Re: [f2fs-dev] [PATCH] f2fs: ensure node page reads complete before f2fs_put_super() finishes

[Jaegeuk Kim <jaegeuk@kernel.org>]

On 10/09, Chao Yu wrote:
> On 10/6/2025 4:46 PM, Jan Prusakowski via Linux-f2fs-devel wrote:
> > Xfstests generic/335, generic/336 sometimes crash with the following message:
> > 
> > F2FS-fs (dm-0): detect filesystem reference count leak during umount, type: 9, count: 1
> > ------------[ cut here ]------------
> > kernel BUG at fs/f2fs/super.c:1939!
> > Oops: invalid opcode: 0000 [#1] SMP NOPTI
> > CPU: 1 UID: 0 PID: 609351 Comm: umount Tainted: G        W           6.17.0-rc5-xfstests-g9dd1835ecda5 #1 PREEMPT(none)
> > Tainted: [W]=WARN
> > Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
> > RIP: 0010:f2fs_put_super+0x3b3/0x3c0
> > Call Trace:
> >   <TASK>
> >   generic_shutdown_super+0x7e/0x190
> >   kill_block_super+0x1a/0x40
> >   kill_f2fs_super+0x9d/0x190
> >   deactivate_locked_super+0x30/0xb0
> >   cleanup_mnt+0xba/0x150
> >   task_work_run+0x5c/0xa0
> >   exit_to_user_mode_loop+0xb7/0xc0
> >   do_syscall_64+0x1ae/0x1c0
> >   entry_SYSCALL_64_after_hwframe+0x76/0x7e
> >   </TASK>
> > ---[ end trace 0000000000000000 ]---
> > 
> > It appears that sometimes it is possible that f2fs_put_super() is called before
> > all node page reads are completed.
> > Adding a call to f2fs_wait_on_all_pages() for F2FS_RD_NODE fixes the problem.
> > 
> > Fixes: bf22c3cc8ce7 ("f2fs: fix the panic in do_checkpoint()")
> > 
> > Signed-off-by: Jan Prusakowski <jprusakowski@google.com>
> > ---
> >   fs/f2fs/super.c | 1 +
> >   1 file changed, 1 insertion(+)
> > 
> > diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
> > index 1e0678e37a30..5c94bc42b8a1 100644
> > --- a/fs/f2fs/super.c
> > +++ b/fs/f2fs/super.c
> > @@ -1976,6 +1976,7 @@ static void f2fs_put_super(struct super_block *sb)
> >   	f2fs_flush_merged_writes(sbi);
> >   	f2fs_wait_on_all_pages(sbi, F2FS_WB_CP_DATA);
> > +	f2fs_wait_on_all_pages(sbi, F2FS_RD_NODE);
> 
> Jan,
> 
> At this stage, GC and checkpoint are both stopped, why there is still read
> IOs on node page? Who is reading node page? Can you please dig more details
> for this issue?

We don't actually wait for completing read IOs. So, I think it doesn't matter
the threads are stopped?

> 
> Thanks,
> 
> >   	if (err || f2fs_cp_error(sbi)) {
> >   		truncate_inode_pages_final(NODE_MAPPING(sbi));