[LU-2501] (osc_cache.c:1344:osc_release_write_grant()) ASSERTION( __raw_spin_is_locked(&(&cli->cl_loi_list_lock.lock)->raw_lock) ) fai led

• Status: Resolved
• Project: Lustre
• Affects Version/s: Lustre 2.4.0
• Fix Version/s: Lustre 2.4.0
• Type: Bug
• Priority: Blocker
• Reporter: Oleg Drokin
• Assignee: Jinshan Xiong (Inactive)
• Resolution: Fixed
• Labels: HB
• Severity: 3

Description

Just hit this running racer.
This is on master + this patch: http://review.whamcloud.com/#change,4617,patchset=2
Not sure if the failure is due to the patch or not:

[73834.653582] LustreError: 12616:0:(osc_cache.c:1344:osc_release_write_grant())
 ASSERTION( __raw_spin_is_locked(&(&cli->cl_loi_list_lock.lock)->raw_lock) ) fai
led: 
[73834.654273] LustreError: 12616:0:(osc_cache.c:1344:osc_release_write_grant())
 LBUG
[73834.654693] Pid: 12616, comm: cp
[73834.654890] 
[73834.654890] Call Trace:
[73834.655252]  [<ffffffffa0d69915>] libcfs_debug_dumpstack+0x55/0x80 [libcfs]
[73834.655516]  [<ffffffffa0d69f27>] lbug_with_loc+0x47/0xb0 [libcfs]
[73834.655776]  [<ffffffffa053b15b>] osc_queue_async_io+0x146b/0x14a8 [osc]
[73834.656032]  [<ffffffff81277e32>] ? number+0x222/0x320
[73834.656261]  [<ffffffff810385d8>] ? pvclock_clocksource_read+0x58/0xd0
[73834.656512]  [<ffffffff8127a646>] ? vsnprintf+0x2b6/0x5f0
[73834.656762]  [<ffffffffa0520169>] osc_page_cache_add+0xc9/0x1d0 [osc]
[73834.657042]  [<ffffffffa0ee34f7>] cl_page_cache_add+0x77/0x250 [obdclass]
[73834.657312]  [<ffffffffa05a0b55>] lov_page_cache_add+0x85/0x210 [lov]
[73834.657578]  [<ffffffffa0ee34f7>] cl_page_cache_add+0x77/0x250 [obdclass]
[73834.657856]  [<ffffffffa0edfa39>] ? cl_env_hops_keycmp+0x19/0x70 [obdclass]
[73834.658132]  [<ffffffffa0d7ff3d>] ? cfs_hash_bd_lookup_intent+0xdd/0x130 [libcfs]
[73834.658902]  [<ffffffffa0a5bdfd>] vvp_io_commit_write+0x3ad/0x590 [lustre]
[73834.659227]  [<ffffffffa0ef2d2d>] cl_io_commit_write+0xad/0x1e0 [obdclass]
[73834.659548]  [<ffffffffa0a312ce>] ll_commit_write+0xee/0x360 [lustre]
[73834.659858]  [<ffffffffa0a49020>] ll_write_end+0x30/0x60 [lustre]
[73834.660153]  [<ffffffff81112d8a>] generic_file_buffered_write+0x18a/0x300
[73834.660462]  [<ffffffff8106fea7>] ? current_fs_time+0x27/0x30
[73834.660785]  [<ffffffff811147e0>] __generic_file_aio_write+0x250/0x480
[73834.661238]  [<ffffffff81114a7f>] generic_file_aio_write+0x6f/0xe0
[73834.661546]  [<ffffffffa0a5c71c>] vvp_io_write_start+0x9c/0x240 [lustre]
[73834.661881]  [<ffffffffa0eef78a>] cl_io_start+0x6a/0x140 [obdclass]
[73834.662201]  [<ffffffffa0ef3f64>] cl_io_loop+0xb4/0x1b0 [obdclass]
[73834.662479]  [<ffffffffa0a0863b>] ll_file_io_generic+0x42b/0x550 [lustre]
[73834.662780]  [<ffffffffa0a09542>] ll_file_aio_write+0x142/0x2c0 [lustre]
[73834.663096]  [<ffffffffa0a0982c>] ll_file_write+0x16c/0x2a0 [lustre]
[73834.663393]  [<ffffffff8117b2e8>] vfs_write+0xb8/0x1a0
[73834.663660]  [<ffffffff8117bbb1>] sys_write+0x51/0x90
[73834.663943]  [<ffffffff8100b0f2>] system_call_fastpath+0x16/0x1b
[73834.664235] 
[73834.682530] Kernel panic - not syncing: LBUG

crashdump + modules are in /exports/crashdumps/192.168.10.211-2012-12-15-14:13:58

Comments

[Jinshan Xiong (Inactive)]

obviously the cli list lock is not held when osc_release_write_grant() is called. patch is at: http://review.whamcloud.com/4834

[Andreas Dilger]

Assign to Jinshan, since he already has a patch.

[Jodi Levi (Inactive)]

Since Change, 4834 has landed, can this ticket be closed?