Merge tag 'xfs-4.15-merge-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux

Pull xfs updates from Darrick Wong:
 "xfs: great scads of new stuff for 4.15.

  This merge cycle, we're making some substantive changes to XFS. The
  in-core extent mappings have been refactored to use proper iterators
  and a btree to handle heavily fragmented files without needing
  high-order memory allocations; some important log recovery bug fixes;
  and the first part of the online fsck functionality.

  (The online fsck feature is disabled by default and more pieces of it
  will be coming in future release cycles.)

  This giant pile of patches has been run through a full xfstests run
  over the weekend and through a quick xfstests run against this
  morning's master, with no major failures reported.

  New in this version:

   - Refactor the incore extent map manipulations to use a cursor
     instead of directly modifying extent data.

   - Refactor the incore extent map cursor to use an in-memory btree
     instead of a single high-order allocation. This eliminates a major
     source of complaints about insufficient memory when opening a
     heavily fragmented file into a system whose memory is also heavily
     fragmented.

   - Fix a longstanding bug where deleting a file with a complex
     extended attribute btree incorrectly handled memory pointers, which
     could lead to memory corruption.

   - Improve metadata validation to eliminate crashing problems found
     while fuzzing xfs.

   - Move the error injection tag definitions into libxfs to be shared
     with userspace components.

   - Fix some log recovery bugs where we'd underflow log block position
     vector and incorrectly fail log recovery.

   - Drain the buffer lru after log recovery to force recovered buffers
     back through the verifiers after mount. On a v4 filesystem the log
     never attaches verifiers during log replay (v5 does), so we could
     end up with buffers marked verified but without having ever been
     verified.

   - Fix various other bugs.

   - Introduce the first part of a new online fsck tool. The new fsck
     tool will be able to iterate every piece of metadata in the
     filesystem to look for obvious errors and corruptions. In the next
     release cycle the checking will be extended to cross-reference with
     the other fs metadata, so this feature should only be used by the
     developers in the mean time"

* tag 'xfs-4.15-merge-1' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux: (131 commits)
  xfs: on failed mount, force-reclaim inodes after unmounting quota controls
  xfs: check the uniqueness of the AGFL entries
  xfs: remove u_int* type usage
  xfs: handle zero entries case in xfs_iext_rebalance_leaf
  xfs: add comments documenting the rebalance algorithm
  xfs: trivial indentation fixup for xfs_iext_remove_node
  xfs: remove a superflous assignment in xfs_iext_remove_node
  xfs: add some comments to xfs_iext_insert/xfs_iext_insert_node
  xfs: fix number of records handling in xfs_iext_split_leaf
  fs/xfs: Remove NULL check before kmem_cache_destroy
  xfs: only check da node header padding on v5 filesystems
  xfs: fix btree scrub deref check
  xfs: fix uninitialized return values in scrub code
  xfs: pass inode number to xfs_scrub_ino_set_{preen,warning}
  xfs: refactor the directory data block bestfree checks
  xfs: mark xlog_verify_dest_ptr STATIC
  xfs: mark xlog_recover_check_summary STATIC
  xfs: mark xfs_btree_check_lblock and xfs_btree_check_ptr static
  xfs: remove unreachable error injection code in xfs_qm_dqget
  xfs: remove unused debug counts for xfs_lock_inodes
  ...

fs/xfs/Kconfig

@@ -71,6 +71,23 @@ config XFS_RT
 
 	  If unsure, say N.
 
+config XFS_ONLINE_SCRUB
+	bool "XFS online metadata check support"
+	default n
+	depends on XFS_FS
+	help
+	  If you say Y here you will be able to check metadata on a
+	  mounted XFS filesystem.  This feature is intended to reduce
+	  filesystem downtime by supplementing xfs_repair.  The key
+	  advantage here is to look for problems proactively so that
+	  they can be dealt with in a controlled manner.
+
+	  This feature is considered EXPERIMENTAL.  Use with caution!
+
+	  See the xfs_scrub man page in section 8 for additional information.
+
+	  If unsure, say N.
+
 config XFS_WARN
 	bool "XFS Verbose Warnings"
 	depends on XFS_FS && !XFS_DEBUG

fs/xfs/Makefile

@@ -49,6 +49,7 @@ xfs-y				+= $(addprefix libxfs/, \
 				   xfs_dquot_buf.o \
 				   xfs_ialloc.o \
 				   xfs_ialloc_btree.o \
+				   xfs_iext_tree.o \
 				   xfs_inode_fork.o \
 				   xfs_inode_buf.o \
 				   xfs_log_rlimit.o \
@@ -135,3 +136,31 @@ xfs-$(CONFIG_XFS_POSIX_ACL)	+= xfs_acl.o
 xfs-$(CONFIG_SYSCTL)		+= xfs_sysctl.o
 xfs-$(CONFIG_COMPAT)		+= xfs_ioctl32.o
 xfs-$(CONFIG_EXPORTFS_BLOCK_OPS)	+= xfs_pnfs.o
+
+# online scrub/repair
+ifeq ($(CONFIG_XFS_ONLINE_SCRUB),y)
+
+# Tracepoints like to blow up, so build that before everything else
+
+xfs-y				+= $(addprefix scrub/, \
+				   trace.o \
+				   agheader.o \
+				   alloc.o \
+				   attr.o \
+				   bmap.o \
+				   btree.o \
+				   common.o \
+				   dabtree.o \
+				   dir.o \
+				   ialloc.o \
+				   inode.o \
+				   parent.o \
+				   refcount.o \
+				   rmap.o \
+				   scrub.o \
+				   symlink.o \
+				   )
+
+xfs-$(CONFIG_XFS_RT)		+= scrub/rtbitmap.o
+xfs-$(CONFIG_XFS_QUOTA)		+= scrub/quota.o
+endif

fs/xfs/kmem.h

@@ -119,7 +119,6 @@ kmem_zone_free(kmem_zone_t *zone, void *ptr)
 static inline void
 kmem_zone_destroy(kmem_zone_t *zone)
 {
-	if (zone)
 	kmem_cache_destroy(zone);
 }
 

fs/xfs/libxfs/xfs_ag_resv.c

@@ -27,6 +27,7 @@
 #include "xfs_mount.h"
 #include "xfs_defer.h"
 #include "xfs_alloc.h"
+#include "xfs_errortag.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
 #include "xfs_cksum.h"

fs/xfs/libxfs/xfs_alloc.c

@@ -31,6 +31,7 @@
 #include "xfs_alloc_btree.h"
 #include "xfs_alloc.h"
 #include "xfs_extent_busy.h"
+#include "xfs_errortag.h"
 #include "xfs_error.h"
 #include "xfs_cksum.h"
 #include "xfs_trace.h"
@@ -2931,3 +2932,52 @@ xfs_alloc_query_all(
 	query.fn = fn;
 	return xfs_btree_query_all(cur, xfs_alloc_query_range_helper, &query);
 }
+
+/* Find the size of the AG, in blocks. */
+xfs_agblock_t
+xfs_ag_block_count(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		agno)
+{
+	ASSERT(agno < mp->m_sb.sb_agcount);
+
+	if (agno < mp->m_sb.sb_agcount - 1)
+		return mp->m_sb.sb_agblocks;
+	return mp->m_sb.sb_dblocks - (agno * mp->m_sb.sb_agblocks);
+}
+
+/*
+ * Verify that an AG block number pointer neither points outside the AG
+ * nor points at static metadata.
+ */
+bool
+xfs_verify_agbno(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		agno,
+	xfs_agblock_t		agbno)
+{
+	xfs_agblock_t		eoag;
+
+	eoag = xfs_ag_block_count(mp, agno);
+	if (agbno >= eoag)
+		return false;
+	if (agbno <= XFS_AGFL_BLOCK(mp))
+		return false;
+	return true;
+}
+
+/*
+ * Verify that an FS block number pointer neither points outside the
+ * filesystem nor points at static AG metadata.
+ */
+bool
+xfs_verify_fsbno(
+	struct xfs_mount	*mp,
+	xfs_fsblock_t		fsbno)
+{
+	xfs_agnumber_t		agno = XFS_FSB_TO_AGNO(mp, fsbno);
+
+	if (agno >= mp->m_sb.sb_agcount)
+		return false;
+	return xfs_verify_agbno(mp, agno, XFS_FSB_TO_AGBNO(mp, fsbno));
+}

fs/xfs/libxfs/xfs_alloc.h

@@ -232,5 +232,9 @@ int xfs_alloc_query_range(struct xfs_btree_cur *cur,
 		xfs_alloc_query_range_fn fn, void *priv);
 int xfs_alloc_query_all(struct xfs_btree_cur *cur, xfs_alloc_query_range_fn fn,
 		void *priv);
+xfs_agblock_t xfs_ag_block_count(struct xfs_mount *mp, xfs_agnumber_t agno);
+bool xfs_verify_agbno(struct xfs_mount *mp, xfs_agnumber_t agno,
+		xfs_agblock_t agbno);
+bool xfs_verify_fsbno(struct xfs_mount *mp, xfs_fsblock_t fsbno);
 
 #endif	/* __XFS_ALLOC_H__ */

fs/xfs/libxfs/xfs_attr_leaf.c

@@ -397,13 +397,9 @@ xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes)
 	/* rounded down */
 	offset = (XFS_LITINO(mp, dp->i_d.di_version) - bytes) >> 3;
 
-	switch (dp->i_d.di_format) {
-	case XFS_DINODE_FMT_DEV:
+	if (dp->i_d.di_format == XFS_DINODE_FMT_DEV) {
 		minforkoff = roundup(sizeof(xfs_dev_t), 8) >> 3;
 		return (offset >= minforkoff) ? minforkoff : 0;
-	case XFS_DINODE_FMT_UUID:
-		minforkoff = roundup(sizeof(uuid_t), 8) >> 3;
-		return (offset >= minforkoff) ? minforkoff : 0;
 	}
 
 	/*

fs/xfs/libxfs/xfs_bmap.h

@@ -43,7 +43,7 @@ struct xfs_bmalloca {
 	xfs_fsblock_t		blkno;	/* starting block of new extent */
 
 	struct xfs_btree_cur	*cur;	/* btree cursor */
-	xfs_extnum_t		idx;	/* current extent index */
+	struct xfs_iext_cursor	icur;	/* incore extent cursor */
 	int			nallocs;/* number of extents alloc'd */
 	int			logflags;/* flags for transaction logging */
 
@@ -113,6 +113,9 @@ struct xfs_extent_free_item
 /* Only convert delalloc space, don't allocate entirely new extents */
 #define XFS_BMAPI_DELALLOC	0x400
 
+/* Only convert unwritten extents, don't allocate new blocks */
+#define XFS_BMAPI_CONVERT_ONLY	0x800
+
 #define XFS_BMAPI_FLAGS \
 	{ XFS_BMAPI_ENTIRE,	"ENTIRE" }, \
 	{ XFS_BMAPI_METADATA,	"METADATA" }, \
@@ -124,7 +127,8 @@ struct xfs_extent_free_item
 	{ XFS_BMAPI_ZERO,	"ZERO" }, \
 	{ XFS_BMAPI_REMAP,	"REMAP" }, \
 	{ XFS_BMAPI_COWFORK,	"COWFORK" }, \
-	{ XFS_BMAPI_DELALLOC,	"DELALLOC" }
+	{ XFS_BMAPI_DELALLOC,	"DELALLOC" }, \
+	{ XFS_BMAPI_CONVERT_ONLY, "CONVERT_ONLY" }
 
 
 static inline int xfs_bmapi_aflag(int w)
@@ -183,29 +187,6 @@ static inline bool xfs_bmap_is_real_extent(struct xfs_bmbt_irec *irec)
 		!isnullstartblock(irec->br_startblock);
 }
 
-/*
- * This macro is used to determine how many extents will be shifted
- * in one write transaction. We could require two splits,
- * an extent move on the first and an extent merge on the second,
- * So it is proper that one extent is shifted inside write transaction
- * at a time.
- */
-#define XFS_BMAP_MAX_SHIFT_EXTENTS	1
-
-enum shift_direction {
-	SHIFT_LEFT = 0,
-	SHIFT_RIGHT,
-};
-
-#ifdef DEBUG
-void	xfs_bmap_trace_exlist(struct xfs_inode *ip, xfs_extnum_t cnt,
-		int whichfork, unsigned long caller_ip);
-#define	XFS_BMAP_TRACE_EXLIST(ip,c,w)	\
-	xfs_bmap_trace_exlist(ip,c,w, _THIS_IP_)
-#else
-#define	XFS_BMAP_TRACE_EXLIST(ip,c,w)
-#endif
-
 void	xfs_trim_extent(struct xfs_bmbt_irec *irec, xfs_fileoff_t bno,
 		xfs_filblks_t len);
 void	xfs_trim_extent_eof(struct xfs_bmbt_irec *, struct xfs_inode *);
@@ -222,8 +203,6 @@ int	xfs_bmap_last_before(struct xfs_trans *tp, struct xfs_inode *ip,
 int	xfs_bmap_last_offset(struct xfs_inode *ip, xfs_fileoff_t *unused,
 		int whichfork);
 int	xfs_bmap_one_block(struct xfs_inode *ip, int whichfork);
-int	xfs_bmap_read_extents(struct xfs_trans *tp, struct xfs_inode *ip,
-		int whichfork);
 int	xfs_bmapi_read(struct xfs_inode *ip, xfs_fileoff_t bno,
 		xfs_filblks_t len, struct xfs_bmbt_irec *mval,
 		int *nmap, int flags);
@@ -241,20 +220,25 @@ int	xfs_bunmapi(struct xfs_trans *tp, struct xfs_inode *ip,
 		xfs_extnum_t nexts, xfs_fsblock_t *firstblock,
 		struct xfs_defer_ops *dfops, int *done);
 int	xfs_bmap_del_extent_delay(struct xfs_inode *ip, int whichfork,
-		xfs_extnum_t *idx, struct xfs_bmbt_irec *got,
+		struct xfs_iext_cursor *cur, struct xfs_bmbt_irec *got,
+		struct xfs_bmbt_irec *del);
+void	xfs_bmap_del_extent_cow(struct xfs_inode *ip,
+		struct xfs_iext_cursor *cur, struct xfs_bmbt_irec *got,
 		struct xfs_bmbt_irec *del);
-void	xfs_bmap_del_extent_cow(struct xfs_inode *ip, xfs_extnum_t *idx,
-		struct xfs_bmbt_irec *got, struct xfs_bmbt_irec *del);
 uint	xfs_default_attroffset(struct xfs_inode *ip);
-int	xfs_bmap_shift_extents(struct xfs_trans *tp, struct xfs_inode *ip,
+int	xfs_bmap_collapse_extents(struct xfs_trans *tp, struct xfs_inode *ip,
+		xfs_fileoff_t *next_fsb, xfs_fileoff_t offset_shift_fsb,
+		bool *done, xfs_fileoff_t stop_fsb, xfs_fsblock_t *firstblock,
+		struct xfs_defer_ops *dfops);
+int	xfs_bmap_insert_extents(struct xfs_trans *tp, struct xfs_inode *ip,
 		xfs_fileoff_t *next_fsb, xfs_fileoff_t offset_shift_fsb,
-		int *done, xfs_fileoff_t stop_fsb, xfs_fsblock_t *firstblock,
-		struct xfs_defer_ops *dfops, enum shift_direction direction,
-		int num_exts);
+		bool *done, xfs_fileoff_t stop_fsb, xfs_fsblock_t *firstblock,
+		struct xfs_defer_ops *dfops);
 int	xfs_bmap_split_extent(struct xfs_inode *ip, xfs_fileoff_t split_offset);
 int	xfs_bmapi_reserve_delalloc(struct xfs_inode *ip, int whichfork,
 		xfs_fileoff_t off, xfs_filblks_t len, xfs_filblks_t prealloc,
-		struct xfs_bmbt_irec *got, xfs_extnum_t *lastx, int eof);
+		struct xfs_bmbt_irec *got, struct xfs_iext_cursor *cur,
+		int eof);
 
 enum xfs_bmap_intent_type {
 	XFS_BMAP_MAP = 1,
@@ -278,4 +262,16 @@ int	xfs_bmap_map_extent(struct xfs_mount *mp, struct xfs_defer_ops *dfops,
 int	xfs_bmap_unmap_extent(struct xfs_mount *mp, struct xfs_defer_ops *dfops,
 		struct xfs_inode *ip, struct xfs_bmbt_irec *imap);
 
+static inline int xfs_bmap_fork_to_state(int whichfork)
+{
+	switch (whichfork) {
+	case XFS_ATTR_FORK:
+		return BMAP_ATTRFORK;
+	case XFS_COW_FORK:
+		return BMAP_COWFORK;
+	default:
+		return 0;
+	}
+}
+
 #endif	/* __XFS_BMAP_H__ */

fs/xfs/libxfs/xfs_bmap_btree.c

@@ -37,22 +37,6 @@
 #include "xfs_cksum.h"
 #include "xfs_rmap.h"
 
-/*
- * Determine the extent state.
- */
-/* ARGSUSED */
-STATIC xfs_exntst_t
-xfs_extent_state(
-	xfs_filblks_t		blks,
-	int			extent_flag)
-{
-	if (extent_flag) {
-		ASSERT(blks != 0);	/* saved for DMIG */
-		return XFS_EXT_UNWRITTEN;
-	}
-	return XFS_EXT_NORM;
-}
-
 /*
  * Convert on-disk form of btree root to in-memory form.
  */
@@ -87,84 +71,21 @@ xfs_bmdr_to_bmbt(
 	memcpy(tpp, fpp, sizeof(*fpp) * dmxr);
 }
 
-/*
- * Convert a compressed bmap extent record to an uncompressed form.
- * This code must be in sync with the routines xfs_bmbt_get_startoff,
- * xfs_bmbt_get_startblock, xfs_bmbt_get_blockcount and xfs_bmbt_get_state.
- */
-STATIC void
-__xfs_bmbt_get_all(
-		uint64_t l0,
-		uint64_t l1,
-		xfs_bmbt_irec_t *s)
-{
-	int	ext_flag;
-	xfs_exntst_t st;
-
-	ext_flag = (int)(l0 >> (64 - BMBT_EXNTFLAG_BITLEN));
-	s->br_startoff = ((xfs_fileoff_t)l0 &
-			   xfs_mask64lo(64 - BMBT_EXNTFLAG_BITLEN)) >> 9;
-	s->br_startblock = (((xfs_fsblock_t)l0 & xfs_mask64lo(9)) << 43) |
-			   (((xfs_fsblock_t)l1) >> 21);
-	s->br_blockcount = (xfs_filblks_t)(l1 & xfs_mask64lo(21));
-	/* This is xfs_extent_state() in-line */
-	if (ext_flag) {
-		ASSERT(s->br_blockcount != 0);	/* saved for DMIG */
-		st = XFS_EXT_UNWRITTEN;
-	} else
-		st = XFS_EXT_NORM;
-	s->br_state = st;
-}
-
 void
-xfs_bmbt_get_all(
-	xfs_bmbt_rec_host_t *r,
-	xfs_bmbt_irec_t *s)
-{
-	__xfs_bmbt_get_all(r->l0, r->l1, s);
-}
-
-/*
- * Extract the blockcount field from an in memory bmap extent record.
- */
-xfs_filblks_t
-xfs_bmbt_get_blockcount(
-	xfs_bmbt_rec_host_t	*r)
-{
-	return (xfs_filblks_t)(r->l1 & xfs_mask64lo(21));
-}
-
-/*
- * Extract the startblock field from an in memory bmap extent record.
- */
-xfs_fsblock_t
-xfs_bmbt_get_startblock(
-	xfs_bmbt_rec_host_t	*r)
-{
-	return (((xfs_fsblock_t)r->l0 & xfs_mask64lo(9)) << 43) |
-	       (((xfs_fsblock_t)r->l1) >> 21);
-}
-
-/*
- * Extract the startoff field from an in memory bmap extent record.
- */
-xfs_fileoff_t
-xfs_bmbt_get_startoff(
-	xfs_bmbt_rec_host_t	*r)
-{
-	return ((xfs_fileoff_t)r->l0 &
-		 xfs_mask64lo(64 - BMBT_EXNTFLAG_BITLEN)) >> 9;
-}
-
-xfs_exntst_t
-xfs_bmbt_get_state(
-	xfs_bmbt_rec_host_t	*r)
-{
-	int	ext_flag;
-
-	ext_flag = (int)((r->l0) >> (64 - BMBT_EXNTFLAG_BITLEN));
-	return xfs_extent_state(xfs_bmbt_get_blockcount(r),
-				ext_flag);
+xfs_bmbt_disk_get_all(
+	struct xfs_bmbt_rec	*rec,
+	struct xfs_bmbt_irec	*irec)
+{
+	uint64_t		l0 = get_unaligned_be64(&rec->l0);
+	uint64_t		l1 = get_unaligned_be64(&rec->l1);
+
+	irec->br_startoff = (l0 & xfs_mask64lo(64 - BMBT_EXNTFLAG_BITLEN)) >> 9;
+	irec->br_startblock = ((l0 & xfs_mask64lo(9)) << 43) | (l1 >> 21);
+	irec->br_blockcount = l1 & xfs_mask64lo(21);
+	if (l0 >> (64 - BMBT_EXNTFLAG_BITLEN))
+		irec->br_state = XFS_EXT_UNWRITTEN;
+	else
+		irec->br_state = XFS_EXT_NORM;
 }
 
 /*
@@ -188,142 +109,29 @@ xfs_bmbt_disk_get_startoff(
 		 xfs_mask64lo(64 - BMBT_EXNTFLAG_BITLEN)) >> 9;
 }
 
-
-/*
- * Set all the fields in a bmap extent record from the arguments.
- */
-void
-xfs_bmbt_set_allf(
-	xfs_bmbt_rec_host_t	*r,
-	xfs_fileoff_t		startoff,
-	xfs_fsblock_t		startblock,
-	xfs_filblks_t		blockcount,
-	xfs_exntst_t		state)
-{
-	int		extent_flag = (state == XFS_EXT_NORM) ? 0 : 1;
-
-	ASSERT(state == XFS_EXT_NORM || state == XFS_EXT_UNWRITTEN);
-	ASSERT((startoff & xfs_mask64hi(64-BMBT_STARTOFF_BITLEN)) == 0);
-	ASSERT((blockcount & xfs_mask64hi(64-BMBT_BLOCKCOUNT_BITLEN)) == 0);
-
-	ASSERT((startblock & xfs_mask64hi(64-BMBT_STARTBLOCK_BITLEN)) == 0);
-
-	r->l0 = ((xfs_bmbt_rec_base_t)extent_flag << 63) |
-		((xfs_bmbt_rec_base_t)startoff << 9) |
-		((xfs_bmbt_rec_base_t)startblock >> 43);
-	r->l1 = ((xfs_bmbt_rec_base_t)startblock << 21) |
-		((xfs_bmbt_rec_base_t)blockcount &
-		(xfs_bmbt_rec_base_t)xfs_mask64lo(21));
-}
-
 /*
  * Set all the fields in a bmap extent record from the uncompressed form.
  */
 void
-xfs_bmbt_set_all(
-	xfs_bmbt_rec_host_t *r,
-	xfs_bmbt_irec_t	*s)
-{
-	xfs_bmbt_set_allf(r, s->br_startoff, s->br_startblock,
-			     s->br_blockcount, s->br_state);
-}
-
-
-/*
- * Set all the fields in a disk format bmap extent record from the arguments.
- */
-void
-xfs_bmbt_disk_set_allf(
-	xfs_bmbt_rec_t		*r,
-	xfs_fileoff_t		startoff,
-	xfs_fsblock_t		startblock,
-	xfs_filblks_t		blockcount,
-	xfs_exntst_t		state)
-{
-	int			extent_flag = (state == XFS_EXT_NORM) ? 0 : 1;
-
-	ASSERT(state == XFS_EXT_NORM || state == XFS_EXT_UNWRITTEN);
-	ASSERT((startoff & xfs_mask64hi(64-BMBT_STARTOFF_BITLEN)) == 0);
-	ASSERT((blockcount & xfs_mask64hi(64-BMBT_BLOCKCOUNT_BITLEN)) == 0);
-	ASSERT((startblock & xfs_mask64hi(64-BMBT_STARTBLOCK_BITLEN)) == 0);
-
-	r->l0 = cpu_to_be64(
-		((xfs_bmbt_rec_base_t)extent_flag << 63) |
-		 ((xfs_bmbt_rec_base_t)startoff << 9) |
-		 ((xfs_bmbt_rec_base_t)startblock >> 43));
-	r->l1 = cpu_to_be64(
-		((xfs_bmbt_rec_base_t)startblock << 21) |
-		 ((xfs_bmbt_rec_base_t)blockcount &
-		  (xfs_bmbt_rec_base_t)xfs_mask64lo(21)));
-}
-
-/*
- * Set all the fields in a bmap extent record from the uncompressed form.
- */
-STATIC void
 xfs_bmbt_disk_set_all(
-	xfs_bmbt_rec_t	*r,
-	xfs_bmbt_irec_t *s)
+	struct xfs_bmbt_rec	*r,
+	struct xfs_bmbt_irec	*s)
 {
-	xfs_bmbt_disk_set_allf(r, s->br_startoff, s->br_startblock,
-				  s->br_blockcount, s->br_state);
-}
+	int			extent_flag = (s->br_state != XFS_EXT_NORM);
 
-/*
- * Set the blockcount field in a bmap extent record.
- */
-void
-xfs_bmbt_set_blockcount(
-	xfs_bmbt_rec_host_t *r,
-	xfs_filblks_t	v)
-{
-	ASSERT((v & xfs_mask64hi(43)) == 0);
-	r->l1 = (r->l1 & (xfs_bmbt_rec_base_t)xfs_mask64hi(43)) |
-		  (xfs_bmbt_rec_base_t)(v & xfs_mask64lo(21));
-}
-
-/*
- * Set the startblock field in a bmap extent record.
- */
-void
-xfs_bmbt_set_startblock(
-	xfs_bmbt_rec_host_t *r,
-	xfs_fsblock_t	v)
-{
-	ASSERT((v & xfs_mask64hi(12)) == 0);
-	r->l0 = (r->l0 & (xfs_bmbt_rec_base_t)xfs_mask64hi(55)) |
-		  (xfs_bmbt_rec_base_t)(v >> 43);
-	r->l1 = (r->l1 & (xfs_bmbt_rec_base_t)xfs_mask64lo(21)) |
-		  (xfs_bmbt_rec_base_t)(v << 21);
-}
-
-/*
- * Set the startoff field in a bmap extent record.
- */
-void
-xfs_bmbt_set_startoff(
-	xfs_bmbt_rec_host_t *r,
-	xfs_fileoff_t	v)
-{
-	ASSERT((v & xfs_mask64hi(9)) == 0);
-	r->l0 = (r->l0 & (xfs_bmbt_rec_base_t) xfs_mask64hi(1)) |
-		((xfs_bmbt_rec_base_t)v << 9) |
-		  (r->l0 & (xfs_bmbt_rec_base_t)xfs_mask64lo(9));
-}
+	ASSERT(s->br_state == XFS_EXT_NORM || s->br_state == XFS_EXT_UNWRITTEN);
+	ASSERT(!(s->br_startoff & xfs_mask64hi(64-BMBT_STARTOFF_BITLEN)));
+	ASSERT(!(s->br_blockcount & xfs_mask64hi(64-BMBT_BLOCKCOUNT_BITLEN)));
+	ASSERT(!(s->br_startblock & xfs_mask64hi(64-BMBT_STARTBLOCK_BITLEN)));
 
-/*
- * Set the extent state field in a bmap extent record.
- */
-void
-xfs_bmbt_set_state(
-	xfs_bmbt_rec_host_t *r,
-	xfs_exntst_t	v)
-{
-	ASSERT(v == XFS_EXT_NORM || v == XFS_EXT_UNWRITTEN);
-	if (v == XFS_EXT_NORM)
-		r->l0 &= xfs_mask64lo(64 - BMBT_EXNTFLAG_BITLEN);
-	else
-		r->l0 |= xfs_mask64hi(BMBT_EXNTFLAG_BITLEN);
+	put_unaligned_be64(
+		((xfs_bmbt_rec_base_t)extent_flag << 63) |
+		 ((xfs_bmbt_rec_base_t)s->br_startoff << 9) |
+		 ((xfs_bmbt_rec_base_t)s->br_startblock >> 43), &r->l0);
+	put_unaligned_be64(
+		((xfs_bmbt_rec_base_t)s->br_startblock << 21) |
+		 ((xfs_bmbt_rec_base_t)s->br_blockcount &
+		  (xfs_bmbt_rec_base_t)xfs_mask64lo(21)), &r->l1);
 }
 
 /*

fs/xfs/libxfs/xfs_bmap_btree.h

@@ -98,25 +98,11 @@ struct xfs_trans;
  */
 extern void xfs_bmdr_to_bmbt(struct xfs_inode *, xfs_bmdr_block_t *, int,
 			struct xfs_btree_block *, int);
-extern void xfs_bmbt_get_all(xfs_bmbt_rec_host_t *r, xfs_bmbt_irec_t *s);
-extern xfs_filblks_t xfs_bmbt_get_blockcount(xfs_bmbt_rec_host_t *r);
-extern xfs_fsblock_t xfs_bmbt_get_startblock(xfs_bmbt_rec_host_t *r);
-extern xfs_fileoff_t xfs_bmbt_get_startoff(xfs_bmbt_rec_host_t *r);
-extern xfs_exntst_t xfs_bmbt_get_state(xfs_bmbt_rec_host_t *r);
 
+void xfs_bmbt_disk_set_all(struct xfs_bmbt_rec *r, struct xfs_bmbt_irec *s);
 extern xfs_filblks_t xfs_bmbt_disk_get_blockcount(xfs_bmbt_rec_t *r);
 extern xfs_fileoff_t xfs_bmbt_disk_get_startoff(xfs_bmbt_rec_t *r);
-
-extern void xfs_bmbt_set_all(xfs_bmbt_rec_host_t *r, xfs_bmbt_irec_t *s);
-extern void xfs_bmbt_set_allf(xfs_bmbt_rec_host_t *r, xfs_fileoff_t o,
-			xfs_fsblock_t b, xfs_filblks_t c, xfs_exntst_t v);
-extern void xfs_bmbt_set_blockcount(xfs_bmbt_rec_host_t *r, xfs_filblks_t v);
-extern void xfs_bmbt_set_startblock(xfs_bmbt_rec_host_t *r, xfs_fsblock_t v);
-extern void xfs_bmbt_set_startoff(xfs_bmbt_rec_host_t *r, xfs_fileoff_t v);
-extern void xfs_bmbt_set_state(xfs_bmbt_rec_host_t *r, xfs_exntst_t v);
-
-extern void xfs_bmbt_disk_set_allf(xfs_bmbt_rec_t *r, xfs_fileoff_t o,
-			xfs_fsblock_t b, xfs_filblks_t c, xfs_exntst_t v);
+extern void xfs_bmbt_disk_get_all(xfs_bmbt_rec_t *r, xfs_bmbt_irec_t *s);
 
 extern void xfs_bmbt_to_bmdr(struct xfs_mount *, struct xfs_btree_block *, int,
 			xfs_bmdr_block_t *, int);
@@ -136,9 +122,9 @@ extern struct xfs_btree_cur *xfs_bmbt_init_cursor(struct xfs_mount *,
  * Check that the extent does not contain an invalid unwritten extent flag.
  */
 static inline bool xfs_bmbt_validate_extent(struct xfs_mount *mp, int whichfork,
-		struct xfs_bmbt_rec_host *ep)
+		struct xfs_bmbt_irec *irec)
 {
-	if (ep->l0 >> (64 - BMBT_EXNTFLAG_BITLEN) == 0)
+	if (irec->br_state == XFS_EXT_NORM)
 		return true;
 	if (whichfork == XFS_DATA_FORK &&
 	    xfs_sb_version_hasextflgbit(&mp->m_sb))

fs/xfs/libxfs/xfs_btree.h

@@ -255,6 +255,14 @@ typedef struct xfs_btree_cur
  */
 #define	XFS_BUF_TO_BLOCK(bp)	((struct xfs_btree_block *)((bp)->b_addr))
 
+/*
+ * Internal long and short btree block checks.  They return NULL if the
+ * block is ok or the address of the failed check otherwise.
+ */
+xfs_failaddr_t __xfs_btree_check_lblock(struct xfs_btree_cur *cur,
+		struct xfs_btree_block *block, int level, struct xfs_buf *bp);
+xfs_failaddr_t __xfs_btree_check_sblock(struct xfs_btree_cur *cur,
+		struct xfs_btree_block *block, int level, struct xfs_buf *bp);
 
 /*
  * Check that block header is ok.
@@ -269,10 +277,19 @@ xfs_btree_check_block(
 /*
  * Check that (long) pointer is ok.
  */
-int					/* error (0 or EFSCORRUPTED) */
+bool					/* error (0 or EFSCORRUPTED) */
 xfs_btree_check_lptr(
 	struct xfs_btree_cur	*cur,	/* btree cursor */
-	xfs_fsblock_t		ptr,	/* btree block disk address */
+	xfs_fsblock_t		fsbno,	/* btree block disk address */
+	int			level);	/* btree block level */
+
+/*
+ * Check that (short) pointer is ok.
+ */
+bool					/* error (0 or EFSCORRUPTED) */
+xfs_btree_check_sptr(
+	struct xfs_btree_cur	*cur,	/* btree cursor */
+	xfs_agblock_t		agbno,	/* btree block disk address */
 	int			level);	/* btree block level */
 
 /*
@@ -517,5 +534,16 @@ int xfs_btree_lookup_get_block(struct xfs_btree_cur *cur, int level,
 		union xfs_btree_ptr *pp, struct xfs_btree_block **blkp);
 struct xfs_btree_block *xfs_btree_get_block(struct xfs_btree_cur *cur,
 		int level, struct xfs_buf **bpp);
+bool xfs_btree_ptr_is_null(struct xfs_btree_cur *cur, union xfs_btree_ptr *ptr);
+int64_t xfs_btree_diff_two_ptrs(struct xfs_btree_cur *cur,
+				const union xfs_btree_ptr *a,
+				const union xfs_btree_ptr *b);
+void xfs_btree_get_sibling(struct xfs_btree_cur *cur,
+			   struct xfs_btree_block *block,
+			   union xfs_btree_ptr *ptr, int lr);
+void xfs_btree_get_keys(struct xfs_btree_cur *cur,
+		struct xfs_btree_block *block, union xfs_btree_key *key);
+union xfs_btree_key *xfs_btree_high_key_from_key(struct xfs_btree_cur *cur,
+		union xfs_btree_key *key);
 
 #endif	/* __XFS_BTREE_H__ */

fs/xfs/libxfs/xfs_da_btree.c

@@ -1466,6 +1466,7 @@ xfs_da3_node_lookup_int(
 	int			max;
 	int			error;
 	int			retval;
+	unsigned int		expected_level = 0;
 	struct xfs_inode	*dp = state->args->dp;
 
 	args = state->args;
@@ -1474,7 +1475,7 @@ xfs_da3_node_lookup_int(
 	 * Descend thru the B-tree searching each level for the right
 	 * node to use, until the right hashval is found.
 	 */
-	blkno = (args->whichfork == XFS_DATA_FORK)? args->geo->leafblk : 0;
+	blkno = args->geo->leafblk;
 	for (blk = &state->path.blk[0], state->path.active = 1;
 			 state->path.active <= XFS_DA_NODE_MAXDEPTH;
 			 blk++, state->path.active++) {
@@ -1517,6 +1518,18 @@ xfs_da3_node_lookup_int(
 		dp->d_ops->node_hdr_from_disk(&nodehdr, node);
 		btree = dp->d_ops->node_tree_p(node);
 
+		/* Tree taller than we can handle; bail out! */
+		if (nodehdr.level >= XFS_DA_NODE_MAXDEPTH)
+			return -EFSCORRUPTED;
+
+		/* Check the level from the root. */
+		if (blkno == args->geo->leafblk)
+			expected_level = nodehdr.level - 1;
+		else if (expected_level != nodehdr.level)
+			return -EFSCORRUPTED;
+		else
+			expected_level--;
+
 		max = nodehdr.count;
 		blk->hashval = be32_to_cpu(btree[max - 1].hashval);
 
@@ -1562,8 +1575,15 @@ xfs_da3_node_lookup_int(
 			blk->index = probe;
 			blkno = be32_to_cpu(btree[probe].before);
 		}
+
+		/* We can't point back to the root. */
+		if (blkno == args->geo->leafblk)
+			return -EFSCORRUPTED;
 	}
 
+	if (expected_level != 0)
+		return -EFSCORRUPTED;
+
 	/*
 	 * A leaf block that ends in the hashval that we are interested in
 	 * (final hashval == search hashval) means that the next block may

fs/xfs/libxfs/xfs_dir2.c

@@ -30,6 +30,8 @@
 #include "xfs_bmap.h"
 #include "xfs_dir2.h"
 #include "xfs_dir2_priv.h"
+#include "xfs_ialloc.h"
+#include "xfs_errortag.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
 
@@ -38,7 +40,9 @@ struct xfs_name xfs_name_dotdot = { (unsigned char *)"..", 2, XFS_DIR3_FT_DIR };
 /*
  * Convert inode mode to directory entry filetype
  */
-unsigned char xfs_mode_to_ftype(int mode)
+unsigned char
+xfs_mode_to_ftype(
+	int		mode)
 {
 	switch (mode & S_IFMT) {
 	case S_IFREG:
@@ -202,22 +206,8 @@ xfs_dir_ino_validate(
 	xfs_mount_t	*mp,
 	xfs_ino_t	ino)
 {
-	xfs_agblock_t	agblkno;
-	xfs_agino_t	agino;
-	xfs_agnumber_t	agno;
-	int		ino_ok;
-	int		ioff;
-
-	agno = XFS_INO_TO_AGNO(mp, ino);
-	agblkno = XFS_INO_TO_AGBNO(mp, ino);
-	ioff = XFS_INO_TO_OFFSET(mp, ino);
-	agino = XFS_OFFBNO_TO_AGINO(mp, agblkno, ioff);
-	ino_ok =
-		agno < mp->m_sb.sb_agcount &&
-		agblkno < mp->m_sb.sb_agblocks &&
-		agblkno != 0 &&
-		ioff < (1 << mp->m_sb.sb_inopblog) &&
-		XFS_AGINO_TO_INO(mp, agno, agino) == ino;
+	bool		ino_ok = xfs_verify_dir_ino(mp, ino);
+
 	if (unlikely(XFS_TEST_ERROR(!ino_ok, mp, XFS_ERRTAG_DIR_INO_VALIDATE))) {
 		xfs_warn(mp, "Invalid inode number 0x%Lx",
 				(unsigned long long) ino);

fs/xfs/libxfs/xfs_dir2.h

@@ -324,4 +324,21 @@ xfs_dir2_leaf_tail_p(struct xfs_da_geometry *geo, struct xfs_dir2_leaf *lp)
 		  sizeof(struct xfs_dir2_leaf_tail));
 }
 
+/*
+ * The Linux API doesn't pass down the total size of the buffer
+ * we read into down to the filesystem.  With the filldir concept
+ * it's not needed for correct information, but the XFS dir2 leaf
+ * code wants an estimate of the buffer size to calculate it's
+ * readahead window and size the buffers used for mapping to
+ * physical blocks.
+ *
+ * Try to give it an estimate that's good enough, maybe at some
+ * point we can change the ->readdir prototype to include the
+ * buffer size.  For now we use the current glibc buffer size.
+ * musl libc hardcodes 2k and dietlibc uses PAGE_SIZE.
+ */
+#define XFS_READDIR_BUFSIZE	(32768)
+
+unsigned char xfs_dir3_get_dtype(struct xfs_mount *mp, uint8_t filetype);
+
 #endif	/* __XFS_DIR2_H__ */

fs/xfs/libxfs/xfs_errortag.h

+/*
+ * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc.
+ * Copyright (C) 2017 Oracle.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301, USA.
+ */
+#ifndef __XFS_ERRORTAG_H_
+#define __XFS_ERRORTAG_H_
+
+/*
+ * error injection tags - the labels can be anything you want
+ * but each tag should have its own unique number
+ */
+
+#define XFS_ERRTAG_NOERROR				0
+#define XFS_ERRTAG_IFLUSH_1				1
+#define XFS_ERRTAG_IFLUSH_2				2
+#define XFS_ERRTAG_IFLUSH_3				3
+#define XFS_ERRTAG_IFLUSH_4				4
+#define XFS_ERRTAG_IFLUSH_5				5
+#define XFS_ERRTAG_IFLUSH_6				6
+#define XFS_ERRTAG_DA_READ_BUF				7
+#define XFS_ERRTAG_BTREE_CHECK_LBLOCK			8
+#define XFS_ERRTAG_BTREE_CHECK_SBLOCK			9
+#define XFS_ERRTAG_ALLOC_READ_AGF			10
+#define XFS_ERRTAG_IALLOC_READ_AGI			11
+#define XFS_ERRTAG_ITOBP_INOTOBP			12
+#define XFS_ERRTAG_IUNLINK				13
+#define XFS_ERRTAG_IUNLINK_REMOVE			14
+#define XFS_ERRTAG_DIR_INO_VALIDATE			15
+#define XFS_ERRTAG_BULKSTAT_READ_CHUNK			16
+#define XFS_ERRTAG_IODONE_IOERR				17
+#define XFS_ERRTAG_STRATREAD_IOERR			18
+#define XFS_ERRTAG_STRATCMPL_IOERR			19
+#define XFS_ERRTAG_DIOWRITE_IOERR			20
+#define XFS_ERRTAG_BMAPIFORMAT				21
+#define XFS_ERRTAG_FREE_EXTENT				22
+#define XFS_ERRTAG_RMAP_FINISH_ONE			23
+#define XFS_ERRTAG_REFCOUNT_CONTINUE_UPDATE		24
+#define XFS_ERRTAG_REFCOUNT_FINISH_ONE			25
+#define XFS_ERRTAG_BMAP_FINISH_ONE			26
+#define XFS_ERRTAG_AG_RESV_CRITICAL			27
+/*
+ * DEBUG mode instrumentation to test and/or trigger delayed allocation
+ * block killing in the event of failed writes. When enabled, all
+ * buffered writes are silenty dropped and handled as if they failed.
+ * All delalloc blocks in the range of the write (including pre-existing
+ * delalloc blocks!) are tossed as part of the write failure error
+ * handling sequence.
+ */
+#define XFS_ERRTAG_DROP_WRITES				28
+#define XFS_ERRTAG_LOG_BAD_CRC				29
+#define XFS_ERRTAG_LOG_ITEM_PIN				30
+#define XFS_ERRTAG_BUF_LRU_REF				31
+#define XFS_ERRTAG_MAX					32
+
+/*
+ * Random factors for above tags, 1 means always, 2 means 1/2 time, etc.
+ */
+#define XFS_RANDOM_DEFAULT				100
+#define XFS_RANDOM_IFLUSH_1				XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_IFLUSH_2				XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_IFLUSH_3				XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_IFLUSH_4				XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_IFLUSH_5				XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_IFLUSH_6				XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_DA_READ_BUF				XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_BTREE_CHECK_LBLOCK			(XFS_RANDOM_DEFAULT/4)
+#define XFS_RANDOM_BTREE_CHECK_SBLOCK			XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_ALLOC_READ_AGF			XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_IALLOC_READ_AGI			XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_ITOBP_INOTOBP			XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_IUNLINK				XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_IUNLINK_REMOVE			XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_DIR_INO_VALIDATE			XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_BULKSTAT_READ_CHUNK			XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_IODONE_IOERR				(XFS_RANDOM_DEFAULT/10)
+#define XFS_RANDOM_STRATREAD_IOERR			(XFS_RANDOM_DEFAULT/10)
+#define XFS_RANDOM_STRATCMPL_IOERR			(XFS_RANDOM_DEFAULT/10)
+#define XFS_RANDOM_DIOWRITE_IOERR			(XFS_RANDOM_DEFAULT/10)
+#define XFS_RANDOM_BMAPIFORMAT				XFS_RANDOM_DEFAULT
+#define XFS_RANDOM_FREE_EXTENT				1
+#define XFS_RANDOM_RMAP_FINISH_ONE			1
+#define XFS_RANDOM_REFCOUNT_CONTINUE_UPDATE		1
+#define XFS_RANDOM_REFCOUNT_FINISH_ONE			1
+#define XFS_RANDOM_BMAP_FINISH_ONE			1
+#define XFS_RANDOM_AG_RESV_CRITICAL			4
+#define XFS_RANDOM_DROP_WRITES				1
+#define XFS_RANDOM_LOG_BAD_CRC				1
+#define XFS_RANDOM_LOG_ITEM_PIN				1
+#define XFS_RANDOM_BUF_LRU_REF				2
+
+#endif /* __XFS_ERRORTAG_H_ */

fs/xfs/libxfs/xfs_format.h

@@ -315,6 +315,11 @@ static inline bool xfs_sb_good_version(struct xfs_sb *sbp)
 	return false;
 }
 
+static inline bool xfs_sb_version_hasrealtime(struct xfs_sb *sbp)
+{
+	return sbp->sb_rblocks > 0;
+}
+
 /*
  * Detect a mismatched features2 field.  Older kernels read/wrote
  * this into the wrong slot, so to be safe we keep them in sync.
@@ -500,12 +505,12 @@ xfs_sb_has_incompat_log_feature(
 /*
  * V5 superblock specific feature checks
  */
-static inline int xfs_sb_version_hascrc(struct xfs_sb *sbp)
+static inline bool xfs_sb_version_hascrc(struct xfs_sb *sbp)
 {
 	return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5;
 }
 
-static inline int xfs_sb_version_has_pquotino(struct xfs_sb *sbp)
+static inline bool xfs_sb_version_has_pquotino(struct xfs_sb *sbp)
 {
 	return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5;
 }
@@ -518,7 +523,7 @@ static inline int xfs_sb_version_hasftype(struct xfs_sb *sbp)
 		 (sbp->sb_features2 & XFS_SB_VERSION2_FTYPE));
 }
 
-static inline int xfs_sb_version_hasfinobt(xfs_sb_t *sbp)
+static inline bool xfs_sb_version_hasfinobt(xfs_sb_t *sbp)
 {
 	return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) &&
 		(sbp->sb_features_ro_compat & XFS_SB_FEAT_RO_COMPAT_FINOBT);
@@ -941,7 +946,7 @@ typedef enum xfs_dinode_fmt {
 	XFS_DINODE_FMT_LOCAL,		/* bulk data */
 	XFS_DINODE_FMT_EXTENTS,		/* struct xfs_bmbt_rec */
 	XFS_DINODE_FMT_BTREE,		/* struct xfs_bmdr_block */
-	XFS_DINODE_FMT_UUID		/* uuid_t */
+	XFS_DINODE_FMT_UUID		/* added long ago, but never used */
 } xfs_dinode_fmt_t;
 
 /*
@@ -1142,7 +1147,7 @@ static inline void xfs_dinode_put_rdev(struct xfs_dinode *dip, xfs_dev_t rdev)
  * Dquot and dquot block format definitions
  */
 #define XFS_DQUOT_MAGIC		0x4451		/* 'DQ' */
-#define XFS_DQUOT_VERSION	(u_int8_t)0x01	/* latest version number */
+#define XFS_DQUOT_VERSION	(uint8_t)0x01	/* latest version number */
 
 /*
  * This is the main portion of the on-disk representation of quota
@@ -1548,10 +1553,6 @@ typedef struct xfs_bmbt_rec {
 typedef uint64_t	xfs_bmbt_rec_base_t;	/* use this for casts */
 typedef xfs_bmbt_rec_t xfs_bmdr_rec_t;
 
-typedef struct xfs_bmbt_rec_host {
-	uint64_t		l0, l1;
-} xfs_bmbt_rec_host_t;
-
 /*
  * Values and macros for delayed-allocation startblock fields.
  */
@@ -1576,24 +1577,6 @@ static inline xfs_filblks_t startblockval(xfs_fsblock_t x)
 	return (xfs_filblks_t)((x) & ~STARTBLOCKMASK);
 }
 
-/*
- * Possible extent states.
- */
-typedef enum {
-	XFS_EXT_NORM, XFS_EXT_UNWRITTEN,
-} xfs_exntst_t;
-
-/*
- * Incore version of above.
- */
-typedef struct xfs_bmbt_irec
-{
-	xfs_fileoff_t	br_startoff;	/* starting file offset */
-	xfs_fsblock_t	br_startblock;	/* starting block number */
-	xfs_filblks_t	br_blockcount;	/* number of blocks */
-	xfs_exntst_t	br_state;	/* extent state */
-} xfs_bmbt_irec_t;
-
 /*
  * Key structure for non-leaf levels of the tree.
  */

fs/xfs/libxfs/xfs_fs.h

@@ -468,6 +468,82 @@ typedef struct xfs_swapext
 #define XFS_FSOP_GOING_FLAGS_LOGFLUSH		0x1	/* flush log but not data */
 #define XFS_FSOP_GOING_FLAGS_NOLOGFLUSH		0x2	/* don't flush log nor data */
 
+/* metadata scrubbing */
+struct xfs_scrub_metadata {
+	__u32 sm_type;		/* What to check? */
+	__u32 sm_flags;		/* flags; see below. */
+	__u64 sm_ino;		/* inode number. */
+	__u32 sm_gen;		/* inode generation. */
+	__u32 sm_agno;		/* ag number. */
+	__u64 sm_reserved[5];	/* pad to 64 bytes */
+};
+
+/*
+ * Metadata types and flags for scrub operation.
+ */
+
+/* Scrub subcommands. */
+#define XFS_SCRUB_TYPE_PROBE	0	/* presence test ioctl */
+#define XFS_SCRUB_TYPE_SB	1	/* superblock */
+#define XFS_SCRUB_TYPE_AGF	2	/* AG free header */
+#define XFS_SCRUB_TYPE_AGFL	3	/* AG free list */
+#define XFS_SCRUB_TYPE_AGI	4	/* AG inode header */
+#define XFS_SCRUB_TYPE_BNOBT	5	/* freesp by block btree */
+#define XFS_SCRUB_TYPE_CNTBT	6	/* freesp by length btree */
+#define XFS_SCRUB_TYPE_INOBT	7	/* inode btree */
+#define XFS_SCRUB_TYPE_FINOBT	8	/* free inode btree */
+#define XFS_SCRUB_TYPE_RMAPBT	9	/* reverse mapping btree */
+#define XFS_SCRUB_TYPE_REFCNTBT	10	/* reference count btree */
+#define XFS_SCRUB_TYPE_INODE	11	/* inode record */
+#define XFS_SCRUB_TYPE_BMBTD	12	/* data fork block mapping */
+#define XFS_SCRUB_TYPE_BMBTA	13	/* attr fork block mapping */
+#define XFS_SCRUB_TYPE_BMBTC	14	/* CoW fork block mapping */
+#define XFS_SCRUB_TYPE_DIR	15	/* directory */
+#define XFS_SCRUB_TYPE_XATTR	16	/* extended attribute */
+#define XFS_SCRUB_TYPE_SYMLINK	17	/* symbolic link */
+#define XFS_SCRUB_TYPE_PARENT	18	/* parent pointers */
+#define XFS_SCRUB_TYPE_RTBITMAP	19	/* realtime bitmap */
+#define XFS_SCRUB_TYPE_RTSUM	20	/* realtime summary */
+#define XFS_SCRUB_TYPE_UQUOTA	21	/* user quotas */
+#define XFS_SCRUB_TYPE_GQUOTA	22	/* group quotas */
+#define XFS_SCRUB_TYPE_PQUOTA	23	/* project quotas */
+
+/* Number of scrub subcommands. */
+#define XFS_SCRUB_TYPE_NR	24
+
+/* i: Repair this metadata. */
+#define XFS_SCRUB_IFLAG_REPAIR		(1 << 0)
+
+/* o: Metadata object needs repair. */
+#define XFS_SCRUB_OFLAG_CORRUPT		(1 << 1)
+
+/*
+ * o: Metadata object could be optimized.  It's not corrupt, but
+ *    we could improve on it somehow.
+ */
+#define XFS_SCRUB_OFLAG_PREEN		(1 << 2)
+
+/* o: Cross-referencing failed. */
+#define XFS_SCRUB_OFLAG_XFAIL		(1 << 3)
+
+/* o: Metadata object disagrees with cross-referenced metadata. */
+#define XFS_SCRUB_OFLAG_XCORRUPT	(1 << 4)
+
+/* o: Scan was not complete. */
+#define XFS_SCRUB_OFLAG_INCOMPLETE	(1 << 5)
+
+/* o: Metadata object looked funny but isn't corrupt. */
+#define XFS_SCRUB_OFLAG_WARNING		(1 << 6)
+
+#define XFS_SCRUB_FLAGS_IN	(XFS_SCRUB_IFLAG_REPAIR)
+#define XFS_SCRUB_FLAGS_OUT	(XFS_SCRUB_OFLAG_CORRUPT | \
+				 XFS_SCRUB_OFLAG_PREEN | \
+				 XFS_SCRUB_OFLAG_XFAIL | \
+				 XFS_SCRUB_OFLAG_XCORRUPT | \
+				 XFS_SCRUB_OFLAG_INCOMPLETE | \
+				 XFS_SCRUB_OFLAG_WARNING)
+#define XFS_SCRUB_FLAGS_ALL	(XFS_SCRUB_FLAGS_IN | XFS_SCRUB_FLAGS_OUT)
+
 /*
  * ioctl limits
  */
@@ -511,6 +587,7 @@ typedef struct xfs_swapext
 #define XFS_IOC_ZERO_RANGE	_IOW ('X', 57, struct xfs_flock64)
 #define XFS_IOC_FREE_EOFBLOCKS	_IOR ('X', 58, struct xfs_fs_eofblocks)
 /*	XFS_IOC_GETFSMAP ------ hoisted 59         */
+#define XFS_IOC_SCRUB_METADATA	_IOWR('X', 60, struct xfs_scrub_metadata)
 
 /*
  * ioctl commands that replace IRIX syssgi()'s

fs/xfs/libxfs/xfs_ialloc.c

@@ -31,6 +31,7 @@
 #include "xfs_ialloc_btree.h"
 #include "xfs_alloc.h"
 #include "xfs_rtalloc.h"
+#include "xfs_errortag.h"
 #include "xfs_error.h"
 #include "xfs_bmap.h"
 #include "xfs_cksum.h"
@@ -2664,3 +2665,93 @@ xfs_ialloc_pagi_init(
 		xfs_trans_brelse(tp, bp);
 	return 0;
 }
+
+/* Calculate the first and last possible inode number in an AG. */
+void
+xfs_ialloc_agino_range(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		agno,
+	xfs_agino_t		*first,
+	xfs_agino_t		*last)
+{
+	xfs_agblock_t		bno;
+	xfs_agblock_t		eoag;
+
+	eoag = xfs_ag_block_count(mp, agno);
+
+	/*
+	 * Calculate the first inode, which will be in the first
+	 * cluster-aligned block after the AGFL.
+	 */
+	bno = round_up(XFS_AGFL_BLOCK(mp) + 1,
+			xfs_ialloc_cluster_alignment(mp));
+	*first = XFS_OFFBNO_TO_AGINO(mp, bno, 0);
+
+	/*
+	 * Calculate the last inode, which will be at the end of the
+	 * last (aligned) cluster that can be allocated in the AG.
+	 */
+	bno = round_down(eoag, xfs_ialloc_cluster_alignment(mp));
+	*last = XFS_OFFBNO_TO_AGINO(mp, bno, 0) - 1;
+}
+
+/*
+ * Verify that an AG inode number pointer neither points outside the AG
+ * nor points at static metadata.
+ */
+bool
+xfs_verify_agino(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		agno,
+	xfs_agino_t		agino)
+{
+	xfs_agino_t		first;
+	xfs_agino_t		last;
+
+	xfs_ialloc_agino_range(mp, agno, &first, &last);
+	return agino >= first && agino <= last;
+}
+
+/*
+ * Verify that an FS inode number pointer neither points outside the
+ * filesystem nor points at static AG metadata.
+ */
+bool
+xfs_verify_ino(
+	struct xfs_mount	*mp,
+	xfs_ino_t		ino)
+{
+	xfs_agnumber_t		agno = XFS_INO_TO_AGNO(mp, ino);
+	xfs_agino_t		agino = XFS_INO_TO_AGINO(mp, ino);
+
+	if (agno >= mp->m_sb.sb_agcount)
+		return false;
+	if (XFS_AGINO_TO_INO(mp, agno, agino) != ino)
+		return false;
+	return xfs_verify_agino(mp, agno, agino);
+}
+
+/* Is this an internal inode number? */
+bool
+xfs_internal_inum(
+	struct xfs_mount	*mp,
+	xfs_ino_t		ino)
+{
+	return ino == mp->m_sb.sb_rbmino || ino == mp->m_sb.sb_rsumino ||
+		(xfs_sb_version_hasquota(&mp->m_sb) &&
+		 xfs_is_quota_inode(&mp->m_sb, ino));
+}
+
+/*
+ * Verify that a directory entry's inode number doesn't point at an internal
+ * inode, empty space, or static AG metadata.
+ */
+bool
+xfs_verify_dir_ino(
+	struct xfs_mount	*mp,
+	xfs_ino_t		ino)
+{
+	if (xfs_internal_inum(mp, ino))
+		return false;
+	return xfs_verify_ino(mp, ino);
+}

fs/xfs/libxfs/xfs_ialloc.h

@@ -173,5 +173,12 @@ void xfs_inobt_btrec_to_irec(struct xfs_mount *mp, union xfs_btree_rec *rec,
 		struct xfs_inobt_rec_incore *irec);
 
 int xfs_ialloc_cluster_alignment(struct xfs_mount *mp);
+void xfs_ialloc_agino_range(struct xfs_mount *mp, xfs_agnumber_t agno,
+		xfs_agino_t *first, xfs_agino_t *last);
+bool xfs_verify_agino(struct xfs_mount *mp, xfs_agnumber_t agno,
+		xfs_agino_t agino);
+bool xfs_verify_ino(struct xfs_mount *mp, xfs_ino_t ino);
+bool xfs_internal_inum(struct xfs_mount *mp, xfs_ino_t ino);
+bool xfs_verify_dir_ino(struct xfs_mount *mp, xfs_ino_t ino);
 
 #endif	/* __XFS_IALLOC_H__ */

fs/xfs/libxfs/xfs_inode_buf.c

@@ -24,6 +24,7 @@
 #include "xfs_mount.h"
 #include "xfs_defer.h"
 #include "xfs_inode.h"
+#include "xfs_errortag.h"
 #include "xfs_error.h"
 #include "xfs_cksum.h"
 #include "xfs_icache.h"

fs/xfs/libxfs/xfs_inode_fork.h

@@ -21,57 +21,20 @@
 struct xfs_inode_log_item;
 struct xfs_dinode;
 
-/*
- * The following xfs_ext_irec_t struct introduces a second (top) level
- * to the in-core extent allocation scheme. These structs are allocated
- * in a contiguous block, creating an indirection array where each entry
- * (irec) contains a pointer to a buffer of in-core extent records which
- * it manages. Each extent buffer is 4k in size, since 4k is the system
- * page size on Linux i386 and systems with larger page sizes don't seem
- * to gain much, if anything, by using their native page size as the
- * extent buffer size. Also, using 4k extent buffers everywhere provides
- * a consistent interface for CXFS across different platforms.
- *
- * There is currently no limit on the number of irec's (extent lists)
- * allowed, so heavily fragmented files may require an indirection array
- * which spans multiple system pages of memory. The number of extents
- * which would require this amount of contiguous memory is very large
- * and should not cause problems in the foreseeable future. However,
- * if the memory needed for the contiguous array ever becomes a problem,
- * it is possible that a third level of indirection may be required.
- */
-typedef struct xfs_ext_irec {
-	xfs_bmbt_rec_host_t *er_extbuf;	/* block of extent records */
-	xfs_extnum_t	er_extoff;	/* extent offset in file */
-	xfs_extnum_t	er_extcount;	/* number of extents in page/block */
-} xfs_ext_irec_t;
-
 /*
  * File incore extent information, present for each of data & attr forks.
  */
-#define	XFS_IEXT_BUFSZ		4096
-#define	XFS_LINEAR_EXTS		(XFS_IEXT_BUFSZ / (uint)sizeof(xfs_bmbt_rec_t))
-#define	XFS_INLINE_EXTS		2
-#define	XFS_INLINE_DATA		32
 typedef struct xfs_ifork {
 	int			if_bytes;	/* bytes in if_u1 */
 	int			if_real_bytes;	/* bytes allocated in if_u1 */
 	struct xfs_btree_block	*if_broot;	/* file's incore btree root */
 	short			if_broot_bytes;	/* bytes allocated for root */
 	unsigned char		if_flags;	/* per-fork flags */
+	int			if_height;	/* height of the extent tree */
 	union {
-		xfs_bmbt_rec_host_t *if_extents;/* linear map file exts */
-		xfs_ext_irec_t	*if_ext_irec;	/* irec map file exts */
+		void		*if_root;	/* extent tree root */
 		char		*if_data;	/* inline file data */
 	} if_u1;
-	union {
-		xfs_bmbt_rec_host_t if_inline_ext[XFS_INLINE_EXTS];
-						/* very small file extents */
-		char		if_inline_data[XFS_INLINE_DATA];
-						/* very small file data */
-		xfs_dev_t	if_rdev;	/* dev number if special */
-		uuid_t		if_uuid;	/* mount point value */
-	} if_u2;
 } xfs_ifork_t;
 
 /*
@@ -80,7 +43,6 @@ typedef struct xfs_ifork {
 #define	XFS_IFINLINE	0x01	/* Inline data is read in */
 #define	XFS_IFEXTENTS	0x02	/* All extent pointers are read in */
 #define	XFS_IFBROOT	0x04	/* i_broot points to the bmap b-tree root */
-#define	XFS_IFEXTIREC	0x08	/* Indirection array of extent blocks */
 
 /*
  * Fork handling.
@@ -150,46 +112,76 @@ int		xfs_iextents_copy(struct xfs_inode *, struct xfs_bmbt_rec *,
 				  int);
 void		xfs_init_local_fork(struct xfs_inode *, int, const void *, int);
 
-struct xfs_bmbt_rec_host *
-		xfs_iext_get_ext(struct xfs_ifork *, xfs_extnum_t);
-xfs_extnum_t	xfs_iext_count(struct xfs_ifork *);
-void		xfs_iext_insert(struct xfs_inode *, xfs_extnum_t, xfs_extnum_t,
+xfs_extnum_t	xfs_iext_count(struct xfs_ifork *ifp);
+void		xfs_iext_insert(struct xfs_inode *, struct xfs_iext_cursor *cur,
 			struct xfs_bmbt_irec *, int);
-void		xfs_iext_add(struct xfs_ifork *, xfs_extnum_t, int);
-void		xfs_iext_add_indirect_multi(struct xfs_ifork *, int,
-					    xfs_extnum_t, int);
-void		xfs_iext_remove(struct xfs_inode *, xfs_extnum_t, int, int);
-void		xfs_iext_remove_inline(struct xfs_ifork *, xfs_extnum_t, int);
-void		xfs_iext_remove_direct(struct xfs_ifork *, xfs_extnum_t, int);
-void		xfs_iext_remove_indirect(struct xfs_ifork *, xfs_extnum_t, int);
-void		xfs_iext_realloc_direct(struct xfs_ifork *, int);
-void		xfs_iext_direct_to_inline(struct xfs_ifork *, xfs_extnum_t);
-void		xfs_iext_inline_to_direct(struct xfs_ifork *, int);
-void		xfs_iext_destroy(struct xfs_ifork *);
-struct xfs_bmbt_rec_host *
-		xfs_iext_bno_to_ext(struct xfs_ifork *, xfs_fileoff_t, int *);
-struct xfs_ext_irec *
-		xfs_iext_bno_to_irec(struct xfs_ifork *, xfs_fileoff_t, int *);
-struct xfs_ext_irec *
-		xfs_iext_idx_to_irec(struct xfs_ifork *, xfs_extnum_t *, int *,
+void		xfs_iext_remove(struct xfs_inode *, struct xfs_iext_cursor *,
 			int);
-void		xfs_iext_irec_init(struct xfs_ifork *);
-struct xfs_ext_irec *
-		xfs_iext_irec_new(struct xfs_ifork *, int);
-void		xfs_iext_irec_remove(struct xfs_ifork *, int);
-void		xfs_iext_irec_compact(struct xfs_ifork *);
-void		xfs_iext_irec_compact_pages(struct xfs_ifork *);
-void		xfs_iext_irec_compact_full(struct xfs_ifork *);
-void		xfs_iext_irec_update_extoffs(struct xfs_ifork *, int, int);
+void		xfs_iext_destroy(struct xfs_ifork *);
 
 bool		xfs_iext_lookup_extent(struct xfs_inode *ip,
 			struct xfs_ifork *ifp, xfs_fileoff_t bno,
-			xfs_extnum_t *idxp, struct xfs_bmbt_irec *gotp);
-bool		xfs_iext_get_extent(struct xfs_ifork *ifp, xfs_extnum_t idx,
+			struct xfs_iext_cursor *cur,
+			struct xfs_bmbt_irec *gotp);
+bool		xfs_iext_lookup_extent_before(struct xfs_inode *ip,
+			struct xfs_ifork *ifp, xfs_fileoff_t *end,
+			struct xfs_iext_cursor *cur,
+			struct xfs_bmbt_irec *gotp);
+bool		xfs_iext_get_extent(struct xfs_ifork *ifp,
+			struct xfs_iext_cursor *cur,
 			struct xfs_bmbt_irec *gotp);
-void		xfs_iext_update_extent(struct xfs_ifork *ifp, xfs_extnum_t idx,
+void		xfs_iext_update_extent(struct xfs_inode *ip, int state,
+			struct xfs_iext_cursor *cur,
 			struct xfs_bmbt_irec *gotp);
 
+void		xfs_iext_first(struct xfs_ifork *, struct xfs_iext_cursor *);
+void		xfs_iext_last(struct xfs_ifork *, struct xfs_iext_cursor *);
+void		xfs_iext_next(struct xfs_ifork *, struct xfs_iext_cursor *);
+void		xfs_iext_prev(struct xfs_ifork *, struct xfs_iext_cursor *);
+
+static inline bool xfs_iext_next_extent(struct xfs_ifork *ifp,
+		struct xfs_iext_cursor *cur, struct xfs_bmbt_irec *gotp)
+{
+	xfs_iext_next(ifp, cur);
+	return xfs_iext_get_extent(ifp, cur, gotp);
+}
+
+static inline bool xfs_iext_prev_extent(struct xfs_ifork *ifp,
+		struct xfs_iext_cursor *cur, struct xfs_bmbt_irec *gotp)
+{
+	xfs_iext_prev(ifp, cur);
+	return xfs_iext_get_extent(ifp, cur, gotp);
+}
+
+/*
+ * Return the extent after cur in gotp without updating the cursor.
+ */
+static inline bool xfs_iext_peek_next_extent(struct xfs_ifork *ifp,
+		struct xfs_iext_cursor *cur, struct xfs_bmbt_irec *gotp)
+{
+	struct xfs_iext_cursor ncur = *cur;
+
+	xfs_iext_next(ifp, &ncur);
+	return xfs_iext_get_extent(ifp, &ncur, gotp);
+}
+
+/*
+ * Return the extent before cur in gotp without updating the cursor.
+ */
+static inline bool xfs_iext_peek_prev_extent(struct xfs_ifork *ifp,
+		struct xfs_iext_cursor *cur, struct xfs_bmbt_irec *gotp)
+{
+	struct xfs_iext_cursor ncur = *cur;
+
+	xfs_iext_prev(ifp, &ncur);
+	return xfs_iext_get_extent(ifp, &ncur, gotp);
+}
+
+#define for_each_xfs_iext(ifp, ext, got)		\
+	for (xfs_iext_first((ifp), (ext));		\
+	     xfs_iext_get_extent((ifp), (ext), (got));	\
+	     xfs_iext_next((ifp), (ext)))
+
 extern struct kmem_zone	*xfs_ifork_zone;
 
 extern void xfs_ifork_init_cow(struct xfs_inode *ip);

fs/xfs/libxfs/xfs_log_format.h

@@ -264,7 +264,7 @@ typedef struct xfs_trans_header {
  * (if any) is indicated in the ilf_dsize field.  Changes to this structure
  * must be added on to the end.
  */
-typedef struct xfs_inode_log_format {
+struct xfs_inode_log_format {
 	uint16_t		ilf_type;	/* inode log item type */
 	uint16_t		ilf_size;	/* size of this item */
 	uint32_t		ilf_fields;	/* flags for fields logged */
@@ -274,12 +274,12 @@ typedef struct xfs_inode_log_format {
 	uint64_t		ilf_ino;	/* inode number */
 	union {
 		uint32_t	ilfu_rdev;	/* rdev value for dev inode*/
-		uuid_t		ilfu_uuid;	/* mount point value */
+		u8		__pad[16];	/* unused */
 	} ilf_u;
 	int64_t			ilf_blkno;	/* blkno of inode buffer */
 	int32_t			ilf_len;	/* len of inode buffer */
 	int32_t			ilf_boffset;	/* off of inode in buffer */
-} xfs_inode_log_format_t;
+};
 
 /*
  * Old 32 bit systems will log in this format without the 64 bit
@@ -295,7 +295,7 @@ struct xfs_inode_log_format_32 {
 	uint64_t		ilf_ino;	/* inode number */
 	union {
 		uint32_t	ilfu_rdev;	/* rdev value for dev inode*/
-		uuid_t		ilfu_uuid;	/* mount point value */
+		u8		__pad[16];	/* unused */
 	} ilf_u;
 	int64_t			ilf_blkno;	/* blkno of inode buffer */
 	int32_t			ilf_len;	/* len of inode buffer */
@@ -311,7 +311,7 @@ struct xfs_inode_log_format_32 {
 #define	XFS_ILOG_DEXT	0x004	/* log i_df.if_extents */
 #define	XFS_ILOG_DBROOT	0x008	/* log i_df.i_broot */
 #define	XFS_ILOG_DEV	0x010	/* log the dev field */
-#define	XFS_ILOG_UUID	0x020	/* log the uuid field */
+#define	XFS_ILOG_UUID	0x020	/* added long ago, but never used */
 #define	XFS_ILOG_ADATA	0x040	/* log i_af.if_data */
 #define	XFS_ILOG_AEXT	0x080	/* log i_af.if_extents */
 #define	XFS_ILOG_ABROOT	0x100	/* log i_af.i_broot */
@@ -329,9 +329,9 @@ struct xfs_inode_log_format_32 {
 
 #define	XFS_ILOG_NONCORE	(XFS_ILOG_DDATA | XFS_ILOG_DEXT | \
 				 XFS_ILOG_DBROOT | XFS_ILOG_DEV | \
-				 XFS_ILOG_UUID | XFS_ILOG_ADATA | \
-				 XFS_ILOG_AEXT | XFS_ILOG_ABROOT | \
-				 XFS_ILOG_DOWNER | XFS_ILOG_AOWNER)
+				 XFS_ILOG_ADATA | XFS_ILOG_AEXT | \
+				 XFS_ILOG_ABROOT | XFS_ILOG_DOWNER | \
+				 XFS_ILOG_AOWNER)
 
 #define	XFS_ILOG_DFORK		(XFS_ILOG_DDATA | XFS_ILOG_DEXT | \
 				 XFS_ILOG_DBROOT)
@@ -341,10 +341,10 @@ struct xfs_inode_log_format_32 {
 
 #define	XFS_ILOG_ALL		(XFS_ILOG_CORE | XFS_ILOG_DDATA | \
 				 XFS_ILOG_DEXT | XFS_ILOG_DBROOT | \
-				 XFS_ILOG_DEV | XFS_ILOG_UUID | \
-				 XFS_ILOG_ADATA | XFS_ILOG_AEXT | \
-				 XFS_ILOG_ABROOT | XFS_ILOG_TIMESTAMP | \
-				 XFS_ILOG_DOWNER | XFS_ILOG_AOWNER)
+				 XFS_ILOG_DEV | XFS_ILOG_ADATA | \
+				 XFS_ILOG_AEXT | XFS_ILOG_ABROOT | \
+				 XFS_ILOG_TIMESTAMP | XFS_ILOG_DOWNER | \
+				 XFS_ILOG_AOWNER)
 
 static inline int xfs_ilog_fbroot(int w)
 {

fs/xfs/libxfs/xfs_refcount.c

@@ -30,6 +30,7 @@
 #include "xfs_bmap.h"
 #include "xfs_refcount_btree.h"
 #include "xfs_alloc.h"
+#include "xfs_errortag.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
 #include "xfs_cksum.h"

fs/xfs/libxfs/xfs_rmap.c

@@ -34,6 +34,7 @@
 #include "xfs_rmap_btree.h"
 #include "xfs_trans_space.h"
 #include "xfs_trace.h"
+#include "xfs_errortag.h"
 #include "xfs_error.h"
 #include "xfs_extent_busy.h"
 #include "xfs_bmap.h"

fs/xfs/libxfs/xfs_rtbitmap.c

@@ -672,7 +672,6 @@ xfs_rtmodify_range(
 		/*
 		 * Compute a mask of relevant bits.
 		 */
-		bit = 0;
 		mask = ((xfs_rtword_t)1 << lastbit) - 1;
 		/*
 		 * Set/clear the active bits.
@@ -1086,3 +1085,15 @@ xfs_rtalloc_query_all(
 
 	return xfs_rtalloc_query_range(tp, &keys[0], &keys[1], fn, priv);
 }
+
+/*
+ * Verify that an realtime block number pointer doesn't point off the
+ * end of the realtime device.
+ */
+bool
+xfs_verify_rtbno(
+	struct xfs_mount	*mp,
+	xfs_rtblock_t		rtbno)
+{
+	return rtbno < mp->m_sb.sb_rblocks;
+}

fs/xfs/libxfs/xfs_types.h

@@ -47,6 +47,12 @@ typedef uint64_t	xfs_filblks_t;	/* number of blocks in a file */
 typedef int64_t		xfs_srtblock_t;	/* signed version of xfs_rtblock_t */
 typedef int64_t		xfs_sfiloff_t;	/* signed block number in a file */
 
+/*
+ * New verifiers will return the instruction address of the failing check.
+ * NULL means everything is ok.
+ */
+typedef void *		xfs_failaddr_t;
+
 /*
  * Null values for the types.
  */
@@ -136,5 +142,21 @@ typedef uint32_t	xfs_dqid_t;
 #define	XFS_NBWORD	(1 << XFS_NBWORDLOG)
 #define	XFS_WORDMASK	((1 << XFS_WORDLOG) - 1)
 
+struct xfs_iext_cursor {
+	struct xfs_iext_leaf	*leaf;
+	int			pos;
+};
+
+typedef enum {
+	XFS_EXT_NORM, XFS_EXT_UNWRITTEN,
+} xfs_exntst_t;
+
+typedef struct xfs_bmbt_irec
+{
+	xfs_fileoff_t	br_startoff;	/* starting file offset */
+	xfs_fsblock_t	br_startblock;	/* starting block number */
+	xfs_filblks_t	br_blockcount;	/* number of blocks */
+	xfs_exntst_t	br_state;	/* extent state */
+} xfs_bmbt_irec_t;
 
 #endif	/* __XFS_TYPES_H__ */