CPU Lock Up in Function ib_fmr_pool_map_phys()

(Doc ID 2106855.1)

Last updated on FEBRUARY 22, 2016

Applies to:

Linux OS - Version Oracle Linux 5.6 with Unbreakable Enterprise Kernel [2.6.32] and later
Linux x86-64
This note describes a specific kernel bug and the kernel version which has the fix for.

Symptoms

You might see the below error messages in the syslog:

137844 Jan 26 06:35:30 ex01dbadm01 kernel: NMI backtrace for cpu 3
137845 Jan 26 06:35:30 ex01dbadm01 kernel: CPU 3
137847 Jan 26 06:35:30 ex01dbadm01 kernel:
137848 Jan 26 06:35:30 ex01dbadm01 kernel: Pid: 5884, comm: oracle Tainted: G W 2.6.39-400.128.17.el5uek #1 Oracle Corporation SUN FIRE X4170 M3 /ASSY,MOTHERBOARD,1U
137849 Jan 26 06:35:30 ex01dbadm01 kernel: RIP: 0010:[<ffffffffa02e30a0>] [<ffffffffa02e30a0>] ib_fmr_pool_map_phys+0xe0/0x470 [ib_core]
...
137861 Jan 26 06:35:30 ex01dbadm01 kernel: Process oracle (pid: 5884, threadinfo ffff8819587c4000, task ffff8809b55aa0c0)
137862 Jan 26 06:35:30 ex01dbadm01 kernel: Stack:
137863 Jan 26 06:35:30 ex01dbadm01 kernel: ffff8819587c5488 ffffffff8126a61b 0000000000000000 ffff880a359c1000
137864 Jan 26 06:35:30 ex01dbadm01 kernel: ffff8819587c54d8 ffff8819587c5538 0000000000000000 000000000000000c
137865 Jan 26 06:35:30 ex01dbadm01 kernel: ffff881974710000 ffff880a55ff2670 ffff8819587c54d8 ffffffffa02aeaeb
137866 Jan 26 06:35:30 ex01dbadm01 kernel: Call Trace:
137867 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8126a61b>] ? is_swiotlb_buffer+0x2b/0x40
137868 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffffa02aeaeb>] srp_map_finish_fmr+0x3b/0xa0 [ib_srp]
137869 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffffa02b05a6>] srp_map_data+0x1d6/0x5a0 [ib_srp]
137870 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffffa02b0eab>] srp_queuecommand+0x16b/0x2c0 [ib_srp]
137871 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8135d7dc>] scsi_dispatch_cmd+0x13c/0x2d0
137872 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff81367229>] scsi_request_fn+0x139/0x4a0
137873 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8123268a>] __blk_run_queue+0x1a/0x20
137874 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8123160f>] __elv_add_request+0x9f/0x1b0
137875 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8123455d>] add_acct_request+0x3d/0x50
137876 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff81506b84>] ? _raw_spin_lock_irqsave+0x34/0x50
137877 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff81236434>] blk_insert_cloned_request+0x64/0xa0
137878 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffffa0099c4e>] dm_dispatch_request+0x3e/0x60 [dm_mod]
137879 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffffa0099d9e>] map_request+0x8e/0xf0 [dm_mod]
137880 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffffa009acd6>] dm_request_fn+0x76/0x170 [dm_mod]
137881 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8123268a>] __blk_run_queue+0x1a/0x20
137882 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff81238327>] __make_request+0x1e7/0x280
137883 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffffa0097666>] dm_make_request+0x16/0x20 [dm_mod]
137884 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffffa009a547>] dm_request+0x37/0x50 [dm_mod]
137885 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff812367c0>] generic_make_request+0x260/0x4b0
137886 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff81236a86>] submit_bio+0x76/0x120
137887 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff81041009>] ? default_spin_lock_flags+0x9/0x10
137888 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff81506b84>] ? _raw_spin_lock_irqsave+0x34/0x50
137889 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff811a16ed>] dio_send_cur_page+0x42d/0x460
137890 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff811a1da8>] do_direct_IO+0x278/0x800
137891 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff811a28e6>] __blockdev_direct_IO+0x2c6/0x370
137892 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8119fd00>] ? ioctl_by_bdev+0x50/0x50
137893 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8119f897>] blkdev_direct_IO+0x57/0x60
137894 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8119fd00>] ? ioctl_by_bdev+0x50/0x50
137895 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8110f425>] mapping_direct_IO+0x45/0x60
137896 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff811109be>] generic_file_direct_write_iter+0x9e/0x150
137897 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8111238f>] __generic_file_write_iter+0x2df/0x3a0
137898 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff810cecbf>] ? audit_sockaddr+0x6f/0xa0
137899 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff811124d5>] __generic_file_aio_write+0x85/0xa0
137900 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8119e966>] blkdev_aio_write+0x26/0x80
137901 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8116baf1>] do_sync_write+0xe1/0x120
137902 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff810cda15>] ? audit_filter_rules+0x4d5/0x660
137903 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff811fb593>] ? security_file_permission+0x23/0x90
137904 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8116bfce>] vfs_write+0xce/0x190
137905 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8116c538>] sys_pwrite64+0x88/0x90
137906 Jan 26 06:35:30 ex01dbadm01 kernel: [<ffffffff8150f302>] system_call_fastpath+0x16/0x1b

Changes

 No change were made to the system.

Cause

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms