linux-toradex.git/fs, branch v2.6.20

[PATCH] revert blockdev direct io back to 2.6.19 version

2007-02-03T19:26:06+00:00

Andrew Vasquez is reporting as-iosched oopses and a 65% throughput
slowdown due to the recent special-casing of direct-io against
blockdevs.  We don't know why either of these things are occurring.

The patch minimally reverts us back to the 2.6.19 code for a 2.6.20
release.

Cc: Andrew Vasquez 
Cc: Ken Chen 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] aio: fix buggy put_ioctx call in aio_complete - v2

2007-02-03T19:26:06+00:00

An AIO bug was reported that sleeping function is being called in softirq
context:

BUG: warning at kernel/mutex.c:132/__mutex_lock_common()
Call Trace:
     [] __mutex_lock_slowpath+0x640/0x6c0
     [] mutex_lock+0x20/0x40
     [] flush_workqueue+0xb0/0x1a0
     [] __put_ioctx+0xc0/0x240
     [] aio_complete+0x2f0/0x420
     [] finished_one_bio+0x200/0x2a0
     [] dio_bio_complete+0x1c0/0x200
     [] dio_bio_end_aio+0x60/0x80
     [] bio_endio+0x110/0x1c0
     [] __end_that_request_first+0x180/0xba0
     [] end_that_request_chunk+0x30/0x60
     [] scsi_end_request+0x50/0x300 [scsi_mod]
     [] scsi_io_completion+0x200/0x8a0 [scsi_mod]
     [] sd_rw_intr+0x330/0x860 [sd_mod]
     [] scsi_finish_command+0x100/0x1c0 [scsi_mod]
     [] scsi_softirq_done+0x230/0x300 [scsi_mod]
     [] blk_done_softirq+0x160/0x1c0
     [] __do_softirq+0x200/0x240
     [] do_softirq+0x70/0xc0

See report: http://marc.theaimsgroup.com/?l=linux-kernel&m=116599593200888&w=2

flush_workqueue() is not allowed to be called in the softirq context.
However, aio_complete() called from I/O interrupt can potentially call
put_ioctx with last ref count on ioctx and triggers bug.  It is simply
incorrect to perform ioctx freeing from aio_complete.

The bug is trigger-able from a race between io_destroy() and aio_complete().
A possible scenario:

cpu0                               cpu1
io_destroy                         aio_complete
  wait_for_all_aios {                __aio_put_req
     ...                                 ctx->reqs_active--;
     if (!ctx->reqs_active)
        return;
  }
  ...
  put_ioctx(ioctx)

                                     put_ioctx(ctx);
                                        __put_ioctx
                                          bam! Bug trigger!

The real problem is that the condition check of ctx->reqs_active in
wait_for_all_aios() is incorrect that access to reqs_active is not
being properly protected by spin lock.

This patch adds that protective spin lock, and at the same time removes
all duplicate ref counting for each kiocb as reqs_active is already used
as a ref count for each active ioctx.  This also ensures that buggy call
to flush_workqueue() in softirq context is eliminated.

Signed-off-by: "Ken Chen" 
Cc: Zach Brown 
Cc: Suparna Bhattacharya 
Cc: Benjamin LaHaise 
Cc: Badari Pulavarty 
Cc: 
Acked-by: Jeff Moyer 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] procfs: Fix listing of /proc/NOT_A_TGID/task

2007-02-02T00:22:41+00:00

Listing /proc/PID/task were PID is not a TGID should not result in
duplicated entries.

	[g ~]$ pidof thunderbird-bin
	2751
	[g ~]$ ls /proc/2751/task
	2751  2770  2771  2824  2826  2834  2835  2851  2853
	[g ~]$ ls /proc/2770/task
	2751  2770  2771  2824  2826  2834  2835  2851  2853
	2770  2771  2824  2826  2834  2835  2851  2853
	[g ~]$

Signed-off-by: Guillaume Chazarain 
Acked-by: "Eric W. Biederman" 
Cc: Oleg Nesterov 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] endianness bug: ntohl() misspelled as >> 24 in fh_verify().

2007-02-02T00:17:06+00:00

Signed-off-by: Al Viro 
Signed-off-by: Linus Torvalds

[PATCH] ntfs: kmap_atomic() atomicity fix

2007-01-31T00:01:35+00:00

The KM_BIO_SRC_IRQ kmap slot requires local irq protection.

Acked-by: Anton Altaparmakov 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] Remove warning: VFS is out of sync with lock manager

2007-01-31T00:01:35+00:00

But keep it as a dprintk

The message can be generated in a quite normal situation:
 If a 'lock' request is interrupted, then the lock client needs to
  record that the server has the lock, incase it does.
 When we come the unlock, the server might say it doesn't, even
  though we think it does (or might) and this generates the message.

Signed-off-by: Neil Brown 
Acked-by: Trond Myklebust 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] ufs: reallocation fix

2007-01-30T16:26:45+00:00

In blocks reallocation function sometimes does not update some of
buffer_head::b_blocknr, which may and cause data damage.

Signed-off-by: Evgeniy Dushistov 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] ufs: truncate negative to unsigned fix

2007-01-30T16:26:45+00:00

During ufs_trunc_direct which is subroutine of ufs::truncate, we try the first
of all free parts of block and then whole blocks.  But we calculate size of
block's part to free in the wrong way.

This may cause bad update of used blocks and fragments statistic, and you can
got report that you have free 32T on 1Gb partition.

Signed-off-by: Evgeniy Dushistov 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] ufs: alloc metadata null page fix

2007-01-30T16:26:45+00:00

These series of patches result of UFS1 write support stress testing, like
running fsx-linux, untar and build linux kernel etc

We pass from ufs::get_block_t to levels below: pointer to the current page, to
make possible things like reallocation of blocks on the fly, and we also uses
this pointer for indication, what actually we allocate data block or meta data
block, but currently we make decision about what we allocate on the wrong
level, this may and cause oops if we allocate blocks in some special order.

Signed-off-by: Evgeniy Dushistov 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] fuse: fix bug in control filesystem mount

2007-01-30T16:26:45+00:00

The BUG in fuse_ctl_add_dentry() could be triggered if the control
filesystem was unmounted and mounted again while one or more fuse
filesystems were present.

The fix is to reset the dentry counter in fuse_ctl_kill_sb().

Bug reported by Florent Mertens.

Signed-off-by: Miklos Szeredi 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds