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
  ...

fs/iomap.c

@@ -910,6 +910,9 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
 			break;
 		}
 		pos += ret;
+
+		if (iov_iter_rw(iter) == READ && pos >= dio->i_size)
+			break;
 	} while ((count = iov_iter_count(iter)) > 0);
 	blk_finish_plug(&plug);
 

fs/xfs/Makefile

@@ -79,6 +79,7 @@ xfs-y				+= xfs_aops.o \
 				   xfs_extent_busy.o \
 				   xfs_file.o \
 				   xfs_filestream.o \
+				   xfs_fsmap.o \
 				   xfs_fsops.o \
 				   xfs_globals.o \
 				   xfs_icache.o \

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);
+}

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__ */

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]);
 	}
 

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;
 }
 
@@ -3852,60 +3814,6 @@ xfs_bmap_btalloc(
 	return 0;
 }
 
-/*
- * 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:

fs/xfs/libxfs/xfs_bmap.h

@@ -171,6 +171,18 @@ static inline int xfs_bmapi_whichfork(int bmapi_flags)
 	{ BMAP_COWFORK,		"COW" }
 
 
+/*
+ * 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,
@@ -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,

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

fs/xfs/libxfs/xfs_bmap_btree.h

@@ -24,14 +24,6 @@ struct xfs_mount;
 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.
  */
@@ -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__ */

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.

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);

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);
 }
 
 /*

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
@@ -1577,20 +1575,11 @@ static inline xfs_filblks_t startblockval(xfs_fsblock_t x)
 	return (xfs_filblks_t)((x) & ~STARTBLOCKMASK);
 }
 
-/*
- * 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;
 
 /*

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

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);

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));
 }

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;
+}

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__ */

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);
+}

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)	\

fs/xfs/xfs_aops.c

@@ -111,11 +111,11 @@ xfs_finish_page_writeback(
 
 	bsize = bh->b_size;
 	do {
+		if (off > end)
+			break;
 		next = bh->b_this_page;
 		if (off < bvec->bv_offset)
 			goto next_bh;
-		if (off > end)
-			break;
 		bh->b_end_io(bh, !error);
 next_bh:
 		off += bsize;
@@ -1261,8 +1261,8 @@ xfs_get_blocks(
 
 	if (nimaps) {
 		trace_xfs_get_blocks_found(ip, offset, size,
-				ISUNWRITTEN(&imap) ? XFS_IO_UNWRITTEN
-						   : XFS_IO_OVERWRITE, &imap);
+			imap.br_state == XFS_EXT_UNWRITTEN ?
+				XFS_IO_UNWRITTEN : XFS_IO_OVERWRITE, &imap);
 		xfs_iunlock(ip, lockmode);
 	} else {
 		trace_xfs_get_blocks_notfound(ip, offset, size);
@@ -1276,9 +1276,7 @@ xfs_get_blocks(
 	 * For unwritten extents do not report a disk address in the buffered
 	 * read case (treat as if we're reading into a hole).
 	 */
-	if (imap.br_startblock != HOLESTARTBLOCK &&
-	    imap.br_startblock != DELAYSTARTBLOCK &&
-	    !ISUNWRITTEN(&imap))
+	if (xfs_bmap_is_real_extent(&imap))
 		xfs_map_buffer(inode, bh_result, &imap, offset);
 
 	/*

fs/xfs/xfs_bmap_item.c

@@ -34,6 +34,8 @@
 #include "xfs_bmap.h"
 #include "xfs_icache.h"
 #include "xfs_trace.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_trans_space.h"
 
 
 kmem_zone_t	*xfs_bui_zone;
@@ -215,6 +217,7 @@ void
 xfs_bui_release(
 	struct xfs_bui_log_item	*buip)
 {
+	ASSERT(atomic_read(&buip->bui_refcount) > 0);
 	if (atomic_dec_and_test(&buip->bui_refcount)) {
 		xfs_trans_ail_remove(&buip->bui_item, SHUTDOWN_LOG_IO_ERROR);
 		xfs_bui_item_free(buip);
@@ -446,7 +449,8 @@ xfs_bui_recover(
 		return -EIO;
 	}
 
-	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_itruncate, 0, 0, 0, &tp);
+	error = xfs_trans_alloc(mp, &M_RES(mp)->tr_itruncate,
+			XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK), 0, 0, &tp);
 	if (error)
 		return error;
 	budp = xfs_trans_get_bud(tp, buip);

fs/xfs/xfs_bmap_util.c

@@ -448,10 +448,9 @@ xfs_getbmap_adjust_shared(
 	next_map->br_blockcount = 0;
 
 	/* Only written data blocks can be shared. */
-	if (!xfs_is_reflink_inode(ip) || whichfork != XFS_DATA_FORK ||
-	    map->br_startblock == DELAYSTARTBLOCK ||
-	    map->br_startblock == HOLESTARTBLOCK ||
-	    ISUNWRITTEN(map))
+	if (!xfs_is_reflink_inode(ip) ||
+	    whichfork != XFS_DATA_FORK ||
+	    !xfs_bmap_is_real_extent(map))
 		return 0;
 
 	agno = XFS_FSB_TO_AGNO(mp, map->br_startblock);
@@ -904,9 +903,9 @@ xfs_can_free_eofblocks(struct xfs_inode *ip, bool force)
 }
 
 /*
- * This is called by xfs_inactive to free any blocks beyond eof
- * when the link count isn't zero and by xfs_dm_punch_hole() when
- * punching a hole to EOF.
+ * This is called to free any blocks beyond eof. The caller must hold
+ * IOLOCK_EXCL unless we are in the inode reclaim path and have the only
+ * reference to the inode.
  */
 int
 xfs_free_eofblocks(
@@ -921,8 +920,6 @@ xfs_free_eofblocks(
 	struct xfs_bmbt_irec	imap;
 	struct xfs_mount	*mp = ip->i_mount;
 
-	ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
-
 	/*
 	 * Figure out if there are any blocks beyond the end
 	 * of the file.  If not, then there is nothing to do.
@@ -1209,11 +1206,8 @@ xfs_adjust_extent_unmap_boundaries(
 		return error;
 
 	if (nimap && imap.br_startblock != HOLESTARTBLOCK) {
-		xfs_daddr_t	block;
-
 		ASSERT(imap.br_startblock != DELAYSTARTBLOCK);
-		block = imap.br_startblock;
-		mod = do_div(block, mp->m_sb.sb_rextsize);
+		mod = do_mod(imap.br_startblock, mp->m_sb.sb_rextsize);
 		if (mod)
 			*startoffset_fsb += mp->m_sb.sb_rextsize - mod;
 	}

fs/xfs/xfs_buf.c

@@ -1079,6 +1079,8 @@ void
 xfs_buf_unlock(
 	struct xfs_buf		*bp)
 {
+	ASSERT(xfs_buf_islocked(bp));
+
 	XB_CLEAR_OWNER(bp);
 	up(&bp->b_sema);
 
@@ -1814,6 +1816,28 @@ error:
 	return NULL;
 }
 
+/*
+ * Cancel a delayed write list.
+ *
+ * Remove each buffer from the list, clear the delwri queue flag and drop the
+ * associated buffer reference.
+ */
+void
+xfs_buf_delwri_cancel(
+	struct list_head	*list)
+{
+	struct xfs_buf		*bp;
+
+	while (!list_empty(list)) {
+		bp = list_first_entry(list, struct xfs_buf, b_list);
+
+		xfs_buf_lock(bp);
+		bp->b_flags &= ~_XBF_DELWRI_Q;
+		list_del_init(&bp->b_list);
+		xfs_buf_relse(bp);
+	}
+}
+
 /*
  * Add a buffer to the delayed write list.
  *

fs/xfs/xfs_buf.h

@@ -291,7 +291,6 @@ xfs_buf_readahead(
 	return xfs_buf_readahead_map(target, &map, 1, ops);
 }
 
-struct xfs_buf *xfs_buf_get_empty(struct xfs_buftarg *target, size_t numblks);
 void xfs_buf_set_empty(struct xfs_buf *bp, size_t numblks);
 int xfs_buf_associate_memory(struct xfs_buf *bp, void *mem, size_t length);
 
@@ -330,6 +329,7 @@ extern void *xfs_buf_offset(struct xfs_buf *, size_t);
 extern void xfs_buf_stale(struct xfs_buf *bp);
 
 /* Delayed Write Buffer Routines */
+extern void xfs_buf_delwri_cancel(struct list_head *);
 extern bool xfs_buf_delwri_queue(struct xfs_buf *, struct list_head *);
 extern int xfs_buf_delwri_submit(struct list_head *);
 extern int xfs_buf_delwri_submit_nowait(struct list_head *);

fs/xfs/xfs_dir2_readdir.c

@@ -394,6 +394,7 @@ xfs_dir2_leaf_readbuf(
 
 	/*
 	 * Do we need more readahead?
+	 * Each loop tries to process 1 full dir blk; last may be partial.
 	 */
 	blk_start_plug(&plug);
 	for (mip->ra_index = mip->ra_offset = i = 0;
@@ -404,7 +405,8 @@ xfs_dir2_leaf_readbuf(
 		 * Read-ahead a contiguous directory block.
 		 */
 		if (i > mip->ra_current &&
-		    map[mip->ra_index].br_blockcount >= geo->fsbcount) {
+		    (map[mip->ra_index].br_blockcount - mip->ra_offset) >=
+		    geo->fsbcount) {
 			xfs_dir3_data_readahead(dp,
 				map[mip->ra_index].br_startoff + mip->ra_offset,
 				XFS_FSB_TO_DADDR(dp->i_mount,
@@ -425,14 +427,19 @@ xfs_dir2_leaf_readbuf(
 		}
 
 		/*
-		 * Advance offset through the mapping table.
+		 * Advance offset through the mapping table, processing a full
+		 * dir block even if it is fragmented into several extents.
+		 * But stop if we have consumed all valid mappings, even if
+		 * it's not yet a full directory block.
 		 */
-		for (j = 0; j < geo->fsbcount; j += length ) {
+		for (j = 0;
+		     j < geo->fsbcount && mip->ra_index < mip->map_valid;
+		     j += length ) {
 			/*
 			 * The rest of this extent but not more than a dir
 			 * block.
 			 */
-			length = min_t(int, geo->fsbcount,
+			length = min_t(int, geo->fsbcount - j,
 					map[mip->ra_index].br_blockcount -
 							mip->ra_offset);
 			mip->ra_offset += length;

fs/xfs/xfs_discard.c

@@ -132,6 +132,11 @@ next_extent:
 		error = xfs_btree_decrement(cur, 0, &i);
 		if (error)
 			goto out_del_cursor;
+
+		if (fatal_signal_pending(current)) {
+			error = -ERESTARTSYS;
+			goto out_del_cursor;
+		}
 	}
 
 out_del_cursor:
@@ -196,8 +201,11 @@ xfs_ioc_trim(
 	for (agno = start_agno; agno <= end_agno; agno++) {
 		error = xfs_trim_extents(mp, agno, start, end, minlen,
 					  &blocks_trimmed);
-		if (error)
+		if (error) {
 			last_error = error;
+			if (error == -ERESTARTSYS)
+				break;
+		}
 	}
 
 	if (last_error)

fs/xfs/xfs_extfree_item.c

@@ -290,6 +290,7 @@ void
 xfs_efi_release(
 	struct xfs_efi_log_item	*efip)
 {
+	ASSERT(atomic_read(&efip->efi_refcount) > 0);
 	if (atomic_dec_and_test(&efip->efi_refcount)) {
 		xfs_trans_ail_remove(&efip->efi_item, SHUTDOWN_LOG_IO_ERROR);
 		xfs_efi_item_free(efip);

fs/xfs/xfs_fsmap.c

@@ -0,0 +1,940 @@
+/*
+ * Copyright (C) 2017 Oracle.  All Rights Reserved.
+ *
+ * Author: Darrick J. Wong <darrick.wong@oracle.com>
+ *
+ * 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.
+ */
+#include "xfs.h"
+#include "xfs_fs.h"
+#include "xfs_shared.h"
+#include "xfs_format.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
+#include "xfs_sb.h"
+#include "xfs_mount.h"
+#include "xfs_defer.h"
+#include "xfs_inode.h"
+#include "xfs_trans.h"
+#include "xfs_error.h"
+#include "xfs_btree.h"
+#include "xfs_rmap_btree.h"
+#include "xfs_trace.h"
+#include "xfs_log.h"
+#include "xfs_rmap.h"
+#include "xfs_alloc.h"
+#include "xfs_bit.h"
+#include <linux/fsmap.h>
+#include "xfs_fsmap.h"
+#include "xfs_refcount.h"
+#include "xfs_refcount_btree.h"
+#include "xfs_alloc_btree.h"
+#include "xfs_rtalloc.h"
+
+/* Convert an xfs_fsmap to an fsmap. */
+void
+xfs_fsmap_from_internal(
+	struct fsmap		*dest,
+	struct xfs_fsmap	*src)
+{
+	dest->fmr_device = src->fmr_device;
+	dest->fmr_flags = src->fmr_flags;
+	dest->fmr_physical = BBTOB(src->fmr_physical);
+	dest->fmr_owner = src->fmr_owner;
+	dest->fmr_offset = BBTOB(src->fmr_offset);
+	dest->fmr_length = BBTOB(src->fmr_length);
+	dest->fmr_reserved[0] = 0;
+	dest->fmr_reserved[1] = 0;
+	dest->fmr_reserved[2] = 0;
+}
+
+/* Convert an fsmap to an xfs_fsmap. */
+void
+xfs_fsmap_to_internal(
+	struct xfs_fsmap	*dest,
+	struct fsmap		*src)
+{
+	dest->fmr_device = src->fmr_device;
+	dest->fmr_flags = src->fmr_flags;
+	dest->fmr_physical = BTOBBT(src->fmr_physical);
+	dest->fmr_owner = src->fmr_owner;
+	dest->fmr_offset = BTOBBT(src->fmr_offset);
+	dest->fmr_length = BTOBBT(src->fmr_length);
+}
+
+/* Convert an fsmap owner into an rmapbt owner. */
+static int
+xfs_fsmap_owner_to_rmap(
+	struct xfs_rmap_irec	*dest,
+	struct xfs_fsmap	*src)
+{
+	if (!(src->fmr_flags & FMR_OF_SPECIAL_OWNER)) {
+		dest->rm_owner = src->fmr_owner;
+		return 0;
+	}
+
+	switch (src->fmr_owner) {
+	case 0:			/* "lowest owner id possible" */
+	case -1ULL:		/* "highest owner id possible" */
+		dest->rm_owner = 0;
+		break;
+	case XFS_FMR_OWN_FREE:
+		dest->rm_owner = XFS_RMAP_OWN_NULL;
+		break;
+	case XFS_FMR_OWN_UNKNOWN:
+		dest->rm_owner = XFS_RMAP_OWN_UNKNOWN;
+		break;
+	case XFS_FMR_OWN_FS:
+		dest->rm_owner = XFS_RMAP_OWN_FS;
+		break;
+	case XFS_FMR_OWN_LOG:
+		dest->rm_owner = XFS_RMAP_OWN_LOG;
+		break;
+	case XFS_FMR_OWN_AG:
+		dest->rm_owner = XFS_RMAP_OWN_AG;
+		break;
+	case XFS_FMR_OWN_INOBT:
+		dest->rm_owner = XFS_RMAP_OWN_INOBT;
+		break;
+	case XFS_FMR_OWN_INODES:
+		dest->rm_owner = XFS_RMAP_OWN_INODES;
+		break;
+	case XFS_FMR_OWN_REFC:
+		dest->rm_owner = XFS_RMAP_OWN_REFC;
+		break;
+	case XFS_FMR_OWN_COW:
+		dest->rm_owner = XFS_RMAP_OWN_COW;
+		break;
+	case XFS_FMR_OWN_DEFECTIVE:	/* not implemented */
+		/* fall through */
+	default:
+		return -EINVAL;
+	}
+	return 0;
+}
+
+/* Convert an rmapbt owner into an fsmap owner. */
+static int
+xfs_fsmap_owner_from_rmap(
+	struct xfs_fsmap	*dest,
+	struct xfs_rmap_irec	*src)
+{
+	dest->fmr_flags = 0;
+	if (!XFS_RMAP_NON_INODE_OWNER(src->rm_owner)) {
+		dest->fmr_owner = src->rm_owner;
+		return 0;
+	}
+	dest->fmr_flags |= FMR_OF_SPECIAL_OWNER;
+
+	switch (src->rm_owner) {
+	case XFS_RMAP_OWN_FS:
+		dest->fmr_owner = XFS_FMR_OWN_FS;
+		break;
+	case XFS_RMAP_OWN_LOG:
+		dest->fmr_owner = XFS_FMR_OWN_LOG;
+		break;
+	case XFS_RMAP_OWN_AG:
+		dest->fmr_owner = XFS_FMR_OWN_AG;
+		break;
+	case XFS_RMAP_OWN_INOBT:
+		dest->fmr_owner = XFS_FMR_OWN_INOBT;
+		break;
+	case XFS_RMAP_OWN_INODES:
+		dest->fmr_owner = XFS_FMR_OWN_INODES;
+		break;
+	case XFS_RMAP_OWN_REFC:
+		dest->fmr_owner = XFS_FMR_OWN_REFC;
+		break;
+	case XFS_RMAP_OWN_COW:
+		dest->fmr_owner = XFS_FMR_OWN_COW;
+		break;
+	case XFS_RMAP_OWN_NULL:	/* "free" */
+		dest->fmr_owner = XFS_FMR_OWN_FREE;
+		break;
+	default:
+		return -EFSCORRUPTED;
+	}
+	return 0;
+}
+
+/* getfsmap query state */
+struct xfs_getfsmap_info {
+	struct xfs_fsmap_head	*head;
+	xfs_fsmap_format_t	formatter;	/* formatting fn */
+	void			*format_arg;	/* format buffer */
+	struct xfs_buf		*agf_bp;	/* AGF, for refcount queries */
+	xfs_daddr_t		next_daddr;	/* next daddr we expect */
+	u64			missing_owner;	/* owner of holes */
+	u32			dev;		/* device id */
+	xfs_agnumber_t		agno;		/* AG number, if applicable */
+	struct xfs_rmap_irec	low;		/* low rmap key */
+	struct xfs_rmap_irec	high;		/* high rmap key */
+	bool			last;		/* last extent? */
+};
+
+/* Associate a device with a getfsmap handler. */
+struct xfs_getfsmap_dev {
+	u32			dev;
+	int			(*fn)(struct xfs_trans *tp,
+				      struct xfs_fsmap *keys,
+				      struct xfs_getfsmap_info *info);
+};
+
+/* Compare two getfsmap device handlers. */
+static int
+xfs_getfsmap_dev_compare(
+	const void			*p1,
+	const void			*p2)
+{
+	const struct xfs_getfsmap_dev	*d1 = p1;
+	const struct xfs_getfsmap_dev	*d2 = p2;
+
+	return d1->dev - d2->dev;
+}
+
+/* Decide if this mapping is shared. */
+STATIC int
+xfs_getfsmap_is_shared(
+	struct xfs_trans		*tp,
+	struct xfs_getfsmap_info	*info,
+	struct xfs_rmap_irec		*rec,
+	bool				*stat)
+{
+	struct xfs_mount		*mp = tp->t_mountp;
+	struct xfs_btree_cur		*cur;
+	xfs_agblock_t			fbno;
+	xfs_extlen_t			flen;
+	int				error;
+
+	*stat = false;
+	if (!xfs_sb_version_hasreflink(&mp->m_sb))
+		return 0;
+	/* rt files will have agno set to NULLAGNUMBER */
+	if (info->agno == NULLAGNUMBER)
+		return 0;
+
+	/* Are there any shared blocks here? */
+	flen = 0;
+	cur = xfs_refcountbt_init_cursor(mp, tp, info->agf_bp,
+			info->agno, NULL);
+
+	error = xfs_refcount_find_shared(cur, rec->rm_startblock,
+			rec->rm_blockcount, &fbno, &flen, false);
+
+	xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
+	if (error)
+		return error;
+
+	*stat = flen > 0;
+	return 0;
+}
+
+/*
+ * Format a reverse mapping for getfsmap, having translated rm_startblock
+ * into the appropriate daddr units.
+ */
+STATIC int
+xfs_getfsmap_helper(
+	struct xfs_trans		*tp,
+	struct xfs_getfsmap_info	*info,
+	struct xfs_rmap_irec		*rec,
+	xfs_daddr_t			rec_daddr)
+{
+	struct xfs_fsmap		fmr;
+	struct xfs_mount		*mp = tp->t_mountp;
+	bool				shared;
+	int				error;
+
+	if (fatal_signal_pending(current))
+		return -EINTR;
+
+	/*
+	 * Filter out records that start before our startpoint, if the
+	 * caller requested that.
+	 */
+	if (xfs_rmap_compare(rec, &info->low) < 0) {
+		rec_daddr += XFS_FSB_TO_BB(mp, rec->rm_blockcount);
+		if (info->next_daddr < rec_daddr)
+			info->next_daddr = rec_daddr;
+		return XFS_BTREE_QUERY_RANGE_CONTINUE;
+	}
+
+	/* Are we just counting mappings? */
+	if (info->head->fmh_count == 0) {
+		if (rec_daddr > info->next_daddr)
+			info->head->fmh_entries++;
+
+		if (info->last)
+			return XFS_BTREE_QUERY_RANGE_CONTINUE;
+
+		info->head->fmh_entries++;
+
+		rec_daddr += XFS_FSB_TO_BB(mp, rec->rm_blockcount);
+		if (info->next_daddr < rec_daddr)
+			info->next_daddr = rec_daddr;
+		return XFS_BTREE_QUERY_RANGE_CONTINUE;
+	}
+
+	/*
+	 * If the record starts past the last physical block we saw,
+	 * then we've found a gap.  Report the gap as being owned by
+	 * whatever the caller specified is the missing owner.
+	 */
+	if (rec_daddr > info->next_daddr) {
+		if (info->head->fmh_entries >= info->head->fmh_count)
+			return XFS_BTREE_QUERY_RANGE_ABORT;
+
+		fmr.fmr_device = info->dev;
+		fmr.fmr_physical = info->next_daddr;
+		fmr.fmr_owner = info->missing_owner;
+		fmr.fmr_offset = 0;
+		fmr.fmr_length = rec_daddr - info->next_daddr;
+		fmr.fmr_flags = FMR_OF_SPECIAL_OWNER;
+		error = info->formatter(&fmr, info->format_arg);
+		if (error)
+			return error;
+		info->head->fmh_entries++;
+	}
+
+	if (info->last)
+		goto out;
+
+	/* Fill out the extent we found */
+	if (info->head->fmh_entries >= info->head->fmh_count)
+		return XFS_BTREE_QUERY_RANGE_ABORT;
+
+	trace_xfs_fsmap_mapping(mp, info->dev, info->agno, rec);
+
+	fmr.fmr_device = info->dev;
+	fmr.fmr_physical = rec_daddr;
+	error = xfs_fsmap_owner_from_rmap(&fmr, rec);
+	if (error)
+		return error;
+	fmr.fmr_offset = XFS_FSB_TO_BB(mp, rec->rm_offset);
+	fmr.fmr_length = XFS_FSB_TO_BB(mp, rec->rm_blockcount);
+	if (rec->rm_flags & XFS_RMAP_UNWRITTEN)
+		fmr.fmr_flags |= FMR_OF_PREALLOC;
+	if (rec->rm_flags & XFS_RMAP_ATTR_FORK)
+		fmr.fmr_flags |= FMR_OF_ATTR_FORK;
+	if (rec->rm_flags & XFS_RMAP_BMBT_BLOCK)
+		fmr.fmr_flags |= FMR_OF_EXTENT_MAP;
+	if (fmr.fmr_flags == 0) {
+		error = xfs_getfsmap_is_shared(tp, info, rec, &shared);
+		if (error)
+			return error;
+		if (shared)
+			fmr.fmr_flags |= FMR_OF_SHARED;
+	}
+	error = info->formatter(&fmr, info->format_arg);
+	if (error)
+		return error;
+	info->head->fmh_entries++;
+
+out:
+	rec_daddr += XFS_FSB_TO_BB(mp, rec->rm_blockcount);
+	if (info->next_daddr < rec_daddr)
+		info->next_daddr = rec_daddr;
+	return XFS_BTREE_QUERY_RANGE_CONTINUE;
+}
+
+/* Transform a rmapbt irec into a fsmap */
+STATIC int
+xfs_getfsmap_datadev_helper(
+	struct xfs_btree_cur		*cur,
+	struct xfs_rmap_irec		*rec,
+	void				*priv)
+{
+	struct xfs_mount		*mp = cur->bc_mp;
+	struct xfs_getfsmap_info	*info = priv;
+	xfs_fsblock_t			fsb;
+	xfs_daddr_t			rec_daddr;
+
+	fsb = XFS_AGB_TO_FSB(mp, cur->bc_private.a.agno, rec->rm_startblock);
+	rec_daddr = XFS_FSB_TO_DADDR(mp, fsb);
+
+	return xfs_getfsmap_helper(cur->bc_tp, info, rec, rec_daddr);
+}
+
+/* Transform a rtbitmap "record" into a fsmap */
+STATIC int
+xfs_getfsmap_rtdev_rtbitmap_helper(
+	struct xfs_trans		*tp,
+	struct xfs_rtalloc_rec		*rec,
+	void				*priv)
+{
+	struct xfs_mount		*mp = tp->t_mountp;
+	struct xfs_getfsmap_info	*info = priv;
+	struct xfs_rmap_irec		irec;
+	xfs_daddr_t			rec_daddr;
+
+	rec_daddr = XFS_FSB_TO_BB(mp, rec->ar_startblock);
+
+	irec.rm_startblock = rec->ar_startblock;
+	irec.rm_blockcount = rec->ar_blockcount;
+	irec.rm_owner = XFS_RMAP_OWN_NULL;	/* "free" */
+	irec.rm_offset = 0;
+	irec.rm_flags = 0;
+
+	return xfs_getfsmap_helper(tp, info, &irec, rec_daddr);
+}
+
+/* Transform a bnobt irec into a fsmap */
+STATIC int
+xfs_getfsmap_datadev_bnobt_helper(
+	struct xfs_btree_cur		*cur,
+	struct xfs_alloc_rec_incore	*rec,
+	void				*priv)
+{
+	struct xfs_mount		*mp = cur->bc_mp;
+	struct xfs_getfsmap_info	*info = priv;
+	struct xfs_rmap_irec		irec;
+	xfs_daddr_t			rec_daddr;
+
+	rec_daddr = XFS_AGB_TO_DADDR(mp, cur->bc_private.a.agno,
+			rec->ar_startblock);
+
+	irec.rm_startblock = rec->ar_startblock;
+	irec.rm_blockcount = rec->ar_blockcount;
+	irec.rm_owner = XFS_RMAP_OWN_NULL;	/* "free" */
+	irec.rm_offset = 0;
+	irec.rm_flags = 0;
+
+	return xfs_getfsmap_helper(cur->bc_tp, info, &irec, rec_daddr);
+}
+
+/* Set rmap flags based on the getfsmap flags */
+static void
+xfs_getfsmap_set_irec_flags(
+	struct xfs_rmap_irec	*irec,
+	struct xfs_fsmap	*fmr)
+{
+	irec->rm_flags = 0;
+	if (fmr->fmr_flags & FMR_OF_ATTR_FORK)
+		irec->rm_flags |= XFS_RMAP_ATTR_FORK;
+	if (fmr->fmr_flags & FMR_OF_EXTENT_MAP)
+		irec->rm_flags |= XFS_RMAP_BMBT_BLOCK;
+	if (fmr->fmr_flags & FMR_OF_PREALLOC)
+		irec->rm_flags |= XFS_RMAP_UNWRITTEN;
+}
+
+/* Execute a getfsmap query against the log device. */
+STATIC int
+xfs_getfsmap_logdev(
+	struct xfs_trans		*tp,
+	struct xfs_fsmap		*keys,
+	struct xfs_getfsmap_info	*info)
+{
+	struct xfs_mount		*mp = tp->t_mountp;
+	struct xfs_rmap_irec		rmap;
+	int				error;
+
+	/* Set up search keys */
+	info->low.rm_startblock = XFS_BB_TO_FSBT(mp, keys[0].fmr_physical);
+	info->low.rm_offset = XFS_BB_TO_FSBT(mp, keys[0].fmr_offset);
+	error = xfs_fsmap_owner_to_rmap(&info->low, keys);
+	if (error)
+		return error;
+	info->low.rm_blockcount = 0;
+	xfs_getfsmap_set_irec_flags(&info->low, &keys[0]);
+
+	error = xfs_fsmap_owner_to_rmap(&info->high, keys + 1);
+	if (error)
+		return error;
+	info->high.rm_startblock = -1U;
+	info->high.rm_owner = ULLONG_MAX;
+	info->high.rm_offset = ULLONG_MAX;
+	info->high.rm_blockcount = 0;
+	info->high.rm_flags = XFS_RMAP_KEY_FLAGS | XFS_RMAP_REC_FLAGS;
+	info->missing_owner = XFS_FMR_OWN_FREE;
+
+	trace_xfs_fsmap_low_key(mp, info->dev, info->agno, &info->low);
+	trace_xfs_fsmap_high_key(mp, info->dev, info->agno, &info->high);
+
+	if (keys[0].fmr_physical > 0)
+		return 0;
+
+	/* Fabricate an rmap entry for the external log device. */
+	rmap.rm_startblock = 0;
+	rmap.rm_blockcount = mp->m_sb.sb_logblocks;
+	rmap.rm_owner = XFS_RMAP_OWN_LOG;
+	rmap.rm_offset = 0;
+	rmap.rm_flags = 0;
+
+	return xfs_getfsmap_helper(tp, info, &rmap, 0);
+}
+
+/* Execute a getfsmap query against the realtime device. */
+STATIC int
+__xfs_getfsmap_rtdev(
+	struct xfs_trans		*tp,
+	struct xfs_fsmap		*keys,
+	int				(*query_fn)(struct xfs_trans *,
+						    struct xfs_getfsmap_info *),
+	struct xfs_getfsmap_info	*info)
+{
+	struct xfs_mount		*mp = tp->t_mountp;
+	xfs_fsblock_t			start_fsb;
+	xfs_fsblock_t			end_fsb;
+	xfs_daddr_t			eofs;
+	int				error = 0;
+
+	eofs = XFS_FSB_TO_BB(mp, mp->m_sb.sb_rblocks);
+	if (keys[0].fmr_physical >= eofs)
+		return 0;
+	if (keys[1].fmr_physical >= eofs)
+		keys[1].fmr_physical = eofs - 1;
+	start_fsb = XFS_BB_TO_FSBT(mp, keys[0].fmr_physical);
+	end_fsb = XFS_BB_TO_FSB(mp, keys[1].fmr_physical);
+
+	/* Set up search keys */
+	info->low.rm_startblock = start_fsb;
+	error = xfs_fsmap_owner_to_rmap(&info->low, &keys[0]);
+	if (error)
+		return error;
+	info->low.rm_offset = XFS_BB_TO_FSBT(mp, keys[0].fmr_offset);
+	info->low.rm_blockcount = 0;
+	xfs_getfsmap_set_irec_flags(&info->low, &keys[0]);
+
+	info->high.rm_startblock = end_fsb;
+	error = xfs_fsmap_owner_to_rmap(&info->high, &keys[1]);
+	if (error)
+		return error;
+	info->high.rm_offset = XFS_BB_TO_FSBT(mp, keys[1].fmr_offset);
+	info->high.rm_blockcount = 0;
+	xfs_getfsmap_set_irec_flags(&info->high, &keys[1]);
+
+	trace_xfs_fsmap_low_key(mp, info->dev, info->agno, &info->low);
+	trace_xfs_fsmap_high_key(mp, info->dev, info->agno, &info->high);
+
+	return query_fn(tp, info);
+}
+
+/* Actually query the realtime bitmap. */
+STATIC int
+xfs_getfsmap_rtdev_rtbitmap_query(
+	struct xfs_trans		*tp,
+	struct xfs_getfsmap_info	*info)
+{
+	struct xfs_rtalloc_rec		alow;
+	struct xfs_rtalloc_rec		ahigh;
+	int				error;
+
+	xfs_ilock(tp->t_mountp->m_rbmip, XFS_ILOCK_SHARED);
+
+	alow.ar_startblock = info->low.rm_startblock;
+	ahigh.ar_startblock = info->high.rm_startblock;
+	error = xfs_rtalloc_query_range(tp, &alow, &ahigh,
+			xfs_getfsmap_rtdev_rtbitmap_helper, info);
+	if (error)
+		goto err;
+
+	/* Report any gaps at the end of the rtbitmap */
+	info->last = true;
+	error = xfs_getfsmap_rtdev_rtbitmap_helper(tp, &ahigh, info);
+	if (error)
+		goto err;
+err:
+	xfs_iunlock(tp->t_mountp->m_rbmip, XFS_ILOCK_SHARED);
+	return error;
+}
+
+/* Execute a getfsmap query against the realtime device rtbitmap. */
+STATIC int
+xfs_getfsmap_rtdev_rtbitmap(
+	struct xfs_trans		*tp,
+	struct xfs_fsmap		*keys,
+	struct xfs_getfsmap_info	*info)
+{
+	info->missing_owner = XFS_FMR_OWN_UNKNOWN;
+	return __xfs_getfsmap_rtdev(tp, keys, xfs_getfsmap_rtdev_rtbitmap_query,
+			info);
+}
+
+/* Execute a getfsmap query against the regular data device. */
+STATIC int
+__xfs_getfsmap_datadev(
+	struct xfs_trans		*tp,
+	struct xfs_fsmap		*keys,
+	struct xfs_getfsmap_info	*info,
+	int				(*query_fn)(struct xfs_trans *,
+						    struct xfs_getfsmap_info *,
+						    struct xfs_btree_cur **,
+						    void *),
+	void				*priv)
+{
+	struct xfs_mount		*mp = tp->t_mountp;
+	struct xfs_btree_cur		*bt_cur = NULL;
+	xfs_fsblock_t			start_fsb;
+	xfs_fsblock_t			end_fsb;
+	xfs_agnumber_t			start_ag;
+	xfs_agnumber_t			end_ag;
+	xfs_daddr_t			eofs;
+	int				error = 0;
+
+	eofs = XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks);
+	if (keys[0].fmr_physical >= eofs)
+		return 0;
+	if (keys[1].fmr_physical >= eofs)
+		keys[1].fmr_physical = eofs - 1;
+	start_fsb = XFS_DADDR_TO_FSB(mp, keys[0].fmr_physical);
+	end_fsb = XFS_DADDR_TO_FSB(mp, keys[1].fmr_physical);
+
+	/*
+	 * Convert the fsmap low/high keys to AG based keys.  Initialize
+	 * low to the fsmap low key and max out the high key to the end
+	 * of the AG.
+	 */
+	info->low.rm_startblock = XFS_FSB_TO_AGBNO(mp, start_fsb);
+	info->low.rm_offset = XFS_BB_TO_FSBT(mp, keys[0].fmr_offset);
+	error = xfs_fsmap_owner_to_rmap(&info->low, &keys[0]);
+	if (error)
+		return error;
+	info->low.rm_blockcount = 0;
+	xfs_getfsmap_set_irec_flags(&info->low, &keys[0]);
+
+	info->high.rm_startblock = -1U;
+	info->high.rm_owner = ULLONG_MAX;
+	info->high.rm_offset = ULLONG_MAX;
+	info->high.rm_blockcount = 0;
+	info->high.rm_flags = XFS_RMAP_KEY_FLAGS | XFS_RMAP_REC_FLAGS;
+
+	start_ag = XFS_FSB_TO_AGNO(mp, start_fsb);
+	end_ag = XFS_FSB_TO_AGNO(mp, end_fsb);
+
+	/* Query each AG */
+	for (info->agno = start_ag; info->agno <= end_ag; info->agno++) {
+		/*
+		 * Set the AG high key from the fsmap high key if this
+		 * is the last AG that we're querying.
+		 */
+		if (info->agno == end_ag) {
+			info->high.rm_startblock = XFS_FSB_TO_AGBNO(mp,
+					end_fsb);
+			info->high.rm_offset = XFS_BB_TO_FSBT(mp,
+					keys[1].fmr_offset);
+			error = xfs_fsmap_owner_to_rmap(&info->high, &keys[1]);
+			if (error)
+				goto err;
+			xfs_getfsmap_set_irec_flags(&info->high, &keys[1]);
+		}
+
+		if (bt_cur) {
+			xfs_btree_del_cursor(bt_cur, XFS_BTREE_NOERROR);
+			bt_cur = NULL;
+			xfs_trans_brelse(tp, info->agf_bp);
+			info->agf_bp = NULL;
+		}
+
+		error = xfs_alloc_read_agf(mp, tp, info->agno, 0,
+				&info->agf_bp);
+		if (error)
+			goto err;
+
+		trace_xfs_fsmap_low_key(mp, info->dev, info->agno, &info->low);
+		trace_xfs_fsmap_high_key(mp, info->dev, info->agno,
+				&info->high);
+
+		error = query_fn(tp, info, &bt_cur, priv);
+		if (error)
+			goto err;
+
+		/*
+		 * Set the AG low key to the start of the AG prior to
+		 * moving on to the next AG.
+		 */
+		if (info->agno == start_ag) {
+			info->low.rm_startblock = 0;
+			info->low.rm_owner = 0;
+			info->low.rm_offset = 0;
+			info->low.rm_flags = 0;
+		}
+	}
+
+	/* Report any gap at the end of the AG */
+	info->last = true;
+	error = query_fn(tp, info, &bt_cur, priv);
+	if (error)
+		goto err;
+
+err:
+	if (bt_cur)
+		xfs_btree_del_cursor(bt_cur, error < 0 ? XFS_BTREE_ERROR :
+							 XFS_BTREE_NOERROR);
+	if (info->agf_bp) {
+		xfs_trans_brelse(tp, info->agf_bp);
+		info->agf_bp = NULL;
+	}
+
+	return error;
+}
+
+/* Actually query the rmap btree. */
+STATIC int
+xfs_getfsmap_datadev_rmapbt_query(
+	struct xfs_trans		*tp,
+	struct xfs_getfsmap_info	*info,
+	struct xfs_btree_cur		**curpp,
+	void				*priv)
+{
+	/* Report any gap at the end of the last AG. */
+	if (info->last)
+		return xfs_getfsmap_datadev_helper(*curpp, &info->high, info);
+
+	/* Allocate cursor for this AG and query_range it. */
+	*curpp = xfs_rmapbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
+			info->agno);
+	return xfs_rmap_query_range(*curpp, &info->low, &info->high,
+			xfs_getfsmap_datadev_helper, info);
+}
+
+/* Execute a getfsmap query against the regular data device rmapbt. */
+STATIC int
+xfs_getfsmap_datadev_rmapbt(
+	struct xfs_trans		*tp,
+	struct xfs_fsmap		*keys,
+	struct xfs_getfsmap_info	*info)
+{
+	info->missing_owner = XFS_FMR_OWN_FREE;
+	return __xfs_getfsmap_datadev(tp, keys, info,
+			xfs_getfsmap_datadev_rmapbt_query, NULL);
+}
+
+/* Actually query the bno btree. */
+STATIC int
+xfs_getfsmap_datadev_bnobt_query(
+	struct xfs_trans		*tp,
+	struct xfs_getfsmap_info	*info,
+	struct xfs_btree_cur		**curpp,
+	void				*priv)
+{
+	struct xfs_alloc_rec_incore	*key = priv;
+
+	/* Report any gap at the end of the last AG. */
+	if (info->last)
+		return xfs_getfsmap_datadev_bnobt_helper(*curpp, &key[1], info);
+
+	/* Allocate cursor for this AG and query_range it. */
+	*curpp = xfs_allocbt_init_cursor(tp->t_mountp, tp, info->agf_bp,
+			info->agno, XFS_BTNUM_BNO);
+	key->ar_startblock = info->low.rm_startblock;
+	key[1].ar_startblock = info->high.rm_startblock;
+	return xfs_alloc_query_range(*curpp, key, &key[1],
+			xfs_getfsmap_datadev_bnobt_helper, info);
+}
+
+/* Execute a getfsmap query against the regular data device's bnobt. */
+STATIC int
+xfs_getfsmap_datadev_bnobt(
+	struct xfs_trans		*tp,
+	struct xfs_fsmap		*keys,
+	struct xfs_getfsmap_info	*info)
+{
+	struct xfs_alloc_rec_incore	akeys[2];
+
+	info->missing_owner = XFS_FMR_OWN_UNKNOWN;
+	return __xfs_getfsmap_datadev(tp, keys, info,
+			xfs_getfsmap_datadev_bnobt_query, &akeys[0]);
+}
+
+/* Do we recognize the device? */
+STATIC bool
+xfs_getfsmap_is_valid_device(
+	struct xfs_mount	*mp,
+	struct xfs_fsmap	*fm)
+{
+	if (fm->fmr_device == 0 || fm->fmr_device == UINT_MAX ||
+	    fm->fmr_device == new_encode_dev(mp->m_ddev_targp->bt_dev))
+		return true;
+	if (mp->m_logdev_targp &&
+	    fm->fmr_device == new_encode_dev(mp->m_logdev_targp->bt_dev))
+		return true;
+	if (mp->m_rtdev_targp &&
+	    fm->fmr_device == new_encode_dev(mp->m_rtdev_targp->bt_dev))
+		return true;
+	return false;
+}
+
+/* Ensure that the low key is less than the high key. */
+STATIC bool
+xfs_getfsmap_check_keys(
+	struct xfs_fsmap		*low_key,
+	struct xfs_fsmap		*high_key)
+{
+	if (low_key->fmr_device > high_key->fmr_device)
+		return false;
+	if (low_key->fmr_device < high_key->fmr_device)
+		return true;
+
+	if (low_key->fmr_physical > high_key->fmr_physical)
+		return false;
+	if (low_key->fmr_physical < high_key->fmr_physical)
+		return true;
+
+	if (low_key->fmr_owner > high_key->fmr_owner)
+		return false;
+	if (low_key->fmr_owner < high_key->fmr_owner)
+		return true;
+
+	if (low_key->fmr_offset > high_key->fmr_offset)
+		return false;
+	if (low_key->fmr_offset < high_key->fmr_offset)
+		return true;
+
+	return false;
+}
+
+#define XFS_GETFSMAP_DEVS	3
+/*
+ * Get filesystem's extents as described in head, and format for
+ * output.  Calls formatter to fill the user's buffer until all
+ * extents are mapped, until the passed-in head->fmh_count slots have
+ * been filled, or until the formatter short-circuits the loop, if it
+ * is tracking filled-in extents on its own.
+ *
+ * Key to Confusion
+ * ----------------
+ * There are multiple levels of keys and counters at work here:
+ * xfs_fsmap_head.fmh_keys	-- low and high fsmap keys passed in;
+ * 				   these reflect fs-wide sector addrs.
+ * dkeys			-- fmh_keys used to query each device;
+ * 				   these are fmh_keys but w/ the low key
+ * 				   bumped up by fmr_length.
+ * xfs_getfsmap_info.next_daddr	-- next disk addr we expect to see; this
+ *				   is how we detect gaps in the fsmap
+				   records and report them.
+ * xfs_getfsmap_info.low/high	-- per-AG low/high keys computed from
+ * 				   dkeys; used to query the metadata.
+ */
+int
+xfs_getfsmap(
+	struct xfs_mount		*mp,
+	struct xfs_fsmap_head		*head,
+	xfs_fsmap_format_t		formatter,
+	void				*arg)
+{
+	struct xfs_trans		*tp = NULL;
+	struct xfs_fsmap		dkeys[2];	/* per-dev keys */
+	struct xfs_getfsmap_dev		handlers[XFS_GETFSMAP_DEVS];
+	struct xfs_getfsmap_info	info = { NULL };
+	int				i;
+	int				error = 0;
+
+	if (head->fmh_iflags & ~FMH_IF_VALID)
+		return -EINVAL;
+	if (!xfs_getfsmap_is_valid_device(mp, &head->fmh_keys[0]) ||
+	    !xfs_getfsmap_is_valid_device(mp, &head->fmh_keys[1]))
+		return -EINVAL;
+
+	head->fmh_entries = 0;
+
+	/* Set up our device handlers. */
+	memset(handlers, 0, sizeof(handlers));
+	handlers[0].dev = new_encode_dev(mp->m_ddev_targp->bt_dev);
+	if (xfs_sb_version_hasrmapbt(&mp->m_sb))
+		handlers[0].fn = xfs_getfsmap_datadev_rmapbt;
+	else
+		handlers[0].fn = xfs_getfsmap_datadev_bnobt;
+	if (mp->m_logdev_targp != mp->m_ddev_targp) {
+		handlers[1].dev = new_encode_dev(mp->m_logdev_targp->bt_dev);
+		handlers[1].fn = xfs_getfsmap_logdev;
+	}
+	if (mp->m_rtdev_targp) {
+		handlers[2].dev = new_encode_dev(mp->m_rtdev_targp->bt_dev);
+		handlers[2].fn = xfs_getfsmap_rtdev_rtbitmap;
+	}
+
+	xfs_sort(handlers, XFS_GETFSMAP_DEVS, sizeof(struct xfs_getfsmap_dev),
+			xfs_getfsmap_dev_compare);
+
+	/*
+	 * To continue where we left off, we allow userspace to use the
+	 * last mapping from a previous call as the low key of the next.
+	 * This is identified by a non-zero length in the low key. We
+	 * have to increment the low key in this scenario to ensure we
+	 * don't return the same mapping again, and instead return the
+	 * very next mapping.
+	 *
+	 * If the low key mapping refers to file data, the same physical
+	 * blocks could be mapped to several other files/offsets.
+	 * According to rmapbt record ordering, the minimal next
+	 * possible record for the block range is the next starting
+	 * offset in the same inode. Therefore, bump the file offset to
+	 * continue the search appropriately.  For all other low key
+	 * mapping types (attr blocks, metadata), bump the physical
+	 * offset as there can be no other mapping for the same physical
+	 * block range.
+	 */
+	dkeys[0] = head->fmh_keys[0];
+	if (dkeys[0].fmr_flags & (FMR_OF_SPECIAL_OWNER | FMR_OF_EXTENT_MAP)) {
+		dkeys[0].fmr_physical += dkeys[0].fmr_length;
+		dkeys[0].fmr_owner = 0;
+		if (dkeys[0].fmr_offset)
+			return -EINVAL;
+	} else
+		dkeys[0].fmr_offset += dkeys[0].fmr_length;
+	dkeys[0].fmr_length = 0;
+	memset(&dkeys[1], 0xFF, sizeof(struct xfs_fsmap));
+
+	if (!xfs_getfsmap_check_keys(dkeys, &head->fmh_keys[1]))
+		return -EINVAL;
+
+	info.next_daddr = head->fmh_keys[0].fmr_physical +
+			  head->fmh_keys[0].fmr_length;
+	info.formatter = formatter;
+	info.format_arg = arg;
+	info.head = head;
+
+	/* For each device we support... */
+	for (i = 0; i < XFS_GETFSMAP_DEVS; i++) {
+		/* Is this device within the range the user asked for? */
+		if (!handlers[i].fn)
+			continue;
+		if (head->fmh_keys[0].fmr_device > handlers[i].dev)
+			continue;
+		if (head->fmh_keys[1].fmr_device < handlers[i].dev)
+			break;
+
+		/*
+		 * If this device number matches the high key, we have
+		 * to pass the high key to the handler to limit the
+		 * query results.  If the device number exceeds the
+		 * low key, zero out the low key so that we get
+		 * everything from the beginning.
+		 */
+		if (handlers[i].dev == head->fmh_keys[1].fmr_device)
+			dkeys[1] = head->fmh_keys[1];
+		if (handlers[i].dev > head->fmh_keys[0].fmr_device)
+			memset(&dkeys[0], 0, sizeof(struct xfs_fsmap));
+
+		error = xfs_trans_alloc_empty(mp, &tp);
+		if (error)
+			break;
+
+		info.dev = handlers[i].dev;
+		info.last = false;
+		info.agno = NULLAGNUMBER;
+		error = handlers[i].fn(tp, dkeys, &info);
+		if (error)
+			break;
+		xfs_trans_cancel(tp);
+		tp = NULL;
+		info.next_daddr = 0;
+	}
+
+	if (tp)
+		xfs_trans_cancel(tp);
+	head->fmh_oflags = FMH_OF_DEV_T;
+	return error;
+}

fs/xfs/xfs_fsmap.h

@@ -0,0 +1,53 @@
+/*
+ * Copyright (C) 2017 Oracle.  All Rights Reserved.
+ *
+ * Author: Darrick J. Wong <darrick.wong@oracle.com>
+ *
+ * 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_FSMAP_H__
+#define __XFS_FSMAP_H__
+
+struct fsmap;
+
+/* internal fsmap representation */
+struct xfs_fsmap {
+	dev_t		fmr_device;	/* device id */
+	uint32_t	fmr_flags;	/* mapping flags */
+	uint64_t	fmr_physical;	/* device offset of segment */
+	uint64_t	fmr_owner;	/* owner id */
+	xfs_fileoff_t	fmr_offset;	/* file offset of segment */
+	xfs_filblks_t	fmr_length;	/* length of segment, blocks */
+};
+
+struct xfs_fsmap_head {
+	uint32_t	fmh_iflags;	/* control flags */
+	uint32_t	fmh_oflags;	/* output flags */
+	unsigned int	fmh_count;	/* # of entries in array incl. input */
+	unsigned int	fmh_entries;	/* # of entries filled in (output). */
+
+	struct xfs_fsmap fmh_keys[2];	/* low and high keys */
+};
+
+void xfs_fsmap_from_internal(struct fsmap *dest, struct xfs_fsmap *src);
+void xfs_fsmap_to_internal(struct xfs_fsmap *dest, struct fsmap *src);
+
+/* fsmap to userspace formatter - copy to user & advance pointer */
+typedef int (*xfs_fsmap_format_t)(struct xfs_fsmap *, void *);
+
+int xfs_getfsmap(struct xfs_mount *mp, struct xfs_fsmap_head *head,
+		xfs_fsmap_format_t formatter, void *arg);
+
+#endif /* __XFS_FSMAP_H__ */

fs/xfs/xfs_icache.c

@@ -262,6 +262,22 @@ xfs_inode_clear_reclaim_tag(
 	xfs_perag_clear_reclaim_tag(pag);
 }
 
+static void
+xfs_inew_wait(
+	struct xfs_inode	*ip)
+{
+	wait_queue_head_t *wq = bit_waitqueue(&ip->i_flags, __XFS_INEW_BIT);
+	DEFINE_WAIT_BIT(wait, &ip->i_flags, __XFS_INEW_BIT);
+
+	do {
+		prepare_to_wait(wq, &wait.wait, TASK_UNINTERRUPTIBLE);
+		if (!xfs_iflags_test(ip, XFS_INEW))
+			break;
+		schedule();
+	} while (true);
+	finish_wait(wq, &wait.wait);
+}
+
 /*
  * When we recycle a reclaimable inode, we need to re-initialise the VFS inode
  * part of the structure. This is made more complex by the fact we store
@@ -366,14 +382,17 @@ xfs_iget_cache_hit(
 
 		error = xfs_reinit_inode(mp, inode);
 		if (error) {
+			bool wake;
 			/*
 			 * Re-initializing the inode failed, and we are in deep
 			 * trouble.  Try to re-add it to the reclaim list.
 			 */
 			rcu_read_lock();
 			spin_lock(&ip->i_flags_lock);
-
+			wake = !!__xfs_iflags_test(ip, XFS_INEW);
 			ip->i_flags &= ~(XFS_INEW | XFS_IRECLAIM);
+			if (wake)
+				wake_up_bit(&ip->i_flags, __XFS_INEW_BIT);
 			ASSERT(ip->i_flags & XFS_IRECLAIMABLE);
 			trace_xfs_iget_reclaim_fail(ip);
 			goto out_error;
@@ -623,9 +642,11 @@ out_error_or_again:
 
 STATIC int
 xfs_inode_ag_walk_grab(
-	struct xfs_inode	*ip)
+	struct xfs_inode	*ip,
+	int			flags)
 {
 	struct inode		*inode = VFS_I(ip);
+	bool			newinos = !!(flags & XFS_AGITER_INEW_WAIT);
 
 	ASSERT(rcu_read_lock_held());
 
@@ -643,7 +664,8 @@ xfs_inode_ag_walk_grab(
 		goto out_unlock_noent;
 
 	/* avoid new or reclaimable inodes. Leave for reclaim code to flush */
-	if (__xfs_iflags_test(ip, XFS_INEW | XFS_IRECLAIMABLE | XFS_IRECLAIM))
+	if ((!newinos && __xfs_iflags_test(ip, XFS_INEW)) ||
+	    __xfs_iflags_test(ip, XFS_IRECLAIMABLE | XFS_IRECLAIM))
 		goto out_unlock_noent;
 	spin_unlock(&ip->i_flags_lock);
 
@@ -671,7 +693,8 @@ xfs_inode_ag_walk(
 					   void *args),
 	int			flags,
 	void			*args,
-	int			tag)
+	int			tag,
+	int			iter_flags)
 {
 	uint32_t		first_index;
 	int			last_error = 0;
@@ -713,7 +736,7 @@ restart:
 		for (i = 0; i < nr_found; i++) {
 			struct xfs_inode *ip = batch[i];
 
-			if (done || xfs_inode_ag_walk_grab(ip))
+			if (done || xfs_inode_ag_walk_grab(ip, iter_flags))
 				batch[i] = NULL;
 
 			/*
@@ -741,6 +764,9 @@ restart:
 		for (i = 0; i < nr_found; i++) {
 			if (!batch[i])
 				continue;
+			if ((iter_flags & XFS_AGITER_INEW_WAIT) &&
+			    xfs_iflags_test(batch[i], XFS_INEW))
+				xfs_inew_wait(batch[i]);
 			error = execute(batch[i], flags, args);
 			IRELE(batch[i]);
 			if (error == -EAGAIN) {
@@ -820,12 +846,13 @@ xfs_cowblocks_worker(
 }
 
 int
-xfs_inode_ag_iterator(
+xfs_inode_ag_iterator_flags(
 	struct xfs_mount	*mp,
 	int			(*execute)(struct xfs_inode *ip, int flags,
 					   void *args),
 	int			flags,
-	void			*args)
+	void			*args,
+	int			iter_flags)
 {
 	struct xfs_perag	*pag;
 	int			error = 0;
@@ -835,7 +862,8 @@ xfs_inode_ag_iterator(
 	ag = 0;
 	while ((pag = xfs_perag_get(mp, ag))) {
 		ag = pag->pag_agno + 1;
-		error = xfs_inode_ag_walk(mp, pag, execute, flags, args, -1);
+		error = xfs_inode_ag_walk(mp, pag, execute, flags, args, -1,
+					  iter_flags);
 		xfs_perag_put(pag);
 		if (error) {
 			last_error = error;
@@ -846,6 +874,17 @@ xfs_inode_ag_iterator(
 	return last_error;
 }
 
+int
+xfs_inode_ag_iterator(
+	struct xfs_mount	*mp,
+	int			(*execute)(struct xfs_inode *ip, int flags,
+					   void *args),
+	int			flags,
+	void			*args)
+{
+	return xfs_inode_ag_iterator_flags(mp, execute, flags, args, 0);
+}
+
 int
 xfs_inode_ag_iterator_tag(
 	struct xfs_mount	*mp,
@@ -863,7 +902,8 @@ xfs_inode_ag_iterator_tag(
 	ag = 0;
 	while ((pag = xfs_perag_get_tag(mp, ag, tag))) {
 		ag = pag->pag_agno + 1;
-		error = xfs_inode_ag_walk(mp, pag, execute, flags, args, tag);
+		error = xfs_inode_ag_walk(mp, pag, execute, flags, args, tag,
+					  0);
 		xfs_perag_put(pag);
 		if (error) {
 			last_error = error;

fs/xfs/xfs_icache.h

@@ -48,6 +48,11 @@ struct xfs_eofblocks {
 #define XFS_IGET_UNTRUSTED	0x2
 #define XFS_IGET_DONTCACHE	0x4
 
+/*
+ * flags for AG inode iterator
+ */
+#define XFS_AGITER_INEW_WAIT	0x1	/* wait on new inodes */
+
 int xfs_iget(struct xfs_mount *mp, struct xfs_trans *tp, xfs_ino_t ino,
 	     uint flags, uint lock_flags, xfs_inode_t **ipp);
 
@@ -79,6 +84,9 @@ void xfs_cowblocks_worker(struct work_struct *);
 int xfs_inode_ag_iterator(struct xfs_mount *mp,
 	int (*execute)(struct xfs_inode *ip, int flags, void *args),
 	int flags, void *args);
+int xfs_inode_ag_iterator_flags(struct xfs_mount *mp,
+	int (*execute)(struct xfs_inode *ip, int flags, void *args),
+	int flags, void *args, int iter_flags);
 int xfs_inode_ag_iterator_tag(struct xfs_mount *mp,
 	int (*execute)(struct xfs_inode *ip, int flags, void *args),
 	int flags, void *args, int tag);

fs/xfs/xfs_inode.c

@@ -1906,12 +1906,13 @@ xfs_inactive(
 		 * force is true because we are evicting an inode from the
 		 * cache. Post-eof blocks must be freed, lest we end up with
 		 * broken free space accounting.
+		 *
+		 * Note: don't bother with iolock here since lockdep complains
+		 * about acquiring it in reclaim context. We have the only
+		 * reference to the inode at this point anyways.
 		 */
-		if (xfs_can_free_eofblocks(ip, true)) {
-			xfs_ilock(ip, XFS_IOLOCK_EXCL);
+		if (xfs_can_free_eofblocks(ip, true))
 			xfs_free_eofblocks(ip);
-			xfs_iunlock(ip, XFS_IOLOCK_EXCL);
-		}
 
 		return;
 	}

fs/xfs/xfs_inode.h

@@ -216,7 +216,8 @@ static inline bool xfs_is_reflink_inode(struct xfs_inode *ip)
 #define XFS_IRECLAIM		(1 << 0) /* started reclaiming this inode */
 #define XFS_ISTALE		(1 << 1) /* inode has been staled */
 #define XFS_IRECLAIMABLE	(1 << 2) /* inode can be reclaimed */
-#define XFS_INEW		(1 << 3) /* inode has just been allocated */
+#define __XFS_INEW_BIT		3	 /* inode has just been allocated */
+#define XFS_INEW		(1 << __XFS_INEW_BIT)
 #define XFS_ITRUNCATED		(1 << 5) /* truncated down so flush-on-close */
 #define XFS_IDIRTY_RELEASE	(1 << 6) /* dirty release already seen */
 #define __XFS_IFLOCK_BIT	7	 /* inode is being flushed right now */
@@ -464,6 +465,7 @@ static inline void xfs_finish_inode_setup(struct xfs_inode *ip)
 	xfs_iflags_clear(ip, XFS_INEW);
 	barrier();
 	unlock_new_inode(VFS_I(ip));
+	wake_up_bit(&ip->i_flags, __XFS_INEW_BIT);
 }
 
 static inline void xfs_setup_existing_inode(struct xfs_inode *ip)

fs/xfs/xfs_inode_item.c

@@ -731,22 +731,27 @@ xfs_iflush_done(
 	 * holding the lock before removing the inode from the AIL.
 	 */
 	if (need_ail) {
-		struct xfs_log_item *log_items[need_ail];
-		int i = 0;
+		bool			mlip_changed = false;
+
+		/* this is an opencoded batch version of xfs_trans_ail_delete */
 		spin_lock(&ailp->xa_lock);
 		for (blip = lip; blip; blip = blip->li_bio_list) {
-			iip = INODE_ITEM(blip);
-			if (iip->ili_logged &&
-			    blip->li_lsn == iip->ili_flush_lsn) {
-				log_items[i++] = blip;
-			}
-			ASSERT(i <= need_ail);
+			if (INODE_ITEM(blip)->ili_logged &&
+			    blip->li_lsn == INODE_ITEM(blip)->ili_flush_lsn)
+				mlip_changed |= xfs_ail_delete_one(ailp, blip);
 		}
-		/* xfs_trans_ail_delete_bulk() drops the AIL lock. */
-		xfs_trans_ail_delete_bulk(ailp, log_items, i,
-					  SHUTDOWN_CORRUPT_INCORE);
-	}
 
+		if (mlip_changed) {
+			if (!XFS_FORCED_SHUTDOWN(ailp->xa_mount))
+				xlog_assign_tail_lsn_locked(ailp->xa_mount);
+			if (list_empty(&ailp->xa_ail))
+				wake_up_all(&ailp->xa_empty);
+		}
+		spin_unlock(&ailp->xa_lock);
+
+		if (mlip_changed)
+			xfs_log_space_wake(ailp->xa_mount);
+	}
 
 	/*
 	 * clean up and unlock the flush lock now we are done. We can clear the

fs/xfs/xfs_ioctl.c

@@ -41,6 +41,9 @@
 #include "xfs_trans.h"
 #include "xfs_pnfs.h"
 #include "xfs_acl.h"
+#include "xfs_btree.h"
+#include <linux/fsmap.h>
+#include "xfs_fsmap.h"
 
 #include <linux/capability.h>
 #include <linux/cred.h>
@@ -1543,10 +1546,11 @@ xfs_ioc_getbmap(
 	unsigned int		cmd,
 	void			__user *arg)
 {
-	struct getbmapx		bmx;
+	struct getbmapx		bmx = { 0 };
 	int			error;
 
-	if (copy_from_user(&bmx, arg, sizeof(struct getbmapx)))
+	/* struct getbmap is a strict subset of struct getbmapx. */
+	if (copy_from_user(&bmx, arg, offsetof(struct getbmapx, bmv_iflags)))
 		return -EFAULT;
 
 	if (bmx.bmv_count < 2)
@@ -1608,6 +1612,84 @@ xfs_ioc_getbmapx(
 	return 0;
 }
 
+struct getfsmap_info {
+	struct xfs_mount	*mp;
+	struct fsmap_head __user *data;
+	unsigned int		idx;
+	__u32			last_flags;
+};
+
+STATIC int
+xfs_getfsmap_format(struct xfs_fsmap *xfm, void *priv)
+{
+	struct getfsmap_info	*info = priv;
+	struct fsmap		fm;
+
+	trace_xfs_getfsmap_mapping(info->mp, xfm);
+
+	info->last_flags = xfm->fmr_flags;
+	xfs_fsmap_from_internal(&fm, xfm);
+	if (copy_to_user(&info->data->fmh_recs[info->idx++], &fm,
+			sizeof(struct fsmap)))
+		return -EFAULT;
+
+	return 0;
+}
+
+STATIC int
+xfs_ioc_getfsmap(
+	struct xfs_inode	*ip,
+	struct fsmap_head	__user *arg)
+{
+	struct getfsmap_info	info = { NULL };
+	struct xfs_fsmap_head	xhead = {0};
+	struct fsmap_head	head;
+	bool			aborted = false;
+	int			error;
+
+	if (copy_from_user(&head, arg, sizeof(struct fsmap_head)))
+		return -EFAULT;
+	if (memchr_inv(head.fmh_reserved, 0, sizeof(head.fmh_reserved)) ||
+	    memchr_inv(head.fmh_keys[0].fmr_reserved, 0,
+		       sizeof(head.fmh_keys[0].fmr_reserved)) ||
+	    memchr_inv(head.fmh_keys[1].fmr_reserved, 0,
+		       sizeof(head.fmh_keys[1].fmr_reserved)))
+		return -EINVAL;
+
+	xhead.fmh_iflags = head.fmh_iflags;
+	xhead.fmh_count = head.fmh_count;
+	xfs_fsmap_to_internal(&xhead.fmh_keys[0], &head.fmh_keys[0]);
+	xfs_fsmap_to_internal(&xhead.fmh_keys[1], &head.fmh_keys[1]);
+
+	trace_xfs_getfsmap_low_key(ip->i_mount, &xhead.fmh_keys[0]);
+	trace_xfs_getfsmap_high_key(ip->i_mount, &xhead.fmh_keys[1]);
+
+	info.mp = ip->i_mount;
+	info.data = arg;
+	error = xfs_getfsmap(ip->i_mount, &xhead, xfs_getfsmap_format, &info);
+	if (error == XFS_BTREE_QUERY_RANGE_ABORT) {
+		error = 0;
+		aborted = true;
+	} else if (error)
+		return error;
+
+	/* If we didn't abort, set the "last" flag in the last fmx */
+	if (!aborted && info.idx) {
+		info.last_flags |= FMR_OF_LAST;
+		if (copy_to_user(&info.data->fmh_recs[info.idx - 1].fmr_flags,
+				&info.last_flags, sizeof(info.last_flags)))
+			return -EFAULT;
+	}
+
+	/* copy back header */
+	head.fmh_entries = xhead.fmh_entries;
+	head.fmh_oflags = xhead.fmh_oflags;
+	if (copy_to_user(arg, &head, sizeof(struct fsmap_head)))
+		return -EFAULT;
+
+	return 0;
+}
+
 int
 xfs_ioc_swapext(
 	xfs_swapext_t	*sxp)
@@ -1788,6 +1870,9 @@ xfs_file_ioctl(
 	case XFS_IOC_GETBMAPX:
 		return xfs_ioc_getbmapx(ip, arg);
 
+	case FS_IOC_GETFSMAP:
+		return xfs_ioc_getfsmap(ip, arg);
+
 	case XFS_IOC_FD_TO_HANDLE:
 	case XFS_IOC_PATH_TO_HANDLE:
 	case XFS_IOC_PATH_TO_FSHANDLE: {

fs/xfs/xfs_ioctl32.c

@@ -20,6 +20,7 @@
 #include <linux/mount.h>
 #include <linux/slab.h>
 #include <linux/uaccess.h>
+#include <linux/fsmap.h>
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_format.h"
@@ -554,6 +555,7 @@ xfs_file_compat_ioctl(
 	case XFS_IOC_GOINGDOWN:
 	case XFS_IOC_ERROR_INJECTION:
 	case XFS_IOC_ERROR_CLEARALL:
+	case FS_IOC_GETFSMAP:
 		return xfs_file_ioctl(filp, cmd, p);
 #ifndef BROKEN_X86_ALIGNMENT
 	/* These are handled fine if no alignment issues */

fs/xfs/xfs_iomap.c

@@ -240,7 +240,7 @@ xfs_iomap_write_direct(
 	 */
 	if (IS_DAX(VFS_I(ip))) {
 		bmapi_flags = XFS_BMAPI_CONVERT | XFS_BMAPI_ZERO;
-		if (ISUNWRITTEN(imap)) {
+		if (imap->br_state == XFS_EXT_UNWRITTEN) {
 			tflags |= XFS_TRANS_RESERVE;
 			resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1;
 		}
@@ -945,7 +945,7 @@ static inline bool imap_needs_alloc(struct inode *inode,
 	return !nimaps ||
 		imap->br_startblock == HOLESTARTBLOCK ||
 		imap->br_startblock == DELAYSTARTBLOCK ||
-		(IS_DAX(inode) && ISUNWRITTEN(imap));
+		(IS_DAX(inode) && imap->br_state == XFS_EXT_UNWRITTEN);
 }
 
 static inline bool need_excl_ilock(struct xfs_inode *ip, unsigned flags)
@@ -1180,10 +1180,10 @@ xfs_xattr_iomap_begin(
 	if (XFS_FORCED_SHUTDOWN(mp))
 		return -EIO;
 
-	lockmode = xfs_ilock_data_map_shared(ip);
+	lockmode = xfs_ilock_attr_map_shared(ip);
 
 	/* if there are no attribute fork or extents, return ENOENT */
-	if (XFS_IFORK_Q(ip) || !ip->i_d.di_anextents) {
+	if (!XFS_IFORK_Q(ip) || !ip->i_d.di_anextents) {
 		error = -ENOENT;
 		goto out_unlock;
 	}

fs/xfs/xfs_linux.h

@@ -212,88 +212,6 @@ static inline kgid_t xfs_gid_to_kgid(__uint32_t gid)
 #define xfs_sort(a,n,s,fn)	sort(a,n,s,fn,NULL)
 #define xfs_stack_trace()	dump_stack()
 
-
-/* Move the kernel do_div definition off to one side */
-
-#if defined __i386__
-/* For ia32 we need to pull some tricks to get past various versions
- * of the compiler which do not like us using do_div in the middle
- * of large functions.
- */
-static inline __u32 xfs_do_div(void *a, __u32 b, int n)
-{
-	__u32	mod;
-
-	switch (n) {
-		case 4:
-			mod = *(__u32 *)a % b;
-			*(__u32 *)a = *(__u32 *)a / b;
-			return mod;
-		case 8:
-			{
-			unsigned long __upper, __low, __high, __mod;
-			__u64	c = *(__u64 *)a;
-			__upper = __high = c >> 32;
-			__low = c;
-			if (__high) {
-				__upper = __high % (b);
-				__high = __high / (b);
-			}
-			asm("divl %2":"=a" (__low), "=d" (__mod):"rm" (b), "0" (__low), "1" (__upper));
-			asm("":"=A" (c):"a" (__low),"d" (__high));
-			*(__u64 *)a = c;
-			return __mod;
-			}
-	}
-
-	/* NOTREACHED */
-	return 0;
-}
-
-/* Side effect free 64 bit mod operation */
-static inline __u32 xfs_do_mod(void *a, __u32 b, int n)
-{
-	switch (n) {
-		case 4:
-			return *(__u32 *)a % b;
-		case 8:
-			{
-			unsigned long __upper, __low, __high, __mod;
-			__u64	c = *(__u64 *)a;
-			__upper = __high = c >> 32;
-			__low = c;
-			if (__high) {
-				__upper = __high % (b);
-				__high = __high / (b);
-			}
-			asm("divl %2":"=a" (__low), "=d" (__mod):"rm" (b), "0" (__low), "1" (__upper));
-			asm("":"=A" (c):"a" (__low),"d" (__high));
-			return __mod;
-			}
-	}
-
-	/* NOTREACHED */
-	return 0;
-}
-#else
-static inline __u32 xfs_do_div(void *a, __u32 b, int n)
-{
-	__u32	mod;
-
-	switch (n) {
-		case 4:
-			mod = *(__u32 *)a % b;
-			*(__u32 *)a = *(__u32 *)a / b;
-			return mod;
-		case 8:
-			mod = do_div(*(__u64 *)a, b);
-			return mod;
-	}
-
-	/* NOTREACHED */
-	return 0;
-}
-
 /* Side effect free 64 bit mod operation */
 static inline __u32 xfs_do_mod(void *a, __u32 b, int n)
 {
@@ -310,10 +228,7 @@ static inline __u32 xfs_do_mod(void *a, __u32 b, int n)
 	/* NOTREACHED */
 	return 0;
 }
-#endif
 
-#undef do_div
-#define do_div(a, b)	xfs_do_div(&(a), (b), sizeof(a))
 #define do_mod(a, b)	xfs_do_mod(&(a), (b), sizeof(a))
 
 static inline __uint64_t roundup_64(__uint64_t x, __uint32_t y)

fs/xfs/xfs_log.c

@@ -1293,7 +1293,7 @@ void
 xfs_log_work_queue(
 	struct xfs_mount        *mp)
 {
-	queue_delayed_work(mp->m_log_workqueue, &mp->m_log->l_work,
+	queue_delayed_work(mp->m_sync_workqueue, &mp->m_log->l_work,
 				msecs_to_jiffies(xfs_syncd_centisecs * 10));
 }
 
@@ -1852,7 +1852,7 @@ xlog_sync(
 	 */
 	if (log->l_badcrc_factor &&
 	    (prandom_u32() % log->l_badcrc_factor == 0)) {
-		iclog->ic_header.h_crc &= 0xAAAAAAAA;
+		iclog->ic_header.h_crc &= cpu_to_le32(0xAAAAAAAA);
 		iclog->ic_state |= XLOG_STATE_IOABORT;
 		xfs_warn(log->l_mp,
 	"Intentionally corrupted log record at LSN 0x%llx. Shutdown imminent.",

fs/xfs/xfs_mount.c

@@ -73,6 +73,10 @@ xfs_uuid_mount(
 	uuid_t			*uuid = &mp->m_sb.sb_uuid;
 	int			hole, i;
 
+	/* Publish UUID in struct super_block */
+	BUILD_BUG_ON(sizeof(mp->m_super->s_uuid) != sizeof(uuid_t));
+	memcpy(&mp->m_super->s_uuid, uuid, sizeof(uuid_t));
+
 	if (mp->m_flags & XFS_MOUNT_NOUUID)
 		return 0;
 

fs/xfs/xfs_mount.h

@@ -183,6 +183,7 @@ typedef struct xfs_mount {
 	struct workqueue_struct	*m_reclaim_workqueue;
 	struct workqueue_struct	*m_log_workqueue;
 	struct workqueue_struct *m_eofblocks_workqueue;
+	struct workqueue_struct	*m_sync_workqueue;
 
 	/*
 	 * Generation of the filesysyem layout.  This is incremented by each
@@ -312,7 +313,7 @@ void xfs_do_force_shutdown(struct xfs_mount *mp, int flags, char *fname,
 static inline xfs_agnumber_t
 xfs_daddr_to_agno(struct xfs_mount *mp, xfs_daddr_t d)
 {
-	xfs_daddr_t ld = XFS_BB_TO_FSBT(mp, d);
+	xfs_rfsblock_t ld = XFS_BB_TO_FSBT(mp, d);
 	do_div(ld, mp->m_sb.sb_agblocks);
 	return (xfs_agnumber_t) ld;
 }
@@ -320,7 +321,7 @@ xfs_daddr_to_agno(struct xfs_mount *mp, xfs_daddr_t d)
 static inline xfs_agblock_t
 xfs_daddr_to_agbno(struct xfs_mount *mp, xfs_daddr_t d)
 {
-	xfs_daddr_t ld = XFS_BB_TO_FSBT(mp, d);
+	xfs_rfsblock_t ld = XFS_BB_TO_FSBT(mp, d);
 	return (xfs_agblock_t) do_div(ld, mp->m_sb.sb_agblocks);
 }
 

fs/xfs/xfs_qm.c

@@ -851,8 +851,8 @@ xfs_qm_reset_dqcounts(
 	 * started afresh by xfs_qm_quotacheck.
 	 */
 #ifdef DEBUG
-	j = XFS_FSB_TO_B(mp, XFS_DQUOT_CLUSTER_SIZE_FSB);
-	do_div(j, sizeof(xfs_dqblk_t));
+	j = (int)XFS_FSB_TO_B(mp, XFS_DQUOT_CLUSTER_SIZE_FSB) /
+		sizeof(xfs_dqblk_t);
 	ASSERT(mp->m_quotainfo->qi_dqperchunk == j);
 #endif
 	dqb = bp->b_addr;
@@ -1384,12 +1384,7 @@ xfs_qm_quotacheck(
 	mp->m_qflags |= flags;
 
  error_return:
-	while (!list_empty(&buffer_list)) {
-		struct xfs_buf *bp =
-			list_first_entry(&buffer_list, struct xfs_buf, b_list);
-		list_del_init(&bp->b_list);
-		xfs_buf_relse(bp);
-	}
+	xfs_buf_delwri_cancel(&buffer_list);
 
 	if (error) {
 		xfs_warn(mp,

fs/xfs/xfs_qm_syscalls.c

@@ -759,5 +759,6 @@ xfs_qm_dqrele_all_inodes(
 	uint		 flags)
 {
 	ASSERT(mp->m_quotainfo);
-	xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags, NULL);
+	xfs_inode_ag_iterator_flags(mp, xfs_dqrele_inode, flags, NULL,
+				    XFS_AGITER_INEW_WAIT);
 }

fs/xfs/xfs_refcount_item.c

@@ -221,6 +221,7 @@ void
 xfs_cui_release(
 	struct xfs_cui_log_item	*cuip)
 {
+	ASSERT(atomic_read(&cuip->cui_refcount) > 0);
 	if (atomic_dec_and_test(&cuip->cui_refcount)) {
 		xfs_trans_ail_remove(&cuip->cui_item, SHUTDOWN_LOG_IO_ERROR);
 		xfs_cui_item_free(cuip);

fs/xfs/xfs_reflink.c

@@ -206,11 +206,7 @@ xfs_reflink_trim_around_shared(
 	int			error = 0;
 
 	/* Holes, unwritten, and delalloc extents cannot be shared */
-	if (!xfs_is_reflink_inode(ip) ||
-	    ISUNWRITTEN(irec) ||
-	    irec->br_startblock == HOLESTARTBLOCK ||
-	    irec->br_startblock == DELAYSTARTBLOCK ||
-	    isnullstartblock(irec->br_startblock)) {
+	if (!xfs_is_reflink_inode(ip) || !xfs_bmap_is_real_extent(irec)) {
 		*shared = false;
 		return 0;
 	}
@@ -709,8 +705,22 @@ xfs_reflink_end_cow(
 	offset_fsb = XFS_B_TO_FSBT(ip->i_mount, offset);
 	end_fsb = XFS_B_TO_FSB(ip->i_mount, offset + count);
 
-	/* Start a rolling transaction to switch the mappings */
-	resblks = XFS_EXTENTADD_SPACE_RES(ip->i_mount, XFS_DATA_FORK);
+	/*
+	 * Start a rolling transaction to switch the mappings.  We're
+	 * unlikely ever to have to remap 16T worth of single-block
+	 * extents, so just cap the worst case extent count to 2^32-1.
+	 * Stick a warning in just in case, and avoid 64-bit division.
+	 */
+	BUILD_BUG_ON(MAX_RW_COUNT > UINT_MAX);
+	if (end_fsb - offset_fsb > UINT_MAX) {
+		error = -EFSCORRUPTED;
+		xfs_force_shutdown(ip->i_mount, SHUTDOWN_CORRUPT_INCORE);
+		ASSERT(0);
+		goto out;
+	}
+	resblks = XFS_NEXTENTADD_SPACE_RES(ip->i_mount,
+			(unsigned int)(end_fsb - offset_fsb),
+			XFS_DATA_FORK);
 	error = xfs_trans_alloc(ip->i_mount, &M_RES(ip->i_mount)->tr_write,
 			resblks, 0, 0, &tp);
 	if (error)
@@ -1045,12 +1055,12 @@ xfs_reflink_remap_extent(
 	xfs_off_t		new_isize)
 {
 	struct xfs_mount	*mp = ip->i_mount;
+	bool			real_extent = xfs_bmap_is_real_extent(irec);
 	struct xfs_trans	*tp;
 	xfs_fsblock_t		firstfsb;
 	unsigned int		resblks;
 	struct xfs_defer_ops	dfops;
 	struct xfs_bmbt_irec	uirec;
-	bool			real_extent;
 	xfs_filblks_t		rlen;
 	xfs_filblks_t		unmap_len;
 	xfs_off_t		newlen;
@@ -1059,11 +1069,6 @@ xfs_reflink_remap_extent(
 	unmap_len = irec->br_startoff + irec->br_blockcount - destoff;
 	trace_xfs_reflink_punch_range(ip, destoff, unmap_len);
 
-	/* Only remap normal extents. */
-	real_extent =  (irec->br_startblock != HOLESTARTBLOCK &&
-			irec->br_startblock != DELAYSTARTBLOCK &&
-			!ISUNWRITTEN(irec));
-
 	/* No reflinking if we're low on space */
 	if (real_extent) {
 		error = xfs_reflink_ag_has_free_space(mp,
@@ -1359,9 +1364,7 @@ xfs_reflink_dirty_extents(
 			goto out;
 		if (nmaps == 0)
 			break;
-		if (map[0].br_startblock == HOLESTARTBLOCK ||
-		    map[0].br_startblock == DELAYSTARTBLOCK ||
-		    ISUNWRITTEN(&map[0]))
+		if (!xfs_bmap_is_real_extent(&map[0]))
 			goto next;
 
 		map[1] = map[0];
@@ -1435,9 +1438,7 @@ xfs_reflink_clear_inode_flag(
 			return error;
 		if (nmaps == 0)
 			break;
-		if (map.br_startblock == HOLESTARTBLOCK ||
-		    map.br_startblock == DELAYSTARTBLOCK ||
-		    ISUNWRITTEN(&map))
+		if (!xfs_bmap_is_real_extent(&map))
 			goto next;
 
 		agno = XFS_FSB_TO_AGNO(mp, map.br_startblock);

fs/xfs/xfs_rmap_item.c

@@ -243,6 +243,7 @@ void
 xfs_rui_release(
 	struct xfs_rui_log_item	*ruip)
 {
+	ASSERT(atomic_read(&ruip->rui_refcount) > 0);
 	if (atomic_dec_and_test(&ruip->rui_refcount)) {
 		xfs_trans_ail_remove(&ruip->rui_item, SHUTDOWN_LOG_IO_ERROR);
 		xfs_rui_item_free(ruip);

fs/xfs/xfs_rtalloc.h

@@ -23,6 +23,16 @@
 struct xfs_mount;
 struct xfs_trans;
 
+struct xfs_rtalloc_rec {
+	xfs_rtblock_t		ar_startblock;
+	xfs_rtblock_t		ar_blockcount;
+};
+
+typedef int (*xfs_rtalloc_query_range_fn)(
+	struct xfs_trans	*tp,
+	struct xfs_rtalloc_rec	*rec,
+	void			*priv);
+
 #ifdef CONFIG_XFS_RT
 /*
  * Function prototypes for exported functions.
@@ -118,13 +128,21 @@ int xfs_rtmodify_summary(struct xfs_mount *mp, struct xfs_trans *tp, int log,
 int xfs_rtfree_range(struct xfs_mount *mp, struct xfs_trans *tp,
 		     xfs_rtblock_t start, xfs_extlen_t len,
 		     struct xfs_buf **rbpp, xfs_fsblock_t *rsb);
-
-
+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);
+int xfs_rtalloc_query_all(struct xfs_trans *tp,
+			  xfs_rtalloc_query_range_fn fn,
+			  void *priv);
 #else
 # define xfs_rtallocate_extent(t,b,min,max,l,f,p,rb)    (ENOSYS)
 # define xfs_rtfree_extent(t,b,l)                       (ENOSYS)
 # define xfs_rtpick_extent(m,t,l,rb)                    (ENOSYS)
 # define xfs_growfs_rt(mp,in)                           (ENOSYS)
+# define xfs_rtalloc_query_range(t,l,h,f,p)             (ENOSYS)
+# define xfs_rtalloc_query_all(t,f,p)                   (ENOSYS)
 static inline int		/* error */
 xfs_rtmount_init(
 	xfs_mount_t	*mp)	/* file system mount structure */

fs/xfs/xfs_super.c

@@ -877,8 +877,15 @@ xfs_init_mount_workqueues(
 	if (!mp->m_eofblocks_workqueue)
 		goto out_destroy_log;
 
+	mp->m_sync_workqueue = alloc_workqueue("xfs-sync/%s", WQ_FREEZABLE, 0,
+					       mp->m_fsname);
+	if (!mp->m_sync_workqueue)
+		goto out_destroy_eofb;
+
 	return 0;
 
+out_destroy_eofb:
+	destroy_workqueue(mp->m_eofblocks_workqueue);
 out_destroy_log:
 	destroy_workqueue(mp->m_log_workqueue);
 out_destroy_reclaim:
@@ -899,6 +906,7 @@ STATIC void
 xfs_destroy_mount_workqueues(
 	struct xfs_mount	*mp)
 {
+	destroy_workqueue(mp->m_sync_workqueue);
 	destroy_workqueue(mp->m_eofblocks_workqueue);
 	destroy_workqueue(mp->m_log_workqueue);
 	destroy_workqueue(mp->m_reclaim_workqueue);

fs/xfs/xfs_trace.c

@@ -47,6 +47,7 @@
 #include "xfs_inode_item.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_filestream.h"
+#include "xfs_fsmap.h"
 
 /*
  * We include this last to have the helpers above available for the trace

fs/xfs/xfs_trace.h

@@ -40,6 +40,8 @@ struct xfs_inode_log_format;
 struct xfs_bmbt_irec;
 struct xfs_btree_cur;
 struct xfs_refcount_irec;
+struct xfs_fsmap;
+struct xfs_rmap_irec;
 
 DECLARE_EVENT_CLASS(xfs_attr_list_class,
 	TP_PROTO(struct xfs_attr_list_context *ctx),
@@ -2190,7 +2192,7 @@ DECLARE_EVENT_CLASS(xfs_discard_class,
 		__entry->agbno = agbno;
 		__entry->len = len;
 	),
-	TP_printk("dev %d:%d agno %u agbno %u len %u\n",
+	TP_printk("dev %d:%d agno %u agbno %u len %u",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->agno,
 		  __entry->agbno,
@@ -2253,8 +2255,8 @@ DECLARE_EVENT_CLASS(xfs_defer_class,
 	TP_STRUCT__entry(
 		__field(dev_t, dev)
 		__field(void *, dop)
-		__field(bool, committed)
-		__field(bool, low)
+		__field(char, committed)
+		__field(char, low)
 	),
 	TP_fast_assign(
 		__entry->dev = mp ? mp->m_super->s_dev : 0;
@@ -2262,7 +2264,7 @@ DECLARE_EVENT_CLASS(xfs_defer_class,
 		__entry->committed = dop->dop_committed;
 		__entry->low = dop->dop_low;
 	),
-	TP_printk("dev %d:%d ops %p committed %d low %d\n",
+	TP_printk("dev %d:%d ops %p committed %d low %d",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->dop,
 		  __entry->committed,
@@ -2279,8 +2281,8 @@ DECLARE_EVENT_CLASS(xfs_defer_error_class,
 	TP_STRUCT__entry(
 		__field(dev_t, dev)
 		__field(void *, dop)
-		__field(bool, committed)
-		__field(bool, low)
+		__field(char, committed)
+		__field(char, low)
 		__field(int, error)
 	),
 	TP_fast_assign(
@@ -2290,7 +2292,7 @@ DECLARE_EVENT_CLASS(xfs_defer_error_class,
 		__entry->low = dop->dop_low;
 		__entry->error = error;
 	),
-	TP_printk("dev %d:%d ops %p committed %d low %d err %d\n",
+	TP_printk("dev %d:%d ops %p committed %d low %d err %d",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->dop,
 		  __entry->committed,
@@ -2309,7 +2311,7 @@ DECLARE_EVENT_CLASS(xfs_defer_pending_class,
 		__field(dev_t, dev)
 		__field(int, type)
 		__field(void *, intent)
-		__field(bool, committed)
+		__field(char, committed)
 		__field(int, nr)
 	),
 	TP_fast_assign(
@@ -2319,7 +2321,7 @@ DECLARE_EVENT_CLASS(xfs_defer_pending_class,
 		__entry->committed = dfp->dfp_done != NULL;
 		__entry->nr = dfp->dfp_count;
 	),
-	TP_printk("dev %d:%d optype %d intent %p committed %d nr %d\n",
+	TP_printk("dev %d:%d optype %d intent %p committed %d nr %d",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->type,
 		  __entry->intent,
@@ -2614,7 +2616,8 @@ DECLARE_EVENT_CLASS(xfs_ag_resv_class,
 		__entry->asked = r ? r->ar_asked : 0;
 		__entry->len = len;
 	),
-	TP_printk("dev %d:%d agno %u resv %d freeblks %u flcount %u resv %u ask %u len %u\n",
+	TP_printk("dev %d:%d agno %u resv %d freeblks %u flcount %u "
+		  "resv %u ask %u len %u",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->agno,
 		  __entry->resv,
@@ -2667,7 +2670,7 @@ DECLARE_EVENT_CLASS(xfs_ag_btree_lookup_class,
 		__entry->agbno = agbno;
 		__entry->dir = dir;
 	),
-	TP_printk("dev %d:%d agno %u agbno %u cmp %s(%d)\n",
+	TP_printk("dev %d:%d agno %u agbno %u cmp %s(%d)",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->agno,
 		  __entry->agbno,
@@ -2700,7 +2703,7 @@ DECLARE_EVENT_CLASS(xfs_refcount_extent_class,
 		__entry->blockcount = irec->rc_blockcount;
 		__entry->refcount = irec->rc_refcount;
 	),
-	TP_printk("dev %d:%d agno %u agbno %u len %u refcount %u\n",
+	TP_printk("dev %d:%d agno %u agbno %u len %u refcount %u",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->agno,
 		  __entry->startblock,
@@ -2735,7 +2738,7 @@ DECLARE_EVENT_CLASS(xfs_refcount_extent_at_class,
 		__entry->refcount = irec->rc_refcount;
 		__entry->agbno = agbno;
 	),
-	TP_printk("dev %d:%d agno %u agbno %u len %u refcount %u @ agbno %u\n",
+	TP_printk("dev %d:%d agno %u agbno %u len %u refcount %u @ agbno %u",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->agno,
 		  __entry->startblock,
@@ -2776,7 +2779,7 @@ DECLARE_EVENT_CLASS(xfs_refcount_double_extent_class,
 		__entry->i2_refcount = i2->rc_refcount;
 	),
 	TP_printk("dev %d:%d agno %u agbno %u len %u refcount %u -- "
-		  "agbno %u len %u refcount %u\n",
+		  "agbno %u len %u refcount %u",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->agno,
 		  __entry->i1_startblock,
@@ -2822,7 +2825,7 @@ DECLARE_EVENT_CLASS(xfs_refcount_double_extent_at_class,
 		__entry->agbno = agbno;
 	),
 	TP_printk("dev %d:%d agno %u agbno %u len %u refcount %u -- "
-		  "agbno %u len %u refcount %u @ agbno %u\n",
+		  "agbno %u len %u refcount %u @ agbno %u",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->agno,
 		  __entry->i1_startblock,
@@ -2875,7 +2878,7 @@ DECLARE_EVENT_CLASS(xfs_refcount_triple_extent_class,
 	),
 	TP_printk("dev %d:%d agno %u agbno %u len %u refcount %u -- "
 		  "agbno %u len %u refcount %u -- "
-		  "agbno %u len %u refcount %u\n",
+		  "agbno %u len %u refcount %u",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->agno,
 		  __entry->i1_startblock,
@@ -3001,31 +3004,6 @@ DEFINE_EVENT(xfs_inode_error_class, name, \
 		 unsigned long caller_ip), \
 	TP_ARGS(ip, error, caller_ip))
 
-/* reflink allocator */
-TRACE_EVENT(xfs_bmap_remap_alloc,
-	TP_PROTO(struct xfs_inode *ip, xfs_fsblock_t fsbno,
-		 xfs_extlen_t len),
-	TP_ARGS(ip, fsbno, len),
-	TP_STRUCT__entry(
-		__field(dev_t, dev)
-		__field(xfs_ino_t, ino)
-		__field(xfs_fsblock_t, fsbno)
-		__field(xfs_extlen_t, len)
-	),
-	TP_fast_assign(
-		__entry->dev = VFS_I(ip)->i_sb->s_dev;
-		__entry->ino = ip->i_ino;
-		__entry->fsbno = fsbno;
-		__entry->len = len;
-	),
-	TP_printk("dev %d:%d ino 0x%llx fsbno 0x%llx len %x",
-		  MAJOR(__entry->dev), MINOR(__entry->dev),
-		  __entry->ino,
-		  __entry->fsbno,
-		  __entry->len)
-);
-DEFINE_INODE_ERROR_EVENT(xfs_bmap_remap_alloc_error);
-
 /* reflink tracepoint classes */
 
 /* two-file io tracepoint class */
@@ -3227,7 +3205,7 @@ TRACE_EVENT(xfs_ioctl_clone,
 	),
 	TP_printk("dev %d:%d "
 		  "ino 0x%lx isize 0x%llx -> "
-		  "ino 0x%lx isize 0x%llx\n",
+		  "ino 0x%lx isize 0x%llx",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->src_ino,
 		  __entry->src_isize,
@@ -3267,6 +3245,88 @@ DEFINE_INODE_IREC_EVENT(xfs_swap_extent_rmap_remap);
 DEFINE_INODE_IREC_EVENT(xfs_swap_extent_rmap_remap_piece);
 DEFINE_INODE_ERROR_EVENT(xfs_swap_extent_rmap_error);
 
+/* fsmap traces */
+DECLARE_EVENT_CLASS(xfs_fsmap_class,
+	TP_PROTO(struct xfs_mount *mp, u32 keydev, xfs_agnumber_t agno,
+		 struct xfs_rmap_irec *rmap),
+	TP_ARGS(mp, keydev, agno, rmap),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(dev_t, keydev)
+		__field(xfs_agnumber_t, agno)
+		__field(xfs_fsblock_t, bno)
+		__field(xfs_filblks_t, len)
+		__field(__uint64_t, owner)
+		__field(__uint64_t, offset)
+		__field(unsigned int, flags)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->keydev = new_decode_dev(keydev);
+		__entry->agno = agno;
+		__entry->bno = rmap->rm_startblock;
+		__entry->len = rmap->rm_blockcount;
+		__entry->owner = rmap->rm_owner;
+		__entry->offset = rmap->rm_offset;
+		__entry->flags = rmap->rm_flags;
+	),
+	TP_printk("dev %d:%d keydev %d:%d agno %u bno %llu len %llu owner %lld offset %llu flags 0x%x",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  MAJOR(__entry->keydev), MINOR(__entry->keydev),
+		  __entry->agno,
+		  __entry->bno,
+		  __entry->len,
+		  __entry->owner,
+		  __entry->offset,
+		  __entry->flags)
+)
+#define DEFINE_FSMAP_EVENT(name) \
+DEFINE_EVENT(xfs_fsmap_class, name, \
+	TP_PROTO(struct xfs_mount *mp, u32 keydev, xfs_agnumber_t agno, \
+		 struct xfs_rmap_irec *rmap), \
+	TP_ARGS(mp, keydev, agno, rmap))
+DEFINE_FSMAP_EVENT(xfs_fsmap_low_key);
+DEFINE_FSMAP_EVENT(xfs_fsmap_high_key);
+DEFINE_FSMAP_EVENT(xfs_fsmap_mapping);
+
+DECLARE_EVENT_CLASS(xfs_getfsmap_class,
+	TP_PROTO(struct xfs_mount *mp, struct xfs_fsmap *fsmap),
+	TP_ARGS(mp, fsmap),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(dev_t, keydev)
+		__field(xfs_daddr_t, block)
+		__field(xfs_daddr_t, len)
+		__field(__uint64_t, owner)
+		__field(__uint64_t, offset)
+		__field(__uint64_t, flags)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->keydev = new_decode_dev(fsmap->fmr_device);
+		__entry->block = fsmap->fmr_physical;
+		__entry->len = fsmap->fmr_length;
+		__entry->owner = fsmap->fmr_owner;
+		__entry->offset = fsmap->fmr_offset;
+		__entry->flags = fsmap->fmr_flags;
+	),
+	TP_printk("dev %d:%d keydev %d:%d block %llu len %llu owner %lld offset %llu flags 0x%llx",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  MAJOR(__entry->keydev), MINOR(__entry->keydev),
+		  __entry->block,
+		  __entry->len,
+		  __entry->owner,
+		  __entry->offset,
+		  __entry->flags)
+)
+#define DEFINE_GETFSMAP_EVENT(name) \
+DEFINE_EVENT(xfs_getfsmap_class, name, \
+	TP_PROTO(struct xfs_mount *mp, struct xfs_fsmap *fsmap), \
+	TP_ARGS(mp, fsmap))
+DEFINE_GETFSMAP_EVENT(xfs_getfsmap_low_key);
+DEFINE_GETFSMAP_EVENT(xfs_getfsmap_high_key);
+DEFINE_GETFSMAP_EVENT(xfs_getfsmap_mapping);
+
 #endif /* _TRACE_XFS_H */
 
 #undef TRACE_INCLUDE_PATH

fs/xfs/xfs_trans.c

@@ -262,6 +262,28 @@ xfs_trans_alloc(
 	return 0;
 }
 
+/*
+ * Create an empty transaction with no reservation.  This is a defensive
+ * mechanism for routines that query metadata without actually modifying
+ * them -- if the metadata being queried is somehow cross-linked (think a
+ * btree block pointer that points higher in the tree), we risk deadlock.
+ * However, blocks grabbed as part of a transaction can be re-grabbed.
+ * The verifiers will notice the corrupt block and the operation will fail
+ * back to userspace without deadlocking.
+ *
+ * Note the zero-length reservation; this transaction MUST be cancelled
+ * without any dirty data.
+ */
+int
+xfs_trans_alloc_empty(
+	struct xfs_mount		*mp,
+	struct xfs_trans		**tpp)
+{
+	struct xfs_trans_res		resv = {0};
+
+	return xfs_trans_alloc(mp, &resv, 0, 0, XFS_TRANS_NO_WRITECOUNT, tpp);
+}
+
 /*
  * Record the indicated change to the given field for application
  * to the file system's superblock when the transaction commits.
@@ -1012,17 +1034,14 @@ xfs_trans_cancel(
  * chunk we've been working on and get a new transaction to continue.
  */
 int
-__xfs_trans_roll(
+xfs_trans_roll(
 	struct xfs_trans	**tpp,
-	struct xfs_inode	*dp,
-	int			*committed)
+	struct xfs_inode	*dp)
 {
 	struct xfs_trans	*trans;
 	struct xfs_trans_res	tres;
 	int			error;
 
-	*committed = 0;
-
 	/*
 	 * Ensure that the inode is always logged.
 	 */
@@ -1048,7 +1067,6 @@ __xfs_trans_roll(
 	if (error)
 		return error;
 
-	*committed = 1;
 	trans = *tpp;
 
 	/*
@@ -1071,12 +1089,3 @@ __xfs_trans_roll(
 		xfs_trans_ijoin(trans, dp, 0);
 	return 0;
 }
-
-int
-xfs_trans_roll(
-	struct xfs_trans	**tpp,
-	struct xfs_inode	*dp)
-{
-	int			committed;
-	return __xfs_trans_roll(tpp, dp, &committed);
-}

fs/xfs/xfs_trans.h

@@ -158,6 +158,8 @@ typedef struct xfs_trans {
 int		xfs_trans_alloc(struct xfs_mount *mp, struct xfs_trans_res *resp,
 			uint blocks, uint rtextents, uint flags,
 			struct xfs_trans **tpp);
+int		xfs_trans_alloc_empty(struct xfs_mount *mp,
+			struct xfs_trans **tpp);
 void		xfs_trans_mod_sb(xfs_trans_t *, uint, int64_t);
 
 struct xfs_buf	*xfs_trans_get_buf_map(struct xfs_trans *tp,
@@ -226,7 +228,6 @@ int		xfs_trans_free_extent(struct xfs_trans *,
 				      struct xfs_efd_log_item *, xfs_fsblock_t,
 				      xfs_extlen_t, struct xfs_owner_info *);
 int		xfs_trans_commit(struct xfs_trans *);
-int		__xfs_trans_roll(struct xfs_trans **, struct xfs_inode *, int *);
 int		xfs_trans_roll(struct xfs_trans **, struct xfs_inode *);
 void		xfs_trans_cancel(xfs_trans_t *);
 int		xfs_trans_ail_init(struct xfs_mount *);

fs/xfs/xfs_trans_ail.c

@@ -684,8 +684,23 @@ xfs_trans_ail_update_bulk(
 	}
 }
 
-/*
- * xfs_trans_ail_delete_bulk - remove multiple log items from the AIL
+bool
+xfs_ail_delete_one(
+	struct xfs_ail		*ailp,
+	struct xfs_log_item 	*lip)
+{
+	struct xfs_log_item	*mlip = xfs_ail_min(ailp);
+
+	trace_xfs_ail_delete(lip, mlip->li_lsn, lip->li_lsn);
+	xfs_ail_delete(ailp, lip);
+	lip->li_flags &= ~XFS_LI_IN_AIL;
+	lip->li_lsn = 0;
+
+	return mlip == lip;
+}
+
+/**
+ * Remove a log items from the AIL
  *
  * @xfs_trans_ail_delete_bulk takes an array of log items that all need to
  * removed from the AIL. The caller is already holding the AIL lock, and done
@@ -706,52 +721,36 @@ xfs_trans_ail_update_bulk(
  * before returning.
  */
 void
-xfs_trans_ail_delete_bulk(
+xfs_trans_ail_delete(
 	struct xfs_ail		*ailp,
-	struct xfs_log_item	**log_items,
-	int			nr_items,
+	struct xfs_log_item	*lip,
 	int			shutdown_type) __releases(ailp->xa_lock)
 {
-	xfs_log_item_t		*mlip;
-	int			mlip_changed = 0;
-	int			i;
+	struct xfs_mount	*mp = ailp->xa_mount;
+	bool			mlip_changed;
 
-	mlip = xfs_ail_min(ailp);
-
-	for (i = 0; i < nr_items; i++) {
-		struct xfs_log_item *lip = log_items[i];
-		if (!(lip->li_flags & XFS_LI_IN_AIL)) {
-			struct xfs_mount	*mp = ailp->xa_mount;
-
-			spin_unlock(&ailp->xa_lock);
-			if (!XFS_FORCED_SHUTDOWN(mp)) {
-				xfs_alert_tag(mp, XFS_PTAG_AILDELETE,
-		"%s: attempting to delete a log item that is not in the AIL",
-						__func__);
-				xfs_force_shutdown(mp, shutdown_type);
-			}
-			return;
+	if (!(lip->li_flags & XFS_LI_IN_AIL)) {
+		spin_unlock(&ailp->xa_lock);
+		if (!XFS_FORCED_SHUTDOWN(mp)) {
+			xfs_alert_tag(mp, XFS_PTAG_AILDELETE,
+	"%s: attempting to delete a log item that is not in the AIL",
+					__func__);
+			xfs_force_shutdown(mp, shutdown_type);
 		}
-
-		trace_xfs_ail_delete(lip, mlip->li_lsn, lip->li_lsn);
-		xfs_ail_delete(ailp, lip);
-		lip->li_flags &= ~XFS_LI_IN_AIL;
-		lip->li_lsn = 0;
-		if (mlip == lip)
-			mlip_changed = 1;
+		return;
 	}
 
+	mlip_changed = xfs_ail_delete_one(ailp, lip);
 	if (mlip_changed) {
-		if (!XFS_FORCED_SHUTDOWN(ailp->xa_mount))
-			xlog_assign_tail_lsn_locked(ailp->xa_mount);
+		if (!XFS_FORCED_SHUTDOWN(mp))
+			xlog_assign_tail_lsn_locked(mp);
 		if (list_empty(&ailp->xa_ail))
 			wake_up_all(&ailp->xa_empty);
-		spin_unlock(&ailp->xa_lock);
+	}
 
+	spin_unlock(&ailp->xa_lock);
+	if (mlip_changed)
 		xfs_log_space_wake(ailp->xa_mount);
-	} else {
-		spin_unlock(&ailp->xa_lock);
-	}
 }
 
 int

fs/xfs/xfs_trans_priv.h

@@ -106,18 +106,9 @@ xfs_trans_ail_update(
 	xfs_trans_ail_update_bulk(ailp, NULL, &lip, 1, lsn);
 }
 
-void	xfs_trans_ail_delete_bulk(struct xfs_ail *ailp,
-				struct xfs_log_item **log_items, int nr_items,
-				int shutdown_type)
-				__releases(ailp->xa_lock);
-static inline void
-xfs_trans_ail_delete(
-	struct xfs_ail	*ailp,
-	xfs_log_item_t	*lip,
-	int		shutdown_type) __releases(ailp->xa_lock)
-{
-	xfs_trans_ail_delete_bulk(ailp, &lip, 1, shutdown_type);
-}
+bool xfs_ail_delete_one(struct xfs_ail *ailp, struct xfs_log_item *lip);
+void xfs_trans_ail_delete(struct xfs_ail *ailp, struct xfs_log_item *lip,
+		int shutdown_type) __releases(ailp->xa_lock);
 
 static inline void
 xfs_trans_ail_remove(

include/uapi/linux/fsmap.h

@@ -0,0 +1,112 @@
+/*
+ * FS_IOC_GETFSMAP ioctl infrastructure.
+ *
+ * Copyright (C) 2017 Oracle.  All Rights Reserved.
+ *
+ * Author: Darrick J. Wong <darrick.wong@oracle.com>
+ */
+#ifndef _LINUX_FSMAP_H
+#define _LINUX_FSMAP_H
+
+#include <linux/types.h>
+
+/*
+ *	Structure for FS_IOC_GETFSMAP.
+ *
+ *	The memory layout for this call are the scalar values defined in
+ *	struct fsmap_head, followed by two struct fsmap that describe
+ *	the lower and upper bound of mappings to return, followed by an
+ *	array of struct fsmap mappings.
+ *
+ *	fmh_iflags control the output of the call, whereas fmh_oflags report
+ *	on the overall record output.  fmh_count should be set to the
+ *	length of the fmh_recs array, and fmh_entries will be set to the
+ *	number of entries filled out during each call.  If fmh_count is
+ *	zero, the number of reverse mappings will be returned in
+ *	fmh_entries, though no mappings will be returned.  fmh_reserved
+ *	must be set to zero.
+ *
+ *	The two elements in the fmh_keys array are used to constrain the
+ *	output.  The first element in the array should represent the
+ *	lowest disk mapping ("low key") that the user wants to learn
+ *	about.  If this value is all zeroes, the filesystem will return
+ *	the first entry it knows about.  For a subsequent call, the
+ *	contents of fsmap_head.fmh_recs[fsmap_head.fmh_count - 1] should be
+ *	copied into fmh_keys[0] to have the kernel start where it left off.
+ *
+ *	The second element in the fmh_keys array should represent the
+ *	highest disk mapping ("high key") that the user wants to learn
+ *	about.  If this value is all ones, the filesystem will not stop
+ *	until it runs out of mapping to return or runs out of space in
+ *	fmh_recs.
+ *
+ *	fmr_device can be either a 32-bit cookie representing a device, or
+ *	a 32-bit dev_t if the FMH_OF_DEV_T flag is set.  fmr_physical,
+ *	fmr_offset, and fmr_length are expressed in units of bytes.
+ *	fmr_owner is either an inode number, or a special value if
+ *	FMR_OF_SPECIAL_OWNER is set in fmr_flags.
+ */
+struct fsmap {
+	__u32		fmr_device;	/* device id */
+	__u32		fmr_flags;	/* mapping flags */
+	__u64		fmr_physical;	/* device offset of segment */
+	__u64		fmr_owner;	/* owner id */
+	__u64		fmr_offset;	/* file offset of segment */
+	__u64		fmr_length;	/* length of segment */
+	__u64		fmr_reserved[3];	/* must be zero */
+};
+
+struct fsmap_head {
+	__u32		fmh_iflags;	/* control flags */
+	__u32		fmh_oflags;	/* output flags */
+	__u32		fmh_count;	/* # of entries in array incl. input */
+	__u32		fmh_entries;	/* # of entries filled in (output). */
+	__u64		fmh_reserved[6];	/* must be zero */
+
+	struct fsmap	fmh_keys[2];	/* low and high keys for the mapping search */
+	struct fsmap	fmh_recs[];	/* returned records */
+};
+
+/* Size of an fsmap_head with room for nr records. */
+static inline size_t
+fsmap_sizeof(
+	unsigned int	nr)
+{
+	return sizeof(struct fsmap_head) + nr * sizeof(struct fsmap);
+}
+
+/* Start the next fsmap query at the end of the current query results. */
+static inline void
+fsmap_advance(
+	struct fsmap_head	*head)
+{
+	head->fmh_keys[0] = head->fmh_recs[head->fmh_entries - 1];
+}
+
+/*	fmh_iflags values - set by FS_IOC_GETFSMAP caller in the header. */
+/* no flags defined yet */
+#define FMH_IF_VALID		0
+
+/*	fmh_oflags values - returned in the header segment only. */
+#define FMH_OF_DEV_T		0x1	/* fmr_device values will be dev_t */
+
+/*	fmr_flags values - returned for each non-header segment */
+#define FMR_OF_PREALLOC		0x1	/* segment = unwritten pre-allocation */
+#define FMR_OF_ATTR_FORK	0x2	/* segment = attribute fork */
+#define FMR_OF_EXTENT_MAP	0x4	/* segment = extent map */
+#define FMR_OF_SHARED		0x8	/* segment = shared with another file */
+#define FMR_OF_SPECIAL_OWNER	0x10	/* owner is a special value */
+#define FMR_OF_LAST		0x20	/* segment is the last in the FS */
+
+/* Each FS gets to define its own special owner codes. */
+#define FMR_OWNER(type, code)	(((__u64)type << 32) | \
+				 ((__u64)code & 0xFFFFFFFFULL))
+#define FMR_OWNER_TYPE(owner)	((__u32)((__u64)owner >> 32))
+#define FMR_OWNER_CODE(owner)	((__u32)(((__u64)owner & 0xFFFFFFFFULL)))
+#define FMR_OWN_FREE		FMR_OWNER(0, 1) /* free space */
+#define FMR_OWN_UNKNOWN		FMR_OWNER(0, 2) /* unknown owner */
+#define FMR_OWN_METADATA	FMR_OWNER(0, 3) /* metadata */
+
+#define FS_IOC_GETFSMAP		_IOWR('X', 59, struct fsmap_head)
+
+#endif /* _LINUX_FSMAP_H */