[ Upstream commit c3be57b6f35ef96a980ce84e59d6a5a8ca6184ad ]

On Sun, Jan 03, 2010 at 12:23:14AM +0000, Russell King wrote:
> - with IDE
>   - locks the interrupt line, and makes the machine extremely painful -
>     about an hour to get to the point of being able to unload the
>     pdc202xx_old module.

Having manually bisected kernel versions, I've narrowed it down to some
change between 2.6.30 and 2.6.31.  There's not much which has changed
between the two kernels, but one change stands out like a sore thumb:

+static int pdc202xx_test_irq(ide_hwif_t *hwif)
+{
+       struct pci_dev *dev     = to_pci_dev(hwif->dev);
+       unsigned long high_16   = pci_resource_start(dev, 4);
+       u8 sc1d                 = inb(high_16 + 0x1d);
+
+       if (hwif->channel) {
+               /*
+                * bit 7: error, bit 6: interrupting,
+                * bit 5: FIFO full, bit 4: FIFO empty
+                */
+               return ((sc1d & 0x50) == 0x40) ? 1 : 0;
+       } else  {
+               /*
+                * bit 3: error, bit 2: interrupting,
+                * bit 1: FIFO full, bit 0: FIFO empty
+                */
+               return ((sc1d & 0x05) == 0x04) ? 1 : 0;
+       }
+}

Reading the (documented as a 32-bit) system control register when the
interface is idle gives: 0x01da110c

So, the byte at 0x1d is 0x11, which is documented as meaning that the
primary and secondary FIFOs are empty.

The code above, which is trying to see whether an IRQ is pending, checks
for the IRQ bit to be one, and the FIFO bit to be zero - or in English,
to be non-empty.

Since during a BM-DMA read, the FIFOs will naturally be drained to the
PCI bus, the chance of us getting to the interface before this happens
are extremely small - and if we don't, it means we decide not to service
the interrupt.  Hence, the screaming interrupt problem with drivers/ide.

Fix this by only indicating an interrupt is ready if both the interrupt
and FIFO empty bits are at '1'.

This bug only affects PDC20246/PDC20247 (Promise Ultra33) based cards,
and has been tested on 2.6.31 and 2.6.33-rc2.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Tested-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

[ Upstream commit 9ce41aed0d392246eb788786253f242e829fd5e1 ]

This reverts commit a20b2a44eca52818ef52a94959480b7e6ea2f528.

As requested by David Fries.  This makes CDROMs which are slave drives
on a ribbon without a master disappear and causes other similar kinds
of badness.

Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

[ Upstream commit f75d4a238770d83d3a0475ce7f34e3fa37de161e ]

Bring back ->maskproc method since it is still needed for proper operation,
as noticed by Russell King:

> This change is bogus.
>
>         writeb(0, base + ICS_ARCIN_V6_INTROFFSET_1);
>         readb(base + ICS_ARCIN_V6_INTROFFSET_2);
>
>         writeb(0, base + ICS_ARCIN_V6_INTROFFSET_2);
>         readb(base + ICS_ARCIN_V6_INTROFFSET_1);
>
> This sequence of code does:
>
> 1. enable interrupt 1
> 2. disable interrupt 2
> 3. enable interrupt 2
> 4. disable interrupt 1
>
> which results in the interrupt for the second channel being enabled -
> leaving channel 1 blocked.
>
> Firstly, icside shares its two IDE channels with one DMA engine - so it's
> a simplex interface.  IDE supports those (or did when the code was written)
> serializing requests between the two interfaces.  libata does not.
>
> Secondly, the interrupt lines on icside float when there's no drive connected
> or when the drive has its NIEN bit set, which means that you get spurious
> screaming interrupts which can kill off all expansion card interrupts on
> the machine unless you disable the channel interrupt on the card.
>
> Since libata can not serialize the operation of the two channels like IDE
> can, the libata version of the icside driver does not contain the interrupt
> stearing logic.  Instead, it looks at the status after reset, and if
> nothing was found on that channel, it masks the interrupt from that
> channel.

This patch reverts changes done in commit dff8817 (I became confused due to
non-standard & undocumented ->maskproc method, anyway sorry about that).

Noticed-by: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc: (151 commits)
  powerpc: Fix usage of 64-bit instruction in 32-bit altivec code
  MAINTAINERS: Add PowerPC patterns
  powerpc/pseries: Track previous CPPR values to correctly EOI interrupts
  powerpc/pseries: Correct pseries/dlpar.c build break without CONFIG_SMP
  powerpc: Make "intspec" pointers in irq_host->xlate() const
  powerpc/8xx: DTLB Miss cleanup
  powerpc/8xx: Remove DIRTY pte handling in DTLB Error.
  powerpc/8xx: Start using dcbX instructions in various copy routines
  powerpc/8xx: Restore _PAGE_WRITETHRU
  powerpc/8xx: Add missing Guarded setting in DTLB Error.
  powerpc/8xx: Fixup DAR from buggy dcbX instructions.
  powerpc/8xx: Tag DAR with 0x00f0 to catch buggy instructions.
  powerpc/8xx: Update TLB asm so it behaves as linux mm expects.
  powerpc/8xx: Invalidate non present TLBs
  powerpc/pseries: Serialize cpu hotplug operations during deactivate Vs deallocate
  pseries/pseries: Add code to online/offline CPUs of a DLPAR node
  powerpc: stop_this_cpu: remove the cpu from the online map.
  powerpc/pseries: Add kernel based CPU DLPAR handling
  sysfs/cpu: Add probe/release files
  powerpc/pseries: Kernel DLPAR Infrastructure
  ...

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/ide-next-2.6:
  sl82c105: remove no longer needed debugging code
  sis5513: remove stale TODO
  pdc202xx_old: remove no longer needed debugging code
  cy82c693: remove no longer needed debugging code
  cmd64x: remove no longer needed debugging code
  alim15x3: remove obsolete and dangerous wdc_udma parameter
  ide: Increase WAIT_DRQ to accomodate some CF cards and SSD drives.
  cs5535: add pci id for AMD based CS5535 controllers
  slc90e66: fix UDMA handling
  drivers/ide/tx4938ide.c: use resource_size()
  drivers/ide/ide_platform.c: use resource_size()
  drivers/ide/au1xxx-ide.c: use resource_size()
  hpt366: remove dead old timing tables
  ide: update Kconfig text to mark as deprecated
  ide-tape: remove the BKL
  hpt366: kill unused #define's

Conflicts:
	include/linux/kvm.h

The hotplug mediabay has tendrils deep into drivers/ide code
which makes a libata port reather difficult. In addition it's
ugly and could be done better.

This reworks the interface between the mediabay and the rest
of the world so that:

   - Any macio_driver can now have a mediabay_event callback
which will be called when that driver sits on a mediabay and
it's been either plugged or unplugged. The device type is
passed as an argument. We can now move all the IDE cruft
into the IDE driver itself

   - A check_media_bay() function can be used to take a peek
at the type of device currently in the bay if any, a cleaner
variant of the previous function with the same name.

   - A pair of lock/unlock functions are exposed to allow the
IDE driver to block the hotplug callbacks during the initial
setup and probing of the bay in order to avoid nasty race
conditions.

   - The mediabay code no longer needs to spin on the status
register of the IDE interface when it detects an IDE device,
this is done just fine by the IDE code itself

Overall, less code, simpler, and allows for another driver
than our old drivers/ide based one.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>

* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev: (54 commits)
  Revert "pata_sis: Implement MWDMA for the UDMA 133 capable chips"
  libata: Clarify ata_set_lba_range_entries function
  libata: Report zeroed read after TRIM and max discard size
  pata_hpt3x2n: fix overclocked MWDMA0 timing
  pata_it8213: MWDMA0 is unsupported
  [libata] MWDMA0 is unsupported on PIIX-like PATA controllers
  pata_via: clear UDMA transfer mode bit for PIO and MWDMA
  pata_sis: Power Management fix
  pata_rz1000: Power Management fix
  pata_radisys: fix UDMA handling
  pata_ns87415: Power Management fix
  pata_marvell: fix marvell_pre_reset() documentation
  pata_legacy: add pointers to QDI65x0 documentation
  pata_legacy: fix access to control register for QDI6580
  pata_legacy: fix QDI6580DP support
  pata_it8213: fix it8213_pre_reset() documentation
  pata_it8213: fix wrong MWDMA timings being programmed
  pata_it8213: fix PIO2 underclocking
  pata_it8213: fix wrong PIO timings being programmed
  pata_it8213: fix UDMA handling
  ...

* git://git.kernel.org/pub/scm/linux/kernel/git/brodo/pcmcia-2.6: (50 commits)
  pcmcia: rework the irq_req_t typedef
  pcmcia: remove deprecated handle_to_dev() macro
  pcmcia: pcmcia_request_window() doesn't need a pointer to a pointer
  pcmcia: remove unused "window_t" typedef
  pcmcia: move some window-related code to pcmcia_ioctl.c
  pcmcia: Change window_handle_t logic to unsigned long
  pcmcia: Pass struct pcmcia_socket to pcmcia_get_mem_page()
  pcmcia: Pass struct pcmcia_device to pcmcia_map_mem_page()
  pcmcia: Pass struct pcmcia_device to pcmcia_release_window()
  drivers/pcmcia: remove unnecessary kzalloc
  pcmcia: correct handling for Zoomed Video registers in topic.h
  pcmcia: fix printk formats
  pcmcia: autoload module pcmcia
  pcmcia/staging: update comedi drivers
  PCMCIA: stop duplicating pci_irq in soc_pcmcia_socket
  PCMCIA: ss: allow PCI IRQs > 255
  PCMCIA: soc_common: remove 'dev' member from soc_pcmcia_socket
  PCMCIA: soc_common: constify soc_pcmcia_socket ops member
  PCMCIA: sa1111: remove duplicated initializers
  PCMCIA: sa1111: wrap soc_pcmcia_socket to contain sa1111 specific data
  ...