xfs: btree lookup shouldn't ASSERT on empty btree nodes

If a btree lookup encounters an empty btree node or an empty btree leaf
on a multi-level btree, that's evidence of a corrupt on-disk btree.
Therefore, we should return -EFSCORRUPTED to the upper levels, not an
ASSERT failure.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
index c825c81..3d59eb6 100644
--- a/fs/xfs/libxfs/xfs_btree.c
+++ b/fs/xfs/libxfs/xfs_btree.c
@@ -1896,7 +1896,12 @@ xfs_btree_lookup(
 			high = xfs_btree_get_numrecs(block);
 			if (!high) {
 				/* Block is empty, must be an empty leaf. */
-				ASSERT(level == 0 && cur->bc_nlevels == 1);
+				if (level != 0 || cur->bc_nlevels != 1) {
+					XFS_CORRUPTION_ERROR(__func__,
+							XFS_ERRLEVEL_LOW,
+							cur->bc_mp, block);
+					return -EFSCORRUPTED;
+				}
 
 				cur->bc_ptrs[0] = dir != XFS_LOOKUP_LE;
 				*stat = 0;