[LU-15117] ofd_read_lock vs transaction deadlock while allocating buffers

• Type: Bug
• Resolution: Fixed
• Priority: Critical
• Fix Version/s: Lustre 2.16.0
• Severity: 3
• Assignee: Andriy Skulysh
• Reporter: Andriy Skulysh

Description

PID: 154236  TASK: ffff9ab9b2f330c0  CPU: 9   COMMAND: "ll_ost_io01_002"
 #0 [ffff9ab9b2f6af58] __schedule at ffffffff9876ab17
 #1 [ffff9ab9b2f6afe0] schedule at ffffffff9876b019
 #2 [ffff9ab9b2f6aff0] wait_transaction_locked at ffffffffc0760085 [jbd2]
 #3 [ffff9ab9b2f6b048] add_transaction_credits at ffffffffc0760368 [jbd2]
 #4 [ffff9ab9b2f6b0a8] start_this_handle at ffffffffc07605e1 [jbd2]
 #5 [ffff9ab9b2f6b140] jbd2__journal_start at ffffffffc0760a93 [jbd2]
 #6 [ffff9ab9b2f6b188] __ldiskfs_journal_start_sb at ffffffffc19c1c79 [ldiskfs]
 #7 [ffff9ab9b2f6b1c8] ldiskfs_release_dquot at ffffffffc19b92ec [ldiskfs]
 #8 [ffff9ab9b2f6b1e8] dqput at ffffffff982aeb5d
 #9 [ffff9ab9b2f6b210] __dquot_drop at ffffffff982b0215
#10 [ffff9ab9b2f6b248] dquot_drop at ffffffff982b0285
#11 [ffff9ab9b2f6b258] ldiskfs_clear_inode at ffffffffc19bdcf2 [ldiskfs]
#12 [ffff9ab9b2f6b270] ldiskfs_evict_inode at ffffffffc19dccdf [ldiskfs]
#13 [ffff9ab9b2f6b2b0] evict at ffffffff9825ee14
#14 [ffff9ab9b2f6b2d8] dispose_list at ffffffff9825ef1e
#15 [ffff9ab9b2f6b300] prune_icache_sb at ffffffff9825ff2c
#16 [ffff9ab9b2f6b368] prune_super at ffffffff98244323
#17 [ffff9ab9b2f6b3a0] shrink_slab at ffffffff981ca105
#18 [ffff9ab9b2f6b440] do_try_to_free_pages at ffffffff981cd3c2
#19 [ffff9ab9b2f6b4b8] try_to_free_pages at ffffffff981cd5dc
#20 [ffff9ab9b2f6b550] __alloc_pages_slowpath at ffffffff987601ef
#21 [ffff9ab9b2f6b640] __alloc_pages_nodemask at ffffffff981c1465
#22 [ffff9ab9b2f6b6f0] alloc_pages_current at ffffffff9820e2c8
#23 [ffff9ab9b2f6b738] new_slab at ffffffff982192d5
#24 [ffff9ab9b2f6b770] ___slab_alloc at ffffffff9821ad4c
#25 [ffff9ab9b2f6b840] __slab_alloc at ffffffff9876160c
#26 [ffff9ab9b2f6b880] kmem_cache_alloc at ffffffff9821c3eb
#27 [ffff9ab9b2f6b8c0] __radix_tree_preload at ffffffff9837b7b9
#28 [ffff9ab9b2f6b8f0] radix_tree_maybe_preload at ffffffff9837bd0e
#29 [ffff9ab9b2f6b900] __add_to_page_cache_locked at ffffffff981b734a
#30 [ffff9ab9b2f6b940] add_to_page_cache_lru at ffffffff981b74b7
#31 [ffff9ab9b2f6b970] find_or_create_page at ffffffff981b783e
#32 [ffff9ab9b2f6b9b0] osd_bufs_get at ffffffffc1a773c3 [osd_ldiskfs]
#33 [ffff9ab9b2f6ba10] ofd_preprw_write at ffffffffc144f156 [ofd]
#34 [ffff9ab9b2f6ba90] ofd_preprw at ffffffffc14500ce [ofd]
#35 [ffff9ab9b2f6bb28] tgt_brw_write at ffffffffc0ece6e9 [ptlrpc]
#36 [ffff9ab9b2f6bca0] tgt_request_handle at ffffffffc0eccd4a [ptlrpc]
#37 [ffff9ab9b2f6bd30] ptlrpc_server_handle_request at ffffffffc0e72586 [ptlrpc]
#38 [ffff9ab9b2f6bde8] ptlrpc_main at ffffffffc0e7625a [ptlrpc]
#39 [ffff9ab9b2f6bec8] kthread at ffffffff980c1f81
#40 [ffff9ab9b2f6bf50] ret_from_fork_nospec_begin at ffffffff98777c1d

Issue Links

• is related to: LU-9728 out of memory on OSS causing allocation failures or hung threads