Merge tag 'xfs-4.11-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 4.11. We aren't introducing any major
  features in this release cycle except for this being the first merge
  window I've managed on my own. :)

  Changes since last update:

   - Various cleanups

   - Livelock fixes for eofblocks scanning

   - Improved input verification for on-disk metadata

   - Fix races in the copy on write remap mechanism

   - Fix buffer io error timeout controls

   - Streamlining of directio copy on write

   - Asynchronous discard support

   - Fix asserts when splitting delalloc reservations

   - Don't bloat bmbt when right shifting extents

   - Inode alignment fixes for 32k block sizes"

* tag 'xfs-4.11-merge-7' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux: (39 commits)
  xfs: remove XFS_ALLOCTYPE_ANY_AG and XFS_ALLOCTYPE_START_AG
  xfs: simplify xfs_rtallocate_extent
  xfs: tune down agno asserts in the bmap code
  xfs: Use xfs_icluster_size_fsb() to calculate inode chunk alignment
  xfs: don't reserve blocks for right shift transactions
  xfs: fix len comparison in xfs_extent_busy_trim
  xfs: fix uninitialized variable in _reflink_convert_cow
  xfs: split indlen reservations fairly when under reserved
  xfs: handle indlen shortage on delalloc extent merge
  xfs: resurrect debug mode drop buffered writes mechanism
  xfs: clear delalloc and cache on buffered write failure
  xfs: don't block the log commit handler for discards
  xfs: improve busy extent sorting
  xfs: improve handling of busy extents in the low-level allocator
  xfs: don't fail xfs_extent_busy allocation
  xfs: correct null checks and error processing in xfs_initialize_perag
  xfs: update ctime and mtime on clone destinatation inodes
  xfs: allocate direct I/O COW blocks in iomap_begin
  xfs: go straight to real allocations for direct I/O COW writes
  xfs: return the converted extent in __xfs_reflink_convert_cow
  ...

42 files changed, 891 insertions, 609 deletions

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 9f06a211e157..369adcc18c02 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -221,20 +221,22 @@ xfs_alloc_get_rec(
  * Compute aligned version of the found extent.
  * Takes alignment and min length into account.
  */
-STATIC void
+STATIC bool
 xfs_alloc_compute_aligned(
 	xfs_alloc_arg_t	*args,		/* allocation argument structure */
 	xfs_agblock_t	foundbno,	/* starting block in found extent */
 	xfs_extlen_t	foundlen,	/* length in found extent */
 	xfs_agblock_t	*resbno,	/* result block number */
-	xfs_extlen_t	*reslen)	/* result length */
+	xfs_extlen_t	*reslen,	/* result length */
+	unsigned	*busy_gen)
 {
-	xfs_agblock_t	bno;
-	xfs_extlen_t	len;
+	xfs_agblock_t	bno = foundbno;
+	xfs_extlen_t	len = foundlen;
 	xfs_extlen_t	diff;
+	bool		busy;
 
 	/* Trim busy sections out of found extent */
-	xfs_extent_busy_trim(args, foundbno, foundlen, &bno, &len);
+	busy = xfs_extent_busy_trim(args, &bno, &len, busy_gen);
 
 	/*
 	 * If we have a largish extent that happens to start before min_agbno,
@@ -259,6 +261,8 @@ xfs_alloc_compute_aligned(
 		*resbno = bno;
 		*reslen = len;
 	}
+
+	return busy;
 }
 
 /*
@@ -737,10 +741,11 @@ xfs_alloc_ag_vextent_exact(
 	int		error;
 	xfs_agblock_t	fbno;	/* start block of found extent */
 	xfs_extlen_t	flen;	/* length of found extent */
-	xfs_agblock_t	tbno;	/* start block of trimmed extent */
-	xfs_extlen_t	tlen;	/* length of trimmed extent */
-	xfs_agblock_t	tend;	/* end block of trimmed extent */
+	xfs_agblock_t	tbno;	/* start block of busy extent */
+	xfs_extlen_t	tlen;	/* length of busy extent */
+	xfs_agblock_t	tend;	/* end block of busy extent */
 	int		i;	/* success/failure of operation */
+	unsigned	busy_gen;
 
 	ASSERT(args->alignment == 1);
 
@@ -773,7 +778,9 @@ xfs_alloc_ag_vextent_exact(
 	/*
 	 * Check for overlapping busy extents.
 	 */
-	xfs_extent_busy_trim(args, fbno, flen, &tbno, &tlen);
+	tbno = fbno;
+	tlen = flen;
+	xfs_extent_busy_trim(args, &tbno, &tlen, &busy_gen);
 
 	/*
 	 * Give up if the start of the extent is busy, or the freespace isn't
@@ -853,6 +860,7 @@ xfs_alloc_find_best_extent(
 	xfs_agblock_t		sdiff;
 	int			error;
 	int			i;
+	unsigned		busy_gen;
 
 	/* The good extent is perfect, no need to  search. */
 	if (!gdiff)
@@ -866,7 +874,8 @@ xfs_alloc_find_best_extent(
 		if (error)
 			goto error0;
 		XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-		xfs_alloc_compute_aligned(args, *sbno, *slen, sbnoa, slena);
+		xfs_alloc_compute_aligned(args, *sbno, *slen,
+				sbnoa, slena, &busy_gen);
 
 		/*
 		 * The good extent is closer than this one.
@@ -955,7 +964,8 @@ xfs_alloc_ag_vextent_near(
 	xfs_extlen_t	ltlena;		/* aligned ... */
 	xfs_agblock_t	ltnew;		/* useful start bno of left side */
 	xfs_extlen_t	rlen;		/* length of returned extent */
-	int		forced = 0;
+	bool		busy;
+	unsigned	busy_gen;
 #ifdef DEBUG
 	/*
 	 * Randomly don't execute the first algorithm.
@@ -982,6 +992,7 @@ restart:
 	ltlen = 0;
 	gtlena = 0;
 	ltlena = 0;
+	busy = false;
 
 	/*
 	 * Get a cursor for the by-size btree.
@@ -1064,8 +1075,8 @@ restart:
 			if ((error = xfs_alloc_get_rec(cnt_cur, &ltbno, &ltlen, &i)))
 				goto error0;
 			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-			xfs_alloc_compute_aligned(args, ltbno, ltlen,
-						  &ltbnoa, &ltlena);
+			busy = xfs_alloc_compute_aligned(args, ltbno, ltlen,
+					&ltbnoa, &ltlena, &busy_gen);
 			if (ltlena < args->minlen)
 				continue;
 			if (ltbnoa < args->min_agbno || ltbnoa > args->max_agbno)
@@ -1183,8 +1194,8 @@ restart:
 			if ((error = xfs_alloc_get_rec(bno_cur_lt, &ltbno, &ltlen, &i)))
 				goto error0;
 			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-			xfs_alloc_compute_aligned(args, ltbno, ltlen,
-						  &ltbnoa, &ltlena);
+			busy |= xfs_alloc_compute_aligned(args, ltbno, ltlen,
+					&ltbnoa, &ltlena, &busy_gen);
 			if (ltlena >= args->minlen && ltbnoa >= args->min_agbno)
 				break;
 			if ((error = xfs_btree_decrement(bno_cur_lt, 0, &i)))
@@ -1199,8 +1210,8 @@ restart:
 			if ((error = xfs_alloc_get_rec(bno_cur_gt, &gtbno, &gtlen, &i)))
 				goto error0;
 			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
-			xfs_alloc_compute_aligned(args, gtbno, gtlen,
-						  &gtbnoa, &gtlena);
+			busy |= xfs_alloc_compute_aligned(args, gtbno, gtlen,
+					&gtbnoa, &gtlena, &busy_gen);
 			if (gtlena >= args->minlen && gtbnoa <= args->max_agbno)
 				break;
 			if ((error = xfs_btree_increment(bno_cur_gt, 0, &i)))
@@ -1261,9 +1272,9 @@ restart:
 	if (bno_cur_lt == NULL && bno_cur_gt == NULL) {
 		xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
 
-		if (!forced++) {
+		if (busy) {
 			trace_xfs_alloc_near_busy(args);
-			xfs_log_force(args->mp, XFS_LOG_SYNC);
+			xfs_extent_busy_flush(args->mp, args->pag, busy_gen);
 			goto restart;
 		}
 		trace_xfs_alloc_size_neither(args);
@@ -1344,7 +1355,8 @@ xfs_alloc_ag_vextent_size(
 	int		i;		/* temp status variable */
 	xfs_agblock_t	rbno;		/* returned block number */
 	xfs_extlen_t	rlen;		/* length of returned extent */
-	int		forced = 0;
+	bool		busy;
+	unsigned	busy_gen;
 
 restart:
 	/*
@@ -1353,6 +1365,7 @@ restart:
 	cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, args->agbp,
 		args->agno, XFS_BTNUM_CNT);
 	bno_cur = NULL;
+	busy = false;
 
 	/*
 	 * Look for an entry >= maxlen+alignment-1 blocks.
@@ -1362,14 +1375,13 @@ restart:
 		goto error0;
 
 	/*
-	 * If none or we have busy extents that we cannot allocate from, then
-	 * we have to settle for a smaller extent. In the case that there are
-	 * no large extents, this will return the last entry in the tree unless
-	 * the tree is empty. In the case that there are only busy large
-	 * extents, this will return the largest small extent unless there
+	 * If none then we have to settle for a smaller extent. In the case that
+	 * there are no large extents, this will return the last entry in the
+	 * tree unless the tree is empty. In the case that there are only busy
+	 * large extents, this will return the largest small extent unless there
 	 * are no smaller extents available.
 	 */
-	if (!i || forced > 1) {
+	if (!i) {
 		error = xfs_alloc_ag_vextent_small(args, cnt_cur,
 						   &fbno, &flen, &i);
 		if (error)
@@ -1380,13 +1392,11 @@ restart:
 			return 0;
 		}
 		ASSERT(i == 1);
-		xfs_alloc_compute_aligned(args, fbno, flen, &rbno, &rlen);
+		busy = xfs_alloc_compute_aligned(args, fbno, flen, &rbno,
+				&rlen, &busy_gen);
 	} else {
 		/*
 		 * Search for a non-busy extent that is large enough.
-		 * If we are at low space, don't check, or if we fall of
-		 * the end of the btree, turn off the busy check and
-		 * restart.
 		 */
 		for (;;) {
 			error = xfs_alloc_get_rec(cnt_cur, &fbno, &flen, &i);
@@ -1394,8 +1404,8 @@ restart:
 				goto error0;
 			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
 
-			xfs_alloc_compute_aligned(args, fbno, flen,
-						  &rbno, &rlen);
+			busy = xfs_alloc_compute_aligned(args, fbno, flen,
+					&rbno, &rlen, &busy_gen);
 
 			if (rlen >= args->maxlen)
 				break;
@@ -1407,18 +1417,13 @@ restart:
 				/*
 				 * Our only valid extents must have been busy.
 				 * Make it unbusy by forcing the log out and
-				 * retrying. If we've been here before, forcing
-				 * the log isn't making the extents available,
-				 * which means they have probably been freed in
-				 * this transaction.  In that case, we have to
-				 * give up on them and we'll attempt a minlen
-				 * allocation the next time around.
+				 * retrying.
 				 */
 				xfs_btree_del_cursor(cnt_cur,
 						     XFS_BTREE_NOERROR);
 				trace_xfs_alloc_size_busy(args);
-				if (!forced++)
-					xfs_log_force(args->mp, XFS_LOG_SYNC);
+				xfs_extent_busy_flush(args->mp,
+							args->pag, busy_gen);
 				goto restart;
 			}
 		}
@@ -1454,8 +1459,8 @@ restart:
 			XFS_WANT_CORRUPTED_GOTO(args->mp, i == 1, error0);
 			if (flen < bestrlen)
 				break;
-			xfs_alloc_compute_aligned(args, fbno, flen,
-						  &rbno, &rlen);
+			busy = xfs_alloc_compute_aligned(args, fbno, flen,
+					&rbno, &rlen, &busy_gen);
 			rlen = XFS_EXTLEN_MIN(args->maxlen, rlen);
 			XFS_WANT_CORRUPTED_GOTO(args->mp, rlen == 0 ||
 				(rlen <= flen && rbno + rlen <= fbno + flen),
@@ -1484,10 +1489,10 @@ restart:
 	 */
 	args->len = rlen;
 	if (rlen < args->minlen) {
-		if (!forced++) {
+		if (busy) {
 			xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
 			trace_xfs_alloc_size_busy(args);
-			xfs_log_force(args->mp, XFS_LOG_SYNC);
+			xfs_extent_busy_flush(args->mp, args->pag, busy_gen);
 			goto restart;
 		}
 		goto out_nominleft;
@@ -2659,21 +2664,11 @@ xfs_alloc_vextent(
 		args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno);
 		args->type = XFS_ALLOCTYPE_NEAR_BNO;
 		/* FALLTHROUGH */
-	case XFS_ALLOCTYPE_ANY_AG:
-	case XFS_ALLOCTYPE_START_AG:
 	case XFS_ALLOCTYPE_FIRST_AG:
 		/*
 		 * Rotate through the allocation groups looking for a winner.
 		 */
-		if (type == XFS_ALLOCTYPE_ANY_AG) {
-			/*
-			 * Start with the last place we left off.
-			 */
-			args->agno = sagno = (mp->m_agfrotor / rotorstep) %
-					mp->m_sb.sb_agcount;
-			args->type = XFS_ALLOCTYPE_THIS_AG;
-			flags = XFS_ALLOC_FLAG_TRYLOCK;
-		} else if (type == XFS_ALLOCTYPE_FIRST_AG) {
+		if (type == XFS_ALLOCTYPE_FIRST_AG) {
 			/*
 			 * Start with allocation group given by bno.
 			 */
@@ -2682,8 +2677,6 @@ xfs_alloc_vextent(
 			sagno = 0;
 			flags = 0;
 		} else {
-			if (type == XFS_ALLOCTYPE_START_AG)
-				args->type = XFS_ALLOCTYPE_THIS_AG;
 			/*
 			 * Start with the given allocation group.
 			 */
@@ -2751,7 +2744,7 @@ xfs_alloc_vextent(
 			}
 			xfs_perag_put(args->pag);
 		}
-		if (bump_rotor || (type == XFS_ALLOCTYPE_ANY_AG)) {
+		if (bump_rotor) {
 			if (args->agno == sagno)
 				mp->m_agfrotor = (mp->m_agfrotor + 1) %
 					(mp->m_sb.sb_agcount * rotorstep);
diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h
index 1d0f48a501a3..2a8d0fa6fbbe 100644
--- a/fs/xfs/libxfs/xfs_alloc.h
+++ b/fs/xfs/libxfs/xfs_alloc.h
@@ -29,9 +29,7 @@ extern struct workqueue_struct *xfs_alloc_wq;
 /*
  * Freespace allocation types.  Argument to xfs_alloc_[v]extent.
  */
-#define XFS_ALLOCTYPE_ANY_AG	0x01	/* allocate anywhere, use rotor */
 #define XFS_ALLOCTYPE_FIRST_AG	0x02	/* ... start at ag 0 */
-#define XFS_ALLOCTYPE_START_AG	0x04	/* anywhere, start in this a.g. */
 #define XFS_ALLOCTYPE_THIS_AG	0x08	/* anywhere in this a.g. */
 #define XFS_ALLOCTYPE_START_BNO	0x10	/* near this block else anywhere */
 #define XFS_ALLOCTYPE_NEAR_BNO	0x20	/* in this a.g. and near this block */
@@ -41,9 +39,7 @@ extern struct workqueue_struct *xfs_alloc_wq;
 typedef unsigned int xfs_alloctype_t;
 
 #define XFS_ALLOC_TYPES \
-	{ XFS_ALLOCTYPE_ANY_AG,		"ANY_AG" }, \
 	{ XFS_ALLOCTYPE_FIRST_AG,	"FIRST_AG" }, \
-	{ XFS_ALLOCTYPE_START_AG,	"START_AG" }, \
 	{ XFS_ALLOCTYPE_THIS_AG,	"THIS_AG" }, \
 	{ XFS_ALLOCTYPE_START_BNO,	"START_BNO" }, \
 	{ XFS_ALLOCTYPE_NEAR_BNO,	"NEAR_BNO" }, \
diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index bfc00de5c6f1..a9c66d47757a 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -740,15 +740,9 @@ xfs_bmap_extents_to_btree(
 	 * Fill in the root.
 	 */
 	block = ifp->if_broot;
-	if (xfs_sb_version_hascrc(&mp->m_sb))
-		xfs_btree_init_block_int(mp, block, XFS_BUF_DADDR_NULL,
-				 XFS_BMAP_CRC_MAGIC, 1, 1, ip->i_ino,
-				 XFS_BTREE_LONG_PTRS | XFS_BTREE_CRC_BLOCKS);
-	else
-		xfs_btree_init_block_int(mp, block, XFS_BUF_DADDR_NULL,
-				 XFS_BMAP_MAGIC, 1, 1, ip->i_ino,
+	xfs_btree_init_block_int(mp, block, XFS_BUF_DADDR_NULL,
+				 XFS_BTNUM_BMAP, 1, 1, ip->i_ino,
 				 XFS_BTREE_LONG_PTRS);
-
 	/*
 	 * Need a cursor.  Can't allocate until bb_level is filled in.
 	 */
@@ -804,9 +798,7 @@ try_another_ag:
 	 */
 	ASSERT(args.fsbno != NULLFSBLOCK);
 	ASSERT(*firstblock == NULLFSBLOCK ||
-	       args.agno == XFS_FSB_TO_AGNO(mp, *firstblock) ||
-	       (dfops->dop_low &&
-		args.agno > XFS_FSB_TO_AGNO(mp, *firstblock)));
+	       args.agno >= XFS_FSB_TO_AGNO(mp, *firstblock));
 	*firstblock = cur->bc_private.b.firstblock = args.fsbno;
 	cur->bc_private.b.allocated++;
 	ip->i_d.di_nblocks++;
@@ -817,13 +809,8 @@ try_another_ag:
 	 */
 	abp->b_ops = &xfs_bmbt_buf_ops;
 	ablock = XFS_BUF_TO_BLOCK(abp);
-	if (xfs_sb_version_hascrc(&mp->m_sb))
-		xfs_btree_init_block_int(mp, ablock, abp->b_bn,
-				XFS_BMAP_CRC_MAGIC, 0, 0, ip->i_ino,
-				XFS_BTREE_LONG_PTRS | XFS_BTREE_CRC_BLOCKS);
-	else
-		xfs_btree_init_block_int(mp, ablock, abp->b_bn,
-				XFS_BMAP_MAGIC, 0, 0, ip->i_ino,
+	xfs_btree_init_block_int(mp, ablock, abp->b_bn,
+				XFS_BTNUM_BMAP, 0, 0, ip->i_ino,
 				XFS_BTREE_LONG_PTRS);
 
 	arp = XFS_BMBT_REC_ADDR(mp, ablock, 1);
@@ -1278,7 +1265,6 @@ xfs_bmap_read_extents(
 	/* REFERENCED */
 	xfs_extnum_t		room;	/* number of entries there's room for */
 
-	bno = NULLFSBLOCK;
 	mp = ip->i_mount;
 	ifp = XFS_IFORK_PTR(ip, whichfork);
 	exntf = (whichfork != XFS_DATA_FORK) ? XFS_EXTFMT_NOSTATE :
@@ -1291,9 +1277,7 @@ xfs_bmap_read_extents(
 	ASSERT(level > 0);
 	pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
 	bno = be64_to_cpu(*pp);
-	ASSERT(bno != NULLFSBLOCK);
-	ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
-	ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
+
 	/*
 	 * Go down the tree until leaf level is reached, following the first
 	 * pointer (leftmost) at each level.
@@ -1864,6 +1848,7 @@ xfs_bmap_add_extent_delay_real(
 		 */
 		trace_xfs_bmap_pre_update(bma->ip, bma->idx, state, _THIS_IP_);
 		xfs_bmbt_set_startblock(ep, new->br_startblock);
+		xfs_bmbt_set_state(ep, new->br_state);
 		trace_xfs_bmap_post_update(bma->ip, bma->idx, state, _THIS_IP_);
 
 		(*nextents)++;
@@ -2202,6 +2187,7 @@ STATIC int				/* error */
 xfs_bmap_add_extent_unwritten_real(
 	struct xfs_trans	*tp,
 	xfs_inode_t		*ip,	/* incore inode pointer */
+	int			whichfork,
 	xfs_extnum_t		*idx,	/* extent number to update/insert */
 	xfs_btree_cur_t		**curp,	/* if *curp is null, not a btree */
 	xfs_bmbt_irec_t		*new,	/* new data to add to file extents */
@@ -2221,12 +2207,14 @@ xfs_bmap_add_extent_unwritten_real(
 					/* left is 0, right is 1, prev is 2 */
 	int			rval=0;	/* return value (logging flags) */
 	int			state = 0;/* state bits, accessed thru macros */
-	struct xfs_mount	*mp = tp->t_mountp;
+	struct xfs_mount	*mp = ip->i_mount;
 
 	*logflagsp = 0;
 
 	cur = *curp;
-	ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
+	ifp = XFS_IFORK_PTR(ip, whichfork);
+	if (whichfork == XFS_COW_FORK)
+		state |= BMAP_COWFORK;
 
 	ASSERT(*idx >= 0);
 	ASSERT(*idx <= xfs_iext_count(ifp));
@@ -2285,7 +2273,7 @@ xfs_bmap_add_extent_unwritten_real(
 	 * Don't set contiguous if the combined extent would be too large.
 	 * Also check for all-three-contiguous being too large.
 	 */
-	if (*idx < xfs_iext_count(&ip->i_df) - 1) {
+	if (*idx < xfs_iext_count(ifp) - 1) {
 		state |= BMAP_RIGHT_VALID;
 		xfs_bmbt_get_all(xfs_iext_get_ext(ifp, *idx + 1), &RIGHT);
 		if (isnullstartblock(RIGHT.br_startblock))
@@ -2325,7 +2313,8 @@ xfs_bmap_add_extent_unwritten_real(
 		trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
 
 		xfs_iext_remove(ip, *idx + 1, 2, state);
-		ip->i_d.di_nextents -= 2;
+		XFS_IFORK_NEXT_SET(ip, whichfork,
+				XFS_IFORK_NEXTENTS(ip, whichfork) - 2);
 		if (cur == NULL)
 			rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
 		else {
@@ -2368,7 +2357,8 @@ xfs_bmap_add_extent_unwritten_real(
 		trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
 
 		xfs_iext_remove(ip, *idx + 1, 1, state);
-		ip->i_d.di_nextents--;
+		XFS_IFORK_NEXT_SET(ip, whichfork,
+				XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
 		if (cur == NULL)
 			rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
 		else {
@@ -2403,7 +2393,8 @@ xfs_bmap_add_extent_unwritten_real(
 		xfs_bmbt_set_state(ep, newext);
 		trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
 		xfs_iext_remove(ip, *idx + 1, 1, state);
-		ip->i_d.di_nextents--;
+		XFS_IFORK_NEXT_SET(ip, whichfork,
+				XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
 		if (cur == NULL)
 			rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
 		else {
@@ -2515,7 +2506,8 @@ xfs_bmap_add_extent_unwritten_real(
 		trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
 
 		xfs_iext_insert(ip, *idx, 1, new, state);
-		ip->i_d.di_nextents++;
+		XFS_IFORK_NEXT_SET(ip, whichfork,
+				XFS_IFORK_NEXTENTS(ip, whichfork) + 1);
 		if (cur == NULL)
 			rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
 		else {
@@ -2593,7 +2585,8 @@ xfs_bmap_add_extent_unwritten_real(
 		++*idx;
 		xfs_iext_insert(ip, *idx, 1, new, state);
 
-		ip->i_d.di_nextents++;
+		XFS_IFORK_NEXT_SET(ip, whichfork,
+				XFS_IFORK_NEXTENTS(ip, whichfork) + 1);
 		if (cur == NULL)
 			rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
 		else {
@@ -2641,7 +2634,8 @@ xfs_bmap_add_extent_unwritten_real(
 		++*idx;
 		xfs_iext_insert(ip, *idx, 2, &r[0], state);
 
-		ip->i_d.di_nextents += 2;
+		XFS_IFORK_NEXT_SET(ip, whichfork,
+				XFS_IFORK_NEXTENTS(ip, whichfork) + 2);
 		if (cur == NULL)
 			rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
 		else {
@@ -2695,17 +2689,17 @@ xfs_bmap_add_extent_unwritten_real(
 	}
 
 	/* update reverse mappings */
-	error = xfs_rmap_convert_extent(mp, dfops, ip, XFS_DATA_FORK, new);
+	error = xfs_rmap_convert_extent(mp, dfops, ip, whichfork, new);
 	if (error)
 		goto done;
 
 	/* convert to a btree if necessary */
-	if (xfs_bmap_needs_btree(ip, XFS_DATA_FORK)) {
+	if (xfs_bmap_needs_btree(ip, whichfork)) {
 		int	tmp_logflags;	/* partial log flag return val */
 
 		ASSERT(cur == NULL);
 		error = xfs_bmap_extents_to_btree(tp, ip, first, dfops, &cur,
-				0, &tmp_logflags, XFS_DATA_FORK);
+				0, &tmp_logflags, whichfork);
 		*logflagsp |= tmp_logflags;
 		if (error)
 			goto done;
@@ -2717,7 +2711,7 @@ xfs_bmap_add_extent_unwritten_real(
 		*curp = cur;
 	}
 
-	xfs_bmap_check_leaf_extents(*curp, ip, XFS_DATA_FORK);
+	xfs_bmap_check_leaf_extents(*curp, ip, whichfork);
 done:
 	*logflagsp |= rval;
 	return error;
@@ -2809,7 +2803,8 @@ xfs_bmap_add_extent_hole_delay(
 		oldlen = startblockval(left.br_startblock) +
 			startblockval(new->br_startblock) +
 			startblockval(right.br_startblock);
-		newlen = xfs_bmap_worst_indlen(ip, temp);
+		newlen = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
+					 oldlen);
 		xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx),
 			nullstartblock((int)newlen));
 		trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
@@ -2830,7 +2825,8 @@ xfs_bmap_add_extent_hole_delay(
 		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), temp);
 		oldlen = startblockval(left.br_startblock) +
 			startblockval(new->br_startblock);
-		newlen = xfs_bmap_worst_indlen(ip, temp);
+		newlen = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
+					 oldlen);
 		xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx),
 			nullstartblock((int)newlen));
 		trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
@@ -2846,7 +2842,8 @@ xfs_bmap_add_extent_hole_delay(
 		temp = new->br_blockcount + right.br_blockcount;
 		oldlen = startblockval(new->br_startblock) +
 			startblockval(right.br_startblock);
-		newlen = xfs_bmap_worst_indlen(ip, temp);
+		newlen = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
+					 oldlen);
 		xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, *idx),
 			new->br_startoff,
 			nullstartblock((int)newlen), temp, right.br_state);
@@ -2899,13 +2896,14 @@ xfs_bmap_add_extent_hole_real(
 	ASSERT(!isnullstartblock(new->br_startblock));
 	ASSERT(!bma->cur ||
 	       !(bma->cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
-	ASSERT(whichfork != XFS_COW_FORK);
 
 	XFS_STATS_INC(mp, xs_add_exlist);
 
 	state = 0;
 	if (whichfork == XFS_ATTR_FORK)
 		state |= BMAP_ATTRFORK;
+	if (whichfork == XFS_COW_FORK)
+		state |= BMAP_COWFORK;
 
 	/*
 	 * Check and set flags if this segment has a left neighbor.
@@ -3822,17 +3820,13 @@ xfs_bmap_btalloc(
 		 * the first block that was allocated.
 		 */
 		ASSERT(*ap->firstblock == NULLFSBLOCK ||
-		       XFS_FSB_TO_AGNO(mp, *ap->firstblock) ==
-		       XFS_FSB_TO_AGNO(mp, args.fsbno) ||
-		       (ap->dfops->dop_low &&
-			XFS_FSB_TO_AGNO(mp, *ap->firstblock) <
-			XFS_FSB_TO_AGNO(mp, args.fsbno)));
+		       XFS_FSB_TO_AGNO(mp, *ap->firstblock) <=
+		       XFS_FSB_TO_AGNO(mp, args.fsbno));
 
 		ap->blkno = args.fsbno;
 		if (*ap->firstblock == NULLFSBLOCK)
 			*ap->firstblock = args.fsbno;
-		ASSERT(nullfb || fb_agno == args.agno ||
-		       (ap->dfops->dop_low && fb_agno < args.agno));
+		ASSERT(nullfb || fb_agno <= args.agno);
 		ap->length = args.len;
 		if (!(ap->flags & XFS_BMAPI_COWFORK))
 			ap->ip->i_d.di_nblocks += args.len;
@@ -4368,10 +4362,16 @@ xfs_bmapi_allocate(
 	bma->got.br_state = XFS_EXT_NORM;
 
 	/*
-	 * A wasdelay extent has been initialized, so shouldn't be flagged
-	 * as unwritten.
+	 * In the data fork, a wasdelay extent has been initialized, so
+	 * shouldn't be flagged as unwritten.
+	 *
+	 * For the cow fork, however, we convert delalloc reservations
+	 * (extents allocated for speculative preallocation) to
+	 * allocated unwritten extents, and only convert the unwritten
+	 * extents to real extents when we're about to write the data.
 	 */
-	if (!bma->wasdel && (bma->flags & XFS_BMAPI_PREALLOC) &&
+	if ((!bma->wasdel || (bma->flags & XFS_BMAPI_COWFORK)) &&
+	    (bma->flags & XFS_BMAPI_PREALLOC) &&
 	    xfs_sb_version_hasextflgbit(&mp->m_sb))
 		bma->got.br_state = XFS_EXT_UNWRITTEN;
 
@@ -4422,8 +4422,6 @@ xfs_bmapi_convert_unwritten(
 			(XFS_BMAPI_PREALLOC | XFS_BMAPI_CONVERT))
 		return 0;
 
-	ASSERT(whichfork != XFS_COW_FORK);
-
 	/*
 	 * Modify (by adding) the state flag, if writing.
 	 */
@@ -4448,8 +4446,8 @@ xfs_bmapi_convert_unwritten(
 			return error;
 	}
 
-	error = xfs_bmap_add_extent_unwritten_real(bma->tp, bma->ip, &bma->idx,
-			&bma->cur, mval, bma->firstblock, bma->dfops,
+	error = xfs_bmap_add_extent_unwritten_real(bma->tp, bma->ip, whichfork,
+			&bma->idx, &bma->cur, mval, bma->firstblock, bma->dfops,
 			&tmp_logflags);
 	/*
 	 * Log the inode core unconditionally in the unwritten extent conversion
@@ -4458,8 +4456,12 @@ xfs_bmapi_convert_unwritten(
 	 * in the transaction for the sake of fsync(), even if nothing has
 	 * changed, because fsync() will not force the log for this transaction
 	 * unless it sees the inode pinned.
+	 *
+	 * Note: If we're only converting cow fork extents, there aren't
+	 * any on-disk updates to make, so we don't need to log anything.
 	 */
-	bma->logflags |= tmp_logflags | XFS_ILOG_CORE;
+	if (whichfork != XFS_COW_FORK)
+		bma->logflags |= tmp_logflags | XFS_ILOG_CORE;
 	if (error)
 		return error;
 
@@ -4533,15 +4535,15 @@ xfs_bmapi_write(
 	ASSERT(*nmap >= 1);
 	ASSERT(*nmap <= XFS_BMAP_MAX_NMAP);
 	ASSERT(!(flags & XFS_BMAPI_IGSTATE));
-	ASSERT(tp != NULL);
+	ASSERT(tp != NULL ||
+	       (flags & (XFS_BMAPI_CONVERT | XFS_BMAPI_COWFORK)) ==
+			(XFS_BMAPI_CONVERT | XFS_BMA