https://bugzilla.kernel.org/show_bug.cgi?id=216343 --- Comment #3 from zhoukete@xxxxxxx --- (In reply to Amir Goldstein from comment #2) > > I don't think so. > I think this buffer write is in-flight. But how to explain the XFS_DONE flag is set.Is XFS_DONE flag means the io had been done？ > > Not sure if space cannot be released or just takes a lot of time. crash> ps -m 2006840 [0 14:35:38.418] [UN] PID: 2006840 TASK: ffff035f80de8f80 CPU: 51 COMMAND: "onestor-peon" pid 2006840 hang 14 hours. It is waiting for the xfs buf io finish. So I think it is ‘cannot be released’ not take a lot of time. crash> bt 2006840 PID: 2006840 TASK: ffff035f80de8f80 CPU: 51 COMMAND: "onestor-peon" #0 [ffff80002a873580] __switch_to at ffff800010016dec #1 [ffff80002a8735b0] __schedule at ffff800010a99d60 #2 [ffff80002a873650] schedule at ffff800010a9a1cc #3 [ffff80002a873670] schedule_timeout at ffff800010a9ee00 #4 [ffff80002a873710] __down at ffff800010a9d2ec #5 [ffff80002a8737a0] down at ffff8000100ea30c #6 [ffff80002a8737c0] xfs_buf_lock at ffff800008f5bd4c [xfs] #7 [ffff80002a8737f0] xfs_buf_find at ffff800008f5c3f0 [xfs] #8 [ffff80002a8738b0] xfs_buf_get_map at ffff800008f5c7d4 [xfs] #9 [ffff80002a873930] xfs_buf_read_map at ffff800008f5cf10 [xfs] #10 [ffff80002a8739a0] xfs_trans_read_buf_map at ffff800008f9c198 [xfs] #11 [ffff80002a873a20] xfs_da_read_buf at ffff800008f2b51c [xfs] #12 [ffff80002a873aa0] xfs_da3_node_read at ffff800008f2b5a4 [xfs] #13 [ffff80002a873ac0] xfs_da3_node_lookup_int at ffff800008f2c778 [xfs] #14 [ffff80002a873b60] xfs_dir2_node_removename at ffff800008f38964 [xfs] #15 [ffff80002a873c30] xfs_dir_removename at ffff800008f2fc6c [xfs] #16 [ffff80002a873c90] xfs_remove at ffff800008f77f70 [xfs] #17 [ffff80002a873d00] xfs_vn_unlink at ffff800008f72744 [xfs] #18 [ffff80002a873d50] vfs_unlink at ffff8000103264e8 #19 [ffff80002a873d90] do_unlinkat at ffff80001032ad44 #20 [ffff80002a873e20] __arm64_sys_unlinkat at ffff80001032ae30 > There are several AIL/CIL improvements in upstream kernel and > none of them are going to land in 5.10.y. > The reported kernel version 5.10.38 has almost no upstream fixes > at all, but I don't think that any of the fixes in 5.10.y are relevant for > this case anyway. > > If this hang happens often with your workload, I suggest using > a newer kernel and/or formatting xfs with a larger log to meet > the demands of your workload. > It is just happens only once. -- You may reply to this email to add a comment. You are receiving this mail because: You are watching the assignee of the bug.
