gfs2: Recursive gfs2_quota_hold in gfs2_iomap_end

commit 7009fa9cd9a5262944b30eb7efb1f0561d074b68 upstream.

When starting an iomap write, gfs2_quota_lock_check -> gfs2_quota_lock
-> gfs2_quota_hold is called from gfs2_iomap_begin.  At the end of the
write, before unlocking the quotas, punch_hole -> gfs2_quota_hold can be
called again in gfs2_iomap_end, which is incorrect and leads to a failed
assertion.  Instead, move the call to gfs2_quota_unlock before the call
to punch_hole to fix that.

Fixes: 64bc06bb32 ("gfs2: iomap buffered write support")
Cc: stable@vger.kernel.org # v4.19+
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This commit is contained in:
Andreas Gruenbacher 2021-02-09 18:32:32 +01:00 committed by Greg Kroah-Hartman
parent fa0201d390
commit fd3b7e07d3
1 changed files with 3 additions and 3 deletions

View File

@ -1228,6 +1228,9 @@ static int gfs2_iomap_end(struct inode *inode, loff_t pos, loff_t length,
gfs2_inplace_release(ip);
if (ip->i_qadata && ip->i_qadata->qa_qd_num)
gfs2_quota_unlock(ip);
if (length != written && (iomap->flags & IOMAP_F_NEW)) {
/* Deallocate blocks that were just allocated. */
loff_t blockmask = i_blocksize(inode) - 1;
@ -1240,9 +1243,6 @@ static int gfs2_iomap_end(struct inode *inode, loff_t pos, loff_t length,
}
}
if (ip->i_qadata && ip->i_qadata->qa_qd_num)
gfs2_quota_unlock(ip);
if (unlikely(!written))
goto out_unlock;