Merge tag 'xfs-4.12-merge-7' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux

Pull xfs updates from Darrick Wong:
 "Here are the XFS changes for 4.12. The big new feature for this
  release is the new space mapping ioctl that we've been discussing
  since LSF2016, but other than that most of the patches are larger bug
  fixes, memory corruption prevention, and other cleanups.

  Summary:
   - various code cleanups
   - introduce GETFSMAP ioctl
   - various refactoring
   - avoid dio reads past eof
   - fix memory corruption and other errors with fragmented directory blocks
   - fix accidental userspace memory corruptions
   - publish fs uuid in superblock
   - make fstrim terminatable
   - fix race between quotaoff and in-core inode creation
   - avoid use-after-free when finishing up w/ buffer heads
   - reserve enough space to handle bmap tree resizing during cow remap"

* tag 'xfs-4.12-merge-7' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux: (53 commits)
  xfs: fix use-after-free in xfs_finish_page_writeback
  xfs: reserve enough blocks to handle btree splits when remapping
  xfs: wait on new inodes during quotaoff dquot release
  xfs: update ag iterator to support wait on new inodes
  xfs: support ability to wait on new inodes
  xfs: publish UUID in struct super_block
  xfs: Allow user to kill fstrim process
  xfs: better log intent item refcount checking
  xfs: fix up quotacheck buffer list error handling
  xfs: remove xfs_trans_ail_delete_bulk
  xfs: don't use bool values in trace buffers
  xfs: fix getfsmap userspace memory corruption while setting OF_LAST
  xfs: fix __user annotations for xfs_ioc_getfsmap
  xfs: corruption needs to respect endianess too!
  xfs: use NULL instead of 0 to initialize a pointer in xfs_ioc_getfsmap
  xfs: use NULL instead of 0 to initialize a pointer in xfs_getfsmap
  xfs: simplify validation of the unwritten extent bit
  xfs: remove unused values from xfs_exntst_t
  xfs: remove the unused XFS_MAXLINK_1 define
  xfs: more do_div cleanups
  ...

18 files changed, 584 insertions, 383 deletions

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 369adcc18c02..7486401ccbd3 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -2868,3 +2868,60 @@ err:
 	xfs_trans_brelse(tp, agbp);
 	return error;
 }
+
+struct xfs_alloc_query_range_info {
+	xfs_alloc_query_range_fn	fn;
+	void				*priv;
+};
+
+/* Format btree record and pass to our callback. */
+STATIC int
+xfs_alloc_query_range_helper(
+	struct xfs_btree_cur		*cur,
+	union xfs_btree_rec		*rec,
+	void				*priv)
+{
+	struct xfs_alloc_query_range_info	*query = priv;
+	struct xfs_alloc_rec_incore		irec;
+
+	irec.ar_startblock = be32_to_cpu(rec->alloc.ar_startblock);
+	irec.ar_blockcount = be32_to_cpu(rec->alloc.ar_blockcount);
+	return query->fn(cur, &irec, query->priv);
+}
+
+/* Find all free space within a given range of blocks. */
+int
+xfs_alloc_query_range(
+	struct xfs_btree_cur			*cur,
+	struct xfs_alloc_rec_incore		*low_rec,
+	struct xfs_alloc_rec_incore		*high_rec,
+	xfs_alloc_query_range_fn		fn,
+	void					*priv)
+{
+	union xfs_btree_irec			low_brec;
+	union xfs_btree_irec			high_brec;
+	struct xfs_alloc_query_range_info	query;
+
+	ASSERT(cur->bc_btnum == XFS_BTNUM_BNO);
+	low_brec.a = *low_rec;
+	high_brec.a = *high_rec;
+	query.priv = priv;
+	query.fn = fn;
+	return xfs_btree_query_range(cur, &low_brec, &high_brec,
+			xfs_alloc_query_range_helper, &query);
+}
+
+/* Find all free space records. */
+int
+xfs_alloc_query_all(
+	struct xfs_btree_cur			*cur,
+	xfs_alloc_query_range_fn		fn,
+	void					*priv)
+{
+	struct xfs_alloc_query_range_info	query;
+
+	ASSERT(cur->bc_btnum == XFS_BTNUM_BNO);
+	query.priv = priv;
+	query.fn = fn;
+	return xfs_btree_query_all(cur, xfs_alloc_query_range_helper, &query);
+}
diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h
index 2a8d0fa6fbbe..77d9c27330ab 100644
--- a/fs/xfs/libxfs/xfs_alloc.h
+++ b/fs/xfs/libxfs/xfs_alloc.h
@@ -219,4 +219,16 @@ int xfs_free_extent_fix_freelist(struct xfs_trans *tp, xfs_agnumber_t agno,
 
 xfs_extlen_t xfs_prealloc_blocks(struct xfs_mount *mp);
 
+typedef int (*xfs_alloc_query_range_fn)(
+	struct xfs_btree_cur		*cur,
+	struct xfs_alloc_rec_incore	*rec,
+	void				*priv);
+
+int xfs_alloc_query_range(struct xfs_btree_cur *cur,
+		struct xfs_alloc_rec_incore *low_rec,
+		struct xfs_alloc_rec_incore *high_rec,
+		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);
+
 #endif	/* __XFS_ALLOC_H__ */
diff --git a/fs/xfs/libxfs/xfs_alloc_btree.c b/fs/xfs/libxfs/xfs_alloc_btree.c
index efb467b10a71..e1fcfe7f0a9a 100644
--- a/fs/xfs/libxfs/xfs_alloc_btree.c
+++ b/fs/xfs/libxfs/xfs_alloc_btree.c
@@ -205,19 +205,37 @@ xfs_allocbt_init_key_from_rec(
 	union xfs_btree_key	*key,
 	union xfs_btree_rec	*rec)
 {
-	ASSERT(rec->alloc.ar_startblock != 0);
-
 	key->alloc.ar_startblock = rec->alloc.ar_startblock;
 	key->alloc.ar_blockcount = rec->alloc.ar_blockcount;
 }
 
 STATIC void
+xfs_bnobt_init_high_key_from_rec(
+	union xfs_btree_key	*key,
+	union xfs_btree_rec	*rec)
+{
+	__u32			x;
+
+	x = be32_to_cpu(rec->alloc.ar_startblock);
+	x += be32_to_cpu(rec->alloc.ar_blockcount) - 1;
+	key->alloc.ar_startblock = cpu_to_be32(x);
+	key->alloc.ar_blockcount = 0;
+}
+
+STATIC void
+xfs_cntbt_init_high_key_from_rec(
+	union xfs_btree_key	*key,
+	union xfs_btree_rec	*rec)
+{
+	key->alloc.ar_blockcount = rec->alloc.ar_blockcount;
+	key->alloc.ar_startblock = 0;
+}
+
+STATIC void
 xfs_allocbt_init_rec_from_cur(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_rec	*rec)
 {
-	ASSERT(cur->bc_rec.a.ar_startblock != 0);
-
 	rec->alloc.ar_startblock = cpu_to_be32(cur->bc_rec.a.ar_startblock);
 	rec->alloc.ar_blockcount = cpu_to_be32(cur->bc_rec.a.ar_blockcount);
 }
@@ -236,18 +254,24 @@ xfs_allocbt_init_ptr_from_cur(
 }
 
 STATIC __int64_t
-xfs_allocbt_key_diff(
+xfs_bnobt_key_diff(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_key	*key)
 {
 	xfs_alloc_rec_incore_t	*rec = &cur->bc_rec.a;
 	xfs_alloc_key_t		*kp = &key->alloc;
-	__int64_t		diff;
 
-	if (cur->bc_btnum == XFS_BTNUM_BNO) {
-		return (__int64_t)be32_to_cpu(kp->ar_startblock) -
-				rec->ar_startblock;
-	}
+	return (__int64_t)be32_to_cpu(kp->ar_startblock) - rec->ar_startblock;
+}
+
+STATIC __int64_t
+xfs_cntbt_key_diff(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_key	*key)
+{
+	xfs_alloc_rec_incore_t	*rec = &cur->bc_rec.a;
+	xfs_alloc_key_t		*kp = &key->alloc;
+	__int64_t		diff;
 
 	diff = (__int64_t)be32_to_cpu(kp->ar_blockcount) - rec->ar_blockcount;
 	if (diff)
@@ -256,6 +280,33 @@ xfs_allocbt_key_diff(
 	return (__int64_t)be32_to_cpu(kp->ar_startblock) - rec->ar_startblock;
 }
 
+STATIC __int64_t
+xfs_bnobt_diff_two_keys(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_key	*k1,
+	union xfs_btree_key	*k2)
+{
+	return (__int64_t)be32_to_cpu(k1->alloc.ar_startblock) -
+			  be32_to_cpu(k2->alloc.ar_startblock);
+}
+
+STATIC __int64_t
+xfs_cntbt_diff_two_keys(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_key	*k1,
+	union xfs_btree_key	*k2)
+{
+	__int64_t		diff;
+
+	diff =  be32_to_cpu(k1->alloc.ar_blockcount) -
+		be32_to_cpu(k2->alloc.ar_blockcount);
+	if (diff)
+		return diff;
+
+	return  be32_to_cpu(k1->alloc.ar_startblock) -
+		be32_to_cpu(k2->alloc.ar_startblock);
+}
+
 static bool
 xfs_allocbt_verify(
 	struct xfs_buf		*bp)
@@ -346,44 +397,54 @@ const struct xfs_buf_ops xfs_allocbt_buf_ops = {
 
 #if defined(DEBUG) || defined(XFS_WARN)
 STATIC int
-xfs_allocbt_keys_inorder(
+xfs_bnobt_keys_inorder(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_key	*k1,
 	union xfs_btree_key	*k2)
 {
-	if (cur->bc_btnum == XFS_BTNUM_BNO) {
-		return be32_to_cpu(k1->alloc.ar_startblock) <
-		       be32_to_cpu(k2->alloc.ar_startblock);
-	} else {
-		return be32_to_cpu(k1->alloc.ar_blockcount) <
-			be32_to_cpu(k2->alloc.ar_blockcount) ||
-			(k1->alloc.ar_blockcount == k2->alloc.ar_blockcount &&
-			 be32_to_cpu(k1->alloc.ar_startblock) <
-			 be32_to_cpu(k2->alloc.ar_startblock));
-	}
+	return be32_to_cpu(k1->alloc.ar_startblock) <
+	       be32_to_cpu(k2->alloc.ar_startblock);
 }
 
 STATIC int
-xfs_allocbt_recs_inorder(
+xfs_bnobt_recs_inorder(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_rec	*r1,
 	union xfs_btree_rec	*r2)
 {
-	if (cur->bc_btnum == XFS_BTNUM_BNO) {
-		return be32_to_cpu(r1->alloc.ar_startblock) +
-			be32_to_cpu(r1->alloc.ar_blockcount) <=
-			be32_to_cpu(r2->alloc.ar_startblock);
-	} else {
-		return be32_to_cpu(r1->alloc.ar_blockcount) <
-			be32_to_cpu(r2->alloc.ar_blockcount) ||
-			(r1->alloc.ar_blockcount == r2->alloc.ar_blockcount &&
-			 be32_to_cpu(r1->alloc.ar_startblock) <
-			 be32_to_cpu(r2->alloc.ar_startblock));
-	}
+	return be32_to_cpu(r1->alloc.ar_startblock) +
+		be32_to_cpu(r1->alloc.ar_blockcount) <=
+		be32_to_cpu(r2->alloc.ar_startblock);
+}
+
+STATIC int
+xfs_cntbt_keys_inorder(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_key	*k1,
+	union xfs_btree_key	*k2)
+{
+	return be32_to_cpu(k1->alloc.ar_blockcount) <
+		be32_to_cpu(k2->alloc.ar_blockcount) ||
+		(k1->alloc.ar_blockcount == k2->alloc.ar_blockcount &&
+		 be32_to_cpu(k1->alloc.ar_startblock) <
+		 be32_to_cpu(k2->alloc.ar_startblock));
 }
-#endif	/* DEBUG */
 
-static const struct xfs_btree_ops xfs_allocbt_ops = {
+STATIC int
+xfs_cntbt_recs_inorder(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_rec	*r1,
+	union xfs_btree_rec	*r2)
+{
+	return be32_to_cpu(r1->alloc.ar_blockcount) <
+		be32_to_cpu(r2->alloc.ar_blockcount) ||
+		(r1->alloc.ar_blockcount == r2->alloc.ar_blockcount &&
+		 be32_to_cpu(r1->alloc.ar_startblock) <
+		 be32_to_cpu(r2->alloc.ar_startblock));
+}
+#endif /* DEBUG */
+
+static const struct xfs_btree_ops xfs_bnobt_ops = {
 	.rec_len		= sizeof(xfs_alloc_rec_t),
 	.key_len		= sizeof(xfs_alloc_key_t),
 
@@ -395,13 +456,39 @@ static const struct xfs_btree_ops xfs_allocbt_ops = {
 	.get_minrecs		= xfs_allocbt_get_minrecs,
 	.get_maxrecs		= xfs_allocbt_get_maxrecs,
 	.init_key_from_rec	= xfs_allocbt_init_key_from_rec,
+	.init_high_key_from_rec	= xfs_bnobt_init_high_key_from_rec,
 	.init_rec_from_cur	= xfs_allocbt_init_rec_from_cur,
 	.init_ptr_from_cur	= xfs_allocbt_init_ptr_from_cur,
-	.key_diff		= xfs_allocbt_key_diff,
+	.key_diff		= xfs_bnobt_key_diff,
 	.buf_ops		= &xfs_allocbt_buf_ops,
+	.diff_two_keys		= xfs_bnobt_diff_two_keys,
 #if defined(DEBUG) || defined(XFS_WARN)
-	.keys_inorder		= xfs_allocbt_keys_inorder,
-	.recs_inorder		= xfs_allocbt_recs_inorder,
+	.keys_inorder		= xfs_bnobt_keys_inorder,
+	.recs_inorder		= xfs_bnobt_recs_inorder,
+#endif
+};
+
+static const struct xfs_btree_ops xfs_cntbt_ops = {
+	.rec_len		= sizeof(xfs_alloc_rec_t),
+	.key_len		= sizeof(xfs_alloc_key_t),
+
+	.dup_cursor		= xfs_allocbt_dup_cursor,
+	.set_root		= xfs_allocbt_set_root,
+	.alloc_block		= xfs_allocbt_alloc_block,
+	.free_block		= xfs_allocbt_free_block,
+	.update_lastrec		= xfs_allocbt_update_lastrec,
+	.get_minrecs		= xfs_allocbt_get_minrecs,
+	.get_maxrecs		= xfs_allocbt_get_maxrecs,
+	.init_key_from_rec	= xfs_allocbt_init_key_from_rec,
+	.init_high_key_from_rec	= xfs_cntbt_init_high_key_from_rec,
+	.init_rec_from_cur	= xfs_allocbt_init_rec_from_cur,
+	.init_ptr_from_cur	= xfs_allocbt_init_ptr_from_cur,
+	.key_diff		= xfs_cntbt_key_diff,
+	.buf_ops		= &xfs_allocbt_buf_ops,
+	.diff_two_keys		= xfs_cntbt_diff_two_keys,
+#if defined(DEBUG) || defined(XFS_WARN)
+	.keys_inorder		= xfs_cntbt_keys_inorder,
+	.recs_inorder		= xfs_cntbt_recs_inorder,
 #endif
 };
 
@@ -427,16 +514,15 @@ xfs_allocbt_init_cursor(
 	cur->bc_mp = mp;
 	cur->bc_btnum = btnum;
 	cur->bc_blocklog = mp->m_sb.sb_blocklog;
-	cur->bc_ops = &xfs_allocbt_ops;
-	if (btnum == XFS_BTNUM_BNO)
-		cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_abtb_2);
-	else
-		cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_abtc_2);
 
 	if (btnum == XFS_BTNUM_CNT) {
+		cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_abtc_2);
+		cur->bc_ops = &xfs_cntbt_ops;
 		cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]);
 		cur->bc_flags = XFS_BTREE_LASTREC_UPDATE;
 	} else {
+		cur->bc_statoff = XFS_STATS_CALC_INDEX(xs_abtb_2);
+		cur->bc_ops = &xfs_bnobt_ops;
 		cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]);
 	}
 
diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 9bd104f32908..f02eb7673392 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -764,7 +764,6 @@ xfs_bmap_extents_to_btree(
 		args.fsbno = XFS_INO_TO_FSB(mp, ip->i_ino);
 	} else if (dfops->dop_low) {
 		args.type = XFS_ALLOCTYPE_START_BNO;
-try_another_ag:
 		args.fsbno = *firstblock;
 	} else {
 		args.type = XFS_ALLOCTYPE_NEAR_BNO;
@@ -779,20 +778,6 @@ try_another_ag:
 		return error;
 	}
 
-	/*
-	 * During a CoW operation, the allocation and bmbt updates occur in
-	 * different transactions.  The mapping code tries to put new bmbt
-	 * blocks near extents being mapped, but the only way to guarantee this
-	 * is if the alloc and the mapping happen in a single transaction that
-	 * has a block reservation.  That isn't the case here, so if we run out
-	 * of space we'll try again with another AG.
-	 */
-	if (xfs_sb_version_hasreflink(&cur->bc_mp->m_sb) &&
-	    args.fsbno == NULLFSBLOCK &&
-	    args.type == XFS_ALLOCTYPE_NEAR_BNO) {
-		args.type = XFS_ALLOCTYPE_FIRST_AG;
-		goto try_another_ag;
-	}
 	if (WARN_ON_ONCE(args.fsbno == NULLFSBLOCK)) {
 		xfs_iroot_realloc(ip, -1, whichfork);
 		xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
@@ -925,7 +910,6 @@ xfs_bmap_local_to_extents(
 	 * file currently fits in an inode.
 	 */
 	if (*firstblock == NULLFSBLOCK) {
-try_another_ag:
 		args.fsbno = XFS_INO_TO_FSB(args.mp, ip->i_ino);
 		args.type = XFS_ALLOCTYPE_START_BNO;
 	} else {
@@ -938,19 +922,6 @@ try_another_ag:
 	if (error)
 		goto done;
 
-	/*
-	 * During a CoW operation, the allocation and bmbt updates occur in
-	 * different transactions.  The mapping code tries to put new bmbt
-	 * blocks near extents being mapped, but the only way to guarantee this
-	 * is if the alloc and the mapping happen in a single transaction that
-	 * has a block reservation.  That isn't the case here, so if we run out
-	 * of space we'll try again with another AG.
-	 */
-	if (xfs_sb_version_hasreflink(&ip->i_mount->m_sb) &&
-	    args.fsbno == NULLFSBLOCK &&
-	    args.type == XFS_ALLOCTYPE_NEAR_BNO) {
-		goto try_another_ag;
-	}
 	/* Can't fail, the space was reserved. */
 	ASSERT(args.fsbno != NULLFSBLOCK);
 	ASSERT(args.len == 1);
@@ -1260,7 +1231,6 @@ xfs_bmap_read_extents(
 	xfs_fsblock_t		bno;	/* block # of "block" */
 	xfs_buf_t		*bp;	/* buffer for "block" */
 	int			error;	/* error return value */
-	xfs_exntfmt_t		exntf;	/* XFS_EXTFMT_NOSTATE, if checking */
 	xfs_extnum_t		i, j;	/* index into the extents list */
 	xfs_ifork_t		*ifp;	/* fork structure */
 	int			level;	/* btree level, for checking */
@@ -1271,8 +1241,6 @@ xfs_bmap_read_extents(
 
 	mp = ip->i_mount;
 	ifp = XFS_IFORK_PTR(ip, whichfork);
-	exntf = (whichfork != XFS_DATA_FORK) ? XFS_EXTFMT_NOSTATE :
-					XFS_EXTFMT_INODE(ip);
 	block = ifp->if_broot;
 	/*
 	 * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out.
@@ -1340,18 +1308,9 @@ xfs_bmap_read_extents(
 			xfs_bmbt_rec_host_t *trp = xfs_iext_get_ext(ifp, i);
 			trp->l0 = be64_to_cpu(frp->l0);
 			trp->l1 = be64_to_cpu(frp->l1);
-		}
-		if (exntf == XFS_EXTFMT_NOSTATE) {
-			/*
-			 * Check all attribute bmap btree records and
-			 * any "older" data bmap btree records for a
-			 * set bit in the "extent flag" position.
-			 */
-			if (unlikely(xfs_check_nostate_extents(ifp,
-					start, num_recs))) {
+			if (!xfs_bmbt_validate_extent(mp, whichfork, trp)) {
 				XFS_ERROR_REPORT("xfs_bmap_read_extents(2)",
-						 XFS_ERRLEVEL_LOW,
-						 ip->i_mount);
+						 XFS_ERRLEVEL_LOW, mp);
 				goto error0;
 			}
 		}
@@ -2879,27 +2838,30 @@ xfs_bmap_add_extent_hole_delay(
  */
 STATIC int				/* error */
 xfs_bmap_add_extent_hole_real(
-	struct xfs_bmalloca	*bma,
-	int			whichfork)
+	struct xfs_trans	*tp,
+	struct xfs_inode	*ip,
+	int			whichfork,
+	xfs_extnum_t		*idx,
+	struct xfs_btree_cur	**curp,
+	struct xfs_bmbt_irec	*new,
+	xfs_fsblock_t		*first,
+	struct xfs_defer_ops	*dfops,
+	int			*logflagsp)
 {
-	struct xfs_bmbt_irec	*new = &bma->got;
+	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_btree_cur	*cur = *curp;
 	int			error;	/* error return value */
 	int			i;	/* temp state */
-	xfs_ifork_t		*ifp;	/* inode fork pointer */
 	xfs_bmbt_irec_t		left;	/* left neighbor extent entry */
 	xfs_bmbt_irec_t		right;	/* right neighbor extent entry */
 	int			rval=0;	/* return value (logging flags) */
 	int			state;	/* state bits, accessed thru macros */
-	struct xfs_mount	*mp;
 
-	mp = bma->ip->i_mount;
-	ifp = XFS_IFORK_PTR(bma->ip, whichfork);
-
-	ASSERT(bma->idx >= 0);
-	ASSERT(bma->idx <= xfs_iext_count(ifp));
+	ASSERT(*idx >= 0);
+	ASSERT(*idx <= xfs_iext_count(ifp));
 	ASSERT(!isnullstartblock(new->br_startblock));
-	ASSERT(!bma->cur ||
-	       !(bma->cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
+	ASSERT(!cur || !(cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
 
 	XFS_STATS_INC(mp, xs_add_exlist);
 
@@ -2912,9 +2874,9 @@ xfs_bmap_add_extent_hole_real(
 	/*
 	 * Check and set flags if this segment has a left neighbor.
 	 */
-	if (bma->idx > 0) {
+	if (*idx > 0) {
 		state |= BMAP_LEFT_VALID;
-		xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx - 1), &left);
+		xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx - 1), &left);
 		if (isnullstartblock(left.br_startblock))
 			state |= BMAP_LEFT_DELAY;
 	}
@@ -2923,9 +2885,9 @@ xfs_bmap_add_extent_hole_real(
 	 * Check and set flags if this segment has a current value.
 	 * Not true if we're inserting into the "hole" at eof.
 	 */
-	if (bma->idx < xfs_iext_count(ifp)) {
+	if (*idx < xfs_iext_count(ifp)) {
 		state |= BMAP_RIGHT_VALID;
-		xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx), &right);
+		xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx), &right);
 		if (isnullstartblock(right.br_startblock))
 			state |= BMAP_RIGHT_DELAY;
 	}
@@ -2962,36 +2924,36 @@ xfs_bmap_add_extent_hole_real(
 		 * left and on the right.
 		 * Merge all three into a single extent record.
 		 */
-		--bma->idx;
-		trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
-		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx),
+		--*idx;
+		trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx),
 			left.br_blockcount + new->br_blockcount +
 			right.br_blockcount);
-		trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+		trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
 
-		xfs_iext_remove(bma->ip, bma->idx + 1, 1, state);
+		xfs_iext_remove(ip, *idx + 1, 1, state);
 
-		XFS_IFORK_NEXT_SET(bma->ip, whichfork,
-			XFS_IFORK_NEXTENTS(bma->ip, whichfork) - 1);
-		if (bma->cur == NULL) {
+		XFS_IFORK_NEXT_SET(ip, whichfork,
+			XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
+		if (cur == NULL) {
 			rval = XFS_ILOG_CORE | xfs_ilog_fext(whichfork);
 		} else {
 			rval = XFS_ILOG_CORE;
-			error = xfs_bmbt_lookup_eq(bma->cur, right.br_startoff,
+			error = xfs_bmbt_lookup_eq(cur, right.br_startoff,
 					right.br_startblock, right.br_blockcount,
 					&i);
 			if (error)
 				goto done;
 			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
-			error = xfs_btree_delete(bma->cur, &i);
+			error = xfs_btree_delete(cur, &i);
 			if (error)
 				goto done;
 			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
-			error = xfs_btree_decrement(bma->cur, 0, &i);
+			error = xfs_btree_decrement(cur, 0, &i);
 			if (error)
 				goto done;
 			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
-			error = xfs_bmbt_update(bma->cur, left.br_startoff,
+			error = xfs_bmbt_update(cur, left.br_startoff,
 					left.br_startblock,
 					left.br_blockcount +
 						new->br_blockcount +
@@ -3008,23 +2970,23 @@ xfs_bmap_add_extent_hole_real(
 		 * on the left.
 		 * Merge the new allocation with the left neighbor.
 		 */
-		--bma->idx;
-		trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
-		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, bma->idx),
+		--*idx;
+		trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx),
 			left.br_blockcount + new->br_blockcount);
-		trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+		trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
 
-		if (bma->cur == NULL) {
+		if (cur == NULL) {
 			rval = xfs_ilog_fext(whichfork);
 		} else {
 			rval = 0;
-			error = xfs_bmbt_lookup_eq(bma->cur, left.br_startoff,
+			error = xfs_bmbt_lookup_eq(cur, left.br_startoff,
 					left.br_startblock, left.br_blockcount,
 					&i);
 			if (error)
 				goto done;
 			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
-			error = xfs_bmbt_update(bma->cur, left.br_startoff,
+			error = xfs_bmbt_update(cur, left.br_startoff,
 					left.br_startblock,
 					left.br_blockcount +
 						new->br_blockcount,
@@ -3040,25 +3002,25 @@ xfs_bmap_add_extent_hole_real(
 		 * on the right.
 		 * Merge the new allocation with the right neighbor.
 		 */
-		trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
-		xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, bma->idx),
+		trace_xfs_bmap_pre_update(ip, *idx, state, _THIS_IP_);
+		xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, *idx),
 			new->br_startoff, new->br_startblock,
 			new->br_blockcount + right.br_blockcount,
 			right.br_state);
-		trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
+		trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
 
-		if (bma->cur == NULL) {
+		if (cur == NULL) {
 			rval = xfs_ilog_fext(whichfork);
 		} else {
 			rval = 0;
-			error = xfs_bmbt_lookup_eq(bma->cur,
+			error = xfs_bmbt_lookup_eq(cur,
 					right.br_startoff,
 					right.br_startblock,
 					right.br_blockcount, &i);
 			if (error)
 				goto done;
 			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
-			error = xfs_bmbt_update(bma->cur, new->br_startoff,
+			error = xfs_bmbt_update(cur, new->br_startoff,
 					new->br_startblock,
 					new->br_blockcount +
 						right.br_blockcount,
@@ -3074,22 +3036,22 @@ xfs_bmap_add_extent_hole_real(
 		 * real allocation.
 		 * Insert a new entry.
 		 */
-		xfs_iext_insert(bma->ip, bma->idx, 1, new, state);
-		XFS_IFORK_NEXT_SET(bma->ip, whichfork,
-			XFS_IFORK_NEXTENTS(bma->ip, whichfork) + 1);
-		if (bma->cur == NULL) {
+		xfs_iext_insert(ip, *idx, 1, new, state);
+		XFS_IFORK_NEXT_SET(ip, whichfork,
+			XFS_IFORK_NEXTENTS(ip, whichfork) + 1);
+		if (cur == NULL) {
 			rval = XFS_ILOG_CORE | xfs_ilog_fext(whichfork);
 		} else {
 			rval = XFS_ILOG_CORE;
-			error = xfs_bmbt_lookup_eq(bma->cur,
+			error = xfs_bmbt_lookup_eq(cur,
 					new->br_startoff,
 					new->br_startblock,
 					new->br_blockcount, &i);
 			if (error)
 				goto done;
 			XFS_WANT_CORRUPTED_GOTO(mp, i == 0, done);
-			bma->cur->bc_rec.b.br_state = new->br_state;
-			error = xfs_btree_insert(bma->cur, &i);
+			cur->bc_rec.b.br_state = new->br_state;
+			error = xfs_btree_insert(cur, &i);
 			if (error)
 				goto done;
 			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
@@ -3098,30 +3060,30 @@ xfs_bmap_add_extent_hole_real(
 	}
 
 	/* add reverse mapping */
-	error = xfs_rmap_map_extent(mp, bma->dfops, bma->ip, whichfork, new);
+	error = xfs_rmap_map_extent(mp, dfops, ip, whichfork, new);
 	if (error)
 		goto done;
 
 	/* convert to a btree if necessary */
-	if (xfs_bmap_needs_btree(bma->ip, whichfork)) {
+	if (xfs_bmap_needs_btree(ip, whichfork)) {
 		int	tmp_logflags;	/* partial log flag return val */
 
-		ASSERT(bma->cur == NULL);
-		error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
-				bma->firstblock, bma->dfops, &bma->cur,
+		ASSERT(cur == NULL);
+		error = xfs_bmap_extents_to_btree(tp, ip, first, dfops, curp,
 				0, &tmp_logflags, whichfork);
-		bma->logflags |= tmp_logflags;
+		*logflagsp |= tmp_logflags;
+		cur = *curp;
 		if (error)
 			goto done;
 	}
 
 	/* clear out the allocated field, done with it now in any case. */
-	if (bma->cur)
-		bma->cur->bc_private.b.allocated = 0;
+	if (cur)
+		cur->bc_private.b.allocated = 0;
 
-	xfs_bmap_check_leaf_extents(bma->cur, bma->ip, whichfork);
+	xfs_bmap_check_leaf_extents(cur, ip, whichfork);
 done:
-	bma->logflags |= rval;
+	*logflagsp |= rval;
 	return error;
 }
 
@@ -3853,60 +3815,6 @@ xfs_bmap_btalloc(
 }
 
 /*
- * For a remap operation, just "allocate" an extent at the address that the
- * caller passed in, and ensure that the AGFL is the right size.  The caller
- * will then map the "allocated" extent into the file somewhere.
- */
-STATIC int
-xfs_bmap_remap_alloc(
-	struct xfs_bmalloca	*ap)
-{
-	struct xfs_trans	*tp = ap->tp;
-	struct xfs_mount	*mp = tp->t_mountp;
-	xfs_agblock_t		bno;
-	struct xfs_alloc_arg	args;
-	int			error;
-
-	/*
-	 * validate that the block number is legal - the enables us to detect
-	 * and handle a silent filesystem corruption rather than crashing.
-	 */
-	memset(&args, 0, sizeof(struct xfs_alloc_arg));
-	args.tp = ap->tp;
-	args.mp = ap->tp->t_mountp;
-	bno = *ap->firstblock;
-	args.agno = XFS_FSB_TO_AGNO(mp, bno);
-	args.agbno = XFS_FSB_TO_AGBNO(mp, bno);
-	if (args.agno >= mp->m_sb.sb_agcount ||
-	    args.agbno >= mp->m_sb.sb_agblocks)
-		return -EFSCORRUPTED;
-
-	/* "Allocate" the extent from the range we passed in. */
-	trace_xfs_bmap_remap_alloc(ap->ip, *ap->firstblock, ap->length);
-	ap->blkno = bno;
-	ap->ip->i_d.di_nblocks += ap->length;
-	xfs_trans_log_inode(ap->tp, ap->ip, XFS_ILOG_CORE);
-
-	/* Fix the freelist, like a real allocator does. */
-	args.datatype = ap->datatype;
-	args.pag = xfs_perag_get(args.mp, args.agno);
-	ASSERT(args.pag);
-
-	/*
-	 * The freelist fixing code will decline the allocation if
-	 * the size and shape of the free space doesn't allow for
-	 * allocating the extent and updating all the metadata that
-	 * happens during an allocation.  We're remapping, not
-	 * allocating, so skip that check by pretending to be freeing.
-	 */
-	error = xfs_alloc_fix_freelist(&args, XFS_ALLOC_FLAG_FREEING);
-	xfs_perag_put(args.pag);
-	if (error)
-		trace_xfs_bmap_remap_alloc_error(ap->ip, error, _RET_IP_);
-	return error;
-}
-
-/*
  * xfs_bmap_alloc is called by xfs_bmapi to allocate an extent for a file.
  * It figures out where to ask the underlying allocator to put the new extent.
  */
@@ -3914,8 +3822,6 @@ STATIC int
 xfs_bmap_alloc(
 	struct xfs_bmalloca	*ap)	/* bmap alloc argument struct */
 {
-	if (ap->flags & XFS_BMAPI_REMAP)
-		return xfs_bmap_remap_alloc(ap);
 	if (XFS_IS_REALTIME_INODE(ap->ip) &&
 	    xfs_alloc_is_userdata(ap->datatype))
 		return xfs_bmap_rtalloc(ap);
@@ -4386,7 +4292,9 @@ xfs_bmapi_allocate(
 	if (bma->wasdel)
 		error = xfs_bmap_add_extent_delay_real(bma, whichfork);
 	else
-		error = xfs_bmap_add_extent_hole_real(bma, whichfork);
+		error = xfs_bmap_add_extent_hole_real(bma->tp, bma->ip,
+				whichfork, &bma->idx, &bma->cur, &bma->got,
+				bma->firstblock, bma->dfops, &bma->logflags);
 
 	bma->logflags |= tmp_logflags;
 	if (error)
@@ -4549,9 +4457,7 @@ xfs_bmapi_write(
 	ASSERT(len > 0);
 	ASSERT(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_LOCAL);
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-	ASSERT(!(flags & XFS_BMAPI_REMAP) || whichfork == XFS_DATA_FORK);
-	ASSERT(!(flags & XFS_BMAPI_PREALLOC) || !(flags & XFS_BMAPI_REMAP));
-	ASSERT(!(flags & XFS_BMAPI_CONVERT) || !(flags & XFS_BMAPI_REMAP));
+	ASSERT(!(flags & XFS_BMAPI_REMAP));
 
 	/* zeroing is for currently only for data extents, not metadata */
 	ASSERT((flags & (XFS_BMAPI_METADATA | XFS_BMAPI_ZERO)) !=
@@ -4635,13 +4541,8 @@ xfs_bmapi_write(
 			} else {
 				need_alloc = true;
 			}
-		} else {
-			/*
-			 * Make sure we only reflink into a hole.
-			 */
-			ASSERT(!(flags & XFS_BMAPI_REMAP));
-			if (isnullstartblock(bma.got.br_startblock))
-				wasdelay = true;
+		} else if (isnullstartblock(bma.got.br_startblock)) {
+			wasdelay = true;
 		}
 
 		/*
@@ -4770,6 +4671,93 @@ error0:
 	return error;
 }
 
+static int
+xfs_bmapi_remap(
+	struct xfs_trans	*tp,
+	struct xfs_inode	*ip,
+	xfs_fileoff_t		bno,
+	xfs_filblks_t		len,
+	xfs_fsblock_t		startblock,
+	struct xfs_defer_ops	*dfops)
+{
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
+	struct xfs_btree_cur	*cur = NULL;
+	xfs_fsblock_t		firstblock = NULLFSBLOCK;
+	struct xfs_bmbt_irec	got;
+	xfs_extnum_t		idx;
+	int			logflags = 0, error;
+
+	ASSERT(len > 0);
+	ASSERT(len <= (xfs_filblks_t)MAXEXTLEN);
+	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
+
+	if (unlikely(XFS_TEST_ERROR(
+	    (XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_EXTENTS &&
+	     XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_BTREE),
+	     mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) {
+		XFS_ERROR_REPORT("xfs_bmapi_remap", XFS_ERRLEVEL_LOW, mp);
+		return -EFSCORRUPTED;
+	}
+
+	if (XFS_FORCED_SHUTDOWN(mp))
+		return -EIO;
+
+	if (!(ifp->if_flags & XFS_IFEXTENTS)) {
+		error = xfs_iread_extents(NULL, ip, XFS_DATA_FORK);
+		if (error)
+			return error;
+	}
+
+	if (xfs_iext_lookup_extent(ip, ifp, bno, &idx, &got)) {
+		/* make sure we only reflink into a hole. */
+		ASSERT(got.br_startoff > bno);
+		ASSERT(got.br_startoff - bno >= len);
+	}
+
+	ip->i_d.di_nblocks += len;
+	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+
+	if (ifp->if_flags & XFS_IFBROOT) {
+		cur = xfs_bmbt_init_cursor(mp, tp, ip, XFS_DATA_FORK);
+		cur->bc_private.b.firstblock = firstblock;
+		cur->bc_private.b.dfops = dfops;
+		cur->bc_private.b.flags = 0;
+	}
+
+	got.br_startoff = bno;
+	got.br_startblock = startblock;
+	got.br_blockcount = len;
+	got.br_state = XFS_EXT_NORM;
+
+	error = xfs_bmap_add_extent_hole_real(tp, ip, XFS_DATA_FORK, &idx, &cur,
+			&got, &firstblock, dfops, &logflags);
+	if (error)
+		goto error0;
+
+	if (xfs_bmap_wants_extents(ip, XFS_DATA_FORK)) {
+		int		tmp_logflags = 0;
+
+		error = xfs_bmap_btree_to_extents(tp, ip, cur,
+			&tmp_logflags, XFS_DATA_FORK);
+		logflags |= tmp_logflags;
+	}
+
+error0:
+	if (ip->i_d.di_format != XFS_DINODE_FMT_EXTENTS)
+		logflags &= ~XFS_ILOG_DEXT;
+	else if (ip->i_d.di_format != XFS_DINODE_FMT_BTREE)
+		logflags &= ~XFS_ILOG_DBROOT;
+
+	if (logflags)
+		xfs_trans_log_inode(tp, ip, logflags);
+	if (cur) {
+		xfs_btree_del_cursor(cur,
+				error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
+	}
+	return error;
+}
+
 /*
  * When a delalloc extent is split (e.g., due to a hole punch), the original
  * indlen reservation must be shared across the two new extents that are left
@@ -4887,7 +4875,7 @@ xfs_bmap_del_extent_delay(
 	ASSERT(got_endoff >= del_endoff);
 
 	if (isrt) {
-		int64_t rtexts = XFS_FSB_TO_B(mp, del->br_blockcount);
+		uint64_t rtexts = XFS_FSB_TO_B(mp, del->br_blockcount);
 
 		do_div(rtexts, mp->m_sb.sb_rextsize);
 		xfs_mod_frextents(mp, rtexts);
@@ -6488,27 +6476,15 @@ xfs_bmap_finish_one(
 	xfs_filblks_t			blockcount,
 	xfs_exntst_t			state)
 {
-	struct xfs_bmbt_irec		bmap;
-	int				nimaps = 1;
-	xfs_fsblock_t			firstfsb;
-	int				flags = XFS_BMAPI_REMAP;
-	int				done;
-	int				error = 0;
-
-	bmap.br_startblock = startblock;
-	bmap.br_startoff = startoff;
-	bmap.br_blockcount = blockcount;
-	bmap.br_state = state;
+	int				error = 0, done;
 
 	trace_xfs_bmap_deferred(tp->t_mountp,
 			XFS_FSB_TO_AGNO(tp->t_mountp, startblock), type,
 			XFS_FSB_TO_AGBNO(tp->t_mountp, startblock),
 			ip->i_ino, whichfork, startoff, blockcount, state);
 
-	if (whichfork != XFS_DATA_FORK && whichfork != XFS_ATTR_FORK)
+	if (WARN_ON_ONCE(whichfork != XFS_DATA_FORK))
 		return -EFSCORRUPTED;
-	if (whichfork == XFS_ATTR_FORK)
-		flags |= XFS_BMAPI_ATTRFORK;
 
 	if (XFS_TEST_ERROR(false, tp->t_mountp,
 			XFS_ERRTAG_BMAP_FINISH_ONE,
@@ -6517,16 +6493,12 @@ xfs_bmap_finish_one(
 
 	switch (type) {
 	case XFS_BMAP_MAP:
-		firstfsb = bmap.br_startblock;
-		error = xfs_bmapi_write(tp, ip, bmap.br_startoff,
-					bmap.br_blockcount, flags, &firstfsb,
-					bmap.br_blockcount, &bmap, &nimaps,
-					dfops);
+		error = xfs_bmapi_remap(tp, ip, startoff, blockcount,
+				startblock, dfops);
 		break;
 	case XFS_BMAP_UNMAP:
-		error = xfs_bunmapi(tp, ip, bmap.br_startoff,
-				bmap.br_blockcount, flags, 1, &firstfsb,
-				dfops, &done);
+		error = xfs_bunmapi(tp, ip, startoff, blockcount,
+				XFS_BMAPI_REMAP, 1, &startblock, dfops, &done);
 		ASSERT(done);
 		break;
 	default:
diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h
index cdef87db5262..c35a14fa1527 100644
--- a/fs/xfs/libxfs/xfs_bmap.h
+++ b/fs/xfs/libxfs/xfs_bmap.h
@@ -172,6 +172,18 @@ static inline int xfs_bmapi_whichfork(int bmapi_flags)
 
 
 /*
+ * Return true if the extent is a real, allocated extent, or false if it is  a
+ * delayed allocation, and unwritten extent or a hole.
+ */
+static inline bool xfs_bmap_is_real_extent(struct xfs_bmbt_irec *irec)
+{
+	return irec->br_state != XFS_EXT_UNWRITTEN &&
+		irec->br_startblock != HOLESTARTBLOCK &&
+		irec->br_startblock != DELAYSTARTBLOCK &&
+		!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,
@@ -232,8 +244,6 @@ int	xfs_bmap_del_extent_delay(struct xfs_inode *ip, int whichfork,
 		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);
-int	xfs_check_nostate_extents(struct xfs_ifork *ifp, xfs_extnum_t idx,
-		xfs_extnum_t num);
 uint	xfs_default_attroffset(struct xfs_inode *ip);
 int	xfs_bmap_shift_extents(struct xfs_trans *tp, struct xfs_inode *ip,
 		xfs_fileoff_t *next_fsb, xfs_fileoff_t offset_shift_fsb,
diff --git a/fs/xfs/libxfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c
index fd55db479385..6cba69aff077 100644
--- a/fs/xfs/libxfs/xfs_bmap_btree.c
+++ b/fs/xfs/libxfs/xfs_bmap_btree.c
@@ -366,32 +366,6 @@ xfs_bmbt_to_bmdr(
 	memcpy(tpp, fpp, sizeof(*fpp) * dmxr);
 }
 
-/*
- * Check extent records, which have just been read, for
- * any bit in the extent flag field. ASSERT on debug
- * kernels, as this condition should not occur.
- * Return an error condition (1) if any flags found,
- * otherwise return 0.
- */
-
-int
-xfs_check_nostate_extents(
-	xfs_ifork_t		*ifp,
-	xfs_extnum_t		idx,
-	xfs_extnum_t		num)
-{
-	for (; num > 0; num--, idx++) {
-		xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, idx);
-		if ((ep->l0 >>
-		     (64 - BMBT_EXNTFLAG_BITLEN)) != 0) {
-			ASSERT(0);
-			return 1;
-		}
-	}
-	return 0;
-}
-
-
 STATIC struct xfs_btree_cur *
 xfs_bmbt_dup_cursor(
 	struct xfs_btree_cur	*cur)
@@ -448,7 +422,6 @@ xfs_bmbt_alloc_block(
 	if (args.fsbno == NULLFSBLOCK) {
 		args.fsbno = be64_to_cpu(start->l);
 		args.type = XFS_ALLOCTYPE_START_BNO;
-try_another_ag:
 		/*
 		 * Make sure there is sufficient room left in the AG to
 		 * complete a full tree split for an extent insert.  If
@@ -477,22 +450,6 @@ try_another_ag:
 	if (error)
 		goto error0;
 
-	/*
-	 * During a CoW operation, the allocation and bmbt updates occur in
-	 * different transactions.  The mapping code tries to put new bmbt
-	 * blocks near extents being mapped, but the only way to guarantee this
-	 * is if the alloc and the mapping happen in a single transaction that
-	 * has a block reservation.  That isn't the case here, so if we run out
-	 * of space we'll try again with another AG.
-	 */
-	if (xfs_sb_version_hasreflink(&cur->bc_mp->m_sb) &&
-	    args.fsbno == NULLFSBLOCK &&
-	    args.type == XFS_ALLOCTYPE_NEAR_BNO) {
-		args.fsbno = cur->bc_private.b.firstblock;
-		args.type = XFS_ALLOCTYPE_FIRST_AG;
-		goto try_another_ag;
-	}
-
 	if (args.fsbno == NULLFSBLOCK && args.minleft) {
 		/*
 		 * Could not find an AG with enough free space to satisfy
diff --git a/fs/xfs/libxfs/xfs_bmap_btree.h b/fs/xfs/libxfs/xfs_bmap_btree.h
index 819a8a4dee95..9da5a8d4f184 100644
--- a/fs/xfs/libxfs/xfs_bmap_btree.h
+++ b/fs/xfs/libxfs/xfs_bmap_btree.h
@@ -25,14 +25,6 @@ struct xfs_inode;
 struct xfs_trans;
 
 /*
- * Extent state and extent format macros.
- */
-#define XFS_EXTFMT_INODE(x)	\
-	(xfs_sb_version_hasextflgbit(&((x)->i_mount->m_sb)) ? \
-		XFS_EXTFMT_HASSTATE : XFS_EXTFMT_NOSTATE)
-#define ISUNWRITTEN(x)	((x)->br_state == XFS_EXT_UNWRITTEN)
-
-/*
  * Btree block header size depends on a superblock flag.
  */
 #define XFS_BMBT_BLOCK_LEN(mp) \
@@ -140,4 +132,18 @@ extern int xfs_bmbt_change_owner(struct xfs_trans *tp, struct xfs_inode *ip,
 extern struct xfs_btree_cur *xfs_bmbt_init_cursor(struct xfs_mount *,
 		struct xfs_trans *, struct xfs_inode *, int);
 
+/*
+ * 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)
+{
+	if (ep->l0 >> (64 - BMBT_EXNTFLAG_BITLEN) == 0)
+		return true;
+	if (whichfork == XFS_DATA_FORK &&
+	    xfs_sb_version_hasextflgbit(&mp->m_sb))
+		return true;
+	return false;
+}
+
 #endif	/* __XFS_BMAP_BTREE_H__ */
diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
index 3059a3ec7ecb..5392674bf893 100644
--- a/fs/xfs/libxfs/xfs_btree.c
+++ b/fs/xfs/libxfs/xfs_btree.c
@@ -4842,6 +4842,21 @@ xfs_btree_query_range(
 			fn, priv);
 }
 
+/* Query a btree for all records. */
+int
+xfs_btree_query_all(
+	struct xfs_btree_cur		*cur,
+	xfs_btree_query_range_fn	fn,
+	void				*priv)
+{
+	union xfs_btree_irec		low_rec;
+	union xfs_btree_irec		high_rec;
+
+	memset(&low_rec, 0, sizeof(low_rec));
+	memset(&high_rec, 0xFF, sizeof(high_rec));
+	return xfs_btree_query_range(cur, &low_rec, &high_rec, fn, priv);
+}
+
 /*
  * Calculate the number of blocks needed to store a given number of records
  * in a short-format (per-AG metadata) btree.
diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h
index 4bb62580a7fd..27bed08261c5 100644
--- a/fs/xfs/libxfs/xfs_btree.h
+++ b/fs/xfs/libxfs/xfs_btree.h
@@ -496,6 +496,8 @@ typedef int (*xfs_btree_query_range_fn)(struct xfs_btree_cur *cur,
 int xfs_btree_query_range(struct xfs_btree_cur *cur,
 		union xfs_btree_irec *low_rec, union xfs_btree_irec *high_rec,
 		xfs_btree_query_range_fn fn, void *priv);
+int xfs_btree_query_all(struct xfs_btree_cur *cur, xfs_btree_query_range_fn fn,
+		void *priv);
 
 typedef int (*xfs_btree_visit_blocks_fn)(struct xfs_btree_cur *cur, int level,
 		void *data);
diff --git a/fs/xfs/libxfs/xfs_dquot_buf.c b/fs/xfs/libxfs/xfs_dquot_buf.c
index ac9a003dd29a..747085b4ef44 100644
--- a/fs/xfs/libxfs/xfs_dquot_buf.c
+++ b/fs/xfs/libxfs/xfs_dquot_buf.c
@@ -35,13 +35,8 @@ int
 xfs_calc_dquots_per_chunk(
 	unsigned int		nbblks)	/* basic block units */
 {
-	unsigned int	ndquots;
-
 	ASSERT(nbblks > 0);
-	ndquots = BBTOB(nbblks);
-	do_div(ndquots, sizeof(xfs_dqblk_t));
-
-	return ndquots;
+	return BBTOB(nbblks) / sizeof(xfs_dqblk_t);
 }
 
 /*
diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h
index 6b7579e7b60a..a1dccd8d96bc 100644
--- a/fs/xfs/libxfs/xfs_format.h
+++ b/fs/xfs/libxfs/xfs_format.h
@@ -930,10 +930,8 @@ static inline uint xfs_dinode_size(int version)
 /*
  * The 32 bit link count in the inode theoretically maxes out at UINT_MAX.
  * Since the pathconf interface is signed, we use 2^31 - 1 instead.
- * The old inode format had a 16 bit link count, so its maximum is USHRT_MAX.
  */
 #define	XFS_MAXLINK		((1U << 31) - 1U)
-#define	XFS_MAXLINK_1		65535U
 
 /*
  * Values for di_format
@@ -1578,19 +1576,10 @@ static inline xfs_filblks_t startblockval(xfs_fsblock_t x)
 }
 
 /*
- * Possible extent formats.
- */
-typedef enum {
-	XFS_EXTFMT_NOSTATE = 0,
-	XFS_EXTFMT_HASSTATE
-} xfs_exntfmt_t;
-
-/*
  * Possible extent states.
  */
 typedef enum {
 	XFS_EXT_NORM, XFS_EXT_UNWRITTEN,
-	XFS_EXT_DMAPI_OFFLINE, XFS_EXT_INVALID
 } xfs_exntst_t;
 
 /*
diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h
index b72dc821d78b..095bdf049a3f 100644
--- a/fs/xfs/libxfs/xfs_fs.h
+++ b/fs/xfs/libxfs/xfs_fs.h
@@ -92,6 +92,18 @@ struct getbmapx {
 #define BMV_OF_LAST		0x4	/* segment is the last in the file */
 #define BMV_OF_SHARED		0x8	/* segment shared with another file */
 
+/*	fmr_owner special values for FS_IOC_GETFSMAP */
+#define XFS_FMR_OWN_FREE	FMR_OWN_FREE      /* free space */
+#define XFS_FMR_OWN_UNKNOWN	FMR_OWN_UNKNOWN   /* unknown owner */
+#define XFS_FMR_OWN_FS		FMR_OWNER('X', 1) /* static fs metadata */
+#define XFS_FMR_OWN_LOG		FMR_OWNER('X', 2) /* journalling log */
+#define XFS_FMR_OWN_AG		FMR_OWNER('X', 3) /* per-AG metadata */
+#define XFS_FMR_OWN_INOBT	FMR_OWNER('X', 4) /* inode btree blocks */
+#define XFS_FMR_OWN_INODES	FMR_OWNER('X', 5) /* inodes */
+#define XFS_FMR_OWN_REFC	FMR_OWNER('X', 6) /* refcount tree */
+#define XFS_FMR_OWN_COW		FMR_OWNER('X', 7) /* cow staging */
+#define XFS_FMR_OWN_DEFECTIVE	FMR_OWNER('X', 8) /* bad blocks */
+
 /*
  * Structure for XFS_IOC_FSSETDM.
  * For use by backup and restore programs to set the XFS on-disk inode
@@ -502,6 +514,7 @@ typedef struct xfs_swapext
 #define XFS_IOC_GETBMAPX	_IOWR('X', 56, struct getbmap)
 #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         */
 
 /*
  * ioctl commands that replace IRIX syssgi()'s
diff --git a/fs/xfs/libxfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c
index d93f9d918cfc..09c3d1aecef2 100644
--- a/fs/xfs/libxfs/xfs_inode_buf.c
+++ b/fs/xfs/libxfs/xfs_inode_buf.c
@@ -508,7 +508,7 @@ xfs_iread(
 
 	/* even unallocated inodes are verified */
 	if (!xfs_dinode_verify(mp, ip->i_ino, dip)) {
-		xfs_alert(mp, "%s: validation failed for inode %lld failed",
+		xfs_alert(mp, "%s: validation failed for inode %lld",
 				__func__, ip->i_ino);
 
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, dip);
diff --git a/fs/xfs/libxfs/xfs_inode_fork.c b/fs/xfs/libxfs/xfs_inode_fork.c
index 8a37efe04de3..0e80f34fe97c 100644
--- a/fs/xfs/libxfs/xfs_inode_fork.c
+++ b/fs/xfs/libxfs/xfs_inode_fork.c
@@ -42,35 +42,6 @@ STATIC int xfs_iformat_local(xfs_inode_t *, xfs_dinode_t *, int, int);
 STATIC int xfs_iformat_extents(xfs_inode_t *, xfs_dinode_t *, int);
 STATIC int xfs_iformat_btree(xfs_inode_t *, xfs_dinode_t *, int);
 
-#ifdef DEBUG
-/*
- * Make sure that the extents in the given memory buffer
- * are valid.
- */
-void
-xfs_validate_extents(
-	xfs_ifork_t		*ifp,
-	int			nrecs,
-	xfs_exntfmt_t		fmt)
-{
-	xfs_bmbt_irec_t		irec;
-	xfs_bmbt_rec_host_t	rec;
-	int			i;
-
-	for (i = 0; i < nrecs; i++) {
-		xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, i);
-		rec.l0 = get_unaligned(&ep->l0);
-		rec.l1 = get_unaligned(&ep->l1);
-		xfs_bmbt_get_all(&rec, &irec);
-		if (fmt == XFS_EXTFMT_NOSTATE)
-			ASSERT(irec.br_state == XFS_EXT_NORM);
-	}
-}
-#else /* DEBUG */
-#define xfs_validate_extents(ifp, nrecs, fmt)
-#endif /* DEBUG */
-
-
 /*
  * Move inode type and inode format specific information from the
  * on-disk inode to the in-core inode.  For fifos, devs, and sockets
@@ -352,40 +323,33 @@ xfs_iformat_local(
 }
 
 /*
- * The file consists of a set of extents all
- * of which fit into the on-disk inode.
- * If there are few enough extents to fit into
- * the if_inline_ext, then copy them there.
- * Otherwise allocate a buffer for them and copy
- * them into it.  Either way, set if_extents
- * to point at the extents.
+ * The file consists of a set of extents all of which fit into the on-disk
+ * inode.  If there are few enough extents to fit into the if_inline_ext, then
+ * copy them there.  Otherwise allocate a buffer for them and copy them into it.
+ * Either way, set if_extents to point at the extents.
  */
 STATIC int
 xfs_iformat_extents(
-	xfs_inode_t	*ip,
-	xfs_dinode_t	*dip,
-	int		whichfork)
+	struct xfs_inode	*ip,
+	struct xfs_dinode	*dip,
+	int			whichfork)
 {
-	xfs_bmbt_rec_t	*dp;
-	xfs_ifork_t	*ifp;
-	int		nex;
-	int		size;
-	int		i;
-
-	ifp = XFS_IFORK_PTR(ip, whichfork);
-	nex = XFS_DFORK_NEXTENTS(dip, whichfork);
-	size = nex * (uint)sizeof(xfs_bmbt_rec_t);
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_ifork	*ifp = XFS_IFORK_PTR(ip, whichfork);
+	int			nex = XFS_DFORK_NEXTENTS(dip, whichfork);
+	int			size = nex * sizeof(xfs_bmbt_rec_t);
+	struct xfs_bmbt_rec	*dp;
+	int			i;
 
 	/*
-	 * If the number of extents is unreasonable, then something
-	 * is wrong and we just bail out rather than crash in
-	 * kmem_alloc() or memcpy() below.
+	 * If the number of extents is unreasonable, then something is wrong and
+	 * we just bail out rather than crash in kmem_alloc() or memcpy() below.
 	 */
-	if (unlikely(size < 0 || size > XFS_DFORK_SIZE(dip, ip->i_mount, whichfork))) {
+	if (unlikely(size < 0 || size > XFS_DFORK_SIZE(dip, mp, whichfork))) {
 		xfs_warn(ip->i_mount, "corrupt inode %Lu ((a)extents = %d).",
 			(unsigned long long) ip->i_ino, nex);
 		XFS_CORRUPTION_ERROR("xfs_iformat_extents(1)", XFS_ERRLEVEL_LOW,
-				     ip->i_mount, dip);
+				     mp, dip);
 		return -EFSCORRUPTED;
 	}
 
@@ -400,22 +364,17 @@ xfs_iformat_extents(
 	ifp->if_bytes = size;
 	if (size) {
 		dp = (xfs_bmbt_rec_t *) XFS_DFORK_PTR(dip, whichfork);
-		xfs_validate_extents(ifp, nex, XFS_EXTFMT_INODE(ip));
 		for (i = 0; i < nex; i++, dp++) {
 			xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, i);
 			ep->l0 = get_unaligned_be64(&dp->l0);
 			ep->l1 = get_unaligned_be64(&dp->l1);
+			if (!xfs_bmbt_validate_extent(mp, whichfork, ep)) {
+				XFS_ERROR_REPORT("xfs_iformat_extents(2)",
+						 XFS_ERRLEVEL_LOW, mp);
+				return -EFSCORRUPTED;
+			}
 		}
 		XFS_BMAP_TRACE_EXLIST(ip, nex, whichfork);
-		if (whichfork != XFS_DATA_FORK ||
-			XFS_EXTFMT_INODE(ip) == XFS_EXTFMT_NOSTATE)
-				if (unlikely(xfs_check_nostate_extents(
-				    ifp, 0, nex))) {
-					XFS_ERROR_REPORT("xfs_iformat_extents(2)",
-							 XFS_ERRLEVEL_LOW,
-							 ip->i_mount);
-					return -EFSCORRUPTED;
-				}
 	}
 	ifp->if_flags |= XFS_IFEXTENTS;
 	return 0;
@@ -518,7 +477,6 @@ xfs_iread_extents(
 		xfs_iext_destroy(ifp);
 		return error;
 	}
-	xfs_validate_extents(ifp, nextents, XFS_EXTFMT_INODE(ip));
 	ifp->if_flags |= XFS_IFEXTENTS;
 	return 0;
 }
@@ -837,6 +795,9 @@ xfs_iextents_copy(
 	copied = 0;
 	for (i = 0; i < nrecs; i++) {
 		xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, i);
+
+		ASSERT(xfs_bmbt_validate_extent(ip->i_mount, whichfork, ep));
+
 		start_block = xfs_bmbt_get_startblock(ep);
 		if (isnullstartblock(start_block)) {
 			/*
@@ -852,7 +813,6 @@ xfs_iextents_copy(
 		copied++;
 	}
 	ASSERT(copied != 0);
-	xfs_validate_extents(ifp, copied, XFS_EXTFMT_INODE(ip));
 
 	return (copied * (uint)sizeof(xfs_bmbt_rec_t));
 }
diff --git a/fs/xfs/libxfs/xfs_rmap.c b/fs/xfs/libxfs/xfs_rmap.c
index 3a8cc7139912..06cfb93c2ef9 100644
--- a/fs/xfs/libxfs/xfs_rmap.c
+++ b/fs/xfs/libxfs/xfs_rmap.c
@@ -2001,14 +2001,14 @@ xfs_rmap_query_range_helper(
 /* Find all rmaps between two keys. */
 int
 xfs_rmap_query_range(
-	struct xfs_btree_cur		*cur,
-	struct xfs_rmap_irec		*low_rec,
-	struct xfs_rmap_irec		*high_rec,
-	xfs_rmap_query_range_fn	fn,
-	void				*priv)
+	struct xfs_btree_cur			*cur,
+	struct xfs_rmap_irec			*low_rec,
+	struct xfs_rmap_irec			*high_rec,
+	xfs_rmap_query_range_fn			fn,
+	void					*priv)
 {
-	union xfs_btree_irec		low_brec;
-	union xfs_btree_irec		high_brec;
+	union xfs_btree_irec			low_brec;
+	union xfs_btree_irec			high_brec;
 	struct xfs_rmap_query_range_info	query;
 
 	low_brec.r = *low_rec;
@@ -2019,6 +2019,20 @@ xfs_rmap_query_range(
 			xfs_rmap_query_range_helper, &query);
 }
 
+/* Find all rmaps. */
+int
+xfs_rmap_query_all(
+	struct xfs_btree_cur			*cur,
+	xfs_rmap_query_range_fn			fn,
+	void					*priv)
+{
+	struct xfs_rmap_query_range_info	query;
+
+	query.priv = priv;
+	query.fn = fn;
+	return xfs_btree_query_all(cur, xfs_rmap_query_range_helper, &query);
+}
+
 /* Clean up after calling xfs_rmap_finish_one. */
 void
 xfs_rmap_finish_one_cleanup(
@@ -2291,3 +2305,31 @@ xfs_rmap_free_extent(
 	return __xfs_rmap_add(mp, dfops, XFS_RMAP_FREE, owner,
 			XFS_DATA_FORK, &bmap);
 }
+
+/* Compare rmap records.  Returns -1 if a < b, 1 if a > b, and 0 if equal. */
+int
+xfs_rmap_compare(
+	const struct xfs_rmap_irec	*a,
+	const struct xfs_rmap_irec	*b)
+{
+	__u64				oa;
+	__u64				ob;
+
+	oa = xfs_rmap_irec_offset_pack(a);
+	ob = xfs_rmap_irec_offset_pack(b);
+
+	if (a->rm_startblock < b->rm_startblock)
+		return -1;
+	else if (a->rm_startblock > b->rm_startblock)
+		return 1;
+	else if (a->rm_owner < b->rm_owner)
+		return -1;
+	else if (a->rm_owner > b->rm_owner)
+		return 1;
+	else if (oa < ob)
+		return -1;
+	else if (oa > ob)
+		return 1;
+	else
+		return 0;
+}
diff --git a/fs/xfs/libxfs/xfs_rmap.h b/fs/xfs/libxfs/xfs_rmap.h
index 789930599339..98f908fea103 100644
--- a/fs/xfs/libxfs/xfs_rmap.h
+++ b/fs/xfs/libxfs/xfs_rmap.h
@@ -162,6 +162,8 @@ typedef int (*xfs_rmap_query_range_fn)(
 int xfs_rmap_query_range(struct xfs_btree_cur *cur,
 		struct xfs_rmap_irec *low_rec, struct xfs_rmap_irec *high_rec,
 		xfs_rmap_query_range_fn fn, void *priv);
+int xfs_rmap_query_all(struct xfs_btree_cur *cur, xfs_rmap_query_range_fn fn,
+		void *priv);
 
 enum xfs_rmap_intent_type {
 	XFS_RMAP_MAP,
@@ -212,5 +214,7 @@ int xfs_rmap_find_left_neighbor(struct xfs_btree_cur *cur, xfs_agblock_t bno,
 int xfs_rmap_lookup_le_range(struct xfs_btree_cur *cur, xfs_agblock_t bno,
 		uint64_t owner, uint64_t offset, unsigned int flags,
 		struct xfs_rmap_irec *irec, int	*stat);
+int xfs_rmap_compare(const struct xfs_rmap_irec *a,
+		const struct xfs_rmap_irec *b);
 
 #endif	/* __XFS_RMAP_H__ */
diff --git a/fs/xfs/libxfs/xfs_rtbitmap.c b/fs/xfs/libxfs/xfs_rtbitmap.c
index ea45584a9913..e47b99e59f60 100644
--- a/fs/xfs/libxfs/xfs_rtbitmap.c
+++ b/fs/xfs/libxfs/xfs_rtbitmap.c
@@ -1016,3 +1016,73 @@ xfs_rtfree_extent(
 	}
 	return 0;
 }
+
+/* Find all the free records within a given range. */
+int
+xfs_rtalloc_query_range(
+	struct xfs_trans		*tp,
+	struct xfs_rtalloc_rec		*low_rec,
+	struct xfs_rtalloc_rec		*high_rec,
+	xfs_rtalloc_query_range_fn	fn,
+	void				*priv)
+{
+	struct xfs_rtalloc_rec		rec;
+	struct xfs_mount		*mp = tp->t_mountp;
+	xfs_rtblock_t			rtstart;
+	xfs_rtblock_t			rtend;
+	xfs_rtblock_t			rem;
+	int				is_free;
+	int				error = 0;
+
+	if (low_rec->ar_startblock > high_rec->ar_startblock)
+		return -EINVAL;
+	else if (low_rec->ar_startblock == high_rec->ar_startblock)
+		return 0;
+
+	/* Iterate the bitmap, looking for discrepancies. */
+	rtstart = low_rec->ar_startblock;
+	rem = high_rec->ar_startblock - rtstart;
+	while (rem) {
+		/* Is the first block free? */
+		error = xfs_rtcheck_range(mp, tp, rtstart, 1, 1, &rtend,
+				&is_free);
+		if (error)
+			break;
+
+		/* How long does the extent go for? */
+		error = xfs_rtfind_forw(mp, tp, rtstart,
+				high_rec->ar_startblock - 1, &rtend);
+		if (error)
+			break;
+
+		if (is_free) {
+			rec.ar_startblock = rtstart;
+			rec.ar_blockcount = rtend - rtstart + 1;
+
+			error = fn(tp, &rec, priv);
+			if (error)
+				break;
+		}
+
+		rem -= rtend - rtstart + 1;
+		rtstart = rtend + 1;
+	}
+
+	return error;
+}
+
+/* Find all the free records. */
+int
+xfs_rtalloc_query_all(
+	struct xfs_trans		*tp,
+	xfs_rtalloc_query_range_fn	fn,
+	void				*priv)
+{
+	struct xfs_rtalloc_rec		keys[2];
+
+	keys[0].ar_startblock = 0;
+	keys[1].ar_startblock = tp->t_mountp->m_sb.sb_rblocks;
+	keys[0].ar_blockcount = keys[1].ar_blockcount = 0;
+
+	return xfs_rtalloc_query_range(tp, &keys[0], &keys[1], fn, priv);
+}
diff --git a/fs/xfs/libxfs/xfs_trans_space.h b/fs/xfs/libxfs/xfs_trans_space.h
index 7917f6e44286..d787c677d2a3 100644
--- a/fs/xfs/libxfs/xfs_trans_space.h
+++ b/fs/xfs/libxfs/xfs_trans_space.h
@@ -21,8 +21,20 @@
 /*
  * Components of space reservations.
  */
+
+/* Worst case number of rmaps that can be held in a block. */
 #define XFS_MAX_CONTIG_RMAPS_PER_BLOCK(mp)    \
 		(((mp)->m_rmap_mxr[0]) - ((mp)->m_rmap_mnr[0]))
+
+/* Adding one rmap could split every level up to the top of the tree. */
+#define XFS_RMAPADD_SPACE_RES(mp) ((mp)->m_rmap_maxlevels)
+
+/* Blocks we might need to add "b" rmaps to a tree. */
+#define XFS_NRMAPADD_SPACE_RES(mp, b)\
+	(((b + XFS_MAX_CONTIG_RMAPS_PER_BLOCK(mp) - 1) / \
+	  XFS_MAX_CONTIG_RMAPS_PER_BLOCK(mp)) * \
+	  XFS_RMAPADD_SPACE_RES(mp))
+
 #define XFS_MAX_CONTIG_EXTENTS_PER_BLOCK(mp)    \
 		(((mp)->m_alloc_mxr[0]) - ((mp)->m_alloc_mnr[0]))
 #define	XFS_EXTENTADD_SPACE_RES(mp,w)	(XFS_BM_MAXLEVELS(mp,w) - 1)
@@ -30,13 +42,12 @@
 	(((b + XFS_MAX_CONTIG_EXTENTS_PER_BLOCK(mp) - 1) / \
 	  XFS_MAX_CONTIG_EXTENTS_PER_BLOCK(mp)) * \
 	  XFS_EXTENTADD_SPACE_RES(mp,w))
+
+/* Blocks we might need to add "b" mappings & rmappings to a file. */
 #define XFS_SWAP_RMAP_SPACE_RES(mp,b,w)\
-	(((b + XFS_MAX_CONTIG_EXTENTS_PER_BLOCK(mp) - 1) / \
-	  XFS_MAX_CONTIG_EXTENTS_PER_BLOCK(mp)) * \
-	  XFS_EXTENTADD_SPACE_RES(mp,w) + \
-	 ((b + XFS_MAX_CONTIG_RMAPS_PER_BLOCK(mp) - 1) / \
-	  XFS_MAX_CONTIG_RMAPS_PER_BLOCK(mp)) * \
-	  (mp)->m_rmap_maxlevels)
+	(XFS_NEXTENTADD_SPACE_RES((mp), (b), (w)) + \
+	 XFS_NRMAPADD_SPACE_RES((mp), (b)))
+
 #define	XFS_DAENTER_1B(mp,w)	\
 	((w) == XFS_DATA_FORK ? (mp)->m_dir_geo->fsbcount : 1)
 #define	XFS_DAENTER_DBS(mp,w)	\