[2/4] xfs: repair the rmapbt

Commit Message

Darrick J. Wong Dec. 31, 2023, 10:17 p.m. UTC

From: Darrick J. Wong <[e-mail]>

Rebuild the reverse mapping btree from all primary metadata.  This first
patch establishes the bare mechanics of finding records and putting
together a new ondisk tree; more complex pieces are needed to make it
work properly.

Link: https://docs.kernel.org/filesystems/xfs-online-fsck-design.html#case-study-rebuilding-reverse-mapping-records
Link: https://docs.kernel.org/filesystems/xfs-online-fsck-design.html#case-study-reaping-after-repairing-reverse-mapping-btrees
Signed-off-by: Darrick J. Wong <[e-mail]>
---
 libxfs/xfs_bmap.c       |   43 +++++++++++++++++++++++++++++++++++++++++++
 libxfs/xfs_bmap.h       |    8 ++++++++
 libxfs/xfs_rmap.c       |   12 ++++++------
 libxfs/xfs_rmap.h       |    2 +-
 libxfs/xfs_rmap_btree.c |   13 ++++++++++++-
 5 files changed, 70 insertions(+), 8 deletions(-)

Patch

diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index b764b7f79c4..cfc4350d18e 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -6356,3 +6356,46 @@  xfs_bunmapi_range(
 out:
 	return error;
 }
+
+struct xfs_bmap_query_range {
+	xfs_bmap_query_range_fn	fn;
+	void			*priv;
+};
+
+/* Format btree record and pass to our callback. */
+STATIC int
+xfs_bmap_query_range_helper(
+	struct xfs_btree_cur		*cur,
+	const union xfs_btree_rec	*rec,
+	void				*priv)
+{
+	struct xfs_bmap_query_range	*query = priv;
+	struct xfs_bmbt_irec		irec;
+	xfs_failaddr_t			fa;
+
+	xfs_bmbt_disk_get_all(&rec->bmbt, &irec);
+	fa = xfs_bmap_validate_extent(cur->bc_ino.ip, cur->bc_ino.whichfork,
+			&irec);
+	if (fa) {
+		xfs_btree_mark_sick(cur);
+		return xfs_bmap_complain_bad_rec(cur->bc_ino.ip,
+				cur->bc_ino.whichfork, fa, &irec);
+	}
+
+	return query->fn(cur, &irec, query->priv);
+}
+
+/* Find all bmaps. */
+int
+xfs_bmap_query_all(
+	struct xfs_btree_cur		*cur,
+	xfs_bmap_query_range_fn		fn,
+	void				*priv)
+{
+	struct xfs_bmap_query_range	query = {
+		.priv			= priv,
+		.fn			= fn,
+	};
+
+	return xfs_btree_query_all(cur, xfs_bmap_query_range_helper, &query);
+}
diff --git a/libxfs/xfs_bmap.h b/libxfs/xfs_bmap.h
index 4b83f6148e0..9dd631bc2dc 100644
--- a/libxfs/xfs_bmap.h
+++ b/libxfs/xfs_bmap.h
@@ -278,4 +278,12 @@  extern struct kmem_cache	*xfs_bmap_intent_cache;
 int __init xfs_bmap_intent_init_cache(void);
 void xfs_bmap_intent_destroy_cache(void);
 
+typedef int (*xfs_bmap_query_range_fn)(
+	struct xfs_btree_cur	*cur,
+	struct xfs_bmbt_irec	*rec,
+	void			*priv);
+
+int xfs_bmap_query_all(struct xfs_btree_cur *cur, xfs_bmap_query_range_fn fn,
+		void *priv);
+
 #endif	/* __XFS_BMAP_H__ */
diff --git a/libxfs/xfs_rmap.c b/libxfs/xfs_rmap.c
index 0b462d17838..cec1c4e6efe 100644
--- a/libxfs/xfs_rmap.c
+++ b/libxfs/xfs_rmap.c
@@ -214,10 +214,10 @@  xfs_rmap_btrec_to_irec(
 /* Simple checks for rmap records. */
 xfs_failaddr_t
 xfs_rmap_check_irec(
-	struct xfs_btree_cur		*cur,
+	struct xfs_perag		*pag,
 	const struct xfs_rmap_irec	*irec)
 {
-	struct xfs_mount		*mp = cur->bc_mp;
+	struct xfs_mount		*mp = pag->pag_mount;
 	bool				is_inode;
 	bool				is_unwritten;
 	bool				is_bmbt;
@@ -232,8 +232,8 @@  xfs_rmap_check_irec(
 			return __this_address;
 	} else {
 		/* check for valid extent range, including overflow */
-		if (!xfs_verify_agbext(cur->bc_ag.pag, irec->rm_startblock,
-						       irec->rm_blockcount))
+		if (!xfs_verify_agbext(pag, irec->rm_startblock,
+					    irec->rm_blockcount))
 			return __this_address;
 	}
 
@@ -306,7 +306,7 @@  xfs_rmap_get_rec(
 
 	fa = xfs_rmap_btrec_to_irec(rec, irec);
 	if (!fa)
-		fa = xfs_rmap_check_irec(cur, irec);
+		fa = xfs_rmap_check_irec(cur->bc_ag.pag, irec);
 	if (fa)
 		return xfs_rmap_complain_bad_rec(cur, fa, irec);
 
@@ -2441,7 +2441,7 @@  xfs_rmap_query_range_helper(
 
 	fa = xfs_rmap_btrec_to_irec(rec, &irec);
 	if (!fa)
-		fa = xfs_rmap_check_irec(cur, &irec);
+		fa = xfs_rmap_check_irec(cur->bc_ag.pag, &irec);
 	if (fa)
 		return xfs_rmap_complain_bad_rec(cur, fa, &irec);
 
diff --git a/libxfs/xfs_rmap.h b/libxfs/xfs_rmap.h
index 3c98d9d50af..58c67896d12 100644
--- a/libxfs/xfs_rmap.h
+++ b/libxfs/xfs_rmap.h
@@ -195,7 +195,7 @@  int xfs_rmap_compare(const struct xfs_rmap_irec *a,
 union xfs_btree_rec;
 xfs_failaddr_t xfs_rmap_btrec_to_irec(const union xfs_btree_rec *rec,
 		struct xfs_rmap_irec *irec);
-xfs_failaddr_t xfs_rmap_check_irec(struct xfs_btree_cur *cur,
+xfs_failaddr_t xfs_rmap_check_irec(struct xfs_perag *pag,
 		const struct xfs_rmap_irec *irec);
 
 int xfs_rmap_has_records(struct xfs_btree_cur *cur, xfs_agblock_t bno,
diff --git a/libxfs/xfs_rmap_btree.c b/libxfs/xfs_rmap_btree.c
index e894a22e087..6924f7e49d9 100644
--- a/libxfs/xfs_rmap_btree.c
+++ b/libxfs/xfs_rmap_btree.c
@@ -340,7 +340,18 @@  xfs_rmapbt_verify(
 
 	level = be16_to_cpu(block->bb_level);
 	if (pag && xfs_perag_initialised_agf(pag)) {
-		if (level >= pag->pagf_levels[XFS_BTNUM_RMAPi])
+		unsigned int	maxlevel = pag->pagf_levels[XFS_BTNUM_RMAPi];
+
+#ifdef CONFIG_XFS_ONLINE_REPAIR
+		/*
+		 * Online repair could be rewriting the free space btrees, so
+		 * we'll validate against the larger of either tree while this
+		 * is going on.
+		 */
+		maxlevel = max_t(unsigned int, maxlevel,
+				pag->pagf_repair_levels[XFS_BTNUM_RMAPi]);
+#endif
+		if (level >= maxlevel)
 			return __this_address;
 	} else if (level >= mp->m_rmap_maxlevels)
 		return __this_address;