Example backtraces: Mar 4 05:41:40 host kernel: [291932.968664] INFO: task tar:44877 blocked for more than 120 seconds. Mar 4 05:41:40 host kernel: [291932.968826] Tainted: P OE 5.4.0-100-generic #113~18.04.1-Ubuntu Mar 4 05:41:40 host kernel: [291932.969019] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. Mar 4 05:41:40 host kernel: [291932.969222] tar D 0 44877 44865 0x00004000 Mar 4 05:41:40 host kernel: [291932.969226] Call Trace: Mar 4 05:41:40 host kernel: [291932.969239] __schedule+0x292/0x720 Mar 4 05:41:40 host kernel: [291932.969315] ? xfs_buf_find.isra.34+0x205/0x640 [xfs] Mar 4 05:41:40 host kernel: [291932.969321] schedule+0x33/0xa0 Mar 4 05:41:40 host kernel: [291932.969325] schedule_timeout+0x1d3/0x320 Mar 4 05:41:40 host kernel: [291932.969381] ? xfs_btree_update+0x7f/0xe0 [xfs] Mar 4 05:41:40 host kernel: [291932.969444] ? xfs_buf_find.isra.34+0x205/0x640 [xfs] Mar 4 05:41:40 host kernel: [291932.969448] __down+0x91/0xe0 Mar 4 05:41:40 host kernel: [291932.969454] ? page_counter_try_charge+0x10/0xd0 Mar 4 05:41:40 host kernel: [291932.969460] down+0x41/0x50 Mar 4 05:41:40 host kernel: [291932.969463] ? down+0x41/0x50 Mar 4 05:41:40 host kernel: [291932.969521] xfs_buf_lock+0x3c/0xf0 [xfs] Mar 4 05:41:40 host kernel: [291932.969575] xfs_buf_find.isra.34+0x205/0x640 [xfs] Mar 4 05:41:40 host kernel: [291932.969629] xfs_buf_get_map+0x43/0x2b0 [xfs] Mar 4 05:41:40 host kernel: [291932.969684] xfs_buf_read_map+0x2c/0x1c0 [xfs] Mar 4 05:41:40 host kernel: [291932.969757] xfs_trans_read_buf_map+0xd5/0x360 [xfs] Mar 4 05:41:40 host kernel: [291932.969805] xfs_read_agf+0x92/0x120 [xfs] Mar 4 05:41:40 host kernel: [291932.969851] xfs_alloc_read_agf+0x47/0x1c0 [xfs] Mar 4 05:41:40 host kernel: [291932.969896] xfs_alloc_fix_freelist+0x2d2/0x500 [xfs] Mar 4 05:41:40 host kernel: [291932.969899] ? down+0x2e/0x50 Mar 4 05:41:40 host kernel: [291932.969955] ? xfs_buf_find.isra.34+0x205/0x640 [xfs] Mar 4 05:41:40 host kernel: [291932.969960] ? radix_tree_lookup+0xd/0x10 Mar 4 05:41:40 host kernel: [291932.970016] ? xfs_perag_get+0x2c/0xc0 [xfs] Mar 4 05:41:40 host kernel: [291932.970061] xfs_alloc_vextent+0x334/0x590 [xfs] Mar 4 05:41:40 host kernel: [291932.970115] xfs_ialloc_ag_alloc+0x17e/0x710 [xfs] Mar 4 05:41:40 host kernel: [291932.970183] ? xfs_trans_read_buf_map+0x183/0x360 [xfs] Mar 4 05:41:40 host kernel: [291932.970238] xfs_dialloc+0x139/0x280 [xfs] Mar 4 05:41:40 host kernel: [291932.970302] xfs_ialloc+0x7c/0x520 [xfs] Mar 4 05:41:40 host kernel: [291932.970364] xfs_dir_ialloc+0x62/0x1e0 [xfs] Mar 4 05:41:40 host kernel: [291932.970422] xfs_create+0x3d9/0x570 [xfs] Mar 4 05:41:40 host kernel: [291932.970479] xfs_generic_create+0x20e/0x2f0 [xfs] Mar 4 05:41:40 host kernel: [291932.970534] xfs_vn_mknod+0x14/0x20 [xfs] Mar 4 05:41:40 host kernel: [291932.970588] xfs_vn_create+0x13/0x20 [xfs] Mar 4 05:41:40 host kernel: [291932.970593] path_openat+0x12cb/0x16a0 Mar 4 05:41:40 host kernel: [291932.970659] ? xfs_trans_get_bud+0x4/0x70 [xfs] Mar 4 05:41:40 host kernel: [291932.970717] ? xfs_perag_get+0x2c/0xc0 [xfs] Mar 4 05:41:40 host kernel: [291932.970721] do_filp_open+0x9b/0x110 Mar 4 05:41:40 host kernel: [291932.970728] ? list_lru_add+0x6c/0x1b0 Mar 4 05:41:40 host kernel: [291932.970731] ? __check_object_size+0xdb/0x1b0 Mar 4 05:41:40 host kernel: [291932.970736] ? __alloc_fd+0x46/0x170 Mar 4 05:41:40 host kernel: [291932.970741] do_sys_open+0x1ba/0x2e0 Mar 4 05:41:40 host kernel: [291932.970744] ? do_sys_open+0x1ba/0x2e0 Mar 4 05:41:40 host kernel: [291932.970747] ? _cond_resched+0x19/0x40 Mar 4 05:41:40 host kernel: [291932.970751] ? task_work_run+0x46/0xc0 Mar 4 05:41:40 host kernel: [291932.970755] __x64_sys_openat+0x20/0x30 Mar 4 05:41:40 host kernel: [291932.970760] do_syscall_64+0x57/0x190 Mar 4 05:41:40 host kernel: [291932.970765] entry_SYSCALL_64_after_hwframe+0x44/0xa9 Mar 4 05:41:40 host kernel: [291932.970768] RIP: 0033:0x7f5e77b6ff83 Mar 4 05:41:40 host kernel: [291932.970775] Code: Bad RIP value. Mar 4 05:41:40 host kernel: [291932.970777] RSP: 002b:00007ffda22272e0 EFLAGS: 00000246 ORIG_RAX: 0000000000000101 Mar 4 05:41:40 host kernel: [291932.970781] RAX: ffffffffffffffda RBX: 00000000000809c1 RCX: 00007f5e77b6ff83 Mar 4 05:41:40 host kernel: [291932.970782] RDX: 00000000000809c1 RSI: 00005595a1b27d10 RDI: 0000000000000003 Mar 4 05:41:40 host kernel: [291932.970784] RBP: 00000000000001a4 R08: 0000000000000020 R09: 00000000ffffffd0 Mar 4 05:41:40 host kernel: [291932.970786] R10: 00000000000001a4 R11: 0000000000000246 R12: 0000000000000000 Mar 4 05:41:40 host kernel: [291932.970787] R13: 00005595a1b27d10 R14: 00005595a091f6f0 R15: 00007ffda2227377 Mar 4 05:41:40 host kernel: [291932.970796] INFO: task perl:50597 blocked for more than 120 seconds. Mar 4 05:41:40 host kernel: [291932.970952] Tainted: P OE 5.4.0-100-generic #113~18.04.1-Ubuntu Mar 4 05:41:40 host kernel: [291932.971147] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. Mar 4 05:41:40 host kernel: [291932.971354] perl D 0 50597 22548 0x00004000 Mar 4 05:41:40 host kernel: [291932.971357] Call Trace: Mar 4 05:41:40 host kernel: [291932.971362] __schedule+0x292/0x720 Mar 4 05:41:40 host kernel: [291932.971422] ? xfs_buf_find.isra.34+0x205/0x640 [xfs] Mar 4 05:41:40 host kernel: [291932.971425] schedule+0x33/0xa0 Mar 4 05:41:40 host kernel: [291932.971428] schedule_timeout+0x1d3/0x320 Mar 4 05:41:40 host kernel: [291932.971484] ? xfs_buf_trylock+0x1f/0xd0 [xfs] Mar 4 05:41:40 host kernel: [291932.971538] ? xfs_buf_find.isra.34+0x205/0x640 [xfs] Mar 4 05:41:40 host kernel: [291932.971542] __down+0x91/0xe0 Mar 4 05:41:40 host kernel: [291932.971546] down+0x41/0x50 Mar 4 05:41:40 host kernel: [291932.971549] ? down+0x41/0x50 Mar 4 05:41:40 host kernel: [291932.971601] xfs_buf_lock+0x3c/0xf0 [xfs] Mar 4 05:41:40 host kernel: [291932.971652] xfs_buf_find.isra.34+0x205/0x640 [xfs] Mar 4 05:41:40 host kernel: [291932.971705] xfs_buf_get_map+0x43/0x2b0 [xfs] Mar 4 05:41:40 host kernel: [291932.971757] xfs_buf_read_map+0x2c/0x1c0 [xfs] Mar 4 05:41:40 host kernel: [291932.971760] ? _cond_resched+0x19/0x40 Mar 4 05:41:40 host kernel: [291932.971828] xfs_trans_read_buf_map+0xd5/0x360 [xfs] Mar 4 05:41:40 host kernel: [291932.971883] xfs_read_agi+0x93/0x130 [xfs] Mar 4 05:41:40 host kernel: [291932.971945] xfs_iunlink_remove+0x66/0x260 [xfs] Mar 4 05:41:40 host kernel: [291932.972011] ? xfs_trans_reserve+0x14f/0x1d0 [xfs] Mar 4 05:41:40 host kernel: [291932.972072] xfs_ifree+0x42/0x160 [xfs] Mar 4 05:41:40 host kernel: [291932.972130] xfs_inactive_ifree+0xab/0x1c0 [xfs] Mar 4 05:41:40 host kernel: [291932.972185] xfs_inactive+0xa8/0x140 [xfs] Mar 4 05:41:40 host kernel: [291932.972247] xfs_fs_destroy_inode+0xaf/0x1e0 [xfs] Mar 4 05:41:40 host kernel: [291932.972251] destroy_inode+0x41/0x80 Mar 4 05:41:40 host kernel: [291932.972255] evict+0x139/0x1a0 Mar 4 05:41:40 host kernel: [291932.972258] iput+0x143/0x210 Mar 4 05:41:40 host kernel: [291932.972262] do_unlinkat+0x19c/0x320 Mar 4 05:41:40 host kernel: [291932.972266] __x64_sys_unlink+0x23/0x30 Mar 4 05:41:40 host kernel: [291932.972270] do_syscall_64+0x57/0x190 Mar 4 05:41:40 host kernel: [291932.972275] entry_SYSCALL_64_after_hwframe+0x44/0xa9 Mar 4 05:41:40 host kernel: [291932.972277] RIP: 0033:0x7f6d3a775e17 Mar 4 05:41:40 host kernel: [291932.972281] Code: Bad RIP value. Mar 4 05:41:40 host kernel: [291932.972283] RSP: 002b:00007ffe5a2888c8 EFLAGS: 00000206 ORIG_RAX: 0000000000000057 Mar 4 05:41:40 host kernel: [291932.972286] RAX: ffffffffffffffda RBX: 000055dc7dbde260 RCX: 00007f6d3a775e17 Mar 4 05:41:40 host kernel: [291932.972287] RDX: 00007ffe5a288900 RSI: 00007ffe5a288900 RDI: 000055dc7e43d640 Mar 4 05:41:40 host kernel: [291932.972289] RBP: 000055dc7e784ed8 R08: 0000000000000000 R09: 0000000000000006 Mar 4 05:41:40 host kernel: [291932.972291] R10: 000055dc7e623ff8 R11: 0000000000000206 R12: 0000000000000001 Mar 4 05:41:40 host kernel: [291932.972292] R13: 000055dc7e784ed8 R14: 00007ffe5a2888f8 R15: 000055dc7e43d640
** Description changed: SRU Justification: [Impact] * The xfs filesystem suffers from a deadlock issue in kernels < 5.5. This hangs IO to/from the affected filesystem. [Fix] * 93597ae8dac0149b5c00b787cba6bf7ba213e666 93597ae8dac0 "xfs: Fix deadlock between AGI and AGF when target_ip exists in xfs_rename()" + + * This is from the upstream 5.5 kernel. [Test Plan] * Set up an Ubuntu Bionic/Focal installation using kernel 5.4. * Create and mount an XFS filesystem on a block device. * Exercise the filesystem to verify that IO does not hang. [Where problems could occur] * This patch could cause locking issues on XFS filesystems, requiring a system restart to correct. [Other Info] * Bug is difficult to reproduce, a test kernel on affected systems with the above patch prevented the issue. + + * Backports to earlier (4.15 and earlier) kernels have been omitted as + the upstream patch does not apply cleanly and the issue has not been + reproduced on them. ** Description changed: SRU Justification: [Impact] * The xfs filesystem suffers from a deadlock issue in kernels < 5.5. - This hangs IO to/from the affected filesystem. + This hangs IO to/from the affected filesystem. Sample backtraces added + as a comment. [Fix] * 93597ae8dac0149b5c00b787cba6bf7ba213e666 93597ae8dac0 "xfs: Fix deadlock between AGI and AGF when target_ip exists in xfs_rename()" * This is from the upstream 5.5 kernel. [Test Plan] * Set up an Ubuntu Bionic/Focal installation using kernel 5.4. * Create and mount an XFS filesystem on a block device. * Exercise the filesystem to verify that IO does not hang. [Where problems could occur] * This patch could cause locking issues on XFS filesystems, requiring a system restart to correct. [Other Info] * Bug is difficult to reproduce, a test kernel on affected systems with the above patch prevented the issue. * Backports to earlier (4.15 and earlier) kernels have been omitted as the upstream patch does not apply cleanly and the issue has not been reproduced on them. -- You received this bug notification because you are a member of Kernel Packages, which is subscribed to linux in Ubuntu. https://bugs.launchpad.net/bugs/1966803 Title: xfs: Fix deadlock between AGI and AGF when target_ip exists in xfs_rename() Status in linux package in Ubuntu: Incomplete Bug description: SRU Justification: [Impact] * The xfs filesystem suffers from a deadlock issue in kernels < 5.5. This hangs IO to/from the affected filesystem. Sample backtraces added as a comment. [Fix] * 93597ae8dac0149b5c00b787cba6bf7ba213e666 93597ae8dac0 "xfs: Fix deadlock between AGI and AGF when target_ip exists in xfs_rename()" * This is from the upstream 5.5 kernel. [Test Plan] * Set up an Ubuntu Bionic/Focal installation using kernel 5.4. * Create and mount an XFS filesystem on a block device. * Exercise the filesystem to verify that IO does not hang. [Where problems could occur] * This patch could cause locking issues on XFS filesystems, requiring a system restart to correct. [Other Info] * Bug is difficult to reproduce, a test kernel on affected systems with the above patch prevented the issue. * Backports to earlier (4.15 and earlier) kernels have been omitted as the upstream patch does not apply cleanly and the issue has not been reproduced on them. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1966803/+subscriptions -- Mailing list: https://launchpad.net/~kernel-packages Post to : kernel-packages@lists.launchpad.net Unsubscribe : https://launchpad.net/~kernel-packages More help : https://help.launchpad.net/ListHelp
