Re: [PATCH 1/2] generic: add a test for device removal with dirty data

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]



On Sun, Aug 06, 2023 at 10:11:45PM +0800, Zorro Lang wrote:
> On Mon, Jul 24, 2023 at 08:29:26AM -0700, Christoph Hellwig wrote:
> > Test the removal of the underlying device when the file system still
> > has dirty data.
> > 
> > Signed-off-by: Christoph Hellwig <hch@xxxxxx>
> > ---
> 
> After I merge this case, it always hit a kernel warning about a "possible
> circular locking dependency detected", no matter on ext4 [1] or xfs. This
> test isn't hang actually. But it always fail this test (got a .dmesg file).
> Any ideas about that?

That's because xfs and ext4 do the blkdev_put of the non-main devices
in ->put_super and thus inside s_umount.  It's actually an old issue,
we now just trigger it reliably.  We'll need to move the dropping into
->kill_sb after dropping ->s_umount.  This has been on my plate for
a while.



[Index of Archives]     [Linux Filesystems Development]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux