xfs: scrub btree keys and records

Add to the btree scrubber the ability to check that the keys and
records are in the right order and actually call out to our record
iterator to do actual checking of the records.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>

fs/xfs/scrub/btree.c

@@ -92,6 +92,101 @@ xfs_scrub_btree_set_corrupt(
 				__return_address);
 }
 
+/*
+ * Make sure this record is in order and doesn't stray outside of the parent
+ * keys.
+ */
+STATIC void
+xfs_scrub_btree_rec(
+	struct xfs_scrub_btree	*bs)
+{
+	struct xfs_btree_cur	*cur = bs->cur;
+	union xfs_btree_rec	*rec;
+	union xfs_btree_key	key;
+	union xfs_btree_key	hkey;
+	union xfs_btree_key	*keyp;
+	struct xfs_btree_block	*block;
+	struct xfs_btree_block	*keyblock;
+	struct xfs_buf		*bp;
+
+	block = xfs_btree_get_block(cur, 0, &bp);
+	rec = xfs_btree_rec_addr(cur, cur->bc_ptrs[0], block);
+
+	trace_xfs_scrub_btree_rec(bs->sc, cur, 0);
+
+	/* If this isn't the first record, are they in order? */
+	if (!bs->firstrec && !cur->bc_ops->recs_inorder(cur, &bs->lastrec, rec))
+		xfs_scrub_btree_set_corrupt(bs->sc, cur, 0);
+	bs->firstrec = false;
+	memcpy(&bs->lastrec, rec, cur->bc_ops->rec_len);
+
+	if (cur->bc_nlevels == 1)
+		return;
+
+	/* Is this at least as large as the parent low key? */
+	cur->bc_ops->init_key_from_rec(&key, rec);
+	keyblock = xfs_btree_get_block(cur, 1, &bp);
+	keyp = xfs_btree_key_addr(cur, cur->bc_ptrs[1], keyblock);
+	if (cur->bc_ops->diff_two_keys(cur, &key, keyp) < 0)
+		xfs_scrub_btree_set_corrupt(bs->sc, cur, 1);
+
+	if (!(cur->bc_flags & XFS_BTREE_OVERLAPPING))
+		return;
+
+	/* Is this no larger than the parent high key? */
+	cur->bc_ops->init_high_key_from_rec(&hkey, rec);
+	keyp = xfs_btree_high_key_addr(cur, cur->bc_ptrs[1], keyblock);
+	if (cur->bc_ops->diff_two_keys(cur, keyp, &hkey) < 0)
+		xfs_scrub_btree_set_corrupt(bs->sc, cur, 1);
+}
+
+/*
+ * Make sure this key is in order and doesn't stray outside of the parent
+ * keys.
+ */
+STATIC void
+xfs_scrub_btree_key(
+	struct xfs_scrub_btree	*bs,
+	int			level)
+{
+	struct xfs_btree_cur	*cur = bs->cur;
+	union xfs_btree_key	*key;
+	union xfs_btree_key	*keyp;
+	struct xfs_btree_block	*block;
+	struct xfs_btree_block	*keyblock;
+	struct xfs_buf		*bp;
+
+	block = xfs_btree_get_block(cur, level, &bp);
+	key = xfs_btree_key_addr(cur, cur->bc_ptrs[level], block);
+
+	trace_xfs_scrub_btree_key(bs->sc, cur, level);
+
+	/* If this isn't the first key, are they in order? */
+	if (!bs->firstkey[level] &&
+	    !cur->bc_ops->keys_inorder(cur, &bs->lastkey[level], key))
+		xfs_scrub_btree_set_corrupt(bs->sc, cur, level);
+	bs->firstkey[level] = false;
+	memcpy(&bs->lastkey[level], key, cur->bc_ops->key_len);
+
+	if (level + 1 >= cur->bc_nlevels)
+		return;
+
+	/* Is this at least as large as the parent low key? */
+	keyblock = xfs_btree_get_block(cur, level + 1, &bp);
+	keyp = xfs_btree_key_addr(cur, cur->bc_ptrs[level + 1], keyblock);
+	if (cur->bc_ops->diff_two_keys(cur, key, keyp) < 0)
+		xfs_scrub_btree_set_corrupt(bs->sc, cur, level);
+
+	if (!(cur->bc_flags & XFS_BTREE_OVERLAPPING))
+		return;
+
+	/* Is this no larger than the parent high key? */
+	key = xfs_btree_high_key_addr(cur, cur->bc_ptrs[level], block);
+	keyp = xfs_btree_high_key_addr(cur, cur->bc_ptrs[level + 1], keyblock);
+	if (cur->bc_ops->diff_two_keys(cur, keyp, key) < 0)
+		xfs_scrub_btree_set_corrupt(bs->sc, cur, level);
+}
+
 /*
  * Check a btree pointer.  Returns true if it's ok to use this pointer.
  * Callers do not need to set the corrupt flag.
@@ -278,6 +373,7 @@ xfs_scrub_btree(
 	struct xfs_scrub_btree		bs = {0};
 	union xfs_btree_ptr		ptr;
 	union xfs_btree_ptr		*pp;
+	union xfs_btree_rec		*recp;
 	struct xfs_btree_block		*block;
 	int				level;
 	struct xfs_buf			*bp;
@@ -328,7 +424,16 @@ xfs_scrub_btree(
 				continue;
 			}
 
-			if (xfs_scrub_should_terminate(sc, &error))
+			/* Records in order for scrub? */
+			xfs_scrub_btree_rec(&bs);
+
+			/* Call out to the record checker. */
+			recp = xfs_btree_rec_addr(cur, cur->bc_ptrs[0], block);
+			error = bs.scrub_rec(&bs, recp);
+			if (error)
+				break;
+			if (xfs_scrub_should_terminate(sc, &error) ||
+			    (sc->sm->sm_flags & XFS_SCRUB_OFLAG_CORRUPT))
 				break;
 
 			cur->bc_ptrs[level]++;
@@ -343,6 +448,9 @@ xfs_scrub_btree(
 			continue;
 		}
 
+		/* Keys in order for scrub? */
+		xfs_scrub_btree_key(&bs, level);
+
 		/* Drill another level deeper. */
 		pp = xfs_btree_ptr_addr(cur, cur->bc_ptrs[level], block);
 		if (!xfs_scrub_btree_ptr_ok(&bs, level, pp)) {

fs/xfs/scrub/trace.h

@@ -446,6 +446,51 @@ TRACE_EVENT(xfs_scrub_ifork_btree_error,
 		  __entry->ret_ip)
 );
 
+DECLARE_EVENT_CLASS(xfs_scrub_sbtree_class,
+	TP_PROTO(struct xfs_scrub_context *sc, struct xfs_btree_cur *cur,
+		 int level),
+	TP_ARGS(sc, cur, level),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(int, type)
+		__field(xfs_btnum_t, btnum)
+		__field(xfs_agnumber_t, agno)
+		__field(xfs_agblock_t, bno)
+		__field(int, level)
+		__field(int, nlevels)
+		__field(int, ptr)
+	),
+	TP_fast_assign(
+		xfs_fsblock_t fsbno = xfs_scrub_btree_cur_fsbno(cur, level);
+
+		__entry->dev = sc->mp->m_super->s_dev;
+		__entry->type = sc->sm->sm_type;
+		__entry->btnum = cur->bc_btnum;
+		__entry->agno = XFS_FSB_TO_AGNO(cur->bc_mp, fsbno);
+		__entry->bno = XFS_FSB_TO_AGBNO(cur->bc_mp, fsbno);
+		__entry->level = level;
+		__entry->nlevels = cur->bc_nlevels;
+		__entry->ptr = cur->bc_ptrs[level];
+	),
+	TP_printk("dev %d:%d type %u btnum %d agno %u agbno %u level %d nlevels %d ptr %d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->type,
+		  __entry->btnum,
+		  __entry->agno,
+		  __entry->bno,
+		  __entry->level,
+		  __entry->nlevels,
+		  __entry->ptr)
+)
+#define DEFINE_SCRUB_SBTREE_EVENT(name) \
+DEFINE_EVENT(xfs_scrub_sbtree_class, name, \
+	TP_PROTO(struct xfs_scrub_context *sc, struct xfs_btree_cur *cur, \
+		 int level), \
+	TP_ARGS(sc, cur, level))
+
+DEFINE_SCRUB_SBTREE_EVENT(xfs_scrub_btree_rec);
+DEFINE_SCRUB_SBTREE_EVENT(xfs_scrub_btree_key);
+
 #endif /* _TRACE_XFS_SCRUB_TRACE_H */
 
 #undef TRACE_INCLUDE_PATH