[f2fs-dev] [Bug 216050] f2fs_gc occupies 100% cpu

[bugzilla-daemon@kernel.org]

https://bugzilla.kernel.org/show_bug.cgi?id=216050

plague@privacyrequired.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |plague@privacyrequired.com

--- Comment #3 from plague@privacyrequired.com ---
Same for me. Every morning my system is nonoperational due to f2fs_gc takes
100% CPU... 
Mine is on LVM (2xNVMe):

/dev/mapper/vg_ssd-lv_root on / type f2fs
(rw,noatime,lazytime,background_gc=on,nodiscard,no_heap,user_xattr,inline_xattr,acl,inline_data,inline_dentry,flush_merge,extent_cache,mode=adaptive,active_logs=6,alloc_mode=default,nocheckpoint_merge,fsync_mode=posix,discard_unit=block)

/dev/mapper/vg_ssd-lv_home on /home type f2fs
(rw,noatime,lazytime,background_gc=on,nodiscard,no_heap,user_xattr,inline_xattr,acl,inline_data,inline_dentry,flush_merge,extent_cache,mode=adaptive,active_logs=6,alloc_mode=default,nocheckpoint_merge,fsync_mode=posix,discard_unit=block)

/dev/mapper/vg_ssd2-lv_backup on /mnt/backup type f2fs
(rw,noatime,lazytime,background_gc=on,nodiscard,no_heap,user_xattr,inline_xattr,acl,inline_data,inline_dentry,flush_merge,extent_cache,mode=adaptive,active_logs=6,alloc_mode=default,checkpoint_merge,fsync_mode=posix,discard_unit=block)

-------
I cannot see any error in the journal or dmesg, except things timeout:

May 31 21:25:10 mefisto kernel: R13: 0000000000000001 R14: 0000000000000001
R15: 0000000000000015
May 31 21:25:10 mefisto kernel: R10: 0000000000000000 R11: 0000000000000293
R12: 0000000000000015
May 31 21:25:10 mefisto kernel: RBP: 000056041c75b640 R08: 0000000000000000
R09: 00007ffdc8d378c8
May 31 21:25:10 mefisto kernel: RDX: 0000000000000000 RSI: 0000000000000000
RDI: 0000000000000015
May 31 21:25:10 mefisto kernel: RAX: ffffffffffffffda RBX: 000056041c6ff430
RCX: 00007effb2a12a5a
May 31 21:25:10 mefisto kernel: RSP: 002b:00007ffdc8d37820 EFLAGS: 00000293
ORIG_RAX: 000000000000004a
May 31 21:25:10 mefisto kernel: RIP: 0033:0x7effb2a12a5a
May 31 21:25:10 mefisto kernel:  entry_SYSCALL_64_after_hwframe+0x44/0xae
May 31 21:25:10 mefisto kernel:  ? do_syscall_64+0x6b/0x90
May 31 21:25:10 mefisto kernel:  ? syscall_exit_to_user_mode+0x26/0x50
May 31 21:25:10 mefisto kernel:  ? exit_to_user_mode_prepare+0x111/0x140
May 31 21:25:10 mefisto kernel:  ? do_syscall_64+0x6b/0x90
May 31 21:25:10 mefisto kernel:  ? syscall_exit_to_user_mode+0x26/0x50
May 31 21:25:10 mefisto kernel:  ? do_syscall_64+0x6b/0x90
May 31 21:25:10 mefisto kernel:  ? syscall_exit_to_user_mode+0x26/0x50
May 31 21:25:10 mefisto kernel:  do_syscall_64+0x5c/0x90
May 31 21:25:10 mefisto kernel:  __x64_sys_fsync+0x3b/0x60
May 31 21:25:10 mefisto kernel:  f2fs_do_sync_file+0x18b/0x920 [f2fs
b9d199b68c91f6957b9756d57ec16e8bad3561b9]
May 31 21:25:10 mefisto kernel:  f2fs_issue_checkpoint+0x143/0x200 [f2fs
b9d199b68c91f6957b9756d57ec16e8bad3561b9]
May 31 21:25:10 mefisto kernel:  ? terminate_walk+0x61/0xf0
May 31 21:25:10 mefisto kernel:  rwsem_down_write_slowpath+0x1eb/0x5c0
May 31 21:25:10 mefisto kernel:  schedule+0x4f/0xb0
May 31 21:25:10 mefisto kernel:  __schedule+0x37c/0x11f0
May 31 21:25:10 mefisto kernel:  <TASK>
May 31 21:25:10 mefisto kernel: Call Trace:
May 31 21:25:10 mefisto kernel: task:NetworkManager  state:D stack:    0 pid:
3380 ppid:     1 flags:0x00004002
May 31 21:25:10 mefisto kernel: "echo 0 >
/proc/sys/kernel/hung_task_timeout_secs" disables this message.
May 31 21:25:10 mefisto kernel:       Tainted: G     U     OE    
5.18.0-arch1-1 #1
May 31 21:25:10 mefisto kernel: INFO: task NetworkManager:3380 blocked for more
than 368 seconds.
May 31 21:23:07 mefisto kernel:  </TASK>

-- 
You may reply to this email to add a comment.

You are receiving this mail because:
You are watching the assignee of the bug.

_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel