syzbot suspects this issue was fixed by commit:
commit a0e6a017ab56936c0405fe914a793b241ed25ee0
Author: Maíra Canal
Date: Tue May 23 12:32:08 2023 +
drm/vkms: Fix race-condition between the hrtimer and the atomic commit
bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=166c
Overview
Introduce a new way to select the data section first when selecting a
victim in foreground GC. This victim selection method works when the
prefer_data_victim mount option is enabled. If foreground GC migrates only
data sections and runs out of free sections, it cleans dirty node
For some still unknown reason, syzbot's bisections of fs bugs
sometimes end up in drm. There've been quite a few such cases
already..
Please ignore this bot's message.
On Thu, Oct 26, 2023 at 8:59 AM syzbot
wrote:
>
> syzbot suspects this issue was fixed by commit:
>
> commit a0e6a017ab56936c04
Jaegeuk, Chao, any comment on this? It really looks like a filesystem
corruption issue in f2fs when whiteouts are used...
Honza
On Tue 17-10-23 06:50:40, Al Viro wrote:
> [f2fs folks Cc'd]
>
> There's something very odd in f2f
On 10/26, Jan Kara wrote:
> Jaegeuk, Chao, any comment on this? It really looks like a filesystem
> corruption issue in f2fs when whiteouts are used...
Thanks Al and Jan for headsup.
Let us take a look as soon as possible.
>
> Honza
>
From: Daeho Jeong
We need to make sure to finish all the zones except six open zones. In
a case of that the previous mount wasn't successfully unmounted, we have
to change all the current segments.
Signed-off-by: Daeho Jeong
---
v2: skip write pointer check after fixing it.
---
fsck/fsck.c
From: Daeho Jeong
When we detect quota inode corruption, we better deallocate the current
space and allocate new ones for a clean start.
Signed-off-by: Daeho Jeong
---
v2: change node count check when creating a new node block
---
fsck/fsck.c| 148 +++--
