int proc_idedisk_read_sv
 	(char *page, char **start, off_t off, int count, int *eof, void *data)
 {
 	return proc_idedisk_read_smart(page, start, off, count, eof, data,
-				       SMART_READ_VALUES);
+				       ATA_SMART_READ_VALUES);
 }
 
 static int proc_idedisk_read_st
 	(char *page, char **start, off_t off, int count, int *eof, void *data)
 {
 	return proc_idedisk_read_smart(page, start, off, count, eof, data,
-				       SMART_READ_THRESHOLDS);
+				       ATA_SMART_READ_THRESHOLDS);
 }
 
 static ide_proc_entry_t idedisk_proc[] = {
@@ -595,11 +526,11 @@ static void idedisk_prepare_flush(struct request_queue *q, struct request *rq)
 	BUG_ON(task == NULL);
 
 	memset(task, 0, sizeof(*task));
-	if (ide_id_has_flush_cache_ext(drive->id) &&
+	if (ata_id_flush_ext_enabled(drive->id) &&
 	    (drive->capacity64 >= (1UL << 28)))
-		task->tf.command = WIN_FLUSH_CACHE_EXT;
+		task->tf.command = ATA_CMD_FLUSH_EXT;
 	else
-		task->tf.command = WIN_FLUSH_CACHE;
+		task->tf.command = ATA_CMD_FLUSH;
 	task->tf_flags	 = IDE_TFLAG_OUT_TF | IDE_TFLAG_OUT_DEVICE |
 			   IDE_TFLAG_DYN;
 	task->data_phase = TASKFILE_NO_DATA;
@@ -609,6 +540,8 @@ static void idedisk_prepare_flush(struct request_queue *q, struct request *rq)
 	rq->special = task;
 }
 
+ide_devset_get(multcount, mult_count);
+
 /*
  * This is tightly woven into the driver->do_special can not touch.
  * DON'T do it again until a total personality rewrite is committed.
@@ -618,7 +551,7 @@ static int set_multcount(ide_drive_t *drive, int arg)
 	struct request *rq;
 	int error;
 
-	if (arg < 0 || arg > drive->id->max_multsect)
+	if (arg < 0 || arg > (drive->id[ATA_ID_MAX_MULTSECT] & 0xff))
 		return -EINVAL;
 
 	if (drive->special.b.set_multmode)
@@ -635,22 +568,21 @@ static int set_multcount(ide_drive_t *drive, int arg)
 	return (drive->mult_count == arg) ? 0 : -EIO;
 }
 
+ide_devset_get(nowerr, nowerr);
+
 static int set_nowerr(ide_drive_t *drive, int arg)
 {
 	if (arg < 0 || arg > 1)
 		return -EINVAL;
 
-	if (ide_spin_wait_hwgroup(drive))
-		return -EBUSY;
 	drive->nowerr = arg;
 	drive->bad_wstat = arg ? BAD_R_STAT : BAD_W_STAT;
-	spin_unlock_irq(&ide_lock);
 	return 0;
 }
 
 static void update_ordered(ide_drive_t *drive)
 {
-	struct hd_driveid *id = drive->id;
+	u16 *id = drive->id;
 	unsigned ordered = QUEUE_ORDERED_NONE;
 	prepare_flush_fn *prep_fn = NULL;
 
@@ -666,9 +598,9 @@ static void update_ordered(ide_drive_t *drive)
 		 * not available so we don't need to recheck that.
 		 */
 		capacity = idedisk_capacity(drive);
-		barrier = ide_id_has_flush_cache(id) && !drive->noflush &&
+		barrier = ata_id_flush_enabled(id) && !drive->noflush &&
 			(drive->addressing == 0 || capacity <= (1ULL << 28) ||
-			 ide_id_has_flush_cache_ext(id));
+			 ata_id_flush_ext_enabled(id));
 
 		printk(KERN_INFO "%s: cache flushes %ssupported\n",
 		       drive->name, barrier ? "" : "not ");
@@ -683,7 +615,9 @@ static void update_ordered(ide_drive_t *drive)
 	blk_queue_ordered(drive->queue, ordered, prep_fn);
 }
 
-static int write_cache(ide_drive_t *drive, int arg)
+ide_devset_get(wcache, wcache);
+
+static int set_wcache(ide_drive_t *drive, int arg)
 {
 	ide_task_t args;
 	int err = 1;
@@ -691,11 +625,11 @@ static int write_cache(ide_drive_t *drive, int arg)
 	if (arg < 0 || arg > 1)
 		return -EINVAL;
 
-	if (ide_id_has_flush_cache(drive->id)) {
+	if (ata_id_flush_enabled(drive->id)) {
 		memset(&args, 0, sizeof(ide_task_t));
 		args.tf.feature = arg ?
-			SETFEATURES_EN_WCACHE : SETFEATURES_DIS_WCACHE;
-		args.tf.command = WIN_SETFEATURES;
+			SETFEATURES_WC_ON : SETFEATURES_WC_OFF;
+		args.tf.command = ATA_CMD_SET_FEATURES;
 		args.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 		err = ide_no_data_taskfile(drive, &args);
 		if (err == 0)
@@ -712,14 +646,16 @@ static int do_idedisk_flushcache(ide_drive_t *drive)
 	ide_task_t args;
 
 	memset(&args, 0, sizeof(ide_task_t));
-	if (ide_id_has_flush_cache_ext(drive->id))
-		args.tf.command = WIN_FLUSH_CACHE_EXT;
+	if (ata_id_flush_ext_enabled(drive->id))
+		args.tf.command = ATA_CMD_FLUSH_EXT;
 	else
-		args.tf.command = WIN_FLUSH_CACHE;
+		args.tf.command = ATA_CMD_FLUSH;
 	args.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 	return ide_no_data_taskfile(drive, &args);
 }
 
+ide_devset_get(acoustic, acoustic);
+
 static int set_acoustic(ide_drive_t *drive, int arg)
 {
 	ide_task_t args;
@@ -728,22 +664,24 @@ static int set_acoustic(ide_drive_t *drive, int arg)
 		return -EINVAL;
 
 	memset(&args, 0, sizeof(ide_task_t));
-	args.tf.feature = arg ? SETFEATURES_EN_AAM : SETFEATURES_DIS_AAM;
+	args.tf.feature = arg ? SETFEATURES_AAM_ON : SETFEATURES_AAM_OFF;
 	args.tf.nsect   = arg;
-	args.tf.command = WIN_SETFEATURES;
+	args.tf.command = ATA_CMD_SET_FEATURES;
 	args.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 	ide_no_data_taskfile(drive, &args);
 	drive->acoustic = arg;
 	return 0;
 }
 
+ide_devset_get(addressing, addressing);
+
 /*
  * drive->addressing:
  *	0: 28-bit
  *	1: 48-bit
  *	2: 48-bit capable doing 28-bit
  */
-static int set_lba_addressing(ide_drive_t *drive, int arg)
+static int set_addressing(ide_drive_t *drive, int arg)
 {
 	if (arg < 0 || arg > 2)
 		return -EINVAL;
@@ -753,52 +691,54 @@ static int set_lba_addressing(ide_drive_t *drive, int arg)
 	if (drive->hwif->host_flags & IDE_HFLAG_NO_LBA48)
 		return 0;
 
-	if (!idedisk_supports_lba48(drive->id))
+	if (ata_id_lba48_enabled(drive->id) == 0)
 		return -EIO;
+
 	drive->addressing = arg;
+
 	return 0;
 }
 
+ide_devset_rw(acoustic, acoustic);
+ide_devset_rw(address, addressing);
+ide_devset_rw(multcount, multcount);
+ide_devset_rw(wcache, wcache);
+
+ide_devset_rw_sync(nowerr, nowerr);
+
 #ifdef CONFIG_IDE_PROC_FS
-static void idedisk_add_settings(ide_drive_t *drive)
-{
-	struct hd_driveid *id = drive->id;
-
-	ide_add_setting(drive, "bios_cyl", SETTING_RW, TYPE_INT, 0, 65535, 1, 1,
-			&drive->bios_cyl, NULL);
-	ide_add_setting(drive, "bios_head", SETTING_RW, TYPE_BYTE, 0, 255, 1, 1,
-			&drive->bios_head, NULL);
-	ide_add_setting(drive, "bios_sect", SETTING_RW, TYPE_BYTE, 0, 63, 1, 1,
-			&drive->bios_sect, NULL);
-	ide_add_setting(drive, "address", SETTING_RW, TYPE_BYTE, 0, 2, 1, 1,
-			&drive->addressing, set_lba_addressing);
-	ide_add_setting(drive, "multcount", SETTING_RW, TYPE_BYTE, 0,
-			id->max_multsect, 1, 1, &drive->mult_count,
-			set_multcount);
-	ide_add_setting(drive, "nowerr", SETTING_RW, TYPE_BYTE, 0, 1, 1, 1,
-			&drive->nowerr, set_nowerr);
-	ide_add_setting(drive, "lun", SETTING_RW, TYPE_INT, 0, 7, 1, 1,
-			&drive->lun, NULL);
-	ide_add_setting(drive, "wcache", SETTING_RW, TYPE_BYTE, 0, 1, 1, 1,
-			&drive->wcache, write_cache);
-	ide_add_setting(drive, "acoustic", SETTING_RW, TYPE_BYTE, 0, 254, 1, 1,
-			&drive->acoustic, set_acoustic);
-	ide_add_setting(drive, "failures", SETTING_RW, TYPE_INT, 0, 65535, 1, 1,
-			&drive->failures, NULL);
-	ide_add_setting(drive, "max_failures", SETTING_RW, TYPE_INT, 0, 65535,
-			1, 1, &drive->max_failures, NULL);
-}
-#else
-static inline void idedisk_add_settings(ide_drive_t *drive) { ; }
+ide_devset_rw_field(bios_cyl, bios_cyl);
+ide_devset_rw_field(bios_head, bios_head);
+ide_devset_rw_field(bios_sect, bios_sect);
+ide_devset_rw_field(failures, failures);
+ide_devset_rw_field(lun, lun);
+ide_devset_rw_field(max_failures, max_failures);
+
+static const struct ide_proc_devset idedisk_settings[] = {
+	IDE_PROC_DEVSET(acoustic,	0,   254),
+	IDE_PROC_DEVSET(address,	0,     2),
+	IDE_PROC_DEVSET(bios_cyl,	0, 65535),
+	IDE_PROC_DEVSET(bios_head,	0,   255),
+	IDE_PROC_DEVSET(bios_sect,	0,    63),
+	IDE_PROC_DEVSET(failures,	0, 65535),
+	IDE_PROC_DEVSET(lun,		0,     7),
+	IDE_PROC_DEVSET(max_failures,	0, 65535),
+	IDE_PROC_DEVSET(multcount,	0,    16),
+	IDE_PROC_DEVSET(nowerr,		0,     1),
+	IDE_PROC_DEVSET(wcache,		0,     1),
+	{ 0 },
+};
 #endif
 
 static void idedisk_setup(ide_drive_t *drive)
 {
+	struct ide_disk_obj *idkp = drive->driver_data;
 	ide_hwif_t *hwif = drive->hwif;
-	struct hd_driveid *id = drive->id;
+	u16 *id = drive->id;
+	char *m = (char *)&id[ATA_ID_PROD];
 	unsigned long long capacity;
 
-	idedisk_add_settings(drive);
+	ide_proc_register_driver(drive, idkp->driver);
 
 	if (drive->id_read == 0)
 		return;
@@ -807,11 +747,11 @@ static void idedisk_setup(ide_drive_t *drive)
 		/*
 		 * Removable disks (eg. SYQUEST); ignore 'WD' drives
 		 */
-		if (id->model[0] != 'W' || id->model[1] != 'D')
+		if (m[0] != 'W' || m[1] != 'D')
 			drive->doorlocking = 1;
 	}
 
-	(void)set_lba_addressing(drive, 1);
+	(void)set_addressing(drive, 1);
 
 	if (drive->addressing == 1) {
 		int max_s = 2048;
@@ -853,8 +793,7 @@ static void idedisk_setup(ide_drive_t *drive)
 	capacity = idedisk_capacity(drive);
 
 	if (!drive->forced_geom) {
-
-		if (idedisk_supports_lba48(drive->id)) {
+		if (ata_id_lba48_enabled(drive->id)) {
 			/* compatibility */
 			drive->bios_sect = 63;
 			drive->bios_head = 255;
@@ -880,22 +819,22 @@ static void idedisk_setup(ide_drive_t *drive)
 			 drive->name, capacity, sectors_to_MB(capacity));
 
 	/* Only print cache size when it was specified */
-	if (id->buf_size)
-		printk(KERN_CONT " w/%dKiB Cache", id->buf_size / 2);
+	if (id[ATA_ID_BUF_SIZE])
+		printk(KERN_CONT " w/%dKiB Cache", id[ATA_ID_BUF_SIZE] / 2);
 
 	printk(KERN_CONT ", CHS=%d/%d/%d\n",
 			 drive->bios_cyl, drive->bios_head, drive->bios_sect);
 
 	/* write cache enabled? */
-	if ((id->csfo & 1) || (id->cfs_enable_1 & (1 << 5)))
+	if ((id[ATA_ID_CSFO] & 1) || ata_id_wcache_enabled(id))
 		drive->wcache = 1;
 
-	write_cache(drive, 1);
+	set_wcache(drive, 1);
 }
 
 static void ide_cacheflush_p(ide_drive_t *drive)
 {
-	if (!drive->wcache || !ide_id_has_flush_cache(drive->id))
+	if (!drive->wcache || ata_id_flush_enabled(drive->id) == 0)
 		return;
 
 	if (do_idedisk_flushcache(drive))
@@ -937,7 +876,7 @@ static int ide_disk_probe(ide_drive_t *drive);
  */
 static void ide_disk_resume(ide_drive_t *drive)
 {
-	if (idedisk_supports_hpa(drive->id))
+	if (ata_id_hpa_enabled(drive->id))
 		init_idedisk_capacity(drive);
 }
 
@@ -980,12 +919,12 @@ static ide_driver_t idedisk_driver = {
 	.shutdown		= ide_device_shutdown,
 	.version		= IDEDISK_VERSION,
 	.media			= ide_disk,
-	.supports_dsc_overlap	= 0,
 	.do_request		= ide_do_rw_disk,
 	.end_request		= ide_end_request,
 	.error			= __ide_error,
 #ifdef CONFIG_IDE_PROC_FS
 	.proc			= idedisk_proc,
+	.settings		= idedisk_settings,
 #endif
 };
 
@@ -994,7 +933,7 @@ static int idedisk_set_doorlock(ide_drive_t *drive, int on)
 	ide_task_t task;
 
 	memset(&task, 0, sizeof(task));
-	task.tf.command = on ? WIN_DOORLOCK : WIN_DOORUNLOCK;
+	task.tf.command = on ? ATA_CMD_MEDIA_LOCK : ATA_CMD_MEDIA_UNLOCK;
 	task.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 
 	return ide_no_data_taskfile(drive, &task);
@@ -1059,52 +998,28 @@ static int idedisk_getgeo(struct block_device *bdev, struct hd_geometry *geo)
 	return 0;
 }
 
+static const struct ide_ioctl_devset ide_disk_ioctl_settings[] = {
+{ HDIO_GET_ADDRESS,	HDIO_SET_ADDRESS,   &ide_devset_address   },
+{ HDIO_GET_MULTCOUNT,	HDIO_SET_MULTCOUNT, &ide_devset_multcount },
+{ HDIO_GET_NOWERR,	HDIO_SET_NOWERR,    &ide_devset_nowerr	  },
+{ HDIO_GET_WCACHE,	HDIO_SET_WCACHE,    &ide_devset_wcache	  },
+{ HDIO_GET_ACOUSTIC,	HDIO_SET_ACOUSTIC,  &ide_devset_acoustic  },
+{ 0 }
+};
+
 static int idedisk_ioctl(struct inode *inode, struct file *file,
 			unsigned int cmd, unsigned long arg)
 {
-	unsigned long flags;
 	struct block_device *bdev = inode->i_bdev;
 	struct ide_disk_obj *idkp = ide_disk_g(bdev->bd_disk);
 	ide_drive_t *drive = idkp->drive;
-	int err, (*setfunc)(ide_drive_t *, int);
-	u8 *val;
-
-	switch (cmd) {
-	case HDIO_GET_ADDRESS:	 val = &drive->addressing;	goto read_val;
-	case HDIO_GET_MULTCOUNT: val = &drive->mult_count;	goto read_val;
-	case HDIO_GET_NOWERR:	 val = &drive->nowerr;		goto read_val;
-	case HDIO_GET_WCACHE:	 val = &drive->wcache;		goto read_val;
-	case HDIO_GET_ACOUSTIC:	 val = &drive->acoustic;	goto read_val;
-	case HDIO_SET_ADDRESS:	 setfunc = set_lba_addressing;	goto set_val;
-	case HDIO_SET_MULTCOUNT: setfunc = set_multcount;	goto set_val;
-	case HDIO_SET_NOWERR:	 setfunc = set_nowerr;		goto set_val;
-	case HDIO_SET_WCACHE:	 setfunc = write_cache;		goto set_val;
-	case HDIO_SET_ACOUSTIC:	 setfunc = set_acoustic;	goto set_val;
-	}
+	int err;
 
-	return generic_ide_ioctl(drive, file, bdev, cmd, arg);
+	err = ide_setting_ioctl(drive, bdev, cmd, arg, ide_disk_ioctl_settings);
+	if (err != -EOPNOTSUPP)
+		return err;
 
-read_val:
-	mutex_lock(&ide_setting_mtx);
-	spin_lock_irqsave(&ide_lock, flags);
-	err = *val;
-	spin_unlock_irqrestore(&ide_lock, flags);
-	mutex_unlock(&ide_setting_mtx);
-	return err >= 0 ? put_user(err, (long __user *)arg) : err;
-
-set_val:
-	if (bdev != bdev->bd_contains)
-		err = -EINVAL;
-	else {
-		if (!capable(CAP_SYS_ADMIN))
-			err = -EACCES;
-		else {
-			mutex_lock(&ide_setting_mtx);
-			err = setfunc(drive, arg);
-			mutex_unlock(&ide_setting_mtx);
-		}
-	}
-	return err;
+	return generic_ide_ioctl(drive, file, bdev, cmd, arg);
 }
 
 static int idedisk_media_changed(struct gendisk *disk)
@@ -1148,8 +1063,7 @@ static int ide_disk_probe(ide_drive_t *drive)
 	/* strstr("foo", "") is non-NULL */
 	if (!strstr("ide-disk", drive->driver_req))
 		goto failed;
-	if (!drive->present)
-		goto failed;
+
 	if (drive->media != ide_disk)
 		goto failed;
 
@@ -1163,8 +1077,6 @@ static int ide_disk_probe(ide_drive_t *drive)
 
 	ide_init_disk(g, drive);
 
-	ide_proc_register_driver(drive, &idedisk_driver);
-
 	kref_init(&idkp->kref);
 
 	idkp->drive = drive;
diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index 3fa07c0aeaa4..ef2f1504c0d5 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -106,7 +106,7 @@ ide_startstop_t ide_dma_intr (ide_drive_t *drive)
 	dma_stat = hwif->dma_ops->dma_end(drive);
 	stat = hwif->tp_ops->read_status(hwif);
 
-	if (OK_STAT(stat,DRIVE_READY,drive->bad_wstat|DRQ_STAT)) {
+	if (OK_STAT(stat, DRIVE_READY, drive->bad_wstat | ATA_DRQ)) {
 		if (!dma_stat) {
 			struct request *rq = HWGROUP(drive)->rq;
 
@@ -288,7 +288,7 @@ EXPORT_SYMBOL_GPL(ide_destroy_dmatable);
 static int config_drive_for_dma (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	struct hd_driveid *id = drive->id;
+	u16 *id = drive->id;
 
 	if (drive->media != ide_disk) {
 		if (hwif->host_flags & IDE_HFLAG_NO_ATAPI_DMA)
@@ -299,16 +299,17 @@ static int config_drive_for_dma (ide_drive_t *drive)
 	 * Enable DMA on any drive that has
 	 * UltraDMA (mode 0/1/2/3/4/5/6) enabled
 	 */
-	if ((id->field_valid & 4) && ((id->dma_ultra >> 8) & 0x7f))
+	if ((id[ATA_ID_FIELD_VALID] & 4) &&
+	    ((id[ATA_ID_UDMA_MODES] >> 8) & 0x7f))
 		return 1;
 
 	/*
 	 * Enable DMA on any drive that has mode2 DMA
 	 * (multi or single) enabled
 	 */
-	if (id->field_valid & 2)	/* regular DMA */
-		if ((id->dma_mword & 0x404) == 0x404 ||
-		    (id->dma_1word & 0x404) == 0x404)
+	if (id[ATA_ID_FIELD_VALID] & 2)	/* regular DMA */
+		if ((id[ATA_ID_MWDMA_MODES] & 0x404) == 0x404 ||
+		    (id[ATA_ID_SWDMA_MODES] & 0x404) == 0x404)
 			return 1;
 
 	/* Consult the list of known "good" drives */
@@ -591,12 +592,12 @@ static inline int config_drive_for_dma(ide_drive_t *drive) { return 0; }
 
 int __ide_dma_bad_drive (ide_drive_t *drive)
 {
-	struct hd_driveid *id = drive->id;
+	u16 *id = drive->id;
 
 	int blacklist = ide_in_drive_list(id, drive_blacklist);
 	if (blacklist) {
 		printk(KERN_WARNING "%s: Disabling (U)DMA for %s (blacklisted)\n",
-				    drive->name, id->model);
+				    drive->name, (char *)&id[ATA_ID_PROD]);
 		return blacklist;
 	}
 	return 0;
@@ -612,21 +613,21 @@ static const u8 xfer_mode_bases[] = {
 
 static unsigned int ide_get_mode_mask(ide_drive_t *drive, u8 base, u8 req_mode)
 {
-	struct hd_driveid *id = drive->id;
+	u16 *id = drive->id;
 	ide_hwif_t *hwif = drive->hwif;
 	const struct ide_port_ops *port_ops = hwif->port_ops;
 	unsigned int mask = 0;
 
 	switch(base) {
 	case XFER_UDMA_0:
-		if ((id->field_valid & 4) == 0)
+		if ((id[ATA_ID_FIELD_VALID] & 4) == 0)
 			break;
 
 		if (port_ops && port_ops->udma_filter)
 			mask = port_ops->udma_filter(drive);
 		else
 			mask = hwif->ultra_mask;
-		mask &= id->dma_ultra;
+		mask &= id[ATA_ID_UDMA_MODES];
 
 		/*
 		 * avoid false cable warning from eighty_ninty_three()
@@ -637,19 +638,19 @@ static unsigned int ide_get_mode_mask(ide_drive_t *drive, u8 base, u8 req_mode)
 		}
 		break;
 	case XFER_MW_DMA_0:
-		if ((id->field_valid & 2) == 0)
+		if ((id[ATA_ID_FIELD_VALID] & 2) == 0)
 			break;
 		if (port_ops && port_ops->mdma_filter)
 			mask = port_ops->mdma_filter(drive);
 		else
 			mask = hwif->mwdma_mask;
-		mask &= id->dma_mword;
+		mask &= id[ATA_ID_MWDMA_MODES];
 		break;
 	case XFER_SW_DMA_0:
-		if (id->field_valid & 2) {
-			mask = id->dma_1word & hwif->swdma_mask;
-		} else if (id->tDMA) {
-			u8 mode = id->tDMA;
+		if (id[ATA_ID_FIELD_VALID] & 2) {
+			mask = id[ATA_ID_SWDMA_MODES] & hwif->swdma_mask;
+		} else if (id[ATA_ID_OLD_DMA_MODES] >> 8) {
+			u8 mode = id[ATA_ID_OLD_DMA_MODES] >> 8;
 
 			/*
 			 * if the mode is valid convert it to the mask
@@ -706,7 +707,8 @@ u8 ide_find_dma_mode(ide_drive_t *drive, u8 req_mode)
 		/*
 		 * is this correct?
 		 */
-		if (ide_dma_good_drive(drive) && drive->id->eide_dma_time < 150)
+		if (ide_dma_good_drive(drive) &&
+		    drive->id[ATA_ID_EIDE_DMA_TIME] < 150)
 			mode = XFER_MW_DMA_1;
 	}
 
@@ -725,7 +727,7 @@ static int ide_tune_dma(ide_drive_t *drive)
 	ide_hwif_t *hwif = drive->hwif;
 	u8 speed;
 
-	if (drive->nodma || (drive->id->capability & 1) == 0)
+	if (drive->nodma || ata_id_has_dma(drive->id) == 0)
 		return 0;
 
 	/* consult the list of known "bad" drives */
@@ -767,13 +769,15 @@ static int ide_dma_check(ide_drive_t *drive)
 
 int ide_id_dma_bug(ide_drive_t *drive)
 {
-	struct hd_driveid *id = drive->id;
+	u16 *id = drive->id;
 
-	if (id->field_valid & 4) {
-		if ((id->dma_ultra >> 8) && (id->dma_mword >> 8))
+	if (id[ATA_ID_FIELD_VALID] & 4) {
+		if ((id[ATA_ID_UDMA_MODES] >> 8) &&
+		    (id[ATA_ID_MWDMA_MODES] >> 8))
 			goto err_out;
-	} else if (id->field_valid & 2) {
-		if ((id->dma_mword >> 8) && (id->dma_1word >> 8))
+	} else if (id[ATA_ID_FIELD_VALID] & 2) {
+		if ((id[ATA_ID_MWDMA_MODES] >> 8) &&
+		    (id[ATA_ID_SWDMA_MODES] >> 8))
 			goto err_out;
 	}
 	return 0;
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index e9034c0125f3..d36f155470a4 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -15,6 +15,8 @@
  * Documentation/ide/ChangeLog.ide-floppy.1996-2002
  */
 
+#define DRV_NAME "ide-floppy"
+
 #define IDEFLOPPY_VERSION "1.00"
 
 #include <linux/module.h>
@@ -31,8 +33,10 @@
 #include <linux/slab.h>
 #include <linux/cdrom.h>
 #include <linux/ide.h>
+#include <linux/hdreg.h>
 #include <linux/bitops.h>
 #include <linux/mutex.h>
+#include <linux/scatterlist.h>
 
 #include <scsi/scsi_ioctl.h>
 
@@ -42,6 +46,8 @@
 #include <linux/io.h>
 #include <asm/unaligned.h>
 
+#include "ide-floppy.h"
+
 /* define to see debug info */
 #define IDEFLOPPY_DEBUG_LOG		0
 
@@ -55,102 +61,23 @@
 #define debug_log(fmt, args...) do {} while (0)
 #endif
 
-
-/* Some drives require a longer irq timeout. */
-#define IDEFLOPPY_WAIT_CMD		(5 * WAIT_CMD)
-
 /*
  * After each failed packet command we issue a request sense command and retry
  * the packet command IDEFLOPPY_MAX_PC_RETRIES times.
  */
 #define IDEFLOPPY_MAX_PC_RETRIES	3
 
-/*
- * With each packet command, we allocate a buffer of IDEFLOPPY_PC_BUFFER_SIZE
- * bytes.
- */
-#define IDEFLOPPY_PC_BUFFER_SIZE	256
-
-/*
- * In various places in the driver, we need to allocate storage for packet
- * commands and requests, which will remain valid while	we leave the driver to
- * wait for an interrupt or a timeout event.
- */
-#define IDEFLOPPY_PC_STACK		(10 + IDEFLOPPY_MAX_PC_RETRIES)
-
 /* format capacities descriptor codes */
 #define CAPACITY_INVALID	0x00
 #define CAPACITY_UNFORMATTED	0x01
 #define CAPACITY_CURRENT	0x02
 #define CAPACITY_NO_CARTRIDGE	0x03
 
-/*
- * Most of our global data which we need to save even as we leave the driver
- * due to an interrupt or a timer event is stored in a variable of type
- * idefloppy_floppy_t, defined below.
- */
-typedef struct ide_floppy_obj {
-	ide_drive_t	*drive;
-	ide_driver_t	*driver;
-	struct gendisk	*disk;
-	struct kref	kref;
-	unsigned int	openers;	/* protected by BKL for now */
-
-	/* Current packet command */
-	struct ide_atapi_pc *pc;
-	/* Last failed packet command */
-	struct ide_atapi_pc *failed_pc;
-	/* Packet command stack */
-	struct ide_atapi_pc pc_stack[IDEFLOPPY_PC_STACK];
-	/* Next free packet command storage space */
-	int pc_stack_index;
-	struct request rq_stack[IDEFLOPPY_PC_STACK];
-	/* We implement a circular array */
-	int rq_stack_index;
-
-	/* Last error information */
-	u8 sense_key, asc, ascq;
-	/* delay this long before sending packet command */
-	u8 ticks;
-	int progress_indication;
-
-	/* Device information */
-	/* Current format */
-	int blocks, block_size, bs_factor;
-	/* Last format capacity descriptor */
-	u8 cap_desc[8];
-	/* Copy of the flexible disk page */
-	u8 flexible_disk_page[32];
-	/* Write protect */
-	int wp;
-	/* Supports format progress report */
-	int srfp;
-} idefloppy_floppy_t;
-
 #define IDEFLOPPY_TICKS_DELAY	HZ/20	/* default delay for ZIP 100 (50ms) */
 
-/* Defines for the MODE SENSE command */
-#define MODE_SENSE_CURRENT		0x00
-#define MODE_SENSE_CHANGEABLE		0x01
-#define MODE_SENSE_DEFAULT		0x02
-#define MODE_SENSE_SAVED		0x03
-
-/* IOCTLs used in low-level formatting. */
-#define	IDEFLOPPY_IOCTL_FORMAT_SUPPORTED	0x4600
-#define	IDEFLOPPY_IOCTL_FORMAT_GET_CAPACITY	0x4601
-#define	IDEFLOPPY_IOCTL_FORMAT_START		0x4602
-#define IDEFLOPPY_IOCTL_FORMAT_GET_PROGRESS	0x4603
-
 /* Error code returned in rq->errors to the higher part of the driver. */
 #define	IDEFLOPPY_ERROR_GENERAL		101
 
-/*
- * Pages of the SELECT SENSE / MODE SENSE packet commands.
- * See SFF-8070i spec.
- */
-#define	IDEFLOPPY_CAPABILITIES_PAGE	0x1b
-#define IDEFLOPPY_FLEXIBLE_DISK_PAGE	0x05
-
 static DEFINE_MUTEX(idefloppy_ref_mutex);
 
 #define to_ide_floppy(obj) container_of(obj, struct ide_floppy_obj, kref)
@@ -219,44 +146,6 @@ static int idefloppy_end_request(ide_drive_t *drive, int uptodate, int nsecs)
 	return 0;
 }
 
-static void ide_floppy_io_buffers(ide_drive_t *drive, struct ide_atapi_pc *pc,
-				  unsigned int bcount, int direction)
-{
-	ide_hwif_t *hwif = drive->hwif;
-	struct request *rq = pc->rq;
-	struct req_iterator iter;
-	struct bio_vec *bvec;
-	unsigned long flags;
-	int count, done = 0;
-	char *data;
-
-	rq_for_each_segment(bvec, rq, iter) {
-		if (!bcount)
-			break;
-
-		count = min(bvec->bv_len, bcount);
-
-		data = bvec_kmap_irq(bvec, &flags);
-		if (direction)
-			hwif->tp_ops->output_data(drive, NULL, data, count);
-		else
-			hwif->tp_ops->input_data(drive, NULL, data, count);
-		bvec_kunmap_irq(data, &flags);
-
-		bcount -= count;
-		pc->b_count += count;
-		done += count;
-	}
-
-	idefloppy_end_request(drive, 1, done >> 9);
-
-	if (bcount) {
-		printk(KERN_ERR "%s: leftover data in %s, bcount == %d\n",
-				drive->name, __func__, bcount);
-		ide_pad_transfer(drive, direction, bcount);
-	}
-}
-
 static void idefloppy_update_buffers(ide_drive_t *drive,
 				struct ide_atapi_pc *pc)
 {
@@ -267,43 +156,6 @@ static void idefloppy_update_buffers(ide_drive_t *drive,
 		idefloppy_end_request(drive, 1, 0);
 }
 
-/*
- * Generate a new packet command request in front of the request queue, before
- * the current request so that it will be processed immediately, on the next
- * pass through the driver.
- */
-static void idefloppy_queue_pc_head(ide_drive_t *drive, struct ide_atapi_pc *pc,
-		struct request *rq)
-{
-	struct ide_floppy_obj *floppy = drive->driver_data;
-
-	blk_rq_init(NULL, rq);
-	rq->buffer = (char *) pc;
-	rq->cmd_type = REQ_TYPE_SPECIAL;
-	rq->cmd_flags |= REQ_PREEMPT;
-	rq->rq_disk = floppy->disk;
-	memcpy(rq->cmd, pc->c, 12);
-	ide_do_drive_cmd(drive, rq);
-}
-
-static struct ide_atapi_pc *idefloppy_next_pc_storage(ide_drive_t *drive)
-{
-	idefloppy_floppy_t *floppy = drive->driver_data;
-
-	if (floppy->pc_stack_index == IDEFLOPPY_PC_STACK)
-		floppy->pc_stack_index = 0;
-	return (&floppy->pc_stack[floppy->pc_stack_index++]);
-}
-
-static struct request *idefloppy_next_rq_storage(ide_drive_t *drive)
-{
-	idefloppy_floppy_t *floppy = drive->driver_data;
-
-	if (floppy->rq_stack_index == IDEFLOPPY_PC_STACK)
-		floppy->rq_stack_index = 0;
-	return (&floppy->rq_stack[floppy->rq_stack_index++]);
-}
-
 static void ide_floppy_callback(ide_drive_t *drive)
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
@@ -341,16 +193,9 @@ static void ide_floppy_callback(ide_drive_t *drive)
 	idefloppy_end_request(drive, uptodate, 0);
 }
 
-static void idefloppy_init_pc(struct ide_atapi_pc *pc)
-{
-	memset(pc, 0, sizeof(*pc));
-	pc->buf = pc->pc_buf;
-	pc->buf_size = IDEFLOPPY_PC_BUFFER_SIZE;
-}
-
-static void idefloppy_create_request_sense_cmd(struct ide_atapi_pc *pc)
+void ide_floppy_create_request_sense_cmd(struct ide_atapi_pc *pc)
 {
-	idefloppy_init_pc(pc);
+	ide_init_pc(pc);
 	pc->c[0] = GPCMD_REQUEST_SENSE;
 	pc->c[4] = 255;
 	pc->req_xfer = 18;
@@ -362,14 +207,13 @@ static void idefloppy_create_request_sense_cmd(struct ide_atapi_pc *pc)
  */
 static void idefloppy_retry_pc(ide_drive_t *drive)
 {
-	struct ide_atapi_pc *pc;
-	struct request *rq;
+	struct ide_floppy_obj *floppy = drive->driver_data;
+	struct request *rq = &floppy->request_sense_rq;
+	struct ide_atapi_pc *pc = &floppy->request_sense_pc;
 
 	(void)ide_read_error(drive);
-	pc = idefloppy_next_pc_storage(drive);
-	rq = idefloppy_next_rq_storage(drive);
-	idefloppy_create_request_sense_cmd(pc);
-	idefloppy_queue_pc_head(drive, pc, rq);
+	ide_floppy_create_request_sense_cmd(pc);
+	ide_queue_pc_head(drive, floppy->disk, pc, rq);
 }
 
 /* The usual interrupt handler called during a packet command. */
@@ -378,8 +222,8 @@ static ide_startstop_t idefloppy_pc_intr(ide_drive_t *drive)
 	idefloppy_floppy_t *floppy = drive->driver_data;
 
 	return ide_pc_intr(drive, floppy->pc, idefloppy_pc_intr,
-			   IDEFLOPPY_WAIT_CMD, NULL, idefloppy_update_buffers,
-			   idefloppy_retry_pc, NULL, ide_floppy_io_buffers);
+			   WAIT_FLOPPY_CMD, NULL, idefloppy_update_buffers,
+			   idefloppy_retry_pc, NULL, ide_io_buffers);
 }
 
 /*
@@ -396,10 +240,9 @@ static int idefloppy_transfer_pc(ide_drive_t *drive)
 	drive->hwif->tp_ops->output_data(drive, NULL, floppy->pc->c, 12);
 
 	/* Timeout for the packet command */
-	return IDEFLOPPY_WAIT_CMD;
+	return WAIT_FLOPPY_CMD;
 }
 
-
 /*
  * Called as an interrupt (or directly). When the device says it's ready for a
  * packet, we schedule the packet transfer to occur about 2-3 ticks later in
@@ -424,7 +267,7 @@ static ide_startstop_t idefloppy_start_pc_transfer(ide_drive_t *drive)
 		timeout = floppy->ticks;
 		expiry = &idefloppy_transfer_pc;
 	} else {
-		timeout = IDEFLOPPY_WAIT_CMD;
+		timeout = WAIT_FLOPPY_CMD;
 		expiry = NULL;
 	}
 
@@ -474,58 +317,27 @@ static ide_startstop_t idefloppy_issue_pc(ide_drive_t *drive,
 	pc->retries++;
 
 	return ide_issue_pc(drive, pc, idefloppy_start_pc_transfer,
-			    IDEFLOPPY_WAIT_CMD, NULL);
-}
-
-static void idefloppy_create_prevent_cmd(struct ide_atapi_pc *pc, int prevent)
-{
-	debug_log("creating prevent removal command, prevent = %d\n", prevent);
-
-	idefloppy_init_pc(pc);
-	pc->c[0] = GPCMD_PREVENT_ALLOW_MEDIUM_REMOVAL;
-	pc->c[4] = prevent;
+			    WAIT_FLOPPY_CMD, NULL);
 }
 
-static void idefloppy_create_read_capacity_cmd(struct ide_atapi_pc *pc)
+void ide_floppy_create_read_capacity_cmd(struct ide_atapi_pc *pc)
 {
-	idefloppy_init_pc(pc);
+	ide_init_pc(pc);
 	pc->c[0] = GPCMD_READ_FORMAT_CAPACITIES;
 	pc->c[7] = 255;
 	pc->c[8] = 255;
 	pc->req_xfer = 255;
 }
 
-static void idefloppy_create_format_unit_cmd(struct ide_atapi_pc *pc, int b,
-		int l, int flags)
-{
-	idefloppy_init_pc(pc);
-	pc->c[0] = GPCMD_FORMAT_UNIT;
-	pc->c[1] = 0x17;
-
-	memset(pc->buf, 0, 12);
-	pc->buf[1] = 0xA2;
-	/* Default format list header, u8 1: FOV/DCRT/IMM bits set */
-
-	if (flags & 1)				/* Verify bit on... */
-		pc->buf[1] ^= 0x20;		/* ... turn off DCRT bit */
-	pc->buf[3] = 8;
-
-	put_unaligned(cpu_to_be32(b), (unsigned int *)(&pc->buf[4]));
-	put_unaligned(cpu_to_be32(l), (unsigned int *)(&pc->buf[8]));
-	pc->buf_size = 12;
-	pc->flags |= PC_FLAG_WRITING;
-}
-
 /* A mode sense command is used to "sense" floppy parameters. */
-static void idefloppy_create_mode_sense_cmd(struct ide_atapi_pc *pc,
-		u8 page_code, u8 type)
+void ide_floppy_create_mode_sense_cmd(struct ide_atapi_pc *pc, u8 page_code)
 {
 	u16 length = 8; /* sizeof(Mode Parameter Header) = 8 Bytes */
 
-	idefloppy_init_pc(pc);
+	ide_init_pc(pc);
 	pc->c[0] = GPCMD_MODE_SENSE_10;
 	pc->c[1] = 0;
-	pc->c[2] = page_code + (type << 6);
+	pc->c[2] = page_code;
 
 	switch (page_code) {
 	case IDEFLOPPY_CAPABILITIES_PAGE:
@@ -542,13 +354,6 @@ static void idefloppy_create_mode_sense_cmd(struct ide_atapi_pc *pc,
 	pc->req_xfer = length;
 }
 
-static void idefloppy_create_start_stop_cmd(struct ide_atapi_pc *pc, int start)
-{
-	idefloppy_init_pc(pc);
-	pc->c[0] = GPCMD_START_STOP_UNIT;
-	pc->c[4] = start;
-}
-
 static void idefloppy_create_rw_cmd(idefloppy_floppy_t *floppy,
 				    struct ide_atapi_pc *pc, struct request *rq,
 				    unsigned long sector)
@@ -560,7 +365,7 @@ static void idefloppy_create_rw_cmd(idefloppy_floppy_t *floppy,
 	debug_log("create_rw10_cmd: block == %d, blocks == %d\n",
 		block, blocks);
 
-	idefloppy_init_pc(pc);
+	ide_init_pc(pc);
 	pc->c[0] = cmd == READ ? GPCMD_READ_10 : GPCMD_WRITE_10;
 	put_unaligned(cpu_to_be16(blocks), (unsigned short *)&pc->c[7]);
 	put_unaligned(cpu_to_be32(block), (unsigned int *) &pc->c[2]);
@@ -568,7 +373,7 @@ static void idefloppy_create_rw_cmd(idefloppy_floppy_t *floppy,
 	memcpy(rq->cmd, pc->c, 12);
 
 	pc->rq = rq;
-	pc->b_count = cmd == READ ? 0 : rq->bio->bi_size;
+	pc->b_count = 0;
 	if (rq->cmd_flags & REQ_RW)
 		pc->flags |= PC_FLAG_WRITING;
 	pc->buf = NULL;
@@ -579,10 +384,10 @@ static void idefloppy_create_rw_cmd(idefloppy_floppy_t *floppy,
 static void idefloppy_blockpc_cmd(idefloppy_floppy_t *floppy,
 		struct ide_atapi_pc *pc, struct request *rq)
 {
-	idefloppy_init_pc(pc);
+	ide_init_pc(pc);
 	memcpy(pc->c, rq->cmd, sizeof(pc->c));
 	pc->rq = rq;
-	pc->b_count = rq->data_len;
+	pc->b_count = 0;
 	if (rq->data_len && rq_data_dir(rq) == WRITE)
 		pc->flags |= PC_FLAG_WRITING;
 	pc->buf = rq->data;
@@ -599,15 +404,17 @@ static ide_startstop_t idefloppy_do_request(ide_drive_t *drive,
 		struct request *rq, sector_t block_s)
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
+	ide_hwif_t *hwif = drive->hwif;
 	struct ide_atapi_pc *pc;
 	unsigned long block = (unsigned long)block_s;
 
-	debug_log("dev: %s, cmd_type: %x, errors: %d\n",
-			rq->rq_disk ? rq->rq_disk->disk_name : "?",
-			rq->cmd_type, rq->errors);
-	debug_log("sector: %ld, nr_sectors: %ld, "
-			"current_nr_sectors: %d\n", (long)rq->sector,
-			rq->nr_sectors, rq->current_nr_sectors);
+	debug_log("%s: dev: %s, cmd: 0x%x, cmd_type: %x, errors: %d\n",
+		  __func__, rq->rq_disk ? rq->rq_disk->disk_name : "?",
+		  rq->cmd[0], rq->cmd_type, rq->errors);
+
+	debug_log("%s: sector: %ld, nr_sectors: %ld, current_nr_sectors: %d\n",
+		  __func__, (long)rq->sector, rq->nr_sectors,
+		  rq->current_nr_sectors);
 
 	if (rq->errors >= ERROR_MAX) {
 		if (floppy->failed_pc)
@@ -626,12 +433,12 @@ static ide_startstop_t idefloppy_do_request(ide_drive_t *drive,
 			idefloppy_end_request(drive, 0, 0);
 			return ide_stopped;
 		}
-		pc = idefloppy_next_pc_storage(drive);
+		pc = &floppy->queued_pc;
 		idefloppy_create_rw_cmd(floppy, pc, rq, block);
 	} else if (blk_special_request(rq)) {
 		pc = (struct ide_atapi_pc *) rq->buffer;
 	} else if (blk_pc_request(rq)) {
-		pc = idefloppy_next_pc_storage(drive);
+		pc = &floppy->queued_pc;
 		idefloppy_blockpc_cmd(floppy, pc, rq);
 	} else {
 		blk_dump_rq_flags(rq,
@@ -640,29 +447,15 @@ static ide_startstop_t idefloppy_do_request(ide_drive_t *drive,
 		return ide_stopped;
 	}
 
-	pc->rq = rq;
+	ide_init_sg_cmd(drive, rq);
+	ide_map_sg(drive, rq);
 
-	return idefloppy_issue_pc(drive, pc);
-}
+	pc->sg = hwif->sg_table;
+	pc->sg_cnt = hwif->sg_nents;
 
-/*
- * Add a special packet command request to the tail of the request queue,
- * and wait for it to be serviced.
- */
-static int idefloppy_queue_pc_tail(ide_drive_t *drive, struct ide_atapi_pc *pc)
-{
-	struct ide_floppy_obj *floppy = drive->driver_data;
-	struct request *rq;
-	int error;
-
-	rq = blk_get_request(drive->queue, READ, __GFP_WAIT);
-	rq->buffer = (char *) pc;
-	rq->cmd_type = REQ_TYPE_SPECIAL;
-	memcpy(rq->cmd, pc->c, 12);
-	error = blk_execute_rq(drive->queue, floppy->disk, rq, 0);
-	blk_put_request(rq);
+	pc->rq = rq;
 
-	return error;
+	return idefloppy_issue_pc(drive, pc);
 }
 
 /*
@@ -672,22 +465,28 @@ static int idefloppy_queue_pc_tail(ide_drive_t *drive, struct ide_atapi_pc *pc)
 static int ide_floppy_get_flexible_disk_page(ide_drive_t *drive)
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
+	struct gendisk *disk = floppy->disk;
 	struct ide_atapi_pc pc;
 	u8 *page;
 	int capacity, lba_capacity;
 	u16 transfer_rate, sector_size, cyls, rpm;
 	u8 heads, sectors;
 
-	idefloppy_create_mode_sense_cmd(&pc, IDEFLOPPY_FLEXIBLE_DISK_PAGE,
-					MODE_SENSE_CURRENT);
+	ide_floppy_create_mode_sense_cmd(&pc, IDEFLOPPY_FLEXIBLE_DISK_PAGE);
 
-	if (idefloppy_queue_pc_tail(drive, &pc)) {
+	if (ide_queue_pc_tail(drive, disk, &pc)) {
 		printk(KERN_ERR "ide-floppy: Can't get flexible disk page"
 				" parameters\n");
 		return 1;
 	}
-	floppy->wp = !!(pc.buf[3] & 0x80);
-	set_disk_ro(floppy->disk, floppy->wp);
+
+	if (pc.buf[3] & 0x80)
+		drive->atapi_flags |= IDE_AFLAG_WP;
+	else
+		drive->atapi_flags &= ~IDE_AFLAG_WP;
+
+	set_disk_ro(disk, !!(drive->atapi_flags & IDE_AFLAG_WP));
+
 	page = &pc.buf[8];
 
 	transfer_rate = be16_to_cpup((__be16 *)&pc.buf[8 + 2]);
@@ -721,23 +520,6 @@ static int ide_floppy_get_flexible_disk_page(ide_drive_t *drive)
 	return 0;
 }
 
-static int idefloppy_get_sfrp_bit(ide_drive_t *drive)
-{
-	idefloppy_floppy_t *floppy = drive->driver_data;
-	struct ide_atapi_pc pc;
-
-	floppy->srfp = 0;
-	idefloppy_create_mode_sense_cmd(&pc, IDEFLOPPY_CAPABILITIES_PAGE,
-						 MODE_SENSE_CURRENT);
-
-	pc.flags |= PC_FLAG_SUPPRESS_ERROR;
-	if (idefloppy_queue_pc_tail(drive, &pc))
-		return 1;
-
-	floppy->srfp = pc.buf[8 + 2] & 0x40;
-	return (0);
-}
-
 /*
  * Determine if a media is present in the floppy drive, and if so, its LBA
  * capacity.
@@ -745,6 +527,7 @@ static int idefloppy_get_sfrp_bit(ide_drive_t *drive)
 static int ide_floppy_get_capacity(ide_drive_t *drive)
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
+	struct gendisk *disk = floppy->disk;
 	struct ide_atapi_pc pc;
 	u8 *cap_desc;
 	u8 header_len, desc_cnt;
@@ -756,8 +539,8 @@ static int ide_floppy_get_capacity(ide_drive_t *drive)
 	floppy->bs_factor = 1;
 	set_capacity(floppy->disk, 0);
 
-	idefloppy_create_read_capacity_cmd(&pc);
-	if (idefloppy_queue_pc_tail(drive, &pc)) {
+	ide_floppy_create_read_capacity_cmd(&pc);
+	if (ide_queue_pc_tail(drive, disk, &pc)) {
 		printk(KERN_ERR "ide-floppy: Can't get floppy parameters\n");
 		return 1;
 	}
@@ -832,202 +615,55 @@ static int ide_floppy_get_capacity(ide_drive_t *drive)
 	if (!(drive->atapi_flags & IDE_AFLAG_CLIK_DRIVE))
 		(void) ide_floppy_get_flexible_disk_page(drive);
 
-	set_capacity(floppy->disk, floppy->blocks * floppy->bs_factor);
+	set_capacity(disk, floppy->blocks * floppy->bs_factor);
+
 	return rc;
 }
 
-/*
- * Obtain the list of formattable capacities.
- * Very similar to ide_floppy_get_capacity, except that we push the capacity
- * descriptors to userland, instead of our own structures.
- *
- * Userland gives us the following structure:
- *
- * struct idefloppy_format_capacities {
- *	int nformats;
- *	struct {
- *		int nblocks;
- *		int blocksize;
- *	} formats[];
- * };
- *
- * userland initializes nformats to the number of allocated formats[] records.
- * On exit we set nformats to the number of records we've actually initialized.
- */
-
-static int ide_floppy_get_format_capacities(ide_drive_t *drive, int __user *arg)
+static sector_t idefloppy_capacity(ide_drive_t *drive)
 {
-	struct ide_atapi_pc pc;
-	u8 header_len, desc_cnt;
-	int i, blocks, length, u_array_size, u_index;
-	int __user *argp;
-
-	if (get_user(u_array_size, arg))
-		return (-EFAULT);
-
-	if (u_array_size <= 0)
-		return (-EINVAL);
-
-	idefloppy_create_read_capacity_cmd(&pc);
-	if (idefloppy_queue_pc_tail(drive, &pc)) {
-		printk(KERN_ERR "ide-floppy: Can't get floppy parameters\n");
-		return (-EIO);
-	}
-	header_len = pc.buf[3];
-	desc_cnt = header_len / 8; /* capacity descriptor of 8 bytes */
-
-	u_index = 0;
-	argp = arg + 1;
-
-	/*
-	 * We always skip the first capacity descriptor.  That's the current
-	 * capacity.  We are interested in the remaining descriptors, the
-	 * formattable capacities.
-	 */
-	for (i = 1; i < desc_cnt; i++) {
-		unsigned int desc_start = 4 + i*8;
-
-		if (u_index >= u_array_size)
-			break;	/* User-supplied buffer too small */
-
-		blocks = be32_to_cpup((__be32 *)&pc.buf[desc_start]);
-		length = be16_to_cpup((__be16 *)&pc.buf[desc_start + 6]);
-
-		if (put_user(blocks, argp))
-			return(-EFAULT);
-		++argp;
-
-		if (put_user(length, argp))
-			return (-EFAULT);
-		++argp;
-
-		++u_index;
-	}
+	idefloppy_floppy_t *floppy = drive->driver_data;
+	unsigned long capacity = floppy->blocks * floppy->bs_factor;
 
-	if (put_user(u_index, arg))
-		return (-EFAULT);
-	return (0);
+	return capacity;
 }
 
-/*
- * Get ATAPI_FORMAT_UNIT progress indication.
- *
- * Userland gives a pointer to an int.  The int is set to a progress
- * indicator 0-65536, with 65536=100%.
- *
- * If the drive does not support format progress indication, we just check
- * the dsc bit, and return either 0 or 65536.
- */
+#ifdef CONFIG_IDE_PROC_FS
+ide_devset_rw_field(bios_cyl, bios_cyl);
+ide_devset_rw_field(bios_head, bios_head);
+ide_devset_rw_field(bios_sect, bios_sect);
 
-static int idefloppy_get_format_progress(ide_drive_t *drive, int __user *arg)
+static int get_ticks(ide_drive_t *drive)
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
-	struct ide_atapi_pc pc;
-	int progress_indication = 0x10000;
-
-	if (floppy->srfp) {
-		idefloppy_create_request_sense_cmd(&pc);
-		if (idefloppy_queue_pc_tail(drive, &pc))
-			return (-EIO);
-
-		if (floppy->sense_key == 2 &&
-		    floppy->asc == 4 &&
-		    floppy->ascq == 4)
-			progress_indication = floppy->progress_indication;
-
-		/* Else assume format_unit has finished, and we're at 0x10000 */
-	} else {
-		ide_hwif_t *hwif = drive->hwif;
-		unsigned long flags;
-		u8 stat;
-
-		local_irq_save(flags);
-		stat = hwif->tp_ops->read_status(hwif);
-		local_irq_restore(flags);
-
-		progress_indication = ((stat & SEEK_STAT) == 0) ? 0 : 0x10000;
-	}
-	if (put_user(progress_indication, arg))
-		return (-EFAULT);
-
-	return (0);
+	return floppy->ticks;
 }
 
-static sector_t idefloppy_capacity(ide_drive_t *drive)
+static int set_ticks(ide_drive_t *drive, int arg)
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
-	unsigned long capacity = floppy->blocks * floppy->bs_factor;
-
-	return capacity;
-}
-
-/*
- * Check whether we can support a drive, based on the ATAPI IDENTIFY command
- * results.
- */
-static int idefloppy_identify_device(ide_drive_t *drive, struct hd_driveid *id)
-{
-	u8 gcw[2];
-	u8 device_type, protocol, removable, drq_type, packet_size;
-
-	*((u16 *) &gcw) = id->config;
-
-	device_type =  gcw[1] & 0x1F;
-	removable   = (gcw[0] & 0x80) >> 7;
-	protocol    = (gcw[1] & 0xC0) >> 6;
-	drq_type    = (gcw[0] & 0x60) >> 5;
-	packet_size =  gcw[0] & 0x03;
-
-#ifdef CONFIG_PPC
-	/* kludge for Apple PowerBook internal zip */
-	if (device_type == 5 &&
-	    !strstr(id->model, "CD-ROM") && strstr(id->model, "ZIP"))
-		device_type = 0;
-#endif
-
-	if (protocol != 2)
-		printk(KERN_ERR "ide-floppy: Protocol (0x%02x) is not ATAPI\n",
-			protocol);
-	else if (device_type != 0)
-		printk(KERN_ERR "ide-floppy: Device type (0x%02x) is not set "
-				"to floppy\n", device_type);
-	else if (!removable)
-		printk(KERN_ERR "ide-floppy: The removable flag is not set\n");
-	else if (drq_type == 3)
-		printk(KERN_ERR "ide-floppy: Sorry, DRQ type (0x%02x) not "
-				"supported\n", drq_type);
-	else if (packet_size != 0)
-		printk(KERN_ERR "ide-floppy: Packet size (0x%02x) is not 12 "
-				"bytes\n", packet_size);
-	else
-		return 1;
+	floppy->ticks = arg;
 	return 0;
 }
 
-#ifdef CONFIG_IDE_PROC_FS
-static void idefloppy_add_settings(ide_drive_t *drive)
-{
-	idefloppy_floppy_t *floppy = drive->driver_data;
+IDE_DEVSET(ticks, DS_SYNC, get_ticks, set_ticks);
 
-	ide_add_setting(drive, "bios_cyl", SETTING_RW, TYPE_INT, 0, 1023, 1, 1,
-			&drive->bios_cyl, NULL);
-	ide_add_setting(drive, "bios_head", SETTING_RW, TYPE_BYTE, 0, 255, 1, 1,
-			&drive->bios_head, NULL);
-	ide_add_setting(drive, "bios_sect", SETTING_RW,	TYPE_BYTE, 0,  63, 1, 1,
-			&drive->bios_sect, NULL);
-	ide_add_setting(drive, "ticks",	   SETTING_RW, TYPE_BYTE, 0, 255, 1, 1,
-			&floppy->ticks,	 NULL);
-}
-#else
-static inline void idefloppy_add_settings(ide_drive_t *drive) { ; }
+static const struct ide_proc_devset idefloppy_settings[] = {
+	IDE_PROC_DEVSET(bios_cyl,  0, 1023),
+	IDE_PROC_DEVSET(bios_head, 0,  255),
+	IDE_PROC_DEVSET(bios_sect, 0,   63),
+	IDE_PROC_DEVSET(ticks,	   0,  255),
+	{ 0 },
+};
 #endif
 
 static void idefloppy_setup(ide_drive_t *drive, idefloppy_floppy_t *floppy)
 {
+	u16 *id = drive->id;
 	u8 gcw[2];
 
-	*((u16 *) &gcw) = drive->id->config;
-	floppy->pc = floppy->pc_stack;
+	*((u16 *)&gcw) = id[ATA_ID_CONFIG];
+
 	drive->pc_callback = ide_floppy_callback;
 
 	if (((gcw[0] & 0x60) >> 5) == 1)
@@ -1041,7 +677,7 @@ static void idefloppy_setup(ide_drive_t *drive, idefloppy_floppy_t *floppy)
 	 * it. It should be fixed as of version 1.9, but to be on the safe side
 	 * we'll leave the limitation below for the 2.2.x tree.
 	 */
-	if (!strncmp(drive->id->model, "IOMEGA ZIP 100 ATAPI", 20)) {
+	if (!strncmp((char *)&id[ATA_ID_PROD], "IOMEGA ZIP 100 ATAPI", 20)) {
 		drive->atapi_flags |= IDE_AFLAG_ZIP_DRIVE;
 		/* This value will be visible in the /proc/ide/hdx/settings */
 		floppy->ticks = IDEFLOPPY_TICKS_DELAY;
@@ -1052,13 +688,16 @@ static void idefloppy_setup(ide_drive_t *drive, idefloppy_floppy_t *floppy)
 	 * Guess what? The IOMEGA Clik! drive also needs the above fix. It makes
 	 * nasty clicking noises without it, so please don't remove this.
 	 */
-	if (strncmp(drive->id->model, "IOMEGA Clik!", 11) == 0) {
+	if (strncmp((char *)&id[ATA_ID_PROD], "IOMEGA Clik!", 11) == 0) {
 		blk_queue_max_sectors(drive->queue, 64);
 		drive->atapi_flags |= IDE_AFLAG_CLIK_DRIVE;
+		/* IOMEGA Clik! drives do not support lock/unlock commands */
+		drive->atapi_flags |= IDE_AFLAG_NO_DOORLOCK;
 	}
 
 	(void) ide_floppy_get_capacity(drive);
-	idefloppy_add_settings(drive);
+
+	ide_proc_register_driver(drive, floppy->driver);
 }
 
 static void ide_floppy_remove(ide_drive_t *drive)
@@ -1115,12 +754,12 @@ static ide_driver_t idefloppy_driver = {
 	.remove			= ide_floppy_remove,
 	.version		= IDEFLOPPY_VERSION,
 	.media			= ide_floppy,
-	.supports_dsc_overlap	= 0,
 	.do_request		= idefloppy_do_request,
 	.end_request		= idefloppy_end_request,
 	.error			= __ide_error,
 #ifdef CONFIG_IDE_PROC_FS
 	.proc			= idefloppy_proc,
+	.settings		= idefloppy_settings,
 #endif
 };
 
@@ -1129,7 +768,6 @@ static int idefloppy_open(struct inode *inode, struct file *filp)
 	struct gendisk *disk = inode->i_bdev->bd_disk;
 	struct ide_floppy_obj *floppy;
 	ide_drive_t *drive;
-	struct ide_atapi_pc pc;
 	int ret = 0;
 
 	debug_log("Reached %s\n", __func__);
@@ -1146,13 +784,8 @@ static int idefloppy_open(struct inode *inode, struct file *filp)
 		drive->atapi_flags &= ~IDE_AFLAG_FORMAT_IN_PROGRESS;
 		/* Just in case */
 
-		idefloppy_init_pc(&pc);
-		pc.c[0] = GPCMD_TEST_UNIT_READY;
-
-		if (idefloppy_queue_pc_tail(drive, &pc)) {
-			idefloppy_create_start_stop_cmd(&pc, 1);
-			(void) idefloppy_queue_pc_tail(drive, &pc);
-		}
+		if (ide_do_test_unit_ready(drive, disk))
+			ide_do_start_stop(drive, disk, 1);
 
 		if (ide_floppy_get_capacity(drive)
 		   && (filp->f_flags & O_NDELAY) == 0
@@ -1166,16 +799,13 @@ static int idefloppy_open(struct inode *inode, struct file *filp)
 			goto out_put_floppy;
 		}
 
-		if (floppy->wp && (filp->f_mode & 2)) {
+		if ((drive->atapi_flags & IDE_AFLAG_WP) && (filp->f_mode & 2)) {
 			ret = -EROFS;
 			goto out_put_floppy;
 		}
+
 		drive->atapi_flags |= IDE_AFLAG_MEDIA_CHANGED;
-		/* IOMEGA Clik! drives do not support lock/unlock commands */
-		if (!(drive->atapi_flags & IDE_AFLAG_CLIK_DRIVE)) {
-			idefloppy_create_prevent_cmd(&pc, 1);
-			(void) idefloppy_queue_pc_tail(drive, &pc);
-		}
+		ide_set_media_lock(drive, disk, 1);
 		check_disk_change(inode->i_bdev);
 	} else if (drive->atapi_flags & IDE_AFLAG_FORMAT_IN_PROGRESS) {
 		ret = -EBUSY;
@@ -1194,17 +824,11 @@ static int idefloppy_release(struct inode *inode, struct file *filp)
 	struct gendisk *disk = inode->i_bdev->bd_disk;
 	struct ide_floppy_obj *floppy = ide_floppy_g(disk);
 	ide_drive_t *drive = floppy->drive;
-	struct ide_atapi_pc pc;
 
 	debug_log("Reached %s\n", __func__);
 
 	if (floppy->openers == 1) {
-		/* IOMEGA Clik! drives do not support lock/unlock commands */
-		if (!(drive->atapi_flags & IDE_AFLAG_CLIK_DRIVE)) {
-			idefloppy_create_prevent_cmd(&pc, 0);
-			(void) idefloppy_queue_pc_tail(drive, &pc);
-		}
-
+		ide_set_media_lock(drive, disk, 0);
 		drive->atapi_flags &= ~IDE_AFLAG_FORMAT_IN_PROGRESS;
 	}
 
@@ -1230,80 +854,20 @@ static int ide_floppy_lockdoor(ide_drive_t *drive, struct ide_atapi_pc *pc,
 			       unsigned long arg, unsigned int cmd)
 {
 	idefloppy_floppy_t *floppy = drive->driver_data;
+	struct gendisk *disk = floppy->disk;
+	int prevent = (arg && cmd != CDROMEJECT) ? 1 : 0;
 
 	if (floppy->openers > 1)
 		return -EBUSY;
 
-	/* The IOMEGA Clik! Drive doesn't support this command -
-	 * no room for an eject mechanism */
-	if (!(drive->atapi_flags & IDE_AFLAG_CLIK_DRIVE)) {
-		int prevent = arg ? 1 : 0;
-
-		if (cmd == CDROMEJECT)
-			prevent = 0;
+	ide_set_media_lock(drive, disk, prevent);
 
-		idefloppy_create_prevent_cmd(pc, prevent);
-		(void) idefloppy_queue_pc_tail(floppy->drive, pc);
-	}
-
-	if (cmd == CDROMEJECT) {
-		idefloppy_create_start_stop_cmd(pc, 2);
-		(void) idefloppy_queue_pc_tail(floppy->drive, pc);
-	}
+	if (cmd == CDROMEJECT)
+		ide_do_start_stop(drive, disk, 2);
 
 	return 0;
 }
 
-static int ide_floppy_format_unit(idefloppy_floppy_t *floppy,
-				  int __user *arg)
-{
-	struct ide_atapi_pc pc;
-	ide_drive_t *drive = floppy->drive;
-	int blocks, length, flags, err = 0;
-
-	if (floppy->openers > 1) {
-		/* Don't format if someone is using the disk */
-		drive->atapi_flags &= ~IDE_AFLAG_FORMAT_IN_PROGRESS;
-		return -EBUSY;
-	}
-
-	drive->atapi_flags |= IDE_AFLAG_FORMAT_IN_PROGRESS;
-
-	/*
-	 * Send ATAPI_FORMAT_UNIT to the drive.
-	 *
-	 * Userland gives us the following structure:
-	 *
-	 * struct idefloppy_format_command {
-	 *        int nblocks;
-	 *        int blocksize;
-	 *        int flags;
-	 *        } ;
-	 *
-	 * flags is a bitmask, currently, the only defined flag is:
-	 *
-	 *        0x01 - verify media after format.
-	 */
-	if (get_user(blocks, arg) ||
-			get_user(length, arg+1) ||
-			get_user(flags, arg+2)) {
-		err = -EFAULT;
-		goto out;
-	}
-
-	(void) idefloppy_get_sfrp_bit(drive);
-	idefloppy_create_format_unit_cmd(&pc, blocks, length, flags);
-
-	if (idefloppy_queue_pc_tail(drive, &pc))
-		err = -EIO;
-
-out:
-	if (err)
-		drive->atapi_flags &= ~IDE_AFLAG_FORMAT_IN_PROGRESS;
-	return err;
-}
-
-
 static int idefloppy_ioctl(struct inode *inode, struct file *file,
 			unsigned int cmd, unsigned long arg)
 {
@@ -1314,23 +878,12 @@ static int idefloppy_ioctl(struct inode *inode, struct file *file,
 	void __user *argp = (void __user *)arg;
 	int err;
 
-	switch (cmd) {
-	case CDROMEJECT:
-		/* fall through */
-	case CDROM_LOCKDOOR:
+	if (cmd == CDROMEJECT || cmd == CDROM_LOCKDOOR)
 		return ide_floppy_lockdoor(drive, &pc, arg, cmd);
-	case IDEFLOPPY_IOCTL_FORMAT_SUPPORTED:
-		return 0;
-	case IDEFLOPPY_IOCTL_FORMAT_GET_CAPACITY:
-		return ide_floppy_get_format_capacities(drive, argp);
-	case IDEFLOPPY_IOCTL_FORMAT_START:
-		if (!(file->f_mode & 2))
-			return -EPERM;
-
-		return ide_floppy_format_unit(floppy, (int __user *)arg);
-	case IDEFLOPPY_IOCTL_FORMAT_GET_PROGRESS:
-		return idefloppy_get_format_progress(drive, argp);
-	}
+
+	err = ide_floppy_format_ioctl(drive, file, cmd, argp);
+	if (err != -ENOTTY)
+		return err;
 
 	/*
 	 * skip SCSI_IOCTL_SEND_COMMAND (deprecated)
@@ -1339,8 +892,6 @@ static int idefloppy_ioctl(struct inode *inode, struct file *file,
 	if (cmd != CDROM_SEND_PACKET && cmd != SCSI_IOCTL_SEND_COMMAND)
 		err = scsi_cmd_ioctl(file, bdev->bd_disk->queue,
 					bdev->bd_disk, cmd, argp);
-	else
-		err = -ENOTTY;
 
 	if (err == -ENOTTY)
 		err = generic_ide_ioctl(drive, file, bdev, cmd, arg);
@@ -1388,11 +939,11 @@ static int ide_floppy_probe(ide_drive_t *drive)
 
 	if (!strstr("ide-floppy", drive->driver_req))
 		goto failed;
-	if (!drive->present)
-		goto failed;
+
 	if (drive->media != ide_floppy)
 		goto failed;
-	if (!idefloppy_identify_device(drive, drive->id)) {
+
+	if (!ide_check_atapi_device(drive, DRV_NAME)) {
 		printk(KERN_ERR "ide-floppy: %s: not supported by this version"
 				" of ide-floppy\n", drive->name);
 		goto failed;
@@ -1410,8 +961,6 @@ static int ide_floppy_probe(ide_drive_t *drive)
 
 	ide_init_disk(g, drive);
 
-	ide_proc_register_driver(drive, &idefloppy_driver);
-
 	kref_init(&floppy->kref);
 
 	floppy->drive = drive;
@@ -1450,6 +999,7 @@ static int __init idefloppy_init(void)
 }
 
 MODULE_ALIAS("ide:*m-floppy*");
+MODULE_ALIAS("ide-floppy");
 module_init(idefloppy_init);
 module_exit(idefloppy_exit);
 MODULE_LICENSE("GPL");
diff --git a/drivers/ide/ide-floppy.h b/drivers/ide/ide-floppy.h
new file mode 100644
index 000000000000..ecadc2bc322d
--- /dev/null
+++ b/drivers/ide/ide-floppy.h
@@ -0,0 +1,63 @@
+#ifndef __IDE_FLOPPY_H
+#define __IDE_FLOPPY_H
+
+/*
+ * Most of our global data which we need to save even as we leave the driver
+ * due to an interrupt or a timer event is stored in a variable of type
+ * idefloppy_floppy_t, defined below.
+ */
+typedef struct ide_floppy_obj {
+	ide_drive_t	*drive;
+	ide_driver_t	*driver;
+	struct gendisk	*disk;
+	struct kref	kref;
+	unsigned int	openers;	/* protected by BKL for now */
+
+	/* Current packet command */
+	struct ide_atapi_pc *pc;
+	/* Last failed packet command */
+	struct ide_atapi_pc *failed_pc;
+	/* used for blk_{fs,pc}_request() requests */
+	struct ide_atapi_pc queued_pc;
+
+	struct ide_atapi_pc request_sense_pc;
+	struct request request_sense_rq;
+
+	/* Last error information */
+	u8 sense_key, asc, ascq;
+	/* delay this long before sending packet command */
+	u8 ticks;
+	int progress_indication;
+
+	/* Device information */
+	/* Current format */
+	int blocks, block_size, bs_factor;
+	/* Last format capacity descriptor */
+	u8 cap_desc[8];
+	/* Copy of the flexible disk page */
+	u8 flexible_disk_page[32];
+} idefloppy_floppy_t;
+
+/*
+ * Pages of the SELECT SENSE / MODE SENSE packet commands.
+ * See SFF-8070i spec.
+ */
+#define	IDEFLOPPY_CAPABILITIES_PAGE	0x1b
+#define IDEFLOPPY_FLEXIBLE_DISK_PAGE	0x05
+
+/* IOCTLs used in low-level formatting. */
+#define	IDEFLOPPY_IOCTL_FORMAT_SUPPORTED	0x4600
+#define	IDEFLOPPY_IOCTL_FORMAT_GET_CAPACITY	0x4601
+#define	IDEFLOPPY_IOCTL_FORMAT_START		0x4602
+#define IDEFLOPPY_IOCTL_FORMAT_GET_PROGRESS	0x4603
+
+/* ide-floppy.c */
+void ide_floppy_create_mode_sense_cmd(struct ide_atapi_pc *, u8);
+void ide_floppy_create_read_capacity_cmd(struct ide_atapi_pc *);
+void ide_floppy_create_request_sense_cmd(struct ide_atapi_pc *);
+
+/* ide-floppy_ioctl.c */
+int ide_floppy_format_ioctl(ide_drive_t *, struct file *, unsigned int,
+			    void __user *);
+
+#endif /*__IDE_FLOPPY_H */
diff --git a/drivers/ide/ide-floppy_ioctl.c b/drivers/ide/ide-floppy_ioctl.c
new file mode 100644
index 000000000000..5ffc4512d14b
--- /dev/null
+++ b/drivers/ide/ide-floppy_ioctl.c
@@ -0,0 +1,243 @@
+/*
+ * ide-floppy IOCTLs handling.
+ */
+
+#include <linux/kernel.h>
+#include <linux/ide.h>
+#include <linux/cdrom.h>
+
+#include <asm/unaligned.h>
+
+#include <scsi/scsi_ioctl.h>
+
+#include "ide-floppy.h"
+
+/*
+ * Obtain the list of formattable capacities.
+ * Very similar to ide_floppy_get_capacity, except that we push the capacity
+ * descriptors to userland, instead of our own structures.
+ *
+ * Userland gives us the following structure:
+ *
+ * struct idefloppy_format_capacities {
+ *	int nformats;
+ *	struct {
+ *		int nblocks;
+ *		int blocksize;
+ *	} formats[];
+ * };
+ *
+ * userland initializes nformats to the number of allocated formats[] records.
+ * On exit we set nformats to the number of records we've actually initialized.
+ */
+
+static int ide_floppy_get_format_capacities(ide_drive_t *drive, int __user *arg)
+{
+	struct ide_floppy_obj *floppy = drive->driver_data;
+	struct ide_atapi_pc pc;
+	u8 header_len, desc_cnt;
+	int i, blocks, length, u_array_size, u_index;
+	int __user *argp;
+
+	if (get_user(u_array_size, arg))
+		return -EFAULT;
+
+	if (u_array_size <= 0)
+		return -EINVAL;
+
+	ide_floppy_create_read_capacity_cmd(&pc);
+	if (ide_queue_pc_tail(drive, floppy->disk, &pc)) {
+		printk(KERN_ERR "ide-floppy: Can't get floppy parameters\n");
+		return -EIO;
+	}
+
+	header_len = pc.buf[3];
+	desc_cnt = header_len / 8; /* capacity descriptor of 8 bytes */
+
+	u_index = 0;
+	argp = arg + 1;
+
+	/*
+	 * We always skip the first capacity descriptor.  That's the current
+	 * capacity.  We are interested in the remaining descriptors, the
+	 * formattable capacities.
+	 */
+	for (i = 1; i < desc_cnt; i++) {
+		unsigned int desc_start = 4 + i*8;
+
+		if (u_index >= u_array_size)
+			break;	/* User-supplied buffer too small */
+
+		blocks = be32_to_cpup((__be32 *)&pc.buf[desc_start]);
+		length = be16_to_cpup((__be16 *)&pc.buf[desc_start + 6]);
+
+		if (put_user(blocks, argp))
+			return -EFAULT;
+
+		++argp;
+
+		if (put_user(length, argp))
+			return -EFAULT;
+
+		++argp;
+
+		++u_index;
+	}
+
+	if (put_user(u_index, arg))
+		return -EFAULT;
+
+	return 0;
+}
+
+static void ide_floppy_create_format_unit_cmd(struct ide_atapi_pc *pc, int b,
+		int l, int flags)
+{
+	ide_init_pc(pc);
+	pc->c[0] = GPCMD_FORMAT_UNIT;
+	pc->c[1] = 0x17;
+
+	memset(pc->buf, 0, 12);
+	pc->buf[1] = 0xA2;
+	/* Default format list header, u8 1: FOV/DCRT/IMM bits set */
+
+	if (flags & 1)				/* Verify bit on... */
+		pc->buf[1] ^= 0x20;		/* ... turn off DCRT bit */
+	pc->buf[3] = 8;
+
+	put_unaligned(cpu_to_be32(b), (unsigned int *)(&pc->buf[4]));
+	put_unaligned(cpu_to_be32(l), (unsigned int *)(&pc->buf[8]));
+	pc->buf_size = 12;
+	pc->flags |= PC_FLAG_WRITING;
+}
+
+static int ide_floppy_get_sfrp_bit(ide_drive_t *drive)
+{
+	idefloppy_floppy_t *floppy = drive->driver_data;
+	struct ide_atapi_pc pc;
+
+	drive->atapi_flags &= ~IDE_AFLAG_SRFP;
+
+	ide_floppy_create_mode_sense_cmd(&pc, IDEFLOPPY_CAPABILITIES_PAGE);
+	pc.flags |= PC_FLAG_SUPPRESS_ERROR;
+
+	if (ide_queue_pc_tail(drive, floppy->disk, &pc))
+		return 1;
+
+	if (pc.buf[8 + 2] & 0x40)
+		drive->atapi_flags |= IDE_AFLAG_SRFP;
+
+	return 0;
+}
+
+static int ide_floppy_format_unit(ide_drive_t *drive, int __user *arg)
+{
+	idefloppy_floppy_t *floppy = drive->driver_data;
+	struct ide_atapi_pc pc;
+	int blocks, length, flags, err = 0;
+
+	if (floppy->openers > 1) {
+		/* Don't format if someone is using the disk */
+		drive->atapi_flags &= ~IDE_AFLAG_FORMAT_IN_PROGRESS;
+		return -EBUSY;
+	}
+
+	drive->atapi_flags |= IDE_AFLAG_FORMAT_IN_PROGRESS;
+
+	/*
+	 * Send ATAPI_FORMAT_UNIT to the drive.
+	 *
+	 * Userland gives us the following structure:
+	 *
+	 * struct idefloppy_format_command {
+	 *        int nblocks;
+	 *        int blocksize;
+	 *        int flags;
+	 *        } ;
+	 *
+	 * flags is a bitmask, currently, the only defined flag is:
+	 *
+	 *        0x01 - verify media after format.
+	 */
+	if (get_user(blocks, arg) ||
+			get_user(length, arg+1) ||
+			get_user(flags, arg+2)) {
+		err = -EFAULT;
+		goto out;
+	}
+
+	(void)ide_floppy_get_sfrp_bit(drive);
+	ide_floppy_create_format_unit_cmd(&pc, blocks, length, flags);
+
+	if (ide_queue_pc_tail(drive, floppy->disk, &pc))
+		err = -EIO;
+
+out:
+	if (err)
+		drive->atapi_flags &= ~IDE_AFLAG_FORMAT_IN_PROGRESS;
+	return err;
+}
+
+/*
+ * Get ATAPI_FORMAT_UNIT progress indication.
+ *
+ * Userland gives a pointer to an int.  The int is set to a progress
+ * indicator 0-65536, with 65536=100%.
+ *
+ * If the drive does not support format progress indication, we just check
+ * the dsc bit, and return either 0 or 65536.
+ */
+
+static int ide_floppy_get_format_progress(ide_drive_t *drive, int __user *arg)
+{
+	idefloppy_floppy_t *floppy = drive->driver_data;
+	struct ide_atapi_pc pc;
+	int progress_indication = 0x10000;
+
+	if (drive->atapi_flags & IDE_AFLAG_SRFP) {
+		ide_floppy_create_request_sense_cmd(&pc);
+		if (ide_queue_pc_tail(drive, floppy->disk, &pc))
+			return -EIO;
+
+		if (floppy->sense_key == 2 &&
+		    floppy->asc == 4 &&
+		    floppy->ascq == 4)
+			progress_indication = floppy->progress_indication;
+
+		/* Else assume format_unit has finished, and we're at 0x10000 */
+	} else {
+		ide_hwif_t *hwif = drive->hwif;
+		unsigned long flags;
+		u8 stat;
+
+		local_irq_save(flags);
+		stat = hwif->tp_ops->read_status(hwif);
+		local_irq_restore(flags);
+
+		progress_indication = ((stat & ATA_DSC) == 0) ? 0 : 0x10000;
+	}
+
+	if (put_user(progress_indication, arg))
+		return -EFAULT;
+
+	return 0;
+}
+
+int ide_floppy_format_ioctl(ide_drive_t *drive, struct file *file,
+			    unsigned int cmd, void __user *argp)
+{
+	switch (cmd) {
+	case IDEFLOPPY_IOCTL_FORMAT_SUPPORTED:
+		return 0;
+	case IDEFLOPPY_IOCTL_FORMAT_GET_CAPACITY:
+		return ide_floppy_get_format_capacities(drive, argp);
+	case IDEFLOPPY_IOCTL_FORMAT_START:
+		if (!(file->f_mode & 2))
+			return -EPERM;
+		return ide_floppy_format_unit(drive, (int __user *)argp);
+	case IDEFLOPPY_IOCTL_FORMAT_GET_PROGRESS:
+		return ide_floppy_get_format_progress(drive, argp);
+	default:
+		return -ENOTTY;
+	}
+}
diff --git a/drivers/ide/ide-generic.c b/drivers/ide/ide-generic.c
index 8fe8b5b9cf7d..0a3cb0c33ae5 100644
--- a/drivers/ide/ide-generic.c
+++ b/drivers/ide/ide-generic.c
@@ -19,6 +19,7 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/ide.h>
+#include <linux/pci_ids.h>
 
 /* FIXME: convert m32r to use ide_platform host driver */
 #ifdef CONFIG_M32R
@@ -27,7 +28,7 @@
 
 #define DRV_NAME	"ide_generic"
 
-static int probe_mask = 0x03;
+static int probe_mask;
 module_param(probe_mask, int, 0);
 MODULE_PARM_DESC(probe_mask, "probe mask for legacy ISA IDE ports");
 
@@ -100,19 +101,65 @@ static const u16 legacy_bases[] = { 0x1f0, 0x170, 0x1e8, 0x168, 0x1e0, 0x160 };
 static const int legacy_irqs[]  = { 14, 15, 11, 10, 8, 12 };
 #endif
 
+static void ide_generic_check_pci_legacy_iobases(int *primary, int *secondary)
+{
+	struct pci_dev *p = NULL;
+	u16 val;
+
+	for_each_pci_dev(p) {
+
+		if (pci_resource_start(p, 0) == 0x1f0)
+			*primary = 1;
+		if (pci_resource_start(p, 2) == 0x170)
+			*secondary = 1;
+
+		/* Cyrix CS55{1,2}0 pre SFF MWDMA ATA on the bridge */
+		if (p->vendor == PCI_VENDOR_ID_CYRIX &&
+		    (p->device == PCI_DEVICE_ID_CYRIX_5510 ||
+		     p->device == PCI_DEVICE_ID_CYRIX_5520))
+			*primary = *secondary = 1;
+
+		/* Intel MPIIX - PIO ATA on non PCI side of bridge */
+		if (p->vendor == PCI_VENDOR_ID_INTEL &&
+		    p->device == PCI_DEVICE_ID_INTEL_82371MX) {
+
+			pci_read_config_word(p, 0x6C, &val);
+			if (val & 0x8000) {
+				/* ATA port enabled */
+				if (val & 0x4000)
+					*secondary = 1;
+				else
+					*primary = 1;
+			}
+		}
+	}
+}
+
 static int __init ide_generic_init(void)
 {
 	hw_regs_t hw[MAX_HWIFS], *hws[MAX_HWIFS];
 	struct ide_host *host;
 	unsigned long io_addr;
-	int i, rc;
+	int i, rc, primary = 0, secondary = 0;
 
 #ifdef CONFIG_MIPS
 	if (!ide_probe_legacy())
 		return -ENODEV;
 #endif
-	printk(KERN_INFO DRV_NAME ": please use \"probe_mask=0x3f\" module "
-			 "parameter for probing all legacy ISA IDE ports\n");
+	ide_generic_check_pci_legacy_iobases(&primary, &secondary);
+
+	if (!probe_mask) {
+		printk(KERN_INFO DRV_NAME ": please use \"probe_mask=0x3f\" "
+		     "module parameter for probing all legacy ISA IDE ports\n");
+
+		if (primary == 0)
+			probe_mask |= 0x1;
+
+		if (secondary == 0)
+			probe_mask |= 0x2;
+	} else
+		printk(KERN_INFO DRV_NAME ": enforcing probing of I/O ports "
+			"upon user request\n");
 
 	memset(hws, 0, sizeof(hw_regs_t *) * MAX_HWIFS);
 
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index a896a283f27f..1c51949833be 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -40,6 +40,7 @@
 #include <linux/pci.h>
 #include <linux/delay.h>
 #include <linux/ide.h>
+#include <linux/hdreg.h>
 #include <linux/completion.h>
 #include <linux/reboot.h>
 #include <linux/cdrom.h>
@@ -183,18 +184,18 @@ static ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *
 		if (drive->media != ide_disk)
 			break;
 		/* Not supported? Switch to next step now. */
-		if (!drive->wcache || !ide_id_has_flush_cache(drive->id)) {
+		if (!drive->wcache || ata_id_flush_enabled(drive->id) == 0) {
 			ide_complete_power_step(drive, rq, 0, 0);
 			return ide_stopped;
 		}
-		if (ide_id_has_flush_cache_ext(drive->id))
-			args->tf.command = WIN_FLUSH_CACHE_EXT;
+		if (ata_id_flush_ext_enabled(drive->id))
+			args->tf.command = ATA_CMD_FLUSH_EXT;
 		else
-			args->tf.command = WIN_FLUSH_CACHE;
+			args->tf.command = ATA_CMD_FLUSH;
 		goto out_do_tf;
 
 	case idedisk_pm_standby:	/* Suspend step 2 (standby) */
-		args->tf.command = WIN_STANDBYNOW1;
+		args->tf.command = ATA_CMD_STANDBYNOW1;
 		goto out_do_tf;
 
 	case idedisk_pm_restore_pio:	/* Resume step 1 (restore PIO) */
@@ -209,7 +210,7 @@ static ide_startstop_t ide_start_power_step(ide_drive_t *drive, struct request *
 		return ide_stopped;
 
 	case idedisk_pm_idle:		/* Resume step 2 (idle) */
-		args->tf.command = WIN_IDLEIMMEDIATE;
+		args->tf.command = ATA_CMD_IDLEIMMEDIATE;
 		goto out_do_tf;
 
 	case ide_pm_restore_dma:	/* Resume step 3 (restore DMA) */
@@ -322,7 +323,7 @@ void ide_end_drive_cmd (ide_drive_t *drive, u8 stat, u8 err)
 		ide_task_t *task = (ide_task_t *)rq->special;
 
 		if (rq->errors == 0)
-			rq->errors = !OK_STAT(stat, READY_STAT, BAD_STAT);
+			rq->errors = !OK_STAT(stat, ATA_DRDY, BAD_STAT);
 
 		if (task) {
 			struct ide_taskfile *tf = &task->tf;
@@ -373,29 +374,29 @@ static ide_startstop_t ide_ata_error(ide_drive_t *drive, struct request *rq, u8
 {
 	ide_hwif_t *hwif = drive->hwif;
 
-	if (stat & BUSY_STAT || ((stat & WRERR_STAT) && !drive->nowerr)) {
+	if ((stat & ATA_BUSY) || ((stat & ATA_DF) && !drive->nowerr)) {
 		/* other bits are useless when BUSY */
 		rq->errors |= ERROR_RESET;
-	} else if (stat & ERR_STAT) {
+	} else if (stat & ATA_ERR) {
 		/* err has different meaning on cdrom and tape */
-		if (err == ABRT_ERR) {
+		if (err == ATA_ABORTED) {
 			if (drive->select.b.lba &&
-			    /* some newer drives don't support WIN_SPECIFY */
-			    hwif->tp_ops->read_status(hwif) == WIN_SPECIFY)
+			    /* some newer drives don't support ATA_CMD_INIT_DEV_PARAMS */
+			    hwif->tp_ops->read_status(hwif) == ATA_CMD_INIT_DEV_PARAMS)
 				return ide_stopped;
 		} else if ((err & BAD_CRC) == BAD_CRC) {
 			/* UDMA crc error, just retry the operation */
 			drive->crc_count++;
-		} else if (err & (BBD_ERR | ECC_ERR)) {
+		} else if (err & (ATA_BBK | ATA_UNC)) {
 			/* retries won't help these */
 			rq->errors = ERROR_MAX;
-		} else if (err & TRK0_ERR) {
+		} else if (err & ATA_TRK0NF) {
 			/* help it find track zero */
 			rq->errors |= ERROR_RECAL;
 		}
 	}
 
-	if ((stat & DRQ_STAT) && rq_data_dir(rq) == READ &&
+	if ((stat & ATA_DRQ) && rq_data_dir(rq) == READ &&
 	    (hwif->host_flags & IDE_HFLAG_ERROR_STOPS_FIFO) == 0) {
 		int nsect = drive->mult_count ? drive->mult_count : 1;
 
@@ -407,7 +408,7 @@ static ide_startstop_t ide_ata_error(ide_drive_t *drive, struct request *rq, u8
 		return ide_stopped;
 	}
 
-	if (hwif->tp_ops->read_status(hwif) & (BUSY_STAT | DRQ_STAT))
+	if (hwif->tp_ops->read_status(hwif) & (ATA_BUSY | ATA_DRQ))
 		rq->errors |= ERROR_RESET;
 
 	if ((rq->errors & ERROR_RESET) == ERROR_RESET) {
@@ -427,16 +428,16 @@ static ide_startstop_t ide_atapi_error(ide_drive_t *drive, struct request *rq, u
 {
 	ide_hwif_t *hwif = drive->hwif;
 
-	if (stat & BUSY_STAT || ((stat & WRERR_STAT) && !drive->nowerr)) {
+	if ((stat & ATA_BUSY) || ((stat & ATA_DF) && !drive->nowerr)) {
 		/* other bits are useless when BUSY */
 		rq->errors |= ERROR_RESET;
 	} else {
 		/* add decoding error stuff */
 	}
 
-	if (hwif->tp_ops->read_status(hwif) & (BUSY_STAT | DRQ_STAT))
+	if (hwif->tp_ops->read_status(hwif) & (ATA_BUSY | ATA_DRQ))
 		/* force an abort */
-		hwif->tp_ops->exec_command(hwif, WIN_IDLEIMMEDIATE);
+		hwif->tp_ops->exec_command(hwif, ATA_CMD_IDLEIMMEDIATE);
 
 	if (rq->errors >= ERROR_MAX) {
 		ide_kill_rq(drive, rq);
@@ -509,19 +510,19 @@ static void ide_tf_set_specify_cmd(ide_drive_t *drive, struct ide_taskfile *tf)
 	tf->lbam    = drive->cyl;
 	tf->lbah    = drive->cyl >> 8;
 	tf->device  = ((drive->head - 1) | drive->select.all) & ~ATA_LBA;
-	tf->command = WIN_SPECIFY;
+	tf->command = ATA_CMD_INIT_DEV_PARAMS;
 }
 
 static void ide_tf_set_restore_cmd(ide_drive_t *drive, struct ide_taskfile *tf)
 {
 	tf->nsect   = drive->sect;
-	tf->command = WIN_RESTORE;
+	tf->command = ATA_CMD_RESTORE;
 }
 
 static void ide_tf_set_setmult_cmd(ide_drive_t *drive, struct ide_taskfile *tf)
 {
 	tf->nsect   = drive->mult_req;
-	tf->command = WIN_SETMULT;
+	tf->command = ATA_CMD_SET_MULTI;
 }
 
 static ide_startstop_t ide_disk_special(ide_drive_t *drive)
@@ -540,8 +541,6 @@ static ide_startstop_t ide_disk_special(ide_drive_t *drive)
 		ide_tf_set_restore_cmd(drive, &args.tf);
 	} else if (s->b.set_multmode) {
 		s->b.set_multmode = 0;
-		if (drive->mult_req > drive->id->max_multsect)
-			drive->mult_req = drive->id->max_multsect;
 		ide_tf_set_setmult_cmd(drive, &args.tf);
 	} else if (s->all) {
 		int special = s->all;
@@ -586,9 +585,10 @@ static int set_pio_mode_abuse(ide_hwif_t *hwif, u8 req_pio)
  *	do_special		-	issue some special commands
  *	@drive: drive the command is for
  *
- *	do_special() is used to issue WIN_SPECIFY, WIN_RESTORE, and WIN_SETMULT
- *	commands to a drive.  It used to do much more, but has been scaled
- *	back.
+ *	do_special() is used to issue ATA_CMD_INIT_DEV_PARAMS,
+ *	ATA_CMD_RESTORE and ATA_CMD_SET_MULTI commands to a drive.
+ *
+ *	It used to do much more, but has been scaled back.
  */
 
 static ide_startstop_t do_special (ide_drive_t *drive)
@@ -716,9 +716,49 @@ static ide_startstop_t execute_drive_cmd (ide_drive_t *drive,
  	return ide_stopped;
 }
 
+int ide_devset_execute(ide_drive_t *drive, const struct ide_devset *setting,
+		       int arg)
+{
+	struct request_queue *q = drive->queue;
+	struct request *rq;
+	int ret = 0;
+
+	if (!(setting->flags & DS_SYNC))
+		return setting->set(drive, arg);
+
+	rq = blk_get_request(q, READ, GFP_KERNEL);
+	if (!rq)
+		return -ENOMEM;
+
+	rq->cmd_type = REQ_TYPE_SPECIAL;
+	rq->cmd_len = 5;
+	rq->cmd[0] = REQ_DEVSET_EXEC;
+	*(int *)&rq->cmd[1] = arg;
+	rq->special = setting->set;
+
+	if (blk_execute_rq(q, NULL, rq, 0))
+		ret = rq->errors;
+	blk_put_request(rq);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(ide_devset_execute);
+
 static ide_startstop_t ide_special_rq(ide_drive_t *drive, struct request *rq)
 {
 	switch (rq->cmd[0]) {
+	case REQ_DEVSET_EXEC:
+	{
+		int err, (*setfunc)(ide_drive_t *, int) = rq->special;
+
+		err = setfunc(drive, *(int *)&rq->cmd[1]);
+		if (err)
+			rq->errors = err;
+		else
+			err = 1;
+		ide_end_request(drive, err, 0);
+		return ide_stopped;
+	}
 	case REQ_DRIVE_RESET:
 		return ide_do_reset(drive);
 	default:
@@ -766,9 +806,7 @@ static void ide_check_pm_state(ide_drive_t *drive, struct request *rq)
  *	start_request	-	start of I/O and command issuing for IDE
  *
  *	start_request() initiates handling of a new I/O request. It
- *	accepts commands and I/O (read/write) requests. It also does
- *	the final remapping for weird stuff like EZDrive. Once 
- *	device mapper can work sector level the EZDrive stuff can go away
+ *	accepts commands and I/O (read/write) requests.
  *
  *	FIXME: this function needs a rename
  */
@@ -776,7 +814,6 @@ static void ide_check_pm_state(ide_drive_t *drive, struct request *rq)
 static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 {
 	ide_startstop_t startstop;
-	sector_t block;
 
 	BUG_ON(!blk_rq_started(rq));
 
@@ -791,21 +828,12 @@ static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 		goto kill_rq;
 	}
 
-	block    = rq->sector;
-	if (blk_fs_request(rq) &&
-	    (drive->media == ide_disk || drive->media == ide_floppy)) {
-		block += drive->sect0;
-	}
-	/* Yecch - this will shift the entire interval,
-	   possibly killing some innocent following sector */
-	if (block == 0 && drive->remap_0_to_1 == 1)
-		block = 1;  /* redirect MBR access to EZ-Drive partn table */
-
 	if (blk_pm_request(rq))
 		ide_check_pm_state(drive, rq);
 
 	SELECT_DRIVE(drive);
-	if (ide_wait_stat(&startstop, drive, drive->ready_stat, BUSY_STAT|DRQ_STAT, WAIT_READY)) {
+	if (ide_wait_stat(&startstop, drive, drive->ready_stat,
+			  ATA_BUSY | ATA_DRQ, WAIT_READY)) {
 		printk(KERN_ERR "%s: drive not ready for command\n", drive->name);
 		return startstop;
 	}
@@ -844,7 +872,8 @@ static ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 			return ide_special_rq(drive, rq);
 
 		drv = *(ide_driver_t **)rq->rq_disk->private_data;
-		return drv->do_request(drive, rq, block);
+
+		return drv->do_request(drive, rq, rq->sector);
 	}
 	return do_special(drive);
 kill_rq:
@@ -1325,7 +1354,7 @@ static void unexpected_intr (int irq, ide_hwgroup_t *hwgroup)
 		if (hwif->irq == irq) {
 			stat = hwif->tp_ops->read_status(hwif);
 
-			if (!OK_STAT(stat, READY_STAT, BAD_STAT)) {
+			if (!OK_STAT(stat, ATA_DRDY, BAD_STAT)) {
 				/* Try to not flood the console with msgs */
 				static unsigned long last_msgtime, count;
 				++count;
diff --git a/drivers/ide/ide-ioctls.c b/drivers/ide/ide-ioctls.c
new file mode 100644
index 000000000000..cf01564901af
--- /dev/null
+++ b/drivers/ide/ide-ioctls.c
@@ -0,0 +1,290 @@
+/*
+ * IDE ioctls handling.
+ */
+
+#include <linux/hdreg.h>
+#include <linux/ide.h>
+
+static const struct ide_ioctl_devset ide_ioctl_settings[] = {
+{ HDIO_GET_32BIT,	 HDIO_SET_32BIT,	&ide_devset_io_32bit  },
+{ HDIO_GET_KEEPSETTINGS, HDIO_SET_KEEPSETTINGS,	&ide_devset_keepsettings },
+{ HDIO_GET_UNMASKINTR,	 HDIO_SET_UNMASKINTR,	&ide_devset_unmaskirq },
+{ HDIO_GET_DMA,		 HDIO_SET_DMA,		&ide_devset_using_dma },
+{ -1,			 HDIO_SET_PIO_MODE,	&ide_devset_pio_mode  },
+{ 0 }
+};
+
+int ide_setting_ioctl(ide_drive_t *drive, struct block_device *bdev,
+		      unsigned int cmd, unsigned long arg,
+		      const struct ide_ioctl_devset *s)
+{
+	const struct ide_devset *ds;
+	unsigned long flags;
+	int err = -EOPNOTSUPP;
+
+	for (; (ds = s->setting); s++) {
+		if (ds->get && s->get_ioctl == cmd)
+			goto read_val;
+		else if (ds->set && s->set_ioctl == cmd)
+			goto set_val;
+	}
+
+	return err;
+
+read_val:
+	mutex_lock(&ide_setting_mtx);
+	spin_lock_irqsave(&ide_lock, flags);
+	err = ds->get(drive);
+	spin_unlock_irqrestore(&ide_lock, flags);
+	mutex_unlock(&ide_setting_mtx);
+	return err >= 0 ? put_user(err, (long __user *)arg) : err;
+
+set_val:
+	if (bdev != bdev->bd_contains)
+		err = -EINVAL;
+	else {
+		if (!capable(CAP_SYS_ADMIN))
+			err = -EACCES;
+		else {
+			mutex_lock(&ide_setting_mtx);
+			err = ide_devset_execute(drive, ds, arg);
+			mutex_unlock(&ide_setting_mtx);
+		}
+	}
+	return err;
+}
+EXPORT_SYMBOL_GPL(ide_setting_ioctl);
+
+static int ide_get_identity_ioctl(ide_drive_t *drive, unsigned int cmd,
+				  unsigned long arg)
+{
+	u16 *id = NULL;
+	int size = (cmd == HDIO_GET_IDENTITY) ? (ATA_ID_WORDS * 2) : 142;
+	int rc = 0;
+
+	if (drive->id_read == 0) {
+		rc = -ENOMSG;
+		goto out;
+	}
+
+	id = kmalloc(size, GFP_KERNEL);
+	if (id == NULL) {
+		rc = -ENOMEM;
+		goto out;
+	}
+
+	memcpy(id, drive->id, size);
+	ata_id_to_hd_driveid(id);
+
+	if (copy_to_user((void __user *)arg, id, size))
+		rc = -EFAULT;
+
+	kfree(id);
+out:
+	return rc;
+}
+
+static int ide_get_nice_ioctl(ide_drive_t *drive, unsigned long arg)
+{
+	return put_user((drive->dsc_overlap << IDE_NICE_DSC_OVERLAP) |
+			(drive->nice1 << IDE_NICE_1), (long __user *)arg);
+}
+
+static int ide_set_nice_ioctl(ide_drive_t *drive, unsigned long arg)
+{
+	if (arg != (arg & ((1 << IDE_NICE_DSC_OVERLAP) | (1 << IDE_NICE_1))))
+		return -EPERM;
+
+	if (((arg >> IDE_NICE_DSC_OVERLAP) & 1) &&
+	    (drive->media == ide_disk || drive->media == ide_floppy ||
+	     drive->scsi))
+		return -EPERM;
+
+	drive->dsc_overlap = (arg >> IDE_NICE_DSC_OVERLAP) & 1;
+	drive->nice1 = (arg >> IDE_NICE_1) & 1;
+
+	return 0;
+}
+
+static int ide_cmd_ioctl(ide_drive_t *drive, unsigned cmd, unsigned long arg)
+{
+	u8 *buf = NULL;
+	int bufsize = 0, err = 0;
+	u8 args[4], xfer_rate = 0;
+	ide_task_t tfargs;
+	struct ide_taskfile *tf = &tfargs.tf;
+	u16 *id = drive->id;
+
+	if (NULL == (void *) arg) {
+		struct request *rq;
+
+		rq = blk_get_request(drive->queue, READ, __GFP_WAIT);
+		rq->cmd_type = REQ_TYPE_ATA_TASKFILE;
+		err = blk_execute_rq(drive->queue, NULL, rq, 0);
+		blk_put_request(rq);
+
+		return err;
+	}
+
+	if (copy_from_user(args, (void __user *)arg, 4))
+		return -EFAULT;
+
+	memset(&tfargs, 0, sizeof(ide_task_t));
+	tf->feature = args[2];
+	if (args[0] == ATA_CMD_SMART) {
+		tf->nsect = args[3];
+		tf->lbal  = args[1];
+		tf->lbam  = 0x4f;
+		tf->lbah  = 0xc2;
+		tfargs.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_IN_NSECT;
+	} else {
+		tf->nsect = args[1];
+		tfargs.tf_flags = IDE_TFLAG_OUT_FEATURE |
+				  IDE_TFLAG_OUT_NSECT | IDE_TFLAG_IN_NSECT;
+	}
+	tf->command = args[0];
+	tfargs.data_phase = args[3] ? TASKFILE_IN : TASKFILE_NO_DATA;
+
+	if (args[3]) {
+		tfargs.tf_flags |= IDE_TFLAG_IO_16BIT;
+		bufsize = SECTOR_SIZE * args[3];
+		buf = kzalloc(bufsize, GFP_KERNEL);
+		if (buf == NULL)
+			return -ENOMEM;
+	}
+
+	if (tf->command == ATA_CMD_SET_FEATURES &&
+	    tf->feature == SETFEATURES_XFER &&
+	    tf->nsect >= XFER_SW_DMA_0 &&
+	    (id[ATA_ID_UDMA_MODES] ||
+	     id[ATA_ID_MWDMA_MODES] ||
+	     id[ATA_ID_SWDMA_MODES])) {
+		xfer_rate = args[1];
+		if (tf->nsect > XFER_UDMA_2 && !eighty_ninty_three(drive)) {
+			printk(KERN_WARNING "%s: UDMA speeds >UDMA33 cannot "
+					    "be set\n", drive->name);
+			goto abort;
+		}
+	}
+
+	err = ide_raw_taskfile(drive, &tfargs, buf, args[3]);
+
+	args[0] = tf->status;
+	args[1] = tf->error;
+	args[2] = tf->nsect;
+
+	if (!err && xfer_rate) {
+		/* active-retuning-calls future */
+		ide_set_xfer_rate(drive, xfer_rate);
+		ide_driveid_update(drive);
+	}
+abort:
+	if (copy_to_user((void __user *)arg, &args, 4))
+		err = -EFAULT;
+	if (buf) {
+		if (copy_to_user((void __user *)(arg + 4), buf, bufsize))
+			err = -EFAULT;
+		kfree(buf);
+	}
+	return err;
+}
+
+static int ide_task_ioctl(ide_drive_t *drive, unsigned cmd, unsigned long arg)
+{
+	void __user *p = (void __user *)arg;
+	int err = 0;
+	u8 args[7];
+	ide_task_t task;
+
+	if (copy_from_user(args, p, 7))
+		return -EFAULT;
+
+	memset(&task, 0, sizeof(task));
+	memcpy(&task.tf_array[7], &args[1], 6);
+	task.tf.command = args[0];
+	task.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
+
+	err = ide_no_data_taskfile(drive, &task);
+
+	args[0] = task.tf.command;
+	memcpy(&args[1], &task.tf_array[7], 6);
+
+	if (copy_to_user(p, args, 7))
+		err = -EFAULT;
+
+	return err;
+}
+
+static int generic_drive_reset(ide_drive_t *drive)
+{
+	struct request *rq;
+	int ret = 0;
+
+	rq = blk_get_request(drive->queue, READ, __GFP_WAIT);
+	rq->cmd_type = REQ_TYPE_SPECIAL;
+	rq->cmd_len = 1;
+	rq->cmd[0] = REQ_DRIVE_RESET;
+	rq->cmd_flags |= REQ_SOFTBARRIER;
+	if (blk_execute_rq(drive->queue, NULL, rq, 1))
+		ret = rq->errors;
+	blk_put_request(rq);
+	return ret;
+}
+
+int generic_ide_ioctl(ide_drive_t *drive, struct file *file,
+		      struct block_device *bdev,
+		      unsigned int cmd, unsigned long arg)
+{
+	int err;
+
+	err = ide_setting_ioctl(drive, bdev, cmd, arg, ide_ioctl_settings);
+	if (err != -EOPNOTSUPP)
+		return err;
+
+	switch (cmd) {
+	case HDIO_OBSOLETE_IDENTITY:
+	case HDIO_GET_IDENTITY:
+		if (bdev != bdev->bd_contains)
+			return -EINVAL;
+		return ide_get_identity_ioctl(drive, cmd, arg);
+	case HDIO_GET_NICE:
+		return ide_get_nice_ioctl(drive, arg);
+	case HDIO_SET_NICE:
+		if (!capable(CAP_SYS_ADMIN))
+			return -EACCES;
+		return ide_set_nice_ioctl(drive, arg);
+#ifdef CONFIG_IDE_TASK_IOCTL
+	case HDIO_DRIVE_TASKFILE:
+		if (!capable(CAP_SYS_ADMIN) || !capable(CAP_SYS_RAWIO))
+			return -EACCES;
+		if (drive->media == ide_disk)
+			return ide_taskfile_ioctl(drive, cmd, arg);
+		return -ENOMSG;
+#endif
+	case HDIO_DRIVE_CMD:
+		if (!capable(CAP_SYS_RAWIO))
+			return -EACCES;
+		return ide_cmd_ioctl(drive, cmd, arg);
+	case HDIO_DRIVE_TASK:
+		if (!capable(CAP_SYS_RAWIO))
+			return -EACCES;
+		return ide_task_ioctl(drive, cmd, arg);
+	case HDIO_DRIVE_RESET:
+		if (!capable(CAP_SYS_ADMIN))
+			return -EACCES;
+		return generic_drive_reset(drive);
+	case HDIO_GET_BUSSTATE:
+		if (!capable(CAP_SYS_ADMIN))
+			return -EACCES;
+		if (put_user(BUSSTATE_ON, (long __user *)arg))
+			return -EFAULT;
+		return 0;
+	case HDIO_SET_BUSSTATE:
+		if (!capable(CAP_SYS_ADMIN))
+			return -EACCES;
+		return -EOPNOTSUPP;
+	default:
+		return -EINVAL;
+	}
+}
+EXPORT_SYMBOL(generic_ide_ioctl);
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 2cbadffe922e..0a2fd3b37ac4 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -18,7 +18,6 @@
 #include <linux/slab.h>
 #include <linux/pci.h>
 #include <linux/delay.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/bitops.h>
 #include <linux/nmi.h>
@@ -400,97 +399,14 @@ const struct ide_tp_ops default_tp_ops = {
 	.output_data		= ide_output_data,
 };
 
-void ide_fix_driveid (struct hd_driveid *id)
+void ide_fix_driveid(u16 *id)
 {
 #ifndef __LITTLE_ENDIAN
 # ifdef __BIG_ENDIAN
 	int i;
-	u16 *stringcast;
-
-	id->config         = __le16_to_cpu(id->config);
-	id->cyls           = __le16_to_cpu(id->cyls);
-	id->reserved2      = __le16_to_cpu(id->reserved2);
-	id->heads          = __le16_to_cpu(id->heads);
-	id->track_bytes    = __le16_to_cpu(id->track_bytes);
-	id->sector_bytes   = __le16_to_cpu(id->sector_bytes);
-	id->sectors        = __le16_to_cpu(id->sectors);
-	id->vendor0        = __le16_to_cpu(id->vendor0);
-	id->vendor1        = __le16_to_cpu(id->vendor1);
-	id->vendor2        = __le16_to_cpu(id->vendor2);
-	stringcast = (u16 *)&id->serial_no[0];
-	for (i = 0; i < (20/2); i++)
-		stringcast[i] = __le16_to_cpu(stringcast[i]);
-	id->buf_type       = __le16_to_cpu(id->buf_type);
-	id->buf_size       = __le16_to_cpu(id->buf_size);
-	id->ecc_bytes      = __le16_to_cpu(id->ecc_bytes);
-	stringcast = (u16 *)&id->fw_rev[0];
-	for (i = 0; i < (8/2); i++)
-		stringcast[i] = __le16_to_cpu(stringcast[i]);
-	stringcast = (u16 *)&id->model[0];
-	for (i = 0; i < (40/2); i++)
-		stringcast[i] = __le16_to_cpu(stringcast[i]);
-	id->dword_io       = __le16_to_cpu(id->dword_io);
-	id->reserved50     = __le16_to_cpu(id->reserved50);
-	id->field_valid    = __le16_to_cpu(id->field_valid);
-	id->cur_cyls       = __le16_to_cpu(id->cur_cyls);
-	id->cur_heads      = __le16_to_cpu(id->cur_heads);
-	id->cur_sectors    = __le16_to_cpu(id->cur_sectors);
-	id->cur_capacity0  = __le16_to_cpu(id->cur_capacity0);
-	id->cur_capacity1  = __le16_to_cpu(id->cur_capacity1);
-	id->lba_capacity   = __le32_to_cpu(id->lba_capacity);
-	id->dma_1word      = __le16_to_cpu(id->dma_1word);
-	id->dma_mword      = __le16_to_cpu(id->dma_mword);
-	id->eide_pio_modes = __le16_to_cpu(id->eide_pio_modes);
-	id->eide_dma_min   = __le16_to_cpu(id->eide_dma_min);
-	id->eide_dma_time  = __le16_to_cpu(id->eide_dma_time);
-	id->eide_pio       = __le16_to_cpu(id->eide_pio);
-	id->eide_pio_iordy = __le16_to_cpu(id->eide_pio_iordy);
-	for (i = 0; i < 2; ++i)
-		id->words69_70[i] = __le16_to_cpu(id->words69_70[i]);
-	for (i = 0; i < 4; ++i)
-		id->words71_74[i] = __le16_to_cpu(id->words71_74[i]);
-	id->queue_depth    = __le16_to_cpu(id->queue_depth);
-	for (i = 0; i < 4; ++i)
-		id->words76_79[i] = __le16_to_cpu(id->words76_79[i]);
-	id->major_rev_num  = __le16_to_cpu(id->major_rev_num);
-	id->minor_rev_num  = __le16_to_cpu(id->minor_rev_num);
-	id->command_set_1  = __le16_to_cpu(id->command_set_1);
-	id->command_set_2  = __le16_to_cpu(id->command_set_2);
-	id->cfsse          = __le16_to_cpu(id->cfsse);
-	id->cfs_enable_1   = __le16_to_cpu(id->cfs_enable_1);
-	id->cfs_enable_2   = __le16_to_cpu(id->cfs_enable_2);
-	id->csf_default    = __le16_to_cpu(id->csf_default);
-	id->dma_ultra      = __le16_to_cpu(id->dma_ultra);
-	id->trseuc         = __le16_to_cpu(id->trseuc);
-	id->trsEuc         = __le16_to_cpu(id->trsEuc);
-	id->CurAPMvalues   = __le16_to_cpu(id->CurAPMvalues);
-	id->mprc           = __le16_to_cpu(id->mprc);
-	id->hw_config      = __le16_to_cpu(id->hw_config);
-	id->acoustic       = __le16_to_cpu(id->acoustic);
-	id->msrqs          = __le16_to_cpu(id->msrqs);
-	id->sxfert         = __le16_to_cpu(id->sxfert);
-	id->sal            = __le16_to_cpu(id->sal);
-	id->spg            = __le32_to_cpu(id->spg);
-	id->lba_capacity_2 = __le64_to_cpu(id->lba_capacity_2);
-	for (i = 0; i < 22; i++)
-		id->words104_125[i]   = __le16_to_cpu(id->words104_125[i]);
-	id->last_lun       = __le16_to_cpu(id->last_lun);
-	id->word127        = __le16_to_cpu(id->word127);
-	id->dlf            = __le16_to_cpu(id->dlf);
-	id->csfo           = __le16_to_cpu(id->csfo);
-	for (i = 0; i < 26; i++)
-		id->words130_155[i] = __le16_to_cpu(id->words130_155[i]);
-	id->word156        = __le16_to_cpu(id->word156);
-	for (i = 0; i < 3; i++)
-		id->words157_159[i] = __le16_to_cpu(id->words157_159[i]);
-	id->cfa_power      = __le16_to_cpu(id->cfa_power);
-	for (i = 0; i < 15; i++)
-		id->words161_175[i] = __le16_to_cpu(id->words161_175[i]);
-	for (i = 0; i < 30; i++)
-		id->words176_205[i] = __le16_to_cpu(id->words176_205[i]);
-	for (i = 0; i < 49; i++)
-		id->words206_254[i] = __le16_to_cpu(id->words206_254[i]);
-	id->integrity_word  = __le16_to_cpu(id->integrity_word);
+
+	for (i = 0; i < 256; i++)
+		id[i] = __le16_to_cpu(id[i]);
 # else
 #  error "Please fix <asm/byteorder.h>"
 # endif
@@ -501,19 +417,21 @@ void ide_fix_driveid (struct hd_driveid *id)
  * ide_fixstring() cleans up and (optionally) byte-swaps a text string,
  * removing leading/trailing blanks and compressing internal blanks.
  * It is primarily used to tidy up the model name/number fields as
- * returned by the WIN_[P]IDENTIFY commands.
+ * returned by the ATA_CMD_ID_ATA[PI] commands.
  */
 
 void ide_fixstring (u8 *s, const int bytecount, const int byteswap)
 {
-	u8 *p = s, *end = &s[bytecount & ~1]; /* bytecount must be even */
+	u8 *p, *end = &s[bytecount & ~1]; /* bytecount must be even */
 
 	if (byteswap) {
 		/* convert from big-endian to host byte order */
-		for (p = end ; p != s;)
-			be16_to_cpus((u16 *)(p -= 2));
+		for (p = s ; p != end ; p += 2)
+			be16_to_cpus((u16 *) p);
 	}
+
 	/* strip leading blanks */
+	p = s;
 	while (s != end && *s == ' ')
 		++s;
 	/* compress internal blanks and strip trailing blanks */
@@ -556,7 +474,7 @@ int drive_is_ready (ide_drive_t *drive)
 		/* Note: this may clear a pending IRQ!! */
 		stat = hwif->tp_ops->read_status(hwif);
 
-	if (stat & BUSY_STAT)
+	if (stat & ATA_BUSY)
 		/* drive busy:  definitely not interrupting */
 		return 0;
 
@@ -588,10 +506,10 @@ static int __ide_wait_stat(ide_drive_t *drive, u8 good, u8 bad, unsigned long ti
 	udelay(1);	/* spec allows drive 400ns to assert "BUSY" */
 	stat = tp_ops->read_status(hwif);
 
-	if (stat & BUSY_STAT) {
+	if (stat & ATA_BUSY) {
 		local_irq_set(flags);
 		timeout += jiffies;
-		while ((stat = tp_ops->read_status(hwif)) & BUSY_STAT) {
+		while ((stat = tp_ops->read_status(hwif)) & ATA_BUSY) {
 			if (time_after(jiffies, timeout)) {
 				/*
 				 * One last read after the timeout in case
@@ -599,7 +517,7 @@ static int __ide_wait_stat(ide_drive_t *drive, u8 good, u8 bad, unsigned long ti
 				 * progress during the timeout..
 				 */
 				stat = tp_ops->read_status(hwif);
-				if (!(stat & BUSY_STAT))
+				if ((stat & ATA_BUSY) == 0)
 					break;
 
 				local_irq_restore(flags);
@@ -660,18 +578,18 @@ EXPORT_SYMBOL(ide_wait_stat);
 /**
  *	ide_in_drive_list	-	look for drive in black/white list
  *	@id: drive identifier
- *	@drive_table: list to inspect
+ *	@table: list to inspect
  *
  *	Look for a drive in the blacklist and the whitelist tables
  *	Returns 1 if the drive is found in the table.
  */
 
-int ide_in_drive_list(struct hd_driveid *id, const struct drive_list_entry *drive_table)
+int ide_in_drive_list(u16 *id, const struct drive_list_entry *table)
 {
-	for ( ; drive_table->id_model; drive_table++)
-		if ((!strcmp(drive_table->id_model, id->model)) &&
-		    (!drive_table->id_firmware ||
-		     strstr(id->fw_rev, drive_table->id_firmware)))
+	for ( ; table->id_model; table++)
+		if ((!strcmp(table->id_model, (char *)&id[ATA_ID_PROD])) &&
+		    (!table->id_firmware ||
+		     strstr((char *)&id[ATA_ID_FW_REV], table->id_firmware)))
 			return 1;
 	return 0;
 }
@@ -702,7 +620,7 @@ static const struct drive_list_entry ivb_list[] = {
 u8 eighty_ninty_three (ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	struct hd_driveid *id = drive->id;
+	u16 *id = drive->id;
 	int ivb = ide_in_drive_list(id, ivb_list);
 
 	if (hwif->cbl == ATA_CBL_PATA40_SHORT)
@@ -712,7 +630,7 @@ u8 eighty_ninty_three (ide_drive_t *drive)
 		printk(KERN_DEBUG "%s: skipping word 93 validity check\n",
 				  drive->name);
 
-	if (ide_dev_is_sata(id) && !ivb)
+	if (ata_id_is_sata(id) && !ivb)
 		return 1;
 
 	if (hwif->cbl != ATA_CBL_PATA80 && !ivb)
@@ -724,7 +642,8 @@ u8 eighty_ninty_three (ide_drive_t *drive)
 	 * - force bit13 (80c cable present) check also for !ivb devices
 	 *   (unless the slave device is pre-ATA3)
 	 */
-	if ((id->hw_config & 0x4000) || (ivb && (id->hw_config & 0x2000)))
+	if ((id[ATA_ID_HW_CONFIG] & 0x4000) ||
+	    (ivb && (id[ATA_ID_HW_CONFIG] & 0x2000)))
 		return 1;
 
 no_80w:
@@ -745,8 +664,8 @@ int ide_driveid_update(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
-	struct hd_driveid *id;
-	unsigned long timeout, flags;
+	u16 *id;
+	unsigned long flags;
 	u8 stat;
 
 	/*
@@ -757,29 +676,24 @@ int ide_driveid_update(ide_drive_t *drive)
 	SELECT_MASK(drive, 1);
 	tp_ops->set_irq(hwif, 0);
 	msleep(50);
-	tp_ops->exec_command(hwif, WIN_IDENTIFY);
-	timeout = jiffies + WAIT_WORSTCASE;
-	do {
-		if (time_after(jiffies, timeout)) {
-			SELECT_MASK(drive, 0);
-			return 0;	/* drive timed-out */
-		}
+	tp_ops->exec_command(hwif, ATA_CMD_ID_ATA);
 
-		msleep(50);	/* give drive a breather */
-		stat = tp_ops->read_altstatus(hwif);
-	} while (stat & BUSY_STAT);
+	if (ide_busy_sleep(hwif, WAIT_WORSTCASE, 1)) {
+		SELECT_MASK(drive, 0);
+		return 0;
+	}
 
-	msleep(50);	/* wait for IRQ and DRQ_STAT */
+	msleep(50);	/* wait for IRQ and ATA_DRQ */
 	stat = tp_ops->read_status(hwif);
 
-	if (!OK_STAT(stat, DRQ_STAT, BAD_R_STAT)) {
+	if (!OK_STAT(stat, ATA_DRQ, BAD_R_STAT)) {
 		SELECT_MASK(drive, 0);
 		printk("%s: CHECK for good STATUS\n", drive->name);
 		return 0;
 	}
 	local_irq_save(flags);
 	SELECT_MASK(drive, 0);
-	id = kmalloc(SECTOR_WORDS*4, GFP_ATOMIC);
+	id = kmalloc(SECTOR_SIZE, GFP_ATOMIC);
 	if (!id) {
 		local_irq_restore(flags);
 		return 0;
@@ -789,16 +703,16 @@ int ide_driveid_update(ide_drive_t *drive)
 	local_irq_enable();
 	local_irq_restore(flags);
 	ide_fix_driveid(id);
-	if (id) {
-		drive->id->dma_ultra = id->dma_ultra;
-		drive->id->dma_mword = id->dma_mword;
-		drive->id->dma_1word = id->dma_1word;
-		/* anything more ? */
-		kfree(id);
-
-		if (drive->using_dma && ide_id_dma_bug(drive))
-			ide_dma_off(drive);
-	}
+
+	drive->id[ATA_ID_UDMA_MODES]  = id[ATA_ID_UDMA_MODES];
+	drive->id[ATA_ID_MWDMA_MODES] = id[ATA_ID_MWDMA_MODES];
+	drive->id[ATA_ID_SWDMA_MODES] = id[ATA_ID_SWDMA_MODES];
+	/* anything more ? */
+
+	kfree(id);
+
+	if (drive->using_dma && ide_id_dma_bug(drive))
+		ide_dma_off(drive);
 
 	return 1;
 }
@@ -807,6 +721,7 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 {
 	ide_hwif_t *hwif = drive->hwif;
 	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
+	u16 *id = drive->id, i;
 	int error = 0;
 	u8 stat;
 	ide_task_t task;
@@ -817,7 +732,7 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 #endif
 
 	/* Skip setting PIO flow-control modes on pre-EIDE drives */
-	if ((speed & 0xf8) == XFER_PIO_0 && !(drive->id->capability & 0x08))
+	if ((speed & 0xf8) == XFER_PIO_0 && ata_id_has_iordy(drive->id) == 0)
 		goto skip;
 
 	/*
@@ -851,13 +766,13 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 
 	tp_ops->tf_load(drive, &task);
 
-	tp_ops->exec_command(hwif, WIN_SETFEATURES);
+	tp_ops->exec_command(hwif, ATA_CMD_SET_FEATURES);
 
 	if (drive->quirk_list == 2)
 		tp_ops->set_irq(hwif, 1);
 
 	error = __ide_wait_stat(drive, drive->ready_stat,
-				BUSY_STAT|DRQ_STAT|ERR_STAT,
+				ATA_BUSY | ATA_DRQ | ATA_ERR,
 				WAIT_CMD, &stat);
 
 	SELECT_MASK(drive, 0);
@@ -869,9 +784,9 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 		return error;
 	}
 
-	drive->id->dma_ultra &= ~0xFF00;
-	drive->id->dma_mword &= ~0x0F00;
-	drive->id->dma_1word &= ~0x0F00;
+	id[ATA_ID_UDMA_MODES]  &= ~0xFF00;
+	id[ATA_ID_MWDMA_MODES] &= ~0x0F00;
+	id[ATA_ID_SWDMA_MODES] &= ~0x0F00;
 
  skip:
 #ifdef CONFIG_BLK_DEV_IDEDMA
@@ -881,23 +796,17 @@ int ide_config_drive_speed(ide_drive_t *drive, u8 speed)
 		ide_dma_off_quietly(drive);
 #endif
 
-	switch(speed) {
-		case XFER_UDMA_7:   drive->id->dma_ultra |= 0x8080; break;
-		case XFER_UDMA_6:   drive->id->dma_ultra |= 0x4040; break;
-		case XFER_UDMA_5:   drive->id->dma_ultra |= 0x2020; break;
-		case XFER_UDMA_4:   drive->id->dma_ultra |= 0x1010; break;
-		case XFER_UDMA_3:   drive->id->dma_ultra |= 0x0808; break;
-		case XFER_UDMA_2:   drive->id->dma_ultra |= 0x0404; break;
-		case XFER_UDMA_1:   drive->id->dma_ultra |= 0x0202; break;
-		case XFER_UDMA_0:   drive->id->dma_ultra |= 0x0101; break;
-		case XFER_MW_DMA_2: drive->id->dma_mword |= 0x0404; break;
-		case XFER_MW_DMA_1: drive->id->dma_mword |= 0x0202; break;
-		case XFER_MW_DMA_0: drive->id->dma_mword |= 0x0101; break;
-		case XFER_SW_DMA_2: drive->id->dma_1word |= 0x0404; break;
-		case XFER_SW_DMA_1: drive->id->dma_1word |= 0x0202; break;
-		case XFER_SW_DMA_0: drive->id->dma_1word |= 0x0101; break;
-		default: break;
+	if (speed >= XFER_UDMA_0) {
+		i = 1 << (speed - XFER_UDMA_0);
+		id[ATA_ID_UDMA_MODES] |= (i << 8 | i);
+	} else if (speed >= XFER_MW_DMA_0) {
+		i = 1 << (speed - XFER_MW_DMA_0);
+		id[ATA_ID_MWDMA_MODES] |= (i << 8 | i);
+	} else if (speed >= XFER_SW_DMA_0) {
+		i = 1 << (speed - XFER_SW_DMA_0);
+		id[ATA_ID_SWDMA_MODES] |= (i << 8 | i);
 	}
+
 	if (!drive->init_speed)
 		drive->init_speed = speed;
 	drive->current_speed = speed;
@@ -977,7 +886,7 @@ void ide_execute_pkt_cmd(ide_drive_t *drive)
 	unsigned long flags;
 
 	spin_lock_irqsave(&ide_lock, flags);
-	hwif->tp_ops->exec_command(hwif, WIN_PACKETCMD);
+	hwif->tp_ops->exec_command(hwif, ATA_CMD_PACKET);
 	ndelay(400);
 	spin_unlock_irqrestore(&ide_lock, flags);
 }
@@ -1010,7 +919,7 @@ static ide_startstop_t atapi_reset_pollfunc (ide_drive_t *drive)
 	udelay (10);
 	stat = hwif->tp_ops->read_status(hwif);
 
-	if (OK_STAT(stat, 0, BUSY_STAT))
+	if (OK_STAT(stat, 0, ATA_BUSY))
 		printk("%s: ATAPI reset complete\n", drive->name);
 	else {
 		if (time_before(jiffies, hwgroup->poll_timeout)) {
@@ -1056,7 +965,7 @@ static ide_startstop_t reset_pollfunc (ide_drive_t *drive)
 
 	tmp = hwif->tp_ops->read_status(hwif);
 
-	if (!OK_STAT(tmp, 0, BUSY_STAT)) {
+	if (!OK_STAT(tmp, 0, ATA_BUSY)) {
 		if (time_before(jiffies, hwgroup->poll_timeout)) {
 			ide_set_handler(drive, &reset_pollfunc, HZ/20, NULL);
 			/* continue polling */
@@ -1102,7 +1011,7 @@ out:
 
 static void ide_disk_pre_reset(ide_drive_t *drive)
 {
-	int legacy = (drive->id->cfs_enable_2 & 0x0400) ? 0 : 1;
+	int legacy = (drive->id[ATA_ID_CFS_ENABLE_2] & 0x0400) ? 0 : 1;
 
 	drive->special.all = 0;
 	drive->special.b.set_geometry = legacy;
@@ -1187,7 +1096,7 @@ static ide_startstop_t do_reset1 (ide_drive_t *drive, int do_not_try_atapi)
 		pre_reset(drive);
 		SELECT_DRIVE(drive);
 		udelay (20);
-		tp_ops->exec_command(hwif, WIN_SRST);
+		tp_ops->exec_command(hwif, ATA_CMD_DEV_RESET);
 		ndelay(400);
 		hwgroup->poll_timeout = jiffies + WAIT_WORSTCASE;
 		hwgroup->polling = 1;
@@ -1270,7 +1179,7 @@ int ide_wait_not_busy(ide_hwif_t *hwif, unsigned long timeout)
 		 */
 		mdelay(1);
 		stat = hwif->tp_ops->read_status(hwif);
-		if ((stat & BUSY_STAT) == 0)
+		if ((stat & ATA_BUSY) == 0)
 			return 0;
 		/*
 		 * Assume a value of 0xff means nothing is connected to
diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c
index 97fefabea8b8..ed426dd0fdd8 100644
--- a/drivers/ide/ide-lib.c
+++ b/drivers/ide/ide-lib.c
@@ -2,7 +2,6 @@
 #include <linux/string.h>
 #include <linux/kernel.h>
 #include <linux/interrupt.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/bitops.h>
 
@@ -90,29 +89,31 @@ static u8 ide_rate_filter(ide_drive_t *drive, u8 speed)
 
 u8 ide_get_best_pio_mode (ide_drive_t *drive, u8 mode_wanted, u8 max_mode)
 {
-	int pio_mode;
-	struct hd_driveid* id = drive->id;
-	int overridden  = 0;
+	u16 *id = drive->id;
+	int pio_mode = -1, overridden = 0;
 
 	if (mode_wanted != 255)
 		return min_t(u8, mode_wanted, max_mode);
 
-	if ((drive->hwif->host_flags & IDE_HFLAG_PIO_NO_BLACKLIST) == 0 &&
-	    (pio_mode = ide_scan_pio_blacklist(id->model)) != -1) {
+	if ((drive->hwif->host_flags & IDE_HFLAG_PIO_NO_BLACKLIST) == 0)
+		pio_mode = ide_scan_pio_blacklist((char *)&id[ATA_ID_PROD]);
+
+	if (pio_mode != -1) {
 		printk(KERN_INFO "%s: is on PIO blacklist\n", drive->name);
 	} else {
-		pio_mode = id->tPIO;
+		pio_mode = id[ATA_ID_OLD_PIO_MODES] >> 8;
 		if (pio_mode > 2) {	/* 2 is maximum allowed tPIO value */
 			pio_mode = 2;
 			overridden = 1;
 		}
-		if (id->field_valid & 2) {	  /* drive implements ATA2? */
-			if (id->capability & 8) { /* IORDY supported? */
-				if (id->eide_pio_modes & 7) {
+
+		if (id[ATA_ID_FIELD_VALID] & 2) {	      /* ATA2? */
+			if (ata_id_has_iordy(id)) {
+				if (id[ATA_ID_PIO_MODES] & 7) {
 					overridden = 0;
-					if (id->eide_pio_modes & 4)
+					if (id[ATA_ID_PIO_MODES] & 4)
 						pio_mode = 5;
-					else if (id->eide_pio_modes & 2)
+					else if (id[ATA_ID_PIO_MODES] & 2)
 						pio_mode = 4;
 					else
 						pio_mode = 3;
@@ -338,16 +339,16 @@ static void ide_dump_sector(ide_drive_t *drive)
 static void ide_dump_ata_error(ide_drive_t *drive, u8 err)
 {
 	printk("{ ");
-	if (err & ABRT_ERR)	printk("DriveStatusError ");
-	if (err & ICRC_ERR)
-		printk((err & ABRT_ERR) ? "BadCRC " : "BadSector ");
-	if (err & ECC_ERR)	printk("UncorrectableError ");
-	if (err & ID_ERR)	printk("SectorIdNotFound ");
-	if (err & TRK0_ERR)	printk("TrackZeroNotFound ");
-	if (err & MARK_ERR)	printk("AddrMarkNotFound ");
+	if (err & ATA_ABORTED)	printk("DriveStatusError ");
+	if (err & ATA_ICRC)
+		printk((err & ATA_ABORTED) ? "BadCRC " : "BadSector ");
+	if (err & ATA_UNC)	printk("UncorrectableError ");
+	if (err & ATA_IDNF)	printk("SectorIdNotFound ");
+	if (err & ATA_TRK0NF)	printk("TrackZeroNotFound ");
+	if (err & ATA_AMNF)	printk("AddrMarkNotFound ");
 	printk("}");
-	if ((err & (BBD_ERR | ABRT_ERR)) == BBD_ERR ||
-	    (err & (ECC_ERR|ID_ERR|MARK_ERR))) {
+	if ((err & (ATA_BBK | ATA_ABORTED)) == ATA_BBK ||
+	    (err & (ATA_UNC | ATA_IDNF | ATA_AMNF))) {
 		ide_dump_sector(drive);
 		if (HWGROUP(drive) && HWGROUP(drive)->rq)
 			printk(", sector=%llu",
@@ -359,12 +360,12 @@ static void ide_dump_ata_error(ide_drive_t *drive, u8 err)
 static void ide_dump_atapi_error(ide_drive_t *drive, u8 err)
 {
 	printk("{ ");
-	if (err & ILI_ERR)	printk("IllegalLengthIndication ");
-	if (err & EOM_ERR)	printk("EndOfMedia ");
-	if (err & ABRT_ERR)	printk("AbortedCommand ");
-	if (err & MCR_ERR)	printk("MediaChangeRequested ");
-	if (err & LFS_ERR)	printk("LastFailedSense=0x%02x ",
-				       (err & LFS_ERR) >> 4);
+	if (err & ATAPI_ILI)	printk("IllegalLengthIndication ");
+	if (err & ATAPI_EOM)	printk("EndOfMedia ");
+	if (err & ATA_ABORTED)	printk("AbortedCommand ");
+	if (err & ATA_MCR)	printk("MediaChangeRequested ");
+	if (err & ATAPI_LFS)	printk("LastFailedSense=0x%02x ",
+				       (err & ATAPI_LFS) >> 4);
 	printk("}\n");
 }
 
@@ -386,19 +387,19 @@ u8 ide_dump_status(ide_drive_t *drive, const char *msg, u8 stat)
 
 	local_irq_save(flags);
 	printk("%s: %s: status=0x%02x { ", drive->name, msg, stat);
-	if (stat & BUSY_STAT)
+	if (stat & ATA_BUSY)
 		printk("Busy ");
 	else {
-		if (stat & READY_STAT)	printk("DriveReady ");
-		if (stat & WRERR_STAT)	printk("DeviceFault ");
-		if (stat & SEEK_STAT)	printk("SeekComplete ");
-		if (stat & DRQ_STAT)	printk("DataRequest ");
-		if (stat & ECC_STAT)	printk("CorrectedError ");
-		if (stat & INDEX_STAT)	printk("Index ");
-		if (stat & ERR_STAT)	printk("Error ");
+		if (stat & ATA_DRDY)	printk("DriveReady ");
+		if (stat & ATA_DF)	printk("DeviceFault ");
+		if (stat & ATA_DSC)	printk("SeekComplete ");
+		if (stat & ATA_DRQ)	printk("DataRequest ");
+		if (stat & ATA_CORR)	printk("CorrectedError ");
+		if (stat & ATA_IDX)	printk("Index ");
+		if (stat & ATA_ERR)	printk("Error ");
 	}
 	printk("}\n");
-	if ((stat & (BUSY_STAT|ERR_STAT)) == ERR_STAT) {
+	if ((stat & (ATA_BUSY | ATA_ERR)) == ATA_ERR) {
 		err = ide_read_error(drive);
 		printk("%s: %s: error=0x%02x ", drive->name, msg, err);
 		if (drive->media == ide_disk)
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 70aa86c8807e..06575a12b635 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -50,59 +50,54 @@
  
 static void generic_id(ide_drive_t *drive)
 {
-	drive->id->cyls = drive->cyl;
-	drive->id->heads = drive->head;
-	drive->id->sectors = drive->sect;
-	drive->id->cur_cyls = drive->cyl;
-	drive->id->cur_heads = drive->head;
-	drive->id->cur_sectors = drive->sect;
+	u16 *id = drive->id;
+
+	id[ATA_ID_CUR_CYLS]	= id[ATA_ID_CYLS]	= drive->cyl;
+	id[ATA_ID_CUR_HEADS]	= id[ATA_ID_HEADS]	= drive->head;
+	id[ATA_ID_CUR_SECTORS]	= id[ATA_ID_SECTORS]	= drive->sect;
 }
 
 static void ide_disk_init_chs(ide_drive_t *drive)
 {
-	struct hd_driveid *id = drive->id;
+	u16 *id = drive->id;
 
 	/* Extract geometry if we did not already have one for the drive */
 	if (!drive->cyl || !drive->head || !drive->sect) {
-		drive->cyl  = drive->bios_cyl  = id->cyls;
-		drive->head = drive->bios_head = id->heads;
-		drive->sect = drive->bios_sect = id->sectors;
+		drive->cyl  = drive->bios_cyl  = id[ATA_ID_CYLS];
+		drive->head = drive->bios_head = id[ATA_ID_HEADS];
+		drive->sect = drive->bios_sect = id[ATA_ID_SECTORS];
 	}
 
 	/* Handle logical geometry translation by the drive */
-	if ((id->field_valid & 1) && id->cur_cyls &&
-	    id->cur_heads && (id->cur_heads <= 16) && id->cur_sectors) {
-		drive->cyl  = id->cur_cyls;
-		drive->head = id->cur_heads;
-		drive->sect = id->cur_sectors;
+	if (ata_id_current_chs_valid(id)) {
+		drive->cyl  = id[ATA_ID_CUR_CYLS];
+		drive->head = id[ATA_ID_CUR_HEADS];
+		drive->sect = id[ATA_ID_CUR_SECTORS];
 	}
 
 	/* Use physical geometry if what we have still makes no sense */
-	if (drive->head > 16 && id->heads && id->heads <= 16) {
-		drive->cyl  = id->cyls;
-		drive->head = id->heads;
-		drive->sect = id->sectors;
+	if (drive->head > 16 && id[ATA_ID_HEADS] && id[ATA_ID_HEADS] <= 16) {
+		drive->cyl  = id[ATA_ID_CYLS];
+		drive->head = id[ATA_ID_HEADS];
+		drive->sect = id[ATA_ID_SECTORS];
 	}
 }
 
 static void ide_disk_init_mult_count(ide_drive_t *drive)
 {
-	struct hd_driveid *id = drive->id;
-
-	drive->mult_count = 0;
-	if (id->max_multsect) {
-#ifdef CONFIG_IDEDISK_MULTI_MODE
-		id->multsect = ((id->max_multsect/2) > 1) ? id->max_multsect : 0;
-		id->multsect_valid = id->multsect ? 1 : 0;
-		drive->mult_req = id->multsect_valid ? id->max_multsect : 0;
-		drive->special.b.set_multmode = drive->mult_req ? 1 : 0;
-#else	/* original, pre IDE-NFG, per request of AC */
-		drive->mult_req = 0;
-		if (drive->mult_req > id->max_multsect)
-			drive->mult_req = id->max_multsect;
-		if (drive->mult_req || ((id->multsect_valid & 1) && id->multsect))
+	u16 *id = drive->id;
+	u8 max_multsect = id[ATA_ID_MAX_MULTSECT] & 0xff;
+
+	if (max_multsect) {
+		if ((max_multsect / 2) > 1)
+			id[ATA_ID_MULTSECT] = max_multsect | 0x100;
+		else
+			id[ATA_ID_MULTSECT] &= ~0x1ff;
+
+		drive->mult_req = id[ATA_ID_MULTSECT] & 0xff;
+
+		if (drive->mult_req)
 			drive->special.b.set_multmode = 1;
-#endif
 	}
 }
 
@@ -119,10 +114,10 @@ static void ide_disk_init_mult_count(ide_drive_t *drive)
 static inline void do_identify (ide_drive_t *drive, u8 cmd)
 {
 	ide_hwif_t *hwif = HWIF(drive);
-	int bswap = 1;
-	struct hd_driveid *id;
+	u16 *id = drive->id;
+	char *m = (char *)&id[ATA_ID_PROD];
+	int bswap = 1, is_cfa;
 
-	id = drive->id;
 	/* read 512 bytes of id info */
 	hwif->tp_ops->input_data(drive, NULL, id, SECTOR_SIZE);
 
@@ -135,27 +130,28 @@ static inline void do_identify (ide_drive_t *drive, u8 cmd)
 	ide_fix_driveid(id);
 
 	/*
-	 *  WIN_IDENTIFY returns little-endian info,
-	 *  WIN_PIDENTIFY *usually* returns little-endian info.
+	 *  ATA_CMD_ID_ATA returns little-endian info,
+	 *  ATA_CMD_ID_ATAPI *usually* returns little-endian info.
 	 */
-	if (cmd == WIN_PIDENTIFY) {
-		if ((id->model[0] == 'N' && id->model[1] == 'E') /* NEC */
-		 || (id->model[0] == 'F' && id->model[1] == 'X') /* Mitsumi */
-		 || (id->model[0] == 'P' && id->model[1] == 'i'))/* Pioneer */
+	if (cmd == ATA_CMD_ID_ATAPI) {
+		if ((m[0] == 'N' && m[1] == 'E') ||  /* NEC */
+		    (m[0] == 'F' && m[1] == 'X') ||  /* Mitsumi */
+		    (m[0] == 'P' && m[1] == 'i'))    /* Pioneer */
 			/* Vertos drives may still be weird */
-			bswap ^= 1;	
+			bswap ^= 1;
 	}
-	ide_fixstring(id->model,     sizeof(id->model),     bswap);
-	ide_fixstring(id->fw_rev,    sizeof(id->fw_rev),    bswap);
-	ide_fixstring(id->serial_no, sizeof(id->serial_no), bswap);
+
+	ide_fixstring(m, ATA_ID_PROD_LEN, bswap);
+	ide_fixstring((char *)&id[ATA_ID_FW_REV], ATA_ID_FW_REV_LEN, bswap);
+	ide_fixstring((char *)&id[ATA_ID_SERNO], ATA_ID_SERNO_LEN, bswap);
 
 	/* we depend on this a lot! */
-	id->model[sizeof(id->model)-1] = '\0';
+	m[ATA_ID_PROD_LEN - 1] = '\0';
 
-	if (strstr(id->model, "E X A B Y T E N E S T"))
+	if (strstr(m, "E X A B Y T E N E S T"))
 		goto err_misc;
 
-	printk(KERN_INFO "%s: %s, ", drive->name, id->model);
+	printk(KERN_INFO "%s: %s, ", drive->name, m);
 
 	drive->present = 1;
 	drive->dead = 0;
@@ -163,16 +159,16 @@ static inline void do_identify (ide_drive_t *drive, u8 cmd)
 	/*
 	 * Check for an ATAPI device
 	 */
-	if (cmd == WIN_PIDENTIFY) {
-		u8 type = (id->config >> 8) & 0x1f;
+	if (cmd == ATA_CMD_ID_ATAPI) {
+		u8 type = (id[ATA_ID_CONFIG] >> 8) & 0x1f;
 
 		printk(KERN_CONT "ATAPI ");
 		switch (type) {
 			case ide_floppy:
-				if (!strstr(id->model, "CD-ROM")) {
-					if (!strstr(id->model, "oppy") &&
-					    !strstr(id->model, "poyp") &&
-					    !strstr(id->model, "ZIP"))
+				if (!strstr(m, "CD-ROM")) {
+					if (!strstr(m, "oppy") &&
+					    !strstr(m, "poyp") &&
+					    !strstr(m, "ZIP"))
 						printk(KERN_CONT "cdrom or floppy?, assuming ");
 					if (drive->media != ide_cdrom) {
 						printk(KERN_CONT "FLOPPY");
@@ -186,8 +182,7 @@ static inline void do_identify (ide_drive_t *drive, u8 cmd)
 				drive->removable = 1;
 #ifdef CONFIG_PPC
 				/* kludge for Apple PowerBook internal zip */
-				if (!strstr(id->model, "CD-ROM") &&
-				    strstr(id->model, "ZIP")) {
+				if (!strstr(m, "CD-ROM") && strstr(m, "ZIP")) {
 					printk(KERN_CONT "FLOPPY");
 					type = ide_floppy;
 					break;
@@ -217,18 +212,15 @@ static inline void do_identify (ide_drive_t *drive, u8 cmd)
 	 * Not an ATAPI device: looks like a "regular" hard disk
 	 */
 
-	/*
-	 * 0x848a = CompactFlash device
-	 * These are *not* removable in Linux definition of the term
-	 */
+	is_cfa = ata_id_is_cfa(id);
 
-	if ((id->config != 0x848a) && (id->config & (1<<7)))
+	/* CF devices are *not* removable in Linux definition of the term */
+	if (is_cfa == 0 && (id[ATA_ID_CONFIG] & (1 << 7)))
 		drive->removable = 1;
 
 	drive->media = ide_disk;
 
-	printk(KERN_CONT "%s DISK drive\n",
-		(id->config == 0x848a) ? "CFA" : "ATA");
+	printk(KERN_CONT "%s DISK drive\n", is_cfa ? "CFA" : "ATA");
 
 	return;
 
@@ -268,7 +260,7 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 	if (io_ports->ctl_addr) {
 		a = tp_ops->read_altstatus(hwif);
 		s = tp_ops->read_status(hwif);
-		if ((a ^ s) & ~INDEX_STAT)
+		if ((a ^ s) & ~ATA_IDX)
 			/* ancient Seagate drives, broken interfaces */
 			printk(KERN_INFO "%s: probing with STATUS(0x%02x) "
 					 "instead of ALTSTATUS(0x%02x)\n",
@@ -281,7 +273,7 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 	/* set features register for atapi
 	 * identify command to be sure of reply
 	 */
-	if (cmd == WIN_PIDENTIFY) {
+	if (cmd == ATA_CMD_ID_ATAPI) {
 		ide_task_t task;
 
 		memset(&task, 0, sizeof(task));
@@ -294,24 +286,16 @@ static int actual_try_to_identify (ide_drive_t *drive, u8 cmd)
 	/* ask drive for ID */
 	tp_ops->exec_command(hwif, cmd);
 
-	timeout = ((cmd == WIN_IDENTIFY) ? WAIT_WORSTCASE : WAIT_PIDENTIFY) / 2;
-	timeout += jiffies;
-	do {
-		if (time_after(jiffies, timeout)) {
-			/* drive timed-out */
-			return 1;
-		}
-		/* give drive a breather */
-		msleep(50);
-		s = use_altstatus ? tp_ops->read_altstatus(hwif)
-				  : tp_ops->read_status(hwif);
-	} while (s & BUSY_STAT);
+	timeout = ((cmd == ATA_CMD_ID_ATA) ? WAIT_WORSTCASE : WAIT_PIDENTIFY) / 2;
 
-	/* wait for IRQ and DRQ_STAT */
+	if (ide_busy_sleep(hwif, timeout, use_altstatus))
+		return 1;
+
+	/* wait for IRQ and ATA_DRQ */
 	msleep(50);
 	s = tp_ops->read_status(hwif);
 
-	if (OK_STAT(s, DRQ_STAT, BAD_R_STAT)) {
+	if (OK_STAT(s, ATA_DRQ, BAD_R_STAT)) {
 		unsigned long flags;
 
 		/* local CPU only; some systems need this */
@@ -387,19 +371,21 @@ static int try_to_identify (ide_drive_t *drive, u8 cmd)
 	return retval;
 }
 
-static int ide_busy_sleep(ide_hwif_t *hwif)
+int ide_busy_sleep(ide_hwif_t *hwif, unsigned long timeout, int altstatus)
 {
-	unsigned long timeout = jiffies + WAIT_WORSTCASE;
 	u8 stat;
 
+	timeout += jiffies;
+
 	do {
-		msleep(50);
-		stat = hwif->tp_ops->read_status(hwif);
-		if ((stat & BUSY_STAT) == 0)
+		msleep(50);	/* give drive a breather */
+		stat = altstatus ? hwif->tp_ops->read_altstatus(hwif)
+				 : hwif->tp_ops->read_status(hwif);
+		if ((stat & ATA_BUSY) == 0)
 			return 0;
 	} while (time_before(jiffies, timeout));
 
-	return 1;
+	return 1;	/* drive timed-out */
 }
 
 static u8 ide_read_device(ide_drive_t *drive)
@@ -444,13 +430,13 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 
 	if (drive->present) {
 		/* avoid waiting for inappropriate probes */
-		if ((drive->media != ide_disk) && (cmd == WIN_IDENTIFY))
+		if (drive->media != ide_disk && cmd == ATA_CMD_ID_ATA)
 			return 4;
 	}
 #ifdef DEBUG
 	printk(KERN_INFO "probing for %s: present=%d, media=%d, probetype=%s\n",
 		drive->name, drive->present, drive->media,
-		(cmd == WIN_IDENTIFY) ? "ATA" : "ATAPI");
+		(cmd == ATA_CMD_ID_ATA) ? "ATA" : "ATAPI");
 #endif
 
 	/* needed for some systems
@@ -464,7 +450,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 		if (drive->select.b.unit != 0) {
 			/* exit with drive0 selected */
 			SELECT_DRIVE(&hwif->drives[0]);
-			/* allow BUSY_STAT to assert & clear */
+			/* allow ATA_BUSY to assert & clear */
 			msleep(50);
 		}
 		/* no i/f present: mmm.. this should be a 4 -ml */
@@ -473,8 +459,8 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 
 	stat = tp_ops->read_status(hwif);
 
-	if (OK_STAT(stat, READY_STAT, BUSY_STAT) ||
-	    drive->present || cmd == WIN_PIDENTIFY) {
+	if (OK_STAT(stat, ATA_DRDY, ATA_BUSY) ||
+	    drive->present || cmd == ATA_CMD_ID_ATAPI) {
 		/* send cmd and wait */
 		if ((rc = try_to_identify(drive, cmd))) {
 			/* failed: try again */
@@ -483,17 +469,17 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 
 		stat = tp_ops->read_status(hwif);
 
-		if (stat == (BUSY_STAT | READY_STAT))
+		if (stat == (ATA_BUSY | ATA_DRDY))
 			return 4;
 
-		if (rc == 1 && cmd == WIN_PIDENTIFY) {
+		if (rc == 1 && cmd == ATA_CMD_ID_ATAPI) {
 			printk(KERN_ERR "%s: no response (status = 0x%02x), "
 					"resetting drive\n", drive->name, stat);
 			msleep(50);
 			SELECT_DRIVE(drive);
 			msleep(50);
-			tp_ops->exec_command(hwif, WIN_SRST);
-			(void)ide_busy_sleep(hwif);
+			tp_ops->exec_command(hwif, ATA_CMD_DEV_RESET);
+			(void)ide_busy_sleep(hwif, WAIT_WORSTCASE, 0);
 			rc = try_to_identify(drive, cmd);
 		}
 
@@ -526,13 +512,14 @@ static void enable_nest (ide_drive_t *drive)
 	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	u8 stat;
 
-	printk(KERN_INFO "%s: enabling %s -- ", hwif->name, drive->id->model);
+	printk(KERN_INFO "%s: enabling %s -- ",
+		hwif->name, (char *)&drive->id[ATA_ID_PROD]);
 
 	SELECT_DRIVE(drive);
 	msleep(50);
-	tp_ops->exec_command(hwif, EXABYTE_ENABLE_NEST);
+	tp_ops->exec_command(hwif, ATA_EXABYTE_ENABLE_NEST);
 
-	if (ide_busy_sleep(hwif)) {
+	if (ide_busy_sleep(hwif, WAIT_WORSTCASE, 0)) {
 		printk(KERN_CONT "failed (timeout)\n");
 		return;
 	}
@@ -545,12 +532,6 @@ static void enable_nest (ide_drive_t *drive)
 		printk(KERN_CONT "failed (status = 0x%02x)\n", stat);
 	else
 		printk(KERN_CONT "success\n");
-
-	/* if !(success||timed-out) */
-	if (do_probe(drive, WIN_IDENTIFY) >= 2) {
-		/* look for ATAPI device */
-		(void) do_probe(drive, WIN_PIDENTIFY);
-	}
 }
 
 /**
@@ -567,6 +548,8 @@ static void enable_nest (ide_drive_t *drive)
  
 static inline u8 probe_for_drive (ide_drive_t *drive)
 {
+	char *m;
+
 	/*
 	 *	In order to keep things simple we have an id
 	 *	block for all drives at all times. If the device
@@ -576,29 +559,34 @@ static inline u8 probe_for_drive (ide_drive_t *drive)
 	 *	Also note that 0 everywhere means "can't do X"
 	 */
  
-	drive->id = kzalloc(SECTOR_WORDS *4, GFP_KERNEL);
+	drive->id = kzalloc(SECTOR_SIZE, GFP_KERNEL);
 	drive->id_read = 0;
 	if(drive->id == NULL)
 	{
 		printk(KERN_ERR "ide: out of memory for id data.\n");
 		return 0;
 	}
-	strcpy(drive->id->model, "UNKNOWN");
-	
+
+	m = (char *)&drive->id[ATA_ID_PROD];
+	strcpy(m, "UNKNOWN");
+
 	/* skip probing? */
-	if (!drive->noprobe)
-	{
+	if (!drive->noprobe) {
+retry:
 		/* if !(success||timed-out) */
-		if (do_probe(drive, WIN_IDENTIFY) >= 2) {
+		if (do_probe(drive, ATA_CMD_ID_ATA) >= 2)
 			/* look for ATAPI device */
-			(void) do_probe(drive, WIN_PIDENTIFY);
-		}
+			(void)do_probe(drive, ATA_CMD_ID_ATAPI);
+
 		if (!drive->present)
 			/* drive not found */
 			return 0;
-		if (strstr(drive->id->model, "E X A B Y T E N E S T"))
+
+		if (strstr(m, "E X A B Y T E N E S T")) {
 			enable_nest(drive);
-	
+			goto retry;
+		}
+
 		/* identification failed? */
 		if (!drive->id_read) {
 			if (drive->media == ide_disk) {
@@ -740,36 +728,38 @@ out:
 
 /**
  *	ide_undecoded_slave	-	look for bad CF adapters
- *	@drive1: drive
+ *	@dev1: slave device
  *
  *	Analyse the drives on the interface and attempt to decide if we
  *	have the same drive viewed twice. This occurs with crap CF adapters
  *	and PCMCIA sometimes.
  */
 
-void ide_undecoded_slave(ide_drive_t *drive1)
+void ide_undecoded_slave(ide_drive_t *dev1)
 {
-	ide_drive_t *drive0 = &drive1->hwif->drives[0];
+	ide_drive_t *dev0 = &dev1->hwif->drives[0];
 
-	if ((drive1->dn & 1) == 0 || drive0->present == 0)
+	if ((dev1->dn & 1) == 0 || dev0->present == 0)
 		return;
 
 	/* If the models don't match they are not the same product */
-	if (strcmp(drive0->id->model, drive1->id->model))
+	if (strcmp((char *)&dev0->id[ATA_ID_PROD],
+		   (char *)&dev1->id[ATA_ID_PROD]))
 		return;
 
 	/* Serial numbers do not match */
-	if (strncmp(drive0->id->serial_no, drive1->id->serial_no, 20))
+	if (strncmp((char *)&dev0->id[ATA_ID_SERNO],
+		    (char *)&dev1->id[ATA_ID_SERNO], ATA_ID_SERNO_LEN))
 		return;
 
 	/* No serial number, thankfully very rare for CF */
-	if (drive0->id->serial_no[0] == 0)
+	if (*(char *)&dev0->id[ATA_ID_SERNO] == 0)
 		return;
 
 	/* Appears to be an IDE flash adapter with decode bugs */
 	printk(KERN_WARNING "ide-probe: ignoring undecoded slave\n");
 
-	drive1->present = 0;
+	dev1->present = 0;
 }
 
 EXPORT_SYMBOL_GPL(ide_undecoded_slave);
@@ -853,7 +843,7 @@ static void ide_port_tune_devices(ide_hwif_t *hwif)
 		if (hwif->host_flags & IDE_HFLAG_NO_IO_32BIT)
 			drive->no_io_32bit = 1;
 		else
-			drive->no_io_32bit = drive->id->dword_io ? 1 : 0;
+			drive->no_io_32bit = drive->id[ATA_ID_DWORD_IO] ? 1 : 0;
 	}
 }
 
@@ -1037,11 +1027,6 @@ static int init_irq (ide_hwif_t *hwif)
 	ide_hwgroup_t *hwgroup;
 	ide_hwif_t *match = NULL;
 
-
-	BUG_ON(in_interrupt());
-	BUG_ON(irqs_disabled());	
-	BUG_ON(hwif == NULL);
-
 	mutex_lock(&ide_cfg_mtx);
 	hwif->hwgroup = NULL;
 #if MAX_HWIFS > 1
@@ -1116,7 +1101,8 @@ static int init_irq (ide_hwif_t *hwif)
 		sa = IRQF_SHARED;
 #endif /* __mc68000__ */
 
-		if (IDE_CHIPSET_IS_PCI(hwif->chipset))
+		if (hwif->chipset == ide_pci || hwif->chipset == ide_cmd646 ||
+		    hwif->chipset == ide_ali14xx)
 			sa = IRQF_SHARED;
 
 		if (io_ports->ctl_addr)
@@ -1344,8 +1330,6 @@ static void hwif_register_devices(ide_hwif_t *hwif)
 		if (!drive->present)
 			continue;
 
-		ide_add_generic_settings(drive);
-
 		snprintf(dev->bus_id, BUS_ID_SIZE, "%u.%u", hwif->index, i);
 		dev->parent = &hwif->gendev;
 		dev->bus = &ide_bus_type;
@@ -1602,8 +1586,10 @@ struct ide_host *ide_host_alloc_all(const struct ide_port_info *d,
 	if (hws[0])
 		host->dev[0] = hws[0]->dev;
 
-	if (d)
+	if (d) {
+		host->init_chipset = d->init_chipset;
 		host->host_flags = d->host_flags;
+	}
 
 	return host;
 }
diff --git a/drivers/ide/ide-proc.c b/drivers/ide/ide-proc.c
index f66c9c3f6fc6..e7030a491463 100644
--- a/drivers/ide/ide-proc.c
+++ b/drivers/ide/ide-proc.c
@@ -12,14 +12,6 @@
  * "settings" files.  e.g.    "cat /proc/ide0/hda/settings"
  * To write a new value "val" into a specific setting "name", use:
  *   echo "name:val" >/proc/ide/ide0/hda/settings
- *
- * Also useful, "cat /proc/ide0/hda/[identify, smart_values,
- * smart_thresholds, capabilities]" will issue an IDENTIFY /
- * PACKET_IDENTIFY / SMART_READ_VALUES / SMART_READ_THRESHOLDS /
- * SENSE CAPABILITIES command to /dev/hda, and then dump out the
- * returned data as 256 16-bit words.  The "hdparm" utility will
- * be updated someday soon to use this mechanism.
- *
  */
 
 #include <linux/module.h>
@@ -31,7 +23,6 @@
 #include <linux/mm.h>
 #include <linux/pci.h>
 #include <linux/ctype.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/seq_file.h>
 
@@ -109,13 +100,14 @@ static int proc_ide_read_identify
 
 		err = taskfile_lib_get_identify(drive, page);
 		if (!err) {
-			char *out = ((char *)page) + (SECTOR_WORDS * 4);
+			char *out = (char *)page + SECTOR_SIZE;
+
 			page = out;
 			do {
 				out += sprintf(out, "%04x%c",
 					le16_to_cpup(val), (++i & 7) ? ' ' : '\n');
 				val += 1;
-			} while (i < (SECTOR_WORDS * 2));
+			} while (i < SECTOR_SIZE / 2);
 			len = out - page;
 		}
 	}
@@ -123,140 +115,25 @@ static int proc_ide_read_identify
 }
 
 /**
- *	__ide_add_setting	-	add an ide setting option
- *	@drive: drive to use
- *	@name: setting name
- *	@rw: true if the function is read write
- *	@data_type: type of data
- *	@min: range minimum
- *	@max: range maximum
- *	@mul_factor: multiplication scale
- *	@div_factor: divison scale
- *	@data: private data field
- *	@set: setting
- *	@auto_remove: setting auto removal flag
- *
- *	Removes the setting named from the device if it is present.
- *	The function takes the settings_lock to protect against
- *	parallel changes. This function must not be called from IRQ
- *	context. Returns 0 on success or -1 on failure.
- *
- *	BUGS: This code is seriously over-engineered. There is also
- *	magic about how the driver specific features are setup. If
- *	a driver is attached we assume the driver settings are auto
- *	remove.
- */
-
-static int __ide_add_setting(ide_drive_t *drive, const char *name, int rw, int data_type, int min, int max, int mul_factor, int div_factor, void *data, ide_procset_t *set, int auto_remove)
-{
-	ide_settings_t **p = (ide_settings_t **) &drive->settings, *setting = NULL;
-
-	mutex_lock(&ide_setting_mtx);
-	while ((*p) && strcmp((*p)->name, name) < 0)
-		p = &((*p)->next);
-	if ((setting = kzalloc(sizeof(*setting), GFP_KERNEL)) == NULL)
-		goto abort;
-	if ((setting->name = kmalloc(strlen(name) + 1, GFP_KERNEL)) == NULL)
-		goto abort;
-	strcpy(setting->name, name);
-	setting->rw = rw;
-	setting->data_type = data_type;
-	setting->min = min;
-	setting->max = max;
-	setting->mul_factor = mul_factor;
-	setting->div_factor = div_factor;
-	setting->data = data;
-	setting->set = set;
-
-	setting->next = *p;
-	if (auto_remove)
-		setting->auto_remove = 1;
-	*p = setting;
-	mutex_unlock(&ide_setting_mtx);
-	return 0;
-abort:
-	mutex_unlock(&ide_setting_mtx);
-	kfree(setting);
-	return -1;
-}
-
-int ide_add_setting(ide_drive_t *drive, const char *name, int rw, int data_type, int min, int max, int mul_factor, int div_factor, void *data, ide_procset_t *set)
-{
-	return __ide_add_setting(drive, name, rw, data_type, min, max, mul_factor, div_factor, data, set, 1);
-}
-
-EXPORT_SYMBOL(ide_add_setting);
-
-/**
- *	__ide_remove_setting	-	remove an ide setting option
- *	@drive: drive to use
- *	@name: setting name
- *
- *	Removes the setting named from the device if it is present.
- *	The caller must hold the setting semaphore.
- */
-
-static void __ide_remove_setting(ide_drive_t *drive, char *name)
-{
-	ide_settings_t **p, *setting;
-
-	p = (ide_settings_t **) &drive->settings;
-
-	while ((*p) && strcmp((*p)->name, name))
-		p = &((*p)->next);
-	setting = (*p);
-	if (setting == NULL)
-		return;
-
-	(*p) = setting->next;
-
-	kfree(setting->name);
-	kfree(setting);
-}
-
-/**
- *	auto_remove_settings	-	remove driver specific settings
- *	@drive: drive
- *
- *	Automatically remove all the driver specific settings for this
- *	drive. This function may not be called from IRQ context. The
- *	caller must hold ide_setting_mtx.
- */
-
-static void auto_remove_settings(ide_drive_t *drive)
-{
-	ide_settings_t *setting;
-repeat:
-	setting = drive->settings;
-	while (setting) {
-		if (setting->auto_remove) {
-			__ide_remove_setting(drive, setting->name);
-			goto repeat;
-		}
-		setting = setting->next;
-	}
-}
-
-/**
- *	ide_find_setting_by_name	-	find a drive specific setting
- *	@drive: drive to scan
+ *	ide_find_setting	-	find a specific setting
+ *	@st: setting table pointer
  *	@name: setting name
  *
- *	Scan's the device setting table for a matching entry and returns
+ *	Scan's the setting table for a matching entry and returns
  *	this or NULL if no entry is found. The caller must hold the
  *	setting semaphore
  */
 
-static ide_settings_t *ide_find_setting_by_name(ide_drive_t *drive, char *name)
+static
+const struct ide_proc_devset *ide_find_setting(const struct ide_proc_devset *st,
+					       char *name)
 {
-	ide_settings_t *setting = drive->settings;
-
-	while (setting) {
-		if (strcmp(setting->name, name) == 0)
+	while (st->name) {
+		if (strcmp(st->name, name) == 0)
 			break;
-		setting = setting->next;
+		st++;
 	}
-	return setting;
+	return st->name ? st : NULL;
 }
 
 /**
@@ -272,26 +149,20 @@ static ide_settings_t *ide_find_setting_by_name(ide_drive_t *drive, char *name)
  *	be told apart
  */
 
-static int ide_read_setting(ide_drive_t *drive, ide_settings_t *setting)
+static int ide_read_setting(ide_drive_t *drive,
+			    const struct ide_proc_devset *setting)
 {
-	int		val = -EINVAL;
-	unsigned long	flags;
+	const struct ide_devset *ds = setting->setting;
+	int val = -EINVAL;
+
+	if (ds->get) {
+		unsigned long flags;
 
-	if ((setting->rw & SETTING_READ)) {
 		spin_lock_irqsave(&ide_lock, flags);
-		switch (setting->data_type) {
-		case TYPE_BYTE:
-			val = *((u8 *) setting->data);
-			break;
-		case TYPE_SHORT:
-			val = *((u16 *) setting->data);
-			break;
-		case TYPE_INT:
-			val = *((u32 *) setting->data);
-			break;
-		}
+		val = ds->get(drive);
 		spin_unlock_irqrestore(&ide_lock, flags);
 	}
+
 	return val;
 }
 
@@ -313,33 +184,23 @@ static int ide_read_setting(ide_drive_t *drive, ide_settings_t *setting)
  *	The current scheme of polling is kludgy, though safe enough.
  */
 
-static int ide_write_setting(ide_drive_t *drive, ide_settings_t *setting, int val)
+static int ide_write_setting(ide_drive_t *drive,
+			     const struct ide_proc_devset *setting, int val)
 {
+	const struct ide_devset *ds = setting->setting;
+
 	if (!capable(CAP_SYS_ADMIN))
 		return -EACCES;
-	if (setting->set)
-		return setting->set(drive, val);
-	if (!(setting->rw & SETTING_WRITE))
+	if (!ds->set)
 		return -EPERM;
-	if (val < setting->min || val > setting->max)
+	if ((ds->flags & DS_SYNC)
+	    && (val < setting->min || val > setting->max))
 		return -EINVAL;
-	if (ide_spin_wait_hwgroup(drive))
-		return -EBUSY;
-	switch (setting->data_type) {
-	case TYPE_BYTE:
-		*((u8 *) setting->data) = val;
-		break;
-	case TYPE_SHORT:
-		*((u16 *) setting->data) = val;
-		break;
-	case TYPE_INT:
-		*((u32 *) setting->data) = val;
-		break;
-	}
-	spin_unlock_irq(&ide_lock);
-	return 0;
+	return ide_devset_execute(drive, ds, val);
 }
 
+ide_devset_get(xfer_rate, current_speed);
+
 static int set_xfer_rate (ide_drive_t *drive, int arg)
 {
 	ide_task_t task;
@@ -349,7 +210,7 @@ static int set_xfer_rate (ide_drive_t *drive, int arg)
 		return -EINVAL;
 
 	memset(&task, 0, sizeof(task));
-	task.tf.command = WIN_SETFEATURES;
+	task.tf.command = ATA_CMD_SET_FEATURES;
 	task.tf.feature = SETFEATURES_XFER;
 	task.tf.nsect   = (u8)arg;
 	task.tf_flags = IDE_TFLAG_OUT_FEATURE | IDE_TFLAG_OUT_NSECT |
@@ -364,29 +225,23 @@ static int set_xfer_rate (ide_drive_t *drive, int arg)
 	return err;
 }
 
-/**
- *	ide_add_generic_settings	-	generic ide settings
- *	@drive: drive being configured
- *
- *	Add the generic parts of the system settings to the /proc files.
- *	The caller must not be holding the ide_setting_mtx.
- */
-
-void ide_add_generic_settings (ide_drive_t *drive)
-{
-/*
- *			  drive		setting name		read/write access				data type	min	max				mul_factor	div_factor	data pointer			set function
- */
-	__ide_add_setting(drive,	"io_32bit",		drive->no_io_32bit ? SETTING_READ : SETTING_RW,	TYPE_BYTE,	0,	1 + (SUPPORT_VLB_SYNC << 1),	1,		1,		&drive->io_32bit,		set_io_32bit,	0);
-	__ide_add_setting(drive,	"keepsettings",		SETTING_RW,					TYPE_BYTE,	0,	1,				1,		1,		&drive->keep_settings,		NULL,		0);
-	__ide_add_setting(drive,	"nice1",		SETTING_RW,					TYPE_BYTE,	0,	1,				1,		1,		&drive->nice1,			NULL,		0);
-	__ide_add_setting(drive,	"pio_mode",		SETTING_WRITE,					TYPE_BYTE,	0,	255,				1,		1,		NULL,				set_pio_mode,	0);
-	__ide_add_setting(drive,	"unmaskirq",		drive->no_unmask ? SETTING_READ : SETTING_RW,	TYPE_BYTE,	0,	1,				1,		1,		&drive->unmask,			NULL,		0);
-	__ide_add_setting(drive,	"using_dma",		SETTING_RW,					TYPE_BYTE,	0,	1,				1,		1,		&drive->using_dma,		set_using_dma,	0);
-	__ide_add_setting(drive,	"init_speed",		SETTING_RW,					TYPE_BYTE,	0,	70,				1,		1,		&drive->init_speed,		NULL,		0);
-	__ide_add_setting(drive,	"current_speed",	SETTING_RW,					TYPE_BYTE,	0,	70,				1,		1,		&drive->current_speed,		set_xfer_rate,	0);
-	__ide_add_setting(drive,	"number",		SETTING_RW,					TYPE_BYTE,	0,	3,				1,		1,		&drive->dn,			NULL,		0);
-}
+ide_devset_rw(current_speed, xfer_rate);
+ide_devset_rw_field(init_speed, init_speed);
+ide_devset_rw_field(nice1, nice1);
+ide_devset_rw_field(number, dn);
+
+static const struct ide_proc_devset ide_generic_settings[] = {
+	IDE_PROC_DEVSET(current_speed, 0, 70),
+	IDE_PROC_DEVSET(init_speed, 0, 70),
+	IDE_PROC_DEVSET(io_32bit,  0, 1 + (SUPPORT_VLB_SYNC << 1)),
+	IDE_PROC_DEVSET(keepsettings, 0, 1),
+	IDE_PROC_DEVSET(nice1, 0, 1),
+	IDE_PROC_DEVSET(number, 0, 3),
+	IDE_PROC_DEVSET(pio_mode, 0, 255),
+	IDE_PROC_DEVSET(unmaskirq, 0, 1),
+	IDE_PROC_DEVSET(using_dma, 0, 1),
+	{ 0 },
+};
 
 static void proc_ide_settings_warn(void)
 {
@@ -403,19 +258,32 @@ static void proc_ide_settings_warn(void)
 static int proc_ide_read_settings
 	(char *page, char **start, off_t off, int count, int *eof, void *data)
 {
+	const struct ide_proc_devset *setting, *g, *d;
+	const struct ide_devset *ds;
 	ide_drive_t	*drive = (ide_drive_t *) data;
-	ide_settings_t	*setting = (ide_settings_t *) drive->settings;
 	char		*out = page;
 	int		len, rc, mul_factor, div_factor;
 
 	proc_ide_settings_warn();
 
 	mutex_lock(&ide_setting_mtx);
+	g = ide_generic_settings;
+	d = drive->settings;
 	out += sprintf(out, "name\t\t\tvalue\t\tmin\t\tmax\t\tmode\n");
 	out += sprintf(out, "----\t\t\t-----\t\t---\t\t---\t\t----\n");
-	while (setting) {
-		mul_factor = setting->mul_factor;
-		div_factor = setting->div_factor;
+	while (g->name || (d && d->name)) {
+		/* read settings in the alphabetical order */
+		if (g->name && d && d->name) {
+			if (strcmp(d->name, g->name) < 0)
+				setting = d++;
+			else
+				setting = g++;
+		} else if (d && d->name) {
+			setting = d++;
+		} else
+			setting = g++;
+		mul_factor = setting->mulf ? setting->mulf(drive) : 1;
+		div_factor = setting->divf ? setting->divf(drive) : 1;
 		out += sprintf(out, "%-24s", setting->name);
 		rc = ide_read_setting(drive, setting);
 		if (rc >= 0)
@@ -423,12 +291,12 @@ static int proc_ide_read_settings
 		else
 			out += sprintf(out, "%-16s", "write-only");
 		out += sprintf(out, "%-16d%-16d", (setting->min * mul_factor + div_factor - 1) / div_factor, setting->max * mul_factor / div_factor);
-		if (setting->rw & SETTING_READ)
+		ds = setting->setting;
+		if (ds->get)
 			out += sprintf(out, "r");
-		if (setting->rw & SETTING_WRITE)
+		if (ds->set)
 			out += sprintf(out, "w");
 		out += sprintf(out, "\n");
-		setting = setting->next;
 	}
 	len = out - page;
 	mutex_unlock(&ide_setting_mtx);
@@ -442,9 +310,10 @@ static int proc_ide_write_settings(struct file *file, const char __user *buffer,
 {
 	ide_drive_t	*drive = (ide_drive_t *) data;
 	char		name[MAX_LEN + 1];
-	int		for_real = 0;
+	int		for_real = 0, mul_factor, div_factor;
 	unsigned long	n;
-	ide_settings_t	*setting;
+
+	const struct ide_proc_devset *setting;
 	char *buf, *s;
 
 	if (!capable(CAP_SYS_ADMIN))
@@ -512,13 +381,21 @@ static int proc_ide_write_settings(struct file *file, const char __user *buffer,
 			}
 
 			mutex_lock(&ide_setting_mtx);
-			setting = ide_find_setting_by_name(drive, name);
+			/* generic settings first, then driver specific ones */
+			setting = ide_find_setting(ide_generic_settings, name);
 			if (!setting) {
-				mutex_unlock(&ide_setting_mtx);
-				goto parse_error;
+				if (drive->settings)
+					setting = ide_find_setting(drive->settings, name);
+				if (!setting) {
+					mutex_unlock(&ide_setting_mtx);
+					goto parse_error;
+				}
+			}
+			if (for_real) {
+				mul_factor = setting->mulf ? setting->mulf(drive) : 1;
+				div_factor = setting->divf ? setting->divf(drive) : 1;
+				ide_write_setting(drive, setting, val * div_factor / mul_factor);
 			}
-			if (for_real)
-				ide_write_setting(drive, setting, val * setting->div_factor / setting->mul_factor);
 			mutex_unlock(&ide_setting_mtx);
 		}
 	} while (!for_real++);
@@ -561,11 +438,10 @@ static int proc_ide_read_dmodel
 	(char *page, char **start, off_t off, int count, int *eof, void *data)
 {
 	ide_drive_t	*drive = (ide_drive_t *) data;
-	struct hd_driveid *id = drive->id;
+	char		*m = (char *)&drive->id[ATA_ID_PROD];
 	int		len;
 
-	len = sprintf(page, "%.40s\n",
-		(id && id->model[0]) ? (char *)id->model : "(none)");
+	len = sprintf(page, "%.40s\n", m[0] ? m : "(none)");
 	PROC_IDE_READ_RETURN(page, start, off, count, eof, len);
 }
 
@@ -690,6 +566,10 @@ static void ide_remove_proc_entries(struct proc_dir_entry *dir, ide_proc_entry_t
 
 void ide_proc_register_driver(ide_drive_t *drive, ide_driver_t *driver)
 {
+	mutex_lock(&ide_setting_mtx);
+	drive->settings = driver->settings;
+	mutex_unlock(&ide_setting_mtx);
+
 	ide_add_proc_entries(drive->proc, driver->proc, drive);
 }
 
@@ -726,7 +606,7 @@ void ide_proc_unregister_driver(ide_drive_t *drive, ide_driver_t *driver)
 	 * OTOH both ide_{read,write}_setting are only ever used under
 	 * ide_setting_mtx.
 	 */
-	auto_remove_settings(drive);
+	drive->settings = NULL;
 	spin_unlock_irqrestore(&ide_lock, flags);
 	mutex_unlock(&ide_setting_mtx);
 }
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 3833189144ed..f8c84df4a0bc 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -15,6 +15,8 @@
  * Documentation/ide/ChangeLog.ide-tape.1995-2002
  */
 
+#define DRV_NAME "ide-tape"
+
 #define IDETAPE_VERSION "1.20"
 
 #include <linux/module.h>
@@ -54,8 +56,6 @@ enum {
 	DBG_CHRDEV =		(1 << 2),
 	/* all remaining procedures */
 	DBG_PROCS =		(1 << 3),
-	/* buffer alloc info (pc_stack & rq_stack) */
-	DBG_PCRQ_STACK =	(1 << 4),
 };
 
 /* define to see debug info */
@@ -81,26 +81,6 @@ enum {
 #define IDETAPE_MAX_PC_RETRIES		3
 
 /*
- * With each packet command, we allocate a buffer of IDETAPE_PC_BUFFER_SIZE
- * bytes. This is used for several packet commands (Not for READ/WRITE commands)
- */
-#define IDETAPE_PC_BUFFER_SIZE		256
-
-/*
- *	In various places in the driver, we need to allocate storage
- *	for packet commands and requests, which will remain valid while
- *	we leave the driver to wait for an interrupt or a timeout event.
- */
-#define IDETAPE_PC_STACK		(10 + IDETAPE_MAX_PC_RETRIES)
-
-/*
- * Some drives (for example, Seagate STT3401A Travan) require a very long
- * timeout, because they don't return an interrupt or clear their busy bit
- * until after the command completes (even retension commands).
- */
-#define IDETAPE_WAIT_CMD		(900*HZ)
-
-/*
  * The following parameter is used to select the point in the internal tape fifo
  * in which we will start to refill the buffer. Decreasing the following
  * parameter will improve the system's latency and interactive response, while
@@ -172,20 +152,6 @@ struct idetape_bh {
 #define IDETAPE_LU_RETENSION_MASK	2
 #define IDETAPE_LU_EOT_MASK		4
 
-/*
- * Special requests for our block device strategy routine.
- *
- * In order to service a character device command, we add special requests to
- * the tail of our block device request queue and wait for their completion.
- */
-
-enum {
-	REQ_IDETAPE_PC1		= (1 << 0), /* packet command (first stage) */
-	REQ_IDETAPE_PC2		= (1 << 1), /* packet command (second stage) */
-	REQ_IDETAPE_READ	= (1 << 2),
-	REQ_IDETAPE_WRITE	= (1 << 3),
-};
-
 /* Error codes returned in rq->errors to the higher part of the driver. */
 #define IDETAPE_ERROR_GENERAL		101
 #define IDETAPE_ERROR_FILEMARK		102
@@ -206,13 +172,6 @@ typedef struct ide_tape_obj {
 	struct kref	kref;
 
 	/*
-	 *	Since a typical character device operation requires more
-	 *	than one packet command, we provide here enough memory
-	 *	for the maximum of interconnected packet commands.
-	 *	The packet commands are stored in the circular array pc_stack.
-	 *	pc_stack_index points to the last used entry, and warps around
-	 *	to the start when we get to the last array entry.
-	 *
 	 *	pc points to the current processed packet command.
 	 *
 	 *	failed_pc points to the last failed packet command, or contains
@@ -224,13 +183,11 @@ typedef struct ide_tape_obj {
 	struct ide_atapi_pc *pc;
 	/* Last failed packet command */
 	struct ide_atapi_pc *failed_pc;
-	/* Packet command stack */
-	struct ide_atapi_pc pc_stack[IDETAPE_PC_STACK];
-	/* Next free packet command storage space */
-	int pc_stack_index;
-	struct request rq_stack[IDETAPE_PC_STACK];
-	/* We implement a circular array */
-	int rq_stack_index;
+	/* used by REQ_IDETAPE_{READ,WRITE} requests */
+	struct ide_atapi_pc queued_pc;
+
+	struct ide_atapi_pc request_sense_pc;
+	struct request request_sense_rq;
 
 	/*
 	 * DSC polling variables.
@@ -451,47 +408,6 @@ static void idetape_update_buffers(ide_drive_t *drive, struct ide_atapi_pc *pc)
 }
 
 /*
- *	idetape_next_pc_storage returns a pointer to a place in which we can
- *	safely store a packet command, even though we intend to leave the
- *	driver. A storage space for a maximum of IDETAPE_PC_STACK packet
- *	commands is allocated at initialization time.
- */
-static struct ide_atapi_pc *idetape_next_pc_storage(ide_drive_t *drive)
-{
-	idetape_tape_t *tape = drive->driver_data;
-
-	debug_log(DBG_PCRQ_STACK, "pc_stack_index=%d\n", tape->pc_stack_index);
-
-	if (tape->pc_stack_index == IDETAPE_PC_STACK)
-		tape->pc_stack_index = 0;
-	return (&tape->pc_stack[tape->pc_stack_index++]);
-}
-
-/*
- *	idetape_next_rq_storage is used along with idetape_next_pc_storage.
- *	Since we queue packet commands in the request queue, we need to
- *	allocate a request, along with the allocation of a packet command.
- */
-
-/**************************************************************
- *                                                            *
- *  This should get fixed to use kmalloc(.., GFP_ATOMIC)      *
- *  followed later on by kfree().   -ml                       *
- *                                                            *
- **************************************************************/
-
-static struct request *idetape_next_rq_storage(ide_drive_t *drive)
-{
-	idetape_tape_t *tape = drive->driver_data;
-
-	debug_log(DBG_PCRQ_STACK, "rq_stack_index=%d\n", tape->rq_stack_index);
-
-	if (tape->rq_stack_index == IDETAPE_PC_STACK)
-		tape->rq_stack_index = 0;
-	return (&tape->rq_stack[tape->rq_stack_index++]);
-}
-
-/*
  * called on each failed packet command retry to analyze the request sense. We
  * currently do not utilize this information.
  */
@@ -667,61 +583,14 @@ static void ide_tape_callback(ide_drive_t *drive)
 	idetape_end_request(drive, uptodate, 0);
 }
 
-static void idetape_init_pc(struct ide_atapi_pc *pc)
-{
-	memset(pc->c, 0, 12);
-	pc->retries = 0;
-	pc->flags = 0;
-	pc->req_xfer = 0;
-	pc->buf = pc->pc_buf;
-	pc->buf_size = IDETAPE_PC_BUFFER_SIZE;
-	pc->bh = NULL;
-	pc->b_data = NULL;
-}
-
 static void idetape_create_request_sense_cmd(struct ide_atapi_pc *pc)
 {
-	idetape_init_pc(pc);
+	ide_init_pc(pc);
 	pc->c[0] = REQUEST_SENSE;
 	pc->c[4] = 20;
 	pc->req_xfer = 20;
 }
 
-static void idetape_init_rq(struct request *rq, u8 cmd)
-{
-	blk_rq_init(NULL, rq);
-	rq->cmd_type = REQ_TYPE_SPECIAL;
-	rq->cmd[13] = cmd;
-}
-
-/*
- * Generate a new packet command request in front of the request queue, before
- * the current request, so that it will be processed immediately, on the next
- * pass through the driver. The function below is called from the request
- * handling part of the driver (the "bottom" part). Safe storage for the request
- * should be allocated with ide_tape_next_{pc,rq}_storage() prior to that.
- *
- * Memory for those requests is pre-allocated at initialization time, and is
- * limited to IDETAPE_PC_STACK requests. We assume that we have enough space for
- * the maximum possible number of inter-dependent packet commands.
- *
- * The higher level of the driver - The ioctl handler and the character device
- * handling functions should queue request to the lower level part and wait for
- * their completion using idetape_queue_pc_tail or idetape_queue_rw_tail.
- */
-static void idetape_queue_pc_head(ide_drive_t *drive, struct ide_atapi_pc *pc,
-				  struct request *rq)
-{
-	struct ide_tape_obj *tape = drive->driver_data;
-
-	idetape_init_rq(rq, REQ_IDETAPE_PC1);
-	rq->cmd_flags |= REQ_PREEMPT;
-	rq->buffer = (char *) pc;
-	rq->rq_disk = tape->disk;
-	memcpy(rq->cmd, pc->c, 12);
-	ide_do_drive_cmd(drive, rq);
-}
-
 /*
  *	idetape_retry_pc is called when an error was detected during the
  *	last packet command. We queue a request sense packet command in
@@ -729,15 +598,14 @@ static void idetape_queue_pc_head(ide_drive_t *drive, struct ide_atapi_pc *pc,
  */
 static void idetape_retry_pc(ide_drive_t *drive)
 {
-	struct ide_atapi_pc *pc;
-	struct request *rq;
+	struct ide_tape_obj *tape = drive->driver_data;
+	struct request *rq = &tape->request_sense_rq;
+	struct ide_atapi_pc *pc = &tape->request_sense_pc;
 
 	(void)ide_read_error(drive);
-	pc = idetape_next_pc_storage(drive);
-	rq = idetape_next_rq_storage(drive);
 	idetape_create_request_sense_cmd(pc);
 	set_bit(IDE_AFLAG_IGNORE_DSC, &drive->atapi_flags);
-	idetape_queue_pc_head(drive, pc, rq);
+	ide_queue_pc_head(drive, tape->disk, pc, rq);
 }
 
 /*
@@ -766,13 +634,15 @@ static void ide_tape_handle_dsc(ide_drive_t *drive)
 	idetape_postpone_request(drive);
 }
 
-static void ide_tape_io_buffers(ide_drive_t *drive, struct ide_atapi_pc *pc,
+static int ide_tape_io_buffers(ide_drive_t *drive, struct ide_atapi_pc *pc,
 				unsigned int bcount, int write)
 {
 	if (write)
 		idetape_output_buffers(drive, pc, bcount);
 	else
 		idetape_input_buffers(drive, pc, bcount);
+
+	return bcount;
 }
 
 /*
@@ -786,7 +656,7 @@ static ide_startstop_t idetape_pc_intr(ide_drive_t *drive)
 {
 	idetape_tape_t *tape = drive->driver_data;
 
-	return ide_pc_intr(drive, tape->pc, idetape_pc_intr, IDETAPE_WAIT_CMD,
+	return ide_pc_intr(drive, tape->pc, idetape_pc_intr, WAIT_TAPE_CMD,
 			   NULL, idetape_update_buffers, idetape_retry_pc,
 			   ide_tape_handle_dsc, ide_tape_io_buffers);
 }
@@ -832,7 +702,7 @@ static ide_startstop_t idetape_transfer_pc(ide_drive_t *drive)
 	idetape_tape_t *tape = drive->driver_data;
 
 	return ide_transfer_pc(drive, tape->pc, idetape_pc_intr,
-			       IDETAPE_WAIT_CMD, NULL);
+			       WAIT_TAPE_CMD, NULL);
 }
 
 static ide_startstop_t idetape_issue_pc(ide_drive_t *drive,
@@ -881,13 +751,13 @@ static ide_startstop_t idetape_issue_pc(ide_drive_t *drive,
 	pc->retries++;
 
 	return ide_issue_pc(drive, pc, idetape_transfer_pc,
-			    IDETAPE_WAIT_CMD, NULL);
+			    WAIT_TAPE_CMD, NULL);
 }
 
 /* A mode sense command is used to "sense" tape parameters. */
 static void idetape_create_mode_sense_cmd(struct ide_atapi_pc *pc, u8 page_code)
 {
-	idetape_init_pc(pc);
+	ide_init_pc(pc);
 	pc->c[0] = MODE_SENSE;
 	if (page_code != IDETAPE_BLOCK_DESCRIPTOR)
 		/* DBD = 1 - Don't return block descriptors */
@@ -920,8 +790,8 @@ static ide_startstop_t idetape_media_access_finished(ide_drive_t *drive)
 
 	stat = hwif->tp_ops->read_status(hwif);
 
-	if (stat & SEEK_STAT) {
-		if (stat & ERR_STAT) {
+	if (stat & ATA_DSC) {
+		if (stat & ATA_ERR) {
 			/* Error detected */
 			if (pc->c[0] != TEST_UNIT_READY)
 				printk(KERN_ERR "ide-tape: %s: I/O error, ",
@@ -946,7 +816,7 @@ static void ide_tape_create_rw_cmd(idetape_tape_t *tape,
 	struct idetape_bh *bh = (struct idetape_bh *)rq->special;
 	unsigned int length = rq->current_nr_sectors;
 
-	idetape_init_pc(pc);
+	ide_init_pc(pc);
 	put_unaligned(cpu_to_be32(length), (unsigned int *) &pc->c[1]);
 	pc->c[1] = 1;
 	pc->bh = bh;
@@ -978,9 +848,10 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive,
 	struct request *postponed_rq = tape->postponed_rq;
 	u8 stat;
 
-	debug_log(DBG_SENSE, "sector: %ld, nr_sectors: %ld,"
-			" current_nr_sectors: %d\n",
-			rq->sector, rq->nr_sectors, rq->current_nr_sectors);
+	debug_log(DBG_SENSE, "sector: %llu, nr_sectors: %lu,"
+			" current_nr_sectors: %u\n",
+			(unsigned long long)rq->sector, rq->nr_sectors,
+			rq->current_nr_sectors);
 
 	if (!blk_special_request(rq)) {
 		/* We do not support buffer cache originated requests. */
@@ -1021,7 +892,7 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive,
 	}
 
 	if (!test_and_clear_bit(IDE_AFLAG_IGNORE_DSC, &drive->atapi_flags) &&
-	    (stat & SEEK_STAT) == 0) {
+	    (stat & ATA_DSC) == 0) {
 		if (postponed_rq == NULL) {
 			tape->dsc_polling_start = jiffies;
 			tape->dsc_poll_freq = tape->best_dsc_rw_freq;
@@ -1043,12 +914,12 @@ static ide_startstop_t idetape_do_request(ide_drive_t *drive,
 		return ide_stopped;
 	}
 	if (rq->cmd[13] & REQ_IDETAPE_READ) {
-		pc = idetape_next_pc_storage(drive);
+		pc = &tape->queued_pc;
 		ide_tape_create_rw_cmd(tape, pc, rq, READ_6);
 		goto out;
 	}
 	if (rq->cmd[13] & REQ_IDETAPE_WRITE) {
-		pc = idetape_next_pc_storage(drive);
+		pc = &tape->queued_pc;
 		ide_tape_create_rw_cmd(tape, pc, rq, WRITE_6);
 		goto out;
 	}
@@ -1235,77 +1106,30 @@ static void idetape_init_merge_buffer(idetape_tape_t *tape)
 static void idetape_create_write_filemark_cmd(ide_drive_t *drive,
 		struct ide_atapi_pc *pc, int write_filemark)
 {
-	idetape_init_pc(pc);
+	ide_init_pc(pc);
 	pc->c[0] = WRITE_FILEMARKS;
 	pc->c[4] = write_filemark;
 	pc->flags |= PC_FLAG_WAIT_FOR_DSC;
 }
 
-static void idetape_create_test_unit_ready_cmd(struct ide_atapi_pc *pc)
-{
-	idetape_init_pc(pc);
-	pc->c[0] = TEST_UNIT_READY;
-}
-
-/*
- * We add a special packet command request to the tail of the request queue, and
- * wait for it to be serviced. This is not to be called from within the request
- * handling part of the driver! We allocate here data on the stack and it is
- * valid until the request is finished. This is not the case for the bottom part
- * of the driver, where we are always leaving the functions to wait for an
- * interrupt or a timer event.
- *
- * From the bottom part of the driver, we should allocate safe memory using
- * idetape_next_pc_storage() and ide_tape_next_rq_storage(), and add the request
- * to the request list without waiting for it to be serviced! In that case, we
- * usually use idetape_queue_pc_head().
- */
-static int idetape_queue_pc_tail(ide_drive_t *drive, struct ide_atapi_pc *pc)
-{
-	struct ide_tape_obj *tape = drive->driver_data;
-	struct request *rq;
-	int error;
-
-	rq = blk_get_request(drive->queue, READ, __GFP_WAIT);
-	rq->cmd_type = REQ_TYPE_SPECIAL;
-	rq->cmd[13] = REQ_IDETAPE_PC1;
-	rq->buffer = (char *)pc;
-	memcpy(rq->cmd, pc->c, 12);
-	error = blk_execute_rq(drive->queue, tape->disk, rq, 0);
-	blk_put_request(rq);
-	return error;
-}
-
-static void idetape_create_load_unload_cmd(ide_drive_t *drive,
-		struct ide_atapi_pc *pc, int cmd)
-{
-	idetape_init_pc(pc);
-	pc->c[0] = START_STOP;
-	pc->c[4] = cmd;
-	pc->flags |= PC_FLAG_WAIT_FOR_DSC;
-}
-
 static int idetape_wait_ready(ide_drive_t *drive, unsigned long timeout)
 {
 	idetape_tape_t *tape = drive->driver_data;
-	struct ide_atapi_pc pc;
+	struct gendisk *disk = tape->disk;
 	int load_attempted = 0;
 
 	/* Wait for the tape to become ready */
 	set_bit(IDE_AFLAG_MEDIUM_PRESENT, &drive->atapi_flags);
 	timeout += jiffies;
 	while (time_before(jiffies, timeout)) {
-		idetape_create_test_unit_ready_cmd(&pc);
-		if (!idetape_queue_pc_tail(drive, &pc))
+		if (ide_do_test_unit_ready(drive, disk) == 0)
 			return 0;
 		if ((tape->sense_key == 2 && tape->asc == 4 && tape->ascq == 2)
 		    || (tape->asc == 0x3A)) {
 			/* no media */
 			if (load_attempted)
 				return -ENOMEDIUM;
-			idetape_create_load_unload_cmd(drive, &pc,
-							IDETAPE_LU_LOAD_MASK);
-			idetape_queue_pc_tail(drive, &pc);
+			ide_do_start_stop(drive, disk, IDETAPE_LU_LOAD_MASK);
 			load_attempted = 1;
 		/* not about to be ready */
 		} else if (!(tape->sense_key == 2 && tape->asc == 4 &&
@@ -1318,11 +1142,12 @@ static int idetape_wait_ready(ide_drive_t *drive, unsigned long timeout)
 
 static int idetape_flush_tape_buffers(ide_drive_t *drive)
 {
+	struct ide_tape_obj *tape = drive->driver_data;
 	struct ide_atapi_pc pc;
 	int rc;
 
 	idetape_create_write_filemark_cmd(drive, &pc, 0);
-	rc = idetape_queue_pc_tail(drive, &pc);
+	rc = ide_queue_pc_tail(drive, tape->disk, &pc);
 	if (rc)
 		return rc;
 	idetape_wait_ready(drive, 60 * 5 * HZ);
@@ -1331,7 +1156,7 @@ static int idetape_flush_tape_buffers(ide_drive_t *drive)
 
 static void idetape_create_read_position_cmd(struct ide_atapi_pc *pc)
 {
-	idetape_init_pc(pc);
+	ide_init_pc(pc);
 	pc->c[0] = READ_POSITION;
 	pc->req_xfer = 20;
 }
@@ -1345,7 +1170,7 @@ static int idetape_read_position(ide_drive_t *drive)
 	debug_log(DBG_PROCS, "Enter %s\n", __func__);
 
 	idetape_create_read_position_cmd(&pc);
-	if (idetape_queue_pc_tail(drive, &pc))
+	if (ide_queue_pc_tail(drive, tape->disk, &pc))
 		return -1;
 	position = tape->first_frame;
 	return position;
@@ -1355,7 +1180,7 @@ static void idetape_create_locate_cmd(ide_drive_t *drive,
 		struct ide_atapi_pc *pc,
 		unsigned int block, u8 partition, int skip)
 {
-	idetape_init_pc(pc);
+	ide_init_pc(pc);
 	pc->c[0] = POSITION_TO_ELEMENT;
 	pc->c[1] = 2;
 	put_unaligned(cpu_to_be32(block), (unsigned int *) &pc->c[3]);
@@ -1363,21 +1188,6 @@ static void idetape_create_locate_cmd(ide_drive_t *drive,
 	pc->flags |= PC_FLAG_WAIT_FOR_DSC;
 }
 
-static int idetape_create_prevent_cmd(ide_drive_t *drive,
-		struct ide_atapi_pc *pc, int prevent)
-{
-	idetape_tape_t *tape = drive->driver_data;
-
-	/* device supports locking according to capabilities page */
-	if (!(tape->caps[6] & 0x01))
-		return 0;
-
-	idetape_init_pc(pc);
-	pc->c[0] = ALLOW_MEDIUM_REMOVAL;
-	pc->c[4] = prevent;
-	return 1;
-}
-
 static void __ide_tape_discard_merge_buffer(ide_drive_t *drive)
 {
 	idetape_tape_t *tape = drive->driver_data;
@@ -1405,6 +1215,7 @@ static int idetape_position_tape(ide_drive_t *drive, unsigned int block,
 		u8 partition, int skip)
 {
 	idetape_tape_t *tape = drive->driver_data;
+	struct gendisk *disk = tape->disk;
 	int retval;
 	struct ide_atapi_pc pc;
 
@@ -1412,12 +1223,12 @@ static int idetape_position_tape(ide_drive_t *drive, unsigned int block,
 		__ide_tape_discard_merge_buffer(drive);
 	idetape_wait_ready(drive, 60 * 5 * HZ);
 	idetape_create_locate_cmd(drive, &pc, block, partition, skip);
-	retval = idetape_queue_pc_tail(drive, &pc);
+	retval = ide_queue_pc_tail(drive, disk, &pc);
 	if (retval)
 		return (retval);
 
 	idetape_create_read_position_cmd(&pc);
-	return (idetape_queue_pc_tail(drive, &pc));
+	return ide_queue_pc_tail(drive, disk, &pc);
 }
 
 static void ide_tape_discard_merge_buffer(ide_drive_t *drive,
@@ -1477,7 +1288,7 @@ static int idetape_queue_rw_tail(ide_drive_t *drive, int cmd, int blocks,
 
 static void idetape_create_inquiry_cmd(struct ide_atapi_pc *pc)
 {
-	idetape_init_pc(pc);
+	ide_init_pc(pc);
 	pc->c[0] = INQUIRY;
 	pc->c[4] = 254;
 	pc->req_xfer = 254;
@@ -1486,14 +1297,14 @@ static void idetape_create_inquiry_cmd(struct ide_atapi_pc *pc)
 static void idetape_create_rewind_cmd(ide_drive_t *drive,
 		struct ide_atapi_pc *pc)
 {
-	idetape_init_pc(pc);
+	ide_init_pc(pc);
 	pc->c[0] = REZERO_UNIT;
 	pc->flags |= PC_FLAG_WAIT_FOR_DSC;
 }
 
 static void idetape_create_erase_cmd(struct ide_atapi_pc *pc)
 {
-	idetape_init_pc(pc);
+	ide_init_pc(pc);
 	pc->c[0] = ERASE;
 	pc->c[1] = 1;
 	pc->flags |= PC_FLAG_WAIT_FOR_DSC;
@@ -1501,7 +1312,7 @@ static void idetape_create_erase_cmd(struct ide_atapi_pc *pc)
 
 static void idetape_create_space_cmd(struct ide_atapi_pc *pc, int count, u8 cmd)
 {
-	idetape_init_pc(pc);
+	ide_init_pc(pc);
 	pc->c[0] = SPACE;
 	put_unaligned(cpu_to_be32(count), (unsigned int *) &pc->c[1]);
 	pc->c[1] = cmd;
@@ -1664,20 +1475,20 @@ static void idetape_pad_zeros(ide_drive_t *drive, int bcount)
  */
 static int idetape_rewind_tape(ide_drive_t *drive)
 {
+	struct ide_tape_obj *tape = drive->driver_data;
+	struct gendisk *disk = tape->disk;
 	int retval;
 	struct ide_atapi_pc pc;
-	idetape_tape_t *tape;
-	tape = drive->driver_data;
 
 	debug_log(DBG_SENSE, "Enter %s\n", __func__);
 
 	idetape_create_rewind_cmd(drive, &pc);
-	retval = idetape_queue_pc_tail(drive, &pc);
+	retval = ide_queue_pc_tail(drive, disk, &pc);
 	if (retval)
 		return retval;
 
 	idetape_create_read_position_cmd(&pc);
-	retval = idetape_queue_pc_tail(drive, &pc);
+	retval = ide_queue_pc_tail(drive, disk, &pc);
 	if (retval)
 		return retval;
 	return 0;
@@ -1720,6 +1531,7 @@ static int idetape_space_over_filemarks(ide_drive_t *drive, short mt_op,
 					int mt_count)
 {
 	idetape_tape_t *tape = drive->driver_data;
+	struct gendisk *disk = tape->disk;
 	struct ide_atapi_pc pc;
 	int retval, count = 0;
 	int sprev = !!(tape->caps[4] & 0x20);
@@ -1744,7 +1556,7 @@ static int idetape_space_over_filemarks(ide_drive_t *drive, short mt_op,
 	case MTBSF:
 		idetape_create_space_cmd(&pc, mt_count - count,
 					 IDETAPE_SPACE_OVER_FILEMARK);
-		return idetape_queue_pc_tail(drive, &pc);
+		return ide_queue_pc_tail(drive, disk, &pc);
 	case MTFSFM:
 	case MTBSFM:
 		if (!sprev)
@@ -1933,11 +1745,12 @@ static ssize_t idetape_chrdev_write(struct file *file, const char __user *buf,
 
 static int idetape_write_filemark(ide_drive_t *drive)
 {
+	struct ide_tape_obj *tape = drive->driver_data;
 	struct ide_atapi_pc pc;
 
 	/* Write a filemark */
 	idetape_create_write_filemark_cmd(drive, &pc, 1);
-	if (idetape_queue_pc_tail(drive, &pc)) {
+	if (ide_queue_pc_tail(drive, tape->disk, &pc)) {
 		printk(KERN_ERR "ide-tape: Couldn't write a filemark\n");
 		return -EIO;
 	}
@@ -1960,6 +1773,7 @@ static int idetape_write_filemark(ide_drive_t *drive)
 static int idetape_mtioctop(ide_drive_t *drive, short mt_op, int mt_count)
 {
 	idetape_tape_t *tape = drive->driver_data;
+	struct gendisk *disk = tape->disk;
 	struct ide_atapi_pc pc;
 	int i, retval;
 
@@ -1996,9 +1810,7 @@ static int idetape_mtioctop(ide_drive_t *drive, short mt_op, int mt_count)
 		return 0;
 	case MTLOAD:
 		ide_tape_discard_merge_buffer(drive, 0);
-		idetape_create_load_unload_cmd(drive, &pc,
-					       IDETAPE_LU_LOAD_MASK);
-		return idetape_queue_pc_tail(drive, &pc);
+		return ide_do_start_stop(drive, disk, IDETAPE_LU_LOAD_MASK);
 	case MTUNLOAD:
 	case MTOFFL:
 		/*
@@ -2006,14 +1818,11 @@ static int idetape_mtioctop(ide_drive_t *drive, short mt_op, int mt_count)
 		 * attempting to eject.
 		 */
 		if (tape->door_locked) {
-			if (idetape_create_prevent_cmd(drive, &pc, 0))
-				if (!idetape_queue_pc_tail(drive, &pc))
-					tape->door_locked = DOOR_UNLOCKED;
+			if (!ide_set_media_lock(drive, disk, 0))
+				tape->door_locked = DOOR_UNLOCKED;
 		}
 		ide_tape_discard_merge_buffer(drive, 0);
-		idetape_create_load_unload_cmd(drive, &pc,
-					      !IDETAPE_LU_LOAD_MASK);
-		retval = idetape_queue_pc_tail(drive, &pc);
+		retval = ide_do_start_stop(drive, disk, !IDETAPE_LU_LOAD_MASK);
 		if (!retval)
 			clear_bit(IDE_AFLAG_MEDIUM_PRESENT, &drive->atapi_flags);
 		return retval;
@@ -2022,16 +1831,15 @@ static int idetape_mtioctop(ide_drive_t *drive, short mt_op, int mt_count)
 		return idetape_flush_tape_buffers(drive);
 	case MTRETEN:
 		ide_tape_discard_merge_buffer(drive, 0);
-		idetape_create_load_unload_cmd(drive, &pc,
+		return ide_do_start_stop(drive, disk,
 			IDETAPE_LU_RETENSION_MASK | IDETAPE_LU_LOAD_MASK);
-		return idetape_queue_pc_tail(drive, &pc);
 	case MTEOM:
 		idetape_create_space_cmd(&pc, 0, IDETAPE_SPACE_TO_EOD);
-		return idetape_queue_pc_tail(drive, &pc);
+		return ide_queue_pc_tail(drive, disk, &pc);
 	case MTERASE:
 		(void)idetape_rewind_tape(drive);
 		idetape_create_erase_cmd(&pc);
-		return idetape_queue_pc_tail(drive, &pc);
+		return ide_queue_pc_tail(drive, disk, &pc);
 	case MTSETBLK:
 		if (mt_count) {
 			if (mt_count < tape->blk_size ||
@@ -2052,17 +1860,13 @@ static int idetape_mtioctop(ide_drive_t *drive, short mt_op, int mt_count)
 	case MTFSR:
 	case MTBSR:
 	case MTLOCK:
-		if (!idetape_create_prevent_cmd(drive, &pc, 1))
-			return 0;
-		retval = idetape_queue_pc_tail(drive, &pc);
+		retval = ide_set_media_lock(drive, disk, 1);
 		if (retval)
 			return retval;
 		tape->door_locked = DOOR_EXPLICITLY_LOCKED;
 		return 0;
 	case MTUNLOCK:
-		if (!idetape_create_prevent_cmd(drive, &pc, 0))
-			return 0;
-		retval = idetape_queue_pc_tail(drive, &pc);
+		retval = ide_set_media_lock(drive, disk, 0);
 		if (retval)
 			return retval;
 		tape->door_locked = DOOR_UNLOCKED;
@@ -2144,7 +1948,7 @@ static void ide_tape_get_bsize_from_bdesc(ide_drive_t *drive)
 	struct ide_atapi_pc pc;
 
 	idetape_create_mode_sense_cmd(&pc, IDETAPE_BLOCK_DESCRIPTOR);
-	if (idetape_queue_pc_tail(drive, &pc)) {
+	if (ide_queue_pc_tail(drive, tape->disk, &pc)) {
 		printk(KERN_ERR "ide-tape: Can't get block descriptor\n");
 		if (tape->blk_size == 0) {
 			printk(KERN_WARNING "ide-tape: Cannot deal with zero "
@@ -2164,7 +1968,6 @@ static int idetape_chrdev_open(struct inode *inode, struct file *filp)
 	unsigned int minor = iminor(inode), i = minor & ~0xc0;
 	ide_drive_t *drive;
 	idetape_tape_t *tape;
-	struct ide_atapi_pc pc;
 	int retval;
 
 	if (i >= MAX_HWIFS * MAX_DRIVES)
@@ -2227,11 +2030,9 @@ static int idetape_chrdev_open(struct inode *inode, struct file *filp)
 
 	/* Lock the tape drive door so user can't eject. */
 	if (tape->chrdev_dir == IDETAPE_DIR_NONE) {
-		if (idetape_create_prevent_cmd(drive, &pc, 1)) {
-			if (!idetape_queue_pc_tail(drive, &pc)) {
-				if (tape->door_locked != DOOR_EXPLICITLY_LOCKED)
-					tape->door_locked = DOOR_LOCKED;
-			}
+		if (!ide_set_media_lock(drive, tape->disk, 1)) {
+			if (tape->door_locked != DOOR_EXPLICITLY_LOCKED)
+				tape->door_locked = DOOR_LOCKED;
 		}
 	}
 	unlock_kernel();
@@ -2264,7 +2065,6 @@ static int idetape_chrdev_release(struct inode *inode, struct file *filp)
 {
 	struct ide_tape_obj *tape = ide_tape_f(filp);
 	ide_drive_t *drive = tape->drive;
-	struct ide_atapi_pc pc;
 	unsigned int minor = iminor(inode);
 
 	lock_kernel();
@@ -2283,10 +2083,8 @@ static int idetape_chrdev_release(struct inode *inode, struct file *filp)
 		(void) idetape_rewind_tape(drive);
 	if (tape->chrdev_dir == IDETAPE_DIR_NONE) {
 		if (tape->door_locked == DOOR_LOCKED) {
-			if (idetape_create_prevent_cmd(drive, &pc, 0)) {
-				if (!idetape_queue_pc_tail(drive, &pc))
-					tape->door_locked = DOOR_UNLOCKED;
-			}
+			if (!ide_set_media_lock(drive, tape->disk, 0))
+				tape->door_locked = DOOR_UNLOCKED;
 		}
 	}
 	clear_bit(IDE_AFLAG_BUSY, &drive->atapi_flags);
@@ -2295,45 +2093,6 @@ static int idetape_chrdev_release(struct inode *inode, struct file *filp)
 	return 0;
 }
 
-/*
- * check the contents of the ATAPI IDENTIFY command results. We return:
- *
- * 1 - If the tape can be supported by us, based on the information we have so
- * far.
- *
- * 0 - If this tape driver is not currently supported by us.
- */
-static int idetape_identify_device(ide_drive_t *drive)
-{
-	u8 gcw[2], protocol, device_type, removable, packet_size;
-
-	if (drive->id_read == 0)
-		return 1;
-
-	*((unsigned short *) &gcw) = drive->id->config;
-
-	protocol	=   (gcw[1] & 0xC0) >> 6;
-	device_type	=    gcw[1] & 0x1F;
-	removable	= !!(gcw[0] & 0x80);
-	packet_size	=    gcw[0] & 0x3;
-
-	/* Check that we can support this device */
-	if (protocol != 2)
-		printk(KERN_ERR "ide-tape: Protocol (0x%02x) is not ATAPI\n",
-				protocol);
-	else if (device_type != 1)
-		printk(KERN_ERR "ide-tape: Device type (0x%02x) is not set "
-				"to tape\n", device_type);
-	else if (!removable)
-		printk(KERN_ERR "ide-tape: The removable flag is not set\n");
-	else if (packet_size != 0) {
-		printk(KERN_ERR "ide-tape: Packet size (0x%02x) is not 12"
-				" bytes\n", packet_size);
-	} else
-		return 1;
-	return 0;
-}
-
 static void idetape_get_inquiry_results(ide_drive_t *drive)
 {
 	idetape_tape_t *tape = drive->driver_data;
@@ -2341,7 +2100,7 @@ static void idetape_get_inquiry_results(ide_drive_t *drive)
 	char fw_rev[4], vendor_id[8], product_id[16];
 
 	idetape_create_inquiry_cmd(&pc);
-	if (idetape_queue_pc_tail(drive, &pc)) {
+	if (ide_queue_pc_tail(drive, tape->disk, &pc)) {
 		printk(KERN_ERR "ide-tape: %s: can't get INQUIRY results\n",
 				tape->name);
 		return;
@@ -2370,7 +2129,7 @@ static void idetape_get_mode_sense_results(ide_drive_t *drive)
 	u8 speed, max_speed;
 
 	idetape_create_mode_sense_cmd(&pc, IDETAPE_CAPABILITIES_PAGE);
-	if (idetape_queue_pc_tail(drive, &pc)) {
+	if (ide_queue_pc_tail(drive, tape->disk, &pc)) {
 		printk(KERN_ERR "ide-tape: Can't get tape parameters - assuming"
 				" some default values\n");
 		tape->blk_size = 512;
@@ -2402,6 +2161,11 @@ static void idetape_get_mode_sense_results(ide_drive_t *drive)
 	}
 
 	memcpy(&tape->caps, caps, 20);
+
+	/* device lacks locking support according to capabilities page */
+	if ((caps[6] & 1) == 0)
+		drive->atapi_flags |= IDE_AFLAG_NO_DOORLOCK;
+
 	if (caps[7] & 0x02)
 		tape->blk_size = 512;
 	else if (caps[7] & 0x04)
@@ -2409,28 +2173,56 @@ static void idetape_get_mode_sense_results(ide_drive_t *drive)
 }
 
 #ifdef CONFIG_IDE_PROC_FS
-static void idetape_add_settings(ide_drive_t *drive)
-{
-	idetape_tape_t *tape = drive->driver_data;
-
-	ide_add_setting(drive, "buffer", SETTING_READ, TYPE_SHORT, 0, 0xffff,
-			1, 2, (u16 *)&tape->caps[16], NULL);
-	ide_add_setting(drive, "speed", SETTING_READ, TYPE_SHORT, 0, 0xffff,
-			1, 1, (u16 *)&tape->caps[14], NULL);
-	ide_add_setting(drive, "buffer_size", SETTING_READ, TYPE_INT, 0, 0xffff,
-			1, 1024, &tape->buffer_size, NULL);
-	ide_add_setting(drive, "tdsc", SETTING_RW, TYPE_INT, IDETAPE_DSC_RW_MIN,
-			IDETAPE_DSC_RW_MAX, 1000, HZ, &tape->best_dsc_rw_freq,
-			NULL);
-	ide_add_setting(drive, "dsc_overlap", SETTING_RW, TYPE_BYTE, 0, 1, 1,
-			1, &drive->dsc_overlap, NULL);
-	ide_add_setting(drive, "avg_speed", SETTING_READ, TYPE_INT, 0, 0xffff,
-			1, 1, &tape->avg_speed, NULL);
-	ide_add_setting(drive, "debug_mask", SETTING_RW, TYPE_INT, 0, 0xffff, 1,
-			1, &tape->debug_mask, NULL);
-}
-#else
-static inline void idetape_add_settings(ide_drive_t *drive) { ; }
+#define ide_tape_devset_get(name, field) \
+static int get_##name(ide_drive_t *drive) \
+{ \
+	idetape_tape_t *tape = drive->driver_data; \
+	return tape->field; \
+}
+
+#define ide_tape_devset_set(name, field) \
+static int set_##name(ide_drive_t *drive, int arg) \
+{ \
+	idetape_tape_t *tape = drive->driver_data; \
+	tape->field = arg; \
+	return 0; \
+}
+
+#define ide_tape_devset_rw_field(_name, _field) \
+ide_tape_devset_get(_name, _field) \
+ide_tape_devset_set(_name, _field) \
+IDE_DEVSET(_name, DS_SYNC, get_##_name, set_##_name)
+
+#define ide_tape_devset_r_field(_name, _field) \
+ide_tape_devset_get(_name, _field) \
+IDE_DEVSET(_name, 0, get_##_name, NULL)
+
+static int mulf_tdsc(ide_drive_t *drive)	{ return 1000; }
+static int divf_tdsc(ide_drive_t *drive)	{ return   HZ; }
+static int divf_buffer(ide_drive_t *drive)	{ return    2; }
+static int divf_buffer_size(ide_drive_t *drive)	{ return 1024; }
+
+ide_devset_rw_field(dsc_overlap, dsc_overlap);
+
+ide_tape_devset_rw_field(debug_mask, debug_mask);
+ide_tape_devset_rw_field(tdsc, best_dsc_rw_freq);
+
+ide_tape_devset_r_field(avg_speed, avg_speed);
+ide_tape_devset_r_field(speed, caps[14]);
+ide_tape_devset_r_field(buffer, caps[16]);
+ide_tape_devset_r_field(buffer_size, buffer_size);
+
+static const struct ide_proc_devset idetape_settings[] = {
+	__IDE_PROC_DEVSET(avg_speed,	0, 0xffff, NULL, NULL),
+	__IDE_PROC_DEVSET(buffer,	0, 0xffff, NULL, divf_buffer),
+	__IDE_PROC_DEVSET(buffer_size,	0, 0xffff, NULL, divf_buffer_size),
+	__IDE_PROC_DEVSET(debug_mask,	0, 0xffff, NULL, NULL),
+	__IDE_PROC_DEVSET(dsc_overlap,	0,      1, NULL, NULL),
+	__IDE_PROC_DEVSET(speed,	0, 0xffff, NULL, NULL),
+	__IDE_PROC_DEVSET(tdsc,		IDETAPE_DSC_RW_MIN, IDETAPE_DSC_RW_MAX,
+					mulf_tdsc, divf_tdsc),
+	{ 0 },
+};
 #endif
 
 /*
@@ -2462,15 +2254,15 @@ static void idetape_setup(ide_drive_t *drive, idetape_tape_t *tape, int minor)
 		drive->dsc_overlap = 0;
 	}
 	/* Seagate Travan drives do not support DSC overlap. */
-	if (strstr(drive->id->model, "Seagate STT3401"))
+	if (strstr((char *)&drive->id[ATA_ID_PROD], "Seagate STT3401"))
 		drive->dsc_overlap = 0;
 	tape->minor = minor;
 	tape->name[0] = 'h';
 	tape->name[1] = 't';
 	tape->name[2] = '0' + minor;
 	tape->chrdev_dir = IDETAPE_DIR_NONE;
-	tape->pc = tape->pc_stack;
-	*((unsigned short *) &gcw) = drive->id->config;
+
+	*((u16 *)&gcw) = drive->id[ATA_ID_CONFIG];
 
 	/* Command packet DRQ type */
 	if (((gcw[0] & 0x60) >> 5) == 1)
@@ -2512,7 +2304,7 @@ static void idetape_setup(ide_drive_t *drive, idetape_tape_t *tape, int minor)
 		tape->best_dsc_rw_freq * 1000 / HZ,
 		drive->using_dma ? ", DMA":"");
 
-	idetape_add_settings(drive);
+	ide_proc_register_driver(drive, tape->driver);
 }
 
 static void ide_tape_remove(ide_drive_t *drive)
@@ -2577,12 +2369,12 @@ static ide_driver_t idetape_driver = {
 	.remove			= ide_tape_remove,
 	.version		= IDETAPE_VERSION,
 	.media			= ide_tape,
-	.supports_dsc_overlap 	= 1,
 	.do_request		= idetape_do_request,
 	.end_request		= idetape_end_request,
 	.error			= __ide_error,
 #ifdef CONFIG_IDE_PROC_FS
 	.proc			= idetape_proc,
+	.settings		= idetape_settings,
 #endif
 };
 
@@ -2645,11 +2437,11 @@ static int ide_tape_probe(ide_drive_t *drive)
 
 	if (!strstr("ide-tape", drive->driver_req))
 		goto failed;
-	if (!drive->present)
-		goto failed;
+
 	if (drive->media != ide_tape)
 		goto failed;
-	if (!idetape_identify_device(drive)) {
+
+	if (drive->id_read == 1 && !ide_check_atapi_device(drive, DRV_NAME)) {
 		printk(KERN_ERR "ide-tape: %s: not supported by this version of"
 				" the driver\n", drive->name);
 		goto failed;
@@ -2667,8 +2459,6 @@ static int ide_tape_probe(ide_drive_t *drive)
 
 	ide_init_disk(g, drive);
 
-	ide_proc_register_driver(drive, &idetape_driver);
-
 	kref_init(&tape->kref);
 
 	tape->drive = drive;
diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 7fb6f1c86272..487b18b3ebae 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -44,9 +44,9 @@ int taskfile_lib_get_identify (ide_drive_t *drive, u8 *buf)
 	memset(&args, 0, sizeof(ide_task_t));
 	args.tf.nsect = 0x01;
 	if (drive->media == ide_disk)
-		args.tf.command = WIN_IDENTIFY;
+		args.tf.command = ATA_CMD_ID_ATA;
 	else
-		args.tf.command = WIN_PIDENTIFY;
+		args.tf.command = ATA_CMD_ID_ATAPI;
 	args.tf_flags	= IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
 	args.data_phase	= TASKFILE_IN;
 	return ide_raw_taskfile(drive, &args, buf, 1);
@@ -99,12 +99,17 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 	case TASKFILE_NO_DATA:
 		if (handler == NULL)
 			handler = task_no_data_intr;
-		/* WIN_{SPECIFY,RESTORE,SETMULT} use custom handlers */
 		if (task->tf_flags & IDE_TFLAG_CUSTOM_HANDLER) {
 			switch (tf->command) {
-			case WIN_SPECIFY: handler = set_geometry_intr;	break;
-			case WIN_RESTORE: handler = recal_intr;		break;
-			case WIN_SETMULT: handler = set_multmode_intr;	break;
+			case ATA_CMD_INIT_DEV_PARAMS:
+				handler = set_geometry_intr;
+				break;
+			case ATA_CMD_RESTORE:
+				handler = recal_intr;
+				break;
+			case ATA_CMD_SET_MULTI:
+				handler = set_multmode_intr;
+				break;
 			}
 		}
 		ide_execute_command(drive, tf->command, handler,
@@ -121,7 +126,7 @@ ide_startstop_t do_rw_taskfile (ide_drive_t *drive, ide_task_t *task)
 EXPORT_SYMBOL_GPL(do_rw_taskfile);
 
 /*
- * set_multmode_intr() is invoked on completion of a WIN_SETMULT cmd.
+ * set_multmode_intr() is invoked on completion of a ATA_CMD_SET_MULTI cmd.
  */
 static ide_startstop_t set_multmode_intr(ide_drive_t *drive)
 {
@@ -131,7 +136,7 @@ static ide_startstop_t set_multmode_intr(ide_drive_t *drive)
 	local_irq_enable_in_hardirq();
 	stat = hwif->tp_ops->read_status(hwif);
 
-	if (OK_STAT(stat, READY_STAT, BAD_STAT))
+	if (OK_STAT(stat, ATA_DRDY, BAD_STAT))
 		drive->mult_count = drive->mult_req;
 	else {
 		drive->mult_req = drive->mult_count = 0;
@@ -142,7 +147,7 @@ static ide_startstop_t set_multmode_intr(ide_drive_t *drive)
 }
 
 /*
- * set_geometry_intr() is invoked on completion of a WIN_SPECIFY cmd.
+ * set_geometry_intr() is invoked on completion of a ATA_CMD_INIT_DEV_PARAMS cmd.
  */
 static ide_startstop_t set_geometry_intr(ide_drive_t *drive)
 {
@@ -154,15 +159,15 @@ static ide_startstop_t set_geometry_intr(ide_drive_t *drive)
 
 	while (1) {
 		stat = hwif->tp_ops->read_status(hwif);
-		if ((stat & BUSY_STAT) == 0 || retries-- == 0)
+		if ((stat & ATA_BUSY) == 0 || retries-- == 0)
 			break;
 		udelay(10);
 	};
 
-	if (OK_STAT(stat, READY_STAT, BAD_STAT))
+	if (OK_STAT(stat, ATA_DRDY, BAD_STAT))
 		return ide_stopped;
 
-	if (stat & (ERR_STAT|DRQ_STAT))
+	if (stat & (ATA_ERR | ATA_DRQ))
 		return ide_error(drive, "set_geometry_intr", stat);
 
 	ide_set_handler(drive, &set_geometry_intr, WAIT_WORSTCASE, NULL);
@@ -170,7 +175,7 @@ static ide_startstop_t set_geometry_intr(ide_drive_t *drive)
 }
 
 /*
- * recal_intr() is invoked on completion of a WIN_RESTORE (recalibrate) cmd.
+ * recal_intr() is invoked on completion of a ATA_CMD_RESTORE (recalibrate) cmd.
  */
 static ide_startstop_t recal_intr(ide_drive_t *drive)
 {
@@ -180,7 +185,7 @@ static ide_startstop_t recal_intr(ide_drive_t *drive)
 	local_irq_enable_in_hardirq();
 	stat = hwif->tp_ops->read_status(hwif);
 
-	if (!OK_STAT(stat, READY_STAT, BAD_STAT))
+	if (!OK_STAT(stat, ATA_DRDY, BAD_STAT))
 		return ide_error(drive, "recal_intr", stat);
 	return ide_stopped;
 }
@@ -197,7 +202,7 @@ static ide_startstop_t task_no_data_intr(ide_drive_t *drive)
 	local_irq_enable_in_hardirq();
 	stat = hwif->tp_ops->read_status(hwif);
 
-	if (!OK_STAT(stat, READY_STAT, BAD_STAT))
+	if (!OK_STAT(stat, ATA_DRDY, BAD_STAT))
 		return ide_error(drive, "task_no_data_intr", stat);
 		/* calls ide_end_drive_cmd */
 
@@ -220,13 +225,13 @@ static u8 wait_drive_not_busy(ide_drive_t *drive)
 	for (retries = 0; retries < 1000; retries++) {
 		stat = hwif->tp_ops->read_status(hwif);
 
-		if (stat & BUSY_STAT)
+		if (stat & ATA_BUSY)
 			udelay(10);
 		else
 			break;
 	}
 
-	if (stat & BUSY_STAT)
+	if (stat & ATA_BUSY)
 		printk(KERN_ERR "%s: drive still BUSY!\n", drive->name);
 
 	return stat;
@@ -385,7 +390,7 @@ void task_end_request(ide_drive_t *drive, struct request *rq, u8 stat)
 static ide_startstop_t task_in_unexpected(ide_drive_t *drive, struct request *rq, u8 stat)
 {
 	/* Command all done? */
-	if (OK_STAT(stat, READY_STAT, BUSY_STAT)) {
+	if (OK_STAT(stat, ATA_DRDY, ATA_BUSY)) {
 		task_end_request(drive, rq, stat);
 		return ide_stopped;
 	}
@@ -405,11 +410,11 @@ static ide_startstop_t task_in_intr(ide_drive_t *drive)
 	u8 stat = hwif->tp_ops->read_status(hwif);
 
 	/* Error? */
-	if (stat & ERR_STAT)
+	if (stat & ATA_ERR)
 		return task_error(drive, rq, __func__, stat);
 
 	/* Didn't want any data? Odd. */
-	if (!(stat & DRQ_STAT))
+	if ((stat & ATA_DRQ) == 0)
 		return task_in_unexpected(drive, rq, stat);
 
 	ide_pio_datablock(drive, rq, 0);
@@ -442,7 +447,7 @@ static ide_startstop_t task_out_intr (ide_drive_t *drive)
 		return task_error(drive, rq, __func__, stat);
 
 	/* Deal with unexpected ATA data phase. */
-	if (((stat & DRQ_STAT) == 0) ^ !hwif->nleft)
+	if (((stat & ATA_DRQ) == 0) ^ !hwif->nleft)
 		return task_error(drive, rq, __func__, stat);
 
 	if (!hwif->nleft) {
@@ -461,7 +466,7 @@ static ide_startstop_t pre_task_out_intr(ide_drive_t *drive, struct request *rq)
 {
 	ide_startstop_t startstop;
 
-	if (ide_wait_stat(&startstop, drive, DRQ_STAT,
+	if (ide_wait_stat(&startstop, drive, ATA_DRQ,
 			  drive->bad_wstat, WAIT_DRQ)) {
 		printk(KERN_ERR "%s: no DRQ after issuing %sWRITE%s\n",
 				drive->name,
@@ -721,110 +726,3 @@ abort:
 	return err;
 }
 #endif
-
-int ide_cmd_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
-{
-	u8 *buf = NULL;
-	int bufsize = 0, err = 0;
-	u8 args[4], xfer_rate = 0;
-	ide_task_t tfargs;
-	struct ide_taskfile *tf = &tfargs.tf;
-	struct hd_driveid *id = drive->id;
-
-	if (NULL == (void *) arg) {
-		struct request *rq;
-
-		rq = blk_get_request(drive->queue, READ, __GFP_WAIT);
-		rq->cmd_type = REQ_TYPE_ATA_TASKFILE;
-		err = blk_execute_rq(drive->queue, NULL, rq, 0);
-		blk_put_request(rq);
-
-		return err;
-	}
-
-	if (copy_from_user(args, (void __user *)arg, 4))
-		return -EFAULT;
-
-	memset(&tfargs, 0, sizeof(ide_task_t));
-	tf->feature = args[2];
-	if (args[0] == WIN_SMART) {
-		tf->nsect = args[3];
-		tf->lbal  = args[1];
-		tf->lbam  = 0x4f;
-		tf->lbah  = 0xc2;
-		tfargs.tf_flags = IDE_TFLAG_OUT_TF | IDE_TFLAG_IN_NSECT;
-	} else {
-		tf->nsect = args[1];
-		tfargs.tf_flags = IDE_TFLAG_OUT_FEATURE |
-				  IDE_TFLAG_OUT_NSECT | IDE_TFLAG_IN_NSECT;
-	}
-	tf->command = args[0];
-	tfargs.data_phase = args[3] ? TASKFILE_IN : TASKFILE_NO_DATA;
-
-	if (args[3]) {
-		tfargs.tf_flags |= IDE_TFLAG_IO_16BIT;
-		bufsize = SECTOR_WORDS * 4 * args[3];
-		buf = kzalloc(bufsize, GFP_KERNEL);
-		if (buf == NULL)
-			return -ENOMEM;
-	}
-
-	if (tf->command == WIN_SETFEATURES &&
-	    tf->feature == SETFEATURES_XFER &&
-	    tf->nsect >= XFER_SW_DMA_0 &&
-	    (id->dma_ultra || id->dma_mword || id->dma_1word)) {
-		xfer_rate = args[1];
-		if (tf->nsect > XFER_UDMA_2 && !eighty_ninty_three(drive)) {
-			printk(KERN_WARNING "%s: UDMA speeds >UDMA33 cannot "
-					    "be set\n", drive->name);
-			goto abort;
-		}
-	}
-
-	err = ide_raw_taskfile(drive, &tfargs, buf, args[3]);
-
-	args[0] = tf->status;
-	args[1] = tf->error;
-	args[2] = tf->nsect;
-
-	if (!err && xfer_rate) {
-		/* active-retuning-calls future */
-		ide_set_xfer_rate(drive, xfer_rate);
-		ide_driveid_update(drive);
-	}
-abort:
-	if (copy_to_user((void __user *)arg, &args, 4))
-		err = -EFAULT;
-	if (buf) {
-		if (copy_to_user((void __user *)(arg + 4), buf, bufsize))
-			err = -EFAULT;
-		kfree(buf);
-	}
-	return err;
-}
-
-int ide_task_ioctl (ide_drive_t *drive, unsigned int cmd, unsigned long arg)
-{
-	void __user *p = (void __user *)arg;
-	int err = 0;
-	u8 args[7];
-	ide_task_t task;
-
-	if (copy_from_user(args, p, 7))
-		return -EFAULT;
-
-	memset(&task, 0, sizeof(task));
-	memcpy(&task.tf_array[7], &args[1], 6);
-	task.tf.command = args[0];
-	task.tf_flags = IDE_TFLAG_TF | IDE_TFLAG_DEVICE;
-
-	err = ide_no_data_taskfile(drive, &task);
-
-	args[0] = task.tf.command;
-	memcpy(&args[1], &task.tf_array[7], 6);
-
-	if (copy_to_user(p, args, 7))
-		err = -EFAULT;
-
-	return err;
-}
diff --git a/drivers/ide/ide-timings.c b/drivers/ide/ide-timings.c
index 8c2f8327f487..81f527af8fae 100644
--- a/drivers/ide/ide-timings.c
+++ b/drivers/ide/ide-timings.c
@@ -22,7 +22,6 @@
  */
 
 #include <linux/kernel.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/module.h>
 
@@ -78,15 +77,15 @@ EXPORT_SYMBOL_GPL(ide_timing_find_mode);
 
 u16 ide_pio_cycle_time(ide_drive_t *drive, u8 pio)
 {
-	struct hd_driveid *id = drive->id;
+	u16 *id = drive->id;
 	struct ide_timing *t = ide_timing_find_mode(XFER_PIO_0 + pio);
 	u16 cycle = 0;
 
-	if (id->field_valid & 2) {
-		if (id->capability & 8)
-			cycle = id->eide_pio_iordy;
+	if (id[ATA_ID_FIELD_VALID] & 2) {
+		if (ata_id_has_iordy(drive->id))
+			cycle = id[ATA_ID_EIDE_PIO_IORDY];
 		else
-			cycle = id->eide_pio;
+			cycle = id[ATA_ID_EIDE_PIO];
 
 		/* conservative "downgrade" for all pre-ATA2 drives */
 		if (pio < 3 && cycle < t->cycle)
@@ -138,7 +137,7 @@ EXPORT_SYMBOL_GPL(ide_timing_merge);
 int ide_timing_compute(ide_drive_t *drive, u8 speed,
 		       struct ide_timing *t, int T, int UT)
 {
-	struct hd_driveid *id = drive->id;
+	u16 *id = drive->id;
 	struct ide_timing *s, p;
 
 	/*
@@ -157,16 +156,15 @@ int ide_timing_compute(ide_drive_t *drive, u8 speed,
 	 * If the drive is an EIDE drive, it can tell us it needs extended
 	 * PIO/MWDMA cycle timing.
 	 */
-	if (id && id->field_valid & 2) {	/* EIDE drive */
-
+	if (id[ATA_ID_FIELD_VALID] & 2) {	/* EIDE drive */
 		memset(&p, 0, sizeof(p));
 
 		if (speed <= XFER_PIO_2)
-			p.cycle = p.cyc8b = id->eide_pio;
+			p.cycle = p.cyc8b = id[ATA_ID_EIDE_PIO];
 		else if (speed <= XFER_PIO_5)
-			p.cycle = p.cyc8b = id->eide_pio_iordy;
+			p.cycle = p.cyc8b = id[ATA_ID_EIDE_PIO_IORDY];
 		else if (speed >= XFER_MW_DMA_0 && speed <= XFER_MW_DMA_2)
-			p.cycle = id->eide_dma_min;
+			p.cycle = id[ATA_ID_EIDE_DMA_MIN];
 
 		ide_timing_merge(&p, t, t, IDE_TIMING_CYCLE | IDE_TIMING_CYC8B);
 	}
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 772451600e4d..9dcf5aed92cb 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -44,8 +44,6 @@
  *  inspiration from lots of linux users, esp.  hamish@zot.apana.org.au
  */
 
-#define _IDE_C			/* Tell ide.h it's really us */
-
 #include <linux/module.h>
 #include <linux/types.h>
 #include <linux/string.h>
@@ -58,6 +56,7 @@
 #include <linux/init.h>
 #include <linux/pci.h>
 #include <linux/ide.h>
+#include <linux/hdreg.h>
 #include <linux/completion.h>
 #include <linux/device.h>
 
@@ -97,8 +96,6 @@ void ide_init_port_data(ide_hwif_t *hwif, unsigned int index)
 	hwif->name[2]	= 'e';
 	hwif->name[3]	= '0' + index;
 
-	hwif->bus_state	= BUSSTATE_ON;
-
 	init_completion(&hwif->gendev_rel_comp);
 
 	hwif->tp_ops = &default_tp_ops;
@@ -119,7 +116,7 @@ static void ide_port_init_devices_data(ide_hwif_t *hwif)
 		drive->media			= ide_disk;
 		drive->select.all		= (unit<<4)|0xa0;
 		drive->hwif			= hwif;
-		drive->ready_stat		= READY_STAT;
+		drive->ready_stat		= ATA_DRDY;
 		drive->bad_wstat		= BAD_W_STAT;
 		drive->special.b.recalibrate	= 1;
 		drive->special.b.set_geometry	= 1;
@@ -253,42 +250,9 @@ void ide_init_port_hw(ide_hwif_t *hwif, hw_regs_t *hw)
 
 DEFINE_MUTEX(ide_setting_mtx);
 
-EXPORT_SYMBOL_GPL(ide_setting_mtx);
-
-/**
- *	ide_spin_wait_hwgroup	-	wait for group
- *	@drive: drive in the group
- *
- *	Wait for an IDE device group to go non busy and then return
- *	holding the ide_lock which guards the hwgroup->busy status
- *	and right to use it.
- */
+ide_devset_get(io_32bit, io_32bit);
 
-int ide_spin_wait_hwgroup (ide_drive_t *drive)
-{
-	ide_hwgroup_t *hwgroup = HWGROUP(drive);
-	unsigned long timeout = jiffies + (3 * HZ);
-
-	spin_lock_irq(&ide_lock);
-
-	while (hwgroup->busy) {
-		unsigned long lflags;
-		spin_unlock_irq(&ide_lock);
-		local_irq_set(lflags);
-		if (time_after(jiffies, timeout)) {
-			local_irq_restore(lflags);
-			printk(KERN_ERR "%s: channel busy\n", drive->name);
-			return -EBUSY;
-		}
-		local_irq_restore(lflags);
-		spin_lock_irq(&ide_lock);
-	}
-	return 0;
-}
-
-EXPORT_SYMBOL(ide_spin_wait_hwgroup);
-
-int set_io_32bit(ide_drive_t *drive, int arg)
+static int set_io_32bit(ide_drive_t *drive, int arg)
 {
 	if (drive->no_io_32bit)
 		return -EPERM;
@@ -296,53 +260,39 @@ int set_io_32bit(ide_drive_t *drive, int arg)
 	if (arg < 0 || arg > 1 + (SUPPORT_VLB_SYNC << 1))
 		return -EINVAL;
 
-	if (ide_spin_wait_hwgroup(drive))
-		return -EBUSY;
-
 	drive->io_32bit = arg;
 
-	spin_unlock_irq(&ide_lock);
-
 	return 0;
 }
 
+ide_devset_get(ksettings, keep_settings);
+
 static int set_ksettings(ide_drive_t *drive, int arg)
 {
 	if (arg < 0 || arg > 1)
 		return -EINVAL;
 
-	if (ide_spin_wait_hwgroup(drive))
-		return -EBUSY;
 	drive->keep_settings = arg;
-	spin_unlock_irq(&ide_lock);
 
 	return 0;
 }
 
-int set_using_dma(ide_drive_t *drive, int arg)
+ide_devset_get(using_dma, using_dma);
+
+static int set_using_dma(ide_drive_t *drive, int arg)
 {
 #ifdef CONFIG_BLK_DEV_IDEDMA
-	ide_hwif_t *hwif = drive->hwif;
 	int err = -EPERM;
 
 	if (arg < 0 || arg > 1)
 		return -EINVAL;
 
-	if (!drive->id || !(drive->id->capability & 1))
+	if (ata_id_has_dma(drive->id) == 0)
 		goto out;
 
-	if (hwif->dma_ops == NULL)
+	if (drive->hwif->dma_ops == NULL)
 		goto out;
 
-	err = -EBUSY;
-	if (ide_spin_wait_hwgroup(drive))
-		goto out;
-	/*
-	 * set ->busy flag, unlock and let it ride
-	 */
-	hwif->hwgroup->busy = 1;
-	spin_unlock_irq(&ide_lock);
-
 	err = 0;
 
 	if (arg) {
@@ -351,12 +301,6 @@ int set_using_dma(ide_drive_t *drive, int arg)
 	} else
 		ide_dma_off(drive);
 
-	/*
-	 * lock, clear ->busy flag and unlock before leaving
-	 */
-	spin_lock_irq(&ide_lock);
-	hwif->hwgroup->busy = 0;
-	spin_unlock_irq(&ide_lock);
 out:
 	return err;
 #else
@@ -367,7 +311,7 @@ out:
 #endif
 }
 
-int set_pio_mode(ide_drive_t *drive, int arg)
+static int set_pio_mode(ide_drive_t *drive, int arg)
 {
 	struct request *rq;
 	ide_hwif_t *hwif = drive->hwif;
@@ -395,6 +339,8 @@ int set_pio_mode(ide_drive_t *drive, int arg)
 	return 0;
 }
 
+ide_devset_get(unmaskirq, unmask);
+
 static int set_unmaskirq(ide_drive_t *drive, int arg)
 {
 	if (drive->no_unmask)
@@ -403,14 +349,20 @@ static int set_unmaskirq(ide_drive_t *drive, int arg)
 	if (arg < 0 || arg > 1)
 		return -EINVAL;
 
-	if (ide_spin_wait_hwgroup(drive))
-		return -EBUSY;
 	drive->unmask = arg;
-	spin_unlock_irq(&ide_lock);
 
 	return 0;
 }
 
+#define ide_gen_devset_rw(_name, _func) \
+__IDE_DEVSET(_name, DS_SYNC, get_##_func, set_##_func)
+
+ide_gen_devset_rw(io_32bit, io_32bit);
+ide_gen_devset_rw(keepsettings, ksettings);
+ide_gen_devset_rw(unmaskirq, unmaskirq);
+ide_gen_devset_rw(using_dma, using_dma);
+__IDE_DEVSET(pio_mode, 0, NULL, set_pio_mode);
+
 static int generic_ide_suspend(struct device *dev, pm_message_t mesg)
 {
 	ide_drive_t *drive = dev->driver_data;
@@ -486,138 +438,6 @@ static int generic_ide_resume(struct device *dev)
 	return err;
 }
 
-static int generic_drive_reset(ide_drive_t *drive)
-{
-	struct request *rq;
-	int ret = 0;
-
-	rq = blk_get_request(drive->queue, READ, __GFP_WAIT);
-	rq->cmd_type = REQ_TYPE_SPECIAL;
-	rq->cmd_len = 1;
-	rq->cmd[0] = REQ_DRIVE_RESET;
-	rq->cmd_flags |= REQ_SOFTBARRIER;
-	if (blk_execute_rq(drive->queue, NULL, rq, 1))
-		ret = rq->errors;
-	blk_put_request(rq);
-	return ret;
-}
-
-int generic_ide_ioctl(ide_drive_t *drive, struct file *file, struct block_device *bdev,
-			unsigned int cmd, unsigned long arg)
-{
-	unsigned long flags;
-	ide_driver_t *drv;
-	void __user *p = (void __user *)arg;
-	int err = 0, (*setfunc)(ide_drive_t *, int);
-	u8 *val;
-
-	switch (cmd) {
-	case HDIO_GET_32BIT:	    val = &drive->io_32bit;	 goto read_val;
-	case HDIO_GET_KEEPSETTINGS: val = &drive->keep_settings; goto read_val;
-	case HDIO_GET_UNMASKINTR:   val = &drive->unmask;	 goto read_val;
-	case HDIO_GET_DMA:	    val = &drive->using_dma;	 goto read_val;
-	case HDIO_SET_32BIT:	    setfunc = set_io_32bit;	 goto set_val;
-	case HDIO_SET_KEEPSETTINGS: setfunc = set_ksettings;	 goto set_val;
-	case HDIO_SET_PIO_MODE:	    setfunc = set_pio_mode;	 goto set_val;
-	case HDIO_SET_UNMASKINTR:   setfunc = set_unmaskirq;	 goto set_val;
-	case HDIO_SET_DMA:	    setfunc = set_using_dma;	 goto set_val;
-	}
-
-	switch (cmd) {
-		case HDIO_OBSOLETE_IDENTITY:
-		case HDIO_GET_IDENTITY:
-			if (bdev != bdev->bd_contains)
-				return -EINVAL;
-			if (drive->id_read == 0)
-				return -ENOMSG;
-			if (copy_to_user(p, drive->id, (cmd == HDIO_GET_IDENTITY) ? sizeof(*drive->id) : 142))
-				return -EFAULT;
-			return 0;
-
-		case HDIO_GET_NICE:
-			return put_user(drive->dsc_overlap	<<	IDE_NICE_DSC_OVERLAP	|
-					drive->atapi_overlap	<<	IDE_NICE_ATAPI_OVERLAP	|
-					drive->nice1 << IDE_NICE_1,
-					(long __user *) arg);
-#ifdef CONFIG_IDE_TASK_IOCTL
-		case HDIO_DRIVE_TASKFILE:
-		        if (!capable(CAP_SYS_ADMIN) || !capable(CAP_SYS_RAWIO))
-				return -EACCES;
-			switch(drive->media) {
-				case ide_disk:
-					return ide_taskfile_ioctl(drive, cmd, arg);
-				default:
-					return -ENOMSG;
-			}
-#endif /* CONFIG_IDE_TASK_IOCTL */
-
-		case HDIO_DRIVE_CMD:
-			if (!capable(CAP_SYS_RAWIO))
-				return -EACCES;
-			return ide_cmd_ioctl(drive, cmd, arg);
-
-		case HDIO_DRIVE_TASK:
-			if (!capable(CAP_SYS_RAWIO))
-				return -EACCES;
-			return ide_task_ioctl(drive, cmd, arg);
-		case HDIO_SET_NICE:
-			if (!capable(CAP_SYS_ADMIN)) return -EACCES;
-			if (arg != (arg & ((1 << IDE_NICE_DSC_OVERLAP) | (1 << IDE_NICE_1))))
-				return -EPERM;
-			drive->dsc_overlap = (arg >> IDE_NICE_DSC_OVERLAP) & 1;
-			drv = *(ide_driver_t **)bdev->bd_disk->private_data;
-			if (drive->dsc_overlap && !drv->supports_dsc_overlap) {
-				drive->dsc_overlap = 0;
-				return -EPERM;
-			}
-			drive->nice1 = (arg >> IDE_NICE_1) & 1;
-			return 0;
-		case HDIO_DRIVE_RESET:
-			if (!capable(CAP_SYS_ADMIN))
-				return -EACCES;
-
-			return generic_drive_reset(drive);
-
-		case HDIO_GET_BUSSTATE:
-			if (!capable(CAP_SYS_ADMIN))
-				return -EACCES;
-			if (put_user(HWIF(drive)->bus_state, (long __user *)arg))
-				return -EFAULT;
-			return 0;
-
-		case HDIO_SET_BUSSTATE:
-			if (!capable(CAP_SYS_ADMIN))
-				return -EACCES;
-			return -EOPNOTSUPP;
-		default:
-			return -EINVAL;
-	}
-
-read_val:
-	mutex_lock(&ide_setting_mtx);
-	spin_lock_irqsave(&ide_lock, flags);
-	err = *val;
-	spin_unlock_irqrestore(&ide_lock, flags);
-	mutex_unlock(&ide_setting_mtx);
-	return err >= 0 ? put_user(err, (long __user *)arg) : err;
-
-set_val:
-	if (bdev != bdev->bd_contains)
-		err = -EINVAL;
-	else {
-		if (!capable(CAP_SYS_ADMIN))
-			err = -EACCES;
-		else {
-			mutex_lock(&ide_setting_mtx);
-			err = setfunc(drive, arg);
-			mutex_unlock(&ide_setting_mtx);
-		}
-	}
-	return err;
-}
-
-EXPORT_SYMBOL(generic_ide_ioctl);
-
 /**
  * ide_device_get	-	get an additional reference to a ide_drive_t
  * @drive:	device to get a reference to
@@ -710,21 +530,21 @@ static ssize_t model_show(struct device *dev, struct device_attribute *attr,
 			  char *buf)
 {
 	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "%s\n", drive->id->model);
+	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_PROD]);
 }
 
 static ssize_t firmware_show(struct device *dev, struct device_attribute *attr,
 			     char *buf)
 {
 	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "%s\n", drive->id->fw_rev);
+	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_FW_REV]);
 }
 
 static ssize_t serial_show(struct device *dev, struct device_attribute *attr,
 			   char *buf)
 {
 	ide_drive_t *drive = to_ide_device(dev);
-	return sprintf(buf, "%s\n", drive->id->serial_no);
+	return sprintf(buf, "%s\n", (char *)&drive->id[ATA_ID_SERNO]);
 }
 
 static struct device_attribute ide_dev_attrs[] = {
@@ -841,7 +661,7 @@ MODULE_PARM_DESC(noprobe, "skip probing for a device");
 static unsigned int ide_nowerr;
 
 module_param_call(nowerr, ide_set_dev_param_mask, NULL, &ide_nowerr, 0);
-MODULE_PARM_DESC(nowerr, "ignore the WRERR_STAT bit for a device");
+MODULE_PARM_DESC(nowerr, "ignore the ATA_DF bit for a device");
 
 static unsigned int ide_cdroms;
 
@@ -906,7 +726,7 @@ static void ide_dev_apply_params(ide_drive_t *drive)
 		drive->noprobe = 1;
 	}
 	if (ide_nowerr & (1 << i)) {
-		printk(KERN_INFO "ide: ignoring the WRERR_STAT bit for %s\n",
+		printk(KERN_INFO "ide: ignoring the ATA_DF bit for %s\n",
 				 drive->name);
 		drive->bad_wstat = BAD_R_STAT;
 	}
@@ -927,7 +747,7 @@ static void ide_dev_apply_params(ide_drive_t *drive)
 				 drive->cyl, drive->head, drive->sect);
 		drive->present = 1;
 		drive->media = ide_disk;
-		drive->ready_stat = READY_STAT;
+		drive->ready_stat = ATA_DRDY;
 	}
 }
 
diff --git a/drivers/ide/legacy/ali14xx.c b/drivers/ide/legacy/ali14xx.c
index 4ec19737f3c5..7276c96aaa2a 100644
--- a/drivers/ide/legacy/ali14xx.c
+++ b/drivers/ide/legacy/ali14xx.c
@@ -43,7 +43,6 @@
 #include <linux/mm.h>
 #include <linux/ioport.h>
 #include <linux/blkdev.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
diff --git a/drivers/ide/legacy/buddha.c b/drivers/ide/legacy/buddha.c
index 7c2afa97f417..c5a3c9ef6a5d 100644
--- a/drivers/ide/legacy/buddha.c
+++ b/drivers/ide/legacy/buddha.c
@@ -20,7 +20,6 @@
 #include <linux/mm.h>
 #include <linux/interrupt.h>
 #include <linux/blkdev.h>
-#include <linux/hdreg.h>
 #include <linux/zorro.h>
 #include <linux/ide.h>
 #include <linux/init.h>
diff --git a/drivers/ide/legacy/dtc2278.c b/drivers/ide/legacy/dtc2278.c
index af791a02a120..689b2e493413 100644
--- a/drivers/ide/legacy/dtc2278.c
+++ b/drivers/ide/legacy/dtc2278.c
@@ -10,7 +10,6 @@
 #include <linux/mm.h>
 #include <linux/ioport.h>
 #include <linux/blkdev.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
diff --git a/drivers/ide/legacy/falconide.c b/drivers/ide/legacy/falconide.c
index 724f95073d80..39d500d84b07 100644
--- a/drivers/ide/legacy/falconide.c
+++ b/drivers/ide/legacy/falconide.c
@@ -13,7 +13,6 @@
 #include <linux/mm.h>
 #include <linux/interrupt.h>
 #include <linux/blkdev.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
diff --git a/drivers/ide/legacy/gayle.c b/drivers/ide/legacy/gayle.c
index 51ba085d7aa8..691506886561 100644
--- a/drivers/ide/legacy/gayle.c
+++ b/drivers/ide/legacy/gayle.c
@@ -12,7 +12,6 @@
 #include <linux/mm.h>
 #include <linux/interrupt.h>
 #include <linux/blkdev.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 #include <linux/zorro.h>
diff --git a/drivers/ide/legacy/ht6560b.c b/drivers/ide/legacy/ht6560b.c
index 98f7c95e39ed..5123ea291d07 100644
--- a/drivers/ide/legacy/ht6560b.c
+++ b/drivers/ide/legacy/ht6560b.c
@@ -24,7 +24,6 @@
 #include <linux/mm.h>
 #include <linux/ioport.h>
 #include <linux/blkdev.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
diff --git a/drivers/ide/legacy/ide-cs.c b/drivers/ide/legacy/ide-cs.c
index 21bfac137844..ee6fc30d5e2b 100644
--- a/drivers/ide/legacy/ide-cs.c
+++ b/drivers/ide/legacy/ide-cs.c
@@ -38,7 +38,6 @@
 #include <linux/timer.h>
 #include <linux/ioport.h>
 #include <linux/ide.h>
-#include <linux/hdreg.h>
 #include <linux/major.h>
 #include <linux/delay.h>
 #include <asm/io.h>
diff --git a/drivers/ide/legacy/macide.c b/drivers/ide/legacy/macide.c
index a0bb167980e7..43f97cc1d30e 100644
--- a/drivers/ide/legacy/macide.c
+++ b/drivers/ide/legacy/macide.c
@@ -15,7 +15,6 @@
 #include <linux/mm.h>
 #include <linux/interrupt.h>
 #include <linux/blkdev.h>
-#include <linux/hdreg.h>
 #include <linux/delay.h>
 #include <linux/ide.h>
 
diff --git a/drivers/ide/legacy/q40ide.c b/drivers/ide/legacy/q40ide.c
index 4abd8fc78197..4af4a8ce4cdf 100644
--- a/drivers/ide/legacy/q40ide.c
+++ b/drivers/ide/legacy/q40ide.c
@@ -14,8 +14,6 @@
 #include <linux/mm.h>
 #include <linux/interrupt.h>
 #include <linux/blkdev.h>
-#include <linux/hdreg.h>
-
 #include <linux/ide.h>
 
     /*
diff --git a/drivers/ide/legacy/qd65xx.c b/drivers/ide/legacy/qd65xx.c
index 2338f344ea24..ec408b3a7100 100644
--- a/drivers/ide/legacy/qd65xx.c
+++ b/drivers/ide/legacy/qd65xx.c
@@ -27,7 +27,6 @@
 #include <linux/mm.h>
 #include <linux/ioport.h>
 #include <linux/blkdev.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 #include <asm/system.h>
@@ -151,12 +150,14 @@ static int qd_find_disk_type (ide_drive_t *drive,
 		int *active_time, int *recovery_time)
 {
 	struct qd65xx_timing_s *p;
-	char model[40];
+	char *m = (char *)&drive->id[ATA_ID_PROD];
+	char model[ATA_ID_PROD_LEN];
 
-	if (!*drive->id->model) return 0;
+	if (*m == 0)
+		return 0;
 
-	strncpy(model,drive->id->model,40);
-	ide_fixstring(model,40,1); /* byte-swap */
+	strncpy(model, m, ATA_ID_PROD_LEN);
+	ide_fixstring(model, ATA_ID_PROD_LEN, 1); /* byte-swap */
 
 	for (p = qd65xx_timing ; p->offset != -1 ; p++) {
 		if (!strncmp(p->model, model+p->offset, 4)) {
@@ -185,20 +186,20 @@ static void qd_set_timing (ide_drive_t *drive, u8 timing)
 
 static void qd6500_set_pio_mode(ide_drive_t *drive, const u8 pio)
 {
+	u16 *id = drive->id;
 	int active_time   = 175;
 	int recovery_time = 415; /* worst case values from the dos driver */
 
 	/*
 	 * FIXME: use "pio" value
 	 */
-	if (drive->id && !qd_find_disk_type(drive, &active_time, &recovery_time)
-		&& drive->id->tPIO && (drive->id->field_valid & 0x02)
-		&& drive->id->eide_pio >= 240) {
-
+	if (!qd_find_disk_type(drive, &active_time, &recovery_time) &&
+	    (id[ATA_ID_OLD_PIO_MODES] & 0xff) && (id[ATA_ID_FIELD_VALID] & 2) &&
+	    id[ATA_ID_EIDE_PIO] >= 240) {
 		printk(KERN_INFO "%s: PIO mode%d\n", drive->name,
-				drive->id->tPIO);
+			id[ATA_ID_OLD_PIO_MODES] & 0xff);
 		active_time = 110;
-		recovery_time = drive->id->eide_pio - 120;
+		recovery_time = drive->id[ATA_ID_EIDE_PIO] - 120;
 	}
 
 	qd_set_timing(drive, qd6500_compute_timing(HWIF(drive), active_time, recovery_time));
diff --git a/drivers/ide/legacy/umc8672.c b/drivers/ide/legacy/umc8672.c
index b54a14a57755..1da076e0c917 100644
--- a/drivers/ide/legacy/umc8672.c
+++ b/drivers/ide/legacy/umc8672.c
@@ -45,7 +45,6 @@
 #include <linux/mm.h>
 #include <linux/ioport.h>
 #include <linux/blkdev.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
diff --git a/drivers/ide/pci/aec62xx.c b/drivers/ide/pci/aec62xx.c
index 3187215e8f89..e7475ba559c7 100644
--- a/drivers/ide/pci/aec62xx.c
+++ b/drivers/ide/pci/aec62xx.c
@@ -7,7 +7,6 @@
 #include <linux/module.h>
 #include <linux/types.h>
 #include <linux/pci.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
@@ -140,7 +139,7 @@ static void aec_set_pio_mode(ide_drive_t *drive, const u8 pio)
 	drive->hwif->port_ops->set_dma_mode(drive, pio + XFER_PIO_0);
 }
 
-static unsigned int __devinit init_chipset_aec62xx(struct pci_dev *dev)
+static unsigned int init_chipset_aec62xx(struct pci_dev *dev)
 {
 	/* These are necessary to get AEC6280 Macintosh cards to work */
 	if ((dev->device == PCI_DEVICE_ID_ARTOP_ATP865) ||
@@ -308,6 +307,8 @@ static struct pci_driver driver = {
 	.id_table	= aec62xx_pci_tbl,
 	.probe		= aec62xx_init_one,
 	.remove		= __devexit_p(aec62xx_remove),
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init aec62xx_ide_init(void)
diff --git a/drivers/ide/pci/alim15x3.c b/drivers/ide/pci/alim15x3.c
index d647526af557..053c75263918 100644
--- a/drivers/ide/pci/alim15x3.c
+++ b/drivers/ide/pci/alim15x3.c
@@ -31,7 +31,6 @@
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/pci.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 #include <linux/dmi.h>
@@ -134,8 +133,8 @@ static u8 ali_udma_filter(ide_drive_t *drive)
 	if (m5229_revision > 0x20 && m5229_revision < 0xC2) {
 		if (drive->media != ide_disk)
 			return 0;
-		if (chip_is_1543c_e && strstr(drive->id->model, "WDC ") &&
-		    wdc_udma == 0)
+		if (wdc_udma == 0 && chip_is_1543c_e &&
+		    strstr((char *)&drive->id[ATA_ID_PROD], "WDC "))
 			return 0;
 	}
 
@@ -214,7 +213,7 @@ static int ali15x3_dma_setup(ide_drive_t *drive)
  *	appropriate also sets up the 1533 southbridge.
  */
 
-static unsigned int __devinit init_chipset_ali15x3(struct pci_dev *dev)
+static unsigned int init_chipset_ali15x3(struct pci_dev *dev)
 {
 	unsigned long flags;
 	u8 tmpbyte;
@@ -582,6 +581,8 @@ static struct pci_driver driver = {
 	.id_table	= alim15x3_pci_tbl,
 	.probe		= alim15x3_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init ali15x3_ide_init(void)
diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c
index 1e66a960a96a..824471f91bf5 100644
--- a/drivers/ide/pci/amd74xx.c
+++ b/drivers/ide/pci/amd74xx.c
@@ -112,13 +112,13 @@ static void amd_set_pio_mode(ide_drive_t *drive, const u8 pio)
 	amd_set_drive(drive, XFER_PIO_0 + pio);
 }
 
-static void __devinit amd7409_cable_detect(struct pci_dev *dev)
+static void amd7409_cable_detect(struct pci_dev *dev)
 {
 	/* no host side cable detection */
 	amd_80w = 0x03;
 }
 
-static void __devinit amd7411_cable_detect(struct pci_dev *dev)
+static void amd7411_cable_detect(struct pci_dev *dev)
 {
 	int i;
 	u32 u = 0;
@@ -140,7 +140,7 @@ static void __devinit amd7411_cable_detect(struct pci_dev *dev)
  * The initialization callback.  Initialize drive independent registers.
  */
 
-static unsigned int __devinit init_chipset_amd74xx(struct pci_dev *dev)
+static unsigned int init_chipset_amd74xx(struct pci_dev *dev)
 {
 	u8 t = 0, offset = amd_offset(dev);
 
@@ -324,6 +324,8 @@ static struct pci_driver driver = {
 	.id_table	= amd74xx_pci_tbl,
 	.probe		= amd74xx_probe,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init amd74xx_ide_init(void)
diff --git a/drivers/ide/pci/atiixp.c b/drivers/ide/pci/atiixp.c
index 41f6cb6c163a..e4437034dd08 100644
--- a/drivers/ide/pci/atiixp.c
+++ b/drivers/ide/pci/atiixp.c
@@ -7,7 +7,6 @@
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/pci.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
@@ -188,6 +187,8 @@ static struct pci_driver driver = {
 	.id_table	= atiixp_pci_tbl,
 	.probe		= atiixp_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init atiixp_ide_init(void)
diff --git a/drivers/ide/pci/cmd640.c b/drivers/ide/pci/cmd640.c
index e6c62006ca1a..7f39cdb41410 100644
--- a/drivers/ide/pci/cmd640.c
+++ b/drivers/ide/pci/cmd640.c
@@ -103,7 +103,6 @@
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/delay.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
@@ -375,6 +374,21 @@ static void cmd640_dump_regs(void)
 }
 #endif
 
+static void __set_prefetch_mode(ide_drive_t *drive, int mode)
+{
+	if (mode) {	/* want prefetch on? */
+#if CMD640_PREFETCH_MASKS
+		drive->no_unmask = 1;
+		drive->unmask = 0;
+#endif
+		drive->no_io_32bit = 0;
+	} else {
+		drive->no_unmask = 0;
+		drive->no_io_32bit = 1;
+		drive->io_32bit = 0;
+	}
+}
+
 #ifndef CONFIG_BLK_DEV_CMD640_ENHANCED
 /*
  * Check whether prefetch is on for a drive,
@@ -384,19 +398,10 @@ static void __init check_prefetch(ide_drive_t *drive, unsigned int index)
 {
 	u8 b = get_cmd640_reg(prefetch_regs[index]);
 
-	if (b & prefetch_masks[index]) {	/* is prefetch off? */
-		drive->no_unmask = 0;
-		drive->no_io_32bit = 1;
-		drive->io_32bit = 0;
-	} else {
-#if CMD640_PREFETCH_MASKS
-		drive->no_unmask = 1;
-		drive->unmask = 0;
-#endif
-		drive->no_io_32bit = 0;
-	}
+	__set_prefetch_mode(drive, (b & prefetch_masks[index]) ? 0 : 1);
 }
 #else
+
 /*
  * Sets prefetch mode for a drive.
  */
@@ -408,19 +413,11 @@ static void set_prefetch_mode(ide_drive_t *drive, unsigned int index, int mode)
 
 	spin_lock_irqsave(&cmd640_lock, flags);
 	b = __get_cmd640_reg(reg);
-	if (mode) {	/* want prefetch on? */
-#if CMD640_PREFETCH_MASKS
-		drive->no_unmask = 1;
-		drive->unmask = 0;
-#endif
-		drive->no_io_32bit = 0;
+	__set_prefetch_mode(drive, mode);
+	if (mode)
 		b &= ~prefetch_masks[index];	/* enable prefetch */
-	} else {
-		drive->no_unmask = 0;
-		drive->no_io_32bit = 1;
-		drive->io_32bit = 0;
+	else
 		b |= prefetch_masks[index];	/* disable prefetch */
-	}
 	__put_cmd640_reg(reg, b);
 	spin_unlock_irqrestore(&cmd640_lock, flags);
 }
diff --git a/drivers/ide/pci/cmd64x.c b/drivers/ide/pci/cmd64x.c
index e064398e03b4..456dee18b660 100644
--- a/drivers/ide/pci/cmd64x.c
+++ b/drivers/ide/pci/cmd64x.c
@@ -13,7 +13,6 @@
 #include <linux/module.h>
 #include <linux/types.h>
 #include <linux/pci.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
@@ -332,7 +331,7 @@ static int cmd646_1_dma_end(ide_drive_t *drive)
 	return (dma_stat & 7) != 4;
 }
 
-static unsigned int __devinit init_chipset_cmd64x(struct pci_dev *dev)
+static unsigned int init_chipset_cmd64x(struct pci_dev *dev)
 {
 	u8 mrdmode = 0;
 
@@ -511,6 +510,8 @@ static struct pci_driver driver = {
 	.id_table	= cmd64x_pci_tbl,
 	.probe		= cmd64x_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init cmd64x_ide_init(void)
diff --git a/drivers/ide/pci/cs5520.c b/drivers/ide/pci/cs5520.c
index 151844fcbb07..d6341f7c4144 100644
--- a/drivers/ide/pci/cs5520.c
+++ b/drivers/ide/pci/cs5520.c
@@ -35,7 +35,6 @@
 #include <linux/module.h>
 #include <linux/types.h>
 #include <linux/kernel.h>
-#include <linux/hdreg.h>
 #include <linux/init.h>
 #include <linux/pci.h>
 #include <linux/ide.h>
@@ -150,6 +149,8 @@ static struct pci_driver driver = {
 	.name		= "Cyrix_IDE",
 	.id_table	= cs5520_pci_tbl,
 	.probe		= cs5520_init_one,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init cs5520_ide_init(void)
diff --git a/drivers/ide/pci/cs5530.c b/drivers/ide/pci/cs5530.c
index f235db8c678b..da42fa7e9f97 100644
--- a/drivers/ide/pci/cs5530.c
+++ b/drivers/ide/pci/cs5530.c
@@ -15,7 +15,6 @@
 #include <linux/module.h>
 #include <linux/types.h>
 #include <linux/kernel.h>
-#include <linux/hdreg.h>
 #include <linux/pci.h>
 #include <linux/init.h>
 #include <linux/ide.h>
@@ -81,17 +80,19 @@ static void cs5530_set_pio_mode(ide_drive_t *drive, const u8 pio)
 static u8 cs5530_udma_filter(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	ide_drive_t *mate = &hwif->drives[(drive->dn & 1) ^ 1];
-	struct hd_driveid *mateid = mate->id;
+	ide_drive_t *mate = ide_get_pair_dev(drive);
+	u16 *mateid = mate->id;
 	u8 mask = hwif->ultra_mask;
 
-	if (mate->present == 0)
+	if (mate == NULL)
 		goto out;
 
-	if ((mateid->capability & 1) && __ide_dma_bad_drive(mate) == 0) {
-		if ((mateid->field_valid & 4) && (mateid->dma_ultra & 7))
+	if (ata_id_has_dma(mateid) && __ide_dma_bad_drive(mate) == 0) {
+		if ((mateid[ATA_ID_FIELD_VALID] & 4) &&
+		    (mateid[ATA_ID_UDMA_MODES] & 7))
 			goto out;
-		if ((mateid->field_valid & 2) && (mateid->dma_mword & 7))
+		if ((mateid[ATA_ID_FIELD_VALID] & 2) &&
+		    (mateid[ATA_ID_MWDMA_MODES] & 7))
 			mask = 0;
 	}
 out:
@@ -133,7 +134,7 @@ static void cs5530_set_dma_mode(ide_drive_t *drive, const u8 mode)
  *	Initialize the cs5530 bridge for reliable IDE DMA operation.
  */
 
-static unsigned int __devinit init_chipset_cs5530(struct pci_dev *dev)
+static unsigned int init_chipset_cs5530(struct pci_dev *dev)
 {
 	struct pci_dev *master_0 = NULL, *cs5530_0 = NULL;
 
@@ -271,6 +272,8 @@ static struct pci_driver driver = {
 	.id_table	= cs5530_pci_tbl,
 	.probe		= cs5530_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init cs5530_ide_init(void)
diff --git a/drivers/ide/pci/cs5535.c b/drivers/ide/pci/cs5535.c
index dd3dc23af995..1e5bc59ea2fb 100644
--- a/drivers/ide/pci/cs5535.c
+++ b/drivers/ide/pci/cs5535.c
@@ -80,12 +80,12 @@ static void cs5535_set_speed(ide_drive_t *drive, const u8 speed)
 
 	/* Set the PIO timings */
 	if (speed < XFER_SW_DMA_0) {
-		ide_drive_t *pair = ide_get_paired_drive(drive);
+		ide_drive_t *pair = ide_get_pair_dev(drive);
 		u8 cmd, pioa;
 
 		cmd = pioa = speed - XFER_PIO_0;
 
-		if (pair->present) {
+		if (pair) {
 			u8 piob = ide_get_best_pio_mode(pair, 255, 4);
 
 			if (piob < cmd)
@@ -193,10 +193,12 @@ static const struct pci_device_id cs5535_pci_tbl[] = {
 MODULE_DEVICE_TABLE(pci, cs5535_pci_tbl);
 
 static struct pci_driver driver = {
-	.name       = "CS5535_IDE",
-	.id_table   = cs5535_pci_tbl,
-	.probe      = cs5535_init_one,
-	.remove     = ide_pci_remove,
+	.name		= "CS5535_IDE",
+	.id_table	= cs5535_pci_tbl,
+	.probe		= cs5535_init_one,
+	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init cs5535_ide_init(void)
diff --git a/drivers/ide/pci/cy82c693.c b/drivers/ide/pci/cy82c693.c
index e6d8ee88d56d..69820e9224d1 100644
--- a/drivers/ide/pci/cy82c693.c
+++ b/drivers/ide/pci/cy82c693.c
@@ -332,7 +332,7 @@ static void cy82c693_set_pio_mode(ide_drive_t *drive, const u8 pio)
 /*
  * this function is called during init and is used to setup the cy82c693 chip
  */
-static unsigned int __devinit init_chipset_cy82c693(struct pci_dev *dev)
+static unsigned int init_chipset_cy82c693(struct pci_dev *dev)
 {
 	if (PCI_FUNC(dev->devfn) != 1)
 		return 0;
@@ -448,6 +448,8 @@ static struct pci_driver driver = {
 	.id_table	= cy82c693_pci_tbl,
 	.probe		= cy82c693_init_one,
 	.remove		= __devexit_p(cy82c693_remove),
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init cy82c693_ide_init(void)
diff --git a/drivers/ide/pci/delkin_cb.c b/drivers/ide/pci/delkin_cb.c
index f84bfb4f600f..83b63b365e51 100644
--- a/drivers/ide/pci/delkin_cb.c
+++ b/drivers/ide/pci/delkin_cb.c
@@ -19,7 +19,6 @@
 
 #include <linux/types.h>
 #include <linux/module.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 #include <linux/pci.h>
diff --git a/drivers/ide/pci/generic.c b/drivers/ide/pci/generic.c
index b07d4f4273b3..092b238cb250 100644
--- a/drivers/ide/pci/generic.c
+++ b/drivers/ide/pci/generic.c
@@ -22,7 +22,6 @@
 #include <linux/types.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
-#include <linux/hdreg.h>
 #include <linux/pci.h>
 #include <linux/ide.h>
 #include <linux/init.h>
@@ -172,6 +171,8 @@ static struct pci_driver driver = {
 	.id_table	= generic_pci_tbl,
 	.probe		= generic_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init generic_ide_init(void)
diff --git a/drivers/ide/pci/hpt34x.c b/drivers/ide/pci/hpt34x.c
index 6009b0b9655d..644de29f8fe4 100644
--- a/drivers/ide/pci/hpt34x.c
+++ b/drivers/ide/pci/hpt34x.c
@@ -27,7 +27,6 @@
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/ioport.h>
-#include <linux/hdreg.h>
 #include <linux/interrupt.h>
 #include <linux/pci.h>
 #include <linux/init.h>
@@ -79,7 +78,7 @@ static void hpt34x_set_pio_mode(ide_drive_t *drive, const u8 pio)
  */
 #define	HPT34X_PCI_INIT_REG		0x80
 
-static unsigned int __devinit init_chipset_hpt34x(struct pci_dev *dev)
+static unsigned int init_chipset_hpt34x(struct pci_dev *dev)
 {
 	int i = 0;
 	unsigned long hpt34xIoBase = pci_resource_start(dev, 4);
@@ -172,6 +171,8 @@ static struct pci_driver driver = {
 	.id_table	= hpt34x_pci_tbl,
 	.probe		= hpt34x_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init hpt34x_ide_init(void)
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index c37ab1743819..a194022b6a61 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -122,7 +122,6 @@
 #include <linux/kernel.h>
 #include <linux/delay.h>
 #include <linux/blkdev.h>
-#include <linux/hdreg.h>
 #include <linux/interrupt.h>
 #include <linux/pci.h>
 #include <linux/init.h>
@@ -605,10 +604,10 @@ static const struct hpt_info hpt371n __devinitdata = {
 
 static int check_in_drive_list(ide_drive_t *drive, const char **list)
 {
-	struct hd_driveid *id = drive->id;
+	char *m = (char *)&drive->id[ATA_ID_PROD];
 
 	while (*list)
-		if (!strcmp(*list++,id->model))
+		if (!strcmp(*list++, m))
 			return 1;
 	return 0;
 }
@@ -655,7 +654,7 @@ static u8 hpt3xx_udma_filter(ide_drive_t *drive)
 	case HPT372A:
 	case HPT372N:
 	case HPT374 :
-		if (ide_dev_is_sata(drive->id))
+		if (ata_id_is_sata(drive->id))
 			mask &= ~0x0e;
 		/* Fall thru */
 	default:
@@ -675,7 +674,7 @@ static u8 hpt3xx_mdma_filter(ide_drive_t *drive)
 	case HPT372A:
 	case HPT372N:
 	case HPT374 :
-		if (ide_dev_is_sata(drive->id))
+		if (ata_id_is_sata(drive->id))
 			return 0x00;
 		/* Fall thru */
 	default:
@@ -731,11 +730,11 @@ static void hpt3xx_set_pio_mode(ide_drive_t *drive, const u8 pio)
 
 static void hpt3xx_quirkproc(ide_drive_t *drive)
 {
-	struct hd_driveid *id	= drive->id;
+	char *m			= (char *)&drive->id[ATA_ID_PROD];
 	const  char **list	= quirk_drives;
 
 	while (*list)
-		if (strstr(id->model, *list++)) {
+		if (strstr(m, *list++)) {
 			drive->quirk_list = 1;
 			return;
 		}
@@ -944,7 +943,7 @@ static void hpt3xxn_rw_disk(ide_drive_t *drive, struct request *rq)
  *	Perform a calibration cycle on the DPLL.
  *	Returns 1 if this succeeds
  */
-static int __devinit hpt37x_calibrate_dpll(struct pci_dev *dev, u16 f_low, u16 f_high)
+static int hpt37x_calibrate_dpll(struct pci_dev *dev, u16 f_low, u16 f_high)
 {
 	u32 dpll = (f_high << 16) | f_low | 0x100;
 	u8  scr2;
@@ -972,7 +971,37 @@ static int __devinit hpt37x_calibrate_dpll(struct pci_dev *dev, u16 f_low, u16 f
 	return 1;
 }
 
-static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev)
+static void hpt3xx_disable_fast_irq(struct pci_dev *dev, u8 mcr_addr)
+{
+	struct ide_host *host	= pci_get_drvdata(dev);
+	struct hpt_info *info	= host->host_priv + (&dev->dev == host->dev[1]);
+	u8  chip_type		= info->chip_type;
+	u8  new_mcr, old_mcr	= 0;
+
+	/*
+	 * Disable the "fast interrupt" prediction.  Don't hold off
+	 * on interrupts. (== 0x01 despite what the docs say)
+	 */
+	pci_read_config_byte(dev, mcr_addr + 1, &old_mcr);
+
+	if (chip_type >= HPT374)
+		new_mcr = old_mcr & ~0x07;
+	else if (chip_type >= HPT370) {
+		new_mcr = old_mcr;
+		new_mcr &= ~0x02;
+#ifdef HPT_DELAY_INTERRUPT
+		new_mcr &= ~0x01;
+#else
+		new_mcr |=  0x01;
+#endif
+	} else					/* HPT366 and HPT368  */
+		new_mcr = old_mcr & ~0x80;
+
+	if (new_mcr != old_mcr)
+		pci_write_config_byte(dev, mcr_addr + 1, new_mcr);
+}
+
+static unsigned int init_chipset_hpt366(struct pci_dev *dev)
 {
 	unsigned long io_base	= pci_resource_start(dev, 4);
 	struct hpt_info *info	= hpt3xx_get_info(&dev->dev);
@@ -1209,9 +1238,11 @@ static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev)
 	 * NOTE: This register is only writeable via I/O space.
 	 */
 	if (chip_type == HPT371N && clock == ATA_CLOCK_66MHZ)
-
 		outb(inb(io_base + 0x9c) | 0x04, io_base + 0x9c);
 
+	hpt3xx_disable_fast_irq(dev, 0x50);
+	hpt3xx_disable_fast_irq(dev, 0x54);
+
 	return dev->irq;
 }
 
@@ -1265,7 +1296,6 @@ static void __devinit init_hwif_hpt366(ide_hwif_t *hwif)
 	struct hpt_info *info	= hpt3xx_get_info(hwif->dev);
 	int serialize		= HPT_SERIALIZE_IO;
 	u8  chip_type		= info->chip_type;
-	u8  new_mcr, old_mcr	= 0;
 
 	/* Cache the channel's MISC. control registers' offset */
 	hwif->select_data	= hwif->channel ? 0x54 : 0x50;
@@ -1288,29 +1318,6 @@ static void __devinit init_hwif_hpt366(ide_hwif_t *hwif)
 	/* Serialize access to this device if needed */
 	if (serialize && hwif->mate)
 		hwif->serialized = hwif->mate->serialized = 1;
-
-	/*
-	 * Disable the "fast interrupt" prediction.  Don't hold off
-	 * on interrupts. (== 0x01 despite what the docs say)
-	 */
-	pci_read_config_byte(dev, hwif->select_data + 1, &old_mcr);
-
-	if (info->chip_type >= HPT374)
-		new_mcr = old_mcr & ~0x07;
-	else if (info->chip_type >= HPT370) {
-		new_mcr = old_mcr;
-		new_mcr &= ~0x02;
-
-#ifdef HPT_DELAY_INTERRUPT
-		new_mcr &= ~0x01;
-#else
-		new_mcr |=  0x01;
-#endif
-	} else					/* HPT366 and HPT368  */
-		new_mcr = old_mcr & ~0x80;
-
-	if (new_mcr != old_mcr)
-		pci_write_config_byte(dev, hwif->select_data + 1, new_mcr);
 }
 
 static int __devinit init_dma_hpt366(ide_hwif_t *hwif,
@@ -1620,6 +1627,8 @@ static struct pci_driver driver = {
 	.id_table	= hpt366_pci_tbl,
 	.probe		= hpt366_init_one,
 	.remove		= __devexit_p(hpt366_remove),
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init hpt366_ide_init(void)
diff --git a/drivers/ide/pci/it8213.c b/drivers/ide/pci/it8213.c
index 652e47dd7e89..0954ccd08d6f 100644
--- a/drivers/ide/pci/it8213.c
+++ b/drivers/ide/pci/it8213.c
@@ -10,7 +10,6 @@
 #include <linux/types.h>
 #include <linux/module.h>
 #include <linux/pci.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
@@ -195,6 +194,8 @@ static struct pci_driver driver = {
 	.id_table	= it8213_pci_tbl,
 	.probe		= it8213_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init it8213_ide_init(void)
diff --git a/drivers/ide/pci/it821x.c b/drivers/ide/pci/it821x.c
index 4a1508a707cc..46edd083b348 100644
--- a/drivers/ide/pci/it821x.c
+++ b/drivers/ide/pci/it821x.c
@@ -63,7 +63,6 @@
 #include <linux/types.h>
 #include <linux/module.h>
 #include <linux/pci.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
@@ -446,8 +445,7 @@ static u8 it821x_cable_detect(ide_hwif_t *hwif)
 static void it821x_quirkproc(ide_drive_t *drive)
 {
 	struct it821x_dev *itdev = ide_get_hwifdata(drive->hwif);
-	struct hd_driveid *id = drive->id;
-	u16 *idbits = (u16 *)drive->id;
+	u16 *id = drive->id;
 
 	if (!itdev->smart) {
 		/*
@@ -466,36 +464,36 @@ static void it821x_quirkproc(ide_drive_t *drive)
 	 */
 
 		/* Check for RAID v native */
-		if(strstr(id->model, "Integrated Technology Express")) {
+		if (strstr((char *)&id[ATA_ID_PROD],
+			   "Integrated Technology Express")) {
 			/* In raid mode the ident block is slightly buggy
 			   We need to set the bits so that the IDE layer knows
 			   LBA28. LBA48 and DMA ar valid */
-			id->capability |= 3;		/* LBA28, DMA */
-			id->command_set_2 |= 0x0400;	/* LBA48 valid */
-			id->cfs_enable_2 |= 0x0400;	/* LBA48 on */
+			id[ATA_ID_CAPABILITY]    |= (3 << 8); /* LBA28, DMA */
+			id[ATA_ID_COMMAND_SET_2] |= 0x0400;   /* LBA48 valid */
+			id[ATA_ID_CFS_ENABLE_2]  |= 0x0400;   /* LBA48 on */
 			/* Reporting logic */
 			printk(KERN_INFO "%s: IT8212 %sRAID %d volume",
-				drive->name,
-				idbits[147] ? "Bootable ":"",
-				idbits[129]);
-				if(idbits[129] != 1)
-					printk("(%dK stripe)", idbits[146]);
-				printk(".\n");
+				drive->name, id[147] ? "Bootable " : "",
+				id[ATA_ID_CSFO]);
+			if (id[ATA_ID_CSFO] != 1)
+				printk(KERN_CONT "(%dK stripe)", id[146]);
+			printk(KERN_CONT ".\n");
 		} else {
 			/* Non RAID volume. Fixups to stop the core code
 			   doing unsupported things */
-			id->field_valid &= 3;
-			id->queue_depth = 0;
-			id->command_set_1 = 0;
-			id->command_set_2 &= 0xC400;
-			id->cfsse &= 0xC000;
-			id->cfs_enable_1 = 0;
-			id->cfs_enable_2 &= 0xC400;
-			id->csf_default &= 0xC000;
-			id->word127 = 0;
-			id->dlf = 0;
-			id->csfo = 0;
-			id->cfa_power = 0;
+			id[ATA_ID_FIELD_VALID]	 &= 3;
+			id[ATA_ID_QUEUE_DEPTH]	  = 0;
+			id[ATA_ID_COMMAND_SET_1]  = 0;
+			id[ATA_ID_COMMAND_SET_2] &= 0xC400;
+			id[ATA_ID_CFSSE]	 &= 0xC000;
+			id[ATA_ID_CFS_ENABLE_1]	  = 0;
+			id[ATA_ID_CFS_ENABLE_2]	 &= 0xC400;
+			id[ATA_ID_CSF_DEFAULT]	 &= 0xC000;
+			id[127]			  = 0;
+			id[ATA_ID_DLF]		  = 0;
+			id[ATA_ID_CSFO]		  = 0;
+			id[ATA_ID_CFA_POWER]	  = 0;
 			printk(KERN_INFO "%s: Performing identify fixups.\n",
 				drive->name);
 		}
@@ -505,8 +503,8 @@ static void it821x_quirkproc(ide_drive_t *drive)
 		 * IDE core that DMA is supported (it821x hardware
 		 * takes care of DMA mode programming).
 		 */
-		if (id->capability & 1) {
-			id->dma_mword |= 0x0101;
+		if (ata_id_has_dma(id)) {
+			id[ATA_ID_MWDMA_MODES] |= 0x0101;
 			drive->current_speed = XFER_MW_DMA_0;
 		}
 	}
@@ -588,7 +586,7 @@ static void __devinit init_hwif_it821x(ide_hwif_t *hwif)
 	hwif->mwdma_mask = ATA_MWDMA2;
 }
 
-static void __devinit it8212_disable_raid(struct pci_dev *dev)
+static void it8212_disable_raid(struct pci_dev *dev)
 {
 	/* Reset local CPU, and set BIOS not ready */
 	pci_write_config_byte(dev, 0x5E, 0x01);
@@ -605,7 +603,7 @@ static void __devinit it8212_disable_raid(struct pci_dev *dev)
 	pci_write_config_byte(dev, PCI_LATENCY_TIMER, 0x20);
 }
 
-static unsigned int __devinit init_chipset_it821x(struct pci_dev *dev)
+static unsigned int init_chipset_it821x(struct pci_dev *dev)
 {
 	u8 conf;
 	static char *mode[2] = { "pass through", "smart" };
@@ -687,6 +685,8 @@ static struct pci_driver driver = {
 	.id_table	= it821x_pci_tbl,
 	.probe		= it821x_init_one,
 	.remove		= __devexit_p(it821x_remove),
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init it821x_ide_init(void)
diff --git a/drivers/ide/pci/jmicron.c b/drivers/ide/pci/jmicron.c
index bb9d09d8f196..acd647110648 100644
--- a/drivers/ide/pci/jmicron.c
+++ b/drivers/ide/pci/jmicron.c
@@ -8,7 +8,6 @@
 #include <linux/types.h>
 #include <linux/module.h>
 #include <linux/pci.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
@@ -155,6 +154,8 @@ static struct pci_driver driver = {
 	.id_table	= jmicron_pci_tbl,
 	.probe		= jmicron_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init jmicron_ide_init(void)
diff --git a/drivers/ide/pci/ns87415.c b/drivers/ide/pci/ns87415.c
index ffefcd15196c..53bd645736d9 100644
--- a/drivers/ide/pci/ns87415.c
+++ b/drivers/ide/pci/ns87415.c
@@ -11,7 +11,6 @@
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/interrupt.h>
-#include <linux/hdreg.h>
 #include <linux/pci.h>
 #include <linux/delay.h>
 #include <linux/ide.h>
@@ -274,9 +273,9 @@ static void __devinit init_hwif_ns87415 (ide_hwif_t *hwif)
 		do {
 			udelay(50);
 			stat = hwif->tp_ops->read_status(hwif);
-                	if (stat == 0xff)
-                        	break;
-        	} while ((stat & BUSY_STAT) && --timeout);
+			if (stat == 0xff)
+				break;
+		} while ((stat & ATA_BUSY) && --timeout);
 #endif
 	}
 
@@ -340,6 +339,8 @@ static struct pci_driver driver = {
 	.id_table	= ns87415_pci_tbl,
 	.probe		= ns87415_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init ns87415_ide_init(void)
diff --git a/drivers/ide/pci/opti621.c b/drivers/ide/pci/opti621.c
index e28e672ddafc..3de11ddcf863 100644
--- a/drivers/ide/pci/opti621.c
+++ b/drivers/ide/pci/opti621.c
@@ -85,7 +85,6 @@
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/pci.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 
 #include <asm/io.h>
@@ -137,7 +136,7 @@ static u8 read_reg(int reg)
 static void opti621_set_pio_mode(ide_drive_t *drive, const u8 pio)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	ide_drive_t *pair = ide_get_paired_drive(drive);
+	ide_drive_t *pair = ide_get_pair_dev(drive);
 	unsigned long flags;
 	u8 tim, misc, addr_pio = pio, clk;
 
@@ -153,7 +152,7 @@ static void opti621_set_pio_mode(ide_drive_t *drive, const u8 pio)
 
 	drive->drive_data = XFER_PIO_0 + pio;
 
-	if (pair->present) {
+	if (pair) {
 		if (pair->drive_data && pair->drive_data < drive->drive_data)
 			addr_pio = pair->drive_data - XFER_PIO_0;
 	}
@@ -226,6 +225,8 @@ static struct pci_driver driver = {
 	.id_table	= opti621_pci_tbl,
 	.probe		= opti621_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init opti621_ide_init(void)
diff --git a/drivers/ide/pci/pdc202xx_new.c b/drivers/ide/pci/pdc202xx_new.c
index d477da6b5858..9fc59962553b 100644
--- a/drivers/ide/pci/pdc202xx_new.c
+++ b/drivers/ide/pci/pdc202xx_new.c
@@ -19,7 +19,6 @@
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/delay.h>
-#include <linux/hdreg.h>
 #include <linux/pci.h>
 #include <linux/init.h>
 #include <linux/ide.h>
@@ -203,10 +202,10 @@ static u8 pdcnew_cable_detect(ide_hwif_t *hwif)
 
 static void pdcnew_quirkproc(ide_drive_t *drive)
 {
-	const char **list, *model = drive->id->model;
+	const char **list, *m = (char *)&drive->id[ATA_ID_PROD];
 
 	for (list = pdc_quirk_drives; *list != NULL; list++)
-		if (strstr(model, *list) != NULL) {
+		if (strstr(m, *list) != NULL) {
 			drive->quirk_list = 2;
 			return;
 		}
@@ -227,7 +226,7 @@ static void pdcnew_reset(ide_drive_t *drive)
  * read_counter - Read the byte count registers
  * @dma_base: for the port address
  */
-static long __devinit read_counter(u32 dma_base)
+static long read_counter(u32 dma_base)
 {
 	u32  pri_dma_base = dma_base, sec_dma_base = dma_base + 0x08;
 	u8   cnt0, cnt1, cnt2, cnt3;
@@ -267,7 +266,7 @@ static long __devinit read_counter(u32 dma_base)
  * @dma_base: for the port address
  * E.g. 16949000 on 33 MHz PCI bus, i.e. half of the PCI clock.
  */
-static long __devinit detect_pll_input_clock(unsigned long dma_base)
+static long detect_pll_input_clock(unsigned long dma_base)
 {
 	struct timeval start_time, end_time;
 	long start_count, end_count;
@@ -310,7 +309,7 @@ static long __devinit detect_pll_input_clock(unsigned long dma_base)
 }
 
 #ifdef CONFIG_PPC_PMAC
-static void __devinit apple_kiwi_init(struct pci_dev *pdev)
+static void apple_kiwi_init(struct pci_dev *pdev)
 {
 	struct device_node *np = pci_device_to_OF_node(pdev);
 	u8 conf;
@@ -326,7 +325,7 @@ static void __devinit apple_kiwi_init(struct pci_dev *pdev)
 }
 #endif /* CONFIG_PPC_PMAC */
 
-static unsigned int __devinit init_chipset_pdcnew(struct pci_dev *dev)
+static unsigned int init_chipset_pdcnew(struct pci_dev *dev)
 {
 	const char *name = DRV_NAME;
 	unsigned long dma_base = pci_resource_start(dev, 4);
@@ -567,6 +566,8 @@ static struct pci_driver driver = {
 	.id_table	= pdc202new_pci_tbl,
 	.probe		= pdc202new_init_one,
 	.remove		= __devexit_p(pdc202new_remove),
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init pdc202new_ide_init(void)
diff --git a/drivers/ide/pci/pdc202xx_old.c b/drivers/ide/pci/pdc202xx_old.c
index de9a27400462..cb6d2a00c514 100644
--- a/drivers/ide/pci/pdc202xx_old.c
+++ b/drivers/ide/pci/pdc202xx_old.c
@@ -13,7 +13,6 @@
 #include <linux/kernel.h>
 #include <linux/delay.h>
 #include <linux/blkdev.h>
-#include <linux/hdreg.h>
 #include <linux/pci.h>
 #include <linux/init.h>
 #include <linux/ide.h>
@@ -86,7 +85,7 @@ static void pdc202xx_set_mode(ide_drive_t *drive, const u8 speed)
 		 * Prefetch_EN / IORDY_EN / PA[3:0] bits of register A
 		 */
 		AP &= ~0x3f;
-		if (drive->id->capability & 4)
+		if (ata_id_iordy_disable(drive->id))
 			AP |= 0x20;	/* set IORDY_EN bit */
 		if (drive->media == ide_disk)
 			AP |= 0x10;	/* set Prefetch_EN bit */
@@ -154,10 +153,10 @@ static void pdc_old_disable_66MHz_clock(ide_hwif_t *hwif)
 
 static void pdc202xx_quirkproc(ide_drive_t *drive)
 {
-	const char **list, *model = drive->id->model;
+	const char **list, *m = (char *)&drive->id[ATA_ID_PROD];
 
 	for (list = pdc_quirk_drives; *list != NULL; list++)
-		if (strstr(model, *list) != NULL) {
+		if (strstr(m, *list) != NULL) {
 			drive->quirk_list = 2;
 			return;
 		}
@@ -265,7 +264,7 @@ static void pdc202xx_dma_timeout(ide_drive_t *drive)
 	ide_dma_timeout(drive);
 }
 
-static unsigned int __devinit init_chipset_pdc202xx(struct pci_dev *dev)
+static unsigned int init_chipset_pdc202xx(struct pci_dev *dev)
 {
 	unsigned long dmabase = pci_resource_start(dev, 4);
 	u8 udma_speed_flag = 0, primary_mode = 0, secondary_mode = 0;
@@ -432,6 +431,8 @@ static struct pci_driver driver = {
 	.id_table	= pdc202xx_pci_tbl,
 	.probe		= pdc202xx_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init pdc202xx_ide_init(void)
diff --git a/drivers/ide/pci/piix.c b/drivers/ide/pci/piix.c
index 30cfc815fe31..a06c03f8e295 100644
--- a/drivers/ide/pci/piix.c
+++ b/drivers/ide/pci/piix.c
@@ -48,7 +48,6 @@
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/pci.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
@@ -205,7 +204,7 @@ static void piix_set_dma_mode(ide_drive_t *drive, const u8 speed)
  *	out to be nice and simple.
  */
 
-static unsigned int __devinit init_chipset_ich(struct pci_dev *dev)
+static unsigned int init_chipset_ich(struct pci_dev *dev)
 {
 	u32 extra = 0;
 
@@ -450,6 +449,8 @@ static struct pci_driver driver = {
 	.id_table	= piix_pci_tbl,
 	.probe		= piix_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init piix_ide_init(void)
diff --git a/drivers/ide/pci/rz1000.c b/drivers/ide/pci/rz1000.c
index 8d11ee838a2a..c117a068761b 100644
--- a/drivers/ide/pci/rz1000.c
+++ b/drivers/ide/pci/rz1000.c
@@ -16,7 +16,6 @@
 #include <linux/types.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
-#include <linux/hdreg.h>
 #include <linux/pci.h>
 #include <linux/ide.h>
 #include <linux/init.h>
diff --git a/drivers/ide/pci/sc1200.c b/drivers/ide/pci/sc1200.c
index 8efaed16fea3..bdc1fed41260 100644
--- a/drivers/ide/pci/sc1200.c
+++ b/drivers/ide/pci/sc1200.c
@@ -14,7 +14,6 @@
 #include <linux/module.h>
 #include <linux/types.h>
 #include <linux/kernel.h>
-#include <linux/hdreg.h>
 #include <linux/pci.h>
 #include <linux/init.h>
 #include <linux/ide.h>
@@ -104,17 +103,19 @@ static void sc1200_tunepio(ide_drive_t *drive, u8 pio)
 static u8 sc1200_udma_filter(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	ide_drive_t *mate = &hwif->drives[(drive->dn & 1) ^ 1];
-	struct hd_driveid *mateid = mate->id;
+	ide_drive_t *mate = ide_get_pair_dev(drive);
+	u16 *mateid = mate->id;
 	u8 mask = hwif->ultra_mask;
 
-	if (mate->present == 0)
+	if (mate == NULL)
 		goto out;
 
-	if ((mateid->capability & 1) && __ide_dma_bad_drive(mate) == 0) {
-		if ((mateid->field_valid & 4) && (mateid->dma_ultra & 7))
+	if (ata_id_has_dma(mateid) && __ide_dma_bad_drive(mate) == 0) {
+		if ((mateid[ATA_ID_FIELD_VALID] & 4) &&
+		    (mateid[ATA_ID_UDMA_MODES] & 7))
 			goto out;
-		if ((mateid->field_valid & 2) && (mateid->dma_mword & 7))
+		if ((mateid[ATA_ID_FIELD_VALID] & 2) &&
+		    (mateid[ATA_ID_MWDMA_MODES] & 7))
 			mask = 0;
 	}
 out:
diff --git a/drivers/ide/pci/scc_pata.c b/drivers/ide/pci/scc_pata.c
index 44cccd1e086a..e92a874b31df 100644
--- a/drivers/ide/pci/scc_pata.c
+++ b/drivers/ide/pci/scc_pata.c
@@ -26,7 +26,6 @@
 #include <linux/module.h>
 #include <linux/pci.h>
 #include <linux/delay.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
@@ -400,7 +399,7 @@ static int scc_dma_end(ide_drive_t *drive)
 	/* errata A308 workaround: Step5 (check data loss) */
 	/* We don't check non ide_disk because it is limited to UDMA4 */
 	if (!(in_be32((void __iomem *)hwif->io_ports.ctl_addr)
-	      & ERR_STAT) &&
+	      & ATA_ERR) &&
 	    drive->media == ide_disk && drive->current_speed > XFER_UDMA_4) {
 		reg = in_be32((void __iomem *)intsts_port);
 		if (!(reg & INTSTS_ACTEINT)) {
@@ -504,7 +503,7 @@ static int scc_dma_test_irq(ide_drive_t *drive)
 
 	/* SCC errata A252,A308 workaround: Step4 */
 	if ((in_be32((void __iomem *)hwif->io_ports.ctl_addr)
-	     & ERR_STAT) &&
+	     & ATA_ERR) &&
 	    (int_stat & INTSTS_INTRQ))
 		return 1;
 
diff --git a/drivers/ide/pci/serverworks.c b/drivers/ide/pci/serverworks.c
index c3bdc6e51a48..3dff2aea317e 100644
--- a/drivers/ide/pci/serverworks.c
+++ b/drivers/ide/pci/serverworks.c
@@ -32,7 +32,6 @@
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/pci.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
@@ -57,8 +56,10 @@ static struct pci_dev *isa_dev;
 
 static int check_in_drive_lists (ide_drive_t *drive, const char **list)
 {
+	char *m = (char *)&drive->id[ATA_ID_PROD];
+
 	while (*list)
-		if (!strcmp(*list++, drive->id->model))
+		if (!strcmp(*list++, m))
 			return 1;
 	return 0;
 }
@@ -174,7 +175,7 @@ static void svwks_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	pci_write_config_byte(dev, 0x54, ultra_enable);
 }
 
-static unsigned int __devinit init_chipset_svwks(struct pci_dev *dev)
+static unsigned int init_chipset_svwks(struct pci_dev *dev)
 {
 	unsigned int reg;
 	u8 btr;
@@ -447,6 +448,8 @@ static struct pci_driver driver = {
 	.id_table	= svwks_pci_tbl,
 	.probe		= svwks_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init svwks_ide_init(void)
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index 681306c9d79b..1017fb4f6317 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -22,7 +22,6 @@
 #include <linux/types.h>
 #include <linux/pci.h>
 #include <linux/delay.h>
-#include <linux/hdreg.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/ioport.h>
diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c
index db2b88a369ab..174a873b4c64 100644
--- a/drivers/ide/pci/siimage.c
+++ b/drivers/ide/pci/siimage.c
@@ -39,7 +39,6 @@
 #include <linux/types.h>
 #include <linux/module.h>
 #include <linux/pci.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 #include <linux/io.h>
@@ -223,7 +222,9 @@ static u8 sil_pata_udma_filter(ide_drive_t *drive)
 
 static u8 sil_sata_udma_filter(ide_drive_t *drive)
 {
-	return strstr(drive->id->model, "Maxtor") ? ATA_UDMA5 : ATA_UDMA6;
+	char *m = (char *)&drive->id[ATA_ID_PROD];
+
+	return strstr(m, "Maxtor") ? ATA_UDMA5 : ATA_UDMA6;
 }
 
 /**
@@ -243,7 +244,7 @@ static void sil_set_pio_mode(ide_drive_t *drive, u8 pio)
 
 	ide_hwif_t *hwif	= HWIF(drive);
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
-	ide_drive_t *pair	= ide_get_paired_drive(drive);
+	ide_drive_t *pair	= ide_get_pair_dev(drive);
 	u32 speedt		= 0;
 	u16 speedp		= 0;
 	unsigned long addr	= siimage_seldev(drive, 0x04);
@@ -257,7 +258,7 @@ static void sil_set_pio_mode(ide_drive_t *drive, u8 pio)
 	u8 unit			= drive->select.b.unit;
 
 	/* trim *taskfile* PIO to the slowest of the master/slave */
-	if (pair->present) {
+	if (pair) {
 		u8 pair_pio = ide_get_best_pio_mode(pair, 255, 4);
 
 		if (pair_pio < tf_pio)
@@ -462,7 +463,7 @@ static void sil_sata_pre_reset(ide_drive_t *drive)
  *	to 133 MHz clocking if the system isn't already set up to do it.
  */
 
-static unsigned int __devinit init_chipset_siimage(struct pci_dev *dev)
+static unsigned int init_chipset_siimage(struct pci_dev *dev)
 {
 	struct ide_host *host = pci_get_drvdata(dev);
 	void __iomem *ioaddr = host->host_priv;
@@ -616,8 +617,8 @@ static void __devinit init_mmio_iops_siimage(ide_hwif_t *hwif)
 
 static int is_dev_seagate_sata(ide_drive_t *drive)
 {
-	const char *s	= &drive->id->model[0];
-	unsigned len	= strnlen(s, sizeof(drive->id->model));
+	const char *s	= (const char *)&drive->id[ATA_ID_PROD];
+	unsigned len	= strnlen(s, ATA_ID_PROD_LEN);
 
 	if ((len > 4) && (!memcmp(s, "ST", 2)))
 		if ((!memcmp(s + len - 2, "AS", 2)) ||
@@ -833,6 +834,8 @@ static struct pci_driver driver = {
 	.id_table	= siimage_pci_tbl,
 	.probe		= siimage_init_one,
 	.remove		= __devexit_p(siimage_remove),
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init siimage_ide_init(void)
diff --git a/drivers/ide/pci/sis5513.c b/drivers/ide/pci/sis5513.c
index 5efe21d6ef97..734dd41f1f67 100644
--- a/drivers/ide/pci/sis5513.c
+++ b/drivers/ide/pci/sis5513.c
@@ -47,7 +47,6 @@
 #include <linux/types.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
-#include <linux/hdreg.h>
 #include <linux/pci.h>
 #include <linux/init.h>
 #include <linux/ide.h>
@@ -448,7 +447,7 @@ static int __devinit sis_find_family(struct pci_dev *dev)
 	return chipset_family;
 }
 
-static unsigned int __devinit init_chipset_sis5513(struct pci_dev *dev)
+static unsigned int init_chipset_sis5513(struct pci_dev *dev)
 {
 	/* Make general config ops here
 	   1/ tell IDE channels to operate in Compatibility mode only
@@ -611,6 +610,8 @@ static struct pci_driver driver = {
 	.id_table	= sis5513_pci_tbl,
 	.probe		= sis5513_init_one,
 	.remove		= __devexit_p(sis5513_remove),
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init sis5513_ide_init(void)
diff --git a/drivers/ide/pci/sl82c105.c b/drivers/ide/pci/sl82c105.c
index 73905bcc08fb..37a6b7bdc040 100644
--- a/drivers/ide/pci/sl82c105.c
+++ b/drivers/ide/pci/sl82c105.c
@@ -17,7 +17,6 @@
 #include <linux/types.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
-#include <linux/hdreg.h>
 #include <linux/pci.h>
 #include <linux/ide.h>
 
@@ -62,7 +61,7 @@ static unsigned int get_pio_timings(ide_drive_t *drive, u8 pio)
 	if (cmd_off == 0)
 		cmd_off = 1;
 
-	if (pio > 2 || ide_dev_has_iordy(drive->id))
+	if (pio > 2 || ata_id_has_iordy(drive->id))
 		iordy = 0x40;
 
 	return (cmd_on - 1) << 8 | (cmd_off - 1) | iordy;
@@ -272,7 +271,7 @@ static u8 sl82c105_bridge_revision(struct pci_dev *dev)
  * channel 0 here at least, but channel 1 has to be enabled by
  * firmware or arch code. We still set both to 16 bits mode.
  */
-static unsigned int __devinit init_chipset_sl82c105(struct pci_dev *dev)
+static unsigned int init_chipset_sl82c105(struct pci_dev *dev)
 {
 	u32 val;
 
@@ -351,6 +350,8 @@ static struct pci_driver driver = {
 	.id_table	= sl82c105_pci_tbl,
 	.probe		= sl82c105_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init sl82c105_ide_init(void)
diff --git a/drivers/ide/pci/slc90e66.c b/drivers/ide/pci/slc90e66.c
index 866d6c65e3a0..a9551a13ac57 100644
--- a/drivers/ide/pci/slc90e66.c
+++ b/drivers/ide/pci/slc90e66.c
@@ -11,7 +11,6 @@
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/pci.h>
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/init.h>
 
@@ -160,6 +159,8 @@ static struct pci_driver driver = {
 	.id_table	= slc90e66_pci_tbl,
 	.probe		= slc90e66_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init slc90e66_ide_init(void)
diff --git a/drivers/ide/pci/triflex.c b/drivers/ide/pci/triflex.c
index b77ec35151b3..be8715dcee05 100644
--- a/drivers/ide/pci/triflex.c
+++ b/drivers/ide/pci/triflex.c
@@ -28,7 +28,6 @@
 #include <linux/types.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
-#include <linux/hdreg.h>
 #include <linux/pci.h>
 #include <linux/ide.h>
 #include <linux/init.h>
@@ -120,6 +119,8 @@ static struct pci_driver driver = {
 	.id_table	= triflex_pci_tbl,
 	.probe		= triflex_init_one,
 	.remove		= ide_pci_remove,
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init triflex_ide_init(void)
diff --git a/drivers/ide/pci/trm290.c b/drivers/ide/pci/trm290.c
index fd28b49977fd..4dfbc6a68b5b 100644
--- a/drivers/ide/pci/trm290.c
+++ b/drivers/ide/pci/trm290.c
@@ -135,7 +135,6 @@
 #include <linux/interrupt.h>
 #include <linux/blkdev.h>
 #include <linux/init.h>
-#include <linux/hdreg.h>
 #include <linux/pci.h>
 #include <linux/ide.h>
 
diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c
index 94fb9ab3223f..acacdaab69c2 100644
--- a/drivers/ide/pci/via82cxxx.c
+++ b/drivers/ide/pci/via82cxxx.c
@@ -154,7 +154,7 @@ static void via_set_speed(ide_hwif_t *hwif, u8 dn, struct ide_timing *timing)
 static void via_set_drive(ide_drive_t *drive, const u8 speed)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	ide_drive_t *peer = hwif->drives + (~drive->dn & 1);
+	ide_drive_t *peer = ide_get_pair_dev(drive);
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
 	struct ide_host *host = pci_get_drvdata(dev);
 	struct via82cxxx_dev *vdev = host->host_priv;
@@ -173,7 +173,7 @@ static void via_set_drive(ide_drive_t *drive, const u8 speed)
 
 	ide_timing_compute(drive, speed, &t, T, UT);
 
-	if (peer->present) {
+	if (peer) {
 		ide_timing_compute(peer, peer->current_speed, &p, T, UT);
 		ide_timing_merge(&p, &t, &t, IDE_TIMING_8BIT);
 	}
@@ -215,7 +215,7 @@ static struct via_isa_bridge *via_config_find(struct pci_dev **isa)
 /*
  * Check and handle 80-wire cable presence
  */
-static void __devinit via_cable_detect(struct via82cxxx_dev *vdev, u32 u)
+static void via_cable_detect(struct via82cxxx_dev *vdev, u32 u)
 {
 	int i;
 
@@ -267,7 +267,7 @@ static void __devinit via_cable_detect(struct via82cxxx_dev *vdev, u32 u)
  *	and initialize its drive independent registers.
  */
 
-static unsigned int __devinit init_chipset_via82cxxx(struct pci_dev *dev)
+static unsigned int init_chipset_via82cxxx(struct pci_dev *dev)
 {
 	struct ide_host *host = pci_get_drvdata(dev);
 	struct via82cxxx_dev *vdev = host->host_priv;
@@ -492,6 +492,8 @@ static struct pci_driver driver = {
 	.id_table 	= via_pci_tbl,
 	.probe 		= via_init_one,
 	.remove		= __devexit_p(via_remove),
+	.suspend	= ide_pci_suspend,
+	.resume		= ide_pci_resume,
 };
 
 static int __init via_ide_init(void)
diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index fa2be26272d5..c3432da78d52 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -669,9 +669,9 @@ static void
 set_timings_mdma(ide_drive_t *drive, int intf_type, u32 *timings, u32 *timings2,
 		 	u8 speed)
 {
+	u16 *id = drive->id;
 	int cycleTime, accessTime = 0, recTime = 0;
 	unsigned accessTicks, recTicks;
-	struct hd_driveid *id = drive->id;
 	struct mdma_timings_t* tm = NULL;
 	int i;
 
@@ -686,8 +686,8 @@ set_timings_mdma(ide_drive_t *drive, int intf_type, u32 *timings, u32 *timings2,
 	}
 
 	/* Check if drive provides explicit DMA cycle time */
-	if ((id->field_valid & 2) && id->eide_dma_time)
-		cycleTime = max_t(int, id->eide_dma_time, cycleTime);
+	if ((id[ATA_ID_FIELD_VALID] & 2) && id[ATA_ID_EIDE_DMA_TIME])
+		cycleTime = max_t(int, id[ATA_ID_EIDE_DMA_TIME], cycleTime);
 
 	/* OHare limits according to some old Apple sources */	
 	if ((intf_type == controller_ohare) && (cycleTime < 150))
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index a8e9e8a69a52..9f1f9163a136 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -659,3 +659,36 @@ void ide_pci_remove(struct pci_dev *dev)
 	pci_disable_device(dev);
 }
 EXPORT_SYMBOL_GPL(ide_pci_remove);
+
+#ifdef CONFIG_PM
+int ide_pci_suspend(struct pci_dev *dev, pm_message_t state)
+{
+	pci_save_state(dev);
+	pci_disable_device(dev);
+	pci_set_power_state(dev, pci_choose_state(dev, state));
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ide_pci_suspend);
+
+int ide_pci_resume(struct pci_dev *dev)
+{
+	struct ide_host *host = pci_get_drvdata(dev);
+	int rc;
+
+	pci_set_power_state(dev, PCI_D0);
+
+	rc = pci_enable_device(dev);
+	if (rc)
+		return rc;
+
+	pci_restore_state(dev);
+	pci_set_master(dev);
+
+	if (host->init_chipset)
+		host->init_chipset(dev);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ide_pci_resume);
+#endif
diff --git a/drivers/input/misc/sparcspkr.c b/drivers/input/misc/sparcspkr.c
index d8765cc93d27..c4f42311fdec 100644
--- a/drivers/input/misc/sparcspkr.c
+++ b/drivers/input/misc/sparcspkr.c
@@ -249,7 +249,7 @@ static int bbc_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id bbc_beep_match[] = {
+static const struct of_device_id bbc_beep_match[] = {
 	{
 		.name = "beep",
 		.compatible = "SUNW,bbc-beep",
@@ -328,7 +328,7 @@ static int grover_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id grover_beep_match[] = {
+static const struct of_device_id grover_beep_match[] = {
 	{
 		.name = "beep",
 		.compatible = "SUNW,smbus-beep",
diff --git a/drivers/input/serio/i8042-sparcio.h b/drivers/input/serio/i8042-sparcio.h
index 692a79ec2a22..5071af2c0604 100644
--- a/drivers/input/serio/i8042-sparcio.h
+++ b/drivers/input/serio/i8042-sparcio.h
@@ -87,7 +87,7 @@ static int __devexit sparc_i8042_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id sparc_i8042_match[] = {
+static const struct of_device_id sparc_i8042_match[] = {
 	{
 		.name = "8042",
 	},
diff --git a/drivers/input/xen-kbdfront.c b/drivers/input/xen-kbdfront.c
index 9ce3b3baf3a2..3ab6362f043c 100644
--- a/drivers/input/xen-kbdfront.c
+++ b/drivers/input/xen-kbdfront.c
@@ -335,11 +335,11 @@ static struct xenbus_driver xenkbd = {
 
 static int __init xenkbd_init(void)
 {
-	if (!is_running_on_xen())
+	if (!xen_domain())
 		return -ENODEV;
 
 	/* Nothing to do if running in dom0. */
-	if (is_initial_xendomain())
+	if (xen_initial_domain())
 		return -ENODEV;
 
 	return xenbus_register_frontend(&xenkbd);
diff --git a/drivers/isdn/capi/capi.c b/drivers/isdn/capi/capi.c
index 871b0cbca5e4..798d7f3e42ef 100644
--- a/drivers/isdn/capi/capi.c
+++ b/drivers/isdn/capi/capi.c
@@ -1231,7 +1231,7 @@ static int capinc_tty_ioctl(struct tty_struct *tty, struct file * file,
 	int error = 0;
 	switch (cmd) {
 	default:
-		error = n_tty_ioctl (tty, file, cmd, arg);
+		error = n_tty_ioctl_helper(tty, file, cmd, arg);
 		break;
 	}
 	return error;
diff --git a/drivers/isdn/gigaset/ser-gigaset.c b/drivers/isdn/gigaset/ser-gigaset.c
index 5e89fa177816..07052ed2a0c5 100644
--- a/drivers/isdn/gigaset/ser-gigaset.c
+++ b/drivers/isdn/gigaset/ser-gigaset.c
@@ -571,6 +571,7 @@ gigaset_tty_close(struct tty_struct *tty)
 	}
 
 	/* prevent other callers from entering ldisc methods */
+	/* FIXME: should use the tty state flags */
 	tty->disc_data = NULL;
 
 	if (!cs->hw.ser)
@@ -642,10 +643,11 @@ gigaset_tty_ioctl(struct tty_struct *tty, struct file *file,
 		return -ENXIO;
 
 	switch (cmd) {
-	case TCGETS:
-	case TCGETA:
-		/* pass through to underlying serial device */
-		rc = n_tty_ioctl(tty, file, cmd, arg);
+
+	case FIONREAD:
+		/* unused, always return zero */
+		val = 0;
+		rc = put_user(val, p);
 		break;
 
 	case TCFLSH:
@@ -659,20 +661,13 @@ gigaset_tty_ioctl(struct tty_struct *tty, struct file *file,
 			flush_send_queue(cs);
 			break;
 		}
-		/* flush the serial port's buffer */
-		rc = n_tty_ioctl(tty, file, cmd, arg);
-		break;
-
-	case FIONREAD:
-		/* unused, always return zero */
-		val = 0;
-		rc = put_user(val, p);
-		break;
+		/* Pass through */
 
 	default:
-		rc = -ENOIOCTLCMD;
+		/* pass through to underlying serial device */
+		rc = n_tty_ioctl_helper(tty, file, cmd, arg);
+		break;
 	}
-
 	cs_put(cs);
 	return rc;
 }
@@ -680,6 +675,8 @@ gigaset_tty_ioctl(struct tty_struct *tty, struct file *file,
 /*
  * Poll on the tty.
  * Unused, always return zero.
+ *
+ * FIXME: should probably return an exception - especially on hangup
  */
 static unsigned int
 gigaset_tty_poll(struct tty_struct *tty, struct file *file, poll_table *wait)
diff --git a/drivers/leds/Kconfig b/drivers/leds/Kconfig
index 5b14262af017..e3e40427e00e 100644
--- a/drivers/leds/Kconfig
+++ b/drivers/leds/Kconfig
@@ -82,6 +82,14 @@ config LEDS_COBALT_RAQ
 	help
 	  This option enables support for the Cobalt Raq series LEDs.
 
+config LEDS_SUNFIRE
+	tristate "LED support for SunFire servers."
+	depends on LEDS_CLASS && SPARC64
+	select LEDS_TRIGGERS
+	help
+	  This option enables support for the Left, Middle, and Right
+	  LEDs on the I/O and CPU boards of SunFire UltraSPARC servers.
+
 config LEDS_HP6XX
 	tristate "LED Support for the HP Jornada 6xx"
 	depends on LEDS_CLASS && SH_HP6XX
diff --git a/drivers/leds/Makefile b/drivers/leds/Makefile
index 3a8e6a04363c..eb186c351a1c 100644
--- a/drivers/leds/Makefile
+++ b/drivers/leds/Makefile
@@ -14,6 +14,7 @@ obj-$(CONFIG_LEDS_WRAP)			+= leds-wrap.o
 obj-$(CONFIG_LEDS_H1940)		+= leds-h1940.o
 obj-$(CONFIG_LEDS_COBALT_QUBE)		+= leds-cobalt-qube.o
 obj-$(CONFIG_LEDS_COBALT_RAQ)		+= leds-cobalt-raq.o
+obj-$(CONFIG_LEDS_SUNFIRE)		+= leds-sunfire.o
 obj-$(CONFIG_LEDS_PCA9532)		+= leds-pca9532.o
 obj-$(CONFIG_LEDS_GPIO)			+= leds-gpio.o
 obj-$(CONFIG_LEDS_CM_X270)              += leds-cm-x270.o
diff --git a/drivers/leds/leds-sunfire.c b/drivers/leds/leds-sunfire.c
new file mode 100644
index 000000000000..6b008f0c3f62
--- /dev/null
+++ b/drivers/leds/leds-sunfire.c
@@ -0,0 +1,273 @@
+/* leds-sunfire.c: SUNW,Ultra-Enterprise LED driver.
+ *
+ * Copyright (C) 2008 David S. Miller <davem@davemloft.net>
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/leds.h>
+#include <linux/io.h>
+#include <linux/platform_device.h>
+
+#include <asm/fhc.h>
+#include <asm/upa.h>
+
+#define DRIVER_NAME	"leds-sunfire"
+#define PFX		DRIVER_NAME ": "
+
+MODULE_AUTHOR("David S. Miller (davem@davemloft.net)");
+MODULE_DESCRIPTION("Sun Fire LED driver");
+MODULE_LICENSE("GPL");
+
+struct sunfire_led {
+	struct led_classdev	led_cdev;
+	void __iomem		*reg;
+};
+#define	to_sunfire_led(d) container_of(d, struct sunfire_led, led_cdev)
+
+static void __clockboard_set(struct led_classdev *led_cdev,
+			     enum led_brightness led_val, u8 bit)
+{
+	struct sunfire_led *p = to_sunfire_led(led_cdev);
+	u8 reg = upa_readb(p->reg);
+
+	switch (bit) {
+	case CLOCK_CTRL_LLED:
+		if (led_val)
+			reg &= ~bit;
+		else
+			reg |= bit;
+		break;
+
+	default:
+		if (led_val)
+			reg |= bit;
+		else
+			reg &= ~bit;
+		break;
+	}
+	upa_writeb(reg, p->reg);
+}
+
+static void clockboard_left_set(struct led_classdev *led_cdev,
+				enum led_brightness led_val)
+{
+	__clockboard_set(led_cdev, led_val, CLOCK_CTRL_LLED);
+}
+
+static void clockboard_middle_set(struct led_classdev *led_cdev,
+				  enum led_brightness led_val)
+{
+	__clockboard_set(led_cdev, led_val, CLOCK_CTRL_MLED);
+}
+
+static void clockboard_right_set(struct led_classdev *led_cdev,
+				 enum led_brightness led_val)
+{
+	__clockboard_set(led_cdev, led_val, CLOCK_CTRL_RLED);
+}
+
+static void __fhc_set(struct led_classdev *led_cdev,
+			     enum led_brightness led_val, u32 bit)
+{
+	struct sunfire_led *p = to_sunfire_led(led_cdev);
+	u32 reg = upa_readl(p->reg);
+
+	switch (bit) {
+	case FHC_CONTROL_LLED:
+		if (led_val)
+			reg &= ~bit;
+		else
+			reg |= bit;
+		break;
+
+	default:
+		if (led_val)
+			reg |= bit;
+		else
+			reg &= ~bit;
+		break;
+	}
+	upa_writel(reg, p->reg);
+}
+
+static void fhc_left_set(struct led_classdev *led_cdev,
+			 enum led_brightness led_val)
+{
+	__fhc_set(led_cdev, led_val, FHC_CONTROL_LLED);
+}
+
+static void fhc_middle_set(struct led_classdev *led_cdev,
+			   enum led_brightness led_val)
+{
+	__fhc_set(led_cdev, led_val, FHC_CONTROL_MLED);
+}
+
+static void fhc_right_set(struct led_classdev *led_cdev,
+			  enum led_brightness led_val)
+{
+	__fhc_set(led_cdev, led_val, FHC_CONTROL_RLED);
+}
+
+typedef void (*set_handler)(struct led_classdev *, enum led_brightness);
+struct led_type {
+	const char	*name;
+	set_handler	handler;
+	const char	*default_trigger;
+};
+
+#define NUM_LEDS_PER_BOARD	3
+struct sunfire_drvdata {
+	struct sunfire_led	leds[NUM_LEDS_PER_BOARD];
+};
+
+static int __devinit sunfire_led_generic_probe(struct platform_device *pdev,
+					       struct led_type *types)
+{
+	struct sunfire_drvdata *p;
+	int i, err = -EINVAL;
+
+	if (pdev->num_resources != 1) {
+		printk(KERN_ERR PFX "Wrong number of resources %d, should be 1\n",
+		       pdev->num_resources);
+		goto out;
+	}
+
+	p = kzalloc(sizeof(*p), GFP_KERNEL);
+	if (!p) {
+		printk(KERN_ERR PFX "Could not allocate struct sunfire_drvdata\n");
+		goto out;
+	}
+
+	for (i = 0; i < NUM_LEDS_PER_BOARD; i++) {
+		struct led_classdev *lp = &p->leds[i].led_cdev;
+
+		p->leds[i].reg = (void __iomem *) pdev->resource[0].start;
+		lp->name = types[i].name;
+		lp->brightness = LED_FULL;
+		lp->brightness_set = types[i].handler;
+		lp->default_trigger = types[i].default_trigger;
+
+		err = led_classdev_register(&pdev->dev, lp);
+		if (err) {
+			printk(KERN_ERR PFX "Could not register %s LED\n",
+			       lp->name);
+			goto out_unregister_led_cdevs;
+		}
+	}
+
+	dev_set_drvdata(&pdev->dev, p);
+
+	err = 0;
+out:
+	return err;
+
+out_unregister_led_cdevs:
+	for (i--; i >= 0; i--)
+		led_classdev_unregister(&p->leds[i].led_cdev);
+	goto out;
+}
+
+static int __devexit sunfire_led_generic_remove(struct platform_device *pdev)
+{
+	struct sunfire_drvdata *p = dev_get_drvdata(&pdev->dev);
+	int i;
+
+	for (i = 0; i < NUM_LEDS_PER_BOARD; i++)
+		led_classdev_unregister(&p->leds[i].led_cdev);
+
+	kfree(p);
+
+	return 0;
+}
+
+static struct led_type clockboard_led_types[NUM_LEDS_PER_BOARD] = {
+	{
+		.name		= "clockboard-left",
+		.handler	= clockboard_left_set,
+	},
+	{
+		.name		= "clockboard-middle",
+		.handler	= clockboard_middle_set,
+	},
+	{
+		.name		= "clockboard-right",
+		.handler	= clockboard_right_set,
+		.default_trigger= "heartbeat",
+	},
+};
+
+static int __devinit sunfire_clockboard_led_probe(struct platform_device *pdev)
+{
+	return sunfire_led_generic_probe(pdev, clockboard_led_types);
+}
+
+static struct led_type fhc_led_types[NUM_LEDS_PER_BOARD] = {
+	{
+		.name		= "fhc-left",
+		.handler	= fhc_left_set,
+	},
+	{
+		.name		= "fhc-middle",
+		.handler	= fhc_middle_set,
+	},
+	{
+		.name		= "fhc-right",
+		.handler	= fhc_right_set,
+		.default_trigger= "heartbeat",
+	},
+};
+
+static int __devinit sunfire_fhc_led_probe(struct platform_device *pdev)
+{
+	return sunfire_led_generic_probe(pdev, fhc_led_types);
+}
+
+MODULE_ALIAS("platform:sunfire-clockboard-leds");
+MODULE_ALIAS("platform:sunfire-fhc-leds");
+
+static struct platform_driver sunfire_clockboard_led_driver = {
+	.probe		= sunfire_clockboard_led_probe,
+	.remove		= __devexit_p(sunfire_led_generic_remove),
+	.driver		= {
+		.name	= "sunfire-clockboard-leds",
+		.owner	= THIS_MODULE,
+	},
+};
+
+static struct platform_driver sunfire_fhc_led_driver = {
+	.probe		= sunfire_fhc_led_probe,
+	.remove		= __devexit_p(sunfire_led_generic_remove),
+	.driver		= {
+		.name	= "sunfire-fhc-leds",
+		.owner	= THIS_MODULE,
+	},
+};
+
+static int __init sunfire_leds_init(void)
+{
+	int err = platform_driver_register(&sunfire_clockboard_led_driver);
+
+	if (err) {
+		printk(KERN_ERR PFX "Could not register clock board LED driver\n");
+		return err;
+	}
+
+	err = platform_driver_register(&sunfire_fhc_led_driver);
+	if (err) {
+		printk(KERN_ERR PFX "Could not register FHC LED driver\n");
+		platform_driver_unregister(&sunfire_clockboard_led_driver);
+	}
+
+	return err;
+}
+
+static void __exit sunfire_leds_exit(void)
+{
+	platform_driver_unregister(&sunfire_clockboard_led_driver);
+	platform_driver_unregister(&sunfire_fhc_led_driver);
+}
+
+module_init(sunfire_leds_init);
+module_exit(sunfire_leds_exit);
diff --git a/drivers/media/video/cafe_ccic.c b/drivers/media/video/cafe_ccic.c
index 5405c30dbb04..08efbe7254ff 100644
--- a/drivers/media/video/cafe_ccic.c
+++ b/drivers/media/video/cafe_ccic.c
@@ -2092,15 +2092,8 @@ static int cafe_pci_probe(struct pci_dev *pdev,
 		const struct pci_device_id *id)
 {
 	int ret;
-	u16 classword;
 	struct cafe_camera *cam;
-	/*
-	 * Make sure we have a camera here - we'll get calls for
-	 * the other cafe devices as well.
-	 */
-	pci_read_config_word(pdev, PCI_CLASS_DEVICE, &classword);
-	if (classword != PCI_CLASS_MULTIMEDIA_VIDEO)
-		return -ENODEV;
+
 	/*
 	 * Start putting together one of our big camera structures.
 	 */
@@ -2288,8 +2281,8 @@ static int cafe_pci_resume(struct pci_dev *pdev)
 
 
 static struct pci_device_id cafe_ids[] = {
-	{ PCI_DEVICE(0x11ab, 0x4100) }, /* Eventual real ID */
-	{ PCI_DEVICE(0x11ab, 0x4102) }, /* Really eventual real ID */
+	{ PCI_DEVICE(PCI_VENDOR_ID_MARVELL,
+		     PCI_DEVICE_ID_MARVELL_88ALP01_CCIC) },
 	{ 0, }
 };
 
diff --git a/drivers/mmc/Kconfig b/drivers/mmc/Kconfig
index c0b41e8bcd9d..f2eeb38efa65 100644
--- a/drivers/mmc/Kconfig
+++ b/drivers/mmc/Kconfig
@@ -3,13 +3,14 @@
 #
 
 menuconfig MMC
-	tristate "MMC/SD card support"
+	tristate "MMC/SD/SDIO card support"
 	depends on HAS_IOMEM
 	help
-	  MMC is the "multi-media card" bus protocol.
+	  This selects MultiMediaCard, Secure Digital and Secure
+	  Digital I/O support.
 
-	  If you want MMC support, you should say Y here and also
-	  to the specific driver for your MMC interface.
+	  If you want MMC/SD/SDIO support, you should say Y here and
+	  also to your specific host controller driver.
 
 config MMC_DEBUG
 	bool "MMC debugging"
diff --git a/drivers/mmc/card/Kconfig b/drivers/mmc/card/Kconfig
index dd0f398ee2f5..3f2a912659af 100644
--- a/drivers/mmc/card/Kconfig
+++ b/drivers/mmc/card/Kconfig
@@ -2,7 +2,7 @@
 # MMC/SD card drivers
 #
 
-comment "MMC/SD Card Drivers"
+comment "MMC/SD/SDIO Card Drivers"
 
 config MMC_BLOCK
 	tristate "MMC block device driver"
@@ -34,7 +34,6 @@ config MMC_BLOCK_BOUNCE
 
 config SDIO_UART
 	tristate "SDIO UART/GPS class support"
-	depends on MMC
 	help
 	  SDIO function driver for SDIO cards that implements the UART
 	  class, as well as the GPS class which appears like a UART.
diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c
index efacee0404a0..24c97d3d16bb 100644
--- a/drivers/mmc/card/block.c
+++ b/drivers/mmc/card/block.c
@@ -58,7 +58,6 @@ struct mmc_blk_data {
 	struct mmc_queue queue;
 
 	unsigned int	usage;
-	unsigned int	block_bits;
 	unsigned int	read_only;
 };
 
@@ -216,8 +215,7 @@ static int mmc_blk_issue_rq(struct mmc_queue *mq, struct request *req)
 	struct mmc_blk_data *md = mq->data;
 	struct mmc_card *card = md->queue.card;
 	struct mmc_blk_request brq;
-	int ret = 1, data_size, i;
-	struct scatterlist *sg;
+	int ret = 1;
 
 	mmc_claim_host(card->host);
 
@@ -233,13 +231,11 @@ static int mmc_blk_issue_rq(struct mmc_queue *mq, struct request *req)
 		if (!mmc_card_blockaddr(card))
 			brq.cmd.arg <<= 9;
 		brq.cmd.flags = MMC_RSP_SPI_R1 | MMC_RSP_R1 | MMC_CMD_ADTC;
-		brq.data.blksz = 1 << md->block_bits;
+		brq.data.blksz = 512;
 		brq.stop.opcode = MMC_STOP_TRANSMISSION;
 		brq.stop.arg = 0;
 		brq.stop.flags = MMC_RSP_SPI_R1B | MMC_RSP_R1B | MMC_CMD_AC;
-		brq.data.blocks = req->nr_sectors >> (md->block_bits - 9);
-		if (brq.data.blocks > card->host->max_blk_count)
-			brq.data.blocks = card->host->max_blk_count;
+		brq.data.blocks = req->nr_sectors;
 
 		if (brq.data.blocks > 1) {
 			/* SPI multiblock writes terminate using a special
@@ -271,24 +267,6 @@ static int mmc_blk_issue_rq(struct mmc_queue *mq, struct request *req)
 
 		mmc_queue_bounce_pre(mq);
 
-		/*
-		 * Adjust the sg list so it is the same size as the
-		 * request.
-		 */
-		if (brq.data.blocks !=
-		    (req->nr_sectors >> (md->block_bits - 9))) {
-			data_size = brq.data.blocks * brq.data.blksz;
-			for_each_sg(brq.data.sg, sg, brq.data.sg_len, i) {
-				data_size -= sg->length;
-				if (data_size <= 0) {
-					sg->length += data_size;
-					i++;
-					break;
-				}
-			}
-			brq.data.sg_len = i;
-		}
-
 		mmc_wait_for_req(card->host, &brq.mrq);
 
 		mmc_queue_bounce_post(mq);
@@ -373,16 +351,11 @@ static int mmc_blk_issue_rq(struct mmc_queue *mq, struct request *req)
 	if (rq_data_dir(req) != READ) {
 		if (mmc_card_sd(card)) {
 			u32 blocks;
-			unsigned int bytes;
 
 			blocks = mmc_sd_num_wr_blocks(card);
 			if (blocks != (u32)-1) {
-				if (card->csd.write_partial)
-					bytes = blocks << md->block_bits;
-				else
-					bytes = blocks << 9;
 				spin_lock_irq(&md->lock);
-				ret = __blk_end_request(req, 0, bytes);
+				ret = __blk_end_request(req, 0, blocks << 9);
 				spin_unlock_irq(&md->lock);
 			}
 		} else {
@@ -432,13 +405,6 @@ static struct mmc_blk_data *mmc_blk_alloc(struct mmc_card *card)
 	 */
 	md->read_only = mmc_blk_readonly(card);
 
-	/*
-	 * Both SD and MMC specifications state (although a bit
-	 * unclearly in the MMC case) that a block size of 512
-	 * bytes must always be supported by the card.
-	 */
-	md->block_bits = 9;
-
 	md->disk = alloc_disk(1 << MMC_SHIFT);
 	if (md->disk == NULL) {
 		ret = -ENOMEM;
@@ -476,7 +442,7 @@ static struct mmc_blk_data *mmc_blk_alloc(struct mmc_card *card)
 
 	sprintf(md->disk->disk_name, "mmcblk%d", devidx);
 
-	blk_queue_hardsect_size(md->queue.queue, 1 << md->block_bits);
+	blk_queue_hardsect_size(md->queue.queue, 512);
 
 	if (!mmc_card_sd(card) && mmc_card_blockaddr(card)) {
 		/*
@@ -514,7 +480,7 @@ mmc_blk_set_blksize(struct mmc_blk_data *md, struct mmc_card *card)
 
 	mmc_claim_host(card->host);
 	cmd.opcode = MMC_SET_BLOCKLEN;
-	cmd.arg = 1 << md->block_bits;
+	cmd.arg = 512;
 	cmd.flags = MMC_RSP_SPI_R1 | MMC_RSP_R1 | MMC_CMD_AC;
 	err = mmc_wait_for_cmd(card->host, &cmd, 5);
 	mmc_release_host(card->host);
diff --git a/drivers/mmc/card/queue.c b/drivers/mmc/card/queue.c
index 3dee97e7d165..406989e992ba 100644
--- a/drivers/mmc/card/queue.c
+++ b/drivers/mmc/card/queue.c
@@ -31,7 +31,7 @@ static int mmc_prep_request(struct request_queue *q, struct request *req)
 	/*
 	 * We only like normal block requests.
 	 */
-	if (!blk_fs_request(req) && !blk_pc_request(req)) {
+	if (!blk_fs_request(req)) {
 		blk_dump_rq_flags(req, "MMC bad request");
 		return BLKPREP_KILL;
 	}
@@ -131,6 +131,7 @@ int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card, spinlock_t *lock
 	mq->req = NULL;
 
 	blk_queue_prep_rq(mq->queue, mmc_prep_request);
+	blk_queue_ordered(mq->queue, QUEUE_ORDERED_DRAIN, NULL);
 
 #ifdef CONFIG_MMC_BLOCK_BOUNCE
 	if (host->max_hw_segs == 1) {
@@ -142,12 +143,19 @@ int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card, spinlock_t *lock
 			bouncesz = host->max_req_size;
 		if (bouncesz > host->max_seg_size)
 			bouncesz = host->max_seg_size;
+		if (bouncesz > (host->max_blk_count * 512))
+			bouncesz = host->max_blk_count * 512;
+
+		if (bouncesz > 512) {
+			mq->bounce_buf = kmalloc(bouncesz, GFP_KERNEL);
+			if (!mq->bounce_buf) {
+				printk(KERN_WARNING "%s: unable to "
+					"allocate bounce buffer\n",
+					mmc_card_name(card));
+			}
+		}
 
-		mq->bounce_buf = kmalloc(bouncesz, GFP_KERNEL);
-		if (!mq->bounce_buf) {
-			printk(KERN_WARNING "%s: unable to allocate "
-				"bounce buffer\n", mmc_card_name(card));
-		} else {
+		if (mq->bounce_buf) {
 			blk_queue_bounce_limit(mq->queue, BLK_BOUNCE_ANY);
 			blk_queue_max_sectors(mq->queue, bouncesz / 512);
 			blk_queue_max_phys_segments(mq->queue, bouncesz / 512);
@@ -175,7 +183,8 @@ int mmc_init_queue(struct mmc_queue *mq, struct mmc_card *card, spinlock_t *lock
 
 	if (!mq->bounce_buf) {
 		blk_queue_bounce_limit(mq->queue, limit);
-		blk_queue_max_sectors(mq->queue, host->max_req_size / 512);
+		blk_queue_max_sectors(mq->queue,
+			min(host->max_blk_count, host->max_req_size / 512));
 		blk_queue_max_phys_segments(mq->queue, host->max_phys_segs);
 		blk_queue_max_hw_segments(mq->queue, host->max_hw_segs);
 		blk_queue_max_segment_size(mq->queue, host->max_seg_size);
diff --git a/drivers/mmc/core/mmc_ops.c b/drivers/mmc/core/mmc_ops.c
index 64b05c6270f2..9c50e6f1c236 100644
--- a/drivers/mmc/core/mmc_ops.c
+++ b/drivers/mmc/core/mmc_ops.c
@@ -248,8 +248,12 @@ mmc_send_cxd_data(struct mmc_card *card, struct mmc_host *host,
 
 	sg_init_one(&sg, data_buf, len);
 
-	if (card)
-		mmc_set_data_timeout(&data, card);
+	/*
+	 * The spec states that CSR and CID accesses have a timeout
+	 * of 64 clock cycles.
+	 */
+	data.timeout_ns = 0;
+	data.timeout_clks = 64;
 
 	mmc_wait_for_req(host, &mrq);
 
diff --git a/drivers/mmc/core/sdio.c b/drivers/mmc/core/sdio.c
index 4eab79e09ccc..fb99ccff9080 100644
--- a/drivers/mmc/core/sdio.c
+++ b/drivers/mmc/core/sdio.c
@@ -165,6 +165,36 @@ static int sdio_enable_wide(struct mmc_card *card)
 }
 
 /*
+ * Test if the card supports high-speed mode and, if so, switch to it.
+ */
+static int sdio_enable_hs(struct mmc_card *card)
+{
+	int ret;
+	u8 speed;
+
+	if (!(card->host->caps & MMC_CAP_SD_HIGHSPEED))
+		return 0;
+
+	if (!card->cccr.high_speed)
+		return 0;
+
+	ret = mmc_io_rw_direct(card, 0, 0, SDIO_CCCR_SPEED, 0, &speed);
+	if (ret)
+		return ret;
+
+	speed |= SDIO_SPEED_EHS;
+
+	ret = mmc_io_rw_direct(card, 1, 0, SDIO_CCCR_SPEED, speed, NULL);
+	if (ret)
+		return ret;
+
+	mmc_card_set_highspeed(card);
+	mmc_set_timing(card->host, MMC_TIMING_SD_HS);
+
+	return 0;
+}
+
+/*
  * Host is being removed. Free up the current card.
  */
 static void mmc_sdio_remove(struct mmc_host *host)
@@ -333,10 +363,26 @@ int mmc_attach_sdio(struct mmc_host *host, u32 ocr)
 		goto remove;
 
 	/*
-	 * No support for high-speed yet, so just set
-	 * the card's maximum speed.
+	 * Switch to high-speed (if supported).
 	 */
-	mmc_set_clock(host, card->cis.max_dtr);
+	err = sdio_enable_hs(card);
+	if (err)
+		goto remove;
+
+	/*
+	 * Change to the card's maximum speed.
+	 */
+	if (mmc_card_highspeed(card)) {
+		/*
+		 * The SDIO specification doesn't mention how
+		 * the CIS transfer speed register relates to
+		 * high-speed, but it seems that 50 MHz is
+		 * mandatory.
+		 */
+		mmc_set_clock(host, 50000000);
+	} else {
+		mmc_set_clock(host, card->cis.max_dtr);
+	}
 
 	/*
 	 * Switch to wider bus (if supported).
diff --git a/drivers/mmc/core/sdio_irq.c b/drivers/mmc/core/sdio_irq.c
index c292e124107a..bb192f90e8e9 100644
--- a/drivers/mmc/core/sdio_irq.c
+++ b/drivers/mmc/core/sdio_irq.c
@@ -5,6 +5,8 @@
  * Created:     June 18, 2007
  * Copyright:   MontaVista Software Inc.
  *
+ * Copyright 2008 Pierre Ossman
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or (at
@@ -107,11 +109,14 @@ static int sdio_irq_thread(void *_host)
 
 		/*
 		 * Give other threads a chance to run in the presence of
-		 * errors.  FIXME: determine if due to card removal and
-		 * possibly exit this thread if so.
+		 * errors.
 		 */
-		if (ret < 0)
-			ssleep(1);
+		if (ret < 0) {
+			set_current_state(TASK_INTERRUPTIBLE);
+			if (!kthread_should_stop())
+				schedule_timeout(HZ);
+			set_current_state(TASK_RUNNING);
+		}
 
 		/*
 		 * Adaptive polling frequency based on the assumption
@@ -154,7 +159,8 @@ static int sdio_card_irq_get(struct mmc_card *card)
 	if (!host->sdio_irqs++) {
 		atomic_set(&host->sdio_irq_thread_abort, 0);
 		host->sdio_irq_thread =
-			kthread_run(sdio_irq_thread, host, "ksdiorqd");
+			kthread_run(sdio_irq_thread, host, "ksdioirqd/%s",
+				mmc_hostname(host));
 		if (IS_ERR(host->sdio_irq_thread)) {
 			int err = PTR_ERR(host->sdio_irq_thread);
 			host->sdio_irqs--;
diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
index ea8d7a3490d9..dfa585f7feaf 100644
--- a/drivers/mmc/host/Kconfig
+++ b/drivers/mmc/host/Kconfig
@@ -2,7 +2,7 @@
 # MMC/SD host controller drivers
 #
 
-comment "MMC/SD Host Controller Drivers"
+comment "MMC/SD/SDIO Host Controller Drivers"
 
 config MMC_ARMMMCI
 	tristate "ARM AMBA Multimedia Card Interface support"
@@ -114,6 +114,17 @@ config MMC_ATMELMCI
 
 	  If unsure, say N.
 
+config MMC_ATMELMCI_DMA
+	bool "Atmel MCI DMA support (EXPERIMENTAL)"
+	depends on MMC_ATMELMCI && DMA_ENGINE && EXPERIMENTAL
+	help
+	  Say Y here to have the Atmel MCI driver use a DMA engine to
+	  do data transfers and thus increase the throughput and
+	  reduce the CPU utilization. Note that this is highly
+	  experimental and may cause the driver to lock up.
+
+	  If unsure, say N.
+
 config MMC_IMX
 	tristate "Motorola i.MX Multimedia Card Interface support"
 	depends on ARCH_IMX
@@ -141,21 +152,22 @@ config MMC_TIFM_SD
 	  module will be called tifm_sd.
 
 config MMC_SPI
-	tristate "MMC/SD over SPI"
-	depends on MMC && SPI_MASTER && !HIGHMEM && HAS_DMA
+	tristate "MMC/SD/SDIO over SPI"
+	depends on SPI_MASTER && !HIGHMEM && HAS_DMA
 	select CRC7
 	select CRC_ITU_T
 	help
-	  Some systems accss MMC/SD cards using a SPI controller instead of
-	  using a "native" MMC/SD controller.  This has a disadvantage of
-	  being relatively high overhead, but a compensating advantage of
-	  working on many systems without dedicated MMC/SD controllers.
+	  Some systems accss MMC/SD/SDIO cards using a SPI controller
+	  instead of using a "native" MMC/SD/SDIO controller.  This has a
+	  disadvantage of being relatively high overhead, but a compensating
+	  advantage of working on many systems without dedicated MMC/SD/SDIO
+	  controllers.
 
 	  If unsure, or if your system has no SPI master driver, say N.
 
 config MMC_S3C
 	tristate "Samsung S3C SD/MMC Card Interface support"
-	depends on ARCH_S3C2410 && MMC
+	depends on ARCH_S3C2410
 	help
 	  This selects a driver for the MCI interface found in
           Samsung's S3C2410, S3C2412, S3C2440, S3C2442 CPUs.
@@ -166,7 +178,7 @@ config MMC_S3C
 
 config MMC_SDRICOH_CS
 	tristate "MMC/SD driver for Ricoh Bay1Controllers (EXPERIMENTAL)"
-	depends on EXPERIMENTAL && MMC && PCI && PCMCIA
+	depends on EXPERIMENTAL && PCI && PCMCIA
 	help
 	  Say Y here if your Notebook reports a Ricoh Bay1Controller PCMCIA
 	  card whenever you insert a MMC or SD card into the card slot.
diff --git a/drivers/mmc/host/atmel-mci-regs.h b/drivers/mmc/host/atmel-mci-regs.h
index 26bd80e65031..b58364ed6bba 100644
--- a/drivers/mmc/host/atmel-mci-regs.h
+++ b/drivers/mmc/host/atmel-mci-regs.h
@@ -25,8 +25,10 @@
 #define MCI_SDCR		0x000c	/* SD Card / SDIO */
 # define MCI_SDCSEL_SLOT_A	(  0 <<  0)	/* Select SD slot A */
 # define MCI_SDCSEL_SLOT_B	(  1 <<  0)	/* Select SD slot A */
-# define MCI_SDCBUS_1BIT	(  0 <<  7)	/* 1-bit data bus */
-# define MCI_SDCBUS_4BIT	(  1 <<  7)	/* 4-bit data bus */
+# define MCI_SDCSEL_MASK	(  3 <<  0)
+# define MCI_SDCBUS_1BIT	(  0 <<  6)	/* 1-bit data bus */
+# define MCI_SDCBUS_4BIT	(  2 <<  6)	/* 4-bit data bus */
+# define MCI_SDCBUS_MASK	(  3 <<  6)
 #define MCI_ARGR		0x0010	/* Command Argument */
 #define MCI_CMDR		0x0014	/* Command */
 # define MCI_CMDR_CMDNB(x)	((x) <<  0)	/* Command Opcode */
diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c
index 00008967ef7a..7a3f2436b011 100644
--- a/drivers/mmc/host/atmel-mci.c
+++ b/drivers/mmc/host/atmel-mci.c
@@ -11,6 +11,8 @@
 #include <linux/clk.h>
 #include <linux/debugfs.h>
 #include <linux/device.h>
+#include <linux/dmaengine.h>
+#include <linux/dma-mapping.h>
 #include <linux/err.h>
 #include <linux/gpio.h>
 #include <linux/init.h>
@@ -33,64 +35,178 @@
 #include "atmel-mci-regs.h"
 
 #define ATMCI_DATA_ERROR_FLAGS	(MCI_DCRCE | MCI_DTOE | MCI_OVRE | MCI_UNRE)
+#define ATMCI_DMA_THRESHOLD	16
 
 enum {
 	EVENT_CMD_COMPLETE = 0,
-	EVENT_DATA_ERROR,
-	EVENT_DATA_COMPLETE,
-	EVENT_STOP_SENT,
-	EVENT_STOP_COMPLETE,
 	EVENT_XFER_COMPLETE,
+	EVENT_DATA_COMPLETE,
+	EVENT_DATA_ERROR,
+};
+
+enum atmel_mci_state {
+	STATE_IDLE = 0,
+	STATE_SENDING_CMD,
+	STATE_SENDING_DATA,
+	STATE_DATA_BUSY,
+	STATE_SENDING_STOP,
+	STATE_DATA_ERROR,
+};
+
+struct atmel_mci_dma {
+#ifdef CONFIG_MMC_ATMELMCI_DMA
+	struct dma_client		client;
+	struct dma_chan			*chan;
+	struct dma_async_tx_descriptor	*data_desc;
+#endif
 };
 
+/**
+ * struct atmel_mci - MMC controller state shared between all slots
+ * @lock: Spinlock protecting the queue and associated data.
+ * @regs: Pointer to MMIO registers.
+ * @sg: Scatterlist entry currently being processed by PIO code, if any.
+ * @pio_offset: Offset into the current scatterlist entry.
+ * @cur_slot: The slot which is currently using the controller.
+ * @mrq: The request currently being processed on @cur_slot,
+ *	or NULL if the controller is idle.
+ * @cmd: The command currently being sent to the card, or NULL.
+ * @data: The data currently being transferred, or NULL if no data
+ *	transfer is in progress.
+ * @dma: DMA client state.
+ * @data_chan: DMA channel being used for the current data transfer.
+ * @cmd_status: Snapshot of SR taken upon completion of the current
+ *	command. Only valid when EVENT_CMD_COMPLETE is pending.
+ * @data_status: Snapshot of SR taken upon completion of the current
+ *	data transfer. Only valid when EVENT_DATA_COMPLETE or
+ *	EVENT_DATA_ERROR is pending.
+ * @stop_cmdr: Value to be loaded into CMDR when the stop command is
+ *	to be sent.
+ * @tasklet: Tasklet running the request state machine.
+ * @pending_events: Bitmask of events flagged by the interrupt handler
+ *	to be processed by the tasklet.
+ * @completed_events: Bitmask of events which the state machine has
+ *	processed.
+ * @state: Tasklet state.
+ * @queue: List of slots waiting for access to the controller.
+ * @need_clock_update: Update the clock rate before the next request.
+ * @need_reset: Reset controller before next request.
+ * @mode_reg: Value of the MR register.
+ * @bus_hz: The rate of @mck in Hz. This forms the basis for MMC bus
+ *	rate and timeout calculations.
+ * @mapbase: Physical address of the MMIO registers.
+ * @mck: The peripheral bus clock hooked up to the MMC controller.
+ * @pdev: Platform device associated with the MMC controller.
+ * @slot: Slots sharing this MMC controller.
+ *
+ * Locking
+ * =======
+ *
+ * @lock is a softirq-safe spinlock protecting @queue as well as
+ * @cur_slot, @mrq and @state. These must always be updated
+ * at the same time while holding @lock.
+ *
+ * @lock also protects mode_reg and need_clock_update since these are
+ * used to synchronize mode register updates with the queue
+ * processing.
+ *
+ * The @mrq field of struct atmel_mci_slot is also protected by @lock,
+ * and must always be written at the same time as the slot is added to
+ * @queue.
+ *
+ * @pending_events and @completed_events are accessed using atomic bit
+ * operations, so they don't need any locking.
+ *
+ * None of the fields touched by the interrupt handler need any
+ * locking. However, ordering is important: Before EVENT_DATA_ERROR or
+ * EVENT_DATA_COMPLETE is set in @pending_events, all data-related
+ * interrupts must be disabled and @data_status updated with a
+ * snapshot of SR. Similarly, before EVENT_CMD_COMPLETE is set, the
+ * CMDRDY interupt must be disabled and @cmd_status updated with a
+ * snapshot of SR, and before EVENT_XFER_COMPLETE can be set, the
+ * bytes_xfered field of @data must be written. This is ensured by
+ * using barriers.
+ */
 struct atmel_mci {
-	struct mmc_host		*mmc;
+	spinlock_t		lock;
 	void __iomem		*regs;
 
 	struct scatterlist	*sg;
 	unsigned int		pio_offset;
 
+	struct atmel_mci_slot	*cur_slot;
 	struct mmc_request	*mrq;
 	struct mmc_command	*cmd;
 	struct mmc_data		*data;
 
+	struct atmel_mci_dma	dma;
+	struct dma_chan		*data_chan;
+
 	u32			cmd_status;
 	u32			data_status;
-	u32			stop_status;
 	u32			stop_cmdr;
 
-	u32			mode_reg;
-	u32			sdc_reg;
-
 	struct tasklet_struct	tasklet;
 	unsigned long		pending_events;
 	unsigned long		completed_events;
+	enum atmel_mci_state	state;
+	struct list_head	queue;
 
-	int			present;
-	int			detect_pin;
-	int			wp_pin;
-
-	/* For detect pin debouncing */
-	struct timer_list	detect_timer;
-
+	bool			need_clock_update;
+	bool			need_reset;
+	u32			mode_reg;
 	unsigned long		bus_hz;
 	unsigned long		mapbase;
 	struct clk		*mck;
 	struct platform_device	*pdev;
+
+	struct atmel_mci_slot	*slot[ATMEL_MCI_MAX_NR_SLOTS];
+};
+
+/**
+ * struct atmel_mci_slot - MMC slot state
+ * @mmc: The mmc_host representing this slot.
+ * @host: The MMC controller this slot is using.
+ * @sdc_reg: Value of SDCR to be written before using this slot.
+ * @mrq: mmc_request currently being processed or waiting to be
+ *	processed, or NULL when the slot is idle.
+ * @queue_node: List node for placing this node in the @queue list of
+ *	&struct atmel_mci.
+ * @clock: Clock rate configured by set_ios(). Protected by host->lock.
+ * @flags: Random state bits associated with the slot.
+ * @detect_pin: GPIO pin used for card detection, or negative if not
+ *	available.
+ * @wp_pin: GPIO pin used for card write protect sending, or negative
+ *	if not available.
+ * @detect_timer: Timer used for debouncing @detect_pin interrupts.
+ */
+struct atmel_mci_slot {
+	struct mmc_host		*mmc;
+	struct atmel_mci	*host;
+
+	u32			sdc_reg;
+
+	struct mmc_request	*mrq;
+	struct list_head	queue_node;
+
+	unsigned int		clock;
+	unsigned long		flags;
+#define ATMCI_CARD_PRESENT	0
+#define ATMCI_CARD_NEED_INIT	1
+#define ATMCI_SHUTDOWN		2
+
+	int			detect_pin;
+	int			wp_pin;
+
+	struct timer_list	detect_timer;
 };
 
-#define atmci_is_completed(host, event)				\
-	test_bit(event, &host->completed_events)
 #define atmci_test_and_clear_pending(host, event)		\
 	test_and_clear_bit(event, &host->pending_events)
-#define atmci_test_and_set_completed(host, event)		\
-	test_and_set_bit(event, &host->completed_events)
 #define atmci_set_completed(host, event)			\
 	set_bit(event, &host->completed_events)
 #define atmci_set_pending(host, event)				\
 	set_bit(event, &host->pending_events)
-#define atmci_clear_pending(host, event)			\
-	clear_bit(event, &host->pending_events)
 
 /*
  * The debugfs stuff below is mostly optimized away when
@@ -98,14 +214,15 @@ struct atmel_mci {
  */
 static int atmci_req_show(struct seq_file *s, void *v)
 {
-	struct atmel_mci	*host = s->private;
-	struct mmc_request	*mrq = host->mrq;
+	struct atmel_mci_slot	*slot = s->private;
+	struct mmc_request	*mrq;
 	struct mmc_command	*cmd;
 	struct mmc_command	*stop;
 	struct mmc_data		*data;
 
 	/* Make sure we get a consistent snapshot */
-	spin_lock_irq(&host->mmc->lock);
+	spin_lock_bh(&slot->host->lock);
+	mrq = slot->mrq;
 
 	if (mrq) {
 		cmd = mrq->cmd;
@@ -130,7 +247,7 @@ static int atmci_req_show(struct seq_file *s, void *v)
 				stop->resp[2], stop->error);
 	}
 
-	spin_unlock_irq(&host->mmc->lock);
+	spin_unlock_bh(&slot->host->lock);
 
 	return 0;
 }
@@ -193,12 +310,16 @@ static int atmci_regs_show(struct seq_file *s, void *v)
 	if (!buf)
 		return -ENOMEM;
 
-	/* Grab a more or less consistent snapshot */
-	spin_lock_irq(&host->mmc->lock);
+	/*
+	 * Grab a more or less consistent snapshot. Note that we're
+	 * not disabling interrupts, so IMR and SR may not be
+	 * consistent.
+	 */
+	spin_lock_bh(&host->lock);
 	clk_enable(host->mck);
 	memcpy_fromio(buf, host->regs, MCI_REGS_SIZE);
 	clk_disable(host->mck);
-	spin_unlock_irq(&host->mmc->lock);
+	spin_unlock_bh(&host->lock);
 
 	seq_printf(s, "MR:\t0x%08x%s%s CLKDIV=%u\n",
 			buf[MCI_MR / 4],
@@ -236,13 +357,13 @@ static const struct file_operations atmci_regs_fops = {
 	.release	= single_release,
 };
 
-static void atmci_init_debugfs(struct atmel_mci *host)
+static void atmci_init_debugfs(struct atmel_mci_slot *slot)
 {
-	struct mmc_host	*mmc;
-	struct dentry	*root;
-	struct dentry	*node;
+	struct mmc_host		*mmc = slot->mmc;
+	struct atmel_mci	*host = slot->host;
+	struct dentry		*root;
+	struct dentry		*node;
 
-	mmc = host->mmc;
 	root = mmc->debugfs_root;
 	if (!root)
 		return;
@@ -254,7 +375,11 @@ static void atmci_init_debugfs(struct atmel_mci *host)
 	if (!node)
 		goto err;
 
-	node = debugfs_create_file("req", S_IRUSR, root, host, &atmci_req_fops);
+	node = debugfs_create_file("req", S_IRUSR, root, slot, &atmci_req_fops);
+	if (!node)
+		goto err;
+
+	node = debugfs_create_u32("state", S_IRUSR, root, (u32 *)&host->state);
 	if (!node)
 		goto err;
 
@@ -271,25 +396,7 @@ static void atmci_init_debugfs(struct atmel_mci *host)
 	return;
 
 err:
-	dev_err(&host->pdev->dev,
-		"failed to initialize debugfs for controller\n");
-}
-
-static void atmci_enable(struct atmel_mci *host)
-{
-	clk_enable(host->mck);
-	mci_writel(host, CR, MCI_CR_MCIEN);
-	mci_writel(host, MR, host->mode_reg);
-	mci_writel(host, SDCR, host->sdc_reg);
-}
-
-static void atmci_disable(struct atmel_mci *host)
-{
-	mci_writel(host, CR, MCI_CR_SWRST);
-
-	/* Stall until write is complete, then disable the bus clock */
-	mci_readl(host, SR);
-	clk_disable(host->mck);
+	dev_err(&mmc->class_dev, "failed to initialize debugfs for slot\n");
 }
 
 static inline unsigned int ns_to_clocks(struct atmel_mci *host,
@@ -299,7 +406,7 @@ static inline unsigned int ns_to_clocks(struct atmel_mci *host,
 }
 
 static void atmci_set_timeout(struct atmel_mci *host,
-			      struct mmc_data *data)
+		struct atmel_mci_slot *slot, struct mmc_data *data)
 {
 	static unsigned	dtomul_to_shift[] = {
 		0, 4, 7, 8, 10, 12, 16, 20
@@ -322,7 +429,7 @@ static void atmci_set_timeout(struct atmel_mci *host,
 		dtocyc = 15;
 	}
 
-	dev_vdbg(&host->mmc->class_dev, "setting timeout to %u cycles\n",
+	dev_vdbg(&slot->mmc->class_dev, "setting timeout to %u cycles\n",
 			dtocyc << dtomul_to_shift[dtomul]);
 	mci_writel(host, DTOR, (MCI_DTOMUL(dtomul) | MCI_DTOCYC(dtocyc)));
 }
@@ -375,15 +482,12 @@ static u32 atmci_prepare_command(struct mmc_host *mmc,
 }
 
 static void atmci_start_command(struct atmel_mci *host,
-				struct mmc_command *cmd,
-				u32 cmd_flags)
+		struct mmc_command *cmd, u32 cmd_flags)
 {
-	/* Must read host->cmd after testing event flags */
-	smp_rmb();
 	WARN_ON(host->cmd);
 	host->cmd = cmd;
 
-	dev_vdbg(&host->mmc->class_dev,
+	dev_vdbg(&host->pdev->dev,
 			"start command: ARGR=0x%08x CMDR=0x%08x\n",
 			cmd->arg, cmd_flags);
 
@@ -391,34 +495,157 @@ static void atmci_start_command(struct atmel_mci *host,
 	mci_writel(host, CMDR, cmd_flags);
 }
 
-static void send_stop_cmd(struct mmc_host *mmc, struct mmc_data *data)
+static void send_stop_cmd(struct atmel_mci *host, struct mmc_data *data)
 {
-	struct atmel_mci *host = mmc_priv(mmc);
-
 	atmci_start_command(host, data->stop, host->stop_cmdr);
 	mci_writel(host, IER, MCI_CMDRDY);
 }
 
-static void atmci_request_end(struct mmc_host *mmc, struct mmc_request *mrq)
+#ifdef CONFIG_MMC_ATMELMCI_DMA
+static void atmci_dma_cleanup(struct atmel_mci *host)
 {
-	struct atmel_mci *host = mmc_priv(mmc);
+	struct mmc_data			*data = host->data;
 
-	WARN_ON(host->cmd || host->data);
-	host->mrq = NULL;
+	dma_unmap_sg(&host->pdev->dev, data->sg, data->sg_len,
+		     ((data->flags & MMC_DATA_WRITE)
+		      ? DMA_TO_DEVICE : DMA_FROM_DEVICE));
+}
+
+static void atmci_stop_dma(struct atmel_mci *host)
+{
+	struct dma_chan *chan = host->data_chan;
+
+	if (chan) {
+		chan->device->device_terminate_all(chan);
+		atmci_dma_cleanup(host);
+	} else {
+		/* Data transfer was stopped by the interrupt handler */
+		atmci_set_pending(host, EVENT_XFER_COMPLETE);
+		mci_writel(host, IER, MCI_NOTBUSY);
+	}
+}
+
+/* This function is called by the DMA driver from tasklet context. */
+static void atmci_dma_complete(void *arg)
+{
+	struct atmel_mci	*host = arg;
+	struct mmc_data		*data = host->data;
+
+	dev_vdbg(&host->pdev->dev, "DMA complete\n");
+
+	atmci_dma_cleanup(host);
+
+	/*
+	 * If the card was removed, data will be NULL. No point trying
+	 * to send the stop command or waiting for NBUSY in this case.
+	 */
+	if (data) {
+		atmci_set_pending(host, EVENT_XFER_COMPLETE);
+		tasklet_schedule(&host->tasklet);
+
+		/*
+		 * Regardless of what the documentation says, we have
+		 * to wait for NOTBUSY even after block read
+		 * operations.
+		 *
+		 * When the DMA transfer is complete, the controller
+		 * may still be reading the CRC from the card, i.e.
+		 * the data transfer is still in progress and we
+		 * haven't seen all the potential error bits yet.
+		 *
+		 * The interrupt handler will schedule a different
+		 * tasklet to finish things up when the data transfer
+		 * is completely done.
+		 *
+		 * We may not complete the mmc request here anyway
+		 * because the mmc layer may call back and cause us to
+		 * violate the "don't submit new operations from the
+		 * completion callback" rule of the dma engine
+		 * framework.
+		 */
+		mci_writel(host, IER, MCI_NOTBUSY);
+	}
+}
+
+static int
+atmci_submit_data_dma(struct atmel_mci *host, struct mmc_data *data)
+{
+	struct dma_chan			*chan;
+	struct dma_async_tx_descriptor	*desc;
+	struct scatterlist		*sg;
+	unsigned int			i;
+	enum dma_data_direction		direction;
+
+	/*
+	 * We don't do DMA on "complex" transfers, i.e. with
+	 * non-word-aligned buffers or lengths. Also, we don't bother
+	 * with all the DMA setup overhead for short transfers.
+	 */
+	if (data->blocks * data->blksz < ATMCI_DMA_THRESHOLD)
+		return -EINVAL;
+	if (data->blksz & 3)
+		return -EINVAL;
+
+	for_each_sg(data->sg, sg, data->sg_len, i) {
+		if (sg->offset & 3 || sg->length & 3)
+			return -EINVAL;
+	}
+
+	/* If we don't have a channel, we can't do DMA */
+	chan = host->dma.chan;
+	if (chan) {
+		dma_chan_get(chan);
+		host->data_chan = chan;
+	}
+
+	if (!chan)
+		return -ENODEV;
+
+	if (data->flags & MMC_DATA_READ)
+		direction = DMA_FROM_DEVICE;
+	else
+		direction = DMA_TO_DEVICE;
+
+	desc = chan->device->device_prep_slave_sg(chan,
+			data->sg, data->sg_len, direction,
+			DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!desc)
+		return -ENOMEM;
 
-	atmci_disable(host);
+	host->dma.data_desc = desc;
+	desc->callback = atmci_dma_complete;
+	desc->callback_param = host;
+	desc->tx_submit(desc);
 
-	mmc_request_done(mmc, mrq);
+	/* Go! */
+	chan->device->device_issue_pending(chan);
+
+	return 0;
+}
+
+#else /* CONFIG_MMC_ATMELMCI_DMA */
+
+static int atmci_submit_data_dma(struct atmel_mci *host, struct mmc_data *data)
+{
+	return -ENOSYS;
 }
 
+static void atmci_stop_dma(struct atmel_mci *host)
+{
+	/* Data transfer was stopped by the interrupt handler */
+	atmci_set_pending(host, EVENT_XFER_COMPLETE);
+	mci_writel(host, IER, MCI_NOTBUSY);
+}
+
+#endif /* CONFIG_MMC_ATMELMCI_DMA */
+
 /*
  * Returns a mask of interrupt flags to be enabled after the whole
  * request has been prepared.
  */
-static u32 atmci_submit_data(struct mmc_host *mmc, struct mmc_data *data)
+static u32 atmci_submit_data(struct atmel_mci *host, struct mmc_data *data)
 {
-	struct atmel_mci	*host = mmc_priv(mmc);
-	u32			iflags;
+	u32 iflags;
 
 	data->error = -EINPROGRESS;
 
@@ -426,77 +653,89 @@ static u32 atmci_submit_data(struct mmc_host *mmc, struct mmc_data *data)
 	host->sg = NULL;
 	host->data = data;
 
-	dev_vdbg(&mmc->class_dev, "BLKR=0x%08x\n",
-			MCI_BCNT(data->blocks) | MCI_BLKLEN(data->blksz));
-
 	iflags = ATMCI_DATA_ERROR_FLAGS;
-	host->sg = data->sg;
-	host->pio_offset = 0;
-	if (data->flags & MMC_DATA_READ)
-		iflags |= MCI_RXRDY;
-	else
-		iflags |= MCI_TXRDY;
+	if (atmci_submit_data_dma(host, data)) {
+		host->data_chan = NULL;
+
+		/*
+		 * Errata: MMC data write operation with less than 12
+		 * bytes is impossible.
+		 *
+		 * Errata: MCI Transmit Data Register (TDR) FIFO
+		 * corruption when length is not multiple of 4.
+		 */
+		if (data->blocks * data->blksz < 12
+				|| (data->blocks * data->blksz) & 3)
+			host->need_reset = true;
+
+		host->sg = data->sg;
+		host->pio_offset = 0;
+		if (data->flags & MMC_DATA_READ)
+			iflags |= MCI_RXRDY;
+		else
+			iflags |= MCI_TXRDY;
+	}
 
 	return iflags;
 }
 
-static void atmci_request(struct mmc_host *mmc, struct mmc_request *mrq)
+static void atmci_start_request(struct atmel_mci *host,
+		struct atmel_mci_slot *slot)
 {
-	struct atmel_mci	*host = mmc_priv(mmc);
-	struct mmc_data		*data;
+	struct mmc_request	*mrq;
 	struct mmc_command	*cmd;
+	struct mmc_data		*data;
 	u32			iflags;
-	u32			cmdflags = 0;
-
-	iflags = mci_readl(host, IMR);
-	if (iflags)
-		dev_warn(&mmc->class_dev, "WARNING: IMR=0x%08x\n",
-				mci_readl(host, IMR));
-
-	WARN_ON(host->mrq != NULL);
-
-	/*
-	 * We may "know" the card is gone even though there's still an
-	 * electrical connection. If so, we really need to communicate
-	 * this to the MMC core since there won't be any more
-	 * interrupts as the card is completely removed. Otherwise,
-	 * the MMC core might believe the card is still there even
-	 * though the card was just removed very slowly.
-	 */
-	if (!host->present) {
-		mrq->cmd->error = -ENOMEDIUM;
-		mmc_request_done(mmc, mrq);
-		return;
-	}
+	u32			cmdflags;
 
+	mrq = slot->mrq;
+	host->cur_slot = slot;
 	host->mrq = mrq;
+
 	host->pending_events = 0;
 	host->completed_events = 0;
+	host->data_status = 0;
 
-	atmci_enable(host);
+	if (host->need_reset) {
+		mci_writel(host, CR, MCI_CR_SWRST);
+		mci_writel(host, CR, MCI_CR_MCIEN);
+		mci_writel(host, MR, host->mode_reg);
+		host->need_reset = false;
+	}
+	mci_writel(host, SDCR, slot->sdc_reg);
 
-	/* We don't support multiple blocks of weird lengths. */
+	iflags = mci_readl(host, IMR);
+	if (iflags)
+		dev_warn(&slot->mmc->class_dev, "WARNING: IMR=0x%08x\n",
+				iflags);
+
+	if (unlikely(test_and_clear_bit(ATMCI_CARD_NEED_INIT, &slot->flags))) {
+		/* Send init sequence (74 clock cycles) */
+		mci_writel(host, CMDR, MCI_CMDR_SPCMD_INIT);
+		while (!(mci_readl(host, SR) & MCI_CMDRDY))
+			cpu_relax();
+	}
 	data = mrq->data;
 	if (data) {
-		if (data->blocks > 1 && data->blksz & 3)
-			goto fail;
-		atmci_set_timeout(host, data);
+		atmci_set_timeout(host, slot, data);
 
 		/* Must set block count/size before sending command */
 		mci_writel(host, BLKR, MCI_BCNT(data->blocks)
 				| MCI_BLKLEN(data->blksz));
+		dev_vdbg(&slot->mmc->class_dev, "BLKR=0x%08x\n",
+			MCI_BCNT(data->blocks) | MCI_BLKLEN(data->blksz));
 	}
 
 	iflags = MCI_CMDRDY;
 	cmd = mrq->cmd;
-	cmdflags = atmci_prepare_command(mmc, cmd);
+	cmdflags = atmci_prepare_command(slot->mmc, cmd);
 	atmci_start_command(host, cmd, cmdflags);
 
 	if (data)
-		iflags |= atmci_submit_data(mmc, data);
+		iflags |= atmci_submit_data(host, data);
 
 	if (mrq->stop) {
-		host->stop_cmdr = atmci_prepare_command(mmc, mrq->stop);
+		host->stop_cmdr = atmci_prepare_command(slot->mmc, mrq->stop);
 		host->stop_cmdr |= MCI_CMDR_STOP_XFER;
 		if (!(data->flags & MMC_DATA_WRITE))
 			host->stop_cmdr |= MCI_CMDR_TRDIR_READ;
@@ -513,59 +752,156 @@ static void atmci_request(struct mmc_host *mmc, struct mmc_request *mrq)
 	 * prepared yet.)
 	 */
 	mci_writel(host, IER, iflags);
+}
 
-	return;
+static void atmci_queue_request(struct atmel_mci *host,
+		struct atmel_mci_slot *slot, struct mmc_request *mrq)
+{
+	dev_vdbg(&slot->mmc->class_dev, "queue request: state=%d\n",
+			host->state);
+
+	spin_lock_bh(&host->lock);
+	slot->mrq = mrq;
+	if (host->state == STATE_IDLE) {
+		host->state = STATE_SENDING_CMD;
+		atmci_start_request(host, slot);
+	} else {
+		list_add_tail(&slot->queue_node, &host->queue);
+	}
+	spin_unlock_bh(&host->lock);
+}
 
-fail:
-	atmci_disable(host);
-	host->mrq = NULL;
-	mrq->cmd->error = -EINVAL;
-	mmc_request_done(mmc, mrq);
+static void atmci_request(struct mmc_host *mmc, struct mmc_request *mrq)
+{
+	struct atmel_mci_slot	*slot = mmc_priv(mmc);
+	struct atmel_mci	*host = slot->host;
+	struct mmc_data		*data;
+
+	WARN_ON(slot->mrq);
+
+	/*
+	 * We may "know" the card is gone even though there's still an
+	 * electrical connection. If so, we really need to communicate
+	 * this to the MMC core since there won't be any more
+	 * interrupts as the card is completely removed. Otherwise,
+	 * the MMC core might believe the card is still there even
+	 * though the card was just removed very slowly.
+	 */
+	if (!test_bit(ATMCI_CARD_PRESENT, &slot->flags)) {
+		mrq->cmd->error = -ENOMEDIUM;
+		mmc_request_done(mmc, mrq);
+		return;
+	}
+
+	/* We don't support multiple blocks of weird lengths. */
+	data = mrq->data;
+	if (data && data->blocks > 1 && data->blksz & 3) {
+		mrq->cmd->error = -EINVAL;
+		mmc_request_done(mmc, mrq);
+	}
+
+	atmci_queue_request(host, slot, mrq);
 }
 
 static void atmci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 {
-	struct atmel_mci	*host = mmc_priv(mmc);
+	struct atmel_mci_slot	*slot = mmc_priv(mmc);
+	struct atmel_mci	*host = slot->host;
+	unsigned int		i;
+
+	slot->sdc_reg &= ~MCI_SDCBUS_MASK;
+	switch (ios->bus_width) {
+	case MMC_BUS_WIDTH_1:
+		slot->sdc_reg |= MCI_SDCBUS_1BIT;
+		break;
+	case MMC_BUS_WIDTH_4:
+		slot->sdc_reg = MCI_SDCBUS_4BIT;
+		break;
+	}
 
 	if (ios->clock) {
+		unsigned int clock_min = ~0U;
 		u32 clkdiv;
 
-		/* Set clock rate */
-		clkdiv = DIV_ROUND_UP(host->bus_hz, 2 * ios->clock) - 1;
+		spin_lock_bh(&host->lock);
+		if (!host->mode_reg) {
+			clk_enable(host->mck);
+			mci_writel(host, CR, MCI_CR_SWRST);
+			mci_writel(host, CR, MCI_CR_MCIEN);
+		}
+
+		/*
+		 * Use mirror of ios->clock to prevent race with mmc
+		 * core ios update when finding the minimum.
+		 */
+		slot->clock = ios->clock;
+		for (i = 0; i < ATMEL_MCI_MAX_NR_SLOTS; i++) {
+			if (host->slot[i] && host->slot[i]->clock
+					&& host->slot[i]->clock < clock_min)
+				clock_min = host->slot[i]->clock;
+		}
+
+		/* Calculate clock divider */
+		clkdiv = DIV_ROUND_UP(host->bus_hz, 2 * clock_min) - 1;
 		if (clkdiv > 255) {
 			dev_warn(&mmc->class_dev,
 				"clock %u too slow; using %lu\n",
-				ios->clock, host->bus_hz / (2 * 256));
+				clock_min, host->bus_hz / (2 * 256));
 			clkdiv = 255;
 		}
 
+		/*
+		 * WRPROOF and RDPROOF prevent overruns/underruns by
+		 * stopping the clock when the FIFO is full/empty.
+		 * This state is not expected to last for long.
+		 */
 		host->mode_reg = MCI_MR_CLKDIV(clkdiv) | MCI_MR_WRPROOF
 					| MCI_MR_RDPROOF;
-	}
 
-	switch (ios->bus_width) {
-	case MMC_BUS_WIDTH_1:
-		host->sdc_reg = 0;
-		break;
-	case MMC_BUS_WIDTH_4:
-		host->sdc_reg = MCI_SDCBUS_4BIT;
-		break;
+		if (list_empty(&host->queue))
+			mci_writel(host, MR, host->mode_reg);
+		else
+			host->need_clock_update = true;
+
+		spin_unlock_bh(&host->lock);
+	} else {
+		bool any_slot_active = false;
+
+		spin_lock_bh(&host->lock);
+		slot->clock = 0;
+		for (i = 0; i < ATMEL_MCI_MAX_NR_SLOTS; i++) {
+			if (host->slot[i] && host->slot[i]->clock) {
+				any_slot_active = true;
+				break;
+			}
+		}
+		if (!any_slot_active) {
+			mci_writel(host, CR, MCI_CR_MCIDIS);
+			if (host->mode_reg) {
+				mci_readl(host, MR);
+				clk_disable(host->mck);
+			}
+			host->mode_reg = 0;
+		}
+		spin_unlock_bh(&host->lock);
 	}
 
 	switch (ios->power_mode) {
-	case MMC_POWER_ON:
-		/* Send init sequence (74 clock cycles) */
-		atmci_enable(host);
-		mci_writel(host, CMDR, MCI_CMDR_SPCMD_INIT);
-		while (!(mci_readl(host, SR) & MCI_CMDRDY))
-			cpu_relax();
-		atmci_disable(host);
+	case MMC_POWER_UP:
+		set_bit(ATMCI_CARD_NEED_INIT, &slot->flags);
 		break;
 	default:
 		/*
 		 * TODO: None of the currently available AVR32-based
 		 * boards allow MMC power to be turned off. Implement
 		 * power control when this can be tested properly.
+		 *
+		 * We also need to hook this into the clock management
+		 * somehow so that newly inserted cards aren't
+		 * subjected to a fast clock before we have a chance
+		 * to figure out what the maximum rate is. Currently,
+		 * there's no way to avoid this, and there never will
+		 * be for boards that don't support power control.
 		 */
 		break;
 	}
@@ -573,31 +909,82 @@ static void atmci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 
 static int atmci_get_ro(struct mmc_host *mmc)
 {
-	int			read_only = 0;
-	struct atmel_mci	*host = mmc_priv(mmc);
+	int			read_only = -ENOSYS;
+	struct atmel_mci_slot	*slot = mmc_priv(mmc);
 
-	if (gpio_is_valid(host->wp_pin)) {
-		read_only = gpio_get_value(host->wp_pin);
+	if (gpio_is_valid(slot->wp_pin)) {
+		read_only = gpio_get_value(slot->wp_pin);
 		dev_dbg(&mmc->class_dev, "card is %s\n",
 				read_only ? "read-only" : "read-write");
-	} else {
-		dev_dbg(&mmc->class_dev,
-			"no pin for checking read-only switch."
-			" Assuming write-enable.\n");
 	}
 
 	return read_only;
 }
 
-static struct mmc_host_ops atmci_ops = {
+static int atmci_get_cd(struct mmc_host *mmc)
+{
+	int			present = -ENOSYS;
+	struct atmel_mci_slot	*slot = mmc_priv(mmc);
+
+	if (gpio_is_valid(slot->detect_pin)) {
+		present = !gpio_get_value(slot->detect_pin);
+		dev_dbg(&mmc->class_dev, "card is %spresent\n",
+				present ? "" : "not ");
+	}
+
+	return present;
+}
+
+static const struct mmc_host_ops atmci_ops = {
 	.request	= atmci_request,
 	.set_ios	= atmci_set_ios,
 	.get_ro		= atmci_get_ro,
+	.get_cd		= atmci_get_cd,
 };
 
+/* Called with host->lock held */
+static void atmci_request_end(struct atmel_mci *host, struct mmc_request *mrq)
+	__releases(&host->lock)
+	__acquires(&host->lock)
+{
+	struct atmel_mci_slot	*slot = NULL;
+	struct mmc_host		*prev_mmc = host->cur_slot->mmc;
+
+	WARN_ON(host->cmd || host->data);
+
+	/*
+	 * Update the MMC clock rate if necessary. This may be
+	 * necessary if set_ios() is called when a different slot is
+	 * busy transfering data.
+	 */
+	if (host->need_clock_update)
+		mci_writel(host, MR, host->mode_reg);
+
+	host->cur_slot->mrq = NULL;
+	host->mrq = NULL;
+	if (!list_empty(&host->queue)) {
+		slot = list_entry(host->queue.next,
+				struct atmel_mci_slot, queue_node);
+		list_del(&slot->queue_node);
+		dev_vdbg(&host->pdev->dev, "list not empty: %s is next\n",
+				mmc_hostname(slot->mmc));
+		host->state = STATE_SENDING_CMD;
+		atmci_start_request(host, slot);
+	} else {
+		dev_vdbg(&host->pdev->dev, "list empty\n");
+		host->state = STATE_IDLE;
+	}
+
+	spin_unlock(&host->lock);
+	mmc_request_done(prev_mmc, mrq);
+	spin_lock(&host->lock);
+}
+
 static void atmci_command_complete(struct atmel_mci *host,
-			struct mmc_command *cmd, u32 status)
+			struct mmc_command *cmd)
 {
+	u32		status = host->cmd_status;
+
 	/* Read the response from the card (up to 16 bytes) */
 	cmd->resp[0] = mci_readl(host, RSPR);
 	cmd->resp[1] = mci_readl(host, RSPR);
@@ -614,11 +1001,12 @@ static void atmci_command_complete(struct atmel_mci *host,
 		cmd->error = 0;
 
 	if (cmd->error) {
-		dev_dbg(&host->mmc->class_dev,
+		dev_dbg(&host->pdev->dev,
 			"command error: status=0x%08x\n", status);
 
 		if (cmd->data) {
 			host->data = NULL;
+			atmci_stop_dma(host);
 			mci_writel(host, IDR, MCI_NOTBUSY
 					| MCI_TXRDY | MCI_RXRDY
 					| ATMCI_DATA_ERROR_FLAGS);
@@ -628,146 +1016,222 @@ static void atmci_command_complete(struct atmel_mci *host,
 
 static void atmci_detect_change(unsigned long data)
 {
-	struct atmel_mci *host = (struct atmel_mci *)data;
-	struct mmc_request *mrq = host->mrq;
-	int present;
+	struct atmel_mci_slot	*slot = (struct atmel_mci_slot *)data;
+	bool			present;
+	bool			present_old;
 
 	/*
-	 * atmci_remove() sets detect_pin to -1 before freeing the
-	 * interrupt. We must not re-enable the interrupt if it has
-	 * been freed.
+	 * atmci_cleanup_slot() sets the ATMCI_SHUTDOWN flag before
+	 * freeing the interrupt. We must not re-enable the interrupt
+	 * if it has been freed, and if we're shutting down, it
+	 * doesn't really matter whether the card is present or not.
 	 */
 	smp_rmb();
-	if (!gpio_is_valid(host->detect_pin))
+	if (test_bit(ATMCI_SHUTDOWN, &slot->flags))
 		return;
 
-	enable_irq(gpio_to_irq(host->detect_pin));
-	present = !gpio_get_value(host->detect_pin);
+	enable_irq(gpio_to_irq(slot->detect_pin));
+	present = !gpio_get_value(slot->detect_pin);
+	present_old = test_bit(ATMCI_CARD_PRESENT, &slot->flags);
+
+	dev_vdbg(&slot->mmc->class_dev, "detect change: %d (was %d)\n",
+			present, present_old);
 
-	dev_vdbg(&host->pdev->dev, "detect change: %d (was %d)\n",
-			present, host->present);
+	if (present != present_old) {
+		struct atmel_mci	*host = slot->host;
+		struct mmc_request	*mrq;
 
-	if (present != host->present) {
-		dev_dbg(&host->mmc->class_dev, "card %s\n",
+		dev_dbg(&slot->mmc->class_dev, "card %s\n",
 			present ? "inserted" : "removed");
-		host->present = present;
 
-		/* Reset controller if card is gone */
-		if (!present) {
-			mci_writel(host, CR, MCI_CR_SWRST);
-			mci_writel(host, IDR, ~0UL);
-			mci_writel(host, CR, MCI_CR_MCIEN);
-		}
+		spin_lock(&host->lock);
+
+		if (!present)
+			clear_bit(ATMCI_CARD_PRESENT, &slot->flags);
+		else
+			set_bit(ATMCI_CARD_PRESENT, &slot->flags);
 
 		/* Clean up queue if present */
+		mrq = slot->mrq;
 		if (mrq) {
-			/*
-			 * Reset controller to terminate any ongoing
-			 * commands or data transfers.
-			 */
-			mci_writel(host, CR, MCI_CR_SWRST);
+			if (mrq == host->mrq) {
+				/*
+				 * Reset controller to terminate any ongoing
+				 * commands or data transfers.
+				 */
+				mci_writel(host, CR, MCI_CR_SWRST);
+				mci_writel(host, CR, MCI_CR_MCIEN);
+				mci_writel(host, MR, host->mode_reg);
 
-			if (!atmci_is_completed(host, EVENT_CMD_COMPLETE))
-				mrq->cmd->error = -ENOMEDIUM;
-
-			if (mrq->data && !atmci_is_completed(host,
-						EVENT_DATA_COMPLETE)) {
 				host->data = NULL;
-				mrq->data->error = -ENOMEDIUM;
+				host->cmd = NULL;
+
+				switch (host->state) {
+				case STATE_IDLE:
+					break;
+				case STATE_SENDING_CMD:
+					mrq->cmd->error = -ENOMEDIUM;
+					if (!mrq->data)
+						break;
+					/* fall through */
+				case STATE_SENDING_DATA:
+					mrq->data->error = -ENOMEDIUM;
+					atmci_stop_dma(host);
+					break;
+				case STATE_DATA_BUSY:
+				case STATE_DATA_ERROR:
+					if (mrq->data->error == -EINPROGRESS)
+						mrq->data->error = -ENOMEDIUM;
+					if (!mrq->stop)
+						break;
+					/* fall through */
+				case STATE_SENDING_STOP:
+					mrq->stop->error = -ENOMEDIUM;
+					break;
+				}
+
+				atmci_request_end(host, mrq);
+			} else {
+				list_del(&slot->queue_node);
+				mrq->cmd->error = -ENOMEDIUM;
+				if (mrq->data)
+					mrq->data->error = -ENOMEDIUM;
+				if (mrq->stop)
+					mrq->stop->error = -ENOMEDIUM;
+
+				spin_unlock(&host->lock);
+				mmc_request_done(slot->mmc, mrq);
+				spin_lock(&host->lock);
 			}
-			if (mrq->stop && !atmci_is_completed(host,
-						EVENT_STOP_COMPLETE))
-				mrq->stop->error = -ENOMEDIUM;
-
-			host->cmd = NULL;
-			atmci_request_end(host->mmc, mrq);
 		}
+		spin_unlock(&host->lock);
 
-		mmc_detect_change(host->mmc, 0);
+		mmc_detect_change(slot->mmc, 0);
 	}
 }
 
 static void atmci_tasklet_func(unsigned long priv)
 {
-	struct mmc_host		*mmc = (struct mmc_host *)priv;
-	struct atmel_mci	*host = mmc_priv(mmc);
+	struct atmel_mci	*host = (struct atmel_mci *)priv;
 	struct mmc_request	*mrq = host->mrq;
 	struct mmc_data		*data = host->data;
+	struct mmc_command	*cmd = host->cmd;
+	enum atmel_mci_state	state = host->state;
+	enum atmel_mci_state	prev_state;
+	u32			status;
+
+	spin_lock(&host->lock);
 
-	dev_vdbg(&mmc->class_dev,
-		"tasklet: pending/completed/mask %lx/%lx/%x\n",
-		host->pending_events, host->completed_events,
+	state = host->state;
+
+	dev_vdbg(&host->pdev->dev,
+		"tasklet: state %u pending/completed/mask %lx/%lx/%x\n",
+		state, host->pending_events, host->completed_events,
 		mci_readl(host, IMR));
 
-	if (atmci_test_and_clear_pending(host, EVENT_CMD_COMPLETE)) {
-		/*
-		 * host->cmd must be set to NULL before the interrupt
-		 * handler sees EVENT_CMD_COMPLETE
-		 */
-		host->cmd = NULL;
-		smp_wmb();
-		atmci_set_completed(host, EVENT_CMD_COMPLETE);
-		atmci_command_complete(host, mrq->cmd, host->cmd_status);
-
-		if (!mrq->cmd->error && mrq->stop
-				&& atmci_is_completed(host, EVENT_XFER_COMPLETE)
-				&& !atmci_test_and_set_completed(host,
-					EVENT_STOP_SENT))
-			send_stop_cmd(host->mmc, mrq->data);
-	}
-	if (atmci_test_and_clear_pending(host, EVENT_STOP_COMPLETE)) {
-		/*
-		 * host->cmd must be set to NULL before the interrupt
-		 * handler sees EVENT_STOP_COMPLETE
-		 */
-		host->cmd = NULL;
-		smp_wmb();
-		atmci_set_completed(host, EVENT_STOP_COMPLETE);
-		atmci_command_complete(host, mrq->stop, host->stop_status);
-	}
-	if (atmci_test_and_clear_pending(host, EVENT_DATA_ERROR)) {
-		u32 status = host->data_status;
+	do {
+		prev_state = state;
 
-		dev_vdbg(&mmc->class_dev, "data error: status=%08x\n", status);
+		switch (state) {
+		case STATE_IDLE:
+			break;
 
-		atmci_set_completed(host, EVENT_DATA_ERROR);
-		atmci_set_completed(host, EVENT_DATA_COMPLETE);
+		case STATE_SENDING_CMD:
+			if (!atmci_test_and_clear_pending(host,
+						EVENT_CMD_COMPLETE))
+				break;
 
-		if (status & MCI_DTOE) {
-			dev_dbg(&mmc->class_dev,
-					"data timeout error\n");
-			data->error = -ETIMEDOUT;
-		} else if (status & MCI_DCRCE) {
-			dev_dbg(&mmc->class_dev, "data CRC error\n");
-			data->error = -EILSEQ;
-		} else {
-			dev_dbg(&mmc->class_dev,
-					"data FIFO error (status=%08x)\n",
-					status);
-			data->error = -EIO;
-		}
+			host->cmd = NULL;
+			atmci_set_completed(host, EVENT_CMD_COMPLETE);
+			atmci_command_complete(host, mrq->cmd);
+			if (!mrq->data || cmd->error) {
+				atmci_request_end(host, host->mrq);
+				goto unlock;
+			}
+
+			prev_state = state = STATE_SENDING_DATA;
+			/* fall through */
+
+		case STATE_SENDING_DATA:
+			if (atmci_test_and_clear_pending(host,
+						EVENT_DATA_ERROR)) {
+				atmci_stop_dma(host);
+				if (data->stop)
+					send_stop_cmd(host, data);
+				state = STATE_DATA_ERROR;
+				break;
+			}
 
-		if (host->present && data->stop
-				&& atmci_is_completed(host, EVENT_CMD_COMPLETE)
-				&& !atmci_test_and_set_completed(
-					host, EVENT_STOP_SENT))
-			send_stop_cmd(host->mmc, data);
+			if (!atmci_test_and_clear_pending(host,
+						EVENT_XFER_COMPLETE))
+				break;
 
-		host->data = NULL;
-	}
-	if (atmci_test_and_clear_pending(host, EVENT_DATA_COMPLETE)) {
-		atmci_set_completed(host, EVENT_DATA_COMPLETE);
+			atmci_set_completed(host, EVENT_XFER_COMPLETE);
+			prev_state = state = STATE_DATA_BUSY;
+			/* fall through */
+
+		case STATE_DATA_BUSY:
+			if (!atmci_test_and_clear_pending(host,
+						EVENT_DATA_COMPLETE))
+				break;
 
-		if (!atmci_is_completed(host, EVENT_DATA_ERROR)) {
-			data->bytes_xfered = data->blocks * data->blksz;
-			data->error = 0;
+			host->data = NULL;
+			atmci_set_completed(host, EVENT_DATA_COMPLETE);
+			status = host->data_status;
+			if (unlikely(status & ATMCI_DATA_ERROR_FLAGS)) {
+				if (status & MCI_DTOE) {
+					dev_dbg(&host->pdev->dev,
+							"data timeout error\n");
+					data->error = -ETIMEDOUT;
+				} else if (status & MCI_DCRCE) {
+					dev_dbg(&host->pdev->dev,
+							"data CRC error\n");
+					data->error = -EILSEQ;
+				} else {
+					dev_dbg(&host->pdev->dev,
+						"data FIFO error (status=%08x)\n",
+						status);
+					data->error = -EIO;
+				}
+			} else {
+				data->bytes_xfered = data->blocks * data->blksz;
+				data->error = 0;
+			}
+
+			if (!data->stop) {
+				atmci_request_end(host, host->mrq);
+				goto unlock;
+			}
+
+			prev_state = state = STATE_SENDING_STOP;
+			if (!data->error)
+				send_stop_cmd(host, data);
+			/* fall through */
+
+		case STATE_SENDING_STOP:
+			if (!atmci_test_and_clear_pending(host,
+						EVENT_CMD_COMPLETE))
+				break;
+
+			host->cmd = NULL;
+			atmci_command_complete(host, mrq->stop);
+			atmci_request_end(host, host->mrq);
+			goto unlock;
+
+		case STATE_DATA_ERROR:
+			if (!atmci_test_and_clear_pending(host,
+						EVENT_XFER_COMPLETE))
+				break;
+
+			state = STATE_DATA_BUSY;
+			break;
 		}
+	} while (state != prev_state);
 
-		host->data = NULL;
-	}
+	host->state = state;
 
-	if (host->mrq && !host->cmd && !host->data)
-		atmci_request_end(mmc, host->mrq);
+unlock:
+	spin_unlock(&host->lock);
 }
 
 static void atmci_read_data_pio(struct atmel_mci *host)
@@ -789,6 +1253,7 @@ static void atmci_read_data_pio(struct atmel_mci *host)
 			nbytes += 4;
 
 			if (offset == sg->length) {
+				flush_dcache_page(sg_page(sg));
 				host->sg = sg = sg_next(sg);
 				if (!sg)
 					goto done;
@@ -817,9 +1282,11 @@ static void atmci_read_data_pio(struct atmel_mci *host)
 			mci_writel(host, IDR, (MCI_NOTBUSY | MCI_RXRDY
 						| ATMCI_DATA_ERROR_FLAGS));
 			host->data_status = status;
+			data->bytes_xfered += nbytes;
+			smp_wmb();
 			atmci_set_pending(host, EVENT_DATA_ERROR);
 			tasklet_schedule(&host->tasklet);
-			break;
+			return;
 		}
 	} while (status & MCI_RXRDY);
 
@@ -832,10 +1299,8 @@ done:
 	mci_writel(host, IDR, MCI_RXRDY);
 	mci_writel(host, IER, MCI_NOTBUSY);
 	data->bytes_xfered += nbytes;
-	atmci_set_completed(host, EVENT_XFER_COMPLETE);
-	if (data->stop && atmci_is_completed(host, EVENT_CMD_COMPLETE)
-			&& !atmci_test_and_set_completed(host, EVENT_STOP_SENT))
-		send_stop_cmd(host->mmc, data);
+	smp_wmb();
+	atmci_set_pending(host, EVENT_XFER_COMPLETE);
 }
 
 static void atmci_write_data_pio(struct atmel_mci *host)
@@ -888,9 +1353,11 @@ static void atmci_write_data_pio(struct atmel_mci *host)
 			mci_writel(host, IDR, (MCI_NOTBUSY | MCI_TXRDY
 						| ATMCI_DATA_ERROR_FLAGS));
 			host->data_status = status;
+			data->bytes_xfered += nbytes;
+			smp_wmb();
 			atmci_set_pending(host, EVENT_DATA_ERROR);
 			tasklet_schedule(&host->tasklet);
-			break;
+			return;
 		}
 	} while (status & MCI_TXRDY);
 
@@ -903,38 +1370,26 @@ done:
 	mci_writel(host, IDR, MCI_TXRDY);
 	mci_writel(host, IER, MCI_NOTBUSY);
 	data->bytes_xfered += nbytes;
-	atmci_set_completed(host, EVENT_XFER_COMPLETE);
-	if (data->stop && atmci_is_completed(host, EVENT_CMD_COMPLETE)
-			&& !atmci_test_and_set_completed(host, EVENT_STOP_SENT))
-		send_stop_cmd(host->mmc, data);
+	smp_wmb();
+	atmci_set_pending(host, EVENT_XFER_COMPLETE);
 }
 
-static void atmci_cmd_interrupt(struct mmc_host *mmc, u32 status)
+static void atmci_cmd_interrupt(struct atmel_mci *host, u32 status)
 {
-	struct atmel_mci	*host = mmc_priv(mmc);
-
 	mci_writel(host, IDR, MCI_CMDRDY);
 
-	if (atmci_is_completed(host, EVENT_STOP_SENT)) {
-		host->stop_status = status;
-		atmci_set_pending(host, EVENT_STOP_COMPLETE);
-	} else {
-		host->cmd_status = status;
-		atmci_set_pending(host, EVENT_CMD_COMPLETE);
-	}
-
+	host->cmd_status = status;
+	smp_wmb();
+	atmci_set_pending(host, EVENT_CMD_COMPLETE);
 	tasklet_schedule(&host->tasklet);
 }
 
 static irqreturn_t atmci_interrupt(int irq, void *dev_id)
 {
-	struct mmc_host		*mmc = dev_id;
-	struct atmel_mci	*host = mmc_priv(mmc);
+	struct atmel_mci	*host = dev_id;
 	u32			status, mask, pending;
 	unsigned int		pass_count = 0;
 
-	spin_lock(&mmc->lock);
-
 	do {
 		status = mci_readl(host, SR);
 		mask = mci_readl(host, IMR);
@@ -946,13 +1401,18 @@ static irqreturn_t atmci_interrupt(int irq, void *dev_id)
 			mci_writel(host, IDR, ATMCI_DATA_ERROR_FLAGS
 					| MCI_RXRDY | MCI_TXRDY);
 			pending &= mci_readl(host, IMR);
+
 			host->data_status = status;
+			smp_wmb();
 			atmci_set_pending(host, EVENT_DATA_ERROR);
 			tasklet_schedule(&host->tasklet);
 		}
 		if (pending & MCI_NOTBUSY) {
-			mci_writel(host, IDR, (MCI_NOTBUSY
-					       | ATMCI_DATA_ERROR_FLAGS));
+			mci_writel(host, IDR,
+					ATMCI_DATA_ERROR_FLAGS | MCI_NOTBUSY);
+			if (!host->data_status)
+				host->data_status = status;
+			smp_wmb();
 			atmci_set_pending(host, EVENT_DATA_COMPLETE);
 			tasklet_schedule(&host->tasklet);
 		}
@@ -962,18 +1422,15 @@ static irqreturn_t atmci_interrupt(int irq, void *dev_id)
 			atmci_write_data_pio(host);
 
 		if (pending & MCI_CMDRDY)
-			atmci_cmd_interrupt(mmc, status);
+			atmci_cmd_interrupt(host, status);
 	} while (pass_count++ < 5);
 
-	spin_unlock(&mmc->lock);
-
 	return pass_count ? IRQ_HANDLED : IRQ_NONE;
 }
 
 static irqreturn_t atmci_detect_interrupt(int irq, void *dev_id)
 {
-	struct mmc_host		*mmc = dev_id;
-	struct atmel_mci	*host = mmc_priv(mmc);
+	struct atmel_mci_slot	*slot = dev_id;
 
 	/*
 	 * Disable interrupts until the pin has stabilized and check
@@ -981,19 +1438,176 @@ static irqreturn_t atmci_detect_interrupt(int irq, void *dev_id)
 	 * middle of the timer routine when this interrupt triggers.
 	 */
 	disable_irq_nosync(irq);
-	mod_timer(&host->detect_timer, jiffies + msecs_to_jiffies(20));
+	mod_timer(&slot->detect_timer, jiffies + msecs_to_jiffies(20));
 
 	return IRQ_HANDLED;
 }
 
+#ifdef CONFIG_MMC_ATMELMCI_DMA
+
+static inline struct atmel_mci *
+dma_client_to_atmel_mci(struct dma_client *client)
+{
+	return container_of(client, struct atmel_mci, dma.client);
+}
+
+static enum dma_state_client atmci_dma_event(struct dma_client *client,
+		struct dma_chan *chan, enum dma_state state)
+{
+	struct atmel_mci	*host;
+	enum dma_state_client	ret = DMA_NAK;
+
+	host = dma_client_to_atmel_mci(client);
+
+	switch (state) {
+	case DMA_RESOURCE_AVAILABLE:
+		spin_lock_bh(&host->lock);
+		if (!host->dma.chan) {
+			host->dma.chan = chan;
+			ret = DMA_ACK;
+		}
+		spin_unlock_bh(&host->lock);
+
+		if (ret == DMA_ACK)
+			dev_info(&host->pdev->dev,
+					"Using %s for DMA transfers\n",
+					chan->dev.bus_id);
+		break;
+
+	case DMA_RESOURCE_REMOVED:
+		spin_lock_bh(&host->lock);
+		if (host->dma.chan == chan) {
+			host->dma.chan = NULL;
+			ret = DMA_ACK;
+		}
+		spin_unlock_bh(&host->lock);
+
+		if (ret == DMA_ACK)
+			dev_info(&host->pdev->dev,
+					"Lost %s, falling back to PIO\n",
+					chan->dev.bus_id);
+		break;
+
+	default:
+		break;
+	}
+
+
+	return ret;
+}
+#endif /* CONFIG_MMC_ATMELMCI_DMA */
+
+static int __init atmci_init_slot(struct atmel_mci *host,
+		struct mci_slot_pdata *slot_data, unsigned int id,
+		u32 sdc_reg)
+{
+	struct mmc_host			*mmc;
+	struct atmel_mci_slot		*slot;
+
+	mmc = mmc_alloc_host(sizeof(struct atmel_mci_slot), &host->pdev->dev);
+	if (!mmc)
+		return -ENOMEM;
+
+	slot = mmc_priv(mmc);
+	slot->mmc = mmc;
+	slot->host = host;
+	slot->detect_pin = slot_data->detect_pin;
+	slot->wp_pin = slot_data->wp_pin;
+	slot->sdc_reg = sdc_reg;
+
+	mmc->ops = &atmci_ops;
+	mmc->f_min = DIV_ROUND_UP(host->bus_hz, 512);
+	mmc->f_max = host->bus_hz / 2;
+	mmc->ocr_avail	= MMC_VDD_32_33 | MMC_VDD_33_34;
+	if (slot_data->bus_width >= 4)
+		mmc->caps |= MMC_CAP_4_BIT_DATA;
+
+	mmc->max_hw_segs = 64;
+	mmc->max_phys_segs = 64;
+	mmc->max_req_size = 32768 * 512;
+	mmc->max_blk_size = 32768;
+	mmc->max_blk_count = 512;
+
+	/* Assume card is present initially */
+	set_bit(ATMCI_CARD_PRESENT, &slot->flags);
+	if (gpio_is_valid(slot->detect_pin)) {
+		if (gpio_request(slot->detect_pin, "mmc_detect")) {
+			dev_dbg(&mmc->class_dev, "no detect pin available\n");
+			slot->detect_pin = -EBUSY;
+		} else if (gpio_get_value(slot->detect_pin)) {
+			clear_bit(ATMCI_CARD_PRESENT, &slot->flags);
+		}
+	}
+
+	if (!gpio_is_valid(slot->detect_pin))
+		mmc->caps |= MMC_CAP_NEEDS_POLL;
+
+	if (gpio_is_valid(slot->wp_pin)) {
+		if (gpio_request(slot->wp_pin, "mmc_wp")) {
+			dev_dbg(&mmc->class_dev, "no WP pin available\n");
+			slot->wp_pin = -EBUSY;
+		}
+	}
+
+	host->slot[id] = slot;
+	mmc_add_host(mmc);
+
+	if (gpio_is_valid(slot->detect_pin)) {
+		int ret;
+
+		setup_timer(&slot->detect_timer, atmci_detect_change,
+				(unsigned long)slot);
+
+		ret = request_irq(gpio_to_irq(slot->detect_pin),
+				atmci_detect_interrupt,
+				IRQF_TRIGGER_FALLING | IRQF_TRIGGER_RISING,
+				"mmc-detect", slot);
+		if (ret) {
+			dev_dbg(&mmc->class_dev,
+				"could not request IRQ %d for detect pin\n",
+				gpio_to_irq(slot->detect_pin));
+			gpio_free(slot->detect_pin);
+			slot->detect_pin = -EBUSY;
+		}
+	}
+
+	atmci_init_debugfs(slot);
+
+	return 0;
+}
+
+static void __exit atmci_cleanup_slot(struct atmel_mci_slot *slot,
+		unsigned int id)
+{
+	/* Debugfs stuff is cleaned up by mmc core */
+
+	set_bit(ATMCI_SHUTDOWN, &slot->flags);
+	smp_wmb();
+
+	mmc_remove_host(slot->mmc);
+
+	if (gpio_is_valid(slot->detect_pin)) {
+		int pin = slot->detect_pin;
+
+		free_irq(gpio_to_irq(pin), slot);
+		del_timer_sync(&slot->detect_timer);
+		gpio_free(pin);
+	}
+	if (gpio_is_valid(slot->wp_pin))
+		gpio_free(slot->wp_pin);
+
+	slot->host->slot[id] = NULL;
+	mmc_free_host(slot->mmc);
+}
+
 static int __init atmci_probe(struct platform_device *pdev)
 {
 	struct mci_platform_data	*pdata;
-	struct atmel_mci *host;
-	struct mmc_host *mmc;
-	struct resource *regs;
-	int irq;
-	int ret;
+	struct atmel_mci		*host;
+	struct resource			*regs;
+	unsigned int			nr_slots;
+	int				irq;
+	int				ret;
 
 	regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (!regs)
@@ -1005,15 +1619,13 @@ static int __init atmci_probe(struct platform_device *pdev)
 	if (irq < 0)
 		return irq;
 
-	mmc = mmc_alloc_host(sizeof(struct atmel_mci), &pdev->dev);
-	if (!mmc)
+	host = kzalloc(sizeof(struct atmel_mci), GFP_KERNEL);
+	if (!host)
 		return -ENOMEM;
 
-	host = mmc_priv(mmc);
 	host->pdev = pdev;
-	host->mmc = mmc;
-	host->detect_pin = pdata->detect_pin;
-	host->wp_pin = pdata->wp_pin;
+	spin_lock_init(&host->lock);
+	INIT_LIST_HEAD(&host->queue);
 
 	host->mck = clk_get(&pdev->dev, "mci_clk");
 	if (IS_ERR(host->mck)) {
@@ -1033,122 +1645,102 @@ static int __init atmci_probe(struct platform_device *pdev)
 
 	host->mapbase = regs->start;
 
-	mmc->ops = &atmci_ops;
-	mmc->f_min = (host->bus_hz + 511) / 512;
-	mmc->f_max = host->bus_hz / 2;
-	mmc->ocr_avail	= MMC_VDD_32_33 | MMC_VDD_33_34;
-	mmc->caps |= MMC_CAP_4_BIT_DATA;
-
-	mmc->max_hw_segs = 64;
-	mmc->max_phys_segs = 64;
-	mmc->max_req_size = 32768 * 512;
-	mmc->max_blk_size = 32768;
-	mmc->max_blk_count = 512;
-
-	tasklet_init(&host->tasklet, atmci_tasklet_func, (unsigned long)mmc);
+	tasklet_init(&host->tasklet, atmci_tasklet_func, (unsigned long)host);
 
-	ret = request_irq(irq, atmci_interrupt, 0, pdev->dev.bus_id, mmc);
+	ret = request_irq(irq, atmci_interrupt, 0, pdev->dev.bus_id, host);
 	if (ret)
 		goto err_request_irq;
 
-	/* Assume card is present if we don't have a detect pin */
-	host->present = 1;
-	if (gpio_is_valid(host->detect_pin)) {
-		if (gpio_request(host->detect_pin, "mmc_detect")) {
-			dev_dbg(&mmc->class_dev, "no detect pin available\n");
-			host->detect_pin = -1;
-		} else {
-			host->present = !gpio_get_value(host->detect_pin);
-		}
-	}
+#ifdef CONFIG_MMC_ATMELMCI_DMA
+	if (pdata->dma_slave) {
+		struct dma_slave *slave = pdata->dma_slave;
 
-	if (!gpio_is_valid(host->detect_pin))
-		mmc->caps |= MMC_CAP_NEEDS_POLL;
+		slave->tx_reg = regs->start + MCI_TDR;
+		slave->rx_reg = regs->start + MCI_RDR;
 
-	if (gpio_is_valid(host->wp_pin)) {
-		if (gpio_request(host->wp_pin, "mmc_wp")) {
-			dev_dbg(&mmc->class_dev, "no WP pin available\n");
-			host->wp_pin = -1;
-		}
+		/* Try to grab a DMA channel */
+		host->dma.client.event_callback = atmci_dma_event;
+		dma_cap_set(DMA_SLAVE, host->dma.client.cap_mask);
+		host->dma.client.slave = slave;
+
+		dma_async_client_register(&host->dma.client);
+		dma_async_client_chan_request(&host->dma.client);
+	} else {
+		dev_notice(&pdev->dev, "DMA not available, using PIO\n");
 	}
+#endif /* CONFIG_MMC_ATMELMCI_DMA */
 
 	platform_set_drvdata(pdev, host);
 
-	mmc_add_host(mmc);
-
-	if (gpio_is_valid(host->detect_pin)) {
-		setup_timer(&host->detect_timer, atmci_detect_change,
-				(unsigned long)host);
-
-		ret = request_irq(gpio_to_irq(host->detect_pin),
-				atmci_detect_interrupt,
-				IRQF_TRIGGER_FALLING | IRQF_TRIGGER_RISING,
-				"mmc-detect", mmc);
-		if (ret) {
-			dev_dbg(&mmc->class_dev,
-				"could not request IRQ %d for detect pin\n",
-				gpio_to_irq(host->detect_pin));
-			gpio_free(host->detect_pin);
-			host->detect_pin = -1;
-		}
+	/* We need at least one slot to succeed */
+	nr_slots = 0;
+	ret = -ENODEV;
+	if (pdata->slot[0].bus_width) {
+		ret = atmci_init_slot(host, &pdata->slot[0],
+				MCI_SDCSEL_SLOT_A, 0);
+		if (!ret)
+			nr_slots++;
+	}
+	if (pdata->slot[1].bus_width) {
+		ret = atmci_init_slot(host, &pdata->slot[1],
+				MCI_SDCSEL_SLOT_B, 1);
+		if (!ret)
+			nr_slots++;
 	}
 
-	dev_info(&mmc->class_dev,
-			"Atmel MCI controller at 0x%08lx irq %d\n",
-			host->mapbase, irq);
+	if (!nr_slots)
+		goto err_init_slot;
 
-	atmci_init_debugfs(host);
+	dev_info(&pdev->dev,
+			"Atmel MCI controller at 0x%08lx irq %d, %u slots\n",
+			host->mapbase, irq, nr_slots);
 
 	return 0;
 
+err_init_slot:
+#ifdef CONFIG_MMC_ATMELMCI_DMA
+	if (pdata->dma_slave)
+		dma_async_client_unregister(&host->dma.client);
+#endif
+	free_irq(irq, host);
 err_request_irq:
 	iounmap(host->regs);
 err_ioremap:
 	clk_put(host->mck);
 err_clk_get:
-	mmc_free_host(mmc);
+	kfree(host);
 	return ret;
 }
 
 static int __exit atmci_remove(struct platform_device *pdev)
 {
-	struct atmel_mci *host = platform_get_drvdata(pdev);
+	struct atmel_mci	*host = platform_get_drvdata(pdev);
+	unsigned int		i;
 
 	platform_set_drvdata(pdev, NULL);
 
-	if (host) {
-		/* Debugfs stuff is cleaned up by mmc core */
-
-		if (gpio_is_valid(host->detect_pin)) {
-			int pin = host->detect_pin;
-
-			/* Make sure the timer doesn't enable the interrupt */
-			host->detect_pin = -1;
-			smp_wmb();
-
-			free_irq(gpio_to_irq(pin), host->mmc);
-			del_timer_sync(&host->detect_timer);
-			gpio_free(pin);
-		}
-
-		mmc_remove_host(host->mmc);
+	for (i = 0; i < ATMEL_MCI_MAX_NR_SLOTS; i++) {
+		if (host->slot[i])
+			atmci_cleanup_slot(host->slot[i], i);
+	}
 
-		clk_enable(host->mck);
-		mci_writel(host, IDR, ~0UL);
-		mci_writel(host, CR, MCI_CR_MCIDIS);
-		mci_readl(host, SR);
-		clk_disable(host->mck);
+	clk_enable(host->mck);
+	mci_writel(host, IDR, ~0UL);
+	mci_writel(host, CR, MCI_CR_MCIDIS);
+	mci_readl(host, SR);
+	clk_disable(host->mck);
 
-		if (gpio_is_valid(host->wp_pin))
-			gpio_free(host->wp_pin);
+#ifdef CONFIG_MMC_ATMELMCI_DMA
+	if (host->dma.client.slave)
+		dma_async_client_unregister(&host->dma.client);
+#endif
 
-		free_irq(platform_get_irq(pdev, 0), host->mmc);
-		iounmap(host->regs);
+	free_irq(platform_get_irq(pdev, 0), host);
+	iounmap(host->regs);
 
-		clk_put(host->mck);
+	clk_put(host->mck);
+	kfree(host);
 
-		mmc_free_host(host->mmc);
-	}
 	return 0;
 }
 
diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c
index 7503b81374e0..07faf5412a1f 100644
--- a/drivers/mmc/host/mmc_spi.c
+++ b/drivers/mmc/host/mmc_spi.c
@@ -95,8 +95,6 @@
  * reads which takes nowhere near that long.  Older cards may be able to use
  * shorter timeouts ... but why bother?
  */
-#define readblock_timeout	ktime_set(0, 100 * 1000 * 1000)
-#define writeblock_timeout	ktime_set(0, 250 * 1000 * 1000)
 #define r1b_timeout		ktime_set(3, 0)
 
 
@@ -220,9 +218,9 @@ mmc_spi_wait_unbusy(struct mmc_spi_host *host, ktime_t timeout)
 	return mmc_spi_skip(host, timeout, sizeof(host->data->status), 0);
 }
 
-static int mmc_spi_readtoken(struct mmc_spi_host *host)
+static int mmc_spi_readtoken(struct mmc_spi_host *host, ktime_t timeout)
 {
-	return mmc_spi_skip(host, readblock_timeout, 1, 0xff);
+	return mmc_spi_skip(host, timeout, 1, 0xff);
 }
 
 
@@ -605,7 +603,8 @@ mmc_spi_setup_data_message(
  * Return negative errno, else success.
  */
 static int
-mmc_spi_writeblock(struct mmc_spi_host *host, struct spi_transfer *t)
+mmc_spi_writeblock(struct mmc_spi_host *host, struct spi_transfer *t,
+	ktime_t timeout)
 {
 	struct spi_device	*spi = host->spi;
 	int			status, i;
@@ -673,7 +672,7 @@ mmc_spi_writeblock(struct mmc_spi_host *host, struct spi_transfer *t)
 		if (scratch->status[i] != 0)
 			return 0;
 	}
-	return mmc_spi_wait_unbusy(host, writeblock_timeout);
+	return mmc_spi_wait_unbusy(host, timeout);
 }
 
 /*
@@ -693,7 +692,8 @@ mmc_spi_writeblock(struct mmc_spi_host *host, struct spi_transfer *t)
  * STOP_TRANSMISSION command.
  */
 static int
-mmc_spi_readblock(struct mmc_spi_host *host, struct spi_transfer *t)
+mmc_spi_readblock(struct mmc_spi_host *host, struct spi_transfer *t,
+	ktime_t timeout)
 {
 	struct spi_device	*spi = host->spi;
 	int			status;
@@ -707,7 +707,7 @@ mmc_spi_readblock(struct mmc_spi_host *host, struct spi_transfer *t)
 		return status;
 	status = scratch->status[0];
 	if (status == 0xff || status == 0)
-		status = mmc_spi_readtoken(host);
+		status = mmc_spi_readtoken(host, timeout);
 
 	if (status == SPI_TOKEN_SINGLE) {
 		if (host->dma_dev) {
@@ -778,6 +778,8 @@ mmc_spi_data_do(struct mmc_spi_host *host, struct mmc_command *cmd,
 	struct scatterlist	*sg;
 	unsigned		n_sg;
 	int			multiple = (data->blocks > 1);
+	u32			clock_rate;
+	ktime_t			timeout;
 
 	if (data->flags & MMC_DATA_READ)
 		direction = DMA_FROM_DEVICE;
@@ -786,6 +788,14 @@ mmc_spi_data_do(struct mmc_spi_host *host, struct mmc_command *cmd,
 	mmc_spi_setup_data_message(host, multiple, direction);
 	t = &host->t;
 
+	if (t->speed_hz)
+		clock_rate = t->speed_hz;
+	else
+		clock_rate = spi->max_speed_hz;
+
+	timeout = ktime_add_ns(ktime_set(0, 0), data->timeout_ns +
+			data->timeout_clks * 1000000 / clock_rate);
+
 	/* Handle scatterlist segments one at a time, with synch for
 	 * each 512-byte block
 	 */
@@ -832,9 +842,9 @@ mmc_spi_data_do(struct mmc_spi_host *host, struct mmc_command *cmd,
 				t->len);
 
 			if (direction == DMA_TO_DEVICE)
-				status = mmc_spi_writeblock(host, t);
+				status = mmc_spi_writeblock(host, t, timeout);
 			else
-				status = mmc_spi_readblock(host, t);
+				status = mmc_spi_readblock(host, t, timeout);
 			if (status < 0)
 				break;
 
@@ -917,7 +927,7 @@ mmc_spi_data_do(struct mmc_spi_host *host, struct mmc_command *cmd,
 			if (scratch->status[tmp] != 0)
 				return;
 		}
-		tmp = mmc_spi_wait_unbusy(host, writeblock_timeout);
+		tmp = mmc_spi_wait_unbusy(host, timeout);
 		if (tmp < 0 && !data->error)
 			data->error = tmp;
 	}
diff --git a/drivers/mmc/host/sdhci-pci.c b/drivers/mmc/host/sdhci-pci.c
index fcb14c2346cc..9bd7026b0021 100644
--- a/drivers/mmc/host/sdhci-pci.c
+++ b/drivers/mmc/host/sdhci-pci.c
@@ -144,7 +144,8 @@ static int jmicron_probe(struct sdhci_pci_chip *chip)
 			  SDHCI_QUIRK_32BIT_DMA_SIZE |
 			  SDHCI_QUIRK_32BIT_ADMA_SIZE |
 			  SDHCI_QUIRK_RESET_AFTER_REQUEST |
-			  SDHCI_QUIRK_BROKEN_SMALL_PIO;
+			  SDHCI_QUIRK_BROKEN_SMALL_PIO |
+			  SDHCI_QUIRK_FORCE_HIGHSPEED;
 	}
 
 	/*
@@ -326,7 +327,7 @@ static const struct pci_device_id pci_ids[] __devinitdata = {
 
 	{
 		.vendor         = PCI_VENDOR_ID_MARVELL,
-		.device         = PCI_DEVICE_ID_MARVELL_CAFE_SD,
+		.device         = PCI_DEVICE_ID_MARVELL_88ALP01_SD,
 		.subvendor      = PCI_ANY_ID,
 		.subdevice      = PCI_ANY_ID,
 		.driver_data    = (kernel_ulong_t)&sdhci_cafe,
diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index e3a8133560a2..30f64b1f2354 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -177,7 +177,7 @@ static void sdhci_read_block_pio(struct sdhci_host *host)
 {
 	unsigned long flags;
 	size_t blksize, len, chunk;
-	u32 scratch;
+	u32 uninitialized_var(scratch);
 	u8 *buf;
 
 	DBG("PIO reading\n");
@@ -1154,7 +1154,7 @@ static void sdhci_tasklet_card(unsigned long param)
 
 	spin_unlock_irqrestore(&host->lock, flags);
 
-	mmc_detect_change(host->mmc, msecs_to_jiffies(500));
+	mmc_detect_change(host->mmc, msecs_to_jiffies(200));
 }
 
 static void sdhci_tasklet_finish(unsigned long param)
@@ -1266,9 +1266,31 @@ static void sdhci_cmd_irq(struct sdhci_host *host, u32 intmask)
 			SDHCI_INT_INDEX))
 		host->cmd->error = -EILSEQ;
 
-	if (host->cmd->error)
+	if (host->cmd->error) {
 		tasklet_schedule(&host->finish_tasklet);
-	else if (intmask & SDHCI_INT_RESPONSE)
+		return;
+	}
+
+	/*
+	 * The host can send and interrupt when the busy state has
+	 * ended, allowing us to wait without wasting CPU cycles.
+	 * Unfortunately this is overloaded on the "data complete"
+	 * interrupt, so we need to take some care when handling
+	 * it.
+	 *
+	 * Note: The 1.0 specification is a bit ambiguous about this
+	 *       feature so there might be some problems with older
+	 *       controllers.
+	 */
+	if (host->cmd->flags & MMC_RSP_BUSY) {
+		if (host->cmd->data)
+			DBG("Cannot wait for busy signal when also "
+				"doing a data transfer");
+		else
+			return;
+	}
+
+	if (intmask & SDHCI_INT_RESPONSE)
 		sdhci_finish_command(host);
 }
 
@@ -1278,11 +1300,16 @@ static void sdhci_data_irq(struct sdhci_host *host, u32 intmask)
 
 	if (!host->data) {
 		/*
-		 * A data end interrupt is sent together with the response
-		 * for the stop command.
+		 * The "data complete" interrupt is also used to
+		 * indicate that a busy state has ended. See comment
+		 * above in sdhci_cmd_irq().
 		 */
-		if (intmask & SDHCI_INT_DATA_END)
-			return;
+		if (host->cmd && (host->cmd->flags & MMC_RSP_BUSY)) {
+			if (intmask & SDHCI_INT_DATA_END) {
+				sdhci_finish_command(host);
+				return;
+			}
+		}
 
 		printk(KERN_ERR "%s: Got data interrupt 0x%08x even "
 			"though no data operation was in progress.\n",
@@ -1604,7 +1631,8 @@ int sdhci_add_host(struct sdhci_host *host)
 	mmc->f_max = host->max_clk;
 	mmc->caps = MMC_CAP_4_BIT_DATA | MMC_CAP_SDIO_IRQ;
 
-	if (caps & SDHCI_CAN_DO_HISPD)
+	if ((caps & SDHCI_CAN_DO_HISPD) ||
+		(host->quirks & SDHCI_QUIRK_FORCE_HIGHSPEED))
 		mmc->caps |= MMC_CAP_SD_HIGHSPEED;
 
 	mmc->ocr_avail = 0;
diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index 197d4a05f4ae..31f4b1528e76 100644
--- a/drivers/mmc/host/sdhci.h
+++ b/drivers/mmc/host/sdhci.h
@@ -208,6 +208,8 @@ struct sdhci_host {
 #define SDHCI_QUIRK_BROKEN_TIMEOUT_VAL			(1<<12)
 /* Controller has an issue with buffer bits for small transfers */
 #define SDHCI_QUIRK_BROKEN_SMALL_PIO			(1<<13)
+/* Controller supports high speed but doesn't have the caps bit set */
+#define SDHCI_QUIRK_FORCE_HIGHSPEED			(1<<14)
 
 	int			irq;		/* Device IRQ */
 	void __iomem *		ioaddr;		/* Mapped address */
diff --git a/drivers/mtd/maps/sun_uflash.c b/drivers/mtd/maps/sun_uflash.c
index 0d7c88396c88..fd7a1017399a 100644
--- a/drivers/mtd/maps/sun_uflash.c
+++ b/drivers/mtd/maps/sun_uflash.c
@@ -1,13 +1,10 @@
-/*
- *
- * sun_uflash - Driver implementation for user-programmable flash
- * present on many Sun Microsystems SME boardsets.
+/* sun_uflash.c - Driver for user-programmable flash on
+ *                Sun Microsystems SME boardsets.
  *
  * This driver does NOT provide access to the OBP-flash for
  * safety reasons-- use <linux>/drivers/sbus/char/flash.c instead.
  *
  * Copyright (c) 2001 Eric Brower (ebrower@usa.net)
- *
  */
 
 #include <linux/kernel.h>
@@ -16,8 +13,8 @@
 #include <linux/errno.h>
 #include <linux/init.h>
 #include <linux/ioport.h>
-#include <asm/ebus.h>
-#include <asm/oplib.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 #include <asm/prom.h>
 #include <asm/uaccess.h>
 #include <asm/io.h>
@@ -26,67 +23,65 @@
 #include <linux/mtd/map.h>
 
 #define UFLASH_OBPNAME	"flashprom"
-#define UFLASH_DEVNAME 	"userflash"
+#define DRIVER_NAME	"sun_uflash"
+#define PFX		DRIVER_NAME ": "
 
 #define UFLASH_WINDOW_SIZE	0x200000
 #define UFLASH_BUSWIDTH		1			/* EBus is 8-bit */
 
 MODULE_AUTHOR("Eric Brower <ebrower@usa.net>");
 MODULE_DESCRIPTION("User-programmable flash device on Sun Microsystems boardsets");
-MODULE_SUPPORTED_DEVICE("userflash");
+MODULE_SUPPORTED_DEVICE(DRIVER_NAME);
 MODULE_LICENSE("GPL");
-MODULE_VERSION("2.0");
+MODULE_VERSION("2.1");
 
-static LIST_HEAD(device_list);
 struct uflash_dev {
 	const char		*name;	/* device name */
 	struct map_info 	map;	/* mtd map info */
 	struct mtd_info		*mtd;	/* mtd info */
 };
 
-
 struct map_info uflash_map_templ = {
 	.name =		"SUNW,???-????",
 	.size =		UFLASH_WINDOW_SIZE,
 	.bankwidth =	UFLASH_BUSWIDTH,
 };
 
-int uflash_devinit(struct linux_ebus_device *edev, struct device_node *dp)
+int uflash_devinit(struct of_device *op, struct device_node *dp)
 {
 	struct uflash_dev *up;
-	struct resource *res;
 
-	res = &edev->resource[0];
-
-	if (edev->num_addrs != 1) {
+	if (op->resource[1].flags) {
 		/* Non-CFI userflash device-- once I find one we
 		 * can work on supporting it.
 		 */
-		printk("%s: unsupported device at 0x%llx (%d regs): " \
-			"email ebrower@usa.net\n",
-		       dp->full_name, (unsigned long long)res->start,
-		       edev->num_addrs);
+		printk(KERN_ERR PFX "Unsupported device at %s, 0x%llx\n",
+		       dp->full_name, (unsigned long long)op->resource[0].start);
 
 		return -ENODEV;
 	}
 
 	up = kzalloc(sizeof(struct uflash_dev), GFP_KERNEL);
-	if (!up)
+	if (!up) {
+		printk(KERN_ERR PFX "Cannot allocate struct uflash_dev\n");
 		return -ENOMEM;
+	}
 
 	/* copy defaults and tweak parameters */
 	memcpy(&up->map, &uflash_map_templ, sizeof(uflash_map_templ));
-	up->map.size = (res->end - res->start) + 1UL;
+
+	up->map.size = resource_size(&op->resource[0]);
 
 	up->name = of_get_property(dp, "model", NULL);
 	if (up->name && 0 < strlen(up->name))
 		up->map.name = (char *)up->name;
 
-	up->map.phys = res->start;
+	up->map.phys = op->resource[0].start;
 
-	up->map.virt = ioremap_nocache(res->start, up->map.size);
+	up->map.virt = of_ioremap(&op->resource[0], 0, up->map.size,
+				  DRIVER_NAME);
 	if (!up->map.virt) {
-		printk("%s: Failed to map device.\n", dp->full_name);
+		printk(KERN_ERR PFX "Failed to map device.\n");
 		kfree(up);
 
 		return -EINVAL;
@@ -97,7 +92,7 @@ int uflash_devinit(struct linux_ebus_device *edev, struct device_node *dp)
 	/* MTD registration */
 	up->mtd = do_map_probe("cfi_probe", &up->map);
 	if (!up->mtd) {
-		iounmap(up->map.virt);
+		of_iounmap(&op->resource[0], up->map.virt, up->map.size);
 		kfree(up);
 
 		return -ENXIO;
@@ -107,32 +102,34 @@ int uflash_devinit(struct linux_ebus_device *edev, struct device_node *dp)
 
 	add_mtd_device(up->mtd);
 
-	dev_set_drvdata(&edev->ofdev.dev, up);
+	dev_set_drvdata(&op->dev, up);
 
 	return 0;
 }
 
-static int __devinit uflash_probe(struct of_device *dev, const struct of_device_id *match)
+static int __devinit uflash_probe(struct of_device *op, const struct of_device_id *match)
 {
-	struct linux_ebus_device *edev = to_ebus_device(&dev->dev);
-	struct device_node *dp = dev->node;
+	struct device_node *dp = op->node;
 
-	if (of_find_property(dp, "user", NULL))
+	/* Flashprom must have the "user" property in order to
+	 * be used by this driver.
+	 */
+	if (!of_find_property(dp, "user", NULL))
 		return -ENODEV;
 
-	return uflash_devinit(edev, dp);
+	return uflash_devinit(op, dp);
 }
 
-static int __devexit uflash_remove(struct of_device *dev)
+static int __devexit uflash_remove(struct of_device *op)
 {
-	struct uflash_dev *up = dev_get_drvdata(&dev->dev);
+	struct uflash_dev *up = dev_get_drvdata(&op->dev);
 
 	if (up->mtd) {
 		del_mtd_device(up->mtd);
 		map_destroy(up->mtd);
 	}
 	if (up->map.virt) {
-		iounmap(up->map.virt);
+		of_iounmap(&op->resource[0], up->map.virt, up->map.size);
 		up->map.virt = NULL;
 	}
 
@@ -141,7 +138,7 @@ static int __devexit uflash_remove(struct of_device *dev)
 	return 0;
 }
 
-static struct of_device_id uflash_match[] = {
+static const struct of_device_id uflash_match[] = {
 	{
 		.name = UFLASH_OBPNAME,
 	},
@@ -151,7 +148,7 @@ static struct of_device_id uflash_match[] = {
 MODULE_DEVICE_TABLE(of, uflash_match);
 
 static struct of_platform_driver uflash_driver = {
-	.name		= UFLASH_DEVNAME,
+	.name		= DRIVER_NAME,
 	.match_table	= uflash_match,
 	.probe		= uflash_probe,
 	.remove		= __devexit_p(uflash_remove),
@@ -159,7 +156,7 @@ static struct of_platform_driver uflash_driver = {
 
 static int __init uflash_init(void)
 {
-	return of_register_driver(&uflash_driver, &ebus_bus_type);
+	return of_register_driver(&uflash_driver, &of_bus_type);
 }
 
 static void __exit uflash_exit(void)
diff --git a/drivers/mtd/nand/cafe_nand.c b/drivers/mtd/nand/cafe_nand.c
index 95345d051579..b8064bf3aee4 100644
--- a/drivers/mtd/nand/cafe_nand.c
+++ b/drivers/mtd/nand/cafe_nand.c
@@ -1,6 +1,9 @@
 /*
  * Driver for One Laptop Per Child ‘CAFÉ’ controller, aka Marvell 88ALP01
  *
+ * The data sheet for this device can be found at:
+ *    http://www.marvell.com/products/pcconn/88ALP01.jsp
+ *
  * Copyright © 2006 Red Hat, Inc.
  * Copyright © 2006 David Woodhouse <dwmw2@infradead.org>
  */
@@ -842,7 +845,8 @@ static void __devexit cafe_nand_remove(struct pci_dev *pdev)
 }
 
 static struct pci_device_id cafe_nand_tbl[] = {
-	{ 0x11ab, 0x4100, PCI_ANY_ID, PCI_ANY_ID },
+	{ PCI_VENDOR_ID_MARVELL, PCI_DEVICE_ID_MARVELL_88ALP01_NAND,
+	  PCI_ANY_ID, PCI_ANY_ID },
 	{ }
 };
 
diff --git a/drivers/net/myri_sbus.c b/drivers/net/myri_sbus.c
index 06e682334c7e..3ad7589d6a1c 100644
--- a/drivers/net/myri_sbus.c
+++ b/drivers/net/myri_sbus.c
@@ -1,6 +1,6 @@
 /* myri_sbus.c: MyriCOM MyriNET SBUS card driver.
  *
- * Copyright (C) 1996, 1999, 2006 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 1996, 1999, 2006, 2008 David S. Miller (davem@davemloft.net)
  */
 
 static char version[] =
@@ -22,6 +22,9 @@ static char version[] =
 #include <linux/etherdevice.h>
 #include <linux/skbuff.h>
 #include <linux/bitops.h>
+#include <linux/dma-mapping.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <net/dst.h>
 #include <net/arp.h>
@@ -33,7 +36,6 @@ static char version[] =
 #include <asm/dma.h>
 #include <asm/byteorder.h>
 #include <asm/idprom.h>
-#include <asm/sbus.h>
 #include <asm/openprom.h>
 #include <asm/oplib.h>
 #include <asm/auxio.h>
@@ -243,7 +245,8 @@ static void myri_clean_rings(struct myri_eth *mp)
 			u32 dma_addr;
 
 			dma_addr = sbus_readl(&rxd->myri_scatters[0].addr);
-			sbus_unmap_single(mp->myri_sdev, dma_addr, RX_ALLOC_SIZE, SBUS_DMA_FROMDEVICE);
+			dma_unmap_single(&mp->myri_op->dev, dma_addr,
+					 RX_ALLOC_SIZE, DMA_FROM_DEVICE);
 			dev_kfree_skb(mp->rx_skbs[i]);
 			mp->rx_skbs[i] = NULL;
 		}
@@ -259,7 +262,9 @@ static void myri_clean_rings(struct myri_eth *mp)
 			u32 dma_addr;
 
 			dma_addr = sbus_readl(&txd->myri_gathers[0].addr);
-			sbus_unmap_single(mp->myri_sdev, dma_addr, (skb->len + 3) & ~3, SBUS_DMA_TODEVICE);
+			dma_unmap_single(&mp->myri_op->dev, dma_addr,
+					 (skb->len + 3) & ~3,
+					 DMA_TO_DEVICE);
 			dev_kfree_skb(mp->tx_skbs[i]);
 			mp->tx_skbs[i] = NULL;
 		}
@@ -288,7 +293,9 @@ static void myri_init_rings(struct myri_eth *mp, int from_irq)
 		skb->dev = dev;
 		skb_put(skb, RX_ALLOC_SIZE);
 
-		dma_addr = sbus_map_single(mp->myri_sdev, skb->data, RX_ALLOC_SIZE, SBUS_DMA_FROMDEVICE);
+		dma_addr = dma_map_single(&mp->myri_op->dev,
+					  skb->data, RX_ALLOC_SIZE,
+					  DMA_FROM_DEVICE);
 		sbus_writel(dma_addr, &rxd[i].myri_scatters[0].addr);
 		sbus_writel(RX_ALLOC_SIZE, &rxd[i].myri_scatters[0].len);
 		sbus_writel(i, &rxd[i].ctx);
@@ -344,7 +351,8 @@ static void myri_tx(struct myri_eth *mp, struct net_device *dev)
 
 		DTX(("SKB[%d] ", entry));
 		dma_addr = sbus_readl(&sq->myri_txd[entry].myri_gathers[0].addr);
-		sbus_unmap_single(mp->myri_sdev, dma_addr, skb->len, SBUS_DMA_TODEVICE);
+		dma_unmap_single(&mp->myri_op->dev, dma_addr,
+				 skb->len, DMA_TO_DEVICE);
 		dev_kfree_skb(skb);
 		mp->tx_skbs[entry] = NULL;
 		dev->stats.tx_packets++;
@@ -423,9 +431,9 @@ static void myri_rx(struct myri_eth *mp, struct net_device *dev)
 
 		/* Check for errors. */
 		DRX(("rxd[%d]: %p len[%d] csum[%08x] ", entry, rxd, len, csum));
-		sbus_dma_sync_single_for_cpu(mp->myri_sdev,
-					     sbus_readl(&rxd->myri_scatters[0].addr),
-					     RX_ALLOC_SIZE, SBUS_DMA_FROMDEVICE);
+		dma_sync_single_for_cpu(&mp->myri_op->dev,
+					sbus_readl(&rxd->myri_scatters[0].addr),
+					RX_ALLOC_SIZE, DMA_FROM_DEVICE);
 		if (len < (ETH_HLEN + MYRI_PAD_LEN) || (skb->data[0] != MYRI_PAD_LEN)) {
 			DRX(("ERROR["));
 			dev->stats.rx_errors++;
@@ -442,10 +450,10 @@ static void myri_rx(struct myri_eth *mp, struct net_device *dev)
 			drops++;
 			DRX(("DROP "));
 			dev->stats.rx_dropped++;
-			sbus_dma_sync_single_for_device(mp->myri_sdev,
-							sbus_readl(&rxd->myri_scatters[0].addr),
-							RX_ALLOC_SIZE,
-							SBUS_DMA_FROMDEVICE);
+			dma_sync_single_for_device(&mp->myri_op->dev,
+						   sbus_readl(&rxd->myri_scatters[0].addr),
+						   RX_ALLOC_SIZE,
+						   DMA_FROM_DEVICE);
 			sbus_writel(RX_ALLOC_SIZE, &rxd->myri_scatters[0].len);
 			sbus_writel(index, &rxd->ctx);
 			sbus_writel(1, &rxd->num_sg);
@@ -464,17 +472,17 @@ static void myri_rx(struct myri_eth *mp, struct net_device *dev)
 				DRX(("skb_alloc(FAILED) "));
 				goto drop_it;
 			}
-			sbus_unmap_single(mp->myri_sdev,
-					  sbus_readl(&rxd->myri_scatters[0].addr),
-					  RX_ALLOC_SIZE,
-					  SBUS_DMA_FROMDEVICE);
+			dma_unmap_single(&mp->myri_op->dev,
+					 sbus_readl(&rxd->myri_scatters[0].addr),
+					 RX_ALLOC_SIZE,
+					 DMA_FROM_DEVICE);
 			mp->rx_skbs[index] = new_skb;
 			new_skb->dev = dev;
 			skb_put(new_skb, RX_ALLOC_SIZE);
-			dma_addr = sbus_map_single(mp->myri_sdev,
-						   new_skb->data,
-						   RX_ALLOC_SIZE,
-						   SBUS_DMA_FROMDEVICE);
+			dma_addr = dma_map_single(&mp->myri_op->dev,
+						  new_skb->data,
+						  RX_ALLOC_SIZE,
+						  DMA_FROM_DEVICE);
 			sbus_writel(dma_addr, &rxd->myri_scatters[0].addr);
 			sbus_writel(RX_ALLOC_SIZE, &rxd->myri_scatters[0].len);
 			sbus_writel(index, &rxd->ctx);
@@ -500,10 +508,10 @@ static void myri_rx(struct myri_eth *mp, struct net_device *dev)
 
 			/* Reuse original ring buffer. */
 			DRX(("reuse "));
-			sbus_dma_sync_single_for_device(mp->myri_sdev,
-							sbus_readl(&rxd->myri_scatters[0].addr),
-							RX_ALLOC_SIZE,
-							SBUS_DMA_FROMDEVICE);
+			dma_sync_single_for_device(&mp->myri_op->dev,
+						   sbus_readl(&rxd->myri_scatters[0].addr),
+						   RX_ALLOC_SIZE,
+						   DMA_FROM_DEVICE);
 			sbus_writel(RX_ALLOC_SIZE, &rxd->myri_scatters[0].len);
 			sbus_writel(index, &rxd->ctx);
 			sbus_writel(1, &rxd->num_sg);
@@ -652,7 +660,8 @@ static int myri_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		sbus_writew((skb->data[4] << 8) | skb->data[5], &txd->addr[3]);
 	}
 
-	dma_addr = sbus_map_single(mp->myri_sdev, skb->data, len, SBUS_DMA_TODEVICE);
+	dma_addr = dma_map_single(&mp->myri_op->dev, skb->data,
+				  len, DMA_TO_DEVICE);
 	sbus_writel(dma_addr, &txd->myri_gathers[0].addr);
 	sbus_writel(len, &txd->myri_gathers[0].len);
 	sbus_writel(1, &txd->num_sg);
@@ -891,30 +900,30 @@ static const struct header_ops myri_header_ops = {
 	.cache_update	= myri_header_cache_update,
 };
 
-static int __devinit myri_ether_init(struct sbus_dev *sdev)
+static int __devinit myri_sbus_probe(struct of_device *op, const struct of_device_id *match)
 {
-	static int num;
+	struct device_node *dp = op->node;
 	static unsigned version_printed;
 	struct net_device *dev;
-	struct myri_eth *mp;
-	unsigned char prop_buf[32];
-	int i;
 	DECLARE_MAC_BUF(mac);
+	struct myri_eth *mp;
+	const void *prop;
+	static int num;
+	int i, len;
 
-	DET(("myri_ether_init(%p,%d):\n", sdev, num));
+	DET(("myri_ether_init(%p,%d):\n", op, num));
 	dev = alloc_etherdev(sizeof(struct myri_eth));
-
 	if (!dev)
 		return -ENOMEM;
 
 	if (version_printed++ == 0)
 		printk(version);
 
-	SET_NETDEV_DEV(dev, &sdev->ofdev.dev);
+	SET_NETDEV_DEV(dev, &op->dev);
 
-	mp = (struct myri_eth *) dev->priv;
+	mp = netdev_priv(dev);
 	spin_lock_init(&mp->irq_lock);
-	mp->myri_sdev = sdev;
+	mp->myri_op = op;
 
 	/* Clean out skb arrays. */
 	for (i = 0; i < (RX_RING_SIZE + 1); i++)
@@ -924,55 +933,44 @@ static int __devinit myri_ether_init(struct sbus_dev *sdev)
 		mp->tx_skbs[i] = NULL;
 
 	/* First check for EEPROM information. */
-	i = prom_getproperty(sdev->prom_node, "myrinet-eeprom-info",
-			     (char *)&mp->eeprom, sizeof(struct myri_eeprom));
-	DET(("prom_getprop(myrinet-eeprom-info) returns %d\n", i));
-	if (i == 0 || i == -1) {
+	prop = of_get_property(dp, "myrinet-eeprom-info", &len);
+
+	if (prop)
+		memcpy(&mp->eeprom, prop, sizeof(struct myri_eeprom));
+	if (!prop) {
 		/* No eeprom property, must cook up the values ourselves. */
 		DET(("No EEPROM: "));
 		mp->eeprom.bus_type = BUS_TYPE_SBUS;
-		mp->eeprom.cpuvers = prom_getintdefault(sdev->prom_node,"cpu_version",0);
-		mp->eeprom.cval = prom_getintdefault(sdev->prom_node,"clock_value",0);
-		mp->eeprom.ramsz = prom_getintdefault(sdev->prom_node,"sram_size",0);
-		DET(("cpuvers[%d] cval[%d] ramsz[%d]\n", mp->eeprom.cpuvers,
-		     mp->eeprom.cval, mp->eeprom.ramsz));
-		if (mp->eeprom.cpuvers == 0) {
-			DET(("EEPROM: cpuvers was zero, setting to %04x\n",CPUVERS_2_3));
+		mp->eeprom.cpuvers =
+			of_getintprop_default(dp, "cpu_version", 0);
+		mp->eeprom.cval =
+			of_getintprop_default(dp, "clock_value", 0);
+		mp->eeprom.ramsz = of_getintprop_default(dp, "sram_size", 0);
+		if (!mp->eeprom.cpuvers)
 			mp->eeprom.cpuvers = CPUVERS_2_3;
-		}
-		if (mp->eeprom.cpuvers < CPUVERS_3_0) {
-			DET(("EEPROM: cpuvers < CPUVERS_3_0, clockval set to zero.\n"));
+		if (mp->eeprom.cpuvers < CPUVERS_3_0)
 			mp->eeprom.cval = 0;
-		}
-		if (mp->eeprom.ramsz == 0) {
-			DET(("EEPROM: ramsz == 0, setting to 128k\n"));
+		if (!mp->eeprom.ramsz)
 			mp->eeprom.ramsz = (128 * 1024);
-		}
-		i = prom_getproperty(sdev->prom_node, "myrinet-board-id",
-				     &prop_buf[0], 10);
-		DET(("EEPROM: prom_getprop(myrinet-board-id) returns %d\n", i));
-		if ((i != 0) && (i != -1))
-			memcpy(&mp->eeprom.id[0], &prop_buf[0], 6);
+
+		prop = of_get_property(dp, "myrinet-board-id", &len);
+		if (prop)
+			memcpy(&mp->eeprom.id[0], prop, 6);
 		else
 			set_boardid_from_idprom(mp, num);
-		i = prom_getproperty(sdev->prom_node, "fpga_version",
-				     &mp->eeprom.fvers[0], 32);
-		DET(("EEPROM: prom_getprop(fpga_version) returns %d\n", i));
-		if (i == 0 || i == -1)
+
+		prop = of_get_property(dp, "fpga_version", &len);
+		if (prop)
+			memcpy(&mp->eeprom.fvers[0], prop, 32);
+		else
 			memset(&mp->eeprom.fvers[0], 0, 32);
 
 		if (mp->eeprom.cpuvers == CPUVERS_4_1) {
-			DET(("EEPROM: cpuvers CPUVERS_4_1, "));
-			if (mp->eeprom.ramsz == (128 * 1024)) {
-				DET(("ramsize 128k, setting to 256k, "));
+			if (mp->eeprom.ramsz == (128 * 1024))
 				mp->eeprom.ramsz = (256 * 1024);
-			}
-			if ((mp->eeprom.cval==0x40414041)||(mp->eeprom.cval==0x90449044)){
-				DET(("changing cval from %08x to %08x ",
-				     mp->eeprom.cval, 0x50e450e4));
+			if ((mp->eeprom.cval == 0x40414041) ||
+			    (mp->eeprom.cval == 0x90449044))
 				mp->eeprom.cval = 0x50e450e4;
-			}
-			DET(("\n"));
 		}
 	}
 #ifdef DEBUG_DETECT
@@ -991,8 +989,8 @@ static int __devinit myri_ether_init(struct sbus_dev *sdev)
 		 * XXX only a valid version for PCI cards?  Ask feldy...
 		 */
 		DET(("Mapping regs for cpuvers < CPUVERS_4_0\n"));
-		mp->regs = sbus_ioremap(&sdev->resource[0], 0,
-					mp->reg_size, "MyriCOM Regs");
+		mp->regs = of_ioremap(&op->resource[0], 0,
+				      mp->reg_size, "MyriCOM Regs");
 		if (!mp->regs) {
 			printk("MyriCOM: Cannot map MyriCOM registers.\n");
 			goto err;
@@ -1001,13 +999,12 @@ static int __devinit myri_ether_init(struct sbus_dev *sdev)
 		mp->lregs = mp->lanai + (0x10000 * 2);
 	} else {
 		DET(("Mapping regs for cpuvers >= CPUVERS_4_0\n"));
-		mp->cregs = sbus_ioremap(&sdev->resource[0], 0,
-					 PAGE_SIZE, "MyriCOM Control Regs");
-		mp->lregs = sbus_ioremap(&sdev->resource[0], (256 * 1024),
+		mp->cregs = of_ioremap(&op->resource[0], 0,
+				       PAGE_SIZE, "MyriCOM Control Regs");
+		mp->lregs = of_ioremap(&op->resource[0], (256 * 1024),
 					 PAGE_SIZE, "MyriCOM LANAI Regs");
-		mp->lanai =
-			sbus_ioremap(&sdev->resource[0], (512 * 1024),
-				     mp->eeprom.ramsz, "MyriCOM SRAM");
+		mp->lanai = of_ioremap(&op->resource[0], (512 * 1024),
+				       mp->eeprom.ramsz, "MyriCOM SRAM");
 	}
 	DET(("Registers mapped: cregs[%p] lregs[%p] lanai[%p]\n",
 	     mp->cregs, mp->lregs, mp->lanai));
@@ -1039,16 +1036,15 @@ static int __devinit myri_ether_init(struct sbus_dev *sdev)
 	myri_reset_on(mp->cregs);
 
 	/* Get the supported DVMA burst sizes from our SBUS. */
-	mp->myri_bursts = prom_getintdefault(mp->myri_sdev->bus->prom_node,
-					     "burst-sizes", 0x00);
-
-	if (!sbus_can_burst64(sdev))
+	mp->myri_bursts = of_getintprop_default(dp->parent,
+						"burst-sizes", 0x00);
+	if (!sbus_can_burst64())
 		mp->myri_bursts &= ~(DMA_BURST64);
 
 	DET(("MYRI bursts %02x\n", mp->myri_bursts));
 
 	/* Encode SBUS interrupt level in second control register. */
-	i = prom_getint(sdev->prom_node, "interrupts");
+	i = of_getintprop_default(dp, "interrupts", 0);
 	if (i == 0)
 		i = 4;
 	DET(("prom_getint(interrupts)==%d, irqlvl set to %04x\n",
@@ -1063,7 +1059,7 @@ static int __devinit myri_ether_init(struct sbus_dev *sdev)
 	dev->tx_timeout = &myri_tx_timeout;
 	dev->watchdog_timeo = 5*HZ;
 	dev->set_multicast_list = &myri_set_multicast;
-	dev->irq = sdev->irqs[0];
+	dev->irq = op->irqs[0];
 
 	/* Register interrupt handler now. */
 	DET(("Requesting MYRIcom IRQ line.\n"));
@@ -1088,7 +1084,7 @@ static int __devinit myri_ether_init(struct sbus_dev *sdev)
 		goto err_free_irq;
 	}
 
-	dev_set_drvdata(&sdev->ofdev.dev, mp);
+	dev_set_drvdata(&op->dev, mp);
 
 	num++;
 
@@ -1105,17 +1101,9 @@ err:
 	return -ENODEV;
 }
 
-
-static int __devinit myri_sbus_probe(struct of_device *dev, const struct of_device_id *match)
-{
-	struct sbus_dev *sdev = to_sbus_device(&dev->dev);
-
-	return myri_ether_init(sdev);
-}
-
-static int __devexit myri_sbus_remove(struct of_device *dev)
+static int __devexit myri_sbus_remove(struct of_device *op)
 {
-	struct myri_eth *mp = dev_get_drvdata(&dev->dev);
+	struct myri_eth *mp = dev_get_drvdata(&op->dev);
 	struct net_device *net_dev = mp->dev;
 
 	unregister_netdev(net_dev);
@@ -1123,21 +1111,21 @@ static int __devexit myri_sbus_remove(struct of_device *dev)
 	free_irq(net_dev->irq, net_dev);
 
 	if (mp->eeprom.cpuvers < CPUVERS_4_0) {
-		sbus_iounmap(mp->regs, mp->reg_size);
+		of_iounmap(&op->resource[0], mp->regs, mp->reg_size);
 	} else {
-		sbus_iounmap(mp->cregs, PAGE_SIZE);
-		sbus_iounmap(mp->lregs, (256 * 1024));
-		sbus_iounmap(mp->lanai, (512 * 1024));
+		of_iounmap(&op->resource[0], mp->cregs, PAGE_SIZE);
+		of_iounmap(&op->resource[0], mp->lregs, (256 * 1024));
+		of_iounmap(&op->resource[0], mp->lanai, (512 * 1024));
 	}
 
 	free_netdev(net_dev);
 
-	dev_set_drvdata(&dev->dev, NULL);
+	dev_set_drvdata(&op->dev, NULL);
 
 	return 0;
 }
 
-static struct of_device_id myri_sbus_match[] = {
+static const struct of_device_id myri_sbus_match[] = {
 	{
 		.name = "MYRICOM,mlanai",
 	},
@@ -1158,7 +1146,7 @@ static struct of_platform_driver myri_sbus_driver = {
 
 static int __init myri_sbus_init(void)
 {
-	return of_register_driver(&myri_sbus_driver, &sbus_bus_type);
+	return of_register_driver(&myri_sbus_driver, &of_bus_type);
 }
 
 static void __exit myri_sbus_exit(void)
diff --git a/drivers/net/myri_sbus.h b/drivers/net/myri_sbus.h
index 5d93fcc95d55..ff363e95d9cf 100644
--- a/drivers/net/myri_sbus.h
+++ b/drivers/net/myri_sbus.h
@@ -288,7 +288,7 @@ struct myri_eth {
 	struct myri_eeprom		eeprom;		/* Local copy of EEPROM.      */
 	unsigned int			reg_size;	/* Size of register space.    */
 	unsigned int			shmem_base;	/* Offset to shared ram.      */
-	struct sbus_dev			*myri_sdev;	/* Our SBUS device struct.    */
+	struct of_device		*myri_op;	/* Our OF device struct.    */
 };
 
 /* We use this to acquire receive skb's that we can DMA directly into. */
diff --git a/drivers/net/niu.c b/drivers/net/niu.c
index e3be81eba8a4..ebc812702903 100644
--- a/drivers/net/niu.c
+++ b/drivers/net/niu.c
@@ -9130,7 +9130,7 @@ static int __devexit niu_of_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id niu_match[] = {
+static const struct of_device_id niu_match[] = {
 	{
 		.name = "network",
 		.compatible = "SUNW,niusl",
diff --git a/drivers/net/sunbmac.c b/drivers/net/sunbmac.c
index 31e7384e312a..018d0fca9422 100644
--- a/drivers/net/sunbmac.c
+++ b/drivers/net/sunbmac.c
@@ -1,7 +1,6 @@
-/* $Id: sunbmac.c,v 1.30 2002/01/15 06:48:55 davem Exp $
- * sunbmac.c: Driver for Sparc BigMAC 100baseT ethernet adapters.
+/* sunbmac.c: Driver for Sparc BigMAC 100baseT ethernet adapters.
  *
- * Copyright (C) 1997, 1998, 1999, 2003 David S. Miller (davem@redhat.com)
+ * Copyright (C) 1997, 1998, 1999, 2003, 2008 David S. Miller (davem@davemloft.net)
  */
 
 #include <linux/module.h>
@@ -23,6 +22,9 @@
 #include <linux/etherdevice.h>
 #include <linux/skbuff.h>
 #include <linux/bitops.h>
+#include <linux/dma-mapping.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/auxio.h>
 #include <asm/byteorder.h>
@@ -32,15 +34,14 @@
 #include <asm/openprom.h>
 #include <asm/oplib.h>
 #include <asm/pgtable.h>
-#include <asm/sbus.h>
 #include <asm/system.h>
 
 #include "sunbmac.h"
 
 #define DRV_NAME	"sunbmac"
-#define DRV_VERSION	"2.0"
-#define DRV_RELDATE	"11/24/03"
-#define DRV_AUTHOR	"David S. Miller (davem@redhat.com)"
+#define DRV_VERSION	"2.1"
+#define DRV_RELDATE	"August 26, 2008"
+#define DRV_AUTHOR	"David S. Miller (davem@davemloft.net)"
 
 static char version[] =
 	DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " " DRV_AUTHOR "\n";
@@ -96,8 +97,8 @@ static int qec_global_reset(void __iomem *gregs)
 
 static void qec_init(struct bigmac *bp)
 {
+	struct of_device *qec_op = bp->qec_op;
 	void __iomem *gregs = bp->gregs;
-	struct sbus_dev *qec_sdev = bp->qec_sdev;
 	u8 bsizes = bp->bigmac_bursts;
 	u32 regval;
 
@@ -112,13 +113,13 @@ static void qec_init(struct bigmac *bp)
 	sbus_writel(GLOB_PSIZE_2048, gregs + GLOB_PSIZE);
 
 	/* All of memsize is given to bigmac. */
-	sbus_writel(qec_sdev->reg_addrs[1].reg_size,
+	sbus_writel(resource_size(&qec_op->resource[1]),
 		    gregs + GLOB_MSIZE);
 
 	/* Half to the transmitter, half to the receiver. */
-	sbus_writel(qec_sdev->reg_addrs[1].reg_size >> 1,
+	sbus_writel(resource_size(&qec_op->resource[1]) >> 1,
 		    gregs + GLOB_TSIZE);
-	sbus_writel(qec_sdev->reg_addrs[1].reg_size >> 1,
+	sbus_writel(resource_size(&qec_op->resource[1]) >> 1,
 		    gregs + GLOB_RSIZE);
 }
 
@@ -239,9 +240,10 @@ static void bigmac_init_rings(struct bigmac *bp, int from_irq)
 		skb_reserve(skb, 34);
 
 		bb->be_rxd[i].rx_addr =
-			sbus_map_single(bp->bigmac_sdev, skb->data,
-					RX_BUF_ALLOC_SIZE - 34,
-					SBUS_DMA_FROMDEVICE);
+			dma_map_single(&bp->bigmac_op->dev,
+				       skb->data,
+				       RX_BUF_ALLOC_SIZE - 34,
+				       DMA_FROM_DEVICE);
 		bb->be_rxd[i].rx_flags =
 			(RXD_OWN | ((RX_BUF_ALLOC_SIZE - 34) & RXD_LENGTH));
 	}
@@ -776,9 +778,9 @@ static void bigmac_tx(struct bigmac *bp)
 		skb = bp->tx_skbs[elem];
 		bp->enet_stats.tx_packets++;
 		bp->enet_stats.tx_bytes += skb->len;
-		sbus_unmap_single(bp->bigmac_sdev,
-				  this->tx_addr, skb->len,
-				  SBUS_DMA_TODEVICE);
+		dma_unmap_single(&bp->bigmac_op->dev,
+				 this->tx_addr, skb->len,
+				 DMA_TO_DEVICE);
 
 		DTX(("skb(%p) ", skb));
 		bp->tx_skbs[elem] = NULL;
@@ -831,18 +833,19 @@ static void bigmac_rx(struct bigmac *bp)
 				drops++;
 				goto drop_it;
 			}
-			sbus_unmap_single(bp->bigmac_sdev,
-					  this->rx_addr,
-					  RX_BUF_ALLOC_SIZE - 34,
-					  SBUS_DMA_FROMDEVICE);
+			dma_unmap_single(&bp->bigmac_op->dev,
+					 this->rx_addr,
+					 RX_BUF_ALLOC_SIZE - 34,
+					 DMA_FROM_DEVICE);
 			bp->rx_skbs[elem] = new_skb;
 			new_skb->dev = bp->dev;
 			skb_put(new_skb, ETH_FRAME_LEN);
 			skb_reserve(new_skb, 34);
-			this->rx_addr = sbus_map_single(bp->bigmac_sdev,
-							new_skb->data,
-							RX_BUF_ALLOC_SIZE - 34,
-							SBUS_DMA_FROMDEVICE);
+			this->rx_addr =
+				dma_map_single(&bp->bigmac_op->dev,
+					       new_skb->data,
+					       RX_BUF_ALLOC_SIZE - 34,
+					       DMA_FROM_DEVICE);
 			this->rx_flags =
 				(RXD_OWN | ((RX_BUF_ALLOC_SIZE - 34) & RXD_LENGTH));
 
@@ -857,13 +860,13 @@ static void bigmac_rx(struct bigmac *bp)
 			}
 			skb_reserve(copy_skb, 2);
 			skb_put(copy_skb, len);
-			sbus_dma_sync_single_for_cpu(bp->bigmac_sdev,
-						     this->rx_addr, len,
-						     SBUS_DMA_FROMDEVICE);
+			dma_sync_single_for_cpu(&bp->bigmac_op->dev,
+						this->rx_addr, len,
+						DMA_FROM_DEVICE);
 			skb_copy_to_linear_data(copy_skb, (unsigned char *)skb->data, len);
-			sbus_dma_sync_single_for_device(bp->bigmac_sdev,
-							this->rx_addr, len,
-							SBUS_DMA_FROMDEVICE);
+			dma_sync_single_for_device(&bp->bigmac_op->dev,
+						   this->rx_addr, len,
+						   DMA_FROM_DEVICE);
 
 			/* Reuse original ring buffer. */
 			this->rx_flags =
@@ -959,7 +962,8 @@ static int bigmac_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	u32 mapping;
 
 	len = skb->len;
-	mapping = sbus_map_single(bp->bigmac_sdev, skb->data, len, SBUS_DMA_TODEVICE);
+	mapping = dma_map_single(&bp->bigmac_op->dev, skb->data,
+				 len, DMA_TO_DEVICE);
 
 	/* Avoid a race... */
 	spin_lock_irq(&bp->lock);
@@ -1051,12 +1055,8 @@ static void bigmac_set_multicast(struct net_device *dev)
 /* Ethtool support... */
 static void bigmac_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
-	struct bigmac *bp = dev->priv;
-
 	strcpy(info->driver, "sunbmac");
 	strcpy(info->version, "2.0");
-	sprintf(info->bus_info, "SBUS:%d",
-		bp->qec_sdev->slot);
 }
 
 static u32 bigmac_get_link(struct net_device *dev)
@@ -1075,14 +1075,15 @@ static const struct ethtool_ops bigmac_ethtool_ops = {
 	.get_link		= bigmac_get_link,
 };
 
-static int __devinit bigmac_ether_init(struct sbus_dev *qec_sdev)
+static int __devinit bigmac_ether_init(struct of_device *op,
+				       struct of_device *qec_op)
 {
-	struct net_device *dev;
 	static int version_printed;
-	struct bigmac *bp;
+	struct net_device *dev;
 	u8 bsizes, bsizes_more;
-	int i;
 	DECLARE_MAC_BUF(mac);
+	struct bigmac *bp;
+	int i;
 
 	/* Get a new device struct for this interface. */
 	dev = alloc_etherdev(sizeof(struct bigmac));
@@ -1092,32 +1093,21 @@ static int __devinit bigmac_ether_init(struct sbus_dev *qec_sdev)
 	if (version_printed++ == 0)
 		printk(KERN_INFO "%s", version);
 
-	dev->base_addr = (long) qec_sdev;
 	for (i = 0; i < 6; i++)
 		dev->dev_addr[i] = idprom->id_ethaddr[i];
 
 	/* Setup softc, with backpointers to QEC and BigMAC SBUS device structs. */
-	bp = dev->priv;
-	bp->qec_sdev = qec_sdev;
-	bp->bigmac_sdev = qec_sdev->child;
+	bp = netdev_priv(dev);
+	bp->qec_op = qec_op;
+	bp->bigmac_op = op;
 
-	SET_NETDEV_DEV(dev, &bp->bigmac_sdev->ofdev.dev);
+	SET_NETDEV_DEV(dev, &op->dev);
 
 	spin_lock_init(&bp->lock);
 
-	/* Verify the registers we expect, are actually there. */
-	if ((bp->bigmac_sdev->num_registers != 3) ||
-	   (bp->qec_sdev->num_registers != 2)) {
-		printk(KERN_ERR "BIGMAC: Device does not have 2 and 3 regs, it has %d and %d.\n",
-		       bp->qec_sdev->num_registers,
-		       bp->bigmac_sdev->num_registers);
-		printk(KERN_ERR "BIGMAC: Would you like that for here or to go?\n");
-		goto fail_and_cleanup;
-	}
-
 	/* Map in QEC global control registers. */
-	bp->gregs = sbus_ioremap(&bp->qec_sdev->resource[0], 0,
-				 GLOB_REG_SIZE, "BigMAC QEC GLobal Regs");
+	bp->gregs = of_ioremap(&qec_op->resource[0], 0,
+			       GLOB_REG_SIZE, "BigMAC QEC GLobal Regs");
 	if (!bp->gregs) {
 		printk(KERN_ERR "BIGMAC: Cannot map QEC global registers.\n");
 		goto fail_and_cleanup;
@@ -1134,13 +1124,8 @@ static int __devinit bigmac_ether_init(struct sbus_dev *qec_sdev)
 		goto fail_and_cleanup;
 
 	/* Get supported SBUS burst sizes. */
-	bsizes = prom_getintdefault(bp->qec_sdev->prom_node,
-				    "burst-sizes",
-				    0xff);
-
-	bsizes_more = prom_getintdefault(bp->qec_sdev->bus->prom_node,
-					 "burst-sizes",
-					 0xff);
+	bsizes = of_getintprop_default(qec_op->node, "burst-sizes", 0xff);
+	bsizes_more = of_getintprop_default(qec_op->node, "burst-sizes", 0xff);
 
 	bsizes &= 0xff;
 	if (bsizes_more != 0xff)
@@ -1154,16 +1139,16 @@ static int __devinit bigmac_ether_init(struct sbus_dev *qec_sdev)
 	qec_init(bp);
 
 	/* Map in the BigMAC channel registers. */
-	bp->creg = sbus_ioremap(&bp->bigmac_sdev->resource[0], 0,
-				CREG_REG_SIZE, "BigMAC QEC Channel Regs");
+	bp->creg = of_ioremap(&op->resource[0], 0,
+			      CREG_REG_SIZE, "BigMAC QEC Channel Regs");
 	if (!bp->creg) {
 		printk(KERN_ERR "BIGMAC: Cannot map QEC channel registers.\n");
 		goto fail_and_cleanup;
 	}
 
 	/* Map in the BigMAC control registers. */
-	bp->bregs = sbus_ioremap(&bp->bigmac_sdev->resource[1], 0,
-				 BMAC_REG_SIZE, "BigMAC Primary Regs");
+	bp->bregs = of_ioremap(&op->resource[1], 0,
+			       BMAC_REG_SIZE, "BigMAC Primary Regs");
 	if (!bp->bregs) {
 		printk(KERN_ERR "BIGMAC: Cannot map BigMAC primary registers.\n");
 		goto fail_and_cleanup;
@@ -1172,8 +1157,8 @@ static int __devinit bigmac_ether_init(struct sbus_dev *qec_sdev)
 	/* Map in the BigMAC transceiver registers, this is how you poke at
 	 * the BigMAC's PHY.
 	 */
-	bp->tregs = sbus_ioremap(&bp->bigmac_sdev->resource[2], 0,
-				 TCVR_REG_SIZE, "BigMAC Transceiver Regs");
+	bp->tregs = of_ioremap(&op->resource[2], 0,
+			       TCVR_REG_SIZE, "BigMAC Transceiver Regs");
 	if (!bp->tregs) {
 		printk(KERN_ERR "BIGMAC: Cannot map BigMAC transceiver registers.\n");
 		goto fail_and_cleanup;
@@ -1183,17 +1168,17 @@ static int __devinit bigmac_ether_init(struct sbus_dev *qec_sdev)
 	bigmac_stop(bp);
 
 	/* Allocate transmit/receive descriptor DVMA block. */
-	bp->bmac_block = sbus_alloc_consistent(bp->bigmac_sdev,
-					       PAGE_SIZE,
-					       &bp->bblock_dvma);
+	bp->bmac_block = dma_alloc_coherent(&bp->bigmac_op->dev,
+					    PAGE_SIZE,
+					    &bp->bblock_dvma, GFP_ATOMIC);
 	if (bp->bmac_block == NULL || bp->bblock_dvma == 0) {
 		printk(KERN_ERR "BIGMAC: Cannot allocate consistent DMA.\n");
 		goto fail_and_cleanup;
 	}
 
 	/* Get the board revision of this BigMAC. */
-	bp->board_rev = prom_getintdefault(bp->bigmac_sdev->prom_node,
-					   "board-version", 1);
+	bp->board_rev = of_getintprop_default(bp->bigmac_op->node,
+					      "board-version", 1);
 
 	/* Init auto-negotiation timer state. */
 	init_timer(&bp->bigmac_timer);
@@ -1217,7 +1202,7 @@ static int __devinit bigmac_ether_init(struct sbus_dev *qec_sdev)
 	dev->watchdog_timeo = 5*HZ;
 
 	/* Finish net device registration. */
-	dev->irq = bp->bigmac_sdev->irqs[0];
+	dev->irq = bp->bigmac_op->irqs[0];
 	dev->dma = 0;
 
 	if (register_netdev(dev)) {
@@ -1225,7 +1210,7 @@ static int __devinit bigmac_ether_init(struct sbus_dev *qec_sdev)
 		goto fail_and_cleanup;
 	}
 
-	dev_set_drvdata(&bp->bigmac_sdev->ofdev.dev, bp);
+	dev_set_drvdata(&bp->bigmac_op->dev, bp);
 
 	printk(KERN_INFO "%s: BigMAC 100baseT Ethernet %s\n",
 	       dev->name, print_mac(mac, dev->dev_addr));
@@ -1236,66 +1221,67 @@ fail_and_cleanup:
 	/* Something went wrong, undo whatever we did so far. */
 	/* Free register mappings if any. */
 	if (bp->gregs)
-		sbus_iounmap(bp->gregs, GLOB_REG_SIZE);
+		of_iounmap(&qec_op->resource[0], bp->gregs, GLOB_REG_SIZE);
 	if (bp->creg)
-		sbus_iounmap(bp->creg, CREG_REG_SIZE);
+		of_iounmap(&op->resource[0], bp->creg, CREG_REG_SIZE);
 	if (bp->bregs)
-		sbus_iounmap(bp->bregs, BMAC_REG_SIZE);
+		of_iounmap(&op->resource[1], bp->bregs, BMAC_REG_SIZE);
 	if (bp->tregs)
-		sbus_iounmap(bp->tregs, TCVR_REG_SIZE);
+		of_iounmap(&op->resource[2], bp->tregs, TCVR_REG_SIZE);
 
 	if (bp->bmac_block)
-		sbus_free_consistent(bp->bigmac_sdev,
-				     PAGE_SIZE,
-				     bp->bmac_block,
-				     bp->bblock_dvma);
+		dma_free_coherent(&bp->bigmac_op->dev,
+				  PAGE_SIZE,
+				  bp->bmac_block,
+				  bp->bblock_dvma);
 
 	/* This also frees the co-located 'dev->priv' */
 	free_netdev(dev);
 	return -ENODEV;
 }
 
-/* QEC can be the parent of either QuadEthernet or
- * a BigMAC.  We want the latter.
+/* QEC can be the parent of either QuadEthernet or a BigMAC.  We want
+ * the latter.
  */
-static int __devinit bigmac_sbus_probe(struct of_device *dev, const struct of_device_id *match)
+static int __devinit bigmac_sbus_probe(struct of_device *op,
+				       const struct of_device_id *match)
 {
-	struct sbus_dev *sdev = to_sbus_device(&dev->dev);
-	struct device_node *dp = dev->node;
+	struct device *parent = op->dev.parent;
+	struct of_device *qec_op;
 
-	if (!strcmp(dp->name, "be"))
-		sdev = sdev->parent;
+	qec_op = to_of_device(parent);
 
-	return bigmac_ether_init(sdev);
+	return bigmac_ether_init(op, qec_op);
 }
 
-static int __devexit bigmac_sbus_remove(struct of_device *dev)
+static int __devexit bigmac_sbus_remove(struct of_device *op)
 {
-	struct bigmac *bp = dev_get_drvdata(&dev->dev);
+	struct bigmac *bp = dev_get_drvdata(&op->dev);
+	struct device *parent = op->dev.parent;
 	struct net_device *net_dev = bp->dev;
+	struct of_device *qec_op;
+
+	qec_op = to_of_device(parent);
 
 	unregister_netdev(net_dev);
 
-	sbus_iounmap(bp->gregs, GLOB_REG_SIZE);
-	sbus_iounmap(bp->creg, CREG_REG_SIZE);
-	sbus_iounmap(bp->bregs, BMAC_REG_SIZE);
-	sbus_iounmap(bp->tregs, TCVR_REG_SIZE);
-	sbus_free_consistent(bp->bigmac_sdev,
-			     PAGE_SIZE,
-			     bp->bmac_block,
-			     bp->bblock_dvma);
+	of_iounmap(&qec_op->resource[0], bp->gregs, GLOB_REG_SIZE);
+	of_iounmap(&op->resource[0], bp->creg, CREG_REG_SIZE);
+	of_iounmap(&op->resource[1], bp->bregs, BMAC_REG_SIZE);
+	of_iounmap(&op->resource[2], bp->tregs, TCVR_REG_SIZE);
+	dma_free_coherent(&op->dev,
+			  PAGE_SIZE,
+			  bp->bmac_block,
+			  bp->bblock_dvma);
 
 	free_netdev(net_dev);
 
-	dev_set_drvdata(&dev->dev, NULL);
+	dev_set_drvdata(&op->dev, NULL);
 
 	return 0;
 }
 
-static struct of_device_id bigmac_sbus_match[] = {
-	{
-		.name = "qec",
-	},
+static const struct of_device_id bigmac_sbus_match[] = {
 	{
 		.name = "be",
 	},
@@ -1313,7 +1299,7 @@ static struct of_platform_driver bigmac_sbus_driver = {
 
 static int __init bigmac_init(void)
 {
-	return of_register_driver(&bigmac_sbus_driver, &sbus_bus_type);
+	return of_register_driver(&bigmac_sbus_driver, &of_bus_type);
 }
 
 static void __exit bigmac_exit(void)
diff --git a/drivers/net/sunbmac.h b/drivers/net/sunbmac.h
index b563d3c2993e..8840bc0b840b 100644
--- a/drivers/net/sunbmac.h
+++ b/drivers/net/sunbmac.h
@@ -329,8 +329,8 @@ struct bigmac {
 	unsigned int		timer_ticks;
 
 	struct net_device_stats	enet_stats;
-	struct sbus_dev		*qec_sdev;
-	struct sbus_dev		*bigmac_sdev;
+	struct of_device	*qec_op;
+	struct of_device	*bigmac_op;
 	struct net_device	*dev;
 };
 
diff --git a/drivers/net/sunhme.c b/drivers/net/sunhme.c
index b79d5f018f79..f1ebeb5f65b2 100644
--- a/drivers/net/sunhme.c
+++ b/drivers/net/sunhme.c
@@ -3,7 +3,7 @@
  *           "Happy Meal Ethernet" found on SunSwift SBUS cards.
  *
  * Copyright (C) 1996, 1998, 1999, 2002, 2003,
-		 2006 David S. Miller (davem@davemloft.net)
+ *		2006, 2008 David S. Miller (davem@davemloft.net)
  *
  * Changes :
  * 2000/11/11 Willy Tarreau <willy AT meta-x.org>
@@ -34,6 +34,7 @@
 #include <linux/skbuff.h>
 #include <linux/mm.h>
 #include <linux/bitops.h>
+#include <linux/dma-mapping.h>
 
 #include <asm/system.h>
 #include <asm/io.h>
@@ -41,8 +42,9 @@
 #include <asm/byteorder.h>
 
 #ifdef CONFIG_SPARC
+#include <linux/of.h>
+#include <linux/of_device.h>
 #include <asm/idprom.h>
-#include <asm/sbus.h>
 #include <asm/openprom.h>
 #include <asm/oplib.h>
 #include <asm/prom.h>
@@ -60,8 +62,8 @@
 #include "sunhme.h"
 
 #define DRV_NAME	"sunhme"
-#define DRV_VERSION	"3.00"
-#define DRV_RELDATE	"June 23, 2006"
+#define DRV_VERSION	"3.10"
+#define DRV_RELDATE	"August 26, 2008"
 #define DRV_AUTHOR	"David S. Miller (davem@davemloft.net)"
 
 static char version[] =
@@ -251,13 +253,13 @@ static u32 pci_hme_read_desc32(hme32 *p)
 #define hme_read_desc32(__hp, __p) \
 	((__hp)->read_desc32(__p))
 #define hme_dma_map(__hp, __ptr, __size, __dir) \
-	((__hp)->dma_map((__hp)->happy_dev, (__ptr), (__size), (__dir)))
+	((__hp)->dma_map((__hp)->dma_dev, (__ptr), (__size), (__dir)))
 #define hme_dma_unmap(__hp, __addr, __size, __dir) \
-	((__hp)->dma_unmap((__hp)->happy_dev, (__addr), (__size), (__dir)))
+	((__hp)->dma_unmap((__hp)->dma_dev, (__addr), (__size), (__dir)))
 #define hme_dma_sync_for_cpu(__hp, __addr, __size, __dir) \
-	((__hp)->dma_sync_for_cpu((__hp)->happy_dev, (__addr), (__size), (__dir)))
+	((__hp)->dma_sync_for_cpu((__hp)->dma_dev, (__addr), (__size), (__dir)))
 #define hme_dma_sync_for_device(__hp, __addr, __size, __dir) \
-	((__hp)->dma_sync_for_device((__hp)->happy_dev, (__addr), (__size), (__dir)))
+	((__hp)->dma_sync_for_device((__hp)->dma_dev, (__addr), (__size), (__dir)))
 #else
 #ifdef CONFIG_SBUS
 /* SBUS only compilation */
@@ -277,13 +279,13 @@ do {	(__txd)->tx_addr = (__force hme32)(u32)(__addr); \
 } while(0)
 #define hme_read_desc32(__hp, __p)	((__force u32)(hme32)*(__p))
 #define hme_dma_map(__hp, __ptr, __size, __dir) \
-	sbus_map_single((__hp)->happy_dev, (__ptr), (__size), (__dir))
+	dma_map_single((__hp)->dma_dev, (__ptr), (__size), (__dir))
 #define hme_dma_unmap(__hp, __addr, __size, __dir) \
-	sbus_unmap_single((__hp)->happy_dev, (__addr), (__size), (__dir))
+	dma_unmap_single((__hp)->dma_dev, (__addr), (__size), (__dir))
 #define hme_dma_sync_for_cpu(__hp, __addr, __size, __dir) \
-	sbus_dma_sync_single_for_cpu((__hp)->happy_dev, (__addr), (__size), (__dir))
+	dma_dma_sync_single_for_cpu((__hp)->dma_dev, (__addr), (__size), (__dir))
 #define hme_dma_sync_for_device(__hp, __addr, __size, __dir) \
-	sbus_dma_sync_single_for_device((__hp)->happy_dev, (__addr), (__size), (__dir))
+	dma_dma_sync_single_for_device((__hp)->dma_dev, (__addr), (__size), (__dir))
 #else
 /* PCI only compilation */
 #define hme_write32(__hp, __reg, __val) \
@@ -305,36 +307,17 @@ static inline u32 hme_read_desc32(struct happy_meal *hp, hme32 *p)
 	return le32_to_cpup((__le32 *)p);
 }
 #define hme_dma_map(__hp, __ptr, __size, __dir) \
-	pci_map_single((__hp)->happy_dev, (__ptr), (__size), (__dir))
+	pci_map_single((__hp)->dma_dev, (__ptr), (__size), (__dir))
 #define hme_dma_unmap(__hp, __addr, __size, __dir) \
-	pci_unmap_single((__hp)->happy_dev, (__addr), (__size), (__dir))
+	pci_unmap_single((__hp)->dma_dev, (__addr), (__size), (__dir))
 #define hme_dma_sync_for_cpu(__hp, __addr, __size, __dir) \
-	pci_dma_sync_single_for_cpu((__hp)->happy_dev, (__addr), (__size), (__dir))
+	pci_dma_sync_single_for_cpu((__hp)->dma_dev, (__addr), (__size), (__dir))
 #define hme_dma_sync_for_device(__hp, __addr, __size, __dir) \
-	pci_dma_sync_single_for_device((__hp)->happy_dev, (__addr), (__size), (__dir))
+	pci_dma_sync_single_for_device((__hp)->dma_dev, (__addr), (__size), (__dir))
 #endif
 #endif
 
 
-#ifdef SBUS_DMA_BIDIRECTIONAL
-#	define DMA_BIDIRECTIONAL	SBUS_DMA_BIDIRECTIONAL
-#else
-#	define DMA_BIDIRECTIONAL	0
-#endif
-
-#ifdef SBUS_DMA_FROMDEVICE
-#	define DMA_FROMDEVICE		SBUS_DMA_FROMDEVICE
-#else
-#	define DMA_TODEVICE		1
-#endif
-
-#ifdef SBUS_DMA_TODEVICE
-#	define DMA_TODEVICE		SBUS_DMA_TODEVICE
-#else
-#	define DMA_FROMDEVICE		2
-#endif
-
-
 /* Oh yes, the MIF BitBang is mighty fun to program.  BitBucket is more like it. */
 static void BB_PUT_BIT(struct happy_meal *hp, void __iomem *tregs, int bit)
 {
@@ -1224,7 +1207,8 @@ static void happy_meal_clean_rings(struct happy_meal *hp)
 
 			rxd = &hp->happy_block->happy_meal_rxd[i];
 			dma_addr = hme_read_desc32(hp, &rxd->rx_addr);
-			hme_dma_unmap(hp, dma_addr, RX_BUF_ALLOC_SIZE, DMA_FROMDEVICE);
+			dma_unmap_single(hp->dma_dev, dma_addr,
+					 RX_BUF_ALLOC_SIZE, DMA_FROM_DEVICE);
 			dev_kfree_skb_any(skb);
 			hp->rx_skbs[i] = NULL;
 		}
@@ -1242,10 +1226,10 @@ static void happy_meal_clean_rings(struct happy_meal *hp)
 			for (frag = 0; frag <= skb_shinfo(skb)->nr_frags; frag++) {
 				txd = &hp->happy_block->happy_meal_txd[i];
 				dma_addr = hme_read_desc32(hp, &txd->tx_addr);
-				hme_dma_unmap(hp, dma_addr,
-					      (hme_read_desc32(hp, &txd->tx_flags)
-					       & TXFLAG_SIZE),
-					      DMA_TODEVICE);
+				dma_unmap_single(hp->dma_dev, dma_addr,
+						 (hme_read_desc32(hp, &txd->tx_flags)
+						  & TXFLAG_SIZE),
+						 DMA_TO_DEVICE);
 
 				if (frag != skb_shinfo(skb)->nr_frags)
 					i++;
@@ -1287,7 +1271,8 @@ static void happy_meal_init_rings(struct happy_meal *hp)
 		skb_put(skb, (ETH_FRAME_LEN + RX_OFFSET + 4));
 		hme_write_rxd(hp, &hb->happy_meal_rxd[i],
 			      (RXFLAG_OWN | ((RX_BUF_ALLOC_SIZE - RX_OFFSET) << 16)),
-			      hme_dma_map(hp, skb->data, RX_BUF_ALLOC_SIZE, DMA_FROMDEVICE));
+			      dma_map_single(hp->dma_dev, skb->data, RX_BUF_ALLOC_SIZE,
+					     DMA_FROM_DEVICE));
 		skb_reserve(skb, RX_OFFSET);
 	}
 
@@ -1593,7 +1578,7 @@ static int happy_meal_init(struct happy_meal *hp)
 	if ((hp->happy_bursts & DMA_BURST64) &&
 	    ((hp->happy_flags & HFLAG_PCI) != 0
 #ifdef CONFIG_SBUS
-	     || sbus_can_burst64(hp->happy_dev)
+	     || sbus_can_burst64()
 #endif
 	     || 0)) {
 		u32 gcfg = GREG_CFG_BURST64;
@@ -1603,11 +1588,13 @@ static int happy_meal_init(struct happy_meal *hp)
 		 * do not.  -DaveM
 		 */
 #ifdef CONFIG_SBUS
-		if ((hp->happy_flags & HFLAG_PCI) == 0 &&
-		    sbus_can_dma_64bit(hp->happy_dev)) {
-			sbus_set_sbus64(hp->happy_dev,
-					hp->happy_bursts);
-			gcfg |= GREG_CFG_64BIT;
+		if ((hp->happy_flags & HFLAG_PCI) == 0) {
+			struct of_device *op = hp->happy_dev;
+			if (sbus_can_dma_64bit()) {
+				sbus_set_sbus64(&op->dev,
+						hp->happy_bursts);
+				gcfg |= GREG_CFG_64BIT;
+			}
 		}
 #endif
 
@@ -1966,7 +1953,7 @@ static void happy_meal_tx(struct happy_meal *hp)
 			dma_len = hme_read_desc32(hp, &this->tx_flags);
 
 			dma_len &= TXFLAG_SIZE;
-			hme_dma_unmap(hp, dma_addr, dma_len, DMA_TODEVICE);
+			dma_unmap_single(hp->dma_dev, dma_addr, dma_len, DMA_TO_DEVICE);
 
 			elem = NEXT_TX(elem);
 			this = &txbase[elem];
@@ -2044,13 +2031,14 @@ static void happy_meal_rx(struct happy_meal *hp, struct net_device *dev)
 				drops++;
 				goto drop_it;
 			}
-			hme_dma_unmap(hp, dma_addr, RX_BUF_ALLOC_SIZE, DMA_FROMDEVICE);
+			dma_unmap_single(hp->dma_dev, dma_addr, RX_BUF_ALLOC_SIZE, DMA_FROM_DEVICE);
 			hp->rx_skbs[elem] = new_skb;
 			new_skb->dev = dev;
 			skb_put(new_skb, (ETH_FRAME_LEN + RX_OFFSET + 4));
 			hme_write_rxd(hp, this,
 				      (RXFLAG_OWN|((RX_BUF_ALLOC_SIZE-RX_OFFSET)<<16)),
-				      hme_dma_map(hp, new_skb->data, RX_BUF_ALLOC_SIZE, DMA_FROMDEVICE));
+				      dma_map_single(hp->dma_dev, new_skb->data, RX_BUF_ALLOC_SIZE,
+						     DMA_FROM_DEVICE));
 			skb_reserve(new_skb, RX_OFFSET);
 
 			/* Trim the original skb for the netif. */
@@ -2065,10 +2053,9 @@ static void happy_meal_rx(struct happy_meal *hp, struct net_device *dev)
 
 			skb_reserve(copy_skb, 2);
 			skb_put(copy_skb, len);
-			hme_dma_sync_for_cpu(hp, dma_addr, len, DMA_FROMDEVICE);
+			dma_sync_single_for_cpu(hp->dma_dev, dma_addr, len, DMA_FROM_DEVICE);
 			skb_copy_from_linear_data(skb, copy_skb->data, len);
-			hme_dma_sync_for_device(hp, dma_addr, len, DMA_FROMDEVICE);
-
+			dma_sync_single_for_device(hp->dma_dev, dma_addr, len, DMA_FROM_DEVICE);
 			/* Reuse original ring buffer. */
 			hme_write_rxd(hp, this,
 				      (RXFLAG_OWN|((RX_BUF_ALLOC_SIZE-RX_OFFSET)<<16)),
@@ -2300,7 +2287,7 @@ static int happy_meal_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		u32 mapping, len;
 
 		len = skb->len;
-		mapping = hme_dma_map(hp, skb->data, len, DMA_TODEVICE);
+		mapping = dma_map_single(hp->dma_dev, skb->data, len, DMA_TO_DEVICE);
 		tx_flags |= (TXFLAG_SOP | TXFLAG_EOP);
 		hme_write_txd(hp, &hp->happy_block->happy_meal_txd[entry],
 			      (tx_flags | (len & TXFLAG_SIZE)),
@@ -2314,7 +2301,8 @@ static int happy_meal_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		 * Otherwise we could race with the device.
 		 */
 		first_len = skb_headlen(skb);
-		first_mapping = hme_dma_map(hp, skb->data, first_len, DMA_TODEVICE);
+		first_mapping = dma_map_single(hp->dma_dev, skb->data, first_len,
+					       DMA_TO_DEVICE);
 		entry = NEXT_TX(entry);
 
 		for (frag = 0; frag < skb_shinfo(skb)->nr_frags; frag++) {
@@ -2322,10 +2310,9 @@ static int happy_meal_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			u32 len, mapping, this_txflags;
 
 			len = this_frag->size;
-			mapping = hme_dma_map(hp,
-					      ((void *) page_address(this_frag->page) +
-					       this_frag->page_offset),
-					      len, DMA_TODEVICE);
+			mapping = dma_map_page(hp->dma_dev, this_frag->page,
+					       this_frag->page_offset, len,
+					       DMA_TO_DEVICE);
 			this_txflags = tx_flags;
 			if (frag == skb_shinfo(skb)->nr_frags - 1)
 				this_txflags |= TXFLAG_EOP;
@@ -2493,9 +2480,12 @@ static void hme_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info
 	}
 #ifdef CONFIG_SBUS
 	else {
-		struct sbus_dev *sdev = hp->happy_dev;
-		sprintf(info->bus_info, "SBUS:%d",
-			sdev->slot);
+		const struct linux_prom_registers *regs;
+		struct of_device *op = hp->happy_dev;
+		regs = of_get_property(op->node, "regs", NULL);
+		if (regs)
+			sprintf(info->bus_info, "SBUS:%d",
+				regs->which_io);
 	}
 #endif
 }
@@ -2521,63 +2511,21 @@ static const struct ethtool_ops hme_ethtool_ops = {
 static int hme_version_printed;
 
 #ifdef CONFIG_SBUS
-void __devinit quattro_get_ranges(struct quattro *qp)
-{
-	struct sbus_dev *sdev = qp->quattro_dev;
-	int err;
-
-	err = prom_getproperty(sdev->prom_node,
-			       "ranges",
-			       (char *)&qp->ranges[0],
-			       sizeof(qp->ranges));
-	if (err == 0 || err == -1) {
-		qp->nranges = 0;
-		return;
-	}
-	qp->nranges = (err / sizeof(struct linux_prom_ranges));
-}
-
-static void __devinit quattro_apply_ranges(struct quattro *qp, struct happy_meal *hp)
-{
-	struct sbus_dev *sdev = hp->happy_dev;
-	int rng;
-
-	for (rng = 0; rng < qp->nranges; rng++) {
-		struct linux_prom_ranges *rngp = &qp->ranges[rng];
-		int reg;
-
-		for (reg = 0; reg < 5; reg++) {
-			if (sdev->reg_addrs[reg].which_io ==
-			    rngp->ot_child_space)
-				break;
-		}
-		if (reg == 5)
-			continue;
-
-		sdev->reg_addrs[reg].which_io = rngp->ot_parent_space;
-		sdev->reg_addrs[reg].phys_addr += rngp->ot_parent_base;
-	}
-}
-
 /* Given a happy meal sbus device, find it's quattro parent.
  * If none exist, allocate and return a new one.
  *
  * Return NULL on failure.
  */
-static struct quattro * __devinit quattro_sbus_find(struct sbus_dev *goal_sdev)
+static struct quattro * __devinit quattro_sbus_find(struct of_device *child)
 {
-	struct sbus_dev *sdev;
+	struct device *parent = child->dev.parent;
+	struct of_device *op;
 	struct quattro *qp;
-	int i;
 
-	for (qp = qfe_sbus_list; qp != NULL; qp = qp->next) {
-		for (i = 0, sdev = qp->quattro_dev;
-		     (sdev != NULL) && (i < 4);
-		     sdev = sdev->next, i++) {
-			if (sdev == goal_sdev)
-				return qp;
-		}
-	}
+	op = to_of_device(parent);
+	qp = dev_get_drvdata(&op->dev);
+	if (qp)
+		return qp;
 
 	qp = kmalloc(sizeof(struct quattro), GFP_KERNEL);
 	if (qp != NULL) {
@@ -2586,10 +2534,11 @@ static struct quattro * __devinit quattro_sbus_find(struct sbus_dev *goal_sdev)
 		for (i = 0; i < 4; i++)
 			qp->happy_meals[i] = NULL;
 
-		qp->quattro_dev = goal_sdev;
+		qp->quattro_dev = child;
 		qp->next = qfe_sbus_list;
 		qfe_sbus_list = qp;
-		quattro_get_ranges(qp);
+
+		dev_set_drvdata(&op->dev, qp);
 	}
 	return qp;
 }
@@ -2602,10 +2551,10 @@ static void __init quattro_sbus_register_irqs(void)
 	struct quattro *qp;
 
 	for (qp = qfe_sbus_list; qp != NULL; qp = qp->next) {
-		struct sbus_dev *sdev = qp->quattro_dev;
+		struct of_device *op = qp->quattro_dev;
 		int err;
 
-		err = request_irq(sdev->irqs[0],
+		err = request_irq(op->irqs[0],
 				  quattro_sbus_interrupt,
 				  IRQF_SHARED, "Quattro",
 				  qp);
@@ -2621,9 +2570,9 @@ static void quattro_sbus_free_irqs(void)
 	struct quattro *qp;
 
 	for (qp = qfe_sbus_list; qp != NULL; qp = qp->next) {
-		struct sbus_dev *sdev = qp->quattro_dev;
+		struct of_device *op = qp->quattro_dev;
 
-		free_irq(sdev->irqs[0], qp);
+		free_irq(op->irqs[0], qp);
 	}
 }
 #endif /* CONFIG_SBUS */
@@ -2660,9 +2609,9 @@ static struct quattro * __devinit quattro_pci_find(struct pci_dev *pdev)
 #endif /* CONFIG_PCI */
 
 #ifdef CONFIG_SBUS
-static int __devinit happy_meal_sbus_probe_one(struct sbus_dev *sdev, int is_qfe)
+static int __devinit happy_meal_sbus_probe_one(struct of_device *op, int is_qfe)
 {
-	struct device_node *dp = sdev->ofdev.node;
+	struct device_node *dp = op->node, *sbus_dp;
 	struct quattro *qp = NULL;
 	struct happy_meal *hp;
 	struct net_device *dev;
@@ -2671,7 +2620,7 @@ static int __devinit happy_meal_sbus_probe_one(struct sbus_dev *sdev, int is_qfe
 	DECLARE_MAC_BUF(mac);
 
 	if (is_qfe) {
-		qp = quattro_sbus_find(sdev);
+		qp = quattro_sbus_find(op);
 		if (qp == NULL)
 			goto err_out;
 		for (qfe_slot = 0; qfe_slot < 4; qfe_slot++)
@@ -2685,7 +2634,7 @@ static int __devinit happy_meal_sbus_probe_one(struct sbus_dev *sdev, int is_qfe
 	dev = alloc_etherdev(sizeof(struct happy_meal));
 	if (!dev)
 		goto err_out;
-	SET_NETDEV_DEV(dev, &sdev->ofdev.dev);
+	SET_NETDEV_DEV(dev, &op->dev);
 
 	if (hme_version_printed++ == 0)
 		printk(KERN_INFO "%s", version);
@@ -2713,56 +2662,50 @@ static int __devinit happy_meal_sbus_probe_one(struct sbus_dev *sdev, int is_qfe
 			memcpy(dev->dev_addr, idprom->id_ethaddr, 6);
 	}
 
-	hp = dev->priv;
+	hp = netdev_priv(dev);
 
-	hp->happy_dev = sdev;
+	hp->happy_dev = op;
+	hp->dma_dev = &op->dev;
 
 	spin_lock_init(&hp->happy_lock);
 
 	err = -ENODEV;
-	if (sdev->num_registers != 5) {
-		printk(KERN_ERR "happymeal: Device needs 5 regs, has %d.\n",
-		       sdev->num_registers);
-		goto err_out_free_netdev;
-	}
-
 	if (qp != NULL) {
 		hp->qfe_parent = qp;
 		hp->qfe_ent = qfe_slot;
 		qp->happy_meals[qfe_slot] = dev;
-		quattro_apply_ranges(qp, hp);
 	}
 
-	hp->gregs = sbus_ioremap(&sdev->resource[0], 0,
-				 GREG_REG_SIZE, "HME Global Regs");
+	hp->gregs = of_ioremap(&op->resource[0], 0,
+			       GREG_REG_SIZE, "HME Global Regs");
 	if (!hp->gregs) {
 		printk(KERN_ERR "happymeal: Cannot map global registers.\n");
 		goto err_out_free_netdev;
 	}
 
-	hp->etxregs = sbus_ioremap(&sdev->resource[1], 0,
-				   ETX_REG_SIZE, "HME TX Regs");
+	hp->etxregs = of_ioremap(&op->resource[1], 0,
+				 ETX_REG_SIZE, "HME TX Regs");
 	if (!hp->etxregs) {
 		printk(KERN_ERR "happymeal: Cannot map MAC TX registers.\n");
 		goto err_out_iounmap;
 	}
 
-	hp->erxregs = sbus_ioremap(&sdev->resource[2], 0,
-				   ERX_REG_SIZE, "HME RX Regs");
+	hp->erxregs = of_ioremap(&op->resource[2], 0,
+				 ERX_REG_SIZE, "HME RX Regs");
 	if (!hp->erxregs) {
 		printk(KERN_ERR "happymeal: Cannot map MAC RX registers.\n");
 		goto err_out_iounmap;
 	}
 
-	hp->bigmacregs = sbus_ioremap(&sdev->resource[3], 0,
-				      BMAC_REG_SIZE, "HME BIGMAC Regs");
+	hp->bigmacregs = of_ioremap(&op->resource[3], 0,
+				    BMAC_REG_SIZE, "HME BIGMAC Regs");
 	if (!hp->bigmacregs) {
 		printk(KERN_ERR "happymeal: Cannot map BIGMAC registers.\n");
 		goto err_out_iounmap;
 	}
 
-	hp->tcvregs = sbus_ioremap(&sdev->resource[4], 0,
-				   TCVR_REG_SIZE, "HME Tranceiver Regs");
+	hp->tcvregs = of_ioremap(&op->resource[4], 0,
+				 TCVR_REG_SIZE, "HME Tranceiver Regs");
 	if (!hp->tcvregs) {
 		printk(KERN_ERR "happymeal: Cannot map TCVR registers.\n");
 		goto err_out_iounmap;
@@ -2781,13 +2724,18 @@ static int __devinit happy_meal_sbus_probe_one(struct sbus_dev *sdev, int is_qfe
 	if (qp != NULL)
 		hp->happy_flags |= HFLAG_QUATTRO;
 
+	sbus_dp = to_of_device(op->dev.parent)->node;
+	if (is_qfe)
+		sbus_dp = to_of_device(op->dev.parent->parent)->node;
+
 	/* Get the supported DVMA burst sizes from our Happy SBUS. */
-	hp->happy_bursts = of_getintprop_default(sdev->bus->ofdev.node,
+	hp->happy_bursts = of_getintprop_default(sbus_dp,
 						 "burst-sizes", 0x00);
 
-	hp->happy_block = sbus_alloc_consistent(hp->happy_dev,
-						PAGE_SIZE,
-						&hp->hblock_dvma);
+	hp->happy_block = dma_alloc_coherent(hp->dma_dev,
+					     PAGE_SIZE,
+					     &hp->hblock_dvma,
+					     GFP_ATOMIC);
 	err = -ENOMEM;
 	if (!hp->happy_block) {
 		printk(KERN_ERR "happymeal: Cannot allocate descriptors.\n");
@@ -2816,19 +2764,13 @@ static int __devinit happy_meal_sbus_probe_one(struct sbus_dev *sdev, int is_qfe
 	/* Happy Meal can do it all... */
 	dev->features |= NETIF_F_SG | NETIF_F_HW_CSUM;
 
-	dev->irq = sdev->irqs[0];
+	dev->irq = op->irqs[0];
 
 #if defined(CONFIG_SBUS) && defined(CONFIG_PCI)
-	/* Hook up PCI register/dma accessors. */
+	/* Hook up SBUS register/descriptor accessors. */
 	hp->read_desc32 = sbus_hme_read_desc32;
 	hp->write_txd = sbus_hme_write_txd;
 	hp->write_rxd = sbus_hme_write_rxd;
-	hp->dma_map = (u32 (*)(void *, void *, long, int))sbus_map_single;
-	hp->dma_unmap = (void (*)(void *, u32, long, int))sbus_unmap_single;
-	hp->dma_sync_for_cpu = (void (*)(void *, u32, long, int))
-		sbus_dma_sync_single_for_cpu;
-	hp->dma_sync_for_device = (void (*)(void *, u32, long, int))
-		sbus_dma_sync_single_for_device;
 	hp->read32 = sbus_hme_read32;
 	hp->write32 = sbus_hme_write32;
 #endif
@@ -2843,10 +2785,10 @@ static int __devinit happy_meal_sbus_probe_one(struct sbus_dev *sdev, int is_qfe
 	if (register_netdev(hp->dev)) {
 		printk(KERN_ERR "happymeal: Cannot register net device, "
 		       "aborting.\n");
-		goto err_out_free_consistent;
+		goto err_out_free_coherent;
 	}
 
-	dev_set_drvdata(&sdev->ofdev.dev, hp);
+	dev_set_drvdata(&op->dev, hp);
 
 	if (qfe_slot != -1)
 		printk(KERN_INFO "%s: Quattro HME slot %d (SBUS) 10/100baseT Ethernet ",
@@ -2859,23 +2801,23 @@ static int __devinit happy_meal_sbus_probe_one(struct sbus_dev *sdev, int is_qfe
 
 	return 0;
 
-err_out_free_consistent:
-	sbus_free_consistent(hp->happy_dev,
-			     PAGE_SIZE,
-			     hp->happy_block,
-			     hp->hblock_dvma);
+err_out_free_coherent:
+	dma_free_coherent(hp->dma_dev,
+			  PAGE_SIZE,
+			  hp->happy_block,
+			  hp->hblock_dvma);
 
 err_out_iounmap:
 	if (hp->gregs)
-		sbus_iounmap(hp->gregs, GREG_REG_SIZE);
+		of_iounmap(&op->resource[0], hp->gregs, GREG_REG_SIZE);
 	if (hp->etxregs)
-		sbus_iounmap(hp->etxregs, ETX_REG_SIZE);
+		of_iounmap(&op->resource[1], hp->etxregs, ETX_REG_SIZE);
 	if (hp->erxregs)
-		sbus_iounmap(hp->erxregs, ERX_REG_SIZE);
+		of_iounmap(&op->resource[2], hp->erxregs, ERX_REG_SIZE);
 	if (hp->bigmacregs)
-		sbus_iounmap(hp->bigmacregs, BMAC_REG_SIZE);
+		of_iounmap(&op->resource[3], hp->bigmacregs, BMAC_REG_SIZE);
 	if (hp->tcvregs)
-		sbus_iounmap(hp->tcvregs, TCVR_REG_SIZE);
+		of_iounmap(&op->resource[4], hp->tcvregs, TCVR_REG_SIZE);
 
 err_out_free_netdev:
 	free_netdev(dev);
@@ -3035,6 +2977,7 @@ static int __devinit happy_meal_pci_probe(struct pci_dev *pdev,
 	memset(hp, 0, sizeof(*hp));
 
 	hp->happy_dev = pdev;
+	hp->dma_dev = &pdev->dev;
 
 	spin_lock_init(&hp->happy_lock);
 
@@ -3121,7 +3064,7 @@ static int __devinit happy_meal_pci_probe(struct pci_dev *pdev,
 #endif
 
 	hp->happy_block = (struct hmeal_init_block *)
-		pci_alloc_consistent(pdev, PAGE_SIZE, &hp->hblock_dvma);
+		dma_alloc_coherent(&pdev->dev, PAGE_SIZE, &hp->hblock_dvma, GFP_KERNEL);
 
 	err = -ENODEV;
 	if (!hp->happy_block) {
@@ -3151,16 +3094,10 @@ static int __devinit happy_meal_pci_probe(struct pci_dev *pdev,
 	dev->features |= NETIF_F_SG | NETIF_F_HW_CSUM;
 
 #if defined(CONFIG_SBUS) && defined(CONFIG_PCI)
-	/* Hook up PCI register/dma accessors. */
+	/* Hook up PCI register/descriptor accessors. */
 	hp->read_desc32 = pci_hme_read_desc32;
 	hp->write_txd = pci_hme_write_txd;
 	hp->write_rxd = pci_hme_write_rxd;
-	hp->dma_map = (u32 (*)(void *, void *, long, int))pci_map_single;
-	hp->dma_unmap = (void (*)(void *, u32, long, int))pci_unmap_single;
-	hp->dma_sync_for_cpu = (void (*)(void *, u32, long, int))
-		pci_dma_sync_single_for_cpu;
-	hp->dma_sync_for_device = (void (*)(void *, u32, long, int))
-		pci_dma_sync_single_for_device;
 	hp->read32 = pci_hme_read32;
 	hp->write32 = pci_hme_write32;
 #endif
@@ -3231,10 +3168,8 @@ static void __devexit happy_meal_pci_remove(struct pci_dev *pdev)
 
 	unregister_netdev(net_dev);
 
-	pci_free_consistent(hp->happy_dev,
-			    PAGE_SIZE,
-			    hp->happy_block,
-			    hp->hblock_dvma);
+	dma_free_coherent(hp->dma_dev, PAGE_SIZE,
+			  hp->happy_block, hp->hblock_dvma);
 	iounmap(hp->gregs);
 	pci_release_regions(hp->happy_dev);
 
@@ -3279,46 +3214,45 @@ static void happy_meal_pci_exit(void)
 #endif
 
 #ifdef CONFIG_SBUS
-static int __devinit hme_sbus_probe(struct of_device *dev, const struct of_device_id *match)
+static int __devinit hme_sbus_probe(struct of_device *op, const struct of_device_id *match)
 {
-	struct sbus_dev *sdev = to_sbus_device(&dev->dev);
-	struct device_node *dp = dev->node;
+	struct device_node *dp = op->node;
 	const char *model = of_get_property(dp, "model", NULL);
 	int is_qfe = (match->data != NULL);
 
 	if (!is_qfe && model && !strcmp(model, "SUNW,sbus-qfe"))
 		is_qfe = 1;
 
-	return happy_meal_sbus_probe_one(sdev, is_qfe);
+	return happy_meal_sbus_probe_one(op, is_qfe);
 }
 
-static int __devexit hme_sbus_remove(struct of_device *dev)
+static int __devexit hme_sbus_remove(struct of_device *op)
 {
-	struct happy_meal *hp = dev_get_drvdata(&dev->dev);
+	struct happy_meal *hp = dev_get_drvdata(&op->dev);
 	struct net_device *net_dev = hp->dev;
 
 	unregister_netdev(net_dev);
 
 	/* XXX qfe parent interrupt... */
 
-	sbus_iounmap(hp->gregs, GREG_REG_SIZE);
-	sbus_iounmap(hp->etxregs, ETX_REG_SIZE);
-	sbus_iounmap(hp->erxregs, ERX_REG_SIZE);
-	sbus_iounmap(hp->bigmacregs, BMAC_REG_SIZE);
-	sbus_iounmap(hp->tcvregs, TCVR_REG_SIZE);
-	sbus_free_consistent(hp->happy_dev,
-			     PAGE_SIZE,
-			     hp->happy_block,
-			     hp->hblock_dvma);
+	of_iounmap(&op->resource[0], hp->gregs, GREG_REG_SIZE);
+	of_iounmap(&op->resource[1], hp->etxregs, ETX_REG_SIZE);
+	of_iounmap(&op->resource[2], hp->erxregs, ERX_REG_SIZE);
+	of_iounmap(&op->resource[3], hp->bigmacregs, BMAC_REG_SIZE);
+	of_iounmap(&op->resource[4], hp->tcvregs, TCVR_REG_SIZE);
+	dma_free_coherent(hp->dma_dev,
+			  PAGE_SIZE,
+			  hp->happy_block,
+			  hp->hblock_dvma);
 
 	free_netdev(net_dev);
 
-	dev_set_drvdata(&dev->dev, NULL);
+	dev_set_drvdata(&op->dev, NULL);
 
 	return 0;
 }
 
-static struct of_device_id hme_sbus_match[] = {
+static const struct of_device_id hme_sbus_match[] = {
 	{
 		.name = "SUNW,hme",
 	},
@@ -3346,7 +3280,7 @@ static int __init happy_meal_sbus_init(void)
 {
 	int err;
 
-	err = of_register_driver(&hme_sbus_driver, &sbus_bus_type);
+	err = of_register_driver(&hme_sbus_driver, &of_bus_type);
 	if (!err)
 		quattro_sbus_register_irqs();
 
diff --git a/drivers/net/sunhme.h b/drivers/net/sunhme.h
index 4da5539fac7b..efd2ca0fcad3 100644
--- a/drivers/net/sunhme.h
+++ b/drivers/net/sunhme.h
@@ -405,14 +405,11 @@ struct happy_meal {
 	u32 (*read_desc32)(hme32 *);
 	void (*write_txd)(struct happy_meal_txd *, u32, u32);
 	void (*write_rxd)(struct happy_meal_rxd *, u32, u32);
-	u32 (*dma_map)(void *, void *, long, int);
-	void (*dma_unmap)(void *, u32, long, int);
-	void (*dma_sync_for_cpu)(void *, u32, long, int);
-	void (*dma_sync_for_device)(void *, u32, long, int);
 #endif
 
-	/* This is either a sbus_dev or a pci_dev. */
+	/* This is either an of_device or a pci_dev. */
 	void			  *happy_dev;
+	struct device		  *dma_dev;
 
 	spinlock_t		  happy_lock;
 
diff --git a/drivers/net/sunlance.c b/drivers/net/sunlance.c
index 4e994f87469e..704301a5a7ff 100644
--- a/drivers/net/sunlance.c
+++ b/drivers/net/sunlance.c
@@ -91,6 +91,9 @@ static char lancestr[] = "LANCE";
 #include <linux/skbuff.h>
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
+#include <linux/dma-mapping.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/system.h>
 #include <asm/io.h>
@@ -98,7 +101,6 @@ static char lancestr[] = "LANCE";
 #include <asm/pgtable.h>
 #include <asm/byteorder.h>	/* Used by the checksum routines */
 #include <asm/idprom.h>
-#include <asm/sbus.h>
 #include <asm/prom.h>
 #include <asm/auxio.h>		/* For tpe-link-test? setting */
 #include <asm/irq.h>
@@ -248,7 +250,7 @@ struct lance_private {
 	int		rx_new, tx_new;
 	int		rx_old, tx_old;
 
-	struct sbus_dma *ledma;	/* If set this points to ledma	*/
+	struct of_device *ledma;	/* If set this points to ledma	*/
 	char		tpe;		/* cable-selection is TPE	*/
 	char		auto_select;	/* cable-selection by carrier	*/
 	char		burst_sizes;	/* ledma SBus burst sizes	*/
@@ -263,7 +265,8 @@ struct lance_private {
 	char	       	       *name;
 	dma_addr_t		init_block_dvma;
 	struct net_device      *dev;		  /* Backpointer	*/
-	struct sbus_dev	       *sdev;
+	struct of_device       *op;
+	struct of_device       *lebuffer;
 	struct timer_list       multicast_timer;
 };
 
@@ -1272,27 +1275,29 @@ static void lance_set_multicast_retry(unsigned long _opaque)
 static void lance_free_hwresources(struct lance_private *lp)
 {
 	if (lp->lregs)
-		sbus_iounmap(lp->lregs, LANCE_REG_SIZE);
+		of_iounmap(&lp->op->resource[0], lp->lregs, LANCE_REG_SIZE);
+	if (lp->dregs) {
+		struct of_device *ledma = lp->ledma;
+
+		of_iounmap(&ledma->resource[0], lp->dregs,
+			   resource_size(&ledma->resource[0]));
+	}
 	if (lp->init_block_iomem) {
-		sbus_iounmap(lp->init_block_iomem,
-			     sizeof(struct lance_init_block));
+		of_iounmap(&lp->lebuffer->resource[0], lp->init_block_iomem,
+			   sizeof(struct lance_init_block));
 	} else if (lp->init_block_mem) {
-		sbus_free_consistent(lp->sdev,
-				     sizeof(struct lance_init_block),
-				     lp->init_block_mem,
-				     lp->init_block_dvma);
+		dma_free_coherent(&lp->op->dev,
+				  sizeof(struct lance_init_block),
+				  lp->init_block_mem,
+				  lp->init_block_dvma);
 	}
 }
 
 /* Ethtool support... */
 static void sparc_lance_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
-	struct lance_private *lp = netdev_priv(dev);
-
 	strcpy(info->driver, "sunlance");
 	strcpy(info->version, "2.02");
-	sprintf(info->bus_info, "SBUS:%d",
-		lp->sdev->slot);
 }
 
 static u32 sparc_lance_get_link(struct net_device *dev)
@@ -1308,16 +1313,16 @@ static const struct ethtool_ops sparc_lance_ethtool_ops = {
 	.get_link		= sparc_lance_get_link,
 };
 
-static int __devinit sparc_lance_probe_one(struct sbus_dev *sdev,
-					   struct sbus_dma *ledma,
-					   struct sbus_dev *lebuffer)
+static int __devinit sparc_lance_probe_one(struct of_device *op,
+					   struct of_device *ledma,
+					   struct of_device *lebuffer)
 {
+	struct device_node *dp = op->node;
 	static unsigned version_printed;
-	struct device_node *dp = sdev->ofdev.node;
-	struct net_device *dev;
 	struct lance_private *lp;
-	int    i;
+	struct net_device *dev;
 	DECLARE_MAC_BUF(mac);
+	int    i;
 
 	dev = alloc_etherdev(sizeof(struct lance_private) + 8);
 	if (!dev)
@@ -1338,14 +1343,27 @@ static int __devinit sparc_lance_probe_one(struct sbus_dev *sdev,
 		dev->dev_addr[i] = idprom->id_ethaddr[i];
 
 	/* Get the IO region */
-	lp->lregs = sbus_ioremap(&sdev->resource[0], 0,
-				 LANCE_REG_SIZE, lancestr);
+	lp->lregs = of_ioremap(&op->resource[0], 0,
+			       LANCE_REG_SIZE, lancestr);
 	if (!lp->lregs) {
 		printk(KERN_ERR "SunLance: Cannot map registers.\n");
 		goto fail;
 	}
 
-	lp->sdev = sdev;
+	lp->ledma = ledma;
+	if (lp->ledma) {
+		lp->dregs = of_ioremap(&ledma->resource[0], 0,
+				       resource_size(&ledma->resource[0]),
+				       "ledma");
+		if (!lp->dregs) {
+			printk(KERN_ERR "SunLance: Cannot map "
+			       "ledma registers.\n");
+			goto fail;
+		}
+	}
+
+	lp->op = op;
+	lp->lebuffer = lebuffer;
 	if (lebuffer) {
 		/* sanity check */
 		if (lebuffer->resource[0].start & 7) {
@@ -1353,8 +1371,8 @@ static int __devinit sparc_lance_probe_one(struct sbus_dev *sdev,
 			goto fail;
 		}
 		lp->init_block_iomem =
-			sbus_ioremap(&lebuffer->resource[0], 0,
-				     sizeof(struct lance_init_block), "lebuffer");
+			of_ioremap(&lebuffer->resource[0], 0,
+				   sizeof(struct lance_init_block), "lebuffer");
 		if (!lp->init_block_iomem) {
 			printk(KERN_ERR "SunLance: Cannot map PIO buffer.\n");
 			goto fail;
@@ -1366,9 +1384,10 @@ static int __devinit sparc_lance_probe_one(struct sbus_dev *sdev,
 		lp->tx = lance_tx_pio;
 	} else {
 		lp->init_block_mem =
-			sbus_alloc_consistent(sdev, sizeof(struct lance_init_block),
-					      &lp->init_block_dvma);
-		if (!lp->init_block_mem || lp->init_block_dvma == 0) {
+			dma_alloc_coherent(&op->dev,
+					   sizeof(struct lance_init_block),
+					   &lp->init_block_dvma, GFP_ATOMIC);
+		if (!lp->init_block_mem) {
 			printk(KERN_ERR "SunLance: Cannot allocate consistent DMA memory.\n");
 			goto fail;
 		}
@@ -1383,13 +1402,13 @@ static int __devinit sparc_lance_probe_one(struct sbus_dev *sdev,
 						      LE_C3_BCON));
 
 	lp->name = lancestr;
-	lp->ledma = ledma;
 
 	lp->burst_sizes = 0;
 	if (lp->ledma) {
-		struct device_node *ledma_dp = ledma->sdev->ofdev.node;
-		const char *prop;
+		struct device_node *ledma_dp = ledma->node;
+		struct device_node *sbus_dp;
 		unsigned int sbmask;
+		const char *prop;
 		u32 csr;
 
 		/* Find burst-size property for ledma */
@@ -1397,7 +1416,8 @@ static int __devinit sparc_lance_probe_one(struct sbus_dev *sdev,
 							"burst-sizes", 0);
 
 		/* ledma may be capable of fast bursts, but sbus may not. */
-		sbmask = of_getintprop_default(ledma_dp, "burst-sizes",
+		sbus_dp = ledma_dp->parent;
+		sbmask = of_getintprop_default(sbus_dp, "burst-sizes",
 					       DMA_BURSTBITS);
 		lp->burst_sizes &= sbmask;
 
@@ -1435,8 +1455,6 @@ no_link_test:
 			lp->tpe = 1;
 		}
 
-		lp->dregs = ledma->regs;
-
 		/* Reset ledma */
 		csr = sbus_readl(lp->dregs + DMA_CSR);
 		sbus_writel(csr | DMA_RST_ENET, lp->dregs + DMA_CSR);
@@ -1446,7 +1464,7 @@ no_link_test:
 		lp->dregs = NULL;
 
 	lp->dev = dev;
-	SET_NETDEV_DEV(dev, &sdev->ofdev.dev);
+	SET_NETDEV_DEV(dev, &op->dev);
 	dev->open = &lance_open;
 	dev->stop = &lance_close;
 	dev->hard_start_xmit = &lance_start_xmit;
@@ -1455,9 +1473,7 @@ no_link_test:
 	dev->set_multicast_list = &lance_set_multicast;
 	dev->ethtool_ops = &sparc_lance_ethtool_ops;
 
-	dev->irq = sdev->irqs[0];
-
-	dev->dma = 0;
+	dev->irq = op->irqs[0];
 
 	/* We cannot sleep if the chip is busy during a
 	 * multicast list update event, because such events
@@ -1473,7 +1489,7 @@ no_link_test:
 		goto fail;
 	}
 
-	dev_set_drvdata(&sdev->ofdev.dev, lp);
+	dev_set_drvdata(&op->dev, lp);
 
 	printk(KERN_INFO "%s: LANCE %s\n",
 	       dev->name, print_mac(mac, dev->dev_addr));
@@ -1486,80 +1502,25 @@ fail:
 	return -ENODEV;
 }
 
-/* On 4m, find the associated dma for the lance chip */
-static struct sbus_dma * __devinit find_ledma(struct sbus_dev *sdev)
-{
-	struct sbus_dma *p;
-
-	for_each_dvma(p) {
-		if (p->sdev == sdev)
-			return p;
-	}
-	return NULL;
-}
-
-#ifdef CONFIG_SUN4
-
-#include <asm/sun4paddr.h>
-#include <asm/machines.h>
-
-/* Find all the lance cards on the system and initialize them */
-static struct sbus_dev sun4_sdev;
-static int __devinit sparc_lance_init(void)
-{
-	if ((idprom->id_machtype == (SM_SUN4|SM_4_330)) ||
-	    (idprom->id_machtype == (SM_SUN4|SM_4_470))) {
-		memset(&sun4_sdev, 0, sizeof(struct sbus_dev));
-		sun4_sdev.reg_addrs[0].phys_addr = sun4_eth_physaddr;
-		sun4_sdev.irqs[0] = 6;
-		return sparc_lance_probe_one(&sun4_sdev, NULL, NULL);
-	}
-	return -ENODEV;
-}
-
-static int __exit sunlance_sun4_remove(void)
+static int __devinit sunlance_sbus_probe(struct of_device *op, const struct of_device_id *match)
 {
-	struct lance_private *lp = dev_get_drvdata(&sun4_sdev.ofdev.dev);
-	struct net_device *net_dev = lp->dev;
-
-	unregister_netdev(net_dev);
-
-	lance_free_hwresources(lp);
-
-	free_netdev(net_dev);
-
-	dev_set_drvdata(&sun4_sdev.ofdev.dev, NULL);
-
-	return 0;
-}
-
-#else /* !CONFIG_SUN4 */
-
-static int __devinit sunlance_sbus_probe(struct of_device *dev, const struct of_device_id *match)
-{
-	struct sbus_dev *sdev = to_sbus_device(&dev->dev);
+	struct of_device *parent = to_of_device(op->dev.parent);
+	struct device_node *parent_dp = parent->node;
 	int err;
 
-	if (sdev->parent) {
-		struct of_device *parent = &sdev->parent->ofdev;
-
-		if (!strcmp(parent->node->name, "ledma")) {
-			struct sbus_dma *ledma = find_ledma(to_sbus_device(&parent->dev));
-
-			err = sparc_lance_probe_one(sdev, ledma, NULL);
-		} else if (!strcmp(parent->node->name, "lebuffer")) {
-			err = sparc_lance_probe_one(sdev, NULL, to_sbus_device(&parent->dev));
-		} else
-			err = sparc_lance_probe_one(sdev, NULL, NULL);
+	if (!strcmp(parent_dp->name, "ledma")) {
+		err = sparc_lance_probe_one(op, parent, NULL);
+	} else if (!strcmp(parent_dp->name, "lebuffer")) {
+		err = sparc_lance_probe_one(op, NULL, parent);
 	} else
-		err = sparc_lance_probe_one(sdev, NULL, NULL);
+		err = sparc_lance_probe_one(op, NULL, NULL);
 
 	return err;
 }
 
-static int __devexit sunlance_sbus_remove(struct of_device *dev)
+static int __devexit sunlance_sbus_remove(struct of_device *op)
 {
-	struct lance_private *lp = dev_get_drvdata(&dev->dev);
+	struct lance_private *lp = dev_get_drvdata(&op->dev);
 	struct net_device *net_dev = lp->dev;
 
 	unregister_netdev(net_dev);
@@ -1568,12 +1529,12 @@ static int __devexit sunlance_sbus_remove(struct of_device *dev)
 
 	free_netdev(net_dev);
 
-	dev_set_drvdata(&dev->dev, NULL);
+	dev_set_drvdata(&op->dev, NULL);
 
 	return 0;
 }
 
-static struct of_device_id sunlance_sbus_match[] = {
+static const struct of_device_id sunlance_sbus_match[] = {
 	{
 		.name = "le",
 	},
@@ -1593,17 +1554,12 @@ static struct of_platform_driver sunlance_sbus_driver = {
 /* Find all the lance cards on the system and initialize them */
 static int __init sparc_lance_init(void)
 {
-	return of_register_driver(&sunlance_sbus_driver, &sbus_bus_type);
+	return of_register_driver(&sunlance_sbus_driver, &of_bus_type);
 }
-#endif /* !CONFIG_SUN4 */
 
 static void __exit sparc_lance_exit(void)
 {
-#ifdef CONFIG_SUN4
-	sunlance_sun4_remove();
-#else
 	of_unregister_driver(&sunlance_sbus_driver);
-#endif
 }
 
 module_init(sparc_lance_init);
diff --git a/drivers/net/sunqe.c b/drivers/net/sunqe.c
index e811331d4608..f63644744ff9 100644
--- a/drivers/net/sunqe.c
+++ b/drivers/net/sunqe.c
@@ -3,7 +3,7 @@
  *          controller out there can be most efficiently programmed
  *          if you make it look like a LANCE.
  *
- * Copyright (C) 1996, 1999, 2003, 2006 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 1996, 1999, 2003, 2006, 2008 David S. Miller (davem@davemloft.net)
  */
 
 #include <linux/module.h>
@@ -24,13 +24,15 @@
 #include <linux/skbuff.h>
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
+#include <linux/dma-mapping.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/system.h>
 #include <asm/io.h>
 #include <asm/dma.h>
 #include <asm/byteorder.h>
 #include <asm/idprom.h>
-#include <asm/sbus.h>
 #include <asm/openprom.h>
 #include <asm/oplib.h>
 #include <asm/auxio.h>
@@ -40,8 +42,8 @@
 #include "sunqe.h"
 
 #define DRV_NAME	"sunqe"
-#define DRV_VERSION	"4.0"
-#define DRV_RELDATE	"June 23, 2006"
+#define DRV_VERSION	"4.1"
+#define DRV_RELDATE	"August 27, 2008"
 #define DRV_AUTHOR	"David S. Miller (davem@davemloft.net)"
 
 static char version[] =
@@ -690,12 +692,18 @@ static void qe_set_multicast(struct net_device *dev)
 /* Ethtool support... */
 static void qe_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
+	const struct linux_prom_registers *regs;
 	struct sunqe *qep = dev->priv;
+	struct of_device *op;
 
 	strcpy(info->driver, "sunqe");
 	strcpy(info->version, "3.0");
-	sprintf(info->bus_info, "SBUS:%d",
-		qep->qe_sdev->slot);
+
+	op = qep->op;
+	regs = of_get_property(op->node, "reg", NULL);
+	if (regs)
+		sprintf(info->bus_info, "SBUS:%d", regs->which_io);
+
 }
 
 static u32 qe_get_link(struct net_device *dev)
@@ -717,11 +725,11 @@ static const struct ethtool_ops qe_ethtool_ops = {
 };
 
 /* This is only called once at boot time for each card probed. */
-static inline void qec_init_once(struct sunqec *qecp, struct sbus_dev *qsdev)
+static void qec_init_once(struct sunqec *qecp, struct of_device *op)
 {
 	u8 bsizes = qecp->qec_bursts;
 
-	if (sbus_can_burst64(qsdev) && (bsizes & DMA_BURST64)) {
+	if (sbus_can_burst64() && (bsizes & DMA_BURST64)) {
 		sbus_writel(GLOB_CTRL_B64, qecp->gregs + GLOB_CTRL);
 	} else if (bsizes & DMA_BURST32) {
 		sbus_writel(GLOB_CTRL_B32, qecp->gregs + GLOB_CTRL);
@@ -735,15 +743,15 @@ static inline void qec_init_once(struct sunqec *qecp, struct sbus_dev *qsdev)
 	sbus_writel(GLOB_PSIZE_2048, qecp->gregs + GLOB_PSIZE);
 
 	/* Set the local memsize register, divided up to one piece per QE channel. */
-	sbus_writel((qsdev->reg_addrs[1].reg_size >> 2),
+	sbus_writel((resource_size(&op->resource[1]) >> 2),
 		    qecp->gregs + GLOB_MSIZE);
 
 	/* Divide up the local QEC memory amongst the 4 QE receiver and
 	 * transmitter FIFOs.  Basically it is (total / 2 / num_channels).
 	 */
-	sbus_writel((qsdev->reg_addrs[1].reg_size >> 2) >> 1,
+	sbus_writel((resource_size(&op->resource[1]) >> 2) >> 1,
 		    qecp->gregs + GLOB_TSIZE);
-	sbus_writel((qsdev->reg_addrs[1].reg_size >> 2) >> 1,
+	sbus_writel((resource_size(&op->resource[1]) >> 2) >> 1,
 		    qecp->gregs + GLOB_RSIZE);
 }
 
@@ -767,24 +775,21 @@ static u8 __devinit qec_get_burst(struct device_node *dp)
 	return bsizes;
 }
 
-static struct sunqec * __devinit get_qec(struct sbus_dev *child_sdev)
+static struct sunqec * __devinit get_qec(struct of_device *child)
 {
-	struct sbus_dev *qec_sdev = child_sdev->parent;
+	struct of_device *op = to_of_device(child->dev.parent);
 	struct sunqec *qecp;
 
-	for (qecp = root_qec_dev; qecp; qecp = qecp->next_module) {
-		if (qecp->qec_sdev == qec_sdev)
-			break;
-	}
+	qecp = dev_get_drvdata(&op->dev);
 	if (!qecp) {
 		qecp = kzalloc(sizeof(struct sunqec), GFP_KERNEL);
 		if (qecp) {
 			u32 ctrl;
 
-			qecp->qec_sdev = qec_sdev;
-			qecp->gregs = sbus_ioremap(&qec_sdev->resource[0], 0,
-						   GLOB_REG_SIZE,
-						   "QEC Global Registers");
+			qecp->op = op;
+			qecp->gregs = of_ioremap(&op->resource[0], 0,
+						 GLOB_REG_SIZE,
+						 "QEC Global Registers");
 			if (!qecp->gregs)
 				goto fail;
 
@@ -799,16 +804,18 @@ static struct sunqec * __devinit get_qec(struct sbus_dev *child_sdev)
 			if (qec_global_reset(qecp->gregs))
 				goto fail;
 
-			qecp->qec_bursts = qec_get_burst(qec_sdev->ofdev.node);
+			qecp->qec_bursts = qec_get_burst(op->node);
 
-			qec_init_once(qecp, qec_sdev);
+			qec_init_once(qecp, op);
 
-			if (request_irq(qec_sdev->irqs[0], &qec_interrupt,
+			if (request_irq(op->irqs[0], &qec_interrupt,
 					IRQF_SHARED, "qec", (void *) qecp)) {
 				printk(KERN_ERR "qec: Can't register irq.\n");
 				goto fail;
 			}
 
+			dev_set_drvdata(&op->dev, qecp);
+
 			qecp->next_module = root_qec_dev;
 			root_qec_dev = qecp;
 		}
@@ -818,17 +825,17 @@ static struct sunqec * __devinit get_qec(struct sbus_dev *child_sdev)
 
 fail:
 	if (qecp->gregs)
-		sbus_iounmap(qecp->gregs, GLOB_REG_SIZE);
+		of_iounmap(&op->resource[0], qecp->gregs, GLOB_REG_SIZE);
 	kfree(qecp);
 	return NULL;
 }
 
-static int __devinit qec_ether_init(struct sbus_dev *sdev)
+static int __devinit qec_ether_init(struct of_device *op)
 {
 	static unsigned version_printed;
 	struct net_device *dev;
-	struct sunqe *qe;
 	struct sunqec *qecp;
+	struct sunqe *qe;
 	int i, res;
 
 	if (version_printed++ == 0)
@@ -842,49 +849,42 @@ static int __devinit qec_ether_init(struct sbus_dev *sdev)
 
 	qe = netdev_priv(dev);
 
-	i = of_getintprop_default(sdev->ofdev.node, "channel#", -1);
-	if (i == -1) {
-		struct sbus_dev *td = sdev->parent->child;
-		i = 0;
-		while (td != sdev) {
-			td = td->next;
-			i++;
-		}
-	}
+	res = -ENODEV;
+
+	i = of_getintprop_default(op->node, "channel#", -1);
+	if (i == -1)
+		goto fail;
 	qe->channel = i;
 	spin_lock_init(&qe->lock);
 
-	res = -ENODEV;
-	qecp = get_qec(sdev);
+	qecp = get_qec(op);
 	if (!qecp)
 		goto fail;
 
 	qecp->qes[qe->channel] = qe;
 	qe->dev = dev;
 	qe->parent = qecp;
-	qe->qe_sdev = sdev;
+	qe->op = op;
 
 	res = -ENOMEM;
-	qe->qcregs = sbus_ioremap(&qe->qe_sdev->resource[0], 0,
-				  CREG_REG_SIZE, "QEC Channel Registers");
+	qe->qcregs = of_ioremap(&op->resource[0], 0,
+				CREG_REG_SIZE, "QEC Channel Registers");
 	if (!qe->qcregs) {
 		printk(KERN_ERR "qe: Cannot map channel registers.\n");
 		goto fail;
 	}
 
-	qe->mregs = sbus_ioremap(&qe->qe_sdev->resource[1], 0,
-				 MREGS_REG_SIZE, "QE MACE Registers");
+	qe->mregs = of_ioremap(&op->resource[1], 0,
+			       MREGS_REG_SIZE, "QE MACE Registers");
 	if (!qe->mregs) {
 		printk(KERN_ERR "qe: Cannot map MACE registers.\n");
 		goto fail;
 	}
 
-	qe->qe_block = sbus_alloc_consistent(qe->qe_sdev,
-					     PAGE_SIZE,
-					     &qe->qblock_dvma);
-	qe->buffers = sbus_alloc_consistent(qe->qe_sdev,
-					    sizeof(struct sunqe_buffers),
-					    &qe->buffers_dvma);
+	qe->qe_block = dma_alloc_coherent(&op->dev, PAGE_SIZE,
+					  &qe->qblock_dvma, GFP_ATOMIC);
+	qe->buffers = dma_alloc_coherent(&op->dev, sizeof(struct sunqe_buffers),
+					 &qe->buffers_dvma, GFP_ATOMIC);
 	if (qe->qe_block == NULL || qe->qblock_dvma == 0 ||
 	    qe->buffers == NULL || qe->buffers_dvma == 0)
 		goto fail;
@@ -892,7 +892,7 @@ static int __devinit qec_ether_init(struct sbus_dev *sdev)
 	/* Stop this QE. */
 	qe_stop(qe);
 
-	SET_NETDEV_DEV(dev, &sdev->ofdev.dev);
+	SET_NETDEV_DEV(dev, &op->dev);
 
 	dev->open = qe_open;
 	dev->stop = qe_close;
@@ -900,7 +900,7 @@ static int __devinit qec_ether_init(struct sbus_dev *sdev)
 	dev->set_multicast_list = qe_set_multicast;
 	dev->tx_timeout = qe_tx_timeout;
 	dev->watchdog_timeo = 5*HZ;
-	dev->irq = sdev->irqs[0];
+	dev->irq = op->irqs[0];
 	dev->dma = 0;
 	dev->ethtool_ops = &qe_ethtool_ops;
 
@@ -908,7 +908,7 @@ static int __devinit qec_ether_init(struct sbus_dev *sdev)
 	if (res)
 		goto fail;
 
-	dev_set_drvdata(&sdev->ofdev.dev, qe);
+	dev_set_drvdata(&op->dev, qe);
 
 	printk(KERN_INFO "%s: qe channel[%d] ", dev->name, qe->channel);
 	for (i = 0; i < 6; i++)
@@ -922,58 +922,50 @@ static int __devinit qec_ether_init(struct sbus_dev *sdev)
 
 fail:
 	if (qe->qcregs)
-		sbus_iounmap(qe->qcregs, CREG_REG_SIZE);
+		of_iounmap(&op->resource[0], qe->qcregs, CREG_REG_SIZE);
 	if (qe->mregs)
-		sbus_iounmap(qe->mregs, MREGS_REG_SIZE);
+		of_iounmap(&op->resource[1], qe->mregs, MREGS_REG_SIZE);
 	if (qe->qe_block)
-		sbus_free_consistent(qe->qe_sdev,
-				     PAGE_SIZE,
-				     qe->qe_block,
-				     qe->qblock_dvma);
+		dma_free_coherent(&op->dev, PAGE_SIZE,
+				  qe->qe_block, qe->qblock_dvma);
 	if (qe->buffers)
-		sbus_free_consistent(qe->qe_sdev,
-				     sizeof(struct sunqe_buffers),
-				     qe->buffers,
-				     qe->buffers_dvma);
+		dma_free_coherent(&op->dev,
+				  sizeof(struct sunqe_buffers),
+				  qe->buffers,
+				  qe->buffers_dvma);
 
 	free_netdev(dev);
 
 	return res;
 }
 
-static int __devinit qec_sbus_probe(struct of_device *dev, const struct of_device_id *match)
+static int __devinit qec_sbus_probe(struct of_device *op, const struct of_device_id *match)
 {
-	struct sbus_dev *sdev = to_sbus_device(&dev->dev);
-
-	return qec_ether_init(sdev);
+	return qec_ether_init(op);
 }
 
-static int __devexit qec_sbus_remove(struct of_device *dev)
+static int __devexit qec_sbus_remove(struct of_device *op)
 {
-	struct sunqe *qp = dev_get_drvdata(&dev->dev);
+	struct sunqe *qp = dev_get_drvdata(&op->dev);
 	struct net_device *net_dev = qp->dev;
 
 	unregister_netdev(net_dev);
 
-	sbus_iounmap(qp->qcregs, CREG_REG_SIZE);
-	sbus_iounmap(qp->mregs, MREGS_REG_SIZE);
-	sbus_free_consistent(qp->qe_sdev,
-			     PAGE_SIZE,
-			     qp->qe_block,
-			     qp->qblock_dvma);
-	sbus_free_consistent(qp->qe_sdev,
-			     sizeof(struct sunqe_buffers),
-			     qp->buffers,
-			     qp->buffers_dvma);
+	of_iounmap(&op->resource[0], qp->qcregs, CREG_REG_SIZE);
+	of_iounmap(&op->resource[1], qp->mregs, MREGS_REG_SIZE);
+	dma_free_coherent(&op->dev, PAGE_SIZE,
+			  qp->qe_block, qp->qblock_dvma);
+	dma_free_coherent(&op->dev, sizeof(struct sunqe_buffers),
+			  qp->buffers, qp->buffers_dvma);
 
 	free_netdev(net_dev);
 
-	dev_set_drvdata(&dev->dev, NULL);
+	dev_set_drvdata(&op->dev, NULL);
 
 	return 0;
 }
 
-static struct of_device_id qec_sbus_match[] = {
+static const struct of_device_id qec_sbus_match[] = {
 	{
 		.name = "qe",
 	},
@@ -991,7 +983,7 @@ static struct of_platform_driver qec_sbus_driver = {
 
 static int __init qec_init(void)
 {
-	return of_register_driver(&qec_sbus_driver, &sbus_bus_type);
+	return of_register_driver(&qec_sbus_driver, &of_bus_type);
 }
 
 static void __exit qec_exit(void)
@@ -1000,11 +992,11 @@ static void __exit qec_exit(void)
 
 	while (root_qec_dev) {
 		struct sunqec *next = root_qec_dev->next_module;
+		struct of_device *op = root_qec_dev->op;
 
-		free_irq(root_qec_dev->qec_sdev->irqs[0],
-			 (void *) root_qec_dev);
-		sbus_iounmap(root_qec_dev->gregs, GLOB_REG_SIZE);
-
+		free_irq(op->irqs[0], (void *) root_qec_dev);
+		of_iounmap(&op->resource[0], root_qec_dev->gregs,
+			   GLOB_REG_SIZE);
 		kfree(root_qec_dev);
 
 		root_qec_dev = next;
diff --git a/drivers/net/sunqe.h b/drivers/net/sunqe.h
index 347c8ddc1592..5813a7b2faa5 100644
--- a/drivers/net/sunqe.h
+++ b/drivers/net/sunqe.h
@@ -314,7 +314,7 @@ struct sunqec {
 	void __iomem		*gregs;		/* QEC Global Registers         */
 	struct sunqe		*qes[4];	/* Each child MACE              */
 	unsigned int            qec_bursts;	/* Support burst sizes          */
-	struct sbus_dev		*qec_sdev;	/* QEC's SBUS device            */
+	struct of_device	*op;		/* QEC's OF device              */
 	struct sunqec		*next_module;	/* List of all QECs in system   */
 };
 
@@ -342,7 +342,7 @@ struct sunqe {
 	__u32				buffers_dvma;	/* DVMA visible address.       */
 	struct sunqec			*parent;
 	u8				mconfig;	/* Base MACE mconfig value     */
-	struct sbus_dev			*qe_sdev;	/* QE's SBUS device struct     */
+	struct of_device		*op;		/* QE's OF device struct       */
 	struct net_device		*dev;		/* QE's netdevice struct       */
 	int				channel;	/* Who am I?                   */
 };
diff --git a/drivers/net/sunvnet.c b/drivers/net/sunvnet.c
index 6415ce15c2ef..a720065553df 100644
--- a/drivers/net/sunvnet.c
+++ b/drivers/net/sunvnet.c
@@ -1,6 +1,6 @@
 /* sunvnet.c: Sun LDOM Virtual Network Driver.
  *
- * Copyright (C) 2007 David S. Miller <davem@davemloft.net>
+ * Copyright (C) 2007, 2008 David S. Miller <davem@davemloft.net>
  */
 
 #include <linux/module.h>
@@ -1260,7 +1260,7 @@ static int vnet_port_remove(struct vio_dev *vdev)
 	return 0;
 }
 
-static struct vio_device_id vnet_port_match[] = {
+static const struct vio_device_id vnet_port_match[] = {
 	{
 		.type = "vnet-port",
 	},
diff --git a/drivers/net/wan/Kconfig b/drivers/net/wan/Kconfig
index 2ae2ec40015d..21efd99b9294 100644
--- a/drivers/net/wan/Kconfig
+++ b/drivers/net/wan/Kconfig
@@ -205,7 +205,7 @@ config WANXL_BUILD_FIRMWARE
 
 config PC300
 	tristate "Cyclades-PC300 support (RS-232/V.35, X.21, T1/E1 boards)"
-	depends on HDLC && PCI
+	depends on HDLC && PCI && BROKEN
 	---help---
 	  Driver for the Cyclades-PC300 synchronous communication boards.
 
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index c749bdba214c..3c3dd403f5dd 100644
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -1794,10 +1794,10 @@ static struct xenbus_driver netfront = {
 
 static int __init netif_init(void)
 {
-	if (!is_running_on_xen())
+	if (!xen_domain())
 		return -ENODEV;
 
-	if (is_initial_xendomain())
+	if (xen_initial_domain())
 		return 0;
 
 	printk(KERN_INFO "Initialising Xen virtual ethernet driver.\n");
@@ -1809,7 +1809,7 @@ module_init(netif_init);
 
 static void __exit netif_exit(void)
 {
-	if (is_initial_xendomain())
+	if (xen_initial_domain())
 		return;
 
 	xenbus_unregister_driver(&netfront);
diff --git a/drivers/oprofile/buffer_sync.c b/drivers/oprofile/buffer_sync.c
index 9304c4555079..ed982273fb8b 100644
--- a/drivers/oprofile/buffer_sync.c
+++ b/drivers/oprofile/buffer_sync.c
@@ -5,6 +5,7 @@
  * @remark Read the file COPYING
  *
  * @author John Levon <levon@movementarian.org>
+ * @author Barry Kasindorf
  *
  * This is the core of the buffer management. Each
  * CPU buffer is processed and entered into the
@@ -33,7 +34,7 @@
 #include "event_buffer.h"
 #include "cpu_buffer.h"
 #include "buffer_sync.h"
- 
+
 static LIST_HEAD(dying_tasks);
 static LIST_HEAD(dead_tasks);
 static cpumask_t marked_cpus = CPU_MASK_NONE;
@@ -48,10 +49,11 @@ static void process_task_mortuary(void);
  * Can be invoked from softirq via RCU callback due to
  * call_rcu() of the task struct, hence the _irqsave.
  */
-static int task_free_notify(struct notifier_block * self, unsigned long val, void * data)
+static int
+task_free_notify(struct notifier_block *self, unsigned long val, void *data)
 {
 	unsigned long flags;
-	struct task_struct * task = data;
+	struct task_struct *task = data;
 	spin_lock_irqsave(&task_mortuary, flags);
 	list_add(&task->tasks, &dying_tasks);
 	spin_unlock_irqrestore(&task_mortuary, flags);
@@ -62,13 +64,14 @@ static int task_free_notify(struct notifier_block * self, unsigned long val, voi
 /* The task is on its way out. A sync of the buffer means we can catch
  * any remaining samples for this task.
  */
-static int task_exit_notify(struct notifier_block * self, unsigned long val, void * data)
+static int
+task_exit_notify(struct notifier_block *self, unsigned long val, void *data)
 {
 	/* To avoid latency problems, we only process the current CPU,
 	 * hoping that most samples for the task are on this CPU
 	 */
 	sync_buffer(raw_smp_processor_id());
-  	return 0;
+	return 0;
 }
 
 
@@ -77,11 +80,12 @@ static int task_exit_notify(struct notifier_block * self, unsigned long val, voi
  * we don't lose any. This does not have to be exact, it's a QoI issue
  * only.
  */
-static int munmap_notify(struct notifier_block * self, unsigned long val, void * data)
+static int
+munmap_notify(struct notifier_block *self, unsigned long val, void *data)
 {
 	unsigned long addr = (unsigned long)data;
-	struct mm_struct * mm = current->mm;
-	struct vm_area_struct * mpnt;
+	struct mm_struct *mm = current->mm;
+	struct vm_area_struct *mpnt;
 
 	down_read(&mm->mmap_sem);
 
@@ -99,11 +103,12 @@ static int munmap_notify(struct notifier_block * self, unsigned long val, void *
 	return 0;
 }
 
- 
+
 /* We need to be told about new modules so we don't attribute to a previously
  * loaded module, or drop the samples on the floor.
  */
-static int module_load_notify(struct notifier_block * self, unsigned long val, void * data)
+static int
+module_load_notify(struct notifier_block *self, unsigned long val, void *data)
 {
 #ifdef CONFIG_MODULES
 	if (val != MODULE_STATE_COMING)
@@ -118,7 +123,7 @@ static int module_load_notify(struct notifier_block * self, unsigned long val, v
 	return 0;
 }
 
- 
+
 static struct notifier_block task_free_nb = {
 	.notifier_call	= task_free_notify,
 };
@@ -135,7 +140,7 @@ static struct notifier_block module_load_nb = {
 	.notifier_call = module_load_notify,
 };
 
- 
+
 static void end_sync(void)
 {
 	end_cpu_work();
@@ -208,14 +213,14 @@ static inline unsigned long fast_get_dcookie(struct path *path)
  * not strictly necessary but allows oprofile to associate
  * shared-library samples with particular applications
  */
-static unsigned long get_exec_dcookie(struct mm_struct * mm)
+static unsigned long get_exec_dcookie(struct mm_struct *mm)
 {
 	unsigned long cookie = NO_COOKIE;
-	struct vm_area_struct * vma;
- 
+	struct vm_area_struct *vma;
+
 	if (!mm)
 		goto out;
- 
+
 	for (vma = mm->mmap; vma; vma = vma->vm_next) {
 		if (!vma->vm_file)
 			continue;
@@ -235,13 +240,14 @@ out:
  * sure to do this lookup before a mm->mmap modification happens so
  * we don't lose track.
  */
-static unsigned long lookup_dcookie(struct mm_struct * mm, unsigned long addr, off_t * offset)
+static unsigned long
+lookup_dcookie(struct mm_struct *mm, unsigned long addr, off_t *offset)
 {
 	unsigned long cookie = NO_COOKIE;
-	struct vm_area_struct * vma;
+	struct vm_area_struct *vma;
 
 	for (vma = find_vma(mm, addr); vma; vma = vma->vm_next) {
- 
+
 		if (addr < vma->vm_start || addr >= vma->vm_end)
 			continue;
 
@@ -263,9 +269,20 @@ static unsigned long lookup_dcookie(struct mm_struct * mm, unsigned long addr, o
 	return cookie;
 }
 
+static void increment_tail(struct oprofile_cpu_buffer *b)
+{
+	unsigned long new_tail = b->tail_pos + 1;
+
+	rmb();	/* be sure fifo pointers are synchromized */
+
+	if (new_tail < b->buffer_size)
+		b->tail_pos = new_tail;
+	else
+		b->tail_pos = 0;
+}
 
 static unsigned long last_cookie = INVALID_COOKIE;
- 
+
 static void add_cpu_switch(int i)
 {
 	add_event_entry(ESCAPE_CODE);
@@ -278,16 +295,16 @@ static void add_kernel_ctx_switch(unsigned int in_kernel)
 {
 	add_event_entry(ESCAPE_CODE);
 	if (in_kernel)
-		add_event_entry(KERNEL_ENTER_SWITCH_CODE); 
+		add_event_entry(KERNEL_ENTER_SWITCH_CODE);
 	else
-		add_event_entry(KERNEL_EXIT_SWITCH_CODE); 
+		add_event_entry(KERNEL_EXIT_SWITCH_CODE);
 }
- 
+
 static void
-add_user_ctx_switch(struct task_struct const * task, unsigned long cookie)
+add_user_ctx_switch(struct task_struct const *task, unsigned long cookie)
 {
 	add_event_entry(ESCAPE_CODE);
-	add_event_entry(CTX_SWITCH_CODE); 
+	add_event_entry(CTX_SWITCH_CODE);
 	add_event_entry(task->pid);
 	add_event_entry(cookie);
 	/* Another code for daemon back-compat */
@@ -296,7 +313,7 @@ add_user_ctx_switch(struct task_struct const * task, unsigned long cookie)
 	add_event_entry(task->tgid);
 }
 
- 
+
 static void add_cookie_switch(unsigned long cookie)
 {
 	add_event_entry(ESCAPE_CODE);
@@ -304,13 +321,78 @@ static void add_cookie_switch(unsigned long cookie)
 	add_event_entry(cookie);
 }
 
- 
+
 static void add_trace_begin(void)
 {
 	add_event_entry(ESCAPE_CODE);
 	add_event_entry(TRACE_BEGIN_CODE);
 }
 
+#ifdef CONFIG_OPROFILE_IBS
+
+#define IBS_FETCH_CODE_SIZE	2
+#define IBS_OP_CODE_SIZE	5
+#define IBS_EIP(offset)				\
+	(((struct op_sample *)&cpu_buf->buffer[(offset)])->eip)
+#define IBS_EVENT(offset)				\
+	(((struct op_sample *)&cpu_buf->buffer[(offset)])->event)
+
+/*
+ * Add IBS fetch and op entries to event buffer
+ */
+static void add_ibs_begin(struct oprofile_cpu_buffer *cpu_buf, int code,
+	int in_kernel, struct mm_struct *mm)
+{
+	unsigned long rip;
+	int i, count;
+	unsigned long ibs_cookie = 0;
+	off_t offset;
+
+	increment_tail(cpu_buf);	/* move to RIP entry */
+
+	rip = IBS_EIP(cpu_buf->tail_pos);
+
+#ifdef __LP64__
+	rip += IBS_EVENT(cpu_buf->tail_pos) << 32;
+#endif
+
+	if (mm) {
+		ibs_cookie = lookup_dcookie(mm, rip, &offset);
+
+		if (ibs_cookie == NO_COOKIE)
+			offset = rip;
+		if (ibs_cookie == INVALID_COOKIE) {
+			atomic_inc(&oprofile_stats.sample_lost_no_mapping);
+			offset = rip;
+		}
+		if (ibs_cookie != last_cookie) {
+			add_cookie_switch(ibs_cookie);
+			last_cookie = ibs_cookie;
+		}
+	} else
+		offset = rip;
+
+	add_event_entry(ESCAPE_CODE);
+	add_event_entry(code);
+	add_event_entry(offset);	/* Offset from Dcookie */
+
+	/* we send the Dcookie offset, but send the raw Linear Add also*/
+	add_event_entry(IBS_EIP(cpu_buf->tail_pos));
+	add_event_entry(IBS_EVENT(cpu_buf->tail_pos));
+
+	if (code == IBS_FETCH_CODE)
+		count = IBS_FETCH_CODE_SIZE;	/*IBS FETCH is 2 int64s*/
+	else
+		count = IBS_OP_CODE_SIZE;	/*IBS OP is 5 int64s*/
+
+	for (i = 0; i < count; i++) {
+		increment_tail(cpu_buf);
+		add_event_entry(IBS_EIP(cpu_buf->tail_pos));
+		add_event_entry(IBS_EVENT(cpu_buf->tail_pos));
+	}
+}
+
+#endif
 
 static void add_sample_entry(unsigned long offset, unsigned long event)
 {
@@ -319,13 +401,13 @@ static void add_sample_entry(unsigned long offset, unsigned long event)
 }
 
 
-static int add_us_sample(struct mm_struct * mm, struct op_sample * s)
+static int add_us_sample(struct mm_struct *mm, struct op_sample *s)
 {
 	unsigned long cookie;
 	off_t offset;
- 
- 	cookie = lookup_dcookie(mm, s->eip, &offset);
- 
+
+	cookie = lookup_dcookie(mm, s->eip, &offset);
+
 	if (cookie == INVALID_COOKIE) {
 		atomic_inc(&oprofile_stats.sample_lost_no_mapping);
 		return 0;
@@ -341,13 +423,13 @@ static int add_us_sample(struct mm_struct * mm, struct op_sample * s)
 	return 1;
 }
 
- 
+
 /* Add a sample to the global event buffer. If possible the
  * sample is converted into a persistent dentry/offset pair
  * for later lookup from userspace.
  */
 static int
-add_sample(struct mm_struct * mm, struct op_sample * s, int in_kernel)
+add_sample(struct mm_struct *mm, struct op_sample *s, int in_kernel)
 {
 	if (in_kernel) {
 		add_sample_entry(s->eip, s->event);
@@ -359,9 +441,9 @@ add_sample(struct mm_struct * mm, struct op_sample * s, int in_kernel)
 	}
 	return 0;
 }
- 
 
-static void release_mm(struct mm_struct * mm)
+
+static void release_mm(struct mm_struct *mm)
 {
 	if (!mm)
 		return;
@@ -370,9 +452,9 @@ static void release_mm(struct mm_struct * mm)
 }
 
 
-static struct mm_struct * take_tasks_mm(struct task_struct * task)
+static struct mm_struct *take_tasks_mm(struct task_struct *task)
 {
-	struct mm_struct * mm = get_task_mm(task);
+	struct mm_struct *mm = get_task_mm(task);
 	if (mm)
 		down_read(&mm->mmap_sem);
 	return mm;
@@ -383,10 +465,10 @@ static inline int is_code(unsigned long val)
 {
 	return val == ESCAPE_CODE;
 }
- 
+
 
 /* "acquire" as many cpu buffer slots as we can */
-static unsigned long get_slots(struct oprofile_cpu_buffer * b)
+static unsigned long get_slots(struct oprofile_cpu_buffer *b)
 {
 	unsigned long head = b->head_pos;
 	unsigned long tail = b->tail_pos;
@@ -412,19 +494,6 @@ static unsigned long get_slots(struct oprofile_cpu_buffer * b)
 }
 
 
-static void increment_tail(struct oprofile_cpu_buffer * b)
-{
-	unsigned long new_tail = b->tail_pos + 1;
-
-	rmb();
-
-	if (new_tail < b->buffer_size)
-		b->tail_pos = new_tail;
-	else
-		b->tail_pos = 0;
-}
-
-
 /* Move tasks along towards death. Any tasks on dead_tasks
  * will definitely have no remaining references in any
  * CPU buffers at this point, because we use two lists,
@@ -435,8 +504,8 @@ static void process_task_mortuary(void)
 {
 	unsigned long flags;
 	LIST_HEAD(local_dead_tasks);
-	struct task_struct * task;
-	struct task_struct * ttask;
+	struct task_struct *task;
+	struct task_struct *ttask;
 
 	spin_lock_irqsave(&task_mortuary, flags);
 
@@ -493,7 +562,7 @@ void sync_buffer(int cpu)
 {
 	struct oprofile_cpu_buffer *cpu_buf = &per_cpu(cpu_buffer, cpu);
 	struct mm_struct *mm = NULL;
-	struct task_struct * new;
+	struct task_struct *new;
 	unsigned long cookie = 0;
 	int in_kernel = 1;
 	unsigned int i;
@@ -501,7 +570,7 @@ void sync_buffer(int cpu)
 	unsigned long available;
 
 	mutex_lock(&buffer_mutex);
- 
+
 	add_cpu_switch(cpu);
 
 	/* Remember, only we can modify tail_pos */
@@ -509,8 +578,8 @@ void sync_buffer(int cpu)
 	available = get_slots(cpu_buf);
 
 	for (i = 0; i < available; ++i) {
-		struct op_sample * s = &cpu_buf->buffer[cpu_buf->tail_pos];
- 
+		struct op_sample *s = &cpu_buf->buffer[cpu_buf->tail_pos];
+
 		if (is_code(s->eip)) {
 			if (s->event <= CPU_IS_KERNEL) {
 				/* kernel/userspace switch */
@@ -521,8 +590,18 @@ void sync_buffer(int cpu)
 			} else if (s->event == CPU_TRACE_BEGIN) {
 				state = sb_bt_start;
 				add_trace_begin();
+#ifdef CONFIG_OPROFILE_IBS
+			} else if (s->event == IBS_FETCH_BEGIN) {
+				state = sb_bt_start;
+				add_ibs_begin(cpu_buf,
+					IBS_FETCH_CODE, in_kernel, mm);
+			} else if (s->event == IBS_OP_BEGIN) {
+				state = sb_bt_start;
+				add_ibs_begin(cpu_buf,
+					IBS_OP_CODE, in_kernel, mm);
+#endif
 			} else {
-				struct mm_struct * oldmm = mm;
+				struct mm_struct *oldmm = mm;
 
 				/* userspace context switch */
 				new = (struct task_struct *)s->event;
@@ -533,13 +612,11 @@ void sync_buffer(int cpu)
 					cookie = get_exec_dcookie(mm);
 				add_user_ctx_switch(new, cookie);
 			}
-		} else {
-			if (state >= sb_bt_start &&
-			    !add_sample(mm, s, in_kernel)) {
-				if (state == sb_bt_start) {
-					state = sb_bt_ignore;
-					atomic_inc(&oprofile_stats.bt_lost_no_mapping);
-				}
+		} else if (state >= sb_bt_start &&
+			   !add_sample(mm, s, in_kernel)) {
+			if (state == sb_bt_start) {
+				state = sb_bt_ignore;
+				atomic_inc(&oprofile_stats.bt_lost_no_mapping);
 			}
 		}
 
diff --git a/drivers/oprofile/cpu_buffer.c b/drivers/oprofile/cpu_buffer.c
index 7ba78e6d210e..e1bd5a937f6c 100644
--- a/drivers/oprofile/cpu_buffer.c
+++ b/drivers/oprofile/cpu_buffer.c
@@ -5,6 +5,7 @@
  * @remark Read the file COPYING
  *
  * @author John Levon <levon@movementarian.org>
+ * @author Barry Kasindorf <barry.kasindorf@amd.com>
  *
  * Each CPU has a local buffer that stores PC value/event
  * pairs. We also log context switches when we notice them.
@@ -209,7 +210,7 @@ static int log_sample(struct oprofile_cpu_buffer * cpu_buf, unsigned long pc,
 	return 1;
 }
 
-static int oprofile_begin_trace(struct oprofile_cpu_buffer * cpu_buf)
+static int oprofile_begin_trace(struct oprofile_cpu_buffer *cpu_buf)
 {
 	if (nr_available_slots(cpu_buf) < 4) {
 		cpu_buf->sample_lost_overflow++;
@@ -254,6 +255,75 @@ void oprofile_add_sample(struct pt_regs * const regs, unsigned long event)
 	oprofile_add_ext_sample(pc, regs, event, is_kernel);
 }
 
+#ifdef CONFIG_OPROFILE_IBS
+
+#define MAX_IBS_SAMPLE_SIZE	14
+static int log_ibs_sample(struct oprofile_cpu_buffer *cpu_buf,
+	unsigned long pc, int is_kernel, unsigned  int *ibs, int ibs_code)
+{
+	struct task_struct *task;
+
+	cpu_buf->sample_received++;
+
+	if (nr_available_slots(cpu_buf) < MAX_IBS_SAMPLE_SIZE) {
+		cpu_buf->sample_lost_overflow++;
+		return 0;
+	}
+
+	is_kernel = !!is_kernel;
+
+	/* notice a switch from user->kernel or vice versa */
+	if (cpu_buf->last_is_kernel != is_kernel) {
+		cpu_buf->last_is_kernel = is_kernel;
+		add_code(cpu_buf, is_kernel);
+	}
+
+	/* notice a task switch */
+	if (!is_kernel) {
+		task = current;
+
+		if (cpu_buf->last_task != task) {
+			cpu_buf->last_task = task;
+			add_code(cpu_buf, (unsigned long)task);
+		}
+	}
+
+	add_code(cpu_buf, ibs_code);
+	add_sample(cpu_buf, ibs[0], ibs[1]);
+	add_sample(cpu_buf, ibs[2], ibs[3]);
+	add_sample(cpu_buf, ibs[4], ibs[5]);
+
+	if (ibs_code == IBS_OP_BEGIN) {
+	add_sample(cpu_buf, ibs[6], ibs[7]);
+	add_sample(cpu_buf, ibs[8], ibs[9]);
+	add_sample(cpu_buf, ibs[10], ibs[11]);
+	}
+
+	return 1;
+}
+
+void oprofile_add_ibs_sample(struct pt_regs *const regs,
+				unsigned int * const ibs_sample, u8 code)
+{
+	int is_kernel = !user_mode(regs);
+	unsigned long pc = profile_pc(regs);
+
+	struct oprofile_cpu_buffer *cpu_buf =
+			 &per_cpu(cpu_buffer, smp_processor_id());
+
+	if (!backtrace_depth) {
+		log_ibs_sample(cpu_buf, pc, is_kernel, ibs_sample, code);
+		return;
+	}
+
+	/* if log_sample() fails we can't backtrace since we lost the source
+	* of this event */
+	if (log_ibs_sample(cpu_buf, pc, is_kernel, ibs_sample, code))
+		oprofile_ops.backtrace(regs, backtrace_depth);
+}
+
+#endif
+
 void oprofile_add_pc(unsigned long pc, int is_kernel, unsigned long event)
 {
 	struct oprofile_cpu_buffer *cpu_buf = &__get_cpu_var(cpu_buffer);
@@ -296,7 +366,7 @@ static void wq_sync_buffer(struct work_struct *work)
 	struct oprofile_cpu_buffer * b =
 		container_of(work, struct oprofile_cpu_buffer, work.work);
 	if (b->cpu != smp_processor_id()) {
-		printk("WQ on CPU%d, prefer CPU%d\n",
+		printk(KERN_DEBUG "WQ on CPU%d, prefer CPU%d\n",
 		       smp_processor_id(), b->cpu);
 	}
 	sync_buffer(b->cpu);
diff --git a/drivers/oprofile/cpu_buffer.h b/drivers/oprofile/cpu_buffer.h
index c3e366b52261..9c44d004da69 100644
--- a/drivers/oprofile/cpu_buffer.h
+++ b/drivers/oprofile/cpu_buffer.h
@@ -55,5 +55,7 @@ void cpu_buffer_reset(struct oprofile_cpu_buffer * cpu_buf);
 /* transient events for the CPU buffer -> event buffer */
 #define CPU_IS_KERNEL 1
 #define CPU_TRACE_BEGIN 2
+#define IBS_FETCH_BEGIN 3
+#define IBS_OP_BEGIN    4
 
 #endif /* OPROFILE_CPU_BUFFER_H */
diff --git a/drivers/parport/parport_sunbpp.c b/drivers/parport/parport_sunbpp.c
index 9d595aa91e46..065f229580d5 100644
--- a/drivers/parport/parport_sunbpp.c
+++ b/drivers/parport/parport_sunbpp.c
@@ -26,6 +26,8 @@
 #include <linux/kernel.h>
 #include <linux/slab.h>
 #include <linux/init.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <linux/parport.h>
 
@@ -34,7 +36,6 @@
 
 #include <asm/io.h>
 #include <asm/oplib.h>           /* OpenProm Library */
-#include <asm/sbus.h>
 #include <asm/dma.h>             /* BPP uses LSI 64854 for DMA */
 #include <asm/irq.h>
 #include <asm/sunbpp.h>
@@ -285,38 +286,37 @@ static struct parport_operations parport_sunbpp_ops =
 	.owner		= THIS_MODULE,
 };
 
-static int __devinit init_one_port(struct sbus_dev *sdev)
+static int __devinit bpp_probe(struct of_device *op, const struct of_device_id *match)
 {
-	struct parport *p;
-	/* at least in theory there may be a "we don't dma" case */
 	struct parport_operations *ops;
-	void __iomem *base;
-	int irq, dma, err = 0, size;
 	struct bpp_regs __iomem *regs;
+	int irq, dma, err = 0, size;
 	unsigned char value_tcr;
+	void __iomem *base;
+	struct parport *p;
 
-	irq = sdev->irqs[0];
-	base = sbus_ioremap(&sdev->resource[0], 0,
-			    sdev->reg_addrs[0].reg_size, 
-			    "sunbpp");
+	irq = op->irqs[0];
+	base = of_ioremap(&op->resource[0], 0,
+			  resource_size(&op->resource[0]),
+			  "sunbpp");
 	if (!base)
 		return -ENODEV;
 
-	size = sdev->reg_addrs[0].reg_size;
+	size = resource_size(&op->resource[0]);
 	dma = PARPORT_DMA_NONE;
 
 	ops = kmalloc(sizeof(struct parport_operations), GFP_KERNEL);
         if (!ops)
 		goto out_unmap;
 
-        memcpy (ops, &parport_sunbpp_ops, sizeof (struct parport_operations));
+        memcpy (ops, &parport_sunbpp_ops, sizeof(struct parport_operations));
 
 	dprintk(("register_port\n"));
 	if (!(p = parport_register_port((unsigned long)base, irq, dma, ops)))
 		goto out_free_ops;
 
 	p->size = size;
-	p->dev = &sdev->ofdev.dev;
+	p->dev = &op->dev;
 
 	if ((err = request_irq(p->irq, parport_irq_handler,
 			       IRQF_SHARED, p->name, p)) != 0) {
@@ -333,7 +333,7 @@ static int __devinit init_one_port(struct sbus_dev *sdev)
 
 	printk(KERN_INFO "%s: sunbpp at 0x%lx\n", p->name, p->base);
 
-	dev_set_drvdata(&sdev->ofdev.dev, p);
+	dev_set_drvdata(&op->dev, p);
 
 	parport_announce_port(p);
 
@@ -346,21 +346,14 @@ out_free_ops:
 	kfree(ops);
 
 out_unmap:
-	sbus_iounmap(base, size);
+	of_iounmap(&op->resource[0], base, size);
 
 	return err;
 }
 
-static int __devinit bpp_probe(struct of_device *dev, const struct of_device_id *match)
-{
-	struct sbus_dev *sdev = to_sbus_device(&dev->dev);
-
-	return init_one_port(sdev);
-}
-
-static int __devexit bpp_remove(struct of_device *dev)
+static int __devexit bpp_remove(struct of_device *op)
 {
-	struct parport *p = dev_get_drvdata(&dev->dev);
+	struct parport *p = dev_get_drvdata(&op->dev);
 	struct parport_operations *ops = p->ops;
 
 	parport_remove_port(p);
@@ -370,16 +363,16 @@ static int __devexit bpp_remove(struct of_device *dev)
 		free_irq(p->irq, p);
 	}
 
-	sbus_iounmap((void __iomem *) p->base, p->size);
+	of_iounmap(&op->resource[0], (void __iomem *) p->base, p->size);
 	parport_put_port(p);
 	kfree(ops);
 
-	dev_set_drvdata(&dev->dev, NULL);
+	dev_set_drvdata(&op->dev, NULL);
 
 	return 0;
 }
 
-static struct of_device_id bpp_match[] = {
+static const struct of_device_id bpp_match[] = {
 	{
 		.name = "SUNW,bpp",
 	},
@@ -397,7 +390,7 @@ static struct of_platform_driver bpp_sbus_driver = {
 
 static int __init parport_sunbpp_init(void)
 {
-	return of_register_driver(&bpp_sbus_driver, &sbus_bus_type);
+	return of_register_driver(&bpp_sbus_driver, &of_bus_type);
 }
 
 static void __exit parport_sunbpp_exit(void)
diff --git a/drivers/pci/Makefile b/drivers/pci/Makefile
index 7d63f8ced24b..4b47f4ece5b7 100644
--- a/drivers/pci/Makefile
+++ b/drivers/pci/Makefile
@@ -26,6 +26,8 @@ obj-$(CONFIG_HT_IRQ) += htirq.o
 # Build Intel IOMMU support
 obj-$(CONFIG_DMAR) += dmar.o iova.o intel-iommu.o
 
+obj-$(CONFIG_INTR_REMAP) += dmar.o intr_remapping.o
+
 #
 # Some architectures use the generic PCI setup functions
 #
diff --git a/drivers/pci/dma_remapping.h b/drivers/pci/dma_remapping.h
new file mode 100644
index 000000000000..bff5c65f81dc
--- /dev/null
+++ b/drivers/pci/dma_remapping.h
@@ -0,0 +1,157 @@
+#ifndef _DMA_REMAPPING_H
+#define _DMA_REMAPPING_H
+
+/*
+ * We need a fixed PAGE_SIZE of 4K irrespective of
+ * arch PAGE_SIZE for IOMMU page tables.
+ */
+#define PAGE_SHIFT_4K		(12)
+#define PAGE_SIZE_4K		(1UL << PAGE_SHIFT_4K)
+#define PAGE_MASK_4K		(((u64)-1) << PAGE_SHIFT_4K)
+#define PAGE_ALIGN_4K(addr)	(((addr) + PAGE_SIZE_4K - 1) & PAGE_MASK_4K)
+
+#define IOVA_PFN(addr)		((addr) >> PAGE_SHIFT_4K)
+#define DMA_32BIT_PFN		IOVA_PFN(DMA_32BIT_MASK)
+#define DMA_64BIT_PFN		IOVA_PFN(DMA_64BIT_MASK)
+
+
+/*
+ * 0: Present
+ * 1-11: Reserved
+ * 12-63: Context Ptr (12 - (haw-1))
+ * 64-127: Reserved
+ */
+struct root_entry {
+	u64	val;
+	u64	rsvd1;
+};
+#define ROOT_ENTRY_NR (PAGE_SIZE_4K/sizeof(struct root_entry))
+static inline bool root_present(struct root_entry *root)
+{
+	return (root->val & 1);
+}
+static inline void set_root_present(struct root_entry *root)
+{
+	root->val |= 1;
+}
+static inline void set_root_value(struct root_entry *root, unsigned long value)
+{
+	root->val |= value & PAGE_MASK_4K;
+}
+
+struct context_entry;
+static inline struct context_entry *
+get_context_addr_from_root(struct root_entry *root)
+{
+	return (struct context_entry *)
+		(root_present(root)?phys_to_virt(
+		root->val & PAGE_MASK_4K):
+		NULL);
+}
+
+/*
+ * low 64 bits:
+ * 0: present
+ * 1: fault processing disable
+ * 2-3: translation type
+ * 12-63: address space root
+ * high 64 bits:
+ * 0-2: address width
+ * 3-6: aval
+ * 8-23: domain id
+ */
+struct context_entry {
+	u64 lo;
+	u64 hi;
+};
+#define context_present(c) ((c).lo & 1)
+#define context_fault_disable(c) (((c).lo >> 1) & 1)
+#define context_translation_type(c) (((c).lo >> 2) & 3)
+#define context_address_root(c) ((c).lo & PAGE_MASK_4K)
+#define context_address_width(c) ((c).hi &  7)
+#define context_domain_id(c) (((c).hi >> 8) & ((1 << 16) - 1))
+
+#define context_set_present(c) do {(c).lo |= 1;} while (0)
+#define context_set_fault_enable(c) \
+	do {(c).lo &= (((u64)-1) << 2) | 1;} while (0)
+#define context_set_translation_type(c, val) \
+	do { \
+		(c).lo &= (((u64)-1) << 4) | 3; \
+		(c).lo |= ((val) & 3) << 2; \
+	} while (0)
+#define CONTEXT_TT_MULTI_LEVEL 0
+#define context_set_address_root(c, val) \
+	do {(c).lo |= (val) & PAGE_MASK_4K;} while (0)
+#define context_set_address_width(c, val) do {(c).hi |= (val) & 7;} while (0)
+#define context_set_domain_id(c, val) \
+	do {(c).hi |= ((val) & ((1 << 16) - 1)) << 8;} while (0)
+#define context_clear_entry(c) do {(c).lo = 0; (c).hi = 0;} while (0)
+
+/*
+ * 0: readable
+ * 1: writable
+ * 2-6: reserved
+ * 7: super page
+ * 8-11: available
+ * 12-63: Host physcial address
+ */
+struct dma_pte {
+	u64 val;
+};
+#define dma_clear_pte(p)	do {(p).val = 0;} while (0)
+
+#define DMA_PTE_READ (1)
+#define DMA_PTE_WRITE (2)
+
+#define dma_set_pte_readable(p) do {(p).val |= DMA_PTE_READ;} while (0)
+#define dma_set_pte_writable(p) do {(p).val |= DMA_PTE_WRITE;} while (0)
+#define dma_set_pte_prot(p, prot) \
+		do {(p).val = ((p).val & ~3) | ((prot) & 3); } while (0)
+#define dma_pte_addr(p) ((p).val & PAGE_MASK_4K)
+#define dma_set_pte_addr(p, addr) do {\
+		(p).val |= ((addr) & PAGE_MASK_4K); } while (0)
+#define dma_pte_present(p) (((p).val & 3) != 0)
+
+struct intel_iommu;
+
+struct dmar_domain {
+	int	id;			/* domain id */
+	struct intel_iommu *iommu;	/* back pointer to owning iommu */
+
+	struct list_head devices; 	/* all devices' list */
+	struct iova_domain iovad;	/* iova's that belong to this domain */
+
+	struct dma_pte	*pgd;		/* virtual address */
+	spinlock_t	mapping_lock;	/* page table lock */
+	int		gaw;		/* max guest address width */
+
+	/* adjusted guest address width, 0 is level 2 30-bit */
+	int		agaw;
+
+#define DOMAIN_FLAG_MULTIPLE_DEVICES 1
+	int		flags;
+};
+
+/* PCI domain-device relationship */
+struct device_domain_info {
+	struct list_head link;	/* link to domain siblings */
+	struct list_head global; /* link to global list */
+	u8 bus;			/* PCI bus numer */
+	u8 devfn;		/* PCI devfn number */
+	struct pci_dev *dev; /* it's NULL for PCIE-to-PCI bridge */
+	struct dmar_domain *domain; /* pointer to domain */
+};
+
+extern int init_dmars(void);
+extern void free_dmar_iommu(struct intel_iommu *iommu);
+
+extern int dmar_disabled;
+
+#ifndef CONFIG_DMAR_GFX_WA
+static inline void iommu_prepare_gfx_mapping(void)
+{
+	return;
+}
+#endif /* !CONFIG_DMAR_GFX_WA */
+
+#endif
diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index 8bf86ae2333f..bd2c01674f5e 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -19,13 +19,16 @@
  * Author: Shaohua Li <shaohua.li@intel.com>
  * Author: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
  *
- * This file implements early detection/parsing of DMA Remapping Devices
+ * This file implements early detection/parsing of Remapping Devices
  * reported to OS through BIOS via DMA remapping reporting (DMAR) ACPI
  * tables.
+ *
+ * These routines are used by both DMA-remapping and Interrupt-remapping
  */
 
 #include <linux/pci.h>
 #include <linux/dmar.h>
+#include <linux/timer.h>
 #include "iova.h"
 #include "intel-iommu.h"
 
@@ -37,7 +40,6 @@
  * these units are not supported by the architecture.
  */
 LIST_HEAD(dmar_drhd_units);
-LIST_HEAD(dmar_rmrr_units);
 
 static struct acpi_table_header * __initdata dmar_tbl;
 
@@ -53,11 +55,6 @@ static void __init dmar_register_drhd_unit(struct dmar_drhd_unit *drhd)
 		list_add(&drhd->list, &dmar_drhd_units);
 }
 
-static void __init dmar_register_rmrr_unit(struct dmar_rmrr_unit *rmrr)
-{
-	list_add(&rmrr->list, &dmar_rmrr_units);
-}
-
 static int __init dmar_parse_one_dev_scope(struct acpi_dmar_device_scope *scope,
 					   struct pci_dev **dev, u16 segment)
 {
@@ -172,19 +169,37 @@ dmar_parse_one_drhd(struct acpi_dmar_header *header)
 	struct acpi_dmar_hardware_unit *drhd;
 	struct dmar_drhd_unit *dmaru;
 	int ret = 0;
-	static int include_all;
 
 	dmaru = kzalloc(sizeof(*dmaru), GFP_KERNEL);
 	if (!dmaru)
 		return -ENOMEM;
 
+	dmaru->hdr = header;
 	drhd = (struct acpi_dmar_hardware_unit *)header;
 	dmaru->reg_base_addr = drhd->address;
 	dmaru->include_all = drhd->flags & 0x1; /* BIT0: INCLUDE_ALL */
 
+	ret = alloc_iommu(dmaru);
+	if (ret) {
+		kfree(dmaru);
+		return ret;
+	}
+	dmar_register_drhd_unit(dmaru);
+	return 0;
+}
+
+static int __init
+dmar_parse_dev(struct dmar_drhd_unit *dmaru)
+{
+	struct acpi_dmar_hardware_unit *drhd;
+	static int include_all;
+	int ret;
+
+	drhd = (struct acpi_dmar_hardware_unit *) dmaru->hdr;
+
 	if (!dmaru->include_all)
 		ret = dmar_parse_dev_scope((void *)(drhd + 1),
-				((void *)drhd) + header->length,
+				((void *)drhd) + drhd->header.length,
 				&dmaru->devices_cnt, &dmaru->devices,
 				drhd->segment);
 	else {
@@ -197,37 +212,59 @@ dmar_parse_one_drhd(struct acpi_dmar_header *header)
 		include_all = 1;
 	}
 
-	if (ret || (dmaru->devices_cnt == 0 && !dmaru->include_all))
+	if (ret || (dmaru->devices_cnt == 0 && !dmaru->include_all)) {
+		list_del(&dmaru->list);
 		kfree(dmaru);
-	else
-		dmar_register_drhd_unit(dmaru);
+	}
 	return ret;
 }
 
+#ifdef CONFIG_DMAR
+LIST_HEAD(dmar_rmrr_units);
+
+static void __init dmar_register_rmrr_unit(struct dmar_rmrr_unit *rmrr)
+{
+	list_add(&rmrr->list, &dmar_rmrr_units);
+}
+
+
 static int __init
 dmar_parse_one_rmrr(struct acpi_dmar_header *header)
 {
 	struct acpi_dmar_reserved_memory *rmrr;
 	struct dmar_rmrr_unit *rmrru;
-	int ret = 0;
 
 	rmrru = kzalloc(sizeof(*rmrru), GFP_KERNEL);
 	if (!rmrru)
 		return -ENOMEM;
 
+	rmrru->hdr = header;
 	rmrr = (struct acpi_dmar_reserved_memory *)header;
 	rmrru->base_address = rmrr->base_address;
 	rmrru->end_address = rmrr->end_address;
+
+	dmar_register_rmrr_unit(rmrru);
+	return 0;
+}
+
+static int __init
+rmrr_parse_dev(struct dmar_rmrr_unit *rmrru)
+{
+	struct acpi_dmar_reserved_memory *rmrr;
+	int ret;
+
+	rmrr = (struct acpi_dmar_reserved_memory *) rmrru->hdr;
 	ret = dmar_parse_dev_scope((void *)(rmrr + 1),
-		((void *)rmrr) + header->length,
+		((void *)rmrr) + rmrr->header.length,
 		&rmrru->devices_cnt, &rmrru->devices, rmrr->segment);
 
-	if (ret || (rmrru->devices_cnt == 0))
+	if (ret || (rmrru->devices_cnt == 0)) {
+		list_del(&rmrru->list);
 		kfree(rmrru);
-	else
-		dmar_register_rmrr_unit(rmrru);
+	}
 	return ret;
 }
+#endif
 
 static void __init
 dmar_table_print_dmar_entry(struct acpi_dmar_header *header)
@@ -252,6 +289,7 @@ dmar_table_print_dmar_entry(struct acpi_dmar_header *header)
 	}
 }
 
+
 /**
  * parse_dmar_table - parses the DMA reporting table
  */
@@ -284,7 +322,9 @@ parse_dmar_table(void)
 			ret = dmar_parse_one_drhd(entry_header);
 			break;
 		case ACPI_DMAR_TYPE_RESERVED_MEMORY:
+#ifdef CONFIG_DMAR
 			ret = dmar_parse_one_rmrr(entry_header);
+#endif
 			break;
 		default:
 			printk(KERN_WARNING PREFIX
@@ -300,15 +340,77 @@ parse_dmar_table(void)
 	return ret;
 }
 
+int dmar_pci_device_match(struct pci_dev *devices[], int cnt,
+			  struct pci_dev *dev)
+{
+	int index;
+
+	while (dev) {
+		for (index = 0; index < cnt; index++)
+			if (dev == devices[index])
+				return 1;
 
-int __init dmar_table_init(void)
+		/* Check our parent */
+		dev = dev->bus->self;
+	}
+
+	return 0;
+}
+
+struct dmar_drhd_unit *
+dmar_find_matched_drhd_unit(struct pci_dev *dev)
 {
+	struct dmar_drhd_unit *drhd = NULL;
+
+	list_for_each_entry(drhd, &dmar_drhd_units, list) {
+		if (drhd->include_all || dmar_pci_device_match(drhd->devices,
+						drhd->devices_cnt, dev))
+			return drhd;
+	}
+
+	return NULL;
+}
+
+int __init dmar_dev_scope_init(void)
+{
+	struct dmar_drhd_unit *drhd;
+	int ret = -ENODEV;
+
+	for_each_drhd_unit(drhd) {
+		ret = dmar_parse_dev(drhd);
+		if (ret)
+			return ret;
+	}
+
+#ifdef CONFIG_DMAR
+	{
+		struct dmar_rmrr_unit *rmrr;
+		for_each_rmrr_units(rmrr) {
+			ret = rmrr_parse_dev(rmrr);
+			if (ret)
+				return ret;
+		}
+	}
+#endif
+
+	return ret;
+}
 
+
+int __init dmar_table_init(void)
+{
+	static int dmar_table_initialized;
 	int ret;
 
+	if (dmar_table_initialized)
+		return 0;
+
+	dmar_table_initialized = 1;
+
 	ret = parse_dmar_table();
 	if (ret) {
-		printk(KERN_INFO PREFIX "parse DMAR table failure.\n");
+		if (ret != -ENODEV)
+			printk(KERN_INFO PREFIX "parse DMAR table failure.\n");
 		return ret;
 	}
 
@@ -317,9 +419,14 @@ int __init dmar_table_init(void)
 		return -ENODEV;
 	}
 
+#ifdef CONFIG_DMAR
 	if (list_empty(&dmar_rmrr_units))
 		printk(KERN_INFO PREFIX "No RMRR found\n");
+#endif
 
+#ifdef CONFIG_INTR_REMAP
+	parse_ioapics_under_ir();
+#endif
 	return 0;
 }
 
@@ -341,3 +448,255 @@ int __init early_dmar_detect(void)
 
 	return (ACPI_SUCCESS(status) ? 1 : 0);
 }
+
+void __init detect_intel_iommu(void)
+{
+	int ret;
+
+	ret = early_dmar_detect();
+
+#ifdef CONFIG_DMAR
+	{
+		struct acpi_table_dmar *dmar;
+		/*
+		 * for now we will disable dma-remapping when interrupt
+		 * remapping is enabled.
+		 * When support for queued invalidation for IOTLB invalidation
+		 * is added, we will not need this any more.
+		 */
+		dmar = (struct acpi_table_dmar *) dmar_tbl;
+		if (ret && cpu_has_x2apic && dmar->flags & 0x1) {
+			printk(KERN_INFO
+			       "Queued invalidation will be enabled to support "
+			       "x2apic and Intr-remapping.\n");
+			printk(KERN_INFO
+			       "Disabling IOMMU detection, because of missing "
+			       "queued invalidation support for IOTLB "
+			       "invalidation\n");
+			printk(KERN_INFO
+			       "Use \"nox2apic\", if you want to use Intel "
+			       " IOMMU for DMA-remapping and don't care about "
+			       " x2apic support\n");
+
+			dmar_disabled = 1;
+			return;
+		}
+
+		if (ret && !no_iommu && !iommu_detected && !swiotlb &&
+		    !dmar_disabled)
+			iommu_detected = 1;
+	}
+#endif
+}
+
+
+int alloc_iommu(struct dmar_drhd_unit *drhd)
+{
+	struct intel_iommu *iommu;
+	int map_size;
+	u32 ver;
+	static int iommu_allocated = 0;
+
+	iommu = kzalloc(sizeof(*iommu), GFP_KERNEL);
+	if (!iommu)
+		return -ENOMEM;
+
+	iommu->seq_id = iommu_allocated++;
+
+	iommu->reg = ioremap(drhd->reg_base_addr, PAGE_SIZE_4K);
+	if (!iommu->reg) {
+		printk(KERN_ERR "IOMMU: can't map the region\n");
+		goto error;
+	}
+	iommu->cap = dmar_readq(iommu->reg + DMAR_CAP_REG);
+	iommu->ecap = dmar_readq(iommu->reg + DMAR_ECAP_REG);
+
+	/* the registers might be more than one page */
+	map_size = max_t(int, ecap_max_iotlb_offset(iommu->ecap),
+		cap_max_fault_reg_offset(iommu->cap));
+	map_size = PAGE_ALIGN_4K(map_size);
+	if (map_size > PAGE_SIZE_4K) {
+		iounmap(iommu->reg);
+		iommu->reg = ioremap(drhd->reg_base_addr, map_size);
+		if (!iommu->reg) {
+			printk(KERN_ERR "IOMMU: can't map the region\n");
+			goto error;
+		}
+	}
+
+	ver = readl(iommu->reg + DMAR_VER_REG);
+	pr_debug("IOMMU %llx: ver %d:%d cap %llx ecap %llx\n",
+		drhd->reg_base_addr, DMAR_VER_MAJOR(ver), DMAR_VER_MINOR(ver),
+		iommu->cap, iommu->ecap);
+
+	spin_lock_init(&iommu->register_lock);
+
+	drhd->iommu = iommu;
+	return 0;
+error:
+	kfree(iommu);
+	return -1;
+}
+
+void free_iommu(struct intel_iommu *iommu)
+{
+	if (!iommu)
+		return;
+
+#ifdef CONFIG_DMAR
+	free_dmar_iommu(iommu);
+#endif
+
+	if (iommu->reg)
+		iounmap(iommu->reg);
+	kfree(iommu);
+}
+
+/*
+ * Reclaim all the submitted descriptors which have completed its work.
+ */
+static inline void reclaim_free_desc(struct q_inval *qi)
+{
+	while (qi->desc_status[qi->free_tail] == QI_DONE) {
+		qi->desc_status[qi->free_tail] = QI_FREE;
+		qi->free_tail = (qi->free_tail + 1) % QI_LENGTH;
+		qi->free_cnt++;
+	}
+}
+
+/*
+ * Submit the queued invalidation descriptor to the remapping
+ * hardware unit and wait for its completion.
+ */
+void qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu)
+{
+	struct q_inval *qi = iommu->qi;
+	struct qi_desc *hw, wait_desc;
+	int wait_index, index;
+	unsigned long flags;
+
+	if (!qi)
+		return;
+
+	hw = qi->desc;
+
+	spin_lock(&qi->q_lock);
+	while (qi->free_cnt < 3) {
+		spin_unlock(&qi->q_lock);
+		cpu_relax();
+		spin_lock(&qi->q_lock);
+	}
+
+	index = qi->free_head;
+	wait_index = (index + 1) % QI_LENGTH;
+
+	qi->desc_status[index] = qi->desc_status[wait_index] = QI_IN_USE;
+
+	hw[index] = *desc;
+
+	wait_desc.low = QI_IWD_STATUS_DATA(2) | QI_IWD_STATUS_WRITE | QI_IWD_TYPE;
+	wait_desc.high = virt_to_phys(&qi->desc_status[wait_index]);
+
+	hw[wait_index] = wait_desc;
+
+	__iommu_flush_cache(iommu, &hw[index], sizeof(struct qi_desc));
+	__iommu_flush_cache(iommu, &hw[wait_index], sizeof(struct qi_desc));
+
+	qi->free_head = (qi->free_head + 2) % QI_LENGTH;
+	qi->free_cnt -= 2;
+
+	spin_lock_irqsave(&iommu->register_lock, flags);
+	/*
+	 * update the HW tail register indicating the presence of
+	 * new descriptors.
+	 */
+	writel(qi->free_head << 4, iommu->reg + DMAR_IQT_REG);
+	spin_unlock_irqrestore(&iommu->register_lock, flags);
+
+	while (qi->desc_status[wait_index] != QI_DONE) {
+		spin_unlock(&qi->q_lock);
+		cpu_relax();
+		spin_lock(&qi->q_lock);
+	}
+
+	qi->desc_status[index] = QI_DONE;
+
+	reclaim_free_desc(qi);
+	spin_unlock(&qi->q_lock);
+}
+
+/*
+ * Flush the global interrupt entry cache.
+ */
+void qi_global_iec(struct intel_iommu *iommu)
+{
+	struct qi_desc desc;
+
+	desc.low = QI_IEC_TYPE;
+	desc.high = 0;
+
+	qi_submit_sync(&desc, iommu);
+}
+
+/*
+ * Enable Queued Invalidation interface. This is a must to support
+ * interrupt-remapping. Also used by DMA-remapping, which replaces
+ * register based IOTLB invalidation.
+ */
+int dmar_enable_qi(struct intel_iommu *iommu)
+{
+	u32 cmd, sts;
+	unsigned long flags;
+	struct q_inval *qi;
+
+	if (!ecap_qis(iommu->ecap))
+		return -ENOENT;
+
+	/*
+	 * queued invalidation is already setup and enabled.
+	 */
+	if (iommu->qi)
+		return 0;
+
+	iommu->qi = kmalloc(sizeof(*qi), GFP_KERNEL);
+	if (!iommu->qi)
+		return -ENOMEM;
+
+	qi = iommu->qi;
+
+	qi->desc = (void *)(get_zeroed_page(GFP_KERNEL));
+	if (!qi->desc) {
+		kfree(qi);
+		iommu->qi = 0;
+		return -ENOMEM;
+	}
+
+	qi->desc_status = kmalloc(QI_LENGTH * sizeof(int), GFP_KERNEL);
+	if (!qi->desc_status) {
+		free_page((unsigned long) qi->desc);
+		kfree(qi);
+		iommu->qi = 0;
+		return -ENOMEM;
+	}
+
+	qi->free_head = qi->free_tail = 0;
+	qi->free_cnt = QI_LENGTH;
+
+	spin_lock_init(&qi->q_lock);
+
+	spin_lock_irqsave(&iommu->register_lock, flags);
+	/* write zero to the tail reg */
+	writel(0, iommu->reg + DMAR_IQT_REG);
+
+	dmar_writeq(iommu->reg + DMAR_IQA_REG, virt_to_phys(qi->desc));
+
+	cmd = iommu->gcmd | DMA_GCMD_QIE;
+	iommu->gcmd |= DMA_GCMD_QIE;
+	writel(cmd, iommu->reg + DMAR_GCMD_REG);
+
+	/* Make sure hardware complete it */
+	IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, readl, (sts & DMA_GSTS_QIES), sts);
+	spin_unlock_irqrestore(&iommu->register_lock, flags);
+
+	return 0;
+}
diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 6c4c1c3c50ee..389fdd6f4a9f 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -49,8 +49,6 @@
 
 #define DEFAULT_DOMAIN_ADDRESS_WIDTH 48
 
-#define DMAR_OPERATION_TIMEOUT ((cycles_t) tsc_khz*10*1000) /* 10sec */
-
 #define DOMAIN_MAX_ADDR(gaw) ((((u64)1) << gaw) - 1)
 
 
@@ -58,8 +56,6 @@ static void flush_unmaps_timeout(unsigned long data);
 
 DEFINE_TIMER(unmap_timer,  flush_unmaps_timeout, 0, 0);
 
-static struct intel_iommu *g_iommus;
-
 #define HIGH_WATER_MARK 250
 struct deferred_flush_tables {
 	int next;
@@ -185,13 +181,6 @@ void free_iova_mem(struct iova *iova)
 	kmem_cache_free(iommu_iova_cache, iova);
 }
 
-static inline void __iommu_flush_cache(
-	struct intel_iommu *iommu, void *addr, int size)
-{
-	if (!ecap_coherent(iommu->ecap))
-		clflush_cache_range(addr, size);
-}
-
 /* Gets context entry for a given bus and devfn */
 static struct context_entry * device_to_context_entry(struct intel_iommu *iommu,
 		u8 bus, u8 devfn)
@@ -488,19 +477,6 @@ static int iommu_alloc_root_entry(struct intel_iommu *iommu)
 	return 0;
 }
 
-#define IOMMU_WAIT_OP(iommu, offset, op, cond, sts) \
-{\
-	cycles_t start_time = get_cycles();\
-	while (1) {\
-		sts = op (iommu->reg + offset);\
-		if (cond)\
-			break;\
-		if (DMAR_OPERATION_TIMEOUT < (get_cycles() - start_time))\
-			panic("DMAR hardware is malfunctioning\n");\
-		cpu_relax();\
-	}\
-}
-
 static void iommu_set_root_entry(struct intel_iommu *iommu)
 {
 	void *addr;
@@ -990,6 +966,8 @@ static int iommu_init_domains(struct intel_iommu *iommu)
 		return -ENOMEM;
 	}
 
+	spin_lock_init(&iommu->lock);
+
 	/*
 	 * if Caching mode is set, then invalid translations are tagged
 	 * with domainid 0. Hence we need to pre-allocate it.
@@ -998,62 +976,15 @@ static int iommu_init_domains(struct intel_iommu *iommu)
 		set_bit(0, iommu->domain_ids);
 	return 0;
 }
-static struct intel_iommu *alloc_iommu(struct intel_iommu *iommu,
-					struct dmar_drhd_unit *drhd)
-{
-	int ret;
-	int map_size;
-	u32 ver;
-
-	iommu->reg = ioremap(drhd->reg_base_addr, PAGE_SIZE_4K);
-	if (!iommu->reg) {
-		printk(KERN_ERR "IOMMU: can't map the region\n");
-		goto error;
-	}
-	iommu->cap = dmar_readq(iommu->reg + DMAR_CAP_REG);
-	iommu->ecap = dmar_readq(iommu->reg + DMAR_ECAP_REG);
-
-	/* the registers might be more than one page */
-	map_size = max_t(int, ecap_max_iotlb_offset(iommu->ecap),
-		cap_max_fault_reg_offset(iommu->cap));
-	map_size = PAGE_ALIGN_4K(map_size);
-	if (map_size > PAGE_SIZE_4K) {
-		iounmap(iommu->reg);
-		iommu->reg = ioremap(drhd->reg_base_addr, map_size);
-		if (!iommu->reg) {
-			printk(KERN_ERR "IOMMU: can't map the region\n");
-			goto error;
-		}
-	}
-
-	ver = readl(iommu->reg + DMAR_VER_REG);
-	pr_debug("IOMMU %llx: ver %d:%d cap %llx ecap %llx\n",
-		drhd->reg_base_addr, DMAR_VER_MAJOR(ver), DMAR_VER_MINOR(ver),
-		iommu->cap, iommu->ecap);
-	ret = iommu_init_domains(iommu);
-	if (ret)
-		goto error_unmap;
-	spin_lock_init(&iommu->lock);
-	spin_lock_init(&iommu->register_lock);
 
-	drhd->iommu = iommu;
-	return iommu;
-error_unmap:
-	iounmap(iommu->reg);
-error:
-	kfree(iommu);
-	return NULL;
-}
 
 static void domain_exit(struct dmar_domain *domain);
-static void free_iommu(struct intel_iommu *iommu)
+
+void free_dmar_iommu(struct intel_iommu *iommu)
 {
 	struct dmar_domain *domain;
 	int i;
 
-	if (!iommu)
-		return;
-
 	i = find_first_bit(iommu->domain_ids, cap_ndoms(iommu->cap));
 	for (; i < cap_ndoms(iommu->cap); ) {
 		domain = iommu->domains[i];
@@ -1078,10 +1009,6 @@ static void free_iommu(struct intel_iommu *iommu)
 
 	/* free context mapping */
 	free_context_table(iommu);
-
-	if (iommu->reg)
-		iounmap(iommu->reg);
-	kfree(iommu);
 }
 
 static struct dmar_domain * iommu_alloc_domain(struct intel_iommu *iommu)
@@ -1426,37 +1353,6 @@ find_domain(struct pci_dev *pdev)
 	return NULL;
 }
 
-static int dmar_pci_device_match(struct pci_dev *devices[], int cnt,
-     struct pci_dev *dev)
-{
-	int index;
-
-	while (dev) {
-		for (index = 0; index < cnt; index++)
-			if (dev == devices[index])
-				return 1;
-
-		/* Check our parent */
-		dev = dev->bus->self;
-	}
-
-	return 0;
-}
-
-static struct dmar_drhd_unit *
-dmar_find_matched_drhd_unit(struct pci_dev *dev)
-{
-	struct dmar_drhd_unit *drhd = NULL;
-
-	list_for_each_entry(drhd, &dmar_drhd_units, list) {
-		if (drhd->include_all || dmar_pci_device_match(drhd->devices,
-						drhd->devices_cnt, dev))
-			return drhd;
-	}
-
-	return NULL;
-}
-
 /* domain is initialized */
 static struct dmar_domain *get_domain_for_dev(struct pci_dev *pdev, int gaw)
 {
@@ -1729,8 +1625,6 @@ int __init init_dmars(void)
 	 * endfor
 	 */
 	for_each_drhd_unit(drhd) {
-		if (drhd->ignored)
-			continue;
 		g_num_of_iommus++;
 		/*
 		 * lock not needed as this is only incremented in the single
@@ -1739,12 +1633,6 @@ int __init init_dmars(void)
 		 */
 	}
 
-	g_iommus = kzalloc(g_num_of_iommus * sizeof(*iommu), GFP_KERNEL);
-	if (!g_iommus) {
-		ret = -ENOMEM;
-		goto error;
-	}
-
 	deferred_flush = kzalloc(g_num_of_iommus *
 		sizeof(struct deferred_flush_tables), GFP_KERNEL);
 	if (!deferred_flush) {
@@ -1752,16 +1640,15 @@ int __init init_dmars(void)
 		goto error;
 	}
 
-	i = 0;
 	for_each_drhd_unit(drhd) {
 		if (drhd->ignored)
 			continue;
-		iommu = alloc_iommu(&g_iommus[i], drhd);
-		i++;
-		if (!iommu) {
-			ret = -ENOMEM;
+
+		iommu = drhd->iommu;
+
+		ret = iommu_init_domains(iommu);
+		if (ret)
 			goto error;
-		}
 
 		/*
 		 * TBD:
@@ -1845,7 +1732,6 @@ error:
 		iommu = drhd->iommu;
 		free_iommu(iommu);
 	}
-	kfree(g_iommus);
 	return ret;
 }
 
@@ -2002,7 +1888,10 @@ static void flush_unmaps(void)
 	/* just flush them all */
 	for (i = 0; i < g_num_of_iommus; i++) {
 		if (deferred_flush[i].next) {
-			iommu_flush_iotlb_global(&g_iommus[i], 0);
+			struct intel_iommu *iommu =
+				deferred_flush[i].domain[0]->iommu;
+
+			iommu_flush_iotlb_global(iommu, 0);
 			for (j = 0; j < deferred_flush[i].next; j++) {
 				__free_iova(&deferred_flush[i].domain[j]->iovad,
 						deferred_flush[i].iova[j]);
@@ -2032,7 +1921,8 @@ static void add_unmap(struct dmar_domain *dom, struct iova *iova)
 	if (list_size == HIGH_WATER_MARK)
 		flush_unmaps();
 
-	iommu_id = dom->iommu - g_iommus;
+	iommu_id = dom->iommu->seq_id;
+
 	next = deferred_flush[iommu_id].next;
 	deferred_flush[iommu_id].domain[next] = dom;
 	deferred_flush[iommu_id].iova[next] = iova;
@@ -2348,38 +2238,6 @@ static void __init iommu_exit_mempool(void)
 
 }
 
-static int blacklist_iommu(const struct dmi_system_id *id)
-{
-	printk(KERN_INFO "%s detected; disabling IOMMU\n",
-	       id->ident);
-	dmar_disabled = 1;
-	return 0;
-}
-
-static struct dmi_system_id __initdata intel_iommu_dmi_table[] = {
-	{	/* Some DG33BU BIOS revisions advertised non-existent VT-d */
-		.callback = blacklist_iommu,
-		.ident = "Intel DG33BU",
-		{	DMI_MATCH(DMI_BOARD_VENDOR, "Intel Corporation"),
-			DMI_MATCH(DMI_BOARD_NAME, "DG33BU"),
-		}
-	},
-	{ }
-};
-
-
-void __init detect_intel_iommu(void)
-{
-	if (swiotlb || no_iommu || iommu_detected || dmar_disabled)
-		return;
-	if (early_dmar_detect()) {
-		dmi_check_system(intel_iommu_dmi_table);
-		if (dmar_disabled)
-			return;
-		iommu_detected = 1;
-	}
-}
-
 static void __init init_no_remapping_devices(void)
 {
 	struct dmar_drhd_unit *drhd;
@@ -2426,12 +2284,19 @@ int __init intel_iommu_init(void)
 {
 	int ret = 0;
 
-	if (no_iommu || swiotlb || dmar_disabled)
-		return -ENODEV;
-
 	if (dmar_table_init())
 		return 	-ENODEV;
 
+	if (dmar_dev_scope_init())
+		return 	-ENODEV;
+
+	/*
+	 * Check the need for DMA-remapping initialization now.
+	 * Above initialization will also be used by Interrupt-remapping.
+	 */
+	if (no_iommu || swiotlb || dmar_disabled)
+		return -ENODEV;
+
 	iommu_init_mempool();
 	dmar_init_reserved_ranges();
 
diff --git a/drivers/pci/intel-iommu.h b/drivers/pci/intel-iommu.h
index afc0ad96122e..2142c01e0143 100644
--- a/drivers/pci/intel-iommu.h
+++ b/drivers/pci/intel-iommu.h
@@ -27,19 +27,8 @@
 #include <linux/sysdev.h>
 #include "iova.h"
 #include <linux/io.h>
-
-/*
- * We need a fixed PAGE_SIZE of 4K irrespective of
- * arch PAGE_SIZE for IOMMU page tables.
- */
-#define PAGE_SHIFT_4K		(12)
-#define PAGE_SIZE_4K		(1UL << PAGE_SHIFT_4K)
-#define PAGE_MASK_4K		(((u64)-1) << PAGE_SHIFT_4K)
-#define PAGE_ALIGN_4K(addr)	(((addr) + PAGE_SIZE_4K - 1) & PAGE_MASK_4K)
-
-#define IOVA_PFN(addr)		((addr) >> PAGE_SHIFT_4K)
-#define DMA_32BIT_PFN		IOVA_PFN(DMA_32BIT_MASK)
-#define DMA_64BIT_PFN		IOVA_PFN(DMA_64BIT_MASK)
+#include <asm/cacheflush.h>
+#include "dma_remapping.h"
 
 /*
  * Intel IOMMU register specification per version 1.0 public spec.
@@ -63,6 +52,11 @@
 #define	DMAR_PLMLIMIT_REG 0x6c	/* PMRR low limit */
 #define	DMAR_PHMBASE_REG 0x70	/* pmrr high base addr */
 #define	DMAR_PHMLIMIT_REG 0x78	/* pmrr high limit */
+#define DMAR_IQH_REG	0x80	/* Invalidation queue head register */
+#define DMAR_IQT_REG	0x88	/* Invalidation queue tail register */
+#define DMAR_IQA_REG	0x90	/* Invalidation queue addr register */
+#define DMAR_ICS_REG	0x98	/* Invalidation complete status register */
+#define DMAR_IRTA_REG	0xb8    /* Interrupt remapping table addr register */
 
 #define OFFSET_STRIDE		(9)
 /*
@@ -126,6 +120,10 @@ static inline void dmar_writeq(void __iomem *addr, u64 val)
 #define ecap_max_iotlb_offset(e) \
 	(ecap_iotlb_offset(e) + ecap_niotlb_iunits(e) * 16)
 #define ecap_coherent(e)	((e) & 0x1)
+#define ecap_qis(e)		((e) & 0x2)
+#define ecap_eim_support(e)	((e >> 4) & 0x1)
+#define ecap_ir_support(e)	((e >> 3) & 0x1)
+#define ecap_max_handle_mask(e) ((e >> 20) & 0xf)
 
 
 /* IOTLB_REG */
@@ -141,6 +139,17 @@ static inline void dmar_writeq(void __iomem *addr, u64 val)
 #define DMA_TLB_IH_NONLEAF (((u64)1) << 6)
 #define DMA_TLB_MAX_SIZE (0x3f)
 
+/* INVALID_DESC */
+#define DMA_ID_TLB_GLOBAL_FLUSH	(((u64)1) << 3)
+#define DMA_ID_TLB_DSI_FLUSH	(((u64)2) << 3)
+#define DMA_ID_TLB_PSI_FLUSH	(((u64)3) << 3)
+#define DMA_ID_TLB_READ_DRAIN	(((u64)1) << 7)
+#define DMA_ID_TLB_WRITE_DRAIN	(((u64)1) << 6)
+#define DMA_ID_TLB_DID(id)	(((u64)((id & 0xffff) << 16)))
+#define DMA_ID_TLB_IH_NONLEAF	(((u64)1) << 6)
+#define DMA_ID_TLB_ADDR(addr)	(addr)
+#define DMA_ID_TLB_ADDR_MASK(mask)	(mask)
+
 /* PMEN_REG */
 #define DMA_PMEN_EPM (((u32)1)<<31)
 #define DMA_PMEN_PRS (((u32)1)<<0)
@@ -151,6 +160,9 @@ static inline void dmar_writeq(void __iomem *addr, u64 val)
 #define DMA_GCMD_SFL (((u32)1) << 29)
 #define DMA_GCMD_EAFL (((u32)1) << 28)
 #define DMA_GCMD_WBF (((u32)1) << 27)
+#define DMA_GCMD_QIE (((u32)1) << 26)
+#define DMA_GCMD_SIRTP (((u32)1) << 24)
+#define DMA_GCMD_IRE (((u32) 1) << 25)
 
 /* GSTS_REG */
 #define DMA_GSTS_TES (((u32)1) << 31)
@@ -158,6 +170,9 @@ static inline void dmar_writeq(void __iomem *addr, u64 val)
 #define DMA_GSTS_FLS (((u32)1) << 29)
 #define DMA_GSTS_AFLS (((u32)1) << 28)
 #define DMA_GSTS_WBFS (((u32)1) << 27)
+#define DMA_GSTS_QIES (((u32)1) << 26)
+#define DMA_GSTS_IRTPS (((u32)1) << 24)
+#define DMA_GSTS_IRES (((u32)1) << 25)
 
 /* CCMD_REG */
 #define DMA_CCMD_ICC (((u64)1) << 63)
@@ -187,158 +202,106 @@ static inline void dmar_writeq(void __iomem *addr, u64 val)
 #define dma_frcd_source_id(c) (c & 0xffff)
 #define dma_frcd_page_addr(d) (d & (((u64)-1) << 12)) /* low 64 bit */
 
-/*
- * 0: Present
- * 1-11: Reserved
- * 12-63: Context Ptr (12 - (haw-1))
- * 64-127: Reserved
- */
-struct root_entry {
-	u64	val;
-	u64	rsvd1;
-};
-#define ROOT_ENTRY_NR (PAGE_SIZE_4K/sizeof(struct root_entry))
-static inline bool root_present(struct root_entry *root)
-{
-	return (root->val & 1);
-}
-static inline void set_root_present(struct root_entry *root)
-{
-	root->val |= 1;
-}
-static inline void set_root_value(struct root_entry *root, unsigned long value)
-{
-	root->val |= value & PAGE_MASK_4K;
+#define DMAR_OPERATION_TIMEOUT ((cycles_t) tsc_khz*10*1000) /* 10sec */
+
+#define IOMMU_WAIT_OP(iommu, offset, op, cond, sts) \
+{\
+	cycles_t start_time = get_cycles();\
+	while (1) {\
+		sts = op (iommu->reg + offset);\
+		if (cond)\
+			break;\
+		if (DMAR_OPERATION_TIMEOUT < (get_cycles() - start_time))\
+			panic("DMAR hardware is malfunctioning\n");\
+		cpu_relax();\
+	}\
 }
 
-struct context_entry;
-static inline struct context_entry *
-get_context_addr_from_root(struct root_entry *root)
-{
-	return (struct context_entry *)
-		(root_present(root)?phys_to_virt(
-		root->val & PAGE_MASK_4K):
-		NULL);
-}
-
-/*
- * low 64 bits:
- * 0: present
- * 1: fault processing disable
- * 2-3: translation type
- * 12-63: address space root
- * high 64 bits:
- * 0-2: address width
- * 3-6: aval
- * 8-23: domain id
- */
-struct context_entry {
-	u64 lo;
-	u64 hi;
-};
-#define context_present(c) ((c).lo & 1)
-#define context_fault_disable(c) (((c).lo >> 1) & 1)
-#define context_translation_type(c) (((c).lo >> 2) & 3)
-#define context_address_root(c) ((c).lo & PAGE_MASK_4K)
-#define context_address_width(c) ((c).hi &  7)
-#define context_domain_id(c) (((c).hi >> 8) & ((1 << 16) - 1))
-
-#define context_set_present(c) do {(c).lo |= 1;} while (0)
-#define context_set_fault_enable(c) \
-	do {(c).lo &= (((u64)-1) << 2) | 1;} while (0)
-#define context_set_translation_type(c, val) \
-	do { \
-		(c).lo &= (((u64)-1) << 4) | 3; \
-		(c).lo |= ((val) & 3) << 2; \
-	} while (0)
-#define CONTEXT_TT_MULTI_LEVEL 0
-#define context_set_address_root(c, val) \
-	do {(c).lo |= (val) & PAGE_MASK_4K;} while (0)
-#define context_set_address_width(c, val) do {(c).hi |= (val) & 7;} while (0)
-#define context_set_domain_id(c, val) \
-	do {(c).hi |= ((val) & ((1 << 16) - 1)) << 8;} while (0)
-#define context_clear_entry(c) do {(c).lo = 0; (c).hi = 0;} while (0)
+#define QI_LENGTH	256	/* queue length */
 
-/*
- * 0: readable
- * 1: writable
- * 2-6: reserved
- * 7: super page
- * 8-11: available
- * 12-63: Host physcial address
- */
-struct dma_pte {
-	u64 val;
+enum {
+	QI_FREE,
+	QI_IN_USE,
+	QI_DONE
 };
-#define dma_clear_pte(p)	do {(p).val = 0;} while (0)
-
-#define DMA_PTE_READ (1)
-#define DMA_PTE_WRITE (2)
 
-#define dma_set_pte_readable(p) do {(p).val |= DMA_PTE_READ;} while (0)
-#define dma_set_pte_writable(p) do {(p).val |= DMA_PTE_WRITE;} while (0)
-#define dma_set_pte_prot(p, prot) \
-		do {(p).val = ((p).val & ~3) | ((prot) & 3); } while (0)
-#define dma_pte_addr(p) ((p).val & PAGE_MASK_4K)
-#define dma_set_pte_addr(p, addr) do {\
-		(p).val |= ((addr) & PAGE_MASK_4K); } while (0)
-#define dma_pte_present(p) (((p).val & 3) != 0)
+#define QI_CC_TYPE		0x1
+#define QI_IOTLB_TYPE		0x2
+#define QI_DIOTLB_TYPE		0x3
+#define QI_IEC_TYPE		0x4
+#define QI_IWD_TYPE		0x5
 
-struct intel_iommu;
+#define QI_IEC_SELECTIVE	(((u64)1) << 4)
+#define QI_IEC_IIDEX(idx)	(((u64)(idx & 0xffff) << 32))
+#define QI_IEC_IM(m)		(((u64)(m & 0x1f) << 27))
 
-struct dmar_domain {
-	int	id;			/* domain id */
-	struct intel_iommu *iommu;	/* back pointer to owning iommu */
+#define QI_IWD_STATUS_DATA(d)	(((u64)d) << 32)
+#define QI_IWD_STATUS_WRITE	(((u64)1) << 5)
 
-	struct list_head devices; 	/* all devices' list */
-	struct iova_domain iovad;	/* iova's that belong to this domain */
+struct qi_desc {
+	u64 low, high;
+};
 
-	struct dma_pte	*pgd;		/* virtual address */
-	spinlock_t	mapping_lock;	/* page table lock */
-	int		gaw;		/* max guest address width */
+struct q_inval {
+	spinlock_t      q_lock;
+	struct qi_desc  *desc;          /* invalidation queue */
+	int             *desc_status;   /* desc status */
+	int             free_head;      /* first free entry */
+	int             free_tail;      /* last free entry */
+	int             free_cnt;
+};
 
-	/* adjusted guest address width, 0 is level 2 30-bit */
-	int		agaw;
+#ifdef CONFIG_INTR_REMAP
+/* 1MB - maximum possible interrupt remapping table size */
+#define INTR_REMAP_PAGE_ORDER	8
+#define INTR_REMAP_TABLE_REG_SIZE	0xf
 
-#define DOMAIN_FLAG_MULTIPLE_DEVICES 1
-	int		flags;
-};
+#define INTR_REMAP_TABLE_ENTRIES	65536
 
-/* PCI domain-device relationship */
-struct device_domain_info {
-	struct list_head link;	/* link to domain siblings */
-	struct list_head global; /* link to global list */
-	u8 bus;			/* PCI bus numer */
-	u8 devfn;		/* PCI devfn number */
-	struct pci_dev *dev; /* it's NULL for PCIE-to-PCI bridge */
-	struct dmar_domain *domain; /* pointer to domain */
+struct ir_table {
+	struct irte *base;
 };
-
-extern int init_dmars(void);
+#endif
 
 struct intel_iommu {
 	void __iomem	*reg; /* Pointer to hardware regs, virtual addr */
 	u64		cap;
 	u64		ecap;
-	unsigned long 	*domain_ids; /* bitmap of domains */
-	struct dmar_domain **domains; /* ptr to domains */
 	int		seg;
 	u32		gcmd; /* Holds TE, EAFL. Don't need SRTP, SFL, WBF */
-	spinlock_t	lock; /* protect context, domain ids */
 	spinlock_t	register_lock; /* protect register handling */
+	int		seq_id;	/* sequence id of the iommu */
+
+#ifdef CONFIG_DMAR
+	unsigned long 	*domain_ids; /* bitmap of domains */
+	struct dmar_domain **domains; /* ptr to domains */
+	spinlock_t	lock; /* protect context, domain ids */
 	struct root_entry *root_entry; /* virtual address */
 
 	unsigned int irq;
 	unsigned char name[7];    /* Device Name */
 	struct msi_msg saved_msg;
 	struct sys_device sysdev;
+#endif
+	struct q_inval  *qi;            /* Queued invalidation info */
+#ifdef CONFIG_INTR_REMAP
+	struct ir_table *ir_table;	/* Interrupt remapping info */
+#endif
 };
 
-#ifndef CONFIG_DMAR_GFX_WA
-static inline void iommu_prepare_gfx_mapping(void)
+static inline void __iommu_flush_cache(
+	struct intel_iommu *iommu, void *addr, int size)
 {
-	return;
+	if (!ecap_coherent(iommu->ecap))
+		clflush_cache_range(addr, size);
 }
-#endif /* !CONFIG_DMAR_GFX_WA */
 
+extern struct dmar_drhd_unit * dmar_find_matched_drhd_unit(struct pci_dev *dev);
+
+extern int alloc_iommu(struct dmar_drhd_unit *drhd);
+extern void free_iommu(struct intel_iommu *iommu);
+extern int dmar_enable_qi(struct intel_iommu *iommu);
+extern void qi_global_iec(struct intel_iommu *iommu);
+
+extern void qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu);
 #endif
diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
new file mode 100644
index 000000000000..bb642cc5e18c
--- /dev/null
+++ b/drivers/pci/intr_remapping.c
@@ -0,0 +1,471 @@
+#include <linux/dmar.h>
+#include <linux/spinlock.h>
+#include <linux/jiffies.h>
+#include <linux/pci.h>
+#include <linux/irq.h>
+#include <asm/io_apic.h>
+#include "intel-iommu.h"
+#include "intr_remapping.h"
+
+static struct ioapic_scope ir_ioapic[MAX_IO_APICS];
+static int ir_ioapic_num;
+int intr_remapping_enabled;
+
+static struct {
+	struct intel_iommu *iommu;
+	u16 irte_index;
+	u16 sub_handle;
+	u8  irte_mask;
+} irq_2_iommu[NR_IRQS];
+
+static DEFINE_SPINLOCK(irq_2_ir_lock);
+
+int irq_remapped(int irq)
+{
+	if (irq > NR_IRQS)
+		return 0;
+
+	if (!irq_2_iommu[irq].iommu)
+		return 0;
+
+	return 1;
+}
+
+int get_irte(int irq, struct irte *entry)
+{
+	int index;
+
+	if (!entry || irq > NR_IRQS)
+		return -1;
+
+	spin_lock(&irq_2_ir_lock);
+	if (!irq_2_iommu[irq].iommu) {
+		spin_unlock(&irq_2_ir_lock);
+		return -1;
+	}
+
+	index = irq_2_iommu[irq].irte_index + irq_2_iommu[irq].sub_handle;
+	*entry = *(irq_2_iommu[irq].iommu->ir_table->base + index);
+
+	spin_unlock(&irq_2_ir_lock);
+	return 0;
+}
+
+int alloc_irte(struct intel_iommu *iommu, int irq, u16 count)
+{
+	struct ir_table *table = iommu->ir_table;
+	u16 index, start_index;
+	unsigned int mask = 0;
+	int i;
+
+	if (!count)
+		return -1;
+
+	/*
+	 * start the IRTE search from index 0.
+	 */
+	index = start_index = 0;
+
+	if (count > 1) {
+		count = __roundup_pow_of_two(count);
+		mask = ilog2(count);
+	}
+
+	if (mask > ecap_max_handle_mask(iommu->ecap)) {
+		printk(KERN_ERR
+		       "Requested mask %x exceeds the max invalidation handle"
+		       " mask value %Lx\n", mask,
+		       ecap_max_handle_mask(iommu->ecap));
+		return -1;
+	}
+
+	spin_lock(&irq_2_ir_lock);
+	do {
+		for (i = index; i < index + count; i++)
+			if  (table->base[i].present)
+				break;
+		/* empty index found */
+		if (i == index + count)
+			break;
+
+		index = (index + count) % INTR_REMAP_TABLE_ENTRIES;
+
+		if (index == start_index) {
+			spin_unlock(&irq_2_ir_lock);
+			printk(KERN_ERR "can't allocate an IRTE\n");
+			return -1;
+		}
+	} while (1);
+
+	for (i = index; i < index + count; i++)
+		table->base[i].present = 1;
+
+	irq_2_iommu[irq].iommu = iommu;
+	irq_2_iommu[irq].irte_index =  index;
+	irq_2_iommu[irq].sub_handle = 0;
+	irq_2_iommu[irq].irte_mask = mask;
+
+	spin_unlock(&irq_2_ir_lock);
+
+	return index;
+}
+
+static void qi_flush_iec(struct intel_iommu *iommu, int index, int mask)
+{
+	struct qi_desc desc;
+
+	desc.low = QI_IEC_IIDEX(index) | QI_IEC_TYPE | QI_IEC_IM(mask)
+		   | QI_IEC_SELECTIVE;
+	desc.high = 0;
+
+	qi_submit_sync(&desc, iommu);
+}
+
+int map_irq_to_irte_handle(int irq, u16 *sub_handle)
+{
+	int index;
+
+	spin_lock(&irq_2_ir_lock);
+	if (irq >= NR_IRQS || !irq_2_iommu[irq].iommu) {
+		spin_unlock(&irq_2_ir_lock);
+		return -1;
+	}
+
+	*sub_handle = irq_2_iommu[irq].sub_handle;
+	index = irq_2_iommu[irq].irte_index;
+	spin_unlock(&irq_2_ir_lock);
+	return index;
+}
+
+int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index, u16 subhandle)
+{
+	spin_lock(&irq_2_ir_lock);
+	if (irq >= NR_IRQS || irq_2_iommu[irq].iommu) {
+		spin_unlock(&irq_2_ir_lock);
+		return -1;
+	}
+
+	irq_2_iommu[irq].iommu = iommu;
+	irq_2_iommu[irq].irte_index = index;
+	irq_2_iommu[irq].sub_handle = subhandle;
+	irq_2_iommu[irq].irte_mask = 0;
+
+	spin_unlock(&irq_2_ir_lock);
+
+	return 0;
+}
+
+int clear_irte_irq(int irq, struct intel_iommu *iommu, u16 index)
+{
+	spin_lock(&irq_2_ir_lock);
+	if (irq >= NR_IRQS || !irq_2_iommu[irq].iommu) {
+		spin_unlock(&irq_2_ir_lock);
+		return -1;
+	}
+
+	irq_2_iommu[irq].iommu = NULL;
+	irq_2_iommu[irq].irte_index = 0;
+	irq_2_iommu[irq].sub_handle = 0;
+	irq_2_iommu[irq].irte_mask = 0;
+
+	spin_unlock(&irq_2_ir_lock);
+
+	return 0;
+}
+
+int modify_irte(int irq, struct irte *irte_modified)
+{
+	int index;
+	struct irte *irte;
+	struct intel_iommu *iommu;
+
+	spin_lock(&irq_2_ir_lock);
+	if (irq >= NR_IRQS || !irq_2_iommu[irq].iommu) {
+		spin_unlock(&irq_2_ir_lock);
+		return -1;
+	}
+
+	iommu = irq_2_iommu[irq].iommu;
+
+	index = irq_2_iommu[irq].irte_index + irq_2_iommu[irq].sub_handle;
+	irte = &iommu->ir_table->base[index];
+
+	set_64bit((unsigned long *)irte, irte_modified->low | (1 << 1));
+	__iommu_flush_cache(iommu, irte, sizeof(*irte));
+
+	qi_flush_iec(iommu, index, 0);
+
+	spin_unlock(&irq_2_ir_lock);
+	return 0;
+}
+
+int flush_irte(int irq)
+{
+	int index;
+	struct intel_iommu *iommu;
+
+	spin_lock(&irq_2_ir_lock);
+	if (irq >= NR_IRQS || !irq_2_iommu[irq].iommu) {
+		spin_unlock(&irq_2_ir_lock);
+		return -1;
+	}
+
+	iommu = irq_2_iommu[irq].iommu;
+
+	index = irq_2_iommu[irq].irte_index + irq_2_iommu[irq].sub_handle;
+
+	qi_flush_iec(iommu, index, irq_2_iommu[irq].irte_mask);
+	spin_unlock(&irq_2_ir_lock);
+
+	return 0;
+}
+
+struct intel_iommu *map_ioapic_to_ir(int apic)
+{
+	int i;
+
+	for (i = 0; i < MAX_IO_APICS; i++)
+		if (ir_ioapic[i].id == apic)
+			return ir_ioapic[i].iommu;
+	return NULL;
+}
+
+struct intel_iommu *map_dev_to_ir(struct pci_dev *dev)
+{
+	struct dmar_drhd_unit *drhd;
+
+	drhd = dmar_find_matched_drhd_unit(dev);
+	if (!drhd)
+		return NULL;
+
+	return drhd->iommu;
+}
+
+int free_irte(int irq)
+{
+	int index, i;
+	struct irte *irte;
+	struct intel_iommu *iommu;
+
+	spin_lock(&irq_2_ir_lock);
+	if (irq >= NR_IRQS || !irq_2_iommu[irq].iommu) {
+		spin_unlock(&irq_2_ir_lock);
+		return -1;
+	}
+
+	iommu = irq_2_iommu[irq].iommu;
+
+	index = irq_2_iommu[irq].irte_index + irq_2_iommu[irq].sub_handle;
+	irte = &iommu->ir_table->base[index];
+
+	if (!irq_2_iommu[irq].sub_handle) {
+		for (i = 0; i < (1 << irq_2_iommu[irq].irte_mask); i++)
+			set_64bit((unsigned long *)irte, 0);
+		qi_flush_iec(iommu, index, irq_2_iommu[irq].irte_mask);
+	}
+
+	irq_2_iommu[irq].iommu = NULL;
+	irq_2_iommu[irq].irte_index = 0;
+	irq_2_iommu[irq].sub_handle = 0;
+	irq_2_iommu[irq].irte_mask = 0;
+
+	spin_unlock(&irq_2_ir_lock);
+
+	return 0;
+}
+
+static void iommu_set_intr_remapping(struct intel_iommu *iommu, int mode)
+{
+	u64 addr;
+	u32 cmd, sts;
+	unsigned long flags;
+
+	addr = virt_to_phys((void *)iommu->ir_table->base);
+
+	spin_lock_irqsave(&iommu->register_lock, flags);
+
+	dmar_writeq(iommu->reg + DMAR_IRTA_REG,
+		    (addr) | IR_X2APIC_MODE(mode) | INTR_REMAP_TABLE_REG_SIZE);
+
+	/* Set interrupt-remapping table pointer */
+	cmd = iommu->gcmd | DMA_GCMD_SIRTP;
+	writel(cmd, iommu->reg + DMAR_GCMD_REG);
+
+	IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG,
+		      readl, (sts & DMA_GSTS_IRTPS), sts);
+	spin_unlock_irqrestore(&iommu->register_lock, flags);
+
+	/*
+	 * global invalidation of interrupt entry cache before enabling
+	 * interrupt-remapping.
+	 */
+	qi_global_iec(iommu);
+
+	spin_lock_irqsave(&iommu->register_lock, flags);
+
+	/* Enable interrupt-remapping */
+	cmd = iommu->gcmd | DMA_GCMD_IRE;
+	iommu->gcmd |= DMA_GCMD_IRE;
+	writel(cmd, iommu->reg + DMAR_GCMD_REG);
+
+	IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG,
+		      readl, (sts & DMA_GSTS_IRES), sts);
+
+	spin_unlock_irqrestore(&iommu->register_lock, flags);
+}
+
+
+static int setup_intr_remapping(struct intel_iommu *iommu, int mode)
+{
+	struct ir_table *ir_table;
+	struct page *pages;
+
+	ir_table = iommu->ir_table = kzalloc(sizeof(struct ir_table),
+					     GFP_KERNEL);
+
+	if (!iommu->ir_table)
+		return -ENOMEM;
+
+	pages = alloc_pages(GFP_KERNEL | __GFP_ZERO, INTR_REMAP_PAGE_ORDER);
+
+	if (!pages) {
+		printk(KERN_ERR "failed to allocate pages of order %d\n",
+		       INTR_REMAP_PAGE_ORDER);
+		kfree(iommu->ir_table);
+		return -ENOMEM;
+	}
+
+	ir_table->base = page_address(pages);
+
+	iommu_set_intr_remapping(iommu, mode);
+	return 0;
+}
+
+int __init enable_intr_remapping(int eim)
+{
+	struct dmar_drhd_unit *drhd;
+	int setup = 0;
+
+	/*
+	 * check for the Interrupt-remapping support
+	 */
+	for_each_drhd_unit(drhd) {
+		struct intel_iommu *iommu = drhd->iommu;
+
+		if (!ecap_ir_support(iommu->ecap))
+			continue;
+
+		if (eim && !ecap_eim_support(iommu->ecap)) {
+			printk(KERN_INFO "DRHD %Lx: EIM not supported by DRHD, "
+			       " ecap %Lx\n", drhd->reg_base_addr, iommu->ecap);
+			return -1;
+		}
+	}
+
+	/*
+	 * Enable queued invalidation for all the DRHD's.
+	 */
+	for_each_drhd_unit(drhd) {
+		int ret;
+		struct intel_iommu *iommu = drhd->iommu;
+		ret = dmar_enable_qi(iommu);
+
+		if (ret) {
+			printk(KERN_ERR "DRHD %Lx: failed to enable queued, "
+			       " invalidation, ecap %Lx, ret %d\n",
+			       drhd->reg_base_addr, iommu->ecap, ret);
+			return -1;
+		}
+	}
+
+	/*
+	 * Setup Interrupt-remapping for all the DRHD's now.
+	 */
+	for_each_drhd_unit(drhd) {
+		struct intel_iommu *iommu = drhd->iommu;
+
+		if (!ecap_ir_support(iommu->ecap))
+			continue;
+
+		if (setup_intr_remapping(iommu, eim))
+			goto error;
+
+		setup = 1;
+	}
+
+	if (!setup)
+		goto error;
+
+	intr_remapping_enabled = 1;
+
+	return 0;
+
+error:
+	/*
+	 * handle error condition gracefully here!
+	 */
+	return -1;
+}
+
+static int ir_parse_ioapic_scope(struct acpi_dmar_header *header,
+				 struct intel_iommu *iommu)
+{
+	struct acpi_dmar_hardware_unit *drhd;
+	struct acpi_dmar_device_scope *scope;
+	void *start, *end;
+
+	drhd = (struct acpi_dmar_hardware_unit *)header;
+
+	start = (void *)(drhd + 1);
+	end = ((void *)drhd) + header->length;
+
+	while (start < end) {
+		scope = start;
+		if (scope->entry_type == ACPI_DMAR_SCOPE_TYPE_IOAPIC) {
+			if (ir_ioapic_num == MAX_IO_APICS) {
+				printk(KERN_WARNING "Exceeded Max IO APICS\n");
+				return -1;
+			}
+
+			printk(KERN_INFO "IOAPIC id %d under DRHD base"
+			       " 0x%Lx\n", scope->enumeration_id,
+			       drhd->address);
+
+			ir_ioapic[ir_ioapic_num].iommu = iommu;
+			ir_ioapic[ir_ioapic_num].id = scope->enumeration_id;
+			ir_ioapic_num++;
+		}
+		start += scope->length;
+	}
+
+	return 0;
+}
+
+/*
+ * Finds the assocaition between IOAPIC's and its Interrupt-remapping
+ * hardware unit.
+ */
+int __init parse_ioapics_under_ir(void)
+{
+	struct dmar_drhd_unit *drhd;
+	int ir_supported = 0;
+
+	for_each_drhd_unit(drhd) {
+		struct intel_iommu *iommu = drhd->iommu;
+
+		if (ecap_ir_support(iommu->ecap)) {
+			if (ir_parse_ioapic_scope(drhd->hdr, iommu))
+				return -1;
+
+			ir_supported = 1;
+		}
+	}
+
+	if (ir_supported && ir_ioapic_num != nr_ioapics) {
+		printk(KERN_WARNING
+		       "Not all IO-APIC's listed under remapping hardware\n");
+		return -1;
+	}
+
+	return ir_supported;
+}
diff --git a/drivers/pci/intr_remapping.h b/drivers/pci/intr_remapping.h
new file mode 100644
index 000000000000..05f2635bbe4e
--- /dev/null
+++ b/drivers/pci/intr_remapping.h
@@ -0,0 +1,8 @@
+#include "intel-iommu.h"
+
+struct ioapic_scope {
+	struct intel_iommu *iommu;
+	unsigned int id;
+};
+
+#define IR_X2APIC_MODE(mode) (mode ? (1 << 11) : 0)
diff --git a/drivers/power/Kconfig b/drivers/power/Kconfig
index 1982f8b42782..63bb57910445 100644
--- a/drivers/power/Kconfig
+++ b/drivers/power/Kconfig
@@ -58,7 +58,7 @@ config BATTERY_TOSA
 
 config BATTERY_WM97XX
 	bool "WM97xx generic battery driver"
-	depends on TOUCHSCREEN_WM97XX
+	depends on TOUCHSCREEN_WM97XX=y
 	help
 	  Say Y to enable support for battery measured by WM97xx aux port.
 
diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig
index 9a9755c92fad..b57fba5c6d02 100644
--- a/drivers/rtc/Kconfig
+++ b/drivers/rtc/Kconfig
@@ -329,7 +329,7 @@ comment "Platform RTC drivers"
 
 config RTC_DRV_CMOS
 	tristate "PC-style 'CMOS'"
-	depends on X86 || ALPHA || ARM || M32R || ATARI || PPC || MIPS
+	depends on X86 || ALPHA || ARM || M32R || ATARI || PPC || MIPS || SPARC64
 	default y if X86
 	help
 	  Say "yes" here to get direct support for the real time clock
@@ -406,14 +406,26 @@ config RTC_DRV_M48T86
 	  will be called rtc-m48t86.
 
 config RTC_DRV_M48T59
-	tristate "ST M48T59"
+	tristate "ST M48T59/M48T08/M48T02"
 	help
 	  If you say Y here you will get support for the
-	  ST M48T59 RTC chip.
+	  ST M48T59 RTC chip and compatible ST M48T08 and M48T02.
+
+	  These chips are usually found in Sun SPARC and UltraSPARC
+	  workstations.
 
 	  This driver can also be built as a module, if so, the module
 	  will be called "rtc-m48t59".
 
+config RTC_DRV_BQ4802
+	tristate "TI BQ4802"
+	help
+	  If you say Y here you will get support for the TI
+	  BQ4802 RTC chip.
+
+	  This driver can also be built as a module. If so, the module
+	  will be called rtc-bq4802.
+
 config RTC_DRV_V3020
 	tristate "EM Microelectronic V3020"
 	help
@@ -583,4 +595,18 @@ config RTC_DRV_PPC
 	 the RTC. This exposes that functionality through the generic RTC
 	 class.
 
+config RTC_DRV_SUN4V
+	bool "SUN4V Hypervisor RTC"
+	depends on SPARC64
+	help
+	  If you say Y here you will get support for the Hypervisor
+	  based RTC on SUN4V systems.
+
+config RTC_DRV_STARFIRE
+	bool "Starfire RTC"
+	depends on SPARC64
+	help
+	  If you say Y here you will get support for the RTC found on
+	  Starfire systems.
+
 endif # RTC_CLASS
diff --git a/drivers/rtc/Makefile b/drivers/rtc/Makefile
index 18622ef84cab..10f41f85c38a 100644
--- a/drivers/rtc/Makefile
+++ b/drivers/rtc/Makefile
@@ -38,6 +38,9 @@ obj-$(CONFIG_RTC_DRV_M41T80)	+= rtc-m41t80.o
 obj-$(CONFIG_RTC_DRV_M41T94)	+= rtc-m41t94.o
 obj-$(CONFIG_RTC_DRV_M48T59)	+= rtc-m48t59.o
 obj-$(CONFIG_RTC_DRV_M48T86)	+= rtc-m48t86.o
+obj-$(CONFIG_RTC_DRV_BQ4802)	+= rtc-bq4802.o
+obj-$(CONFIG_RTC_DRV_SUN4V)	+= rtc-sun4v.o
+obj-$(CONFIG_RTC_DRV_STARFIRE)	+= rtc-starfire.o
 obj-$(CONFIG_RTC_DRV_MAX6900)	+= rtc-max6900.o
 obj-$(CONFIG_RTC_DRV_MAX6902)	+= rtc-max6902.o
 obj-$(CONFIG_RTC_DRV_OMAP)	+= rtc-omap.o
diff --git a/drivers/rtc/rtc-bq4802.c b/drivers/rtc/rtc-bq4802.c
new file mode 100644
index 000000000000..189a018bdf34
--- /dev/null
+++ b/drivers/rtc/rtc-bq4802.c
@@ -0,0 +1,230 @@
+/* rtc-bq4802.c: TI BQ4802 RTC driver.
+ *
+ * Copyright (C) 2008 David S. Miller <davem@davemloft.net>
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/platform_device.h>
+#include <linux/rtc.h>
+#include <linux/bcd.h>
+
+MODULE_AUTHOR("David S. Miller <davem@davemloft.net>");
+MODULE_DESCRIPTION("TI BQ4802 RTC driver");
+MODULE_LICENSE("GPL");
+
+struct bq4802 {
+	void __iomem		*regs;
+	unsigned long		ioport;
+	struct rtc_device	*rtc;
+	spinlock_t		lock;
+	struct resource		*r;
+	u8 (*read)(struct bq4802 *, int);
+	void (*write)(struct bq4802 *, int, u8);
+};
+
+static u8 bq4802_read_io(struct bq4802 *p, int off)
+{
+	return inb(p->ioport + off);
+}
+
+static void bq4802_write_io(struct bq4802 *p, int off, u8 val)
+{
+	outb(val, p->ioport + off);
+}
+
+static u8 bq4802_read_mem(struct bq4802 *p, int off)
+{
+	return readb(p->regs + off);
+}
+
+static void bq4802_write_mem(struct bq4802 *p, int off, u8 val)
+{
+	writeb(val, p->regs + off);
+}
+
+static int bq4802_read_time(struct device *dev, struct rtc_time *tm)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct bq4802 *p = platform_get_drvdata(pdev);
+	unsigned long flags;
+	unsigned int century;
+	u8 val;
+
+	spin_lock_irqsave(&p->lock, flags);
+
+	val = p->read(p, 0x0e);
+	p->write(p, 0xe, val | 0x08);
+
+	tm->tm_sec = p->read(p, 0x00);
+	tm->tm_min = p->read(p, 0x02);
+	tm->tm_hour = p->read(p, 0x04);
+	tm->tm_mday = p->read(p, 0x06);
+	tm->tm_mon = p->read(p, 0x09);
+	tm->tm_year = p->read(p, 0x0a);
+	tm->tm_wday = p->read(p, 0x08);
+	century = p->read(p, 0x0f);
+
+	p->write(p, 0x0e, val);
+
+	spin_unlock_irqrestore(&p->lock, flags);
+
+	BCD_TO_BIN(tm->tm_sec);
+	BCD_TO_BIN(tm->tm_min);
+	BCD_TO_BIN(tm->tm_hour);
+	BCD_TO_BIN(tm->tm_mday);
+	BCD_TO_BIN(tm->tm_mon);
+	BCD_TO_BIN(tm->tm_year);
+	BCD_TO_BIN(tm->tm_wday);
+	BCD_TO_BIN(century);
+
+	tm->tm_year += (century * 100);
+	tm->tm_year -= 1900;
+
+	tm->tm_mon--;
+
+	return 0;
+}
+
+static int bq4802_set_time(struct device *dev, struct rtc_time *tm)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct bq4802 *p = platform_get_drvdata(pdev);
+	u8 sec, min, hrs, day, mon, yrs, century, val;
+	unsigned long flags;
+	unsigned int year;
+
+	year = tm->tm_year + 1900;
+	century = year / 100;
+	yrs = year % 100;
+
+	mon = tm->tm_mon + 1;   /* tm_mon starts at zero */
+	day = tm->tm_mday;
+	hrs = tm->tm_hour;
+	min = tm->tm_min;
+	sec = tm->tm_sec;
+
+	BIN_TO_BCD(sec);
+	BIN_TO_BCD(min);
+	BIN_TO_BCD(hrs);
+	BIN_TO_BCD(day);
+	BIN_TO_BCD(mon);
+	BIN_TO_BCD(yrs);
+	BIN_TO_BCD(century);
+
+	spin_lock_irqsave(&p->lock, flags);
+
+	val = p->read(p, 0x0e);
+	p->write(p, 0x0e, val | 0x08);
+
+	p->write(p, 0x00, sec);
+	p->write(p, 0x02, min);
+	p->write(p, 0x04, hrs);
+	p->write(p, 0x06, day);
+	p->write(p, 0x09, mon);
+	p->write(p, 0x0a, yrs);
+	p->write(p, 0x0f, century);
+
+	p->write(p, 0x0e, val);
+
+	spin_unlock_irqrestore(&p->lock, flags);
+
+	return 0;
+}
+
+static const struct rtc_class_ops bq4802_ops = {
+	.read_time	= bq4802_read_time,
+	.set_time	= bq4802_set_time,
+};
+
+static int __devinit bq4802_probe(struct platform_device *pdev)
+{
+	struct bq4802 *p = kzalloc(sizeof(*p), GFP_KERNEL);
+	int err = -ENOMEM;
+
+	if (!p)
+		goto out;
+
+	spin_lock_init(&p->lock);
+
+	p->r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!p->r) {
+		p->r = platform_get_resource(pdev, IORESOURCE_IO, 0);
+		err = -EINVAL;
+		if (!p->r)
+			goto out_free;
+	}
+	if (p->r->flags & IORESOURCE_IO) {
+		p->ioport = p->r->start;
+		p->read = bq4802_read_io;
+		p->write = bq4802_write_io;
+	} else if (p->r->flags & IORESOURCE_MEM) {
+		p->regs = ioremap(p->r->start, resource_size(p->r));
+		p->read = bq4802_read_mem;
+		p->write = bq4802_write_mem;
+	} else {
+		err = -EINVAL;
+		goto out_free;
+	}
+
+	p->rtc = rtc_device_register("bq4802", &pdev->dev,
+				     &bq4802_ops, THIS_MODULE);
+	if (IS_ERR(p->rtc)) {
+		err = PTR_ERR(p->rtc);
+		goto out_iounmap;
+	}
+
+	platform_set_drvdata(pdev, p);
+	err = 0;
+out:
+	return err;
+
+out_iounmap:
+	if (p->r->flags & IORESOURCE_MEM)
+		iounmap(p->regs);
+out_free:
+	kfree(p);
+	goto out;
+}
+
+static int __devexit bq4802_remove(struct platform_device *pdev)
+{
+	struct bq4802 *p = platform_get_drvdata(pdev);
+
+	rtc_device_unregister(p->rtc);
+	if (p->r->flags & IORESOURCE_MEM)
+		iounmap(p->regs);
+
+	platform_set_drvdata(pdev, NULL);
+
+	kfree(p);
+
+	return 0;
+}
+
+/* work with hotplug and coldplug */
+MODULE_ALIAS("platform:rtc-bq4802");
+
+static struct platform_driver bq4802_driver = {
+	.driver		= {
+		.name	= "rtc-bq4802",
+		.owner	= THIS_MODULE,
+	},
+	.probe		= bq4802_probe,
+	.remove		= __devexit_p(bq4802_remove),
+};
+
+static int __init bq4802_init(void)
+{
+	return platform_driver_register(&bq4802_driver);
+}
+
+static void __exit bq4802_exit(void)
+{
+	platform_driver_unregister(&bq4802_driver);
+}
+
+module_init(bq4802_init);
+module_exit(bq4802_exit);
diff --git a/drivers/rtc/rtc-cmos.c b/drivers/rtc/rtc-cmos.c
index b184367637d0..b23af0c2a869 100644
--- a/drivers/rtc/rtc-cmos.c
+++ b/drivers/rtc/rtc-cmos.c
@@ -636,7 +636,7 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq)
 	 */
 #if	defined(CONFIG_ATARI)
 	address_space = 64;
-#elif defined(__i386__) || defined(__x86_64__) || defined(__arm__)
+#elif defined(__i386__) || defined(__x86_64__) || defined(__arm__) || defined(__sparc__)
 	address_space = 128;
 #else
 #warning Assuming 128 bytes of RTC+NVRAM address space, not 64 bytes.
@@ -699,7 +699,8 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq)
 	/* FIXME teach the alarm code how to handle binary mode;
 	 * <asm-generic/rtc.h> doesn't know 12-hour mode either.
 	 */
-	if (!(rtc_control & RTC_24H) || (rtc_control & (RTC_DM_BINARY))) {
+	if (is_valid_irq(rtc_irq) &&
+	    (!(rtc_control & RTC_24H) || (rtc_control & (RTC_DM_BINARY)))) {
 		dev_dbg(dev, "only 24-hr BCD mode supported\n");
 		retval = -ENXIO;
 		goto cleanup1;
diff --git a/drivers/rtc/rtc-m48t59.c b/drivers/rtc/rtc-m48t59.c
index 013e6c103b9c..ce4eff6a8d51 100644
--- a/drivers/rtc/rtc-m48t59.c
+++ b/drivers/rtc/rtc-m48t59.c
@@ -24,8 +24,9 @@
 #define NO_IRQ	(-1)
 #endif
 
-#define M48T59_READ(reg)	pdata->read_byte(dev, reg)
-#define M48T59_WRITE(val, reg)	pdata->write_byte(dev, reg, val)
+#define M48T59_READ(reg) (pdata->read_byte(dev, pdata->offset + reg))
+#define M48T59_WRITE(val, reg) \
+	(pdata->write_byte(dev, pdata->offset + reg, val))
 
 #define M48T59_SET_BITS(mask, reg)	\
 	M48T59_WRITE((M48T59_READ(reg) | (mask)), (reg))
@@ -34,7 +35,6 @@
 
 struct m48t59_private {
 	void __iomem *ioaddr;
-	unsigned int size; /* iomem size */
 	int irq;
 	struct rtc_device *rtc;
 	spinlock_t lock; /* serialize the NVRAM and RTC access */
@@ -82,7 +82,8 @@ static int m48t59_rtc_read_time(struct device *dev, struct rtc_time *tm)
 	tm->tm_mday	= BCD2BIN(M48T59_READ(M48T59_MDAY));
 
 	val = M48T59_READ(M48T59_WDAY);
-	if ((val & M48T59_WDAY_CEB) && (val & M48T59_WDAY_CB)) {
+	if ((pdata->type == M48T59RTC_TYPE_M48T59) &&
+	    (val & M48T59_WDAY_CEB) && (val & M48T59_WDAY_CB)) {
 		dev_dbg(dev, "Century bit is enabled\n");
 		tm->tm_year += 100;	/* one century */
 	}
@@ -126,7 +127,7 @@ static int m48t59_rtc_set_time(struct device *dev, struct rtc_time *tm)
 	M48T59_WRITE((BIN2BCD(tm->tm_mon + 1) & 0x1F), M48T59_MONTH);
 	M48T59_WRITE(BIN2BCD(tm->tm_year % 100), M48T59_YEAR);
 
-	if (tm->tm_year/100)
+	if (pdata->type == M48T59RTC_TYPE_M48T59 && (tm->tm_year / 100))
 		val = (M48T59_WDAY_CEB | M48T59_WDAY_CB);
 	val |= (BIN2BCD(tm->tm_wday) & 0x07);
 	M48T59_WRITE(val, M48T59_WDAY);
@@ -310,6 +311,11 @@ static const struct rtc_class_ops m48t59_rtc_ops = {
 	.proc		= m48t59_rtc_proc,
 };
 
+static const struct rtc_class_ops m48t02_rtc_ops = {
+	.read_time	= m48t59_rtc_read_time,
+	.set_time	= m48t59_rtc_set_time,
+};
+
 static ssize_t m48t59_nvram_read(struct kobject *kobj,
 				struct bin_attribute *bin_attr,
 				char *buf, loff_t pos, size_t size)
@@ -321,7 +327,7 @@ static ssize_t m48t59_nvram_read(struct kobject *kobj,
 	ssize_t cnt = 0;
 	unsigned long flags;
 
-	for (; size > 0 && pos < M48T59_NVRAM_SIZE; cnt++, size--) {
+	for (; size > 0 && pos < pdata->offset; cnt++, size--) {
 		spin_lock_irqsave(&m48t59->lock, flags);
 		*buf++ = M48T59_READ(cnt);
 		spin_unlock_irqrestore(&m48t59->lock, flags);
@@ -341,7 +347,7 @@ static ssize_t m48t59_nvram_write(struct kobject *kobj,
 	ssize_t cnt = 0;
 	unsigned long flags;
 
-	for (; size > 0 && pos < M48T59_NVRAM_SIZE; cnt++, size--) {
+	for (; size > 0 && pos < pdata->offset; cnt++, size--) {
 		spin_lock_irqsave(&m48t59->lock, flags);
 		M48T59_WRITE(*buf++, cnt);
 		spin_unlock_irqrestore(&m48t59->lock, flags);
@@ -358,7 +364,6 @@ static struct bin_attribute m48t59_nvram_attr = {
 	},
 	.read = m48t59_nvram_read,
 	.write = m48t59_nvram_write,
-	.size = M48T59_NVRAM_SIZE,
 };
 
 static int __devinit m48t59_rtc_probe(struct platform_device *pdev)
@@ -367,6 +372,8 @@ static int __devinit m48t59_rtc_probe(struct platform_device *pdev)
 	struct m48t59_private *m48t59 = NULL;
 	struct resource *res;
 	int ret = -ENOMEM;
+	char *name;
+	const struct rtc_class_ops *ops;
 
 	/* This chip could be memory-mapped or I/O-mapped */
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
@@ -391,6 +398,8 @@ static int __devinit m48t59_rtc_probe(struct platform_device *pdev)
 			/* Ensure we only kmalloc platform data once */
 			pdev->dev.platform_data = pdata;
 		}
+		if (!pdata->type)
+			pdata->type = M48T59RTC_TYPE_M48T59;
 
 		/* Try to use the generic memory read/write ops */
 		if (!pdata->write_byte)
@@ -403,10 +412,14 @@ static int __devinit m48t59_rtc_probe(struct platform_device *pdev)
 	if (!m48t59)
 		return -ENOMEM;
 
-	m48t59->size = res->end - res->start + 1;
-	m48t59->ioaddr = ioremap(res->start, m48t59->size);
-	if (!m48t59->ioaddr)
-		goto out;
+	m48t59->ioaddr = pdata->ioaddr;
+
+	if (!m48t59->ioaddr) {
+		/* ioaddr not mapped externally */
+		m48t59->ioaddr = ioremap(res->start, res->end - res->start + 1);
+		if (!m48t59->ioaddr)
+			goto out;
+	}
 
 	/* Try to get irq number. We also can work in
 	 * the mode without IRQ.
@@ -421,14 +434,36 @@ static int __devinit m48t59_rtc_probe(struct platform_device *pdev)
 		if (ret)
 			goto out;
 	}
+	switch (pdata->type) {
+	case M48T59RTC_TYPE_M48T59:
+		name = "m48t59";
+		ops = &m48t59_rtc_ops;
+		pdata->offset = 0x1ff0;
+		break;
+	case M48T59RTC_TYPE_M48T02:
+		name = "m48t02";
+		ops = &m48t02_rtc_ops;
+		pdata->offset = 0x7f0;
+		break;
+	case M48T59RTC_TYPE_M48T08:
+		name = "m48t08";
+		ops = &m48t02_rtc_ops;
+		pdata->offset = 0x1ff0;
+		break;
+	default:
+		dev_err(&pdev->dev, "Unknown RTC type\n");
+		ret = -ENODEV;
+		goto out;
+	}
 
-	m48t59->rtc = rtc_device_register("m48t59", &pdev->dev,
-				&m48t59_rtc_ops, THIS_MODULE);
+	m48t59->rtc = rtc_device_register(name, &pdev->dev, ops, THIS_MODULE);
 	if (IS_ERR(m48t59->rtc)) {
 		ret = PTR_ERR(m48t59->rtc);
 		goto out;
 	}
 
+	m48t59_nvram_attr.size = pdata->offset;
+
 	ret = sysfs_create_bin_file(&pdev->dev.kobj, &m48t59_nvram_attr);
 	if (ret)
 		goto out;
@@ -452,11 +487,12 @@ out:
 static int __devexit m48t59_rtc_remove(struct platform_device *pdev)
 {
 	struct m48t59_private *m48t59 = platform_get_drvdata(pdev);
+	struct m48t59_plat_data *pdata = pdev->dev.platform_data;
 
 	sysfs_remove_bin_file(&pdev->dev.kobj, &m48t59_nvram_attr);
 	if (!IS_ERR(m48t59->rtc))
 		rtc_device_unregister(m48t59->rtc);
-	if (m48t59->ioaddr)
+	if (m48t59->ioaddr && !pdata->ioaddr)
 		iounmap(m48t59->ioaddr);
 	if (m48t59->irq != NO_IRQ)
 		free_irq(m48t59->irq, &pdev->dev);
@@ -491,5 +527,5 @@ module_init(m48t59_rtc_init);
 module_exit(m48t59_rtc_exit);
 
 MODULE_AUTHOR("Mark Zhan <rongkai.zhan@windriver.com>");
-MODULE_DESCRIPTION("M48T59 RTC driver");
+MODULE_DESCRIPTION("M48T59/M48T02/M48T08 RTC driver");
 MODULE_LICENSE("GPL");
diff --git a/drivers/rtc/rtc-starfire.c b/drivers/rtc/rtc-starfire.c
new file mode 100644
index 000000000000..7ccb0dd700af
--- /dev/null
+++ b/drivers/rtc/rtc-starfire.c
@@ -0,0 +1,120 @@
+/* rtc-starfire.c: Starfire platform RTC driver.
+ *
+ * Copyright (C) 2008 David S. Miller <davem@davemloft.net>
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/time.h>
+#include <linux/rtc.h>
+#include <linux/platform_device.h>
+
+#include <asm/oplib.h>
+
+MODULE_AUTHOR("David S. Miller <davem@davemloft.net>");
+MODULE_DESCRIPTION("Starfire RTC driver");
+MODULE_LICENSE("GPL");
+
+struct starfire_rtc {
+	struct rtc_device	*rtc;
+	spinlock_t		lock;
+};
+
+static u32 starfire_get_time(void)
+{
+	static char obp_gettod[32];
+	static u32 unix_tod;
+
+	sprintf(obp_gettod, "h# %08x unix-gettod",
+		(unsigned int) (long) &unix_tod);
+	prom_feval(obp_gettod);
+
+	return unix_tod;
+}
+
+static int starfire_read_time(struct device *dev, struct rtc_time *tm)
+{
+	struct starfire_rtc *p = dev_get_drvdata(dev);
+	unsigned long flags, secs;
+
+	spin_lock_irqsave(&p->lock, flags);
+	secs = starfire_get_time();
+	spin_unlock_irqrestore(&p->lock, flags);
+
+	rtc_time_to_tm(secs, tm);
+
+	return 0;
+}
+
+static int starfire_set_time(struct device *dev, struct rtc_time *tm)
+{
+	unsigned long secs;
+	int err;
+
+	err = rtc_tm_to_time(tm, &secs);
+	if (err)
+		return err;
+
+	/* Do nothing, time is set using the service processor
+	 * console on this platform.
+	 */
+	return 0;
+}
+
+static const struct rtc_class_ops starfire_rtc_ops = {
+	.read_time	= starfire_read_time,
+	.set_time	= starfire_set_time,
+};
+
+static int __devinit starfire_rtc_probe(struct platform_device *pdev)
+{
+	struct starfire_rtc *p = kzalloc(sizeof(*p), GFP_KERNEL);
+
+	if (!p)
+		return -ENOMEM;
+
+	spin_lock_init(&p->lock);
+
+	p->rtc = rtc_device_register("starfire", &pdev->dev,
+				     &starfire_rtc_ops, THIS_MODULE);
+	if (IS_ERR(p->rtc)) {
+		int err = PTR_ERR(p->rtc);
+		kfree(p);
+		return err;
+	}
+	platform_set_drvdata(pdev, p);
+	return 0;
+}
+
+static int __devexit starfire_rtc_remove(struct platform_device *pdev)
+{
+	struct starfire_rtc *p = platform_get_drvdata(pdev);
+
+	rtc_device_unregister(p->rtc);
+	kfree(p);
+
+	return 0;
+}
+
+static struct platform_driver starfire_rtc_driver = {
+	.driver		= {
+		.name	= "rtc-starfire",
+		.owner	= THIS_MODULE,
+	},
+	.probe		= starfire_rtc_probe,
+	.remove		= __devexit_p(starfire_rtc_remove),
+};
+
+static int __init starfire_rtc_init(void)
+{
+	return platform_driver_register(&starfire_rtc_driver);
+}
+
+static void __exit starfire_rtc_exit(void)
+{
+	platform_driver_unregister(&starfire_rtc_driver);
+}
+
+module_init(starfire_rtc_init);
+module_exit(starfire_rtc_exit);
diff --git a/drivers/rtc/rtc-sun4v.c b/drivers/rtc/rtc-sun4v.c
new file mode 100644
index 000000000000..2012ccbb4a53
--- /dev/null
+++ b/drivers/rtc/rtc-sun4v.c
@@ -0,0 +1,153 @@
+/* rtc-sun4c.c: Hypervisor based RTC for SUN4V systems.
+ *
+ * Copyright (C) 2008 David S. Miller <davem@davemloft.net>
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/time.h>
+#include <linux/rtc.h>
+#include <linux/platform_device.h>
+
+#include <asm/hypervisor.h>
+
+MODULE_AUTHOR("David S. Miller <davem@davemloft.net>");
+MODULE_DESCRIPTION("SUN4V RTC driver");
+MODULE_LICENSE("GPL");
+
+struct sun4v_rtc {
+	struct rtc_device	*rtc;
+	spinlock_t		lock;
+};
+
+static unsigned long hypervisor_get_time(void)
+{
+	unsigned long ret, time;
+	int retries = 10000;
+
+retry:
+	ret = sun4v_tod_get(&time);
+	if (ret == HV_EOK)
+		return time;
+	if (ret == HV_EWOULDBLOCK) {
+		if (--retries > 0) {
+			udelay(100);
+			goto retry;
+		}
+		printk(KERN_WARNING "SUN4V: tod_get() timed out.\n");
+		return 0;
+	}
+	printk(KERN_WARNING "SUN4V: tod_get() not supported.\n");
+	return 0;
+}
+
+static int sun4v_read_time(struct device *dev, struct rtc_time *tm)
+{
+	struct sun4v_rtc *p = dev_get_drvdata(dev);
+	unsigned long flags, secs;
+
+	spin_lock_irqsave(&p->lock, flags);
+	secs = hypervisor_get_time();
+	spin_unlock_irqrestore(&p->lock, flags);
+
+	rtc_time_to_tm(secs, tm);
+
+	return 0;
+}
+
+static int hypervisor_set_time(unsigned long secs)
+{
+	unsigned long ret;
+	int retries = 10000;
+
+retry:
+	ret = sun4v_tod_set(secs);
+	if (ret == HV_EOK)
+		return 0;
+	if (ret == HV_EWOULDBLOCK) {
+		if (--retries > 0) {
+			udelay(100);
+			goto retry;
+		}
+		printk(KERN_WARNING "SUN4V: tod_set() timed out.\n");
+		return -EAGAIN;
+	}
+	printk(KERN_WARNING "SUN4V: tod_set() not supported.\n");
+	return -EOPNOTSUPP;
+}
+
+static int sun4v_set_time(struct device *dev, struct rtc_time *tm)
+{
+	struct sun4v_rtc *p = dev_get_drvdata(dev);
+	unsigned long flags, secs;
+	int err;
+
+	err = rtc_tm_to_time(tm, &secs);
+	if (err)
+		return err;
+
+	spin_lock_irqsave(&p->lock, flags);
+	err = hypervisor_set_time(secs);
+	spin_unlock_irqrestore(&p->lock, flags);
+
+	return err;
+}
+
+static const struct rtc_class_ops sun4v_rtc_ops = {
+	.read_time	= sun4v_read_time,
+	.set_time	= sun4v_set_time,
+};
+
+static int __devinit sun4v_rtc_probe(struct platform_device *pdev)
+{
+	struct sun4v_rtc *p = kzalloc(sizeof(*p), GFP_KERNEL);
+
+	if (!p)
+		return -ENOMEM;
+
+	spin_lock_init(&p->lock);
+
+	p->rtc = rtc_device_register("sun4v", &pdev->dev,
+				     &sun4v_rtc_ops, THIS_MODULE);
+	if (IS_ERR(p->rtc)) {
+		int err = PTR_ERR(p->rtc);
+		kfree(p);
+		return err;
+	}
+	platform_set_drvdata(pdev, p);
+	return 0;
+}
+
+static int __devexit sun4v_rtc_remove(struct platform_device *pdev)
+{
+	struct sun4v_rtc *p = platform_get_drvdata(pdev);
+
+	rtc_device_unregister(p->rtc);
+	kfree(p);
+
+	return 0;
+}
+
+static struct platform_driver sun4v_rtc_driver = {
+	.driver		= {
+		.name	= "rtc-sun4v",
+		.owner	= THIS_MODULE,
+	},
+	.probe		= sun4v_rtc_probe,
+	.remove		= __devexit_p(sun4v_rtc_remove),
+};
+
+static int __init sun4v_rtc_init(void)
+{
+	return platform_driver_register(&sun4v_rtc_driver);
+}
+
+static void __exit sun4v_rtc_exit(void)
+{
+	platform_driver_unregister(&sun4v_rtc_driver);
+}
+
+module_init(sun4v_rtc_init);
+module_exit(sun4v_rtc_exit);
diff --git a/drivers/s390/char/fs3270.c b/drivers/s390/char/fs3270.c
index d18e6d2e0b49..40759c33477d 100644
--- a/drivers/s390/char/fs3270.c
+++ b/drivers/s390/char/fs3270.c
@@ -418,25 +418,22 @@ fs3270_open(struct inode *inode, struct file *filp)
 {
 	struct fs3270 *fp;
 	struct idal_buffer *ib;
-	int minor, rc;
+	int minor, rc = 0;
 
 	if (imajor(filp->f_path.dentry->d_inode) != IBM_FS3270_MAJOR)
 		return -ENODEV;
-	lock_kernel();
 	minor = iminor(filp->f_path.dentry->d_inode);
 	/* Check for minor 0 multiplexer. */
 	if (minor == 0) {
-		struct tty_struct *tty;
-		mutex_lock(&tty_mutex);
-		tty = get_current_tty();
+		struct tty_struct *tty = get_current_tty();
 		if (!tty || tty->driver->major != IBM_TTY3270_MAJOR) {
-			mutex_unlock(&tty_mutex);
-			rc = -ENODEV;
-			goto out;
+			tty_kref_put(tty);
+			return -ENODEV;
 		}
 		minor = tty->index + RAW3270_FIRSTMINOR;
-		mutex_unlock(&tty_mutex);
+		tty_kref_put(tty);
 	}
+	lock_kernel();
 	/* Check if some other program is already using fullscreen mode. */
 	fp = (struct fs3270 *) raw3270_find_view(&fs3270_fn, minor);
 	if (!IS_ERR(fp)) {
@@ -478,7 +475,7 @@ fs3270_open(struct inode *inode, struct file *filp)
 	filp->private_data = fp;
 out:
 	unlock_kernel();
-	return 0;
+	return rc;
 }
 
 /*
diff --git a/drivers/sbus/Makefile b/drivers/sbus/Makefile
index 7b1d24d95308..e94dc25805f9 100644
--- a/drivers/sbus/Makefile
+++ b/drivers/sbus/Makefile
@@ -2,8 +2,4 @@
 # Makefile for the linux kernel.
 #
 
-ifneq ($(ARCH),m68k)
-obj-y    := sbus.o dvma.o
-endif
-
 obj-$(CONFIG_SBUSCHAR) += char/
diff --git a/drivers/sbus/char/Kconfig b/drivers/sbus/char/Kconfig
index 400c65bfb8c7..73cde85d04d8 100644
--- a/drivers/sbus/char/Kconfig
+++ b/drivers/sbus/char/Kconfig
@@ -13,16 +13,6 @@ config SUN_OPENPROMIO
 
 	  If unsure, say Y.
 
-config SUN_MOSTEK_RTC
-	tristate "Mostek real time clock support"
-	depends on SPARC32
-	help
-	  The Mostek RTC chip is used on all known Sun computers except
-	  some JavaStations. For a JavaStation you need to say Y both here
-	  and to "Enhanced Real Time Clock Support".
-
-	  Say Y here unless you are building a special purpose kernel.
-
 config OBP_FLASH
 	tristate "OBP Flash Device support"
 	depends on SPARC64
@@ -30,26 +20,9 @@ config OBP_FLASH
 	  The OpenBoot PROM on Ultra systems is flashable. If you want to be
 	  able to upgrade the OBP firmware, say Y here.
 
-config SUN_BPP
-	tristate "Bidirectional parallel port support (OBSOLETE)"
-	depends on EXPERIMENTAL
-	help
-	  Say Y here to support Sun's obsolete variant of IEEE1284
-	  bidirectional parallel port protocol as /dev/bppX.  Can be built on
-	  x86 machines.
-
-config SUN_VIDEOPIX
-	tristate "Videopix Frame Grabber (EXPERIMENTAL)"
-	depends on EXPERIMENTAL && (BROKEN || !64BIT)
-	help
-	  Say Y here to support the Videopix Frame Grabber from Sun
-	  Microsystems, commonly found on SPARCstations.  This card, which is
-	  based on the Phillips SAA9051, can handle NTSC and PAL/SECAM and
-	  SVIDEO signals.
-
 config TADPOLE_TS102_UCTRL
 	tristate "Tadpole TS102 Microcontroller support (EXPERIMENTAL)"
-	depends on EXPERIMENTAL && SPARC32
+	depends on EXPERIMENTAL
 	help
 	  Say Y here to directly support the TS102 Microcontroller interface
 	  on the Tadpole Sparcbook 3.  This device handles power-management
diff --git a/drivers/sbus/char/Makefile b/drivers/sbus/char/Makefile
index 7ab060e9a5fe..78b6183c9866 100644
--- a/drivers/sbus/char/Makefile
+++ b/drivers/sbus/char/Makefile
@@ -7,18 +7,12 @@
 # Rewritten to use lists instead of if-statements.
 #
 
-vfc-objs := vfc_dev.o vfc_i2c.o
 bbc-objs := bbc_i2c.o bbc_envctrl.o
 
 obj-$(CONFIG_ENVCTRL)			+= envctrl.o
 obj-$(CONFIG_DISPLAY7SEG)		+= display7seg.o
-obj-$(CONFIG_WATCHDOG_CP1XXX)		+= cpwatchdog.o
-obj-$(CONFIG_WATCHDOG_RIO)		+= riowatchdog.o
 obj-$(CONFIG_OBP_FLASH)			+= flash.o
 obj-$(CONFIG_SUN_OPENPROMIO)		+= openprom.o
-obj-$(CONFIG_SUN_MOSTEK_RTC)		+= rtc.o
-obj-$(CONFIG_SUN_BPP)			+= bpp.o
-obj-$(CONFIG_SUN_VIDEOPIX)		+= vfc.o
 obj-$(CONFIG_TADPOLE_TS102_UCTRL)	+= uctrl.o
 obj-$(CONFIG_SUN_JSFLASH)		+= jsflash.o
 obj-$(CONFIG_BBC_I2C)			+= bbc.o
diff --git a/drivers/sbus/char/bbc_envctrl.c b/drivers/sbus/char/bbc_envctrl.c
index 0bde26989a23..15dab96d05e3 100644
--- a/drivers/sbus/char/bbc_envctrl.c
+++ b/drivers/sbus/char/bbc_envctrl.c
@@ -1,15 +1,15 @@
-/* $Id: bbc_envctrl.c,v 1.4 2001/04/06 16:48:08 davem Exp $
- * bbc_envctrl.c: UltraSPARC-III environment control driver.
+/* bbc_envctrl.c: UltraSPARC-III environment control driver.
  *
- * Copyright (C) 2001 David S. Miller (davem@redhat.com)
+ * Copyright (C) 2001, 2008 David S. Miller (davem@davemloft.net)
  */
 
 #include <linux/kthread.h>
 #include <linux/delay.h>
 #include <linux/kmod.h>
 #include <linux/reboot.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 #include <asm/oplib.h>
-#include <asm/ebus.h>
 
 #include "bbc_i2c.h"
 #include "max1617.h"
@@ -75,43 +75,8 @@ static struct temp_limits amb_temp_limits[2] = {
 	{ 65, 55, 40, 5, -5, -10 },
 };
 
-enum fan_action { FAN_SLOWER, FAN_SAME, FAN_FASTER, FAN_FULLBLAST, FAN_STATE_MAX };
-
-struct bbc_cpu_temperature {
-	struct bbc_cpu_temperature	*next;
-
-	struct bbc_i2c_client		*client;
-	int				index;
-
-	/* Current readings, and history. */
-	s8				curr_cpu_temp;
-	s8				curr_amb_temp;
-	s8				prev_cpu_temp;
-	s8				prev_amb_temp;
-	s8				avg_cpu_temp;
-	s8				avg_amb_temp;
-
-	int				sample_tick;
-
-	enum fan_action			fan_todo[2];
-#define FAN_AMBIENT	0
-#define FAN_CPU		1
-};
-
-struct bbc_cpu_temperature *all_bbc_temps;
-
-struct bbc_fan_control {
-	struct bbc_fan_control 	*next;
-
-	struct bbc_i2c_client 	*client;
-	int 			index;
-
-	int			psupply_fan_on;
-	int			cpu_fan_speed;
-	int			system_fan_speed;
-};
-
-struct bbc_fan_control *all_bbc_fans;
+static LIST_HEAD(all_temps);
+static LIST_HEAD(all_fans);
 
 #define CPU_FAN_REG	0xf0
 #define SYS_FAN_REG	0xf2
@@ -330,7 +295,7 @@ static enum fan_action prioritize_fan_action(int which_fan)
 	 * recommend we do, and perform that action on all the
 	 * fans.
 	 */
-	for (tp = all_bbc_temps; tp; tp = tp->next) {
+	list_for_each_entry(tp, &all_temps, glob_list) {
 		if (tp->fan_todo[which_fan] == FAN_FULLBLAST) {
 			decision = FAN_FULLBLAST;
 			break;
@@ -439,7 +404,7 @@ static void fans_full_blast(void)
 	/* Since we will not be monitoring things anymore, put
 	 * the fans on full blast.
 	 */
-	for (fp = all_bbc_fans; fp; fp = fp->next) {
+	list_for_each_entry(fp, &all_fans, glob_list) {
 		fp->cpu_fan_speed = FAN_SPEED_MAX;
 		fp->system_fan_speed = FAN_SPEED_MAX;
 		fp->psupply_fan_on = 1;
@@ -463,11 +428,11 @@ static int kenvctrld(void *__unused)
 		if (kthread_should_stop())
 			break;
 
-		for (tp = all_bbc_temps; tp; tp = tp->next) {
+		list_for_each_entry(tp, &all_temps, glob_list) {
 			get_current_temps(tp);
 			analyze_temps(tp, &last_warning_jiffies);
 		}
-		for (fp = all_bbc_fans; fp; fp = fp->next)
+		list_for_each_entry(fp, &all_fans, glob_list)
 			maybe_new_fan_speeds(fp);
 	}
 	printk(KERN_INFO "bbc_envctrl: kenvctrld exiting...\n");
@@ -477,7 +442,8 @@ static int kenvctrld(void *__unused)
 	return 0;
 }
 
-static void attach_one_temp(struct linux_ebus_child *echild, int temp_idx)
+static void attach_one_temp(struct bbc_i2c_bus *bp, struct of_device *op,
+			    int temp_idx)
 {
 	struct bbc_cpu_temperature *tp;
 
@@ -485,20 +451,17 @@ static void attach_one_temp(struct linux_ebus_child *echild, int temp_idx)
 	if (!tp)
 		return;
 
-	tp->client = bbc_i2c_attach(echild);
+	tp->client = bbc_i2c_attach(bp, op);
 	if (!tp->client) {
 		kfree(tp);
 		return;
 	}
 
+
 	tp->index = temp_idx;
-	{
-		struct bbc_cpu_temperature **tpp = &all_bbc_temps;
-		while (*tpp)
-			tpp = &((*tpp)->next);
-		tp->next = NULL;
-		*tpp = tp;
-	}
+
+	list_add(&tp->glob_list, &all_temps);
+	list_add(&tp->bp_list, &bp->temps);
 
 	/* Tell it to convert once every 5 seconds, clear all cfg
 	 * bits.
@@ -524,7 +487,8 @@ static void attach_one_temp(struct linux_ebus_child *echild, int temp_idx)
 	tp->fan_todo[FAN_CPU] = FAN_SAME;
 }
 
-static void attach_one_fan(struct linux_ebus_child *echild, int fan_idx)
+static void attach_one_fan(struct bbc_i2c_bus *bp, struct of_device *op,
+			   int fan_idx)
 {
 	struct bbc_fan_control *fp;
 
@@ -532,7 +496,7 @@ static void attach_one_fan(struct linux_ebus_child *echild, int fan_idx)
 	if (!fp)
 		return;
 
-	fp->client = bbc_i2c_attach(echild);
+	fp->client = bbc_i2c_attach(bp, op);
 	if (!fp->client) {
 		kfree(fp);
 		return;
@@ -540,13 +504,8 @@ static void attach_one_fan(struct linux_ebus_child *echild, int fan_idx)
 
 	fp->index = fan_idx;
 
-	{
-		struct bbc_fan_control **fpp = &all_bbc_fans;
-		while (*fpp)
-			fpp = &((*fpp)->next);
-		fp->next = NULL;
-		*fpp = fp;
-	}
+	list_add(&fp->glob_list, &all_fans);
+	list_add(&fp->bp_list, &bp->fans);
 
 	/* The i2c device controlling the fans is write-only.
 	 * So the only way to keep track of the current power
@@ -563,18 +522,18 @@ static void attach_one_fan(struct linux_ebus_child *echild, int fan_idx)
 	set_fan_speeds(fp);
 }
 
-int bbc_envctrl_init(void)
+int bbc_envctrl_init(struct bbc_i2c_bus *bp)
 {
-	struct linux_ebus_child *echild;
+	struct of_device *op;
 	int temp_index = 0;
 	int fan_index = 0;
 	int devidx = 0;
 
-	while ((echild = bbc_i2c_getdev(devidx++)) != NULL) {
-		if (!strcmp(echild->prom_node->name, "temperature"))
-			attach_one_temp(echild, temp_index++);
-		if (!strcmp(echild->prom_node->name, "fan-control"))
-			attach_one_fan(echild, fan_index++);
+	while ((op = bbc_i2c_getdev(bp, devidx++)) != NULL) {
+		if (!strcmp(op->node->name, "temperature"))
+			attach_one_temp(bp, op, temp_index++);
+		if (!strcmp(op->node->name, "fan-control"))
+			attach_one_fan(bp, op, fan_index++);
 	}
 	if (temp_index != 0 && fan_index != 0) {
 		kenvctrld_task = kthread_run(kenvctrld, NULL, "kenvctrld");
@@ -597,26 +556,22 @@ static void destroy_one_fan(struct bbc_fan_control *fp)
 	kfree(fp);
 }
 
-void bbc_envctrl_cleanup(void)
+void bbc_envctrl_cleanup(struct bbc_i2c_bus *bp)
 {
-	struct bbc_cpu_temperature *tp;
-	struct bbc_fan_control *fp;
+	struct bbc_cpu_temperature *tp, *tpos;
+	struct bbc_fan_control *fp, *fpos;
 
 	kthread_stop(kenvctrld_task);
 
-	tp = all_bbc_temps;
-	while (tp != NULL) {
-		struct bbc_cpu_temperature *next = tp->next;
+	list_for_each_entry_safe(tp, tpos, &bp->temps, bp_list) {
+		list_del(&tp->bp_list);
+		list_del(&tp->glob_list);
 		destroy_one_temp(tp);
-		tp = next;
 	}
-	all_bbc_temps = NULL;
 
-	fp = all_bbc_fans;
-	while (fp != NULL) {
-		struct bbc_fan_control *next = fp->next;
+	list_for_each_entry_safe(fp, fpos, &bp->fans, bp_list) {
+		list_del(&fp->bp_list);
+		list_del(&fp->glob_list);
 		destroy_one_fan(fp);
-		fp = next;
 	}
-	all_bbc_fans = NULL;
 }
diff --git a/drivers/sbus/char/bbc_i2c.c b/drivers/sbus/char/bbc_i2c.c
index ac8ef2ce07fb..f08e169ba1b5 100644
--- a/drivers/sbus/char/bbc_i2c.c
+++ b/drivers/sbus/char/bbc_i2c.c
@@ -1,8 +1,7 @@
-/* $Id: bbc_i2c.c,v 1.2 2001/04/02 09:59:08 davem Exp $
- * bbc_i2c.c: I2C low-level driver for BBC device on UltraSPARC-III
+/* bbc_i2c.c: I2C low-level driver for BBC device on UltraSPARC-III
  *            platforms.
  *
- * Copyright (C) 2001 David S. Miller (davem@redhat.com)
+ * Copyright (C) 2001, 2008 David S. Miller (davem@davemloft.net)
  */
 
 #include <linux/module.h>
@@ -14,9 +13,8 @@
 #include <linux/delay.h>
 #include <linux/init.h>
 #include <linux/interrupt.h>
-#include <asm/oplib.h>
-#include <asm/ebus.h>
-#include <asm/spitfire.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 #include <asm/bbc.h>
 #include <asm/io.h>
 
@@ -53,54 +51,12 @@
  * The second controller also connects to the smartcard reader, if present.
  */
 
-#define NUM_CHILDREN	8
-struct bbc_i2c_bus {
-	struct bbc_i2c_bus		*next;
-	int				index;
-	spinlock_t			lock;
-	void				__iomem *i2c_bussel_reg;
-	void				__iomem *i2c_control_regs;
-	unsigned char			own, clock;
-
-	wait_queue_head_t		wq;
-	volatile int			waiting;
-
-	struct linux_ebus_device	*bus_edev;
-	struct {
-		struct linux_ebus_child	*device;
-		int			client_claimed;
-	} devs[NUM_CHILDREN];
-};
-
-static struct bbc_i2c_bus *all_bbc_i2c;
-
-struct bbc_i2c_client {
-	struct bbc_i2c_bus	*bp;
-	struct linux_ebus_child	*echild;
-	int			bus;
-	int			address;
-};
-
-static int find_device(struct bbc_i2c_bus *bp, struct linux_ebus_child *echild)
+static void set_device_claimage(struct bbc_i2c_bus *bp, struct of_device *op, int val)
 {
 	int i;
 
 	for (i = 0; i < NUM_CHILDREN; i++) {
-		if (bp->devs[i].device == echild) {
-			if (bp->devs[i].client_claimed)
-				return 0;
-			return 1;
-		}
-	}
-	return 0;
-}
-
-static void set_device_claimage(struct bbc_i2c_bus *bp, struct linux_ebus_child *echild, int val)
-{
-	int i;
-
-	for (i = 0; i < NUM_CHILDREN; i++) {
-		if (bp->devs[i].device == echild) {
+		if (bp->devs[i].device == op) {
 			bp->devs[i].client_claimed = val;
 			return;
 		}
@@ -110,61 +66,47 @@ static void set_device_claimage(struct bbc_i2c_bus *bp, struct linux_ebus_child
 #define claim_device(BP,ECHILD)		set_device_claimage(BP,ECHILD,1)
 #define release_device(BP,ECHILD)	set_device_claimage(BP,ECHILD,0)
 
-static struct bbc_i2c_bus *find_bus_for_device(struct linux_ebus_child *echild)
+struct of_device *bbc_i2c_getdev(struct bbc_i2c_bus *bp, int index)
 {
-	struct bbc_i2c_bus *bp = all_bbc_i2c;
+	struct of_device *op = NULL;
+	int curidx = 0, i;
 
-	while (bp != NULL) {
-		if (find_device(bp, echild) != 0)
+	for (i = 0; i < NUM_CHILDREN; i++) {
+		if (!(op = bp->devs[i].device))
 			break;
-		bp = bp->next;
+		if (curidx == index)
+			goto out;
+		op = NULL;
+		curidx++;
 	}
 
-	return bp;
-}
-
-struct linux_ebus_child *bbc_i2c_getdev(int index)
-{
-	struct bbc_i2c_bus *bp = all_bbc_i2c;
-	struct linux_ebus_child *echild = NULL;
-	int curidx = 0;
-
-	while (bp != NULL) {
-		struct bbc_i2c_bus *next = bp->next;
-		int i;
-
-		for (i = 0; i < NUM_CHILDREN; i++) {
-			if (!(echild = bp->devs[i].device))
-				break;
-			if (curidx == index)
-				goto out;
-			echild = NULL;
-			curidx++;
-		}
-		bp = next;
-	}
 out:
 	if (curidx == index)
-		return echild;
+		return op;
 	return NULL;
 }
 
-struct bbc_i2c_client *bbc_i2c_attach(struct linux_ebus_child *echild)
+struct bbc_i2c_client *bbc_i2c_attach(struct bbc_i2c_bus *bp, struct of_device *op)
 {
-	struct bbc_i2c_bus *bp = find_bus_for_device(echild);
 	struct bbc_i2c_client *client;
+	const u32 *reg;
 
-	if (!bp)
-		return NULL;
 	client = kzalloc(sizeof(*client), GFP_KERNEL);
 	if (!client)
 		return NULL;
 	client->bp = bp;
-	client->echild = echild;
-	client->bus = echild->resource[0].start;
-	client->address = echild->resource[1].start;
+	client->op = op;
+
+	reg = of_get_property(op->node, "reg", NULL);
+	if (!reg) {
+		kfree(client);
+		return NULL;
+	}
 
-	claim_device(bp, echild);
+	client->bus = reg[0];
+	client->address = reg[1];
+
+	claim_device(bp, op);
 
 	return client;
 }
@@ -172,9 +114,9 @@ struct bbc_i2c_client *bbc_i2c_attach(struct linux_ebus_child *echild)
 void bbc_i2c_detach(struct bbc_i2c_client *client)
 {
 	struct bbc_i2c_bus *bp = client->bp;
-	struct linux_ebus_child *echild = client->echild;
+	struct of_device *op = client->op;
 
-	release_device(bp, echild);
+	release_device(bp, op);
 	kfree(client);
 }
 
@@ -355,44 +297,43 @@ static void __init reset_one_i2c(struct bbc_i2c_bus *bp)
 	writeb(I2C_PCF_IDLE, bp->i2c_control_regs + 0x0);
 }
 
-static int __init attach_one_i2c(struct linux_ebus_device *edev, int index)
+static struct bbc_i2c_bus * __init attach_one_i2c(struct of_device *op, int index)
 {
 	struct bbc_i2c_bus *bp;
-	struct linux_ebus_child *echild;
+	struct device_node *dp;
 	int entry;
 
 	bp = kzalloc(sizeof(*bp), GFP_KERNEL);
 	if (!bp)
-		return -ENOMEM;
+		return NULL;
 
-	bp->i2c_control_regs = ioremap(edev->resource[0].start, 0x2);
+	bp->i2c_control_regs = of_ioremap(&op->resource[0], 0, 0x2, "bbc_i2c_regs");
 	if (!bp->i2c_control_regs)
 		goto fail;
 
-	if (edev->num_addrs == 2) {
-		bp->i2c_bussel_reg = ioremap(edev->resource[1].start, 0x1);
-		if (!bp->i2c_bussel_reg)
-			goto fail;
-	}
+	bp->i2c_bussel_reg = of_ioremap(&op->resource[1], 0, 0x1, "bbc_i2c_bussel");
+	if (!bp->i2c_bussel_reg)
+		goto fail;
 
 	bp->waiting = 0;
 	init_waitqueue_head(&bp->wq);
-	if (request_irq(edev->irqs[0], bbc_i2c_interrupt,
+	if (request_irq(op->irqs[0], bbc_i2c_interrupt,
 			IRQF_SHARED, "bbc_i2c", bp))
 		goto fail;
 
 	bp->index = index;
-	bp->bus_edev = edev;
+	bp->op = op;
 
 	spin_lock_init(&bp->lock);
-	bp->next = all_bbc_i2c;
-	all_bbc_i2c = bp;
 
 	entry = 0;
-	for (echild = edev->children;
-	     echild && entry < 8;
-	     echild = echild->next, entry++) {
-		bp->devs[entry].device = echild;
+	for (dp = op->node->child;
+	     dp && entry < 8;
+	     dp = dp->sibling, entry++) {
+		struct of_device *child_op;
+
+		child_op = of_find_device_by_node(dp);
+		bp->devs[entry].device = child_op;
 		bp->devs[entry].client_claimed = 0;
 	}
 
@@ -406,86 +347,90 @@ static int __init attach_one_i2c(struct linux_ebus_device *edev, int index)
 
 	reset_one_i2c(bp);
 
-	return 0;
+	return bp;
 
 fail:
 	if (bp->i2c_bussel_reg)
-		iounmap(bp->i2c_bussel_reg);
+		of_iounmap(&op->resource[1], bp->i2c_bussel_reg, 1);
 	if (bp->i2c_control_regs)
-		iounmap(bp->i2c_control_regs);
+		of_iounmap(&op->resource[0], bp->i2c_control_regs, 2);
 	kfree(bp);
-	return -EINVAL;
-}
-
-static int __init bbc_present(void)
-{
-	struct linux_ebus *ebus = NULL;
-	struct linux_ebus_device *edev = NULL;
-
-	for_each_ebus(ebus) {
-		for_each_ebusdev(edev, ebus) {
-			if (!strcmp(edev->prom_node->name, "bbc"))
-				return 1;
-		}
-	}
-	return 0;
+	return NULL;
 }
 
-extern int bbc_envctrl_init(void);
-extern void bbc_envctrl_cleanup(void);
-static void bbc_i2c_cleanup(void);
+extern int bbc_envctrl_init(struct bbc_i2c_bus *bp);
+extern void bbc_envctrl_cleanup(struct bbc_i2c_bus *bp);
 
-static int __init bbc_i2c_init(void)
+static int __devinit bbc_i2c_probe(struct of_device *op,
+				   const struct of_device_id *match)
 {
-	struct linux_ebus *ebus = NULL;
-	struct linux_ebus_device *edev = NULL;
+	struct bbc_i2c_bus *bp;
 	int err, index = 0;
 
-	if ((tlb_type != cheetah && tlb_type != cheetah_plus) ||
-	    !bbc_present())
-		return -ENODEV;
+	bp = attach_one_i2c(op, index);
+	if (!bp)
+		return -EINVAL;
 
-	for_each_ebus(ebus) {
-		for_each_ebusdev(edev, ebus) {
-			if (!strcmp(edev->prom_node->name, "i2c")) {
-				if (!attach_one_i2c(edev, index))
-					index++;
-			}
-		}
+	err = bbc_envctrl_init(bp);
+	if (err) {
+		free_irq(op->irqs[0], bp);
+		if (bp->i2c_bussel_reg)
+			of_iounmap(&op->resource[0], bp->i2c_bussel_reg, 1);
+		if (bp->i2c_control_regs)
+			of_iounmap(&op->resource[1], bp->i2c_control_regs, 2);
+		kfree(bp);
+	} else {
+		dev_set_drvdata(&op->dev, bp);
 	}
 
-	if (!index)
-		return -ENODEV;
-
-	err = bbc_envctrl_init();
-	if (err)
-		bbc_i2c_cleanup();
 	return err;
 }
 
-static void bbc_i2c_cleanup(void)
+static int __devexit bbc_i2c_remove(struct of_device *op)
 {
-	struct bbc_i2c_bus *bp = all_bbc_i2c;
+	struct bbc_i2c_bus *bp = dev_get_drvdata(&op->dev);
+
+	bbc_envctrl_cleanup(bp);
+
+	free_irq(op->irqs[0], bp);
 
-	bbc_envctrl_cleanup();
+	if (bp->i2c_bussel_reg)
+		of_iounmap(&op->resource[0], bp->i2c_bussel_reg, 1);
+	if (bp->i2c_control_regs)
+		of_iounmap(&op->resource[1], bp->i2c_control_regs, 2);
 
-	while (bp != NULL) {
-		struct bbc_i2c_bus *next = bp->next;
+	kfree(bp);
 
-		free_irq(bp->bus_edev->irqs[0], bp);
+	return 0;
+}
 
-		if (bp->i2c_bussel_reg)
-			iounmap(bp->i2c_bussel_reg);
-		if (bp->i2c_control_regs)
-			iounmap(bp->i2c_control_regs);
+static const struct of_device_id bbc_i2c_match[] = {
+	{
+		.name = "i2c",
+		.compatible = "SUNW,bbc-i2c",
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, bbc_i2c_match);
 
-		kfree(bp);
+static struct of_platform_driver bbc_i2c_driver = {
+	.name		= "bbc_i2c",
+	.match_table	= bbc_i2c_match,
+	.probe		= bbc_i2c_probe,
+	.remove		= __devexit_p(bbc_i2c_remove),
+};
 
-		bp = next;
-	}
-	all_bbc_i2c = NULL;
+static int __init bbc_i2c_init(void)
+{
+	return of_register_driver(&bbc_i2c_driver, &of_bus_type);
+}
+
+static void __exit bbc_i2c_exit(void)
+{
+	of_unregister_driver(&bbc_i2c_driver);
 }
 
 module_init(bbc_i2c_init);
-module_exit(bbc_i2c_cleanup);
+module_exit(bbc_i2c_exit);
+
 MODULE_LICENSE("GPL");
diff --git a/drivers/sbus/char/bbc_i2c.h b/drivers/sbus/char/bbc_i2c.h
index fb01bd17704b..83c4811b7b5e 100644
--- a/drivers/sbus/char/bbc_i2c.h
+++ b/drivers/sbus/char/bbc_i2c.h
@@ -1,14 +1,79 @@
-/* $Id: bbc_i2c.h,v 1.2 2001/04/02 09:59:25 davem Exp $ */
 #ifndef _BBC_I2C_H
 #define _BBC_I2C_H
 
-#include <asm/ebus.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/list.h>
 
-struct bbc_i2c_client;
+struct bbc_i2c_client {
+	struct bbc_i2c_bus		*bp;
+	struct of_device		*op;
+	int				bus;
+	int				address;
+};
+
+enum fan_action { FAN_SLOWER, FAN_SAME, FAN_FASTER, FAN_FULLBLAST, FAN_STATE_MAX };
+
+struct bbc_cpu_temperature {
+	struct list_head		bp_list;
+	struct list_head		glob_list;
+
+	struct bbc_i2c_client		*client;
+	int				index;
+
+	/* Current readings, and history. */
+	s8				curr_cpu_temp;
+	s8				curr_amb_temp;
+	s8				prev_cpu_temp;
+	s8				prev_amb_temp;
+	s8				avg_cpu_temp;
+	s8				avg_amb_temp;
+
+	int				sample_tick;
+
+	enum fan_action			fan_todo[2];
+#define FAN_AMBIENT	0
+#define FAN_CPU		1
+};
+
+struct bbc_fan_control {
+	struct list_head		bp_list;
+	struct list_head		glob_list;
+
+	struct bbc_i2c_client 		*client;
+	int 				index;
+
+	int				psupply_fan_on;
+	int				cpu_fan_speed;
+	int				system_fan_speed;
+};
+
+#define NUM_CHILDREN	8
+
+struct bbc_i2c_bus {
+	struct bbc_i2c_bus		*next;
+	int				index;
+	spinlock_t			lock;
+	void				__iomem *i2c_bussel_reg;
+	void				__iomem *i2c_control_regs;
+	unsigned char			own, clock;
+
+	wait_queue_head_t		wq;
+	volatile int			waiting;
+
+	struct list_head		temps;
+	struct list_head		fans;
+
+	struct of_device		*op;
+	struct {
+		struct of_device	*device;
+		int			client_claimed;
+	} devs[NUM_CHILDREN];
+};
 
 /* Probing and attachment. */
-extern struct linux_ebus_child *bbc_i2c_getdev(int);
-extern struct bbc_i2c_client *bbc_i2c_attach(struct linux_ebus_child *);
+extern struct of_device *bbc_i2c_getdev(struct bbc_i2c_bus *, int);
+extern struct bbc_i2c_client *bbc_i2c_attach(struct bbc_i2c_bus *bp, struct of_device *);
 extern void bbc_i2c_detach(struct bbc_i2c_client *);
 
 /* Register read/write.  NOTE: Blocking! */
diff --git a/drivers/sbus/char/bpp.c b/drivers/sbus/char/bpp.c
deleted file mode 100644
index bba21e053a1b..000000000000
--- a/drivers/sbus/char/bpp.c
+++ /dev/null
@@ -1,1055 +0,0 @@
-/*
- * drivers/sbus/char/bpp.c
- *
- * Copyright (c) 1995 Picture Elements
- *      Stephen Williams (steve@icarus.com)
- *      Gus Baldauf (gbaldauf@ix.netcom.com)
- *
- * Linux/SPARC port by Peter Zaitcev.
- * Integration into SPARC tree by Tom Dyas.
- */
-
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/fs.h>
-#include <linux/errno.h>
-#include <linux/sched.h>
-#include <linux/spinlock.h>
-#include <linux/timer.h>
-#include <linux/ioport.h>
-#include <linux/major.h>
-#include <linux/smp_lock.h>
-
-#include <asm/uaccess.h>
-#include <asm/io.h>
-
-#if defined(__i386__)
-# include <asm/system.h>
-#endif
-
-#if defined(__sparc__)
-# include <linux/init.h>
-# include <linux/delay.h>         /* udelay() */
-
-# include <asm/oplib.h>           /* OpenProm Library */
-# include <asm/sbus.h>
-#endif
-
-#include <asm/bpp.h>
-
-#define BPP_PROBE_CODE 0x55
-#define BPP_DELAY 100
-
-static const unsigned  BPP_MAJOR = LP_MAJOR;
-static const char *bpp_dev_name = "bpp";
-
-/* When switching from compatibility to a mode where I can read, try
-   the following mode first. */
-
-/* const unsigned char DEFAULT_ECP = 0x10; */
-static const unsigned char DEFAULT_ECP = 0x30;
-static const unsigned char DEFAULT_NIBBLE = 0x00;
-
-/*
- * These are 1284 time constraints, in units of jiffies.
- */
-
-static const unsigned long TIME_PSetup = 1;
-static const unsigned long TIME_PResponse = 6;
-static const unsigned long TIME_IDLE_LIMIT = 2000;
-
-/*
- * One instance per supported subdevice...
- */
-# define BPP_NO 3
-
-enum IEEE_Mode { COMPATIBILITY, NIBBLE, ECP, ECP_RLE, EPP };
-
-struct inst {
-      unsigned present  : 1; /* True if the hardware exists */
-      unsigned enhanced : 1; /* True if the hardware in "enhanced" */
-      unsigned opened   : 1; /* True if the device is opened already */
-      unsigned run_flag : 1; /* True if waiting for a repeate byte */
-
-      unsigned char direction; /* 0 --> out, 0x20 --> IN */
-      unsigned char pp_state; /* State of host controlled pins. */
-      enum IEEE_Mode mode;
-
-      unsigned char run_length;
-      unsigned char repeat_byte;
-};
-
-static struct inst instances[BPP_NO];
-
-#if defined(__i386__)
-
-static const unsigned short base_addrs[BPP_NO] = { 0x278, 0x378, 0x3bc };
-
-/*
- * These are for data access.
- * Control lines accesses are hidden in set_bits() and get_bits().
- * The exception is the probe procedure, which is system-dependent.
- */
-#define bpp_outb_p(data, base)  outb_p((data), (base))
-#define bpp_inb(base)  inb(base)
-#define bpp_inb_p(base)  inb_p(base)
-
-/*
- * This method takes the pin values mask and sets the hardware pins to
- * the requested value: 1 == high voltage, 0 == low voltage. This
- * burries the annoying PC bit inversion and preserves the direction
- * flag.
- */
-static void set_pins(unsigned short pins, unsigned minor)
-{
-      unsigned char bits = instances[minor].direction;  /* == 0x20 */
-
-      if (! (pins & BPP_PP_nStrobe))   bits |= 1;
-      if (! (pins & BPP_PP_nAutoFd))   bits |= 2;
-      if (   pins & BPP_PP_nInit)      bits |= 4;
-      if (! (pins & BPP_PP_nSelectIn)) bits |= 8;
-
-      instances[minor].pp_state = bits;
-
-      outb_p(bits, base_addrs[minor]+2);
-}
-
-static unsigned short get_pins(unsigned minor)
-{
-      unsigned short bits = 0;
-
-      unsigned value = instances[minor].pp_state;
-      if (! (value & 0x01)) bits |= BPP_PP_nStrobe;
-      if (! (value & 0x02)) bits |= BPP_PP_nAutoFd;
-      if (value & 0x04)     bits |= BPP_PP_nInit;
-      if (! (value & 0x08)) bits |= BPP_PP_nSelectIn;
-
-      value = inb_p(base_addrs[minor]+1);
-      if (value & 0x08)     bits |= BPP_GP_nFault;
-      if (value & 0x10)     bits |= BPP_GP_Select;
-      if (value & 0x20)     bits |= BPP_GP_PError;
-      if (value & 0x40)     bits |= BPP_GP_nAck;
-      if (! (value & 0x80)) bits |= BPP_GP_Busy;
-
-      return bits;
-}
-
-#endif /* __i386__ */
-
-#if defined(__sparc__)
-
-/*
- * Register block
- */
-      /* DMA registers */
-#define BPP_CSR      0x00
-#define BPP_ADDR     0x04
-#define BPP_BCNT     0x08
-#define BPP_TST_CSR  0x0C
-      /* Parallel Port registers */
-#define BPP_HCR      0x10
-#define BPP_OCR      0x12
-#define BPP_DR       0x14
-#define BPP_TCR      0x15
-#define BPP_OR       0x16
-#define BPP_IR       0x17
-#define BPP_ICR      0x18
-#define BPP_SIZE     0x1A
-
-/* BPP_CSR.  Bits of type RW1 are cleared with writing '1'. */
-#define P_DEV_ID_MASK   0xf0000000      /* R   */
-#define P_DEV_ID_ZEBRA  0x40000000
-#define P_DEV_ID_L64854 0xa0000000      /*      == NCR 89C100+89C105. Pity. */
-#define P_NA_LOADED     0x08000000      /* R    NA wirtten but was not used */
-#define P_A_LOADED      0x04000000      /* R    */
-#define P_DMA_ON        0x02000000      /* R    DMA is not disabled */
-#define P_EN_NEXT       0x01000000      /* RW   */
-#define P_TCI_DIS       0x00800000      /* RW   TCI forbidden from interrupts */
-#define P_DIAG          0x00100000      /* RW   Disables draining and resetting
-                                                of P-FIFO on loading of P_ADDR*/
-#define P_BURST_SIZE    0x000c0000      /* RW   SBus burst size */
-#define P_BURST_8       0x00000000
-#define P_BURST_4       0x00040000
-#define P_BURST_1       0x00080000      /*      "No burst" write */
-#define P_TC            0x00004000      /* RW1  Term Count, can be cleared when
-                                           P_EN_NEXT=1 */
-#define P_EN_CNT        0x00002000      /* RW   */
-#define P_EN_DMA        0x00000200      /* RW   */
-#define P_WRITE         0x00000100      /* R    DMA dir, 1=to ram, 0=to port */
-#define P_RESET         0x00000080      /* RW   */
-#define P_SLAVE_ERR     0x00000040      /* RW1  Access size error */
-#define P_INVALIDATE    0x00000020      /* W    Drop P-FIFO */
-#define P_INT_EN        0x00000010      /* RW   OK to P_INT_PEND||P_ERR_PEND */
-#define P_DRAINING      0x0000000c      /* R    P-FIFO is draining to memory */
-#define P_ERR_PEND      0x00000002      /* R    */
-#define P_INT_PEND      0x00000001      /* R    */
-
-/* BPP_HCR. Time is in increments of SBus clock. */
-#define P_HCR_TEST      0x8000      /* Allows buried counters to be read */
-#define P_HCR_DSW       0x7f00      /* Data strobe width (in ticks) */
-#define P_HCR_DDS       0x007f      /* Data setup before strobe (in ticks) */
-
-/* BPP_OCR. */
-#define P_OCR_MEM_CLR   0x8000
-#define P_OCR_DATA_SRC  0x4000      /* )                  */
-#define P_OCR_DS_DSEL   0x2000      /* )  Bidirectional      */
-#define P_OCR_BUSY_DSEL 0x1000      /* )    selects            */
-#define P_OCR_ACK_DSEL  0x0800      /* )                  */
-#define P_OCR_EN_DIAG   0x0400
-#define P_OCR_BUSY_OP   0x0200      /* Busy operation */
-#define P_OCR_ACK_OP    0x0100      /* Ack operation */
-#define P_OCR_SRST      0x0080      /* Reset state machines. Not selfcleaning. */
-#define P_OCR_IDLE      0x0008      /* PP data transfer state machine is idle */
-#define P_OCR_V_ILCK    0x0002      /* Versatec faded. Zebra only. */
-#define P_OCR_EN_VER    0x0001      /* Enable Versatec (0 - enable). Zebra only. */
-
-/* BPP_TCR */
-#define P_TCR_DIR       0x08
-#define P_TCR_BUSY      0x04
-#define P_TCR_ACK       0x02
-#define P_TCR_DS        0x01        /* Strobe */
-
-/* BPP_OR */
-#define P_OR_V3         0x20        /* )                 */
-#define P_OR_V2         0x10        /* ) on Zebra only   */
-#define P_OR_V1         0x08        /* )                 */
-#define P_OR_INIT       0x04
-#define P_OR_AFXN       0x02        /* Auto Feed */
-#define P_OR_SLCT_IN    0x01
-
-/* BPP_IR */
-#define P_IR_PE         0x04
-#define P_IR_SLCT       0x02
-#define P_IR_ERR        0x01
-
-/* BPP_ICR */
-#define P_DS_IRQ        0x8000      /* RW1  */
-#define P_ACK_IRQ       0x4000      /* RW1  */
-#define P_BUSY_IRQ      0x2000      /* RW1  */
-#define P_PE_IRQ        0x1000      /* RW1  */
-#define P_SLCT_IRQ      0x0800      /* RW1  */
-#define P_ERR_IRQ       0x0400      /* RW1  */
-#define P_DS_IRQ_EN     0x0200      /* RW   Always on rising edge */
-#define P_ACK_IRQ_EN    0x0100      /* RW   Always on rising edge */
-#define P_BUSY_IRP      0x0080      /* RW   1= rising edge */
-#define P_BUSY_IRQ_EN   0x0040      /* RW   */
-#define P_PE_IRP        0x0020      /* RW   1= rising edge */
-#define P_PE_IRQ_EN     0x0010      /* RW   */
-#define P_SLCT_IRP      0x0008      /* RW   1= rising edge */
-#define P_SLCT_IRQ_EN   0x0004      /* RW   */
-#define P_ERR_IRP       0x0002      /* RW1  1= rising edge */
-#define P_ERR_IRQ_EN    0x0001      /* RW   */
-
-static void __iomem *base_addrs[BPP_NO];
-
-#define bpp_outb_p(data, base)	sbus_writeb(data, (base) + BPP_DR)
-#define bpp_inb_p(base)		sbus_readb((base) + BPP_DR)
-#define bpp_inb(base)		sbus_readb((base) + BPP_DR)
-
-static void set_pins(unsigned short pins, unsigned minor)
-{
-      void __iomem *base = base_addrs[minor];
-      unsigned char bits_tcr = 0, bits_or = 0;
-
-      if (instances[minor].direction & 0x20) bits_tcr |= P_TCR_DIR;
-      if (   pins & BPP_PP_nStrobe)          bits_tcr |= P_TCR_DS;
-
-      if (   pins & BPP_PP_nAutoFd)          bits_or |= P_OR_AFXN;
-      if (! (pins & BPP_PP_nInit))           bits_or |= P_OR_INIT;
-      if (! (pins & BPP_PP_nSelectIn))       bits_or |= P_OR_SLCT_IN;
-
-      sbus_writeb(bits_or, base + BPP_OR);
-      sbus_writeb(bits_tcr, base + BPP_TCR);
-}
-
-/*
- * i386 people read output pins from a software image.
- * We may get them back from hardware.
- * Again, inversion of pins must he buried here.
- */
-static unsigned short get_pins(unsigned minor)
-{
-      void __iomem *base = base_addrs[minor];
-      unsigned short bits = 0;
-      unsigned value_tcr = sbus_readb(base + BPP_TCR);
-      unsigned value_ir = sbus_readb(base + BPP_IR);
-      unsigned value_or = sbus_readb(base + BPP_OR);
-
-      if (value_tcr & P_TCR_DS)         bits |= BPP_PP_nStrobe;
-      if (value_or & P_OR_AFXN)         bits |= BPP_PP_nAutoFd;
-      if (! (value_or & P_OR_INIT))     bits |= BPP_PP_nInit;
-      if (! (value_or & P_OR_SLCT_IN))  bits |= BPP_PP_nSelectIn;
-
-      if (value_ir & P_IR_ERR)          bits |= BPP_GP_nFault;
-      if (! (value_ir & P_IR_SLCT))     bits |= BPP_GP_Select;
-      if (! (value_ir & P_IR_PE))       bits |= BPP_GP_PError;
-      if (! (value_tcr & P_TCR_ACK))    bits |= BPP_GP_nAck;
-      if (value_tcr & P_TCR_BUSY)       bits |= BPP_GP_Busy;
-
-      return bits;
-}
-
-#endif /* __sparc__ */
-
-static void snooze(unsigned long snooze_time, unsigned minor)
-{
-	schedule_timeout_uninterruptible(snooze_time + 1);
-}
-
-static int wait_for(unsigned short set, unsigned short clr,
-               unsigned long delay, unsigned minor)
-{
-      unsigned short pins = get_pins(minor);
-
-      unsigned long extime = 0;
-
-      /*
-       * Try a real fast scan for the first jiffy, in case the device
-       * responds real good. The first while loop guesses an expire
-       * time accounting for possible wraparound of jiffies.
-       */
-      while (time_after_eq(jiffies, extime)) extime = jiffies + 1;
-      while ( (time_before(jiffies, extime))
-              && (((pins & set) != set) || ((pins & clr) != 0)) ) {
-            pins = get_pins(minor);
-      }
-
-      delay -= 1;
-
-      /*
-       * If my delay expired or the pins are still not where I want
-       * them, then resort to using the timer and greatly reduce my
-       * sample rate. If the peripheral is going to be slow, this will
-       * give the CPU up to some more worthy process.
-       */
-      while ( delay && (((pins & set) != set) || ((pins & clr) != 0)) ) {
-
-            snooze(1, minor);
-            pins = get_pins(minor);
-            delay -= 1;
-      }
-
-      if (delay == 0) return -1;
-      else return pins;
-}
-
-/*
- * Return ZERO(0) If the negotiation succeeds, an errno otherwise. An
- * errno means something broke, and I do not yet know how to fix it.
- */
-static int negotiate(unsigned char mode, unsigned minor)
-{
-      int rc;
-      unsigned short pins = get_pins(minor);
-      if (pins & BPP_PP_nSelectIn) return -EIO;
-
-
-        /* Event 0: Write the mode to the data lines */
-      bpp_outb_p(mode, base_addrs[minor]);
-
-      snooze(TIME_PSetup, minor);
-
-        /* Event 1: Strobe the mode code into the peripheral */
-      set_pins(BPP_PP_nSelectIn|BPP_PP_nStrobe|BPP_PP_nInit, minor);
-
-        /* Wait for Event 2: Peripheral responds as a 1284 device. */
-      rc = wait_for(BPP_GP_PError|BPP_GP_Select|BPP_GP_nFault,
-                BPP_GP_nAck,
-                TIME_PResponse,
-                minor);
-
-      if (rc == -1) return -ETIMEDOUT;
-
-        /* Event 3: latch extensibility request */
-      set_pins(BPP_PP_nSelectIn|BPP_PP_nInit, minor);
-
-        /* ... quick nap while peripheral ponders the byte i'm sending...*/
-      snooze(1, minor);
-
-        /* Event 4: restore strobe, to ACK peripheral's response. */
-      set_pins(BPP_PP_nSelectIn|BPP_PP_nAutoFd|BPP_PP_nStrobe|BPP_PP_nInit, minor);
-
-        /* Wait for Event 6: Peripheral latches response bits */
-      rc = wait_for(BPP_GP_nAck, 0, TIME_PSetup+TIME_PResponse, minor);
-      if (rc == -1) return -EIO;
-
-        /* A 1284 device cannot refuse nibble mode */
-      if (mode == DEFAULT_NIBBLE) return 0;
-
-      if (pins & BPP_GP_Select) return 0;
-
-      return -EPROTONOSUPPORT;
-}
-
-static int terminate(unsigned minor)
-{
-      int rc;
-
-        /* Event 22: Request termination of 1284 mode */
-      set_pins(BPP_PP_nAutoFd|BPP_PP_nStrobe|BPP_PP_nInit, minor);
-
-        /* Wait for Events 23 and 24: ACK termination request. */
-      rc = wait_for(BPP_GP_Busy|BPP_GP_nFault,
-                BPP_GP_nAck,
-                TIME_PSetup+TIME_PResponse,
-                minor);
-
-      instances[minor].direction = 0;
-      instances[minor].mode = COMPATIBILITY;
-
-      if (rc == -1) {
-          return -EIO;
-      }
-
-        /* Event 25: Handshake by lowering nAutoFd */
-      set_pins(BPP_PP_nStrobe|BPP_PP_nInit, minor);
-
-        /* Event 26: Peripheral wiggles lines... */
-
-        /* Event 27: Peripheral sets nAck HIGH to ack handshake */
-      rc = wait_for(BPP_GP_nAck, 0, TIME_PResponse, minor);
-      if (rc == -1) {
-          set_pins(BPP_PP_nAutoFd|BPP_PP_nStrobe|BPP_PP_nInit, minor);
-          return -EIO;
-      }
-
-        /* Event 28: Finish phase by raising nAutoFd */
-      set_pins(BPP_PP_nAutoFd|BPP_PP_nStrobe|BPP_PP_nInit, minor);
-
-      return 0;
-}
-
-static DEFINE_SPINLOCK(bpp_open_lock);
-
-/*
- * Allow only one process to open the device at a time.
- */
-static int bpp_open(struct inode *inode, struct file *f)
-{
-      unsigned minor = iminor(inode);
-      int ret;
-
-      lock_kernel();
-      spin_lock(&bpp_open_lock);
-      ret = 0;
-      if (minor >= BPP_NO) {
-	      ret = -ENODEV;
-      } else {
-	      if (! instances[minor].present) {
-		      ret = -ENODEV;
-	      } else {
-		      if (instances[minor].opened) 
-			      ret = -EBUSY;
-		      else
-			      instances[minor].opened = 1;
-	      }
-      }
-      spin_unlock(&bpp_open_lock);
-      unlock_kernel();
-
-      return ret;
-}
-
-/*
- * When the process closes the device, this method is called to clean
- * up and reset the hardware. Always leave the device in compatibility
- * mode as this is a reasonable place to clean up from messes made by
- * ioctls, or other mayhem.
- */
-static int bpp_release(struct inode *inode, struct file *f)
-{
-      unsigned minor = iminor(inode);
-
-      spin_lock(&bpp_open_lock);
-      instances[minor].opened = 0;
-
-      if (instances[minor].mode != COMPATIBILITY)
-	      terminate(minor);
-
-      spin_unlock(&bpp_open_lock);
-
-      return 0;
-}
-
-static long read_nibble(unsigned minor, char __user *c, unsigned long cnt)
-{
-      unsigned long remaining = cnt;
-      long rc;
-
-      while (remaining > 0) {
-          unsigned char byte = 0;
-          int pins;
-
-          /* Event 7: request nibble */
-          set_pins(BPP_PP_nSelectIn|BPP_PP_nStrobe, minor);
-
-          /* Wait for event 9: Peripher strobes first nibble */
-          pins = wait_for(0, BPP_GP_nAck, TIME_IDLE_LIMIT, minor);
-          if (pins == -1) return -ETIMEDOUT;
-
-          /* Event 10: I handshake nibble */
-          set_pins(BPP_PP_nSelectIn|BPP_PP_nStrobe|BPP_PP_nAutoFd, minor);
-          if (pins & BPP_GP_nFault) byte |= 0x01;
-          if (pins & BPP_GP_Select) byte |= 0x02;
-          if (pins & BPP_GP_PError) byte |= 0x04;
-          if (pins & BPP_GP_Busy)   byte |= 0x08;
-
-          /* Wait for event 11: Peripheral handshakes nibble */
-          rc = wait_for(BPP_GP_nAck, 0, TIME_PResponse, minor);
-
-          /* Event 7: request nibble */
-          set_pins(BPP_PP_nSelectIn|BPP_PP_nStrobe, minor);
-
-          /* Wait for event 9: Peripher strobes first nibble */
-          pins = wait_for(0, BPP_GP_nAck, TIME_PResponse, minor);
-          if (rc == -1) return -ETIMEDOUT;
-
-          /* Event 10: I handshake nibble */
-          set_pins(BPP_PP_nSelectIn|BPP_PP_nStrobe|BPP_PP_nAutoFd, minor);
-          if (pins & BPP_GP_nFault) byte |= 0x10;
-          if (pins & BPP_GP_Select) byte |= 0x20;
-          if (pins & BPP_GP_PError) byte |= 0x40;
-          if (pins & BPP_GP_Busy)   byte |= 0x80;
-
-          if (put_user(byte, c))
-		  return -EFAULT;
-          c += 1;
-          remaining -= 1;
-
-          /* Wait for event 11: Peripheral handshakes nibble */
-          rc = wait_for(BPP_GP_nAck, 0, TIME_PResponse, minor);
-          if (rc == -1) return -EIO;
-      }
-
-      return cnt - remaining;
-}
-
-static long read_ecp(unsigned minor, char __user *c, unsigned long cnt)
-{
-      unsigned long remaining;
-      long rc;
-
-        /* Turn ECP mode from forward to reverse if needed. */
-      if (! instances[minor].direction) {
-          unsigned short pins = get_pins(minor);
-
-            /* Event 38: Turn the bus around */
-          instances[minor].direction = 0x20;
-          pins &= ~BPP_PP_nAutoFd;
-          set_pins(pins, minor);
-
-            /* Event 39: Set pins for reverse mode. */
-          snooze(TIME_PSetup, minor);
-          set_pins(BPP_PP_nStrobe|BPP_PP_nSelectIn, minor);
-
-            /* Wait for event 40: Peripheral ready to be strobed */
-          rc = wait_for(0, BPP_GP_PError, TIME_PResponse, minor);
-          if (rc == -1) return -ETIMEDOUT;
-      }
-
-      remaining = cnt;
-
-      while (remaining > 0) {
-
-            /* If there is a run length for a repeated byte, repeat */
-            /* that byte a few times. */
-          if (instances[minor].run_length && !instances[minor].run_flag) {
-
-              char buffer[128];
-              unsigned idx;
-              unsigned repeat = remaining < instances[minor].run_length
-                                     ? remaining
-                               : instances[minor].run_length;
-
-              for (idx = 0 ;  idx < repeat ;  idx += 1)
-                buffer[idx] = instances[minor].repeat_byte;
-
-              if (copy_to_user(c, buffer, repeat))
-		      return -EFAULT;
-              remaining -= repeat;
-              c += repeat;
-              instances[minor].run_length -= repeat;
-          }
-
-          if (remaining == 0) break;
-
-
-            /* Wait for Event 43: Data active on the bus. */
-          rc = wait_for(0, BPP_GP_nAck, TIME_IDLE_LIMIT, minor);
-          if (rc == -1) break;
-
-          if (rc & BPP_GP_Busy) {
-                /* OK, this is data. read it in. */
-              unsigned char byte = bpp_inb(base_addrs[minor]);
-              if (put_user(byte, c))
-		      return -EFAULT;
-              c += 1;
-              remaining -= 1;
-
-              if (instances[minor].run_flag) {
-                  instances[minor].repeat_byte = byte;
-                  instances[minor].run_flag = 0;
-              }
-
-          } else {
-              unsigned char byte = bpp_inb(base_addrs[minor]);
-              if (byte & 0x80) {
-                  printk("bpp%d: "
-                         "Ignoring ECP channel %u from device.\n",
-                         minor, byte & 0x7f);
-              } else {
-                  instances[minor].run_length = byte;
-                  instances[minor].run_flag = 1;
-              }
-          }
-
-            /* Event 44: I got it. */
-          set_pins(BPP_PP_nStrobe|BPP_PP_nAutoFd|BPP_PP_nSelectIn, minor);
-
-            /* Wait for event 45: peripheral handshake */
-          rc = wait_for(BPP_GP_nAck, 0, TIME_PResponse, minor);
-          if (rc == -1) return -ETIMEDOUT;
-
-             /* Event 46: Finish handshake */
-          set_pins(BPP_PP_nStrobe|BPP_PP_nSelectIn, minor);
-
-      }
-
-
-      return cnt - remaining;
-}
-
-static ssize_t bpp_read(struct file *f, char __user *c, size_t cnt, loff_t * ppos)
-{
-      long rc;
-      unsigned minor = iminor(f->f_path.dentry->d_inode);
-      if (minor >= BPP_NO) return -ENODEV;
-      if (!instances[minor].present) return -ENODEV;
-
-      switch (instances[minor].mode) {
-
-        default:
-          if (instances[minor].mode != COMPATIBILITY)
-            terminate(minor);
-
-          if (instances[minor].enhanced) {
-              /* For now, do all reads with ECP-RLE mode */
-              unsigned short pins;
-
-              rc = negotiate(DEFAULT_ECP, minor);
-              if (rc < 0) break;
-
-              instances[minor].mode = ECP_RLE;
-
-              /* Event 30: set nAutoFd low to setup for ECP mode */
-              pins = get_pins(minor);
-              pins &= ~BPP_PP_nAutoFd;
-              set_pins(pins, minor);
-
-              /* Wait for Event 31: peripheral ready */
-              rc = wait_for(BPP_GP_PError, 0, TIME_PResponse, minor);
-              if (rc == -1) return -ETIMEDOUT;
-
-              rc = read_ecp(minor, c, cnt);
-
-          } else {
-              rc = negotiate(DEFAULT_NIBBLE, minor);
-              if (rc < 0) break;
-
-              instances[minor].mode = NIBBLE;
-
-              rc = read_nibble(minor, c, cnt);
-          }
-          break;
-
-        case NIBBLE:
-          rc = read_nibble(minor, c, cnt);
-          break;
-
-        case ECP:
-        case ECP_RLE:
-          rc = read_ecp(minor, c, cnt);
-          break;
-
-      }
-
-
-      return rc;
-}
-
-/*
- * Compatibility mode handshaking is a matter of writing data,
- * strobing it, and waiting for the printer to stop being busy.
- */
-static long write_compat(unsigned minor, const char __user *c, unsigned long cnt)
-{
-      long rc;
-      unsigned short pins = get_pins(minor);
-
-      unsigned long remaining = cnt;
-
-
-      while (remaining > 0) {
-            unsigned char byte;
-
-            if (get_user(byte, c))
-		    return -EFAULT;
-            c += 1;
-
-            rc = wait_for(BPP_GP_nAck, BPP_GP_Busy, TIME_IDLE_LIMIT, minor);
-            if (rc == -1) return -ETIMEDOUT;
-
-            bpp_outb_p(byte, base_addrs[minor]);
-            remaining -= 1;
-          /* snooze(1, minor); */
-
-          pins &= ~BPP_PP_nStrobe;
-          set_pins(pins, minor);
-
-          rc = wait_for(BPP_GP_Busy, 0, TIME_PResponse, minor);
-
-          pins |= BPP_PP_nStrobe;
-          set_pins(pins, minor);
-      }
-
-      return cnt - remaining;
-}
-
-/*
- * Write data using ECP mode. Watch out that the port may be set up
- * for reading. If so, turn the port around.
- */
-static long write_ecp(unsigned minor, const char __user *c, unsigned long cnt)
-{
-      unsigned short pins = get_pins(minor);
-      unsigned long remaining = cnt;
-
-      if (instances[minor].direction) {
-          int rc;
-
-            /* Event 47 Request bus be turned around */
-          pins |= BPP_PP_nInit;
-          set_pins(pins, minor);
-
-            /* Wait for Event 49: Peripheral relinquished bus */
-          rc = wait_for(BPP_GP_PError, 0, TIME_PResponse, minor);
-
-          pins |= BPP_PP_nAutoFd;
-          instances[minor].direction = 0;
-          set_pins(pins, minor);
-      }
-
-      while (remaining > 0) {
-          unsigned char byte;
-          int rc;
-
-          if (get_user(byte, c))
-		  return -EFAULT;
-
-          rc = wait_for(0, BPP_GP_Busy, TIME_PResponse, minor);
-          if (rc == -1) return -ETIMEDOUT;
-
-          c += 1;
-
-          bpp_outb_p(byte, base_addrs[minor]);
-
-          pins &= ~BPP_PP_nStrobe;
-          set_pins(pins, minor);
-
-          pins |= BPP_PP_nStrobe;
-          rc = wait_for(BPP_GP_Busy, 0, TIME_PResponse, minor);
-          if (rc == -1) return -EIO;
-
-          set_pins(pins, minor);
-      }
-
-      return cnt - remaining;
-}
-
-/*
- * Write to the peripheral. Be sensitive of the current mode. If I'm
- * in a mode that can be turned around (ECP) then just do
- * that. Otherwise, terminate and do my writing in compat mode. This
- * is the safest course as any device can handle it.
- */
-static ssize_t bpp_write(struct file *f, const char __user *c, size_t cnt, loff_t * ppos)
-{
-      long errno = 0;
-      unsigned minor = iminor(f->f_path.dentry->d_inode);
-      if (minor >= BPP_NO) return -ENODEV;
-      if (!instances[minor].present) return -ENODEV;
-
-      switch (instances[minor].mode) {
-
-        case ECP:
-        case ECP_RLE:
-          errno = write_ecp(minor, c, cnt);
-          break;
-        case COMPATIBILITY:
-          errno = write_compat(minor, c, cnt);
-          break;
-        default:
-          terminate(minor);
-          errno = write_compat(minor, c, cnt);
-      }
-
-      return errno;
-}
-
-static int bpp_ioctl(struct inode *inode, struct file *f, unsigned int cmd,
-		 unsigned long arg)
-{
-      int errno = 0;
-
-      unsigned minor = iminor(inode);
-      if (minor >= BPP_NO) return -ENODEV;
-      if (!instances[minor].present) return -ENODEV;
-
-
-      switch (cmd) {
-
-        case BPP_PUT_PINS:
-          set_pins(arg, minor);
-          break;
-
-        case BPP_GET_PINS:
-          errno = get_pins(minor);
-          break;
-
-        case BPP_PUT_DATA:
-          bpp_outb_p(arg, base_addrs[minor]);
-          break;
-
-        case BPP_GET_DATA:
-          errno = bpp_inb_p(base_addrs[minor]);
-          break;
-
-        case BPP_SET_INPUT:
-          if (arg)
-            if (instances[minor].enhanced) {
-                unsigned short bits = get_pins(minor);
-                instances[minor].direction = 0x20;
-                set_pins(bits, minor);
-            } else {
-                errno = -ENOTTY;
-            }
-          else {
-              unsigned short bits = get_pins(minor);
-              instances[minor].direction = 0x00;
-              set_pins(bits, minor);
-          }
-          break;
-
-        default:
-            errno = -EINVAL;
-      }
-
-      return errno;
-}
-
-static const struct file_operations bpp_fops = {
-	.owner =	THIS_MODULE,
-	.read =		bpp_read,
-	.write =	bpp_write,
-	.ioctl =	bpp_ioctl,
-	.open =		bpp_open,
-	.release =	bpp_release,
-};
-
-#if defined(__i386__)
-
-#define collectLptPorts()  {}
-
-static void probeLptPort(unsigned idx)
-{
-      unsigned int testvalue;
-      const unsigned short lpAddr = base_addrs[idx];
-
-      instances[idx].present = 0;
-      instances[idx].enhanced = 0;
-      instances[idx].direction = 0;
-      instances[idx].mode = COMPATIBILITY;
-      instances[idx].run_length = 0;
-      instances[idx].run_flag = 0;
-      if (!request_region(lpAddr,3, bpp_dev_name)) return;
-
-      /*
-       * First, make sure the instance exists. Do this by writing to
-       * the data latch and reading the value back. If the port *is*
-       * present, test to see if it supports extended-mode
-       * operation. This will be required for IEEE1284 reverse
-       * transfers.
-       */
-
-      outb_p(BPP_PROBE_CODE, lpAddr);
-      for (testvalue=0; testvalue<BPP_DELAY; testvalue++)
-            ;
-      testvalue = inb_p(lpAddr);
-      if (testvalue == BPP_PROBE_CODE) {
-            unsigned save;
-            instances[idx].present = 1;
-
-            save = inb_p(lpAddr+2);
-            for (testvalue=0; testvalue<BPP_DELAY; testvalue++)
-                  ;
-            outb_p(save|0x20, lpAddr+2);
-            for (testvalue=0; testvalue<BPP_DELAY; testvalue++)
-                  ;
-            outb_p(~BPP_PROBE_CODE, lpAddr);
-            for (testvalue=0; testvalue<BPP_DELAY; testvalue++)
-                  ;
-            testvalue = inb_p(lpAddr);
-            if ((testvalue&0xff) == (0xff&~BPP_PROBE_CODE))
-                  instances[idx].enhanced = 0;
-            else
-                  instances[idx].enhanced = 1;
-            outb_p(save, lpAddr+2);
-      }
-      else {
-            release_region(lpAddr,3);
-      }
-      /*
-       * Leave the port in compat idle mode.
-       */
-      set_pins(BPP_PP_nAutoFd|BPP_PP_nStrobe|BPP_PP_nInit, idx);
-
-      printk("bpp%d: Port at 0x%03x: Enhanced mode %s\n", idx, base_addrs[idx],
-            instances[idx].enhanced? "SUPPORTED" : "UNAVAILABLE");
-}
-
-static inline void freeLptPort(int idx)
-{
-      release_region(base_addrs[idx], 3);
-}
-
-#endif
-
-#if defined(__sparc__)
-
-static void __iomem *map_bpp(struct sbus_dev *dev, int idx)
-{
-      return sbus_ioremap(&dev->resource[0], 0, BPP_SIZE, "bpp");
-}
-
-static int collectLptPorts(void)
-{
-	struct sbus_bus *bus;
-	struct sbus_dev *dev;
-	int count;
-
-	count = 0;
-	for_all_sbusdev(dev, bus) {
-		if (strcmp(dev->prom_name, "SUNW,bpp") == 0) {
-			if (count >= BPP_NO) {
-				printk(KERN_NOTICE
-				       "bpp: More than %d bpp ports,"
-				       " rest is ignored\n", BPP_NO);
-				return count;
-			}
-			base_addrs[count] = map_bpp(dev, count);
-			count++;
-		}
-	}
-	return count;
-}
-
-static void probeLptPort(unsigned idx)
-{
-      void __iomem *rp = base_addrs[idx];
-      __u32 csr;
-      char *brand;
-
-      instances[idx].present = 0;
-      instances[idx].enhanced = 0;
-      instances[idx].direction = 0;
-      instances[idx].mode = COMPATIBILITY;
-      instances[idx].run_length = 0;
-      instances[idx].run_flag = 0;
-
-      if (!rp) return;
-
-      instances[idx].present = 1;
-      instances[idx].enhanced = 1;   /* Sure */
-
-      csr = sbus_readl(rp + BPP_CSR);
-      if ((csr & P_DRAINING) != 0 && (csr & P_ERR_PEND) == 0) {
-            udelay(20);
-            csr = sbus_readl(rp + BPP_CSR);
-            if ((csr & P_DRAINING) != 0 && (csr & P_ERR_PEND) == 0) {
-                  printk("bpp%d: DRAINING still active (0x%08x)\n", idx, csr);
-            }
-      }
-      printk("bpp%d: reset with 0x%08x ..", idx, csr);
-      sbus_writel((csr | P_RESET) & ~P_INT_EN, rp + BPP_CSR);
-      udelay(500);
-      sbus_writel(sbus_readl(rp + BPP_CSR) & ~P_RESET, rp + BPP_CSR);
-      csr = sbus_readl(rp + BPP_CSR);
-      printk(" done with csr=0x%08x ocr=0x%04x\n",
-         csr, sbus_readw(rp + BPP_OCR));
-
-      switch (csr & P_DEV_ID_MASK) {
-      case P_DEV_ID_ZEBRA:
-            brand = "Zebra";
-            break;
-      case P_DEV_ID_L64854:
-            brand = "DMA2";
-            break;
-      default:
-            brand = "Unknown";
-      }
-      printk("bpp%d: %s at %p\n", idx, brand, rp);
-
-      /*
-       * Leave the port in compat idle mode.
-       */
-      set_pins(BPP_PP_nAutoFd|BPP_PP_nStrobe|BPP_PP_nInit, idx);
-
-      return;
-}
-
-static inline void freeLptPort(int idx)
-{
-      sbus_iounmap(base_addrs[idx], BPP_SIZE);
-}
-
-#endif
-
-static int __init bpp_init(void)
-{
-	int rc;
-	unsigned idx;
-
-	rc = collectLptPorts();
-	if (rc == 0)
-		return -ENODEV;
-
-	rc = register_chrdev(BPP_MAJOR, bpp_dev_name, &bpp_fops);
-	if (rc < 0)
-		return rc;
-
-	for (idx = 0; idx < BPP_NO; idx++) {
-		instances[idx].opened = 0;
-		probeLptPort(idx);
-	}
-
-	return 0;
-}
-
-static void __exit bpp_cleanup(void)
-{
-	unsigned idx;
-
-	unregister_chrdev(BPP_MAJOR, bpp_dev_name);
-
-	for (idx = 0;  idx < BPP_NO; idx++) {
-		if (instances[idx].present)
-			freeLptPort(idx);
-	}
-}
-
-module_init(bpp_init);
-module_exit(bpp_cleanup);
-
-MODULE_LICENSE("GPL");
-
diff --git a/drivers/sbus/char/cpwatchdog.c b/drivers/sbus/char/cpwatchdog.c
deleted file mode 100644
index 23abfdfb44f1..000000000000
--- a/drivers/sbus/char/cpwatchdog.c
+++ /dev/null
@@ -1,858 +0,0 @@
-/* cpwatchdog.c - driver implementation for hardware watchdog
- * timers found on Sun Microsystems CP1400 and CP1500 boards.
- *
- * This device supports both the generic Linux watchdog 
- * interface and Solaris-compatible ioctls as best it is
- * able.
- *
- * NOTE: 	CP1400 systems appear to have a defective intr_mask
- * 			register on the PLD, preventing the disabling of
- * 			timer interrupts.  We use a timer to periodically 
- * 			reset 'stopped' watchdogs on affected platforms.
- *
- * Copyright (c) 2000 Eric Brower (ebrower@usa.net)
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/fs.h>
-#include <linux/errno.h>
-#include <linux/major.h>
-#include <linux/init.h>
-#include <linux/miscdevice.h>
-#include <linux/interrupt.h>
-#include <linux/ioport.h>
-#include <linux/timer.h>
-#include <linux/smp_lock.h>
-#include <linux/io.h>
-#include <asm/irq.h>
-#include <asm/ebus.h>
-#include <asm/oplib.h>
-#include <asm/uaccess.h>
-
-#include <asm/watchdog.h>
-
-#define WD_OBPNAME	"watchdog"
-#define WD_BADMODEL "SUNW,501-5336"
-#define WD_BTIMEOUT	(jiffies + (HZ * 1000))
-#define WD_BLIMIT	0xFFFF
-
-#define WD0_DEVNAME "watchdog0"
-#define WD1_DEVNAME "watchdog1"
-#define WD2_DEVNAME "watchdog2"
-
-#define WD0_MINOR	212
-#define WD1_MINOR	213	
-#define WD2_MINOR	214	
-
-
-/* Internal driver definitions
- */
-#define WD0_ID			0		/* Watchdog0						*/
-#define WD1_ID			1		/* Watchdog1						*/
-#define WD2_ID			2		/* Watchdog2						*/
-#define WD_NUMDEVS		3		/* Device contains 3 timers			*/
-
-#define WD_INTR_OFF		0		/* Interrupt disable value			*/
-#define WD_INTR_ON		1		/* Interrupt enable value			*/
-
-#define WD_STAT_INIT	0x01	/* Watchdog timer is initialized	*/
-#define WD_STAT_BSTOP	0x02	/* Watchdog timer is brokenstopped	*/
-#define WD_STAT_SVCD	0x04	/* Watchdog interrupt occurred		*/
-
-/* Register value definitions
- */
-#define WD0_INTR_MASK	0x01	/* Watchdog device interrupt masks	*/
-#define WD1_INTR_MASK	0x02
-#define WD2_INTR_MASK	0x04
-
-#define WD_S_RUNNING	0x01	/* Watchdog device status running	*/
-#define WD_S_EXPIRED	0x02	/* Watchdog device status expired	*/
-
-/* Sun uses Altera PLD EPF8820ATC144-4 
- * providing three hardware watchdogs:
- *
- * 	1) RIC - sends an interrupt when triggered
- * 	2) XIR - asserts XIR_B_RESET when triggered, resets CPU
- * 	3) POR - asserts POR_B_RESET when triggered, resets CPU, backplane, board
- *
- *** Timer register block definition (struct wd_timer_regblk)
- *
- * dcntr and limit registers (halfword access):      
- * -------------------
- * | 15 | ...| 1 | 0 |
- * -------------------
- * |-  counter val  -|
- * -------------------
- * dcntr - 	Current 16-bit downcounter value.
- * 			When downcounter reaches '0' watchdog expires.
- * 			Reading this register resets downcounter with 'limit' value.
- * limit - 	16-bit countdown value in 1/10th second increments.
- * 			Writing this register begins countdown with input value.
- * 			Reading from this register does not affect counter.
- * NOTES:	After watchdog reset, dcntr and limit contain '1'
- *
- * status register (byte access):
- * ---------------------------
- * | 7 | ... | 2 |  1  |  0  |
- * --------------+------------
- * |-   UNUSED  -| EXP | RUN |
- * ---------------------------
- * status-	Bit 0 - Watchdog is running
- * 			Bit 1 - Watchdog has expired
- *
- *** PLD register block definition (struct wd_pld_regblk)
- *
- * intr_mask register (byte access):
- * ---------------------------------
- * | 7 | ... | 3 |  2  |  1  |  0  |
- * +-------------+------------------
- * |-   UNUSED  -| WD3 | WD2 | WD1 |
- * ---------------------------------
- * WD3 -  1 == Interrupt disabled for watchdog 3
- * WD2 -  1 == Interrupt disabled for watchdog 2
- * WD1 -  1 == Interrupt disabled for watchdog 1
- *
- * pld_status register (byte access):
- * UNKNOWN, MAGICAL MYSTERY REGISTER
- *
- */
-#define WD_TIMER_REGSZ	16
-#define WD0_OFF		0
-#define WD1_OFF		(WD_TIMER_REGSZ * 1)
-#define WD2_OFF		(WD_TIMER_REGSZ * 2)
-#define PLD_OFF		(WD_TIMER_REGSZ * 3)
-
-#define WD_DCNTR	0x00
-#define WD_LIMIT	0x04
-#define WD_STATUS	0x08
-
-#define PLD_IMASK	(PLD_OFF + 0x00)
-#define PLD_STATUS	(PLD_OFF + 0x04)
-
-/* Individual timer structure 
- */
-struct wd_timer {
-	__u16			timeout;
-	__u8			intr_mask;
-	unsigned char		runstatus;
-	void __iomem		*regs;
-};
-
-/* Device structure
- */
-struct wd_device {
-	int				irq;
-	spinlock_t		lock;
-	unsigned char	isbaddoggie;	/* defective PLD */
-	unsigned char	opt_enable;
-	unsigned char	opt_reboot;
-	unsigned short	opt_timeout;
-	unsigned char	initialized;
-	struct wd_timer	watchdog[WD_NUMDEVS];
-	void __iomem	*regs;
-};
-
-static struct wd_device wd_dev = { 
-		0, __SPIN_LOCK_UNLOCKED(wd_dev.lock), 0, 0, 0, 0,
-};
-
-static struct timer_list wd_timer;
-
-static int wd0_timeout = 0;
-static int wd1_timeout = 0;
-static int wd2_timeout = 0;
-
-#ifdef MODULE
-module_param	(wd0_timeout, int, 0);
-MODULE_PARM_DESC(wd0_timeout, "Default watchdog0 timeout in 1/10secs");
-module_param 	(wd1_timeout, int, 0);
-MODULE_PARM_DESC(wd1_timeout, "Default watchdog1 timeout in 1/10secs");
-module_param 	(wd2_timeout, int, 0);
-MODULE_PARM_DESC(wd2_timeout, "Default watchdog2 timeout in 1/10secs");
-
-MODULE_AUTHOR
-	("Eric Brower <ebrower@usa.net>");
-MODULE_DESCRIPTION
-	("Hardware watchdog driver for Sun Microsystems CP1400/1500");
-MODULE_LICENSE("GPL");
-MODULE_SUPPORTED_DEVICE
-	("watchdog");
-#endif /* ifdef MODULE */
-
-/* Forward declarations of internal methods
- */
-#ifdef WD_DEBUG
-static void wd_dumpregs(void);
-#endif
-static irqreturn_t wd_interrupt(int irq, void *dev_id);
-static void wd_toggleintr(struct wd_timer* pTimer, int enable);
-static void wd_pingtimer(struct wd_timer* pTimer);
-static void wd_starttimer(struct wd_timer* pTimer);
-static void wd_resetbrokentimer(struct wd_timer* pTimer);
-static void wd_stoptimer(struct wd_timer* pTimer);
-static void wd_brokentimer(unsigned long data);
-static int  wd_getstatus(struct wd_timer* pTimer);
-
-/* PLD expects words to be written in LSB format,
- * so we must flip all words prior to writing them to regs
- */
-static inline unsigned short flip_word(unsigned short word)
-{
-	return ((word & 0xff) << 8) | ((word >> 8) & 0xff);
-}
-
-#define wd_writew(val, addr) 	(writew(flip_word(val), addr))
-#define wd_readw(addr) 			(flip_word(readw(addr)))
-#define wd_writeb(val, addr) 	(writeb(val, addr))
-#define wd_readb(addr) 			(readb(addr))
-
-
-/* CP1400s seem to have broken PLD implementations--
- * the interrupt_mask register cannot be written, so
- * no timer interrupts can be masked within the PLD.
- */
-static inline int wd_isbroken(void)
-{
-	/* we could test this by read/write/read/restore
-	 * on the interrupt mask register only if OBP
-	 * 'watchdog-enable?' == FALSE, but it seems 
-	 * ubiquitous on CP1400s
-	 */
-	char val[32];
-	prom_getproperty(prom_root_node, "model", val, sizeof(val));
-	return((!strcmp(val, WD_BADMODEL)) ? 1 : 0);
-}
-		
-/* Retrieve watchdog-enable? option from OBP
- * Returns 0 if false, 1 if true
- */
-static inline int wd_opt_enable(void)
-{
-	int opt_node;
-
-	opt_node = prom_getchild(prom_root_node);
-	opt_node = prom_searchsiblings(opt_node, "options");
-	return((-1 == prom_getint(opt_node, "watchdog-enable?")) ? 0 : 1);
-}
-
-/* Retrieve watchdog-reboot? option from OBP
- * Returns 0 if false, 1 if true
- */
-static inline int wd_opt_reboot(void)
-{
-	int opt_node;
-
-	opt_node = prom_getchild(prom_root_node);
-	opt_node = prom_searchsiblings(opt_node, "options");
-	return((-1 == prom_getint(opt_node, "watchdog-reboot?")) ? 0 : 1);
-}
-
-/* Retrieve watchdog-timeout option from OBP
- * Returns OBP value, or 0 if not located
- */
-static inline int wd_opt_timeout(void)
-{
-	int opt_node;
-	char value[32];
-	char *p = value;
-
-	opt_node = prom_getchild(prom_root_node);
-	opt_node = prom_searchsiblings(opt_node, "options");
-	opt_node = prom_getproperty(opt_node, 
-								"watchdog-timeout", 
-								value, 
-								sizeof(value));
-	if(-1 != opt_node) {
-		/* atoi implementation */
-		for(opt_node = 0; /* nop */; p++) {
-			if(*p >= '0' && *p <= '9') {
-				opt_node = (10*opt_node)+(*p-'0');
-			}
-			else {
-				break;
-			}
-		}
-	}
-	return((-1 == opt_node) ? (0) : (opt_node)); 
-}
-
-static int wd_open(struct inode *inode, struct file *f)
-{
-	lock_kernel();
-	switch(iminor(inode))
-	{
-		case WD0_MINOR:
-			f->private_data = &wd_dev.watchdog[WD0_ID];
-			break;
-		case WD1_MINOR:
-			f->private_data = &wd_dev.watchdog[WD1_ID];
-			break;
-		case WD2_MINOR:
-			f->private_data = &wd_dev.watchdog[WD2_ID];
-			break;
-		default:
-			unlock_kernel();
-			return(-ENODEV);
-	}
-
-	/* Register IRQ on first open of device */
-	if(0 == wd_dev.initialized)
-	{	
-		if (request_irq(wd_dev.irq, 
-						&wd_interrupt, 
-						IRQF_SHARED,
-						WD_OBPNAME,
-						(void *)wd_dev.regs)) {
-			printk("%s: Cannot register IRQ %d\n", 
-				WD_OBPNAME, wd_dev.irq);
-			unlock_kernel();
-			return(-EBUSY);
-		}
-		wd_dev.initialized = 1;
-	}
-
-	unlock_kernel();
-	return(nonseekable_open(inode, f));
-}
-
-static int wd_release(struct inode *inode, struct file *file)
-{
-	return 0;
-}
-
-static int wd_ioctl(struct inode *inode, struct file *file, 
-		     unsigned int cmd, unsigned long arg)
-{
-	int 	setopt 				= 0;
-	struct 	wd_timer* pTimer 	= (struct wd_timer*)file->private_data;
-	void __user *argp = (void __user *)arg;
-	struct 	watchdog_info info 	= {
-		0,
-		0,
-		"Altera EPF8820ATC144-4"
-	};
-
-	if(NULL == pTimer) {
-		return(-EINVAL);
-	}
-
-	switch(cmd)
-	{
-		/* Generic Linux IOCTLs */
-		case WDIOC_GETSUPPORT:
-			if(copy_to_user(argp, &info, sizeof(struct watchdog_info))) {
-				return(-EFAULT);
-			}
-			break;
-		case WDIOC_GETSTATUS:
-		case WDIOC_GETBOOTSTATUS:
-			if (put_user(0, (int __user *)argp))
-				return -EFAULT;
-			break;
-		case WDIOC_KEEPALIVE:
-			wd_pingtimer(pTimer);
-			break;
-		case WDIOC_SETOPTIONS:
-			if(copy_from_user(&setopt, argp, sizeof(unsigned int))) {
-				return -EFAULT;
-			}
-			if(setopt & WDIOS_DISABLECARD) {
-				if(wd_dev.opt_enable) {
-					printk(
-						"%s: cannot disable watchdog in ENABLED mode\n",
-						WD_OBPNAME);
-					return(-EINVAL);
-				}
-				wd_stoptimer(pTimer);
-			}
-			else if(setopt & WDIOS_ENABLECARD) {
-				wd_starttimer(pTimer);
-			}
-			else {
-				return(-EINVAL);
-			}	
-			break;
-		/* Solaris-compatible IOCTLs */
-		case WIOCGSTAT:
-			setopt = wd_getstatus(pTimer);
-			if(copy_to_user(argp, &setopt, sizeof(unsigned int))) {
-				return(-EFAULT);
-			}
-			break;
-		case WIOCSTART:
-			wd_starttimer(pTimer);
-			break;
-		case WIOCSTOP:
-			if(wd_dev.opt_enable) {
-				printk("%s: cannot disable watchdog in ENABLED mode\n",
-					WD_OBPNAME);
-				return(-EINVAL);
-			}
-			wd_stoptimer(pTimer);
-			break;
-		default:
-			return(-EINVAL);
-	}
-	return(0);
-}
-
-static long wd_compat_ioctl(struct file *file, unsigned int cmd,
-		unsigned long arg)
-{
-	int rval = -ENOIOCTLCMD;
-
-	switch (cmd) {
-	/* solaris ioctls are specific to this driver */
-	case WIOCSTART:
-	case WIOCSTOP:
-	case WIOCGSTAT:
-		lock_kernel();
-		rval = wd_ioctl(file->f_path.dentry->d_inode, file, cmd, arg);
-		unlock_kernel();
-		break;
-	/* everything else is handled by the generic compat layer */
-	default:
-		break;
-	}
-
-	return rval;
-}
-
-static ssize_t wd_write(struct file 	*file, 
-			const char	__user *buf, 
-			size_t 		count, 
-			loff_t 		*ppos)
-{
-	struct wd_timer* pTimer = (struct wd_timer*)file->private_data;
-
-	if(NULL == pTimer) {
-		return(-EINVAL);
-	}
-
-	if (count) {
-		wd_pingtimer(pTimer);
-		return 1;
-	}
-	return 0;
-}
-
-static ssize_t wd_read(struct file * file, char __user *buffer,
-		        size_t count, loff_t *ppos)
-{
-#ifdef WD_DEBUG
-	wd_dumpregs();
-	return(0);
-#else
-	return(-EINVAL);
-#endif /* ifdef WD_DEBUG */
-}
-
-static irqreturn_t wd_interrupt(int irq, void *dev_id)
-{
-	/* Only WD0 will interrupt-- others are NMI and we won't
-	 * see them here....
-	 */
-	spin_lock_irq(&wd_dev.lock);
-	if((unsigned long)wd_dev.regs == (unsigned long)dev_id)
-	{
-		wd_stoptimer(&wd_dev.watchdog[WD0_ID]);
-		wd_dev.watchdog[WD0_ID].runstatus |=  WD_STAT_SVCD;
-	}
-	spin_unlock_irq(&wd_dev.lock);
-	return IRQ_HANDLED;
-}
-
-static const struct file_operations wd_fops = {
-	.owner =	THIS_MODULE,
-	.ioctl =	wd_ioctl,
-	.compat_ioctl =	wd_compat_ioctl,
-	.open =		wd_open,
-	.write =	wd_write,
-	.read =		wd_read,
-	.release =	wd_release,
-};
-
-static struct miscdevice wd0_miscdev = { WD0_MINOR, WD0_DEVNAME, &wd_fops };
-static struct miscdevice wd1_miscdev = { WD1_MINOR, WD1_DEVNAME, &wd_fops };
-static struct miscdevice wd2_miscdev = { WD2_MINOR, WD2_DEVNAME, &wd_fops };
-
-#ifdef WD_DEBUG
-static void wd_dumpregs(void)
-{
-	/* Reading from downcounters initiates watchdog countdown--
-	 * Example is included below for illustration purposes.
-	 */
-	int i;
-	printk("%s: dumping register values\n", WD_OBPNAME);
-	for(i = WD0_ID; i < WD_NUMDEVS; ++i) {
-			/* printk("\t%s%i: dcntr  at 0x%lx: 0x%x\n", 
-			 * 	WD_OBPNAME,
-		 	 *	i,
-			 *	(unsigned long)(&wd_dev.watchdog[i].regs->dcntr), 
-			 *	readw(&wd_dev.watchdog[i].regs->dcntr));
-			 */
-			printk("\t%s%i: limit  at 0x%lx: 0x%x\n", 
-				WD_OBPNAME,
-				i,
-				(unsigned long)(&wd_dev.watchdog[i].regs->limit), 
-				readw(&wd_dev.watchdog[i].regs->limit));
-			printk("\t%s%i: status at 0x%lx: 0x%x\n", 
-				WD_OBPNAME,
-				i,
-				(unsigned long)(&wd_dev.watchdog[i].regs->status), 
-				readb(&wd_dev.watchdog[i].regs->status));
-			printk("\t%s%i: driver status: 0x%x\n",
-				WD_OBPNAME,
-				i,
-				wd_getstatus(&wd_dev.watchdog[i]));
-	}
-	printk("\tintr_mask  at %p: 0x%x\n", 
-		wd_dev.regs + PLD_IMASK,
-		readb(wd_dev.regs + PLD_IMASK));
-	printk("\tpld_status at %p: 0x%x\n", 
-		wd_dev.regs + PLD_STATUS, 
-		readb(wd_dev.regs + PLD_STATUS));
-}
-#endif
-
-/* Enable or disable watchdog interrupts
- * Because of the CP1400 defect this should only be
- * called during initialzation or by wd_[start|stop]timer()
- *
- * pTimer 	- pointer to timer device, or NULL to indicate all timers 
- * enable	- non-zero to enable interrupts, zero to disable
- */
-static void wd_toggleintr(struct wd_timer* pTimer, int enable)
-{
-	unsigned char curregs = wd_readb(wd_dev.regs + PLD_IMASK);
-	unsigned char setregs = 
-		(NULL == pTimer) ? 
-			(WD0_INTR_MASK | WD1_INTR_MASK | WD2_INTR_MASK) : 
-			(pTimer->intr_mask);
-
-	(WD_INTR_ON == enable) ?
-		(curregs &= ~setregs):
-		(curregs |=  setregs);
-
-	wd_writeb(curregs, wd_dev.regs + PLD_IMASK);
-	return;
-}
-
-/* Reset countdown timer with 'limit' value and continue countdown.
- * This will not start a stopped timer.
- *
- * pTimer	- pointer to timer device
- */
-static void wd_pingtimer(struct wd_timer* pTimer)
-{
-	if (wd_readb(pTimer->regs + WD_STATUS) & WD_S_RUNNING) {
-		wd_readw(pTimer->regs + WD_DCNTR);
-	}
-}
-
-/* Stop a running watchdog timer-- the timer actually keeps
- * running, but the interrupt is masked so that no action is
- * taken upon expiration.
- *
- * pTimer	- pointer to timer device
- */
-static void wd_stoptimer(struct wd_timer* pTimer)
-{
-	if(wd_readb(pTimer->regs + WD_STATUS) & WD_S_RUNNING) {
-		wd_toggleintr(pTimer, WD_INTR_OFF);
-
-		if(wd_dev.isbaddoggie) {
-			pTimer->runstatus |= WD_STAT_BSTOP;
-			wd_brokentimer((unsigned long)&wd_dev);
-		}
-	}
-}
-
-/* Start a watchdog timer with the specified limit value
- * If the watchdog is running, it will be restarted with
- * the provided limit value.
- *
- * This function will enable interrupts on the specified
- * watchdog.
- *
- * pTimer	- pointer to timer device
- * limit	- limit (countdown) value in 1/10th seconds
- */
-static void wd_starttimer(struct wd_timer* pTimer)
-{
-	if(wd_dev.isbaddoggie) {
-		pTimer->runstatus &= ~WD_STAT_BSTOP;
-	}
-	pTimer->runstatus &= ~WD_STAT_SVCD;
-
-	wd_writew(pTimer->timeout, pTimer->regs + WD_LIMIT);
-	wd_toggleintr(pTimer, WD_INTR_ON);
-}
-
-/* Restarts timer with maximum limit value and
- * does not unset 'brokenstop' value.
- */
-static void wd_resetbrokentimer(struct wd_timer* pTimer)
-{
-	wd_toggleintr(pTimer, WD_INTR_ON);
-	wd_writew(WD_BLIMIT, pTimer->regs + WD_LIMIT);
-}
-
-/* Timer device initialization helper.
- * Returns 0 on success, other on failure
- */
-static int wd_inittimer(int whichdog)
-{
-	struct miscdevice 				*whichmisc;
-	void __iomem *whichregs;
-	char 							whichident[8];
-	int								whichmask;
-	__u16							whichlimit;
-
-	switch(whichdog)
-	{
-		case WD0_ID:
-			whichmisc = &wd0_miscdev;
-			strcpy(whichident, "RIC");
-			whichregs = wd_dev.regs + WD0_OFF;
-			whichmask = WD0_INTR_MASK;
-			whichlimit= (0 == wd0_timeout) 	? 
-						(wd_dev.opt_timeout): 
-						(wd0_timeout);
-			break;
-		case WD1_ID:
-			whichmisc = &wd1_miscdev;
-			strcpy(whichident, "XIR");
-			whichregs = wd_dev.regs + WD1_OFF;
-			whichmask = WD1_INTR_MASK;
-			whichlimit= (0 == wd1_timeout) 	? 
-						(wd_dev.opt_timeout): 
-						(wd1_timeout);
-			break;
-		case WD2_ID:
-			whichmisc = &wd2_miscdev;
-			strcpy(whichident, "POR");
-			whichregs = wd_dev.regs + WD2_OFF;
-			whichmask = WD2_INTR_MASK;
-			whichlimit= (0 == wd2_timeout) 	? 
-						(wd_dev.opt_timeout): 
-						(wd2_timeout);
-			break;
-		default:
-			printk("%s: %s: invalid watchdog id: %i\n",
-				WD_OBPNAME, __func__, whichdog);
-			return(1);
-	}
-	if(0 != misc_register(whichmisc))
-	{
-		return(1);
-	}
-	wd_dev.watchdog[whichdog].regs			= whichregs;
-	wd_dev.watchdog[whichdog].timeout 		= whichlimit;
-	wd_dev.watchdog[whichdog].intr_mask		= whichmask;
-	wd_dev.watchdog[whichdog].runstatus 	&= ~WD_STAT_BSTOP;
-	wd_dev.watchdog[whichdog].runstatus 	|= WD_STAT_INIT;
-
-	printk("%s%i: %s hardware watchdog [%01i.%i sec] %s\n", 
-		WD_OBPNAME, 
-		whichdog, 
-		whichident, 
-		wd_dev.watchdog[whichdog].timeout / 10,
-		wd_dev.watchdog[whichdog].timeout % 10,
-		(0 != wd_dev.opt_enable) ? "in ENABLED mode" : "");
-	return(0);
-}
-
-/* Timer method called to reset stopped watchdogs--
- * because of the PLD bug on CP1400, we cannot mask
- * interrupts within the PLD so me must continually
- * reset the timers ad infinitum.
- */
-static void wd_brokentimer(unsigned long data)
-{
-	struct wd_device* pDev = (struct wd_device*)data;
-	int id, tripped = 0;
-
-	/* kill a running timer instance, in case we
-	 * were called directly instead of by kernel timer
-	 */
-	if(timer_pending(&wd_timer)) {
-		del_timer(&wd_timer);
-	}
-
-	for(id = WD0_ID; id < WD_NUMDEVS; ++id) {
-		if(pDev->watchdog[id].runstatus & WD_STAT_BSTOP) {
-			++tripped;
-			wd_resetbrokentimer(&pDev->watchdog[id]);
-		}
-	}
-
-	if(tripped) {
-		/* there is at least one timer brokenstopped-- reschedule */
-		init_timer(&wd_timer);
-		wd_timer.expires = WD_BTIMEOUT;
-		add_timer(&wd_timer);
-	}
-}
-
-static int wd_getstatus(struct wd_timer* pTimer)
-{
-	unsigned char stat = wd_readb(pTimer->regs + WD_STATUS);
-	unsigned char intr = wd_readb(wd_dev.regs + PLD_IMASK);
-	unsigned char ret  = WD_STOPPED;
-
-	/* determine STOPPED */
-	if(0 == stat ) { 
-		return(ret);
-	}
-	/* determine EXPIRED vs FREERUN vs RUNNING */
-	else if(WD_S_EXPIRED & stat) {
-		ret = WD_EXPIRED;
-	}
-	else if(WD_S_RUNNING & stat) {
-		if(intr & pTimer->intr_mask) {
-			ret = WD_FREERUN;
-		}
-		else {
-			/* Fudge WD_EXPIRED status for defective CP1400--
-			 * IF timer is running 
-			 * 	AND brokenstop is set 
-			 * 	AND an interrupt has been serviced
-			 * we are WD_EXPIRED.
-			 *
-			 * IF timer is running 
-			 * 	AND brokenstop is set 
-			 * 	AND no interrupt has been serviced
-			 * we are WD_FREERUN.
-			 */
-			if(wd_dev.isbaddoggie && (pTimer->runstatus & WD_STAT_BSTOP)) {
-				if(pTimer->runstatus & WD_STAT_SVCD) {
-					ret = WD_EXPIRED;
-				}
-				else {
-					/* we could as well pretend we are expired */
-					ret = WD_FREERUN;
-				}
-			}
-			else {
-				ret = WD_RUNNING;
-			}
-		}
-	}
-
-	/* determine SERVICED */
-	if(pTimer->runstatus & WD_STAT_SVCD) {
-		ret |= WD_SERVICED;
-	}
-
-	return(ret);
-}
-
-static int __init wd_init(void)
-{
-	int 	id;
-	struct 	linux_ebus *ebus = NULL;
-	struct 	linux_ebus_device *edev = NULL;
-
-	for_each_ebus(ebus) {
-		for_each_ebusdev(edev, ebus) {
-			if (!strcmp(edev->ofdev.node->name, WD_OBPNAME))
-				goto ebus_done;
-		}
-	}
-
-ebus_done:
-	if(!edev) {
-		printk("%s: unable to locate device\n", WD_OBPNAME);
-		return -ENODEV;
-	}
-
-	wd_dev.regs = 
-		ioremap(edev->resource[0].start, 4 * WD_TIMER_REGSZ); /* ? */
-
-	if(NULL == wd_dev.regs) {
-		printk("%s: unable to map registers\n", WD_OBPNAME);
-		return(-ENODEV);
-	}
-
-	/* initialize device structure from OBP parameters */
-	wd_dev.irq 			= edev->irqs[0];
-	wd_dev.opt_enable	= wd_opt_enable();
-	wd_dev.opt_reboot	= wd_opt_reboot();
-	wd_dev.opt_timeout	= wd_opt_timeout();
-	wd_dev.isbaddoggie	= wd_isbroken();
-
-	/* disable all interrupts unless watchdog-enabled? == true */
-	if(! wd_dev.opt_enable) {
-		wd_toggleintr(NULL, WD_INTR_OFF);
-	}
-
-	/* register miscellaneous devices */
-	for(id = WD0_ID; id < WD_NUMDEVS; ++id) {
-		if(0 != wd_inittimer(id)) {
-			printk("%s%i: unable to initialize\n", WD_OBPNAME, id);
-		}
-	}
-
-	/* warn about possible defective PLD */
-	if(wd_dev.isbaddoggie) {
-		init_timer(&wd_timer);
-		wd_timer.function 	= wd_brokentimer;
-		wd_timer.data		= (unsigned long)&wd_dev;
-		wd_timer.expires	= WD_BTIMEOUT;
-
-		printk("%s: PLD defect workaround enabled for model %s\n",
-			WD_OBPNAME, WD_BADMODEL);
-	}
-	return(0);
-}
-
-static void __exit wd_cleanup(void)
-{
-	int id;
-
-	/* if 'watchdog-enable?' == TRUE, timers are not stopped 
-	 * when module is unloaded.  All brokenstopped timers will
-	 * also now eventually trip. 
-	 */
-	for(id = WD0_ID; id < WD_NUMDEVS; ++id) {
-		if(WD_S_RUNNING == wd_readb(wd_dev.watchdog[id].regs + WD_STATUS)) {
-			if(wd_dev.opt_enable) {
-				printk(KERN_WARNING "%s%i: timer not stopped at release\n",
-					WD_OBPNAME, id);
-			}
-			else {
-				wd_stoptimer(&wd_dev.watchdog[id]);
-				if(wd_dev.watchdog[id].runstatus & WD_STAT_BSTOP) {
-					wd_resetbrokentimer(&wd_dev.watchdog[id]);
-					printk(KERN_WARNING 
-							"%s%i: defect workaround disabled at release, "\
-							"timer expires in ~%01i sec\n",
-							WD_OBPNAME, id, 
-							wd_readw(wd_dev.watchdog[id].regs + WD_LIMIT) / 10);
-				}
-			}
-		}
-	}
-
-	if(wd_dev.isbaddoggie && timer_pending(&wd_timer)) {
-		del_timer(&wd_timer);
-	}
-	if(0 != (wd_dev.watchdog[WD0_ID].runstatus & WD_STAT_INIT)) {
-		misc_deregister(&wd0_miscdev);
-	}
-	if(0 != (wd_dev.watchdog[WD1_ID].runstatus & WD_STAT_INIT)) {
-		misc_deregister(&wd1_miscdev);
-	}
-	if(0 != (wd_dev.watchdog[WD2_ID].runstatus & WD_STAT_INIT)) {
-		misc_deregister(&wd2_miscdev);
-	}
-	if(0 != wd_dev.initialized) {
-		free_irq(wd_dev.irq, (void *)wd_dev.regs);
-	}
-	iounmap(wd_dev.regs);
-}
-
-module_init(wd_init);
-module_exit(wd_cleanup);
diff --git a/drivers/sbus/char/display7seg.c b/drivers/sbus/char/display7seg.c
index d8f5c0ca236d..2550af4ae432 100644
--- a/drivers/sbus/char/display7seg.c
+++ b/drivers/sbus/char/display7seg.c
@@ -1,10 +1,7 @@
-/* $Id: display7seg.c,v 1.6 2002/01/08 16:00:16 davem Exp $
- *
- * display7seg - Driver implementation for the 7-segment display
- * present on Sun Microsystems CP1400 and CP1500
+/* display7seg.c - Driver implementation for the 7-segment display
+ *                 present on Sun Microsystems CP1400 and CP1500
  *
  * Copyright (c) 2000 Eric Brower (ebrower@usa.net)
- *
  */
 
 #include <linux/kernel.h>
@@ -16,22 +13,20 @@
 #include <linux/miscdevice.h>
 #include <linux/ioport.h>		/* request_region */
 #include <linux/smp_lock.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 #include <asm/atomic.h>
-#include <asm/ebus.h>			/* EBus device					*/
-#include <asm/oplib.h>			/* OpenProm Library 			*/
 #include <asm/uaccess.h>		/* put_/get_user			*/
 #include <asm/io.h>
 
 #include <asm/display7seg.h>
 
 #define D7S_MINOR	193
-#define D7S_OBPNAME	"display7seg"
-#define D7S_DEVNAME "d7s"
+#define DRIVER_NAME	"d7s"
+#define PFX		DRIVER_NAME ": "
 
 static int sol_compat = 0;		/* Solaris compatibility mode	*/
 
-#ifdef MODULE
-
 /* Solaris compatibility flag -
  * The Solaris implementation omits support for several
  * documented driver features (ref Sun doc 806-0180-03).  
@@ -46,20 +41,20 @@ static int sol_compat = 0;		/* Solaris compatibility mode	*/
  * If you wish the device to operate as under Solaris,
  * omitting above features, set this parameter to non-zero.
  */
-module_param
-	(sol_compat, int, 0);
-MODULE_PARM_DESC
-	(sol_compat, 
-	 "Disables documented functionality omitted from Solaris driver");
-
-MODULE_AUTHOR
-	("Eric Brower <ebrower@usa.net>");
-MODULE_DESCRIPTION
-	("7-Segment Display driver for Sun Microsystems CP1400/1500");
+module_param(sol_compat, int, 0);
+MODULE_PARM_DESC(sol_compat, 
+		 "Disables documented functionality omitted from Solaris driver");
+
+MODULE_AUTHOR("Eric Brower <ebrower@usa.net>");
+MODULE_DESCRIPTION("7-Segment Display driver for Sun Microsystems CP1400/1500");
 MODULE_LICENSE("GPL");
-MODULE_SUPPORTED_DEVICE
-	("d7s");
-#endif /* ifdef MODULE */
+MODULE_SUPPORTED_DEVICE("d7s");
+
+struct d7s {
+	void __iomem	*regs;
+	bool		flipped;
+};
+struct d7s *d7s_device;
 
 /*
  * Register block address- see header for details
@@ -72,22 +67,6 @@ MODULE_SUPPORTED_DEVICE
  * FLIP		- Inverts display for upside-down mounted board
  * bits 0-4	- 7-segment display contents
  */
-static void __iomem* d7s_regs;
-
-static inline void d7s_free(void)
-{
-	iounmap(d7s_regs);
-}
-
-static inline int d7s_obpflipped(void)
-{
-	int opt_node;
-
-	opt_node = prom_getchild(prom_root_node);
-	opt_node = prom_searchsiblings(opt_node, "options");
-	return ((-1 != prom_getintdefault(opt_node, "d7s-flipped?", -1)) ? 0 : 1);
-}
-
 static atomic_t d7s_users = ATOMIC_INIT(0);
 
 static int d7s_open(struct inode *inode, struct file *f)
@@ -106,12 +85,15 @@ static int d7s_release(struct inode *inode, struct file *f)
 	 * are not operating in solaris-compat mode
 	 */
 	if (atomic_dec_and_test(&d7s_users) && !sol_compat) {
-		int regval = 0;
-
-		regval = readb(d7s_regs);
-		(0 == d7s_obpflipped())	? 
-			writeb(regval |= D7S_FLIP,  d7s_regs): 
-			writeb(regval &= ~D7S_FLIP, d7s_regs);
+		struct d7s *p = d7s_device;
+		u8 regval = 0;
+
+		regval = readb(p->regs);
+		if (p->flipped)
+			regval |= D7S_FLIP;
+		else
+			regval &= ~D7S_FLIP;
+		writeb(regval, p->regs);
 	}
 
 	return 0;
@@ -119,9 +101,10 @@ static int d7s_release(struct inode *inode, struct file *f)
 
 static long d7s_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
-	__u8 regs = readb(d7s_regs);
-	__u8 ireg = 0;
+	struct d7s *p = d7s_device;
+	u8 regs = readb(p->regs);
 	int error = 0;
+	u8 ireg = 0;
 
 	if (D7S_MINOR != iminor(file->f_path.dentry->d_inode))
 		return -ENODEV;
@@ -129,18 +112,20 @@ static long d7s_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	lock_kernel();
 	switch (cmd) {
 	case D7SIOCWR:
-		/* assign device register values
-		 * we mask-out D7S_FLIP if in sol_compat mode
+		/* assign device register values we mask-out D7S_FLIP
+		 * if in sol_compat mode
 		 */
 		if (get_user(ireg, (int __user *) arg)) {
 			error = -EFAULT;
 			break;
 		}
-		if (0 != sol_compat) {
-			(regs & D7S_FLIP) ? 
-				(ireg |= D7S_FLIP) : (ireg &= ~D7S_FLIP);
+		if (sol_compat) {
+			if (regs & D7S_FLIP)
+				ireg |= D7S_FLIP;
+			else
+				ireg &= ~D7S_FLIP;
 		}
-		writeb(ireg, d7s_regs);
+		writeb(ireg, p->regs);
 		break;
 
 	case D7SIOCRD:
@@ -158,9 +143,11 @@ static long d7s_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 
 	case D7SIOCTM:
 		/* toggle device mode-- flip display orientation */
-		(regs & D7S_FLIP) ? 
-			(regs &= ~D7S_FLIP) : (regs |= D7S_FLIP);
-		writeb(regs, d7s_regs);
+		if (regs & D7S_FLIP)
+			regs &= ~D7S_FLIP;
+		else
+			regs |= D7S_FLIP;
+		writeb(regs, p->regs);
 		break;
 	};
 	unlock_kernel();
@@ -176,69 +163,123 @@ static const struct file_operations d7s_fops = {
 	.release =		d7s_release,
 };
 
-static struct miscdevice d7s_miscdev = { D7S_MINOR, D7S_DEVNAME, &d7s_fops };
+static struct miscdevice d7s_miscdev = {
+	.minor		= D7S_MINOR,
+	.name		= DRIVER_NAME,
+	.fops		= &d7s_fops
+};
 
-static int __init d7s_init(void)
+static int __devinit d7s_probe(struct of_device *op,
+			       const struct of_device_id *match)
 {
-	struct linux_ebus *ebus = NULL;
-	struct linux_ebus_device *edev = NULL;
-	int iTmp = 0, regs = 0;
-
-	for_each_ebus(ebus) {
-		for_each_ebusdev(edev, ebus) {
-			if (!strcmp(edev->prom_node->name, D7S_OBPNAME))
-				goto ebus_done;
-		}
+	struct device_node *opts;
+	int err = -EINVAL;
+	struct d7s *p;
+	u8 regs;
+
+	if (d7s_device)
+		goto out;
+
+	p = kzalloc(sizeof(*p), GFP_KERNEL);
+	err = -ENOMEM;
+	if (!p)
+		goto out;
+
+	p->regs = of_ioremap(&op->resource[0], 0, sizeof(u8), "d7s");
+	if (!p->regs) {
+		printk(KERN_ERR PFX "Cannot map chip registers\n");
+		goto out_free;
 	}
 
-ebus_done:
-	if(!edev) {
-		printk("%s: unable to locate device\n", D7S_DEVNAME);
-		return -ENODEV;
+	err = misc_register(&d7s_miscdev);
+	if (err) {
+		printk(KERN_ERR PFX "Unable to acquire miscdevice minor %i\n",
+		       D7S_MINOR);
+		goto out_iounmap;
 	}
 
-	d7s_regs = ioremap(edev->resource[0].start, sizeof(__u8));
-
-	iTmp = misc_register(&d7s_miscdev);
-	if (0 != iTmp) {
-		printk("%s: unable to acquire miscdevice minor %i\n",
-		       D7S_DEVNAME, D7S_MINOR);
-		iounmap(d7s_regs);
-		return iTmp;
-	}
-
-	/* OBP option "d7s-flipped?" is honored as default
-	 * for the device, and reset default when detached
+	/* OBP option "d7s-flipped?" is honored as default for the
+	 * device, and reset default when detached
 	 */
-	regs = readb(d7s_regs);
-	iTmp = d7s_obpflipped();
-	(0 == iTmp) ? 
-		writeb(regs |= D7S_FLIP,  d7s_regs): 
-		writeb(regs &= ~D7S_FLIP, d7s_regs);
-
-	printk("%s: 7-Segment Display%s at 0x%lx %s\n", 
-	       D7S_DEVNAME,
-	       (0 == iTmp) ? (" (FLIPPED)") : (""),
-	       edev->resource[0].start,
-	       (0 != sol_compat) ? ("in sol_compat mode") : (""));
-
-	return 0;
+	regs = readb(p->regs);
+	opts = of_find_node_by_path("/options");
+	if (opts &&
+	    of_get_property(opts, "d7s-flipped?", NULL))
+		p->flipped = true;
+
+	if (p->flipped)
+		regs |= D7S_FLIP;
+	else
+		regs &= ~D7S_FLIP;
+
+	writeb(regs,  p->regs);
+
+	printk(KERN_INFO PFX "7-Segment Display%s at [%s:0x%lx] %s\n", 
+	       op->node->full_name,
+	       (regs & D7S_FLIP) ? " (FLIPPED)" : "",
+	       op->resource[0].start,
+	       sol_compat ? "in sol_compat mode" : "");
+
+	dev_set_drvdata(&op->dev, p);
+	d7s_device = p;
+	err = 0;
+
+out:
+	return err;
+
+out_iounmap:
+	of_iounmap(&op->resource[0], p->regs, sizeof(u8));
+
+out_free:
+	kfree(p);
+	goto out;
 }
 
-static void __exit d7s_cleanup(void)
+static int __devexit d7s_remove(struct of_device *op)
 {
-	int regs = readb(d7s_regs);
+	struct d7s *p = dev_get_drvdata(&op->dev);
+	u8 regs = readb(p->regs);
 
 	/* Honor OBP d7s-flipped? unless operating in solaris-compat mode */
-	if (0 == sol_compat) {
-		(0 == d7s_obpflipped())	? 
-			writeb(regs |= D7S_FLIP,  d7s_regs):
-			writeb(regs &= ~D7S_FLIP, d7s_regs);
+	if (sol_compat) {
+		if (p->flipped)
+			regs |= D7S_FLIP;
+		else
+			regs &= ~D7S_FLIP;
+		writeb(regs, p->regs);
 	}
 
 	misc_deregister(&d7s_miscdev);
-	d7s_free();
+	of_iounmap(&op->resource[0], p->regs, sizeof(u8));
+	kfree(p);
+
+	return 0;
+}
+
+static const struct of_device_id d7s_match[] = {
+	{
+		.name = "display7seg",
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, d7s_match);
+
+static struct of_platform_driver d7s_driver = {
+	.name		= DRIVER_NAME,
+	.match_table	= d7s_match,
+	.probe		= d7s_probe,
+	.remove		= __devexit_p(d7s_remove),
+};
+
+static int __init d7s_init(void)
+{
+	return of_register_driver(&d7s_driver, &of_bus_type);
+}
+
+static void __exit d7s_exit(void)
+{
+	of_unregister_driver(&d7s_driver);
 }
 
 module_init(d7s_init);
-module_exit(d7s_cleanup);
+module_exit(d7s_exit);
diff --git a/drivers/sbus/char/envctrl.c b/drivers/sbus/char/envctrl.c
index a408402426f8..58e583b61e60 100644
--- a/drivers/sbus/char/envctrl.c
+++ b/drivers/sbus/char/envctrl.c
@@ -1,5 +1,4 @@
-/* $Id: envctrl.c,v 1.25 2002/01/15 09:01:26 davem Exp $
- * envctrl.c: Temperature and Fan monitoring on Machines providing it.
+/* envctrl.c: Temperature and Fan monitoring on Machines providing it.
  *
  * Copyright (C) 1998  Eddie C. Dost  (ecd@skynet.be)
  * Copyright (C) 2000  Vinh Truong    (vinh.truong@eng.sun.com)
@@ -28,12 +27,16 @@
 #include <linux/kmod.h>
 #include <linux/reboot.h>
 #include <linux/smp_lock.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
-#include <asm/ebus.h>
 #include <asm/uaccess.h>
 #include <asm/envctrl.h>
 #include <asm/io.h>
 
+#define DRIVER_NAME	"envctrl"
+#define PFX		DRIVER_NAME ": "
+
 #define ENVCTRL_MINOR	162
 
 #define PCF8584_ADDRESS	0x55
@@ -193,7 +196,7 @@ static void envtrl_i2c_test_pin(void)
 	} 
 
 	if (limit <= 0)
-		printk(KERN_INFO "envctrl: Pin status will not clear.\n");
+		printk(KERN_INFO PFX "Pin status will not clear.\n");
 }
 
 /* Function Description: Test busy bit.
@@ -211,7 +214,7 @@ static void envctrl_i2c_test_bb(void)
 	} 
 
 	if (limit <= 0)
-		printk(KERN_INFO "envctrl: Busy bit will not clear.\n");
+		printk(KERN_INFO PFX "Busy bit will not clear.\n");
 }
 
 /* Function Description: Send the address for a read access.
@@ -858,11 +861,10 @@ static void envctrl_init_voltage_status(struct i2c_child_t *pchild)
 /* Function Description: Initialize i2c child device.
  * Return: None.
  */
-static void envctrl_init_i2c_child(struct linux_ebus_child *edev_child,
+static void envctrl_init_i2c_child(struct device_node *dp,
 				   struct i2c_child_t *pchild)
 {
 	int len, i, tbls_size = 0;
-	struct device_node *dp = edev_child->prom_node;
 	const void *pval;
 
 	/* Get device address. */
@@ -882,12 +884,12 @@ static void envctrl_init_i2c_child(struct linux_ebus_child *edev_child,
 
                 pchild->tables = kmalloc(tbls_size, GFP_KERNEL);
 		if (pchild->tables == NULL){
-			printk("envctrl: Failed to allocate table.\n");
+			printk(KERN_ERR PFX "Failed to allocate table.\n");
 			return;
 		}
 		pval = of_get_property(dp, "tables", &len);
                 if (!pval || len <= 0) {
-			printk("envctrl: Failed to get table.\n");
+			printk(KERN_ERR PFX "Failed to get table.\n");
 			return;
 		}
 		memcpy(pchild->tables, pval, len);
@@ -993,14 +995,14 @@ static int kenvctrld(void *__unused)
 	struct i2c_child_t *cputemp;
 
 	if (NULL == (cputemp = envctrl_get_i2c_child(ENVCTRL_CPUTEMP_MON))) {
-		printk(KERN_ERR 
-		       "envctrl: kenvctrld unable to monitor CPU temp-- exiting\n");
+		printk(KERN_ERR  PFX
+		       "kenvctrld unable to monitor CPU temp-- exiting\n");
 		return -ENODEV;
 	}
 
 	poll_interval = 5000; /* TODO env_mon_interval */
 
-	printk(KERN_INFO "envctrl: %s starting...\n", current->comm);
+	printk(KERN_INFO PFX "%s starting...\n", current->comm);
 	for (;;) {
 		msleep_interruptible(poll_interval);
 
@@ -1022,54 +1024,35 @@ static int kenvctrld(void *__unused)
 			}
 		}
 	}
-	printk(KERN_INFO "envctrl: %s exiting...\n", current->comm);
+	printk(KERN_INFO PFX "%s exiting...\n", current->comm);
 	return 0;
 }
 
-static int __init envctrl_init(void)
+static int __devinit envctrl_probe(struct of_device *op,
+				   const struct of_device_id *match)
 {
-	struct linux_ebus *ebus = NULL;
-	struct linux_ebus_device *edev = NULL;
-	struct linux_ebus_child *edev_child = NULL;
-	int err, i = 0;
-
-	for_each_ebus(ebus) {
-		for_each_ebusdev(edev, ebus) {
-			if (!strcmp(edev->prom_node->name, "bbc")) {
-				/* If we find a boot-bus controller node,
-				 * then this envctrl driver is not for us.
-				 */
-				return -ENODEV;
-			}
-		}
-	}
+	struct device_node *dp;
+	int index, err;
 
-	/* Traverse through ebus and ebus device list for i2c device and
-	 * adc and gpio nodes.
-	 */
-	for_each_ebus(ebus) {
-		for_each_ebusdev(edev, ebus) {
-			if (!strcmp(edev->prom_node->name, "i2c")) {
-				i2c = ioremap(edev->resource[0].start, 0x2);
-				for_each_edevchild(edev, edev_child) {
-					if (!strcmp("gpio", edev_child->prom_node->name)) {
-						i2c_childlist[i].i2ctype = I2C_GPIO;
-						envctrl_init_i2c_child(edev_child, &(i2c_childlist[i++]));
-					}
-					if (!strcmp("adc", edev_child->prom_node->name)) {
-						i2c_childlist[i].i2ctype = I2C_ADC;
-						envctrl_init_i2c_child(edev_child, &(i2c_childlist[i++]));
-					}
-				}
-				goto done;
-			}
+	if (i2c)
+		return -EINVAL;
+
+	i2c = of_ioremap(&op->resource[0], 0, 0x2, DRIVER_NAME);
+	if (!i2c)
+		return -ENOMEM;
+
+	index = 0;
+	dp = op->node->child;
+	while (dp) {
+		if (!strcmp(dp->name, "gpio")) {
+			i2c_childlist[index].i2ctype = I2C_GPIO;
+			envctrl_init_i2c_child(dp, &(i2c_childlist[index++]));
+		} else if (!strcmp(dp->name, "adc")) {
+			i2c_childlist[index].i2ctype = I2C_ADC;
+			envctrl_init_i2c_child(dp, &(i2c_childlist[index++]));
 		}
-	}
 
-done:
-	if (!edev) {
-		printk("envctrl: I2C device not found.\n");
-		return -ENODEV;
+		dp = dp->sibling;
 	}
 
 	/* Set device address. */
@@ -1087,7 +1070,7 @@ done:
 	/* Register the device as a minor miscellaneous device. */
 	err = misc_register(&envctrl_dev);
 	if (err) {
-		printk("envctrl: Unable to get misc minor %d\n",
+		printk(KERN_ERR PFX "Unable to get misc minor %d\n",
 		       envctrl_dev.minor);
 		goto out_iounmap;
 	}
@@ -1096,12 +1079,12 @@ done:
 	 * a next child device, so we decrement before reverse-traversal of
 	 * child devices.
 	 */
-	printk("envctrl: initialized ");
-	for (--i; i >= 0; --i) {
+	printk(KERN_INFO PFX "Initialized ");
+	for (--index; index >= 0; --index) {
 		printk("[%s 0x%lx]%s", 
-			(I2C_ADC == i2c_childlist[i].i2ctype) ? ("adc") : 
-			((I2C_GPIO == i2c_childlist[i].i2ctype) ? ("gpio") : ("unknown")), 
-			i2c_childlist[i].addr, (0 == i) ? ("\n") : (" "));
+			(I2C_ADC == i2c_childlist[index].i2ctype) ? "adc" : 
+			((I2C_GPIO == i2c_childlist[index].i2ctype) ? "gpio" : "unknown"), 
+			i2c_childlist[index].addr, (0 == index) ? "\n" : " ");
 	}
 
 	kenvctrld_task = kthread_run(kenvctrld, NULL, "kenvctrld");
@@ -1115,26 +1098,54 @@ done:
 out_deregister:
 	misc_deregister(&envctrl_dev);
 out_iounmap:
-	iounmap(i2c);
-	for (i = 0; i < ENVCTRL_MAX_CPU * 2; i++)
-		kfree(i2c_childlist[i].tables);
+	of_iounmap(&op->resource[0], i2c, 0x2);
+	for (index = 0; index < ENVCTRL_MAX_CPU * 2; index++)
+		kfree(i2c_childlist[index].tables);
 
 	return err;
 }
 
-static void __exit envctrl_cleanup(void)
+static int __devexit envctrl_remove(struct of_device *op)
 {
-	int i;
+	int index;
 
 	kthread_stop(kenvctrld_task);
 
-	iounmap(i2c);
+	of_iounmap(&op->resource[0], i2c, 0x2);
 	misc_deregister(&envctrl_dev);
 
-	for (i = 0; i < ENVCTRL_MAX_CPU * 2; i++)
-		kfree(i2c_childlist[i].tables);
+	for (index = 0; index < ENVCTRL_MAX_CPU * 2; index++)
+		kfree(i2c_childlist[index].tables);
+
+	return 0;
+}
+
+static const struct of_device_id envctrl_match[] = {
+	{
+		.name = "i2c",
+		.compatible = "i2cpcf,8584",
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, envctrl_match);
+
+static struct of_platform_driver envctrl_driver = {
+	.name		= DRIVER_NAME,
+	.match_table	= envctrl_match,
+	.probe		= envctrl_probe,
+	.remove		= __devexit_p(envctrl_remove),
+};
+
+static int __init envctrl_init(void)
+{
+	return of_register_driver(&envctrl_driver, &of_bus_type);
+}
+
+static void __exit envctrl_exit(void)
+{
+	of_unregister_driver(&envctrl_driver);
 }
 
 module_init(envctrl_init);
-module_exit(envctrl_cleanup);
+module_exit(envctrl_exit);
 MODULE_LICENSE("GPL");
diff --git a/drivers/sbus/char/flash.c b/drivers/sbus/char/flash.c
index 7d95e151513a..41083472ff4f 100644
--- a/drivers/sbus/char/flash.c
+++ b/drivers/sbus/char/flash.c
@@ -1,5 +1,4 @@
-/* $Id: flash.c,v 1.25 2001/12/21 04:56:16 davem Exp $
- * flash.c: Allow mmap access to the OBP Flash, for OBP updates.
+/* flash.c: Allow mmap access to the OBP Flash, for OBP updates.
  *
  * Copyright (C) 1997  Eddie C. Dost  (ecd@skynet.be)
  */
@@ -15,13 +14,13 @@
 #include <linux/smp_lock.h>
 #include <linux/spinlock.h>
 #include <linux/mm.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/system.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 #include <asm/io.h>
-#include <asm/sbus.h>
-#include <asm/ebus.h>
 #include <asm/upa.h>
 
 static DEFINE_SPINLOCK(flash_lock);
@@ -161,97 +160,68 @@ static const struct file_operations flash_fops = {
 
 static struct miscdevice flash_dev = { FLASH_MINOR, "flash", &flash_fops };
 
-static int __init flash_init(void)
+static int __devinit flash_probe(struct of_device *op,
+				 const struct of_device_id *match)
 {
-	struct sbus_bus *sbus;
-	struct sbus_dev *sdev = NULL;
-#ifdef CONFIG_PCI
-	struct linux_ebus *ebus;
-	struct linux_ebus_device *edev = NULL;
-	struct linux_prom_registers regs[2];
-	int len, nregs;
-#endif
-	int err;
-
-	for_all_sbusdev(sdev, sbus) {
-		if (!strcmp(sdev->prom_name, "flashprom")) {
-			if (sdev->reg_addrs[0].phys_addr == sdev->reg_addrs[1].phys_addr) {
-				flash.read_base = ((unsigned long)sdev->reg_addrs[0].phys_addr) |
-					(((unsigned long)sdev->reg_addrs[0].which_io)<<32UL);
-				flash.read_size = sdev->reg_addrs[0].reg_size;
-				flash.write_base = flash.read_base;
-				flash.write_size = flash.read_size;
-			} else {
-				flash.read_base = ((unsigned long)sdev->reg_addrs[0].phys_addr) |
-					(((unsigned long)sdev->reg_addrs[0].which_io)<<32UL);
-				flash.read_size = sdev->reg_addrs[0].reg_size;
-				flash.write_base = ((unsigned long)sdev->reg_addrs[1].phys_addr) |
-					(((unsigned long)sdev->reg_addrs[1].which_io)<<32UL);
-				flash.write_size = sdev->reg_addrs[1].reg_size;
-			}
-			flash.busy = 0;
-			break;
-		}
-	}
-	if (!sdev) {
-#ifdef CONFIG_PCI
-		const struct linux_prom_registers *ebus_regs;
-
-		for_each_ebus(ebus) {
-			for_each_ebusdev(edev, ebus) {
-				if (!strcmp(edev->prom_node->name, "flashprom"))
-					goto ebus_done;
-			}
-		}
-	ebus_done:
-		if (!edev)
-			return -ENODEV;
-
-		ebus_regs = of_get_property(edev->prom_node, "reg", &len);
-		if (!ebus_regs || (len % sizeof(regs[0])) != 0) {
-			printk("flash: Strange reg property size %d\n", len);
-			return -ENODEV;
-		}
-
-		nregs = len / sizeof(ebus_regs[0]);
+	struct device_node *dp = op->node;
+	struct device_node *parent;
 
-		flash.read_base = edev->resource[0].start;
-		flash.read_size = ebus_regs[0].reg_size;
+	parent = dp->parent;
 
-		if (nregs == 1) {
-			flash.write_base = edev->resource[0].start;
-			flash.write_size = ebus_regs[0].reg_size;
-		} else if (nregs == 2) {
-			flash.write_base = edev->resource[1].start;
-			flash.write_size = ebus_regs[1].reg_size;
-		} else {
-			printk("flash: Strange number of regs %d\n", nregs);
-			return -ENODEV;
-		}
-
-		flash.busy = 0;
-
-#else
+	if (strcmp(parent->name, "sbus") &&
+	    strcmp(parent->name, "sbi") &&
+	    strcmp(parent->name, "ebus"))
 		return -ENODEV;
-#endif
+
+	flash.read_base = op->resource[0].start;
+	flash.read_size = resource_size(&op->resource[0]);
+	if (op->resource[1].flags) {
+		flash.write_base = op->resource[1].start;
+		flash.write_size = resource_size(&op->resource[1]);
+	} else {
+		flash.write_base = op->resource[0].start;
+		flash.write_size = resource_size(&op->resource[0]);
 	}
+	flash.busy = 0;
 
-	printk("OBP Flash: RD %lx[%lx] WR %lx[%lx]\n",
+	printk(KERN_INFO "%s: OBP Flash, RD %lx[%lx] WR %lx[%lx]\n",
+	       op->node->full_name,
 	       flash.read_base, flash.read_size,
 	       flash.write_base, flash.write_size);
 
-	err = misc_register(&flash_dev);
-	if (err) {
-		printk(KERN_ERR "flash: unable to get misc minor\n");
-		return err;
-	}
+	return misc_register(&flash_dev);
+}
+
+static int __devexit flash_remove(struct of_device *op)
+{
+	misc_deregister(&flash_dev);
 
 	return 0;
 }
 
+static const struct of_device_id flash_match[] = {
+	{
+		.name = "flashprom",
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, flash_match);
+
+static struct of_platform_driver flash_driver = {
+	.name		= "flash",
+	.match_table	= flash_match,
+	.probe		= flash_probe,
+	.remove		= __devexit_p(flash_remove),
+};
+
+static int __init flash_init(void)
+{
+	return of_register_driver(&flash_driver, &of_bus_type);
+}
+
 static void __exit flash_cleanup(void)
 {
-	misc_deregister(&flash_dev);
+	of_unregister_driver(&flash_driver);
 }
 
 module_init(flash_init);
diff --git a/drivers/sbus/char/rtc.c b/drivers/sbus/char/rtc.c
deleted file mode 100644
index b0429917154d..000000000000
--- a/drivers/sbus/char/rtc.c
+++ /dev/null
@@ -1,275 +0,0 @@
-/* $Id: rtc.c,v 1.28 2001/10/08 22:19:51 davem Exp $
- *
- * Linux/SPARC Real Time Clock Driver
- * Copyright (C) 1996 Thomas K. Dyas (tdyas@eden.rutgers.edu)
- *
- * This is a little driver that lets a user-level program access
- * the SPARC Mostek real time clock chip. It is no use unless you
- * use the modified clock utility.
- *
- * Get the modified clock utility from:
- *   ftp://vger.kernel.org/pub/linux/Sparc/userland/clock.c
- */
-
-#include <linux/module.h>
-#include <linux/smp_lock.h>
-#include <linux/types.h>
-#include <linux/errno.h>
-#include <linux/miscdevice.h>
-#include <linux/slab.h>
-#include <linux/fcntl.h>
-#include <linux/poll.h>
-#include <linux/init.h>
-#include <asm/io.h>
-#include <asm/mostek.h>
-#include <asm/system.h>
-#include <asm/uaccess.h>
-#include <asm/rtc.h>
-
-static int rtc_busy = 0;
-
-/* This is the structure layout used by drivers/char/rtc.c, we
- * support that driver's ioctls so that things are less messy in
- * userspace.
- */
-struct rtc_time_generic {
-	int tm_sec;
-	int tm_min;
-	int tm_hour;
-	int tm_mday;
-	int tm_mon;
-	int tm_year;
-	int tm_wday;
-	int tm_yday;
-	int tm_isdst;
-};
-#define RTC_AIE_ON	_IO('p', 0x01)	/* Alarm int. enable on		*/
-#define RTC_AIE_OFF	_IO('p', 0x02)	/* ... off			*/
-#define RTC_UIE_ON	_IO('p', 0x03)	/* Update int. enable on	*/
-#define RTC_UIE_OFF	_IO('p', 0x04)	/* ... off			*/
-#define RTC_PIE_ON	_IO('p', 0x05)	/* Periodic int. enable on	*/
-#define RTC_PIE_OFF	_IO('p', 0x06)	/* ... off			*/
-#define RTC_WIE_ON	_IO('p', 0x0f)  /* Watchdog int. enable on	*/
-#define RTC_WIE_OFF	_IO('p', 0x10)  /* ... off			*/
-#define RTC_RD_TIME	_IOR('p', 0x09, struct rtc_time_generic) /* Read RTC time   */
-#define RTC_SET_TIME	_IOW('p', 0x0a, struct rtc_time_generic) /* Set RTC time    */
-#define RTC_ALM_SET	_IOW('p', 0x07, struct rtc_time) /* Set alarm time  */
-#define RTC_ALM_READ	_IOR('p', 0x08, struct rtc_time) /* Read alarm time */
-#define RTC_IRQP_READ	_IOR('p', 0x0b, unsigned long)	 /* Read IRQ rate   */
-#define RTC_IRQP_SET	_IOW('p', 0x0c, unsigned long)	 /* Set IRQ rate    */
-#define RTC_EPOCH_READ	_IOR('p', 0x0d, unsigned long)	 /* Read epoch      */
-#define RTC_EPOCH_SET	_IOW('p', 0x0e, unsigned long)	 /* Set epoch       */
-#define RTC_WKALM_SET	_IOW('p', 0x0f, struct rtc_wkalrm)/* Set wakeup alarm*/
-#define RTC_WKALM_RD	_IOR('p', 0x10, struct rtc_wkalrm)/* Get wakeup alarm*/
-#define RTC_PLL_GET	_IOR('p', 0x11, struct rtc_pll_info)  /* Get PLL correction */
-#define RTC_PLL_SET	_IOW('p', 0x12, struct rtc_pll_info)  /* Set PLL correction */
-
-/* Retrieve the current date and time from the real time clock. */
-static void get_rtc_time(struct rtc_time *t)
-{
-	void __iomem *regs = mstk48t02_regs;
-	u8 tmp;
-
-	spin_lock_irq(&mostek_lock);
-
-	tmp = mostek_read(regs + MOSTEK_CREG);
-	tmp |= MSTK_CREG_READ;
-	mostek_write(regs + MOSTEK_CREG, tmp);
-
-	t->sec = MSTK_REG_SEC(regs);
-	t->min = MSTK_REG_MIN(regs);
-	t->hour = MSTK_REG_HOUR(regs);
-	t->dow = MSTK_REG_DOW(regs);
-	t->dom = MSTK_REG_DOM(regs);
-	t->month = MSTK_REG_MONTH(regs);
-	t->year = MSTK_CVT_YEAR( MSTK_REG_YEAR(regs) );
-
-	tmp = mostek_read(regs + MOSTEK_CREG);
-	tmp &= ~MSTK_CREG_READ;
-	mostek_write(regs + MOSTEK_CREG, tmp);
-
-	spin_unlock_irq(&mostek_lock);
-}
-
-/* Set the current date and time inthe real time clock. */
-void set_rtc_time(struct rtc_time *t)
-{
-	void __iomem *regs = mstk48t02_regs;
-	u8 tmp;
-
-	spin_lock_irq(&mostek_lock);
-
-	tmp = mostek_read(regs + MOSTEK_CREG);
-	tmp |= MSTK_CREG_WRITE;
-	mostek_write(regs + MOSTEK_CREG, tmp);
-
-	MSTK_SET_REG_SEC(regs,t->sec);
-	MSTK_SET_REG_MIN(regs,t->min);
-	MSTK_SET_REG_HOUR(regs,t->hour);
-	MSTK_SET_REG_DOW(regs,t->dow);
-	MSTK_SET_REG_DOM(regs,t->dom);
-	MSTK_SET_REG_MONTH(regs,t->month);
-	MSTK_SET_REG_YEAR(regs,t->year - MSTK_YEAR_ZERO);
-
-	tmp = mostek_read(regs + MOSTEK_CREG);
-	tmp &= ~MSTK_CREG_WRITE;
-	mostek_write(regs + MOSTEK_CREG, tmp);
-
-	spin_unlock_irq(&mostek_lock);
-}
-
-static int put_rtc_time_generic(void __user *argp, struct rtc_time *tm)
-{
-	struct rtc_time_generic __user *utm = argp;
-
-	if (__put_user(tm->sec, &utm->tm_sec) ||
-	    __put_user(tm->min, &utm->tm_min) ||
-	    __put_user(tm->hour, &utm->tm_hour) ||
-	    __put_user(tm->dom, &utm->tm_mday) ||
-	    __put_user(tm->month, &utm->tm_mon) ||
-	    __put_user(tm->year, &utm->tm_year) ||
-	    __put_user(tm->dow, &utm->tm_wday) ||
-	    __put_user(0, &utm->tm_yday) ||
-	    __put_user(0, &utm->tm_isdst))
-		return -EFAULT;
-
-	return 0;
-}
-
-static int get_rtc_time_generic(struct rtc_time *tm, void __user *argp)
-{
-	struct rtc_time_generic __user *utm = argp;
-
-	if (__get_user(tm->sec, &utm->tm_sec) ||
-	    __get_user(tm->min, &utm->tm_min) ||
-	    __get_user(tm->hour, &utm->tm_hour) ||
-	    __get_user(tm->dom, &utm->tm_mday) ||
-	    __get_user(tm->month, &utm->tm_mon) ||
-	    __get_user(tm->year, &utm->tm_year) ||
-	    __get_user(tm->dow, &utm->tm_wday))
-		return -EFAULT;
-
-	return 0;
-}
-
-static int rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
-	unsigned long arg)
-{
-	struct rtc_time rtc_tm;
-	void __user *argp = (void __user *)arg;
-
-	switch (cmd) {
-	/* No interrupt support, return an error
-	 * compatible with drivers/char/rtc.c
-	 */
-	case RTC_AIE_OFF:
-	case RTC_AIE_ON:
-	case RTC_PIE_OFF:
-	case RTC_PIE_ON:
-	case RTC_UIE_OFF:
-	case RTC_UIE_ON:
-	case RTC_IRQP_READ:
-	case RTC_IRQP_SET:
-	case RTC_EPOCH_SET:
-	case RTC_EPOCH_READ:
-		return -EINVAL;
-
-	case RTCGET:
-	case RTC_RD_TIME:
-		memset(&rtc_tm, 0, sizeof(struct rtc_time));
-		get_rtc_time(&rtc_tm);
-
-		if (cmd == RTCGET) {
-			if (copy_to_user(argp, &rtc_tm,
-					 sizeof(struct rtc_time)))
-				return -EFAULT;
-		} else if (put_rtc_time_generic(argp, &rtc_tm))
-			return -EFAULT;
-
-		return 0;
-
-
-	case RTCSET:
-	case RTC_SET_TIME:
-		if (!capable(CAP_SYS_TIME))
-			return -EPERM;
-
-		if (cmd == RTCSET) {
-			if (copy_from_user(&rtc_tm, argp,
-					   sizeof(struct rtc_time)))
-				return -EFAULT;
-		} else if (get_rtc_time_generic(&rtc_tm, argp))
-			return -EFAULT;
-
-		set_rtc_time(&rtc_tm);
-
-		return 0;
-
-	default:
-		return -EINVAL;
-	}
-}
-
-static int rtc_open(struct inode *inode, struct file *file)
-{
-	int ret;
-
-	lock_kernel();
-	spin_lock_irq(&mostek_lock);
-	if (rtc_busy) {
-		ret = -EBUSY;
-	} else {
-		rtc_busy = 1;
-		ret = 0;
-	}
-	spin_unlock_irq(&mostek_lock);
-	unlock_kernel();
-
-	return ret;
-}
-
-static int rtc_release(struct inode *inode, struct file *file)
-{
-	rtc_busy = 0;
-
-	return 0;
-}
-
-static const struct file_operations rtc_fops = {
-	.owner =	THIS_MODULE,
-	.llseek =	no_llseek,
-	.ioctl =	rtc_ioctl,
-	.open =		rtc_open,
-	.release =	rtc_release,
-};
-
-static struct miscdevice rtc_dev = { RTC_MINOR, "rtc", &rtc_fops };
-
-static int __init rtc_sun_init(void)
-{
-	int error;
-
-	/* It is possible we are being driven by some other RTC chip
-	 * and thus another RTC driver is handling things.
-	 */
-	if (!mstk48t02_regs)
-		return -ENODEV;
-
-	error = misc_register(&rtc_dev);
-	if (error) {
-		printk(KERN_ERR "rtc: unable to get misc minor for Mostek\n");
-		return error;
-	}
-	printk("rtc_sun_init: Registered Mostek RTC driver.\n");
-
-	return 0;
-}
-
-static void __exit rtc_sun_cleanup(void)
-{
-	misc_deregister(&rtc_dev);
-}
-
-module_init(rtc_sun_init);
-module_exit(rtc_sun_cleanup);
-MODULE_LICENSE("GPL");
diff --git a/drivers/sbus/char/uctrl.c b/drivers/sbus/char/uctrl.c
index 777637594acd..27993c37775d 100644
--- a/drivers/sbus/char/uctrl.c
+++ b/drivers/sbus/char/uctrl.c
@@ -1,7 +1,7 @@
-/* $Id: uctrl.c,v 1.12 2001/10/08 22:19:51 davem Exp $
- * uctrl.c: TS102 Microcontroller interface on Tadpole Sparcbook 3
+/* uctrl.c: TS102 Microcontroller interface on Tadpole Sparcbook 3
  *
  * Copyright 1999 Derrick J Brashear (shadow@dementia.org)
+ * Copyright 2008 David S. Miller (davem@davemloft.net)
  */
 
 #include <linux/module.h>
@@ -14,6 +14,8 @@
 #include <linux/init.h>
 #include <linux/miscdevice.h>
 #include <linux/mm.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/openprom.h>
 #include <asm/oplib.h>
@@ -21,7 +23,6 @@
 #include <asm/irq.h>
 #include <asm/io.h>
 #include <asm/pgtable.h>
-#include <asm/sbus.h>
 
 #define UCTRL_MINOR	174
 
@@ -33,26 +34,26 @@
 #endif
 
 struct uctrl_regs {
-	volatile u32 uctrl_intr;
-	volatile u32 uctrl_data;
-	volatile u32 uctrl_stat;
-	volatile u32 uctrl_xxx[5];
+	u32 uctrl_intr;
+	u32 uctrl_data;
+	u32 uctrl_stat;
+	u32 uctrl_xxx[5];
 };
 
 struct ts102_regs {
-	volatile u32 card_a_intr;
-	volatile u32 card_a_stat;
-	volatile u32 card_a_ctrl;
-	volatile u32 card_a_xxx;
-	volatile u32 card_b_intr;
-	volatile u32 card_b_stat;
-	volatile u32 card_b_ctrl;
-	volatile u32 card_b_xxx;
-	volatile u32 uctrl_intr;
-	volatile u32 uctrl_data;
-	volatile u32 uctrl_stat;
-	volatile u32 uctrl_xxx;
-	volatile u32 ts102_xxx[4];
+	u32 card_a_intr;
+	u32 card_a_stat;
+	u32 card_a_ctrl;
+	u32 card_a_xxx;
+	u32 card_b_intr;
+	u32 card_b_stat;
+	u32 card_b_ctrl;
+	u32 card_b_xxx;
+	u32 uctrl_intr;
+	u32 uctrl_data;
+	u32 uctrl_stat;
+	u32 uctrl_xxx;
+	u32 ts102_xxx[4];
 };
 
 /* Bits for uctrl_intr register */
@@ -186,17 +187,15 @@ enum uctrl_opcode {
   POWER_RESTART=0x83,
 };
 
-struct uctrl_driver {
-	struct uctrl_regs *regs;
+static struct uctrl_driver {
+	struct uctrl_regs __iomem *regs;
 	int irq;
 	int pending;
 	struct uctrl_status status;
-};
-
-static struct uctrl_driver drv;
+} *global_driver;
 
-static void uctrl_get_event_status(void);
-static void uctrl_get_external_status(void);
+static void uctrl_get_event_status(struct uctrl_driver *);
+static void uctrl_get_external_status(struct uctrl_driver *);
 
 static int
 uctrl_ioctl(struct inode *inode, struct file *file,
@@ -213,16 +212,14 @@ static int
 uctrl_open(struct inode *inode, struct file *file)
 {
 	lock_kernel();
-	uctrl_get_event_status();
-	uctrl_get_external_status();
+	uctrl_get_event_status(global_driver);
+	uctrl_get_external_status(global_driver);
 	unlock_kernel();
 	return 0;
 }
 
 static irqreturn_t uctrl_interrupt(int irq, void *dev_id)
 {
-	struct uctrl_driver *driver = (struct uctrl_driver *)dev_id;
-	printk("in uctrl_interrupt\n");
 	return IRQ_HANDLED;
 }
 
@@ -244,11 +241,11 @@ static struct miscdevice uctrl_dev = {
 { \
   unsigned int i; \
   for (i = 0; i < 10000; i++) { \
-    if (UCTRL_STAT_TXNF_STA & driver->regs->uctrl_stat) \
+      if (UCTRL_STAT_TXNF_STA & sbus_readl(&driver->regs->uctrl_stat)) \
       break; \
   } \
   dprintk(("write data 0x%02x\n", value)); \
-  driver->regs->uctrl_data = value; \
+  sbus_writel(value, &driver->regs->uctrl_data); \
 }
 
 /* Wait for something to read, read it, then clear the bit */
@@ -257,24 +254,23 @@ static struct miscdevice uctrl_dev = {
   unsigned int i; \
   value = 0; \
   for (i = 0; i < 10000; i++) { \
-    if ((UCTRL_STAT_RXNE_STA & driver->regs->uctrl_stat) == 0) \
+      if ((UCTRL_STAT_RXNE_STA & sbus_readl(&driver->regs->uctrl_stat)) == 0) \
       break; \
     udelay(1); \
   } \
-  value = driver->regs->uctrl_data; \
+  value = sbus_readl(&driver->regs->uctrl_data); \
   dprintk(("read data 0x%02x\n", value)); \
-  driver->regs->uctrl_stat = UCTRL_STAT_RXNE_STA; \
+  sbus_writel(UCTRL_STAT_RXNE_STA, &driver->regs->uctrl_stat); \
 }
 
-static void uctrl_do_txn(struct uctrl_txn *txn)
+static void uctrl_do_txn(struct uctrl_driver *driver, struct uctrl_txn *txn)
 {
-	struct uctrl_driver *driver = &drv;
 	int stat, incnt, outcnt, bytecnt, intr;
 	u32 byte;
 
-	stat = driver->regs->uctrl_stat;
-	intr = driver->regs->uctrl_intr;
-	driver->regs->uctrl_stat = stat;
+	stat = sbus_readl(&driver->regs->uctrl_stat);
+	intr = sbus_readl(&driver->regs->uctrl_intr);
+	sbus_writel(stat, &driver->regs->uctrl_stat);
 
 	dprintk(("interrupt stat 0x%x int 0x%x\n", stat, intr));
 
@@ -305,9 +301,8 @@ static void uctrl_do_txn(struct uctrl_txn *txn)
 	}
 }
 
-static void uctrl_get_event_status(void)
+static void uctrl_get_event_status(struct uctrl_driver *driver)
 {
-	struct uctrl_driver *driver = &drv;
 	struct uctrl_txn txn;
 	u8 outbits[2];
 
@@ -317,7 +312,7 @@ static void uctrl_get_event_status(void)
 	txn.inbuf = NULL;
 	txn.outbuf = outbits;
 
-	uctrl_do_txn(&txn);
+	uctrl_do_txn(driver, &txn);
 
 	dprintk(("bytes %x %x\n", (outbits[0] & 0xff), (outbits[1] & 0xff)));
 	driver->status.event_status = 
@@ -325,9 +320,8 @@ static void uctrl_get_event_status(void)
 	dprintk(("ev is %x\n", driver->status.event_status));
 }
 
-static void uctrl_get_external_status(void)
+static void uctrl_get_external_status(struct uctrl_driver *driver)
 {
-	struct uctrl_driver *driver = &drv;
 	struct uctrl_txn txn;
 	u8 outbits[2];
 	int i, v;
@@ -338,7 +332,7 @@ static void uctrl_get_external_status(void)
 	txn.inbuf = NULL;
 	txn.outbuf = outbits;
 
-	uctrl_do_txn(&txn);
+	uctrl_do_txn(driver, &txn);
 
 	dprintk(("bytes %x %x\n", (outbits[0] & 0xff), (outbits[1] & 0xff)));
 	driver->status.external_status = 
@@ -354,71 +348,101 @@ static void uctrl_get_external_status(void)
 	
 }
 
-static int __init ts102_uctrl_init(void)
+static int __devinit uctrl_probe(struct of_device *op,
+				 const struct of_device_id *match)
 {
-	struct uctrl_driver *driver = &drv;
-	int len;
-	struct linux_prom_irqs tmp_irq[2];
-        unsigned int vaddr[2] = { 0, 0 };
-	int tmpnode, uctrlnode = prom_getchild(prom_root_node);
-	int err;
+	struct uctrl_driver *p;
+	int err = -ENOMEM;
 
-	tmpnode = prom_searchsiblings(uctrlnode, "obio");
+	p = kzalloc(sizeof(*p), GFP_KERNEL);
+	if (!p) {
+		printk(KERN_ERR "uctrl: Unable to allocate device struct.\n");
+		goto out;
+	}
 
-	if (tmpnode)
-	  uctrlnode = prom_getchild(tmpnode);
+	p->regs = of_ioremap(&op->resource[0], 0,
+			     resource_size(&op->resource[0]),
+			     "uctrl");
+	if (!p->regs) {
+		printk(KERN_ERR "uctrl: Unable to map registers.\n");
+		goto out_free;
+	}
 
-	uctrlnode = prom_searchsiblings(uctrlnode, "uctrl");
+	p->irq = op->irqs[0];
+	err = request_irq(p->irq, uctrl_interrupt, 0, "uctrl", p);
+	if (err) {
+		printk(KERN_ERR "uctrl: Unable to register irq.\n");
+		goto out_iounmap;
+	}
 
-	if (!uctrlnode)
-		return -ENODEV;
+	err = misc_register(&uctrl_dev);
+	if (err) {
+		printk(KERN_ERR "uctrl: Unable to register misc device.\n");
+		goto out_free_irq;
+	}
 
-	/* the prom mapped it for us */
-	len = prom_getproperty(uctrlnode, "address", (void *) vaddr,
-			       sizeof(vaddr));
-	driver->regs = (struct uctrl_regs *)vaddr[0];
+	sbus_writel(UCTRL_INTR_RXNE_REQ|UCTRL_INTR_RXNE_MSK, &p->regs->uctrl_intr);
+	printk(KERN_INFO "%s: uctrl regs[0x%p] (irq %d)\n",
+	       op->node->full_name, p->regs, p->irq);
+	uctrl_get_event_status(p);
+	uctrl_get_external_status(p);
 
-	len = prom_getproperty(uctrlnode, "intr", (char *) tmp_irq,
-			       sizeof(tmp_irq));
+	dev_set_drvdata(&op->dev, p);
+	global_driver = p;
 
-	/* Flush device */
-	READUCTLDATA(len);
+out:
+	return err;
 
-	if(!driver->irq) 
-		driver->irq = tmp_irq[0].pri;
+out_free_irq:
+	free_irq(p->irq, p);
 
-	err = request_irq(driver->irq, uctrl_interrupt, 0, "uctrl", driver);
-	if (err) {
-		printk("%s: unable to register irq %d\n",
-		       __func__, driver->irq);
-		return err;
-	}
+out_iounmap:
+	of_iounmap(&op->resource[0], p->regs, resource_size(&op->resource[0]));
 
-	if (misc_register(&uctrl_dev)) {
-		printk("%s: unable to get misc minor %d\n",
-		       __func__, uctrl_dev.minor);
-		free_irq(driver->irq, driver);
-		return -ENODEV;
-	}
+out_free:
+	kfree(p);
+	goto out;
+}
 
-	driver->regs->uctrl_intr = UCTRL_INTR_RXNE_REQ|UCTRL_INTR_RXNE_MSK;
-	printk("uctrl: 0x%p (irq %d)\n", driver->regs, driver->irq);
-	uctrl_get_event_status();
-	uctrl_get_external_status();
-        return 0;
+static int __devexit uctrl_remove(struct of_device *op)
+{
+	struct uctrl_driver *p = dev_get_drvdata(&op->dev);
+
+	if (p) {
+		misc_deregister(&uctrl_dev);
+		free_irq(p->irq, p);
+		of_iounmap(&op->resource[0], p->regs, resource_size(&op->resource[0]));
+		kfree(p);
+	}
+	return 0;
 }
 
-static void __exit ts102_uctrl_cleanup(void)
+static const struct of_device_id uctrl_match[] = {
+	{
+		.name = "uctrl",
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, uctrl_match);
+
+static struct of_platform_driver uctrl_driver = {
+	.name		= "uctrl",
+	.match_table	= uctrl_match,
+	.probe		= uctrl_probe,
+	.remove		= __devexit_p(uctrl_remove),
+};
+
+
+static int __init uctrl_init(void)
 {
-	struct uctrl_driver *driver = &drv;
+	return of_register_driver(&uctrl_driver, &of_bus_type);
+}
 
-	misc_deregister(&uctrl_dev);
-	if (driver->irq)
-		free_irq(driver->irq, driver);
-	if (driver->regs)
-		driver->regs = NULL;
+static void __exit uctrl_exit(void)
+{
+	of_unregister_driver(&uctrl_driver);
 }
 
-module_init(ts102_uctrl_init);
-module_exit(ts102_uctrl_cleanup);
+module_init(uctrl_init);
+module_exit(uctrl_exit);
 MODULE_LICENSE("GPL");
diff --git a/drivers/sbus/char/vfc.h b/drivers/sbus/char/vfc.h
deleted file mode 100644
index a5240c52aa0b..000000000000
--- a/drivers/sbus/char/vfc.h
+++ /dev/null
@@ -1,171 +0,0 @@
-#ifndef _LINUX_VFC_H_
-#define _LINUX_VFC_H_
-
-/*
- * The control register for the vfc is at offset 0x4000
- * The first field ram bank is located at offset 0x5000
- * The second field ram bank is at offset 0x7000
- * i2c_reg address the Phillips PCF8584(see notes in vfc_i2c.c) 
- *    data and transmit register.
- * i2c_s1 controls register s1 of the PCF8584
- * i2c_write seems to be similar to i2c_write but I am not 
- *    quite sure why sun uses it
- * 
- * I am also not sure whether or not you can read the fram bank as a
- * whole or whether you must read each word individually from offset
- * 0x5000 as soon as I figure it out I will update this file */
-
-struct vfc_regs {
-	char pad1[0x4000];
-	unsigned int control;  /* Offset 0x4000 */
-	char pad2[0xffb];      /* from offset 0x4004 to 0x5000 */
-	unsigned int fram_bank1; /* Offset 0x5000 */
-	char pad3[0xffb];        /* from offset 0x5004 to 0x6000 */
-	unsigned int i2c_reg; /* Offset 0x6000 */
-	unsigned int i2c_magic2; /* Offset 0x6004 */
-	unsigned int i2c_s1;  /* Offset 0x6008 */
-	unsigned int i2c_write; /* Offset 0x600c */
-	char pad4[0xff0];     /* from offset 0x6010 to 0x7000 */
-	unsigned int fram_bank2; /* Offset 0x7000 */
-	char pad5[0x1000];
-};
-
-#define VFC_SAA9051_NR (13)
-#define VFC_SAA9051_ADDR (0x8a)
-	/* The saa9051 returns the following for its status 
-	 * bit 0 - 0
-	 * bit 1 - SECAM color detected (1=found,0=not found)
-	 * bit 2 - COLOR detected (1=found,0=not found)
-	 * bit 3 - 0
-	 * bit 4 - Field frequency bit (1=60Hz (NTSC), 0=50Hz (PAL))
-	 * bit 5 - 1
-	 * bit 6 - horizontal frequency lock (1=transmitter found,
-	 *                                    0=no transmitter)
-	 * bit 7 - Power on reset bit (1=reset,0=at least one successful 
-	 *                                       read of the status byte)
-	 */
-
-#define VFC_SAA9051_PONRES (0x80)
-#define VFC_SAA9051_HLOCK (0x40)
-#define VFC_SAA9051_FD (0x10)
-#define VFC_SAA9051_CD (0x04)
-#define VFC_SAA9051_CS (0x02)
-
-
-/* The various saa9051 sub addresses */
-
-#define VFC_SAA9051_IDEL (0) 
-#define VFC_SAA9051_HSY_START (1)
-#define VFC_SAA9051_HSY_STOP (2)
-#define VFC_SAA9051_HC_START (3)
-#define VFC_SAA9051_HC_STOP (4)
-#define VFC_SAA9051_HS_START (5)
-#define VFC_SAA9051_HORIZ_PEAK (6)
-#define VFC_SAA9051_HUE (7)
-#define VFC_SAA9051_C1 (8)
-#define VFC_SAA9051_C2 (9)
-#define VFC_SAA9051_C3 (0xa)
-#define VFC_SAA9051_SECAM_DELAY (0xb)
-
-
-/* Bit settings for saa9051 sub address 0x06 */
-
-#define VFC_SAA9051_AP1 (0x01)
-#define VFC_SAA9051_AP2 (0x02)
-#define VFC_SAA9051_COR1 (0x04)
-#define VFC_SAA9051_COR2 (0x08)
-#define VFC_SAA9051_BP1 (0x10)
-#define VFC_SAA9051_BP2 (0x20)
-#define VFC_SAA9051_PF (0x40)
-#define VFC_SAA9051_BY (0x80)
-
-
-/* Bit settings for saa9051 sub address 0x08 */
-
-#define VFC_SAA9051_CCFR0 (0x01)
-#define VFC_SAA9051_CCFR1 (0x02)
-#define VFC_SAA9051_YPN (0x04)
-#define VFC_SAA9051_ALT (0x08)
-#define VFC_SAA9051_CO (0x10)
-#define VFC_SAA9051_VTR (0x20)
-#define VFC_SAA9051_FS (0x40)
-#define VFC_SAA9051_HPLL (0x80)
-
-
-/* Bit settings for saa9051 sub address 9 */
-
-#define VFC_SAA9051_SS0 (0x01)
-#define VFC_SAA9051_SS1 (0x02)
-#define VFC_SAA9051_AFCC (0x04)
-#define VFC_SAA9051_CI (0x08)
-#define VFC_SAA9051_SA9D4 (0x10) /* Don't care bit */
-#define VFC_SAA9051_OEC (0x20)
-#define VFC_SAA9051_OEY (0x40)
-#define VFC_SAA9051_VNL (0x80)
-
-
-/* Bit settings for saa9051 sub address 0x0A */
-
-#define VFC_SAA9051_YDL0 (0x01)
-#define VFC_SAA9051_YDL1 (0x02)
-#define VFC_SAA9051_YDL2 (0x04)
-#define VFC_SAA9051_SS2 (0x08)
-#define VFC_SAA9051_SS3 (0x10)
-#define VFC_SAA9051_YC (0x20)
-#define VFC_SAA9051_CT (0x40)
-#define VFC_SAA9051_SYC (0x80)
-
-
-#define VFC_SAA9051_SA(a,b) ((a)->saa9051_state_array[(b)+1])
-#define vfc_update_saa9051(a) (vfc_i2c_sendbuf((a),VFC_SAA9051_ADDR,\
-					    (a)->saa9051_state_array,\
-					    VFC_SAA9051_NR))
-
-
-struct vfc_dev {
-	volatile struct vfc_regs __iomem *regs;
-	struct vfc_regs *phys_regs;
-	unsigned int control_reg;
-	struct mutex device_lock_mtx;
-	int instance;
-	int busy;
-	unsigned long which_io;
-	unsigned char saa9051_state_array[VFC_SAA9051_NR];
-};
-
-void captstat_reset(struct vfc_dev *);
-void memptr_reset(struct vfc_dev *);
-
-int vfc_pcf8584_init(struct vfc_dev *);
-void vfc_i2c_delay_no_busy(struct vfc_dev *, unsigned long);
-void vfc_i2c_delay(struct vfc_dev *);
-int vfc_i2c_sendbuf(struct vfc_dev *, unsigned char, char *, int) ;
-int vfc_i2c_recvbuf(struct vfc_dev *, unsigned char, char *, int) ;
-int vfc_i2c_reset_bus(struct vfc_dev *);
-int vfc_init_i2c_bus(struct vfc_dev *);
-
-#define VFC_CONTROL_DIAGMODE  0x10000000
-#define VFC_CONTROL_MEMPTR    0x20000000
-#define VFC_CONTROL_CAPTURE   0x02000000
-#define VFC_CONTROL_CAPTRESET 0x04000000
-
-#define VFC_STATUS_CAPTURE    0x08000000
-
-#ifdef VFC_IOCTL_DEBUG
-#define VFC_IOCTL_DEBUG_PRINTK(a) printk a
-#else
-#define VFC_IOCTL_DEBUG_PRINTK(a)
-#endif
-
-#ifdef VFC_I2C_DEBUG
-#define VFC_I2C_DEBUG_PRINTK(a) printk a
-#else
-#define VFC_I2C_DEBUG_PRINTK(a)
-#endif
-
-#endif /* _LINUX_VFC_H_ */
-
-
-
-
-
diff --git a/drivers/sbus/char/vfc_dev.c b/drivers/sbus/char/vfc_dev.c
deleted file mode 100644
index 25181bb7d627..000000000000
--- a/drivers/sbus/char/vfc_dev.c
+++ /dev/null
@@ -1,736 +0,0 @@
-/*
- * drivers/sbus/char/vfc_dev.c
- *
- * Driver for the Videopix Frame Grabber.
- * 
- * In order to use the VFC you need to program the video controller
- * chip. This chip is the Phillips SAA9051.  You need to call their
- * documentation ordering line to get the docs.
- *
- * There is very little documentation on the VFC itself.  There is
- * some useful info that can be found in the manuals that come with
- * the card.  I will hopefully write some better docs at a later date.
- *
- * Copyright (C) 1996 Manish Vachharajani (mvachhar@noc.rutgers.edu)
- * */
-
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/slab.h>
-#include <linux/errno.h>
-#include <linux/fs.h>
-#include <linux/delay.h>
-#include <linux/spinlock.h>
-#include <linux/mutex.h>
-#include <linux/mm.h>
-#include <linux/smp_lock.h>
-
-#include <asm/openprom.h>
-#include <asm/oplib.h>
-#include <asm/io.h>
-#include <asm/system.h>
-#include <asm/sbus.h>
-#include <asm/page.h>
-#include <asm/pgtable.h>
-#include <asm/uaccess.h>
-
-#define VFC_MAJOR (60)
-
-#if 0
-#define VFC_IOCTL_DEBUG
-#endif
-
-#include "vfc.h"
-#include <asm/vfc_ioctls.h>
-
-static const struct file_operations vfc_fops;
-static struct vfc_dev **vfc_dev_lst;
-static char vfcstr[]="vfc";
-static unsigned char saa9051_init_array[VFC_SAA9051_NR] = {
-	0x00, 0x64, 0x72, 0x52,
-	0x36, 0x18, 0xff, 0x20,
-	0xfc, 0x77, 0xe3, 0x50,
-	0x3e
-};
-
-static void vfc_lock_device(struct vfc_dev *dev)
-{
-	mutex_lock(&dev->device_lock_mtx);
-}
-
-static void vfc_unlock_device(struct vfc_dev *dev)
-{
-	mutex_unlock(&dev->device_lock_mtx);
-}
-
-
-static void vfc_captstat_reset(struct vfc_dev *dev)
-{
-	dev->control_reg |= VFC_CONTROL_CAPTRESET;
-	sbus_writel(dev->control_reg, &dev->regs->control);
-	dev->control_reg &= ~VFC_CONTROL_CAPTRESET;
-	sbus_writel(dev->control_reg, &dev->regs->control);
-	dev->control_reg |= VFC_CONTROL_CAPTRESET;
-	sbus_writel(dev->control_reg, &dev->regs->control);
-}
-
-static void vfc_memptr_reset(struct vfc_dev *dev)
-{
-	dev->control_reg |= VFC_CONTROL_MEMPTR;
-	sbus_writel(dev->control_reg, &dev->regs->control);
-	dev->control_reg &= ~VFC_CONTROL_MEMPTR;
-	sbus_writel(dev->control_reg, &dev->regs->control);
-	dev->control_reg |= VFC_CONTROL_MEMPTR; 
-	sbus_writel(dev->control_reg, &dev->regs->control);
-}
-
-static int vfc_csr_init(struct vfc_dev *dev)
-{
-	dev->control_reg = 0x80000000;
-	sbus_writel(dev->control_reg, &dev->regs->control);
-	udelay(200); 
-	dev->control_reg &= ~0x80000000;
-	sbus_writel(dev->control_reg, &dev->regs->control);
-	udelay(100); 
-	sbus_writel(0x0f000000, &dev->regs->i2c_magic2);
-
-	vfc_memptr_reset(dev);
-
-	dev->control_reg &= ~VFC_CONTROL_DIAGMODE;
-	dev->control_reg &= ~VFC_CONTROL_CAPTURE;
-	dev->control_reg |= 0x40000000;
-	sbus_writel(dev->control_reg, &dev->regs->control);
-
-	vfc_captstat_reset(dev);
-
-	return 0;
-}
-
-static int vfc_saa9051_init(struct vfc_dev *dev)
-{
-	int i;
-
-	for (i = 0; i < VFC_SAA9051_NR; i++)
-		dev->saa9051_state_array[i] = saa9051_init_array[i];
-
-	vfc_i2c_sendbuf(dev,VFC_SAA9051_ADDR,
-			dev->saa9051_state_array, VFC_SAA9051_NR);
-	return 0;
-}
-
-static int init_vfc_hw(struct vfc_dev *dev)
-{
-	vfc_lock_device(dev);
-	vfc_csr_init(dev);
-
-	vfc_pcf8584_init(dev);
-	vfc_init_i2c_bus(dev); /* hopefully this doesn't undo the magic
-				  sun code above*/
-	vfc_saa9051_init(dev);
-	vfc_unlock_device(dev);
-	return 0; 
-}
-
-static int init_vfc_devstruct(struct vfc_dev *dev, int instance)
-{
-	dev->instance=instance;
-	mutex_init(&dev->device_lock_mtx);
-	dev->control_reg=0;
-	dev->busy=0;
-	return 0;
-}
-
-static int init_vfc_device(struct sbus_dev *sdev,struct vfc_dev *dev,
-			   int instance)
-{
-	if(dev == NULL) {
-		printk(KERN_ERR "VFC: Bogus pointer passed\n");
-		return -ENOMEM;
-	}
-	printk("Initializing vfc%d\n",instance);
-	dev->regs = NULL;
-	dev->regs = (volatile struct vfc_regs __iomem *)
-		sbus_ioremap(&sdev->resource[0], 0,
-			     sizeof(struct vfc_regs), vfcstr);
-	dev->which_io = sdev->reg_addrs[0].which_io;
-	dev->phys_regs = (struct vfc_regs *) sdev->reg_addrs[0].phys_addr;
-	if (dev->regs == NULL)
-		return -EIO;
-
-	printk("vfc%d: registers mapped at phys_addr: 0x%lx\n    virt_addr: 0x%lx\n",
-	       instance,(unsigned long)sdev->reg_addrs[0].phys_addr,(unsigned long)dev->regs);
-
-	if (init_vfc_devstruct(dev, instance))
-		return -EINVAL;
-	if (init_vfc_hw(dev))
-		return -EIO;
-	return 0;
-}
-
-
-static struct vfc_dev *vfc_get_dev_ptr(int instance)
-{
-	return vfc_dev_lst[instance];
-}
-
-static DEFINE_SPINLOCK(vfc_dev_lock);
-
-static int vfc_open(struct inode *inode, struct file *file) 
-{
-	struct vfc_dev *dev;
-
-	lock_kernel();
-	spin_lock(&vfc_dev_lock);
-	dev = vfc_get_dev_ptr(iminor(inode));
-	if (dev == NULL) {
-		spin_unlock(&vfc_dev_lock);
-		unlock_kernel();
-		return -ENODEV;
-	}
-	if (dev->busy) {
-		spin_unlock(&vfc_dev_lock);
-		unlock_kernel();
-		return -EBUSY;
-	}
-
-	dev->busy = 1;
-	spin_unlock(&vfc_dev_lock);
-
-	vfc_lock_device(dev);
-	
-	vfc_csr_init(dev);
-	vfc_pcf8584_init(dev);
-	vfc_init_i2c_bus(dev);
-	vfc_saa9051_init(dev);
-	vfc_memptr_reset(dev);
-	vfc_captstat_reset(dev);
-	
-	vfc_unlock_device(dev);
-	unlock_kernel();
-	return 0;
-}
-
-static int vfc_release(struct inode *inode,struct file *file) 
-{
-	struct vfc_dev *dev;
-
-	spin_lock(&vfc_dev_lock);
-	dev = vfc_get_dev_ptr(iminor(inode));
-	if (!dev || !dev->busy) {
-		spin_unlock(&vfc_dev_lock);
-		return -EINVAL;
-	}
-	dev->busy = 0;
-	spin_unlock(&vfc_dev_lock);
-	return 0;
-}
-
-static int vfc_debug(struct vfc_dev *dev, int cmd, void __user *argp)
-{
-	struct vfc_debug_inout inout;
-	unsigned char *buffer;
-
-	if (!capable(CAP_SYS_ADMIN))
-		return -EPERM;
-
-	switch(cmd) {
-	case VFC_I2C_SEND:
-		if(copy_from_user(&inout, argp, sizeof(inout)))
-			return -EFAULT;
-
-		buffer = kmalloc(inout.len, GFP_KERNEL);
-		if (buffer == NULL)
-			return -ENOMEM;
-
-		if(copy_from_user(buffer, inout.buffer, inout.len)) {
-			kfree(buffer);
-			return -EFAULT;
-		}
-		
-
-		vfc_lock_device(dev);
-		inout.ret=
-			vfc_i2c_sendbuf(dev,inout.addr & 0xff,
-					buffer,inout.len);
-
-		if (copy_to_user(argp,&inout,sizeof(inout))) {
-			vfc_unlock_device(dev);
-			kfree(buffer);
-			return -EFAULT;
-		}
-		vfc_unlock_device(dev);
-
-		break;
-	case VFC_I2C_RECV:
-		if (copy_from_user(&inout, argp, sizeof(inout)))
-			return -EFAULT;
-
-		buffer = kzalloc(inout.len, GFP_KERNEL);
-		if (buffer == NULL)
-			return -ENOMEM;
-
-		vfc_lock_device(dev);
-		inout.ret=
-			vfc_i2c_recvbuf(dev,inout.addr & 0xff
-					,buffer,inout.len);
-		vfc_unlock_device(dev);
-		
-		if (copy_to_user(inout.buffer, buffer, inout.len)) {
-			kfree(buffer);
-			return -EFAULT;
-		}
-		if (copy_to_user(argp,&inout,sizeof(inout))) {
-			kfree(buffer);
-			return -EFAULT;
-		}
-		kfree(buffer);
-		break;
-	default:
-		return -EINVAL;
-	};
-
-	return 0;
-}
-
-static int vfc_capture_start(struct vfc_dev *dev)
-{
-	vfc_captstat_reset(dev);
-	dev->control_reg = sbus_readl(&dev->regs->control);
-	if((dev->control_reg & VFC_STATUS_CAPTURE)) {
-		printk(KERN_ERR "vfc%d: vfc capture status not reset\n",
-		       dev->instance);
-		return -EIO;
-	}
-
-	vfc_lock_device(dev);
-	dev->control_reg &= ~VFC_CONTROL_CAPTURE;
-	sbus_writel(dev->control_reg, &dev->regs->control);
-	dev->control_reg |= VFC_CONTROL_CAPTURE;
-	sbus_writel(dev->control_reg, &dev->regs->control);
-	dev->control_reg &= ~VFC_CONTROL_CAPTURE;
-	sbus_writel(dev->control_reg, &dev->regs->control);
-	vfc_unlock_device(dev);
-
-	return 0;
-}
-
-static int vfc_capture_poll(struct vfc_dev *dev)
-{
-	int timeout = 1000;
-
-	while (!timeout--) {
-		if (sbus_readl(&dev->regs->control) & VFC_STATUS_CAPTURE)
-			break;
-		vfc_i2c_delay_no_busy(dev, 100);
-	}
-	if(!timeout) {
-		printk(KERN_WARNING "vfc%d: capture timed out\n",
-		       dev->instance);
-		return -ETIMEDOUT;
-	}
-	return 0;
-}
-
-
-
-static int vfc_set_control_ioctl(struct inode *inode, struct file *file, 
-			  struct vfc_dev *dev, unsigned long arg) 
-{
-	int setcmd, ret = 0;
-
-	if (copy_from_user(&setcmd,(void __user *)arg,sizeof(unsigned int)))
-		return -EFAULT;
-
-	VFC_IOCTL_DEBUG_PRINTK(("vfc%d: IOCTL(VFCSCTRL) arg=0x%x\n",
-				dev->instance,setcmd));
-
-	switch(setcmd) {
-	case MEMPRST:
-		vfc_lock_device(dev);
-		vfc_memptr_reset(dev);
-		vfc_unlock_device(dev);
-		ret=0;
-		break;
-	case CAPTRCMD:
-		vfc_capture_start(dev);
-		vfc_capture_poll(dev);
-		break;
-	case DIAGMODE:
-		if(capable(CAP_SYS_ADMIN)) {
-			vfc_lock_device(dev);
-			dev->control_reg |= VFC_CONTROL_DIAGMODE;
-			sbus_writel(dev->control_reg, &dev->regs->control);
-			vfc_unlock_device(dev);
-			ret = 0;
-		} else {
-			ret = -EPERM; 
-		}
-		break;
-	case NORMMODE:
-		vfc_lock_device(dev);
-		dev->control_reg &= ~VFC_CONTROL_DIAGMODE;
-		sbus_writel(dev->control_reg, &dev->regs->control);
-		vfc_unlock_device(dev);
-		ret = 0;
-		break;
-	case CAPTRSTR:
-		vfc_capture_start(dev);
-		ret = 0;
-		break;
-	case CAPTRWAIT:
-		vfc_capture_poll(dev);
-		ret = 0;
-		break;
-	default:
-		ret = -EINVAL;
-		break;
-	};
-
-	return ret;
-}
-
-
-static int vfc_port_change_ioctl(struct inode *inode, struct file *file,
-				 struct vfc_dev *dev, unsigned long arg)
-{
-	int ret = 0;
-	int cmd;
-
-	if(copy_from_user(&cmd, (void __user *)arg, sizeof(unsigned int))) {
-		VFC_IOCTL_DEBUG_PRINTK(("vfc%d: User passed bogus pointer to "
-					"vfc_port_change_ioctl\n",
-					dev->instance));
-		return -EFAULT;
-	}
-	
-	VFC_IOCTL_DEBUG_PRINTK(("vfc%d: IOCTL(VFCPORTCHG) arg=0x%x\n",
-				dev->instance, cmd));
-
-	switch(cmd) {
-	case 1:
-	case 2:
-		VFC_SAA9051_SA(dev,VFC_SAA9051_HSY_START) = 0x72; 
-		VFC_SAA9051_SA(dev,VFC_SAA9051_HSY_STOP) = 0x52;
-		VFC_SAA9051_SA(dev,VFC_SAA9051_HC_START) = 0x36;
-		VFC_SAA9051_SA(dev,VFC_SAA9051_HC_STOP) = 0x18;
-		VFC_SAA9051_SA(dev,VFC_SAA9051_HORIZ_PEAK) = VFC_SAA9051_BP2;
-		VFC_SAA9051_SA(dev,VFC_SAA9051_C3) = VFC_SAA9051_CT | VFC_SAA9051_SS3;
-		VFC_SAA9051_SA(dev,VFC_SAA9051_SECAM_DELAY) = 0x3e;
-		break;
-	case 3:
-		VFC_SAA9051_SA(dev,VFC_SAA9051_HSY_START) = 0x3a;
-		VFC_SAA9051_SA(dev,VFC_SAA9051_HSY_STOP) = 0x17;
-		VFC_SAA9051_SA(dev,VFC_SAA9051_HC_START) = 0xfa;
-		VFC_SAA9051_SA(dev,VFC_SAA9051_HC_STOP) = 0xde;
-		VFC_SAA9051_SA(dev,VFC_SAA9051_HORIZ_PEAK) =
-			VFC_SAA9051_BY | VFC_SAA9051_PF | VFC_SAA9051_BP2;
-		VFC_SAA9051_SA(dev,VFC_SAA9051_C3) = VFC_SAA9051_YC;
-		VFC_SAA9051_SA(dev,VFC_SAA9051_SECAM_DELAY) = 0;
-		VFC_SAA9051_SA(dev,VFC_SAA9051_C2) &=
-			~(VFC_SAA9051_SS0 | VFC_SAA9051_SS1);
-		break;
-	default:
-		ret = -EINVAL;
-		return ret;
-		break;
-	}
-
-	switch(cmd) {
-	case 1:
-		VFC_SAA9051_SA(dev,VFC_SAA9051_C2) |=
-			(VFC_SAA9051_SS0 | VFC_SAA9051_SS1);
-		break;
-	case 2:
-		VFC_SAA9051_SA(dev,VFC_SAA9051_C2) &=
-			~(VFC_SAA9051_SS0 | VFC_SAA9051_SS1);
-		VFC_SAA9051_SA(dev,VFC_SAA9051_C2) |= VFC_SAA9051_SS0; 
-		break;
-	case 3:
-		break;
-	default:
-		ret = -EINVAL;
-		return ret;
-		break;
-	}
-	VFC_SAA9051_SA(dev,VFC_SAA9051_C3) &= ~(VFC_SAA9051_SS2);
-	ret=vfc_update_saa9051(dev);
-	udelay(500);
-	VFC_SAA9051_SA(dev,VFC_SAA9051_C3) |= (VFC_SAA9051_SS2);
-	ret=vfc_update_saa9051(dev);
-	return ret;
-}
-
-static int vfc_set_video_ioctl(struct inode *inode, struct file *file,
-			       struct vfc_dev *dev, unsigned long arg)
-{
-	int ret = 0;
-	int cmd;
-
-	if(copy_from_user(&cmd, (void __user *)arg, sizeof(unsigned int))) {
-		VFC_IOCTL_DEBUG_PRINTK(("vfc%d: User passed bogus pointer to "
-					"vfc_set_video_ioctl\n",
-					dev->instance));
-		return ret;
-	}
-	
-	VFC_IOCTL_DEBUG_PRINTK(("vfc%d: IOCTL(VFCSVID) arg=0x%x\n",
-				dev->instance, cmd));
-	switch(cmd) {
-	case STD_NTSC:
-		VFC_SAA9051_SA(dev,VFC_SAA9051_C1) &= ~VFC_SAA9051_ALT;
-		VFC_SAA9051_SA(dev,VFC_SAA9051_C1) |= VFC_SAA9051_YPN | 
-			VFC_SAA9051_CCFR0 | VFC_SAA9051_CCFR1 | VFC_SAA9051_FS;
-		ret = vfc_update_saa9051(dev);
-		break;
-	case STD_PAL:
-		VFC_SAA9051_SA(dev,VFC_SAA9051_C1) &= ~(VFC_SAA9051_YPN | 
-							VFC_SAA9051_CCFR1 | 
-							VFC_SAA9051_CCFR0 |
-							VFC_SAA9051_FS);
-		VFC_SAA9051_SA(dev,VFC_SAA9051_C1) |= VFC_SAA9051_ALT;
-		ret = vfc_update_saa9051(dev);
-		break;
-
-	case COLOR_ON:
-		VFC_SAA9051_SA(dev,VFC_SAA9051_C1) |= VFC_SAA9051_CO;
-		VFC_SAA9051_SA(dev,VFC_SAA9051_HORIZ_PEAK) &=
-			~(VFC_SAA9051_BY | VFC_SAA9051_PF);
-		ret = vfc_update_saa9051(dev);
-		break;
-	case MONO:
-		VFC_SAA9051_SA(dev,VFC_SAA9051_C1) &= ~(VFC_SAA9051_CO);
-		VFC_SAA9051_SA(dev,VFC_SAA9051_HORIZ_PEAK) |=
-			(VFC_SAA9051_BY | VFC_SAA9051_PF);
-		ret = vfc_update_saa9051(dev);
-		break;
-	default:
-		ret = -EINVAL;
-		break;
-	};
-
-	return ret;
-}
-
-static int vfc_get_video_ioctl(struct inode *inode, struct file *file,
-			       struct vfc_dev *dev, unsigned long arg)
-{
-	int ret = 0;
-	unsigned int status = NO_LOCK;
-	unsigned char buf[1];
-
-	if(vfc_i2c_recvbuf(dev, VFC_SAA9051_ADDR, buf, 1)) {
-		printk(KERN_ERR "vfc%d: Unable to get status\n",
-		       dev->instance);
-		return -EIO;
-	}
-
-	if(buf[0] & VFC_SAA9051_HLOCK) {
-		status = NO_LOCK;
-	} else if(buf[0] & VFC_SAA9051_FD) {
-		if(buf[0] & VFC_SAA9051_CD)
-			status = NTSC_COLOR;
-		else
-			status = NTSC_NOCOLOR;
-	} else {
-		if(buf[0] & VFC_SAA9051_CD)
-			status = PAL_COLOR;
-		else
-			status = PAL_NOCOLOR;
-	}
-	VFC_IOCTL_DEBUG_PRINTK(("vfc%d: IOCTL(VFCGVID) returning status 0x%x; "
-				"buf[0]=%x\n", dev->instance, status, buf[0]));
-
-	if (copy_to_user((void __user *)arg,&status,sizeof(unsigned int))) {
-		VFC_IOCTL_DEBUG_PRINTK(("vfc%d: User passed bogus pointer to "
-					"vfc_get_video_ioctl\n",
-					dev->instance));
-		return ret;
-	}
-	return ret;
-}
-
-static int vfc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
-	      unsigned long arg) 
-{
-	int ret = 0;
-	unsigned int tmp;
-	struct vfc_dev *dev;
-	void __user *argp = (void __user *)arg;
-
-	dev = vfc_get_dev_ptr(iminor(inode));
-	if(dev == NULL)
-		return -ENODEV;
-	
-	switch(cmd & 0x0000ffff) {
-	case VFCGCTRL:
-#if 0
-		VFC_IOCTL_DEBUG_PRINTK(("vfc%d: IOCTL(VFCGCTRL)\n", dev->instance));
-#endif
-		tmp = sbus_readl(&dev->regs->control);
-		if(copy_to_user(argp, &tmp, sizeof(unsigned int))) {
-			ret = -EFAULT;
-			break;
-		}
-		ret = 0;
-		break;
-	case VFCSCTRL:
-		ret = vfc_set_control_ioctl(inode, file, dev, arg);
-		break;
-	case VFCGVID:
-		ret = vfc_get_video_ioctl(inode, file, dev, arg);
-		break;
-	case VFCSVID:
-		ret = vfc_set_video_ioctl(inode, file, dev, arg);
-		break;
-	case VFCHUE:
-		VFC_IOCTL_DEBUG_PRINTK(("vfc%d: IOCTL(VFCHUE)\n", dev->instance));
-		if(copy_from_user(&tmp,argp,sizeof(unsigned int))) {
-			VFC_IOCTL_DEBUG_PRINTK(("vfc%d: User passed bogus pointer "
-						"to IOCTL(VFCHUE)", dev->instance));
-			ret = -EFAULT;
-		} else {
-			VFC_SAA9051_SA(dev,VFC_SAA9051_HUE) = tmp;
-			vfc_update_saa9051(dev);
-			ret = 0;
-		}
-		break;
-	case VFCPORTCHG:
-		ret = vfc_port_change_ioctl(inode, file, dev, arg);
-		break;
-	case VFCRDINFO:
-		ret = -EINVAL;
-		VFC_IOCTL_DEBUG_PRINTK(("vfc%d: IOCTL(VFCRDINFO)\n", dev->instance));
-		break;
-	default:
-		ret = vfc_debug(vfc_get_dev_ptr(iminor(inode)), cmd, argp);
-		break;
-	};
-
-	return ret;
-}
-
-static int vfc_mmap(struct file *file, struct vm_area_struct *vma) 
-{
-	unsigned int map_size, ret, map_offset;
-	struct vfc_dev *dev;
-	
-	dev = vfc_get_dev_ptr(iminor(file->f_path.dentry->d_inode));
-	if(dev == NULL)
-		return -ENODEV;
-
-	map_size = vma->vm_end - vma->vm_start;
-	if(map_size > sizeof(struct vfc_regs)) 
-		map_size = sizeof(struct vfc_regs);
-
-	vma->vm_flags |=
-		(VM_MAYREAD | VM_MAYWRITE | VM_MAYSHARE);
-	map_offset = (unsigned int) (long)dev->phys_regs;
-	ret = io_remap_pfn_range(vma, vma->vm_start,
-				  MK_IOSPACE_PFN(dev->which_io,
-					map_offset >> PAGE_SHIFT),
-				  map_size, vma->vm_page_prot);
-
-	if(ret)
-		return -EAGAIN;
-
-	return 0;
-}
-
-
-static const struct file_operations vfc_fops = {
-	.owner =	THIS_MODULE,
-	.llseek =	no_llseek,
-	.ioctl =	vfc_ioctl,
-	.mmap =		vfc_mmap,
-	.open =		vfc_open,
-	.release =	vfc_release,
-};
-
-static int vfc_probe(void)
-{
-	struct sbus_bus *sbus;
-	struct sbus_dev *sdev = NULL;
-	int ret;
-	int instance = 0, cards = 0;
-
-	for_all_sbusdev(sdev, sbus) {
-		if (strcmp(sdev->prom_name, "vfc") == 0) {
-			cards++;
-			continue;
-		}
-	}
-
-	if (!cards)
-		return -ENODEV;
-
-	vfc_dev_lst = kcalloc(cards + 1, sizeof(struct vfc_dev*), GFP_KERNEL);
-	if (vfc_dev_lst == NULL)
-		return -ENOMEM;
-	vfc_dev_lst[cards] = NULL;
-
-	ret = register_chrdev(VFC_MAJOR, vfcstr, &vfc_fops);
-	if(ret) {
-		printk(KERN_ERR "Unable to get major number %d\n", VFC_MAJOR);
-		kfree(vfc_dev_lst);
-		return -EIO;
-	}
-	instance = 0;
-	for_all_sbusdev(sdev, sbus) {
-		if (strcmp(sdev->prom_name, "vfc") == 0) {
-			vfc_dev_lst[instance]=(struct vfc_dev *)
-				kmalloc(sizeof(struct vfc_dev), GFP_KERNEL);
-			if (vfc_dev_lst[instance] == NULL)
-				return -ENOMEM;
-			ret = init_vfc_device(sdev,
-					      vfc_dev_lst[instance],
-					      instance);
-			if(ret) {
-				printk(KERN_ERR "Unable to initialize"
-				       " vfc%d device\n",
-				       instance);
-			} else {
-			}
-		
-			instance++;
-			continue;
-		}
-	}
-
-	return 0;
-}
-
-#ifdef MODULE
-int init_module(void)
-#else 
-int vfc_init(void)
-#endif
-{
-	return vfc_probe();
-}
-
-#ifdef MODULE
-static void deinit_vfc_device(struct vfc_dev *dev)
-{
-	if(dev == NULL)
-		return;
-	sbus_iounmap(dev->regs, sizeof(struct vfc_regs));
-	kfree(dev);
-}
-
-void cleanup_module(void)
-{
-	struct vfc_dev **devp;
-
-	unregister_chrdev(VFC_MAJOR,vfcstr);
-
-	for (devp = vfc_dev_lst; *devp; devp++)
-		deinit_vfc_device(*devp);
-
-	kfree(vfc_dev_lst);
-	return;
-}
-#endif
-
-MODULE_LICENSE("GPL");
-
diff --git a/drivers/sbus/char/vfc_i2c.c b/drivers/sbus/char/vfc_i2c.c
deleted file mode 100644
index 32b986e0ed78..000000000000
--- a/drivers/sbus/char/vfc_i2c.c
+++ /dev/null
@@ -1,335 +0,0 @@
-/*
- * drivers/sbus/char/vfc_i2c.c
- *
- * Driver for the Videopix Frame Grabber.
- * 
- * Functions that support the Phillips i2c(I squared C) bus on the vfc
- *  Documentation for the Phillips I2C bus can be found on the 
- *  phillips home page
- *
- * Copyright (C) 1996 Manish Vachharajani (mvachhar@noc.rutgers.edu)
- *
- */
-
-/* NOTE: It seems to me that the documentation regarding the
-pcd8584t/pcf8584 does not show the correct way to address the i2c bus.
-Based on the information on the I2C bus itself and the remainder of
-the Phillips docs the following algorithms appear to be correct.  I am
-fairly certain that the flowcharts in the phillips docs are wrong. */
-
-
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/slab.h>
-#include <linux/errno.h>
-#include <linux/sched.h>
-#include <linux/wait.h>
-#include <linux/delay.h>
-#include <asm/openprom.h>
-#include <asm/oplib.h>
-#include <asm/io.h>
-#include <asm/system.h>
-#include <asm/sbus.h>
-
-#if 0 
-#define VFC_I2C_DEBUG
-#endif
-
-#include "vfc.h"
-#include "vfc_i2c.h"
-
-#define WRITE_S1(__val) \
-	sbus_writel(__val, &dev->regs->i2c_s1)
-#define WRITE_REG(__val) \
-	sbus_writel(__val, &dev->regs->i2c_reg)
-
-#define VFC_I2C_READ (0x1)
-#define VFC_I2C_WRITE (0x0)
-     
-/****** 
-  The i2c bus controller chip on the VFC is a pcd8584t, but
-  phillips claims it doesn't exist.  As far as I can tell it is
-  identical to the PCF8584 so I treat it like it is the pcf8584.
-  
-  NOTE: The pcf8584 only cares
-  about the msb of the word you feed it 
-*****/
-
-int vfc_pcf8584_init(struct vfc_dev *dev) 
-{
-	/* This will also choose register S0_OWN so we can set it. */
-	WRITE_S1(RESET);
-
-	/* The pcf8584 shifts this value left one bit and uses
-	 * it as its i2c bus address.
-	 */
-	WRITE_REG(0x55000000);
-
-	/* This will set the i2c bus at the same speed sun uses,
-	 * and set another magic bit.
-	 */
-	WRITE_S1(SELECT(S2));
-	WRITE_REG(0x14000000);
-	
-	/* Enable the serial port, idle the i2c bus and set
-	 * the data reg to s0.
-	 */
-	WRITE_S1(CLEAR_I2C_BUS);
-	udelay(100);
-	return 0;
-}
-
-void vfc_i2c_delay_no_busy(struct vfc_dev *dev, unsigned long usecs) 
-{
-	schedule_timeout_uninterruptible(usecs_to_jiffies(usecs));
-}
-
-void inline vfc_i2c_delay(struct vfc_dev *dev) 
-{ 
-	vfc_i2c_delay_no_busy(dev, 100);
-}
-
-int vfc_init_i2c_bus(struct vfc_dev *dev)
-{
-	WRITE_S1(ENABLE_SERIAL | SELECT(S0) | ACK);
-	vfc_i2c_reset_bus(dev);
-	return 0;
-}
-
-int vfc_i2c_reset_bus(struct vfc_dev *dev) 
-{
-	VFC_I2C_DEBUG_PRINTK((KERN_DEBUG "vfc%d: Resetting the i2c bus\n",
-			      dev->instance));
-	if(dev == NULL)
-		return -EINVAL;
-	if(dev->regs == NULL)
-		return -EINVAL;
-	WRITE_S1(SEND_I2C_STOP);
-	WRITE_S1(SEND_I2C_STOP | ACK);
-	vfc_i2c_delay(dev);
-	WRITE_S1(CLEAR_I2C_BUS);
-	VFC_I2C_DEBUG_PRINTK((KERN_DEBUG "vfc%d: I2C status %x\n",
-			      dev->instance,
-			      sbus_readl(&dev->regs->i2c_s1)));
-	return 0;
-}
-
-static int vfc_i2c_wait_for_bus(struct vfc_dev *dev)
-{
-	int timeout = 1000; 
-
-	while(!(sbus_readl(&dev->regs->i2c_s1) & BB)) {
-		if(!(timeout--))
-			return -ETIMEDOUT;
-		vfc_i2c_delay(dev);
-	}
-	return 0;
-}
-
-static int vfc_i2c_wait_for_pin(struct vfc_dev *dev, int ack)
-{
-	int timeout = 1000; 
-	int s1;
-
-	while ((s1 = sbus_readl(&dev->regs->i2c_s1)) & PIN) {
-		if (!(timeout--))
-			return -ETIMEDOUT;
-		vfc_i2c_delay(dev);
-	}
-	if (ack == VFC_I2C_ACK_CHECK) {
-		if(s1 & LRB)
-			return -EIO; 
-	}
-	return 0;
-}
-
-#define SHIFT(a) ((a) << 24)
-static int vfc_i2c_xmit_addr(struct vfc_dev *dev, unsigned char addr,
-			     char mode)
-{ 
-	int ret, raddr;
-#if 1
-	WRITE_S1(SEND_I2C_STOP | ACK);
-	WRITE_S1(SELECT(S0) | ENABLE_SERIAL);
-	vfc_i2c_delay(dev);
-#endif
-
-	switch(mode) {
-	case VFC_I2C_READ:
-		raddr = SHIFT(((unsigned int)addr | 0x1));
-		WRITE_REG(raddr);
-		VFC_I2C_DEBUG_PRINTK(("vfc%d: receiving from i2c addr 0x%x\n",
-				      dev->instance, addr | 0x1));
-		break;
-	case VFC_I2C_WRITE:
-		raddr = SHIFT((unsigned int)addr & ~0x1);
-		WRITE_REG(raddr);
-		VFC_I2C_DEBUG_PRINTK(("vfc%d: sending to i2c addr 0x%x\n",
-				      dev->instance, addr & ~0x1));
-		break;
-	default:
-		return -EINVAL;
-	};
-
-	WRITE_S1(SEND_I2C_START);
-	vfc_i2c_delay(dev);
-	ret = vfc_i2c_wait_for_pin(dev,VFC_I2C_ACK_CHECK); /* We wait
-							      for the
-							      i2c send
-							      to finish
-							      here but
-							      Sun
-							      doesn't,
-							      hmm */
-	if (ret) {
-		printk(KERN_ERR "vfc%d: VFC xmit addr timed out or no ack\n",
-		       dev->instance);
-		return ret;
-	} else if (mode == VFC_I2C_READ) {
-		if ((ret = sbus_readl(&dev->regs->i2c_reg) & 0xff000000) != raddr) {
-			printk(KERN_WARNING 
-			       "vfc%d: returned slave address "
-			       "mismatch(%x,%x)\n",
-			       dev->instance, raddr, ret);
-		}
-	}	
-	return 0;
-}
-
-static int vfc_i2c_xmit_byte(struct vfc_dev *dev,unsigned char *byte)
-{
-	int ret;
-	u32 val = SHIFT((unsigned int)*byte);
-
-	WRITE_REG(val);
-
-	ret = vfc_i2c_wait_for_pin(dev, VFC_I2C_ACK_CHECK); 
-	switch(ret) {
-	case -ETIMEDOUT: 
-		printk(KERN_ERR "vfc%d: VFC xmit byte timed out or no ack\n",
-		       dev->instance);
-		break;
-	case -EIO:
-		ret = XMIT_LAST_BYTE;
-		break;
-	default:
-		break;
-	};
-
-	return ret;
-}
-
-static int vfc_i2c_recv_byte(struct vfc_dev *dev, unsigned char *byte,
-			     int last)
-{
-	int ret;
-
-	if (last) {
-		WRITE_REG(NEGATIVE_ACK);
-		VFC_I2C_DEBUG_PRINTK(("vfc%d: sending negative ack\n",
-				      dev->instance));
-	} else {
-		WRITE_S1(ACK);
-	}
-
-	ret = vfc_i2c_wait_for_pin(dev, VFC_I2C_NO_ACK_CHECK);
-	if(ret) {
-		printk(KERN_ERR "vfc%d: "
-		       "VFC recv byte timed out\n",
-		       dev->instance);
-	}
-	*byte = (sbus_readl(&dev->regs->i2c_reg)) >> 24;
-	return ret;
-}
-
-int vfc_i2c_recvbuf(struct vfc_dev *dev, unsigned char addr,
-		    char *buf, int count)
-{
-	int ret, last;
-
-	if(!(count && buf && dev && dev->regs) )
-		return -EINVAL;
-
-	if ((ret = vfc_i2c_wait_for_bus(dev))) {
-		printk(KERN_ERR "vfc%d: VFC I2C bus busy\n", dev->instance);
-		return ret;
-	}
-
-	if ((ret = vfc_i2c_xmit_addr(dev, addr, VFC_I2C_READ))) {
-		WRITE_S1(SEND_I2C_STOP);
-		vfc_i2c_delay(dev);
-		return ret;
-	}
-	
-	last = 0;
-	while (count--) {
-		if (!count)
-			last = 1;
-		if ((ret = vfc_i2c_recv_byte(dev, buf, last))) {
-			printk(KERN_ERR "vfc%d: "
-			       "VFC error while receiving byte\n",
-			       dev->instance);
-			WRITE_S1(SEND_I2C_STOP);
-			ret = -EINVAL;
-		}
-		buf++;
-	}
-	WRITE_S1(SEND_I2C_STOP | ACK);
-	vfc_i2c_delay(dev);
-	return ret;
-}
-
-int vfc_i2c_sendbuf(struct vfc_dev *dev, unsigned char addr, 
-		    char *buf, int count) 
-{
-	int ret;
-	
-	if (!(buf && dev && dev->regs))
-		return -EINVAL;
-	
-	if ((ret = vfc_i2c_wait_for_bus(dev))) {
-		printk(KERN_ERR "vfc%d: VFC I2C bus busy\n", dev->instance);
-		return ret;
-	}
-	
-	if ((ret = vfc_i2c_xmit_addr(dev, addr, VFC_I2C_WRITE))) {
-		WRITE_S1(SEND_I2C_STOP);
-		vfc_i2c_delay(dev);
-		return ret;
-	}
-	
-	while(count--) {
-		ret = vfc_i2c_xmit_byte(dev, buf);
-		switch(ret) {
-		case XMIT_LAST_BYTE:
-			VFC_I2C_DEBUG_PRINTK(("vfc%d: "
-					      "Receiver ended transmission with "
-					      " %d bytes remaining\n",
-					      dev->instance, count));
-			ret = 0;
-			goto done;
-			break;
-		case 0:
-			break;
-		default:
-			printk(KERN_ERR "vfc%d: "
-			       "VFC error while sending byte\n", dev->instance);
-			break;
-		};
-
-		buf++;
-	}
-done:
-	WRITE_S1(SEND_I2C_STOP | ACK);
-	vfc_i2c_delay(dev);
-	return ret;
-}
-
-
-
-
-
-
-
-
-
diff --git a/drivers/sbus/char/vfc_i2c.h b/drivers/sbus/char/vfc_i2c.h
deleted file mode 100644
index a2e6973209d5..000000000000
--- a/drivers/sbus/char/vfc_i2c.h
+++ /dev/null
@@ -1,44 +0,0 @@
-#ifndef _LINUX_VFC_I2C_H_
-#define _LINUX_VFC_I2C_H_
-
-/* control bits */
-#define PIN  (0x80000000)
-#define ESO  (0x40000000)
-#define ES1  (0x20000000)
-#define ES2  (0x10000000)
-#define ENI  (0x08000000)
-#define STA  (0x04000000)
-#define STO  (0x02000000)
-#define ACK  (0x01000000)
-
-/* status bits */
-#define STS  (0x20000000)
-#define BER  (0x10000000)
-#define LRB  (0x08000000)
-#define AAS  (0x04000000)
-#define LAB  (0x02000000)
-#define BB   (0x01000000)
-
-#define SEND_I2C_START (PIN | ESO | STA)
-#define SEND_I2C_STOP (PIN | ESO | STO)
-#define CLEAR_I2C_BUS (PIN | ESO | ACK)
-#define NEGATIVE_ACK ((ESO) & ~ACK)
-
-#define SELECT(a) (a)
-#define S0 (PIN | ESO | ES1)
-#define S0_OWN (PIN)
-#define S2 (PIN | ES1)
-#define S3 (PIN | ES2)
-
-#define ENABLE_SERIAL (PIN | ESO)
-#define DISABLE_SERIAL (PIN)
-#define RESET (PIN)
-
-#define XMIT_LAST_BYTE (1)
-#define VFC_I2C_ACK_CHECK (1)
-#define VFC_I2C_NO_ACK_CHECK (0)
-
-#endif /* _LINUX_VFC_I2C_H_ */
-
-
-
diff --git a/drivers/sbus/dvma.c b/drivers/sbus/dvma.c
deleted file mode 100644
index ab0d2de3324c..000000000000
--- a/drivers/sbus/dvma.c
+++ /dev/null
@@ -1,136 +0,0 @@
-/* dvma.c:  Routines that are used to access DMA on the Sparc SBus.
- *
- * Copyright (C) 1995 David S. Miller (davem@caip.rutgers.edu)
- */
-
-#include <linux/string.h>
-#include <linux/kernel.h>
-#include <linux/slab.h>
-#include <linux/init.h>
-#include <linux/delay.h>
-
-#include <asm/oplib.h>
-#include <asm/io.h>
-#include <asm/dma.h>
-#include <asm/sbus.h>
-
-struct sbus_dma *dma_chain;
-
-static void __init init_one_dvma(struct sbus_dma *dma, int num_dma)
-{
-	printk("dma%d: ", num_dma);
-	
-	dma->next = NULL;
-	dma->running = 0;      /* No transfers going on as of yet */
-	dma->allocated = 0;    /* No one has allocated us yet */
-	switch(sbus_readl(dma->regs + DMA_CSR)&DMA_DEVICE_ID) {
-	case DMA_VERS0:
-		dma->revision = dvmarev0;
-		printk("Revision 0 ");
-		break;
-	case DMA_ESCV1:
-		dma->revision = dvmaesc1;
-		printk("ESC Revision 1 ");
-		break;
-	case DMA_VERS1:
-		dma->revision = dvmarev1;
-		printk("Revision 1 ");
-		break;
-	case DMA_VERS2:
-		dma->revision = dvmarev2;
-		printk("Revision 2 ");
-		break;
-	case DMA_VERHME:
-		dma->revision = dvmahme;
-		printk("HME DVMA gate array ");
-		break;
-	case DMA_VERSPLUS:
-		dma->revision = dvmarevplus;
-		printk("Revision 1 PLUS ");
-		break;
-	default:
-		printk("unknown dma version %08x",
-		       sbus_readl(dma->regs + DMA_CSR) & DMA_DEVICE_ID);
-		dma->allocated = 1;
-		break;
-	}
-	printk("\n");
-}
-
-/* Probe this SBus DMA module(s) */
-void __init dvma_init(struct sbus_bus *sbus)
-{
-	struct sbus_dev *this_dev;
-	struct sbus_dma *dma;
-	struct sbus_dma *dchain;
-	static int num_dma = 0;
-
-	for_each_sbusdev(this_dev, sbus) {
-		char *name = this_dev->prom_name;
-		int hme = 0;
-
-		if(!strcmp(name, "SUNW,fas"))
-			hme = 1;
-		else if(strcmp(name, "dma") &&
-			strcmp(name, "ledma") &&
-			strcmp(name, "espdma"))
-			continue;
-
-		/* Found one... */
-		dma = kmalloc(sizeof(struct sbus_dma), GFP_ATOMIC);
-
-		dma->sdev = this_dev;
-
-		/* Put at end of dma chain */
-		dchain = dma_chain;
-		if(dchain) {
-			while(dchain->next)
-				dchain = dchain->next;
-			dchain->next = dma;
-		} else {
-			/* We're the first in line */
-			dma_chain = dma;
-		}
-
-		dma->regs = sbus_ioremap(&dma->sdev->resource[0], 0,
-					 dma->sdev->resource[0].end - dma->sdev->resource[0].start + 1,
-					 "dma");
-
-		dma->node = dma->sdev->prom_node;
-		
-		init_one_dvma(dma, num_dma++);
-	}
-}
-
-#ifdef CONFIG_SUN4
-
-#include <asm/sun4paddr.h>
-
-void __init sun4_dvma_init(void)
-{
-	struct sbus_dma *dma;
-	struct resource r;
-
-	if(sun4_dma_physaddr) {
-		dma = kmalloc(sizeof(struct sbus_dma), GFP_ATOMIC);
-
-		/* No SBUS */
-		dma->sdev = NULL;
-
-		/* Only one DMA device */
-		dma_chain = dma;
-
-		memset(&r, 0, sizeof(r));
-		r.start = sun4_dma_physaddr;
-		dma->regs = sbus_ioremap(&r, 0, PAGE_SIZE, "dma");
-
-		/* No prom node */
-		dma->node = 0x0;
-
-		init_one_dvma(dma, 0);
-	} else {
-	  	dma_chain = NULL;
-	}
-}
-
-#endif
diff --git a/drivers/sbus/sbus.c b/drivers/sbus/sbus.c
deleted file mode 100644
index 9c129248466c..000000000000
--- a/drivers/sbus/sbus.c
+++ /dev/null
@@ -1,316 +0,0 @@
-/* sbus.c: SBus support routines.
- *
- * Copyright (C) 1995, 2006 David S. Miller (davem@davemloft.net)
- */
-
-#include <linux/kernel.h>
-#include <linux/slab.h>
-#include <linux/init.h>
-#include <linux/device.h>
-#include <linux/of_device.h>
-
-#include <asm/system.h>
-#include <asm/sbus.h>
-#include <asm/dma.h>
-#include <asm/oplib.h>
-#include <asm/prom.h>
-#include <asm/bpp.h>
-#include <asm/irq.h>
-
-static ssize_t
-show_sbusobppath_attr(struct device * dev, struct device_attribute * attr, char * buf)
-{
-	struct sbus_dev *sbus;
-
-	sbus = to_sbus_device(dev);
-
-	return snprintf (buf, PAGE_SIZE, "%s\n", sbus->ofdev.node->full_name);
-}
-
-static DEVICE_ATTR(obppath, S_IRUSR | S_IRGRP | S_IROTH, show_sbusobppath_attr, NULL);
-
-struct sbus_bus *sbus_root;
-
-static void __init fill_sbus_device(struct device_node *dp, struct sbus_dev *sdev)
-{
-	struct dev_archdata *sd;
-	unsigned long base;
-	const void *pval;
-	int len, err;
-
-	sdev->prom_node = dp->node;
-	strcpy(sdev->prom_name, dp->name);
-
-	pval = of_get_property(dp, "reg", &len);
-	sdev->num_registers = 0;
-	if (pval) {
-		memcpy(sdev->reg_addrs, pval, len);
-
-		sdev->num_registers =
-			len / sizeof(struct linux_prom_registers);
-
-		base = (unsigned long) sdev->reg_addrs[0].phys_addr;
-
-		/* Compute the slot number. */
-		if (base >= SUN_SBUS_BVADDR && sparc_cpu_model == sun4m)
-			sdev->slot = sbus_dev_slot(base);
-		else
-			sdev->slot = sdev->reg_addrs[0].which_io;
-	}
-
-	pval = of_get_property(dp, "ranges", &len);
-	sdev->num_device_ranges = 0;
-	if (pval) {
-		memcpy(sdev->device_ranges, pval, len);
-		sdev->num_device_ranges =
-			len / sizeof(struct linux_prom_ranges);
-	}
-
-	sbus_fill_device_irq(sdev);
-
-	sd = &sdev->ofdev.dev.archdata;
-	sd->prom_node = dp;
-	sd->op = &sdev->ofdev;
-
-	sdev->ofdev.node = dp;
-	if (sdev->parent)
-		sdev->ofdev.dev.parent = &sdev->parent->ofdev.dev;
-	else
-		sdev->ofdev.dev.parent = &sdev->bus->ofdev.dev;
-	sdev->ofdev.dev.bus = &sbus_bus_type;
-	dev_set_name(&sdev->ofdev.dev, "sbus[%08x]", dp->node);
-
-	if (of_device_register(&sdev->ofdev) != 0)
-		printk(KERN_DEBUG "sbus: device registration error for %s!\n",
-		       dp->path_component_name);
-
-	/* WE HAVE BEEN INVADED BY ALIENS! */
-	err = sysfs_create_file(&sdev->ofdev.dev.kobj, &dev_attr_obppath.attr);
-}
-
-static void __init sbus_bus_ranges_init(struct device_node *dp, struct sbus_bus *sbus)
-{
-	const void *pval;
-	int len;
-
-	pval = of_get_property(dp, "ranges", &len);
-	sbus->num_sbus_ranges = 0;
-	if (pval) {
-		memcpy(sbus->sbus_ranges, pval, len);
-		sbus->num_sbus_ranges =
-			len / sizeof(struct linux_prom_ranges);
-
-		sbus_arch_bus_ranges_init(dp->parent, sbus);
-	}
-}
-
-static void __init __apply_ranges_to_regs(struct linux_prom_ranges *ranges,
-					  int num_ranges,
-					  struct linux_prom_registers *regs,
-					  int num_regs)
-{
-	if (num_ranges) {
-		int regnum;
-
-		for (regnum = 0; regnum < num_regs; regnum++) {
-			int rngnum;
-
-			for (rngnum = 0; rngnum < num_ranges; rngnum++) {
-				if (regs[regnum].which_io == ranges[rngnum].ot_child_space)
-					break;
-			}
-			if (rngnum == num_ranges) {
-				/* We used to flag this as an error.  Actually
-				 * some devices do not report the regs as we expect.
-				 * For example, see SUNW,pln device.  In that case
-				 * the reg property is in a format internal to that
-				 * node, ie. it is not in the SBUS register space
-				 * per se. -DaveM
-				 */
-				return;
-			}
-			regs[regnum].which_io = ranges[rngnum].ot_parent_space;
-			regs[regnum].phys_addr -= ranges[rngnum].ot_child_base;
-			regs[regnum].phys_addr += ranges[rngnum].ot_parent_base;
-		}
-	}
-}
-
-static void __init __fixup_regs_sdev(struct sbus_dev *sdev)
-{
-	if (sdev->num_registers != 0) {
-		struct sbus_dev *parent = sdev->parent;
-		int i;
-
-		while (parent != NULL) {
-			__apply_ranges_to_regs(parent->device_ranges,
-					       parent->num_device_ranges,
-					       sdev->reg_addrs,
-					       sdev->num_registers);
-
-			parent = parent->parent;
-		}
-
-		__apply_ranges_to_regs(sdev->bus->sbus_ranges,
-				       sdev->bus->num_sbus_ranges,
-				       sdev->reg_addrs,
-				       sdev->num_registers);
-
-		for (i = 0; i < sdev->num_registers; i++) {
-			struct resource *res = &sdev->resource[i];
-
-			res->start = sdev->reg_addrs[i].phys_addr;
-			res->end = (res->start +
-				    (unsigned long)sdev->reg_addrs[i].reg_size - 1UL);
-			res->flags = IORESOURCE_IO |
-				(sdev->reg_addrs[i].which_io & 0xff);
-		}
-	}
-}
-
-static void __init sbus_fixup_all_regs(struct sbus_dev *first_sdev)
-{
-	struct sbus_dev *sdev;
-
-	for (sdev = first_sdev; sdev; sdev = sdev->next) {
-		if (sdev->child)
-			sbus_fixup_all_regs(sdev->child);
-		__fixup_regs_sdev(sdev);
-	}
-}
-
-/* We preserve the "probe order" of these bus and device lists to give
- * the same ordering as the old code.
- */
-static void __init sbus_insert(struct sbus_bus *sbus, struct sbus_bus **root)
-{
-	while (*root)
-		root = &(*root)->next;
-	*root = sbus;
-	sbus->next = NULL;
-}
-
-static void __init sdev_insert(struct sbus_dev *sdev, struct sbus_dev **root)
-{
-	while (*root)
-		root = &(*root)->next;
-	*root = sdev;
-	sdev->next = NULL;
-}
-
-static void __init walk_children(struct device_node *dp, struct sbus_dev *parent, struct sbus_bus *sbus)
-{
-	dp = dp->child;
-	while (dp) {
-		struct sbus_dev *sdev;
-
-		sdev = kzalloc(sizeof(struct sbus_dev), GFP_ATOMIC);
-		if (sdev) {
-			sdev_insert(sdev, &parent->child);
-
-			sdev->bus = sbus;
-			sdev->parent = parent;
-			sdev->ofdev.dev.archdata.iommu =
-				sbus->ofdev.dev.archdata.iommu;
-			sdev->ofdev.dev.archdata.stc =
-				sbus->ofdev.dev.archdata.stc;
-
-			fill_sbus_device(dp, sdev);
-
-			walk_children(dp, sdev, sbus);
-		}
-		dp = dp->sibling;
-	}
-}
-
-static void __init build_one_sbus(struct device_node *dp, int num_sbus)
-{
-	struct sbus_bus *sbus;
-	unsigned int sbus_clock;
-	struct device_node *dev_dp;
-
-	sbus = kzalloc(sizeof(struct sbus_bus), GFP_ATOMIC);
-	if (!sbus)
-		return;
-
-	sbus_insert(sbus, &sbus_root);
-	sbus->prom_node = dp->node;
-
-	sbus_setup_iommu(sbus, dp);
-
-	printk("sbus%d: ", num_sbus);
-
-	sbus_clock = of_getintprop_default(dp, "clock-frequency",
-					   (25*1000*1000));
-	sbus->clock_freq = sbus_clock;
-
-	printk("Clock %d.%d MHz\n", (int) ((sbus_clock/1000)/1000),
-	       (int) (((sbus_clock/1000)%1000 != 0) ? 
-		      (((sbus_clock/1000)%1000) + 1000) : 0));
-
-	strcpy(sbus->prom_name, dp->name);
-
-	sbus_setup_arch_props(sbus, dp);
-
-	sbus_bus_ranges_init(dp, sbus);
-
-	sbus->ofdev.node = dp;
-	sbus->ofdev.dev.parent = NULL;
-	sbus->ofdev.dev.bus = &sbus_bus_type;
-	dev_set_name(&sbus->ofdev.dev, "sbus%d", num_sbus);
-
-	if (of_device_register(&sbus->ofdev) != 0)
-		printk(KERN_DEBUG "sbus: device registration error for %s!\n",
-		       dev_name(&sbus->ofdev.dev));
-
-	dev_dp = dp->child;
-	while (dev_dp) {
-		struct sbus_dev *sdev;
-
-		sdev = kzalloc(sizeof(struct sbus_dev), GFP_ATOMIC);
-		if (sdev) {
-			sdev_insert(sdev, &sbus->devices);
-
-			sdev->bus = sbus;
-			sdev->parent = NULL;
-			sdev->ofdev.dev.archdata.iommu =
-				sbus->ofdev.dev.archdata.iommu;
-			sdev->ofdev.dev.archdata.stc =
-				sbus->ofdev.dev.archdata.stc;
-
-			fill_sbus_device(dev_dp, sdev);
-
-			walk_children(dev_dp, sdev, sbus);
-		}
-		dev_dp = dev_dp->sibling;
-	}
-
-	sbus_fixup_all_regs(sbus->devices);
-
-	dvma_init(sbus);
-}
-
-static int __init sbus_init(void)
-{
-	struct device_node *dp;
-	const char *sbus_name = "sbus";
-	int num_sbus = 0;
-
-	if (sbus_arch_preinit())
-		return 0;
-
-	if (sparc_cpu_model == sun4d)
-		sbus_name = "sbi";
-
-	for_each_node_by_name(dp, sbus_name) {
-		build_one_sbus(dp, num_sbus);
-		num_sbus++;
-
-	}
-
-	sbus_arch_postinit();
-
-	return 0;
-}
-
-subsys_initcall(sbus_init);
diff --git a/drivers/scsi/esp_scsi.h b/drivers/scsi/esp_scsi.h
index bb43a1388188..28e22acf87ea 100644
--- a/drivers/scsi/esp_scsi.h
+++ b/drivers/scsi/esp_scsi.h
@@ -521,7 +521,8 @@ struct esp {
 
 	struct completion	*eh_reset;
 
-	struct sbus_dma		*dma;
+	void			*dma;
+	int			dmarev;
 };
 
 /* A front-end driver for the ESP chip should do the following in
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 81c16cba5417..90212ac33be3 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -40,7 +40,6 @@
 #include <linux/ioport.h>
 #include <linux/blkdev.h>
 #include <linux/errno.h>
-#include <linux/hdreg.h>
 #include <linux/slab.h>
 #include <linux/ide.h>
 #include <linux/scatterlist.h>
@@ -131,50 +130,6 @@ static inline idescsi_scsi_t *drive_to_idescsi(ide_drive_t *ide_drive)
 	return scsihost_to_idescsi(ide_drive->driver_data);
 }
 
-/*
- *	PIO data transfer routine using the scatter gather table.
- */
-static void ide_scsi_io_buffers(ide_drive_t *drive, struct ide_atapi_pc *pc,
-				unsigned int bcount, int write)
-{
-	ide_hwif_t *hwif = drive->hwif;
-	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
-	xfer_func_t *xf = write ? tp_ops->output_data : tp_ops->input_data;
-	char *buf;
-	int count;
-
-	while (bcount) {
-		count = min(pc->sg->length - pc->b_count, bcount);
-		if (PageHighMem(sg_page(pc->sg))) {
-			unsigned long flags;
-
-			local_irq_save(flags);
-			buf = kmap_atomic(sg_page(pc->sg), KM_IRQ0) +
-					  pc->sg->offset;
-			xf(drive, NULL, buf + pc->b_count, count);
-			kunmap_atomic(buf - pc->sg->offset, KM_IRQ0);
-			local_irq_restore(flags);
-		} else {
-			buf = sg_virt(pc->sg);
-			xf(drive, NULL, buf + pc->b_count, count);
-		}
-		bcount -= count; pc->b_count += count;
-		if (pc->b_count == pc->sg->length) {
-			if (!--pc->sg_cnt)
-				break;
-			pc->sg = sg_next(pc->sg);
-			pc->b_count = 0;
-		}
-	}
-
-	if (bcount) {
-		printk(KERN_ERR "%s: scatter gather table too small, %s\n",
-				drive->name, write ? "padding with zeros"
-						   : "discarding data");
-		ide_pad_transfer(drive, write, bcount);
-	}
-}
-
 static void ide_scsi_hex_dump(u8 *data, int len)
 {
 	print_hex_dump(KERN_CONT, "", DUMP_PREFIX_NONE, 16, 1, data, len, 0);
@@ -244,9 +199,9 @@ idescsi_atapi_error(ide_drive_t *drive, struct request *rq, u8 stat, u8 err)
 {
 	ide_hwif_t *hwif = drive->hwif;
 
-	if (hwif->tp_ops->read_status(hwif) & (BUSY_STAT | DRQ_STAT))
+	if (hwif->tp_ops->read_status(hwif) & (ATA_BUSY | ATA_DRQ))
 		/* force an abort */
-		hwif->tp_ops->exec_command(hwif, WIN_IDLEIMMEDIATE);
+		hwif->tp_ops->exec_command(hwif, ATA_CMD_IDLEIMMEDIATE);
 
 	rq->errors++;
 
@@ -344,7 +299,7 @@ static ide_startstop_t idescsi_pc_intr (ide_drive_t *drive)
 
 	return ide_pc_intr(drive, pc, idescsi_pc_intr, get_timeout(pc),
 			   idescsi_expiry, NULL, NULL, NULL,
-			   ide_scsi_io_buffers);
+			   ide_io_buffers);
 }
 
 static ide_startstop_t idescsi_transfer_pc(ide_drive_t *drive)
@@ -430,21 +385,41 @@ static ide_startstop_t idescsi_do_request (ide_drive_t *drive, struct request *r
 }
 
 #ifdef CONFIG_IDE_PROC_FS
-static void idescsi_add_settings(ide_drive_t *drive)
-{
-	idescsi_scsi_t *scsi = drive_to_idescsi(drive);
-
-/*
- *			drive	setting name	read/write	data type	min	max	mul_factor	div_factor	data pointer		set function
- */
-	ide_add_setting(drive,	"bios_cyl",	SETTING_RW,	TYPE_INT,	0,	1023,	1,		1,		&drive->bios_cyl,	NULL);
-	ide_add_setting(drive,	"bios_head",	SETTING_RW,	TYPE_BYTE,	0,	255,	1,		1,		&drive->bios_head,	NULL);
-	ide_add_setting(drive,	"bios_sect",	SETTING_RW,	TYPE_BYTE,	0,	63,	1,		1,		&drive->bios_sect,	NULL);
-	ide_add_setting(drive,	"transform",	SETTING_RW,	TYPE_INT,	0,	3,	1,		1,		&scsi->transform,	NULL);
-	ide_add_setting(drive,	"log",		SETTING_RW,	TYPE_INT,	0,	1,	1,		1,		&scsi->log,		NULL);
-}
-#else
-static inline void idescsi_add_settings(ide_drive_t *drive) { ; }
+#define ide_scsi_devset_get(name, field) \
+static int get_##name(ide_drive_t *drive) \
+{ \
+	idescsi_scsi_t *scsi = drive_to_idescsi(drive); \
+	return scsi->field; \
+}
+
+#define ide_scsi_devset_set(name, field) \
+static int set_##name(ide_drive_t *drive, int arg) \
+{ \
+	idescsi_scsi_t *scsi = drive_to_idescsi(drive); \
+	scsi->field = arg; \
+	return 0; \
+}
+
+#define ide_scsi_devset_rw_field(_name, _field) \
+ide_scsi_devset_get(_name, _field); \
+ide_scsi_devset_set(_name, _field); \
+IDE_DEVSET(_name, DS_SYNC, get_##_name, set_##_name);
+
+ide_devset_rw_field(bios_cyl, bios_cyl);
+ide_devset_rw_field(bios_head, bios_head);
+ide_devset_rw_field(bios_sect, bios_sect);
+
+ide_scsi_devset_rw_field(transform, transform);
+ide_scsi_devset_rw_field(log, log);
+
+static const struct ide_proc_devset idescsi_settings[] = {
+	IDE_PROC_DEVSET(bios_cyl,  0, 1023),
+	IDE_PROC_DEVSET(bios_head, 0,  255),
+	IDE_PROC_DEVSET(bios_sect, 0,	63),
+	IDE_PROC_DEVSET(log,	   0,	 1),
+	IDE_PROC_DEVSET(transform, 0,	 3),
+	{ 0 },
+};
 #endif
 
 /*
@@ -452,7 +427,7 @@ static inline void idescsi_add_settings(ide_drive_t *drive) { ; }
  */
 static void idescsi_setup (ide_drive_t *drive, idescsi_scsi_t *scsi)
 {
-	if (drive->id && (drive->id->config & 0x0060) == 0x20)
+	if ((drive->id[ATA_ID_CONFIG] & 0x0060) == 0x20)
 		set_bit(IDE_AFLAG_DRQ_INTERRUPT, &drive->atapi_flags);
 	clear_bit(IDESCSI_SG_TRANSFORM, &scsi->transform);
 #if IDESCSI_DEBUG_LOG
@@ -461,7 +436,7 @@ static void idescsi_setup (ide_drive_t *drive, idescsi_scsi_t *scsi)
 
 	drive->pc_callback = ide_scsi_callback;
 
-	idescsi_add_settings(drive);
+	ide_proc_register_driver(drive, scsi->driver);
 }
 
 static void ide_scsi_remove(ide_drive_t *drive)
@@ -503,12 +478,12 @@ static ide_driver_t idescsi_driver = {
 	.remove			= ide_scsi_remove,
 	.version		= IDESCSI_VERSION,
 	.media			= ide_scsi,
-	.supports_dsc_overlap	= 0,
 	.do_request		= idescsi_do_request,
 	.end_request		= idescsi_end_request,
 	.error                  = idescsi_atapi_error,
 #ifdef CONFIG_IDE_PROC_FS
 	.proc			= idescsi_proc,
+	.settings		= idescsi_settings,
 #endif
 };
 
@@ -811,6 +786,7 @@ static int ide_scsi_probe(ide_drive_t *drive)
 	struct gendisk *g;
 	static int warned;
 	int err = -ENOMEM;
+	u16 last_lun;
 
 	if (!warned && drive->media == ide_cdrom) {
 		printk(KERN_WARNING "ide-scsi is deprecated for cd burning! Use ide-cd and give dev=/dev/hdX as device\n");
@@ -821,7 +797,6 @@ static int ide_scsi_probe(ide_drive_t *drive)
 		return -ENODEV;
 
 	if (!strstr("ide-scsi", drive->driver_req) ||
-	    !drive->present ||
 	    drive->media == ide_disk ||
 	    !(host = scsi_host_alloc(&idescsi_template,sizeof(idescsi_scsi_t))))
 		return -ENODEV;
@@ -836,12 +811,12 @@ static int ide_scsi_probe(ide_drive_t *drive)
 
 	host->max_id = 1;
 
-	if (drive->id->last_lun)
-		debug_log("%s: id->last_lun=%u\n", drive->name,
-			  drive->id->last_lun);
+	last_lun = drive->id[ATA_ID_LAST_LUN];
+	if (last_lun)
+		debug_log("%s: last_lun=%u\n", drive->name, last_lun);
 
-	if ((drive->id->last_lun & 0x7) != 7)
-		host->max_lun = (drive->id->last_lun & 0x7) + 1;
+	if ((last_lun & 7) != 7)
+		host->max_lun = (last_lun & 7) + 1;
 	else
 		host->max_lun = 1;
 
@@ -852,7 +827,6 @@ static int ide_scsi_probe(ide_drive_t *drive)
 	idescsi->host = host;
 	idescsi->disk = g;
 	g->private_data = &idescsi->driver;
-	ide_proc_register_driver(drive, &idescsi_driver);
 	err = 0;
 	idescsi_setup(drive, idescsi);
 	g->fops = &idescsi_ops;
diff --git a/drivers/scsi/qlogicpti.c b/drivers/scsi/qlogicpti.c
index 905350896725..69d6ad862b60 100644
--- a/drivers/scsi/qlogicpti.c
+++ b/drivers/scsi/qlogicpti.c
@@ -1,6 +1,6 @@
 /* qlogicpti.c: Performance Technologies QlogicISP sbus card driver.
  *
- * Copyright (C) 1996, 2006 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 1996, 2006, 2008 David S. Miller (davem@davemloft.net)
  *
  * A lot of this driver was directly stolen from Erik H. Moe's PCI
  * Qlogic ISP driver.  Mucho kudos to him for this code.
@@ -25,12 +25,14 @@
 #include <linux/interrupt.h>
 #include <linux/module.h>
 #include <linux/jiffies.h>
+#include <linux/dma-mapping.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/byteorder.h>
 
 #include "qlogicpti.h"
 
-#include <asm/sbus.h>
 #include <asm/dma.h>
 #include <asm/system.h>
 #include <asm/ptrace.h>
@@ -157,7 +159,7 @@ static inline void set_sbus_cfg1(struct qlogicpti *qpti)
 	 * is a nop and the chip ends up using the smallest burst
 	 * size. -DaveM
 	 */
-	if (sbus_can_burst64(qpti->sdev) && (bursts & DMA_BURST64)) {
+	if (sbus_can_burst64() && (bursts & DMA_BURST64)) {
 		val = (SBUS_CFG1_BENAB | SBUS_CFG1_B64);
 	} else
 #endif
@@ -684,19 +686,19 @@ static void __devexit qpti_chain_del(struct qlogicpti *qpti)
 
 static int __devinit qpti_map_regs(struct qlogicpti *qpti)
 {
-	struct sbus_dev *sdev = qpti->sdev;
+	struct of_device *op = qpti->op;
 
-	qpti->qregs = sbus_ioremap(&sdev->resource[0], 0,
-				   sdev->reg_addrs[0].reg_size,
-				   "PTI Qlogic/ISP");
+	qpti->qregs = of_ioremap(&op->resource[0], 0,
+				 resource_size(&op->resource[0]),
+				 "PTI Qlogic/ISP");
 	if (!qpti->qregs) {
 		printk("PTI: Qlogic/ISP registers are unmappable\n");
 		return -1;
 	}
 	if (qpti->is_pti) {
-		qpti->sreg = sbus_ioremap(&sdev->resource[0], (16 * 4096),
-					  sizeof(unsigned char),
-					  "PTI Qlogic/ISP statreg");
+		qpti->sreg = of_ioremap(&op->resource[0], (16 * 4096),
+					sizeof(unsigned char),
+					"PTI Qlogic/ISP statreg");
 		if (!qpti->sreg) {
 			printk("PTI: Qlogic/ISP status register is unmappable\n");
 			return -1;
@@ -707,9 +709,9 @@ static int __devinit qpti_map_regs(struct qlogicpti *qpti)
 
 static int __devinit qpti_register_irq(struct qlogicpti *qpti)
 {
-	struct sbus_dev *sdev = qpti->sdev;
+	struct of_device *op = qpti->op;
 
-	qpti->qhost->irq = qpti->irq = sdev->irqs[0];
+	qpti->qhost->irq = qpti->irq = op->irqs[0];
 
 	/* We used to try various overly-clever things to
 	 * reduce the interrupt processing overhead on
@@ -732,17 +734,19 @@ fail:
 
 static void __devinit qpti_get_scsi_id(struct qlogicpti *qpti)
 {
-	qpti->scsi_id = prom_getintdefault(qpti->prom_node,
-					   "initiator-id",
-					   -1);
+	struct of_device *op = qpti->op;
+	struct device_node *dp;
+
+	dp = op->node;
+
+	qpti->scsi_id = of_getintprop_default(dp, "initiator-id", -1);
 	if (qpti->scsi_id == -1)
-		qpti->scsi_id = prom_getintdefault(qpti->prom_node,
-						   "scsi-initiator-id",
-						   -1);
+		qpti->scsi_id = of_getintprop_default(dp, "scsi-initiator-id",
+						      -1);
 	if (qpti->scsi_id == -1)
 		qpti->scsi_id =
-			prom_getintdefault(qpti->sdev->bus->prom_node,
-					   "scsi-initiator-id", 7);
+			of_getintprop_default(dp->parent,
+					      "scsi-initiator-id", 7);
 	qpti->qhost->this_id = qpti->scsi_id;
 	qpti->qhost->max_sectors = 64;
 
@@ -751,12 +755,11 @@ static void __devinit qpti_get_scsi_id(struct qlogicpti *qpti)
 
 static void qpti_get_bursts(struct qlogicpti *qpti)
 {
-	struct sbus_dev *sdev = qpti->sdev;
+	struct of_device *op = qpti->op;
 	u8 bursts, bmask;
 
-	bursts = prom_getintdefault(qpti->prom_node, "burst-sizes", 0xff);
-	bmask = prom_getintdefault(sdev->bus->prom_node,
-				   "burst-sizes", 0xff);
+	bursts = of_getintprop_default(op->node, "burst-sizes", 0xff);
+	bmask = of_getintprop_default(op->node->parent, "burst-sizes", 0xff);
 	if (bmask != 0xff)
 		bursts &= bmask;
 	if (bursts == 0xff ||
@@ -785,25 +788,25 @@ static void qpti_get_clock(struct qlogicpti *qpti)
  */
 static int __devinit qpti_map_queues(struct qlogicpti *qpti)
 {
-	struct sbus_dev *sdev = qpti->sdev;
+	struct of_device *op = qpti->op;
 
 #define QSIZE(entries)	(((entries) + 1) * QUEUE_ENTRY_LEN)
-	qpti->res_cpu = sbus_alloc_consistent(sdev,
-					      QSIZE(RES_QUEUE_LEN),
-					      &qpti->res_dvma);
+	qpti->res_cpu = dma_alloc_coherent(&op->dev,
+					   QSIZE(RES_QUEUE_LEN),
+					   &qpti->res_dvma, GFP_ATOMIC);
 	if (qpti->res_cpu == NULL ||
 	    qpti->res_dvma == 0) {
 		printk("QPTI: Cannot map response queue.\n");
 		return -1;
 	}
 
-	qpti->req_cpu = sbus_alloc_consistent(sdev,
-					      QSIZE(QLOGICPTI_REQ_QUEUE_LEN),
-					      &qpti->req_dvma);
+	qpti->req_cpu = dma_alloc_coherent(&op->dev,
+					   QSIZE(QLOGICPTI_REQ_QUEUE_LEN),
+					   &qpti->req_dvma, GFP_ATOMIC);
 	if (qpti->req_cpu == NULL ||
 	    qpti->req_dvma == 0) {
-		sbus_free_consistent(sdev, QSIZE(RES_QUEUE_LEN),
-				     qpti->res_cpu, qpti->res_dvma);
+		dma_free_coherent(&op->dev, QSIZE(RES_QUEUE_LEN),
+				  qpti->res_cpu, qpti->res_dvma);
 		printk("QPTI: Cannot map request queue.\n");
 		return -1;
 	}
@@ -875,8 +878,9 @@ static inline int load_cmd(struct scsi_cmnd *Cmnd, struct Command_Entry *cmd,
 		int sg_count;
 
 		sg = scsi_sglist(Cmnd);
-		sg_count = sbus_map_sg(qpti->sdev, sg, scsi_sg_count(Cmnd),
-		                                      Cmnd->sc_data_direction);
+		sg_count = dma_map_sg(&qpti->op->dev, sg,
+				      scsi_sg_count(Cmnd),
+				      Cmnd->sc_data_direction);
 
 		ds = cmd->dataseg;
 		cmd->segment_cnt = sg_count;
@@ -1152,9 +1156,9 @@ static struct scsi_cmnd *qlogicpti_intr_handler(struct qlogicpti *qpti)
 			Cmnd->result = DID_ERROR << 16;
 
 		if (scsi_bufflen(Cmnd))
-			sbus_unmap_sg(qpti->sdev,
-				      scsi_sglist(Cmnd), scsi_sg_count(Cmnd),
-				      Cmnd->sc_data_direction);
+			dma_unmap_sg(&qpti->op->dev,
+				     scsi_sglist(Cmnd), scsi_sg_count(Cmnd),
+				     Cmnd->sc_data_direction);
 
 		qpti->cmd_count[Cmnd->device->id]--;
 		sbus_writew(out_ptr, qpti->qregs + MBOX5);
@@ -1268,34 +1272,32 @@ static struct scsi_host_template qpti_template = {
 	.use_clustering		= ENABLE_CLUSTERING,
 };
 
-static int __devinit qpti_sbus_probe(struct of_device *dev, const struct of_device_id *match)
+static int __devinit qpti_sbus_probe(struct of_device *op, const struct of_device_id *match)
 {
-	static int nqptis;
-	struct sbus_dev *sdev = to_sbus_device(&dev->dev);
-	struct device_node *dp = dev->node;
 	struct scsi_host_template *tpnt = match->data;
+	struct device_node *dp = op->node;
 	struct Scsi_Host *host;
 	struct qlogicpti *qpti;
+	static int nqptis;
 	const char *fcode;
 
 	/* Sometimes Antares cards come up not completely
 	 * setup, and we get a report of a zero IRQ.
 	 */
-	if (sdev->irqs[0] == 0)
+	if (op->irqs[0] == 0)
 		return -ENODEV;
 
 	host = scsi_host_alloc(tpnt, sizeof(struct qlogicpti));
 	if (!host)
 		return -ENOMEM;
 
-	qpti = (struct qlogicpti *) host->hostdata;
+	qpti = shost_priv(host);
 
 	host->max_id = MAX_TARGETS;
 	qpti->qhost = host;
-	qpti->sdev = sdev;
+	qpti->op = op;
 	qpti->qpti_id = nqptis;
-	qpti->prom_node = sdev->prom_node;
-	strcpy(qpti->prom_name, sdev->ofdev.node->name);
+	strcpy(qpti->prom_name, op->node->name);
 	qpti->is_pti = strcmp(qpti->prom_name, "QLGC,isp");
 
 	if (qpti_map_regs(qpti) < 0)
@@ -1341,12 +1343,12 @@ static int __devinit qpti_sbus_probe(struct of_device *dev, const struct of_devi
 		(qpti->ultra ? "Ultra" : "Fast"),
 		(qpti->differential ? "differential" : "single ended"));
 
-	if (scsi_add_host(host, &dev->dev)) {
+	if (scsi_add_host(host, &op->dev)) {
 		printk("qlogicpti%d: Failed scsi_add_host\n", qpti->qpti_id);
 		goto fail_unmap_queues;
 	}
 
-	dev_set_drvdata(&sdev->ofdev.dev, qpti);
+	dev_set_drvdata(&op->dev, qpti);
 
 	qpti_chain_add(qpti);
 
@@ -1357,19 +1359,20 @@ static int __devinit qpti_sbus_probe(struct of_device *dev, const struct of_devi
 
 fail_unmap_queues:
 #define QSIZE(entries)	(((entries) + 1) * QUEUE_ENTRY_LEN)
-	sbus_free_consistent(qpti->sdev,
-			     QSIZE(RES_QUEUE_LEN),
-			     qpti->res_cpu, qpti->res_dvma);
-	sbus_free_consistent(qpti->sdev,
-			     QSIZE(QLOGICPTI_REQ_QUEUE_LEN),
-			     qpti->req_cpu, qpti->req_dvma);
+	dma_free_coherent(&op->dev,
+			  QSIZE(RES_QUEUE_LEN),
+			  qpti->res_cpu, qpti->res_dvma);
+	dma_free_coherent(&op->dev,
+			  QSIZE(QLOGICPTI_REQ_QUEUE_LEN),
+			  qpti->req_cpu, qpti->req_dvma);
 #undef QSIZE
 
 fail_unmap_regs:
-	sbus_iounmap(qpti->qregs,
-		     qpti->sdev->reg_addrs[0].reg_size);
+	of_iounmap(&op->resource[0], qpti->qregs,
+		   resource_size(&op->resource[0]));
 	if (qpti->is_pti)
-		sbus_iounmap(qpti->sreg, sizeof(unsigned char));
+		of_iounmap(&op->resource[0], qpti->sreg,
+			   sizeof(unsigned char));
 
 fail_free_irq:
 	free_irq(qpti->irq, qpti);
@@ -1380,9 +1383,9 @@ fail_unlink:
 	return -ENODEV;
 }
 
-static int __devexit qpti_sbus_remove(struct of_device *dev)
+static int __devexit qpti_sbus_remove(struct of_device *op)
 {
-	struct qlogicpti *qpti = dev_get_drvdata(&dev->dev);
+	struct qlogicpti *qpti = dev_get_drvdata(&op->dev);
 
 	qpti_chain_del(qpti);
 
@@ -1395,24 +1398,25 @@ static int __devexit qpti_sbus_remove(struct of_device *dev)
 	free_irq(qpti->irq, qpti);
 
 #define QSIZE(entries)	(((entries) + 1) * QUEUE_ENTRY_LEN)
-	sbus_free_consistent(qpti->sdev,
-			     QSIZE(RES_QUEUE_LEN),
-			     qpti->res_cpu, qpti->res_dvma);
-	sbus_free_consistent(qpti->sdev,
-			     QSIZE(QLOGICPTI_REQ_QUEUE_LEN),
-			     qpti->req_cpu, qpti->req_dvma);
+	dma_free_coherent(&op->dev,
+			  QSIZE(RES_QUEUE_LEN),
+			  qpti->res_cpu, qpti->res_dvma);
+	dma_free_coherent(&op->dev,
+			  QSIZE(QLOGICPTI_REQ_QUEUE_LEN),
+			  qpti->req_cpu, qpti->req_dvma);
 #undef QSIZE
 
-	sbus_iounmap(qpti->qregs, qpti->sdev->reg_addrs[0].reg_size);
+	of_iounmap(&op->resource[0], qpti->qregs,
+		   resource_size(&op->resource[0]));
 	if (qpti->is_pti)
-		sbus_iounmap(qpti->sreg, sizeof(unsigned char));
+		of_iounmap(&op->resource[0], qpti->sreg, sizeof(unsigned char));
 
 	scsi_host_put(qpti->qhost);
 
 	return 0;
 }
 
-static struct of_device_id qpti_match[] = {
+static const struct of_device_id qpti_match[] = {
 	{
 		.name = "ptisp",
 		.data = &qpti_template,
@@ -1442,7 +1446,7 @@ static struct of_platform_driver qpti_sbus_driver = {
 
 static int __init qpti_init(void)
 {
-	return of_register_driver(&qpti_sbus_driver, &sbus_bus_type);
+	return of_register_driver(&qpti_sbus_driver, &of_bus_type);
 }
 
 static void __exit qpti_exit(void)
@@ -1453,7 +1457,7 @@ static void __exit qpti_exit(void)
 MODULE_DESCRIPTION("QlogicISP SBUS driver");
 MODULE_AUTHOR("David S. Miller (davem@davemloft.net)");
 MODULE_LICENSE("GPL");
-MODULE_VERSION("2.0");
+MODULE_VERSION("2.1");
 
 module_init(qpti_init);
 module_exit(qpti_exit);
diff --git a/drivers/scsi/qlogicpti.h b/drivers/scsi/qlogicpti.h
index ef6da2df584b..9c053bbaa877 100644
--- a/drivers/scsi/qlogicpti.h
+++ b/drivers/scsi/qlogicpti.h
@@ -342,7 +342,7 @@ struct qlogicpti {
 	u_int	                  req_in_ptr;		/* index of next request slot */
 	u_int	                  res_out_ptr;		/* index of next result slot  */
 	long	                  send_marker;		/* must we send a marker?     */
-	struct sbus_dev		 *sdev;
+	struct of_device	 *op;
 	unsigned long		  __pad;
 
 	int                       cmd_count[MAX_TARGETS];
diff --git a/drivers/scsi/sun_esp.c b/drivers/scsi/sun_esp.c
index f9cf70151366..3d73aad4bc82 100644
--- a/drivers/scsi/sun_esp.c
+++ b/drivers/scsi/sun_esp.c
@@ -1,6 +1,6 @@
 /* sun_esp.c: ESP front-end for Sparc SBUS systems.
  *
- * Copyright (C) 2007 David S. Miller (davem@davemloft.net)
+ * Copyright (C) 2007, 2008 David S. Miller (davem@davemloft.net)
  */
 
 #include <linux/kernel.h>
@@ -9,60 +9,70 @@
 #include <linux/module.h>
 #include <linux/mm.h>
 #include <linux/init.h>
+#include <linux/dma-mapping.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/irq.h>
 #include <asm/io.h>
 #include <asm/dma.h>
 
-#include <asm/sbus.h>
-
 #include <scsi/scsi_host.h>
 
 #include "esp_scsi.h"
 
 #define DRV_MODULE_NAME		"sun_esp"
 #define PFX DRV_MODULE_NAME	": "
-#define DRV_VERSION		"1.000"
-#define DRV_MODULE_RELDATE	"April 19, 2007"
+#define DRV_VERSION		"1.100"
+#define DRV_MODULE_RELDATE	"August 27, 2008"
 
 #define dma_read32(REG) \
 	sbus_readl(esp->dma_regs + (REG))
 #define dma_write32(VAL, REG) \
 	sbus_writel((VAL), esp->dma_regs + (REG))
 
-static int __devinit esp_sbus_find_dma(struct esp *esp, struct sbus_dev *dma_sdev)
-{
-	struct sbus_dev *sdev = esp->dev;
-	struct sbus_dma *dma;
+/* DVMA chip revisions */
+enum dvma_rev {
+	dvmarev0,
+	dvmaesc1,
+	dvmarev1,
+	dvmarev2,
+	dvmarev3,
+	dvmarevplus,
+	dvmahme
+};
 
-	if (dma_sdev != NULL) {
-		for_each_dvma(dma) {
-			if (dma->sdev == dma_sdev)
-				break;
-		}
-	} else {
-		for_each_dvma(dma) {
-			if (dma->sdev == NULL)
-				break;
+static int __devinit esp_sbus_setup_dma(struct esp *esp,
+					struct of_device *dma_of)
+{
+	esp->dma = dma_of;
 
-			/* If bus + slot are the same and it has the
-			 * correct OBP name, it's ours.
-			 */
-			if (sdev->bus == dma->sdev->bus &&
-			    sdev->slot == dma->sdev->slot &&
-			    (!strcmp(dma->sdev->prom_name, "dma") ||
-			     !strcmp(dma->sdev->prom_name, "espdma")))
-				break;
-		}
-	}
+	esp->dma_regs = of_ioremap(&dma_of->resource[0], 0,
+				   resource_size(&dma_of->resource[0]),
+				   "espdma");
+	if (!esp->dma_regs)
+		return -ENOMEM;
 
-	if (dma == NULL) {
-		printk(KERN_ERR PFX "[%s] Cannot find dma.\n",
-		       sdev->ofdev.node->full_name);
-		return -ENODEV;
+	switch (dma_read32(DMA_CSR) & DMA_DEVICE_ID) {
+	case DMA_VERS0:
+		esp->dmarev = dvmarev0;
+		break;
+	case DMA_ESCV1:
+		esp->dmarev = dvmaesc1;
+		break;
+	case DMA_VERS1:
+		esp->dmarev = dvmarev1;
+		break;
+	case DMA_VERS2:
+		esp->dmarev = dvmarev2;
+		break;
+	case DMA_VERHME:
+		esp->dmarev = dvmahme;
+		break;
+	case DMA_VERSPLUS:
+		esp->dmarev = dvmarevplus;
+		break;
 	}
-	esp->dma = dma;
-	esp->dma_regs = dma->regs;
 
 	return 0;
 
@@ -70,18 +80,18 @@ static int __devinit esp_sbus_find_dma(struct esp *esp, struct sbus_dev *dma_sde
 
 static int __devinit esp_sbus_map_regs(struct esp *esp, int hme)
 {
-	struct sbus_dev *sdev = esp->dev;
+	struct of_device *op = esp->dev;
 	struct resource *res;
 
 	/* On HME, two reg sets exist, first is DVMA,
 	 * second is ESP registers.
 	 */
 	if (hme)
-		res = &sdev->resource[1];
+		res = &op->resource[1];
 	else
-		res = &sdev->resource[0];
+		res = &op->resource[0];
 
-	esp->regs = sbus_ioremap(res, 0, SBUS_ESP_REG_SIZE, "ESP");
+	esp->regs = of_ioremap(res, 0, SBUS_ESP_REG_SIZE, "ESP");
 	if (!esp->regs)
 		return -ENOMEM;
 
@@ -90,10 +100,11 @@ static int __devinit esp_sbus_map_regs(struct esp *esp, int hme)
 
 static int __devinit esp_sbus_map_command_block(struct esp *esp)
 {
-	struct sbus_dev *sdev = esp->dev;
+	struct of_device *op = esp->dev;
 
-	esp->command_block = sbus_alloc_consistent(sdev, 16,
-						   &esp->command_block_dma);
+	esp->command_block = dma_alloc_coherent(&op->dev, 16,
+						&esp->command_block_dma,
+						GFP_ATOMIC);
 	if (!esp->command_block)
 		return -ENOMEM;
 	return 0;
@@ -102,17 +113,18 @@ static int __devinit esp_sbus_map_command_block(struct esp *esp)
 static int __devinit esp_sbus_register_irq(struct esp *esp)
 {
 	struct Scsi_Host *host = esp->host;
-	struct sbus_dev *sdev = esp->dev;
+	struct of_device *op = esp->dev;
 
-	host->irq = sdev->irqs[0];
+	host->irq = op->irqs[0];
 	return request_irq(host->irq, scsi_esp_intr, IRQF_SHARED, "ESP", esp);
 }
 
-static void __devinit esp_get_scsi_id(struct esp *esp)
+static void __devinit esp_get_scsi_id(struct esp *esp, struct of_device *espdma)
 {
-	struct sbus_dev *sdev = esp->dev;
-	struct device_node *dp = sdev->ofdev.node;
+	struct of_device *op = esp->dev;
+	struct device_node *dp;
 
+	dp = op->node;
 	esp->scsi_id = of_getintprop_default(dp, "initiator-id", 0xff);
 	if (esp->scsi_id != 0xff)
 		goto done;
@@ -121,13 +133,7 @@ static void __devinit esp_get_scsi_id(struct esp *esp)
 	if (esp->scsi_id != 0xff)
 		goto done;
 
-	if (!sdev->bus) {
-		/* SUN4 */
-		esp->scsi_id = 7;
-		goto done;
-	}
-
-	esp->scsi_id = of_getintprop_default(sdev->bus->ofdev.node,
+	esp->scsi_id = of_getintprop_default(espdma->node,
 					     "scsi-initiator-id", 7);
 
 done:
@@ -137,9 +143,10 @@ done:
 
 static void __devinit esp_get_differential(struct esp *esp)
 {
-	struct sbus_dev *sdev = esp->dev;
-	struct device_node *dp = sdev->ofdev.node;
+	struct of_device *op = esp->dev;
+	struct device_node *dp;
 
+	dp = op->node;
 	if (of_find_property(dp, "differential", NULL))
 		esp->flags |= ESP_FLAG_DIFFERENTIAL;
 	else
@@ -148,43 +155,36 @@ static void __devinit esp_get_differential(struct esp *esp)
 
 static void __devinit esp_get_clock_params(struct esp *esp)
 {
-	struct sbus_dev *sdev = esp->dev;
-	struct device_node *dp = sdev->ofdev.node;
-	struct device_node *bus_dp;
+	struct of_device *op = esp->dev;
+	struct device_node *bus_dp, *dp;
 	int fmhz;
 
-	bus_dp = NULL;
-	if (sdev != NULL && sdev->bus != NULL)
-		bus_dp = sdev->bus->ofdev.node;
+	dp = op->node;
+	bus_dp = dp->parent;
 
 	fmhz = of_getintprop_default(dp, "clock-frequency", 0);
 	if (fmhz == 0)
-		fmhz = (!bus_dp) ? 0 :
-			of_getintprop_default(bus_dp, "clock-frequency", 0);
+		fmhz = of_getintprop_default(bus_dp, "clock-frequency", 0);
 
 	esp->cfreq = fmhz;
 }
 
-static void __devinit esp_get_bursts(struct esp *esp, struct sbus_dev *dma)
+static void __devinit esp_get_bursts(struct esp *esp, struct of_device *dma_of)
 {
-	struct sbus_dev *sdev = esp->dev;
-	struct device_node *dp = sdev->ofdev.node;
-	u8 bursts;
+	struct device_node *dma_dp = dma_of->node;
+	struct of_device *op = esp->dev;
+	struct device_node *dp;
+	u8 bursts, val;
 
+	dp = op->node;
 	bursts = of_getintprop_default(dp, "burst-sizes", 0xff);
-	if (dma) {
-		struct device_node *dma_dp = dma->ofdev.node;
-		u8 val = of_getintprop_default(dma_dp, "burst-sizes", 0xff);
-		if (val != 0xff)
-			bursts &= val;
-	}
+	val = of_getintprop_default(dma_dp, "burst-sizes", 0xff);
+	if (val != 0xff)
+		bursts &= val;
 
-	if (sdev->bus) {
-		u8 val = of_getintprop_default(sdev->bus->ofdev.node,
-					       "burst-sizes", 0xff);
-		if (val != 0xff)
-			bursts &= val;
-	}
+	val = of_getintprop_default(dma_dp->parent, "burst-sizes", 0xff);
+	if (val != 0xff)
+		bursts &= val;
 
 	if (bursts == 0xff ||
 	    (bursts & DMA_BURST16) == 0 ||
@@ -194,9 +194,9 @@ static void __devinit esp_get_bursts(struct esp *esp, struct sbus_dev *dma)
 	esp->bursts = bursts;
 }
 
-static void __devinit esp_sbus_get_props(struct esp *esp, struct sbus_dev *espdma)
+static void __devinit esp_sbus_get_props(struct esp *esp, struct of_device *espdma)
 {
-	esp_get_scsi_id(esp);
+	esp_get_scsi_id(esp, espdma);
 	esp_get_differential(esp);
 	esp_get_clock_params(esp);
 	esp_get_bursts(esp, espdma);
@@ -215,25 +215,33 @@ static u8 sbus_esp_read8(struct esp *esp, unsigned long reg)
 static dma_addr_t sbus_esp_map_single(struct esp *esp, void *buf,
 				      size_t sz, int dir)
 {
-	return sbus_map_single(esp->dev, buf, sz, dir);
+	struct of_device *op = esp->dev;
+
+	return dma_map_single(&op->dev, buf, sz, dir);
 }
 
 static int sbus_esp_map_sg(struct esp *esp, struct scatterlist *sg,
 				  int num_sg, int dir)
 {
-	return sbus_map_sg(esp->dev, sg, num_sg, dir);
+	struct of_device *op = esp->dev;
+
+	return dma_map_sg(&op->dev, sg, num_sg, dir);
 }
 
 static void sbus_esp_unmap_single(struct esp *esp, dma_addr_t addr,
 				  size_t sz, int dir)
 {
-	sbus_unmap_single(esp->dev, addr, sz, dir);
+	struct of_device *op = esp->dev;
+
+	dma_unmap_single(&op->dev, addr, sz, dir);
 }
 
 static void sbus_esp_unmap_sg(struct esp *esp, struct scatterlist *sg,
 			      int num_sg, int dir)
 {
-	sbus_unmap_sg(esp->dev, sg, num_sg, dir);
+	struct of_device *op = esp->dev;
+
+	dma_unmap_sg(&op->dev, sg, num_sg, dir);
 }
 
 static int sbus_esp_irq_pending(struct esp *esp)
@@ -247,24 +255,26 @@ static void sbus_esp_reset_dma(struct esp *esp)
 {
 	int can_do_burst16, can_do_burst32, can_do_burst64;
 	int can_do_sbus64, lim;
+	struct of_device *op;
 	u32 val;
 
 	can_do_burst16 = (esp->bursts & DMA_BURST16) != 0;
 	can_do_burst32 = (esp->bursts & DMA_BURST32) != 0;
 	can_do_burst64 = 0;
 	can_do_sbus64 = 0;
-	if (sbus_can_dma_64bit(esp->dev))
+	op = esp->dev;
+	if (sbus_can_dma_64bit())
 		can_do_sbus64 = 1;
-	if (sbus_can_burst64(esp->sdev))
+	if (sbus_can_burst64())
 		can_do_burst64 = (esp->bursts & DMA_BURST64) != 0;
 
 	/* Put the DVMA into a known state. */
-	if (esp->dma->revision != dvmahme) {
+	if (esp->dmarev != dvmahme) {
 		val = dma_read32(DMA_CSR);
 		dma_write32(val | DMA_RST_SCSI, DMA_CSR);
 		dma_write32(val & ~DMA_RST_SCSI, DMA_CSR);
 	}
-	switch (esp->dma->revision) {
+	switch (esp->dmarev) {
 	case dvmahme:
 		dma_write32(DMA_RESET_FAS366, DMA_CSR);
 		dma_write32(DMA_RST_SCSI, DMA_CSR);
@@ -282,7 +292,7 @@ static void sbus_esp_reset_dma(struct esp *esp)
 
 		if (can_do_sbus64) {
 			esp->prev_hme_dmacsr |= DMA_SCSI_SBUS64;
-			sbus_set_sbus64(esp->dev, esp->bursts);
+			sbus_set_sbus64(&op->dev, esp->bursts);
 		}
 
 		lim = 1000;
@@ -346,14 +356,14 @@ static void sbus_esp_dma_drain(struct esp *esp)
 	u32 csr;
 	int lim;
 
-	if (esp->dma->revision == dvmahme)
+	if (esp->dmarev == dvmahme)
 		return;
 
 	csr = dma_read32(DMA_CSR);
 	if (!(csr & DMA_FIFO_ISDRAIN))
 		return;
 
-	if (esp->dma->revision != dvmarev3 && esp->dma->revision != dvmaesc1)
+	if (esp->dmarev != dvmarev3 && esp->dmarev != dvmaesc1)
 		dma_write32(csr | DMA_FIFO_STDRAIN, DMA_CSR);
 
 	lim = 1000;
@@ -369,7 +379,7 @@ static void sbus_esp_dma_drain(struct esp *esp)
 
 static void sbus_esp_dma_invalidate(struct esp *esp)
 {
-	if (esp->dma->revision == dvmahme) {
+	if (esp->dmarev == dvmahme) {
 		dma_write32(DMA_RST_SCSI, DMA_CSR);
 
 		esp->prev_hme_dmacsr = ((esp->prev_hme_dmacsr |
@@ -440,7 +450,7 @@ static void sbus_esp_send_dma_cmd(struct esp *esp, u32 addr, u32 esp_count,
 		else
 			csr &= ~DMA_ST_WRITE;
 		dma_write32(csr, DMA_CSR);
-		if (esp->dma->revision == dvmaesc1) {
+		if (esp->dmarev == dvmaesc1) {
 			u32 end = PAGE_ALIGN(addr + dma_count + 16U);
 			dma_write32(end - addr, DMA_COUNT);
 		}
@@ -476,10 +486,8 @@ static const struct esp_driver_ops sbus_esp_ops = {
 	.dma_error	=	sbus_esp_dma_error,
 };
 
-static int __devinit esp_sbus_probe_one(struct device *dev,
-					struct sbus_dev *esp_dev,
-					struct sbus_dev *espdma,
-					struct sbus_bus *sbus,
+static int __devinit esp_sbus_probe_one(struct of_device *op,
+					struct of_device *espdma,
 					int hme)
 {
 	struct scsi_host_template *tpnt = &scsi_esp_template;
@@ -497,13 +505,13 @@ static int __devinit esp_sbus_probe_one(struct device *dev,
 	esp = shost_priv(host);
 
 	esp->host = host;
-	esp->dev = esp_dev;
+	esp->dev = op;
 	esp->ops = &sbus_esp_ops;
 
 	if (hme)
 		esp->flags |= ESP_FLAG_WIDE_CAPABLE;
 
-	err = esp_sbus_find_dma(esp, espdma);
+	err = esp_sbus_setup_dma(esp, espdma);
 	if (err < 0)
 		goto fail_unlink;
 
@@ -525,15 +533,15 @@ static int __devinit esp_sbus_probe_one(struct device *dev,
 	 * come up with the reset bit set, so make sure that
 	 * is clear first.
 	 */
-	if (esp->dma->revision == dvmaesc1) {
+	if (esp->dmarev == dvmaesc1) {
 		u32 val = dma_read32(DMA_CSR);
 
 		dma_write32(val & ~DMA_RST_SCSI, DMA_CSR);
 	}
 
-	dev_set_drvdata(&esp_dev->ofdev.dev, esp);
+	dev_set_drvdata(&op->dev, esp);
 
-	err = scsi_esp_register(esp, dev);
+	err = scsi_esp_register(esp, &op->dev);
 	if (err)
 		goto fail_free_irq;
 
@@ -542,41 +550,46 @@ static int __devinit esp_sbus_probe_one(struct device *dev,
 fail_free_irq:
 	free_irq(host->irq, esp);
 fail_unmap_command_block:
-	sbus_free_consistent(esp->dev, 16,
-			     esp->command_block,
-			     esp->command_block_dma);
+	dma_free_coherent(&op->dev, 16,
+			  esp->command_block,
+			  esp->command_block_dma);
 fail_unmap_regs:
-	sbus_iounmap(esp->regs, SBUS_ESP_REG_SIZE);
+	of_iounmap(&op->resource[(hme ? 1 : 0)], esp->regs, SBUS_ESP_REG_SIZE);
 fail_unlink:
 	scsi_host_put(host);
 fail:
 	return err;
 }
 
-static int __devinit esp_sbus_probe(struct of_device *dev, const struct of_device_id *match)
+static int __devinit esp_sbus_probe(struct of_device *op, const struct of_device_id *match)
 {
-	struct sbus_dev *sdev = to_sbus_device(&dev->dev);
-	struct device_node *dp = dev->node;
-	struct sbus_dev *dma_sdev = NULL;
+	struct device_node *dma_node = NULL;
+	struct device_node *dp = op->node;
+	struct of_device *dma_of = NULL;
 	int hme = 0;
 
 	if (dp->parent &&
 	    (!strcmp(dp->parent->name, "espdma") ||
 	     !strcmp(dp->parent->name, "dma")))
-		dma_sdev = sdev->parent;
+		dma_node = dp->parent;
 	else if (!strcmp(dp->name, "SUNW,fas")) {
-		dma_sdev = sdev;
+		dma_node = op->node;
 		hme = 1;
 	}
+	if (dma_node)
+		dma_of = of_find_device_by_node(dma_node);
+	if (!dma_of)
+		return -ENODEV;
 
-	return esp_sbus_probe_one(&dev->dev, sdev, dma_sdev,
-				  sdev->bus, hme);
+	return esp_sbus_probe_one(op, dma_of, hme);
 }
 
-static int __devexit esp_sbus_remove(struct of_device *dev)
+static int __devexit esp_sbus_remove(struct of_device *op)
 {
-	struct esp *esp = dev_get_drvdata(&dev->dev);
+	struct esp *esp = dev_get_drvdata(&op->dev);
+	struct of_device *dma_of = esp->dma;
 	unsigned int irq = esp->host->irq;
+	bool is_hme;
 	u32 val;
 
 	scsi_esp_unregister(esp);
@@ -586,17 +599,25 @@ static int __devexit esp_sbus_remove(struct of_device *dev)
 	dma_write32(val & ~DMA_INT_ENAB, DMA_CSR);
 
 	free_irq(irq, esp);
-	sbus_free_consistent(esp->dev, 16,
-			     esp->command_block,
-			     esp->command_block_dma);
-	sbus_iounmap(esp->regs, SBUS_ESP_REG_SIZE);
+
+	is_hme = (esp->dmarev == dvmahme);
+
+	dma_free_coherent(&op->dev, 16,
+			  esp->command_block,
+			  esp->command_block_dma);
+	of_iounmap(&op->resource[(is_hme ? 1 : 0)], esp->regs,
+		   SBUS_ESP_REG_SIZE);
+	of_iounmap(&dma_of->resource[0], esp->dma_regs,
+		   resource_size(&dma_of->resource[0]));
 
 	scsi_host_put(esp->host);
 
+	dev_set_drvdata(&op->dev, NULL);
+
 	return 0;
 }
 
-static struct of_device_id esp_match[] = {
+static const struct of_device_id esp_match[] = {
 	{
 		.name = "SUNW,esp",
 	},
@@ -619,7 +640,7 @@ static struct of_platform_driver esp_sbus_driver = {
 
 static int __init sunesp_init(void)
 {
-	return of_register_driver(&esp_sbus_driver, &sbus_bus_type);
+	return of_register_driver(&esp_sbus_driver, &of_bus_type);
 }
 
 static void __exit sunesp_exit(void)
diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index 9ccc563d8730..d4104a3bbe87 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -44,6 +44,10 @@
 
 #include "8250.h"
 
+#ifdef CONFIG_SPARC
+#include "suncore.h"
+#endif
+
 /*
  * Configuration:
  *   share_irqs - whether we pass IRQF_SHARED to request_irq().  This option
@@ -53,6 +57,13 @@ static unsigned int share_irqs = SERIAL8250_SHARE_IRQS;
 
 static unsigned int nr_uarts = CONFIG_SERIAL_8250_RUNTIME_UARTS;
 
+static struct uart_driver serial8250_reg;
+
+static int serial_index(struct uart_port *port)
+{
+	return (serial8250_reg.minor - 64) + port->line;
+}
+
 /*
  * Debugging.
  */
@@ -536,7 +547,7 @@ static unsigned int serial_icr_read(struct uart_8250_port *up, int offset)
 /*
  * FIFO support.
  */
-static inline void serial8250_clear_fifos(struct uart_8250_port *p)
+static void serial8250_clear_fifos(struct uart_8250_port *p)
 {
 	if (p->capabilities & UART_CAP_FIFO) {
 		serial_outp(p, UART_FCR, UART_FCR_ENABLE_FIFO);
@@ -551,7 +562,7 @@ static inline void serial8250_clear_fifos(struct uart_8250_port *p)
  * capability" bit enabled.  Note that on XR16C850s, we need to
  * reset LCR to write to IER.
  */
-static inline void serial8250_set_sleep(struct uart_8250_port *p, int sleep)
+static void serial8250_set_sleep(struct uart_8250_port *p, int sleep)
 {
 	if (p->capabilities & UART_CAP_SLEEP) {
 		if (p->capabilities & UART_CAP_EFR) {
@@ -993,7 +1004,7 @@ static void autoconfig(struct uart_8250_port *up, unsigned int probeflags)
 		return;
 
 	DEBUG_AUTOCONF("ttyS%d: autoconf (0x%04x, 0x%p): ",
-			up->port.line, up->port.iobase, up->port.membase);
+		       serial_index(&up->port), up->port.iobase, up->port.membase);
 
 	/*
 	 * We really do need global IRQs disabled here - we're going to
@@ -1128,8 +1139,8 @@ static void autoconfig(struct uart_8250_port *up, unsigned int probeflags)
 	if (up->capabilities != uart_config[up->port.type].flags) {
 		printk(KERN_WARNING
 		       "ttyS%d: detected caps %08x should be %08x\n",
-			up->port.line, up->capabilities,
-			uart_config[up->port.type].flags);
+		       serial_index(&up->port), up->capabilities,
+		       uart_config[up->port.type].flags);
 	}
 
 	up->port.fifosize = uart_config[up->port.type].fifo_size;
@@ -1424,8 +1435,7 @@ static unsigned int check_modem_status(struct uart_8250_port *up)
 /*
  * This handles the interrupt from one port.
  */
-static inline void
-serial8250_handle_port(struct uart_8250_port *up)
+static void serial8250_handle_port(struct uart_8250_port *up)
 {
 	unsigned int status;
 	unsigned long flags;
@@ -1719,7 +1729,7 @@ static void serial8250_break_ctl(struct uart_port *port, int break_state)
 /*
  *	Wait for transmitter & holding register to empty
  */
-static inline void wait_for_xmitr(struct uart_8250_port *up, int bits)
+static void wait_for_xmitr(struct uart_8250_port *up, int bits)
 {
 	unsigned int status, tmout = 10000;
 
@@ -1854,7 +1864,8 @@ static int serial8250_startup(struct uart_port *port)
 	 */
 	if (!(up->port.flags & UPF_BUGGY_UART) &&
 	    (serial_inp(up, UART_LSR) == 0xff)) {
-		printk("ttyS%d: LSR safety check engaged!\n", up->port.line);
+		printk(KERN_INFO "ttyS%d: LSR safety check engaged!\n",
+		       serial_index(&up->port));
 		return -ENODEV;
 	}
 
@@ -1909,7 +1920,8 @@ static int serial8250_startup(struct uart_port *port)
 		 */
 		if (!(iir1 & UART_IIR_NO_INT) && (iir & UART_IIR_NO_INT)) {
 			up->bugs |= UART_BUG_THRE;
-			pr_debug("ttyS%d - using backup timer\n", port->line);
+			pr_debug("ttyS%d - using backup timer\n",
+				 serial_index(port));
 		}
 	}
 
@@ -1969,7 +1981,7 @@ static int serial8250_startup(struct uart_port *port)
 		if (!(up->bugs & UART_BUG_TXEN)) {
 			up->bugs |= UART_BUG_TXEN;
 			pr_debug("ttyS%d - enabling bad tx status workarounds\n",
-				 port->line);
+				 serial_index(port));
 		}
 	} else {
 		up->bugs &= ~UART_BUG_TXEN;
@@ -2630,7 +2642,6 @@ static int serial8250_console_early_setup(void)
 	return serial8250_find_port_for_earlycon();
 }
 
-static struct uart_driver serial8250_reg;
 static struct console serial8250_console = {
 	.name		= "ttyS",
 	.write		= serial8250_console_write,
@@ -2677,7 +2688,6 @@ static struct uart_driver serial8250_reg = {
 	.dev_name		= "ttyS",
 	.major			= TTY_MAJOR,
 	.minor			= 64,
-	.nr			= UART_NR,
 	.cons			= SERIAL8250_CONSOLE,
 };
 
@@ -2959,10 +2969,12 @@ static int __init serial8250_init(void)
 		"%d ports, IRQ sharing %sabled\n", nr_uarts,
 		share_irqs ? "en" : "dis");
 
-	for (i = 0; i < NR_IRQS; i++)
-		spin_lock_init(&irq_lists[i].lock);
-
+#ifdef CONFIG_SPARC
+	ret = sunserial_register_minors(&serial8250_reg, UART_NR);
+#else
+	serial8250_reg.nr = UART_NR;
 	ret = uart_register_driver(&serial8250_reg);
+#endif
 	if (ret)
 		goto out;
 
@@ -2987,7 +2999,11 @@ static int __init serial8250_init(void)
  put_dev:
 	platform_device_put(serial8250_isa_devs);
  unreg_uart_drv:
+#ifdef CONFIG_SPARC
+	sunserial_unregister_minors(&serial8250_reg, UART_NR);
+#else
 	uart_unregister_driver(&serial8250_reg);
+#endif
  out:
 	return ret;
 }
@@ -3006,7 +3022,11 @@ static void __exit serial8250_exit(void)
 	platform_driver_unregister(&serial8250_isa_driver);
 	platform_device_unregister(isa_dev);
 
+#ifdef CONFIG_SPARC
+	sunserial_unregister_minors(&serial8250_reg, UART_NR);
+#else
 	uart_unregister_driver(&serial8250_reg);
+#endif
 }
 
 module_init(serial8250_init);
diff --git a/drivers/serial/8250_pci.c b/drivers/serial/8250_pci.c
index c2f23933155b..c014ffb110e9 100644
--- a/drivers/serial/8250_pci.c
+++ b/drivers/serial/8250_pci.c
@@ -2041,9 +2041,9 @@ static int pciserial_resume_one(struct pci_dev *dev)
 		 * The device may have been disabled.  Re-enable it.
 		 */
 		err = pci_enable_device(dev);
+		/* FIXME: We cannot simply error out here */
 		if (err)
-			return err;
-
+			printk(KERN_ERR "pciserial: Unable to re-enable ports, trying to continue.\n");
 		pciserial_resume_ports(priv);
 	}
 	return 0;
diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig
index 77cb34270fc1..31786b3b0a68 100644
--- a/drivers/serial/Kconfig
+++ b/drivers/serial/Kconfig
@@ -9,7 +9,6 @@ menu "Serial drivers"
 # The new 8250/16550 serial drivers
 config SERIAL_8250
 	tristate "8250/16550 and compatible serial support"
-	depends on (BROKEN || !SPARC)
 	select SERIAL_CORE
 	---help---
 	  This selects whether you want to include the driver for the standard
@@ -994,24 +993,12 @@ config SERIAL_68328_RTS_CTS
 	bool "Support RTS/CTS on 68328 serial port"
 	depends on SERIAL_68328
 
-config SERIAL_COLDFIRE
-	bool "ColdFire serial support (DEPRECATED)"
-	depends on COLDFIRE
-	help
-	  This driver supports the built-in serial ports of the Motorola ColdFire
-	  family of CPUs.
-	  This driver is deprecated because it supports only the old interface
-	  for serial drivers and features like magic keys are not working.
-	  Please switch to the new style driver because this driver will be
-	  removed soon.
-
 config SERIAL_MCF
-	bool "Coldfire serial support (new style driver)"
+	bool "Coldfire serial support"
 	depends on COLDFIRE
 	select SERIAL_CORE
 	help
-	  This new serial driver supports the Freescale Coldfire serial ports
-	  using the new serial driver subsystem.
+	  This serial driver supports the Freescale Coldfire serial ports.
 
 config SERIAL_MCF_BAUDRATE
 	int "Default baudrate for Coldfire serial ports"
diff --git a/drivers/serial/Makefile b/drivers/serial/Makefile
index 7e7383e890d8..0c17c8ddb19d 100644
--- a/drivers/serial/Makefile
+++ b/drivers/serial/Makefile
@@ -4,6 +4,16 @@
 
 obj-$(CONFIG_SERIAL_CORE) += serial_core.o
 obj-$(CONFIG_SERIAL_21285) += 21285.o
+
+# These Sparc drivers have to appear before others such as 8250
+# which share ttySx minor node space.  Otherwise console device
+# names change and other unplesantries.
+obj-$(CONFIG_SERIAL_SUNCORE) += suncore.o
+obj-$(CONFIG_SERIAL_SUNHV) += sunhv.o
+obj-$(CONFIG_SERIAL_SUNZILOG) += sunzilog.o
+obj-$(CONFIG_SERIAL_SUNSU) += sunsu.o
+obj-$(CONFIG_SERIAL_SUNSAB) += sunsab.o
+
 obj-$(CONFIG_SERIAL_8250) += 8250.o
 obj-$(CONFIG_SERIAL_8250_PNP) += 8250_pnp.o
 obj-$(CONFIG_SERIAL_8250_GSC) += 8250_gsc.o
@@ -31,16 +41,10 @@ obj-$(CONFIG_SERIAL_S3C2400) += s3c2400.o
 obj-$(CONFIG_SERIAL_S3C2410) += s3c2410.o
 obj-$(CONFIG_SERIAL_S3C2412) += s3c2412.o
 obj-$(CONFIG_SERIAL_S3C2440) += s3c2440.o
-obj-$(CONFIG_SERIAL_SUNCORE) += suncore.o
-obj-$(CONFIG_SERIAL_SUNHV) += sunhv.o
-obj-$(CONFIG_SERIAL_SUNZILOG) += sunzilog.o
 obj-$(CONFIG_SERIAL_IP22_ZILOG) += ip22zilog.o
-obj-$(CONFIG_SERIAL_SUNSU) += sunsu.o
-obj-$(CONFIG_SERIAL_SUNSAB) += sunsab.o
 obj-$(CONFIG_SERIAL_MUX) += mux.o
 obj-$(CONFIG_SERIAL_68328) += 68328serial.o
 obj-$(CONFIG_SERIAL_68360) += 68360serial.o
-obj-$(CONFIG_SERIAL_COLDFIRE) += mcfserial.o
 obj-$(CONFIG_SERIAL_MCF) += mcf.o
 obj-$(CONFIG_SERIAL_PMACZILOG) += pmac_zilog.o
 obj-$(CONFIG_SERIAL_LH7A40X) += serial_lh7a40x.o
diff --git a/drivers/serial/bfin_5xx.c b/drivers/serial/bfin_5xx.c
index 4a0d30bed9f1..569f0e2476c6 100644
--- a/drivers/serial/bfin_5xx.c
+++ b/drivers/serial/bfin_5xx.c
@@ -1,7 +1,7 @@
 /*
  * Blackfin On-Chip Serial Driver
  *
- * Copyright 2006-2007 Analog Devices Inc.
+ * Copyright 2006-2008 Analog Devices Inc.
  *
  * Enter bugs at http://blackfin.uclinux.org/
  *
@@ -42,6 +42,9 @@
 #define BFIN_SERIAL_MAJOR	204
 #define BFIN_SERIAL_MINOR	64
 
+static struct bfin_serial_port bfin_serial_ports[BFIN_UART_NR_PORTS];
+static int nr_active_ports = ARRAY_SIZE(bfin_serial_resource);
+
 /*
  * Setup for console. Argument comes from the menuconfig
  */
@@ -126,13 +129,13 @@ static int kgdb_entry_state;
 void kgdb_put_debug_char(int chr)
 {
 	struct bfin_serial_port *uart;
-	
+
 	if (CONFIG_KGDB_UART_PORT < 0
 		|| CONFIG_KGDB_UART_PORT >= BFIN_UART_NR_PORTS)
 		uart = &bfin_serial_ports[0];
 	else
 		uart = &bfin_serial_ports[CONFIG_KGDB_UART_PORT];
-	
+
 	while (!(UART_GET_LSR(uart) & THRE)) {
 		SSYNC();
 	}
@@ -152,7 +155,7 @@ int kgdb_get_debug_char(void)
 		uart = &bfin_serial_ports[0];
 	else
 		uart = &bfin_serial_ports[CONFIG_KGDB_UART_PORT];
-	
+
 	while(!(UART_GET_LSR(uart) & DR)) {
 		SSYNC();
 	}
@@ -298,7 +301,11 @@ static void bfin_serial_tx_chars(struct bfin_serial_port *uart)
 	bfin_serial_mctrl_check(uart);
 
 	if (uart_circ_empty(xmit) || uart_tx_stopped(&uart->port)) {
-		bfin_serial_stop_tx(&uart->port);
+#ifdef CONFIG_BF54x
+		/* Clear TFI bit */
+		UART_PUT_LSR(uart, TFI);
+#endif
+		UART_CLEAR_IER(uart, ETBEI);
 		return;
 	}
 
@@ -317,9 +324,6 @@ static void bfin_serial_tx_chars(struct bfin_serial_port *uart)
 
 	if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
 		uart_write_wakeup(&uart->port);
-
-	if (uart_circ_empty(xmit))
-		bfin_serial_stop_tx(&uart->port);
 }
 
 static irqreturn_t bfin_serial_rx_int(int irq, void *dev_id)
@@ -645,6 +649,42 @@ static int bfin_serial_startup(struct uart_port *port)
 		free_irq(uart->port.irq, uart);
 		return -EBUSY;
 	}
+
+# ifdef CONFIG_BF54x
+	{
+		unsigned uart_dma_ch_rx, uart_dma_ch_tx;
+
+		switch (uart->port.irq) {
+		case IRQ_UART3_RX:
+			uart_dma_ch_rx = CH_UART3_RX;
+			uart_dma_ch_tx = CH_UART3_TX;
+			break;
+		case IRQ_UART2_RX:
+			uart_dma_ch_rx = CH_UART2_RX;
+			uart_dma_ch_tx = CH_UART2_TX;
+			break;
+		default:
+			uart_dma_ch_rx = uart_dma_ch_tx = 0;
+			break;
+		};
+
+		if (uart_dma_ch_rx &&
+			request_dma(uart_dma_ch_rx, "BFIN_UART_RX") < 0) {
+			printk(KERN_NOTICE"Fail to attach UART interrupt\n");
+			free_irq(uart->port.irq, uart);
+			free_irq(uart->port.irq + 1, uart);
+			return -EBUSY;
+		}
+		if (uart_dma_ch_tx &&
+			request_dma(uart_dma_ch_tx, "BFIN_UART_TX") < 0) {
+			printk(KERN_NOTICE "Fail to attach UART interrupt\n");
+			free_dma(uart_dma_ch_rx);
+			free_irq(uart->port.irq, uart);
+			free_irq(uart->port.irq + 1, uart);
+			return -EBUSY;
+		}
+	}
+# endif
 #endif
 	UART_SET_IER(uart, ERBFI);
 	return 0;
@@ -662,6 +702,20 @@ static void bfin_serial_shutdown(struct uart_port *port)
 	del_timer(&(uart->rx_dma_timer));
 	dma_free_coherent(NULL, PAGE_SIZE, uart->rx_dma_buf.buf, 0);
 #else
+#ifdef CONFIG_BF54x
+	switch (uart->port.irq) {
+	case IRQ_UART3_RX:
+		free_dma(CH_UART3_RX);
+		free_dma(CH_UART3_TX);
+		break;
+	case IRQ_UART2_RX:
+		free_dma(CH_UART2_RX);
+		free_dma(CH_UART2_TX);
+		break;
+	default:
+		break;
+	};
+#endif
 #ifdef	CONFIG_KGDB_UART
 	if (uart->port.line != CONFIG_KGDB_UART_PORT)
 #endif
@@ -757,6 +811,9 @@ bfin_serial_set_termios(struct uart_port *port, struct ktermios *termios,
 	val |= UCEN;
 	UART_PUT_GCTL(uart, val);
 
+	/* Port speed changed, update the per-port timeout. */
+	uart_update_timeout(port, termios->c_cflag, baud);
+
 	spin_unlock_irqrestore(&uart->port.lock, flags);
 }
 
@@ -859,8 +916,9 @@ static void __init bfin_serial_init_ports(void)
 		return;
 	first = 0;
 
-	for (i = 0; i < nr_ports; i++) {
+	for (i = 0; i < nr_active_ports; i++) {
 		bfin_serial_ports[i].port.uartclk   = get_sclk();
+		bfin_serial_ports[i].port.fifosize  = BFIN_UART_TX_FIFO_SIZE;
 		bfin_serial_ports[i].port.ops       = &bfin_serial_pops;
 		bfin_serial_ports[i].port.line      = i;
 		bfin_serial_ports[i].port.iotype    = UPIO_MEM;
@@ -961,7 +1019,7 @@ bfin_serial_console_setup(struct console *co, char *options)
 	 * if so, search for the first available port that does have
 	 * console support.
 	 */
-	if (co->index == -1 || co->index >= nr_ports)
+	if (co->index == -1 || co->index >= nr_active_ports)
 		co->index = 0;
 	uart = &bfin_serial_ports[co->index];
 
@@ -1056,7 +1114,7 @@ static __init void early_serial_write(struct console *con, const char *s,
 	}
 }
 
-static struct __init console bfin_early_serial_console = {
+static struct __initdata console bfin_early_serial_console = {
 	.name = "early_BFuart",
 	.write = early_serial_write,
 	.device = uart_console_device,
@@ -1072,7 +1130,7 @@ struct console __init *bfin_earlyserial_init(unsigned int port,
 	struct bfin_serial_port *uart;
 	struct ktermios t;
 
-	if (port == -1 || port >= nr_ports)
+	if (port == -1 || port >= nr_active_ports)
 		port = 0;
 	bfin_serial_init_ports();
 	bfin_early_serial_console.index = port;
@@ -1100,20 +1158,26 @@ static struct uart_driver bfin_serial_reg = {
 
 static int bfin_serial_suspend(struct platform_device *dev, pm_message_t state)
 {
-	struct bfin_serial_port *uart = platform_get_drvdata(dev);
+	int i;
 
-	if (uart)
-		uart_suspend_port(&bfin_serial_reg, &uart->port);
+	for (i = 0; i < nr_active_ports; i++) {
+		if (bfin_serial_ports[i].port.dev != &dev->dev)
+			continue;
+		uart_suspend_port(&bfin_serial_reg, &bfin_serial_ports[i].port);
+	}
 
 	return 0;
 }
 
 static int bfin_serial_resume(struct platform_device *dev)
 {
-	struct bfin_serial_port *uart = platform_get_drvdata(dev);
+	int i;
 
-	if (uart)
-		uart_resume_port(&bfin_serial_reg, &uart->port);
+	for (i = 0; i < nr_active_ports; i++) {
+		if (bfin_serial_ports[i].port.dev != &dev->dev)
+			continue;
+		uart_resume_port(&bfin_serial_reg, &bfin_serial_ports[i].port);
+	}
 
 	return 0;
 }
@@ -1128,32 +1192,31 @@ static int bfin_serial_probe(struct platform_device *dev)
 			break;
 
 	if (i < dev->num_resources) {
-		for (i = 0; i < nr_ports; i++, res++) {
+		for (i = 0; i < nr_active_ports; i++, res++) {
 			if (bfin_serial_ports[i].port.mapbase != res->start)
 				continue;
 			bfin_serial_ports[i].port.dev = &dev->dev;
 			uart_add_one_port(&bfin_serial_reg, &bfin_serial_ports[i].port);
-			platform_set_drvdata(dev, &bfin_serial_ports[i]);
 		}
 	}
 
 	return 0;
 }
 
-static int bfin_serial_remove(struct platform_device *pdev)
+static int bfin_serial_remove(struct platform_device *dev)
 {
-	struct bfin_serial_port *uart = platform_get_drvdata(pdev);
-
+	int i;
 
+	for (i = 0; i < nr_active_ports; i++) {
+		if (bfin_serial_ports[i].port.dev != &dev->dev)
+			continue;
+		uart_remove_one_port(&bfin_serial_reg, &bfin_serial_ports[i].port);
+		bfin_serial_ports[i].port.dev = NULL;
 #ifdef CONFIG_SERIAL_BFIN_CTSRTS
-	gpio_free(uart->cts_pin);
-	gpio_free(uart->rts_pin);
+		gpio_free(bfin_serial_ports[i].cts_pin);
+		gpio_free(bfin_serial_ports[i].rts_pin);
 #endif
-
-	platform_set_drvdata(pdev, NULL);
-
-	if (uart)
-		uart_remove_one_port(&bfin_serial_reg, &uart->port);
+	}
 
 	return 0;
 }
diff --git a/drivers/serial/crisv10.c b/drivers/serial/crisv10.c
index bf94a770bb44..211c21797ce0 100644
--- a/drivers/serial/crisv10.c
+++ b/drivers/serial/crisv10.c
@@ -457,7 +457,6 @@ static struct e100_serial rs_table[] = {
 #define NR_PORTS (sizeof(rs_table)/sizeof(struct e100_serial))
 
 static struct ktermios *serial_termios[NR_PORTS];
-static struct ktermios *serial_termios_locked[NR_PORTS];
 #ifdef CONFIG_ETRAX_SERIAL_FAST_TIMER
 static struct fast_timer fast_timers[NR_PORTS];
 #endif
@@ -4419,6 +4418,7 @@ rs_init(void)
 			rs485_pa_bit)) {
 		printk(KERN_CRIT "ETRAX100LX serial: Could not allocate "
 			"RS485 pin\n");
+		put_tty_driver(driver);
 		return -EBUSY;
 	}
 #endif
@@ -4427,6 +4427,7 @@ rs_init(void)
 			rs485_port_g_bit)) {
 		printk(KERN_CRIT "ETRAX100LX serial: Could not allocate "
 			"RS485 pin\n");
+		put_tty_driver(driver);
 		return -EBUSY;
 	}
 #endif
@@ -4446,8 +4447,6 @@ rs_init(void)
 	driver->init_termios.c_ispeed = 115200;
 	driver->init_termios.c_ospeed = 115200;
 	driver->flags = TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV;
-	driver->termios = serial_termios;
-	driver->termios_locked = serial_termios_locked;
 
 	tty_set_operations(driver, &rs_ops);
         serial_driver = driver;
diff --git a/drivers/serial/mcfserial.c b/drivers/serial/mcfserial.c
deleted file mode 100644
index fbe3835f6b77..000000000000
--- a/drivers/serial/mcfserial.c
+++ /dev/null
@@ -1,1965 +0,0 @@
-#warning This driver is deprecated. Check Kconfig for details.
-/*
- * mcfserial.c -- serial driver for ColdFire internal UARTS.
- *
- * Copyright (C) 1999-2003 Greg Ungerer <gerg@snapgear.com>
- * Copyright (c) 2000-2001 Lineo, Inc. <www.lineo.com> 
- * Copyright (C) 2001-2002 SnapGear Inc. <www.snapgear.com> 
- *
- * Based on code from 68332serial.c which was:
- *
- * Copyright (C) 1995 David S. Miller (davem@caip.rutgers.edu)
- * Copyright (C) 1998 TSHG
- * Copyright (c) 1999 Rt-Control Inc. <jeff@uclinux.org>
- *
- * Changes:
- * 08/07/2003    Daniele Bellucci <bellucda@tiscali.it>
- *               some cleanups in mcfrs_write.
- *
- */
- 
-#include <linux/module.h>
-#include <linux/errno.h>
-#include <linux/signal.h>
-#include <linux/sched.h>
-#include <linux/timer.h>
-#include <linux/wait.h>
-#include <linux/interrupt.h>
-#include <linux/tty.h>
-#include <linux/tty_flip.h>
-#include <linux/string.h>
-#include <linux/fcntl.h>
-#include <linux/mm.h>
-#include <linux/kernel.h>
-#include <linux/serial.h>
-#include <linux/serialP.h>
-#include <linux/console.h>
-#include <linux/init.h>
-#include <linux/bitops.h>
-#include <linux/delay.h>
-
-#include <asm/io.h>
-#include <asm/irq.h>
-#include <asm/system.h>
-#include <asm/delay.h>
-#include <asm/coldfire.h>
-#include <asm/mcfsim.h>
-#include <asm/mcfuart.h>
-#include <asm/nettel.h>
-#include <asm/uaccess.h>
-#include "mcfserial.h"
-
-struct timer_list mcfrs_timer_struct;
-
-/*
- *	Default console baud rate,  we use this as the default
- *	for all ports so init can just open /dev/console and
- *	keep going.  Perhaps one day the cflag settings for the
- *	console can be used instead.
- */
-#if defined(CONFIG_HW_FEITH)
-#define	CONSOLE_BAUD_RATE	38400
-#define	DEFAULT_CBAUD		B38400
-#elif defined(CONFIG_MOD5272) || defined(CONFIG_M5208EVB) || \
-      defined(CONFIG_M5329EVB) || defined(CONFIG_GILBARCO)
-#define CONSOLE_BAUD_RATE 	115200
-#define DEFAULT_CBAUD		B115200
-#elif defined(CONFIG_ARNEWSH) || defined(CONFIG_FREESCALE) || \
-      defined(CONFIG_senTec) || defined(CONFIG_SNEHA) || defined(CONFIG_AVNET)
-#define	CONSOLE_BAUD_RATE	19200
-#define	DEFAULT_CBAUD		B19200
-#endif
-
-#ifndef CONSOLE_BAUD_RATE
-#define	CONSOLE_BAUD_RATE	9600
-#define	DEFAULT_CBAUD		B9600
-#endif
-
-int mcfrs_console_inited = 0;
-int mcfrs_console_port = -1;
-int mcfrs_console_baud = CONSOLE_BAUD_RATE;
-int mcfrs_console_cbaud = DEFAULT_CBAUD;
-
-/*
- *	Driver data structures.
- */
-static struct tty_driver *mcfrs_serial_driver;
-
-/* number of characters left in xmit buffer before we ask for more */
-#define WAKEUP_CHARS 256
-
-/* Debugging...
- */
-#undef SERIAL_DEBUG_OPEN
-#undef SERIAL_DEBUG_FLOW
-
-#if defined(CONFIG_M523x) || defined(CONFIG_M527x) || defined(CONFIG_M528x) || \
-    defined(CONFIG_M520x) || defined(CONFIG_M532x)
-#define	IRQBASE	(MCFINT_VECBASE+MCFINT_UART0)
-#else
-#define	IRQBASE	73
-#endif
-
-/*
- *	Configuration table, UARTs to look for at startup.
- */
-static struct mcf_serial mcfrs_table[] = {
-	{  /* ttyS0 */
-		.magic = 0,
-		.addr = (volatile unsigned char *) (MCF_MBAR+MCFUART_BASE1),
-		.irq = IRQBASE,
-		.flags = ASYNC_BOOT_AUTOCONF,
-	},
-#ifdef MCFUART_BASE2
-	{  /* ttyS1 */
-		.magic = 0,
-		.addr = (volatile unsigned char *) (MCF_MBAR+MCFUART_BASE2),
-		.irq = IRQBASE+1,
-		.flags = ASYNC_BOOT_AUTOCONF,
-	},
-#endif
-#ifdef MCFUART_BASE3
-	{  /* ttyS2 */
-		.magic = 0,
-		.addr = (volatile unsigned char *) (MCF_MBAR+MCFUART_BASE3),
-		.irq = IRQBASE+2,
-		.flags = ASYNC_BOOT_AUTOCONF,
-	},
-#endif
-#ifdef MCFUART_BASE4
-	{  /* ttyS3 */
-		.magic = 0,
-		.addr = (volatile unsigned char *) (MCF_MBAR+MCFUART_BASE4),
-		.irq = IRQBASE+3,
-		.flags = ASYNC_BOOT_AUTOCONF,
-	},
-#endif
-};
-
-
-#define	NR_PORTS	(sizeof(mcfrs_table) / sizeof(struct mcf_serial))
-
-/*
- * This is used to figure out the divisor speeds and the timeouts.
- */
-static int mcfrs_baud_table[] = {
-	0, 50, 75, 110, 134, 150, 200, 300, 600, 1200, 1800, 2400, 4800,
-	9600, 19200, 38400, 57600, 115200, 230400, 460800, 0
-};
-#define MCFRS_BAUD_TABLE_SIZE \
-			(sizeof(mcfrs_baud_table)/sizeof(mcfrs_baud_table[0]))
-
-
-#ifdef CONFIG_MAGIC_SYSRQ
-/*
- *	Magic system request keys. Used for debugging...
- */
-extern int	magic_sysrq_key(int ch);
-#endif
-
-
-/*
- *	Forware declarations...
- */
-static void	mcfrs_change_speed(struct mcf_serial *info);
-static void	mcfrs_wait_until_sent(struct tty_struct *tty, int timeout);
-
-
-static inline int serial_paranoia_check(struct mcf_serial *info,
-					char *name, const char *routine)
-{
-#ifdef SERIAL_PARANOIA_CHECK
-	static const char badmagic[] =
-		"MCFRS(warning): bad magic number for serial struct %s in %s\n";
-	static const char badinfo[] =
-		"MCFRS(warning): null mcf_serial for %s in %s\n";
-
-	if (!info) {
-		printk(badinfo, name, routine);
-		return 1;
-	}
-	if (info->magic != SERIAL_MAGIC) {
-		printk(badmagic, name, routine);
-		return 1;
-	}
-#endif
-	return 0;
-}
-
-/*
- *	Sets or clears DTR and RTS on the requested line.
- */
-static void mcfrs_setsignals(struct mcf_serial *info, int dtr, int rts)
-{
-	volatile unsigned char	*uartp;
-	unsigned long		flags;
-	
-#if 0
-	printk("%s(%d): mcfrs_setsignals(info=%x,dtr=%d,rts=%d)\n",
-		__FILE__, __LINE__, info, dtr, rts);
-#endif
-
-	local_irq_save(flags);
-	if (dtr >= 0) {
-#ifdef MCFPP_DTR0
-		if (info->line)
-			mcf_setppdata(MCFPP_DTR1, (dtr ? 0 : MCFPP_DTR1));
-		else
-			mcf_setppdata(MCFPP_DTR0, (dtr ? 0 : MCFPP_DTR0));
-#endif
-	}
-	if (rts >= 0) {
-		uartp = info->addr;
-		if (rts) {
-			info->sigs |= TIOCM_RTS;
-			uartp[MCFUART_UOP1] = MCFUART_UOP_RTS;
-		} else {
-			info->sigs &= ~TIOCM_RTS;
-			uartp[MCFUART_UOP0] = MCFUART_UOP_RTS;
-		}
-	}
-	local_irq_restore(flags);
-	return;
-}
-
-/*
- *	Gets values of serial signals.
- */
-static int mcfrs_getsignals(struct mcf_serial *info)
-{
-	volatile unsigned char	*uartp;
-	unsigned long		flags;
-	int			sigs;
-#if defined(CONFIG_NETtel) && defined(CONFIG_M5307)
-	unsigned short		ppdata;
-#endif
-
-#if 0
-	printk("%s(%d): mcfrs_getsignals(info=%x)\n", __FILE__, __LINE__);
-#endif
-
-	local_irq_save(flags);
-	uartp = info->addr;
-	sigs = (uartp[MCFUART_UIPR] & MCFUART_UIPR_CTS) ? 0 : TIOCM_CTS;
-	sigs |= (info->sigs & TIOCM_RTS);
-
-#ifdef MCFPP_DCD0
-{
-	unsigned int ppdata;
-	ppdata = mcf_getppdata();
-	if (info->line == 0) {
-		sigs |= (ppdata & MCFPP_DCD0) ? 0 : TIOCM_CD;
-		sigs |= (ppdata & MCFPP_DTR0) ? 0 : TIOCM_DTR;
-	} else if (info->line == 1) {
-		sigs |= (ppdata & MCFPP_DCD1) ? 0 : TIOCM_CD;
-		sigs |= (ppdata & MCFPP_DTR1) ? 0 : TIOCM_DTR;
-	}
-}
-#endif
-
-	local_irq_restore(flags);
-	return(sigs);
-}
-
-/*
- * ------------------------------------------------------------
- * mcfrs_stop() and mcfrs_start()
- *
- * This routines are called before setting or resetting tty->stopped.
- * They enable or disable transmitter interrupts, as necessary.
- * ------------------------------------------------------------
- */
-static void mcfrs_stop(struct tty_struct *tty)
-{
-	volatile unsigned char	*uartp;
-	struct mcf_serial	*info = (struct mcf_serial *)tty->driver_data;
-	unsigned long		flags;
-
-	if (serial_paranoia_check(info, tty->name, "mcfrs_stop"))
-		return;
-	
-	local_irq_save(flags);
-	uartp = info->addr;
-	info->imr &= ~MCFUART_UIR_TXREADY;
-	uartp[MCFUART_UIMR] = info->imr;
-	local_irq_restore(flags);
-}
-
-static void mcfrs_start(struct tty_struct *tty)
-{
-	volatile unsigned char	*uartp;
-	struct mcf_serial	*info = (struct mcf_serial *)tty->driver_data;
-	unsigned long		flags;
-	
-	if (serial_paranoia_check(info, tty->name, "mcfrs_start"))
-		return;
-
-	local_irq_save(flags);
-	if (info->xmit_cnt && info->xmit_buf) {
-		uartp = info->addr;
-		info->imr |= MCFUART_UIR_TXREADY;
-		uartp[MCFUART_UIMR] = info->imr;
-	}
-	local_irq_restore(flags);
-}
-
-/*
- * ----------------------------------------------------------------------
- *
- * Here starts the interrupt handling routines.  All of the following
- * subroutines are declared as inline and are folded into
- * mcfrs_interrupt().  They were separated out for readability's sake.
- *
- * Note: mcfrs_interrupt() is a "fast" interrupt, which means that it
- * runs with interrupts turned off.  People who may want to modify
- * mcfrs_interrupt() should try to keep the interrupt handler as fast as
- * possible.  After you are done making modifications, it is not a bad
- * idea to do:
- * 
- * gcc -S -DKERNEL -Wall -Wstrict-prototypes -O6 -fomit-frame-pointer serial.c
- *
- * and look at the resulting assemble code in serial.s.
- *
- * 				- Ted Ts'o (tytso@mit.edu), 7-Mar-93
- * -----------------------------------------------------------------------
- */
-
-static inline void receive_chars(struct mcf_serial *info)
-{
-	volatile unsigned char	*uartp;
-	struct tty_struct	*tty = info->port.tty;
-	unsigned char		status, ch, flag;
-
-	if (!tty)
-		return;
-
-	uartp = info->addr;
-
-	while ((status = uartp[MCFUART_USR]) & MCFUART_USR_RXREADY) {
-		ch = uartp[MCFUART_URB];
-		info->stats.rx++;
-
-#ifdef CONFIG_MAGIC_SYSRQ
-		if (mcfrs_console_inited && (info->line == mcfrs_console_port)) {
-			if (magic_sysrq_key(ch))
-				continue;
-		}
-#endif
-
-		flag = TTY_NORMAL;
-		if (status & MCFUART_USR_RXERR) {
-			uartp[MCFUART_UCR] = MCFUART_UCR_CMDRESETERR;
-			if (status & MCFUART_USR_RXBREAK) {
-				info->stats.rxbreak++;
-				flag = TTY_BREAK;
-			} else if (status & MCFUART_USR_RXPARITY) {
-				info->stats.rxparity++;
-				flag = TTY_PARITY;
-			} else if (status & MCFUART_USR_RXOVERRUN) {
-				info->stats.rxoverrun++;
-				flag = TTY_OVERRUN;
-			} else if (status & MCFUART_USR_RXFRAMING) {
-				info->stats.rxframing++;
-				flag = TTY_FRAME;
-			}
-		}
-		tty_insert_flip_char(tty, ch, flag);
-	}
-	tty_schedule_flip(tty);
-	return;
-}
-
-static inline void transmit_chars(struct mcf_serial *info)
-{
-	volatile unsigned char	*uartp;
-
-	uartp = info->addr;
-
-	if (info->x_char) {
-		/* Send special char - probably flow control */
-		uartp[MCFUART_UTB] = info->x_char;
-		info->x_char = 0;
-		info->stats.tx++;
-	}
-
-	if ((info->xmit_cnt <= 0) || info->port.tty->stopped) {
-		info->imr &= ~MCFUART_UIR_TXREADY;
-		uartp[MCFUART_UIMR] = info->imr;
-		return;
-	}
-
-	while (uartp[MCFUART_USR] & MCFUART_USR_TXREADY) {
-		uartp[MCFUART_UTB] = info->xmit_buf[info->xmit_tail++];
-		info->xmit_tail = info->xmit_tail & (SERIAL_XMIT_SIZE-1);
-		info->stats.tx++;
-		if (--info->xmit_cnt <= 0)
-			break;
-	}
-
-	if (info->xmit_cnt < WAKEUP_CHARS)
-		schedule_work(&info->tqueue);
-	return;
-}
-
-/*
- * This is the serial driver's generic interrupt routine
- */
-irqreturn_t mcfrs_interrupt(int irq, void *dev_id)
-{
-	struct mcf_serial	*info;
-	unsigned char		isr;
-
-	info = &mcfrs_table[(irq - IRQBASE)];
-	isr = info->addr[MCFUART_UISR] & info->imr;
-
-	if (isr & MCFUART_UIR_RXREADY)
-		receive_chars(info);
-	if (isr & MCFUART_UIR_TXREADY)
-		transmit_chars(info);
-	return IRQ_HANDLED;
-}
-
-/*
- * -------------------------------------------------------------------
- * Here ends the serial interrupt routines.
- * -------------------------------------------------------------------
- */
-
-static void mcfrs_offintr(struct work_struct *work)
-{
-	struct mcf_serial *info = container_of(work, struct mcf_serial, tqueue);
-	struct tty_struct *tty = info->port.tty;
-	
-	if (tty)
-		tty_wakeup(tty);
-}
-
-
-/*
- *	Change of state on a DCD line.
- */
-void mcfrs_modem_change(struct mcf_serial *info, int dcd)
-{
-	if (info->count == 0)
-		return;
-
-	if (info->flags & ASYNC_CHECK_CD) {
-		if (dcd)
-			wake_up_interruptible(&info->open_wait);
-		else 
-			schedule_work(&info->tqueue_hangup);
-	}
-}
-
-
-#ifdef MCFPP_DCD0
-
-unsigned short	mcfrs_ppstatus;
-
-/*
- * This subroutine is called when the RS_TIMER goes off. It is used
- * to monitor the state of the DCD lines - since they have no edge
- * sensors and interrupt generators.
- */
-static void mcfrs_timer(void)
-{
-	unsigned int	ppstatus, dcdval, i;
-
-	ppstatus = mcf_getppdata() & (MCFPP_DCD0 | MCFPP_DCD1);
-
-	if (ppstatus != mcfrs_ppstatus) {
-		for (i = 0; (i < 2); i++) {
-			dcdval = (i ? MCFPP_DCD1 : MCFPP_DCD0);
-			if ((ppstatus & dcdval) != (mcfrs_ppstatus & dcdval)) {
-				mcfrs_modem_change(&mcfrs_table[i],
-					((ppstatus & dcdval) ? 0 : 1));
-			}
-		}
-	}
-	mcfrs_ppstatus = ppstatus;
-
-	/* Re-arm timer */
-	mcfrs_timer_struct.expires = jiffies + HZ/25;
-	add_timer(&mcfrs_timer_struct);
-}
-
-#endif	/* MCFPP_DCD0 */
-
-
-/*
- * This routine is called from the scheduler tqueue when the interrupt
- * routine has signalled that a hangup has occurred. The path of
- * hangup processing is:
- *
- * 	serial interrupt routine -> (scheduler tqueue) ->
- * 	do_serial_hangup() -> tty->hangup() -> mcfrs_hangup()
- * 
- */
-static void do_serial_hangup(struct work_struct *work)
-{
-	struct mcf_serial *info = container_of(work, struct mcf_serial, tqueue_hangup);
-	struct tty_struct *tty = info->port.tty;
-	
-	if (tty)
-		tty_hangup(tty);
-}
-
-static int startup(struct mcf_serial * info)
-{
-	volatile unsigned char	*uartp;
-	unsigned long		flags;
-	
-	if (info->flags & ASYNC_INITIALIZED)
-		return 0;
-
-	if (!info->xmit_buf) {
-		info->xmit_buf = (unsigned char *) __get_free_page(GFP_KERNEL);
-		if (!info->xmit_buf)
-			return -ENOMEM;
-	}
-
-	local_irq_save(flags);
-
-#ifdef SERIAL_DEBUG_OPEN
-	printk("starting up ttyS%d (irq %d)...\n", info->line, info->irq);
-#endif
-
-	/*
-	 *	Reset UART, get it into known state...
-	 */
-	uartp = info->addr;
-	uartp[MCFUART_UCR] = MCFUART_UCR_CMDRESETRX;  /* reset RX */
-	uartp[MCFUART_UCR] = MCFUART_UCR_CMDRESETTX;  /* reset TX */
-	mcfrs_setsignals(info, 1, 1);
-
-	if (info->port.tty)
-		clear_bit(TTY_IO_ERROR, &info->port.tty->flags);
-	info->xmit_cnt = info->xmit_head = info->xmit_tail = 0;
-
-	/*
-	 * and set the speed of the serial port
-	 */
-	mcfrs_change_speed(info);
-
-	/*
-	 * Lastly enable the UART transmitter and receiver, and
-	 * interrupt enables.
-	 */
-	info->imr = MCFUART_UIR_RXREADY;
-	uartp[MCFUART_UCR] = MCFUART_UCR_RXENABLE | MCFUART_UCR_TXENABLE;
-	uartp[MCFUART_UIMR] = info->imr;
-
-	info->flags |= ASYNC_INITIALIZED;
-	local_irq_restore(flags);
-	return 0;
-}
-
-/*
- * This routine will shutdown a serial port; interrupts are disabled, and
- * DTR is dropped if the hangup on close termio flag is on.
- */
-static void shutdown(struct mcf_serial * info)
-{
-	volatile unsigned char	*uartp;
-	unsigned long		flags;
-
-	if (!(info->flags & ASYNC_INITIALIZED))
-		return;
-
-#ifdef SERIAL_DEBUG_OPEN
-	printk("Shutting down serial port %d (irq %d)....\n", info->line,
-	       info->irq);
-#endif
-	
-	local_irq_save(flags);
-
-	uartp = info->addr;
-	uartp[MCFUART_UIMR] = 0;  /* mask all interrupts */
-	uartp[MCFUART_UCR] = MCFUART_UCR_CMDRESETRX;  /* reset RX */
-	uartp[MCFUART_UCR] = MCFUART_UCR_CMDRESETTX;  /* reset TX */
-
-	if (!info->port.tty || (info->port.tty->termios->c_cflag & HUPCL))
-		mcfrs_setsignals(info, 0, 0);
-
-	if (info->xmit_buf) {
-		free_page((unsigned long) info->xmit_buf);
-		info->xmit_buf = 0;
-	}
-
-	if (info->port.tty)
-		set_bit(TTY_IO_ERROR, &info->port.tty->flags);
-	
-	info->flags &= ~ASYNC_INITIALIZED;
-	local_irq_restore(flags);
-}
-
-
-/*
- * This routine is called to set the UART divisor registers to match
- * the specified baud rate for a serial port.
- */
-static void mcfrs_change_speed(struct mcf_serial *info)
-{
-	volatile unsigned char	*uartp;
-	unsigned int		baudclk, cflag;
-	unsigned long		flags;
-	unsigned char		mr1, mr2;
-	int			i;
-#ifdef	CONFIG_M5272
-	unsigned int		fraction;
-#endif
-
-	if (!info->port.tty || !info->port.tty->termios)
-		return;
-	cflag = info->port.tty->termios->c_cflag;
-	if (info->addr == 0)
-		return;
-
-#if 0
-	printk("%s(%d): mcfrs_change_speed()\n", __FILE__, __LINE__);
-#endif
-
-	i = cflag & CBAUD;
-	if (i & CBAUDEX) {
-		i &= ~CBAUDEX;
-		if (i < 1 || i > 4)
-			info->port.tty->termios->c_cflag &= ~CBAUDEX;
-		else
-			i += 15;
-	}
-	if (i == 0) {
-		mcfrs_setsignals(info, 0, -1);
-		return;
-	}
-
-	/* compute the baudrate clock */
-#ifdef	CONFIG_M5272
-	/*
-	 * For the MCF5272, also compute the baudrate fraction.
-	 */
-	baudclk = (MCF_BUSCLK / mcfrs_baud_table[i]) / 32;
-	fraction = MCF_BUSCLK - (baudclk * 32 * mcfrs_baud_table[i]);
-	fraction *= 16;
-	fraction /= (32 * mcfrs_baud_table[i]);
-#else
-	baudclk = ((MCF_BUSCLK / mcfrs_baud_table[i]) + 16) / 32;
-#endif
-
-	info->baud = mcfrs_baud_table[i];
-
-	mr1 = MCFUART_MR1_RXIRQRDY | MCFUART_MR1_RXERRCHAR;
-	mr2 = 0;
-
-	switch (cflag & CSIZE) {
-	case CS5:	mr1 |= MCFUART_MR1_CS5; break;
-	case CS6:	mr1 |= MCFUART_MR1_CS6; break;
-	case CS7:	mr1 |= MCFUART_MR1_CS7; break;
-	case CS8:
-	default:	mr1 |= MCFUART_MR1_CS8; break;
-	}
-
-	if (cflag & PARENB) {
-		if (cflag & CMSPAR) {
-			if (cflag & PARODD)
-				mr1 |= MCFUART_MR1_PARITYMARK;
-			else
-				mr1 |= MCFUART_MR1_PARITYSPACE;
-		} else {
-			if (cflag & PARODD)
-				mr1 |= MCFUART_MR1_PARITYODD;
-			else
-				mr1 |= MCFUART_MR1_PARITYEVEN;
-		}
-	} else {
-		mr1 |= MCFUART_MR1_PARITYNONE;
-	}
-
-	if (cflag & CSTOPB)
-		mr2 |= MCFUART_MR2_STOP2;
-	else
-		mr2 |= MCFUART_MR2_STOP1;
-
-	if (cflag & CRTSCTS) {
-		mr1 |= MCFUART_MR1_RXRTS;
-		mr2 |= MCFUART_MR2_TXCTS;
-	}
-
-	if (cflag & CLOCAL)
-		info->flags &= ~ASYNC_CHECK_CD;
-	else
-		info->flags |= ASYNC_CHECK_CD;
-
-	uartp = info->addr;
-
-	local_irq_save(flags);
-#if 0
-	printk("%s(%d): mr1=%x mr2=%x baudclk=%x\n", __FILE__, __LINE__,
-		mr1, mr2, baudclk);
-#endif
-	/*
-	  Note: pg 12-16 of MCF5206e User's Manual states that a
-	  software reset should be performed prior to changing
-	  UMR1,2, UCSR, UACR, bit 7
-	*/
-	uartp[MCFUART_UCR] = MCFUART_UCR_CMDRESETRX;    /* reset RX */
-	uartp[MCFUART_UCR] = MCFUART_UCR_CMDRESETTX;    /* reset TX */
-	uartp[MCFUART_UCR] = MCFUART_UCR_CMDRESETMRPTR;	/* reset MR pointer */
-	uartp[MCFUART_UMR] = mr1;
-	uartp[MCFUART_UMR] = mr2;
-	uartp[MCFUART_UBG1] = (baudclk & 0xff00) >> 8;	/* set msb byte */
-	uartp[MCFUART_UBG2] = (baudclk & 0xff);		/* set lsb byte */
-#ifdef	CONFIG_M5272
-	uartp[MCFUART_UFPD] = (fraction & 0xf);		/* set fraction */
-#endif
-	uartp[MCFUART_UCSR] = MCFUART_UCSR_RXCLKTIMER | MCFUART_UCSR_TXCLKTIMER;
-	uartp[MCFUART_UCR] = MCFUART_UCR_RXENABLE | MCFUART_UCR_TXENABLE;
-	mcfrs_setsignals(info, 1, -1);
-	local_irq_restore(flags);
-	return;
-}
-
-static void mcfrs_flush_chars(struct tty_struct *tty)
-{
-	volatile unsigned char	*uartp;
-	struct mcf_serial	*info = (struct mcf_serial *)tty->driver_data;
-	unsigned long		flags;
-
-	if (serial_paranoia_check(info, tty->name, "mcfrs_flush_chars"))
-		return;
-
-	uartp = (volatile unsigned char *) info->addr;
-
-	/*
-	 * re-enable receiver interrupt
-	 */
-	local_irq_save(flags);
-	if ((!(info->imr & MCFUART_UIR_RXREADY)) &&
-	    (info->flags & ASYNC_INITIALIZED) ) {
-		info->imr |= MCFUART_UIR_RXREADY;
-		uartp[MCFUART_UIMR] = info->imr;
-	}
-	local_irq_restore(flags);
-
-	if (info->xmit_cnt <= 0 || tty->stopped || tty->hw_stopped ||
-	    !info->xmit_buf)
-		return;
-
-	/* Enable transmitter */
-	local_irq_save(flags);
-	info->imr |= MCFUART_UIR_TXREADY;
-	uartp[MCFUART_UIMR] = info->imr;
-	local_irq_restore(flags);
-}
-
-static int mcfrs_write(struct tty_struct * tty,
-		    const unsigned char *buf, int count)
-{
-	volatile unsigned char	*uartp;
-	struct mcf_serial	*info = (struct mcf_serial *)tty->driver_data;
-	unsigned long		flags;
-	int			c, total = 0;
-
-#if 0
-	printk("%s(%d): mcfrs_write(tty=%x,buf=%x,count=%d)\n",
-		__FILE__, __LINE__, (int)tty, (int)buf, count);
-#endif
-
-	if (serial_paranoia_check(info, tty->name, "mcfrs_write"))
-		return 0;
-
-	if (!tty || !info->xmit_buf)
-		return 0;
-	
-	local_save_flags(flags);
-	while (1) {
-		local_irq_disable();		
-		c = min(count, (int) min(((int)SERIAL_XMIT_SIZE) - info->xmit_cnt - 1,
-			((int)SERIAL_XMIT_SIZE) - info->xmit_head));
-		local_irq_restore(flags);
-
-		if (c <= 0)
-			break;
-
-		memcpy(info->xmit_buf + info->xmit_head, buf, c);
-
-		local_irq_disable();
-		info->xmit_head = (info->xmit_head + c) & (SERIAL_XMIT_SIZE-1);
-		info->xmit_cnt += c;
-		local_irq_restore(flags);
-
-		buf += c;
-		count -= c;
-		total += c;
-	}
-
-	local_irq_disable();
-	uartp = info->addr;
-	info->imr |= MCFUART_UIR_TXREADY;
-	uartp[MCFUART_UIMR] = info->imr;
-	local_irq_restore(flags);
-
-	return total;
-}
-
-static int mcfrs_write_room(struct tty_struct *tty)
-{
-	struct mcf_serial *info = (struct mcf_serial *)tty->driver_data;
-	int	ret;
-
-	if (serial_paranoia_check(info, tty->name, "mcfrs_write_room"))
-		return 0;
-	ret = SERIAL_XMIT_SIZE - info->xmit_cnt - 1;
-	if (ret < 0)
-		ret = 0;
-	return ret;
-}
-
-static int mcfrs_chars_in_buffer(struct tty_struct *tty)
-{
-	struct mcf_serial *info = (struct mcf_serial *)tty->driver_data;
-
-	if (serial_paranoia_check(info, tty->name, "mcfrs_chars_in_buffer"))
-		return 0;
-	return info->xmit_cnt;
-}
-
-static void mcfrs_flush_buffer(struct tty_struct *tty)
-{
-	struct mcf_serial	*info = (struct mcf_serial *)tty->driver_data;
-	unsigned long		flags;
-
-	if (serial_paranoia_check(info, tty->name, "mcfrs_flush_buffer"))
-		return;
-
-	local_irq_save(flags);
-	info->xmit_cnt = info->xmit_head = info->xmit_tail = 0;
-	local_irq_restore(flags);
-
-	tty_wakeup(tty);
-}
-
-/*
- * ------------------------------------------------------------
- * mcfrs_throttle()
- * 
- * This routine is called by the upper-layer tty layer to signal that
- * incoming characters should be throttled.
- * ------------------------------------------------------------
- */
-static void mcfrs_throttle(struct tty_struct * tty)
-{
-	struct mcf_serial *info = (struct mcf_serial *)tty->driver_data;
-#ifdef SERIAL_DEBUG_THROTTLE
-	char	buf[64];
-	
-	printk("throttle %s: %d....\n", tty_name(tty, buf),
-	       tty->ldisc.chars_in_buffer(tty));
-#endif
-
-	if (serial_paranoia_check(info, tty->name, "mcfrs_throttle"))
-		return;
-	
-	if (I_IXOFF(tty))
-		info->x_char = STOP_CHAR(tty);
-
-	/* Turn off RTS line (do this atomic) */
-}
-
-static void mcfrs_unthrottle(struct tty_struct * tty)
-{
-	struct mcf_serial *info = (struct mcf_serial *)tty->driver_data;
-#ifdef SERIAL_DEBUG_THROTTLE
-	char	buf[64];
-	
-	printk("unthrottle %s: %d....\n", tty_name(tty, buf),
-	       tty->ldisc.chars_in_buffer(tty));
-#endif
-
-	if (serial_paranoia_check(info, tty->name, "mcfrs_unthrottle"))
-		return;
-	
-	if (I_IXOFF(tty)) {
-		if (info->x_char)
-			info->x_char = 0;
-		else
-			info->x_char = START_CHAR(tty);
-	}
-
-	/* Assert RTS line (do this atomic) */
-}
-
-/*
- * ------------------------------------------------------------
- * mcfrs_ioctl() and friends
- * ------------------------------------------------------------
- */
-
-static int get_serial_info(struct mcf_serial * info,
-			   struct serial_struct * retinfo)
-{
-	struct serial_struct tmp;
-  
-	if (!retinfo)
-		return -EFAULT;
-	memset(&tmp, 0, sizeof(tmp));
-	tmp.type = info->type;
-	tmp.line = info->line;
-	tmp.port = (unsigned int) info->addr;
-	tmp.irq = info->irq;
-	tmp.flags = info->flags;
-	tmp.baud_base = info->baud_base;
-	tmp.close_delay = info->close_delay;
-	tmp.closing_wait = info->closing_wait;
-	tmp.custom_divisor = info->custom_divisor;
-	return copy_to_user(retinfo,&tmp,sizeof(*retinfo)) ? -EFAULT : 0;
-}
-
-static int set_serial_info(struct mcf_serial * info,
-			   struct serial_struct * new_info)
-{
-	struct serial_struct new_serial;
-	struct mcf_serial old_info;
-	int 	retval = 0;
-
-	if (!new_info)
-		return -EFAULT;
-	if (copy_from_user(&new_serial,new_info,sizeof(new_serial)))
-		return -EFAULT;
-	old_info = *info;
-
-	if (!capable(CAP_SYS_ADMIN)) {
-		if ((new_serial.baud_base != info->baud_base) ||
-		    (new_serial.type != info->type) ||
-		    (new_serial.close_delay != info->close_delay) ||
-		    ((new_serial.flags & ~ASYNC_USR_MASK) !=
-		     (info->flags & ~ASYNC_USR_MASK)))
-			return -EPERM;
-		info->flags = ((info->flags & ~ASYNC_USR_MASK) |
-			       (new_serial.flags & ASYNC_USR_MASK));
-		info->custom_divisor = new_serial.custom_divisor;
-		goto check_and_exit;
-	}
-
-	if (info->count > 1)
-		return -EBUSY;
-
-	/*
-	 * OK, past this point, all the error checking has been done.
-	 * At this point, we start making changes.....
-	 */
-
-	info->baud_base = new_serial.baud_base;
-	info->flags = ((info->flags & ~ASYNC_FLAGS) |
-			(new_serial.flags & ASYNC_FLAGS));
-	info->type = new_serial.type;
-	info->close_delay = new_serial.close_delay;
-	info->closing_wait = new_serial.closing_wait;
-
-check_and_exit:
-	retval = startup(info);
-	return retval;
-}
-
-/*
- * get_lsr_info - get line status register info
- *
- * Purpose: Let user call ioctl() to get info when the UART physically
- * 	    is emptied.  On bus types like RS485, the transmitter must
- * 	    release the bus after transmitting. This must be done when
- * 	    the transmit shift register is empty, not be done when the
- * 	    transmit holding register is empty.  This functionality
- * 	    allows an RS485 driver to be written in user space. 
- */
-static int get_lsr_info(struct mcf_serial * info, unsigned int *value)
-{
-	volatile unsigned char	*uartp;
-	unsigned long		flags;
-	unsigned char		status;
-
-	local_irq_save(flags);
-	uartp = info->addr;
-	status = (uartp[MCFUART_USR] & MCFUART_USR_TXEMPTY) ? TIOCSER_TEMT : 0;
-	local_irq_restore(flags);
-
-	return put_user(status,value);
-}
-
-/*
- * This routine sends a break character out the serial port.
- */
-static void send_break(	struct mcf_serial * info, int duration)
-{
-	volatile unsigned char	*uartp;
-	unsigned long		flags;
-
-	if (!info->addr)
-		return;
-	set_current_state(TASK_INTERRUPTIBLE);
-	uartp = info->addr;
-
-	local_irq_save(flags);
-	uartp[MCFUART_UCR] = MCFUART_UCR_CMDBREAKSTART;
-	schedule_timeout(duration);
-	uartp[MCFUART_UCR] = MCFUART_UCR_CMDBREAKSTOP;
-	local_irq_restore(flags);
-}
-
-static int mcfrs_tiocmget(struct tty_struct *tty, struct file *file)
-{
-	struct mcf_serial * info = (struct mcf_serial *)tty->driver_data;
-
-	if (serial_paranoia_check(info, tty->name, "mcfrs_ioctl"))
-		return -ENODEV;
-	if (tty->flags & (1 << TTY_IO_ERROR))
-		return -EIO;
-
-	return mcfrs_getsignals(info);
-}
-
-static int mcfrs_tiocmset(struct tty_struct *tty, struct file *file,
-			  unsigned int set, unsigned int clear)
-{
-	struct mcf_serial * info = (struct mcf_serial *)tty->driver_data;
-	int rts = -1, dtr = -1;
-
-	if (serial_paranoia_check(info, tty->name, "mcfrs_ioctl"))
-		return -ENODEV;
-	if (tty->flags & (1 << TTY_IO_ERROR))
-		return -EIO;
-
-	if (set & TIOCM_RTS)
-		rts = 1;
-	if (set & TIOCM_DTR)
-		dtr = 1;
-	if (clear & TIOCM_RTS)
-		rts = 0;
-	if (clear & TIOCM_DTR)
-		dtr = 0;
-
-	mcfrs_setsignals(info, dtr, rts);
-
-	return 0;
-}
-
-static int mcfrs_ioctl(struct tty_struct *tty, struct file * file,
-		    unsigned int cmd, unsigned long arg)
-{
-	struct mcf_serial * info = (struct mcf_serial *)tty->driver_data;
-	int retval, error;
-
-	if (serial_paranoia_check(info, tty->name, "mcfrs_ioctl"))
-		return -ENODEV;
-
-	if ((cmd != TIOCGSERIAL) && (cmd != TIOCSSERIAL) &&
-	    (cmd != TIOCSERCONFIG) && (cmd != TIOCSERGWILD)  &&
-	    (cmd != TIOCSERSWILD) && (cmd != TIOCSERGSTRUCT)) {
-		if (tty->flags & (1 << TTY_IO_ERROR))
-		    return -EIO;
-	}
-	
-	switch (cmd) {
-		case TCSBRK:	/* SVID version: non-zero arg --> no break */
-			retval = tty_check_change(tty);
-			if (retval)
-				return retval;
-			tty_wait_until_sent(tty, 0);
-			if (!arg)
-				send_break(info, HZ/4);	/* 1/4 second */
-			return 0;
-		case TCSBRKP:	/* support for POSIX tcsendbreak() */
-			retval = tty_check_change(tty);
-			if (retval)
-				return retval;
-			tty_wait_until_sent(tty, 0);
-			send_break(info, arg ? arg*(HZ/10) : HZ/4);
-			return 0;
-		case TIOCGSERIAL:
-			if (access_ok(VERIFY_WRITE, (void *) arg,
-						sizeof(struct serial_struct)))
-				return get_serial_info(info,
-					       (struct serial_struct *) arg);
-			return -EFAULT;
-		case TIOCSSERIAL:
-			return set_serial_info(info,
-					       (struct serial_struct *) arg);
-		case TIOCSERGETLSR: /* Get line status register */
-			if (access_ok(VERIFY_WRITE, (void *) arg,
-						sizeof(unsigned int)))
-				return get_lsr_info(info, (unsigned int *) arg);
-			return -EFAULT;
-		case TIOCSERGSTRUCT:
-			error = copy_to_user((struct mcf_serial *) arg,
-				    info, sizeof(struct mcf_serial));
-			if (error)
-				return -EFAULT;
-			return 0;
-			
-#ifdef TIOCSET422
-		case TIOCSET422: {
-			unsigned int val;
-			get_user(val, (unsigned int *) arg);
-			mcf_setpa(MCFPP_PA11, (val ? 0 : MCFPP_PA11));
-			break;
-		}
-		case TIOCGET422: {
-			unsigned int val;
-			val = (mcf_getpa() & MCFPP_PA11) ? 0 : 1;
-			put_user(val, (unsigned int *) arg);
-			break;
-		}
-#endif
-
-		default:
-			return -ENOIOCTLCMD;
-		}
-	return 0;
-}
-
-static void mcfrs_set_termios(struct tty_struct *tty, struct ktermios *old_termios)
-{
-	struct mcf_serial *info = (struct mcf_serial *)tty->driver_data;
-
-	if (tty->termios->c_cflag == old_termios->c_cflag)
-		return;
-
-	mcfrs_change_speed(info);
-
-	if ((old_termios->c_cflag & CRTSCTS) &&
-	    !(tty->termios->c_cflag & CRTSCTS)) {
-		tty->hw_stopped = 0;
-		mcfrs_setsignals(info, -1, 1);
-#if 0
-		mcfrs_start(tty);
-#endif
-	}
-}
-
-/*
- * ------------------------------------------------------------
- * mcfrs_close()
- * 
- * This routine is called when the serial port gets closed.  First, we
- * wait for the last remaining data to be sent.  Then, we unlink its
- * S structure from the interrupt chain if necessary, and we free
- * that IRQ if nothing is left in the chain.
- * ------------------------------------------------------------
- */
-static void mcfrs_close(struct tty_struct *tty, struct file * filp)
-{
-	volatile unsigned char	*uartp;
-	struct mcf_serial	*info = (struct mcf_serial *)tty->driver_data;
-	unsigned long		flags;
-
-	if (!info || serial_paranoia_check(info, tty->name, "mcfrs_close"))
-		return;
-	
-	local_irq_save(flags);
-	
-	if (tty_hung_up_p(filp)) {
-		local_irq_restore(flags);
-		return;
-	}
-	
-#ifdef SERIAL_DEBUG_OPEN
-	printk("mcfrs_close ttyS%d, count = %d\n", info->line, info->count);
-#endif
-	if ((tty->count == 1) && (info->count != 1)) {
-		/*
-		 * Uh, oh.  tty->count is 1, which means that the tty
-		 * structure will be freed.  Info->count should always
-		 * be one in these conditions.  If it's greater than
-		 * one, we've got real problems, since it means the
-		 * serial port won't be shutdown.
-		 */
-		printk("MCFRS: bad serial port count; tty->count is 1, "
-		       "info->count is %d\n", info->count);
-		info->count = 1;
-	}
-	if (--info->count < 0) {
-		printk("MCFRS: bad serial port count for ttyS%d: %d\n",
-		       info->line, info->count);
-		info->count = 0;
-	}
-	if (info->count) {
-		local_irq_restore(flags);
-		return;
-	}
-	info->flags |= ASYNC_CLOSING;
-
-	/*
-	 * Now we wait for the transmit buffer to clear; and we notify 
-	 * the line discipline to only process XON/XOFF characters.
-	 */
-	tty->closing = 1;
-	if (info->closing_wait != ASYNC_CLOSING_WAIT_NONE)
-		tty_wait_until_sent(tty, info->closing_wait);
-
-	/*
-	 * At this point we stop accepting input.  To do this, we
-	 * disable the receive line status interrupts, and tell the
-	 * interrupt driver to stop checking the data ready bit in the
-	 * line status register.
-	 */
-	info->imr &= ~MCFUART_UIR_RXREADY;
-	uartp = info->addr;
-	uartp[MCFUART_UIMR] = info->imr;
-
-#if 0
-	/* FIXME: do we need to keep this enabled for console?? */
-	if (mcfrs_console_inited && (mcfrs_console_port == info->line)) {
-		/* Do not disable the UART */ ;
-	} else
-#endif
-	shutdown(info);
-	mcfrs_flush_buffer(tty);
-	tty_ldisc_flush(tty);
-	
-	tty->closing = 0;
-	info->event = 0;
-	info->port.tty = NULL;
-#if 0	
-	if (tty->ldisc.num != ldiscs[N_TTY].num) {
-		if (tty->ldisc.close)
-			(tty->ldisc.close)(tty);
-		tty->ldisc = ldiscs[N_TTY];
-		tty->termios->c_line = N_TTY;
-		if (tty->ldisc.open)
-			(tty->ldisc.open)(tty);
-	}
-#endif	
-	if (info->blocked_open) {
-		if (info->close_delay) {
-			msleep_interruptible(jiffies_to_msecs(info->close_delay));
-		}
-		wake_up_interruptible(&info->open_wait);
-	}
-	info->flags &= ~(ASYNC_NORMAL_ACTIVE|ASYNC_CLOSING);
-	wake_up_interruptible(&info->close_wait);
-	local_irq_restore(flags);
-}
-
-/*
- * mcfrs_wait_until_sent() --- wait until the transmitter is empty
- */
-static void
-mcfrs_wait_until_sent(struct tty_struct *tty, int timeout)
-{
-#ifdef	CONFIG_M5272
-#define	MCF5272_FIFO_SIZE	25		/* fifo size + shift reg */
-
-	struct mcf_serial * info = (struct mcf_serial *)tty->driver_data;
-	volatile unsigned char *uartp;
-	unsigned long orig_jiffies, fifo_time, char_time, fifo_cnt;
-
-	if (serial_paranoia_check(info, tty->name, "mcfrs_wait_until_sent"))
-		return;
-
-	orig_jiffies = jiffies;
-
-	/*
-	 * Set the check interval to be 1/5 of the approximate time
-	 * to send the entire fifo, and make it at least 1.  The check
-	 * interval should also be less than the timeout.
-	 *
-	 * Note: we have to use pretty tight timings here to satisfy
-	 * the NIST-PCTS.
-	 */
-	lock_kernel();
-
-	fifo_time = (MCF5272_FIFO_SIZE * HZ * 10) / info->baud;
-	char_time = fifo_time / 5;
-	if (char_time == 0)
-		char_time = 1;
-	if (timeout && timeout < char_time)
-		char_time = timeout;
-
-	/*
-	 * Clamp the timeout period at 2 * the time to empty the
-	 * fifo.  Just to be safe, set the minimum at .5 seconds.
-	 */
-	fifo_time *= 2;
-	if (fifo_time < (HZ/2))
-		fifo_time = HZ/2;
-	if (!timeout || timeout > fifo_time)
-		timeout = fifo_time;
-
-	/*
-	 * Account for the number of bytes in the UART
-	 * transmitter FIFO plus any byte being shifted out.
-	 */
-	uartp = (volatile unsigned char *) info->addr;
-	for (;;) {
-		fifo_cnt = (uartp[MCFUART_UTF] & MCFUART_UTF_TXB);
-		if ((uartp[MCFUART_USR] & (MCFUART_USR_TXREADY|
-				MCFUART_USR_TXEMPTY)) ==
-			MCFUART_USR_TXREADY)
-			fifo_cnt++;
-		if (fifo_cnt == 0)
-			break;
-		msleep_interruptible(jiffies_to_msecs(char_time));
-		if (signal_pending(current))
-			break;
-		if (timeout && time_after(jiffies, orig_jiffies + timeout))
-			break;
-	}
-	unlock_kernel();
-#else
-	/*
-	 * For the other coldfire models, assume all data has been sent
-	 */
-#endif
-}
-
-/*
- * mcfrs_hangup() --- called by tty_hangup() when a hangup is signaled.
- */
-void mcfrs_hangup(struct tty_struct *tty)
-{
-	struct mcf_serial * info = (struct mcf_serial *)tty->driver_data;
-	
-	if (serial_paranoia_check(info, tty->name, "mcfrs_hangup"))
-		return;
-	
-	mcfrs_flush_buffer(tty);
-	shutdown(info);
-	info->event = 0;
-	info->count = 0;
-	info->flags &= ~ASYNC_NORMAL_ACTIVE;
-	info->port.tty = NULL;
-	wake_up_interruptible(&info->open_wait);
-}
-
-/*
- * ------------------------------------------------------------
- * mcfrs_open() and friends
- * ------------------------------------------------------------
- */
-static int block_til_ready(struct tty_struct *tty, struct file * filp,
-			   struct mcf_serial *info)
-{
-	DECLARE_WAITQUEUE(wait, current);
-	int	retval;
-	int	do_clocal = 0;
-
-	/*
-	 * If the device is in the middle of being closed, then block
-	 * until it's done, and then try again.
-	 */
-	if (info->flags & ASYNC_CLOSING) {
-		interruptible_sleep_on(&info->close_wait);
-#ifdef SERIAL_DO_RESTART
-		if (info->flags & ASYNC_HUP_NOTIFY)
-			return -EAGAIN;
-		else
-			return -ERESTARTSYS;
-#else
-		return -EAGAIN;
-#endif
-	}
-	
-	/*
-	 * If non-blocking mode is set, or the port is not enabled,
-	 * then make the check up front and then exit.
-	 */
-	if ((filp->f_flags & O_NONBLOCK) ||
-	    (tty->flags & (1 << TTY_IO_ERROR))) {
-		info->flags |= ASYNC_NORMAL_ACTIVE;
-		return 0;
-	}
-
-	if (tty->termios->c_cflag & CLOCAL)
-		do_clocal = 1;
-
-	/*
-	 * Block waiting for the carrier detect and the line to become
-	 * free (i.e., not in use by the callout).  While we are in
-	 * this loop, info->count is dropped by one, so that
-	 * mcfrs_close() knows when to free things.  We restore it upon
-	 * exit, either normal or abnormal.
-	 */
-	retval = 0;
-	add_wait_queue(&info->open_wait, &wait);
-#ifdef SERIAL_DEBUG_OPEN
-	printk("block_til_ready before block: ttyS%d, count = %d\n",
-	       info->line, info->count);
-#endif
-	info->count--;
-	info->blocked_open++;
-	while (1) {
-		local_irq_disable();
-		mcfrs_setsignals(info, 1, 1);
-		local_irq_enable();
-		current->state = TASK_INTERRUPTIBLE;
-		if (tty_hung_up_p(filp) ||
-		    !(info->flags & ASYNC_INITIALIZED)) {
-#ifdef SERIAL_DO_RESTART
-			if (info->flags & ASYNC_HUP_NOTIFY)
-				retval = -EAGAIN;
-			else
-				retval = -ERESTARTSYS;	
-#else
-			retval = -EAGAIN;
-#endif
-			break;
-		}
-		if (!(info->flags & ASYNC_CLOSING) &&
-		    (do_clocal || (mcfrs_getsignals(info) & TIOCM_CD)))
-			break;
-		if (signal_pending(current)) {
-			retval = -ERESTARTSYS;
-			break;
-		}
-#ifdef SERIAL_DEBUG_OPEN
-		printk("block_til_ready blocking: ttyS%d, count = %d\n",
-		       info->line, info->count);
-#endif
-		schedule();
-	}
-	current->state = TASK_RUNNING;
-	remove_wait_queue(&info->open_wait, &wait);
-	if (!tty_hung_up_p(filp))
-		info->count++;
-	info->blocked_open--;
-#ifdef SERIAL_DEBUG_OPEN
-	printk("block_til_ready after blocking: ttyS%d, count = %d\n",
-	       info->line, info->count);
-#endif
-	if (retval)
-		return retval;
-	info->flags |= ASYNC_NORMAL_ACTIVE;
-	return 0;
-}	
-
-/*
- * This routine is called whenever a serial port is opened. It
- * enables interrupts for a serial port, linking in its structure into
- * the IRQ chain.   It also performs the serial-specific
- * initialization for the tty structure.
- */
-int mcfrs_open(struct tty_struct *tty, struct file * filp)
-{
-	struct mcf_serial	*info;
-	int 			retval, line;
-
-	line = tty->index;
-	if ((line < 0) || (line >= NR_PORTS))
-		return -ENODEV;
-	info = mcfrs_table + line;
-	if (serial_paranoia_check(info, tty->name, "mcfrs_open"))
-		return -ENODEV;
-#ifdef SERIAL_DEBUG_OPEN
-	printk("mcfrs_open %s, count = %d\n", tty->name, info->count);
-#endif
-	info->count++;
-	tty->driver_data = info;
-	info->port.tty = tty;
-
-	/*
-	 * Start up serial port
-	 */
-	retval = startup(info);
-	if (retval)
-		return retval;
-
-	retval = block_til_ready(tty, filp, info);
-	if (retval) {
-#ifdef SERIAL_DEBUG_OPEN
-		printk("mcfrs_open returning after block_til_ready with %d\n",
-		       retval);
-#endif
-		return retval;
-	}
-
-#ifdef SERIAL_DEBUG_OPEN
-	printk("mcfrs_open %s successful...\n", tty->name);
-#endif
-	return 0;
-}
-
-/*
- *	Based on the line number set up the internal interrupt stuff.
- */
-static void mcfrs_irqinit(struct mcf_serial *info)
-{
-#if defined(CONFIG_M5272)
-	volatile unsigned long	*icrp;
-	volatile unsigned long	*portp;
-	volatile unsigned char	*uartp;
-
-	uartp = info->addr;
-	icrp = (volatile unsigned long *) (MCF_MBAR + MCFSIM_ICR2);
-
-	switch (info->line) {
-	case 0:
-		*icrp = 0xe0000000;
-		break;
-	case 1:
-		*icrp = 0x0e000000;
-		break;
-	default:
-		printk("MCFRS: don't know how to handle UART %d interrupt?\n",
-			info->line);
-		return;
-	}
-
-	/* Enable the output lines for the serial ports */
-	portp = (volatile unsigned long *) (MCF_MBAR + MCFSIM_PBCNT);
-	*portp = (*portp & ~0x000000ff) | 0x00000055;
-	portp = (volatile unsigned long *) (MCF_MBAR + MCFSIM_PDCNT);
-	*portp = (*portp & ~0x000003fc) | 0x000002a8;
-#elif defined(CONFIG_M523x) || defined(CONFIG_M527x) || defined(CONFIG_M528x)
-	volatile unsigned char *icrp, *uartp;
-	volatile unsigned long *imrp;
-
-	uartp = info->addr;
-
-	icrp = (volatile unsigned char *) (MCF_MBAR + MCFICM_INTC0 +
-		MCFINTC_ICR0 + MCFINT_UART0 + info->line);
-	*icrp = 0x30 + info->line; /* level 6, line based priority */
-
-	imrp = (volatile unsigned long *) (MCF_MBAR + MCFICM_INTC0 +
-		MCFINTC_IMRL);
-	*imrp &= ~((1 << (info->irq - MCFINT_VECBASE)) | 1);
-#if defined(CONFIG_M527x)
-	{
-		/*
-		 * External Pin Mask Setting & Enable External Pin for Interface
-		 * mrcbis@aliceposta.it
-        	 */
-		u16 *serpin_enable_mask;
-		serpin_enable_mask = (u16 *) (MCF_IPSBAR + MCF_GPIO_PAR_UART);
-		if (info->line == 0)
-			*serpin_enable_mask |= UART0_ENABLE_MASK;
-		else if (info->line == 1)
-			*serpin_enable_mask |= UART1_ENABLE_MASK;
-		else if (info->line == 2)
-			*serpin_enable_mask |= UART2_ENABLE_MASK;
-	}
-#endif
-#if defined(CONFIG_M528x)
-	/* make sure PUAPAR is set for UART0 and UART1 */
-	if (info->line < 2) {
-		volatile unsigned char *portp = (volatile unsigned char *) (MCF_MBAR + MCF5282_GPIO_PUAPAR);
-		*portp |= (0x03 << (info->line * 2));
-	}
-#endif
-#elif defined(CONFIG_M520x)
-	volatile unsigned char *icrp, *uartp;
-	volatile unsigned long *imrp;
-
-	uartp = info->addr;
-
-	icrp = (volatile unsigned char *) (MCF_MBAR + MCFICM_INTC0 +
-		MCFINTC_ICR0 + MCFINT_UART0 + info->line);
-	*icrp = 0x03;
-
-	imrp = (volatile unsigned long *) (MCF_MBAR + MCFICM_INTC0 +
-		MCFINTC_IMRL);
-	*imrp &= ~((1 << (info->irq - MCFINT_VECBASE)) | 1);
-	if (info->line < 2) {
-		unsigned short *uart_par;
-		uart_par = (unsigned short *)(MCF_IPSBAR + MCF_GPIO_PAR_UART);
-		if (info->line == 0)
-			*uart_par |=  MCF_GPIO_PAR_UART_PAR_UTXD0
-				  | MCF_GPIO_PAR_UART_PAR_URXD0;
-		else if (info->line == 1)
-			*uart_par |=  MCF_GPIO_PAR_UART_PAR_UTXD1
-				  | MCF_GPIO_PAR_UART_PAR_URXD1;
-		} else if (info->line == 2) {
-			unsigned char *feci2c_par;
-			feci2c_par = (unsigned char *)(MCF_IPSBAR +  MCF_GPIO_PAR_FECI2C);
-			*feci2c_par &= ~0x0F;
-			*feci2c_par |=  MCF_GPIO_PAR_FECI2C_PAR_SCL_UTXD2
-				    | MCF_GPIO_PAR_FECI2C_PAR_SDA_URXD2;
-		}
-#elif defined(CONFIG_M532x)
-	volatile unsigned char *uartp;
-	uartp = info->addr;
-	switch (info->line) {
-	case 0:
-		MCF_INTC0_ICR26 = 0x3;
-		MCF_INTC0_CIMR = 26;
-		/* GPIO initialization */
-		MCF_GPIO_PAR_UART |= 0x000F;
-		break;
-	case 1:
-		MCF_INTC0_ICR27 = 0x3;
-		MCF_INTC0_CIMR = 27;
-		/* GPIO initialization */
-		MCF_GPIO_PAR_UART |= 0x0FF0;
-		break;
-	case 2:
-		MCF_INTC0_ICR28 = 0x3;
-		MCF_INTC0_CIMR = 28;
-		/* GPIOs also must be initalized, depends on board */
-		break;
-	}
-#else
-	volatile unsigned char	*icrp, *uartp;
-
-	switch (info->line) {
-	case 0:
-		icrp = (volatile unsigned char *) (MCF_MBAR + MCFSIM_UART1ICR);
-		*icrp = /*MCFSIM_ICR_AUTOVEC |*/ MCFSIM_ICR_LEVEL6 |
-			MCFSIM_ICR_PRI1;
-		mcf_setimr(mcf_getimr() & ~MCFSIM_IMR_UART1);
-		break;
-	case 1:
-		icrp = (volatile unsigned char *) (MCF_MBAR + MCFSIM_UART2ICR);
-		*icrp = /*MCFSIM_ICR_AUTOVEC |*/ MCFSIM_ICR_LEVEL6 |
-			MCFSIM_ICR_PRI2;
-		mcf_setimr(mcf_getimr() & ~MCFSIM_IMR_UART2);
-		break;
-	default:
-		printk("MCFRS: don't know how to handle UART %d interrupt?\n",
-			info->line);
-		return;
-	}
-
-	uartp = info->addr;
-	uartp[MCFUART_UIVR] = info->irq;
-#endif
-
-	/* Clear mask, so no surprise interrupts. */
-	uartp[MCFUART_UIMR] = 0;
-
-	if (request_irq(info->irq, mcfrs_interrupt, IRQF_DISABLED,
-	    "ColdFire UART", NULL)) {
-		printk("MCFRS: Unable to attach ColdFire UART %d interrupt "
-			"vector=%d\n", info->line, info->irq);
-	}
-
-	return;
-}
-
-
-char *mcfrs_drivername = "ColdFire internal UART serial driver version 1.00\n";
-
-
-/*
- * Serial stats reporting...
- */
-int mcfrs_readproc(char *page, char **start, off_t off, int count,
-		         int *eof, void *data)
-{
-	struct mcf_serial	*info;
-	char			str[20];
-	int			len, sigs, i;
-
-	len = sprintf(page, mcfrs_drivername);
-	for (i = 0; (i < NR_PORTS); i++) {
-		info = &mcfrs_table[i];
-		len += sprintf((page + len), "%d: port:%x irq=%d baud:%d ",
-			i, (unsigned int) info->addr, info->irq, info->baud);
-		if (info->stats.rx || info->stats.tx)
-			len += sprintf((page + len), "tx:%d rx:%d ",
-			info->stats.tx, info->stats.rx);
-		if (info->stats.rxframing)
-			len += sprintf((page + len), "fe:%d ",
-			info->stats.rxframing);
-		if (info->stats.rxparity)
-			len += sprintf((page + len), "pe:%d ",
-			info->stats.rxparity);
-		if (info->stats.rxbreak)
-			len += sprintf((page + len), "brk:%d ",
-			info->stats.rxbreak);
-		if (info->stats.rxoverrun)
-			len += sprintf((page + len), "oe:%d ",
-			info->stats.rxoverrun);
-
-		str[0] = str[1] = 0;
-		if ((sigs = mcfrs_getsignals(info))) {
-			if (sigs & TIOCM_RTS)
-				strcat(str, "|RTS");
-			if (sigs & TIOCM_CTS)
-				strcat(str, "|CTS");
-			if (sigs & TIOCM_DTR)
-				strcat(str, "|DTR");
-			if (sigs & TIOCM_CD)
-				strcat(str, "|CD");
-		}
-
-		len += sprintf((page + len), "%s\n", &str[1]);
-	}
-
-	return(len);
-}
-
-
-/* Finally, routines used to initialize the serial driver. */
-
-static void show_serial_version(void)
-{
-	printk(mcfrs_drivername);
-}
-
-static const struct tty_operations mcfrs_ops = {
-	.open = mcfrs_open,
-	.close = mcfrs_close,
-	.write = mcfrs_write,
-	.flush_chars = mcfrs_flush_chars,
-	.write_room = mcfrs_write_room,
-	.chars_in_buffer = mcfrs_chars_in_buffer,
-	.flush_buffer = mcfrs_flush_buffer,
-	.ioctl = mcfrs_ioctl,
-	.throttle = mcfrs_throttle,
-	.unthrottle = mcfrs_unthrottle,
-	.set_termios = mcfrs_set_termios,
-	.stop = mcfrs_stop,
-	.start = mcfrs_start,
-	.hangup = mcfrs_hangup,
-	.read_proc = mcfrs_readproc,
-	.wait_until_sent = mcfrs_wait_until_sent,
- 	.tiocmget = mcfrs_tiocmget,
-	.tiocmset = mcfrs_tiocmset,
-};
-
-/* mcfrs_init inits the driver */
-static int __init
-mcfrs_init(void)
-{
-	struct mcf_serial	*info;
-	unsigned long		flags;
-	int			i;
-
-	/* Setup base handler, and timer table. */
-#ifdef MCFPP_DCD0
-	init_timer(&mcfrs_timer_struct);
-	mcfrs_timer_struct.function = mcfrs_timer;
-	mcfrs_timer_struct.data = 0;
-	mcfrs_timer_struct.expires = jiffies + HZ/25;
-	add_timer(&mcfrs_timer_struct);
-	mcfrs_ppstatus = mcf_getppdata() & (MCFPP_DCD0 | MCFPP_DCD1);
-#endif
-	mcfrs_serial_driver = alloc_tty_driver(NR_PORTS);
-	if (!mcfrs_serial_driver)
-		return -ENOMEM;
-
-	show_serial_version();
-
-	/* Initialize the tty_driver structure */
-	mcfrs_serial_driver->owner = THIS_MODULE;
-	mcfrs_serial_driver->name = "ttyS";
-	mcfrs_serial_driver->driver_name = "mcfserial";
-	mcfrs_serial_driver->major = TTY_MAJOR;
-	mcfrs_serial_driver->minor_start = 64;
-	mcfrs_serial_driver->type = TTY_DRIVER_TYPE_SERIAL;
-	mcfrs_serial_driver->subtype = SERIAL_TYPE_NORMAL;
-	mcfrs_serial_driver->init_termios = tty_std_termios;
-
-	mcfrs_serial_driver->init_termios.c_cflag =
-		mcfrs_console_cbaud | CS8 | CREAD | HUPCL | CLOCAL;
-	mcfrs_serial_driver->flags = TTY_DRIVER_REAL_RAW;
-
-	tty_set_operations(mcfrs_serial_driver, &mcfrs_ops);
-
-	if (tty_register_driver(mcfrs_serial_driver)) {
-		printk("MCFRS: Couldn't register serial driver\n");
-		put_tty_driver(mcfrs_serial_driver);
-		return(-EBUSY);
-	}
-
-	local_irq_save(flags);
-
-	/*
-	 *	Configure all the attached serial ports.
-	 */
-	for (i = 0, info = mcfrs_table; (i < NR_PORTS); i++, info++) {
-		info->magic = SERIAL_MAGIC;
-		info->line = i;
-		info->port.tty = NULL;
-		info->custom_divisor = 16;
-		info->close_delay = 50;
-		info->closing_wait = 3000;
-		info->x_char = 0;
-		info->event = 0;
-		info->count = 0;
-		info->blocked_open = 0;
-		INIT_WORK(&info->tqueue, mcfrs_offintr);
-		INIT_WORK(&info->tqueue_hangup, do_serial_hangup);
-		init_waitqueue_head(&info->open_wait);
-		init_waitqueue_head(&info->close_wait);
-
-		info->imr = 0;
-		mcfrs_setsignals(info, 0, 0);
-		mcfrs_irqinit(info);
-
-		printk("ttyS%d at 0x%04x (irq = %d)", info->line,
-			(unsigned int) info->addr, info->irq);
-		printk(" is a builtin ColdFire UART\n");
-	}
-
-	local_irq_restore(flags);
-	return 0;
-}
-
-module_init(mcfrs_init);
-
-/****************************************************************************/
-/*                          Serial Console                                  */
-/****************************************************************************/
-
-/*
- *	Quick and dirty UART initialization, for console output.
- */
-
-void mcfrs_init_console(void)
-{
-	volatile unsigned char	*uartp;
-	unsigned int		clk;
-
-	/*
-	 *	Reset UART, get it into known state...
-	 */
-	uartp = (volatile unsigned char *) (MCF_MBAR +
-		(mcfrs_console_port ? MCFUART_BASE2 : MCFUART_BASE1));
-
-	uartp[MCFUART_UCR] = MCFUART_UCR_CMDRESETRX;  /* reset RX */
-	uartp[MCFUART_UCR] = MCFUART_UCR_CMDRESETTX;  /* reset TX */
-	uartp[MCFUART_UCR] = MCFUART_UCR_CMDRESETMRPTR;  /* reset MR pointer */
-
-	/*
-	 * Set port for defined baud , 8 data bits, 1 stop bit, no parity.
-	 */
-	uartp[MCFUART_UMR] = MCFUART_MR1_PARITYNONE | MCFUART_MR1_CS8;
-	uartp[MCFUART_UMR] = MCFUART_MR2_STOP1;
-
-#ifdef	CONFIG_M5272
-{
-	/*
-	 * For the MCF5272, also compute the baudrate fraction.
-	 */
-	int fraction = MCF_BUSCLK - (clk * 32 * mcfrs_console_baud);
-	fraction *= 16;
-	fraction /= (32 * mcfrs_console_baud);
-	uartp[MCFUART_UFPD] = (fraction & 0xf);		/* set fraction */
-	clk = (MCF_BUSCLK / mcfrs_console_baud) / 32;
-}
-#else
-	clk = ((MCF_BUSCLK / mcfrs_console_baud) + 16) / 32; /* set baud */
-#endif
-
-	uartp[MCFUART_UBG1] = (clk & 0xff00) >> 8;  /* set msb baud */
-	uartp[MCFUART_UBG2] = (clk & 0xff);  /* set lsb baud */
-	uartp[MCFUART_UCSR] = MCFUART_UCSR_RXCLKTIMER | MCFUART_UCSR_TXCLKTIMER;
-	uartp[MCFUART_UCR] = MCFUART_UCR_RXENABLE | MCFUART_UCR_TXENABLE;
-
-	mcfrs_console_inited++;
-	return;
-}
-
-
-/*
- *	Setup for console. Argument comes from the boot command line.
- */
-
-int mcfrs_console_setup(struct console *cp, char *arg)
-{
-	int		i, n = CONSOLE_BAUD_RATE;
-
-	if (!cp)
-		return(-1);
-
-	if (!strncmp(cp->name, "ttyS", 4))
-		mcfrs_console_port = cp->index;
-	else if (!strncmp(cp->name, "cua", 3))
-		mcfrs_console_port = cp->index;
-	else
-		return(-1);
-
-	if (arg)
-		n = simple_strtoul(arg,NULL,0);
-	for (i = 0; i < MCFRS_BAUD_TABLE_SIZE; i++)
-		if (mcfrs_baud_table[i] == n)
-			break;
-	if (i < MCFRS_BAUD_TABLE_SIZE) {
-		mcfrs_console_baud = n;
-		mcfrs_console_cbaud = 0;
-		if (i > 15) {
-			mcfrs_console_cbaud |= CBAUDEX;
-			i -= 15;
-		}
-		mcfrs_console_cbaud |= i;
-	}
-	mcfrs_init_console(); /* make sure baud rate changes */
-	return(0);
-}
-
-
-static struct tty_driver *mcfrs_console_device(struct console *c, int *index)
-{
-	*index = c->index;
-	return mcfrs_serial_driver;
-}
-
-
-/*
- *	Output a single character, using UART polled mode.
- *	This is used for console output.
- */
-
-int mcfrs_put_char(char ch)
-{
-	volatile unsigned char	*uartp;
-	unsigned long		flags;
-	int			i;
-
-	uartp = (volatile unsigned char *) (MCF_MBAR +
-		(mcfrs_console_port ? MCFUART_BASE2 : MCFUART_BASE1));
-
-	local_irq_save(flags);
-	for (i = 0; (i < 0x10000); i++) {
-		if (uartp[MCFUART_USR] & MCFUART_USR_TXREADY)
-			break;
-	}
-	if (i < 0x10000) {
-		uartp[MCFUART_UTB] = ch;
-		for (i = 0; (i < 0x10000); i++)
-			if (uartp[MCFUART_USR] & MCFUART_USR_TXEMPTY)
-				break;
-	}
-	if (i >= 0x10000)
-		mcfrs_init_console(); /* try and get it back */
-	local_irq_restore(flags);
-
-	return 1;
-}
-
-
-/*
- * rs_console_write is registered for printk output.
- */
-
-void mcfrs_console_write(struct console *cp, const char *p, unsigned len)
-{
-	if (!mcfrs_console_inited)
-		mcfrs_init_console();
-	while (len-- > 0) {
-		if (*p == '\n')
-			mcfrs_put_char('\r');
-		mcfrs_put_char(*p++);
-	}
-}
-
-/*
- * declare our consoles
- */
-
-struct console mcfrs_console = {
-	.name		= "ttyS",
-	.write		= mcfrs_console_write,
-	.device		= mcfrs_console_device,
-	.setup		= mcfrs_console_setup,
-	.flags		= CON_PRINTBUFFER,
-	.index		= -1,
-};
-
-static int __init mcfrs_console_init(void)
-{
-	register_console(&mcfrs_console);
-	return 0;
-}
-
-console_initcall(mcfrs_console_init);
-
-/****************************************************************************/
diff --git a/drivers/serial/mcfserial.h b/drivers/serial/mcfserial.h
deleted file mode 100644
index 56420e2cb110..000000000000
--- a/drivers/serial/mcfserial.h
+++ /dev/null
@@ -1,74 +0,0 @@
-/*
- * mcfserial.c -- serial driver for ColdFire internal UARTS.
- *
- * Copyright (c) 1999 Greg Ungerer <gerg@snapgear.com>
- * Copyright (c) 2000-2001 Lineo, Inc. <www.lineo.com>
- * Copyright (c) 2002 SnapGear Inc., <www.snapgear.com>
- *
- * Based on code from 68332serial.c which was:
- *
- * Copyright (C) 1995 David S. Miller (davem@caip.rutgers.edu)
- * Copyright (C) 1998 TSHG
- * Copyright (c) 1999 Rt-Control Inc. <jeff@uclinux.org>
- */ 
-#ifndef _MCF_SERIAL_H
-#define _MCF_SERIAL_H
-
-#include <linux/serial.h>
-
-#ifdef __KERNEL__
-
-/*
- *	Define a local serial stats structure.
- */
-
-struct mcf_stats {
-	unsigned int	rx;
-	unsigned int	tx;
-	unsigned int	rxbreak;
-	unsigned int	rxframing;
-	unsigned int	rxparity;
-	unsigned int	rxoverrun;
-};
-
-
-/*
- * This is our internal structure for each serial port's state.
- * Each serial port has one of these structures associated with it.
- */
-
-struct mcf_serial {
-	int			magic;
-	volatile unsigned char	*addr;		/* UART memory address */
-	int			irq;
-	int			flags; 		/* defined in tty.h */
-	int			type; 		/* UART type */
-	struct tty_struct 	*tty;
-	unsigned char		imr;		/* Software imr register */
-	unsigned int		baud;
-	int			sigs;
-	int			custom_divisor;
-	int			x_char;	/* xon/xoff character */
-	int			baud_base;
-	int			close_delay;
-	unsigned short		closing_wait;
-	unsigned short		closing_wait2;
-	unsigned long		event;
-	int			line;
-	int			count;	    /* # of fd on device */
-	int			blocked_open; /* # of blocked opens */
-	unsigned char 		*xmit_buf;
-	int			xmit_head;
-	int			xmit_tail;
-	int			xmit_cnt;
-	struct mcf_stats	stats;
-	struct work_struct	tqueue;
-	struct work_struct	tqueue_hangup;
-	wait_queue_head_t	open_wait;
-	wait_queue_head_t	close_wait;
-
-};
-
-#endif /* __KERNEL__ */
-
-#endif /* _MCF_SERIAL_H */
diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c
index f977c98cfa95..6bdf3362e3b1 100644
--- a/drivers/serial/serial_core.c
+++ b/drivers/serial/serial_core.c
@@ -2051,7 +2051,8 @@ int uart_suspend_port(struct uart_driver *drv, struct uart_port *port)
 					"transmitter\n",
 			       port->dev ? port->dev->bus_id : "",
 			       port->dev ? ": " : "",
-			       drv->dev_name, port->line);
+			       drv->dev_name,
+			       drv->tty_driver->name_base + port->line);
 
 		ops->shutdown(port);
 	}
@@ -2154,12 +2155,11 @@ uart_report_port(struct uart_driver *drv, struct uart_port *port)
 
 	switch (port->iotype) {
 	case UPIO_PORT:
-		snprintf(address, sizeof(address),
-			 "I/O 0x%x", port->iobase);
+		snprintf(address, sizeof(address), "I/O 0x%lx", port->iobase);
 		break;
 	case UPIO_HUB6:
 		snprintf(address, sizeof(address),
-			 "I/O 0x%x offset 0x%x", port->iobase, port->hub6);
+			 "I/O 0x%lx offset 0x%x", port->iobase, port->hub6);
 		break;
 	case UPIO_MEM:
 	case UPIO_MEM32:
@@ -2177,7 +2177,9 @@ uart_report_port(struct uart_driver *drv, struct uart_port *port)
 	printk(KERN_INFO "%s%s%s%d at %s (irq = %d) is a %s\n",
 	       port->dev ? port->dev->bus_id : "",
 	       port->dev ? ": " : "",
-	       drv->dev_name, port->line, address, port->irq, uart_type(port));
+	       drv->dev_name,
+	       drv->tty_driver->name_base + port->line,
+	       address, port->irq, uart_type(port));
 }
 
 static void
diff --git a/drivers/serial/sunhv.c b/drivers/serial/sunhv.c
index e41766d08035..a94a2ab4b571 100644
--- a/drivers/serial/sunhv.c
+++ b/drivers/serial/sunhv.c
@@ -616,7 +616,7 @@ static int __devexit hv_remove(struct of_device *dev)
 	return 0;
 }
 
-static struct of_device_id hv_match[] = {
+static const struct of_device_id hv_match[] = {
 	{
 		.name = "console",
 		.compatible = "qcn",
diff --git a/drivers/serial/sunsab.c b/drivers/serial/sunsab.c
index 29b4458abf74..0355efe115d9 100644
--- a/drivers/serial/sunsab.c
+++ b/drivers/serial/sunsab.c
@@ -1078,7 +1078,7 @@ static int __devexit sab_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id sab_match[] = {
+static const struct of_device_id sab_match[] = {
 	{
 		.name = "se",
 	},
diff --git a/drivers/serial/sunsu.c b/drivers/serial/sunsu.c
index a378464f9292..a4dc79b1d7ab 100644
--- a/drivers/serial/sunsu.c
+++ b/drivers/serial/sunsu.c
@@ -1506,7 +1506,7 @@ static int __devexit su_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id su_match[] = {
+static const struct of_device_id su_match[] = {
 	{
 		.name = "su",
 	},
diff --git a/drivers/serial/sunzilog.c b/drivers/serial/sunzilog.c
index 3cb4c8aee13f..45a299f35617 100644
--- a/drivers/serial/sunzilog.c
+++ b/drivers/serial/sunzilog.c
@@ -1480,7 +1480,7 @@ static int __devexit zs_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id zs_match[] = {
+static const struct of_device_id zs_match[] = {
 	{
 		.name = "zs",
 	},
diff --git a/drivers/usb/host/ehci.h b/drivers/usb/host/ehci.h
index 5799298364fb..b697a13364ec 100644
--- a/drivers/usb/host/ehci.h
+++ b/drivers/usb/host/ehci.h
@@ -210,143 +210,7 @@ timer_action (struct ehci_hcd *ehci, enum ehci_timer_action action)
 
 /*-------------------------------------------------------------------------*/
 
-/* EHCI register interface, corresponds to EHCI Revision 0.95 specification */
-
-/* Section 2.2 Host Controller Capability Registers */
-struct ehci_caps {
-	/* these fields are specified as 8 and 16 bit registers,
-	 * but some hosts can't perform 8 or 16 bit PCI accesses.
-	 */
-	u32		hc_capbase;
-#define HC_LENGTH(p)		(((p)>>00)&0x00ff)	/* bits 7:0 */
-#define HC_VERSION(p)		(((p)>>16)&0xffff)	/* bits 31:16 */
-	u32		hcs_params;     /* HCSPARAMS - offset 0x4 */
-#define HCS_DEBUG_PORT(p)	(((p)>>20)&0xf)	/* bits 23:20, debug port? */
-#define HCS_INDICATOR(p)	((p)&(1 << 16))	/* true: has port indicators */
-#define HCS_N_CC(p)		(((p)>>12)&0xf)	/* bits 15:12, #companion HCs */
-#define HCS_N_PCC(p)		(((p)>>8)&0xf)	/* bits 11:8, ports per CC */
-#define HCS_PORTROUTED(p)	((p)&(1 << 7))	/* true: port routing */
-#define HCS_PPC(p)		((p)&(1 << 4))	/* true: port power control */
-#define HCS_N_PORTS(p)		(((p)>>0)&0xf)	/* bits 3:0, ports on HC */
-
-	u32		hcc_params;      /* HCCPARAMS - offset 0x8 */
-#define HCC_EXT_CAPS(p)		(((p)>>8)&0xff)	/* for pci extended caps */
-#define HCC_ISOC_CACHE(p)       ((p)&(1 << 7))  /* true: can cache isoc frame */
-#define HCC_ISOC_THRES(p)       (((p)>>4)&0x7)  /* bits 6:4, uframes cached */
-#define HCC_CANPARK(p)		((p)&(1 << 2))  /* true: can park on async qh */
-#define HCC_PGM_FRAMELISTLEN(p) ((p)&(1 << 1))  /* true: periodic_size changes*/
-#define HCC_64BIT_ADDR(p)       ((p)&(1))       /* true: can use 64-bit addr */
-	u8		portroute [8];	 /* nibbles for routing - offset 0xC */
-} __attribute__ ((packed));
-
-
-/* Section 2.3 Host Controller Operational Registers */
-struct ehci_regs {
-
-	/* USBCMD: offset 0x00 */
-	u32		command;
-/* 23:16 is r/w intr rate, in microframes; default "8" == 1/msec */
-#define CMD_PARK	(1<<11)		/* enable "park" on async qh */
-#define CMD_PARK_CNT(c)	(((c)>>8)&3)	/* how many transfers to park for */
-#define CMD_LRESET	(1<<7)		/* partial reset (no ports, etc) */
-#define CMD_IAAD	(1<<6)		/* "doorbell" interrupt async advance */
-#define CMD_ASE		(1<<5)		/* async schedule enable */
-#define CMD_PSE		(1<<4)		/* periodic schedule enable */
-/* 3:2 is periodic frame list size */
-#define CMD_RESET	(1<<1)		/* reset HC not bus */
-#define CMD_RUN		(1<<0)		/* start/stop HC */
-
-	/* USBSTS: offset 0x04 */
-	u32		status;
-#define STS_ASS		(1<<15)		/* Async Schedule Status */
-#define STS_PSS		(1<<14)		/* Periodic Schedule Status */
-#define STS_RECL	(1<<13)		/* Reclamation */
-#define STS_HALT	(1<<12)		/* Not running (any reason) */
-/* some bits reserved */
-	/* these STS_* flags are also intr_enable bits (USBINTR) */
-#define STS_IAA		(1<<5)		/* Interrupted on async advance */
-#define STS_FATAL	(1<<4)		/* such as some PCI access errors */
-#define STS_FLR		(1<<3)		/* frame list rolled over */
-#define STS_PCD		(1<<2)		/* port change detect */
-#define STS_ERR		(1<<1)		/* "error" completion (overflow, ...) */
-#define STS_INT		(1<<0)		/* "normal" completion (short, ...) */
-
-	/* USBINTR: offset 0x08 */
-	u32		intr_enable;
-
-	/* FRINDEX: offset 0x0C */
-	u32		frame_index;	/* current microframe number */
-	/* CTRLDSSEGMENT: offset 0x10 */
-	u32		segment;	/* address bits 63:32 if needed */
-	/* PERIODICLISTBASE: offset 0x14 */
-	u32		frame_list;	/* points to periodic list */
-	/* ASYNCLISTADDR: offset 0x18 */
-	u32		async_next;	/* address of next async queue head */
-
-	u32		reserved [9];
-
-	/* CONFIGFLAG: offset 0x40 */
-	u32		configured_flag;
-#define FLAG_CF		(1<<0)		/* true: we'll support "high speed" */
-
-	/* PORTSC: offset 0x44 */
-	u32		port_status [0];	/* up to N_PORTS */
-/* 31:23 reserved */
-#define PORT_WKOC_E	(1<<22)		/* wake on overcurrent (enable) */
-#define PORT_WKDISC_E	(1<<21)		/* wake on disconnect (enable) */
-#define PORT_WKCONN_E	(1<<20)		/* wake on connect (enable) */
-/* 19:16 for port testing */
-#define PORT_LED_OFF	(0<<14)
-#define PORT_LED_AMBER	(1<<14)
-#define PORT_LED_GREEN	(2<<14)
-#define PORT_LED_MASK	(3<<14)
-#define PORT_OWNER	(1<<13)		/* true: companion hc owns this port */
-#define PORT_POWER	(1<<12)		/* true: has power (see PPC) */
-#define PORT_USB11(x) (((x)&(3<<10))==(1<<10))	/* USB 1.1 device */
-/* 11:10 for detecting lowspeed devices (reset vs release ownership) */
-/* 9 reserved */
-#define PORT_RESET	(1<<8)		/* reset port */
-#define PORT_SUSPEND	(1<<7)		/* suspend port */
-#define PORT_RESUME	(1<<6)		/* resume it */
-#define PORT_OCC	(1<<5)		/* over current change */
-#define PORT_OC		(1<<4)		/* over current active */
-#define PORT_PEC	(1<<3)		/* port enable change */
-#define PORT_PE		(1<<2)		/* port enable */
-#define PORT_CSC	(1<<1)		/* connect status change */
-#define PORT_CONNECT	(1<<0)		/* device connected */
-#define PORT_RWC_BITS   (PORT_CSC | PORT_PEC | PORT_OCC)
-} __attribute__ ((packed));
-
-#define USBMODE		0x68		/* USB Device mode */
-#define USBMODE_SDIS	(1<<3)		/* Stream disable */
-#define USBMODE_BE	(1<<2)		/* BE/LE endianness select */
-#define USBMODE_CM_HC	(3<<0)		/* host controller mode */
-#define USBMODE_CM_IDLE	(0<<0)		/* idle state */
-
-/* Appendix C, Debug port ... intended for use with special "debug devices"
- * that can help if there's no serial console.  (nonstandard enumeration.)
- */
-struct ehci_dbg_port {
-	u32	control;
-#define DBGP_OWNER	(1<<30)
-#define DBGP_ENABLED	(1<<28)
-#define DBGP_DONE	(1<<16)
-#define DBGP_INUSE	(1<<10)
-#define DBGP_ERRCODE(x)	(((x)>>7)&0x07)
-#	define DBGP_ERR_BAD	1
-#	define DBGP_ERR_SIGNAL	2
-#define DBGP_ERROR	(1<<6)
-#define DBGP_GO		(1<<5)
-#define DBGP_OUT	(1<<4)
-#define DBGP_LEN(x)	(((x)>>0)&0x0f)
-	u32	pids;
-#define DBGP_PID_GET(x)		(((x)>>16)&0xff)
-#define DBGP_PID_SET(data,tok)	(((data)<<8)|(tok))
-	u32	data03;
-	u32	data47;
-	u32	address;
-#define DBGP_EPADDR(dev,ep)	(((dev)<<8)|(ep))
-} __attribute__ ((packed));
+#include <linux/usb/ehci_def.h>
 
 /*-------------------------------------------------------------------------*/
 
diff --git a/drivers/usb/serial/aircable.c b/drivers/usb/serial/aircable.c
index 79ea98c66fa8..99fb7dc59c45 100644
--- a/drivers/usb/serial/aircable.c
+++ b/drivers/usb/serial/aircable.c
@@ -272,7 +272,7 @@ static void aircable_read(struct work_struct *work)
 	 * 64 bytes, to ensure I do not get throttled.
 	 * Ask USB mailing list for better aproach.
 	 */
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 
 	if (!tty) {
 		schedule_work(&priv->rx_work);
@@ -283,12 +283,13 @@ static void aircable_read(struct work_struct *work)
 	count = min(64, serial_buf_data_avail(priv->rx_buf));
 
 	if (count <= 0)
-		return; /* We have finished sending everything. */
+		goto out; /* We have finished sending everything. */
 
 	tty_prepare_flip_string(tty, &data, count);
 	if (!data) {
-		err("%s- kzalloc(%d) failed.", __func__, count);
-		return;
+		dev_err(&port->dev, "%s- kzalloc(%d) failed.",
+							__func__, count);
+		goto out;
 	}
 
 	serial_buf_get(priv->rx_buf, data, count);
@@ -297,7 +298,8 @@ static void aircable_read(struct work_struct *work)
 
 	if (serial_buf_data_avail(priv->rx_buf))
 		schedule_work(&priv->rx_work);
-
+out:		
+	tty_kref_put(tty);
 	return;
 }
 /* End of private methods */
@@ -495,7 +497,7 @@ static void aircable_read_bulk_callback(struct urb *urb)
 	usb_serial_debug_data(debug, &port->dev, __func__,
 				urb->actual_length, urb->transfer_buffer);
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (tty && urb->actual_length) {
 		if (urb->actual_length <= 2) {
 			/* This is an incomplete package */
@@ -527,6 +529,7 @@ static void aircable_read_bulk_callback(struct urb *urb)
 		}
 		aircable_read(&priv->rx_work);
 	}
+	tty_kref_put(tty);
 
 	/* Schedule the next read _if_ we are still open */
 	if (port->port.count) {
diff --git a/drivers/usb/serial/belkin_sa.c b/drivers/usb/serial/belkin_sa.c
index 2ebe06c3405a..1913bc7c5f0b 100644
--- a/drivers/usb/serial/belkin_sa.c
+++ b/drivers/usb/serial/belkin_sa.c
@@ -322,7 +322,7 @@ static void belkin_sa_read_int_callback(struct urb *urb)
 	 * to look in to this before committing any code.
 	 */
 	if (priv->last_lsr & BELKIN_SA_LSR_ERR) {
-		tty = port->port.tty;
+		tty = tty_port_tty_get(&port->port);
 		/* Overrun Error */
 		if (priv->last_lsr & BELKIN_SA_LSR_OE) {
 		}
@@ -335,6 +335,7 @@ static void belkin_sa_read_int_callback(struct urb *urb)
 		/* Break Indicator */
 		if (priv->last_lsr & BELKIN_SA_LSR_BI) {
 		}
+		tty_kref_put(tty);
 	}
 #endif
 	spin_unlock_irqrestore(&priv->lock, flags);
diff --git a/drivers/usb/serial/console.c b/drivers/usb/serial/console.c
index e980766bb84b..5b20de130e08 100644
--- a/drivers/usb/serial/console.c
+++ b/drivers/usb/serial/console.c
@@ -117,7 +117,7 @@ static int usb_console_setup(struct console *co, char *options)
 	}
 
 	port = serial->port[0];
-	port->port.tty = NULL;
+	tty_port_tty_set(&port->port, NULL);
 
 	info->port = port;
 
@@ -143,7 +143,7 @@ static int usb_console_setup(struct console *co, char *options)
 			}
 			memset(&dummy, 0, sizeof(struct ktermios));
 			tty->termios = termios;
-			port->port.tty = tty;
+			tty_port_tty_set(&port->port, tty);
 		}
 
 		/* only call the device specific open if this
@@ -163,7 +163,7 @@ static int usb_console_setup(struct console *co, char *options)
 			tty_termios_encode_baud_rate(termios, baud, baud);
 			serial->type->set_termios(tty, port, &dummy);
 
-			port->port.tty = NULL;
+			tty_port_tty_set(&port->port, NULL);
 			kfree(termios);
 			kfree(tty);
 		}
@@ -176,7 +176,7 @@ out:
 	return retval;
 free_termios:
 	kfree(termios);
-	port->port.tty = NULL;
+	tty_port_tty_set(&port->port, NULL);
 free_tty:
 	kfree(tty);
 reset_open_count:
diff --git a/drivers/usb/serial/cyberjack.c b/drivers/usb/serial/cyberjack.c
index b4d72351cb96..94ef36c4764b 100644
--- a/drivers/usb/serial/cyberjack.c
+++ b/drivers/usb/serial/cyberjack.c
@@ -384,7 +384,7 @@ static void cyberjack_read_bulk_callback(struct urb *urb)
 		return;
 	}
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (!tty) {
 		dbg("%s - ignoring since device not open\n", __func__);
 		return;
@@ -394,6 +394,7 @@ static void cyberjack_read_bulk_callback(struct urb *urb)
 		tty_insert_flip_string(tty, data, urb->actual_length);
 		tty_flip_buffer_push(tty);
 	}
+	tty_kref_put(tty);
 
 	spin_lock(&priv->lock);
 
diff --git a/drivers/usb/serial/cypress_m8.c b/drivers/usb/serial/cypress_m8.c
index 22837a3f2f89..f3514a91f915 100644
--- a/drivers/usb/serial/cypress_m8.c
+++ b/drivers/usb/serial/cypress_m8.c
@@ -1286,7 +1286,7 @@ static void cypress_read_int_callback(struct urb *urb)
 	}
 	spin_unlock_irqrestore(&priv->lock, flags);
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (!tty) {
 		dbg("%s - bad tty pointer - exiting", __func__);
 		return;
@@ -1362,7 +1362,7 @@ static void cypress_read_int_callback(struct urb *urb)
 					data[i]);
 			tty_insert_flip_char(tty, data[i], tty_flag);
 		}
-		tty_flip_buffer_push(port->port.tty);
+		tty_flip_buffer_push(tty);
 	}
 
 	spin_lock_irqsave(&priv->lock, flags);
@@ -1371,6 +1371,7 @@ static void cypress_read_int_callback(struct urb *urb)
 	spin_unlock_irqrestore(&priv->lock, flags);
 
 continue_read:
+	tty_kref_put(tty);
 
 	/* Continue trying to always read... unless the port has closed. */
 
diff --git a/drivers/usb/serial/digi_acceleport.c b/drivers/usb/serial/digi_acceleport.c
index 240aad1acaab..5756ac6d6c92 100644
--- a/drivers/usb/serial/digi_acceleport.c
+++ b/drivers/usb/serial/digi_acceleport.c
@@ -604,7 +604,9 @@ static void digi_wakeup_write_lock(struct work_struct *work)
 
 static void digi_wakeup_write(struct usb_serial_port *port)
 {
-	tty_wakeup(port->port.tty);
+	struct tty_struct *tty = tty_port_tty_get(&port->port);
+	tty_wakeup(tty);
+	tty_kref_put(tty);
 }
 
 
@@ -1668,7 +1670,7 @@ static int digi_read_inb_callback(struct urb *urb)
 {
 
 	struct usb_serial_port *port = urb->context;
-	struct tty_struct *tty = port->port.tty;
+	struct tty_struct *tty;
 	struct digi_port *priv = usb_get_serial_port_data(port);
 	int opcode = ((unsigned char *)urb->transfer_buffer)[0];
 	int len = ((unsigned char *)urb->transfer_buffer)[1];
@@ -1692,6 +1694,7 @@ static int digi_read_inb_callback(struct urb *urb)
 		return -1;
 	}
 
+	tty = tty_port_tty_get(&port->port);
 	spin_lock(&priv->dp_port_lock);
 
 	/* check for throttle; if set, do not resubmit read urb */
@@ -1735,6 +1738,7 @@ static int digi_read_inb_callback(struct urb *urb)
 		}
 	}
 	spin_unlock(&priv->dp_port_lock);
+	tty_kref_put(tty);
 
 	if (opcode == DIGI_CMD_RECEIVE_DISABLE)
 		dbg("%s: got RECEIVE_DISABLE", __func__);
@@ -1760,6 +1764,7 @@ static int digi_read_oob_callback(struct urb *urb)
 
 	struct usb_serial_port *port = urb->context;
 	struct usb_serial *serial = port->serial;
+	struct tty_struct *tty;
 	struct digi_port *priv = usb_get_serial_port_data(port);
 	int opcode, line, status, val;
 	int i;
@@ -1787,10 +1792,11 @@ static int digi_read_oob_callback(struct urb *urb)
 		if (priv == NULL)
 			return -1;
 
+		tty = tty_port_tty_get(&port->port);
 		rts = 0;
 		if (port->port.count)
-			rts = port->port.tty->termios->c_cflag & CRTSCTS;
-
+			rts = tty->termios->c_cflag & CRTSCTS;
+		
 		if (opcode == DIGI_CMD_READ_INPUT_SIGNALS) {
 			spin_lock(&priv->dp_port_lock);
 			/* convert from digi flags to termiox flags */
@@ -1798,14 +1804,14 @@ static int digi_read_oob_callback(struct urb *urb)
 				priv->dp_modem_signals |= TIOCM_CTS;
 				/* port must be open to use tty struct */
 				if (rts) {
-					port->port.tty->hw_stopped = 0;
+					tty->hw_stopped = 0;
 					digi_wakeup_write(port);
 				}
 			} else {
 				priv->dp_modem_signals &= ~TIOCM_CTS;
 				/* port must be open to use tty struct */
 				if (rts)
-					port->port.tty->hw_stopped = 1;
+					tty->hw_stopped = 1;
 			}
 			if (val & DIGI_READ_INPUT_SIGNALS_DSR)
 				priv->dp_modem_signals |= TIOCM_DSR;
@@ -1830,6 +1836,7 @@ static int digi_read_oob_callback(struct urb *urb)
 		} else if (opcode == DIGI_CMD_IFLUSH_FIFO) {
 			wake_up_interruptible(&priv->dp_flush_wait);
 		}
+		tty_kref_put(tty);
 	}
 	return 0;
 
diff --git a/drivers/usb/serial/empeg.c b/drivers/usb/serial/empeg.c
index a6ab5b58d9ca..1072e847280f 100644
--- a/drivers/usb/serial/empeg.c
+++ b/drivers/usb/serial/empeg.c
@@ -33,9 +33,8 @@
  *	Moved MOD_DEC_USE_COUNT to end of empeg_close().
  *
  * (12/03/2000) gb
- *	Added port->port.tty->ldisc.set_termios(port->port.tty, NULL) to
- *	empeg_open(). This notifies the tty driver that the termios have
- *	changed.
+ *	Added tty->ldisc.set_termios(port, tty, NULL) to empeg_open().
+ *	This notifies the tty driver that the termios have changed.
  *
  * (11/13/2000) gb
  *	Moved tty->low_latency = 1 from empeg_read_bulk_callback() to
@@ -354,7 +353,7 @@ static void empeg_read_bulk_callback(struct urb *urb)
 
 	usb_serial_debug_data(debug, &port->dev, __func__,
 						urb->actual_length, data);
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 
 	if (urb->actual_length) {
 		tty_buffer_request_room(tty, urb->actual_length);
@@ -362,6 +361,7 @@ static void empeg_read_bulk_callback(struct urb *urb)
 		tty_flip_buffer_push(tty);
 		bytes_in += urb->actual_length;
 	}
+	tty_kref_put(tty);
 
 	/* Continue trying to always read  */
 	usb_fill_bulk_urb(
diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c
index 3dc93b542b30..c2ac129557aa 100644
--- a/drivers/usb/serial/ftdi_sio.c
+++ b/drivers/usb/serial/ftdi_sio.c
@@ -860,7 +860,7 @@ static int update_mctrl(struct usb_serial_port *port, unsigned int set,
 
 	kfree(buf);
 	if (rv < 0) {
-		err("%s Error from MODEM_CTRL urb: DTR %s, RTS %s",
+		dbg("%s Error from MODEM_CTRL urb: DTR %s, RTS %s",
 				__func__,
 				(set & TIOCM_DTR) ? "HIGH" :
 				(clear & TIOCM_DTR) ? "LOW" : "unchanged",
@@ -1808,7 +1808,7 @@ static void ftdi_read_bulk_callback(struct urb *urb)
 	if (port->port.count <= 0)
 		return;
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (!tty) {
 		dbg("%s - bad tty pointer - exiting", __func__);
 		return;
@@ -1817,7 +1817,7 @@ static void ftdi_read_bulk_callback(struct urb *urb)
 	priv = usb_get_serial_port_data(port);
 	if (!priv) {
 		dbg("%s - bad port private data pointer - exiting", __func__);
-		return;
+		goto out;
 	}
 
 	if (urb != port->read_urb)
@@ -1827,7 +1827,7 @@ static void ftdi_read_bulk_callback(struct urb *urb)
 		/* This will happen at close every time so it is a dbg not an
 		   err */
 		dbg("(this is ok on close) nonzero read bulk status received: %d", status);
-		return;
+		goto out;
 	}
 
 	/* count data bytes, but not status bytes */
@@ -1838,7 +1838,8 @@ static void ftdi_read_bulk_callback(struct urb *urb)
 	spin_unlock_irqrestore(&priv->rx_lock, flags);
 
 	ftdi_process_read(&priv->rx_work.work);
-
+out:
+	tty_kref_put(tty);
 } /* ftdi_read_bulk_callback */
 
 
@@ -1863,7 +1864,7 @@ static void ftdi_process_read(struct work_struct *work)
 	if (port->port.count <= 0)
 		return;
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (!tty) {
 		dbg("%s - bad tty pointer - exiting", __func__);
 		return;
@@ -1872,13 +1873,13 @@ static void ftdi_process_read(struct work_struct *work)
 	priv = usb_get_serial_port_data(port);
 	if (!priv) {
 		dbg("%s - bad port private data pointer - exiting", __func__);
-		return;
+		goto out;
 	}
 
 	urb = port->read_urb;
 	if (!urb) {
 		dbg("%s - bad read_urb pointer - exiting", __func__);
-		return;
+		goto out;
 	}
 
 	data = urb->transfer_buffer;
@@ -2020,7 +2021,7 @@ static void ftdi_process_read(struct work_struct *work)
 			schedule_delayed_work(&priv->rx_work, 1);
 		else
 			dbg("%s - port is closed", __func__);
-		return;
+		goto out;
 	}
 
 	/* urb is completely processed */
@@ -2041,6 +2042,8 @@ static void ftdi_process_read(struct work_struct *work)
 			err("%s - failed resubmitting read urb, error %d",
 							__func__, result);
 	}
+out:
+	tty_kref_put(tty);
 } /* ftdi_process_read */
 
 
@@ -2256,7 +2259,7 @@ static int ftdi_tiocmget(struct tty_struct *tty, struct file *file)
 			   0, 0,
 			   buf, 1, WDR_TIMEOUT);
 		if (ret < 0) {
-			err("%s Could not get modem status of device - err: %d", __func__,
+			dbg("%s Could not get modem status of device - err: %d", __func__,
 			    ret);
 			return ret;
 		}
@@ -2275,7 +2278,7 @@ static int ftdi_tiocmget(struct tty_struct *tty, struct file *file)
 				   0, priv->interface,
 				   buf, 2, WDR_TIMEOUT);
 		if (ret < 0) {
-			err("%s Could not get modem status of device - err: %d", __func__,
+			dbg("%s Could not get modem status of device - err: %d", __func__,
 			    ret);
 			return ret;
 		}
diff --git a/drivers/usb/serial/garmin_gps.c b/drivers/usb/serial/garmin_gps.c
index d95382088075..2ad0569bcf19 100644
--- a/drivers/usb/serial/garmin_gps.c
+++ b/drivers/usb/serial/garmin_gps.c
@@ -276,7 +276,7 @@ static inline int isAbortTrfCmnd(const unsigned char *buf)
 static void send_to_tty(struct usb_serial_port *port,
 			char *data, unsigned int actual_length)
 {
-	struct tty_struct *tty = port->port.tty;
+	struct tty_struct *tty = tty_port_tty_get(&port->port);
 
 	if (tty && actual_length) {
 
@@ -287,6 +287,7 @@ static void send_to_tty(struct usb_serial_port *port,
 		tty_insert_flip_string(tty, data, actual_length);
 		tty_flip_buffer_push(tty);
 	}
+	tty_kref_put(tty);
 }
 
 
diff --git a/drivers/usb/serial/generic.c b/drivers/usb/serial/generic.c
index fe84c88ec20c..814909f1ee63 100644
--- a/drivers/usb/serial/generic.c
+++ b/drivers/usb/serial/generic.c
@@ -330,7 +330,7 @@ static void resubmit_read_urb(struct usb_serial_port *port, gfp_t mem_flags)
 static void flush_and_resubmit_read_urb(struct usb_serial_port *port)
 {
 	struct urb *urb = port->read_urb;
-	struct tty_struct *tty = port->port.tty;
+	struct tty_struct *tty = tty_port_tty_get(&port->port);
 	int room;
 
 	/* Push data to tty */
@@ -341,6 +341,7 @@ static void flush_and_resubmit_read_urb(struct usb_serial_port *port)
 			tty_flip_buffer_push(tty);
 		}
 	}
+	tty_kref_put(tty);
 
 	resubmit_read_urb(port, GFP_ATOMIC);
 }
diff --git a/drivers/usb/serial/io_edgeport.c b/drivers/usb/serial/io_edgeport.c
index bfa508ddb0fe..611f97fd62f1 100644
--- a/drivers/usb/serial/io_edgeport.c
+++ b/drivers/usb/serial/io_edgeport.c
@@ -600,6 +600,7 @@ static void edge_interrupt_callback(struct urb *urb)
 	struct edgeport_serial	*edge_serial = urb->context;
 	struct edgeport_port *edge_port;
 	struct usb_serial_port *port;
+	struct tty_struct *tty;
 	unsigned char *data = urb->transfer_buffer;
 	int length = urb->actual_length;
 	int bytes_avail;
@@ -675,9 +676,12 @@ static void edge_interrupt_callback(struct urb *urb)
 
 					/* tell the tty driver that something
 					   has changed */
-					if (edge_port->port->port.tty)
-						tty_wakeup(edge_port->port->port.tty);
-
+					tty = tty_port_tty_get(
+						&edge_port->port->port);
+					if (tty) {
+						tty_wakeup(tty);
+						tty_kref_put(tty);
+					}
 					/* Since we have more credit, check
 					   if more data can be sent */
 					send_more_port_data(edge_serial,
@@ -778,13 +782,14 @@ static void edge_bulk_out_data_callback(struct urb *urb)
 		    __func__, status);
 	}
 
-	tty = edge_port->port->port.tty;
+	tty = tty_port_tty_get(&edge_port->port->port);
 
 	if (tty && edge_port->open) {
 		/* let the tty driver wakeup if it has a special
 		   write_wakeup function */
 		tty_wakeup(tty);
 	}
+	tty_kref_put(tty);
 
 	/* Release the Write URB */
 	edge_port->write_in_progress = false;
@@ -826,11 +831,12 @@ static void edge_bulk_out_cmd_callback(struct urb *urb)
 	}
 
 	/* Get pointer to tty */
-	tty = edge_port->port->port.tty;
+	tty = tty_port_tty_get(&edge_port->port->port);
 
 	/* tell the tty driver that something has changed */
 	if (tty && edge_port->open)
 		tty_wakeup(tty);
+	tty_kref_put(tty);
 
 	/* we have completed the command */
 	edge_port->commandPending = false;
@@ -1932,11 +1938,13 @@ static void process_rcvd_data(struct edgeport_serial *edge_serial,
 							edge_serial->rxPort];
 				edge_port = usb_get_serial_port_data(port);
 				if (edge_port->open) {
-					tty = edge_port->port->port.tty;
+					tty = tty_port_tty_get(
+						&edge_port->port->port);
 					if (tty) {
 						dbg("%s - Sending %d bytes to TTY for port %d",
 							__func__, rxLen, edge_serial->rxPort);
 						edge_tty_recv(&edge_serial->serial->dev->dev, tty, buffer, rxLen);
+						tty_kref_put(tty);
 					}
 					edge_port->icount.rx += rxLen;
 				}
@@ -1971,6 +1979,7 @@ static void process_rcvd_status(struct edgeport_serial *edge_serial,
 {
 	struct usb_serial_port *port;
 	struct edgeport_port *edge_port;
+	struct tty_struct *tty;
 	__u8 code = edge_serial->rxStatusCode;
 
 	/* switch the port pointer to the one being currently talked about */
@@ -2020,10 +2029,12 @@ static void process_rcvd_status(struct edgeport_serial *edge_serial,
 
 		/* send the current line settings to the port so we are
 		   in sync with any further termios calls */
-		/* FIXME: locking on tty */
-		if (edge_port->port->port.tty)
-			change_port_settings(edge_port->port->port.tty,
-				edge_port, edge_port->port->port.tty->termios);
+		tty = tty_port_tty_get(&edge_port->port->port);
+		if (tty) {
+			change_port_settings(tty,
+				edge_port, tty->termios);
+			tty_kref_put(tty);
+		}
 
 		/* we have completed the open */
 		edge_port->openPending = false;
@@ -2163,10 +2174,14 @@ static void handle_new_lsr(struct edgeport_port *edge_port, __u8 lsrData,
 	}
 
 	/* Place LSR data byte into Rx buffer */
-	if (lsrData && edge_port->port->port.tty)
-		edge_tty_recv(&edge_port->port->dev,
-					edge_port->port->port.tty, &data, 1);
-
+	if (lsrData) {
+		struct tty_struct *tty =
+				tty_port_tty_get(&edge_port->port->port);
+		if (tty) {
+			edge_tty_recv(&edge_port->port->dev, tty, &data, 1);
+			tty_kref_put(tty);
+		}
+	}
 	/* update input line counters */
 	icount = &edge_port->icount;
 	if (newLsr & LSR_BREAK)
diff --git a/drivers/usb/serial/io_ti.c b/drivers/usb/serial/io_ti.c
index cb4c54316cf5..541dd8e6e7a2 100644
--- a/drivers/usb/serial/io_ti.c
+++ b/drivers/usb/serial/io_ti.c
@@ -572,7 +572,7 @@ static void chase_port(struct edgeport_port *port, unsigned long timeout,
 								int flush)
 {
 	int baud_rate;
-	struct tty_struct *tty = port->port->port.tty;
+	struct tty_struct *tty = tty_port_tty_get(&port->port->port);
 	wait_queue_t wait;
 	unsigned long flags;
 
@@ -599,6 +599,7 @@ static void chase_port(struct edgeport_port *port, unsigned long timeout,
 	if (flush)
 		edge_buf_clear(port->ep_out_buf);
 	spin_unlock_irqrestore(&port->ep_lock, flags);
+	tty_kref_put(tty);
 
 	/* wait for data to drain from the device */
 	timeout += jiffies;
@@ -1554,7 +1555,7 @@ static void handle_new_msr(struct edgeport_port *edge_port, __u8 msr)
 	/* Save the new modem status */
 	edge_port->shadow_msr = msr & 0xf0;
 
-	tty = edge_port->port->port.tty;
+	tty = tty_port_tty_get(&edge_port->port->port);
 	/* handle CTS flow control */
 	if (tty && C_CRTSCTS(tty)) {
 		if (msr & EDGEPORT_MSR_CTS) {
@@ -1564,6 +1565,7 @@ static void handle_new_msr(struct edgeport_port *edge_port, __u8 msr)
 			tty->hw_stopped = 1;
 		}
 	}
+	tty_kref_put(tty);
 
 	return;
 }
@@ -1574,6 +1576,7 @@ static void handle_new_lsr(struct edgeport_port *edge_port, int lsr_data,
 	struct async_icount *icount;
 	__u8 new_lsr = (__u8)(lsr & (__u8)(LSR_OVER_ERR | LSR_PAR_ERR |
 						LSR_FRM_ERR | LSR_BREAK));
+	struct tty_struct *tty;
 
 	dbg("%s - %02x", __func__, new_lsr);
 
@@ -1587,8 +1590,13 @@ static void handle_new_lsr(struct edgeport_port *edge_port, int lsr_data,
 		new_lsr &= (__u8)(LSR_OVER_ERR | LSR_BREAK);
 
 	/* Place LSR data byte into Rx buffer */
-	if (lsr_data && edge_port->port->port.tty)
-		edge_tty_recv(&edge_port->port->dev, edge_port->port->port.tty, &data, 1);
+	if (lsr_data) {
+		tty = tty_port_tty_get(&edge_port->port->port);
+		if (tty) {
+			edge_tty_recv(&edge_port->port->dev, tty, &data, 1);
+			tty_kref_put(tty);
+		}
+	}
 
 	/* update input line counters */
 	icount = &edge_port->icount;
@@ -1749,7 +1757,7 @@ static void edge_bulk_in_callback(struct urb *urb)
 		++data;
 	}
 
-	tty = edge_port->port->port.tty;
+	tty = tty_port_tty_get(&edge_port->port->port);
 	if (tty && urb->actual_length) {
 		usb_serial_debug_data(debug, &edge_port->port->dev,
 					__func__, urb->actual_length, data);
@@ -1761,6 +1769,7 @@ static void edge_bulk_in_callback(struct urb *urb)
 							urb->actual_length);
 		edge_port->icount.rx += urb->actual_length;
 	}
+	tty_kref_put(tty);
 
 exit:
 	/* continue read unless stopped */
@@ -1796,6 +1805,7 @@ static void edge_bulk_out_callback(struct urb *urb)
 	struct usb_serial_port *port = urb->context;
 	struct edgeport_port *edge_port = usb_get_serial_port_data(port);
 	int status = urb->status;
+	struct tty_struct *tty;
 
 	dbg("%s - port %d", __func__, port->number);
 
@@ -1818,7 +1828,9 @@ static void edge_bulk_out_callback(struct urb *urb)
 	}
 
 	/* send any buffered data */
-	edge_send(port->port.tty);
+	tty = tty_port_tty_get(&port->port);
+	edge_send(tty);
+	tty_kref_put(tty);
 }
 
 static int edge_open(struct tty_struct *tty,
@@ -1876,7 +1888,7 @@ static int edge_open(struct tty_struct *tty,
 
 	/* set up the port settings */
 	if (tty)
-		edge_set_termios(tty, port, port->port.tty->termios);
+		edge_set_termios(tty, port, tty->termios);
 
 	/* open up the port */
 
diff --git a/drivers/usb/serial/ipaq.c b/drivers/usb/serial/ipaq.c
index cd9a2e138c8b..2affa9c118b2 100644
--- a/drivers/usb/serial/ipaq.c
+++ b/drivers/usb/serial/ipaq.c
@@ -764,13 +764,14 @@ static void ipaq_read_bulk_callback(struct urb *urb)
 	usb_serial_debug_data(debug, &port->dev, __func__,
 						urb->actual_length, data);
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (tty && urb->actual_length) {
 		tty_buffer_request_room(tty, urb->actual_length);
 		tty_insert_flip_string(tty, data, urb->actual_length);
 		tty_flip_buffer_push(tty);
 		bytes_in += urb->actual_length;
 	}
+	tty_kref_put(tty);
 
 	/* Continue trying to always read  */
 	usb_fill_bulk_urb(port->read_urb, port->serial->dev,
diff --git a/drivers/usb/serial/ipw.c b/drivers/usb/serial/ipw.c
index a842025b9b57..480cac27d646 100644
--- a/drivers/usb/serial/ipw.c
+++ b/drivers/usb/serial/ipw.c
@@ -170,12 +170,13 @@ static void ipw_read_bulk_callback(struct urb *urb)
 	usb_serial_debug_data(debug, &port->dev, __func__,
 					urb->actual_length, data);
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (tty && urb->actual_length) {
 		tty_buffer_request_room(tty, urb->actual_length);
 		tty_insert_flip_string(tty, data, urb->actual_length);
 		tty_flip_buffer_push(tty);
 	}
+	tty_kref_put(tty);
 
 	/* Continue trying to always read  */
 	usb_fill_bulk_urb(port->read_urb, port->serial->dev,
diff --git a/drivers/usb/serial/ir-usb.c b/drivers/usb/serial/ir-usb.c
index e59155c6607d..45d4043e04ab 100644
--- a/drivers/usb/serial/ir-usb.c
+++ b/drivers/usb/serial/ir-usb.c
@@ -465,11 +465,12 @@ static void ir_read_bulk_callback(struct urb *urb)
 			ir_baud = *data & 0x0f;
 		usb_serial_debug_data(debug, &port->dev, __func__,
 						urb->actual_length, data);
- 		tty = port->port.tty;
+		tty = tty_port_tty_get(&port->port);
 		if (tty_buffer_request_room(tty, urb->actual_length - 1)) {
 			tty_insert_flip_string(tty, data+1, urb->actual_length - 1);
 			tty_flip_buffer_push(tty);
 		}
+		tty_kref_put(tty);
 
 		/*
 		 * No break here.
diff --git a/drivers/usb/serial/iuu_phoenix.c b/drivers/usb/serial/iuu_phoenix.c
index ddff37fa6339..53710aa7eadd 100644
--- a/drivers/usb/serial/iuu_phoenix.c
+++ b/drivers/usb/serial/iuu_phoenix.c
@@ -629,13 +629,14 @@ static void read_buf_callback(struct urb *urb)
 	}
 
 	dbg("%s - %i chars to write", __func__, urb->actual_length);
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (data == NULL)
 		dbg("%s - data is NULL !!!", __func__);
 	if (tty && urb->actual_length && data) {
 		tty_insert_flip_string(tty, data, urb->actual_length);
 		tty_flip_buffer_push(tty);
 	}
+	tty_kref_put(tty);
 	iuu_led_activity_on(urb);
 }
 
diff --git a/drivers/usb/serial/keyspan.c b/drivers/usb/serial/keyspan.c
index 704716f6f6d3..15447af48691 100644
--- a/drivers/usb/serial/keyspan.c
+++ b/drivers/usb/serial/keyspan.c
@@ -430,7 +430,7 @@ static void	usa26_indat_callback(struct urb *urb)
 	}
 
 	port =  urb->context;
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (tty && urb->actual_length) {
 		/* 0x80 bit is error flag */
 		if ((data[0] & 0x80) == 0) {
@@ -459,6 +459,7 @@ static void	usa26_indat_callback(struct urb *urb)
 		}
 		tty_flip_buffer_push(tty);
 	}
+	tty_kref_put(tty);
 
 	/* Resubmit urb so we continue receiving */
 	urb->dev = port->serial->dev;
@@ -513,6 +514,7 @@ static void	usa26_instat_callback(struct urb *urb)
 	struct usb_serial			*serial;
 	struct usb_serial_port			*port;
 	struct keyspan_port_private	 	*p_priv;
+	struct tty_struct			*tty;
 	int old_dcd_state, err;
 	int status = urb->status;
 
@@ -553,12 +555,11 @@ static void	usa26_instat_callback(struct urb *urb)
 	p_priv->dcd_state = ((msg->gpia_dcd) ? 1 : 0);
 	p_priv->ri_state = ((msg->ri) ? 1 : 0);
 
-	if (port->port.tty && !C_CLOCAL(port->port.tty)
-	    && old_dcd_state != p_priv->dcd_state) {
-		if (old_dcd_state)
-			tty_hangup(port->port.tty);
-		/*  else */
-		/*	wake_up_interruptible(&p_priv->open_wait); */
+	if (old_dcd_state != p_priv->dcd_state) {
+		tty = tty_port_tty_get(&port->port);
+		if (tty && !C_CLOCAL(tty))
+			tty_hangup(tty);
+		tty_kref_put(tty);
 	}
 
 	/* Resubmit urb so we continue receiving */
@@ -604,11 +605,12 @@ static void usa28_indat_callback(struct urb *urb)
 		p_priv = usb_get_serial_port_data(port);
 		data = urb->transfer_buffer;
 
-		tty = port->port.tty;
-		if (urb->actual_length) {
+		tty =tty_port_tty_get(&port->port);
+		if (tty && urb->actual_length) {
 			tty_insert_flip_string(tty, data, urb->actual_length);
 			tty_flip_buffer_push(tty);
 		}
+		tty_kref_put(tty);
 
 		/* Resubmit urb so we continue receiving */
 		urb->dev = port->serial->dev;
@@ -652,6 +654,7 @@ static void	usa28_instat_callback(struct urb *urb)
 	struct usb_serial			*serial;
 	struct usb_serial_port			*port;
 	struct keyspan_port_private	 	*p_priv;
+	struct tty_struct			*tty;
 	int old_dcd_state;
 	int status = urb->status;
 
@@ -689,12 +692,11 @@ static void	usa28_instat_callback(struct urb *urb)
 	p_priv->dcd_state = ((msg->dcd) ? 1 : 0);
 	p_priv->ri_state = ((msg->ri) ? 1 : 0);
 
-	if (port->port.tty && !C_CLOCAL(port->port.tty)
-	    && old_dcd_state != p_priv->dcd_state) {
-		if (old_dcd_state)
-			tty_hangup(port->port.tty);
-		/*  else */
-		/*	wake_up_interruptible(&p_priv->open_wait); */
+	if( old_dcd_state != p_priv->dcd_state && old_dcd_state) {
+		tty = tty_port_tty_get(&port->port);
+		if (tty && !C_CLOCAL(tty)) 
+			tty_hangup(tty);
+		tty_kref_put(tty);
 	}
 
 		/* Resubmit urb so we continue receiving */
@@ -785,12 +787,11 @@ static void	usa49_instat_callback(struct urb *urb)
 	p_priv->dcd_state = ((msg->dcd) ? 1 : 0);
 	p_priv->ri_state = ((msg->ri) ? 1 : 0);
 
-	if (port->port.tty && !C_CLOCAL(port->port.tty)
-	    && old_dcd_state != p_priv->dcd_state) {
-		if (old_dcd_state)
-			tty_hangup(port->port.tty);
-		/*  else */
-		/*	wake_up_interruptible(&p_priv->open_wait); */
+	if (old_dcd_state != p_priv->dcd_state && old_dcd_state) {
+		struct tty_struct *tty = tty_port_tty_get(&port->port);
+		if (tty && !C_CLOCAL(tty))
+			tty_hangup(tty);
+		tty_kref_put(tty);
 	}
 
 	/* Resubmit urb so we continue receiving */
@@ -827,7 +828,7 @@ static void	usa49_indat_callback(struct urb *urb)
 	}
 
 	port =  urb->context;
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (tty && urb->actual_length) {
 		/* 0x80 bit is error flag */
 		if ((data[0] & 0x80) == 0) {
@@ -850,6 +851,7 @@ static void	usa49_indat_callback(struct urb *urb)
 		}
 		tty_flip_buffer_push(tty);
 	}
+	tty_kref_put(tty);
 
 	/* Resubmit urb so we continue receiving */
 	urb->dev = port->serial->dev;
@@ -893,7 +895,7 @@ static void usa49wg_indat_callback(struct urb *urb)
 				return;
 			}
 			port = serial->port[data[i++]];
-			tty = port->port.tty;
+			tty = tty_port_tty_get(&port->port);
 			len = data[i++];
 
 			/* 0x80 bit is error flag */
@@ -927,6 +929,7 @@ static void usa49wg_indat_callback(struct urb *urb)
 			}
 			if (port->port.count)
 				tty_flip_buffer_push(tty);
+			tty_kref_put(tty);
 		}
 	}
 
@@ -967,8 +970,8 @@ static void usa90_indat_callback(struct urb *urb)
 	port =  urb->context;
 	p_priv = usb_get_serial_port_data(port);
 
-	tty = port->port.tty;
 	if (urb->actual_length) {
+		tty = tty_port_tty_get(&port->port);
 		/* if current mode is DMA, looks like usa28 format
 		   otherwise looks like usa26 data format */
 
@@ -1004,6 +1007,7 @@ static void usa90_indat_callback(struct urb *urb)
 			}
 		}
 		tty_flip_buffer_push(tty);
+		tty_kref_put(tty);
 	}
 
 	/* Resubmit urb so we continue receiving */
@@ -1025,6 +1029,7 @@ static void	usa90_instat_callback(struct urb *urb)
 	struct usb_serial			*serial;
 	struct usb_serial_port			*port;
 	struct keyspan_port_private	 	*p_priv;
+	struct tty_struct			*tty;
 	int old_dcd_state, err;
 	int status = urb->status;
 
@@ -1053,12 +1058,11 @@ static void	usa90_instat_callback(struct urb *urb)
 	p_priv->dcd_state = ((msg->dcd) ? 1 : 0);
 	p_priv->ri_state = ((msg->ri) ? 1 : 0);
 
-	if (port->port.tty && !C_CLOCAL(port->port.tty)
-	    && old_dcd_state != p_priv->dcd_state) {
-		if (old_dcd_state)
-			tty_hangup(port->port.tty);
-		/*  else */
-		/*	wake_up_interruptible(&p_priv->open_wait); */
+	if (old_dcd_state != p_priv->dcd_state && old_dcd_state) {
+		tty = tty_port_tty_get(&port->port);
+		if (tty && !C_CLOCAL(tty))
+			tty_hangup(tty);
+		tty_kref_put(tty);
 	}
 
 	/* Resubmit urb so we continue receiving */
@@ -1130,12 +1134,11 @@ static void	usa67_instat_callback(struct urb *urb)
 	p_priv->cts_state = ((msg->hskia_cts) ? 1 : 0);
 	p_priv->dcd_state = ((msg->gpia_dcd) ? 1 : 0);
 
-	if (port->port.tty && !C_CLOCAL(port->port.tty)
-	    && old_dcd_state != p_priv->dcd_state) {
-		if (old_dcd_state)
-			tty_hangup(port->port.tty);
-		/*  else */
-		/*	wake_up_interruptible(&p_priv->open_wait); */
+	if (old_dcd_state != p_priv->dcd_state && old_dcd_state) {
+		struct tty_struct *tty = tty_port_tty_get(&port->port);
+		if (tty && !C_CLOCAL(tty))
+			tty_hangup(tty);
+		tty_kref_put(tty);
 	}
 
 	/* Resubmit urb so we continue receiving */
@@ -1332,7 +1335,7 @@ static void keyspan_close(struct tty_struct *tty,
 			stop_urb(p_priv->out_urbs[i]);
 		}
 	}
-	port->port.tty = NULL;
+	tty_port_tty_set(&port->port, NULL);
 }
 
 /* download the firmware to a pre-renumeration device */
diff --git a/drivers/usb/serial/keyspan_pda.c b/drivers/usb/serial/keyspan_pda.c
index 040040a267d9..99e9a14c5bf6 100644
--- a/drivers/usb/serial/keyspan_pda.c
+++ b/drivers/usb/serial/keyspan_pda.c
@@ -172,8 +172,9 @@ static void keyspan_pda_wakeup_write(struct work_struct *work)
 	struct keyspan_pda_private *priv =
 		container_of(work, struct keyspan_pda_private, wakeup_work);
 	struct usb_serial_port *port = priv->port;
-
-	tty_wakeup(port->port.tty);
+	struct tty_struct *tty = tty_port_tty_get(&port->port);
+	tty_wakeup(tty);
+	tty_kref_put(tty);
 }
 
 static void keyspan_pda_request_unthrottle(struct work_struct *work)
@@ -205,7 +206,7 @@ static void keyspan_pda_request_unthrottle(struct work_struct *work)
 static void keyspan_pda_rx_interrupt(struct urb *urb)
 {
 	struct usb_serial_port *port = urb->context;
-	struct tty_struct *tty = port->port.tty;
+	struct tty_struct *tty = tty_port_tty_get(&port->port);
 	unsigned char *data = urb->transfer_buffer;
 	int retval;
 	int status = urb->status;
@@ -222,7 +223,7 @@ static void keyspan_pda_rx_interrupt(struct urb *urb)
 		/* this urb is terminated, clean up */
 		dbg("%s - urb shutting down with status: %d",
 		    __func__, status);
-		return;
+		goto out;
 	default:
 		dbg("%s - nonzero urb status received: %d",
 		    __func__, status);
@@ -261,8 +262,11 @@ static void keyspan_pda_rx_interrupt(struct urb *urb)
 exit:
 	retval = usb_submit_urb(urb, GFP_ATOMIC);
 	if (retval)
-		err("%s - usb_submit_urb failed with result %d",
-		     __func__, retval);
+		dev_err(&port->dev,
+			"%s - usb_submit_urb failed with result %d",
+			__func__, retval);
+out:
+	tty_kref_put(tty);		     
 }
 
 
diff --git a/drivers/usb/serial/kl5kusb105.c b/drivers/usb/serial/kl5kusb105.c
index b84dddc71124..ff3a07f5102f 100644
--- a/drivers/usb/serial/kl5kusb105.c
+++ b/drivers/usb/serial/kl5kusb105.c
@@ -658,7 +658,7 @@ static void klsi_105_read_bulk_callback(struct urb *urb)
 	} else {
 		int bytes_sent = ((__u8 *) data)[0] +
 				 ((unsigned int) ((__u8 *) data)[1] << 8);
-		tty = port->port.tty;
+		tty = tty_port_tty_get(&port->port);
 		/* we should immediately resubmit the URB, before attempting
 		 * to pass the data on to the tty layer. But that needs locking
 		 * against re-entry an then mixed-up data because of
@@ -679,6 +679,7 @@ static void klsi_105_read_bulk_callback(struct urb *urb)
 		tty_buffer_request_room(tty, bytes_sent);
 		tty_insert_flip_string(tty, data + 2, bytes_sent);
 		tty_flip_buffer_push(tty);
+		tty_kref_put(tty);
 
 		/* again lockless, but debug info only */
 		priv->bytes_in += bytes_sent;
diff --git a/drivers/usb/serial/kobil_sct.c b/drivers/usb/serial/kobil_sct.c
index deba28ec77e8..cfcf37c2b957 100644
--- a/drivers/usb/serial/kobil_sct.c
+++ b/drivers/usb/serial/kobil_sct.c
@@ -383,7 +383,7 @@ static void kobil_read_int_callback(struct urb *urb)
 		return;
 	}
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (urb->actual_length) {
 
 		/* BEGIN DEBUG */
@@ -405,6 +405,7 @@ static void kobil_read_int_callback(struct urb *urb)
 		tty_insert_flip_string(tty, data, urb->actual_length);
 		tty_flip_buffer_push(tty);
 	}
+	tty_kref_put(tty);
 	/* someone sets the dev to 0 if the close method has been called */
 	port->interrupt_in_urb->dev = port->serial->dev;
 
diff --git a/drivers/usb/serial/mct_u232.c b/drivers/usb/serial/mct_u232.c
index 0ded8bd6ec85..9b2cef81cde0 100644
--- a/drivers/usb/serial/mct_u232.c
+++ b/drivers/usb/serial/mct_u232.c
@@ -563,10 +563,11 @@ static void mct_u232_read_int_callback(struct urb *urb)
 	 * Work-a-round: handle the 'usual' bulk-in pipe here
 	 */
 	if (urb->transfer_buffer_length > 2) {
-		tty = port->port.tty;
+		tty = tty_port_tty_get(&port->port);
 		if (urb->actual_length) {
 			tty_insert_flip_string(tty, data, urb->actual_length);
 			tty_flip_buffer_push(tty);
+			tty_kref_put(tty);
 		}
 		goto exit;
 	}
@@ -591,7 +592,7 @@ static void mct_u232_read_int_callback(struct urb *urb)
 	 * to look in to this before committing any code.
 	 */
 	if (priv->last_lsr & MCT_U232_LSR_ERR) {
-		tty = port->port.tty;
+		tty = tty_port_tty_get(&port->port);
 		/* Overrun Error */
 		if (priv->last_lsr & MCT_U232_LSR_OE) {
 		}
@@ -604,6 +605,7 @@ static void mct_u232_read_int_callback(struct urb *urb)
 		/* Break Indicator */
 		if (priv->last_lsr & MCT_U232_LSR_BI) {
 		}
+		tty_kref_put(tty);
 	}
 #endif
 	spin_unlock_irqrestore(&priv->lock, flags);
diff --git a/drivers/usb/serial/mos7720.c b/drivers/usb/serial/mos7720.c
index 7c4917d77c0a..7b538caec37f 100644
--- a/drivers/usb/serial/mos7720.c
+++ b/drivers/usb/serial/mos7720.c
@@ -216,12 +216,13 @@ static void mos7720_bulk_in_callback(struct urb *urb)
 
 	data = urb->transfer_buffer;
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (tty && urb->actual_length) {
 		tty_buffer_request_room(tty, urb->actual_length);
 		tty_insert_flip_string(tty, data, urb->actual_length);
 		tty_flip_buffer_push(tty);
 	}
+	tty_kref_put(tty);
 
 	if (!port->read_urb) {
 		dbg("URB KILLED !!!");
@@ -262,10 +263,11 @@ static void mos7720_bulk_out_data_callback(struct urb *urb)
 
 	dbg("Entering .........");
 
-	tty = mos7720_port->port->port.tty;
+	tty = tty_port_tty_get(&mos7720_port->port->port);
 
 	if (tty && mos7720_port->open)
 		tty_wakeup(tty);
+	tty_kref_put(tty);
 }
 
 /*
@@ -1267,29 +1269,6 @@ static int get_lsr_info(struct tty_struct *tty,
 	return 0;
 }
 
-/*
- * get_number_bytes_avail - get number of bytes available
- *
- * Purpose: Let user call ioctl to get the count of number of bytes available.
- */
-static int get_number_bytes_avail(struct moschip_port *mos7720_port,
-				  unsigned int __user *value)
-{
-	unsigned int result = 0;
-	struct tty_struct *tty = mos7720_port->port->port.tty;
-
-	if (!tty)
-		return -ENOIOCTLCMD;
-
-	result = tty->read_cnt;
-
-	dbg("%s(%d) = %d", __func__,  mos7720_port->port->number, result);
-	if (copy_to_user(value, &result, sizeof(int)))
-		return -EFAULT;
-
-	return -ENOIOCTLCMD;
-}
-
 static int set_modem_info(struct moschip_port *mos7720_port, unsigned int cmd,
 			  unsigned int __user *value)
 {
@@ -1409,13 +1388,6 @@ static int mos7720_ioctl(struct tty_struct *tty, struct file *file,
 	dbg("%s - port %d, cmd = 0x%x", __func__, port->number, cmd);
 
 	switch (cmd) {
-	case TIOCINQ:
-		/* return number of bytes available */
-		dbg("%s (%d) TIOCINQ", __func__,  port->number);
-		return get_number_bytes_avail(mos7720_port,
-					      (unsigned int __user *)arg);
-		break;
-
 	case TIOCSERGETLSR:
 		dbg("%s (%d) TIOCSERGETLSR", __func__,  port->number);
 		return get_lsr_info(tty, mos7720_port,
diff --git a/drivers/usb/serial/mos7840.c b/drivers/usb/serial/mos7840.c
index 09d82062b973..60543d79ef56 100644
--- a/drivers/usb/serial/mos7840.c
+++ b/drivers/usb/serial/mos7840.c
@@ -709,12 +709,13 @@ static void mos7840_bulk_in_callback(struct urb *urb)
 	dbg("%s", "Entering ........... \n");
 
 	if (urb->actual_length) {
-		tty = mos7840_port->port->port.tty;
+		tty = tty_port_tty_get(&mos7840_port->port->port);
 		if (tty) {
 			tty_buffer_request_room(tty, urb->actual_length);
 			tty_insert_flip_string(tty, data, urb->actual_length);
 			dbg(" %s \n", data);
 			tty_flip_buffer_push(tty);
+			tty_kref_put(tty);
 		}
 		mos7840_port->icount.rx += urb->actual_length;
 		smp_wmb();
@@ -773,10 +774,10 @@ static void mos7840_bulk_out_data_callback(struct urb *urb)
 
 	dbg("%s \n", "Entering .........");
 
-	tty = mos7840_port->port->port.tty;
-
+	tty = tty_port_tty_get(&mos7840_port->port->port);
 	if (tty && mos7840_port->open)
 		tty_wakeup(tty);
+	tty_kref_put(tty);
 
 }
 
diff --git a/drivers/usb/serial/navman.c b/drivers/usb/serial/navman.c
index d6736531a0fa..bcdcbb822705 100644
--- a/drivers/usb/serial/navman.c
+++ b/drivers/usb/serial/navman.c
@@ -64,12 +64,13 @@ static void navman_read_int_callback(struct urb *urb)
 	usb_serial_debug_data(debug, &port->dev, __func__,
 			      urb->actual_length, data);
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (tty && urb->actual_length) {
 		tty_buffer_request_room(tty, urb->actual_length);
 		tty_insert_flip_string(tty, data, urb->actual_length);
 		tty_flip_buffer_push(tty);
 	}
+	tty_kref_put(tty);
 
 exit:
 	result = usb_submit_urb(urb, GFP_ATOMIC);
diff --git a/drivers/usb/serial/omninet.c b/drivers/usb/serial/omninet.c
index ae8e227f3db2..c4d70b0f1e48 100644
--- a/drivers/usb/serial/omninet.c
+++ b/drivers/usb/serial/omninet.c
@@ -172,7 +172,7 @@ static int omninet_open(struct tty_struct *tty,
 	dbg("%s - port %d", __func__, port->number);
 
 	wport = serial->port[1];
-	wport->port.tty = tty;		/* FIXME */
+	tty_port_tty_set(&wport->port, tty);
 
 	/* Start reading from the device */
 	usb_fill_bulk_urb(port->read_urb, serial->dev,
@@ -229,9 +229,11 @@ static void omninet_read_bulk_callback(struct urb *urb)
 	}
 
 	if (urb->actual_length && header->oh_len) {
-		tty_insert_flip_string(port->port.tty,
-			data + OMNINET_DATAOFFSET, header->oh_len);
-		tty_flip_buffer_push(port->port.tty);
+		struct tty_struct *tty = tty_port_tty_get(&port->port);
+		tty_insert_flip_string(tty, data + OMNINET_DATAOFFSET,
+							header->oh_len);
+		tty_flip_buffer_push(tty);
+		tty_kref_put(tty);
 	}
 
 	/* Continue trying to always read  */
diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c
index 73f8277f88f2..6b1727e751e3 100644
--- a/drivers/usb/serial/option.c
+++ b/drivers/usb/serial/option.c
@@ -571,14 +571,14 @@ static void option_indat_callback(struct urb *urb)
 		dbg("%s: nonzero status: %d on endpoint %02x.",
 		    __func__, status, endpoint);
 	} else {
-		tty = port->port.tty;
+		tty = tty_port_tty_get(&port->port);
 		if (urb->actual_length) {
 			tty_buffer_request_room(tty, urb->actual_length);
 			tty_insert_flip_string(tty, data, urb->actual_length);
 			tty_flip_buffer_push(tty);
-		} else {
+		} else 
 			dbg("%s: empty read urb received", __func__);
-		}
+		tty_kref_put(tty);
 
 		/* Resubmit urb so we continue receiving */
 		if (port->port.count && status != -ESHUTDOWN) {
@@ -647,9 +647,13 @@ static void option_instat_callback(struct urb *urb)
 			portdata->dsr_state = ((signals & 0x02) ? 1 : 0);
 			portdata->ri_state = ((signals & 0x08) ? 1 : 0);
 
-			if (port->port.tty && !C_CLOCAL(port->port.tty) &&
-					old_dcd_state && !portdata->dcd_state)
-				tty_hangup(port->port.tty);
+			if (old_dcd_state && !portdata->dcd_state) {
+				struct tty_struct *tty =
+						tty_port_tty_get(&port->port);
+				if (tty && !C_CLOCAL(tty))
+					tty_hangup(tty);
+				tty_kref_put(tty);
+			}
 		} else {
 			dbg("%s: type %x req %x", __func__,
 				req_pkt->bRequestType, req_pkt->bRequest);
@@ -793,7 +797,7 @@ static void option_close(struct tty_struct *tty,
 		for (i = 0; i < N_OUT_URB; i++)
 			usb_kill_urb(portdata->out_urbs[i]);
 	}
-	port->port.tty = NULL;	/* FIXME */
+	tty_port_tty_set(&port->port, NULL);
 }
 
 /* Helper functions used by option_setup_urbs */
diff --git a/drivers/usb/serial/oti6858.c b/drivers/usb/serial/oti6858.c
index 81db5715ee25..ba551f00f16f 100644
--- a/drivers/usb/serial/oti6858.c
+++ b/drivers/usb/serial/oti6858.c
@@ -224,10 +224,6 @@ struct oti6858_private {
 	struct usb_serial_port *port;   /* USB port with which associated */
 };
 
-#undef dbg
-/* #define dbg(format, arg...) printk(KERN_INFO "%s: " format "\n", __FILE__, ## arg) */
-#define dbg(format, arg...) printk(KERN_INFO "" format "\n", ## arg)
-
 static void setup_line(struct work_struct *work)
 {
 	struct oti6858_private *priv = container_of(work,
@@ -1002,11 +998,12 @@ static void oti6858_read_bulk_callback(struct urb *urb)
 		return;
 	}
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (tty != NULL && urb->actual_length > 0) {
 		tty_insert_flip_string(tty, data, urb->actual_length);
 		tty_flip_buffer_push(tty);
 	}
+	tty_kref_put(tty);
 
 	/* schedule the interrupt urb if we are still open */
 	if (port->port.count != 0) {
diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c
index 1ede1441cb1b..908437847165 100644
--- a/drivers/usb/serial/pl2303.c
+++ b/drivers/usb/serial/pl2303.c
@@ -154,7 +154,6 @@ struct pl2303_private {
 	wait_queue_head_t delta_msr_wait;
 	u8 line_control;
 	u8 line_status;
-	u8 termios_initialized;
 	enum pl2303_type type;
 };
 
@@ -526,16 +525,6 @@ static void pl2303_set_termios(struct tty_struct *tty,
 
 	dbg("%s -  port %d", __func__, port->number);
 
-	spin_lock_irqsave(&priv->lock, flags);
-	if (!priv->termios_initialized) {
-		*(tty->termios) = tty_std_termios;
-		tty->termios->c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL;
-		tty->termios->c_ispeed = 9600;
-		tty->termios->c_ospeed = 9600;
-		priv->termios_initialized = 1;
-	}
-	spin_unlock_irqrestore(&priv->lock, flags);
-
 	/* The PL2303 is reported to lose bytes if you change
 	   serial settings even to the same values as before. Thus
 	   we actually need to filter in this specific case */
@@ -1057,7 +1046,7 @@ static void pl2303_read_bulk_callback(struct urb *urb)
 		tty_flag = TTY_FRAME;
 	dbg("%s - tty_flag = %d", __func__, tty_flag);
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (tty && urb->actual_length) {
 		tty_buffer_request_room(tty, urb->actual_length + 1);
 		/* overrun is special, not associated with a char */
@@ -1067,7 +1056,7 @@ static void pl2303_read_bulk_callback(struct urb *urb)
 			tty_insert_flip_char(tty, data[i], tty_flag);
 		tty_flip_buffer_push(tty);
 	}
-
+	tty_kref_put(tty);
 	/* Schedule the next read _if_ we are still open */
 	if (port->port.count) {
 		urb->dev = port->serial->dev;
diff --git a/drivers/usb/serial/safe_serial.c b/drivers/usb/serial/safe_serial.c
index def52d07a4ea..72903ac9f5c0 100644
--- a/drivers/usb/serial/safe_serial.c
+++ b/drivers/usb/serial/safe_serial.c
@@ -217,6 +217,7 @@ static void safe_read_bulk_callback(struct urb *urb)
 	struct usb_serial_port *port =  urb->context;
 	unsigned char *data = urb->transfer_buffer;
 	unsigned char length = urb->actual_length;
+	struct tty_struct *tty;
 	int result;
 	int status = urb->status;
 
@@ -242,6 +243,7 @@ static void safe_read_bulk_callback(struct urb *urb)
 		printk("\n");
 	}
 #endif
+	tty = tty_port_tty_get(&port->port);
 	if (safe) {
 		__u16 fcs;
 		fcs = fcs_compute10(data, length, CRC10_INITFCS);
@@ -250,9 +252,9 @@ static void safe_read_bulk_callback(struct urb *urb)
 			if (actual_length <= (length - 2)) {
 				info("%s - actual: %d", __func__,
 							actual_length);
-				tty_insert_flip_string(port->port.tty,
+				tty_insert_flip_string(tty,
 							data, actual_length);
-				tty_flip_buffer_push(port->port.tty);
+				tty_flip_buffer_push(tty);
 			} else {
 				err("%s - inconsistent lengths %d:%d",
 					__func__, actual_length, length);
@@ -261,9 +263,10 @@ static void safe_read_bulk_callback(struct urb *urb)
 			err("%s - bad CRC %x", __func__, fcs);
 		}
 	} else {
-		tty_insert_flip_string(port->port.tty, data, length);
-		tty_flip_buffer_push(port->port.tty);
+		tty_insert_flip_string(tty, data, length);
+		tty_flip_buffer_push(tty);
 	}
+	tty_kref_put(tty);
 
 	/* Continue trying to always read  */
 	usb_fill_bulk_urb(urb, port->serial->dev,
diff --git a/drivers/usb/serial/sierra.c b/drivers/usb/serial/sierra.c
index ea1a103c99be..8b9eaf383679 100644
--- a/drivers/usb/serial/sierra.c
+++ b/drivers/usb/serial/sierra.c
@@ -440,14 +440,14 @@ static void sierra_indat_callback(struct urb *urb)
 		dbg("%s: nonzero status: %d on endpoint %02x.",
 		    __func__, status, endpoint);
 	} else {
-		tty = port->port.tty;
 		if (urb->actual_length) {
+		tty = tty_port_tty_get(&port->port);
 			tty_buffer_request_room(tty, urb->actual_length);
 			tty_insert_flip_string(tty, data, urb->actual_length);
 			tty_flip_buffer_push(tty);
-		} else {
+			tty_kref_put(tty);
+		} else
 			dbg("%s: empty read urb received", __func__);
-		}
 
 		/* Resubmit urb so we continue receiving */
 		if (port->port.count && status != -ESHUTDOWN) {
@@ -485,6 +485,7 @@ static void sierra_instat_callback(struct urb *urb)
 			unsigned char signals = *((unsigned char *)
 					urb->transfer_buffer +
 					sizeof(struct usb_ctrlrequest));
+			struct tty_struct *tty;
 
 			dbg("%s: signal x%x", __func__, signals);
 
@@ -494,9 +495,11 @@ static void sierra_instat_callback(struct urb *urb)
 			portdata->dsr_state = ((signals & 0x02) ? 1 : 0);
 			portdata->ri_state = ((signals & 0x08) ? 1 : 0);
 
-			if (port->port.tty && !C_CLOCAL(port->port.tty) &&
+			tty = tty_port_tty_get(&port->port);
+			if (tty && !C_CLOCAL(tty) &&
 					old_dcd_state && !portdata->dcd_state)
-				tty_hangup(port->port.tty);
+				tty_hangup(tty);
+			tty_kref_put(tty);
 		} else {
 			dbg("%s: type %x req %x", __func__,
 				req_pkt->bRequestType, req_pkt->bRequest);
@@ -616,8 +619,7 @@ static void sierra_close(struct tty_struct *tty,
 	}
 
 	usb_kill_urb(port->interrupt_in_urb);
-
-	port->port.tty = NULL;	/* FIXME */
+	tty_port_tty_set(&port->port, NULL);
 }
 
 static int sierra_startup(struct usb_serial *serial)
diff --git a/drivers/usb/serial/spcp8x5.c b/drivers/usb/serial/spcp8x5.c
index 283cf6b36b2c..1533d6e12238 100644
--- a/drivers/usb/serial/spcp8x5.c
+++ b/drivers/usb/serial/spcp8x5.c
@@ -755,7 +755,7 @@ static void spcp8x5_read_bulk_callback(struct urb *urb)
 		tty_flag = TTY_FRAME;
 	dev_dbg(&port->dev, "tty_flag = %d\n", tty_flag);
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (tty && urb->actual_length) {
 		tty_buffer_request_room(tty, urb->actual_length + 1);
 		/* overrun is special, not associated with a char */
@@ -765,6 +765,7 @@ static void spcp8x5_read_bulk_callback(struct urb *urb)
 			tty_insert_flip_char(tty, data[i], tty_flag);
 		tty_flip_buffer_push(tty);
 	}
+	tty_kref_put(tty);
 
 	/* Schedule the next read _if_ we are still open */
 	if (port->port.count) {
diff --git a/drivers/usb/serial/ti_usb_3410_5052.c b/drivers/usb/serial/ti_usb_3410_5052.c
index 9a3e495c769c..c90237d48b0e 100644
--- a/drivers/usb/serial/ti_usb_3410_5052.c
+++ b/drivers/usb/serial/ti_usb_3410_5052.c
@@ -179,7 +179,7 @@ static int ti_set_mcr(struct ti_port *tport, unsigned int mcr);
 static int ti_get_lsr(struct ti_port *tport);
 static int ti_get_serial_info(struct ti_port *tport,
 	struct serial_struct __user *ret_arg);
-static int ti_set_serial_info(struct ti_port *tport,
+static int ti_set_serial_info(struct tty_struct *tty, struct ti_port *tport,
 	struct serial_struct __user *new_arg);
 static void ti_handle_new_msr(struct ti_port *tport, __u8 msr);
 
@@ -857,8 +857,8 @@ static int ti_ioctl(struct tty_struct *tty, struct file *file,
 				(struct serial_struct __user *)arg);
 	case TIOCSSERIAL:
 		dbg("%s - (%d) TIOCSSERIAL", __func__, port->number);
-		return ti_set_serial_info(tport,
-					(struct serial_struct __user *)arg);
+		return ti_set_serial_info(tty, tport,
+				(struct serial_struct __user *)arg);
 	case TIOCMIWAIT:
 		dbg("%s - (%d) TIOCMIWAIT", __func__, port->number);
 		cprev = tport->tp_icount;
@@ -1211,6 +1211,7 @@ static void ti_bulk_in_callback(struct urb *urb)
 	struct device *dev = &urb->dev->dev;
 	int status = urb->status;
 	int retval = 0;
+	struct tty_struct *tty;
 
 	dbg("%s", __func__);
 
@@ -1239,20 +1240,22 @@ static void ti_bulk_in_callback(struct urb *urb)
 		return;
 	}
 
-	if (port->port.tty && urb->actual_length) {
+	tty = tty_port_tty_get(&port->port);
+	if (tty && urb->actual_length) {
 		usb_serial_debug_data(debug, dev, __func__,
 			urb->actual_length, urb->transfer_buffer);
 
 		if (!tport->tp_is_open)
 			dbg("%s - port closed, dropping data", __func__);
 		else
-			ti_recv(&urb->dev->dev, port->port.tty,
+			ti_recv(&urb->dev->dev, tty,
 						urb->transfer_buffer,
 						urb->actual_length);
 
 		spin_lock(&tport->tp_lock);
 		tport->tp_icount.rx += urb->actual_length;
 		spin_unlock(&tport->tp_lock);
+		tty_kref_put(tty);
 	}
 
 exit:
@@ -1330,7 +1333,7 @@ static void ti_send(struct ti_port *tport)
 {
 	int count, result;
 	struct usb_serial_port *port = tport->tp_port;
-	struct tty_struct *tty = port->port.tty;	/* FIXME */
+	struct tty_struct *tty = tty_port_tty_get(&port->port);	/* FIXME */
 	unsigned long flags;
 
 
@@ -1338,19 +1341,15 @@ static void ti_send(struct ti_port *tport)
 
 	spin_lock_irqsave(&tport->tp_lock, flags);
 
-	if (tport->tp_write_urb_in_use) {
-		spin_unlock_irqrestore(&tport->tp_lock, flags);
-		return;
-	}
+	if (tport->tp_write_urb_in_use)
+		goto unlock;
 
 	count = ti_buf_get(tport->tp_write_buf,
 				port->write_urb->transfer_buffer,
 				port->bulk_out_size);
 
-	if (count == 0) {
-		spin_unlock_irqrestore(&tport->tp_lock, flags);
-		return;
-	}
+	if (count == 0)
+		goto unlock;
 
 	tport->tp_write_urb_in_use = 1;
 
@@ -1380,7 +1379,13 @@ static void ti_send(struct ti_port *tport)
 	/* more room in the buffer for new writes, wakeup */
 	if (tty)
 		tty_wakeup(tty);
+	tty_kref_put(tty);
 	wake_up_interruptible(&tport->tp_write_wait);
+	return;
+unlock:
+	spin_unlock_irqrestore(&tport->tp_lock, flags);
+	tty_kref_put(tty);
+	return;
 }
 
 
@@ -1464,20 +1469,16 @@ static int ti_get_serial_info(struct ti_port *tport,
 }
 
 
-static int ti_set_serial_info(struct ti_port *tport,
+static int ti_set_serial_info(struct tty_struct *tty, struct ti_port *tport,
 	struct serial_struct __user *new_arg)
 {
-	struct usb_serial_port *port = tport->tp_port;
 	struct serial_struct new_serial;
 
 	if (copy_from_user(&new_serial, new_arg, sizeof(new_serial)))
 		return -EFAULT;
 
 	tport->tp_flags = new_serial.flags & TI_SET_SERIAL_FLAGS;
-	/* FIXME */
-	if (port->port.tty)
-		port->port.tty->low_latency =
-			(tport->tp_flags & ASYNC_LOW_LATENCY) ? 1 : 0;
+	tty->low_latency = (tport->tp_flags & ASYNC_LOW_LATENCY) ? 1 : 0;
 	tport->tp_closing_wait = new_serial.closing_wait;
 
 	return 0;
@@ -1510,7 +1511,7 @@ static void ti_handle_new_msr(struct ti_port *tport, __u8 msr)
 	tport->tp_msr = msr & TI_MSR_MASK;
 
 	/* handle CTS flow control */
-	tty = tport->tp_port->port.tty;
+	tty = tty_port_tty_get(&tport->tp_port->port);
 	if (tty && C_CRTSCTS(tty)) {
 		if (msr & TI_MSR_CTS) {
 			tty->hw_stopped = 0;
@@ -1519,6 +1520,7 @@ static void ti_handle_new_msr(struct ti_port *tport, __u8 msr)
 			tty->hw_stopped = 1;
 		}
 	}
+	tty_kref_put(tty);
 }
 
 
diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c
index 4f7f9e3ae0a4..e7d4246027b2 100644
--- a/drivers/usb/serial/usb-serial.c
+++ b/drivers/usb/serial/usb-serial.c
@@ -214,7 +214,7 @@ static int serial_open (struct tty_struct *tty, struct file *filp)
 	/* set up our port structure making the tty driver
 	 * remember our port object, and us it */
 	tty->driver_data = port;
-	port->port.tty = tty;
+	tty_port_tty_set(&port->port, tty);
 
 	if (port->port.count == 1) {
 
@@ -246,7 +246,7 @@ bailout_module_put:
 bailout_mutex_unlock:
 	port->port.count = 0;
 	tty->driver_data = NULL;
-	port->port.tty = NULL;
+	tty_port_tty_set(&port->port, NULL);
 	mutex_unlock(&port->mutex);
 bailout_kref_put:
 	usb_serial_put(serial);
@@ -276,10 +276,11 @@ static void serial_close(struct tty_struct *tty, struct file *filp)
 		port->serial->type->close(tty, port, filp);
 
 	if (port->port.count == (port->console? 1 : 0)) {
-		if (port->port.tty) {
-			if (port->port.tty->driver_data)
-				port->port.tty->driver_data = NULL;
-			port->port.tty = NULL;
+		struct tty_struct *tty = tty_port_tty_get(&port->port);
+		if (tty) {
+			if (tty->driver_data)
+				tty->driver_data = NULL;
+			tty_port_tty_set(&port->port, NULL);
 		}
 	}
 
@@ -508,11 +509,12 @@ static void usb_serial_port_work(struct work_struct *work)
 	if (!port)
 		return;
 
-	tty = port->port.tty;
+	tty = tty_port_tty_get(&port->port);
 	if (!tty)
 		return;
 
 	tty_wakeup(tty);
+	tty_kref_put(tty);
 }
 
 static void port_release(struct device *dev)
@@ -819,6 +821,7 @@ int usb_serial_probe(struct usb_interface *interface,
 		port = kzalloc(sizeof(struct usb_serial_port), GFP_KERNEL);
 		if (!port)
 			goto probe_error;
+		tty_port_init(&port->port);
 		port->serial = serial;
 		spin_lock_init(&port->lock);
 		mutex_init(&port->mutex);
@@ -1040,8 +1043,11 @@ void usb_serial_disconnect(struct usb_interface *interface)
 	for (i = 0; i < serial->num_ports; ++i) {
 		port = serial->port[i];
 		if (port) {
-			if (port->port.tty)
-				tty_hangup(port->port.tty);
+			struct tty_struct *tty = tty_port_tty_get(&port->port);
+			if (tty) {
+				tty_hangup(tty);
+				tty_kref_put(tty);
+			}
 			kill_traffic(port);
 		}
 	}
diff --git a/drivers/usb/serial/visor.c b/drivers/usb/serial/visor.c
index cf8924f9a2cc..a6d1c75a1c89 100644
--- a/drivers/usb/serial/visor.c
+++ b/drivers/usb/serial/visor.c
@@ -499,7 +499,7 @@ static void visor_read_bulk_callback(struct urb *urb)
 	int status = urb->status;
 	struct tty_struct *tty;
 	int result;
-	int available_room;
+	int available_room = 0;
 
 	dbg("%s - port %d", __func__, port->number);
 
@@ -512,13 +512,17 @@ static void visor_read_bulk_callback(struct urb *urb)
 	usb_serial_debug_data(debug, &port->dev, __func__,
 						urb->actual_length, data);
 
-	tty = port->port.tty;
-	if (tty && urb->actual_length) {
-		available_room = tty_buffer_request_room(tty,
+	if (urb->actual_length) {
+		tty = tty_port_tty_get(&port->port);
+		if (tty) {
+			available_room = tty_buffer_request_room(tty,
 							urb->actual_length);
-		if (available_room) {
-			tty_insert_flip_string(tty, data, available_room);
-			tty_flip_buffer_push(tty);
+			if (available_room) {
+				tty_insert_flip_string(tty, data,
+							available_room);
+				tty_flip_buffer_push(tty);
+			}
+			tty_kref_put(tty);
 		}
 		spin_lock(&priv->lock);
 		priv->bytes_in += available_room;
diff --git a/drivers/usb/serial/whiteheat.c b/drivers/usb/serial/whiteheat.c
index 3a9d14384a43..11c8b97a5177 100644
--- a/drivers/usb/serial/whiteheat.c
+++ b/drivers/usb/serial/whiteheat.c
@@ -1481,7 +1481,7 @@ static void rx_data_softint(struct work_struct *work)
 	struct whiteheat_private *info =
 		container_of(work, struct whiteheat_private, rx_work);
 	struct usb_serial_port *port = info->port;
-	struct tty_struct *tty = port->port.tty;
+	struct tty_struct *tty = tty_port_tty_get(&port->port);
 	struct whiteheat_urb_wrap *wrap;
 	struct urb *urb;
 	unsigned long flags;
@@ -1493,7 +1493,7 @@ static void rx_data_softint(struct work_struct *work)
 	spin_lock_irqsave(&info->lock, flags);
 	if (info->flags & THROTTLED) {
 		spin_unlock_irqrestore(&info->lock, flags);
-		return;
+		goto out;
 	}
 
 	list_for_each_safe(tmp, tmp2, &info->rx_urb_q) {
@@ -1513,7 +1513,7 @@ static void rx_data_softint(struct work_struct *work)
 				spin_unlock_irqrestore(&info->lock, flags);
 				tty_flip_buffer_push(tty);
 				schedule_work(&info->rx_work);
-				return;
+				goto out;
 			}
 			tty_insert_flip_string(tty, urb->transfer_buffer, len);
 			sent += len;
@@ -1536,6 +1536,8 @@ static void rx_data_softint(struct work_struct *work)
 
 	if (sent)
 		tty_flip_buffer_push(tty);
+out:
+	tty_kref_put(tty);
 }
 
 
diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index d85a74c64b54..f79c2040758b 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -673,7 +673,6 @@ config FB_VESA
 	select FB_CFB_FILLRECT
 	select FB_CFB_COPYAREA
 	select FB_CFB_IMAGEBLIT
-	select VIDEO_SELECT
 	help
 	  This is the frame buffer device driver for generic VESA 2.0
 	  compliant graphic cards. The older VESA 1.2 cards are not supported.
@@ -1578,7 +1577,6 @@ config FB_CYBLA
 	tristate "Cyberblade/i1 support"
 	depends on FB && PCI && X86_32 && !64BIT
 	select FB_CFB_IMAGEBLIT
-	select VIDEO_SELECT
 	---help---
 	  This driver is supposed to support the Trident Cyberblade/i1
 	  graphics core integrated in the VIA VT8601A North Bridge,
diff --git a/drivers/video/backlight/mbp_nvidia_bl.c b/drivers/video/backlight/mbp_nvidia_bl.c
index 385cba40ea87..06964af761c6 100644
--- a/drivers/video/backlight/mbp_nvidia_bl.c
+++ b/drivers/video/backlight/mbp_nvidia_bl.c
@@ -111,6 +111,4 @@ module_exit(mbp_exit);
 MODULE_AUTHOR("Matthew Garrett <mjg@redhat.com>");
 MODULE_DESCRIPTION("Nvidia-based Macbook Pro Backlight Driver");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS("svnAppleInc.:pnMacBookPro3,1");
-MODULE_ALIAS("svnAppleInc.:pnMacBookPro3,2");
-MODULE_ALIAS("svnAppleInc.:pnMacBookPro4,1");
+MODULE_DEVICE_TABLE(dmi, mbp_device_table);
diff --git a/drivers/video/bw2.c b/drivers/video/bw2.c
index e721644bad74..1e35ba6f18e0 100644
--- a/drivers/video/bw2.c
+++ b/drivers/video/bw2.c
@@ -372,7 +372,7 @@ static int __devexit bw2_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id bw2_match[] = {
+static const struct of_device_id bw2_match[] = {
 	{
 		.name = "bwtwo",
 	},
diff --git a/drivers/video/cg14.c b/drivers/video/cg14.c
index b17e74671779..a2d1882791a5 100644
--- a/drivers/video/cg14.c
+++ b/drivers/video/cg14.c
@@ -589,7 +589,7 @@ static int __devexit cg14_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id cg14_match[] = {
+static const struct of_device_id cg14_match[] = {
 	{
 		.name = "cgfourteen",
 	},
diff --git a/drivers/video/cg3.c b/drivers/video/cg3.c
index 3aa7b6cb0268..99f87fb61d05 100644
--- a/drivers/video/cg3.c
+++ b/drivers/video/cg3.c
@@ -456,7 +456,7 @@ static int __devexit cg3_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id cg3_match[] = {
+static const struct of_device_id cg3_match[] = {
 	{
 		.name = "cgthree",
 	},
diff --git a/drivers/video/cg6.c b/drivers/video/cg6.c
index 2f64bb3bd254..940ec04f0f1b 100644
--- a/drivers/video/cg6.c
+++ b/drivers/video/cg6.c
@@ -34,10 +34,11 @@ static int cg6_blank(int, struct fb_info *);
 
 static void cg6_imageblit(struct fb_info *, const struct fb_image *);
 static void cg6_fillrect(struct fb_info *, const struct fb_fillrect *);
+static void cg6_copyarea(struct fb_info *info, const struct fb_copyarea *area);
 static int cg6_sync(struct fb_info *);
 static int cg6_mmap(struct fb_info *, struct vm_area_struct *);
 static int cg6_ioctl(struct fb_info *, unsigned int, unsigned long);
-static void cg6_copyarea(struct fb_info *info, const struct fb_copyarea *area);
+static int cg6_pan_display(struct fb_var_screeninfo *, struct fb_info *);
 
 /*
  *  Frame buffer operations
@@ -47,6 +48,7 @@ static struct fb_ops cg6_ops = {
 	.owner			= THIS_MODULE,
 	.fb_setcolreg		= cg6_setcolreg,
 	.fb_blank		= cg6_blank,
+	.fb_pan_display		= cg6_pan_display,
 	.fb_fillrect		= cg6_fillrect,
 	.fb_copyarea		= cg6_copyarea,
 	.fb_imageblit		= cg6_imageblit,
@@ -161,6 +163,7 @@ static struct fb_ops cg6_ops = {
 #define CG6_THC_MISC_INT_ENAB		(1 << 5)
 #define CG6_THC_MISC_INT		(1 << 4)
 #define CG6_THC_MISC_INIT		0x9f
+#define CG6_THC_CURSOFF			((65536-32) | ((65536-32) << 16))
 
 /* The contents are unknown */
 struct cg6_tec {
@@ -280,6 +283,33 @@ static int cg6_sync(struct fb_info *info)
 	return 0;
 }
 
+static void cg6_switch_from_graph(struct cg6_par *par)
+{
+	struct cg6_thc __iomem *thc = par->thc;
+	unsigned long flags;
+
+	spin_lock_irqsave(&par->lock, flags);
+
+	/* Hide the cursor. */
+	sbus_writel(CG6_THC_CURSOFF, &thc->thc_cursxy);
+
+	spin_unlock_irqrestore(&par->lock, flags);
+}
+
+static int cg6_pan_display(struct fb_var_screeninfo *var, struct fb_info *info)
+{
+	struct cg6_par *par = (struct cg6_par *)info->par;
+
+	/* We just use this to catch switches out of
+	 * graphics mode.
+	 */
+	cg6_switch_from_graph(par);
+
+	if (var->xoffset || var->yoffset || var->vmode)
+		return -EINVAL;
+	return 0;
+}
+
 /**
  *	cg6_fillrect -	Draws a rectangle on the screen.
  *
@@ -643,9 +673,13 @@ static void __devinit cg6_chip_init(struct fb_info *info)
 	struct cg6_par *par = (struct cg6_par *)info->par;
 	struct cg6_tec __iomem *tec = par->tec;
 	struct cg6_fbc __iomem *fbc = par->fbc;
+	struct cg6_thc __iomem *thc = par->thc;
 	u32 rev, conf, mode;
 	int i;
 
+	/* Hide the cursor. */
+	sbus_writel(CG6_THC_CURSOFF, &thc->thc_cursxy);
+
 	/* Turn off stuff in the Transform Engine. */
 	sbus_writel(0, &tec->tec_matrix);
 	sbus_writel(0, &tec->tec_clip);
@@ -814,7 +848,7 @@ static int __devexit cg6_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id cg6_match[] = {
+static const struct of_device_id cg6_match[] = {
 	{
 		.name = "cgsix",
 	},
diff --git a/drivers/video/console/Kconfig b/drivers/video/console/Kconfig
index 06f87b04f207..2f50a80b413e 100644
--- a/drivers/video/console/Kconfig
+++ b/drivers/video/console/Kconfig
@@ -43,22 +43,6 @@ config VGACON_SOFT_SCROLLBACK_SIZE
 	 buffer.  Each 64KB will give you approximately 16 80x25
 	 screenfuls of scrollback buffer
 
-config VIDEO_SELECT
-	bool "Video mode selection support"
-	depends on  X86 && VGA_CONSOLE
-	---help---
-	  This enables support for text mode selection on kernel startup. If
-	  you want to take advantage of some high-resolution text mode your
-	  card's BIOS offers, but the traditional Linux utilities like
-	  SVGATextMode don't, you can say Y here and set the mode using the
-	  "vga=" option from your boot loader (lilo or loadlin) or set
-	  "vga=ask" which brings up a video mode menu on kernel startup. (Try
-	  "man bootparam" or see the documentation of your boot loader about
-	  how to pass options to the kernel.)
-
-	  Read the file <file:Documentation/svga.txt> for more information
-	  about the Video mode selection support. If unsure, say N.
-
 config MDA_CONSOLE
 	depends on !M68K && !PARISC && ISA
 	tristate "MDA text console (dual-headed) (EXPERIMENTAL)"
diff --git a/drivers/video/ffb.c b/drivers/video/ffb.c
index 7992b13ee68f..9dbb9646081f 100644
--- a/drivers/video/ffb.c
+++ b/drivers/video/ffb.c
@@ -1042,7 +1042,7 @@ static int __devexit ffb_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id ffb_match[] = {
+static const struct of_device_id ffb_match[] = {
 	{
 		.name = "SUNW,ffb",
 	},
diff --git a/drivers/video/leo.c b/drivers/video/leo.c
index 13fea61d6ae4..7c7e8c2da9d9 100644
--- a/drivers/video/leo.c
+++ b/drivers/video/leo.c
@@ -33,6 +33,7 @@ static int leo_blank(int, struct fb_info *);
 
 static int leo_mmap(struct fb_info *, struct vm_area_struct *);
 static int leo_ioctl(struct fb_info *, unsigned int, unsigned long);
+static int leo_pan_display(struct fb_var_screeninfo *, struct fb_info *);
 
 /*
  *  Frame buffer operations
@@ -42,6 +43,7 @@ static struct fb_ops leo_ops = {
 	.owner			= THIS_MODULE,
 	.fb_setcolreg		= leo_setcolreg,
 	.fb_blank		= leo_blank,
+	.fb_pan_display		= leo_pan_display,
 	.fb_fillrect		= cfb_fillrect,
 	.fb_copyarea		= cfb_copyarea,
 	.fb_imageblit		= cfb_imageblit,
@@ -206,6 +208,60 @@ static void leo_wait(struct leo_lx_krn __iomem *lx_krn)
 	return;
 }
 
+static void leo_switch_from_graph(struct fb_info *info)
+{
+	struct leo_par *par = (struct leo_par *) info->par;
+	struct leo_ld_ss0 __iomem *ss = par->ld_ss0;
+	struct leo_cursor __iomem *cursor = par->cursor;
+	unsigned long flags;
+	u32 val;
+
+	spin_lock_irqsave(&par->lock, flags);
+
+	par->extent = ((info->var.xres - 1) |
+		       ((info->var.yres - 1) << 16));
+
+	sbus_writel(0xffffffff, &ss->wid);
+	sbus_writel(0xffff, &ss->wmask);
+	sbus_writel(0, &ss->vclipmin);
+	sbus_writel(par->extent, &ss->vclipmax);
+	sbus_writel(0, &ss->fg);
+	sbus_writel(0xff000000, &ss->planemask);
+	sbus_writel(0x310850, &ss->rop);
+	sbus_writel(0, &ss->widclip);
+	sbus_writel((info->var.xres-1) | ((info->var.yres-1) << 11),
+		    &par->lc_ss0_usr->extent);
+	sbus_writel(4, &par->lc_ss0_usr->addrspace);
+	sbus_writel(0x80000000, &par->lc_ss0_usr->fill);
+	sbus_writel(0, &par->lc_ss0_usr->fontt);
+	do {
+		val = sbus_readl(&par->lc_ss0_usr->csr);
+	} while (val & 0x20000000);
+
+	/* setup screen buffer for cfb_* functions */
+	sbus_writel(1, &ss->wid);
+	sbus_writel(0x00ffffff, &ss->planemask);
+	sbus_writel(0x310b90, &ss->rop);
+	sbus_writel(0, &par->lc_ss0_usr->addrspace);
+
+	/* hide cursor */
+	sbus_writel(sbus_readl(&cursor->cur_misc) & ~LEO_CUR_ENABLE, &cursor->cur_misc);
+
+	spin_unlock_irqrestore(&par->lock, flags);
+}
+
+static int leo_pan_display(struct fb_var_screeninfo *var, struct fb_info *info)
+{
+	/* We just use this to catch switches out of
+	 * graphics mode.
+	 */
+	leo_switch_from_graph(info);
+
+	if (var->xoffset || var->yoffset || var->vmode)
+		return -EINVAL;
+	return 0;
+}
+
 /**
  *      leo_setcolreg - Optional function. Sets a color register.
  *      @regno: boolean, 0 copy local, 1 get_user() function
@@ -454,44 +510,6 @@ static void leo_init_wids(struct fb_info *info)
 	leo_wid_put(info, &wl);
 }
 
-static void leo_switch_from_graph(struct fb_info *info)
-{
-	struct leo_par *par = (struct leo_par *) info->par;
-	struct leo_ld_ss0 __iomem *ss = par->ld_ss0;
-	unsigned long flags;
-	u32 val;
-
-	spin_lock_irqsave(&par->lock, flags);
-
-	par->extent = ((info->var.xres - 1) |
-		       ((info->var.yres - 1) << 16));
-
-	sbus_writel(0xffffffff, &ss->wid);
-	sbus_writel(0xffff, &ss->wmask);
-	sbus_writel(0, &ss->vclipmin);
-	sbus_writel(par->extent, &ss->vclipmax);
-	sbus_writel(0, &ss->fg);
-	sbus_writel(0xff000000, &ss->planemask);
-	sbus_writel(0x310850, &ss->rop);
-	sbus_writel(0, &ss->widclip);
-	sbus_writel((info->var.xres-1) | ((info->var.yres-1) << 11),
-		    &par->lc_ss0_usr->extent);
-	sbus_writel(4, &par->lc_ss0_usr->addrspace);
-	sbus_writel(0x80000000, &par->lc_ss0_usr->fill);
-	sbus_writel(0, &par->lc_ss0_usr->fontt);
-	do {
-		val = sbus_readl(&par->lc_ss0_usr->csr);
-	} while (val & 0x20000000);
-
-	/* setup screen buffer for cfb_* functions */
-	sbus_writel(1, &ss->wid);
-	sbus_writel(0x00ffffff, &ss->planemask);
-	sbus_writel(0x310b90, &ss->rop);
-	sbus_writel(0, &par->lc_ss0_usr->addrspace);
-
-	spin_unlock_irqrestore(&par->lock, flags);
-}
-
 static void leo_init_hw(struct fb_info *info)
 {
 	struct leo_par *par = (struct leo_par *) info->par;
@@ -641,7 +659,7 @@ static int __devexit leo_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id leo_match[] = {
+static const struct of_device_id leo_match[] = {
 	{
 		.name = "SUNW,leo",
 	},
diff --git a/drivers/video/p9100.c b/drivers/video/p9100.c
index 9e903454ffc1..7000f2cd5854 100644
--- a/drivers/video/p9100.c
+++ b/drivers/video/p9100.c
@@ -349,7 +349,7 @@ static int __devexit p9100_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id p9100_match[] = {
+static const struct of_device_id p9100_match[] = {
 	{
 		.name = "p9100",
 	},
diff --git a/drivers/video/tcx.c b/drivers/video/tcx.c
index 2a03f78bbb0d..643afbfe8277 100644
--- a/drivers/video/tcx.c
+++ b/drivers/video/tcx.c
@@ -505,7 +505,7 @@ static int __devexit tcx_remove(struct of_device *op)
 	return 0;
 }
 
-static struct of_device_id tcx_match[] = {
+static const struct of_device_id tcx_match[] = {
 	{
 		.name = "SUNW,tcx",
 	},
diff --git a/drivers/video/xen-fbfront.c b/drivers/video/xen-fbfront.c
index 47ed39b52f9c..a463b3dd837b 100644
--- a/drivers/video/xen-fbfront.c
+++ b/drivers/video/xen-fbfront.c
@@ -680,11 +680,11 @@ static struct xenbus_driver xenfb = {
 
 static int __init xenfb_init(void)
 {
-	if (!is_running_on_xen())
+	if (!xen_domain())
 		return -ENODEV;
 
 	/* Nothing to do if running in dom0. */
-	if (is_initial_xendomain())
+	if (xen_initial_domain())
 		return -ENODEV;
 
 	return xenbus_register_frontend(&xenfb);
diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index c51036716700..1a22fe782a27 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -66,6 +66,13 @@ config AT91RM9200_WATCHDOG
 	  Watchdog timer embedded into AT91RM9200 chips. This will reboot your
 	  system when the timeout is reached.
 
+config AT91SAM9X_WATCHDOG
+	tristate "AT91SAM9X watchdog"
+	depends on WATCHDOG && (ARCH_AT91SAM9260 || ARCH_AT91SAM9261)
+	help
+	  Watchdog timer embedded into AT91SAM9X chips. This will reboot your
+	  system when the timeout is reached.
+
 config 21285_WATCHDOG
 	tristate "DC21285 watchdog"
 	depends on FOOTBRIDGE
@@ -217,6 +224,15 @@ config DAVINCI_WATCHDOG
 	  NOTE: once enabled, this timer cannot be disabled.
 	  Say N if you are unsure.
 
+config ORION5X_WATCHDOG
+	tristate "Orion5x watchdog"
+	depends on ARCH_ORION5X
+	help
+	  Say Y here if to include support for the watchdog timer
+	  in the Orion5x ARM SoCs.
+	  To compile this driver as a module, choose M here: the
+	  module will be called orion5x_wdt.
+
 # ARM26 Architecture
 
 # AVR32 Architecture
@@ -416,6 +432,18 @@ config IT8712F_WDT
 	  To compile this driver as a module, choose M here: the
 	  module will be called it8712f_wdt.
 
+config IT87_WDT
+	tristate "IT87 Watchdog Timer"
+	depends on X86 && EXPERIMENTAL
+	---help---
+	  This is the driver for the hardware watchdog on the ITE IT8716,
+	  IT8718, IT8726, IT8712(Version J,K) Super I/O chips. This watchdog
+	  simply watches your kernel to make sure it doesn't freeze, and if
+	  it does, it reboots your computer after a certain amount of time.
+
+	  To compile this driver as a module, choose M here: the module will
+	  be called it87_wdt.
+
 config HP_WATCHDOG
 	tristate "HP Proliant iLO 2 Hardware Watchdog Timer"
 	depends on X86
@@ -573,6 +601,21 @@ config W83697HF_WDT
 
 	  Most people will say N.
 
+config W83697UG_WDT
+	tristate "W83697UG/W83697UF Watchdog Timer"
+	depends on X86
+	---help---
+	  This is the driver for the hardware watchdog on the W83697UG/UF
+	  chipset as used in MSI Fuzzy CX700 VIA motherboards (and likely others).
+	  This watchdog simply watches your kernel to make sure it doesn't
+	  freeze, and if it does, it reboots your computer after a certain
+	  amount of time.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called w83697ug_wdt.
+
+	  Most people will say N.
+
 config W83877F_WDT
 	tristate "W83877F (EMACS) Watchdog Timer"
 	depends on X86
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index e0ef123fbdea..e352bbb7630b 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -26,6 +26,7 @@ obj-$(CONFIG_USBPCWATCHDOG) += pcwd_usb.o
 
 # ARM Architecture
 obj-$(CONFIG_AT91RM9200_WATCHDOG) += at91rm9200_wdt.o
+obj-$(CONFIG_AT91SAM9X_WATCHDOG) += at91sam9_wdt.o
 obj-$(CONFIG_OMAP_WATCHDOG) += omap_wdt.o
 obj-$(CONFIG_21285_WATCHDOG) += wdt285.o
 obj-$(CONFIG_977_WATCHDOG) += wdt977.o
@@ -39,6 +40,7 @@ obj-$(CONFIG_EP93XX_WATCHDOG) += ep93xx_wdt.o
 obj-$(CONFIG_PNX4008_WATCHDOG) += pnx4008_wdt.o
 obj-$(CONFIG_IOP_WATCHDOG) += iop_wdt.o
 obj-$(CONFIG_DAVINCI_WATCHDOG) += davinci_wdt.o
+obj-$(CONFIG_ORION5X_WATCHDOG) += orion5x_wdt.o
 
 # ARM26 Architecture
 
@@ -71,6 +73,7 @@ ifeq ($(CONFIG_ITCO_VENDOR_SUPPORT),y)
 obj-$(CONFIG_ITCO_WDT) += iTCO_vendor_support.o
 endif
 obj-$(CONFIG_IT8712F_WDT) += it8712f_wdt.o
+obj-$(CONFIG_IT87_WDT) += it87_wdt.o
 obj-$(CONFIG_HP_WATCHDOG) += hpwdt.o
 obj-$(CONFIG_SC1200_WDT) += sc1200wdt.o
 obj-$(CONFIG_SCx200_WDT) += scx200_wdt.o
@@ -83,6 +86,7 @@ obj-$(CONFIG_CPU5_WDT) += cpu5wdt.o
 obj-$(CONFIG_SMSC37B787_WDT) += smsc37b787_wdt.o
 obj-$(CONFIG_W83627HF_WDT) += w83627hf_wdt.o
 obj-$(CONFIG_W83697HF_WDT) += w83697hf_wdt.o
+obj-$(CONFIG_W83697UG_WDT) += w83697ug_wdt.o
 obj-$(CONFIG_W83877F_WDT) += w83877f_wdt.o
 obj-$(CONFIG_W83977F_WDT) += w83977f_wdt.o
 obj-$(CONFIG_MACHZ_WDT) += machzwd.o
@@ -123,6 +127,9 @@ obj-$(CONFIG_SH_WDT) += shwdt.o
 
 # SPARC64 Architecture
 
+obj-$(CONFIG_WATCHDOG_RIO)		+= riowd.o
+obj-$(CONFIG_WATCHDOG_CP1XXX)		+= cpwd.o
+
 # XTENSA Architecture
 
 # Architecture Independant
diff --git a/drivers/watchdog/at91sam9_wdt.c b/drivers/watchdog/at91sam9_wdt.c
new file mode 100644
index 000000000000..b4babfc31586
--- /dev/null
+++ b/drivers/watchdog/at91sam9_wdt.c
@@ -0,0 +1,328 @@
+/*
+ * Watchdog driver for Atmel AT91SAM9x processors.
+ *
+ * Copyright (C) 2008 Renaud CERRATO r.cerrato@til-technologies.fr
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+/*
+ * The Watchdog Timer Mode Register can be only written to once. If the
+ * timeout need to be set from Linux, be sure that the bootstrap or the
+ * bootloader doesn't write to this register.
+ */
+
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/platform_device.h>
+#include <linux/types.h>
+#include <linux/watchdog.h>
+#include <linux/jiffies.h>
+#include <linux/timer.h>
+#include <linux/bitops.h>
+#include <linux/uaccess.h>
+
+#include <asm/arch/at91_wdt.h>
+
+#define DRV_NAME "AT91SAM9 Watchdog"
+
+/* AT91SAM9 watchdog runs a 12bit counter @ 256Hz,
+ * use this to convert a watchdog
+ * value from/to milliseconds.
+ */
+#define ms_to_ticks(t)	(((t << 8) / 1000) - 1)
+#define ticks_to_ms(t)	(((t + 1) * 1000) >> 8)
+
+/* Hardware timeout in seconds */
+#define WDT_HW_TIMEOUT 2
+
+/* Timer heartbeat (500ms) */
+#define WDT_TIMEOUT	(HZ/2)
+
+/* User land timeout */
+#define WDT_HEARTBEAT 15
+static int heartbeat = WDT_HEARTBEAT;
+module_param(heartbeat, int, 0);
+MODULE_PARM_DESC(heartbeat, "Watchdog heartbeats in seconds. "
+	"(default = " __MODULE_STRING(WDT_HEARTBEAT) ")");
+
+static int nowayout = WATCHDOG_NOWAYOUT;
+module_param(nowayout, int, 0);
+MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started "
+	"(default=" __MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
+
+static void at91_ping(unsigned long data);
+
+static struct {
+	unsigned long next_heartbeat;	/* the next_heartbeat for the timer */
+	unsigned long open;
+	char expect_close;
+	struct timer_list timer;	/* The timer that pings the watchdog */
+} at91wdt_private;
+
+/* ......................................................................... */
+
+
+/*
+ * Reload the watchdog timer.  (ie, pat the watchdog)
+ */
+static inline void at91_wdt_reset(void)
+{
+	at91_sys_write(AT91_WDT_CR, AT91_WDT_KEY | AT91_WDT_WDRSTT);
+}
+
+/*
+ * Timer tick
+ */
+static void at91_ping(unsigned long data)
+{
+	if (time_before(jiffies, at91wdt_private.next_heartbeat) ||
+			(!nowayout && !at91wdt_private.open)) {
+		at91_wdt_reset();
+		mod_timer(&at91wdt_private.timer, jiffies + WDT_TIMEOUT);
+	} else
+		printk(KERN_CRIT DRV_NAME": I will reset your machine !\n");
+}
+
+/*
+ * Watchdog device is opened, and watchdog starts running.
+ */
+static int at91_wdt_open(struct inode *inode, struct file *file)
+{
+	if (test_and_set_bit(0, &at91wdt_private.open))
+		return -EBUSY;
+
+	at91wdt_private.next_heartbeat = jiffies + heartbeat * HZ;
+	mod_timer(&at91wdt_private.timer, jiffies + WDT_TIMEOUT);
+
+	return nonseekable_open(inode, file);
+}
+
+/*
+ * Close the watchdog device.
+ */
+static int at91_wdt_close(struct inode *inode, struct file *file)
+{
+	clear_bit(0, &at91wdt_private.open);
+
+	/* stop internal ping */
+	if (!at91wdt_private.expect_close)
+		del_timer(&at91wdt_private.timer);
+
+	at91wdt_private.expect_close = 0;
+	return 0;
+}
+
+/*
+ * Set the watchdog time interval in 1/256Hz (write-once)
+ * Counter is 12 bit.
+ */
+static int at91_wdt_settimeout(unsigned int timeout)
+{
+	unsigned int reg;
+	unsigned int mr;
+
+	/* Check if disabled */
+	mr = at91_sys_read(AT91_WDT_MR);
+	if (mr & AT91_WDT_WDDIS) {
+		printk(KERN_ERR DRV_NAME": sorry, watchdog is disabled\n");
+		return -EIO;
+	}
+
+	/*
+	 * All counting occurs at SLOW_CLOCK / 128 = 256 Hz
+	 *
+	 * Since WDV is a 12-bit counter, the maximum period is
+	 * 4096 / 256 = 16 seconds.
+	 */
+	reg = AT91_WDT_WDRSTEN	/* causes watchdog reset */
+		/* | AT91_WDT_WDRPROC	causes processor reset only */
+		| AT91_WDT_WDDBGHLT	/* disabled in debug mode */
+		| AT91_WDT_WDD		/* restart at any time */
+		| (timeout & AT91_WDT_WDV);  /* timer value */
+	at91_sys_write(AT91_WDT_MR, reg);
+
+	return 0;
+}
+
+static const struct watchdog_info at91_wdt_info = {
+	.identity	= DRV_NAME,
+	.options	= WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING,
+};
+
+/*
+ * Handle commands from user-space.
+ */
+static long at91_wdt_ioctl(struct file *file,
+		unsigned int cmd, unsigned long arg)
+{
+	void __user *argp = (void __user *)arg;
+	int __user *p = argp;
+	int new_value;
+
+	switch (cmd) {
+	case WDIOC_GETSUPPORT:
+		return copy_to_user(argp, &at91_wdt_info,
+				    sizeof(at91_wdt_info)) ? -EFAULT : 0;
+
+	case WDIOC_GETSTATUS:
+	case WDIOC_GETBOOTSTATUS:
+		return put_user(0, p);
+
+	case WDIOC_KEEPALIVE:
+		at91wdt_private.next_heartbeat = jiffies + heartbeat * HZ;
+		return 0;
+
+	case WDIOC_SETTIMEOUT:
+		if (get_user(new_value, p))
+			return -EFAULT;
+
+		heartbeat = new_value;
+		at91wdt_private.next_heartbeat = jiffies + heartbeat * HZ;
+
+		return put_user(new_value, p);  /* return current value */
+
+	case WDIOC_GETTIMEOUT:
+		return put_user(heartbeat, p);
+	}
+	return -ENOTTY;
+}
+
+/*
+ * Pat the watchdog whenever device is written to.
+ */
+static ssize_t at91_wdt_write(struct file *file, const char *data, size_t len,
+      loff_t *ppos)
+{
+	if (!len)
+		return 0;
+
+	/* Scan for magic character */
+	if (!nowayout) {
+		size_t i;
+
+		at91wdt_private.expect_close = 0;
+
+		for (i = 0; i < len; i++) {
+			char c;
+			if (get_user(c, data + i))
+				return -EFAULT;
+			if (c == 'V') {
+				at91wdt_private.expect_close = 42;
+				break;
+			}
+		}
+	}
+
+	at91wdt_private.next_heartbeat = jiffies + heartbeat * HZ;
+
+	return len;
+}
+
+/* ......................................................................... */
+
+static const struct file_operations at91wdt_fops = {
+	.owner			= THIS_MODULE,
+	.llseek			= no_llseek,
+	.unlocked_ioctl	= at91_wdt_ioctl,
+	.open			= at91_wdt_open,
+	.release		= at91_wdt_close,
+	.write			= at91_wdt_write,
+};
+
+static struct miscdevice at91wdt_miscdev = {
+	.minor		= WATCHDOG_MINOR,
+	.name		= "watchdog",
+	.fops		= &at91wdt_fops,
+};
+
+static int __init at91wdt_probe(struct platform_device *pdev)
+{
+	int res;
+
+	if (at91wdt_miscdev.parent)
+		return -EBUSY;
+	at91wdt_miscdev.parent = &pdev->dev;
+
+	/* Set watchdog */
+	res = at91_wdt_settimeout(ms_to_ticks(WDT_HW_TIMEOUT * 1000));
+	if (res)
+		return res;
+
+	res = misc_register(&at91wdt_miscdev);
+	if (res)
+		return res;
+
+	at91wdt_private.next_heartbeat = jiffies + heartbeat * HZ;
+	setup_timer(&at91wdt_private.timer, at91_ping, 0);
+	mod_timer(&at91wdt_private.timer, jiffies + WDT_TIMEOUT);
+
+	printk(KERN_INFO DRV_NAME " enabled (heartbeat=%d sec, nowayout=%d)\n",
+		heartbeat, nowayout);
+
+	return 0;
+}
+
+static int __exit at91wdt_remove(struct platform_device *pdev)
+{
+	int res;
+
+	res = misc_deregister(&at91wdt_miscdev);
+	if (!res)
+		at91wdt_miscdev.parent = NULL;
+
+	return res;
+}
+
+#ifdef CONFIG_PM
+
+static int at91wdt_suspend(struct platform_device *pdev, pm_message_t message)
+{
+	return 0;
+}
+
+static int at91wdt_resume(struct platform_device *pdev)
+{
+	return 0;
+}
+
+#else
+#define at91wdt_suspend	NULL
+#define at91wdt_resume	NULL
+#endif
+
+static struct platform_driver at91wdt_driver = {
+	.remove		= __exit_p(at91wdt_remove),
+	.suspend	= at91wdt_suspend,
+	.resume		= at91wdt_resume,
+	.driver		= {
+		.name	= "at91_wdt",
+		.owner	= THIS_MODULE,
+	},
+};
+
+static int __init at91sam_wdt_init(void)
+{
+	return platform_driver_probe(&at91wdt_driver, at91wdt_probe);
+}
+
+static void __exit at91sam_wdt_exit(void)
+{
+	platform_driver_unregister(&at91wdt_driver);
+}
+
+module_init(at91sam_wdt_init);
+module_exit(at91sam_wdt_exit);
+
+MODULE_AUTHOR("Renaud CERRATO <r.cerrato@til-technologies.fr>");
+MODULE_DESCRIPTION("Watchdog driver for Atmel AT91SAM9x processors");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/cpwd.c b/drivers/watchdog/cpwd.c
new file mode 100644
index 000000000000..084dfe9cecfb
--- /dev/null
+++ b/drivers/watchdog/cpwd.c
@@ -0,0 +1,695 @@
+/* cpwd.c - driver implementation for hardware watchdog
+ * timers found on Sun Microsystems CP1400 and CP1500 boards.
+ *
+ * This device supports both the generic Linux watchdog 
+ * interface and Solaris-compatible ioctls as best it is
+ * able.
+ *
+ * NOTE: 	CP1400 systems appear to have a defective intr_mask
+ * 			register on the PLD, preventing the disabling of
+ * 			timer interrupts.  We use a timer to periodically 
+ * 			reset 'stopped' watchdogs on affected platforms.
+ *
+ * Copyright (c) 2000 Eric Brower (ebrower@usa.net)
+ * Copyright (C) 2008 David S. Miller <davem@davemloft.net>
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/errno.h>
+#include <linux/major.h>
+#include <linux/init.h>
+#include <linux/miscdevice.h>
+#include <linux/interrupt.h>
+#include <linux/ioport.h>
+#include <linux/timer.h>
+#include <linux/smp_lock.h>
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+
+#include <asm/irq.h>
+#include <asm/uaccess.h>
+
+#include <asm/watchdog.h>
+
+#define DRIVER_NAME	"cpwd"
+#define PFX		DRIVER_NAME ": "
+
+#define WD_OBPNAME	"watchdog"
+#define WD_BADMODEL	"SUNW,501-5336"
+#define WD_BTIMEOUT	(jiffies + (HZ * 1000))
+#define WD_BLIMIT	0xFFFF
+
+#define WD0_MINOR	212
+#define WD1_MINOR	213	
+#define WD2_MINOR	214	
+
+/* Internal driver definitions.  */
+#define WD0_ID			0
+#define WD1_ID			1
+#define WD2_ID			2
+#define WD_NUMDEVS		3
+
+#define WD_INTR_OFF		0
+#define WD_INTR_ON		1
+
+#define WD_STAT_INIT	0x01	/* Watchdog timer is initialized	*/
+#define WD_STAT_BSTOP	0x02	/* Watchdog timer is brokenstopped	*/
+#define WD_STAT_SVCD	0x04	/* Watchdog interrupt occurred		*/
+
+/* Register value definitions
+ */
+#define WD0_INTR_MASK	0x01	/* Watchdog device interrupt masks	*/
+#define WD1_INTR_MASK	0x02
+#define WD2_INTR_MASK	0x04
+
+#define WD_S_RUNNING	0x01	/* Watchdog device status running	*/
+#define WD_S_EXPIRED	0x02	/* Watchdog device status expired	*/
+
+struct cpwd {
+	void __iomem	*regs;
+	spinlock_t	lock;
+
+	unsigned int	irq;
+
+	unsigned long	timeout;
+	bool		enabled;
+	bool		reboot;
+	bool		broken;
+	bool		initialized;
+
+	struct {
+		struct miscdevice	misc;
+		void __iomem		*regs;
+		u8			intr_mask;
+		u8			runstatus;
+		u16			timeout;
+	} devs[WD_NUMDEVS];
+};
+
+static struct cpwd *cpwd_device;
+
+/* Sun uses Altera PLD EPF8820ATC144-4 
+ * providing three hardware watchdogs:
+ *
+ * 	1) RIC - sends an interrupt when triggered
+ * 	2) XIR - asserts XIR_B_RESET when triggered, resets CPU
+ * 	3) POR - asserts POR_B_RESET when triggered, resets CPU, backplane, board
+ *
+ *** Timer register block definition (struct wd_timer_regblk)
+ *
+ * dcntr and limit registers (halfword access):      
+ * -------------------
+ * | 15 | ...| 1 | 0 |
+ * -------------------
+ * |-  counter val  -|
+ * -------------------
+ * dcntr - 	Current 16-bit downcounter value.
+ * 			When downcounter reaches '0' watchdog expires.
+ * 			Reading this register resets downcounter with 'limit' value.
+ * limit - 	16-bit countdown value in 1/10th second increments.
+ * 			Writing this register begins countdown with input value.
+ * 			Reading from this register does not affect counter.
+ * NOTES:	After watchdog reset, dcntr and limit contain '1'
+ *
+ * status register (byte access):
+ * ---------------------------
+ * | 7 | ... | 2 |  1  |  0  |
+ * --------------+------------
+ * |-   UNUSED  -| EXP | RUN |
+ * ---------------------------
+ * status-	Bit 0 - Watchdog is running
+ * 			Bit 1 - Watchdog has expired
+ *
+ *** PLD register block definition (struct wd_pld_regblk)
+ *
+ * intr_mask register (byte access):
+ * ---------------------------------
+ * | 7 | ... | 3 |  2  |  1  |  0  |
+ * +-------------+------------------
+ * |-   UNUSED  -| WD3 | WD2 | WD1 |
+ * ---------------------------------
+ * WD3 -  1 == Interrupt disabled for watchdog 3
+ * WD2 -  1 == Interrupt disabled for watchdog 2
+ * WD1 -  1 == Interrupt disabled for watchdog 1
+ *
+ * pld_status register (byte access):
+ * UNKNOWN, MAGICAL MYSTERY REGISTER
+ *
+ */
+#define WD_TIMER_REGSZ	16
+#define WD0_OFF		0
+#define WD1_OFF		(WD_TIMER_REGSZ * 1)
+#define WD2_OFF		(WD_TIMER_REGSZ * 2)
+#define PLD_OFF		(WD_TIMER_REGSZ * 3)
+
+#define WD_DCNTR	0x00
+#define WD_LIMIT	0x04
+#define WD_STATUS	0x08
+
+#define PLD_IMASK	(PLD_OFF + 0x00)
+#define PLD_STATUS	(PLD_OFF + 0x04)
+
+static struct timer_list cpwd_timer;
+
+static int wd0_timeout = 0;
+static int wd1_timeout = 0;
+static int wd2_timeout = 0;
+
+module_param	(wd0_timeout, int, 0);
+MODULE_PARM_DESC(wd0_timeout, "Default watchdog0 timeout in 1/10secs");
+module_param 	(wd1_timeout, int, 0);
+MODULE_PARM_DESC(wd1_timeout, "Default watchdog1 timeout in 1/10secs");
+module_param 	(wd2_timeout, int, 0);
+MODULE_PARM_DESC(wd2_timeout, "Default watchdog2 timeout in 1/10secs");
+
+MODULE_AUTHOR("Eric Brower <ebrower@usa.net>");
+MODULE_DESCRIPTION("Hardware watchdog driver for Sun Microsystems CP1400/1500");
+MODULE_LICENSE("GPL");
+MODULE_SUPPORTED_DEVICE("watchdog");
+
+static void cpwd_writew(u16 val, void __iomem *addr)
+{
+	writew(cpu_to_le16(val), addr);
+}
+static u16 cpwd_readw(void __iomem *addr)
+{
+	u16 val = readw(addr);
+
+	return le16_to_cpu(val);
+}
+
+static void cpwd_writeb(u8 val, void __iomem *addr)
+{
+	writeb(val, addr);
+}
+
+static u8 cpwd_readb(void __iomem *addr)
+{
+	return readb(addr);
+}
+
+/* Enable or disable watchdog interrupts
+ * Because of the CP1400 defect this should only be
+ * called during initialzation or by wd_[start|stop]timer()
+ *
+ * index 	- sub-device index, or -1 for 'all'
+ * enable	- non-zero to enable interrupts, zero to disable
+ */
+static void cpwd_toggleintr(struct cpwd *p, int index, int enable)
+{
+	unsigned char curregs = cpwd_readb(p->regs + PLD_IMASK);
+	unsigned char setregs = 
+		(index == -1) ? 
+		(WD0_INTR_MASK | WD1_INTR_MASK | WD2_INTR_MASK) : 
+		(p->devs[index].intr_mask);
+
+	if (enable == WD_INTR_ON)
+		curregs &= ~setregs;
+	else
+		curregs |= setregs;
+
+	cpwd_writeb(curregs, p->regs + PLD_IMASK);
+}
+
+/* Restarts timer with maximum limit value and
+ * does not unset 'brokenstop' value.
+ */
+static void cpwd_resetbrokentimer(struct cpwd *p, int index)
+{
+	cpwd_toggleintr(p, index, WD_INTR_ON);
+	cpwd_writew(WD_BLIMIT, p->devs[index].regs + WD_LIMIT);
+}
+
+/* Timer method called to reset stopped watchdogs--
+ * because of the PLD bug on CP1400, we cannot mask
+ * interrupts within the PLD so me must continually
+ * reset the timers ad infinitum.
+ */
+static void cpwd_brokentimer(unsigned long data)
+{
+	struct cpwd *p = (struct cpwd *) data;
+	int id, tripped = 0;
+
+	/* kill a running timer instance, in case we
+	 * were called directly instead of by kernel timer
+	 */
+	if (timer_pending(&cpwd_timer))
+		del_timer(&cpwd_timer);
+
+	for (id = 0; id < WD_NUMDEVS; id++) {
+		if (p->devs[id].runstatus & WD_STAT_BSTOP) {
+			++tripped;
+			cpwd_resetbrokentimer(p, id);
+		}
+	}
+
+	if (tripped) {
+		/* there is at least one timer brokenstopped-- reschedule */
+		cpwd_timer.expires = WD_BTIMEOUT;
+		add_timer(&cpwd_timer);
+	}
+}
+
+/* Reset countdown timer with 'limit' value and continue countdown.
+ * This will not start a stopped timer.
+ */
+static void cpwd_pingtimer(struct cpwd *p, int index)
+{
+	if (cpwd_readb(p->devs[index].regs + WD_STATUS) & WD_S_RUNNING)
+		cpwd_readw(p->devs[index].regs + WD_DCNTR);
+}
+
+/* Stop a running watchdog timer-- the timer actually keeps
+ * running, but the interrupt is masked so that no action is
+ * taken upon expiration.
+ */
+static void cpwd_stoptimer(struct cpwd *p, int index)
+{
+	if (cpwd_readb(p->devs[index].regs + WD_STATUS) & WD_S_RUNNING) {
+		cpwd_toggleintr(p, index, WD_INTR_OFF);
+
+		if (p->broken) {
+			p->devs[index].runstatus |= WD_STAT_BSTOP;
+			cpwd_brokentimer((unsigned long) p);
+		}
+	}
+}
+
+/* Start a watchdog timer with the specified limit value
+ * If the watchdog is running, it will be restarted with
+ * the provided limit value.
+ *
+ * This function will enable interrupts on the specified
+ * watchdog.
+ */
+static void cpwd_starttimer(struct cpwd *p, int index)
+{
+	if (p->broken)
+		p->devs[index].runstatus &= ~WD_STAT_BSTOP;
+
+	p->devs[index].runstatus &= ~WD_STAT_SVCD;
+
+	cpwd_writew(p->devs[index].timeout, p->devs[index].regs + WD_LIMIT);
+	cpwd_toggleintr(p, index, WD_INTR_ON);
+}
+
+static int cpwd_getstatus(struct cpwd *p, int index)
+{
+	unsigned char stat = cpwd_readb(p->devs[index].regs + WD_STATUS);
+	unsigned char intr = cpwd_readb(p->devs[index].regs + PLD_IMASK);
+	unsigned char ret  = WD_STOPPED;
+
+	/* determine STOPPED */
+	if (!stat) 
+		return ret;
+
+	/* determine EXPIRED vs FREERUN vs RUNNING */
+	else if (WD_S_EXPIRED & stat) {
+		ret = WD_EXPIRED;
+	} else if(WD_S_RUNNING & stat) {
+		if (intr & p->devs[index].intr_mask) {
+			ret = WD_FREERUN;
+		} else {
+			/* Fudge WD_EXPIRED status for defective CP1400--
+			 * IF timer is running 
+			 * 	AND brokenstop is set 
+			 * 	AND an interrupt has been serviced
+			 * we are WD_EXPIRED.
+			 *
+			 * IF timer is running 
+			 * 	AND brokenstop is set 
+			 * 	AND no interrupt has been serviced
+			 * we are WD_FREERUN.
+			 */
+			if (p->broken &&
+			    (p->devs[index].runstatus & WD_STAT_BSTOP)) {
+				if (p->devs[index].runstatus & WD_STAT_SVCD) {
+					ret = WD_EXPIRED;
+				} else {
+					/* we could as well pretend we are expired */
+					ret = WD_FREERUN;
+				}
+			} else {
+				ret = WD_RUNNING;
+			}
+		}
+	}
+
+	/* determine SERVICED */
+	if (p->devs[index].runstatus & WD_STAT_SVCD)
+		ret |= WD_SERVICED;
+
+	return(ret);
+}
+
+static irqreturn_t cpwd_interrupt(int irq, void *dev_id)
+{
+	struct cpwd *p = dev_id;
+
+	/* Only WD0 will interrupt-- others are NMI and we won't
+	 * see them here....
+	 */
+	spin_lock_irq(&p->lock);
+
+	cpwd_stoptimer(p, WD0_ID);
+	p->devs[WD0_ID].runstatus |=  WD_STAT_SVCD;
+
+	spin_unlock_irq(&p->lock);
+
+	return IRQ_HANDLED;
+}
+
+static int cpwd_open(struct inode *inode, struct file *f)
+{
+	struct cpwd *p = cpwd_device;
+
+	lock_kernel();
+	switch(iminor(inode)) {
+		case WD0_MINOR:
+		case WD1_MINOR:
+		case WD2_MINOR:
+			break;
+
+		default:
+			unlock_kernel();
+			return -ENODEV;
+	}
+
+	/* Register IRQ on first open of device */
+	if (!p->initialized) {
+		if (request_irq(p->irq, &cpwd_interrupt, 
+				IRQF_SHARED, DRIVER_NAME, p)) {
+			printk(KERN_ERR PFX "Cannot register IRQ %d\n", 
+				p->irq);
+			unlock_kernel();
+			return -EBUSY;
+		}
+		p->initialized = true;
+	}
+
+	unlock_kernel();
+
+	return nonseekable_open(inode, f);
+}
+
+static int cpwd_release(struct inode *inode, struct file *file)
+{
+	return 0;
+}
+
+static int cpwd_ioctl(struct inode *inode, struct file *file, 
+		      unsigned int cmd, unsigned long arg)
+{
+	static struct watchdog_info info = {
+		.options		= WDIOF_SETTIMEOUT,
+		.firmware_version	= 1,
+		.identity		= DRIVER_NAME,
+	};
+	void __user *argp = (void __user *)arg;
+	int index = iminor(inode) - WD0_MINOR;
+	struct cpwd *p = cpwd_device;
+	int setopt = 0;
+
+	switch (cmd) {
+	/* Generic Linux IOCTLs */
+	case WDIOC_GETSUPPORT:
+		if (copy_to_user(argp, &info, sizeof(struct watchdog_info)))
+			return -EFAULT;
+		break;
+
+	case WDIOC_GETSTATUS:
+	case WDIOC_GETBOOTSTATUS:
+		if (put_user(0, (int __user *)argp))
+			return -EFAULT;
+		break;
+
+	case WDIOC_KEEPALIVE:
+		cpwd_pingtimer(p, index);
+		break;
+
+	case WDIOC_SETOPTIONS:
+		if (copy_from_user(&setopt, argp, sizeof(unsigned int)))
+			return -EFAULT;
+
+		if (setopt & WDIOS_DISABLECARD) {
+			if (p->enabled)
+				return -EINVAL;
+			cpwd_stoptimer(p, index);
+		} else if (setopt & WDIOS_ENABLECARD) {
+			cpwd_starttimer(p, index);
+		} else {
+			return -EINVAL;
+		}	
+		break;
+
+	/* Solaris-compatible IOCTLs */
+	case WIOCGSTAT:
+		setopt = cpwd_getstatus(p, index);
+		if (copy_to_user(argp, &setopt, sizeof(unsigned int)))
+			return -EFAULT;
+		break;
+
+	case WIOCSTART:
+		cpwd_starttimer(p, index);
+		break;
+
+	case WIOCSTOP:
+		if (p->enabled)
+			return(-EINVAL);
+
+		cpwd_stoptimer(p, index);
+		break;
+
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static long cpwd_compat_ioctl(struct file *file, unsigned int cmd,
+			      unsigned long arg)
+{
+	int rval = -ENOIOCTLCMD;
+
+	switch (cmd) {
+	/* solaris ioctls are specific to this driver */
+	case WIOCSTART:
+	case WIOCSTOP:
+	case WIOCGSTAT:
+		lock_kernel();
+		rval = cpwd_ioctl(file->f_path.dentry->d_inode, file, cmd, arg);
+		unlock_kernel();
+		break;
+
+	/* everything else is handled by the generic compat layer */
+	default:
+		break;
+	}
+
+	return rval;
+}
+
+static ssize_t cpwd_write(struct file *file, const char __user *buf, 
+			  size_t count, loff_t *ppos)
+{
+	struct inode *inode = file->f_path.dentry->d_inode;
+	struct cpwd *p = cpwd_device;
+	int index = iminor(inode);
+
+	if (count) {
+		cpwd_pingtimer(p, index);
+		return 1;
+	}
+
+	return 0;
+}
+
+static ssize_t cpwd_read(struct file * file, char __user *buffer,
+			 size_t count, loff_t *ppos)
+{
+	return -EINVAL;
+}
+
+static const struct file_operations cpwd_fops = {
+	.owner =	THIS_MODULE,
+	.ioctl =	cpwd_ioctl,
+	.compat_ioctl =	cpwd_compat_ioctl,
+	.open =		cpwd_open,
+	.write =	cpwd_write,
+	.read =		cpwd_read,
+	.release =	cpwd_release,
+};
+
+static int __devinit cpwd_probe(struct of_device *op,
+				const struct of_device_id *match)
+{
+	struct device_node *options;
+	const char *str_prop;
+	const void *prop_val;
+	int i, err = -EINVAL;
+	struct cpwd *p;
+
+	if (cpwd_device)
+		return -EINVAL;
+
+	p = kzalloc(sizeof(*p), GFP_KERNEL);
+	err = -ENOMEM;
+	if (!p) {
+		printk(KERN_ERR PFX "Unable to allocate struct cpwd.\n");
+		goto out;
+	}
+
+	p->irq = op->irqs[0];
+
+	spin_lock_init(&p->lock);
+
+	p->regs = of_ioremap(&op->resource[0], 0,
+			     4 * WD_TIMER_REGSZ, DRIVER_NAME);
+	if (!p->regs) {
+		printk(KERN_ERR PFX "Unable to map registers.\n");
+		goto out_free;
+	}
+
+	options = of_find_node_by_path("/options");
+	err = -ENODEV;
+	if (!options) {
+		printk(KERN_ERR PFX "Unable to find /options node.\n");
+		goto out_iounmap;
+	}
+
+	prop_val = of_get_property(options, "watchdog-enable?", NULL);
+	p->enabled = (prop_val ? true : false);
+
+	prop_val = of_get_property(options, "watchdog-reboot?", NULL);
+	p->reboot = (prop_val ? true : false);
+
+	str_prop = of_get_property(options, "watchdog-timeout", NULL);
+	if (str_prop)
+		p->timeout = simple_strtoul(str_prop, NULL, 10);
+
+	/* CP1400s seem to have broken PLD implementations-- the
+	 * interrupt_mask register cannot be written, so no timer
+	 * interrupts can be masked within the PLD.
+	 */
+	str_prop = of_get_property(op->node, "model", NULL);
+	p->broken = (str_prop && !strcmp(str_prop, WD_BADMODEL));
+
+	if (!p->enabled)
+		cpwd_toggleintr(p, -1, WD_INTR_OFF);
+
+	for (i = 0; i < WD_NUMDEVS; i++) {
+		static const char *cpwd_names[] = { "RIC", "XIR", "POR" };
+		static int *parms[] = { &wd0_timeout,
+					&wd1_timeout,
+					&wd2_timeout };
+		struct miscdevice *mp = &p->devs[i].misc;
+
+		mp->minor = WD0_MINOR + i;
+		mp->name = cpwd_names[i];
+		mp->fops = &cpwd_fops;
+
+		p->devs[i].regs = p->regs + (i * WD_TIMER_REGSZ);
+		p->devs[i].intr_mask = (WD0_INTR_MASK << i);
+		p->devs[i].runstatus &= ~WD_STAT_BSTOP;
+		p->devs[i].runstatus |= WD_STAT_INIT;
+		p->devs[i].timeout = p->timeout;
+		if (*parms[i])
+			p->devs[i].timeout = *parms[i];
+
+		err = misc_register(&p->devs[i].misc);
+		if (err) {
+			printk(KERN_ERR "Could not register misc device for "
+			       "dev %d\n", i);
+			goto out_unregister;
+		}
+	}
+
+	if (p->broken) {
+		init_timer(&cpwd_timer);
+		cpwd_timer.function 	= cpwd_brokentimer;
+		cpwd_timer.data		= (unsigned long) p;
+		cpwd_timer.expires	= WD_BTIMEOUT;
+
+		printk(KERN_INFO PFX "PLD defect workaround enabled for "
+		       "model " WD_BADMODEL ".\n");
+	}
+
+	dev_set_drvdata(&op->dev, p);
+	cpwd_device = p;
+	err = 0;
+
+out:
+	return err;
+
+out_unregister:
+	for (i--; i >= 0; i--)
+		misc_deregister(&p->devs[i].misc);
+
+out_iounmap:
+	of_iounmap(&op->resource[0], p->regs, 4 * WD_TIMER_REGSZ);
+
+out_free:
+	kfree(p);
+	goto out;
+}
+
+static int __devexit cpwd_remove(struct of_device *op)
+{
+	struct cpwd *p = dev_get_drvdata(&op->dev);
+	int i;
+
+	for (i = 0; i < 4; i++) {
+		misc_deregister(&p->devs[i].misc);
+
+		if (!p->enabled) {
+			cpwd_stoptimer(p, i);
+			if (p->devs[i].runstatus & WD_STAT_BSTOP)
+				cpwd_resetbrokentimer(p, i);
+		}
+	}
+
+	if (p->broken)
+		del_timer_sync(&cpwd_timer);
+
+	if (p->initialized)
+		free_irq(p->irq, p);
+
+	of_iounmap(&op->resource[0], p->regs, 4 * WD_TIMER_REGSZ);
+	kfree(p);
+
+	cpwd_device = NULL;
+
+	return 0;
+}
+
+static const struct of_device_id cpwd_match[] = {
+	{
+		.name = "watchdog",
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, cpwd_match);
+
+static struct of_platform_driver cpwd_driver = {
+	.name		= DRIVER_NAME,
+	.match_table	= cpwd_match,
+	.probe		= cpwd_probe,
+	.remove		= __devexit_p(cpwd_remove),
+};
+
+static int __init cpwd_init(void)
+{
+	return of_register_driver(&cpwd_driver, &of_bus_type);
+}
+
+static void __exit cpwd_exit(void)
+{
+	of_unregister_driver(&cpwd_driver);
+}
+
+module_init(cpwd_init);
+module_exit(cpwd_exit);
diff --git a/drivers/watchdog/it87_wdt.c b/drivers/watchdog/it87_wdt.c
new file mode 100644
index 000000000000..afb8af397a9f
--- /dev/null
+++ b/drivers/watchdog/it87_wdt.c
@@ -0,0 +1,725 @@
+/*
+ *	Watchdog Timer Driver
+ *	   for ITE IT87xx Environment Control - Low Pin Count Input / Output
+ *
+ *	(c) Copyright 2007  Oliver Schuster <olivers137@aol.com>
+ *
+ *	Based on softdog.c	by Alan Cox,
+ *		 83977f_wdt.c	by Jose Goncalves,
+ *		 it87.c		by Chris Gauthron, Jean Delvare
+ *
+ *	Data-sheets: Publicly available at the ITE website
+ *		    http://www.ite.com.tw/
+ *
+ *	Support of the watchdog timers, which are available on
+ *	IT8716, IT8718, IT8726 and IT8712 (J,K version).
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License
+ *	as published by the Free Software Foundation; either version
+ *	2 of the License, or (at your option) any later version.
+ *
+ *	This program is distributed in the hope that it will be useful,
+ *	but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *	MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *	GNU General Public License for more details.
+ *
+ *	You should have received a copy of the GNU General Public License
+ *	along with this program; if not, write to the Free Software
+ *	Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/fs.h>
+#include <linux/miscdevice.h>
+#include <linux/init.h>
+#include <linux/ioport.h>
+#include <linux/watchdog.h>
+#include <linux/notifier.h>
+#include <linux/reboot.h>
+#include <linux/uaccess.h>
+#include <linux/io.h>
+
+#include <asm/system.h>
+
+#define WATCHDOG_VERSION	"1.12"
+#define WATCHDOG_NAME		"IT87 WDT"
+#define PFX			WATCHDOG_NAME ": "
+#define DRIVER_VERSION		WATCHDOG_NAME " driver, v" WATCHDOG_VERSION "\n"
+#define WD_MAGIC		'V'
+
+/* Defaults for Module Parameter */
+#define DEFAULT_NOGAMEPORT	0
+#define DEFAULT_EXCLUSIVE	1
+#define DEFAULT_TIMEOUT 	60
+#define DEFAULT_TESTMODE	0
+#define DEFAULT_NOWAYOUT	WATCHDOG_NOWAYOUT
+
+/* IO Ports */
+#define REG		0x2e
+#define VAL		0x2f
+
+/* Logical device Numbers LDN */
+#define GPIO		0x07
+#define GAMEPORT	0x09
+#define CIR		0x0a
+
+/* Configuration Registers and Functions */
+#define LDNREG		0x07
+#define CHIPID		0x20
+#define CHIPREV 	0x22
+#define ACTREG		0x30
+#define BASEREG 	0x60
+
+/* Chip Id numbers */
+#define NO_DEV_ID	0xffff
+#define IT8705_ID	0x8705
+#define IT8712_ID	0x8712
+#define IT8716_ID	0x8716
+#define IT8718_ID	0x8718
+#define IT8726_ID	0x8726	/* the data sheet suggest wrongly 0x8716 */
+
+/* GPIO Configuration Registers LDN=0x07 */
+#define WDTCTRL 	0x71
+#define WDTCFG		0x72
+#define WDTVALLSB	0x73
+#define WDTVALMSB	0x74
+
+/* GPIO Bits WDTCTRL */
+#define WDT_CIRINT	0x80
+#define WDT_MOUSEINT	0x40
+#define WDT_KYBINT	0x20
+#define WDT_GAMEPORT	0x10 /* not it8718 */
+#define WDT_FORCE	0x02
+#define WDT_ZERO	0x01
+
+/* GPIO Bits WDTCFG */
+#define WDT_TOV1	0x80
+#define WDT_KRST	0x40
+#define WDT_TOVE	0x20
+#define WDT_PWROK	0x10
+#define WDT_INT_MASK	0x0f
+
+/* CIR Configuration Register LDN=0x0a */
+#define CIR_ILS 	0x70
+
+/* The default Base address is not always available, we use this */
+#define CIR_BASE	0x0208
+
+/* CIR Controller */
+#define CIR_DR(b)	(b)
+#define CIR_IER(b)	(b + 1)
+#define CIR_RCR(b)	(b + 2)
+#define CIR_TCR1(b)	(b + 3)
+#define CIR_TCR2(b)	(b + 4)
+#define CIR_TSR(b)	(b + 5)
+#define CIR_RSR(b)	(b + 6)
+#define CIR_BDLR(b)	(b + 5)
+#define CIR_BDHR(b)	(b + 6)
+#define CIR_IIR(b)	(b + 7)
+
+/* Default Base address of Game port */
+#define GP_BASE_DEFAULT	0x0201
+
+/* wdt_status */
+#define WDTS_TIMER_RUN	0
+#define WDTS_DEV_OPEN	1
+#define WDTS_KEEPALIVE	2
+#define WDTS_LOCKED	3
+#define WDTS_USE_GP	4
+#define WDTS_EXPECTED	5
+
+static	unsigned int base, gpact, ciract;
+static	unsigned long wdt_status;
+static	DEFINE_SPINLOCK(spinlock);
+
+static	int nogameport = DEFAULT_NOGAMEPORT;
+static	int exclusive  = DEFAULT_EXCLUSIVE;
+static	int timeout    = DEFAULT_TIMEOUT;
+static	int testmode   = DEFAULT_TESTMODE;
+static	int nowayout   = DEFAULT_NOWAYOUT;
+
+module_param(nogameport, int, 0);
+MODULE_PARM_DESC(nogameport, "Forbid the activation of game port, default="
+		__MODULE_STRING(DEFAULT_NOGAMEPORT));
+module_param(exclusive, int, 0);
+MODULE_PARM_DESC(exclusive, "Watchdog exclusive device open, default="
+		__MODULE_STRING(DEFAULT_EXCLUSIVE));
+module_param(timeout, int, 0);
+MODULE_PARM_DESC(timeout, "Watchdog timeout in seconds, default="
+		__MODULE_STRING(DEFAULT_TIMEOUT));
+module_param(testmode, int, 0);
+MODULE_PARM_DESC(testmode, "Watchdog test mode (1 = no reboot), default="
+		__MODULE_STRING(DEFAULT_TESTMODE));
+module_param(nowayout, int, 0);
+MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started, default="
+		__MODULE_STRING(WATCHDOG_NOWAYOUT));
+
+/* Superio Chip */
+
+static inline void superio_enter(void)
+{
+	outb(0x87, REG);
+	outb(0x01, REG);
+	outb(0x55, REG);
+	outb(0x55, REG);
+}
+
+static inline void superio_exit(void)
+{
+	outb(0x02, REG);
+	outb(0x02, VAL);
+}
+
+static inline void superio_select(int ldn)
+{
+	outb(LDNREG, REG);
+	outb(ldn, VAL);
+}
+
+static inline int superio_inb(int reg)
+{
+	outb(reg, REG);
+	return inb(VAL);
+}
+
+static inline void superio_outb(int val, int reg)
+{
+       outb(reg, REG);
+       outb(val, VAL);
+}
+
+static inline int superio_inw(int reg)
+{
+	int val;
+	outb(reg++, REG);
+	val = inb(VAL) << 8;
+	outb(reg, REG);
+	val |= inb(VAL);
+	return val;
+}
+
+static inline void superio_outw(int val, int reg)
+{
+       outb(reg++, REG);
+       outb(val >> 8, VAL);
+       outb(reg, REG);
+       outb(val, VAL);
+}
+
+/* watchdog timer handling */
+
+static void wdt_keepalive(void)
+{
+	if (test_bit(WDTS_USE_GP, &wdt_status))
+		inb(base);
+	else
+		/* The timer reloads with around 5 msec delay */
+		outb(0x55, CIR_DR(base));
+	set_bit(WDTS_KEEPALIVE, &wdt_status);
+}
+
+static void wdt_start(void)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&spinlock, flags);
+	superio_enter();
+
+	superio_select(GPIO);
+	if (test_bit(WDTS_USE_GP, &wdt_status))
+		superio_outb(WDT_GAMEPORT, WDTCTRL);
+	else
+		superio_outb(WDT_CIRINT, WDTCTRL);
+	if (!testmode)
+		superio_outb(WDT_TOV1 | WDT_KRST | WDT_PWROK, WDTCFG);
+	else
+		superio_outb(WDT_TOV1, WDTCFG);
+	superio_outb(timeout>>8, WDTVALMSB);
+	superio_outb(timeout, WDTVALLSB);
+
+	superio_exit();
+	spin_unlock_irqrestore(&spinlock, flags);
+}
+
+static void wdt_stop(void)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&spinlock, flags);
+	superio_enter();
+
+	superio_select(GPIO);
+	superio_outb(0x00, WDTCTRL);
+	superio_outb(WDT_TOV1, WDTCFG);
+	superio_outb(0x00, WDTVALMSB);
+	superio_outb(0x00, WDTVALLSB);
+
+	superio_exit();
+	spin_unlock_irqrestore(&spinlock, flags);
+}
+
+/**
+ *	wdt_set_timeout - set a new timeout value with watchdog ioctl
+ *	@t: timeout value in seconds
+ *
+ *	The hardware device has a 16 bit watchdog timer, thus the
+ *	timeout time ranges between 1 and 65535 seconds.
+ *
+ *	Used within WDIOC_SETTIMEOUT watchdog device ioctl.
+ */
+
+static int wdt_set_timeout(int t)
+{
+	unsigned long flags;
+
+	if (t < 1 || t > 65535)
+		return -EINVAL;
+
+	timeout = t;
+
+	spin_lock_irqsave(&spinlock, flags);
+	if (test_bit(WDTS_TIMER_RUN, &wdt_status)) {
+		superio_enter();
+
+		superio_select(GPIO);
+		superio_outb(t>>8, WDTVALMSB);
+		superio_outb(t, WDTVALLSB);
+
+		superio_exit();
+	}
+	spin_unlock_irqrestore(&spinlock, flags);
+	return 0;
+}
+
+/**
+ *	wdt_get_status - determines the status supported by watchdog ioctl
+ *	@status: status returned to user space
+ *
+ *	The status bit of the device does not allow to distinguish
+ *	between a regular system reset and a watchdog forced reset.
+ *	But, in test mode it is useful, so it is supported through
+ *	WDIOC_GETSTATUS watchdog ioctl. Additionally the driver
+ *	reports the keepalive signal and the acception of the magic.
+ *
+ *	Used within WDIOC_GETSTATUS watchdog device ioctl.
+ */
+
+static int wdt_get_status(int *status)
+{
+	unsigned long flags;
+
+	*status = 0;
+	if (testmode) {
+		spin_lock_irqsave(&spinlock, flags);
+		superio_enter();
+		superio_select(GPIO);
+		if (superio_inb(WDTCTRL) & WDT_ZERO) {
+			superio_outb(0x00, WDTCTRL);
+			clear_bit(WDTS_TIMER_RUN, &wdt_status);
+			*status |= WDIOF_CARDRESET;
+		}
+
+		superio_exit();
+		spin_unlock_irqrestore(&spinlock, flags);
+	}
+	if (test_and_clear_bit(WDTS_KEEPALIVE, &wdt_status))
+		*status |= WDIOF_KEEPALIVEPING;
+	if (test_bit(WDTS_EXPECTED, &wdt_status))
+		*status |= WDIOF_MAGICCLOSE;
+	return 0;
+}
+
+/* /dev/watchdog handling */
+
+/**
+ *	wdt_open - watchdog file_operations .open
+ *	@inode: inode of the device
+ *	@file: file handle to the device
+ *
+ *	The watchdog timer starts by opening the device.
+ *
+ *	Used within the file operation of the watchdog device.
+ */
+
+static int wdt_open(struct inode *inode, struct file *file)
+{
+	if (exclusive && test_and_set_bit(WDTS_DEV_OPEN, &wdt_status))
+		return -EBUSY;
+	if (!test_and_set_bit(WDTS_TIMER_RUN, &wdt_status)) {
+		if (nowayout && !test_and_set_bit(WDTS_LOCKED, &wdt_status))
+			__module_get(THIS_MODULE);
+		wdt_start();
+	}
+	return nonseekable_open(inode, file);
+}
+
+/**
+ *	wdt_release - watchdog file_operations .release
+ *	@inode: inode of the device
+ *	@file: file handle to the device
+ *
+ *	Closing the watchdog device either stops the watchdog timer
+ *	or in the case, that nowayout is set or the magic character
+ *	wasn't written, a critical warning about an running watchdog
+ *	timer is given.
+ *
+ *	Used within the file operation of the watchdog device.
+ */
+
+static int wdt_release(struct inode *inode, struct file *file)
+{
+	if (test_bit(WDTS_TIMER_RUN, &wdt_status)) {
+		if (test_and_clear_bit(WDTS_EXPECTED, &wdt_status)) {
+			wdt_stop();
+			clear_bit(WDTS_TIMER_RUN, &wdt_status);
+		} else {
+			wdt_keepalive();
+			printk(KERN_CRIT PFX
+			       "unexpected close, not stopping watchdog!\n");
+		}
+	}
+	clear_bit(WDTS_DEV_OPEN, &wdt_status);
+	return 0;
+}
+
+/**
+ *	wdt_write - watchdog file_operations .write
+ *	@file: file handle to the watchdog
+ *	@buf: buffer to write
+ *	@count: count of bytes
+ *	@ppos: pointer to the position to write. No seeks allowed
+ *
+ *	A write to a watchdog device is defined as a keepalive signal. Any
+ *	write of data will do, as we don't define content meaning.
+ *
+ *	Used within the file operation of the watchdog device.
+ */
+
+static ssize_t wdt_write(struct file *file, const char __user *buf,
+			    size_t count, loff_t *ppos)
+{
+	if (count) {
+		clear_bit(WDTS_EXPECTED, &wdt_status);
+		wdt_keepalive();
+	}
+	if (!nowayout) {
+		size_t ofs;
+
+	/* note: just in case someone wrote the magic character long ago */
+		for (ofs = 0; ofs != count; ofs++) {
+			char c;
+			if (get_user(c, buf + ofs))
+				return -EFAULT;
+			if (c == WD_MAGIC)
+				set_bit(WDTS_EXPECTED, &wdt_status);
+		}
+	}
+	return count;
+}
+
+static struct watchdog_info ident = {
+	.options = WDIOF_SETTIMEOUT | WDIOF_MAGICCLOSE | WDIOF_KEEPALIVEPING,
+	.firmware_version =	1,
+	.identity = WATCHDOG_NAME,
+};
+
+/**
+ *	wdt_ioctl - watchdog file_operations .unlocked_ioctl
+ *	@file: file handle to the device
+ *	@cmd: watchdog command
+ *	@arg: argument pointer
+ *
+ *	The watchdog API defines a common set of functions for all watchdogs
+ *	according to their available features.
+ *
+ *	Used within the file operation of the watchdog device.
+ */
+
+static long wdt_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	int rc = 0, status, new_options, new_timeout;
+	union {
+		struct watchdog_info __user *ident;
+		int __user *i;
+	} uarg;
+
+	uarg.i = (int __user *)arg;
+
+	switch (cmd) {
+	case WDIOC_GETSUPPORT:
+		return copy_to_user(uarg.ident,
+				    &ident, sizeof(ident)) ? -EFAULT : 0;
+
+	case WDIOC_GETSTATUS:
+		wdt_get_status(&status);
+		return put_user(status, uarg.i);
+
+	case WDIOC_GETBOOTSTATUS:
+		return put_user(0, uarg.i);
+
+	case WDIOC_KEEPALIVE:
+		wdt_keepalive();
+		return 0;
+
+	case WDIOC_SETOPTIONS:
+		if (get_user(new_options, uarg.i))
+			return -EFAULT;
+
+		switch (new_options) {
+		case WDIOS_DISABLECARD:
+			if (test_bit(WDTS_TIMER_RUN, &wdt_status))
+				wdt_stop();
+			clear_bit(WDTS_TIMER_RUN, &wdt_status);
+			return 0;
+
+		case WDIOS_ENABLECARD:
+			if (!test_and_set_bit(WDTS_TIMER_RUN, &wdt_status))
+				wdt_start();
+			return 0;
+
+		default:
+			return -EFAULT;
+		}
+
+	case WDIOC_SETTIMEOUT:
+		if (get_user(new_timeout, uarg.i))
+			return -EFAULT;
+		rc = wdt_set_timeout(new_timeout);
+	case WDIOC_GETTIMEOUT:
+		if (put_user(timeout, uarg.i))
+			return -EFAULT;
+		return rc;
+
+	default:
+		return -ENOTTY;
+	}
+}
+
+static int wdt_notify_sys(struct notifier_block *this, unsigned long code,
+	void *unused)
+{
+	if (code == SYS_DOWN || code == SYS_HALT)
+		wdt_stop();
+	return NOTIFY_DONE;
+}
+
+static const struct file_operations wdt_fops = {
+	.owner		= THIS_MODULE,
+	.llseek		= no_llseek,
+	.write		= wdt_write,
+	.unlocked_ioctl	= wdt_ioctl,
+	.open		= wdt_open,
+	.release	= wdt_release,
+};
+
+static struct miscdevice wdt_miscdev = {
+	.minor		= WATCHDOG_MINOR,
+	.name		= "watchdog",
+	.fops		= &wdt_fops,
+};
+
+static struct notifier_block wdt_notifier = {
+	.notifier_call = wdt_notify_sys,
+};
+
+static int __init it87_wdt_init(void)
+{
+	int rc = 0;
+	u16 chip_type;
+	u8  chip_rev;
+	unsigned long flags;
+
+	spin_lock_irqsave(&spinlock, flags);
+	superio_enter();
+	chip_type = superio_inw(CHIPID);
+	chip_rev  = superio_inb(CHIPREV) & 0x0f;
+	superio_exit();
+	spin_unlock_irqrestore(&spinlock, flags);
+
+	switch (chip_type) {
+	case IT8716_ID:
+	case IT8718_ID:
+	case IT8726_ID:
+		break;
+	case IT8712_ID:
+		if (chip_rev > 7)
+			break;
+	case IT8705_ID:
+		printk(KERN_ERR PFX
+		       "Unsupported Chip found, Chip %04x Revision %02x\n",
+		       chip_type, chip_rev);
+		return -ENODEV;
+	case NO_DEV_ID:
+		printk(KERN_ERR PFX "no device\n");
+		return -ENODEV;
+	default:
+		printk(KERN_ERR PFX
+		       "Unknown Chip found, Chip %04x Revision %04x\n",
+		       chip_type, chip_rev);
+		return -ENODEV;
+	}
+
+	spin_lock_irqsave(&spinlock, flags);
+	superio_enter();
+
+	superio_select(GPIO);
+	superio_outb(WDT_TOV1, WDTCFG);
+	superio_outb(0x00, WDTCTRL);
+
+	/* First try to get Gameport support */
+	if (chip_type != IT8718_ID && !nogameport) {
+		superio_select(GAMEPORT);
+		base = superio_inw(BASEREG);
+		if (!base) {
+			base = GP_BASE_DEFAULT;
+			superio_outw(base, BASEREG);
+		}
+		gpact = superio_inb(ACTREG);
+		superio_outb(0x01, ACTREG);
+		superio_exit();
+		spin_unlock_irqrestore(&spinlock, flags);
+		if (request_region(base, 1, WATCHDOG_NAME))
+			set_bit(WDTS_USE_GP, &wdt_status);
+		else
+			rc = -EIO;
+	} else {
+		superio_exit();
+		spin_unlock_irqrestore(&spinlock, flags);
+	}
+
+	/* If we haven't Gameport support, try to get CIR support */
+	if (!test_bit(WDTS_USE_GP, &wdt_status)) {
+		if (!request_region(CIR_BASE, 8, WATCHDOG_NAME)) {
+			if (rc == -EIO)
+				printk(KERN_ERR PFX
+					"I/O Address 0x%04x and 0x%04x"
+					" already in use\n", base, CIR_BASE);
+			else
+				printk(KERN_ERR PFX
+					"I/O Address 0x%04x already in use\n",
+					CIR_BASE);
+			rc = -EIO;
+			goto err_out;
+		}
+		base = CIR_BASE;
+		spin_lock_irqsave(&spinlock, flags);
+		superio_enter();
+
+		superio_select(CIR);
+		superio_outw(base, BASEREG);
+		superio_outb(0x00, CIR_ILS);
+		ciract = superio_inb(ACTREG);
+		superio_outb(0x01, ACTREG);
+		if (rc == -EIO) {
+			superio_select(GAMEPORT);
+			superio_outb(gpact, ACTREG);
+		}
+
+		superio_exit();
+		spin_unlock_irqrestore(&spinlock, flags);
+	}
+
+	if (timeout < 1 || timeout > 65535) {
+		timeout = DEFAULT_TIMEOUT;
+		printk(KERN_WARNING PFX
+		       "Timeout value out of range, use default %d sec\n",
+		       DEFAULT_TIMEOUT);
+	}
+
+	rc = register_reboot_notifier(&wdt_notifier);
+	if (rc) {
+		printk(KERN_ERR PFX
+		       "Cannot register reboot notifier (err=%d)\n", rc);
+		goto err_out_region;
+	}
+
+	rc = misc_register(&wdt_miscdev);
+	if (rc) {
+		printk(KERN_ERR PFX
+		       "Cannot register miscdev on minor=%d (err=%d)\n",
+			wdt_miscdev.minor, rc);
+		goto err_out_reboot;
+	}
+
+	/* Initialize CIR to use it as keepalive source */
+	if (!test_bit(WDTS_USE_GP, &wdt_status)) {
+		outb(0x00, CIR_RCR(base));
+		outb(0xc0, CIR_TCR1(base));
+		outb(0x5c, CIR_TCR2(base));
+		outb(0x10, CIR_IER(base));
+		outb(0x00, CIR_BDHR(base));
+		outb(0x01, CIR_BDLR(base));
+		outb(0x09, CIR_IER(base));
+	}
+
+	printk(KERN_INFO PFX "Chip it%04x revision %d initialized. "
+		"timeout=%d sec (nowayout=%d testmode=%d exclusive=%d "
+		"nogameport=%d)\n", chip_type, chip_rev, timeout,
+		nowayout, testmode, exclusive, nogameport);
+
+	return 0;
+
+err_out_reboot:
+	unregister_reboot_notifier(&wdt_notifier);
+err_out_region:
+	release_region(base, test_bit(WDTS_USE_GP, &wdt_status) ? 1 : 8);
+	if (!test_bit(WDTS_USE_GP, &wdt_status)) {
+		spin_lock_irqsave(&spinlock, flags);
+		superio_enter();
+		superio_select(CIR);
+		superio_outb(ciract, ACTREG);
+		superio_exit();
+		spin_unlock_irqrestore(&spinlock, flags);
+	}
+err_out:
+	if (chip_type != IT8718_ID && !nogameport) {
+		spin_lock_irqsave(&spinlock, flags);
+		superio_enter();
+		superio_select(GAMEPORT);
+		superio_outb(gpact, ACTREG);
+		superio_exit();
+		spin_unlock_irqrestore(&spinlock, flags);
+	}
+
+	return rc;
+}
+
+static void __exit it87_wdt_exit(void)
+{
+	unsigned long flags;
+	int nolock;
+
+	nolock = !spin_trylock_irqsave(&spinlock, flags);
+	superio_enter();
+	superio_select(GPIO);
+	superio_outb(0x00, WDTCTRL);
+	superio_outb(0x00, WDTCFG);
+	superio_outb(0x00, WDTVALMSB);
+	superio_outb(0x00, WDTVALLSB);
+	if (test_bit(WDTS_USE_GP, &wdt_status)) {
+		superio_select(GAMEPORT);
+		superio_outb(gpact, ACTREG);
+	} else {
+		superio_select(CIR);
+		superio_outb(ciract, ACTREG);
+	}
+	superio_exit();
+	if (!nolock)
+		spin_unlock_irqrestore(&spinlock, flags);
+
+	misc_deregister(&wdt_miscdev);
+	unregister_reboot_notifier(&wdt_notifier);
+	release_region(base, test_bit(WDTS_USE_GP, &wdt_status) ? 1 : 8);
+}
+
+module_init(it87_wdt_init);
+module_exit(it87_wdt_exit);
+
+MODULE_AUTHOR("Oliver Schuster");
+MODULE_DESCRIPTION("Hardware Watchdog Device Driver for IT87xx EC-LPC I/O");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/watchdog/omap_wdt.c b/drivers/watchdog/omap_wdt.c
index 3a11dadfd8e7..7bcbb7f4745f 100644
--- a/drivers/watchdog/omap_wdt.c
+++ b/drivers/watchdog/omap_wdt.c
@@ -1,7 +1,7 @@
 /*
- * linux/drivers/char/watchdog/omap_wdt.c
+ * omap_wdt.c
  *
- * Watchdog driver for the TI OMAP 16xx & 24xx 32KHz (non-secure) watchdog
+ * Watchdog driver for the TI OMAP 16xx & 24xx/34xx 32KHz (non-secure) watchdog
  *
  * Author: MontaVista Software, Inc.
  *	 <gdavis@mvista.com> or <source@mvista.com>
@@ -47,50 +47,68 @@
 
 #include "omap_wdt.h"
 
+static struct platform_device *omap_wdt_dev;
+
 static unsigned timer_margin;
 module_param(timer_margin, uint, 0);
 MODULE_PARM_DESC(timer_margin, "initial watchdog timeout (in seconds)");
 
-static int omap_wdt_users;
-static struct clk *armwdt_ck;
-static struct clk *mpu_wdt_ick;
-static struct clk *mpu_wdt_fck;
-
 static unsigned int wdt_trgr_pattern = 0x1234;
 static spinlock_t wdt_lock;
 
-static void omap_wdt_ping(void)
+struct omap_wdt_dev {
+	void __iomem    *base;          /* physical */
+	struct device   *dev;
+	int             omap_wdt_users;
+	struct clk      *armwdt_ck;
+	struct clk      *mpu_wdt_ick;
+	struct clk      *mpu_wdt_fck;
+	struct resource *mem;
+	struct miscdevice omap_wdt_miscdev;
+};
+
+static void omap_wdt_ping(struct omap_wdt_dev *wdev)
 {
+	void __iomem    *base = wdev->base;
+
 	/* wait for posted write to complete */
-	while ((omap_readl(OMAP_WATCHDOG_WPS)) & 0x08)
+	while ((__raw_readl(base + OMAP_WATCHDOG_WPS)) & 0x08)
 		cpu_relax();
+
 	wdt_trgr_pattern = ~wdt_trgr_pattern;
-	omap_writel(wdt_trgr_pattern, (OMAP_WATCHDOG_TGR));
+	__raw_writel(wdt_trgr_pattern, (base + OMAP_WATCHDOG_TGR));
+
 	/* wait for posted write to complete */
-	while ((omap_readl(OMAP_WATCHDOG_WPS)) & 0x08)
+	while ((__raw_readl(base + OMAP_WATCHDOG_WPS)) & 0x08)
 		cpu_relax();
 	/* reloaded WCRR from WLDR */
 }
 
-static void omap_wdt_enable(void)
+static void omap_wdt_enable(struct omap_wdt_dev *wdev)
 {
+	void __iomem *base = wdev->base;
+
 	/* Sequence to enable the watchdog */
-	omap_writel(0xBBBB, OMAP_WATCHDOG_SPR);
-	while ((omap_readl(OMAP_WATCHDOG_WPS)) & 0x10)
+	__raw_writel(0xBBBB, base + OMAP_WATCHDOG_SPR);
+	while ((__raw_readl(base + OMAP_WATCHDOG_WPS)) & 0x10)
 		cpu_relax();
-	omap_writel(0x4444, OMAP_WATCHDOG_SPR);
-	while ((omap_readl(OMAP_WATCHDOG_WPS)) & 0x10)
+
+	__raw_writel(0x4444, base + OMAP_WATCHDOG_SPR);
+	while ((__raw_readl(base + OMAP_WATCHDOG_WPS)) & 0x10)
 		cpu_relax();
 }
 
-static void omap_wdt_disable(void)
+static void omap_wdt_disable(struct omap_wdt_dev *wdev)
 {
+	void __iomem *base = wdev->base;
+
 	/* sequence required to disable watchdog */
-	omap_writel(0xAAAA, OMAP_WATCHDOG_SPR);	/* TIMER_MODE */
-	while (omap_readl(OMAP_WATCHDOG_WPS) & 0x10)
+	__raw_writel(0xAAAA, base + OMAP_WATCHDOG_SPR);	/* TIMER_MODE */
+	while (__raw_readl(base + OMAP_WATCHDOG_WPS) & 0x10)
 		cpu_relax();
-	omap_writel(0x5555, OMAP_WATCHDOG_SPR);	/* TIMER_MODE */
-	while (omap_readl(OMAP_WATCHDOG_WPS) & 0x10)
+
+	__raw_writel(0x5555, base + OMAP_WATCHDOG_SPR);	/* TIMER_MODE */
+	while (__raw_readl(base + OMAP_WATCHDOG_WPS) & 0x10)
 		cpu_relax();
 }
 
@@ -103,83 +121,90 @@ static void omap_wdt_adjust_timeout(unsigned new_timeout)
 	timer_margin = new_timeout;
 }
 
-static void omap_wdt_set_timeout(void)
+static void omap_wdt_set_timeout(struct omap_wdt_dev *wdev)
 {
 	u32 pre_margin = GET_WLDR_VAL(timer_margin);
+	void __iomem *base = wdev->base;
 
 	/* just count up at 32 KHz */
-	while (omap_readl(OMAP_WATCHDOG_WPS) & 0x04)
+	while (__raw_readl(base + OMAP_WATCHDOG_WPS) & 0x04)
 		cpu_relax();
-	omap_writel(pre_margin, OMAP_WATCHDOG_LDR);
-	while (omap_readl(OMAP_WATCHDOG_WPS) & 0x04)
+
+	__raw_writel(pre_margin, base + OMAP_WATCHDOG_LDR);
+	while (__raw_readl(base + OMAP_WATCHDOG_WPS) & 0x04)
 		cpu_relax();
 }
 
 /*
  *	Allow only one task to hold it open
  */
-
 static int omap_wdt_open(struct inode *inode, struct file *file)
 {
-	if (test_and_set_bit(1, (unsigned long *)&omap_wdt_users))
+	struct omap_wdt_dev *wdev = platform_get_drvdata(omap_wdt_dev);
+	void __iomem *base = wdev->base;
+
+	if (test_and_set_bit(1, (unsigned long *)&(wdev->omap_wdt_users)))
 		return -EBUSY;
 
 	if (cpu_is_omap16xx())
-		clk_enable(armwdt_ck);	/* Enable the clock */
+		clk_enable(wdev->armwdt_ck);	/* Enable the clock */
 
-	if (cpu_is_omap24xx()) {
-		clk_enable(mpu_wdt_ick);    /* Enable the interface clock */
-		clk_enable(mpu_wdt_fck);    /* Enable the functional clock */
+	if (cpu_is_omap24xx() || cpu_is_omap34xx()) {
+		clk_enable(wdev->mpu_wdt_ick);    /* Enable the interface clock */
+		clk_enable(wdev->mpu_wdt_fck);    /* Enable the functional clock */
 	}
 
 	/* initialize prescaler */
-	while (omap_readl(OMAP_WATCHDOG_WPS) & 0x01)
+	while (__raw_readl(base + OMAP_WATCHDOG_WPS) & 0x01)
 		cpu_relax();
-	omap_writel((1 << 5) | (PTV << 2), OMAP_WATCHDOG_CNTRL);
-	while (omap_readl(OMAP_WATCHDOG_WPS) & 0x01)
+
+	__raw_writel((1 << 5) | (PTV << 2), base + OMAP_WATCHDOG_CNTRL);
+	while (__raw_readl(base + OMAP_WATCHDOG_WPS) & 0x01)
 		cpu_relax();
 
-	omap_wdt_set_timeout();
-	omap_wdt_enable();
+	file->private_data = (void *) wdev;
+
+	omap_wdt_set_timeout(wdev);
+	omap_wdt_enable(wdev);
+
 	return nonseekable_open(inode, file);
 }
 
 static int omap_wdt_release(struct inode *inode, struct file *file)
 {
+	struct omap_wdt_dev *wdev = file->private_data;
+
 	/*
 	 *      Shut off the timer unless NOWAYOUT is defined.
 	 */
 #ifndef CONFIG_WATCHDOG_NOWAYOUT
-	omap_wdt_disable();
 
-	if (cpu_is_omap16xx()) {
-		clk_disable(armwdt_ck);	/* Disable the clock */
-		clk_put(armwdt_ck);
-		armwdt_ck = NULL;
-	}
+	omap_wdt_disable(wdev);
 
-	if (cpu_is_omap24xx()) {
-		clk_disable(mpu_wdt_ick);	/* Disable the clock */
-		clk_disable(mpu_wdt_fck);	/* Disable the clock */
-		clk_put(mpu_wdt_ick);
-		clk_put(mpu_wdt_fck);
-		mpu_wdt_ick = NULL;
-		mpu_wdt_fck = NULL;
+	if (cpu_is_omap16xx())
+		clk_disable(wdev->armwdt_ck);	/* Disable the clock */
+
+	if (cpu_is_omap24xx() || cpu_is_omap34xx()) {
+		clk_disable(wdev->mpu_wdt_ick);	/* Disable the clock */
+		clk_disable(wdev->mpu_wdt_fck);	/* Disable the clock */
 	}
 #else
 	printk(KERN_CRIT "omap_wdt: Unexpected close, not stopping!\n");
 #endif
-	omap_wdt_users = 0;
+	wdev->omap_wdt_users = 0;
+
 	return 0;
 }
 
 static ssize_t omap_wdt_write(struct file *file, const char __user *data,
 		size_t len, loff_t *ppos)
 {
+	struct omap_wdt_dev *wdev = file->private_data;
+
 	/* Refresh LOAD_TIME. */
 	if (len) {
 		spin_lock(&wdt_lock);
-		omap_wdt_ping();
+		omap_wdt_ping(wdev);
 		spin_unlock(&wdt_lock);
 	}
 	return len;
@@ -188,6 +213,7 @@ static ssize_t omap_wdt_write(struct file *file, const char __user *data,
 static long omap_wdt_ioctl(struct file *file, unsigned int cmd,
 						unsigned long arg)
 {
+	struct omap_wdt_dev *wdev;
 	int new_margin;
 	static const struct watchdog_info ident = {
 		.identity = "OMAP Watchdog",
@@ -195,6 +221,8 @@ static long omap_wdt_ioctl(struct file *file, unsigned int cmd,
 		.firmware_version = 0,
 	};
 
+	wdev = file->private_data;
+
 	switch (cmd) {
 	case WDIOC_GETSUPPORT:
 		return copy_to_user((struct watchdog_info __user *)arg, &ident,
@@ -203,14 +231,14 @@ static long omap_wdt_ioctl(struct file *file, unsigned int cmd,
 		return put_user(0, (int __user *)arg);
 	case WDIOC_GETBOOTSTATUS:
 		if (cpu_is_omap16xx())
-			return put_user(omap_readw(ARM_SYSST),
+			return put_user(__raw_readw(ARM_SYSST),
 					(int __user *)arg);
 		if (cpu_is_omap24xx())
 			return put_user(omap_prcm_get_reset_sources(),
 					(int __user *)arg);
 	case WDIOC_KEEPALIVE:
 		spin_lock(&wdt_lock);
-		omap_wdt_ping();
+		omap_wdt_ping(wdev);
 		spin_unlock(&wdt_lock);
 		return 0;
 	case WDIOC_SETTIMEOUT:
@@ -219,11 +247,11 @@ static long omap_wdt_ioctl(struct file *file, unsigned int cmd,
 		omap_wdt_adjust_timeout(new_margin);
 
 		spin_lock(&wdt_lock);
-		omap_wdt_disable();
-		omap_wdt_set_timeout();
-		omap_wdt_enable();
+		omap_wdt_disable(wdev);
+		omap_wdt_set_timeout(wdev);
+		omap_wdt_enable(wdev);
 
-		omap_wdt_ping();
+		omap_wdt_ping(wdev);
 		spin_unlock(&wdt_lock);
 		/* Fall */
 	case WDIOC_GETTIMEOUT:
@@ -241,96 +269,173 @@ static const struct file_operations omap_wdt_fops = {
 	.release = omap_wdt_release,
 };
 
-static struct miscdevice omap_wdt_miscdev = {
-	.minor = WATCHDOG_MINOR,
-	.name = "watchdog",
-	.fops = &omap_wdt_fops,
-};
-
 static int __init omap_wdt_probe(struct platform_device *pdev)
 {
 	struct resource *res, *mem;
+	struct omap_wdt_dev *wdev;
 	int ret;
 
 	/* reserve static register mappings */
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res)
-		return -ENOENT;
+	if (!res) {
+		ret = -ENOENT;
+		goto err_get_resource;
+	}
+
+	if (omap_wdt_dev) {
+		ret = -EBUSY;
+		goto err_busy;
+	}
 
 	mem = request_mem_region(res->start, res->end - res->start + 1,
 				 pdev->name);
-	if (mem == NULL)
-		return -EBUSY;
+	if (!mem) {
+		ret = -EBUSY;
+		goto err_busy;
+	}
 
-	platform_set_drvdata(pdev, mem);
+	wdev = kzalloc(sizeof(struct omap_wdt_dev), GFP_KERNEL);
+	if (!wdev) {
+		ret = -ENOMEM;
+		goto err_kzalloc;
+	}
 
-	omap_wdt_users = 0;
+	wdev->omap_wdt_users = 0;
+	wdev->mem = mem;
 
 	if (cpu_is_omap16xx()) {
-		armwdt_ck = clk_get(&pdev->dev, "armwdt_ck");
-		if (IS_ERR(armwdt_ck)) {
-			ret = PTR_ERR(armwdt_ck);
-			armwdt_ck = NULL;
-			goto fail;
+		wdev->armwdt_ck = clk_get(&pdev->dev, "armwdt_ck");
+		if (IS_ERR(wdev->armwdt_ck)) {
+			ret = PTR_ERR(wdev->armwdt_ck);
+			wdev->armwdt_ck = NULL;
+			goto err_clk;
 		}
 	}
 
 	if (cpu_is_omap24xx()) {
-		mpu_wdt_ick = clk_get(&pdev->dev, "mpu_wdt_ick");
-		if (IS_ERR(mpu_wdt_ick)) {
-			ret = PTR_ERR(mpu_wdt_ick);
-			mpu_wdt_ick = NULL;
-			goto fail;
+		wdev->mpu_wdt_ick = clk_get(&pdev->dev, "mpu_wdt_ick");
+		if (IS_ERR(wdev->mpu_wdt_ick)) {
+			ret = PTR_ERR(wdev->mpu_wdt_ick);
+			wdev->mpu_wdt_ick = NULL;
+			goto err_clk;
 		}
-		mpu_wdt_fck = clk_get(&pdev->dev, "mpu_wdt_fck");
-		if (IS_ERR(mpu_wdt_fck)) {
-			ret = PTR_ERR(mpu_wdt_fck);
-			mpu_wdt_fck = NULL;
-			goto fail;
+		wdev->mpu_wdt_fck = clk_get(&pdev->dev, "mpu_wdt_fck");
+		if (IS_ERR(wdev->mpu_wdt_fck)) {
+			ret = PTR_ERR(wdev->mpu_wdt_fck);
+			wdev->mpu_wdt_fck = NULL;
+			goto err_clk;
 		}
 	}
 
-	omap_wdt_disable();
+	if (cpu_is_omap34xx()) {
+		wdev->mpu_wdt_ick = clk_get(&pdev->dev, "wdt2_ick");
+		if (IS_ERR(wdev->mpu_wdt_ick)) {
+			ret = PTR_ERR(wdev->mpu_wdt_ick);
+			wdev->mpu_wdt_ick = NULL;
+			goto err_clk;
+		}
+		wdev->mpu_wdt_fck = clk_get(&pdev->dev, "wdt2_fck");
+		if (IS_ERR(wdev->mpu_wdt_fck)) {
+			ret = PTR_ERR(wdev->mpu_wdt_fck);
+			wdev->mpu_wdt_fck = NULL;
+			goto err_clk;
+		}
+	}
+	wdev->base = ioremap(res->start, res->end - res->start + 1);
+	if (!wdev->base) {
+		ret = -ENOMEM;
+		goto err_ioremap;
+	}
+
+	platform_set_drvdata(pdev, wdev);
+
+	omap_wdt_disable(wdev);
 	omap_wdt_adjust_timeout(timer_margin);
 
-	omap_wdt_miscdev.parent = &pdev->dev;
-	ret = misc_register(&omap_wdt_miscdev);
+	wdev->omap_wdt_miscdev.parent = &pdev->dev;
+	wdev->omap_wdt_miscdev.minor = WATCHDOG_MINOR;
+	wdev->omap_wdt_miscdev.name = "watchdog";
+	wdev->omap_wdt_miscdev.fops = &omap_wdt_fops;
+
+	ret = misc_register(&(wdev->omap_wdt_miscdev));
 	if (ret)
-		goto fail;
+		goto err_misc;
 
-	pr_info("OMAP Watchdog Timer: initial timeout %d sec\n", timer_margin);
+	pr_info("OMAP Watchdog Timer Rev 0x%02x: initial timeout %d sec\n",
+		__raw_readl(wdev->base + OMAP_WATCHDOG_REV) & 0xFF,
+		timer_margin);
 
 	/* autogate OCP interface clock */
-	omap_writel(0x01, OMAP_WATCHDOG_SYS_CONFIG);
+	__raw_writel(0x01, wdev->base + OMAP_WATCHDOG_SYS_CONFIG);
+
+	omap_wdt_dev = pdev;
+
 	return 0;
 
-fail:
-	if (armwdt_ck)
-		clk_put(armwdt_ck);
-	if (mpu_wdt_ick)
-		clk_put(mpu_wdt_ick);
-	if (mpu_wdt_fck)
-		clk_put(mpu_wdt_fck);
-	release_resource(mem);
+err_misc:
+	platform_set_drvdata(pdev, NULL);
+	iounmap(wdev->base);
+
+err_ioremap:
+	wdev->base = NULL;
+
+err_clk:
+	if (wdev->armwdt_ck)
+		clk_put(wdev->armwdt_ck);
+	if (wdev->mpu_wdt_ick)
+		clk_put(wdev->mpu_wdt_ick);
+	if (wdev->mpu_wdt_fck)
+		clk_put(wdev->mpu_wdt_fck);
+	kfree(wdev);
+
+err_kzalloc:
+	release_mem_region(res->start, res->end - res->start + 1);
+
+err_busy:
+err_get_resource:
+
 	return ret;
 }
 
 static void omap_wdt_shutdown(struct platform_device *pdev)
 {
-	omap_wdt_disable();
+	struct omap_wdt_dev *wdev = platform_get_drvdata(pdev);
+
+	if (wdev->omap_wdt_users)
+		omap_wdt_disable(wdev);
 }
 
 static int omap_wdt_remove(struct platform_device *pdev)
 {
-	struct resource *mem = platform_get_drvdata(pdev);
-	misc_deregister(&omap_wdt_miscdev);
-	release_resource(mem);
-	if (armwdt_ck)
-		clk_put(armwdt_ck);
-	if (mpu_wdt_ick)
-		clk_put(mpu_wdt_ick);
-	if (mpu_wdt_fck)
-		clk_put(mpu_wdt_fck);
+	struct omap_wdt_dev *wdev = platform_get_drvdata(pdev);
+	struct resource *res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+
+	if (!res)
+		return -ENOENT;
+
+	misc_deregister(&(wdev->omap_wdt_miscdev));
+	release_mem_region(res->start, res->end - res->start + 1);
+	platform_set_drvdata(pdev, NULL);
+
+	if (wdev->armwdt_ck) {
+		clk_put(wdev->armwdt_ck);
+		wdev->armwdt_ck = NULL;
+	}
+
+	if (wdev->mpu_wdt_ick) {
+		clk_put(wdev->mpu_wdt_ick);
+		wdev->mpu_wdt_ick = NULL;
+	}
+
+	if (wdev->mpu_wdt_fck) {
+		clk_put(wdev->mpu_wdt_fck);
+		wdev->mpu_wdt_fck = NULL;
+	}
+	iounmap(wdev->base);
+
+	kfree(wdev);
+	omap_wdt_dev = NULL;
+
 	return 0;
 }
 
@@ -344,17 +449,23 @@ static int omap_wdt_remove(struct platform_device *pdev)
 
 static int omap_wdt_suspend(struct platform_device *pdev, pm_message_t state)
 {
-	if (omap_wdt_users)
-		omap_wdt_disable();
+	struct omap_wdt_dev *wdev = platform_get_drvdata(pdev);
+
+	if (wdev->omap_wdt_users)
+		omap_wdt_disable(wdev);
+
 	return 0;
 }
 
 static int omap_wdt_resume(struct platform_device *pdev)
 {
-	if (omap_wdt_users) {
-		omap_wdt_enable();
-		omap_wdt_ping();
+	struct omap_wdt_dev *wdev = platform_get_drvdata(pdev);
+
+	if (wdev->omap_wdt_users) {
+		omap_wdt_enable(wdev);
+		omap_wdt_ping(wdev);
 	}
+
 	return 0;
 }
 
diff --git a/drivers/watchdog/omap_wdt.h b/drivers/watchdog/omap_wdt.h
index 52a532a5114a..fc02ec6a0386 100644
--- a/drivers/watchdog/omap_wdt.h
+++ b/drivers/watchdog/omap_wdt.h
@@ -30,25 +30,15 @@
 #ifndef _OMAP_WATCHDOG_H
 #define _OMAP_WATCHDOG_H
 
-#define OMAP1610_WATCHDOG_BASE		0xfffeb000
-#define OMAP2420_WATCHDOG_BASE		0x48022000	/*WDT Timer 2 */
-
-#ifdef CONFIG_ARCH_OMAP24XX
-#define OMAP_WATCHDOG_BASE 		OMAP2420_WATCHDOG_BASE
-#else
-#define OMAP_WATCHDOG_BASE 		OMAP1610_WATCHDOG_BASE
-#define RM_RSTST_WKUP			0
-#endif
-
-#define OMAP_WATCHDOG_REV		(OMAP_WATCHDOG_BASE + 0x00)
-#define OMAP_WATCHDOG_SYS_CONFIG	(OMAP_WATCHDOG_BASE + 0x10)
-#define OMAP_WATCHDOG_STATUS		(OMAP_WATCHDOG_BASE + 0x14)
-#define OMAP_WATCHDOG_CNTRL		(OMAP_WATCHDOG_BASE + 0x24)
-#define OMAP_WATCHDOG_CRR		(OMAP_WATCHDOG_BASE + 0x28)
-#define OMAP_WATCHDOG_LDR		(OMAP_WATCHDOG_BASE + 0x2c)
-#define OMAP_WATCHDOG_TGR		(OMAP_WATCHDOG_BASE + 0x30)
-#define OMAP_WATCHDOG_WPS		(OMAP_WATCHDOG_BASE + 0x34)
-#define OMAP_WATCHDOG_SPR		(OMAP_WATCHDOG_BASE + 0x48)
+#define OMAP_WATCHDOG_REV		(0x00)
+#define OMAP_WATCHDOG_SYS_CONFIG	(0x10)
+#define OMAP_WATCHDOG_STATUS		(0x14)
+#define OMAP_WATCHDOG_CNTRL		(0x24)
+#define OMAP_WATCHDOG_CRR		(0x28)
+#define OMAP_WATCHDOG_LDR		(0x2c)
+#define OMAP_WATCHDOG_TGR		(0x30)
+#define OMAP_WATCHDOG_WPS		(0x34)
+#define OMAP_WATCHDOG_SPR		(0x48)
 
 /* Using the prescaler, the OMAP watchdog could go for many
  * months before firing.  These limits work without scaling,
diff --git a/drivers/watchdog/orion5x_wdt.c b/drivers/watchdog/orion5x_wdt.c
new file mode 100644
index 000000000000..14a339f58b6a
--- /dev/null
+++ b/drivers/watchdog/orion5x_wdt.c
@@ -0,0 +1,245 @@
+/*
+ * drivers/watchdog/orion5x_wdt.c
+ *
+ * Watchdog driver for Orion5x processors
+ *
+ * Author: Sylver Bruneau <sylver.bruneau@googlemail.com>
+ *
+ * This file is licensed under  the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/fs.h>
+#include <linux/miscdevice.h>
+#include <linux/watchdog.h>
+#include <linux/init.h>
+#include <linux/uaccess.h>
+#include <linux/io.h>
+#include <linux/spinlock.h>
+
+/*
+ * Watchdog timer block registers.
+ */
+#define TIMER_CTRL		(TIMER_VIRT_BASE + 0x0000)
+#define  WDT_EN			0x0010
+#define WDT_VAL			(TIMER_VIRT_BASE + 0x0024)
+
+#define WDT_MAX_DURATION	(0xffffffff / ORION5X_TCLK)
+#define WDT_IN_USE		0
+#define WDT_OK_TO_CLOSE		1
+
+static int nowayout = WATCHDOG_NOWAYOUT;
+static int heartbeat =  WDT_MAX_DURATION;	/* (seconds) */
+static unsigned long wdt_status;
+static spinlock_t wdt_lock;
+
+static void wdt_enable(void)
+{
+	u32 reg;
+
+	spin_lock(&wdt_lock);
+
+	/* Set watchdog duration */
+	writel(ORION5X_TCLK * heartbeat, WDT_VAL);
+
+	/* Clear watchdog timer interrupt */
+	reg = readl(BRIDGE_CAUSE);
+	reg &= ~WDT_INT_REQ;
+	writel(reg, BRIDGE_CAUSE);
+
+	/* Enable watchdog timer */
+	reg = readl(TIMER_CTRL);
+	reg |= WDT_EN;
+	writel(reg, TIMER_CTRL);
+
+	/* Enable reset on watchdog */
+	reg = readl(CPU_RESET_MASK);
+	reg |= WDT_RESET;
+	writel(reg, CPU_RESET_MASK);
+
+	spin_unlock(&wdt_lock);
+}
+
+static void wdt_disable(void)
+{
+	u32 reg;
+
+	spin_lock(&wdt_lock);
+
+	/* Disable reset on watchdog */
+	reg = readl(CPU_RESET_MASK);
+	reg &= ~WDT_RESET;
+	writel(reg, CPU_RESET_MASK);
+
+	/* Disable watchdog timer */
+	reg = readl(TIMER_CTRL);
+	reg &= ~WDT_EN;
+	writel(reg, TIMER_CTRL);
+
+	spin_unlock(&wdt_lock);
+}
+
+static int orion5x_wdt_get_timeleft(int *time_left)
+{
+	spin_lock(&wdt_lock);
+	*time_left = readl(WDT_VAL) / ORION5X_TCLK;
+	spin_unlock(&wdt_lock);
+	return 0;
+}
+
+static int orion5x_wdt_open(struct inode *inode, struct file *file)
+{
+	if (test_and_set_bit(WDT_IN_USE, &wdt_status))
+		return -EBUSY;
+	clear_bit(WDT_OK_TO_CLOSE, &wdt_status);
+	wdt_enable();
+	return nonseekable_open(inode, file);
+}
+
+static ssize_t orion5x_wdt_write(struct file *file, const char *data,
+					size_t len, loff_t *ppos)
+{
+	if (len) {
+		if (!nowayout) {
+			size_t i;
+
+			clear_bit(WDT_OK_TO_CLOSE, &wdt_status);
+			for (i = 0; i != len; i++) {
+				char c;
+
+				if (get_user(c, data + i))
+					return -EFAULT;
+				if (c == 'V')
+					set_bit(WDT_OK_TO_CLOSE, &wdt_status);
+			}
+		}
+		wdt_enable();
+	}
+	return len;
+}
+
+static struct watchdog_info ident = {
+	.options	= WDIOF_MAGICCLOSE | WDIOF_SETTIMEOUT |
+			  WDIOF_KEEPALIVEPING,
+	.identity	= "Orion5x Watchdog",
+};
+
+
+static long orion5x_wdt_ioctl(struct file *file, unsigned int cmd,
+				unsigned long arg)
+{
+	int ret = -ENOTTY;
+	int time;
+
+	switch (cmd) {
+	case WDIOC_GETSUPPORT:
+		ret = copy_to_user((struct watchdog_info *)arg, &ident,
+				   sizeof(ident)) ? -EFAULT : 0;
+		break;
+
+	case WDIOC_GETSTATUS:
+	case WDIOC_GETBOOTSTATUS:
+		ret = put_user(0, (int *)arg);
+		break;
+
+	case WDIOC_KEEPALIVE:
+		wdt_enable();
+		ret = 0;
+		break;
+
+	case WDIOC_SETTIMEOUT:
+		ret = get_user(time, (int *)arg);
+		if (ret)
+			break;
+
+		if (time <= 0 || time > WDT_MAX_DURATION) {
+			ret = -EINVAL;
+			break;
+		}
+		heartbeat = time;
+		wdt_enable();
+		/* Fall through */
+
+	case WDIOC_GETTIMEOUT:
+		ret = put_user(heartbeat, (int *)arg);
+		break;
+
+	case WDIOC_GETTIMELEFT:
+		if (orion5x_wdt_get_timeleft(&time)) {
+			ret = -EINVAL;
+			break;
+		}
+		ret = put_user(time, (int *)arg);
+		break;
+	}
+	return ret;
+}
+
+static int orion5x_wdt_release(struct inode *inode, struct file *file)
+{
+	if (test_bit(WDT_OK_TO_CLOSE, &wdt_status))
+		wdt_disable();
+	else
+		printk(KERN_CRIT "WATCHDOG: Device closed unexpectedly - "
+					"timer will not stop\n");
+	clear_bit(WDT_IN_USE, &wdt_status);
+	clear_bit(WDT_OK_TO_CLOSE, &wdt_status);
+
+	return 0;
+}
+
+
+static const struct file_operations orion5x_wdt_fops = {
+	.owner		= THIS_MODULE,
+	.llseek		= no_llseek,
+	.write		= orion5x_wdt_write,
+	.unlocked_ioctl	= orion5x_wdt_ioctl,
+	.open		= orion5x_wdt_open,
+	.release	= orion5x_wdt_release,
+};
+
+static struct miscdevice orion5x_wdt_miscdev = {
+	.minor		= WATCHDOG_MINOR,
+	.name		= "watchdog",
+	.fops		= &orion5x_wdt_fops,
+};
+
+static int __init orion5x_wdt_init(void)
+{
+	int ret;
+
+	spin_lock_init(&wdt_lock);
+
+	ret = misc_register(&orion5x_wdt_miscdev);
+	if (ret == 0)
+		printk("Orion5x Watchdog Timer: heartbeat %d sec\n",
+								heartbeat);
+
+	return ret;
+}
+
+static void __exit orion5x_wdt_exit(void)
+{
+	misc_deregister(&orion5x_wdt_miscdev);
+}
+
+module_init(orion5x_wdt_init);
+module_exit(orion5x_wdt_exit);
+
+MODULE_AUTHOR("Sylver Bruneau <sylver.bruneau@googlemail.com>");
+MODULE_DESCRIPTION("Orion5x Processor Watchdog");
+
+module_param(heartbeat, int, 0);
+MODULE_PARM_DESC(heartbeat, "Watchdog heartbeat in seconds (default is "
+					__MODULE_STRING(WDT_MAX_DURATION) ")");
+
+module_param(nowayout, int, 0);
+MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started");
+
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/sbus/char/riowatchdog.c b/drivers/watchdog/riowd.c
index 88c0fc6395e1..09cb1833ea27 100644
--- a/drivers/sbus/char/riowatchdog.c
+++ b/drivers/watchdog/riowd.c
@@ -1,7 +1,6 @@
-/* $Id: riowatchdog.c,v 1.3.2.2 2002/01/23 18:48:02 davem Exp $
- * riowatchdog.c - driver for hw watchdog inside Super I/O of RIO
+/* riowd.c - driver for hw watchdog inside Super I/O of RIO
  *
- * Copyright (C) 2001 David S. Miller (davem@redhat.com)
+ * Copyright (C) 2001, 2008 David S. Miller (davem@davemloft.net)
  */
 
 #include <linux/kernel.h>
@@ -12,14 +11,13 @@
 #include <linux/init.h>
 #include <linux/miscdevice.h>
 #include <linux/smp_lock.h>
+#include <linux/watchdog.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <asm/io.h>
-#include <asm/ebus.h>
-#include <asm/bbc.h>
-#include <asm/oplib.h>
 #include <asm/uaccess.h>
 
-#include <asm/watchdog.h>
 
 /* RIO uses the NatSemi Super I/O power management logical device
  * as its' watchdog.
@@ -45,74 +43,35 @@
  * The watchdog device generates no interrupts.
  */
 
-MODULE_AUTHOR("David S. Miller <davem@redhat.com>");
+MODULE_AUTHOR("David S. Miller <davem@davemloft.net>");
 MODULE_DESCRIPTION("Hardware watchdog driver for Sun RIO");
 MODULE_SUPPORTED_DEVICE("watchdog");
 MODULE_LICENSE("GPL");
 
-#define RIOWD_NAME	"pmc"
-#define RIOWD_MINOR	215
+#define DRIVER_NAME	"riowd"
+#define PFX		DRIVER_NAME ": "
 
-static DEFINE_SPINLOCK(riowd_lock);
+struct riowd {
+	void __iomem		*regs;
+	spinlock_t		lock;
+};
+
+static struct riowd *riowd_device;
 
-static void __iomem *bbc_regs;
-static void __iomem *riowd_regs;
 #define WDTO_INDEX	0x05
 
 static int riowd_timeout = 1;		/* in minutes */
 module_param(riowd_timeout, int, 0);
 MODULE_PARM_DESC(riowd_timeout, "Watchdog timeout in minutes");
 
-#if 0 /* Currently unused. */
-static u8 riowd_readreg(int index)
+static void riowd_writereg(struct riowd *p, u8 val, int index)
 {
 	unsigned long flags;
-	u8 ret;
 
-	spin_lock_irqsave(&riowd_lock, flags);
-	writeb(index, riowd_regs + 0);
-	ret = readb(riowd_regs + 1);
-	spin_unlock_irqrestore(&riowd_lock, flags);
-
-	return ret;
-}
-#endif
-
-static void riowd_writereg(u8 val, int index)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&riowd_lock, flags);
-	writeb(index, riowd_regs + 0);
-	writeb(val, riowd_regs + 1);
-	spin_unlock_irqrestore(&riowd_lock, flags);
-}
-
-static void riowd_pingtimer(void)
-{
-	riowd_writereg(riowd_timeout, WDTO_INDEX);
-}
-
-static void riowd_stoptimer(void)
-{
-	u8 val;
-
-	riowd_writereg(0, WDTO_INDEX);
-
-	val = readb(bbc_regs + BBC_WDACTION);
-	val &= ~BBC_WDACTION_RST;
-	writeb(val, bbc_regs + BBC_WDACTION);
-}
-
-static void riowd_starttimer(void)
-{
-	u8 val;
-
-	riowd_writereg(riowd_timeout, WDTO_INDEX);
-
-	val = readb(bbc_regs + BBC_WDACTION);
-	val |= BBC_WDACTION_RST;
-	writeb(val, bbc_regs + BBC_WDACTION);
+	spin_lock_irqsave(&p->lock, flags);
+	writeb(index, p->regs + 0);
+	writeb(val, p->regs + 1);
+	spin_unlock_irqrestore(&p->lock, flags);
 }
 
 static int riowd_open(struct inode *inode, struct file *filp)
@@ -131,9 +90,12 @@ static int riowd_ioctl(struct inode *inode, struct file *filp,
 		       unsigned int cmd, unsigned long arg)
 {
 	static struct watchdog_info info = {
-	       	WDIOF_SETTIMEOUT, 0, "Natl. Semiconductor PC97317"
+		.options		= WDIOF_SETTIMEOUT,
+		.firmware_version	= 1,
+		.identity		= DRIVER_NAME,
 	};
 	void __user *argp = (void __user *)arg;
+	struct riowd *p = riowd_device;
 	unsigned int options;
 	int new_margin;
 
@@ -150,7 +112,7 @@ static int riowd_ioctl(struct inode *inode, struct file *filp,
 		break;
 
 	case WDIOC_KEEPALIVE:
-		riowd_pingtimer();
+		riowd_writereg(p, riowd_timeout, WDTO_INDEX);
 		break;
 
 	case WDIOC_SETOPTIONS:
@@ -158,9 +120,9 @@ static int riowd_ioctl(struct inode *inode, struct file *filp,
 			return -EFAULT;
 
 		if (options & WDIOS_DISABLECARD)
-			riowd_stoptimer();
+			riowd_writereg(p, 0, WDTO_INDEX);
 		else if (options & WDIOS_ENABLECARD)
-			riowd_starttimer();
+			riowd_writereg(p, riowd_timeout, WDTO_INDEX);
 		else
 			return -EINVAL;
 
@@ -170,9 +132,9 @@ static int riowd_ioctl(struct inode *inode, struct file *filp,
 		if (get_user(new_margin, (int __user *)argp))
 			return -EFAULT;
 		if ((new_margin < 60) || (new_margin > (255 * 60)))
-		    return -EINVAL;
+			return -EINVAL;
 		riowd_timeout = (new_margin + 59) / 60;
-		riowd_pingtimer();
+		riowd_writereg(p, riowd_timeout, WDTO_INDEX);
 		/* Fall */
 
 	case WDIOC_GETTIMEOUT:
@@ -187,8 +149,10 @@ static int riowd_ioctl(struct inode *inode, struct file *filp,
 
 static ssize_t riowd_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
 {
+	struct riowd *p = riowd_device;
+
 	if (count) {
-		riowd_pingtimer();
+		riowd_writereg(p, riowd_timeout, WDTO_INDEX);
 		return 1;
 	}
 
@@ -197,99 +161,99 @@ static ssize_t riowd_write(struct file *file, const char __user *buf, size_t cou
 
 static const struct file_operations riowd_fops = {
 	.owner =	THIS_MODULE,
+	.llseek =	no_llseek,
 	.ioctl =	riowd_ioctl,
 	.open =		riowd_open,
 	.write =	riowd_write,
 	.release =	riowd_release,
 };
 
-static struct miscdevice riowd_miscdev = { RIOWD_MINOR, RIOWD_NAME, &riowd_fops };
+static struct miscdevice riowd_miscdev = {
+	.minor	= WATCHDOG_MINOR,
+	.name	= "watchdog",
+	.fops	= &riowd_fops
+};
 
-static int __init riowd_bbc_init(void)
+static int __devinit riowd_probe(struct of_device *op,
+				 const struct of_device_id *match)
 {
-	struct 	linux_ebus *ebus = NULL;
-	struct 	linux_ebus_device *edev = NULL;
-	u8 val;
-
-	for_each_ebus(ebus) {
-		for_each_ebusdev(edev, ebus) {
-			if (!strcmp(edev->ofdev.node->name, "bbc"))
-				goto found_bbc;
-		}
-	}
+	struct riowd *p;
+	int err = -EINVAL;
 
-found_bbc:
-	if (!edev)
-		return -ENODEV;
-	bbc_regs = ioremap(edev->resource[0].start, BBC_REGS_SIZE);
-	if (!bbc_regs)
-		return -ENODEV;
+	if (riowd_device)
+		goto out;
 
-	/* Turn it off. */
-	val = readb(bbc_regs + BBC_WDACTION);
-	val &= ~BBC_WDACTION_RST;
-	writeb(val, bbc_regs + BBC_WDACTION);
+	err = -ENOMEM;
+	p = kzalloc(sizeof(*p), GFP_KERNEL);
+	if (!p)
+		goto out;
 
-	return 0;
-}
+	spin_lock_init(&p->lock);
 
-static int __init riowd_init(void)
-{
-	struct 	linux_ebus *ebus = NULL;
-	struct 	linux_ebus_device *edev = NULL;
-
-	for_each_ebus(ebus) {
-		for_each_ebusdev(edev, ebus) {
-			if (!strcmp(edev->ofdev.node->name, RIOWD_NAME))
-				goto ebus_done;
-		}
+	p->regs = of_ioremap(&op->resource[0], 0, 2, DRIVER_NAME);
+	if (!p->regs) {
+		printk(KERN_ERR PFX "Cannot map registers.\n");
+		goto out_free;
 	}
 
-ebus_done:
-	if (!edev)
-		goto fail;
-
-	riowd_regs = ioremap(edev->resource[0].start, 2);
-	if (riowd_regs == NULL) {
-		printk(KERN_ERR "pmc: Cannot map registers.\n");
-		return -ENODEV;
+	err = misc_register(&riowd_miscdev);
+	if (err) {
+		printk(KERN_ERR PFX "Cannot register watchdog misc device.\n");
+		goto out_iounmap;
 	}
 
-	if (riowd_bbc_init()) {
-		printk(KERN_ERR "pmc: Failure initializing BBC config.\n");
-		goto fail;
-	}
+	printk(KERN_INFO PFX "Hardware watchdog [%i minutes], "
+	       "regs at %p\n", riowd_timeout, p->regs);
 
-	if (misc_register(&riowd_miscdev)) {
-		printk(KERN_ERR "pmc: Cannot register watchdog misc device.\n");
-		goto fail;
-	}
+	dev_set_drvdata(&op->dev, p);
+	riowd_device = p;
+	err = 0;
 
-	printk(KERN_INFO "pmc: Hardware watchdog [%i minutes], "
-	       "regs at %p\n", riowd_timeout, riowd_regs);
+out_iounmap:
+	of_iounmap(&op->resource[0], p->regs, 2);
 
-	return 0;
+out_free:
+	kfree(p);
 
-fail:
-	if (riowd_regs) {
-		iounmap(riowd_regs);
-		riowd_regs = NULL;
-	}
-	if (bbc_regs) {
-		iounmap(bbc_regs);
-		bbc_regs = NULL;
-	}
-	return -ENODEV;
+out:
+	return err;
 }
 
-static void __exit riowd_cleanup(void)
+static int __devexit riowd_remove(struct of_device *op)
 {
+	struct riowd *p = dev_get_drvdata(&op->dev);
+
 	misc_deregister(&riowd_miscdev);
-	iounmap(riowd_regs);
-	riowd_regs = NULL;
-	iounmap(bbc_regs);
-	bbc_regs = NULL;
+	of_iounmap(&op->resource[0], p->regs, 2);
+	kfree(p);
+
+	return 0;
+}
+
+static const struct of_device_id riowd_match[] = {
+	{
+		.name = "pmc",
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, riowd_match);
+
+static struct of_platform_driver riowd_driver = {
+	.name		= DRIVER_NAME,
+	.match_table	= riowd_match,
+	.probe		= riowd_probe,
+	.remove		= __devexit_p(riowd_remove),
+};
+
+static int __init riowd_init(void)
+{
+	return of_register_driver(&riowd_driver, &of_bus_type);
+}
+
+static void __exit riowd_exit(void)
+{
+	of_unregister_driver(&riowd_driver);
 }
 
 module_init(riowd_init);
-module_exit(riowd_cleanup);
+module_exit(riowd_exit);
diff --git a/drivers/watchdog/w83697ug_wdt.c b/drivers/watchdog/w83697ug_wdt.c
new file mode 100644
index 000000000000..c73b5e2919c6
--- /dev/null
+++ b/drivers/watchdog/w83697ug_wdt.c
@@ -0,0 +1,392 @@
+/*
+ *	w83697ug/uf WDT driver
+ *
+ *	(c) Copyright 2008 Flemming Fransen <ff@nrvissing.net>
+ *              reused original code to supoprt w83697ug/uf.
+ *
+ *	Based on w83627hf_wdt.c which is based on advantechwdt.c
+ *	which is based on wdt.c.
+ *	Original copyright messages:
+ *
+ *	(c) Copyright 2007 Vlad Drukker <vlad@storewiz.com>
+ *		added support for W83627THF.
+ *
+ *	(c) Copyright 2003 Pádraig Brady <P@draigBrady.com>
+ *
+ *	(c) Copyright 2000-2001 Marek Michalkiewicz <marekm@linux.org.pl>
+ *
+ *	(c) Copyright 1996 Alan Cox <alan@redhat.com>, All Rights Reserved.
+ *				http://www.redhat.com
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License
+ *	as published by the Free Software Foundation; either version
+ *	2 of the License, or (at your option) any later version.
+ *
+ *	Neither Alan Cox nor CymruNet Ltd. admit liability nor provide
+ *	warranty for any of this software. This material is provided
+ *	"AS-IS" and at no charge.
+ *
+ *	(c) Copyright 1995    Alan Cox <alan@redhat.com>
+ */
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/types.h>
+#include <linux/miscdevice.h>
+#include <linux/watchdog.h>
+#include <linux/fs.h>
+#include <linux/ioport.h>
+#include <linux/notifier.h>
+#include <linux/reboot.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/io.h>
+#include <linux/uaccess.h>
+
+#include <asm/system.h>
+
+#define WATCHDOG_NAME "w83697ug/uf WDT"
+#define PFX WATCHDOG_NAME ": "
+#define WATCHDOG_TIMEOUT 60		/* 60 sec default timeout */
+
+static unsigned long wdt_is_open;
+static char expect_close;
+static DEFINE_SPINLOCK(io_lock);
+
+static int wdt_io = 0x2e;
+module_param(wdt_io, int, 0);
+MODULE_PARM_DESC(wdt_io, "w83697ug/uf WDT io port (default 0x2e)");
+
+static int timeout = WATCHDOG_TIMEOUT;	/* in seconds */
+module_param(timeout, int, 0);
+MODULE_PARM_DESC(timeout,
+	"Watchdog timeout in seconds. 1<= timeout <=255 (default="
+				__MODULE_STRING(WATCHDOG_TIMEOUT) ")");
+
+static int nowayout = WATCHDOG_NOWAYOUT;
+module_param(nowayout, int, 0);
+MODULE_PARM_DESC(nowayout,
+	"Watchdog cannot be stopped once started (default="
+				__MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
+
+/*
+ *	Kernel methods.
+ */
+
+#define WDT_EFER (wdt_io+0)   /* Extended Function Enable Registers */
+#define WDT_EFIR (wdt_io+0)   /* Extended Function Index Register
+							(same as EFER) */
+#define WDT_EFDR (WDT_EFIR+1) /* Extended Function Data Register */
+
+static void w83697ug_select_wd_register(void)
+{
+	unsigned char c;
+	unsigned char version;
+
+	outb_p(0x87, WDT_EFER); /* Enter extended function mode */
+	outb_p(0x87, WDT_EFER); /* Again according to manual */
+
+	outb(0x20, WDT_EFER); 	/* check chip version	*/
+	version = inb(WDT_EFDR);
+
+	if (version == 0x68) {	/* W83697UG 		*/
+		printk(KERN_INFO PFX "Watchdog chip version 0x%02x = "
+			"W83697UG/UF found at 0x%04x\n", version, wdt_io);
+
+		outb_p(0x2b, WDT_EFER);
+		c = inb_p(WDT_EFDR);    /* select WDT0 */
+		c &= ~0x04;
+		outb_p(0x2b, WDT_EFER);
+		outb_p(c, WDT_EFDR);	/* set pin118 to WDT0 */
+
+	} else {
+		printk(KERN_ERR PFX "No W83697UG/UF could be found\n");
+		return -EIO;
+	}
+
+	outb_p(0x07, WDT_EFER); /* point to logical device number reg */
+	outb_p(0x08, WDT_EFDR); /* select logical device 8 (GPIO2) */
+	outb_p(0x30, WDT_EFER); /* select CR30 */
+	c = inb_p(WDT_EFDR);
+	outb_p(c || 0x01, WDT_EFDR); /* set bit 0 to activate GPIO2 */
+}
+
+static void w83697ug_unselect_wd_register(void)
+{
+	outb_p(0xAA, WDT_EFER); /* Leave extended function mode */
+}
+
+static void w83697ug_init(void)
+{
+	unsigned char t;
+
+	w83697ug_select_wd_register();
+
+	outb_p(0xF6, WDT_EFER); /* Select CRF6 */
+	t = inb_p(WDT_EFDR);    /* read CRF6 */
+	if (t != 0) {
+		printk(KERN_INFO PFX "Watchdog already running."
+			" Resetting timeout to %d sec\n", timeout);
+		outb_p(timeout, WDT_EFDR);    /* Write back to CRF6 */
+	}
+	outb_p(0xF5, WDT_EFER); /* Select CRF5 */
+	t = inb_p(WDT_EFDR);    /* read CRF5 */
+	t &= ~0x0C;             /* set second mode &
+					disable keyboard turning off watchdog */
+	outb_p(t, WDT_EFDR);    /* Write back to CRF5 */
+
+	w83697ug_unselect_wd_register();
+}
+
+static void wdt_ctrl(int timeout)
+{
+	spin_lock(&io_lock);
+
+	w83697ug_select_wd_register();
+
+	outb_p(0xF4, WDT_EFER);    /* Select CRF4 */
+	outb_p(timeout, WDT_EFDR); /* Write Timeout counter to CRF4 */
+
+	w83697ug_unselect_wd_register();
+
+	spin_unlock(&io_lock);
+}
+
+static int wdt_ping(void)
+{
+	wdt_ctrl(timeout);
+	return 0;
+}
+
+static int wdt_disable(void)
+{
+	wdt_ctrl(0);
+	return 0;
+}
+
+static int wdt_set_heartbeat(int t)
+{
+	if (t < 1 || t > 255)
+		return -EINVAL;
+
+	timeout = t;
+	return 0;
+}
+
+static ssize_t wdt_write(struct file *file, const char __user *buf,
+						size_t count, loff_t *ppos)
+{
+	if (count) {
+		if (!nowayout) {
+			size_t i;
+
+			expect_close = 0;
+
+			for (i = 0; i != count; i++) {
+				char c;
+				if (get_user(c, buf + i))
+					return -EFAULT;
+				if (c == 'V')
+					expect_close = 42;
+			}
+		}
+		wdt_ping();
+	}
+	return count;
+}
+
+static long wdt_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	void __user *argp = (void __user *)arg;
+	int __user *p = argp;
+	int new_timeout;
+	static const struct watchdog_info ident = {
+		.options =		WDIOF_KEEPALIVEPING |
+					WDIOF_SETTIMEOUT |
+					WDIOF_MAGICCLOSE,
+		.firmware_version =	1,
+		.identity =		"W83697UG WDT",
+	};
+
+	switch (cmd) {
+	case WDIOC_GETSUPPORT:
+		if (copy_to_user(argp, &ident, sizeof(ident)))
+			return -EFAULT;
+		break;
+
+	case WDIOC_GETSTATUS:
+	case WDIOC_GETBOOTSTATUS:
+		return put_user(0, p);
+
+	case WDIOC_SETOPTIONS:
+	{
+		int options, retval = -EINVAL;
+
+		if (get_user(options, p))
+			return -EFAULT;
+
+		if (options & WDIOS_DISABLECARD) {
+			wdt_disable();
+			retval = 0;
+		}
+
+		if (options & WDIOS_ENABLECARD) {
+			wdt_ping();
+			retval = 0;
+		}
+
+		return retval;
+	}
+
+	case WDIOC_KEEPALIVE:
+		wdt_ping();
+		break;
+
+	case WDIOC_SETTIMEOUT:
+		if (get_user(new_timeout, p))
+			return -EFAULT;
+		if (wdt_set_heartbeat(new_timeout))
+			return -EINVAL;
+		wdt_ping();
+		/* Fall */
+
+	case WDIOC_GETTIMEOUT:
+		return put_user(timeout, p);
+
+	default:
+		return -ENOTTY;
+	}
+	return 0;
+}
+
+static int wdt_open(struct inode *inode, struct file *file)
+{
+	if (test_and_set_bit(0, &wdt_is_open))
+		return -EBUSY;
+	/*
+	 *	Activate
+	 */
+
+	wdt_ping();
+	return nonseekable_open(inode, file);
+}
+
+static int wdt_close(struct inode *inode, struct file *file)
+{
+	if (expect_close == 42)
+		wdt_disable();
+	else {
+		printk(KERN_CRIT PFX
+			"Unexpected close, not stopping watchdog!\n");
+		wdt_ping();
+	}
+	expect_close = 0;
+	clear_bit(0, &wdt_is_open);
+	return 0;
+}
+
+/*
+ *	Notifier for system down
+ */
+
+static int wdt_notify_sys(struct notifier_block *this, unsigned long code,
+	void *unused)
+{
+	if (code == SYS_DOWN || code == SYS_HALT)
+		wdt_disable();	/* Turn the WDT off */
+
+	return NOTIFY_DONE;
+}
+
+/*
+ *	Kernel Interfaces
+ */
+
+static const struct file_operations wdt_fops = {
+	.owner		= THIS_MODULE,
+	.llseek		= no_llseek,
+	.write		= wdt_write,
+	.unlocked_ioctl	= wdt_ioctl,
+	.open		= wdt_open,
+	.release	= wdt_close,
+};
+
+static struct miscdevice wdt_miscdev = {
+	.minor = WATCHDOG_MINOR,
+	.name = "watchdog",
+	.fops = &wdt_fops,
+};
+
+/*
+ *	The WDT needs to learn about soft shutdowns in order to
+ *	turn the timebomb registers off.
+ */
+
+static struct notifier_block wdt_notifier = {
+	.notifier_call = wdt_notify_sys,
+};
+
+static int __init wdt_init(void)
+{
+	int ret;
+
+	printk(KERN_INFO "WDT driver for the Winbond(TM) W83697UG/UF Super I/O chip initialising.\n");
+
+	if (wdt_set_heartbeat(timeout)) {
+		wdt_set_heartbeat(WATCHDOG_TIMEOUT);
+		printk(KERN_INFO PFX
+			"timeout value must be 1<=timeout<=255, using %d\n",
+			WATCHDOG_TIMEOUT);
+	}
+
+	if (!request_region(wdt_io, 1, WATCHDOG_NAME)) {
+		printk(KERN_ERR PFX "I/O address 0x%04x already in use\n",
+			wdt_io);
+		ret = -EIO;
+		goto out;
+	}
+
+	w83697ug_init();
+
+	ret = register_reboot_notifier(&wdt_notifier);
+	if (ret != 0) {
+		printk(KERN_ERR PFX
+			"cannot register reboot notifier (err=%d)\n", ret);
+		goto unreg_regions;
+	}
+
+	ret = misc_register(&wdt_miscdev);
+	if (ret != 0) {
+		printk(KERN_ERR PFX
+			"cannot register miscdev on minor=%d (err=%d)\n",
+			WATCHDOG_MINOR, ret);
+		goto unreg_reboot;
+	}
+
+	printk(KERN_INFO PFX "initialized. timeout=%d sec (nowayout=%d)\n",
+		timeout, nowayout);
+
+out:
+	return ret;
+unreg_reboot:
+	unregister_reboot_notifier(&wdt_notifier);
+unreg_regions:
+	release_region(wdt_io, 1);
+	goto out;
+}
+
+static void __exit wdt_exit(void)
+{
+	misc_deregister(&wdt_miscdev);
+	unregister_reboot_notifier(&wdt_notifier);
+	release_region(wdt_io, 1);
+}
+
+module_init(wdt_init);
+module_exit(wdt_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Flemming Frandsen <ff@nrvissing.net>");
+MODULE_DESCRIPTION("w83697ug/uf WDT driver");
+MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile
index 363286c54290..d2a8fdf0e191 100644
--- a/drivers/xen/Makefile
+++ b/drivers/xen/Makefile
@@ -1,4 +1,5 @@
 obj-y	+= grant-table.o features.o events.o manage.o
 obj-y	+= xenbus/
+obj-$(CONFIG_HOTPLUG_CPU)	+= cpu_hotplug.o
 obj-$(CONFIG_XEN_XENCOMM)	+= xencomm.o
 obj-$(CONFIG_XEN_BALLOON)	+= balloon.o
diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c
index 2e15da5459cf..8c83abc73400 100644
--- a/drivers/xen/balloon.c
+++ b/drivers/xen/balloon.c
@@ -53,7 +53,6 @@
 #include <asm/tlb.h>
 
 #include <xen/interface/memory.h>
-#include <xen/balloon.h>
 #include <xen/xenbus.h>
 #include <xen/features.h>
 #include <xen/page.h>
@@ -226,9 +225,8 @@ static int increase_reservation(unsigned long nr_pages)
 	}
 
 	set_xen_guest_handle(reservation.extent_start, frame_list);
-	reservation.nr_extents   = nr_pages;
-	rc = HYPERVISOR_memory_op(
-		XENMEM_populate_physmap, &reservation);
+	reservation.nr_extents = nr_pages;
+	rc = HYPERVISOR_memory_op(XENMEM_populate_physmap, &reservation);
 	if (rc < nr_pages) {
 		if (rc > 0) {
 			int ret;
@@ -236,7 +234,7 @@ static int increase_reservation(unsigned long nr_pages)
 			/* We hit the Xen hard limit: reprobe. */
 			reservation.nr_extents = rc;
 			ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
-					&reservation);
+						   &reservation);
 			BUG_ON(ret != rc);
 		}
 		if (rc >= 0)
@@ -420,7 +418,7 @@ static int __init balloon_init(void)
 	unsigned long pfn;
 	struct page *page;
 
-	if (!is_running_on_xen())
+	if (!xen_pv_domain())
 		return -ENODEV;
 
 	pr_info("xen_balloon: Initialising balloon driver.\n");
@@ -464,136 +462,13 @@ static void balloon_exit(void)
 
 module_exit(balloon_exit);
 
-static void balloon_update_driver_allowance(long delta)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&balloon_lock, flags);
-	balloon_stats.driver_pages += delta;
-	spin_unlock_irqrestore(&balloon_lock, flags);
-}
-
-static int dealloc_pte_fn(
-	pte_t *pte, struct page *pmd_page, unsigned long addr, void *data)
-{
-	unsigned long mfn = pte_mfn(*pte);
-	int ret;
-	struct xen_memory_reservation reservation = {
-		.nr_extents   = 1,
-		.extent_order = 0,
-		.domid        = DOMID_SELF
-	};
-	set_xen_guest_handle(reservation.extent_start, &mfn);
-	set_pte_at(&init_mm, addr, pte, __pte_ma(0ull));
-	set_phys_to_machine(__pa(addr) >> PAGE_SHIFT, INVALID_P2M_ENTRY);
-	ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation);
-	BUG_ON(ret != 1);
-	return 0;
-}
-
-static struct page **alloc_empty_pages_and_pagevec(int nr_pages)
-{
-	unsigned long vaddr, flags;
-	struct page *page, **pagevec;
-	int i, ret;
-
-	pagevec = kmalloc(sizeof(page) * nr_pages, GFP_KERNEL);
-	if (pagevec == NULL)
-		return NULL;
-
-	for (i = 0; i < nr_pages; i++) {
-		page = pagevec[i] = alloc_page(GFP_KERNEL);
-		if (page == NULL)
-			goto err;
-
-		vaddr = (unsigned long)page_address(page);
-
-		scrub_page(page);
-
-		spin_lock_irqsave(&balloon_lock, flags);
-
-		if (xen_feature(XENFEAT_auto_translated_physmap)) {
-			unsigned long gmfn = page_to_pfn(page);
-			struct xen_memory_reservation reservation = {
-				.nr_extents   = 1,
-				.extent_order = 0,
-				.domid        = DOMID_SELF
-			};
-			set_xen_guest_handle(reservation.extent_start, &gmfn);
-			ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
-						   &reservation);
-			if (ret == 1)
-				ret = 0; /* success */
-		} else {
-			ret = apply_to_page_range(&init_mm, vaddr, PAGE_SIZE,
-						  dealloc_pte_fn, NULL);
-		}
-
-		if (ret != 0) {
-			spin_unlock_irqrestore(&balloon_lock, flags);
-			__free_page(page);
-			goto err;
-		}
-
-		totalram_pages = --balloon_stats.current_pages;
-
-		spin_unlock_irqrestore(&balloon_lock, flags);
-	}
-
- out:
-	schedule_work(&balloon_worker);
-	flush_tlb_all();
-	return pagevec;
-
- err:
-	spin_lock_irqsave(&balloon_lock, flags);
-	while (--i >= 0)
-		balloon_append(pagevec[i]);
-	spin_unlock_irqrestore(&balloon_lock, flags);
-	kfree(pagevec);
-	pagevec = NULL;
-	goto out;
-}
-
-static void free_empty_pages_and_pagevec(struct page **pagevec, int nr_pages)
-{
-	unsigned long flags;
-	int i;
-
-	if (pagevec == NULL)
-		return;
-
-	spin_lock_irqsave(&balloon_lock, flags);
-	for (i = 0; i < nr_pages; i++) {
-		BUG_ON(page_count(pagevec[i]) != 1);
-		balloon_append(pagevec[i]);
-	}
-	spin_unlock_irqrestore(&balloon_lock, flags);
-
-	kfree(pagevec);
-
-	schedule_work(&balloon_worker);
-}
-
-static void balloon_release_driver_page(struct page *page)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&balloon_lock, flags);
-	balloon_append(page);
-	balloon_stats.driver_pages--;
-	spin_unlock_irqrestore(&balloon_lock, flags);
-
-	schedule_work(&balloon_worker);
-}
-
-
-#define BALLOON_SHOW(name, format, args...)			\
-	static ssize_t show_##name(struct sys_device *dev,	\
-				   char *buf)			\
-	{							\
-		return sprintf(buf, format, ##args);		\
-	}							\
+#define BALLOON_SHOW(name, format, args...)				\
+	static ssize_t show_##name(struct sys_device *dev,		\
+				   struct sysdev_attribute *attr,	\
+				   char *buf)				\
+	{								\
+		return sprintf(buf, format, ##args);			\
+	}								\
 	static SYSDEV_ATTR(name, S_IRUGO, show_##name, NULL)
 
 BALLOON_SHOW(current_kb, "%lu\n", PAGES2KB(balloon_stats.current_pages));
@@ -604,7 +479,8 @@ BALLOON_SHOW(hard_limit_kb,
 	     (balloon_stats.hard_limit!=~0UL) ? PAGES2KB(balloon_stats.hard_limit) : 0);
 BALLOON_SHOW(driver_kb, "%lu\n", PAGES2KB(balloon_stats.driver_pages));
 
-static ssize_t show_target_kb(struct sys_device *dev, char *buf)
+static ssize_t show_target_kb(struct sys_device *dev, struct sysdev_attribute *attr,
+			      char *buf)
 {
 	return sprintf(buf, "%lu\n", PAGES2KB(balloon_stats.target_pages));
 }
@@ -614,19 +490,14 @@ static ssize_t store_target_kb(struct sys_device *dev,
 			       const char *buf,
 			       size_t count)
 {
-	char memstring[64], *endchar;
+	char *endchar;
 	unsigned long long target_bytes;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	if (count <= 1)
-		return -EBADMSG; /* runt */
-	if (count > sizeof(memstring))
-		return -EFBIG;   /* too long */
-	strcpy(memstring, buf);
+	target_bytes = memparse(buf, &endchar);
 
-	target_bytes = memparse(memstring, &endchar);
 	balloon_set_new_target(target_bytes >> PAGE_SHIFT);
 
 	return count;
@@ -694,20 +565,4 @@ static int register_balloon(struct sys_device *sysdev)
 	return error;
 }
 
-static void unregister_balloon(struct sys_device *sysdev)
-{
-	int i;
-
-	sysfs_remove_group(&sysdev->kobj, &balloon_info_group);
-	for (i = 0; i < ARRAY_SIZE(balloon_attrs); i++)
-		sysdev_remove_file(sysdev, balloon_attrs[i]);
-	sysdev_unregister(sysdev);
-	sysdev_class_unregister(&balloon_sysdev_class);
-}
-
-static void balloon_sysfs_exit(void)
-{
-	unregister_balloon(&balloon_sysdev);
-}
-
 MODULE_LICENSE("GPL");
diff --git a/drivers/xen/cpu_hotplug.c b/drivers/xen/cpu_hotplug.c
new file mode 100644
index 000000000000..565280ec1c6a
--- /dev/null
+++ b/drivers/xen/cpu_hotplug.c
@@ -0,0 +1,90 @@
+#include <linux/notifier.h>
+
+#include <xen/xenbus.h>
+
+#include <asm-x86/xen/hypervisor.h>
+#include <asm/cpu.h>
+
+static void enable_hotplug_cpu(int cpu)
+{
+	if (!cpu_present(cpu))
+		arch_register_cpu(cpu);
+
+	cpu_set(cpu, cpu_present_map);
+}
+
+static void disable_hotplug_cpu(int cpu)
+{
+	if (cpu_present(cpu))
+		arch_unregister_cpu(cpu);
+
+	cpu_clear(cpu, cpu_present_map);
+}
+
+static void vcpu_hotplug(unsigned int cpu)
+{
+	int err;
+	char dir[32], state[32];
+
+	if (!cpu_possible(cpu))
+		return;
+
+	sprintf(dir, "cpu/%u", cpu);
+	err = xenbus_scanf(XBT_NIL, dir, "availability", "%s", state);
+	if (err != 1) {
+		printk(KERN_ERR "XENBUS: Unable to read cpu state\n");
+		return;
+	}
+
+	if (strcmp(state, "online") == 0) {
+		enable_hotplug_cpu(cpu);
+	} else if (strcmp(state, "offline") == 0) {
+		(void)cpu_down(cpu);
+		disable_hotplug_cpu(cpu);
+	} else {
+		printk(KERN_ERR "XENBUS: unknown state(%s) on CPU%d\n",
+		       state, cpu);
+	}
+}
+
+static void handle_vcpu_hotplug_event(struct xenbus_watch *watch,
+					const char **vec, unsigned int len)
+{
+	unsigned int cpu;
+	char *cpustr;
+	const char *node = vec[XS_WATCH_PATH];
+
+	cpustr = strstr(node, "cpu/");
+	if (cpustr != NULL) {
+		sscanf(cpustr, "cpu/%u", &cpu);
+		vcpu_hotplug(cpu);
+	}
+}
+
+static int setup_cpu_watcher(struct notifier_block *notifier,
+			      unsigned long event, void *data)
+{
+	static struct xenbus_watch cpu_watch = {
+		.node = "cpu",
+		.callback = handle_vcpu_hotplug_event};
+
+	(void)register_xenbus_watch(&cpu_watch);
+
+	return NOTIFY_DONE;
+}
+
+static int __init setup_vcpu_hotplug_event(void)
+{
+	static struct notifier_block xsn_cpu = {
+		.notifier_call = setup_cpu_watcher };
+
+	if (!xen_pv_domain())
+		return -ENODEV;
+
+	register_xenstore_notifier(&xsn_cpu);
+
+	return 0;
+}
+
+arch_initcall(setup_vcpu_hotplug_event);
+
diff --git a/drivers/xen/events.c b/drivers/xen/events.c
index 0e0c28574af8..c3290bc186a0 100644
--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -84,17 +84,6 @@ static int irq_bindcount[NR_IRQS];
 /* Xen will never allocate port zero for any purpose. */
 #define VALID_EVTCHN(chn)	((chn) != 0)
 
-/*
- * Force a proper event-channel callback from Xen after clearing the
- * callback mask. We do this in a very simple manner, by making a call
- * down into Xen. The pending flag will be checked by Xen on return.
- */
-void force_evtchn_callback(void)
-{
-	(void)HYPERVISOR_xen_version(0, NULL);
-}
-EXPORT_SYMBOL_GPL(force_evtchn_callback);
-
 static struct irq_chip xen_dynamic_chip;
 
 /* Constructor for packed IRQ information. */
@@ -175,6 +164,12 @@ static inline void set_evtchn(int port)
 	sync_set_bit(port, &s->evtchn_pending[0]);
 }
 
+static inline int test_evtchn(int port)
+{
+	struct shared_info *s = HYPERVISOR_shared_info;
+	return sync_test_bit(port, &s->evtchn_pending[0]);
+}
+
 
 /**
  * notify_remote_via_irq - send event to remote end of event channel via irq
@@ -365,6 +360,10 @@ static void unbind_from_irq(unsigned int irq)
 			per_cpu(virq_to_irq, cpu_from_evtchn(evtchn))
 				[index_from_irq(irq)] = -1;
 			break;
+		case IRQT_IPI:
+			per_cpu(ipi_to_irq, cpu_from_evtchn(evtchn))
+				[index_from_irq(irq)] = -1;
+			break;
 		default:
 			break;
 		}
@@ -743,6 +742,25 @@ void xen_clear_irq_pending(int irq)
 		clear_evtchn(evtchn);
 }
 
+void xen_set_irq_pending(int irq)
+{
+	int evtchn = evtchn_from_irq(irq);
+
+	if (VALID_EVTCHN(evtchn))
+		set_evtchn(evtchn);
+}
+
+bool xen_test_irq_pending(int irq)
+{
+	int evtchn = evtchn_from_irq(irq);
+	bool ret = false;
+
+	if (VALID_EVTCHN(evtchn))
+		ret = test_evtchn(evtchn);
+
+	return ret;
+}
+
 /* Poll waiting for an irq to become pending.  In the usual case, the
    irq will be disabled so it won't deliver an interrupt. */
 void xen_poll_irq(int irq)
diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
index e9e11168616a..06592b9da83c 100644
--- a/drivers/xen/grant-table.c
+++ b/drivers/xen/grant-table.c
@@ -508,7 +508,7 @@ static int __devinit gnttab_init(void)
 	unsigned int max_nr_glist_frames, nr_glist_frames;
 	unsigned int nr_init_grefs;
 
-	if (!is_running_on_xen())
+	if (!xen_domain())
 		return -ENODEV;
 
 	nr_grant_frames = 1;
diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c
index 57ceb5346b74..7f24a98a446f 100644
--- a/drivers/xen/xenbus/xenbus_probe.c
+++ b/drivers/xen/xenbus/xenbus_probe.c
@@ -814,7 +814,7 @@ static int __init xenbus_probe_init(void)
 	DPRINTK("");
 
 	err = -ENODEV;
-	if (!is_running_on_xen())
+	if (!xen_domain())
 		goto out_error;
 
 	/* Register ourselves with the kernel bus subsystem */
@@ -829,7 +829,7 @@ static int __init xenbus_probe_init(void)
 	/*
 	 * Domain0 doesn't have a store_evtchn or store_mfn yet.
 	 */
-	if (is_initial_xendomain()) {
+	if (xen_initial_domain()) {
 		/* dom0 not yet supported */
 	} else {
 		xenstored_ready = 1;
@@ -846,7 +846,7 @@ static int __init xenbus_probe_init(void)
 		goto out_unreg_back;
 	}
 
-	if (!is_initial_xendomain())
+	if (!xen_initial_domain())
 		xenbus_probe(NULL);
 
 	return 0;
@@ -937,7 +937,7 @@ static void wait_for_devices(struct xenbus_driver *xendrv)
 	unsigned long timeout = jiffies + 10*HZ;
 	struct device_driver *drv = xendrv ? &xendrv->driver : NULL;
 
-	if (!ready_to_wait_for_devices || !is_running_on_xen())
+	if (!ready_to_wait_for_devices || !xen_domain())
 		return;
 
 	while (exists_disconnected_device(drv)) {
diff --git a/fs/Kconfig b/fs/Kconfig
index abccb5dab9a8..f54a157a0296 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -136,37 +136,51 @@ config EXT3_FS_SECURITY
 	  If you are not using a security module that requires using
 	  extended attributes for file security labels, say N.
 
-config EXT4DEV_FS
-	tristate "Ext4dev/ext4 extended fs support development (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
+config EXT4_FS
+	tristate "The Extended 4 (ext4) filesystem"
 	select JBD2
 	select CRC16
 	help
-	  Ext4dev is a predecessor filesystem of the next generation
-	  extended fs ext4, based on ext3 filesystem code. It will be
-	  renamed ext4 fs later, once ext4dev is mature and stabilized.
+	  This is the next generation of the ext3 filesystem.
 
 	  Unlike the change from ext2 filesystem to ext3 filesystem,
-	  the on-disk format of ext4dev is not the same as ext3 any more:
-	  it is based on extent maps and it supports 48-bit physical block
-	  numbers. These combined on-disk format changes will allow
-	  ext4dev/ext4 to handle more than 16 TB filesystem volumes --
-	  a hard limit that ext3 cannot overcome without changing the
-	  on-disk format.
-
-	  Other than extent maps and 48-bit block numbers, ext4dev also is
-	  likely to have other new features such as persistent preallocation,
-	  high resolution time stamps, and larger file support etc.  These
-	  features will be added to ext4dev gradually.
+	  the on-disk format of ext4 is not forwards compatible with
+	  ext3; it is based on extent maps and it supports 48-bit
+	  physical block numbers.  The ext4 filesystem also supports delayed
+	  allocation, persistent preallocation, high resolution time stamps,
+	  and a number of other features to improve performance and speed
+	  up fsck time.  For more information, please see the web pages at
+	  http://ext4.wiki.kernel.org.
+
+	  The ext4 filesystem will support mounting an ext3
+	  filesystem; while there will be some performance gains from
+	  the delayed allocation and inode table readahead, the best
+	  performance gains will require enabling ext4 features in the
+	  filesystem, or formating a new filesystem as an ext4
+	  filesystem initially.
 
 	  To compile this file system support as a module, choose M here. The
 	  module will be called ext4dev.
 
 	  If unsure, say N.
 
-config EXT4DEV_FS_XATTR
-	bool "Ext4dev extended attributes"
-	depends on EXT4DEV_FS
+config EXT4DEV_COMPAT
+	bool "Enable ext4dev compatibility"
+	depends on EXT4_FS
+	help
+	  Starting with 2.6.28, the name of the ext4 filesystem was
+	  renamed from ext4dev to ext4.  Unfortunately there are some
+	  legacy userspace programs (such as klibc's fstype) have
+	  "ext4dev" hardcoded.
+
+	  To enable backwards compatibility so that systems that are
+	  still expecting to mount ext4 filesystems using ext4dev,
+	  chose Y here.   This feature will go away by 2.6.31, so
+	  please arrange to get your userspace programs fixed!
+
+config EXT4_FS_XATTR
+	bool "Ext4 extended attributes"
+	depends on EXT4_FS
 	default y
 	help
 	  Extended attributes are name:value pairs associated with inodes by
@@ -175,11 +189,11 @@ config EXT4DEV_FS_XATTR
 
 	  If unsure, say N.
 
-	  You need this for POSIX ACL support on ext4dev/ext4.
+	  You need this for POSIX ACL support on ext4.
 
-config EXT4DEV_FS_POSIX_ACL
-	bool "Ext4dev POSIX Access Control Lists"
-	depends on EXT4DEV_FS_XATTR
+config EXT4_FS_POSIX_ACL
+	bool "Ext4 POSIX Access Control Lists"
+	depends on EXT4_FS_XATTR
 	select FS_POSIX_ACL
 	help
 	  POSIX Access Control Lists (ACLs) support permissions for users and
@@ -190,14 +204,14 @@ config EXT4DEV_FS_POSIX_ACL
 
 	  If you don't know what Access Control Lists are, say N
 
-config EXT4DEV_FS_SECURITY
-	bool "Ext4dev Security Labels"
-	depends on EXT4DEV_FS_XATTR
+config EXT4_FS_SECURITY
+	bool "Ext4 Security Labels"
+	depends on EXT4_FS_XATTR
 	help
 	  Security labels support alternative access control models
 	  implemented by security modules like SELinux.  This option
 	  enables an extended attribute handler for file security
-	  labels in the ext4dev/ext4 filesystem.
+	  labels in the ext4 filesystem.
 
 	  If you are not using a security module that requires using
 	  extended attributes for file security labels, say N.
@@ -240,22 +254,22 @@ config JBD2
 	help
 	  This is a generic journaling layer for block devices that support
 	  both 32-bit and 64-bit block numbers.  It is currently used by
-	  the ext4dev/ext4 filesystem, but it could also be used to add
+	  the ext4 filesystem, but it could also be used to add
 	  journal support to other file systems or block devices such
 	  as RAID or LVM.
 
-	  If you are using ext4dev/ext4, you need to say Y here. If you are not
-	  using ext4dev/ext4 then you will probably want to say N.
+	  If you are using ext4, you need to say Y here. If you are not
+	  using ext4 then you will probably want to say N.
 
 	  To compile this device as a module, choose M here. The module will be
-	  called jbd2.  If you are compiling ext4dev/ext4 into the kernel,
+	  called jbd2.  If you are compiling ext4 into the kernel,
 	  you cannot compile this code as a module.
 
 config JBD2_DEBUG
-	bool "JBD2 (ext4dev/ext4) debugging support"
+	bool "JBD2 (ext4) debugging support"
 	depends on JBD2 && DEBUG_FS
 	help
-	  If you are using the ext4dev/ext4 journaled file system (or
+	  If you are using the ext4 journaled file system (or
 	  potentially any other filesystem/device using JBD2), this option
 	  allows you to enable debugging output while the system is running,
 	  in order to help track down any problems you are having.
@@ -270,9 +284,9 @@ config JBD2_DEBUG
 config FS_MBCACHE
 # Meta block cache for Extended Attributes (ext2/ext3/ext4)
 	tristate
-	depends on EXT2_FS_XATTR || EXT3_FS_XATTR || EXT4DEV_FS_XATTR
-	default y if EXT2_FS=y || EXT3_FS=y || EXT4DEV_FS=y
-	default m if EXT2_FS=m || EXT3_FS=m || EXT4DEV_FS=m
+	depends on EXT2_FS_XATTR || EXT3_FS_XATTR || EXT4_FS_XATTR
+	default y if EXT2_FS=y || EXT3_FS=y || EXT4_FS=y
+	default m if EXT2_FS=m || EXT3_FS=m || EXT4_FS=m
 
 config REISERFS_FS
 	tristate "Reiserfs support"
diff --git a/fs/Kconfig.binfmt b/fs/Kconfig.binfmt
index 4a551af6f3fc..17c9c5ec14c5 100644
--- a/fs/Kconfig.binfmt
+++ b/fs/Kconfig.binfmt
@@ -59,10 +59,12 @@ config BINFMT_SHARED_FLAT
 	help
 	  Support FLAT shared libraries
 
+config HAVE_AOUT
+       def_bool n
+
 config BINFMT_AOUT
 	tristate "Kernel support for a.out and ECOFF binaries"
-	depends on ARCH_SUPPORTS_AOUT && \
-		(X86_32 || ALPHA || ARM || M68K)
+	depends on HAVE_AOUT
 	---help---
 	  A.out (Assembler.OUTput) is a set of formats for libraries and
 	  executables used in the earliest versions of UNIX.  Linux used
diff --git a/fs/Makefile b/fs/Makefile
index a1482a5eff15..de404b00eb0c 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -69,7 +69,7 @@ obj-$(CONFIG_DLM)		+= dlm/
 # Do not add any filesystems before this line
 obj-$(CONFIG_REISERFS_FS)	+= reiserfs/
 obj-$(CONFIG_EXT3_FS)		+= ext3/ # Before ext2 so root fs can be ext3
-obj-$(CONFIG_EXT4DEV_FS)	+= ext4/ # Before ext2 so root fs can be ext4dev
+obj-$(CONFIG_EXT4_FS)		+= ext4/ # Before ext2 so root fs can be ext4dev
 obj-$(CONFIG_JBD)		+= jbd/
 obj-$(CONFIG_JBD2)		+= jbd2/
 obj-$(CONFIG_EXT2_FS)		+= ext2/
diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c
index 654d972a88f4..88786ba02d27 100644
--- a/fs/cifs/misc.c
+++ b/fs/cifs/misc.c
@@ -311,8 +311,6 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ ,
 	buffer->Flags2 = SMBFLG2_KNOWS_LONG_NAMES;
 	buffer->Pid = cpu_to_le16((__u16)current->tgid);
 	buffer->PidHigh = cpu_to_le16((__u16)(current->tgid >> 16));
-	spin_lock(&GlobalMid_Lock);
-	spin_unlock(&GlobalMid_Lock);
 	if (treeCon) {
 		buffer->Tid = treeCon->tid;
 		if (treeCon->ses) {
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index 08e28c9bb416..3dbe2169cf36 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -26,8 +26,7 @@
 #include <linux/debugfs.h>
 #include <linux/fsnotify.h>
 #include <linux/string.h>
-
-#define DEBUGFS_MAGIC	0x64626720
+#include <linux/magic.h>
 
 static struct vfsmount *debugfs_mount;
 static int debugfs_mount_count;
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index 488eb424f662..a70d5d0890c7 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -27,6 +27,7 @@
 #define DEVPTS_SUPER_MAGIC 0x1cd1
 
 #define DEVPTS_DEFAULT_MODE 0600
+#define PTMX_MINOR	2
 
 extern int pty_limit;			/* Config limit on Unix98 ptys */
 static DEFINE_IDA(allocated_ptys);
@@ -169,15 +170,7 @@ static struct file_system_type devpts_fs_type = {
  * to the System V naming convention
  */
 
-static struct dentry *get_node(int num)
-{
-	char s[12];
-	struct dentry *root = devpts_root;
-	mutex_lock(&root->d_inode->i_mutex);
-	return lookup_one_len(s, root, sprintf(s, "%d", num));
-}
-
-int devpts_new_index(void)
+int devpts_new_index(struct inode *ptmx_inode)
 {
 	int index;
 	int ida_ret;
@@ -205,20 +198,21 @@ retry:
 	return index;
 }
 
-void devpts_kill_index(int idx)
+void devpts_kill_index(struct inode *ptmx_inode, int idx)
 {
 	mutex_lock(&allocated_ptys_lock);
 	ida_remove(&allocated_ptys, idx);
 	mutex_unlock(&allocated_ptys_lock);
 }
 
-int devpts_pty_new(struct tty_struct *tty)
+int devpts_pty_new(struct inode *ptmx_inode, struct tty_struct *tty)
 {
 	int number = tty->index; /* tty layer puts index from devpts_new_index() in here */
 	struct tty_driver *driver = tty->driver;
 	dev_t device = MKDEV(driver->major, driver->minor_start+number);
 	struct dentry *dentry;
 	struct inode *inode = new_inode(devpts_mnt->mnt_sb);
+	char s[12];
 
 	/* We're supposed to be given the slave end of a pty */
 	BUG_ON(driver->type != TTY_DRIVER_TYPE_PTY);
@@ -233,10 +227,15 @@ int devpts_pty_new(struct tty_struct *tty)
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
 	init_special_inode(inode, S_IFCHR|config.mode, device);
 	inode->i_private = tty;
+	tty->driver_data = inode;
 
-	dentry = get_node(number);
-	if (!IS_ERR(dentry) && !dentry->d_inode) {
-		d_instantiate(dentry, inode);
+	sprintf(s, "%d", number);
+
+	mutex_lock(&devpts_root->d_inode->i_mutex);
+
+	dentry = d_alloc_name(devpts_root, s);
+	if (!IS_ERR(dentry)) {
+		d_add(dentry, inode);
 		fsnotify_create(devpts_root->d_inode, dentry);
 	}
 
@@ -245,36 +244,31 @@ int devpts_pty_new(struct tty_struct *tty)
 	return 0;
 }
 
-struct tty_struct *devpts_get_tty(int number)
+struct tty_struct *devpts_get_tty(struct inode *pts_inode, int number)
 {
-	struct dentry *dentry = get_node(number);
-	struct tty_struct *tty;
-
-	tty = NULL;
-	if (!IS_ERR(dentry)) {
-		if (dentry->d_inode)
-			tty = dentry->d_inode->i_private;
-		dput(dentry);
-	}
+	BUG_ON(pts_inode->i_rdev == MKDEV(TTYAUX_MAJOR, PTMX_MINOR));
 
-	mutex_unlock(&devpts_root->d_inode->i_mutex);
-
-	return tty;
+	if (pts_inode->i_sb->s_magic == DEVPTS_SUPER_MAGIC)
+		return (struct tty_struct *)pts_inode->i_private;
+	return NULL;
 }
 
-void devpts_pty_kill(int number)
+void devpts_pty_kill(struct tty_struct *tty)
 {
-	struct dentry *dentry = get_node(number);
+	struct inode *inode = tty->driver_data;
+	struct dentry *dentry;
 
-	if (!IS_ERR(dentry)) {
-		struct inode *inode = dentry->d_inode;
-		if (inode) {
-			inode->i_nlink--;
-			d_delete(dentry);
-			dput(dentry);
-		}
+	BUG_ON(inode->i_rdev == MKDEV(TTYAUX_MAJOR, PTMX_MINOR));
+
+	mutex_lock(&devpts_root->d_inode->i_mutex);
+
+	dentry = d_find_alias(inode);
+	if (dentry && !IS_ERR(dentry)) {
+		inode->i_nlink--;
+		d_delete(dentry);
 		dput(dentry);
 	}
+
 	mutex_unlock(&devpts_root->d_inode->i_mutex);
 }
 
diff --git a/fs/dquot.c b/fs/dquot.c
index 8ec4d6cc7633..ad7e59003e04 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -895,10 +895,9 @@ static void print_warning(struct dquot *dquot, const int warntype)
 	    warntype == QUOTA_NL_BSOFTBELOW || !need_print_warning(dquot))
 		return;
 
-	mutex_lock(&tty_mutex);
 	tty = get_current_tty();
 	if (!tty)
-		goto out_lock;
+		return;
 	tty_write_message(tty, dquot->dq_sb->s_id);
 	if (warntype == QUOTA_NL_ISOFTWARN || warntype == QUOTA_NL_BSOFTWARN)
 		tty_write_message(tty, ": warning, ");
@@ -926,8 +925,7 @@ static void print_warning(struct dquot *dquot, const int warntype)
 			break;
 	}
 	tty_write_message(tty, msg);
-out_lock:
-	mutex_unlock(&tty_mutex);
+	tty_kref_put(tty);
 }
 #endif
 
diff --git a/fs/efs/super.c b/fs/efs/super.c
index 567b134fa1f1..73b19cfc91fc 100644
--- a/fs/efs/super.c
+++ b/fs/efs/super.c
@@ -341,8 +341,6 @@ static int efs_statfs(struct dentry *dentry, struct kstatfs *buf) {
 			sb->inode_blocks *
 			(EFS_BLOCKSIZE / sizeof(struct efs_dinode));
 	buf->f_ffree   = sb->inode_free;	/* free inodes */
-	buf->f_fsid.val[0] = (sb->fs_magic >> 16) & 0xffff; /* fs ID */
-	buf->f_fsid.val[1] =  sb->fs_magic        & 0xffff; /* fs ID */
 	buf->f_namelen = EFS_MAXNAMELEN;	/* max filename length */
 
 	return 0;
diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h
index 47d88da2d33b..bae998c1e44e 100644
--- a/fs/ext2/ext2.h
+++ b/fs/ext2/ext2.h
@@ -133,6 +133,8 @@ extern void ext2_truncate (struct inode *);
 extern int ext2_setattr (struct dentry *, struct iattr *);
 extern void ext2_set_inode_flags(struct inode *inode);
 extern void ext2_get_inode_flags(struct ext2_inode_info *);
+extern int ext2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+		       u64 start, u64 len);
 int __ext2_write_begin(struct file *file, struct address_space *mapping,
 		loff_t pos, unsigned len, unsigned flags,
 		struct page **pagep, void **fsdata);
diff --git a/fs/ext2/file.c b/fs/ext2/file.c
index 5f2fa9c36293..45ed07122182 100644
--- a/fs/ext2/file.c
+++ b/fs/ext2/file.c
@@ -86,4 +86,5 @@ const struct inode_operations ext2_file_inode_operations = {
 #endif
 	.setattr	= ext2_setattr,
 	.permission	= ext2_permission,
+	.fiemap		= ext2_fiemap,
 };
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index 991d6dfeb51f..7658b33e2653 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -31,6 +31,7 @@
 #include <linux/writeback.h>
 #include <linux/buffer_head.h>
 #include <linux/mpage.h>
+#include <linux/fiemap.h>
 #include "ext2.h"
 #include "acl.h"
 #include "xip.h"
@@ -704,6 +705,13 @@ int ext2_get_block(struct inode *inode, sector_t iblock, struct buffer_head *bh_
 
 }
 
+int ext2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+		u64 start, u64 len)
+{
+	return generic_block_fiemap(inode, fieinfo, start, len,
+				    ext2_get_block);
+}
+
 static int ext2_writepage(struct page *page, struct writeback_control *wbc)
 {
 	return block_write_full_page(page, ext2_get_block, wbc);
diff --git a/fs/ext3/file.c b/fs/ext3/file.c
index acc4913d3019..3be1e0689c9a 100644
--- a/fs/ext3/file.c
+++ b/fs/ext3/file.c
@@ -134,5 +134,6 @@ const struct inode_operations ext3_file_inode_operations = {
 	.removexattr	= generic_removexattr,
 #endif
 	.permission	= ext3_permission,
+	.fiemap		= ext3_fiemap,
 };
 
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 507d8689b111..ebfec4d0148e 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -36,6 +36,7 @@
 #include <linux/mpage.h>
 #include <linux/uio.h>
 #include <linux/bio.h>
+#include <linux/fiemap.h>
 #include "xattr.h"
 #include "acl.h"
 
@@ -981,6 +982,13 @@ out:
 	return ret;
 }
 
+int ext3_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+		u64 start, u64 len)
+{
+	return generic_block_fiemap(inode, fieinfo, start, len,
+				    ext3_get_block);
+}
+
 /*
  * `handle' can be NULL if create is zero
  */
diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile
index ac6fa8ca0a2f..a8ff003a00f7 100644
--- a/fs/ext4/Makefile
+++ b/fs/ext4/Makefile
@@ -2,12 +2,12 @@
 # Makefile for the linux ext4-filesystem routines.
 #
 
-obj-$(CONFIG_EXT4DEV_FS) += ext4dev.o
+obj-$(CONFIG_EXT4_FS) += ext4.o
 
-ext4dev-y	:= balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \
+ext4-y	:= balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \
 		   ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o \
 		   ext4_jbd2.o migrate.o mballoc.o
 
-ext4dev-$(CONFIG_EXT4DEV_FS_XATTR)	+= xattr.o xattr_user.o xattr_trusted.o
-ext4dev-$(CONFIG_EXT4DEV_FS_POSIX_ACL)	+= acl.o
-ext4dev-$(CONFIG_EXT4DEV_FS_SECURITY)	+= xattr_security.o
+ext4-$(CONFIG_EXT4_FS_XATTR)		+= xattr.o xattr_user.o xattr_trusted.o
+ext4-$(CONFIG_EXT4_FS_POSIX_ACL)	+= acl.o
+ext4-$(CONFIG_EXT4_FS_SECURITY)		+= xattr_security.o
diff --git a/fs/ext4/acl.h b/fs/ext4/acl.h
index cd2b855a07d6..cb45257a246e 100644
--- a/fs/ext4/acl.h
+++ b/fs/ext4/acl.h
@@ -51,18 +51,18 @@ static inline int ext4_acl_count(size_t size)
 	}
 }
 
-#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
+#ifdef CONFIG_EXT4_FS_POSIX_ACL
 
 /* Value for inode->u.ext4_i.i_acl and inode->u.ext4_i.i_default_acl
    if the ACL has not been cached */
 #define EXT4_ACL_NOT_CACHED ((void *)-1)
 
 /* acl.c */
-extern int ext4_permission (struct inode *, int);
-extern int ext4_acl_chmod (struct inode *);
-extern int ext4_init_acl (handle_t *, struct inode *, struct inode *);
+extern int ext4_permission(struct inode *, int);
+extern int ext4_acl_chmod(struct inode *);
+extern int ext4_init_acl(handle_t *, struct inode *, struct inode *);
 
-#else  /* CONFIG_EXT4DEV_FS_POSIX_ACL */
+#else  /* CONFIG_EXT4_FS_POSIX_ACL */
 #include <linux/sched.h>
 #define ext4_permission NULL
 
@@ -77,5 +77,5 @@ ext4_init_acl(handle_t *handle, struct inode *inode, struct inode *dir)
 {
 	return 0;
 }
-#endif  /* CONFIG_EXT4DEV_FS_POSIX_ACL */
+#endif  /* CONFIG_EXT4_FS_POSIX_ACL */
 
diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index e9fa960ba6da..bd2ece228827 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -83,6 +83,7 @@ static int ext4_group_used_meta_blocks(struct super_block *sb,
 	}
 	return used_blocks;
 }
+
 /* Initializes an uninitialized block bitmap if given, and returns the
  * number of blocks free in the group. */
 unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh,
@@ -132,7 +133,7 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh,
 		 */
 		group_blocks = ext4_blocks_count(sbi->s_es) -
 			le32_to_cpu(sbi->s_es->s_first_data_block) -
-			(EXT4_BLOCKS_PER_GROUP(sb) * (sbi->s_groups_count -1));
+			(EXT4_BLOCKS_PER_GROUP(sb) * (sbi->s_groups_count - 1));
 	} else {
 		group_blocks = EXT4_BLOCKS_PER_GROUP(sb);
 	}
@@ -200,20 +201,20 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh,
  * @bh:			pointer to the buffer head to store the block
  *			group descriptor
  */
-struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb,
+struct ext4_group_desc * ext4_get_group_desc(struct super_block *sb,
 					     ext4_group_t block_group,
-					     struct buffer_head ** bh)
+					     struct buffer_head **bh)
 {
 	unsigned long group_desc;
 	unsigned long offset;
-	struct ext4_group_desc * desc;
+	struct ext4_group_desc *desc;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 
 	if (block_group >= sbi->s_groups_count) {
-		ext4_error (sb, "ext4_get_group_desc",
-			    "block_group >= groups_count - "
-			    "block_group = %lu, groups_count = %lu",
-			    block_group, sbi->s_groups_count);
+		ext4_error(sb, "ext4_get_group_desc",
+			   "block_group >= groups_count - "
+			   "block_group = %lu, groups_count = %lu",
+			   block_group, sbi->s_groups_count);
 
 		return NULL;
 	}
@@ -222,10 +223,10 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb,
 	group_desc = block_group >> EXT4_DESC_PER_BLOCK_BITS(sb);
 	offset = block_group & (EXT4_DESC_PER_BLOCK(sb) - 1);
 	if (!sbi->s_group_desc[group_desc]) {
-		ext4_error (sb, "ext4_get_group_desc",
-			    "Group descriptor not loaded - "
-			    "block_group = %lu, group_desc = %lu, desc = %lu",
-			     block_group, group_desc, offset);
+		ext4_error(sb, "ext4_get_group_desc",
+			   "Group descriptor not loaded - "
+			   "block_group = %lu, group_desc = %lu, desc = %lu",
+			   block_group, group_desc, offset);
 		return NULL;
 	}
 
@@ -302,8 +303,8 @@ err_out:
 struct buffer_head *
 ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group)
 {
-	struct ext4_group_desc * desc;
-	struct buffer_head * bh = NULL;
+	struct ext4_group_desc *desc;
+	struct buffer_head *bh = NULL;
 	ext4_fsblk_t bitmap_blk;
 
 	desc = ext4_get_group_desc(sb, block_group, NULL);
@@ -318,9 +319,11 @@ ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group)
 			    block_group, bitmap_blk);
 		return NULL;
 	}
-	if (bh_uptodate_or_lock(bh))
+	if (buffer_uptodate(bh) &&
+	    !(desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)))
 		return bh;
 
+	lock_buffer(bh);
 	spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group));
 	if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
 		ext4_init_block_bitmap(sb, bh, block_group, desc);
@@ -345,301 +348,6 @@ ext4_read_block_bitmap(struct super_block *sb, ext4_group_t block_group)
 	 */
 	return bh;
 }
-/*
- * The reservation window structure operations
- * --------------------------------------------
- * Operations include:
- * dump, find, add, remove, is_empty, find_next_reservable_window, etc.
- *
- * We use a red-black tree to represent per-filesystem reservation
- * windows.
- *
- */
-
-/**
- * __rsv_window_dump() -- Dump the filesystem block allocation reservation map
- * @rb_root:		root of per-filesystem reservation rb tree
- * @verbose:		verbose mode
- * @fn:			function which wishes to dump the reservation map
- *
- * If verbose is turned on, it will print the whole block reservation
- * windows(start, end).	Otherwise, it will only print out the "bad" windows,
- * those windows that overlap with their immediate neighbors.
- */
-#if 1
-static void __rsv_window_dump(struct rb_root *root, int verbose,
-			      const char *fn)
-{
-	struct rb_node *n;
-	struct ext4_reserve_window_node *rsv, *prev;
-	int bad;
-
-restart:
-	n = rb_first(root);
-	bad = 0;
-	prev = NULL;
-
-	printk("Block Allocation Reservation Windows Map (%s):\n", fn);
-	while (n) {
-		rsv = rb_entry(n, struct ext4_reserve_window_node, rsv_node);
-		if (verbose)
-			printk("reservation window 0x%p "
-			       "start:  %llu, end:  %llu\n",
-			       rsv, rsv->rsv_start, rsv->rsv_end);
-		if (rsv->rsv_start && rsv->rsv_start >= rsv->rsv_end) {
-			printk("Bad reservation %p (start >= end)\n",
-			       rsv);
-			bad = 1;
-		}
-		if (prev && prev->rsv_end >= rsv->rsv_start) {
-			printk("Bad reservation %p (prev->end >= start)\n",
-			       rsv);
-			bad = 1;
-		}
-		if (bad) {
-			if (!verbose) {
-				printk("Restarting reservation walk in verbose mode\n");
-				verbose = 1;
-				goto restart;
-			}
-		}
-		n = rb_next(n);
-		prev = rsv;
-	}
-	printk("Window map complete.\n");
-	BUG_ON(bad);
-}
-#define rsv_window_dump(root, verbose) \
-	__rsv_window_dump((root), (verbose), __func__)
-#else
-#define rsv_window_dump(root, verbose) do {} while (0)
-#endif
-
-/**
- * goal_in_my_reservation()
- * @rsv:		inode's reservation window
- * @grp_goal:		given goal block relative to the allocation block group
- * @group:		the current allocation block group
- * @sb:			filesystem super block
- *
- * Test if the given goal block (group relative) is within the file's
- * own block reservation window range.
- *
- * If the reservation window is outside the goal allocation group, return 0;
- * grp_goal (given goal block) could be -1, which means no specific
- * goal block. In this case, always return 1.
- * If the goal block is within the reservation window, return 1;
- * otherwise, return 0;
- */
-static int
-goal_in_my_reservation(struct ext4_reserve_window *rsv, ext4_grpblk_t grp_goal,
-			ext4_group_t group, struct super_block *sb)
-{
-	ext4_fsblk_t group_first_block, group_last_block;
-
-	group_first_block = ext4_group_first_block_no(sb, group);
-	group_last_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1);
-
-	if ((rsv->_rsv_start > group_last_block) ||
-	    (rsv->_rsv_end < group_first_block))
-		return 0;
-	if ((grp_goal >= 0) && ((grp_goal + group_first_block < rsv->_rsv_start)
-		|| (grp_goal + group_first_block > rsv->_rsv_end)))
-		return 0;
-	return 1;
-}
-
-/**
- * search_reserve_window()
- * @rb_root:		root of reservation tree
- * @goal:		target allocation block
- *
- * Find the reserved window which includes the goal, or the previous one
- * if the goal is not in any window.
- * Returns NULL if there are no windows or if all windows start after the goal.
- */
-static struct ext4_reserve_window_node *
-search_reserve_window(struct rb_root *root, ext4_fsblk_t goal)
-{
-	struct rb_node *n = root->rb_node;
-	struct ext4_reserve_window_node *rsv;
-
-	if (!n)
-		return NULL;
-
-	do {
-		rsv = rb_entry(n, struct ext4_reserve_window_node, rsv_node);
-
-		if (goal < rsv->rsv_start)
-			n = n->rb_left;
-		else if (goal > rsv->rsv_end)
-			n = n->rb_right;
-		else
-			return rsv;
-	} while (n);
-	/*
-	 * We've fallen off the end of the tree: the goal wasn't inside
-	 * any particular node.  OK, the previous node must be to one
-	 * side of the interval containing the goal.  If it's the RHS,
-	 * we need to back up one.
-	 */
-	if (rsv->rsv_start > goal) {
-		n = rb_prev(&rsv->rsv_node);
-		rsv = rb_entry(n, struct ext4_reserve_window_node, rsv_node);
-	}
-	return rsv;
-}
-
-/**
- * ext4_rsv_window_add() -- Insert a window to the block reservation rb tree.
- * @sb:			super block
- * @rsv:		reservation window to add
- *
- * Must be called with rsv_lock hold.
- */
-void ext4_rsv_window_add(struct super_block *sb,
-		    struct ext4_reserve_window_node *rsv)
-{
-	struct rb_root *root = &EXT4_SB(sb)->s_rsv_window_root;
-	struct rb_node *node = &rsv->rsv_node;
-	ext4_fsblk_t start = rsv->rsv_start;
-
-	struct rb_node ** p = &root->rb_node;
-	struct rb_node * parent = NULL;
-	struct ext4_reserve_window_node *this;
-
-	while (*p)
-	{
-		parent = *p;
-		this = rb_entry(parent, struct ext4_reserve_window_node, rsv_node);
-
-		if (start < this->rsv_start)
-			p = &(*p)->rb_left;
-		else if (start > this->rsv_end)
-			p = &(*p)->rb_right;
-		else {
-			rsv_window_dump(root, 1);
-			BUG();
-		}
-	}
-
-	rb_link_node(node, parent, p);
-	rb_insert_color(node, root);
-}
-
-/**
- * ext4_rsv_window_remove() -- unlink a window from the reservation rb tree
- * @sb:			super block
- * @rsv:		reservation window to remove
- *
- * Mark the block reservation window as not allocated, and unlink it
- * from the filesystem reservation window rb tree. Must be called with
- * rsv_lock hold.
- */
-static void rsv_window_remove(struct super_block *sb,
-			      struct ext4_reserve_window_node *rsv)
-{
-	rsv->rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED;
-	rsv->rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED;
-	rsv->rsv_alloc_hit = 0;
-	rb_erase(&rsv->rsv_node, &EXT4_SB(sb)->s_rsv_window_root);
-}
-
-/*
- * rsv_is_empty() -- Check if the reservation window is allocated.
- * @rsv:		given reservation window to check
- *
- * returns 1 if the end block is EXT4_RESERVE_WINDOW_NOT_ALLOCATED.
- */
-static inline int rsv_is_empty(struct ext4_reserve_window *rsv)
-{
-	/* a valid reservation end block could not be 0 */
-	return rsv->_rsv_end == EXT4_RESERVE_WINDOW_NOT_ALLOCATED;
-}
-
-/**
- * ext4_init_block_alloc_info()
- * @inode:		file inode structure
- *
- * Allocate and initialize the	reservation window structure, and
- * link the window to the ext4 inode structure at last
- *
- * The reservation window structure is only dynamically allocated
- * and linked to ext4 inode the first time the open file
- * needs a new block. So, before every ext4_new_block(s) call, for
- * regular files, we should check whether the reservation window
- * structure exists or not. In the latter case, this function is called.
- * Fail to do so will result in block reservation being turned off for that
- * open file.
- *
- * This function is called from ext4_get_blocks_handle(), also called
- * when setting the reservation window size through ioctl before the file
- * is open for write (needs block allocation).
- *
- * Needs down_write(i_data_sem) protection prior to call this function.
- */
-void ext4_init_block_alloc_info(struct inode *inode)
-{
-	struct ext4_inode_info *ei = EXT4_I(inode);
-	struct ext4_block_alloc_info *block_i = ei->i_block_alloc_info;
-	struct super_block *sb = inode->i_sb;
-
-	block_i = kmalloc(sizeof(*block_i), GFP_NOFS);
-	if (block_i) {
-		struct ext4_reserve_window_node *rsv = &block_i->rsv_window_node;
-
-		rsv->rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED;
-		rsv->rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED;
-
-		/*
-		 * if filesystem is mounted with NORESERVATION, the goal
-		 * reservation window size is set to zero to indicate
-		 * block reservation is off
-		 */
-		if (!test_opt(sb, RESERVATION))
-			rsv->rsv_goal_size = 0;
-		else
-			rsv->rsv_goal_size = EXT4_DEFAULT_RESERVE_BLOCKS;
-		rsv->rsv_alloc_hit = 0;
-		block_i->last_alloc_logical_block = 0;
-		block_i->last_alloc_physical_block = 0;
-	}
-	ei->i_block_alloc_info = block_i;
-}
-
-/**
- * ext4_discard_reservation()
- * @inode:		inode
- *
- * Discard(free) block reservation window on last file close, or truncate
- * or at last iput().
- *
- * It is being called in three cases:
- *	ext4_release_file(): last writer close the file
- *	ext4_clear_inode(): last iput(), when nobody link to this file.
- *	ext4_truncate(): when the block indirect map is about to change.
- *
- */
-void ext4_discard_reservation(struct inode *inode)
-{
-	struct ext4_inode_info *ei = EXT4_I(inode);
-	struct ext4_block_alloc_info *block_i = ei->i_block_alloc_info;
-	struct ext4_reserve_window_node *rsv;
-	spinlock_t *rsv_lock = &EXT4_SB(inode->i_sb)->s_rsv_window_lock;
-
-	ext4_mb_discard_inode_preallocations(inode);
-
-	if (!block_i)
-		return;
-
-	rsv = &block_i->rsv_window_node;
-	if (!rsv_is_empty(&rsv->rsv_window)) {
-		spin_lock(rsv_lock);
-		if (!rsv_is_empty(&rsv->rsv_window))
-			rsv_window_remove(inode->i_sb, rsv);
-		spin_unlock(rsv_lock);
-	}
-}
 
 /**
  * ext4_free_blocks_sb() -- Free given blocks and update quota
@@ -648,6 +356,13 @@ void ext4_discard_reservation(struct inode *inode)
  * @block:			start physcial block to free
  * @count:			number of blocks to free
  * @pdquot_freed_blocks:	pointer to quota
+ *
+ * XXX This function is only used by the on-line resizing code, which
+ * should probably be fixed up to call the mballoc variant.  There
+ * this needs to be cleaned up later; in fact, I'm not convinced this
+ * is 100% correct in the face of the mballoc code.  The online resizing
+ * code needs to be fixed up to more tightly (and correctly) interlock
+ * with the mballoc code.
  */
 void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb,
 			 ext4_fsblk_t block, unsigned long count,
@@ -659,8 +374,8 @@ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb,
 	ext4_grpblk_t bit;
 	unsigned long i;
 	unsigned long overflow;
-	struct ext4_group_desc * desc;
-	struct ext4_super_block * es;
+	struct ext4_group_desc *desc;
+	struct ext4_super_block *es;
 	struct ext4_sb_info *sbi;
 	int err = 0, ret;
 	ext4_grpblk_t group_freed;
@@ -671,13 +386,13 @@ void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb,
 	if (block < le32_to_cpu(es->s_first_data_block) ||
 	    block + count < block ||
 	    block + count > ext4_blocks_count(es)) {
-		ext4_error (sb, "ext4_free_blocks",
-			    "Freeing blocks not in datazone - "
-			    "block = %llu, count = %lu", block, count);
+		ext4_error(sb, "ext4_free_blocks",
+			   "Freeing blocks not in datazone - "
+			   "block = %llu, count = %lu", block, count);
 		goto error_return;
 	}
 
-	ext4_debug ("freeing block(s) %llu-%llu\n", block, block + count - 1);
+	ext4_debug("freeing block(s) %llu-%llu\n", block, block + count - 1);
 
 do_more:
 	overflow = 0;
@@ -694,7 +409,7 @@ do_more:
 	bitmap_bh = ext4_read_block_bitmap(sb, block_group);
 	if (!bitmap_bh)
 		goto error_return;
-	desc = ext4_get_group_desc (sb, block_group, &gd_bh);
+	desc = ext4_get_group_desc(sb, block_group, &gd_bh);
 	if (!desc)
 		goto error_return;
 
@@ -703,10 +418,10 @@ do_more:
 	    in_range(block, ext4_inode_table(sb, desc), sbi->s_itb_per_group) ||
 	    in_range(block + count - 1, ext4_inode_table(sb, desc),
 		     sbi->s_itb_per_group)) {
-		ext4_error (sb, "ext4_free_blocks",
-			    "Freeing blocks in system zones - "
-			    "Block = %llu, count = %lu",
-			    block, count);
+		ext4_error(sb, "ext4_free_blocks",
+			   "Freeing blocks in system zones - "
+			   "Block = %llu, count = %lu",
+			   block, count);
 		goto error_return;
 	}
 
@@ -848,7 +563,7 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
 			ext4_fsblk_t block, unsigned long count,
 			int metadata)
 {
-	struct super_block * sb;
+	struct super_block *sb;
 	unsigned long dquot_freed_blocks;
 
 	/* this isn't the right place to decide whether block is metadata
@@ -859,748 +574,52 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
 
 	sb = inode->i_sb;
 
-	if (!test_opt(sb, MBALLOC) || !EXT4_SB(sb)->s_group_info)
-		ext4_free_blocks_sb(handle, sb, block, count,
-						&dquot_freed_blocks);
-	else
-		ext4_mb_free_blocks(handle, inode, block, count,
-						metadata, &dquot_freed_blocks);
+	ext4_mb_free_blocks(handle, inode, block, count,
+			    metadata, &dquot_freed_blocks);
 	if (dquot_freed_blocks)
 		DQUOT_FREE_BLOCK(inode, dquot_freed_blocks);
 	return;
 }
 
-/**
- * ext4_test_allocatable()
- * @nr:			given allocation block group
- * @bh:			bufferhead contains the bitmap of the given block group
- *
- * For ext4 allocations, we must not reuse any blocks which are
- * allocated in the bitmap buffer's "last committed data" copy.  This
- * prevents deletes from freeing up the page for reuse until we have
- * committed the delete transaction.
- *
- * If we didn't do this, then deleting something and reallocating it as
- * data would allow the old block to be overwritten before the
- * transaction committed (because we force data to disk before commit).
- * This would lead to corruption if we crashed between overwriting the
- * data and committing the delete.
- *
- * @@@ We may want to make this allocation behaviour conditional on
- * data-writes at some point, and disable it for metadata allocations or
- * sync-data inodes.
- */
-static int ext4_test_allocatable(ext4_grpblk_t nr, struct buffer_head *bh)
-{
-	int ret;
-	struct journal_head *jh = bh2jh(bh);
-
-	if (ext4_test_bit(nr, bh->b_data))
-		return 0;
-
-	jbd_lock_bh_state(bh);
-	if (!jh->b_committed_data)
-		ret = 1;
-	else
-		ret = !ext4_test_bit(nr, jh->b_committed_data);
-	jbd_unlock_bh_state(bh);
-	return ret;
-}
-
-/**
- * bitmap_search_next_usable_block()
- * @start:		the starting block (group relative) of the search
- * @bh:			bufferhead contains the block group bitmap
- * @maxblocks:		the ending block (group relative) of the reservation
- *
- * The bitmap search --- search forward alternately through the actual
- * bitmap on disk and the last-committed copy in journal, until we find a
- * bit free in both bitmaps.
- */
-static ext4_grpblk_t
-bitmap_search_next_usable_block(ext4_grpblk_t start, struct buffer_head *bh,
-					ext4_grpblk_t maxblocks)
+int ext4_claim_free_blocks(struct ext4_sb_info *sbi,
+						s64 nblocks)
 {
-	ext4_grpblk_t next;
-	struct journal_head *jh = bh2jh(bh);
-
-	while (start < maxblocks) {
-		next = ext4_find_next_zero_bit(bh->b_data, maxblocks, start);
-		if (next >= maxblocks)
-			return -1;
-		if (ext4_test_allocatable(next, bh))
-			return next;
-		jbd_lock_bh_state(bh);
-		if (jh->b_committed_data)
-			start = ext4_find_next_zero_bit(jh->b_committed_data,
-							maxblocks, next);
-		jbd_unlock_bh_state(bh);
-	}
-	return -1;
-}
+	s64 free_blocks, dirty_blocks;
+	s64 root_blocks = 0;
+	struct percpu_counter *fbc = &sbi->s_freeblocks_counter;
+	struct percpu_counter *dbc = &sbi->s_dirtyblocks_counter;
 
-/**
- * find_next_usable_block()
- * @start:		the starting block (group relative) to find next
- *			allocatable block in bitmap.
- * @bh:			bufferhead contains the block group bitmap
- * @maxblocks:		the ending block (group relative) for the search
- *
- * Find an allocatable block in a bitmap.  We honor both the bitmap and
- * its last-committed copy (if that exists), and perform the "most
- * appropriate allocation" algorithm of looking for a free block near
- * the initial goal; then for a free byte somewhere in the bitmap; then
- * for any free bit in the bitmap.
- */
-static ext4_grpblk_t
-find_next_usable_block(ext4_grpblk_t start, struct buffer_head *bh,
-			ext4_grpblk_t maxblocks)
-{
-	ext4_grpblk_t here, next;
-	char *p, *r;
-
-	if (start > 0) {
-		/*
-		 * The goal was occupied; search forward for a free
-		 * block within the next XX blocks.
-		 *
-		 * end_goal is more or less random, but it has to be
-		 * less than EXT4_BLOCKS_PER_GROUP. Aligning up to the
-		 * next 64-bit boundary is simple..
-		 */
-		ext4_grpblk_t end_goal = (start + 63) & ~63;
-		if (end_goal > maxblocks)
-			end_goal = maxblocks;
-		here = ext4_find_next_zero_bit(bh->b_data, end_goal, start);
-		if (here < end_goal && ext4_test_allocatable(here, bh))
-			return here;
-		ext4_debug("Bit not found near goal\n");
-	}
-
-	here = start;
-	if (here < 0)
-		here = 0;
-
-	p = ((char *)bh->b_data) + (here >> 3);
-	r = memscan(p, 0, ((maxblocks + 7) >> 3) - (here >> 3));
-	next = (r - ((char *)bh->b_data)) << 3;
-
-	if (next < maxblocks && next >= start && ext4_test_allocatable(next, bh))
-		return next;
-
-	/*
-	 * The bitmap search --- search forward alternately through the actual
-	 * bitmap and the last-committed copy until we find a bit free in
-	 * both
-	 */
-	here = bitmap_search_next_usable_block(here, bh, maxblocks);
-	return here;
-}
-
-/**
- * claim_block()
- * @block:		the free block (group relative) to allocate
- * @bh:			the bufferhead containts the block group bitmap
- *
- * We think we can allocate this block in this bitmap.  Try to set the bit.
- * If that succeeds then check that nobody has allocated and then freed the
- * block since we saw that is was not marked in b_committed_data.  If it _was_
- * allocated and freed then clear the bit in the bitmap again and return
- * zero (failure).
- */
-static inline int
-claim_block(spinlock_t *lock, ext4_grpblk_t block, struct buffer_head *bh)
-{
-	struct journal_head *jh = bh2jh(bh);
-	int ret;
-
-	if (ext4_set_bit_atomic(lock, block, bh->b_data))
-		return 0;
-	jbd_lock_bh_state(bh);
-	if (jh->b_committed_data && ext4_test_bit(block,jh->b_committed_data)) {
-		ext4_clear_bit_atomic(lock, block, bh->b_data);
-		ret = 0;
-	} else {
-		ret = 1;
-	}
-	jbd_unlock_bh_state(bh);
-	return ret;
-}
+	free_blocks  = percpu_counter_read_positive(fbc);
+	dirty_blocks = percpu_counter_read_positive(dbc);
 
-/**
- * ext4_try_to_allocate()
- * @sb:			superblock
- * @handle:		handle to this transaction
- * @group:		given allocation block group
- * @bitmap_bh:		bufferhead holds the block bitmap
- * @grp_goal:		given target block within the group
- * @count:		target number of blocks to allocate
- * @my_rsv:		reservation window
- *
- * Attempt to allocate blocks within a give range. Set the range of allocation
- * first, then find the first free bit(s) from the bitmap (within the range),
- * and at last, allocate the blocks by claiming the found free bit as allocated.
- *
- * To set the range of this allocation:
- *	if there is a reservation window, only try to allocate block(s) from the
- *	file's own reservation window;
- *	Otherwise, the allocation range starts from the give goal block, ends at
- *	the block group's last block.
- *
- * If we failed to allocate the desired block then we may end up crossing to a
- * new bitmap.  In that case we must release write access to the old one via
- * ext4_journal_release_buffer(), else we'll run out of credits.
- */
-static ext4_grpblk_t
-ext4_try_to_allocate(struct super_block *sb, handle_t *handle,
-			ext4_group_t group, struct buffer_head *bitmap_bh,
-			ext4_grpblk_t grp_goal, unsigned long *count,
-			struct ext4_reserve_window *my_rsv)
-{
-	ext4_fsblk_t group_first_block;
-	ext4_grpblk_t start, end;
-	unsigned long num = 0;
-
-	/* we do allocation within the reservation window if we have a window */
-	if (my_rsv) {
-		group_first_block = ext4_group_first_block_no(sb, group);
-		if (my_rsv->_rsv_start >= group_first_block)
-			start = my_rsv->_rsv_start - group_first_block;
-		else
-			/* reservation window cross group boundary */
-			start = 0;
-		end = my_rsv->_rsv_end - group_first_block + 1;
-		if (end > EXT4_BLOCKS_PER_GROUP(sb))
-			/* reservation window crosses group boundary */
-			end = EXT4_BLOCKS_PER_GROUP(sb);
-		if ((start <= grp_goal) && (grp_goal < end))
-			start = grp_goal;
-		else
-			grp_goal = -1;
-	} else {
-		if (grp_goal > 0)
-			start = grp_goal;
-		else
-			start = 0;
-		end = EXT4_BLOCKS_PER_GROUP(sb);
-	}
-
-	BUG_ON(start > EXT4_BLOCKS_PER_GROUP(sb));
-
-repeat:
-	if (grp_goal < 0 || !ext4_test_allocatable(grp_goal, bitmap_bh)) {
-		grp_goal = find_next_usable_block(start, bitmap_bh, end);
-		if (grp_goal < 0)
-			goto fail_access;
-		if (!my_rsv) {
-			int i;
-
-			for (i = 0; i < 7 && grp_goal > start &&
-					ext4_test_allocatable(grp_goal - 1,
-								bitmap_bh);
-					i++, grp_goal--)
-				;
-		}
-	}
-	start = grp_goal;
-
-	if (!claim_block(sb_bgl_lock(EXT4_SB(sb), group),
-		grp_goal, bitmap_bh)) {
-		/*
-		 * The block was allocated by another thread, or it was
-		 * allocated and then freed by another thread
-		 */
-		start++;
-		grp_goal++;
-		if (start >= end)
-			goto fail_access;
-		goto repeat;
-	}
-	num++;
-	grp_goal++;
-	while (num < *count && grp_goal < end
-		&& ext4_test_allocatable(grp_goal, bitmap_bh)
-		&& claim_block(sb_bgl_lock(EXT4_SB(sb), group),
-				grp_goal, bitmap_bh)) {
-		num++;
-		grp_goal++;
-	}
-	*count = num;
-	return grp_goal - num;
-fail_access:
-	*count = num;
-	return -1;
-}
-
-/**
- *	find_next_reservable_window():
- *		find a reservable space within the given range.
- *		It does not allocate the reservation window for now:
- *		alloc_new_reservation() will do the work later.
- *
- *	@search_head: the head of the searching list;
- *		This is not necessarily the list head of the whole filesystem
- *
- *		We have both head and start_block to assist the search
- *		for the reservable space. The list starts from head,
- *		but we will shift to the place where start_block is,
- *		then start from there, when looking for a reservable space.
- *
- *	@size: the target new reservation window size
- *
- *	@group_first_block: the first block we consider to start
- *			the real search from
- *
- *	@last_block:
- *		the maximum block number that our goal reservable space
- *		could start from. This is normally the last block in this
- *		group. The search will end when we found the start of next
- *		possible reservable space is out of this boundary.
- *		This could handle the cross boundary reservation window
- *		request.
- *
- *	basically we search from the given range, rather than the whole
- *	reservation double linked list, (start_block, last_block)
- *	to find a free region that is of my size and has not
- *	been reserved.
- *
- */
-static int find_next_reservable_window(
-				struct ext4_reserve_window_node *search_head,
-				struct ext4_reserve_window_node *my_rsv,
-				struct super_block * sb,
-				ext4_fsblk_t start_block,
-				ext4_fsblk_t last_block)
-{
-	struct rb_node *next;
-	struct ext4_reserve_window_node *rsv, *prev;
-	ext4_fsblk_t cur;
-	int size = my_rsv->rsv_goal_size;
-
-	/* TODO: make the start of the reservation window byte-aligned */
-	/* cur = *start_block & ~7;*/
-	cur = start_block;
-	rsv = search_head;
-	if (!rsv)
-		return -1;
-
-	while (1) {
-		if (cur <= rsv->rsv_end)
-			cur = rsv->rsv_end + 1;
-
-		/* TODO?
-		 * in the case we could not find a reservable space
-		 * that is what is expected, during the re-search, we could
-		 * remember what's the largest reservable space we could have
-		 * and return that one.
-		 *
-		 * For now it will fail if we could not find the reservable
-		 * space with expected-size (or more)...
-		 */
-		if (cur > last_block)
-			return -1;		/* fail */
-
-		prev = rsv;
-		next = rb_next(&rsv->rsv_node);
-		rsv = rb_entry(next,struct ext4_reserve_window_node,rsv_node);
+	if (!capable(CAP_SYS_RESOURCE) &&
+		sbi->s_resuid != current->fsuid &&
+		(sbi->s_resgid == 0 || !in_group_p(sbi->s_resgid)))
+		root_blocks = ext4_r_blocks_count(sbi->s_es);
 
-		/*
-		 * Reached the last reservation, we can just append to the
-		 * previous one.
-		 */
-		if (!next)
-			break;
-
-		if (cur + size <= rsv->rsv_start) {
-			/*
-			 * Found a reserveable space big enough.  We could
-			 * have a reservation across the group boundary here
-			 */
-			break;
+	if (free_blocks - (nblocks + root_blocks + dirty_blocks) <
+						EXT4_FREEBLOCKS_WATERMARK) {
+		free_blocks  = percpu_counter_sum(fbc);
+		dirty_blocks = percpu_counter_sum(dbc);
+		if (dirty_blocks < 0) {
+			printk(KERN_CRIT "Dirty block accounting "
+					"went wrong %lld\n",
+					dirty_blocks);
 		}
 	}
-	/*
-	 * we come here either :
-	 * when we reach the end of the whole list,
-	 * and there is empty reservable space after last entry in the list.
-	 * append it to the end of the list.
-	 *
-	 * or we found one reservable space in the middle of the list,
-	 * return the reservation window that we could append to.
-	 * succeed.
+	/* Check whether we have space after
+	 * accounting for current dirty blocks
 	 */
+	if (free_blocks < ((root_blocks + nblocks) + dirty_blocks))
+		/* we don't have free space */
+		return -ENOSPC;
 
-	if ((prev != my_rsv) && (!rsv_is_empty(&my_rsv->rsv_window)))
-		rsv_window_remove(sb, my_rsv);
-
-	/*
-	 * Let's book the whole avaliable window for now.  We will check the
-	 * disk bitmap later and then, if there are free blocks then we adjust
-	 * the window size if it's larger than requested.
-	 * Otherwise, we will remove this node from the tree next time
-	 * call find_next_reservable_window.
-	 */
-	my_rsv->rsv_start = cur;
-	my_rsv->rsv_end = cur + size - 1;
-	my_rsv->rsv_alloc_hit = 0;
-
-	if (prev != my_rsv)
-		ext4_rsv_window_add(sb, my_rsv);
-
+	/* Add the blocks to nblocks */
+	percpu_counter_add(dbc, nblocks);
 	return 0;
 }
 
 /**
- *	alloc_new_reservation()--allocate a new reservation window
- *
- *		To make a new reservation, we search part of the filesystem
- *		reservation list (the list that inside the group). We try to
- *		allocate a new reservation window near the allocation goal,
- *		or the beginning of the group, if there is no goal.
- *
- *		We first find a reservable space after the goal, then from
- *		there, we check the bitmap for the first free block after
- *		it. If there is no free block until the end of group, then the
- *		whole group is full, we failed. Otherwise, check if the free
- *		block is inside the expected reservable space, if so, we
- *		succeed.
- *		If the first free block is outside the reservable space, then
- *		start from the first free block, we search for next available
- *		space, and go on.
- *
- *	on succeed, a new reservation will be found and inserted into the list
- *	It contains at least one free block, and it does not overlap with other
- *	reservation windows.
- *
- *	failed: we failed to find a reservation window in this group
- *
- *	@rsv: the reservation
- *
- *	@grp_goal: The goal (group-relative).  It is where the search for a
- *		free reservable space should start from.
- *		if we have a grp_goal(grp_goal >0 ), then start from there,
- *		no grp_goal(grp_goal = -1), we start from the first block
- *		of the group.
- *
- *	@sb: the super block
- *	@group: the group we are trying to allocate in
- *	@bitmap_bh: the block group block bitmap
- *
- */
-static int alloc_new_reservation(struct ext4_reserve_window_node *my_rsv,
-		ext4_grpblk_t grp_goal, struct super_block *sb,
-		ext4_group_t group, struct buffer_head *bitmap_bh)
-{
-	struct ext4_reserve_window_node *search_head;
-	ext4_fsblk_t group_first_block, group_end_block, start_block;
-	ext4_grpblk_t first_free_block;
-	struct rb_root *fs_rsv_root = &EXT4_SB(sb)->s_rsv_window_root;
-	unsigned long size;
-	int ret;
-	spinlock_t *rsv_lock = &EXT4_SB(sb)->s_rsv_window_lock;
-
-	group_first_block = ext4_group_first_block_no(sb, group);
-	group_end_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1);
-
-	if (grp_goal < 0)
-		start_block = group_first_block;
-	else
-		start_block = grp_goal + group_first_block;
-
-	size = my_rsv->rsv_goal_size;
-
-	if (!rsv_is_empty(&my_rsv->rsv_window)) {
-		/*
-		 * if the old reservation is cross group boundary
-		 * and if the goal is inside the old reservation window,
-		 * we will come here when we just failed to allocate from
-		 * the first part of the window. We still have another part
-		 * that belongs to the next group. In this case, there is no
-		 * point to discard our window and try to allocate a new one
-		 * in this group(which will fail). we should
-		 * keep the reservation window, just simply move on.
-		 *
-		 * Maybe we could shift the start block of the reservation
-		 * window to the first block of next group.
-		 */
-
-		if ((my_rsv->rsv_start <= group_end_block) &&
-				(my_rsv->rsv_end > group_end_block) &&
-				(start_block >= my_rsv->rsv_start))
-			return -1;
-
-		if ((my_rsv->rsv_alloc_hit >
-		     (my_rsv->rsv_end - my_rsv->rsv_start + 1) / 2)) {
-			/*
-			 * if the previously allocation hit ratio is
-			 * greater than 1/2, then we double the size of
-			 * the reservation window the next time,
-			 * otherwise we keep the same size window
-			 */
-			size = size * 2;
-			if (size > EXT4_MAX_RESERVE_BLOCKS)
-				size = EXT4_MAX_RESERVE_BLOCKS;
-			my_rsv->rsv_goal_size= size;
-		}
-	}
-
-	spin_lock(rsv_lock);
-	/*
-	 * shift the search start to the window near the goal block
-	 */
-	search_head = search_reserve_window(fs_rsv_root, start_block);
-
-	/*
-	 * find_next_reservable_window() simply finds a reservable window
-	 * inside the given range(start_block, group_end_block).
-	 *
-	 * To make sure the reservation window has a free bit inside it, we
-	 * need to check the bitmap after we found a reservable window.
-	 */
-retry:
-	ret = find_next_reservable_window(search_head, my_rsv, sb,
-						start_block, group_end_block);
-
-	if (ret == -1) {
-		if (!rsv_is_empty(&my_rsv->rsv_window))
-			rsv_window_remove(sb, my_rsv);
-		spin_unlock(rsv_lock);
-		return -1;
-	}
-
-	/*
-	 * On success, find_next_reservable_window() returns the
-	 * reservation window where there is a reservable space after it.
-	 * Before we reserve this reservable space, we need
-	 * to make sure there is at least a free block inside this region.
-	 *
-	 * searching the first free bit on the block bitmap and copy of
-	 * last committed bitmap alternatively, until we found a allocatable
-	 * block. Search start from the start block of the reservable space
-	 * we just found.
-	 */
-	spin_unlock(rsv_lock);
-	first_free_block = bitmap_search_next_usable_block(
-			my_rsv->rsv_start - group_first_block,
-			bitmap_bh, group_end_block - group_first_block + 1);
-
-	if (first_free_block < 0) {
-		/*
-		 * no free block left on the bitmap, no point
-		 * to reserve the space. return failed.
-		 */
-		spin_lock(rsv_lock);
-		if (!rsv_is_empty(&my_rsv->rsv_window))
-			rsv_window_remove(sb, my_rsv);
-		spin_unlock(rsv_lock);
-		return -1;		/* failed */
-	}
-
-	start_block = first_free_block + group_first_block;
-	/*
-	 * check if the first free block is within the
-	 * free space we just reserved
-	 */
-	if (start_block >= my_rsv->rsv_start && start_block <= my_rsv->rsv_end)
-		return 0;		/* success */
-	/*
-	 * if the first free bit we found is out of the reservable space
-	 * continue search for next reservable space,
-	 * start from where the free block is,
-	 * we also shift the list head to where we stopped last time
-	 */
-	search_head = my_rsv;
-	spin_lock(rsv_lock);
-	goto retry;
-}
-
-/**
- * try_to_extend_reservation()
- * @my_rsv:		given reservation window
- * @sb:			super block
- * @size:		the delta to extend
- *
- * Attempt to expand the reservation window large enough to have
- * required number of free blocks
- *
- * Since ext4_try_to_allocate() will always allocate blocks within
- * the reservation window range, if the window size is too small,
- * multiple blocks allocation has to stop at the end of the reservation
- * window. To make this more efficient, given the total number of
- * blocks needed and the current size of the window, we try to
- * expand the reservation window size if necessary on a best-effort
- * basis before ext4_new_blocks() tries to allocate blocks,
- */
-static void try_to_extend_reservation(struct ext4_reserve_window_node *my_rsv,
-			struct super_block *sb, int size)
-{
-	struct ext4_reserve_window_node *next_rsv;
-	struct rb_node *next;
-	spinlock_t *rsv_lock = &EXT4_SB(sb)->s_rsv_window_lock;
-
-	if (!spin_trylock(rsv_lock))
-		return;
-
-	next = rb_next(&my_rsv->rsv_node);
-
-	if (!next)
-		my_rsv->rsv_end += size;
-	else {
-		next_rsv = rb_entry(next, struct ext4_reserve_window_node, rsv_node);
-
-		if ((next_rsv->rsv_start - my_rsv->rsv_end - 1) >= size)
-			my_rsv->rsv_end += size;
-		else
-			my_rsv->rsv_end = next_rsv->rsv_start - 1;
-	}
-	spin_unlock(rsv_lock);
-}
-
-/**
- * ext4_try_to_allocate_with_rsv()
- * @sb:			superblock
- * @handle:		handle to this transaction
- * @group:		given allocation block group
- * @bitmap_bh:		bufferhead holds the block bitmap
- * @grp_goal:		given target block within the group
- * @count:		target number of blocks to allocate
- * @my_rsv:		reservation window
- * @errp:		pointer to store the error code
- *
- * This is the main function used to allocate a new block and its reservation
- * window.
- *
- * Each time when a new block allocation is need, first try to allocate from
- * its own reservation.  If it does not have a reservation window, instead of
- * looking for a free bit on bitmap first, then look up the reservation list to
- * see if it is inside somebody else's reservation window, we try to allocate a
- * reservation window for it starting from the goal first. Then do the block
- * allocation within the reservation window.
- *
- * This will avoid keeping on searching the reservation list again and
- * again when somebody is looking for a free block (without
- * reservation), and there are lots of free blocks, but they are all
- * being reserved.
- *
- * We use a red-black tree for the per-filesystem reservation list.
- *
- */
-static ext4_grpblk_t
-ext4_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle,
-			ext4_group_t group, struct buffer_head *bitmap_bh,
-			ext4_grpblk_t grp_goal,
-			struct ext4_reserve_window_node * my_rsv,
-			unsigned long *count, int *errp)
-{
-	ext4_fsblk_t group_first_block, group_last_block;
-	ext4_grpblk_t ret = 0;
-	int fatal;
-	unsigned long num = *count;
-
-	*errp = 0;
-
-	/*
-	 * Make sure we use undo access for the bitmap, because it is critical
-	 * that we do the frozen_data COW on bitmap buffers in all cases even
-	 * if the buffer is in BJ_Forget state in the committing transaction.
-	 */
-	BUFFER_TRACE(bitmap_bh, "get undo access for new block");
-	fatal = ext4_journal_get_undo_access(handle, bitmap_bh);
-	if (fatal) {
-		*errp = fatal;
-		return -1;
-	}
-
-	/*
-	 * we don't deal with reservation when
-	 * filesystem is mounted without reservation
-	 * or the file is not a regular file
-	 * or last attempt to allocate a block with reservation turned on failed
-	 */
-	if (my_rsv == NULL ) {
-		ret = ext4_try_to_allocate(sb, handle, group, bitmap_bh,
-						grp_goal, count, NULL);
-		goto out;
-	}
-	/*
-	 * grp_goal is a group relative block number (if there is a goal)
-	 * 0 <= grp_goal < EXT4_BLOCKS_PER_GROUP(sb)
-	 * first block is a filesystem wide block number
-	 * first block is the block number of the first block in this group
-	 */
-	group_first_block = ext4_group_first_block_no(sb, group);
-	group_last_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1);
-
-	/*
-	 * Basically we will allocate a new block from inode's reservation
-	 * window.
-	 *
-	 * We need to allocate a new reservation window, if:
-	 * a) inode does not have a reservation window; or
-	 * b) last attempt to allocate a block from existing reservation
-	 *    failed; or
-	 * c) we come here with a goal and with a reservation window
-	 *
-	 * We do not need to allocate a new reservation window if we come here
-	 * at the beginning with a goal and the goal is inside the window, or
-	 * we don't have a goal but already have a reservation window.
-	 * then we could go to allocate from the reservation window directly.
-	 */
-	while (1) {
-		if (rsv_is_empty(&my_rsv->rsv_window) || (ret < 0) ||
-			!goal_in_my_reservation(&my_rsv->rsv_window,
-						grp_goal, group, sb)) {
-			if (my_rsv->rsv_goal_size < *count)
-				my_rsv->rsv_goal_size = *count;
-			ret = alloc_new_reservation(my_rsv, grp_goal, sb,
-							group, bitmap_bh);
-			if (ret < 0)
-				break;			/* failed */
-
-			if (!goal_in_my_reservation(&my_rsv->rsv_window,
-							grp_goal, group, sb))
-				grp_goal = -1;
-		} else if (grp_goal >= 0) {
-			int curr = my_rsv->rsv_end -
-					(grp_goal + group_first_block) + 1;
-
-			if (curr < *count)
-				try_to_extend_reservation(my_rsv, sb,
-							*count - curr);
-		}
-
-		if ((my_rsv->rsv_start > group_last_block) ||
-				(my_rsv->rsv_end < group_first_block)) {
-			rsv_window_dump(&EXT4_SB(sb)->s_rsv_window_root, 1);
-			BUG();
-		}
-		ret = ext4_try_to_allocate(sb, handle, group, bitmap_bh,
-					   grp_goal, &num, &my_rsv->rsv_window);
-		if (ret >= 0) {
-			my_rsv->rsv_alloc_hit += num;
-			*count = num;
-			break;				/* succeed */
-		}
-		num = *count;
-	}
-out:
-	if (ret >= 0) {
-		BUFFER_TRACE(bitmap_bh, "journal_dirty_metadata for "
-					"bitmap block");
-		fatal = ext4_journal_dirty_metadata(handle, bitmap_bh);
-		if (fatal) {
-			*errp = fatal;
-			return -1;
-		}
-		return ret;
-	}
-
-	BUFFER_TRACE(bitmap_bh, "journal_release_buffer");
-	ext4_journal_release_buffer(handle, bitmap_bh);
-	return ret;
-}
-
-/**
  * ext4_has_free_blocks()
  * @sbi:	in-core super block structure.
  * @nblocks:	number of neeed blocks
@@ -1610,29 +629,34 @@ out:
  * On success, return nblocks
  */
 ext4_fsblk_t ext4_has_free_blocks(struct ext4_sb_info *sbi,
-						ext4_fsblk_t nblocks)
+						s64 nblocks)
 {
-	ext4_fsblk_t free_blocks;
-	ext4_fsblk_t root_blocks = 0;
+	s64 free_blocks, dirty_blocks;
+	s64 root_blocks = 0;
+	struct percpu_counter *fbc = &sbi->s_freeblocks_counter;
+	struct percpu_counter *dbc = &sbi->s_dirtyblocks_counter;
 
-	free_blocks = percpu_counter_read_positive(&sbi->s_freeblocks_counter);
+	free_blocks  = percpu_counter_read_positive(fbc);
+	dirty_blocks = percpu_counter_read_positive(dbc);
 
 	if (!capable(CAP_SYS_RESOURCE) &&
 		sbi->s_resuid != current->fsuid &&
 		(sbi->s_resgid == 0 || !in_group_p(sbi->s_resgid)))
 		root_blocks = ext4_r_blocks_count(sbi->s_es);
-#ifdef CONFIG_SMP
-	if (free_blocks - root_blocks < FBC_BATCH)
-		free_blocks =
-			percpu_counter_sum_and_set(&sbi->s_freeblocks_counter);
-#endif
-	if (free_blocks <= root_blocks)
+
+	if (free_blocks - (nblocks + root_blocks + dirty_blocks) <
+						EXT4_FREEBLOCKS_WATERMARK) {
+		free_blocks  = percpu_counter_sum(fbc);
+		dirty_blocks = percpu_counter_sum(dbc);
+	}
+	if (free_blocks <= (root_blocks + dirty_blocks))
 		/* we don't have free space */
 		return 0;
-	if (free_blocks - root_blocks < nblocks)
-		return free_blocks - root_blocks;
+
+	if (free_blocks - (root_blocks + dirty_blocks) < nblocks)
+		return free_blocks - (root_blocks + dirty_blocks);
 	return nblocks;
- }
+}
 
 
 /**
@@ -1657,303 +681,6 @@ int ext4_should_retry_alloc(struct super_block *sb, int *retries)
 	return jbd2_journal_force_commit_nested(EXT4_SB(sb)->s_journal);
 }
 
-/**
- * ext4_old_new_blocks() -- core block bitmap based block allocation function
- *
- * @handle:		handle to this transaction
- * @inode:		file inode
- * @goal:		given target block(filesystem wide)
- * @count:		target number of blocks to allocate
- * @errp:		error code
- *
- * ext4_old_new_blocks uses a goal block to assist allocation and look up
- * the block bitmap directly to do block allocation.  It tries to
- * allocate block(s) from the block group contains the goal block first. If
- * that fails, it will try to allocate block(s) from other block groups
- * without any specific goal block.
- *
- * This function is called when -o nomballoc mount option is enabled
- *
- */
-ext4_fsblk_t ext4_old_new_blocks(handle_t *handle, struct inode *inode,
-			ext4_fsblk_t goal, unsigned long *count, int *errp)
-{
-	struct buffer_head *bitmap_bh = NULL;
-	struct buffer_head *gdp_bh;
-	ext4_group_t group_no;
-	ext4_group_t goal_group;
-	ext4_grpblk_t grp_target_blk;	/* blockgroup relative goal block */
-	ext4_grpblk_t grp_alloc_blk;	/* blockgroup-relative allocated block*/
-	ext4_fsblk_t ret_block;		/* filesyetem-wide allocated block */
-	ext4_group_t bgi;			/* blockgroup iteration index */
-	int fatal = 0, err;
-	int performed_allocation = 0;
-	ext4_grpblk_t free_blocks;	/* number of free blocks in a group */
-	struct super_block *sb;
-	struct ext4_group_desc *gdp;
-	struct ext4_super_block *es;
-	struct ext4_sb_info *sbi;
-	struct ext4_reserve_window_node *my_rsv = NULL;
-	struct ext4_block_alloc_info *block_i;
-	unsigned short windowsz = 0;
-	ext4_group_t ngroups;
-	unsigned long num = *count;
-
-	sb = inode->i_sb;
-	if (!sb) {
-		*errp = -ENODEV;
-		printk("ext4_new_block: nonexistent device");
-		return 0;
-	}
-
-	sbi = EXT4_SB(sb);
-	if (!EXT4_I(inode)->i_delalloc_reserved_flag) {
-		/*
-		 * With delalloc we already reserved the blocks
-		 */
-		*count = ext4_has_free_blocks(sbi, *count);
-	}
-	if (*count == 0) {
-		*errp = -ENOSPC;
-		return 0;	/*return with ENOSPC error */
-	}
-	num = *count;
-
-	/*
-	 * Check quota for allocation of this block.
-	 */
-	if (DQUOT_ALLOC_BLOCK(inode, num)) {
-		*errp = -EDQUOT;
-		return 0;
-	}
-
-	sbi = EXT4_SB(sb);
-	es = EXT4_SB(sb)->s_es;
-	ext4_debug("goal=%llu.\n", goal);
-	/*
-	 * Allocate a block from reservation only when
-	 * filesystem is mounted with reservation(default,-o reservation), and
-	 * it's a regular file, and
-	 * the desired window size is greater than 0 (One could use ioctl
-	 * command EXT4_IOC_SETRSVSZ to set the window size to 0 to turn off
-	 * reservation on that particular file)
-	 */
-	block_i = EXT4_I(inode)->i_block_alloc_info;
-	if (block_i && ((windowsz = block_i->rsv_window_node.rsv_goal_size) > 0))
-		my_rsv = &block_i->rsv_window_node;
-
-	/*
-	 * First, test whether the goal block is free.
-	 */
-	if (goal < le32_to_cpu(es->s_first_data_block) ||
-	    goal >= ext4_blocks_count(es))
-		goal = le32_to_cpu(es->s_first_data_block);
-	ext4_get_group_no_and_offset(sb, goal, &group_no, &grp_target_blk);
-	goal_group = group_no;
-retry_alloc:
-	gdp = ext4_get_group_desc(sb, group_no, &gdp_bh);
-	if (!gdp)
-		goto io_error;
-
-	free_blocks = le16_to_cpu(gdp->bg_free_blocks_count);
-	/*
-	 * if there is not enough free blocks to make a new resevation
-	 * turn off reservation for this allocation
-	 */
-	if (my_rsv && (free_blocks < windowsz)
-		&& (rsv_is_empty(&my_rsv->rsv_window)))
-		my_rsv = NULL;
-
-	if (free_blocks > 0) {
-		bitmap_bh = ext4_read_block_bitmap(sb, group_no);
-		if (!bitmap_bh)
-			goto io_error;
-		grp_alloc_blk = ext4_try_to_allocate_with_rsv(sb, handle,
-					group_no, bitmap_bh, grp_target_blk,
-					my_rsv,	&num, &fatal);
-		if (fatal)
-			goto out;
-		if (grp_alloc_blk >= 0)
-			goto allocated;
-	}
-
-	ngroups = EXT4_SB(sb)->s_groups_count;
-	smp_rmb();
-
-	/*
-	 * Now search the rest of the groups.  We assume that
-	 * group_no and gdp correctly point to the last group visited.
-	 */
-	for (bgi = 0; bgi < ngroups; bgi++) {
-		group_no++;
-		if (group_no >= ngroups)
-			group_no = 0;
-		gdp = ext4_get_group_desc(sb, group_no, &gdp_bh);
-		if (!gdp)
-			goto io_error;
-		free_blocks = le16_to_cpu(gdp->bg_free_blocks_count);
-		/*
-		 * skip this group if the number of
-		 * free blocks is less than half of the reservation
-		 * window size.
-		 */
-		if (free_blocks <= (windowsz/2))
-			continue;
-
-		brelse(bitmap_bh);
-		bitmap_bh = ext4_read_block_bitmap(sb, group_no);
-		if (!bitmap_bh)
-			goto io_error;
-		/*
-		 * try to allocate block(s) from this group, without a goal(-1).
-		 */
-		grp_alloc_blk = ext4_try_to_allocate_with_rsv(sb, handle,
-					group_no, bitmap_bh, -1, my_rsv,
-					&num, &fatal);
-		if (fatal)
-			goto out;
-		if (grp_alloc_blk >= 0)
-			goto allocated;
-	}
-	/*
-	 * We may end up a bogus ealier ENOSPC error due to
-	 * filesystem is "full" of reservations, but
-	 * there maybe indeed free blocks avaliable on disk
-	 * In this case, we just forget about the reservations
-	 * just do block allocation as without reservations.
-	 */
-	if (my_rsv) {
-		my_rsv = NULL;
-		windowsz = 0;
-		group_no = goal_group;
-		goto retry_alloc;
-	}
-	/* No space left on the device */
-	*errp = -ENOSPC;
-	goto out;
-
-allocated:
-
-	ext4_debug("using block group %lu(%d)\n",
-			group_no, gdp->bg_free_blocks_count);
-
-	BUFFER_TRACE(gdp_bh, "get_write_access");
-	fatal = ext4_journal_get_write_access(handle, gdp_bh);
-	if (fatal)
-		goto out;
-
-	ret_block = grp_alloc_blk + ext4_group_first_block_no(sb, group_no);
-
-	if (in_range(ext4_block_bitmap(sb, gdp), ret_block, num) ||
-	    in_range(ext4_inode_bitmap(sb, gdp), ret_block, num) ||
-	    in_range(ret_block, ext4_inode_table(sb, gdp),
-		     EXT4_SB(sb)->s_itb_per_group) ||
-	    in_range(ret_block + num - 1, ext4_inode_table(sb, gdp),
-		     EXT4_SB(sb)->s_itb_per_group)) {
-		ext4_error(sb, "ext4_new_block",
-			    "Allocating block in system zone - "
-			    "blocks from %llu, length %lu",
-			     ret_block, num);
-		/*
-		 * claim_block marked the blocks we allocated
-		 * as in use. So we may want to selectively
-		 * mark some of the blocks as free
-		 */
-		goto retry_alloc;
-	}
-
-	performed_allocation = 1;
-
-#ifdef CONFIG_JBD2_DEBUG
-	{
-		struct buffer_head *debug_bh;
-
-		/* Record bitmap buffer state in the newly allocated block */
-		debug_bh = sb_find_get_block(sb, ret_block);
-		if (debug_bh) {
-			BUFFER_TRACE(debug_bh, "state when allocated");
-			BUFFER_TRACE2(debug_bh, bitmap_bh, "bitmap state");
-			brelse(debug_bh);
-		}
-	}
-	jbd_lock_bh_state(bitmap_bh);
-	spin_lock(sb_bgl_lock(sbi, group_no));
-	if (buffer_jbd(bitmap_bh) && bh2jh(bitmap_bh)->b_committed_data) {
-		int i;
-
-		for (i = 0; i < num; i++) {
-			if (ext4_test_bit(grp_alloc_blk+i,
-					bh2jh(bitmap_bh)->b_committed_data)) {
-				printk("%s: block was unexpectedly set in "
-					"b_committed_data\n", __func__);
-			}
-		}
-	}
-	ext4_debug("found bit %d\n", grp_alloc_blk);
-	spin_unlock(sb_bgl_lock(sbi, group_no));
-	jbd_unlock_bh_state(bitmap_bh);
-#endif
-
-	if (ret_block + num - 1 >= ext4_blocks_count(es)) {
-		ext4_error(sb, "ext4_new_block",
-			    "block(%llu) >= blocks count(%llu) - "
-			    "block_group = %lu, es == %p ", ret_block,
-			ext4_blocks_count(es), group_no, es);
-		goto out;
-	}
-
-	/*
-	 * It is up to the caller to add the new buffer to a journal
-	 * list of some description.  We don't know in advance whether
-	 * the caller wants to use it as metadata or data.
-	 */
-	spin_lock(sb_bgl_lock(sbi, group_no));
-	if (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT))
-		gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT);
-	le16_add_cpu(&gdp->bg_free_blocks_count, -num);
-	gdp->bg_checksum = ext4_group_desc_csum(sbi, group_no, gdp);
-	spin_unlock(sb_bgl_lock(sbi, group_no));
-	if (!EXT4_I(inode)->i_delalloc_reserved_flag)
-		percpu_counter_sub(&sbi->s_freeblocks_counter, num);
-
-	if (sbi->s_log_groups_per_flex) {
-		ext4_group_t flex_group = ext4_flex_group(sbi, group_no);
-		spin_lock(sb_bgl_lock(sbi, flex_group));
-		sbi->s_flex_groups[flex_group].free_blocks -= num;
-		spin_unlock(sb_bgl_lock(sbi, flex_group));
-	}
-
-	BUFFER_TRACE(gdp_bh, "journal_dirty_metadata for group descriptor");
-	err = ext4_journal_dirty_metadata(handle, gdp_bh);
-	if (!fatal)
-		fatal = err;
-
-	sb->s_dirt = 1;
-	if (fatal)
-		goto out;
-
-	*errp = 0;
-	brelse(bitmap_bh);
-	DQUOT_FREE_BLOCK(inode, *count-num);
-	*count = num;
-	return ret_block;
-
-io_error:
-	*errp = -EIO;
-out:
-	if (fatal) {
-		*errp = fatal;
-		ext4_std_error(sb, fatal);
-	}
-	/*
-	 * Undo the block allocation
-	 */
-	if (!performed_allocation)
-		DQUOT_FREE_BLOCK(inode, *count);
-	brelse(bitmap_bh);
-	return 0;
-}
-
 #define EXT4_META_BLOCK 0x1
 
 static ext4_fsblk_t do_blk_alloc(handle_t *handle, struct inode *inode,
@@ -1963,10 +690,6 @@ static ext4_fsblk_t do_blk_alloc(handle_t *handle, struct inode *inode,
 	struct ext4_allocation_request ar;
 	ext4_fsblk_t ret;
 
-	if (!test_opt(inode->i_sb, MBALLOC)) {
-		return ext4_old_new_blocks(handle, inode, goal, count, errp);
-	}
-
 	memset(&ar, 0, sizeof(ar));
 	/* Fill with neighbour allocated blocks */
 
@@ -2008,7 +731,7 @@ ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode,
 	/*
 	 * Account for the allocated meta blocks
 	 */
-	if (!(*errp)) {
+	if (!(*errp) && EXT4_I(inode)->i_delalloc_reserved_flag) {
 		spin_lock(&EXT4_I(inode)->i_block_reservation_lock);
 		EXT4_I(inode)->i_allocated_meta_blocks += *count;
 		spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
@@ -2093,10 +816,9 @@ ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb)
 		bitmap_count += x;
 	}
 	brelse(bitmap_bh);
-	printk("ext4_count_free_blocks: stored = %llu"
-		", computed = %llu, %llu\n",
-		ext4_free_blocks_count(es),
-		desc_count, bitmap_count);
+	printk(KERN_DEBUG "ext4_count_free_blocks: stored = %llu"
+		", computed = %llu, %llu\n", ext4_free_blocks_count(es),
+	       desc_count, bitmap_count);
 	return bitmap_count;
 #else
 	desc_count = 0;
@@ -2183,8 +905,9 @@ unsigned long ext4_bg_num_gdb(struct super_block *sb, ext4_group_t group)
 
 	if (!EXT4_HAS_INCOMPAT_FEATURE(sb,EXT4_FEATURE_INCOMPAT_META_BG) ||
 			metagroup < first_meta_bg)
-		return ext4_bg_num_gdb_nometa(sb,group);
+		return ext4_bg_num_gdb_nometa(sb, group);
 
 	return ext4_bg_num_gdb_meta(sb,group);
 
 }
+
diff --git a/fs/ext4/bitmap.c b/fs/ext4/bitmap.c
index d37ea6750454..0a7a6663c190 100644
--- a/fs/ext4/bitmap.c
+++ b/fs/ext4/bitmap.c
@@ -15,17 +15,17 @@
 
 static const int nibblemap[] = {4, 3, 3, 2, 3, 2, 2, 1, 3, 2, 2, 1, 2, 1, 1, 0};
 
-unsigned long ext4_count_free (struct buffer_head * map, unsigned int numchars)
+unsigned long ext4_count_free(struct buffer_head *map, unsigned int numchars)
 {
 	unsigned int i;
 	unsigned long sum = 0;
 
 	if (!map)
-		return (0);
+		return 0;
 	for (i = 0; i < numchars; i++)
 		sum += nibblemap[map->b_data[i] & 0xf] +
 			nibblemap[(map->b_data[i] >> 4) & 0xf];
-	return (sum);
+	return sum;
 }
 
 #endif  /*  EXT4FS_DEBUG  */
diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c
index ec8e33b45219..3ca6a2b7632d 100644
--- a/fs/ext4/dir.c
+++ b/fs/ext4/dir.c
@@ -33,10 +33,10 @@ static unsigned char ext4_filetype_table[] = {
 };
 
 static int ext4_readdir(struct file *, void *, filldir_t);
-static int ext4_dx_readdir(struct file * filp,
-			   void * dirent, filldir_t filldir);
-static int ext4_release_dir (struct inode * inode,
-				struct file * filp);
+static int ext4_dx_readdir(struct file *filp,
+			   void *dirent, filldir_t filldir);
+static int ext4_release_dir(struct inode *inode,
+				struct file *filp);
 
 const struct file_operations ext4_dir_operations = {
 	.llseek		= generic_file_llseek,
@@ -61,12 +61,12 @@ static unsigned char get_dtype(struct super_block *sb, int filetype)
 }
 
 
-int ext4_check_dir_entry (const char * function, struct inode * dir,
-			  struct ext4_dir_entry_2 * de,
-			  struct buffer_head * bh,
-			  unsigned long offset)
+int ext4_check_dir_entry(const char *function, struct inode *dir,
+			 struct ext4_dir_entry_2 *de,
+			 struct buffer_head *bh,
+			 unsigned long offset)
 {
-	const char * error_msg = NULL;
+	const char *error_msg = NULL;
 	const int rlen = ext4_rec_len_from_disk(de->rec_len);
 
 	if (rlen < EXT4_DIR_REC_LEN(1))
@@ -82,7 +82,7 @@ int ext4_check_dir_entry (const char * function, struct inode * dir,
 		error_msg = "inode out of bounds";
 
 	if (error_msg != NULL)
-		ext4_error (dir->i_sb, function,
+		ext4_error(dir->i_sb, function,
 			"bad entry in directory #%lu: %s - "
 			"offset=%lu, inode=%lu, rec_len=%d, name_len=%d",
 			dir->i_ino, error_msg, offset,
@@ -91,8 +91,8 @@ int ext4_check_dir_entry (const char * function, struct inode * dir,
 	return error_msg == NULL ? 1 : 0;
 }
 
-static int ext4_readdir(struct file * filp,
-			 void * dirent, filldir_t filldir)
+static int ext4_readdir(struct file *filp,
+			 void *dirent, filldir_t filldir)
 {
 	int error = 0;
 	unsigned long offset;
@@ -102,6 +102,7 @@ static int ext4_readdir(struct file * filp,
 	int err;
 	struct inode *inode = filp->f_path.dentry->d_inode;
 	int ret = 0;
+	int dir_has_error = 0;
 
 	sb = inode->i_sb;
 
@@ -148,9 +149,13 @@ static int ext4_readdir(struct file * filp,
 		 * of recovering data when there's a bad sector
 		 */
 		if (!bh) {
-			ext4_error (sb, "ext4_readdir",
-				"directory #%lu contains a hole at offset %lu",
-				inode->i_ino, (unsigned long)filp->f_pos);
+			if (!dir_has_error) {
+				ext4_error(sb, __func__, "directory #%lu "
+					   "contains a hole at offset %Lu",
+					   inode->i_ino,
+					   (unsigned long long) filp->f_pos);
+				dir_has_error = 1;
+			}
 			/* corrupt size?  Maybe no more blocks to read */
 			if (filp->f_pos > inode->i_blocks << 9)
 				break;
@@ -187,14 +192,14 @@ revalidate:
 		while (!error && filp->f_pos < inode->i_size
 		       && offset < sb->s_blocksize) {
 			de = (struct ext4_dir_entry_2 *) (bh->b_data + offset);
-			if (!ext4_check_dir_entry ("ext4_readdir", inode, de,
-						   bh, offset)) {
+			if (!ext4_check_dir_entry("ext4_readdir", inode, de,
+						  bh, offset)) {
 				/*
 				 * On error, skip the f_pos to the next block
 				 */
 				filp->f_pos = (filp->f_pos |
 						(sb->s_blocksize - 1)) + 1;
-				brelse (bh);
+				brelse(bh);
 				ret = stored;
 				goto out;
 			}
@@ -218,12 +223,12 @@ revalidate:
 					break;
 				if (version != filp->f_version)
 					goto revalidate;
-				stored ++;
+				stored++;
 			}
 			filp->f_pos += ext4_rec_len_from_disk(de->rec_len);
 		}
 		offset = 0;
-		brelse (bh);
+		brelse(bh);
 	}
 out:
 	return ret;
@@ -290,9 +295,9 @@ static void free_rb_tree_fname(struct rb_root *root)
 		parent = rb_parent(n);
 		fname = rb_entry(n, struct fname, rb_hash);
 		while (fname) {
-			struct fname * old = fname;
+			struct fname *old = fname;
 			fname = fname->next;
-			kfree (old);
+			kfree(old);
 		}
 		if (!parent)
 			root->rb_node = NULL;
@@ -331,7 +336,7 @@ int ext4_htree_store_dirent(struct file *dir_file, __u32 hash,
 			     struct ext4_dir_entry_2 *dirent)
 {
 	struct rb_node **p, *parent = NULL;
-	struct fname * fname, *new_fn;
+	struct fname *fname, *new_fn;
 	struct dir_private_info *info;
 	int len;
 
@@ -388,19 +393,20 @@ int ext4_htree_store_dirent(struct file *dir_file, __u32 hash,
  * for all entres on the fname linked list.  (Normally there is only
  * one entry on the linked list, unless there are 62 bit hash collisions.)
  */
-static int call_filldir(struct file * filp, void * dirent,
+static int call_filldir(struct file *filp, void *dirent,
 			filldir_t filldir, struct fname *fname)
 {
 	struct dir_private_info *info = filp->private_data;
 	loff_t	curr_pos;
 	struct inode *inode = filp->f_path.dentry->d_inode;
-	struct super_block * sb;
+	struct super_block *sb;
 	int error;
 
 	sb = inode->i_sb;
 
 	if (!fname) {
-		printk("call_filldir: called with null fname?!?\n");
+		printk(KERN_ERR "ext4: call_filldir: called with "
+		       "null fname?!?\n");
 		return 0;
 	}
 	curr_pos = hash2pos(fname->hash, fname->minor_hash);
@@ -419,8 +425,8 @@ static int call_filldir(struct file * filp, void * dirent,
 	return 0;
 }
 
-static int ext4_dx_readdir(struct file * filp,
-			 void * dirent, filldir_t filldir)
+static int ext4_dx_readdir(struct file *filp,
+			 void *dirent, filldir_t filldir)
 {
 	struct dir_private_info *info = filp->private_data;
 	struct inode *inode = filp->f_path.dentry->d_inode;
@@ -511,7 +517,7 @@ finished:
 	return 0;
 }
 
-static int ext4_release_dir (struct inode * inode, struct file * filp)
+static int ext4_release_dir(struct inode *inode, struct file *filp)
 {
 	if (filp->private_data)
 		ext4_htree_free_dir_info(filp->private_data);
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 295003241d3d..6690a41cdd9f 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -44,9 +44,9 @@
 #ifdef EXT4FS_DEBUG
 #define ext4_debug(f, a...)						\
 	do {								\
-		printk (KERN_DEBUG "EXT4-fs DEBUG (%s, %d): %s:",	\
+		printk(KERN_DEBUG "EXT4-fs DEBUG (%s, %d): %s:",	\
 			__FILE__, __LINE__, __func__);			\
-		printk (KERN_DEBUG f, ## a);				\
+		printk(KERN_DEBUG f, ## a);				\
 	} while (0)
 #else
 #define ext4_debug(f, a...)	do {} while (0)
@@ -128,7 +128,7 @@ struct ext4_allocation_request {
 #else
 # define EXT4_BLOCK_SIZE(s)		(EXT4_MIN_BLOCK_SIZE << (s)->s_log_block_size)
 #endif
-#define	EXT4_ADDR_PER_BLOCK(s)		(EXT4_BLOCK_SIZE(s) / sizeof (__u32))
+#define	EXT4_ADDR_PER_BLOCK(s)		(EXT4_BLOCK_SIZE(s) / sizeof(__u32))
 #ifdef __KERNEL__
 # define EXT4_BLOCK_SIZE_BITS(s)	((s)->s_blocksize_bits)
 #else
@@ -245,7 +245,7 @@ struct flex_groups {
 #define EXT4_RESERVED_FL		0x80000000 /* reserved for ext4 lib */
 
 #define EXT4_FL_USER_VISIBLE		0x000BDFFF /* User visible flags */
-#define EXT4_FL_USER_MODIFIABLE		0x000380FF /* User modifiable flags */
+#define EXT4_FL_USER_MODIFIABLE		0x000B80FF /* User modifiable flags */
 
 /*
  * Inode dynamic state flags
@@ -291,8 +291,6 @@ struct ext4_new_group_data {
 #define	EXT4_IOC_SETFLAGS		FS_IOC_SETFLAGS
 #define	EXT4_IOC_GETVERSION		_IOR('f', 3, long)
 #define	EXT4_IOC_SETVERSION		_IOW('f', 4, long)
-#define EXT4_IOC_GROUP_EXTEND		_IOW('f', 7, unsigned long)
-#define EXT4_IOC_GROUP_ADD		_IOW('f', 8,struct ext4_new_group_input)
 #define	EXT4_IOC_GETVERSION_OLD		FS_IOC_GETVERSION
 #define	EXT4_IOC_SETVERSION_OLD		FS_IOC_SETVERSION
 #ifdef CONFIG_JBD2_DEBUG
@@ -300,7 +298,10 @@ struct ext4_new_group_data {
 #endif
 #define EXT4_IOC_GETRSVSZ		_IOR('f', 5, long)
 #define EXT4_IOC_SETRSVSZ		_IOW('f', 6, long)
-#define EXT4_IOC_MIGRATE		_IO('f', 7)
+#define EXT4_IOC_GROUP_EXTEND		_IOW('f', 7, unsigned long)
+#define EXT4_IOC_GROUP_ADD		_IOW('f', 8, struct ext4_new_group_input)
+#define EXT4_IOC_MIGRATE		_IO('f', 9)
+ /* note ioctl 11 reserved for filesystem-independent FIEMAP ioctl */
 
 /*
  * ioctl commands in 32 bit emulation
@@ -538,8 +539,9 @@ do {									       \
 #define EXT4_MOUNT_JOURNAL_CHECKSUM	0x800000 /* Journal checksums */
 #define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT	0x1000000 /* Journal Async Commit */
 #define EXT4_MOUNT_I_VERSION            0x2000000 /* i_version support */
-#define EXT4_MOUNT_MBALLOC		0x4000000 /* Buddy allocation support */
 #define EXT4_MOUNT_DELALLOC		0x8000000 /* Delalloc support */
+#define EXT4_MOUNT_DATA_ERR_ABORT	0x10000000 /* Abort on file data write */
+
 /* Compatibility, for having both ext2_fs.h and ext4_fs.h included at once */
 #ifndef _LINUX_EXT2_FS_H
 #define clear_opt(o, opt)		o &= ~EXT4_MOUNT_##opt
@@ -667,7 +669,7 @@ struct ext4_super_block {
 };
 
 #ifdef __KERNEL__
-static inline struct ext4_sb_info * EXT4_SB(struct super_block *sb)
+static inline struct ext4_sb_info *EXT4_SB(struct super_block *sb)
 {
 	return sb->s_fs_info;
 }
@@ -725,11 +727,11 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino)
  */
 
 #define EXT4_HAS_COMPAT_FEATURE(sb,mask)			\
-	( EXT4_SB(sb)->s_es->s_feature_compat & cpu_to_le32(mask) )
+	(EXT4_SB(sb)->s_es->s_feature_compat & cpu_to_le32(mask))
 #define EXT4_HAS_RO_COMPAT_FEATURE(sb,mask)			\
-	( EXT4_SB(sb)->s_es->s_feature_ro_compat & cpu_to_le32(mask) )
+	(EXT4_SB(sb)->s_es->s_feature_ro_compat & cpu_to_le32(mask))
 #define EXT4_HAS_INCOMPAT_FEATURE(sb,mask)			\
-	( EXT4_SB(sb)->s_es->s_feature_incompat & cpu_to_le32(mask) )
+	(EXT4_SB(sb)->s_es->s_feature_incompat & cpu_to_le32(mask))
 #define EXT4_SET_COMPAT_FEATURE(sb,mask)			\
 	EXT4_SB(sb)->s_es->s_feature_compat |= cpu_to_le32(mask)
 #define EXT4_SET_RO_COMPAT_FEATURE(sb,mask)			\
@@ -789,6 +791,8 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino)
 #define	EXT4_DEF_RESUID		0
 #define	EXT4_DEF_RESGID		0
 
+#define EXT4_DEF_INODE_READAHEAD_BLKS	32
+
 /*
  * Default mount options
  */
@@ -954,6 +958,24 @@ ext4_group_first_block_no(struct super_block *sb, ext4_group_t group_no)
 void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr,
 			unsigned long *blockgrpp, ext4_grpblk_t *offsetp);
 
+extern struct proc_dir_entry *ext4_proc_root;
+
+#ifdef CONFIG_PROC_FS
+extern const struct file_operations ext4_ui_proc_fops;
+
+#define	EXT4_PROC_HANDLER(name, var)					\
+do {									\
+	proc = proc_create_data(name, mode, sbi->s_proc,		\
+				&ext4_ui_proc_fops, &sbi->s_##var);	\
+	if (proc == NULL) {						\
+		printk(KERN_ERR "EXT4-fs: can't create %s\n", name);	\
+		goto err_out;						\
+	}								\
+} while (0)
+#else
+#define EXT4_PROC_HANDLER(name, var)
+#endif
+
 /*
  * Function prototypes
  */
@@ -981,23 +1003,20 @@ extern ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode,
 extern ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode,
 					ext4_lblk_t iblock, ext4_fsblk_t goal,
 					unsigned long *count, int *errp);
-extern ext4_fsblk_t ext4_old_new_blocks(handle_t *handle, struct inode *inode,
-			ext4_fsblk_t goal, unsigned long *count, int *errp);
+extern int ext4_claim_free_blocks(struct ext4_sb_info *sbi, s64 nblocks);
 extern ext4_fsblk_t ext4_has_free_blocks(struct ext4_sb_info *sbi,
-						ext4_fsblk_t nblocks);
-extern void ext4_free_blocks (handle_t *handle, struct inode *inode,
+					 s64 nblocks);
+extern void ext4_free_blocks(handle_t *handle, struct inode *inode,
 			ext4_fsblk_t block, unsigned long count, int metadata);
-extern void ext4_free_blocks_sb (handle_t *handle, struct super_block *sb,
-				 ext4_fsblk_t block, unsigned long count,
+extern void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb,
+				ext4_fsblk_t block, unsigned long count,
 				unsigned long *pdquot_freed_blocks);
-extern ext4_fsblk_t ext4_count_free_blocks (struct super_block *);
-extern void ext4_check_blocks_bitmap (struct super_block *);
+extern ext4_fsblk_t ext4_count_free_blocks(struct super_block *);
+extern void ext4_check_blocks_bitmap(struct super_block *);
 extern struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb,
 						    ext4_group_t block_group,
 						    struct buffer_head ** bh);
 extern int ext4_should_retry_alloc(struct super_block *sb, int *retries);
-extern void ext4_init_block_alloc_info(struct inode *);
-extern void ext4_rsv_window_add(struct super_block *sb, struct ext4_reserve_window_node *rsv);
 
 /* dir.c */
 extern int ext4_check_dir_entry(const char *, struct inode *,
@@ -1009,20 +1028,20 @@ extern int ext4_htree_store_dirent(struct file *dir_file, __u32 hash,
 extern void ext4_htree_free_dir_info(struct dir_private_info *p);
 
 /* fsync.c */
-extern int ext4_sync_file (struct file *, struct dentry *, int);
+extern int ext4_sync_file(struct file *, struct dentry *, int);
 
 /* hash.c */
 extern int ext4fs_dirhash(const char *name, int len, struct
 			  dx_hash_info *hinfo);
 
 /* ialloc.c */
-extern struct inode * ext4_new_inode (handle_t *, struct inode *, int);
-extern void ext4_free_inode (handle_t *, struct inode *);
-extern struct inode * ext4_orphan_get (struct super_block *, unsigned long);
-extern unsigned long ext4_count_free_inodes (struct super_block *);
-extern unsigned long ext4_count_dirs (struct super_block *);
-extern void ext4_check_inodes_bitmap (struct super_block *);
-extern unsigned long ext4_count_free (struct buffer_head *, unsigned);
+extern struct inode * ext4_new_inode(handle_t *, struct inode *, int);
+extern void ext4_free_inode(handle_t *, struct inode *);
+extern struct inode * ext4_orphan_get(struct super_block *, unsigned long);
+extern unsigned long ext4_count_free_inodes(struct super_block *);
+extern unsigned long ext4_count_dirs(struct super_block *);
+extern void ext4_check_inodes_bitmap(struct super_block *);
+extern unsigned long ext4_count_free(struct buffer_head *, unsigned);
 
 /* mballoc.c */
 extern long ext4_mb_stats;
@@ -1032,7 +1051,7 @@ extern int ext4_mb_release(struct super_block *);
 extern ext4_fsblk_t ext4_mb_new_blocks(handle_t *,
 				struct ext4_allocation_request *, int *);
 extern int ext4_mb_reserve_blocks(struct super_block *, int);
-extern void ext4_mb_discard_inode_preallocations(struct inode *);
+extern void ext4_discard_preallocations(struct inode *);
 extern int __init init_ext4_mballoc(void);
 extern void exit_ext4_mballoc(void);
 extern void ext4_mb_free_blocks(handle_t *, struct inode *,
@@ -1050,24 +1069,25 @@ struct buffer_head *ext4_getblk(handle_t *, struct inode *,
 						ext4_lblk_t, int, int *);
 struct buffer_head *ext4_bread(handle_t *, struct inode *,
 						ext4_lblk_t, int, int *);
+int ext4_get_block(struct inode *inode, sector_t iblock,
+				struct buffer_head *bh_result, int create);
 int ext4_get_blocks_handle(handle_t *handle, struct inode *inode,
 				ext4_lblk_t iblock, unsigned long maxblocks,
 				struct buffer_head *bh_result,
 				int create, int extend_disksize);
 
 extern struct inode *ext4_iget(struct super_block *, unsigned long);
-extern int  ext4_write_inode (struct inode *, int);
-extern int  ext4_setattr (struct dentry *, struct iattr *);
+extern int  ext4_write_inode(struct inode *, int);
+extern int  ext4_setattr(struct dentry *, struct iattr *);
 extern int  ext4_getattr(struct vfsmount *mnt, struct dentry *dentry,
 				struct kstat *stat);
-extern void ext4_delete_inode (struct inode *);
-extern int  ext4_sync_inode (handle_t *, struct inode *);
-extern void ext4_discard_reservation (struct inode *);
+extern void ext4_delete_inode(struct inode *);
+extern int  ext4_sync_inode(handle_t *, struct inode *);
 extern void ext4_dirty_inode(struct inode *);
 extern int ext4_change_inode_journal_flag(struct inode *, int);
 extern int ext4_get_inode_loc(struct inode *, struct ext4_iloc *);
 extern int ext4_can_truncate(struct inode *inode);
-extern void ext4_truncate (struct inode *);
+extern void ext4_truncate(struct inode *);
 extern void ext4_set_inode_flags(struct inode *);
 extern void ext4_get_inode_flags(struct ext4_inode_info *);
 extern void ext4_set_aops(struct inode *inode);
@@ -1080,11 +1100,10 @@ extern int ext4_page_mkwrite(struct vm_area_struct *vma, struct page *page);
 
 /* ioctl.c */
 extern long ext4_ioctl(struct file *, unsigned int, unsigned long);
-extern long ext4_compat_ioctl (struct file *, unsigned int, unsigned long);
+extern long ext4_compat_ioctl(struct file *, unsigned int, unsigned long);
 
 /* migrate.c */
-extern int ext4_ext_migrate(struct inode *, struct file *, unsigned int,
-		       unsigned long);
+extern int ext4_ext_migrate(struct inode *);
 /* namei.c */
 extern int ext4_orphan_add(handle_t *, struct inode *);
 extern int ext4_orphan_del(handle_t *, struct inode *);
@@ -1099,14 +1118,14 @@ extern int ext4_group_extend(struct super_block *sb,
 				ext4_fsblk_t n_blocks_count);
 
 /* super.c */
-extern void ext4_error (struct super_block *, const char *, const char *, ...)
+extern void ext4_error(struct super_block *, const char *, const char *, ...)
 	__attribute__ ((format (printf, 3, 4)));
-extern void __ext4_std_error (struct super_block *, const char *, int);
-extern void ext4_abort (struct super_block *, const char *, const char *, ...)
+extern void __ext4_std_error(struct super_block *, const char *, int);
+extern void ext4_abort(struct super_block *, const char *, const char *, ...)
 	__attribute__ ((format (printf, 3, 4)));
-extern void ext4_warning (struct super_block *, const char *, const char *, ...)
+extern void ext4_warning(struct super_block *, const char *, const char *, ...)
 	__attribute__ ((format (printf, 3, 4)));
-extern void ext4_update_dynamic_rev (struct super_block *sb);
+extern void ext4_update_dynamic_rev(struct super_block *sb);
 extern int ext4_update_compat_feature(handle_t *handle, struct super_block *sb,
 					__u32 compat);
 extern int ext4_update_rocompat_feature(handle_t *handle,
@@ -1179,7 +1198,7 @@ static inline void ext4_isize_set(struct ext4_inode *raw_inode, loff_t i_size)
 
 static inline
 struct ext4_group_info *ext4_get_group_info(struct super_block *sb,
-							ext4_group_t group)
+					    ext4_group_t group)
 {
 	 struct ext4_group_info ***grp_info;
 	 long indexv, indexh;
@@ -1207,6 +1226,28 @@ do {								\
 		__ext4_std_error((sb), __func__, (errno));	\
 } while (0)
 
+#ifdef CONFIG_SMP
+/* Each CPU can accumulate FBC_BATCH blocks in their local
+ * counters. So we need to make sure we have free blocks more
+ * than FBC_BATCH  * nr_cpu_ids. Also add a window of 4 times.
+ */
+#define EXT4_FREEBLOCKS_WATERMARK (4 * (FBC_BATCH * nr_cpu_ids))
+#else
+#define EXT4_FREEBLOCKS_WATERMARK 0
+#endif
+
+static inline void ext4_update_i_disksize(struct inode *inode, loff_t newsize)
+{
+	/*
+	 * XXX: replace with spinlock if seen contended -bzzz
+	 */
+	down_write(&EXT4_I(inode)->i_data_sem);
+	if (newsize > EXT4_I(inode)->i_disksize)
+		EXT4_I(inode)->i_disksize = newsize;
+	up_write(&EXT4_I(inode)->i_data_sem);
+	return ;
+}
+
 /*
  * Inodes and files operations
  */
diff --git a/fs/ext4/ext4_extents.h b/fs/ext4/ext4_extents.h
index d33dc56d6986..bec7ce59fc0d 100644
--- a/fs/ext4/ext4_extents.h
+++ b/fs/ext4/ext4_extents.h
@@ -124,6 +124,19 @@ struct ext4_ext_path {
 #define EXT4_EXT_CACHE_GAP	1
 #define EXT4_EXT_CACHE_EXTENT	2
 
+/*
+ * to be called by ext4_ext_walk_space()
+ * negative retcode - error
+ * positive retcode - signal for ext4_ext_walk_space(), see below
+ * callback must return valid extent (passed or newly created)
+ */
+typedef int (*ext_prepare_callback)(struct inode *, struct ext4_ext_path *,
+					struct ext4_ext_cache *,
+					struct ext4_extent *, void *);
+
+#define EXT_CONTINUE   0
+#define EXT_BREAK      1
+#define EXT_REPEAT     2
 
 #define EXT_MAX_BLOCK	0xffffffff
 
@@ -224,6 +237,8 @@ extern int ext4_ext_try_to_merge(struct inode *inode,
 				 struct ext4_extent *);
 extern unsigned int ext4_ext_check_overlap(struct inode *, struct ext4_extent *, struct ext4_ext_path *);
 extern int ext4_ext_insert_extent(handle_t *, struct inode *, struct ext4_ext_path *, struct ext4_extent *);
+extern int ext4_ext_walk_space(struct inode *, ext4_lblk_t, ext4_lblk_t,
+							ext_prepare_callback, void *);
 extern struct ext4_ext_path *ext4_ext_find_extent(struct inode *, ext4_lblk_t,
 							struct ext4_ext_path *);
 extern int ext4_ext_search_left(struct inode *, struct ext4_ext_path *,
diff --git a/fs/ext4/ext4_i.h b/fs/ext4/ext4_i.h
index ef7409f0e7e4..5c124c0ac6d3 100644
--- a/fs/ext4/ext4_i.h
+++ b/fs/ext4/ext4_i.h
@@ -33,38 +33,6 @@ typedef __u32 ext4_lblk_t;
 /* data type for block group number */
 typedef unsigned long ext4_group_t;
 
-struct ext4_reserve_window {
-	ext4_fsblk_t	_rsv_start;	/* First byte reserved */
-	ext4_fsblk_t	_rsv_end;	/* Last byte reserved or 0 */
-};
-
-struct ext4_reserve_window_node {
-	struct rb_node		rsv_node;
-	__u32			rsv_goal_size;
-	__u32			rsv_alloc_hit;
-	struct ext4_reserve_window	rsv_window;
-};
-
-struct ext4_block_alloc_info {
-	/* information about reservation window */
-	struct ext4_reserve_window_node rsv_window_node;
-	/*
-	 * was i_next_alloc_block in ext4_inode_info
-	 * is the logical (file-relative) number of the
-	 * most-recently-allocated block in this file.
-	 * We use this for detecting linearly ascending allocation requests.
-	 */
-	ext4_lblk_t last_alloc_logical_block;
-	/*
-	 * Was i_next_alloc_goal in ext4_inode_info
-	 * is the *physical* companion to i_next_alloc_block.
-	 * it the physical block number of the block which was most-recentl
-	 * allocated to this file.  This give us the goal (target) for the next
-	 * allocation when we detect linearly ascending requests.
-	 */
-	ext4_fsblk_t last_alloc_physical_block;
-};
-
 #define rsv_start rsv_window._rsv_start
 #define rsv_end rsv_window._rsv_end
 
@@ -97,11 +65,8 @@ struct ext4_inode_info {
 	ext4_group_t	i_block_group;
 	__u32	i_state;		/* Dynamic state flags for ext4 */
 
-	/* block reservation info */
-	struct ext4_block_alloc_info *i_block_alloc_info;
-
 	ext4_lblk_t		i_dir_start_lookup;
-#ifdef CONFIG_EXT4DEV_FS_XATTR
+#ifdef CONFIG_EXT4_FS_XATTR
 	/*
 	 * Extended attributes can be read independently of the main file
 	 * data. Taking i_mutex even when reading would cause contention
@@ -111,7 +76,7 @@ struct ext4_inode_info {
 	 */
 	struct rw_semaphore xattr_sem;
 #endif
-#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
+#ifdef CONFIG_EXT4_FS_POSIX_ACL
 	struct posix_acl	*i_acl;
 	struct posix_acl	*i_default_acl;
 #endif
diff --git a/fs/ext4/ext4_sb.h b/fs/ext4/ext4_sb.h
index 6300226d5531..6a0b40d43264 100644
--- a/fs/ext4/ext4_sb.h
+++ b/fs/ext4/ext4_sb.h
@@ -40,8 +40,8 @@ struct ext4_sb_info {
 	unsigned long s_blocks_last;    /* Last seen block count */
 	loff_t s_bitmap_maxbytes;	/* max bytes for bitmap files */
 	struct buffer_head * s_sbh;	/* Buffer containing the super block */
-	struct ext4_super_block * s_es;	/* Pointer to the super block in the buffer */
-	struct buffer_head ** s_group_desc;
+	struct ext4_super_block *s_es;	/* Pointer to the super block in the buffer */
+	struct buffer_head **s_group_desc;
 	unsigned long  s_mount_opt;
 	ext4_fsblk_t s_sb_block;
 	uid_t s_resuid;
@@ -52,6 +52,7 @@ struct ext4_sb_info {
 	int s_desc_per_block_bits;
 	int s_inode_size;
 	int s_first_ino;
+	unsigned int s_inode_readahead_blks;
 	spinlock_t s_next_gen_lock;
 	u32 s_next_generation;
 	u32 s_hash_seed[4];
@@ -59,16 +60,17 @@ struct ext4_sb_info {
 	struct percpu_counter s_freeblocks_counter;
 	struct percpu_counter s_freeinodes_counter;
 	struct percpu_counter s_dirs_counter;
+	struct percpu_counter s_dirtyblocks_counter;
 	struct blockgroup_lock s_blockgroup_lock;
+	struct proc_dir_entry *s_proc;
 
 	/* root of the per fs reservation window tree */
 	spinlock_t s_rsv_window_lock;
 	struct rb_root s_rsv_window_root;
-	struct ext4_reserve_window_node s_rsv_window_head;
 
 	/* Journaling */
-	struct inode * s_journal_inode;
-	struct journal_s * s_journal;
+	struct inode *s_journal_inode;
+	struct journal_s *s_journal;
 	struct list_head s_orphan;
 	unsigned long s_commit_interval;
 	struct block_device *journal_bdev;
@@ -106,12 +108,12 @@ struct ext4_sb_info {
 
 	/* tunables */
 	unsigned long s_stripe;
-	unsigned long s_mb_stream_request;
-	unsigned long s_mb_max_to_scan;
-	unsigned long s_mb_min_to_scan;
-	unsigned long s_mb_stats;
-	unsigned long s_mb_order2_reqs;
-	unsigned long s_mb_group_prealloc;
+	unsigned int s_mb_stream_request;
+	unsigned int s_mb_max_to_scan;
+	unsigned int s_mb_min_to_scan;
+	unsigned int s_mb_stats;
+	unsigned int s_mb_order2_reqs;
+	unsigned int s_mb_group_prealloc;
 	/* where last allocation was done - for stream allocation */
 	unsigned long s_mb_last_group;
 	unsigned long s_mb_last_start;
@@ -121,7 +123,6 @@ struct ext4_sb_info {
 	int s_mb_history_cur;
 	int s_mb_history_max;
 	int s_mb_history_num;
-	struct proc_dir_entry *s_mb_proc;
 	spinlock_t s_mb_history_lock;
 	int s_mb_history_filter;
 
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index b24d3c53f20c..ea2ce3c0ae66 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -40,6 +40,7 @@
 #include <linux/slab.h>
 #include <linux/falloc.h>
 #include <asm/uaccess.h>
+#include <linux/fiemap.h>
 #include "ext4_jbd2.h"
 #include "ext4_extents.h"
 
@@ -383,8 +384,8 @@ static void ext4_ext_show_leaf(struct inode *inode, struct ext4_ext_path *path)
 	ext_debug("\n");
 }
 #else
-#define ext4_ext_show_path(inode,path)
-#define ext4_ext_show_leaf(inode,path)
+#define ext4_ext_show_path(inode, path)
+#define ext4_ext_show_leaf(inode, path)
 #endif
 
 void ext4_ext_drop_refs(struct ext4_ext_path *path)
@@ -440,9 +441,10 @@ ext4_ext_binsearch_idx(struct inode *inode,
 		for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ix++) {
 		  if (k != 0 &&
 		      le32_to_cpu(ix->ei_block) <= le32_to_cpu(ix[-1].ei_block)) {
-				printk("k=%d, ix=0x%p, first=0x%p\n", k,
-					ix, EXT_FIRST_INDEX(eh));
-				printk("%u <= %u\n",
+				printk(KERN_DEBUG "k=%d, ix=0x%p, "
+				       "first=0x%p\n", k,
+				       ix, EXT_FIRST_INDEX(eh));
+				printk(KERN_DEBUG "%u <= %u\n",
 				       le32_to_cpu(ix->ei_block),
 				       le32_to_cpu(ix[-1].ei_block));
 			}
@@ -1475,7 +1477,7 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode,
 				struct ext4_ext_path *path,
 				struct ext4_extent *newext)
 {
-	struct ext4_extent_header * eh;
+	struct ext4_extent_header *eh;
 	struct ext4_extent *ex, *fex;
 	struct ext4_extent *nearex; /* nearest extent */
 	struct ext4_ext_path *npath = NULL;
@@ -1625,6 +1627,113 @@ cleanup:
 	return err;
 }
 
+int ext4_ext_walk_space(struct inode *inode, ext4_lblk_t block,
+			ext4_lblk_t num, ext_prepare_callback func,
+			void *cbdata)
+{
+	struct ext4_ext_path *path = NULL;
+	struct ext4_ext_cache cbex;
+	struct ext4_extent *ex;
+	ext4_lblk_t next, start = 0, end = 0;
+	ext4_lblk_t last = block + num;
+	int depth, exists, err = 0;
+
+	BUG_ON(func == NULL);
+	BUG_ON(inode == NULL);
+
+	while (block < last && block != EXT_MAX_BLOCK) {
+		num = last - block;
+		/* find extent for this block */
+		path = ext4_ext_find_extent(inode, block, path);
+		if (IS_ERR(path)) {
+			err = PTR_ERR(path);
+			path = NULL;
+			break;
+		}
+
+		depth = ext_depth(inode);
+		BUG_ON(path[depth].p_hdr == NULL);
+		ex = path[depth].p_ext;
+		next = ext4_ext_next_allocated_block(path);
+
+		exists = 0;
+		if (!ex) {
+			/* there is no extent yet, so try to allocate
+			 * all requested space */
+			start = block;
+			end = block + num;
+		} else if (le32_to_cpu(ex->ee_block) > block) {
+			/* need to allocate space before found extent */
+			start = block;
+			end = le32_to_cpu(ex->ee_block);
+			if (block + num < end)
+				end = block + num;
+		} else if (block >= le32_to_cpu(ex->ee_block)
+					+ ext4_ext_get_actual_len(ex)) {
+			/* need to allocate space after found extent */
+			start = block;
+			end = block + num;
+			if (end >= next)
+				end = next;
+		} else if (block >= le32_to_cpu(ex->ee_block)) {
+			/*
+			 * some part of requested space is covered
+			 * by found extent
+			 */
+			start = block;
+			end = le32_to_cpu(ex->ee_block)
+				+ ext4_ext_get_actual_len(ex);
+			if (block + num < end)
+				end = block + num;
+			exists = 1;
+		} else {
+			BUG();
+		}
+		BUG_ON(end <= start);
+
+		if (!exists) {
+			cbex.ec_block = start;
+			cbex.ec_len = end - start;
+			cbex.ec_start = 0;
+			cbex.ec_type = EXT4_EXT_CACHE_GAP;
+		} else {
+			cbex.ec_block = le32_to_cpu(ex->ee_block);
+			cbex.ec_len = ext4_ext_get_actual_len(ex);
+			cbex.ec_start = ext_pblock(ex);
+			cbex.ec_type = EXT4_EXT_CACHE_EXTENT;
+		}
+
+		BUG_ON(cbex.ec_len == 0);
+		err = func(inode, path, &cbex, ex, cbdata);
+		ext4_ext_drop_refs(path);
+
+		if (err < 0)
+			break;
+
+		if (err == EXT_REPEAT)
+			continue;
+		else if (err == EXT_BREAK) {
+			err = 0;
+			break;
+		}
+
+		if (ext_depth(inode) != depth) {
+			/* depth was changed. we have to realloc path */
+			kfree(path);
+			path = NULL;
+		}
+
+		block = cbex.ec_block + cbex.ec_len;
+	}
+
+	if (path) {
+		ext4_ext_drop_refs(path);
+		kfree(path);
+	}
+
+	return err;
+}
+
 static void
 ext4_ext_put_in_cache(struct inode *inode, ext4_lblk_t block,
 			__u32 len, ext4_fsblk_t start, int type)
@@ -2142,7 +2251,7 @@ void ext4_ext_init(struct super_block *sb)
 	 */
 
 	if (test_opt(sb, EXTENTS)) {
-		printk("EXT4-fs: file extents enabled");
+		printk(KERN_INFO "EXT4-fs: file extents enabled");
 #ifdef AGGRESSIVE_TEST
 		printk(", aggressive tests");
 #endif
@@ -2696,11 +2805,8 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 		goto out2;
 	}
 	/*
-	 * Okay, we need to do block allocation.  Lazily initialize the block
-	 * allocation info here if necessary.
+	 * Okay, we need to do block allocation.
 	 */
-	if (S_ISREG(inode->i_mode) && (!EXT4_I(inode)->i_block_alloc_info))
-		ext4_init_block_alloc_info(inode);
 
 	/* find neighbour allocated blocks */
 	ar.lleft = iblock;
@@ -2760,7 +2866,7 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 		/* free data blocks we just allocated */
 		/* not a good idea to call discard here directly,
 		 * but otherwise we'd need to call it every free() */
-		ext4_mb_discard_inode_preallocations(inode);
+		ext4_discard_preallocations(inode);
 		ext4_free_blocks(handle, inode, ext_pblock(&newex),
 					ext4_ext_get_actual_len(&newex), 0);
 		goto out2;
@@ -2824,7 +2930,7 @@ void ext4_ext_truncate(struct inode *inode)
 	down_write(&EXT4_I(inode)->i_data_sem);
 	ext4_ext_invalidate_cache(inode);
 
-	ext4_discard_reservation(inode);
+	ext4_discard_preallocations(inode);
 
 	/*
 	 * TODO: optimization is possible here.
@@ -2877,10 +2983,11 @@ static void ext4_falloc_update_inode(struct inode *inode,
 	 * Update only when preallocation was requested beyond
 	 * the file size.
 	 */
-	if (!(mode & FALLOC_FL_KEEP_SIZE) &&
-				new_size > i_size_read(inode)) {
-		i_size_write(inode, new_size);
-		EXT4_I(inode)->i_disksize = new_size;
+	if (!(mode & FALLOC_FL_KEEP_SIZE)) {
+		if (new_size > i_size_read(inode))
+			i_size_write(inode, new_size);
+		if (new_size > EXT4_I(inode)->i_disksize)
+			ext4_update_i_disksize(inode, new_size);
 	}
 
 }
@@ -2972,3 +3079,143 @@ retry:
 	mutex_unlock(&inode->i_mutex);
 	return ret > 0 ? ret2 : ret;
 }
+
+/*
+ * Callback function called for each extent to gather FIEMAP information.
+ */
+int ext4_ext_fiemap_cb(struct inode *inode, struct ext4_ext_path *path,
+		       struct ext4_ext_cache *newex, struct ext4_extent *ex,
+		       void *data)
+{
+	struct fiemap_extent_info *fieinfo = data;
+	unsigned long blksize_bits = inode->i_sb->s_blocksize_bits;
+	__u64	logical;
+	__u64	physical;
+	__u64	length;
+	__u32	flags = 0;
+	int	error;
+
+	logical =  (__u64)newex->ec_block << blksize_bits;
+
+	if (newex->ec_type == EXT4_EXT_CACHE_GAP) {
+		pgoff_t offset;
+		struct page *page;
+		struct buffer_head *bh = NULL;
+
+		offset = logical >> PAGE_SHIFT;
+		page = find_get_page(inode->i_mapping, offset);
+		if (!page || !page_has_buffers(page))
+			return EXT_CONTINUE;
+
+		bh = page_buffers(page);
+
+		if (!bh)
+			return EXT_CONTINUE;
+
+		if (buffer_delay(bh)) {
+			flags |= FIEMAP_EXTENT_DELALLOC;
+			page_cache_release(page);
+		} else {
+			page_cache_release(page);
+			return EXT_CONTINUE;
+		}
+	}
+
+	physical = (__u64)newex->ec_start << blksize_bits;
+	length =   (__u64)newex->ec_len << blksize_bits;
+
+	if (ex && ext4_ext_is_uninitialized(ex))
+		flags |= FIEMAP_EXTENT_UNWRITTEN;
+
+	/*
+	 * If this extent reaches EXT_MAX_BLOCK, it must be last.
+	 *
+	 * Or if ext4_ext_next_allocated_block is EXT_MAX_BLOCK,
+	 * this also indicates no more allocated blocks.
+	 *
+	 * XXX this might miss a single-block extent at EXT_MAX_BLOCK
+	 */
+	if (logical + length - 1 == EXT_MAX_BLOCK ||
+	    ext4_ext_next_allocated_block(path) == EXT_MAX_BLOCK)
+		flags |= FIEMAP_EXTENT_LAST;
+
+	error = fiemap_fill_next_extent(fieinfo, logical, physical,
+					length, flags);
+	if (error < 0)
+		return error;
+	if (error == 1)
+		return EXT_BREAK;
+
+	return EXT_CONTINUE;
+}
+
+/* fiemap flags we can handle specified here */
+#define EXT4_FIEMAP_FLAGS	(FIEMAP_FLAG_SYNC|FIEMAP_FLAG_XATTR)
+
+int ext4_xattr_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo)
+{
+	__u64 physical = 0;
+	__u64 length;
+	__u32 flags = FIEMAP_EXTENT_LAST;
+	int blockbits = inode->i_sb->s_blocksize_bits;
+	int error = 0;
+
+	/* in-inode? */
+	if (EXT4_I(inode)->i_state & EXT4_STATE_XATTR) {
+		struct ext4_iloc iloc;
+		int offset;	/* offset of xattr in inode */
+
+		error = ext4_get_inode_loc(inode, &iloc);
+		if (error)
+			return error;
+		physical = iloc.bh->b_blocknr << blockbits;
+		offset = EXT4_GOOD_OLD_INODE_SIZE +
+				EXT4_I(inode)->i_extra_isize;
+		physical += offset;
+		length = EXT4_SB(inode->i_sb)->s_inode_size - offset;
+		flags |= FIEMAP_EXTENT_DATA_INLINE;
+	} else { /* external block */
+		physical = EXT4_I(inode)->i_file_acl << blockbits;
+		length = inode->i_sb->s_blocksize;
+	}
+
+	if (physical)
+		error = fiemap_fill_next_extent(fieinfo, 0, physical,
+						length, flags);
+	return (error < 0 ? error : 0);
+}
+
+int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+		__u64 start, __u64 len)
+{
+	ext4_lblk_t start_blk;
+	ext4_lblk_t len_blks;
+	int error = 0;
+
+	/* fallback to generic here if not in extents fmt */
+	if (!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL))
+		return generic_block_fiemap(inode, fieinfo, start, len,
+			ext4_get_block);
+
+	if (fiemap_check_flags(fieinfo, EXT4_FIEMAP_FLAGS))
+		return -EBADR;
+
+	if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR) {
+		error = ext4_xattr_fiemap(inode, fieinfo);
+	} else {
+		start_blk = start >> inode->i_sb->s_blocksize_bits;
+		len_blks = len >> inode->i_sb->s_blocksize_bits;
+
+		/*
+		 * Walk the extent tree gathering extent information.
+		 * ext4_ext_fiemap_cb will push extents back to user.
+		 */
+		down_write(&EXT4_I(inode)->i_data_sem);
+		error = ext4_ext_walk_space(inode, start_blk, len_blks,
+					  ext4_ext_fiemap_cb, fieinfo);
+		up_write(&EXT4_I(inode)->i_data_sem);
+	}
+
+	return error;
+}
+
diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index 430eb7978db4..6bd11fba71f7 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -31,14 +31,14 @@
  * from ext4_file_open: open gets called at every open, but release
  * gets called only when /all/ the files are closed.
  */
-static int ext4_release_file (struct inode * inode, struct file * filp)
+static int ext4_release_file(struct inode *inode, struct file *filp)
 {
 	/* if we are the last writer on the inode, drop the block reservation */
 	if ((filp->f_mode & FMODE_WRITE) &&
 			(atomic_read(&inode->i_writecount) == 1))
 	{
 		down_write(&EXT4_I(inode)->i_data_sem);
-		ext4_discard_reservation(inode);
+		ext4_discard_preallocations(inode);
 		up_write(&EXT4_I(inode)->i_data_sem);
 	}
 	if (is_dx(inode) && filp->private_data)
@@ -140,6 +140,9 @@ static int ext4_file_mmap(struct file *file, struct vm_area_struct *vma)
 	return 0;
 }
 
+extern int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+		__u64 start, __u64 len);
+
 const struct file_operations ext4_file_operations = {
 	.llseek		= generic_file_llseek,
 	.read		= do_sync_read,
@@ -162,7 +165,7 @@ const struct inode_operations ext4_file_inode_operations = {
 	.truncate	= ext4_truncate,
 	.setattr	= ext4_setattr,
 	.getattr	= ext4_getattr,
-#ifdef CONFIG_EXT4DEV_FS_XATTR
+#ifdef CONFIG_EXT4_FS_XATTR
 	.setxattr	= generic_setxattr,
 	.getxattr	= generic_getxattr,
 	.listxattr	= ext4_listxattr,
@@ -170,5 +173,6 @@ const struct inode_operations ext4_file_inode_operations = {
 #endif
 	.permission	= ext4_permission,
 	.fallocate	= ext4_fallocate,
+	.fiemap		= ext4_fiemap,
 };
 
diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c
index a45c3737ad31..5afe4370840b 100644
--- a/fs/ext4/fsync.c
+++ b/fs/ext4/fsync.c
@@ -28,6 +28,7 @@
 #include <linux/writeback.h>
 #include <linux/jbd2.h>
 #include <linux/blkdev.h>
+#include <linux/marker.h>
 #include "ext4.h"
 #include "ext4_jbd2.h"
 
@@ -43,7 +44,7 @@
  * inode to disk.
  */
 
-int ext4_sync_file(struct file * file, struct dentry *dentry, int datasync)
+int ext4_sync_file(struct file *file, struct dentry *dentry, int datasync)
 {
 	struct inode *inode = dentry->d_inode;
 	journal_t *journal = EXT4_SB(inode->i_sb)->s_journal;
@@ -51,6 +52,10 @@ int ext4_sync_file(struct file * file, struct dentry *dentry, int datasync)
 
 	J_ASSERT(ext4_journal_current_handle() == NULL);
 
+	trace_mark(ext4_sync_file, "dev %s datasync %d ino %ld parent %ld",
+		   inode->i_sb->s_id, datasync, inode->i_ino,
+		   dentry->d_parent->d_inode->i_ino);
+
 	/*
 	 * data=writeback:
 	 *  The caller's filemap_fdatawrite()/wait will sync the data.
diff --git a/fs/ext4/hash.c b/fs/ext4/hash.c
index 1d6329dbe390..556ca8eba3db 100644
--- a/fs/ext4/hash.c
+++ b/fs/ext4/hash.c
@@ -27,7 +27,7 @@ static void TEA_transform(__u32 buf[4], __u32 const in[])
 		sum += DELTA;
 		b0 += ((b1 << 4)+a) ^ (b1+sum) ^ ((b1 >> 5)+b);
 		b1 += ((b0 << 4)+c) ^ (b0+sum) ^ ((b0 >> 5)+d);
-	} while(--n);
+	} while (--n);
 
 	buf[0] += b0;
 	buf[1] += b1;
@@ -35,7 +35,7 @@ static void TEA_transform(__u32 buf[4], __u32 const in[])
 
 
 /* The old legacy hash */
-static __u32 dx_hack_hash (const char *name, int len)
+static __u32 dx_hack_hash(const char *name, int len)
 {
 	__u32 hash0 = 0x12a3fe2d, hash1 = 0x37abe8f9;
 	while (len--) {
@@ -59,7 +59,7 @@ static void str2hashbuf(const char *msg, int len, __u32 *buf, int num)
 	val = pad;
 	if (len > num*4)
 		len = num * 4;
-	for (i=0; i < len; i++) {
+	for (i = 0; i < len; i++) {
 		if ((i % 4) == 0)
 			val = pad;
 		val = msg[i] + (val << 8);
@@ -104,7 +104,7 @@ int ext4fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo)
 
 	/* Check to see if the seed is all zero's */
 	if (hinfo->seed) {
-		for (i=0; i < 4; i++) {
+		for (i = 0; i < 4; i++) {
 			if (hinfo->seed[i])
 				break;
 		}
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index f344834bbf58..fe34d74cfb19 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -115,9 +115,11 @@ ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group)
 			    block_group, bitmap_blk);
 		return NULL;
 	}
-	if (bh_uptodate_or_lock(bh))
+	if (buffer_uptodate(bh) &&
+	    !(desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)))
 		return bh;
 
+	lock_buffer(bh);
 	spin_lock(sb_bgl_lock(EXT4_SB(sb), block_group));
 	if (desc->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT)) {
 		ext4_init_inode_bitmap(sb, bh, block_group, desc);
@@ -154,39 +156,40 @@ ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group)
  * though), and then we'd have two inodes sharing the
  * same inode number and space on the harddisk.
  */
-void ext4_free_inode (handle_t *handle, struct inode * inode)
+void ext4_free_inode(handle_t *handle, struct inode *inode)
 {
-	struct super_block * sb = inode->i_sb;
+	struct super_block *sb = inode->i_sb;
 	int is_directory;
 	unsigned long ino;
 	struct buffer_head *bitmap_bh = NULL;
 	struct buffer_head *bh2;
 	ext4_group_t block_group;
 	unsigned long bit;
-	struct ext4_group_desc * gdp;
-	struct ext4_super_block * es;
+	struct ext4_group_desc *gdp;
+	struct ext4_super_block *es;
 	struct ext4_sb_info *sbi;
 	int fatal = 0, err;
 	ext4_group_t flex_group;
 
 	if (atomic_read(&inode->i_count) > 1) {
-		printk ("ext4_free_inode: inode has count=%d\n",
-					atomic_read(&inode->i_count));
+		printk(KERN_ERR "ext4_free_inode: inode has count=%d\n",
+		       atomic_read(&inode->i_count));
 		return;
 	}
 	if (inode->i_nlink) {
-		printk ("ext4_free_inode: inode has nlink=%d\n",
-			inode->i_nlink);
+		printk(KERN_ERR "ext4_free_inode: inode has nlink=%d\n",
+		       inode->i_nlink);
 		return;
 	}
 	if (!sb) {
-		printk("ext4_free_inode: inode on nonexistent device\n");
+		printk(KERN_ERR "ext4_free_inode: inode on "
+		       "nonexistent device\n");
 		return;
 	}
 	sbi = EXT4_SB(sb);
 
 	ino = inode->i_ino;
-	ext4_debug ("freeing inode %lu\n", ino);
+	ext4_debug("freeing inode %lu\n", ino);
 
 	/*
 	 * Note: we must free any quota before locking the superblock,
@@ -200,12 +203,12 @@ void ext4_free_inode (handle_t *handle, struct inode * inode)
 	is_directory = S_ISDIR(inode->i_mode);
 
 	/* Do this BEFORE marking the inode not in use or returning an error */
-	clear_inode (inode);
+	clear_inode(inode);
 
 	es = EXT4_SB(sb)->s_es;
 	if (ino < EXT4_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) {
-		ext4_error (sb, "ext4_free_inode",
-			    "reserved or nonexistent inode %lu", ino);
+		ext4_error(sb, "ext4_free_inode",
+			   "reserved or nonexistent inode %lu", ino);
 		goto error_return;
 	}
 	block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
@@ -222,10 +225,10 @@ void ext4_free_inode (handle_t *handle, struct inode * inode)
 	/* Ok, now we can actually update the inode bitmaps.. */
 	if (!ext4_clear_bit_atomic(sb_bgl_lock(sbi, block_group),
 					bit, bitmap_bh->b_data))
-		ext4_error (sb, "ext4_free_inode",
-			      "bit already cleared for inode %lu", ino);
+		ext4_error(sb, "ext4_free_inode",
+			   "bit already cleared for inode %lu", ino);
 	else {
-		gdp = ext4_get_group_desc (sb, block_group, &bh2);
+		gdp = ext4_get_group_desc(sb, block_group, &bh2);
 
 		BUFFER_TRACE(bh2, "get_write_access");
 		fatal = ext4_journal_get_write_access(handle, bh2);
@@ -287,7 +290,7 @@ static int find_group_dir(struct super_block *sb, struct inode *parent,
 	avefreei = freei / ngroups;
 
 	for (group = 0; group < ngroups; group++) {
-		desc = ext4_get_group_desc (sb, group, NULL);
+		desc = ext4_get_group_desc(sb, group, NULL);
 		if (!desc || !desc->bg_free_inodes_count)
 			continue;
 		if (le16_to_cpu(desc->bg_free_inodes_count) < avefreei)
@@ -576,16 +579,16 @@ static int find_group_other(struct super_block *sb, struct inode *parent,
  * For other inodes, search forward from the parent directory's block
  * group to find a free inode.
  */
-struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode)
+struct inode *ext4_new_inode(handle_t *handle, struct inode *dir, int mode)
 {
 	struct super_block *sb;
 	struct buffer_head *bitmap_bh = NULL;
 	struct buffer_head *bh2;
 	ext4_group_t group = 0;
 	unsigned long ino = 0;
-	struct inode * inode;
-	struct ext4_group_desc * gdp = NULL;
-	struct ext4_super_block * es;
+	struct inode *inode;
+	struct ext4_group_desc *gdp = NULL;
+	struct ext4_super_block *es;
 	struct ext4_inode_info *ei;
 	struct ext4_sb_info *sbi;
 	int ret2, err = 0;
@@ -613,7 +616,7 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode)
 	}
 
 	if (S_ISDIR(mode)) {
-		if (test_opt (sb, OLDALLOC))
+		if (test_opt(sb, OLDALLOC))
 			ret2 = find_group_dir(sb, dir, &group);
 		else
 			ret2 = find_group_orlov(sb, dir, &group);
@@ -783,7 +786,7 @@ got:
 	}
 
 	inode->i_uid = current->fsuid;
-	if (test_opt (sb, GRPID))
+	if (test_opt(sb, GRPID))
 		inode->i_gid = dir->i_gid;
 	else if (dir->i_mode & S_ISGID) {
 		inode->i_gid = dir->i_gid;
@@ -816,7 +819,6 @@ got:
 		ei->i_flags &= ~EXT4_DIRSYNC_FL;
 	ei->i_file_acl = 0;
 	ei->i_dtime = 0;
-	ei->i_block_alloc_info = NULL;
 	ei->i_block_group = group;
 
 	ext4_set_inode_flags(inode);
@@ -832,7 +834,7 @@ got:
 	ei->i_extra_isize = EXT4_SB(sb)->s_want_extra_isize;
 
 	ret = inode;
-	if(DQUOT_ALLOC_INODE(inode)) {
+	if (DQUOT_ALLOC_INODE(inode)) {
 		err = -EDQUOT;
 		goto fail_drop;
 	}
@@ -841,7 +843,7 @@ got:
 	if (err)
 		goto fail_free_drop;
 
-	err = ext4_init_security(handle,inode, dir);
+	err = ext4_init_security(handle, inode, dir);
 	if (err)
 		goto fail_free_drop;
 
@@ -959,7 +961,7 @@ error:
 	return ERR_PTR(err);
 }
 
-unsigned long ext4_count_free_inodes (struct super_block * sb)
+unsigned long ext4_count_free_inodes(struct super_block *sb)
 {
 	unsigned long desc_count;
 	struct ext4_group_desc *gdp;
@@ -974,7 +976,7 @@ unsigned long ext4_count_free_inodes (struct super_block * sb)
 	bitmap_count = 0;
 	gdp = NULL;
 	for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) {
-		gdp = ext4_get_group_desc (sb, i, NULL);
+		gdp = ext4_get_group_desc(sb, i, NULL);
 		if (!gdp)
 			continue;
 		desc_count += le16_to_cpu(gdp->bg_free_inodes_count);
@@ -989,13 +991,14 @@ unsigned long ext4_count_free_inodes (struct super_block * sb)
 		bitmap_count += x;
 	}
 	brelse(bitmap_bh);
-	printk("ext4_count_free_inodes: stored = %u, computed = %lu, %lu\n",
-		le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count);
+	printk(KERN_DEBUG "ext4_count_free_inodes: "
+	       "stored = %u, computed = %lu, %lu\n",
+	       le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count);
 	return desc_count;
 #else
 	desc_count = 0;
 	for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) {
-		gdp = ext4_get_group_desc (sb, i, NULL);
+		gdp = ext4_get_group_desc(sb, i, NULL);
 		if (!gdp)
 			continue;
 		desc_count += le16_to_cpu(gdp->bg_free_inodes_count);
@@ -1006,13 +1009,13 @@ unsigned long ext4_count_free_inodes (struct super_block * sb)
 }
 
 /* Called at mount-time, super-block is locked */
-unsigned long ext4_count_dirs (struct super_block * sb)
+unsigned long ext4_count_dirs(struct super_block * sb)
 {
 	unsigned long count = 0;
 	ext4_group_t i;
 
 	for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) {
-		struct ext4_group_desc *gdp = ext4_get_group_desc (sb, i, NULL);
+		struct ext4_group_desc *gdp = ext4_get_group_desc(sb, i, NULL);
 		if (!gdp)
 			continue;
 		count += le16_to_cpu(gdp->bg_used_dirs_count);
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 7e91913e325b..9b4ec9decfd1 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -190,7 +190,7 @@ static int ext4_journal_test_restart(handle_t *handle, struct inode *inode)
 /*
  * Called at the last iput() if i_nlink is zero.
  */
-void ext4_delete_inode (struct inode * inode)
+void ext4_delete_inode(struct inode *inode)
 {
 	handle_t *handle;
 	int err;
@@ -330,11 +330,11 @@ static int ext4_block_to_path(struct inode *inode,
 	int final = 0;
 
 	if (i_block < 0) {
-		ext4_warning (inode->i_sb, "ext4_block_to_path", "block < 0");
+		ext4_warning(inode->i_sb, "ext4_block_to_path", "block < 0");
 	} else if (i_block < direct_blocks) {
 		offsets[n++] = i_block;
 		final = direct_blocks;
-	} else if ( (i_block -= direct_blocks) < indirect_blocks) {
+	} else if ((i_block -= direct_blocks) < indirect_blocks) {
 		offsets[n++] = EXT4_IND_BLOCK;
 		offsets[n++] = i_block;
 		final = ptrs;
@@ -400,14 +400,14 @@ static Indirect *ext4_get_branch(struct inode *inode, int depth,
 
 	*err = 0;
 	/* i_data is not going away, no lock needed */
-	add_chain (chain, NULL, EXT4_I(inode)->i_data + *offsets);
+	add_chain(chain, NULL, EXT4_I(inode)->i_data + *offsets);
 	if (!p->key)
 		goto no_block;
 	while (--depth) {
 		bh = sb_bread(sb, le32_to_cpu(p->key));
 		if (!bh)
 			goto failure;
-		add_chain(++p, bh, (__le32*)bh->b_data + *++offsets);
+		add_chain(++p, bh, (__le32 *)bh->b_data + *++offsets);
 		/* Reader: end */
 		if (!p->key)
 			goto no_block;
@@ -443,7 +443,7 @@ no_block:
 static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind)
 {
 	struct ext4_inode_info *ei = EXT4_I(inode);
-	__le32 *start = ind->bh ? (__le32*) ind->bh->b_data : ei->i_data;
+	__le32 *start = ind->bh ? (__le32 *) ind->bh->b_data : ei->i_data;
 	__le32 *p;
 	ext4_fsblk_t bg_start;
 	ext4_fsblk_t last_block;
@@ -486,18 +486,9 @@ static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind)
 static ext4_fsblk_t ext4_find_goal(struct inode *inode, ext4_lblk_t block,
 		Indirect *partial)
 {
-	struct ext4_block_alloc_info *block_i;
-
-	block_i =  EXT4_I(inode)->i_block_alloc_info;
-
 	/*
-	 * try the heuristic for sequential allocation,
-	 * failing that at least try to get decent locality.
+	 * XXX need to get goal block from mballoc's data structures
 	 */
-	if (block_i && (block == block_i->last_alloc_logical_block + 1)
-		&& (block_i->last_alloc_physical_block != 0)) {
-		return block_i->last_alloc_physical_block + 1;
-	}
 
 	return ext4_find_near(inode, partial);
 }
@@ -630,7 +621,7 @@ allocated:
 	*err = 0;
 	return ret;
 failed_out:
-	for (i = 0; i <index; i++)
+	for (i = 0; i < index; i++)
 		ext4_free_blocks(handle, inode, new_blocks[i], 1, 0);
 	return ret;
 }
@@ -703,7 +694,7 @@ static int ext4_alloc_branch(handle_t *handle, struct inode *inode,
 		branch[n].p = (__le32 *) bh->b_data + offsets[n];
 		branch[n].key = cpu_to_le32(new_blocks[n]);
 		*branch[n].p = branch[n].key;
-		if ( n == indirect_blks) {
+		if (n == indirect_blks) {
 			current_block = new_blocks[n];
 			/*
 			 * End of chain, update the last new metablock of
@@ -730,7 +721,7 @@ failed:
 		BUFFER_TRACE(branch[i].bh, "call jbd2_journal_forget");
 		ext4_journal_forget(handle, branch[i].bh);
 	}
-	for (i = 0; i <indirect_blks; i++)
+	for (i = 0; i < indirect_blks; i++)
 		ext4_free_blocks(handle, inode, new_blocks[i], 1, 0);
 
 	ext4_free_blocks(handle, inode, new_blocks[i], num, 0);
@@ -757,10 +748,8 @@ static int ext4_splice_branch(handle_t *handle, struct inode *inode,
 {
 	int i;
 	int err = 0;
-	struct ext4_block_alloc_info *block_i;
 	ext4_fsblk_t current_block;
 
-	block_i = EXT4_I(inode)->i_block_alloc_info;
 	/*
 	 * If we're splicing into a [td]indirect block (as opposed to the
 	 * inode) then we need to get write access to the [td]indirect block
@@ -783,18 +772,7 @@ static int ext4_splice_branch(handle_t *handle, struct inode *inode,
 	if (num == 0 && blks > 1) {
 		current_block = le32_to_cpu(where->key) + 1;
 		for (i = 1; i < blks; i++)
-			*(where->p + i ) = cpu_to_le32(current_block++);
-	}
-
-	/*
-	 * update the most recently allocated logical & physical block
-	 * in i_block_alloc_info, to assist find the proper goal block for next
-	 * allocation
-	 */
-	if (block_i) {
-		block_i->last_alloc_logical_block = block + blks - 1;
-		block_i->last_alloc_physical_block =
-				le32_to_cpu(where[num].key) + blks - 1;
+			*(where->p + i) = cpu_to_le32(current_block++);
 	}
 
 	/* We are done with atomic stuff, now do the rest of housekeeping */
@@ -914,12 +892,8 @@ int ext4_get_blocks_handle(handle_t *handle, struct inode *inode,
 		goto cleanup;
 
 	/*
-	 * Okay, we need to do block allocation.  Lazily initialize the block
-	 * allocation info here if necessary
+	 * Okay, we need to do block allocation.
 	*/
-	if (S_ISREG(inode->i_mode) && (!ei->i_block_alloc_info))
-		ext4_init_block_alloc_info(inode);
-
 	goal = ext4_find_goal(inode, iblock, partial);
 
 	/* the number of blocks need to allocate for [d,t]indirect blocks */
@@ -1030,19 +1004,20 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used)
 	BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks);
 	mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb;
 
-	/* Account for allocated meta_blocks */
-	mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks;
+	if (mdb_free) {
+		/* Account for allocated meta_blocks */
+		mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks;
 
-	/* update fs free blocks counter for truncate case */
-	percpu_counter_add(&sbi->s_freeblocks_counter, mdb_free);
+		/* update fs dirty blocks counter */
+		percpu_counter_sub(&sbi->s_dirtyblocks_counter, mdb_free);
+		EXT4_I(inode)->i_allocated_meta_blocks = 0;
+		EXT4_I(inode)->i_reserved_meta_blocks = mdb;
+	}
 
 	/* update per-inode reservations */
 	BUG_ON(used  > EXT4_I(inode)->i_reserved_data_blocks);
 	EXT4_I(inode)->i_reserved_data_blocks -= used;
 
-	BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks);
-	EXT4_I(inode)->i_reserved_meta_blocks = mdb;
-	EXT4_I(inode)->i_allocated_meta_blocks = 0;
 	spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
 }
 
@@ -1160,8 +1135,8 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block,
 /* Maximum number of blocks we map for direct IO at once. */
 #define DIO_MAX_BLOCKS 4096
 
-static int ext4_get_block(struct inode *inode, sector_t iblock,
-			struct buffer_head *bh_result, int create)
+int ext4_get_block(struct inode *inode, sector_t iblock,
+		   struct buffer_head *bh_result, int create)
 {
 	handle_t *handle = ext4_journal_current_handle();
 	int ret = 0, started = 0;
@@ -1241,7 +1216,7 @@ struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode,
 			BUFFER_TRACE(bh, "call get_create_access");
 			fatal = ext4_journal_get_create_access(handle, bh);
 			if (!fatal && !buffer_uptodate(bh)) {
-				memset(bh->b_data,0,inode->i_sb->s_blocksize);
+				memset(bh->b_data, 0, inode->i_sb->s_blocksize);
 				set_buffer_uptodate(bh);
 			}
 			unlock_buffer(bh);
@@ -1266,7 +1241,7 @@ err:
 struct buffer_head *ext4_bread(handle_t *handle, struct inode *inode,
 			       ext4_lblk_t block, int create, int *err)
 {
-	struct buffer_head * bh;
+	struct buffer_head *bh;
 
 	bh = ext4_getblk(handle, inode, block, create, err);
 	if (!bh)
@@ -1282,13 +1257,13 @@ struct buffer_head *ext4_bread(handle_t *handle, struct inode *inode,
 	return NULL;
 }
 
-static int walk_page_buffers(	handle_t *handle,
-				struct buffer_head *head,
-				unsigned from,
-				unsigned to,
-				int *partial,
-				int (*fn)(	handle_t *handle,
-						struct buffer_head *bh))
+static int walk_page_buffers(handle_t *handle,
+			     struct buffer_head *head,
+			     unsigned from,
+			     unsigned to,
+			     int *partial,
+			     int (*fn)(handle_t *handle,
+				       struct buffer_head *bh))
 {
 	struct buffer_head *bh;
 	unsigned block_start, block_end;
@@ -1296,9 +1271,9 @@ static int walk_page_buffers(	handle_t *handle,
 	int err, ret = 0;
 	struct buffer_head *next;
 
-	for (	bh = head, block_start = 0;
-		ret == 0 && (bh != head || !block_start);
-		block_start = block_end, bh = next)
+	for (bh = head, block_start = 0;
+	     ret == 0 && (bh != head || !block_start);
+	     block_start = block_end, bh = next)
 	{
 		next = bh->b_this_page;
 		block_end = block_start + blocksize;
@@ -1351,23 +1326,23 @@ static int ext4_write_begin(struct file *file, struct address_space *mapping,
 				loff_t pos, unsigned len, unsigned flags,
 				struct page **pagep, void **fsdata)
 {
- 	struct inode *inode = mapping->host;
+	struct inode *inode = mapping->host;
 	int ret, needed_blocks = ext4_writepage_trans_blocks(inode);
 	handle_t *handle;
 	int retries = 0;
- 	struct page *page;
+	struct page *page;
  	pgoff_t index;
- 	unsigned from, to;
+	unsigned from, to;
 
  	index = pos >> PAGE_CACHE_SHIFT;
- 	from = pos & (PAGE_CACHE_SIZE - 1);
- 	to = from + len;
+	from = pos & (PAGE_CACHE_SIZE - 1);
+	to = from + len;
 
 retry:
-  	handle = ext4_journal_start(inode, needed_blocks);
-  	if (IS_ERR(handle)) {
-  		ret = PTR_ERR(handle);
-  		goto out;
+	handle = ext4_journal_start(inode, needed_blocks);
+	if (IS_ERR(handle)) {
+		ret = PTR_ERR(handle);
+		goto out;
 	}
 
 	page = __grab_cache_page(mapping, index);
@@ -1387,9 +1362,16 @@ retry:
 	}
 
 	if (ret) {
- 		unlock_page(page);
+		unlock_page(page);
 		ext4_journal_stop(handle);
- 		page_cache_release(page);
+		page_cache_release(page);
+		/*
+		 * block_write_begin may have instantiated a few blocks
+		 * outside i_size.  Trim these off again. Don't need
+		 * i_size_read because we hold i_mutex.
+		 */
+		if (pos + len > inode->i_size)
+			vmtruncate(inode, inode->i_size);
 	}
 
 	if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
@@ -1426,16 +1408,18 @@ static int ext4_ordered_write_end(struct file *file,
 	ret = ext4_jbd2_file_inode(handle, inode);
 
 	if (ret == 0) {
-		/*
-		 * generic_write_end() will run mark_inode_dirty() if i_size
-		 * changes.  So let's piggyback the i_disksize mark_inode_dirty
-		 * into that.
-		 */
 		loff_t new_i_size;
 
 		new_i_size = pos + copied;
-		if (new_i_size > EXT4_I(inode)->i_disksize)
-			EXT4_I(inode)->i_disksize = new_i_size;
+		if (new_i_size > EXT4_I(inode)->i_disksize) {
+			ext4_update_i_disksize(inode, new_i_size);
+			/* We need to mark inode dirty even if
+			 * new_i_size is less that inode->i_size
+			 * bu greater than i_disksize.(hint delalloc)
+			 */
+			ext4_mark_inode_dirty(handle, inode);
+		}
+
 		ret2 = generic_write_end(file, mapping, pos, len, copied,
 							page, fsdata);
 		copied = ret2;
@@ -1460,8 +1444,14 @@ static int ext4_writeback_write_end(struct file *file,
 	loff_t new_i_size;
 
 	new_i_size = pos + copied;
-	if (new_i_size > EXT4_I(inode)->i_disksize)
-		EXT4_I(inode)->i_disksize = new_i_size;
+	if (new_i_size > EXT4_I(inode)->i_disksize) {
+		ext4_update_i_disksize(inode, new_i_size);
+		/* We need to mark inode dirty even if
+		 * new_i_size is less that inode->i_size
+		 * bu greater than i_disksize.(hint delalloc)
+		 */
+		ext4_mark_inode_dirty(handle, inode);
+	}
 
 	ret2 = generic_write_end(file, mapping, pos, len, copied,
 							page, fsdata);
@@ -1486,6 +1476,7 @@ static int ext4_journalled_write_end(struct file *file,
 	int ret = 0, ret2;
 	int partial = 0;
 	unsigned from, to;
+	loff_t new_i_size;
 
 	from = pos & (PAGE_CACHE_SIZE - 1);
 	to = from + len;
@@ -1500,11 +1491,12 @@ static int ext4_journalled_write_end(struct file *file,
 				to, &partial, write_end_fn);
 	if (!partial)
 		SetPageUptodate(page);
-	if (pos+copied > inode->i_size)
+	new_i_size = pos + copied;
+	if (new_i_size > inode->i_size)
 		i_size_write(inode, pos+copied);
 	EXT4_I(inode)->i_state |= EXT4_STATE_JDATA;
-	if (inode->i_size > EXT4_I(inode)->i_disksize) {
-		EXT4_I(inode)->i_disksize = inode->i_size;
+	if (new_i_size > EXT4_I(inode)->i_disksize) {
+		ext4_update_i_disksize(inode, new_i_size);
 		ret2 = ext4_mark_inode_dirty(handle, inode);
 		if (!ret)
 			ret = ret2;
@@ -1521,6 +1513,7 @@ static int ext4_journalled_write_end(struct file *file,
 
 static int ext4_da_reserve_space(struct inode *inode, int nrblocks)
 {
+	int retries = 0;
        struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
        unsigned long md_needed, mdblocks, total = 0;
 
@@ -1529,6 +1522,7 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks)
 	 * in order to allocate nrblocks
 	 * worse case is one extent per block
 	 */
+repeat:
 	spin_lock(&EXT4_I(inode)->i_block_reservation_lock);
 	total = EXT4_I(inode)->i_reserved_data_blocks + nrblocks;
 	mdblocks = ext4_calc_metadata_amount(inode, total);
@@ -1537,13 +1531,14 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks)
 	md_needed = mdblocks - EXT4_I(inode)->i_reserved_meta_blocks;
 	total = md_needed + nrblocks;
 
-	if (ext4_has_free_blocks(sbi, total) < total) {
+	if (ext4_claim_free_blocks(sbi, total)) {
 		spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
+		if (ext4_should_retry_alloc(inode->i_sb, &retries)) {
+			yield();
+			goto repeat;
+		}
 		return -ENOSPC;
 	}
-	/* reduce fs free blocks counter */
-	percpu_counter_sub(&sbi->s_freeblocks_counter, total);
-
 	EXT4_I(inode)->i_reserved_data_blocks += nrblocks;
 	EXT4_I(inode)->i_reserved_meta_blocks = mdblocks;
 
@@ -1585,8 +1580,8 @@ static void ext4_da_release_space(struct inode *inode, int to_free)
 
 	release = to_free + mdb_free;
 
-	/* update fs free blocks counter for truncate case */
-	percpu_counter_add(&sbi->s_freeblocks_counter, release);
+	/* update fs dirty blocks counter for truncate case */
+	percpu_counter_sub(&sbi->s_dirtyblocks_counter, release);
 
 	/* update per-inode reservations */
 	BUG_ON(to_free > EXT4_I(inode)->i_reserved_data_blocks);
@@ -1630,6 +1625,7 @@ struct mpage_da_data {
 	struct writeback_control *wbc;
 	int io_done;
 	long pages_written;
+	int retval;
 };
 
 /*
@@ -1783,6 +1779,57 @@ static inline void __unmap_underlying_blocks(struct inode *inode,
 		unmap_underlying_metadata(bdev, bh->b_blocknr + i);
 }
 
+static void ext4_da_block_invalidatepages(struct mpage_da_data *mpd,
+					sector_t logical, long blk_cnt)
+{
+	int nr_pages, i;
+	pgoff_t index, end;
+	struct pagevec pvec;
+	struct inode *inode = mpd->inode;
+	struct address_space *mapping = inode->i_mapping;
+
+	index = logical >> (PAGE_CACHE_SHIFT - inode->i_blkbits);
+	end   = (logical + blk_cnt - 1) >>
+				(PAGE_CACHE_SHIFT - inode->i_blkbits);
+	while (index <= end) {
+		nr_pages = pagevec_lookup(&pvec, mapping, index, PAGEVEC_SIZE);
+		if (nr_pages == 0)
+			break;
+		for (i = 0; i < nr_pages; i++) {
+			struct page *page = pvec.pages[i];
+			index = page->index;
+			if (index > end)
+				break;
+			index++;
+
+			BUG_ON(!PageLocked(page));
+			BUG_ON(PageWriteback(page));
+			block_invalidatepage(page, 0);
+			ClearPageUptodate(page);
+			unlock_page(page);
+		}
+	}
+	return;
+}
+
+static void ext4_print_free_blocks(struct inode *inode)
+{
+	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
+	printk(KERN_EMERG "Total free blocks count %lld\n",
+			ext4_count_free_blocks(inode->i_sb));
+	printk(KERN_EMERG "Free/Dirty block details\n");
+	printk(KERN_EMERG "free_blocks=%lld\n",
+			percpu_counter_sum(&sbi->s_freeblocks_counter));
+	printk(KERN_EMERG "dirty_blocks=%lld\n",
+			percpu_counter_sum(&sbi->s_dirtyblocks_counter));
+	printk(KERN_EMERG "Block reservation details\n");
+	printk(KERN_EMERG "i_reserved_data_blocks=%lu\n",
+			EXT4_I(inode)->i_reserved_data_blocks);
+	printk(KERN_EMERG "i_reserved_meta_blocks=%lu\n",
+			EXT4_I(inode)->i_reserved_meta_blocks);
+	return;
+}
+
 /*
  * mpage_da_map_blocks - go through given space
  *
@@ -1792,32 +1839,69 @@ static inline void __unmap_underlying_blocks(struct inode *inode,
  * The function skips space we know is already mapped to disk blocks.
  *
  */
-static void mpage_da_map_blocks(struct mpage_da_data *mpd)
+static int  mpage_da_map_blocks(struct mpage_da_data *mpd)
 {
 	int err = 0;
-	struct buffer_head *lbh = &mpd->lbh;
-	sector_t next = lbh->b_blocknr;
 	struct buffer_head new;
+	struct buffer_head *lbh = &mpd->lbh;
+	sector_t next;
 
 	/*
 	 * We consider only non-mapped and non-allocated blocks
 	 */
 	if (buffer_mapped(lbh) && !buffer_delay(lbh))
-		return;
-
+		return 0;
 	new.b_state = lbh->b_state;
 	new.b_blocknr = 0;
 	new.b_size = lbh->b_size;
-
+	next = lbh->b_blocknr;
 	/*
 	 * If we didn't accumulate anything
 	 * to write simply return
 	 */
 	if (!new.b_size)
-		return;
+		return 0;
 	err = mpd->get_block(mpd->inode, next, &new, 1);
-	if (err)
-		return;
+	if (err) {
+
+		/* If get block returns with error
+		 * we simply return. Later writepage
+		 * will redirty the page and writepages
+		 * will find the dirty page again
+		 */
+		if (err == -EAGAIN)
+			return 0;
+
+		if (err == -ENOSPC &&
+				ext4_count_free_blocks(mpd->inode->i_sb)) {
+			mpd->retval = err;
+			return 0;
+		}
+
+		/*
+		 * get block failure will cause us
+		 * to loop in writepages. Because
+		 * a_ops->writepage won't be able to
+		 * make progress. The page will be redirtied
+		 * by writepage and writepages will again
+		 * try to write the same.
+		 */
+		printk(KERN_EMERG "%s block allocation failed for inode %lu "
+				  "at logical offset %llu with max blocks "
+				  "%zd with error %d\n",
+				  __func__, mpd->inode->i_ino,
+				  (unsigned long long)next,
+				  lbh->b_size >> mpd->inode->i_blkbits, err);
+		printk(KERN_EMERG "This should not happen.!! "
+					"Data will be lost\n");
+		if (err == -ENOSPC) {
+			ext4_print_free_blocks(mpd->inode);
+		}
+		/* invlaidate all the pages */
+		ext4_da_block_invalidatepages(mpd, next,
+				lbh->b_size >> mpd->inode->i_blkbits);
+		return err;
+	}
 	BUG_ON(new.b_size == 0);
 
 	if (buffer_new(&new))
@@ -1830,7 +1914,7 @@ static void mpage_da_map_blocks(struct mpage_da_data *mpd)
 	if (buffer_delay(lbh) || buffer_unwritten(lbh))
 		mpage_put_bnr_to_bhs(mpd, next, &new);
 
-	return;
+	return 0;
 }
 
 #define BH_FLAGS ((1 << BH_Uptodate) | (1 << BH_Mapped) | \
@@ -1899,8 +1983,8 @@ flush_it:
 	 * We couldn't merge the block to our extent, so we
 	 * need to flush current  extent and start new one
 	 */
-	mpage_da_map_blocks(mpd);
-	mpage_da_submit_io(mpd);
+	if (mpage_da_map_blocks(mpd) == 0)
+		mpage_da_submit_io(mpd);
 	mpd->io_done = 1;
 	return;
 }
@@ -1942,8 +2026,8 @@ static int __mpage_da_writepage(struct page *page,
 		 * and start IO on them using writepage()
 		 */
 		if (mpd->next_page != mpd->first_page) {
-			mpage_da_map_blocks(mpd);
-			mpage_da_submit_io(mpd);
+			if (mpage_da_map_blocks(mpd) == 0)
+				mpage_da_submit_io(mpd);
 			/*
 			 * skip rest of the page in the page_vec
 			 */
@@ -2018,39 +2102,36 @@ static int __mpage_da_writepage(struct page *page,
  */
 static int mpage_da_writepages(struct address_space *mapping,
 			       struct writeback_control *wbc,
-			       get_block_t get_block)
+			       struct mpage_da_data *mpd)
 {
-	struct mpage_da_data mpd;
 	long to_write;
 	int ret;
 
-	if (!get_block)
+	if (!mpd->get_block)
 		return generic_writepages(mapping, wbc);
 
-	mpd.wbc = wbc;
-	mpd.inode = mapping->host;
-	mpd.lbh.b_size = 0;
-	mpd.lbh.b_state = 0;
-	mpd.lbh.b_blocknr = 0;
-	mpd.first_page = 0;
-	mpd.next_page = 0;
-	mpd.get_block = get_block;
-	mpd.io_done = 0;
-	mpd.pages_written = 0;
+	mpd->lbh.b_size = 0;
+	mpd->lbh.b_state = 0;
+	mpd->lbh.b_blocknr = 0;
+	mpd->first_page = 0;
+	mpd->next_page = 0;
+	mpd->io_done = 0;
+	mpd->pages_written = 0;
+	mpd->retval = 0;
 
 	to_write = wbc->nr_to_write;
 
-	ret = write_cache_pages(mapping, wbc, __mpage_da_writepage, &mpd);
+	ret = write_cache_pages(mapping, wbc, __mpage_da_writepage, mpd);
 
 	/*
 	 * Handle last extent of pages
 	 */
-	if (!mpd.io_done && mpd.next_page != mpd.first_page) {
-		mpage_da_map_blocks(&mpd);
-		mpage_da_submit_io(&mpd);
+	if (!mpd->io_done && mpd->next_page != mpd->first_page) {
+		if (mpage_da_map_blocks(mpd) == 0)
+			mpage_da_submit_io(mpd);
 	}
 
-	wbc->nr_to_write = to_write - mpd.pages_written;
+	wbc->nr_to_write = to_write - mpd->pages_written;
 	return ret;
 }
 
@@ -2103,18 +2184,24 @@ static int ext4_da_get_block_write(struct inode *inode, sector_t iblock,
 	handle_t *handle = NULL;
 
 	handle = ext4_journal_current_handle();
-	if (!handle) {
-		ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks,
-				   bh_result, 0, 0, 0);
-		BUG_ON(!ret);
-	} else {
-		ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks,
-				   bh_result, create, 0, EXT4_DELALLOC_RSVED);
-	}
-
+	BUG_ON(!handle);
+	ret = ext4_get_blocks_wrap(handle, inode, iblock, max_blocks,
+			bh_result, create, 0, EXT4_DELALLOC_RSVED);
 	if (ret > 0) {
+
 		bh_result->b_size = (ret << inode->i_blkbits);
 
+		if (ext4_should_order_data(inode)) {
+			int retval;
+			retval = ext4_jbd2_file_inode(handle, inode);
+			if (retval)
+				/*
+				 * Failed to add inode for ordered
+				 * mode. Don't update file size
+				 */
+				return retval;
+		}
+
 		/*
 		 * Update on-disk size along with block allocation
 		 * we don't use 'extend_disksize' as size may change
@@ -2124,18 +2211,9 @@ static int ext4_da_get_block_write(struct inode *inode, sector_t iblock,
 		if (disksize > i_size_read(inode))
 			disksize = i_size_read(inode);
 		if (disksize > EXT4_I(inode)->i_disksize) {
-			/*
-			 * XXX: replace with spinlock if seen contended -bzzz
-			 */
-			down_write(&EXT4_I(inode)->i_data_sem);
-			if (disksize > EXT4_I(inode)->i_disksize)
-				EXT4_I(inode)->i_disksize = disksize;
-			up_write(&EXT4_I(inode)->i_data_sem);
-
-			if (EXT4_I(inode)->i_disksize == disksize) {
-				ret = ext4_mark_inode_dirty(handle, inode);
-				return ret;
-			}
+			ext4_update_i_disksize(inode, disksize);
+			ret = ext4_mark_inode_dirty(handle, inode);
+			return ret;
 		}
 		ret = 0;
 	}
@@ -2284,6 +2362,7 @@ static int ext4_da_writepages(struct address_space *mapping,
 {
 	handle_t *handle = NULL;
 	loff_t range_start = 0;
+	struct mpage_da_data mpd;
 	struct inode *inode = mapping->host;
 	int needed_blocks, ret = 0, nr_to_writebump = 0;
 	long to_write, pages_skipped = 0;
@@ -2317,6 +2396,9 @@ static int ext4_da_writepages(struct address_space *mapping,
 	range_start =  wbc->range_start;
 	pages_skipped = wbc->pages_skipped;
 
+	mpd.wbc = wbc;
+	mpd.inode = mapping->host;
+
 restart_loop:
 	to_write = wbc->nr_to_write;
 	while (!ret && to_write > 0) {
@@ -2340,23 +2422,17 @@ restart_loop:
 			dump_stack();
 			goto out_writepages;
 		}
-		if (ext4_should_order_data(inode)) {
-			/*
-			 * With ordered mode we need to add
-			 * the inode to the journal handl
-			 * when we do block allocation.
-			 */
-			ret = ext4_jbd2_file_inode(handle, inode);
-			if (ret) {
-				ext4_journal_stop(handle);
-				goto out_writepages;
-			}
-		}
-
 		to_write -= wbc->nr_to_write;
-		ret = mpage_da_writepages(mapping, wbc,
-					  ext4_da_get_block_write);
+
+		mpd.get_block = ext4_da_get_block_write;
+		ret = mpage_da_writepages(mapping, wbc, &mpd);
+
 		ext4_journal_stop(handle);
+
+		if (mpd.retval == -ENOSPC)
+			jbd2_journal_force_commit_nested(sbi->s_journal);
+
+		/* reset the retry count */
 		if (ret == MPAGE_DA_EXTENT_TAIL) {
 			/*
 			 * got one extent now try with
@@ -2391,6 +2467,33 @@ out_writepages:
 	return ret;
 }
 
+#define FALL_BACK_TO_NONDELALLOC 1
+static int ext4_nonda_switch(struct super_block *sb)
+{
+	s64 free_blocks, dirty_blocks;
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+
+	/*
+	 * switch to non delalloc mode if we are running low
+	 * on free block. The free block accounting via percpu
+	 * counters can get slightly wrong with FBC_BATCH getting
+	 * accumulated on each CPU without updating global counters
+	 * Delalloc need an accurate free block accounting. So switch
+	 * to non delalloc when we are near to error range.
+	 */
+	free_blocks  = percpu_counter_read_positive(&sbi->s_freeblocks_counter);
+	dirty_blocks = percpu_counter_read_positive(&sbi->s_dirtyblocks_counter);
+	if (2 * free_blocks < 3 * dirty_blocks ||
+		free_blocks < (dirty_blocks + EXT4_FREEBLOCKS_WATERMARK)) {
+		/*
+		 * free block count is less that 150% of dirty blocks
+		 * or free blocks is less that watermark
+		 */
+		return 1;
+	}
+	return 0;
+}
+
 static int ext4_da_write_begin(struct file *file, struct address_space *mapping,
 				loff_t pos, unsigned len, unsigned flags,
 				struct page **pagep, void **fsdata)
@@ -2406,6 +2509,12 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping,
 	from = pos & (PAGE_CACHE_SIZE - 1);
 	to = from + len;
 
+	if (ext4_nonda_switch(inode->i_sb)) {
+		*fsdata = (void *)FALL_BACK_TO_NONDELALLOC;
+		return ext4_write_begin(file, mapping, pos,
+					len, flags, pagep, fsdata);
+	}
+	*fsdata = (void *)0;
 retry:
 	/*
 	 * With delayed allocation, we don't log the i_disksize update
@@ -2433,6 +2542,13 @@ retry:
 		unlock_page(page);
 		ext4_journal_stop(handle);
 		page_cache_release(page);
+		/*
+		 * block_write_begin may have instantiated a few blocks
+		 * outside i_size.  Trim these off again. Don't need
+		 * i_size_read because we hold i_mutex.
+		 */
+		if (pos + len > inode->i_size)
+			vmtruncate(inode, inode->i_size);
 	}
 
 	if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
@@ -2456,7 +2572,7 @@ static int ext4_da_should_update_i_disksize(struct page *page,
 	bh = page_buffers(page);
 	idx = offset >> inode->i_blkbits;
 
-	for (i=0; i < idx; i++)
+	for (i = 0; i < idx; i++)
 		bh = bh->b_this_page;
 
 	if (!buffer_mapped(bh) || (buffer_delay(bh)))
@@ -2474,9 +2590,22 @@ static int ext4_da_write_end(struct file *file,
 	handle_t *handle = ext4_journal_current_handle();
 	loff_t new_i_size;
 	unsigned long start, end;
+	int write_mode = (int)(unsigned long)fsdata;
+
+	if (write_mode == FALL_BACK_TO_NONDELALLOC) {
+		if (ext4_should_order_data(inode)) {
+			return ext4_ordered_write_end(file, mapping, pos,
+					len, copied, page, fsdata);
+		} else if (ext4_should_writeback_data(inode)) {
+			return ext4_writeback_write_end(file, mapping, pos,
+					len, copied, page, fsdata);
+		} else {
+			BUG();
+		}
+	}
 
 	start = pos & (PAGE_CACHE_SIZE - 1);
-	end = start + copied -1;
+	end = start + copied - 1;
 
 	/*
 	 * generic_write_end() will run mark_inode_dirty() if i_size
@@ -2500,6 +2629,11 @@ static int ext4_da_write_end(struct file *file,
 				EXT4_I(inode)->i_disksize = new_i_size;
 			}
 			up_write(&EXT4_I(inode)->i_data_sem);
+			/* We need to mark inode dirty even if
+			 * new_i_size is less that inode->i_size
+			 * bu greater than i_disksize.(hint delalloc)
+			 */
+			ext4_mark_inode_dirty(handle, inode);
 		}
 	}
 	ret2 = generic_write_end(file, mapping, pos, len, copied,
@@ -2591,7 +2725,7 @@ static sector_t ext4_bmap(struct address_space *mapping, sector_t block)
 			return 0;
 	}
 
-	return generic_block_bmap(mapping,block,ext4_get_block);
+	return generic_block_bmap(mapping, block, ext4_get_block);
 }
 
 static int bget_one(handle_t *handle, struct buffer_head *bh)
@@ -3197,7 +3331,7 @@ static Indirect *ext4_find_shared(struct inode *inode, int depth,
 	if (!partial->key && *partial->p)
 		/* Writer: end */
 		goto no_top;
-	for (p=partial; p>chain && all_zeroes((__le32*)p->bh->b_data,p->p); p--)
+	for (p = partial; (p > chain) && all_zeroes((__le32 *) p->bh->b_data, p->p); p--)
 		;
 	/*
 	 * OK, we've found the last block that must survive. The rest of our
@@ -3216,7 +3350,7 @@ static Indirect *ext4_find_shared(struct inode *inode, int depth,
 	}
 	/* Writer: end */
 
-	while(partial > p) {
+	while (partial > p) {
 		brelse(partial->bh);
 		partial--;
 	}
@@ -3408,9 +3542,9 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode,
 			/* This zaps the entire block.  Bottom up. */
 			BUFFER_TRACE(bh, "free child branches");
 			ext4_free_branches(handle, inode, bh,
-					   (__le32*)bh->b_data,
-					   (__le32*)bh->b_data + addr_per_block,
-					   depth);
+					(__le32 *) bh->b_data,
+					(__le32 *) bh->b_data + addr_per_block,
+					depth);
 
 			/*
 			 * We've probably journalled the indirect block several
@@ -3578,7 +3712,7 @@ void ext4_truncate(struct inode *inode)
 	 */
 	down_write(&ei->i_data_sem);
 
-	ext4_discard_reservation(inode);
+	ext4_discard_preallocations(inode);
 
 	/*
 	 * The orphan list entry will now protect us from any crash which
@@ -3673,41 +3807,6 @@ out_stop:
 	ext4_journal_stop(handle);
 }
 
-static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb,
-		unsigned long ino, struct ext4_iloc *iloc)
-{
-	ext4_group_t block_group;
-	unsigned long offset;
-	ext4_fsblk_t block;
-	struct ext4_group_desc *gdp;
-
-	if (!ext4_valid_inum(sb, ino)) {
-		/*
-		 * This error is already checked for in namei.c unless we are
-		 * looking at an NFS filehandle, in which case no error
-		 * report is needed
-		 */
-		return 0;
-	}
-
-	block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
-	gdp = ext4_get_group_desc(sb, block_group, NULL);
-	if (!gdp)
-		return 0;
-
-	/*
-	 * Figure out the offset within the block group inode table
-	 */
-	offset = ((ino - 1) % EXT4_INODES_PER_GROUP(sb)) *
-		EXT4_INODE_SIZE(sb);
-	block = ext4_inode_table(sb, gdp) +
-		(offset >> EXT4_BLOCK_SIZE_BITS(sb));
-
-	iloc->block_group = block_group;
-	iloc->offset = offset & (EXT4_BLOCK_SIZE(sb) - 1);
-	return block;
-}
-
 /*
  * ext4_get_inode_loc returns with an extra refcount against the inode's
  * underlying buffer_head on success. If 'in_mem' is true, we have all
@@ -3717,19 +3816,35 @@ static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb,
 static int __ext4_get_inode_loc(struct inode *inode,
 				struct ext4_iloc *iloc, int in_mem)
 {
-	ext4_fsblk_t block;
-	struct buffer_head *bh;
+	struct ext4_group_desc	*gdp;
+	struct buffer_head	*bh;
+	struct super_block	*sb = inode->i_sb;
+	ext4_fsblk_t		block;
+	int			inodes_per_block, inode_offset;
+
+	iloc->bh = 0;
+	if (!ext4_valid_inum(sb, inode->i_ino))
+		return -EIO;
 
-	block = ext4_get_inode_block(inode->i_sb, inode->i_ino, iloc);
-	if (!block)
+	iloc->block_group = (inode->i_ino - 1) / EXT4_INODES_PER_GROUP(sb);
+	gdp = ext4_get_group_desc(sb, iloc->block_group, NULL);
+	if (!gdp)
 		return -EIO;
 
-	bh = sb_getblk(inode->i_sb, block);
+	/*
+	 * Figure out the offset within the block group inode table
+	 */
+	inodes_per_block = (EXT4_BLOCK_SIZE(sb) / EXT4_INODE_SIZE(sb));
+	inode_offset = ((inode->i_ino - 1) %
+			EXT4_INODES_PER_GROUP(sb));
+	block = ext4_inode_table(sb, gdp) + (inode_offset / inodes_per_block);
+	iloc->offset = (inode_offset % inodes_per_block) * EXT4_INODE_SIZE(sb);
+
+	bh = sb_getblk(sb, block);
 	if (!bh) {
-		ext4_error (inode->i_sb, "ext4_get_inode_loc",
-				"unable to read inode block - "
-				"inode=%lu, block=%llu",
-				 inode->i_ino, block);
+		ext4_error(sb, "ext4_get_inode_loc", "unable to read "
+			   "inode block - inode=%lu, block=%llu",
+			   inode->i_ino, block);
 		return -EIO;
 	}
 	if (!buffer_uptodate(bh)) {
@@ -3757,28 +3872,12 @@ static int __ext4_get_inode_loc(struct inode *inode,
 		 */
 		if (in_mem) {
 			struct buffer_head *bitmap_bh;
-			struct ext4_group_desc *desc;
-			int inodes_per_buffer;
-			int inode_offset, i;
-			ext4_group_t block_group;
-			int start;
-
-			block_group = (inode->i_ino - 1) /
-					EXT4_INODES_PER_GROUP(inode->i_sb);
-			inodes_per_buffer = bh->b_size /
-				EXT4_INODE_SIZE(inode->i_sb);
-			inode_offset = ((inode->i_ino - 1) %
-					EXT4_INODES_PER_GROUP(inode->i_sb));
-			start = inode_offset & ~(inodes_per_buffer - 1);
+			int i, start;
 
-			/* Is the inode bitmap in cache? */
-			desc = ext4_get_group_desc(inode->i_sb,
-						block_group, NULL);
-			if (!desc)
-				goto make_io;
+			start = inode_offset & ~(inodes_per_block - 1);
 
-			bitmap_bh = sb_getblk(inode->i_sb,
-				ext4_inode_bitmap(inode->i_sb, desc));
+			/* Is the inode bitmap in cache? */
+			bitmap_bh = sb_getblk(sb, ext4_inode_bitmap(sb, gdp));
 			if (!bitmap_bh)
 				goto make_io;
 
@@ -3791,14 +3890,14 @@ static int __ext4_get_inode_loc(struct inode *inode,
 				brelse(bitmap_bh);
 				goto make_io;
 			}
-			for (i = start; i < start + inodes_per_buffer; i++) {
+			for (i = start; i < start + inodes_per_block; i++) {
 				if (i == inode_offset)
 					continue;
 				if (ext4_test_bit(i, bitmap_bh->b_data))
 					break;
 			}
 			brelse(bitmap_bh);
-			if (i == start + inodes_per_buffer) {
+			if (i == start + inodes_per_block) {
 				/* all other inodes are free, so skip I/O */
 				memset(bh->b_data, 0, bh->b_size);
 				set_buffer_uptodate(bh);
@@ -3809,6 +3908,36 @@ static int __ext4_get_inode_loc(struct inode *inode,
 
 make_io:
 		/*
+		 * If we need to do any I/O, try to pre-readahead extra
+		 * blocks from the inode table.
+		 */
+		if (EXT4_SB(sb)->s_inode_readahead_blks) {
+			ext4_fsblk_t b, end, table;
+			unsigned num;
+
+			table = ext4_inode_table(sb, gdp);
+			/* Make sure s_inode_readahead_blks is a power of 2 */
+			while (EXT4_SB(sb)->s_inode_readahead_blks &
+			       (EXT4_SB(sb)->s_inode_readahead_blks-1))
+				EXT4_SB(sb)->s_inode_readahead_blks = 
+				   (EXT4_SB(sb)->s_inode_readahead_blks &
+				    (EXT4_SB(sb)->s_inode_readahead_blks-1));
+			b = block & ~(EXT4_SB(sb)->s_inode_readahead_blks-1);
+			if (table > b)
+				b = table;
+			end = b + EXT4_SB(sb)->s_inode_readahead_blks;
+			num = EXT4_INODES_PER_GROUP(sb);
+			if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
+				       EXT4_FEATURE_RO_COMPAT_GDT_CSUM))
+				num -= le16_to_cpu(gdp->bg_itable_unused);
+			table += num / inodes_per_block;
+			if (end > table)
+				end = table;
+			while (b <= end)
+				sb_breadahead(sb, b++);
+		}
+
+		/*
 		 * There are other valid inodes in the buffer, this inode
 		 * has in-inode xattrs, or we don't have this inode in memory.
 		 * Read the block from disk.
@@ -3818,10 +3947,9 @@ make_io:
 		submit_bh(READ_META, bh);
 		wait_on_buffer(bh);
 		if (!buffer_uptodate(bh)) {
-			ext4_error(inode->i_sb, "ext4_get_inode_loc",
-					"unable to read inode block - "
-					"inode=%lu, block=%llu",
-					inode->i_ino, block);
+			ext4_error(sb, __func__,
+				   "unable to read inode block - inode=%lu, "
+				   "block=%llu", inode->i_ino, block);
 			brelse(bh);
 			return -EIO;
 		}
@@ -3913,11 +4041,10 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 		return inode;
 
 	ei = EXT4_I(inode);
-#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
+#ifdef CONFIG_EXT4_FS_POSIX_ACL
 	ei->i_acl = EXT4_ACL_NOT_CACHED;
 	ei->i_default_acl = EXT4_ACL_NOT_CACHED;
 #endif
-	ei->i_block_alloc_info = NULL;
 
 	ret = __ext4_get_inode_loc(inode, &iloc, 0);
 	if (ret < 0)
@@ -3927,7 +4054,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 	inode->i_mode = le16_to_cpu(raw_inode->i_mode);
 	inode->i_uid = (uid_t)le16_to_cpu(raw_inode->i_uid_low);
 	inode->i_gid = (gid_t)le16_to_cpu(raw_inode->i_gid_low);
-	if(!(test_opt (inode->i_sb, NO_UID32))) {
+	if (!(test_opt(inode->i_sb, NO_UID32))) {
 		inode->i_uid |= le16_to_cpu(raw_inode->i_uid_high) << 16;
 		inode->i_gid |= le16_to_cpu(raw_inode->i_gid_high) << 16;
 	}
@@ -3945,7 +4072,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 		if (inode->i_mode == 0 ||
 		    !(EXT4_SB(inode->i_sb)->s_mount_state & EXT4_ORPHAN_FS)) {
 			/* this inode is deleted */
-			brelse (bh);
+			brelse(bh);
 			ret = -ESTALE;
 			goto bad_inode;
 		}
@@ -3978,7 +4105,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 		ei->i_extra_isize = le16_to_cpu(raw_inode->i_extra_isize);
 		if (EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize >
 		    EXT4_INODE_SIZE(inode->i_sb)) {
-			brelse (bh);
+			brelse(bh);
 			ret = -EIO;
 			goto bad_inode;
 		}
@@ -4031,7 +4158,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 			init_special_inode(inode, inode->i_mode,
 			   new_decode_dev(le32_to_cpu(raw_inode->i_block[1])));
 	}
-	brelse (iloc.bh);
+	brelse(iloc.bh);
 	ext4_set_inode_flags(inode);
 	unlock_new_inode(inode);
 	return inode;
@@ -4113,14 +4240,14 @@ static int ext4_do_update_inode(handle_t *handle,
 
 	ext4_get_inode_flags(ei);
 	raw_inode->i_mode = cpu_to_le16(inode->i_mode);
-	if(!(test_opt(inode->i_sb, NO_UID32))) {
+	if (!(test_opt(inode->i_sb, NO_UID32))) {
 		raw_inode->i_uid_low = cpu_to_le16(low_16_bits(inode->i_uid));
 		raw_inode->i_gid_low = cpu_to_le16(low_16_bits(inode->i_gid));
 /*
  * Fix up interoperability with old kernels. Otherwise, old inodes get
  * re-used with the upper 16 bits of the uid/gid intact
  */
-		if(!ei->i_dtime) {
+		if (!ei->i_dtime) {
 			raw_inode->i_uid_high =
 				cpu_to_le16(high_16_bits(inode->i_uid));
 			raw_inode->i_gid_high =
@@ -4208,7 +4335,7 @@ static int ext4_do_update_inode(handle_t *handle,
 	ei->i_state &= ~EXT4_STATE_NEW;
 
 out_brelse:
-	brelse (bh);
+	brelse(bh);
 	ext4_std_error(inode->i_sb, err);
 	return err;
 }
@@ -4811,6 +4938,7 @@ int ext4_page_mkwrite(struct vm_area_struct *vma, struct page *page)
 	loff_t size;
 	unsigned long len;
 	int ret = -EINVAL;
+	void *fsdata;
 	struct file *file = vma->vm_file;
 	struct inode *inode = file->f_path.dentry->d_inode;
 	struct address_space *mapping = inode->i_mapping;
@@ -4849,11 +4977,11 @@ int ext4_page_mkwrite(struct vm_area_struct *vma, struct page *page)
 	 * on the same page though
 	 */
 	ret = mapping->a_ops->write_begin(file, mapping, page_offset(page),
-			len, AOP_FLAG_UNINTERRUPTIBLE, &page, NULL);
+			len, AOP_FLAG_UNINTERRUPTIBLE, &page, &fsdata);
 	if (ret < 0)
 		goto out_unlock;
 	ret = mapping->a_ops->write_end(file, mapping, page_offset(page),
-			len, len, page, NULL);
+			len, len, page, fsdata);
 	if (ret < 0)
 		goto out_unlock;
 	ret = 0;
diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c
index 7a6c2f1faba6..dc99b4776d58 100644
--- a/fs/ext4/ioctl.c
+++ b/fs/ext4/ioctl.c
@@ -23,9 +23,8 @@ long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	struct inode *inode = filp->f_dentry->d_inode;
 	struct ext4_inode_info *ei = EXT4_I(inode);
 	unsigned int flags;
-	unsigned short rsv_window_size;
 
-	ext4_debug ("cmd = %u, arg = %lu\n", cmd, arg);
+	ext4_debug("cmd = %u, arg = %lu\n", cmd, arg);
 
 	switch (cmd) {
 	case EXT4_IOC_GETFLAGS:
@@ -34,7 +33,7 @@ long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 		return put_user(flags, (int __user *) arg);
 	case EXT4_IOC_SETFLAGS: {
 		handle_t *handle = NULL;
-		int err;
+		int err, migrate = 0;
 		struct ext4_iloc iloc;
 		unsigned int oldflags;
 		unsigned int jflag;
@@ -82,6 +81,17 @@ long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 			if (!capable(CAP_SYS_RESOURCE))
 				goto flags_out;
 		}
+		if (oldflags & EXT4_EXTENTS_FL) {
+			/* We don't support clearning extent flags */
+			if (!(flags & EXT4_EXTENTS_FL)) {
+				err = -EOPNOTSUPP;
+				goto flags_out;
+			}
+		} else if (flags & EXT4_EXTENTS_FL) {
+			/* migrate the file */
+			migrate = 1;
+			flags &= ~EXT4_EXTENTS_FL;
+		}
 
 		handle = ext4_journal_start(inode, 1);
 		if (IS_ERR(handle)) {
@@ -109,6 +119,10 @@ flags_err:
 
 		if ((jflag ^ oldflags) & (EXT4_JOURNAL_DATA_FL))
 			err = ext4_change_inode_journal_flag(inode, jflag);
+		if (err)
+			goto flags_out;
+		if (migrate)
+			err = ext4_ext_migrate(inode);
 flags_out:
 		mutex_unlock(&inode->i_mutex);
 		mnt_drop_write(filp->f_path.mnt);
@@ -175,53 +189,10 @@ setversion_out:
 			return ret;
 		}
 #endif
-	case EXT4_IOC_GETRSVSZ:
-		if (test_opt(inode->i_sb, RESERVATION)
-			&& S_ISREG(inode->i_mode)
-			&& ei->i_block_alloc_info) {
-			rsv_window_size = ei->i_block_alloc_info->rsv_window_node.rsv_goal_size;
-			return put_user(rsv_window_size, (int __user *)arg);
-		}
-		return -ENOTTY;
-	case EXT4_IOC_SETRSVSZ: {
-		int err;
-
-		if (!test_opt(inode->i_sb, RESERVATION) ||!S_ISREG(inode->i_mode))
-			return -ENOTTY;
-
-		if (!is_owner_or_cap(inode))
-			return -EACCES;
-
-		if (get_user(rsv_window_size, (int __user *)arg))
-			return -EFAULT;
-
-		err = mnt_want_write(filp->f_path.mnt);
-		if (err)
-			return err;
-
-		if (rsv_window_size > EXT4_MAX_RESERVE_BLOCKS)
-			rsv_window_size = EXT4_MAX_RESERVE_BLOCKS;
-
-		/*
-		 * need to allocate reservation structure for this inode
-		 * before set the window size
-		 */
-		down_write(&ei->i_data_sem);
-		if (!ei->i_block_alloc_info)
-			ext4_init_block_alloc_info(inode);
-
-		if (ei->i_block_alloc_info){
-			struct ext4_reserve_window_node *rsv = &ei->i_block_alloc_info->rsv_window_node;
-			rsv->rsv_goal_size = rsv_window_size;
-		}
-		up_write(&ei->i_data_sem);
-		mnt_drop_write(filp->f_path.mnt);
-		return 0;
-	}
 	case EXT4_IOC_GROUP_EXTEND: {
 		ext4_fsblk_t n_blocks_count;
 		struct super_block *sb = inode->i_sb;
-		int err;
+		int err, err2;
 
 		if (!capable(CAP_SYS_RESOURCE))
 			return -EPERM;
@@ -235,8 +206,10 @@ setversion_out:
 
 		err = ext4_group_extend(sb, EXT4_SB(sb)->s_es, n_blocks_count);
 		jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal);
-		jbd2_journal_flush(EXT4_SB(sb)->s_journal);
+		err2 = jbd2_journal_flush(EXT4_SB(sb)->s_journal);
 		jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal);
+		if (err == 0)
+			err = err2;
 		mnt_drop_write(filp->f_path.mnt);
 
 		return err;
@@ -244,7 +217,7 @@ setversion_out:
 	case EXT4_IOC_GROUP_ADD: {
 		struct ext4_new_group_data input;
 		struct super_block *sb = inode->i_sb;
-		int err;
+		int err, err2;
 
 		if (!capable(CAP_SYS_RESOURCE))
 			return -EPERM;
@@ -259,15 +232,36 @@ setversion_out:
 
 		err = ext4_group_add(sb, &input);
 		jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal);
-		jbd2_journal_flush(EXT4_SB(sb)->s_journal);
+		err2 = jbd2_journal_flush(EXT4_SB(sb)->s_journal);
 		jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal);
+		if (err == 0)
+			err = err2;
 		mnt_drop_write(filp->f_path.mnt);
 
 		return err;
 	}
 
 	case EXT4_IOC_MIGRATE:
-		return ext4_ext_migrate(inode, filp, cmd, arg);
+	{
+		int err;
+		if (!is_owner_or_cap(inode))
+			return -EACCES;
+
+		err = mnt_want_write(filp->f_path.mnt);
+		if (err)
+			return err;
+		/*
+		 * inode_mutex prevent write and truncate on the file.
+		 * Read still goes through. We take i_data_sem in
+		 * ext4_ext_swap_inode_data before we switch the
+		 * inode format to prevent read.
+		 */
+		mutex_lock(&(inode->i_mutex));
+		err = ext4_ext_migrate(inode);
+		mutex_unlock(&(inode->i_mutex));
+		mnt_drop_write(filp->f_path.mnt);
+		return err;
+	}
 
 	default:
 		return -ENOTTY;
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index e0e3a5eb1ddb..b580714f0d85 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -477,9 +477,10 @@ static void mb_cmp_bitmaps(struct ext4_buddy *e4b, void *bitmap)
 		b2 = (unsigned char *) bitmap;
 		for (i = 0; i < e4b->bd_sb->s_blocksize; i++) {
 			if (b1[i] != b2[i]) {
-				printk("corruption in group %lu at byte %u(%u):"
-				       " %x in copy != %x on disk/prealloc\n",
-					e4b->bd_group, i, i * 8, b1[i], b2[i]);
+				printk(KERN_ERR "corruption in group %lu "
+				       "at byte %u(%u): %x in copy != %x "
+				       "on disk/prealloc\n",
+				       e4b->bd_group, i, i * 8, b1[i], b2[i]);
 				BUG();
 			}
 		}
@@ -533,9 +534,6 @@ static int __mb_check_buddy(struct ext4_buddy *e4b, char *file,
 	void *buddy;
 	void *buddy2;
 
-	if (!test_opt(sb, MBALLOC))
-		return 0;
-
 	{
 		static int mb_check_counter;
 		if (mb_check_counter++ % 100 != 0)
@@ -784,9 +782,11 @@ static int ext4_mb_init_cache(struct page *page, char *incore)
 		if (bh[i] == NULL)
 			goto out;
 
-		if (bh_uptodate_or_lock(bh[i]))
+		if (buffer_uptodate(bh[i]) &&
+		    !(desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)))
 			continue;
 
+		lock_buffer(bh[i]);
 		spin_lock(sb_bgl_lock(EXT4_SB(sb), first_group + i));
 		if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
 			ext4_init_block_bitmap(sb, bh[i],
@@ -2169,9 +2169,10 @@ static void ext4_mb_history_release(struct super_block *sb)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 
-	remove_proc_entry("mb_groups", sbi->s_mb_proc);
-	remove_proc_entry("mb_history", sbi->s_mb_proc);
-
+	if (sbi->s_proc != NULL) {
+		remove_proc_entry("mb_groups", sbi->s_proc);
+		remove_proc_entry("mb_history", sbi->s_proc);
+	}
 	kfree(sbi->s_mb_history);
 }
 
@@ -2180,10 +2181,10 @@ static void ext4_mb_history_init(struct super_block *sb)
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	int i;
 
-	if (sbi->s_mb_proc != NULL) {
-		proc_create_data("mb_history", S_IRUGO, sbi->s_mb_proc,
+	if (sbi->s_proc != NULL) {
+		proc_create_data("mb_history", S_IRUGO, sbi->s_proc,
 				 &ext4_mb_seq_history_fops, sb);
-		proc_create_data("mb_groups", S_IRUGO, sbi->s_mb_proc,
+		proc_create_data("mb_groups", S_IRUGO, sbi->s_proc,
 				 &ext4_mb_seq_groups_fops, sb);
 	}
 
@@ -2485,19 +2486,14 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery)
 	unsigned max;
 	int ret;
 
-	if (!test_opt(sb, MBALLOC))
-		return 0;
-
 	i = (sb->s_blocksize_bits + 2) * sizeof(unsigned short);
 
 	sbi->s_mb_offsets = kmalloc(i, GFP_KERNEL);
 	if (sbi->s_mb_offsets == NULL) {
-		clear_opt(sbi->s_mount_opt, MBALLOC);
 		return -ENOMEM;
 	}
 	sbi->s_mb_maxs = kmalloc(i, GFP_KERNEL);
 	if (sbi->s_mb_maxs == NULL) {
-		clear_opt(sbi->s_mount_opt, MBALLOC);
 		kfree(sbi->s_mb_maxs);
 		return -ENOMEM;
 	}
@@ -2520,7 +2516,6 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery)
 	/* init file for buddy data */
 	ret = ext4_mb_init_backend(sb);
 	if (ret != 0) {
-		clear_opt(sbi->s_mount_opt, MBALLOC);
 		kfree(sbi->s_mb_offsets);
 		kfree(sbi->s_mb_maxs);
 		return ret;
@@ -2540,17 +2535,15 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery)
 	sbi->s_mb_history_filter = EXT4_MB_HISTORY_DEFAULT;
 	sbi->s_mb_group_prealloc = MB_DEFAULT_GROUP_PREALLOC;
 
-	i = sizeof(struct ext4_locality_group) * nr_cpu_ids;
-	sbi->s_locality_groups = kmalloc(i, GFP_KERNEL);
+	sbi->s_locality_groups = alloc_percpu(struct ext4_locality_group);
 	if (sbi->s_locality_groups == NULL) {
-		clear_opt(sbi->s_mount_opt, MBALLOC);
 		kfree(sbi->s_mb_offsets);
 		kfree(sbi->s_mb_maxs);
 		return -ENOMEM;
 	}
-	for (i = 0; i < nr_cpu_ids; i++) {
+	for_each_possible_cpu(i) {
 		struct ext4_locality_group *lg;
-		lg = &sbi->s_locality_groups[i];
+		lg = per_cpu_ptr(sbi->s_locality_groups, i);
 		mutex_init(&lg->lg_mutex);
 		for (j = 0; j < PREALLOC_TB_SIZE; j++)
 			INIT_LIST_HEAD(&lg->lg_prealloc_list[j]);
@@ -2560,7 +2553,7 @@ int ext4_mb_init(struct super_block *sb, int needs_recovery)
 	ext4_mb_init_per_dev_proc(sb);
 	ext4_mb_history_init(sb);
 
-	printk("EXT4-fs: mballoc enabled\n");
+	printk(KERN_INFO "EXT4-fs: mballoc enabled\n");
 	return 0;
 }
 
@@ -2589,9 +2582,6 @@ int ext4_mb_release(struct super_block *sb)
 	struct ext4_group_info *grinfo;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 
-	if (!test_opt(sb, MBALLOC))
-		return 0;
-
 	/* release freed, non-committed blocks */
 	spin_lock(&sbi->s_md_lock);
 	list_splice_init(&sbi->s_closed_transaction,
@@ -2647,8 +2637,7 @@ int ext4_mb_release(struct super_block *sb)
 				atomic_read(&sbi->s_mb_discarded));
 	}
 
-	kfree(sbi->s_locality_groups);
-
+	free_percpu(sbi->s_locality_groups);
 	ext4_mb_history_release(sb);
 	ext4_mb_destroy_per_dev_proc(sb);
 
@@ -2721,118 +2710,46 @@ ext4_mb_free_committed_blocks(struct super_block *sb)
 #define EXT4_MB_STREAM_REQ		"stream_req"
 #define EXT4_MB_GROUP_PREALLOC		"group_prealloc"
 
-
-
-#define MB_PROC_FOPS(name)					\
-static int ext4_mb_##name##_proc_show(struct seq_file *m, void *v)	\
-{								\
-	struct ext4_sb_info *sbi = m->private;			\
-								\
-	seq_printf(m, "%ld\n", sbi->s_mb_##name);		\
-	return 0;						\
-}								\
-								\
-static int ext4_mb_##name##_proc_open(struct inode *inode, struct file *file)\
-{								\
-	return single_open(file, ext4_mb_##name##_proc_show, PDE(inode)->data);\
-}								\
-								\
-static ssize_t ext4_mb_##name##_proc_write(struct file *file,	\
-		const char __user *buf, size_t cnt, loff_t *ppos)	\
-{								\
-	struct ext4_sb_info *sbi = PDE(file->f_path.dentry->d_inode)->data;\
-	char str[32];						\
-	long value;						\
-	if (cnt >= sizeof(str))					\
-		return -EINVAL;					\
-	if (copy_from_user(str, buf, cnt))			\
-		return -EFAULT;					\
-	value = simple_strtol(str, NULL, 0);			\
-	if (value <= 0)						\
-		return -ERANGE;					\
-	sbi->s_mb_##name = value;				\
-	return cnt;						\
-}								\
-								\
-static const struct file_operations ext4_mb_##name##_proc_fops = {	\
-	.owner		= THIS_MODULE,				\
-	.open		= ext4_mb_##name##_proc_open,		\
-	.read		= seq_read,				\
-	.llseek		= seq_lseek,				\
-	.release	= single_release,			\
-	.write		= ext4_mb_##name##_proc_write,		\
-};
-
-MB_PROC_FOPS(stats);
-MB_PROC_FOPS(max_to_scan);
-MB_PROC_FOPS(min_to_scan);
-MB_PROC_FOPS(order2_reqs);
-MB_PROC_FOPS(stream_request);
-MB_PROC_FOPS(group_prealloc);
-
-#define	MB_PROC_HANDLER(name, var)					\
-do {									\
-	proc = proc_create_data(name, mode, sbi->s_mb_proc,		\
-				&ext4_mb_##var##_proc_fops, sbi);	\
-	if (proc == NULL) {						\
-		printk(KERN_ERR "EXT4-fs: can't to create %s\n", name);	\
-		goto err_out;						\
-	}								\
-} while (0)
-
 static int ext4_mb_init_per_dev_proc(struct super_block *sb)
 {
 	mode_t mode = S_IFREG | S_IRUGO | S_IWUSR;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	struct proc_dir_entry *proc;
-	char devname[64];
 
-	if (proc_root_ext4 == NULL) {
-		sbi->s_mb_proc = NULL;
+	if (sbi->s_proc == NULL)
 		return -EINVAL;
-	}
-	bdevname(sb->s_bdev, devname);
-	sbi->s_mb_proc = proc_mkdir(devname, proc_root_ext4);
-
-	MB_PROC_HANDLER(EXT4_MB_STATS_NAME, stats);
-	MB_PROC_HANDLER(EXT4_MB_MAX_TO_SCAN_NAME, max_to_scan);
-	MB_PROC_HANDLER(EXT4_MB_MIN_TO_SCAN_NAME, min_to_scan);
-	MB_PROC_HANDLER(EXT4_MB_ORDER2_REQ, order2_reqs);
-	MB_PROC_HANDLER(EXT4_MB_STREAM_REQ, stream_request);
-	MB_PROC_HANDLER(EXT4_MB_GROUP_PREALLOC, group_prealloc);
 
+	EXT4_PROC_HANDLER(EXT4_MB_STATS_NAME, mb_stats);
+	EXT4_PROC_HANDLER(EXT4_MB_MAX_TO_SCAN_NAME, mb_max_to_scan);
+	EXT4_PROC_HANDLER(EXT4_MB_MIN_TO_SCAN_NAME, mb_min_to_scan);
+	EXT4_PROC_HANDLER(EXT4_MB_ORDER2_REQ, mb_order2_reqs);
+	EXT4_PROC_HANDLER(EXT4_MB_STREAM_REQ, mb_stream_request);
+	EXT4_PROC_HANDLER(EXT4_MB_GROUP_PREALLOC, mb_group_prealloc);
 	return 0;
 
 err_out:
-	printk(KERN_ERR "EXT4-fs: Unable to create %s\n", devname);
-	remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc);
-	remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc);
-	remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc);
-	remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_mb_proc);
-	remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_mb_proc);
-	remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_mb_proc);
-	remove_proc_entry(devname, proc_root_ext4);
-	sbi->s_mb_proc = NULL;
-
+	remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_proc);
+	remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_proc);
+	remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_proc);
+	remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_proc);
+	remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_proc);
+	remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_proc);
 	return -ENOMEM;
 }
 
 static int ext4_mb_destroy_per_dev_proc(struct super_block *sb)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
-	char devname[64];
 
-	if (sbi->s_mb_proc == NULL)
+	if (sbi->s_proc == NULL)
 		return -EINVAL;
 
-	bdevname(sb->s_bdev, devname);
-	remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc);
-	remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc);
-	remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc);
-	remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_mb_proc);
-	remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_mb_proc);
-	remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_mb_proc);
-	remove_proc_entry(devname, proc_root_ext4);
+	remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_proc);
+	remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_proc);
+	remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_proc);
+	remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_proc);
+	remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_proc);
+	remove_proc_entry(EXT4_MB_STATS_NAME, sbi->s_proc);
 
 	return 0;
 }
@@ -2854,11 +2771,6 @@ int __init init_ext4_mballoc(void)
 		kmem_cache_destroy(ext4_pspace_cachep);
 		return -ENOMEM;
 	}
-#ifdef CONFIG_PROC_FS
-	proc_root_ext4 = proc_mkdir("fs/ext4", NULL);
-	if (proc_root_ext4 == NULL)
-		printk(KERN_ERR "EXT4-fs: Unable to create fs/ext4\n");
-#endif
 	return 0;
 }
 
@@ -2867,9 +2779,6 @@ void exit_ext4_mballoc(void)
 	/* XXX: synchronize_rcu(); */
 	kmem_cache_destroy(ext4_pspace_cachep);
 	kmem_cache_destroy(ext4_ac_cachep);
-#ifdef CONFIG_PROC_FS
-	remove_proc_entry("fs/ext4", NULL);
-#endif
 }
 
 
@@ -2879,7 +2788,7 @@ void exit_ext4_mballoc(void)
  */
 static noinline_for_stack int
 ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
-				handle_t *handle)
+				handle_t *handle, unsigned long reserv_blks)
 {
 	struct buffer_head *bitmap_bh = NULL;
 	struct ext4_super_block *es;
@@ -2968,15 +2877,16 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
 	le16_add_cpu(&gdp->bg_free_blocks_count, -ac->ac_b_ex.fe_len);
 	gdp->bg_checksum = ext4_group_desc_csum(sbi, ac->ac_b_ex.fe_group, gdp);
 	spin_unlock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group));
-
+	percpu_counter_sub(&sbi->s_freeblocks_counter, ac->ac_b_ex.fe_len);
 	/*
-	 * free blocks account has already be reduced/reserved
-	 * at write_begin() time for delayed allocation
-	 * do not double accounting
+	 * Now reduce the dirty block count also. Should not go negative
 	 */
 	if (!(ac->ac_flags & EXT4_MB_DELALLOC_RESERVED))
-		percpu_counter_sub(&sbi->s_freeblocks_counter,
-					ac->ac_b_ex.fe_len);
+		/* release all the reserved blocks if non delalloc */
+		percpu_counter_sub(&sbi->s_dirtyblocks_counter, reserv_blks);
+	else
+		percpu_counter_sub(&sbi->s_dirtyblocks_counter,
+						ac->ac_b_ex.fe_len);
 
 	if (sbi->s_log_groups_per_flex) {
 		ext4_group_t flex_group = ext4_flex_group(sbi,
@@ -3884,7 +3794,7 @@ out:
  *
  * FIXME!! Make sure it is valid at all the call sites
  */
-void ext4_mb_discard_inode_preallocations(struct inode *inode)
+void ext4_discard_preallocations(struct inode *inode)
 {
 	struct ext4_inode_info *ei = EXT4_I(inode);
 	struct super_block *sb = inode->i_sb;
@@ -3896,7 +3806,7 @@ void ext4_mb_discard_inode_preallocations(struct inode *inode)
 	struct ext4_buddy e4b;
 	int err;
 
-	if (!test_opt(sb, MBALLOC) || !S_ISREG(inode->i_mode)) {
+	if (!S_ISREG(inode->i_mode)) {
 		/*BUG_ON(!list_empty(&ei->i_prealloc_list));*/
 		return;
 	}
@@ -4094,8 +4004,7 @@ static void ext4_mb_group_or_file(struct ext4_allocation_context *ac)
 	 * per cpu locality group is to reduce the contention between block
 	 * request from multiple CPUs.
 	 */
-	ac->ac_lg = &sbi->s_locality_groups[get_cpu()];
-	put_cpu();
+	ac->ac_lg = per_cpu_ptr(sbi->s_locality_groups, raw_smp_processor_id());
 
 	/* we're going to use group allocation */
 	ac->ac_flags |= EXT4_MB_HINT_GROUP_ALLOC;
@@ -4369,33 +4278,32 @@ static int ext4_mb_discard_preallocations(struct super_block *sb, int needed)
 ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle,
 				 struct ext4_allocation_request *ar, int *errp)
 {
+	int freed;
 	struct ext4_allocation_context *ac = NULL;
 	struct ext4_sb_info *sbi;
 	struct super_block *sb;
 	ext4_fsblk_t block = 0;
-	int freed;
-	int inquota;
+	unsigned long inquota;
+	unsigned long reserv_blks = 0;
 
 	sb = ar->inode->i_sb;
 	sbi = EXT4_SB(sb);
 
-	if (!test_opt(sb, MBALLOC)) {
-		block = ext4_old_new_blocks(handle, ar->inode, ar->goal,
-					    &(ar->len), errp);
-		return block;
-	}
 	if (!EXT4_I(ar->inode)->i_delalloc_reserved_flag) {
 		/*
 		 * With delalloc we already reserved the blocks
 		 */
-		ar->len = ext4_has_free_blocks(sbi, ar->len);
-	}
-
-	if (ar->len == 0) {
-		*errp = -ENOSPC;
-		return 0;
+		while (ar->len && ext4_claim_free_blocks(sbi, ar->len)) {
+			/* let others to free the space */
+			yield();
+			ar->len = ar->len >> 1;
+		}
+		if (!ar->len) {
+			*errp = -ENOSPC;
+			return 0;
+		}
+		reserv_blks = ar->len;
 	}
-
 	while (ar->len && DQUOT_ALLOC_BLOCK(ar->inode, ar->len)) {
 		ar->flags |= EXT4_MB_HINT_NOPREALLOC;
 		ar->len--;
@@ -4441,7 +4349,7 @@ repeat:
 	}
 
 	if (likely(ac->ac_status == AC_STATUS_FOUND)) {
-		*errp = ext4_mb_mark_diskspace_used(ac, handle);
+		*errp = ext4_mb_mark_diskspace_used(ac, handle, reserv_blks);
 		if (*errp ==  -EAGAIN) {
 			ac->ac_b_ex.fe_group = 0;
 			ac->ac_b_ex.fe_start = 0;
diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h
index c7c9906c2a75..b3b4828f8b89 100644
--- a/fs/ext4/mballoc.h
+++ b/fs/ext4/mballoc.h
@@ -257,7 +257,6 @@ static void ext4_mb_store_history(struct ext4_allocation_context *ac);
 
 #define in_range(b, first, len)	((b) >= (first) && (b) <= (first) + (len) - 1)
 
-static struct proc_dir_entry *proc_root_ext4;
 struct buffer_head *read_block_bitmap(struct super_block *, ext4_group_t);
 
 static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap,
diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c
index 46fc0b5b12ba..f2a9cf498ecd 100644
--- a/fs/ext4/migrate.c
+++ b/fs/ext4/migrate.c
@@ -447,8 +447,7 @@ static int free_ext_block(handle_t *handle, struct inode *inode)
 
 }
 
-int ext4_ext_migrate(struct inode *inode, struct file *filp,
-				unsigned int cmd, unsigned long arg)
+int ext4_ext_migrate(struct inode *inode)
 {
 	handle_t *handle;
 	int retval = 0, i;
@@ -516,12 +515,6 @@ int ext4_ext_migrate(struct inode *inode, struct file *filp,
 	 * when we add extents we extent the journal
 	 */
 	/*
-	 * inode_mutex prevent write and truncate on the file. Read still goes
-	 * through. We take i_data_sem in ext4_ext_swap_inode_data before we
-	 * switch the inode format to prevent read.
-	 */
-	mutex_lock(&(inode->i_mutex));
-	/*
 	 * Even though we take i_mutex we can still cause block allocation
 	 * via mmap write to holes. If we have allocated new blocks we fail
 	 * migrate.  New block allocation will clear EXT4_EXT_MIGRATE flag.
@@ -623,7 +616,6 @@ err_out:
 	tmp_inode->i_nlink = 0;
 
 	ext4_journal_stop(handle);
-	mutex_unlock(&(inode->i_mutex));
 
 	if (tmp_inode)
 		iput(tmp_inode);
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index 387ad98350c3..92db9e945147 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -151,34 +151,36 @@ struct dx_map_entry
 
 static inline ext4_lblk_t dx_get_block(struct dx_entry *entry);
 static void dx_set_block(struct dx_entry *entry, ext4_lblk_t value);
-static inline unsigned dx_get_hash (struct dx_entry *entry);
-static void dx_set_hash (struct dx_entry *entry, unsigned value);
-static unsigned dx_get_count (struct dx_entry *entries);
-static unsigned dx_get_limit (struct dx_entry *entries);
-static void dx_set_count (struct dx_entry *entries, unsigned value);
-static void dx_set_limit (struct dx_entry *entries, unsigned value);
-static unsigned dx_root_limit (struct inode *dir, unsigned infosize);
-static unsigned dx_node_limit (struct inode *dir);
-static struct dx_frame *dx_probe(struct dentry *dentry,
+static inline unsigned dx_get_hash(struct dx_entry *entry);
+static void dx_set_hash(struct dx_entry *entry, unsigned value);
+static unsigned dx_get_count(struct dx_entry *entries);
+static unsigned dx_get_limit(struct dx_entry *entries);
+static void dx_set_count(struct dx_entry *entries, unsigned value);
+static void dx_set_limit(struct dx_entry *entries, unsigned value);
+static unsigned dx_root_limit(struct inode *dir, unsigned infosize);
+static unsigned dx_node_limit(struct inode *dir);
+static struct dx_frame *dx_probe(const struct qstr *d_name,
 				 struct inode *dir,
 				 struct dx_hash_info *hinfo,
 				 struct dx_frame *frame,
 				 int *err);
-static void dx_release (struct dx_frame *frames);
-static int dx_make_map (struct ext4_dir_entry_2 *de, int size,
-			struct dx_hash_info *hinfo, struct dx_map_entry map[]);
+static void dx_release(struct dx_frame *frames);
+static int dx_make_map(struct ext4_dir_entry_2 *de, int size,
+		       struct dx_hash_info *hinfo, struct dx_map_entry map[]);
 static void dx_sort_map(struct dx_map_entry *map, unsigned count);
-static struct ext4_dir_entry_2 *dx_move_dirents (char *from, char *to,
+static struct ext4_dir_entry_2 *dx_move_dirents(char *from, char *to,
 		struct dx_map_entry *offsets, int count);
-static struct ext4_dir_entry_2* dx_pack_dirents (char *base, int size);
+static struct ext4_dir_entry_2* dx_pack_dirents(char *base, int size);
 static void dx_insert_block(struct dx_frame *frame,
 					u32 hash, ext4_lblk_t block);
 static int ext4_htree_next_block(struct inode *dir, __u32 hash,
 				 struct dx_frame *frame,
 				 struct dx_frame *frames,
 				 __u32 *start_hash);
-static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry,
-		       struct ext4_dir_entry_2 **res_dir, int *err);
+static struct buffer_head * ext4_dx_find_entry(struct inode *dir,
+		const struct qstr *d_name,
+		struct ext4_dir_entry_2 **res_dir,
+		int *err);
 static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
 			     struct inode *inode);
 
@@ -207,44 +209,44 @@ static inline void dx_set_block(struct dx_entry *entry, ext4_lblk_t value)
 	entry->block = cpu_to_le32(value);
 }
 
-static inline unsigned dx_get_hash (struct dx_entry *entry)
+static inline unsigned dx_get_hash(struct dx_entry *entry)
 {
 	return le32_to_cpu(entry->hash);
 }
 
-static inline void dx_set_hash (struct dx_entry *entry, unsigned value)
+static inline void dx_set_hash(struct dx_entry *entry, unsigned value)
 {
 	entry->hash = cpu_to_le32(value);
 }
 
-static inline unsigned dx_get_count (struct dx_entry *entries)
+static inline unsigned dx_get_count(struct dx_entry *entries)
 {
 	return le16_to_cpu(((struct dx_countlimit *) entries)->count);
 }
 
-static inline unsigned dx_get_limit (struct dx_entry *entries)
+static inline unsigned dx_get_limit(struct dx_entry *entries)
 {
 	return le16_to_cpu(((struct dx_countlimit *) entries)->limit);
 }
 
-static inline void dx_set_count (struct dx_entry *entries, unsigned value)
+static inline void dx_set_count(struct dx_entry *entries, unsigned value)
 {
 	((struct dx_countlimit *) entries)->count = cpu_to_le16(value);
 }
 
-static inline void dx_set_limit (struct dx_entry *entries, unsigned value)
+static inline void dx_set_limit(struct dx_entry *entries, unsigned value)
 {
 	((struct dx_countlimit *) entries)->limit = cpu_to_le16(value);
 }
 
-static inline unsigned dx_root_limit (struct inode *dir, unsigned infosize)
+static inline unsigned dx_root_limit(struct inode *dir, unsigned infosize)
 {
 	unsigned entry_space = dir->i_sb->s_blocksize - EXT4_DIR_REC_LEN(1) -
 		EXT4_DIR_REC_LEN(2) - infosize;
 	return entry_space / sizeof(struct dx_entry);
 }
 
-static inline unsigned dx_node_limit (struct inode *dir)
+static inline unsigned dx_node_limit(struct inode *dir)
 {
 	unsigned entry_space = dir->i_sb->s_blocksize - EXT4_DIR_REC_LEN(0);
 	return entry_space / sizeof(struct dx_entry);
@@ -254,12 +256,12 @@ static inline unsigned dx_node_limit (struct inode *dir)
  * Debug
  */
 #ifdef DX_DEBUG
-static void dx_show_index (char * label, struct dx_entry *entries)
+static void dx_show_index(char * label, struct dx_entry *entries)
 {
 	int i, n = dx_get_count (entries);
-	printk("%s index ", label);
+	printk(KERN_DEBUG "%s index ", label);
 	for (i = 0; i < n; i++) {
-		printk("%x->%lu ", i? dx_get_hash(entries + i) :
+		printk("%x->%lu ", i ? dx_get_hash(entries + i) :
 				0, (unsigned long)dx_get_block(entries + i));
 	}
 	printk("\n");
@@ -306,7 +308,7 @@ struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir,
 			     struct dx_entry *entries, int levels)
 {
 	unsigned blocksize = dir->i_sb->s_blocksize;
-	unsigned count = dx_get_count (entries), names = 0, space = 0, i;
+	unsigned count = dx_get_count(entries), names = 0, space = 0, i;
 	unsigned bcount = 0;
 	struct buffer_head *bh;
 	int err;
@@ -325,11 +327,12 @@ struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir,
 		names += stats.names;
 		space += stats.space;
 		bcount += stats.bcount;
-		brelse (bh);
+		brelse(bh);
 	}
 	if (bcount)
-		printk("%snames %u, fullness %u (%u%%)\n", levels?"":"   ",
-			names, space/bcount,(space/bcount)*100/blocksize);
+		printk(KERN_DEBUG "%snames %u, fullness %u (%u%%)\n", 
+		       levels ? "" : "   ", names, space/bcount,
+		       (space/bcount)*100/blocksize);
 	return (struct stats) { names, space, bcount};
 }
 #endif /* DX_DEBUG */
@@ -344,7 +347,7 @@ struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir,
  * back to userspace.
  */
 static struct dx_frame *
-dx_probe(struct dentry *dentry, struct inode *dir,
+dx_probe(const struct qstr *d_name, struct inode *dir,
 	 struct dx_hash_info *hinfo, struct dx_frame *frame_in, int *err)
 {
 	unsigned count, indirect;
@@ -355,8 +358,6 @@ dx_probe(struct dentry *dentry, struct inode *dir,
 	u32 hash;
 
 	frame->bh = NULL;
-	if (dentry)
-		dir = dentry->d_parent->d_inode;
 	if (!(bh = ext4_bread (NULL,dir, 0, 0, err)))
 		goto fail;
 	root = (struct dx_root *) bh->b_data;
@@ -372,8 +373,8 @@ dx_probe(struct dentry *dentry, struct inode *dir,
 	}
 	hinfo->hash_version = root->info.hash_version;
 	hinfo->seed = EXT4_SB(dir->i_sb)->s_hash_seed;
-	if (dentry)
-		ext4fs_dirhash(dentry->d_name.name, dentry->d_name.len, hinfo);
+	if (d_name)
+		ext4fs_dirhash(d_name->name, d_name->len, hinfo);
 	hash = hinfo->hash;
 
 	if (root->info.unused_flags & 1) {
@@ -406,7 +407,7 @@ dx_probe(struct dentry *dentry, struct inode *dir,
 		goto fail;
 	}
 
-	dxtrace (printk("Look up %x", hash));
+	dxtrace(printk("Look up %x", hash));
 	while (1)
 	{
 		count = dx_get_count(entries);
@@ -555,7 +556,7 @@ static int ext4_htree_next_block(struct inode *dir, __u32 hash,
 				      0, &err)))
 			return err; /* Failure */
 		p++;
-		brelse (p->bh);
+		brelse(p->bh);
 		p->bh = bh;
 		p->at = p->entries = ((struct dx_node *) bh->b_data)->entries;
 	}
@@ -593,7 +594,7 @@ static int htree_dirblock_to_tree(struct file *dir_file,
 			/* On error, skip the f_pos to the next block. */
 			dir_file->f_pos = (dir_file->f_pos |
 					(dir->i_sb->s_blocksize - 1)) + 1;
-			brelse (bh);
+			brelse(bh);
 			return count;
 		}
 		ext4fs_dirhash(de->name, de->name_len, hinfo);
@@ -635,8 +636,8 @@ int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash,
 	int ret, err;
 	__u32 hashval;
 
-	dxtrace(printk("In htree_fill_tree, start hash: %x:%x\n", start_hash,
-		       start_minor_hash));
+	dxtrace(printk(KERN_DEBUG "In htree_fill_tree, start hash: %x:%x\n", 
+		       start_hash, start_minor_hash));
 	dir = dir_file->f_path.dentry->d_inode;
 	if (!(EXT4_I(dir)->i_flags & EXT4_INDEX_FL)) {
 		hinfo.hash_version = EXT4_SB(dir->i_sb)->s_def_hash_version;
@@ -648,7 +649,7 @@ int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash,
 	}
 	hinfo.hash = start_hash;
 	hinfo.minor_hash = 0;
-	frame = dx_probe(NULL, dir_file->f_path.dentry->d_inode, &hinfo, frames, &err);
+	frame = dx_probe(NULL, dir, &hinfo, frames, &err);
 	if (!frame)
 		return err;
 
@@ -694,8 +695,8 @@ int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash,
 			break;
 	}
 	dx_release(frames);
-	dxtrace(printk("Fill tree: returned %d entries, next hash: %x\n",
-		       count, *next_hash));
+	dxtrace(printk(KERN_DEBUG "Fill tree: returned %d entries, "
+		       "next hash: %x\n", count, *next_hash));
 	return count;
 errout:
 	dx_release(frames);
@@ -802,17 +803,17 @@ static inline int ext4_match (int len, const char * const name,
 /*
  * Returns 0 if not found, -1 on failure, and 1 on success
  */
-static inline int search_dirblock(struct buffer_head * bh,
+static inline int search_dirblock(struct buffer_head *bh,
 				  struct inode *dir,
-				  struct dentry *dentry,
+				  const struct qstr *d_name,
 				  unsigned long offset,
 				  struct ext4_dir_entry_2 ** res_dir)
 {
 	struct ext4_dir_entry_2 * de;
 	char * dlimit;
 	int de_len;
-	const char *name = dentry->d_name.name;
-	int namelen = dentry->d_name.len;
+	const char *name = d_name->name;
+	int namelen = d_name->len;
 
 	de = (struct ext4_dir_entry_2 *) bh->b_data;
 	dlimit = bh->b_data + dir->i_sb->s_blocksize;
@@ -851,12 +852,13 @@ static inline int search_dirblock(struct buffer_head * bh,
  * The returned buffer_head has ->b_count elevated.  The caller is expected
  * to brelse() it when appropriate.
  */
-static struct buffer_head * ext4_find_entry (struct dentry *dentry,
+static struct buffer_head * ext4_find_entry (struct inode *dir,
+					const struct qstr *d_name,
 					struct ext4_dir_entry_2 ** res_dir)
 {
-	struct super_block * sb;
-	struct buffer_head * bh_use[NAMEI_RA_SIZE];
-	struct buffer_head * bh, *ret = NULL;
+	struct super_block *sb;
+	struct buffer_head *bh_use[NAMEI_RA_SIZE];
+	struct buffer_head *bh, *ret = NULL;
 	ext4_lblk_t start, block, b;
 	int ra_max = 0;		/* Number of bh's in the readahead
 				   buffer, bh_use[] */
@@ -865,16 +867,15 @@ static struct buffer_head * ext4_find_entry (struct dentry *dentry,
 	int num = 0;
 	ext4_lblk_t  nblocks;
 	int i, err;
-	struct inode *dir = dentry->d_parent->d_inode;
 	int namelen;
 
 	*res_dir = NULL;
 	sb = dir->i_sb;
-	namelen = dentry->d_name.len;
+	namelen = d_name->len;
 	if (namelen > EXT4_NAME_LEN)
 		return NULL;
 	if (is_dx(dir)) {
-		bh = ext4_dx_find_entry(dentry, res_dir, &err);
+		bh = ext4_dx_find_entry(dir, d_name, res_dir, &err);
 		/*
 		 * On success, or if the error was file not found,
 		 * return.  Otherwise, fall back to doing a search the
@@ -882,7 +883,8 @@ static struct buffer_head * ext4_find_entry (struct dentry *dentry,
 		 */
 		if (bh || (err != ERR_BAD_DX_DIR))
 			return bh;
-		dxtrace(printk("ext4_find_entry: dx failed, falling back\n"));
+		dxtrace(printk(KERN_DEBUG "ext4_find_entry: dx failed, "
+			       "falling back\n"));
 	}
 	nblocks = dir->i_size >> EXT4_BLOCK_SIZE_BITS(sb);
 	start = EXT4_I(dir)->i_dir_start_lookup;
@@ -926,7 +928,7 @@ restart:
 			brelse(bh);
 			goto next;
 		}
-		i = search_dirblock(bh, dir, dentry,
+		i = search_dirblock(bh, dir, d_name,
 			    block << EXT4_BLOCK_SIZE_BITS(sb), res_dir);
 		if (i == 1) {
 			EXT4_I(dir)->i_dir_start_lookup = block;
@@ -956,11 +958,11 @@ restart:
 cleanup_and_exit:
 	/* Clean up the read-ahead blocks */
 	for (; ra_ptr < ra_max; ra_ptr++)
-		brelse (bh_use[ra_ptr]);
+		brelse(bh_use[ra_ptr]);
 	return ret;
 }
 
-static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry,
+static struct buffer_head * ext4_dx_find_entry(struct inode *dir, const struct qstr *d_name,
 		       struct ext4_dir_entry_2 **res_dir, int *err)
 {
 	struct super_block * sb;
@@ -971,14 +973,13 @@ static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry,
 	struct buffer_head *bh;
 	ext4_lblk_t block;
 	int retval;
-	int namelen = dentry->d_name.len;
-	const u8 *name = dentry->d_name.name;
-	struct inode *dir = dentry->d_parent->d_inode;
+	int namelen = d_name->len;
+	const u8 *name = d_name->name;
 
 	sb = dir->i_sb;
 	/* NFS may look up ".." - look at dx_root directory block */
 	if (namelen > 2 || name[0] != '.'||(name[1] != '.' && name[1] != '\0')){
-		if (!(frame = dx_probe(dentry, NULL, &hinfo, frames, err)))
+		if (!(frame = dx_probe(d_name, dir, &hinfo, frames, err)))
 			return NULL;
 	} else {
 		frame = frames;
@@ -1010,7 +1011,7 @@ static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry,
 				return bh;
 			}
 		}
-		brelse (bh);
+		brelse(bh);
 		/* Check to see if we should continue to search */
 		retval = ext4_htree_next_block(dir, hash, frame,
 					       frames, NULL);
@@ -1025,25 +1026,25 @@ static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry,
 
 	*err = -ENOENT;
 errout:
-	dxtrace(printk("%s not found\n", name));
+	dxtrace(printk(KERN_DEBUG "%s not found\n", name));
 	dx_release (frames);
 	return NULL;
 }
 
-static struct dentry *ext4_lookup(struct inode * dir, struct dentry *dentry, struct nameidata *nd)
+static struct dentry *ext4_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 {
-	struct inode * inode;
-	struct ext4_dir_entry_2 * de;
-	struct buffer_head * bh;
+	struct inode *inode;
+	struct ext4_dir_entry_2 *de;
+	struct buffer_head *bh;
 
 	if (dentry->d_name.len > EXT4_NAME_LEN)
 		return ERR_PTR(-ENAMETOOLONG);
 
-	bh = ext4_find_entry(dentry, &de);
+	bh = ext4_find_entry(dir, &dentry->d_name, &de);
 	inode = NULL;
 	if (bh) {
 		unsigned long ino = le32_to_cpu(de->inode);
-		brelse (bh);
+		brelse(bh);
 		if (!ext4_valid_inum(dir->i_sb, ino)) {
 			ext4_error(dir->i_sb, "ext4_lookup",
 				   "bad inode number: %lu", ino);
@@ -1062,15 +1063,14 @@ struct dentry *ext4_get_parent(struct dentry *child)
 	unsigned long ino;
 	struct dentry *parent;
 	struct inode *inode;
-	struct dentry dotdot;
+	static const struct qstr dotdot = {
+		.name = "..",
+		.len = 2,
+	};
 	struct ext4_dir_entry_2 * de;
 	struct buffer_head *bh;
 
-	dotdot.d_name.name = "..";
-	dotdot.d_name.len = 2;
-	dotdot.d_parent = child; /* confusing, isn't it! */
-
-	bh = ext4_find_entry(&dotdot, &de);
+	bh = ext4_find_entry(child->d_inode, &dotdot, &de);
 	inode = NULL;
 	if (!bh)
 		return ERR_PTR(-ENOENT);
@@ -1201,10 +1201,10 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir,
 
 	/* create map in the end of data2 block */
 	map = (struct dx_map_entry *) (data2 + blocksize);
-	count = dx_make_map ((struct ext4_dir_entry_2 *) data1,
+	count = dx_make_map((struct ext4_dir_entry_2 *) data1,
 			     blocksize, hinfo, map);
 	map -= count;
-	dx_sort_map (map, count);
+	dx_sort_map(map, count);
 	/* Split the existing block in the middle, size-wise */
 	size = 0;
 	move = 0;
@@ -1225,7 +1225,7 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir,
 
 	/* Fancy dance to stay within two buffers */
 	de2 = dx_move_dirents(data1, data2, map + split, count - split);
-	de = dx_pack_dirents(data1,blocksize);
+	de = dx_pack_dirents(data1, blocksize);
 	de->rec_len = ext4_rec_len_to_disk(data1 + blocksize - (char *) de);
 	de2->rec_len = ext4_rec_len_to_disk(data2 + blocksize - (char *) de2);
 	dxtrace(dx_show_leaf (hinfo, (struct ext4_dir_entry_2 *) data1, blocksize, 1));
@@ -1237,15 +1237,15 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir,
 		swap(*bh, bh2);
 		de = de2;
 	}
-	dx_insert_block (frame, hash2 + continued, newblock);
-	err = ext4_journal_dirty_metadata (handle, bh2);
+	dx_insert_block(frame, hash2 + continued, newblock);
+	err = ext4_journal_dirty_metadata(handle, bh2);
 	if (err)
 		goto journal_error;
-	err = ext4_journal_dirty_metadata (handle, frame->bh);
+	err = ext4_journal_dirty_metadata(handle, frame->bh);
 	if (err)
 		goto journal_error;
-	brelse (bh2);
-	dxtrace(dx_show_index ("frame", frame->entries));
+	brelse(bh2);
+	dxtrace(dx_show_index("frame", frame->entries));
 	return de;
 
 journal_error:
@@ -1271,7 +1271,7 @@ errout:
  */
 static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry,
 			     struct inode *inode, struct ext4_dir_entry_2 *de,
-			     struct buffer_head * bh)
+			     struct buffer_head *bh)
 {
 	struct inode	*dir = dentry->d_parent->d_inode;
 	const char	*name = dentry->d_name.name;
@@ -1288,11 +1288,11 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry,
 		while ((char *) de <= top) {
 			if (!ext4_check_dir_entry("ext4_add_entry", dir, de,
 						  bh, offset)) {
-				brelse (bh);
+				brelse(bh);
 				return -EIO;
 			}
-			if (ext4_match (namelen, name, de)) {
-				brelse (bh);
+			if (ext4_match(namelen, name, de)) {
+				brelse(bh);
 				return -EEXIST;
 			}
 			nlen = EXT4_DIR_REC_LEN(de->name_len);
@@ -1329,7 +1329,7 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry,
 	} else
 		de->inode = 0;
 	de->name_len = namelen;
-	memcpy (de->name, name, namelen);
+	memcpy(de->name, name, namelen);
 	/*
 	 * XXX shouldn't update any times until successful
 	 * completion of syscall, but too many callers depend
@@ -1377,7 +1377,7 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry,
 	struct fake_dirent *fde;
 
 	blocksize =  dir->i_sb->s_blocksize;
-	dxtrace(printk("Creating index\n"));
+	dxtrace(printk(KERN_DEBUG "Creating index\n"));
 	retval = ext4_journal_get_write_access(handle, bh);
 	if (retval) {
 		ext4_std_error(dir->i_sb, retval);
@@ -1386,7 +1386,7 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry,
 	}
 	root = (struct dx_root *) bh->b_data;
 
-	bh2 = ext4_append (handle, dir, &block, &retval);
+	bh2 = ext4_append(handle, dir, &block, &retval);
 	if (!(bh2)) {
 		brelse(bh);
 		return retval;
@@ -1412,9 +1412,9 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry,
 	root->info.info_length = sizeof(root->info);
 	root->info.hash_version = EXT4_SB(dir->i_sb)->s_def_hash_version;
 	entries = root->entries;
-	dx_set_block (entries, 1);
-	dx_set_count (entries, 1);
-	dx_set_limit (entries, dx_root_limit(dir, sizeof(root->info)));
+	dx_set_block(entries, 1);
+	dx_set_count(entries, 1);
+	dx_set_limit(entries, dx_root_limit(dir, sizeof(root->info)));
 
 	/* Initialize as for dx_probe */
 	hinfo.hash_version = root->info.hash_version;
@@ -1443,14 +1443,14 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry,
  * may not sleep between calling this and putting something into
  * the entry, as someone else might have used it while you slept.
  */
-static int ext4_add_entry (handle_t *handle, struct dentry *dentry,
-	struct inode *inode)
+static int ext4_add_entry(handle_t *handle, struct dentry *dentry,
+			  struct inode *inode)
 {
 	struct inode *dir = dentry->d_parent->d_inode;
 	unsigned long offset;
-	struct buffer_head * bh;
+	struct buffer_head *bh;
 	struct ext4_dir_entry_2 *de;
-	struct super_block * sb;
+	struct super_block *sb;
 	int	retval;
 	int	dx_fallback=0;
 	unsigned blocksize;
@@ -1500,13 +1500,13 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
 	struct dx_frame frames[2], *frame;
 	struct dx_entry *entries, *at;
 	struct dx_hash_info hinfo;
-	struct buffer_head * bh;
+	struct buffer_head *bh;
 	struct inode *dir = dentry->d_parent->d_inode;
-	struct super_block * sb = dir->i_sb;
+	struct super_block *sb = dir->i_sb;
 	struct ext4_dir_entry_2 *de;
 	int err;
 
-	frame = dx_probe(dentry, NULL, &hinfo, frames, &err);
+	frame = dx_probe(&dentry->d_name, dir, &hinfo, frames, &err);
 	if (!frame)
 		return err;
 	entries = frame->entries;
@@ -1527,7 +1527,7 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
 	}
 
 	/* Block full, should compress but for now just split */
-	dxtrace(printk("using %u of %u node entries\n",
+	dxtrace(printk(KERN_DEBUG "using %u of %u node entries\n",
 		       dx_get_count(entries), dx_get_limit(entries)));
 	/* Need to split index? */
 	if (dx_get_count(entries) == dx_get_limit(entries)) {
@@ -1559,7 +1559,8 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
 		if (levels) {
 			unsigned icount1 = icount/2, icount2 = icount - icount1;
 			unsigned hash2 = dx_get_hash(entries + icount1);
-			dxtrace(printk("Split index %i/%i\n", icount1, icount2));
+			dxtrace(printk(KERN_DEBUG "Split index %i/%i\n",
+				       icount1, icount2));
 
 			BUFFER_TRACE(frame->bh, "get_write_access"); /* index root */
 			err = ext4_journal_get_write_access(handle,
@@ -1567,11 +1568,11 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
 			if (err)
 				goto journal_error;
 
-			memcpy ((char *) entries2, (char *) (entries + icount1),
-				icount2 * sizeof(struct dx_entry));
-			dx_set_count (entries, icount1);
-			dx_set_count (entries2, icount2);
-			dx_set_limit (entries2, dx_node_limit(dir));
+			memcpy((char *) entries2, (char *) (entries + icount1),
+			       icount2 * sizeof(struct dx_entry));
+			dx_set_count(entries, icount1);
+			dx_set_count(entries2, icount2);
+			dx_set_limit(entries2, dx_node_limit(dir));
 
 			/* Which index block gets the new entry? */
 			if (at - entries >= icount1) {
@@ -1579,16 +1580,17 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
 				frame->entries = entries = entries2;
 				swap(frame->bh, bh2);
 			}
-			dx_insert_block (frames + 0, hash2, newblock);
-			dxtrace(dx_show_index ("node", frames[1].entries));
-			dxtrace(dx_show_index ("node",
+			dx_insert_block(frames + 0, hash2, newblock);
+			dxtrace(dx_show_index("node", frames[1].entries));
+			dxtrace(dx_show_index("node",
 			       ((struct dx_node *) bh2->b_data)->entries));
 			err = ext4_journal_dirty_metadata(handle, bh2);
 			if (err)
 				goto journal_error;
 			brelse (bh2);
 		} else {
-			dxtrace(printk("Creating second level index...\n"));
+			dxtrace(printk(KERN_DEBUG
+				       "Creating second level index...\n"));
 			memcpy((char *) entries2, (char *) entries,
 			       icount * sizeof(struct dx_entry));
 			dx_set_limit(entries2, dx_node_limit(dir));
@@ -1630,12 +1632,12 @@ cleanup:
  * ext4_delete_entry deletes a directory entry by merging it with the
  * previous entry
  */
-static int ext4_delete_entry (handle_t *handle,
-			      struct inode * dir,
-			      struct ext4_dir_entry_2 * de_del,
-			      struct buffer_head * bh)
+static int ext4_delete_entry(handle_t *handle,
+			     struct inode *dir,
+			     struct ext4_dir_entry_2 *de_del,
+			     struct buffer_head *bh)
 {
-	struct ext4_dir_entry_2 * de, * pde;
+	struct ext4_dir_entry_2 *de, *pde;
 	int i;
 
 	i = 0;
@@ -1716,11 +1718,11 @@ static int ext4_add_nondir(handle_t *handle,
  * If the create succeeds, we fill in the inode information
  * with d_instantiate().
  */
-static int ext4_create (struct inode * dir, struct dentry * dentry, int mode,
-		struct nameidata *nd)
+static int ext4_create(struct inode *dir, struct dentry *dentry, int mode,
+		       struct nameidata *nd)
 {
 	handle_t *handle;
-	struct inode * inode;
+	struct inode *inode;
 	int err, retries = 0;
 
 retry:
@@ -1747,8 +1749,8 @@ retry:
 	return err;
 }
 
-static int ext4_mknod (struct inode * dir, struct dentry *dentry,
-			int mode, dev_t rdev)
+static int ext4_mknod(struct inode *dir, struct dentry *dentry,
+		      int mode, dev_t rdev)
 {
 	handle_t *handle;
 	struct inode *inode;
@@ -1767,11 +1769,11 @@ retry:
 	if (IS_DIRSYNC(dir))
 		handle->h_sync = 1;
 
-	inode = ext4_new_inode (handle, dir, mode);
+	inode = ext4_new_inode(handle, dir, mode);
 	err = PTR_ERR(inode);
 	if (!IS_ERR(inode)) {
 		init_special_inode(inode, inode->i_mode, rdev);
-#ifdef CONFIG_EXT4DEV_FS_XATTR
+#ifdef CONFIG_EXT4_FS_XATTR
 		inode->i_op = &ext4_special_inode_operations;
 #endif
 		err = ext4_add_nondir(handle, dentry, inode);
@@ -1782,12 +1784,12 @@ retry:
 	return err;
 }
 
-static int ext4_mkdir(struct inode * dir, struct dentry * dentry, int mode)
+static int ext4_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 {
 	handle_t *handle;
-	struct inode * inode;
-	struct buffer_head * dir_block;
-	struct ext4_dir_entry_2 * de;
+	struct inode *inode;
+	struct buffer_head *dir_block;
+	struct ext4_dir_entry_2 *de;
 	int err, retries = 0;
 
 	if (EXT4_DIR_LINK_MAX(dir))
@@ -1803,7 +1805,7 @@ retry:
 	if (IS_DIRSYNC(dir))
 		handle->h_sync = 1;
 
-	inode = ext4_new_inode (handle, dir, S_IFDIR | mode);
+	inode = ext4_new_inode(handle, dir, S_IFDIR | mode);
 	err = PTR_ERR(inode);
 	if (IS_ERR(inode))
 		goto out_stop;
@@ -1811,7 +1813,7 @@ retry:
 	inode->i_op = &ext4_dir_inode_operations;
 	inode->i_fop = &ext4_dir_operations;
 	inode->i_size = EXT4_I(inode)->i_disksize = inode->i_sb->s_blocksize;
-	dir_block = ext4_bread (handle, inode, 0, 1, &err);
+	dir_block = ext4_bread(handle, inode, 0, 1, &err);
 	if (!dir_block)
 		goto out_clear_inode;
 	BUFFER_TRACE(dir_block, "get_write_access");
@@ -1820,26 +1822,26 @@ retry:
 	de->inode = cpu_to_le32(inode->i_ino);
 	de->name_len = 1;
 	de->rec_len = ext4_rec_len_to_disk(EXT4_DIR_REC_LEN(de->name_len));
-	strcpy (de->name, ".");
+	strcpy(de->name, ".");
 	ext4_set_de_type(dir->i_sb, de, S_IFDIR);
 	de = ext4_next_entry(de);
 	de->inode = cpu_to_le32(dir->i_ino);
 	de->rec_len = ext4_rec_len_to_disk(inode->i_sb->s_blocksize -
 						EXT4_DIR_REC_LEN(1));
 	de->name_len = 2;
-	strcpy (de->name, "..");
+	strcpy(de->name, "..");
 	ext4_set_de_type(dir->i_sb, de, S_IFDIR);
 	inode->i_nlink = 2;
 	BUFFER_TRACE(dir_block, "call ext4_journal_dirty_metadata");
 	ext4_journal_dirty_metadata(handle, dir_block);
-	brelse (dir_block);
+	brelse(dir_block);
 	ext4_mark_inode_dirty(handle, inode);
-	err = ext4_add_entry (handle, dentry, inode);
+	err = ext4_add_entry(handle, dentry, inode);
 	if (err) {
 out_clear_inode:
 		clear_nlink(inode);
 		ext4_mark_inode_dirty(handle, inode);
-		iput (inode);
+		iput(inode);
 		goto out_stop;
 	}
 	ext4_inc_count(handle, dir);
@@ -1856,17 +1858,17 @@ out_stop:
 /*
  * routine to check that the specified directory is empty (for rmdir)
  */
-static int empty_dir (struct inode * inode)
+static int empty_dir(struct inode *inode)
 {
 	unsigned long offset;
-	struct buffer_head * bh;
-	struct ext4_dir_entry_2 * de, * de1;
-	struct super_block * sb;
+	struct buffer_head *bh;
+	struct ext4_dir_entry_2 *de, *de1;
+	struct super_block *sb;
 	int err = 0;
 
 	sb = inode->i_sb;
 	if (inode->i_size < EXT4_DIR_REC_LEN(1) + EXT4_DIR_REC_LEN(2) ||
-	    !(bh = ext4_bread (NULL, inode, 0, 0, &err))) {
+	    !(bh = ext4_bread(NULL, inode, 0, 0, &err))) {
 		if (err)
 			ext4_error(inode->i_sb, __func__,
 				   "error %d reading directory #%lu offset 0",
@@ -1881,23 +1883,23 @@ static int empty_dir (struct inode * inode)
 	de1 = ext4_next_entry(de);
 	if (le32_to_cpu(de->inode) != inode->i_ino ||
 			!le32_to_cpu(de1->inode) ||
-			strcmp (".", de->name) ||
-			strcmp ("..", de1->name)) {
-		ext4_warning (inode->i_sb, "empty_dir",
-			      "bad directory (dir #%lu) - no `.' or `..'",
-			      inode->i_ino);
-		brelse (bh);
+			strcmp(".", de->name) ||
+			strcmp("..", de1->name)) {
+		ext4_warning(inode->i_sb, "empty_dir",
+			     "bad directory (dir #%lu) - no `.' or `..'",
+			     inode->i_ino);
+		brelse(bh);
 		return 1;
 	}
 	offset = ext4_rec_len_from_disk(de->rec_len) +
 		 ext4_rec_len_from_disk(de1->rec_len);
 	de = ext4_next_entry(de1);
-	while (offset < inode->i_size ) {
+	while (offset < inode->i_size) {
 		if (!bh ||
 			(void *) de >= (void *) (bh->b_data+sb->s_blocksize)) {
 			err = 0;
-			brelse (bh);
-			bh = ext4_bread (NULL, inode,
+			brelse(bh);
+			bh = ext4_bread(NULL, inode,
 				offset >> EXT4_BLOCK_SIZE_BITS(sb), 0, &err);
 			if (!bh) {
 				if (err)
@@ -1917,13 +1919,13 @@ static int empty_dir (struct inode * inode)
 			continue;
 		}
 		if (le32_to_cpu(de->inode)) {
-			brelse (bh);
+			brelse(bh);
 			return 0;
 		}
 		offset += ext4_rec_len_from_disk(de->rec_len);
 		de = ext4_next_entry(de);
 	}
-	brelse (bh);
+	brelse(bh);
 	return 1;
 }
 
@@ -1954,8 +1956,8 @@ int ext4_orphan_add(handle_t *handle, struct inode *inode)
 	 * ->i_nlink. For, say it, character device. Not a regular file,
 	 * not a directory, not a symlink and ->i_nlink > 0.
 	 */
-	J_ASSERT ((S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) ||
-		S_ISLNK(inode->i_mode)) || inode->i_nlink == 0);
+	J_ASSERT((S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) ||
+		  S_ISLNK(inode->i_mode)) || inode->i_nlink == 0);
 
 	BUFFER_TRACE(EXT4_SB(sb)->s_sbh, "get_write_access");
 	err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh);
@@ -2069,12 +2071,12 @@ out_brelse:
 	goto out_err;
 }
 
-static int ext4_rmdir (struct inode * dir, struct dentry *dentry)
+static int ext4_rmdir(struct inode *dir, struct dentry *dentry)
 {
 	int retval;
-	struct inode * inode;
-	struct buffer_head * bh;
-	struct ext4_dir_entry_2 * de;
+	struct inode *inode;
+	struct buffer_head *bh;
+	struct ext4_dir_entry_2 *de;
 	handle_t *handle;
 
 	/* Initialize quotas before so that eventual writes go in
@@ -2085,7 +2087,7 @@ static int ext4_rmdir (struct inode * dir, struct dentry *dentry)
 		return PTR_ERR(handle);
 
 	retval = -ENOENT;
-	bh = ext4_find_entry (dentry, &de);
+	bh = ext4_find_entry(dir, &dentry->d_name, &de);
 	if (!bh)
 		goto end_rmdir;
 
@@ -2099,16 +2101,16 @@ static int ext4_rmdir (struct inode * dir, struct dentry *dentry)
 		goto end_rmdir;
 
 	retval = -ENOTEMPTY;
-	if (!empty_dir (inode))
+	if (!empty_dir(inode))
 		goto end_rmdir;
 
 	retval = ext4_delete_entry(handle, dir, de, bh);
 	if (retval)
 		goto end_rmdir;
 	if (!EXT4_DIR_LINK_EMPTY(inode))
-		ext4_warning (inode->i_sb, "ext4_rmdir",
-			      "empty directory has too many links (%d)",
-			      inode->i_nlink);
+		ext4_warning(inode->i_sb, "ext4_rmdir",
+			     "empty directory has too many links (%d)",
+			     inode->i_nlink);
 	inode->i_version++;
 	clear_nlink(inode);
 	/* There's no need to set i_disksize: the fact that i_nlink is
@@ -2124,16 +2126,16 @@ static int ext4_rmdir (struct inode * dir, struct dentry *dentry)
 
 end_rmdir:
 	ext4_journal_stop(handle);
-	brelse (bh);
+	brelse(bh);
 	return retval;
 }
 
-static int ext4_unlink(struct inode * dir, struct dentry *dentry)
+static int ext4_unlink(struct inode *dir, struct dentry *dentry)
 {
 	int retval;
-	struct inode * inode;
-	struct buffer_head * bh;
-	struct ext4_dir_entry_2 * de;
+	struct inode *inode;
+	struct buffer_head *bh;
+	struct ext4_dir_entry_2 *de;
 	handle_t *handle;
 
 	/* Initialize quotas before so that eventual writes go
@@ -2147,7 +2149,7 @@ static int ext4_unlink(struct inode * dir, struct dentry *dentry)
 		handle->h_sync = 1;
 
 	retval = -ENOENT;
-	bh = ext4_find_entry (dentry, &de);
+	bh = ext4_find_entry(dir, &dentry->d_name, &de);
 	if (!bh)
 		goto end_unlink;
 
@@ -2158,9 +2160,9 @@ static int ext4_unlink(struct inode * dir, struct dentry *dentry)
 		goto end_unlink;
 
 	if (!inode->i_nlink) {
-		ext4_warning (inode->i_sb, "ext4_unlink",
-			      "Deleting nonexistent file (%lu), %d",
-			      inode->i_ino, inode->i_nlink);
+		ext4_warning(inode->i_sb, "ext4_unlink",
+			     "Deleting nonexistent file (%lu), %d",
+			     inode->i_ino, inode->i_nlink);
 		inode->i_nlink = 1;
 	}
 	retval = ext4_delete_entry(handle, dir, de, bh);
@@ -2178,15 +2180,15 @@ static int ext4_unlink(struct inode * dir, struct dentry *dentry)
 
 end_unlink:
 	ext4_journal_stop(handle);
-	brelse (bh);
+	brelse(bh);
 	return retval;
 }
 
-static int ext4_symlink (struct inode * dir,
-		struct dentry *dentry, const char * symname)
+static int ext4_symlink(struct inode *dir,
+			struct dentry *dentry, const char *symname)
 {
 	handle_t *handle;
-	struct inode * inode;
+	struct inode *inode;
 	int l, err, retries = 0;
 
 	l = strlen(symname)+1;
@@ -2203,12 +2205,12 @@ retry:
 	if (IS_DIRSYNC(dir))
 		handle->h_sync = 1;
 
-	inode = ext4_new_inode (handle, dir, S_IFLNK|S_IRWXUGO);
+	inode = ext4_new_inode(handle, dir, S_IFLNK|S_IRWXUGO);
 	err = PTR_ERR(inode);
 	if (IS_ERR(inode))
 		goto out_stop;
 
-	if (l > sizeof (EXT4_I(inode)->i_data)) {
+	if (l > sizeof(EXT4_I(inode)->i_data)) {
 		inode->i_op = &ext4_symlink_inode_operations;
 		ext4_set_aops(inode);
 		/*
@@ -2221,14 +2223,14 @@ retry:
 		if (err) {
 			clear_nlink(inode);
 			ext4_mark_inode_dirty(handle, inode);
-			iput (inode);
+			iput(inode);
 			goto out_stop;
 		}
 	} else {
 		/* clear the extent format for fast symlink */
 		EXT4_I(inode)->i_flags &= ~EXT4_EXTENTS_FL;
 		inode->i_op = &ext4_fast_symlink_inode_operations;
-		memcpy((char*)&EXT4_I(inode)->i_data,symname,l);
+		memcpy((char *)&EXT4_I(inode)->i_data, symname, l);
 		inode->i_size = l-1;
 	}
 	EXT4_I(inode)->i_disksize = inode->i_size;
@@ -2240,8 +2242,8 @@ out_stop:
 	return err;
 }
 
-static int ext4_link (struct dentry * old_dentry,
-		struct inode * dir, struct dentry *dentry)
+static int ext4_link(struct dentry *old_dentry,
+		     struct inode *dir, struct dentry *dentry)
 {
 	handle_t *handle;
 	struct inode *inode = old_dentry->d_inode;
@@ -2284,13 +2286,13 @@ retry:
  * Anybody can rename anything with this: the permission checks are left to the
  * higher-level routines.
  */
-static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry,
-			   struct inode * new_dir,struct dentry *new_dentry)
+static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry,
+		       struct inode *new_dir, struct dentry *new_dentry)
 {
 	handle_t *handle;
-	struct inode * old_inode, * new_inode;
-	struct buffer_head * old_bh, * new_bh, * dir_bh;
-	struct ext4_dir_entry_2 * old_de, * new_de;
+	struct inode *old_inode, *new_inode;
+	struct buffer_head *old_bh, *new_bh, *dir_bh;
+	struct ext4_dir_entry_2 *old_de, *new_de;
 	int retval;
 
 	old_bh = new_bh = dir_bh = NULL;
@@ -2308,7 +2310,7 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry,
 	if (IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir))
 		handle->h_sync = 1;
 
-	old_bh = ext4_find_entry (old_dentry, &old_de);
+	old_bh = ext4_find_entry(old_dir, &old_dentry->d_name, &old_de);
 	/*
 	 *  Check for inode number is _not_ due to possible IO errors.
 	 *  We might rmdir the source, keep it as pwd of some process
@@ -2321,32 +2323,32 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry,
 		goto end_rename;
 
 	new_inode = new_dentry->d_inode;
-	new_bh = ext4_find_entry (new_dentry, &new_de);
+	new_bh = ext4_find_entry(new_dir, &new_dentry->d_name, &new_de);
 	if (new_bh) {
 		if (!new_inode) {
-			brelse (new_bh);
+			brelse(new_bh);
 			new_bh = NULL;
 		}
 	}
 	if (S_ISDIR(old_inode->i_mode)) {
 		if (new_inode) {
 			retval = -ENOTEMPTY;
-			if (!empty_dir (new_inode))
+			if (!empty_dir(new_inode))
 				goto end_rename;
 		}
 		retval = -EIO;
-		dir_bh = ext4_bread (handle, old_inode, 0, 0, &retval);
+		dir_bh = ext4_bread(handle, old_inode, 0, 0, &retval);
 		if (!dir_bh)
 			goto end_rename;
 		if (le32_to_cpu(PARENT_INO(dir_bh->b_data)) != old_dir->i_ino)
 			goto end_rename;
 		retval = -EMLINK;
-		if (!new_inode && new_dir!=old_dir &&
+		if (!new_inode && new_dir != old_dir &&
 				new_dir->i_nlink >= EXT4_LINK_MAX)
 			goto end_rename;
 	}
 	if (!new_bh) {
-		retval = ext4_add_entry (handle, new_dentry, old_inode);
+		retval = ext4_add_entry(handle, new_dentry, old_inode);
 		if (retval)
 			goto end_rename;
 	} else {
@@ -2388,7 +2390,7 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry,
 		struct buffer_head *old_bh2;
 		struct ext4_dir_entry_2 *old_de2;
 
-		old_bh2 = ext4_find_entry(old_dentry, &old_de2);
+		old_bh2 = ext4_find_entry(old_dir, &old_dentry->d_name, &old_de2);
 		if (old_bh2) {
 			retval = ext4_delete_entry(handle, old_dir,
 						   old_de2, old_bh2);
@@ -2433,9 +2435,9 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry,
 	retval = 0;
 
 end_rename:
-	brelse (dir_bh);
-	brelse (old_bh);
-	brelse (new_bh);
+	brelse(dir_bh);
+	brelse(old_bh);
+	brelse(new_bh);
 	ext4_journal_stop(handle);
 	return retval;
 }
@@ -2454,7 +2456,7 @@ const struct inode_operations ext4_dir_inode_operations = {
 	.mknod		= ext4_mknod,
 	.rename		= ext4_rename,
 	.setattr	= ext4_setattr,
-#ifdef CONFIG_EXT4DEV_FS_XATTR
+#ifdef CONFIG_EXT4_FS_XATTR
 	.setxattr	= generic_setxattr,
 	.getxattr	= generic_getxattr,
 	.listxattr	= ext4_listxattr,
@@ -2465,7 +2467,7 @@ const struct inode_operations ext4_dir_inode_operations = {
 
 const struct inode_operations ext4_special_inode_operations = {
 	.setattr	= ext4_setattr,
-#ifdef CONFIG_EXT4DEV_FS_XATTR
+#ifdef CONFIG_EXT4_FS_XATTR
 	.setxattr	= generic_setxattr,
 	.getxattr	= generic_getxattr,
 	.listxattr	= ext4_listxattr,
diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c
index b3d35604ea18..b6ec1843a015 100644
--- a/fs/ext4/resize.c
+++ b/fs/ext4/resize.c
@@ -416,8 +416,8 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 		       "EXT4-fs: ext4_add_new_gdb: adding group block %lu\n",
 		       gdb_num);
 
-        /*
-         * If we are not using the primary superblock/GDT copy don't resize,
+	/*
+	 * If we are not using the primary superblock/GDT copy don't resize,
          * because the user tools have no way of handling this.  Probably a
          * bad time to do it anyways.
          */
@@ -870,11 +870,10 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input)
 	 * We can allocate memory for mb_alloc based on the new group
 	 * descriptor
 	 */
-	if (test_opt(sb, MBALLOC)) {
-		err = ext4_mb_add_more_groupinfo(sb, input->group, gdp);
-		if (err)
-			goto exit_journal;
-	}
+	err = ext4_mb_add_more_groupinfo(sb, input->group, gdp);
+	if (err)
+		goto exit_journal;
+
 	/*
 	 * Make the new blocks and inodes valid next.  We do this before
 	 * increasing the group count so that once the group is enabled,
@@ -929,6 +928,15 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input)
 	percpu_counter_add(&sbi->s_freeinodes_counter,
 			   EXT4_INODES_PER_GROUP(sb));
 
+	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG)) {
+		ext4_group_t flex_group;
+		flex_group = ext4_flex_group(sbi, input->group);
+		sbi->s_flex_groups[flex_group].free_blocks +=
+			input->free_blocks_count;
+		sbi->s_flex_groups[flex_group].free_inodes +=
+			EXT4_INODES_PER_GROUP(sb);
+	}
+
 	ext4_journal_dirty_metadata(handle, sbi->s_sbh);
 	sb->s_dirt = 1;
 
@@ -964,7 +972,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
 	ext4_group_t o_groups_count;
 	ext4_grpblk_t last;
 	ext4_grpblk_t add;
-	struct buffer_head * bh;
+	struct buffer_head *bh;
 	handle_t *handle;
 	int err;
 	unsigned long freed_blocks;
@@ -1077,8 +1085,15 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
 	/*
 	 * Mark mballoc pages as not up to date so that they will be updated
 	 * next time they are loaded by ext4_mb_load_buddy.
+	 *
+	 * XXX Bad, Bad, BAD!!!  We should not be overloading the
+	 * Uptodate flag, particularly on thte bitmap bh, as way of
+	 * hinting to ext4_mb_load_buddy() that it needs to be
+	 * overloaded.  A user could take a LVM snapshot, then do an
+	 * on-line fsck, and clear the uptodate flag, and this would
+	 * not be a bug in userspace, but a bug in the kernel.  FIXME!!!
 	 */
-	if (test_opt(sb, MBALLOC)) {
+	{
 		struct ext4_sb_info *sbi = EXT4_SB(sb);
 		struct inode *inode = sbi->s_buddy_cache;
 		int blocks_per_page;
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 566344b926b7..fb940c22ab0d 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -34,6 +34,8 @@
 #include <linux/namei.h>
 #include <linux/quotaops.h>
 #include <linux/seq_file.h>
+#include <linux/proc_fs.h>
+#include <linux/marker.h>
 #include <linux/log2.h>
 #include <linux/crc16.h>
 #include <asm/uaccess.h>
@@ -45,6 +47,8 @@
 #include "namei.h"
 #include "group.h"
 
+struct proc_dir_entry *ext4_proc_root;
+
 static int ext4_load_journal(struct super_block *, struct ext4_super_block *,
 			     unsigned long journal_devnum);
 static int ext4_create_journal(struct super_block *, struct ext4_super_block *,
@@ -503,15 +507,18 @@ static void ext4_put_super(struct super_block *sb)
 	ext4_mb_release(sb);
 	ext4_ext_release(sb);
 	ext4_xattr_put_super(sb);
-	jbd2_journal_destroy(sbi->s_journal);
+	if (jbd2_journal_destroy(sbi->s_journal) < 0)
+		ext4_abort(sb, __func__, "Couldn't clean up the journal");
 	sbi->s_journal = NULL;
 	if (!(sb->s_flags & MS_RDONLY)) {
 		EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
 		es->s_state = cpu_to_le16(sbi->s_mount_state);
-		BUFFER_TRACE(sbi->s_sbh, "marking dirty");
-		mark_buffer_dirty(sbi->s_sbh);
 		ext4_commit_super(sb, es, 1);
 	}
+	if (sbi->s_proc) {
+		remove_proc_entry("inode_readahead_blks", sbi->s_proc);
+		remove_proc_entry(sb->s_id, ext4_proc_root);
+	}
 
 	for (i = 0; i < sbi->s_gdb_count; i++)
 		brelse(sbi->s_group_desc[i]);
@@ -520,6 +527,7 @@ static void ext4_put_super(struct super_block *sb)
 	percpu_counter_destroy(&sbi->s_freeblocks_counter);
 	percpu_counter_destroy(&sbi->s_freeinodes_counter);
 	percpu_counter_destroy(&sbi->s_dirs_counter);
+	percpu_counter_destroy(&sbi->s_dirtyblocks_counter);
 	brelse(sbi->s_sbh);
 #ifdef CONFIG_QUOTA
 	for (i = 0; i < MAXQUOTAS; i++)
@@ -562,11 +570,10 @@ static struct inode *ext4_alloc_inode(struct super_block *sb)
 	ei = kmem_cache_alloc(ext4_inode_cachep, GFP_NOFS);
 	if (!ei)
 		return NULL;
-#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
+#ifdef CONFIG_EXT4_FS_POSIX_ACL
 	ei->i_acl = EXT4_ACL_NOT_CACHED;
 	ei->i_default_acl = EXT4_ACL_NOT_CACHED;
 #endif
-	ei->i_block_alloc_info = NULL;
 	ei->vfs_inode.i_version = 1;
 	ei->vfs_inode.i_data.writeback_index = 0;
 	memset(&ei->i_cached_extent, 0, sizeof(struct ext4_ext_cache));
@@ -599,7 +606,7 @@ static void init_once(void *foo)
 	struct ext4_inode_info *ei = (struct ext4_inode_info *) foo;
 
 	INIT_LIST_HEAD(&ei->i_orphan);
-#ifdef CONFIG_EXT4DEV_FS_XATTR
+#ifdef CONFIG_EXT4_FS_XATTR
 	init_rwsem(&ei->xattr_sem);
 #endif
 	init_rwsem(&ei->i_data_sem);
@@ -625,8 +632,7 @@ static void destroy_inodecache(void)
 
 static void ext4_clear_inode(struct inode *inode)
 {
-	struct ext4_block_alloc_info *rsv = EXT4_I(inode)->i_block_alloc_info;
-#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
+#ifdef CONFIG_EXT4_FS_POSIX_ACL
 	if (EXT4_I(inode)->i_acl &&
 			EXT4_I(inode)->i_acl != EXT4_ACL_NOT_CACHED) {
 		posix_acl_release(EXT4_I(inode)->i_acl);
@@ -638,10 +644,7 @@ static void ext4_clear_inode(struct inode *inode)
 		EXT4_I(inode)->i_default_acl = EXT4_ACL_NOT_CACHED;
 	}
 #endif
-	ext4_discard_reservation(inode);
-	EXT4_I(inode)->i_block_alloc_info = NULL;
-	if (unlikely(rsv))
-		kfree(rsv);
+	ext4_discard_preallocations(inode);
 	jbd2_journal_release_jbd_inode(EXT4_SB(inode->i_sb)->s_journal,
 				       &EXT4_I(inode)->jinode);
 }
@@ -654,7 +657,7 @@ static inline void ext4_show_quota_options(struct seq_file *seq,
 
 	if (sbi->s_jquota_fmt)
 		seq_printf(seq, ",jqfmt=%s",
-		(sbi->s_jquota_fmt == QFMT_VFS_OLD) ? "vfsold": "vfsv0");
+		(sbi->s_jquota_fmt == QFMT_VFS_OLD) ? "vfsold" : "vfsv0");
 
 	if (sbi->s_qf_names[USRQUOTA])
 		seq_printf(seq, ",usrjquota=%s", sbi->s_qf_names[USRQUOTA]);
@@ -718,7 +721,7 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs)
 		seq_puts(seq, ",debug");
 	if (test_opt(sb, OLDALLOC))
 		seq_puts(seq, ",oldalloc");
-#ifdef CONFIG_EXT4DEV_FS_XATTR
+#ifdef CONFIG_EXT4_FS_XATTR
 	if (test_opt(sb, XATTR_USER) &&
 		!(def_mount_opts & EXT4_DEFM_XATTR_USER))
 		seq_puts(seq, ",user_xattr");
@@ -727,7 +730,7 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs)
 		seq_puts(seq, ",nouser_xattr");
 	}
 #endif
-#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
+#ifdef CONFIG_EXT4_FS_POSIX_ACL
 	if (test_opt(sb, POSIX_ACL) && !(def_mount_opts & EXT4_DEFM_ACL))
 		seq_puts(seq, ",acl");
 	if (!test_opt(sb, POSIX_ACL) && (def_mount_opts & EXT4_DEFM_ACL))
@@ -752,8 +755,6 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs)
 		seq_puts(seq, ",nobh");
 	if (!test_opt(sb, EXTENTS))
 		seq_puts(seq, ",noextents");
-	if (!test_opt(sb, MBALLOC))
-		seq_puts(seq, ",nomballoc");
 	if (test_opt(sb, I_VERSION))
 		seq_puts(seq, ",i_version");
 	if (!test_opt(sb, DELALLOC))
@@ -773,6 +774,13 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs)
 	else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA)
 		seq_puts(seq, ",data=writeback");
 
+	if (sbi->s_inode_readahead_blks != EXT4_DEF_INODE_READAHEAD_BLKS)
+		seq_printf(seq, ",inode_readahead_blks=%u",
+			   sbi->s_inode_readahead_blks);
+
+	if (test_opt(sb, DATA_ERR_ABORT))
+		seq_puts(seq, ",data_err=abort");
+
 	ext4_show_quota_options(seq, sb);
 	return 0;
 }
@@ -822,7 +830,7 @@ static struct dentry *ext4_fh_to_parent(struct super_block *sb, struct fid *fid,
 }
 
 #ifdef CONFIG_QUOTA
-#define QTYPE2NAME(t) ((t) == USRQUOTA?"user":"group")
+#define QTYPE2NAME(t) ((t) == USRQUOTA ? "user" : "group")
 #define QTYPE2MOPT(on, t) ((t) == USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA))
 
 static int ext4_dquot_initialize(struct inode *inode, int type);
@@ -902,11 +910,13 @@ enum {
 	Opt_commit, Opt_journal_update, Opt_journal_inum, Opt_journal_dev,
 	Opt_journal_checksum, Opt_journal_async_commit,
 	Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
+	Opt_data_err_abort, Opt_data_err_ignore,
 	Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
 	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota,
 	Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota,
 	Opt_grpquota, Opt_extents, Opt_noextents, Opt_i_version,
 	Opt_mballoc, Opt_nomballoc, Opt_stripe, Opt_delalloc, Opt_nodelalloc,
+	Opt_inode_readahead_blks
 };
 
 static match_table_t tokens = {
@@ -947,6 +957,8 @@ static match_table_t tokens = {
 	{Opt_data_journal, "data=journal"},
 	{Opt_data_ordered, "data=ordered"},
 	{Opt_data_writeback, "data=writeback"},
+	{Opt_data_err_abort, "data_err=abort"},
+	{Opt_data_err_ignore, "data_err=ignore"},
 	{Opt_offusrjquota, "usrjquota="},
 	{Opt_usrjquota, "usrjquota=%s"},
 	{Opt_offgrpjquota, "grpjquota="},
@@ -967,6 +979,7 @@ static match_table_t tokens = {
 	{Opt_resize, "resize"},
 	{Opt_delalloc, "delalloc"},
 	{Opt_nodelalloc, "nodelalloc"},
+	{Opt_inode_readahead_blks, "inode_readahead_blks=%u"},
 	{Opt_err, NULL},
 };
 
@@ -981,7 +994,7 @@ static ext4_fsblk_t get_sb_block(void **data)
 	/*todo: use simple_strtoll with >32bit ext4 */
 	sb_block = simple_strtoul(options, &options, 0);
 	if (*options && *options != ',') {
-		printk("EXT4-fs: Invalid sb specification: %s\n",
+		printk(KERN_ERR "EXT4-fs: Invalid sb specification: %s\n",
 		       (char *) *data);
 		return 1;
 	}
@@ -1072,7 +1085,7 @@ static int parse_options(char *options, struct super_block *sb,
 		case Opt_orlov:
 			clear_opt(sbi->s_mount_opt, OLDALLOC);
 			break;
-#ifdef CONFIG_EXT4DEV_FS_XATTR
+#ifdef CONFIG_EXT4_FS_XATTR
 		case Opt_user_xattr:
 			set_opt(sbi->s_mount_opt, XATTR_USER);
 			break;
@@ -1082,10 +1095,11 @@ static int parse_options(char *options, struct super_block *sb,
 #else
 		case Opt_user_xattr:
 		case Opt_nouser_xattr:
-			printk("EXT4 (no)user_xattr options not supported\n");
+			printk(KERN_ERR "EXT4 (no)user_xattr options "
+			       "not supported\n");
 			break;
 #endif
-#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
+#ifdef CONFIG_EXT4_FS_POSIX_ACL
 		case Opt_acl:
 			set_opt(sbi->s_mount_opt, POSIX_ACL);
 			break;
@@ -1095,7 +1109,8 @@ static int parse_options(char *options, struct super_block *sb,
 #else
 		case Opt_acl:
 		case Opt_noacl:
-			printk("EXT4 (no)acl options not supported\n");
+			printk(KERN_ERR "EXT4 (no)acl options "
+			       "not supported\n");
 			break;
 #endif
 		case Opt_reservation:
@@ -1178,6 +1193,12 @@ static int parse_options(char *options, struct super_block *sb,
 				sbi->s_mount_opt |= data_opt;
 			}
 			break;
+		case Opt_data_err_abort:
+			set_opt(sbi->s_mount_opt, DATA_ERR_ABORT);
+			break;
+		case Opt_data_err_ignore:
+			clear_opt(sbi->s_mount_opt, DATA_ERR_ABORT);
+			break;
 #ifdef CONFIG_QUOTA
 		case Opt_usrjquota:
 			qtype = USRQUOTA;
@@ -1189,8 +1210,8 @@ set_qf_name:
 			     sb_any_quota_suspended(sb)) &&
 			    !sbi->s_qf_names[qtype]) {
 				printk(KERN_ERR
-					"EXT4-fs: Cannot change journaled "
-					"quota options when quota turned on.\n");
+				       "EXT4-fs: Cannot change journaled "
+				       "quota options when quota turned on.\n");
 				return 0;
 			}
 			qname = match_strdup(&args[0]);
@@ -1357,12 +1378,6 @@ set_qf_format:
 		case Opt_nodelalloc:
 			clear_opt(sbi->s_mount_opt, DELALLOC);
 			break;
-		case Opt_mballoc:
-			set_opt(sbi->s_mount_opt, MBALLOC);
-			break;
-		case Opt_nomballoc:
-			clear_opt(sbi->s_mount_opt, MBALLOC);
-			break;
 		case Opt_stripe:
 			if (match_int(&args[0], &option))
 				return 0;
@@ -1373,6 +1388,13 @@ set_qf_format:
 		case Opt_delalloc:
 			set_opt(sbi->s_mount_opt, DELALLOC);
 			break;
+		case Opt_inode_readahead_blks:
+			if (match_int(&args[0], &option))
+				return 0;
+			if (option < 0 || option > (1 << 30))
+				return 0;
+			sbi->s_inode_readahead_blks = option;
+			break;
 		default:
 			printk(KERN_ERR
 			       "EXT4-fs: Unrecognized mount option \"%s\" "
@@ -1473,15 +1495,9 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es,
 			EXT4_INODES_PER_GROUP(sb),
 			sbi->s_mount_opt);
 
-	printk(KERN_INFO "EXT4 FS on %s, ", sb->s_id);
-	if (EXT4_SB(sb)->s_journal->j_inode == NULL) {
-		char b[BDEVNAME_SIZE];
-
-		printk("external journal on %s\n",
-			bdevname(EXT4_SB(sb)->s_journal->j_dev, b));
-	} else {
-		printk("internal journal\n");
-	}
+	printk(KERN_INFO "EXT4 FS on %s, %s journal on %s\n",
+	       sb->s_id, EXT4_SB(sb)->s_journal->j_inode ? "internal" :
+	       "external", EXT4_SB(sb)->s_journal->j_devname);
 	return res;
 }
 
@@ -1504,8 +1520,11 @@ static int ext4_fill_flex_info(struct super_block *sb)
 	sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex;
 	groups_per_flex = 1 << sbi->s_log_groups_per_flex;
 
-	flex_group_count = (sbi->s_groups_count + groups_per_flex - 1) /
-		groups_per_flex;
+	/* We allocate both existing and potentially added groups */
+	flex_group_count = ((sbi->s_groups_count + groups_per_flex - 1) +
+			    ((sbi->s_es->s_reserved_gdt_blocks +1 ) <<
+			      EXT4_DESC_PER_BLOCK_BITS(sb))) /
+			   groups_per_flex;
 	sbi->s_flex_groups = kzalloc(flex_group_count *
 				     sizeof(struct flex_groups), GFP_KERNEL);
 	if (sbi->s_flex_groups == NULL) {
@@ -1584,7 +1603,7 @@ static int ext4_check_descriptors(struct super_block *sb)
 	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FLEX_BG))
 		flexbg_flag = 1;
 
-	ext4_debug ("Checking group descriptors");
+	ext4_debug("Checking group descriptors");
 
 	for (i = 0; i < sbi->s_groups_count; i++) {
 		struct ext4_group_desc *gdp = ext4_get_group_desc(sb, i, NULL);
@@ -1623,8 +1642,10 @@ static int ext4_check_descriptors(struct super_block *sb)
 			       "Checksum for group %lu failed (%u!=%u)\n",
 			       i, le16_to_cpu(ext4_group_desc_csum(sbi, i,
 			       gdp)), le16_to_cpu(gdp->bg_checksum));
-			if (!(sb->s_flags & MS_RDONLY))
+			if (!(sb->s_flags & MS_RDONLY)) {
+				spin_unlock(sb_bgl_lock(sbi, i));
 				return 0;
+			}
 		}
 		spin_unlock(sb_bgl_lock(sbi, i));
 		if (!flexbg_flag)
@@ -1714,9 +1735,9 @@ static void ext4_orphan_cleanup(struct super_block *sb,
 		DQUOT_INIT(inode);
 		if (inode->i_nlink) {
 			printk(KERN_DEBUG
-				"%s: truncating inode %lu to %Ld bytes\n",
+				"%s: truncating inode %lu to %lld bytes\n",
 				__func__, inode->i_ino, inode->i_size);
-			jbd_debug(2, "truncating inode %lu to %Ld bytes\n",
+			jbd_debug(2, "truncating inode %lu to %lld bytes\n",
 				  inode->i_ino, inode->i_size);
 			ext4_truncate(inode);
 			nr_truncates++;
@@ -1914,6 +1935,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	unsigned long journal_devnum = 0;
 	unsigned long def_mount_opts;
 	struct inode *root;
+	char *cp;
 	int ret = -EINVAL;
 	int blocksize;
 	int db_count;
@@ -1930,10 +1952,15 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	sbi->s_mount_opt = 0;
 	sbi->s_resuid = EXT4_DEF_RESUID;
 	sbi->s_resgid = EXT4_DEF_RESGID;
+	sbi->s_inode_readahead_blks = EXT4_DEF_INODE_READAHEAD_BLKS;
 	sbi->s_sb_block = sb_block;
 
 	unlock_kernel();
 
+	/* Cleanup superblock name */
+	for (cp = sb->s_id; (cp = strchr(cp, '/'));)
+		*cp = '!';
+
 	blocksize = sb_min_blocksize(sb, EXT4_MIN_BLOCK_SIZE);
 	if (!blocksize) {
 		printk(KERN_ERR "EXT4-fs: unable to set blocksize\n");
@@ -1973,11 +2000,11 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 		set_opt(sbi->s_mount_opt, GRPID);
 	if (def_mount_opts & EXT4_DEFM_UID16)
 		set_opt(sbi->s_mount_opt, NO_UID32);
-#ifdef CONFIG_EXT4DEV_FS_XATTR
+#ifdef CONFIG_EXT4_FS_XATTR
 	if (def_mount_opts & EXT4_DEFM_XATTR_USER)
 		set_opt(sbi->s_mount_opt, XATTR_USER);
 #endif
-#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
+#ifdef CONFIG_EXT4_FS_POSIX_ACL
 	if (def_mount_opts & EXT4_DEFM_ACL)
 		set_opt(sbi->s_mount_opt, POSIX_ACL);
 #endif
@@ -2012,11 +2039,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 		ext4_warning(sb, __func__,
 			"extents feature not enabled on this filesystem, "
 			"use tune2fs.\n");
-	/*
-	 * turn on mballoc code by default in ext4 filesystem
-	 * Use -o nomballoc to turn it off
-	 */
-	set_opt(sbi->s_mount_opt, MBALLOC);
 
 	/*
 	 * enable delayed allocation by default
@@ -2041,16 +2063,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 		       "running e2fsck is recommended\n");
 
 	/*
-	 * Since ext4 is still considered development code, we require
-	 * that the TEST_FILESYS flag in s->flags be set.
-	 */
-	if (!(le32_to_cpu(es->s_flags) & EXT2_FLAGS_TEST_FILESYS)) {
-		printk(KERN_WARNING "EXT4-fs: %s: not marked "
-		       "OK to use with test code.\n", sb->s_id);
-		goto failed_mount;
-	}
-
-	/*
 	 * Check feature flags regardless of the revision level, since we
 	 * previously didn't change the revision level when setting the flags,
 	 * so there is a chance incompat flags are set on a rev 0 filesystem.
@@ -2219,6 +2231,16 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 		goto failed_mount;
 	}
 
+#ifdef CONFIG_PROC_FS
+	if (ext4_proc_root)
+		sbi->s_proc = proc_mkdir(sb->s_id, ext4_proc_root);
+
+	if (sbi->s_proc)
+		proc_create_data("inode_readahead_blks", 0644, sbi->s_proc,
+				 &ext4_ui_proc_fops,
+				 &sbi->s_inode_readahead_blks);
+#endif
+
 	bgl_lock_init(&sbi->s_blockgroup_lock);
 
 	for (i = 0; i < db_count; i++) {
@@ -2257,24 +2279,14 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 		err = percpu_counter_init(&sbi->s_dirs_counter,
 				ext4_count_dirs(sb));
 	}
+	if (!err) {
+		err = percpu_counter_init(&sbi->s_dirtyblocks_counter, 0);
+	}
 	if (err) {
 		printk(KERN_ERR "EXT4-fs: insufficient memory\n");
 		goto failed_mount3;
 	}
 
-	/* per fileystem reservation list head & lock */
-	spin_lock_init(&sbi->s_rsv_window_lock);
-	sbi->s_rsv_window_root = RB_ROOT;
-	/* Add a single, static dummy reservation to the start of the
-	 * reservation window list --- it gives us a placeholder for
-	 * append-at-start-of-list which makes the allocation logic
-	 * _much_ simpler. */
-	sbi->s_rsv_window_head.rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED;
-	sbi->s_rsv_window_head.rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED;
-	sbi->s_rsv_window_head.rsv_alloc_hit = 0;
-	sbi->s_rsv_window_head.rsv_goal_size = 0;
-	ext4_rsv_window_add(sb, &sbi->s_rsv_window_head);
-
 	sbi->s_stripe = ext4_get_stripe_size(sbi);
 
 	/*
@@ -2471,7 +2483,12 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 		printk(KERN_INFO "EXT4-fs: delayed allocation enabled\n");
 
 	ext4_ext_init(sb);
-	ext4_mb_init(sb, needs_recovery);
+	err = ext4_mb_init(sb, needs_recovery);
+	if (err) {
+		printk(KERN_ERR "EXT4-fs: failed to initalize mballoc (%d)\n",
+		       err);
+		goto failed_mount4;
+	}
 
 	lock_kernel();
 	return 0;
@@ -2489,11 +2506,16 @@ failed_mount3:
 	percpu_counter_destroy(&sbi->s_freeblocks_counter);
 	percpu_counter_destroy(&sbi->s_freeinodes_counter);
 	percpu_counter_destroy(&sbi->s_dirs_counter);
+	percpu_counter_destroy(&sbi->s_dirtyblocks_counter);
 failed_mount2:
 	for (i = 0; i < db_count; i++)
 		brelse(sbi->s_group_desc[i]);
 	kfree(sbi->s_group_desc);
 failed_mount:
+	if (sbi->s_proc) {
+		remove_proc_entry("inode_readahead_blks", sbi->s_proc);
+		remove_proc_entry(sb->s_id, ext4_proc_root);
+	}
 #ifdef CONFIG_QUOTA
 	for (i = 0; i < MAXQUOTAS; i++)
 		kfree(sbi->s_qf_names[i]);
@@ -2527,6 +2549,10 @@ static void ext4_init_journal_params(struct super_block *sb, journal_t *journal)
 		journal->j_flags |= JBD2_BARRIER;
 	else
 		journal->j_flags &= ~JBD2_BARRIER;
+	if (test_opt(sb, DATA_ERR_ABORT))
+		journal->j_flags |= JBD2_ABORT_ON_SYNCDATA_ERR;
+	else
+		journal->j_flags &= ~JBD2_ABORT_ON_SYNCDATA_ERR;
 	spin_unlock(&journal->j_state_lock);
 }
 
@@ -2552,7 +2578,7 @@ static journal_t *ext4_get_journal(struct super_block *sb,
 		return NULL;
 	}
 
-	jbd_debug(2, "Journal inode found at %p: %Ld bytes\n",
+	jbd_debug(2, "Journal inode found at %p: %lld bytes\n",
 		  journal_inode, journal_inode->i_size);
 	if (!S_ISREG(journal_inode->i_mode)) {
 		printk(KERN_ERR "EXT4-fs: invalid journal inode.\n");
@@ -2715,6 +2741,11 @@ static int ext4_load_journal(struct super_block *sb,
 			return -EINVAL;
 	}
 
+	if (journal->j_flags & JBD2_BARRIER)
+		printk(KERN_INFO "EXT4-fs: barriers enabled\n");
+	else
+		printk(KERN_INFO "EXT4-fs: barriers disabled\n");
+
 	if (!really_read_only && test_opt(sb, UPDATE_JOURNAL)) {
 		err = jbd2_journal_update_format(journal);
 		if (err)  {
@@ -2799,13 +2830,34 @@ static void ext4_commit_super(struct super_block *sb,
 
 	if (!sbh)
 		return;
+	if (buffer_write_io_error(sbh)) {
+		/*
+		 * Oh, dear.  A previous attempt to write the
+		 * superblock failed.  This could happen because the
+		 * USB device was yanked out.  Or it could happen to
+		 * be a transient write error and maybe the block will
+		 * be remapped.  Nothing we can do but to retry the
+		 * write and hope for the best.
+		 */
+		printk(KERN_ERR "ext4: previous I/O error to "
+		       "superblock detected for %s.\n", sb->s_id);
+		clear_buffer_write_io_error(sbh);
+		set_buffer_uptodate(sbh);
+	}
 	es->s_wtime = cpu_to_le32(get_seconds());
 	ext4_free_blocks_count_set(es, ext4_count_free_blocks(sb));
 	es->s_free_inodes_count = cpu_to_le32(ext4_count_free_inodes(sb));
 	BUFFER_TRACE(sbh, "marking dirty");
 	mark_buffer_dirty(sbh);
-	if (sync)
+	if (sync) {
 		sync_dirty_buffer(sbh);
+		if (buffer_write_io_error(sbh)) {
+			printk(KERN_ERR "ext4: I/O error while writing "
+			       "superblock for %s.\n", sb->s_id);
+			clear_buffer_write_io_error(sbh);
+			set_buffer_uptodate(sbh);
+		}
+	}
 }
 
 
@@ -2820,7 +2872,9 @@ static void ext4_mark_recovery_complete(struct super_block *sb,
 	journal_t *journal = EXT4_SB(sb)->s_journal;
 
 	jbd2_journal_lock_updates(journal);
-	jbd2_journal_flush(journal);
+	if (jbd2_journal_flush(journal) < 0)
+		goto out;
+
 	lock_super(sb);
 	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER) &&
 	    sb->s_flags & MS_RDONLY) {
@@ -2829,6 +2883,8 @@ static void ext4_mark_recovery_complete(struct super_block *sb,
 		ext4_commit_super(sb, es, 1);
 	}
 	unlock_super(sb);
+
+out:
 	jbd2_journal_unlock_updates(journal);
 }
 
@@ -2907,6 +2963,7 @@ static int ext4_sync_fs(struct super_block *sb, int wait)
 {
 	tid_t target;
 
+	trace_mark(ext4_sync_fs, "dev %s wait %d", sb->s_id, wait);
 	sb->s_dirt = 0;
 	if (jbd2_journal_start_commit(EXT4_SB(sb)->s_journal, &target)) {
 		if (wait)
@@ -2928,7 +2985,13 @@ static void ext4_write_super_lockfs(struct super_block *sb)
 
 		/* Now we set up the journal barrier. */
 		jbd2_journal_lock_updates(journal);
-		jbd2_journal_flush(journal);
+
+		/*
+		 * We don't want to clear needs_recovery flag when we failed
+		 * to flush the journal.
+		 */
+		if (jbd2_journal_flush(journal) < 0)
+			return;
 
 		/* Journal blocked and flushed, clear needs_recovery flag. */
 		EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
@@ -3162,7 +3225,8 @@ static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf)
 	buf->f_type = EXT4_SUPER_MAGIC;
 	buf->f_bsize = sb->s_blocksize;
 	buf->f_blocks = ext4_blocks_count(es) - sbi->s_overhead_last;
-	buf->f_bfree = percpu_counter_sum_positive(&sbi->s_freeblocks_counter);
+	buf->f_bfree = percpu_counter_sum_positive(&sbi->s_freeblocks_counter) -
+		       percpu_counter_sum_positive(&sbi->s_dirtyblocks_counter);
 	ext4_free_blocks_count_set(es, buf->f_bfree);
 	buf->f_bavail = buf->f_bfree - ext4_r_blocks_count(es);
 	if (buf->f_bfree < ext4_r_blocks_count(es))
@@ -3367,8 +3431,12 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id,
 		 * otherwise be livelocked...
 		 */
 		jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal);
-		jbd2_journal_flush(EXT4_SB(sb)->s_journal);
+		err = jbd2_journal_flush(EXT4_SB(sb)->s_journal);
 		jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal);
+		if (err) {
+			path_put(&nd.path);
+			return err;
+		}
 	}
 
 	err = vfs_quota_on_path(sb, type, format_id, &nd.path);
@@ -3432,7 +3500,7 @@ static ssize_t ext4_quota_write(struct super_block *sb, int type,
 	handle_t *handle = journal_current_handle();
 
 	if (!handle) {
-		printk(KERN_WARNING "EXT4-fs: Quota write (off=%Lu, len=%Lu)"
+		printk(KERN_WARNING "EXT4-fs: Quota write (off=%llu, len=%llu)"
 			" cancelled because transaction is not started.\n",
 			(unsigned long long)off, (unsigned long long)len);
 		return -EIO;
@@ -3493,18 +3561,82 @@ static int ext4_get_sb(struct file_system_type *fs_type,
 	return get_sb_bdev(fs_type, flags, dev_name, data, ext4_fill_super, mnt);
 }
 
+#ifdef CONFIG_PROC_FS
+static int ext4_ui_proc_show(struct seq_file *m, void *v)
+{
+	unsigned int *p = m->private;
+
+	seq_printf(m, "%u\n", *p);
+	return 0;
+}
+
+static int ext4_ui_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, ext4_ui_proc_show, PDE(inode)->data);
+}
+
+static ssize_t ext4_ui_proc_write(struct file *file, const char __user *buf,
+			       size_t cnt, loff_t *ppos)
+{
+	unsigned int *p = PDE(file->f_path.dentry->d_inode)->data;
+	char str[32];
+	unsigned long value;
+
+	if (cnt >= sizeof(str))
+		return -EINVAL;
+	if (copy_from_user(str, buf, cnt))
+		return -EFAULT;
+	value = simple_strtol(str, NULL, 0);
+	if (value < 0)
+		return -ERANGE;
+	*p = value;
+	return cnt;
+}
+
+const struct file_operations ext4_ui_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= ext4_ui_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+	.write		= ext4_ui_proc_write,
+};
+#endif
+
+static struct file_system_type ext4_fs_type = {
+	.owner		= THIS_MODULE,
+	.name		= "ext4",
+	.get_sb		= ext4_get_sb,
+	.kill_sb	= kill_block_super,
+	.fs_flags	= FS_REQUIRES_DEV,
+};
+
+#ifdef CONFIG_EXT4DEV_COMPAT
+static int ext4dev_get_sb(struct file_system_type *fs_type,
+	int flags, const char *dev_name, void *data, struct vfsmount *mnt)
+{
+	printk(KERN_WARNING "EXT4-fs: Update your userspace programs "
+	       "to mount using ext4\n");
+	printk(KERN_WARNING "EXT4-fs: ext4dev backwards compatibility "
+	       "will go away by 2.6.31\n");
+	return get_sb_bdev(fs_type, flags, dev_name, data, ext4_fill_super, mnt);
+}
+
 static struct file_system_type ext4dev_fs_type = {
 	.owner		= THIS_MODULE,
 	.name		= "ext4dev",
-	.get_sb		= ext4_get_sb,
+	.get_sb		= ext4dev_get_sb,
 	.kill_sb	= kill_block_super,
 	.fs_flags	= FS_REQUIRES_DEV,
 };
+MODULE_ALIAS("ext4dev");
+#endif
 
 static int __init init_ext4_fs(void)
 {
 	int err;
 
+	ext4_proc_root = proc_mkdir("fs/ext4", NULL);
 	err = init_ext4_mballoc();
 	if (err)
 		return err;
@@ -3515,9 +3647,16 @@ static int __init init_ext4_fs(void)
 	err = init_inodecache();
 	if (err)
 		goto out1;
-	err = register_filesystem(&ext4dev_fs_type);
+	err = register_filesystem(&ext4_fs_type);
 	if (err)
 		goto out;
+#ifdef CONFIG_EXT4DEV_COMPAT
+	err = register_filesystem(&ext4dev_fs_type);
+	if (err) {
+		unregister_filesystem(&ext4_fs_type);
+		goto out;
+	}
+#endif
 	return 0;
 out:
 	destroy_inodecache();
@@ -3530,10 +3669,14 @@ out2:
 
 static void __exit exit_ext4_fs(void)
 {
+	unregister_filesystem(&ext4_fs_type);
+#ifdef CONFIG_EXT4DEV_COMPAT
 	unregister_filesystem(&ext4dev_fs_type);
+#endif
 	destroy_inodecache();
 	exit_ext4_xattr();
 	exit_ext4_mballoc();
+	remove_proc_entry("fs/ext4", NULL);
 }
 
 MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others");
diff --git a/fs/ext4/symlink.c b/fs/ext4/symlink.c
index e9178643dc01..00740cb32be3 100644
--- a/fs/ext4/symlink.c
+++ b/fs/ext4/symlink.c
@@ -23,10 +23,10 @@
 #include "ext4.h"
 #include "xattr.h"
 
-static void * ext4_follow_link(struct dentry *dentry, struct nameidata *nd)
+static void *ext4_follow_link(struct dentry *dentry, struct nameidata *nd)
 {
 	struct ext4_inode_info *ei = EXT4_I(dentry->d_inode);
-	nd_set_link(nd, (char*)ei->i_data);
+	nd_set_link(nd, (char *) ei->i_data);
 	return NULL;
 }
 
@@ -34,7 +34,7 @@ const struct inode_operations ext4_symlink_inode_operations = {
 	.readlink	= generic_readlink,
 	.follow_link	= page_follow_link_light,
 	.put_link	= page_put_link,
-#ifdef CONFIG_EXT4DEV_FS_XATTR
+#ifdef CONFIG_EXT4_FS_XATTR
 	.setxattr	= generic_setxattr,
 	.getxattr	= generic_getxattr,
 	.listxattr	= ext4_listxattr,
@@ -45,7 +45,7 @@ const struct inode_operations ext4_symlink_inode_operations = {
 const struct inode_operations ext4_fast_symlink_inode_operations = {
 	.readlink	= generic_readlink,
 	.follow_link	= ext4_follow_link,
-#ifdef CONFIG_EXT4DEV_FS_XATTR
+#ifdef CONFIG_EXT4_FS_XATTR
 	.setxattr	= generic_setxattr,
 	.getxattr	= generic_getxattr,
 	.listxattr	= ext4_listxattr,
diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c
index 8954208b4893..80626d516fee 100644
--- a/fs/ext4/xattr.c
+++ b/fs/ext4/xattr.c
@@ -99,12 +99,12 @@ static struct mb_cache *ext4_xattr_cache;
 
 static struct xattr_handler *ext4_xattr_handler_map[] = {
 	[EXT4_XATTR_INDEX_USER]		     = &ext4_xattr_user_handler,
-#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
+#ifdef CONFIG_EXT4_FS_POSIX_ACL
 	[EXT4_XATTR_INDEX_POSIX_ACL_ACCESS]  = &ext4_xattr_acl_access_handler,
 	[EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT] = &ext4_xattr_acl_default_handler,
 #endif
 	[EXT4_XATTR_INDEX_TRUSTED]	     = &ext4_xattr_trusted_handler,
-#ifdef CONFIG_EXT4DEV_FS_SECURITY
+#ifdef CONFIG_EXT4_FS_SECURITY
 	[EXT4_XATTR_INDEX_SECURITY]	     = &ext4_xattr_security_handler,
 #endif
 };
@@ -112,11 +112,11 @@ static struct xattr_handler *ext4_xattr_handler_map[] = {
 struct xattr_handler *ext4_xattr_handlers[] = {
 	&ext4_xattr_user_handler,
 	&ext4_xattr_trusted_handler,
-#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL
+#ifdef CONFIG_EXT4_FS_POSIX_ACL
 	&ext4_xattr_acl_access_handler,
 	&ext4_xattr_acl_default_handler,
 #endif
-#ifdef CONFIG_EXT4DEV_FS_SECURITY
+#ifdef CONFIG_EXT4_FS_SECURITY
 	&ext4_xattr_security_handler,
 #endif
 	NULL
@@ -959,6 +959,7 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index,
 	struct ext4_xattr_block_find bs = {
 		.s = { .not_found = -ENODATA, },
 	};
+	unsigned long no_expand;
 	int error;
 
 	if (!name)
@@ -966,6 +967,9 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index,
 	if (strlen(name) > 255)
 		return -ERANGE;
 	down_write(&EXT4_I(inode)->xattr_sem);
+	no_expand = EXT4_I(inode)->i_state & EXT4_STATE_NO_EXPAND;
+	EXT4_I(inode)->i_state |= EXT4_STATE_NO_EXPAND;
+
 	error = ext4_get_inode_loc(inode, &is.iloc);
 	if (error)
 		goto cleanup;
@@ -1042,6 +1046,8 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index,
 cleanup:
 	brelse(is.iloc.bh);
 	brelse(bs.bh);
+	if (no_expand == 0)
+		EXT4_I(inode)->i_state &= ~EXT4_STATE_NO_EXPAND;
 	up_write(&EXT4_I(inode)->xattr_sem);
 	return error;
 }
diff --git a/fs/ext4/xattr.h b/fs/ext4/xattr.h
index 5992fe979bb9..8ede88b18c29 100644
--- a/fs/ext4/xattr.h
+++ b/fs/ext4/xattr.h
@@ -51,8 +51,8 @@ struct ext4_xattr_entry {
 	(((name_len) + EXT4_XATTR_ROUND + \
 	sizeof(struct ext4_xattr_entry)) & ~EXT4_XATTR_ROUND)
 #define EXT4_XATTR_NEXT(entry) \
-	( (struct ext4_xattr_entry *)( \
-	  (char *)(entry) + EXT4_XATTR_LEN((entry)->e_name_len)) )
+	((struct ext4_xattr_entry *)( \
+	 (char *)(entry) + EXT4_XATTR_LEN((entry)->e_name_len)))
 #define EXT4_XATTR_SIZE(size) \
 	(((size) + EXT4_XATTR_ROUND) & ~EXT4_XATTR_ROUND)
 
@@ -63,7 +63,7 @@ struct ext4_xattr_entry {
 		EXT4_I(inode)->i_extra_isize))
 #define IFIRST(hdr) ((struct ext4_xattr_entry *)((hdr)+1))
 
-# ifdef CONFIG_EXT4DEV_FS_XATTR
+# ifdef CONFIG_EXT4_FS_XATTR
 
 extern struct xattr_handler ext4_xattr_user_handler;
 extern struct xattr_handler ext4_xattr_trusted_handler;
@@ -88,7 +88,7 @@ extern void exit_ext4_xattr(void);
 
 extern struct xattr_handler *ext4_xattr_handlers[];
 
-# else  /* CONFIG_EXT4DEV_FS_XATTR */
+# else  /* CONFIG_EXT4_FS_XATTR */
 
 static inline int
 ext4_xattr_get(struct inode *inode, int name_index, const char *name,
@@ -141,9 +141,9 @@ ext4_expand_extra_isize_ea(struct inode *inode, int new_extra_isize,
 
 #define ext4_xattr_handlers	NULL
 
-# endif  /* CONFIG_EXT4DEV_FS_XATTR */
+# endif  /* CONFIG_EXT4_FS_XATTR */
 
-#ifdef CONFIG_EXT4DEV_FS_SECURITY
+#ifdef CONFIG_EXT4_FS_SECURITY
 extern int ext4_init_security(handle_t *handle, struct inode *inode,
 				struct inode *dir);
 #else
diff --git a/fs/ioctl.c b/fs/ioctl.c
index 7db32b3382d3..d152856c371b 100644
--- a/fs/ioctl.c
+++ b/fs/ioctl.c
@@ -13,9 +13,14 @@
 #include <linux/security.h>
 #include <linux/module.h>
 #include <linux/uaccess.h>
+#include <linux/writeback.h>
+#include <linux/buffer_head.h>
 
 #include <asm/ioctls.h>
 
+/* So that the fiemap access checks can't overflow on 32 bit machines. */
+#define FIEMAP_MAX_EXTENTS	(UINT_MAX / sizeof(struct fiemap_extent))
+
 /**
  * vfs_ioctl - call filesystem specific ioctl methods
  * @filp:	open file to invoke ioctl method on
@@ -71,6 +76,276 @@ static int ioctl_fibmap(struct file *filp, int __user *p)
 	return put_user(res, p);
 }
 
+/**
+ * fiemap_fill_next_extent - Fiemap helper function
+ * @fieinfo:	Fiemap context passed into ->fiemap
+ * @logical:	Extent logical start offset, in bytes
+ * @phys:	Extent physical start offset, in bytes
+ * @len:	Extent length, in bytes
+ * @flags:	FIEMAP_EXTENT flags that describe this extent
+ *
+ * Called from file system ->fiemap callback. Will populate extent
+ * info as passed in via arguments and copy to user memory. On
+ * success, extent count on fieinfo is incremented.
+ *
+ * Returns 0 on success, -errno on error, 1 if this was the last
+ * extent that will fit in user array.
+ */
+#define SET_UNKNOWN_FLAGS	(FIEMAP_EXTENT_DELALLOC)
+#define SET_NO_UNMOUNTED_IO_FLAGS	(FIEMAP_EXTENT_DATA_ENCRYPTED)
+#define SET_NOT_ALIGNED_FLAGS	(FIEMAP_EXTENT_DATA_TAIL|FIEMAP_EXTENT_DATA_INLINE)
+int fiemap_fill_next_extent(struct fiemap_extent_info *fieinfo, u64 logical,
+			    u64 phys, u64 len, u32 flags)
+{
+	struct fiemap_extent extent;
+	struct fiemap_extent *dest = fieinfo->fi_extents_start;
+
+	/* only count the extents */
+	if (fieinfo->fi_extents_max == 0) {
+		fieinfo->fi_extents_mapped++;
+		return (flags & FIEMAP_EXTENT_LAST) ? 1 : 0;
+	}
+
+	if (fieinfo->fi_extents_mapped >= fieinfo->fi_extents_max)
+		return 1;
+
+	if (flags & SET_UNKNOWN_FLAGS)
+		flags |= FIEMAP_EXTENT_UNKNOWN;
+	if (flags & SET_NO_UNMOUNTED_IO_FLAGS)
+		flags |= FIEMAP_EXTENT_ENCODED;
+	if (flags & SET_NOT_ALIGNED_FLAGS)
+		flags |= FIEMAP_EXTENT_NOT_ALIGNED;
+
+	memset(&extent, 0, sizeof(extent));
+	extent.fe_logical = logical;
+	extent.fe_physical = phys;
+	extent.fe_length = len;
+	extent.fe_flags = flags;
+
+	dest += fieinfo->fi_extents_mapped;
+	if (copy_to_user(dest, &extent, sizeof(extent)))
+		return -EFAULT;
+
+	fieinfo->fi_extents_mapped++;
+	if (fieinfo->fi_extents_mapped == fieinfo->fi_extents_max)
+		return 1;
+	return (flags & FIEMAP_EXTENT_LAST) ? 1 : 0;
+}
+EXPORT_SYMBOL(fiemap_fill_next_extent);
+
+/**
+ * fiemap_check_flags - check validity of requested flags for fiemap
+ * @fieinfo:	Fiemap context passed into ->fiemap
+ * @fs_flags:	Set of fiemap flags that the file system understands
+ *
+ * Called from file system ->fiemap callback. This will compute the
+ * intersection of valid fiemap flags and those that the fs supports. That
+ * value is then compared against the user supplied flags. In case of bad user
+ * flags, the invalid values will be written into the fieinfo structure, and
+ * -EBADR is returned, which tells ioctl_fiemap() to return those values to
+ * userspace. For this reason, a return code of -EBADR should be preserved.
+ *
+ * Returns 0 on success, -EBADR on bad flags.
+ */
+int fiemap_check_flags(struct fiemap_extent_info *fieinfo, u32 fs_flags)
+{
+	u32 incompat_flags;
+
+	incompat_flags = fieinfo->fi_flags & ~(FIEMAP_FLAGS_COMPAT & fs_flags);
+	if (incompat_flags) {
+		fieinfo->fi_flags = incompat_flags;
+		return -EBADR;
+	}
+	return 0;
+}
+EXPORT_SYMBOL(fiemap_check_flags);
+
+static int fiemap_check_ranges(struct super_block *sb,
+			       u64 start, u64 len, u64 *new_len)
+{
+	*new_len = len;
+
+	if (len == 0)
+		return -EINVAL;
+
+	if (start > sb->s_maxbytes)
+		return -EFBIG;
+
+	/*
+	 * Shrink request scope to what the fs can actually handle.
+	 */
+	if ((len > sb->s_maxbytes) ||
+	    (sb->s_maxbytes - len) < start)
+		*new_len = sb->s_maxbytes - start;
+
+	return 0;
+}
+
+static int ioctl_fiemap(struct file *filp, unsigned long arg)
+{
+	struct fiemap fiemap;
+	struct fiemap_extent_info fieinfo = { 0, };
+	struct inode *inode = filp->f_path.dentry->d_inode;
+	struct super_block *sb = inode->i_sb;
+	u64 len;
+	int error;
+
+	if (!inode->i_op->fiemap)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&fiemap, (struct fiemap __user *)arg,
+			   sizeof(struct fiemap)))
+		return -EFAULT;
+
+	if (fiemap.fm_extent_count > FIEMAP_MAX_EXTENTS)
+		return -EINVAL;
+
+	error = fiemap_check_ranges(sb, fiemap.fm_start, fiemap.fm_length,
+				    &len);
+	if (error)
+		return error;
+
+	fieinfo.fi_flags = fiemap.fm_flags;
+	fieinfo.fi_extents_max = fiemap.fm_extent_count;
+	fieinfo.fi_extents_start = (struct fiemap_extent *)(arg + sizeof(fiemap));
+
+	if (fiemap.fm_extent_count != 0 &&
+	    !access_ok(VERIFY_WRITE, fieinfo.fi_extents_start,
+		       fieinfo.fi_extents_max * sizeof(struct fiemap_extent)))
+		return -EFAULT;
+
+	if (fieinfo.fi_flags & FIEMAP_FLAG_SYNC)
+		filemap_write_and_wait(inode->i_mapping);
+
+	error = inode->i_op->fiemap(inode, &fieinfo, fiemap.fm_start, len);
+	fiemap.fm_flags = fieinfo.fi_flags;
+	fiemap.fm_mapped_extents = fieinfo.fi_extents_mapped;
+	if (copy_to_user((char *)arg, &fiemap, sizeof(fiemap)))
+		error = -EFAULT;
+
+	return error;
+}
+
+#ifdef CONFIG_BLOCK
+
+#define blk_to_logical(inode, blk) (blk << (inode)->i_blkbits)
+#define logical_to_blk(inode, offset) (offset >> (inode)->i_blkbits);
+
+/*
+ * @inode - the inode to map
+ * @arg - the pointer to userspace where we copy everything to
+ * @get_block - the fs's get_block function
+ *
+ * This does FIEMAP for block based inodes.  Basically it will just loop
+ * through get_block until we hit the number of extents we want to map, or we
+ * go past the end of the file and hit a hole.
+ *
+ * If it is possible to have data blocks beyond a hole past @inode->i_size, then
+ * please do not use this function, it will stop at the first unmapped block
+ * beyond i_size
+ */
+int generic_block_fiemap(struct inode *inode,
+			 struct fiemap_extent_info *fieinfo, u64 start,
+			 u64 len, get_block_t *get_block)
+{
+	struct buffer_head tmp;
+	unsigned int start_blk;
+	long long length = 0, map_len = 0;
+	u64 logical = 0, phys = 0, size = 0;
+	u32 flags = FIEMAP_EXTENT_MERGED;
+	int ret = 0;
+
+	if ((ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC)))
+		return ret;
+
+	start_blk = logical_to_blk(inode, start);
+
+	/* guard against change */
+	mutex_lock(&inode->i_mutex);
+
+	length = (long long)min_t(u64, len, i_size_read(inode));
+	map_len = length;
+
+	do {
+		/*
+		 * we set b_size to the total size we want so it will map as
+		 * many contiguous blocks as possible at once
+		 */
+		memset(&tmp, 0, sizeof(struct buffer_head));
+		tmp.b_size = map_len;
+
+		ret = get_block(inode, start_blk, &tmp, 0);
+		if (ret)
+			break;
+
+		/* HOLE */
+		if (!buffer_mapped(&tmp)) {
+			/*
+			 * first hole after going past the EOF, this is our
+			 * last extent
+			 */
+			if (length <= 0) {
+				flags = FIEMAP_EXTENT_MERGED|FIEMAP_EXTENT_LAST;
+				ret = fiemap_fill_next_extent(fieinfo, logical,
+							      phys, size,
+							      flags);
+				break;
+			}
+
+			length -= blk_to_logical(inode, 1);
+
+			/* if we have holes up to/past EOF then we're done */
+			if (length <= 0)
+				break;
+
+			start_blk++;
+		} else {
+			if (length <= 0 && size) {
+				ret = fiemap_fill_next_extent(fieinfo, logical,
+							      phys, size,
+							      flags);
+				if (ret)
+					break;
+			}
+
+			logical = blk_to_logical(inode, start_blk);
+			phys = blk_to_logical(inode, tmp.b_blocknr);
+			size = tmp.b_size;
+			flags = FIEMAP_EXTENT_MERGED;
+
+			length -= tmp.b_size;
+			start_blk += logical_to_blk(inode, size);
+
+			/*
+			 * if we are past the EOF we need to loop again to see
+			 * if there is a hole so we can mark this extent as the
+			 * last one, and if not keep mapping things until we
+			 * find a hole, or we run out of slots in the extent
+			 * array
+			 */
+			if (length <= 0)
+				continue;
+
+			ret = fiemap_fill_next_extent(fieinfo, logical, phys,
+						      size, flags);
+			if (ret)
+				break;
+		}
+		cond_resched();
+	} while (1);
+
+	mutex_unlock(&inode->i_mutex);
+
+	/* if ret is 1 then we just hit the end of the extent array */
+	if (ret == 1)
+		ret = 0;
+
+	return ret;
+}
+EXPORT_SYMBOL(generic_block_fiemap);
+
+#endif  /*  CONFIG_BLOCK  */
+
 static int file_ioctl(struct file *filp, unsigned int cmd,
 		unsigned long arg)
 {
@@ -80,6 +355,8 @@ static int file_ioctl(struct file *filp, unsigned int cmd,
 	switch (cmd) {
 	case FIBMAP:
 		return ioctl_fibmap(filp, p);
+	case FS_IOC_FIEMAP:
+		return ioctl_fiemap(filp, arg);
 	case FIGETBSZ:
 		return put_user(inode->i_sb->s_blocksize, p);
 	case FIONREAD:
diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index 91389c8aee8a..9203c3332f17 100644
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -20,6 +20,7 @@
 #include <linux/time.h>
 #include <linux/fs.h>
 #include <linux/jbd2.h>
+#include <linux/marker.h>
 #include <linux/errno.h>
 #include <linux/slab.h>
 
@@ -93,7 +94,8 @@ static int __try_to_free_cp_buf(struct journal_head *jh)
 	int ret = 0;
 	struct buffer_head *bh = jh2bh(jh);
 
-	if (jh->b_jlist == BJ_None && !buffer_locked(bh) && !buffer_dirty(bh)) {
+	if (jh->b_jlist == BJ_None && !buffer_locked(bh) &&
+	    !buffer_dirty(bh) && !buffer_write_io_error(bh)) {
 		JBUFFER_TRACE(jh, "remove from checkpoint list");
 		ret = __jbd2_journal_remove_checkpoint(jh) + 1;
 		jbd_unlock_bh_state(bh);
@@ -126,14 +128,29 @@ void __jbd2_log_wait_for_space(journal_t *journal)
 
 		/*
 		 * Test again, another process may have checkpointed while we
-		 * were waiting for the checkpoint lock
+		 * were waiting for the checkpoint lock. If there are no
+		 * outstanding transactions there is nothing to checkpoint and
+		 * we can't make progress. Abort the journal in this case.
 		 */
 		spin_lock(&journal->j_state_lock);
+		spin_lock(&journal->j_list_lock);
 		nblocks = jbd_space_needed(journal);
 		if (__jbd2_log_space_left(journal) < nblocks) {
+			int chkpt = journal->j_checkpoint_transactions != NULL;
+
+			spin_unlock(&journal->j_list_lock);
 			spin_unlock(&journal->j_state_lock);
-			jbd2_log_do_checkpoint(journal);
+			if (chkpt) {
+				jbd2_log_do_checkpoint(journal);
+			} else {
+				printk(KERN_ERR "%s: no transactions\n",
+				       __func__);
+				jbd2_journal_abort(journal, 0);
+			}
+
 			spin_lock(&journal->j_state_lock);
+		} else {
+			spin_unlock(&journal->j_list_lock);
 		}
 		mutex_unlock(&journal->j_checkpoint_mutex);
 	}
@@ -160,21 +177,25 @@ static void jbd_sync_bh(journal_t *journal, struct buffer_head *bh)
  * buffers. Note that we take the buffers in the opposite ordering
  * from the one in which they were submitted for IO.
  *
+ * Return 0 on success, and return <0 if some buffers have failed
+ * to be written out.
+ *
  * Called with j_list_lock held.
  */
-static void __wait_cp_io(journal_t *journal, transaction_t *transaction)
+static int __wait_cp_io(journal_t *journal, transaction_t *transaction)
 {
 	struct journal_head *jh;
 	struct buffer_head *bh;
 	tid_t this_tid;
 	int released = 0;
+	int ret = 0;
 
 	this_tid = transaction->t_tid;
 restart:
 	/* Did somebody clean up the transaction in the meanwhile? */
 	if (journal->j_checkpoint_transactions != transaction ||
 			transaction->t_tid != this_tid)
-		return;
+		return ret;
 	while (!released && transaction->t_checkpoint_io_list) {
 		jh = transaction->t_checkpoint_io_list;
 		bh = jh2bh(jh);
@@ -194,6 +215,9 @@ restart:
 			spin_lock(&journal->j_list_lock);
 			goto restart;
 		}
+		if (unlikely(buffer_write_io_error(bh)))
+			ret = -EIO;
+
 		/*
 		 * Now in whatever state the buffer currently is, we know that
 		 * it has been written out and so we can drop it from the list
@@ -203,6 +227,8 @@ restart:
 		jbd2_journal_remove_journal_head(bh);
 		__brelse(bh);
 	}
+
+	return ret;
 }
 
 #define NR_BATCH	64
@@ -226,7 +252,8 @@ __flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count)
  * Try to flush one buffer from the checkpoint list to disk.
  *
  * Return 1 if something happened which requires us to abort the current
- * scan of the checkpoint list.
+ * scan of the checkpoint list.  Return <0 if the buffer has failed to
+ * be written out.
  *
  * Called with j_list_lock held and drops it if 1 is returned
  * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it
@@ -258,6 +285,9 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
 		jbd2_log_wait_commit(journal, tid);
 		ret = 1;
 	} else if (!buffer_dirty(bh)) {
+		ret = 1;
+		if (unlikely(buffer_write_io_error(bh)))
+			ret = -EIO;
 		J_ASSERT_JH(jh, !buffer_jbddirty(bh));
 		BUFFER_TRACE(bh, "remove from checkpoint");
 		__jbd2_journal_remove_checkpoint(jh);
@@ -265,7 +295,6 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
 		jbd_unlock_bh_state(bh);
 		jbd2_journal_remove_journal_head(bh);
 		__brelse(bh);
-		ret = 1;
 	} else {
 		/*
 		 * Important: we are about to write the buffer, and
@@ -298,6 +327,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
  * to disk. We submit larger chunks of data at once.
  *
  * The journal should be locked before calling this function.
+ * Called with j_checkpoint_mutex held.
  */
 int jbd2_log_do_checkpoint(journal_t *journal)
 {
@@ -313,6 +343,8 @@ int jbd2_log_do_checkpoint(journal_t *journal)
 	 * journal straight away.
 	 */
 	result = jbd2_cleanup_journal_tail(journal);
+	trace_mark(jbd2_checkpoint, "dev %s need_checkpoint %d",
+		   journal->j_devname, result);
 	jbd_debug(1, "cleanup_journal_tail returned %d\n", result);
 	if (result <= 0)
 		return result;
@@ -321,6 +353,7 @@ int jbd2_log_do_checkpoint(journal_t *journal)
 	 * OK, we need to start writing disk blocks.  Take one transaction
 	 * and write it.
 	 */
+	result = 0;
 	spin_lock(&journal->j_list_lock);
 	if (!journal->j_checkpoint_transactions)
 		goto out;
@@ -339,7 +372,7 @@ restart:
 		int batch_count = 0;
 		struct buffer_head *bhs[NR_BATCH];
 		struct journal_head *jh;
-		int retry = 0;
+		int retry = 0, err;
 
 		while (!retry && transaction->t_checkpoint_list) {
 			struct buffer_head *bh;
@@ -353,6 +386,8 @@ restart:
 			}
 			retry = __process_buffer(journal, jh, bhs, &batch_count,
 						 transaction);
+			if (retry < 0 && !result)
+				result = retry;
 			if (!retry && (need_resched() ||
 				spin_needbreak(&journal->j_list_lock))) {
 				spin_unlock(&journal->j_list_lock);
@@ -377,14 +412,18 @@ restart:
 		 * Now we have cleaned up the first transaction's checkpoint
 		 * list. Let's clean up the second one
 		 */
-		__wait_cp_io(journal, transaction);
+		err = __wait_cp_io(journal, transaction);
+		if (!result)
+			result = err;
 	}
 out:
 	spin_unlock(&journal->j_list_lock);
-	result = jbd2_cleanup_journal_tail(journal);
 	if (result < 0)
-		return result;
-	return 0;
+		jbd2_journal_abort(journal, result);
+	else
+		result = jbd2_cleanup_journal_tail(journal);
+
+	return (result < 0) ? result : 0;
 }
 
 /*
@@ -400,8 +439,9 @@ out:
  * This is the only part of the journaling code which really needs to be
  * aware of transaction aborts.  Checkpointing involves writing to the
  * main filesystem area rather than to the journal, so it can proceed
- * even in abort state, but we must not update the journal superblock if
- * we have an abort error outstanding.
+ * even in abort state, but we must not update the super block if
+ * checkpointing may have failed.  Otherwise, we would lose some metadata
+ * buffers which should be written-back to the filesystem.
  */
 
 int jbd2_cleanup_journal_tail(journal_t *journal)
@@ -410,6 +450,9 @@ int jbd2_cleanup_journal_tail(journal_t *journal)
 	tid_t		first_tid;
 	unsigned long	blocknr, freed;
 
+	if (is_journal_aborted(journal))
+		return 1;
+
 	/* OK, work out the oldest transaction remaining in the log, and
 	 * the log block it starts at.
 	 *
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index f2ad061e95ec..0abe02c4242a 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -16,6 +16,7 @@
 #include <linux/time.h>
 #include <linux/fs.h>
 #include <linux/jbd2.h>
+#include <linux/marker.h>
 #include <linux/errno.h>
 #include <linux/slab.h>
 #include <linux/mm.h>
@@ -126,8 +127,7 @@ static int journal_submit_commit_record(journal_t *journal,
 
 	JBUFFER_TRACE(descriptor, "submit commit block");
 	lock_buffer(bh);
-	get_bh(bh);
-	set_buffer_dirty(bh);
+	clear_buffer_dirty(bh);
 	set_buffer_uptodate(bh);
 	bh->b_end_io = journal_end_buffer_io_sync;
 
@@ -147,12 +147,9 @@ static int journal_submit_commit_record(journal_t *journal,
 	 * to remember if we sent a barrier request
 	 */
 	if (ret == -EOPNOTSUPP && barrier_done) {
-		char b[BDEVNAME_SIZE];
-
 		printk(KERN_WARNING
-			"JBD: barrier-based sync failed on %s - "
-			"disabling barriers\n",
-			bdevname(journal->j_dev, b));
+		       "JBD: barrier-based sync failed on %s - "
+		       "disabling barriers\n", journal->j_devname);
 		spin_lock(&journal->j_state_lock);
 		journal->j_flags &= ~JBD2_BARRIER;
 		spin_unlock(&journal->j_state_lock);
@@ -160,7 +157,7 @@ static int journal_submit_commit_record(journal_t *journal,
 		/* And try again, without the barrier */
 		lock_buffer(bh);
 		set_buffer_uptodate(bh);
-		set_buffer_dirty(bh);
+		clear_buffer_dirty(bh);
 		ret = submit_bh(WRITE, bh);
 	}
 	*cbh = bh;
@@ -371,6 +368,8 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	commit_transaction = journal->j_running_transaction;
 	J_ASSERT(commit_transaction->t_state == T_RUNNING);
 
+	trace_mark(jbd2_start_commit, "dev %s transaction %d",
+		   journal->j_devname, commit_transaction->t_tid);
 	jbd_debug(1, "JBD: starting commit of transaction %d\n",
 			commit_transaction->t_tid);
 
@@ -505,9 +504,10 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 		jh = commit_transaction->t_buffers;
 
 		/* If we're in abort mode, we just un-journal the buffer and
-		   release it for background writing. */
+		   release it. */
 
 		if (is_journal_aborted(journal)) {
+			clear_buffer_jbddirty(jh2bh(jh));
 			JBUFFER_TRACE(jh, "journal is aborting: refile");
 			jbd2_journal_refile_buffer(journal, jh);
 			/* If that was the last one, we need to clean up
@@ -681,11 +681,11 @@ start_journal_io:
 	 */
 	err = journal_finish_inode_data_buffers(journal, commit_transaction);
 	if (err) {
-		char b[BDEVNAME_SIZE];
-
 		printk(KERN_WARNING
 			"JBD2: Detected IO errors while flushing file data "
-			"on %s\n", bdevname(journal->j_fs_dev, b));
+		       "on %s\n", journal->j_devname);
+		if (journal->j_flags & JBD2_ABORT_ON_SYNCDATA_ERR)
+			jbd2_journal_abort(journal, err);
 		err = 0;
 	}
 
@@ -786,6 +786,9 @@ wait_for_iobuf:
 		/* AKPM: bforget here */
 	}
 
+	if (err)
+		jbd2_journal_abort(journal, err);
+
 	jbd_debug(3, "JBD: commit phase 5\n");
 
 	if (!JBD2_HAS_INCOMPAT_FEATURE(journal,
@@ -884,6 +887,8 @@ restart_loop:
 		if (buffer_jbddirty(bh)) {
 			JBUFFER_TRACE(jh, "add to new checkpointing trans");
 			__jbd2_journal_insert_checkpoint(jh, commit_transaction);
+			if (is_journal_aborted(journal))
+				clear_buffer_jbddirty(bh);
 			JBUFFER_TRACE(jh, "refile for checkpoint writeback");
 			__jbd2_journal_refile_buffer(jh);
 			jbd_unlock_bh_state(bh);
@@ -990,6 +995,9 @@ restart_loop:
 	}
 	spin_unlock(&journal->j_list_lock);
 
+	trace_mark(jbd2_end_commit, "dev %s transaction %d head %d",
+		   journal->j_devname, commit_transaction->t_tid,
+		   journal->j_tail_sequence);
 	jbd_debug(1, "JBD: commit %d complete, head %d\n",
 		  journal->j_commit_sequence, journal->j_tail_sequence);
 
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 8207a01c4edb..783de118de92 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -597,13 +597,9 @@ int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr,
 		if (ret)
 			*retp = ret;
 		else {
-			char b[BDEVNAME_SIZE];
-
 			printk(KERN_ALERT "%s: journal block not found "
 					"at offset %lu on %s\n",
-				__func__,
-				blocknr,
-				bdevname(journal->j_dev, b));
+			       __func__, blocknr, journal->j_devname);
 			err = -EIO;
 			__journal_abort_soft(journal, err);
 		}
@@ -901,10 +897,7 @@ static struct proc_dir_entry *proc_jbd2_stats;
 
 static void jbd2_stats_proc_init(journal_t *journal)
 {
-	char name[BDEVNAME_SIZE];
-
-	bdevname(journal->j_dev, name);
-	journal->j_proc_entry = proc_mkdir(name, proc_jbd2_stats);
+	journal->j_proc_entry = proc_mkdir(journal->j_devname, proc_jbd2_stats);
 	if (journal->j_proc_entry) {
 		proc_create_data("history", S_IRUGO, journal->j_proc_entry,
 				 &jbd2_seq_history_fops, journal);
@@ -915,12 +908,9 @@ static void jbd2_stats_proc_init(journal_t *journal)
 
 static void jbd2_stats_proc_exit(journal_t *journal)
 {
-	char name[BDEVNAME_SIZE];
-
-	bdevname(journal->j_dev, name);
 	remove_proc_entry("info", journal->j_proc_entry);
 	remove_proc_entry("history", journal->j_proc_entry);
-	remove_proc_entry(name, proc_jbd2_stats);
+	remove_proc_entry(journal->j_devname, proc_jbd2_stats);
 }
 
 static void journal_init_stats(journal_t *journal)
@@ -1018,6 +1008,7 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev,
 {
 	journal_t *journal = journal_init_common();
 	struct buffer_head *bh;
+	char *p;
 	int n;
 
 	if (!journal)
@@ -1039,6 +1030,10 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev,
 	journal->j_fs_dev = fs_dev;
 	journal->j_blk_offset = start;
 	journal->j_maxlen = len;
+	bdevname(journal->j_dev, journal->j_devname);
+	p = journal->j_devname;
+	while ((p = strchr(p, '/')))
+		*p = '!';
 	jbd2_stats_proc_init(journal);
 
 	bh = __getblk(journal->j_dev, start, journal->j_blocksize);
@@ -1061,6 +1056,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode)
 {
 	struct buffer_head *bh;
 	journal_t *journal = journal_init_common();
+	char *p;
 	int err;
 	int n;
 	unsigned long long blocknr;
@@ -1070,6 +1066,12 @@ journal_t * jbd2_journal_init_inode (struct inode *inode)
 
 	journal->j_dev = journal->j_fs_dev = inode->i_sb->s_bdev;
 	journal->j_inode = inode;
+	bdevname(journal->j_dev, journal->j_devname);
+	p = journal->j_devname;
+	while ((p = strchr(p, '/')))
+		*p = '!';
+	p = journal->j_devname + strlen(journal->j_devname);
+	sprintf(p, ":%lu", journal->j_inode->i_ino);
 	jbd_debug(1,
 		  "journal %p: inode %s/%ld, size %Ld, bits %d, blksize %ld\n",
 		  journal, inode->i_sb->s_id, inode->i_ino,
@@ -1253,6 +1255,22 @@ void jbd2_journal_update_superblock(journal_t *journal, int wait)
 		goto out;
 	}
 
+	if (buffer_write_io_error(bh)) {
+		/*
+		 * Oh, dear.  A previous attempt to write the journal
+		 * superblock failed.  This could happen because the
+		 * USB device was yanked out.  Or it could happen to
+		 * be a transient write error and maybe the block will
+		 * be remapped.  Nothing we can do but to retry the
+		 * write and hope for the best.
+		 */
+		printk(KERN_ERR "JBD2: previous I/O error detected "
+		       "for journal superblock update for %s.\n",
+		       journal->j_devname);
+		clear_buffer_write_io_error(bh);
+		set_buffer_uptodate(bh);
+	}
+
 	spin_lock(&journal->j_state_lock);
 	jbd_debug(1,"JBD: updating superblock (start %ld, seq %d, errno %d)\n",
 		  journal->j_tail, journal->j_tail_sequence, journal->j_errno);
@@ -1264,9 +1282,16 @@ void jbd2_journal_update_superblock(journal_t *journal, int wait)
 
 	BUFFER_TRACE(bh, "marking dirty");
 	mark_buffer_dirty(bh);
-	if (wait)
+	if (wait) {
 		sync_dirty_buffer(bh);
-	else
+		if (buffer_write_io_error(bh)) {
+			printk(KERN_ERR "JBD2: I/O error detected "
+			       "when updating journal superblock for %s.\n",
+			       journal->j_devname);
+			clear_buffer_write_io_error(bh);
+			set_buffer_uptodate(bh);
+		}
+	} else
 		ll_rw_block(SWRITE, 1, &bh);
 
 out:
@@ -1426,9 +1451,12 @@ recovery_error:
  *
  * Release a journal_t structure once it is no longer in use by the
  * journaled object.
+ * Return <0 if we couldn't clean up the journal.
  */
-void jbd2_journal_destroy(journal_t *journal)
+int jbd2_journal_destroy(journal_t *journal)
 {
+	int err = 0;
+
 	/* Wait for the commit thread to wake up and die. */
 	journal_kill_thread(journal);
 
@@ -1451,11 +1479,16 @@ void jbd2_journal_destroy(journal_t *journal)
 	J_ASSERT(journal->j_checkpoint_transactions == NULL);
 	spin_unlock(&journal->j_list_lock);
 
-	/* We can now mark the journal as empty. */
-	journal->j_tail = 0;
-	journal->j_tail_sequence = ++journal->j_transaction_sequence;
 	if (journal->j_sb_buffer) {
-		jbd2_journal_update_superblock(journal, 1);
+		if (!is_journal_aborted(journal)) {
+			/* We can now mark the journal as empty. */
+			journal->j_tail = 0;
+			journal->j_tail_sequence =
+				++journal->j_transaction_sequence;
+			jbd2_journal_update_superblock(journal, 1);
+		} else {
+			err = -EIO;
+		}
 		brelse(journal->j_sb_buffer);
 	}
 
@@ -1467,6 +1500,8 @@ void jbd2_journal_destroy(journal_t *journal)
 		jbd2_journal_destroy_revoke(journal);
 	kfree(journal->j_wbuf);
 	kfree(journal);
+
+	return err;
 }
 
 
@@ -1692,10 +1727,16 @@ int jbd2_journal_flush(journal_t *journal)
 	spin_lock(&journal->j_list_lock);
 	while (!err && journal->j_checkpoint_transactions != NULL) {
 		spin_unlock(&journal->j_list_lock);
+		mutex_lock(&journal->j_checkpoint_mutex);
 		err = jbd2_log_do_checkpoint(journal);
+		mutex_unlock(&journal->j_checkpoint_mutex);
 		spin_lock(&journal->j_list_lock);
 	}
 	spin_unlock(&journal->j_list_lock);
+
+	if (is_journal_aborted(journal))
+		return -EIO;
+
 	jbd2_cleanup_journal_tail(journal);
 
 	/* Finally, mark the journal as really needing no recovery.
@@ -1717,7 +1758,7 @@ int jbd2_journal_flush(journal_t *journal)
 	J_ASSERT(journal->j_head == journal->j_tail);
 	J_ASSERT(journal->j_tail_sequence == journal->j_transaction_sequence);
 	spin_unlock(&journal->j_state_lock);
-	return err;
+	return 0;
 }
 
 /**
@@ -1761,23 +1802,6 @@ int jbd2_journal_wipe(journal_t *journal, int write)
 }
 
 /*
- * journal_dev_name: format a character string to describe on what
- * device this journal is present.
- */
-
-static const char *journal_dev_name(journal_t *journal, char *buffer)
-{
-	struct block_device *bdev;
-
-	if (journal->j_inode)
-		bdev = journal->j_inode->i_sb->s_bdev;
-	else
-		bdev = journal->j_dev;
-
-	return bdevname(bdev, buffer);
-}
-
-/*
  * Journal abort has very specific semantics, which we describe
  * for journal abort.
  *
@@ -1793,13 +1817,12 @@ static const char *journal_dev_name(journal_t *journal, char *buffer)
 void __jbd2_journal_abort_hard(journal_t *journal)
 {
 	transaction_t *transaction;
-	char b[BDEVNAME_SIZE];
 
 	if (journal->j_flags & JBD2_ABORT)
 		return;
 
 	printk(KERN_ERR "Aborting journal on device %s.\n",
-		journal_dev_name(journal, b));
+	       journal->j_devname);
 
 	spin_lock(&journal->j_state_lock);
 	journal->j_flags |= JBD2_ABORT;
diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c
index 058f50f65b76..73063285b13f 100644
--- a/fs/jbd2/recovery.c
+++ b/fs/jbd2/recovery.c
@@ -225,7 +225,7 @@ do {									\
  */
 int jbd2_journal_recover(journal_t *journal)
 {
-	int			err;
+	int			err, err2;
 	journal_superblock_t *	sb;
 
 	struct recovery_info	info;
@@ -263,7 +263,10 @@ int jbd2_journal_recover(journal_t *journal)
 	journal->j_transaction_sequence = ++info.end_transaction;
 
 	jbd2_journal_clear_revoke(journal);
-	sync_blockdev(journal->j_fs_dev);
+	err2 = sync_blockdev(journal->j_fs_dev);
+	if (!err)
+		err = err2;
+
 	return err;
 }
 
diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index 10bfb466e068..29ff57ec5d1f 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -990,15 +990,6 @@ out:
 }
 
 /*
- * This is only valid for leaf nodes, which are the only ones that can
- * have empty extents anyway.
- */
-static inline int ocfs2_is_empty_extent(struct ocfs2_extent_rec *rec)
-{
-	return !rec->e_leaf_clusters;
-}
-
-/*
  * This function will discard the rightmost extent record.
  */
 static void ocfs2_shift_records_right(struct ocfs2_extent_list *el)
diff --git a/fs/ocfs2/alloc.h b/fs/ocfs2/alloc.h
index 42ff94bd8011..60cd3d59230c 100644
--- a/fs/ocfs2/alloc.h
+++ b/fs/ocfs2/alloc.h
@@ -146,4 +146,13 @@ static inline unsigned int ocfs2_rec_clusters(struct ocfs2_extent_list *el,
 		return le16_to_cpu(rec->e_leaf_clusters);
 }
 
+/*
+ * This is only valid for leaf nodes, which are the only ones that can
+ * have empty extents anyway.
+ */
+static inline int ocfs2_is_empty_extent(struct ocfs2_extent_rec *rec)
+{
+	return !rec->e_leaf_clusters;
+}
+
 #endif /* OCFS2_ALLOC_H */
diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c
index c58668a326fe..aed268e80b49 100644
--- a/fs/ocfs2/extent_map.c
+++ b/fs/ocfs2/extent_map.c
@@ -25,6 +25,7 @@
 #include <linux/fs.h>
 #include <linux/init.h>
 #include <linux/types.h>
+#include <linux/fiemap.h>
 
 #define MLOG_MASK_PREFIX ML_EXTENT_MAP
 #include <cluster/masklog.h>
@@ -32,6 +33,7 @@
 #include "ocfs2.h"
 
 #include "alloc.h"
+#include "dlmglue.h"
 #include "extent_map.h"
 #include "inode.h"
 #include "super.h"
@@ -282,6 +284,51 @@ out:
 		kfree(new_emi);
 }
 
+static int ocfs2_last_eb_is_empty(struct inode *inode,
+				  struct ocfs2_dinode *di)
+{
+	int ret, next_free;
+	u64 last_eb_blk = le64_to_cpu(di->i_last_eb_blk);
+	struct buffer_head *eb_bh = NULL;
+	struct ocfs2_extent_block *eb;
+	struct ocfs2_extent_list *el;
+
+	ret = ocfs2_read_block(OCFS2_SB(inode->i_sb), last_eb_blk,
+			       &eb_bh, OCFS2_BH_CACHED, inode);
+	if (ret) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+	eb = (struct ocfs2_extent_block *) eb_bh->b_data;
+	el = &eb->h_list;
+
+	if (!OCFS2_IS_VALID_EXTENT_BLOCK(eb)) {
+		ret = -EROFS;
+		OCFS2_RO_ON_INVALID_EXTENT_BLOCK(inode->i_sb, eb);
+		goto out;
+	}
+
+	if (el->l_tree_depth) {
+		ocfs2_error(inode->i_sb,
+			    "Inode %lu has non zero tree depth in "
+			    "leaf block %llu\n", inode->i_ino,
+			    (unsigned long long)eb_bh->b_blocknr);
+		ret = -EROFS;
+		goto out;
+	}
+
+	next_free = le16_to_cpu(el->l_next_free_rec);
+
+	if (next_free == 0 ||
+	    (next_free == 1 && ocfs2_is_empty_extent(&el->l_recs[0])))
+		ret = 1;
+
+out:
+	brelse(eb_bh);
+	return ret;
+}
+
 /*
  * Return the 1st index within el which contains an extent start
  * larger than v_cluster.
@@ -373,42 +420,28 @@ out:
 	return ret;
 }
 
-int ocfs2_get_clusters(struct inode *inode, u32 v_cluster,
-		       u32 *p_cluster, u32 *num_clusters,
-		       unsigned int *extent_flags)
+static int ocfs2_get_clusters_nocache(struct inode *inode,
+				      struct buffer_head *di_bh,
+				      u32 v_cluster, unsigned int *hole_len,
+				      struct ocfs2_extent_rec *ret_rec,
+				      unsigned int *is_last)
 {
-	int ret, i;
-	unsigned int flags = 0;
-	struct buffer_head *di_bh = NULL;
-	struct buffer_head *eb_bh = NULL;
+	int i, ret, tree_height, len;
 	struct ocfs2_dinode *di;
-	struct ocfs2_extent_block *eb;
+	struct ocfs2_extent_block *uninitialized_var(eb);
 	struct ocfs2_extent_list *el;
 	struct ocfs2_extent_rec *rec;
-	u32 coff;
-
-	if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) {
-		ret = -ERANGE;
-		mlog_errno(ret);
-		goto out;
-	}
-
-	ret = ocfs2_extent_map_lookup(inode, v_cluster, p_cluster,
-				      num_clusters, extent_flags);
-	if (ret == 0)
-		goto out;
+	struct buffer_head *eb_bh = NULL;
 
-	ret = ocfs2_read_block(OCFS2_SB(inode->i_sb), OCFS2_I(inode)->ip_blkno,
-			       &di_bh, OCFS2_BH_CACHED, inode);
-	if (ret) {
-		mlog_errno(ret);
-		goto out;
-	}
+	memset(ret_rec, 0, sizeof(*ret_rec));
+	if (is_last)
+		*is_last = 0;
 
 	di = (struct ocfs2_dinode *) di_bh->b_data;
 	el = &di->id2.i_list;
+	tree_height = le16_to_cpu(el->l_tree_depth);
 
-	if (el->l_tree_depth) {
+	if (tree_height > 0) {
 		ret = ocfs2_find_leaf(inode, el, v_cluster, &eb_bh);
 		if (ret) {
 			mlog_errno(ret);
@@ -431,46 +464,143 @@ int ocfs2_get_clusters(struct inode *inode, u32 v_cluster,
 	i = ocfs2_search_extent_list(el, v_cluster);
 	if (i == -1) {
 		/*
-		 * A hole was found. Return some canned values that
-		 * callers can key on. If asked for, num_clusters will
-		 * be populated with the size of the hole.
+		 * Holes can be larger than the maximum size of an
+		 * extent, so we return their lengths in a seperate
+		 * field.
 		 */
-		*p_cluster = 0;
-		if (num_clusters) {
+		if (hole_len) {
 			ret = ocfs2_figure_hole_clusters(inode, el, eb_bh,
-							 v_cluster,
-							 num_clusters);
+							 v_cluster, &len);
 			if (ret) {
 				mlog_errno(ret);
 				goto out;
 			}
+
+			*hole_len = len;
 		}
-	} else {
-		rec = &el->l_recs[i];
+		goto out_hole;
+	}
 
-		BUG_ON(v_cluster < le32_to_cpu(rec->e_cpos));
+	rec = &el->l_recs[i];
 
-		if (!rec->e_blkno) {
-			ocfs2_error(inode->i_sb, "Inode %lu has bad extent "
-				    "record (%u, %u, 0)", inode->i_ino,
-				    le32_to_cpu(rec->e_cpos),
-				    ocfs2_rec_clusters(el, rec));
-			ret = -EROFS;
-			goto out;
+	BUG_ON(v_cluster < le32_to_cpu(rec->e_cpos));
+
+	if (!rec->e_blkno) {
+		ocfs2_error(inode->i_sb, "Inode %lu has bad extent "
+			    "record (%u, %u, 0)", inode->i_ino,
+			    le32_to_cpu(rec->e_cpos),
+			    ocfs2_rec_clusters(el, rec));
+		ret = -EROFS;
+		goto out;
+	}
+
+	*ret_rec = *rec;
+
+	/*
+	 * Checking for last extent is potentially expensive - we
+	 * might have to look at the next leaf over to see if it's
+	 * empty.
+	 *
+	 * The first two checks are to see whether the caller even
+	 * cares for this information, and if the extent is at least
+	 * the last in it's list.
+	 *
+	 * If those hold true, then the extent is last if any of the
+	 * additional conditions hold true:
+	 *  - Extent list is in-inode
+	 *  - Extent list is right-most
+	 *  - Extent list is 2nd to rightmost, with empty right-most
+	 */
+	if (is_last) {
+		if (i == (le16_to_cpu(el->l_next_free_rec) - 1)) {
+			if (tree_height == 0)
+				*is_last = 1;
+			else if (eb->h_blkno == di->i_last_eb_blk)
+				*is_last = 1;
+			else if (eb->h_next_leaf_blk == di->i_last_eb_blk) {
+				ret = ocfs2_last_eb_is_empty(inode, di);
+				if (ret < 0) {
+					mlog_errno(ret);
+					goto out;
+				}
+				if (ret == 1)
+					*is_last = 1;
+			}
 		}
+	}
+
+out_hole:
+	ret = 0;
+out:
+	brelse(eb_bh);
+	return ret;
+}
+
+static void ocfs2_relative_extent_offsets(struct super_block *sb,
+					  u32 v_cluster,
+					  struct ocfs2_extent_rec *rec,
+					  u32 *p_cluster, u32 *num_clusters)
+
+{
+	u32 coff = v_cluster - le32_to_cpu(rec->e_cpos);
+
+	*p_cluster = ocfs2_blocks_to_clusters(sb, le64_to_cpu(rec->e_blkno));
+	*p_cluster = *p_cluster + coff;
+
+	if (num_clusters)
+		*num_clusters = le16_to_cpu(rec->e_leaf_clusters) - coff;
+}
+
+int ocfs2_get_clusters(struct inode *inode, u32 v_cluster,
+		       u32 *p_cluster, u32 *num_clusters,
+		       unsigned int *extent_flags)
+{
+	int ret;
+	unsigned int uninitialized_var(hole_len), flags = 0;
+	struct buffer_head *di_bh = NULL;
+	struct ocfs2_extent_rec rec;
 
-		coff = v_cluster - le32_to_cpu(rec->e_cpos);
+	if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) {
+		ret = -ERANGE;
+		mlog_errno(ret);
+		goto out;
+	}
 
-		*p_cluster = ocfs2_blocks_to_clusters(inode->i_sb,
-						    le64_to_cpu(rec->e_blkno));
-		*p_cluster = *p_cluster + coff;
+	ret = ocfs2_extent_map_lookup(inode, v_cluster, p_cluster,
+				      num_clusters, extent_flags);
+	if (ret == 0)
+		goto out;
 
-		if (num_clusters)
-			*num_clusters = ocfs2_rec_clusters(el, rec) - coff;
+	ret = ocfs2_read_block(OCFS2_SB(inode->i_sb), OCFS2_I(inode)->ip_blkno,
+			       &di_bh, OCFS2_BH_CACHED, inode);
+	if (ret) {
+		mlog_errno(ret);
+		goto out;
+	}
 
-		flags = rec->e_flags;
+	ret = ocfs2_get_clusters_nocache(inode, di_bh, v_cluster, &hole_len,
+					 &rec, NULL);
+	if (ret) {
+		mlog_errno(ret);
+		goto out;
+	}
 
-		ocfs2_extent_map_insert_rec(inode, rec);
+	if (rec.e_blkno == 0ULL) {
+		/*
+		 * A hole was found. Return some canned values that
+		 * callers can key on. If asked for, num_clusters will
+		 * be populated with the size of the hole.
+		 */
+		*p_cluster = 0;
+		if (num_clusters) {
+			*num_clusters = hole_len;
+		}
+	} else {
+		ocfs2_relative_extent_offsets(inode->i_sb, v_cluster, &rec,
+					      p_cluster, num_clusters);
+		flags = rec.e_flags;
+
+		ocfs2_extent_map_insert_rec(inode, &rec);
 	}
 
 	if (extent_flags)
@@ -478,7 +608,6 @@ int ocfs2_get_clusters(struct inode *inode, u32 v_cluster,
 
 out:
 	brelse(di_bh);
-	brelse(eb_bh);
 	return ret;
 }
 
@@ -521,3 +650,114 @@ int ocfs2_extent_map_get_blocks(struct inode *inode, u64 v_blkno, u64 *p_blkno,
 out:
 	return ret;
 }
+
+static int ocfs2_fiemap_inline(struct inode *inode, struct buffer_head *di_bh,
+			       struct fiemap_extent_info *fieinfo,
+			       u64 map_start)
+{
+	int ret;
+	unsigned int id_count;
+	struct ocfs2_dinode *di;
+	u64 phys;
+	u32 flags = FIEMAP_EXTENT_DATA_INLINE|FIEMAP_EXTENT_LAST;
+	struct ocfs2_inode_info *oi = OCFS2_I(inode);
+
+	di = (struct ocfs2_dinode *)di_bh->b_data;
+	id_count = le16_to_cpu(di->id2.i_data.id_count);
+
+	if (map_start < id_count) {
+		phys = oi->ip_blkno << inode->i_sb->s_blocksize_bits;
+		phys += offsetof(struct ocfs2_dinode, id2.i_data.id_data);
+
+		ret = fiemap_fill_next_extent(fieinfo, 0, phys, id_count,
+					      flags);
+		if (ret < 0)
+			return ret;
+	}
+
+	return 0;
+}
+
+#define OCFS2_FIEMAP_FLAGS	(FIEMAP_FLAG_SYNC)
+
+int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+		 u64 map_start, u64 map_len)
+{
+	int ret, is_last;
+	u32 mapping_end, cpos;
+	unsigned int hole_size;
+	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+	u64 len_bytes, phys_bytes, virt_bytes;
+	struct buffer_head *di_bh = NULL;
+	struct ocfs2_extent_rec rec;
+
+	ret = fiemap_check_flags(fieinfo, OCFS2_FIEMAP_FLAGS);
+	if (ret)
+		return ret;
+
+	ret = ocfs2_inode_lock(inode, &di_bh, 0);
+	if (ret) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+	down_read(&OCFS2_I(inode)->ip_alloc_sem);
+
+	/*
+	 * Handle inline-data separately.
+	 */
+	if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) {
+		ret = ocfs2_fiemap_inline(inode, di_bh, fieinfo, map_start);
+		goto out_unlock;
+	}
+
+	cpos = map_start >> osb->s_clustersize_bits;
+	mapping_end = ocfs2_clusters_for_bytes(inode->i_sb,
+					       map_start + map_len);
+	mapping_end -= cpos;
+	is_last = 0;
+	while (cpos < mapping_end && !is_last) {
+		u32 fe_flags;
+
+		ret = ocfs2_get_clusters_nocache(inode, di_bh, cpos,
+						 &hole_size, &rec, &is_last);
+		if (ret) {
+			mlog_errno(ret);
+			goto out;
+		}
+
+		if (rec.e_blkno == 0ULL) {
+			cpos += hole_size;
+			continue;
+		}
+
+		fe_flags = 0;
+		if (rec.e_flags & OCFS2_EXT_UNWRITTEN)
+			fe_flags |= FIEMAP_EXTENT_UNWRITTEN;
+		if (is_last)
+			fe_flags |= FIEMAP_EXTENT_LAST;
+		len_bytes = (u64)le16_to_cpu(rec.e_leaf_clusters) << osb->s_clustersize_bits;
+		phys_bytes = le64_to_cpu(rec.e_blkno) << osb->sb->s_blocksize_bits;
+		virt_bytes = (u64)le32_to_cpu(rec.e_cpos) << osb->s_clustersize_bits;
+
+		ret = fiemap_fill_next_extent(fieinfo, virt_bytes, phys_bytes,
+					      len_bytes, fe_flags);
+		if (ret)
+			break;
+
+		cpos = le32_to_cpu(rec.e_cpos)+ le16_to_cpu(rec.e_leaf_clusters);
+	}
+
+	if (ret > 0)
+		ret = 0;
+
+out_unlock:
+	brelse(di_bh);
+
+	up_read(&OCFS2_I(inode)->ip_alloc_sem);
+
+	ocfs2_inode_unlock(inode, 0);
+out:
+
+	return ret;
+}
diff --git a/fs/ocfs2/extent_map.h b/fs/ocfs2/extent_map.h
index de91e3e41a22..1b97490e1ea8 100644
--- a/fs/ocfs2/extent_map.h
+++ b/fs/ocfs2/extent_map.h
@@ -50,4 +50,7 @@ int ocfs2_get_clusters(struct inode *inode, u32 v_cluster, u32 *p_cluster,
 int ocfs2_extent_map_get_blocks(struct inode *inode, u64 v_blkno, u64 *p_blkno,
 				u64 *ret_count, unsigned int *extent_flags);
 
+int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+		 u64 map_start, u64 map_len);
+
 #endif  /* _EXTENT_MAP_H */
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index ec2ed15c3daa..ed38796052d2 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -2228,6 +2228,7 @@ const struct inode_operations ocfs2_file_iops = {
 	.getattr	= ocfs2_getattr,
 	.permission	= ocfs2_permission,
 	.fallocate	= ocfs2_fallocate,
+	.fiemap		= ocfs2_fiemap,
 };
 
 const struct inode_operations ocfs2_special_file_iops = {
diff --git a/fs/open.c b/fs/open.c
index 07da9359481c..5596049863bf 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -1141,8 +1141,7 @@ EXPORT_SYMBOL(sys_close);
 asmlinkage long sys_vhangup(void)
 {
 	if (capable(CAP_SYS_TTY_CONFIG)) {
-		/* XXX: this needs locking */
-		tty_vhangup(current->signal->tty);
+		tty_vhangup_self();
 		return 0;
 	}
 	return -EPERM;
diff --git a/fs/proc/Kconfig b/fs/proc/Kconfig
index 73cd7a418f06..50f8f0600f06 100644
--- a/fs/proc/Kconfig
+++ b/fs/proc/Kconfig
@@ -57,3 +57,13 @@ config PROC_SYSCTL
 	  As it is generally a good thing, you should say Y here unless
 	  building a kernel for install/rescue disks or your system is very
 	  limited in memory.
+
+config PROC_PAGE_MONITOR
+ 	default y
+	depends on PROC_FS && MMU
+	bool "Enable /proc page monitoring" if EMBEDDED
+ 	help
+	  Various /proc files exist to monitor process memory utilization:
+	  /proc/pid/smaps, /proc/pid/clear_refs, /proc/pid/pagemap,
+	  /proc/kpagecount, and /proc/kpageflags. Disabling these
+          interfaces will reduce the size of the kernel by approximately 4kb.
diff --git a/fs/proc/array.c b/fs/proc/array.c
index 71c9be59c9c2..f4bc0e789539 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -86,11 +86,6 @@
 #include <asm/processor.h>
 #include "internal.h"
 
-/* Gcc optimizes away "strlen(x)" for constant x */
-#define ADDBUF(buffer, string) \
-do { memcpy(buffer, string, strlen(string)); \
-     buffer += strlen(string); } while (0)
-
 static inline void task_name(struct seq_file *m, struct task_struct *p)
 {
 	int i;
@@ -261,7 +256,6 @@ static inline void task_sig(struct seq_file *m, struct task_struct *p)
 	sigemptyset(&ignored);
 	sigemptyset(&caught);
 
-	rcu_read_lock();
 	if (lock_task_sighand(p, &flags)) {
 		pending = p->pending.signal;
 		shpending = p->signal->shared_pending.signal;
@@ -272,7 +266,6 @@ static inline void task_sig(struct seq_file *m, struct task_struct *p)
 		qlim = p->signal->rlim[RLIMIT_SIGPENDING].rlim_cur;
 		unlock_task_sighand(p, &flags);
 	}
-	rcu_read_unlock();
 
 	seq_printf(m, "Threads:\t%d\n", num_threads);
 	seq_printf(m, "SigQ:\t%lu/%lu\n", qsize, qlim);
diff --git a/fs/proc/base.c b/fs/proc/base.c
index a28840b11b89..b5918ae8ca79 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -148,9 +148,6 @@ static unsigned int pid_entry_count_dirs(const struct pid_entry *entries,
 	return count;
 }
 
-int maps_protect;
-EXPORT_SYMBOL(maps_protect);
-
 static struct fs_struct *get_fs_struct(struct task_struct *task)
 {
 	struct fs_struct *fs;
@@ -164,7 +161,6 @@ static struct fs_struct *get_fs_struct(struct task_struct *task)
 
 static int get_nr_threads(struct task_struct *tsk)
 {
-	/* Must be called with the rcu_read_lock held */
 	unsigned long flags;
 	int count = 0;
 
@@ -471,14 +467,10 @@ static int proc_pid_limits(struct task_struct *task, char *buffer)
 
 	struct rlimit rlim[RLIM_NLIMITS];
 
-	rcu_read_lock();
-	if (!lock_task_sighand(task,&flags)) {
-		rcu_read_unlock();
+	if (!lock_task_sighand(task, &flags))
 		return 0;
-	}
 	memcpy(rlim, task->signal->rlim, sizeof(struct rlimit) * RLIM_NLIMITS);
 	unlock_task_sighand(task, &flags);
-	rcu_read_unlock();
 
 	/*
 	 * print the file header
@@ -2443,6 +2435,13 @@ static int proc_tgid_io_accounting(struct task_struct *task, char *buffer)
 }
 #endif /* CONFIG_TASK_IO_ACCOUNTING */
 
+static int proc_pid_personality(struct seq_file *m, struct pid_namespace *ns,
+				struct pid *pid, struct task_struct *task)
+{
+	seq_printf(m, "%08x\n", task->personality);
+	return 0;
+}
+
 /*
  * Thread groups
  */
@@ -2459,6 +2458,7 @@ static const struct pid_entry tgid_base_stuff[] = {
 	REG("environ",    S_IRUSR, environ),
 	INF("auxv",       S_IRUSR, pid_auxv),
 	ONE("status",     S_IRUGO, pid_status),
+	ONE("personality", S_IRUSR, pid_personality),
 	INF("limits",	  S_IRUSR, pid_limits),
 #ifdef CONFIG_SCHED_DEBUG
 	REG("sched",      S_IRUGO|S_IWUSR, pid_sched),
@@ -2794,6 +2794,7 @@ static const struct pid_entry tid_base_stuff[] = {
 	REG("environ",   S_IRUSR, environ),
 	INF("auxv",      S_IRUSR, pid_auxv),
 	ONE("status",    S_IRUGO, pid_status),
+	ONE("personality", S_IRUSR, pid_personality),
 	INF("limits",	 S_IRUSR, pid_limits),
 #ifdef CONFIG_SCHED_DEBUG
 	REG("sched",     S_IRUGO|S_IWUSR, pid_sched),
@@ -3088,9 +3089,7 @@ static int proc_task_getattr(struct vfsmount *mnt, struct dentry *dentry, struct
 	generic_fillattr(inode, stat);
 
 	if (p) {
-		rcu_read_lock();
 		stat->nlink += get_nr_threads(p);
-		rcu_read_unlock();
 		put_task_struct(p);
 	}
 
diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index 8bb03f056c28..c6b4fa7e3b49 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -342,7 +342,7 @@ static int proc_reg_open(struct inode *inode, struct file *file)
 	if (!pde->proc_fops) {
 		spin_unlock(&pde->pde_unload_lock);
 		kfree(pdeo);
-		return rv;
+		return -EINVAL;
 	}
 	pde->pde_users++;
 	open = pde->proc_fops->open;
diff --git a/fs/proc/internal.h b/fs/proc/internal.h
index 442202314d53..3bfb7b8747b3 100644
--- a/fs/proc/internal.h
+++ b/fs/proc/internal.h
@@ -45,8 +45,6 @@ do {						\
 extern int nommu_vma_show(struct seq_file *, struct vm_area_struct *);
 #endif
 
-extern int maps_protect;
-
 extern int proc_tid_stat(struct seq_file *m, struct pid_namespace *ns,
 				struct pid *pid, struct task_struct *task);
 extern int proc_tgid_stat(struct seq_file *m, struct pid_namespace *ns,
diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c
index 29e20c6b1f7f..66c1ab87656c 100644
--- a/fs/proc/proc_misc.c
+++ b/fs/proc/proc_misc.c
@@ -68,7 +68,6 @@
 extern int get_hardware_list(char *);
 extern int get_stram_list(char *);
 extern int get_exec_domain_list(char *);
-extern int get_dma_list(char *);
 
 static int proc_calc_metrics(char *page, char **start, off_t off,
 				 int count, int *eof, int len)
diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
index f9a8b892718f..945a81043ba2 100644
--- a/fs/proc/proc_sysctl.c
+++ b/fs/proc/proc_sysctl.c
@@ -66,7 +66,7 @@ static struct ctl_table *find_in_table(struct ctl_table *p, struct qstr *name)
 	return NULL;
 }
 
-struct ctl_table_header *grab_header(struct inode *inode)
+static struct ctl_table_header *grab_header(struct inode *inode)
 {
 	if (PROC_I(inode)->sysctl)
 		return sysctl_head_grab(PROC_I(inode)->sysctl);
@@ -395,10 +395,10 @@ static struct dentry_operations proc_sys_dentry_operations = {
 	.d_compare	= proc_sys_compare,
 };
 
-static struct proc_dir_entry *proc_sys_root;
-
 int proc_sys_init(void)
 {
+	struct proc_dir_entry *proc_sys_root;
+
 	proc_sys_root = proc_mkdir("sys", NULL);
 	proc_sys_root->proc_iops = &proc_sys_dir_operations;
 	proc_sys_root->proc_fops = &proc_sys_dir_file_operations;
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 73d1891ee625..4806830ea2a1 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -210,9 +210,6 @@ static int show_map(struct seq_file *m, void *v)
 	dev_t dev = 0;
 	int len;
 
-	if (maps_protect && !ptrace_may_access(task, PTRACE_MODE_READ))
-		return -EACCES;
-
 	if (file) {
 		struct inode *inode = vma->vm_file->f_path.dentry->d_inode;
 		dev = inode->i_sb->s_dev;
@@ -742,22 +739,11 @@ const struct file_operations proc_pagemap_operations = {
 #ifdef CONFIG_NUMA
 extern int show_numa_map(struct seq_file *m, void *v);
 
-static int show_numa_map_checked(struct seq_file *m, void *v)
-{
-	struct proc_maps_private *priv = m->private;
-	struct task_struct *task = priv->task;
-
-	if (maps_protect && !ptrace_may_access(task, PTRACE_MODE_READ))
-		return -EACCES;
-
-	return show_numa_map(m, v);
-}
-
 static const struct seq_operations proc_pid_numa_maps_op = {
         .start  = m_start,
         .next   = m_next,
         .stop   = m_stop,
-        .show   = show_numa_map_checked
+        .show   = show_numa_map,
 };
 
 static int numa_maps_open(struct inode *inode, struct file *file)
diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c
index 5d84e7121df8..219bd79ea894 100644
--- a/fs/proc/task_nommu.c
+++ b/fs/proc/task_nommu.c
@@ -110,11 +110,6 @@ int task_statm(struct mm_struct *mm, int *shared, int *text,
 static int show_map(struct seq_file *m, void *_vml)
 {
 	struct vm_list_struct *vml = _vml;
-	struct proc_maps_private *priv = m->private;
-	struct task_struct *task = priv->task;
-
-	if (maps_protect && !ptrace_may_access(task, PTRACE_MODE_READ))
-		return -EACCES;
 
 	return nommu_vma_show(m, vml->vma);
 }
diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c
index 9ac0f5e064e0..841368b87a29 100644
--- a/fs/proc/vmcore.c
+++ b/fs/proc/vmcore.c
@@ -165,14 +165,8 @@ static ssize_t read_vmcore(struct file *file, char __user *buffer,
 	return acc;
 }
 
-static int open_vmcore(struct inode *inode, struct file *filp)
-{
-	return 0;
-}
-
 const struct file_operations proc_vmcore_operations = {
 	.read		= read_vmcore,
-	.open		= open_vmcore,
 };
 
 static struct vmcore* __init get_new_element(void)
diff --git a/include/asm-cris/a.out.h b/include/asm-cris/a.out.h
deleted file mode 100644
index c82e9f9b75f6..000000000000
--- a/include/asm-cris/a.out.h
+++ /dev/null
@@ -1,26 +0,0 @@
-#ifndef __CRIS_A_OUT_H__
-#define __CRIS_A_OUT_H__
-
-/* we don't support a.out binaries on Linux/CRIS anyway, so this is
- * not really used but still needed because binfmt_elf.c for some reason
- * wants to know about a.out even if there is no interpreter available...
- */
-
-struct exec
-{
-  unsigned long a_info;		/* Use macros N_MAGIC, etc for access */
-  unsigned a_text;		/* length of text, in bytes */
-  unsigned a_data;		/* length of data, in bytes */
-  unsigned a_bss;		/* length of uninitialized data area for file, in bytes */
-  unsigned a_syms;		/* length of symbol table data in file, in bytes */
-  unsigned a_entry;		/* start address */
-  unsigned a_trsize;		/* length of relocation info for text, in bytes */
-  unsigned a_drsize;		/* length of relocation info for data, in bytes */
-};
-
-
-#define N_TRSIZE(a)	((a).a_trsize)
-#define N_DRSIZE(a)	((a).a_drsize)
-#define N_SYMSIZE(a)	((a).a_syms)
-
-#endif
diff --git a/include/asm-generic/siginfo.h b/include/asm-generic/siginfo.h
index 8786e01e0db8..969570167e9e 100644
--- a/include/asm-generic/siginfo.h
+++ b/include/asm-generic/siginfo.h
@@ -199,6 +199,8 @@ typedef struct siginfo {
  */
 #define TRAP_BRKPT	(__SI_FAULT|1)	/* process breakpoint */
 #define TRAP_TRACE	(__SI_FAULT|2)	/* process trace trap */
+#define TRAP_BRANCH     (__SI_FAULT|3)  /* process taken branch trap */
+#define TRAP_HWBKPT     (__SI_FAULT|4)  /* hardware breakpoint/watchpoint */
 #define NSIGTRAP	2
 
 /*
diff --git a/include/asm-generic/statfs.h b/include/asm-generic/statfs.h
index 1d01043e797d..6129d6802149 100644
--- a/include/asm-generic/statfs.h
+++ b/include/asm-generic/statfs.h
@@ -6,33 +6,64 @@
 typedef __kernel_fsid_t	fsid_t;
 #endif
 
+/*
+ * Most 64-bit platforms use 'long', while most 32-bit platforms use '__u32'.
+ * Yes, they differ in signedness as well as size.
+ * Special cases can override it for themselves -- except for S390x, which
+ * is just a little too special for us. And MIPS, which I'm not touching
+ * with a 10' pole.
+ */
+#ifndef __statfs_word
+#if BITS_PER_LONG == 64
+#define __statfs_word long
+#else
+#define __statfs_word __u32
+#endif
+#endif
+
 struct statfs {
-	__u32 f_type;
-	__u32 f_bsize;
-	__u32 f_blocks;
-	__u32 f_bfree;
-	__u32 f_bavail;
-	__u32 f_files;
-	__u32 f_ffree;
+	__statfs_word f_type;
+	__statfs_word f_bsize;
+	__statfs_word f_blocks;
+	__statfs_word f_bfree;
+	__statfs_word f_bavail;
+	__statfs_word f_files;
+	__statfs_word f_ffree;
 	__kernel_fsid_t f_fsid;
-	__u32 f_namelen;
-	__u32 f_frsize;
-	__u32 f_spare[5];
+	__statfs_word f_namelen;
+	__statfs_word f_frsize;
+	__statfs_word f_spare[5];
 };
 
+/*
+ * ARM needs to avoid the 32-bit padding at the end, for consistency
+ * between EABI and OABI 
+ */
+#ifndef ARCH_PACK_STATFS64
+#define ARCH_PACK_STATFS64
+#endif
+
 struct statfs64 {
-	__u32 f_type;
-	__u32 f_bsize;
+	__statfs_word f_type;
+	__statfs_word f_bsize;
 	__u64 f_blocks;
 	__u64 f_bfree;
 	__u64 f_bavail;
 	__u64 f_files;
 	__u64 f_ffree;
 	__kernel_fsid_t f_fsid;
-	__u32 f_namelen;
-	__u32 f_frsize;
-	__u32 f_spare[5];
-};
+	__statfs_word f_namelen;
+	__statfs_word f_frsize;
+	__statfs_word f_spare[5];
+} ARCH_PACK_STATFS64;
+
+/* 
+ * IA64 and x86_64 need to avoid the 32-bit padding at the end,
+ * to be compatible with the i386 ABI
+ */
+#ifndef ARCH_PACK_COMPAT_STATFS64
+#define ARCH_PACK_COMPAT_STATFS64
+#endif
 
 struct compat_statfs64 {
 	__u32 f_type;
@@ -46,6 +77,6 @@ struct compat_statfs64 {
 	__u32 f_namelen;
 	__u32 f_frsize;
 	__u32 f_spare[5];
-};
+} ARCH_PACK_COMPAT_STATFS64;
 
 #endif
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index cb752ba72466..7440a0dceddb 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -385,6 +385,7 @@
 	. = ALIGN(align);						\
 	VMLINUX_SYMBOL(__per_cpu_start) = .;				\
 	.data.percpu  : AT(ADDR(.data.percpu) - LOAD_OFFSET) {		\
+		*(.data.percpu.page_aligned)				\
 		*(.data.percpu)						\
 		*(.data.percpu.shared_aligned)				\
 	}								\
diff --git a/include/asm-m32r/a.out.h b/include/asm-m32r/a.out.h
deleted file mode 100644
index ab150f5c1666..000000000000
--- a/include/asm-m32r/a.out.h
+++ /dev/null
@@ -1,20 +0,0 @@
-#ifndef _ASM_M32R_A_OUT_H
-#define _ASM_M32R_A_OUT_H
-
-struct exec
-{
-  unsigned long a_info;		/* Use macros N_MAGIC, etc for access */
-  unsigned a_text;		/* length of text, in bytes */
-  unsigned a_data;		/* length of data, in bytes */
-  unsigned a_bss;		/* length of uninitialized data area for file, in bytes */
-  unsigned a_syms;		/* length of symbol table data in file, in bytes */
-  unsigned a_entry;		/* start address */
-  unsigned a_trsize;		/* length of relocation info for text, in bytes */
-  unsigned a_drsize;		/* length of relocation info for data, in bytes */
-};
-
-#define N_TRSIZE(a)	((a).a_trsize)
-#define N_DRSIZE(a)	((a).a_drsize)
-#define N_SYMSIZE(a)	((a).a_syms)
-
-#endif /* _ASM_M32R_A_OUT_H */
diff --git a/include/asm-parisc/a.out.h b/include/asm-parisc/a.out.h
deleted file mode 100644
index eb04e34c5bb1..000000000000
--- a/include/asm-parisc/a.out.h
+++ /dev/null
@@ -1,20 +0,0 @@
-#ifndef __PARISC_A_OUT_H__
-#define __PARISC_A_OUT_H__
-
-struct exec
-{
-  unsigned int a_info;		/* Use macros N_MAGIC, etc for access */
-  unsigned a_text;		/* length of text, in bytes */
-  unsigned a_data;		/* length of data, in bytes */
-  unsigned a_bss;		/* length of uninitialized data area for file, in bytes */
-  unsigned a_syms;		/* length of symbol table data in file, in bytes */
-  unsigned a_entry;		/* start address */
-  unsigned a_trsize;		/* length of relocation info for text, in bytes */
-  unsigned a_drsize;		/* length of relocation info for data, in bytes */
-};
-
-#define N_TRSIZE(a)	((a).a_trsize)
-#define N_DRSIZE(a)	((a).a_drsize)
-#define N_SYMSIZE(a)	((a).a_syms)
-
-#endif /* __A_OUT_GNU_H__ */
diff --git a/include/asm-parisc/siginfo.h b/include/asm-parisc/siginfo.h
index d4909f55fe35..d7034728f377 100644
--- a/include/asm-parisc/siginfo.h
+++ b/include/asm-parisc/siginfo.h
@@ -3,11 +3,6 @@
 
 #include <asm-generic/siginfo.h>
 
-/*
- * SIGTRAP si_codes
- */
-#define TRAP_BRANCH	(__SI_FAULT|3)	/* process taken branch trap */
-#define TRAP_HWBKPT	(__SI_FAULT|4)	/* hardware breakpoint or watchpoint */
 #undef NSIGTRAP
 #define NSIGTRAP	4
 
diff --git a/include/asm-parisc/statfs.h b/include/asm-parisc/statfs.h
index 1d2b8130b23d..324bea905dc6 100644
--- a/include/asm-parisc/statfs.h
+++ b/include/asm-parisc/statfs.h
@@ -1,58 +1,7 @@
 #ifndef _PARISC_STATFS_H
 #define _PARISC_STATFS_H
 
-#ifndef __KERNEL_STRICT_NAMES
-
-#include <linux/types.h>
-
-typedef __kernel_fsid_t	fsid_t;
-
-#endif
-
-/*
- * It appears that PARISC could be 64 _or_ 32 bit.
- * 64-bit fields must be explicitly 64-bit in statfs64.
- */
-struct statfs {
-	long f_type;
-	long f_bsize;
-	long f_blocks;
-	long f_bfree;
-	long f_bavail;
-	long f_files;
-	long f_ffree;
-	__kernel_fsid_t f_fsid;
-	long f_namelen;
-	long f_frsize;
-	long f_spare[5];
-};
-
-struct statfs64 {
-	long f_type;
-	long f_bsize;
-	__u64 f_blocks;
-	__u64 f_bfree;
-	__u64 f_bavail;
-	__u64 f_files;
-	__u64 f_ffree;
-	__kernel_fsid_t f_fsid;
-	long f_namelen;
-	long f_frsize;
-	long f_spare[5];
-};
-
-struct compat_statfs64 {
-	__u32 f_type;
-	__u32 f_bsize;
-	__u64 f_blocks;
-	__u64 f_bfree;
-	__u64 f_bavail;
-	__u64 f_files;
-	__u64 f_ffree;
-	__kernel_fsid_t f_fsid;
-	__u32 f_namelen;
-	__u32 f_frsize;
-	__u32 f_spare[5];
-};
+#define __statfs_word long
+#include <asm-generic/statfs.h>
 
 #endif
diff --git a/include/asm-x86/apic.h b/include/asm-x86/apic.h
index 65590c9aecd4..d76a0839abe9 100644
--- a/include/asm-x86/apic.h
+++ b/include/asm-x86/apic.h
@@ -9,6 +9,8 @@
 #include <asm/apicdef.h>
 #include <asm/processor.h>
 #include <asm/system.h>
+#include <asm/cpufeature.h>
+#include <asm/msr.h>
 
 #define ARCH_APICTIMER_STOPS_ON_C3	1
 
@@ -47,8 +49,6 @@ extern int disable_apic;
 #ifdef CONFIG_PARAVIRT
 #include <asm/paravirt.h>
 #else
-#define apic_write native_apic_write
-#define apic_read native_apic_read
 #define setup_boot_clock setup_boot_APIC_clock
 #define setup_secondary_clock setup_secondary_APIC_clock
 #endif
@@ -60,7 +60,7 @@ extern u64 xapic_icr_read(void);
 extern void xapic_icr_write(u32, u32);
 extern int setup_profiling_timer(unsigned int);
 
-static inline void native_apic_write(unsigned long reg, u32 v)
+static inline void native_apic_mem_write(u32 reg, u32 v)
 {
 	volatile u32 *addr = (volatile u32 *)(APIC_BASE + reg);
 
@@ -69,15 +69,68 @@ static inline void native_apic_write(unsigned long reg, u32 v)
 		       ASM_OUTPUT2("0" (v), "m" (*addr)));
 }
 
-static inline u32 native_apic_read(unsigned long reg)
+static inline u32 native_apic_mem_read(u32 reg)
 {
 	return *((volatile u32 *)(APIC_BASE + reg));
 }
 
-extern void apic_wait_icr_idle(void);
-extern u32 safe_apic_wait_icr_idle(void);
+static inline void native_apic_msr_write(u32 reg, u32 v)
+{
+	if (reg == APIC_DFR || reg == APIC_ID || reg == APIC_LDR ||
+	    reg == APIC_LVR)
+		return;
+
+	wrmsr(APIC_BASE_MSR + (reg >> 4), v, 0);
+}
+
+static inline u32 native_apic_msr_read(u32 reg)
+{
+	u32 low, high;
+
+	if (reg == APIC_DFR)
+		return -1;
+
+	rdmsr(APIC_BASE_MSR + (reg >> 4), low, high);
+	return low;
+}
+
+#ifndef CONFIG_X86_32
+extern int x2apic, x2apic_preenabled;
+extern void check_x2apic(void);
+extern void enable_x2apic(void);
+extern void enable_IR_x2apic(void);
+extern void x2apic_icr_write(u32 low, u32 id);
+#endif
+
+struct apic_ops {
+	u32 (*read)(u32 reg);
+	void (*write)(u32 reg, u32 v);
+	u64 (*icr_read)(void);
+	void (*icr_write)(u32 low, u32 high);
+	void (*wait_icr_idle)(void);
+	u32 (*safe_wait_icr_idle)(void);
+};
+
+extern struct apic_ops *apic_ops;
+
+#define apic_read (apic_ops->read)
+#define apic_write (apic_ops->write)
+#define apic_icr_read (apic_ops->icr_read)
+#define apic_icr_write (apic_ops->icr_write)
+#define apic_wait_icr_idle (apic_ops->wait_icr_idle)
+#define safe_apic_wait_icr_idle (apic_ops->safe_wait_icr_idle)
+
 extern int get_physical_broadcast(void);
 
+#ifdef CONFIG_X86_64
+static inline void ack_x2APIC_irq(void)
+{
+	/* Docs say use 0 for future compatibility */
+	native_apic_msr_write(APIC_EOI, 0);
+}
+#endif
+
+
 static inline void ack_APIC_irq(void)
 {
 	/*
diff --git a/include/asm-x86/apicdef.h b/include/asm-x86/apicdef.h
index c40687da20fc..b922c85ac91d 100644
--- a/include/asm-x86/apicdef.h
+++ b/include/asm-x86/apicdef.h
@@ -105,6 +105,7 @@
 #define	APIC_TMICT	0x380
 #define	APIC_TMCCT	0x390
 #define	APIC_TDCR	0x3E0
+#define APIC_SELF_IPI	0x3F0
 #define		APIC_TDR_DIV_TMBASE	(1 << 2)
 #define		APIC_TDR_DIV_1		0xB
 #define		APIC_TDR_DIV_2		0x0
@@ -128,6 +129,8 @@
 #define	APIC_EILVT3     0x530
 
 #define APIC_BASE (fix_to_virt(FIX_APIC_BASE))
+#define APIC_BASE_MSR	0x800
+#define X2APIC_ENABLE	(1UL << 10)
 
 #ifdef CONFIG_X86_32
 # define MAX_IO_APICS 64
diff --git a/include/asm-x86/arch_hooks.h b/include/asm-x86/arch_hooks.h
index 72adc3a109cc..de4596b24c23 100644
--- a/include/asm-x86/arch_hooks.h
+++ b/include/asm-x86/arch_hooks.h
@@ -12,8 +12,6 @@
 /* these aren't arch hooks, they are generic routines
  * that can be used by the hooks */
 extern void init_ISA_irqs(void);
-extern void apic_intr_init(void);
-extern void smp_intr_init(void);
 extern irqreturn_t timer_interrupt(int irq, void *dev_id);
 
 /* these are the defined hooks */
diff --git a/include/asm-x86/mach-bigsmp/mach_apic.h b/include/asm-x86/bigsmp/apic.h
index 05362d44a3ee..0a9cd7c5ca0c 100644
--- a/include/asm-x86/mach-bigsmp/mach_apic.h
+++ b/include/asm-x86/bigsmp/apic.h
@@ -1,5 +1,5 @@
-#ifndef ASM_X86__MACH_BIGSMP__MACH_APIC_H
-#define ASM_X86__MACH_BIGSMP__MACH_APIC_H
+#ifndef __ASM_MACH_APIC_H
+#define __ASM_MACH_APIC_H
 
 #define xapic_phys_to_log_apicid(cpu) (per_cpu(x86_bios_cpu_apicid, cpu))
 #define esr_disable (1)
@@ -11,7 +11,7 @@ static inline int apic_id_registered(void)
 
 /* Round robin the irqs amoung the online cpus */
 static inline cpumask_t target_cpus(void)
-{ 
+{
 	static unsigned long cpu = NR_CPUS;
 	do {
 		if (cpu >= NR_CPUS)
@@ -23,7 +23,7 @@ static inline cpumask_t target_cpus(void)
 }
 
 #undef APIC_DEST_LOGICAL
-#define APIC_DEST_LOGICAL 	0
+#define APIC_DEST_LOGICAL	0
 #define TARGET_CPUS		(target_cpus())
 #define APIC_DFR_VALUE		(APIC_DFR_FLAT)
 #define INT_DELIVERY_MODE	(dest_Fixed)
@@ -141,4 +141,4 @@ static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb)
 	return cpuid_apic >> index_msb;
 }
 
-#endif /* ASM_X86__MACH_BIGSMP__MACH_APIC_H */
+#endif /* __ASM_MACH_APIC_H */
diff --git a/include/asm-x86/bigsmp/apicdef.h b/include/asm-x86/bigsmp/apicdef.h
new file mode 100644
index 000000000000..392c3f5ef2fe
--- /dev/null
+++ b/include/asm-x86/bigsmp/apicdef.h
@@ -0,0 +1,13 @@
+#ifndef __ASM_MACH_APICDEF_H
+#define __ASM_MACH_APICDEF_H
+
+#define		APIC_ID_MASK		(0xFF<<24)
+
+static inline unsigned get_apic_id(unsigned long x)
+{
+	return (((x)>>24)&0xFF);
+}
+
+#define		GET_APIC_ID(x)	get_apic_id(x)
+
+#endif
diff --git a/include/asm-x86/mach-summit/mach_ipi.h b/include/asm-x86/bigsmp/ipi.h
index a3b31c528d90..9404c535b7ec 100644
--- a/include/asm-x86/mach-summit/mach_ipi.h
+++ b/include/asm-x86/bigsmp/ipi.h
@@ -1,5 +1,5 @@
-#ifndef ASM_X86__MACH_SUMMIT__MACH_IPI_H
-#define ASM_X86__MACH_SUMMIT__MACH_IPI_H
+#ifndef __ASM_MACH_IPI_H
+#define __ASM_MACH_IPI_H
 
 void send_IPI_mask_sequence(cpumask_t mask, int vector);
 
@@ -22,4 +22,4 @@ static inline void send_IPI_all(int vector)
 	send_IPI_mask(cpu_online_map, vector);
 }
 
-#endif /* ASM_X86__MACH_SUMMIT__MACH_IPI_H */
+#endif /* __ASM_MACH_IPI_H */
diff --git a/include/asm-x86/bios_ebda.h b/include/asm-x86/bios_ebda.h
index ec42ed874591..79b4b88505d7 100644
--- a/include/asm-x86/bios_ebda.h
+++ b/include/asm-x86/bios_ebda.h
@@ -16,4 +16,21 @@ static inline unsigned int get_bios_ebda(void)
 
 void reserve_ebda_region(void);
 
+#ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION
+/*
+ * This is obviously not a great place for this, but we want to be
+ * able to scatter it around anywhere in the kernel.
+ */
+void check_for_bios_corruption(void);
+void start_periodic_check_for_corruption(void);
+#else
+static inline void check_for_bios_corruption(void)
+{
+}
+
+static inline void start_periodic_check_for_corruption(void)
+{
+}
+#endif
+
 #endif /* ASM_X86__BIOS_EBDA_H */
diff --git a/include/asm-x86/boot.h b/include/asm-x86/boot.h
index 825de5dc867c..1d63bd5d5946 100644
--- a/include/asm-x86/boot.h
+++ b/include/asm-x86/boot.h
@@ -2,9 +2,7 @@
 #define ASM_X86__BOOT_H
 
 /* Don't touch these, unless you really know what you're doing. */
-#define DEF_INITSEG	0x9000
 #define DEF_SYSSEG	0x1000
-#define DEF_SETUPSEG	0x9020
 #define DEF_SYSSIZE	0x7F00
 
 /* Internal svga startup constants */
diff --git a/include/asm-x86/bugs.h b/include/asm-x86/bugs.h
index 4761c461d23a..dc604985f2ad 100644
--- a/include/asm-x86/bugs.h
+++ b/include/asm-x86/bugs.h
@@ -2,6 +2,11 @@
 #define ASM_X86__BUGS_H
 
 extern void check_bugs(void);
+
+#if defined(CONFIG_CPU_SUP_INTEL) && defined(CONFIG_X86_32)
 int ppro_with_ram_bug(void);
+#else
+static inline int ppro_with_ram_bug(void) { return 0; }
+#endif
 
 #endif /* ASM_X86__BUGS_H */
diff --git a/include/asm-x86/cpufeature.h b/include/asm-x86/cpufeature.h
index 250fa0cb144b..adfeae6586e1 100644
--- a/include/asm-x86/cpufeature.h
+++ b/include/asm-x86/cpufeature.h
@@ -6,7 +6,13 @@
 
 #include <asm/required-features.h>
 
-#define NCAPINTS	8	/* N 32-bit words worth of info */
+#define NCAPINTS	9	/* N 32-bit words worth of info */
+
+/*
+ * Note: If the comment begins with a quoted string, that string is used
+ * in /proc/cpuinfo instead of the macro name.  If the string is "",
+ * this feature bit is not displayed in /proc/cpuinfo at all.
+ */
 
 /* Intel-defined CPU features, CPUID level 0x00000001 (edx), word 0 */
 #define X86_FEATURE_FPU		(0*32+ 0) /* Onboard FPU */
@@ -14,7 +20,7 @@
 #define X86_FEATURE_DE		(0*32+ 2) /* Debugging Extensions */
 #define X86_FEATURE_PSE		(0*32+ 3) /* Page Size Extensions */
 #define X86_FEATURE_TSC		(0*32+ 4) /* Time Stamp Counter */
-#define X86_FEATURE_MSR		(0*32+ 5) /* Model-Specific Registers, RDMSR, WRMSR */
+#define X86_FEATURE_MSR		(0*32+ 5) /* Model-Specific Registers */
 #define X86_FEATURE_PAE		(0*32+ 6) /* Physical Address Extensions */
 #define X86_FEATURE_MCE		(0*32+ 7) /* Machine Check Architecture */
 #define X86_FEATURE_CX8		(0*32+ 8) /* CMPXCHG8 instruction */
@@ -23,22 +29,23 @@
 #define X86_FEATURE_MTRR	(0*32+12) /* Memory Type Range Registers */
 #define X86_FEATURE_PGE		(0*32+13) /* Page Global Enable */
 #define X86_FEATURE_MCA		(0*32+14) /* Machine Check Architecture */
-#define X86_FEATURE_CMOV	(0*32+15) /* CMOV instruction (FCMOVCC and FCOMI too if FPU present) */
+#define X86_FEATURE_CMOV	(0*32+15) /* CMOV instructions */
+					  /* (plus FCMOVcc, FCOMI with FPU) */
 #define X86_FEATURE_PAT		(0*32+16) /* Page Attribute Table */
 #define X86_FEATURE_PSE36	(0*32+17) /* 36-bit PSEs */
 #define X86_FEATURE_PN		(0*32+18) /* Processor serial number */
-#define X86_FEATURE_CLFLSH	(0*32+19) /* Supports the CLFLUSH instruction */
-#define X86_FEATURE_DS		(0*32+21) /* Debug Store */
+#define X86_FEATURE_CLFLSH	(0*32+19) /* "clflush" CLFLUSH instruction */
+#define X86_FEATURE_DS		(0*32+21) /* "dts" Debug Store */
 #define X86_FEATURE_ACPI	(0*32+22) /* ACPI via MSR */
 #define X86_FEATURE_MMX		(0*32+23) /* Multimedia Extensions */
-#define X86_FEATURE_FXSR	(0*32+24) /* FXSAVE and FXRSTOR instructions (fast save and restore */
-					  /* of FPU context), and CR4.OSFXSR available */
-#define X86_FEATURE_XMM		(0*32+25) /* Streaming SIMD Extensions */
-#define X86_FEATURE_XMM2	(0*32+26) /* Streaming SIMD Extensions-2 */
-#define X86_FEATURE_SELFSNOOP	(0*32+27) /* CPU self snoop */
+#define X86_FEATURE_FXSR	(0*32+24) /* FXSAVE/FXRSTOR, CR4.OSFXSR */
+#define X86_FEATURE_XMM		(0*32+25) /* "sse" */
+#define X86_FEATURE_XMM2	(0*32+26) /* "sse2" */
+#define X86_FEATURE_SELFSNOOP	(0*32+27) /* "ss" CPU self snoop */
 #define X86_FEATURE_HT		(0*32+28) /* Hyper-Threading */
-#define X86_FEATURE_ACC		(0*32+29) /* Automatic clock control */
+#define X86_FEATURE_ACC		(0*32+29) /* "tm" Automatic clock control */
 #define X86_FEATURE_IA64	(0*32+30) /* IA-64 processor */
+#define X86_FEATURE_PBE		(0*32+31) /* Pending Break Enable */
 
 /* AMD-defined CPU features, CPUID level 0x80000001, word 1 */
 /* Don't duplicate feature flags which are redundant with Intel! */
@@ -46,7 +53,8 @@
 #define X86_FEATURE_MP		(1*32+19) /* MP Capable. */
 #define X86_FEATURE_NX		(1*32+20) /* Execute Disable */
 #define X86_FEATURE_MMXEXT	(1*32+22) /* AMD MMX extensions */
-#define X86_FEATURE_GBPAGES	(1*32+26) /* GB pages */
+#define X86_FEATURE_FXSR_OPT	(1*32+25) /* FXSAVE/FXRSTOR optimizations */
+#define X86_FEATURE_GBPAGES	(1*32+26) /* "pdpe1gb" GB pages */
 #define X86_FEATURE_RDTSCP	(1*32+27) /* RDTSCP */
 #define X86_FEATURE_LM		(1*32+29) /* Long Mode (x86-64) */
 #define X86_FEATURE_3DNOWEXT	(1*32+30) /* AMD 3DNow! extensions */
@@ -64,53 +72,79 @@
 #define X86_FEATURE_CYRIX_ARR	(3*32+ 2) /* Cyrix ARRs (= MTRRs) */
 #define X86_FEATURE_CENTAUR_MCR	(3*32+ 3) /* Centaur MCRs (= MTRRs) */
 /* cpu types for specific tunings: */
-#define X86_FEATURE_K8		(3*32+ 4) /* Opteron, Athlon64 */
-#define X86_FEATURE_K7		(3*32+ 5) /* Athlon */
-#define X86_FEATURE_P3		(3*32+ 6) /* P3 */
-#define X86_FEATURE_P4		(3*32+ 7) /* P4 */
+#define X86_FEATURE_K8		(3*32+ 4) /* "" Opteron, Athlon64 */
+#define X86_FEATURE_K7		(3*32+ 5) /* "" Athlon */
+#define X86_FEATURE_P3		(3*32+ 6) /* "" P3 */
+#define X86_FEATURE_P4		(3*32+ 7) /* "" P4 */
 #define X86_FEATURE_CONSTANT_TSC (3*32+ 8) /* TSC ticks at a constant rate */
 #define X86_FEATURE_UP		(3*32+ 9) /* smp kernel running on up */
-#define X86_FEATURE_FXSAVE_LEAK (3*32+10) /* FXSAVE leaks FOP/FIP/FOP */
+#define X86_FEATURE_FXSAVE_LEAK (3*32+10) /* "" FXSAVE leaks FOP/FIP/FOP */
 #define X86_FEATURE_ARCH_PERFMON (3*32+11) /* Intel Architectural PerfMon */
+#define X86_FEATURE_NOPL	(3*32+20) /* The NOPL (0F 1F) instructions */
 #define X86_FEATURE_PEBS	(3*32+12) /* Precise-Event Based Sampling */
 #define X86_FEATURE_BTS		(3*32+13) /* Branch Trace Store */
-#define X86_FEATURE_SYSCALL32	(3*32+14) /* syscall in ia32 userspace */
-#define X86_FEATURE_SYSENTER32	(3*32+15) /* sysenter in ia32 userspace */
-#define X86_FEATURE_REP_GOOD	(3*32+16) /* rep microcode works well on this CPU */
-#define X86_FEATURE_MFENCE_RDTSC (3*32+17) /* Mfence synchronizes RDTSC */
-#define X86_FEATURE_LFENCE_RDTSC (3*32+18) /* Lfence synchronizes RDTSC */
-#define X86_FEATURE_11AP	(3*32+19) /* Bad local APIC aka 11AP */
+#define X86_FEATURE_SYSCALL32	(3*32+14) /* "" syscall in ia32 userspace */
+#define X86_FEATURE_SYSENTER32	(3*32+15) /* "" sysenter in ia32 userspace */
+#define X86_FEATURE_REP_GOOD	(3*32+16) /* rep microcode works well */
+#define X86_FEATURE_MFENCE_RDTSC (3*32+17) /* "" Mfence synchronizes RDTSC */
+#define X86_FEATURE_LFENCE_RDTSC (3*32+18) /* "" Lfence synchronizes RDTSC */
+#define X86_FEATURE_11AP	(3*32+19) /* "" Bad local APIC aka 11AP */
 #define X86_FEATURE_NOPL	(3*32+20) /* The NOPL (0F 1F) instructions */
 #define X86_FEATURE_AMDC1E	(3*32+21) /* AMD C1E detected */
+#define X86_FEATURE_XTOPOLOGY	(3*32+21) /* cpu topology enum extensions */
 
 /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */
-#define X86_FEATURE_XMM3	(4*32+ 0) /* Streaming SIMD Extensions-3 */
-#define X86_FEATURE_MWAIT	(4*32+ 3) /* Monitor/Mwait support */
-#define X86_FEATURE_DSCPL	(4*32+ 4) /* CPL Qualified Debug Store */
+#define X86_FEATURE_XMM3	(4*32+ 0) /* "pni" SSE-3 */
+#define X86_FEATURE_PCLMULQDQ	(4*32+ 1) /* PCLMULQDQ instruction */
+#define X86_FEATURE_DTES64	(4*32+ 2) /* 64-bit Debug Store */
+#define X86_FEATURE_MWAIT	(4*32+ 3) /* "monitor" Monitor/Mwait support */
+#define X86_FEATURE_DSCPL	(4*32+ 4) /* "ds_cpl" CPL Qual. Debug Store */
+#define X86_FEATURE_VMX		(4*32+ 5) /* Hardware virtualization */
+#define X86_FEATURE_SMX		(4*32+ 6) /* Safer mode */
 #define X86_FEATURE_EST		(4*32+ 7) /* Enhanced SpeedStep */
 #define X86_FEATURE_TM2		(4*32+ 8) /* Thermal Monitor 2 */
+#define X86_FEATURE_SSSE3	(4*32+ 9) /* Supplemental SSE-3 */
 #define X86_FEATURE_CID		(4*32+10) /* Context ID */
+#define X86_FEATURE_FMA		(4*32+12) /* Fused multiply-add */
 #define X86_FEATURE_CX16	(4*32+13) /* CMPXCHG16B */
 #define X86_FEATURE_XTPR	(4*32+14) /* Send Task Priority Messages */
+#define X86_FEATURE_PDCM	(4*32+15) /* Performance Capabilities */
 #define X86_FEATURE_DCA		(4*32+18) /* Direct Cache Access */
-#define X86_FEATURE_XMM4_2	(4*32+20) /* Streaming SIMD Extensions-4.2 */
+#define X86_FEATURE_XMM4_1	(4*32+19) /* "sse4_1" SSE-4.1 */
+#define X86_FEATURE_XMM4_2	(4*32+20) /* "sse4_2" SSE-4.2 */
+#define X86_FEATURE_X2APIC	(4*32+21) /* x2APIC */
+#define X86_FEATURE_AES		(4*32+25) /* AES instructions */
+#define X86_FEATURE_XSAVE	(4*32+26) /* XSAVE/XRSTOR/XSETBV/XGETBV */
+#define X86_FEATURE_OSXSAVE	(4*32+27) /* "" XSAVE enabled in the OS */
+#define X86_FEATURE_AVX		(4*32+28) /* Advanced Vector Extensions */
 
 /* VIA/Cyrix/Centaur-defined CPU features, CPUID level 0xC0000001, word 5 */
-#define X86_FEATURE_XSTORE	(5*32+ 2) /* on-CPU RNG present (xstore insn) */
-#define X86_FEATURE_XSTORE_EN	(5*32+ 3) /* on-CPU RNG enabled */
-#define X86_FEATURE_XCRYPT	(5*32+ 6) /* on-CPU crypto (xcrypt insn) */
-#define X86_FEATURE_XCRYPT_EN	(5*32+ 7) /* on-CPU crypto enabled */
+#define X86_FEATURE_XSTORE	(5*32+ 2) /* "rng" RNG present (xstore) */
+#define X86_FEATURE_XSTORE_EN	(5*32+ 3) /* "rng_en" RNG enabled */
+#define X86_FEATURE_XCRYPT	(5*32+ 6) /* "ace" on-CPU crypto (xcrypt) */
+#define X86_FEATURE_XCRYPT_EN	(5*32+ 7) /* "ace_en" on-CPU crypto enabled */
 #define X86_FEATURE_ACE2	(5*32+ 8) /* Advanced Cryptography Engine v2 */
 #define X86_FEATURE_ACE2_EN	(5*32+ 9) /* ACE v2 enabled */
-#define X86_FEATURE_PHE		(5*32+ 10) /* PadLock Hash Engine */
-#define X86_FEATURE_PHE_EN	(5*32+ 11) /* PHE enabled */
-#define X86_FEATURE_PMM		(5*32+ 12) /* PadLock Montgomery Multiplier */
-#define X86_FEATURE_PMM_EN	(5*32+ 13) /* PMM enabled */
+#define X86_FEATURE_PHE		(5*32+10) /* PadLock Hash Engine */
+#define X86_FEATURE_PHE_EN	(5*32+11) /* PHE enabled */
+#define X86_FEATURE_PMM		(5*32+12) /* PadLock Montgomery Multiplier */
+#define X86_FEATURE_PMM_EN	(5*32+13) /* PMM enabled */
 
 /* More extended AMD flags: CPUID level 0x80000001, ecx, word 6 */
 #define X86_FEATURE_LAHF_LM	(6*32+ 0) /* LAHF/SAHF in long mode */
 #define X86_FEATURE_CMP_LEGACY	(6*32+ 1) /* If yes HyperThreading not valid */
-#define X86_FEATURE_IBS		(6*32+ 10) /* Instruction Based Sampling */
+#define X86_FEATURE_SVM		(6*32+ 2) /* Secure virtual machine */
+#define X86_FEATURE_EXTAPIC	(6*32+ 3) /* Extended APIC space */
+#define X86_FEATURE_CR8_LEGACY	(6*32+ 4) /* CR8 in 32-bit mode */
+#define X86_FEATURE_ABM		(6*32+ 5) /* Advanced bit manipulation */
+#define X86_FEATURE_SSE4A	(6*32+ 6) /* SSE-4A */
+#define X86_FEATURE_MISALIGNSSE (6*32+ 7) /* Misaligned SSE mode */
+#define X86_FEATURE_3DNOWPREFETCH (6*32+ 8) /* 3DNow prefetch instructions */
+#define X86_FEATURE_OSVW	(6*32+ 9) /* OS Visible Workaround */
+#define X86_FEATURE_IBS		(6*32+10) /* Instruction Based Sampling */
+#define X86_FEATURE_SSE5	(6*32+11) /* SSE-5 */
+#define X86_FEATURE_SKINIT	(6*32+12) /* SKINIT/STGI instructions */
+#define X86_FEATURE_WDT		(6*32+13) /* Watchdog timer */
 
 /*
  * Auxiliary flags: Linux defined - For features scattered in various
@@ -118,6 +152,13 @@
  */
 #define X86_FEATURE_IDA		(7*32+ 0) /* Intel Dynamic Acceleration */
 
+/* Virtualization flags: Linux defined */
+#define X86_FEATURE_TPR_SHADOW  (8*32+ 0) /* Intel TPR Shadow */
+#define X86_FEATURE_VNMI        (8*32+ 1) /* Intel Virtual NMI */
+#define X86_FEATURE_FLEXPRIORITY (8*32+ 2) /* Intel FlexPriority */
+#define X86_FEATURE_EPT         (8*32+ 3) /* Intel Extended Page Table */
+#define X86_FEATURE_VPID        (8*32+ 4) /* Intel Virtual Processor ID */
+
 #if defined(__KERNEL__) && !defined(__ASSEMBLY__)
 
 #include <linux/bitops.h>
@@ -151,7 +192,7 @@ extern const char * const x86_power_flags[32];
 } while (0)
 #define setup_force_cpu_cap(bit) do { \
 	set_cpu_cap(&boot_cpu_data, bit);	\
-	clear_bit(bit, (unsigned long *)cleared_cpu_caps); 	\
+	clear_bit(bit, (unsigned long *)cleared_cpu_caps);	\
 } while (0)
 
 #define cpu_has_fpu		boot_cpu_has(X86_FEATURE_FPU)
@@ -192,7 +233,10 @@ extern const char * const x86_power_flags[32];
 #define cpu_has_gbpages		boot_cpu_has(X86_FEATURE_GBPAGES)
 #define cpu_has_arch_perfmon	boot_cpu_has(X86_FEATURE_ARCH_PERFMON)
 #define cpu_has_pat		boot_cpu_has(X86_FEATURE_PAT)
+#define cpu_has_xmm4_1		boot_cpu_has(X86_FEATURE_XMM4_1)
 #define cpu_has_xmm4_2		boot_cpu_has(X86_FEATURE_XMM4_2)
+#define cpu_has_x2apic		boot_cpu_has(X86_FEATURE_X2APIC)
+#define cpu_has_xsave		boot_cpu_has(X86_FEATURE_XSAVE)
 
 #if defined(CONFIG_X86_INVLPG) || defined(CONFIG_X86_64)
 # define cpu_has_invlpg		1
diff --git a/include/asm-x86/desc.h b/include/asm-x86/desc.h
index b73fea54def2..f06adac7938c 100644
--- a/include/asm-x86/desc.h
+++ b/include/asm-x86/desc.h
@@ -24,6 +24,11 @@ static inline void fill_ldt(struct desc_struct *desc,
 	desc->d = info->seg_32bit;
 	desc->g = info->limit_in_pages;
 	desc->base2 = (info->base_addr & 0xff000000) >> 24;
+	/*
+	 * Don't allow setting of the lm bit. It is useless anyway
+	 * because 64bit system calls require __USER_CS:
+	 */
+	desc->l = 0;
 }
 
 extern struct desc_ptr idt_descr;
@@ -97,7 +102,15 @@ static inline int desc_empty(const void *ptr)
 	native_write_gdt_entry(dt, entry, desc, type)
 #define write_idt_entry(dt, entry, g)		\
 	native_write_idt_entry(dt, entry, g)
-#endif
+
+static inline void paravirt_alloc_ldt(struct desc_struct *ldt, unsigned entries)
+{
+}
+
+static inline void paravirt_free_ldt(struct desc_struct *ldt, unsigned entries)
+{
+}
+#endif	/* CONFIG_PARAVIRT */
 
 static inline void native_write_idt_entry(gate_desc *idt, int entry,
 					  const gate_desc *gate)
@@ -338,20 +351,16 @@ static inline void set_system_intr_gate(unsigned int n, void *addr)
 	_set_gate(n, GATE_INTERRUPT, addr, 0x3, 0, __KERNEL_CS);
 }
 
-static inline void set_trap_gate(unsigned int n, void *addr)
+static inline void set_system_trap_gate(unsigned int n, void *addr)
 {
 	BUG_ON((unsigned)n > 0xFF);
-	_set_gate(n, GATE_TRAP, addr, 0, 0, __KERNEL_CS);
+	_set_gate(n, GATE_TRAP, addr, 0x3, 0, __KERNEL_CS);
 }
 
-static inline void set_system_gate(unsigned int n, void *addr)
+static inline void set_trap_gate(unsigned int n, void *addr)
 {
 	BUG_ON((unsigned)n > 0xFF);
-#ifdef CONFIG_X86_32
-	_set_gate(n, GATE_TRAP, addr, 0x3, 0, __KERNEL_CS);
-#else
-	_set_gate(n, GATE_INTERRUPT, addr, 0x3, 0, __KERNEL_CS);
-#endif
+	_set_gate(n, GATE_TRAP, addr, 0, 0, __KERNEL_CS);
 }
 
 static inline void set_task_gate(unsigned int n, unsigned int gdt_entry)
@@ -366,7 +375,7 @@ static inline void set_intr_gate_ist(int n, void *addr, unsigned ist)
 	_set_gate(n, GATE_INTERRUPT, addr, 0, ist, __KERNEL_CS);
 }
 
-static inline void set_system_gate_ist(int n, void *addr, unsigned ist)
+static inline void set_system_intr_gate_ist(int n, void *addr, unsigned ist)
 {
 	BUG_ON((unsigned)n > 0xFF);
 	_set_gate(n, GATE_INTERRUPT, addr, 0x3, ist, __KERNEL_CS);
diff --git a/include/asm-x86/e820.h b/include/asm-x86/e820.h
index f52daf176bcb..5abbdec06bd2 100644
--- a/include/asm-x86/e820.h
+++ b/include/asm-x86/e820.h
@@ -43,6 +43,7 @@
 #define E820_RESERVED	2
 #define E820_ACPI	3
 #define E820_NVS	4
+#define E820_UNUSABLE	5
 
 /* reserved RAM used by kernel itself */
 #define E820_RESERVED_KERN        128
@@ -121,6 +122,7 @@ extern void e820_register_active_regions(int nid, unsigned long start_pfn,
 extern u64 e820_hole_size(u64 start, u64 end);
 extern void finish_e820_parsing(void);
 extern void e820_reserve_resources(void);
+extern void e820_reserve_resources_late(void);
 extern void setup_memory_map(void);
 extern char *default_machine_specific_memory_setup(void);
 extern char *machine_specific_memory_setup(void);
diff --git a/include/asm-x86/mach-es7000/mach_apic.h b/include/asm-x86/es7000/apic.h
index c1f6f682d619..bd2c44d1f7ac 100644
--- a/include/asm-x86/mach-es7000/mach_apic.h
+++ b/include/asm-x86/es7000/apic.h
@@ -1,5 +1,5 @@
-#ifndef ASM_X86__MACH_ES7000__MACH_APIC_H
-#define ASM_X86__MACH_ES7000__MACH_APIC_H
+#ifndef __ASM_ES7000_APIC_H
+#define __ASM_ES7000_APIC_H
 
 #define xapic_phys_to_log_apicid(cpu) per_cpu(x86_bios_cpu_apicid, cpu)
 #define esr_disable (1)
@@ -10,7 +10,7 @@ static inline int apic_id_registered(void)
 }
 
 static inline cpumask_t target_cpus(void)
-{ 
+{
 #if defined CONFIG_ES7000_CLUSTERED_APIC
 	return CPU_MASK_ALL;
 #else
@@ -23,24 +23,24 @@ static inline cpumask_t target_cpus(void)
 #define APIC_DFR_VALUE		(APIC_DFR_CLUSTER)
 #define INT_DELIVERY_MODE	(dest_LowestPrio)
 #define INT_DEST_MODE		(1)    /* logical delivery broadcast to all procs */
-#define NO_BALANCE_IRQ 		(1)
+#define NO_BALANCE_IRQ		(1)
 #undef  WAKE_SECONDARY_VIA_INIT
 #define WAKE_SECONDARY_VIA_MIP
 #else
 #define APIC_DFR_VALUE		(APIC_DFR_FLAT)
 #define INT_DELIVERY_MODE	(dest_Fixed)
 #define INT_DEST_MODE		(0)    /* phys delivery to target procs */
-#define NO_BALANCE_IRQ 		(0)
+#define NO_BALANCE_IRQ		(0)
 #undef  APIC_DEST_LOGICAL
 #define APIC_DEST_LOGICAL	0x0
 #define WAKE_SECONDARY_VIA_INIT
 #endif
 
 static inline unsigned long check_apicid_used(physid_mask_t bitmap, int apicid)
-{ 
+{
 	return 0;
-} 
-static inline unsigned long check_apicid_present(int bit) 
+}
+static inline unsigned long check_apicid_present(int bit)
 {
 	return physid_isset(bit, phys_cpu_present_map);
 }
@@ -80,7 +80,7 @@ static inline void setup_apic_routing(void)
 {
 	int apic = per_cpu(x86_bios_cpu_apicid, smp_processor_id());
 	printk("Enabling APIC mode:  %s.  Using %d I/O APICs, target cpus %lx\n",
-		(apic_version[apic] == 0x14) ? 
+		(apic_version[apic] == 0x14) ?
 		"Physical Cluster" : "Logical Cluster", nr_ioapics, cpus_addr(TARGET_CPUS)[0]);
 }
 
@@ -141,7 +141,7 @@ static inline void setup_portio_remap(void)
 extern unsigned int boot_cpu_physical_apicid;
 static inline int check_phys_apicid_present(int cpu_physical_apicid)
 {
-	boot_cpu_physical_apicid = GET_APIC_ID(read_apic_id());
+	boot_cpu_physical_apicid = read_apic_id();
 	return (1);
 }
 
@@ -150,7 +150,7 @@ static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask)
 	int num_bits_set;
 	int cpus_found = 0;
 	int cpu;
-	int apicid;	
+	int apicid;
 
 	num_bits_set = cpus_weight(cpumask);
 	/* Return id to all */
@@ -160,16 +160,16 @@ static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask)
 #else
 		return cpu_to_logical_apicid(0);
 #endif
-	/* 
-	 * The cpus in the mask must all be on the apic cluster.  If are not 
-	 * on the same apicid cluster return default value of TARGET_CPUS. 
+	/*
+	 * The cpus in the mask must all be on the apic cluster.  If are not
+	 * on the same apicid cluster return default value of TARGET_CPUS.
 	 */
 	cpu = first_cpu(cpumask);
 	apicid = cpu_to_logical_apicid(cpu);
 	while (cpus_found < num_bits_set) {
 		if (cpu_isset(cpu, cpumask)) {
 			int new_apicid = cpu_to_logical_apicid(cpu);
-			if (apicid_cluster(apicid) != 
+			if (apicid_cluster(apicid) !=
 					apicid_cluster(new_apicid)){
 				printk ("%s: Not a valid mask!\n",__FUNCTION__);
 #if defined CONFIG_ES7000_CLUSTERED_APIC
@@ -191,4 +191,4 @@ static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb)
 	return cpuid_apic >> index_msb;
 }
 
-#endif /* ASM_X86__MACH_ES7000__MACH_APIC_H */
+#endif /* __ASM_ES7000_APIC_H */
diff --git a/include/asm-x86/es7000/apicdef.h b/include/asm-x86/es7000/apicdef.h
new file mode 100644
index 000000000000..8b234a3cb851
--- /dev/null
+++ b/include/asm-x86/es7000/apicdef.h
@@ -0,0 +1,13 @@
+#ifndef __ASM_ES7000_APICDEF_H
+#define __ASM_ES7000_APICDEF_H
+
+#define		APIC_ID_MASK		(0xFF<<24)
+
+static inline unsigned get_apic_id(unsigned long x)
+{
+	return (((x)>>24)&0xFF);
+}
+
+#define		GET_APIC_ID(x)	get_apic_id(x)
+
+#endif
diff --git a/include/asm-x86/mach-es7000/mach_ipi.h b/include/asm-x86/es7000/ipi.h
index 3a21240e03dc..632a955fcc0a 100644
--- a/include/asm-x86/mach-es7000/mach_ipi.h
+++ b/include/asm-x86/es7000/ipi.h
@@ -1,5 +1,5 @@
-#ifndef ASM_X86__MACH_ES7000__MACH_IPI_H
-#define ASM_X86__MACH_ES7000__MACH_IPI_H
+#ifndef __ASM_ES7000_IPI_H
+#define __ASM_ES7000_IPI_H
 
 void send_IPI_mask_sequence(cpumask_t mask, int vector);
 
@@ -21,4 +21,4 @@ static inline void send_IPI_all(int vector)
 	send_IPI_mask(cpu_online_map, vector);
 }
 
-#endif /* ASM_X86__MACH_ES7000__MACH_IPI_H */
+#endif /* __ASM_ES7000_IPI_H */
diff --git a/include/asm-x86/mach-es7000/mach_mpparse.h b/include/asm-x86/es7000/mpparse.h
index befde24705b7..ed5a3caae141 100644
--- a/include/asm-x86/mach-es7000/mach_mpparse.h
+++ b/include/asm-x86/es7000/mpparse.h
@@ -1,10 +1,11 @@
-#ifndef ASM_X86__MACH_ES7000__MACH_MPPARSE_H
-#define ASM_X86__MACH_ES7000__MACH_MPPARSE_H
+#ifndef __ASM_ES7000_MPPARSE_H
+#define __ASM_ES7000_MPPARSE_H
 
 #include <linux/acpi.h>
 
 extern int parse_unisys_oem (char *oemptr);
 extern int find_unisys_acpi_oem_table(unsigned long *oem_addr);
+extern void unmap_unisys_acpi_oem_table(unsigned long oem_addr);
 extern void setup_unisys(void);
 
 #ifndef CONFIG_X86_GENERICARCH
@@ -26,4 +27,4 @@ static inline int es7000_check_dsdt(void)
 }
 #endif
 
-#endif /* ASM_X86__MACH_ES7000__MACH_MPPARSE_H */
+#endif /* __ASM_MACH_MPPARSE_H */
diff --git a/include/asm-x86/mach-es7000/mach_wakecpu.h b/include/asm-x86/es7000/wakecpu.h
index 97c776ce13f2..3ffc5a7bf667 100644
--- a/include/asm-x86/mach-es7000/mach_wakecpu.h
+++ b/include/asm-x86/es7000/wakecpu.h
@@ -1,7 +1,7 @@
-#ifndef ASM_X86__MACH_ES7000__MACH_WAKECPU_H
-#define ASM_X86__MACH_ES7000__MACH_WAKECPU_H
+#ifndef __ASM_ES7000_WAKECPU_H
+#define __ASM_ES7000_WAKECPU_H
 
-/* 
+/*
  * This file copes with machines that wakeup secondary CPUs by the
  * INIT, INIT, STARTUP sequence.
  */
@@ -56,4 +56,4 @@ static inline void restore_NMI_vector(unsigned short *high, unsigned short *low)
  #define inquire_remote_apic(apicid) {}
 #endif
 
-#endif /* ASM_X86__MACH_ES7000__MACH_WAKECPU_H */
+#endif /* __ASM_MACH_WAKECPU_H */
diff --git a/include/asm-x86/fixmap_32.h b/include/asm-x86/fixmap_32.h
index 784e3e759866..8844002da0e0 100644
--- a/include/asm-x86/fixmap_32.h
+++ b/include/asm-x86/fixmap_32.h
@@ -94,10 +94,10 @@ enum fixed_addresses {
 	 * can have a single pgd entry and a single pte table:
 	 */
 #define NR_FIX_BTMAPS		64
-#define FIX_BTMAPS_NESTING	4
+#define FIX_BTMAPS_SLOTS	4
 	FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 -
 			(__end_of_permanent_fixed_addresses & 255),
-	FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_NESTING - 1,
+	FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_SLOTS - 1,
 	FIX_WP_TEST,
 #ifdef CONFIG_ACPI
 	FIX_ACPI_BEGIN,
diff --git a/include/asm-x86/fixmap_64.h b/include/asm-x86/fixmap_64.h
index dafb24bc0424..dab4751d1307 100644
--- a/include/asm-x86/fixmap_64.h
+++ b/include/asm-x86/fixmap_64.h
@@ -49,6 +49,7 @@ enum fixed_addresses {
 #ifdef CONFIG_PARAVIRT
 	FIX_PARAVIRT_BOOTMAP,
 #endif
+	__end_of_permanent_fixed_addresses,
 #ifdef CONFIG_ACPI
 	FIX_ACPI_BEGIN,
 	FIX_ACPI_END = FIX_ACPI_BEGIN + FIX_ACPI_PAGES - 1,
@@ -56,19 +57,18 @@ enum fixed_addresses {
 #ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT
 	FIX_OHCI1394_BASE,
 #endif
-	__end_of_permanent_fixed_addresses,
 	/*
 	 * 256 temporary boot-time mappings, used by early_ioremap(),
 	 * before ioremap() is functional.
 	 *
-	 * We round it up to the next 512 pages boundary so that we
+	 * We round it up to the next 256 pages boundary so that we
 	 * can have a single pgd entry and a single pte table:
 	 */
 #define NR_FIX_BTMAPS		64
-#define FIX_BTMAPS_NESTING	4
-	FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 512 -
-			(__end_of_permanent_fixed_addresses & 511),
-	FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_NESTING - 1,
+#define FIX_BTMAPS_SLOTS	4
+	FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 -
+			(__end_of_permanent_fixed_addresses & 255),
+	FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_SLOTS - 1,
 	__end_of_fixed_addresses
 };
 
diff --git a/include/asm-x86/genapic_64.h b/include/asm-x86/genapic_64.h
index 25097a8cc5ef..ed6a4886c082 100644
--- a/include/asm-x86/genapic_64.h
+++ b/include/asm-x86/genapic_64.h
@@ -14,6 +14,7 @@
 
 struct genapic {
 	char *name;
+	int (*acpi_madt_oem_check)(char *oem_id, char *oem_table_id);
 	u32 int_delivery_mode;
 	u32 int_dest_mode;
 	int (*apic_id_registered)(void);
@@ -24,17 +25,24 @@ struct genapic {
 	void (*send_IPI_mask)(cpumask_t mask, int vector);
 	void (*send_IPI_allbutself)(int vector);
 	void (*send_IPI_all)(int vector);
+	void (*send_IPI_self)(int vector);
 	/* */
 	unsigned int (*cpu_mask_to_apicid)(cpumask_t cpumask);
 	unsigned int (*phys_pkg_id)(int index_msb);
+	unsigned int (*get_apic_id)(unsigned long x);
+	unsigned long (*set_apic_id)(unsigned int id);
+	unsigned long apic_id_mask;
 };
 
 extern struct genapic *genapic;
 
 extern struct genapic apic_flat;
 extern struct genapic apic_physflat;
+extern struct genapic apic_x2apic_cluster;
+extern struct genapic apic_x2apic_phys;
 extern int acpi_madt_oem_check(char *, char *);
 
+extern void apic_send_IPI_self(int vector);
 enum uv_system_type {UV_NONE, UV_LEGACY_APIC, UV_X2APIC, UV_NON_UNIQUE_APIC};
 extern enum uv_system_type get_uv_system_type(void);
 extern int is_uv_system(void);
diff --git a/include/asm-x86/hw_irq.h b/include/asm-x86/hw_irq.h
index 65997b15d56a..50f6e0316b50 100644
--- a/include/asm-x86/hw_irq.h
+++ b/include/asm-x86/hw_irq.h
@@ -64,7 +64,6 @@ extern unsigned long io_apic_irqs;
 extern void init_VISWS_APIC_irqs(void);
 extern void setup_IO_APIC(void);
 extern void disable_IO_APIC(void);
-extern void print_IO_APIC(void);
 extern int IO_APIC_get_PCI_irq_vector(int bus, int slot, int fn);
 extern void setup_ioapic_dest(void);
 
@@ -73,7 +72,9 @@ extern void enable_IO_APIC(void);
 #endif
 
 /* IPI functions */
+#ifdef CONFIG_X86_32
 extern void send_IPI_self(int vector);
+#endif
 extern void send_IPI(int dest, int vector);
 
 /* Statistics */
diff --git a/include/asm-x86/i387.h b/include/asm-x86/i387.h
index 1ecdc3ed96e4..9ba862a4eac0 100644
--- a/include/asm-x86/i387.h
+++ b/include/asm-x86/i387.h
@@ -19,7 +19,9 @@
 #include <asm/sigcontext.h>
 #include <asm/user.h>
 #include <asm/uaccess.h>
+#include <asm/xsave.h>
 
+extern unsigned int sig_xstate_size;
 extern void fpu_init(void);
 extern void mxcsr_feature_mask_init(void);
 extern int init_fpu(struct task_struct *child);
@@ -31,12 +33,18 @@ extern user_regset_active_fn fpregs_active, xfpregs_active;
 extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get;
 extern user_regset_set_fn fpregs_set, xfpregs_set, fpregs_soft_set;
 
+extern struct _fpx_sw_bytes fx_sw_reserved;
 #ifdef CONFIG_IA32_EMULATION
+extern unsigned int sig_xstate_ia32_size;
+extern struct _fpx_sw_bytes fx_sw_reserved_ia32;
 struct _fpstate_ia32;
-extern int save_i387_ia32(struct _fpstate_ia32 __user *buf);
-extern int restore_i387_ia32(struct _fpstate_ia32 __user *buf);
+struct _xstate_ia32;
+extern int save_i387_xstate_ia32(void __user *buf);
+extern int restore_i387_xstate_ia32(void __user *buf);
 #endif
 
+#define X87_FSW_ES (1 << 7)	/* Exception Summary */
+
 #ifdef CONFIG_X86_64
 
 /* Ignore delayed exceptions from user space */
@@ -47,7 +55,7 @@ static inline void tolerant_fwait(void)
 		     _ASM_EXTABLE(1b, 2b));
 }
 
-static inline int restore_fpu_checking(struct i387_fxsave_struct *fx)
+static inline int fxrstor_checking(struct i387_fxsave_struct *fx)
 {
 	int err;
 
@@ -67,15 +75,31 @@ static inline int restore_fpu_checking(struct i387_fxsave_struct *fx)
 	return err;
 }
 
-#define X87_FSW_ES (1 << 7)	/* Exception Summary */
+static inline int restore_fpu_checking(struct task_struct *tsk)
+{
+	if (task_thread_info(tsk)->status & TS_XSAVE)
+		return xrstor_checking(&tsk->thread.xstate->xsave);
+	else
+		return fxrstor_checking(&tsk->thread.xstate->fxsave);
+}
 
 /* AMD CPUs don't save/restore FDP/FIP/FOP unless an exception
    is pending. Clear the x87 state here by setting it to fixed
    values. The kernel data segment can be sometimes 0 and sometimes
    new user value. Both should be ok.
    Use the PDA as safe address because it should be already in L1. */
-static inline void clear_fpu_state(struct i387_fxsave_struct *fx)
+static inline void clear_fpu_state(struct task_struct *tsk)
 {
+	struct xsave_struct *xstate = &tsk->thread.xstate->xsave;
+	struct i387_fxsave_struct *fx = &tsk->thread.xstate->fxsave;
+
+	/*
+	 * xsave header may indicate the init state of the FP.
+	 */
+	if ((task_thread_info(tsk)->status & TS_XSAVE) &&
+	    !(xstate->xsave_hdr.xstate_bv & XSTATE_FP))
+		return;
+
 	if (unlikely(fx->swd & X87_FSW_ES))
 		asm volatile("fnclex");
 	alternative_input(ASM_NOP8 ASM_NOP2,
@@ -84,7 +108,7 @@ static inline void clear_fpu_state(struct i387_fxsave_struct *fx)
 			  X86_FEATURE_FXSAVE_LEAK);
 }
 
-static inline int save_i387_checking(struct i387_fxsave_struct __user *fx)
+static inline int fxsave_user(struct i387_fxsave_struct __user *fx)
 {
 	int err;
 
@@ -108,7 +132,7 @@ static inline int save_i387_checking(struct i387_fxsave_struct __user *fx)
 	return err;
 }
 
-static inline void __save_init_fpu(struct task_struct *tsk)
+static inline void fxsave(struct task_struct *tsk)
 {
 	/* Using "rex64; fxsave %0" is broken because, if the memory operand
 	   uses any extended registers for addressing, a second REX prefix
@@ -133,7 +157,16 @@ static inline void __save_init_fpu(struct task_struct *tsk)
 			     : "=m" (tsk->thread.xstate->fxsave)
 			     : "cdaSDb" (&tsk->thread.xstate->fxsave));
 #endif
-	clear_fpu_state(&tsk->thread.xstate->fxsave);
+}
+
+static inline void __save_init_fpu(struct task_struct *tsk)
+{
+	if (task_thread_info(tsk)->status & TS_XSAVE)
+		xsave(tsk);
+	else
+		fxsave(tsk);
+
+	clear_fpu_state(tsk);
 	task_thread_info(tsk)->status &= ~TS_USEDFPU;
 }
 
@@ -148,6 +181,10 @@ static inline void tolerant_fwait(void)
 
 static inline void restore_fpu(struct task_struct *tsk)
 {
+	if (task_thread_info(tsk)->status & TS_XSAVE) {
+		xrstor_checking(&tsk->thread.xstate->xsave);
+		return;
+	}
 	/*
 	 * The "nop" is needed to make the instructions the same
 	 * length.
@@ -173,6 +210,27 @@ static inline void restore_fpu(struct task_struct *tsk)
  */
 static inline void __save_init_fpu(struct task_struct *tsk)
 {
+	if (task_thread_info(tsk)->status & TS_XSAVE) {
+		struct xsave_struct *xstate = &tsk->thread.xstate->xsave;
+		struct i387_fxsave_struct *fx = &tsk->thread.xstate->fxsave;
+
+		xsave(tsk);
+
+		/*
+		 * xsave header may indicate the init state of the FP.
+		 */
+		if (!(xstate->xsave_hdr.xstate_bv & XSTATE_FP))
+			goto end;
+
+		if (unlikely(fx->swd & X87_FSW_ES))
+			asm volatile("fnclex");
+
+		/*
+		 * we can do a simple return here or be paranoid :)
+		 */
+		goto clear_state;
+	}
+
 	/* Use more nops than strictly needed in case the compiler
 	   varies code */
 	alternative_input(
@@ -182,6 +240,7 @@ static inline void __save_init_fpu(struct task_struct *tsk)
 		X86_FEATURE_FXSR,
 		[fx] "m" (tsk->thread.xstate->fxsave),
 		[fsw] "m" (tsk->thread.xstate->fxsave.swd) : "memory");
+clear_state:
 	/* AMD K7/K8 CPUs don't save/restore FDP/FIP/FOP unless an exception
 	   is pending.  Clear the x87 state here by setting it to fixed
 	   values. safe_address is a random variable that should be in L1 */
@@ -191,16 +250,17 @@ static inline void __save_init_fpu(struct task_struct *tsk)
 		"fildl %[addr]", 	/* set F?P to defined value */
 		X86_FEATURE_FXSAVE_LEAK,
 		[addr] "m" (safe_address));
+end:
 	task_thread_info(tsk)->status &= ~TS_USEDFPU;
 }
 
+#endif	/* CONFIG_X86_64 */
+
 /*
  * Signal frame handlers...
  */
-extern int save_i387(struct _fpstate __user *buf);
-extern int restore_i387(struct _fpstate __user *buf);
-
-#endif	/* CONFIG_X86_64 */
+extern int save_i387_xstate(void __user *buf);
+extern int restore_i387_xstate(void __user *buf);
 
 static inline void __unlazy_fpu(struct task_struct *tsk)
 {
diff --git a/include/asm-x86/i8259.h b/include/asm-x86/i8259.h
index c586559a6957..23c1b3baaecd 100644
--- a/include/asm-x86/i8259.h
+++ b/include/asm-x86/i8259.h
@@ -57,4 +57,7 @@ static inline void outb_pic(unsigned char value, unsigned int port)
 
 extern struct irq_chip i8259A_chip;
 
+extern void mask_8259A(void);
+extern void unmask_8259A(void);
+
 #endif /* ASM_X86__I8259_H */
diff --git a/include/asm-x86/io.h b/include/asm-x86/io.h
index 72b7719523bf..a233f835e0b5 100644
--- a/include/asm-x86/io.h
+++ b/include/asm-x86/io.h
@@ -5,20 +5,6 @@
 
 #include <linux/compiler.h>
 
-/*
- * early_ioremap() and early_iounmap() are for temporary early boot-time
- * mappings, before the real ioremap() is functional.
- * A boot-time mapping is currently limited to at most 16 pages.
- */
-#ifndef __ASSEMBLY__
-extern void early_ioremap_init(void);
-extern void early_ioremap_clear(void);
-extern void early_ioremap_reset(void);
-extern void *early_ioremap(unsigned long offset, unsigned long size);
-extern void early_iounmap(void *addr, unsigned long size);
-extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys);
-#endif
-
 #define build_mmio_read(name, size, type, reg, barrier) \
 static inline type name(const volatile void __iomem *addr) \
 { type ret; asm volatile("mov" size " %1,%0":reg (ret) \
@@ -97,6 +83,7 @@ extern void early_ioremap_init(void);
 extern void early_ioremap_clear(void);
 extern void early_ioremap_reset(void);
 extern void *early_ioremap(unsigned long offset, unsigned long size);
+extern void *early_memremap(unsigned long offset, unsigned long size);
 extern void early_iounmap(void *addr, unsigned long size);
 extern void __iomem *fix_ioremap(unsigned idx, unsigned long phys);
 
diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h
index 64429e9431a8..ee6e086b7dfe 100644
--- a/include/asm-x86/io_64.h
+++ b/include/asm-x86/io_64.h
@@ -165,9 +165,6 @@ static inline void *phys_to_virt(unsigned long address)
 
 #include <asm-generic/iomap.h>
 
-extern void *early_ioremap(unsigned long addr, unsigned long size);
-extern void early_iounmap(void *addr, unsigned long size);
-
 /*
  * This one maps high address device memory and turns off caching for that area.
  * it's useful if some control registers are in such an area and write combining
diff --git a/include/asm-x86/io_apic.h b/include/asm-x86/io_apic.h
index be62847ab07e..8ec68a50cf10 100644
--- a/include/asm-x86/io_apic.h
+++ b/include/asm-x86/io_apic.h
@@ -107,6 +107,20 @@ struct IO_APIC_route_entry {
 
 } __attribute__ ((packed));
 
+struct IR_IO_APIC_route_entry {
+	__u64	vector		: 8,
+		zero		: 3,
+		index2		: 1,
+		delivery_status : 1,
+		polarity	: 1,
+		irr		: 1,
+		trigger		: 1,
+		mask		: 1,
+		reserved	: 31,
+		format		: 1,
+		index		: 15;
+} __attribute__ ((packed));
+
 #ifdef CONFIG_X86_IO_APIC
 
 /*
@@ -183,6 +197,12 @@ extern int io_apic_set_pci_routing(int ioapic, int pin, int irq,
 extern int (*ioapic_renumber_irq)(int ioapic, int irq);
 extern void ioapic_init_mappings(void);
 
+#ifdef CONFIG_X86_64
+extern int save_mask_IO_APIC_setup(void);
+extern void restore_IO_APIC_setup(void);
+extern void reinit_intr_remapped_IO_APIC(int);
+#endif
+
 #else  /* !CONFIG_X86_IO_APIC */
 #define io_apic_assign_pci_irqs 0
 static const int timer_through_8259 = 0;
diff --git a/include/asm-x86/ioctls.h b/include/asm-x86/ioctls.h
index 336603512399..06752a649044 100644
--- a/include/asm-x86/ioctls.h
+++ b/include/asm-x86/ioctls.h
@@ -51,9 +51,15 @@
 #define TCSETS2		_IOW('T', 0x2B, struct termios2)
 #define TCSETSW2	_IOW('T', 0x2C, struct termios2)
 #define TCSETSF2	_IOW('T', 0x2D, struct termios2)
+#define TIOCGRS485	0x542E
+#define TIOCSRS485	0x542F
 #define TIOCGPTN	_IOR('T', 0x30, unsigned int)
 				/* Get Pty Number (of pty-mux device) */
 #define TIOCSPTLCK	_IOW('T', 0x31, int)  /* Lock/unlock Pty */
+#define TCGETX		0x5432 /* SYS5 TCGETX compatibility */
+#define TCSETX		0x5433
+#define TCSETXF		0x5434
+#define TCSETXW		0x5435
 
 #define FIONCLEX	0x5450
 #define FIOCLEX		0x5451
diff --git a/include/asm-x86/ipi.h b/include/asm-x86/ipi.h
index c1b226797518..30a692cfaff8 100644
--- a/include/asm-x86/ipi.h
+++ b/include/asm-x86/ipi.h
@@ -49,6 +49,12 @@ static inline int __prepare_ICR2(unsigned int mask)
 	return SET_APIC_DEST_FIELD(mask);
 }
 
+static inline void __xapic_wait_icr_idle(void)
+{
+	while (native_apic_mem_read(APIC_ICR) & APIC_ICR_BUSY)
+		cpu_relax();
+}
+
 static inline void __send_IPI_shortcut(unsigned int shortcut, int vector,
 				       unsigned int dest)
 {
@@ -64,7 +70,7 @@ static inline void __send_IPI_shortcut(unsigned int shortcut, int vector,
 	/*
 	 * Wait for idle.
 	 */
-	apic_wait_icr_idle();
+	__xapic_wait_icr_idle();
 
 	/*
 	 * No need to touch the target chip field
@@ -74,7 +80,7 @@ static inline void __send_IPI_shortcut(unsigned int shortcut, int vector,
 	/*
 	 * Send the IPI. The write to APIC_ICR fires this off.
 	 */
-	apic_write(APIC_ICR, cfg);
+	native_apic_mem_write(APIC_ICR, cfg);
 }
 
 /*
@@ -92,13 +98,13 @@ static inline void __send_IPI_dest_field(unsigned int mask, int vector,
 	if (unlikely(vector == NMI_VECTOR))
 		safe_apic_wait_icr_idle();
 	else
-		apic_wait_icr_idle();
+		__xapic_wait_icr_idle();
 
 	/*
 	 * prepare target chip field
 	 */
 	cfg = __prepare_ICR2(mask);
-	apic_write(APIC_ICR2, cfg);
+	native_apic_mem_write(APIC_ICR2, cfg);
 
 	/*
 	 * program the ICR
@@ -108,7 +114,7 @@ static inline void __send_IPI_dest_field(unsigned int mask, int vector,
 	/*
 	 * Send the IPI. The write to APIC_ICR fires this off.
 	 */
-	apic_write(APIC_ICR, cfg);
+	native_apic_mem_write(APIC_ICR, cfg);
 }
 
 static inline void send_IPI_mask_sequence(cpumask_t mask, int vector)
diff --git a/include/asm-x86/irq_remapping.h b/include/asm-x86/irq_remapping.h
new file mode 100644
index 000000000000..78242c6ffa58
--- /dev/null
+++ b/include/asm-x86/irq_remapping.h
@@ -0,0 +1,8 @@
+#ifndef _ASM_IRQ_REMAPPING_H
+#define _ASM_IRQ_REMAPPING_H
+
+extern int x2apic;
+
+#define IRTE_DEST(dest) ((x2apic) ? dest : dest << 8)
+
+#endif
diff --git a/include/asm-x86/irqflags.h b/include/asm-x86/irqflags.h
index 424acb48cd61..2bdab21f0898 100644
--- a/include/asm-x86/irqflags.h
+++ b/include/asm-x86/irqflags.h
@@ -166,27 +166,6 @@ static inline int raw_irqs_disabled(void)
 	return raw_irqs_disabled_flags(flags);
 }
 
-/*
- * makes the traced hardirq state match with the machine state
- *
- * should be a rarely used function, only in places where its
- * otherwise impossible to know the irq state, like in traps.
- */
-static inline void trace_hardirqs_fixup_flags(unsigned long flags)
-{
-	if (raw_irqs_disabled_flags(flags))
-		trace_hardirqs_off();
-	else
-		trace_hardirqs_on();
-}
-
-static inline void trace_hardirqs_fixup(void)
-{
-	unsigned long flags = __raw_local_save_flags();
-
-	trace_hardirqs_fixup_flags(flags);
-}
-
 #else
 
 #ifdef CONFIG_X86_64
diff --git a/include/asm-x86/kdebug.h b/include/asm-x86/kdebug.h
index 5ec3ad3e825c..fbbab66ee9df 100644
--- a/include/asm-x86/kdebug.h
+++ b/include/asm-x86/kdebug.h
@@ -27,10 +27,9 @@ extern void printk_address(unsigned long address, int reliable);
 extern void die(const char *, struct pt_regs *,long);
 extern int __must_check __die(const char *, struct pt_regs *, long);
 extern void show_registers(struct pt_regs *regs);
-extern void __show_registers(struct pt_regs *, int all);
 extern void show_trace(struct task_struct *t, struct pt_regs *regs,
 		       unsigned long *sp, unsigned long bp);
-extern void __show_regs(struct pt_regs *regs);
+extern void __show_regs(struct pt_regs *regs, int all);
 extern void show_regs(struct pt_regs *regs);
 extern unsigned long oops_begin(void);
 extern void oops_end(unsigned long, struct pt_regs *, int signr);
diff --git a/include/asm-x86/kprobes.h b/include/asm-x86/kprobes.h
index bd8407863c13..8a0748d01036 100644
--- a/include/asm-x86/kprobes.h
+++ b/include/asm-x86/kprobes.h
@@ -82,15 +82,6 @@ struct kprobe_ctlblk {
 	struct prev_kprobe prev_kprobe;
 };
 
-/* trap3/1 are intr gates for kprobes.  So, restore the status of IF,
- * if necessary, before executing the original int3/1 (trap) handler.
- */
-static inline void restore_interrupts(struct pt_regs *regs)
-{
-	if (regs->flags & X86_EFLAGS_IF)
-		local_irq_enable();
-}
-
 extern int kprobe_fault_handler(struct pt_regs *regs, int trapnr);
 extern int kprobe_exceptions_notify(struct notifier_block *self,
 				    unsigned long val, void *data);
diff --git a/include/asm-x86/mach-bigsmp/mach_apicdef.h b/include/asm-x86/mach-bigsmp/mach_apicdef.h
deleted file mode 100644
index 811935d9d49b..000000000000
--- a/include/asm-x86/mach-bigsmp/mach_apicdef.h
+++ /dev/null
@@ -1,13 +0,0 @@
-#ifndef ASM_X86__MACH_BIGSMP__MACH_APICDEF_H
-#define ASM_X86__MACH_BIGSMP__MACH_APICDEF_H
-
-#define		APIC_ID_MASK		(0xFF<<24)
-
-static inline unsigned get_apic_id(unsigned long x) 
-{ 
-	return (((x)>>24)&0xFF);
-} 
-
-#define		GET_APIC_ID(x)	get_apic_id(x)
-
-#endif /* ASM_X86__MACH_BIGSMP__MACH_APICDEF_H */
diff --git a/include/asm-x86/mach-default/mach_apic.h b/include/asm-x86/mach-default/mach_apic.h
index b615f40736be..2a330a41b3dd 100644
--- a/include/asm-x86/mach-default/mach_apic.h
+++ b/include/asm-x86/mach-default/mach_apic.h
@@ -30,6 +30,8 @@ static inline cpumask_t target_cpus(void)
 #define cpu_mask_to_apicid (genapic->cpu_mask_to_apicid)
 #define phys_pkg_id	(genapic->phys_pkg_id)
 #define vector_allocation_domain    (genapic->vector_allocation_domain)
+#define read_apic_id()  (GET_APIC_ID(apic_read(APIC_ID)))
+#define send_IPI_self (genapic->send_IPI_self)
 extern void setup_apic_routing(void);
 #else
 #define INT_DELIVERY_MODE dest_LowestPrio
@@ -54,7 +56,7 @@ static inline void init_apic_ldr(void)
 
 static inline int apic_id_registered(void)
 {
-	return physid_isset(GET_APIC_ID(read_apic_id()), phys_cpu_present_map);
+	return physid_isset(read_apic_id(), phys_cpu_present_map);
 }
 
 static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask)
diff --git a/include/asm-x86/mach-default/mach_apicdef.h b/include/asm-x86/mach-default/mach_apicdef.h
index 936704f816d6..0c2d41c41b20 100644
--- a/include/asm-x86/mach-default/mach_apicdef.h
+++ b/include/asm-x86/mach-default/mach_apicdef.h
@@ -4,9 +4,9 @@
 #include <asm/apic.h>
 
 #ifdef CONFIG_X86_64
-#define	APIC_ID_MASK		(0xFFu<<24)
-#define GET_APIC_ID(x)          (((x)>>24)&0xFFu)
-#define	SET_APIC_ID(x)		(((x)<<24))
+#define	APIC_ID_MASK		(genapic->apic_id_mask)
+#define GET_APIC_ID(x)		(genapic->get_apic_id(x))
+#define	SET_APIC_ID(x)		(genapic->set_apic_id(x))
 #else
 #define		APIC_ID_MASK		(0xF<<24)
 static inline unsigned get_apic_id(unsigned long x) 
diff --git a/include/asm-x86/mach-default/mach_traps.h b/include/asm-x86/mach-default/mach_traps.h
index de9ac3f5c4ce..ff8778f26b84 100644
--- a/include/asm-x86/mach-default/mach_traps.h
+++ b/include/asm-x86/mach-default/mach_traps.h
@@ -7,12 +7,6 @@
 
 #include <asm/mc146818rtc.h>
 
-static inline void clear_mem_error(unsigned char reason)
-{
-	reason = (reason & 0xf) | 4;
-	outb(reason, 0x61);
-}
-
 static inline unsigned char get_nmi_reason(void)
 {
 	return inb(0x61);
diff --git a/include/asm-x86/mach-es7000/mach_apicdef.h b/include/asm-x86/mach-es7000/mach_apicdef.h
deleted file mode 100644
index a07e56744028..000000000000
--- a/include/asm-x86/mach-es7000/mach_apicdef.h
+++ /dev/null
@@ -1,13 +0,0 @@
-#ifndef ASM_X86__MACH_ES7000__MACH_APICDEF_H
-#define ASM_X86__MACH_ES7000__MACH_APICDEF_H
-
-#define		APIC_ID_MASK		(0xFF<<24)
-
-static inline unsigned get_apic_id(unsigned long x) 
-{ 
-	return (((x)>>24)&0xFF);
-} 
-
-#define		GET_APIC_ID(x)	get_apic_id(x)
-
-#endif /* ASM_X86__MACH_ES7000__MACH_APICDEF_H */
diff --git a/include/asm-x86/mach-numaq/mach_mpparse.h b/include/asm-x86/mach-numaq/mach_mpparse.h
deleted file mode 100644
index 74ade184920b..000000000000
--- a/include/asm-x86/mach-numaq/mach_mpparse.h
+++ /dev/null
@@ -1,7 +0,0 @@
-#ifndef ASM_X86__MACH_NUMAQ__MACH_MPPARSE_H
-#define ASM_X86__MACH_NUMAQ__MACH_MPPARSE_H
-
-extern void numaq_mps_oem_check(struct mp_config_table *mpc, char *oem,
-				char *productid);
-
-#endif /* ASM_X86__MACH_NUMAQ__MACH_MPPARSE_H */
diff --git a/include/asm-x86/mach-summit/mach_apicdef.h b/include/asm-x86/mach-summit/mach_apicdef.h
deleted file mode 100644
index d4bc8590c4f6..000000000000
--- a/include/asm-x86/mach-summit/mach_apicdef.h
+++ /dev/null
@@ -1,13 +0,0 @@
-#ifndef ASM_X86__MACH_SUMMIT__MACH_APICDEF_H
-#define ASM_X86__MACH_SUMMIT__MACH_APICDEF_H
-
-#define		APIC_ID_MASK		(0xFF<<24)
-
-static inline unsigned get_apic_id(unsigned long x) 
-{ 
-	return (((x)>>24)&0xFF);
-} 
-
-#define		GET_APIC_ID(x)	get_apic_id(x)
-
-#endif /* ASM_X86__MACH_SUMMIT__MACH_APICDEF_H */
diff --git a/include/asm-x86/microcode.h b/include/asm-x86/microcode.h
new file mode 100644
index 000000000000..62c793bb70ca
--- /dev/null
+++ b/include/asm-x86/microcode.h
@@ -0,0 +1,47 @@
+#ifndef ASM_X86__MICROCODE_H
+#define ASM_X86__MICROCODE_H
+
+struct cpu_signature {
+	unsigned int sig;
+	unsigned int pf;
+	unsigned int rev;
+};
+
+struct device;
+
+struct microcode_ops {
+	int  (*request_microcode_user) (int cpu, const void __user *buf, size_t size);
+	int  (*request_microcode_fw) (int cpu, struct device *device);
+
+	void (*apply_microcode) (int cpu);
+
+	int  (*collect_cpu_info) (int cpu, struct cpu_signature *csig);
+	void (*microcode_fini_cpu) (int cpu);
+};
+
+struct ucode_cpu_info {
+	struct cpu_signature cpu_sig;
+	int valid;
+	void *mc;
+};
+extern struct ucode_cpu_info ucode_cpu_info[];
+
+#ifdef CONFIG_MICROCODE_INTEL
+extern struct microcode_ops * __init init_intel_microcode(void);
+#else
+static inline struct microcode_ops * __init init_intel_microcode(void)
+{
+	return NULL;
+}
+#endif /* CONFIG_MICROCODE_INTEL */
+
+#ifdef CONFIG_MICROCODE_AMD
+extern struct microcode_ops * __init init_amd_microcode(void);
+#else
+static inline struct microcode_ops * __init init_amd_microcode(void)
+{
+	return NULL;
+}
+#endif
+
+#endif /* ASM_X86__MICROCODE_H */
diff --git a/include/asm-x86/mmzone_64.h b/include/asm-x86/mmzone_64.h
index 626b03a14875..6480f3333b2a 100644
--- a/include/asm-x86/mmzone_64.h
+++ b/include/asm-x86/mmzone_64.h
@@ -7,7 +7,7 @@
 
 #ifdef CONFIG_NUMA
 
-#define VIRTUAL_BUG_ON(x)
+#include <linux/mmdebug.h>
 
 #include <asm/smp.h>
 
@@ -29,7 +29,6 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr)
 {
 	unsigned nid;
 	VIRTUAL_BUG_ON(!memnodemap);
-	VIRTUAL_BUG_ON((addr >> memnode_shift) >= memnodemapsize);
 	nid = memnodemap[addr >> memnode_shift];
 	VIRTUAL_BUG_ON(nid >= MAX_NUMNODES || !node_data[nid]);
 	return nid;
diff --git a/include/asm-x86/module.h b/include/asm-x86/module.h
index 48dc3e0c07d9..864f2005fc1d 100644
--- a/include/asm-x86/module.h
+++ b/include/asm-x86/module.h
@@ -52,8 +52,6 @@ struct mod_arch_specific {};
 #define MODULE_PROC_FAMILY "EFFICEON "
 #elif defined CONFIG_MWINCHIPC6
 #define MODULE_PROC_FAMILY "WINCHIPC6 "
-#elif defined CONFIG_MWINCHIP2
-#define MODULE_PROC_FAMILY "WINCHIP2 "
 #elif defined CONFIG_MWINCHIP3D
 #define MODULE_PROC_FAMILY "WINCHIP3D "
 #elif defined CONFIG_MCYRIXIII
diff --git a/include/asm-x86/mpspec.h b/include/asm-x86/mpspec.h
index 118da365e371..be2241a818f1 100644
--- a/include/asm-x86/mpspec.h
+++ b/include/asm-x86/mpspec.h
@@ -5,11 +5,12 @@
 
 #include <asm/mpspec_def.h>
 
+extern int apic_version[MAX_APICS];
+
 #ifdef CONFIG_X86_32
 #include <mach_mpspec.h>
 
 extern unsigned int def_to_bigsmp;
-extern int apic_version[MAX_APICS];
 extern u8 apicid_2_node[];
 extern int pic_mode;
 
diff --git a/include/asm-x86/msidef.h b/include/asm-x86/msidef.h
index 3139666a94fa..ed9190246876 100644
--- a/include/asm-x86/msidef.h
+++ b/include/asm-x86/msidef.h
@@ -48,4 +48,8 @@
 #define  MSI_ADDR_DEST_ID(dest)		(((dest) << MSI_ADDR_DEST_ID_SHIFT) & \
 					 MSI_ADDR_DEST_ID_MASK)
 
+#define MSI_ADDR_IR_EXT_INT		(1 << 4)
+#define MSI_ADDR_IR_SHV			(1 << 3)
+#define MSI_ADDR_IR_INDEX1(index)	((index & 0x8000) >> 13)
+#define MSI_ADDR_IR_INDEX2(index)	((index & 0x7fff) << 5)
 #endif /* ASM_X86__MSIDEF_H */
diff --git a/include/asm-x86/msr-index.h b/include/asm-x86/msr-index.h
index 3052f058ab06..0bb43301a202 100644
--- a/include/asm-x86/msr-index.h
+++ b/include/asm-x86/msr-index.h
@@ -176,6 +176,7 @@
 #define MSR_IA32_TSC			0x00000010
 #define MSR_IA32_PLATFORM_ID		0x00000017
 #define MSR_IA32_EBL_CR_POWERON		0x0000002a
+#define MSR_IA32_FEATURE_CONTROL        0x0000003a
 
 #define MSR_IA32_APICBASE		0x0000001b
 #define MSR_IA32_APICBASE_BSP		(1<<8)
@@ -310,4 +311,19 @@
 /* Geode defined MSRs */
 #define MSR_GEODE_BUSCONT_CONF0		0x00001900
 
+/* Intel VT MSRs */
+#define MSR_IA32_VMX_BASIC              0x00000480
+#define MSR_IA32_VMX_PINBASED_CTLS      0x00000481
+#define MSR_IA32_VMX_PROCBASED_CTLS     0x00000482
+#define MSR_IA32_VMX_EXIT_CTLS          0x00000483
+#define MSR_IA32_VMX_ENTRY_CTLS         0x00000484
+#define MSR_IA32_VMX_MISC               0x00000485
+#define MSR_IA32_VMX_CR0_FIXED0         0x00000486
+#define MSR_IA32_VMX_CR0_FIXED1         0x00000487
+#define MSR_IA32_VMX_CR4_FIXED0         0x00000488
+#define MSR_IA32_VMX_CR4_FIXED1         0x00000489
+#define MSR_IA32_VMX_VMCS_ENUM          0x0000048a
+#define MSR_IA32_VMX_PROCBASED_CTLS2    0x0000048b
+#define MSR_IA32_VMX_EPT_VPID_CAP       0x0000048c
+
 #endif /* ASM_X86__MSR_INDEX_H */
diff --git a/include/asm-x86/nmi.h b/include/asm-x86/nmi.h
index d5e715f024dc..a53f829a97c5 100644
--- a/include/asm-x86/nmi.h
+++ b/include/asm-x86/nmi.h
@@ -15,10 +15,6 @@
  */
 int do_nmi_callback(struct pt_regs *regs, int cpu);
 
-#ifdef CONFIG_X86_64
-extern void default_do_nmi(struct pt_regs *);
-#endif
-
 extern void die_nmi(char *str, struct pt_regs *regs, int do_panic);
 extern int check_nmi_watchdog(void);
 extern int nmi_watchdog_enabled;
diff --git a/include/asm-x86/mach-numaq/mach_apic.h b/include/asm-x86/numaq/apic.h
index 7a0d39edfcfa..a8344ba6ea15 100644
--- a/include/asm-x86/mach-numaq/mach_apic.h
+++ b/include/asm-x86/numaq/apic.h
@@ -1,5 +1,5 @@
-#ifndef ASM_X86__MACH_NUMAQ__MACH_APIC_H
-#define ASM_X86__MACH_NUMAQ__MACH_APIC_H
+#ifndef __ASM_NUMAQ_APIC_H
+#define __ASM_NUMAQ_APIC_H
 
 #include <asm/io.h>
 #include <linux/mmzone.h>
@@ -135,4 +135,4 @@ static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb)
 	return cpuid_apic >> index_msb;
 }
 
-#endif /* ASM_X86__MACH_NUMAQ__MACH_APIC_H */
+#endif /* __ASM_NUMAQ_APIC_H */
diff --git a/include/asm-x86/mach-numaq/mach_apicdef.h b/include/asm-x86/numaq/apicdef.h
index f870ec5f7782..e012a46cc22a 100644
--- a/include/asm-x86/mach-numaq/mach_apicdef.h
+++ b/include/asm-x86/numaq/apicdef.h
@@ -1,5 +1,5 @@
-#ifndef ASM_X86__MACH_NUMAQ__MACH_APICDEF_H
-#define ASM_X86__MACH_NUMAQ__MACH_APICDEF_H
+#ifndef __ASM_NUMAQ_APICDEF_H
+#define __ASM_NUMAQ_APICDEF_H
 
 
 #define APIC_ID_MASK (0xF<<24)
@@ -11,4 +11,4 @@ static inline unsigned get_apic_id(unsigned long x)
 
 #define         GET_APIC_ID(x)  get_apic_id(x)
 
-#endif /* ASM_X86__MACH_NUMAQ__MACH_APICDEF_H */
+#endif
diff --git a/include/asm-x86/mach-numaq/mach_ipi.h b/include/asm-x86/numaq/ipi.h
index 1e835823f4bc..935588d286cf 100644
--- a/include/asm-x86/mach-numaq/mach_ipi.h
+++ b/include/asm-x86/numaq/ipi.h
@@ -1,5 +1,5 @@
-#ifndef ASM_X86__MACH_NUMAQ__MACH_IPI_H
-#define ASM_X86__MACH_NUMAQ__MACH_IPI_H
+#ifndef __ASM_NUMAQ_IPI_H
+#define __ASM_NUMAQ_IPI_H
 
 void send_IPI_mask_sequence(cpumask_t, int vector);
 
@@ -22,4 +22,4 @@ static inline void send_IPI_all(int vector)
 	send_IPI_mask(cpu_online_map, vector);
 }
 
-#endif /* ASM_X86__MACH_NUMAQ__MACH_IPI_H */
+#endif /* __ASM_NUMAQ_IPI_H */
diff --git a/include/asm-x86/numaq/mpparse.h b/include/asm-x86/numaq/mpparse.h
new file mode 100644
index 000000000000..252292e077b6
--- /dev/null
+++ b/include/asm-x86/numaq/mpparse.h
@@ -0,0 +1,7 @@
+#ifndef __ASM_NUMAQ_MPPARSE_H
+#define __ASM_NUMAQ_MPPARSE_H
+
+extern void numaq_mps_oem_check(struct mp_config_table *mpc, char *oem,
+				char *productid);
+
+#endif /* __ASM_NUMAQ_MPPARSE_H */
diff --git a/include/asm-x86/mach-numaq/mach_wakecpu.h b/include/asm-x86/numaq/wakecpu.h
index 0db8cea643c0..c577bda5b1c5 100644
--- a/include/asm-x86/mach-numaq/mach_wakecpu.h
+++ b/include/asm-x86/numaq/wakecpu.h
@@ -1,5 +1,5 @@
-#ifndef ASM_X86__MACH_NUMAQ__MACH_WAKECPU_H
-#define ASM_X86__MACH_NUMAQ__MACH_WAKECPU_H
+#ifndef __ASM_NUMAQ_WAKECPU_H
+#define __ASM_NUMAQ_WAKECPU_H
 
 /* This file copes with machines that wakeup secondary CPUs by NMIs */
 
@@ -40,4 +40,4 @@ static inline void restore_NMI_vector(unsigned short *high, unsigned short *low)
 
 #define inquire_remote_apic(apicid) {}
 
-#endif /* ASM_X86__MACH_NUMAQ__MACH_WAKECPU_H */
+#endif /* __ASM_NUMAQ_WAKECPU_H */
diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index c91574776751..d4f1d5791fc1 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -179,6 +179,7 @@ static inline pteval_t native_pte_flags(pte_t pte)
 #endif	/* CONFIG_PARAVIRT */
 
 #define __pa(x)		__phys_addr((unsigned long)(x))
+#define __pa_nodebug(x)	__phys_addr_nodebug((unsigned long)(x))
 /* __pa_symbol should be used for C visible symbols.
    This seems to be the official gcc blessed way to do such arithmetic. */
 #define __pa_symbol(x)	__pa(__phys_reloc_hide((unsigned long)(x)))
@@ -188,9 +189,14 @@ static inline pteval_t native_pte_flags(pte_t pte)
 #define __boot_va(x)		__va(x)
 #define __boot_pa(x)		__pa(x)
 
+/*
+ * virt_to_page(kaddr) returns a valid pointer if and only if
+ * virt_addr_valid(kaddr) returns true.
+ */
 #define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
 #define pfn_to_kaddr(pfn)      __va((pfn) << PAGE_SHIFT)
-#define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
+extern bool __virt_addr_valid(unsigned long kaddr);
+#define virt_addr_valid(kaddr)	__virt_addr_valid((unsigned long) (kaddr))
 
 #endif	/* __ASSEMBLY__ */
 
diff --git a/include/asm-x86/page_32.h b/include/asm-x86/page_32.h
index 72f7305682c6..e8d80d1de237 100644
--- a/include/asm-x86/page_32.h
+++ b/include/asm-x86/page_32.h
@@ -20,6 +20,12 @@
 #endif
 #define THREAD_SIZE 	(PAGE_SIZE << THREAD_ORDER)
 
+#define STACKFAULT_STACK 0
+#define DOUBLEFAULT_STACK 1
+#define NMI_STACK 0
+#define DEBUG_STACK 0
+#define MCE_STACK 0
+#define N_EXCEPTION_STACKS 1
 
 #ifdef CONFIG_X86_PAE
 /* 44=32+12, the limit we can fit into an unsigned long pfn */
@@ -73,7 +79,12 @@ typedef struct page *pgtable_t;
 #endif
 
 #ifndef __ASSEMBLY__
-#define __phys_addr(x)		((x) - PAGE_OFFSET)
+#define __phys_addr_nodebug(x)	((x) - PAGE_OFFSET)
+#ifdef CONFIG_DEBUG_VIRTUAL
+extern unsigned long __phys_addr(unsigned long);
+#else
+#define __phys_addr(x)		__phys_addr_nodebug(x)
+#endif
 #define __phys_reloc_hide(x)	RELOC_HIDE((x), 0)
 
 #ifdef CONFIG_FLATMEM
diff --git a/include/asm-x86/paravirt.h b/include/asm-x86/paravirt.h
index 891971f57d35..8d6ae2f760d0 100644
--- a/include/asm-x86/paravirt.h
+++ b/include/asm-x86/paravirt.h
@@ -124,6 +124,9 @@ struct pv_cpu_ops {
 				int entrynum, const void *desc, int size);
 	void (*write_idt_entry)(gate_desc *,
 				int entrynum, const gate_desc *gate);
+	void (*alloc_ldt)(struct desc_struct *ldt, unsigned entries);
+	void (*free_ldt)(struct desc_struct *ldt, unsigned entries);
+
 	void (*load_sp0)(struct tss_struct *tss, struct thread_struct *t);
 
 	void (*set_iopl_mask)(unsigned mask);
@@ -201,12 +204,6 @@ struct pv_irq_ops {
 
 struct pv_apic_ops {
 #ifdef CONFIG_X86_LOCAL_APIC
-	/*
-	 * Direct APIC operations, principally for VMI.  Ideally
-	 * these shouldn't be in this interface.
-	 */
-	void (*apic_write)(unsigned long reg, u32 v);
-	u32 (*apic_read)(unsigned long reg);
 	void (*setup_boot_clock)(void);
 	void (*setup_secondary_clock)(void);
 
@@ -331,6 +328,7 @@ struct pv_lock_ops {
 	int (*spin_is_locked)(struct raw_spinlock *lock);
 	int (*spin_is_contended)(struct raw_spinlock *lock);
 	void (*spin_lock)(struct raw_spinlock *lock);
+	void (*spin_lock_flags)(struct raw_spinlock *lock, unsigned long flags);
 	int (*spin_trylock)(struct raw_spinlock *lock);
 	void (*spin_unlock)(struct raw_spinlock *lock);
 };
@@ -836,6 +834,16 @@ do {							\
 	(aux) = __aux;					\
 } while (0)
 
+static inline void paravirt_alloc_ldt(struct desc_struct *ldt, unsigned entries)
+{
+	PVOP_VCALL2(pv_cpu_ops.alloc_ldt, ldt, entries);
+}
+
+static inline void paravirt_free_ldt(struct desc_struct *ldt, unsigned entries)
+{
+	PVOP_VCALL2(pv_cpu_ops.free_ldt, ldt, entries);
+}
+
 static inline void load_TR_desc(void)
 {
 	PVOP_VCALL0(pv_cpu_ops.load_tr_desc);
@@ -910,19 +918,6 @@ static inline void slow_down_io(void)
 }
 
 #ifdef CONFIG_X86_LOCAL_APIC
-/*
- * Basic functions accessing APICs.
- */
-static inline void apic_write(unsigned long reg, u32 v)
-{
-	PVOP_VCALL2(pv_apic_ops.apic_write, reg, v);
-}
-
-static inline u32 apic_read(unsigned long reg)
-{
-	return PVOP_CALL1(unsigned long, pv_apic_ops.apic_read, reg);
-}
-
 static inline void setup_boot_clock(void)
 {
 	PVOP_VCALL0(pv_apic_ops.setup_boot_clock);
@@ -1413,6 +1408,12 @@ static __always_inline void __raw_spin_lock(struct raw_spinlock *lock)
 	PVOP_VCALL1(pv_lock_ops.spin_lock, lock);
 }
 
+static __always_inline void __raw_spin_lock_flags(struct raw_spinlock *lock,
+						  unsigned long flags)
+{
+	PVOP_VCALL2(pv_lock_ops.spin_lock_flags, lock, flags);
+}
+
 static __always_inline int __raw_spin_trylock(struct raw_spinlock *lock)
 {
 	return PVOP_CALL1(int, pv_lock_ops.spin_trylock, lock);
diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index ed932453ef26..182f9d4c570f 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -15,7 +15,7 @@
 #define _PAGE_BIT_PAT		7	/* on 4KB pages */
 #define _PAGE_BIT_GLOBAL	8	/* Global TLB entry PPro+ */
 #define _PAGE_BIT_UNUSED1	9	/* available for programmer */
-#define _PAGE_BIT_UNUSED2	10
+#define _PAGE_BIT_IOMAP		10	/* flag used to indicate IO mapping */
 #define _PAGE_BIT_UNUSED3	11
 #define _PAGE_BIT_PAT_LARGE	12	/* On 2MB or 1GB pages */
 #define _PAGE_BIT_SPECIAL	_PAGE_BIT_UNUSED1
@@ -32,7 +32,7 @@
 #define _PAGE_PSE	(_AT(pteval_t, 1) << _PAGE_BIT_PSE)
 #define _PAGE_GLOBAL	(_AT(pteval_t, 1) << _PAGE_BIT_GLOBAL)
 #define _PAGE_UNUSED1	(_AT(pteval_t, 1) << _PAGE_BIT_UNUSED1)
-#define _PAGE_UNUSED2	(_AT(pteval_t, 1) << _PAGE_BIT_UNUSED2)
+#define _PAGE_IOMAP	(_AT(pteval_t, 1) << _PAGE_BIT_IOMAP)
 #define _PAGE_UNUSED3	(_AT(pteval_t, 1) << _PAGE_BIT_UNUSED3)
 #define _PAGE_PAT	(_AT(pteval_t, 1) << _PAGE_BIT_PAT)
 #define _PAGE_PAT_LARGE (_AT(pteval_t, 1) << _PAGE_BIT_PAT_LARGE)
@@ -99,6 +99,11 @@
 #define __PAGE_KERNEL_LARGE_NOCACHE	(__PAGE_KERNEL | _PAGE_CACHE_UC | _PAGE_PSE)
 #define __PAGE_KERNEL_LARGE_EXEC	(__PAGE_KERNEL_EXEC | _PAGE_PSE)
 
+#define __PAGE_KERNEL_IO		(__PAGE_KERNEL | _PAGE_IOMAP)
+#define __PAGE_KERNEL_IO_NOCACHE	(__PAGE_KERNEL_NOCACHE | _PAGE_IOMAP)
+#define __PAGE_KERNEL_IO_UC_MINUS	(__PAGE_KERNEL_UC_MINUS | _PAGE_IOMAP)
+#define __PAGE_KERNEL_IO_WC		(__PAGE_KERNEL_WC | _PAGE_IOMAP)
+
 #define PAGE_KERNEL			__pgprot(__PAGE_KERNEL)
 #define PAGE_KERNEL_RO			__pgprot(__PAGE_KERNEL_RO)
 #define PAGE_KERNEL_EXEC		__pgprot(__PAGE_KERNEL_EXEC)
@@ -113,6 +118,11 @@
 #define PAGE_KERNEL_VSYSCALL		__pgprot(__PAGE_KERNEL_VSYSCALL)
 #define PAGE_KERNEL_VSYSCALL_NOCACHE	__pgprot(__PAGE_KERNEL_VSYSCALL_NOCACHE)
 
+#define PAGE_KERNEL_IO			__pgprot(__PAGE_KERNEL_IO)
+#define PAGE_KERNEL_IO_NOCACHE		__pgprot(__PAGE_KERNEL_IO_NOCACHE)
+#define PAGE_KERNEL_IO_UC_MINUS		__pgprot(__PAGE_KERNEL_IO_UC_MINUS)
+#define PAGE_KERNEL_IO_WC		__pgprot(__PAGE_KERNEL_IO_WC)
+
 /*         xwr */
 #define __P000	PAGE_NONE
 #define __P001	PAGE_READONLY
@@ -196,7 +206,7 @@ static inline int pte_exec(pte_t pte)
 
 static inline int pte_special(pte_t pte)
 {
-	return pte_val(pte) & _PAGE_SPECIAL;
+	return pte_flags(pte) & _PAGE_SPECIAL;
 }
 
 static inline unsigned long pte_pfn(pte_t pte)
diff --git a/include/asm-x86/processor-cyrix.h b/include/asm-x86/processor-cyrix.h
index 97568ada1f97..1198f2a0e42c 100644
--- a/include/asm-x86/processor-cyrix.h
+++ b/include/asm-x86/processor-cyrix.h
@@ -28,3 +28,11 @@ static inline void setCx86(u8 reg, u8 data)
 	outb(reg, 0x22);
 	outb(data, 0x23);
 }
+
+#define getCx86_old(reg) ({ outb((reg), 0x22); inb(0x23); })
+
+#define setCx86_old(reg, data) do { \
+	outb((reg), 0x22); \
+	outb((data), 0x23); \
+} while (0)
+
diff --git a/include/asm-x86/processor-flags.h b/include/asm-x86/processor-flags.h
index 5dd79774f693..dc5f0712f9fa 100644
--- a/include/asm-x86/processor-flags.h
+++ b/include/asm-x86/processor-flags.h
@@ -59,6 +59,7 @@
 #define X86_CR4_OSFXSR	0x00000200 /* enable fast FPU save and restore */
 #define X86_CR4_OSXMMEXCPT 0x00000400 /* enable unmasked SSE exceptions */
 #define X86_CR4_VMXE	0x00002000 /* enable VMX virtualization */
+#define X86_CR4_OSXSAVE 0x00040000 /* enable xsave and xrestore */
 
 /*
  * x86-64 Task Priority Register, CR8
diff --git a/include/asm-x86/processor.h b/include/asm-x86/processor.h
index 5eaf9bf0a623..ee7cbb30773a 100644
--- a/include/asm-x86/processor.h
+++ b/include/asm-x86/processor.h
@@ -76,11 +76,11 @@ struct cpuinfo_x86 {
 	int			 x86_tlbsize;
 	__u8			x86_virt_bits;
 	__u8			x86_phys_bits;
+#endif
 	/* CPUID returned core id bits: */
 	__u8			x86_coreid_bits;
 	/* Max extended CPUID function supported: */
 	__u32			extended_cpuid_level;
-#endif
 	/* Maximum supported CPUID level, -1=no CPUID: */
 	int			cpuid_level;
 	__u32			x86_capability[NCAPINTS];
@@ -166,11 +166,8 @@ extern void init_scattered_cpuid_features(struct cpuinfo_x86 *c);
 extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
 extern unsigned short num_cache_leaves;
 
-#if defined(CONFIG_X86_HT) || defined(CONFIG_X86_64)
+extern void detect_extended_topology(struct cpuinfo_x86 *c);
 extern void detect_ht(struct cpuinfo_x86 *c);
-#else
-static inline void detect_ht(struct cpuinfo_x86 *c) {}
-#endif
 
 static inline void native_cpuid(unsigned int *eax, unsigned int *ebx,
 				unsigned int *ecx, unsigned int *edx)
@@ -327,7 +324,12 @@ struct i387_fxsave_struct {
 	/* 16*16 bytes for each XMM-reg = 256 bytes:			*/
 	u32			xmm_space[64];
 
-	u32			padding[24];
+	u32			padding[12];
+
+	union {
+		u32		padding1[12];
+		u32		sw_reserved[12];
+	};
 
 } __attribute__((aligned(16)));
 
@@ -351,10 +353,23 @@ struct i387_soft_struct {
 	u32			entry_eip;
 };
 
+struct xsave_hdr_struct {
+	u64 xstate_bv;
+	u64 reserved1[2];
+	u64 reserved2[5];
+} __attribute__((packed));
+
+struct xsave_struct {
+	struct i387_fxsave_struct i387;
+	struct xsave_hdr_struct xsave_hdr;
+	/* new processor state extensions will go here */
+} __attribute__ ((packed, aligned (64)));
+
 union thread_xstate {
 	struct i387_fsave_struct	fsave;
 	struct i387_fxsave_struct	fxsave;
 	struct i387_soft_struct		soft;
+	struct xsave_struct		xsave;
 };
 
 #ifdef CONFIG_X86_64
@@ -571,41 +586,6 @@ static inline void clear_in_cr4(unsigned long mask)
 	write_cr4(cr4);
 }
 
-struct microcode_header {
-	unsigned int		hdrver;
-	unsigned int		rev;
-	unsigned int		date;
-	unsigned int		sig;
-	unsigned int		cksum;
-	unsigned int		ldrver;
-	unsigned int		pf;
-	unsigned int		datasize;
-	unsigned int		totalsize;
-	unsigned int		reserved[3];
-};
-
-struct microcode {
-	struct microcode_header	hdr;
-	unsigned int		bits[0];
-};
-
-typedef struct microcode	microcode_t;
-typedef struct microcode_header	microcode_header_t;
-
-/* microcode format is extended from prescott processors */
-struct extended_signature {
-	unsigned int		sig;
-	unsigned int		pf;
-	unsigned int		cksum;
-};
-
-struct extended_sigtable {
-	unsigned int		count;
-	unsigned int		cksum;
-	unsigned int		reserved[3];
-	struct extended_signature sigs[0];
-};
-
 typedef struct {
 	unsigned long		seg;
 } mm_segment_t;
diff --git a/include/asm-x86/ptrace.h b/include/asm-x86/ptrace.h
index d64a61097165..a2025525a15a 100644
--- a/include/asm-x86/ptrace.h
+++ b/include/asm-x86/ptrace.h
@@ -174,13 +174,9 @@ extern unsigned long profile_pc(struct pt_regs *regs);
 
 extern unsigned long
 convert_ip_to_linear(struct task_struct *child, struct pt_regs *regs);
-
-#ifdef CONFIG_X86_32
 extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs,
-			 int error_code);
-#else
+			 int error_code, int si_code);
 void signal_fault(struct pt_regs *regs, void __user *frame, char *where);
-#endif
 
 extern long syscall_trace_enter(struct pt_regs *);
 extern void syscall_trace_leave(struct pt_regs *);
diff --git a/include/asm-x86/segment.h b/include/asm-x86/segment.h
index ea5f0a8686f7..5d6e69454891 100644
--- a/include/asm-x86/segment.h
+++ b/include/asm-x86/segment.h
@@ -131,12 +131,6 @@
  * Matching rules for certain types of segments.
  */
 
-/* Matches only __KERNEL_CS, ignoring PnP / USER / APM segments */
-#define SEGMENT_IS_KERNEL_CODE(x) (((x) & 0xfc) == GDT_ENTRY_KERNEL_CS * 8)
-
-/* Matches __KERNEL_CS and __USER_CS (they must be 2 entries apart) */
-#define SEGMENT_IS_FLAT_CODE(x)  (((x) & 0xec) == GDT_ENTRY_KERNEL_CS * 8)
-
 /* Matches PNP_CS32 and PNP_CS16 (they must be consecutive) */
 #define SEGMENT_IS_PNP_CODE(x)   (((x) & 0xf4) == GDT_ENTRY_PNPBIOS_BASE * 8)
 
diff --git a/include/asm-x86/setup.h b/include/asm-x86/setup.h
index 9030cb73c4d7..11b6cc14b289 100644
--- a/include/asm-x86/setup.h
+++ b/include/asm-x86/setup.h
@@ -38,6 +38,7 @@ struct x86_quirks {
 	void (*mpc_oem_pci_bus)(struct mpc_config_bus *m);
 	void (*smp_read_mpc_oem)(struct mp_config_oemtable *oemtable,
                                     unsigned short oemsize);
+	int (*setup_ioapic_ids)(void);
 };
 
 extern struct x86_quirks *x86_quirks;
diff --git a/include/asm-x86/sigcontext.h b/include/asm-x86/sigcontext.h
index 24879c85b291..ee813f4fe5d5 100644
--- a/include/asm-x86/sigcontext.h
+++ b/include/asm-x86/sigcontext.h
@@ -4,6 +4,40 @@
 #include <linux/compiler.h>
 #include <asm/types.h>
 
+#define FP_XSTATE_MAGIC1	0x46505853U
+#define FP_XSTATE_MAGIC2	0x46505845U
+#define FP_XSTATE_MAGIC2_SIZE	sizeof(FP_XSTATE_MAGIC2)
+
+/*
+ * bytes 464..511 in the current 512byte layout of fxsave/fxrstor frame
+ * are reserved for SW usage. On cpu's supporting xsave/xrstor, these bytes
+ * are used to extended the fpstate pointer in the sigcontext, which now
+ * includes the extended state information along with fpstate information.
+ *
+ * Presence of FP_XSTATE_MAGIC1 at the beginning of this SW reserved
+ * area and FP_XSTATE_MAGIC2 at the end of memory layout
+ * (extended_size - FP_XSTATE_MAGIC2_SIZE) indicates the presence of the
+ * extended state information in the memory layout pointed by the fpstate
+ * pointer in sigcontext.
+ */
+struct _fpx_sw_bytes {
+	__u32 magic1;		/* FP_XSTATE_MAGIC1 */
+	__u32 extended_size;	/* total size of the layout referred by
+				 * fpstate pointer in the sigcontext.
+				 */
+	__u64 xstate_bv;
+				/* feature bit mask (including fp/sse/extended
+				 * state) that is present in the memory
+				 * layout.
+				 */
+	__u32 xstate_size;	/* actual xsave state size, based on the
+				 * features saved in the layout.
+				 * 'extended_size' will be greater than
+				 * 'xstate_size'.
+				 */
+	__u32 padding[7];	/*  for future use. */
+};
+
 #ifdef __i386__
 /*
  * As documented in the iBCS2 standard..
@@ -53,7 +87,13 @@ struct _fpstate {
 	unsigned long	reserved;
 	struct _fpxreg	_fxsr_st[8];	/* FXSR FPU reg data is ignored */
 	struct _xmmreg	_xmm[8];
-	unsigned long	padding[56];
+	unsigned long	padding1[44];
+
+	union {
+		unsigned long	padding2[12];
+		struct _fpx_sw_bytes sw_reserved; /* represents the extended
+						   * state info */
+	};
 };
 
 #define X86_FXSR_MAGIC		0x0000
@@ -79,7 +119,15 @@ struct sigcontext {
 	unsigned long flags;
 	unsigned long sp_at_signal;
 	unsigned short ss, __ssh;
-	struct _fpstate __user *fpstate;
+
+	/*
+	 * fpstate is really (struct _fpstate *) or (struct _xstate *)
+	 * depending on the FP_XSTATE_MAGIC1 encoded in the SW reserved
+	 * bytes of (struct _fpstate) and FP_XSTATE_MAGIC2 present at the end
+	 * of extended memory layout. See comments at the defintion of
+	 * (struct _fpx_sw_bytes)
+	 */
+	void __user *fpstate;		/* zero when no FPU/extended context */
 	unsigned long oldmask;
 	unsigned long cr2;
 };
@@ -130,7 +178,12 @@ struct _fpstate {
 	__u32	mxcsr_mask;
 	__u32	st_space[32];	/* 8*16 bytes for each FP-reg */
 	__u32	xmm_space[64];	/* 16*16 bytes for each XMM-reg  */
-	__u32	reserved2[24];
+	__u32	reserved2[12];
+	union {
+		__u32	reserved3[12];
+		struct _fpx_sw_bytes sw_reserved; /* represents the extended
+						   * state information */
+	};
 };
 
 #ifdef __KERNEL__
@@ -161,7 +214,15 @@ struct sigcontext {
 	unsigned long trapno;
 	unsigned long oldmask;
 	unsigned long cr2;
-	struct _fpstate __user *fpstate;	/* zero when no FPU context */
+
+	/*
+	 * fpstate is really (struct _fpstate *) or (struct _xstate *)
+	 * depending on the FP_XSTATE_MAGIC1 encoded in the SW reserved
+	 * bytes of (struct _fpstate) and FP_XSTATE_MAGIC2 present at the end
+	 * of extended memory layout. See comments at the defintion of
+	 * (struct _fpx_sw_bytes)
+	 */
+	void __user *fpstate;		/* zero when no FPU/extended context */
 	unsigned long reserved1[8];
 };
 #else /* __KERNEL__ */
@@ -202,4 +263,22 @@ struct sigcontext {
 
 #endif /* !__i386__ */
 
+struct _xsave_hdr {
+	__u64 xstate_bv;
+	__u64 reserved1[2];
+	__u64 reserved2[5];
+};
+
+/*
+ * Extended state pointed by the fpstate pointer in the sigcontext.
+ * In addition to the fpstate, information encoded in the xstate_hdr
+ * indicates the presence of other extended state information
+ * supported by the processor and OS.
+ */
+struct _xstate {
+	struct _fpstate fpstate;
+	struct _xsave_hdr xstate_hdr;
+	/* new processor state extensions go here */
+};
+
 #endif /* ASM_X86__SIGCONTEXT_H */
diff --git a/include/asm-x86/sigcontext32.h b/include/asm-x86/sigcontext32.h
index 4e2ec732dd01..8c347032c2f2 100644
--- a/include/asm-x86/sigcontext32.h
+++ b/include/asm-x86/sigcontext32.h
@@ -40,7 +40,11 @@ struct _fpstate_ia32 {
 	__u32	reserved;
 	struct _fpxreg	_fxsr_st[8];
 	struct _xmmreg	_xmm[8];	/* It's actually 16 */
-	__u32	padding[56];
+	__u32	padding[44];
+	union {
+		__u32 padding2[12];
+		struct _fpx_sw_bytes sw_reserved;
+	};
 };
 
 struct sigcontext_ia32 {
diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h
index 04f84f4e2c8b..a6afc29f2dd9 100644
--- a/include/asm-x86/smp.h
+++ b/include/asm-x86/smp.h
@@ -50,12 +50,16 @@ extern struct {
 struct smp_ops {
 	void (*smp_prepare_boot_cpu)(void);
 	void (*smp_prepare_cpus)(unsigned max_cpus);
-	int (*cpu_up)(unsigned cpu);
 	void (*smp_cpus_done)(unsigned max_cpus);
 
 	void (*smp_send_stop)(void);
 	void (*smp_send_reschedule)(int cpu);
 
+	int (*cpu_up)(unsigned cpu);
+	int (*cpu_disable)(void);
+	void (*cpu_die)(unsigned int cpu);
+	void (*play_dead)(void);
+
 	void (*send_call_func_ipi)(cpumask_t mask);
 	void (*send_call_func_single_ipi)(int cpu);
 };
@@ -94,6 +98,21 @@ static inline int __cpu_up(unsigned int cpu)
 	return smp_ops.cpu_up(cpu);
 }
 
+static inline int __cpu_disable(void)
+{
+	return smp_ops.cpu_disable();
+}
+
+static inline void __cpu_die(unsigned int cpu)
+{
+	smp_ops.cpu_die(cpu);
+}
+
+static inline void play_dead(void)
+{
+	smp_ops.play_dead();
+}
+
 static inline void smp_send_reschedule(int cpu)
 {
 	smp_ops.smp_send_reschedule(cpu);
@@ -109,15 +128,20 @@ static inline void arch_send_call_function_ipi(cpumask_t mask)
 	smp_ops.send_call_func_ipi(mask);
 }
 
+void cpu_disable_common(void);
 void native_smp_prepare_boot_cpu(void);
 void native_smp_prepare_cpus(unsigned int max_cpus);
 void native_smp_cpus_done(unsigned int max_cpus);
 int native_cpu_up(unsigned int cpunum);
+int native_cpu_disable(void);
+void native_cpu_die(unsigned int cpu);
+void native_play_dead(void);
+void play_dead_common(void);
+
 void native_send_call_func_ipi(cpumask_t mask);
 void native_send_call_func_single_ipi(int cpu);
 
-extern int __cpu_disable(void);
-extern void __cpu_die(unsigned int cpu);
+extern void prefill_possible_map(void);
 
 void smp_store_cpu_info(int id);
 #define cpu_physical_id(cpu)	per_cpu(x86_cpu_to_apicid, cpu)
@@ -127,15 +151,11 @@ static inline int num_booting_cpus(void)
 {
 	return cpus_weight(cpu_callout_map);
 }
-#endif /* CONFIG_SMP */
-
-#if defined(CONFIG_SMP) && defined(CONFIG_HOTPLUG_CPU)
-extern void prefill_possible_map(void);
 #else
 static inline void prefill_possible_map(void)
 {
 }
-#endif
+#endif /* CONFIG_SMP */
 
 extern unsigned disabled_cpus __cpuinitdata;
 
@@ -167,30 +187,33 @@ extern int safe_smp_processor_id(void);
 
 #ifdef CONFIG_X86_LOCAL_APIC
 
+#ifndef CONFIG_X86_64
 static inline int logical_smp_processor_id(void)
 {
 	/* we don't want to mark this access volatile - bad code generation */
 	return GET_APIC_LOGICAL_ID(*(u32 *)(APIC_BASE + APIC_LDR));
 }
 
-#ifndef CONFIG_X86_64
+#include <mach_apicdef.h>
 static inline unsigned int read_apic_id(void)
 {
-	return *(u32 *)(APIC_BASE + APIC_ID);
+	unsigned int reg;
+
+	reg = *(u32 *)(APIC_BASE + APIC_ID);
+
+	return GET_APIC_ID(reg);
 }
-#else
-extern unsigned int read_apic_id(void);
 #endif
 
 
-# ifdef APIC_DEFINITION
+# if defined(APIC_DEFINITION) || defined(CONFIG_X86_64)
 extern int hard_smp_processor_id(void);
 # else
-#  include <mach_apicdef.h>
+#include <mach_apicdef.h>
 static inline int hard_smp_processor_id(void)
 {
 	/* we don't want to mark this access volatile - bad code generation */
-	return GET_APIC_ID(read_apic_id());
+	return read_apic_id();
 }
 # endif /* APIC_DEFINITION */
 
@@ -202,9 +225,5 @@ static inline int hard_smp_processor_id(void)
 
 #endif /* CONFIG_X86_LOCAL_APIC */
 
-#ifdef CONFIG_HOTPLUG_CPU
-extern void cpu_uninit(void);
-#endif
-
 #endif /* __ASSEMBLY__ */
 #endif /* ASM_X86__SMP_H */
diff --git a/include/asm-x86/spinlock.h b/include/asm-x86/spinlock.h
index 93adae338ac6..157ff7fab97a 100644
--- a/include/asm-x86/spinlock.h
+++ b/include/asm-x86/spinlock.h
@@ -21,8 +21,10 @@
 
 #ifdef CONFIG_X86_32
 # define LOCK_PTR_REG "a"
+# define REG_PTR_MODE "k"
 #else
 # define LOCK_PTR_REG "D"
+# define REG_PTR_MODE "q"
 #endif
 
 #if defined(CONFIG_X86_32) && \
@@ -54,19 +56,7 @@
  * much between them in performance though, especially as locks are out of line.
  */
 #if (NR_CPUS < 256)
-static inline int __ticket_spin_is_locked(raw_spinlock_t *lock)
-{
-	int tmp = ACCESS_ONCE(lock->slock);
-
-	return (((tmp >> 8) & 0xff) != (tmp & 0xff));
-}
-
-static inline int __ticket_spin_is_contended(raw_spinlock_t *lock)
-{
-	int tmp = ACCESS_ONCE(lock->slock);
-
-	return (((tmp >> 8) - tmp) & 0xff) > 1;
-}
+#define TICKET_SHIFT 8
 
 static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock)
 {
@@ -89,19 +79,17 @@ static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock)
 
 static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock)
 {
-	int tmp;
-	short new;
+	int tmp, new;
 
-	asm volatile("movw %2,%w0\n\t"
+	asm volatile("movzwl %2, %0\n\t"
 		     "cmpb %h0,%b0\n\t"
+		     "leal 0x100(%" REG_PTR_MODE "0), %1\n\t"
 		     "jne 1f\n\t"
-		     "movw %w0,%w1\n\t"
-		     "incb %h1\n\t"
 		     LOCK_PREFIX "cmpxchgw %w1,%2\n\t"
 		     "1:"
 		     "sete %b1\n\t"
 		     "movzbl %b1,%0\n\t"
-		     : "=&a" (tmp), "=Q" (new), "+m" (lock->slock)
+		     : "=&a" (tmp), "=&q" (new), "+m" (lock->slock)
 		     :
 		     : "memory", "cc");
 
@@ -116,19 +104,7 @@ static __always_inline void __ticket_spin_unlock(raw_spinlock_t *lock)
 		     : "memory", "cc");
 }
 #else
-static inline int __ticket_spin_is_locked(raw_spinlock_t *lock)
-{
-	int tmp = ACCESS_ONCE(lock->slock);
-
-	return (((tmp >> 16) & 0xffff) != (tmp & 0xffff));
-}
-
-static inline int __ticket_spin_is_contended(raw_spinlock_t *lock)
-{
-	int tmp = ACCESS_ONCE(lock->slock);
-
-	return (((tmp >> 16) - tmp) & 0xffff) > 1;
-}
+#define TICKET_SHIFT 16
 
 static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock)
 {
@@ -146,7 +122,7 @@ static __always_inline void __ticket_spin_lock(raw_spinlock_t *lock)
 		     /* don't need lfence here, because loads are in-order */
 		     "jmp 1b\n"
 		     "2:"
-		     : "+Q" (inc), "+m" (lock->slock), "=r" (tmp)
+		     : "+r" (inc), "+m" (lock->slock), "=&r" (tmp)
 		     :
 		     : "memory", "cc");
 }
@@ -160,13 +136,13 @@ static __always_inline int __ticket_spin_trylock(raw_spinlock_t *lock)
 		     "movl %0,%1\n\t"
 		     "roll $16, %0\n\t"
 		     "cmpl %0,%1\n\t"
+		     "leal 0x00010000(%" REG_PTR_MODE "0), %1\n\t"
 		     "jne 1f\n\t"
-		     "addl $0x00010000, %1\n\t"
 		     LOCK_PREFIX "cmpxchgl %1,%2\n\t"
 		     "1:"
 		     "sete %b1\n\t"
 		     "movzbl %b1,%0\n\t"
-		     : "=&a" (tmp), "=r" (new), "+m" (lock->slock)
+		     : "=&a" (tmp), "=&q" (new), "+m" (lock->slock)
 		     :
 		     : "memory", "cc");
 
@@ -182,7 +158,19 @@ static __always_inline void __ticket_spin_unlock(raw_spinlock_t *lock)
 }
 #endif
 
-#define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
+static inline int __ticket_spin_is_locked(raw_spinlock_t *lock)
+{
+	int tmp = ACCESS_ONCE(lock->slock);
+
+	return !!(((tmp >> TICKET_SHIFT) ^ tmp) & ((1 << TICKET_SHIFT) - 1));
+}
+
+static inline int __ticket_spin_is_contended(raw_spinlock_t *lock)
+{
+	int tmp = ACCESS_ONCE(lock->slock);
+
+	return (((tmp >> TICKET_SHIFT) - tmp) & ((1 << TICKET_SHIFT) - 1)) > 1;
+}
 
 #ifdef CONFIG_PARAVIRT
 /*
@@ -272,6 +260,13 @@ static __always_inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
 	__ticket_spin_unlock(lock);
 }
+
+static __always_inline void __raw_spin_lock_flags(raw_spinlock_t *lock,
+						  unsigned long flags)
+{
+	__raw_spin_lock(lock);
+}
+
 #endif	/* CONFIG_PARAVIRT */
 
 static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
diff --git a/include/asm-x86/statfs.h b/include/asm-x86/statfs.h
index 3f005bc3aa5b..ca5dc19dd461 100644
--- a/include/asm-x86/statfs.h
+++ b/include/asm-x86/statfs.h
@@ -1,63 +1,12 @@
 #ifndef ASM_X86__STATFS_H
 #define ASM_X86__STATFS_H
 
-#ifdef __i386__
-#include <asm-generic/statfs.h>
-#else
-
-#ifndef __KERNEL_STRICT_NAMES
-
-#include <linux/types.h>
-
-typedef __kernel_fsid_t	fsid_t;
-
-#endif
-
 /*
- * This is ugly -- we're already 64-bit clean, so just duplicate the
- * definitions.
+ * We need compat_statfs64 to be packed, because the i386 ABI won't
+ * add padding at the end to bring it to a multiple of 8 bytes, but
+ * the x86_64 ABI will.
  */
-struct statfs {
-	long f_type;
-	long f_bsize;
-	long f_blocks;
-	long f_bfree;
-	long f_bavail;
-	long f_files;
-	long f_ffree;
-	__kernel_fsid_t f_fsid;
-	long f_namelen;
-	long f_frsize;
-	long f_spare[5];
-};
-
-struct statfs64 {
-	long f_type;
-	long f_bsize;
-	long f_blocks;
-	long f_bfree;
-	long f_bavail;
-	long f_files;
-	long f_ffree;
-	__kernel_fsid_t f_fsid;
-	long f_namelen;
-	long f_frsize;
-	long f_spare[5];
-};
+#define ARCH_PACK_COMPAT_STATFS64 __attribute__((packed,aligned(4)))
 
-struct compat_statfs64 {
-	__u32 f_type;
-	__u32 f_bsize;
-	__u64 f_blocks;
-	__u64 f_bfree;
-	__u64 f_bavail;
-	__u64 f_files;
-	__u64 f_ffree;
-	__kernel_fsid_t f_fsid;
-	__u32 f_namelen;
-	__u32 f_frsize;
-	__u32 f_spare[5];
-} __attribute__((packed));
-
-#endif /* !__i386__ */
+#include <asm-generic/statfs.h>
 #endif /* ASM_X86__STATFS_H */
diff --git a/include/asm-x86/mach-summit/mach_apic.h b/include/asm-x86/summit/apic.h
index 7a66758d701d..c5b2e4b10358 100644
--- a/include/asm-x86/mach-summit/mach_apic.h
+++ b/include/asm-x86/summit/apic.h
@@ -1,5 +1,5 @@
-#ifndef ASM_X86__MACH_SUMMIT__MACH_APIC_H
-#define ASM_X86__MACH_SUMMIT__MACH_APIC_H
+#ifndef __ASM_SUMMIT_APIC_H
+#define __ASM_SUMMIT_APIC_H
 
 #include <asm/smp.h>
 
@@ -21,7 +21,7 @@ static inline cpumask_t target_cpus(void)
 	 * Just start on cpu 0.  IRQ balancing will spread load
 	 */
 	return cpumask_of_cpu(0);
-} 
+}
 #define TARGET_CPUS	(target_cpus())
 
 #define INT_DELIVERY_MODE (dest_LowestPrio)
@@ -30,10 +30,10 @@ static inline cpumask_t target_cpus(void)
 static inline unsigned long check_apicid_used(physid_mask_t bitmap, int apicid)
 {
 	return 0;
-} 
+}
 
 /* we don't use the phys_cpu_present_map to indicate apicid presence */
-static inline unsigned long check_apicid_present(int bit) 
+static inline unsigned long check_apicid_present(int bit)
 {
 	return 1;
 }
@@ -122,7 +122,7 @@ static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_id_map)
 
 static inline physid_mask_t apicid_to_cpu_present(int apicid)
 {
-	return physid_mask_of_physid(apicid);
+	return physid_mask_of_physid(0);
 }
 
 static inline void setup_portio_remap(void)
@@ -143,22 +143,22 @@ static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask)
 	int num_bits_set;
 	int cpus_found = 0;
 	int cpu;
-	int apicid;	
+	int apicid;
 
 	num_bits_set = cpus_weight(cpumask);
 	/* Return id to all */
 	if (num_bits_set == NR_CPUS)
 		return (int) 0xFF;
-	/* 
-	 * The cpus in the mask must all be on the apic cluster.  If are not 
-	 * on the same apicid cluster return default value of TARGET_CPUS. 
+	/*
+	 * The cpus in the mask must all be on the apic cluster.  If are not
+	 * on the same apicid cluster return default value of TARGET_CPUS.
 	 */
 	cpu = first_cpu(cpumask);
 	apicid = cpu_to_logical_apicid(cpu);
 	while (cpus_found < num_bits_set) {
 		if (cpu_isset(cpu, cpumask)) {
 			int new_apicid = cpu_to_logical_apicid(cpu);
-			if (apicid_cluster(apicid) != 
+			if (apicid_cluster(apicid) !=
 					apicid_cluster(new_apicid)){
 				printk ("%s: Not a valid mask!\n",__FUNCTION__);
 				return 0xFF;
@@ -182,4 +182,4 @@ static inline u32 phys_pkg_id(u32 cpuid_apic, int index_msb)
 	return hard_smp_processor_id() >> index_msb;
 }
 
-#endif /* ASM_X86__MACH_SUMMIT__MACH_APIC_H */
+#endif /* __ASM_SUMMIT_APIC_H */
diff --git a/include/asm-x86/summit/apicdef.h b/include/asm-x86/summit/apicdef.h
new file mode 100644
index 000000000000..f3fbca1f61c1
--- /dev/null
+++ b/include/asm-x86/summit/apicdef.h
@@ -0,0 +1,13 @@
+#ifndef __ASM_SUMMIT_APICDEF_H
+#define __ASM_SUMMIT_APICDEF_H
+
+#define		APIC_ID_MASK		(0xFF<<24)
+
+static inline unsigned get_apic_id(unsigned long x)
+{
+	return (x>>24)&0xFF;
+}
+
+#define		GET_APIC_ID(x)	get_apic_id(x)
+
+#endif
diff --git a/include/asm-x86/mach-bigsmp/mach_ipi.h b/include/asm-x86/summit/ipi.h
index b1b0f966a009..53bd1e7bd7b4 100644
--- a/include/asm-x86/mach-bigsmp/mach_ipi.h
+++ b/include/asm-x86/summit/ipi.h
@@ -1,5 +1,5 @@
-#ifndef ASM_X86__MACH_BIGSMP__MACH_IPI_H
-#define ASM_X86__MACH_BIGSMP__MACH_IPI_H
+#ifndef __ASM_SUMMIT_IPI_H
+#define __ASM_SUMMIT_IPI_H
 
 void send_IPI_mask_sequence(cpumask_t mask, int vector);
 
@@ -22,4 +22,4 @@ static inline void send_IPI_all(int vector)
 	send_IPI_mask(cpu_online_map, vector);
 }
 
-#endif /* ASM_X86__MACH_BIGSMP__MACH_IPI_H */
+#endif /* __ASM_SUMMIT_IPI_H */
diff --git a/include/asm-x86/mach-summit/irq_vectors_limits.h b/include/asm-x86/summit/irq_vectors_limits.h
index 22f376ad68e1..890ce3f5e09a 100644
--- a/include/asm-x86/mach-summit/irq_vectors_limits.h
+++ b/include/asm-x86/summit/irq_vectors_limits.h
@@ -1,5 +1,5 @@
-#ifndef ASM_X86__MACH_SUMMIT__IRQ_VECTORS_LIMITS_H
-#define ASM_X86__MACH_SUMMIT__IRQ_VECTORS_LIMITS_H
+#ifndef _ASM_IRQ_VECTORS_LIMITS_H
+#define _ASM_IRQ_VECTORS_LIMITS_H
 
 /*
  * For Summit or generic (i.e. installer) kernels, we have lots of I/O APICs,
@@ -11,4 +11,4 @@
 #define NR_IRQS	224
 #define NR_IRQ_VECTORS	1024
 
-#endif /* ASM_X86__MACH_SUMMIT__IRQ_VECTORS_LIMITS_H */
+#endif /* _ASM_IRQ_VECTORS_LIMITS_H */
diff --git a/include/asm-x86/mach-summit/mach_mpparse.h b/include/asm-x86/summit/mpparse.h
index 92396f28772b..013ce6fab2d5 100644
--- a/include/asm-x86/mach-summit/mach_mpparse.h
+++ b/include/asm-x86/summit/mpparse.h
@@ -1,7 +1,6 @@
-#ifndef ASM_X86__MACH_SUMMIT__MACH_MPPARSE_H
-#define ASM_X86__MACH_SUMMIT__MACH_MPPARSE_H
+#ifndef __ASM_SUMMIT_MPPARSE_H
+#define __ASM_SUMMIT_MPPARSE_H
 
-#include <mach_apic.h>
 #include <asm/tsc.h>
 
 extern int use_cyclone;
@@ -12,11 +11,11 @@ extern void setup_summit(void);
 #define setup_summit()	{}
 #endif
 
-static inline int mps_oem_check(struct mp_config_table *mpc, char *oem, 
+static inline int mps_oem_check(struct mp_config_table *mpc, char *oem,
 		char *productid)
 {
-	if (!strncmp(oem, "IBM ENSW", 8) && 
-			(!strncmp(productid, "VIGIL SMP", 9) 
+	if (!strncmp(oem, "IBM ENSW", 8) &&
+			(!strncmp(productid, "VIGIL SMP", 9)
 			 || !strncmp(productid, "EXA", 3)
 			 || !strncmp(productid, "RUTHLESS SMP", 12))){
 		mark_tsc_unstable("Summit based system");
@@ -107,4 +106,4 @@ static inline int is_WPEG(struct rio_detail *rio){
 		rio->type == LookOutAWPEG || rio->type == LookOutBWPEG);
 }
 
-#endif /* ASM_X86__MACH_SUMMIT__MACH_MPPARSE_H */
+#endif /* __ASM_SUMMIT_MPPARSE_H */
diff --git a/include/asm-x86/system.h b/include/asm-x86/system.h
index 34505dd7b24d..b20c894660f9 100644
--- a/include/asm-x86/system.h
+++ b/include/asm-x86/system.h
@@ -64,7 +64,10 @@ do {									\
 		       							\
 		       /* regparm parameters for __switch_to(): */	\
 		       [prev]     "a" (prev),				\
-		       [next]     "d" (next));				\
+		       [next]     "d" (next)				\
+									\
+		     : /* reloaded segment registers */			\
+			"memory");					\
 } while (0)
 
 /*
diff --git a/include/asm-x86/thread_info.h b/include/asm-x86/thread_info.h
index 4db0066a3a35..3f4e52bb77f5 100644
--- a/include/asm-x86/thread_info.h
+++ b/include/asm-x86/thread_info.h
@@ -241,6 +241,7 @@ static inline struct thread_info *stack_thread_info(void)
 #define TS_POLLING		0x0004	/* true if in idle loop
 					   and not sleeping */
 #define TS_RESTORE_SIGMASK	0x0008	/* restore signal mask in do_signal() */
+#define TS_XSAVE		0x0010	/* Use xsave/xrstor */
 
 #define tsk_is_polling(t) (task_thread_info(t)->status & TS_POLLING)
 
diff --git a/include/asm-x86/tlbflush.h b/include/asm-x86/tlbflush.h
index ef68b76dc3c5..3cdd08b5bdb7 100644
--- a/include/asm-x86/tlbflush.h
+++ b/include/asm-x86/tlbflush.h
@@ -119,6 +119,10 @@ static inline void native_flush_tlb_others(const cpumask_t *cpumask,
 {
 }
 
+static inline void reset_lazy_tlbstate(void)
+{
+}
+
 #else  /* SMP */
 
 #include <asm/smp.h>
@@ -151,6 +155,12 @@ struct tlb_state {
 	char __cacheline_padding[L1_CACHE_BYTES-8];
 };
 DECLARE_PER_CPU(struct tlb_state, cpu_tlbstate);
+
+void reset_lazy_tlbstate(void);
+#else
+static inline void reset_lazy_tlbstate(void)
+{
+}
 #endif
 
 #endif	/* SMP */
diff --git a/include/asm-x86/traps.h b/include/asm-x86/traps.h
index 2ccebc6fb0b0..6c3dc2c65751 100644
--- a/include/asm-x86/traps.h
+++ b/include/asm-x86/traps.h
@@ -1,7 +1,14 @@
 #ifndef ASM_X86__TRAPS_H
 #define ASM_X86__TRAPS_H
 
-/* Common in X86_32 and X86_64 */
+#include <asm/debugreg.h>
+
+#ifdef CONFIG_X86_32
+#define dotraplinkage
+#else
+#define dotraplinkage asmlinkage
+#endif
+
 asmlinkage void divide_error(void);
 asmlinkage void debug(void);
 asmlinkage void nmi(void);
@@ -10,61 +17,65 @@ asmlinkage void overflow(void);
 asmlinkage void bounds(void);
 asmlinkage void invalid_op(void);
 asmlinkage void device_not_available(void);
+#ifdef CONFIG_X86_64
+asmlinkage void double_fault(void);
+#endif
 asmlinkage void coprocessor_segment_overrun(void);
 asmlinkage void invalid_TSS(void);
 asmlinkage void segment_not_present(void);
 asmlinkage void stack_segment(void);
 asmlinkage void general_protection(void);
 asmlinkage void page_fault(void);
+asmlinkage void spurious_interrupt_bug(void);
 asmlinkage void coprocessor_error(void);
-asmlinkage void simd_coprocessor_error(void);
 asmlinkage void alignment_check(void);
-asmlinkage void spurious_interrupt_bug(void);
 #ifdef CONFIG_X86_MCE
 asmlinkage void machine_check(void);
 #endif /* CONFIG_X86_MCE */
+asmlinkage void simd_coprocessor_error(void);
+
+dotraplinkage void do_divide_error(struct pt_regs *, long);
+dotraplinkage void do_debug(struct pt_regs *, long);
+dotraplinkage void do_nmi(struct pt_regs *, long);
+dotraplinkage void do_int3(struct pt_regs *, long);
+dotraplinkage void do_overflow(struct pt_regs *, long);
+dotraplinkage void do_bounds(struct pt_regs *, long);
+dotraplinkage void do_invalid_op(struct pt_regs *, long);
+dotraplinkage void do_device_not_available(struct pt_regs *, long);
+dotraplinkage void do_coprocessor_segment_overrun(struct pt_regs *, long);
+dotraplinkage void do_invalid_TSS(struct pt_regs *, long);
+dotraplinkage void do_segment_not_present(struct pt_regs *, long);
+dotraplinkage void do_stack_segment(struct pt_regs *, long);
+dotraplinkage void do_general_protection(struct pt_regs *, long);
+dotraplinkage void do_page_fault(struct pt_regs *, unsigned long);
+dotraplinkage void do_spurious_interrupt_bug(struct pt_regs *, long);
+dotraplinkage void do_coprocessor_error(struct pt_regs *, long);
+dotraplinkage void do_alignment_check(struct pt_regs *, long);
+#ifdef CONFIG_X86_MCE
+dotraplinkage void do_machine_check(struct pt_regs *, long);
+#endif
+dotraplinkage void do_simd_coprocessor_error(struct pt_regs *, long);
+#ifdef CONFIG_X86_32
+dotraplinkage void do_iret_error(struct pt_regs *, long);
+#endif
 
-void do_divide_error(struct pt_regs *, long);
-void do_overflow(struct pt_regs *, long);
-void do_bounds(struct pt_regs *, long);
-void do_coprocessor_segment_overrun(struct pt_regs *, long);
-void do_invalid_TSS(struct pt_regs *, long);
-void do_segment_not_present(struct pt_regs *, long);
-void do_stack_segment(struct pt_regs *, long);
-void do_alignment_check(struct pt_regs *, long);
-void do_invalid_op(struct pt_regs *, long);
-void do_general_protection(struct pt_regs *, long);
-void do_nmi(struct pt_regs *, long);
+static inline int get_si_code(unsigned long condition)
+{
+	if (condition & DR_STEP)
+		return TRAP_TRACE;
+	else if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3))
+		return TRAP_HWBKPT;
+	else
+		return TRAP_BRKPT;
+}
 
 extern int panic_on_unrecovered_nmi;
 extern int kstack_depth_to_print;
 
 #ifdef CONFIG_X86_32
-
-void do_iret_error(struct pt_regs *, long);
-void do_int3(struct pt_regs *, long);
-void do_debug(struct pt_regs *, long);
 void math_error(void __user *);
-void do_coprocessor_error(struct pt_regs *, long);
-void do_simd_coprocessor_error(struct pt_regs *, long);
-void do_spurious_interrupt_bug(struct pt_regs *, long);
 unsigned long patch_espfix_desc(unsigned long, unsigned long);
 asmlinkage void math_emulate(long);
+#endif
 
-void do_page_fault(struct pt_regs *regs, unsigned long error_code);
-
-#else /* CONFIG_X86_32 */
-
-asmlinkage void double_fault(void);
-
-asmlinkage void do_int3(struct pt_regs *, long);
-asmlinkage void do_stack_segment(struct pt_regs *, long);
-asmlinkage void do_debug(struct pt_regs *, unsigned long);
-asmlinkage void do_coprocessor_error(struct pt_regs *);
-asmlinkage void do_simd_coprocessor_error(struct pt_regs *);
-asmlinkage void do_spurious_interrupt_bug(struct pt_regs *);
-
-asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long error_code);
-
-#endif /* CONFIG_X86_32 */
 #endif /* ASM_X86__TRAPS_H */
diff --git a/include/asm-x86/ucontext.h b/include/asm-x86/ucontext.h
index 9948dd328084..89eaa5456a7e 100644
--- a/include/asm-x86/ucontext.h
+++ b/include/asm-x86/ucontext.h
@@ -1,6 +1,12 @@
 #ifndef ASM_X86__UCONTEXT_H
 #define ASM_X86__UCONTEXT_H
 
+#define UC_FP_XSTATE	0x1	/* indicates the presence of extended state
+				 * information in the memory layout pointed
+				 * by the fpstate pointer in the ucontext's
+				 * sigcontext struct (uc_mcontext).
+				 */
+
 struct ucontext {
 	unsigned long	  uc_flags;
 	struct ucontext  *uc_link;
diff --git a/include/asm-x86/xcr.h b/include/asm-x86/xcr.h
new file mode 100644
index 000000000000..f2cba4e79a23
--- /dev/null
+++ b/include/asm-x86/xcr.h
@@ -0,0 +1,49 @@
+/* -*- linux-c -*- ------------------------------------------------------- *
+ *
+ *   Copyright 2008 rPath, Inc. - All Rights Reserved
+ *
+ *   This file is part of the Linux kernel, and is made available under
+ *   the terms of the GNU General Public License version 2 or (at your
+ *   option) any later version; incorporated herein by reference.
+ *
+ * ----------------------------------------------------------------------- */
+
+/*
+ * asm-x86/xcr.h
+ *
+ * Definitions for the eXtended Control Register instructions
+ */
+
+#ifndef _ASM_X86_XCR_H
+#define _ASM_X86_XCR_H
+
+#define XCR_XFEATURE_ENABLED_MASK	0x00000000
+
+#ifdef __KERNEL__
+# ifndef __ASSEMBLY__
+
+#include <linux/types.h>
+
+static inline u64 xgetbv(u32 index)
+{
+	u32 eax, edx;
+
+	asm volatile(".byte 0x0f,0x01,0xd0" /* xgetbv */
+		     : "=a" (eax), "=d" (edx)
+		     : "c" (index));
+	return eax + ((u64)edx << 32);
+}
+
+static inline void xsetbv(u32 index, u64 value)
+{
+	u32 eax = value;
+	u32 edx = value >> 32;
+
+	asm volatile(".byte 0x0f,0x01,0xd1" /* xsetbv */
+		     : : "a" (eax), "d" (edx), "c" (index));
+}
+
+# endif /* __ASSEMBLY__ */
+#endif /* __KERNEL__ */
+
+#endif /* _ASM_X86_XCR_H */
diff --git a/include/asm-x86/xen/hypervisor.h b/include/asm-x86/xen/hypervisor.h
index 0ef3a88b869d..445a24759560 100644
--- a/include/asm-x86/xen/hypervisor.h
+++ b/include/asm-x86/xen/hypervisor.h
@@ -54,7 +54,6 @@
 /* arch/i386/kernel/setup.c */
 extern struct shared_info *HYPERVISOR_shared_info;
 extern struct start_info *xen_start_info;
-#define is_initial_xendomain() (xen_start_info->flags & SIF_INITDOMAIN)
 
 /* arch/i386/mach-xen/evtchn.c */
 /* Force a proper event-channel callback from Xen. */
@@ -67,6 +66,17 @@ u64 jiffies_to_st(unsigned long jiffies);
 #define MULTI_UVMFLAGS_INDEX 3
 #define MULTI_UVMDOMID_INDEX 4
 
-#define is_running_on_xen()	(xen_start_info ? 1 : 0)
+enum xen_domain_type {
+	XEN_NATIVE,
+	XEN_PV_DOMAIN,
+	XEN_HVM_DOMAIN,
+};
+
+extern enum xen_domain_type xen_domain_type;
+
+#define xen_domain()		(xen_domain_type != XEN_NATIVE)
+#define xen_pv_domain()		(xen_domain_type == XEN_PV_DOMAIN)
+#define xen_initial_domain()	(xen_pv_domain() && xen_start_info->flags & SIF_INITDOMAIN)
+#define xen_hvm_domain()	(xen_domain_type == XEN_HVM_DOMAIN)
 
 #endif /* ASM_X86__XEN__HYPERVISOR_H */
diff --git a/include/asm-x86/xsave.h b/include/asm-x86/xsave.h
new file mode 100644
index 000000000000..08e9a1ac07a9
--- /dev/null
+++ b/include/asm-x86/xsave.h
@@ -0,0 +1,118 @@
+#ifndef __ASM_X86_XSAVE_H
+#define __ASM_X86_XSAVE_H
+
+#include <linux/types.h>
+#include <asm/processor.h>
+#include <asm/i387.h>
+
+#define XSTATE_FP	0x1
+#define XSTATE_SSE	0x2
+
+#define XSTATE_FPSSE	(XSTATE_FP | XSTATE_SSE)
+
+#define FXSAVE_SIZE	512
+
+/*
+ * These are the features that the OS can handle currently.
+ */
+#define XCNTXT_MASK	(XSTATE_FP | XSTATE_SSE)
+
+#ifdef CONFIG_X86_64
+#define REX_PREFIX	"0x48, "
+#else
+#define REX_PREFIX
+#endif
+
+extern unsigned int xstate_size;
+extern u64 pcntxt_mask;
+extern struct xsave_struct *init_xstate_buf;
+
+extern void xsave_cntxt_init(void);
+extern void xsave_init(void);
+extern int init_fpu(struct task_struct *child);
+extern int check_for_xstate(struct i387_fxsave_struct __user *buf,
+			    void __user *fpstate,
+			    struct _fpx_sw_bytes *sw);
+
+static inline int xrstor_checking(struct xsave_struct *fx)
+{
+	int err;
+
+	asm volatile("1: .byte " REX_PREFIX "0x0f,0xae,0x2f\n\t"
+		     "2:\n"
+		     ".section .fixup,\"ax\"\n"
+		     "3:  movl $-1,%[err]\n"
+		     "    jmp  2b\n"
+		     ".previous\n"
+		     _ASM_EXTABLE(1b, 3b)
+		     : [err] "=r" (err)
+		     : "D" (fx), "m" (*fx), "a" (-1), "d" (-1), "0" (0)
+		     : "memory");
+
+	return err;
+}
+
+static inline int xsave_user(struct xsave_struct __user *buf)
+{
+	int err;
+	__asm__ __volatile__("1: .byte " REX_PREFIX "0x0f,0xae,0x27\n"
+			     "2:\n"
+			     ".section .fixup,\"ax\"\n"
+			     "3:  movl $-1,%[err]\n"
+			     "    jmp  2b\n"
+			     ".previous\n"
+			     ".section __ex_table,\"a\"\n"
+			     _ASM_ALIGN "\n"
+			     _ASM_PTR "1b,3b\n"
+			     ".previous"
+			     : [err] "=r" (err)
+			     : "D" (buf), "a" (-1), "d" (-1), "0" (0)
+			     : "memory");
+	if (unlikely(err) && __clear_user(buf, xstate_size))
+		err = -EFAULT;
+	/* No need to clear here because the caller clears USED_MATH */
+	return err;
+}
+
+static inline int xrestore_user(struct xsave_struct __user *buf, u64 mask)
+{
+	int err;
+	struct xsave_struct *xstate = ((__force struct xsave_struct *)buf);
+	u32 lmask = mask;
+	u32 hmask = mask >> 32;
+
+	__asm__ __volatile__("1: .byte " REX_PREFIX "0x0f,0xae,0x2f\n"
+			     "2:\n"
+			     ".section .fixup,\"ax\"\n"
+			     "3:  movl $-1,%[err]\n"
+			     "    jmp  2b\n"
+			     ".previous\n"
+			     ".section __ex_table,\"a\"\n"
+			     _ASM_ALIGN "\n"
+			     _ASM_PTR "1b,3b\n"
+			     ".previous"
+			     : [err] "=r" (err)
+			     : "D" (xstate), "a" (lmask), "d" (hmask), "0" (0)
+			     : "memory");	/* memory required? */
+	return err;
+}
+
+static inline void xrstor_state(struct xsave_struct *fx, u64 mask)
+{
+	u32 lmask = mask;
+	u32 hmask = mask >> 32;
+
+	asm volatile(".byte " REX_PREFIX "0x0f,0xae,0x2f\n\t"
+		     : : "D" (fx), "m" (*fx), "a" (lmask), "d" (hmask)
+		     :   "memory");
+}
+
+static inline void xsave(struct task_struct *tsk)
+{
+	/* This, however, we can work around by forcing the compiler to select
+	   an addressing mode that doesn't require extended registers. */
+	__asm__ __volatile__(".byte " REX_PREFIX "0x0f,0xae,0x27"
+			     : : "D" (&(tsk->thread.xstate->xsave)),
+				 "a" (-1), "d"(-1) : "memory");
+}
+#endif
diff --git a/include/asm-xtensa/a.out.h b/include/asm-xtensa/a.out.h
deleted file mode 100644
index fdf13702924a..000000000000
--- a/include/asm-xtensa/a.out.h
+++ /dev/null
@@ -1,29 +0,0 @@
-/*
- * include/asm-xtensa/a.out.h
- *
- * Dummy a.out file. Xtensa does not support the a.out format, but the kernel
- * seems to depend on it.
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 2001 - 2005 Tensilica Inc.
- */
-
-#ifndef _XTENSA_A_OUT_H
-#define _XTENSA_A_OUT_H
-
-struct exec
-{
-  unsigned long a_info;
-  unsigned a_text;
-  unsigned a_data;
-  unsigned a_bss;
-  unsigned a_syms;
-  unsigned a_entry;
-  unsigned a_trsize;
-  unsigned a_drsize;
-};
-
-#endif /* _XTENSA_A_OUT_H */
diff --git a/include/linux/ata.h b/include/linux/ata.h
index be00973d1a8c..a53318b8cbd0 100644
--- a/include/linux/ata.h
+++ b/include/linux/ata.h
@@ -30,6 +30,7 @@
 #define __LINUX_ATA_H__
 
 #include <linux/types.h>
+#include <asm/byteorder.h>
 
 /* defines only for the constants which don't work well as enums */
 #define ATA_DMA_BOUNDARY	0xffffUL
@@ -558,6 +559,15 @@ static inline int ata_id_has_flush(const u16 *id)
 	return id[ATA_ID_COMMAND_SET_2] & (1 << 12);
 }
 
+static inline int ata_id_flush_enabled(const u16 *id)
+{
+	if (ata_id_has_flush(id) == 0)
+		return 0;
+	if ((id[ATA_ID_CSF_DEFAULT] & 0xC000) != 0x4000)
+		return 0;
+	return id[ATA_ID_CFS_ENABLE_2] & (1 << 12);
+}
+
 static inline int ata_id_has_flush_ext(const u16 *id)
 {
 	if ((id[ATA_ID_COMMAND_SET_2] & 0xC000) != 0x4000)
@@ -565,6 +575,19 @@ static inline int ata_id_has_flush_ext(const u16 *id)
 	return id[ATA_ID_COMMAND_SET_2] & (1 << 13);
 }
 
+static inline int ata_id_flush_ext_enabled(const u16 *id)
+{
+	if (ata_id_has_flush_ext(id) == 0)
+		return 0;
+	if ((id[ATA_ID_CSF_DEFAULT] & 0xC000) != 0x4000)
+		return 0;
+	/*
+	 * some Maxtor disks have bit 13 defined incorrectly
+	 * so check bit 10 too
+	 */
+	return (id[ATA_ID_CFS_ENABLE_2] & 0x2400) == 0x2400;
+}
+
 static inline int ata_id_has_lba48(const u16 *id)
 {
 	if ((id[ATA_ID_COMMAND_SET_2] & 0xC000) != 0x4000)
@@ -574,6 +597,15 @@ static inline int ata_id_has_lba48(const u16 *id)
 	return id[ATA_ID_COMMAND_SET_2] & (1 << 10);
 }
 
+static inline int ata_id_lba48_enabled(const u16 *id)
+{
+	if (ata_id_has_lba48(id) == 0)
+		return 0;
+	if ((id[ATA_ID_CSF_DEFAULT] & 0xC000) != 0x4000)
+		return 0;
+	return id[ATA_ID_CFS_ENABLE_2] & (1 << 10);
+}
+
 static inline int ata_id_hpa_enabled(const u16 *id)
 {
 	/* Yes children, word 83 valid bits cover word 82 data */
@@ -645,7 +677,15 @@ static inline unsigned int ata_id_major_version(const u16 *id)
 
 static inline int ata_id_is_sata(const u16 *id)
 {
-	return ata_id_major_version(id) >= 5 && id[ATA_ID_HW_CONFIG] == 0;
+	/*
+	 * See if word 93 is 0 AND drive is at least ATA-5 compatible
+	 * verifying that word 80 by casting it to a signed type --
+	 * this trick allows us to filter out the reserved values of
+	 * 0x0000 and 0xffff along with the earlier ATA revisions...
+	 */
+	if (id[ATA_ID_HW_CONFIG] == 0 && (short)id[ATA_ID_MAJOR_VER] >= 0x0020)
+		return 1;
+	return 0;
 }
 
 static inline int ata_id_has_tpm(const u16 *id)
@@ -742,6 +782,76 @@ static inline int atapi_id_dmadir(const u16 *dev_id)
 	return ata_id_major_version(dev_id) >= 7 && (dev_id[62] & 0x8000);
 }
 
+/*
+ * ata_id_is_lba_capacity_ok() performs a sanity check on
+ * the claimed LBA capacity value for the device.
+ *
+ * Returns 1 if LBA capacity looks sensible, 0 otherwise.
+ *
+ * It is called only once for each device.
+ */
+static inline int ata_id_is_lba_capacity_ok(u16 *id)
+{
+	unsigned long lba_sects, chs_sects, head, tail;
+
+	/* No non-LBA info .. so valid! */
+	if (id[ATA_ID_CYLS] == 0)
+		return 1;
+
+	lba_sects = ata_id_u32(id, ATA_ID_LBA_CAPACITY);
+
+	/*
+	 * The ATA spec tells large drives to return
+	 * C/H/S = 16383/16/63 independent of their size.
+	 * Some drives can be jumpered to use 15 heads instead of 16.
+	 * Some drives can be jumpered to use 4092 cyls instead of 16383.
+	 */
+	if ((id[ATA_ID_CYLS] == 16383 ||
+	     (id[ATA_ID_CYLS] == 4092 && id[ATA_ID_CUR_CYLS] == 16383)) &&
+	    id[ATA_ID_SECTORS] == 63 &&
+	    (id[ATA_ID_HEADS] == 15 || id[ATA_ID_HEADS] == 16) &&
+	    (lba_sects >= 16383 * 63 * id[ATA_ID_HEADS]))
+		return 1;
+
+	chs_sects = id[ATA_ID_CYLS] * id[ATA_ID_HEADS] * id[ATA_ID_SECTORS];
+
+	/* perform a rough sanity check on lba_sects: within 10% is OK */
+	if (lba_sects - chs_sects < chs_sects/10)
+		return 1;
+
+	/* some drives have the word order reversed */
+	head = (lba_sects >> 16) & 0xffff;
+	tail = lba_sects & 0xffff;
+	lba_sects = head | (tail << 16);
+
+	if (lba_sects - chs_sects < chs_sects/10) {
+		*(__le32 *)&id[ATA_ID_LBA_CAPACITY] = __cpu_to_le32(lba_sects);
+		return 1;	/* LBA capacity is (now) good */
+	}
+
+	return 0;	/* LBA capacity value may be bad */
+}
+
+static inline void ata_id_to_hd_driveid(u16 *id)
+{
+#ifdef __BIG_ENDIAN
+	/* accessed in struct hd_driveid as 8-bit values */
+	id[ATA_ID_MAX_MULTSECT]	 = __cpu_to_le16(id[ATA_ID_MAX_MULTSECT]);
+	id[ATA_ID_CAPABILITY]	 = __cpu_to_le16(id[ATA_ID_CAPABILITY]);
+	id[ATA_ID_OLD_PIO_MODES] = __cpu_to_le16(id[ATA_ID_OLD_PIO_MODES]);
+	id[ATA_ID_OLD_DMA_MODES] = __cpu_to_le16(id[ATA_ID_OLD_DMA_MODES]);
+	id[ATA_ID_MULTSECT]	 = __cpu_to_le16(id[ATA_ID_MULTSECT]);
+
+	/* as 32-bit values */
+	*(u32 *)&id[ATA_ID_LBA_CAPACITY] = ata_id_u32(id, ATA_ID_LBA_CAPACITY);
+	*(u32 *)&id[ATA_ID_SPG]		 = ata_id_u32(id, ATA_ID_SPG);
+
+	/* as 64-bit value */
+	*(u64 *)&id[ATA_ID_LBA_CAPACITY_2] =
+		ata_id_u64(id, ATA_ID_LBA_CAPACITY_2);
+#endif
+}
+
 static inline int is_multi_taskfile(struct ata_taskfile *tf)
 {
 	return (tf->command == ATA_CMD_READ_MULTI) ||
diff --git a/include/linux/devpts_fs.h b/include/linux/devpts_fs.h
index 154769cad3f3..5ce0e5fd712e 100644
--- a/include/linux/devpts_fs.h
+++ b/include/linux/devpts_fs.h
@@ -17,20 +17,31 @@
 
 #ifdef CONFIG_UNIX98_PTYS
 
-int devpts_new_index(void);
-void devpts_kill_index(int idx);
-int devpts_pty_new(struct tty_struct *tty);      /* mknod in devpts */
-struct tty_struct *devpts_get_tty(int number);	 /* get tty structure */
-void devpts_pty_kill(int number);		 /* unlink */
+int devpts_new_index(struct inode *ptmx_inode);
+void devpts_kill_index(struct inode *ptmx_inode, int idx);
+/* mknod in devpts */
+int devpts_pty_new(struct inode *ptmx_inode, struct tty_struct *tty);
+/* get tty structure */
+struct tty_struct *devpts_get_tty(struct inode *pts_inode, int number);
+/* unlink */
+void devpts_pty_kill(struct tty_struct *tty);
 
 #else
 
 /* Dummy stubs in the no-pty case */
-static inline int devpts_new_index(void) { return -EINVAL; }
-static inline void devpts_kill_index(int idx) { }
-static inline int devpts_pty_new(struct tty_struct *tty) { return -EINVAL; }
-static inline struct tty_struct *devpts_get_tty(int number) { return NULL; }
-static inline void devpts_pty_kill(int number) { }
+static inline int devpts_new_index(struct inode *ptmx_inode) { return -EINVAL; }
+static inline void devpts_kill_index(struct inode *ptmx_inode, int idx) { }
+static inline int devpts_pty_new(struct inode *ptmx_inode,
+				struct tty_struct *tty)
+{
+	return -EINVAL;
+}
+static inline struct tty_struct *devpts_get_tty(struct inode *pts_inode,
+		int number)
+{
+	return NULL;
+}
+static inline void devpts_pty_kill(struct tty_struct *tty) { }
 
 #endif
 
diff --git a/include/linux/dmar.h b/include/linux/dmar.h
index 56c73b847551..c360c558e59e 100644
--- a/include/linux/dmar.h
+++ b/include/linux/dmar.h
@@ -25,9 +25,99 @@
 #include <linux/types.h>
 #include <linux/msi.h>
 
-#ifdef CONFIG_DMAR
+#if defined(CONFIG_DMAR) || defined(CONFIG_INTR_REMAP)
 struct intel_iommu;
 
+struct dmar_drhd_unit {
+	struct list_head list;		/* list of drhd units	*/
+	struct  acpi_dmar_header *hdr;	/* ACPI header		*/
+	u64	reg_base_addr;		/* register base address*/
+	struct	pci_dev **devices; 	/* target device array	*/
+	int	devices_cnt;		/* target device count	*/
+	u8	ignored:1; 		/* ignore drhd		*/
+	u8	include_all:1;
+	struct intel_iommu *iommu;
+};
+
+extern struct list_head dmar_drhd_units;
+
+#define for_each_drhd_unit(drhd) \
+	list_for_each_entry(drhd, &dmar_drhd_units, list)
+
+extern int dmar_table_init(void);
+extern int early_dmar_detect(void);
+extern int dmar_dev_scope_init(void);
+
+/* Intel IOMMU detection */
+extern void detect_intel_iommu(void);
+
+
+extern int parse_ioapics_under_ir(void);
+extern int alloc_iommu(struct dmar_drhd_unit *);
+#else
+static inline void detect_intel_iommu(void)
+{
+	return;
+}
+
+static inline int dmar_table_init(void)
+{
+	return -ENODEV;
+}
+#endif /* !CONFIG_DMAR && !CONFIG_INTR_REMAP */
+
+#ifdef CONFIG_INTR_REMAP
+extern int intr_remapping_enabled;
+extern int enable_intr_remapping(int);
+
+struct irte {
+	union {
+		struct {
+			__u64	present 	: 1,
+				fpd		: 1,
+				dst_mode	: 1,
+				redir_hint	: 1,
+				trigger_mode	: 1,
+				dlvry_mode	: 3,
+				avail		: 4,
+				__reserved_1	: 4,
+				vector		: 8,
+				__reserved_2	: 8,
+				dest_id		: 32;
+		};
+		__u64 low;
+	};
+
+	union {
+		struct {
+			__u64	sid		: 16,
+				sq		: 2,
+				svt		: 2,
+				__reserved_3	: 44;
+		};
+		__u64 high;
+	};
+};
+extern int get_irte(int irq, struct irte *entry);
+extern int modify_irte(int irq, struct irte *irte_modified);
+extern int alloc_irte(struct intel_iommu *iommu, int irq, u16 count);
+extern int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index,
+   			u16 sub_handle);
+extern int map_irq_to_irte_handle(int irq, u16 *sub_handle);
+extern int clear_irte_irq(int irq, struct intel_iommu *iommu, u16 index);
+extern int flush_irte(int irq);
+extern int free_irte(int irq);
+
+extern int irq_remapped(int irq);
+extern struct intel_iommu *map_dev_to_ir(struct pci_dev *dev);
+extern struct intel_iommu *map_ioapic_to_ir(int apic);
+#else
+#define irq_remapped(irq)		(0)
+#define enable_intr_remapping(mode)	(-1)
+#define intr_remapping_enabled		(0)
+#endif
+
+#ifdef CONFIG_DMAR
 extern const char *dmar_get_fault_reason(u8 fault_reason);
 
 /* Can't use the common MSI interrupt functions
@@ -40,47 +130,30 @@ extern void dmar_msi_write(int irq, struct msi_msg *msg);
 extern int dmar_set_interrupt(struct intel_iommu *iommu);
 extern int arch_setup_dmar_msi(unsigned int irq);
 
-/* Intel IOMMU detection and initialization functions */
-extern void detect_intel_iommu(void);
-extern int intel_iommu_init(void);
-
-extern int dmar_table_init(void);
-extern int early_dmar_detect(void);
-
-extern struct list_head dmar_drhd_units;
+extern int iommu_detected, no_iommu;
 extern struct list_head dmar_rmrr_units;
-
-struct dmar_drhd_unit {
-	struct list_head list;		/* list of drhd units	*/
-	u64	reg_base_addr;		/* register base address*/
-	struct	pci_dev **devices; 	/* target device array	*/
-	int	devices_cnt;		/* target device count	*/
-	u8	ignored:1; 		/* ignore drhd		*/
-	u8	include_all:1;
-	struct intel_iommu *iommu;
-};
-
 struct dmar_rmrr_unit {
 	struct list_head list;		/* list of rmrr units	*/
+	struct acpi_dmar_header *hdr;	/* ACPI header		*/
 	u64	base_address;		/* reserved base address*/
 	u64	end_address;		/* reserved end address */
 	struct pci_dev **devices;	/* target devices */
 	int	devices_cnt;		/* target device count */
 };
 
-#define for_each_drhd_unit(drhd) \
-	list_for_each_entry(drhd, &dmar_drhd_units, list)
 #define for_each_rmrr_units(rmrr) \
 	list_for_each_entry(rmrr, &dmar_rmrr_units, list)
+/* Intel DMAR  initialization functions */
+extern int intel_iommu_init(void);
+extern int dmar_disabled;
 #else
-static inline void detect_intel_iommu(void)
-{
-	return;
-}
 static inline int intel_iommu_init(void)
 {
+#ifdef CONFIG_INTR_REMAP
+	return dmar_dev_scope_init();
+#else
 	return -ENODEV;
+#endif
 }
-
 #endif /* !CONFIG_DMAR */
 #endif /* __DMAR_H__ */
diff --git a/include/linux/dmi.h b/include/linux/dmi.h
index 2a063b64133f..e5084eb5943a 100644
--- a/include/linux/dmi.h
+++ b/include/linux/dmi.h
@@ -2,29 +2,9 @@
 #define __DMI_H__
 
 #include <linux/list.h>
+#include <linux/mod_devicetable.h>
 
-enum dmi_field {
-	DMI_NONE,
-	DMI_BIOS_VENDOR,
-	DMI_BIOS_VERSION,
-	DMI_BIOS_DATE,
-	DMI_SYS_VENDOR,
-	DMI_PRODUCT_NAME,
-	DMI_PRODUCT_VERSION,
-	DMI_PRODUCT_SERIAL,
-	DMI_PRODUCT_UUID,
-	DMI_BOARD_VENDOR,
-	DMI_BOARD_NAME,
-	DMI_BOARD_VERSION,
-	DMI_BOARD_SERIAL,
-	DMI_BOARD_ASSET_TAG,
-	DMI_CHASSIS_VENDOR,
-	DMI_CHASSIS_TYPE,
-	DMI_CHASSIS_VERSION,
-	DMI_CHASSIS_SERIAL,
-	DMI_CHASSIS_ASSET_TAG,
-	DMI_STRING_MAX,
-};
+/* enum dmi_field is in mod_devicetable.h */
 
 enum dmi_device_type {
 	DMI_DEV_TYPE_ANY = 0,
@@ -48,23 +28,6 @@ struct dmi_header {
 	u16 handle;
 };
 
-/*
- *	DMI callbacks for problem boards
- */
-struct dmi_strmatch {
-	u8 slot;
-	char *substr;
-};
-
-struct dmi_system_id {
-	int (*callback)(const struct dmi_system_id *);
-	const char *ident;
-	struct dmi_strmatch matches[4];
-	void *driver_data;
-};
-
-#define DMI_MATCH(a, b)	{ a, b }
-
 struct dmi_device {
 	struct list_head list;
 	int type;
diff --git a/include/linux/elf.h b/include/linux/elf.h
index edc3dac3f02f..0b61ca41a044 100644
--- a/include/linux/elf.h
+++ b/include/linux/elf.h
@@ -360,6 +360,7 @@ typedef struct elf64_shdr {
 #define NT_PPC_SPE	0x101		/* PowerPC SPE/EVR registers */
 #define NT_PPC_VSX	0x102		/* PowerPC VSX registers */
 #define NT_386_TLS	0x200		/* i386 TLS slots (struct user_desc) */
+#define NT_386_IOPERM	0x201		/* x86 io permission bitmap (1=deny) */
 
 
 /* Note header in a PT_NOTE section */
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index 80171ee89a22..8120fa1bc235 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -837,6 +837,8 @@ extern void ext3_truncate (struct inode *);
 extern void ext3_set_inode_flags(struct inode *);
 extern void ext3_get_inode_flags(struct ext3_inode_info *);
 extern void ext3_set_aops(struct inode *inode);
+extern int ext3_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+		       u64 start, u64 len);
 
 /* ioctl.c */
 extern int ext3_ioctl (struct inode *, struct file *, unsigned int,
diff --git a/include/linux/fiemap.h b/include/linux/fiemap.h
new file mode 100644
index 000000000000..671decbd2aeb
--- /dev/null
+++ b/include/linux/fiemap.h
@@ -0,0 +1,64 @@
+/*
+ * FS_IOC_FIEMAP ioctl infrastructure.
+ *
+ * Some portions copyright (C) 2007 Cluster File Systems, Inc
+ *
+ * Authors: Mark Fasheh <mfasheh@suse.com>
+ *          Kalpak Shah <kalpak.shah@sun.com>
+ *          Andreas Dilger <adilger@sun.com>
+ */
+
+#ifndef _LINUX_FIEMAP_H
+#define _LINUX_FIEMAP_H
+
+struct fiemap_extent {
+	__u64 fe_logical;  /* logical offset in bytes for the start of
+			    * the extent from the beginning of the file */
+	__u64 fe_physical; /* physical offset in bytes for the start
+			    * of the extent from the beginning of the disk */
+	__u64 fe_length;   /* length in bytes for this extent */
+	__u64 fe_reserved64[2];
+	__u32 fe_flags;    /* FIEMAP_EXTENT_* flags for this extent */
+	__u32 fe_reserved[3];
+};
+
+struct fiemap {
+	__u64 fm_start;		/* logical offset (inclusive) at
+				 * which to start mapping (in) */
+	__u64 fm_length;	/* logical length of mapping which
+				 * userspace wants (in) */
+	__u32 fm_flags;		/* FIEMAP_FLAG_* flags for request (in/out) */
+	__u32 fm_mapped_extents;/* number of extents that were mapped (out) */
+	__u32 fm_extent_count;  /* size of fm_extents array (in) */
+	__u32 fm_reserved;
+	struct fiemap_extent fm_extents[0]; /* array of mapped extents (out) */
+};
+
+#define FIEMAP_MAX_OFFSET	(~0ULL)
+
+#define FIEMAP_FLAG_SYNC	0x00000001 /* sync file data before map */
+#define FIEMAP_FLAG_XATTR	0x00000002 /* map extended attribute tree */
+
+#define FIEMAP_FLAGS_COMPAT	(FIEMAP_FLAG_SYNC | FIEMAP_FLAG_XATTR)
+
+#define FIEMAP_EXTENT_LAST		0x00000001 /* Last extent in file. */
+#define FIEMAP_EXTENT_UNKNOWN		0x00000002 /* Data location unknown. */
+#define FIEMAP_EXTENT_DELALLOC		0x00000004 /* Location still pending.
+						    * Sets EXTENT_UNKNOWN. */
+#define FIEMAP_EXTENT_ENCODED		0x00000008 /* Data can not be read
+						    * while fs is unmounted */
+#define FIEMAP_EXTENT_DATA_ENCRYPTED	0x00000080 /* Data is encrypted by fs.
+						    * Sets EXTENT_NO_BYPASS. */
+#define FIEMAP_EXTENT_NOT_ALIGNED	0x00000100 /* Extent offsets may not be
+						    * block aligned. */
+#define FIEMAP_EXTENT_DATA_INLINE	0x00000200 /* Data mixed with metadata.
+						    * Sets EXTENT_NOT_ALIGNED.*/
+#define FIEMAP_EXTENT_DATA_TAIL		0x00000400 /* Multiple files in block.
+						    * Sets EXTENT_NOT_ALIGNED.*/
+#define FIEMAP_EXTENT_UNWRITTEN		0x00000800 /* Space allocated, but
+						    * no data (i.e. zero). */
+#define FIEMAP_EXTENT_MERGED		0x00001000 /* File does not natively
+						    * support extents. Result
+						    * merged for efficiency. */
+
+#endif /* _LINUX_FIEMAP_H */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 32477e8872d5..44e3cb2f1966 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -234,6 +234,7 @@ extern int dir_notify_enable;
 #define	FS_IOC_SETFLAGS			_IOW('f', 2, long)
 #define	FS_IOC_GETVERSION		_IOR('v', 1, long)
 #define	FS_IOC_SETVERSION		_IOW('v', 2, long)
+#define FS_IOC_FIEMAP			_IOWR('f', 11, struct fiemap)
 #define FS_IOC32_GETFLAGS		_IOR('f', 1, int)
 #define FS_IOC32_SETFLAGS		_IOW('f', 2, int)
 #define FS_IOC32_GETVERSION		_IOR('v', 1, int)
@@ -294,6 +295,7 @@ extern int dir_notify_enable;
 #include <linux/mutex.h>
 #include <linux/capability.h>
 #include <linux/semaphore.h>
+#include <linux/fiemap.h>
 
 #include <asm/atomic.h>
 #include <asm/byteorder.h>
@@ -1182,6 +1184,20 @@ extern void dentry_unhash(struct dentry *dentry);
 extern int file_permission(struct file *, int);
 
 /*
+ * VFS FS_IOC_FIEMAP helper definitions.
+ */
+struct fiemap_extent_info {
+	unsigned int fi_flags;		/* Flags as passed from user */
+	unsigned int fi_extents_mapped;	/* Number of mapped extents */
+	unsigned int fi_extents_max;	/* Size of fiemap_extent array */
+	struct fiemap_extent *fi_extents_start; /* Start of fiemap_extent
+						 * array */
+};
+int fiemap_fill_next_extent(struct fiemap_extent_info *info, u64 logical,
+			    u64 phys, u64 len, u32 flags);
+int fiemap_check_flags(struct fiemap_extent_info *fieinfo, u32 fs_flags);
+
+/*
  * File types
  *
  * NOTE! These match bits 12..15 of stat.st_mode
@@ -1290,6 +1306,8 @@ struct inode_operations {
 	void (*truncate_range)(struct inode *, loff_t, loff_t);
 	long (*fallocate)(struct inode *inode, int mode, loff_t offset,
 			  loff_t len);
+	int (*fiemap)(struct inode *, struct fiemap_extent_info *, u64 start,
+		      u64 len);
 };
 
 struct seq_file;
@@ -1987,6 +2005,9 @@ extern int vfs_fstat(unsigned int, struct kstat *);
 
 extern int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd,
 		    unsigned long arg);
+extern int generic_block_fiemap(struct inode *inode,
+				struct fiemap_extent_info *fieinfo, u64 start,
+				u64 len, get_block_t *get_block);
 
 extern void get_filesystem(struct file_system_type *fs);
 extern void put_filesystem(struct file_system_type *fs);
diff --git a/include/linux/hpet.h b/include/linux/hpet.h
index 2dc29ce6c8e4..79f63a27bcef 100644
--- a/include/linux/hpet.h
+++ b/include/linux/hpet.h
@@ -37,6 +37,7 @@ struct hpet {
 #define	hpet_compare	_u1._hpet_compare
 
 #define	HPET_MAX_TIMERS	(32)
+#define	HPET_MAX_IRQ	(32)
 
 /*
  * HPET general capabilities register
@@ -64,7 +65,7 @@ struct hpet {
  */
 
 #define	Tn_INT_ROUTE_CAP_MASK		(0xffffffff00000000ULL)
-#define	Tn_INI_ROUTE_CAP_SHIFT		(32UL)
+#define	Tn_INT_ROUTE_CAP_SHIFT		(32UL)
 #define	Tn_FSB_INT_DELCAP_MASK		(0x8000UL)
 #define	Tn_FSB_INT_DELCAP_SHIFT		(15)
 #define	Tn_FSB_EN_CNF_MASK		(0x4000UL)
@@ -91,23 +92,14 @@ struct hpet {
  * exported interfaces
  */
 
-struct hpet_task {
-	void (*ht_func) (void *);
-	void *ht_data;
-	void *ht_opaque;
-};
-
 struct hpet_data {
 	unsigned long hd_phys_address;
 	void __iomem *hd_address;
 	unsigned short hd_nirqs;
-	unsigned short hd_flags;
 	unsigned int hd_state;	/* timer allocated */
 	unsigned int hd_irq[HPET_MAX_TIMERS];
 };
 
-#define	HPET_DATA_PLATFORM	0x0001	/* platform call to hpet_alloc */
-
 static inline void hpet_reserve_timer(struct hpet_data *hd, int timer)
 {
 	hd->hd_state |= (1 << timer);
@@ -125,7 +117,7 @@ struct hpet_info {
 	unsigned short hi_timer;
 };
 
-#define	HPET_INFO_PERIODIC	0x0001	/* timer is periodic */
+#define HPET_INFO_PERIODIC	0x0010	/* periodic-capable comparator */
 
 #define	HPET_IE_ON	_IO('h', 0x01)	/* interrupt on */
 #define	HPET_IE_OFF	_IO('h', 0x02)	/* interrupt off */
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 6514db8fd2e4..a9d82d6e6bdd 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -8,7 +8,7 @@
 
 #include <linux/init.h>
 #include <linux/ioport.h>
-#include <linux/hdreg.h>
+#include <linux/ata.h>
 #include <linux/blkdev.h>
 #include <linux/proc_fs.h>
 #include <linux/interrupt.h>
@@ -17,6 +17,7 @@
 #include <linux/device.h>
 #include <linux/pci.h>
 #include <linux/completion.h>
+#include <linux/pm.h>
 #ifdef CONFIG_BLK_DEV_IDEACPI
 #include <acpi/acpi.h>
 #endif
@@ -87,12 +88,13 @@ struct ide_io_ports {
 };
 
 #define OK_STAT(stat,good,bad)	(((stat)&((good)|(bad)))==(good))
-#define BAD_R_STAT		(BUSY_STAT   | ERR_STAT)
-#define BAD_W_STAT		(BAD_R_STAT  | WRERR_STAT)
-#define BAD_STAT		(BAD_R_STAT  | DRQ_STAT)
-#define DRIVE_READY		(READY_STAT  | SEEK_STAT)
 
-#define BAD_CRC			(ABRT_ERR    | ICRC_ERR)
+#define BAD_R_STAT	(ATA_BUSY | ATA_ERR)
+#define BAD_W_STAT	(BAD_R_STAT | ATA_DF)
+#define BAD_STAT	(BAD_R_STAT | ATA_DRQ)
+#define DRIVE_READY	(ATA_DRDY | ATA_DSC)
+
+#define BAD_CRC		(ATA_ABORTED | ATA_ICRC)
 
 #define SATA_NR_PORTS		(3)	/* 16 possible ?? */
 
@@ -125,24 +127,41 @@ struct ide_io_ports {
 #define PARTN_BITS	6	/* number of minor dev bits for partitions */
 #define MAX_DRIVES	2	/* per interface; 2 assumed by lots of code */
 #define SECTOR_SIZE	512
-#define SECTOR_WORDS	(SECTOR_SIZE / 4)	/* number of 32bit words per sector */
+
 #define IDE_LARGE_SEEK(b1,b2,t)	(((b1) > (b2) + (t)) || ((b2) > (b1) + (t)))
 
 /*
  * Timeouts for various operations:
  */
-#define WAIT_DRQ	(HZ/10)		/* 100msec - spec allows up to 20ms */
-#define WAIT_READY	(5*HZ)		/* 5sec - some laptops are very slow */
-#define WAIT_PIDENTIFY	(10*HZ)	/* 10sec  - should be less than 3ms (?), if all ATAPI CD is closed at boot */
-#define WAIT_WORSTCASE	(30*HZ)	/* 30sec  - worst case when spinning up */
-#define WAIT_CMD	(10*HZ)	/* 10sec  - maximum wait for an IRQ to happen */
-#define WAIT_MIN_SLEEP	(2*HZ/100)	/* 20msec - minimum sleep time */
+enum {
+	/* spec allows up to 20ms */
+	WAIT_DRQ	= HZ / 10,	/* 100ms */
+	/* some laptops are very slow */
+	WAIT_READY	= 5 * HZ,	/* 5s */
+	/* should be less than 3ms (?), if all ATAPI CD is closed at boot */
+	WAIT_PIDENTIFY	= 10 * HZ,	/* 10s */
+	/* worst case when spinning up */
+	WAIT_WORSTCASE	= 30 * HZ,	/* 30s */
+	/* maximum wait for an IRQ to happen */
+	WAIT_CMD	= 10 * HZ,	/* 10s */
+	/* Some drives require a longer IRQ timeout. */
+	WAIT_FLOPPY_CMD	= 50 * HZ,	/* 50s */
+	/*
+	 * Some drives (for example, Seagate STT3401A Travan) require a very
+	 * long timeout, because they don't return an interrupt or clear their
+	 * BSY bit until after the command completes (even retension commands).
+	 */
+	WAIT_TAPE_CMD	= 900 * HZ,	/* 900s */
+	/* minimum sleep time */
+	WAIT_MIN_SLEEP	= HZ / 50,	/* 20ms */
+};
 
 /*
  * Op codes for special requests to be handled by ide_special_rq().
  * Values should be in the range of 0x20 to 0x3f.
  */
 #define REQ_DRIVE_RESET		0x20
+#define REQ_DEVSET_EXEC		0x21
 
 /*
  * Check for an interrupt and acknowledge the interrupt status
@@ -303,8 +322,8 @@ typedef enum {
 	ide_started,	/* a drive operation was started, handler was set */
 } ide_startstop_t;
 
+struct ide_devset;
 struct ide_driver_s;
-struct ide_settings_s;
 
 #ifdef CONFIG_BLK_DEV_IDEACPI
 struct ide_acpi_drive_link;
@@ -315,10 +334,10 @@ struct ide_acpi_hwif_link;
 enum {
 	IDE_AFLAG_DRQ_INTERRUPT		= (1 << 0),
 	IDE_AFLAG_MEDIA_CHANGED		= (1 << 1),
-
-	/* ide-cd */
 	/* Drive cannot lock the door. */
 	IDE_AFLAG_NO_DOORLOCK		= (1 << 2),
+
+	/* ide-cd */
 	/* Drive cannot eject the disc. */
 	IDE_AFLAG_NO_EJECT		= (1 << 3),
 	/* Drive is a pre ATAPI 1.2 drive. */
@@ -354,21 +373,25 @@ enum {
 	IDE_AFLAG_CLIK_DRIVE		= (1 << 19),
 	/* Requires BH algorithm for packets */
 	IDE_AFLAG_ZIP_DRIVE		= (1 << 20),
+	/* Write protect */
+	IDE_AFLAG_WP			= (1 << 21),
+	/* Supports format progress report */
+	IDE_AFLAG_SRFP			= (1 << 22),
 
 	/* ide-tape */
-	IDE_AFLAG_IGNORE_DSC		= (1 << 21),
+	IDE_AFLAG_IGNORE_DSC		= (1 << 23),
 	/* 0 When the tape position is unknown */
-	IDE_AFLAG_ADDRESS_VALID		= (1 <<	22),
+	IDE_AFLAG_ADDRESS_VALID		= (1 <<	24),
 	/* Device already opened */
-	IDE_AFLAG_BUSY			= (1 << 23),
+	IDE_AFLAG_BUSY			= (1 << 25),
 	/* Attempt to auto-detect the current user block size */
-	IDE_AFLAG_DETECT_BS		= (1 << 24),
+	IDE_AFLAG_DETECT_BS		= (1 << 26),
 	/* Currently on a filemark */
-	IDE_AFLAG_FILEMARK		= (1 << 25),
+	IDE_AFLAG_FILEMARK		= (1 << 27),
 	/* 0 = no tape is loaded, so we don't rewind after ejecting */
-	IDE_AFLAG_MEDIUM_PRESENT	= (1 << 26),
+	IDE_AFLAG_MEDIUM_PRESENT	= (1 << 28),
 
-	IDE_AFLAG_NO_AUTOCLOSE		= (1 << 27),
+	IDE_AFLAG_NO_AUTOCLOSE		= (1 << 29),
 };
 
 struct ide_drive_s {
@@ -380,10 +403,10 @@ struct ide_drive_s {
 	struct request		*rq;	/* current request */
 	struct ide_drive_s 	*next;	/* circular list of hwgroup drives */
 	void		*driver_data;	/* extra driver data */
-	struct hd_driveid	*id;	/* drive model identification info */
+	u16			*id;	/* identification info */
 #ifdef CONFIG_IDE_PROC_FS
 	struct proc_dir_entry *proc;	/* /proc/ide/ directory entry */
-	struct ide_settings_s *settings;/* /proc/ide/ drive settings */
+	const struct ide_proc_devset *settings; /* /proc/ide/ drive settings */
 #endif
 	struct hwif_s		*hwif;	/* actually (ide_hwif_t *) */
 
@@ -395,16 +418,16 @@ struct ide_drive_s {
 	special_t	special;	/* special action flags */
 	select_t	select;		/* basic drive/head select reg value */
 
-	u8	keep_settings;		/* restore settings after drive reset */
-	u8	using_dma;		/* disk is using dma for read/write */
 	u8	retry_pio;		/* retrying dma capable host in pio */
 	u8	state;			/* retry state */
 	u8	waiting_for_dma;	/* dma currently in progress */
-	u8	unmask;			/* okay to unmask other irqs */
-	u8	noflush;		/* don't attempt flushes */
-	u8	dsc_overlap;		/* DSC overlap */
-	u8	nice1;			/* give potential excess bandwidth */
 
+	unsigned keep_settings	: 1;	/* restore settings after drive reset */
+	unsigned using_dma	: 1;	/* disk is using dma for read/write */
+	unsigned unmask		: 1;	/* okay to unmask other irqs */
+	unsigned noflush	: 1;	/* don't attempt flushes */
+	unsigned dsc_overlap	: 1;	/* DSC overlap */
+	unsigned nice1		: 1;	/* give potential excess bandwidth */
 	unsigned present	: 1;	/* drive is physically present */
 	unsigned dead		: 1;	/* device ejected hint */
 	unsigned id_read	: 1;	/* 1=id read from disk 0 = synthetic */
@@ -414,23 +437,22 @@ struct ide_drive_s {
 	unsigned forced_geom	: 1;	/* 1 if hdx=c,h,s was given at boot */
 	unsigned no_unmask	: 1;	/* disallow setting unmask bit */
 	unsigned no_io_32bit	: 1;	/* disallow enabling 32bit I/O */
-	unsigned atapi_overlap	: 1;	/* ATAPI overlap (not supported) */
 	unsigned doorlocking	: 1;	/* for removable only: door lock/unlock works */
 	unsigned nodma		: 1;	/* disallow DMA */
-	unsigned remap_0_to_1	: 1;	/* 0=noremap, 1=remap 0->1 (for EZDrive) */
 	unsigned blocked        : 1;	/* 1=powermanagment told us not to do anything, so sleep nicely */
 	unsigned scsi		: 1;	/* 0=default, 1=ide-scsi emulation */
 	unsigned sleeping	: 1;	/* 1=sleeping & sleep field valid */
 	unsigned post_reset	: 1;
 	unsigned udma33_warned	: 1;
+	unsigned addressing	: 2;	/* 0=28-bit, 1=48-bit, 2=48-bit doing 28-bit */
+	unsigned wcache		: 1;	/* status of write cache */
+	unsigned nowerr		: 1;	/* used for ignoring ATA_DF */
 
-	u8	addressing;	/* 0=28-bit, 1=48-bit, 2=48-bit doing 28-bit */
         u8	quirk_list;	/* considered quirky, set for a specific host */
         u8	init_speed;	/* transfer rate set at boot */
         u8	current_speed;	/* current transfer rate set */
 	u8	desired_speed;	/* desired transfer rate set */
         u8	dn;		/* now wide spread use */
-        u8	wcache;		/* status of write cache */
 	u8	acoustic;	/* acoustic management */
 	u8	media;		/* disk, cdrom, tape, floppy, ... */
 	u8	ready_stat;	/* min status value for drive ready */
@@ -438,9 +460,7 @@ struct ide_drive_s {
 	u8	mult_req;	/* requested multiple sector setting */
 	u8	tune_req;	/* requested drive tuning setting */
 	u8	io_32bit;	/* 0=16-bit, 1=32-bit, 2/3=32bit+sync */
-	u8	bad_wstat;	/* used for ignoring WRERR_STAT */
-	u8	nowerr;		/* used for ignoring WRERR_STAT */
-	u8	sect0;		/* offset of first sector for DM6:DDO */
+	u8	bad_wstat;	/* used for ignoring ATA_DF */
 	u8	head;		/* "real" number of heads */
 	u8	sect;		/* "real" sectors per track */
 	u8	bios_head;	/* BIOS/fdisk/LILO number of heads */
@@ -474,10 +494,6 @@ typedef struct ide_drive_s ide_drive_t;
 
 #define to_ide_device(dev)container_of(dev, ide_drive_t, gendev)
 
-#define IDE_CHIPSET_PCI_MASK	\
-    ((1<<ide_pci)|(1<<ide_cmd646)|(1<<ide_ali14xx))
-#define IDE_CHIPSET_IS_PCI(c)	((IDE_CHIPSET_PCI_MASK >> (c)) & 1)
-
 struct ide_task_s;
 struct ide_port_info;
 
@@ -567,7 +583,6 @@ typedef struct hwif_s {
 	u8 major;	/* our major number */
 	u8 index;	/* 0 for ide0; 1 for ide1; ... */
 	u8 channel;	/* for dual-port chips: 0=primary, 1=secondary */
-	u8 bus_state;	/* power state of the IDE bus */
 
 	u32 host_flags;
 
@@ -645,6 +660,7 @@ struct ide_host {
 	ide_hwif_t	*ports[MAX_HWIFS];
 	unsigned int	n_ports;
 	struct device	*dev[2];
+	unsigned int	(*init_chipset)(struct pci_dev *);
 	unsigned long	host_flags;
 	void		*host_priv;
 };
@@ -692,9 +708,61 @@ typedef struct ide_driver_s ide_driver_t;
 
 extern struct mutex ide_setting_mtx;
 
-int set_io_32bit(ide_drive_t *, int);
-int set_pio_mode(ide_drive_t *, int);
-int set_using_dma(ide_drive_t *, int);
+/*
+ * configurable drive settings
+ */
+
+#define DS_SYNC	(1 << 0)
+
+struct ide_devset {
+	int		(*get)(ide_drive_t *);
+	int		(*set)(ide_drive_t *, int);
+	unsigned int	flags;
+};
+
+#define __DEVSET(_flags, _get, _set) { \
+	.flags	= _flags, \
+	.get	= _get,	\
+	.set	= _set,	\
+}
+
+#define ide_devset_get(name, field) \
+static int get_##name(ide_drive_t *drive) \
+{ \
+	return drive->field; \
+}
+
+#define ide_devset_set(name, field) \
+static int set_##name(ide_drive_t *drive, int arg) \
+{ \
+	drive->field = arg; \
+	return 0; \
+}
+
+#define __IDE_DEVSET(_name, _flags, _get, _set) \
+const struct ide_devset ide_devset_##_name = \
+	__DEVSET(_flags, _get, _set)
+
+#define IDE_DEVSET(_name, _flags, _get, _set) \
+static __IDE_DEVSET(_name, _flags, _get, _set)
+
+#define ide_devset_rw(_name, _func) \
+IDE_DEVSET(_name, 0, get_##_func, set_##_func)
+
+#define ide_devset_w(_name, _func) \
+IDE_DEVSET(_name, 0, NULL, set_##_func)
+
+#define ide_devset_rw_sync(_name, _func) \
+IDE_DEVSET(_name, DS_SYNC, get_##_func, set_##_func)
+
+#define ide_decl_devset(_name) \
+extern const struct ide_devset ide_devset_##_name
+
+ide_decl_devset(io_32bit);
+ide_decl_devset(keepsettings);
+ide_decl_devset(pio_mode);
+ide_decl_devset(unmaskirq);
+ide_decl_devset(using_dma);
 
 /* ATAPI packet command flags */
 enum {
@@ -710,6 +778,12 @@ enum {
 	PC_FLAG_TIMEDOUT		= (1 << 7),
 };
 
+/*
+ * With each packet command, we allocate a buffer of IDE_PC_BUFFER_SIZE bytes.
+ * This is used for several packet commands (not for READ/WRITE commands).
+ */
+#define IDE_PC_BUFFER_SIZE	256
+
 struct ide_atapi_pc {
 	/* actual packet bytes */
 	u8 c[12];
@@ -739,7 +813,7 @@ struct ide_atapi_pc {
 	 * those are more or less driver-specific and some of them are subject
 	 * to change/removal later.
 	 */
-	u8 pc_buf[256];
+	u8 pc_buf[IDE_PC_BUFFER_SIZE];
 
 	/* idetape only */
 	struct idetape_bh *bh;
@@ -757,37 +831,34 @@ struct ide_atapi_pc {
 
 #ifdef CONFIG_IDE_PROC_FS
 /*
- * configurable drive settings
+ * /proc/ide interface
  */
 
-#define TYPE_INT	0
-#define TYPE_BYTE	1
-#define TYPE_SHORT	2
+#define ide_devset_rw_field(_name, _field) \
+ide_devset_get(_name, _field); \
+ide_devset_set(_name, _field); \
+IDE_DEVSET(_name, DS_SYNC, get_##_name, set_##_name)
+
+struct ide_proc_devset {
+	const char		*name;
+	const struct ide_devset	*setting;
+	int			min, max;
+	int			(*mulf)(ide_drive_t *);
+	int			(*divf)(ide_drive_t *);
+};
 
-#define SETTING_READ	(1 << 0)
-#define SETTING_WRITE	(1 << 1)
-#define SETTING_RW	(SETTING_READ | SETTING_WRITE)
+#define __IDE_PROC_DEVSET(_name, _min, _max, _mulf, _divf) { \
+	.name = __stringify(_name), \
+	.setting = &ide_devset_##_name, \
+	.min = _min, \
+	.max = _max, \
+	.mulf = _mulf, \
+	.divf = _divf, \
+}
 
-typedef int (ide_procset_t)(ide_drive_t *, int);
-typedef struct ide_settings_s {
-	char			*name;
-	int			rw;
-	int			data_type;
-	int			min;
-	int			max;
-	int			mul_factor;
-	int			div_factor;
-	void			*data;
-	ide_procset_t		*set;
-	int			auto_remove;
-	struct ide_settings_s	*next;
-} ide_settings_t;
-
-int ide_add_setting(ide_drive_t *, const char *, int, int, int, int, int, int, void *, ide_procset_t *set);
+#define IDE_PROC_DEVSET(_name, _min, _max) \
+__IDE_PROC_DEVSET(_name, _min, _max, NULL, NULL)
 
-/*
- * /proc/ide interface
- */
 typedef struct {
 	const char	*name;
 	mode_t		mode;
@@ -804,8 +875,6 @@ void ide_proc_unregister_port(ide_hwif_t *);
 void ide_proc_register_driver(ide_drive_t *, ide_driver_t *);
 void ide_proc_unregister_driver(ide_drive_t *, ide_driver_t *);
 
-void ide_add_generic_settings(ide_drive_t *);
-
 read_proc_t proc_ide_read_capacity;
 read_proc_t proc_ide_read_geometry;
 
@@ -833,7 +902,6 @@ static inline void ide_proc_unregister_device(ide_drive_t *drive) { ; }
 static inline void ide_proc_unregister_port(ide_hwif_t *hwif) { ; }
 static inline void ide_proc_register_driver(ide_drive_t *drive, ide_driver_t *driver) { ; }
 static inline void ide_proc_unregister_driver(ide_drive_t *drive, ide_driver_t *driver) { ; }
-static inline void ide_add_generic_settings(ide_drive_t *drive) { ; }
 #define PROC_IDE_READ_RETURN(page,start,off,count,eof,len) return 0;
 #endif
 
@@ -879,7 +947,6 @@ enum {
 struct ide_driver_s {
 	const char			*version;
 	u8				media;
-	unsigned supports_dsc_overlap	: 1;
 	ide_startstop_t	(*do_request)(ide_drive_t *, struct request *, sector_t);
 	int		(*end_request)(ide_drive_t *, int, int);
 	ide_startstop_t	(*error)(ide_drive_t *, struct request *rq, u8, u8);
@@ -889,7 +956,8 @@ struct ide_driver_s {
 	void		(*resume)(ide_drive_t *);
 	void		(*shutdown)(ide_drive_t *);
 #ifdef CONFIG_IDE_PROC_FS
-	ide_proc_entry_t	*proc;
+	ide_proc_entry_t		*proc;
+	const struct ide_proc_devset	*settings;
 #endif
 };
 
@@ -898,7 +966,17 @@ struct ide_driver_s {
 int ide_device_get(ide_drive_t *);
 void ide_device_put(ide_drive_t *);
 
-int generic_ide_ioctl(ide_drive_t *, struct file *, struct block_device *, unsigned, unsigned long);
+struct ide_ioctl_devset {
+	unsigned int	get_ioctl;
+	unsigned int	set_ioctl;
+	const struct ide_devset *setting;
+};
+
+int ide_setting_ioctl(ide_drive_t *, struct block_device *, unsigned int,
+		      unsigned long, const struct ide_ioctl_devset *);
+
+int generic_ide_ioctl(ide_drive_t *, struct file *, struct block_device *,
+		      unsigned, unsigned long);
 
 extern int ide_vlb_clk;
 extern int ide_pci_clk;
@@ -920,14 +998,19 @@ ide_startstop_t __ide_error(ide_drive_t *, struct request *, u8, u8);
 
 ide_startstop_t ide_error (ide_drive_t *drive, const char *msg, byte stat);
 
-extern void ide_fix_driveid(struct hd_driveid *);
+void ide_fix_driveid(u16 *);
 
 extern void ide_fixstring(u8 *, const int, const int);
 
+int ide_busy_sleep(ide_hwif_t *, unsigned long, int);
+
 int ide_wait_stat(ide_startstop_t *, ide_drive_t *, u8, u8, unsigned long);
 
 extern ide_startstop_t ide_do_reset (ide_drive_t *);
 
+extern int ide_devset_execute(ide_drive_t *drive,
+			      const struct ide_devset *setting, int arg);
+
 extern void ide_do_drive_cmd(ide_drive_t *, struct request *);
 
 extern void ide_end_drive_cmd(ide_drive_t *, u8, u8);
@@ -1051,6 +1134,8 @@ void ide_tf_read(ide_drive_t *, ide_task_t *);
 void ide_input_data(ide_drive_t *, struct request *, void *, unsigned int);
 void ide_output_data(ide_drive_t *, struct request *, void *, unsigned int);
 
+int ide_io_buffers(ide_drive_t *, struct ide_atapi_pc *, unsigned int, int);
+
 extern void SELECT_DRIVE(ide_drive_t *);
 void SELECT_MASK(ide_drive_t *, int);
 
@@ -1061,11 +1146,36 @@ extern int drive_is_ready(ide_drive_t *);
 
 void ide_pktcmd_tf_load(ide_drive_t *, u32, u16, u8);
 
+int ide_check_atapi_device(ide_drive_t *, const char *);
+
+void ide_init_pc(struct ide_atapi_pc *);
+
+/*
+ * Special requests for ide-tape block device strategy routine.
+ *
+ * In order to service a character device command, we add special requests to
+ * the tail of our block device request queue and wait for their completion.
+ */
+enum {
+	REQ_IDETAPE_PC1		= (1 << 0), /* packet command (first stage) */
+	REQ_IDETAPE_PC2		= (1 << 1), /* packet command (second stage) */
+	REQ_IDETAPE_READ	= (1 << 2),
+	REQ_IDETAPE_WRITE	= (1 << 3),
+};
+
+void ide_queue_pc_head(ide_drive_t *, struct gendisk *, struct ide_atapi_pc *,
+		       struct request *);
+int ide_queue_pc_tail(ide_drive_t *, struct gendisk *, struct ide_atapi_pc *);
+
+int ide_do_test_unit_ready(ide_drive_t *, struct gendisk *);
+int ide_do_start_stop(ide_drive_t *, struct gendisk *, int);
+int ide_set_media_lock(ide_drive_t *, struct gendisk *, int);
+
 ide_startstop_t ide_pc_intr(ide_drive_t *drive, struct ide_atapi_pc *pc,
 	ide_handler_t *handler, unsigned int timeout, ide_expiry_t *expiry,
 	void (*update_buffers)(ide_drive_t *, struct ide_atapi_pc *),
 	void (*retry_pc)(ide_drive_t *), void (*dsc_handle)(ide_drive_t *),
-	void (*io_buffers)(ide_drive_t *, struct ide_atapi_pc *, unsigned int,
+	int (*io_buffers)(ide_drive_t *, struct ide_atapi_pc *, unsigned int,
 			   int));
 ide_startstop_t ide_transfer_pc(ide_drive_t *, struct ide_atapi_pc *,
 				ide_handler_t *, unsigned int, ide_expiry_t *);
@@ -1080,8 +1190,6 @@ int ide_raw_taskfile(ide_drive_t *, ide_task_t *, u8 *, u16);
 int ide_no_data_taskfile(ide_drive_t *, ide_task_t *);
 
 int ide_taskfile_ioctl(ide_drive_t *, unsigned int, unsigned long);
-int ide_cmd_ioctl(ide_drive_t *, unsigned int, unsigned long);
-int ide_task_ioctl(ide_drive_t *, unsigned int, unsigned long);
 
 extern int ide_driveid_update(ide_drive_t *);
 extern int ide_config_drive_speed(ide_drive_t *, u8);
@@ -1092,7 +1200,6 @@ extern int ide_wait_not_busy(ide_hwif_t *hwif, unsigned long timeout);
 
 extern void ide_stall_queue(ide_drive_t *drive, unsigned long timeout);
 
-extern int ide_spin_wait_hwgroup(ide_drive_t *);
 extern void ide_timer_expiry(unsigned long);
 extern irqreturn_t ide_intr(int irq, void *dev_id);
 extern void do_ide_request(struct request_queue *);
@@ -1229,6 +1336,14 @@ int ide_pci_init_two(struct pci_dev *, struct pci_dev *,
 		     const struct ide_port_info *, void *);
 void ide_pci_remove(struct pci_dev *);
 
+#ifdef CONFIG_PM
+int ide_pci_suspend(struct pci_dev *, pm_message_t);
+int ide_pci_resume(struct pci_dev *);
+#else
+#define ide_pci_suspend NULL
+#define ide_pci_resume NULL
+#endif
+
 void ide_map_sg(ide_drive_t *, struct request *);
 void ide_init_sg_cmd(ide_drive_t *, struct request *);
 
@@ -1240,7 +1355,7 @@ struct drive_list_entry {
 	const char *id_firmware;
 };
 
-int ide_in_drive_list(struct hd_driveid *, const struct drive_list_entry *);
+int ide_in_drive_list(u16 *, const struct drive_list_entry *);
 
 #ifdef CONFIG_BLK_DEV_IDEDMA
 int __ide_dma_bad_drive(ide_drive_t *);
@@ -1347,24 +1462,6 @@ const char *ide_xfer_verbose(u8 mode);
 extern void ide_toggle_bounce(ide_drive_t *drive, int on);
 extern int ide_set_xfer_rate(ide_drive_t *drive, u8 rate);
 
-static inline int ide_dev_has_iordy(struct hd_driveid *id)
-{
-	return ((id->field_valid & 2) && (id->capability & 8)) ? 1 : 0;
-}
-
-static inline int ide_dev_is_sata(struct hd_driveid *id)
-{
-	/*
-	 * See if word 93 is 0 AND drive is at least ATA-5 compatible
-	 * verifying that word 80 by casting it to a signed type --
-	 * this trick allows us to filter out the reserved values of
-	 * 0x0000 and 0xffff along with the earlier ATA revisions...
-	 */
-	if (id->hw_config == 0 && (short)id->major_rev_num >= 0x0020)
-		return 1;
-	return 0;
-}
-
 u64 ide_get_lba_addr(struct ide_taskfile *, int);
 u8 ide_dump_status(ide_drive_t *, const char *, u8);
 
@@ -1436,13 +1533,6 @@ extern struct mutex ide_cfg_mtx;
 extern struct bus_type ide_bus_type;
 extern struct class *ide_port_class;
 
-/* check if CACHE FLUSH (EXT) command is supported (bits defined in ATA-6) */
-#define ide_id_has_flush_cache(id)	((id)->cfs_enable_2 & 0x3000)
-
-/* some Maxtor disks have bit 13 defined incorrectly so check bit 10 too */
-#define ide_id_has_flush_cache_ext(id)	\
-	(((id)->cfs_enable_2 & 0x2400) == 0x2400)
-
 static inline void ide_dump_identify(u8 *id)
 {
 	print_hex_dump(KERN_INFO, "", DUMP_PREFIX_NONE, 16, 2, id, 512, 0);
@@ -1453,10 +1543,10 @@ static inline int hwif_to_node(ide_hwif_t *hwif)
 	return hwif->dev ? dev_to_node(hwif->dev) : -1;
 }
 
-static inline ide_drive_t *ide_get_paired_drive(ide_drive_t *drive)
+static inline ide_drive_t *ide_get_pair_dev(ide_drive_t *drive)
 {
-	ide_hwif_t *hwif	= HWIF(drive);
+	ide_drive_t *peer = &drive->hwif->drives[(drive->dn ^ 1) & 1];
 
-	return &hwif->drives[(drive->dn ^ 1) & 1];
+	return peer->present ? peer : NULL;
 }
 #endif /* _IDE_H */
diff --git a/include/linux/ioport.h b/include/linux/ioport.h
index 350033e8f4e1..ee9bcc6f32b6 100644
--- a/include/linux/ioport.h
+++ b/include/linux/ioport.h
@@ -108,6 +108,9 @@ extern struct resource iomem_resource;
 
 extern int request_resource(struct resource *root, struct resource *new);
 extern int release_resource(struct resource *new);
+extern void reserve_region_with_split(struct resource *root,
+			     resource_size_t start, resource_size_t end,
+			     const char *name);
 extern int insert_resource(struct resource *parent, struct resource *new);
 extern void insert_resource_expand_to_fit(struct resource *root, struct resource *new);
 extern int allocate_resource(struct resource *root, struct resource *new,
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 8ccb462ea42c..8d9411bc60f6 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -62,6 +62,7 @@ typedef	void (*irq_flow_handler_t)(unsigned int irq,
 #define IRQ_MOVE_PENDING	0x00200000	/* need to re-target IRQ destination */
 #define IRQ_NO_BALANCING	0x00400000	/* IRQ is excluded from balancing */
 #define IRQ_SPURIOUS_DISABLED	0x00800000	/* IRQ was disabled by the spurious trap */
+#define IRQ_MOVE_PCNTXT	0x01000000	/* IRQ migration from process context */
 
 #ifdef CONFIG_IRQ_PER_CPU
 # define CHECK_IRQ_PER_CPU(var) ((var) & IRQ_PER_CPU)
diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
index 3dd209007098..d2e91ea998fd 100644
--- a/include/linux/jbd2.h
+++ b/include/linux/jbd2.h
@@ -850,7 +850,8 @@ struct journal_s
 	 */
 	struct block_device	*j_dev;
 	int			j_blocksize;
-	unsigned long long		j_blk_offset;
+	unsigned long long	j_blk_offset;
+	char			j_devname[BDEVNAME_SIZE+24];
 
 	/*
 	 * Device which holds the client fs.  For internal journal this will be
@@ -966,6 +967,9 @@ struct journal_s
 #define JBD2_FLUSHED	0x008	/* The journal superblock has been flushed */
 #define JBD2_LOADED	0x010	/* The journal superblock has been loaded */
 #define JBD2_BARRIER	0x020	/* Use IDE barriers */
+#define JBD2_ABORT_ON_SYNCDATA_ERR	0x040	/* Abort the journal on file
+						 * data write error in ordered
+						 * mode */
 
 /*
  * Function declarations for the journaling transaction and buffer
@@ -1059,7 +1063,7 @@ extern void	   jbd2_journal_clear_features
 		   (journal_t *, unsigned long, unsigned long, unsigned long);
 extern int	   jbd2_journal_create     (journal_t *);
 extern int	   jbd2_journal_load       (journal_t *journal);
-extern void	   jbd2_journal_destroy    (journal_t *);
+extern int	   jbd2_journal_destroy    (journal_t *);
 extern int	   jbd2_journal_recover    (journal_t *journal);
 extern int	   jbd2_journal_wipe       (journal_t *, int);
 extern int	   jbd2_journal_skip_recovery	(journal_t *);
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 2651f805ba6d..75d81f157d2e 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -182,7 +182,7 @@ extern int vsscanf(const char *, const char *, va_list)
 
 extern int get_option(char **str, int *pint);
 extern char *get_options(const char *str, int nints, int *ints);
-extern unsigned long long memparse(char *ptr, char **retptr);
+extern unsigned long long memparse(const char *ptr, char **retptr);
 
 extern int core_kernel_text(unsigned long addr);
 extern int __kernel_text_address(unsigned long addr);
diff --git a/include/linux/key.h b/include/linux/key.h
index c45c962d1cc5..1b70e35a71e3 100644
--- a/include/linux/key.h
+++ b/include/linux/key.h
@@ -299,6 +299,7 @@ extern void key_init(void);
 #define key_validate(k)			0
 #define key_serial(k)			0
 #define key_get(k) 			({ NULL; })
+#define key_revoke(k)			do { } while(0)
 #define key_put(k)			do { } while(0)
 #define key_ref_put(k)			do { } while(0)
 #define make_key_ref(k, p)			({ NULL; })
diff --git a/include/linux/magic.h b/include/linux/magic.h
index 1fa0c2ce4dec..f7f3fdddbef0 100644
--- a/include/linux/magic.h
+++ b/include/linux/magic.h
@@ -6,6 +6,10 @@
 #define AFS_SUPER_MAGIC                0x5346414F
 #define AUTOFS_SUPER_MAGIC	0x0187
 #define CODA_SUPER_MAGIC	0x73757245
+#define DEBUGFS_MAGIC          0x64626720
+#define SYSFS_MAGIC		0x62656572
+#define SECURITYFS_MAGIC	0x73636673
+#define TMPFS_MAGIC		0x01021994
 #define EFS_SUPER_MAGIC		0x414A53
 #define EXT2_SUPER_MAGIC	0xEF53
 #define EXT3_SUPER_MAGIC	0xEF53
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 72a15dc26bbf..c61ba10768ea 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -7,6 +7,7 @@
 
 #include <linux/gfp.h>
 #include <linux/list.h>
+#include <linux/mmdebug.h>
 #include <linux/mmzone.h>
 #include <linux/rbtree.h>
 #include <linux/prio_tree.h>
@@ -219,12 +220,6 @@ struct inode;
  */
 #include <linux/page-flags.h>
 
-#ifdef CONFIG_DEBUG_VM
-#define VM_BUG_ON(cond) BUG_ON(cond)
-#else
-#define VM_BUG_ON(condition) do { } while(0)
-#endif
-
 /*
  * Methods to modify the page usage count.
  *
@@ -919,7 +914,7 @@ static inline pmd_t *pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long a
 }
 #endif /* CONFIG_MMU && !__ARCH_HAS_4LEVEL_HACK */
 
-#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
+#if USE_SPLIT_PTLOCKS
 /*
  * We tuck a spinlock to guard each pagetable page into its struct page,
  * at page->private, with BUILD_BUG_ON to make sure that this will not
@@ -932,14 +927,14 @@ static inline pmd_t *pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long a
 } while (0)
 #define pte_lock_deinit(page)	((page)->mapping = NULL)
 #define pte_lockptr(mm, pmd)	({(void)(mm); __pte_lockptr(pmd_page(*(pmd)));})
-#else
+#else	/* !USE_SPLIT_PTLOCKS */
 /*
  * We use mm->page_table_lock to guard all pagetable pages of the mm.
  */
 #define pte_lock_init(page)	do {} while (0)
 #define pte_lock_deinit(page)	do {} while (0)
 #define pte_lockptr(mm, pmd)	({(void)(pmd); &(mm)->page_table_lock;})
-#endif /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */
+#endif /* USE_SPLIT_PTLOCKS */
 
 static inline void pgtable_page_ctor(struct page *page)
 {
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index bf334138c7c1..9d49fa36bbef 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -21,11 +21,13 @@
 
 struct address_space;
 
-#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
+#define USE_SPLIT_PTLOCKS	(NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS)
+
+#if USE_SPLIT_PTLOCKS
 typedef atomic_long_t mm_counter_t;
-#else  /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */
+#else  /* !USE_SPLIT_PTLOCKS */
 typedef unsigned long mm_counter_t;
-#endif /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */
+#endif /* !USE_SPLIT_PTLOCKS */
 
 /*
  * Each physical page in the system has a struct page associated with
@@ -65,7 +67,7 @@ struct page {
 						 * see PAGE_MAPPING_ANON below.
 						 */
 	    };
-#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
+#if USE_SPLIT_PTLOCKS
 	    spinlock_t ptl;
 #endif
 	    struct kmem_cache *slab;	/* SLUB: Pointer to slab */
diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
index 9c288c909878..bde891f64591 100644
--- a/include/linux/mmc/host.h
+++ b/include/linux/mmc/host.h
@@ -65,7 +65,7 @@ struct mmc_host_ops {
 	 *   -ENOSYS when not supported (equal to NULL callback)
 	 *   or a negative errno value when something bad happened
 	 *
-	 * Return values for the get_ro callback should be:
+	 * Return values for the get_cd callback should be:
 	 *   0 for a absent card
 	 *   1 for a present card
 	 *   -ENOSYS when not supported (equal to NULL callback)
diff --git a/include/linux/mmdebug.h b/include/linux/mmdebug.h
new file mode 100644
index 000000000000..8a5509877192
--- /dev/null
+++ b/include/linux/mmdebug.h
@@ -0,0 +1,18 @@
+#ifndef LINUX_MM_DEBUG_H
+#define LINUX_MM_DEBUG_H 1
+
+#include <linux/autoconf.h>
+
+#ifdef CONFIG_DEBUG_VM
+#define VM_BUG_ON(cond) BUG_ON(cond)
+#else
+#define VM_BUG_ON(cond) do { } while (0)
+#endif
+
+#ifdef CONFIG_DEBUG_VIRTUAL
+#define VIRTUAL_BUG_ON(cond) BUG_ON(cond)
+#else
+#define VIRTUAL_BUG_ON(cond) do { } while (0)
+#endif
+
+#endif
diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h
index c4db5827963d..3481a7d5bc0a 100644
--- a/include/linux/mod_devicetable.h
+++ b/include/linux/mod_devicetable.h
@@ -388,5 +388,52 @@ struct i2c_device_id {
 			__attribute__((aligned(sizeof(kernel_ulong_t))));
 };
 
+/* dmi */
+enum dmi_field {
+	DMI_NONE,
+	DMI_BIOS_VENDOR,
+	DMI_BIOS_VERSION,
+	DMI_BIOS_DATE,
+	DMI_SYS_VENDOR,
+	DMI_PRODUCT_NAME,
+	DMI_PRODUCT_VERSION,
+	DMI_PRODUCT_SERIAL,
+	DMI_PRODUCT_UUID,
+	DMI_BOARD_VENDOR,
+	DMI_BOARD_NAME,
+	DMI_BOARD_VERSION,
+	DMI_BOARD_SERIAL,
+	DMI_BOARD_ASSET_TAG,
+	DMI_CHASSIS_VENDOR,
+	DMI_CHASSIS_TYPE,
+	DMI_CHASSIS_VERSION,
+	DMI_CHASSIS_SERIAL,
+	DMI_CHASSIS_ASSET_TAG,
+	DMI_STRING_MAX,
+};
+
+struct dmi_strmatch {
+	unsigned char slot;
+	char substr[79];
+};
+
+#ifndef __KERNEL__
+struct dmi_system_id {
+	kernel_ulong_t callback;
+	kernel_ulong_t ident;
+	struct dmi_strmatch matches[4];
+	kernel_ulong_t driver_data
+			__attribute__((aligned(sizeof(kernel_ulong_t))));
+};
+#else
+struct dmi_system_id {
+	int (*callback)(const struct dmi_system_id *);
+	const char *ident;
+	struct dmi_strmatch matches[4];
+	void *driver_data;
+};
+#endif
+
+#define DMI_MATCH(a, b)	{ a, b }
 
 #endif /* LINUX_MOD_DEVICETABLE_H */
diff --git a/include/linux/oprofile.h b/include/linux/oprofile.h
index 041bb31100f4..bcb8f725427c 100644
--- a/include/linux/oprofile.h
+++ b/include/linux/oprofile.h
@@ -36,6 +36,8 @@
 #define XEN_ENTER_SWITCH_CODE		10
 #define SPU_PROFILING_CODE		11
 #define SPU_CTX_SWITCH_CODE		12
+#define IBS_FETCH_CODE			13
+#define IBS_OP_CODE			14
 
 struct super_block;
 struct dentry;
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index f63b5455801c..1176f1f177e2 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1533,7 +1533,9 @@
 #define PCI_DEVICE_ID_MARVELL_GT64260	0x6430
 #define PCI_DEVICE_ID_MARVELL_MV64360	0x6460
 #define PCI_DEVICE_ID_MARVELL_MV64460	0x6480
-#define PCI_DEVICE_ID_MARVELL_CAFE_SD	0x4101
+#define PCI_DEVICE_ID_MARVELL_88ALP01_NAND	0x4100
+#define PCI_DEVICE_ID_MARVELL_88ALP01_SD	0x4101
+#define PCI_DEVICE_ID_MARVELL_88ALP01_CCIC	0x4102
 
 #define PCI_VENDOR_ID_V3		0x11b0
 #define PCI_DEVICE_ID_V3_V960		0x0001
diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index fac3337547eb..9f2a3751873a 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -23,12 +23,19 @@
 	__attribute__((__section__(SHARED_ALIGNED_SECTION)))		\
 	PER_CPU_ATTRIBUTES __typeof__(type) per_cpu__##name		\
 	____cacheline_aligned_in_smp
+
+#define DEFINE_PER_CPU_PAGE_ALIGNED(type, name)			\
+	__attribute__((__section__(".data.percpu.page_aligned")))	\
+	PER_CPU_ATTRIBUTES __typeof__(type) per_cpu__##name
 #else
 #define DEFINE_PER_CPU(type, name)					\
 	PER_CPU_ATTRIBUTES __typeof__(type) per_cpu__##name
 
 #define DEFINE_PER_CPU_SHARED_ALIGNED(type, name)		      \
 	DEFINE_PER_CPU(type, name)
+
+#define DEFINE_PER_CPU_PAGE_ALIGNED(type, name)		      \
+	DEFINE_PER_CPU(type, name)
 #endif
 
 #define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(per_cpu__##var)
diff --git a/include/linux/percpu_counter.h b/include/linux/percpu_counter.h
index 208388835357..9007ccdfc112 100644
--- a/include/linux/percpu_counter.h
+++ b/include/linux/percpu_counter.h
@@ -35,7 +35,7 @@ int percpu_counter_init_irq(struct percpu_counter *fbc, s64 amount);
 void percpu_counter_destroy(struct percpu_counter *fbc);
 void percpu_counter_set(struct percpu_counter *fbc, s64 amount);
 void __percpu_counter_add(struct percpu_counter *fbc, s64 amount, s32 batch);
-s64 __percpu_counter_sum(struct percpu_counter *fbc, int set);
+s64 __percpu_counter_sum(struct percpu_counter *fbc);
 
 static inline void percpu_counter_add(struct percpu_counter *fbc, s64 amount)
 {
@@ -44,19 +44,13 @@ static inline void percpu_counter_add(struct percpu_counter *fbc, s64 amount)
 
 static inline s64 percpu_counter_sum_positive(struct percpu_counter *fbc)
 {
-	s64 ret = __percpu_counter_sum(fbc, 0);
+	s64 ret = __percpu_counter_sum(fbc);
 	return ret < 0 ? 0 : ret;
 }
 
-static inline s64 percpu_counter_sum_and_set(struct percpu_counter *fbc)
-{
-	return __percpu_counter_sum(fbc, 1);
-}
-
-
 static inline s64 percpu_counter_sum(struct percpu_counter *fbc)
 {
-	return __percpu_counter_sum(fbc, 0);
+	return __percpu_counter_sum(fbc);
 }
 
 static inline s64 percpu_counter_read(struct percpu_counter *fbc)
diff --git a/include/linux/rtc/m48t59.h b/include/linux/rtc/m48t59.h
index e8c7c21ceb1f..6fc961459b4a 100644
--- a/include/linux/rtc/m48t59.h
+++ b/include/linux/rtc/m48t59.h
@@ -18,40 +18,47 @@
 /*
  * M48T59 Register Offset
  */
-#define M48T59_YEAR		0x1fff
-#define M48T59_MONTH		0x1ffe
-#define M48T59_MDAY		0x1ffd	/* Day of Month */
-#define M48T59_WDAY		0x1ffc	/* Day of Week */
+#define M48T59_YEAR		0xf
+#define M48T59_MONTH		0xe
+#define M48T59_MDAY		0xd	/* Day of Month */
+#define M48T59_WDAY		0xc	/* Day of Week */
 #define M48T59_WDAY_CB			0x20	/* Century Bit */
 #define M48T59_WDAY_CEB			0x10	/* Century Enable Bit */
-#define M48T59_HOUR		0x1ffb
-#define M48T59_MIN		0x1ffa
-#define M48T59_SEC		0x1ff9
-#define M48T59_CNTL		0x1ff8
+#define M48T59_HOUR		0xb
+#define M48T59_MIN		0xa
+#define M48T59_SEC		0x9
+#define M48T59_CNTL		0x8
 #define M48T59_CNTL_READ		0x40
 #define M48T59_CNTL_WRITE		0x80
-#define M48T59_WATCHDOG		0x1ff7
-#define M48T59_INTR		0x1ff6
+#define M48T59_WATCHDOG		0x7
+#define M48T59_INTR		0x6
 #define M48T59_INTR_AFE			0x80	/* Alarm Interrupt Enable */
 #define M48T59_INTR_ABE			0x20
-#define M48T59_ALARM_DATE	0x1ff5
-#define M48T59_ALARM_HOUR	0x1ff4
-#define M48T59_ALARM_MIN	0x1ff3
-#define M48T59_ALARM_SEC	0x1ff2
-#define M48T59_UNUSED		0x1ff1
-#define M48T59_FLAGS		0x1ff0
+#define M48T59_ALARM_DATE	0x5
+#define M48T59_ALARM_HOUR	0x4
+#define M48T59_ALARM_MIN	0x3
+#define M48T59_ALARM_SEC	0x2
+#define M48T59_UNUSED		0x1
+#define M48T59_FLAGS		0x0
 #define M48T59_FLAGS_WDT		0x80	/* watchdog timer expired */
 #define M48T59_FLAGS_AF			0x40	/* alarm */
 #define M48T59_FLAGS_BF			0x10	/* low battery */
 
-#define M48T59_NVRAM_SIZE	0x1ff0
+#define M48T59RTC_TYPE_M48T59	0 /* to keep compatibility */
+#define M48T59RTC_TYPE_M48T02	1
+#define M48T59RTC_TYPE_M48T08	2
 
 struct m48t59_plat_data {
-	/* The method to access M48T59 registers,
-	 * NOTE: The 'ofs' should be 0x00~0x1fff
-	 */
+	/* The method to access M48T59 registers */
 	void (*write_byte)(struct device *dev, u32 ofs, u8 val);
 	unsigned char (*read_byte)(struct device *dev, u32 ofs);
+
+	int type; /* RTC model */
+
+	/* ioaddr mapped externally */
+	void __iomem *ioaddr;
+	/* offset to RTC registers, automatically set according to the type */
+	unsigned int offset;
 };
 
 #endif /* _LINUX_RTC_M48T59_H_ */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 5d0819ee442a..c226c7b82946 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -352,7 +352,7 @@ arch_get_unmapped_area_topdown(struct file *filp, unsigned long addr,
 extern void arch_unmap_area(struct mm_struct *, unsigned long);
 extern void arch_unmap_area_topdown(struct mm_struct *, unsigned long);
 
-#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
+#if USE_SPLIT_PTLOCKS
 /*
  * The mm counters are not protected by its page_table_lock,
  * so must be incremented atomically.
@@ -363,7 +363,7 @@ extern void arch_unmap_area_topdown(struct mm_struct *, unsigned long);
 #define inc_mm_counter(mm, member) atomic_long_inc(&(mm)->_##member)
 #define dec_mm_counter(mm, member) atomic_long_dec(&(mm)->_##member)
 
-#else  /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */
+#else  /* !USE_SPLIT_PTLOCKS */
 /*
  * The mm counters are protected by its page_table_lock,
  * so can be incremented directly.
@@ -374,7 +374,7 @@ extern void arch_unmap_area_topdown(struct mm_struct *, unsigned long);
 #define inc_mm_counter(mm, member) (mm)->_##member++
 #define dec_mm_counter(mm, member) (mm)->_##member--
 
-#endif /* NR_CPUS < CONFIG_SPLIT_PTLOCK_CPUS */
+#endif /* !USE_SPLIT_PTLOCKS */
 
 #define get_mm_rss(mm)					\
 	(get_mm_counter(mm, file_rss) + get_mm_counter(mm, anon_rss))
diff --git a/include/linux/serial.h b/include/linux/serial.h
index deb714314fb1..1ea8d9265bf6 100644
--- a/include/linux/serial.h
+++ b/include/linux/serial.h
@@ -173,6 +173,22 @@ struct serial_icounter_struct {
 	int reserved[9];
 };
 
+/*
+ * Serial interface for controlling RS485 settings on chips with suitable
+ * support. Set with TIOCSRS485 and get with TIOCGRS485 if supported by your
+ * platform. The set function returns the new state, with any unsupported bits
+ * reverted appropriately.
+ */
+
+struct serial_rs485 {
+	__u32	flags;			/* RS485 feature flags */
+#define SER_RS485_ENABLED		(1 << 0)
+#define SER_RS485_RTS_ON_SEND		(1 << 1)
+#define SER_RS485_RTS_AFTER_SEND	(1 << 2)
+	__u32	delay_rts_before_send;	/* Milliseconds */
+	__u32	padding[6];		/* Memory is cheap, new structs
+					   are a royal PITA .. */
+};
 
 #ifdef __KERNEL__
 #include <linux/compiler.h>
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index 3b2f6c04855e..e27f216361fc 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -241,7 +241,7 @@ typedef unsigned int __bitwise__ upf_t;
 
 struct uart_port {
 	spinlock_t		lock;			/* port lock */
-	unsigned int		iobase;			/* in/out[bwl] */
+	unsigned long		iobase;			/* in/out[bwl] */
 	unsigned char __iomem	*membase;		/* read/write[bwl] */
 	unsigned int		irq;			/* irq number */
 	unsigned int		uartclk;		/* base uart clock */
diff --git a/include/linux/termios.h b/include/linux/termios.h
index 478662889f48..2acd0c1f8a2a 100644
--- a/include/linux/termios.h
+++ b/include/linux/termios.h
@@ -4,4 +4,19 @@
 #include <linux/types.h>
 #include <asm/termios.h>
 
+#define NFF	5
+
+struct termiox
+{
+	__u16	x_hflag;
+	__u16	x_cflag;
+	__u16	x_rflag[NFF];
+	__u16	x_sflag;
+};
+
+#define	RTSXOFF		0x0001		/* RTS flow control on input */
+#define	CTSXON		0x0002		/* CTS flow control on output */
+#define	DTRXOFF		0x0004		/* DTR flow control on input */
+#define DSRXON		0x0008		/* DCD flow control on output */
+
 #endif
diff --git a/include/linux/tty.h b/include/linux/tty.h
index 0cbec74ec086..3b8121d4e36f 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -23,7 +23,7 @@
  */
 #define NR_UNIX98_PTY_DEFAULT	4096      /* Default maximum for Unix98 ptys */
 #define NR_UNIX98_PTY_MAX	(1 << MINORBITS) /* Absolute limit */
-#define NR_LDISCS		18
+#define NR_LDISCS		19
 
 /* line disciplines */
 #define N_TTY		0
@@ -45,6 +45,7 @@
 #define N_HCI		15	/* Bluetooth HCI UART */
 #define N_GIGASET_M101	16	/* Siemens Gigaset M101 serial DECT adapter */
 #define N_SLCAN		17	/* Serial / USB serial CAN Adaptors */
+#define N_PPS		18	/* Pulse per Second */
 
 /*
  * This character is the same as _POSIX_VDISABLE: it cannot be used as
@@ -181,6 +182,7 @@ struct signal_struct;
 
 struct tty_port {
 	struct tty_struct	*tty;		/* Back pointer */
+	spinlock_t		lock;		/* Lock protecting tty field */
 	int			blocked_open;	/* Waiting to open */
 	int			count;		/* Usage count */
 	wait_queue_head_t	open_wait;	/* Open waiters */
@@ -208,6 +210,7 @@ struct tty_operations;
 
 struct tty_struct {
 	int	magic;
+	struct kref kref;
 	struct tty_driver *driver;
 	const struct tty_operations *ops;
 	int index;
@@ -217,6 +220,7 @@ struct tty_struct {
 	spinlock_t ctrl_lock;
 	/* Termios values are protected by the termios mutex */
 	struct ktermios *termios, *termios_locked;
+	struct termiox *termiox;	/* May be NULL for unsupported */
 	char name[64];
 	struct pid *pgrp;		/* Protected by ctrl lock */
 	struct pid *session;
@@ -310,6 +314,25 @@ extern int kmsg_redirect;
 extern void console_init(void);
 extern int vcs_init(void);
 
+extern struct class *tty_class;
+
+/**
+ *	tty_kref_get		-	get a tty reference
+ *	@tty: tty device
+ *
+ *	Return a new reference to a tty object. The caller must hold
+ *	sufficient locks/counts to ensure that their existing reference cannot
+ *	go away
+ */
+
+extern inline struct tty_struct *tty_kref_get(struct tty_struct *tty)
+{
+	if (tty)
+		kref_get(&tty->kref);
+	return tty;
+}
+extern void tty_kref_put(struct tty_struct *tty);
+
 extern int tty_paranoia_check(struct tty_struct *tty, struct inode *inode,
 			      const char *routine);
 extern char *tty_name(struct tty_struct *tty, char *buf);
@@ -333,13 +356,15 @@ extern void tty_throttle(struct tty_struct *tty);
 extern void tty_unthrottle(struct tty_struct *tty);
 extern int tty_do_resize(struct tty_struct *tty, struct tty_struct *real_tty,
 						struct winsize *ws);
-
+extern void tty_shutdown(struct tty_struct *tty);
+extern void tty_free_termios(struct tty_struct *tty);
 extern int is_current_pgrp_orphaned(void);
 extern struct pid *tty_get_pgrp(struct tty_struct *tty);
 extern int is_ignored(int sig);
 extern int tty_signal(int sig, struct tty_struct *tty);
 extern void tty_hangup(struct tty_struct *tty);
 extern void tty_vhangup(struct tty_struct *tty);
+extern void tty_vhangup_self(void);
 extern void tty_unhangup(struct file *filp);
 extern int tty_hung_up_p(struct file *filp);
 extern void do_SAK(struct tty_struct *tty);
@@ -347,6 +372,9 @@ extern void __do_SAK(struct tty_struct *tty);
 extern void disassociate_ctty(int priv);
 extern void no_tty(void);
 extern void tty_flip_buffer_push(struct tty_struct *tty);
+extern void tty_buffer_free_all(struct tty_struct *tty);
+extern void tty_buffer_flush(struct tty_struct *tty);
+extern void tty_buffer_init(struct tty_struct *tty);
 extern speed_t tty_get_baud_rate(struct tty_struct *tty);
 extern speed_t tty_termios_baud_rate(struct ktermios *termios);
 extern speed_t tty_termios_input_baud_rate(struct ktermios *termios);
@@ -372,6 +400,15 @@ extern int tty_perform_flush(struct tty_struct *tty, unsigned long arg);
 extern dev_t tty_devnum(struct tty_struct *tty);
 extern void proc_clear_tty(struct task_struct *p);
 extern struct tty_struct *get_current_tty(void);
+extern void tty_default_fops(struct file_operations *fops);
+extern struct tty_struct *alloc_tty_struct(void);
+extern void free_tty_struct(struct tty_struct *tty);
+extern void initialize_tty_struct(struct tty_struct *tty,
+		struct tty_driver *driver, int idx);
+extern struct tty_struct *tty_init_dev(struct tty_driver *driver, int idx,
+								int first_ok);
+extern void tty_release_dev(struct file *filp);
+extern int tty_init_termios(struct tty_struct *tty);
 
 extern struct mutex tty_mutex;
 
@@ -382,6 +419,8 @@ extern int tty_write_lock(struct tty_struct *tty, int ndelay);
 extern void tty_port_init(struct tty_port *port);
 extern int tty_port_alloc_xmit_buf(struct tty_port *port);
 extern void tty_port_free_xmit_buf(struct tty_port *port);
+extern struct tty_struct *tty_port_tty_get(struct tty_port *port);
+extern void tty_port_tty_set(struct tty_port *port, struct tty_struct *tty);
 
 extern int tty_register_ldisc(int disc, struct tty_ldisc_ops *new_ldisc);
 extern int tty_unregister_ldisc(int disc);
@@ -427,7 +466,7 @@ static inline void tty_audit_push_task(struct task_struct *tsk,
 #endif
 
 /* tty_ioctl.c */
-extern int n_tty_ioctl(struct tty_struct *tty, struct file *file,
+extern int n_tty_ioctl_helper(struct tty_struct *tty, struct file *file,
 		       unsigned int cmd, unsigned long arg);
 
 /* serial.c */
diff --git a/include/linux/tty_driver.h b/include/linux/tty_driver.h
index 16d27944c321..78416b901589 100644
--- a/include/linux/tty_driver.h
+++ b/include/linux/tty_driver.h
@@ -7,6 +7,28 @@
  * defined; unless noted otherwise, they are optional, and can be
  * filled in with a null pointer.
  *
+ * struct tty_struct * (*lookup)(struct tty_driver *self, int idx)
+ *
+ *	Return the tty device corresponding to idx, NULL if there is not
+ *	one currently in use and an ERR_PTR value on error. Called under
+ *	tty_mutex (for now!)
+ *
+ *	Optional method. Default behaviour is to use the ttys array
+ *
+ * int (*install)(struct tty_driver *self, struct tty_struct *tty)
+ *
+ *	Install a new tty into the tty driver internal tables. Used in
+ *	conjunction with lookup and remove methods.
+ *
+ *	Optional method. Default behaviour is to use the ttys array
+ *
+ * void (*remove)(struct tty_driver *self, struct tty_struct *tty)
+ *
+ *	Remove a closed tty from the tty driver internal tables. Used in
+ *	conjunction with lookup and remove methods.
+ *
+ *	Optional method. Default behaviour is to use the ttys array
+ *
  * int  (*open)(struct tty_struct * tty, struct file * filp);
  *
  * 	This routine is called when a particular tty device is opened.
@@ -21,6 +43,11 @@
  *
  *	Required method.
  *
+ * void (*shutdown)(struct tty_struct * tty);
+ *
+ * 	This routine is called when a particular tty device is closed for
+ *	the last time freeing up the resources.
+ *
  * int (*write)(struct tty_struct * tty,
  * 		 const unsigned char *buf, int count);
  *
@@ -180,6 +207,14 @@
  *	not force errors here if they are not resizable objects (eg a serial
  *	line). See tty_do_resize() if you need to wrap the standard method
  *	in your own logic - the usual case.
+ *
+ * void (*set_termiox)(struct tty_struct *tty, struct termiox *new);
+ *
+ *	Called when the device receives a termiox based ioctl. Passes down
+ *	the requested data from user space. This method will not be invoked
+ *	unless the tty also has a valid tty->termiox pointer.
+ *
+ *	Optional: Called under the termios lock
  */
 
 #include <linux/fs.h>
@@ -190,8 +225,13 @@ struct tty_struct;
 struct tty_driver;
 
 struct tty_operations {
+	struct tty_struct * (*lookup)(struct tty_driver *driver,
+			struct inode *inode, int idx);
+	int  (*install)(struct tty_driver *driver, struct tty_struct *tty);
+	void (*remove)(struct tty_driver *driver, struct tty_struct *tty);
 	int  (*open)(struct tty_struct * tty, struct file * filp);
 	void (*close)(struct tty_struct * tty, struct file * filp);
+	void (*shutdown)(struct tty_struct *tty);
 	int  (*write)(struct tty_struct * tty,
 		      const unsigned char *buf, int count);
 	int  (*put_char)(struct tty_struct *tty, unsigned char ch);
@@ -220,6 +260,7 @@ struct tty_operations {
 			unsigned int set, unsigned int clear);
 	int (*resize)(struct tty_struct *tty, struct tty_struct *real_tty,
 				struct winsize *ws);
+	int (*set_termiox)(struct tty_struct *tty, struct termiox *tnew);
 #ifdef CONFIG_CONSOLE_POLL
 	int (*poll_init)(struct tty_driver *driver, int line, char *options);
 	int (*poll_get_char)(struct tty_driver *driver, int line);
@@ -229,6 +270,7 @@ struct tty_operations {
 
 struct tty_driver {
 	int	magic;		/* magic number for this structure */
+	struct kref kref;	/* Reference management */
 	struct cdev cdev;
 	struct module	*owner;
 	const char	*driver_name;
@@ -242,7 +284,6 @@ struct tty_driver {
 	short	subtype;	/* subtype of tty driver */
 	struct ktermios init_termios; /* Initial termios */
 	int	flags;		/* tty driver flags */
-	int	refcount;	/* for loadable tty drivers */
 	struct proc_dir_entry *proc_entry; /* /proc fs entry */
 	struct tty_driver *other; /* only used for the PTY driver */
 
@@ -264,12 +305,19 @@ struct tty_driver {
 
 extern struct list_head tty_drivers;
 
-struct tty_driver *alloc_tty_driver(int lines);
-void put_tty_driver(struct tty_driver *driver);
-void tty_set_operations(struct tty_driver *driver,
+extern struct tty_driver *alloc_tty_driver(int lines);
+extern void put_tty_driver(struct tty_driver *driver);
+extern void tty_set_operations(struct tty_driver *driver,
 			const struct tty_operations *op);
 extern struct tty_driver *tty_find_polling_driver(char *name, int *line);
 
+extern void tty_driver_kref_put(struct tty_driver *driver);
+extern inline struct tty_driver *tty_driver_kref_get(struct tty_driver *d)
+{
+	kref_get(&d->kref);
+	return d;
+}
+
 /* tty driver magic number */
 #define TTY_DRIVER_MAGIC		0x5402
 
diff --git a/include/linux/usb/ehci_def.h b/include/linux/usb/ehci_def.h
new file mode 100644
index 000000000000..5b88e36c9103
--- /dev/null
+++ b/include/linux/usb/ehci_def.h
@@ -0,0 +1,160 @@
+/*
+ * Copyright (c) 2001-2002 by David Brownell
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2 of the License, or (at your
+ * option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ * for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software Foundation,
+ * Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#ifndef __LINUX_USB_EHCI_DEF_H
+#define __LINUX_USB_EHCI_DEF_H
+
+/* EHCI register interface, corresponds to EHCI Revision 0.95 specification */
+
+/* Section 2.2 Host Controller Capability Registers */
+struct ehci_caps {
+	/* these fields are specified as 8 and 16 bit registers,
+	 * but some hosts can't perform 8 or 16 bit PCI accesses.
+	 */
+	u32		hc_capbase;
+#define HC_LENGTH(p)		(((p)>>00)&0x00ff)	/* bits 7:0 */
+#define HC_VERSION(p)		(((p)>>16)&0xffff)	/* bits 31:16 */
+	u32		hcs_params;     /* HCSPARAMS - offset 0x4 */
+#define HCS_DEBUG_PORT(p)	(((p)>>20)&0xf)	/* bits 23:20, debug port? */
+#define HCS_INDICATOR(p)	((p)&(1 << 16))	/* true: has port indicators */
+#define HCS_N_CC(p)		(((p)>>12)&0xf)	/* bits 15:12, #companion HCs */
+#define HCS_N_PCC(p)		(((p)>>8)&0xf)	/* bits 11:8, ports per CC */
+#define HCS_PORTROUTED(p)	((p)&(1 << 7))	/* true: port routing */
+#define HCS_PPC(p)		((p)&(1 << 4))	/* true: port power control */
+#define HCS_N_PORTS(p)		(((p)>>0)&0xf)	/* bits 3:0, ports on HC */
+
+	u32		hcc_params;      /* HCCPARAMS - offset 0x8 */
+#define HCC_EXT_CAPS(p)		(((p)>>8)&0xff)	/* for pci extended caps */
+#define HCC_ISOC_CACHE(p)       ((p)&(1 << 7))  /* true: can cache isoc frame */
+#define HCC_ISOC_THRES(p)       (((p)>>4)&0x7)  /* bits 6:4, uframes cached */
+#define HCC_CANPARK(p)		((p)&(1 << 2))  /* true: can park on async qh */
+#define HCC_PGM_FRAMELISTLEN(p) ((p)&(1 << 1))  /* true: periodic_size changes*/
+#define HCC_64BIT_ADDR(p)       ((p)&(1))       /* true: can use 64-bit addr */
+	u8		portroute [8];	 /* nibbles for routing - offset 0xC */
+} __attribute__ ((packed));
+
+
+/* Section 2.3 Host Controller Operational Registers */
+struct ehci_regs {
+
+	/* USBCMD: offset 0x00 */
+	u32		command;
+/* 23:16 is r/w intr rate, in microframes; default "8" == 1/msec */
+#define CMD_PARK	(1<<11)		/* enable "park" on async qh */
+#define CMD_PARK_CNT(c)	(((c)>>8)&3)	/* how many transfers to park for */
+#define CMD_LRESET	(1<<7)		/* partial reset (no ports, etc) */
+#define CMD_IAAD	(1<<6)		/* "doorbell" interrupt async advance */
+#define CMD_ASE		(1<<5)		/* async schedule enable */
+#define CMD_PSE		(1<<4)		/* periodic schedule enable */
+/* 3:2 is periodic frame list size */
+#define CMD_RESET	(1<<1)		/* reset HC not bus */
+#define CMD_RUN		(1<<0)		/* start/stop HC */
+
+	/* USBSTS: offset 0x04 */
+	u32		status;
+#define STS_ASS		(1<<15)		/* Async Schedule Status */
+#define STS_PSS		(1<<14)		/* Periodic Schedule Status */
+#define STS_RECL	(1<<13)		/* Reclamation */
+#define STS_HALT	(1<<12)		/* Not running (any reason) */
+/* some bits reserved */
+	/* these STS_* flags are also intr_enable bits (USBINTR) */
+#define STS_IAA		(1<<5)		/* Interrupted on async advance */
+#define STS_FATAL	(1<<4)		/* such as some PCI access errors */
+#define STS_FLR		(1<<3)		/* frame list rolled over */
+#define STS_PCD		(1<<2)		/* port change detect */
+#define STS_ERR		(1<<1)		/* "error" completion (overflow, ...) */
+#define STS_INT		(1<<0)		/* "normal" completion (short, ...) */
+
+	/* USBINTR: offset 0x08 */
+	u32		intr_enable;
+
+	/* FRINDEX: offset 0x0C */
+	u32		frame_index;	/* current microframe number */
+	/* CTRLDSSEGMENT: offset 0x10 */
+	u32		segment;	/* address bits 63:32 if needed */
+	/* PERIODICLISTBASE: offset 0x14 */
+	u32		frame_list;	/* points to periodic list */
+	/* ASYNCLISTADDR: offset 0x18 */
+	u32		async_next;	/* address of next async queue head */
+
+	u32		reserved [9];
+
+	/* CONFIGFLAG: offset 0x40 */
+	u32		configured_flag;
+#define FLAG_CF		(1<<0)		/* true: we'll support "high speed" */
+
+	/* PORTSC: offset 0x44 */
+	u32		port_status [0];	/* up to N_PORTS */
+/* 31:23 reserved */
+#define PORT_WKOC_E	(1<<22)		/* wake on overcurrent (enable) */
+#define PORT_WKDISC_E	(1<<21)		/* wake on disconnect (enable) */
+#define PORT_WKCONN_E	(1<<20)		/* wake on connect (enable) */
+/* 19:16 for port testing */
+#define PORT_LED_OFF	(0<<14)
+#define PORT_LED_AMBER	(1<<14)
+#define PORT_LED_GREEN	(2<<14)
+#define PORT_LED_MASK	(3<<14)
+#define PORT_OWNER	(1<<13)		/* true: companion hc owns this port */
+#define PORT_POWER	(1<<12)		/* true: has power (see PPC) */
+#define PORT_USB11(x) (((x)&(3<<10)) == (1<<10))	/* USB 1.1 device */
+/* 11:10 for detecting lowspeed devices (reset vs release ownership) */
+/* 9 reserved */
+#define PORT_RESET	(1<<8)		/* reset port */
+#define PORT_SUSPEND	(1<<7)		/* suspend port */
+#define PORT_RESUME	(1<<6)		/* resume it */
+#define PORT_OCC	(1<<5)		/* over current change */
+#define PORT_OC		(1<<4)		/* over current active */
+#define PORT_PEC	(1<<3)		/* port enable change */
+#define PORT_PE		(1<<2)		/* port enable */
+#define PORT_CSC	(1<<1)		/* connect status change */
+#define PORT_CONNECT	(1<<0)		/* device connected */
+#define PORT_RWC_BITS   (PORT_CSC | PORT_PEC | PORT_OCC)
+} __attribute__ ((packed));
+
+#define USBMODE		0x68		/* USB Device mode */
+#define USBMODE_SDIS	(1<<3)		/* Stream disable */
+#define USBMODE_BE	(1<<2)		/* BE/LE endianness select */
+#define USBMODE_CM_HC	(3<<0)		/* host controller mode */
+#define USBMODE_CM_IDLE	(0<<0)		/* idle state */
+
+/* Appendix C, Debug port ... intended for use with special "debug devices"
+ * that can help if there's no serial console.  (nonstandard enumeration.)
+ */
+struct ehci_dbg_port {
+	u32	control;
+#define DBGP_OWNER	(1<<30)
+#define DBGP_ENABLED	(1<<28)
+#define DBGP_DONE	(1<<16)
+#define DBGP_INUSE	(1<<10)
+#define DBGP_ERRCODE(x)	(((x)>>7)&0x07)
+#	define DBGP_ERR_BAD	1
+#	define DBGP_ERR_SIGNAL	2
+#define DBGP_ERROR	(1<<6)
+#define DBGP_GO		(1<<5)
+#define DBGP_OUT	(1<<4)
+#define DBGP_LEN(x)	(((x)>>0)&0x0f)
+	u32	pids;
+#define DBGP_PID_GET(x)		(((x)>>16)&0xff)
+#define DBGP_PID_SET(data, tok)	(((data)<<8)|(tok))
+	u32	data03;
+	u32	data47;
+	u32	address;
+#define DBGP_EPADDR(dev, ep)	(((dev)<<8)|(ep))
+} __attribute__ ((packed));
+
+#endif /* __LINUX_USB_EHCI_DEF_H */
diff --git a/include/linux/vt_kern.h b/include/linux/vt_kern.h
index 1cbd0a7db4e6..2f1113467f70 100644
--- a/include/linux/vt_kern.h
+++ b/include/linux/vt_kern.h
@@ -96,7 +96,7 @@ void change_console(struct vc_data *new_vc);
 void reset_vc(struct vc_data *vc);
 extern int unbind_con_driver(const struct consw *csw, int first, int last,
 			     int deflt);
-int vty_init(void);
+int vty_init(const struct file_operations *console_fops);
 
 /*
  * vc_screen.c shares this temporary buffer with the console write code so that
diff --git a/include/net/cipso_ipv4.h b/include/net/cipso_ipv4.h
index a6bb94530cfd..9909774eb998 100644
--- a/include/net/cipso_ipv4.h
+++ b/include/net/cipso_ipv4.h
@@ -40,11 +40,12 @@
 #include <linux/net.h>
 #include <linux/skbuff.h>
 #include <net/netlabel.h>
+#include <asm/atomic.h>
 
 /* known doi values */
 #define CIPSO_V4_DOI_UNKNOWN          0x00000000
 
-/* tag types */
+/* standard tag types */
 #define CIPSO_V4_TAG_INVALID          0
 #define CIPSO_V4_TAG_RBITMAP          1
 #define CIPSO_V4_TAG_ENUM             2
@@ -52,10 +53,14 @@
 #define CIPSO_V4_TAG_PBITMAP          6
 #define CIPSO_V4_TAG_FREEFORM         7
 
+/* non-standard tag types (tags > 127) */
+#define CIPSO_V4_TAG_LOCAL            128
+
 /* doi mapping types */
 #define CIPSO_V4_MAP_UNKNOWN          0
-#define CIPSO_V4_MAP_STD              1
+#define CIPSO_V4_MAP_TRANS            1
 #define CIPSO_V4_MAP_PASS             2
+#define CIPSO_V4_MAP_LOCAL            3
 
 /* limits */
 #define CIPSO_V4_MAX_REM_LVLS         255
@@ -79,10 +84,9 @@ struct cipso_v4_doi {
 	} map;
 	u8 tags[CIPSO_V4_TAG_MAXCNT];
 
-	u32 valid;
+	atomic_t refcount;
 	struct list_head list;
 	struct rcu_head rcu;
-	struct list_head dom_list;
 };
 
 /* Standard CIPSO mapping table */
@@ -128,25 +132,26 @@ extern int cipso_v4_rbm_strictvalid;
 
 #ifdef CONFIG_NETLABEL
 int cipso_v4_doi_add(struct cipso_v4_doi *doi_def);
-int cipso_v4_doi_remove(u32 doi,
-			struct netlbl_audit *audit_info,
-			void (*callback) (struct rcu_head * head));
+void cipso_v4_doi_free(struct cipso_v4_doi *doi_def);
+int cipso_v4_doi_remove(u32 doi, struct netlbl_audit *audit_info);
 struct cipso_v4_doi *cipso_v4_doi_getdef(u32 doi);
+void cipso_v4_doi_putdef(struct cipso_v4_doi *doi_def);
 int cipso_v4_doi_walk(u32 *skip_cnt,
 		     int (*callback) (struct cipso_v4_doi *doi_def, void *arg),
 	             void *cb_arg);
-int cipso_v4_doi_domhsh_add(struct cipso_v4_doi *doi_def, const char *domain);
-int cipso_v4_doi_domhsh_remove(struct cipso_v4_doi *doi_def,
-			       const char *domain);
 #else
 static inline int cipso_v4_doi_add(struct cipso_v4_doi *doi_def)
 {
 	return -ENOSYS;
 }
 
+static inline void cipso_v4_doi_free(struct cipso_v4_doi *doi_def)
+{
+	return;
+}
+
 static inline int cipso_v4_doi_remove(u32 doi,
-				    struct netlbl_audit *audit_info,
-				    void (*callback) (struct rcu_head * head))
+				      struct netlbl_audit *audit_info)
 {
 	return 0;
 }
@@ -206,10 +211,15 @@ void cipso_v4_error(struct sk_buff *skb, int error, u32 gateway);
 int cipso_v4_sock_setattr(struct sock *sk,
 			  const struct cipso_v4_doi *doi_def,
 			  const struct netlbl_lsm_secattr *secattr);
+void cipso_v4_sock_delattr(struct sock *sk);
 int cipso_v4_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr);
+int cipso_v4_skbuff_setattr(struct sk_buff *skb,
+			    const struct cipso_v4_doi *doi_def,
+			    const struct netlbl_lsm_secattr *secattr);
+int cipso_v4_skbuff_delattr(struct sk_buff *skb);
 int cipso_v4_skbuff_getattr(const struct sk_buff *skb,
 			    struct netlbl_lsm_secattr *secattr);
-int cipso_v4_validate(unsigned char **option);
+int cipso_v4_validate(const struct sk_buff *skb, unsigned char **option);
 #else
 static inline void cipso_v4_error(struct sk_buff *skb,
 				  int error,
@@ -225,19 +235,36 @@ static inline int cipso_v4_sock_setattr(struct sock *sk,
 	return -ENOSYS;
 }
 
+static inline void cipso_v4_sock_delattr(struct sock *sk)
+{
+}
+
 static inline int cipso_v4_sock_getattr(struct sock *sk,
 					struct netlbl_lsm_secattr *secattr)
 {
 	return -ENOSYS;
 }
 
+static inline int cipso_v4_skbuff_setattr(struct sk_buff *skb,
+				      const struct cipso_v4_doi *doi_def,
+				      const struct netlbl_lsm_secattr *secattr)
+{
+	return -ENOSYS;
+}
+
+static inline int cipso_v4_skbuff_delattr(struct sk_buff *skb)
+{
+	return -ENOSYS;
+}
+
 static inline int cipso_v4_skbuff_getattr(const struct sk_buff *skb,
 					  struct netlbl_lsm_secattr *secattr)
 {
 	return -ENOSYS;
 }
 
-static inline int cipso_v4_validate(unsigned char **option)
+static inline int cipso_v4_validate(const struct sk_buff *skb,
+				    unsigned char **option)
 {
 	return -ENOSYS;
 }
diff --git a/include/net/netlabel.h b/include/net/netlabel.h
index e4d2d6baa983..17c442a4514e 100644
--- a/include/net/netlabel.h
+++ b/include/net/netlabel.h
@@ -9,7 +9,7 @@
  */
 
 /*
- * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006, 2008
  *
  * This program is free software;  you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -72,8 +72,10 @@ struct cipso_v4_doi;
 /* NetLabel NETLINK protocol version
  *  1: initial version
  *  2: added static labels for unlabeled connections
+ *  3: network selectors added to the NetLabel/LSM domain mapping and the
+ *     CIPSO_V4_MAP_LOCAL CIPSO mapping was added
  */
-#define NETLBL_PROTO_VERSION            2
+#define NETLBL_PROTO_VERSION            3
 
 /* NetLabel NETLINK types/families */
 #define NETLBL_NLTYPE_NONE              0
@@ -87,6 +89,8 @@ struct cipso_v4_doi;
 #define NETLBL_NLTYPE_CIPSOV6_NAME      "NLBL_CIPSOv6"
 #define NETLBL_NLTYPE_UNLABELED         5
 #define NETLBL_NLTYPE_UNLABELED_NAME    "NLBL_UNLBL"
+#define NETLBL_NLTYPE_ADDRSELECT        6
+#define NETLBL_NLTYPE_ADDRSELECT_NAME   "NLBL_ADRSEL"
 
 /*
  * NetLabel - Kernel API for accessing the network packet label mappings.
@@ -200,7 +204,7 @@ struct netlbl_lsm_secattr {
 	u32 type;
 	char *domain;
 	struct netlbl_lsm_cache *cache;
-	union {
+	struct {
 		struct {
 			struct netlbl_lsm_secattr_catmap *cat;
 			u32 lvl;
@@ -352,12 +356,9 @@ static inline void netlbl_secattr_free(struct netlbl_lsm_secattr *secattr)
 int netlbl_cfg_map_del(const char *domain, struct netlbl_audit *audit_info);
 int netlbl_cfg_unlbl_add_map(const char *domain,
 			     struct netlbl_audit *audit_info);
-int netlbl_cfg_cipsov4_add(struct cipso_v4_doi *doi_def,
-			   struct netlbl_audit *audit_info);
 int netlbl_cfg_cipsov4_add_map(struct cipso_v4_doi *doi_def,
 			       const char *domain,
 			       struct netlbl_audit *audit_info);
-int netlbl_cfg_cipsov4_del(u32 doi, struct netlbl_audit *audit_info);
 
 /*
  * LSM security attribute operations
@@ -380,12 +381,19 @@ int netlbl_secattr_catmap_setrng(struct netlbl_lsm_secattr_catmap *catmap,
 int netlbl_enabled(void);
 int netlbl_sock_setattr(struct sock *sk,
 			const struct netlbl_lsm_secattr *secattr);
+void netlbl_sock_delattr(struct sock *sk);
 int netlbl_sock_getattr(struct sock *sk,
 			struct netlbl_lsm_secattr *secattr);
+int netlbl_conn_setattr(struct sock *sk,
+			struct sockaddr *addr,
+			const struct netlbl_lsm_secattr *secattr);
+int netlbl_skbuff_setattr(struct sk_buff *skb,
+			  u16 family,
+			  const struct netlbl_lsm_secattr *secattr);
 int netlbl_skbuff_getattr(const struct sk_buff *skb,
 			  u16 family,
 			  struct netlbl_lsm_secattr *secattr);
-void netlbl_skbuff_err(struct sk_buff *skb, int error);
+void netlbl_skbuff_err(struct sk_buff *skb, int error, int gateway);
 
 /*
  * LSM label mapping cache operations
@@ -404,22 +412,12 @@ static inline int netlbl_cfg_unlbl_add_map(const char *domain,
 {
 	return -ENOSYS;
 }
-static inline int netlbl_cfg_cipsov4_add(struct cipso_v4_doi *doi_def,
-					 struct netlbl_audit *audit_info)
-{
-	return -ENOSYS;
-}
 static inline int netlbl_cfg_cipsov4_add_map(struct cipso_v4_doi *doi_def,
 					     const char *domain,
 					     struct netlbl_audit *audit_info)
 {
 	return -ENOSYS;
 }
-static inline int netlbl_cfg_cipsov4_del(u32 doi,
-					 struct netlbl_audit *audit_info)
-{
-	return -ENOSYS;
-}
 static inline int netlbl_secattr_catmap_walk(
 	                              struct netlbl_lsm_secattr_catmap *catmap,
 				      u32 offset)
@@ -456,18 +454,35 @@ static inline int netlbl_sock_setattr(struct sock *sk,
 {
 	return -ENOSYS;
 }
+static inline void netlbl_sock_delattr(struct sock *sk)
+{
+}
 static inline int netlbl_sock_getattr(struct sock *sk,
 				      struct netlbl_lsm_secattr *secattr)
 {
 	return -ENOSYS;
 }
+static inline int netlbl_conn_setattr(struct sock *sk,
+				      struct sockaddr *addr,
+				      const struct netlbl_lsm_secattr *secattr)
+{
+	return -ENOSYS;
+}
+static inline int netlbl_skbuff_setattr(struct sk_buff *skb,
+				      u16 family,
+				      const struct netlbl_lsm_secattr *secattr)
+{
+	return -ENOSYS;
+}
 static inline int netlbl_skbuff_getattr(const struct sk_buff *skb,
 					u16 family,
 					struct netlbl_lsm_secattr *secattr)
 {
 	return -ENOSYS;
 }
-static inline void netlbl_skbuff_err(struct sk_buff *skb, int error)
+static inline void netlbl_skbuff_err(struct sk_buff *skb,
+				     int error,
+				     int gateway)
 {
 	return;
 }
diff --git a/include/sound/core.h b/include/sound/core.h
index e5eec5f73502..35424a971b7a 100644
--- a/include/sound/core.h
+++ b/include/sound/core.h
@@ -43,9 +43,6 @@
 #ifdef CONFIG_PCI
 struct pci_dev;
 #endif
-#ifdef CONFIG_SBUS
-struct sbus_dev;
-#endif
 
 /* device allocation stuff */
 
diff --git a/include/sound/memalloc.h b/include/sound/memalloc.h
index d787a6b4a101..7ccce94a5255 100644
--- a/include/sound/memalloc.h
+++ b/include/sound/memalloc.h
@@ -37,7 +37,6 @@ struct snd_dma_device {
 #ifndef snd_dma_pci_data
 #define snd_dma_pci_data(pci)	(&(pci)->dev)
 #define snd_dma_isa_data()	NULL
-#define snd_dma_sbus_data(sbus)	((struct device *)(sbus))
 #define snd_dma_continuous_data(x)	((struct device *)(unsigned long)(x))
 #endif
 
@@ -49,7 +48,6 @@ struct snd_dma_device {
 #define SNDRV_DMA_TYPE_CONTINUOUS	1	/* continuous no-DMA memory */
 #define SNDRV_DMA_TYPE_DEV		2	/* generic device continuous */
 #define SNDRV_DMA_TYPE_DEV_SG		3	/* generic device SG-buffer */
-#define SNDRV_DMA_TYPE_SBUS		4	/* SBUS continuous */
 
 /*
  * info for buffer allocation
diff --git a/include/sound/soc-dapm.h b/include/sound/soc-dapm.h
index c1b26fcc0b5c..ca699a3017f3 100644
--- a/include/sound/soc-dapm.h
+++ b/include/sound/soc-dapm.h
@@ -240,6 +240,7 @@ int snd_soc_dapm_sys_add(struct device *dev);
 /* dapm audio pin control and status */
 int snd_soc_dapm_enable_pin(struct snd_soc_codec *codec, char *pin);
 int snd_soc_dapm_disable_pin(struct snd_soc_codec *codec, char *pin);
+int snd_soc_dapm_nc_pin(struct snd_soc_codec *codec, char *pin);
 int snd_soc_dapm_get_pin_status(struct snd_soc_codec *codec, char *pin);
 int snd_soc_dapm_sync(struct snd_soc_codec *codec);
 
diff --git a/include/xen/balloon.h b/include/xen/balloon.h
deleted file mode 100644
index fe43b0f3c86a..000000000000
--- a/include/xen/balloon.h
+++ /dev/null
@@ -1,61 +0,0 @@
-/******************************************************************************
- * balloon.h
- *
- * Xen balloon driver - enables returning/claiming memory to/from Xen.
- *
- * Copyright (c) 2003, B Dragovic
- * Copyright (c) 2003-2004, M Williamson, K Fraser
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License version 2
- * as published by the Free Software Foundation; or, when distributed
- * separately from the Linux kernel or incorporated into other
- * software packages, subject to the following license:
- *
- * Permission is hereby granted, free of charge, to any person obtaining a copy
- * of this source file (the "Software"), to deal in the Software without
- * restriction, including without limitation the rights to use, copy, modify,
- * merge, publish, distribute, sublicense, and/or sell copies of the Software,
- * and to permit persons to whom the Software is furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
- * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
- * IN THE SOFTWARE.
- */
-
-#ifndef __XEN_BALLOON_H__
-#define __XEN_BALLOON_H__
-
-#include <linux/spinlock.h>
-
-#if 0
-/*
- * Inform the balloon driver that it should allow some slop for device-driver
- * memory activities.
- */
-void balloon_update_driver_allowance(long delta);
-
-/* Allocate/free a set of empty pages in low memory (i.e., no RAM mapped). */
-struct page **alloc_empty_pages_and_pagevec(int nr_pages);
-void free_empty_pages_and_pagevec(struct page **pagevec, int nr_pages);
-
-void balloon_release_driver_page(struct page *page);
-
-/*
- * Prevent the balloon driver from changing the memory reservation during
- * a driver critical region.
- */
-extern spinlock_t balloon_lock;
-#define balloon_lock(__flags)   spin_lock_irqsave(&balloon_lock, __flags)
-#define balloon_unlock(__flags) spin_unlock_irqrestore(&balloon_lock, __flags)
-#endif
-
-#endif /* __XEN_BALLOON_H__ */
diff --git a/include/xen/events.h b/include/xen/events.h
index 4680ff3fbc91..0d5f1adc0363 100644
--- a/include/xen/events.h
+++ b/include/xen/events.h
@@ -46,6 +46,8 @@ extern void xen_irq_resume(void);
 
 /* Clear an irq's pending state, in preparation for polling on it */
 void xen_clear_irq_pending(int irq);
+void xen_set_irq_pending(int irq);
+bool xen_test_irq_pending(int irq);
 
 /* Poll waiting for an irq to become pending.  In the usual case, the
    irq will be disabled so it won't deliver an interrupt. */
diff --git a/init/Kconfig b/init/Kconfig
index c11da38837e5..8a8e2d00c40e 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -779,16 +779,6 @@ config MARKERS
 
 source "arch/Kconfig"
 
-config PROC_PAGE_MONITOR
- 	default y
-	depends on PROC_FS && MMU
-	bool "Enable /proc page monitoring" if EMBEDDED
- 	help
-	  Various /proc files exist to monitor process memory utilization:
-	  /proc/pid/smaps, /proc/pid/clear_refs, /proc/pid/pagemap,
-	  /proc/kpagecount, and /proc/kpageflags. Disabling these
-          interfaces will reduce the size of the kernel by approximately 4kb.
-
 endmenu		# General setup
 
 config HAVE_GENERIC_DMA_COHERENT
diff --git a/kernel/acct.c b/kernel/acct.c
index dd68b9059418..f6006a60df5d 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -548,7 +548,7 @@ static void do_acct_process(struct bsd_acct_struct *acct,
 #endif
 
 	spin_lock_irq(&current->sighand->siglock);
-	tty = current->signal->tty;
+	tty = current->signal->tty;	/* Safe as we hold the siglock */
 	ac.ac_tty = tty ? old_encode_dev(tty_devnum(tty)) : 0;
 	ac.ac_utime = encode_comp_t(jiffies_to_AHZ(cputime_to_jiffies(pacct->ac_utime)));
 	ac.ac_stime = encode_comp_t(jiffies_to_AHZ(cputime_to_jiffies(pacct->ac_stime)));
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 59cedfb040e7..cf5bc2f5f9c3 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -246,8 +246,8 @@ static int audit_match_perm(struct audit_context *ctx, int mask)
 	unsigned n;
 	if (unlikely(!ctx))
 		return 0;
-
 	n = ctx->major;
+
 	switch (audit_classify_syscall(ctx->arch, n)) {
 	case 0:	/* native */
 		if ((mask & AUDIT_PERM_WRITE) &&
@@ -1204,13 +1204,13 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 				 (context->return_valid==AUDITSC_SUCCESS)?"yes":"no",
 				 context->return_code);
 
-	mutex_lock(&tty_mutex);
-	read_lock(&tasklist_lock);
+	spin_lock_irq(&tsk->sighand->siglock);
 	if (tsk->signal && tsk->signal->tty && tsk->signal->tty->name)
 		tty = tsk->signal->tty->name;
 	else
 		tty = "(none)";
-	read_unlock(&tasklist_lock);
+	spin_unlock_irq(&tsk->sighand->siglock);
+
 	audit_log_format(ab,
 		  " a0=%lx a1=%lx a2=%lx a3=%lx items=%d"
 		  " ppid=%d pid=%d auid=%u uid=%u gid=%u"
@@ -1230,7 +1230,6 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 		  context->egid, context->sgid, context->fsgid, tty,
 		  tsk->sessionid);
 
-	mutex_unlock(&tty_mutex);
 
 	audit_log_task_info(ab, tsk);
 	if (context->filterkey) {
diff --git a/kernel/fork.c b/kernel/fork.c
index 7ce2ebe84796..30de644a40c4 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -802,6 +802,7 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
 
 	sig->leader = 0;	/* session leadership doesn't inherit */
 	sig->tty_old_pgrp = NULL;
+	sig->tty = NULL;
 
 	sig->utime = sig->stime = sig->cutime = sig->cstime = cputime_zero;
 	sig->gtime = cputime_zero;
@@ -838,6 +839,7 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
 void __cleanup_signal(struct signal_struct *sig)
 {
 	exit_thread_group_keys(sig);
+	tty_kref_put(sig->tty);
 	kmem_cache_free(signal_cachep, sig);
 }
 
@@ -1227,7 +1229,8 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 				p->nsproxy->pid_ns->child_reaper = p;
 
 			p->signal->leader_pid = pid;
-			p->signal->tty = current->signal->tty;
+			tty_kref_put(p->signal->tty);
+			p->signal->tty = tty_kref_get(current->signal->tty);
 			set_task_pgrp(p, task_pgrp_nr(current));
 			set_task_session(p, task_session_nr(current));
 			attach_pid(p, PIDTYPE_PGID, task_pgrp(current));
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 0314074fa232..60c49e324390 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -89,7 +89,14 @@ int irq_set_affinity(unsigned int irq, cpumask_t cpumask)
 	set_balance_irq_affinity(irq, cpumask);
 
 #ifdef CONFIG_GENERIC_PENDING_IRQ
-	set_pending_irq(irq, cpumask);
+	if (desc->status & IRQ_MOVE_PCNTXT) {
+		unsigned long flags;
+
+		spin_lock_irqsave(&desc->lock, flags);
+		desc->chip->set_affinity(irq, cpumask);
+		spin_unlock_irqrestore(&desc->lock, flags);
+	} else
+		set_pending_irq(irq, cpumask);
 #else
 	desc->affinity = cpumask;
 	desc->chip->set_affinity(irq, cpumask);
diff --git a/kernel/printk.c b/kernel/printk.c
index b51b1567bb55..a430fd04008b 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -1291,22 +1291,6 @@ static int __init disable_boot_consoles(void)
 }
 late_initcall(disable_boot_consoles);
 
-/**
- * tty_write_message - write a message to a certain tty, not just the console.
- * @tty: the destination tty_struct
- * @msg: the message to write
- *
- * This is used for messages that need to be redirected to a specific tty.
- * We don't put it into the syslog queue right now maybe in the future if
- * really needed.
- */
-void tty_write_message(struct tty_struct *tty, char *msg)
-{
-	if (tty && tty->ops->write)
-		tty->ops->write(tty, msg, strlen(msg));
-	return;
-}
-
 #if defined CONFIG_PRINTK
 
 /*
diff --git a/kernel/resource.c b/kernel/resource.c
index 03d796c1b2e9..414d6fc9131e 100644
--- a/kernel/resource.c
+++ b/kernel/resource.c
@@ -516,6 +516,74 @@ int adjust_resource(struct resource *res, resource_size_t start, resource_size_t
 	return result;
 }
 
+static void __init __reserve_region_with_split(struct resource *root,
+		resource_size_t start, resource_size_t end,
+		const char *name)
+{
+	struct resource *parent = root;
+	struct resource *conflict;
+	struct resource *res = kzalloc(sizeof(*res), GFP_KERNEL);
+
+	if (!res)
+		return;
+
+	res->name = name;
+	res->start = start;
+	res->end = end;
+	res->flags = IORESOURCE_BUSY;
+
+	for (;;) {
+		conflict = __request_resource(parent, res);
+		if (!conflict)
+			break;
+		if (conflict != parent) {
+			parent = conflict;
+			if (!(conflict->flags & IORESOURCE_BUSY))
+				continue;
+		}
+
+		/* Uhhuh, that didn't work out.. */
+		kfree(res);
+		res = NULL;
+		break;
+	}
+
+	if (!res) {
+		printk(KERN_DEBUG "    __reserve_region_with_split: (%s) [%llx, %llx], res: (%s) [%llx, %llx]\n",
+			 conflict->name, conflict->start, conflict->end,
+			 name, start, end);
+
+		/* failed, split and try again */
+
+		/* conflict coverred whole area */
+		if (conflict->start <= start && conflict->end >= end)
+			return;
+
+		if (conflict->start > start)
+			__reserve_region_with_split(root, start, conflict->start-1, name);
+		if (!(conflict->flags & IORESOURCE_BUSY)) {
+			resource_size_t common_start, common_end;
+
+			common_start = max(conflict->start, start);
+			common_end = min(conflict->end, end);
+			if (common_start < common_end)
+				__reserve_region_with_split(root, common_start, common_end, name);
+		}
+		if (conflict->end < end)
+			__reserve_region_with_split(root, conflict->end+1, end, name);
+	}
+
+}
+
+void reserve_region_with_split(struct resource *root,
+		resource_size_t start, resource_size_t end,
+		const char *name)
+{
+	write_lock(&resource_lock);
+	__reserve_region_with_split(root, start, end, name);
+	write_unlock(&resource_lock);
+}
+
 EXPORT_SYMBOL(adjust_resource);
 
 /**
diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c
index bbe6b31c3c56..ad958c1ec708 100644
--- a/kernel/sched_debug.c
+++ b/kernel/sched_debug.c
@@ -333,12 +333,10 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m)
 	unsigned long flags;
 	int num_threads = 1;
 
-	rcu_read_lock();
 	if (lock_task_sighand(p, &flags)) {
 		num_threads = atomic_read(&p->signal->count);
 		unlock_task_sighand(p, &flags);
 	}
-	rcu_read_unlock();
 
 	SEQ_printf(m, "%s (%d, #threads: %d)\n", p->comm, p->pid, num_threads);
 	SEQ_printf(m,
diff --git a/kernel/sys.c b/kernel/sys.c
index 038a7bc0901d..234d9454294e 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -1060,9 +1060,7 @@ asmlinkage long sys_setsid(void)
 	group_leader->signal->leader = 1;
 	__set_special_pids(sid);
 
-	spin_lock(&group_leader->sighand->siglock);
-	group_leader->signal->tty = NULL;
-	spin_unlock(&group_leader->sighand->siglock);
+	proc_clear_tty(group_leader);
 
 	err = session;
 out:
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 50ec0886fa3d..c468c3c6dfc5 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -80,7 +80,6 @@ extern int pid_max_min, pid_max_max;
 extern int sysctl_drop_caches;
 extern int percpu_pagelist_fraction;
 extern int compat_log;
-extern int maps_protect;
 extern int latencytop_enabled;
 extern int sysctl_nr_open_min, sysctl_nr_open_max;
 #ifdef CONFIG_RCU_TORTURE_TEST
@@ -118,10 +117,8 @@ extern char modprobe_path[];
 extern int sg_big_buff;
 #endif
 
-#ifdef __sparc__
-extern char reboot_command [];
-extern int stop_a_enabled;
-extern int scons_pwroff;
+#ifdef CONFIG_SPARC
+#include <asm/system.h>
 #endif
 
 #ifdef __hppa__
@@ -415,7 +412,7 @@ static struct ctl_table kern_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
 	},
-#ifdef __sparc__
+#ifdef CONFIG_SPARC
 	{
 		.ctl_name	= KERN_SPARC_REBOOT,
 		.procname	= "reboot-cmd",
@@ -810,16 +807,6 @@ static struct ctl_table kern_table[] = {
 		.proc_handler	= &proc_dointvec,
 	},
 #endif
-#ifdef CONFIG_PROC_FS
-	{
-		.ctl_name       = CTL_UNNUMBERED,
-		.procname       = "maps_protect",
-		.data           = &maps_protect,
-		.maxlen         = sizeof(int),
-		.mode           = 0644,
-		.proc_handler   = &proc_dointvec,
-	},
-#endif
 	{
 		.ctl_name	= CTL_UNNUMBERED,
 		.procname	= "poweroff_cmd",
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index ce697e0b319e..aa81d2848448 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -495,6 +495,15 @@ config DEBUG_VM
 
 	  If unsure, say N.
 
+config DEBUG_VIRTUAL
+	bool "Debug VM translations"
+	depends on DEBUG_KERNEL && X86
+	help
+	  Enable some costly sanity checks in virtual to page code. This can
+	  catch mistakes with virt_to_page() and friends.
+
+	  If unsure, say N.
+
 config DEBUG_WRITECOUNT
 	bool "Debug filesystem writers count"
 	depends on DEBUG_KERNEL
diff --git a/lib/cmdline.c b/lib/cmdline.c
index 5ba8a942a478..f5f3ad8b62ff 100644
--- a/lib/cmdline.c
+++ b/lib/cmdline.c
@@ -126,7 +126,7 @@ char *get_options(const char *str, int nints, int *ints)
  *	megabyte, or one gigabyte, respectively.
  */
 
-unsigned long long memparse(char *ptr, char **retptr)
+unsigned long long memparse(const char *ptr, char **retptr)
 {
 	char *endptr;	/* local pointer to end of parsed string */
 
diff --git a/lib/percpu_counter.c b/lib/percpu_counter.c
index 4a8ba4bf5f6f..a8663890a88c 100644
--- a/lib/percpu_counter.c
+++ b/lib/percpu_counter.c
@@ -52,7 +52,7 @@ EXPORT_SYMBOL(__percpu_counter_add);
  * Add up all the per-cpu counts, return the result.  This is a more accurate
  * but much slower version of percpu_counter_read_positive()
  */
-s64 __percpu_counter_sum(struct percpu_counter *fbc, int set)
+s64 __percpu_counter_sum(struct percpu_counter *fbc)
 {
 	s64 ret;
 	int cpu;
@@ -62,11 +62,9 @@ s64 __percpu_counter_sum(struct percpu_counter *fbc, int set)
 	for_each_online_cpu(cpu) {
 		s32 *pcount = per_cpu_ptr(fbc->counters, cpu);
 		ret += *pcount;
-		if (set)
-			*pcount = 0;
+		*pcount = 0;
 	}
-	if (set)
-		fbc->count = ret;
+	fbc->count = ret;
 
 	spin_unlock(&fbc->lock);
 	return ret;
diff --git a/mm/shmem.c b/mm/shmem.c
index 04fb4f1ab88e..bf66d0191baf 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -50,14 +50,12 @@
 #include <linux/migrate.h>
 #include <linux/highmem.h>
 #include <linux/seq_file.h>
+#include <linux/magic.h>
 
 #include <asm/uaccess.h>
 #include <asm/div64.h>
 #include <asm/pgtable.h>
 
-/* This magic number is used in glibc for posix shared memory */
-#define TMPFS_MAGIC	0x01021994
-
 #define ENTRIES_PER_PAGE (PAGE_CACHE_SIZE/sizeof(unsigned long))
 #define ENTRIES_PER_PAGEPAGE (ENTRIES_PER_PAGE*ENTRIES_PER_PAGE)
 #define BLOCKS_PER_PAGE  (PAGE_CACHE_SIZE/512)
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 85b9a0d2c877..bba06c41fc59 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -180,6 +180,13 @@ struct page *vmalloc_to_page(const void *vmalloc_addr)
 	pmd_t *pmd;
 	pte_t *ptep, pte;
 
+	/*
+	 * XXX we might need to change this if we add VIRTUAL_BUG_ON for
+	 * architectures that do not vmalloc module space
+	 */
+	VIRTUAL_BUG_ON(!is_vmalloc_addr(vmalloc_addr) &&
+			!is_module_address(addr));
+
 	if (!pgd_none(*pgd)) {
 		pud = pud_offset(pgd, addr);
 		if (!pud_none(*pud)) {
diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
index 2c0e4572cc90..490e035c6d90 100644
--- a/net/ipv4/cipso_ipv4.c
+++ b/net/ipv4/cipso_ipv4.c
@@ -13,7 +13,7 @@
  */
 
 /*
- * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006, 2008
  *
  * This program is free software;  you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -47,17 +47,7 @@
 #include <asm/bug.h>
 #include <asm/unaligned.h>
 
-struct cipso_v4_domhsh_entry {
-	char *domain;
-	u32 valid;
-	struct list_head list;
-	struct rcu_head rcu;
-};
-
 /* List of available DOI definitions */
-/* XXX - Updates should be minimal so having a single lock for the
- * cipso_v4_doi_list and the cipso_v4_doi_list->dom_list should be
- * okay. */
 /* XXX - This currently assumes a minimal number of different DOIs in use,
  * if in practice there are a lot of different DOIs this list should
  * probably be turned into a hash table or something similar so we
@@ -119,6 +109,19 @@ int cipso_v4_rbm_strictvalid = 1;
  * be omitted. */
 #define CIPSO_V4_TAG_RNG_CAT_MAX      8
 
+/* Base length of the local tag (non-standard tag).
+ *  Tag definition (may change between kernel versions)
+ *
+ * 0          8          16         24         32
+ * +----------+----------+----------+----------+
+ * | 10000000 | 00000110 | 32-bit secid value  |
+ * +----------+----------+----------+----------+
+ * | in (host byte order)|
+ * +----------+----------+
+ *
+ */
+#define CIPSO_V4_TAG_LOC_BLEN         6
+
 /*
  * Helper Functions
  */
@@ -194,25 +197,6 @@ static void cipso_v4_bitmap_setbit(unsigned char *bitmap,
 }
 
 /**
- * cipso_v4_doi_domhsh_free - Frees a domain list entry
- * @entry: the entry's RCU field
- *
- * Description:
- * This function is designed to be used as a callback to the call_rcu()
- * function so that the memory allocated to a domain list entry can be released
- * safely.
- *
- */
-static void cipso_v4_doi_domhsh_free(struct rcu_head *entry)
-{
-	struct cipso_v4_domhsh_entry *ptr;
-
-	ptr = container_of(entry, struct cipso_v4_domhsh_entry, rcu);
-	kfree(ptr->domain);
-	kfree(ptr);
-}
-
-/**
  * cipso_v4_cache_entry_free - Frees a cache entry
  * @entry: the entry to free
  *
@@ -457,7 +441,7 @@ static struct cipso_v4_doi *cipso_v4_doi_search(u32 doi)
 	struct cipso_v4_doi *iter;
 
 	list_for_each_entry_rcu(iter, &cipso_v4_doi_list, list)
-		if (iter->doi == doi && iter->valid)
+		if (iter->doi == doi && atomic_read(&iter->refcount))
 			return iter;
 	return NULL;
 }
@@ -496,14 +480,17 @@ int cipso_v4_doi_add(struct cipso_v4_doi *doi_def)
 			if (doi_def->type != CIPSO_V4_MAP_PASS)
 				return -EINVAL;
 			break;
+		case CIPSO_V4_TAG_LOCAL:
+			if (doi_def->type != CIPSO_V4_MAP_LOCAL)
+				return -EINVAL;
+			break;
 		default:
 			return -EINVAL;
 		}
 	}
 
-	doi_def->valid = 1;
+	atomic_set(&doi_def->refcount, 1);
 	INIT_RCU_HEAD(&doi_def->rcu);
-	INIT_LIST_HEAD(&doi_def->dom_list);
 
 	spin_lock(&cipso_v4_doi_list_lock);
 	if (cipso_v4_doi_search(doi_def->doi) != NULL)
@@ -519,59 +506,129 @@ doi_add_failure:
 }
 
 /**
+ * cipso_v4_doi_free - Frees a DOI definition
+ * @entry: the entry's RCU field
+ *
+ * Description:
+ * This function frees all of the memory associated with a DOI definition.
+ *
+ */
+void cipso_v4_doi_free(struct cipso_v4_doi *doi_def)
+{
+	if (doi_def == NULL)
+		return;
+
+	switch (doi_def->type) {
+	case CIPSO_V4_MAP_TRANS:
+		kfree(doi_def->map.std->lvl.cipso);
+		kfree(doi_def->map.std->lvl.local);
+		kfree(doi_def->map.std->cat.cipso);
+		kfree(doi_def->map.std->cat.local);
+		break;
+	}
+	kfree(doi_def);
+}
+
+/**
+ * cipso_v4_doi_free_rcu - Frees a DOI definition via the RCU pointer
+ * @entry: the entry's RCU field
+ *
+ * Description:
+ * This function is designed to be used as a callback to the call_rcu()
+ * function so that the memory allocated to the DOI definition can be released
+ * safely.
+ *
+ */
+static void cipso_v4_doi_free_rcu(struct rcu_head *entry)
+{
+	struct cipso_v4_doi *doi_def;
+
+	doi_def = container_of(entry, struct cipso_v4_doi, rcu);
+	cipso_v4_doi_free(doi_def);
+}
+
+/**
  * cipso_v4_doi_remove - Remove an existing DOI from the CIPSO protocol engine
  * @doi: the DOI value
  * @audit_secid: the LSM secid to use in the audit message
- * @callback: the DOI cleanup/free callback
  *
  * Description:
- * Removes a DOI definition from the CIPSO engine, @callback is called to
- * free any memory.  The NetLabel routines will be called to release their own
- * LSM domain mappings as well as our own domain list.  Returns zero on
- * success and negative values on failure.
+ * Removes a DOI definition from the CIPSO engine.  The NetLabel routines will
+ * be called to release their own LSM domain mappings as well as our own
+ * domain list.  Returns zero on success and negative values on failure.
  *
  */
-int cipso_v4_doi_remove(u32 doi,
-			struct netlbl_audit *audit_info,
-			void (*callback) (struct rcu_head * head))
+int cipso_v4_doi_remove(u32 doi, struct netlbl_audit *audit_info)
 {
 	struct cipso_v4_doi *doi_def;
-	struct cipso_v4_domhsh_entry *dom_iter;
 
 	spin_lock(&cipso_v4_doi_list_lock);
 	doi_def = cipso_v4_doi_search(doi);
-	if (doi_def != NULL) {
-		doi_def->valid = 0;
-		list_del_rcu(&doi_def->list);
+	if (doi_def == NULL) {
 		spin_unlock(&cipso_v4_doi_list_lock);
-		rcu_read_lock();
-		list_for_each_entry_rcu(dom_iter, &doi_def->dom_list, list)
-			if (dom_iter->valid)
-				netlbl_cfg_map_del(dom_iter->domain,
-						   audit_info);
-		rcu_read_unlock();
-		cipso_v4_cache_invalidate();
-		call_rcu(&doi_def->rcu, callback);
-		return 0;
+		return -ENOENT;
+	}
+	if (!atomic_dec_and_test(&doi_def->refcount)) {
+		spin_unlock(&cipso_v4_doi_list_lock);
+		return -EBUSY;
 	}
+	list_del_rcu(&doi_def->list);
 	spin_unlock(&cipso_v4_doi_list_lock);
 
-	return -ENOENT;
+	cipso_v4_cache_invalidate();
+	call_rcu(&doi_def->rcu, cipso_v4_doi_free_rcu);
+
+	return 0;
 }
 
 /**
- * cipso_v4_doi_getdef - Returns a pointer to a valid DOI definition
+ * cipso_v4_doi_getdef - Returns a reference to a valid DOI definition
  * @doi: the DOI value
  *
  * Description:
  * Searches for a valid DOI definition and if one is found it is returned to
  * the caller.  Otherwise NULL is returned.  The caller must ensure that
- * rcu_read_lock() is held while accessing the returned definition.
+ * rcu_read_lock() is held while accessing the returned definition and the DOI
+ * definition reference count is decremented when the caller is done.
  *
  */
 struct cipso_v4_doi *cipso_v4_doi_getdef(u32 doi)
 {
-	return cipso_v4_doi_search(doi);
+	struct cipso_v4_doi *doi_def;
+
+	rcu_read_lock();
+	doi_def = cipso_v4_doi_search(doi);
+	if (doi_def == NULL)
+		goto doi_getdef_return;
+	if (!atomic_inc_not_zero(&doi_def->refcount))
+		doi_def = NULL;
+
+doi_getdef_return:
+	rcu_read_unlock();
+	return doi_def;
+}
+
+/**
+ * cipso_v4_doi_putdef - Releases a reference for the given DOI definition
+ * @doi_def: the DOI definition
+ *
+ * Description:
+ * Releases a DOI definition reference obtained from cipso_v4_doi_getdef().
+ *
+ */
+void cipso_v4_doi_putdef(struct cipso_v4_doi *doi_def)
+{
+	if (doi_def == NULL)
+		return;
+
+	if (!atomic_dec_and_test(&doi_def->refcount))
+		return;
+	spin_lock(&cipso_v4_doi_list_lock);
+	list_del_rcu(&doi_def->list);
+	spin_unlock(&cipso_v4_doi_list_lock);
+
+	cipso_v4_cache_invalidate();
+	call_rcu(&doi_def->rcu, cipso_v4_doi_free_rcu);
 }
 
 /**
@@ -597,7 +654,7 @@ int cipso_v4_doi_walk(u32 *skip_cnt,
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(iter_doi, &cipso_v4_doi_list, list)
-		if (iter_doi->valid) {
+		if (atomic_read(&iter_doi->refcount) > 0) {
 			if (doi_cnt++ < *skip_cnt)
 				continue;
 			ret_val = callback(iter_doi, cb_arg);
@@ -613,85 +670,6 @@ doi_walk_return:
 	return ret_val;
 }
 
-/**
- * cipso_v4_doi_domhsh_add - Adds a domain entry to a DOI definition
- * @doi_def: the DOI definition
- * @domain: the domain to add
- *
- * Description:
- * Adds the @domain to the DOI specified by @doi_def, this function
- * should only be called by external functions (i.e. NetLabel).  This function
- * does allocate memory.  Returns zero on success, negative values on failure.
- *
- */
-int cipso_v4_doi_domhsh_add(struct cipso_v4_doi *doi_def, const char *domain)
-{
-	struct cipso_v4_domhsh_entry *iter;
-	struct cipso_v4_domhsh_entry *new_dom;
-
-	new_dom = kzalloc(sizeof(*new_dom), GFP_KERNEL);
-	if (new_dom == NULL)
-		return -ENOMEM;
-	if (domain) {
-		new_dom->domain = kstrdup(domain, GFP_KERNEL);
-		if (new_dom->domain == NULL) {
-			kfree(new_dom);
-			return -ENOMEM;
-		}
-	}
-	new_dom->valid = 1;
-	INIT_RCU_HEAD(&new_dom->rcu);
-
-	spin_lock(&cipso_v4_doi_list_lock);
-	list_for_each_entry(iter, &doi_def->dom_list, list)
-		if (iter->valid &&
-		    ((domain != NULL && iter->domain != NULL &&
-		      strcmp(iter->domain, domain) == 0) ||
-		     (domain == NULL && iter->domain == NULL))) {
-			spin_unlock(&cipso_v4_doi_list_lock);
-			kfree(new_dom->domain);
-			kfree(new_dom);
-			return -EEXIST;
-		}
-	list_add_tail_rcu(&new_dom->list, &doi_def->dom_list);
-	spin_unlock(&cipso_v4_doi_list_lock);
-
-	return 0;
-}
-
-/**
- * cipso_v4_doi_domhsh_remove - Removes a domain entry from a DOI definition
- * @doi_def: the DOI definition
- * @domain: the domain to remove
- *
- * Description:
- * Removes the @domain from the DOI specified by @doi_def, this function
- * should only be called by external functions (i.e. NetLabel).   Returns zero
- * on success and negative values on error.
- *
- */
-int cipso_v4_doi_domhsh_remove(struct cipso_v4_doi *doi_def,
-			       const char *domain)
-{
-	struct cipso_v4_domhsh_entry *iter;
-
-	spin_lock(&cipso_v4_doi_list_lock);
-	list_for_each_entry(iter, &doi_def->dom_list, list)
-		if (iter->valid &&
-		    ((domain != NULL && iter->domain != NULL &&
-		      strcmp(iter->domain, domain) == 0) ||
-		     (domain == NULL && iter->domain == NULL))) {
-			iter->valid = 0;
-			list_del_rcu(&iter->list);
-			spin_unlock(&cipso_v4_doi_list_lock);
-			call_rcu(&iter->rcu, cipso_v4_doi_domhsh_free);
-			return 0;
-		}
-	spin_unlock(&cipso_v4_doi_list_lock);
-
-	return -ENOENT;
-}
-
 /*
  * Label Mapping Functions
  */
@@ -712,7 +690,7 @@ static int cipso_v4_map_lvl_valid(const struct cipso_v4_doi *doi_def, u8 level)
 	switch (doi_def->type) {
 	case CIPSO_V4_MAP_PASS:
 		return 0;
-	case CIPSO_V4_MAP_STD:
+	case CIPSO_V4_MAP_TRANS:
 		if (doi_def->map.std->lvl.cipso[level] < CIPSO_V4_INV_LVL)
 			return 0;
 		break;
@@ -741,7 +719,7 @@ static int cipso_v4_map_lvl_hton(const struct cipso_v4_doi *doi_def,
 	case CIPSO_V4_MAP_PASS:
 		*net_lvl = host_lvl;
 		return 0;
-	case CIPSO_V4_MAP_STD:
+	case CIPSO_V4_MAP_TRANS:
 		if (host_lvl < doi_def->map.std->lvl.local_size &&
 		    doi_def->map.std->lvl.local[host_lvl] < CIPSO_V4_INV_LVL) {
 			*net_lvl = doi_def->map.std->lvl.local[host_lvl];
@@ -775,7 +753,7 @@ static int cipso_v4_map_lvl_ntoh(const struct cipso_v4_doi *doi_def,
 	case CIPSO_V4_MAP_PASS:
 		*host_lvl = net_lvl;
 		return 0;
-	case CIPSO_V4_MAP_STD:
+	case CIPSO_V4_MAP_TRANS:
 		map_tbl = doi_def->map.std;
 		if (net_lvl < map_tbl->lvl.cipso_size &&
 		    map_tbl->lvl.cipso[net_lvl] < CIPSO_V4_INV_LVL) {
@@ -812,7 +790,7 @@ static int cipso_v4_map_cat_rbm_valid(const struct cipso_v4_doi *doi_def,
 	switch (doi_def->type) {
 	case CIPSO_V4_MAP_PASS:
 		return 0;
-	case CIPSO_V4_MAP_STD:
+	case CIPSO_V4_MAP_TRANS:
 		cipso_cat_size = doi_def->map.std->cat.cipso_size;
 		cipso_array = doi_def->map.std->cat.cipso;
 		for (;;) {
@@ -860,7 +838,7 @@ static int cipso_v4_map_cat_rbm_hton(const struct cipso_v4_doi *doi_def,
 	u32 host_cat_size = 0;
 	u32 *host_cat_array = NULL;
 
-	if (doi_def->type == CIPSO_V4_MAP_STD) {
+	if (doi_def->type == CIPSO_V4_MAP_TRANS) {
 		host_cat_size = doi_def->map.std->cat.local_size;
 		host_cat_array = doi_def->map.std->cat.local;
 	}
@@ -875,7 +853,7 @@ static int cipso_v4_map_cat_rbm_hton(const struct cipso_v4_doi *doi_def,
 		case CIPSO_V4_MAP_PASS:
 			net_spot = host_spot;
 			break;
-		case CIPSO_V4_MAP_STD:
+		case CIPSO_V4_MAP_TRANS:
 			if (host_spot >= host_cat_size)
 				return -EPERM;
 			net_spot = host_cat_array[host_spot];
@@ -921,7 +899,7 @@ static int cipso_v4_map_cat_rbm_ntoh(const struct cipso_v4_doi *doi_def,
 	u32 net_cat_size = 0;
 	u32 *net_cat_array = NULL;
 
-	if (doi_def->type == CIPSO_V4_MAP_STD) {
+	if (doi_def->type == CIPSO_V4_MAP_TRANS) {
 		net_cat_size = doi_def->map.std->cat.cipso_size;
 		net_cat_array = doi_def->map.std->cat.cipso;
 	}
@@ -941,7 +919,7 @@ static int cipso_v4_map_cat_rbm_ntoh(const struct cipso_v4_doi *doi_def,
 		case CIPSO_V4_MAP_PASS:
 			host_spot = net_spot;
 			break;
-		case CIPSO_V4_MAP_STD:
+		case CIPSO_V4_MAP_TRANS:
 			if (net_spot >= net_cat_size)
 				return -EPERM;
 			host_spot = net_cat_array[net_spot];
@@ -1277,7 +1255,7 @@ static int cipso_v4_gentag_rbm(const struct cipso_v4_doi *doi_def,
 	} else
 		tag_len = 4;
 
-	buffer[0] = 0x01;
+	buffer[0] = CIPSO_V4_TAG_RBITMAP;
 	buffer[1] = tag_len;
 	buffer[3] = level;
 
@@ -1373,7 +1351,7 @@ static int cipso_v4_gentag_enum(const struct cipso_v4_doi *doi_def,
 	} else
 		tag_len = 4;
 
-	buffer[0] = 0x02;
+	buffer[0] = CIPSO_V4_TAG_ENUM;
 	buffer[1] = tag_len;
 	buffer[3] = level;
 
@@ -1469,7 +1447,7 @@ static int cipso_v4_gentag_rng(const struct cipso_v4_doi *doi_def,
 	} else
 		tag_len = 4;
 
-	buffer[0] = 0x05;
+	buffer[0] = CIPSO_V4_TAG_RANGE;
 	buffer[1] = tag_len;
 	buffer[3] = level;
 
@@ -1523,6 +1501,54 @@ static int cipso_v4_parsetag_rng(const struct cipso_v4_doi *doi_def,
 }
 
 /**
+ * cipso_v4_gentag_loc - Generate a CIPSO local tag (non-standard)
+ * @doi_def: the DOI definition
+ * @secattr: the security attributes
+ * @buffer: the option buffer
+ * @buffer_len: length of buffer in bytes
+ *
+ * Description:
+ * Generate a CIPSO option using the local tag.  Returns the size of the tag
+ * on success, negative values on failure.
+ *
+ */
+static int cipso_v4_gentag_loc(const struct cipso_v4_doi *doi_def,
+			       const struct netlbl_lsm_secattr *secattr,
+			       unsigned char *buffer,
+			       u32 buffer_len)
+{
+	if (!(secattr->flags & NETLBL_SECATTR_SECID))
+		return -EPERM;
+
+	buffer[0] = CIPSO_V4_TAG_LOCAL;
+	buffer[1] = CIPSO_V4_TAG_LOC_BLEN;
+	*(u32 *)&buffer[2] = secattr->attr.secid;
+
+	return CIPSO_V4_TAG_LOC_BLEN;
+}
+
+/**
+ * cipso_v4_parsetag_loc - Parse a CIPSO local tag
+ * @doi_def: the DOI definition
+ * @tag: the CIPSO tag
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Parse a CIPSO local tag and return the security attributes in @secattr.
+ * Return zero on success, negatives values on failure.
+ *
+ */
+static int cipso_v4_parsetag_loc(const struct cipso_v4_doi *doi_def,
+				 const unsigned char *tag,
+				 struct netlbl_lsm_secattr *secattr)
+{
+	secattr->attr.secid = *(u32 *)&tag[2];
+	secattr->flags |= NETLBL_SECATTR_SECID;
+
+	return 0;
+}
+
+/**
  * cipso_v4_validate - Validate a CIPSO option
  * @option: the start of the option, on error it is set to point to the error
  *
@@ -1541,7 +1567,7 @@ static int cipso_v4_parsetag_rng(const struct cipso_v4_doi *doi_def,
  *   that is unrecognized."
  *
  */
-int cipso_v4_validate(unsigned char **option)
+int cipso_v4_validate(const struct sk_buff *skb, unsigned char **option)
 {
 	unsigned char *opt = *option;
 	unsigned char *tag;
@@ -1566,7 +1592,7 @@ int cipso_v4_validate(unsigned char **option)
 		goto validate_return_locked;
 	}
 
-	opt_iter = 6;
+	opt_iter = CIPSO_V4_HDR_LEN;
 	tag = opt + opt_iter;
 	while (opt_iter < opt_len) {
 		for (tag_iter = 0; doi_def->tags[tag_iter] != tag[0];)
@@ -1584,7 +1610,7 @@ int cipso_v4_validate(unsigned char **option)
 
 		switch (tag[0]) {
 		case CIPSO_V4_TAG_RBITMAP:
-			if (tag_len < 4) {
+			if (tag_len < CIPSO_V4_TAG_RBM_BLEN) {
 				err_offset = opt_iter + 1;
 				goto validate_return_locked;
 			}
@@ -1602,7 +1628,7 @@ int cipso_v4_validate(unsigned char **option)
 					err_offset = opt_iter + 3;
 					goto validate_return_locked;
 				}
-				if (tag_len > 4 &&
+				if (tag_len > CIPSO_V4_TAG_RBM_BLEN &&
 				    cipso_v4_map_cat_rbm_valid(doi_def,
 							    &tag[4],
 							    tag_len - 4) < 0) {
@@ -1612,7 +1638,7 @@ int cipso_v4_validate(unsigned char **option)
 			}
 			break;
 		case CIPSO_V4_TAG_ENUM:
-			if (tag_len < 4) {
+			if (tag_len < CIPSO_V4_TAG_ENUM_BLEN) {
 				err_offset = opt_iter + 1;
 				goto validate_return_locked;
 			}
@@ -1622,7 +1648,7 @@ int cipso_v4_validate(unsigned char **option)
 				err_offset = opt_iter + 3;
 				goto validate_return_locked;
 			}
-			if (tag_len > 4 &&
+			if (tag_len > CIPSO_V4_TAG_ENUM_BLEN &&
 			    cipso_v4_map_cat_enum_valid(doi_def,
 							&tag[4],
 							tag_len - 4) < 0) {
@@ -1631,7 +1657,7 @@ int cipso_v4_validate(unsigned char **option)
 			}
 			break;
 		case CIPSO_V4_TAG_RANGE:
-			if (tag_len < 4) {
+			if (tag_len < CIPSO_V4_TAG_RNG_BLEN) {
 				err_offset = opt_iter + 1;
 				goto validate_return_locked;
 			}
@@ -1641,7 +1667,7 @@ int cipso_v4_validate(unsigned char **option)
 				err_offset = opt_iter + 3;
 				goto validate_return_locked;
 			}
-			if (tag_len > 4 &&
+			if (tag_len > CIPSO_V4_TAG_RNG_BLEN &&
 			    cipso_v4_map_cat_rng_valid(doi_def,
 						       &tag[4],
 						       tag_len - 4) < 0) {
@@ -1649,6 +1675,19 @@ int cipso_v4_validate(unsigned char **option)
 				goto validate_return_locked;
 			}
 			break;
+		case CIPSO_V4_TAG_LOCAL:
+			/* This is a non-standard tag that we only allow for
+			 * local connections, so if the incoming interface is
+			 * not the loopback device drop the packet. */
+			if (!(skb->dev->flags & IFF_LOOPBACK)) {
+				err_offset = opt_iter;
+				goto validate_return_locked;
+			}
+			if (tag_len != CIPSO_V4_TAG_LOC_BLEN) {
+				err_offset = opt_iter + 1;
+				goto validate_return_locked;
+			}
+			break;
 		default:
 			err_offset = opt_iter;
 			goto validate_return_locked;
@@ -1704,48 +1743,27 @@ void cipso_v4_error(struct sk_buff *skb, int error, u32 gateway)
 }
 
 /**
- * cipso_v4_sock_setattr - Add a CIPSO option to a socket
- * @sk: the socket
+ * cipso_v4_genopt - Generate a CIPSO option
+ * @buf: the option buffer
+ * @buf_len: the size of opt_buf
  * @doi_def: the CIPSO DOI to use
- * @secattr: the specific security attributes of the socket
+ * @secattr: the security attributes
  *
  * Description:
- * Set the CIPSO option on the given socket using the DOI definition and
- * security attributes passed to the function.  This function requires
- * exclusive access to @sk, which means it either needs to be in the
- * process of being created or locked.  Returns zero on success and negative
- * values on failure.
+ * Generate a CIPSO option using the DOI definition and security attributes
+ * passed to the function.  Returns the length of the option on success and
+ * negative values on failure.
  *
  */
-int cipso_v4_sock_setattr(struct sock *sk,
-			  const struct cipso_v4_doi *doi_def,
-			  const struct netlbl_lsm_secattr *secattr)
+static int cipso_v4_genopt(unsigned char *buf, u32 buf_len,
+			   const struct cipso_v4_doi *doi_def,
+			   const struct netlbl_lsm_secattr *secattr)
 {
-	int ret_val = -EPERM;
+	int ret_val;
 	u32 iter;
-	unsigned char *buf;
-	u32 buf_len = 0;
-	u32 opt_len;
-	struct ip_options *opt = NULL;
-	struct inet_sock *sk_inet;
-	struct inet_connection_sock *sk_conn;
 
-	/* In the case of sock_create_lite(), the sock->sk field is not
-	 * defined yet but it is not a problem as the only users of these
-	 * "lite" PF_INET sockets are functions which do an accept() call
-	 * afterwards so we will label the socket as part of the accept(). */
-	if (sk == NULL)
-		return 0;
-
-	/* We allocate the maximum CIPSO option size here so we are probably
-	 * being a little wasteful, but it makes our life _much_ easier later
-	 * on and after all we are only talking about 40 bytes. */
-	buf_len = CIPSO_V4_OPT_LEN_MAX;
-	buf = kmalloc(buf_len, GFP_ATOMIC);
-	if (buf == NULL) {
-		ret_val = -ENOMEM;
-		goto socket_setattr_failure;
-	}
+	if (buf_len <= CIPSO_V4_HDR_LEN)
+		return -ENOSPC;
 
 	/* XXX - This code assumes only one tag per CIPSO option which isn't
 	 * really a good assumption to make but since we only support the MAC
@@ -1772,9 +1790,14 @@ int cipso_v4_sock_setattr(struct sock *sk,
 						   &buf[CIPSO_V4_HDR_LEN],
 						   buf_len - CIPSO_V4_HDR_LEN);
 			break;
+		case CIPSO_V4_TAG_LOCAL:
+			ret_val = cipso_v4_gentag_loc(doi_def,
+						   secattr,
+						   &buf[CIPSO_V4_HDR_LEN],
+						   buf_len - CIPSO_V4_HDR_LEN);
+			break;
 		default:
-			ret_val = -EPERM;
-			goto socket_setattr_failure;
+			return -EPERM;
 		}
 
 		iter++;
@@ -1782,9 +1805,58 @@ int cipso_v4_sock_setattr(struct sock *sk,
 		 iter < CIPSO_V4_TAG_MAXCNT &&
 		 doi_def->tags[iter] != CIPSO_V4_TAG_INVALID);
 	if (ret_val < 0)
-		goto socket_setattr_failure;
+		return ret_val;
 	cipso_v4_gentag_hdr(doi_def, buf, ret_val);
-	buf_len = CIPSO_V4_HDR_LEN + ret_val;
+	return CIPSO_V4_HDR_LEN + ret_val;
+}
+
+/**
+ * cipso_v4_sock_setattr - Add a CIPSO option to a socket
+ * @sk: the socket
+ * @doi_def: the CIPSO DOI to use
+ * @secattr: the specific security attributes of the socket
+ *
+ * Description:
+ * Set the CIPSO option on the given socket using the DOI definition and
+ * security attributes passed to the function.  This function requires
+ * exclusive access to @sk, which means it either needs to be in the
+ * process of being created or locked.  Returns zero on success and negative
+ * values on failure.
+ *
+ */
+int cipso_v4_sock_setattr(struct sock *sk,
+			  const struct cipso_v4_doi *doi_def,
+			  const struct netlbl_lsm_secattr *secattr)
+{
+	int ret_val = -EPERM;
+	unsigned char *buf = NULL;
+	u32 buf_len;
+	u32 opt_len;
+	struct ip_options *opt = NULL;
+	struct inet_sock *sk_inet;
+	struct inet_connection_sock *sk_conn;
+
+	/* In the case of sock_create_lite(), the sock->sk field is not
+	 * defined yet but it is not a problem as the only users of these
+	 * "lite" PF_INET sockets are functions which do an accept() call
+	 * afterwards so we will label the socket as part of the accept(). */
+	if (sk == NULL)
+		return 0;
+
+	/* We allocate the maximum CIPSO option size here so we are probably
+	 * being a little wasteful, but it makes our life _much_ easier later
+	 * on and after all we are only talking about 40 bytes. */
+	buf_len = CIPSO_V4_OPT_LEN_MAX;
+	buf = kmalloc(buf_len, GFP_ATOMIC);
+	if (buf == NULL) {
+		ret_val = -ENOMEM;
+		goto socket_setattr_failure;
+	}
+
+	ret_val = cipso_v4_genopt(buf, buf_len, doi_def, secattr);
+	if (ret_val < 0)
+		goto socket_setattr_failure;
+	buf_len = ret_val;
 
 	/* We can't use ip_options_get() directly because it makes a call to
 	 * ip_options_get_alloc() which allocates memory with GFP_KERNEL and
@@ -1822,6 +1894,80 @@ socket_setattr_failure:
 }
 
 /**
+ * cipso_v4_sock_delattr - Delete the CIPSO option from a socket
+ * @sk: the socket
+ *
+ * Description:
+ * Removes the CIPSO option from a socket, if present.
+ *
+ */
+void cipso_v4_sock_delattr(struct sock *sk)
+{
+	u8 hdr_delta;
+	struct ip_options *opt;
+	struct inet_sock *sk_inet;
+
+	sk_inet = inet_sk(sk);
+	opt = sk_inet->opt;
+	if (opt == NULL || opt->cipso == 0)
+		return;
+
+	if (opt->srr || opt->rr || opt->ts || opt->router_alert) {
+		u8 cipso_len;
+		u8 cipso_off;
+		unsigned char *cipso_ptr;
+		int iter;
+		int optlen_new;
+
+		cipso_off = opt->cipso - sizeof(struct iphdr);
+		cipso_ptr = &opt->__data[cipso_off];
+		cipso_len = cipso_ptr[1];
+
+		if (opt->srr > opt->cipso)
+			opt->srr -= cipso_len;
+		if (opt->rr > opt->cipso)
+			opt->rr -= cipso_len;
+		if (opt->ts > opt->cipso)
+			opt->ts -= cipso_len;
+		if (opt->router_alert > opt->cipso)
+			opt->router_alert -= cipso_len;
+		opt->cipso = 0;
+
+		memmove(cipso_ptr, cipso_ptr + cipso_len,
+			opt->optlen - cipso_off - cipso_len);
+
+		/* determining the new total option length is tricky because of
+		 * the padding necessary, the only thing i can think to do at
+		 * this point is walk the options one-by-one, skipping the
+		 * padding at the end to determine the actual option size and
+		 * from there we can determine the new total option length */
+		iter = 0;
+		optlen_new = 0;
+		while (iter < opt->optlen)
+			if (opt->__data[iter] != IPOPT_NOP) {
+				iter += opt->__data[iter + 1];
+				optlen_new = iter;
+			} else
+				iter++;
+		hdr_delta = opt->optlen;
+		opt->optlen = (optlen_new + 3) & ~3;
+		hdr_delta -= opt->optlen;
+	} else {
+		/* only the cipso option was present on the socket so we can
+		 * remove the entire option struct */
+		sk_inet->opt = NULL;
+		hdr_delta = opt->optlen;
+		kfree(opt);
+	}
+
+	if (sk_inet->is_icsk && hdr_delta > 0) {
+		struct inet_connection_sock *sk_conn = inet_csk(sk);
+		sk_conn->icsk_ext_hdr_len -= hdr_delta;
+		sk_conn->icsk_sync_mss(sk, sk_conn->icsk_pmtu_cookie);
+	}
+}
+
+/**
  * cipso_v4_getattr - Helper function for the cipso_v4_*_getattr functions
  * @cipso: the CIPSO v4 option
  * @secattr: the security attributes
@@ -1859,6 +2005,9 @@ static int cipso_v4_getattr(const unsigned char *cipso,
 	case CIPSO_V4_TAG_RANGE:
 		ret_val = cipso_v4_parsetag_rng(doi_def, &cipso[6], secattr);
 		break;
+	case CIPSO_V4_TAG_LOCAL:
+		ret_val = cipso_v4_parsetag_loc(doi_def, &cipso[6], secattr);
+		break;
 	}
 	if (ret_val == 0)
 		secattr->type = NETLBL_NLTYPE_CIPSOV4;
@@ -1893,6 +2042,123 @@ int cipso_v4_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr)
 }
 
 /**
+ * cipso_v4_skbuff_setattr - Set the CIPSO option on a packet
+ * @skb: the packet
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Set the CIPSO option on the given packet based on the security attributes.
+ * Returns a pointer to the IP header on success and NULL on failure.
+ *
+ */
+int cipso_v4_skbuff_setattr(struct sk_buff *skb,
+			    const struct cipso_v4_doi *doi_def,
+			    const struct netlbl_lsm_secattr *secattr)
+{
+	int ret_val;
+	struct iphdr *iph;
+	struct ip_options *opt = &IPCB(skb)->opt;
+	unsigned char buf[CIPSO_V4_OPT_LEN_MAX];
+	u32 buf_len = CIPSO_V4_OPT_LEN_MAX;
+	u32 opt_len;
+	int len_delta;
+
+	buf_len = cipso_v4_genopt(buf, buf_len, doi_def, secattr);
+	if (buf_len < 0)
+		return buf_len;
+	opt_len = (buf_len + 3) & ~3;
+
+	/* we overwrite any existing options to ensure that we have enough
+	 * room for the CIPSO option, the reason is that we _need_ to guarantee
+	 * that the security label is applied to the packet - we do the same
+	 * thing when using the socket options and it hasn't caused a problem,
+	 * if we need to we can always revisit this choice later */
+
+	len_delta = opt_len - opt->optlen;
+	/* if we don't ensure enough headroom we could panic on the skb_push()
+	 * call below so make sure we have enough, we are also "mangling" the
+	 * packet so we should probably do a copy-on-write call anyway */
+	ret_val = skb_cow(skb, skb_headroom(skb) + len_delta);
+	if (ret_val < 0)
+		return ret_val;
+
+	if (len_delta > 0) {
+		/* we assume that the header + opt->optlen have already been
+		 * "pushed" in ip_options_build() or similar */
+		iph = ip_hdr(skb);
+		skb_push(skb, len_delta);
+		memmove((char *)iph - len_delta, iph, iph->ihl << 2);
+		skb_reset_network_header(skb);
+		iph = ip_hdr(skb);
+	} else if (len_delta < 0) {
+		iph = ip_hdr(skb);
+		memset(iph + 1, IPOPT_NOP, opt->optlen);
+	} else
+		iph = ip_hdr(skb);
+
+	if (opt->optlen > 0)
+		memset(opt, 0, sizeof(*opt));
+	opt->optlen = opt_len;
+	opt->cipso = sizeof(struct iphdr);
+	opt->is_changed = 1;
+
+	/* we have to do the following because we are being called from a
+	 * netfilter hook which means the packet already has had the header
+	 * fields populated and the checksum calculated - yes this means we
+	 * are doing more work than needed but we do it to keep the core
+	 * stack clean and tidy */
+	memcpy(iph + 1, buf, buf_len);
+	if (opt_len > buf_len)
+		memset((char *)(iph + 1) + buf_len, 0, opt_len - buf_len);
+	if (len_delta != 0) {
+		iph->ihl = 5 + (opt_len >> 2);
+		iph->tot_len = htons(skb->len);
+	}
+	ip_send_check(iph);
+
+	return 0;
+}
+
+/**
+ * cipso_v4_skbuff_delattr - Delete any CIPSO options from a packet
+ * @skb: the packet
+ *
+ * Description:
+ * Removes any and all CIPSO options from the given packet.  Returns zero on
+ * success, negative values on failure.
+ *
+ */
+int cipso_v4_skbuff_delattr(struct sk_buff *skb)
+{
+	int ret_val;
+	struct iphdr *iph;
+	struct ip_options *opt = &IPCB(skb)->opt;
+	unsigned char *cipso_ptr;
+
+	if (opt->cipso == 0)
+		return 0;
+
+	/* since we are changing the packet we should make a copy */
+	ret_val = skb_cow(skb, skb_headroom(skb));
+	if (ret_val < 0)
+		return ret_val;
+
+	/* the easiest thing to do is just replace the cipso option with noop
+	 * options since we don't change the size of the packet, although we
+	 * still need to recalculate the checksum */
+
+	iph = ip_hdr(skb);
+	cipso_ptr = (unsigned char *)iph + opt->cipso;
+	memset(cipso_ptr, IPOPT_NOOP, cipso_ptr[1]);
+	opt->cipso = 0;
+	opt->is_changed = 1;
+
+	ip_send_check(iph);
+
+	return 0;
+}
+
+/**
  * cipso_v4_skbuff_getattr - Get the security attributes from the CIPSO option
  * @skb: the packet
  * @secattr: the security attributes
diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c
index be3f18a7a40e..2c88da6e7862 100644
--- a/net/ipv4/ip_options.c
+++ b/net/ipv4/ip_options.c
@@ -438,7 +438,7 @@ int ip_options_compile(struct net *net,
 				goto error;
 			}
 			opt->cipso = optptr - iph;
-			if (cipso_v4_validate(&optptr)) {
+			if (cipso_v4_validate(skb, &optptr)) {
 				pp_ptr = optptr;
 				goto error;
 			}
diff --git a/net/netlabel/Makefile b/net/netlabel/Makefile
index 8af18c0a47d9..ea750e9df65f 100644
--- a/net/netlabel/Makefile
+++ b/net/netlabel/Makefile
@@ -5,7 +5,8 @@
 #
 
 # base objects
-obj-y	:= netlabel_user.o netlabel_kapi.o netlabel_domainhash.o
+obj-y	:= netlabel_user.o netlabel_kapi.o
+obj-y	+= netlabel_domainhash.o netlabel_addrlist.o
 
 # management objects
 obj-y	+= netlabel_mgmt.o
diff --git a/net/netlabel/netlabel_addrlist.c b/net/netlabel/netlabel_addrlist.c
new file mode 100644
index 000000000000..b0925a303353
--- /dev/null
+++ b/net/netlabel/netlabel_addrlist.c
@@ -0,0 +1,388 @@
+/*
+ * NetLabel Network Address Lists
+ *
+ * This file contains network address list functions used to manage ordered
+ * lists of network addresses for use by the NetLabel subsystem.  The NetLabel
+ * system manages static and dynamic label mappings for network protocols such
+ * as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2008
+ *
+ * This program is free software;  you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY;  without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program;  if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#include <linux/types.h>
+#include <linux/rcupdate.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+#include <linux/in.h>
+#include <linux/in6.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <net/ip.h>
+#include <net/ipv6.h>
+#include <linux/audit.h>
+
+#include "netlabel_addrlist.h"
+
+/*
+ * Address List Functions
+ */
+
+/**
+ * netlbl_af4list_search - Search for a matching IPv4 address entry
+ * @addr: IPv4 address
+ * @head: the list head
+ *
+ * Description:
+ * Searches the IPv4 address list given by @head.  If a matching address entry
+ * is found it is returned, otherwise NULL is returned.  The caller is
+ * responsible for calling the rcu_read_[un]lock() functions.
+ *
+ */
+struct netlbl_af4list *netlbl_af4list_search(__be32 addr,
+					     struct list_head *head)
+{
+	struct netlbl_af4list *iter;
+
+	list_for_each_entry_rcu(iter, head, list)
+		if (iter->valid && (addr & iter->mask) == iter->addr)
+			return iter;
+
+	return NULL;
+}
+
+/**
+ * netlbl_af4list_search_exact - Search for an exact IPv4 address entry
+ * @addr: IPv4 address
+ * @mask: IPv4 address mask
+ * @head: the list head
+ *
+ * Description:
+ * Searches the IPv4 address list given by @head.  If an exact match if found
+ * it is returned, otherwise NULL is returned.  The caller is responsible for
+ * calling the rcu_read_[un]lock() functions.
+ *
+ */
+struct netlbl_af4list *netlbl_af4list_search_exact(__be32 addr,
+						   __be32 mask,
+						   struct list_head *head)
+{
+	struct netlbl_af4list *iter;
+
+	list_for_each_entry_rcu(iter, head, list)
+		if (iter->valid && iter->addr == addr && iter->mask == mask)
+			return iter;
+
+	return NULL;
+}
+
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+/**
+ * netlbl_af6list_search - Search for a matching IPv6 address entry
+ * @addr: IPv6 address
+ * @head: the list head
+ *
+ * Description:
+ * Searches the IPv6 address list given by @head.  If a matching address entry
+ * is found it is returned, otherwise NULL is returned.  The caller is
+ * responsible for calling the rcu_read_[un]lock() functions.
+ *
+ */
+struct netlbl_af6list *netlbl_af6list_search(const struct in6_addr *addr,
+					     struct list_head *head)
+{
+	struct netlbl_af6list *iter;
+
+	list_for_each_entry_rcu(iter, head, list)
+		if (iter->valid &&
+		    ipv6_masked_addr_cmp(&iter->addr, &iter->mask, addr) == 0)
+			return iter;
+
+	return NULL;
+}
+
+/**
+ * netlbl_af6list_search_exact - Search for an exact IPv6 address entry
+ * @addr: IPv6 address
+ * @mask: IPv6 address mask
+ * @head: the list head
+ *
+ * Description:
+ * Searches the IPv6 address list given by @head.  If an exact match if found
+ * it is returned, otherwise NULL is returned.  The caller is responsible for
+ * calling the rcu_read_[un]lock() functions.
+ *
+ */
+struct netlbl_af6list *netlbl_af6list_search_exact(const struct in6_addr *addr,
+						   const struct in6_addr *mask,
+						   struct list_head *head)
+{
+	struct netlbl_af6list *iter;
+
+	list_for_each_entry_rcu(iter, head, list)
+		if (iter->valid &&
+		    ipv6_addr_equal(&iter->addr, addr) &&
+		    ipv6_addr_equal(&iter->mask, mask))
+			return iter;
+
+	return NULL;
+}
+#endif /* IPv6 */
+
+/**
+ * netlbl_af4list_add - Add a new IPv4 address entry to a list
+ * @entry: address entry
+ * @head: the list head
+ *
+ * Description:
+ * Add a new address entry to the list pointed to by @head.  On success zero is
+ * returned, otherwise a negative value is returned.  The caller is responsible
+ * for calling the necessary locking functions.
+ *
+ */
+int netlbl_af4list_add(struct netlbl_af4list *entry, struct list_head *head)
+{
+	struct netlbl_af4list *iter;
+
+	iter = netlbl_af4list_search(entry->addr, head);
+	if (iter != NULL &&
+	    iter->addr == entry->addr && iter->mask == entry->mask)
+		return -EEXIST;
+
+	/* in order to speed up address searches through the list (the common
+	 * case) we need to keep the list in order based on the size of the
+	 * address mask such that the entry with the widest mask (smallest
+	 * numerical value) appears first in the list */
+	list_for_each_entry_rcu(iter, head, list)
+		if (iter->valid &&
+		    ntohl(entry->mask) > ntohl(iter->mask)) {
+			__list_add_rcu(&entry->list,
+				       iter->list.prev,
+				       &iter->list);
+			return 0;
+		}
+	list_add_tail_rcu(&entry->list, head);
+	return 0;
+}
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+/**
+ * netlbl_af6list_add - Add a new IPv6 address entry to a list
+ * @entry: address entry
+ * @head: the list head
+ *
+ * Description:
+ * Add a new address entry to the list pointed to by @head.  On success zero is
+ * returned, otherwise a negative value is returned.  The caller is responsible
+ * for calling the necessary locking functions.
+ *
+ */
+int netlbl_af6list_add(struct netlbl_af6list *entry, struct list_head *head)
+{
+	struct netlbl_af6list *iter;
+
+	iter = netlbl_af6list_search(&entry->addr, head);
+	if (iter != NULL &&
+	    ipv6_addr_equal(&iter->addr, &entry->addr) &&
+	    ipv6_addr_equal(&iter->mask, &entry->mask))
+		return -EEXIST;
+
+	/* in order to speed up address searches through the list (the common
+	 * case) we need to keep the list in order based on the size of the
+	 * address mask such that the entry with the widest mask (smallest
+	 * numerical value) appears first in the list */
+	list_for_each_entry_rcu(iter, head, list)
+		if (iter->valid &&
+		    ipv6_addr_cmp(&entry->mask, &iter->mask) > 0) {
+			__list_add_rcu(&entry->list,
+				       iter->list.prev,
+				       &iter->list);
+			return 0;
+		}
+	list_add_tail_rcu(&entry->list, head);
+	return 0;
+}
+#endif /* IPv6 */
+
+/**
+ * netlbl_af4list_remove_entry - Remove an IPv4 address entry
+ * @entry: address entry
+ *
+ * Description:
+ * Remove the specified IP address entry.  The caller is responsible for
+ * calling the necessary locking functions.
+ *
+ */
+void netlbl_af4list_remove_entry(struct netlbl_af4list *entry)
+{
+	entry->valid = 0;
+	list_del_rcu(&entry->list);
+}
+
+/**
+ * netlbl_af4list_remove - Remove an IPv4 address entry
+ * @addr: IP address
+ * @mask: IP address mask
+ * @head: the list head
+ *
+ * Description:
+ * Remove an IP address entry from the list pointed to by @head.  Returns the
+ * entry on success, NULL on failure.  The caller is responsible for calling
+ * the necessary locking functions.
+ *
+ */
+struct netlbl_af4list *netlbl_af4list_remove(__be32 addr, __be32 mask,
+					     struct list_head *head)
+{
+	struct netlbl_af4list *entry;
+
+	entry = netlbl_af4list_search(addr, head);
+	if (entry != NULL && entry->addr == addr && entry->mask == mask) {
+		netlbl_af4list_remove_entry(entry);
+		return entry;
+	}
+
+	return NULL;
+}
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+/**
+ * netlbl_af6list_remove_entry - Remove an IPv6 address entry
+ * @entry: address entry
+ *
+ * Description:
+ * Remove the specified IP address entry.  The caller is responsible for
+ * calling the necessary locking functions.
+ *
+ */
+void netlbl_af6list_remove_entry(struct netlbl_af6list *entry)
+{
+	entry->valid = 0;
+	list_del_rcu(&entry->list);
+}
+
+/**
+ * netlbl_af6list_remove - Remove an IPv6 address entry
+ * @addr: IP address
+ * @mask: IP address mask
+ * @head: the list head
+ *
+ * Description:
+ * Remove an IP address entry from the list pointed to by @head.  Returns the
+ * entry on success, NULL on failure.  The caller is responsible for calling
+ * the necessary locking functions.
+ *
+ */
+struct netlbl_af6list *netlbl_af6list_remove(const struct in6_addr *addr,
+					     const struct in6_addr *mask,
+					     struct list_head *head)
+{
+	struct netlbl_af6list *entry;
+
+	entry = netlbl_af6list_search(addr, head);
+	if (entry != NULL &&
+	    ipv6_addr_equal(&entry->addr, addr) &&
+	    ipv6_addr_equal(&entry->mask, mask)) {
+		netlbl_af6list_remove_entry(entry);
+		return entry;
+	}
+
+	return NULL;
+}
+#endif /* IPv6 */
+
+/*
+ * Audit Helper Functions
+ */
+
+/**
+ * netlbl_af4list_audit_addr - Audit an IPv4 address
+ * @audit_buf: audit buffer
+ * @src: true if source address, false if destination
+ * @dev: network interface
+ * @addr: IP address
+ * @mask: IP address mask
+ *
+ * Description:
+ * Write the IPv4 address and address mask, if necessary, to @audit_buf.
+ *
+ */
+void netlbl_af4list_audit_addr(struct audit_buffer *audit_buf,
+					int src, const char *dev,
+					__be32 addr, __be32 mask)
+{
+	u32 mask_val = ntohl(mask);
+	char *dir = (src ? "src" : "dst");
+
+	if (dev != NULL)
+		audit_log_format(audit_buf, " netif=%s", dev);
+	audit_log_format(audit_buf, " %s=" NIPQUAD_FMT, dir, NIPQUAD(addr));
+	if (mask_val != 0xffffffff) {
+		u32 mask_len = 0;
+		while (mask_val > 0) {
+			mask_val <<= 1;
+			mask_len++;
+		}
+		audit_log_format(audit_buf, " %s_prefixlen=%d", dir, mask_len);
+	}
+}
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+/**
+ * netlbl_af6list_audit_addr - Audit an IPv6 address
+ * @audit_buf: audit buffer
+ * @src: true if source address, false if destination
+ * @dev: network interface
+ * @addr: IP address
+ * @mask: IP address mask
+ *
+ * Description:
+ * Write the IPv6 address and address mask, if necessary, to @audit_buf.
+ *
+ */
+void netlbl_af6list_audit_addr(struct audit_buffer *audit_buf,
+				 int src,
+				 const char *dev,
+				 const struct in6_addr *addr,
+				 const struct in6_addr *mask)
+{
+	char *dir = (src ? "src" : "dst");
+
+	if (dev != NULL)
+		audit_log_format(audit_buf, " netif=%s", dev);
+	audit_log_format(audit_buf, " %s=" NIP6_FMT, dir, NIP6(*addr));
+	if (ntohl(mask->s6_addr32[3]) != 0xffffffff) {
+		u32 mask_len = 0;
+		u32 mask_val;
+		int iter = -1;
+		while (ntohl(mask->s6_addr32[++iter]) == 0xffffffff)
+			mask_len += 32;
+		mask_val = ntohl(mask->s6_addr32[iter]);
+		while (mask_val > 0) {
+			mask_val <<= 1;
+			mask_len++;
+		}
+		audit_log_format(audit_buf, " %s_prefixlen=%d", dir, mask_len);
+	}
+}
+#endif /* IPv6 */
diff --git a/net/netlabel/netlabel_addrlist.h b/net/netlabel/netlabel_addrlist.h
new file mode 100644
index 000000000000..0242bead405f
--- /dev/null
+++ b/net/netlabel/netlabel_addrlist.h
@@ -0,0 +1,189 @@
+/*
+ * NetLabel Network Address Lists
+ *
+ * This file contains network address list functions used to manage ordered
+ * lists of network addresses for use by the NetLabel subsystem.  The NetLabel
+ * system manages static and dynamic label mappings for network protocols such
+ * as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul.moore@hp.com>
+ *
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2008
+ *
+ * This program is free software;  you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY;  without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program;  if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ */
+
+#ifndef _NETLABEL_ADDRLIST_H
+#define _NETLABEL_ADDRLIST_H
+
+#include <linux/types.h>
+#include <linux/rcupdate.h>
+#include <linux/list.h>
+#include <linux/in6.h>
+#include <linux/audit.h>
+
+/**
+ * struct netlbl_af4list - NetLabel IPv4 address list
+ * @addr: IPv4 address
+ * @mask: IPv4 address mask
+ * @valid: valid flag
+ * @list: list structure, used internally
+ */
+struct netlbl_af4list {
+	__be32 addr;
+	__be32 mask;
+
+	u32 valid;
+	struct list_head list;
+};
+
+/**
+ * struct netlbl_af6list - NetLabel IPv6 address list
+ * @addr: IPv6 address
+ * @mask: IPv6 address mask
+ * @valid: valid flag
+ * @list: list structure, used internally
+ */
+struct netlbl_af6list {
+	struct in6_addr addr;
+	struct in6_addr mask;
+
+	u32 valid;
+	struct list_head list;
+};
+
+#define __af4list_entry(ptr) container_of(ptr, struct netlbl_af4list, list)
+
+static inline struct netlbl_af4list *__af4list_valid(struct list_head *s,
+						     struct list_head *h)
+{
+	struct list_head *i = s;
+	struct netlbl_af4list *n = __af4list_entry(s);
+	while (i != h && !n->valid) {
+		i = i->next;
+		n = __af4list_entry(i);
+	}
+	return n;
+}
+
+static inline struct netlbl_af4list *__af4list_valid_rcu(struct list_head *s,
+							 struct list_head *h)
+{
+	struct list_head *i = s;
+	struct netlbl_af4list *n = __af4list_entry(s);
+	while (i != h && !n->valid) {
+		i = rcu_dereference(i->next);
+		n = __af4list_entry(i);
+	}
+	return n;
+}
+
+#define netlbl_af4list_foreach(iter, head)				\
+	for (iter = __af4list_valid((head)->next, head);		\
+	     prefetch(iter->list.next), &iter->list != (head);		\
+	     iter = __af4list_valid(iter->list.next, head))
+
+#define netlbl_af4list_foreach_rcu(iter, head)				\
+	for (iter = __af4list_valid_rcu((head)->next, head);		\
+	     prefetch(iter->list.next),	&iter->list != (head);		\
+	     iter = __af4list_valid_rcu(iter->list.next, head))
+
+#define netlbl_af4list_foreach_safe(iter, tmp, head)			\
+	for (iter = __af4list_valid((head)->next, head),		\
+		     tmp = __af4list_valid(iter->list.next, head);	\
+	     &iter->list != (head);					\
+	     iter = tmp, tmp = __af4list_valid(iter->list.next, head))
+
+int netlbl_af4list_add(struct netlbl_af4list *entry,
+		       struct list_head *head);
+struct netlbl_af4list *netlbl_af4list_remove(__be32 addr, __be32 mask,
+					     struct list_head *head);
+void netlbl_af4list_remove_entry(struct netlbl_af4list *entry);
+struct netlbl_af4list *netlbl_af4list_search(__be32 addr,
+					     struct list_head *head);
+struct netlbl_af4list *netlbl_af4list_search_exact(__be32 addr,
+						   __be32 mask,
+						   struct list_head *head);
+void netlbl_af4list_audit_addr(struct audit_buffer *audit_buf,
+			       int src, const char *dev,
+			       __be32 addr, __be32 mask);
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+
+#define __af6list_entry(ptr) container_of(ptr, struct netlbl_af6list, list)
+
+static inline struct netlbl_af6list *__af6list_valid(struct list_head *s,
+						     struct list_head *h)
+{
+	struct list_head *i = s;
+	struct netlbl_af6list *n = __af6list_entry(s);
+	while (i != h && !n->valid) {
+		i = i->next;
+		n = __af6list_entry(i);
+	}
+	return n;
+}
+
+static inline struct netlbl_af6list *__af6list_valid_rcu(struct list_head *s,
+							 struct list_head *h)
+{
+	struct list_head *i = s;
+	struct netlbl_af6list *n = __af6list_entry(s);
+	while (i != h && !n->valid) {
+		i = rcu_dereference(i->next);
+		n = __af6list_entry(i);
+	}
+	return n;
+}
+
+#define netlbl_af6list_foreach(iter, head)				\
+	for (iter = __af6list_valid((head)->next, head);		\
+	     prefetch(iter->list.next),	&iter->list != (head);		\
+	     iter = __af6list_valid(iter->list.next, head))
+
+#define netlbl_af6list_foreach_rcu(iter, head)				\
+	for (iter = __af6list_valid_rcu((head)->next, head);		\
+	     prefetch(iter->list.next),	&iter->list != (head);		\
+	     iter = __af6list_valid_rcu(iter->list.next, head))
+
+#define netlbl_af6list_foreach_safe(iter, tmp, head)			\
+	for (iter = __af6list_valid((head)->next, head),		\
+		     tmp = __af6list_valid(iter->list.next, head);	\
+	     &iter->list != (head);					\
+	     iter = tmp, tmp = __af6list_valid(iter->list.next, head))
+
+int netlbl_af6list_add(struct netlbl_af6list *entry,
+		       struct list_head *head);
+struct netlbl_af6list *netlbl_af6list_remove(const struct in6_addr *addr,
+					     const struct in6_addr *mask,
+					     struct list_head *head);
+void netlbl_af6list_remove_entry(struct netlbl_af6list *entry);
+struct netlbl_af6list *netlbl_af6list_search(const struct in6_addr *addr,
+					     struct list_head *head);
+struct netlbl_af6list *netlbl_af6list_search_exact(const struct in6_addr *addr,
+						   const struct in6_addr *mask,
+						   struct list_head *head);
+void netlbl_af6list_audit_addr(struct audit_buffer *audit_buf,
+			       int src,
+			       const char *dev,
+			       const struct in6_addr *addr,
+			       const struct in6_addr *mask);
+#endif /* IPV6 */
+
+#endif
diff --git a/net/netlabel/netlabel_cipso_v4.c b/net/netlabel/netlabel_cipso_v4.c
index 0aec318bf0ef..fff32b70efa9 100644
--- a/net/netlabel/netlabel_cipso_v4.c
+++ b/net/netlabel/netlabel_cipso_v4.c
@@ -43,6 +43,7 @@
 #include "netlabel_user.h"
 #include "netlabel_cipso_v4.h"
 #include "netlabel_mgmt.h"
+#include "netlabel_domainhash.h"
 
 /* Argument struct for cipso_v4_doi_walk() */
 struct netlbl_cipsov4_doiwalk_arg {
@@ -51,6 +52,12 @@ struct netlbl_cipsov4_doiwalk_arg {
 	u32 seq;
 };
 
+/* Argument struct for netlbl_domhsh_walk() */
+struct netlbl_domhsh_walk_arg {
+	struct netlbl_audit *audit_info;
+	u32 doi;
+};
+
 /* NetLabel Generic NETLINK CIPSOv4 family */
 static struct genl_family netlbl_cipsov4_gnl_family = {
 	.id = GENL_ID_GENERATE,
@@ -81,32 +88,6 @@ static const struct nla_policy netlbl_cipsov4_genl_policy[NLBL_CIPSOV4_A_MAX + 1
  */
 
 /**
- * netlbl_cipsov4_doi_free - Frees a CIPSO V4 DOI definition
- * @entry: the entry's RCU field
- *
- * Description:
- * This function is designed to be used as a callback to the call_rcu()
- * function so that the memory allocated to the DOI definition can be released
- * safely.
- *
- */
-void netlbl_cipsov4_doi_free(struct rcu_head *entry)
-{
-	struct cipso_v4_doi *ptr;
-
-	ptr = container_of(entry, struct cipso_v4_doi, rcu);
-	switch (ptr->type) {
-	case CIPSO_V4_MAP_STD:
-		kfree(ptr->map.std->lvl.cipso);
-		kfree(ptr->map.std->lvl.local);
-		kfree(ptr->map.std->cat.cipso);
-		kfree(ptr->map.std->cat.local);
-		break;
-	}
-	kfree(ptr);
-}
-
-/**
  * netlbl_cipsov4_add_common - Parse the common sections of a ADD message
  * @info: the Generic NETLINK info block
  * @doi_def: the CIPSO V4 DOI definition
@@ -151,9 +132,9 @@ static int netlbl_cipsov4_add_common(struct genl_info *info,
  * @info: the Generic NETLINK info block
  *
  * Description:
- * Create a new CIPSO_V4_MAP_STD DOI definition based on the given ADD message
- * and add it to the CIPSO V4 engine.  Return zero on success and non-zero on
- * error.
+ * Create a new CIPSO_V4_MAP_TRANS DOI definition based on the given ADD
+ * message and add it to the CIPSO V4 engine.  Return zero on success and
+ * non-zero on error.
  *
  */
 static int netlbl_cipsov4_add_std(struct genl_info *info)
@@ -183,7 +164,7 @@ static int netlbl_cipsov4_add_std(struct genl_info *info)
 		ret_val = -ENOMEM;
 		goto add_std_failure;
 	}
-	doi_def->type = CIPSO_V4_MAP_STD;
+	doi_def->type = CIPSO_V4_MAP_TRANS;
 
 	ret_val = netlbl_cipsov4_add_common(info, doi_def);
 	if (ret_val != 0)
@@ -342,7 +323,7 @@ static int netlbl_cipsov4_add_std(struct genl_info *info)
 
 add_std_failure:
 	if (doi_def)
-		netlbl_cipsov4_doi_free(&doi_def->rcu);
+		cipso_v4_doi_free(doi_def);
 	return ret_val;
 }
 
@@ -379,7 +360,44 @@ static int netlbl_cipsov4_add_pass(struct genl_info *info)
 	return 0;
 
 add_pass_failure:
-	netlbl_cipsov4_doi_free(&doi_def->rcu);
+	cipso_v4_doi_free(doi_def);
+	return ret_val;
+}
+
+/**
+ * netlbl_cipsov4_add_local - Adds a CIPSO V4 DOI definition
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Create a new CIPSO_V4_MAP_LOCAL DOI definition based on the given ADD
+ * message and add it to the CIPSO V4 engine.  Return zero on success and
+ * non-zero on error.
+ *
+ */
+static int netlbl_cipsov4_add_local(struct genl_info *info)
+{
+	int ret_val;
+	struct cipso_v4_doi *doi_def = NULL;
+
+	if (!info->attrs[NLBL_CIPSOV4_A_TAGLST])
+		return -EINVAL;
+
+	doi_def = kmalloc(sizeof(*doi_def), GFP_KERNEL);
+	if (doi_def == NULL)
+		return -ENOMEM;
+	doi_def->type = CIPSO_V4_MAP_LOCAL;
+
+	ret_val = netlbl_cipsov4_add_common(info, doi_def);
+	if (ret_val != 0)
+		goto add_local_failure;
+
+	ret_val = cipso_v4_doi_add(doi_def);
+	if (ret_val != 0)
+		goto add_local_failure;
+	return 0;
+
+add_local_failure:
+	cipso_v4_doi_free(doi_def);
 	return ret_val;
 }
 
@@ -412,14 +430,18 @@ static int netlbl_cipsov4_add(struct sk_buff *skb, struct genl_info *info)
 
 	type = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_MTYPE]);
 	switch (type) {
-	case CIPSO_V4_MAP_STD:
-		type_str = "std";
+	case CIPSO_V4_MAP_TRANS:
+		type_str = "trans";
 		ret_val = netlbl_cipsov4_add_std(info);
 		break;
 	case CIPSO_V4_MAP_PASS:
 		type_str = "pass";
 		ret_val = netlbl_cipsov4_add_pass(info);
 		break;
+	case CIPSO_V4_MAP_LOCAL:
+		type_str = "local";
+		ret_val = netlbl_cipsov4_add_local(info);
+		break;
 	}
 	if (ret_val == 0)
 		atomic_inc(&netlabel_mgmt_protocount);
@@ -491,7 +513,7 @@ list_start:
 	doi_def = cipso_v4_doi_getdef(doi);
 	if (doi_def == NULL) {
 		ret_val = -EINVAL;
-		goto list_failure;
+		goto list_failure_lock;
 	}
 
 	ret_val = nla_put_u32(ans_skb, NLBL_CIPSOV4_A_MTYPE, doi_def->type);
@@ -516,7 +538,7 @@ list_start:
 	nla_nest_end(ans_skb, nla_a);
 
 	switch (doi_def->type) {
-	case CIPSO_V4_MAP_STD:
+	case CIPSO_V4_MAP_TRANS:
 		nla_a = nla_nest_start(ans_skb, NLBL_CIPSOV4_A_MLSLVLLST);
 		if (nla_a == NULL) {
 			ret_val = -ENOMEM;
@@ -655,7 +677,7 @@ static int netlbl_cipsov4_listall(struct sk_buff *skb,
 				  struct netlink_callback *cb)
 {
 	struct netlbl_cipsov4_doiwalk_arg cb_arg;
-	int doi_skip = cb->args[0];
+	u32 doi_skip = cb->args[0];
 
 	cb_arg.nl_cb = cb;
 	cb_arg.skb = skb;
@@ -668,6 +690,29 @@ static int netlbl_cipsov4_listall(struct sk_buff *skb,
 }
 
 /**
+ * netlbl_cipsov4_remove_cb - netlbl_cipsov4_remove() callback for REMOVE
+ * @entry: LSM domain mapping entry
+ * @arg: the netlbl_domhsh_walk_arg structure
+ *
+ * Description:
+ * This function is intended for use by netlbl_cipsov4_remove() as the callback
+ * for the netlbl_domhsh_walk() function; it removes LSM domain map entries
+ * which are associated with the CIPSO DOI specified in @arg.  Returns zero on
+ * success, negative values on failure.
+ *
+ */
+static int netlbl_cipsov4_remove_cb(struct netlbl_dom_map *entry, void *arg)
+{
+	struct netlbl_domhsh_walk_arg *cb_arg = arg;
+
+	if (entry->type == NETLBL_NLTYPE_CIPSOV4 &&
+	    entry->type_def.cipsov4->doi == cb_arg->doi)
+		return netlbl_domhsh_remove_entry(entry, cb_arg->audit_info);
+
+	return 0;
+}
+
+/**
  * netlbl_cipsov4_remove - Handle a REMOVE message
  * @skb: the NETLINK buffer
  * @info: the Generic NETLINK info block
@@ -681,8 +726,11 @@ static int netlbl_cipsov4_remove(struct sk_buff *skb, struct genl_info *info)
 {
 	int ret_val = -EINVAL;
 	u32 doi = 0;
+	struct netlbl_domhsh_walk_arg cb_arg;
 	struct audit_buffer *audit_buf;
 	struct netlbl_audit audit_info;
+	u32 skip_bkt = 0;
+	u32 skip_chain = 0;
 
 	if (!info->attrs[NLBL_CIPSOV4_A_DOI])
 		return -EINVAL;
@@ -690,11 +738,15 @@ static int netlbl_cipsov4_remove(struct sk_buff *skb, struct genl_info *info)
 	doi = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_DOI]);
 	netlbl_netlink_auditinfo(skb, &audit_info);
 
-	ret_val = cipso_v4_doi_remove(doi,
-				      &audit_info,
-				      netlbl_cipsov4_doi_free);
-	if (ret_val == 0)
-		atomic_dec(&netlabel_mgmt_protocount);
+	cb_arg.doi = doi;
+	cb_arg.audit_info = &audit_info;
+	ret_val = netlbl_domhsh_walk(&skip_bkt, &skip_chain,
+				     netlbl_cipsov4_remove_cb, &cb_arg);
+	if (ret_val == 0 || ret_val == -ENOENT) {
+		ret_val = cipso_v4_doi_remove(doi, &audit_info);
+		if (ret_val == 0)
+			atomic_dec(&netlabel_mgmt_protocount);
+	}
 
 	audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_DEL,
 					      &audit_info);
diff --git a/net/netlabel/netlabel_cipso_v4.h b/net/netlabel/netlabel_cipso_v4.h
index 220cb9d06b49..c8a4079261f0 100644
--- a/net/netlabel/netlabel_cipso_v4.h
+++ b/net/netlabel/netlabel_cipso_v4.h
@@ -45,12 +45,13 @@
  *     NLBL_CIPSOV4_A_MTYPE
  *     NLBL_CIPSOV4_A_TAGLST
  *
- *   If using CIPSO_V4_MAP_STD the following attributes are required:
+ *   If using CIPSO_V4_MAP_TRANS the following attributes are required:
  *
  *     NLBL_CIPSOV4_A_MLSLVLLST
  *     NLBL_CIPSOV4_A_MLSCATLST
  *
- *   If using CIPSO_V4_MAP_PASS no additional attributes are required.
+ *   If using CIPSO_V4_MAP_PASS or CIPSO_V4_MAP_LOCAL no additional attributes
+ *   are required.
  *
  * o REMOVE:
  *   Sent by an application to remove a specific DOI mapping table from the
@@ -76,12 +77,13 @@
  *     NLBL_CIPSOV4_A_MTYPE
  *     NLBL_CIPSOV4_A_TAGLST
  *
- *   If using CIPSO_V4_MAP_STD the following attributes are required:
+ *   If using CIPSO_V4_MAP_TRANS the following attributes are required:
  *
  *     NLBL_CIPSOV4_A_MLSLVLLST
  *     NLBL_CIPSOV4_A_MLSCATLST
  *
- *   If using CIPSO_V4_MAP_PASS no additional attributes are required.
+ *   If using CIPSO_V4_MAP_PASS or CIPSO_V4_MAP_LOCAL no additional attributes
+ *   are required.
  *
  * o LISTALL:
  *   This message is sent by an application to list the valid DOIs on the
diff --git a/net/netlabel/netlabel_domainhash.c b/net/netlabel/netlabel_domainhash.c
index 643c032a3a57..5fadf10e5ddf 100644
--- a/net/netlabel/netlabel_domainhash.c
+++ b/net/netlabel/netlabel_domainhash.c
@@ -11,7 +11,7 @@
  */
 
 /*
- * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006, 2008
  *
  * This program is free software;  you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -40,6 +40,7 @@
 #include <asm/bug.h>
 
 #include "netlabel_mgmt.h"
+#include "netlabel_addrlist.h"
 #include "netlabel_domainhash.h"
 #include "netlabel_user.h"
 
@@ -72,8 +73,28 @@ static struct netlbl_dom_map *netlbl_domhsh_def = NULL;
 static void netlbl_domhsh_free_entry(struct rcu_head *entry)
 {
 	struct netlbl_dom_map *ptr;
+	struct netlbl_af4list *iter4;
+	struct netlbl_af4list *tmp4;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	struct netlbl_af6list *iter6;
+	struct netlbl_af6list *tmp6;
+#endif /* IPv6 */
 
 	ptr = container_of(entry, struct netlbl_dom_map, rcu);
+	if (ptr->type == NETLBL_NLTYPE_ADDRSELECT) {
+		netlbl_af4list_foreach_safe(iter4, tmp4,
+					    &ptr->type_def.addrsel->list4) {
+			netlbl_af4list_remove_entry(iter4);
+			kfree(netlbl_domhsh_addr4_entry(iter4));
+		}
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+		netlbl_af6list_foreach_safe(iter6, tmp6,
+					    &ptr->type_def.addrsel->list6) {
+			netlbl_af6list_remove_entry(iter6);
+			kfree(netlbl_domhsh_addr6_entry(iter6));
+		}
+#endif /* IPv6 */
+	}
 	kfree(ptr->domain);
 	kfree(ptr);
 }
@@ -115,13 +136,13 @@ static u32 netlbl_domhsh_hash(const char *key)
 static struct netlbl_dom_map *netlbl_domhsh_search(const char *domain)
 {
 	u32 bkt;
+	struct list_head *bkt_list;
 	struct netlbl_dom_map *iter;
 
 	if (domain != NULL) {
 		bkt = netlbl_domhsh_hash(domain);
-		list_for_each_entry_rcu(iter,
-				     &rcu_dereference(netlbl_domhsh)->tbl[bkt],
-				     list)
+		bkt_list = &rcu_dereference(netlbl_domhsh)->tbl[bkt];
+		list_for_each_entry_rcu(iter, bkt_list, list)
 			if (iter->valid && strcmp(iter->domain, domain) == 0)
 				return iter;
 	}
@@ -156,6 +177,69 @@ static struct netlbl_dom_map *netlbl_domhsh_search_def(const char *domain)
 	return entry;
 }
 
+/**
+ * netlbl_domhsh_audit_add - Generate an audit entry for an add event
+ * @entry: the entry being added
+ * @addr4: the IPv4 address information
+ * @addr6: the IPv6 address information
+ * @result: the result code
+ * @audit_info: NetLabel audit information
+ *
+ * Description:
+ * Generate an audit record for adding a new NetLabel/LSM mapping entry with
+ * the given information.  Caller is responsibile for holding the necessary
+ * locks.
+ *
+ */
+static void netlbl_domhsh_audit_add(struct netlbl_dom_map *entry,
+				    struct netlbl_af4list *addr4,
+				    struct netlbl_af6list *addr6,
+				    int result,
+				    struct netlbl_audit *audit_info)
+{
+	struct audit_buffer *audit_buf;
+	struct cipso_v4_doi *cipsov4 = NULL;
+	u32 type;
+
+	audit_buf = netlbl_audit_start_common(AUDIT_MAC_MAP_ADD, audit_info);
+	if (audit_buf != NULL) {
+		audit_log_format(audit_buf, " nlbl_domain=%s",
+				 entry->domain ? entry->domain : "(default)");
+		if (addr4 != NULL) {
+			struct netlbl_domaddr4_map *map4;
+			map4 = netlbl_domhsh_addr4_entry(addr4);
+			type = map4->type;
+			cipsov4 = map4->type_def.cipsov4;
+			netlbl_af4list_audit_addr(audit_buf, 0, NULL,
+						  addr4->addr, addr4->mask);
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+		} else if (addr6 != NULL) {
+			struct netlbl_domaddr6_map *map6;
+			map6 = netlbl_domhsh_addr6_entry(addr6);
+			type = map6->type;
+			netlbl_af6list_audit_addr(audit_buf, 0, NULL,
+						  &addr6->addr, &addr6->mask);
+#endif /* IPv6 */
+		} else {
+			type = entry->type;
+			cipsov4 = entry->type_def.cipsov4;
+		}
+		switch (type) {
+		case NETLBL_NLTYPE_UNLABELED:
+			audit_log_format(audit_buf, " nlbl_protocol=unlbl");
+			break;
+		case NETLBL_NLTYPE_CIPSOV4:
+			BUG_ON(cipsov4 == NULL);
+			audit_log_format(audit_buf,
+					 " nlbl_protocol=cipsov4 cipso_doi=%u",
+					 cipsov4->doi);
+			break;
+		}
+		audit_log_format(audit_buf, " res=%u", result == 0 ? 1 : 0);
+		audit_log_end(audit_buf);
+	}
+}
+
 /*
  * Domain Hash Table Functions
  */
@@ -213,74 +297,106 @@ int __init netlbl_domhsh_init(u32 size)
 int netlbl_domhsh_add(struct netlbl_dom_map *entry,
 		      struct netlbl_audit *audit_info)
 {
-	int ret_val;
-	u32 bkt;
-	struct audit_buffer *audit_buf;
-
-	switch (entry->type) {
-	case NETLBL_NLTYPE_UNLABELED:
-		ret_val = 0;
-		break;
-	case NETLBL_NLTYPE_CIPSOV4:
-		ret_val = cipso_v4_doi_domhsh_add(entry->type_def.cipsov4,
-						  entry->domain);
-		break;
-	default:
-		return -EINVAL;
-	}
-	if (ret_val != 0)
-		return ret_val;
-
-	entry->valid = 1;
-	INIT_RCU_HEAD(&entry->rcu);
+	int ret_val = 0;
+	struct netlbl_dom_map *entry_old;
+	struct netlbl_af4list *iter4;
+	struct netlbl_af4list *tmp4;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	struct netlbl_af6list *iter6;
+	struct netlbl_af6list *tmp6;
+#endif /* IPv6 */
 
 	rcu_read_lock();
+
 	spin_lock(&netlbl_domhsh_lock);
-	if (entry->domain != NULL) {
-		bkt = netlbl_domhsh_hash(entry->domain);
-		if (netlbl_domhsh_search(entry->domain) == NULL)
+	if (entry->domain != NULL)
+		entry_old = netlbl_domhsh_search(entry->domain);
+	else
+		entry_old = netlbl_domhsh_search_def(entry->domain);
+	if (entry_old == NULL) {
+		entry->valid = 1;
+		INIT_RCU_HEAD(&entry->rcu);
+
+		if (entry->domain != NULL) {
+			u32 bkt = netlbl_domhsh_hash(entry->domain);
 			list_add_tail_rcu(&entry->list,
 				    &rcu_dereference(netlbl_domhsh)->tbl[bkt]);
-		else
-			ret_val = -EEXIST;
-	} else {
-		INIT_LIST_HEAD(&entry->list);
-		if (rcu_dereference(netlbl_domhsh_def) == NULL)
+		} else {
+			INIT_LIST_HEAD(&entry->list);
 			rcu_assign_pointer(netlbl_domhsh_def, entry);
-		else
-			ret_val = -EEXIST;
-	}
-	spin_unlock(&netlbl_domhsh_lock);
-	audit_buf = netlbl_audit_start_common(AUDIT_MAC_MAP_ADD, audit_info);
-	if (audit_buf != NULL) {
-		audit_log_format(audit_buf,
-				 " nlbl_domain=%s",
-				 entry->domain ? entry->domain : "(default)");
-		switch (entry->type) {
-		case NETLBL_NLTYPE_UNLABELED:
-			audit_log_format(audit_buf, " nlbl_protocol=unlbl");
-			break;
-		case NETLBL_NLTYPE_CIPSOV4:
-			audit_log_format(audit_buf,
-					 " nlbl_protocol=cipsov4 cipso_doi=%u",
-					 entry->type_def.cipsov4->doi);
-			break;
 		}
-		audit_log_format(audit_buf, " res=%u", ret_val == 0 ? 1 : 0);
-		audit_log_end(audit_buf);
-	}
-	rcu_read_unlock();
 
-	if (ret_val != 0) {
-		switch (entry->type) {
-		case NETLBL_NLTYPE_CIPSOV4:
-			if (cipso_v4_doi_domhsh_remove(entry->type_def.cipsov4,
-						       entry->domain) != 0)
-				BUG();
-			break;
+		if (entry->type == NETLBL_NLTYPE_ADDRSELECT) {
+			netlbl_af4list_foreach_rcu(iter4,
+					       &entry->type_def.addrsel->list4)
+				netlbl_domhsh_audit_add(entry, iter4, NULL,
+							ret_val, audit_info);
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+			netlbl_af6list_foreach_rcu(iter6,
+					       &entry->type_def.addrsel->list6)
+				netlbl_domhsh_audit_add(entry, NULL, iter6,
+							ret_val, audit_info);
+#endif /* IPv6 */
+		} else
+			netlbl_domhsh_audit_add(entry, NULL, NULL,
+						ret_val, audit_info);
+	} else if (entry_old->type == NETLBL_NLTYPE_ADDRSELECT &&
+		   entry->type == NETLBL_NLTYPE_ADDRSELECT) {
+		struct list_head *old_list4;
+		struct list_head *old_list6;
+
+		old_list4 = &entry_old->type_def.addrsel->list4;
+		old_list6 = &entry_old->type_def.addrsel->list6;
+
+		/* we only allow the addition of address selectors if all of
+		 * the selectors do not exist in the existing domain map */
+		netlbl_af4list_foreach_rcu(iter4,
+					   &entry->type_def.addrsel->list4)
+			if (netlbl_af4list_search_exact(iter4->addr,
+							iter4->mask,
+							old_list4)) {
+				ret_val = -EEXIST;
+				goto add_return;
+			}
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+		netlbl_af6list_foreach_rcu(iter6,
+					   &entry->type_def.addrsel->list6)
+			if (netlbl_af6list_search_exact(&iter6->addr,
+							&iter6->mask,
+							old_list6)) {
+				ret_val = -EEXIST;
+				goto add_return;
+			}
+#endif /* IPv6 */
+
+		netlbl_af4list_foreach_safe(iter4, tmp4,
+					    &entry->type_def.addrsel->list4) {
+			netlbl_af4list_remove_entry(iter4);
+			iter4->valid = 1;
+			ret_val = netlbl_af4list_add(iter4, old_list4);
+			netlbl_domhsh_audit_add(entry_old, iter4, NULL,
+						ret_val, audit_info);
+			if (ret_val != 0)
+				goto add_return;
 		}
-	}
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+		netlbl_af6list_foreach_safe(iter6, tmp6,
+					    &entry->type_def.addrsel->list6) {
+			netlbl_af6list_remove_entry(iter6);
+			iter6->valid = 1;
+			ret_val = netlbl_af6list_add(iter6, old_list6);
+			netlbl_domhsh_audit_add(entry_old, NULL, iter6,
+						ret_val, audit_info);
+			if (ret_val != 0)
+				goto add_return;
+		}
+#endif /* IPv6 */
+	} else
+		ret_val = -EINVAL;
 
+add_return:
+	spin_unlock(&netlbl_domhsh_lock);
+	rcu_read_unlock();
 	return ret_val;
 }
 
@@ -302,35 +418,26 @@ int netlbl_domhsh_add_default(struct netlbl_dom_map *entry,
 }
 
 /**
- * netlbl_domhsh_remove - Removes an entry from the domain hash table
- * @domain: the domain to remove
+ * netlbl_domhsh_remove_entry - Removes a given entry from the domain table
+ * @entry: the entry to remove
  * @audit_info: NetLabel audit information
  *
  * Description:
  * Removes an entry from the domain hash table and handles any updates to the
- * lower level protocol handler (i.e. CIPSO).  Returns zero on success,
- * negative on failure.
+ * lower level protocol handler (i.e. CIPSO).  Caller is responsible for
+ * ensuring that the RCU read lock is held.  Returns zero on success, negative
+ * on failure.
  *
  */
-int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info)
+int netlbl_domhsh_remove_entry(struct netlbl_dom_map *entry,
+			       struct netlbl_audit *audit_info)
 {
-	int ret_val = -ENOENT;
-	struct netlbl_dom_map *entry;
+	int ret_val = 0;
 	struct audit_buffer *audit_buf;
 
-	rcu_read_lock();
-	if (domain)
-		entry = netlbl_domhsh_search(domain);
-	else
-		entry = netlbl_domhsh_search_def(domain);
 	if (entry == NULL)
-		goto remove_return;
-	switch (entry->type) {
-	case NETLBL_NLTYPE_CIPSOV4:
-		cipso_v4_doi_domhsh_remove(entry->type_def.cipsov4,
-					   entry->domain);
-		break;
-	}
+		return -ENOENT;
+
 	spin_lock(&netlbl_domhsh_lock);
 	if (entry->valid) {
 		entry->valid = 0;
@@ -338,8 +445,8 @@ int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info)
 			list_del_rcu(&entry->list);
 		else
 			rcu_assign_pointer(netlbl_domhsh_def, NULL);
-		ret_val = 0;
-	}
+	} else
+		ret_val = -ENOENT;
 	spin_unlock(&netlbl_domhsh_lock);
 
 	audit_buf = netlbl_audit_start_common(AUDIT_MAC_MAP_DEL, audit_info);
@@ -351,10 +458,54 @@ int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info)
 		audit_log_end(audit_buf);
 	}
 
-remove_return:
-	rcu_read_unlock();
-	if (ret_val == 0)
+	if (ret_val == 0) {
+		struct netlbl_af4list *iter4;
+		struct netlbl_domaddr4_map *map4;
+
+		switch (entry->type) {
+		case NETLBL_NLTYPE_ADDRSELECT:
+			netlbl_af4list_foreach_rcu(iter4,
+					     &entry->type_def.addrsel->list4) {
+				map4 = netlbl_domhsh_addr4_entry(iter4);
+				cipso_v4_doi_putdef(map4->type_def.cipsov4);
+			}
+			/* no need to check the IPv6 list since we currently
+			 * support only unlabeled protocols for IPv6 */
+			break;
+		case NETLBL_NLTYPE_CIPSOV4:
+			cipso_v4_doi_putdef(entry->type_def.cipsov4);
+			break;
+		}
 		call_rcu(&entry->rcu, netlbl_domhsh_free_entry);
+	}
+
+	return ret_val;
+}
+
+/**
+ * netlbl_domhsh_remove - Removes an entry from the domain hash table
+ * @domain: the domain to remove
+ * @audit_info: NetLabel audit information
+ *
+ * Description:
+ * Removes an entry from the domain hash table and handles any updates to the
+ * lower level protocol handler (i.e. CIPSO).  Returns zero on success,
+ * negative on failure.
+ *
+ */
+int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info)
+{
+	int ret_val;
+	struct netlbl_dom_map *entry;
+
+	rcu_read_lock();
+	if (domain)
+		entry = netlbl_domhsh_search(domain);
+	else
+		entry = netlbl_domhsh_search_def(domain);
+	ret_val = netlbl_domhsh_remove_entry(entry, audit_info);
+	rcu_read_unlock();
+
 	return ret_val;
 }
 
@@ -389,6 +540,70 @@ struct netlbl_dom_map *netlbl_domhsh_getentry(const char *domain)
 }
 
 /**
+ * netlbl_domhsh_getentry_af4 - Get an entry from the domain hash table
+ * @domain: the domain name to search for
+ * @addr: the IP address to search for
+ *
+ * Description:
+ * Look through the domain hash table searching for an entry to match @domain
+ * and @addr, return a pointer to a copy of the entry or NULL.  The caller is
+ * responsible for ensuring that rcu_read_[un]lock() is called.
+ *
+ */
+struct netlbl_domaddr4_map *netlbl_domhsh_getentry_af4(const char *domain,
+						       __be32 addr)
+{
+	struct netlbl_dom_map *dom_iter;
+	struct netlbl_af4list *addr_iter;
+
+	dom_iter = netlbl_domhsh_search_def(domain);
+	if (dom_iter == NULL)
+		return NULL;
+	if (dom_iter->type != NETLBL_NLTYPE_ADDRSELECT)
+		return NULL;
+
+	addr_iter = netlbl_af4list_search(addr,
+					  &dom_iter->type_def.addrsel->list4);
+	if (addr_iter == NULL)
+		return NULL;
+
+	return netlbl_domhsh_addr4_entry(addr_iter);
+}
+
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+/**
+ * netlbl_domhsh_getentry_af6 - Get an entry from the domain hash table
+ * @domain: the domain name to search for
+ * @addr: the IP address to search for
+ *
+ * Description:
+ * Look through the domain hash table searching for an entry to match @domain
+ * and @addr, return a pointer to a copy of the entry or NULL.  The caller is
+ * responsible for ensuring that rcu_read_[un]lock() is called.
+ *
+ */
+struct netlbl_domaddr6_map *netlbl_domhsh_getentry_af6(const char *domain,
+						   const struct in6_addr *addr)
+{
+	struct netlbl_dom_map *dom_iter;
+	struct netlbl_af6list *addr_iter;
+
+	dom_iter = netlbl_domhsh_search_def(domain);
+	if (dom_iter == NULL)
+		return NULL;
+	if (dom_iter->type != NETLBL_NLTYPE_ADDRSELECT)
+		return NULL;
+
+	addr_iter = netlbl_af6list_search(addr,
+					  &dom_iter->type_def.addrsel->list6);
+	if (addr_iter == NULL)
+		return NULL;
+
+	return netlbl_domhsh_addr6_entry(addr_iter);
+}
+#endif /* IPv6 */
+
+/**
  * netlbl_domhsh_walk - Iterate through the domain mapping hash table
  * @skip_bkt: the number of buckets to skip at the start
  * @skip_chain: the number of entries to skip in the first iterated bucket
@@ -410,6 +625,7 @@ int netlbl_domhsh_walk(u32 *skip_bkt,
 {
 	int ret_val = -ENOENT;
 	u32 iter_bkt;
+	struct list_head *iter_list;
 	struct netlbl_dom_map *iter_entry;
 	u32 chain_cnt = 0;
 
@@ -417,9 +633,8 @@ int netlbl_domhsh_walk(u32 *skip_bkt,
 	for (iter_bkt = *skip_bkt;
 	     iter_bkt < rcu_dereference(netlbl_domhsh)->size;
 	     iter_bkt++, chain_cnt = 0) {
-		list_for_each_entry_rcu(iter_entry,
-				&rcu_dereference(netlbl_domhsh)->tbl[iter_bkt],
-				list)
+		iter_list = &rcu_dereference(netlbl_domhsh)->tbl[iter_bkt];
+		list_for_each_entry_rcu(iter_entry, iter_list, list)
 			if (iter_entry->valid) {
 				if (chain_cnt++ < *skip_chain)
 					continue;
diff --git a/net/netlabel/netlabel_domainhash.h b/net/netlabel/netlabel_domainhash.h
index 8220990ceb96..bfcb6763a1a1 100644
--- a/net/netlabel/netlabel_domainhash.h
+++ b/net/netlabel/netlabel_domainhash.h
@@ -11,7 +11,7 @@
  */
 
 /*
- * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006, 2008
  *
  * This program is free software;  you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -36,16 +36,43 @@
 #include <linux/rcupdate.h>
 #include <linux/list.h>
 
+#include "netlabel_addrlist.h"
+
 /* Domain hash table size */
 /* XXX - currently this number is an uneducated guess */
 #define NETLBL_DOMHSH_BITSIZE       7
 
-/* Domain mapping definition struct */
+/* Domain mapping definition structures */
+#define netlbl_domhsh_addr4_entry(iter) \
+	container_of(iter, struct netlbl_domaddr4_map, list)
+struct netlbl_domaddr4_map {
+	u32 type;
+	union {
+		struct cipso_v4_doi *cipsov4;
+	} type_def;
+
+	struct netlbl_af4list list;
+};
+#define netlbl_domhsh_addr6_entry(iter) \
+	container_of(iter, struct netlbl_domaddr6_map, list)
+struct netlbl_domaddr6_map {
+	u32 type;
+
+	/* NOTE: no 'type_def' union needed at present since we don't currently
+	 *       support any IPv6 labeling protocols */
+
+	struct netlbl_af6list list;
+};
+struct netlbl_domaddr_map {
+	struct list_head list4;
+	struct list_head list6;
+};
 struct netlbl_dom_map {
 	char *domain;
 	u32 type;
 	union {
 		struct cipso_v4_doi *cipsov4;
+		struct netlbl_domaddr_map *addrsel;
 	} type_def;
 
 	u32 valid;
@@ -61,12 +88,21 @@ int netlbl_domhsh_add(struct netlbl_dom_map *entry,
 		      struct netlbl_audit *audit_info);
 int netlbl_domhsh_add_default(struct netlbl_dom_map *entry,
 			      struct netlbl_audit *audit_info);
+int netlbl_domhsh_remove_entry(struct netlbl_dom_map *entry,
+			       struct netlbl_audit *audit_info);
 int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info);
 int netlbl_domhsh_remove_default(struct netlbl_audit *audit_info);
 struct netlbl_dom_map *netlbl_domhsh_getentry(const char *domain);
+struct netlbl_domaddr4_map *netlbl_domhsh_getentry_af4(const char *domain,
+						       __be32 addr);
 int netlbl_domhsh_walk(u32 *skip_bkt,
 		     u32 *skip_chain,
 		     int (*callback) (struct netlbl_dom_map *entry, void *arg),
 		     void *cb_arg);
 
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+struct netlbl_domaddr6_map *netlbl_domhsh_getentry_af6(const char *domain,
+						  const struct in6_addr *addr);
+#endif /* IPv6 */
+
 #endif
diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c
index 39793a1a93aa..b32eceb3ab0d 100644
--- a/net/netlabel/netlabel_kapi.c
+++ b/net/netlabel/netlabel_kapi.c
@@ -10,7 +10,7 @@
  */
 
 /*
- * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006, 2008
  *
  * This program is free software;  you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -82,7 +82,7 @@ int netlbl_cfg_unlbl_add_map(const char *domain,
 
 	entry = kzalloc(sizeof(*entry), GFP_ATOMIC);
 	if (entry == NULL)
-		goto cfg_unlbl_add_map_failure;
+		return -ENOMEM;
 	if (domain != NULL) {
 		entry->domain = kstrdup(domain, GFP_ATOMIC);
 		if (entry->domain == NULL)
@@ -104,49 +104,6 @@ cfg_unlbl_add_map_failure:
 }
 
 /**
- * netlbl_cfg_cipsov4_add - Add a new CIPSOv4 DOI definition
- * @doi_def: the DOI definition
- * @audit_info: NetLabel audit information
- *
- * Description:
- * Add a new CIPSOv4 DOI definition to the NetLabel subsystem.  Returns zero on
- * success, negative values on failure.
- *
- */
-int netlbl_cfg_cipsov4_add(struct cipso_v4_doi *doi_def,
-			   struct netlbl_audit *audit_info)
-{
-	int ret_val;
-	const char *type_str;
-	struct audit_buffer *audit_buf;
-
-	ret_val = cipso_v4_doi_add(doi_def);
-
-	audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_ADD,
-					      audit_info);
-	if (audit_buf != NULL) {
-		switch (doi_def->type) {
-		case CIPSO_V4_MAP_STD:
-			type_str = "std";
-			break;
-		case CIPSO_V4_MAP_PASS:
-			type_str = "pass";
-			break;
-		default:
-			type_str = "(unknown)";
-		}
-		audit_log_format(audit_buf,
-				 " cipso_doi=%u cipso_type=%s res=%u",
-				 doi_def->doi,
-				 type_str,
-				 ret_val == 0 ? 1 : 0);
-		audit_log_end(audit_buf);
-	}
-
-	return ret_val;
-}
-
-/**
  * netlbl_cfg_cipsov4_add_map - Add a new CIPSOv4 DOI definition and mapping
  * @doi_def: the DOI definition
  * @domain: the domain mapping to add
@@ -164,58 +121,71 @@ int netlbl_cfg_cipsov4_add_map(struct cipso_v4_doi *doi_def,
 			       struct netlbl_audit *audit_info)
 {
 	int ret_val = -ENOMEM;
+	u32 doi;
+	u32 doi_type;
 	struct netlbl_dom_map *entry;
+	const char *type_str;
+	struct audit_buffer *audit_buf;
+
+	doi = doi_def->doi;
+	doi_type = doi_def->type;
 
 	entry = kzalloc(sizeof(*entry), GFP_ATOMIC);
 	if (entry == NULL)
-		goto cfg_cipsov4_add_map_failure;
+		return -ENOMEM;
 	if (domain != NULL) {
 		entry->domain = kstrdup(domain, GFP_ATOMIC);
 		if (entry->domain == NULL)
 			goto cfg_cipsov4_add_map_failure;
 	}
-	entry->type = NETLBL_NLTYPE_CIPSOV4;
-	entry->type_def.cipsov4 = doi_def;
-
-	/* Grab a RCU read lock here so nothing happens to the doi_def variable
-	 * between adding it to the CIPSOv4 protocol engine and adding a
-	 * domain mapping for it. */
 
-	rcu_read_lock();
-	ret_val = netlbl_cfg_cipsov4_add(doi_def, audit_info);
+	ret_val = cipso_v4_doi_add(doi_def);
 	if (ret_val != 0)
-		goto cfg_cipsov4_add_map_failure_unlock;
+		goto cfg_cipsov4_add_map_failure_remove_doi;
+	entry->type = NETLBL_NLTYPE_CIPSOV4;
+	entry->type_def.cipsov4 = cipso_v4_doi_getdef(doi);
+	if (entry->type_def.cipsov4 == NULL) {
+		ret_val = -ENOENT;
+		goto cfg_cipsov4_add_map_failure_remove_doi;
+	}
 	ret_val = netlbl_domhsh_add(entry, audit_info);
 	if (ret_val != 0)
-		goto cfg_cipsov4_add_map_failure_remove_doi;
-	rcu_read_unlock();
+		goto cfg_cipsov4_add_map_failure_release_doi;
 
-	return 0;
+cfg_cipsov4_add_map_return:
+	audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_ADD,
+					      audit_info);
+	if (audit_buf != NULL) {
+		switch (doi_type) {
+		case CIPSO_V4_MAP_TRANS:
+			type_str = "trans";
+			break;
+		case CIPSO_V4_MAP_PASS:
+			type_str = "pass";
+			break;
+		case CIPSO_V4_MAP_LOCAL:
+			type_str = "local";
+			break;
+		default:
+			type_str = "(unknown)";
+		}
+		audit_log_format(audit_buf,
+				 " cipso_doi=%u cipso_type=%s res=%u",
+				 doi, type_str, ret_val == 0 ? 1 : 0);
+		audit_log_end(audit_buf);
+	}
 
+	return ret_val;
+
+cfg_cipsov4_add_map_failure_release_doi:
+	cipso_v4_doi_putdef(doi_def);
 cfg_cipsov4_add_map_failure_remove_doi:
-	cipso_v4_doi_remove(doi_def->doi, audit_info, netlbl_cipsov4_doi_free);
-cfg_cipsov4_add_map_failure_unlock:
-	rcu_read_unlock();
+	cipso_v4_doi_remove(doi, audit_info);
 cfg_cipsov4_add_map_failure:
 	if (entry != NULL)
 		kfree(entry->domain);
 	kfree(entry);
-	return ret_val;
-}
-
-/**
- * netlbl_cfg_cipsov4_del - Removean existing CIPSOv4 DOI definition
- * @doi: the CIPSO DOI value
- * @audit_info: NetLabel audit information
- *
- * Description:
- * Removes an existing CIPSOv4 DOI definition from the NetLabel subsystem.
- * Returns zero on success, negative values on failure.
- *
- */
-int netlbl_cfg_cipsov4_del(u32 doi, struct netlbl_audit *audit_info)
-{
-	return cipso_v4_doi_remove(doi, audit_info, netlbl_cipsov4_doi_free);
+	goto cfg_cipsov4_add_map_return;
 }
 
 /*
@@ -452,7 +422,9 @@ int netlbl_enabled(void)
  * Attach the correct label to the given socket using the security attributes
  * specified in @secattr.  This function requires exclusive access to @sk,
  * which means it either needs to be in the process of being created or locked.
- * Returns zero on success, negative values on failure.
+ * Returns zero on success, -EDESTADDRREQ if the domain is configured to use
+ * network address selectors (can't blindly label the socket), and negative
+ * values on all other failures.
  *
  */
 int netlbl_sock_setattr(struct sock *sk,
@@ -466,6 +438,9 @@ int netlbl_sock_setattr(struct sock *sk,
 	if (dom_entry == NULL)
 		goto socket_setattr_return;
 	switch (dom_entry->type) {
+	case NETLBL_NLTYPE_ADDRSELECT:
+		ret_val = -EDESTADDRREQ;
+		break;
 	case NETLBL_NLTYPE_CIPSOV4:
 		ret_val = cipso_v4_sock_setattr(sk,
 						dom_entry->type_def.cipsov4,
@@ -484,6 +459,20 @@ socket_setattr_return:
 }
 
 /**
+ * netlbl_sock_delattr - Delete all the NetLabel labels on a socket
+ * @sk: the socket
+ *
+ * Description:
+ * Remove all the NetLabel labeling from @sk.  The caller is responsible for
+ * ensuring that @sk is locked.
+ *
+ */
+void netlbl_sock_delattr(struct sock *sk)
+{
+	cipso_v4_sock_delattr(sk);
+}
+
+/**
  * netlbl_sock_getattr - Determine the security attributes of a sock
  * @sk: the sock
  * @secattr: the security attributes
@@ -501,6 +490,128 @@ int netlbl_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr)
 }
 
 /**
+ * netlbl_conn_setattr - Label a connected socket using the correct protocol
+ * @sk: the socket to label
+ * @addr: the destination address
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Attach the correct label to the given connected socket using the security
+ * attributes specified in @secattr.  The caller is responsible for ensuring
+ * that @sk is locked.  Returns zero on success, negative values on failure.
+ *
+ */
+int netlbl_conn_setattr(struct sock *sk,
+			struct sockaddr *addr,
+			const struct netlbl_lsm_secattr *secattr)
+{
+	int ret_val;
+	struct sockaddr_in *addr4;
+	struct netlbl_domaddr4_map *af4_entry;
+
+	rcu_read_lock();
+	switch (addr->sa_family) {
+	case AF_INET:
+		addr4 = (struct sockaddr_in *)addr;
+		af4_entry = netlbl_domhsh_getentry_af4(secattr->domain,
+						       addr4->sin_addr.s_addr);
+		if (af4_entry == NULL) {
+			ret_val = -ENOENT;
+			goto conn_setattr_return;
+		}
+		switch (af4_entry->type) {
+		case NETLBL_NLTYPE_CIPSOV4:
+			ret_val = cipso_v4_sock_setattr(sk,
+						   af4_entry->type_def.cipsov4,
+						   secattr);
+			break;
+		case NETLBL_NLTYPE_UNLABELED:
+			/* just delete the protocols we support for right now
+			 * but we could remove other protocols if needed */
+			cipso_v4_sock_delattr(sk);
+			ret_val = 0;
+			break;
+		default:
+			ret_val = -ENOENT;
+		}
+		break;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	case AF_INET6:
+		/* since we don't support any IPv6 labeling protocols right
+		 * now we can optimize everything away until we do */
+		ret_val = 0;
+		break;
+#endif /* IPv6 */
+	default:
+		ret_val = 0;
+	}
+
+conn_setattr_return:
+	rcu_read_unlock();
+	return ret_val;
+}
+
+/**
+ * netlbl_skbuff_setattr - Label a packet using the correct protocol
+ * @skb: the packet
+ * @family: protocol family
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Attach the correct label to the given packet using the security attributes
+ * specified in @secattr.  Returns zero on success, negative values on failure.
+ *
+ */
+int netlbl_skbuff_setattr(struct sk_buff *skb,
+			  u16 family,
+			  const struct netlbl_lsm_secattr *secattr)
+{
+	int ret_val;
+	struct iphdr *hdr4;
+	struct netlbl_domaddr4_map *af4_entry;
+
+	rcu_read_lock();
+	switch (family) {
+	case AF_INET:
+		hdr4 = ip_hdr(skb);
+		af4_entry = netlbl_domhsh_getentry_af4(secattr->domain,
+						       hdr4->daddr);
+		if (af4_entry == NULL) {
+			ret_val = -ENOENT;
+			goto skbuff_setattr_return;
+		}
+		switch (af4_entry->type) {
+		case NETLBL_NLTYPE_CIPSOV4:
+			ret_val = cipso_v4_skbuff_setattr(skb,
+						   af4_entry->type_def.cipsov4,
+						   secattr);
+			break;
+		case NETLBL_NLTYPE_UNLABELED:
+			/* just delete the protocols we support for right now
+			 * but we could remove other protocols if needed */
+			ret_val = cipso_v4_skbuff_delattr(skb);
+			break;
+		default:
+			ret_val = -ENOENT;
+		}
+		break;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	case AF_INET6:
+		/* since we don't support any IPv6 labeling protocols right
+		 * now we can optimize everything away until we do */
+		ret_val = 0;
+		break;
+#endif /* IPv6 */
+	default:
+		ret_val = 0;
+	}
+
+skbuff_setattr_return:
+	rcu_read_unlock();
+	return ret_val;
+}
+
+/**
  * netlbl_skbuff_getattr - Determine the security attributes of a packet
  * @skb: the packet
  * @family: protocol family
@@ -528,6 +639,7 @@ int netlbl_skbuff_getattr(const struct sk_buff *skb,
  * netlbl_skbuff_err - Handle a LSM error on a sk_buff
  * @skb: the packet
  * @error: the error code
+ * @gateway: true if host is acting as a gateway, false otherwise
  *
  * Description:
  * Deal with a LSM problem when handling the packet in @skb, typically this is
@@ -535,10 +647,10 @@ int netlbl_skbuff_getattr(const struct sk_buff *skb,
  * according to the packet's labeling protocol.
  *
  */
-void netlbl_skbuff_err(struct sk_buff *skb, int error)
+void netlbl_skbuff_err(struct sk_buff *skb, int error, int gateway)
 {
 	if (CIPSO_V4_OPTEXIST(skb))
-		cipso_v4_error(skb, error, 0);
+		cipso_v4_error(skb, error, gateway);
 }
 
 /**
diff --git a/net/netlabel/netlabel_mgmt.c b/net/netlabel/netlabel_mgmt.c
index 44be5d5261f4..ee769ecaa13c 100644
--- a/net/netlabel/netlabel_mgmt.c
+++ b/net/netlabel/netlabel_mgmt.c
@@ -10,7 +10,7 @@
  */
 
 /*
- * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006, 2008
  *
  * This program is free software;  you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -32,9 +32,13 @@
 #include <linux/socket.h>
 #include <linux/string.h>
 #include <linux/skbuff.h>
+#include <linux/in.h>
+#include <linux/in6.h>
 #include <net/sock.h>
 #include <net/netlink.h>
 #include <net/genetlink.h>
+#include <net/ip.h>
+#include <net/ipv6.h>
 #include <net/netlabel.h>
 #include <net/cipso_ipv4.h>
 #include <asm/atomic.h>
@@ -71,86 +75,337 @@ static const struct nla_policy netlbl_mgmt_genl_policy[NLBL_MGMT_A_MAX + 1] = {
 };
 
 /*
- * NetLabel Command Handlers
+ * Helper Functions
  */
 
 /**
  * netlbl_mgmt_add - Handle an ADD message
- * @skb: the NETLINK buffer
  * @info: the Generic NETLINK info block
+ * @audit_info: NetLabel audit information
  *
  * Description:
- * Process a user generated ADD message and add the domains from the message
- * to the hash table.  See netlabel.h for a description of the message format.
- * Returns zero on success, negative values on failure.
+ * Helper function for the ADD and ADDDEF messages to add the domain mappings
+ * from the message to the hash table.  See netlabel.h for a description of the
+ * message format.  Returns zero on success, negative values on failure.
  *
  */
-static int netlbl_mgmt_add(struct sk_buff *skb, struct genl_info *info)
+static int netlbl_mgmt_add_common(struct genl_info *info,
+				  struct netlbl_audit *audit_info)
 {
 	int ret_val = -EINVAL;
 	struct netlbl_dom_map *entry = NULL;
-	size_t tmp_size;
+	struct netlbl_domaddr_map *addrmap = NULL;
+	struct cipso_v4_doi *cipsov4 = NULL;
 	u32 tmp_val;
-	struct netlbl_audit audit_info;
-
-	if (!info->attrs[NLBL_MGMT_A_DOMAIN] ||
-	    !info->attrs[NLBL_MGMT_A_PROTOCOL])
-		goto add_failure;
-
-	netlbl_netlink_auditinfo(skb, &audit_info);
 
 	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
 	if (entry == NULL) {
 		ret_val = -ENOMEM;
 		goto add_failure;
 	}
-	tmp_size = nla_len(info->attrs[NLBL_MGMT_A_DOMAIN]);
-	entry->domain = kmalloc(tmp_size, GFP_KERNEL);
-	if (entry->domain == NULL) {
-		ret_val = -ENOMEM;
-		goto add_failure;
-	}
 	entry->type = nla_get_u32(info->attrs[NLBL_MGMT_A_PROTOCOL]);
-	nla_strlcpy(entry->domain, info->attrs[NLBL_MGMT_A_DOMAIN], tmp_size);
+	if (info->attrs[NLBL_MGMT_A_DOMAIN]) {
+		size_t tmp_size = nla_len(info->attrs[NLBL_MGMT_A_DOMAIN]);
+		entry->domain = kmalloc(tmp_size, GFP_KERNEL);
+		if (entry->domain == NULL) {
+			ret_val = -ENOMEM;
+			goto add_failure;
+		}
+		nla_strlcpy(entry->domain,
+			    info->attrs[NLBL_MGMT_A_DOMAIN], tmp_size);
+	}
+
+	/* NOTE: internally we allow/use a entry->type value of
+	 *       NETLBL_NLTYPE_ADDRSELECT but we don't currently allow users
+	 *       to pass that as a protocol value because we need to know the
+	 *       "real" protocol */
 
 	switch (entry->type) {
 	case NETLBL_NLTYPE_UNLABELED:
-		ret_val = netlbl_domhsh_add(entry, &audit_info);
 		break;
 	case NETLBL_NLTYPE_CIPSOV4:
 		if (!info->attrs[NLBL_MGMT_A_CV4DOI])
 			goto add_failure;
 
 		tmp_val = nla_get_u32(info->attrs[NLBL_MGMT_A_CV4DOI]);
-		/* We should be holding a rcu_read_lock() here while we hold
-		 * the result but since the entry will always be deleted when
-		 * the CIPSO DOI is deleted we aren't going to keep the
-		 * lock. */
-		rcu_read_lock();
-		entry->type_def.cipsov4 = cipso_v4_doi_getdef(tmp_val);
-		if (entry->type_def.cipsov4 == NULL) {
-			rcu_read_unlock();
+		cipsov4 = cipso_v4_doi_getdef(tmp_val);
+		if (cipsov4 == NULL)
 			goto add_failure;
-		}
-		ret_val = netlbl_domhsh_add(entry, &audit_info);
-		rcu_read_unlock();
+		entry->type_def.cipsov4 = cipsov4;
 		break;
 	default:
 		goto add_failure;
 	}
+
+	if (info->attrs[NLBL_MGMT_A_IPV4ADDR]) {
+		struct in_addr *addr;
+		struct in_addr *mask;
+		struct netlbl_domaddr4_map *map;
+
+		addrmap = kzalloc(sizeof(*addrmap), GFP_KERNEL);
+		if (addrmap == NULL) {
+			ret_val = -ENOMEM;
+			goto add_failure;
+		}
+		INIT_LIST_HEAD(&addrmap->list4);
+		INIT_LIST_HEAD(&addrmap->list6);
+
+		if (nla_len(info->attrs[NLBL_MGMT_A_IPV4ADDR]) !=
+		    sizeof(struct in_addr)) {
+			ret_val = -EINVAL;
+			goto add_failure;
+		}
+		if (nla_len(info->attrs[NLBL_MGMT_A_IPV4MASK]) !=
+		    sizeof(struct in_addr)) {
+			ret_val = -EINVAL;
+			goto add_failure;
+		}
+		addr = nla_data(info->attrs[NLBL_MGMT_A_IPV4ADDR]);
+		mask = nla_data(info->attrs[NLBL_MGMT_A_IPV4MASK]);
+
+		map = kzalloc(sizeof(*map), GFP_KERNEL);
+		if (map == NULL) {
+			ret_val = -ENOMEM;
+			goto add_failure;
+		}
+		map->list.addr = addr->s_addr & mask->s_addr;
+		map->list.mask = mask->s_addr;
+		map->list.valid = 1;
+		map->type = entry->type;
+		if (cipsov4)
+			map->type_def.cipsov4 = cipsov4;
+
+		ret_val = netlbl_af4list_add(&map->list, &addrmap->list4);
+		if (ret_val != 0) {
+			kfree(map);
+			goto add_failure;
+		}
+
+		entry->type = NETLBL_NLTYPE_ADDRSELECT;
+		entry->type_def.addrsel = addrmap;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	} else if (info->attrs[NLBL_MGMT_A_IPV6ADDR]) {
+		struct in6_addr *addr;
+		struct in6_addr *mask;
+		struct netlbl_domaddr6_map *map;
+
+		addrmap = kzalloc(sizeof(*addrmap), GFP_KERNEL);
+		if (addrmap == NULL) {
+			ret_val = -ENOMEM;
+			goto add_failure;
+		}
+		INIT_LIST_HEAD(&addrmap->list4);
+		INIT_LIST_HEAD(&addrmap->list6);
+
+		if (nla_len(info->attrs[NLBL_MGMT_A_IPV6ADDR]) !=
+		    sizeof(struct in6_addr)) {
+			ret_val = -EINVAL;
+			goto add_failure;
+		}
+		if (nla_len(info->attrs[NLBL_MGMT_A_IPV6MASK]) !=
+		    sizeof(struct in6_addr)) {
+			ret_val = -EINVAL;
+			goto add_failure;
+		}
+		addr = nla_data(info->attrs[NLBL_MGMT_A_IPV6ADDR]);
+		mask = nla_data(info->attrs[NLBL_MGMT_A_IPV6MASK]);
+
+		map = kzalloc(sizeof(*map), GFP_KERNEL);
+		if (map == NULL) {
+			ret_val = -ENOMEM;
+			goto add_failure;
+		}
+		ipv6_addr_copy(&map->list.addr, addr);
+		map->list.addr.s6_addr32[0] &= mask->s6_addr32[0];
+		map->list.addr.s6_addr32[1] &= mask->s6_addr32[1];
+		map->list.addr.s6_addr32[2] &= mask->s6_addr32[2];
+		map->list.addr.s6_addr32[3] &= mask->s6_addr32[3];
+		ipv6_addr_copy(&map->list.mask, mask);
+		map->list.valid = 1;
+		map->type = entry->type;
+
+		ret_val = netlbl_af6list_add(&map->list, &addrmap->list6);
+		if (ret_val != 0) {
+			kfree(map);
+			goto add_failure;
+		}
+
+		entry->type = NETLBL_NLTYPE_ADDRSELECT;
+		entry->type_def.addrsel = addrmap;
+#endif /* IPv6 */
+	}
+
+	ret_val = netlbl_domhsh_add(entry, audit_info);
 	if (ret_val != 0)
 		goto add_failure;
 
 	return 0;
 
 add_failure:
+	if (cipsov4)
+		cipso_v4_doi_putdef(cipsov4);
 	if (entry)
 		kfree(entry->domain);
+	kfree(addrmap);
 	kfree(entry);
 	return ret_val;
 }
 
 /**
+ * netlbl_mgmt_listentry - List a NetLabel/LSM domain map entry
+ * @skb: the NETLINK buffer
+ * @entry: the map entry
+ *
+ * Description:
+ * This function is a helper function used by the LISTALL and LISTDEF command
+ * handlers.  The caller is responsibile for ensuring that the RCU read lock
+ * is held.  Returns zero on success, negative values on failure.
+ *
+ */
+static int netlbl_mgmt_listentry(struct sk_buff *skb,
+				 struct netlbl_dom_map *entry)
+{
+	int ret_val;
+	struct nlattr *nla_a;
+	struct nlattr *nla_b;
+	struct netlbl_af4list *iter4;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	struct netlbl_af6list *iter6;
+#endif
+
+	if (entry->domain != NULL) {
+		ret_val = nla_put_string(skb,
+					 NLBL_MGMT_A_DOMAIN, entry->domain);
+		if (ret_val != 0)
+			return ret_val;
+	}
+
+	switch (entry->type) {
+	case NETLBL_NLTYPE_ADDRSELECT:
+		nla_a = nla_nest_start(skb, NLBL_MGMT_A_SELECTORLIST);
+		if (nla_a == NULL)
+			return -ENOMEM;
+
+		netlbl_af4list_foreach_rcu(iter4,
+					   &entry->type_def.addrsel->list4) {
+			struct netlbl_domaddr4_map *map4;
+			struct in_addr addr_struct;
+
+			nla_b = nla_nest_start(skb, NLBL_MGMT_A_ADDRSELECTOR);
+			if (nla_b == NULL)
+				return -ENOMEM;
+
+			addr_struct.s_addr = iter4->addr;
+			ret_val = nla_put(skb, NLBL_MGMT_A_IPV4ADDR,
+					  sizeof(struct in_addr),
+					  &addr_struct);
+			if (ret_val != 0)
+				return ret_val;
+			addr_struct.s_addr = iter4->mask;
+			ret_val = nla_put(skb, NLBL_MGMT_A_IPV4MASK,
+					  sizeof(struct in_addr),
+					  &addr_struct);
+			if (ret_val != 0)
+				return ret_val;
+			map4 = netlbl_domhsh_addr4_entry(iter4);
+			ret_val = nla_put_u32(skb, NLBL_MGMT_A_PROTOCOL,
+					      map4->type);
+			if (ret_val != 0)
+				return ret_val;
+			switch (map4->type) {
+			case NETLBL_NLTYPE_CIPSOV4:
+				ret_val = nla_put_u32(skb, NLBL_MGMT_A_CV4DOI,
+						  map4->type_def.cipsov4->doi);
+				if (ret_val != 0)
+					return ret_val;
+				break;
+			}
+
+			nla_nest_end(skb, nla_b);
+		}
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+		netlbl_af6list_foreach_rcu(iter6,
+					   &entry->type_def.addrsel->list6) {
+			struct netlbl_domaddr6_map *map6;
+
+			nla_b = nla_nest_start(skb, NLBL_MGMT_A_ADDRSELECTOR);
+			if (nla_b == NULL)
+				return -ENOMEM;
+
+			ret_val = nla_put(skb, NLBL_MGMT_A_IPV6ADDR,
+					  sizeof(struct in6_addr),
+					  &iter6->addr);
+			if (ret_val != 0)
+				return ret_val;
+			ret_val = nla_put(skb, NLBL_MGMT_A_IPV6MASK,
+					  sizeof(struct in6_addr),
+					  &iter6->mask);
+			if (ret_val != 0)
+				return ret_val;
+			map6 = netlbl_domhsh_addr6_entry(iter6);
+			ret_val = nla_put_u32(skb, NLBL_MGMT_A_PROTOCOL,
+					      map6->type);
+			if (ret_val != 0)
+				return ret_val;
+
+			nla_nest_end(skb, nla_b);
+		}
+#endif /* IPv6 */
+
+		nla_nest_end(skb, nla_a);
+		break;
+	case NETLBL_NLTYPE_UNLABELED:
+		ret_val = nla_put_u32(skb, NLBL_MGMT_A_PROTOCOL, entry->type);
+		break;
+	case NETLBL_NLTYPE_CIPSOV4:
+		ret_val = nla_put_u32(skb, NLBL_MGMT_A_PROTOCOL, entry->type);
+		if (ret_val != 0)
+			return ret_val;
+		ret_val = nla_put_u32(skb, NLBL_MGMT_A_CV4DOI,
+				      entry->type_def.cipsov4->doi);
+		break;
+	}
+
+	return ret_val;
+}
+
+/*
+ * NetLabel Command Handlers
+ */
+
+/**
+ * netlbl_mgmt_add - Handle an ADD message
+ * @skb: the NETLINK buffer
+ * @info: the Generic NETLINK info block
+ *
+ * Description:
+ * Process a user generated ADD message and add the domains from the message
+ * to the hash table.  See netlabel.h for a description of the message format.
+ * Returns zero on success, negative values on failure.
+ *
+ */
+static int netlbl_mgmt_add(struct sk_buff *skb, struct genl_info *info)
+{
+	struct netlbl_audit audit_info;
+
+	if ((!info->attrs[NLBL_MGMT_A_DOMAIN]) ||
+	    (!info->attrs[NLBL_MGMT_A_PROTOCOL]) ||
+	    (info->attrs[NLBL_MGMT_A_IPV4ADDR] &&
+	     info->attrs[NLBL_MGMT_A_IPV6ADDR]) ||
+	    (info->attrs[NLBL_MGMT_A_IPV4MASK] &&
+	     info->attrs[NLBL_MGMT_A_IPV6MASK]) ||
+	    ((info->attrs[NLBL_MGMT_A_IPV4ADDR] != NULL) ^
+	     (info->attrs[NLBL_MGMT_A_IPV4MASK] != NULL)) ||
+	    ((info->attrs[NLBL_MGMT_A_IPV6ADDR] != NULL) ^
+	     (info->attrs[NLBL_MGMT_A_IPV6MASK] != NULL)))
+		return -EINVAL;
+
+	netlbl_netlink_auditinfo(skb, &audit_info);
+
+	return netlbl_mgmt_add_common(info, &audit_info);
+}
+
+/**
  * netlbl_mgmt_remove - Handle a REMOVE message
  * @skb: the NETLINK buffer
  * @info: the Generic NETLINK info block
@@ -198,23 +453,9 @@ static int netlbl_mgmt_listall_cb(struct netlbl_dom_map *entry, void *arg)
 	if (data == NULL)
 		goto listall_cb_failure;
 
-	ret_val = nla_put_string(cb_arg->skb,
-				 NLBL_MGMT_A_DOMAIN,
-				 entry->domain);
+	ret_val = netlbl_mgmt_listentry(cb_arg->skb, entry);
 	if (ret_val != 0)
 		goto listall_cb_failure;
-	ret_val = nla_put_u32(cb_arg->skb, NLBL_MGMT_A_PROTOCOL, entry->type);
-	if (ret_val != 0)
-		goto listall_cb_failure;
-	switch (entry->type) {
-	case NETLBL_NLTYPE_CIPSOV4:
-		ret_val = nla_put_u32(cb_arg->skb,
-				      NLBL_MGMT_A_CV4DOI,
-				      entry->type_def.cipsov4->doi);
-		if (ret_val != 0)
-			goto listall_cb_failure;
-		break;
-	}
 
 	cb_arg->seq++;
 	return genlmsg_end(cb_arg->skb, data);
@@ -268,56 +509,22 @@ static int netlbl_mgmt_listall(struct sk_buff *skb,
  */
 static int netlbl_mgmt_adddef(struct sk_buff *skb, struct genl_info *info)
 {
-	int ret_val = -EINVAL;
-	struct netlbl_dom_map *entry = NULL;
-	u32 tmp_val;
 	struct netlbl_audit audit_info;
 
-	if (!info->attrs[NLBL_MGMT_A_PROTOCOL])
-		goto adddef_failure;
+	if ((!info->attrs[NLBL_MGMT_A_PROTOCOL]) ||
+	    (info->attrs[NLBL_MGMT_A_IPV4ADDR] &&
+	     info->attrs[NLBL_MGMT_A_IPV6ADDR]) ||
+	    (info->attrs[NLBL_MGMT_A_IPV4MASK] &&
+	     info->attrs[NLBL_MGMT_A_IPV6MASK]) ||
+	    ((info->attrs[NLBL_MGMT_A_IPV4ADDR] != NULL) ^
+	     (info->attrs[NLBL_MGMT_A_IPV4MASK] != NULL)) ||
+	    ((info->attrs[NLBL_MGMT_A_IPV6ADDR] != NULL) ^
+	     (info->attrs[NLBL_MGMT_A_IPV6MASK] != NULL)))
+		return -EINVAL;
 
 	netlbl_netlink_auditinfo(skb, &audit_info);
 
-	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
-	if (entry == NULL) {
-		ret_val = -ENOMEM;
-		goto adddef_failure;
-	}
-	entry->type = nla_get_u32(info->attrs[NLBL_MGMT_A_PROTOCOL]);
-
-	switch (entry->type) {
-	case NETLBL_NLTYPE_UNLABELED:
-		ret_val = netlbl_domhsh_add_default(entry, &audit_info);
-		break;
-	case NETLBL_NLTYPE_CIPSOV4:
-		if (!info->attrs[NLBL_MGMT_A_CV4DOI])
-			goto adddef_failure;
-
-		tmp_val = nla_get_u32(info->attrs[NLBL_MGMT_A_CV4DOI]);
-		/* We should be holding a rcu_read_lock() here while we hold
-		 * the result but since the entry will always be deleted when
-		 * the CIPSO DOI is deleted we aren't going to keep the
-		 * lock. */
-		rcu_read_lock();
-		entry->type_def.cipsov4 = cipso_v4_doi_getdef(tmp_val);
-		if (entry->type_def.cipsov4 == NULL) {
-			rcu_read_unlock();
-			goto adddef_failure;
-		}
-		ret_val = netlbl_domhsh_add_default(entry, &audit_info);
-		rcu_read_unlock();
-		break;
-	default:
-		goto adddef_failure;
-	}
-	if (ret_val != 0)
-		goto adddef_failure;
-
-	return 0;
-
-adddef_failure:
-	kfree(entry);
-	return ret_val;
+	return netlbl_mgmt_add_common(info, &audit_info);
 }
 
 /**
@@ -371,19 +578,10 @@ static int netlbl_mgmt_listdef(struct sk_buff *skb, struct genl_info *info)
 		ret_val = -ENOENT;
 		goto listdef_failure_lock;
 	}
-	ret_val = nla_put_u32(ans_skb, NLBL_MGMT_A_PROTOCOL, entry->type);
-	if (ret_val != 0)
-		goto listdef_failure_lock;
-	switch (entry->type) {
-	case NETLBL_NLTYPE_CIPSOV4:
-		ret_val = nla_put_u32(ans_skb,
-				      NLBL_MGMT_A_CV4DOI,
-				      entry->type_def.cipsov4->doi);
-		if (ret_val != 0)
-			goto listdef_failure_lock;
-		break;
-	}
+	ret_val = netlbl_mgmt_listentry(ans_skb, entry);
 	rcu_read_unlock();
+	if (ret_val != 0)
+		goto listdef_failure;
 
 	genlmsg_end(ans_skb, data);
 	return genlmsg_reply(ans_skb, info);
diff --git a/net/netlabel/netlabel_mgmt.h b/net/netlabel/netlabel_mgmt.h
index a43bff169d6b..05d96431f819 100644
--- a/net/netlabel/netlabel_mgmt.h
+++ b/net/netlabel/netlabel_mgmt.h
@@ -45,6 +45,16 @@
  *     NLBL_MGMT_A_DOMAIN
  *     NLBL_MGMT_A_PROTOCOL
  *
+ *   If IPv4 is specified the following attributes are required:
+ *
+ *     NLBL_MGMT_A_IPV4ADDR
+ *     NLBL_MGMT_A_IPV4MASK
+ *
+ *   If IPv6 is specified the following attributes are required:
+ *
+ *     NLBL_MGMT_A_IPV6ADDR
+ *     NLBL_MGMT_A_IPV6MASK
+ *
  *   If using NETLBL_NLTYPE_CIPSOV4 the following attributes are required:
  *
  *     NLBL_MGMT_A_CV4DOI
@@ -68,13 +78,24 @@
  *   Required attributes:
  *
  *     NLBL_MGMT_A_DOMAIN
+ *
+ *   If the IP address selectors are not used the following attribute is
+ *   required:
+ *
  *     NLBL_MGMT_A_PROTOCOL
  *
- *   If using NETLBL_NLTYPE_CIPSOV4 the following attributes are required:
+ *   If the IP address selectors are used then the following attritbute is
+ *   required:
+ *
+ *     NLBL_MGMT_A_SELECTORLIST
+ *
+ *   If the mapping is using the NETLBL_NLTYPE_CIPSOV4 type then the following
+ *   attributes are required:
  *
  *     NLBL_MGMT_A_CV4DOI
  *
- *   If using NETLBL_NLTYPE_UNLABELED no other attributes are required.
+ *   If the mapping is using the NETLBL_NLTYPE_UNLABELED type no other
+ *   attributes are required.
  *
  * o ADDDEF:
  *   Sent by an application to set the default domain mapping for the NetLabel
@@ -100,15 +121,23 @@
  *   application there is no payload.  On success the kernel should send a
  *   response using the following format.
  *
- *   Required attributes:
+ *   If the IP address selectors are not used the following attribute is
+ *   required:
  *
  *     NLBL_MGMT_A_PROTOCOL
  *
- *   If using NETLBL_NLTYPE_CIPSOV4 the following attributes are required:
+ *   If the IP address selectors are used then the following attritbute is
+ *   required:
+ *
+ *     NLBL_MGMT_A_SELECTORLIST
+ *
+ *   If the mapping is using the NETLBL_NLTYPE_CIPSOV4 type then the following
+ *   attributes are required:
  *
  *     NLBL_MGMT_A_CV4DOI
  *
- *   If using NETLBL_NLTYPE_UNLABELED no other attributes are required.
+ *   If the mapping is using the NETLBL_NLTYPE_UNLABELED type no other
+ *   attributes are required.
  *
  * o PROTOCOLS:
  *   Sent by an application to request a list of configured NetLabel protocols
@@ -162,6 +191,26 @@ enum {
 	NLBL_MGMT_A_CV4DOI,
 	/* (NLA_U32)
 	 * the CIPSOv4 DOI value */
+	NLBL_MGMT_A_IPV6ADDR,
+	/* (NLA_BINARY, struct in6_addr)
+	 * an IPv6 address */
+	NLBL_MGMT_A_IPV6MASK,
+	/* (NLA_BINARY, struct in6_addr)
+	 * an IPv6 address mask */
+	NLBL_MGMT_A_IPV4ADDR,
+	/* (NLA_BINARY, struct in_addr)
+	 * an IPv4 address */
+	NLBL_MGMT_A_IPV4MASK,
+	/* (NLA_BINARY, struct in_addr)
+	 * and IPv4 address mask */
+	NLBL_MGMT_A_ADDRSELECTOR,
+	/* (NLA_NESTED)
+	 * an IP address selector, must contain an address, mask, and protocol
+	 * attribute plus any protocol specific attributes */
+	NLBL_MGMT_A_SELECTORLIST,
+	/* (NLA_NESTED)
+	 * the selector list, there must be at least one
+	 * NLBL_MGMT_A_ADDRSELECTOR attribute */
 	__NLBL_MGMT_A_MAX,
 };
 #define NLBL_MGMT_A_MAX (__NLBL_MGMT_A_MAX - 1)
diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c
index 921c118ead89..e8a5c32b0f10 100644
--- a/net/netlabel/netlabel_unlabeled.c
+++ b/net/netlabel/netlabel_unlabeled.c
@@ -10,7 +10,7 @@
  */
 
 /*
- * (c) Copyright Hewlett-Packard Development Company, L.P., 2006 - 2007
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006 - 2008
  *
  * This program is free software;  you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -54,6 +54,7 @@
 #include <asm/atomic.h>
 
 #include "netlabel_user.h"
+#include "netlabel_addrlist.h"
 #include "netlabel_domainhash.h"
 #include "netlabel_unlabeled.h"
 #include "netlabel_mgmt.h"
@@ -76,22 +77,20 @@ struct netlbl_unlhsh_tbl {
 	struct list_head *tbl;
 	u32 size;
 };
+#define netlbl_unlhsh_addr4_entry(iter) \
+	container_of(iter, struct netlbl_unlhsh_addr4, list)
 struct netlbl_unlhsh_addr4 {
-	__be32 addr;
-	__be32 mask;
 	u32 secid;
 
-	u32 valid;
-	struct list_head list;
+	struct netlbl_af4list list;
 	struct rcu_head rcu;
 };
+#define netlbl_unlhsh_addr6_entry(iter) \
+	container_of(iter, struct netlbl_unlhsh_addr6, list)
 struct netlbl_unlhsh_addr6 {
-	struct in6_addr addr;
-	struct in6_addr mask;
 	u32 secid;
 
-	u32 valid;
-	struct list_head list;
+	struct netlbl_af6list list;
 	struct rcu_head rcu;
 };
 struct netlbl_unlhsh_iface {
@@ -147,76 +146,6 @@ static const struct nla_policy netlbl_unlabel_genl_policy[NLBL_UNLABEL_A_MAX + 1
 };
 
 /*
- * Audit Helper Functions
- */
-
-/**
- * netlbl_unlabel_audit_addr4 - Audit an IPv4 address
- * @audit_buf: audit buffer
- * @dev: network interface
- * @addr: IP address
- * @mask: IP address mask
- *
- * Description:
- * Write the IPv4 address and address mask, if necessary, to @audit_buf.
- *
- */
-static void netlbl_unlabel_audit_addr4(struct audit_buffer *audit_buf,
-				     const char *dev,
-				     __be32 addr, __be32 mask)
-{
-	u32 mask_val = ntohl(mask);
-
-	if (dev != NULL)
-		audit_log_format(audit_buf, " netif=%s", dev);
-	audit_log_format(audit_buf, " src=" NIPQUAD_FMT, NIPQUAD(addr));
-	if (mask_val != 0xffffffff) {
-		u32 mask_len = 0;
-		while (mask_val > 0) {
-			mask_val <<= 1;
-			mask_len++;
-		}
-		audit_log_format(audit_buf, " src_prefixlen=%d", mask_len);
-	}
-}
-
-#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
-/**
- * netlbl_unlabel_audit_addr6 - Audit an IPv6 address
- * @audit_buf: audit buffer
- * @dev: network interface
- * @addr: IP address
- * @mask: IP address mask
- *
- * Description:
- * Write the IPv6 address and address mask, if necessary, to @audit_buf.
- *
- */
-static void netlbl_unlabel_audit_addr6(struct audit_buffer *audit_buf,
-				     const char *dev,
-				     const struct in6_addr *addr,
-				     const struct in6_addr *mask)
-{
-	if (dev != NULL)
-		audit_log_format(audit_buf, " netif=%s", dev);
-	audit_log_format(audit_buf, " src=" NIP6_FMT, NIP6(*addr));
-	if (ntohl(mask->s6_addr32[3]) != 0xffffffff) {
-		u32 mask_len = 0;
-		u32 mask_val;
-		int iter = -1;
-		while (ntohl(mask->s6_addr32[++iter]) == 0xffffffff)
-			mask_len += 32;
-		mask_val = ntohl(mask->s6_addr32[iter]);
-		while (mask_val > 0) {
-			mask_val <<= 1;
-			mask_len++;
-		}
-		audit_log_format(audit_buf, " src_prefixlen=%d", mask_len);
-	}
-}
-#endif /* IPv6 */
-
-/*
  * Unlabeled Connection Hash Table Functions
  */
 
@@ -274,26 +203,28 @@ static void netlbl_unlhsh_free_addr6(struct rcu_head *entry)
 static void netlbl_unlhsh_free_iface(struct rcu_head *entry)
 {
 	struct netlbl_unlhsh_iface *iface;
-	struct netlbl_unlhsh_addr4 *iter4;
-	struct netlbl_unlhsh_addr4 *tmp4;
-	struct netlbl_unlhsh_addr6 *iter6;
-	struct netlbl_unlhsh_addr6 *tmp6;
+	struct netlbl_af4list *iter4;
+	struct netlbl_af4list *tmp4;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	struct netlbl_af6list *iter6;
+	struct netlbl_af6list *tmp6;
+#endif /* IPv6 */
 
 	iface = container_of(entry, struct netlbl_unlhsh_iface, rcu);
 
 	/* no need for locks here since we are the only one with access to this
 	 * structure */
 
-	list_for_each_entry_safe(iter4, tmp4, &iface->addr4_list, list)
-		if (iter4->valid) {
-			list_del_rcu(&iter4->list);
-			kfree(iter4);
-		}
-	list_for_each_entry_safe(iter6, tmp6, &iface->addr6_list, list)
-		if (iter6->valid) {
-			list_del_rcu(&iter6->list);
-			kfree(iter6);
-		}
+	netlbl_af4list_foreach_safe(iter4, tmp4, &iface->addr4_list) {
+		netlbl_af4list_remove_entry(iter4);
+		kfree(netlbl_unlhsh_addr4_entry(iter4));
+	}
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	netlbl_af6list_foreach_safe(iter6, tmp6, &iface->addr6_list) {
+		netlbl_af6list_remove_entry(iter6);
+		kfree(netlbl_unlhsh_addr6_entry(iter6));
+	}
+#endif /* IPv6 */
 	kfree(iface);
 }
 
@@ -316,59 +247,6 @@ static u32 netlbl_unlhsh_hash(int ifindex)
 }
 
 /**
- * netlbl_unlhsh_search_addr4 - Search for a matching IPv4 address entry
- * @addr: IPv4 address
- * @iface: the network interface entry
- *
- * Description:
- * Searches the IPv4 address list of the network interface specified by @iface.
- * If a matching address entry is found it is returned, otherwise NULL is
- * returned.  The caller is responsible for calling the rcu_read_[un]lock()
- * functions.
- *
- */
-static struct netlbl_unlhsh_addr4 *netlbl_unlhsh_search_addr4(
-	                               __be32 addr,
-	                               const struct netlbl_unlhsh_iface *iface)
-{
-	struct netlbl_unlhsh_addr4 *iter;
-
-	list_for_each_entry_rcu(iter, &iface->addr4_list, list)
-		if (iter->valid && (addr & iter->mask) == iter->addr)
-			return iter;
-
-	return NULL;
-}
-
-#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
-/**
- * netlbl_unlhsh_search_addr6 - Search for a matching IPv6 address entry
- * @addr: IPv6 address
- * @iface: the network interface entry
- *
- * Description:
- * Searches the IPv6 address list of the network interface specified by @iface.
- * If a matching address entry is found it is returned, otherwise NULL is
- * returned.  The caller is responsible for calling the rcu_read_[un]lock()
- * functions.
- *
- */
-static struct netlbl_unlhsh_addr6 *netlbl_unlhsh_search_addr6(
-	                               const struct in6_addr *addr,
-	                               const struct netlbl_unlhsh_iface *iface)
-{
-	struct netlbl_unlhsh_addr6 *iter;
-
-	list_for_each_entry_rcu(iter, &iface->addr6_list, list)
-		if (iter->valid &&
-		    ipv6_masked_addr_cmp(&iter->addr, &iter->mask, addr) == 0)
-		return iter;
-
-	return NULL;
-}
-#endif /* IPv6 */
-
-/**
  * netlbl_unlhsh_search_iface - Search for a matching interface entry
  * @ifindex: the network interface
  *
@@ -381,12 +259,12 @@ static struct netlbl_unlhsh_addr6 *netlbl_unlhsh_search_addr6(
 static struct netlbl_unlhsh_iface *netlbl_unlhsh_search_iface(int ifindex)
 {
 	u32 bkt;
+	struct list_head *bkt_list;
 	struct netlbl_unlhsh_iface *iter;
 
 	bkt = netlbl_unlhsh_hash(ifindex);
-	list_for_each_entry_rcu(iter,
-				&rcu_dereference(netlbl_unlhsh)->tbl[bkt],
-				list)
+	bkt_list = &rcu_dereference(netlbl_unlhsh)->tbl[bkt];
+	list_for_each_entry_rcu(iter, bkt_list, list)
 		if (iter->valid && iter->ifindex == ifindex)
 			return iter;
 
@@ -439,43 +317,26 @@ static int netlbl_unlhsh_add_addr4(struct netlbl_unlhsh_iface *iface,
 				   const struct in_addr *mask,
 				   u32 secid)
 {
+	int ret_val;
 	struct netlbl_unlhsh_addr4 *entry;
-	struct netlbl_unlhsh_addr4 *iter;
 
 	entry = kzalloc(sizeof(*entry), GFP_ATOMIC);
 	if (entry == NULL)
 		return -ENOMEM;
 
-	entry->addr = addr->s_addr & mask->s_addr;
-	entry->mask = mask->s_addr;
-	entry->secid = secid;
-	entry->valid = 1;
+	entry->list.addr = addr->s_addr & mask->s_addr;
+	entry->list.mask = mask->s_addr;
+	entry->list.valid = 1;
 	INIT_RCU_HEAD(&entry->rcu);
+	entry->secid = secid;
 
 	spin_lock(&netlbl_unlhsh_lock);
-	iter = netlbl_unlhsh_search_addr4(entry->addr, iface);
-	if (iter != NULL &&
-	    iter->addr == addr->s_addr && iter->mask == mask->s_addr) {
-		spin_unlock(&netlbl_unlhsh_lock);
-		kfree(entry);
-		return -EEXIST;
-	}
-	/* in order to speed up address searches through the list (the common
-	 * case) we need to keep the list in order based on the size of the
-	 * address mask such that the entry with the widest mask (smallest
-	 * numerical value) appears first in the list */
-	list_for_each_entry_rcu(iter, &iface->addr4_list, list)
-		if (iter->valid &&
-		    ntohl(entry->mask) > ntohl(iter->mask)) {
-			__list_add_rcu(&entry->list,
-				       iter->list.prev,
-				       &iter->list);
-			spin_unlock(&netlbl_unlhsh_lock);
-			return 0;
-		}
-	list_add_tail_rcu(&entry->list, &iface->addr4_list);
+	ret_val = netlbl_af4list_add(&entry->list, &iface->addr4_list);
 	spin_unlock(&netlbl_unlhsh_lock);
-	return 0;
+
+	if (ret_val != 0)
+		kfree(entry);
+	return ret_val;
 }
 
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
@@ -498,47 +359,29 @@ static int netlbl_unlhsh_add_addr6(struct netlbl_unlhsh_iface *iface,
 				   const struct in6_addr *mask,
 				   u32 secid)
 {
+	int ret_val;
 	struct netlbl_unlhsh_addr6 *entry;
-	struct netlbl_unlhsh_addr6 *iter;
 
 	entry = kzalloc(sizeof(*entry), GFP_ATOMIC);
 	if (entry == NULL)
 		return -ENOMEM;
 
-	ipv6_addr_copy(&entry->addr, addr);
-	entry->addr.s6_addr32[0] &= mask->s6_addr32[0];
-	entry->addr.s6_addr32[1] &= mask->s6_addr32[1];
-	entry->addr.s6_addr32[2] &= mask->s6_addr32[2];
-	entry->addr.s6_addr32[3] &= mask->s6_addr32[3];
-	ipv6_addr_copy(&entry->mask, mask);
-	entry->secid = secid;
-	entry->valid = 1;
+	ipv6_addr_copy(&entry->list.addr, addr);
+	entry->list.addr.s6_addr32[0] &= mask->s6_addr32[0];
+	entry->list.addr.s6_addr32[1] &= mask->s6_addr32[1];
+	entry->list.addr.s6_addr32[2] &= mask->s6_addr32[2];
+	entry->list.addr.s6_addr32[3] &= mask->s6_addr32[3];
+	ipv6_addr_copy(&entry->list.mask, mask);
+	entry->list.valid = 1;
 	INIT_RCU_HEAD(&entry->rcu);
+	entry->secid = secid;
 
 	spin_lock(&netlbl_unlhsh_lock);
-	iter = netlbl_unlhsh_search_addr6(&entry->addr, iface);
-	if (iter != NULL &&
-	    (ipv6_addr_equal(&iter->addr, addr) &&
-	     ipv6_addr_equal(&iter->mask, mask))) {
-		spin_unlock(&netlbl_unlhsh_lock);
-		kfree(entry);
-		return -EEXIST;
-	}
-	/* in order to speed up address searches through the list (the common
-	 * case) we need to keep the list in order based on the size of the
-	 * address mask such that the entry with the widest mask (smallest
-	 * numerical value) appears first in the list */
-	list_for_each_entry_rcu(iter, &iface->addr6_list, list)
-		if (iter->valid &&
-		    ipv6_addr_cmp(&entry->mask, &iter->mask) > 0) {
-			__list_add_rcu(&entry->list,
-				       iter->list.prev,
-				       &iter->list);
-			spin_unlock(&netlbl_unlhsh_lock);
-			return 0;
-		}
-	list_add_tail_rcu(&entry->list, &iface->addr6_list);
+	ret_val = netlbl_af6list_add(&entry->list, &iface->addr6_list);
 	spin_unlock(&netlbl_unlhsh_lock);
+
+	if (ret_val != 0)
+		kfree(entry);
 	return 0;
 }
 #endif /* IPv6 */
@@ -658,10 +501,10 @@ static int netlbl_unlhsh_add(struct net *net,
 		mask4 = (struct in_addr *)mask;
 		ret_val = netlbl_unlhsh_add_addr4(iface, addr4, mask4, secid);
 		if (audit_buf != NULL)
-			netlbl_unlabel_audit_addr4(audit_buf,
-						   dev_name,
-						   addr4->s_addr,
-						   mask4->s_addr);
+			netlbl_af4list_audit_addr(audit_buf, 1,
+						  dev_name,
+						  addr4->s_addr,
+						  mask4->s_addr);
 		break;
 	}
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
@@ -672,9 +515,9 @@ static int netlbl_unlhsh_add(struct net *net,
 		mask6 = (struct in6_addr *)mask;
 		ret_val = netlbl_unlhsh_add_addr6(iface, addr6, mask6, secid);
 		if (audit_buf != NULL)
-			netlbl_unlabel_audit_addr6(audit_buf,
-						   dev_name,
-						   addr6, mask6);
+			netlbl_af6list_audit_addr(audit_buf, 1,
+						  dev_name,
+						  addr6, mask6);
 		break;
 	}
 #endif /* IPv6 */
@@ -719,35 +562,34 @@ static int netlbl_unlhsh_remove_addr4(struct net *net,
 				      const struct in_addr *mask,
 				      struct netlbl_audit *audit_info)
 {
-	int ret_val = -ENOENT;
+	int ret_val = 0;
+	struct netlbl_af4list *list_entry;
 	struct netlbl_unlhsh_addr4 *entry;
-	struct audit_buffer *audit_buf = NULL;
+	struct audit_buffer *audit_buf;
 	struct net_device *dev;
-	char *secctx = NULL;
+	char *secctx;
 	u32 secctx_len;
 
 	spin_lock(&netlbl_unlhsh_lock);
-	entry = netlbl_unlhsh_search_addr4(addr->s_addr, iface);
-	if (entry != NULL &&
-	    entry->addr == addr->s_addr && entry->mask == mask->s_addr) {
-		entry->valid = 0;
-		list_del_rcu(&entry->list);
-		ret_val = 0;
-	}
+	list_entry = netlbl_af4list_remove(addr->s_addr, mask->s_addr,
+					   &iface->addr4_list);
 	spin_unlock(&netlbl_unlhsh_lock);
+	if (list_entry == NULL)
+		ret_val = -ENOENT;
+	entry = netlbl_unlhsh_addr4_entry(list_entry);
 
 	audit_buf = netlbl_audit_start_common(AUDIT_MAC_UNLBL_STCDEL,
 					      audit_info);
 	if (audit_buf != NULL) {
 		dev = dev_get_by_index(net, iface->ifindex);
-		netlbl_unlabel_audit_addr4(audit_buf,
-					   (dev != NULL ? dev->name : NULL),
-					   entry->addr, entry->mask);
+		netlbl_af4list_audit_addr(audit_buf, 1,
+					  (dev != NULL ? dev->name : NULL),
+					  addr->s_addr, mask->s_addr);
 		if (dev != NULL)
 			dev_put(dev);
-		if (security_secid_to_secctx(entry->secid,
-					     &secctx,
-					     &secctx_len) == 0) {
+		if (entry && security_secid_to_secctx(entry->secid,
+						      &secctx,
+						      &secctx_len) == 0) {
 			audit_log_format(audit_buf, " sec_obj=%s", secctx);
 			security_release_secctx(secctx, secctx_len);
 		}
@@ -781,36 +623,33 @@ static int netlbl_unlhsh_remove_addr6(struct net *net,
 				      const struct in6_addr *mask,
 				      struct netlbl_audit *audit_info)
 {
-	int ret_val = -ENOENT;
+	int ret_val = 0;
+	struct netlbl_af6list *list_entry;
 	struct netlbl_unlhsh_addr6 *entry;
-	struct audit_buffer *audit_buf = NULL;
+	struct audit_buffer *audit_buf;
 	struct net_device *dev;
-	char *secctx = NULL;
+	char *secctx;
 	u32 secctx_len;
 
 	spin_lock(&netlbl_unlhsh_lock);
-	entry = netlbl_unlhsh_search_addr6(addr, iface);
-	if (entry != NULL &&
-	    (ipv6_addr_equal(&entry->addr, addr) &&
-	     ipv6_addr_equal(&entry->mask, mask))) {
-		entry->valid = 0;
-		list_del_rcu(&entry->list);
-		ret_val = 0;
-	}
+	list_entry = netlbl_af6list_remove(addr, mask, &iface->addr6_list);
 	spin_unlock(&netlbl_unlhsh_lock);
+	if (list_entry == NULL)
+		ret_val = -ENOENT;
+	entry = netlbl_unlhsh_addr6_entry(list_entry);
 
 	audit_buf = netlbl_audit_start_common(AUDIT_MAC_UNLBL_STCDEL,
 					      audit_info);
 	if (audit_buf != NULL) {
 		dev = dev_get_by_index(net, iface->ifindex);
-		netlbl_unlabel_audit_addr6(audit_buf,
-					   (dev != NULL ? dev->name : NULL),
-					   addr, mask);
+		netlbl_af6list_audit_addr(audit_buf, 1,
+					  (dev != NULL ? dev->name : NULL),
+					  addr, mask);
 		if (dev != NULL)
 			dev_put(dev);
-		if (security_secid_to_secctx(entry->secid,
-					     &secctx,
-					     &secctx_len) == 0) {
+		if (entry && security_secid_to_secctx(entry->secid,
+						      &secctx,
+						      &secctx_len) == 0) {
 			audit_log_format(audit_buf, " sec_obj=%s", secctx);
 			security_release_secctx(secctx, secctx_len);
 		}
@@ -836,16 +675,18 @@ static int netlbl_unlhsh_remove_addr6(struct net *net,
  */
 static void netlbl_unlhsh_condremove_iface(struct netlbl_unlhsh_iface *iface)
 {
-	struct netlbl_unlhsh_addr4 *iter4;
-	struct netlbl_unlhsh_addr6 *iter6;
+	struct netlbl_af4list *iter4;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	struct netlbl_af6list *iter6;
+#endif /* IPv6 */
 
 	spin_lock(&netlbl_unlhsh_lock);
-	list_for_each_entry_rcu(iter4, &iface->addr4_list, list)
-		if (iter4->valid)
-			goto unlhsh_condremove_failure;
-	list_for_each_entry_rcu(iter6, &iface->addr6_list, list)
-		if (iter6->valid)
-			goto unlhsh_condremove_failure;
+	netlbl_af4list_foreach_rcu(iter4, &iface->addr4_list)
+		goto unlhsh_condremove_failure;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	netlbl_af6list_foreach_rcu(iter6, &iface->addr6_list)
+		goto unlhsh_condremove_failure;
+#endif /* IPv6 */
 	iface->valid = 0;
 	if (iface->ifindex > 0)
 		list_del_rcu(&iface->list);
@@ -1349,7 +1190,7 @@ static int netlbl_unlabel_staticlist_gen(u32 cmd,
 	if (addr4) {
 		struct in_addr addr_struct;
 
-		addr_struct.s_addr = addr4->addr;
+		addr_struct.s_addr = addr4->list.addr;
 		ret_val = nla_put(cb_arg->skb,
 				  NLBL_UNLABEL_A_IPV4ADDR,
 				  sizeof(struct in_addr),
@@ -1357,7 +1198,7 @@ static int netlbl_unlabel_staticlist_gen(u32 cmd,
 		if (ret_val != 0)
 			goto list_cb_failure;
 
-		addr_struct.s_addr = addr4->mask;
+		addr_struct.s_addr = addr4->list.mask;
 		ret_val = nla_put(cb_arg->skb,
 				  NLBL_UNLABEL_A_IPV4MASK,
 				  sizeof(struct in_addr),
@@ -1370,14 +1211,14 @@ static int netlbl_unlabel_staticlist_gen(u32 cmd,
 		ret_val = nla_put(cb_arg->skb,
 				  NLBL_UNLABEL_A_IPV6ADDR,
 				  sizeof(struct in6_addr),
-				  &addr6->addr);
+				  &addr6->list.addr);
 		if (ret_val != 0)
 			goto list_cb_failure;
 
 		ret_val = nla_put(cb_arg->skb,
 				  NLBL_UNLABEL_A_IPV6MASK,
 				  sizeof(struct in6_addr),
-				  &addr6->mask);
+				  &addr6->list.mask);
 		if (ret_val != 0)
 			goto list_cb_failure;
 
@@ -1425,8 +1266,11 @@ static int netlbl_unlabel_staticlist(struct sk_buff *skb,
 	u32 iter_bkt;
 	u32 iter_chain = 0, iter_addr4 = 0, iter_addr6 = 0;
 	struct netlbl_unlhsh_iface *iface;
-	struct netlbl_unlhsh_addr4 *addr4;
-	struct netlbl_unlhsh_addr6 *addr6;
+	struct list_head *iter_list;
+	struct netlbl_af4list *addr4;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	struct netlbl_af6list *addr6;
+#endif
 
 	cb_arg.nl_cb = cb;
 	cb_arg.skb = skb;
@@ -1436,44 +1280,43 @@ static int netlbl_unlabel_staticlist(struct sk_buff *skb,
 	for (iter_bkt = skip_bkt;
 	     iter_bkt < rcu_dereference(netlbl_unlhsh)->size;
 	     iter_bkt++, iter_chain = 0, iter_addr4 = 0, iter_addr6 = 0) {
-		list_for_each_entry_rcu(iface,
-			        &rcu_dereference(netlbl_unlhsh)->tbl[iter_bkt],
-				list) {
+		iter_list = &rcu_dereference(netlbl_unlhsh)->tbl[iter_bkt];
+		list_for_each_entry_rcu(iface, iter_list, list) {
 			if (!iface->valid ||
 			    iter_chain++ < skip_chain)
 				continue;
-			list_for_each_entry_rcu(addr4,
-						&iface->addr4_list,
-						list) {
-				if (!addr4->valid || iter_addr4++ < skip_addr4)
+			netlbl_af4list_foreach_rcu(addr4,
+						   &iface->addr4_list) {
+				if (iter_addr4++ < skip_addr4)
 					continue;
 				if (netlbl_unlabel_staticlist_gen(
-					             NLBL_UNLABEL_C_STATICLIST,
-						     iface,
-						     addr4,
-						     NULL,
-						     &cb_arg) < 0) {
+					      NLBL_UNLABEL_C_STATICLIST,
+					      iface,
+					      netlbl_unlhsh_addr4_entry(addr4),
+					      NULL,
+					      &cb_arg) < 0) {
 					iter_addr4--;
 					iter_chain--;
 					goto unlabel_staticlist_return;
 				}
 			}
-			list_for_each_entry_rcu(addr6,
-						&iface->addr6_list,
-						list) {
-				if (!addr6->valid || iter_addr6++ < skip_addr6)
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+			netlbl_af6list_foreach_rcu(addr6,
+						   &iface->addr6_list) {
+				if (iter_addr6++ < skip_addr6)
 					continue;
 				if (netlbl_unlabel_staticlist_gen(
-						     NLBL_UNLABEL_C_STATICLIST,
-						     iface,
-						     NULL,
-						     addr6,
-						     &cb_arg) < 0) {
+					      NLBL_UNLABEL_C_STATICLIST,
+					      iface,
+					      NULL,
+					      netlbl_unlhsh_addr6_entry(addr6),
+					      &cb_arg) < 0) {
 					iter_addr6--;
 					iter_chain--;
 					goto unlabel_staticlist_return;
 				}
 			}
+#endif /* IPv6 */
 		}
 	}
 
@@ -1504,9 +1347,12 @@ static int netlbl_unlabel_staticlistdef(struct sk_buff *skb,
 	struct netlbl_unlhsh_iface *iface;
 	u32 skip_addr4 = cb->args[0];
 	u32 skip_addr6 = cb->args[1];
-	u32 iter_addr4 = 0, iter_addr6 = 0;
-	struct netlbl_unlhsh_addr4 *addr4;
-	struct netlbl_unlhsh_addr6 *addr6;
+	u32 iter_addr4 = 0;
+	struct netlbl_af4list *addr4;
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	u32 iter_addr6 = 0;
+	struct netlbl_af6list *addr6;
+#endif
 
 	cb_arg.nl_cb = cb;
 	cb_arg.skb = skb;
@@ -1517,30 +1363,32 @@ static int netlbl_unlabel_staticlistdef(struct sk_buff *skb,
 	if (iface == NULL || !iface->valid)
 		goto unlabel_staticlistdef_return;
 
-	list_for_each_entry_rcu(addr4, &iface->addr4_list, list) {
-		if (!addr4->valid || iter_addr4++ < skip_addr4)
+	netlbl_af4list_foreach_rcu(addr4, &iface->addr4_list) {
+		if (iter_addr4++ < skip_addr4)
 			continue;
 		if (netlbl_unlabel_staticlist_gen(NLBL_UNLABEL_C_STATICLISTDEF,
-					   iface,
-					   addr4,
-					   NULL,
-					   &cb_arg) < 0) {
+					      iface,
+					      netlbl_unlhsh_addr4_entry(addr4),
+					      NULL,
+					      &cb_arg) < 0) {
 			iter_addr4--;
 			goto unlabel_staticlistdef_return;
 		}
 	}
-	list_for_each_entry_rcu(addr6, &iface->addr6_list, list) {
-		if (!addr6->valid || iter_addr6++ < skip_addr6)
+#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
+	netlbl_af6list_foreach_rcu(addr6, &iface->addr6_list) {
+		if (iter_addr6++ < skip_addr6)
 			continue;
 		if (netlbl_unlabel_staticlist_gen(NLBL_UNLABEL_C_STATICLISTDEF,
-					   iface,
-					   NULL,
-					   addr6,
-					   &cb_arg) < 0) {
+					      iface,
+					      NULL,
+					      netlbl_unlhsh_addr6_entry(addr6),
+					      &cb_arg) < 0) {
 			iter_addr6--;
 			goto unlabel_staticlistdef_return;
 		}
 	}
+#endif /* IPv6 */
 
 unlabel_staticlistdef_return:
 	rcu_read_unlock();
@@ -1718,25 +1566,27 @@ int netlbl_unlabel_getattr(const struct sk_buff *skb,
 	switch (family) {
 	case PF_INET: {
 		struct iphdr *hdr4;
-		struct netlbl_unlhsh_addr4 *addr4;
+		struct netlbl_af4list *addr4;
 
 		hdr4 = ip_hdr(skb);
-		addr4 = netlbl_unlhsh_search_addr4(hdr4->saddr, iface);
+		addr4 = netlbl_af4list_search(hdr4->saddr,
+					      &iface->addr4_list);
 		if (addr4 == NULL)
 			goto unlabel_getattr_nolabel;
-		secattr->attr.secid = addr4->secid;
+		secattr->attr.secid = netlbl_unlhsh_addr4_entry(addr4)->secid;
 		break;
 	}
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
 	case PF_INET6: {
 		struct ipv6hdr *hdr6;
-		struct netlbl_unlhsh_addr6 *addr6;
+		struct netlbl_af6list *addr6;
 
 		hdr6 = ipv6_hdr(skb);
-		addr6 = netlbl_unlhsh_search_addr6(&hdr6->saddr, iface);
+		addr6 = netlbl_af6list_search(&hdr6->saddr,
+					      &iface->addr6_list);
 		if (addr6 == NULL)
 			goto unlabel_getattr_nolabel;
-		secattr->attr.secid = addr6->secid;
+		secattr->attr.secid = netlbl_unlhsh_addr6_entry(addr6)->secid;
 		break;
 	}
 #endif /* IPv6 */
diff --git a/scripts/Kbuild.include b/scripts/Kbuild.include
index d64e6badc942..982dcae7bbe2 100644
--- a/scripts/Kbuild.include
+++ b/scripts/Kbuild.include
@@ -105,12 +105,12 @@ as-instr = $(call try-run,\
 # Usage: cflags-y += $(call cc-option,-march=winchip-c6,-march=i586)
 
 cc-option = $(call try-run,\
-	$(CC) $(KBUILD_CFLAGS) $(1) -S -xc /dev/null -o "$$TMP",$(1),$(2))
+	$(CC) $(KBUILD_CFLAGS) $(1) -c -xc /dev/null -o "$$TMP",$(1),$(2))
 
 # cc-option-yn
 # Usage: flag := $(call cc-option-yn,-march=winchip-c6)
 cc-option-yn = $(call try-run,\
-	$(CC) $(KBUILD_CFLAGS) $(1) -S -xc /dev/null -o "$$TMP",y,n)
+	$(CC) $(KBUILD_CFLAGS) $(1) -c -xc /dev/null -o "$$TMP",y,n)
 
 # cc-option-align
 # Prefix align with either -falign or -malign
diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c
index 4c9890ec2528..473f94e56ead 100644
--- a/scripts/mod/file2alias.c
+++ b/scripts/mod/file2alias.c
@@ -629,6 +629,59 @@ static int do_i2c_entry(const char *filename, struct i2c_device_id *id,
 	return 1;
 }
 
+static const struct dmifield {
+	const char *prefix;
+	int field;
+} dmi_fields[] = {
+	{ "bvn", DMI_BIOS_VENDOR },
+	{ "bvr", DMI_BIOS_VERSION },
+	{ "bd",  DMI_BIOS_DATE },
+	{ "svn", DMI_SYS_VENDOR },
+	{ "pn",  DMI_PRODUCT_NAME },
+	{ "pvr", DMI_PRODUCT_VERSION },
+	{ "rvn", DMI_BOARD_VENDOR },
+	{ "rn",  DMI_BOARD_NAME },
+	{ "rvr", DMI_BOARD_VERSION },
+	{ "cvn", DMI_CHASSIS_VENDOR },
+	{ "ct",  DMI_CHASSIS_TYPE },
+	{ "cvr", DMI_CHASSIS_VERSION },
+	{ NULL,  DMI_NONE }
+};
+
+static void dmi_ascii_filter(char *d, const char *s)
+{
+	/* Filter out characters we don't want to see in the modalias string */
+	for (; *s; s++)
+		if (*s > ' ' && *s < 127 && *s != ':')
+			*(d++) = *s;
+
+	*d = 0;
+}
+
+
+static int do_dmi_entry(const char *filename, struct dmi_system_id *id,
+			char *alias)
+{
+	int i, j;
+
+	sprintf(alias, "dmi*");
+
+	for (i = 0; i < ARRAY_SIZE(dmi_fields); i++) {
+		for (j = 0; j < 4; j++) {
+			if (id->matches[j].slot &&
+			    id->matches[j].slot == dmi_fields[i].field) {
+				sprintf(alias + strlen(alias), ":%s*",
+					dmi_fields[i].prefix);
+				dmi_ascii_filter(alias + strlen(alias),
+						 id->matches[j].substr);
+				strcat(alias, "*");
+			}
+		}
+	}
+
+	strcat(alias, ":");
+	return 1;
+}
 /* Ignore any prefix, eg. some architectures prepend _ */
 static inline int sym_is(const char *symbol, const char *name)
 {
@@ -760,6 +813,10 @@ void handle_moddevtable(struct module *mod, struct elf_info *info,
 		do_table(symval, sym->st_size,
 			 sizeof(struct i2c_device_id), "i2c",
 			 do_i2c_entry, mod);
+	else if (sym_is(symname, "__mod_dmi_device_table"))
+		do_table(symval, sym->st_size,
+			 sizeof(struct dmi_system_id), "dmi",
+			 do_dmi_entry, mod);
 	free(zeros);
 }
 
diff --git a/security/inode.c b/security/inode.c
index ca4958ebad8d..efea5a605466 100644
--- a/security/inode.c
+++ b/security/inode.c
@@ -20,8 +20,7 @@
 #include <linux/init.h>
 #include <linux/namei.h>
 #include <linux/security.h>
-
-#define SECURITYFS_MAGIC	0x73636673
+#include <linux/magic.h>
 
 static struct vfsmount *mount;
 static int mount_count;
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 4a7374c12d9c..88f19536efad 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -291,6 +291,7 @@ static void sk_free_security(struct sock *sk)
 	struct sk_security_struct *ssec = sk->sk_security;
 
 	sk->sk_security = NULL;
+	selinux_netlbl_sk_security_free(ssec);
 	kfree(ssec);
 }
 
@@ -2121,7 +2122,6 @@ static inline void flush_unauthorized_files(struct files_struct *files)
 	long j = -1;
 	int drop_tty = 0;
 
-	mutex_lock(&tty_mutex);
 	tty = get_current_tty();
 	if (tty) {
 		file_list_lock();
@@ -2139,8 +2139,8 @@ static inline void flush_unauthorized_files(struct files_struct *files)
 			}
 		}
 		file_list_unlock();
+		tty_kref_put(tty);
 	}
-	mutex_unlock(&tty_mutex);
 	/* Reset controlling tty. */
 	if (drop_tty)
 		no_tty();
@@ -3801,6 +3801,7 @@ out:
 
 static int selinux_socket_connect(struct socket *sock, struct sockaddr *address, int addrlen)
 {
+	struct sock *sk = sock->sk;
 	struct inode_security_struct *isec;
 	int err;
 
@@ -3814,7 +3815,6 @@ static int selinux_socket_connect(struct socket *sock, struct sockaddr *address,
 	isec = SOCK_INODE(sock)->i_security;
 	if (isec->sclass == SECCLASS_TCP_SOCKET ||
 	    isec->sclass == SECCLASS_DCCP_SOCKET) {
-		struct sock *sk = sock->sk;
 		struct avc_audit_data ad;
 		struct sockaddr_in *addr4 = NULL;
 		struct sockaddr_in6 *addr6 = NULL;
@@ -3848,6 +3848,8 @@ static int selinux_socket_connect(struct socket *sock, struct sockaddr *address,
 			goto out;
 	}
 
+	err = selinux_netlbl_socket_connect(sk, address);
+
 out:
 	return err;
 }
@@ -4077,20 +4079,28 @@ static int selinux_sock_rcv_skb_iptables_compat(struct sock *sk,
 }
 
 static int selinux_sock_rcv_skb_compat(struct sock *sk, struct sk_buff *skb,
-				       struct avc_audit_data *ad,
-				       u16 family, char *addrp)
+				       u16 family)
 {
 	int err;
 	struct sk_security_struct *sksec = sk->sk_security;
 	u32 peer_sid;
 	u32 sk_sid = sksec->sid;
+	struct avc_audit_data ad;
+	char *addrp;
+
+	AVC_AUDIT_DATA_INIT(&ad, NET);
+	ad.u.net.netif = skb->iif;
+	ad.u.net.family = family;
+	err = selinux_parse_skb(skb, &ad, &addrp, 1, NULL);
+	if (err)
+		return err;
 
 	if (selinux_compat_net)
-		err = selinux_sock_rcv_skb_iptables_compat(sk, skb, ad,
+		err = selinux_sock_rcv_skb_iptables_compat(sk, skb, &ad,
 							   family, addrp);
 	else
 		err = avc_has_perm(sk_sid, skb->secmark, SECCLASS_PACKET,
-				   PACKET__RECV, ad);
+				   PACKET__RECV, &ad);
 	if (err)
 		return err;
 
@@ -4099,12 +4109,14 @@ static int selinux_sock_rcv_skb_compat(struct sock *sk, struct sk_buff *skb,
 		if (err)
 			return err;
 		err = avc_has_perm(sk_sid, peer_sid,
-				   SECCLASS_PEER, PEER__RECV, ad);
+				   SECCLASS_PEER, PEER__RECV, &ad);
+		if (err)
+			selinux_netlbl_err(skb, err, 0);
 	} else {
-		err = selinux_netlbl_sock_rcv_skb(sksec, skb, family, ad);
+		err = selinux_netlbl_sock_rcv_skb(sksec, skb, family, &ad);
 		if (err)
 			return err;
-		err = selinux_xfrm_sock_rcv_skb(sksec->sid, skb, ad);
+		err = selinux_xfrm_sock_rcv_skb(sksec->sid, skb, &ad);
 	}
 
 	return err;
@@ -4118,6 +4130,8 @@ static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
 	u32 sk_sid = sksec->sid;
 	struct avc_audit_data ad;
 	char *addrp;
+	u8 secmark_active;
+	u8 peerlbl_active;
 
 	if (family != PF_INET && family != PF_INET6)
 		return 0;
@@ -4126,6 +4140,18 @@ static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
 	if (family == PF_INET6 && skb->protocol == htons(ETH_P_IP))
 		family = PF_INET;
 
+	/* If any sort of compatibility mode is enabled then handoff processing
+	 * to the selinux_sock_rcv_skb_compat() function to deal with the
+	 * special handling.  We do this in an attempt to keep this function
+	 * as fast and as clean as possible. */
+	if (selinux_compat_net || !selinux_policycap_netpeer)
+		return selinux_sock_rcv_skb_compat(sk, skb, family);
+
+	secmark_active = selinux_secmark_enabled();
+	peerlbl_active = netlbl_enabled() || selinux_xfrm_enabled();
+	if (!secmark_active && !peerlbl_active)
+		return 0;
+
 	AVC_AUDIT_DATA_INIT(&ad, NET);
 	ad.u.net.netif = skb->iif;
 	ad.u.net.family = family;
@@ -4133,15 +4159,7 @@ static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
 	if (err)
 		return err;
 
-	/* If any sort of compatibility mode is enabled then handoff processing
-	 * to the selinux_sock_rcv_skb_compat() function to deal with the
-	 * special handling.  We do this in an attempt to keep this function
-	 * as fast and as clean as possible. */
-	if (selinux_compat_net || !selinux_policycap_netpeer)
-		return selinux_sock_rcv_skb_compat(sk, skb, &ad,
-						   family, addrp);
-
-	if (netlbl_enabled() || selinux_xfrm_enabled()) {
+	if (peerlbl_active) {
 		u32 peer_sid;
 
 		err = selinux_skb_peerlbl_sid(skb, family, &peer_sid);
@@ -4149,13 +4167,17 @@ static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
 			return err;
 		err = selinux_inet_sys_rcv_skb(skb->iif, addrp, family,
 					       peer_sid, &ad);
-		if (err)
+		if (err) {
+			selinux_netlbl_err(skb, err, 0);
 			return err;
+		}
 		err = avc_has_perm(sk_sid, peer_sid, SECCLASS_PEER,
 				   PEER__RECV, &ad);
+		if (err)
+			selinux_netlbl_err(skb, err, 0);
 	}
 
-	if (selinux_secmark_enabled()) {
+	if (secmark_active) {
 		err = avc_has_perm(sk_sid, skb->secmark, SECCLASS_PACKET,
 				   PACKET__RECV, &ad);
 		if (err)
@@ -4214,10 +4236,12 @@ static int selinux_socket_getpeersec_dgram(struct socket *sock, struct sk_buff *
 	u32 peer_secid = SECSID_NULL;
 	u16 family;
 
-	if (sock)
+	if (skb && skb->protocol == htons(ETH_P_IP))
+		family = PF_INET;
+	else if (skb && skb->protocol == htons(ETH_P_IPV6))
+		family = PF_INET6;
+	else if (sock)
 		family = sock->sk->sk_family;
-	else if (skb && skb->sk)
-		family = skb->sk->sk_family;
 	else
 		goto out;
 
@@ -4275,8 +4299,6 @@ static void selinux_sock_graft(struct sock *sk, struct socket *parent)
 	    sk->sk_family == PF_UNIX)
 		isec->sid = sksec->sid;
 	sksec->sclass = isec->sclass;
-
-	selinux_netlbl_sock_graft(sk, parent);
 }
 
 static int selinux_inet_conn_request(struct sock *sk, struct sk_buff *skb,
@@ -4284,10 +4306,15 @@ static int selinux_inet_conn_request(struct sock *sk, struct sk_buff *skb,
 {
 	struct sk_security_struct *sksec = sk->sk_security;
 	int err;
+	u16 family = sk->sk_family;
 	u32 newsid;
 	u32 peersid;
 
-	err = selinux_skb_peerlbl_sid(skb, sk->sk_family, &peersid);
+	/* handle mapped IPv4 packets arriving via IPv6 sockets */
+	if (family == PF_INET6 && skb->protocol == htons(ETH_P_IP))
+		family = PF_INET;
+
+	err = selinux_skb_peerlbl_sid(skb, family, &peersid);
 	if (err)
 		return err;
 	if (peersid == SECSID_NULL) {
@@ -4322,12 +4349,18 @@ static void selinux_inet_csk_clone(struct sock *newsk,
 	selinux_netlbl_sk_security_reset(newsksec, req->rsk_ops->family);
 }
 
-static void selinux_inet_conn_established(struct sock *sk,
-				struct sk_buff *skb)
+static void selinux_inet_conn_established(struct sock *sk, struct sk_buff *skb)
 {
+	u16 family = sk->sk_family;
 	struct sk_security_struct *sksec = sk->sk_security;
 
-	selinux_skb_peerlbl_sid(skb, sk->sk_family, &sksec->peer_sid);
+	/* handle mapped IPv4 packets arriving via IPv6 sockets */
+	if (family == PF_INET6 && skb->protocol == htons(ETH_P_IP))
+		family = PF_INET;
+
+	selinux_skb_peerlbl_sid(skb, family, &sksec->peer_sid);
+
+	selinux_netlbl_inet_conn_established(sk, family);
 }
 
 static void selinux_req_classify_flow(const struct request_sock *req,
@@ -4377,39 +4410,54 @@ out:
 static unsigned int selinux_ip_forward(struct sk_buff *skb, int ifindex,
 				       u16 family)
 {
+	int err;
 	char *addrp;
 	u32 peer_sid;
 	struct avc_audit_data ad;
 	u8 secmark_active;
+	u8 netlbl_active;
 	u8 peerlbl_active;
 
 	if (!selinux_policycap_netpeer)
 		return NF_ACCEPT;
 
 	secmark_active = selinux_secmark_enabled();
-	peerlbl_active = netlbl_enabled() || selinux_xfrm_enabled();
+	netlbl_active = netlbl_enabled();
+	peerlbl_active = netlbl_active || selinux_xfrm_enabled();
 	if (!secmark_active && !peerlbl_active)
 		return NF_ACCEPT;
 
+	if (selinux_skb_peerlbl_sid(skb, family, &peer_sid) != 0)
+		return NF_DROP;
+
 	AVC_AUDIT_DATA_INIT(&ad, NET);
 	ad.u.net.netif = ifindex;
 	ad.u.net.family = family;
 	if (selinux_parse_skb(skb, &ad, &addrp, 1, NULL) != 0)
 		return NF_DROP;
 
-	if (selinux_skb_peerlbl_sid(skb, family, &peer_sid) != 0)
-		return NF_DROP;
-
-	if (peerlbl_active)
-		if (selinux_inet_sys_rcv_skb(ifindex, addrp, family,
-					     peer_sid, &ad) != 0)
+	if (peerlbl_active) {
+		err = selinux_inet_sys_rcv_skb(ifindex, addrp, family,
+					       peer_sid, &ad);
+		if (err) {
+			selinux_netlbl_err(skb, err, 1);
 			return NF_DROP;
+		}
+	}
 
 	if (secmark_active)
 		if (avc_has_perm(peer_sid, skb->secmark,
 				 SECCLASS_PACKET, PACKET__FORWARD_IN, &ad))
 			return NF_DROP;
 
+	if (netlbl_active)
+		/* we do this in the FORWARD path and not the POST_ROUTING
+		 * path because we want to make sure we apply the necessary
+		 * labeling before IPsec is applied so we can leverage AH
+		 * protection */
+		if (selinux_netlbl_skbuff_setsid(skb, family, peer_sid) != 0)
+			return NF_DROP;
+
 	return NF_ACCEPT;
 }
 
@@ -4433,6 +4481,37 @@ static unsigned int selinux_ipv6_forward(unsigned int hooknum,
 }
 #endif	/* IPV6 */
 
+static unsigned int selinux_ip_output(struct sk_buff *skb,
+				      u16 family)
+{
+	u32 sid;
+
+	if (!netlbl_enabled())
+		return NF_ACCEPT;
+
+	/* we do this in the LOCAL_OUT path and not the POST_ROUTING path
+	 * because we want to make sure we apply the necessary labeling
+	 * before IPsec is applied so we can leverage AH protection */
+	if (skb->sk) {
+		struct sk_security_struct *sksec = skb->sk->sk_security;
+		sid = sksec->sid;
+	} else
+		sid = SECINITSID_KERNEL;
+	if (selinux_netlbl_skbuff_setsid(skb, family, sid) != 0)
+		return NF_DROP;
+
+	return NF_ACCEPT;
+}
+
+static unsigned int selinux_ipv4_output(unsigned int hooknum,
+					struct sk_buff *skb,
+					const struct net_device *in,
+					const struct net_device *out,
+					int (*okfn)(struct sk_buff *))
+{
+	return selinux_ip_output(skb, PF_INET);
+}
+
 static int selinux_ip_postroute_iptables_compat(struct sock *sk,
 						int ifindex,
 						struct avc_audit_data *ad,
@@ -4500,30 +4579,36 @@ static int selinux_ip_postroute_iptables_compat(struct sock *sk,
 
 static unsigned int selinux_ip_postroute_compat(struct sk_buff *skb,
 						int ifindex,
-						struct avc_audit_data *ad,
-						u16 family,
-						char *addrp,
-						u8 proto)
+						u16 family)
 {
 	struct sock *sk = skb->sk;
 	struct sk_security_struct *sksec;
+	struct avc_audit_data ad;
+	char *addrp;
+	u8 proto;
 
 	if (sk == NULL)
 		return NF_ACCEPT;
 	sksec = sk->sk_security;
 
+	AVC_AUDIT_DATA_INIT(&ad, NET);
+	ad.u.net.netif = ifindex;
+	ad.u.net.family = family;
+	if (selinux_parse_skb(skb, &ad, &addrp, 0, &proto))
+		return NF_DROP;
+
 	if (selinux_compat_net) {
 		if (selinux_ip_postroute_iptables_compat(skb->sk, ifindex,
-							 ad, family, addrp))
+							 &ad, family, addrp))
 			return NF_DROP;
 	} else {
 		if (avc_has_perm(sksec->sid, skb->secmark,
-				 SECCLASS_PACKET, PACKET__SEND, ad))
+				 SECCLASS_PACKET, PACKET__SEND, &ad))
 			return NF_DROP;
 	}
 
 	if (selinux_policycap_netpeer)
-		if (selinux_xfrm_postroute_last(sksec->sid, skb, ad, proto))
+		if (selinux_xfrm_postroute_last(sksec->sid, skb, &ad, proto))
 			return NF_DROP;
 
 	return NF_ACCEPT;
@@ -4537,23 +4622,15 @@ static unsigned int selinux_ip_postroute(struct sk_buff *skb, int ifindex,
 	struct sock *sk;
 	struct avc_audit_data ad;
 	char *addrp;
-	u8 proto;
 	u8 secmark_active;
 	u8 peerlbl_active;
 
-	AVC_AUDIT_DATA_INIT(&ad, NET);
-	ad.u.net.netif = ifindex;
-	ad.u.net.family = family;
-	if (selinux_parse_skb(skb, &ad, &addrp, 0, &proto))
-		return NF_DROP;
-
 	/* If any sort of compatibility mode is enabled then handoff processing
 	 * to the selinux_ip_postroute_compat() function to deal with the
 	 * special handling.  We do this in an attempt to keep this function
 	 * as fast and as clean as possible. */
 	if (selinux_compat_net || !selinux_policycap_netpeer)
-		return selinux_ip_postroute_compat(skb, ifindex, &ad,
-						   family, addrp, proto);
+		return selinux_ip_postroute_compat(skb, ifindex, family);
 
 	/* If skb->dst->xfrm is non-NULL then the packet is undergoing an IPsec
 	 * packet transformation so allow the packet to pass without any checks
@@ -4569,21 +4646,45 @@ static unsigned int selinux_ip_postroute(struct sk_buff *skb, int ifindex,
 	if (!secmark_active && !peerlbl_active)
 		return NF_ACCEPT;
 
-	/* if the packet is locally generated (skb->sk != NULL) then use the
-	 * socket's label as the peer label, otherwise the packet is being
-	 * forwarded through this system and we need to fetch the peer label
-	 * directly from the packet */
+	/* if the packet is being forwarded then get the peer label from the
+	 * packet itself; otherwise check to see if it is from a local
+	 * application or the kernel, if from an application get the peer label
+	 * from the sending socket, otherwise use the kernel's sid */
 	sk = skb->sk;
-	if (sk) {
+	if (sk == NULL) {
+		switch (family) {
+		case PF_INET:
+			if (IPCB(skb)->flags & IPSKB_FORWARDED)
+				secmark_perm = PACKET__FORWARD_OUT;
+			else
+				secmark_perm = PACKET__SEND;
+			break;
+		case PF_INET6:
+			if (IP6CB(skb)->flags & IP6SKB_FORWARDED)
+				secmark_perm = PACKET__FORWARD_OUT;
+			else
+				secmark_perm = PACKET__SEND;
+			break;
+		default:
+			return NF_DROP;
+		}
+		if (secmark_perm == PACKET__FORWARD_OUT) {
+			if (selinux_skb_peerlbl_sid(skb, family, &peer_sid))
+				return NF_DROP;
+		} else
+			peer_sid = SECINITSID_KERNEL;
+	} else {
 		struct sk_security_struct *sksec = sk->sk_security;
 		peer_sid = sksec->sid;
 		secmark_perm = PACKET__SEND;
-	} else {
-		if (selinux_skb_peerlbl_sid(skb, family, &peer_sid))
-				return NF_DROP;
-		secmark_perm = PACKET__FORWARD_OUT;
 	}
 
+	AVC_AUDIT_DATA_INIT(&ad, NET);
+	ad.u.net.netif = ifindex;
+	ad.u.net.family = family;
+	if (selinux_parse_skb(skb, &ad, &addrp, 0, NULL))
+		return NF_DROP;
+
 	if (secmark_active)
 		if (avc_has_perm(peer_sid, skb->secmark,
 				 SECCLASS_PACKET, secmark_perm, &ad))
@@ -5657,6 +5758,13 @@ static struct nf_hook_ops selinux_ipv4_ops[] = {
 		.pf =		PF_INET,
 		.hooknum =	NF_INET_FORWARD,
 		.priority =	NF_IP_PRI_SELINUX_FIRST,
+	},
+	{
+		.hook =		selinux_ipv4_output,
+		.owner =	THIS_MODULE,
+		.pf =		PF_INET,
+		.hooknum =	NF_INET_LOCAL_OUT,
+		.priority =	NF_IP_PRI_SELINUX_FIRST,
 	}
 };
 
diff --git a/security/selinux/include/netlabel.h b/security/selinux/include/netlabel.h
index 487a7d81fe20..b913c8d06038 100644
--- a/security/selinux/include/netlabel.h
+++ b/security/selinux/include/netlabel.h
@@ -39,6 +39,9 @@
 #ifdef CONFIG_NETLABEL
 void selinux_netlbl_cache_invalidate(void);
 
+void selinux_netlbl_err(struct sk_buff *skb, int error, int gateway);
+
+void selinux_netlbl_sk_security_free(struct sk_security_struct *ssec);
 void selinux_netlbl_sk_security_reset(struct sk_security_struct *ssec,
 				      int family);
 
@@ -46,8 +49,11 @@ int selinux_netlbl_skbuff_getsid(struct sk_buff *skb,
 				 u16 family,
 				 u32 *type,
 				 u32 *sid);
+int selinux_netlbl_skbuff_setsid(struct sk_buff *skb,
+				 u16 family,
+				 u32 sid);
 
-void selinux_netlbl_sock_graft(struct sock *sk, struct socket *sock);
+void selinux_netlbl_inet_conn_established(struct sock *sk, u16 family);
 int selinux_netlbl_socket_post_create(struct socket *sock);
 int selinux_netlbl_inode_permission(struct inode *inode, int mask);
 int selinux_netlbl_sock_rcv_skb(struct sk_security_struct *sksec,
@@ -57,12 +63,27 @@ int selinux_netlbl_sock_rcv_skb(struct sk_security_struct *sksec,
 int selinux_netlbl_socket_setsockopt(struct socket *sock,
 				     int level,
 				     int optname);
+int selinux_netlbl_socket_connect(struct sock *sk, struct sockaddr *addr);
+
 #else
 static inline void selinux_netlbl_cache_invalidate(void)
 {
 	return;
 }
 
+static inline void selinux_netlbl_err(struct sk_buff *skb,
+				      int error,
+				      int gateway)
+{
+	return;
+}
+
+static inline void selinux_netlbl_sk_security_free(
+					       struct sk_security_struct *ssec)
+{
+	return;
+}
+
 static inline void selinux_netlbl_sk_security_reset(
 					       struct sk_security_struct *ssec,
 					       int family)
@@ -79,9 +100,21 @@ static inline int selinux_netlbl_skbuff_getsid(struct sk_buff *skb,
 	*sid = SECSID_NULL;
 	return 0;
 }
+static inline int selinux_netlbl_skbuff_setsid(struct sk_buff *skb,
+					       u16 family,
+					       u32 sid)
+{
+	return 0;
+}
 
-static inline void selinux_netlbl_sock_graft(struct sock *sk,
-					     struct socket *sock)
+static inline int selinux_netlbl_conn_setsid(struct sock *sk,
+					     struct sockaddr *addr)
+{
+	return 0;
+}
+
+static inline void selinux_netlbl_inet_conn_established(struct sock *sk,
+							u16 family)
 {
 	return;
 }
@@ -107,6 +140,11 @@ static inline int selinux_netlbl_socket_setsockopt(struct socket *sock,
 {
 	return 0;
 }
+static inline int selinux_netlbl_socket_connect(struct sock *sk,
+						struct sockaddr *addr)
+{
+	return 0;
+}
 #endif /* CONFIG_NETLABEL */
 
 #endif
diff --git a/security/selinux/include/objsec.h b/security/selinux/include/objsec.h
index 91070ab874ce..f8be8d7fa26d 100644
--- a/security/selinux/include/objsec.h
+++ b/security/selinux/include/objsec.h
@@ -109,16 +109,19 @@ struct netport_security_struct {
 };
 
 struct sk_security_struct {
-	u32 sid;			/* SID of this object */
-	u32 peer_sid;			/* SID of peer */
-	u16 sclass;			/* sock security class */
 #ifdef CONFIG_NETLABEL
 	enum {				/* NetLabel state */
 		NLBL_UNSET = 0,
 		NLBL_REQUIRE,
 		NLBL_LABELED,
+		NLBL_REQSKB,
+		NLBL_CONNLABELED,
 	} nlbl_state;
+	struct netlbl_lsm_secattr *nlbl_secattr; /* NetLabel sec attributes */
 #endif
+	u32 sid;			/* SID of this object */
+	u32 peer_sid;			/* SID of peer */
+	u16 sclass;			/* sock security class */
 };
 
 struct key_security_struct {
diff --git a/security/selinux/netlabel.c b/security/selinux/netlabel.c
index 89b418392f11..f58701a7b728 100644
--- a/security/selinux/netlabel.c
+++ b/security/selinux/netlabel.c
@@ -9,7 +9,7 @@
  */
 
 /*
- * (c) Copyright Hewlett-Packard Development Company, L.P., 2007
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2007, 2008
  *
  * This program is free software;  you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -29,8 +29,12 @@
 
 #include <linux/spinlock.h>
 #include <linux/rcupdate.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
 #include <net/sock.h>
 #include <net/netlabel.h>
+#include <net/ip.h>
+#include <net/ipv6.h>
 
 #include "objsec.h"
 #include "security.h"
@@ -64,32 +68,69 @@ static int selinux_netlbl_sidlookup_cached(struct sk_buff *skb,
 }
 
 /**
+ * selinux_netlbl_sock_genattr - Generate the NetLabel socket secattr
+ * @sk: the socket
+ *
+ * Description:
+ * Generate the NetLabel security attributes for a socket, making full use of
+ * the socket's attribute cache.  Returns a pointer to the security attributes
+ * on success, NULL on failure.
+ *
+ */
+static struct netlbl_lsm_secattr *selinux_netlbl_sock_genattr(struct sock *sk)
+{
+	int rc;
+	struct sk_security_struct *sksec = sk->sk_security;
+	struct netlbl_lsm_secattr *secattr;
+
+	if (sksec->nlbl_secattr != NULL)
+		return sksec->nlbl_secattr;
+
+	secattr = netlbl_secattr_alloc(GFP_ATOMIC);
+	if (secattr == NULL)
+		return NULL;
+	rc = security_netlbl_sid_to_secattr(sksec->sid, secattr);
+	if (rc != 0) {
+		netlbl_secattr_free(secattr);
+		return NULL;
+	}
+	sksec->nlbl_secattr = secattr;
+
+	return secattr;
+}
+
+/**
  * selinux_netlbl_sock_setsid - Label a socket using the NetLabel mechanism
  * @sk: the socket to label
- * @sid: the SID to use
  *
  * Description:
- * Attempt to label a socket using the NetLabel mechanism using the given
- * SID.  Returns zero values on success, negative values on failure.
+ * Attempt to label a socket using the NetLabel mechanism.  Returns zero values
+ * on success, negative values on failure.
  *
  */
-static int selinux_netlbl_sock_setsid(struct sock *sk, u32 sid)
+static int selinux_netlbl_sock_setsid(struct sock *sk)
 {
 	int rc;
 	struct sk_security_struct *sksec = sk->sk_security;
-	struct netlbl_lsm_secattr secattr;
+	struct netlbl_lsm_secattr *secattr;
 
-	netlbl_secattr_init(&secattr);
+	if (sksec->nlbl_state != NLBL_REQUIRE)
+		return 0;
 
-	rc = security_netlbl_sid_to_secattr(sid, &secattr);
-	if (rc != 0)
-		goto sock_setsid_return;
-	rc = netlbl_sock_setattr(sk, &secattr);
-	if (rc == 0)
+	secattr = selinux_netlbl_sock_genattr(sk);
+	if (secattr == NULL)
+		return -ENOMEM;
+	rc = netlbl_sock_setattr(sk, secattr);
+	switch (rc) {
+	case 0:
 		sksec->nlbl_state = NLBL_LABELED;
+		break;
+	case -EDESTADDRREQ:
+		sksec->nlbl_state = NLBL_REQSKB;
+		rc = 0;
+		break;
+	}
 
-sock_setsid_return:
-	netlbl_secattr_destroy(&secattr);
 	return rc;
 }
 
@@ -106,6 +147,38 @@ void selinux_netlbl_cache_invalidate(void)
 }
 
 /**
+ * selinux_netlbl_err - Handle a NetLabel packet error
+ * @skb: the packet
+ * @error: the error code
+ * @gateway: true if host is acting as a gateway, false otherwise
+ *
+ * Description:
+ * When a packet is dropped due to a call to avc_has_perm() pass the error
+ * code to the NetLabel subsystem so any protocol specific processing can be
+ * done.  This is safe to call even if you are unsure if NetLabel labeling is
+ * present on the packet, NetLabel is smart enough to only act when it should.
+ *
+ */
+void selinux_netlbl_err(struct sk_buff *skb, int error, int gateway)
+{
+	netlbl_skbuff_err(skb, error, gateway);
+}
+
+/**
+ * selinux_netlbl_sk_security_free - Free the NetLabel fields
+ * @sssec: the sk_security_struct
+ *
+ * Description:
+ * Free all of the memory in the NetLabel fields of a sk_security_struct.
+ *
+ */
+void selinux_netlbl_sk_security_free(struct sk_security_struct *ssec)
+{
+	if (ssec->nlbl_secattr != NULL)
+		netlbl_secattr_free(ssec->nlbl_secattr);
+}
+
+/**
  * selinux_netlbl_sk_security_reset - Reset the NetLabel fields
  * @ssec: the sk_security_struct
  * @family: the socket family
@@ -163,35 +236,118 @@ int selinux_netlbl_skbuff_getsid(struct sk_buff *skb,
 }
 
 /**
- * selinux_netlbl_sock_graft - Netlabel the new socket
+ * selinux_netlbl_skbuff_setsid - Set the NetLabel on a packet given a sid
+ * @skb: the packet
+ * @family: protocol family
+ * @sid: the SID
+ *
+ * Description
+ * Call the NetLabel mechanism to set the label of a packet using @sid.
+ * Returns zero on auccess, negative values on failure.
+ *
+ */
+int selinux_netlbl_skbuff_setsid(struct sk_buff *skb,
+				 u16 family,
+				 u32 sid)
+{
+	int rc;
+	struct netlbl_lsm_secattr secattr_storage;
+	struct netlbl_lsm_secattr *secattr = NULL;
+	struct sock *sk;
+
+	/* if this is a locally generated packet check to see if it is already
+	 * being labeled by it's parent socket, if it is just exit */
+	sk = skb->sk;
+	if (sk != NULL) {
+		struct sk_security_struct *sksec = sk->sk_security;
+		if (sksec->nlbl_state != NLBL_REQSKB)
+			return 0;
+		secattr = sksec->nlbl_secattr;
+	}
+	if (secattr == NULL) {
+		secattr = &secattr_storage;
+		netlbl_secattr_init(secattr);
+		rc = security_netlbl_sid_to_secattr(sid, secattr);
+		if (rc != 0)
+			goto skbuff_setsid_return;
+	}
+
+	rc = netlbl_skbuff_setattr(skb, family, secattr);
+
+skbuff_setsid_return:
+	if (secattr == &secattr_storage)
+		netlbl_secattr_destroy(secattr);
+	return rc;
+}
+
+/**
+ * selinux_netlbl_inet_conn_established - Netlabel the newly accepted connection
  * @sk: the new connection
- * @sock: the new socket
  *
  * Description:
- * The connection represented by @sk is being grafted onto @sock so set the
- * socket's NetLabel to match the SID of @sk.
+ * A new connection has been established on @sk so make sure it is labeled
+ * correctly with the NetLabel susbsystem.
  *
  */
-void selinux_netlbl_sock_graft(struct sock *sk, struct socket *sock)
+void selinux_netlbl_inet_conn_established(struct sock *sk, u16 family)
 {
+	int rc;
 	struct sk_security_struct *sksec = sk->sk_security;
-	struct netlbl_lsm_secattr secattr;
-	u32 nlbl_peer_sid;
+	struct netlbl_lsm_secattr *secattr;
+	struct inet_sock *sk_inet = inet_sk(sk);
+	struct sockaddr_in addr;
 
 	if (sksec->nlbl_state != NLBL_REQUIRE)
 		return;
 
-	netlbl_secattr_init(&secattr);
-	if (netlbl_sock_getattr(sk, &secattr) == 0 &&
-	    secattr.flags != NETLBL_SECATTR_NONE &&
-	    security_netlbl_secattr_to_sid(&secattr, &nlbl_peer_sid) == 0)
-		sksec->peer_sid = nlbl_peer_sid;
-	netlbl_secattr_destroy(&secattr);
+	secattr = selinux_netlbl_sock_genattr(sk);
+	if (secattr == NULL)
+		return;
 
-	/* Try to set the NetLabel on the socket to save time later, if we fail
-	 * here we will pick up the pieces in later calls to
-	 * selinux_netlbl_inode_permission(). */
-	selinux_netlbl_sock_setsid(sk, sksec->sid);
+	rc = netlbl_sock_setattr(sk, secattr);
+	switch (rc) {
+	case 0:
+		sksec->nlbl_state = NLBL_LABELED;
+		break;
+	case -EDESTADDRREQ:
+		/* no PF_INET6 support yet because we don't support any IPv6
+		 * labeling protocols */
+		if (family != PF_INET) {
+			sksec->nlbl_state = NLBL_UNSET;
+			return;
+		}
+
+		addr.sin_family = family;
+		addr.sin_addr.s_addr = sk_inet->daddr;
+		if (netlbl_conn_setattr(sk, (struct sockaddr *)&addr,
+					secattr) != 0) {
+			/* we failed to label the connected socket (could be
+			 * for a variety of reasons, the actual "why" isn't
+			 * important here) so we have to go to our backup plan,
+			 * labeling the packets individually in the netfilter
+			 * local output hook.  this is okay but we need to
+			 * adjust the MSS of the connection to take into
+			 * account any labeling overhead, since we don't know
+			 * the exact overhead at this point we'll use the worst
+			 * case value which is 40 bytes for IPv4 */
+			struct inet_connection_sock *sk_conn = inet_csk(sk);
+			sk_conn->icsk_ext_hdr_len += 40 -
+				      (sk_inet->opt ? sk_inet->opt->optlen : 0);
+			sk_conn->icsk_sync_mss(sk, sk_conn->icsk_pmtu_cookie);
+
+			sksec->nlbl_state = NLBL_REQSKB;
+		} else
+			sksec->nlbl_state = NLBL_CONNLABELED;
+		break;
+	default:
+		/* note that we are failing to label the socket which could be
+		 * a bad thing since it means traffic could leave the system
+		 * without the desired labeling, however, all is not lost as
+		 * we have a check in selinux_netlbl_inode_permission() to
+		 * pick up the pieces that we might drop here because we can't
+		 * return an error code */
+		break;
+	}
 }
 
 /**
@@ -205,13 +361,7 @@ void selinux_netlbl_sock_graft(struct sock *sk, struct socket *sock)
  */
 int selinux_netlbl_socket_post_create(struct socket *sock)
 {
-	struct sock *sk = sock->sk;
-	struct sk_security_struct *sksec = sk->sk_security;
-
-	if (sksec->nlbl_state != NLBL_REQUIRE)
-		return 0;
-
-	return selinux_netlbl_sock_setsid(sk, sksec->sid);
+	return selinux_netlbl_sock_setsid(sock->sk);
 }
 
 /**
@@ -246,7 +396,7 @@ int selinux_netlbl_inode_permission(struct inode *inode, int mask)
 	local_bh_disable();
 	bh_lock_sock_nested(sk);
 	if (likely(sksec->nlbl_state == NLBL_REQUIRE))
-		rc = selinux_netlbl_sock_setsid(sk, sksec->sid);
+		rc = selinux_netlbl_sock_setsid(sk);
 	else
 		rc = 0;
 	bh_unlock_sock(sk);
@@ -307,7 +457,7 @@ int selinux_netlbl_sock_rcv_skb(struct sk_security_struct *sksec,
 		return 0;
 
 	if (nlbl_sid != SECINITSID_UNLABELED)
-		netlbl_skbuff_err(skb, rc);
+		netlbl_skbuff_err(skb, rc, 0);
 	return rc;
 }
 
@@ -334,7 +484,8 @@ int selinux_netlbl_socket_setsockopt(struct socket *sock,
 	struct netlbl_lsm_secattr secattr;
 
 	if (level == IPPROTO_IP && optname == IP_OPTIONS &&
-	    sksec->nlbl_state == NLBL_LABELED) {
+	    (sksec->nlbl_state == NLBL_LABELED ||
+	     sksec->nlbl_state == NLBL_CONNLABELED)) {
 		netlbl_secattr_init(&secattr);
 		lock_sock(sk);
 		rc = netlbl_sock_getattr(sk, &secattr);
@@ -346,3 +497,50 @@ int selinux_netlbl_socket_setsockopt(struct socket *sock,
 
 	return rc;
 }
+
+/**
+ * selinux_netlbl_socket_connect - Label a client-side socket on connect
+ * @sk: the socket to label
+ * @addr: the destination address
+ *
+ * Description:
+ * Attempt to label a connected socket with NetLabel using the given address.
+ * Returns zero values on success, negative values on failure.
+ *
+ */
+int selinux_netlbl_socket_connect(struct sock *sk, struct sockaddr *addr)
+{
+	int rc;
+	struct sk_security_struct *sksec = sk->sk_security;
+	struct netlbl_lsm_secattr *secattr;
+
+	if (sksec->nlbl_state != NLBL_REQSKB &&
+	    sksec->nlbl_state != NLBL_CONNLABELED)
+		return 0;
+
+	local_bh_disable();
+	bh_lock_sock_nested(sk);
+
+	/* connected sockets are allowed to disconnect when the address family
+	 * is set to AF_UNSPEC, if that is what is happening we want to reset
+	 * the socket */
+	if (addr->sa_family == AF_UNSPEC) {
+		netlbl_sock_delattr(sk);
+		sksec->nlbl_state = NLBL_REQSKB;
+		rc = 0;
+		goto socket_connect_return;
+	}
+	secattr = selinux_netlbl_sock_genattr(sk);
+	if (secattr == NULL) {
+		rc = -ENOMEM;
+		goto socket_connect_return;
+	}
+	rc = netlbl_conn_setattr(sk, addr, secattr);
+	if (rc == 0)
+		sksec->nlbl_state = NLBL_CONNLABELED;
+
+socket_connect_return:
+	bh_unlock_sock(sk);
+	local_bh_enable();
+	return rc;
+}
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index ab0cc0c7b944..343c8ab14af0 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -2955,7 +2955,7 @@ netlbl_secattr_to_sid_return_cleanup:
  */
 int security_netlbl_sid_to_secattr(u32 sid, struct netlbl_lsm_secattr *secattr)
 {
-	int rc = -ENOENT;
+	int rc;
 	struct context *ctx;
 
 	if (!ss_initialized)
@@ -2963,11 +2963,18 @@ int security_netlbl_sid_to_secattr(u32 sid, struct netlbl_lsm_secattr *secattr)
 
 	read_lock(&policy_rwlock);
 	ctx = sidtab_search(&sidtab, sid);
-	if (ctx == NULL)
+	if (ctx == NULL) {
+		rc = -ENOENT;
 		goto netlbl_sid_to_secattr_failure;
+	}
 	secattr->domain = kstrdup(policydb.p_type_val_to_name[ctx->type - 1],
 				  GFP_ATOMIC);
-	secattr->flags |= NETLBL_SECATTR_DOMAIN_CPY;
+	if (secattr->domain == NULL) {
+		rc = -ENOMEM;
+		goto netlbl_sid_to_secattr_failure;
+	}
+	secattr->attr.secid = sid;
+	secattr->flags |= NETLBL_SECATTR_DOMAIN_CPY | NETLBL_SECATTR_SECID;
 	mls_export_netlbl_lvl(ctx, secattr);
 	rc = mls_export_netlbl_cat(ctx, secattr);
 	if (rc != 0)
diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c
index 87d75417ea93..6e2dc0bab70d 100644
--- a/security/smack/smack_lsm.c
+++ b/security/smack/smack_lsm.c
@@ -2179,7 +2179,10 @@ static int smack_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb)
 	 * This is the simplist possible security model
 	 * for networking.
 	 */
-	return smk_access(smack, ssp->smk_in, MAY_WRITE);
+	rc = smk_access(smack, ssp->smk_in, MAY_WRITE);
+	if (rc != 0)
+		netlbl_skbuff_err(skb, rc, 0);
+	return rc;
 }
 
 /**
diff --git a/security/smack/smackfs.c b/security/smack/smackfs.c
index e7c642458ec9..c21d8c8bf0c7 100644
--- a/security/smack/smackfs.c
+++ b/security/smack/smackfs.c
@@ -354,9 +354,11 @@ static void smk_cipso_doi(void)
 		doip->tags[rc] = CIPSO_V4_TAG_INVALID;
 
 	rc = netlbl_cfg_cipsov4_add_map(doip, NULL, &audit_info);
-	if (rc != 0)
+	if (rc != 0) {
 		printk(KERN_WARNING "%s:%d add rc = %d\n",
 		       __func__, __LINE__, rc);
+		kfree(doip);
+	}
 }
 
 /**
diff --git a/sound/core/memalloc.c b/sound/core/memalloc.c
index a7b46ec72f32..1b3534d67686 100644
--- a/sound/core/memalloc.c
+++ b/sound/core/memalloc.c
@@ -33,9 +33,6 @@
 #include <linux/moduleparam.h>
 #include <linux/mutex.h>
 #include <sound/memalloc.h>
-#ifdef CONFIG_SBUS
-#include <asm/sbus.h>
-#endif
 
 
 MODULE_AUTHOR("Takashi Iwai <tiwai@suse.de>, Jaroslav Kysela <perex@perex.cz>");
@@ -162,39 +159,6 @@ static void snd_free_dev_pages(struct device *dev, size_t size, void *ptr,
 }
 #endif /* CONFIG_HAS_DMA */
 
-#ifdef CONFIG_SBUS
-
-static void *snd_malloc_sbus_pages(struct device *dev, size_t size,
-				   dma_addr_t *dma_addr)
-{
-	struct sbus_dev *sdev = (struct sbus_dev *)dev;
-	int pg;
-	void *res;
-
-	if (WARN_ON(!dma_addr))
-		return NULL;
-	pg = get_order(size);
-	res = sbus_alloc_consistent(sdev, PAGE_SIZE * (1 << pg), dma_addr);
-	if (res != NULL)
-		inc_snd_pages(pg);
-	return res;
-}
-
-static void snd_free_sbus_pages(struct device *dev, size_t size,
-				void *ptr, dma_addr_t dma_addr)
-{
-	struct sbus_dev *sdev = (struct sbus_dev *)dev;
-	int pg;
-
-	if (ptr == NULL)
-		return;
-	pg = get_order(size);
-	dec_snd_pages(pg);
-	sbus_free_consistent(sdev, PAGE_SIZE * (1 << pg), ptr, dma_addr);
-}
-
-#endif /* CONFIG_SBUS */
-
 /*
  *
  *  ALSA generic memory management
@@ -231,11 +195,6 @@ int snd_dma_alloc_pages(int type, struct device *device, size_t size,
 		dmab->area = snd_malloc_pages(size, (unsigned long)device);
 		dmab->addr = 0;
 		break;
-#ifdef CONFIG_SBUS
-	case SNDRV_DMA_TYPE_SBUS:
-		dmab->area = snd_malloc_sbus_pages(device, size, &dmab->addr);
-		break;
-#endif
 #ifdef CONFIG_HAS_DMA
 	case SNDRV_DMA_TYPE_DEV:
 		dmab->area = snd_malloc_dev_pages(device, size, &dmab->addr);
@@ -306,11 +265,6 @@ void snd_dma_free_pages(struct snd_dma_buffer *dmab)
 	case SNDRV_DMA_TYPE_CONTINUOUS:
 		snd_free_pages(dmab->area, dmab->bytes);
 		break;
-#ifdef CONFIG_SBUS
-	case SNDRV_DMA_TYPE_SBUS:
-		snd_free_sbus_pages(dmab->dev.dev, dmab->bytes, dmab->area, dmab->addr);
-		break;
-#endif
 #ifdef CONFIG_HAS_DMA
 	case SNDRV_DMA_TYPE_DEV:
 		snd_free_dev_pages(dmab->dev.dev, dmab->bytes, dmab->area, dmab->addr);
@@ -419,7 +373,7 @@ static int snd_mem_proc_read(struct seq_file *seq, void *offset)
 	long pages = snd_allocated_pages >> (PAGE_SHIFT-12);
 	struct snd_mem_list *mem;
 	int devno;
-	static char *types[] = { "UNKNOWN", "CONT", "DEV", "DEV-SG", "SBUS" };
+	static char *types[] = { "UNKNOWN", "CONT", "DEV", "DEV-SG" };
 
 	mutex_lock(&list_mutex);
 	seq_printf(seq, "pages  : %li bytes (%li pages per %likB)\n",
diff --git a/sound/oss/ac97_codec.c b/sound/oss/ac97_codec.c
index b63839e8f9bd..456a1b4d7832 100644
--- a/sound/oss/ac97_codec.c
+++ b/sound/oss/ac97_codec.c
@@ -30,7 +30,7 @@
  **************************************************************************
  *
  * History
- * May 02, 2003 Liam Girdwood <liam.girdwood@wolfsonmicro.com>
+ * May 02, 2003 Liam Girdwood <lrg@slimlogic.co.uk>
  *	Removed non existant WM9700
  *	Added support for WM9705, WM9708, WM9709, WM9710, WM9711
  *	WM9712 and WM9717
diff --git a/sound/pci/ac97/ac97_patch.c b/sound/pci/ac97/ac97_patch.c
index 6ce3cbe98a6a..6e831aff1bd0 100644
--- a/sound/pci/ac97/ac97_patch.c
+++ b/sound/pci/ac97/ac97_patch.c
@@ -476,7 +476,7 @@ static int patch_yamaha_ymf753(struct snd_ac97 * ac97)
 }
 
 /*
- * May 2, 2003 Liam Girdwood <liam.girdwood@wolfsonmicro.com>
+ * May 2, 2003 Liam Girdwood <lrg@slimlogic.co.uk>
  *  removed broken wolfson00 patch.
  *  added support for WM9705,WM9708,WM9709,WM9710,WM9711,WM9712 and WM9717.
  */
diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c
index c461baa83c2a..c59065513118 100644
--- a/sound/pci/hda/patch_sigmatel.c
+++ b/sound/pci/hda/patch_sigmatel.c
@@ -322,8 +322,8 @@ static hda_nid_t stac92hd71bxx_mux_nids[2] = {
 	0x1a, 0x1b
 };
 
-static hda_nid_t stac92hd71bxx_dmux_nids[1] = {
-	0x1c,
+static hda_nid_t stac92hd71bxx_dmux_nids[2] = {
+	0x1c, 0x1d,
 };
 
 static hda_nid_t stac92hd71bxx_smux_nids[2] = {
@@ -861,20 +861,18 @@ static struct hda_verb stac92hd71bxx_core_init[] = {
 	{ 0x28, AC_VERB_SET_VOLUME_KNOB_CONTROL, 0xff},
 	/* connect headphone jack to dac1 */
 	{ 0x0a, AC_VERB_SET_CONNECT_SEL, 0x01},
-	{ 0x0f, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, /* Speaker */
 	/* unmute right and left channels for nodes 0x0a, 0xd, 0x0f */
 	{ 0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)},
 	{ 0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)},
 	{ 0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)},
 };
 
-#define HD_DISABLE_PORTF 3
+#define HD_DISABLE_PORTF 2
 static struct hda_verb stac92hd71bxx_analog_core_init[] = {
 	/* start of config #1 */
 
 	/* connect port 0f to audio mixer */
 	{ 0x0f, AC_VERB_SET_CONNECT_SEL, 0x2},
-	{ 0x0f, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, /* Speaker */
 	/* unmute right and left channels for node 0x0f */
 	{ 0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)},
 	/* start of config #2 */
@@ -883,10 +881,6 @@ static struct hda_verb stac92hd71bxx_analog_core_init[] = {
 	{ 0x28, AC_VERB_SET_VOLUME_KNOB_CONTROL, 0xff},
 	/* connect headphone jack to dac1 */
 	{ 0x0a, AC_VERB_SET_CONNECT_SEL, 0x01},
-	/* connect port 0d to audio mixer */
-	{ 0x0d, AC_VERB_SET_CONNECT_SEL, 0x2},
-	/* unmute dac0 input in audio mixer */
-	{ 0x17, AC_VERB_SET_AMP_GAIN_MUTE, 0x701f},
 	/* unmute right and left channels for nodes 0x0a, 0xd */
 	{ 0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)},
 	{ 0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)},
@@ -1107,6 +1101,7 @@ static struct snd_kcontrol_new stac92hd83xxx_mixer[] = {
 
 static struct snd_kcontrol_new stac92hd71bxx_analog_mixer[] = {
 	STAC_INPUT_SOURCE(2),
+	STAC_ANALOG_LOOPBACK(0xFA0, 0x7A0, 2),
 
 	HDA_CODEC_VOLUME_IDX("Capture Volume", 0x0, 0x1c, 0x0, HDA_OUTPUT),
 	HDA_CODEC_MUTE_IDX("Capture Switch", 0x0, 0x1c, 0x0, HDA_OUTPUT),
@@ -1119,8 +1114,17 @@ static struct snd_kcontrol_new stac92hd71bxx_analog_mixer[] = {
 	HDA_CODEC_MUTE("PC Beep Switch", 0x17, 0x2, HDA_INPUT),
 	*/
 
-	HDA_CODEC_MUTE("Analog Loopback 1", 0x17, 0x3, HDA_INPUT),
-	HDA_CODEC_MUTE("Analog Loopback 2", 0x17, 0x4, HDA_INPUT),
+	HDA_CODEC_MUTE("Import0 Mux Capture Switch", 0x17, 0x0, HDA_INPUT),
+	HDA_CODEC_VOLUME("Import0 Mux Capture Volume", 0x17, 0x0, HDA_INPUT),
+
+	HDA_CODEC_MUTE("Import1 Mux Capture Switch", 0x17, 0x1, HDA_INPUT),
+	HDA_CODEC_VOLUME("Import1 Mux Capture Volume", 0x17, 0x1, HDA_INPUT),
+
+	HDA_CODEC_MUTE("DAC0 Capture Switch", 0x17, 0x3, HDA_INPUT),
+	HDA_CODEC_VOLUME("DAC0 Capture Volume", 0x17, 0x3, HDA_INPUT),
+
+	HDA_CODEC_MUTE("DAC1 Capture Switch", 0x17, 0x4, HDA_INPUT),
+	HDA_CODEC_VOLUME("DAC1 Capture Volume", 0x17, 0x4, HDA_INPUT),
 	{ } /* end */
 };
 
@@ -1649,7 +1653,7 @@ static struct snd_pci_quirk stac92hd83xxx_cfg_tbl[] = {
 
 static unsigned int ref92hd71bxx_pin_configs[11] = {
 	0x02214030, 0x02a19040, 0x01a19020, 0x01014010,
-	0x0181302e, 0x01114010, 0x01019020, 0x90a000f0,
+	0x0181302e, 0x01014010, 0x01019020, 0x90a000f0,
 	0x90a000f0, 0x01452050, 0x01452050,
 };
 
@@ -3000,7 +3004,7 @@ static int stac92xx_auto_create_mono_output_ctls(struct hda_codec *codec)
 
 /* labels for amp mux outputs */
 static const char *stac92xx_amp_labels[3] = {
-	"Front Microphone", "Microphone", "Line In"
+	"Front Microphone", "Microphone", "Line In",
 };
 
 /* create amp out controls mux on capable codecs */
@@ -4327,6 +4331,16 @@ static struct hda_codec_ops stac92hd71bxx_patch_ops = {
 #endif
 };
 
+static struct hda_input_mux stac92hd71bxx_dmux = {
+	.num_items = 4,
+	.items = {
+		{ "Analog Inputs", 0x00 },
+		{ "Mixer", 0x01 },
+		{ "Digital Mic 1", 0x02 },
+		{ "Digital Mic 2", 0x03 },
+	}
+};
+
 static int patch_stac92hd71bxx(struct hda_codec *codec)
 {
 	struct sigmatel_spec *spec;
@@ -4341,6 +4355,8 @@ static int patch_stac92hd71bxx(struct hda_codec *codec)
 	spec->num_pins = ARRAY_SIZE(stac92hd71bxx_pin_nids);
 	spec->num_pwrs = ARRAY_SIZE(stac92hd71bxx_pwr_nids);
 	spec->pin_nids = stac92hd71bxx_pin_nids;
+	memcpy(&spec->private_dimux, &stac92hd71bxx_dmux,
+			sizeof(stac92hd71bxx_dmux));
 	spec->board_config = snd_hda_check_board_config(codec,
 							STAC_92HD71BXX_MODELS,
 							stac92hd71bxx_models,
@@ -4392,6 +4408,7 @@ again:
 		/* no output amps */
 		spec->num_pwrs = 0;
 		spec->mixer = stac92hd71bxx_analog_mixer;
+		spec->dinput_mux = &spec->private_dimux;
 
 		/* disable VSW */
 		spec->init = &stac92hd71bxx_analog_core_init[HD_DISABLE_PORTF];
@@ -4409,12 +4426,13 @@ again:
 		spec->num_pwrs = 0;
 		/* fallthru */
 	default:
+		spec->dinput_mux = &spec->private_dimux;
 		spec->mixer = stac92hd71bxx_analog_mixer;
 		spec->init = stac92hd71bxx_analog_core_init;
 		codec->slave_dig_outs = stac92hd71bxx_slave_dig_outs;
 	}
 
-	spec->aloopback_mask = 0x20;
+	spec->aloopback_mask = 0x50;
 	spec->aloopback_shift = 0;
 
 	if (spec->board_config > STAC_92HD71BXX_REF) {
@@ -4456,6 +4474,10 @@ again:
 	spec->multiout.num_dacs = 1;
 	spec->multiout.hp_nid = 0x11;
 	spec->multiout.dac_nids = stac92hd71bxx_dac_nids;
+	if (spec->dinput_mux)
+		spec->private_dimux.num_items +=
+			spec->num_dmics -
+				(ARRAY_SIZE(stac92hd71bxx_dmic_nids) - 1);
 
 	err = stac92xx_parse_auto_config(codec, 0x21, 0x23);
 	if (!err) {
diff --git a/sound/soc/at91/Kconfig b/sound/soc/at91/Kconfig
index 905186502e00..85a883299c2e 100644
--- a/sound/soc/at91/Kconfig
+++ b/sound/soc/at91/Kconfig
@@ -8,20 +8,3 @@ config SND_AT91_SOC
 
 config SND_AT91_SOC_SSC
 	tristate
-
-config SND_AT91_SOC_ETI_B1_WM8731
-	tristate "SoC Audio support for WM8731-based Endrelia ETI-B1 boards"
-	depends on SND_AT91_SOC && (MACH_ETI_B1 || MACH_ETI_C1)
-	select SND_AT91_SOC_SSC
-	select SND_SOC_WM8731
-	help
-	  Say Y if you want to add support for SoC audio on WM8731-based
-	  Endrelia Technologies Inc ETI-B1 or ETI-C1 boards.
-
-config SND_AT91_SOC_ETI_SLAVE
-	bool "Run codec in slave Mode on Endrelia boards"
-	depends on SND_AT91_SOC_ETI_B1_WM8731
-	default n
-	help
-	  Say Y if you want to run with the AT91 SSC generating the BCLK
-	  and LRC signals on Endrelia boards.
diff --git a/sound/soc/at91/Makefile b/sound/soc/at91/Makefile
index f23da17cc328..b817f11df286 100644
--- a/sound/soc/at91/Makefile
+++ b/sound/soc/at91/Makefile
@@ -4,8 +4,3 @@ snd-soc-at91-ssc-objs := at91-ssc.o
 
 obj-$(CONFIG_SND_AT91_SOC) += snd-soc-at91.o
 obj-$(CONFIG_SND_AT91_SOC_SSC) += snd-soc-at91-ssc.o
-
-# AT91 Machine Support
-snd-soc-eti-b1-wm8731-objs := eti_b1_wm8731.o
-
-obj-$(CONFIG_SND_AT91_SOC_ETI_B1_WM8731) += snd-soc-eti-b1-wm8731.o
diff --git a/sound/soc/at91/at91-ssc.c b/sound/soc/at91/at91-ssc.c
index a5b1a79ebffb..1b61cc461261 100644
--- a/sound/soc/at91/at91-ssc.c
+++ b/sound/soc/at91/at91-ssc.c
@@ -5,7 +5,7 @@
  *         Endrelia Technologies Inc.
  *
  * Based on pxa2xx Platform drivers by
- * Liam Girdwood <liam.girdwood@wolfsonmicro.com>
+ * Liam Girdwood <lrg@slimlogic.co.uk>
  *
  *  This program is free software; you can redistribute  it and/or modify it
  *  under  the terms of  the GNU General  Public License as published by the
diff --git a/sound/soc/at91/eti_b1_wm8731.c b/sound/soc/at91/eti_b1_wm8731.c
deleted file mode 100644
index 684781e4088b..000000000000
--- a/sound/soc/at91/eti_b1_wm8731.c
+++ /dev/null
@@ -1,349 +0,0 @@
-/*
- * eti_b1_wm8731  --  SoC audio for AT91RM9200-based Endrelia ETI_B1 board.
- *
- * Author:	Frank Mandarino <fmandarino@endrelia.com>
- *		Endrelia Technologies Inc.
- * Created:	Mar 29, 2006
- *
- * Based on corgi.c by:
- *
- * Copyright 2005 Wolfson Microelectronics PLC.
- * Copyright 2005 Openedhand Ltd.
- *
- * Authors: Liam Girdwood <liam.girdwood@wolfsonmicro.com>
- *          Richard Purdie <richard@openedhand.com>
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- */
-
-#include <linux/module.h>
-#include <linux/moduleparam.h>
-#include <linux/kernel.h>
-#include <linux/clk.h>
-#include <linux/timer.h>
-#include <linux/interrupt.h>
-#include <linux/platform_device.h>
-#include <sound/core.h>
-#include <sound/pcm.h>
-#include <sound/soc.h>
-#include <sound/soc-dapm.h>
-
-#include <mach/hardware.h>
-#include <mach/gpio.h>
-
-#include "../codecs/wm8731.h"
-#include "at91-pcm.h"
-#include "at91-ssc.h"
-
-#if 0
-#define	DBG(x...)	printk(KERN_INFO "eti_b1_wm8731: " x)
-#else
-#define	DBG(x...)
-#endif
-
-static struct clk *pck1_clk;
-static struct clk *pllb_clk;
-
-
-static int eti_b1_startup(struct snd_pcm_substream *substream)
-{
-	struct snd_soc_pcm_runtime *rtd = substream->private_data;
-	struct snd_soc_dai *codec_dai = rtd->dai->codec_dai;
-	struct snd_soc_dai *cpu_dai = rtd->dai->cpu_dai;
-	int ret;
-
-	/* cpu clock is the AT91 master clock sent to the SSC */
-	ret = snd_soc_dai_set_sysclk(cpu_dai, AT91_SYSCLK_MCK,
-		60000000, SND_SOC_CLOCK_IN);
-	if (ret < 0)
-		return ret;
-
-	/* codec system clock is supplied by PCK1, set to 12MHz */
-	ret = snd_soc_dai_set_sysclk(codec_dai, WM8731_SYSCLK,
-		12000000, SND_SOC_CLOCK_IN);
-	if (ret < 0)
-		return ret;
-
-	/* Start PCK1 clock. */
-	clk_enable(pck1_clk);
-	DBG("pck1 started\n");
-
-	return 0;
-}
-
-static void eti_b1_shutdown(struct snd_pcm_substream *substream)
-{
-	/* Stop PCK1 clock. */
-	clk_disable(pck1_clk);
-	DBG("pck1 stopped\n");
-}
-
-static int eti_b1_hw_params(struct snd_pcm_substream *substream,
-	struct snd_pcm_hw_params *params)
-{
-	struct snd_soc_pcm_runtime *rtd = substream->private_data;
-	struct snd_soc_dai *codec_dai = rtd->dai->codec_dai;
-	struct snd_soc_dai *cpu_dai = rtd->dai->cpu_dai;
-	int ret;
-
-#ifdef CONFIG_SND_AT91_SOC_ETI_SLAVE
-	unsigned int rate;
-	int cmr_div, period;
-
-	/* set codec DAI configuration */
-	ret = snd_soc_dai_set_fmt(codec_dai, SND_SOC_DAIFMT_I2S |
-		SND_SOC_DAIFMT_NB_NF | SND_SOC_DAIFMT_CBS_CFS);
-	if (ret < 0)
-		return ret;
-
-	/* set cpu DAI configuration */
-	ret = snd_soc_dai_set_fmt(cpu_dai, SND_SOC_DAIFMT_I2S |
-		SND_SOC_DAIFMT_NB_NF | SND_SOC_DAIFMT_CBS_CFS);
-	if (ret < 0)
-		return ret;
-
-	/*
-	 * The SSC clock dividers depend on the sample rate.  The CMR.DIV
-	 * field divides the system master clock MCK to drive the SSC TK
-	 * signal which provides the codec BCLK.  The TCMR.PERIOD and
-	 * RCMR.PERIOD fields further divide the BCLK signal to drive
-	 * the SSC TF and RF signals which provide the codec DACLRC and
-	 * ADCLRC clocks.
-	 *
-	 * The dividers were determined through trial and error, where a
-	 * CMR.DIV value is chosen such that the resulting BCLK value is
-	 * divisible, or almost divisible, by (2 * sample rate), and then
-	 * the TCMR.PERIOD or RCMR.PERIOD is BCLK / (2 * sample rate) - 1.
-	 */
-	rate = params_rate(params);
-
-	switch (rate) {
-	case 8000:
-		cmr_div = 25;	/* BCLK = 60MHz/(2*25) = 1.2MHz */
-		period = 74;	/* LRC = BCLK/(2*(74+1)) = 8000Hz */
-		break;
-	case 32000:
-		cmr_div = 7;	/* BCLK = 60MHz/(2*7) ~= 4.28571428MHz */
-		period = 66;	/* LRC = BCLK/(2*(66+1)) = 31982.942Hz */
-		break;
-	case 48000:
-		cmr_div = 13;	/* BCLK = 60MHz/(2*13) ~= 2.3076923MHz */
-		period = 23;	/* LRC = BCLK/(2*(23+1)) = 48076.923Hz */
-		break;
-	default:
-		printk(KERN_WARNING "unsupported rate %d on ETI-B1 board\n", rate);
-		return -EINVAL;
-	}
-
-	/* set the MCK divider for BCLK */
-	ret = snd_soc_dai_set_clkdiv(cpu_dai, AT91SSC_CMR_DIV, cmr_div);
-	if (ret < 0)
-		return ret;
-
-	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
-		/* set the BCLK divider for DACLRC */
-		ret = snd_soc_dai_set_clkdiv(cpu_dai,
-						AT91SSC_TCMR_PERIOD, period);
-	} else {
-		/* set the BCLK divider for ADCLRC */
-		ret = snd_soc_dai_set_clkdiv(cpu_dai,
-						AT91SSC_RCMR_PERIOD, period);
-	}
-	if (ret < 0)
-		return ret;
-
-#else /* CONFIG_SND_AT91_SOC_ETI_SLAVE */
-	/*
-	 * Codec in Master Mode.
-	 */
-
-	/* set codec DAI configuration */
-	ret = snd_soc_dai_set_fmt(codec_dai, SND_SOC_DAIFMT_I2S |
-		SND_SOC_DAIFMT_NB_NF | SND_SOC_DAIFMT_CBM_CFM);
-	if (ret < 0)
-		return ret;
-
-	/* set cpu DAI configuration */
-	ret = snd_soc_dai_set_fmt(cpu_dai, SND_SOC_DAIFMT_I2S |
-		SND_SOC_DAIFMT_NB_NF | SND_SOC_DAIFMT_CBM_CFM);
-	if (ret < 0)
-		return ret;
-
-#endif /* CONFIG_SND_AT91_SOC_ETI_SLAVE */
-
-	return 0;
-}
-
-static struct snd_soc_ops eti_b1_ops = {
-	.startup = eti_b1_startup,
-	.hw_params = eti_b1_hw_params,
-	.shutdown = eti_b1_shutdown,
-};
-
-
-static const struct snd_soc_dapm_widget eti_b1_dapm_widgets[] = {
-	SND_SOC_DAPM_MIC("Int Mic", NULL),
-	SND_SOC_DAPM_SPK("Ext Spk", NULL),
-};
-
-static const struct snd_soc_dapm_route intercon[] = {
-
-	/* speaker connected to LHPOUT */
-	{"Ext Spk", NULL, "LHPOUT"},
-
-	/* mic is connected to Mic Jack, with WM8731 Mic Bias */
-	{"MICIN", NULL, "Mic Bias"},
-	{"Mic Bias", NULL, "Int Mic"},
-};
-
-/*
- * Logic for a wm8731 as connected on a Endrelia ETI-B1 board.
- */
-static int eti_b1_wm8731_init(struct snd_soc_codec *codec)
-{
-	DBG("eti_b1_wm8731_init() called\n");
-
-	/* Add specific widgets */
-	snd_soc_dapm_new_controls(codec, eti_b1_dapm_widgets,
-				  ARRAY_SIZE(eti_b1_dapm_widgets));
-
-	/* Set up specific audio path interconnects */
-	snd_soc_dapm_add_route(codec, intercon, ARRAY_SIZE(intercon));
-
-	/* not connected */
-	snd_soc_dapm_disable_pin(codec, "RLINEIN");
-	snd_soc_dapm_disable_pin(codec, "LLINEIN");
-
-	/* always connected */
-	snd_soc_dapm_enable_pin(codec, "Int Mic");
-	snd_soc_dapm_enable_pin(codec, "Ext Spk");
-
-	snd_soc_dapm_sync(codec);
-
-	return 0;
-}
-
-static struct snd_soc_dai_link eti_b1_dai = {
-	.name = "WM8731",
-	.stream_name = "WM8731 PCM",
-	.cpu_dai = &at91_ssc_dai[1],
-	.codec_dai = &wm8731_dai,
-	.init = eti_b1_wm8731_init,
-	.ops = &eti_b1_ops,
-};
-
-static struct snd_soc_machine snd_soc_machine_eti_b1 = {
-	.name = "ETI_B1_WM8731",
-	.dai_link = &eti_b1_dai,
-	.num_links = 1,
-};
-
-static struct wm8731_setup_data eti_b1_wm8731_setup = {
-	.i2c_bus = 0,
-	.i2c_address = 0x1a,
-};
-
-static struct snd_soc_device eti_b1_snd_devdata = {
-	.machine = &snd_soc_machine_eti_b1,
-	.platform = &at91_soc_platform,
-	.codec_dev = &soc_codec_dev_wm8731,
-	.codec_data = &eti_b1_wm8731_setup,
-};
-
-static struct platform_device *eti_b1_snd_device;
-
-static int __init eti_b1_init(void)
-{
-	int ret;
-	struct at91_ssc_periph *ssc = eti_b1_dai.cpu_dai->private_data;
-
-	if (!request_mem_region(AT91RM9200_BASE_SSC1, SZ_16K, "soc-audio")) {
-		DBG("SSC1 memory region is busy\n");
-		return -EBUSY;
-	}
-
-	ssc->base = ioremap(AT91RM9200_BASE_SSC1, SZ_16K);
-	if (!ssc->base) {
-		DBG("SSC1 memory ioremap failed\n");
-		ret = -ENOMEM;
-		goto fail_release_mem;
-	}
-
-	ssc->pid = AT91RM9200_ID_SSC1;
-
-	eti_b1_snd_device = platform_device_alloc("soc-audio", -1);
-	if (!eti_b1_snd_device) {
-		DBG("platform device allocation failed\n");
-		ret = -ENOMEM;
-		goto fail_io_unmap;
-	}
-
-	platform_set_drvdata(eti_b1_snd_device, &eti_b1_snd_devdata);
-	eti_b1_snd_devdata.dev = &eti_b1_snd_device->dev;
-
-	ret = platform_device_add(eti_b1_snd_device);
-	if (ret) {
-		DBG("platform device add failed\n");
-		platform_device_put(eti_b1_snd_device);
-		goto fail_io_unmap;
-	}
-
-	at91_set_A_periph(AT91_PIN_PB6, 0);	/* TF1 */
-	at91_set_A_periph(AT91_PIN_PB7, 0);	/* TK1 */
-	at91_set_A_periph(AT91_PIN_PB8, 0);	/* TD1 */
-	at91_set_A_periph(AT91_PIN_PB9, 0);	/* RD1 */
-/*	at91_set_A_periph(AT91_PIN_PB10, 0);*/	/* RK1 */
-	at91_set_A_periph(AT91_PIN_PB11, 0);	/* RF1 */
-
-	/*
-	 * Set PCK1 parent to PLLB and its rate to 12 Mhz.
-	 */
-	pllb_clk = clk_get(NULL, "pllb");
-	pck1_clk = clk_get(NULL, "pck1");
-
-	clk_set_parent(pck1_clk, pllb_clk);
-	clk_set_rate(pck1_clk, 12000000);
-
-	DBG("MCLK rate %luHz\n", clk_get_rate(pck1_clk));
-
-	/* assign the GPIO pin to PCK1 */
-	at91_set_B_periph(AT91_PIN_PA24, 0);
-
-#ifdef CONFIG_SND_AT91_SOC_ETI_SLAVE
-	printk(KERN_INFO "eti_b1_wm8731: Codec in Slave Mode\n");
-#else
-	printk(KERN_INFO "eti_b1_wm8731: Codec in Master Mode\n");
-#endif
-	return ret;
-
-fail_io_unmap:
-	iounmap(ssc->base);
-fail_release_mem:
-	release_mem_region(AT91RM9200_BASE_SSC1, SZ_16K);
-	return ret;
-}
-
-static void __exit eti_b1_exit(void)
-{
-	struct at91_ssc_periph *ssc = eti_b1_dai.cpu_dai->private_data;
-
-	clk_put(pck1_clk);
-	clk_put(pllb_clk);
-
-	platform_device_unregister(eti_b1_snd_device);
-
-	iounmap(ssc->base);
-	release_mem_region(AT91RM9200_BASE_SSC1, SZ_16K);
-}
-
-module_init(eti_b1_init);
-module_exit(eti_b1_exit);
-
-/* Module information */
-MODULE_AUTHOR("Frank Mandarino <fmandarino@endrelia.com>");
-MODULE_DESCRIPTION("ALSA SoC ETI-B1-WM8731");
-MODULE_LICENSE("GPL");
diff --git a/sound/soc/blackfin/Kconfig b/sound/soc/blackfin/Kconfig
index f98331d099e7..dc006206f622 100644
--- a/sound/soc/blackfin/Kconfig
+++ b/sound/soc/blackfin/Kconfig
@@ -17,6 +17,22 @@ config SND_BF5XX_SOC_SSM2602
 	help
 	  Say Y if you want to add support for SoC audio on BF527-EZKIT.
 
+config SND_BF5XX_SOC_AD73311
+	tristate "SoC AD73311 Audio support for Blackfin"
+	depends on SND_BF5XX_I2S
+	select SND_BF5XX_SOC_I2S
+	select SND_SOC_AD73311
+	help
+	  Say Y if you want to add support for AD73311 codec on Blackfin.
+
+config SND_BFIN_AD73311_SE
+	int "PF pin for AD73311L Chip Select"
+	depends on SND_BF5XX_SOC_AD73311
+	default 4
+	help
+	  Enter the GPIO used to control AD73311's SE pin. Acceptable
+	  values are 0 to 7
+
 config SND_BF5XX_AC97
 	tristate "SoC AC97 Audio for the ADI BF5xx chip"
 	depends on BLACKFIN && SND_SOC
diff --git a/sound/soc/blackfin/Makefile b/sound/soc/blackfin/Makefile
index 9ea8bd9e0ba3..97bb37a6359c 100644
--- a/sound/soc/blackfin/Makefile
+++ b/sound/soc/blackfin/Makefile
@@ -14,7 +14,8 @@ obj-$(CONFIG_SND_BF5XX_SOC_I2S) += snd-soc-bf5xx-i2s.o
 # Blackfin Machine Support
 snd-ad1980-objs := bf5xx-ad1980.o
 snd-ssm2602-objs := bf5xx-ssm2602.o
-
+snd-ad73311-objs := bf5xx-ad73311.o
 
 obj-$(CONFIG_SND_BF5XX_SOC_AD1980) += snd-ad1980.o
 obj-$(CONFIG_SND_BF5XX_SOC_SSM2602) += snd-ssm2602.o
+obj-$(CONFIG_SND_BF5XX_SOC_AD73311) += snd-ad73311.o
diff --git a/sound/soc/blackfin/bf5xx-ac97-pcm.c b/sound/soc/blackfin/bf5xx-ac97-pcm.c
index 51f4907c4831..25e50d2ea1ec 100644
--- a/sound/soc/blackfin/bf5xx-ac97-pcm.c
+++ b/sound/soc/blackfin/bf5xx-ac97-pcm.c
@@ -56,6 +56,7 @@ static void bf5xx_mmap_copy(struct snd_pcm_substream *substream,
 		sport->tx_pos += runtime->period_size;
 		if (sport->tx_pos >= runtime->buffer_size)
 			sport->tx_pos %= runtime->buffer_size;
+		sport->tx_delay_pos = sport->tx_pos;
 	} else {
 		bf5xx_ac97_to_pcm(
 			(struct ac97_frame *)sport->rx_dma_buf + sport->rx_pos,
@@ -72,7 +73,15 @@ static void bf5xx_dma_irq(void *data)
 	struct snd_pcm_substream *pcm = data;
 #if defined(CONFIG_SND_MMAP_SUPPORT)
 	struct snd_pcm_runtime *runtime = pcm->runtime;
+	struct sport_device *sport = runtime->private_data;
 	bf5xx_mmap_copy(pcm, runtime->period_size);
+	if (pcm->stream == SNDRV_PCM_STREAM_PLAYBACK) {
+		if (sport->once == 0) {
+			snd_pcm_period_elapsed(pcm);
+			bf5xx_mmap_copy(pcm, runtime->period_size);
+			sport->once = 1;
+		}
+	}
 #endif
 	snd_pcm_period_elapsed(pcm);
 }
@@ -114,6 +123,10 @@ static int bf5xx_pcm_hw_params(struct snd_pcm_substream *substream,
 
 static int bf5xx_pcm_hw_free(struct snd_pcm_substream *substream)
 {
+	struct snd_pcm_runtime *runtime = substream->runtime;
+
+	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+	memset(runtime->dma_area, 0, runtime->buffer_size);
 	snd_pcm_lib_free_pages(substream);
 	return 0;
 }
@@ -127,16 +140,11 @@ static int bf5xx_pcm_prepare(struct snd_pcm_substream *substream)
 	 * SPORT working in TMD mode(include AC97).
 	 */
 #if defined(CONFIG_SND_MMAP_SUPPORT)
-	size_t size = bf5xx_pcm_hardware.buffer_bytes_max
-			* sizeof(struct ac97_frame) / 4;
-	/*clean up intermediate buffer*/
 	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
-		memset(sport->tx_dma_buf, 0, size);
 		sport_set_tx_callback(sport, bf5xx_dma_irq, substream);
 		sport_config_tx_dma(sport, sport->tx_dma_buf, runtime->periods,
 			runtime->period_size * sizeof(struct ac97_frame));
 	} else {
-		memset(sport->rx_dma_buf, 0, size);
 		sport_set_rx_callback(sport, bf5xx_dma_irq, substream);
 		sport_config_rx_dma(sport, sport->rx_dma_buf, runtime->periods,
 			runtime->period_size * sizeof(struct ac97_frame));
@@ -164,8 +172,12 @@ static int bf5xx_pcm_trigger(struct snd_pcm_substream *substream, int cmd)
 	pr_debug("%s enter\n", __func__);
 	switch (cmd) {
 	case SNDRV_PCM_TRIGGER_START:
-		if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
+		if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
+			bf5xx_mmap_copy(substream, runtime->period_size);
+			snd_pcm_period_elapsed(substream);
+			sport->tx_delay_pos = 0;
 			sport_tx_start(sport);
+		}
 		else
 			sport_rx_start(sport);
 		break;
@@ -198,7 +210,7 @@ static snd_pcm_uframes_t bf5xx_pcm_pointer(struct snd_pcm_substream *substream)
 
 #if defined(CONFIG_SND_MMAP_SUPPORT)
 	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
-		curr = sport->tx_pos;
+		curr = sport->tx_delay_pos;
 	else
 		curr = sport->rx_pos;
 #else
@@ -237,6 +249,21 @@ static int bf5xx_pcm_open(struct snd_pcm_substream *substream)
 	return ret;
 }
 
+static int bf5xx_pcm_close(struct snd_pcm_substream *substream)
+{
+	struct snd_pcm_runtime *runtime = substream->runtime;
+	struct sport_device *sport = runtime->private_data;
+
+	pr_debug("%s enter\n", __func__);
+	if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
+		sport->once = 0;
+		memset(sport->tx_dma_buf, 0, runtime->buffer_size * sizeof(struct ac97_frame));
+	} else
+		memset(sport->rx_dma_buf, 0, runtime->buffer_size * sizeof(struct ac97_frame));
+
+	return 0;
+}
+
 #ifdef CONFIG_SND_MMAP_SUPPORT
 static int bf5xx_pcm_mmap(struct snd_pcm_substream *substream,
 	struct vm_area_struct *vma)
@@ -272,6 +299,7 @@ static	int bf5xx_pcm_copy(struct snd_pcm_substream *substream, int channel,
 
 struct snd_pcm_ops bf5xx_pcm_ac97_ops = {
 	.open		= bf5xx_pcm_open,
+	.close		= bf5xx_pcm_close,
 	.ioctl		= snd_pcm_lib_ioctl,
 	.hw_params	= bf5xx_pcm_hw_params,
 	.hw_free	= bf5xx_pcm_hw_free,
diff --git a/sound/soc/blackfin/bf5xx-ac97.c b/sound/soc/blackfin/bf5xx-ac97.c
index c782e311fd56..5e5aafb6485f 100644
--- a/sound/soc/blackfin/bf5xx-ac97.c
+++ b/sound/soc/blackfin/bf5xx-ac97.c
@@ -129,7 +129,6 @@ static void enqueue_cmd(struct snd_ac97 *ac97, __u16 addr, __u16 data)
 	struct ac97_frame *nextwrite;
 
 	sport_incfrag(sport, &nextfrag, 1);
-	sport_incfrag(sport, &nextfrag, 1);
 
 	nextwrite = (struct ac97_frame *)(sport->tx_buf + \
 			nextfrag * sport->tx_fragsize);
diff --git a/sound/soc/blackfin/bf5xx-ad73311.c b/sound/soc/blackfin/bf5xx-ad73311.c
new file mode 100644
index 000000000000..622c9b909532
--- /dev/null
+++ b/sound/soc/blackfin/bf5xx-ad73311.c
@@ -0,0 +1,240 @@
+/*
+ * File:         sound/soc/blackfin/bf5xx-ad73311.c
+ * Author:       Cliff Cai <Cliff.Cai@analog.com>
+ *
+ * Created:      Thur Sep 25 2008
+ * Description:  Board driver for ad73311 sound chip
+ *
+ * Modified:
+ *               Copyright 2008 Analog Devices Inc.
+ *
+ * Bugs:         Enter bugs at http://blackfin.uclinux.org/
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see the file COPYING, or write
+ * to the Free Software Foundation, Inc.,
+ * 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/device.h>
+#include <linux/delay.h>
+#include <linux/gpio.h>
+
+#include <sound/core.h>
+#include <sound/pcm.h>
+#include <sound/soc.h>
+#include <sound/soc-dapm.h>
+#include <sound/pcm_params.h>
+
+#include <asm/blackfin.h>
+#include <asm/cacheflush.h>
+#include <asm/irq.h>
+#include <asm/dma.h>
+#include <asm/portmux.h>
+
+#include "../codecs/ad73311.h"
+#include "bf5xx-sport.h"
+#include "bf5xx-i2s-pcm.h"
+#include "bf5xx-i2s.h"
+
+#if CONFIG_SND_BF5XX_SPORT_NUM == 0
+#define bfin_write_SPORT_TCR1	bfin_write_SPORT0_TCR1
+#define bfin_read_SPORT_TCR1	bfin_read_SPORT0_TCR1
+#define bfin_write_SPORT_TCR2	bfin_write_SPORT0_TCR2
+#define bfin_write_SPORT_TX16	bfin_write_SPORT0_TX16
+#define bfin_read_SPORT_STAT	bfin_read_SPORT0_STAT
+#else
+#define bfin_write_SPORT_TCR1	bfin_write_SPORT1_TCR1
+#define bfin_read_SPORT_TCR1	bfin_read_SPORT1_TCR1
+#define bfin_write_SPORT_TCR2	bfin_write_SPORT1_TCR2
+#define bfin_write_SPORT_TX16	bfin_write_SPORT1_TX16
+#define bfin_read_SPORT_STAT	bfin_read_SPORT1_STAT
+#endif
+
+#define GPIO_SE CONFIG_SND_BFIN_AD73311_SE
+
+static struct snd_soc_machine bf5xx_ad73311;
+
+static int snd_ad73311_startup(void)
+{
+	pr_debug("%s enter\n", __func__);
+
+	/* Pull up SE pin on AD73311L */
+	gpio_set_value(GPIO_SE, 1);
+	return 0;
+}
+
+static int snd_ad73311_configure(void)
+{
+	unsigned short ctrl_regs[6];
+	unsigned short status = 0;
+	int count = 0;
+
+	/* DMCLK = MCLK = 16.384 MHz
+	 * SCLK = DMCLK/8 = 2.048 MHz
+	 * Sample Rate = DMCLK/2048  = 8 KHz
+	 */
+	ctrl_regs[0] = AD_CONTROL | AD_WRITE | CTRL_REG_B | REGB_MCDIV(0) | \
+			REGB_SCDIV(0) | REGB_DIRATE(0);
+	ctrl_regs[1] = AD_CONTROL | AD_WRITE | CTRL_REG_C | REGC_PUDEV | \
+			REGC_PUADC | REGC_PUDAC | REGC_PUREF | REGC_REFUSE ;
+	ctrl_regs[2] = AD_CONTROL | AD_WRITE | CTRL_REG_D | REGD_OGS(2) | \
+			REGD_IGS(2);
+	ctrl_regs[3] = AD_CONTROL | AD_WRITE | CTRL_REG_E | REGE_DA(0x1f);
+	ctrl_regs[4] = AD_CONTROL | AD_WRITE | CTRL_REG_F | REGF_SEEN ;
+	ctrl_regs[5] = AD_CONTROL | AD_WRITE | CTRL_REG_A | REGA_MODE_DATA;
+
+	local_irq_disable();
+	snd_ad73311_startup();
+	udelay(1);
+
+	bfin_write_SPORT_TCR1(TFSR);
+	bfin_write_SPORT_TCR2(0xF);
+	SSYNC();
+
+	/* SPORT Tx Register is a 8 x 16 FIFO, all the data can be put to
+	 * FIFO before enable SPORT to transfer the data
+	 */
+	for (count = 0; count < 6; count++)
+		bfin_write_SPORT_TX16(ctrl_regs[count]);
+	SSYNC();
+	bfin_write_SPORT_TCR1(bfin_read_SPORT_TCR1() | TSPEN);
+	SSYNC();
+
+	/* When TUVF is set, the data is already send out */
+	while (!(status & TUVF) && count++ < 10000) {
+		udelay(1);
+		status = bfin_read_SPORT_STAT();
+		SSYNC();
+	}
+	bfin_write_SPORT_TCR1(bfin_read_SPORT_TCR1() & ~TSPEN);
+	SSYNC();
+	local_irq_enable();
+
+	if (count == 10000) {
+		printk(KERN_ERR "ad73311: failed to configure codec\n");
+		return -1;
+	}
+	return 0;
+}
+
+static int bf5xx_probe(struct platform_device *pdev)
+{
+	int err;
+	if (gpio_request(GPIO_SE, "AD73311_SE")) {
+		printk(KERN_ERR "%s: Failed ro request GPIO_%d\n", __func__, GPIO_SE);
+		return -EBUSY;
+	}
+
+	gpio_direction_output(GPIO_SE, 0);
+
+	err = snd_ad73311_configure();
+	if (err < 0)
+		return -EFAULT;
+
+	return 0;
+}
+
+static int bf5xx_ad73311_startup(struct snd_pcm_substream *substream)
+{
+	struct snd_soc_pcm_runtime *rtd = substream->private_data;
+	struct snd_soc_dai *cpu_dai = rtd->dai->cpu_dai;
+
+	pr_debug("%s enter\n", __func__);
+	cpu_dai->private_data = sport_handle;
+	return 0;
+}
+
+static int bf5xx_ad73311_hw_params(struct snd_pcm_substream *substream,
+	struct snd_pcm_hw_params *params)
+{
+	struct snd_soc_pcm_runtime *rtd = substream->private_data;
+	struct snd_soc_dai *cpu_dai = rtd->dai->cpu_dai;
+	int ret = 0;
+
+	pr_debug("%s rate %d format %x\n", __func__, params_rate(params),
+		params_format(params));
+
+	/* set cpu DAI configuration */
+	ret = cpu_dai->dai_ops.set_fmt(cpu_dai, SND_SOC_DAIFMT_DSP_A |
+		SND_SOC_DAIFMT_NB_NF | SND_SOC_DAIFMT_CBM_CFM);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+
+static struct snd_soc_ops bf5xx_ad73311_ops = {
+	.startup = bf5xx_ad73311_startup,
+	.hw_params = bf5xx_ad73311_hw_params,
+};
+
+static struct snd_soc_dai_link bf5xx_ad73311_dai = {
+	.name = "ad73311",
+	.stream_name = "AD73311",
+	.cpu_dai = &bf5xx_i2s_dai,
+	.codec_dai = &ad73311_dai,
+	.ops = &bf5xx_ad73311_ops,
+};
+
+static struct snd_soc_machine bf5xx_ad73311 = {
+	.name = "bf5xx_ad73311",
+	.probe = bf5xx_probe,
+	.dai_link = &bf5xx_ad73311_dai,
+	.num_links = 1,
+};
+
+static struct snd_soc_device bf5xx_ad73311_snd_devdata = {
+	.machine = &bf5xx_ad73311,
+	.platform = &bf5xx_i2s_soc_platform,
+	.codec_dev = &soc_codec_dev_ad73311,
+};
+
+static struct platform_device *bf52x_ad73311_snd_device;
+
+static int __init bf5xx_ad73311_init(void)
+{
+	int ret;
+
+	pr_debug("%s enter\n", __func__);
+	bf52x_ad73311_snd_device = platform_device_alloc("soc-audio", -1);
+	if (!bf52x_ad73311_snd_device)
+		return -ENOMEM;
+
+	platform_set_drvdata(bf52x_ad73311_snd_device, &bf5xx_ad73311_snd_devdata);
+	bf5xx_ad73311_snd_devdata.dev = &bf52x_ad73311_snd_device->dev;
+	ret = platform_device_add(bf52x_ad73311_snd_device);
+
+	if (ret)
+		platform_device_put(bf52x_ad73311_snd_device);
+
+	return ret;
+}
+
+static void __exit bf5xx_ad73311_exit(void)
+{
+	pr_debug("%s enter\n", __func__);
+	platform_device_unregister(bf52x_ad73311_snd_device);
+}
+
+module_init(bf5xx_ad73311_init);
+module_exit(bf5xx_ad73311_exit);
+
+/* Module information */
+MODULE_AUTHOR("Cliff Cai");
+MODULE_DESCRIPTION("ALSA SoC AD73311 Blackfin");
+MODULE_LICENSE("GPL");
+
diff --git a/sound/soc/blackfin/bf5xx-i2s.c b/sound/soc/blackfin/bf5xx-i2s.c
index 43a4092eeb89..827587f08180 100644
--- a/sound/soc/blackfin/bf5xx-i2s.c
+++ b/sound/soc/blackfin/bf5xx-i2s.c
@@ -70,6 +70,13 @@ static struct sport_param sport_params[2] = {
 	}
 };
 
+static u16 sport_req[][7] = {
+		{ P_SPORT0_DTPRI, P_SPORT0_TSCLK, P_SPORT0_RFS,
+		  P_SPORT0_DRPRI, P_SPORT0_RSCLK, 0},
+		{ P_SPORT1_DTPRI, P_SPORT1_TSCLK, P_SPORT1_RFS,
+		  P_SPORT1_DRPRI, P_SPORT1_RSCLK, 0},
+};
+
 static int bf5xx_i2s_set_dai_fmt(struct snd_soc_dai *cpu_dai,
 		unsigned int fmt)
 {
@@ -78,6 +85,14 @@ static int bf5xx_i2s_set_dai_fmt(struct snd_soc_dai *cpu_dai,
 	/* interface format:support I2S,slave mode */
 	switch (fmt & SND_SOC_DAIFMT_FORMAT_MASK) {
 	case SND_SOC_DAIFMT_I2S:
+		bf5xx_i2s.tcr1 |= TFSR | TCKFE;
+		bf5xx_i2s.rcr1 |= RFSR | RCKFE;
+		bf5xx_i2s.tcr2 |= TSFSE;
+		bf5xx_i2s.rcr2 |= RSFSE;
+		break;
+	case SND_SOC_DAIFMT_DSP_A:
+		bf5xx_i2s.tcr1 |= TFSR;
+		bf5xx_i2s.rcr1 |= RFSR;
 		break;
 	case SND_SOC_DAIFMT_LEFT_J:
 		ret = -EINVAL;
@@ -127,14 +142,17 @@ static int bf5xx_i2s_hw_params(struct snd_pcm_substream *substream,
 	case SNDRV_PCM_FORMAT_S16_LE:
 		bf5xx_i2s.tcr2 |= 15;
 		bf5xx_i2s.rcr2 |= 15;
+		sport_handle->wdsize = 2;
 		break;
 	case SNDRV_PCM_FORMAT_S24_LE:
 		bf5xx_i2s.tcr2 |= 23;
 		bf5xx_i2s.rcr2 |= 23;
+		sport_handle->wdsize = 3;
 		break;
 	case SNDRV_PCM_FORMAT_S32_LE:
 		bf5xx_i2s.tcr2 |= 31;
 		bf5xx_i2s.rcr2 |= 31;
+		sport_handle->wdsize = 4;
 		break;
 	}
 
@@ -145,17 +163,17 @@ static int bf5xx_i2s_hw_params(struct snd_pcm_substream *substream,
 		 * need to configure both of them at the time when the first
 		 * stream is opened.
 		 *
-		 * CPU DAI format:I2S, slave mode.
+		 * CPU DAI:slave mode.
 		 */
-		ret = sport_config_rx(sport_handle, RFSR | RCKFE,
-				      RSFSE|bf5xx_i2s.rcr2, 0, 0);
+		ret = sport_config_rx(sport_handle, bf5xx_i2s.rcr1,
+				      bf5xx_i2s.rcr2, 0, 0);
 		if (ret) {
 			pr_err("SPORT is busy!\n");
 			return -EBUSY;
 		}
 
-		ret = sport_config_tx(sport_handle, TFSR | TCKFE,
-				      TSFSE|bf5xx_i2s.tcr2, 0, 0);
+		ret = sport_config_tx(sport_handle, bf5xx_i2s.tcr1,
+				      bf5xx_i2s.tcr2, 0, 0);
 		if (ret) {
 			pr_err("SPORT is busy!\n");
 			return -EBUSY;
@@ -174,13 +192,6 @@ static void bf5xx_i2s_shutdown(struct snd_pcm_substream *substream)
 static int bf5xx_i2s_probe(struct platform_device *pdev,
 			   struct snd_soc_dai *dai)
 {
-	u16 sport_req[][7] = {
-		{ P_SPORT0_DTPRI, P_SPORT0_TSCLK, P_SPORT0_RFS,
-		  P_SPORT0_DRPRI, P_SPORT0_RSCLK, 0},
-		{ P_SPORT1_DTPRI, P_SPORT1_TSCLK, P_SPORT1_RFS,
-		  P_SPORT1_DRPRI, P_SPORT1_RSCLK, 0},
-	};
-
 	pr_debug("%s enter\n", __func__);
 	if (peripheral_request_list(&sport_req[sport_num][0], "soc-audio")) {
 		pr_err("Requesting Peripherals failed\n");
@@ -198,6 +209,13 @@ static int bf5xx_i2s_probe(struct platform_device *pdev,
 	return 0;
 }
 
+static void bf5xx_i2s_remove(struct platform_device *pdev,
+			   struct snd_soc_dai *dai)
+{
+	pr_debug("%s enter\n", __func__);
+	peripheral_free_list(&sport_req[sport_num][0]);
+}
+
 #ifdef CONFIG_PM
 static int bf5xx_i2s_suspend(struct platform_device *dev,
 			     struct snd_soc_dai *dai)
@@ -263,15 +281,16 @@ struct snd_soc_dai bf5xx_i2s_dai = {
 	.id = 0,
 	.type = SND_SOC_DAI_I2S,
 	.probe = bf5xx_i2s_probe,
+	.remove = bf5xx_i2s_remove,
 	.suspend = bf5xx_i2s_suspend,
 	.resume = bf5xx_i2s_resume,
 	.playback = {
-		.channels_min = 2,
+		.channels_min = 1,
 		.channels_max = 2,
 		.rates = BF5XX_I2S_RATES,
 		.formats = BF5XX_I2S_FORMATS,},
 	.capture = {
-		.channels_min = 2,
+		.channels_min = 1,
 		.channels_max = 2,
 		.rates = BF5XX_I2S_RATES,
 		.formats = BF5XX_I2S_FORMATS,},
diff --git a/sound/soc/blackfin/bf5xx-sport.h b/sound/soc/blackfin/bf5xx-sport.h
index 4c163454bbf8..fcadcc081f7f 100644
--- a/sound/soc/blackfin/bf5xx-sport.h
+++ b/sound/soc/blackfin/bf5xx-sport.h
@@ -123,6 +123,8 @@ struct sport_device {
 	int rx_pos;
 	unsigned int tx_buffer_size;
 	unsigned int rx_buffer_size;
+	int tx_delay_pos;
+	int once;
 #endif
 	void *private_data;
 };
diff --git a/sound/soc/codecs/Kconfig b/sound/soc/codecs/Kconfig
index e0b9869df0f1..4975d8573e4f 100644
--- a/sound/soc/codecs/Kconfig
+++ b/sound/soc/codecs/Kconfig
@@ -3,9 +3,11 @@ config SND_SOC_ALL_CODECS
 	depends on I2C
 	select SPI
 	select SPI_MASTER
+	select SND_SOC_AD73311
 	select SND_SOC_AK4535
 	select SND_SOC_CS4270
 	select SND_SOC_SSM2602
+	select SND_SOC_TLV320AIC23
 	select SND_SOC_TLV320AIC26
 	select SND_SOC_TLV320AIC3X
 	select SND_SOC_UDA1380
@@ -34,6 +36,9 @@ config SND_SOC_AC97_CODEC
 config SND_SOC_AD1980
 	tristate
 
+config SND_SOC_AD73311
+	tristate
+
 config SND_SOC_AK4535
 	tristate
 
@@ -58,9 +63,13 @@ config SND_SOC_CS4270_VD33_ERRATA
 config SND_SOC_SSM2602
 	tristate
 
+config SND_SOC_TLV320AIC23
+	tristate
+	depends on I2C
+
 config SND_SOC_TLV320AIC26
 	tristate "TI TLV320AIC26 Codec support"
-	depends on SND_SOC && SPI
+	depends on SPI
 
 config SND_SOC_TLV320AIC3X
 	tristate
diff --git a/sound/soc/codecs/Makefile b/sound/soc/codecs/Makefile
index f977978a3409..90f0a585fc70 100644
--- a/sound/soc/codecs/Makefile
+++ b/sound/soc/codecs/Makefile
@@ -1,8 +1,10 @@
 snd-soc-ac97-objs := ac97.o
 snd-soc-ad1980-objs := ad1980.o
+snd-soc-ad73311-objs := ad73311.o
 snd-soc-ak4535-objs := ak4535.o
 snd-soc-cs4270-objs := cs4270.o
 snd-soc-ssm2602-objs := ssm2602.o
+snd-soc-tlv320aic23-objs := tlv320aic23.o
 snd-soc-tlv320aic26-objs := tlv320aic26.o
 snd-soc-tlv320aic3x-objs := tlv320aic3x.o
 snd-soc-uda1380-objs := uda1380.o
@@ -20,9 +22,11 @@ snd-soc-wm9713-objs := wm9713.o
 
 obj-$(CONFIG_SND_SOC_AC97_CODEC)	+= snd-soc-ac97.o
 obj-$(CONFIG_SND_SOC_AD1980)	+= snd-soc-ad1980.o
+obj-$(CONFIG_SND_SOC_AD73311) += snd-soc-ad73311.o
 obj-$(CONFIG_SND_SOC_AK4535)	+= snd-soc-ak4535.o
 obj-$(CONFIG_SND_SOC_CS4270)	+= snd-soc-cs4270.o
 obj-$(CONFIG_SND_SOC_SSM2602)	+= snd-soc-ssm2602.o
+obj-$(CONFIG_SND_SOC_TLV320AIC23)	+= snd-soc-tlv320aic23.o
 obj-$(CONFIG_SND_SOC_TLV320AIC26)	+= snd-soc-tlv320aic26.o
 obj-$(CONFIG_SND_SOC_TLV320AIC3X)	+= snd-soc-tlv320aic3x.o
 obj-$(CONFIG_SND_SOC_UDA1380)	+= snd-soc-uda1380.o
diff --git a/sound/soc/codecs/ac97.c b/sound/soc/codecs/ac97.c
index 61fd96ca7bc7..bd1ebdc6c86c 100644
--- a/sound/soc/codecs/ac97.c
+++ b/sound/soc/codecs/ac97.c
@@ -2,8 +2,7 @@
  * ac97.c  --  ALSA Soc AC97 codec support
  *
  * Copyright 2005 Wolfson Microelectronics PLC.
- * Author: Liam Girdwood
- *         liam.girdwood@wolfsonmicro.com or linux@wolfsonmicro.com
+ * Author: Liam Girdwood <lrg@slimlogic.co.uk>
  *
  *  This program is free software; you can redistribute  it and/or modify it
  *  under  the terms of  the GNU General  Public License as published by the
diff --git a/sound/soc/codecs/ad1980.c b/sound/soc/codecs/ad1980.c
index 4e09c1f2c063..1397b8e06c0b 100644
--- a/sound/soc/codecs/ad1980.c
+++ b/sound/soc/codecs/ad1980.c
@@ -13,7 +13,6 @@
 
 #include <linux/init.h>
 #include <linux/module.h>
-#include <linux/version.h>
 #include <linux/kernel.h>
 #include <linux/device.h>
 #include <sound/core.h>
diff --git a/sound/soc/codecs/ad73311.c b/sound/soc/codecs/ad73311.c
new file mode 100644
index 000000000000..37af8607b00a
--- /dev/null
+++ b/sound/soc/codecs/ad73311.c
@@ -0,0 +1,107 @@
+/*
+ * ad73311.c  --  ALSA Soc AD73311 codec support
+ *
+ * Copyright:	Analog Device Inc.
+ * Author:	Cliff Cai <cliff.cai@analog.com>
+ *
+ *  This program is free software; you can redistribute  it and/or modify it
+ *  under  the terms of  the GNU General  Public License as published by the
+ *  Free Software Foundation;  either version 2 of the  License, or (at your
+ *  option) any later version.
+ *
+ *  Revision history
+ *    25th Sep 2008   Initial version.
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/version.h>
+#include <linux/kernel.h>
+#include <linux/device.h>
+#include <sound/core.h>
+#include <sound/pcm.h>
+#include <sound/ac97_codec.h>
+#include <sound/initval.h>
+#include <sound/soc.h>
+
+#include "ad73311.h"
+
+struct snd_soc_dai ad73311_dai = {
+	.name = "AD73311",
+	.playback = {
+		.stream_name = "Playback",
+		.channels_min = 1,
+		.channels_max = 1,
+		.rates = SNDRV_PCM_RATE_8000,
+		.formats = SNDRV_PCM_FMTBIT_S16_LE, },
+	.capture = {
+		.stream_name = "Capture",
+		.channels_min = 1,
+		.channels_max = 1,
+		.rates = SNDRV_PCM_RATE_8000,
+		.formats = SNDRV_PCM_FMTBIT_S16_LE, },
+};
+EXPORT_SYMBOL_GPL(ad73311_dai);
+
+static int ad73311_soc_probe(struct platform_device *pdev)
+{
+	struct snd_soc_device *socdev = platform_get_drvdata(pdev);
+	struct snd_soc_codec *codec;
+	int ret = 0;
+
+	codec = kzalloc(sizeof(struct snd_soc_codec), GFP_KERNEL);
+	if (codec == NULL)
+		return -ENOMEM;
+	mutex_init(&codec->mutex);
+	codec->name = "AD73311";
+	codec->owner = THIS_MODULE;
+	codec->dai = &ad73311_dai;
+	codec->num_dai = 1;
+	socdev->codec = codec;
+	INIT_LIST_HEAD(&codec->dapm_widgets);
+	INIT_LIST_HEAD(&codec->dapm_paths);
+
+	/* register pcms */
+	ret = snd_soc_new_pcms(socdev, SNDRV_DEFAULT_IDX1, SNDRV_DEFAULT_STR1);
+	if (ret < 0) {
+		printk(KERN_ERR "ad73311: failed to create pcms\n");
+		goto pcm_err;
+	}
+
+	ret = snd_soc_register_card(socdev);
+	if (ret < 0) {
+		printk(KERN_ERR "ad73311: failed to register card\n");
+		goto register_err;
+	}
+
+	return ret;
+
+register_err:
+	snd_soc_free_pcms(socdev);
+pcm_err:
+	kfree(socdev->codec);
+	socdev->codec = NULL;
+	return ret;
+}
+
+static int ad73311_soc_remove(struct platform_device *pdev)
+{
+	struct snd_soc_device *socdev = platform_get_drvdata(pdev);
+	struct snd_soc_codec *codec = socdev->codec;
+
+	if (codec == NULL)
+		return 0;
+	snd_soc_free_pcms(socdev);
+	kfree(codec);
+	return 0;
+}
+
+struct snd_soc_codec_device soc_codec_dev_ad73311 = {
+	.probe = 	ad73311_soc_probe,
+	.remove = 	ad73311_soc_remove,
+};
+EXPORT_SYMBOL_GPL(soc_codec_dev_ad73311);
+
+MODULE_DESCRIPTION("ASoC ad73311 driver");
+MODULE_AUTHOR("Cliff Cai ");
+MODULE_LICENSE("GPL");
diff --git a/sound/soc/codecs/ad73311.h b/sound/soc/codecs/ad73311.h
new file mode 100644
index 000000000000..507ce0c30edf
--- /dev/null
+++ b/sound/soc/codecs/ad73311.h
@@ -0,0 +1,90 @@
+/*
+ * File:         sound/soc/codec/ad73311.h
+ * Based on:
+ * Author:       Cliff Cai <cliff.cai@analog.com>
+ *
+ * Created:      Thur Sep 25, 2008
+ * Description:  definitions for AD73311 registers
+ *
+ *
+ * Modified:
+ *               Copyright 2006 Analog Devices Inc.
+ *
+ * Bugs:         Enter bugs at http://blackfin.uclinux.org/
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see the file COPYING, or write
+ * to the Free Software Foundation, Inc.,
+ * 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+
+#ifndef __AD73311_H__
+#define __AD73311_H__
+
+#define AD_CONTROL	0x8000
+#define AD_DATA		0x0000
+#define AD_READ		0x4000
+#define AD_WRITE	0x0000
+
+/* Control register A */
+#define CTRL_REG_A	(0 << 8)
+
+#define REGA_MODE_PRO	0x00
+#define REGA_MODE_DATA	0x01
+#define REGA_MODE_MIXED	0x03
+#define REGA_DLB		0x04
+#define REGA_SLB		0x08
+#define REGA_DEVC(x)		((x & 0x7) << 4)
+#define REGA_RESET		0x80
+
+/* Control register B */
+#define CTRL_REG_B	(1 << 8)
+
+#define REGB_DIRATE(x)	(x & 0x3)
+#define REGB_SCDIV(x)	((x & 0x3) << 2)
+#define REGB_MCDIV(x)	((x & 0x7) << 4)
+#define REGB_CEE		(1 << 7)
+
+/* Control register C */
+#define CTRL_REG_C	(2 << 8)
+
+#define REGC_PUDEV		(1 << 0)
+#define REGC_PUADC		(1 << 3)
+#define REGC_PUDAC		(1 << 4)
+#define REGC_PUREF		(1 << 5)
+#define REGC_REFUSE		(1 << 6)
+
+/* Control register D */
+#define CTRL_REG_D	(3 << 8)
+
+#define REGD_IGS(x)		(x & 0x7)
+#define REGD_RMOD		(1 << 3)
+#define REGD_OGS(x)		((x & 0x7) << 4)
+#define REGD_MUTE		(x << 7)
+
+/* Control register E */
+#define CTRL_REG_E	(4 << 8)
+
+#define REGE_DA(x)		(x & 0x1f)
+#define REGE_IBYP		(1 << 5)
+
+/* Control register F */
+#define CTRL_REG_F	(5 << 8)
+
+#define REGF_SEEN		(1 << 5)
+#define REGF_INV		(1 << 6)
+#define REGF_ALB		(1 << 7)
+
+extern struct snd_soc_dai ad73311_dai;
+extern struct snd_soc_codec_device soc_codec_dev_ad73311;
+#endif
diff --git a/sound/soc/codecs/ak4535.c b/sound/soc/codecs/ak4535.c
index 088cf9927720..2a89b5888e11 100644
--- a/sound/soc/codecs/ak4535.c
+++ b/sound/soc/codecs/ak4535.c
@@ -28,7 +28,6 @@
 
 #include "ak4535.h"
 
-#define AUDIO_NAME "ak4535"
 #define AK4535_VERSION "0.3"
 
 struct snd_soc_codec_device soc_codec_dev_ak4535;
diff --git a/sound/soc/codecs/ssm2602.c b/sound/soc/codecs/ssm2602.c
index 940ce1c3522e..44ef0dacd564 100644
--- a/sound/soc/codecs/ssm2602.c
+++ b/sound/soc/codecs/ssm2602.c
@@ -42,7 +42,6 @@
 
 #include "ssm2602.h"
 
-#define AUDIO_NAME "ssm2602"
 #define SSM2602_VERSION "0.1"
 
 struct snd_soc_codec_device soc_codec_dev_ssm2602;
diff --git a/sound/soc/codecs/tlv320aic23.c b/sound/soc/codecs/tlv320aic23.c
new file mode 100644
index 000000000000..bac7815e00fb
--- /dev/null
+++ b/sound/soc/codecs/tlv320aic23.c
@@ -0,0 +1,714 @@
+/*
+ * ALSA SoC TLV320AIC23 codec driver
+ *
+ * Author:      Arun KS, <arunks@mistralsolutions.com>
+ * Copyright:   (C) 2008 Mistral Solutions Pvt Ltd.,
+ *
+ * Based on sound/soc/codecs/wm8731.c by Richard Purdie
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Notes:
+ *  The AIC23 is a driver for a low power stereo audio
+ *  codec tlv320aic23
+ *
+ *  The machine layer should disable unsupported inputs/outputs by
+ *  snd_soc_dapm_disable_pin(codec, "LHPOUT"), etc.
+ */
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/pm.h>
+#include <linux/i2c.h>
+#include <linux/platform_device.h>
+#include <sound/core.h>
+#include <sound/pcm.h>
+#include <sound/pcm_params.h>
+#include <sound/soc.h>
+#include <sound/soc-dapm.h>
+#include <sound/tlv.h>
+#include <sound/initval.h>
+
+#include "tlv320aic23.h"
+
+#define AIC23_VERSION "0.1"
+
+struct tlv320aic23_srate_reg_info {
+	u32 sample_rate;
+	u8 control;		/* SR3, SR2, SR1, SR0 and BOSR */
+	u8 divider;		/* if 0 CLKIN = MCLK, if 1 CLKIN = MCLK/2 */
+};
+
+/*
+ * AIC23 register cache
+ */
+static const u16 tlv320aic23_reg[] = {
+	0x0097, 0x0097, 0x00F9, 0x00F9,	/* 0 */
+	0x001A, 0x0004, 0x0007, 0x0001,	/* 4 */
+	0x0020, 0x0000, 0x0000, 0x0000,	/* 8 */
+	0x0000, 0x0000, 0x0000, 0x0000,	/* 12 */
+};
+
+/*
+ * read tlv320aic23 register cache
+ */
+static inline unsigned int tlv320aic23_read_reg_cache(struct snd_soc_codec
+						      *codec, unsigned int reg)
+{
+	u16 *cache = codec->reg_cache;
+	if (reg >= ARRAY_SIZE(tlv320aic23_reg))
+		return -1;
+	return cache[reg];
+}
+
+/*
+ * write tlv320aic23 register cache
+ */
+static inline void tlv320aic23_write_reg_cache(struct snd_soc_codec *codec,
+					       u8 reg, u16 value)
+{
+	u16 *cache = codec->reg_cache;
+	if (reg >= ARRAY_SIZE(tlv320aic23_reg))
+		return;
+	cache[reg] = value;
+}
+
+/*
+ * write to the tlv320aic23 register space
+ */
+static int tlv320aic23_write(struct snd_soc_codec *codec, unsigned int reg,
+			     unsigned int value)
+{
+
+	u8 data;
+
+	/* TLV320AIC23 has 7 bit address and 9 bits of data
+	 * so we need to switch one data bit into reg and rest
+	 * of data into val
+	 */
+
+	if ((reg < 0 || reg > 9) && (reg != 15)) {
+		printk(KERN_WARNING "%s Invalid register R%d\n", __func__, reg);
+		return -1;
+	}
+
+	data = (reg << 1) | (value >> 8 & 0x01);
+
+	tlv320aic23_write_reg_cache(codec, reg, value);
+
+	if (codec->hw_write(codec->control_data, data,
+			    (value & 0xff)) == 0)
+		return 0;
+
+	printk(KERN_ERR "%s cannot write %03x to register R%d\n", __func__,
+	       value, reg);
+
+	return -EIO;
+}
+
+static const char *rec_src_text[] = { "Line", "Mic" };
+static const char *deemph_text[] = {"None", "32Khz", "44.1Khz", "48Khz"};
+
+static const struct soc_enum rec_src_enum =
+	SOC_ENUM_SINGLE(TLV320AIC23_ANLG, 2, 2, rec_src_text);
+
+static const struct snd_kcontrol_new tlv320aic23_rec_src_mux_controls =
+SOC_DAPM_ENUM("Input Select", rec_src_enum);
+
+static const struct soc_enum tlv320aic23_rec_src =
+	SOC_ENUM_SINGLE(TLV320AIC23_ANLG, 2, 2, rec_src_text);
+static const struct soc_enum tlv320aic23_deemph =
+	SOC_ENUM_SINGLE(TLV320AIC23_DIGT, 1, 4, deemph_text);
+
+static const DECLARE_TLV_DB_SCALE(out_gain_tlv, -12100, 100, 0);
+static const DECLARE_TLV_DB_SCALE(input_gain_tlv, -1725, 75, 0);
+static const DECLARE_TLV_DB_SCALE(sidetone_vol_tlv, -1800, 300, 0);
+
+static int snd_soc_tlv320aic23_put_volsw(struct snd_kcontrol *kcontrol,
+	struct snd_ctl_elem_value *ucontrol)
+{
+	struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
+	u16 val, reg;
+
+	val = (ucontrol->value.integer.value[0] & 0x07);
+
+	/* linear conversion to userspace
+	* 000	=	-6db
+	* 001	=	-9db
+	* 010	=	-12db
+	* 011	=	-18db (Min)
+	* 100	=	0db (Max)
+	*/
+	val = (val >= 4) ? 4  : (3 - val);
+
+	reg = tlv320aic23_read_reg_cache(codec, TLV320AIC23_ANLG) & (~0x1C0);
+	tlv320aic23_write(codec, TLV320AIC23_ANLG, reg | (val << 6));
+
+	return 0;
+}
+
+static int snd_soc_tlv320aic23_get_volsw(struct snd_kcontrol *kcontrol,
+	struct snd_ctl_elem_value *ucontrol)
+{
+	struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
+	u16 val;
+
+	val = tlv320aic23_read_reg_cache(codec, TLV320AIC23_ANLG) & (0x1C0);
+	val = val >> 6;
+	val = (val >= 4) ? 4  : (3 -  val);
+	ucontrol->value.integer.value[0] = val;
+	return 0;
+
+}
+
+#define SOC_TLV320AIC23_SINGLE_TLV(xname, reg, shift, max, invert, tlv_array) \
+{	.iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = xname, \
+	.access = SNDRV_CTL_ELEM_ACCESS_TLV_READ |\
+		 SNDRV_CTL_ELEM_ACCESS_READWRITE,\
+	.tlv.p = (tlv_array), \
+	.info = snd_soc_info_volsw, .get = snd_soc_tlv320aic23_get_volsw,\
+	.put = snd_soc_tlv320aic23_put_volsw, \
+	.private_value =  SOC_SINGLE_VALUE(reg, shift, max, invert) }
+
+static const struct snd_kcontrol_new tlv320aic23_snd_controls[] = {
+	SOC_DOUBLE_R_TLV("Digital Playback Volume", TLV320AIC23_LCHNVOL,
+			 TLV320AIC23_RCHNVOL, 0, 127, 0, out_gain_tlv),
+	SOC_SINGLE("Digital Playback Switch", TLV320AIC23_DIGT, 3, 1, 1),
+	SOC_DOUBLE_R("Line Input Switch", TLV320AIC23_LINVOL,
+		     TLV320AIC23_RINVOL, 7, 1, 0),
+	SOC_DOUBLE_R_TLV("Line Input Volume", TLV320AIC23_LINVOL,
+			 TLV320AIC23_RINVOL, 0, 31, 0, input_gain_tlv),
+	SOC_SINGLE("Mic Input Switch", TLV320AIC23_ANLG, 1, 1, 1),
+	SOC_SINGLE("Mic Booster Switch", TLV320AIC23_ANLG, 0, 1, 0),
+	SOC_TLV320AIC23_SINGLE_TLV("Sidetone Volume", TLV320AIC23_ANLG,
+				  6, 4, 0, sidetone_vol_tlv),
+	SOC_ENUM("Playback De-emphasis", tlv320aic23_deemph),
+};
+
+/* add non dapm controls */
+static int tlv320aic23_add_controls(struct snd_soc_codec *codec)
+{
+
+	int err, i;
+
+	for (i = 0; i < ARRAY_SIZE(tlv320aic23_snd_controls); i++) {
+		err = snd_ctl_add(codec->card,
+				  snd_soc_cnew(&tlv320aic23_snd_controls[i],
+					       codec, NULL));
+		if (err < 0)
+			return err;
+	}
+
+	return 0;
+
+}
+
+/* PGA Mixer controls for Line and Mic switch */
+static const struct snd_kcontrol_new tlv320aic23_output_mixer_controls[] = {
+	SOC_DAPM_SINGLE("Line Bypass Switch", TLV320AIC23_ANLG, 3, 1, 0),
+	SOC_DAPM_SINGLE("Mic Sidetone Switch", TLV320AIC23_ANLG, 5, 1, 0),
+	SOC_DAPM_SINGLE("Playback Switch", TLV320AIC23_ANLG, 4, 1, 0),
+};
+
+static const struct snd_soc_dapm_widget tlv320aic23_dapm_widgets[] = {
+	SND_SOC_DAPM_DAC("DAC", "Playback", TLV320AIC23_PWR, 3, 1),
+	SND_SOC_DAPM_ADC("ADC", "Capture", TLV320AIC23_PWR, 2, 1),
+	SND_SOC_DAPM_MUX("Capture Source", SND_SOC_NOPM, 0, 0,
+			 &tlv320aic23_rec_src_mux_controls),
+	SND_SOC_DAPM_MIXER("Output Mixer", TLV320AIC23_PWR, 4, 1,
+			   &tlv320aic23_output_mixer_controls[0],
+			   ARRAY_SIZE(tlv320aic23_output_mixer_controls)),
+	SND_SOC_DAPM_PGA("Line Input", TLV320AIC23_PWR, 0, 1, NULL, 0),
+	SND_SOC_DAPM_PGA("Mic Input", TLV320AIC23_PWR, 1, 1, NULL, 0),
+
+	SND_SOC_DAPM_OUTPUT("LHPOUT"),
+	SND_SOC_DAPM_OUTPUT("RHPOUT"),
+	SND_SOC_DAPM_OUTPUT("LOUT"),
+	SND_SOC_DAPM_OUTPUT("ROUT"),
+
+	SND_SOC_DAPM_INPUT("LLINEIN"),
+	SND_SOC_DAPM_INPUT("RLINEIN"),
+
+	SND_SOC_DAPM_INPUT("MICIN"),
+};
+
+static const struct snd_soc_dapm_route intercon[] = {
+	/* Output Mixer */
+	{"Output Mixer", "Line Bypass Switch", "Line Input"},
+	{"Output Mixer", "Playback Switch", "DAC"},
+	{"Output Mixer", "Mic Sidetone Switch", "Mic Input"},
+
+	/* Outputs */
+	{"RHPOUT", NULL, "Output Mixer"},
+	{"LHPOUT", NULL, "Output Mixer"},
+	{"LOUT", NULL, "Output Mixer"},
+	{"ROUT", NULL, "Output Mixer"},
+
+	/* Inputs */
+	{"Line Input", "NULL", "LLINEIN"},
+	{"Line Input", "NULL", "RLINEIN"},
+
+	{"Mic Input", "NULL", "MICIN"},
+
+	/* input mux */
+	{"Capture Source", "Line", "Line Input"},
+	{"Capture Source", "Mic", "Mic Input"},
+	{"ADC", NULL, "Capture Source"},
+
+};
+
+/* tlv320aic23 related */
+static const struct tlv320aic23_srate_reg_info srate_reg_info[] = {
+	{4000, 0x06, 1},	/*  4000 */
+	{8000, 0x06, 0},	/*  8000 */
+	{16000, 0x0C, 1},	/* 16000 */
+	{22050, 0x11, 1},	/* 22050 */
+	{24000, 0x00, 1},	/* 24000 */
+	{32000, 0x0C, 0},	/* 32000 */
+	{44100, 0x11, 0},	/* 44100 */
+	{48000, 0x00, 0},	/* 48000 */
+	{88200, 0x1F, 0},	/* 88200 */
+	{96000, 0x0E, 0},	/* 96000 */
+};
+
+static int tlv320aic23_add_widgets(struct snd_soc_codec *codec)
+{
+	snd_soc_dapm_new_controls(codec, tlv320aic23_dapm_widgets,
+				  ARRAY_SIZE(tlv320aic23_dapm_widgets));
+
+	/* set up audio path interconnects */
+	snd_soc_dapm_add_routes(codec, intercon, ARRAY_SIZE(intercon));
+
+	snd_soc_dapm_new_widgets(codec);
+	return 0;
+}
+
+static int tlv320aic23_hw_params(struct snd_pcm_substream *substream,
+				 struct snd_pcm_hw_params *params)
+{
+	struct snd_soc_pcm_runtime *rtd = substream->private_data;
+	struct snd_soc_device *socdev = rtd->socdev;
+	struct snd_soc_codec *codec = socdev->codec;
+	u16 iface_reg, data;
+	u8 count = 0;
+
+	iface_reg =
+	    tlv320aic23_read_reg_cache(codec,
+				       TLV320AIC23_DIGT_FMT) & ~(0x03 << 2);
+
+	/* Search for the right sample rate */
+	/* Verify what happens if the rate is not supported
+	 * now it goes to 96Khz */
+	while ((srate_reg_info[count].sample_rate != params_rate(params)) &&
+	       (count < ARRAY_SIZE(srate_reg_info))) {
+		count++;
+	}
+
+	data =  (srate_reg_info[count].divider << TLV320AIC23_CLKIN_SHIFT) |
+		(srate_reg_info[count]. control << TLV320AIC23_BOSR_SHIFT) |
+		TLV320AIC23_USB_CLK_ON;
+
+	tlv320aic23_write(codec, TLV320AIC23_SRATE, data);
+
+	switch (params_format(params)) {
+	case SNDRV_PCM_FORMAT_S16_LE:
+		break;
+	case SNDRV_PCM_FORMAT_S20_3LE:
+		iface_reg |= (0x01 << 2);
+		break;
+	case SNDRV_PCM_FORMAT_S24_LE:
+		iface_reg |= (0x02 << 2);
+		break;
+	case SNDRV_PCM_FORMAT_S32_LE:
+		iface_reg |= (0x03 << 2);
+		break;
+	}
+	tlv320aic23_write(codec, TLV320AIC23_DIGT_FMT, iface_reg);
+
+	return 0;
+}
+
+static int tlv320aic23_pcm_prepare(struct snd_pcm_substream *substream)
+{
+	struct snd_soc_pcm_runtime *rtd = substream->private_data;
+	struct snd_soc_device *socdev = rtd->socdev;
+	struct snd_soc_codec *codec = socdev->codec;
+
+	/* set active */
+	tlv320aic23_write(codec, TLV320AIC23_ACTIVE, 0x0001);
+
+	return 0;
+}
+
+static void tlv320aic23_shutdown(struct snd_pcm_substream *substream)
+{
+	struct snd_soc_pcm_runtime *rtd = substream->private_data;
+	struct snd_soc_device *socdev = rtd->socdev;
+	struct snd_soc_codec *codec = socdev->codec;
+
+	/* deactivate */
+	if (!codec->active) {
+		udelay(50);
+		tlv320aic23_write(codec, TLV320AIC23_ACTIVE, 0x0);
+	}
+}
+
+static int tlv320aic23_mute(struct snd_soc_dai *dai, int mute)
+{
+	struct snd_soc_codec *codec = dai->codec;
+	u16 reg;
+
+	reg = tlv320aic23_read_reg_cache(codec, TLV320AIC23_DIGT);
+	if (mute)
+		reg |= TLV320AIC23_DACM_MUTE;
+
+	else
+		reg &= ~TLV320AIC23_DACM_MUTE;
+
+	tlv320aic23_write(codec, TLV320AIC23_DIGT, reg);
+
+	return 0;
+}
+
+static int tlv320aic23_set_dai_fmt(struct snd_soc_dai *codec_dai,
+				   unsigned int fmt)
+{
+	struct snd_soc_codec *codec = codec_dai->codec;
+	u16 iface_reg;
+
+	iface_reg =
+	    tlv320aic23_read_reg_cache(codec, TLV320AIC23_DIGT_FMT) & (~0x03);
+
+	/* set master/slave audio interface */
+	switch (fmt & SND_SOC_DAIFMT_MASTER_MASK) {
+	case SND_SOC_DAIFMT_CBM_CFM:
+		iface_reg |= TLV320AIC23_MS_MASTER;
+		break;
+	case SND_SOC_DAIFMT_CBS_CFS:
+		break;
+	default:
+		return -EINVAL;
+
+	}
+
+	/* interface format */
+	switch (fmt & SND_SOC_DAIFMT_FORMAT_MASK) {
+	case SND_SOC_DAIFMT_I2S:
+		iface_reg |= TLV320AIC23_FOR_I2S;
+		break;
+	case SND_SOC_DAIFMT_DSP_A:
+		iface_reg |= TLV320AIC23_FOR_DSP;
+		break;
+	case SND_SOC_DAIFMT_RIGHT_J:
+		break;
+	case SND_SOC_DAIFMT_LEFT_J:
+		iface_reg |= TLV320AIC23_FOR_LJUST;
+		break;
+	default:
+		return -EINVAL;
+
+	}
+
+	tlv320aic23_write(codec, TLV320AIC23_DIGT_FMT, iface_reg);
+
+	return 0;
+}
+
+static int tlv320aic23_set_dai_sysclk(struct snd_soc_dai *codec_dai,
+				      int clk_id, unsigned int freq, int dir)
+{
+	struct snd_soc_codec *codec = codec_dai->codec;
+
+	switch (freq) {
+	case 12000000:
+		return 0;
+	}
+	return -EINVAL;
+}
+
+static int tlv320aic23_set_bias_level(struct snd_soc_codec *codec,
+				      enum snd_soc_bias_level level)
+{
+	u16 reg = tlv320aic23_read_reg_cache(codec, TLV320AIC23_PWR) & 0xff7f;
+
+	switch (level) {
+	case SND_SOC_BIAS_ON:
+		/* vref/mid, osc on, dac unmute */
+		tlv320aic23_write(codec, TLV320AIC23_PWR, reg);
+		break;
+	case SND_SOC_BIAS_PREPARE:
+		break;
+	case SND_SOC_BIAS_STANDBY:
+		/* everything off except vref/vmid, */
+		tlv320aic23_write(codec, TLV320AIC23_PWR, reg | 0x0040);
+		break;
+	case SND_SOC_BIAS_OFF:
+		/* everything off, dac mute, inactive */
+		tlv320aic23_write(codec, TLV320AIC23_ACTIVE, 0x0);
+		tlv320aic23_write(codec, TLV320AIC23_PWR, 0xffff);
+		break;
+	}
+	codec->bias_level = level;
+	return 0;
+}
+
+#define AIC23_RATES	SNDRV_PCM_RATE_8000_96000
+#define AIC23_FORMATS	(SNDRV_PCM_FMTBIT_S16_LE | SNDRV_PCM_FMTBIT_S20_3LE | \
+			 SNDRV_PCM_FMTBIT_S24_3LE | SNDRV_PCM_FMTBIT_S32_LE)
+
+struct snd_soc_dai tlv320aic23_dai = {
+	.name = "tlv320aic23",
+	.playback = {
+		     .stream_name = "Playback",
+		     .channels_min = 2,
+		     .channels_max = 2,
+		     .rates = AIC23_RATES,
+		     .formats = AIC23_FORMATS,},
+	.capture = {
+		    .stream_name = "Capture",
+		    .channels_min = 2,
+		    .channels_max = 2,
+		    .rates = AIC23_RATES,
+		    .formats = AIC23_FORMATS,},
+	.ops = {
+		.prepare = tlv320aic23_pcm_prepare,
+		.hw_params = tlv320aic23_hw_params,
+		.shutdown = tlv320aic23_shutdown,
+		},
+	.dai_ops = {
+		    .digital_mute = tlv320aic23_mute,
+		    .set_fmt = tlv320aic23_set_dai_fmt,
+		    .set_sysclk = tlv320aic23_set_dai_sysclk,
+		    }
+};
+EXPORT_SYMBOL_GPL(tlv320aic23_dai);
+
+static int tlv320aic23_suspend(struct platform_device *pdev,
+			       pm_message_t state)
+{
+	struct snd_soc_device *socdev = platform_get_drvdata(pdev);
+	struct snd_soc_codec *codec = socdev->codec;
+
+	tlv320aic23_write(codec, TLV320AIC23_ACTIVE, 0x0);
+	tlv320aic23_set_bias_level(codec, SND_SOC_BIAS_OFF);
+
+	return 0;
+}
+
+static int tlv320aic23_resume(struct platform_device *pdev)
+{
+	struct snd_soc_device *socdev = platform_get_drvdata(pdev);
+	struct snd_soc_codec *codec = socdev->codec;
+	int i;
+	u16 reg;
+
+	/* Sync reg_cache with the hardware */
+	for (reg = 0; reg < ARRAY_SIZE(tlv320aic23_reg); i++) {
+		u16 val = tlv320aic23_read_reg_cache(codec, reg);
+		tlv320aic23_write(codec, reg, val);
+	}
+
+	tlv320aic23_set_bias_level(codec, SND_SOC_BIAS_STANDBY);
+	tlv320aic23_set_bias_level(codec, codec->suspend_bias_level);
+
+	return 0;
+}
+
+/*
+ * initialise the AIC23 driver
+ * register the mixer and dsp interfaces with the kernel
+ */
+static int tlv320aic23_init(struct snd_soc_device *socdev)
+{
+	struct snd_soc_codec *codec = socdev->codec;
+	int ret = 0;
+	u16 reg;
+
+	codec->name = "tlv320aic23";
+	codec->owner = THIS_MODULE;
+	codec->read = tlv320aic23_read_reg_cache;
+	codec->write = tlv320aic23_write;
+	codec->set_bias_level = tlv320aic23_set_bias_level;
+	codec->dai = &tlv320aic23_dai;
+	codec->num_dai = 1;
+	codec->reg_cache_size = ARRAY_SIZE(tlv320aic23_reg);
+	codec->reg_cache =
+	    kmemdup(tlv320aic23_reg, sizeof(tlv320aic23_reg), GFP_KERNEL);
+	if (codec->reg_cache == NULL)
+		return -ENOMEM;
+
+	/* Reset codec */
+	tlv320aic23_write(codec, TLV320AIC23_RESET, 0);
+
+	/* register pcms */
+	ret = snd_soc_new_pcms(socdev, SNDRV_DEFAULT_IDX1, SNDRV_DEFAULT_STR1);
+	if (ret < 0) {
+		printk(KERN_ERR "tlv320aic23: failed to create pcms\n");
+		goto pcm_err;
+	}
+
+	/* power on device */
+	tlv320aic23_set_bias_level(codec, SND_SOC_BIAS_STANDBY);
+
+	tlv320aic23_write(codec, TLV320AIC23_DIGT, TLV320AIC23_DEEMP_44K);
+
+	/* Unmute input */
+	reg = tlv320aic23_read_reg_cache(codec, TLV320AIC23_LINVOL);
+	tlv320aic23_write(codec, TLV320AIC23_LINVOL,
+			  (reg & (~TLV320AIC23_LIM_MUTED)) |
+			  (TLV320AIC23_LRS_ENABLED));
+
+	reg = tlv320aic23_read_reg_cache(codec, TLV320AIC23_RINVOL);
+	tlv320aic23_write(codec, TLV320AIC23_RINVOL,
+			  (reg & (~TLV320AIC23_LIM_MUTED)) |
+			  TLV320AIC23_LRS_ENABLED);
+
+	reg = tlv320aic23_read_reg_cache(codec, TLV320AIC23_ANLG);
+	tlv320aic23_write(codec, TLV320AIC23_ANLG,
+			 (reg) & (~TLV320AIC23_BYPASS_ON) &
+			 (~TLV320AIC23_MICM_MUTED));
+
+	/* Default output volume */
+	tlv320aic23_write(codec, TLV320AIC23_LCHNVOL,
+			  TLV320AIC23_DEFAULT_OUT_VOL &
+			  TLV320AIC23_OUT_VOL_MASK);
+	tlv320aic23_write(codec, TLV320AIC23_RCHNVOL,
+			  TLV320AIC23_DEFAULT_OUT_VOL &
+			  TLV320AIC23_OUT_VOL_MASK);
+
+	tlv320aic23_write(codec, TLV320AIC23_ACTIVE, 0x1);
+
+	tlv320aic23_add_controls(codec);
+	tlv320aic23_add_widgets(codec);
+	ret = snd_soc_register_card(socdev);
+	if (ret < 0) {
+		printk(KERN_ERR "tlv320aic23: failed to register card\n");
+		goto card_err;
+	}
+
+	return ret;
+
+card_err:
+	snd_soc_free_pcms(socdev);
+	snd_soc_dapm_free(socdev);
+pcm_err:
+	kfree(codec->reg_cache);
+	return ret;
+}
+static struct snd_soc_device *tlv320aic23_socdev;
+
+#if defined(CONFIG_I2C) || defined(CONFIG_I2C_MODULE)
+/*
+ * If the i2c layer weren't so broken, we could pass this kind of data
+ * around
+ */
+static int tlv320aic23_codec_probe(struct i2c_client *i2c,
+				   const struct i2c_device_id *i2c_id)
+{
+	struct snd_soc_device *socdev = tlv320aic23_socdev;
+	struct snd_soc_codec *codec = socdev->codec;
+	int ret;
+
+	if (!i2c_check_functionality(i2c->adapter, I2C_FUNC_SMBUS_BYTE_DATA))
+		return -EINVAL;
+
+	i2c_set_clientdata(i2c, codec);
+	codec->control_data = i2c;
+
+	ret = tlv320aic23_init(socdev);
+	if (ret < 0) {
+		printk(KERN_ERR "tlv320aic23: failed to initialise AIC23\n");
+		goto err;
+	}
+	return ret;
+
+err:
+	kfree(codec);
+	kfree(i2c);
+	return ret;
+}
+static int __exit tlv320aic23_i2c_remove(struct i2c_client *i2c)
+{
+	put_device(&i2c->dev);
+	return 0;
+}
+
+static const struct i2c_device_id tlv320aic23_id[] = {
+	{"tlv320aic23", 0},
+	{}
+};
+
+MODULE_DEVICE_TABLE(i2c, tlv320aic23_id);
+
+static struct i2c_driver tlv320aic23_i2c_driver = {
+	.driver = {
+		   .name = "tlv320aic23",
+		   },
+	.probe = tlv320aic23_codec_probe,
+	.remove = __exit_p(tlv320aic23_i2c_remove),
+	.id_table = tlv320aic23_id,
+};
+
+#endif
+
+static int tlv320aic23_probe(struct platform_device *pdev)
+{
+	struct snd_soc_device *socdev = platform_get_drvdata(pdev);
+	struct snd_soc_codec *codec;
+	int ret = 0;
+
+	printk(KERN_INFO "AIC23 Audio Codec %s\n", AIC23_VERSION);
+
+	codec = kzalloc(sizeof(struct snd_soc_codec), GFP_KERNEL);
+	if (codec == NULL)
+		return -ENOMEM;
+
+	socdev->codec = codec;
+	mutex_init(&codec->mutex);
+	INIT_LIST_HEAD(&codec->dapm_widgets);
+	INIT_LIST_HEAD(&codec->dapm_paths);
+
+	tlv320aic23_socdev = socdev;
+#if defined(CONFIG_I2C) || defined(CONFIG_I2C_MODULE)
+	codec->hw_write = (hw_write_t) i2c_smbus_write_byte_data;
+	codec->hw_read = NULL;
+	ret = i2c_add_driver(&tlv320aic23_i2c_driver);
+	if (ret != 0)
+		printk(KERN_ERR "can't add i2c driver");
+#endif
+	return ret;
+}
+
+static int tlv320aic23_remove(struct platform_device *pdev)
+{
+	struct snd_soc_device *socdev = platform_get_drvdata(pdev);
+	struct snd_soc_codec *codec = socdev->codec;
+
+	if (codec->control_data)
+		tlv320aic23_set_bias_level(codec, SND_SOC_BIAS_OFF);
+
+	snd_soc_free_pcms(socdev);
+	snd_soc_dapm_free(socdev);
+#if defined(CONFIG_I2C) || defined(CONFIG_I2C_MODULE)
+	i2c_del_driver(&tlv320aic23_i2c_driver);
+#endif
+	kfree(codec->reg_cache);
+	kfree(codec);
+
+	return 0;
+}
+struct snd_soc_codec_device soc_codec_dev_tlv320aic23 = {
+	.probe = tlv320aic23_probe,
+	.remove = tlv320aic23_remove,
+	.suspend = tlv320aic23_suspend,
+	.resume = tlv320aic23_resume,
+};
+EXPORT_SYMBOL_GPL(soc_codec_dev_tlv320aic23);
+
+MODULE_DESCRIPTION("ASoC TLV320AIC23 codec driver");
+MODULE_AUTHOR("Arun KS <arunks@mistralsolutions.com>");
+MODULE_LICENSE("GPL");
diff --git a/sound/soc/codecs/tlv320aic23.h b/sound/soc/codecs/tlv320aic23.h
new file mode 100644
index 000000000000..79d1faf8e570
--- /dev/null
+++ b/sound/soc/codecs/tlv320aic23.h
@@ -0,0 +1,122 @@
+/*
+ * ALSA SoC TLV320AIC23 codec driver
+ *
+ * Author:      Arun KS, <arunks@mistralsolutions.com>
+ * Copyright:   (C) 2008 Mistral Solutions Pvt Ltd
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef _TLV320AIC23_H
+#define _TLV320AIC23_H
+
+/* Codec TLV320AIC23 */
+#define TLV320AIC23_LINVOL		0x00
+#define TLV320AIC23_RINVOL		0x01
+#define TLV320AIC23_LCHNVOL		0x02
+#define TLV320AIC23_RCHNVOL		0x03
+#define TLV320AIC23_ANLG		0x04
+#define TLV320AIC23_DIGT		0x05
+#define TLV320AIC23_PWR			0x06
+#define TLV320AIC23_DIGT_FMT		0x07
+#define TLV320AIC23_SRATE		0x08
+#define TLV320AIC23_ACTIVE		0x09
+#define TLV320AIC23_RESET		0x0F
+
+/* Left (right) line input volume control register */
+#define TLV320AIC23_LRS_ENABLED		0x0100
+#define TLV320AIC23_LIM_MUTED		0x0080
+#define TLV320AIC23_LIV_DEFAULT		0x0017
+#define TLV320AIC23_LIV_MAX		0x001f
+#define TLV320AIC23_LIV_MIN		0x0000
+
+/* Left (right) channel headphone volume control register */
+#define TLV320AIC23_LZC_ON		0x0080
+#define TLV320AIC23_LHV_DEFAULT		0x0079
+#define TLV320AIC23_LHV_MAX		0x007f
+#define TLV320AIC23_LHV_MIN		0x0000
+
+/* Analog audio path control register */
+#define TLV320AIC23_STA_REG(x)		((x)<<6)
+#define TLV320AIC23_STE_ENABLED		0x0020
+#define TLV320AIC23_DAC_SELECTED	0x0010
+#define TLV320AIC23_BYPASS_ON		0x0008
+#define TLV320AIC23_INSEL_MIC		0x0004
+#define TLV320AIC23_MICM_MUTED		0x0002
+#define TLV320AIC23_MICB_20DB		0x0001
+
+/* Digital audio path control register */
+#define TLV320AIC23_DACM_MUTE		0x0008
+#define TLV320AIC23_DEEMP_32K		0x0002
+#define TLV320AIC23_DEEMP_44K		0x0004
+#define TLV320AIC23_DEEMP_48K		0x0006
+#define TLV320AIC23_ADCHP_ON		0x0001
+
+/* Power control down register */
+#define TLV320AIC23_DEVICE_PWR_OFF  	0x0080
+#define TLV320AIC23_CLK_OFF		0x0040
+#define TLV320AIC23_OSC_OFF		0x0020
+#define TLV320AIC23_OUT_OFF		0x0010
+#define TLV320AIC23_DAC_OFF		0x0008
+#define TLV320AIC23_ADC_OFF		0x0004
+#define TLV320AIC23_MIC_OFF		0x0002
+#define TLV320AIC23_LINE_OFF		0x0001
+
+/* Digital audio interface register */
+#define TLV320AIC23_MS_MASTER		0x0040
+#define TLV320AIC23_LRSWAP_ON		0x0020
+#define TLV320AIC23_LRP_ON		0x0010
+#define TLV320AIC23_IWL_16		0x0000
+#define TLV320AIC23_IWL_20		0x0004
+#define TLV320AIC23_IWL_24		0x0008
+#define TLV320AIC23_IWL_32		0x000C
+#define TLV320AIC23_FOR_I2S		0x0002
+#define TLV320AIC23_FOR_DSP		0x0003
+#define TLV320AIC23_FOR_LJUST		0x0001
+
+/* Sample rate control register */
+#define TLV320AIC23_CLKOUT_HALF		0x0080
+#define TLV320AIC23_CLKIN_HALF		0x0040
+#define TLV320AIC23_BOSR_384fs		0x0002	/* BOSR_272fs in USB mode */
+#define TLV320AIC23_USB_CLK_ON		0x0001
+#define TLV320AIC23_SR_MASK             0xf
+#define TLV320AIC23_CLKOUT_SHIFT        7
+#define TLV320AIC23_CLKIN_SHIFT         6
+#define TLV320AIC23_SR_SHIFT            2
+#define TLV320AIC23_BOSR_SHIFT          1
+
+/* Digital interface register */
+#define TLV320AIC23_ACT_ON		0x0001
+
+/*
+ * AUDIO related MACROS
+ */
+
+#define TLV320AIC23_DEFAULT_OUT_VOL	0x70
+#define TLV320AIC23_DEFAULT_IN_VOLUME	0x10
+
+#define TLV320AIC23_OUT_VOL_MIN		TLV320AIC23_LHV_MIN
+#define TLV320AIC23_OUT_VOL_MAX		TLV320AIC23_LHV_MAX
+#define TLV320AIC23_OUT_VO_RANGE	(TLV320AIC23_OUT_VOL_MAX - \
+					TLV320AIC23_OUT_VOL_MIN)
+#define TLV320AIC23_OUT_VOL_MASK	TLV320AIC23_OUT_VOL_MAX
+
+#define TLV320AIC23_IN_VOL_MIN		TLV320AIC23_LIV_MIN
+#define TLV320AIC23_IN_VOL_MAX		TLV320AIC23_LIV_MAX
+#define TLV320AIC23_IN_VOL_RANGE	(TLV320AIC23_IN_VOL_MAX - \
+					TLV320AIC23_IN_VOL_MIN)
+#define TLV320AIC23_IN_VOL_MASK		TLV320AIC23_IN_VOL_MAX
+
+#define TLV320AIC23_SIDETONE_MASK	0x1c0
+#define TLV320AIC23_SIDETONE_0		0x100
+#define TLV320AIC23_SIDETONE_6		0x000
+#define TLV320AIC23_SIDETONE_9		0x040
+#define TLV320AIC23_SIDETONE_12		0x080
+#define TLV320AIC23_SIDETONE_18		0x0c0
+
+extern struct snd_soc_dai tlv320aic23_dai;
+extern struct snd_soc_codec_device soc_codec_dev_tlv320aic23;
+
+#endif /* _TLV320AIC23_H */
diff --git a/sound/soc/codecs/tlv320aic3x.c b/sound/soc/codecs/tlv320aic3x.c
index 566a427c928f..05336ed7e493 100644
--- a/sound/soc/codecs/tlv320aic3x.c
+++ b/sound/soc/codecs/tlv320aic3x.c
@@ -48,7 +48,6 @@
 
 #include "tlv320aic3x.h"
 
-#define AUDIO_NAME "aic3x"
 #define AIC3X_VERSION "0.2"
 
 /* codec private data */
@@ -991,7 +990,7 @@ EXPORT_SYMBOL_GPL(aic3x_headset_detected);
 			 SNDRV_PCM_FMTBIT_S24_3LE | SNDRV_PCM_FMTBIT_S32_LE)
 
 struct snd_soc_dai aic3x_dai = {
-	.name = "aic3x",
+	.name = "tlv320aic3x",
 	.playback = {
 		.stream_name = "Playback",
 		.channels_min = 1,
@@ -1055,7 +1054,7 @@ static int aic3x_init(struct snd_soc_device *socdev)
 	struct aic3x_setup_data *setup = socdev->codec_data;
 	int reg, ret = 0;
 
-	codec->name = "aic3x";
+	codec->name = "tlv320aic3x";
 	codec->owner = THIS_MODULE;
 	codec->read = aic3x_read_reg_cache;
 	codec->write = aic3x_write;
diff --git a/sound/soc/codecs/uda1380.c b/sound/soc/codecs/uda1380.c
index d206d7f892b6..a69ee72a7af5 100644
--- a/sound/soc/codecs/uda1380.c
+++ b/sound/soc/codecs/uda1380.c
@@ -36,7 +36,6 @@
 #include "uda1380.h"
 
 #define UDA1380_VERSION "0.6"
-#define AUDIO_NAME "uda1380"
 
 /*
  * uda1380 register cache
diff --git a/sound/soc/codecs/wm8510.c b/sound/soc/codecs/wm8510.c
index 9a37c8d95ed2..d8ca2da8d634 100644
--- a/sound/soc/codecs/wm8510.c
+++ b/sound/soc/codecs/wm8510.c
@@ -3,7 +3,7 @@
  *
  * Copyright 2006 Wolfson Microelectronics PLC.
  *
- * Author: Liam Girdwood <liam.girdwood@wolfsonmicro.com>
+ * Author: Liam Girdwood <lrg@slimlogic.co.uk>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -18,6 +18,7 @@
 #include <linux/pm.h>
 #include <linux/i2c.h>
 #include <linux/platform_device.h>
+#include <linux/spi/spi.h>
 #include <sound/core.h>
 #include <sound/pcm.h>
 #include <sound/pcm_params.h>
@@ -27,7 +28,6 @@
 
 #include "wm8510.h"
 
-#define AUDIO_NAME "wm8510"
 #define WM8510_VERSION "0.6"
 
 struct snd_soc_codec_device soc_codec_dev_wm8510;
@@ -55,6 +55,9 @@ static const u16 wm8510_reg[WM8510_CACHEREGNUM] = {
 	0x0001,
 };
 
+#define WM8510_POWER1_BIASEN  0x08
+#define WM8510_POWER1_BUFIOEN 0x10
+
 /*
  * read wm8510 register cache
  */
@@ -224,9 +227,9 @@ SND_SOC_DAPM_PGA("SpkN Out", WM8510_POWER3, 5, 0, NULL, 0),
 SND_SOC_DAPM_PGA("SpkP Out", WM8510_POWER3, 6, 0, NULL, 0),
 SND_SOC_DAPM_PGA("Mono Out", WM8510_POWER3, 7, 0, NULL, 0),
 
-SND_SOC_DAPM_PGA("Mic PGA", WM8510_POWER2, 2, 0,
-		 &wm8510_micpga_controls[0],
-		 ARRAY_SIZE(wm8510_micpga_controls)),
+SND_SOC_DAPM_MIXER("Mic PGA", WM8510_POWER2, 2, 0,
+		   &wm8510_micpga_controls[0],
+		   ARRAY_SIZE(wm8510_micpga_controls)),
 SND_SOC_DAPM_MIXER("Boost Mixer", WM8510_POWER2, 4, 0,
 	&wm8510_boost_controls[0],
 	ARRAY_SIZE(wm8510_boost_controls)),
@@ -526,23 +529,35 @@ static int wm8510_mute(struct snd_soc_dai *dai, int mute)
 static int wm8510_set_bias_level(struct snd_soc_codec *codec,
 	enum snd_soc_bias_level level)
 {
+	u16 power1 = wm8510_read_reg_cache(codec, WM8510_POWER1) & ~0x3;
 
 	switch (level) {
 	case SND_SOC_BIAS_ON:
-		wm8510_write(codec, WM8510_POWER1, 0x1ff);
-		wm8510_write(codec, WM8510_POWER2, 0x1ff);
-		wm8510_write(codec, WM8510_POWER3, 0x1ff);
-		break;
 	case SND_SOC_BIAS_PREPARE:
+		power1 |= 0x1;  /* VMID 50k */
+		wm8510_write(codec, WM8510_POWER1, power1);
+		break;
+
 	case SND_SOC_BIAS_STANDBY:
+		power1 |= WM8510_POWER1_BIASEN | WM8510_POWER1_BUFIOEN;
+
+		if (codec->bias_level == SND_SOC_BIAS_OFF) {
+			/* Initial cap charge at VMID 5k */
+			wm8510_write(codec, WM8510_POWER1, power1 | 0x3);
+			mdelay(100);
+		}
+
+		power1 |= 0x2;  /* VMID 500k */
+		wm8510_write(codec, WM8510_POWER1, power1);
 		break;
+
 	case SND_SOC_BIAS_OFF:
-		/* everything off, dac mute, inactive */
-		wm8510_write(codec, WM8510_POWER1, 0x0);
-		wm8510_write(codec, WM8510_POWER2, 0x0);
-		wm8510_write(codec, WM8510_POWER3, 0x0);
+		wm8510_write(codec, WM8510_POWER1, 0);
+		wm8510_write(codec, WM8510_POWER2, 0);
+		wm8510_write(codec, WM8510_POWER3, 0);
 		break;
 	}
+
 	codec->bias_level = level;
 	return 0;
 }
@@ -640,6 +655,7 @@ static int wm8510_init(struct snd_soc_device *socdev)
 	}
 
 	/* power on device */
+	codec->bias_level = SND_SOC_BIAS_OFF;
 	wm8510_set_bias_level(codec, SND_SOC_BIAS_STANDBY);
 	wm8510_add_controls(codec);
 	wm8510_add_widgets(codec);
@@ -747,6 +763,62 @@ err_driver:
 }
 #endif
 
+#if defined(CONFIG_SPI_MASTER)
+static int __devinit wm8510_spi_probe(struct spi_device *spi)
+{
+	struct snd_soc_device *socdev = wm8510_socdev;
+	struct snd_soc_codec *codec = socdev->codec;
+	int ret;
+
+	codec->control_data = spi;
+
+	ret = wm8510_init(socdev);
+	if (ret < 0)
+		dev_err(&spi->dev, "failed to initialise WM8510\n");
+
+	return ret;
+}
+
+static int __devexit wm8510_spi_remove(struct spi_device *spi)
+{
+	return 0;
+}
+
+static struct spi_driver wm8510_spi_driver = {
+	.driver = {
+		.name	= "wm8510",
+		.bus	= &spi_bus_type,
+		.owner	= THIS_MODULE,
+	},
+	.probe		= wm8510_spi_probe,
+	.remove		= __devexit_p(wm8510_spi_remove),
+};
+
+static int wm8510_spi_write(struct spi_device *spi, const char *data, int len)
+{
+	struct spi_transfer t;
+	struct spi_message m;
+	u8 msg[2];
+
+	if (len <= 0)
+		return 0;
+
+	msg[0] = data[0];
+	msg[1] = data[1];
+
+	spi_message_init(&m);
+	memset(&t, 0, (sizeof t));
+
+	t.tx_buf = &msg[0];
+	t.len = len;
+
+	spi_message_add_tail(&t, &m);
+	spi_sync(spi, &m);
+
+	return len;
+}
+#endif /* CONFIG_SPI_MASTER */
+
 static int wm8510_probe(struct platform_device *pdev)
 {
 	struct snd_soc_device *socdev = platform_get_drvdata(pdev);
@@ -772,8 +844,14 @@ static int wm8510_probe(struct platform_device *pdev)
 		codec->hw_write = (hw_write_t)i2c_master_send;
 		ret = wm8510_add_i2c_device(pdev, setup);
 	}
-#else
-	/* Add other interfaces here */
+#endif
+#if defined(CONFIG_SPI_MASTER)
+	if (setup->spi) {
+		codec->hw_write = (hw_write_t)wm8510_spi_write;
+		ret = spi_register_driver(&wm8510_spi_driver);
+		if (ret != 0)
+			printk(KERN_ERR "can't add spi driver");
+	}
 #endif
 
 	if (ret != 0)
@@ -796,6 +874,9 @@ static int wm8510_remove(struct platform_device *pdev)
 	i2c_unregister_device(codec->control_data);
 	i2c_del_driver(&wm8510_i2c_driver);
 #endif
+#if defined(CONFIG_SPI_MASTER)
+	spi_unregister_driver(&wm8510_spi_driver);
+#endif
 	kfree(codec);
 
 	return 0;
diff --git a/sound/soc/codecs/wm8510.h b/sound/soc/codecs/wm8510.h
index c53683960456..bdefcf5c69ff 100644
--- a/sound/soc/codecs/wm8510.h
+++ b/sound/soc/codecs/wm8510.h
@@ -94,6 +94,7 @@
 #define WM8510_MCLKDIV_12	(7 << 5)
 
 struct wm8510_setup_data {
+	int spi;
 	int i2c_bus;
 	unsigned short i2c_address;
 };
diff --git a/sound/soc/codecs/wm8580.c b/sound/soc/codecs/wm8580.c
index df1ffbe305bf..627ebfb4209b 100644
--- a/sound/soc/codecs/wm8580.c
+++ b/sound/soc/codecs/wm8580.c
@@ -18,7 +18,6 @@
 
 #include <linux/module.h>
 #include <linux/moduleparam.h>
-#include <linux/version.h>
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/delay.h>
@@ -36,7 +35,6 @@
 
 #include "wm8580.h"
 
-#define AUDIO_NAME "wm8580"
 #define WM8580_VERSION "0.1"
 
 struct pll_state {
diff --git a/sound/soc/codecs/wm8731.c b/sound/soc/codecs/wm8731.c
index 7b64d9a7ff76..7f8a7e36b33e 100644
--- a/sound/soc/codecs/wm8731.c
+++ b/sound/soc/codecs/wm8731.c
@@ -29,7 +29,6 @@
 
 #include "wm8731.h"
 
-#define AUDIO_NAME "wm8731"
 #define WM8731_VERSION "0.13"
 
 struct snd_soc_codec_device soc_codec_dev_wm8731;
diff --git a/sound/soc/codecs/wm8750.c b/sound/soc/codecs/wm8750.c
index 4892e398a598..9b7296ee5b08 100644
--- a/sound/soc/codecs/wm8750.c
+++ b/sound/soc/codecs/wm8750.c
@@ -29,7 +29,6 @@
 
 #include "wm8750.h"
 
-#define AUDIO_NAME "WM8750"
 #define WM8750_VERSION "0.12"
 
 /* codec private data */
diff --git a/sound/soc/codecs/wm8753.c b/sound/soc/codecs/wm8753.c
index 8c4df44f3345..d426eaa22185 100644
--- a/sound/soc/codecs/wm8753.c
+++ b/sound/soc/codecs/wm8753.c
@@ -2,8 +2,7 @@
  * wm8753.c  --  WM8753 ALSA Soc Audio driver
  *
  * Copyright 2003 Wolfson Microelectronics PLC.
- * Author: Liam Girdwood
- *         liam.girdwood@wolfsonmicro.com or linux@wolfsonmicro.com
+ * Author: Liam Girdwood <lrg@slimlogic.co.uk>
  *
  *  This program is free software; you can redistribute  it and/or modify it
  *  under  the terms of  the GNU General  Public License as published by the
@@ -40,6 +39,7 @@
 #include <linux/pm.h>
 #include <linux/i2c.h>
 #include <linux/platform_device.h>
+#include <linux/spi/spi.h>
 #include <sound/core.h>
 #include <sound/pcm.h>
 #include <sound/pcm_params.h>
@@ -51,7 +51,6 @@
 
 #include "wm8753.h"
 
-#define AUDIO_NAME "wm8753"
 #define WM8753_VERSION "0.16"
 
 static int caps_charge = 2000;
@@ -1719,6 +1718,63 @@ err_driver:
 }
 #endif
 
+#if defined(CONFIG_SPI_MASTER)
+static int __devinit wm8753_spi_probe(struct spi_device *spi)
+{
+	struct snd_soc_device *socdev = wm8753_socdev;
+	struct snd_soc_codec *codec = socdev->codec;
+	int ret;
+
+	codec->control_data = spi;
+
+	ret = wm8753_init(socdev);
+	if (ret < 0)
+		dev_err(&spi->dev, "failed to initialise WM8753\n");
+
+	return ret;
+}
+
+static int __devexit wm8753_spi_remove(struct spi_device *spi)
+{
+	return 0;
+}
+
+static struct spi_driver wm8753_spi_driver = {
+	.driver = {
+		.name	= "wm8753",
+		.bus	= &spi_bus_type,
+		.owner	= THIS_MODULE,
+	},
+	.probe		= wm8753_spi_probe,
+	.remove		= __devexit_p(wm8753_spi_remove),
+};
+
+static int wm8753_spi_write(struct spi_device *spi, const char *data, int len)
+{
+	struct spi_transfer t;
+	struct spi_message m;
+	u8 msg[2];
+
+	if (len <= 0)
+		return 0;
+
+	msg[0] = data[0];
+	msg[1] = data[1];
+
+	spi_message_init(&m);
+	memset(&t, 0, (sizeof t));
+
+	t.tx_buf = &msg[0];
+	t.len = len;
+
+	spi_message_add_tail(&t, &m);
+	spi_sync(spi, &m);
+
+	return len;
+}
+#endif
+
+
 static int wm8753_probe(struct platform_device *pdev)
 {
 	struct snd_soc_device *socdev = platform_get_drvdata(pdev);
@@ -1753,8 +1809,14 @@ static int wm8753_probe(struct platform_device *pdev)
 		codec->hw_write = (hw_write_t)i2c_master_send;
 		ret = wm8753_add_i2c_device(pdev, setup);
 	}
-#else
-		/* Add other interfaces here */
+#endif
+#if defined(CONFIG_SPI_MASTER)
+	if (setup->spi) {
+		codec->hw_write = (hw_write_t)wm8753_spi_write;
+		ret = spi_register_driver(&wm8753_spi_driver);
+		if (ret != 0)
+			printk(KERN_ERR "can't add spi driver");
+	}
 #endif
 
 	if (ret != 0) {
@@ -1798,6 +1860,9 @@ static int wm8753_remove(struct platform_device *pdev)
 	i2c_unregister_device(codec->control_data);
 	i2c_del_driver(&wm8753_i2c_driver);
 #endif
+#if defined(CONFIG_SPI_MASTER)
+	spi_unregister_driver(&wm8753_spi_driver);
+#endif
 	kfree(codec->private_data);
 	kfree(codec);
 
diff --git a/sound/soc/codecs/wm8753.h b/sound/soc/codecs/wm8753.h
index 7defde069f1d..f55704ce931b 100644
--- a/sound/soc/codecs/wm8753.h
+++ b/sound/soc/codecs/wm8753.h
@@ -2,8 +2,7 @@
  * wm8753.h  --  audio driver for WM8753
  *
  * Copyright 2003 Wolfson Microelectronics PLC.
- * Author: Liam Girdwood
- *         liam.girdwood@wolfsonmicro.com or linux@wolfsonmicro.com
+ * Author: Liam Girdwood <lrg@slimlogic.co.uk>
  *
  *  This program is free software; you can redistribute  it and/or modify it
  *  under  the terms of  the GNU General  Public License as published by the
@@ -79,6 +78,7 @@
 #define WM8753_ADCTL2		0x3f
 
 struct wm8753_setup_data {
+	int spi;
 	int i2c_bus;
 	unsigned short i2c_address;
 };
diff --git a/sound/soc/codecs/wm8900.c b/sound/soc/codecs/wm8900.c
index 0b8c6d38b48f..3b326c9b5586 100644
--- a/sound/soc/codecs/wm8900.c
+++ b/sound/soc/codecs/wm8900.c
@@ -18,7 +18,6 @@
 
 #include <linux/module.h>
 #include <linux/moduleparam.h>
-#include <linux/version.h>
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/delay.h>
diff --git a/sound/soc/codecs/wm8903.c b/sound/soc/codecs/wm8903.c
index a3f54ec4226e..ce40d7877605 100644
--- a/sound/soc/codecs/wm8903.c
+++ b/sound/soc/codecs/wm8903.c
@@ -653,14 +653,14 @@ static const struct snd_kcontrol_new wm8903_snd_controls[] = {
 
 /* Input PGAs - No TLV since the scale depends on PGA mode */
 SOC_SINGLE("Left Input PGA Switch", WM8903_ANALOGUE_LEFT_INPUT_0,
-	   7, 1, 0),
+	   7, 1, 1),
 SOC_SINGLE("Left Input PGA Volume", WM8903_ANALOGUE_LEFT_INPUT_0,
 	   0, 31, 0),
 SOC_SINGLE("Left Input PGA Common Mode Switch", WM8903_ANALOGUE_LEFT_INPUT_1,
 	   6, 1, 0),
 
 SOC_SINGLE("Right Input PGA Switch", WM8903_ANALOGUE_RIGHT_INPUT_0,
-	   7, 1, 0),
+	   7, 1, 1),
 SOC_SINGLE("Right Input PGA Volume", WM8903_ANALOGUE_RIGHT_INPUT_0,
 	   0, 31, 0),
 SOC_SINGLE("Right Input PGA Common Mode Switch", WM8903_ANALOGUE_RIGHT_INPUT_1,
diff --git a/sound/soc/codecs/wm8971.c b/sound/soc/codecs/wm8971.c
index 974a4cd0f3fd..f41a578ddd4f 100644
--- a/sound/soc/codecs/wm8971.c
+++ b/sound/soc/codecs/wm8971.c
@@ -29,7 +29,6 @@
 
 #include "wm8971.h"
 
-#define AUDIO_NAME "wm8971"
 #define WM8971_VERSION "0.9"
 
 #define	WM8971_REG_COUNT		43
diff --git a/sound/soc/codecs/wm8990.c b/sound/soc/codecs/wm8990.c
index 63410d7b5efb..572d22b0880b 100644
--- a/sound/soc/codecs/wm8990.c
+++ b/sound/soc/codecs/wm8990.c
@@ -30,7 +30,6 @@
 
 #include "wm8990.h"
 
-#define AUDIO_NAME "wm8990"
 #define WM8990_VERSION "0.2"
 
 /* codec private data */
diff --git a/sound/soc/codecs/wm9712.c b/sound/soc/codecs/wm9712.c
index 2f1c91b1d556..ffb471e420e2 100644
--- a/sound/soc/codecs/wm9712.c
+++ b/sound/soc/codecs/wm9712.c
@@ -2,8 +2,7 @@
  * wm9712.c  --  ALSA Soc WM9712 codec support
  *
  * Copyright 2006 Wolfson Microelectronics PLC.
- * Author: Liam Girdwood
- *         liam.girdwood@wolfsonmicro.com or linux@wolfsonmicro.com
+ * Author: Liam Girdwood <lrg@slimlogic.co.uk>
  *
  *  This program is free software; you can redistribute  it and/or modify it
  *  under  the terms of  the GNU General  Public License as published by the
diff --git a/sound/soc/codecs/wm9713.c b/sound/soc/codecs/wm9713.c
index 441d0580db1f..aba402b3c999 100644
--- a/sound/soc/codecs/wm9713.c
+++ b/sound/soc/codecs/wm9713.c
@@ -2,8 +2,7 @@
  * wm9713.c  --  ALSA Soc WM9713 codec support
  *
  * Copyright 2006 Wolfson Microelectronics PLC.
- * Author: Liam Girdwood
- *         liam.girdwood@wolfsonmicro.com or linux@wolfsonmicro.com
+ * Author: Liam Girdwood <lrg@slimlogic.co.uk>
  *
  *  This program is free software; you can redistribute  it and/or modify it
  *  under  the terms of  the GNU General  Public License as published by the
diff --git a/sound/soc/omap/Kconfig b/sound/soc/omap/Kconfig
index aea27e70043c..8b7766b998d7 100644
--- a/sound/soc/omap/Kconfig
+++ b/sound/soc/omap/Kconfig
@@ -13,3 +13,11 @@ config SND_OMAP_SOC_N810
 	select SND_SOC_TLV320AIC3X
 	help
 	  Say Y if you want to add support for SoC audio on Nokia N810.
+
+config SND_OMAP_SOC_OSK5912
+	tristate "SoC Audio support for omap osk5912"
+	depends on SND_OMAP_SOC && MACH_OMAP_OSK
+	select SND_OMAP_SOC_MCBSP
+	select SND_SOC_TLV320AIC23
+	help
+	  Say Y if you want to add support for SoC audio on osk5912.
diff --git a/sound/soc/omap/Makefile b/sound/soc/omap/Makefile
index d8d8d58075e3..e09d1f297f64 100644
--- a/sound/soc/omap/Makefile
+++ b/sound/soc/omap/Makefile
@@ -7,5 +7,7 @@ obj-$(CONFIG_SND_OMAP_SOC_MCBSP) += snd-soc-omap-mcbsp.o
 
 # OMAP Machine Support
 snd-soc-n810-objs := n810.o
+snd-soc-osk5912-objs := osk5912.o
 
 obj-$(CONFIG_SND_OMAP_SOC_N810) += snd-soc-n810.o
+obj-$(CONFIG_SND_OMAP_SOC_OSK5912) += snd-soc-osk5912.o
diff --git a/sound/soc/omap/n810.c b/sound/soc/omap/n810.c
index d166b6b2a60d..fae3ad36e0bf 100644
--- a/sound/soc/omap/n810.c
+++ b/sound/soc/omap/n810.c
@@ -247,9 +247,9 @@ static int n810_aic33_init(struct snd_soc_codec *codec)
 	int i, err;
 
 	/* Not connected */
-	snd_soc_dapm_disable_pin(codec, "MONO_LOUT");
-	snd_soc_dapm_disable_pin(codec, "HPLCOM");
-	snd_soc_dapm_disable_pin(codec, "HPRCOM");
+	snd_soc_dapm_nc_pin(codec, "MONO_LOUT");
+	snd_soc_dapm_nc_pin(codec, "HPLCOM");
+	snd_soc_dapm_nc_pin(codec, "HPRCOM");
 
 	/* Add N810 specific controls */
 	for (i = 0; i < ARRAY_SIZE(aic33_n810_controls); i++) {
diff --git a/sound/soc/omap/omap-mcbsp.c b/sound/soc/omap/omap-mcbsp.c
index 35310e16d7f3..0a063a98a661 100644
--- a/sound/soc/omap/omap-mcbsp.c
+++ b/sound/soc/omap/omap-mcbsp.c
@@ -59,12 +59,7 @@ static struct omap_mcbsp_data mcbsp_data[NUM_LINKS];
  * Stream DMA parameters. DMA request line and port address are set runtime
  * since they are different between OMAP1 and later OMAPs
  */
-static struct omap_pcm_dma_data omap_mcbsp_dai_dma_params[NUM_LINKS][2] = {
-{
-	{ .name		= "I2S PCM Stereo out", },
-	{ .name		= "I2S PCM Stereo in", },
-},
-};
+static struct omap_pcm_dma_data omap_mcbsp_dai_dma_params[NUM_LINKS][2];
 
 #if defined(CONFIG_ARCH_OMAP15XX) || defined(CONFIG_ARCH_OMAP16XX)
 static const int omap1_dma_reqs[][2] = {
@@ -84,11 +79,22 @@ static const unsigned long omap1_mcbsp_port[][2] = {
 static const int omap1_dma_reqs[][2] = {};
 static const unsigned long omap1_mcbsp_port[][2] = {};
 #endif
-#if defined(CONFIG_ARCH_OMAP2420)
-static const int omap2420_dma_reqs[][2] = {
+
+#if defined(CONFIG_ARCH_OMAP24XX) || defined(CONFIG_ARCH_OMAP34XX)
+static const int omap24xx_dma_reqs[][2] = {
 	{ OMAP24XX_DMA_MCBSP1_TX, OMAP24XX_DMA_MCBSP1_RX },
 	{ OMAP24XX_DMA_MCBSP2_TX, OMAP24XX_DMA_MCBSP2_RX },
+#if defined(CONFIG_ARCH_OMAP2430) || defined(CONFIG_ARCH_OMAP34XX)
+	{ OMAP24XX_DMA_MCBSP3_TX, OMAP24XX_DMA_MCBSP3_RX },
+	{ OMAP24XX_DMA_MCBSP4_TX, OMAP24XX_DMA_MCBSP4_RX },
+	{ OMAP24XX_DMA_MCBSP5_TX, OMAP24XX_DMA_MCBSP5_RX },
+#endif
 };
+#else
+static const int omap24xx_dma_reqs[][2] = {};
+#endif
+
+#if defined(CONFIG_ARCH_OMAP2420)
 static const unsigned long omap2420_mcbsp_port[][2] = {
 	{ OMAP24XX_MCBSP1_BASE + OMAP_MCBSP_REG_DXR1,
 	  OMAP24XX_MCBSP1_BASE + OMAP_MCBSP_REG_DRR1 },
@@ -96,10 +102,43 @@ static const unsigned long omap2420_mcbsp_port[][2] = {
 	  OMAP24XX_MCBSP2_BASE + OMAP_MCBSP_REG_DRR1 },
 };
 #else
-static const int omap2420_dma_reqs[][2] = {};
 static const unsigned long omap2420_mcbsp_port[][2] = {};
 #endif
 
+#if defined(CONFIG_ARCH_OMAP2430)
+static const unsigned long omap2430_mcbsp_port[][2] = {
+	{ OMAP24XX_MCBSP1_BASE + OMAP_MCBSP_REG_DXR,
+	  OMAP24XX_MCBSP1_BASE + OMAP_MCBSP_REG_DRR },
+	{ OMAP24XX_MCBSP2_BASE + OMAP_MCBSP_REG_DXR,
+	  OMAP24XX_MCBSP2_BASE + OMAP_MCBSP_REG_DRR },
+	{ OMAP2430_MCBSP3_BASE + OMAP_MCBSP_REG_DXR,
+	  OMAP2430_MCBSP3_BASE + OMAP_MCBSP_REG_DRR },
+	{ OMAP2430_MCBSP4_BASE + OMAP_MCBSP_REG_DXR,
+	  OMAP2430_MCBSP4_BASE + OMAP_MCBSP_REG_DRR },
+	{ OMAP2430_MCBSP5_BASE + OMAP_MCBSP_REG_DXR,
+	  OMAP2430_MCBSP5_BASE + OMAP_MCBSP_REG_DRR },
+};
+#else
+static const unsigned long omap2430_mcbsp_port[][2] = {};
+#endif
+
+#if defined(CONFIG_ARCH_OMAP34XX)
+static const unsigned long omap34xx_mcbsp_port[][2] = {
+	{ OMAP34XX_MCBSP1_BASE + OMAP_MCBSP_REG_DXR,
+	  OMAP34XX_MCBSP1_BASE + OMAP_MCBSP_REG_DRR },
+	{ OMAP34XX_MCBSP2_BASE + OMAP_MCBSP_REG_DXR,
+	  OMAP34XX_MCBSP2_BASE + OMAP_MCBSP_REG_DRR },
+	{ OMAP34XX_MCBSP3_BASE + OMAP_MCBSP_REG_DXR,
+	  OMAP34XX_MCBSP3_BASE + OMAP_MCBSP_REG_DRR },
+	{ OMAP34XX_MCBSP4_BASE + OMAP_MCBSP_REG_DXR,
+	  OMAP34XX_MCBSP4_BASE + OMAP_MCBSP_REG_DRR },
+	{ OMAP34XX_MCBSP5_BASE + OMAP_MCBSP_REG_DXR,
+	  OMAP34XX_MCBSP5_BASE + OMAP_MCBSP_REG_DRR },
+};
+#else
+static const unsigned long omap34xx_mcbsp_port[][2] = {};
+#endif
+
 static int omap_mcbsp_dai_startup(struct snd_pcm_substream *substream)
 {
 	struct snd_soc_pcm_runtime *rtd = substream->private_data;
@@ -167,14 +206,19 @@ static int omap_mcbsp_dai_hw_params(struct snd_pcm_substream *substream,
 		dma = omap1_dma_reqs[bus_id][substream->stream];
 		port = omap1_mcbsp_port[bus_id][substream->stream];
 	} else if (cpu_is_omap2420()) {
-		dma = omap2420_dma_reqs[bus_id][substream->stream];
+		dma = omap24xx_dma_reqs[bus_id][substream->stream];
 		port = omap2420_mcbsp_port[bus_id][substream->stream];
+	} else if (cpu_is_omap2430()) {
+		dma = omap24xx_dma_reqs[bus_id][substream->stream];
+		port = omap2430_mcbsp_port[bus_id][substream->stream];
+	} else if (cpu_is_omap343x()) {
+		dma = omap24xx_dma_reqs[bus_id][substream->stream];
+		port = omap34xx_mcbsp_port[bus_id][substream->stream];
 	} else {
-		/*
-		 * TODO: Add support for 2430 and 3430
-		 */
 		return -ENODEV;
 	}
+	omap_mcbsp_dai_dma_params[id][substream->stream].name =
+		substream->stream ? "Audio Capture" : "Audio Playback";
 	omap_mcbsp_dai_dma_params[id][substream->stream].dma_req = dma;
 	omap_mcbsp_dai_dma_params[id][substream->stream].port_addr = port;
 	cpu_dai->dma_data = &omap_mcbsp_dai_dma_params[id][substream->stream];
@@ -245,6 +289,11 @@ static int omap_mcbsp_dai_set_dai_fmt(struct snd_soc_dai *cpu_dai,
 		regs->rcr2	|= RDATDLY(1);
 		regs->xcr2	|= XDATDLY(1);
 		break;
+	case SND_SOC_DAIFMT_DSP_A:
+		/* 0-bit data delay */
+		regs->rcr2      |= RDATDLY(0);
+		regs->xcr2      |= XDATDLY(0);
+		break;
 	default:
 		/* Unsupported data format */
 		return -EINVAL;
@@ -310,7 +359,7 @@ static int omap_mcbsp_dai_set_clks_src(struct omap_mcbsp_data *mcbsp_data,
 				       int clk_id)
 {
 	int sel_bit;
-	u16 reg;
+	u16 reg, reg_devconf1 = OMAP243X_CONTROL_DEVCONF1;
 
 	if (cpu_class_is_omap1()) {
 		/* OMAP1's can use only external source clock */
@@ -320,6 +369,12 @@ static int omap_mcbsp_dai_set_clks_src(struct omap_mcbsp_data *mcbsp_data,
 			return 0;
 	}
 
+	if (cpu_is_omap2420() && mcbsp_data->bus_id > 1)
+		return -EINVAL;
+
+	if (cpu_is_omap343x())
+		reg_devconf1 = OMAP343X_CONTROL_DEVCONF1;
+
 	switch (mcbsp_data->bus_id) {
 	case 0:
 		reg = OMAP2_CONTROL_DEVCONF0;
@@ -329,20 +384,26 @@ static int omap_mcbsp_dai_set_clks_src(struct omap_mcbsp_data *mcbsp_data,
 		reg = OMAP2_CONTROL_DEVCONF0;
 		sel_bit = 6;
 		break;
-	/* TODO: Support for ports 3 - 5 in OMAP2430 and OMAP34xx */
+	case 2:
+		reg = reg_devconf1;
+		sel_bit = 0;
+		break;
+	case 3:
+		reg = reg_devconf1;
+		sel_bit = 2;
+		break;
+	case 4:
+		reg = reg_devconf1;
+		sel_bit = 4;
+		break;
 	default:
 		return -EINVAL;
 	}
 
-	if (cpu_class_is_omap2()) {
-		if (clk_id == OMAP_MCBSP_SYSCLK_CLKS_FCLK) {
-			omap_ctrl_writel(omap_ctrl_readl(reg) &
-					 ~(1 << sel_bit), reg);
-		} else {
-			omap_ctrl_writel(omap_ctrl_readl(reg) |
-					 (1 << sel_bit), reg);
-		}
-	}
+	if (clk_id == OMAP_MCBSP_SYSCLK_CLKS_FCLK)
+		omap_ctrl_writel(omap_ctrl_readl(reg) & ~(1 << sel_bit), reg);
+	else
+		omap_ctrl_writel(omap_ctrl_readl(reg) | (1 << sel_bit), reg);
 
 	return 0;
 }
@@ -376,37 +437,49 @@ static int omap_mcbsp_dai_set_dai_sysclk(struct snd_soc_dai *cpu_dai,
 	return err;
 }
 
-struct snd_soc_dai omap_mcbsp_dai[NUM_LINKS] = {
-{
-	.name = "omap-mcbsp-dai",
-	.id = 0,
-	.type = SND_SOC_DAI_I2S,
-	.playback = {
-		.channels_min = 2,
-		.channels_max = 2,
-		.rates = OMAP_MCBSP_RATES,
-		.formats = SNDRV_PCM_FMTBIT_S16_LE,
-	},
-	.capture = {
-		.channels_min = 2,
-		.channels_max = 2,
-		.rates = OMAP_MCBSP_RATES,
-		.formats = SNDRV_PCM_FMTBIT_S16_LE,
-	},
-	.ops = {
-		.startup = omap_mcbsp_dai_startup,
-		.shutdown = omap_mcbsp_dai_shutdown,
-		.trigger = omap_mcbsp_dai_trigger,
-		.hw_params = omap_mcbsp_dai_hw_params,
-	},
-	.dai_ops = {
-		.set_fmt = omap_mcbsp_dai_set_dai_fmt,
-		.set_clkdiv = omap_mcbsp_dai_set_clkdiv,
-		.set_sysclk = omap_mcbsp_dai_set_dai_sysclk,
-	},
-	.private_data = &mcbsp_data[0].bus_id,
-},
+#define OMAP_MCBSP_DAI_BUILDER(link_id)				\
+{								\
+	.name = "omap-mcbsp-dai-(link_id)",			\
+	.id = (link_id),					\
+	.type = SND_SOC_DAI_I2S,				\
+	.playback = {						\
+		.channels_min = 2,				\
+		.channels_max = 2,				\
+		.rates = OMAP_MCBSP_RATES,			\
+		.formats = SNDRV_PCM_FMTBIT_S16_LE,		\
+	},							\
+	.capture = {						\
+		.channels_min = 2,				\
+		.channels_max = 2,				\
+		.rates = OMAP_MCBSP_RATES,			\
+		.formats = SNDRV_PCM_FMTBIT_S16_LE,		\
+	},							\
+	.ops = {						\
+		.startup = omap_mcbsp_dai_startup,		\
+		.shutdown = omap_mcbsp_dai_shutdown,		\
+		.trigger = omap_mcbsp_dai_trigger,		\
+		.hw_params = omap_mcbsp_dai_hw_params,		\
+	},							\
+	.dai_ops = {						\
+		.set_fmt = omap_mcbsp_dai_set_dai_fmt,		\
+		.set_clkdiv = omap_mcbsp_dai_set_clkdiv,	\
+		.set_sysclk = omap_mcbsp_dai_set_dai_sysclk,	\
+	},							\
+	.private_data = &mcbsp_data[(link_id)].bus_id,		\
+}
+
+struct snd_soc_dai omap_mcbsp_dai[] = {
+	OMAP_MCBSP_DAI_BUILDER(0),
+	OMAP_MCBSP_DAI_BUILDER(1),
+#if NUM_LINKS >= 3
+	OMAP_MCBSP_DAI_BUILDER(2),
+#endif
+#if NUM_LINKS == 5
+	OMAP_MCBSP_DAI_BUILDER(3),
+	OMAP_MCBSP_DAI_BUILDER(4),
+#endif
 };
+
 EXPORT_SYMBOL_GPL(omap_mcbsp_dai);
 
 MODULE_AUTHOR("Jarkko Nikula <jarkko.nikula@nokia.com>");
diff --git a/sound/soc/omap/omap-mcbsp.h b/sound/soc/omap/omap-mcbsp.h
index ed8afb550671..df7ad13ba73d 100644
--- a/sound/soc/omap/omap-mcbsp.h
+++ b/sound/soc/omap/omap-mcbsp.h
@@ -38,11 +38,17 @@ enum omap_mcbsp_div {
 	OMAP_MCBSP_CLKGDV,		/* Sample rate generator divider */
 };
 
-/*
- * REVISIT: Preparation for the ASoC v2. Let the number of available links to
- * be same than number of McBSP ports found in OMAP(s) we are compiling for.
- */
-#define NUM_LINKS	1
+#if defined(CONFIG_ARCH_OMAP2420)
+#define NUM_LINKS	2
+#endif
+#if defined(CONFIG_ARCH_OMAP15XX) || defined(CONFIG_ARCH_OMAP16XX)
+#undef  NUM_LINKS
+#define NUM_LINKS	3
+#endif
+#if defined(CONFIG_ARCH_OMAP2430) || defined(CONFIG_ARCH_OMAP34XX)
+#undef  NUM_LINKS
+#define NUM_LINKS	5
+#endif
 
 extern struct snd_soc_dai omap_mcbsp_dai[NUM_LINKS];
 
diff --git a/sound/soc/omap/omap-pcm.c b/sound/soc/omap/omap-pcm.c
index 690bfeaec4a0..e9084fdd2082 100644
--- a/sound/soc/omap/omap-pcm.c
+++ b/sound/soc/omap/omap-pcm.c
@@ -97,7 +97,7 @@ static int omap_pcm_hw_params(struct snd_pcm_substream *substream,
 	prtd->dma_data = dma_data;
 	err = omap_request_dma(dma_data->dma_req, dma_data->name,
 			       omap_pcm_dma_irq, substream, &prtd->dma_ch);
-	if (!cpu_is_omap1510()) {
+	if (!err & !cpu_is_omap1510()) {
 		/*
 		 * Link channel with itself so DMA doesn't need any
 		 * reprogramming while looping the buffer
@@ -147,12 +147,14 @@ static int omap_pcm_prepare(struct snd_pcm_substream *substream)
 		dma_params.src_or_dst_synch	= OMAP_DMA_DST_SYNC;
 		dma_params.src_start		= runtime->dma_addr;
 		dma_params.dst_start		= dma_data->port_addr;
+		dma_params.dst_port		= OMAP_DMA_PORT_MPUI;
 	} else {
 		dma_params.src_amode		= OMAP_DMA_AMODE_CONSTANT;
 		dma_params.dst_amode		= OMAP_DMA_AMODE_POST_INC;
 		dma_params.src_or_dst_synch	= OMAP_DMA_SRC_SYNC;
 		dma_params.src_start		= dma_data->port_addr;
 		dma_params.dst_start		= runtime->dma_addr;
+		dma_params.src_port		= OMAP_DMA_PORT_MPUI;
 	}
 	/*
 	 * Set DMA transfer frame size equal to ALSA period size and frame
diff --git a/sound/soc/omap/osk5912.c b/sound/soc/omap/osk5912.c
new file mode 100644
index 000000000000..0fe733796898
--- /dev/null
+++ b/sound/soc/omap/osk5912.c
@@ -0,0 +1,232 @@
+/*
+ * osk5912.c  --  SoC audio for OSK 5912
+ *
+ * Copyright (C) 2008 Mistral Solutions
+ *
+ * Contact: Arun KS  <arunks@mistralsolutions.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
+ * 02110-1301 USA
+ *
+ */
+
+#include <linux/clk.h>
+#include <linux/platform_device.h>
+#include <sound/core.h>
+#include <sound/pcm.h>
+#include <sound/soc.h>
+#include <sound/soc-dapm.h>
+
+#include <asm/mach-types.h>
+#include <mach/hardware.h>
+#include <linux/gpio.h>
+#include <mach/mcbsp.h>
+
+#include "omap-mcbsp.h"
+#include "omap-pcm.h"
+#include "../codecs/tlv320aic23.h"
+
+#define CODEC_CLOCK 	12000000
+
+static struct clk *tlv320aic23_mclk;
+
+static int osk_startup(struct snd_pcm_substream *substream)
+{
+	return clk_enable(tlv320aic23_mclk);
+}
+
+static void osk_shutdown(struct snd_pcm_substream *substream)
+{
+	clk_disable(tlv320aic23_mclk);
+}
+
+static int osk_hw_params(struct snd_pcm_substream *substream,
+			 struct snd_pcm_hw_params *params)
+{
+	struct snd_soc_pcm_runtime *rtd = substream->private_data;
+	struct snd_soc_dai *codec_dai = rtd->dai->codec_dai;
+	struct snd_soc_dai *cpu_dai = rtd->dai->cpu_dai;
+	int err;
+
+	/* Set codec DAI configuration */
+	err = snd_soc_dai_set_fmt(codec_dai,
+				  SND_SOC_DAIFMT_DSP_A |
+				  SND_SOC_DAIFMT_NB_IF |
+				  SND_SOC_DAIFMT_CBM_CFM);
+	if (err < 0) {
+		printk(KERN_ERR "can't set codec DAI configuration\n");
+		return err;
+	}
+
+	/* Set cpu DAI configuration */
+	err = snd_soc_dai_set_fmt(cpu_dai,
+				  SND_SOC_DAIFMT_DSP_A |
+				  SND_SOC_DAIFMT_NB_IF |
+				  SND_SOC_DAIFMT_CBM_CFM);
+	if (err < 0) {
+		printk(KERN_ERR "can't set cpu DAI configuration\n");
+		return err;
+	}
+
+	/* Set the codec system clock for DAC and ADC */
+	err =
+	    snd_soc_dai_set_sysclk(codec_dai, 0, CODEC_CLOCK, SND_SOC_CLOCK_IN);
+
+	if (err < 0) {
+		printk(KERN_ERR "can't set codec system clock\n");
+		return err;
+	}
+
+	return err;
+}
+
+static struct snd_soc_ops osk_ops = {
+	.startup = osk_startup,
+	.hw_params = osk_hw_params,
+	.shutdown = osk_shutdown,
+};
+
+static const struct snd_soc_dapm_widget tlv320aic23_dapm_widgets[] = {
+	SND_SOC_DAPM_HP("Headphone Jack", NULL),
+	SND_SOC_DAPM_LINE("Line In", NULL),
+	SND_SOC_DAPM_MIC("Mic Jack", NULL),
+};
+
+static const struct snd_soc_dapm_route audio_map[] = {
+	{"Headphone Jack", NULL, "LHPOUT"},
+	{"Headphone Jack", NULL, "RHPOUT"},
+
+	{"LLINEIN", NULL, "Line In"},
+	{"RLINEIN", NULL, "Line In"},
+
+	{"MICIN", NULL, "Mic Jack"},
+};
+
+static int osk_tlv320aic23_init(struct snd_soc_codec *codec)
+{
+
+	/* Add osk5912 specific widgets */
+	snd_soc_dapm_new_controls(codec, tlv320aic23_dapm_widgets,
+				  ARRAY_SIZE(tlv320aic23_dapm_widgets));
+
+	/* Set up osk5912 specific audio path audio_map */
+	snd_soc_dapm_add_routes(codec, audio_map, ARRAY_SIZE(audio_map));
+
+	snd_soc_dapm_enable_pin(codec, "Headphone Jack");
+	snd_soc_dapm_enable_pin(codec, "Line In");
+	snd_soc_dapm_enable_pin(codec, "Mic Jack");
+
+	snd_soc_dapm_sync(codec);
+
+	return 0;
+}
+
+/* Digital audio interface glue - connects codec <--> CPU */
+static struct snd_soc_dai_link osk_dai = {
+	.name = "TLV320AIC23",
+	.stream_name = "AIC23",
+	.cpu_dai = &omap_mcbsp_dai[0],
+	.codec_dai = &tlv320aic23_dai,
+	.init = osk_tlv320aic23_init,
+	.ops = &osk_ops,
+};
+
+/* Audio machine driver */
+static struct snd_soc_machine snd_soc_machine_osk = {
+	.name = "OSK5912",
+	.dai_link = &osk_dai,
+	.num_links = 1,
+};
+
+/* Audio subsystem */
+static struct snd_soc_device osk_snd_devdata = {
+	.machine = &snd_soc_machine_osk,
+	.platform = &omap_soc_platform,
+	.codec_dev = &soc_codec_dev_tlv320aic23,
+};
+
+static struct platform_device *osk_snd_device;
+
+static int __init osk_soc_init(void)
+{
+	int err;
+	u32 curRate;
+	struct device *dev;
+
+	if (!(machine_is_omap_osk()))
+		return -ENODEV;
+
+	osk_snd_device = platform_device_alloc("soc-audio", -1);
+	if (!osk_snd_device)
+		return -ENOMEM;
+
+	platform_set_drvdata(osk_snd_device, &osk_snd_devdata);
+	osk_snd_devdata.dev = &osk_snd_device->dev;
+	*(unsigned int *)osk_dai.cpu_dai->private_data = 0;	/* McBSP1 */
+	err = platform_device_add(osk_snd_device);
+	if (err)
+		goto err1;
+
+	dev = &osk_snd_device->dev;
+
+	tlv320aic23_mclk = clk_get(dev, "mclk");
+	if (IS_ERR(tlv320aic23_mclk)) {
+		printk(KERN_ERR "Could not get mclk clock\n");
+		return -ENODEV;
+	}
+
+	if (clk_get_usecount(tlv320aic23_mclk) > 0) {
+		/* MCLK is already in use */
+		printk(KERN_WARNING
+		       "MCLK in use at %d Hz. We change it to %d Hz\n",
+		       (uint) clk_get_rate(tlv320aic23_mclk), CODEC_CLOCK);
+	}
+
+	/*
+	 * Configure 12 MHz output on MCLK.
+	 */
+	curRate = (uint) clk_get_rate(tlv320aic23_mclk);
+	if (curRate != CODEC_CLOCK) {
+		if (clk_set_rate(tlv320aic23_mclk, CODEC_CLOCK)) {
+			printk(KERN_ERR "Cannot set MCLK for AIC23 CODEC\n");
+			err = -ECANCELED;
+			goto err1;
+		}
+	}
+
+	printk(KERN_INFO "MCLK = %d [%d], usecount = %d\n",
+	       (uint) clk_get_rate(tlv320aic23_mclk), CODEC_CLOCK,
+	       clk_get_usecount(tlv320aic23_mclk));
+
+	return 0;
+err1:
+	clk_put(tlv320aic23_mclk);
+	platform_device_del(osk_snd_device);
+	platform_device_put(osk_snd_device);
+
+	return err;
+
+}
+
+static void __exit osk_soc_exit(void)
+{
+	platform_device_unregister(osk_snd_device);
+}
+
+module_init(osk_soc_init);
+module_exit(osk_soc_exit);
+
+MODULE_AUTHOR("Arun KS <arunks@mistralsolutions.com>");
+MODULE_DESCRIPTION("ALSA SoC OSK 5912");
+MODULE_LICENSE("GPL");
diff --git a/sound/soc/pxa/corgi.c b/sound/soc/pxa/corgi.c
index 1a8373de7f3a..2718eaf7895f 100644
--- a/sound/soc/pxa/corgi.c
+++ b/sound/soc/pxa/corgi.c
@@ -4,7 +4,7 @@
  * Copyright 2005 Wolfson Microelectronics PLC.
  * Copyright 2005 Openedhand Ltd.
  *
- * Authors: Liam Girdwood <liam.girdwood@wolfsonmicro.com>
+ * Authors: Liam Girdwood <lrg@slimlogic.co.uk>
  *          Richard Purdie <richard@openedhand.com>
  *
  *  This program is free software; you can redistribute  it and/or modify it
@@ -281,8 +281,8 @@ static int corgi_wm8731_init(struct snd_soc_codec *codec)
 {
 	int i, err;
 
-	snd_soc_dapm_disable_pin(codec, "LLINEIN");
-	snd_soc_dapm_disable_pin(codec, "RLINEIN");
+	snd_soc_dapm_nc_pin(codec, "LLINEIN");
+	snd_soc_dapm_nc_pin(codec, "RLINEIN");
 
 	/* Add corgi specific controls */
 	for (i = 0; i < ARRAY_SIZE(wm8731_corgi_controls); i++) {
diff --git a/sound/soc/pxa/em-x270.c b/sound/soc/pxa/em-x270.c
index d9c3f7b28be2..e6ff6929ab4b 100644
--- a/sound/soc/pxa/em-x270.c
+++ b/sound/soc/pxa/em-x270.c
@@ -9,7 +9,7 @@
  * Copyright 2005 Wolfson Microelectronics PLC.
  * Copyright 2005 Openedhand Ltd.
  *
- * Authors: Liam Girdwood <liam.girdwood@wolfsonmicro.com>
+ * Authors: Liam Girdwood <lrg@slimlogic.co.uk>
  *          Richard Purdie <richard@openedhand.com>
  *
  *  This program is free software; you can redistribute  it and/or modify it
diff --git a/sound/soc/pxa/poodle.c b/sound/soc/pxa/poodle.c
index f84f7d8db09a..4d9930c52789 100644
--- a/sound/soc/pxa/poodle.c
+++ b/sound/soc/pxa/poodle.c
@@ -4,7 +4,7 @@
  * Copyright 2005 Wolfson Microelectronics PLC.
  * Copyright 2005 Openedhand Ltd.
  *
- * Authors: Liam Girdwood <liam.girdwood@wolfsonmicro.com>
+ * Authors: Liam Girdwood <lrg@slimlogic.co.uk>
  *          Richard Purdie <richard@openedhand.com>
  *
  *  This program is free software; you can redistribute  it and/or modify it
@@ -242,8 +242,8 @@ static int poodle_wm8731_init(struct snd_soc_codec *codec)
 {
 	int i, err;
 
-	snd_soc_dapm_disable_pin(codec, "LLINEIN");
-	snd_soc_dapm_disable_pin(codec, "RLINEIN");
+	snd_soc_dapm_nc_pin(codec, "LLINEIN");
+	snd_soc_dapm_nc_pin(codec, "RLINEIN");
 	snd_soc_dapm_enable_pin(codec, "MICIN");
 
 	/* Add poodle specific controls */
diff --git a/sound/soc/pxa/pxa2xx-i2s.c b/sound/soc/pxa/pxa2xx-i2s.c
index 2fb58298513b..e758034db5c3 100644
--- a/sound/soc/pxa/pxa2xx-i2s.c
+++ b/sound/soc/pxa/pxa2xx-i2s.c
@@ -3,7 +3,7 @@
  *
  * Copyright 2005 Wolfson Microelectronics PLC.
  * Author: Liam Girdwood
- *         liam.girdwood@wolfsonmicro.com or linux@wolfsonmicro.com
+ *         lrg@slimlogic.co.uk
  *
  *  This program is free software; you can redistribute  it and/or modify it
  *  under  the terms of  the GNU General  Public License as published by the
@@ -405,6 +405,6 @@ module_init(pxa2xx_i2s_init);
 module_exit(pxa2xx_i2s_exit);
 
 /* Module information */
-MODULE_AUTHOR("Liam Girdwood, liam.girdwood@wolfsonmicro.com, www.wolfsonmicro.com");
+MODULE_AUTHOR("Liam Girdwood, lrg@slimlogic.co.uk");
 MODULE_DESCRIPTION("pxa2xx I2S SoC Interface");
 MODULE_LICENSE("GPL");
diff --git a/sound/soc/pxa/spitz.c b/sound/soc/pxa/spitz.c
index 9a70b00fc30e..d307b6757e95 100644
--- a/sound/soc/pxa/spitz.c
+++ b/sound/soc/pxa/spitz.c
@@ -4,7 +4,7 @@
  * Copyright 2005 Wolfson Microelectronics PLC.
  * Copyright 2005 Openedhand Ltd.
  *
- * Authors: Liam Girdwood <liam.girdwood@wolfsonmicro.com>
+ * Authors: Liam Girdwood <lrg@slimlogic.co.uk>
  *          Richard Purdie <richard@openedhand.com>
  *
  *  This program is free software; you can redistribute  it and/or modify it
@@ -281,13 +281,13 @@ static int spitz_wm8750_init(struct snd_soc_codec *codec)
 	int i, err;
 
 	/* NC codec pins */
-	snd_soc_dapm_disable_pin(codec, "RINPUT1");
-	snd_soc_dapm_disable_pin(codec, "LINPUT2");
-	snd_soc_dapm_disable_pin(codec, "RINPUT2");
-	snd_soc_dapm_disable_pin(codec, "LINPUT3");
-	snd_soc_dapm_disable_pin(codec, "RINPUT3");
-	snd_soc_dapm_disable_pin(codec, "OUT3");
-	snd_soc_dapm_disable_pin(codec, "MONO1");
+	snd_soc_dapm_nc_pin(codec, "RINPUT1");
+	snd_soc_dapm_nc_pin(codec, "LINPUT2");
+	snd_soc_dapm_nc_pin(codec, "RINPUT2");
+	snd_soc_dapm_nc_pin(codec, "LINPUT3");
+	snd_soc_dapm_nc_pin(codec, "RINPUT3");
+	snd_soc_dapm_nc_pin(codec, "OUT3");
+	snd_soc_dapm_nc_pin(codec, "MONO1");
 
 	/* Add spitz specific controls */
 	for (i = 0; i < ARRAY_SIZE(wm8750_spitz_controls); i++) {
diff --git a/sound/soc/pxa/tosa.c b/sound/soc/pxa/tosa.c
index 2baaa750f123..afefe41b8c46 100644
--- a/sound/soc/pxa/tosa.c
+++ b/sound/soc/pxa/tosa.c
@@ -4,7 +4,7 @@
  * Copyright 2005 Wolfson Microelectronics PLC.
  * Copyright 2005 Openedhand Ltd.
  *
- * Authors: Liam Girdwood <liam.girdwood@wolfsonmicro.com>
+ * Authors: Liam Girdwood <lrg@slimlogic.co.uk>
  *          Richard Purdie <richard@openedhand.com>
  *
  *  This program is free software; you can redistribute  it and/or modify it
@@ -190,8 +190,8 @@ static int tosa_ac97_init(struct snd_soc_codec *codec)
 {
 	int i, err;
 
-	snd_soc_dapm_disable_pin(codec, "OUT3");
-	snd_soc_dapm_disable_pin(codec, "MONOOUT");
+	snd_soc_dapm_nc_pin(codec, "OUT3");
+	snd_soc_dapm_nc_pin(codec, "MONOOUT");
 
 	/* add tosa specific controls */
 	for (i = 0; i < ARRAY_SIZE(tosa_controls); i++) {
diff --git a/sound/soc/s3c24xx/neo1973_wm8753.c b/sound/soc/s3c24xx/neo1973_wm8753.c
index 73a50e93a9a2..87ddfefcc2fb 100644
--- a/sound/soc/s3c24xx/neo1973_wm8753.c
+++ b/sound/soc/s3c24xx/neo1973_wm8753.c
@@ -511,21 +511,20 @@ static int neo1973_wm8753_init(struct snd_soc_codec *codec)
 	DBG("Entered %s\n", __func__);
 
 	/* set up NC codec pins */
-	snd_soc_dapm_disable_pin(codec, "LOUT2");
-	snd_soc_dapm_disable_pin(codec, "ROUT2");
-	snd_soc_dapm_disable_pin(codec, "OUT3");
-	snd_soc_dapm_disable_pin(codec, "OUT4");
-	snd_soc_dapm_disable_pin(codec, "LINE1");
-	snd_soc_dapm_disable_pin(codec, "LINE2");
-
-
-	/* set endpoints to default mode */
-	set_scenario_endpoints(codec, NEO_AUDIO_OFF);
+	snd_soc_dapm_nc_pin(codec, "LOUT2");
+	snd_soc_dapm_nc_pin(codec, "ROUT2");
+	snd_soc_dapm_nc_pin(codec, "OUT3");
+	snd_soc_dapm_nc_pin(codec, "OUT4");
+	snd_soc_dapm_nc_pin(codec, "LINE1");
+	snd_soc_dapm_nc_pin(codec, "LINE2");
 
 	/* Add neo1973 specific widgets */
 	snd_soc_dapm_new_controls(codec, wm8753_dapm_widgets,
 				  ARRAY_SIZE(wm8753_dapm_widgets));
 
+	/* set endpoints to default mode */
+	set_scenario_endpoints(codec, NEO_AUDIO_OFF);
+
 	/* add neo1973 specific controls */
 	for (i = 0; i < ARRAY_SIZE(wm8753_neo1973_controls); i++) {
 		err = snd_ctl_add(codec->card,
@@ -603,6 +602,8 @@ static int lm4857_i2c_probe(struct i2c_client *client,
 {
 	DBG("Entered %s\n", __func__);
 
+	i2c = client;
+
 	lm4857_write_regs();
 	return 0;
 }
@@ -611,6 +612,8 @@ static int lm4857_i2c_remove(struct i2c_client *client)
 {
 	DBG("Entered %s\n", __func__);
 
+	i2c = NULL;
+
 	return 0;
 }
 
@@ -650,7 +653,7 @@ static void lm4857_shutdown(struct i2c_client *dev)
 }
 
 static const struct i2c_device_id lm4857_i2c_id[] = {
-	{ "neo1973_lm4857", 0 }
+	{ "neo1973_lm4857", 0 },
 	{ }
 };
 
@@ -668,48 +671,6 @@ static struct i2c_driver lm4857_i2c_driver = {
 };
 
 static struct platform_device *neo1973_snd_device;
-static struct i2c_client *lm4857_client;
-
-static int __init neo1973_add_lm4857_device(struct platform_device *pdev,
-					    int i2c_bus,
-					    unsigned short i2c_address)
-{
-	struct i2c_board_info info;
-	struct i2c_adapter *adapter;
-	struct i2c_client *client;
-	int ret;
-
-	ret = i2c_add_driver(&lm4857_i2c_driver);
-	if (ret != 0) {
-		dev_err(&pdev->dev, "can't add lm4857 driver\n");
-		return ret;
-	}
-
-	memset(&info, 0, sizeof(struct i2c_board_info));
-	info.addr = i2c_address;
-	strlcpy(info.type, "neo1973_lm4857", I2C_NAME_SIZE);
-
-	adapter = i2c_get_adapter(i2c_bus);
-	if (!adapter) {
-		dev_err(&pdev->dev, "can't get i2c adapter %d\n", i2c_bus);
-		goto err_driver;
-	}
-
-	client = i2c_new_device(adapter, &info);
-	i2c_put_adapter(adapter);
-	if (!client) {
-		dev_err(&pdev->dev, "can't add lm4857 device at 0x%x\n",
-			(unsigned int)info.addr);
-		goto err_driver;
-	}
-
-	lm4857_client = client;
-	return 0;
-
-err_driver:
-	i2c_del_driver(&lm4857_i2c_driver);
-	return -ENODEV;
-}
 
 static int __init neo1973_init(void)
 {
@@ -736,8 +697,8 @@ static int __init neo1973_init(void)
 		return ret;
 	}
 
-	ret = neo1973_add_lm4857_device(neo1973_snd_device,
-					neo1973_wm8753_setup, 0x7C);
+	ret = i2c_add_driver(&lm4857_i2c_driver);
+
 	if (ret != 0)
 		platform_device_unregister(neo1973_snd_device);
 
@@ -748,7 +709,6 @@ static void __exit neo1973_exit(void)
 {
 	DBG("Entered %s\n", __func__);
 
-	i2c_unregister_device(lm4857_client);
 	i2c_del_driver(&lm4857_i2c_driver);
 	platform_device_unregister(neo1973_snd_device);
 }
diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c
index ad381138fc2e..462e635dfc74 100644
--- a/sound/soc/soc-core.c
+++ b/sound/soc/soc-core.c
@@ -4,8 +4,7 @@
  * Copyright 2005 Wolfson Microelectronics PLC.
  * Copyright 2005 Openedhand Ltd.
  *
- * Author: Liam Girdwood
- *         liam.girdwood@wolfsonmicro.com or linux@wolfsonmicro.com
+ * Author: Liam Girdwood <lrg@slimlogic.co.uk>
  *         with code, comments and ideas from :-
  *         Richard Purdie <richard@openedhand.com>
  *
@@ -1886,7 +1885,7 @@ module_init(snd_soc_init);
 module_exit(snd_soc_exit);
 
 /* Module information */
-MODULE_AUTHOR("Liam Girdwood, liam.girdwood@wolfsonmicro.com, www.wolfsonmicro.com");
+MODULE_AUTHOR("Liam Girdwood, lrg@slimlogic.co.uk");
 MODULE_DESCRIPTION("ALSA SoC Core");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS("platform:soc-audio");
diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c
index 9ca9c08610fa..efbd0b37810a 100644
--- a/sound/soc/soc-dapm.c
+++ b/sound/soc/soc-dapm.c
@@ -2,8 +2,7 @@
  * soc-dapm.c  --  ALSA SoC Dynamic Audio Power Management
  *
  * Copyright 2005 Wolfson Microelectronics PLC.
- * Author: Liam Girdwood
- *         liam.girdwood@wolfsonmicro.com or linux@wolfsonmicro.com
+ * Author: Liam Girdwood <lrg@slimlogic.co.uk>
  *
  *  This program is free software; you can redistribute  it and/or modify it
  *  under  the terms of  the GNU General  Public License as published by the
@@ -1484,6 +1483,26 @@ int snd_soc_dapm_disable_pin(struct snd_soc_codec *codec, char *pin)
 EXPORT_SYMBOL_GPL(snd_soc_dapm_disable_pin);
 
 /**
+ * snd_soc_dapm_nc_pin - permanently disable pin.
+ * @codec: SoC codec
+ * @pin: pin name
+ *
+ * Marks the specified pin as being not connected, disabling it along
+ * any parent or child widgets.  At present this is identical to
+ * snd_soc_dapm_disable_pin() but in future it will be extended to do
+ * additional things such as disabling controls which only affect
+ * paths through the pin.
+ *
+ * NOTE: snd_soc_dapm_sync() needs to be called after this for DAPM to
+ * do any widget power switching.
+ */
+int snd_soc_dapm_nc_pin(struct snd_soc_codec *codec, char *pin)
+{
+	return snd_soc_dapm_set_pin(codec, pin, 0);
+}
+EXPORT_SYMBOL_GPL(snd_soc_dapm_nc_pin);
+
+/**
  * snd_soc_dapm_get_pin_status - get audio pin status
  * @codec: audio codec
  * @pin: audio signal pin endpoint (or start point)
@@ -1521,6 +1540,6 @@ void snd_soc_dapm_free(struct snd_soc_device *socdev)
 EXPORT_SYMBOL_GPL(snd_soc_dapm_free);
 
 /* Module information */
-MODULE_AUTHOR("Liam Girdwood, liam.girdwood@wolfsonmicro.com, www.wolfsonmicro.com");
+MODULE_AUTHOR("Liam Girdwood, lrg@slimlogic.co.uk");
 MODULE_DESCRIPTION("Dynamic Audio Power Management core for ALSA SoC");
 MODULE_LICENSE("GPL");
diff --git a/sound/sparc/amd7930.c b/sound/sparc/amd7930.c
index 49acee0c4840..f87933e48812 100644
--- a/sound/sparc/amd7930.c
+++ b/sound/sparc/amd7930.c
@@ -1,6 +1,6 @@
 /*
  * Driver for AMD7930 sound chips found on Sparcs.
- * Copyright (C) 2002 David S. Miller <davem@redhat.com>
+ * Copyright (C) 2002, 2008 David S. Miller <davem@davemloft.net>
  *
  * Based entirely upon drivers/sbus/audio/amd7930.c which is:
  * Copyright (C) 1996,1997 Thomas K. Dyas (tdyas@eden.rutgers.edu)
@@ -35,6 +35,8 @@
 #include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/moduleparam.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <sound/core.h>
 #include <sound/pcm.h>
@@ -44,7 +46,6 @@
 
 #include <asm/io.h>
 #include <asm/irq.h>
-#include <asm/sbus.h>
 #include <asm/prom.h>
 
 static int index[SNDRV_CARDS] = SNDRV_DEFAULT_IDX;	/* Index 0-MAX */
@@ -335,8 +336,8 @@ struct snd_amd7930 {
 	int			pgain;
 	int			mgain;
 
+	struct of_device	*op;
 	unsigned int		irq;
-	unsigned int		regs_size;
 	struct snd_amd7930	*next;
 };
 
@@ -905,13 +906,16 @@ static int __devinit snd_amd7930_mixer(struct snd_amd7930 *amd)
 
 static int snd_amd7930_free(struct snd_amd7930 *amd)
 {
+	struct of_device *op = amd->op;
+
 	amd7930_idle(amd);
 
 	if (amd->irq)
 		free_irq(amd->irq, amd);
 
 	if (amd->regs)
-		sbus_iounmap(amd->regs, amd->regs_size);
+		of_iounmap(&op->resource[0], amd->regs,
+			   resource_size(&op->resource[0]));
 
 	kfree(amd);
 
@@ -930,13 +934,12 @@ static struct snd_device_ops snd_amd7930_dev_ops = {
 };
 
 static int __devinit snd_amd7930_create(struct snd_card *card,
-					struct resource *rp,
-					unsigned int reg_size,
+					struct of_device *op,
 					int irq, int dev,
 					struct snd_amd7930 **ramd)
 {
-	unsigned long flags;
 	struct snd_amd7930 *amd;
+	unsigned long flags;
 	int err;
 
 	*ramd = NULL;
@@ -946,9 +949,10 @@ static int __devinit snd_amd7930_create(struct snd_card *card,
 
 	spin_lock_init(&amd->lock);
 	amd->card = card;
-	amd->regs_size = reg_size;
+	amd->op = op;
 
-	amd->regs = sbus_ioremap(rp, 0, amd->regs_size, "amd7930");
+	amd->regs = of_ioremap(&op->resource[0], 0,
+			       resource_size(&op->resource[0]), "amd7930");
 	if (!amd->regs) {
 		snd_printk("amd7930-%d: Unable to map chip registers.\n", dev);
 		return -EIO;
@@ -997,12 +1001,15 @@ static int __devinit snd_amd7930_create(struct snd_card *card,
 	return 0;
 }
 
-static int __devinit amd7930_attach_common(struct resource *rp, int irq)
+static int __devinit amd7930_sbus_probe(struct of_device *op, const struct of_device_id *match)
 {
+	struct resource *rp = &op->resource[0];
 	static int dev_num;
 	struct snd_card *card;
 	struct snd_amd7930 *amd;
-	int err;
+	int err, irq;
+
+	irq = op->irqs[0];
 
 	if (dev_num >= SNDRV_CARDS)
 		return -ENODEV;
@@ -1023,8 +1030,7 @@ static int __devinit amd7930_attach_common(struct resource *rp, int irq)
 		(unsigned long long)rp->start,
 		irq);
 
-	if ((err = snd_amd7930_create(card, rp,
-				      (rp->end - rp->start) + 1,
+	if ((err = snd_amd7930_create(card, op,
 				      irq, dev_num, &amd)) < 0)
 		goto out_err;
 
@@ -1049,43 +1055,7 @@ out_err:
 	return err;
 }
 
-static int __devinit amd7930_obio_attach(struct device_node *dp)
-{
-	const struct linux_prom_registers *regs;
-	const struct linux_prom_irqs *irqp;
-	struct resource res, *rp;
-	int len;
-
-	irqp = of_get_property(dp, "intr", &len);
-	if (!irqp) {
-		snd_printk("%s: Firmware node lacks IRQ property.\n",
-			   dp->full_name);
-		return -ENODEV;
-	}
-
-	regs = of_get_property(dp, "reg", &len);
-	if (!regs) {
-		snd_printk("%s: Firmware node lacks register property.\n",
-			   dp->full_name);
-		return -ENODEV;
-	}
-
-	rp = &res;
-	rp->start = regs->phys_addr;
-	rp->end = rp->start + regs->reg_size - 1;
-	rp->flags = IORESOURCE_IO | (regs->which_io & 0xff);
-
-	return amd7930_attach_common(rp, irqp->pri);
-}
-
-static int __devinit amd7930_sbus_probe(struct of_device *dev, const struct of_device_id *match)
-{
-	struct sbus_dev *sdev = to_sbus_device(&dev->dev);
-
-	return amd7930_attach_common(&sdev->resource[0], sdev->irqs[0]);
-}
-
-static struct of_device_id amd7930_match[] = {
+static const struct of_device_id amd7930_match[] = {
 	{
 		.name = "audio",
 	},
@@ -1100,20 +1070,7 @@ static struct of_platform_driver amd7930_sbus_driver = {
 
 static int __init amd7930_init(void)
 {
-	struct device_node *dp;
-
-	/* Try to find the sun4c "audio" node first. */
-	dp = of_find_node_by_path("/");
-	dp = dp->child;
-	while (dp) {
-		if (!strcmp(dp->name, "audio"))
-			amd7930_obio_attach(dp);
-
-		dp = dp->sibling;
-	}
-
-	/* Probe each SBUS for amd7930 chips. */
-	return of_register_driver(&amd7930_sbus_driver, &sbus_bus_type);
+	return of_register_driver(&amd7930_sbus_driver, &of_bus_type);
 }
 
 static void __exit amd7930_exit(void)
diff --git a/sound/sparc/cs4231.c b/sound/sparc/cs4231.c
index 791d2fb821d1..d44bf98e965e 100644
--- a/sound/sparc/cs4231.c
+++ b/sound/sparc/cs4231.c
@@ -1,6 +1,6 @@
 /*
  * Driver for CS4231 sound chips found on Sparcs.
- * Copyright (C) 2002 David S. Miller <davem@redhat.com>
+ * Copyright (C) 2002, 2008 David S. Miller <davem@davemloft.net>
  *
  * Based entirely upon drivers/sbus/audio/cs4231.c which is:
  * Copyright (C) 1996, 1997, 1998 Derrick J Brashear (shadow@andrew.cmu.edu)
@@ -17,7 +17,8 @@
 #include <linux/moduleparam.h>
 #include <linux/irq.h>
 #include <linux/io.h>
-
+#include <linux/of.h>
+#include <linux/of_device.h>
 
 #include <sound/core.h>
 #include <sound/pcm.h>
@@ -29,13 +30,12 @@
 
 #ifdef CONFIG_SBUS
 #define SBUS_SUPPORT
-#include <asm/sbus.h>
 #endif
 
 #if defined(CONFIG_PCI) && defined(CONFIG_SPARC64)
 #define EBUS_SUPPORT
 #include <linux/pci.h>
-#include <asm/ebus.h>
+#include <asm/ebus_dma.h>
 #endif
 
 static int index[SNDRV_CARDS] = SNDRV_DEFAULT_IDX;	/* Index 0-MAX */
@@ -70,8 +70,6 @@ struct cs4231_dma_control {
 	int		(*request)(struct cs4231_dma_control *dma_cont,
 				   dma_addr_t bus_addr, size_t len);
 	unsigned int	(*address)(struct cs4231_dma_control *dma_cont);
-	void		(*preallocate)(struct snd_cs4231 *chip,
-				       struct snd_pcm *pcm);
 #ifdef EBUS_SUPPORT
 	struct		ebus_dma_info	ebus_info;
 #endif
@@ -114,21 +112,12 @@ struct snd_cs4231 {
 	struct mutex		mce_mutex;	/* mutex for mce register */
 	struct mutex		open_mutex;	/* mutex for ALSA open/close */
 
-	union {
-#ifdef SBUS_SUPPORT
-		struct sbus_dev		*sdev;
-#endif
-#ifdef EBUS_SUPPORT
-		struct pci_dev		*pdev;
-#endif
-	} dev_u;
+	struct of_device	*op;
 	unsigned int		irq[2];
 	unsigned int		regs_size;
 	struct snd_cs4231	*next;
 };
 
-static struct snd_cs4231 *cs4231_list;
-
 /* Eventually we can use sound/isa/cs423x/cs4231_lib.c directly, but for
  * now....  -DaveM
  */
@@ -267,27 +256,19 @@ static unsigned char snd_cs4231_original_image[32] =
 
 static u8 __cs4231_readb(struct snd_cs4231 *cp, void __iomem *reg_addr)
 {
-#ifdef EBUS_SUPPORT
 	if (cp->flags & CS4231_FLAG_EBUS)
 		return readb(reg_addr);
 	else
-#endif
-#ifdef SBUS_SUPPORT
 		return sbus_readb(reg_addr);
-#endif
 }
 
 static void __cs4231_writeb(struct snd_cs4231 *cp, u8 val,
 			    void __iomem *reg_addr)
 {
-#ifdef EBUS_SUPPORT
 	if (cp->flags & CS4231_FLAG_EBUS)
 		return writeb(val, reg_addr);
 	else
-#endif
-#ifdef SBUS_SUPPORT
 		return sbus_writeb(val, reg_addr);
-#endif
 }
 
 /*
@@ -1258,7 +1239,9 @@ static int __init snd_cs4231_pcm(struct snd_card *card)
 	pcm->info_flags = SNDRV_PCM_INFO_JOINT_DUPLEX;
 	strcpy(pcm->name, "CS4231");
 
-	chip->p_dma.preallocate(chip, pcm);
+	snd_pcm_lib_preallocate_pages_for_all(pcm, SNDRV_DMA_TYPE_DEV,
+					      &chip->op->dev,
+					      64 * 1024, 128 * 1024);
 
 	chip->pcm = pcm;
 
@@ -1627,8 +1610,7 @@ static int __init cs4231_attach_finish(struct snd_card *card)
 	if (err < 0)
 		goto out_err;
 
-	chip->next = cs4231_list;
-	cs4231_list = chip;
+	dev_set_drvdata(&chip->op->dev, chip);
 
 	dev++;
 	return 0;
@@ -1783,24 +1765,19 @@ static unsigned int sbus_dma_addr(struct cs4231_dma_control *dma_cont)
 	return sbus_readl(base->regs + base->dir + APCVA);
 }
 
-static void sbus_dma_preallocate(struct snd_cs4231 *chip, struct snd_pcm *pcm)
-{
-	snd_pcm_lib_preallocate_pages_for_all(pcm, SNDRV_DMA_TYPE_SBUS,
-					snd_dma_sbus_data(chip->dev_u.sdev),
-					64 * 1024, 128 * 1024);
-}
-
 /*
  * Init and exit routines
  */
 
 static int snd_cs4231_sbus_free(struct snd_cs4231 *chip)
 {
+	struct of_device *op = chip->op;
+
 	if (chip->irq[0])
 		free_irq(chip->irq[0], chip);
 
 	if (chip->port)
-		sbus_iounmap(chip->port, chip->regs_size);
+		of_iounmap(&op->resource[0], chip->port, chip->regs_size);
 
 	return 0;
 }
@@ -1817,7 +1794,7 @@ static struct snd_device_ops snd_cs4231_sbus_dev_ops = {
 };
 
 static int __init snd_cs4231_sbus_create(struct snd_card *card,
-					 struct sbus_dev *sdev,
+					 struct of_device *op,
 					 int dev)
 {
 	struct snd_cs4231 *chip = card->private_data;
@@ -1828,13 +1805,13 @@ static int __init snd_cs4231_sbus_create(struct snd_card *card,
 	spin_lock_init(&chip->p_dma.sbus_info.lock);
 	mutex_init(&chip->mce_mutex);
 	mutex_init(&chip->open_mutex);
-	chip->dev_u.sdev = sdev;
-	chip->regs_size = sdev->reg_addrs[0].reg_size;
+	chip->op = op;
+	chip->regs_size = resource_size(&op->resource[0]);
 	memcpy(&chip->image, &snd_cs4231_original_image,
 	       sizeof(snd_cs4231_original_image));
 
-	chip->port = sbus_ioremap(&sdev->resource[0], 0,
-				  chip->regs_size, "cs4231");
+	chip->port = of_ioremap(&op->resource[0], 0,
+				chip->regs_size, "cs4231");
 	if (!chip->port) {
 		snd_printdd("cs4231-%d: Unable to map chip registers.\n", dev);
 		return -EIO;
@@ -1849,22 +1826,20 @@ static int __init snd_cs4231_sbus_create(struct snd_card *card,
 	chip->p_dma.enable = sbus_dma_enable;
 	chip->p_dma.request = sbus_dma_request;
 	chip->p_dma.address = sbus_dma_addr;
-	chip->p_dma.preallocate = sbus_dma_preallocate;
 
 	chip->c_dma.prepare = sbus_dma_prepare;
 	chip->c_dma.enable = sbus_dma_enable;
 	chip->c_dma.request = sbus_dma_request;
 	chip->c_dma.address = sbus_dma_addr;
-	chip->c_dma.preallocate = sbus_dma_preallocate;
 
-	if (request_irq(sdev->irqs[0], snd_cs4231_sbus_interrupt,
+	if (request_irq(op->irqs[0], snd_cs4231_sbus_interrupt,
 			IRQF_SHARED, "cs4231", chip)) {
 		snd_printdd("cs4231-%d: Unable to grab SBUS IRQ %d\n",
-			    dev, sdev->irqs[0]);
+			    dev, op->irqs[0]);
 		snd_cs4231_sbus_free(chip);
 		return -EBUSY;
 	}
-	chip->irq[0] = sdev->irqs[0];
+	chip->irq[0] = op->irqs[0];
 
 	if (snd_cs4231_probe(chip) < 0) {
 		snd_cs4231_sbus_free(chip);
@@ -1881,9 +1856,9 @@ static int __init snd_cs4231_sbus_create(struct snd_card *card,
 	return 0;
 }
 
-static int __init cs4231_sbus_attach(struct sbus_dev *sdev)
+static int __devinit cs4231_sbus_probe(struct of_device *op, const struct of_device_id *match)
 {
-	struct resource *rp = &sdev->resource[0];
+	struct resource *rp = &op->resource[0];
 	struct snd_card *card;
 	int err;
 
@@ -1895,9 +1870,9 @@ static int __init cs4231_sbus_attach(struct sbus_dev *sdev)
 		card->shortname,
 		rp->flags & 0xffL,
 		(unsigned long long)rp->start,
-		sdev->irqs[0]);
+		op->irqs[0]);
 
-	err = snd_cs4231_sbus_create(card, sdev, dev);
+	err = snd_cs4231_sbus_create(card, op, dev);
 	if (err < 0) {
 		snd_card_free(card);
 		return err;
@@ -1950,30 +1925,25 @@ static unsigned int _ebus_dma_addr(struct cs4231_dma_control *dma_cont)
 	return ebus_dma_addr(&dma_cont->ebus_info);
 }
 
-static void _ebus_dma_preallocate(struct snd_cs4231 *chip, struct snd_pcm *pcm)
-{
-	snd_pcm_lib_preallocate_pages_for_all(pcm, SNDRV_DMA_TYPE_DEV,
-				      snd_dma_pci_data(chip->dev_u.pdev),
-				      64*1024, 128*1024);
-}
-
 /*
  * Init and exit routines
  */
 
 static int snd_cs4231_ebus_free(struct snd_cs4231 *chip)
 {
+	struct of_device *op = chip->op;
+
 	if (chip->c_dma.ebus_info.regs) {
 		ebus_dma_unregister(&chip->c_dma.ebus_info);
-		iounmap(chip->c_dma.ebus_info.regs);
+		of_iounmap(&op->resource[2], chip->c_dma.ebus_info.regs, 0x10);
 	}
 	if (chip->p_dma.ebus_info.regs) {
 		ebus_dma_unregister(&chip->p_dma.ebus_info);
-		iounmap(chip->p_dma.ebus_info.regs);
+		of_iounmap(&op->resource[1], chip->p_dma.ebus_info.regs, 0x10);
 	}
 
 	if (chip->port)
-		iounmap(chip->port);
+		of_iounmap(&op->resource[0], chip->port, 0x10);
 
 	return 0;
 }
@@ -1990,7 +1960,7 @@ static struct snd_device_ops snd_cs4231_ebus_dev_ops = {
 };
 
 static int __init snd_cs4231_ebus_create(struct snd_card *card,
-					 struct linux_ebus_device *edev,
+					 struct of_device *op,
 					 int dev)
 {
 	struct snd_cs4231 *chip = card->private_data;
@@ -2002,35 +1972,35 @@ static int __init snd_cs4231_ebus_create(struct snd_card *card,
 	mutex_init(&chip->mce_mutex);
 	mutex_init(&chip->open_mutex);
 	chip->flags |= CS4231_FLAG_EBUS;
-	chip->dev_u.pdev = edev->bus->self;
+	chip->op = op;
 	memcpy(&chip->image, &snd_cs4231_original_image,
 	       sizeof(snd_cs4231_original_image));
 	strcpy(chip->c_dma.ebus_info.name, "cs4231(capture)");
 	chip->c_dma.ebus_info.flags = EBUS_DMA_FLAG_USE_EBDMA_HANDLER;
 	chip->c_dma.ebus_info.callback = snd_cs4231_ebus_capture_callback;
 	chip->c_dma.ebus_info.client_cookie = chip;
-	chip->c_dma.ebus_info.irq = edev->irqs[0];
+	chip->c_dma.ebus_info.irq = op->irqs[0];
 	strcpy(chip->p_dma.ebus_info.name, "cs4231(play)");
 	chip->p_dma.ebus_info.flags = EBUS_DMA_FLAG_USE_EBDMA_HANDLER;
 	chip->p_dma.ebus_info.callback = snd_cs4231_ebus_play_callback;
 	chip->p_dma.ebus_info.client_cookie = chip;
-	chip->p_dma.ebus_info.irq = edev->irqs[1];
+	chip->p_dma.ebus_info.irq = op->irqs[1];
 
 	chip->p_dma.prepare = _ebus_dma_prepare;
 	chip->p_dma.enable = _ebus_dma_enable;
 	chip->p_dma.request = _ebus_dma_request;
 	chip->p_dma.address = _ebus_dma_addr;
-	chip->p_dma.preallocate = _ebus_dma_preallocate;
 
 	chip->c_dma.prepare = _ebus_dma_prepare;
 	chip->c_dma.enable = _ebus_dma_enable;
 	chip->c_dma.request = _ebus_dma_request;
 	chip->c_dma.address = _ebus_dma_addr;
-	chip->c_dma.preallocate = _ebus_dma_preallocate;
 
-	chip->port = ioremap(edev->resource[0].start, 0x10);
-	chip->p_dma.ebus_info.regs = ioremap(edev->resource[1].start, 0x10);
-	chip->c_dma.ebus_info.regs = ioremap(edev->resource[2].start, 0x10);
+	chip->port = of_ioremap(&op->resource[0], 0, 0x10, "cs4231");
+	chip->p_dma.ebus_info.regs =
+		of_ioremap(&op->resource[1], 0, 0x10, "cs4231_pdma");
+	chip->c_dma.ebus_info.regs =
+		of_ioremap(&op->resource[2], 0, 0x10, "cs4231_cdma");
 	if (!chip->port || !chip->p_dma.ebus_info.regs ||
 	    !chip->c_dma.ebus_info.regs) {
 		snd_cs4231_ebus_free(chip);
@@ -2078,7 +2048,7 @@ static int __init snd_cs4231_ebus_create(struct snd_card *card,
 	return 0;
 }
 
-static int __init cs4231_ebus_attach(struct linux_ebus_device *edev)
+static int __devinit cs4231_ebus_probe(struct of_device *op, const struct of_device_id *match)
 {
 	struct snd_card *card;
 	int err;
@@ -2089,10 +2059,10 @@ static int __init cs4231_ebus_attach(struct linux_ebus_device *edev)
 
 	sprintf(card->longname, "%s at 0x%lx, irq %d",
 		card->shortname,
-		edev->resource[0].start,
-		edev->irqs[0]);
+		op->resource[0].start,
+		op->irqs[0]);
 
-	err = snd_cs4231_ebus_create(card, edev, dev);
+	err = snd_cs4231_ebus_create(card, op, dev);
 	if (err < 0) {
 		snd_card_free(card);
 		return err;
@@ -2102,68 +2072,57 @@ static int __init cs4231_ebus_attach(struct linux_ebus_device *edev)
 }
 #endif
 
-static int __init cs4231_init(void)
+static int __devinit cs4231_probe(struct of_device *op, const struct of_device_id *match)
 {
-#ifdef SBUS_SUPPORT
-	struct sbus_bus *sbus;
-	struct sbus_dev *sdev;
-#endif
 #ifdef EBUS_SUPPORT
-	struct linux_ebus *ebus;
-	struct linux_ebus_device *edev;
+	if (!strcmp(op->node->parent->name, "ebus"))
+		return cs4231_ebus_probe(op, match);
 #endif
-	int found;
-
-	found = 0;
-
 #ifdef SBUS_SUPPORT
-	for_all_sbusdev(sdev, sbus) {
-		if (!strcmp(sdev->prom_name, "SUNW,CS4231")) {
-			if (cs4231_sbus_attach(sdev) == 0)
-				found++;
-		}
-	}
+	if (!strcmp(op->node->parent->name, "sbus") ||
+	    !strcmp(op->node->parent->name, "sbi"))
+		return cs4231_sbus_probe(op, match);
 #endif
-#ifdef EBUS_SUPPORT
-	for_each_ebus(ebus) {
-		for_each_ebusdev(edev, ebus) {
-			int match = 0;
-
-			if (!strcmp(edev->prom_node->name, "SUNW,CS4231")) {
-				match = 1;
-			} else if (!strcmp(edev->prom_node->name, "audio")) {
-				const char *compat;
-
-				compat = of_get_property(edev->prom_node,
-							 "compatible", NULL);
-				if (compat && !strcmp(compat, "SUNW,CS4231"))
-					match = 1;
-			}
+	return -ENODEV;
+}
 
-			if (match &&
-			    cs4231_ebus_attach(edev) == 0)
-				found++;
-		}
-	}
-#endif
+static int __devexit cs4231_remove(struct of_device *op)
+{
+	struct snd_cs4231 *chip = dev_get_drvdata(&op->dev);
 
+	snd_card_free(chip->card);
 
-	return (found > 0) ? 0 : -EIO;
+	return 0;
 }
 
-static void __exit cs4231_exit(void)
-{
-	struct snd_cs4231 *p = cs4231_list;
+static const struct of_device_id cs4231_match[] = {
+	{
+		.name = "SUNW,CS4231",
+	},
+	{
+		.name = "audio",
+		.compatible = "SUNW,CS4231",
+	},
+	{},
+};
 
-	while (p != NULL) {
-		struct snd_cs4231 *next = p->next;
+MODULE_DEVICE_TABLE(of, cs4231_match);
 
-		snd_card_free(p->card);
+static struct of_platform_driver cs4231_driver = {
+	.name		= "audio",
+	.match_table	= cs4231_match,
+	.probe		= cs4231_probe,
+	.remove		= __devexit_p(cs4231_remove),
+};
 
-		p = next;
-	}
+static int __init cs4231_init(void)
+{
+	return of_register_driver(&cs4231_driver, &of_bus_type);
+}
 
-	cs4231_list = NULL;
+static void __exit cs4231_exit(void)
+{
+	of_unregister_driver(&cs4231_driver);
 }
 
 module_init(cs4231_init);
diff --git a/sound/sparc/dbri.c b/sound/sparc/dbri.c
index c534a2a849fa..c257ad8bdfbc 100644
--- a/sound/sparc/dbri.c
+++ b/sound/sparc/dbri.c
@@ -57,6 +57,7 @@
 #include <linux/delay.h>
 #include <linux/irq.h>
 #include <linux/io.h>
+#include <linux/dma-mapping.h>
 
 #include <sound/core.h>
 #include <sound/pcm.h>
@@ -66,7 +67,7 @@
 #include <sound/initval.h>
 
 #include <linux/of.h>
-#include <asm/sbus.h>
+#include <linux/of_device.h>
 #include <asm/atomic.h>
 
 MODULE_AUTHOR("Rudolf Koenig, Brent Baccala and Martin Habets");
@@ -297,7 +298,7 @@ struct dbri_streaminfo {
 /* This structure holds the information for both chips (DBRI & CS4215) */
 struct snd_dbri {
 	int regs_size, irq;	/* Needed for unload */
-	struct sbus_dev *sdev;	/* SBUS device info */
+	struct of_device *op;	/* OF device info */
 	spinlock_t lock;
 
 	struct dbri_dma *dma;	/* Pointer to our DMA block */
@@ -2093,14 +2094,15 @@ static int snd_dbri_hw_params(struct snd_pcm_substream *substream,
 	 */
 	if (info->dvma_buffer == 0) {
 		if (DBRI_STREAMNO(substream) == DBRI_PLAY)
-			direction = SBUS_DMA_TODEVICE;
+			direction = DMA_TO_DEVICE;
 		else
-			direction = SBUS_DMA_FROMDEVICE;
+			direction = DMA_FROM_DEVICE;
 
-		info->dvma_buffer = sbus_map_single(dbri->sdev,
-					runtime->dma_area,
-					params_buffer_bytes(hw_params),
-					direction);
+		info->dvma_buffer =
+			dma_map_single(&dbri->op->dev,
+				       runtime->dma_area,
+				       params_buffer_bytes(hw_params),
+				       direction);
 	}
 
 	direction = params_buffer_bytes(hw_params);
@@ -2121,12 +2123,12 @@ static int snd_dbri_hw_free(struct snd_pcm_substream *substream)
 	 */
 	if (info->dvma_buffer) {
 		if (DBRI_STREAMNO(substream) == DBRI_PLAY)
-			direction = SBUS_DMA_TODEVICE;
+			direction = DMA_TO_DEVICE;
 		else
-			direction = SBUS_DMA_FROMDEVICE;
+			direction = DMA_FROM_DEVICE;
 
-		sbus_unmap_single(dbri->sdev, info->dvma_buffer,
-				  substream->runtime->buffer_size, direction);
+		dma_unmap_single(&dbri->op->dev, info->dvma_buffer,
+				 substream->runtime->buffer_size, direction);
 		info->dvma_buffer = 0;
 	}
 	if (info->pipe != -1) {
@@ -2519,31 +2521,32 @@ static void __devinit snd_dbri_proc(struct snd_card *card)
 static void snd_dbri_free(struct snd_dbri *dbri);
 
 static int __devinit snd_dbri_create(struct snd_card *card,
-				  struct sbus_dev *sdev,
-				  int irq, int dev)
+				     struct of_device *op,
+				     int irq, int dev)
 {
 	struct snd_dbri *dbri = card->private_data;
 	int err;
 
 	spin_lock_init(&dbri->lock);
-	dbri->sdev = sdev;
+	dbri->op = op;
 	dbri->irq = irq;
 
-	dbri->dma = sbus_alloc_consistent(sdev, sizeof(struct dbri_dma),
-					  &dbri->dma_dvma);
+	dbri->dma = dma_alloc_coherent(&op->dev,
+				       sizeof(struct dbri_dma),
+				       &dbri->dma_dvma, GFP_ATOMIC);
 	memset((void *)dbri->dma, 0, sizeof(struct dbri_dma));
 
 	dprintk(D_GEN, "DMA Cmd Block 0x%p (0x%08x)\n",
 		dbri->dma, dbri->dma_dvma);
 
 	/* Map the registers into memory. */
-	dbri->regs_size = sdev->reg_addrs[0].reg_size;
-	dbri->regs = sbus_ioremap(&sdev->resource[0], 0,
-				  dbri->regs_size, "DBRI Registers");
+	dbri->regs_size = resource_size(&op->resource[0]);
+	dbri->regs = of_ioremap(&op->resource[0], 0,
+				dbri->regs_size, "DBRI Registers");
 	if (!dbri->regs) {
 		printk(KERN_ERR "DBRI: could not allocate registers\n");
-		sbus_free_consistent(sdev, sizeof(struct dbri_dma),
-				     (void *)dbri->dma, dbri->dma_dvma);
+		dma_free_coherent(&op->dev, sizeof(struct dbri_dma),
+				  (void *)dbri->dma, dbri->dma_dvma);
 		return -EIO;
 	}
 
@@ -2551,9 +2554,9 @@ static int __devinit snd_dbri_create(struct snd_card *card,
 			  "DBRI audio", dbri);
 	if (err) {
 		printk(KERN_ERR "DBRI: Can't get irq %d\n", dbri->irq);
-		sbus_iounmap(dbri->regs, dbri->regs_size);
-		sbus_free_consistent(sdev, sizeof(struct dbri_dma),
-				     (void *)dbri->dma, dbri->dma_dvma);
+		of_iounmap(&op->resource[0], dbri->regs, dbri->regs_size);
+		dma_free_coherent(&op->dev, sizeof(struct dbri_dma),
+				  (void *)dbri->dma, dbri->dma_dvma);
 		return err;
 	}
 
@@ -2577,27 +2580,23 @@ static void snd_dbri_free(struct snd_dbri *dbri)
 		free_irq(dbri->irq, dbri);
 
 	if (dbri->regs)
-		sbus_iounmap(dbri->regs, dbri->regs_size);
+		of_iounmap(&dbri->op->resource[0], dbri->regs, dbri->regs_size);
 
 	if (dbri->dma)
-		sbus_free_consistent(dbri->sdev, sizeof(struct dbri_dma),
-				     (void *)dbri->dma, dbri->dma_dvma);
+		dma_free_coherent(&dbri->op->dev,
+				  sizeof(struct dbri_dma),
+				  (void *)dbri->dma, dbri->dma_dvma);
 }
 
-static int __devinit dbri_probe(struct of_device *of_dev,
-				const struct of_device_id *match)
+static int __devinit dbri_probe(struct of_device *op, const struct of_device_id *match)
 {
-	struct sbus_dev *sdev = to_sbus_device(&of_dev->dev);
 	struct snd_dbri *dbri;
-	int irq;
 	struct resource *rp;
 	struct snd_card *card;
 	static int dev = 0;
+	int irq;
 	int err;
 
-	dprintk(D_GEN, "DBRI: Found %s in SBUS slot %d\n",
-		sdev->prom_name, sdev->slot);
-
 	if (dev >= SNDRV_CARDS)
 		return -ENODEV;
 	if (!enable[dev]) {
@@ -2605,7 +2604,7 @@ static int __devinit dbri_probe(struct of_device *of_dev,
 		return -ENOENT;
 	}
 
-	irq = sdev->irqs[0];
+	irq = op->irqs[0];
 	if (irq <= 0) {
 		printk(KERN_ERR "DBRI-%d: No IRQ.\n", dev);
 		return -ENODEV;
@@ -2618,12 +2617,12 @@ static int __devinit dbri_probe(struct of_device *of_dev,
 
 	strcpy(card->driver, "DBRI");
 	strcpy(card->shortname, "Sun DBRI");
-	rp = &sdev->resource[0];
+	rp = &op->resource[0];
 	sprintf(card->longname, "%s at 0x%02lx:0x%016Lx, irq %d",
 		card->shortname,
 		rp->flags & 0xffL, (unsigned long long)rp->start, irq);
 
-	err = snd_dbri_create(card, sdev, irq, dev);
+	err = snd_dbri_create(card, op, irq, dev);
 	if (err < 0) {
 		snd_card_free(card);
 		return err;
@@ -2640,7 +2639,7 @@ static int __devinit dbri_probe(struct of_device *of_dev,
 
 	/* /proc file handling */
 	snd_dbri_proc(card);
-	dev_set_drvdata(&of_dev->dev, card);
+	dev_set_drvdata(&op->dev, card);
 
 	err = snd_card_register(card);
 	if (err < 0)
@@ -2648,7 +2647,7 @@ static int __devinit dbri_probe(struct of_device *of_dev,
 
 	printk(KERN_INFO "audio%d at %p (irq %d) is DBRI(%c)+CS4215(%d)\n",
 	       dev, dbri->regs,
-	       dbri->irq, sdev->prom_name[9], dbri->mm.version);
+	       dbri->irq, op->node->name[9], dbri->mm.version);
 	dev++;
 
 	return 0;
@@ -2659,19 +2658,19 @@ _err:
 	return err;
 }
 
-static int __devexit dbri_remove(struct of_device *dev)
+static int __devexit dbri_remove(struct of_device *op)
 {
-	struct snd_card *card = dev_get_drvdata(&dev->dev);
+	struct snd_card *card = dev_get_drvdata(&op->dev);
 
 	snd_dbri_free(card->private_data);
 	snd_card_free(card);
 
-	dev_set_drvdata(&dev->dev, NULL);
+	dev_set_drvdata(&op->dev, NULL);
 
 	return 0;
 }
 
-static struct of_device_id dbri_match[] = {
+static const struct of_device_id dbri_match[] = {
 	{
 		.name = "SUNW,DBRIe",
 	},
@@ -2693,7 +2692,7 @@ static struct of_platform_driver dbri_sbus_driver = {
 /* Probe for the dbri chip and then attach the driver. */
 static int __init dbri_init(void)
 {
-	return of_register_driver(&dbri_sbus_driver, &sbus_bus_type);
+	return of_register_driver(&dbri_sbus_driver, &of_bus_type);
 }
 
 static void __exit dbri_exit(void)