From: Darrick J. Wong <redacted>

Since the sb write verifier trips on bad icounts, we should also force a
mount time recalculation of the summary counters if the icount is bad.
This helps us avoid blowing up at freeze/unmount time when the bad
counter gets written back out.

Signed-off-by: Darrick J. Wong <redacted>
---
  fs/xfs/xfs_mount.c |    1 +
  1 file changed, 1 insertion(+)

diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index 99db27d6ac8a..02d15098dbee 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c

@@ -637,6 +637,7 @@ xfs_check_summary_counts(
  	 */
  	if (XFS_LAST_UNMOUNT_WAS_CLEAN(mp) &&
  	    (mp->m_sb.sb_fdblocks > mp->m_sb.sb_dblocks ||
+	     !xfs_verify_icount(mp, mp->m_sb.sb_icount) ||
  	     mp->m_sb.sb_ifree > mp->m_sb.sb_icount))
  		mp->m_flags |= XFS_MOUNT_BAD_SUMMARY;