Re: linux-next: Tree for June 10

From: Bartlomiej Zolnierkiewicz
Date: Wed Jun 10 2009 - 18:22:37 EST


On Wednesday 10 June 2009 11:43:09 Stephen Rothwell wrote:
> Hi all,
>
> Changes since 20090609:

commit 9df1bb9b516daeece159ab7fb262d01a0359247c
Author: Jens Axboe <jens.axboe@xxxxxxxxxx>
Date: Tue Jun 9 06:22:57 2009 +0200

Revert "block: Fix bounce limit setting in DM"

This reverts commit a05c0205ba031c01bba33a21bf0a35920eb64833.

DM doesn't need to access the bounce_pfn directly.

Signed-off-by: Jens Axboe <jens.axboe@xxxxxxxxxx>

and the following warning is back:

BUG: sleeping function called from invalid context at mm/page_alloc.c:1493
in_atomic(): 1, irqs_disabled(): 0, pid: 21882, name: cryptsetup
INFO: lockdep is turned off.
Pid: 21882, comm: cryptsetup Tainted: G M 2.6.30-rc8-next-20090610-08093-g6597e49-dirty #19
Call Trace:
[<c0122ca6>] __might_sleep+0xfb/0x102
[<c0162e4c>] __alloc_pages_internal+0x62/0x388
[<c014689d>] ? lockdep_init_map+0x8a/0x3fb
[<c015fced>] mempool_alloc_pages+0x1b/0x21
[<c0176d47>] mempool_alloc_pages_isa+0xb/0x10
[<c01600ed>] mempool_create_node+0x92/0xcc
[<c015fcc8>] ? mempool_free_pages+0x0/0xa
[<c0176d3c>] ? mempool_alloc_pages_isa+0x0/0x10
[<c0160134>] mempool_create+0xd/0x11
[<c0176d1c>] init_emergency_isa_pool+0x22/0x42
[<c0223749>] blk_queue_bounce_limit+0x2c/0x41
[<c030b3a7>] dm_table_set_restrictions+0x76/0x9b
[<c03094b0>] dm_swap_table+0x117/0x132
[<c030d9ce>] dev_suspend+0xf4/0x16f
[<c030e246>] dm_ctl_ioctl+0x18e/0x1cc
[<c030d8da>] ? dev_suspend+0x0/0x16f
[<c030e0b8>] ? dm_ctl_ioctl+0x0/0x1cc
[<c018b81f>] vfs_ioctl+0x22/0x69
[<c018bce9>] do_vfs_ioctl+0x483/0x4bd
[<c0193606>] ? mntput_no_expire+0x19/0xb4
[<c0181ee2>] ? __fput+0x185/0x18d
[<c0181f03>] ? fput+0x19/0x1b
[<c017f6c2>] ? filp_close+0x51/0x5b
[<c018bd4f>] sys_ioctl+0x2c/0x45
[<c0102988>] sysenter_do_call+0x12/0x36

[ The problem is in init_emergency_isa_pool() being called for DM. ]

The fix was probably lost because the original commit missed information
about the warning and also cc: to the reporter (me in this particular case).
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/