xfs: fix perag reference leak on iteration race with growfs

The for_each_perag*() set of macros are hacky in that some (i.e.
those based on sb_agcount) rely on the assumption that perag
iteration terminates naturally with a NULL perag at the specified
end_agno. Others allow for the final AG to have a valid perag and
require the calling function to clean up any potential leftover
xfs_perag reference on termination of the loop.

Aside from providing a subtly inconsistent interface, the former
variant is racy with growfs because growfs can create discoverable
post-eofs perags before the final superblock update that completes
the grow operation and increases sb_agcount. This leads to the
following assert failure (reproduced by xfs/104) in the perag free
path during unmount:

 XFS: Assertion failed: atomic_read(&pag->pag_ref) == 0, file: fs/xfs/libxfs/xfs_ag.c, line: 195

This occurs because one of the many for_each_perag() loops in the
code that is expected to terminate with a NULL pag (and thus has no
post-loop xfs_perag_put() check) raced with a growfs and found a
non-NULL post-EOFS perag, but terminated naturally based on the
end_agno check without releasing the post-EOFS perag.

Rework the iteration logic to lift the agno check from the main for
loop conditional to the iteration helper function. The for loop now
purely terminates on a NULL pag and xfs_perag_next() avoids taking a
reference to any perag beyond end_agno in the first place.

Fixes: f250eedcf762 ("xfs: make for_each_perag... a first class citizen")
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>

diff --git a/fs/xfs/libxfs/xfs_ag.h b/fs/xfs/libxfs/xfs_ag.h
index 4585ebb..3f597ca 100644 (file)
--- a/fs/xfs/libxfs/xfs_ag.h
+++ b/fs/xfs/libxfs/xfs_ag.h
@@ -116,30 +116,26 @@ void xfs_perag_put(struct xfs_perag *pag);
 
 /*
  * Perag iteration APIs
- *
- * XXX: for_each_perag_range() usage really needs an iterator to clean up when
- * we terminate at end_agno because we may have taken a reference to the perag
- * beyond end_agno. Right now callers have to be careful to catch and clean that
- * up themselves. This is not necessary for the callers of for_each_perag() and
- * for_each_perag_from() because they terminate at sb_agcount where there are
- * no perag structures in tree beyond end_agno.
  */
 static inline struct xfs_perag *
 xfs_perag_next(
        struct xfs_perag        *pag,
-       xfs_agnumber_t          *agno)
+       xfs_agnumber_t          *agno,
+       xfs_agnumber_t          end_agno)
 {
        struct xfs_mount        *mp = pag->pag_mount;
 
        *agno = pag->pag_agno + 1;
        xfs_perag_put(pag);
+       if (*agno > end_agno)
+               return NULL;
        return xfs_perag_get(mp, *agno);
 }
 
 #define for_each_perag_range(mp, agno, end_agno, pag) \
        for ((pag) = xfs_perag_get((mp), (agno)); \
-               (pag) != NULL && (agno) <= (end_agno); \
-               (pag) = xfs_perag_next((pag), &(agno)))
+               (pag) != NULL; \
+               (pag) = xfs_perag_next((pag), &(agno), (end_agno)))
 
 #define for_each_perag_from(mp, agno, pag) \
        for_each_perag_range((mp), (agno), (mp)->m_sb.sb_agcount - 1, (pag))