Hi all,
I just ran across a kernel bug seemingly in md. It's not something
people should run across normally, but here are details.
* System is a regular x86 running a Core2 processor in 32-bit mode.
* The issue happens with raid1, raid5, and raid6 personalities (that's
all I tested with).
* I was not able to reproduce the issue without md.
* It showed up for me starting with vanilla 2.6.27, and is still an
issue with 2.6.27.7 and 2.6.28-rc6, but it did not happen with 2.6.26.
It is very easy to reproduce:
1) Create an md array with >= 1 disk
2) Start a task writing to the array ("dd if=/dev/zero of=/dev/md0
bs=1M count=10000 &" does the trick for me)
3) Force an improper reboot with reboot -fn
If you need any more information, let me know.
Console capture is below:
nv6-j1:/mnt# reboot -nf
md: stopping all md devices.
------------[ cut here ]------------
Kernel BUG at c0682ca7 [verbose debug info unavailable]
invalid opcode: 0000 [#1] SMP
Modules linked in:
Pid: 3661, comm: kjournald Not tainted (2.6.27.6 #1)
EIP: 0060:[<c0682ca7>] EFLAGS: 00010246 CPU: 1
EIP is at md_write_start+0x157/0x160
EAX: 00000001 EBX: f71c7e00 ECX: f2bb7940 EDX: f2bb7940
ESI: 000000ff EDI: f2bb7940 EBP: 00000008 ESP: f65d3da4
DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068
Process kjournald (pid: 3661, ti=f65d2000 task=f6cef540 task.ti=f65d2000)
Stack: f717bdc0 c06890e5 007f40a0 00000000 00000000 f7144920 000000ff f2bb7940
c06783e0 00000000 00000001 00000008 f73c1ca0 007f4218 00000000 f702b96c
f2bb7840 f71b2bc0 11027f60 f2bb7940 f702b880 f71c7e00 007f4220 ef326398
Call Trace:
[<c06890e5>] __map_bio+0x35/0xa0
[<c06783e0>] make_request+0x40/0x6b0
[<c068a444>] dm_request+0xd4/0x120
[<c0581b1f>] generic_make_request+0x12f/0x280
[<c044a5ed>] mempool_alloc+0x2d/0xe0
[<c0583108>] submit_bio+0x58/0xf0
[<c048d5c5>] bvec_alloc_bs+0x65/0x110
[<c048d891>] bio_alloc_bioset+0x61/0x90
[<c0489a31>] submit_bh+0xd1/0x110
[<c04c8eaa>] journal_do_submit_data+0x2a/0x40
[<c04c9c0d>] journal_commit_transaction+0xcad/0xcf0
[<c0430fe0>] autoremove_wake_function+0x0/0x40
[<c0427b75>] try_to_del_timer_sync+0x45/0x50
[<c04cc429>] kjournald+0xa9/0x1c0
[<c0430fe0>] autoremove_wake_function+0x0/0x40
[<c04cc380>] kjournald+0x0/0x1c0
[<c0430ce2>] kthread+0x42/0x70
[<c0430ca0>] kthread+0x0/0x70
[<c0403c13>] kernel_thread_helper+0x7/0x14
=======================
Code: ff ff e9 f9 fe ff ff c7 83 34 01 00 00 00 00 00 00 f0 80 4b 18
02 8b 83 dc 00 00 00 be 01 00 0
0 00 e8 ee 9a ff ff e9 7c ff ff ff <0f> 0b eb fe 90 8d 74 26 00 83 ec
10 89 74 24 04 89 7c 24 08 89
EIP: [<c0682ca7>] md_write_start+0x157/0x160 SS:ESP 0068:f65d3da4
------------[ cut here ]------------
---[ end trace 02aba934bad77262 ]---
Kernel BUG at c0682ca7 [verbose debug info unavailable]
invalid opcode: 0000 [#2] SMP
Modules linked in:
Pid: 3665, comm: pdflush Tainted: G D (2.6.27.6 #1)
EIP: 0060:[<c0682ca7>] EFLAGS: 00010246 CPU: 0
EIP is at md_write_start+0x157/0x160
EAX: 00000001 EBX: f71c7e00 ECX: f2bb7cc0 EDX: f2bb7cc0
ESI: 000000ff EDI: f2bb7cc0 EBP: 00000008 ESP: c5b25c44
DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068
Process pdflush (pid: 3665, ti=c5b24000 task=f701a040 task.ti=c5b24000)
Stack: f717bdc0 c06890e5 007f50a0 00000000 00000000 f7144920 000000ff f2bb7cc0
c06783e0 00000000 00000001 00000008 f73c1ca0 007f4210 00000000 f702b96c
f2bb7a40 f71b2bc0 11026f60 f2bb7cc0 f702b880 f71c7e00 007f4218 efc04fa8
Call Trace:
[<c06890e5>] __map_bio+0x35/0xa0
[<c06783e0>] make_request+0x40/0x6b0
[<c068a444>] dm_request+0xd4/0x120
[<c0581b1f>] generic_make_request+0x12f/0x280
[<c044a5ed>] mempool_alloc+0x2d/0xe0
[<c0583108>] submit_bio+0x58/0xf0
[<c048d5c5>] bvec_alloc_bs+0x65/0x110
[<c048d891>] bio_alloc_bioset+0x61/0x90
[<c0489a31>] submit_bh+0xd1/0x110
[<c048b8bd>] __block_write_full_page+0x1ed/0x360
[<c04c7d9f>] start_this_handle+0x8f/0x330
[<c04b7e00>] ext3_get_block+0x0/0x110
[<c048bb1a>] block_write_full_page+0xea/0x100
[<c04b7e00>] ext3_get_block+0x0/0x110
[<c04b9713>] ext3_ordered_writepage+0xa3/0x170
[<c04b64d0>] bget_one+0x0/0x10
[<c044dbe8>] __writepage+0x8/0x30
[<c044e13f>] write_cache_pages+0x20f/0x320
[<c044dbe0>] __writepage+0x0/0x30
[<c044e270>] generic_writepages+0x20/0x30
[<c044e2c9>] do_writepages+0x49/0x50
[<c0485aea>] __writeback_single_inode+0x8a/0x2e0
[<c068bb7a>] dm_table_any_congested+0x2a/0x60
[<c04860fe>] generic_sync_sb_inodes+0x1ce/0x2d0
[<c048656b>] writeback_inodes+0x7b/0xa0
[<c044eba8>] background_writeout+0x98/0xc0
[<c044f0b0>] pdflush+0x0/0x1a0
[<c044f19b>] pdflush+0xeb/0x1a0
[<c044eb10>] background_writeout+0x0/0xc0
[<c0430ce2>] kthread+0x42/0x70
[<c0430ca0>] kthread+0x0/0x70
[<c0403c13>] kernel_thread_helper+0x7/0x14
=======================
Code: ff ff e9 f9 fe ff ff c7 83 34 01 00 00 00 00 00 00 f0 80 4b 18
02 8b 83 dc 00 00 00 be 01 00 0
0 00 e8 ee 9a ff ff e9 7c ff ff ff <0f> 0b eb fe 90 8d 74 26 00 83 ec
10 89 74 24 04 89 7c 24 08 89
EIP: [<c0682ca7>] md_write_start+0x157/0x160 SS:ESP 0068:c5b25c44
---[ end trace 02aba934bad77262 ]---
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
