[LU-6118] locking flaw generates logged errors

Duplicate — Tue, 24 Jul 2018 13:45:26 +0000

From time to time, we can see the following stack trace on our Lustre clients.

 WARNING: at fs/ext4/inode.c:3929 ext4_flush_unwritten_io+0x74/0x80 [ext4]() (Tainted: P           ---------------   )
Hardware name: bullx blade
Modules linked in: sha512_generic crc32c_intel libcfs(U) panfs(P)(U) sunrpc acpi_cpufreq freq_table mperf rdma_ucm(U) rdma_cm(U) iw_cm(U) ib_addr(U) ib_ipoib(U) ib_cm(U) ipv6 ib_uverbs(U) ib_umad(U) mlx4_ib(U) ib_sa(U) mlx4_core(U) ib_mthca(U) ib_mad(U) dm_mirror dm_region_hash dm_log dm_mod mic(U) ipmi_devintf ipmi_si ipmi_msghandler sg lpc_ich mfd_core igb dca i2c_algo_bit i2c_core ptp pps_core mlx5_ib(U) mlx5_core(U) ib_core(U) compat(U) ext4 jbd2 mbcache sd_mod crc_t10dif ahci xhci_hcd megaraid_sas [last unloaded: lvfs]
Pid: 18257, comm: lctl Tainted: P           ---------------    2.6.32-504.1.3.el6.x86_64 #1
Call Trace:
 [<ffffffff81074df7>] ? warn_slowpath_common+0x87/0xc0
 [<ffffffff81074e4a>] ? warn_slowpath_null+0x1a/0x20
 [<ffffffffa009fbb4>] ? ext4_flush_unwritten_io+0x74/0x80 [ext4]
 [<ffffffffa009bfc8>] ? ext4_sync_file+0x88/0x1d0 [ext4]
 [<ffffffffa11e43e8>] ? cfs_tracefile_dump_all_pages+0x188/0x2d0 [libcfs]
 [<ffffffffa11e45bb>] ? cfs_trace_dump_debug_buffer_usrstr+0x8b/0x90 [libcfs]
 [<ffffffffa11db6f3>] ? __proc_dump_kernel+0x23/0x30 [libcfs]
 [<ffffffffa11db0eb>] ? proc_call_handler+0x2b/0x70 [libcfs]
 [<ffffffffa11db185>] ? proc_dump_kernel+0x25/0x30 [libcfs]
 [<ffffffff81203b07>] ? proc_sys_call_handler+0x97/0xd0
 [<ffffffff81203b54>] ? proc_sys_write+0x14/0x20
 [<ffffffff8118e058>] ? vfs_write+0xb8/0x1a0
 [<ffffffff8118ea21>] ? sys_write+0x51/0x90
 [<ffffffff8100b072>] ? system_call_fastpath+0x16/0x1b

This issue is a duplicate of ~~LU-5912~~, but for Lustre 2.5.3. Could you provide a backport of http://review.whamcloud.com/12731/ ?

Whamcloud Community JIRA

[LU-6118] locking flaw generates logged errors