On Tue, Jan 30, 2018 at 11:40:04PM +0500, mikhail wrote: > Hi. > > I launched several application which highly use I/O on start and it > caused freezing system for several second. > > All traces lead to xfs. > > Whether there is a useful info in trace or just it means that disk is slow? Could be a disk that is slow, or could be many other things. More information required: http://xfs.org/index.php/XFS_FAQ#Q:_What_information_should_I_include_when_reporting_a_problem.3F .... > [ 369.301111] disk_cache:0 D12928 5241 5081 0x00000000 Your "disk_cache" process is walking the inobt during inode allocation: > [ 369.301118] Call Trace: > [ 369.301124] __schedule+0x2dc/0xba0 > [ 369.301133] ? wait_for_completion+0x10e/0x1a0 > [ 369.301137] schedule+0x33/0x90 > [ 369.301140] schedule_timeout+0x25a/0x5b0 > [ 369.301146] ? mark_held_locks+0x5f/0x90 > [ 369.301150] ? _raw_spin_unlock_irq+0x2c/0x40 > [ 369.301153] ? wait_for_completion+0x10e/0x1a0 > [ 369.301157] ? trace_hardirqs_on_caller+0xf4/0x190 > [ 369.301162] ? wait_for_completion+0x10e/0x1a0 > [ 369.301166] wait_for_completion+0x136/0x1a0 > [ 369.301172] ? wake_up_q+0x80/0x80 > [ 369.301203] ? _xfs_buf_read+0x23/0x30 [xfs] > [ 369.301232] xfs_buf_submit_wait+0xb2/0x530 [xfs] > [ 369.301262] _xfs_buf_read+0x23/0x30 [xfs] > [ 369.301290] xfs_buf_read_map+0x14b/0x300 [xfs] > [ 369.301324] ? xfs_trans_read_buf_map+0xc4/0x5d0 [xfs] > [ 369.301360] xfs_trans_read_buf_map+0xc4/0x5d0 [xfs] > [ 369.301390] xfs_btree_read_buf_block.constprop.36+0x72/0xc0 [xfs] > [ 369.301423] xfs_btree_lookup_get_block+0x88/0x180 [xfs] > [ 369.301454] xfs_btree_lookup+0xcd/0x410 [xfs] > [ 369.301462] ? rcu_read_lock_sched_held+0x79/0x80 > [ 369.301495] ? kmem_zone_alloc+0x6c/0xf0 [xfs] > [ 369.301530] xfs_dialloc_ag_update_inobt+0x49/0x120 [xfs] > [ 369.301557] ? xfs_inobt_init_cursor+0x3e/0xe0 [xfs] > [ 369.301588] xfs_dialloc_ag+0x17c/0x260 [xfs] > [ 369.301616] ? xfs_dialloc+0x236/0x270 [xfs] > [ 369.301652] xfs_dialloc+0x59/0x270 [xfs] > [ 369.301718] xfs_ialloc+0x6a/0x520 [xfs] > [ 369.301724] ? find_held_lock+0x3c/0xb0 > [ 369.301757] xfs_dir_ialloc+0x67/0x210 [xfs] > [ 369.301792] xfs_create+0x514/0x840 [xfs] > [ 369.301833] xfs_generic_create+0x1fa/0x2d0 [xfs] > [ 369.301865] xfs_vn_mknod+0x14/0x20 [xfs] > [ 369.301889] xfs_vn_mkdir+0x16/0x20 [xfs] > [ 369.301893] vfs_mkdir+0x10c/0x1d0 > [ 369.301900] SyS_mkdir+0x7e/0xf0 > [ 369.301909] entry_SYSCALL_64_fastpath+0x1f/0x96 And everything else is backed up behind it trying to allocate inodes. There could be many, many reasons for that, and that's why we need more information to begin to isolate the cause. Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>
