Merge branch 'master' into for-next

Sync with Linus' tree so that patches against newer codebase can be applied.

Signed-off-by: Jiri Kosina <jkosina@suse.cz>

98 files changed, 4226 insertions, 2705 deletions

diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index a096841bd06c..3542d94fddce 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -84,6 +84,7 @@ xfs-y				+= xfs_aops.o \
 				   xfs_message.o \
 				   xfs_mount.o \
 				   xfs_mru_cache.o \
+				   xfs_stats.o \
 				   xfs_super.o \
 				   xfs_symlink.o \
 				   xfs_sysfs.o \
@@ -118,7 +119,7 @@ xfs-$(CONFIG_XFS_QUOTA)		+= xfs_dquot.o \
 xfs-$(CONFIG_XFS_RT)		+= xfs_rtalloc.o
 
 xfs-$(CONFIG_XFS_POSIX_ACL)	+= xfs_acl.o
-xfs-$(CONFIG_PROC_FS)		+= xfs_stats.o
 xfs-$(CONFIG_SYSCTL)		+= xfs_sysctl.o
 xfs-$(CONFIG_COMPAT)		+= xfs_ioctl32.o
-xfs-$(CONFIG_NFSD_PNFS)		+= xfs_pnfs.o
+xfs-$(CONFIG_NFSD_BLOCKLAYOUT)	+= xfs_pnfs.o
+xfs-$(CONFIG_NFSD_SCSILAYOUT)	+= xfs_pnfs.o
diff --git a/fs/xfs/kmem.c b/fs/xfs/kmem.c
index a7a3a63bb360..686ba6fb20dd 100644
--- a/fs/xfs/kmem.c
+++ b/fs/xfs/kmem.c
@@ -55,8 +55,9 @@ kmem_alloc(size_t size, xfs_km_flags_t flags)
 			return ptr;
 		if (!(++retries % 100))
 			xfs_err(NULL,
-		"possible memory allocation deadlock in %s (mode:0x%x)",
-					__func__, lflags);
+	"%s(%u) possible memory allocation deadlock size %u in %s (mode:0x%x)",
+				current->comm, current->pid,
+				(unsigned int)size, __func__, lflags);
 		congestion_wait(BLK_RW_ASYNC, HZ/50);
 	} while (1);
 }
@@ -120,8 +121,9 @@ kmem_zone_alloc(kmem_zone_t *zone, xfs_km_flags_t flags)
 			return ptr;
 		if (!(++retries % 100))
 			xfs_err(NULL,
-		"possible memory allocation deadlock in %s (mode:0x%x)",
-					__func__, lflags);
+		"%s(%u) possible memory allocation deadlock in %s (mode:0x%x)",
+				current->comm, current->pid,
+				__func__, lflags);
 		congestion_wait(BLK_RW_ASYNC, HZ/50);
 	} while (1);
 }
diff --git a/fs/xfs/kmem.h b/fs/xfs/kmem.h
index cc6b768fc068..d1c66e465ca5 100644
--- a/fs/xfs/kmem.h
+++ b/fs/xfs/kmem.h
@@ -84,6 +84,7 @@ kmem_zalloc(size_t size, xfs_km_flags_t flags)
 #define KM_ZONE_HWALIGN	SLAB_HWCACHE_ALIGN
 #define KM_ZONE_RECLAIM	SLAB_RECLAIM_ACCOUNT
 #define KM_ZONE_SPREAD	SLAB_MEM_SPREAD
+#define KM_ZONE_ACCOUNT	SLAB_ACCOUNT
 
 #define kmem_zone	kmem_cache
 #define kmem_zone_t	struct kmem_cache
diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index ffad7f20342f..a708e38b494c 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -482,7 +482,9 @@ xfs_agfl_verify(
 		    be32_to_cpu(agfl->agfl_bno[i]) >= mp->m_sb.sb_agblocks)
 			return false;
 	}
-	return true;
+
+	return xfs_log_check_lsn(mp,
+				 be64_to_cpu(XFS_BUF_TO_AGFL(bp)->agfl_lsn));
 }
 
 static void
@@ -533,6 +535,7 @@ xfs_agfl_write_verify(
 }
 
 const struct xfs_buf_ops xfs_agfl_buf_ops = {
+	.name = "xfs_agfl",
 	.verify_read = xfs_agfl_read_verify,
 	.verify_write = xfs_agfl_write_verify,
 };
@@ -651,8 +654,8 @@ xfs_alloc_ag_vextent(
 				 -((long)(args->len)));
 	}
 
-	XFS_STATS_INC(xs_allocx);
-	XFS_STATS_ADD(xs_allocb, args->len);
+	XFS_STATS_INC(args->mp, xs_allocx);
+	XFS_STATS_ADD(args->mp, xs_allocb, args->len);
 	return error;
 }
 
@@ -1808,8 +1811,8 @@ xfs_free_ag_extent(
 
 	if (!isfl)
 		xfs_trans_mod_sb(tp, XFS_TRANS_SB_FDBLOCKS, (long)len);
-	XFS_STATS_INC(xs_freex);
-	XFS_STATS_ADD(xs_freeb, len);
+	XFS_STATS_INC(mp, xs_freex);
+	XFS_STATS_ADD(mp, xs_freeb, len);
 
 	trace_xfs_free_extent(mp, agno, bno, len, isfl, haveleft, haveright);
 
@@ -1924,7 +1927,7 @@ xfs_alloc_space_available(
  * Decide whether to use this allocation group for this allocation.
  * If so, fix up the btree freelist's size.
  */
-STATIC int			/* error */
+int			/* error */
 xfs_alloc_fix_freelist(
 	struct xfs_alloc_arg	*args,	/* allocation argument structure */
 	int			flags)	/* XFS_ALLOC_FLAG_... */
@@ -2259,9 +2262,13 @@ xfs_agf_verify(
  {
 	struct xfs_agf	*agf = XFS_BUF_TO_AGF(bp);
 
-	if (xfs_sb_version_hascrc(&mp->m_sb) &&
-	    !uuid_equal(&agf->agf_uuid, &mp->m_sb.sb_meta_uuid))
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		if (!uuid_equal(&agf->agf_uuid, &mp->m_sb.sb_meta_uuid))
+			return false;
+		if (!xfs_log_check_lsn(mp,
+				be64_to_cpu(XFS_BUF_TO_AGF(bp)->agf_lsn)))
 			return false;
+	}
 
 	if (!(agf->agf_magicnum == cpu_to_be32(XFS_AGF_MAGIC) &&
 	      XFS_AGF_GOOD_VERSION(be32_to_cpu(agf->agf_versionnum)) &&
@@ -2333,6 +2340,7 @@ xfs_agf_write_verify(
 }
 
 const struct xfs_buf_ops xfs_agf_buf_ops = {
+	.name = "xfs_agf",
 	.verify_read = xfs_agf_read_verify,
 	.verify_write = xfs_agf_write_verify,
 };
@@ -2503,7 +2511,7 @@ xfs_alloc_vextent(
 		 * Try near allocation first, then anywhere-in-ag after
 		 * the first a.g. fails.
 		 */
-		if ((args->userdata  == XFS_ALLOC_INITIAL_USER_DATA) &&
+		if ((args->userdata & XFS_ALLOC_INITIAL_USER_DATA) &&
 		    (mp->m_flags & XFS_MOUNT_32BITINODES)) {
 			args->fsbno = XFS_AGB_TO_FSB(mp,
 					((mp->m_agfrotor / rotorstep) %
@@ -2634,6 +2642,14 @@ xfs_alloc_vextent(
 		XFS_AG_CHECK_DADDR(mp, XFS_FSB_TO_DADDR(mp, args->fsbno),
 			args->len);
 #endif
+
+		/* Zero the extent if we were asked to do so */
+		if (args->userdata & XFS_ALLOC_USERDATA_ZERO) {
+			error = xfs_zero_extent(args->ip, args->fsbno, args->len);
+			if (error)
+				goto error0;
+		}
+
 	}
 	xfs_perag_put(args->pag);
 	return 0;
diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h
index ca1c8168373a..135eb3d24db7 100644
--- a/fs/xfs/libxfs/xfs_alloc.h
+++ b/fs/xfs/libxfs/xfs_alloc.h
@@ -101,6 +101,7 @@ typedef struct xfs_alloc_arg {
 	struct xfs_mount *mp;		/* file system mount point */
 	struct xfs_buf	*agbp;		/* buffer for a.g. freelist header */
 	struct xfs_perag *pag;		/* per-ag struct for this agno */
+	struct xfs_inode *ip;		/* for userdata zeroing method */
 	xfs_fsblock_t	fsbno;		/* file system block number */
 	xfs_agnumber_t	agno;		/* allocation group number */
 	xfs_agblock_t	agbno;		/* allocation group-relative block # */
@@ -120,15 +121,16 @@ typedef struct xfs_alloc_arg {
 	char		wasdel;		/* set if allocation was prev delayed */
 	char		wasfromfl;	/* set if allocation is from freelist */
 	char		isfl;		/* set if is freelist blocks - !acctg */
-	char		userdata;	/* set if this is user data */
+	char		userdata;	/* mask defining userdata treatment */
 	xfs_fsblock_t	firstblock;	/* io first block allocated */
 } xfs_alloc_arg_t;
 
 /*
  * Defines for userdata
  */
-#define XFS_ALLOC_USERDATA		1	/* allocation is for user data*/
-#define XFS_ALLOC_INITIAL_USER_DATA	2	/* special case start of file */
+#define XFS_ALLOC_USERDATA		(1 << 0)/* allocation is for user data*/
+#define XFS_ALLOC_INITIAL_USER_DATA	(1 << 1)/* special case start of file */
+#define XFS_ALLOC_USERDATA_ZERO		(1 << 2)/* zero extent on allocation */
 
 xfs_extlen_t xfs_alloc_longest_free_extent(struct xfs_mount *mp,
 		struct xfs_perag *pag, xfs_extlen_t need);
@@ -233,5 +235,6 @@ xfs_alloc_get_rec(
 
 int xfs_read_agf(struct xfs_mount *mp, struct xfs_trans *tp,
 			xfs_agnumber_t agno, int flags, struct xfs_buf **bpp);
+int xfs_alloc_fix_freelist(struct xfs_alloc_arg *args, int flags);
 
 #endif	/* __XFS_ALLOC_H__ */
diff --git a/fs/xfs/libxfs/xfs_alloc_btree.c b/fs/xfs/libxfs/xfs_alloc_btree.c
index 90de071dd4c2..d9b42425291e 100644
--- a/fs/xfs/libxfs/xfs_alloc_btree.c
+++ b/fs/xfs/libxfs/xfs_alloc_btree.c
@@ -118,8 +118,6 @@ xfs_allocbt_free_block(
 	xfs_extent_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1,
 			      XFS_EXTENT_BUSY_SKIP_DISCARD);
 	xfs_trans_agbtree_delta(cur->bc_tp, -1);
-
-	xfs_trans_binval(cur->bc_tp, bp);
 	return 0;
 }
 
@@ -293,14 +291,7 @@ xfs_allocbt_verify(
 	level = be16_to_cpu(block->bb_level);
 	switch (block->bb_magic) {
 	case cpu_to_be32(XFS_ABTB_CRC_MAGIC):
-		if (!xfs_sb_version_hascrc(&mp->m_sb))
-			return false;
-		if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid))
-			return false;
-		if (block->bb_u.s.bb_blkno != cpu_to_be64(bp->b_bn))
-			return false;
-		if (pag &&
-		    be32_to_cpu(block->bb_u.s.bb_owner) != pag->pag_agno)
+		if (!xfs_btree_sblock_v5hdr_verify(bp))
 			return false;
 		/* fall through */
 	case cpu_to_be32(XFS_ABTB_MAGIC):
@@ -311,14 +302,7 @@ xfs_allocbt_verify(
 			return false;
 		break;
 	case cpu_to_be32(XFS_ABTC_CRC_MAGIC):
-		if (!xfs_sb_version_hascrc(&mp->m_sb))
-			return false;
-		if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid))
-			return false;
-		if (block->bb_u.s.bb_blkno != cpu_to_be64(bp->b_bn))
-			return false;
-		if (pag &&
-		    be32_to_cpu(block->bb_u.s.bb_owner) != pag->pag_agno)
+		if (!xfs_btree_sblock_v5hdr_verify(bp))
 			return false;
 		/* fall through */
 	case cpu_to_be32(XFS_ABTC_MAGIC):
@@ -332,21 +316,7 @@ xfs_allocbt_verify(
 		return false;
 	}
 
-	/* numrecs verification */
-	if (be16_to_cpu(block->bb_numrecs) > mp->m_alloc_mxr[level != 0])
-		return false;
-
-	/* sibling pointer verification */
-	if (!block->bb_u.s.bb_leftsib ||
-	    (be32_to_cpu(block->bb_u.s.bb_leftsib) >= mp->m_sb.sb_agblocks &&
-	     block->bb_u.s.bb_leftsib != cpu_to_be32(NULLAGBLOCK)))
-		return false;
-	if (!block->bb_u.s.bb_rightsib ||
-	    (be32_to_cpu(block->bb_u.s.bb_rightsib) >= mp->m_sb.sb_agblocks &&
-	     block->bb_u.s.bb_rightsib != cpu_to_be32(NULLAGBLOCK)))
-		return false;
-
-	return true;
+	return xfs_btree_sblock_verify(bp, mp->m_alloc_mxr[level != 0]);
 }
 
 static void
@@ -379,6 +349,7 @@ xfs_allocbt_write_verify(
 }
 
 const struct xfs_buf_ops xfs_allocbt_buf_ops = {
+	.name = "xfs_allocbt",
 	.verify_read = xfs_allocbt_read_verify,
 	.verify_write = xfs_allocbt_write_verify,
 };
diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c
index ff065578969f..fa3b948ef9c2 100644
--- a/fs/xfs/libxfs/xfs_attr.c
+++ b/fs/xfs/libxfs/xfs_attr.c
@@ -125,7 +125,7 @@ xfs_attr_get(
 	uint			lock_mode;
 	int			error;
 
-	XFS_STATS_INC(xs_attr_get);
+	XFS_STATS_INC(ip->i_mount, xs_attr_get);
 
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
 		return -EIO;
@@ -207,9 +207,9 @@ xfs_attr_set(
 	struct xfs_trans_res	tres;
 	xfs_fsblock_t		firstblock;
 	int			rsvd = (flags & ATTR_ROOT) != 0;
-	int			error, err2, committed, local;
+	int			error, err2, local;
 
-	XFS_STATS_INC(xs_attr_set);
+	XFS_STATS_INC(mp, xs_attr_set);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
 		return -EIO;
@@ -334,25 +334,15 @@ xfs_attr_set(
 		 */
 		xfs_bmap_init(args.flist, args.firstblock);
 		error = xfs_attr_shortform_to_leaf(&args);
-		if (!error) {
-			error = xfs_bmap_finish(&args.trans, args.flist,
-						&committed);
-		}
+		if (!error)
+			error = xfs_bmap_finish(&args.trans, args.flist, dp);
 		if (error) {
-			ASSERT(committed);
 			args.trans = NULL;
 			xfs_bmap_cancel(&flist);
 			goto out;
 		}
 
 		/*
-		 * bmap_finish() may have committed the last trans and started
-		 * a new one.  We need the inode to be in all transactions.
-		 */
-		if (committed)
-			xfs_trans_ijoin(args.trans, dp, 0);
-
-		/*
 		 * Commit the leaf transformation.  We'll need another (linked)
 		 * transaction to add the new attribute to the leaf.
 		 */
@@ -412,7 +402,7 @@ xfs_attr_remove(
 	xfs_fsblock_t		firstblock;
 	int			error;
 
-	XFS_STATS_INC(xs_attr_remove);
+	XFS_STATS_INC(mp, xs_attr_remove);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
 		return -EIO;
@@ -568,7 +558,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 {
 	xfs_inode_t *dp;
 	struct xfs_buf *bp;
-	int retval, error, committed, forkoff;
+	int retval, error, forkoff;
 
 	trace_xfs_attr_leaf_addname(args);
 
@@ -628,25 +618,15 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 		 */
 		xfs_bmap_init(args->flist, args->firstblock);
 		error = xfs_attr3_leaf_to_node(args);
-		if (!error) {
-			error = xfs_bmap_finish(&args->trans, args->flist,
-						&committed);
-		}
+		if (!error)
+			error = xfs_bmap_finish(&args->trans, args->flist, dp);
 		if (error) {
-			ASSERT(committed);
 			args->trans = NULL;
 			xfs_bmap_cancel(args->flist);
 			return error;
 		}
 
 		/*
-		 * bmap_finish() may have committed the last trans and started
-		 * a new one.  We need the inode to be in all transactions.
-		 */
-		if (committed)
-			xfs_trans_ijoin(args->trans, dp, 0);
-
-		/*
 		 * Commit the current trans (including the inode) and start
 		 * a new one.
 		 */
@@ -729,25 +709,14 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 			xfs_bmap_init(args->flist, args->firstblock);
 			error = xfs_attr3_leaf_to_shortform(bp, args, forkoff);
 			/* bp is gone due to xfs_da_shrink_inode */
-			if (!error) {
+			if (!error)
 				error = xfs_bmap_finish(&args->trans,
-							args->flist,
-							&committed);
-			}
+							args->flist, dp);
 			if (error) {
-				ASSERT(committed);
 				args->trans = NULL;
 				xfs_bmap_cancel(args->flist);
 				return error;
 			}
-
-			/*
-			 * bmap_finish() may have committed the last trans
-			 * and started a new one.  We need the inode to be
-			 * in all transactions.
-			 */
-			if (committed)
-				xfs_trans_ijoin(args->trans, dp, 0);
 		}
 
 		/*
@@ -775,7 +744,7 @@ xfs_attr_leaf_removename(xfs_da_args_t *args)
 {
 	xfs_inode_t *dp;
 	struct xfs_buf *bp;
-	int error, committed, forkoff;
+	int error, forkoff;
 
 	trace_xfs_attr_leaf_removename(args);
 
@@ -803,23 +772,13 @@ xfs_attr_leaf_removename(xfs_da_args_t *args)
 		xfs_bmap_init(args->flist, args->firstblock);
 		error = xfs_attr3_leaf_to_shortform(bp, args, forkoff);
 		/* bp is gone due to xfs_da_shrink_inode */
-		if (!error) {
-			error = xfs_bmap_finish(&args->trans, args->flist,
-						&committed);
-		}
+		if (!error)
+			error = xfs_bmap_finish(&args->trans, args->flist, dp);
 		if (error) {
-			ASSERT(committed);
 			args->trans = NULL;
 			xfs_bmap_cancel(args->flist);
 			return error;
 		}
-
-		/*
-		 * bmap_finish() may have committed the last trans and started
-		 * a new one.  We need the inode to be in all transactions.
-		 */
-		if (committed)
-			xfs_trans_ijoin(args->trans, dp, 0);
 	}
 	return 0;
 }
@@ -877,7 +836,7 @@ xfs_attr_node_addname(xfs_da_args_t *args)
 	xfs_da_state_blk_t *blk;
 	xfs_inode_t *dp;
 	xfs_mount_t *mp;
-	int committed, retval, error;
+	int retval, error;
 
 	trace_xfs_attr_node_addname(args);
 
@@ -938,27 +897,16 @@ restart:
 			state = NULL;
 			xfs_bmap_init(args->flist, args->firstblock);
 			error = xfs_attr3_leaf_to_node(args);
-			if (!error) {
+			if (!error)
 				error = xfs_bmap_finish(&args->trans,
-							args->flist,
-							&committed);
-			}
+							args->flist, dp);
 			if (error) {
-				ASSERT(committed);
 				args->trans = NULL;
 				xfs_bmap_cancel(args->flist);
 				goto out;
 			}
 
 			/*
-			 * bmap_finish() may have committed the last trans
-			 * and started a new one.  We need the inode to be
-			 * in all transactions.
-			 */
-			if (committed)
-				xfs_trans_ijoin(args->trans, dp, 0);
-
-			/*
 			 * Commit the node conversion and start the next
 			 * trans in the chain.
 			 */
@@ -977,23 +925,13 @@ restart:
 		 */
 		xfs_bmap_init(args->flist, args->firstblock);
 		error = xfs_da3_split(state);
-		if (!error) {
-			error = xfs_bmap_finish(&args->trans, args->flist,
-						&committed);
-		}
+		if (!error)
+			error = xfs_bmap_finish(&args->trans, args->flist, dp);
 		if (error) {
-			ASSERT(committed);
 			args->trans = NULL;
 			xfs_bmap_cancel(args->flist);
 			goto out;
 		}
-
-		/*
-		 * bmap_finish() may have committed the last trans and started
-		 * a new one.  We need the inode to be in all transactions.
-		 */
-		if (committed)
-			xfs_trans_ijoin(args->trans, dp, 0);
 	} else {
 		/*
 		 * Addition succeeded, update Btree hashvals.
@@ -1086,25 +1024,14 @@ restart:
 		if (retval && (state->path.active > 1)) {
 			xfs_bmap_init(args->flist, args->firstblock);
 			error = xfs_da3_join(state);
-			if (!error) {
+			if (!error)
 				error = xfs_bmap_finish(&args->trans,
-							args->flist,
-							&committed);
-			}
+							args->flist, dp);
 			if (error) {
-				ASSERT(committed);
 				args->trans = NULL;
 				xfs_bmap_cancel(args->flist);
 				goto out;
 			}
-
-			/*
-			 * bmap_finish() may have committed the last trans
-			 * and started a new one.  We need the inode to be
-			 * in all transactions.
-			 */
-			if (committed)
-				xfs_trans_ijoin(args->trans, dp, 0);
 		}
 
 		/*
@@ -1146,7 +1073,7 @@ xfs_attr_node_removename(xfs_da_args_t *args)
 	xfs_da_state_blk_t *blk;
 	xfs_inode_t *dp;
 	struct xfs_buf *bp;
-	int retval, error, committed, forkoff;
+	int retval, error, forkoff;
 
 	trace_xfs_attr_node_removename(args);
 
@@ -1220,24 +1147,13 @@ xfs_attr_node_removename(xfs_da_args_t *args)
 	if (retval && (state->path.active > 1)) {
 		xfs_bmap_init(args->flist, args->firstblock);
 		error = xfs_da3_join(state);
-		if (!error) {
-			error = xfs_bmap_finish(&args->trans, args->flist,
-						&committed);
-		}
+		if (!error)
+			error = xfs_bmap_finish(&args->trans, args->flist, dp);
 		if (error) {
-			ASSERT(committed);
 			args->trans = NULL;
 			xfs_bmap_cancel(args->flist);
 			goto out;
 		}
-
-		/*
-		 * bmap_finish() may have committed the last trans and started
-		 * a new one.  We need the inode to be in all transactions.
-		 */
-		if (committed)
-			xfs_trans_ijoin(args->trans, dp, 0);
-
 		/*
 		 * Commit the Btree join operation and start a new trans.
 		 */
@@ -1265,25 +1181,14 @@ xfs_attr_node_removename(xfs_da_args_t *args)
 			xfs_bmap_init(args->flist, args->firstblock);
 			error = xfs_attr3_leaf_to_shortform(bp, args, forkoff);
 			/* bp is gone due to xfs_da_shrink_inode */
-			if (!error) {
+			if (!error)
 				error = xfs_bmap_finish(&args->trans,
-							args->flist,
-							&committed);
-			}
+							args->flist, dp);
 			if (error) {
-				ASSERT(committed);
 				args->trans = NULL;
 				xfs_bmap_cancel(args->flist);
 				goto out;
 			}
-
-			/*
-			 * bmap_finish() may have committed the last trans
-			 * and started a new one.  We need the inode to be
-			 * in all transactions.
-			 */
-			if (committed)
-				xfs_trans_ijoin(args->trans, dp, 0);
 		} else
 			xfs_trans_brelse(args->trans, bp);
 	}
diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c
index 33df52d97ec7..01a5ecfedfcf 100644
--- a/fs/xfs/libxfs/xfs_attr_leaf.c
+++ b/fs/xfs/libxfs/xfs_attr_leaf.c
@@ -41,6 +41,7 @@
 #include "xfs_buf_item.h"
 #include "xfs_cksum.h"
 #include "xfs_dir2.h"
+#include "xfs_log.h"
 
 
 /*
@@ -266,6 +267,8 @@ xfs_attr3_leaf_verify(
 			return false;
 		if (be64_to_cpu(hdr3->info.blkno) != bp->b_bn)
 			return false;
+		if (!xfs_log_check_lsn(mp, be64_to_cpu(hdr3->info.lsn)))
+			return false;
 	} else {
 		if (ichdr.magic != XFS_ATTR_LEAF_MAGIC)
 			return false;
@@ -325,6 +328,7 @@ xfs_attr3_leaf_read_verify(
 }
 
 const struct xfs_buf_ops xfs_attr3_leaf_buf_ops = {
+	.name = "xfs_attr3_leaf",
 	.verify_read = xfs_attr3_leaf_read_verify,
 	.verify_write = xfs_attr3_leaf_write_verify,
 };
diff --git a/fs/xfs/libxfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c
index f38f9bd81557..a572532a55cd 100644
--- a/fs/xfs/libxfs/xfs_attr_remote.c
+++ b/fs/xfs/libxfs/xfs_attr_remote.c
@@ -107,7 +107,7 @@ xfs_attr3_rmt_verify(
 	if (be32_to_cpu(rmt->rm_bytes) > fsbsize - sizeof(*rmt))
 		return false;
 	if (be32_to_cpu(rmt->rm_offset) +
-				be32_to_cpu(rmt->rm_bytes) > XATTR_SIZE_MAX)
+				be32_to_cpu(rmt->rm_bytes) > XFS_XATTR_SIZE_MAX)
 		return false;
 	if (rmt->rm_owner == 0)
 		return false;
@@ -201,6 +201,7 @@ xfs_attr3_rmt_write_verify(
 }
 
 const struct xfs_buf_ops xfs_attr3_rmt_buf_ops = {
+	.name = "xfs_attr3_rmt",
 	.verify_read = xfs_attr3_rmt_read_verify,
 	.verify_write = xfs_attr3_rmt_write_verify,
 };
@@ -447,8 +448,6 @@ xfs_attr_rmtval_set(
 	 * Roll through the "value", allocating blocks on disk as required.
 	 */
 	while (blkcnt > 0) {
-		int	committed;
-
 		/*
 		 * Allocate a single extent, up to the size of the value.
 		 *
@@ -466,24 +465,14 @@ xfs_attr_rmtval_set(
 		error = xfs_bmapi_write(args->trans, dp, (xfs_fileoff_t)lblkno,
 				  blkcnt, XFS_BMAPI_ATTRFORK, args->firstblock,
 				  args->total, &map, &nmap, args->flist);
-		if (!error) {
-			error = xfs_bmap_finish(&args->trans, args->flist,
-						&committed);
-		}
+		if (!error)
+			error = xfs_bmap_finish(&args->trans, args->flist, dp);
 		if (error) {
-			ASSERT(committed);
 			args->trans = NULL;
 			xfs_bmap_cancel(args->flist);
 			return error;
 		}
 
-		/*
-		 * bmap_finish() may have committed the last trans and started
-		 * a new one.  We need the inode to be in all transactions.
-		 */
-		if (committed)
-			xfs_trans_ijoin(args->trans, dp, 0);
-
 		ASSERT(nmap == 1);
 		ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
 		       (map.br_startblock != HOLESTARTBLOCK));
@@ -614,31 +603,20 @@ xfs_attr_rmtval_remove(
 	blkcnt = args->rmtblkcnt;
 	done = 0;
 	while (!done) {
-		int committed;
-
 		xfs_bmap_init(args->flist, args->firstblock);
 		error = xfs_bunmapi(args->trans, args->dp, lblkno, blkcnt,
 				    XFS_BMAPI_ATTRFORK, 1, args->firstblock,
 				    args->flist, &done);
-		if (!error) {
+		if (!error)
 			error = xfs_bmap_finish(&args->trans, args->flist,
-						&committed);
-		}
+						args->dp);
 		if (error) {
-			ASSERT(committed);
 			args->trans = NULL;
 			xfs_bmap_cancel(args->flist);
 			return error;
 		}
 
 		/*
-		 * bmap_finish() may have committed the last trans and started
-		 * a new one.  We need the inode to be in all transactions.
-		 */
-		if (committed)
-			xfs_trans_ijoin(args->trans, args->dp, 0);
-
-		/*
 		 * Close out trans and start the next one in the chain.
 		 */
 		error = xfs_trans_roll(&args->trans, args->dp);
diff --git a/fs/xfs/libxfs/xfs_attr_sf.h b/fs/xfs/libxfs/xfs_attr_sf.h
index 919756e3ba53..90928bbe693c 100644
--- a/fs/xfs/libxfs/xfs_attr_sf.h
+++ b/fs/xfs/libxfs/xfs_attr_sf.h
@@ -24,22 +24,6 @@
  * Small attribute lists are packed as tightly as possible so as
  * to fit into the literal area of the inode.
  */
-
-/*
- * Entries are packed toward the top as tight as possible.
- */
-typedef struct xfs_attr_shortform {
-	struct xfs_attr_sf_hdr {	/* constant-structure header block */
-		__be16	totsize;	/* total bytes in shortform list */
-		__u8	count;	/* count of active entries */
-	} hdr;
-	struct xfs_attr_sf_entry {
-		__uint8_t namelen;	/* actual length of name (no NULL) */
-		__uint8_t valuelen;	/* actual length of value (no NULL) */
-		__uint8_t flags;	/* flags bits (see xfs_attr_leaf.h) */
-		__uint8_t nameval[1];	/* name & value bytes concatenated */
-	} list[1];			/* variable sized array */
-} xfs_attr_shortform_t;
 typedef struct xfs_attr_sf_hdr xfs_attr_sf_hdr_t;
 typedef struct xfs_attr_sf_entry xfs_attr_sf_entry_t;
 
diff --git a/fs/xfs/libxfs/xfs_bit.c b/fs/xfs/libxfs/xfs_bit.c
index 0e8885a59646..0a94cce5ea35 100644
--- a/fs/xfs/libxfs/xfs_bit.c
+++ b/fs/xfs/libxfs/xfs_bit.c
@@ -32,13 +32,13 @@ int
 xfs_bitmap_empty(uint *map, uint size)
 {
 	uint i;
-	uint ret = 0;
 
 	for (i = 0; i < size; i++) {
-		ret |= map[i];
+		if (map[i] != 0)
+			return 0;
 	}
 
-	return (ret == 0);
+	return 1;
 }
 
 /*
diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 8e2010d53b07..ce41d7fe753c 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -325,9 +325,11 @@ xfs_check_block(
 
 /*
  * Check that the extents for the inode ip are in the right order in all
- * btree leaves.
+ * btree leaves. THis becomes prohibitively expensive for large extent count
+ * files, so don't bother with inodes that have more than 10,000 extents in
+ * them. The btree record ordering checks will still be done, so for such large
+ * bmapbt constructs that is going to catch most corruptions.
  */
-
 STATIC void
 xfs_bmap_check_leaf_extents(
 	xfs_btree_cur_t		*cur,	/* btree cursor or null */
@@ -352,6 +354,10 @@ xfs_bmap_check_leaf_extents(
 		return;
 	}
 
+	/* skip large extent count inodes */
+	if (ip->i_d.di_nextents > 10000)
+		return;
+
 	bno = NULLFSBLOCK;
 	mp = ip->i_mount;
 	ifp = XFS_IFORK_PTR(ip, whichfork);
@@ -471,10 +477,7 @@ xfs_bmap_check_leaf_extents(
 		}
 		block = XFS_BUF_TO_BLOCK(bp);
 	}
-	if (bp_release) {
-		bp_release = 0;
-		xfs_trans_brelse(NULL, bp);
-	}
+
 	return;
 
 error0:
@@ -906,7 +909,7 @@ xfs_bmap_local_to_extents(
 	 * We don't want to deal with the case of keeping inode data inline yet.
 	 * So sending the data fork of a regular inode is invalid.
 	 */
-	ASSERT(!(S_ISREG(ip->i_d.di_mode) && whichfork == XFS_DATA_FORK));
+	ASSERT(!(S_ISREG(VFS_I(ip)->i_mode) && whichfork == XFS_DATA_FORK));
 	ifp = XFS_IFORK_PTR(ip, whichfork);
 	ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL);
 
@@ -948,14 +951,16 @@ xfs_bmap_local_to_extents(
 	bp = xfs_btree_get_bufl(args.mp, tp, args.fsbno, 0);
 
 	/*
-	 * Initialise the block and copy the data
+	 * Initialize the block, copy the data and log the remote buffer.
 	 *
-	 * Note: init_fn must set the buffer log item type correctly!
+	 * The callout is responsible for logging because the remote format
+	 * might differ from the local format and thus we don't know how much to
+	 * log here. Note that init_fn must also set the buffer log item type
+	 * correctly.
 	 */
 	init_fn(tp, bp, ip, ifp);
 
-	/* account for the change in fork size and log everything */
-	xfs_trans_log_buf(tp, bp, 0, ifp->if_bytes - 1);
+	/* account for the change in fork size */
 	xfs_idata_realloc(ip, -ifp->if_bytes, whichfork);
 	xfs_bmap_local_to_extents_empty(ip, whichfork);
 	flags |= XFS_ILOG_CORE;
@@ -1071,7 +1076,7 @@ xfs_bmap_add_attrfork_local(
 	if (ip->i_df.if_bytes <= XFS_IFORK_DSIZE(ip))
 		return 0;
 
-	if (S_ISDIR(ip->i_d.di_mode)) {
+	if (S_ISDIR(VFS_I(ip)->i_mode)) {
 		memset(&dargs, 0, sizeof(dargs));
 		dargs.geo = ip->i_mount->m_dir_geo;
 		dargs.dp = ip;
@@ -1083,7 +1088,7 @@ xfs_bmap_add_attrfork_local(
 		return xfs_dir2_sf_to_block(&dargs);
 	}
 
-	if (S_ISLNK(ip->i_d.di_mode))
+	if (S_ISLNK(VFS_I(ip)->i_mode))
 		return xfs_bmap_local_to_extents(tp, ip, firstblock, 1,
 						 flags, XFS_DATA_FORK,
 						 xfs_symlink_local_to_remote);
@@ -1109,7 +1114,6 @@ xfs_bmap_add_attrfork(
 	xfs_trans_t		*tp;		/* transaction pointer */
 	int			blks;		/* space reservation */
 	int			version = 1;	/* superblock attr version */
-	int			committed;	/* xaction was committed */
 	int			logflags;	/* logging flags */
 	int			error;		/* error return value */
 
@@ -1212,7 +1216,7 @@ xfs_bmap_add_attrfork(
 			xfs_log_sb(tp);
 	}
 
-	error = xfs_bmap_finish(&tp, &flist, &committed);
+	error = xfs_bmap_finish(&tp, &flist, NULL);
 	if (error)
 		goto bmap_cancel;
 	error = xfs_trans_commit(tp);
@@ -1435,7 +1439,7 @@ xfs_bmap_search_extents(
 	xfs_ifork_t	*ifp;		/* inode fork pointer */
 	xfs_bmbt_rec_host_t  *ep;            /* extent record pointer */
 
-	XFS_STATS_INC(xs_look_exlist);
+	XFS_STATS_INC(ip->i_mount, xs_look_exlist);
 	ifp = XFS_IFORK_PTR(ip, fork);
 
 	ep = xfs_bmap_search_multi_extents(ifp, bno, eofp, lastxp, gotp, prevp);
@@ -1721,10 +1725,11 @@ xfs_bmap_add_extent_delay_real(
 	xfs_filblks_t		temp=0;	/* value for da_new calculations */
 	xfs_filblks_t		temp2=0;/* value for da_new calculations */
 	int			tmp_rval;	/* partial logging flags */
+	int			whichfork = XFS_DATA_FORK;
 	struct xfs_mount	*mp;
 
-	mp  = bma->tp ? bma->tp->t_mountp : NULL;
-	ifp = XFS_IFORK_PTR(bma->ip, XFS_DATA_FORK);
+	mp = bma->ip->i_mount;
+	ifp = XFS_IFORK_PTR(bma->ip, whichfork);
 
 	ASSERT(bma->idx >= 0);
 	ASSERT(bma->idx <= ifp->if_bytes / sizeof(struct xfs_bmbt_rec));
@@ -1732,7 +1737,7 @@ xfs_bmap_add_extent_delay_real(
 	ASSERT(!bma->cur ||
 	       (bma->cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
 
-	XFS_STATS_INC(xs_add_exlist);
+	XFS_STATS_INC(mp, xs_add_exlist);
 
 #define	LEFT		r[0]
 #define	RIGHT		r[1]
@@ -1783,7 +1788,7 @@ xfs_bmap_add_extent_delay_real(
 	 * Don't set contiguous if the combined extent would be too large.
 	 * Also check for all-three-contiguous being too large.
 	 */
-	if (bma->idx < bma->ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) {
+	if (bma->idx < ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) {
 		state |= BMAP_RIGHT_VALID;
 		xfs_bmbt_get_all(xfs_iext_get_ext(ifp, bma->idx + 1), &RIGHT);
 
@@ -2014,10 +2019,10 @@ xfs_bmap_add_extent_delay_real(
 			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
 		}
 
-		if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
+		if (xfs_bmap_needs_btree(bma->ip, whichfork)) {
 			error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
 					bma->firstblock, bma->flist,
-					&bma->cur, 1, &tmp_rval, XFS_DATA_FORK);
+					&bma->cur, 1, &tmp_rval, whichfork);
 			rval |= tmp_rval;
 			if (error)
 				goto done;
@@ -2098,10 +2103,10 @@ xfs_bmap_add_extent_delay_real(
 			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
 		}
 
-		if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
+		if (xfs_bmap_needs_btree(bma->ip, whichfork)) {
 			error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
 				bma->firstblock, bma->flist, &bma->cur, 1,
-				&tmp_rval, XFS_DATA_FORK);
+				&tmp_rval, whichfork);
 			rval |= tmp_rval;
 			if (error)
 				goto done;
@@ -2167,10 +2172,10 @@ xfs_bmap_add_extent_delay_real(
 			XFS_WANT_CORRUPTED_GOTO(mp, i == 1, done);
 		}
 
-		if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
+		if (xfs_bmap_needs_btree(bma->ip, whichfork)) {
 			error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
 					bma->firstblock, bma->flist, &bma->cur,
-					1, &tmp_rval, XFS_DATA_FORK);
+					1, &tmp_rval, whichfork);
 			rval |= tmp_rval;
 			if (error)
 				goto done;
@@ -2213,13 +2218,13 @@ xfs_bmap_add_extent_delay_real(
 	}
 
 	/* convert to a btree if necessary */
-	if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
+	if (xfs_bmap_needs_btree(bma->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->flist, &bma->cur,
-				da_old > 0, &tmp_logflags, XFS_DATA_FORK);
+				da_old > 0, &tmp_logflags, whichfork);
 		bma->logflags |= tmp_logflags;
 		if (error)
 			goto done;
@@ -2240,7 +2245,7 @@ xfs_bmap_add_extent_delay_real(
 	if (bma->cur)
 		bma->cur->bc_private.b.allocated = 0;
 
-	xfs_bmap_check_leaf_extents(bma->cur, bma->ip, XFS_DATA_FORK);
+	xfs_bmap_check_leaf_extents(bma->cur, bma->ip, whichfork);
 done:
 	bma->logflags |= rval;
 	return error;
@@ -2286,7 +2291,7 @@ xfs_bmap_add_extent_unwritten_real(
 	ASSERT(*idx <= ifp->if_bytes / sizeof(struct xfs_bmbt_rec));
 	ASSERT(!isnullstartblock(new->br_startblock));
 
-	XFS_STATS_INC(xs_add_exlist);
+	XFS_STATS_INC(mp, xs_add_exlist);
 
 #define	LEFT		r[0]
 #define	RIGHT		r[1]
@@ -2937,7 +2942,7 @@ xfs_bmap_add_extent_hole_real(
 	int			state;	/* state bits, accessed thru macros */
 	struct xfs_mount	*mp;
 
-	mp = bma->tp ? bma->tp->t_mountp : NULL;
+	mp = bma->ip->i_mount;
 	ifp = XFS_IFORK_PTR(bma->ip, whichfork);
 
 	ASSERT(bma->idx >= 0);
@@ -2946,7 +2951,7 @@ xfs_bmap_add_extent_hole_real(
 	ASSERT(!bma->cur ||
 	       !(bma->cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL));
 
-	XFS_STATS_INC(xs_add_exlist);
+	XFS_STATS_INC(mp, xs_add_exlist);
 
 	state = 0;
 	if (whichfork == XFS_ATTR_FORK)
@@ -3737,11 +3742,11 @@ xfs_bmap_btalloc(
 		args.prod = align;
 		if ((args.mod = (xfs_extlen_t)do_mod(ap->offset, args.prod)))
 			args.mod = (xfs_extlen_t)(args.prod - args.mod);
-	} else if (mp->m_sb.sb_blocksize >= PAGE_CACHE_SIZE) {
+	} else if (mp->m_sb.sb_blocksize >= PAGE_SIZE) {
 		args.prod = 1;
 		args.mod = 0;
 	} else {
-		args.prod = PAGE_CACHE_SIZE >> mp->m_sb.sb_blocklog;
+		args.prod = PAGE_SIZE >> mp->m_sb.sb_blocklog;
 		if ((args.mod = (xfs_extlen_t)(do_mod(ap->offset, args.prod))))
 			args.mod = (xfs_extlen_t)(args.prod - args.mod);
 	}
@@ -3800,8 +3805,13 @@ xfs_bmap_btalloc(
 	args.wasdel = ap->wasdel;
 	args.isfl = 0;
 	args.userdata = ap->userdata;
-	if ((error = xfs_alloc_vextent(&args)))
+	if (ap->userdata & XFS_ALLOC_USERDATA_ZERO)
+		args.ip = ap->ip;
+
+	error = xfs_alloc_vextent(&args);
+	if (error)
 		return error;
+
 	if (tryagain && args.fsbno == NULLFSBLOCK) {
 		/*
 		 * Exact allocation failed. Now try with alignment
@@ -4036,7 +4046,7 @@ xfs_bmapi_read(
 	if (XFS_FORCED_SHUTDOWN(mp))
 		return -EIO;
 
-	XFS_STATS_INC(xs_blk_mapr);
+	XFS_STATS_INC(mp, xs_blk_mapr);
 
 	ifp = XFS_IFORK_PTR(ip, whichfork);
 
@@ -4221,7 +4231,7 @@ xfs_bmapi_delay(
 	if (XFS_FORCED_SHUTDOWN(mp))
 		return -EIO;
 
-	XFS_STATS_INC(xs_blk_mapw);
+	XFS_STATS_INC(mp, xs_blk_mapw);
 
 	if (!(ifp->if_flags & XFS_IFEXTENTS)) {
 		error = xfs_iread_extents(NULL, ip, XFS_DATA_FORK);
@@ -4300,11 +4310,14 @@ xfs_bmapi_allocate(
 
 	/*
 	 * Indicate if this is the first user data in the file, or just any
-	 * user data.
+	 * user data. And if it is userdata, indicate whether it needs to
+	 * be initialised to zero during allocation.
 	 */
 	if (!(bma->flags & XFS_BMAPI_METADATA)) {
 		bma->userdata = (bma->offset == 0) ?
 			XFS_ALLOC_INITIAL_USER_DATA : XFS_ALLOC_USERDATA;
+		if (bma->flags & XFS_BMAPI_ZERO)
+			bma->userdata |= XFS_ALLOC_USERDATA_ZERO;
 	}
 
 	bma->minlen = (bma->flags & XFS_BMAPI_CONTIG) ? bma->length : 1;
@@ -4419,6 +4432,17 @@ xfs_bmapi_convert_unwritten(
 	mval->br_state = (mval->br_state == XFS_EXT_UNWRITTEN)
 				? XFS_EXT_NORM : XFS_EXT_UNWRITTEN;
 
+	/*
+	 * Before insertion into the bmbt, zero the range being converted
+	 * if required.
+	 */
+	if (flags & XFS_BMAPI_ZERO) {
+		error = xfs_zero_extent(bma->ip, mval->br_startblock,
+					mval->br_blockcount);
+		if (error)
+			return error;
+	}
+
 	error = xfs_bmap_add_extent_unwritten_real(bma->tp, bma->ip, &bma->idx,
 			&bma->cur, mval, bma->firstblock, bma->flist,
 			&tmp_logflags);
@@ -4512,6 +4536,18 @@ xfs_bmapi_write(
 	ASSERT(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_LOCAL);
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
+	/* zeroing is for currently only for data extents, not metadata */
+	ASSERT((flags & (XFS_BMAPI_METADATA | XFS_BMAPI_ZERO)) !=
+			(XFS_BMAPI_METADATA | XFS_BMAPI_ZERO));
+	/*
+	 * we can allocate unwritten extents or pre-zero allocated blocks,
+	 * but it makes no sense to do both at once. This would result in
+	 * zeroing the unwritten extent twice, but it still being an
+	 * unwritten extent....
+	 */
+	ASSERT((flags & (XFS_BMAPI_PREALLOC | XFS_BMAPI_ZERO)) !=
+			(XFS_BMAPI_PREALLOC | XFS_BMAPI_ZERO));
+
 	if (unlikely(XFS_TEST_ERROR(
 	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
 	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
@@ -4525,7 +4561,7 @@ xfs_bmapi_write(
 
 	ifp = XFS_IFORK_PTR(ip, whichfork);
 
-	XFS_STATS_INC(xs_blk_mapw);
+	XFS_STATS_INC(mp, xs_blk_mapw);
 
 	if (*firstblock == NULLFSBLOCK) {
 		if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE)
@@ -4682,6 +4718,66 @@ error0:
 }
 
 /*
+ * 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
+ * behind.
+ *
+ * Given the original reservation and the worst case indlen for the two new
+ * extents (as calculated by xfs_bmap_worst_indlen()), split the original
+ * reservation fairly across the two new extents. If necessary, steal available
+ * blocks from a deleted extent to make up a reservation deficiency (e.g., if
+ * ores == 1). The number of stolen blocks is returned. The availability and
+ * subsequent accounting of stolen blocks is the responsibility of the caller.
+ */
+static xfs_filblks_t
+xfs_bmap_split_indlen(
+	xfs_filblks_t			ores,		/* original res. */
+	xfs_filblks_t			*indlen1,	/* ext1 worst indlen */
+	xfs_filblks_t			*indlen2,	/* ext2 worst indlen */
+	xfs_filblks_t			avail)		/* stealable blocks */
+{
+	xfs_filblks_t			len1 = *indlen1;
+	xfs_filblks_t			len2 = *indlen2;
+	xfs_filblks_t			nres = len1 + len2; /* new total res. */
+	xfs_filblks_t			stolen = 0;
+
+	/*
+	 * Steal as many blocks as we can to try and satisfy the worst case
+	 * indlen for both new extents.
+	 */
+	while (nres > ores && avail) {
+		nres--;
+		avail--;
+		stolen++;
+	}
+
+	/*
+	 * The only blocks available are those reserved for the original
+	 * extent and what we can steal from the extent being removed.
+	 * If this still isn't enough to satisfy the combined
+	 * requirements for the two new extents, skim blocks off of each
+	 * of the new reservations until they match what is available.
+	 */
+	while (nres > ores) {
+		if (len1) {
+			len1--;
+			nres--;
+		}
+		if (nres == ores)
+			break;
+		if (len2) {
+			len2--;
+			nres--;
+		}
+	}
+
+	*indlen1 = len1;
+	*indlen2 = len2;
+
+	return stolen;
+}
+
+/*
  * Called by xfs_bmapi to update file extent records and the btree
  * after removing space (or undoing a delayed allocation).
  */
@@ -4718,12 +4814,12 @@ xfs_bmap_del_extent(
 	xfs_filblks_t		temp2;	/* for indirect length calculations */
 	int			state = 0;
 
-	XFS_STATS_INC(xs_del_exlist);
+	mp = ip->i_mount;
+	XFS_STATS_INC(mp, xs_del_exlist);
 
 	if (whichfork == XFS_ATTR_FORK)
 		state |= BMAP_ATTRFORK;
 
-	mp = ip->i_mount;
 	ifp = XFS_IFORK_PTR(ip, whichfork);
 	ASSERT((*idx >= 0) && (*idx < ifp->if_bytes /
 		(uint)sizeof(xfs_bmbt_rec_t)));
@@ -4945,28 +5041,29 @@ xfs_bmap_del_extent(
 			XFS_IFORK_NEXT_SET(ip, whichfork,
 				XFS_IFORK_NEXTENTS(ip, whichfork) + 1);
 		} else {
+			xfs_filblks_t	stolen;
 			ASSERT(whichfork == XFS_DATA_FORK);
-			temp = xfs_bmap_worst_indlen(ip, temp);
+
+			/*
+			 * Distribute the original indlen reservation across the
+			 * two new extents. Steal blocks from the deleted extent
+			 * if necessary. Stealing blocks simply fudges the
+			 * fdblocks accounting in xfs_bunmapi().
+			 */
+			temp = xfs_bmap_worst_indlen(ip, got.br_blockcount);
+			temp2 = xfs_bmap_worst_indlen(ip, new.br_blockcount);
+			stolen = xfs_bmap_split_indlen(da_old, &temp, &temp2,
+						       del->br_blockcount);
+			da_new = temp + temp2 - stolen;
+			del->br_blockcount -= stolen;
+
+			/*
+			 * Set the reservation for each extent. Warn if either
+			 * is zero as this can lead to delalloc problems.
+			 */
+			WARN_ON_ONCE(!temp || !temp2);
 			xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
-			temp2 = xfs_bmap_worst_indlen(ip, temp2);
 			new.br_startblock = nullstartblock((int)temp2);
-			da_new = temp + temp2;
-			while (da_new > da_old) {
-				if (temp) {
-					temp--;
-					da_new--;
-					xfs_bmbt_set_startblock(ep,
-						nullstartblock((int)temp));
-				}
-				if (da_new == da_old)
-					break;
-				if (temp2) {
-					temp2--;
-					da_new--;
-					new.br_startblock =
-						nullstartblock((int)temp2);
-				}
-			}
 		}
 		trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);
 		xfs_iext_insert(ip, *idx + 1, 1, &new, state);
@@ -5070,7 +5167,7 @@ xfs_bunmapi(
 		*done = 1;
 		return 0;
 	}
-	XFS_STATS_INC(xs_blk_unmap);
+	XFS_STATS_INC(mp, xs_blk_unmap);
 	isrt = (whichfork == XFS_DATA_FORK) && XFS_IS_REALTIME_INODE(ip);
 	start = bno;
 	bno = start + len - 1;
@@ -5171,7 +5268,7 @@ xfs_bunmapi(
 			 * This is better than zeroing it.
 			 */
 			ASSERT(del.br_state == XFS_EXT_NORM);
-			ASSERT(xfs_trans_get_block_res(tp) > 0);
+			ASSERT(tp->t_blk_res > 0);
 			/*
 			 * If this spans a realtime extent boundary,
 			 * chop it back to the start of the one we end at.
@@ -5202,7 +5299,7 @@ xfs_bunmapi(
 				del.br_startblock += mod;
 			} else if ((del.br_startoff == start &&
 				    (del.br_state == XFS_EXT_UNWRITTEN ||
-				     xfs_trans_get_block_res(tp) == 0)) ||
+				     tp->t_blk_res == 0)) ||
 				   !xfs_sb_version_hasextflgbit(&mp->m_sb)) {
 				/*
 				 * Can't make it unwritten.  There isn't
@@ -5257,9 +5354,37 @@ xfs_bunmapi(
 				goto nodelete;
 			}
 		}
+
+		/*
+		 * If it's the case where the directory code is running
+		 * with no block reservation, and the deleted block is in
+		 * the middle of its extent, and the resulting insert
+		 * of an extent would cause transformation to btree format,
+		 * then reject it.  The calling code will then swap
+		 * blocks around instead.
+		 * We have to do this now, rather than waiting for the
+		 * conversion to btree format, since the transaction
+		 * will be dirty.
+		 */
+		if (!wasdel && tp->t_blk_res == 0 &&
+		    XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS &&
+		    XFS_IFORK_NEXTENTS(ip, whichfork) >= /* Note the >= */
+			XFS_IFORK_MAXEXT(ip, whichfork) &&
+		    del.br_startoff > got.br_startoff &&
+		    del.br_startoff + del.br_blockcount <
+		    got.br_startoff + got.br_blockcount) {
+			error = -ENOSPC;
+			goto error0;
+		}
+
+		/*
+		 * Unreserve quota and update realtime free space, if
+		 * appropriate. If delayed allocation, update the inode delalloc
+		 * counter now and wait to update the sb counters as
+		 * xfs_bmap_del_extent() might need to borrow some blocks.
+		 */
 		if (wasdel) {
 			ASSERT(startblockval(del.br_startblock) > 0);
-			/* Update realtime/data freespace, unreserve quota */
 			if (isrt) {
 				xfs_filblks_t rtexts;
 
@@ -5270,8 +5395,6 @@ xfs_bunmapi(
 					ip, -((long)del.br_blockcount), 0,
 					XFS_QMOPT_RES_RTBLKS);
 			} else {
-				xfs_mod_fdblocks(mp, (int64_t)del.br_blockcount,
-						 false);
 				(void)xfs_trans_reserve_quota_nblks(NULL,
 					ip, -((long)del.br_blockcount), 0,
 					XFS_QMOPT_RES_REGBLKS);
@@ -5282,32 +5405,16 @@ xfs_bunmapi(
 					XFS_BTCUR_BPRV_WASDEL;
 		} else if (cur)
 			cur->bc_private.b.flags &= ~XFS_BTCUR_BPRV_WASDEL;
-		/*
-		 * If it's the case where the directory code is running
-		 * with no block reservation, and the deleted block is in
-		 * the middle of its extent, and the resulting insert
-		 * of an extent would cause transformation to btree format,
-		 * then reject it.  The calling code will then swap
-		 * blocks around instead.
-		 * We have to do this now, rather than waiting for the
-		 * conversion to btree format, since the transaction
-		 * will be dirty.
-		 */
-		if (!wasdel && xfs_trans_get_block_res(tp) == 0 &&
-		    XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS &&
-		    XFS_IFORK_NEXTENTS(ip, whichfork) >= /* Note the >= */
-			XFS_IFORK_MAXEXT(ip, whichfork) &&
-		    del.br_startoff > got.br_startoff &&
-		    del.br_startoff + del.br_blockcount <
-		    got.br_startoff + got.br_blockcount) {
-			error = -ENOSPC;
-			goto error0;
-		}
+
 		error = xfs_bmap_del_extent(ip, tp, &lastx, flist, cur, &del,
 				&tmp_logflags, whichfork);
 		logflags |= tmp_logflags;
 		if (error)
 			goto error0;
+
+		if (!isrt && wasdel)
+			xfs_mod_fdblocks(mp, (int64_t)del.br_blockcount, false);
+
 		bno = del.br_startoff - 1;
 nodelete:
 		/*
@@ -5917,7 +6024,6 @@ xfs_bmap_split_extent(
 	struct xfs_trans        *tp;
 	struct xfs_bmap_free    free_list;
 	xfs_fsblock_t           firstfsb;
-	int                     committed;
 	int                     error;
 
 	tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT);
@@ -5938,7 +6044,7 @@ xfs_bmap_split_extent(
 	if (error)
 		goto out;
 
-	error = xfs_bmap_finish(&tp, &free_list, &committed);
+	error = xfs_bmap_finish(&tp, &free_list, NULL);
 	if (error)
 		goto out;
 
diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h
index 6aaa0c1c7200..423a34e832bd 100644
--- a/fs/xfs/libxfs/xfs_bmap.h
+++ b/fs/xfs/libxfs/xfs_bmap.h
@@ -52,9 +52,9 @@ struct xfs_bmalloca {
 	xfs_extlen_t		minleft; /* amount must be left after alloc */
 	bool			eof;	/* set if allocating past last extent */
 	bool			wasdel;	/* replacing a delayed allocation */
-	bool			userdata;/* set if is user data */
 	bool			aeof;	/* allocated space at eof */
 	bool			conv;	/* overwriting unwritten extents */
+	char			userdata;/* userdata mask */
 	int			flags;
 };
 
@@ -109,6 +109,14 @@ typedef	struct xfs_bmap_free
  */
 #define XFS_BMAPI_CONVERT	0x040
 
+/*
+ * allocate zeroed extents - this requires all newly allocated user data extents
+ * to be initialised to zero. It will be ignored if XFS_BMAPI_METADATA is set.
+ * Use in conjunction with XFS_BMAPI_CONVERT to convert unwritten extents found
+ * during the allocation range to zeroed written extents.
+ */
+#define XFS_BMAPI_ZERO		0x080
+
 #define XFS_BMAPI_FLAGS \
 	{ XFS_BMAPI_ENTIRE,	"ENTIRE" }, \
 	{ XFS_BMAPI_METADATA,	"METADATA" }, \
@@ -116,7 +124,8 @@ typedef	struct xfs_bmap_free
 	{ XFS_BMAPI_PREALLOC,	"PREALLOC" }, \
 	{ XFS_BMAPI_IGSTATE,	"IGSTATE" }, \
 	{ XFS_BMAPI_CONTIG,	"CONTIG" }, \
-	{ XFS_BMAPI_CONVERT,	"CONVERT" }
+	{ XFS_BMAPI_CONVERT,	"CONVERT" }, \
+	{ XFS_BMAPI_ZERO,	"ZERO" }
 
 
 static inline int xfs_bmapi_aflag(int w)
@@ -186,7 +195,7 @@ void	xfs_bmap_add_free(xfs_fsblock_t bno, xfs_filblks_t len,
 		struct xfs_bmap_free *flist, struct xfs_mount *mp);
 void	xfs_bmap_cancel(struct xfs_bmap_free *flist);
 int	xfs_bmap_finish(struct xfs_trans **tp, struct xfs_bmap_free *flist,
-			int *committed);
+			struct xfs_inode *ip);
 void	xfs_bmap_compute_maxlevels(struct xfs_mount *mp, int whichfork);
 int	xfs_bmap_first_unused(struct xfs_trans *tp, struct xfs_inode *ip,
 		xfs_extlen_t len, xfs_fileoff_t *unused, int whichfork);
diff --git a/fs/xfs/libxfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c
index 6b0cf6546a82..6282f6e708af 100644
--- a/fs/xfs/libxfs/xfs_bmap_btree.c
+++ b/fs/xfs/libxfs/xfs_bmap_btree.c
@@ -461,7 +461,7 @@ xfs_bmbt_alloc_block(
 		 * reservation amount is insufficient then we may fail a
 		 * block allocation here and corrupt the filesystem.
 		 */
-		args.minleft = xfs_trans_get_block_res(args.tp);
+		args.minleft = args.tp->t_blk_res;
 	} else if (cur->bc_private.b.flist->xbf_low) {
 		args.type = XFS_ALLOCTYPE_START_BNO;
 	} else {
@@ -470,7 +470,7 @@ xfs_bmbt_alloc_block(
 
 	args.minlen = args.maxlen = args.prod = 1;
 	args.wasdel = cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL;
-	if (!args.wasdel && xfs_trans_get_block_res(args.tp) == 0) {
+	if (!args.wasdel && args.tp->t_blk_res == 0) {
 		error = -ENOSPC;
 		goto error0;
 	}
@@ -531,7 +531,6 @@ xfs_bmbt_free_block(
 
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 	xfs_trans_mod_dquot_byino(tp, ip, XFS_TRANS_DQ_BCOUNT, -1L);
-	xfs_trans_binval(tp, bp);
 	return 0;
 }
 
@@ -720,6 +719,7 @@ xfs_bmbt_write_verify(
 }
 
 const struct xfs_buf_ops xfs_bmbt_buf_ops = {
+	.name = "xfs_bmbt",
 	.verify_read = xfs_bmbt_read_verify,
 	.verify_write = xfs_bmbt_write_verify,
 };
diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
index f7d7ee7a2607..1f88e1ce770f 100644
--- a/fs/xfs/libxfs/xfs_btree.c
+++ b/fs/xfs/libxfs/xfs_btree.c
@@ -32,6 +32,7 @@
 #include "xfs_trace.h"
 #include "xfs_cksum.h"
 #include "xfs_alloc.h"
+#include "xfs_log.h"
 
 /*
  * Cursor allocation zone.
@@ -222,7 +223,7 @@ xfs_btree_check_ptr(
  * long-form btree header.
  *
  * Prior to calculting the CRC, pull the LSN out of the buffer log item and put
- * it into the buffer so recovery knows what the last modifcation was that made
+ * it into the buffer so recovery knows what the last modification was that made
  * it to disk.
  */
 void
@@ -243,8 +244,14 @@ bool
 xfs_btree_lblock_verify_crc(
 	struct xfs_buf		*bp)
 {
-	if (xfs_sb_version_hascrc(&bp->b_target->bt_mount->m_sb))
+	struct xfs_btree_block	*block = XFS_BUF_TO_BLOCK(bp);
+	struct xfs_mount	*mp = bp->b_target->bt_mount;
+
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		if (!xfs_log_check_lsn(mp, be64_to_cpu(block->bb_u.l.bb_lsn)))
+			return false;
 		return xfs_buf_verify_cksum(bp, XFS_BTREE_LBLOCK_CRC_OFF);
+	}
 
 	return true;
 }
@@ -254,7 +261,7 @@ xfs_btree_lblock_verify_crc(
  * short-form btree header.
  *
  * Prior to calculting the CRC, pull the LSN out of the buffer log item and put
- * it into the buffer so recovery knows what the last modifcation was that made
+ * it into the buffer so recovery knows what the last modification was that made
  * it to disk.
  */
 void
@@ -275,12 +282,33 @@ bool
 xfs_btree_sblock_verify_crc(
 	struct xfs_buf		*bp)
 {
-	if (xfs_sb_version_hascrc(&bp->b_target->bt_mount->m_sb))
+	struct xfs_btree_block  *block = XFS_BUF_TO_BLOCK(bp);
+	struct xfs_mount	*mp = bp->b_target->bt_mount;
+
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		if (!xfs_log_check_lsn(mp, be64_to_cpu(block->bb_u.s.bb_lsn)))
+			return false;
 		return xfs_buf_verify_cksum(bp, XFS_BTREE_SBLOCK_CRC_OFF);
+	}
 
 	return true;
 }
 
+static int
+xfs_btree_free_block(
+	struct xfs_btree_cur	*cur,
+	struct xfs_buf		*bp)
+{
+	int			error;
+
+	error = cur->bc_ops->free_block(cur, bp);
+	if (!error) {
+		xfs_trans_binval(cur->bc_tp, bp);
+		XFS_BTREE_STATS_INC(cur, free);
+	}
+	return error;
+}
+
 /*
  * Delete the btree cursor.
  */
@@ -3196,6 +3224,7 @@ xfs_btree_kill_iroot(
 	int			level;
 	int			index;
 	int			numrecs;
+	int			error;
 #ifdef DEBUG
 	union xfs_btree_ptr	ptr;
 	int			i;
@@ -3259,8 +3288,6 @@ xfs_btree_kill_iroot(
 	cpp = xfs_btree_ptr_addr(cur, 1, cblock);
 #ifdef DEBUG
 	for (i = 0; i < numrecs; i++) {
-		int		error;
-
 		error = xfs_btree_check_ptr(cur, cpp, i, level - 1);
 		if (error) {
 			XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR);
@@ -3270,8 +3297,11 @@ xfs_btree_kill_iroot(
 #endif
 	xfs_btree_copy_ptrs(cur, pp, cpp, numrecs);
 
-	cur->bc_ops->free_block(cur, cbp);
-	XFS_BTREE_STATS_INC(cur, free);
+	error = xfs_btree_free_block(cur, cbp);
+	if (error) {
+		XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR);
+		return error;
+	}
 
 	cur->bc_bufs[level - 1] = NULL;
 	be16_add_cpu(&block->bb_level, -1);
@@ -3304,14 +3334,12 @@ xfs_btree_kill_root(
 	 */
 	cur->bc_ops->set_root(cur, newroot, -1);
 
-	error = cur->bc_ops->free_block(cur, bp);
+	error = xfs_btree_free_block(cur, bp);
 	if (error) {
 		XFS_BTREE_TRACE_CURSOR(cur, XBT_ERROR);
 		return error;
 	}
 
-	XFS_BTREE_STATS_INC(cur, free);
-
 	cur->bc_bufs[level] = NULL;
 	cur->bc_ra[level] = 0;
 	cur->bc_nlevels--;
@@ -3817,10 +3845,9 @@ xfs_btree_delrec(
 	}
 
 	/* Free the deleted block. */
-	error = cur->bc_ops->free_block(cur, rbp);
+	error = xfs_btree_free_block(cur, rbp);
 	if (error)
 		goto error0;
-	XFS_BTREE_STATS_INC(cur, free);
 
 	/*
 	 * If we joined with the left neighbor, set the buffer in the
@@ -4067,3 +4094,61 @@ xfs_btree_change_owner(
 
 	return 0;
 }
+
+/**
+ * xfs_btree_sblock_v5hdr_verify() -- verify the v5 fields of a short-format
+ *				      btree block
+ *
+ * @bp: buffer containing the btree block
+ * @max_recs: pointer to the m_*_mxr max records field in the xfs mount
+ * @pag_max_level: pointer to the per-ag max level field
+ */
+bool
+xfs_btree_sblock_v5hdr_verify(
+	struct xfs_buf		*bp)
+{
+	struct xfs_mount	*mp = bp->b_target->bt_mount;
+	struct xfs_btree_block	*block = XFS_BUF_TO_BLOCK(bp);
+	struct xfs_perag	*pag = bp->b_pag;
+
+	if (!xfs_sb_version_hascrc(&mp->m_sb))
+		return false;
+	if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid))
+		return false;
+	if (block->bb_u.s.bb_blkno != cpu_to_be64(bp->b_bn))
+		return false;
+	if (pag && be32_to_cpu(block->bb_u.s.bb_owner) != pag->pag_agno)
+		return false;
+	return true;
+}
+
+/**
+ * xfs_btree_sblock_verify() -- verify a short-format btree block
+ *
+ * @bp: buffer containing the btree block
+ * @max_recs: maximum records allowed in this btree node
+ */
+bool
+xfs_btree_sblock_verify(
+	struct xfs_buf		*bp,
+	unsigned int		max_recs)
+{
+	struct xfs_mount	*mp = bp->b_target->bt_mount;
+	struct xfs_btree_block	*block = XFS_BUF_TO_BLOCK(bp);
+
+	/* numrecs verification */
+	if (be16_to_cpu(block->bb_numrecs) > max_recs)
+		return false;
+
+	/* sibling pointer verification */
+	if (!block->bb_u.s.bb_leftsib ||
+	    (be32_to_cpu(block->bb_u.s.bb_leftsib) >= mp->m_sb.sb_agblocks &&
+	     block->bb_u.s.bb_leftsib != cpu_to_be32(NULLAGBLOCK)))
+		return false;
+	if (!block->bb_u.s.bb_rightsib ||
+	    (be32_to_cpu(block->bb_u.s.bb_rightsib) >= mp->m_sb.sb_agblocks &&
+	     block->bb_u.s.bb_rightsib != cpu_to_be32(NULLAGBLOCK)))
+		return false;
+
+	return true;
+}
diff --git a/fs/xfs/libxfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h
index 8f18bab73ea5..2e874be70209 100644
--- a/fs/xfs/libxfs/xfs_btree.h
+++ b/fs/xfs/libxfs/xfs_btree.h
@@ -84,31 +84,38 @@ union xfs_btree_rec {
 /*
  * Generic stats interface
  */
-#define __XFS_BTREE_STATS_INC(type, stat) \
-	XFS_STATS_INC(xs_ ## type ## _2_ ## stat)
-#define XFS_BTREE_STATS_INC(cur, stat)  \
+#define __XFS_BTREE_STATS_INC(mp, type, stat) \
+	XFS_STATS_INC(mp, xs_ ## type ## _2_ ## stat)
+#define XFS_BTREE_STATS_INC(cur, stat)	\
 do {    \
+	struct xfs_mount *__mp = cur->bc_mp; \
 	switch (cur->bc_btnum) {  \
-	case XFS_BTNUM_BNO: __XFS_BTREE_STATS_INC(abtb, stat); break;	\
-	case XFS_BTNUM_CNT: __XFS_BTREE_STATS_INC(abtc, stat); break;	\
-	case XFS_BTNUM_BMAP: __XFS_BTREE_STATS_INC(bmbt, stat); break;	\
-	case XFS_BTNUM_INO: __XFS_BTREE_STATS_INC(ibt, stat); break;	\
-	case XFS_BTNUM_FINO: __XFS_BTREE_STATS_INC(fibt, stat); break;	\
+	case XFS_BTNUM_BNO: __XFS_BTREE_STATS_INC(__mp, abtb, stat); break; \
+	case XFS_BTNUM_CNT: __XFS_BTREE_STATS_INC(__mp, abtc, stat); break; \
+	case XFS_BTNUM_BMAP: __XFS_BTREE_STATS_INC(__mp, bmbt, stat); break; \
+	case XFS_BTNUM_INO: __XFS_BTREE_STATS_INC(__mp, ibt, stat); break; \
+	case XFS_BTNUM_FINO: __XFS_BTREE_STATS_INC(__mp, fibt, stat); break; \
 	case XFS_BTNUM_MAX: ASSERT(0); /* fucking gcc */ ; break;	\
 	}       \
 } while (0)
 
-#define __XFS_BTREE_STATS_ADD(type, stat, val) \
-	XFS_STATS_ADD(xs_ ## type ## _2_ ## stat, val)
+#define __XFS_BTREE_STATS_ADD(mp, type, stat, val) \
+	XFS_STATS_ADD(mp, xs_ ## type ## _2_ ## stat, val)
 #define XFS_BTREE_STATS_ADD(cur, stat, val)  \
 do {    \
+	struct xfs_mount *__mp = cur->bc_mp; \
 	switch (cur->bc_btnum) {  \
-	case XFS_BTNUM_BNO: __XFS_BTREE_STATS_ADD(abtb, stat, val); break; \
-	case XFS_BTNUM_CNT: __XFS_BTREE_STATS_ADD(abtc, stat, val); break; \
-	case XFS_BTNUM_BMAP: __XFS_BTREE_STATS_ADD(bmbt, stat, val); break; \
-	case XFS_BTNUM_INO: __XFS_BTREE_STATS_ADD(ibt, stat, val); break; \
-	case XFS_BTNUM_FINO: __XFS_BTREE_STATS_ADD(fibt, stat, val); break; \
-	case XFS_BTNUM_MAX: ASSERT(0); /* fucking gcc */ ; break;	\
+	case XFS_BTNUM_BNO:	\
+		__XFS_BTREE_STATS_ADD(__mp, abtb, stat, val); break; \
+	case XFS_BTNUM_CNT:	\
+		__XFS_BTREE_STATS_ADD(__mp, abtc, stat, val); break; \
+	case XFS_BTNUM_BMAP:	\
+		__XFS_BTREE_STATS_ADD(__mp, bmbt, stat, val); break; \
+	case XFS_BTNUM_INO:	\
+		__XFS_BTREE_STATS_ADD(__mp, ibt, stat, val); break; \
+	case XFS_BTNUM_FINO:	\
+		__XFS_BTREE_STATS_ADD(__mp, fibt, stat, val); break; \
+	case XFS_BTNUM_MAX: ASSERT(0); /* fucking gcc */ ; break; \
 	}       \
 } while (0)
 
@@ -465,4 +472,7 @@ static inline int xfs_btree_get_level(struct xfs_btree_block *block)
 #define XFS_BTREE_TRACE_ARGR(c, r)
 #define	XFS_BTREE_TRACE_CURSOR(c, t)
 
+bool xfs_btree_sblock_v5hdr_verify(struct xfs_buf *bp);
+bool xfs_btree_sblock_verify(struct xfs_buf *bp, unsigned int max_recs);
+
 #endif	/* __XFS_BTREE_H__ */
diff --git a/fs/xfs/libxfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c
index be43248a5822..097bf7717d80 100644
--- a/fs/xfs/libxfs/xfs_da_btree.c
+++ b/fs/xfs/libxfs/xfs_da_btree.c
@@ -39,6 +39,7 @@
 #include "xfs_trace.h"
 #include "xfs_cksum.h"
 #include "xfs_buf_item.h"
+#include "xfs_log.h"
 
 /*
  * xfs_da_btree.c
@@ -150,6 +151,8 @@ xfs_da3_node_verify(
 			return false;
 		if (be64_to_cpu(hdr3->info.blkno) != bp->b_bn)
 			return false;
+		if (!xfs_log_check_lsn(mp, be64_to_cpu(hdr3->info.lsn)))
+			return false;
 	} else {
 		if (ichdr.magic != XFS_DA_NODE_MAGIC)
 			return false;
@@ -242,6 +245,7 @@ xfs_da3_node_read_verify(
 }
 
 const struct xfs_buf_ops xfs_da3_node_buf_ops = {
+	.name = "xfs_da3_node",
 	.verify_read = xfs_da3_node_read_verify,
 	.verify_write = xfs_da3_node_write_verify,
 };
@@ -322,6 +326,7 @@ xfs_da3_node_create(
 	if (xfs_sb_version_hascrc(&mp->m_sb)) {
 		struct xfs_da3_node_hdr *hdr3 = bp->b_addr;
 
+		memset(hdr3, 0, sizeof(struct xfs_da3_node_hdr));
 		ichdr.magic = XFS_DA3_NODE_MAGIC;
 		hdr3->info.blkno = cpu_to_be64(bp->b_bn);
 		hdr3->info.owner = cpu_to_be64(args->dp->i_ino);
diff --git a/fs/xfs/libxfs/xfs_da_format.h b/fs/xfs/libxfs/xfs_da_format.h
index b14bbd6bb05f..8d4d8bce41bf 100644
--- a/fs/xfs/libxfs/xfs_da_format.h
+++ b/fs/xfs/libxfs/xfs_da_format.h
@@ -641,6 +641,22 @@ xfs_dir2_block_leaf_p(struct xfs_dir2_block_tail *btp)
  */
 #define XFS_ATTR_LEAF_MAPSIZE	3	/* how many freespace slots */
 
+/*
+ * Entries are packed toward the top as tight as possible.
+ */
+typedef struct xfs_attr_shortform {
+	struct xfs_attr_sf_hdr {	/* constant-structure header block */
+		__be16	totsize;	/* total bytes in shortform list */
+		__u8	count;	/* count of active entries */
+	} hdr;
+	struct xfs_attr_sf_entry {
+		__uint8_t namelen;	/* actual length of name (no NULL) */
+		__uint8_t valuelen;	/* actual length of value (no NULL) */
+		__uint8_t flags;	/* flags bits (see xfs_attr_leaf.h) */
+		__uint8_t nameval[1];	/* name & value bytes concatenated */
+	} list[1];			/* variable sized array */
+} xfs_attr_shortform_t;
+
 typedef struct xfs_attr_leaf_map {	/* RLE map of free bytes */
 	__be16	base;			  /* base of free region */
 	__be16	size;			  /* length of free region */
diff --git a/fs/xfs/libxfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c
index 9de401d297e5..af0f9d171f8a 100644
--- a/fs/xfs/libxfs/xfs_dir2.c
+++ b/fs/xfs/libxfs/xfs_dir2.c
@@ -176,7 +176,7 @@ xfs_dir_isempty(
 {
 	xfs_dir2_sf_hdr_t	*sfp;
 
-	ASSERT(S_ISDIR(dp->i_d.di_mode));
+	ASSERT(S_ISDIR(VFS_I(dp)->i_mode));
 	if (dp->i_d.di_size == 0)	/* might happen during shutdown. */
 		return 1;
 	if (dp->i_d.di_size > XFS_IFORK_DSIZE(dp))
@@ -231,7 +231,7 @@ xfs_dir_init(
 	struct xfs_da_args *args;
 	int		error;
 
-	ASSERT(S_ISDIR(dp->i_d.di_mode));
+	ASSERT(S_ISDIR(VFS_I(dp)->i_mode));
 	error = xfs_dir_ino_validate(tp->t_mountp, pdp->i_ino);
 	if (error)
 		return error;
@@ -266,12 +266,12 @@ xfs_dir_createname(
 	int			rval;
 	int			v;		/* type-checking value */
 
-	ASSERT(S_ISDIR(dp->i_d.di_mode));
+	ASSERT(S_ISDIR(VFS_I(dp)->i_mode));
 	if (inum) {
 		rval = xfs_dir_ino_validate(tp->t_mountp, inum);
 		if (rval)
 			return rval;
-		XFS_STATS_INC(xs_dir_create);
+		XFS_STATS_INC(dp->i_mount, xs_dir_create);
 	}
 
 	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
@@ -364,8 +364,8 @@ xfs_dir_lookup(
 	int		v;		/* type-checking value */
 	int		lock_mode;
 
-	ASSERT(S_ISDIR(dp->i_d.di_mode));
-	XFS_STATS_INC(xs_dir_lookup);
+	ASSERT(S_ISDIR(VFS_I(dp)->i_mode));
+	XFS_STATS_INC(dp->i_mount, xs_dir_lookup);
 
 	/*
 	 * We need to use KM_NOFS here so that lockdep will not throw false
@@ -443,8 +443,8 @@ xfs_dir_removename(
 	int		rval;
 	int		v;		/* type-checking value */
 
-	ASSERT(S_ISDIR(dp->i_d.di_mode));
-	XFS_STATS_INC(xs_dir_remove);
+	ASSERT(S_ISDIR(VFS_I(dp)->i_mode));
+	XFS_STATS_INC(dp->i_mount, xs_dir_remove);
 
 	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
 	if (!args)
@@ -505,7 +505,7 @@ xfs_dir_replace(
 	int		rval;
 	int		v;		/* type-checking value */
 
-	ASSERT(S_ISDIR(dp->i_d.di_mode));
+	ASSERT(S_ISDIR(VFS_I(dp)->i_mode));
 
 	rval = xfs_dir_ino_validate(tp->t_mountp, inum);
 	if (rval)
diff --git a/fs/xfs/libxfs/xfs_dir2_block.c b/fs/xfs/libxfs/xfs_dir2_block.c
index 4778d1dd511a..aa17cb788946 100644
--- a/fs/xfs/libxfs/xfs_dir2_block.c
+++ b/fs/xfs/libxfs/xfs_dir2_block.c
@@ -34,6 +34,7 @@
 #include "xfs_error.h"
 #include "xfs_trace.h"
 #include "xfs_cksum.h"
+#include "xfs_log.h"
 
 /*
  * Local function prototypes.
@@ -71,6 +72,8 @@ xfs_dir3_block_verify(
 			return false;
 		if (be64_to_cpu(hdr3->blkno) != bp->b_bn)
 			return false;
+		if (!xfs_log_check_lsn(mp, be64_to_cpu(hdr3->lsn)))
+			return false;
 	} else {
 		if (hdr3->magic != cpu_to_be32(XFS_DIR2_BLOCK_MAGIC))
 			return false;
@@ -120,6 +123,7 @@ xfs_dir3_block_write_verify(
 }
 
 const struct xfs_buf_ops xfs_dir3_block_buf_ops = {
+	.name = "xfs_dir3_block",
 	.verify_read = xfs_dir3_block_read_verify,
 	.verify_write = xfs_dir3_block_write_verify,
 };
diff --git a/fs/xfs/libxfs/xfs_dir2_data.c b/fs/xfs/libxfs/xfs_dir2_data.c
index 824131e71bc5..725fc7841fde 100644
--- a/fs/xfs/libxfs/xfs_dir2_data.c
+++ b/fs/xfs/libxfs/xfs_dir2_data.c
@@ -31,6 +31,7 @@
 #include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_cksum.h"
+#include "xfs_log.h"
 
 /*
  * Check the consistency of the data block.
@@ -224,6 +225,8 @@ xfs_dir3_data_verify(
 			return false;
 		if (be64_to_cpu(hdr3->blkno) != bp->b_bn)
 			return false;
+		if (!xfs_log_check_lsn(mp, be64_to_cpu(hdr3->lsn)))
+			return false;
 	} else {
 		if (hdr3->magic != cpu_to_be32(XFS_DIR2_DATA_MAGIC))
 			return false;
@@ -302,11 +305,13 @@ xfs_dir3_data_write_verify(
 }
 
 const struct xfs_buf_ops xfs_dir3_data_buf_ops = {
+	.name = "xfs_dir3_data",
 	.verify_read = xfs_dir3_data_read_verify,
 	.verify_write = xfs_dir3_data_write_verify,
 };
 
 static const struct xfs_buf_ops xfs_dir3_data_reada_buf_ops = {
+	.name = "xfs_dir3_data_reada",
 	.verify_read = xfs_dir3_data_reada_verify,
 	.verify_write = xfs_dir3_data_write_verify,
 };
diff --git a/fs/xfs/libxfs/xfs_dir2_leaf.c b/fs/xfs/libxfs/xfs_dir2_leaf.c
index f300240ebb8d..b887fb2a2bcf 100644
--- a/fs/xfs/libxfs/xfs_dir2_leaf.c
+++ b/fs/xfs/libxfs/xfs_dir2_leaf.c
@@ -33,6 +33,7 @@
 #include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_cksum.h"
+#include "xfs_log.h"
 
 /*
  * Local function declarations.
@@ -164,6 +165,8 @@ xfs_dir3_leaf_verify(
 			return false;
 		if (be64_to_cpu(leaf3->info.blkno) != bp->b_bn)
 			return false;
+		if (!xfs_log_check_lsn(mp, be64_to_cpu(leaf3->info.lsn)))
+			return false;
 	} else {
 		if (leaf->hdr.info.magic != cpu_to_be16(magic))
 			return false;
@@ -242,11 +245,13 @@ xfs_dir3_leafn_write_verify(
 }
 
 const struct xfs_buf_ops xfs_dir3_leaf1_buf_ops = {
+	.name = "xfs_dir3_leaf1",
 	.verify_read = xfs_dir3_leaf1_read_verify,
 	.verify_write = xfs_dir3_leaf1_write_verify,
 };
 
 const struct xfs_buf_ops xfs_dir3_leafn_buf_ops = {
+	.name = "xfs_dir3_leafn",
 	.verify_read = xfs_dir3_leafn_read_verify,
 	.verify_write = xfs_dir3_leafn_write_verify,
 };
diff --git a/fs/xfs/libxfs/xfs_dir2_node.c b/fs/xfs/libxfs/xfs_dir2_node.c
index cc28e924545b..75a557432d0f 100644
--- a/fs/xfs/libxfs/xfs_dir2_node.c
+++ b/fs/xfs/libxfs/xfs_dir2_node.c
@@ -33,6 +33,7 @@
 #include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_cksum.h"
+#include "xfs_log.h"
 
 /*
  * Function declarations.
@@ -97,6 +98,8 @@ xfs_dir3_free_verify(
 			return false;
 		if (be64_to_cpu(hdr3->blkno) != bp->b_bn)
 			return false;
+		if (!xfs_log_check_lsn(mp, be64_to_cpu(hdr3->lsn)))
+			return false;
 	} else {
 		if (hdr->magic != cpu_to_be32(XFS_DIR2_FREE_MAGIC))
 			return false;
@@ -147,6 +150,7 @@ xfs_dir3_free_write_verify(
 }
 
 const struct xfs_buf_ops xfs_dir3_free_buf_ops = {
+	.name = "xfs_dir3_free",
 	.verify_read = xfs_dir3_free_read_verify,
 	.verify_write = xfs_dir3_free_write_verify,
 };
@@ -2231,6 +2235,9 @@ xfs_dir2_node_trim_free(
 
 	dp = args->dp;
 	tp = args->trans;
+
+	*rvalp = 0;
+
 	/*
 	 * Read the freespace block.
 	 */
@@ -2251,7 +2258,6 @@ xfs_dir2_node_trim_free(
 	 */
 	if (freehdr.nused > 0) {
 		xfs_trans_brelse(tp, bp);
-		*rvalp = 0;
 		return 0;
 	}
 	/*
diff --git a/fs/xfs/libxfs/xfs_dquot_buf.c b/fs/xfs/libxfs/xfs_dquot_buf.c
index 5331b7f0460c..3cc3cf767474 100644
--- a/fs/xfs/libxfs/xfs_dquot_buf.c
+++ b/fs/xfs/libxfs/xfs_dquot_buf.c
@@ -54,7 +54,7 @@ xfs_dqcheck(
 	xfs_dqid_t	 id,
 	uint		 type,	  /* used only when IO_dorepair is true */
 	uint		 flags,
-	char		 *str)
+	const char	 *str)
 {
 	xfs_dqblk_t	 *d = (xfs_dqblk_t *)ddq;
 	int		errs = 0;
@@ -207,7 +207,8 @@ xfs_dquot_buf_verify_crc(
 STATIC bool
 xfs_dquot_buf_verify(
 	struct xfs_mount	*mp,
-	struct xfs_buf		*bp)
+	struct xfs_buf		*bp,
+	int			warn)
 {
 	struct xfs_dqblk	*d = (struct xfs_dqblk *)bp->b_addr;
 	xfs_dqid_t		id = 0;
@@ -240,8 +241,7 @@ xfs_dquot_buf_verify(
 		if (i == 0)
 			id = be32_to_cpu(ddq->d_id);
 
-		error = xfs_dqcheck(mp, ddq, id + i, 0, XFS_QMOPT_DOWARN,
-				       "xfs_dquot_buf_verify");
+		error = xfs_dqcheck(mp, ddq, id + i, 0, warn, __func__);
 		if (error)
 			return false;
 	}
@@ -256,7 +256,7 @@ xfs_dquot_buf_read_verify(
 
 	if (!xfs_dquot_buf_verify_crc(mp, bp))
 		xfs_buf_ioerror(bp, -EFSBADCRC);
-	else if (!xfs_dquot_buf_verify(mp, bp))
+	else if (!xfs_dquot_buf_verify(mp, bp, XFS_QMOPT_DOWARN))
 		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
@@ -264,6 +264,25 @@ xfs_dquot_buf_read_verify(
 }
 
 /*
+ * readahead errors are silent and simply leave the buffer as !done so a real
+ * read will then be run with the xfs_dquot_buf_ops verifier. See
+ * xfs_inode_buf_verify() for why we use EIO and ~XBF_DONE here rather than
+ * reporting the failure.
+ */
+static void
+xfs_dquot_buf_readahead_verify(
+	struct xfs_buf	*bp)
+{
+	struct xfs_mount	*mp = bp->b_target->bt_mount;
+
+	if (!xfs_dquot_buf_verify_crc(mp, bp) ||
+	    !xfs_dquot_buf_verify(mp, bp, 0)) {
+		xfs_buf_ioerror(bp, -EIO);
+		bp->b_flags &= ~XBF_DONE;
+	}
+}
+
+/*
  * we don't calculate the CRC here as that is done when the dquot is flushed to
  * the buffer after the update is done. This ensures that the dquot in the
  * buffer always has an up-to-date CRC value.
@@ -274,7 +293,7 @@ xfs_dquot_buf_write_verify(
 {
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 
-	if (!xfs_dquot_buf_verify(mp, bp)) {
+	if (!xfs_dquot_buf_verify(mp, bp, XFS_QMOPT_DOWARN)) {
 		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
@@ -282,7 +301,13 @@ xfs_dquot_buf_write_verify(
 }
 
 const struct xfs_buf_ops xfs_dquot_buf_ops = {
+	.name = "xfs_dquot",
 	.verify_read = xfs_dquot_buf_read_verify,
 	.verify_write = xfs_dquot_buf_write_verify,
 };
 
+const struct xfs_buf_ops xfs_dquot_buf_ra_ops = {
+	.name = "xfs_dquot_ra",
+	.verify_read = xfs_dquot_buf_readahead_verify,
+	.verify_write = xfs_dquot_buf_write_verify,
+};
diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h
index 9590a069e556..dc97eb21af07 100644
--- a/fs/xfs/libxfs/xfs_format.h
+++ b/fs/xfs/libxfs/xfs_format.h
@@ -60,6 +60,14 @@ struct xfs_ifork;
 #define	XFS_SB_VERSION_MOREBITSBIT	0x8000
 
 /*
+ * The size of a single extended attribute on disk is limited by
+ * the size of index values within the attribute entries themselves.
+ * These are be16 fields, so we can only support attribute data
+ * sizes up to 2^16 bytes in length.
+ */
+#define XFS_XATTR_SIZE_MAX (1 << 16)
+
+/*
  * Supported feature bit list is just all bits in the versionnum field because
  * we've used them all up and understand them all. Except, of course, for the
  * shared superblock bit, which nobody knows what it does and so is unsupported.
@@ -778,7 +786,7 @@ typedef struct xfs_agfl {
 	__be64		agfl_lsn;
 	__be32		agfl_crc;
 	__be32		agfl_bno[];	/* actually XFS_AGFL_SIZE(mp) */
-} xfs_agfl_t;
+} __attribute__((packed)) xfs_agfl_t;
 
 #define XFS_AGFL_CRC_OFF	offsetof(struct xfs_agfl, agfl_crc)
 
@@ -976,8 +984,6 @@ static inline void xfs_dinode_put_rdev(struct xfs_dinode *dip, xfs_dev_t rdev)
 
 /*
  * Values for di_flags
- * There should be a one-to-one correspondence between these flags and the
- * XFS_XFLAG_s.
  */
 #define XFS_DIFLAG_REALTIME_BIT  0	/* file's blocks come from rt area */
 #define XFS_DIFLAG_PREALLOC_BIT  1	/* file space has been preallocated */
@@ -1018,6 +1024,15 @@ static inline void xfs_dinode_put_rdev(struct xfs_dinode *dip, xfs_dev_t rdev)
 	 XFS_DIFLAG_EXTSZINHERIT | XFS_DIFLAG_NODEFRAG | XFS_DIFLAG_FILESTREAM)
 
 /*
+ * Values for di_flags2 These start by being exposed to userspace in the upper
+ * 16 bits of the XFS_XFLAG_s range.
+ */
+#define XFS_DIFLAG2_DAX_BIT	0	/* use DAX for this inode */
+#define XFS_DIFLAG2_DAX		(1 << XFS_DIFLAG2_DAX_BIT)
+
+#define XFS_DIFLAG2_ANY		(XFS_DIFLAG2_DAX)
+
+/*
  * Inode number format:
  * low inopblog bits - offset in block
  * next agblklog bits - block number in ag
@@ -1483,13 +1498,17 @@ struct xfs_acl {
  */
 #define XFS_ACL_MAX_ENTRIES(mp)	\
 	(xfs_sb_version_hascrc(&mp->m_sb) \
-		?  (XATTR_SIZE_MAX - sizeof(struct xfs_acl)) / \
+		?  (XFS_XATTR_SIZE_MAX - sizeof(struct xfs_acl)) / \
 						sizeof(struct xfs_acl_entry) \
 		: 25)
 
-#define XFS_ACL_MAX_SIZE(mp) \
+#define XFS_ACL_SIZE(cnt) \
 	(sizeof(struct xfs_acl) + \
-		sizeof(struct xfs_acl_entry) * XFS_ACL_MAX_ENTRIES((mp)))
+		sizeof(struct xfs_acl_entry) * cnt)
+
+#define XFS_ACL_MAX_SIZE(mp) \
+	XFS_ACL_SIZE(XFS_ACL_MAX_ENTRIES((mp)))
+
 
 /* On-disk XFS extended attribute names */
 #define SGI_ACL_FILE		"SGI_ACL_FILE"
diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h
index 89689c6a43e2..fffe3d01bd9f 100644
--- a/fs/xfs/libxfs/xfs_fs.h
+++ b/fs/xfs/libxfs/xfs_fs.h
@@ -36,40 +36,6 @@ struct dioattr {
 #endif
 
 /*
- * Structure for XFS_IOC_FSGETXATTR[A] and XFS_IOC_FSSETXATTR.
- */
-#ifndef HAVE_FSXATTR
-struct fsxattr {
-	__u32		fsx_xflags;	/* xflags field value (get/set) */
-	__u32		fsx_extsize;	/* extsize field value (get/set)*/
-	__u32		fsx_nextents;	/* nextents field value (get)	*/
-	__u32		fsx_projid;	/* project identifier (get/set) */
-	unsigned char	fsx_pad[12];
-};
-#endif
-
-/*
- * Flags for the bs_xflags/fsx_xflags field
- * There should be a one-to-one correspondence between these flags and the
- * XFS_DIFLAG_s.
- */
-#define XFS_XFLAG_REALTIME	0x00000001	/* data in realtime volume */
-#define XFS_XFLAG_PREALLOC	0x00000002	/* preallocated file extents */
-#define XFS_XFLAG_IMMUTABLE	0x00000008	/* file cannot be modified */
-#define XFS_XFLAG_APPEND	0x00000010	/* all writes append */
-#define XFS_XFLAG_SYNC		0x00000020	/* all writes synchronous */
-#define XFS_XFLAG_NOATIME	0x00000040	/* do not update access time */
-#define XFS_XFLAG_NODUMP	0x00000080	/* do not include in backups */
-#define XFS_XFLAG_RTINHERIT	0x00000100	/* create with rt bit set */
-#define XFS_XFLAG_PROJINHERIT	0x00000200	/* create with parents projid */
-#define XFS_XFLAG_NOSYMLINKS	0x00000400	/* disallow symlink creation */
-#define XFS_XFLAG_EXTSIZE	0x00000800	/* extent size allocator hint */
-#define XFS_XFLAG_EXTSZINHERIT	0x00001000	/* inherit inode extent size */
-#define XFS_XFLAG_NODEFRAG	0x00002000  	/* do not defragment */
-#define XFS_XFLAG_FILESTREAM	0x00004000	/* use filestream allocator */
-#define XFS_XFLAG_HASATTR	0x80000000	/* no DIFLAG for this	*/
-
-/*
  * Structure for XFS_IOC_GETBMAP.
  * On input, fill in bmv_offset and bmv_length of the first structure
  * to indicate the area of interest in the file, and bmv_entries with
@@ -490,6 +456,16 @@ typedef struct xfs_swapext
 #define XFS_FSOP_GOING_FLAGS_NOLOGFLUSH		0x2	/* don't flush log nor data */
 
 /*
+ * ioctl limits
+ */
+#ifdef XATTR_LIST_MAX
+#  define XFS_XATTR_LIST_MAX XATTR_LIST_MAX
+#else
+#  define XFS_XATTR_LIST_MAX 65536
+#endif
+
+
+/*
  * ioctl commands that are used by Linux filesystems
  */
 #define XFS_IOC_GETXFLAGS	FS_IOC_GETFLAGS
@@ -504,8 +480,8 @@ typedef struct xfs_swapext
 #define XFS_IOC_ALLOCSP		_IOW ('X', 10, struct xfs_flock64)
 #define XFS_IOC_FREESP		_IOW ('X', 11, struct xfs_flock64)
 #define XFS_IOC_DIOINFO		_IOR ('X', 30, struct dioattr)
-#define XFS_IOC_FSGETXATTR	_IOR ('X', 31, struct fsxattr)
-#define XFS_IOC_FSSETXATTR	_IOW ('X', 32, struct fsxattr)
+#define XFS_IOC_FSGETXATTR	FS_IOC_FSGETXATTR
+#define XFS_IOC_FSSETXATTR	FS_IOC_FSSETXATTR
 #define XFS_IOC_ALLOCSP64	_IOW ('X', 36, struct xfs_flock64)
 #define XFS_IOC_FREESP64	_IOW ('X', 37, struct xfs_flock64)
 #define XFS_IOC_GETBMAP		_IOWR('X', 38, struct getbmap)
diff --git a/fs/xfs/libxfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c
index 54deb2d12ac6..22297f9b0fd5 100644
--- a/fs/xfs/libxfs/xfs_ialloc.c
+++ b/fs/xfs/libxfs/xfs_ialloc.c
@@ -38,6 +38,7 @@
 #include "xfs_icreate_item.h"
 #include "xfs_icache.h"
 #include "xfs_trace.h"
+#include "xfs_log.h"
 
 
 /*
@@ -2402,8 +2403,8 @@ xfs_ialloc_compute_maxlevels(
 
 	maxleafents = (1LL << XFS_INO_AGINO_BITS(mp)) >>
 		XFS_INODES_PER_CHUNK_LOG;
-	minleafrecs = mp->m_alloc_mnr[0];
-	minnoderecs = mp->m_alloc_mnr[1];
+	minleafrecs = mp->m_inobt_mnr[0];
+	minnoderecs = mp->m_inobt_mnr[1];
 	maxblocks = (maxleafents + minleafrecs - 1) / minleafrecs;
 	for (level = 1; maxblocks > 1; level++)
 		maxblocks = (maxblocks + minnoderecs - 1) / minnoderecs;
@@ -2500,9 +2501,14 @@ xfs_agi_verify(
 	struct xfs_mount *mp = bp->b_target->bt_mount;
 	struct xfs_agi	*agi = XFS_BUF_TO_AGI(bp);
 
-	if (xfs_sb_version_hascrc(&mp->m_sb) &&
-	    !uuid_equal(&agi->agi_uuid, &mp->m_sb.sb_meta_uuid))
+	if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		if (!uuid_equal(&agi->agi_uuid, &mp->m_sb.sb_meta_uuid))
+			return false;
+		if (!xfs_log_check_lsn(mp,
+				be64_to_cpu(XFS_BUF_TO_AGI(bp)->agi_lsn)))
 			return false;
+	}
+
 	/*
 	 * Validate the magic number of the agi block.
 	 */
@@ -2566,6 +2572,7 @@ xfs_agi_write_verify(
 }
 
 const struct xfs_buf_ops xfs_agi_buf_ops = {
+	.name = "xfs_agi",
 	.verify_read = xfs_agi_read_verify,
 	.verify_write = xfs_agi_write_verify,
 };
diff --git a/fs/xfs/libxfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c
index f39b285beb19..89c21d771e35 100644
--- a/fs/xfs/libxfs/xfs_ialloc_btree.c
+++ b/fs/xfs/libxfs/xfs_ialloc_btree.c
@@ -125,16 +125,8 @@ xfs_inobt_free_block(
 	struct xfs_btree_cur	*cur,
 	struct xfs_buf		*bp)
 {
-	xfs_fsblock_t		fsbno;
-	int			error;
-
-	fsbno = XFS_DADDR_TO_FSB(cur->bc_mp, XFS_BUF_ADDR(bp));
-	error = xfs_free_extent(cur->bc_tp, fsbno, 1);
-	if (error)
-		return error;
-
-	xfs_trans_binval(cur->bc_tp, bp);
-	return error;
+	return xfs_free_extent(cur->bc_tp,
+			XFS_DADDR_TO_FSB(cur->bc_mp, XFS_BUF_ADDR(bp)), 1);
 }
 
 STATIC int
@@ -221,7 +213,6 @@ xfs_inobt_verify(
 {
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 	struct xfs_btree_block	*block = XFS_BUF_TO_BLOCK(bp);
-	struct xfs_perag	*pag = bp->b_pag;
 	unsigned int		level;
 
 	/*
@@ -237,14 +228,7 @@ xfs_inobt_verify(
 	switch (block->bb_magic) {
 	case cpu_to_be32(XFS_IBT_CRC_MAGIC):
 	case cpu_to_be32(XFS_FIBT_CRC_MAGIC):
-		if (!xfs_sb_version_hascrc(&mp->m_sb))
-			return false;
-		if (!uuid_equal(&block->bb_u.s.bb_uuid, &mp->m_sb.sb_meta_uuid))
-			return false;
-		if (block->bb_u.s.bb_blkno != cpu_to_be64(bp->b_bn))
-			return false;
-		if (pag &&
-		    be32_to_cpu(block->bb_u.s.bb_owner) != pag->pag_agno)
+		if (!xfs_btree_sblock_v5hdr_verify(bp))
 			return false;
 		/* fall through */
 	case cpu_to_be32(XFS_IBT_MAGIC):
@@ -254,24 +238,12 @@ xfs_inobt_verify(
 		return 0;
 	}
 
-	/* numrecs and level verification */
+	/* level verification */
 	level = be16_to_cpu(block->bb_level);
 	if (level >= mp->m_in_maxlevels)
 		return false;
-	if (be16_to_cpu(block->bb_numrecs) > mp->m_inobt_mxr[level != 0])
-		return false;
-
-	/* sibling pointer verification */
-	if (!block->bb_u.s.bb_leftsib ||
-	    (be32_to_cpu(block->bb_u.s.bb_leftsib) >= mp->m_sb.sb_agblocks &&
-	     block->bb_u.s.bb_leftsib != cpu_to_be32(NULLAGBLOCK)))
-		return false;
-	if (!block->bb_u.s.bb_rightsib ||
-	    (be32_to_cpu(block->bb_u.s.bb_rightsib) >= mp->m_sb.sb_agblocks &&
-	     block->bb_u.s.bb_rightsib != cpu_to_be32(NULLAGBLOCK)))
-		return false;
 
-	return true;
+	return xfs_btree_sblock_verify(bp, mp->m_inobt_mxr[level != 0]);
 }
 
 static void
@@ -304,6 +276,7 @@ xfs_inobt_write_verify(
 }
 
 const struct xfs_buf_ops xfs_inobt_buf_ops = {
+	.name = "xfs_inobt",
 	.verify_read = xfs_inobt_read_verify,
 	.verify_write = xfs_inobt_write_verify,
 };
diff --git a/fs/xfs/libxfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c
index 268c00f4f83a..9d9559eb2835 100644
--- a/fs/xfs/libxfs/xfs_inode_buf.c
+++ b/fs/xfs/libxfs/xfs_inode_buf.c
@@ -62,11 +62,14 @@ xfs_inobp_check(
  * has not had the inode cores stamped into it. Hence for readahead, the buffer
  * may be potentially invalid.
  *
- * If the readahead buffer is invalid, we don't want to mark it with an error,
- * but we do want to clear the DONE status of the buffer so that a followup read
- * will re-read it from disk. This will ensure that we don't get an unnecessary
- * warnings during log recovery and we don't get unnecssary panics on debug
- * kernels.
+ * If the readahead buffer is invalid, we need to mark it with an error and
+ * clear the DONE status of the buffer so that a followup read will re-read it
+ * from disk. We don't report the error otherwise to avoid warnings during log
+ * recovery and we don't get unnecssary panics on debug kernels. We use EIO here
+ * because all we want to do is say readahead failed; there is no-one to report
+ * the error to, so this will distinguish it from a non-ra verifier failure.
+ * Changes to this readahead error behavour also need to be reflected in
+ * xfs_dquot_buf_readahead_verify().
  */
 static void
 xfs_inode_buf_verify(
@@ -93,6 +96,7 @@ xfs_inode_buf_verify(
 						XFS_RANDOM_ITOBP_INOTOBP))) {
 			if (readahead) {
 				bp->b_flags &= ~XBF_DONE;
+				xfs_buf_ioerror(bp, -EIO);
 				return;
 			}
 
@@ -132,11 +136,13 @@ xfs_inode_buf_write_verify(
 }
 
 const struct xfs_buf_ops xfs_inode_buf_ops = {
+	.name = "xfs_inode",
 	.verify_read = xfs_inode_buf_read_verify,
 	.verify_write = xfs_inode_buf_write_verify,
 };
 
 const struct xfs_buf_ops xfs_inode_buf_ra_ops = {
+	.name = "xxfs_inode_ra",
 	.verify_read = xfs_inode_buf_readahead_verify,
 	.verify_write = xfs_inode_buf_write_verify,
 };
@@ -189,28 +195,50 @@ xfs_imap_to_bp(
 }
 
 void
-xfs_dinode_from_disk(
-	xfs_icdinode_t		*to,
-	xfs_dinode_t		*from)
+xfs_inode_from_disk(
+	struct xfs_inode	*ip,
+	struct xfs_dinode	*from)
 {
-	to->di_magic = be16_to_cpu(from->di_magic);
-	to->di_mode = be16_to_cpu(from->di_mode);
-	to->di_version = from ->di_version;
+	struct xfs_icdinode	*to = &ip->i_d;
+	struct inode		*inode = VFS_I(ip);
+
+
+	/*
+	 * Convert v1 inodes immediately to v2 inode format as this is the
+	 * minimum inode version format we support in the rest of the code.
+	 */
+	to->di_version = from->di_version;
+	if (to->di_version == 1) {
+		set_nlink(inode, be16_to_cpu(from->di_onlink));
+		to->di_projid_lo = 0;
+		to->di_projid_hi = 0;
+		to->di_version = 2;
+	} else {
+		set_nlink(inode, be32_to_cpu(from->di_nlink));
+		to->di_projid_lo = be16_to_cpu(from->di_projid_lo);
+		to->di_projid_hi = be16_to_cpu(from->di_projid_hi);
+	}
+
 	to->di_format = from->di_format;
-	to->di_onlink = be16_to_cpu(from->di_onlink);
 	to->di_uid = be32_to_cpu(from->di_uid);
 	to->di_gid = be32_to_cpu(from->di_gid);
-	to->di_nlink = be32_to_cpu(from->di_nlink);
-	to->di_projid_lo = be16_to_cpu(from->di_projid_lo);
-	to->di_projid_hi = be16_to_cpu(from->di_projid_hi);
-	memcpy(to->di_pad, from->di_pad, sizeof(to->di_pad));
 	to->di_flushiter = be16_to_cpu(from->di_flushiter);
-	to->di_atime.t_sec = be32_to_cpu(from->di_atime.t_sec);
-	to->di_atime.t_nsec = be32_to_cpu(from->di_atime.t_nsec);
-	to->di_mtime.t_sec = be32_to_cpu(from->di_mtime.t_sec);
-	to->di_mtime.t_nsec = be32_to_cpu(from->di_mtime.t_nsec);
-	to->di_ctime.t_sec = be32_to_cpu(from->di_ctime.t_sec);
-	to->di_ctime.t_nsec = be32_to_cpu(from->di_ctime.t_nsec);
+
+	/*
+	 * Time is signed, so need to convert to signed 32 bit before
+	 * storing in inode timestamp which may be 64 bit. Otherwise
+	 * a time before epoch is converted to a time long after epoch
+	 * on 64 bit systems.
+	 */
+	inode->i_atime.tv_sec = (int)be32_to_cpu(from->di_atime.t_sec);
+	inode->i_atime.tv_nsec = (int)be32_to_cpu(from->di_atime.t_nsec);
+	inode->i_mtime.tv_sec = (int)be32_to_cpu(from->di_mtime.t_sec);
+	inode->i_mtime.tv_nsec = (int)be32_to_cpu(from->di_mtime.t_nsec);
+	inode->i_ctime.tv_sec = (int)be32_to_cpu(from->di_ctime.t_sec);
+	inode->i_ctime.tv_nsec = (int)be32_to_cpu(from->di_ctime.t_nsec);
+	inode->i_generation = be32_to_cpu(from->di_gen);
+	inode->i_mode = be16_to_cpu(from->di_mode);
+
 	to->di_size = be64_to_cpu(from->di_size);
 	to->di_nblocks = be64_to_cpu(from->di_nblocks);
 	to->di_extsize = be32_to_cpu(from->di_extsize);
@@ -221,42 +249,96 @@ xfs_dinode_from_disk(
 	to->di_dmevmask	= be32_to_cpu(from->di_dmevmask);
 	to->di_dmstate	= be16_to_cpu(from->di_dmstate);
 	to->di_flags	= be16_to_cpu(from->di_flags);
-	to->di_gen	= be32_to_cpu(from->di_gen);
 
 	if (to->di_version == 3) {
-		to->di_changecount = be64_to_cpu(from->di_changecount);
+		inode->i_version = be64_to_cpu(from->di_changecount);
 		to->di_crtime.t_sec = be32_to_cpu(from->di_crtime.t_sec);
 		to->di_crtime.t_nsec = be32_to_cpu(from->di_crtime.t_nsec);
 		to->di_flags2 = be64_to_cpu(from->di_flags2);
-		to->di_ino = be64_to_cpu(from->di_ino);
-		to->di_lsn = be64_to_cpu(from->di_lsn);
-		memcpy(to->di_pad2, from->di_pad2, sizeof(to->di_pad2));
-		uuid_copy(&to->di_uuid, &from->di_uuid);
 	}
 }
 
 void
-xfs_dinode_to_disk(
-	xfs_dinode_t		*to,
-	xfs_icdinode_t		*from)
+xfs_inode_to_disk(
+	struct xfs_inode	*ip,
+	struct xfs_dinode	*to,
+	xfs_lsn_t		lsn)
+{
+	struct xfs_icdinode	*from = &ip->i_d;
+	struct inode		*inode = VFS_I(ip);
+
+	to->di_magic = cpu_to_be16(XFS_DINODE_MAGIC);
+	to->di_onlink = 0;
+
+	to->di_version = from->di_version;
+	to->di_format = from->di_format;
+	to->di_uid = cpu_to_be32(from->di_uid);
+	to->di_gid = cpu_to_be32(from->di_gid);
+	to->di_projid_lo = cpu_to_be16(from->di_projid_lo);
+	to->di_projid_hi = cpu_to_be16(from->di_projid_hi);
+
+	memset(to->di_pad, 0, sizeof(to->di_pad));
+	to->di_atime.t_sec = cpu_to_be32(inode->i_atime.tv_sec);
+	to->di_atime.t_nsec = cpu_to_be32(inode->i_atime.tv_nsec);
+	to->di_mtime.t_sec = cpu_to_be32(inode->i_mtime.tv_sec);
+	to->di_mtime.t_nsec = cpu_to_be32(inode->i_mtime.tv_nsec);
+	to->di_ctime.t_sec = cpu_to_be32(inode->i_ctime.tv_sec);
+	to->di_ctime.t_nsec = cpu_to_be32(inode->i_ctime.tv_nsec);
+	to->di_nlink = cpu_to_be32(inode->i_nlink);
+	to->di_gen = cpu_to_be32(inode->i_generation);
+	to->di_mode = cpu_to_be16(inode->i_mode);
+
+	to->di_size = cpu_to_be64(from->di_size);
+	to->di_nblocks = cpu_to_be64(from->di_nblocks);
+	to->di_extsize = cpu_to_be32(from->di_extsize);
+	to->di_nextents = cpu_to_be32(from->di_nextents);
+	to->di_anextents = cpu_to_be16(from->di_anextents);
+	to->di_forkoff = from->di_forkoff;
+	to->di_aformat = from->di_aformat;
+	to->di_dmevmask = cpu_to_be32(from->di_dmevmask);
+	to->di_dmstate = cpu_to_be16(from->di_dmstate);
+	to->di_flags = cpu_to_be16(from->di_flags);
+
+	if (from->di_version == 3) {
+		to->di_changecount = cpu_to_be64(inode->i_version);
+		to->di_crtime.t_sec = cpu_to_be32(from->di_crtime.t_sec);
+		to->di_crtime.t_nsec = cpu_to_be32(from->di_crtime.t_nsec);
+		to->di_flags2 = cpu_to_be64(from->di_flags2);
+
+		to->di_ino = cpu_to_be64(ip->i_ino);
+		to->di_lsn = cpu_to_be64(lsn);
+		memset(to->di_pad2, 0, sizeof(to->di_pad2));
+		uuid_copy(&to->di_uuid, &ip->i_mount->m_sb.sb_meta_uuid);
+		to->di_flushiter = 0;
+	} else {
+		to->di_flushiter = cpu_to_be16(from->di_flushiter);
+	}
+}
+
+void
+xfs_log_dinode_to_disk(
+	struct xfs_log_dinode	*from,
+	struct xfs_dinode	*to)
 {
 	to->di_magic = cpu_to_be16(from->di_magic);
 	to->di_mode = cpu_to_be16(from->di_mode);
-	to->di_version = from ->di_version;
+	to->di_version = from->di_version;
 	to->di_format = from->di_format;
-	to->di_onlink = cpu_to_be16(from->di_onlink);
+	to->di_onlink = 0;
 	to->di_uid = cpu_to_be32(from->di_uid);
 	to->di_gid = cpu_to_be32(from->di_gid);
 	to->di_nlink = cpu_to_be32(from->di_nlink);
 	to->di_projid_lo = cpu_to_be16(from->di_projid_lo);
 	to->di_projid_hi = cpu_to_be16(from->di_projid_hi);
 	memcpy(to->di_pad, from->di_pad, sizeof(to->di_pad));
+
 	to->di_atime.t_sec = cpu_to_be32(from->di_atime.t_sec);
 	to->di_atime.t_nsec = cpu_to_be32(from->di_atime.t_nsec);
 	to->di_mtime.t_sec = cpu_to_be32(from->di_mtime.t_sec);
 	to->di_mtime.t_nsec = cpu_to_be32(from->di_mtime.t_nsec);
 	to->di_ctime.t_sec = cpu_to_be32(from->di_ctime.t_sec);
 	to->di_ctime.t_nsec = cpu_to_be32(from->di_ctime.t_nsec);
+
 	to->di_size = cpu_to_be64(from->di_size);
 	to->di_nblocks = cpu_to_be64(from->di_nblocks);
 	to->di_extsize = cpu_to_be32(from->di_extsize);
@@ -361,13 +443,10 @@ xfs_iread(
 	    !(mp->m_flags & XFS_MOUNT_IKEEP)) {
 		/* initialise the on-disk inode core */
 		memset(&ip->i_d, 0, sizeof(ip->i_d));
-		ip->i_d.di_magic = XFS_DINODE_MAGIC;
-		ip->i_d.di_gen = prandom_u32();
-		if (xfs_sb_version_hascrc(&mp->m_sb)) {
+		VFS_I(ip)->i_generation = prandom_u32();
+		if (xfs_sb_version_hascrc(&mp->m_sb))
 			ip->i_d.di_version = 3;
-			ip->i_d.di_ino = ip->i_ino;
-			uuid_copy(&ip->i_d.di_uuid, &mp->m_sb.sb_meta_uuid);
-		} else
+		else
 			ip->i_d.di_version = 2;
 		return 0;
 	}
@@ -397,7 +476,7 @@ xfs_iread(
 	 * Otherwise, just get the truly permanent information.
 	 */
 	if (dip->di_mode) {
-		xfs_dinode_from_disk(&ip->i_d, dip);
+		xfs_inode_from_disk(ip, dip);
 		error = xfs_iformat_fork(ip, dip);
 		if (error)  {
 #ifdef DEBUG
@@ -411,16 +490,10 @@ xfs_iread(
 		 * Partial initialisation of the in-core inode. Just the bits
 		 * that xfs_ialloc won't overwrite or relies on being correct.
 		 */
-		ip->i_d.di_magic = be16_to_cpu(dip->di_magic);
 		ip->i_d.di_version = dip->di_version;
-		ip->i_d.di_gen = be32_to_cpu(dip->di_gen);
+		VFS_I(ip)->i_generation = be32_to_cpu(dip->di_gen);
 		ip->i_d.di_flushiter = be16_to_cpu(dip->di_flushiter);
 
-		if (dip->di_version == 3) {
-			ip->i_d.di_ino = be64_to_cpu(dip->di_ino);
-			uuid_copy(&ip->i_d.di_uuid, &dip->di_uuid);
-		}
-
 		/*
 		 * Make sure to pull in the mode here as well in
 		 * case the inode is released without being used.
@@ -428,25 +501,10 @@ xfs_iread(
 		 * the inode is already free and not try to mess
 		 * with the uninitialized part of it.
 		 */
-		ip->i_d.di_mode = 0;
-	}
-
-	/*
-	 * Automatically convert version 1 inode formats in memory to version 2
-	 * inode format. If the inode is modified, it will get logged and
-	 * rewritten as a version 2 inode. We can do this because we set the
-	 * superblock feature bit for v2 inodes unconditionally during mount
-	 * and it means the reast of the code can assume the inode version is 2
-	 * or higher.
-	 */
-	if (ip->i_d.di_version == 1) {
-		ip->i_d.di_version = 2;
-		memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad));
-		ip->i_d.di_nlink = ip->i_d.di_onlink;
-		ip->i_d.di_onlink = 0;
-		xfs_set_projid(ip, 0);
+		VFS_I(ip)->i_mode = 0;
 	}
 
+	ASSERT(ip->i_d.di_version >= 2);
 	ip->i_delayed_blks = 0;
 
 	/*
diff --git a/fs/xfs/libxfs/xfs_inode_buf.h b/fs/xfs/libxfs/xfs_inode_buf.h
index 9308c47f2a52..7c4dd321b215 100644
--- a/fs/xfs/libxfs/xfs_inode_buf.h
+++ b/fs/xfs/libxfs/xfs_inode_buf.h
@@ -20,7 +20,36 @@
 
 struct xfs_inode;
 struct xfs_dinode;
-struct xfs_icdinode;
+
+/*
+ * In memory representation of the XFS inode. This is held in the in-core struct
+ * xfs_inode and represents the current on disk values but the structure is not
+ * in on-disk format.  That is, this structure is always translated to on-disk
+ * format specific structures at the appropriate time.
+ */
+struct xfs_icdinode {
+	__int8_t	di_version;	/* inode version */
+	__int8_t	di_format;	/* format of di_c data */
+	__uint16_t	di_flushiter;	/* incremented on flush */
+	__uint32_t	di_uid;		/* owner's user id */
+	__uint32_t	di_gid;		/* owner's group id */
+	__uint16_t	di_projid_lo;	/* lower part of owner's project id */
+	__uint16_t	di_projid_hi;	/* higher part of owner's project id */
+	xfs_fsize_t	di_size;	/* number of bytes in file */
+	xfs_rfsblock_t	di_nblocks;	/* # of direct & btree blocks used */
+	xfs_extlen_t	di_extsize;	/* basic/minimum extent size for file */
+	xfs_extnum_t	di_nextents;	/* number of extents in data fork */
+	xfs_aextnum_t	di_anextents;	/* number of extents in attribute fork*/
+	__uint8_t	di_forkoff;	/* attr fork offs, <<3 for 64b align */
+	__int8_t	di_aformat;	/* format of attr fork's data */
+	__uint32_t	di_dmevmask;	/* DMIG event mask */
+	__uint16_t	di_dmstate;	/* DMIG state info */
+	__uint16_t	di_flags;	/* random flags, XFS_DIFLAG_... */
+
+	__uint64_t	di_flags2;	/* more random flags */
+
+	xfs_ictimestamp_t di_crtime;	/* time created */
+};
 
 /*
  * Inode location information.  Stored in the inode and passed to
@@ -38,8 +67,11 @@ int	xfs_imap_to_bp(struct xfs_mount *, struct xfs_trans *,
 int	xfs_iread(struct xfs_mount *, struct xfs_trans *,
 		  struct xfs_inode *, uint);
 void	xfs_dinode_calc_crc(struct xfs_mount *, struct xfs_dinode *);
-void	xfs_dinode_to_disk(struct xfs_dinode *to, struct xfs_icdinode *from);
-void	xfs_dinode_from_disk(struct xfs_icdinode *to, struct xfs_dinode *from);
+void	xfs_inode_to_disk(struct xfs_inode *ip, struct xfs_dinode *to,
+			  xfs_lsn_t lsn);
+void	xfs_inode_from_disk(struct xfs_inode *ip, struct xfs_dinode *from);
+void	xfs_log_dinode_to_disk(struct xfs_log_dinode *from,
+			       struct xfs_dinode *to);
 
 #if defined(DEBUG)
 void	xfs_inobp_check(struct xfs_mount *, struct xfs_buf *);
diff --git a/fs/xfs/libxfs/xfs_inode_fork.c b/fs/xfs/libxfs/xfs_inode_fork.c
index 0defbd02f62d..11faf7df14c8 100644
--- a/fs/xfs/libxfs/xfs_inode_fork.c
+++ b/fs/xfs/libxfs/xfs_inode_fork.c
@@ -31,6 +31,7 @@
 #include "xfs_error.h"
 #include "xfs_trace.h"
 #include "xfs_attr_sf.h"
+#include "xfs_da_format.h"
 
 kmem_zone_t *xfs_ifork_zone;
 
@@ -120,7 +121,7 @@ xfs_iformat_fork(
 		return -EFSCORRUPTED;
 	}
 
-	switch (ip->i_d.di_mode & S_IFMT) {
+	switch (VFS_I(ip)->i_mode & S_IFMT) {
 	case S_IFIFO:
 	case S_IFCHR:
 	case S_IFBLK:
diff --git a/fs/xfs/libxfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h
index 265314690415..d54a8018b079 100644
--- a/fs/xfs/libxfs/xfs_log_format.h
+++ b/fs/xfs/libxfs/xfs_log_format.h
@@ -290,6 +290,7 @@ typedef struct xfs_inode_log_format_64 {
 	__int32_t		ilf_boffset;	/* off of inode in buffer */
 } xfs_inode_log_format_64_t;
 
+
 /*
  * Flags for xfs_trans_log_inode flags field.
  */
@@ -360,15 +361,15 @@ typedef struct xfs_ictimestamp {
 } xfs_ictimestamp_t;
 
 /*
- * NOTE:  This structure must be kept identical to struct xfs_dinode
- *	  except for the endianness annotations.
+ * Define the format of the inode core that is logged. This structure must be
+ * kept identical to struct xfs_dinode except for the endianness annotations.
  */
-typedef struct xfs_icdinode {
+struct xfs_log_dinode {
 	__uint16_t	di_magic;	/* inode magic # = XFS_DINODE_MAGIC */
 	__uint16_t	di_mode;	/* mode and type of file */
 	__int8_t	di_version;	/* inode version */
 	__int8_t	di_format;	/* format of di_c data */
-	__uint16_t	di_onlink;	/* old number of links to file */
+	__uint8_t	di_pad3[2];	/* unused in v2/3 inodes */
 	__uint32_t	di_uid;		/* owner's user id */
 	__uint32_t	di_gid;		/* owner's group id */
 	__uint32_t	di_nlink;	/* number of links to file */
@@ -407,13 +408,13 @@ typedef struct xfs_icdinode {
 	uuid_t		di_uuid;	/* UUID of the filesystem */
 
 	/* structure must be padded to 64 bit alignment */
-} xfs_icdinode_t;
+};
 
-static inline uint xfs_icdinode_size(int version)
+static inline uint xfs_log_dinode_size(int version)
 {
 	if (version == 3)
-		return sizeof(struct xfs_icdinode);
-	return offsetof(struct xfs_icdinode, di_next_unlinked);
+		return sizeof(struct xfs_log_dinode);
+	return offsetof(struct xfs_log_dinode, di_next_unlinked);
 }
 
 /*
@@ -495,6 +496,8 @@ enum xfs_blft {
 	XFS_BLFT_ATTR_LEAF_BUF,
 	XFS_BLFT_ATTR_RMT_BUF,
 	XFS_BLFT_SB_BUF,
+	XFS_BLFT_RTBITMAP_BUF,
+	XFS_BLFT_RTSUMMARY_BUF,
 	XFS_BLFT_MAX_BUF = (1 << XFS_BLFT_BITS),
 };
 
diff --git a/fs/xfs/libxfs/xfs_log_recover.h b/fs/xfs/libxfs/xfs_log_recover.h
index 1c55ccbb379d..8e385f91d660 100644
--- a/fs/xfs/libxfs/xfs_log_recover.h
+++ b/fs/xfs/libxfs/xfs_log_recover.h
@@ -60,6 +60,7 @@ typedef struct xlog_recover {
  */
 #define	XLOG_BC_TABLE_SIZE	64
 
+#define	XLOG_RECOVER_CRCPASS	0
 #define	XLOG_RECOVER_PASS1	1
 #define	XLOG_RECOVER_PASS2	2
 
diff --git a/fs/xfs/libxfs/xfs_quota_defs.h b/fs/xfs/libxfs/xfs_quota_defs.h
index 1b0a08379759..8eed51275bb3 100644
--- a/fs/xfs/libxfs/xfs_quota_defs.h
+++ b/fs/xfs/libxfs/xfs_quota_defs.h
@@ -37,7 +37,7 @@ typedef __uint16_t	xfs_qwarncnt_t;
 #define XFS_DQ_PROJ		0x0002		/* project quota */
 #define XFS_DQ_GROUP		0x0004		/* a group quota */
 #define XFS_DQ_DIRTY		0x0008		/* dquot is dirty */
-#define XFS_DQ_FREEING		0x0010		/* dquot is beeing torn down */
+#define XFS_DQ_FREEING		0x0010		/* dquot is being torn down */
 
 #define XFS_DQ_ALLTYPES		(XFS_DQ_USER|XFS_DQ_PROJ|XFS_DQ_GROUP)
 
@@ -116,6 +116,7 @@ typedef __uint16_t	xfs_qwarncnt_t;
 #define XFS_QMOPT_DQREPAIR	0x0001000 /* repair dquot if damaged */
 #define XFS_QMOPT_GQUOTA	0x0002000 /* group dquot requested */
 #define XFS_QMOPT_ENOSPC	0x0004000 /* enospc instead of edquot (prj) */
+#define XFS_QMOPT_DQNEXT	0x0008000 /* return next dquot >= this ID */
 
 /*
  * flags to xfs_trans_mod_dquot to indicate which field needs to be
@@ -153,7 +154,7 @@ typedef __uint16_t	xfs_qwarncnt_t;
 #define XFS_QMOPT_RESBLK_MASK	(XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_RES_RTBLKS)
 
 extern int xfs_dqcheck(struct xfs_mount *mp, xfs_disk_dquot_t *ddq,
-		       xfs_dqid_t id, uint type, uint flags, char *str);
+		       xfs_dqid_t id, uint type, uint flags, const char *str);
 extern int xfs_calc_dquots_per_chunk(unsigned int nbblks);
 
 #endif	/* __XFS_QUOTA_H__ */
diff --git a/fs/xfs/libxfs/xfs_rtbitmap.c b/fs/xfs/libxfs/xfs_rtbitmap.c
index 9b59ffa1fc19..951c044e24e4 100644
--- a/fs/xfs/libxfs/xfs_rtbitmap.c
+++ b/fs/xfs/libxfs/xfs_rtbitmap.c
@@ -42,6 +42,31 @@
  */
 
 /*
+ * Real time buffers need verifiers to avoid runtime warnings during IO.
+ * We don't have anything to verify, however, so these are just dummy
+ * operations.
+ */
+static void
+xfs_rtbuf_verify_read(
+	struct xfs_buf	*bp)
+{
+	return;
+}
+
+static void
+xfs_rtbuf_verify_write(
+	struct xfs_buf	*bp)
+{
+	return;
+}
+
+const struct xfs_buf_ops xfs_rtbuf_ops = {
+	.name = "rtbuf",
+	.verify_read = xfs_rtbuf_verify_read,
+	.verify_write = xfs_rtbuf_verify_write,
+};
+
+/*
  * Get a buffer for the bitmap or summary file block specified.
  * The buffer is returned read and locked.
  */
@@ -68,9 +93,12 @@ xfs_rtbuf_get(
 	ASSERT(map.br_startblock != NULLFSBLOCK);
 	error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
 				   XFS_FSB_TO_DADDR(mp, map.br_startblock),
-				   mp->m_bsize, 0, &bp, NULL);
+				   mp->m_bsize, 0, &bp, &xfs_rtbuf_ops);
 	if (error)
 		return error;
+
+	xfs_trans_buf_set_type(tp, bp, issum ? XFS_BLFT_RTSUMMARY_BUF
+					     : XFS_BLFT_RTBITMAP_BUF);
 	*bpp = bp;
 	return 0;
 }
@@ -983,7 +1011,7 @@ xfs_rtfree_extent(
 	    mp->m_sb.sb_rextents) {
 		if (!(mp->m_rbmip->i_d.di_flags & XFS_DIFLAG_NEWRTBM))
 			mp->m_rbmip->i_d.di_flags |= XFS_DIFLAG_NEWRTBM;
-		*(__uint64_t *)&mp->m_rbmip->i_d.di_atime = 0;
+		*(__uint64_t *)&VFS_I(mp->m_rbmip)->i_atime = 0;
 		xfs_trans_log_inode(tp, mp->m_rbmip, XFS_ILOG_CORE);
 	}
 	return 0;
diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c
index 47425140f343..8a53eaa349f4 100644
--- a/fs/xfs/libxfs/xfs_sb.c
+++ b/fs/xfs/libxfs/xfs_sb.c
@@ -35,6 +35,7 @@
 #include "xfs_bmap_btree.h"
 #include "xfs_alloc_btree.h"
 #include "xfs_ialloc_btree.h"
+#include "xfs_log.h"
 
 /*
  * Physical superblock buffer manipulations. Shared with libxfs in userspace.
@@ -163,6 +164,15 @@ xfs_mount_validate_sb(
 "Filesystem can not be safely mounted by this kernel.");
 			return -EINVAL;
 		}
+	} else if (xfs_sb_version_hascrc(sbp)) {
+		/*
+		 * We can't read verify the sb LSN because the read verifier is
+		 * called before the log is allocated and processed. We know the
+		 * log is set up before write verifier (!check_version) calls,
+		 * so just check it here.
+		 */
+		if (!xfs_log_check_lsn(mp, sbp->sb_lsn))
+			return -EFSCORRUPTED;
 	}
 
 	if (xfs_sb_version_has_pquotino(sbp)) {
@@ -669,11 +679,13 @@ xfs_sb_write_verify(
 }
 
 const struct xfs_buf_ops xfs_sb_buf_ops = {
+	.name = "xfs_sb",
 	.verify_read = xfs_sb_read_verify,
 	.verify_write = xfs_sb_write_verify,
 };
 
 const struct xfs_buf_ops xfs_sb_quiet_buf_ops = {
+	.name = "xfs_sb_quiet",
 	.verify_read = xfs_sb_quiet_read_verify,
 	.verify_write = xfs_sb_write_verify,
 };
diff --git a/fs/xfs/libxfs/xfs_sb.h b/fs/xfs/libxfs/xfs_sb.h
index b25bb9a343f3..961e6475a309 100644
--- a/fs/xfs/libxfs/xfs_sb.h
+++ b/fs/xfs/libxfs/xfs_sb.h
@@ -27,7 +27,6 @@ extern struct xfs_perag *xfs_perag_get_tag(struct xfs_mount *, xfs_agnumber_t,
 extern void	xfs_perag_put(struct xfs_perag *pag);
 extern int	xfs_initialize_perag_data(struct xfs_mount *, xfs_agnumber_t);
 
-extern void	xfs_sb_calc_crc(struct xfs_buf *bp);
 extern void	xfs_log_sb(struct xfs_trans *tp);
 extern int	xfs_sync_sb(struct xfs_mount *mp, bool wait);
 extern void	xfs_sb_mount_common(struct xfs_mount *mp, struct xfs_sb *sbp);
diff --git a/fs/xfs/libxfs/xfs_shared.h b/fs/xfs/libxfs/xfs_shared.h
index 5be529707903..81ac870834da 100644
--- a/fs/xfs/libxfs/xfs_shared.h
+++ b/fs/xfs/libxfs/xfs_shared.h
@@ -49,9 +49,11 @@ extern const struct xfs_buf_ops xfs_inobt_buf_ops;
 extern const struct xfs_buf_ops xfs_inode_buf_ops;
 extern const struct xfs_buf_ops xfs_inode_buf_ra_ops;
 extern const struct xfs_buf_ops xfs_dquot_buf_ops;
+extern const struct xfs_buf_ops xfs_dquot_buf_ra_ops;
 extern const struct xfs_buf_ops xfs_sb_buf_ops;
 extern const struct xfs_buf_ops xfs_sb_quiet_buf_ops;
 extern const struct xfs_buf_ops xfs_symlink_buf_ops;
+extern const struct xfs_buf_ops xfs_rtbuf_ops;
 
 /*
  * Transaction types.  Used to distinguish types of buffers. These never reach
diff --git a/fs/xfs/libxfs/xfs_symlink_remote.c b/fs/xfs/libxfs/xfs_symlink_remote.c
index 8f8af05b3f13..2e2c6716b623 100644
--- a/fs/xfs/libxfs/xfs_symlink_remote.c
+++ b/fs/xfs/libxfs/xfs_symlink_remote.c
@@ -31,6 +31,7 @@
 #include "xfs_cksum.h"
 #include "xfs_trans.h"
 #include "xfs_buf_item.h"
+#include "xfs_log.h"
 
 
 /*
@@ -60,6 +61,7 @@ xfs_symlink_hdr_set(
 	if (!xfs_sb_version_hascrc(&mp->m_sb))
 		return 0;
 
+	memset(dsl, 0, sizeof(struct xfs_dsymlink_hdr));
 	dsl->sl_magic = cpu_to_be32(XFS_SYMLINK_MAGIC);
 	dsl->sl_offset = cpu_to_be32(offset);
 	dsl->sl_bytes = cpu_to_be32(size);
@@ -116,6 +118,8 @@ xfs_symlink_verify(
 		return false;
 	if (dsl->sl_owner == 0)
 		return false;
+	if (!xfs_log_check_lsn(mp, be64_to_cpu(dsl->sl_lsn)))
+		return false;
 
 	return true;
 }
@@ -164,6 +168,7 @@ xfs_symlink_write_verify(
 }
 
 const struct xfs_buf_ops xfs_symlink_buf_ops = {
+	.name = "xfs_symlink",
 	.verify_read = xfs_symlink_read_verify,
 	.verify_write = xfs_symlink_write_verify,
 };
@@ -183,6 +188,7 @@ xfs_symlink_local_to_remote(
 	if (!xfs_sb_version_hascrc(&mp->m_sb)) {
 		bp->b_ops = NULL;
 		memcpy(bp->b_addr, ifp->if_u1.if_data, ifp->if_bytes);
+		xfs_trans_log_buf(tp, bp, 0, ifp->if_bytes - 1);
 		return;
 	}
 
@@ -198,4 +204,6 @@ xfs_symlink_local_to_remote(
 	buf = bp->b_addr;
 	buf += xfs_symlink_hdr_set(mp, ip->i_ino, 0, ifp->if_bytes, bp);
 	memcpy(buf, ifp->if_u1.if_data, ifp->if_bytes);
+	xfs_trans_log_buf(tp, bp, 0, sizeof(struct xfs_dsymlink_hdr) +
+					ifp->if_bytes - 1);
 }
diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c
index 4b641676f258..2d5df1f23bbc 100644
--- a/fs/xfs/xfs_acl.c
+++ b/fs/xfs/xfs_acl.c
@@ -37,16 +37,19 @@
 
 STATIC struct posix_acl *
 xfs_acl_from_disk(
-	struct xfs_acl	*aclp,
-	int		max_entries)
+	const struct xfs_acl	*aclp,
+	int			len,
+	int			max_entries)
 {
 	struct posix_acl_entry *acl_e;
 	struct posix_acl *acl;
-	struct xfs_acl_entry *ace;
+	const struct xfs_acl_entry *ace;
 	unsigned int count, i;
 
+	if (len < sizeof(*aclp))
+		return ERR_PTR(-EFSCORRUPTED);
 	count = be32_to_cpu(aclp->acl_cnt);
-	if (count > max_entries)
+	if (count > max_entries || XFS_ACL_SIZE(count) != len)
 		return ERR_PTR(-EFSCORRUPTED);
 
 	acl = posix_acl_alloc(count, GFP_KERNEL);
@@ -160,10 +163,11 @@ xfs_get_acl(struct inode *inode, int type)
 		 */
 		if (error == -ENOATTR)
 			goto out_update_cache;
+		acl = ERR_PTR(error);
 		goto out;
 	}
 
-	acl = xfs_acl_from_disk(xfs_acl, XFS_ACL_MAX_ENTRIES(ip->i_mount));
+	acl = xfs_acl_from_disk(xfs_acl, len, XFS_ACL_MAX_ENTRIES(ip->i_mount));
 	if (IS_ERR(acl))
 		goto out;
 
@@ -248,29 +252,6 @@ xfs_set_mode(struct inode *inode, umode_t mode)
 	return error;
 }
 
-static int
-xfs_acl_exists(struct inode *inode, unsigned char *name)
-{
-	int len = XFS_ACL_MAX_SIZE(XFS_M(inode->i_sb));
-
-	return (xfs_attr_get(XFS_I(inode), name, NULL, &len,
-			    ATTR_ROOT|ATTR_KERNOVAL) == 0);
-}
-
-int
-posix_acl_access_exists(struct inode *inode)
-{
-	return xfs_acl_exists(inode, SGI_ACL_FILE);
-}
-
-int
-posix_acl_default_exists(struct inode *inode)
-{
-	if (!S_ISDIR(inode->i_mode))
-		return 0;
-	return xfs_acl_exists(inode, SGI_ACL_DEFAULT);
-}
-
 int
 xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 {
diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h
index 3841b07f27bf..286fa89217f5 100644
--- a/fs/xfs/xfs_acl.h
+++ b/fs/xfs/xfs_acl.h
@@ -20,20 +20,18 @@
 
 struct inode;
 struct posix_acl;
-struct xfs_inode;
 
 #ifdef CONFIG_XFS_POSIX_ACL
 extern struct posix_acl *xfs_get_acl(struct inode *inode, int type);
 extern int xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type);
-extern int posix_acl_access_exists(struct inode *inode);
-extern int posix_acl_default_exists(struct inode *inode);
 #else
 static inline struct posix_acl *xfs_get_acl(struct inode *inode, int type)
 {
 	return NULL;
 }
 # define xfs_set_acl					NULL
-# define posix_acl_access_exists(inode)			0
-# define posix_acl_default_exists(inode)		0
 #endif /* CONFIG_XFS_POSIX_ACL */
+
+extern void xfs_forget_acl(struct inode *inode, const char *name, int xflags);
+
 #endif	/* __XFS_ACL_H__ */
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index 50ab2879b9da..e49b2406d15d 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -36,6 +36,21 @@
 #include <linux/pagevec.h>
 #include <linux/writeback.h>
 
+/* flags for direct write completions */
+#define XFS_DIO_FLAG_UNWRITTEN	(1 << 0)
+#define XFS_DIO_FLAG_APPEND	(1 << 1)
+
+/*
+ * structure owned by writepages passed to individual writepage calls
+ */
+struct xfs_writepage_ctx {
+	struct xfs_bmbt_irec    imap;
+	bool			imap_valid;
+	unsigned int		io_type;
+	struct xfs_ioend	*ioend;
+	sector_t		last_block;
+};
+
 void
 xfs_count_page_state(
 	struct page		*page,
@@ -55,7 +70,7 @@ xfs_count_page_state(
 	} while ((bh = bh->b_this_page) != head);
 }
 
-STATIC struct block_device *
+struct block_device *
 xfs_find_bdev_for_inode(
 	struct inode		*inode)
 {
@@ -172,6 +187,12 @@ xfs_setfilesize_ioend(
 	current_set_flags_nested(&tp->t_pflags, PF_FSTRANS);
 	__sb_writers_acquired(VFS_I(ip)->i_sb, SB_FREEZE_FS);
 
+	/* we abort the update if there was an IO error */
+	if (ioend->io_error) {
+		xfs_trans_cancel(tp);
+		return ioend->io_error;
+	}
+
 	return xfs_setfilesize(ip, tp, ioend->io_offset, ioend->io_size);
 }
 
@@ -208,18 +229,23 @@ xfs_end_io(
 	struct xfs_inode *ip = XFS_I(ioend->io_inode);
 	int		error = 0;
 
-	if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
+	/*
+	 * Set an error if the mount has shut down and proceed with end I/O
+	 * processing so it can perform whatever cleanups are necessary.
+	 */
+	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
 		ioend->io_error = -EIO;
-		goto done;
-	}
-	if (ioend->io_error)
-		goto done;
 
 	/*
 	 * For unwritten extents we need to issue transactions to convert a
 	 * range to normal written extens after the data I/O has finished.
+	 * Detecting and handling completion IO errors is done individually
+	 * for each case as different cleanup operations need to be performed
+	 * on error.
 	 */
 	if (ioend->io_type == XFS_IO_UNWRITTEN) {
+		if (ioend->io_error)
+			goto done;
 		error = xfs_iomap_write_unwritten(ip, ioend->io_offset,
 						  ioend->io_size);
 	} else if (ioend->io_append_trans) {
@@ -256,7 +282,7 @@ xfs_alloc_ioend(
 	 */
 	atomic_set(&ioend->io_remaining, 1);
 	ioend->io_error = 0;
-	ioend->io_list = NULL;
+	INIT_LIST_HEAD(&ioend->io_list);
 	ioend->io_type = type;
 	ioend->io_inode = inode;
 	ioend->io_buffer_head = NULL;
@@ -274,8 +300,7 @@ xfs_map_blocks(
 	struct inode		*inode,
 	loff_t			offset,
 	struct xfs_bmbt_irec	*imap,
-	int			type,
-	int			nonblocking)
+	int			type)
 {
 	struct xfs_inode	*ip = XFS_I(inode);
 	struct xfs_mount	*mp = ip->i_mount;
@@ -291,12 +316,7 @@ xfs_map_blocks(
 	if (type == XFS_IO_UNWRITTEN)
 		bmapi_flags |= XFS_BMAPI_IGSTATE;
 
-	if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED)) {
-		if (nonblocking)
-			return -EAGAIN;
-		xfs_ilock(ip, XFS_ILOCK_SHARED);
-	}
-
+	xfs_ilock(ip, XFS_ILOCK_SHARED);
 	ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
 	       (ip->i_df.if_flags & XFS_IFEXTENTS));
 	ASSERT(offset <= mp->m_super->s_maxbytes);
@@ -332,7 +352,7 @@ xfs_map_blocks(
 	return 0;
 }
 
-STATIC int
+STATIC bool
 xfs_imap_valid(
 	struct inode		*inode,
 	struct xfs_bmbt_irec	*imap,
@@ -405,8 +425,7 @@ xfs_start_buffer_writeback(
 STATIC void
 xfs_start_page_writeback(
 	struct page		*page,
-	int			clear_dirty,
-	int			buffers)
+	int			clear_dirty)
 {
 	ASSERT(PageLocked(page));
 	ASSERT(!PageWriteback(page));
@@ -425,10 +444,6 @@ xfs_start_page_writeback(
 		set_page_writeback_keepwrite(page);
 
 	unlock_page(page);
-
-	/* If no buffers on the page are to be written, finish it here */
-	if (!buffers)
-		end_page_writeback(page);
 }
 
 static inline int xfs_bio_add_buffer(struct bio *bio, struct buffer_head *bh)
@@ -437,153 +452,101 @@ static inline int xfs_bio_add_buffer(struct bio *bio, struct buffer_head *bh)
 }
 
 /*
- * Submit all of the bios for all of the ioends we have saved up, covering the
- * initial writepage page and also any probed pages.
- *
- * Because we may have multiple ioends spanning a page, we need to start
- * writeback on all the buffers before we submit them for I/O. If we mark the
- * buffers as we got, then we can end up with a page that only has buffers
- * marked async write and I/O complete on can occur before we mark the other
- * buffers async write.
- *
- * The end result of this is that we trip a bug in end_page_writeback() because
- * we call it twice for the one page as the code in end_buffer_async_write()
- * assumes that all buffers on the page are started at the same time.
- *
- * The fix is two passes across the ioend list - one to start writeback on the
- * buffer_heads, and then submit them for I/O on the second pass.
+ * Submit all of the bios for an ioend. We are only passed a single ioend at a
+ * time; the caller is responsible for chaining prior to submission.
  *
  * If @fail is non-zero, it means that we have a situation where some part of
  * the submission process has failed after we have marked paged for writeback
  * and unlocked them. In this situation, we need to fail the ioend chain rather
  * than submit it to IO. This typically only happens on a filesystem shutdown.
  */
-STATIC void
+STATIC int
 xfs_submit_ioend(
 	struct writeback_control *wbc,
 	xfs_ioend_t		*ioend,
-	int			fail)
+	int			status)
 {
-	xfs_ioend_t		*head = ioend;
-	xfs_ioend_t		*next;
 	struct buffer_head	*bh;
 	struct bio		*bio;
 	sector_t		lastblock = 0;
 
-	/* Pass 1 - start writeback */
-	do {
-		next = ioend->io_list;
-		for (bh = ioend->io_buffer_head; bh; bh = bh->b_private)
-			xfs_start_buffer_writeback(bh);
-	} while ((ioend = next) != NULL);
+	/* Reserve log space if we might write beyond the on-disk inode size. */
+	if (!status &&
+	     ioend->io_type != XFS_IO_UNWRITTEN && xfs_ioend_is_append(ioend))
+		status = xfs_setfilesize_trans_alloc(ioend);
+	/*
+	 * If we are failing the IO now, just mark the ioend with an
+	 * error and finish it. This will run IO completion immediately
+	 * as there is only one reference to the ioend at this point in
+	 * time.
+	 */
+	if (status) {
+		ioend->io_error = status;
+		xfs_finish_ioend(ioend);
+		return status;
+	}
 
-	/* Pass 2 - submit I/O */
-	ioend = head;
-	do {
-		next = ioend->io_list;
-		bio = NULL;
+	bio = NULL;
+	for (bh = ioend->io_buffer_head; bh; bh = bh->b_private) {
 
-		/*
-		 * If we are failing the IO now, just mark the ioend with an
-		 * error and finish it. This will run IO completion immediately
-		 * as there is only one reference to the ioend at this point in
-		 * time.
-		 */
-		if (fail) {
-			ioend->io_error = fail;
-			xfs_finish_ioend(ioend);
-			continue;
+		if (!bio) {
+retry:
+			bio = xfs_alloc_ioend_bio(bh);
+		} else if (bh->b_blocknr != lastblock + 1) {
+			xfs_submit_ioend_bio(wbc, ioend, bio);
+			goto retry;
 		}
 
-		for (bh = ioend->io_buffer_head; bh; bh = bh->b_private) {
-
-			if (!bio) {
- retry:
-				bio = xfs_alloc_ioend_bio(bh);
-			} else if (bh->b_blocknr != lastblock + 1) {
-				xfs_submit_ioend_bio(wbc, ioend, bio);
-				goto retry;
-			}
-
-			if (xfs_bio_add_buffer(bio, bh) != bh->b_size) {
-				xfs_submit_ioend_bio(wbc, ioend, bio);
-				goto retry;
-			}
-
-			lastblock = bh->b_blocknr;
-		}
-		if (bio)
+		if (xfs_bio_add_buffer(bio, bh) != bh->b_size) {
 			xfs_submit_ioend_bio(wbc, ioend, bio);
-		xfs_finish_ioend(ioend);
-	} while ((ioend = next) != NULL);
-}
-
-/*
- * Cancel submission of all buffer_heads so far in this endio.
- * Toss the endio too.  Only ever called for the initial page
- * in a writepage request, so only ever one page.
- */
-STATIC void
-xfs_cancel_ioend(
-	xfs_ioend_t		*ioend)
-{
-	xfs_ioend_t		*next;
-	struct buffer_head	*bh, *next_bh;
-
-	do {
-		next = ioend->io_list;
-		bh = ioend->io_buffer_head;
-		do {
-			next_bh = bh->b_private;
-			clear_buffer_async_write(bh);
-			/*
-			 * The unwritten flag is cleared when added to the
-			 * ioend. We're not submitting for I/O so mark the
-			 * buffer unwritten again for next time around.
-			 */
-			if (ioend->io_type == XFS_IO_UNWRITTEN)
-				set_buffer_unwritten(bh);
-			unlock_buffer(bh);
-		} while ((bh = next_bh) != NULL);
+			goto retry;
+		}
 
-		mempool_free(ioend, xfs_ioend_pool);
-	} while ((ioend = next) != NULL);
+		lastblock = bh->b_blocknr;
+	}
+	if (bio)
+		xfs_submit_ioend_bio(wbc, ioend, bio);
+	xfs_finish_ioend(ioend);
+	return 0;
 }
 
 /*
  * Test to see if we've been building up a completion structure for
  * earlier buffers -- if so, we try to append to this ioend if we
  * can, otherwise we finish off any current ioend and start another.
- * Return true if we've finished the given ioend.
+ * Return the ioend we finished off so that the caller can submit it
+ * once it has finished processing the dirty page.
  */
 STATIC void
 xfs_add_to_ioend(
 	struct inode		*inode,
 	struct buffer_head	*bh,
 	xfs_off_t		offset,
-	unsigned int		type,
-	xfs_ioend_t		**result,
-	int			need_ioend)
+	struct xfs_writepage_ctx *wpc,
+	struct list_head	*iolist)
 {
-	xfs_ioend_t		*ioend = *result;
-
-	if (!ioend || need_ioend || type != ioend->io_type) {
-		xfs_ioend_t	*previous = *result;
-
-		ioend = xfs_alloc_ioend(inode, type);
-		ioend->io_offset = offset;
-		ioend->io_buffer_head = bh;
-		ioend->io_buffer_tail = bh;
-		if (previous)
-			previous->io_list = ioend;
-		*result = ioend;
+	if (!wpc->ioend || wpc->io_type != wpc->ioend->io_type ||
+	    bh->b_blocknr != wpc->last_block + 1 ||
+	    offset != wpc->ioend->io_offset + wpc->ioend->io_size) {
+		struct xfs_ioend	*new;
+
+		if (wpc->ioend)
+			list_add(&wpc->ioend->io_list, iolist);
+
+		new = xfs_alloc_ioend(inode, wpc->io_type);
+		new->io_offset = offset;
+		new->io_buffer_head = bh;
+		new->io_buffer_tail = bh;
+		wpc->ioend = new;
 	} else {
-		ioend->io_buffer_tail->b_private = bh;
-		ioend->io_buffer_tail = bh;
+		wpc->ioend->io_buffer_tail->b_private = bh;
+		wpc->ioend->io_buffer_tail = bh;
 	}
 
 	bh->b_private = NULL;
-	ioend->io_size += bh->b_size;
+	wpc->ioend->io_size += bh->b_size;
+	wpc->last_block = bh->b_blocknr;
+	xfs_start_buffer_writeback(bh);
 }
 
 STATIC void
@@ -669,183 +632,6 @@ xfs_check_page_type(
 	return false;
 }
 
-/*
- * Allocate & map buffers for page given the extent map. Write it out.
- * except for the original page of a writepage, this is called on
- * delalloc/unwritten pages only, for the original page it is possible
- * that the page has no mapping at all.
- */
-STATIC int
-xfs_convert_page(
-	struct inode		*inode,
-	struct page		*page,
-	loff_t			tindex,
-	struct xfs_bmbt_irec	*imap,
-	xfs_ioend_t		**ioendp,
-	struct writeback_control *wbc)
-{
-	struct buffer_head	*bh, *head;
-	xfs_off_t		end_offset;
-	unsigned long		p_offset;
-	unsigned int		type;
-	int			len, page_dirty;
-	int			count = 0, done = 0, uptodate = 1;
- 	xfs_off_t		offset = page_offset(page);
-
-	if (page->index != tindex)
-		goto fail;
-	if (!trylock_page(page))
-		goto fail;
-	if (PageWriteback(page))
-		goto fail_unlock_page;
-	if (page->mapping != inode->i_mapping)
-		goto fail_unlock_page;
-	if (!xfs_check_page_type(page, (*ioendp)->io_type, false))
-		goto fail_unlock_page;
-
-	/*
-	 * page_dirty is initially a count of buffers on the page before
-	 * EOF and is decremented as we move each into a cleanable state.
-	 *
-	 * Derivation:
-	 *
-	 * End offset is the highest offset that this page should represent.
-	 * If we are on the last page, (end_offset & (PAGE_CACHE_SIZE - 1))
-	 * will evaluate non-zero and be less than PAGE_CACHE_SIZE and
-	 * hence give us the correct page_dirty count. On any other page,
-	 * it will be zero and in that case we need page_dirty to be the
-	 * count of buffers on the page.
-	 */
-	end_offset = min_t(unsigned long long,
-			(xfs_off_t)(page->index + 1) << PAGE_CACHE_SHIFT,
-			i_size_read(inode));
-
-	/*
-	 * If the current map does not span the entire page we are about to try
-	 * to write, then give up. The only way we can write a page that spans
-	 * multiple mappings in a single writeback iteration is via the
-	 * xfs_vm_writepage() function. Data integrity writeback requires the
-	 * entire page to be written in a single attempt, otherwise the part of
-	 * the page we don't write here doesn't get written as part of the data
-	 * integrity sync.
-	 *
-	 * For normal writeback, we also don't attempt to write partial pages
-	 * here as it simply means that write_cache_pages() will see it under
-	 * writeback and ignore the page until some point in the future, at
-	 * which time this will be the only page in the file that needs
-	 * writeback.  Hence for more optimal IO patterns, we should always
-	 * avoid partial page writeback due to multiple mappings on a page here.
-	 */
-	if (!xfs_imap_valid(inode, imap, end_offset))
-		goto fail_unlock_page;
-
-	len = 1 << inode->i_blkbits;
-	p_offset = min_t(unsigned long, end_offset & (PAGE_CACHE_SIZE - 1),
-					PAGE_CACHE_SIZE);
-	p_offset = p_offset ? roundup(p_offset, len) : PAGE_CACHE_SIZE;
-	page_dirty = p_offset / len;
-
-	/*
-	 * The moment we find a buffer that doesn't match our current type
-	 * specification or can't be written, abort the loop and start
-	 * writeback. As per the above xfs_imap_valid() check, only
-	 * xfs_vm_writepage() can handle partial page writeback fully - we are
-	 * limited here to the buffers that are contiguous with the current
-	 * ioend, and hence a buffer we can't write breaks that contiguity and
-	 * we have to defer the rest of the IO to xfs_vm_writepage().
-	 */
-	bh = head = page_buffers(page);
-	do {
-		if (offset >= end_offset)
-			break;
-		if (!buffer_uptodate(bh))
-			uptodate = 0;
-		if (!(PageUptodate(page) || buffer_uptodate(bh))) {
-			done = 1;
-			break;
-		}
-
-		if (buffer_unwritten(bh) || buffer_delay(bh) ||
-		    buffer_mapped(bh)) {
-			if (buffer_unwritten(bh))
-				type = XFS_IO_UNWRITTEN;
-			else if (buffer_delay(bh))
-				type = XFS_IO_DELALLOC;
-			else
-				type = XFS_IO_OVERWRITE;
-
-			/*
-			 * imap should always be valid because of the above
-			 * partial page end_offset check on the imap.
-			 */
-			ASSERT(xfs_imap_valid(inode, imap, offset));
-
-			lock_buffer(bh);
-			if (type != XFS_IO_OVERWRITE)
-				xfs_map_at_offset(inode, bh, imap, offset);
-			xfs_add_to_ioend(inode, bh, offset, type,
-					 ioendp, done);
-
-			page_dirty--;
-			count++;
-		} else {
-			done = 1;
-			break;
-		}
-	} while (offset += len, (bh = bh->b_this_page) != head);
-
-	if (uptodate && bh == head)
-		SetPageUptodate(page);
-
-	if (count) {
-		if (--wbc->nr_to_write <= 0 &&
-		    wbc->sync_mode == WB_SYNC_NONE)
-			done = 1;
-	}
-	xfs_start_page_writeback(page, !page_dirty, count);
-
-	return done;
- fail_unlock_page:
-	unlock_page(page);
- fail:
-	return 1;
-}
-
-/*
- * Convert & write out a cluster of pages in the same extent as defined
- * by mp and following the start page.
- */
-STATIC void
-xfs_cluster_write(
-	struct inode		*inode,
-	pgoff_t			tindex,
-	struct xfs_bmbt_irec	*imap,
-	xfs_ioend_t		**ioendp,
-	struct writeback_control *wbc,
-	pgoff_t			tlast)
-{
-	struct pagevec		pvec;
-	int			done = 0, i;
-
-	pagevec_init(&pvec, 0);
-	while (!done && tindex <= tlast) {
-		unsigned len = min_t(pgoff_t, PAGEVEC_SIZE, tlast - tindex + 1);
-
-		if (!pagevec_lookup(&pvec, inode->i_mapping, tindex, len))
-			break;
-
-		for (i = 0; i < pagevec_count(&pvec); i++) {
-			done = xfs_convert_page(inode, pvec.pages[i], tindex++,
-					imap, ioendp, wbc);
-			if (done)
-				break;
-		}
-
-		pagevec_release(&pvec);
-		cond_resched();
-	}
-}
-
 STATIC void
 xfs_vm_invalidatepage(
 	struct page		*page,
@@ -918,11 +704,169 @@ next_buffer:
 
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 out_invalidate:
-	xfs_vm_invalidatepage(page, 0, PAGE_CACHE_SIZE);
+	xfs_vm_invalidatepage(page, 0, PAGE_SIZE);
 	return;
 }
 
 /*
+ * We implement an immediate ioend submission policy here to avoid needing to
+ * chain multiple ioends and hence nest mempool allocations which can violate
+ * forward progress guarantees we need to provide. The current ioend we are
+ * adding buffers to is cached on the writepage context, and if the new buffer
+ * does not append to the cached ioend it will create a new ioend and cache that
+ * instead.
+ *
+ * If a new ioend is created and cached, the old ioend is returned and queued
+ * locally for submission once the entire page is processed or an error has been
+ * detected.  While ioends are submitted immediately after they are completed,
+ * batching optimisations are provided by higher level block plugging.
+ *
+ * At the end of a writeback pass, there will be a cached ioend remaining on the
+ * writepage context that the caller will need to submit.
+ */
+static int
+xfs_writepage_map(
+	struct xfs_writepage_ctx *wpc,
+	struct writeback_control *wbc,
+	struct inode		*inode,
+	struct page		*page,
+	loff_t			offset,
+	__uint64_t              end_offset)
+{
+	LIST_HEAD(submit_list);
+	struct xfs_ioend	*ioend, *next;
+	struct buffer_head	*bh, *head;
+	ssize_t			len = 1 << inode->i_blkbits;
+	int			error = 0;
+	int			count = 0;
+	int			uptodate = 1;
+
+	bh = head = page_buffers(page);
+	offset = page_offset(page);
+	do {
+		if (offset >= end_offset)
+			break;
+		if (!buffer_uptodate(bh))
+			uptodate = 0;
+
+		/*
+		 * set_page_dirty dirties all buffers in a page, independent
+		 * of their state.  The dirty state however is entirely
+		 * meaningless for holes (!mapped && uptodate), so skip
+		 * buffers covering holes here.
+		 */
+		if (!buffer_mapped(bh) && buffer_uptodate(bh)) {
+			wpc->imap_valid = false;
+			continue;
+		}
+
+		if (buffer_unwritten(bh)) {
+			if (wpc->io_type != XFS_IO_UNWRITTEN) {
+				wpc->io_type = XFS_IO_UNWRITTEN;
+				wpc->imap_valid = false;
+			}
+		} else if (buffer_delay(bh)) {
+			if (wpc->io_type != XFS_IO_DELALLOC) {
+				wpc->io_type = XFS_IO_DELALLOC;
+				wpc->imap_valid = false;
+			}
+		} else if (buffer_uptodate(bh)) {
+			if (wpc->io_type != XFS_IO_OVERWRITE) {
+				wpc->io_type = XFS_IO_OVERWRITE;
+				wpc->imap_valid = false;
+			}
+		} else {
+			if (PageUptodate(page))
+				ASSERT(buffer_mapped(bh));
+			/*
+			 * This buffer is not uptodate and will not be
+			 * written to disk.  Ensure that we will put any
+			 * subsequent writeable buffers into a new
+			 * ioend.
+			 */
+			wpc->imap_valid = false;
+			continue;
+		}
+
+		if (wpc->imap_valid)
+			wpc->imap_valid = xfs_imap_valid(inode, &wpc->imap,
+							 offset);
+		if (!wpc->imap_valid) {
+			error = xfs_map_blocks(inode, offset, &wpc->imap,
+					     wpc->io_type);
+			if (error)
+				goto out;
+			wpc->imap_valid = xfs_imap_valid(inode, &wpc->imap,
+							 offset);
+		}
+		if (wpc->imap_valid) {
+			lock_buffer(bh);
+			if (wpc->io_type != XFS_IO_OVERWRITE)
+				xfs_map_at_offset(inode, bh, &wpc->imap, offset);
+			xfs_add_to_ioend(inode, bh, offset, wpc, &submit_list);
+			count++;
+		}
+
+	} while (offset += len, ((bh = bh->b_this_page) != head));
+
+	if (uptodate && bh == head)
+		SetPageUptodate(page);
+
+	ASSERT(wpc->ioend || list_empty(&submit_list));
+
+out:
+	/*
+	 * On error, we have to fail the ioend here because we have locked
+	 * buffers in the ioend. If we don't do this, we'll deadlock
+	 * invalidating the page as that tries to lock the buffers on the page.
+	 * Also, because we may have set pages under writeback, we have to make
+	 * sure we run IO completion to mark the error state of the IO
+	 * appropriately, so we can't cancel the ioend directly here. That means
+	 * we have to mark this page as under writeback if we included any
+	 * buffers from it in the ioend chain so that completion treats it
+	 * correctly.
+	 *
+	 * If we didn't include the page in the ioend, the on error we can
+	 * simply discard and unlock it as there are no other users of the page
+	 * or it's buffers right now. The caller will still need to trigger
+	 * submission of outstanding ioends on the writepage context so they are
+	 * treated correctly on error.
+	 */
+	if (count) {
+		xfs_start_page_writeback(page, !error);
+
+		/*
+		 * Preserve the original error if there was one, otherwise catch
+		 * submission errors here and propagate into subsequent ioend
+		 * submissions.
+		 */
+		list_for_each_entry_safe(ioend, next, &submit_list, io_list) {
+			int error2;
+
+			list_del_init(&ioend->io_list);
+			error2 = xfs_submit_ioend(wbc, ioend, error);
+			if (error2 && !error)
+				error = error2;
+		}
+	} else if (error) {
+		xfs_aops_discard_page(page);
+		ClearPageUptodate(page);
+		unlock_page(page);
+	} else {
+		/*
+		 * We can end up here with no error and nothing to write if we
+		 * race with a partial page truncate on a sub-page block sized
+		 * filesystem. In that case we need to mark the page clean.
+		 */
+		xfs_start_page_writeback(page, 1);
+		end_page_writeback(page);
+	}
+
+	mapping_set_error(page->mapping, error);
+	return error;
+}
+
+/*
  * Write out a dirty page.
  *
  * For delalloc space on the page we need to allocate space and flush it.
@@ -931,22 +875,16 @@ out_invalidate:
  * For any other dirty buffer heads on the page we should flush them.
  */
 STATIC int
-xfs_vm_writepage(
+xfs_do_writepage(
 	struct page		*page,
-	struct writeback_control *wbc)
+	struct writeback_control *wbc,
+	void			*data)
 {
+	struct xfs_writepage_ctx *wpc = data;
 	struct inode		*inode = page->mapping->host;
-	struct buffer_head	*bh, *head;
-	struct xfs_bmbt_irec	imap;
-	xfs_ioend_t		*ioend = NULL, *iohead = NULL;
 	loff_t			offset;
-	unsigned int		type;
 	__uint64_t              end_offset;
-	pgoff_t                 end_index, last_index;
-	ssize_t			len;
-	int			err, imap_valid = 0, uptodate = 1;
-	int			count = 0;
-	int			nonblocking = 0;
+	pgoff_t                 end_index;
 
 	trace_xfs_writepage(inode, page, 0, 0);
 
@@ -973,12 +911,9 @@ xfs_vm_writepage(
 	if (WARN_ON_ONCE(current->flags & PF_FSTRANS))
 		goto redirty;
 
-	/* Is this page beyond the end of the file? */
-	offset = i_size_read(inode);
-	end_index = offset >> PAGE_CACHE_SHIFT;
-	last_index = (offset - 1) >> PAGE_CACHE_SHIFT;
-
 	/*
+	 * Is this page beyond the end of the file?
+	 *
 	 * The page index is less than the end_index, adjust the end_offset
 	 * to the highest offset that this page should represent.
 	 * -----------------------------------------------------
@@ -989,8 +924,10 @@ xfs_vm_writepage(
 	 * |     desired writeback range    |      see else    |
 	 * ---------------------------------^------------------|
 	 */
+	offset = i_size_read(inode);
+	end_index = offset >> PAGE_SHIFT;
 	if (page->index < end_index)
-		end_offset = (xfs_off_t)(page->index + 1) << PAGE_CACHE_SHIFT;
+		end_offset = (xfs_off_t)(page->index + 1) << PAGE_SHIFT;
 	else {
 		/*
 		 * Check whether the page to write out is beyond or straddles
@@ -1003,7 +940,7 @@ xfs_vm_writepage(
 		 * |				    |      Straddles     |
 		 * ---------------------------------^-----------|--------|
 		 */
-		unsigned offset_into_page = offset & (PAGE_CACHE_SIZE - 1);
+		unsigned offset_into_page = offset & (PAGE_SIZE - 1);
 
 		/*
 		 * Skip the page if it is fully outside i_size, e.g. due to a
@@ -1034,158 +971,13 @@ xfs_vm_writepage(
 		 * memory is zeroed when mapped, and writes to that region are
 		 * not written out to the file."
 		 */
-		zero_user_segment(page, offset_into_page, PAGE_CACHE_SIZE);
+		zero_user_segment(page, offset_into_page, PAGE_SIZE);
 
 		/* Adjust the end_offset to the end of file */
 		end_offset = offset;
 	}
 
-	len = 1 << inode->i_blkbits;
-
-	bh = head = page_buffers(page);
-	offset = page_offset(page);
-	type = XFS_IO_OVERWRITE;
-
-	if (wbc->sync_mode == WB_SYNC_NONE)
-		nonblocking = 1;
-
-	do {
-		int new_ioend = 0;
-
-		if (offset >= end_offset)
-			break;
-		if (!buffer_uptodate(bh))
-			uptodate = 0;
-
-		/*
-		 * set_page_dirty dirties all buffers in a page, independent
-		 * of their state.  The dirty state however is entirely
-		 * meaningless for holes (!mapped && uptodate), so skip
-		 * buffers covering holes here.
-		 */
-		if (!buffer_mapped(bh) && buffer_uptodate(bh)) {
-			imap_valid = 0;
-			continue;
-		}
-
-		if (buffer_unwritten(bh)) {
-			if (type != XFS_IO_UNWRITTEN) {
-				type = XFS_IO_UNWRITTEN;
-				imap_valid = 0;
-			}
-		} else if (buffer_delay(bh)) {
-			if (type != XFS_IO_DELALLOC) {
-				type = XFS_IO_DELALLOC;
-				imap_valid = 0;
-			}
-		} else if (buffer_uptodate(bh)) {
-			if (type != XFS_IO_OVERWRITE) {
-				type = XFS_IO_OVERWRITE;
-				imap_valid = 0;
-			}
-		} else {
-			if (PageUptodate(page))
-				ASSERT(buffer_mapped(bh));
-			/*
-			 * This buffer is not uptodate and will not be
-			 * written to disk.  Ensure that we will put any
-			 * subsequent writeable buffers into a new
-			 * ioend.
-			 */
-			imap_valid = 0;
-			continue;
-		}
-
-		if (imap_valid)
-			imap_valid = xfs_imap_valid(inode, &imap, offset);
-		if (!imap_valid) {
-			/*
-			 * If we didn't have a valid mapping then we need to
-			 * put the new mapping into a separate ioend structure.
-			 * This ensures non-contiguous extents always have
-			 * separate ioends, which is particularly important
-			 * for unwritten extent conversion at I/O completion
-			 * time.
-			 */
-			new_ioend = 1;
-			err = xfs_map_blocks(inode, offset, &imap, type,
-					     nonblocking);
-			if (err)
-				goto error;
-			imap_valid = xfs_imap_valid(inode, &imap, offset);
-		}
-		if (imap_valid) {
-			lock_buffer(bh);
-			if (type != XFS_IO_OVERWRITE)
-				xfs_map_at_offset(inode, bh, &imap, offset);
-			xfs_add_to_ioend(inode, bh, offset, type, &ioend,
-					 new_ioend);
-			count++;
-		}
-
-		if (!iohead)
-			iohead = ioend;
-
-	} while (offset += len, ((bh = bh->b_this_page) != head));
-
-	if (uptodate && bh == head)
-		SetPageUptodate(page);
-
-	xfs_start_page_writeback(page, 1, count);
-
-	/* if there is no IO to be submitted for this page, we are done */
-	if (!ioend)
-		return 0;
-
-	ASSERT(iohead);
-
-	/*
-	 * Any errors from this point onwards need tobe reported through the IO
-	 * completion path as we have marked the initial page as under writeback
-	 * and unlocked it.
-	 */
-	if (imap_valid) {
-		xfs_off_t		end_index;
-
-		end_index = imap.br_startoff + imap.br_blockcount;
-
-		/* to bytes */
-		end_index <<= inode->i_blkbits;
-
-		/* to pages */
-		end_index = (end_index - 1) >> PAGE_CACHE_SHIFT;
-
-		/* check against file size */
-		if (end_index > last_index)
-			end_index = last_index;
-
-		xfs_cluster_write(inode, page->index + 1, &imap, &ioend,
-				  wbc, end_index);
-	}
-
-
-	/*
-	 * Reserve log space if we might write beyond the on-disk inode size.
-	 */
-	err = 0;
-	if (ioend->io_type != XFS_IO_UNWRITTEN && xfs_ioend_is_append(ioend))
-		err = xfs_setfilesize_trans_alloc(ioend);
-
-	xfs_submit_ioend(wbc, iohead, err);
-
-	return 0;
-
-error:
-	if (iohead)
-		xfs_cancel_ioend(iohead);
-
-	if (err == -EAGAIN)
-		goto redirty;
-
-	xfs_aops_discard_page(page);
-	ClearPageUptodate(page);
-	unlock_page(page);
-	return err;
+	return xfs_writepage_map(wpc, wbc, inode, page, offset, end_offset);
 
 redirty:
 	redirty_page_for_writepage(wbc, page);
@@ -1194,12 +986,40 @@ redirty:
 }
 
 STATIC int
+xfs_vm_writepage(
+	struct page		*page,
+	struct writeback_control *wbc)
+{
+	struct xfs_writepage_ctx wpc = {
+		.io_type = XFS_IO_INVALID,
+	};
+	int			ret;
+
+	ret = xfs_do_writepage(page, wbc, &wpc);
+	if (wpc.ioend)
+		ret = xfs_submit_ioend(wbc, wpc.ioend, ret);
+	return ret;
+}
+
+STATIC int
 xfs_vm_writepages(
 	struct address_space	*mapping,
 	struct writeback_control *wbc)
 {
+	struct xfs_writepage_ctx wpc = {
+		.io_type = XFS_IO_INVALID,
+	};
+	int			ret;
+
 	xfs_iflags_clear(XFS_I(mapping->host), XFS_ITRUNCATED);
-	return generic_writepages(mapping, wbc);
+	if (dax_mapping(mapping))
+		return dax_writeback_mapping_range(mapping,
+				xfs_find_bdev_for_inode(mapping->host), wbc);
+
+	ret = write_cache_pages(mapping, wbc, xfs_do_writepage, &wpc);
+	if (wpc.ioend)
+		ret = xfs_submit_ioend(wbc, wpc.ioend, ret);
+	return ret;
 }
 
 /*
@@ -1229,27 +1049,15 @@ xfs_vm_releasepage(
 }
 
 /*
- * When we map a DIO buffer, we may need to attach an ioend that describes the
- * type of write IO we are doing. This passes to the completion function the
- * operations it needs to perform. If the mapping is for an overwrite wholly
- * within the EOF then we don't need an ioend and so we don't allocate one.
- * This avoids the unnecessary overhead of allocating and freeing ioends for
- * workloads that don't require transactions on IO completion.
- *
- * If we get multiple mappings in a single IO, we might be mapping different
- * types. But because the direct IO can only have a single private pointer, we
- * need to ensure that:
+ * When we map a DIO buffer, we may need to pass flags to
+ * xfs_end_io_direct_write to tell it what kind of write IO we are doing.
  *
- * a) i) the ioend spans the entire region of unwritten mappings; or
- *    ii) the ioend spans all the mappings that cross or are beyond EOF; and
- * b) if it contains unwritten extents, it is *permanently* marked as such
- *
- * We could do this by chaining ioends like buffered IO does, but we only
- * actually get one IO completion callback from the direct IO, and that spans
- * the entire IO regardless of how many mappings and IOs are needed to complete
- * the DIO. There is only going to be one reference to the ioend and its life
- * cycle is constrained by the DIO completion code. hence we don't need
- * reference counting here.
+ * Note that for DIO, an IO to the highest supported file block offset (i.e.
+ * 2^63 - 1FSB bytes) will result in the offset + count overflowing a signed 64
+ * bit variable. Hence if we see this overflow, we have to assume that the IO is
+ * extending the file size. We won't know for sure until IO completion is run
+ * and the actual max write offset is communicated to the IO completion
+ * routine.
  */
 static void
 xfs_map_direct(
@@ -1258,44 +1066,18 @@ xfs_map_direct(
 	struct xfs_bmbt_irec	*imap,
 	xfs_off_t		offset)
 {
-	struct xfs_ioend	*ioend;
+	uintptr_t		*flags = (uintptr_t *)&bh_result->b_private;
 	xfs_off_t		size = bh_result->b_size;
-	int			type;
 
-	if (ISUNWRITTEN(imap))
-		type = XFS_IO_UNWRITTEN;
-	else
-		type = XFS_IO_OVERWRITE;
-
-	trace_xfs_gbmap_direct(XFS_I(inode), offset, size, type, imap);
-
-	if (bh_result->b_private) {
-		ioend = bh_result->b_private;
-		ASSERT(ioend->io_size > 0);
-		ASSERT(offset >= ioend->io_offset);
-		if (offset + size > ioend->io_offset + ioend->io_size)
-			ioend->io_size = offset - ioend->io_offset + size;
-
-		if (type == XFS_IO_UNWRITTEN && type != ioend->io_type)
-			ioend->io_type = XFS_IO_UNWRITTEN;
-
-		trace_xfs_gbmap_direct_update(XFS_I(inode), ioend->io_offset,
-					      ioend->io_size, ioend->io_type,
-					      imap);
-	} else if (type == XFS_IO_UNWRITTEN ||
-		   offset + size > i_size_read(inode)) {
-		ioend = xfs_alloc_ioend(inode, type);
-		ioend->io_offset = offset;
-		ioend->io_size = size;
-
-		bh_result->b_private = ioend;
-		set_buffer_defer_completion(bh_result);
+	trace_xfs_get_blocks_map_direct(XFS_I(inode), offset, size,
+		ISUNWRITTEN(imap) ? XFS_IO_UNWRITTEN : XFS_IO_OVERWRITE, imap);
 
-		trace_xfs_gbmap_direct_new(XFS_I(inode), offset, size, type,
-					   imap);
-	} else {
-		trace_xfs_gbmap_direct_none(XFS_I(inode), offset, size, type,
-					    imap);
+	if (ISUNWRITTEN(imap)) {
+		*flags |= XFS_DIO_FLAG_UNWRITTEN;
+		set_buffer_defer_completion(bh_result);
+	} else if (offset + size > i_size_read(inode) || offset + size < 0) {
+		*flags |= XFS_DIO_FLAG_APPEND;
+		set_buffer_defer_completion(bh_result);
 	}
 }
 
@@ -1345,7 +1127,8 @@ __xfs_get_blocks(
 	sector_t		iblock,
 	struct buffer_head	*bh_result,
 	int			create,
-	bool			direct)
+	bool			direct,
+	bool			dax_fault)
 {
 	struct xfs_inode	*ip = XFS_I(inode);
 	struct xfs_mount	*mp = ip->i_mount;
@@ -1393,18 +1176,20 @@ __xfs_get_blocks(
 	if (error)
 		goto out_unlock;
 
+	/* for DAX, we convert unwritten extents directly */
 	if (create &&
 	    (!nimaps ||
 	     (imap.br_startblock == HOLESTARTBLOCK ||
-	      imap.br_startblock == DELAYSTARTBLOCK))) {
+	      imap.br_startblock == DELAYSTARTBLOCK) ||
+	     (IS_DAX(inode) && ISUNWRITTEN(&imap)))) {
 		if (direct || xfs_get_extsz_hint(ip)) {
 			/*
-			 * Drop the ilock in preparation for starting the block
-			 * allocation transaction.  It will be retaken
-			 * exclusively inside xfs_iomap_write_direct for the
-			 * actual allocation.
+			 * xfs_iomap_write_direct() expects the shared lock. It
+			 * is unlocked on return.
 			 */
-			xfs_iunlock(ip, lockmode);
+			if (lockmode == XFS_ILOCK_EXCL)
+				xfs_ilock_demote(ip, lockmode);
+
 			error = xfs_iomap_write_direct(ip, offset, size,
 						       &imap, nimaps);
 			if (error)
@@ -1441,6 +1226,12 @@ __xfs_get_blocks(
 		goto out_unlock;
 	}
 
+	if (IS_DAX(inode) && create) {
+		ASSERT(!ISUNWRITTEN(&imap));
+		/* zeroing is not needed at a higher layer */
+		new = 0;
+	}
+
 	/* trim mapping down to size requested */
 	if (direct || size > (1 << inode->i_blkbits))
 		xfs_map_trim_size(inode, iblock, bh_result,
@@ -1457,8 +1248,12 @@ __xfs_get_blocks(
 		if (ISUNWRITTEN(&imap))
 			set_buffer_unwritten(bh_result);
 		/* direct IO needs special help */
-		if (create && direct)
-			xfs_map_direct(inode, bh_result, &imap, offset);
+		if (create && direct) {
+			if (dax_fault)
+				ASSERT(!ISUNWRITTEN(&imap));
+			else
+				xfs_map_direct(inode, bh_result, &imap, offset);
+		}
 	}
 
 	/*
@@ -1505,7 +1300,7 @@ xfs_get_blocks(
 	struct buffer_head	*bh_result,
 	int			create)
 {
-	return __xfs_get_blocks(inode, iblock, bh_result, create, false);
+	return __xfs_get_blocks(inode, iblock, bh_result, create, false, false);
 }
 
 int
@@ -1515,45 +1310,63 @@ xfs_get_blocks_direct(
 	struct buffer_head	*bh_result,
 	int			create)
 {
-	return __xfs_get_blocks(inode, iblock, bh_result, create, true);
+	return __xfs_get_blocks(inode, iblock, bh_result, create, true, false);
 }
 
-static void
-__xfs_end_io_direct_write(
+int
+xfs_get_blocks_dax_fault(
 	struct inode		*inode,
-	struct xfs_ioend	*ioend,
+	sector_t		iblock,
+	struct buffer_head	*bh_result,
+	int			create)
+{
+	return __xfs_get_blocks(inode, iblock, bh_result, create, true, true);
+}
+
+/*
+ * Complete a direct I/O write request.
+ *
+ * xfs_map_direct passes us some flags in the private data to tell us what to
+ * do.  If no flags are set, then the write IO is an overwrite wholly within
+ * the existing allocated file size and so there is nothing for us to do.
+ *
+ * Note that in this case the completion can be called in interrupt context,
+ * whereas if we have flags set we will always be called in task context
+ * (i.e. from a workqueue).
+ */
+STATIC int
+xfs_end_io_direct_write(
+	struct kiocb		*iocb,
 	loff_t			offset,
-	ssize_t			size)
+	ssize_t			size,
+	void			*private)
 {
-	struct xfs_mount	*mp = XFS_I(inode)->i_mount;
+	struct inode		*inode = file_inode(iocb->ki_filp);
+	struct xfs_inode	*ip = XFS_I(inode);
+	struct xfs_mount	*mp = ip->i_mount;
+	uintptr_t		flags = (uintptr_t)private;
+	int			error = 0;
 
-	if (XFS_FORCED_SHUTDOWN(mp) || ioend->io_error)
-		goto out_end_io;
+	trace_xfs_end_io_direct_write(ip, offset, size);
 
-	/*
-	 * dio completion end_io functions are only called on writes if more
-	 * than 0 bytes was written.
-	 */
-	ASSERT(size > 0);
+	if (XFS_FORCED_SHUTDOWN(mp))
+		return -EIO;
 
-	/*
-	 * The ioend only maps whole blocks, while the IO may be sector aligned.
-	 * Hence the ioend offset/size may not match the IO offset/size exactly.
-	 * Because we don't map overwrites within EOF into the ioend, the offset
-	 * may not match, but only if the endio spans EOF.  Either way, write
-	 * the IO sizes into the ioend so that completion processing does the
-	 * right thing.
-	 */
-	ASSERT(offset + size <= ioend->io_offset + ioend->io_size);
-	ioend->io_size = size;
-	ioend->io_offset = offset;
+	if (size <= 0)
+		return size;
 
 	/*
-	 * The ioend tells us whether we are doing unwritten extent conversion
+	 * The flags tell us whether we are doing unwritten extent conversions
 	 * or an append transaction that updates the on-disk file size. These
 	 * cases are the only cases where we should *potentially* be needing
 	 * to update the VFS inode size.
-	 *
+	 */
+	if (flags == 0) {
+		ASSERT(offset + size <= i_size_read(inode));
+		return 0;
+	}
+
+	/*
 	 * We need to update the in-core inode size here so that we don't end up
 	 * with the on-disk inode size being outside the in-core inode size. We
 	 * have no other method of updating EOF for AIO, so always do it here
@@ -1564,130 +1377,56 @@ __xfs_end_io_direct_write(
 	 * here can result in EOF moving backwards and Bad Things Happen when
 	 * that occurs.
 	 */
-	spin_lock(&XFS_I(inode)->i_flags_lock);
+	spin_lock(&ip->i_flags_lock);
 	if (offset + size > i_size_read(inode))
 		i_size_write(inode, offset + size);
-	spin_unlock(&XFS_I(inode)->i_flags_lock);
+	spin_unlock(&ip->i_flags_lock);
 
-	/*
-	 * If we are doing an append IO that needs to update the EOF on disk,
-	 * do the transaction reserve now so we can use common end io
-	 * processing. Stashing the error (if there is one) in the ioend will
-	 * result in the ioend processing passing on the error if it is
-	 * possible as we can't return it from here.
-	 */
-	if (ioend->io_type == XFS_IO_OVERWRITE)
-		ioend->io_error = xfs_setfilesize_trans_alloc(ioend);
+	if (flags & XFS_DIO_FLAG_UNWRITTEN) {
+		trace_xfs_end_io_direct_write_unwritten(ip, offset, size);
 
-out_end_io:
-	xfs_end_io(&ioend->io_work);
-	return;
-}
+		error = xfs_iomap_write_unwritten(ip, offset, size);
+	} else if (flags & XFS_DIO_FLAG_APPEND) {
+		struct xfs_trans *tp;
 
-/*
- * Complete a direct I/O write request.
- *
- * The ioend structure is passed from __xfs_get_blocks() to tell us what to do.
- * If no ioend exists (i.e. @private == NULL) then the write IO is an overwrite
- * wholly within the EOF and so there is nothing for us to do. Note that in this
- * case the completion can be called in interrupt context, whereas if we have an
- * ioend we will always be called in task context (i.e. from a workqueue).
- */
-STATIC void
-xfs_end_io_direct_write(
-	struct kiocb		*iocb,
-	loff_t			offset,
-	ssize_t			size,
-	void			*private)
-{
-	struct inode		*inode = file_inode(iocb->ki_filp);
-	struct xfs_ioend	*ioend = private;
+		trace_xfs_end_io_direct_write_append(ip, offset, size);
 
-	trace_xfs_gbmap_direct_endio(XFS_I(inode), offset, size,
-				     ioend ? ioend->io_type : 0, NULL);
-
-	if (!ioend) {
-		ASSERT(offset + size <= i_size_read(inode));
-		return;
+		tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS);
+		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_fsyncts, 0, 0);
+		if (error) {
+			xfs_trans_cancel(tp);
+			return error;
+		}
+		error = xfs_setfilesize(ip, tp, offset, size);
 	}
 
-	__xfs_end_io_direct_write(inode, ioend, offset, size);
-}
-
-/*
- * For DAX we need a mapping buffer callback for unwritten extent conversion
- * when page faults allocate blocks and then zero them. Note that in this
- * case the mapping indicated by the ioend may extend beyond EOF. We most
- * definitely do not want to extend EOF here, so we trim back the ioend size to
- * EOF.
- */
-#ifdef CONFIG_FS_DAX
-void
-xfs_end_io_dax_write(
-	struct buffer_head	*bh,
-	int			uptodate)
-{
-	struct xfs_ioend	*ioend = bh->b_private;
-	struct inode		*inode = ioend->io_inode;
-	ssize_t			size = ioend->io_size;
-
-	ASSERT(IS_DAX(ioend->io_inode));
-
-	/* if there was an error zeroing, then don't convert it */
-	if (!uptodate)
-		ioend->io_error = -EIO;
-
-	/*
-	 * Trim update to EOF, so we don't extend EOF during unwritten extent
-	 * conversion of partial EOF blocks.
-	 */
-	spin_lock(&XFS_I(inode)->i_flags_lock);
-	if (ioend->io_offset + size > i_size_read(inode))
-		size = i_size_read(inode) - ioend->io_offset;
-	spin_unlock(&XFS_I(inode)->i_flags_lock);
-
-	__xfs_end_io_direct_write(inode, ioend, ioend->io_offset, size);
-
+	return error;
 }
-#else
-void xfs_end_io_dax_write(struct buffer_head *bh, int uptodate) { }
-#endif
 
-static inline ssize_t
-xfs_vm_do_dio(
-	struct inode		*inode,
+STATIC ssize_t
+xfs_vm_direct_IO(
 	struct kiocb		*iocb,
 	struct iov_iter		*iter,
-	loff_t			offset,
-	void			(*endio)(struct kiocb	*iocb,
-					 loff_t		offset,
-					 ssize_t	size,
-					 void		*private),
-	int			flags)
+	loff_t			offset)
 {
+	struct inode		*inode = iocb->ki_filp->f_mapping->host;
+	dio_iodone_t		*endio = NULL;
+	int			flags = 0;
 	struct block_device	*bdev;
 
-	if (IS_DAX(inode))
+	if (iov_iter_rw(iter) == WRITE) {
+		endio = xfs_end_io_direct_write;
+		flags = DIO_ASYNC_EXTEND;
+	}
+
+	if (IS_DAX(inode)) {
 		return dax_do_io(iocb, inode, iter, offset,
 				 xfs_get_blocks_direct, endio, 0);
+	}
 
 	bdev = xfs_find_bdev_for_inode(inode);
 	return  __blockdev_direct_IO(iocb, inode, bdev, iter, offset,
-				     xfs_get_blocks_direct, endio, NULL, flags);
-}
-
-STATIC ssize_t
-xfs_vm_direct_IO(
-	struct kiocb		*iocb,
-	struct iov_iter		*iter,
-	loff_t			offset)
-{
-	struct inode		*inode = iocb->ki_filp->f_mapping->host;
-
-	if (iov_iter_rw(iter) == WRITE)
-		return xfs_vm_do_dio(inode, iocb, iter, offset,
-				     xfs_end_io_direct_write, DIO_ASYNC_EXTEND);
-	return xfs_vm_do_dio(inode, iocb, iter, offset, NULL, 0);
+			xfs_get_blocks_direct, endio, NULL, flags);
 }
 
 /*
@@ -1736,9 +1475,10 @@ xfs_vm_write_failed(
 	loff_t			block_offset;
 	loff_t			block_start;
 	loff_t			block_end;
-	loff_t			from = pos & (PAGE_CACHE_SIZE - 1);
+	loff_t			from = pos & (PAGE_SIZE - 1);
 	loff_t			to = from + len;
 	struct buffer_head	*bh, *head;
+	struct xfs_mount	*mp = XFS_I(inode)->i_mount;
 
 	/*
 	 * The request pos offset might be 32 or 64 bit, this is all fine
@@ -1751,7 +1491,7 @@ xfs_vm_write_failed(
 	 * start of the page by using shifts rather than masks the mismatch
 	 * problem.
 	 */
-	block_offset = (pos >> PAGE_CACHE_SHIFT) << PAGE_CACHE_SHIFT;
+	block_offset = (pos >> PAGE_SHIFT) << PAGE_SHIFT;
 
 	ASSERT(block_offset + from == pos);
 
@@ -1770,14 +1510,23 @@ xfs_vm_write_failed(
 		if (block_start >= to)
 			break;
 
-		if (!buffer_delay(bh))
+		/*
+		 * Process delalloc and unwritten buffers beyond EOF. We can
+		 * encounter unwritten buffers in the event that a file has
+		 * post-EOF unwritten extents and an extending write happens to
+		 * fail (e.g., an unaligned write that also involves a delalloc
+		 * to the same page).
+		 */
+		if (!buffer_delay(bh) && !buffer_unwritten(bh))
 			continue;
 
-		if (!buffer_new(bh) && block_offset < i_size_read(inode))
+		if (!xfs_mp_fail_writes(mp) && !buffer_new(bh) &&
+		    block_offset < i_size_read(inode))
 			continue;
 
-		xfs_vm_kill_delalloc_range(inode, block_offset,
-					   block_offset + bh->b_size);
+		if (buffer_delay(bh))
+			xfs_vm_kill_delalloc_range(inode, block_offset,
+						   block_offset + bh->b_size);
 
 		/*
 		 * This buffer does not contain data anymore. make sure anyone
@@ -1788,6 +1537,7 @@ xfs_vm_write_failed(
 		clear_buffer_mapped(bh);
 		clear_buffer_new(bh);
 		clear_buffer_dirty(bh);
+		clear_buffer_unwritten(bh);
 	}
 
 }
@@ -1808,17 +1558,20 @@ xfs_vm_write_begin(
 	struct page		**pagep,
 	void			**fsdata)
 {
-	pgoff_t			index = pos >> PAGE_CACHE_SHIFT;
+	pgoff_t			index = pos >> PAGE_SHIFT;
 	struct page		*page;
 	int			status;
+	struct xfs_mount	*mp = XFS_I(mapping->host)->i_mount;
 
-	ASSERT(len <= PAGE_CACHE_SIZE);
+	ASSERT(len <= PAGE_SIZE);
 
 	page = grab_cache_page_write_begin(mapping, index, flags);
 	if (!page)
 		return -ENOMEM;
 
 	status = __block_write_begin(page, pos, len, xfs_get_blocks);
+	if (xfs_mp_fail_writes(mp))
+		status = -EIO;
 	if (unlikely(status)) {
 		struct inode	*inode = mapping->host;
 		size_t		isize = i_size_read(inode);
@@ -1831,13 +1584,15 @@ xfs_vm_write_begin(
 		 * allocated in this write, not blocks that were previously
 		 * written successfully.
 		 */
+		if (xfs_mp_fail_writes(mp))
+			isize = 0;
 		if (pos + len > isize) {
 			ssize_t start = max_t(ssize_t, pos, isize);
 
 			truncate_pagecache_range(inode, start, pos + len);
 		}
 
-		page_cache_release(page);
+		put_page(page);
 		page = NULL;
 	}
 
@@ -1865,7 +1620,7 @@ xfs_vm_write_end(
 {
 	int			ret;
 
-	ASSERT(len <= PAGE_CACHE_SIZE);
+	ASSERT(len <= PAGE_SIZE);
 
 	ret = generic_write_end(file, mapping, pos, len, copied, page, fsdata);
 	if (unlikely(ret < len)) {
@@ -1904,6 +1659,7 @@ xfs_vm_readpage(
 	struct file		*unused,
 	struct page		*page)
 {
+	trace_xfs_vm_readpage(page->mapping->host, 1);
 	return mpage_readpage(page, xfs_get_blocks);
 }
 
@@ -1914,6 +1670,7 @@ xfs_vm_readpages(
 	struct list_head	*pages,
 	unsigned		nr_pages)
 {
+	trace_xfs_vm_readpages(mapping->host, nr_pages);
 	return mpage_readpages(mapping, pages, nr_pages, xfs_get_blocks);
 }
 
@@ -1938,7 +1695,6 @@ xfs_vm_set_page_dirty(
 	loff_t			end_offset;
 	loff_t			offset;
 	int			newly_dirty;
-	struct mem_cgroup	*memcg;
 
 	if (unlikely(!mapping))
 		return !TestSetPageDirty(page);
@@ -1959,10 +1715,10 @@ xfs_vm_set_page_dirty(
 		} while (bh != head);
 	}
 	/*
-	 * Use mem_group_begin_page_stat() to keep PageDirty synchronized with
-	 * per-memcg dirty page counters.
+	 * Lock out page->mem_cgroup migration to keep PageDirty
+	 * synchronized with per-memcg dirty page counters.
 	 */
-	memcg = mem_cgroup_begin_page_stat(page);
+	lock_page_memcg(page);
 	newly_dirty = !TestSetPageDirty(page);
 	spin_unlock(&mapping->private_lock);
 
@@ -1973,13 +1729,13 @@ xfs_vm_set_page_dirty(
 		spin_lock_irqsave(&mapping->tree_lock, flags);
 		if (page->mapping) {	/* Race with truncate? */
 			WARN_ON_ONCE(!PageUptodate(page));
-			account_page_dirtied(page, mapping, memcg);
+			account_page_dirtied(page, mapping);
 			radix_tree_tag_set(&mapping->page_tree,
 					page_index(page), PAGECACHE_TAG_DIRTY);
 		}
 		spin_unlock_irqrestore(&mapping->tree_lock, flags);
 	}
-	mem_cgroup_end_page_stat(memcg);
+	unlock_page_memcg(page);
 	if (newly_dirty)
 		__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
 	return newly_dirty;
diff --git a/fs/xfs/xfs_aops.h b/fs/xfs/xfs_aops.h
index 86afd1ac7895..b4421177b68d 100644
--- a/fs/xfs/xfs_aops.h
+++ b/fs/xfs/xfs_aops.h
@@ -24,12 +24,14 @@ extern mempool_t *xfs_ioend_pool;
  * Types of I/O for bmap clustering and I/O completion tracking.
  */
 enum {
+	XFS_IO_INVALID,		/* initial state */
 	XFS_IO_DELALLOC,	/* covers delalloc region */
 	XFS_IO_UNWRITTEN,	/* covers allocated but uninitialized data */
 	XFS_IO_OVERWRITE,	/* covers already allocated extent */
 };
 
 #define XFS_IO_TYPES \
+	{ XFS_IO_INVALID,		"invalid" }, \
 	{ XFS_IO_DELALLOC,		"delalloc" }, \
 	{ XFS_IO_UNWRITTEN,		"unwritten" }, \
 	{ XFS_IO_OVERWRITE,		"overwrite" }
@@ -39,7 +41,7 @@ enum {
  * It can manage several multi-page bio's at once.
  */
 typedef struct xfs_ioend {
-	struct xfs_ioend	*io_list;	/* next ioend in chain */
+	struct list_head	io_list;	/* next ioend in chain */
 	unsigned int		io_type;	/* delalloc / unwritten */
 	int			io_error;	/* I/O error code */
 	atomic_t		io_remaining;	/* hold count */
@@ -58,8 +60,10 @@ int	xfs_get_blocks(struct inode *inode, sector_t offset,
 		       struct buffer_head *map_bh, int create);
 int	xfs_get_blocks_direct(struct inode *inode, sector_t offset,
 			      struct buffer_head *map_bh, int create);
-void	xfs_end_io_dax_write(struct buffer_head *bh, int uptodate);
+int	xfs_get_blocks_dax_fault(struct inode *inode, sector_t offset,
+			         struct buffer_head *map_bh, int create);
 
 extern void xfs_count_page_state(struct page *, int *, int *);
+extern struct block_device *xfs_find_bdev_for_inode(struct inode *);
 
 #endif /* __XFS_AOPS_H__ */
diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c
index 65fb37a18e92..4fa14820e2e2 100644
--- a/fs/xfs/xfs_attr_list.c
+++ b/fs/xfs/xfs_attr_list.c
@@ -202,8 +202,10 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 					sbp->namelen,
 					sbp->valuelen,
 					&sbp->name[sbp->namelen]);
-		if (error)
+		if (error) {
+			kmem_free(sbuf);
 			return error;
+		}
 		if (context->seen_enough)
 			break;
 		cursor->offset++;
@@ -454,14 +456,13 @@ xfs_attr3_leaf_list_int(
 				args.rmtblkcnt = xfs_attr3_rmt_blocks(
 							args.dp->i_mount, valuelen);
 				retval = xfs_attr_rmtval_get(&args);
-				if (retval)
-					return retval;
-				retval = context->put_listent(context,
-						entry->flags,
-						name_rmt->name,
-						(int)name_rmt->namelen,
-						valuelen,
-						args.value);
+				if (!retval)
+					retval = context->put_listent(context,
+							entry->flags,
+							name_rmt->name,
+							(int)name_rmt->namelen,
+							valuelen,
+							args.value);
 				kmem_free(args.value);
 			} else {
 				retval = context->put_listent(context,
@@ -511,7 +512,7 @@ xfs_attr_list_int(
 	xfs_inode_t *dp = context->dp;
 	uint		lock_mode;
 
-	XFS_STATS_INC(xs_attr_list);
+	XFS_STATS_INC(dp->i_mount, xs_attr_list);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
 		return -EIO;
diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
index 3bf4ad0d19e4..3b6309865c65 100644
--- a/fs/xfs/xfs_bmap_util.c
+++ b/fs/xfs/xfs_bmap_util.c
@@ -57,37 +57,67 @@ xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb)
 }
 
 /*
+ * Routine to zero an extent on disk allocated to the specific inode.
+ *
+ * The VFS functions take a linearised filesystem block offset, so we have to
+ * convert the sparse xfs fsb to the right format first.
+ * VFS types are real funky, too.
+ */
+int
+xfs_zero_extent(
+	struct xfs_inode *ip,
+	xfs_fsblock_t	start_fsb,
+	xfs_off_t	count_fsb)
+{
+	struct xfs_mount *mp = ip->i_mount;
+	xfs_daddr_t	sector = xfs_fsb_to_db(ip, start_fsb);
+	sector_t	block = XFS_BB_TO_FSBT(mp, sector);
+	ssize_t		size = XFS_FSB_TO_B(mp, count_fsb);
+
+	if (IS_DAX(VFS_I(ip)))
+		return dax_clear_sectors(xfs_find_bdev_for_inode(VFS_I(ip)),
+				sector, size);
+
+	/*
+	 * let the block layer decide on the fastest method of
+	 * implementing the zeroing.
+	 */
+	return sb_issue_zeroout(mp->m_super, block, count_fsb, GFP_NOFS);
+
+}
+
+/*
  * Routine to be called at transaction's end by xfs_bmapi, xfs_bunmapi
  * caller.  Frees all the extents that need freeing, which must be done
  * last due to locking considerations.  We never free any extents in
  * the first transaction.
  *
- * Return 1 if the given transaction was committed and a new one
- * started, and 0 otherwise in the committed parameter.
+ * If an inode *ip is provided, rejoin it to the transaction if
+ * the transaction was committed.
  */
 int						/* error */
 xfs_bmap_finish(
 	struct xfs_trans		**tp,	/* transaction pointer addr */
 	struct xfs_bmap_free		*flist,	/* i/o: list extents to free */
-	int				*committed)/* xact committed or not */
+	struct xfs_inode		*ip)
 {
 	struct xfs_efd_log_item		*efd;	/* extent free data */
 	struct xfs_efi_log_item		*efi;	/* extent free intention */
 	int				error;	/* error return value */
+	int				committed;/* xact committed or not */
 	struct xfs_bmap_free_item	*free;	/* free extent item */
 	struct xfs_bmap_free_item	*next;	/* next item on free list */
 
 	ASSERT((*tp)->t_flags & XFS_TRANS_PERM_LOG_RES);
-	if (flist->xbf_count == 0) {
-		*committed = 0;
+	if (flist->xbf_count == 0)
 		return 0;
-	}
+
 	efi = xfs_trans_get_efi(*tp, flist->xbf_count);
 	for (free = flist->xbf_first; free; free = free->xbfi_next)
 		xfs_trans_log_efi_extent(*tp, efi, free->xbfi_startblock,
 			free->xbfi_blockcount);
 
-	error = __xfs_trans_roll(tp, NULL, committed);
+	error = __xfs_trans_roll(tp, ip, &committed);
 	if (error) {
 		/*
 		 * If the transaction was committed, drop the EFD reference
@@ -99,16 +129,13 @@ xfs_bmap_finish(
 		 * transaction so we should return committed=1 even though we're
 		 * returning an error.
 		 */
-		if (*committed) {
+		if (committed) {
 			xfs_efi_release(efi);
 			xfs_force_shutdown((*tp)->t_mountp,
 				(error == -EFSCORRUPTED) ?
 					SHUTDOWN_CORRUPT_INCORE :
 					SHUTDOWN_META_IO_ERROR);
-		} else {
-			*committed = 1;
 		}
-
 		return error;
 	}
 
@@ -176,10 +203,12 @@ xfs_bmap_rtalloc(
 		ralen = MAXEXTLEN / mp->m_sb.sb_rextsize;
 
 	/*
-	 * Lock out other modifications to the RT bitmap inode.
+	 * Lock out modifications to both the RT bitmap and summary inodes
 	 */
 	xfs_ilock(mp->m_rbmip, XFS_ILOCK_EXCL);
 	xfs_trans_ijoin(ap->tp, mp->m_rbmip, XFS_ILOCK_EXCL);
+	xfs_ilock(mp->m_rsumip, XFS_ILOCK_EXCL);
+	xfs_trans_ijoin(ap->tp, mp->m_rsumip, XFS_ILOCK_EXCL);
 
 	/*
 	 * If it's an allocation to an empty file at offset 0,
@@ -229,6 +258,13 @@ xfs_bmap_rtalloc(
 		xfs_trans_mod_dquot_byino(ap->tp, ap->ip,
 			ap->wasdel ? XFS_TRANS_DQ_DELRTBCOUNT :
 					XFS_TRANS_DQ_RTBCOUNT, (long) ralen);
+
+		/* Zero the extent if we were asked to do so */
+		if (ap->userdata & XFS_ALLOC_USERDATA_ZERO) {
+			error = xfs_zero_extent(ap->ip, ap->blkno, ap->length);
+			if (error)
+				return error;
+		}
 	} else {
 		ap->length = 0;
 	}
@@ -788,7 +824,7 @@ bool
 xfs_can_free_eofblocks(struct xfs_inode *ip, bool force)
 {
 	/* prealloc/delalloc exists only on regular files */
-	if (!S_ISREG(ip->i_d.di_mode))
+	if (!S_ISREG(VFS_I(ip)->i_mode))
 		return false;
 
 	/*
@@ -933,7 +969,6 @@ xfs_alloc_file_space(
 	xfs_bmbt_irec_t		imaps[1], *imapp;
 	xfs_bmap_free_t		free_list;
 	uint			qblocks, resblks, resrtextents;
-	int			committed;
 	int			error;
 
 	trace_xfs_alloc_file_space(ip);
@@ -1027,24 +1062,21 @@ xfs_alloc_file_space(
 		xfs_bmap_init(&free_list, &firstfsb);
 		error = xfs_bmapi_write(tp, ip, startoffset_fsb,
 					allocatesize_fsb, alloc_type, &firstfsb,
-					0, imapp, &nimaps, &free_list);
-		if (error) {
+					resblks, imapp, &nimaps, &free_list);
+		if (error)
 			goto error0;
-		}
 
 		/*
 		 * Complete the transaction
 		 */
-		error = xfs_bmap_finish(&tp, &free_list, &committed);
-		if (error) {
+		error = xfs_bmap_finish(&tp, &free_list, NULL);
+		if (error)
 			goto error0;
-		}
 
 		error = xfs_trans_commit(tp);
 		xfs_iunlock(ip, XFS_ILOCK_EXCL);
-		if (error) {
+		if (error)
 			break;
-		}
 
 		allocated_fsb = imapp->br_blockcount;
 
@@ -1170,7 +1202,6 @@ xfs_free_file_space(
 	xfs_off_t		offset,
 	xfs_off_t		len)
 {
-	int			committed;
 	int			done;
 	xfs_fileoff_t		endoffset_fsb;
 	int			error;
@@ -1206,7 +1237,7 @@ xfs_free_file_space(
 	/* wait for the completion of any pending DIOs */
 	inode_dio_wait(VFS_I(ip));
 
-	rounding = max_t(xfs_off_t, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE);
+	rounding = max_t(xfs_off_t, 1 << mp->m_sb.sb_blocklog, PAGE_SIZE);
 	ioffset = round_down(offset, rounding);
 	iendoffset = round_up(offset + len, rounding) - 1;
 	error = filemap_write_and_wait_range(VFS_I(ip)->i_mapping, ioffset,
@@ -1310,17 +1341,15 @@ xfs_free_file_space(
 		error = xfs_bunmapi(tp, ip, startoffset_fsb,
 				  endoffset_fsb - startoffset_fsb,
 				  0, 2, &firstfsb, &free_list, &done);
-		if (error) {
+		if (error)
 			goto error0;
-		}
 
 		/*
 		 * complete the transaction
 		 */
-		error = xfs_bmap_finish(&tp, &free_list, &committed);
-		if (error) {
+		error = xfs_bmap_finish(&tp, &free_list, NULL);
+		if (error)
 			goto error0;
-		}
 
 		error = xfs_trans_commit(tp);
 		xfs_iunlock(ip, XFS_ILOCK_EXCL);
@@ -1398,7 +1427,6 @@ xfs_shift_file_space(
 	int			error;
 	struct xfs_bmap_free	free_list;
 	xfs_fsblock_t		first_block;
-	int			committed;
 	xfs_fileoff_t		stop_fsb;
 	xfs_fileoff_t		next_fsb;
 	xfs_fileoff_t		shift_fsb;
@@ -1438,7 +1466,7 @@ xfs_shift_file_space(
 	if (error)
 		return error;
 	error = invalidate_inode_pages2_range(VFS_I(ip)->i_mapping,
-					offset >> PAGE_CACHE_SHIFT, -1);
+					offset >> PAGE_SHIFT, -1);
 	if (error)
 		return error;
 
@@ -1490,7 +1518,7 @@ xfs_shift_file_space(
 		if (error)
 			goto out_bmap_cancel;
 
-		error = xfs_bmap_finish(&tp, &free_list, &committed);
+		error = xfs_bmap_finish(&tp, &free_list, NULL);
 		if (error)
 			goto out_bmap_cancel;
 
@@ -1701,7 +1729,7 @@ xfs_swap_extents(
 	xfs_lock_two_inodes(ip, tip, XFS_MMAPLOCK_EXCL);
 
 	/* Verify that both files have the same format */
-	if ((ip->i_d.di_mode & S_IFMT) != (tip->i_d.di_mode & S_IFMT)) {
+	if ((VFS_I(ip)->i_mode & S_IFMT) != (VFS_I(tip)->i_mode & S_IFMT)) {
 		error = -EINVAL;
 		goto out_unlock;
 	}
diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c
index 8ecffb35935b..9a2191b91137 100644
--- a/fs/xfs/xfs_buf.c
+++ b/fs/xfs/xfs_buf.c
@@ -201,7 +201,7 @@ _xfs_buf_alloc(
 	atomic_set(&bp->b_pin_count, 0);
 	init_waitqueue_head(&bp->b_waiters);
 
-	XFS_STATS_INC(xb_create);
+	XFS_STATS_INC(target->bt_mount, xb_create);
 	trace_xfs_buf_init(bp, _RET_IP_);
 
 	return bp;
@@ -354,15 +354,16 @@ retry:
 			 */
 			if (!(++retries % 100))
 				xfs_err(NULL,
-		"possible memory allocation deadlock in %s (mode:0x%x)",
+		"%s(%u) possible memory allocation deadlock in %s (mode:0x%x)",
+					current->comm, current->pid,
 					__func__, gfp_mask);
 
-			XFS_STATS_INC(xb_page_retries);
+			XFS_STATS_INC(bp->b_target->bt_mount, xb_page_retries);
 			congestion_wait(BLK_RW_ASYNC, HZ/50);
 			goto retry;
 		}
 
-		XFS_STATS_INC(xb_page_found);
+		XFS_STATS_INC(bp->b_target->bt_mount, xb_page_found);
 
 		nbytes = min_t(size_t, size, PAGE_SIZE - offset);
 		size -= nbytes;
@@ -516,7 +517,7 @@ _xfs_buf_find(
 		new_bp->b_pag = pag;
 		spin_unlock(&pag->pag_buf_lock);
 	} else {
-		XFS_STATS_INC(xb_miss_locked);
+		XFS_STATS_INC(btp->bt_mount, xb_miss_locked);
 		spin_unlock(&pag->pag_buf_lock);
 		xfs_perag_put(pag);
 	}
@@ -529,11 +530,11 @@ found:
 	if (!xfs_buf_trylock(bp)) {
 		if (flags & XBF_TRYLOCK) {
 			xfs_buf_rele(bp);
-			XFS_STATS_INC(xb_busy_locked);
+			XFS_STATS_INC(btp->bt_mount, xb_busy_locked);
 			return NULL;
 		}
 		xfs_buf_lock(bp);
-		XFS_STATS_INC(xb_get_locked_waited);
+		XFS_STATS_INC(btp->bt_mount, xb_get_locked_waited);
 	}
 
 	/*
@@ -549,7 +550,7 @@ found:
 	}
 
 	trace_xfs_buf_find(bp, flags, _RET_IP_);
-	XFS_STATS_INC(xb_get_locked);
+	XFS_STATS_INC(btp->bt_mount, xb_get_locked);
 	return bp;
 }
 
@@ -603,7 +604,14 @@ found:
 		}
 	}
 
-	XFS_STATS_INC(xb_get);
+	/*
+	 * Clear b_error if this is a lookup from a caller that doesn't expect
+	 * valid data to be found in the buffer.
+	 */
+	if (!(flags & XBF_READ))
+		xfs_buf_ioerror(bp, 0);
+
+	XFS_STATS_INC(target->bt_mount, xb_get);
 	trace_xfs_buf_get(bp, flags, _RET_IP_);
 	return bp;
 }
@@ -642,8 +650,8 @@ xfs_buf_read_map(
 	if (bp) {
 		trace_xfs_buf_read(bp, flags, _RET_IP_);
 
-		if (!XFS_BUF_ISDONE(bp)) {
-			XFS_STATS_INC(xb_get_read);
+		if (!(bp->b_flags & XBF_DONE)) {
+			XFS_STATS_INC(target->bt_mount, xb_get_read);
 			bp->b_ops = ops;
 			_xfs_buf_read(bp, flags);
 		} else if (flags & XBF_ASYNC) {
@@ -1044,7 +1052,7 @@ xfs_buf_ioend_work(
 	xfs_buf_ioend(bp);
 }
 
-void
+static void
 xfs_buf_ioend_async(
 	struct xfs_buf	*bp)
 {
@@ -1519,6 +1527,16 @@ xfs_wait_buftarg(
 	LIST_HEAD(dispose);
 	int loop = 0;
 
+	/*
+	 * We need to flush the buffer workqueue to ensure that all IO
+	 * completion processing is 100% done. Just waiting on buffer locks is
+	 * not sufficient for async IO as the reference count held over IO is
+	 * not released until after the buffer lock is dropped. Hence we need to
+	 * ensure here that all reference counts have been dropped before we
+	 * start walking the LRU list.
+	 */
+	drain_workqueue(btp->bt_mount->m_buf_workqueue);
+
 	/* loop until there is nothing left on the lru list. */
 	while (list_lru_count(&btp->bt_lru)) {
 		list_lru_walk(&btp->bt_lru, xfs_buftarg_wait_rele,
@@ -1631,13 +1649,9 @@ xfs_setsize_buftarg(
 	btp->bt_meta_sectormask = sectorsize - 1;
 
 	if (set_blocksize(btp->bt_bdev, sectorsize)) {
-		char name[BDEVNAME_SIZE];
-
-		bdevname(btp->bt_bdev, name);
-
 		xfs_warn(btp->bt_mount,
-			"Cannot set_blocksize to %u on device %s",
-			sectorsize, name);
+			"Cannot set_blocksize to %u on device %pg",
+			sectorsize, btp->bt_bdev);
 		return -EINVAL;
 	}
 
diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h
index c79b717d9b88..4eb89bd4ee73 100644
--- a/fs/xfs/xfs_buf.h
+++ b/fs/xfs/xfs_buf.h
@@ -132,6 +132,7 @@ struct xfs_buf_map {
 	struct xfs_buf_map (map) = { .bm_bn = (blkno), .bm_len = (numblk) };
 
 struct xfs_buf_ops {
+	char *name;
 	void (*verify_read)(struct xfs_buf *);
 	void (*verify_write)(struct xfs_buf *);
 };
@@ -301,6 +302,7 @@ extern void xfs_buf_iomove(xfs_buf_t *, size_t, size_t, void *,
 
 /* Buffer Utility Routines */
 extern void *xfs_buf_offset(struct xfs_buf *, size_t);
+extern void xfs_buf_stale(struct xfs_buf *bp);
 
 /* Delayed Write Buffer Routines */
 extern bool xfs_buf_delwri_queue(struct xfs_buf *, struct list_head *);
@@ -311,31 +313,6 @@ extern int xfs_buf_delwri_submit_nowait(struct list_head *);
 extern int xfs_buf_init(void);
 extern void xfs_buf_terminate(void);
 
-#define XFS_BUF_ZEROFLAGS(bp) \
-	((bp)->b_flags &= ~(XBF_READ|XBF_WRITE|XBF_ASYNC| \
-			    XBF_SYNCIO|XBF_FUA|XBF_FLUSH| \
-			    XBF_WRITE_FAIL))
-
-void xfs_buf_stale(struct xfs_buf *bp);
-#define XFS_BUF_UNSTALE(bp)	((bp)->b_flags &= ~XBF_STALE)
-#define XFS_BUF_ISSTALE(bp)	((bp)->b_flags & XBF_STALE)
-
-#define XFS_BUF_DONE(bp)	((bp)->b_flags |= XBF_DONE)
-#define XFS_BUF_UNDONE(bp)	((bp)->b_flags &= ~XBF_DONE)
-#define XFS_BUF_ISDONE(bp)	((bp)->b_flags & XBF_DONE)
-
-#define XFS_BUF_ASYNC(bp)	((bp)->b_flags |= XBF_ASYNC)
-#define XFS_BUF_UNASYNC(bp)	((bp)->b_flags &= ~XBF_ASYNC)
-#define XFS_BUF_ISASYNC(bp)	((bp)->b_flags & XBF_ASYNC)
-
-#define XFS_BUF_READ(bp)	((bp)->b_flags |= XBF_READ)
-#define XFS_BUF_UNREAD(bp)	((bp)->b_flags &= ~XBF_READ)
-#define XFS_BUF_ISREAD(bp)	((bp)->b_flags & XBF_READ)
-
-#define XFS_BUF_WRITE(bp)	((bp)->b_flags |= XBF_WRITE)
-#define XFS_BUF_UNWRITE(bp)	((bp)->b_flags &= ~XBF_WRITE)
-#define XFS_BUF_ISWRITE(bp)	((bp)->b_flags & XBF_WRITE)
-
 /*
  * These macros use the IO block map rather than b_bn. b_bn is now really
  * just for the buffer cache index for cached buffers. As IO does not use b_bn
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index 7e986da34f6c..99e91a0e554e 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -431,7 +431,7 @@ xfs_buf_item_unpin(
 	if (freed && stale) {
 		ASSERT(bip->bli_flags & XFS_BLI_STALE);
 		ASSERT(xfs_buf_islocked(bp));
-		ASSERT(XFS_BUF_ISSTALE(bp));
+		ASSERT(bp->b_flags & XBF_STALE);
 		ASSERT(bip->__bli_format.blf_flags & XFS_BLF_CANCEL);
 
 		trace_xfs_buf_item_unpin_stale(bip);
@@ -493,7 +493,7 @@ xfs_buf_item_unpin(
 		xfs_buf_hold(bp);
 		bp->b_flags |= XBF_ASYNC;
 		xfs_buf_ioerror(bp, -EIO);
-		XFS_BUF_UNDONE(bp);
+		bp->b_flags &= ~XBF_DONE;
 		xfs_buf_stale(bp);
 		xfs_buf_ioend(bp);
 	}
@@ -1067,7 +1067,7 @@ xfs_buf_iodone_callbacks(
 	 */
 	if (XFS_FORCED_SHUTDOWN(mp)) {
 		xfs_buf_stale(bp);
-		XFS_BUF_DONE(bp);
+		bp->b_flags |= XBF_DONE;
 		trace_xfs_buf_item_iodone(bp, _RET_IP_);
 		goto do_callbacks;
 	}
@@ -1090,7 +1090,7 @@ xfs_buf_iodone_callbacks(
 	 * errors tend to affect the whole device and a failing log write
 	 * will make us give up.  But we really ought to do better here.
 	 */
-	if (XFS_BUF_ISASYNC(bp)) {
+	if (bp->b_flags & XBF_ASYNC) {
 		ASSERT(bp->b_iodone != NULL);
 
 		trace_xfs_buf_item_iodone_async(bp, _RET_IP_);
@@ -1113,7 +1113,7 @@ xfs_buf_iodone_callbacks(
 	 * sure to return the error to the caller of xfs_bwrite().
 	 */
 	xfs_buf_stale(bp);
-	XFS_BUF_DONE(bp);
+	bp->b_flags |= XBF_DONE;
 
 	trace_xfs_buf_error_relse(bp, _RET_IP_);
 
diff --git a/fs/xfs/xfs_dir2_readdir.c b/fs/xfs/xfs_dir2_readdir.c
index a989a9c7edb7..93b3ab0c5435 100644
--- a/fs/xfs/xfs_dir2_readdir.c
+++ b/fs/xfs/xfs_dir2_readdir.c
@@ -665,8 +665,8 @@ xfs_readdir(
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
 		return -EIO;
 
-	ASSERT(S_ISDIR(dp->i_d.di_mode));
-	XFS_STATS_INC(xs_dir_getdents);
+	ASSERT(S_ISDIR(VFS_I(dp)->i_mode));
+	XFS_STATS_INC(dp->i_mount, xs_dir_getdents);
 
 	args.dp = dp;
 	args.geo = dp->i_mount->m_dir_geo;
diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c
index e85a9519a5ae..272c3f8b6f7d 100644
--- a/fs/xfs/xfs_discard.c
+++ b/fs/xfs/xfs_discard.c
@@ -227,7 +227,7 @@ xfs_discard_extents(
 				GFP_NOFS, 0);
 		if (error && error != -EOPNOTSUPP) {
 			xfs_info(mp,
-	 "discard failed for extent [0x%llu,%u], error %d",
+	 "discard failed for extent [0x%llx,%u], error %d",
 				 (unsigned long long)busyp->bno,
 				 busyp->length,
 				 error);
diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c
index 30cb3afb67f0..316b2a1bdba5 100644
--- a/fs/xfs/xfs_dquot.c
+++ b/fs/xfs/xfs_dquot.c
@@ -75,9 +75,9 @@ xfs_qm_dqdestroy(
 	ASSERT(list_empty(&dqp->q_lru));
 
 	mutex_destroy(&dqp->q_qlock);
-	kmem_zone_free(xfs_qm_dqzone, dqp);
 
-	XFS_STATS_DEC(xs_qm_dquot);
+	XFS_STATS_DEC(dqp->q_mount, xs_qm_dquot);
+	kmem_zone_free(xfs_qm_dqzone, dqp);
 }
 
 /*
@@ -92,26 +92,28 @@ xfs_qm_adjust_dqlimits(
 {
 	struct xfs_quotainfo	*q = mp->m_quotainfo;
 	struct xfs_disk_dquot	*d = &dq->q_core;
+	struct xfs_def_quota	*defq;
 	int			prealloc = 0;
 
 	ASSERT(d->d_id);
+	defq = xfs_get_defquota(dq, q);
 
-	if (q->qi_bsoftlimit && !d->d_blk_softlimit) {
-		d->d_blk_softlimit = cpu_to_be64(q->qi_bsoftlimit);
+	if (defq->bsoftlimit && !d->d_blk_softlimit) {
+		d->d_blk_softlimit = cpu_to_be64(defq->bsoftlimit);
 		prealloc = 1;
 	}
-	if (q->qi_bhardlimit && !d->d_blk_hardlimit) {
-		d->d_blk_hardlimit = cpu_to_be64(q->qi_bhardlimit);
+	if (defq->bhardlimit && !d->d_blk_hardlimit) {
+		d->d_blk_hardlimit = cpu_to_be64(defq->bhardlimit);
 		prealloc = 1;
 	}
-	if (q->qi_isoftlimit && !d->d_ino_softlimit)
-		d->d_ino_softlimit = cpu_to_be64(q->qi_isoftlimit);
-	if (q->qi_ihardlimit && !d->d_ino_hardlimit)
-		d->d_ino_hardlimit = cpu_to_be64(q->qi_ihardlimit);
-	if (q->qi_rtbsoftlimit && !d->d_rtb_softlimit)
-		d->d_rtb_softlimit = cpu_to_be64(q->qi_rtbsoftlimit);
-	if (q->qi_rtbhardlimit && !d->d_rtb_hardlimit)
-		d->d_rtb_hardlimit = cpu_to_be64(q->qi_rtbhardlimit);
+	if (defq->isoftlimit && !d->d_ino_softlimit)
+		d->d_ino_softlimit = cpu_to_be64(defq->isoftlimit);
+	if (defq->ihardlimit && !d->d_ino_hardlimit)
+		d->d_ino_hardlimit = cpu_to_be64(defq->ihardlimit);
+	if (defq->rtbsoftlimit && !d->d_rtb_softlimit)
+		d->d_rtb_softlimit = cpu_to_be64(defq->rtbsoftlimit);
+	if (defq->rtbhardlimit && !d->d_rtb_hardlimit)
+		d->d_rtb_hardlimit = cpu_to_be64(defq->rtbhardlimit);
 
 	if (prealloc)
 		xfs_dquot_set_prealloc_limits(dq);
@@ -232,7 +234,8 @@ xfs_qm_init_dquot_blk(
 {
 	struct xfs_quotainfo	*q = mp->m_quotainfo;
 	xfs_dqblk_t	*d;
-	int		curid, i;
+	xfs_dqid_t	curid;
+	int		i;
 
 	ASSERT(tp);
 	ASSERT(xfs_buf_islocked(bp));
@@ -243,7 +246,6 @@ xfs_qm_init_dquot_blk(
 	 * ID of the first dquot in the block - id's are zero based.
 	 */
 	curid = id - (id % q->qi_dqperchunk);
-	ASSERT(curid >= 0);
 	memset(d, 0, BBTOB(q->qi_dqchunklen));
 	for (i = 0; i < q->qi_dqperchunk; i++, d++, curid++) {
 		d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC);
@@ -306,7 +308,7 @@ xfs_qm_dqalloc(
 	xfs_fsblock_t	firstblock;
 	xfs_bmap_free_t flist;
 	xfs_bmbt_irec_t map;
-	int		nmaps, error, committed;
+	int		nmaps, error;
 	xfs_buf_t	*bp;
 	xfs_trans_t	*tp = *tpp;
 
@@ -379,11 +381,12 @@ xfs_qm_dqalloc(
 
 	xfs_trans_bhold(tp, bp);
 
-	if ((error = xfs_bmap_finish(tpp, &flist, &committed))) {
+	error = xfs_bmap_finish(tpp, &flist, NULL);
+	if (error)
 		goto error1;
-	}
 
-	if (committed) {
+	/* Transaction was committed? */
+	if (*tpp != tp) {
 		tp = *tpp;
 		xfs_trans_bjoin(tp, bp);
 	} else {
@@ -393,9 +396,9 @@ xfs_qm_dqalloc(
 	*O_bpp = bp;
 	return 0;
 
-      error1:
+error1:
 	xfs_bmap_cancel(&flist);
-      error0:
+error0:
 	xfs_iunlock(quotip, XFS_ILOCK_EXCL);
 
 	return error;
@@ -463,12 +466,13 @@ xfs_qm_dqtobp(
 	struct xfs_bmbt_irec	map;
 	int			nmaps = 1, error;
 	struct xfs_buf		*bp;
-	struct xfs_inode	*quotip = xfs_dq_to_quota_inode(dqp);
+	struct xfs_inode	*quotip;
 	struct xfs_mount	*mp = dqp->q_mount;
 	xfs_dqid_t		id = be32_to_cpu(dqp->q_core.d_id);
 	struct xfs_trans	*tp = (tpp ? *tpp : NULL);
 	uint			lock_mode;
 
+	quotip = xfs_quota_inode(dqp->q_mount, dqp->dq_flags);
 	dqp->q_fileoffset = (xfs_fileoff_t)id / mp->m_quotainfo->qi_dqperchunk;
 
 	lock_mode = xfs_ilock_data_map_shared(quotip);
@@ -605,7 +609,7 @@ xfs_qm_dqread(
 		break;
 	}
 
-	XFS_STATS_INC(xs_qm_dquot);
+	XFS_STATS_INC(mp, xs_qm_dquot);
 
 	trace_xfs_dqread(dqp);
 
@@ -684,6 +688,56 @@ error0:
 }
 
 /*
+ * Advance to the next id in the current chunk, or if at the
+ * end of the chunk, skip ahead to first id in next allocated chunk
+ * using the SEEK_DATA interface.
+ */
+int
+xfs_dq_get_next_id(
+	xfs_mount_t		*mp,
+	uint			type,
+	xfs_dqid_t		*id,
+	loff_t			eof)
+{
+	struct xfs_inode	*quotip;
+	xfs_fsblock_t		start;
+	loff_t			offset;
+	uint			lock;
+	xfs_dqid_t		next_id;
+	int			error = 0;
+
+	/* Simple advance */
+	next_id = *id + 1;
+
+	/* If new ID is within the current chunk, advancing it sufficed */
+	if (next_id % mp->m_quotainfo->qi_dqperchunk) {
+		*id = next_id;
+		return 0;
+	}
+
+	/* Nope, next_id is now past the current chunk, so find the next one */
+	start = (xfs_fsblock_t)next_id / mp->m_quotainfo->qi_dqperchunk;
+
+	quotip = xfs_quota_inode(mp, type);
+	lock = xfs_ilock_data_map_shared(quotip);
+
+	offset = __xfs_seek_hole_data(VFS_I(quotip), XFS_FSB_TO_B(mp, start),
+				      eof, SEEK_DATA);
+	if (offset < 0)
+		error = offset;
+
+	xfs_iunlock(quotip, lock);
+
+	/* -ENXIO is essentially "no more data" */
+	if (error)
+		return (error == -ENXIO ? -ENOENT: error);
+
+	/* Convert next data offset back to a quota id */
+	*id = XFS_B_TO_FSB(mp, offset) * mp->m_quotainfo->qi_dqperchunk;
+	return 0;
+}
+
+/*
  * Given the file system, inode OR id, and type (UDQUOT/GDQUOT), return a
  * a locked dquot, doing an allocation (if requested) as needed.
  * When both an inode and an id are given, the inode's id takes precedence.
@@ -703,6 +757,7 @@ xfs_qm_dqget(
 	struct xfs_quotainfo	*qi = mp->m_quotainfo;
 	struct radix_tree_root *tree = xfs_dquot_tree(qi, type);
 	struct xfs_dquot	*dqp;
+	loff_t			eof = 0;
 	int			error;
 
 	ASSERT(XFS_IS_QUOTA_RUNNING(mp));
@@ -730,6 +785,21 @@ xfs_qm_dqget(
 	}
 #endif
 
+	/* Get the end of the quota file if we need it */
+	if (flags & XFS_QMOPT_DQNEXT) {
+		struct xfs_inode	*quotip;
+		xfs_fileoff_t		last;
+		uint			lock_mode;
+
+		quotip = xfs_quota_inode(mp, type);
+		lock_mode = xfs_ilock_data_map_shared(quotip);
+		error = xfs_bmap_last_offset(quotip, &last, XFS_DATA_FORK);
+		xfs_iunlock(quotip, lock_mode);
+		if (error)
+			return error;
+		eof = XFS_FSB_TO_B(mp, last);
+	}
+
 restart:
 	mutex_lock(&qi->qi_tree_lock);
 	dqp = radix_tree_lookup(tree, id);
@@ -743,16 +813,28 @@ restart:
 			goto restart;
 		}
 
+		/* uninit / unused quota found in radix tree, keep looking  */
+		if (flags & XFS_QMOPT_DQNEXT) {
+			if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) {
+				xfs_dqunlock(dqp);
+				mutex_unlock(&qi->qi_tree_lock);
+				error = xfs_dq_get_next_id(mp, type, &id, eof);
+				if (error)
+					return error;
+				goto restart;
+			}
+		}
+
 		dqp->q_nrefs++;
 		mutex_unlock(&qi->qi_tree_lock);
 
 		trace_xfs_dqget_hit(dqp);
-		XFS_STATS_INC(xs_qm_dqcachehits);
+		XFS_STATS_INC(mp, xs_qm_dqcachehits);
 		*O_dqpp = dqp;
 		return 0;
 	}
 	mutex_unlock(&qi->qi_tree_lock);
-	XFS_STATS_INC(xs_qm_dqcachemisses);
+	XFS_STATS_INC(mp, xs_qm_dqcachemisses);
 
 	/*
 	 * Dquot cache miss. We don't want to keep the inode lock across
@@ -769,6 +851,13 @@ restart:
 	if (ip)
 		xfs_ilock(ip, XFS_ILOCK_EXCL);
 
+	/* If we are asked to find next active id, keep looking */
+	if (error == -ENOENT && (flags & XFS_QMOPT_DQNEXT)) {
+		error = xfs_dq_get_next_id(mp, type, &id, eof);
+		if (!error)
+			goto restart;
+	}
+
 	if (error)
 		return error;
 
@@ -806,7 +895,7 @@ restart:
 		mutex_unlock(&qi->qi_tree_lock);
 		trace_xfs_dqget_dup(dqp);
 		xfs_qm_dqdestroy(dqp);
-		XFS_STATS_INC(xs_qm_dquot_dups);
+		XFS_STATS_INC(mp, xs_qm_dquot_dups);
 		goto restart;
 	}
 
@@ -819,6 +908,17 @@ restart:
 	qi->qi_dquots++;
 	mutex_unlock(&qi->qi_tree_lock);
 
+	/* If we are asked to find next active id, keep looking */
+	if (flags & XFS_QMOPT_DQNEXT) {
+		if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) {
+			xfs_qm_dqput(dqp);
+			error = xfs_dq_get_next_id(mp, type, &id, eof);
+			if (error)
+				return error;
+			goto restart;
+		}
+	}
+
  dqret:
 	ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	trace_xfs_dqget_miss(dqp);
@@ -846,7 +946,7 @@ xfs_qm_dqput(
 		trace_xfs_dqput_free(dqp);
 
 		if (list_lru_add(&qi->qi_lru, &dqp->q_lru))
-			XFS_STATS_INC(xs_qm_dquot_unused);
+			XFS_STATS_INC(dqp->q_mount, xs_qm_dquot_unused);
 	}
 	xfs_dqunlock(dqp);
 }
diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c
index 74d0e5966ebc..88693a98fac5 100644
--- a/fs/xfs/xfs_error.c
+++ b/fs/xfs/xfs_error.c
@@ -164,9 +164,9 @@ xfs_verifier_error(
 {
 	struct xfs_mount *mp = bp->b_target->bt_mount;
 
-	xfs_alert(mp, "Metadata %s detected at %pF, block 0x%llx",
+	xfs_alert(mp, "Metadata %s detected at %pF, %s block 0x%llx",
 		  bp->b_error == -EFSBADCRC ? "CRC error" : "corruption",
-		  __return_address, bp->b_bn);
+		  __return_address, bp->b_ops->name, bp->b_bn);
 
 	xfs_alert(mp, "Unmount and run xfs_repair");
 
diff --git a/fs/xfs/xfs_export.c b/fs/xfs/xfs_export.c
index 652cd3c5b58c..a1b2dd828b9d 100644
--- a/fs/xfs/xfs_export.c
+++ b/fs/xfs/xfs_export.c
@@ -152,7 +152,7 @@ xfs_nfs_get_inode(
 		return ERR_PTR(error);
 	}
 
-	if (ip->i_d.di_gen != generation) {
+	if (VFS_I(ip)->i_generation != generation) {
 		IRELE(ip);
 		return ERR_PTR(-ESTALE);
 	}
@@ -246,7 +246,7 @@ const struct export_operations xfs_export_operations = {
 	.fh_to_parent		= xfs_fs_fh_to_parent,
 	.get_parent		= xfs_fs_get_parent,
 	.commit_metadata	= xfs_fs_nfs_commit_metadata,
-#ifdef CONFIG_NFSD_PNFS
+#ifdef CONFIG_NFSD_BLOCKLAYOUT
 	.get_uuid		= xfs_fs_get_uuid,
 	.map_blocks		= xfs_fs_map_blocks,
 	.commit_blocks		= xfs_fs_commit_blocks,
diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index e78feb400e22..569938a4a357 100644
--- a/fs/xfs/xfs_file.c
+++ b/fs/xfs/xfs_file.c
@@ -55,7 +55,7 @@ xfs_rw_ilock(
 	int			type)
 {
 	if (type & XFS_IOLOCK_EXCL)
-		mutex_lock(&VFS_I(ip)->i_mutex);
+		inode_lock(VFS_I(ip));
 	xfs_ilock(ip, type);
 }
 
@@ -66,7 +66,7 @@ xfs_rw_iunlock(
 {
 	xfs_iunlock(ip, type);
 	if (type & XFS_IOLOCK_EXCL)
-		mutex_unlock(&VFS_I(ip)->i_mutex);
+		inode_unlock(VFS_I(ip));
 }
 
 static inline void
@@ -76,7 +76,7 @@ xfs_rw_ilock_demote(
 {
 	xfs_ilock_demote(ip, type);
 	if (type & XFS_IOLOCK_EXCL)
-		mutex_unlock(&VFS_I(ip)->i_mutex);
+		inode_unlock(VFS_I(ip));
 }
 
 /*
@@ -106,8 +106,8 @@ xfs_iozero(
 		unsigned offset, bytes;
 		void *fsdata;
 
-		offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */
-		bytes = PAGE_CACHE_SIZE - offset;
+		offset = (pos & (PAGE_SIZE -1)); /* Within page */
+		bytes = PAGE_SIZE - offset;
 		if (bytes > count)
 			bytes = count;
 
@@ -156,9 +156,9 @@ xfs_update_prealloc_flags(
 	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
 
 	if (!(flags & XFS_PREALLOC_INVISIBLE)) {
-		ip->i_d.di_mode &= ~S_ISUID;
-		if (ip->i_d.di_mode & S_IXGRP)
-			ip->i_d.di_mode &= ~S_ISGID;
+		VFS_I(ip)->i_mode &= ~S_ISUID;
+		if (VFS_I(ip)->i_mode & S_IXGRP)
+			VFS_I(ip)->i_mode &= ~S_ISGID;
 		xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
 	}
 
@@ -242,19 +242,30 @@ xfs_file_fsync(
 	}
 
 	/*
-	 * All metadata updates are logged, which means that we just have
-	 * to flush the log up to the latest LSN that touched the inode.
+	 * All metadata updates are logged, which means that we just have to
+	 * flush the log up to the latest LSN that touched the inode. If we have
+	 * concurrent fsync/fdatasync() calls, we need them to all block on the
+	 * log force before we clear the ili_fsync_fields field. This ensures
+	 * that we don't get a racing sync operation that does not wait for the
+	 * metadata to hit the journal before returning. If we race with
+	 * clearing the ili_fsync_fields, then all that will happen is the log
+	 * force will do nothing as the lsn will already be on disk. We can't
+	 * race with setting ili_fsync_fields because that is done under
+	 * XFS_ILOCK_EXCL, and that can't happen because we hold the lock shared
+	 * until after the ili_fsync_fields is cleared.
 	 */
 	xfs_ilock(ip, XFS_ILOCK_SHARED);
 	if (xfs_ipincount(ip)) {
 		if (!datasync ||
-		    (ip->i_itemp->ili_fields & ~XFS_ILOG_TIMESTAMP))
+		    (ip->i_itemp->ili_fsync_fields & ~XFS_ILOG_TIMESTAMP))
 			lsn = ip->i_itemp->ili_last_lsn;
 	}
-	xfs_iunlock(ip, XFS_ILOCK_SHARED);
 
-	if (lsn)
+	if (lsn) {
 		error = _xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, &log_flushed);
+		ip->i_itemp->ili_fsync_fields = 0;
+	}
+	xfs_iunlock(ip, XFS_ILOCK_SHARED);
 
 	/*
 	 * If we only have a single device, and the log force about was
@@ -287,7 +298,7 @@ xfs_file_read_iter(
 	xfs_fsize_t		n;
 	loff_t			pos = iocb->ki_pos;
 
-	XFS_STATS_INC(xs_read_calls);
+	XFS_STATS_INC(mp, xs_read_calls);
 
 	if (unlikely(iocb->ki_flags & IOCB_DIRECT))
 		ioflags |= XFS_IO_ISDIRECT;
@@ -365,7 +376,7 @@ xfs_file_read_iter(
 
 	ret = generic_file_read_iter(iocb, to);
 	if (ret > 0)
-		XFS_STATS_ADD(xs_read_bytes, ret);
+		XFS_STATS_ADD(mp, xs_read_bytes, ret);
 
 	xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
 	return ret;
@@ -383,7 +394,7 @@ xfs_file_splice_read(
 	int			ioflags = 0;
 	ssize_t			ret;
 
-	XFS_STATS_INC(xs_read_calls);
+	XFS_STATS_INC(ip->i_mount, xs_read_calls);
 
 	if (infilp->f_mode & FMODE_NOCMTIME)
 		ioflags |= XFS_IO_INVIS;
@@ -391,19 +402,26 @@ xfs_file_splice_read(
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
 		return -EIO;
 
-	xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
-
 	trace_xfs_file_splice_read(ip, count, *ppos, ioflags);
 
-	/* for dax, we need to avoid the page cache */
-	if (IS_DAX(VFS_I(ip)))
-		ret = default_file_splice_read(infilp, ppos, pipe, count, flags);
-	else
-		ret = generic_file_splice_read(infilp, ppos, pipe, count, flags);
-	if (ret > 0)
-		XFS_STATS_ADD(xs_read_bytes, ret);
+	/*
+	 * DAX inodes cannot ues the page cache for splice, so we have to push
+	 * them through the VFS IO path. This means it goes through
+	 * ->read_iter, which for us takes the XFS_IOLOCK_SHARED. Hence we
+	 * cannot lock the splice operation at this level for DAX inodes.
+	 */
+	if (IS_DAX(VFS_I(ip))) {
+		ret = default_file_splice_read(infilp, ppos, pipe, count,
+					       flags);
+		goto out;
+	}
 
+	xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
+	ret = generic_file_splice_read(infilp, ppos, pipe, count, flags);
 	xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
+out:
+	if (ret > 0)
+		XFS_STATS_ADD(ip->i_mount, xs_read_bytes, ret);
 	return ret;
 }
 
@@ -482,6 +500,8 @@ xfs_zero_eof(
 	ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
 	ASSERT(offset > isize);
 
+	trace_xfs_zero_eof(ip, isize, offset - isize);
+
 	/*
 	 * First handle zeroing the block on which isize resides.
 	 *
@@ -574,6 +594,7 @@ xfs_file_aio_write_checks(
 	struct xfs_inode	*ip = XFS_I(inode);
 	ssize_t			error = 0;
 	size_t			count = iov_iter_count(from);
+	bool			drained_dio = false;
 
 restart:
 	error = generic_write_checks(iocb, from);
@@ -611,12 +632,13 @@ restart:
 		bool	zero = false;
 
 		spin_unlock(&ip->i_flags_lock);
-		if (*iolock == XFS_IOLOCK_SHARED) {
-			xfs_rw_iunlock(ip, *iolock);
-			*iolock = XFS_IOLOCK_EXCL;
-			xfs_rw_ilock(ip, *iolock);
-			iov_iter_reexpand(from, count);
-
+		if (!drained_dio) {
+			if (*iolock == XFS_IOLOCK_SHARED) {
+				xfs_rw_iunlock(ip, *iolock);
+				*iolock = XFS_IOLOCK_EXCL;
+				xfs_rw_ilock(ip, *iolock);
+				iov_iter_reexpand(from, count);
+			}
 			/*
 			 * We now have an IO submission barrier in place, but
 			 * AIO can do EOF updates during IO completion and hence
@@ -626,6 +648,7 @@ restart:
 			 * no-op.
 			 */
 			inode_dio_wait(inode);
+			drained_dio = true;
 			goto restart;
 		}
 		error = xfs_zero_eof(ip, iocb->ki_pos, i_size_read(inode), &zero);
@@ -776,8 +799,8 @@ xfs_file_dio_aio_write(
 	/* see generic_file_direct_write() for why this is necessary */
 	if (mapping->nrpages) {
 		invalidate_inode_pages2_range(mapping,
-					      pos >> PAGE_CACHE_SHIFT,
-					      end >> PAGE_CACHE_SHIFT);
+					      pos >> PAGE_SHIFT,
+					      end >> PAGE_SHIFT);
 	}
 
 	if (ret > 0) {
@@ -867,7 +890,7 @@ xfs_file_write_iter(
 	ssize_t			ret;
 	size_t			ocount = iov_iter_count(from);
 
-	XFS_STATS_INC(xs_write_calls);
+	XFS_STATS_INC(ip->i_mount, xs_write_calls);
 
 	if (ocount == 0)
 		return 0;
@@ -883,7 +906,7 @@ xfs_file_write_iter(
 	if (ret > 0) {
 		ssize_t err;
 
-		XFS_STATS_ADD(xs_write_bytes, ret);
+		XFS_STATS_ADD(ip->i_mount, xs_write_bytes, ret);
 
 		/* Handle various SYNC-type writes */
 		err = generic_write_sync(file, iocb->ki_pos - ret, ret);
@@ -1184,9 +1207,9 @@ xfs_find_get_desired_pgoff(
 
 	pagevec_init(&pvec, 0);
 
-	index = startoff >> PAGE_CACHE_SHIFT;
+	index = startoff >> PAGE_SHIFT;
 	endoff = XFS_FSB_TO_B(mp, map->br_startoff + map->br_blockcount);
-	end = endoff >> PAGE_CACHE_SHIFT;
+	end = endoff >> PAGE_SHIFT;
 	do {
 		int		want;
 		unsigned	nr_pages;
@@ -1314,31 +1337,31 @@ out:
 	return found;
 }
 
-STATIC loff_t
-xfs_seek_hole_data(
-	struct file		*file,
+/*
+ * caller must lock inode with xfs_ilock_data_map_shared,
+ * can we craft an appropriate ASSERT?
+ *
+ * end is because the VFS-level lseek interface is defined such that any
+ * offset past i_size shall return -ENXIO, but we use this for quota code
+ * which does not maintain i_size, and we want to SEEK_DATA past i_size.
+ */
+loff_t
+__xfs_seek_hole_data(
+	struct inode		*inode,
 	loff_t			start,
+	loff_t			end,
 	int			whence)
 {
-	struct inode		*inode = file->f_mapping->host;
 	struct xfs_inode	*ip = XFS_I(inode);
 	struct xfs_mount	*mp = ip->i_mount;
 	loff_t			uninitialized_var(offset);
-	xfs_fsize_t		isize;
 	xfs_fileoff_t		fsbno;
-	xfs_filblks_t		end;
-	uint			lock;
+	xfs_filblks_t		lastbno;
 	int			error;
 
-	if (XFS_FORCED_SHUTDOWN(mp))
-		return -EIO;
-
-	lock = xfs_ilock_data_map_shared(ip);
-
-	isize = i_size_read(inode);
-	if (start >= isize) {
+	if (start >= end) {
 		error = -ENXIO;
-		goto out_unlock;
+		goto out_error;
 	}
 
 	/*
@@ -1346,22 +1369,22 @@ xfs_seek_hole_data(
 	 * by fsbno to the end block of the file.
 	 */
 	fsbno = XFS_B_TO_FSBT(mp, start);
-	end = XFS_B_TO_FSB(mp, isize);
+	lastbno = XFS_B_TO_FSB(mp, end);
 
 	for (;;) {
 		struct xfs_bmbt_irec	map[2];
 		int			nmap = 2;
 		unsigned int		i;
 
-		error = xfs_bmapi_read(ip, fsbno, end - fsbno, map, &nmap,
+		error = xfs_bmapi_read(ip, fsbno, lastbno - fsbno, map, &nmap,
 				       XFS_BMAPI_ENTIRE);
 		if (error)
-			goto out_unlock;
+			goto out_error;
 
 		/* No extents at given offset, must be beyond EOF */
 		if (nmap == 0) {
 			error = -ENXIO;
-			goto out_unlock;
+			goto out_error;
 		}
 
 		for (i = 0; i < nmap; i++) {
@@ -1403,7 +1426,7 @@ xfs_seek_hole_data(
 			 * hole at the end of any file).
 		 	 */
 			if (whence == SEEK_HOLE) {
-				offset = isize;
+				offset = end;
 				break;
 			}
 			/*
@@ -1411,7 +1434,7 @@ xfs_seek_hole_data(
 			 */
 			ASSERT(whence == SEEK_DATA);
 			error = -ENXIO;
-			goto out_unlock;
+			goto out_error;
 		}
 
 		ASSERT(i > 1);
@@ -1422,14 +1445,14 @@ xfs_seek_hole_data(
 		 */
 		fsbno = map[i - 1].br_startoff + map[i - 1].br_blockcount;
 		start = XFS_FSB_TO_B(mp, fsbno);
-		if (start >= isize) {
+		if (start >= end) {
 			if (whence == SEEK_HOLE) {
-				offset = isize;
+				offset = end;
 				break;
 			}
 			ASSERT(whence == SEEK_DATA);
 			error = -ENXIO;
-			goto out_unlock;
+			goto out_error;
 		}
 	}
 
@@ -1441,7 +1464,39 @@ out:
 	 * situation in particular.
 	 */
 	if (whence == SEEK_HOLE)
-		offset = min_t(loff_t, offset, isize);
+		offset = min_t(loff_t, offset, end);
+
+	return offset;
+
+out_error:
+	return error;
+}
+
+STATIC loff_t
+xfs_seek_hole_data(
+	struct file		*file,
+	loff_t			start,
+	int			whence)
+{
+	struct inode		*inode = file->f_mapping->host;
+	struct xfs_inode	*ip = XFS_I(inode);
+	struct xfs_mount	*mp = ip->i_mount;
+	uint			lock;
+	loff_t			offset, end;
+	int			error = 0;
+
+	if (XFS_FORCED_SHUTDOWN(mp))
+		return -EIO;
+
+	lock = xfs_ilock_data_map_shared(ip);
+
+	end = i_size_read(inode);
+	offset = __xfs_seek_hole_data(inode, start, end, whence);
+	if (offset < 0) {
+		error = offset;
+		goto out_unlock;
+	}
+
 	offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes);
 
 out_unlock:
@@ -1477,7 +1532,7 @@ xfs_file_llseek(
  *
  * mmap_sem (MM)
  *   sb_start_pagefault(vfs, freeze)
- *     i_mmap_lock (XFS - truncate serialisation)
+ *     i_mmaplock (XFS - truncate serialisation)
  *       page_lock (MM)
  *         i_lock (XFS - extent map serialisation)
  */
@@ -1503,10 +1558,9 @@ xfs_filemap_page_mkwrite(
 	xfs_ilock(XFS_I(inode), XFS_MMAPLOCK_SHARED);
 
 	if (IS_DAX(inode)) {
-		ret = __dax_mkwrite(vma, vmf, xfs_get_blocks_direct,
-				    xfs_end_io_dax_write);
+		ret = __dax_mkwrite(vma, vmf, xfs_get_blocks_dax_fault, NULL);
 	} else {
-		ret = __block_page_mkwrite(vma, vmf, xfs_get_blocks);
+		ret = block_page_mkwrite(vma, vmf, xfs_get_blocks);
 		ret = block_page_mkwrite_return(ret);
 	}
 
@@ -1538,7 +1592,7 @@ xfs_filemap_fault(
 		 * changes to xfs_get_blocks_direct() to map unwritten extent
 		 * ioend for conversion on read-only mappings.
 		 */
-		ret = __dax_fault(vma, vmf, xfs_get_blocks_direct, NULL);
+		ret = __dax_fault(vma, vmf, xfs_get_blocks_dax_fault, NULL);
 	} else
 		ret = filemap_fault(vma, vmf);
 	xfs_iunlock(XFS_I(inode), XFS_MMAPLOCK_SHARED);
@@ -1546,6 +1600,13 @@ xfs_filemap_fault(
 	return ret;
 }
 
+/*
+ * Similar to xfs_filemap_fault(), the DAX fault path can call into here on
+ * both read and write faults. Hence we need to handle both cases. There is no
+ * ->pmd_mkwrite callout for huge pages, so we have a single function here to
+ * handle both cases here. @flags carries the information on the type of fault
+ * occuring.
+ */
 STATIC int
 xfs_filemap_pmd_fault(
 	struct vm_area_struct	*vma,
@@ -1562,15 +1623,55 @@ xfs_filemap_pmd_fault(
 
 	trace_xfs_filemap_pmd_fault(ip);
 
-	sb_start_pagefault(inode->i_sb);
-	file_update_time(vma->vm_file);
+	if (flags & FAULT_FLAG_WRITE) {
+		sb_start_pagefault(inode->i_sb);
+		file_update_time(vma->vm_file);
+	}
+
 	xfs_ilock(XFS_I(inode), XFS_MMAPLOCK_SHARED);
-	ret = __dax_pmd_fault(vma, addr, pmd, flags, xfs_get_blocks_direct,
-				    xfs_end_io_dax_write);
+	ret = __dax_pmd_fault(vma, addr, pmd, flags, xfs_get_blocks_dax_fault,
+			      NULL);
 	xfs_iunlock(XFS_I(inode), XFS_MMAPLOCK_SHARED);
-	sb_end_pagefault(inode->i_sb);
 
+	if (flags & FAULT_FLAG_WRITE)
+		sb_end_pagefault(inode->i_sb);
+
+	return ret;
+}
+
+/*
+ * pfn_mkwrite was originally inteneded to ensure we capture time stamp
+ * updates on write faults. In reality, it's need to serialise against
+ * truncate similar to page_mkwrite. Hence we cycle the XFS_MMAPLOCK_SHARED
+ * to ensure we serialise the fault barrier in place.
+ */
+static int
+xfs_filemap_pfn_mkwrite(
+	struct vm_area_struct	*vma,
+	struct vm_fault		*vmf)
+{
+
+	struct inode		*inode = file_inode(vma->vm_file);
+	struct xfs_inode	*ip = XFS_I(inode);
+	int			ret = VM_FAULT_NOPAGE;
+	loff_t			size;
+
+	trace_xfs_filemap_pfn_mkwrite(ip);
+
+	sb_start_pagefault(inode->i_sb);
+	file_update_time(vma->vm_file);
+
+	/* check if the faulting page hasn't raced with truncate */
+	xfs_ilock(ip, XFS_MMAPLOCK_SHARED);
+	size = (i_size_read(inode) + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	if (vmf->pgoff >= size)
+		ret = VM_FAULT_SIGBUS;
+	else if (IS_DAX(inode))
+		ret = dax_pfn_mkwrite(vma, vmf);
+	xfs_iunlock(ip, XFS_MMAPLOCK_SHARED);
+	sb_end_pagefault(inode->i_sb);
 	return ret;
+
 }
 
 static const struct vm_operations_struct xfs_file_vm_ops = {
@@ -1578,6 +1679,7 @@ static const struct vm_operations_struct xfs_file_vm_ops = {
 	.pmd_fault	= xfs_filemap_pmd_fault,
 	.map_pages	= filemap_map_pages,
 	.page_mkwrite	= xfs_filemap_page_mkwrite,
+	.pfn_mkwrite	= xfs_filemap_pfn_mkwrite,
 };
 
 STATIC int
diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c
index c4c130f9bfb6..a51353a1f87f 100644
--- a/fs/xfs/xfs_filestream.c
+++ b/fs/xfs/xfs_filestream.c
@@ -151,7 +151,7 @@ xfs_filestream_pick_ag(
 	xfs_agnumber_t		ag, max_ag = NULLAGNUMBER;
 	int			err, trylock, nscan;
 
-	ASSERT(S_ISDIR(ip->i_d.di_mode));
+	ASSERT(S_ISDIR(VFS_I(ip)->i_mode));
 
 	/* 2% of an AG's blocks must be free for it to be chosen. */
 	minfree = mp->m_sb.sb_agblocks / 50;
@@ -319,7 +319,7 @@ xfs_filestream_lookup_ag(
 	xfs_agnumber_t		startag, ag = NULLAGNUMBER;
 	struct xfs_mru_cache_elem *mru;
 
-	ASSERT(S_ISREG(ip->i_d.di_mode));
+	ASSERT(S_ISREG(VFS_I(ip)->i_mode));
 
 	pip = xfs_filestream_get_parent(ip);
 	if (!pip)
diff --git a/fs/xfs/xfs_fsops.h b/fs/xfs/xfs_fsops.h
index 1b6a98b66886..f32713f14f9a 100644
--- a/fs/xfs/xfs_fsops.h
+++ b/fs/xfs/xfs_fsops.h
@@ -25,6 +25,5 @@ extern int xfs_fs_counts(xfs_mount_t *mp, xfs_fsop_counts_t *cnt);
 extern int xfs_reserve_blocks(xfs_mount_t *mp, __uint64_t *inval,
 				xfs_fsop_resblks_t *outval);
 extern int xfs_fs_goingdown(xfs_mount_t *mp, __uint32_t inflags);
-extern int xfs_fs_log_dummy(struct xfs_mount *mp);
 
 #endif	/* __XFS_FSOPS_H__ */
diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c
index 0a326bd64d4e..bf2d60749278 100644
--- a/fs/xfs/xfs_icache.c
+++ b/fs/xfs/xfs_icache.c
@@ -63,7 +63,10 @@ xfs_inode_alloc(
 		return NULL;
 	}
 
-	XFS_STATS_INC(vn_active);
+	/* VFS doesn't initialise i_mode! */
+	VFS_I(ip)->i_mode = 0;
+
+	XFS_STATS_INC(mp, vn_active);
 	ASSERT(atomic_read(&ip->i_pincount) == 0);
 	ASSERT(!spin_is_locked(&ip->i_flags_lock));
 	ASSERT(!xfs_isiflocked(ip));
@@ -79,7 +82,7 @@ xfs_inode_alloc(
 	memset(&ip->i_df, 0, sizeof(xfs_ifork_t));
 	ip->i_flags = 0;
 	ip->i_delayed_blks = 0;
-	memset(&ip->i_d, 0, sizeof(xfs_icdinode_t));
+	memset(&ip->i_d, 0, sizeof(ip->i_d));
 
 	return ip;
 }
@@ -98,7 +101,7 @@ void
 xfs_inode_free(
 	struct xfs_inode	*ip)
 {
-	switch (ip->i_d.di_mode & S_IFMT) {
+	switch (VFS_I(ip)->i_mode & S_IFMT) {
 	case S_IFREG:
 	case S_IFDIR:
 	case S_IFLNK:
@@ -129,12 +132,40 @@ xfs_inode_free(
 	/* asserts to verify all state is correct here */
 	ASSERT(atomic_read(&ip->i_pincount) == 0);
 	ASSERT(!xfs_isiflocked(ip));
-	XFS_STATS_DEC(vn_active);
+	XFS_STATS_DEC(ip->i_mount, vn_active);
 
 	call_rcu(&VFS_I(ip)->i_rcu, xfs_inode_free_callback);
 }
 
 /*
+ * 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
+ * information about the on-disk values in the VFS inode and so we can't just
+ * overwrite the values unconditionally. Hence we save the parameters we
+ * need to retain across reinitialisation, and rewrite them into the VFS inode
+ * after reinitialisation even if it fails.
+ */
+static int
+xfs_reinit_inode(
+	struct xfs_mount	*mp,
+	struct inode		*inode)
+{
+	int		error;
+	uint32_t	nlink = inode->i_nlink;
+	uint32_t	generation = inode->i_generation;
+	uint64_t	version = inode->i_version;
+	umode_t		mode = inode->i_mode;
+
+	error = inode_init_always(mp->m_super, inode);
+
+	set_nlink(inode, nlink);
+	inode->i_generation = generation;
+	inode->i_version = version;
+	inode->i_mode = mode;
+	return error;
+}
+
+/*
  * Check the validity of the inode we just found it the cache
  */
 static int
@@ -159,7 +190,7 @@ xfs_iget_cache_hit(
 	spin_lock(&ip->i_flags_lock);
 	if (ip->i_ino != ino) {
 		trace_xfs_iget_skip(ip);
-		XFS_STATS_INC(xs_ig_frecycle);
+		XFS_STATS_INC(mp, xs_ig_frecycle);
 		error = -EAGAIN;
 		goto out_error;
 	}
@@ -177,7 +208,7 @@ xfs_iget_cache_hit(
 	 */
 	if (ip->i_flags & (XFS_INEW|XFS_IRECLAIM)) {
 		trace_xfs_iget_skip(ip);
-		XFS_STATS_INC(xs_ig_frecycle);
+		XFS_STATS_INC(mp, xs_ig_frecycle);
 		error = -EAGAIN;
 		goto out_error;
 	}
@@ -185,7 +216,7 @@ xfs_iget_cache_hit(
 	/*
 	 * If lookup is racing with unlink return an error immediately.
 	 */
-	if (ip->i_d.di_mode == 0 && !(flags & XFS_IGET_CREATE)) {
+	if (VFS_I(ip)->i_mode == 0 && !(flags & XFS_IGET_CREATE)) {
 		error = -ENOENT;
 		goto out_error;
 	}
@@ -208,7 +239,7 @@ xfs_iget_cache_hit(
 		spin_unlock(&ip->i_flags_lock);
 		rcu_read_unlock();
 
-		error = inode_init_always(mp->m_super, inode);
+		error = xfs_reinit_inode(mp, inode);
 		if (error) {
 			/*
 			 * Re-initializing the inode failed, and we are in deep
@@ -259,7 +290,7 @@ xfs_iget_cache_hit(
 		xfs_ilock(ip, lock_flags);
 
 	xfs_iflags_clear(ip, XFS_ISTALE | XFS_IDONTCACHE);
-	XFS_STATS_INC(xs_ig_found);
+	XFS_STATS_INC(mp, xs_ig_found);
 
 	return 0;
 
@@ -295,7 +326,7 @@ xfs_iget_cache_miss(
 
 	trace_xfs_iget_miss(ip);
 
-	if ((ip->i_d.di_mode == 0) && !(flags & XFS_IGET_CREATE)) {
+	if ((VFS_I(ip)->i_mode == 0) && !(flags & XFS_IGET_CREATE)) {
 		error = -ENOENT;
 		goto out_destroy;
 	}
@@ -342,7 +373,7 @@ xfs_iget_cache_miss(
 	error = radix_tree_insert(&pag->pag_ici_root, agino, ip);
 	if (unlikely(error)) {
 		WARN_ON(error != -EEXIST);
-		XFS_STATS_INC(xs_ig_dup);
+		XFS_STATS_INC(mp, xs_ig_dup);
 		error = -EAGAIN;
 		goto out_preload_end;
 	}
@@ -412,7 +443,7 @@ xfs_iget(
 	if (!ino || XFS_INO_TO_AGNO(mp, ino) >= mp->m_sb.sb_agcount)
 		return -EINVAL;
 
-	XFS_STATS_INC(xs_ig_attempts);
+	XFS_STATS_INC(mp, xs_ig_attempts);
 
 	/* get the perag structure and ensure that it's inode capable */
 	pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ino));
@@ -429,7 +460,7 @@ again:
 			goto out_error_or_again;
 	} else {
 		rcu_read_unlock();
-		XFS_STATS_INC(xs_ig_missed);
+		XFS_STATS_INC(mp, xs_ig_missed);
 
 		error = xfs_iget_cache_miss(mp, pag, tp, ino, &ip,
 							flags, lock_flags);
@@ -444,7 +475,7 @@ again:
 	 * If we have a real type for an on-disk inode, we can setup the inode
 	 * now.	 If it's a new inode being created, xfs_ialloc will handle it.
 	 */
-	if (xfs_iflags_test(ip, XFS_INEW) && ip->i_d.di_mode != 0)
+	if (xfs_iflags_test(ip, XFS_INEW) && VFS_I(ip)->i_mode != 0)
 		xfs_setup_existing_inode(ip);
 	return 0;
 
@@ -965,7 +996,7 @@ reclaim:
 	xfs_ifunlock(ip);
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 
-	XFS_STATS_INC(xs_ig_reclaims);
+	XFS_STATS_INC(ip->i_mount, xs_ig_reclaims);
 	/*
 	 * Remove the inode from the per-AG radix tree.
 	 *
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index dc40a6d5ae0d..96f606deee31 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -57,9 +57,9 @@ kmem_zone_t *xfs_inode_zone;
  */
 #define	XFS_ITRUNC_MAX_EXTENTS	2
 
-STATIC int xfs_iflush_int(xfs_inode_t *, xfs_buf_t *);
-
-STATIC int xfs_iunlink_remove(xfs_trans_t *, xfs_inode_t *);
+STATIC int xfs_iflush_int(struct xfs_inode *, struct xfs_buf *);
+STATIC int xfs_iunlink(struct xfs_trans *, struct xfs_inode *);
+STATIC int xfs_iunlink_remove(struct xfs_trans *, struct xfs_inode *);
 
 /*
  * helper function to extract extent size hint from inode
@@ -610,60 +610,69 @@ __xfs_iflock(
 
 STATIC uint
 _xfs_dic2xflags(
-	__uint16_t		di_flags)
+	__uint16_t		di_flags,
+	uint64_t		di_flags2,
+	bool			has_attr)
 {
 	uint			flags = 0;
 
 	if (di_flags & XFS_DIFLAG_ANY) {
 		if (di_flags & XFS_DIFLAG_REALTIME)
-			flags |= XFS_XFLAG_REALTIME;
+			flags |= FS_XFLAG_REALTIME;
 		if (di_flags & XFS_DIFLAG_PREALLOC)
-			flags |= XFS_XFLAG_PREALLOC;
+			flags |= FS_XFLAG_PREALLOC;
 		if (di_flags & XFS_DIFLAG_IMMUTABLE)
-			flags |= XFS_XFLAG_IMMUTABLE;
+			flags |= FS_XFLAG_IMMUTABLE;
 		if (di_flags & XFS_DIFLAG_APPEND)
-			flags |= XFS_XFLAG_APPEND;
+			flags |= FS_XFLAG_APPEND;
 		if (di_flags & XFS_DIFLAG_SYNC)
-			flags |= XFS_XFLAG_SYNC;
+			flags |= FS_XFLAG_SYNC;
 		if (di_flags & XFS_DIFLAG_NOATIME)
-			flags |= XFS_XFLAG_NOATIME;
+			flags |= FS_XFLAG_NOATIME;
 		if (di_flags & XFS_DIFLAG_NODUMP)
-			flags |= XFS_XFLAG_NODUMP;
+			flags |= FS_XFLAG_NODUMP;
 		if (di_flags & XFS_DIFLAG_RTINHERIT)
-			flags |= XFS_XFLAG_RTINHERIT;
+			flags |= FS_XFLAG_RTINHERIT;
 		if (di_flags & XFS_DIFLAG_PROJINHERIT)
-			flags |= XFS_XFLAG_PROJINHERIT;
+			flags |= FS_XFLAG_PROJINHERIT;
 		if (di_flags & XFS_DIFLAG_NOSYMLINKS)
-			flags |= XFS_XFLAG_NOSYMLINKS;
+			flags |= FS_XFLAG_NOSYMLINKS;
 		if (di_flags & XFS_DIFLAG_EXTSIZE)
-			flags |= XFS_XFLAG_EXTSIZE;
+			flags |= FS_XFLAG_EXTSIZE;
 		if (di_flags & XFS_DIFLAG_EXTSZINHERIT)
-			flags |= XFS_XFLAG_EXTSZINHERIT;
+			flags |= FS_XFLAG_EXTSZINHERIT;
 		if (di_flags & XFS_DIFLAG_NODEFRAG)
-			flags |= XFS_XFLAG_NODEFRAG;
+			flags |= FS_XFLAG_NODEFRAG;
 		if (di_flags & XFS_DIFLAG_FILESTREAM)
-			flags |= XFS_XFLAG_FILESTREAM;
+			flags |= FS_XFLAG_FILESTREAM;
+	}
+
+	if (di_flags2 & XFS_DIFLAG2_ANY) {
+		if (di_flags2 & XFS_DIFLAG2_DAX)
+			flags |= FS_XFLAG_DAX;
 	}
 
+	if (has_attr)
+		flags |= FS_XFLAG_HASATTR;
+
 	return flags;
 }
 
 uint
 xfs_ip2xflags(
-	xfs_inode_t		*ip)
+	struct xfs_inode	*ip)
 {
-	xfs_icdinode_t		*dic = &ip->i_d;
+	struct xfs_icdinode	*dic = &ip->i_d;
 
-	return _xfs_dic2xflags(dic->di_flags) |
-				(XFS_IFORK_Q(ip) ? XFS_XFLAG_HASATTR : 0);
+	return _xfs_dic2xflags(dic->di_flags, dic->di_flags2, XFS_IFORK_Q(ip));
 }
 
 uint
 xfs_dic2xflags(
-	xfs_dinode_t		*dip)
+	struct xfs_dinode	*dip)
 {
-	return _xfs_dic2xflags(be16_to_cpu(dip->di_flags)) |
-				(XFS_DFORK_Q(dip) ? XFS_XFLAG_HASATTR : 0);
+	return _xfs_dic2xflags(be16_to_cpu(dip->di_flags),
+				be64_to_cpu(dip->di_flags2), XFS_DFORK_Q(dip));
 }
 
 /*
@@ -757,6 +766,7 @@ xfs_ialloc(
 	uint		flags;
 	int		error;
 	struct timespec	tv;
+	struct inode	*inode;
 
 	/*
 	 * Call the space management code to pick
@@ -782,6 +792,7 @@ xfs_ialloc(
 	if (error)
 		return error;
 	ASSERT(ip != NULL);
+	inode = VFS_I(ip);
 
 	/*
 	 * We always convert v1 inodes to v2 now - we only support filesystems
@@ -791,20 +802,16 @@ xfs_ialloc(
 	if (ip->i_d.di_version == 1)
 		ip->i_d.di_version = 2;
 
-	ip->i_d.di_mode = mode;
-	ip->i_d.di_onlink = 0;
-	ip->i_d.di_nlink = nlink;
-	ASSERT(ip->i_d.di_nlink == nlink);
+	inode->i_mode = mode;
+	set_nlink(inode, nlink);
 	ip->i_d.di_uid = xfs_kuid_to_uid(current_fsuid());
 	ip->i_d.di_gid = xfs_kgid_to_gid(current_fsgid());
 	xfs_set_projid(ip, prid);
-	memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad));
 
 	if (pip && XFS_INHERIT_GID(pip)) {
 		ip->i_d.di_gid = pip->i_d.di_gid;
-		if ((pip->i_d.di_mode & S_ISGID) && S_ISDIR(mode)) {
-			ip->i_d.di_mode |= S_ISGID;
-		}
+		if ((VFS_I(pip)->i_mode & S_ISGID) && S_ISDIR(mode))
+			inode->i_mode |= S_ISGID;
 	}
 
 	/*
@@ -813,38 +820,29 @@ xfs_ialloc(
 	 * (and only if the irix_sgid_inherit compatibility variable is set).
 	 */
 	if ((irix_sgid_inherit) &&
-	    (ip->i_d.di_mode & S_ISGID) &&
-	    (!in_group_p(xfs_gid_to_kgid(ip->i_d.di_gid)))) {
-		ip->i_d.di_mode &= ~S_ISGID;
-	}
+	    (inode->i_mode & S_ISGID) &&
+	    (!in_group_p(xfs_gid_to_kgid(ip->i_d.di_gid))))
+		inode->i_mode &= ~S_ISGID;
 
 	ip->i_d.di_size = 0;
 	ip->i_d.di_nextents = 0;
 	ASSERT(ip->i_d.di_nblocks == 0);
 
 	tv = current_fs_time(mp->m_super);
-	ip->i_d.di_mtime.t_sec = (__int32_t)tv.tv_sec;
-	ip->i_d.di_mtime.t_nsec = (__int32_t)tv.tv_nsec;
-	ip->i_d.di_atime = ip->i_d.di_mtime;
-	ip->i_d.di_ctime = ip->i_d.di_mtime;
+	inode->i_mtime = tv;
+	inode->i_atime = tv;
+	inode->i_ctime = tv;
 
-	/*
-	 * di_gen will have been taken care of in xfs_iread.
-	 */
 	ip->i_d.di_extsize = 0;
 	ip->i_d.di_dmevmask = 0;
 	ip->i_d.di_dmstate = 0;
 	ip->i_d.di_flags = 0;
 
 	if (ip->i_d.di_version == 3) {
-		ASSERT(ip->i_d.di_ino == ino);
-		ASSERT(uuid_equal(&ip->i_d.di_uuid, &mp->m_sb.sb_meta_uuid));
-		ip->i_d.di_crc = 0;
-		ip->i_d.di_changecount = 1;
-		ip->i_d.di_lsn = 0;
+		inode->i_version = 1;
 		ip->i_d.di_flags2 = 0;
-		memset(&(ip->i_d.di_pad2[0]), 0, sizeof(ip->i_d.di_pad2));
-		ip->i_d.di_crtime = ip->i_d.di_mtime;
+		ip->i_d.di_crtime.t_sec = (__int32_t)tv.tv_sec;
+		ip->i_d.di_crtime.t_nsec = (__int32_t)tv.tv_nsec;
 	}
 
 
@@ -862,7 +860,8 @@ xfs_ialloc(
 	case S_IFREG:
 	case S_IFDIR:
 		if (pip && (pip->i_d.di_flags & XFS_DIFLAG_ANY)) {
-			uint	di_flags = 0;
+			uint64_t	di_flags2 = 0;
+			uint		di_flags = 0;
 
 			if (S_ISDIR(mode)) {
 				if (pip->i_d.di_flags & XFS_DIFLAG_RTINHERIT)
@@ -898,7 +897,11 @@ xfs_ialloc(
 				di_flags |= XFS_DIFLAG_NODEFRAG;
 			if (pip->i_d.di_flags & XFS_DIFLAG_FILESTREAM)
 				di_flags |= XFS_DIFLAG_FILESTREAM;
+			if (pip->i_d.di_flags2 & XFS_DIFLAG2_DAX)
+				di_flags2 |= XFS_DIFLAG2_DAX;
+
 			ip->i_d.di_flags |= di_flags;
+			ip->i_d.di_flags2 |= di_flags2;
 		}
 		/* FALLTHROUGH */
 	case S_IFLNK:
@@ -1078,35 +1081,24 @@ xfs_dir_ialloc(
 }
 
 /*
- * Decrement the link count on an inode & log the change.
- * If this causes the link count to go to zero, initiate the
- * logging activity required to truncate a file.
+ * Decrement the link count on an inode & log the change.  If this causes the
+ * link count to go to zero, move the inode to AGI unlinked list so that it can
+ * be freed when the last active reference goes away via xfs_inactive().
  */
 int				/* error */
 xfs_droplink(
 	xfs_trans_t *tp,
 	xfs_inode_t *ip)
 {
-	int	error;
-
 	xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG);
 
-	ASSERT (ip->i_d.di_nlink > 0);
-	ip->i_d.di_nlink--;
 	drop_nlink(VFS_I(ip));
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 
-	error = 0;
-	if (ip->i_d.di_nlink == 0) {
-		/*
-		 * We're dropping the last link to this file.
-		 * Move the on-disk inode to the AGI unlinked list.
-		 * From xfs_inactive() we will pull the inode from
-		 * the list and free it.
-		 */
-		error = xfs_iunlink(tp, ip);
-	}
-	return error;
+	if (VFS_I(ip)->i_nlink)
+		return 0;
+
+	return xfs_iunlink(tp, ip);
 }
 
 /*
@@ -1120,8 +1112,6 @@ xfs_bumplink(
 	xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG);
 
 	ASSERT(ip->i_d.di_version > 1);
-	ASSERT(ip->i_d.di_nlink > 0 || (VFS_I(ip)->i_state & I_LINKABLE));
-	ip->i_d.di_nlink++;
 	inc_nlink(VFS_I(ip));
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 	return 0;
@@ -1143,7 +1133,6 @@ xfs_create(
 	xfs_bmap_free_t		free_list;
 	xfs_fsblock_t		first_block;
 	bool                    unlock_dp_on_error = false;
-	int			committed;
 	prid_t			prid;
 	struct xfs_dquot	*udqp = NULL;
 	struct xfs_dquot	*gdqp = NULL;
@@ -1226,7 +1215,7 @@ xfs_create(
 	 * pointing to itself.
 	 */
 	error = xfs_dir_ialloc(&tp, dp, mode, is_dir ? 2 : 1, rdev,
-			       prid, resblks > 0, &ip, &committed);
+			       prid, resblks > 0, &ip, NULL);
 	if (error)
 		goto out_trans_cancel;
 
@@ -1275,7 +1264,7 @@ xfs_create(
 	 */
 	xfs_qm_vop_create_dqattach(tp, ip, udqp, gdqp, pdqp);
 
-	error = xfs_bmap_finish(&tp, &free_list, &committed);
+	error = xfs_bmap_finish(&tp, &free_list, NULL);
 	if (error)
 		goto out_bmap_cancel;
 
@@ -1380,7 +1369,6 @@ xfs_create_tmpfile(
 	 */
 	xfs_qm_vop_create_dqattach(tp, ip, udqp, gdqp, pdqp);
 
-	ip->i_d.di_nlink--;
 	error = xfs_iunlink(tp, ip);
 	if (error)
 		goto out_trans_cancel;
@@ -1427,12 +1415,11 @@ xfs_link(
 	int			error;
 	xfs_bmap_free_t         free_list;
 	xfs_fsblock_t           first_block;
-	int			committed;
 	int			resblks;
 
 	trace_xfs_link(tdp, target_name);
 
-	ASSERT(!S_ISDIR(sip->i_d.di_mode));
+	ASSERT(!S_ISDIR(VFS_I(sip)->i_mode));
 
 	if (XFS_FORCED_SHUTDOWN(mp))
 		return -EIO;
@@ -1480,7 +1467,10 @@ xfs_link(
 
 	xfs_bmap_init(&free_list, &first_block);
 
-	if (sip->i_d.di_nlink == 0) {
+	/*
+	 * Handle initial link state of O_TMPFILE inode
+	 */
+	if (VFS_I(sip)->i_nlink == 0) {
 		error = xfs_iunlink_remove(tp, sip);
 		if (error)
 			goto error_return;
@@ -1502,11 +1492,10 @@ xfs_link(
 	 * link transaction goes to disk before returning to
 	 * the user.
 	 */
-	if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) {
+	if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC))
 		xfs_trans_set_sync(tp);
-	}
 
-	error = xfs_bmap_finish (&tp, &free_list, &committed);
+	error = xfs_bmap_finish(&tp, &free_list, NULL);
 	if (error) {
 		xfs_bmap_cancel(&free_list);
 		goto error_return;
@@ -1555,7 +1544,6 @@ xfs_itruncate_extents(
 	xfs_fileoff_t		first_unmap_block;
 	xfs_fileoff_t		last_block;
 	xfs_filblks_t		unmap_len;
-	int			committed;
 	int			error = 0;
 	int			done = 0;
 
@@ -1601,9 +1589,7 @@ xfs_itruncate_extents(
 		 * Duplicate the transaction that has the permanent
 		 * reservation and commit the old transaction.
 		 */
-		error = xfs_bmap_finish(&tp, &free_list, &committed);
-		if (committed)
-			xfs_trans_ijoin(tp, ip, 0);
+		error = xfs_bmap_finish(&tp, &free_list, ip);
 		if (error)
 			goto out_bmap_cancel;
 
@@ -1640,7 +1626,7 @@ xfs_release(
 	xfs_mount_t	*mp = ip->i_mount;
 	int		error;
 
-	if (!S_ISREG(ip->i_d.di_mode) || (ip->i_d.di_mode == 0))
+	if (!S_ISREG(VFS_I(ip)->i_mode) || (VFS_I(ip)->i_mode == 0))
 		return 0;
 
 	/* If this is a read-only mount, don't do this (would generate I/O) */
@@ -1671,7 +1657,7 @@ xfs_release(
 		}
 	}
 
-	if (ip->i_d.di_nlink == 0)
+	if (VFS_I(ip)->i_nlink == 0)
 		return 0;
 
 	if (xfs_can_free_eofblocks(ip, false)) {
@@ -1774,7 +1760,6 @@ xfs_inactive_ifree(
 {
 	xfs_bmap_free_t		free_list;
 	xfs_fsblock_t		first_block;
-	int			committed;
 	struct xfs_mount	*mp = ip->i_mount;
 	struct xfs_trans	*tp;
 	int			error;
@@ -1841,7 +1826,7 @@ xfs_inactive_ifree(
 	 * Just ignore errors at this point.  There is nothing we can do except
 	 * to try to keep going. Make sure it's not a silent error.
 	 */
-	error = xfs_bmap_finish(&tp,  &free_list, &committed);
+	error = xfs_bmap_finish(&tp, &free_list, NULL);
 	if (error) {
 		xfs_notice(mp, "%s: xfs_bmap_finish returned error %d",
 			__func__, error);
@@ -1876,7 +1861,7 @@ xfs_inactive(
 	 * If the inode is already free, then there can be nothing
 	 * to clean up here.
 	 */
-	if (ip->i_d.di_mode == 0) {
+	if (VFS_I(ip)->i_mode == 0) {
 		ASSERT(ip->i_df.if_real_bytes == 0);
 		ASSERT(ip->i_df.if_broot_bytes == 0);
 		return;
@@ -1888,7 +1873,7 @@ xfs_inactive(
 	if (mp->m_flags & XFS_MOUNT_RDONLY)
 		return;
 
-	if (ip->i_d.di_nlink != 0) {
+	if (VFS_I(ip)->i_nlink != 0) {
 		/*
 		 * force is true because we are evicting an inode from the
 		 * cache. Post-eof blocks must be freed, lest we end up with
@@ -1900,7 +1885,7 @@ xfs_inactive(
 		return;
 	}
 
-	if (S_ISREG(ip->i_d.di_mode) &&
+	if (S_ISREG(VFS_I(ip)->i_mode) &&
 	    (ip->i_d.di_size != 0 || XFS_ISIZE(ip) != 0 ||
 	     ip->i_d.di_nextents > 0 || ip->i_delayed_blks > 0))
 		truncate = 1;
@@ -1909,7 +1894,7 @@ xfs_inactive(
 	if (error)
 		return;
 
-	if (S_ISLNK(ip->i_d.di_mode))
+	if (S_ISLNK(VFS_I(ip)->i_mode))
 		error = xfs_inactive_symlink(ip);
 	else if (truncate)
 		error = xfs_inactive_truncate(ip);
@@ -1945,16 +1930,21 @@ xfs_inactive(
 }
 
 /*
- * This is called when the inode's link count goes to 0.
- * We place the on-disk inode on a list in the AGI.  It
- * will be pulled from this list when the inode is freed.
+ * This is called when the inode's link count goes to 0 or we are creating a
+ * tmpfile via O_TMPFILE. In the case of a tmpfile, @ignore_linkcount will be
+ * set to true as the link count is dropped to zero by the VFS after we've
+ * created the file successfully, so we have to add it to the unlinked list
+ * while the link count is non-zero.
+ *
+ * We place the on-disk inode on a list in the AGI.  It will be pulled from this
+ * list when the inode is freed.
  */
-int
+STATIC int
 xfs_iunlink(
-	xfs_trans_t	*tp,
-	xfs_inode_t	*ip)
+	struct xfs_trans *tp,
+	struct xfs_inode *ip)
 {
-	xfs_mount_t	*mp;
+	xfs_mount_t	*mp = tp->t_mountp;
 	xfs_agi_t	*agi;
 	xfs_dinode_t	*dip;
 	xfs_buf_t	*agibp;
@@ -1964,10 +1954,7 @@ xfs_iunlink(
 	int		offset;
 	int		error;
 
-	ASSERT(ip->i_d.di_nlink == 0);
-	ASSERT(ip->i_d.di_mode != 0);
-
-	mp = tp->t_mountp;
+	ASSERT(VFS_I(ip)->i_mode != 0);
 
 	/*
 	 * Get the agi buffer first.  It ensures lock ordering
@@ -2365,6 +2352,7 @@ retry:
 
 			iip->ili_last_fields = iip->ili_fields;
 			iip->ili_fields = 0;
+			iip->ili_fsync_fields = 0;
 			iip->ili_logged = 1;
 			xfs_trans_ail_copy_lsn(mp->m_ail, &iip->ili_flush_lsn,
 						&iip->ili_item.li_lsn);
@@ -2404,10 +2392,10 @@ xfs_ifree(
 	struct xfs_icluster	xic = { 0 };
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-	ASSERT(ip->i_d.di_nlink == 0);
+	ASSERT(VFS_I(ip)->i_nlink == 0);
 	ASSERT(ip->i_d.di_nextents == 0);
 	ASSERT(ip->i_d.di_anextents == 0);
-	ASSERT(ip->i_d.di_size == 0 || !S_ISREG(ip->i_d.di_mode));
+	ASSERT(ip->i_d.di_size == 0 || !S_ISREG(VFS_I(ip)->i_mode));
 	ASSERT(ip->i_d.di_nblocks == 0);
 
 	/*
@@ -2421,7 +2409,7 @@ xfs_ifree(
 	if (error)
 		return error;
 
-	ip->i_d.di_mode = 0;		/* mark incore inode as free */
+	VFS_I(ip)->i_mode = 0;		/* mark incore inode as free */
 	ip->i_d.di_flags = 0;
 	ip->i_d.di_dmevmask = 0;
 	ip->i_d.di_forkoff = 0;		/* mark the attr fork not in use */
@@ -2431,7 +2419,7 @@ xfs_ifree(
 	 * Bump the generation count so no one will be confused
 	 * by reincarnations of this inode.
 	 */
-	ip->i_d.di_gen++;
+	VFS_I(ip)->i_generation++;
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 
 	if (xic.deleted)
@@ -2518,11 +2506,10 @@ xfs_remove(
 {
 	xfs_mount_t		*mp = dp->i_mount;
 	xfs_trans_t             *tp = NULL;
-	int			is_dir = S_ISDIR(ip->i_d.di_mode);
+	int			is_dir = S_ISDIR(VFS_I(ip)->i_mode);
 	int                     error = 0;
 	xfs_bmap_free_t         free_list;
 	xfs_fsblock_t           first_block;
-	int			committed;
 	uint			resblks;
 
 	trace_xfs_remove(dp, name);
@@ -2573,8 +2560,8 @@ xfs_remove(
 	 * If we're removing a directory perform some additional validation.
 	 */
 	if (is_dir) {
-		ASSERT(ip->i_d.di_nlink >= 2);
-		if (ip->i_d.di_nlink != 2) {
+		ASSERT(VFS_I(ip)->i_nlink >= 2);
+		if (VFS_I(ip)->i_nlink != 2) {
 			error = -ENOTEMPTY;
 			goto out_trans_cancel;
 		}
@@ -2623,7 +2610,7 @@ xfs_remove(
 	if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC))
 		xfs_trans_set_sync(tp);
 
-	error = xfs_bmap_finish(&tp, &free_list, &committed);
+	error = xfs_bmap_finish(&tp, &free_list, NULL);
 	if (error)
 		goto out_bmap_cancel;
 
@@ -2700,7 +2687,6 @@ xfs_finish_rename(
 	struct xfs_trans	*tp,
 	struct xfs_bmap_free	*free_list)
 {
-	int			committed = 0;
 	int			error;
 
 	/*
@@ -2710,7 +2696,7 @@ xfs_finish_rename(
 	if (tp->t_mountp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC))
 		xfs_trans_set_sync(tp);
 
-	error = xfs_bmap_finish(&tp, free_list, &committed);
+	error = xfs_bmap_finish(&tp, free_list, NULL);
 	if (error) {
 		xfs_bmap_cancel(free_list);
 		xfs_trans_cancel(tp);
@@ -2765,7 +2751,7 @@ xfs_cross_rename(
 	if (dp1 != dp2) {
 		dp2_flags = XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG;
 
-		if (S_ISDIR(ip2->i_d.di_mode)) {
+		if (S_ISDIR(VFS_I(ip2)->i_mode)) {
 			error = xfs_dir_replace(tp, ip2, &xfs_name_dotdot,
 						dp1->i_ino, first_block,
 						free_list, spaceres);
@@ -2773,7 +2759,7 @@ xfs_cross_rename(
 				goto out_trans_abort;
 
 			/* transfer ip2 ".." reference to dp1 */
-			if (!S_ISDIR(ip1->i_d.di_mode)) {
+			if (!S_ISDIR(VFS_I(ip1)->i_mode)) {
 				error = xfs_droplink(tp, dp2);
 				if (error)
 					goto out_trans_abort;
@@ -2792,7 +2778,7 @@ xfs_cross_rename(
 			ip2_flags |= XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG;
 		}
 
-		if (S_ISDIR(ip1->i_d.di_mode)) {
+		if (S_ISDIR(VFS_I(ip1)->i_mode)) {
 			error = xfs_dir_replace(tp, ip1, &xfs_name_dotdot,
 						dp2->i_ino, first_block,
 						free_list, spaceres);
@@ -2800,7 +2786,7 @@ xfs_cross_rename(
 				goto out_trans_abort;
 
 			/* transfer ip1 ".." reference to dp2 */
-			if (!S_ISDIR(ip2->i_d.di_mode)) {
+			if (!S_ISDIR(VFS_I(ip2)->i_mode)) {
 				error = xfs_droplink(tp, dp1);
 				if (error)
 					goto out_trans_abort;
@@ -2897,7 +2883,7 @@ xfs_rename(
 	struct xfs_inode	*inodes[__XFS_SORT_INODES];
 	int			num_inodes = __XFS_SORT_INODES;
 	bool			new_parent = (src_dp != target_dp);
-	bool			src_is_directory = S_ISDIR(src_ip->i_d.di_mode);
+	bool			src_is_directory = S_ISDIR(VFS_I(src_ip)->i_mode);
 	int			spaceres;
 	int			error;
 
@@ -3026,12 +3012,12 @@ xfs_rename(
 		 * target and source are directories and that target can be
 		 * destroyed, or that neither is a directory.
 		 */
-		if (S_ISDIR(target_ip->i_d.di_mode)) {
+		if (S_ISDIR(VFS_I(target_ip)->i_mode)) {
 			/*
 			 * Make sure target dir is empty.
 			 */
 			if (!(xfs_dir_isempty(target_ip)) ||
-			    (target_ip->i_d.di_nlink > 2)) {
+			    (VFS_I(target_ip)->i_nlink > 2)) {
 				error = -EEXIST;
 				goto out_trans_cancel;
 			}
@@ -3138,7 +3124,7 @@ xfs_rename(
 	 * intermediate state on disk.
 	 */
 	if (wip) {
-		ASSERT(VFS_I(wip)->i_nlink == 0 && wip->i_d.di_nlink == 0);
+		ASSERT(VFS_I(wip)->i_nlink == 0);
 		error = xfs_bumplink(tp, wip);
 		if (error)
 			goto out_bmap_cancel;
@@ -3271,8 +3257,8 @@ xfs_iflush_cluster(
 	}
 
 	if (clcount) {
-		XFS_STATS_INC(xs_icluster_flushcnt);
-		XFS_STATS_ADD(xs_icluster_flushinode, clcount);
+		XFS_STATS_INC(mp, xs_icluster_flushcnt);
+		XFS_STATS_ADD(mp, xs_icluster_flushinode, clcount);
 	}
 
 out_free:
@@ -3307,7 +3293,7 @@ cluster_corrupt_out:
 		 * mark it as stale and brelse.
 		 */
 		if (bp->b_iodone) {
-			XFS_BUF_UNDONE(bp);
+			bp->b_flags &= ~XBF_DONE;
 			xfs_buf_stale(bp);
 			xfs_buf_ioerror(bp, -EIO);
 			xfs_buf_ioend(bp);
@@ -3345,7 +3331,7 @@ xfs_iflush(
 	struct xfs_dinode	*dip;
 	int			error;
 
-	XFS_STATS_INC(xs_iflush_count);
+	XFS_STATS_INC(mp, xs_iflush_count);
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
 	ASSERT(xfs_isiflocked(ip));
@@ -3456,14 +3442,7 @@ xfs_iflush_int(
 			__func__, ip->i_ino, be16_to_cpu(dip->di_magic), dip);
 		goto corrupt_out;
 	}
-	if (XFS_TEST_ERROR(ip->i_d.di_magic != XFS_DINODE_MAGIC,
-				mp, XFS_ERRTAG_IFLUSH_2, XFS_RANDOM_IFLUSH_2)) {
-		xfs_alert_tag(mp, XFS_PTAG_IFLUSH,
-			"%s: Bad inode %Lu, ptr 0x%p, magic number 0x%x",
-			__func__, ip->i_ino, ip, ip->i_d.di_magic);
-		goto corrupt_out;
-	}
-	if (S_ISREG(ip->i_d.di_mode)) {
+	if (S_ISREG(VFS_I(ip)->i_mode)) {
 		if (XFS_TEST_ERROR(
 		    (ip->i_d.di_format != XFS_DINODE_FMT_EXTENTS) &&
 		    (ip->i_d.di_format != XFS_DINODE_FMT_BTREE),
@@ -3473,7 +3452,7 @@ xfs_iflush_int(
 				__func__, ip->i_ino, ip);
 			goto corrupt_out;
 		}
-	} else if (S_ISDIR(ip->i_d.di_mode)) {
+	} else if (S_ISDIR(VFS_I(ip)->i_mode)) {
 		if (XFS_TEST_ERROR(
 		    (ip->i_d.di_format != XFS_DINODE_FMT_EXTENTS) &&
 		    (ip->i_d.di_format != XFS_DINODE_FMT_BTREE) &&
@@ -3517,12 +3496,11 @@ xfs_iflush_int(
 		ip->i_d.di_flushiter++;
 
 	/*
-	 * Copy the dirty parts of the inode into the on-disk
-	 * inode.  We always copy out the core of the inode,
-	 * because if the inode is dirty at all the core must
-	 * be.
+	 * Copy the dirty parts of the inode into the on-disk inode.  We always
+	 * copy out the core of the inode, because if the inode is dirty at all
+	 * the core must be.
 	 */
-	xfs_dinode_to_disk(dip, &ip->i_d);
+	xfs_inode_to_disk(ip, dip, iip->ili_item.li_lsn);
 
 	/* Wrap, we never let the log put out DI_MAX_FLUSH */
 	if (ip->i_d.di_flushiter == DI_MAX_FLUSH)
@@ -3560,6 +3538,7 @@ xfs_iflush_int(
 	 */
 	iip->ili_last_fields = iip->ili_fields;
 	iip->ili_fields = 0;
+	iip->ili_fsync_fields = 0;
 	iip->ili_logged = 1;
 
 	xfs_trans_ail_copy_lsn(mp->m_ail, &iip->ili_flush_lsn,
@@ -3573,10 +3552,6 @@ xfs_iflush_int(
 	 */
 	xfs_buf_attach_iodone(bp, xfs_iflush_done, &iip->ili_item);
 
-	/* update the lsn in the on disk inode if required */
-	if (ip->i_d.di_version == 3)
-		dip->di_lsn = cpu_to_be64(iip->ili_item.li_lsn);
-
 	/* generate the checksum. */
 	xfs_dinode_calc_crc(mp, dip);
 
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index ca9e11989cbd..43e1d51b15eb 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -63,7 +63,7 @@ typedef struct xfs_inode {
 	unsigned long		i_flags;	/* see defined flags below */
 	unsigned int		i_delayed_blks;	/* count of delay alloc blks */
 
-	xfs_icdinode_t		i_d;		/* most of ondisk inode */
+	struct xfs_icdinode	i_d;		/* most of ondisk inode */
 
 	/* VFS inode */
 	struct inode		i_vnode;	/* embedded VFS inode */
@@ -88,7 +88,7 @@ static inline struct inode *VFS_I(struct xfs_inode *ip)
  */
 static inline xfs_fsize_t XFS_ISIZE(struct xfs_inode *ip)
 {
-	if (S_ISREG(ip->i_d.di_mode))
+	if (S_ISREG(VFS_I(ip)->i_mode))
 		return i_size_read(VFS_I(ip));
 	return ip->i_d.di_size;
 }
@@ -369,7 +369,7 @@ static inline int xfs_isiflocked(struct xfs_inode *ip)
  */
 #define XFS_INHERIT_GID(pip)	\
 	(((pip)->i_mount->m_flags & XFS_MOUNT_GRPID) || \
-	 ((pip)->i_d.di_mode & S_ISGID))
+	 (VFS_I(pip)->i_mode & S_ISGID))
 
 int		xfs_release(struct xfs_inode *ip);
 void		xfs_inactive(struct xfs_inode *ip);
@@ -405,8 +405,6 @@ int		xfs_ifree(struct xfs_trans *, xfs_inode_t *,
 			   struct xfs_bmap_free *);
 int		xfs_itruncate_extents(struct xfs_trans **, struct xfs_inode *,
 				      int, xfs_fsize_t);
-int		xfs_iunlink(struct xfs_trans *, xfs_inode_t *);
-
 void		xfs_iext_realloc(xfs_inode_t *, int, int);
 
 void		xfs_iunpin_wait(xfs_inode_t *);
@@ -437,6 +435,8 @@ int	xfs_update_prealloc_flags(struct xfs_inode *ip,
 int	xfs_zero_eof(struct xfs_inode *ip, xfs_off_t offset,
 		     xfs_fsize_t isize, bool *did_zeroing);
 int	xfs_iozero(struct xfs_inode *ip, loff_t pos, size_t count);
+loff_t	__xfs_seek_hole_data(struct inode *inode, loff_t start,
+			     loff_t eof, int whence);
 
 
 /* from xfs_iops.c */
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index 62bd80f4edd9..c48b5b18d771 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -135,7 +135,7 @@ xfs_inode_item_size(
 
 	*nvecs += 2;
 	*nbytes += sizeof(struct xfs_inode_log_format) +
-		   xfs_icdinode_size(ip->i_d.di_version);
+		   xfs_log_dinode_size(ip->i_d.di_version);
 
 	xfs_inode_item_data_fork_size(iip, nvecs, nbytes);
 	if (XFS_IFORK_Q(ip))
@@ -322,6 +322,81 @@ xfs_inode_item_format_attr_fork(
 	}
 }
 
+static void
+xfs_inode_to_log_dinode(
+	struct xfs_inode	*ip,
+	struct xfs_log_dinode	*to,
+	xfs_lsn_t		lsn)
+{
+	struct xfs_icdinode	*from = &ip->i_d;
+	struct inode		*inode = VFS_I(ip);
+
+	to->di_magic = XFS_DINODE_MAGIC;
+
+	to->di_version = from->di_version;
+	to->di_format = from->di_format;
+	to->di_uid = from->di_uid;
+	to->di_gid = from->di_gid;
+	to->di_projid_lo = from->di_projid_lo;
+	to->di_projid_hi = from->di_projid_hi;
+
+	memset(to->di_pad, 0, sizeof(to->di_pad));
+	memset(to->di_pad3, 0, sizeof(to->di_pad3));
+	to->di_atime.t_sec = inode->i_atime.tv_sec;
+	to->di_atime.t_nsec = inode->i_atime.tv_nsec;
+	to->di_mtime.t_sec = inode->i_mtime.tv_sec;
+	to->di_mtime.t_nsec = inode->i_mtime.tv_nsec;
+	to->di_ctime.t_sec = inode->i_ctime.tv_sec;
+	to->di_ctime.t_nsec = inode->i_ctime.tv_nsec;
+	to->di_nlink = inode->i_nlink;
+	to->di_gen = inode->i_generation;
+	to->di_mode = inode->i_mode;
+
+	to->di_size = from->di_size;
+	to->di_nblocks = from->di_nblocks;
+	to->di_extsize = from->di_extsize;
+	to->di_nextents = from->di_nextents;
+	to->di_anextents = from->di_anextents;
+	to->di_forkoff = from->di_forkoff;
+	to->di_aformat = from->di_aformat;
+	to->di_dmevmask = from->di_dmevmask;
+	to->di_dmstate = from->di_dmstate;
+	to->di_flags = from->di_flags;
+
+	if (from->di_version == 3) {
+		to->di_changecount = inode->i_version;
+		to->di_crtime.t_sec = from->di_crtime.t_sec;
+		to->di_crtime.t_nsec = from->di_crtime.t_nsec;
+		to->di_flags2 = from->di_flags2;
+
+		to->di_ino = ip->i_ino;
+		to->di_lsn = lsn;
+		memset(to->di_pad2, 0, sizeof(to->di_pad2));
+		uuid_copy(&to->di_uuid, &ip->i_mount->m_sb.sb_meta_uuid);
+		to->di_flushiter = 0;
+	} else {
+		to->di_flushiter = from->di_flushiter;
+	}
+}
+
+/*
+ * Format the inode core. Current timestamp data is only in the VFS inode
+ * fields, so we need to grab them from there. Hence rather than just copying
+ * the XFS inode core structure, format the fields directly into the iovec.
+ */
+static void
+xfs_inode_item_format_core(
+	struct xfs_inode	*ip,
+	struct xfs_log_vec	*lv,
+	struct xfs_log_iovec	**vecp)
+{
+	struct xfs_log_dinode	*dic;
+
+	dic = xlog_prepare_iovec(lv, vecp, XLOG_REG_TYPE_ICORE);
+	xfs_inode_to_log_dinode(ip, dic, ip->i_itemp->ili_item.li_lsn);
+	xlog_finish_iovec(lv, *vecp, xfs_log_dinode_size(ip->i_d.di_version));
+}
+
 /*
  * This is called to fill in the vector of log iovecs for the given inode
  * log item.  It fills the first item with an inode log format structure,
@@ -351,10 +426,7 @@ xfs_inode_item_format(
 	ilf->ilf_size = 2; /* format + core */
 	xlog_finish_iovec(lv, vecp, sizeof(struct xfs_inode_log_format));
 
-	xlog_copy_iovec(lv, &vecp, XLOG_REG_TYPE_ICORE,
-			&ip->i_d,
-			xfs_icdinode_size(ip->i_d.di_version));
-
+	xfs_inode_item_format_core(ip, lv, &vecp);
 	xfs_inode_item_format_data_fork(iip, ilf, lv, &vecp);
 	if (XFS_IFORK_Q(ip)) {
 		xfs_inode_item_format_attr_fork(iip, ilf, lv, &vecp);
@@ -719,6 +791,7 @@ xfs_iflush_abort(
 		 * attempted.
 		 */
 		iip->ili_fields = 0;
+		iip->ili_fsync_fields = 0;
 	}
 	/*
 	 * Release the inode's flush lock since we're done with it.
diff --git a/fs/xfs/xfs_inode_item.h b/fs/xfs/xfs_inode_item.h
index 488d81254e28..4c7722e325b3 100644
--- a/fs/xfs/xfs_inode_item.h
+++ b/fs/xfs/xfs_inode_item.h
@@ -34,6 +34,7 @@ typedef struct xfs_inode_log_item {
 	unsigned short		ili_logged;	   /* flushed logged data */
 	unsigned int		ili_last_fields;   /* fields when flushed */
 	unsigned int		ili_fields;	   /* fields to be logged */
+	unsigned int		ili_fsync_fields;  /* logged since last fsync */
 } xfs_inode_log_item_t;
 
 static inline int xfs_inode_clean(xfs_inode_t *ip)
diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c
index ea7d85af5310..bcb6c19ce3ea 100644
--- a/fs/xfs/xfs_ioctl.c
+++ b/fs/xfs/xfs_ioctl.c
@@ -40,6 +40,7 @@
 #include "xfs_symlink.h"
 #include "xfs_trans.h"
 #include "xfs_pnfs.h"
+#include "xfs_acl.h"
 
 #include <linux/capability.h>
 #include <linux/dcache.h>
@@ -113,7 +114,7 @@ xfs_find_handle(
 		handle.ha_fid.fid_len = sizeof(xfs_fid_t) -
 					sizeof(handle.ha_fid.fid_len);
 		handle.ha_fid.fid_pad = 0;
-		handle.ha_fid.fid_gen = ip->i_d.di_gen;
+		handle.ha_fid.fid_gen = inode->i_generation;
 		handle.ha_fid.fid_ino = ip->i_ino;
 
 		hsize = XFS_HSIZE(handle);
@@ -411,7 +412,7 @@ xfs_attrlist_by_handle(
 	if (copy_from_user(&al_hreq, arg, sizeof(xfs_fsop_attrlist_handlereq_t)))
 		return -EFAULT;
 	if (al_hreq.buflen < sizeof(struct attrlist) ||
-	    al_hreq.buflen > XATTR_LIST_MAX)
+	    al_hreq.buflen > XFS_XATTR_LIST_MAX)
 		return -EINVAL;
 
 	/*
@@ -455,7 +456,7 @@ xfs_attrmulti_attr_get(
 	unsigned char		*kbuf;
 	int			error = -EFAULT;
 
-	if (*len > XATTR_SIZE_MAX)
+	if (*len > XFS_XATTR_SIZE_MAX)
 		return -EINVAL;
 	kbuf = kmem_zalloc_large(*len, KM_SLEEP);
 	if (!kbuf)
@@ -482,17 +483,22 @@ xfs_attrmulti_attr_set(
 	__uint32_t		flags)
 {
 	unsigned char		*kbuf;
+	int			error;
 
 	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
 		return -EPERM;
-	if (len > XATTR_SIZE_MAX)
+	if (len > XFS_XATTR_SIZE_MAX)
 		return -EINVAL;
 
 	kbuf = memdup_user(ubuf, len);
 	if (IS_ERR(kbuf))
 		return PTR_ERR(kbuf);
 
-	return xfs_attr_set(XFS_I(inode), name, kbuf, len, flags);
+	error = xfs_attr_set(XFS_I(inode), name, kbuf, len, flags);
+	if (!error)
+		xfs_forget_acl(inode, name, flags);
+	kfree(kbuf);
+	return error;
 }
 
 int
@@ -501,9 +507,14 @@ xfs_attrmulti_attr_remove(
 	unsigned char		*name,
 	__uint32_t		flags)
 {
+	int			error;
+
 	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
 		return -EPERM;
-	return xfs_attr_remove(XFS_I(inode), name, flags);
+	error = xfs_attr_remove(XFS_I(inode), name, flags);
+	if (!error)
+		xfs_forget_acl(inode, name, flags);
+	return error;
 }
 
 STATIC int
@@ -848,25 +859,25 @@ xfs_merge_ioc_xflags(
 	unsigned int	xflags = start;
 
 	if (flags & FS_IMMUTABLE_FL)
-		xflags |= XFS_XFLAG_IMMUTABLE;
+		xflags |= FS_XFLAG_IMMUTABLE;
 	else
-		xflags &= ~XFS_XFLAG_IMMUTABLE;
+		xflags &= ~FS_XFLAG_IMMUTABLE;
 	if (flags & FS_APPEND_FL)
-		xflags |= XFS_XFLAG_APPEND;
+		xflags |= FS_XFLAG_APPEND;
 	else
-		xflags &= ~XFS_XFLAG_APPEND;
+		xflags &= ~FS_XFLAG_APPEND;
 	if (flags & FS_SYNC_FL)
-		xflags |= XFS_XFLAG_SYNC;
+		xflags |= FS_XFLAG_SYNC;
 	else
-		xflags &= ~XFS_XFLAG_SYNC;
+		xflags &= ~FS_XFLAG_SYNC;
 	if (flags & FS_NOATIME_FL)
-		xflags |= XFS_XFLAG_NOATIME;
+		xflags |= FS_XFLAG_NOATIME;
 	else
-		xflags &= ~XFS_XFLAG_NOATIME;
+		xflags &= ~FS_XFLAG_NOATIME;
 	if (flags & FS_NODUMP_FL)
-		xflags |= XFS_XFLAG_NODUMP;
+		xflags |= FS_XFLAG_NODUMP;
 	else
-		xflags &= ~XFS_XFLAG_NODUMP;
+		xflags &= ~FS_XFLAG_NODUMP;
 
 	return xflags;
 }
@@ -934,40 +945,51 @@ xfs_set_diflags(
 	unsigned int		xflags)
 {
 	unsigned int		di_flags;
+	uint64_t		di_flags2;
 
 	/* can't set PREALLOC this way, just preserve it */
 	di_flags = (ip->i_d.di_flags & XFS_DIFLAG_PREALLOC);
-	if (xflags & XFS_XFLAG_IMMUTABLE)
+	if (xflags & FS_XFLAG_IMMUTABLE)
 		di_flags |= XFS_DIFLAG_IMMUTABLE;
-	if (xflags & XFS_XFLAG_APPEND)
+	if (xflags & FS_XFLAG_APPEND)
 		di_flags |= XFS_DIFLAG_APPEND;
-	if (xflags & XFS_XFLAG_SYNC)
+	if (xflags & FS_XFLAG_SYNC)
 		di_flags |= XFS_DIFLAG_SYNC;
-	if (xflags & XFS_XFLAG_NOATIME)
+	if (xflags & FS_XFLAG_NOATIME)
 		di_flags |= XFS_DIFLAG_NOATIME;
-	if (xflags & XFS_XFLAG_NODUMP)
+	if (xflags & FS_XFLAG_NODUMP)
 		di_flags |= XFS_DIFLAG_NODUMP;
-	if (xflags & XFS_XFLAG_NODEFRAG)
+	if (xflags & FS_XFLAG_NODEFRAG)
 		di_flags |= XFS_DIFLAG_NODEFRAG;
-	if (xflags & XFS_XFLAG_FILESTREAM)
+	if (xflags & FS_XFLAG_FILESTREAM)
 		di_flags |= XFS_DIFLAG_FILESTREAM;
-	if (S_ISDIR(ip->i_d.di_mode)) {
-		if (xflags & XFS_XFLAG_RTINHERIT)
+	if (S_ISDIR(VFS_I(ip)->i_mode)) {
+		if (xflags & FS_XFLAG_RTINHERIT)
 			di_flags |= XFS_DIFLAG_RTINHERIT;
-		if (xflags & XFS_XFLAG_NOSYMLINKS)
+		if (xflags & FS_XFLAG_NOSYMLINKS)
 			di_flags |= XFS_DIFLAG_NOSYMLINKS;
-		if (xflags & XFS_XFLAG_EXTSZINHERIT)
+		if (xflags & FS_XFLAG_EXTSZINHERIT)
 			di_flags |= XFS_DIFLAG_EXTSZINHERIT;
-		if (xflags & XFS_XFLAG_PROJINHERIT)
+		if (xflags & FS_XFLAG_PROJINHERIT)
 			di_flags |= XFS_DIFLAG_PROJINHERIT;
-	} else if (S_ISREG(ip->i_d.di_mode)) {
-		if (xflags & XFS_XFLAG_REALTIME)
+	} else if (S_ISREG(VFS_I(ip)->i_mode)) {
+		if (xflags & FS_XFLAG_REALTIME)
 			di_flags |= XFS_DIFLAG_REALTIME;
-		if (xflags & XFS_XFLAG_EXTSIZE)
+		if (xflags & FS_XFLAG_EXTSIZE)
 			di_flags |= XFS_DIFLAG_EXTSIZE;
 	}
-
 	ip->i_d.di_flags = di_flags;
+
+	/* diflags2 only valid for v3 inodes. */
+	if (ip->i_d.di_version < 3)
+		return;
+
+	di_flags2 = 0;
+	if (xflags & FS_XFLAG_DAX)
+		di_flags2 |= XFS_DIFLAG2_DAX;
+
+	ip->i_d.di_flags2 = di_flags2;
+
 }
 
 STATIC void
@@ -977,22 +999,27 @@ xfs_diflags_to_linux(
 	struct inode		*inode = VFS_I(ip);
 	unsigned int		xflags = xfs_ip2xflags(ip);
 
-	if (xflags & XFS_XFLAG_IMMUTABLE)
+	if (xflags & FS_XFLAG_IMMUTABLE)
 		inode->i_flags |= S_IMMUTABLE;
 	else
 		inode->i_flags &= ~S_IMMUTABLE;
-	if (xflags & XFS_XFLAG_APPEND)
+	if (xflags & FS_XFLAG_APPEND)
 		inode->i_flags |= S_APPEND;
 	else
 		inode->i_flags &= ~S_APPEND;
-	if (xflags & XFS_XFLAG_SYNC)
+	if (xflags & FS_XFLAG_SYNC)
 		inode->i_flags |= S_SYNC;
 	else
 		inode->i_flags &= ~S_SYNC;
-	if (xflags & XFS_XFLAG_NOATIME)
+	if (xflags & FS_XFLAG_NOATIME)
 		inode->i_flags |= S_NOATIME;
 	else
 		inode->i_flags &= ~S_NOATIME;
+	if (xflags & FS_XFLAG_DAX)
+		inode->i_flags |= S_DAX;
+	else
+		inode->i_flags &= ~S_DAX;
+
 }
 
 static int
@@ -1005,11 +1032,11 @@ xfs_ioctl_setattr_xflags(
 
 	/* Can't change realtime flag if any extents are allocated. */
 	if ((ip->i_d.di_nextents || ip->i_delayed_blks) &&
-	    XFS_IS_REALTIME_INODE(ip) != (fa->fsx_xflags & XFS_XFLAG_REALTIME))
+	    XFS_IS_REALTIME_INODE(ip) != (fa->fsx_xflags & FS_XFLAG_REALTIME))
 		return -EINVAL;
 
 	/* If realtime flag is set then must have realtime device */
-	if (fa->fsx_xflags & XFS_XFLAG_REALTIME) {
+	if (fa->fsx_xflags & FS_XFLAG_REALTIME) {
 		if (mp->m_sb.sb_rblocks == 0 || mp->m_sb.sb_rextsize == 0 ||
 		    (ip->i_d.di_extsize % mp->m_sb.sb_rextsize))
 			return -EINVAL;
@@ -1020,7 +1047,7 @@ xfs_ioctl_setattr_xflags(
 	 * we have appropriate permission.
 	 */
 	if (((ip->i_d.di_flags & (XFS_DIFLAG_IMMUTABLE | XFS_DIFLAG_APPEND)) ||
-	     (fa->fsx_xflags & (XFS_XFLAG_IMMUTABLE | XFS_XFLAG_APPEND))) &&
+	     (fa->fsx_xflags & (FS_XFLAG_IMMUTABLE | FS_XFLAG_APPEND))) &&
 	    !capable(CAP_LINUX_IMMUTABLE))
 		return -EPERM;
 
@@ -1028,8 +1055,63 @@ xfs_ioctl_setattr_xflags(
 	xfs_diflags_to_linux(ip);
 	xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG);
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-	XFS_STATS_INC(xs_ig_attrchg);
+	XFS_STATS_INC(mp, xs_ig_attrchg);
+	return 0;
+}
+
+/*
+ * If we are changing DAX flags, we have to ensure the file is clean and any
+ * cached objects in the address space are invalidated and removed. This
+ * requires us to lock out other IO and page faults similar to a truncate
+ * operation. The locks need to be held until the transaction has been committed
+ * so that the cache invalidation is atomic with respect to the DAX flag
+ * manipulation.
+ */
+static int
+xfs_ioctl_setattr_dax_invalidate(
+	struct xfs_inode	*ip,
+	struct fsxattr		*fa,
+	int			*join_flags)
+{
+	struct inode		*inode = VFS_I(ip);
+	int			error;
+
+	*join_flags = 0;
+
+	/*
+	 * It is only valid to set the DAX flag on regular files and
+	 * directories on filesystems where the block size is equal to the page
+	 * size. On directories it serves as an inherit hint.
+	 */
+	if (fa->fsx_xflags & FS_XFLAG_DAX) {
+		if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode)))
+			return -EINVAL;
+		if (ip->i_mount->m_sb.sb_blocksize != PAGE_SIZE)
+			return -EINVAL;
+	}
+
+	/* If the DAX state is not changing, we have nothing to do here. */
+	if ((fa->fsx_xflags & FS_XFLAG_DAX) && IS_DAX(inode))
+		return 0;
+	if (!(fa->fsx_xflags & FS_XFLAG_DAX) && !IS_DAX(inode))
+		return 0;
+
+	/* lock, flush and invalidate mapping in preparation for flag change */
+	xfs_ilock(ip, XFS_MMAPLOCK_EXCL | XFS_IOLOCK_EXCL);
+	error = filemap_write_and_wait(inode->i_mapping);
+	if (error)
+		goto out_unlock;
+	error = invalidate_inode_pages2(inode->i_mapping);
+	if (error)
+		goto out_unlock;
+
+	*join_flags = XFS_MMAPLOCK_EXCL | XFS_IOLOCK_EXCL;
 	return 0;
+
+out_unlock:
+	xfs_iunlock(ip, XFS_MMAPLOCK_EXCL | XFS_IOLOCK_EXCL);
+	return error;
+
 }
 
 /*
@@ -1037,19 +1119,27 @@ xfs_ioctl_setattr_xflags(
  * have permission to do so. On success, return a clean transaction and the
  * inode locked exclusively ready for further operation specific checks. On
  * failure, return an error without modifying or locking the inode.
+ *
+ * The inode might already be IO locked on call. If this is the case, it is
+ * indicated in @join_flags and we take full responsibility for ensuring they
+ * are unlocked from now on. Hence if we have an error here, we still have to
+ * unlock them. Otherwise, once they are joined to the transaction, they will
+ * be unlocked on commit/cancel.
  */
 static struct xfs_trans *
 xfs_ioctl_setattr_get_trans(
-	struct xfs_inode	*ip)
+	struct xfs_inode	*ip,
+	int			join_flags)
 {
 	struct xfs_mount	*mp = ip->i_mount;
 	struct xfs_trans	*tp;
-	int			error;
+	int			error = -EROFS;
 
 	if (mp->m_flags & XFS_MOUNT_RDONLY)
-		return ERR_PTR(-EROFS);
+		goto out_unlock;
+	error = -EIO;
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return ERR_PTR(-EIO);
+		goto out_unlock;
 
 	tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE);
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ichange, 0, 0);
@@ -1057,7 +1147,8 @@ xfs_ioctl_setattr_get_trans(
 		goto out_cancel;
 
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL | join_flags);
+	join_flags = 0;
 
 	/*
 	 * CAP_FOWNER overrides the following restrictions:
@@ -1077,6 +1168,9 @@ xfs_ioctl_setattr_get_trans(
 
 out_cancel:
 	xfs_trans_cancel(tp);
+out_unlock:
+	if (join_flags)
+		xfs_iunlock(ip, join_flags);
 	return ERR_PTR(error);
 }
 
@@ -1084,8 +1178,8 @@ out_cancel:
  * extent size hint validation is somewhat cumbersome. Rules are:
  *
  * 1. extent size hint is only valid for directories and regular files
- * 2. XFS_XFLAG_EXTSIZE is only valid for regular files
- * 3. XFS_XFLAG_EXTSZINHERIT is only valid for directories.
+ * 2. FS_XFLAG_EXTSIZE is only valid for regular files
+ * 3. FS_XFLAG_EXTSZINHERIT is only valid for directories.
  * 4. can only be changed on regular files if no extents are allocated
  * 5. can be changed on directories at any time
  * 6. extsize hint of 0 turns off hints, clears inode flags.
@@ -1101,14 +1195,14 @@ xfs_ioctl_setattr_check_extsize(
 {
 	struct xfs_mount	*mp = ip->i_mount;
 
-	if ((fa->fsx_xflags & XFS_XFLAG_EXTSIZE) && !S_ISREG(ip->i_d.di_mode))
+	if ((fa->fsx_xflags & FS_XFLAG_EXTSIZE) && !S_ISREG(VFS_I(ip)->i_mode))
 		return -EINVAL;
 
-	if ((fa->fsx_xflags & XFS_XFLAG_EXTSZINHERIT) &&
-	    !S_ISDIR(ip->i_d.di_mode))
+	if ((fa->fsx_xflags & FS_XFLAG_EXTSZINHERIT) &&
+	    !S_ISDIR(VFS_I(ip)->i_mode))
 		return -EINVAL;
 
-	if (S_ISREG(ip->i_d.di_mode) && ip->i_d.di_nextents &&
+	if (S_ISREG(VFS_I(ip)->i_mode) && ip->i_d.di_nextents &&
 	    ((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) != fa->fsx_extsize))
 		return -EINVAL;
 
@@ -1121,7 +1215,7 @@ xfs_ioctl_setattr_check_extsize(
 			return -EINVAL;
 
 		if (XFS_IS_REALTIME_INODE(ip) ||
-		    (fa->fsx_xflags & XFS_XFLAG_REALTIME)) {
+		    (fa->fsx_xflags & FS_XFLAG_REALTIME)) {
 			size = mp->m_sb.sb_rextsize << mp->m_sb.sb_blocklog;
 		} else {
 			size = mp->m_sb.sb_blocksize;
@@ -1132,7 +1226,7 @@ xfs_ioctl_setattr_check_extsize(
 		if (fa->fsx_extsize % size)
 			return -EINVAL;
 	} else
-		fa->fsx_xflags &= ~(XFS_XFLAG_EXTSIZE | XFS_XFLAG_EXTSZINHERIT);
+		fa->fsx_xflags &= ~(FS_XFLAG_EXTSIZE | FS_XFLAG_EXTSZINHERIT);
 
 	return 0;
 }
@@ -1157,7 +1251,7 @@ xfs_ioctl_setattr_check_projid(
 
 	if (xfs_get_projid(ip) != fa->fsx_projid)
 		return -EINVAL;
-	if ((fa->fsx_xflags & XFS_XFLAG_PROJINHERIT) !=
+	if ((fa->fsx_xflags & FS_XFLAG_PROJINHERIT) !=
 	    (ip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT))
 		return -EINVAL;
 
@@ -1175,6 +1269,7 @@ xfs_ioctl_setattr(
 	struct xfs_dquot	*pdqp = NULL;
 	struct xfs_dquot	*olddquot = NULL;
 	int			code;
+	int			join_flags = 0;
 
 	trace_xfs_ioctl_setattr(ip);
 
@@ -1198,7 +1293,18 @@ xfs_ioctl_setattr(
 			return code;
 	}
 
-	tp = xfs_ioctl_setattr_get_trans(ip);
+	/*
+	 * Changing DAX config may require inode locking for mapping
+	 * invalidation. These need to be held all the way to transaction commit
+	 * or cancel time, so need to be passed through to
+	 * xfs_ioctl_setattr_get_trans() so it can apply them to the join call
+	 * appropriately.
+	 */
+	code = xfs_ioctl_setattr_dax_invalidate(ip, fa, &join_flags);
+	if (code)
+		goto error_free_dquots;
+
+	tp = xfs_ioctl_setattr_get_trans(ip, join_flags);
 	if (IS_ERR(tp)) {
 		code = PTR_ERR(tp);
 		goto error_free_dquots;
@@ -1229,9 +1335,9 @@ xfs_ioctl_setattr(
 	 * successful return from chown()
 	 */
 
-	if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) &&
+	if ((VFS_I(ip)->i_mode & (S_ISUID|S_ISGID)) &&
 	    !capable_wrt_inode_uidgid(VFS_I(ip), CAP_FSETID))
-		ip->i_d.di_mode &= ~(S_ISUID|S_ISGID);
+		VFS_I(ip)->i_mode &= ~(S_ISUID|S_ISGID);
 
 	/* Change the ownerships and register project quota modifications */
 	if (xfs_get_projid(ip) != fa->fsx_projid) {
@@ -1314,6 +1420,7 @@ xfs_ioc_setxflags(
 	struct xfs_trans	*tp;
 	struct fsxattr		fa;
 	unsigned int		flags;
+	int			join_flags = 0;
 	int			error;
 
 	if (copy_from_user(&flags, arg, sizeof(flags)))
@@ -1330,7 +1437,18 @@ xfs_ioc_setxflags(
 	if (error)
 		return error;
 
-	tp = xfs_ioctl_setattr_get_trans(ip);
+	/*
+	 * Changing DAX config may require inode locking for mapping
+	 * invalidation. These need to be held all the way to transaction commit
+	 * or cancel time, so need to be passed through to
+	 * xfs_ioctl_setattr_get_trans() so it can apply them to the join call
+	 * appropriately.
+	 */
+	error = xfs_ioctl_setattr_dax_invalidate(ip, &fa, &join_flags);
+	if (error)
+		goto out_drop_write;
+
+	tp = xfs_ioctl_setattr_get_trans(ip, join_flags);
 	if (IS_ERR(tp)) {
 		error = PTR_ERR(tp);
 		goto out_drop_write;
diff --git a/fs/xfs/xfs_ioctl32.c b/fs/xfs/xfs_ioctl32.c
index b88bdc85dd3d..1a05d8ae327d 100644
--- a/fs/xfs/xfs_ioctl32.c
+++ b/fs/xfs/xfs_ioctl32.c
@@ -356,7 +356,7 @@ xfs_compat_attrlist_by_handle(
 			   sizeof(compat_xfs_fsop_attrlist_handlereq_t)))
 		return -EFAULT;
 	if (al_hreq.buflen < sizeof(struct attrlist) ||
-	    al_hreq.buflen > XATTR_LIST_MAX)
+	    al_hreq.buflen > XFS_XATTR_LIST_MAX)
 		return -EINVAL;
 
 	/*
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 1f86033171c8..d81bdc080370 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -129,22 +129,31 @@ xfs_iomap_write_direct(
 	xfs_trans_t	*tp;
 	xfs_bmap_free_t free_list;
 	uint		qblocks, resblks, resrtextents;
-	int		committed;
 	int		error;
-
-	error = xfs_qm_dqattach(ip, 0);
-	if (error)
-		return error;
+	int		lockmode;
+	int		bmapi_flags = XFS_BMAPI_PREALLOC;
 
 	rt = XFS_IS_REALTIME_INODE(ip);
 	extsz = xfs_get_extsz_hint(ip);
+	lockmode = XFS_ILOCK_SHARED;	/* locked by caller */
+
+	ASSERT(xfs_isilocked(ip, lockmode));
 
 	offset_fsb = XFS_B_TO_FSBT(mp, offset);
 	last_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)(offset + count)));
 	if ((offset + count) > XFS_ISIZE(ip)) {
+		/*
+		 * Assert that the in-core extent list is present since this can
+		 * call xfs_iread_extents() and we only have the ilock shared.
+		 * This should be safe because the lock was held around a bmapi
+		 * call in the caller and we only need it to access the in-core
+		 * list.
+		 */
+		ASSERT(XFS_IFORK_PTR(ip, XFS_DATA_FORK)->if_flags &
+								XFS_IFEXTENTS);
 		error = xfs_iomap_eof_align_last_fsb(mp, ip, extsz, &last_fsb);
 		if (error)
-			return error;
+			goto out_unlock;
 	} else {
 		if (nmaps && (imap->br_startblock == HOLESTARTBLOCK))
 			last_fsb = MIN(last_fsb, (xfs_fileoff_t)
@@ -174,9 +183,40 @@ xfs_iomap_write_direct(
 	}
 
 	/*
+	 * Drop the shared lock acquired by the caller, attach the dquot if
+	 * necessary and move on to transaction setup.
+	 */
+	xfs_iunlock(ip, lockmode);
+	error = xfs_qm_dqattach(ip, 0);
+	if (error)
+		return error;
+
+	/*
 	 * Allocate and setup the transaction
 	 */
 	tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT);
+
+	/*
+	 * For DAX, we do not allocate unwritten extents, but instead we zero
+	 * the block before we commit the transaction.  Ideally we'd like to do
+	 * this outside the transaction context, but if we commit and then crash
+	 * we may not have zeroed the blocks and this will be exposed on
+	 * recovery of the allocation. Hence we must zero before commit.
+	 *
+	 * Further, if we are mapping unwritten extents here, we need to zero
+	 * and convert them to written so that we don't need an unwritten extent
+	 * callback for DAX. This also means that we need to be able to dip into
+	 * the reserve block pool for bmbt block allocation if there is no space
+	 * left but we need to do unwritten extent conversion.
+	 */
+
+	if (IS_DAX(VFS_I(ip))) {
+		bmapi_flags = XFS_BMAPI_CONVERT | XFS_BMAPI_ZERO;
+		if (ISUNWRITTEN(imap)) {
+			tp->t_flags |= XFS_TRANS_RESERVE;
+			resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1;
+		}
+	}
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_write,
 				  resblks, resrtextents);
 	/*
@@ -187,7 +227,8 @@ xfs_iomap_write_direct(
 		return error;
 	}
 
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	lockmode = XFS_ILOCK_EXCL;
+	xfs_ilock(ip, lockmode);
 
 	error = xfs_trans_reserve_quota_nblks(tp, ip, qblocks, 0, quota_flag);
 	if (error)
@@ -202,17 +243,18 @@ xfs_iomap_write_direct(
 	xfs_bmap_init(&free_list, &firstfsb);
 	nimaps = 1;
 	error = xfs_bmapi_write(tp, ip, offset_fsb, count_fsb,
-				XFS_BMAPI_PREALLOC, &firstfsb, 0,
-				imap, &nimaps, &free_list);
+				bmapi_flags, &firstfsb, resblks, imap,
+				&nimaps, &free_list);
 	if (error)
 		goto out_bmap_cancel;
 
 	/*
 	 * Complete the transaction
 	 */
-	error = xfs_bmap_finish(&tp, &free_list, &committed);
+	error = xfs_bmap_finish(&tp, &free_list, NULL);
 	if (error)
 		goto out_bmap_cancel;
+
 	error = xfs_trans_commit(tp);
 	if (error)
 		goto out_unlock;
@@ -229,7 +271,7 @@ xfs_iomap_write_direct(
 		error = xfs_alert_fsblock_zero(ip, imap);
 
 out_unlock:
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	xfs_iunlock(ip, lockmode);
 	return error;
 
 out_bmap_cancel:
@@ -655,7 +697,7 @@ xfs_iomap_write_allocate(
 	xfs_bmap_free_t	free_list;
 	xfs_filblks_t	count_fsb;
 	xfs_trans_t	*tp;
-	int		nimaps, committed;
+	int		nimaps;
 	int		error = 0;
 	int		nres;
 
@@ -670,7 +712,7 @@ xfs_iomap_write_allocate(
 	count_fsb = imap->br_blockcount;
 	map_start_fsb = imap->br_startoff;
 
-	XFS_STATS_ADD(xs_xstrat_bytes, XFS_FSB_TO_B(mp, count_fsb));
+	XFS_STATS_ADD(mp, xs_xstrat_bytes, XFS_FSB_TO_B(mp, count_fsb));
 
 	while (count_fsb != 0) {
 		/*
@@ -750,13 +792,13 @@ xfs_iomap_write_allocate(
 			 * pointer that the caller gave to us.
 			 */
 			error = xfs_bmapi_write(tp, ip, map_start_fsb,
-						count_fsb, 0,
-						&first_block, 1,
-						imap, &nimaps, &free_list);
+						count_fsb, 0, &first_block,
+						nres, imap, &nimaps,
+						&free_list);
 			if (error)
 				goto trans_cancel;
 
-			error = xfs_bmap_finish(&tp, &free_list, &committed);
+			error = xfs_bmap_finish(&tp, &free_list, NULL);
 			if (error)
 				goto trans_cancel;
 
@@ -777,7 +819,7 @@ xfs_iomap_write_allocate(
 		if ((offset_fsb >= imap->br_startoff) &&
 		    (offset_fsb < (imap->br_startoff +
 				   imap->br_blockcount))) {
-			XFS_STATS_INC(xs_xstrat_quick);
+			XFS_STATS_INC(mp, xs_xstrat_quick);
 			return 0;
 		}
 
@@ -814,7 +856,6 @@ xfs_iomap_write_unwritten(
 	xfs_bmap_free_t free_list;
 	xfs_fsize_t	i_size;
 	uint		resblks;
-	int		committed;
 	int		error;
 
 	trace_xfs_unwritten_convert(ip, offset, count);
@@ -866,8 +907,8 @@ xfs_iomap_write_unwritten(
 		xfs_bmap_init(&free_list, &firstfsb);
 		nimaps = 1;
 		error = xfs_bmapi_write(tp, ip, offset_fsb, count_fsb,
-				  XFS_BMAPI_CONVERT, &firstfsb,
-				  1, &imap, &nimaps, &free_list);
+					XFS_BMAPI_CONVERT, &firstfsb, resblks,
+					&imap, &nimaps, &free_list);
 		if (error)
 			goto error_on_bmapi_transaction;
 
@@ -886,7 +927,7 @@ xfs_iomap_write_unwritten(
 			xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 		}
 
-		error = xfs_bmap_finish(&tp, &free_list, &committed);
+		error = xfs_bmap_finish(&tp, &free_list, NULL);
 		if (error)
 			goto error_on_bmapi_transaction;
 
diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c
index 8294132e6a3c..fb7dc61f4a29 100644
--- a/fs/xfs/xfs_iops.c
+++ b/fs/xfs/xfs_iops.c
@@ -414,13 +414,17 @@ xfs_vn_rename(
  * uio is kmalloced for this reason...
  */
 STATIC const char *
-xfs_vn_follow_link(
+xfs_vn_get_link(
 	struct dentry		*dentry,
-	void			**cookie)
+	struct inode		*inode,
+	struct delayed_call	*done)
 {
 	char			*link;
 	int			error = -ENOMEM;
 
+	if (!dentry)
+		return ERR_PTR(-ECHILD);
+
 	link = kmalloc(MAXPATHLEN+1, GFP_KERNEL);
 	if (!link)
 		goto out_err;
@@ -429,7 +433,8 @@ xfs_vn_follow_link(
 	if (unlikely(error))
 		goto out_kfree;
 
-	return *cookie = link;
+	set_delayed_call(done, kfree_link, link);
+	return link;
 
  out_kfree:
 	kfree(link);
@@ -454,8 +459,8 @@ xfs_vn_getattr(
 
 	stat->size = XFS_ISIZE(ip);
 	stat->dev = inode->i_sb->s_dev;
-	stat->mode = ip->i_d.di_mode;
-	stat->nlink = ip->i_d.di_nlink;
+	stat->mode = inode->i_mode;
+	stat->nlink = inode->i_nlink;
 	stat->uid = inode->i_uid;
 	stat->gid = inode->i_gid;
 	stat->ino = ip->i_ino;
@@ -501,9 +506,6 @@ xfs_setattr_mode(
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
-	ip->i_d.di_mode &= S_IFMT;
-	ip->i_d.di_mode |= mode & ~S_IFMT;
-
 	inode->i_mode &= S_IFMT;
 	inode->i_mode |= mode & ~S_IFMT;
 }
@@ -517,21 +519,12 @@ xfs_setattr_time(
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
-	if (iattr->ia_valid & ATTR_ATIME) {
+	if (iattr->ia_valid & ATTR_ATIME)
 		inode->i_atime = iattr->ia_atime;
-		ip->i_d.di_atime.t_sec = iattr->ia_atime.tv_sec;
-		ip->i_d.di_atime.t_nsec = iattr->ia_atime.tv_nsec;
-	}
-	if (iattr->ia_valid & ATTR_CTIME) {
+	if (iattr->ia_valid & ATTR_CTIME)
 		inode->i_ctime = iattr->ia_ctime;
-		ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec;
-		ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec;
-	}
-	if (iattr->ia_valid & ATTR_MTIME) {
+	if (iattr->ia_valid & ATTR_MTIME)
 		inode->i_mtime = iattr->ia_mtime;
-		ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec;
-		ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec;
-	}
 }
 
 int
@@ -656,9 +649,9 @@ xfs_setattr_nonsize(
 		 * The set-user-ID and set-group-ID bits of a file will be
 		 * cleared upon successful return from chown()
 		 */
-		if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) &&
+		if ((inode->i_mode & (S_ISUID|S_ISGID)) &&
 		    !capable(CAP_FSETID))
-			ip->i_d.di_mode &= ~(S_ISUID|S_ISGID);
+			inode->i_mode &= ~(S_ISUID|S_ISGID);
 
 		/*
 		 * Change the ownerships and register quota modifications
@@ -695,7 +688,7 @@ xfs_setattr_nonsize(
 
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 
-	XFS_STATS_INC(xs_ig_attrchg);
+	XFS_STATS_INC(mp, xs_ig_attrchg);
 
 	if (mp->m_flags & XFS_MOUNT_WSYNC)
 		xfs_trans_set_sync(tp);
@@ -768,7 +761,7 @@ xfs_setattr_size(
 
 	ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
 	ASSERT(xfs_isilocked(ip, XFS_MMAPLOCK_EXCL));
-	ASSERT(S_ISREG(ip->i_d.di_mode));
+	ASSERT(S_ISREG(inode->i_mode));
 	ASSERT((iattr->ia_valid & (ATTR_UID|ATTR_GID|ATTR_ATIME|ATTR_ATIME_SET|
 		ATTR_MTIME_SET|ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0);
 
@@ -922,7 +915,7 @@ xfs_setattr_size(
 
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 
-	XFS_STATS_INC(xs_ig_attrchg);
+	XFS_STATS_INC(mp, xs_ig_attrchg);
 
 	if (mp->m_flags & XFS_MOUNT_WSYNC)
 		xfs_trans_set_sync(tp);
@@ -986,21 +979,13 @@ xfs_vn_update_time(
 	}
 
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	if (flags & S_CTIME) {
+	if (flags & S_CTIME)
 		inode->i_ctime = *now;
-		ip->i_d.di_ctime.t_sec = (__int32_t)now->tv_sec;
-		ip->i_d.di_ctime.t_nsec = (__int32_t)now->tv_nsec;
-	}
-	if (flags & S_MTIME) {
+	if (flags & S_MTIME)
 		inode->i_mtime = *now;
-		ip->i_d.di_mtime.t_sec = (__int32_t)now->tv_sec;
-		ip->i_d.di_mtime.t_nsec = (__int32_t)now->tv_nsec;
-	}
-	if (flags & S_ATIME) {
+	if (flags & S_ATIME)
 		inode->i_atime = *now;
-		ip->i_d.di_atime.t_sec = (__int32_t)now->tv_sec;
-		ip->i_d.di_atime.t_nsec = (__int32_t)now->tv_nsec;
-	}
+
 	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_TIMESTAMP);
 	return xfs_trans_commit(tp);
@@ -1172,8 +1157,7 @@ static const struct inode_operations xfs_dir_ci_inode_operations = {
 
 static const struct inode_operations xfs_symlink_inode_operations = {
 	.readlink		= generic_readlink,
-	.follow_link		= xfs_vn_follow_link,
-	.put_link		= kfree_put_link,
+	.get_link		= xfs_vn_get_link,
 	.getattr		= xfs_vn_getattr,
 	.setattr		= xfs_vn_setattr,
 	.setxattr		= generic_setxattr,
@@ -1201,8 +1185,10 @@ xfs_diflags_to_iflags(
 		inode->i_flags |= S_SYNC;
 	if (flags & XFS_DIFLAG_NOATIME)
 		inode->i_flags |= S_NOATIME;
-	/* XXX: Also needs an on-disk per inode flag! */
-	if (ip->i_mount->m_flags & XFS_MOUNT_DAX)
+	if (S_ISREG(inode->i_mode) &&
+	    ip->i_mount->m_sb.sb_blocksize == PAGE_SIZE &&
+	    (ip->i_mount->m_flags & XFS_MOUNT_DAX ||
+	     ip->i_d.di_flags2 & XFS_DIFLAG2_DAX))
 		inode->i_flags |= S_DAX;
 }
 
@@ -1228,8 +1214,6 @@ xfs_setup_inode(
 	/* make the inode look hashed for the writeback code */
 	hlist_add_fake(&inode->i_hash);
 
-	inode->i_mode	= ip->i_d.di_mode;
-	set_nlink(inode, ip->i_d.di_nlink);
 	inode->i_uid    = xfs_uid_to_kuid(ip->i_d.di_uid);
 	inode->i_gid    = xfs_gid_to_kgid(ip->i_d.di_gid);
 
@@ -1245,14 +1229,7 @@ xfs_setup_inode(
 		break;
 	}
 
-	inode->i_generation = ip->i_d.di_gen;
 	i_size_write(inode, ip->i_d.di_size);
-	inode->i_atime.tv_sec	= ip->i_d.di_atime.t_sec;
-	inode->i_atime.tv_nsec	= ip->i_d.di_atime.t_nsec;
-	inode->i_mtime.tv_sec	= ip->i_d.di_mtime.t_sec;
-	inode->i_mtime.tv_nsec	= ip->i_d.di_mtime.t_nsec;
-	inode->i_ctime.tv_sec	= ip->i_d.di_ctime.t_sec;
-	inode->i_ctime.tv_nsec	= ip->i_d.di_ctime.t_nsec;
 	xfs_diflags_to_iflags(inode, ip);
 
 	ip->d_ops = ip->i_mount->m_nondir_inode_ops;
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c
index 930ebd86beba..ce73eb34620d 100644
--- a/fs/xfs/xfs_itable.c
+++ b/fs/xfs/xfs_itable.c
@@ -57,6 +57,7 @@ xfs_bulkstat_one_int(
 {
 	struct xfs_icdinode	*dic;		/* dinode core info pointer */
 	struct xfs_inode	*ip;		/* incore inode pointer */
+	struct inode		*inode;
 	struct xfs_bstat	*buf;		/* return buffer */
 	int			error = 0;	/* error value */
 
@@ -77,30 +78,33 @@ xfs_bulkstat_one_int(
 
 	ASSERT(ip != NULL);
 	ASSERT(ip->i_imap.im_blkno != 0);
+	inode = VFS_I(ip);
 
 	dic = &ip->i_d;
 
 	/* xfs_iget returns the following without needing
 	 * further change.
 	 */
-	buf->bs_nlink = dic->di_nlink;
 	buf->bs_projid_lo = dic->di_projid_lo;
 	buf->bs_projid_hi = dic->di_projid_hi;
 	buf->bs_ino = ino;
-	buf->bs_mode = dic->di_mode;
 	buf->bs_uid = dic->di_uid;
 	buf->bs_gid = dic->di_gid;
 	buf->bs_size = dic->di_size;
-	buf->bs_atime.tv_sec = dic->di_atime.t_sec;
-	buf->bs_atime.tv_nsec = dic->di_atime.t_nsec;
-	buf->bs_mtime.tv_sec = dic->di_mtime.t_sec;
-	buf->bs_mtime.tv_nsec = dic->di_mtime.t_nsec;
-	buf->bs_ctime.tv_sec = dic->di_ctime.t_sec;
-	buf->bs_ctime.tv_nsec = dic->di_ctime.t_nsec;
+
+	buf->bs_nlink = inode->i_nlink;
+	buf->bs_atime.tv_sec = inode->i_atime.tv_sec;
+	buf->bs_atime.tv_nsec = inode->i_atime.tv_nsec;
+	buf->bs_mtime.tv_sec = inode->i_mtime.tv_sec;
+	buf->bs_mtime.tv_nsec = inode->i_mtime.tv_nsec;
+	buf->bs_ctime.tv_sec = inode->i_ctime.tv_sec;
+	buf->bs_ctime.tv_nsec = inode->i_ctime.tv_nsec;
+	buf->bs_gen = inode->i_generation;
+	buf->bs_mode = inode->i_mode;
+
 	buf->bs_xflags = xfs_ip2xflags(ip);
 	buf->bs_extsize = dic->di_extsize << mp->m_sb.sb_blocklog;
 	buf->bs_extents = dic->di_nextents;
-	buf->bs_gen = dic->di_gen;
 	memset(buf->bs_pad, 0, sizeof(buf->bs_pad));
 	buf->bs_dmevmask = dic->di_dmevmask;
 	buf->bs_dmstate = dic->di_dmstate;
diff --git a/fs/xfs/xfs_linux.h b/fs/xfs/xfs_linux.h
index 85f883dd6207..a8192dc797dc 100644
--- a/fs/xfs/xfs_linux.h
+++ b/fs/xfs/xfs_linux.h
@@ -135,7 +135,7 @@ typedef __u32			xfs_nlink_t;
  * Size of block device i/o is parameterized here.
  * Currently the system supports page-sized i/o.
  */
-#define	BLKDEV_IOSHIFT		PAGE_CACHE_SHIFT
+#define	BLKDEV_IOSHIFT		PAGE_SHIFT
 #define	BLKDEV_IOSIZE		(1<<BLKDEV_IOSHIFT)
 /* number of BB's per block device block */
 #define	BLKDEV_BB		BTOBB(BLKDEV_IOSIZE)
@@ -171,6 +171,13 @@ struct xfs_kobj {
 	struct completion	complete;
 };
 
+struct xstats {
+	struct xfsstats __percpu	*xs_stats;
+	struct xfs_kobj			xs_kobj;
+};
+
+extern struct xstats xfsstats;
+
 /* Kernel uid/gid conversion. These are used to convert to/from the on disk
  * uid_t/gid_t types to the kuid_t/kgid_t types that the kernel uses internally.
  * The conversion here is type only, the value will remain the same since we
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index aaadee0969c9..b49ccf5c1d75 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -268,7 +268,7 @@ xlog_grant_head_wait(
 		__set_current_state(TASK_UNINTERRUPTIBLE);
 		spin_unlock(&head->lock);
 
-		XFS_STATS_INC(xs_sleep_logspace);
+		XFS_STATS_INC(log->l_mp, xs_sleep_logspace);
 
 		trace_xfs_log_grant_sleep(log, tic);
 		schedule();
@@ -379,7 +379,7 @@ xfs_log_regrant(
 	if (XLOG_FORCED_SHUTDOWN(log))
 		return -EIO;
 
-	XFS_STATS_INC(xs_try_logspace);
+	XFS_STATS_INC(mp, xs_try_logspace);
 
 	/*
 	 * This is a new transaction on the ticket, so we need to change the
@@ -448,7 +448,7 @@ xfs_log_reserve(
 	if (XLOG_FORCED_SHUTDOWN(log))
 		return -EIO;
 
-	XFS_STATS_INC(xs_try_logspace);
+	XFS_STATS_INC(mp, xs_try_logspace);
 
 	ASSERT(*ticp == NULL);
 	tic = xlog_ticket_alloc(log, unit_bytes, cnt, client, permanent,
@@ -1188,10 +1188,16 @@ xlog_iodone(xfs_buf_t *bp)
 	int			aborted = 0;
 
 	/*
-	 * Race to shutdown the filesystem if we see an error.
+	 * Race to shutdown the filesystem if we see an error or the iclog is in
+	 * IOABORT state. The IOABORT state is only set in DEBUG mode to inject
+	 * CRC errors into log recovery.
 	 */
-	if (XFS_TEST_ERROR(bp->b_error, l->l_mp,
-			XFS_ERRTAG_IODONE_IOERR, XFS_RANDOM_IODONE_IOERR)) {
+	if (XFS_TEST_ERROR(bp->b_error, l->l_mp, XFS_ERRTAG_IODONE_IOERR,
+			   XFS_RANDOM_IODONE_IOERR) ||
+	    iclog->ic_state & XLOG_STATE_IOABORT) {
+		if (iclog->ic_state & XLOG_STATE_IOABORT)
+			iclog->ic_state &= ~XLOG_STATE_IOABORT;
+
 		xfs_buf_ioerror_alert(bp, __func__);
 		xfs_buf_stale(bp);
 		xfs_force_shutdown(l->l_mp, SHUTDOWN_LOG_IO_ERROR);
@@ -1206,7 +1212,7 @@ xlog_iodone(xfs_buf_t *bp)
 	}
 
 	/* log I/O is always issued ASYNC */
-	ASSERT(XFS_BUF_ISASYNC(bp));
+	ASSERT(bp->b_flags & XBF_ASYNC);
 	xlog_state_done_syncing(iclog, aborted);
 
 	/*
@@ -1768,7 +1774,7 @@ xlog_sync(
 	int		v2 = xfs_sb_version_haslogv2(&log->l_mp->m_sb);
 	int		size;
 
-	XFS_STATS_INC(xs_log_writes);
+	XFS_STATS_INC(log->l_mp, xs_log_writes);
 	ASSERT(atomic_read(&iclog->ic_refcnt) == 0);
 
 	/* Add for LR header */
@@ -1805,7 +1811,7 @@ xlog_sync(
 	bp = iclog->ic_bp;
 	XFS_BUF_SET_ADDR(bp, BLOCK_LSN(be64_to_cpu(iclog->ic_header.h_lsn)));
 
-	XFS_STATS_ADD(xs_log_blocks, BTOBB(count));
+	XFS_STATS_ADD(log->l_mp, xs_log_blocks, BTOBB(count));
 
 	/* Do we need to split this write into 2 parts? */
 	if (XFS_BUF_ADDR(bp) + BTOBB(count) > log->l_logBBsize) {
@@ -1838,12 +1844,28 @@ xlog_sync(
 	/* calculcate the checksum */
 	iclog->ic_header.h_crc = xlog_cksum(log, &iclog->ic_header,
 					    iclog->ic_datap, size);
+#ifdef DEBUG
+	/*
+	 * Intentionally corrupt the log record CRC based on the error injection
+	 * frequency, if defined. This facilitates testing log recovery in the
+	 * event of torn writes. Hence, set the IOABORT state to abort the log
+	 * write on I/O completion and shutdown the fs. The subsequent mount
+	 * detects the bad CRC and attempts to recover.
+	 */
+	if (log->l_badcrc_factor &&
+	    (prandom_u32() % log->l_badcrc_factor == 0)) {
+		iclog->ic_header.h_crc &= 0xAAAAAAAA;
+		iclog->ic_state |= XLOG_STATE_IOABORT;
+		xfs_warn(log->l_mp,
+	"Intentionally corrupted log record at LSN 0x%llx. Shutdown imminent.",
+			 be64_to_cpu(iclog->ic_header.h_lsn));
+	}
+#endif
 
 	bp->b_io_length = BTOBB(count);
 	bp->b_fspriv = iclog;
-	XFS_BUF_ZEROFLAGS(bp);
-	XFS_BUF_ASYNC(bp);
-	bp->b_flags |= XBF_SYNCIO;
+	bp->b_flags &= ~(XBF_FUA | XBF_FLUSH);
+	bp->b_flags |= (XBF_ASYNC | XBF_SYNCIO | XBF_WRITE);
 
 	if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) {
 		bp->b_flags |= XBF_FUA;
@@ -1870,12 +1892,11 @@ xlog_sync(
 
 	/* account for log which doesn't start at block #0 */
 	XFS_BUF_SET_ADDR(bp, XFS_BUF_ADDR(bp) + log->l_logBBstart);
+
 	/*
 	 * Don't call xfs_bwrite here. We do log-syncs even when the filesystem
 	 * is shutting down.
 	 */
-	XFS_BUF_WRITE(bp);
-
 	error = xlog_bdstrat(bp);
 	if (error) {
 		xfs_buf_ioerror_alert(bp, "xlog_sync");
@@ -1887,9 +1908,8 @@ xlog_sync(
 		xfs_buf_associate_memory(bp,
 				(char *)&iclog->ic_header + count, split);
 		bp->b_fspriv = iclog;
-		XFS_BUF_ZEROFLAGS(bp);
-		XFS_BUF_ASYNC(bp);
-		bp->b_flags |= XBF_SYNCIO;
+		bp->b_flags &= ~(XBF_FUA | XBF_FLUSH);
+		bp->b_flags |= (XBF_ASYNC | XBF_SYNCIO | XBF_WRITE);
 		if (log->l_mp->m_flags & XFS_MOUNT_BARRIER)
 			bp->b_flags |= XBF_FUA;
 
@@ -1898,7 +1918,6 @@ xlog_sync(
 
 		/* account for internal log which doesn't start at block #0 */
 		XFS_BUF_SET_ADDR(bp, XFS_BUF_ADDR(bp) + log->l_logBBstart);
-		XFS_BUF_WRITE(bp);
 		error = xlog_bdstrat(bp);
 		if (error) {
 			xfs_buf_ioerror_alert(bp, "xlog_sync (split)");
@@ -1989,75 +2008,81 @@ xlog_print_tic_res(
 	uint ophdr_spc = ticket->t_res_num_ophdrs * (uint)sizeof(xlog_op_header_t);
 
 	/* match with XLOG_REG_TYPE_* in xfs_log.h */
-	static char *res_type_str[XLOG_REG_TYPE_MAX] = {
-	    "bformat",
-	    "bchunk",
-	    "efi_format",
-	    "efd_format",
-	    "iformat",
-	    "icore",
-	    "iext",
-	    "ibroot",
-	    "ilocal",
-	    "iattr_ext",
-	    "iattr_broot",
-	    "iattr_local",
-	    "qformat",
-	    "dquot",
-	    "quotaoff",
-	    "LR header",
-	    "unmount",
-	    "commit",
-	    "trans header"
+#define REG_TYPE_STR(type, str)	[XLOG_REG_TYPE_##type] = str
+	static char *res_type_str[XLOG_REG_TYPE_MAX + 1] = {
+	    REG_TYPE_STR(BFORMAT, "bformat"),
+	    REG_TYPE_STR(BCHUNK, "bchunk"),
+	    REG_TYPE_STR(EFI_FORMAT, "efi_format"),
+	    REG_TYPE_STR(EFD_FORMAT, "efd_format"),
+	    REG_TYPE_STR(IFORMAT, "iformat"),
+	    REG_TYPE_STR(ICORE, "icore"),
+	    REG_TYPE_STR(IEXT, "iext"),
+	    REG_TYPE_STR(IBROOT, "ibroot"),
+	    REG_TYPE_STR(ILOCAL, "ilocal"),
+	    REG_TYPE_STR(IATTR_EXT, "iattr_ext"),
+	    REG_TYPE_STR(IATTR_BROOT, "iattr_broot"),
+	    REG_TYPE_STR(IATTR_LOCAL, "iattr_local"),
+	    REG_TYPE_STR(QFORMAT, "qformat"),
+	    REG_TYPE_STR(DQUOT, "dquot"),
+	    REG_TYPE_STR(QUOTAOFF, "quotaoff"),
+	    REG_TYPE_STR(LRHEADER, "LR header"),
+	    REG_TYPE_STR(UNMOUNT, "unmount"),
+	    REG_TYPE_STR(COMMIT, "commit"),
+	    REG_TYPE_STR(TRANSHDR, "trans header"),
+	    REG_TYPE_STR(ICREATE, "inode create")
 	};
+#undef REG_TYPE_STR
+#define TRANS_TYPE_STR(type)	[XFS_TRANS_##type] = #type
 	static char *trans_type_str[XFS_TRANS_TYPE_MAX] = {
-	    "SETATTR_NOT_SIZE",
-	    "SETATTR_SIZE",
-	    "INACTIVE",
-	    "CREATE",
-	    "CREATE_TRUNC",
-	    "TRUNCATE_FILE",
-	    "REMOVE",
-	    "LINK",
-	    "RENAME",
-	    "MKDIR",
-	    "RMDIR",
-	    "SYMLINK",
-	    "SET_DMATTRS",
-	    "GROWFS",
-	    "STRAT_WRITE",
-	    "DIOSTRAT",
-	    "WRITE_SYNC",
-	    "WRITEID",
-	    "ADDAFORK",
-	    "ATTRINVAL",
-	    "ATRUNCATE",
-	    "ATTR_SET",
-	    "ATTR_RM",
-	    "ATTR_FLAG",
-	    "CLEAR_AGI_BUCKET",
-	    "QM_SBCHANGE",
-	    "DUMMY1",
-	    "DUMMY2",
-	    "QM_QUOTAOFF",
-	    "QM_DQALLOC",
-	    "QM_SETQLIM",
-	    "QM_DQCLUSTER",
-	    "QM_QINOCREATE",
-	    "QM_QUOTAOFF_END",
-	    "SB_UNIT",
-	    "FSYNC_TS",
-	    "GROWFSRT_ALLOC",
-	    "GROWFSRT_ZERO",
-	    "GROWFSRT_FREE",
-	    "SWAPEXT"
+	    TRANS_TYPE_STR(SETATTR_NOT_SIZE),
+	    TRANS_TYPE_STR(SETATTR_SIZE),
+	    TRANS_TYPE_STR(INACTIVE),
+	    TRANS_TYPE_STR(CREATE),
+	    TRANS_TYPE_STR(CREATE_TRUNC),
+	    TRANS_TYPE_STR(TRUNCATE_FILE),
+	    TRANS_TYPE_STR(REMOVE),
+	    TRANS_TYPE_STR(LINK),
+	    TRANS_TYPE_STR(RENAME),
+	    TRANS_TYPE_STR(MKDIR),
+	    TRANS_TYPE_STR(RMDIR),
+	    TRANS_TYPE_STR(SYMLINK),
+	    TRANS_TYPE_STR(SET_DMATTRS),
+	    TRANS_TYPE_STR(GROWFS),
+	    TRANS_TYPE_STR(STRAT_WRITE),
+	    TRANS_TYPE_STR(DIOSTRAT),
+	    TRANS_TYPE_STR(WRITEID),
+	    TRANS_TYPE_STR(ADDAFORK),
+	    TRANS_TYPE_STR(ATTRINVAL),
+	    TRANS_TYPE_STR(ATRUNCATE),
+	    TRANS_TYPE_STR(ATTR_SET),
+	    TRANS_TYPE_STR(ATTR_RM),
+	    TRANS_TYPE_STR(ATTR_FLAG),
+	    TRANS_TYPE_STR(CLEAR_AGI_BUCKET),
+	    TRANS_TYPE_STR(SB_CHANGE),
+	    TRANS_TYPE_STR(DUMMY1),
+	    TRANS_TYPE_STR(DUMMY2),
+	    TRANS_TYPE_STR(QM_QUOTAOFF),
+	    TRANS_TYPE_STR(QM_DQALLOC),
+	    TRANS_TYPE_STR(QM_SETQLIM),
+	    TRANS_TYPE_STR(QM_DQCLUSTER),
+	    TRANS_TYPE_STR(QM_QINOCREATE),
+	    TRANS_TYPE_STR(QM_QUOTAOFF_END),
+	    TRANS_TYPE_STR(FSYNC_TS),
+	    TRANS_TYPE_STR(GROWFSRT_ALLOC),
+	    TRANS_TYPE_STR(GROWFSRT_ZERO),
+	    TRANS_TYPE_STR(GROWFSRT_FREE),
+	    TRANS_TYPE_STR(SWAPEXT),
+	    TRANS_TYPE_STR(CHECKPOINT),
+	    TRANS_TYPE_STR(ICREATE),
+	    TRANS_TYPE_STR(CREATE_TMPFILE)
 	};
+#undef TRANS_TYPE_STR
 
 	xfs_warn(mp, "xlog_write: reservation summary:");
 	xfs_warn(mp, "  trans type  = %s (%u)",
 		 ((ticket->t_trans_type <= 0 ||
 		   ticket->t_trans_type > XFS_TRANS_TYPE_MAX) ?
-		  "bad-trans-type" : trans_type_str[ticket->t_trans_type-1]),
+		  "bad-trans-type" : trans_type_str[ticket->t_trans_type]),
 		 ticket->t_trans_type);
 	xfs_warn(mp, "  unit res    = %d bytes",
 		 ticket->t_unit_res);
@@ -2076,7 +2101,7 @@ xlog_print_tic_res(
 		uint r_type = ticket->t_res_arr[i].r_type;
 		xfs_warn(mp, "region[%u]: %s - %u bytes", i,
 			    ((r_type <= 0 || r_type > XLOG_REG_TYPE_MAX) ?
-			    "bad-rtype" : res_type_str[r_type-1]),
+			    "bad-rtype" : res_type_str[r_type]),
 			    ticket->t_res_arr[i].r_len);
 	}
 
@@ -2422,11 +2447,20 @@ xlog_write(
 						     &partial_copy_len);
 			xlog_verify_dest_ptr(log, ptr);
 
-			/* copy region */
+			/*
+			 * Copy region.
+			 *
+			 * Unmount records just log an opheader, so can have
+			 * empty payloads with no data region to copy. Hence we
+			 * only copy the payload if the vector says it has data
+			 * to copy.
+			 */
 			ASSERT(copy_len >= 0);
-			memcpy(ptr, reg->i_addr + copy_off, copy_len);
-			xlog_write_adv_cnt(&ptr, &len, &log_offset, copy_len);
-
+			if (copy_len > 0) {
+				memcpy(ptr, reg->i_addr + copy_off, copy_len);
+				xlog_write_adv_cnt(&ptr, &len, &log_offset,
+						   copy_len);
+			}
 			copy_len += start_rec_copy + sizeof(xlog_op_header_t);
 			record_cnt++;
 			data_cnt += contwr ? copy_len : 0;
@@ -2782,11 +2816,19 @@ xlog_state_do_callback(
 		}
 	} while (!ioerrors && loopdidcallbacks);
 
+#ifdef DEBUG
 	/*
-	 * make one last gasp attempt to see if iclogs are being left in
-	 * limbo..
+	 * Make one last gasp attempt to see if iclogs are being left in limbo.
+	 * If the above loop finds an iclog earlier than the current iclog and
+	 * in one of the syncing states, the current iclog is put into
+	 * DO_CALLBACK and the callbacks are deferred to the completion of the
+	 * earlier iclog. Walk the iclogs in order and make sure that no iclog
+	 * is in DO_CALLBACK unless an earlier iclog is in one of the syncing
+	 * states.
+	 *
+	 * Note that SYNCING|IOABORT is a valid state so we cannot just check
+	 * for ic_state == SYNCING.
 	 */
-#ifdef DEBUG
 	if (funcdidcallbacks) {
 		first_iclog = iclog = log->l_iclog;
 		do {
@@ -2801,7 +2843,7 @@ xlog_state_do_callback(
 			 * IOERROR - give up hope all ye who enter here
 			 */
 			if (iclog->ic_state == XLOG_STATE_WANT_SYNC ||
-			    iclog->ic_state == XLOG_STATE_SYNCING ||
+			    iclog->ic_state & XLOG_STATE_SYNCING ||
 			    iclog->ic_state == XLOG_STATE_DONE_SYNC ||
 			    iclog->ic_state == XLOG_STATE_IOERROR )
 				break;
@@ -2913,7 +2955,7 @@ restart:
 
 	iclog = log->l_iclog;
 	if (iclog->ic_state != XLOG_STATE_ACTIVE) {
-		XFS_STATS_INC(xs_log_noiclogs);
+		XFS_STATS_INC(log->l_mp, xs_log_noiclogs);
 
 		/* Wait for log writes to have flushed */
 		xlog_wait(&log->l_flush_wait, &log->l_icloglock);
@@ -3165,11 +3207,19 @@ xlog_state_switch_iclogs(
 	}
 
 	if (log->l_curr_block >= log->l_logBBsize) {
+		/*
+		 * Rewind the current block before the cycle is bumped to make
+		 * sure that the combined LSN never transiently moves forward
+		 * when the log wraps to the next cycle. This is to support the
+		 * unlocked sample of these fields from xlog_valid_lsn(). Most
+		 * other cases should acquire l_icloglock.
+		 */
+		log->l_curr_block -= log->l_logBBsize;
+		ASSERT(log->l_curr_block >= 0);
+		smp_wmb();
 		log->l_curr_cycle++;
 		if (log->l_curr_cycle == XLOG_HEADER_MAGIC_NUM)
 			log->l_curr_cycle++;
-		log->l_curr_block -= log->l_logBBsize;
-		ASSERT(log->l_curr_block >= 0);
 	}
 	ASSERT(iclog == log->l_iclog);
 	log->l_iclog = iclog->ic_next;
@@ -3212,7 +3262,7 @@ _xfs_log_force(
 	struct xlog_in_core	*iclog;
 	xfs_lsn_t		lsn;
 
-	XFS_STATS_INC(xs_log_force);
+	XFS_STATS_INC(mp, xs_log_force);
 
 	xlog_cil_force(log);
 
@@ -3297,7 +3347,7 @@ maybe_sleep:
 			spin_unlock(&log->l_icloglock);
 			return -EIO;
 		}
-		XFS_STATS_INC(xs_log_force_sleep);
+		XFS_STATS_INC(mp, xs_log_force_sleep);
 		xlog_wait(&iclog->ic_force_wait, &log->l_icloglock);
 		/*
 		 * No need to grab the log lock here since we're
@@ -3362,7 +3412,7 @@ _xfs_log_force_lsn(
 
 	ASSERT(lsn != 0);
 
-	XFS_STATS_INC(xs_log_force);
+	XFS_STATS_INC(mp, xs_log_force);
 
 	lsn = xlog_cil_force_lsn(log, lsn);
 	if (lsn == NULLCOMMITLSN)
@@ -3411,7 +3461,7 @@ try_again:
 			     (XLOG_STATE_WANT_SYNC | XLOG_STATE_SYNCING))) {
 				ASSERT(!(iclog->ic_state & XLOG_STATE_IOERROR));
 
-				XFS_STATS_INC(xs_log_force_sleep);
+				XFS_STATS_INC(mp, xs_log_force_sleep);
 
 				xlog_wait(&iclog->ic_prev->ic_write_wait,
 							&log->l_icloglock);
@@ -3441,7 +3491,7 @@ try_again:
 				spin_unlock(&log->l_icloglock);
 				return -EIO;
 			}
-			XFS_STATS_INC(xs_log_force_sleep);
+			XFS_STATS_INC(mp, xs_log_force_sleep);
 			xlog_wait(&iclog->ic_force_wait, &log->l_icloglock);
 			/*
 			 * No need to grab the log lock here since we're
@@ -3929,7 +3979,7 @@ xfs_log_force_umount(
 	    log->l_flags & XLOG_ACTIVE_RECOVERY) {
 		mp->m_flags |= XFS_MOUNT_FS_SHUTDOWN;
 		if (mp->m_sb_bp)
-			XFS_BUF_DONE(mp->m_sb_bp);
+			mp->m_sb_bp->b_flags |= XBF_DONE;
 		return 0;
 	}
 
@@ -3959,7 +4009,7 @@ xfs_log_force_umount(
 	spin_lock(&log->l_icloglock);
 	mp->m_flags |= XFS_MOUNT_FS_SHUTDOWN;
 	if (mp->m_sb_bp)
-		XFS_BUF_DONE(mp->m_sb_bp);
+		mp->m_sb_bp->b_flags |= XBF_DONE;
 
 	/*
 	 * Mark the log and the iclogs with IO error flags to prevent any
@@ -4023,3 +4073,45 @@ xlog_iclogs_empty(
 	return 1;
 }
 
+/*
+ * Verify that an LSN stamped into a piece of metadata is valid. This is
+ * intended for use in read verifiers on v5 superblocks.
+ */
+bool
+xfs_log_check_lsn(
+	struct xfs_mount	*mp,
+	xfs_lsn_t		lsn)
+{
+	struct xlog		*log = mp->m_log;
+	bool			valid;
+
+	/*
+	 * norecovery mode skips mount-time log processing and unconditionally
+	 * resets the in-core LSN. We can't validate in this mode, but
+	 * modifications are not allowed anyways so just return true.
+	 */
+	if (mp->m_flags & XFS_MOUNT_NORECOVERY)
+		return true;
+
+	/*
+	 * Some metadata LSNs are initialized to NULL (e.g., the agfl). This is
+	 * handled by recovery and thus safe to ignore here.
+	 */
+	if (lsn == NULLCOMMITLSN)
+		return true;
+
+	valid = xlog_valid_lsn(mp->m_log, lsn);
+
+	/* warn the user about what's gone wrong before verifier failure */
+	if (!valid) {
+		spin_lock(&log->l_icloglock);
+		xfs_warn(mp,
+"Corruption warning: Metadata has LSN (%d:%d) ahead of current LSN (%d:%d). "
+"Please unmount and run xfs_repair (>= v4.3) to resolve.",
+			 CYCLE_LSN(lsn), BLOCK_LSN(lsn),
+			 log->l_curr_cycle, log->l_curr_block);
+		spin_unlock(&log->l_icloglock);
+	}
+
+	return valid;
+}
diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h
index 09d91d3166cd..aa533a7d50f2 100644
--- a/fs/xfs/xfs_log.h
+++ b/fs/xfs/xfs_log.h
@@ -181,5 +181,6 @@ bool	xfs_log_item_in_current_chkpt(struct xfs_log_item *lip);
 void	xfs_log_work_queue(struct xfs_mount *mp);
 void	xfs_log_worker(struct work_struct *work);
 void	xfs_log_quiesce(struct xfs_mount *mp);
+bool	xfs_log_check_lsn(struct xfs_mount *, xfs_lsn_t);
 
 #endif	/* __XFS_LOG_H__ */
diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h
index 950f3f94720c..ed8896310c00 100644
--- a/fs/xfs/xfs_log_priv.h
+++ b/fs/xfs/xfs_log_priv.h
@@ -62,6 +62,7 @@ static inline uint xlog_get_client_id(__be32 i)
 #define XLOG_STATE_CALLBACK  0x0020 /* Callback functions now */
 #define XLOG_STATE_DIRTY     0x0040 /* Dirty IC log, not ready for ACTIVE status*/
 #define XLOG_STATE_IOERROR   0x0080 /* IO error happened in sync'ing log */
+#define XLOG_STATE_IOABORT   0x0100 /* force abort on I/O completion (debug) */
 #define XLOG_STATE_ALL	     0x7FFF /* All possible valid flags */
 #define XLOG_STATE_NOTUSED   0x8000 /* This IC log not being used */
 
@@ -410,6 +411,8 @@ struct xlog {
 	/* The following field are used for debugging; need to hold icloglock */
 #ifdef DEBUG
 	void			*l_iclog_bak[XLOG_MAX_ICLOGS];
+	/* log record crc error injection factor */
+	uint32_t		l_badcrc_factor;
 #endif
 
 };
@@ -560,4 +563,55 @@ static inline void xlog_wait(wait_queue_head_t *wq, spinlock_t *lock)
 	remove_wait_queue(wq, &wait);
 }
 
+/*
+ * The LSN is valid so long as it is behind the current LSN. If it isn't, this
+ * means that the next log record that includes this metadata could have a
+ * smaller LSN. In turn, this means that the modification in the log would not
+ * replay.
+ */
+static inline bool
+xlog_valid_lsn(
+	struct xlog	*log,
+	xfs_lsn_t	lsn)
+{
+	int		cur_cycle;
+	int		cur_block;
+	bool		valid = true;
+
+	/*
+	 * First, sample the current lsn without locking to avoid added
+	 * contention from metadata I/O. The current cycle and block are updated
+	 * (in xlog_state_switch_iclogs()) and read here in a particular order
+	 * to avoid false negatives (e.g., thinking the metadata LSN is valid
+	 * when it is not).
+	 *
+	 * The current block is always rewound before the cycle is bumped in
+	 * xlog_state_switch_iclogs() to ensure the current LSN is never seen in
+	 * a transiently forward state. Instead, we can see the LSN in a
+	 * transiently behind state if we happen to race with a cycle wrap.
+	 */
+	cur_cycle = ACCESS_ONCE(log->l_curr_cycle);
+	smp_rmb();
+	cur_block = ACCESS_ONCE(log->l_curr_block);
+
+	if ((CYCLE_LSN(lsn) > cur_cycle) ||
+	    (CYCLE_LSN(lsn) == cur_cycle && BLOCK_LSN(lsn) > cur_block)) {
+		/*
+		 * If the metadata LSN appears invalid, it's possible the check
+		 * above raced with a wrap to the next log cycle. Grab the lock
+		 * to check for sure.
+		 */
+		spin_lock(&log->l_icloglock);
+		cur_cycle = log->l_curr_cycle;
+		cur_block = log->l_curr_block;
+		spin_unlock(&log->l_icloglock);
+
+		if ((CYCLE_LSN(lsn) > cur_cycle) ||
+		    (CYCLE_LSN(lsn) == cur_cycle && BLOCK_LSN(lsn) > cur_block))
+			valid = false;
+	}
+
+	return valid;
+}
+
 #endif	/* __XFS_LOG_PRIV_H__ */
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 512a0945d52a..396565f43247 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -61,6 +61,9 @@ xlog_recover_check_summary(
 #else
 #define	xlog_recover_check_summary(log)
 #endif
+STATIC int
+xlog_do_recovery_pass(
+        struct xlog *, xfs_daddr_t, xfs_daddr_t, int, xfs_daddr_t *);
 
 /*
  * This structure is used during recovery to record the buf log items which
@@ -187,7 +190,7 @@ xlog_bread_noalign(
 	ASSERT(nbblks <= bp->b_length);
 
 	XFS_BUF_SET_ADDR(bp, log->l_logBBstart + blk_no);
-	XFS_BUF_READ(bp);
+	bp->b_flags |= XBF_READ;
 	bp->b_io_length = nbblks;
 	bp->b_error = 0;
 
@@ -272,7 +275,6 @@ xlog_bwrite(
 	ASSERT(nbblks <= bp->b_length);
 
 	XFS_BUF_SET_ADDR(bp, log->l_logBBstart + blk_no);
-	XFS_BUF_ZEROFLAGS(bp);
 	xfs_buf_hold(bp);
 	xfs_buf_lock(bp);
 	bp->b_io_length = nbblks;
@@ -868,136 +870,365 @@ validate_head:
 }
 
 /*
- * Find the sync block number or the tail of the log.
- *
- * This will be the block number of the last record to have its
- * associated buffers synced to disk.  Every log record header has
- * a sync lsn embedded in it.  LSNs hold block numbers, so it is easy
- * to get a sync block number.  The only concern is to figure out which
- * log record header to believe.
- *
- * The following algorithm uses the log record header with the largest
- * lsn.  The entire log record does not need to be valid.  We only care
- * that the header is valid.
+ * Seek backwards in the log for log record headers.
  *
- * We could speed up search by using current head_blk buffer, but it is not
- * available.
+ * Given a starting log block, walk backwards until we find the provided number
+ * of records or hit the provided tail block. The return value is the number of
+ * records encountered or a negative error code. The log block and buffer
+ * pointer of the last record seen are returned in rblk and rhead respectively.
  */
 STATIC int
-xlog_find_tail(
+xlog_rseek_logrec_hdr(
 	struct xlog		*log,
-	xfs_daddr_t		*head_blk,
-	xfs_daddr_t		*tail_blk)
+	xfs_daddr_t		head_blk,
+	xfs_daddr_t		tail_blk,
+	int			count,
+	struct xfs_buf		*bp,
+	xfs_daddr_t		*rblk,
+	struct xlog_rec_header	**rhead,
+	bool			*wrapped)
 {
-	xlog_rec_header_t	*rhead;
-	xlog_op_header_t	*op_head;
+	int			i;
+	int			error;
+	int			found = 0;
 	char			*offset = NULL;
-	xfs_buf_t		*bp;
-	int			error, i, found;
-	xfs_daddr_t		umount_data_blk;
-	xfs_daddr_t		after_umount_blk;
-	xfs_lsn_t		tail_lsn;
-	int			hblks;
+	xfs_daddr_t		end_blk;
 
-	found = 0;
+	*wrapped = false;
 
 	/*
-	 * Find previous log record
+	 * Walk backwards from the head block until we hit the tail or the first
+	 * block in the log.
 	 */
-	if ((error = xlog_find_head(log, head_blk)))
-		return error;
-
-	bp = xlog_get_bp(log, 1);
-	if (!bp)
-		return -ENOMEM;
-	if (*head_blk == 0) {				/* special case */
-		error = xlog_bread(log, 0, 1, bp, &offset);
+	end_blk = head_blk > tail_blk ? tail_blk : 0;
+	for (i = (int) head_blk - 1; i >= end_blk; i--) {
+		error = xlog_bread(log, i, 1, bp, &offset);
 		if (error)
-			goto done;
+			goto out_error;
 
-		if (xlog_get_cycle(offset) == 0) {
-			*tail_blk = 0;
-			/* leave all other log inited values alone */
-			goto done;
+		if (*(__be32 *) offset == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) {
+			*rblk = i;
+			*rhead = (struct xlog_rec_header *) offset;
+			if (++found == count)
+				break;
 		}
 	}
 
 	/*
-	 * Search backwards looking for log record header block
+	 * If we haven't hit the tail block or the log record header count,
+	 * start looking again from the end of the physical log. Note that
+	 * callers can pass head == tail if the tail is not yet known.
 	 */
-	ASSERT(*head_blk < INT_MAX);
-	for (i = (int)(*head_blk) - 1; i >= 0; i--) {
+	if (tail_blk >= head_blk && found != count) {
+		for (i = log->l_logBBsize - 1; i >= (int) tail_blk; i--) {
+			error = xlog_bread(log, i, 1, bp, &offset);
+			if (error)
+				goto out_error;
+
+			if (*(__be32 *)offset ==
+			    cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) {
+				*wrapped = true;
+				*rblk = i;
+				*rhead = (struct xlog_rec_header *) offset;
+				if (++found == count)
+					break;
+			}
+		}
+	}
+
+	return found;
+
+out_error:
+	return error;
+}
+
+/*
+ * Seek forward in the log for log record headers.
+ *
+ * Given head and tail blocks, walk forward from the tail block until we find
+ * the provided number of records or hit the head block. The return value is the
+ * number of records encountered or a negative error code. The log block and
+ * buffer pointer of the last record seen are returned in rblk and rhead
+ * respectively.
+ */
+STATIC int
+xlog_seek_logrec_hdr(
+	struct xlog		*log,
+	xfs_daddr_t		head_blk,
+	xfs_daddr_t		tail_blk,
+	int			count,
+	struct xfs_buf		*bp,
+	xfs_daddr_t		*rblk,
+	struct xlog_rec_header	**rhead,
+	bool			*wrapped)
+{
+	int			i;
+	int			error;
+	int			found = 0;
+	char			*offset = NULL;
+	xfs_daddr_t		end_blk;
+
+	*wrapped = false;
+
+	/*
+	 * Walk forward from the tail block until we hit the head or the last
+	 * block in the log.
+	 */
+	end_blk = head_blk > tail_blk ? head_blk : log->l_logBBsize - 1;
+	for (i = (int) tail_blk; i <= end_blk; i++) {
 		error = xlog_bread(log, i, 1, bp, &offset);
 		if (error)
-			goto done;
+			goto out_error;
 
-		if (*(__be32 *)offset == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) {
-			found = 1;
-			break;
+		if (*(__be32 *) offset == cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) {
+			*rblk = i;
+			*rhead = (struct xlog_rec_header *) offset;
+			if (++found == count)
+				break;
 		}
 	}
+
 	/*
-	 * If we haven't found the log record header block, start looking
-	 * again from the end of the physical log.  XXXmiken: There should be
-	 * a check here to make sure we didn't search more than N blocks in
-	 * the previous code.
+	 * If we haven't hit the head block or the log record header count,
+	 * start looking again from the start of the physical log.
 	 */
-	if (!found) {
-		for (i = log->l_logBBsize - 1; i >= (int)(*head_blk); i--) {
+	if (tail_blk > head_blk && found != count) {
+		for (i = 0; i < (int) head_blk; i++) {
 			error = xlog_bread(log, i, 1, bp, &offset);
 			if (error)
-				goto done;
+				goto out_error;
 
 			if (*(__be32 *)offset ==
 			    cpu_to_be32(XLOG_HEADER_MAGIC_NUM)) {
-				found = 2;
-				break;
+				*wrapped = true;
+				*rblk = i;
+				*rhead = (struct xlog_rec_header *) offset;
+				if (++found == count)
+					break;
 			}
 		}
 	}
-	if (!found) {
-		xfs_warn(log->l_mp, "%s: couldn't find sync record", __func__);
-		xlog_put_bp(bp);
-		ASSERT(0);
-		return -EIO;
+
+	return found;
+
+out_error:
+	return error;
+}
+
+/*
+ * Check the log tail for torn writes. This is required when torn writes are
+ * detected at the head and the head had to be walked back to a previous record.
+ * The tail of the previous record must now be verified to ensure the torn
+ * writes didn't corrupt the previous tail.
+ *
+ * Return an error if CRC verification fails as recovery cannot proceed.
+ */
+STATIC int
+xlog_verify_tail(
+	struct xlog		*log,
+	xfs_daddr_t		head_blk,
+	xfs_daddr_t		tail_blk)
+{
+	struct xlog_rec_header	*thead;
+	struct xfs_buf		*bp;
+	xfs_daddr_t		first_bad;
+	int			count;
+	int			error = 0;
+	bool			wrapped;
+	xfs_daddr_t		tmp_head;
+
+	bp = xlog_get_bp(log, 1);
+	if (!bp)
+		return -ENOMEM;
+
+	/*
+	 * Seek XLOG_MAX_ICLOGS + 1 records past the current tail record to get
+	 * a temporary head block that points after the last possible
+	 * concurrently written record of the tail.
+	 */
+	count = xlog_seek_logrec_hdr(log, head_blk, tail_blk,
+				     XLOG_MAX_ICLOGS + 1, bp, &tmp_head, &thead,
+				     &wrapped);
+	if (count < 0) {
+		error = count;
+		goto out;
 	}
 
-	/* find blk_no of tail of log */
-	rhead = (xlog_rec_header_t *)offset;
-	*tail_blk = BLOCK_LSN(be64_to_cpu(rhead->h_tail_lsn));
+	/*
+	 * If the call above didn't find XLOG_MAX_ICLOGS + 1 records, we ran
+	 * into the actual log head. tmp_head points to the start of the record
+	 * so update it to the actual head block.
+	 */
+	if (count < XLOG_MAX_ICLOGS + 1)
+		tmp_head = head_blk;
 
 	/*
-	 * Reset log values according to the state of the log when we
-	 * crashed.  In the case where head_blk == 0, we bump curr_cycle
-	 * one because the next write starts a new cycle rather than
-	 * continuing the cycle of the last good log record.  At this
-	 * point we have guaranteed that all partial log records have been
-	 * accounted for.  Therefore, we know that the last good log record
-	 * written was complete and ended exactly on the end boundary
-	 * of the physical log.
+	 * We now have a tail and temporary head block that covers at least
+	 * XLOG_MAX_ICLOGS records from the tail. We need to verify that these
+	 * records were completely written. Run a CRC verification pass from
+	 * tail to head and return the result.
 	 */
-	log->l_prev_block = i;
-	log->l_curr_block = (int)*head_blk;
-	log->l_curr_cycle = be32_to_cpu(rhead->h_cycle);
-	if (found == 2)
-		log->l_curr_cycle++;
-	atomic64_set(&log->l_tail_lsn, be64_to_cpu(rhead->h_tail_lsn));
-	atomic64_set(&log->l_last_sync_lsn, be64_to_cpu(rhead->h_lsn));
-	xlog_assign_grant_head(&log->l_reserve_head.grant, log->l_curr_cycle,
-					BBTOB(log->l_curr_block));
-	xlog_assign_grant_head(&log->l_write_head.grant, log->l_curr_cycle,
-					BBTOB(log->l_curr_block));
+	error = xlog_do_recovery_pass(log, tmp_head, tail_blk,
+				      XLOG_RECOVER_CRCPASS, &first_bad);
+
+out:
+	xlog_put_bp(bp);
+	return error;
+}
+
+/*
+ * Detect and trim torn writes from the head of the log.
+ *
+ * Storage without sector atomicity guarantees can result in torn writes in the
+ * log in the event of a crash. Our only means to detect this scenario is via
+ * CRC verification. While we can't always be certain that CRC verification
+ * failure is due to a torn write vs. an unrelated corruption, we do know that
+ * only a certain number (XLOG_MAX_ICLOGS) of log records can be written out at
+ * one time. Therefore, CRC verify up to XLOG_MAX_ICLOGS records at the head of
+ * the log and treat failures in this range as torn writes as a matter of
+ * policy. In the event of CRC failure, the head is walked back to the last good
+ * record in the log and the tail is updated from that record and verified.
+ */
+STATIC int
+xlog_verify_head(
+	struct xlog		*log,
+	xfs_daddr_t		*head_blk,	/* in/out: unverified head */
+	xfs_daddr_t		*tail_blk,	/* out: tail block */
+	struct xfs_buf		*bp,
+	xfs_daddr_t		*rhead_blk,	/* start blk of last record */
+	struct xlog_rec_header	**rhead,	/* ptr to last record */
+	bool			*wrapped)	/* last rec. wraps phys. log */
+{
+	struct xlog_rec_header	*tmp_rhead;
+	struct xfs_buf		*tmp_bp;
+	xfs_daddr_t		first_bad;
+	xfs_daddr_t		tmp_rhead_blk;
+	int			found;
+	int			error;
+	bool			tmp_wrapped;
+
+	/*
+	 * Check the head of the log for torn writes. Search backwards from the
+	 * head until we hit the tail or the maximum number of log record I/Os
+	 * that could have been in flight at one time. Use a temporary buffer so
+	 * we don't trash the rhead/bp pointers from the caller.
+	 */
+	tmp_bp = xlog_get_bp(log, 1);
+	if (!tmp_bp)
+		return -ENOMEM;
+	error = xlog_rseek_logrec_hdr(log, *head_blk, *tail_blk,
+				      XLOG_MAX_ICLOGS, tmp_bp, &tmp_rhead_blk,
+				      &tmp_rhead, &tmp_wrapped);
+	xlog_put_bp(tmp_bp);
+	if (error < 0)
+		return error;
+
+	/*
+	 * Now run a CRC verification pass over the records starting at the
+	 * block found above to the current head. If a CRC failure occurs, the
+	 * log block of the first bad record is saved in first_bad.
+	 */
+	error = xlog_do_recovery_pass(log, *head_blk, tmp_rhead_blk,
+				      XLOG_RECOVER_CRCPASS, &first_bad);
+	if (error == -EFSBADCRC) {
+		/*
+		 * We've hit a potential torn write. Reset the error and warn
+		 * about it.
+		 */
+		error = 0;
+		xfs_warn(log->l_mp,
+"Torn write (CRC failure) detected at log block 0x%llx. Truncating head block from 0x%llx.",
+			 first_bad, *head_blk);
+
+		/*
+		 * Get the header block and buffer pointer for the last good
+		 * record before the bad record.
+		 *
+		 * Note that xlog_find_tail() clears the blocks at the new head
+		 * (i.e., the records with invalid CRC) if the cycle number
+		 * matches the the current cycle.
+		 */
+		found = xlog_rseek_logrec_hdr(log, first_bad, *tail_blk, 1, bp,
+					      rhead_blk, rhead, wrapped);
+		if (found < 0)
+			return found;
+		if (found == 0)		/* XXX: right thing to do here? */
+			return -EIO;
+
+		/*
+		 * Reset the head block to the starting block of the first bad
+		 * log record and set the tail block based on the last good
+		 * record.
+		 *
+		 * Bail out if the updated head/tail match as this indicates
+		 * possible corruption outside of the acceptable
+		 * (XLOG_MAX_ICLOGS) range. This is a job for xfs_repair...
+		 */
+		*head_blk = first_bad;
+		*tail_blk = BLOCK_LSN(be64_to_cpu((*rhead)->h_tail_lsn));
+		if (*head_blk == *tail_blk) {
+			ASSERT(0);
+			return 0;
+		}
+
+		/*
+		 * Now verify the tail based on the updated head. This is
+		 * required because the torn writes trimmed from the head could
+		 * have been written over the tail of a previous record. Return
+		 * any errors since recovery cannot proceed if the tail is
+		 * corrupt.
+		 *
+		 * XXX: This leaves a gap in truly robust protection from torn
+		 * writes in the log. If the head is behind the tail, the tail
+		 * pushes forward to create some space and then a crash occurs
+		 * causing the writes into the previous record's tail region to
+		 * tear, log recovery isn't able to recover.
+		 *
+		 * How likely is this to occur? If possible, can we do something
+		 * more intelligent here? Is it safe to push the tail forward if
+		 * we can determine that the tail is within the range of the
+		 * torn write (e.g., the kernel can only overwrite the tail if
+		 * it has actually been pushed forward)? Alternatively, could we
+		 * somehow prevent this condition at runtime?
+		 */
+		error = xlog_verify_tail(log, *head_blk, *tail_blk);
+	}
+
+	return error;
+}
+
+/*
+ * Check whether the head of the log points to an unmount record. In other
+ * words, determine whether the log is clean. If so, update the in-core state
+ * appropriately.
+ */
+static int
+xlog_check_unmount_rec(
+	struct xlog		*log,
+	xfs_daddr_t		*head_blk,
+	xfs_daddr_t		*tail_blk,
+	struct xlog_rec_header	*rhead,
+	xfs_daddr_t		rhead_blk,
+	struct xfs_buf		*bp,
+	bool			*clean)
+{
+	struct xlog_op_header	*op_head;
+	xfs_daddr_t		umount_data_blk;
+	xfs_daddr_t		after_umount_blk;
+	int			hblks;
+	int			error;
+	char			*offset;
+
+	*clean = false;
 
 	/*
-	 * Look for unmount record.  If we find it, then we know there
-	 * was a clean unmount.  Since 'i' could be the last block in
-	 * the physical log, we convert to a log block before comparing
-	 * to the head_blk.
+	 * Look for unmount record. If we find it, then we know there was a
+	 * clean unmount. Since 'i' could be the last block in the physical
+	 * log, we convert to a log block before comparing to the head_blk.
 	 *
-	 * Save the current tail lsn to use to pass to
-	 * xlog_clear_stale_blocks() below.  We won't want to clear the
-	 * unmount record if there is one, so we pass the lsn of the
-	 * unmount record rather than the block after it.
+	 * Save the current tail lsn to use to pass to xlog_clear_stale_blocks()
+	 * below. We won't want to clear the unmount record if there is one, so
+	 * we pass the lsn of the unmount record rather than the block after it.
 	 */
 	if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) {
 		int	h_size = be32_to_cpu(rhead->h_size);
@@ -1014,22 +1245,22 @@ xlog_find_tail(
 	} else {
 		hblks = 1;
 	}
-	after_umount_blk = (i + hblks + (int)
-		BTOBB(be32_to_cpu(rhead->h_len))) % log->l_logBBsize;
-	tail_lsn = atomic64_read(&log->l_tail_lsn);
+	after_umount_blk = rhead_blk + hblks + BTOBB(be32_to_cpu(rhead->h_len));
+	after_umount_blk = do_mod(after_umount_blk, log->l_logBBsize);
 	if (*head_blk == after_umount_blk &&
 	    be32_to_cpu(rhead->h_num_logops) == 1) {
-		umount_data_blk = (i + hblks) % log->l_logBBsize;
+		umount_data_blk = rhead_blk + hblks;
+		umount_data_blk = do_mod(umount_data_blk, log->l_logBBsize);
 		error = xlog_bread(log, umount_data_blk, 1, bp, &offset);
 		if (error)
-			goto done;
+			return error;
 
-		op_head = (xlog_op_header_t *)offset;
+		op_head = (struct xlog_op_header *)offset;
 		if (op_head->oh_flags & XLOG_UNMOUNT_TRANS) {
 			/*
-			 * Set tail and last sync so that newly written
-			 * log records will point recovery to after the
-			 * current unmount record.
+			 * Set tail and last sync so that newly written log
+			 * records will point recovery to after the current
+			 * unmount record.
 			 */
 			xlog_assign_atomic_lsn(&log->l_tail_lsn,
 					log->l_curr_cycle, after_umount_blk);
@@ -1037,17 +1268,167 @@ xlog_find_tail(
 					log->l_curr_cycle, after_umount_blk);
 			*tail_blk = after_umount_blk;
 
-			/*
-			 * Note that the unmount was clean. If the unmount
-			 * was not clean, we need to know this to rebuild the
-			 * superblock counters from the perag headers if we
-			 * have a filesystem using non-persistent counters.
-			 */
-			log->l_mp->m_flags |= XFS_MOUNT_WAS_CLEAN;
+			*clean = true;
+		}
+	}
+
+	return 0;
+}
+
+static void
+xlog_set_state(
+	struct xlog		*log,
+	xfs_daddr_t		head_blk,
+	struct xlog_rec_header	*rhead,
+	xfs_daddr_t		rhead_blk,
+	bool			bump_cycle)
+{
+	/*
+	 * Reset log values according to the state of the log when we
+	 * crashed.  In the case where head_blk == 0, we bump curr_cycle
+	 * one because the next write starts a new cycle rather than
+	 * continuing the cycle of the last good log record.  At this
+	 * point we have guaranteed that all partial log records have been
+	 * accounted for.  Therefore, we know that the last good log record
+	 * written was complete and ended exactly on the end boundary
+	 * of the physical log.
+	 */
+	log->l_prev_block = rhead_blk;
+	log->l_curr_block = (int)head_blk;
+	log->l_curr_cycle = be32_to_cpu(rhead->h_cycle);
+	if (bump_cycle)
+		log->l_curr_cycle++;
+	atomic64_set(&log->l_tail_lsn, be64_to_cpu(rhead->h_tail_lsn));
+	atomic64_set(&log->l_last_sync_lsn, be64_to_cpu(rhead->h_lsn));
+	xlog_assign_grant_head(&log->l_reserve_head.grant, log->l_curr_cycle,
+					BBTOB(log->l_curr_block));
+	xlog_assign_grant_head(&log->l_write_head.grant, log->l_curr_cycle,
+					BBTOB(log->l_curr_block));
+}
+
+/*
+ * Find the sync block number or the tail of the log.
+ *
+ * This will be the block number of the last record to have its
+ * associated buffers synced to disk.  Every log record header has
+ * a sync lsn embedded in it.  LSNs hold block numbers, so it is easy
+ * to get a sync block number.  The only concern is to figure out which
+ * log record header to believe.
+ *
+ * The following algorithm uses the log record header with the largest
+ * lsn.  The entire log record does not need to be valid.  We only care
+ * that the header is valid.
+ *
+ * We could speed up search by using current head_blk buffer, but it is not
+ * available.
+ */
+STATIC int
+xlog_find_tail(
+	struct xlog		*log,
+	xfs_daddr_t		*head_blk,
+	xfs_daddr_t		*tail_blk)
+{
+	xlog_rec_header_t	*rhead;
+	char			*offset = NULL;
+	xfs_buf_t		*bp;
+	int			error;
+	xfs_daddr_t		rhead_blk;
+	xfs_lsn_t		tail_lsn;
+	bool			wrapped = false;
+	bool			clean = false;
+
+	/*
+	 * Find previous log record
+	 */
+	if ((error = xlog_find_head(log, head_blk)))
+		return error;
+	ASSERT(*head_blk < INT_MAX);
+
+	bp = xlog_get_bp(log, 1);
+	if (!bp)
+		return -ENOMEM;
+	if (*head_blk == 0) {				/* special case */
+		error = xlog_bread(log, 0, 1, bp, &offset);
+		if (error)
+			goto done;
+
+		if (xlog_get_cycle(offset) == 0) {
+			*tail_blk = 0;
+			/* leave all other log inited values alone */
+			goto done;
+		}
+	}
+
+	/*
+	 * Search backwards through the log looking for the log record header
+	 * block. This wraps all the way back around to the head so something is
+	 * seriously wrong if we can't find it.
+	 */
+	error = xlog_rseek_logrec_hdr(log, *head_blk, *head_blk, 1, bp,
+				      &rhead_blk, &rhead, &wrapped);
+	if (error < 0)
+		return error;
+	if (!error) {
+		xfs_warn(log->l_mp, "%s: couldn't find sync record", __func__);
+		return -EIO;
+	}
+	*tail_blk = BLOCK_LSN(be64_to_cpu(rhead->h_tail_lsn));
+
+	/*
+	 * Set the log state based on the current head record.
+	 */
+	xlog_set_state(log, *head_blk, rhead, rhead_blk, wrapped);
+	tail_lsn = atomic64_read(&log->l_tail_lsn);
+
+	/*
+	 * Look for an unmount record at the head of the log. This sets the log
+	 * state to determine whether recovery is necessary.
+	 */
+	error = xlog_check_unmount_rec(log, head_blk, tail_blk, rhead,
+				       rhead_blk, bp, &clean);
+	if (error)
+		goto done;
+
+	/*
+	 * Verify the log head if the log is not clean (e.g., we have anything
+	 * but an unmount record at the head). This uses CRC verification to
+	 * detect and trim torn writes. If discovered, CRC failures are
+	 * considered torn writes and the log head is trimmed accordingly.
+	 *
+	 * Note that we can only run CRC verification when the log is dirty
+	 * because there's no guarantee that the log data behind an unmount
+	 * record is compatible with the current architecture.
+	 */
+	if (!clean) {
+		xfs_daddr_t	orig_head = *head_blk;
+
+		error = xlog_verify_head(log, head_blk, tail_blk, bp,
+					 &rhead_blk, &rhead, &wrapped);
+		if (error)
+			goto done;
+
+		/* update in-core state again if the head changed */
+		if (*head_blk != orig_head) {
+			xlog_set_state(log, *head_blk, rhead, rhead_blk,
+				       wrapped);
+			tail_lsn = atomic64_read(&log->l_tail_lsn);
+			error = xlog_check_unmount_rec(log, head_blk, tail_blk,
+						       rhead, rhead_blk, bp,
+						       &clean);
+			if (error)
+				goto done;
 		}
 	}
 
 	/*
+	 * Note that the unmount was clean. If the unmount was not clean, we
+	 * need to know this to rebuild the superblock counters from the perag
+	 * headers if we have a filesystem using non-persistent counters.
+	 */
+	if (clean)
+		log->l_mp->m_flags |= XFS_MOUNT_WAS_CLEAN;
+
+	/*
 	 * Make sure that there are no blocks in front of the head
 	 * with the same cycle number as the head.  This can happen
 	 * because we allow multiple outstanding log writes concurrently,
@@ -2156,6 +2537,13 @@ xlog_recover_validate_buf_type(
 		}
 		bp->b_ops = &xfs_sb_buf_ops;
 		break;
+#ifdef CONFIG_XFS_RT
+	case XFS_BLFT_RTBITMAP_BUF:
+	case XFS_BLFT_RTSUMMARY_BUF:
+		/* no magic numbers for verification of RT buffers */
+		bp->b_ops = &xfs_rtbuf_ops;
+		break;
+#endif /* CONFIG_XFS_RT */
 	default:
 		xfs_warn(mp, "Unknown buffer type %d!",
 			 xfs_blft_from_flags(buf_f));
@@ -2476,7 +2864,7 @@ xfs_recover_inode_owner_change(
 		return -ENOMEM;
 
 	/* instantiate the inode */
-	xfs_dinode_from_disk(&ip->i_d, dip);
+	xfs_inode_from_disk(ip, dip);
 	ASSERT(ip->i_d.di_version >= 3);
 
 	error = xfs_iformat_fork(ip, dip);
@@ -2522,7 +2910,7 @@ xlog_recover_inode_pass2(
 	int			error;
 	int			attr_index;
 	uint			fields;
-	xfs_icdinode_t		*dicp;
+	struct xfs_log_dinode	*ldip;
 	uint			isize;
 	int			need_free = 0;
 
@@ -2575,8 +2963,8 @@ xlog_recover_inode_pass2(
 		error = -EFSCORRUPTED;
 		goto out_release;
 	}
-	dicp = item->ri_buf[1].i_addr;
-	if (unlikely(dicp->di_magic != XFS_DINODE_MAGIC)) {
+	ldip = item->ri_buf[1].i_addr;
+	if (unlikely(ldip->di_magic != XFS_DINODE_MAGIC)) {
 		xfs_alert(mp,
 			"%s: Bad inode log record, rec ptr 0x%p, ino %Ld",
 			__func__, item, in_f->ilf_ino);
@@ -2612,13 +3000,13 @@ xlog_recover_inode_pass2(
 	 * to skip replay when the on disk inode is newer than the log one
 	 */
 	if (!xfs_sb_version_hascrc(&mp->m_sb) &&
-	    dicp->di_flushiter < be16_to_cpu(dip->di_flushiter)) {
+	    ldip->di_flushiter < be16_to_cpu(dip->di_flushiter)) {
 		/*
 		 * Deal with the wrap case, DI_MAX_FLUSH is less
 		 * than smaller numbers
 		 */
 		if (be16_to_cpu(dip->di_flushiter) == DI_MAX_FLUSH &&
-		    dicp->di_flushiter < (DI_MAX_FLUSH >> 1)) {
+		    ldip->di_flushiter < (DI_MAX_FLUSH >> 1)) {
 			/* do nothing */
 		} else {
 			trace_xfs_log_recover_inode_skip(log, in_f);
@@ -2628,13 +3016,13 @@ xlog_recover_inode_pass2(
 	}
 
 	/* Take the opportunity to reset the flush iteration count */
-	dicp->di_flushiter = 0;
+	ldip->di_flushiter = 0;
 
-	if (unlikely(S_ISREG(dicp->di_mode))) {
-		if ((dicp->di_format != XFS_DINODE_FMT_EXTENTS) &&
-		    (dicp->di_format != XFS_DINODE_FMT_BTREE)) {
+	if (unlikely(S_ISREG(ldip->di_mode))) {
+		if ((ldip->di_format != XFS_DINODE_FMT_EXTENTS) &&
+		    (ldip->di_format != XFS_DINODE_FMT_BTREE)) {
 			XFS_CORRUPTION_ERROR("xlog_recover_inode_pass2(3)",
-					 XFS_ERRLEVEL_LOW, mp, dicp);
+					 XFS_ERRLEVEL_LOW, mp, ldip);
 			xfs_alert(mp,
 		"%s: Bad regular inode log record, rec ptr 0x%p, "
 		"ino ptr = 0x%p, ino bp = 0x%p, ino %Ld",
@@ -2642,12 +3030,12 @@ xlog_recover_inode_pass2(
 			error = -EFSCORRUPTED;
 			goto out_release;
 		}
-	} else if (unlikely(S_ISDIR(dicp->di_mode))) {
-		if ((dicp->di_format != XFS_DINODE_FMT_EXTENTS) &&
-		    (dicp->di_format != XFS_DINODE_FMT_BTREE) &&
-		    (dicp->di_format != XFS_DINODE_FMT_LOCAL)) {
+	} else if (unlikely(S_ISDIR(ldip->di_mode))) {
+		if ((ldip->di_format != XFS_DINODE_FMT_EXTENTS) &&
+		    (ldip->di_format != XFS_DINODE_FMT_BTREE) &&
+		    (ldip->di_format != XFS_DINODE_FMT_LOCAL)) {
 			XFS_CORRUPTION_ERROR("xlog_recover_inode_pass2(4)",
-					     XFS_ERRLEVEL_LOW, mp, dicp);
+					     XFS_ERRLEVEL_LOW, mp, ldip);
 			xfs_alert(mp,
 		"%s: Bad dir inode log record, rec ptr 0x%p, "
 		"ino ptr = 0x%p, ino bp = 0x%p, ino %Ld",
@@ -2656,32 +3044,32 @@ xlog_recover_inode_pass2(
 			goto out_release;
 		}
 	}
-	if (unlikely(dicp->di_nextents + dicp->di_anextents > dicp->di_nblocks)){
+	if (unlikely(ldip->di_nextents + ldip->di_anextents > ldip->di_nblocks)){
 		XFS_CORRUPTION_ERROR("xlog_recover_inode_pass2(5)",
-				     XFS_ERRLEVEL_LOW, mp, dicp);
+				     XFS_ERRLEVEL_LOW, mp, ldip);
 		xfs_alert(mp,
 	"%s: Bad inode log record, rec ptr 0x%p, dino ptr 0x%p, "
 	"dino bp 0x%p, ino %Ld, total extents = %d, nblocks = %Ld",
 			__func__, item, dip, bp, in_f->ilf_ino,
-			dicp->di_nextents + dicp->di_anextents,
-			dicp->di_nblocks);
+			ldip->di_nextents + ldip->di_anextents,
+			ldip->di_nblocks);
 		error = -EFSCORRUPTED;
 		goto out_release;
 	}
-	if (unlikely(dicp->di_forkoff > mp->m_sb.sb_inodesize)) {
+	if (unlikely(ldip->di_forkoff > mp->m_sb.sb_inodesize)) {
 		XFS_CORRUPTION_ERROR("xlog_recover_inode_pass2(6)",
-				     XFS_ERRLEVEL_LOW, mp, dicp);
+				     XFS_ERRLEVEL_LOW, mp, ldip);
 		xfs_alert(mp,
 	"%s: Bad inode log record, rec ptr 0x%p, dino ptr 0x%p, "
 	"dino bp 0x%p, ino %Ld, forkoff 0x%x", __func__,
-			item, dip, bp, in_f->ilf_ino, dicp->di_forkoff);
+			item, dip, bp, in_f->ilf_ino, ldip->di_forkoff);
 		error = -EFSCORRUPTED;
 		goto out_release;
 	}
-	isize = xfs_icdinode_size(dicp->di_version);
+	isize = xfs_log_dinode_size(ldip->di_version);
 	if (unlikely(item->ri_buf[1].i_len > isize)) {
 		XFS_CORRUPTION_ERROR("xlog_recover_inode_pass2(7)",
-				     XFS_ERRLEVEL_LOW, mp, dicp);
+				     XFS_ERRLEVEL_LOW, mp, ldip);
 		xfs_alert(mp,
 			"%s: Bad inode log record length %d, rec ptr 0x%p",
 			__func__, item->ri_buf[1].i_len, item);
@@ -2689,8 +3077,8 @@ xlog_recover_inode_pass2(
 		goto out_release;
 	}
 
-	/* The core is in in-core format */
-	xfs_dinode_to_disk(dip, dicp);
+	/* recover the log dinode inode into the on disk inode */
+	xfs_log_dinode_to_disk(ldip, dip);
 
 	/* the rest is in on-disk format */
 	if (item->ri_buf[1].i_len > isize) {
@@ -3204,6 +3592,7 @@ xlog_recover_dquot_ra_pass2(
 	struct xfs_disk_dquot	*recddq;
 	struct xfs_dq_logformat	*dq_f;
 	uint			type;
+	int			len;
 
 
 	if (mp->m_qflags == 0)
@@ -3224,8 +3613,12 @@ xlog_recover_dquot_ra_pass2(
 	ASSERT(dq_f);
 	ASSERT(dq_f->qlf_len == 1);
 
-	xfs_buf_readahead(mp->m_ddev_targp, dq_f->qlf_blkno,
-			  XFS_FSB_TO_BB(mp, dq_f->qlf_len), NULL);
+	len = XFS_FSB_TO_BB(mp, dq_f->qlf_len);
+	if (xlog_peek_buffer_cancelled(log, dq_f->qlf_blkno, len, 0))
+		return;
+
+	xfs_buf_readahead(mp->m_ddev_targp, dq_f->qlf_blkno, len,
+			  &xfs_dquot_buf_ra_ops);
 }
 
 STATIC void
@@ -3431,7 +3824,7 @@ xlog_recover_add_to_cont_trans(
 	 * previous record. Copy the rest of the header.
 	 */
 	if (list_empty(&trans->r_itemq)) {
-		ASSERT(len < sizeof(struct xfs_trans_header));
+		ASSERT(len <= sizeof(struct xfs_trans_header));
 		if (len > sizeof(struct xfs_trans_header)) {
 			xfs_warn(log->l_mp, "%s: bad header length", __func__);
 			return -EIO;
@@ -4015,8 +4408,8 @@ xlog_recover_process_one_iunlink(
 	if (error)
 		goto fail_iput;
 
-	ASSERT(ip->i_d.di_nlink == 0);
-	ASSERT(ip->i_d.di_mode != 0);
+	ASSERT(VFS_I(ip)->i_nlink == 0);
+	ASSERT(VFS_I(ip)->i_mode != 0);
 
 	/* setup for the next pass */
 	agino = be32_to_cpu(dip->di_next_unlinked);
@@ -4118,25 +4511,68 @@ xlog_recover_process_iunlinks(
 	mp->m_dmevmask = mp_dmevmask;
 }
 
+STATIC int
+xlog_unpack_data(
+	struct xlog_rec_header	*rhead,
+	char			*dp,
+	struct xlog		*log)
+{
+	int			i, j, k;
+
+	for (i = 0; i < BTOBB(be32_to_cpu(rhead->h_len)) &&
+		  i < (XLOG_HEADER_CYCLE_SIZE / BBSIZE); i++) {
+		*(__be32 *)dp = *(__be32 *)&rhead->h_cycle_data[i];
+		dp += BBSIZE;
+	}
+
+	if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) {
+		xlog_in_core_2_t *xhdr = (xlog_in_core_2_t *)rhead;
+		for ( ; i < BTOBB(be32_to_cpu(rhead->h_len)); i++) {
+			j = i / (XLOG_HEADER_CYCLE_SIZE / BBSIZE);
+			k = i % (XLOG_HEADER_CYCLE_SIZE / BBSIZE);
+			*(__be32 *)dp = xhdr[j].hic_xheader.xh_cycle_data[k];
+			dp += BBSIZE;
+		}
+	}
+
+	return 0;
+}
+
 /*
- * Upack the log buffer data and crc check it. If the check fails, issue a
- * warning if and only if the CRC in the header is non-zero. This makes the
- * check an advisory warning, and the zero CRC check will prevent failure
- * warnings from being emitted when upgrading the kernel from one that does not
- * add CRCs by default.
- *
- * When filesystems are CRC enabled, this CRC mismatch becomes a fatal log
- * corruption failure
+ * CRC check, unpack and process a log record.
  */
 STATIC int
-xlog_unpack_data_crc(
+xlog_recover_process(
+	struct xlog		*log,
+	struct hlist_head	rhash[],
 	struct xlog_rec_header	*rhead,
 	char			*dp,
-	struct xlog		*log)
+	int			pass)
 {
+	int			error;
 	__le32			crc;
 
 	crc = xlog_cksum(log, rhead, dp, be32_to_cpu(rhead->h_len));
+
+	/*
+	 * Nothing else to do if this is a CRC verification pass. Just return
+	 * if this a record with a non-zero crc. Unfortunately, mkfs always
+	 * sets h_crc to 0 so we must consider this valid even on v5 supers.
+	 * Otherwise, return EFSBADCRC on failure so the callers up the stack
+	 * know precisely what failed.
+	 */
+	if (pass == XLOG_RECOVER_CRCPASS) {
+		if (rhead->h_crc && crc != rhead->h_crc)
+			return -EFSBADCRC;
+		return 0;
+	}
+
+	/*
+	 * We're in the normal recovery path. Issue a warning if and only if the
+	 * CRC in the header is non-zero. This is an advisory warning and the
+	 * zero CRC check prevents warnings from being emitted when upgrading
+	 * the kernel from one that does not add CRCs by default.
+	 */
 	if (crc != rhead->h_crc) {
 		if (rhead->h_crc || xfs_sb_version_hascrc(&log->l_mp->m_sb)) {
 			xfs_alert(log->l_mp,
@@ -4147,47 +4583,18 @@ xlog_unpack_data_crc(
 		}
 
 		/*
-		 * If we've detected a log record corruption, then we can't
-		 * recover past this point. Abort recovery if we are enforcing
-		 * CRC protection by punting an error back up the stack.
+		 * If the filesystem is CRC enabled, this mismatch becomes a
+		 * fatal log corruption failure.
 		 */
 		if (xfs_sb_version_hascrc(&log->l_mp->m_sb))
 			return -EFSCORRUPTED;
 	}
 
-	return 0;
-}
-
-STATIC int
-xlog_unpack_data(
-	struct xlog_rec_header	*rhead,
-	char			*dp,
-	struct xlog		*log)
-{
-	int			i, j, k;
-	int			error;
-
-	error = xlog_unpack_data_crc(rhead, dp, log);
+	error = xlog_unpack_data(rhead, dp, log);
 	if (error)
 		return error;
 
-	for (i = 0; i < BTOBB(be32_to_cpu(rhead->h_len)) &&
-		  i < (XLOG_HEADER_CYCLE_SIZE / BBSIZE); i++) {
-		*(__be32 *)dp = *(__be32 *)&rhead->h_cycle_data[i];
-		dp += BBSIZE;
-	}
-
-	if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) {
-		xlog_in_core_2_t *xhdr = (xlog_in_core_2_t *)rhead;
-		for ( ; i < BTOBB(be32_to_cpu(rhead->h_len)); i++) {
-			j = i / (XLOG_HEADER_CYCLE_SIZE / BBSIZE);
-			k = i % (XLOG_HEADER_CYCLE_SIZE / BBSIZE);
-			*(__be32 *)dp = xhdr[j].hic_xheader.xh_cycle_data[k];
-			dp += BBSIZE;
-		}
-	}
-
-	return 0;
+	return xlog_recover_process_data(log, rhash, rhead, dp, pass);
 }
 
 STATIC int
@@ -4239,18 +4646,21 @@ xlog_do_recovery_pass(
 	struct xlog		*log,
 	xfs_daddr_t		head_blk,
 	xfs_daddr_t		tail_blk,
-	int			pass)
+	int			pass,
+	xfs_daddr_t		*first_bad)	/* out: first bad log rec */
 {
 	xlog_rec_header_t	*rhead;
 	xfs_daddr_t		blk_no;
+	xfs_daddr_t		rhead_blk;
 	char			*offset;
 	xfs_buf_t		*hbp, *dbp;
-	int			error = 0, h_size;
+	int			error = 0, h_size, h_len;
 	int			bblks, split_bblks;
 	int			hblks, split_hblks, wrapped_hblks;
 	struct hlist_head	rhash[XLOG_RHASH_SIZE];
 
 	ASSERT(head_blk != tail_blk);
+	rhead_blk = 0;
 
 	/*
 	 * Read the header of the tail block and get the iclog buffer size from
@@ -4274,7 +4684,31 @@ xlog_do_recovery_pass(
 		error = xlog_valid_rec_header(log, rhead, tail_blk);
 		if (error)
 			goto bread_err1;
+
+		/*
+		 * xfsprogs has a bug where record length is based on lsunit but
+		 * h_size (iclog size) is hardcoded to 32k. Now that we
+		 * unconditionally CRC verify the unmount record, this means the
+		 * log buffer can be too small for the record and cause an
+		 * overrun.
+		 *
+		 * Detect this condition here. Use lsunit for the buffer size as
+		 * long as this looks like the mkfs case. Otherwise, return an
+		 * error to avoid a buffer overrun.
+		 */
 		h_size = be32_to_cpu(rhead->h_size);
+		h_len = be32_to_cpu(rhead->h_len);
+		if (h_len > h_size) {
+			if (h_len <= log->l_mp->m_logbsize &&
+			    be32_to_cpu(rhead->h_num_logops) == 1) {
+				xfs_warn(log->l_mp,
+		"invalid iclog size (%d bytes), using lsunit (%d bytes)",
+					 h_size, log->l_mp->m_logbsize);
+				h_size = log->l_mp->m_logbsize;
+			} else
+				return -EFSCORRUPTED;
+		}
+
 		if ((be32_to_cpu(rhead->h_version) & XLOG_VERSION_2) &&
 		    (h_size > XLOG_HEADER_CYCLE_SIZE)) {
 			hblks = h_size / XLOG_HEADER_CYCLE_SIZE;
@@ -4301,7 +4735,7 @@ xlog_do_recovery_pass(
 	}
 
 	memset(rhash, 0, sizeof(rhash));
-	blk_no = tail_blk;
+	blk_no = rhead_blk = tail_blk;
 	if (tail_blk > head_blk) {
 		/*
 		 * Perform recovery around the end of the physical log.
@@ -4408,19 +4842,18 @@ xlog_do_recovery_pass(
 					goto bread_err2;
 			}
 
-			error = xlog_unpack_data(rhead, offset, log);
+			error = xlog_recover_process(log, rhash, rhead, offset,
+						     pass);
 			if (error)
 				goto bread_err2;
 
-			error = xlog_recover_process_data(log, rhash,
-							rhead, offset, pass);
-			if (error)
-				goto bread_err2;
 			blk_no += bblks;
+			rhead_blk = blk_no;
 		}
 
 		ASSERT(blk_no >= log->l_logBBsize);
 		blk_no -= log->l_logBBsize;
+		rhead_blk = blk_no;
 	}
 
 	/* read first part of physical log */
@@ -4441,21 +4874,22 @@ xlog_do_recovery_pass(
 		if (error)
 			goto bread_err2;
 
-		error = xlog_unpack_data(rhead, offset, log);
+		error = xlog_recover_process(log, rhash, rhead, offset, pass);
 		if (error)
 			goto bread_err2;
 
-		error = xlog_recover_process_data(log, rhash,
-						rhead, offset, pass);
-		if (error)
-			goto bread_err2;
 		blk_no += bblks + hblks;
+		rhead_blk = blk_no;
 	}
 
  bread_err2:
 	xlog_put_bp(dbp);
  bread_err1:
 	xlog_put_bp(hbp);
+
+	if (error && first_bad)
+		*first_bad = rhead_blk;
+
 	return error;
 }
 
@@ -4493,7 +4927,7 @@ xlog_do_log_recovery(
 		INIT_LIST_HEAD(&log->l_buf_cancel_table[i]);
 
 	error = xlog_do_recovery_pass(log, head_blk, tail_blk,
-				      XLOG_RECOVER_PASS1);
+				      XLOG_RECOVER_PASS1, NULL);
 	if (error != 0) {
 		kmem_free(log->l_buf_cancel_table);
 		log->l_buf_cancel_table = NULL;
@@ -4504,7 +4938,7 @@ xlog_do_log_recovery(
 	 * When it is complete free the table of buf cancel items.
 	 */
 	error = xlog_do_recovery_pass(log, head_blk, tail_blk,
-				      XLOG_RECOVER_PASS2);
+				      XLOG_RECOVER_PASS2, NULL);
 #ifdef DEBUG
 	if (!error) {
 		int	i;
@@ -4529,6 +4963,7 @@ xlog_do_recover(
 	xfs_daddr_t	head_blk,
 	xfs_daddr_t	tail_blk)
 {
+	struct xfs_mount *mp = log->l_mp;
 	int		error;
 	xfs_buf_t	*bp;
 	xfs_sb_t	*sbp;
@@ -4543,7 +4978,7 @@ xlog_do_recover(
 	/*
 	 * If IO errors happened during recovery, bail out.
 	 */
-	if (XFS_FORCED_SHUTDOWN(log->l_mp)) {
+	if (XFS_FORCED_SHUTDOWN(mp)) {
 		return -EIO;
 	}
 
@@ -4556,22 +4991,21 @@ xlog_do_recover(
 	 * or iunlinks they will have some entries in the AIL; so we look at
 	 * the AIL to determine how to set the tail_lsn.
 	 */
-	xlog_assign_tail_lsn(log->l_mp);
+	xlog_assign_tail_lsn(mp);
 
 	/*
 	 * Now that we've finished replaying all buffer and inode
 	 * updates, re-read in the superblock and reverify it.
 	 */
-	bp = xfs_getsb(log->l_mp, 0);
-	XFS_BUF_UNDONE(bp);
-	ASSERT(!(XFS_BUF_ISWRITE(bp)));
-	XFS_BUF_READ(bp);
-	XFS_BUF_UNASYNC(bp);
+	bp = xfs_getsb(mp, 0);
+	bp->b_flags &= ~(XBF_DONE | XBF_ASYNC);
+	ASSERT(!(bp->b_flags & XBF_WRITE));
+	bp->b_flags |= XBF_READ;
 	bp->b_ops = &xfs_sb_buf_ops;
 
 	error = xfs_buf_submit_wait(bp);
 	if (error) {
-		if (!XFS_FORCED_SHUTDOWN(log->l_mp)) {
+		if (!XFS_FORCED_SHUTDOWN(mp)) {
 			xfs_buf_ioerror_alert(bp, __func__);
 			ASSERT(0);
 		}
@@ -4580,14 +5014,17 @@ xlog_do_recover(
 	}
 
 	/* Convert superblock from on-disk format */
-	sbp = &log->l_mp->m_sb;
+	sbp = &mp->m_sb;
 	xfs_sb_from_disk(sbp, XFS_BUF_TO_SBP(bp));
-	ASSERT(sbp->sb_magicnum == XFS_SB_MAGIC);
-	ASSERT(xfs_sb_good_version(sbp));
-	xfs_reinit_percpu_counters(log->l_mp);
-
 	xfs_buf_relse(bp);
 
+	/* re-initialise in-core superblock and geometry structures */
+	xfs_reinit_percpu_counters(mp);
+	error = xfs_initialize_perag(mp, sbp->sb_agcount, &mp->m_maxagi);
+	if (error) {
+		xfs_warn(mp, "Failed post-recovery per-ag init: %d", error);
+		return error;
+	}
 
 	xlog_recover_check_summary(log);
 
@@ -4609,9 +5046,19 @@ xlog_recover(
 	int		error;
 
 	/* find the tail of the log */
-	if ((error = xlog_find_tail(log, &head_blk, &tail_blk)))
+	error = xlog_find_tail(log, &head_blk, &tail_blk);
+	if (error)
 		return error;
 
+	/*
+	 * The superblock was read before the log was available and thus the LSN
+	 * could not be verified. Check the superblock LSN against the current
+	 * LSN now that it's known.
+	 */
+	if (xfs_sb_version_hascrc(&log->l_mp->m_sb) &&
+	    !xfs_log_check_lsn(log->l_mp, log->l_mp->m_sb.sb_lsn))
+		return -EINVAL;
+
 	if (tail_blk != head_blk) {
 		/* There used to be a comment here:
 		 *
diff --git a/fs/xfs/xfs_message.c b/fs/xfs/xfs_message.c
index d8b67547ab34..11792d888e4e 100644
--- a/fs/xfs/xfs_message.c
+++ b/fs/xfs/xfs_message.c
@@ -17,6 +17,7 @@
 
 #include "xfs.h"
 #include "xfs_fs.h"
+#include "xfs_error.h"
 #include "xfs_format.h"
 #include "xfs_log_format.h"
 #include "xfs_trans_resv.h"
@@ -43,6 +44,7 @@ void func(const struct xfs_mount *mp, const char *fmt, ...)	\
 {								\
 	struct va_format	vaf;				\
 	va_list			args;				\
+	int			level;				\
 								\
 	va_start(args, fmt);					\
 								\
@@ -51,6 +53,11 @@ void func(const struct xfs_mount *mp, const char *fmt, ...)	\
 								\
 	__xfs_printk(kern_level, mp, &vaf);			\
 	va_end(args);						\
+								\
+	if (!kstrtoint(kern_level, 0, &level) &&		\
+	    level <= LOGLEVEL_ERR &&				\
+	    xfs_error_level >= XFS_ERRLEVEL_HIGH)		\
+		xfs_stack_trace();				\
 }								\
 
 define_xfs_printk_level(xfs_emerg, KERN_EMERG);
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index bf92e0c037c7..cfd4210dd015 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -47,6 +47,16 @@ static DEFINE_MUTEX(xfs_uuid_table_mutex);
 static int xfs_uuid_table_size;
 static uuid_t *xfs_uuid_table;
 
+void
+xfs_uuid_table_free(void)
+{
+	if (xfs_uuid_table_size == 0)
+		return;
+	kmem_free(xfs_uuid_table);
+	xfs_uuid_table = NULL;
+	xfs_uuid_table_size = 0;
+}
+
 /*
  * See if the UUID is unique among mounted XFS filesystems.
  * Mount fails if UUID is nil or a FS with the same UUID is already mounted.
@@ -161,7 +171,7 @@ xfs_sb_validate_fsb_count(
 	ASSERT(sbp->sb_blocklog >= BBSHIFT);
 
 	/* Limited by ULONG_MAX of page cache index */
-	if (nblocks >> (PAGE_CACHE_SHIFT - sbp->sb_blocklog) > ULONG_MAX)
+	if (nblocks >> (PAGE_SHIFT - sbp->sb_blocklog) > ULONG_MAX)
 		return -EFBIG;
 	return 0;
 }
@@ -175,9 +185,6 @@ xfs_initialize_perag(
 	xfs_agnumber_t	index;
 	xfs_agnumber_t	first_initialised = 0;
 	xfs_perag_t	*pag;
-	xfs_agino_t	agino;
-	xfs_ino_t	ino;
-	xfs_sb_t	*sbp = &mp->m_sb;
 	int		error = -ENOMEM;
 
 	/*
@@ -220,22 +227,7 @@ xfs_initialize_perag(
 		radix_tree_preload_end();
 	}
 
-	/*
-	 * If we mount with the inode64 option, or no inode overflows
-	 * the legacy 32-bit address space clear the inode32 option.
-	 */
-	agino = XFS_OFFBNO_TO_AGINO(mp, sbp->sb_agblocks - 1, 0);
-	ino = XFS_AGINO_TO_INO(mp, agcount - 1, agino);
-
-	if ((mp->m_flags & XFS_MOUNT_SMALL_INUMS) && ino > XFS_MAXINUMBER_32)
-		mp->m_flags |= XFS_MOUNT_32BITINODES;
-	else
-		mp->m_flags &= ~XFS_MOUNT_32BITINODES;
-
-	if (mp->m_flags & XFS_MOUNT_32BITINODES)
-		index = xfs_set_inode32(mp, agcount);
-	else
-		index = xfs_set_inode64(mp, agcount);
+	index = xfs_set_inode_alloc(mp, agcount);
 
 	if (maxagi)
 		*maxagi = index;
@@ -693,10 +685,15 @@ xfs_mountfs(
 	if (error)
 		goto out;
 
-	error = xfs_uuid_mount(mp);
+	error = xfs_sysfs_init(&mp->m_stats.xs_kobj, &xfs_stats_ktype,
+			       &mp->m_kobj, "stats");
 	if (error)
 		goto out_remove_sysfs;
 
+	error = xfs_uuid_mount(mp);
+	if (error)
+		goto out_del_stats;
+
 	/*
 	 * Set the minimum read and write sizes
 	 */
@@ -850,7 +847,7 @@ xfs_mountfs(
 
 	ASSERT(rip != NULL);
 
-	if (unlikely(!S_ISDIR(rip->i_d.di_mode))) {
+	if (unlikely(!S_ISDIR(VFS_I(rip)->i_mode))) {
 		xfs_warn(mp, "corrupted root inode %llu: not a directory",
 			(unsigned long long)rip->i_ino);
 		xfs_iunlock(rip, XFS_ILOCK_EXCL);
@@ -971,6 +968,8 @@ xfs_mountfs(
 	xfs_da_unmount(mp);
  out_remove_uuid:
 	xfs_uuid_unmount(mp);
+ out_del_stats:
+	xfs_sysfs_del(&mp->m_stats.xs_kobj);
  out_remove_sysfs:
 	xfs_sysfs_del(&mp->m_kobj);
  out:
@@ -1047,6 +1046,7 @@ xfs_unmountfs(
 		xfs_warn(mp, "Unable to update superblock counters. "
 				"Freespace may not be correct on next mount.");
 
+
 	xfs_log_unmount(mp);
 	xfs_da_unmount(mp);
 	xfs_uuid_unmount(mp);
@@ -1056,6 +1056,7 @@ xfs_unmountfs(
 #endif
 	xfs_free_perag(mp);
 
+	xfs_sysfs_del(&mp->m_stats.xs_kobj);
 	xfs_sysfs_del(&mp->m_kobj);
 }
 
@@ -1265,7 +1266,7 @@ xfs_getsb(
 	}
 
 	xfs_buf_hold(bp);
-	ASSERT(XFS_BUF_ISDONE(bp));
+	ASSERT(bp->b_flags & XBF_DONE);
 	return bp;
 }
 
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index 7999e91cd49a..eafe257b357a 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -127,6 +127,7 @@ typedef struct xfs_mount {
 	int64_t			m_low_space[XFS_LOWSP_MAX];
 						/* low free space thresholds */
 	struct xfs_kobj		m_kobj;
+	struct xstats		m_stats;	/* per-fs stats */
 
 	struct workqueue_struct *m_buf_workqueue;
 	struct workqueue_struct	*m_data_workqueue;
@@ -146,6 +147,17 @@ typedef struct xfs_mount {
 	 * to various other kinds of pain inflicted on the pNFS server.
 	 */
 	__uint32_t		m_generation;
+
+#ifdef DEBUG
+	/*
+	 * DEBUG mode instrumentation to test and/or trigger delayed allocation
+	 * block killing in the event of failed writes. When enabled, all
+	 * buffered writes are forced to fail. All delalloc blocks in the range
+	 * of the write (including pre-existing delalloc blocks!) are tossed as
+	 * part of the write failure error handling sequence.
+	 */
+	bool			m_fail_writes;
+#endif
 } xfs_mount_t;
 
 /*
@@ -165,9 +177,8 @@ typedef struct xfs_mount {
 #define XFS_MOUNT_GRPID		(1ULL << 9)	/* group-ID assigned from directory */
 #define XFS_MOUNT_NORECOVERY	(1ULL << 10)	/* no recovery - dirty fs */
 #define XFS_MOUNT_DFLT_IOSIZE	(1ULL << 12)	/* set default i/o size */
-#define XFS_MOUNT_32BITINODES	(1ULL << 14)	/* do not create inodes above
-						 * 32 bits in size */
-#define XFS_MOUNT_SMALL_INUMS	(1ULL << 15)	/* users wants 32bit inodes */
+#define XFS_MOUNT_SMALL_INUMS	(1ULL << 14)	/* user wants 32bit inodes */
+#define XFS_MOUNT_32BITINODES	(1ULL << 15)	/* inode32 allocator active */
 #define XFS_MOUNT_NOUUID	(1ULL << 16)	/* ignore uuid during mount */
 #define XFS_MOUNT_BARRIER	(1ULL << 17)
 #define XFS_MOUNT_IKEEP		(1ULL << 18)	/* keep empty inode clusters*/
@@ -220,12 +231,12 @@ static inline unsigned long
 xfs_preferred_iosize(xfs_mount_t *mp)
 {
 	if (mp->m_flags & XFS_MOUNT_COMPAT_IOSIZE)
-		return PAGE_CACHE_SIZE;
+		return PAGE_SIZE;
 	return (mp->m_swidth ?
 		(mp->m_swidth << mp->m_sb.sb_blocklog) :
 		((mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) ?
 			(1 << (int)MAX(mp->m_readio_log, mp->m_writeio_log)) :
-			PAGE_CACHE_SIZE));
+			PAGE_SIZE));
 }
 
 #define XFS_LAST_UNMOUNT_WAS_CLEAN(mp)	\
@@ -263,6 +274,20 @@ xfs_daddr_to_agbno(struct xfs_mount *mp, xfs_daddr_t d)
 	return (xfs_agblock_t) do_div(ld, mp->m_sb.sb_agblocks);
 }
 
+#ifdef DEBUG
+static inline bool
+xfs_mp_fail_writes(struct xfs_mount *mp)
+{
+	return mp->m_fail_writes;
+}
+#else
+static inline bool
+xfs_mp_fail_writes(struct xfs_mount *mp)
+{
+	return 0;
+}
+#endif
+
 /*
  * Per-ag incore structure, copies of information in agf and agi, to improve the
  * performance of allocation group selection.
@@ -312,6 +337,7 @@ typedef struct xfs_perag {
 	int		pagb_count;	/* pagb slots in use */
 } xfs_perag_t;
 
+extern void	xfs_uuid_table_free(void);
 extern int	xfs_log_sbcount(xfs_mount_t *);
 extern __uint64_t xfs_default_resblks(xfs_mount_t *mp);
 extern int	xfs_mountfs(xfs_mount_t *mp);
@@ -325,7 +351,6 @@ extern int	xfs_mod_fdblocks(struct xfs_mount *mp, int64_t delta,
 				 bool reserved);
 extern int	xfs_mod_frextents(struct xfs_mount *mp, int64_t delta);
 
-extern int	xfs_mount_log_sb(xfs_mount_t *);
 extern struct xfs_buf *xfs_getsb(xfs_mount_t *, int);
 extern int	xfs_readsb(xfs_mount_t *, int);
 extern void	xfs_freesb(xfs_mount_t *);
@@ -336,4 +361,7 @@ extern int	xfs_dev_is_read_only(struct xfs_mount *, char *);
 
 extern void	xfs_set_low_space_thresholds(struct xfs_mount *);
 
+int	xfs_zero_extent(struct xfs_inode *ip, xfs_fsblock_t start_fsb,
+			xfs_off_t count_fsb);
+
 #endif	/* __XFS_MOUNT_H__ */
diff --git a/fs/xfs/xfs_ondisk.h b/fs/xfs/xfs_ondisk.h
new file mode 100644
index 000000000000..184c44effdd5
--- /dev/null
+++ b/fs/xfs/xfs_ondisk.h
@@ -0,0 +1,117 @@
+/*
+ * Copyright (c) 2016 Oracle.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * 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_ONDISK_H
+#define __XFS_ONDISK_H
+
+#define XFS_CHECK_STRUCT_SIZE(structname, size) \
+	BUILD_BUG_ON_MSG(sizeof(structname) != (size), "XFS: sizeof(" \
+		#structname ") is wrong, expected " #size)
+
+static inline void __init
+xfs_check_ondisk_structs(void)
+{
+	/* ag/file structures */
+	XFS_CHECK_STRUCT_SIZE(struct xfs_acl,			4);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_acl_entry,		12);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_agf,			224);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_agfl,			36);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_agi,			336);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_bmbt_key,		8);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_bmbt_rec,		16);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_bmdr_block,		4);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_btree_block,		72);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_dinode,		176);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_disk_dquot,		104);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_dqblk,			136);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_dsb,			264);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_dsymlink_hdr,		56);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_inobt_key,		4);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_inobt_rec,		16);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_timestamp,		8);
+	XFS_CHECK_STRUCT_SIZE(xfs_alloc_key_t,			8);
+	XFS_CHECK_STRUCT_SIZE(xfs_alloc_ptr_t,			4);
+	XFS_CHECK_STRUCT_SIZE(xfs_alloc_rec_t,			8);
+	XFS_CHECK_STRUCT_SIZE(xfs_inobt_ptr_t,			4);
+
+	/* dir/attr trees */
+	XFS_CHECK_STRUCT_SIZE(struct xfs_attr3_leaf_hdr,	80);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_attr3_leafblock,	88);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_attr3_rmt_hdr,		56);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_da3_blkinfo,		56);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_da3_intnode,		64);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_da3_node_hdr,		64);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_dir3_blk_hdr,		48);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_dir3_data_hdr,		64);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_dir3_free,		64);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_dir3_free_hdr,		64);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_dir3_leaf,		64);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_dir3_leaf_hdr,		64);
+	XFS_CHECK_STRUCT_SIZE(xfs_attr_leaf_entry_t,		8);
+	XFS_CHECK_STRUCT_SIZE(xfs_attr_leaf_hdr_t,		32);
+	XFS_CHECK_STRUCT_SIZE(xfs_attr_leaf_map_t,		4);
+	XFS_CHECK_STRUCT_SIZE(xfs_attr_leaf_name_local_t,	4);
+
+	/*
+	 * m68k has problems with xfs_attr_leaf_name_remote_t, but we pad it to
+	 * 4 bytes anyway so it's not obviously a problem.  Hence for the moment
+	 * we don't check this structure. This can be re-instated when the attr
+	 * definitions are updated to use c99 VLA definitions.
+	 *
+	XFS_CHECK_STRUCT_SIZE(xfs_attr_leaf_name_remote_t,	12);
+	 */
+
+	XFS_CHECK_STRUCT_SIZE(xfs_attr_leafblock_t,		40);
+	XFS_CHECK_STRUCT_SIZE(xfs_attr_shortform_t,		8);
+	XFS_CHECK_STRUCT_SIZE(xfs_da_blkinfo_t,			12);
+	XFS_CHECK_STRUCT_SIZE(xfs_da_intnode_t,			16);
+	XFS_CHECK_STRUCT_SIZE(xfs_da_node_entry_t,		8);
+	XFS_CHECK_STRUCT_SIZE(xfs_da_node_hdr_t,		16);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_data_free_t,		4);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_data_hdr_t,		16);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_data_unused_t,		6);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_free_hdr_t,		16);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_free_t,			16);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_ino4_t,			4);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_ino8_t,			8);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_inou_t,			8);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_leaf_entry_t,		8);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_leaf_hdr_t,		16);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_leaf_t,			16);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_leaf_tail_t,		4);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_sf_entry_t,		3);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_sf_hdr_t,		10);
+	XFS_CHECK_STRUCT_SIZE(xfs_dir2_sf_off_t,		2);
+
+	/* log structures */
+	XFS_CHECK_STRUCT_SIZE(struct xfs_dq_logformat,		24);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_efd_log_format_32,	28);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_efd_log_format_64,	32);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_efi_log_format_32,	28);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_efi_log_format_64,	32);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_extent_32,		12);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_extent_64,		16);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_log_dinode,		176);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_icreate_log,		28);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_ictimestamp,		8);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_inode_log_format_32,	52);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_inode_log_format_64,	56);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_qoff_logformat,	20);
+	XFS_CHECK_STRUCT_SIZE(struct xfs_trans_header,		16);
+}
+
+#endif /* __XFS_ONDISK_H */
diff --git a/fs/xfs/xfs_pnfs.c b/fs/xfs/xfs_pnfs.c
index ab4a6066f7ca..51ddaf2c2b8c 100644
--- a/fs/xfs/xfs_pnfs.c
+++ b/fs/xfs/xfs_pnfs.c
@@ -42,11 +42,11 @@ xfs_break_layouts(
 	while ((error = break_layout(inode, false) == -EWOULDBLOCK)) {
 		xfs_iunlock(ip, *iolock);
 		if (with_imutex && (*iolock & XFS_IOLOCK_EXCL))
-			mutex_unlock(&inode->i_mutex);
+			inode_unlock(inode);
 		error = break_layout(inode, true);
 		*iolock = XFS_IOLOCK_EXCL;
 		if (with_imutex)
-			mutex_lock(&inode->i_mutex);
+			inode_lock(inode);
 		xfs_ilock(ip, *iolock);
 	}
 
@@ -181,6 +181,11 @@ xfs_fs_map_blocks(
 		ASSERT(imap.br_startblock != DELAYSTARTBLOCK);
 
 		if (!nimaps || imap.br_startblock == HOLESTARTBLOCK) {
+			/*
+			 * xfs_iomap_write_direct() expects to take ownership of
+			 * the shared ilock.
+			 */
+			xfs_ilock(ip, XFS_ILOCK_SHARED);
 			error = xfs_iomap_write_direct(ip, offset, length,
 						       &imap, nimaps);
 			if (error)
@@ -288,8 +293,8 @@ xfs_fs_commit_blocks(
 		 * Make sure reads through the pagecache see the new data.
 		 */
 		error = invalidate_inode_pages2_range(inode->i_mapping,
-					start >> PAGE_CACHE_SHIFT,
-					(end - 1) >> PAGE_CACHE_SHIFT);
+					start >> PAGE_SHIFT,
+					(end - 1) >> PAGE_SHIFT);
 		WARN_ON_ONCE(error);
 
 		error = xfs_iomap_write_unwritten(ip, start, length);
diff --git a/fs/xfs/xfs_pnfs.h b/fs/xfs/xfs_pnfs.h
index 8147ac108820..93f74853961b 100644
--- a/fs/xfs/xfs_pnfs.h
+++ b/fs/xfs/xfs_pnfs.h
@@ -1,7 +1,7 @@
 #ifndef _XFS_PNFS_H
 #define _XFS_PNFS_H 1
 
-#ifdef CONFIG_NFSD_PNFS
+#if defined(CONFIG_NFSD_BLOCKLAYOUT) || defined(CONFIG_NFSD_SCSILAYOUT)
 int xfs_fs_get_uuid(struct super_block *sb, u8 *buf, u32 *len, u64 *offset);
 int xfs_fs_map_blocks(struct inode *inode, loff_t offset, u64 length,
 		struct iomap *iomap, bool write, u32 *device_generation);
diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c
index eac9549efd52..be125e1758c1 100644
--- a/fs/xfs/xfs_qm.c
+++ b/fs/xfs/xfs_qm.c
@@ -184,7 +184,7 @@ xfs_qm_dqpurge(
 	 */
 	ASSERT(!list_empty(&dqp->q_lru));
 	list_lru_del(&qi->qi_lru, &dqp->q_lru);
-	XFS_STATS_DEC(xs_qm_dquot_unused);
+	XFS_STATS_DEC(mp, xs_qm_dquot_unused);
 
 	xfs_qm_dqdestroy(dqp);
 	return 0;
@@ -448,11 +448,11 @@ xfs_qm_dquot_isolate(
 	 */
 	if (dqp->q_nrefs) {
 		xfs_dqunlock(dqp);
-		XFS_STATS_INC(xs_qm_dqwants);
+		XFS_STATS_INC(dqp->q_mount, xs_qm_dqwants);
 
 		trace_xfs_dqreclaim_want(dqp);
 		list_lru_isolate(lru, &dqp->q_lru);
-		XFS_STATS_DEC(xs_qm_dquot_unused);
+		XFS_STATS_DEC(dqp->q_mount, xs_qm_dquot_unused);
 		return LRU_REMOVED;
 	}
 
@@ -496,19 +496,19 @@ xfs_qm_dquot_isolate(
 
 	ASSERT(dqp->q_nrefs == 0);
 	list_lru_isolate_move(lru, &dqp->q_lru, &isol->dispose);
-	XFS_STATS_DEC(xs_qm_dquot_unused);
+	XFS_STATS_DEC(dqp->q_mount, xs_qm_dquot_unused);
 	trace_xfs_dqreclaim_done(dqp);
-	XFS_STATS_INC(xs_qm_dqreclaims);
+	XFS_STATS_INC(dqp->q_mount, xs_qm_dqreclaims);
 	return LRU_REMOVED;
 
 out_miss_busy:
 	trace_xfs_dqreclaim_busy(dqp);
-	XFS_STATS_INC(xs_qm_dqreclaim_misses);
+	XFS_STATS_INC(dqp->q_mount, xs_qm_dqreclaim_misses);
 	return LRU_SKIP;
 
 out_unlock_dirty:
 	trace_xfs_dqreclaim_busy(dqp);
-	XFS_STATS_INC(xs_qm_dqreclaim_misses);
+	XFS_STATS_INC(dqp->q_mount, xs_qm_dqreclaim_misses);
 	xfs_dqunlock(dqp);
 	spin_lock(lru_lock);
 	return LRU_RETRY;
@@ -525,7 +525,7 @@ xfs_qm_shrink_scan(
 	unsigned long		freed;
 	int			error;
 
-	if ((sc->gfp_mask & (__GFP_FS|__GFP_WAIT)) != (__GFP_FS|__GFP_WAIT))
+	if ((sc->gfp_mask & (__GFP_FS|__GFP_DIRECT_RECLAIM)) != (__GFP_FS|__GFP_DIRECT_RECLAIM))
 		return 0;
 
 	INIT_LIST_HEAD(&isol.buffers);
@@ -560,6 +560,37 @@ xfs_qm_shrink_count(
 	return list_lru_shrink_count(&qi->qi_lru, sc);
 }
 
+STATIC void
+xfs_qm_set_defquota(
+	xfs_mount_t	*mp,
+	uint		type,
+	xfs_quotainfo_t	*qinf)
+{
+	xfs_dquot_t		*dqp;
+	struct xfs_def_quota    *defq;
+	int			error;
+
+	error = xfs_qm_dqread(mp, 0, type, XFS_QMOPT_DOWARN, &dqp);
+
+	if (!error) {
+		xfs_disk_dquot_t        *ddqp = &dqp->q_core;
+
+		defq = xfs_get_defquota(dqp, qinf);
+
+		/*
+		 * Timers and warnings have been already set, let's just set the
+		 * default limits for this quota type
+		 */
+		defq->bhardlimit = be64_to_cpu(ddqp->d_blk_hardlimit);
+		defq->bsoftlimit = be64_to_cpu(ddqp->d_blk_softlimit);
+		defq->ihardlimit = be64_to_cpu(ddqp->d_ino_hardlimit);
+		defq->isoftlimit = be64_to_cpu(ddqp->d_ino_softlimit);
+		defq->rtbhardlimit = be64_to_cpu(ddqp->d_rtb_hardlimit);
+		defq->rtbsoftlimit = be64_to_cpu(ddqp->d_rtb_softlimit);
+		xfs_qm_dqdestroy(dqp);
+	}
+}
+
 /*
  * This initializes all the quota information that's kept in the
  * mount structure
@@ -606,19 +637,19 @@ xfs_qm_init_quotainfo(
 	 * We try to get the limits from the superuser's limits fields.
 	 * This is quite hacky, but it is standard quota practice.
 	 *
-	 * We look at the USR dquot with id == 0 first, but if user quotas
-	 * are not enabled we goto the GRP dquot with id == 0.
-	 * We don't really care to keep separate default limits for user
-	 * and group quotas, at least not at this point.
-	 *
 	 * Since we may not have done a quotacheck by this point, just read
 	 * the dquot without attaching it to any hashtables or lists.
+	 *
+	 * Timers and warnings are globally set by the first timer found in
+	 * user/group/proj quota types, otherwise a default value is used.
+	 * This should be split into different fields per quota type.
 	 */
 	error = xfs_qm_dqread(mp, 0,
 			XFS_IS_UQUOTA_RUNNING(mp) ? XFS_DQ_USER :
 			 (XFS_IS_GQUOTA_RUNNING(mp) ? XFS_DQ_GROUP :
 			  XFS_DQ_PROJ),
 			XFS_QMOPT_DOWARN, &dqp);
+
 	if (!error) {
 		xfs_disk_dquot_t	*ddqp = &dqp->q_core;
 
@@ -639,13 +670,6 @@ xfs_qm_init_quotainfo(
 			be16_to_cpu(ddqp->d_iwarns) : XFS_QM_IWARNLIMIT;
 		qinf->qi_rtbwarnlimit = ddqp->d_rtbwarns ?
 			be16_to_cpu(ddqp->d_rtbwarns) : XFS_QM_RTBWARNLIMIT;
-		qinf->qi_bhardlimit = be64_to_cpu(ddqp->d_blk_hardlimit);
-		qinf->qi_bsoftlimit = be64_to_cpu(ddqp->d_blk_softlimit);
-		qinf->qi_ihardlimit = be64_to_cpu(ddqp->d_ino_hardlimit);
-		qinf->qi_isoftlimit = be64_to_cpu(ddqp->d_ino_softlimit);
-		qinf->qi_rtbhardlimit = be64_to_cpu(ddqp->d_rtb_hardlimit);
-		qinf->qi_rtbsoftlimit = be64_to_cpu(ddqp->d_rtb_softlimit);
-
 		xfs_qm_dqdestroy(dqp);
 	} else {
 		qinf->qi_btimelimit = XFS_QM_BTIMELIMIT;
@@ -656,6 +680,13 @@ xfs_qm_init_quotainfo(
 		qinf->qi_rtbwarnlimit = XFS_QM_RTBWARNLIMIT;
 	}
 
+	if (XFS_IS_UQUOTA_RUNNING(mp))
+		xfs_qm_set_defquota(mp, XFS_DQ_USER, qinf);
+	if (XFS_IS_GQUOTA_RUNNING(mp))
+		xfs_qm_set_defquota(mp, XFS_DQ_GROUP, qinf);
+	if (XFS_IS_PQUOTA_RUNNING(mp))
+		xfs_qm_set_defquota(mp, XFS_DQ_PROJ, qinf);
+
 	qinf->qi_shrinker.count_objects = xfs_qm_shrink_count;
 	qinf->qi_shrinker.scan_objects = xfs_qm_shrink_scan;
 	qinf->qi_shrinker.seeks = DEFAULT_SEEKS;
diff --git a/fs/xfs/xfs_qm.h b/fs/xfs/xfs_qm.h
index 996a04064894..2975a822e9f0 100644
--- a/fs/xfs/xfs_qm.h
+++ b/fs/xfs/xfs_qm.h
@@ -53,6 +53,15 @@ extern struct kmem_zone	*xfs_qm_dqtrxzone;
  */
 #define XFS_DQUOT_CLUSTER_SIZE_FSB	(xfs_filblks_t)1
 
+struct xfs_def_quota {
+	xfs_qcnt_t       bhardlimit;     /* default data blk hard limit */
+	xfs_qcnt_t       bsoftlimit;	 /* default data blk soft limit */
+	xfs_qcnt_t       ihardlimit;	 /* default inode count hard limit */
+	xfs_qcnt_t       isoftlimit;	 /* default inode count soft limit */
+	xfs_qcnt_t	 rtbhardlimit;   /* default realtime blk hard limit */
+	xfs_qcnt_t	 rtbsoftlimit;   /* default realtime blk soft limit */
+};
+
 /*
  * Various quota information for individual filesystems.
  * The mount structure keeps a pointer to this.
@@ -76,12 +85,9 @@ typedef struct xfs_quotainfo {
 	struct mutex	 qi_quotaofflock;/* to serialize quotaoff */
 	xfs_filblks_t	 qi_dqchunklen;	 /* # BBs in a chunk of dqs */
 	uint		 qi_dqperchunk;	 /* # ondisk dqs in above chunk */
-	xfs_qcnt_t	 qi_bhardlimit;	 /* default data blk hard limit */
-	xfs_qcnt_t	 qi_bsoftlimit;	 /* default data blk soft limit */
-	xfs_qcnt_t	 qi_ihardlimit;	 /* default inode count hard limit */
-	xfs_qcnt_t	 qi_isoftlimit;	 /* default inode count soft limit */
-	xfs_qcnt_t	 qi_rtbhardlimit;/* default realtime blk hard limit */
-	xfs_qcnt_t	 qi_rtbsoftlimit;/* default realtime blk soft limit */
+	struct xfs_def_quota	qi_usr_default;
+	struct xfs_def_quota	qi_grp_default;
+	struct xfs_def_quota	qi_prj_default;
 	struct shrinker  qi_shrinker;
 } xfs_quotainfo_t;
 
@@ -104,15 +110,15 @@ xfs_dquot_tree(
 }
 
 static inline struct xfs_inode *
-xfs_dq_to_quota_inode(struct xfs_dquot *dqp)
+xfs_quota_inode(xfs_mount_t *mp, uint dq_flags)
 {
-	switch (dqp->dq_flags & XFS_DQ_ALLTYPES) {
+	switch (dq_flags & XFS_DQ_ALLTYPES) {
 	case XFS_DQ_USER:
-		return dqp->q_mount->m_quotainfo->qi_uquotaip;
+		return mp->m_quotainfo->qi_uquotaip;
 	case XFS_DQ_GROUP:
-		return dqp->q_mount->m_quotainfo->qi_gquotaip;
+		return mp->m_quotainfo->qi_gquotaip;
 	case XFS_DQ_PROJ:
-		return dqp->q_mount->m_quotainfo->qi_pquotaip;
+		return mp->m_quotainfo->qi_pquotaip;
 	default:
 		ASSERT(0);
 	}
@@ -164,11 +170,27 @@ extern void		xfs_qm_dqrele_all_inodes(struct xfs_mount *, uint);
 
 /* quota ops */
 extern int		xfs_qm_scall_trunc_qfiles(struct xfs_mount *, uint);
-extern int		xfs_qm_scall_getquota(struct xfs_mount *, xfs_dqid_t,
-					uint, struct qc_dqblk *);
+extern int		xfs_qm_scall_getquota(struct xfs_mount *, xfs_dqid_t *,
+					uint, struct qc_dqblk *, uint);
 extern int		xfs_qm_scall_setqlim(struct xfs_mount *, xfs_dqid_t, uint,
 					struct qc_dqblk *);
 extern int		xfs_qm_scall_quotaon(struct xfs_mount *, uint);
 extern int		xfs_qm_scall_quotaoff(struct xfs_mount *, uint);
 
+static inline struct xfs_def_quota *
+xfs_get_defquota(struct xfs_dquot *dqp, struct xfs_quotainfo *qi)
+{
+	struct xfs_def_quota *defq;
+
+	if (XFS_QM_ISUDQ(dqp))
+		defq = &qi->qi_usr_default;
+	else if (XFS_QM_ISGDQ(dqp))
+		defq = &qi->qi_grp_default;
+	else {
+		ASSERT(XFS_QM_ISPDQ(dqp));
+		defq = &qi->qi_prj_default;
+	}
+	return defq;
+}
+
 #endif /* __XFS_QM_H__ */
diff --git a/fs/xfs/xfs_qm_syscalls.c b/fs/xfs/xfs_qm_syscalls.c
index 3640c6e896af..f4d0e0a8f517 100644
--- a/fs/xfs/xfs_qm_syscalls.c
+++ b/fs/xfs/xfs_qm_syscalls.c
@@ -404,6 +404,7 @@ xfs_qm_scall_setqlim(
 	struct xfs_disk_dquot	*ddq;
 	struct xfs_dquot	*dqp;
 	struct xfs_trans	*tp;
+	struct xfs_def_quota	*defq;
 	int			error;
 	xfs_qcnt_t		hard, soft;
 
@@ -431,6 +432,8 @@ xfs_qm_scall_setqlim(
 		ASSERT(error != -ENOENT);
 		goto out_unlock;
 	}
+
+	defq = xfs_get_defquota(dqp, q);
 	xfs_dqunlock(dqp);
 
 	tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SETQLIM);
@@ -458,8 +461,8 @@ xfs_qm_scall_setqlim(
 		ddq->d_blk_softlimit = cpu_to_be64(soft);
 		xfs_dquot_set_prealloc_limits(dqp);
 		if (id == 0) {
-			q->qi_bhardlimit = hard;
-			q->qi_bsoftlimit = soft;
+			defq->bhardlimit = hard;
+			defq->bsoftlimit = soft;
 		}
 	} else {
 		xfs_debug(mp, "blkhard %Ld < blksoft %Ld", hard, soft);
@@ -474,8 +477,8 @@ xfs_qm_scall_setqlim(
 		ddq->d_rtb_hardlimit = cpu_to_be64(hard);
 		ddq->d_rtb_softlimit = cpu_to_be64(soft);
 		if (id == 0) {
-			q->qi_rtbhardlimit = hard;
-			q->qi_rtbsoftlimit = soft;
+			defq->rtbhardlimit = hard;
+			defq->rtbsoftlimit = soft;
 		}
 	} else {
 		xfs_debug(mp, "rtbhard %Ld < rtbsoft %Ld", hard, soft);
@@ -491,8 +494,8 @@ xfs_qm_scall_setqlim(
 		ddq->d_ino_hardlimit = cpu_to_be64(hard);
 		ddq->d_ino_softlimit = cpu_to_be64(soft);
 		if (id == 0) {
-			q->qi_ihardlimit = hard;
-			q->qi_isoftlimit = soft;
+			defq->ihardlimit = hard;
+			defq->isoftlimit = soft;
 		}
 	} else {
 		xfs_debug(mp, "ihard %Ld < isoft %Ld", hard, soft);
@@ -635,9 +638,10 @@ out:
 int
 xfs_qm_scall_getquota(
 	struct xfs_mount	*mp,
-	xfs_dqid_t		id,
+	xfs_dqid_t		*id,
 	uint			type,
-	struct qc_dqblk		*dst)
+	struct qc_dqblk		*dst,
+	uint			dqget_flags)
 {
 	struct xfs_dquot	*dqp;
 	int			error;
@@ -647,7 +651,7 @@ xfs_qm_scall_getquota(
 	 * we aren't passing the XFS_QMOPT_DOALLOC flag. If it doesn't
 	 * exist, we'll get ENOENT back.
 	 */
-	error = xfs_qm_dqget(mp, NULL, id, type, 0, &dqp);
+	error = xfs_qm_dqget(mp, NULL, *id, type, dqget_flags, &dqp);
 	if (error)
 		return error;
 
@@ -660,6 +664,9 @@ xfs_qm_scall_getquota(
 		goto out_put;
 	}
 
+	/* Fill in the ID we actually read from disk */
+	*id = be32_to_cpu(dqp->q_core.d_id);
+
 	memset(dst, 0, sizeof(*dst));
 	dst->d_spc_hardlimit =
 		XFS_FSB_TO_B(mp, be64_to_cpu(dqp->q_core.d_blk_hardlimit));
@@ -701,7 +708,7 @@ xfs_qm_scall_getquota(
 	if (((XFS_IS_UQUOTA_ENFORCED(mp) && type == XFS_DQ_USER) ||
 	     (XFS_IS_GQUOTA_ENFORCED(mp) && type == XFS_DQ_GROUP) ||
 	     (XFS_IS_PQUOTA_ENFORCED(mp) && type == XFS_DQ_PROJ)) &&
-	    id != 0) {
+	    *id != 0) {
 		if ((dst->d_space > dst->d_spc_softlimit) &&
 		    (dst->d_spc_softlimit > 0)) {
 			ASSERT(dst->d_spc_timer != 0);
diff --git a/fs/xfs/xfs_quotaops.c b/fs/xfs/xfs_quotaops.c
index 7795e0d01382..f82d79a8c694 100644
--- a/fs/xfs/xfs_quotaops.c
+++ b/fs/xfs/xfs_quotaops.c
@@ -231,14 +231,45 @@ xfs_fs_get_dqblk(
 	struct qc_dqblk		*qdq)
 {
 	struct xfs_mount	*mp = XFS_M(sb);
+	xfs_dqid_t		id;
 
 	if (!XFS_IS_QUOTA_RUNNING(mp))
 		return -ENOSYS;
 	if (!XFS_IS_QUOTA_ON(mp))
 		return -ESRCH;
 
-	return xfs_qm_scall_getquota(mp, from_kqid(&init_user_ns, qid),
-				      xfs_quota_type(qid.type), qdq);
+	id = from_kqid(&init_user_ns, qid);
+	return xfs_qm_scall_getquota(mp, &id,
+				      xfs_quota_type(qid.type), qdq, 0);
+}
+
+/* Return quota info for active quota >= this qid */
+STATIC int
+xfs_fs_get_nextdqblk(
+	struct super_block	*sb,
+	struct kqid		*qid,
+	struct qc_dqblk		*qdq)
+{
+	int			ret;
+	struct xfs_mount	*mp = XFS_M(sb);
+	xfs_dqid_t		id;
+
+	if (!XFS_IS_QUOTA_RUNNING(mp))
+		return -ENOSYS;
+	if (!XFS_IS_QUOTA_ON(mp))
+		return -ESRCH;
+
+	id = from_kqid(&init_user_ns, *qid);
+	ret = xfs_qm_scall_getquota(mp, &id,
+				    xfs_quota_type(qid->type), qdq,
+				    XFS_QMOPT_DQNEXT);
+	if (ret)
+		return ret;
+
+	/* ID may be different, so convert back what we got */
+	*qid = make_kqid(current_user_ns(), qid->type, id);
+	return 0;
+	
 }
 
 STATIC int
@@ -267,5 +298,6 @@ const struct quotactl_ops xfs_quotactl_operations = {
 	.quota_disable		= xfs_quota_disable,
 	.rm_xquota		= xfs_fs_rm_xquota,
 	.get_dqblk		= xfs_fs_get_dqblk,
+	.get_nextdqblk		= xfs_fs_get_nextdqblk,
 	.set_dqblk		= xfs_fs_set_dqblk,
 };
diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c
index ab1bac6a3a1c..abf44435d04a 100644
--- a/fs/xfs/xfs_rtalloc.c
+++ b/fs/xfs/xfs_rtalloc.c
@@ -766,7 +766,6 @@ xfs_growfs_rt_alloc(
 {
 	xfs_fileoff_t		bno;		/* block number in file */
 	struct xfs_buf		*bp;	/* temporary buffer for zeroing */
-	int			committed;	/* transaction committed flag */
 	xfs_daddr_t		d;		/* disk block address */
 	int			error;		/* error return value */
 	xfs_fsblock_t		firstblock;/* first block allocated in xaction */
@@ -811,7 +810,7 @@ xfs_growfs_rt_alloc(
 		/*
 		 * Free any blocks freed up in the transaction, then commit.
 		 */
-		error = xfs_bmap_finish(&tp, &flist, &committed);
+		error = xfs_bmap_finish(&tp, &flist, NULL);
 		if (error)
 			goto out_bmap_cancel;
 		error = xfs_trans_commit(tp);
@@ -1273,7 +1272,7 @@ xfs_rtpick_extent(
 
 	ASSERT(xfs_isilocked(mp->m_rbmip, XFS_ILOCK_EXCL));
 
-	seqp = (__uint64_t *)&mp->m_rbmip->i_d.di_atime;
+	seqp = (__uint64_t *)&VFS_I(mp->m_rbmip)->i_atime;
 	if (!(mp->m_rbmip->i_d.di_flags & XFS_DIFLAG_NEWRTBM)) {
 		mp->m_rbmip->i_d.di_flags |= XFS_DIFLAG_NEWRTBM;
 		*seqp = 0;
diff --git a/fs/xfs/xfs_stats.c b/fs/xfs/xfs_stats.c
index f2240383d4bb..8686df6c7609 100644
--- a/fs/xfs/xfs_stats.c
+++ b/fs/xfs/xfs_stats.c
@@ -18,20 +18,21 @@
 #include "xfs.h"
 #include <linux/proc_fs.h>
 
-DEFINE_PER_CPU(struct xfsstats, xfsstats);
+struct xstats xfsstats;
 
-static int counter_val(int idx)
+static int counter_val(struct xfsstats __percpu *stats, int idx)
 {
 	int val = 0, cpu;
 
 	for_each_possible_cpu(cpu)
-		val += *(((__u32 *)&per_cpu(xfsstats, cpu) + idx));
+		val += *(((__u32 *)per_cpu_ptr(stats, cpu) + idx));
 	return val;
 }
 
-static int xfs_stat_proc_show(struct seq_file *m, void *v)
+int xfs_stats_format(struct xfsstats __percpu *stats, char *buf)
 {
 	int		i, j;
+	int		len = 0;
 	__uint64_t	xs_xstrat_bytes = 0;
 	__uint64_t	xs_write_bytes = 0;
 	__uint64_t	xs_read_bytes = 0;
@@ -65,54 +66,59 @@ static int xfs_stat_proc_show(struct seq_file *m, void *v)
 	};
 
 	/* Loop over all stats groups */
+
 	for (i = j = 0; i < ARRAY_SIZE(xstats); i++) {
-		seq_printf(m, "%s", xstats[i].desc);
+		len += snprintf(buf + len, PATH_MAX - len, "%s",
+				xstats[i].desc);
 		/* inner loop does each group */
 		for (; j < xstats[i].endpoint; j++)
-			seq_printf(m, " %u", counter_val(j));
-		seq_putc(m, '\n');
+			len += snprintf(buf + len, PATH_MAX - len, " %u",
+					counter_val(stats, j));
+		len += snprintf(buf + len, PATH_MAX - len, "\n");
 	}
 	/* extra precision counters */
 	for_each_possible_cpu(i) {
-		xs_xstrat_bytes += per_cpu(xfsstats, i).xs_xstrat_bytes;
-		xs_write_bytes += per_cpu(xfsstats, i).xs_write_bytes;
-		xs_read_bytes += per_cpu(xfsstats, i).xs_read_bytes;
+		xs_xstrat_bytes += per_cpu_ptr(stats, i)->xs_xstrat_bytes;
+		xs_write_bytes += per_cpu_ptr(stats, i)->xs_write_bytes;
+		xs_read_bytes += per_cpu_ptr(stats, i)->xs_read_bytes;
 	}
 
-	seq_printf(m, "xpc %Lu %Lu %Lu\n",
+	len += snprintf(buf + len, PATH_MAX-len, "xpc %Lu %Lu %Lu\n",
 			xs_xstrat_bytes, xs_write_bytes, xs_read_bytes);
-	seq_printf(m, "debug %u\n",
+	len += snprintf(buf + len, PATH_MAX-len, "debug %u\n",
 #if defined(DEBUG)
 		1);
 #else
 		0);
 #endif
-	return 0;
+
+	return len;
 }
 
-static int xfs_stat_proc_open(struct inode *inode, struct file *file)
+void xfs_stats_clearall(struct xfsstats __percpu *stats)
 {
-	return single_open(file, xfs_stat_proc_show, NULL);
+	int		c;
+	__uint32_t	vn_active;
+
+	xfs_notice(NULL, "Clearing xfsstats");
+	for_each_possible_cpu(c) {
+		preempt_disable();
+		/* save vn_active, it's a universal truth! */
+		vn_active = per_cpu_ptr(stats, c)->vn_active;
+		memset(per_cpu_ptr(stats, c), 0, sizeof(*stats));
+		per_cpu_ptr(stats, c)->vn_active = vn_active;
+		preempt_enable();
+	}
 }
 
-static const struct file_operations xfs_stat_proc_fops = {
-	.owner		= THIS_MODULE,
-	.open		= xfs_stat_proc_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-};
-
 /* legacy quota interfaces */
 #ifdef CONFIG_XFS_QUOTA
 static int xqm_proc_show(struct seq_file *m, void *v)
 {
 	/* maximum; incore; ratio free to inuse; freelist */
 	seq_printf(m, "%d\t%d\t%d\t%u\n",
-			0,
-			counter_val(XFSSTAT_END_XQMSTAT),
-			0,
-			counter_val(XFSSTAT_END_XQMSTAT + 1));
+		   0, counter_val(xfsstats.xs_stats, XFSSTAT_END_XQMSTAT),
+		   0, counter_val(xfsstats.xs_stats, XFSSTAT_END_XQMSTAT + 1));
 	return 0;
 }
 
@@ -136,7 +142,7 @@ static int xqmstat_proc_show(struct seq_file *m, void *v)
 
 	seq_printf(m, "qm");
 	for (j = XFSSTAT_END_IBT_V2; j < XFSSTAT_END_XQMSTAT; j++)
-		seq_printf(m, " %u", counter_val(j));
+		seq_printf(m, " %u", counter_val(xfsstats.xs_stats, j));
 	seq_putc(m, '\n');
 	return 0;
 }
@@ -155,44 +161,35 @@ static const struct file_operations xqmstat_proc_fops = {
 };
 #endif /* CONFIG_XFS_QUOTA */
 
+#ifdef CONFIG_PROC_FS
 int
 xfs_init_procfs(void)
 {
 	if (!proc_mkdir("fs/xfs", NULL))
+		return -ENOMEM;
+
+	if (!proc_symlink("fs/xfs/stat", NULL,
+			  "/sys/fs/xfs/stats/stats"))
 		goto out;
 
-	if (!proc_create("fs/xfs/stat", 0, NULL,
-			 &xfs_stat_proc_fops))
-		goto out_remove_xfs_dir;
 #ifdef CONFIG_XFS_QUOTA
 	if (!proc_create("fs/xfs/xqmstat", 0, NULL,
 			 &xqmstat_proc_fops))
-		goto out_remove_stat_file;
+		goto out;
 	if (!proc_create("fs/xfs/xqm", 0, NULL,
 			 &xqm_proc_fops))
-		goto out_remove_xqmstat_file;
+		goto out;
 #endif
 	return 0;
 
-#ifdef CONFIG_XFS_QUOTA
- out_remove_xqmstat_file:
-	remove_proc_entry("fs/xfs/xqmstat", NULL);
- out_remove_stat_file:
-	remove_proc_entry("fs/xfs/stat", NULL);
-#endif
- out_remove_xfs_dir:
-	remove_proc_entry("fs/xfs", NULL);
- out:
+out:
+	remove_proc_subtree("fs/xfs", NULL);
 	return -ENOMEM;
 }
 
 void
 xfs_cleanup_procfs(void)
 {
-#ifdef CONFIG_XFS_QUOTA
-	remove_proc_entry("fs/xfs/xqm", NULL);
-	remove_proc_entry("fs/xfs/xqmstat", NULL);
-#endif
-	remove_proc_entry("fs/xfs/stat", NULL);
-	remove_proc_entry("fs/xfs", NULL);
+	remove_proc_subtree("fs/xfs", NULL);
 }
+#endif /* CONFIG_PROC_FS */
diff --git a/fs/xfs/xfs_stats.h b/fs/xfs/xfs_stats.h
index c8f238b8299a..483b0eff1988 100644
--- a/fs/xfs/xfs_stats.h
+++ b/fs/xfs/xfs_stats.h
@@ -19,8 +19,6 @@
 #define __XFS_STATS_H__
 
 
-#if defined(CONFIG_PROC_FS) && !defined(XFS_STATS_OFF)
-
 #include <linux/percpu.h>
 
 /*
@@ -215,15 +213,29 @@ struct xfsstats {
 	__uint64_t		xs_read_bytes;
 };
 
-DECLARE_PER_CPU(struct xfsstats, xfsstats);
+int xfs_stats_format(struct xfsstats __percpu *stats, char *buf);
+void xfs_stats_clearall(struct xfsstats __percpu *stats);
+extern struct xstats xfsstats;
 
-/*
- * We don't disable preempt, not too worried about poking the
- * wrong CPU's stat for now (also aggregated before reporting).
- */
-#define XFS_STATS_INC(v)	(per_cpu(xfsstats, current_cpu()).v++)
-#define XFS_STATS_DEC(v)	(per_cpu(xfsstats, current_cpu()).v--)
-#define XFS_STATS_ADD(v, inc)	(per_cpu(xfsstats, current_cpu()).v += (inc))
+#define XFS_STATS_INC(mp, v)					\
+do {								\
+	per_cpu_ptr(xfsstats.xs_stats, current_cpu())->v++;	\
+	per_cpu_ptr(mp->m_stats.xs_stats, current_cpu())->v++;	\
+} while (0)
+
+#define XFS_STATS_DEC(mp, v)					\
+do {								\
+	per_cpu_ptr(xfsstats.xs_stats, current_cpu())->v--;	\
+	per_cpu_ptr(mp->m_stats.xs_stats, current_cpu())->v--;	\
+} while (0)
+
+#define XFS_STATS_ADD(mp, v, inc)					\
+do {									\
+	per_cpu_ptr(xfsstats.xs_stats, current_cpu())->v += (inc);	\
+	per_cpu_ptr(mp->m_stats.xs_stats, current_cpu())->v += (inc);	\
+} while (0)
+
+#if defined(CONFIG_PROC_FS)
 
 extern int xfs_init_procfs(void);
 extern void xfs_cleanup_procfs(void);
@@ -231,10 +243,6 @@ extern void xfs_cleanup_procfs(void);
 
 #else	/* !CONFIG_PROC_FS */
 
-# define XFS_STATS_INC(count)
-# define XFS_STATS_DEC(count)
-# define XFS_STATS_ADD(count, inc)
-
 static inline int xfs_init_procfs(void)
 {
 	return 0;
diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
index 904f637cfa5f..187e14b696c2 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -45,6 +45,7 @@
 #include "xfs_filestream.h"
 #include "xfs_quota.h"
 #include "xfs_sysfs.h"
+#include "xfs_ondisk.h"
 
 #include <linux/namei.h>
 #include <linux/init.h>
@@ -65,83 +66,85 @@ static struct kset *xfs_kset;		/* top-level xfs sysfs dir */
 static struct xfs_kobj xfs_dbg_kobj;	/* global debug sysfs attrs */
 #endif
 
-#define MNTOPT_LOGBUFS	"logbufs"	/* number of XFS log buffers */
-#define MNTOPT_LOGBSIZE	"logbsize"	/* size of XFS log buffers */
-#define MNTOPT_LOGDEV	"logdev"	/* log device */
-#define MNTOPT_RTDEV	"rtdev"		/* realtime I/O device */
-#define MNTOPT_BIOSIZE	"biosize"	/* log2 of preferred buffered io size */
-#define MNTOPT_WSYNC	"wsync"		/* safe-mode nfs compatible mount */
-#define MNTOPT_NOALIGN	"noalign"	/* turn off stripe alignment */
-#define MNTOPT_SWALLOC	"swalloc"	/* turn on stripe width allocation */
-#define MNTOPT_SUNIT	"sunit"		/* data volume stripe unit */
-#define MNTOPT_SWIDTH	"swidth"	/* data volume stripe width */
-#define MNTOPT_NOUUID	"nouuid"	/* ignore filesystem UUID */
-#define MNTOPT_MTPT	"mtpt"		/* filesystem mount point */
-#define MNTOPT_GRPID	"grpid"		/* group-ID from parent directory */
-#define MNTOPT_NOGRPID	"nogrpid"	/* group-ID from current process */
-#define MNTOPT_BSDGROUPS    "bsdgroups"    /* group-ID from parent directory */
-#define MNTOPT_SYSVGROUPS   "sysvgroups"   /* group-ID from current process */
-#define MNTOPT_ALLOCSIZE    "allocsize"    /* preferred allocation size */
-#define MNTOPT_NORECOVERY   "norecovery"   /* don't run XFS recovery */
-#define MNTOPT_BARRIER	"barrier"	/* use writer barriers for log write and
-					 * unwritten extent conversion */
-#define MNTOPT_NOBARRIER "nobarrier"	/* .. disable */
-#define MNTOPT_64BITINODE   "inode64"	/* inodes can be allocated anywhere */
-#define MNTOPT_32BITINODE   "inode32"	/* inode allocation limited to
-					 * XFS_MAXINUMBER_32 */
-#define MNTOPT_IKEEP	"ikeep"		/* do not free empty inode clusters */
-#define MNTOPT_NOIKEEP	"noikeep"	/* free empty inode clusters */
-#define MNTOPT_LARGEIO	   "largeio"	/* report large I/O sizes in stat() */
-#define MNTOPT_NOLARGEIO   "nolargeio"	/* do not report large I/O sizes
-					 * in stat(). */
-#define MNTOPT_ATTR2	"attr2"		/* do use attr2 attribute format */
-#define MNTOPT_NOATTR2	"noattr2"	/* do not use attr2 attribute format */
-#define MNTOPT_FILESTREAM  "filestreams" /* use filestreams allocator */
-#define MNTOPT_QUOTA	"quota"		/* disk quotas (user) */
-#define MNTOPT_NOQUOTA	"noquota"	/* no quotas */
-#define MNTOPT_USRQUOTA	"usrquota"	/* user quota enabled */
-#define MNTOPT_GRPQUOTA	"grpquota"	/* group quota enabled */
-#define MNTOPT_PRJQUOTA	"prjquota"	/* project quota enabled */
-#define MNTOPT_UQUOTA	"uquota"	/* user quota (IRIX variant) */
-#define MNTOPT_GQUOTA	"gquota"	/* group quota (IRIX variant) */
-#define MNTOPT_PQUOTA	"pquota"	/* project quota (IRIX variant) */
-#define MNTOPT_UQUOTANOENF "uqnoenforce"/* user quota limit enforcement */
-#define MNTOPT_GQUOTANOENF "gqnoenforce"/* group quota limit enforcement */
-#define MNTOPT_PQUOTANOENF "pqnoenforce"/* project quota limit enforcement */
-#define MNTOPT_QUOTANOENF  "qnoenforce"	/* same as uqnoenforce */
-#define MNTOPT_DISCARD	   "discard"	/* Discard unused blocks */
-#define MNTOPT_NODISCARD   "nodiscard"	/* Do not discard unused blocks */
-
-#define MNTOPT_DAX	"dax"		/* Enable direct access to bdev pages */
-
 /*
  * Table driven mount option parser.
- *
- * Currently only used for remount, but it will be used for mount
- * in the future, too.
  */
 enum {
-	Opt_barrier,
-	Opt_nobarrier,
-	Opt_inode64,
-	Opt_inode32,
-	Opt_err
+	Opt_logbufs, Opt_logbsize, Opt_logdev, Opt_rtdev, Opt_biosize,
+	Opt_wsync, Opt_noalign, Opt_swalloc, Opt_sunit, Opt_swidth, Opt_nouuid,
+	Opt_mtpt, Opt_grpid, Opt_nogrpid, Opt_bsdgroups, Opt_sysvgroups,
+	Opt_allocsize, Opt_norecovery, Opt_barrier, Opt_nobarrier,
+	Opt_inode64, Opt_inode32, Opt_ikeep, Opt_noikeep,
+	Opt_largeio, Opt_nolargeio, Opt_attr2, Opt_noattr2, Opt_filestreams,
+	Opt_quota, Opt_noquota, Opt_usrquota, Opt_grpquota, Opt_prjquota,
+	Opt_uquota, Opt_gquota, Opt_pquota,
+	Opt_uqnoenforce, Opt_gqnoenforce, Opt_pqnoenforce, Opt_qnoenforce,
+	Opt_discard, Opt_nodiscard, Opt_dax, Opt_err,
 };
 
 static const match_table_t tokens = {
-	{Opt_barrier, "barrier"},
-	{Opt_nobarrier, "nobarrier"},
-	{Opt_inode64, "inode64"},
-	{Opt_inode32, "inode32"},
-	{Opt_err, NULL}
+	{Opt_logbufs,	"logbufs=%u"},	/* number of XFS log buffers */
+	{Opt_logbsize,	"logbsize=%s"},	/* size of XFS log buffers */
+	{Opt_logdev,	"logdev=%s"},	/* log device */
+	{Opt_rtdev,	"rtdev=%s"},	/* realtime I/O device */
+	{Opt_biosize,	"biosize=%u"},	/* log2 of preferred buffered io size */
+	{Opt_wsync,	"wsync"},	/* safe-mode nfs compatible mount */
+	{Opt_noalign,	"noalign"},	/* turn off stripe alignment */
+	{Opt_swalloc,	"swalloc"},	/* turn on stripe width allocation */
+	{Opt_sunit,	"sunit=%u"},	/* data volume stripe unit */
+	{Opt_swidth,	"swidth=%u"},	/* data volume stripe width */
+	{Opt_nouuid,	"nouuid"},	/* ignore filesystem UUID */
+	{Opt_mtpt,	"mtpt"},	/* filesystem mount point */
+	{Opt_grpid,	"grpid"},	/* group-ID from parent directory */
+	{Opt_nogrpid,	"nogrpid"},	/* group-ID from current process */
+	{Opt_bsdgroups,	"bsdgroups"},	/* group-ID from parent directory */
+	{Opt_sysvgroups,"sysvgroups"},	/* group-ID from current process */
+	{Opt_allocsize,	"allocsize=%s"},/* preferred allocation size */
+	{Opt_norecovery,"norecovery"},	/* don't run XFS recovery */
+	{Opt_barrier,	"barrier"},	/* use writer barriers for log write and
+					 * unwritten extent conversion */
+	{Opt_nobarrier,	"nobarrier"},	/* .. disable */
+	{Opt_inode64,	"inode64"},	/* inodes can be allocated anywhere */
+	{Opt_inode32,   "inode32"},	/* inode allocation limited to
+					 * XFS_MAXINUMBER_32 */
+	{Opt_ikeep,	"ikeep"},	/* do not free empty inode clusters */
+	{Opt_noikeep,	"noikeep"},	/* free empty inode clusters */
+	{Opt_largeio,	"largeio"},	/* report large I/O sizes in stat() */
+	{Opt_nolargeio,	"nolargeio"},	/* do not report large I/O sizes
+					 * in stat(). */
+	{Opt_attr2,	"attr2"},	/* do use attr2 attribute format */
+	{Opt_noattr2,	"noattr2"},	/* do not use attr2 attribute format */
+	{Opt_filestreams,"filestreams"},/* use filestreams allocator */
+	{Opt_quota,	"quota"},	/* disk quotas (user) */
+	{Opt_noquota,	"noquota"},	/* no quotas */
+	{Opt_usrquota,	"usrquota"},	/* user quota enabled */
+	{Opt_grpquota,	"grpquota"},	/* group quota enabled */
+	{Opt_prjquota,	"prjquota"},	/* project quota enabled */
+	{Opt_uquota,	"uquota"},	/* user quota (IRIX variant) */
+	{Opt_gquota,	"gquota"},	/* group quota (IRIX variant) */
+	{Opt_pquota,	"pquota"},	/* project quota (IRIX variant) */
+	{Opt_uqnoenforce,"uqnoenforce"},/* user quota limit enforcement */
+	{Opt_gqnoenforce,"gqnoenforce"},/* group quota limit enforcement */
+	{Opt_pqnoenforce,"pqnoenforce"},/* project quota limit enforcement */
+	{Opt_qnoenforce, "qnoenforce"},	/* same as uqnoenforce */
+	{Opt_discard,	"discard"},	/* Discard unused blocks */
+	{Opt_nodiscard,	"nodiscard"},	/* Do not discard unused blocks */
+
+	{Opt_dax,	"dax"},		/* Enable direct access to bdev pages */
+	{Opt_err,	NULL},
 };
 
 
-STATIC unsigned long
-suffix_kstrtoint(char *s, unsigned int base, int *res)
+STATIC int
+suffix_kstrtoint(const substring_t *s, unsigned int base, int *res)
 {
 	int	last, shift_left_factor = 0, _res;
-	char	*value = s;
+	char	*value;
+	int	ret = 0;
+
+	value = match_strdup(s);
+	if (!value)
+		return -ENOMEM;
 
 	last = strlen(value) - 1;
 	if (value[last] == 'K' || value[last] == 'k') {
@@ -157,10 +160,11 @@ suffix_kstrtoint(char *s, unsigned int base, int *res)
 		value[last] = '\0';
 	}
 
-	if (kstrtoint(s, base, &_res))
-		return -EINVAL;
+	if (kstrtoint(value, base, &_res))
+		ret = -EINVAL;
+	kfree(value);
 	*res = _res << shift_left_factor;
-	return 0;
+	return ret;
 }
 
 /*
@@ -169,14 +173,19 @@ suffix_kstrtoint(char *s, unsigned int base, int *res)
  *
  * Note that this function leaks the various device name allocations on
  * failure.  The caller takes care of them.
+ *
+ * *sb is const because this is also used to test options on the remount
+ * path, and we don't want this to have any side effects at remount time.
+ * Today this function does not change *sb, but just to future-proof...
  */
 STATIC int
 xfs_parseargs(
 	struct xfs_mount	*mp,
 	char			*options)
 {
-	struct super_block	*sb = mp->m_super;
-	char			*this_char, *value;
+	const struct super_block *sb = mp->m_super;
+	char			*p;
+	substring_t		args[MAX_OPT_ARGS];
 	int			dsunit = 0;
 	int			dswidth = 0;
 	int			iosize = 0;
@@ -217,152 +226,152 @@ xfs_parseargs(
 	if (!options)
 		goto done;
 
-	while ((this_char = strsep(&options, ",")) != NULL) {
-		if (!*this_char)
+	while ((p = strsep(&options, ",")) != NULL) {
+		int		token;
+
+		if (!*p)
 			continue;
-		if ((value = strchr(this_char, '=')) != NULL)
-			*value++ = 0;
 
-		if (!strcmp(this_char, MNTOPT_LOGBUFS)) {
-			if (!value || !*value) {
-				xfs_warn(mp, "%s option requires an argument",
-					this_char);
-				return -EINVAL;
-			}
-			if (kstrtoint(value, 10, &mp->m_logbufs))
-				return -EINVAL;
-		} else if (!strcmp(this_char, MNTOPT_LOGBSIZE)) {
-			if (!value || !*value) {
-				xfs_warn(mp, "%s option requires an argument",
-					this_char);
-				return -EINVAL;
-			}
-			if (suffix_kstrtoint(value, 10, &mp->m_logbsize))
+		token = match_token(p, tokens, args);
+		switch (token) {
+		case Opt_logbufs:
+			if (match_int(args, &mp->m_logbufs))
 				return -EINVAL;
-		} else if (!strcmp(this_char, MNTOPT_LOGDEV)) {
-			if (!value || !*value) {
-				xfs_warn(mp, "%s option requires an argument",
-					this_char);
+			break;
+		case Opt_logbsize:
+			if (suffix_kstrtoint(args, 10, &mp->m_logbsize))
 				return -EINVAL;
-			}
-			mp->m_logname = kstrndup(value, MAXNAMELEN, GFP_KERNEL);
+			break;
+		case Opt_logdev:
+			mp->m_logname = match_strdup(args);
 			if (!mp->m_logname)
 				return -ENOMEM;
-		} else if (!strcmp(this_char, MNTOPT_MTPT)) {
-			xfs_warn(mp, "%s option not allowed on this system",
-				this_char);
+			break;
+		case Opt_mtpt:
+			xfs_warn(mp, "%s option not allowed on this system", p);
 			return -EINVAL;
-		} else if (!strcmp(this_char, MNTOPT_RTDEV)) {
-			if (!value || !*value) {
-				xfs_warn(mp, "%s option requires an argument",
-					this_char);
-				return -EINVAL;
-			}
-			mp->m_rtname = kstrndup(value, MAXNAMELEN, GFP_KERNEL);
+		case Opt_rtdev:
+			mp->m_rtname = match_strdup(args);
 			if (!mp->m_rtname)
 				return -ENOMEM;
-		} else if (!strcmp(this_char, MNTOPT_ALLOCSIZE) ||
-			   !strcmp(this_char, MNTOPT_BIOSIZE)) {
-			if (!value || !*value) {
-				xfs_warn(mp, "%s option requires an argument",
-					this_char);
-				return -EINVAL;
-			}
-			if (suffix_kstrtoint(value, 10, &iosize))
+			break;
+		case Opt_allocsize:
+		case Opt_biosize:
+			if (suffix_kstrtoint(args, 10, &iosize))
 				return -EINVAL;
 			iosizelog = ffs(iosize) - 1;
-		} else if (!strcmp(this_char, MNTOPT_GRPID) ||
-			   !strcmp(this_char, MNTOPT_BSDGROUPS)) {
+			break;
+		case Opt_grpid:
+		case Opt_bsdgroups:
 			mp->m_flags |= XFS_MOUNT_GRPID;
-		} else if (!strcmp(this_char, MNTOPT_NOGRPID) ||
-			   !strcmp(this_char, MNTOPT_SYSVGROUPS)) {
+			break;
+		case Opt_nogrpid:
+		case Opt_sysvgroups:
 			mp->m_flags &= ~XFS_MOUNT_GRPID;
-		} else if (!strcmp(this_char, MNTOPT_WSYNC)) {
+			break;
+		case Opt_wsync:
 			mp->m_flags |= XFS_MOUNT_WSYNC;
-		} else if (!strcmp(this_char, MNTOPT_NORECOVERY)) {
+			break;
+		case Opt_norecovery:
 			mp->m_flags |= XFS_MOUNT_NORECOVERY;
-		} else if (!strcmp(this_char, MNTOPT_NOALIGN)) {
+			break;
+		case Opt_noalign:
 			mp->m_flags |= XFS_MOUNT_NOALIGN;
-		} else if (!strcmp(this_char, MNTOPT_SWALLOC)) {
+			break;
+		case Opt_swalloc:
 			mp->m_flags |= XFS_MOUNT_SWALLOC;
-		} else if (!strcmp(this_char, MNTOPT_SUNIT)) {
-			if (!value || !*value) {
-				xfs_warn(mp, "%s option requires an argument",
-					this_char);
-				return -EINVAL;
-			}
-			if (kstrtoint(value, 10, &dsunit))
-				return -EINVAL;
-		} else if (!strcmp(this_char, MNTOPT_SWIDTH)) {
-			if (!value || !*value) {
-				xfs_warn(mp, "%s option requires an argument",
-					this_char);
+			break;
+		case Opt_sunit:
+			if (match_int(args, &dsunit))
 				return -EINVAL;
-			}
-			if (kstrtoint(value, 10, &dswidth))
+			break;
+		case Opt_swidth:
+			if (match_int(args, &dswidth))
 				return -EINVAL;
-		} else if (!strcmp(this_char, MNTOPT_32BITINODE)) {
+			break;
+		case Opt_inode32:
 			mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
-		} else if (!strcmp(this_char, MNTOPT_64BITINODE)) {
+			break;
+		case Opt_inode64:
 			mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
-		} else if (!strcmp(this_char, MNTOPT_NOUUID)) {
+			break;
+		case Opt_nouuid:
 			mp->m_flags |= XFS_MOUNT_NOUUID;
-		} else if (!strcmp(this_char, MNTOPT_BARRIER)) {
+			break;
+		case Opt_barrier:
 			mp->m_flags |= XFS_MOUNT_BARRIER;
-		} else if (!strcmp(this_char, MNTOPT_NOBARRIER)) {
+			break;
+		case Opt_nobarrier:
 			mp->m_flags &= ~XFS_MOUNT_BARRIER;
-		} else if (!strcmp(this_char, MNTOPT_IKEEP)) {
+			break;
+		case Opt_ikeep:
 			mp->m_flags |= XFS_MOUNT_IKEEP;
-		} else if (!strcmp(this_char, MNTOPT_NOIKEEP)) {
+			break;
+		case Opt_noikeep:
 			mp->m_flags &= ~XFS_MOUNT_IKEEP;
-		} else if (!strcmp(this_char, MNTOPT_LARGEIO)) {
+			break;
+		case Opt_largeio:
 			mp->m_flags &= ~XFS_MOUNT_COMPAT_IOSIZE;
-		} else if (!strcmp(this_char, MNTOPT_NOLARGEIO)) {
+			break;
+		case Opt_nolargeio:
 			mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
-		} else if (!strcmp(this_char, MNTOPT_ATTR2)) {
+			break;
+		case Opt_attr2:
 			mp->m_flags |= XFS_MOUNT_ATTR2;
-		} else if (!strcmp(this_char, MNTOPT_NOATTR2)) {
+			break;
+		case Opt_noattr2:
 			mp->m_flags &= ~XFS_MOUNT_ATTR2;
 			mp->m_flags |= XFS_MOUNT_NOATTR2;
-		} else if (!strcmp(this_char, MNTOPT_FILESTREAM)) {
+			break;
+		case Opt_filestreams:
 			mp->m_flags |= XFS_MOUNT_FILESTREAMS;
-		} else if (!strcmp(this_char, MNTOPT_NOQUOTA)) {
+			break;
+		case Opt_noquota:
 			mp->m_qflags &= ~XFS_ALL_QUOTA_ACCT;
 			mp->m_qflags &= ~XFS_ALL_QUOTA_ENFD;
 			mp->m_qflags &= ~XFS_ALL_QUOTA_ACTIVE;
-		} else if (!strcmp(this_char, MNTOPT_QUOTA) ||
-			   !strcmp(this_char, MNTOPT_UQUOTA) ||
-			   !strcmp(this_char, MNTOPT_USRQUOTA)) {
+			break;
+		case Opt_quota:
+		case Opt_uquota:
+		case Opt_usrquota:
 			mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE |
 					 XFS_UQUOTA_ENFD);
-		} else if (!strcmp(this_char, MNTOPT_QUOTANOENF) ||
-			   !strcmp(this_char, MNTOPT_UQUOTANOENF)) {
+			break;
+		case Opt_qnoenforce:
+		case Opt_uqnoenforce:
 			mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE);
 			mp->m_qflags &= ~XFS_UQUOTA_ENFD;
-		} else if (!strcmp(this_char, MNTOPT_PQUOTA) ||
-			   !strcmp(this_char, MNTOPT_PRJQUOTA)) {
+			break;
+		case Opt_pquota:
+		case Opt_prjquota:
 			mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE |
 					 XFS_PQUOTA_ENFD);
-		} else if (!strcmp(this_char, MNTOPT_PQUOTANOENF)) {
+			break;
+		case Opt_pqnoenforce:
 			mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE);
 			mp->m_qflags &= ~XFS_PQUOTA_ENFD;
-		} else if (!strcmp(this_char, MNTOPT_GQUOTA) ||
-			   !strcmp(this_char, MNTOPT_GRPQUOTA)) {
+		case Opt_gquota:
+		case Opt_grpquota:
 			mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE |
 					 XFS_GQUOTA_ENFD);
-		} else if (!strcmp(this_char, MNTOPT_GQUOTANOENF)) {
+			break;
+		case Opt_gqnoenforce:
 			mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE);
 			mp->m_qflags &= ~XFS_GQUOTA_ENFD;
-		} else if (!strcmp(this_char, MNTOPT_DISCARD)) {
+			break;
+		case Opt_discard:
 			mp->m_flags |= XFS_MOUNT_DISCARD;
-		} else if (!strcmp(this_char, MNTOPT_NODISCARD)) {
+			break;
+		case Opt_nodiscard:
 			mp->m_flags &= ~XFS_MOUNT_DISCARD;
+			break;
 #ifdef CONFIG_FS_DAX
-		} else if (!strcmp(this_char, MNTOPT_DAX)) {
+		case Opt_dax:
 			mp->m_flags |= XFS_MOUNT_DAX;
+			break;
 #endif
-		} else {
-			xfs_warn(mp, "unknown mount option [%s].", this_char);
+		default:
+			xfs_warn(mp, "unknown mount option [%s].", p);
 			return -EINVAL;
 		}
 	}
@@ -461,25 +470,25 @@ xfs_showargs(
 {
 	static struct proc_xfs_info xfs_info_set[] = {
 		/* the few simple ones we can get from the mount struct */
-		{ XFS_MOUNT_IKEEP,		"," MNTOPT_IKEEP },
-		{ XFS_MOUNT_WSYNC,		"," MNTOPT_WSYNC },
-		{ XFS_MOUNT_NOALIGN,		"," MNTOPT_NOALIGN },
-		{ XFS_MOUNT_SWALLOC,		"," MNTOPT_SWALLOC },
-		{ XFS_MOUNT_NOUUID,		"," MNTOPT_NOUUID },
-		{ XFS_MOUNT_NORECOVERY,		"," MNTOPT_NORECOVERY },
-		{ XFS_MOUNT_ATTR2,		"," MNTOPT_ATTR2 },
-		{ XFS_MOUNT_FILESTREAMS,	"," MNTOPT_FILESTREAM },
-		{ XFS_MOUNT_GRPID,		"," MNTOPT_GRPID },
-		{ XFS_MOUNT_DISCARD,		"," MNTOPT_DISCARD },
-		{ XFS_MOUNT_SMALL_INUMS,	"," MNTOPT_32BITINODE },
-		{ XFS_MOUNT_DAX,		"," MNTOPT_DAX },
+		{ XFS_MOUNT_IKEEP,		",ikeep" },
+		{ XFS_MOUNT_WSYNC,		",wsync" },
+		{ XFS_MOUNT_NOALIGN,		",noalign" },
+		{ XFS_MOUNT_SWALLOC,		",swalloc" },
+		{ XFS_MOUNT_NOUUID,		",nouuid" },
+		{ XFS_MOUNT_NORECOVERY,		",norecovery" },
+		{ XFS_MOUNT_ATTR2,		",attr2" },
+		{ XFS_MOUNT_FILESTREAMS,	",filestreams" },
+		{ XFS_MOUNT_GRPID,		",grpid" },
+		{ XFS_MOUNT_DISCARD,		",discard" },
+		{ XFS_MOUNT_SMALL_INUMS,	",inode32" },
+		{ XFS_MOUNT_DAX,		",dax" },
 		{ 0, NULL }
 	};
 	static struct proc_xfs_info xfs_info_unset[] = {
 		/* the few simple ones we can get from the mount struct */
-		{ XFS_MOUNT_COMPAT_IOSIZE,	"," MNTOPT_LARGEIO },
-		{ XFS_MOUNT_BARRIER,		"," MNTOPT_NOBARRIER },
-		{ XFS_MOUNT_SMALL_INUMS,	"," MNTOPT_64BITINODE },
+		{ XFS_MOUNT_COMPAT_IOSIZE,	",largeio" },
+		{ XFS_MOUNT_BARRIER,		",nobarrier" },
+		{ XFS_MOUNT_SMALL_INUMS,	",inode64" },
 		{ 0, NULL }
 	};
 	struct proc_xfs_info	*xfs_infop;
@@ -494,46 +503,46 @@ xfs_showargs(
 	}
 
 	if (mp->m_flags & XFS_MOUNT_DFLT_IOSIZE)
-		seq_printf(m, "," MNTOPT_ALLOCSIZE "=%dk",
+		seq_printf(m, ",allocsize=%dk",
 				(int)(1 << mp->m_writeio_log) >> 10);
 
 	if (mp->m_logbufs > 0)
-		seq_printf(m, "," MNTOPT_LOGBUFS "=%d", mp->m_logbufs);
+		seq_printf(m, ",logbufs=%d", mp->m_logbufs);
 	if (mp->m_logbsize > 0)
-		seq_printf(m, "," MNTOPT_LOGBSIZE "=%dk", mp->m_logbsize >> 10);
+		seq_printf(m, ",logbsize=%dk", mp->m_logbsize >> 10);
 
 	if (mp->m_logname)
-		seq_show_option(m, MNTOPT_LOGDEV, mp->m_logname);
+		seq_show_option(m, "logdev", mp->m_logname);
 	if (mp->m_rtname)
-		seq_show_option(m, MNTOPT_RTDEV, mp->m_rtname);
+		seq_show_option(m, "rtdev", mp->m_rtname);
 
 	if (mp->m_dalign > 0)
-		seq_printf(m, "," MNTOPT_SUNIT "=%d",
+		seq_printf(m, ",sunit=%d",
 				(int)XFS_FSB_TO_BB(mp, mp->m_dalign));
 	if (mp->m_swidth > 0)
-		seq_printf(m, "," MNTOPT_SWIDTH "=%d",
+		seq_printf(m, ",swidth=%d",
 				(int)XFS_FSB_TO_BB(mp, mp->m_swidth));
 
 	if (mp->m_qflags & (XFS_UQUOTA_ACCT|XFS_UQUOTA_ENFD))
-		seq_puts(m, "," MNTOPT_USRQUOTA);
+		seq_puts(m, ",usrquota");
 	else if (mp->m_qflags & XFS_UQUOTA_ACCT)
-		seq_puts(m, "," MNTOPT_UQUOTANOENF);
+		seq_puts(m, ",uqnoenforce");
 
 	if (mp->m_qflags & XFS_PQUOTA_ACCT) {
 		if (mp->m_qflags & XFS_PQUOTA_ENFD)
-			seq_puts(m, "," MNTOPT_PRJQUOTA);
+			seq_puts(m, ",prjquota");
 		else
-			seq_puts(m, "," MNTOPT_PQUOTANOENF);
+			seq_puts(m, ",pqnoenforce");
 	}
 	if (mp->m_qflags & XFS_GQUOTA_ACCT) {
 		if (mp->m_qflags & XFS_GQUOTA_ENFD)
-			seq_puts(m, "," MNTOPT_GRPQUOTA);
+			seq_puts(m, ",grpquota");
 		else
-			seq_puts(m, "," MNTOPT_GQUOTANOENF);
+			seq_puts(m, ",gqnoenforce");
 	}
 
 	if (!(mp->m_qflags & XFS_ALL_QUOTA_ACCT))
-		seq_puts(m, "," MNTOPT_NOQUOTA);
+		seq_puts(m, ",noquota");
 
 	return 0;
 }
@@ -547,10 +556,10 @@ xfs_max_file_offset(
 	/* Figure out maximum filesize, on Linux this can depend on
 	 * the filesystem blocksize (on 32 bit platforms).
 	 * __block_write_begin does this in an [unsigned] long...
-	 *      page->index << (PAGE_CACHE_SHIFT - bbits)
+	 *      page->index << (PAGE_SHIFT - bbits)
 	 * So, for page sized blocks (4K on 32 bit platforms),
 	 * this wraps at around 8Tb (hence MAX_LFS_FILESIZE which is
-	 *      (((u64)PAGE_CACHE_SIZE << (BITS_PER_LONG-1))-1)
+	 *      (((u64)PAGE_SIZE << (BITS_PER_LONG-1))-1)
 	 * but for smaller blocksizes it is less (bbits = log2 bsize).
 	 * Note1: get_block_t takes a long (implicit cast from above)
 	 * Note2: The Large Block Device (LBD and HAVE_SECTOR_T) patch
@@ -561,10 +570,10 @@ xfs_max_file_offset(
 #if BITS_PER_LONG == 32
 # if defined(CONFIG_LBDAF)
 	ASSERT(sizeof(sector_t) == 8);
-	pagefactor = PAGE_CACHE_SIZE;
+	pagefactor = PAGE_SIZE;
 	bitshift = BITS_PER_LONG;
 # else
-	pagefactor = PAGE_CACHE_SIZE >> (PAGE_CACHE_SHIFT - blockshift);
+	pagefactor = PAGE_SIZE >> (PAGE_SHIFT - blockshift);
 # endif
 #endif
 
@@ -572,23 +581,35 @@ xfs_max_file_offset(
 }
 
 /*
- * xfs_set_inode32() and xfs_set_inode64() are passed an agcount
- * because in the growfs case, mp->m_sb.sb_agcount is not updated
- * yet to the potentially higher ag count.
+ * Set parameters for inode allocation heuristics, taking into account
+ * filesystem size and inode32/inode64 mount options; i.e. specifically
+ * whether or not XFS_MOUNT_SMALL_INUMS is set.
+ *
+ * Inode allocation patterns are altered only if inode32 is requested
+ * (XFS_MOUNT_SMALL_INUMS), and the filesystem is sufficiently large.
+ * If altered, XFS_MOUNT_32BITINODES is set as well.
+ *
+ * An agcount independent of that in the mount structure is provided
+ * because in the growfs case, mp->m_sb.sb_agcount is not yet updated
+ * to the potentially higher ag count.
+ *
+ * Returns the maximum AG index which may contain inodes.
  */
 xfs_agnumber_t
-xfs_set_inode32(struct xfs_mount *mp, xfs_agnumber_t agcount)
+xfs_set_inode_alloc(
+	struct xfs_mount *mp,
+	xfs_agnumber_t	agcount)
 {
-	xfs_agnumber_t	index = 0;
+	xfs_agnumber_t	index;
 	xfs_agnumber_t	maxagi = 0;
 	xfs_sb_t	*sbp = &mp->m_sb;
 	xfs_agnumber_t	max_metadata;
 	xfs_agino_t	agino;
 	xfs_ino_t	ino;
-	xfs_perag_t	*pag;
 
-	/* Calculate how much should be reserved for inodes to meet
-	 * the max inode percentage.
+	/*
+	 * Calculate how much should be reserved for inodes to meet
+	 * the max inode percentage.  Used only for inode32.
 	 */
 	if (mp->m_maxicount) {
 		__uint64_t	icount;
@@ -602,54 +623,48 @@ xfs_set_inode32(struct xfs_mount *mp, xfs_agnumber_t agcount)
 		max_metadata = agcount;
 	}
 
+	/* Get the last possible inode in the filesystem */
 	agino =	XFS_OFFBNO_TO_AGINO(mp, sbp->sb_agblocks - 1, 0);
+	ino = XFS_AGINO_TO_INO(mp, agcount - 1, agino);
+
+	/*
+	 * If user asked for no more than 32-bit inodes, and the fs is
+	 * sufficiently large, set XFS_MOUNT_32BITINODES if we must alter
+	 * the allocator to accommodate the request.
+	 */
+	if ((mp->m_flags & XFS_MOUNT_SMALL_INUMS) && ino > XFS_MAXINUMBER_32)
+		mp->m_flags |= XFS_MOUNT_32BITINODES;
+	else
+		mp->m_flags &= ~XFS_MOUNT_32BITINODES;
 
 	for (index = 0; index < agcount; index++) {
-		ino = XFS_AGINO_TO_INO(mp, index, agino);
+		struct xfs_perag	*pag;
 
-		if (ino > XFS_MAXINUMBER_32) {
-			pag = xfs_perag_get(mp, index);
-			pag->pagi_inodeok = 0;
-			pag->pagf_metadata = 0;
-			xfs_perag_put(pag);
-			continue;
-		}
+		ino = XFS_AGINO_TO_INO(mp, index, agino);
 
 		pag = xfs_perag_get(mp, index);
-		pag->pagi_inodeok = 1;
-		maxagi++;
-		if (index < max_metadata)
-			pag->pagf_metadata = 1;
-		xfs_perag_put(pag);
-	}
-	mp->m_flags |= (XFS_MOUNT_32BITINODES |
-			XFS_MOUNT_SMALL_INUMS);
 
-	return maxagi;
-}
-
-xfs_agnumber_t
-xfs_set_inode64(struct xfs_mount *mp, xfs_agnumber_t agcount)
-{
-	xfs_agnumber_t index = 0;
-
-	for (index = 0; index < agcount; index++) {
-		struct xfs_perag	*pag;
+		if (mp->m_flags & XFS_MOUNT_32BITINODES) {
+			if (ino > XFS_MAXINUMBER_32) {
+				pag->pagi_inodeok = 0;
+				pag->pagf_metadata = 0;
+			} else {
+				pag->pagi_inodeok = 1;
+				maxagi++;
+				if (index < max_metadata)
+					pag->pagf_metadata = 1;
+				else
+					pag->pagf_metadata = 0;
+			}
+		} else {
+			pag->pagi_inodeok = 1;
+			pag->pagf_metadata = 0;
+		}
 
-		pag = xfs_perag_get(mp, index);
-		pag->pagi_inodeok = 1;
-		pag->pagf_metadata = 0;
 		xfs_perag_put(pag);
 	}
 
-	/* There is no need for lock protection on m_flags,
-	 * the rw_semaphore of the VFS superblock is locked
-	 * during mount/umount/remount operations, so this is
-	 * enough to avoid concurency on the m_flags field
-	 */
-	mp->m_flags &= ~(XFS_MOUNT_32BITINODES |
-			 XFS_MOUNT_SMALL_INUMS);
-	return index;
+	return (mp->m_flags & XFS_MOUNT_32BITINODES) ? maxagi : agcount;
 }
 
 STATIC int
@@ -838,17 +853,18 @@ xfs_init_mount_workqueues(
 		goto out_destroy_unwritten;
 
 	mp->m_reclaim_workqueue = alloc_workqueue("xfs-reclaim/%s",
-			WQ_FREEZABLE, 0, mp->m_fsname);
+			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_fsname);
 	if (!mp->m_reclaim_workqueue)
 		goto out_destroy_cil;
 
 	mp->m_log_workqueue = alloc_workqueue("xfs-log/%s",
-			WQ_FREEZABLE|WQ_HIGHPRI, 0, mp->m_fsname);
+			WQ_MEM_RECLAIM|WQ_FREEZABLE|WQ_HIGHPRI, 0,
+			mp->m_fsname);
 	if (!mp->m_log_workqueue)
 		goto out_destroy_reclaim;
 
 	mp->m_eofblocks_workqueue = alloc_workqueue("xfs-eofblocks/%s",
-			WQ_FREEZABLE, 0, mp->m_fsname);
+			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_fsname);
 	if (!mp->m_eofblocks_workqueue)
 		goto out_destroy_log;
 
@@ -922,7 +938,7 @@ xfs_fs_destroy_inode(
 
 	trace_xfs_destroy_inode(ip);
 
-	XFS_STATS_INC(vn_reclaim);
+	XFS_STATS_INC(ip->i_mount, vn_reclaim);
 
 	ASSERT(XFS_FORCED_SHUTDOWN(ip->i_mount) || ip->i_delayed_blks == 0);
 
@@ -983,8 +999,8 @@ xfs_fs_evict_inode(
 
 	truncate_inode_pages_final(&inode->i_data);
 	clear_inode(inode);
-	XFS_STATS_INC(vn_rele);
-	XFS_STATS_INC(vn_remove);
+	XFS_STATS_INC(ip->i_mount, vn_rele);
+	XFS_STATS_INC(ip->i_mount, vn_remove);
 
 	xfs_inactive(ip);
 }
@@ -1165,6 +1181,27 @@ xfs_quiesce_attr(
 }
 
 STATIC int
+xfs_test_remount_options(
+	struct super_block	*sb,
+	struct xfs_mount	*mp,
+	char			*options)
+{
+	int			error = 0;
+	struct xfs_mount	*tmp_mp;
+
+	tmp_mp = kmem_zalloc(sizeof(*tmp_mp), KM_MAYFAIL);
+	if (!tmp_mp)
+		return -ENOMEM;
+
+	tmp_mp->m_super = sb;
+	error = xfs_parseargs(tmp_mp, options);
+	xfs_free_fsname(tmp_mp);
+	kfree(tmp_mp);
+
+	return error;
+}
+
+STATIC int
 xfs_fs_remount(
 	struct super_block	*sb,
 	int			*flags,
@@ -1176,6 +1213,11 @@ xfs_fs_remount(
 	char			*p;
 	int			error;
 
+	/* First, check for complete junk; i.e. invalid options */
+	error = xfs_test_remount_options(sb, mp, options);
+	if (error)
+		return error;
+
 	sync_filesystem(sb);
 	while ((p = strsep(&options, ",")) != NULL) {
 		int token;
@@ -1192,10 +1234,12 @@ xfs_fs_remount(
 			mp->m_flags &= ~XFS_MOUNT_BARRIER;
 			break;
 		case Opt_inode64:
-			mp->m_maxagi = xfs_set_inode64(mp, sbp->sb_agcount);
+			mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
+			mp->m_maxagi = xfs_set_inode_alloc(mp, sbp->sb_agcount);
 			break;
 		case Opt_inode32:
-			mp->m_maxagi = xfs_set_inode32(mp, sbp->sb_agcount);
+			mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
+			mp->m_maxagi = xfs_set_inode_alloc(mp, sbp->sb_agcount);
 			break;
 		default:
 			/*
@@ -1343,9 +1387,8 @@ xfs_finish_flags(
 	 */
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	    (mp->m_flags & XFS_MOUNT_NOATTR2)) {
-		xfs_warn(mp,
-"Cannot mount a V5 filesystem as %s. %s is always enabled for V5 filesystems.",
-			MNTOPT_NOATTR2, MNTOPT_ATTR2);
+		xfs_warn(mp, "Cannot mount a V5 filesystem as noattr2. "
+			     "attr2 is always enabled for V5 filesystems.");
 		return -EINVAL;
 	}
 
@@ -1474,9 +1517,16 @@ xfs_fs_fill_super(
 	if (error)
 		goto out_destroy_workqueues;
 
+	/* Allocate stats memory before we do operations that might use it */
+	mp->m_stats.xs_stats = alloc_percpu(struct xfsstats);
+	if (!mp->m_stats.xs_stats) {
+		error = -ENOMEM;
+		goto out_destroy_counters;
+	}
+
 	error = xfs_readsb(mp, flags);
 	if (error)
-		goto out_destroy_counters;
+		goto out_free_stats;
 
 	error = xfs_finish_flags(mp);
 	if (error)
@@ -1545,9 +1595,11 @@ xfs_fs_fill_super(
 	xfs_filestream_unmount(mp);
  out_free_sb:
 	xfs_freesb(mp);
+ out_free_stats:
+	free_percpu(mp->m_stats.xs_stats);
  out_destroy_counters:
 	xfs_destroy_percpu_counters(mp);
-out_destroy_workqueues:
+ out_destroy_workqueues:
 	xfs_destroy_mount_workqueues(mp);
  out_close_devices:
 	xfs_close_devices(mp);
@@ -1574,6 +1626,7 @@ xfs_fs_put_super(
 	xfs_unmountfs(mp);
 
 	xfs_freesb(mp);
+	free_percpu(mp->m_stats.xs_stats);
 	xfs_destroy_percpu_counters(mp);
 	xfs_destroy_mount_workqueues(mp);
 	xfs_close_devices(mp);
@@ -1703,8 +1756,8 @@ xfs_init_zones(void)
 
 	xfs_inode_zone =
 		kmem_zone_init_flags(sizeof(xfs_inode_t), "xfs_inode",
-			KM_ZONE_HWALIGN | KM_ZONE_RECLAIM | KM_ZONE_SPREAD,
-			xfs_fs_inode_init_once);
+			KM_ZONE_HWALIGN | KM_ZONE_RECLAIM | KM_ZONE_SPREAD |
+			KM_ZONE_ACCOUNT, xfs_fs_inode_init_once);
 	if (!xfs_inode_zone)
 		goto out_destroy_efi_zone;
 
@@ -1806,6 +1859,8 @@ init_xfs_fs(void)
 {
 	int			error;
 
+	xfs_check_ondisk_structs();
+
 	printk(KERN_INFO XFS_VERSION_STRING " with "
 			 XFS_BUILD_OPTIONS " enabled\n");
 
@@ -1838,19 +1893,32 @@ init_xfs_fs(void)
 	xfs_kset = kset_create_and_add("xfs", NULL, fs_kobj);
 	if (!xfs_kset) {
 		error = -ENOMEM;
-		goto out_sysctl_unregister;;
+		goto out_sysctl_unregister;
 	}
 
+	xfsstats.xs_kobj.kobject.kset = xfs_kset;
+
+	xfsstats.xs_stats = alloc_percpu(struct xfsstats);
+	if (!xfsstats.xs_stats) {
+		error = -ENOMEM;
+		goto out_kset_unregister;
+	}
+
+	error = xfs_sysfs_init(&xfsstats.xs_kobj, &xfs_stats_ktype, NULL,
+			       "stats");
+	if (error)
+		goto out_free_stats;
+
 #ifdef DEBUG
 	xfs_dbg_kobj.kobject.kset = xfs_kset;
 	error = xfs_sysfs_init(&xfs_dbg_kobj, &xfs_dbg_ktype, NULL, "debug");
 	if (error)
-		goto out_kset_unregister;
+		goto out_remove_stats_kobj;
 #endif
 
 	error = xfs_qm_init();
 	if (error)
-		goto out_remove_kobj;
+		goto out_remove_dbg_kobj;
 
 	error = register_filesystem(&xfs_fs_type);
 	if (error)
@@ -1859,11 +1927,15 @@ init_xfs_fs(void)
 
  out_qm_exit:
 	xfs_qm_exit();
- out_remove_kobj:
+ out_remove_dbg_kobj:
 #ifdef DEBUG
 	xfs_sysfs_del(&xfs_dbg_kobj);
- out_kset_unregister:
+ out_remove_stats_kobj:
 #endif
+	xfs_sysfs_del(&xfsstats.xs_kobj);
+ out_free_stats:
+	free_percpu(xfsstats.xs_stats);
+ out_kset_unregister:
 	kset_unregister(xfs_kset);
  out_sysctl_unregister:
 	xfs_sysctl_unregister();
@@ -1889,6 +1961,8 @@ exit_xfs_fs(void)
 #ifdef DEBUG
 	xfs_sysfs_del(&xfs_dbg_kobj);
 #endif
+	xfs_sysfs_del(&xfsstats.xs_kobj);
+	free_percpu(xfsstats.xs_stats);
 	kset_unregister(xfs_kset);
 	xfs_sysctl_unregister();
 	xfs_cleanup_procfs();
@@ -1896,6 +1970,7 @@ exit_xfs_fs(void)
 	xfs_mru_cache_uninit();
 	xfs_destroy_workqueues();
 	xfs_destroy_zones();
+	xfs_uuid_table_free();
 }
 
 module_init(init_xfs_fs);
diff --git a/fs/xfs/xfs_super.h b/fs/xfs/xfs_super.h
index 499058fea303..2dfb1ce4585f 100644
--- a/fs/xfs/xfs_super.h
+++ b/fs/xfs/xfs_super.h
@@ -65,8 +65,8 @@ extern __uint64_t xfs_max_file_offset(unsigned int);
 
 extern void xfs_flush_inodes(struct xfs_mount *mp);
 extern void xfs_blkdev_issue_flush(struct xfs_buftarg *);
-extern xfs_agnumber_t xfs_set_inode32(struct xfs_mount *, xfs_agnumber_t agcount);
-extern xfs_agnumber_t xfs_set_inode64(struct xfs_mount *, xfs_agnumber_t agcount);
+extern xfs_agnumber_t xfs_set_inode_alloc(struct xfs_mount *,
+					   xfs_agnumber_t agcount);
 
 extern const struct export_operations xfs_export_operations;
 extern const struct xattr_handler *xfs_xattr_handlers[];
diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c
index 996481eeb491..b44284c1adda 100644
--- a/fs/xfs/xfs_symlink.c
+++ b/fs/xfs/xfs_symlink.c
@@ -178,7 +178,6 @@ xfs_symlink(
 	struct xfs_bmap_free	free_list;
 	xfs_fsblock_t		first_block;
 	bool                    unlock_dp_on_error = false;
-	int			committed;
 	xfs_fileoff_t		first_fsb;
 	xfs_filblks_t		fs_blocks;
 	int			nmaps;
@@ -387,7 +386,7 @@ xfs_symlink(
 		xfs_trans_set_sync(tp);
 	}
 
-	error = xfs_bmap_finish(&tp, &free_list, &committed);
+	error = xfs_bmap_finish(&tp, &free_list, NULL);
 	if (error)
 		goto out_bmap_cancel;
 
@@ -434,7 +433,6 @@ xfs_inactive_symlink_rmt(
 	struct xfs_inode *ip)
 {
 	xfs_buf_t	*bp;
-	int		committed;
 	int		done;
 	int		error;
 	xfs_fsblock_t	first_block;
@@ -510,16 +508,10 @@ xfs_inactive_symlink_rmt(
 	/*
 	 * Commit the first transaction.  This logs the EFI and the inode.
 	 */
-	error = xfs_bmap_finish(&tp, &free_list, &committed);
+	error = xfs_bmap_finish(&tp, &free_list, ip);
 	if (error)
 		goto error_bmap_cancel;
 	/*
-	 * The transaction must have been committed, since there were
-	 * actually extents freed by xfs_bunmapi.  See xfs_bmap_finish.
-	 * The new tp has the extent freeing and EFDs.
-	 */
-	ASSERT(committed);
-	/*
 	 * The first xact was committed, so add the inode to the new one.
 	 * Mark it dirty so it will be logged and moved forward in the log as
 	 * part of every commit.
diff --git a/fs/xfs/xfs_sysctl.c b/fs/xfs/xfs_sysctl.c
index a0c8067cea6f..aed74d3f8da9 100644
--- a/fs/xfs/xfs_sysctl.c
+++ b/fs/xfs/xfs_sysctl.c
@@ -19,6 +19,7 @@
 #include <linux/sysctl.h>
 #include <linux/proc_fs.h>
 #include "xfs_error.h"
+#include "xfs_stats.h"
 
 static struct ctl_table_header *xfs_table_header;
 
@@ -31,22 +32,12 @@ xfs_stats_clear_proc_handler(
 	size_t			*lenp,
 	loff_t			*ppos)
 {
-	int		c, ret, *valp = ctl->data;
-	__uint32_t	vn_active;
+	int		ret, *valp = ctl->data;
 
 	ret = proc_dointvec_minmax(ctl, write, buffer, lenp, ppos);
 
 	if (!ret && write && *valp) {
-		xfs_notice(NULL, "Clearing xfsstats");
-		for_each_possible_cpu(c) {
-			preempt_disable();
-			/* save vn_active, it's a universal truth! */
-			vn_active = per_cpu(xfsstats, c).vn_active;
-			memset(&per_cpu(xfsstats, c), 0,
-			       sizeof(struct xfsstats));
-			per_cpu(xfsstats, c).vn_active = vn_active;
-			preempt_enable();
-		}
+		xfs_stats_clearall(xfsstats.xs_stats);
 		xfs_stats_clear = 0;
 	}
 
diff --git a/fs/xfs/xfs_sysfs.c b/fs/xfs/xfs_sysfs.c
index aa03670851d8..6ced4f143494 100644
--- a/fs/xfs/xfs_sysfs.c
+++ b/fs/xfs/xfs_sysfs.c
@@ -18,14 +18,19 @@
 
 #include "xfs.h"
 #include "xfs_sysfs.h"
+#include "xfs_format.h"
 #include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_log.h"
 #include "xfs_log_priv.h"
+#include "xfs_stats.h"
+#include "xfs_mount.h"
 
 struct xfs_sysfs_attr {
 	struct attribute attr;
-	ssize_t (*show)(char *buf, void *data);
-	ssize_t (*store)(const char *buf, size_t count, void *data);
+	ssize_t (*show)(struct kobject *kobject, char *buf);
+	ssize_t (*store)(struct kobject *kobject, const char *buf,
+			 size_t count);
 };
 
 static inline struct xfs_sysfs_attr *
@@ -38,17 +43,102 @@ to_attr(struct attribute *attr)
 	static struct xfs_sysfs_attr xfs_sysfs_attr_##name = __ATTR_RW(name)
 #define XFS_SYSFS_ATTR_RO(name) \
 	static struct xfs_sysfs_attr xfs_sysfs_attr_##name = __ATTR_RO(name)
+#define XFS_SYSFS_ATTR_WO(name) \
+	static struct xfs_sysfs_attr xfs_sysfs_attr_##name = __ATTR_WO(name)
 
 #define ATTR_LIST(name) &xfs_sysfs_attr_##name.attr
 
+STATIC ssize_t
+xfs_sysfs_object_show(
+	struct kobject		*kobject,
+	struct attribute	*attr,
+	char			*buf)
+{
+	struct xfs_sysfs_attr *xfs_attr = to_attr(attr);
+
+	return xfs_attr->show ? xfs_attr->show(kobject, buf) : 0;
+}
+
+STATIC ssize_t
+xfs_sysfs_object_store(
+	struct kobject		*kobject,
+	struct attribute	*attr,
+	const char		*buf,
+	size_t			count)
+{
+	struct xfs_sysfs_attr *xfs_attr = to_attr(attr);
+
+	return xfs_attr->store ? xfs_attr->store(kobject, buf, count) : 0;
+}
+
+static const struct sysfs_ops xfs_sysfs_ops = {
+	.show = xfs_sysfs_object_show,
+	.store = xfs_sysfs_object_store,
+};
+
 /*
- * xfs_mount kobject. This currently has no attributes and thus no need for show
- * and store helpers. The mp kobject serves as the per-mount parent object that
- * is identified by the fsname under sysfs.
+ * xfs_mount kobject. The mp kobject also serves as the per-mount parent object
+ * that is identified by the fsname under sysfs.
  */
 
+static inline struct xfs_mount *
+to_mp(struct kobject *kobject)
+{
+	struct xfs_kobj *kobj = to_kobj(kobject);
+
+	return container_of(kobj, struct xfs_mount, m_kobj);
+}
+
+#ifdef DEBUG
+
+STATIC ssize_t
+fail_writes_store(
+	struct kobject		*kobject,
+	const char		*buf,
+	size_t			count)
+{
+	struct xfs_mount	*mp = to_mp(kobject);
+	int			ret;
+	int			val;
+
+	ret = kstrtoint(buf, 0, &val);
+	if (ret)
+		return ret;
+
+	if (val == 1)
+		mp->m_fail_writes = true;
+	else if (val == 0)
+		mp->m_fail_writes = false;
+	else
+		return -EINVAL;
+
+	return count;
+}
+
+STATIC ssize_t
+fail_writes_show(
+	struct kobject		*kobject,
+	char			*buf)
+{
+	struct xfs_mount	*mp = to_mp(kobject);
+
+	return snprintf(buf, PAGE_SIZE, "%d\n", mp->m_fail_writes ? 1 : 0);
+}
+XFS_SYSFS_ATTR_RW(fail_writes);
+
+#endif /* DEBUG */
+
+static struct attribute *xfs_mp_attrs[] = {
+#ifdef DEBUG
+	ATTR_LIST(fail_writes),
+#endif
+	NULL,
+};
+
 struct kobj_type xfs_mp_ktype = {
 	.release = xfs_sysfs_release,
+	.sysfs_ops = &xfs_sysfs_ops,
+	.default_attrs = xfs_mp_attrs,
 };
 
 #ifdef DEBUG
@@ -56,9 +146,9 @@ struct kobj_type xfs_mp_ktype = {
 
 STATIC ssize_t
 log_recovery_delay_store(
+	struct kobject	*kobject,
 	const char	*buf,
-	size_t		count,
-	void		*data)
+	size_t		count)
 {
 	int		ret;
 	int		val;
@@ -77,8 +167,8 @@ log_recovery_delay_store(
 
 STATIC ssize_t
 log_recovery_delay_show(
-	char	*buf,
-	void	*data)
+	struct kobject	*kobject,
+	char		*buf)
 {
 	return snprintf(buf, PAGE_SIZE, "%d\n", xfs_globals.log_recovery_delay);
 }
@@ -89,52 +179,87 @@ static struct attribute *xfs_dbg_attrs[] = {
 	NULL,
 };
 
+struct kobj_type xfs_dbg_ktype = {
+	.release = xfs_sysfs_release,
+	.sysfs_ops = &xfs_sysfs_ops,
+	.default_attrs = xfs_dbg_attrs,
+};
+
+#endif /* DEBUG */
+
+/* stats */
+
+static inline struct xstats *
+to_xstats(struct kobject *kobject)
+{
+	struct xfs_kobj *kobj = to_kobj(kobject);
+
+	return container_of(kobj, struct xstats, xs_kobj);
+}
+
 STATIC ssize_t
-xfs_dbg_show(
-	struct kobject		*kobject,
-	struct attribute	*attr,
-	char			*buf)
+stats_show(
+	struct kobject	*kobject,
+	char		*buf)
 {
-	struct xfs_sysfs_attr *xfs_attr = to_attr(attr);
+	struct xstats	*stats = to_xstats(kobject);
 
-	return xfs_attr->show ? xfs_attr->show(buf, NULL) : 0;
+	return xfs_stats_format(stats->xs_stats, buf);
 }
+XFS_SYSFS_ATTR_RO(stats);
 
 STATIC ssize_t
-xfs_dbg_store(
-	struct kobject		*kobject,
-	struct attribute	*attr,
-	const char		*buf,
-	size_t			count)
+stats_clear_store(
+	struct kobject	*kobject,
+	const char	*buf,
+	size_t		count)
 {
-	struct xfs_sysfs_attr *xfs_attr = to_attr(attr);
+	int		ret;
+	int		val;
+	struct xstats	*stats = to_xstats(kobject);
 
-	return xfs_attr->store ? xfs_attr->store(buf, count, NULL) : 0;
+	ret = kstrtoint(buf, 0, &val);
+	if (ret)
+		return ret;
+
+	if (val != 1)
+		return -EINVAL;
+
+	xfs_stats_clearall(stats->xs_stats);
+	return count;
 }
+XFS_SYSFS_ATTR_WO(stats_clear);
 
-static struct sysfs_ops xfs_dbg_ops = {
-	.show = xfs_dbg_show,
-	.store = xfs_dbg_store,
+static struct attribute *xfs_stats_attrs[] = {
+	ATTR_LIST(stats),
+	ATTR_LIST(stats_clear),
+	NULL,
 };
 
-struct kobj_type xfs_dbg_ktype = {
+struct kobj_type xfs_stats_ktype = {
 	.release = xfs_sysfs_release,
-	.sysfs_ops = &xfs_dbg_ops,
-	.default_attrs = xfs_dbg_attrs,
+	.sysfs_ops = &xfs_sysfs_ops,
+	.default_attrs = xfs_stats_attrs,
 };
 
-#endif /* DEBUG */
-
 /* xlog */
 
+static inline struct xlog *
+to_xlog(struct kobject *kobject)
+{
+	struct xfs_kobj *kobj = to_kobj(kobject);
+
+	return container_of(kobj, struct xlog, l_kobj);
+}
+
 STATIC ssize_t
 log_head_lsn_show(
-	char	*buf,
-	void	*data)
+	struct kobject	*kobject,
+	char		*buf)
 {
-	struct xlog *log = data;
 	int cycle;
 	int block;
+	struct xlog *log = to_xlog(kobject);
 
 	spin_lock(&log->l_icloglock);
 	cycle = log->l_curr_cycle;
@@ -147,12 +272,12 @@ XFS_SYSFS_ATTR_RO(log_head_lsn);
 
 STATIC ssize_t
 log_tail_lsn_show(
-	char	*buf,
-	void	*data)
+	struct kobject	*kobject,
+	char		*buf)
 {
-	struct xlog *log = data;
 	int cycle;
 	int block;
+	struct xlog *log = to_xlog(kobject);
 
 	xlog_crack_atomic_lsn(&log->l_tail_lsn, &cycle, &block);
 	return snprintf(buf, PAGE_SIZE, "%d:%d\n", cycle, block);
@@ -161,12 +286,13 @@ XFS_SYSFS_ATTR_RO(log_tail_lsn);
 
 STATIC ssize_t
 reserve_grant_head_show(
-	char	*buf,
-	void	*data)
+	struct kobject	*kobject,
+	char		*buf)
+
 {
-	struct xlog *log = data;
 	int cycle;
 	int bytes;
+	struct xlog *log = to_xlog(kobject);
 
 	xlog_crack_grant_head(&log->l_reserve_head.grant, &cycle, &bytes);
 	return snprintf(buf, PAGE_SIZE, "%d:%d\n", cycle, bytes);
@@ -175,65 +301,64 @@ XFS_SYSFS_ATTR_RO(reserve_grant_head);
 
 STATIC ssize_t
 write_grant_head_show(
-	char	*buf,
-	void	*data)
+	struct kobject	*kobject,
+	char		*buf)
 {
-	struct xlog *log = data;
 	int cycle;
 	int bytes;
+	struct xlog *log = to_xlog(kobject);
 
 	xlog_crack_grant_head(&log->l_write_head.grant, &cycle, &bytes);
 	return snprintf(buf, PAGE_SIZE, "%d:%d\n", cycle, bytes);
 }
 XFS_SYSFS_ATTR_RO(write_grant_head);
 
-static struct attribute *xfs_log_attrs[] = {
-	ATTR_LIST(log_head_lsn),
-	ATTR_LIST(log_tail_lsn),
-	ATTR_LIST(reserve_grant_head),
-	ATTR_LIST(write_grant_head),
-	NULL,
-};
-
-static inline struct xlog *
-to_xlog(struct kobject *kobject)
-{
-	struct xfs_kobj *kobj = to_kobj(kobject);
-	return container_of(kobj, struct xlog, l_kobj);
-}
-
+#ifdef DEBUG
 STATIC ssize_t
-xfs_log_show(
-	struct kobject		*kobject,
-	struct attribute	*attr,
-	char			*buf)
+log_badcrc_factor_store(
+	struct kobject	*kobject,
+	const char	*buf,
+	size_t		count)
 {
-	struct xlog *log = to_xlog(kobject);
-	struct xfs_sysfs_attr *xfs_attr = to_attr(attr);
+	struct xlog	*log = to_xlog(kobject);
+	int		ret;
+	uint32_t	val;
 
-	return xfs_attr->show ? xfs_attr->show(buf, log) : 0;
+	ret = kstrtouint(buf, 0, &val);
+	if (ret)
+		return ret;
+
+	log->l_badcrc_factor = val;
+
+	return count;
 }
 
 STATIC ssize_t
-xfs_log_store(
-	struct kobject		*kobject,
-	struct attribute	*attr,
-	const char		*buf,
-	size_t			count)
+log_badcrc_factor_show(
+	struct kobject	*kobject,
+	char		*buf)
 {
-	struct xlog *log = to_xlog(kobject);
-	struct xfs_sysfs_attr *xfs_attr = to_attr(attr);
+	struct xlog	*log = to_xlog(kobject);
 
-	return xfs_attr->store ? xfs_attr->store(buf, count, log) : 0;
+	return snprintf(buf, PAGE_SIZE, "%d\n", log->l_badcrc_factor);
 }
 
-static struct sysfs_ops xfs_log_ops = {
-	.show = xfs_log_show,
-	.store = xfs_log_store,
+XFS_SYSFS_ATTR_RW(log_badcrc_factor);
+#endif	/* DEBUG */
+
+static struct attribute *xfs_log_attrs[] = {
+	ATTR_LIST(log_head_lsn),
+	ATTR_LIST(log_tail_lsn),
+	ATTR_LIST(reserve_grant_head),
+	ATTR_LIST(write_grant_head),
+#ifdef DEBUG
+	ATTR_LIST(log_badcrc_factor),
+#endif
+	NULL,
 };
 
 struct kobj_type xfs_log_ktype = {
 	.release = xfs_sysfs_release,
-	.sysfs_ops = &xfs_log_ops,
+	.sysfs_ops = &xfs_sysfs_ops,
 	.default_attrs = xfs_log_attrs,
 };
diff --git a/fs/xfs/xfs_sysfs.h b/fs/xfs/xfs_sysfs.h
index 240eee35f342..be692e59938d 100644
--- a/fs/xfs/xfs_sysfs.h
+++ b/fs/xfs/xfs_sysfs.h
@@ -22,6 +22,7 @@
 extern struct kobj_type xfs_mp_ktype;	/* xfs_mount */
 extern struct kobj_type xfs_dbg_ktype;	/* debug */
 extern struct kobj_type xfs_log_ktype;	/* xlog */
+extern struct kobj_type xfs_stats_ktype;	/* stats */
 
 static inline struct xfs_kobj *
 to_kobj(struct kobject *kobject)
diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
index 5ed36b1e04c1..c8d58426008e 100644
--- a/fs/xfs/xfs_trace.h
+++ b/fs/xfs/xfs_trace.h
@@ -689,6 +689,7 @@ DEFINE_INODE_EVENT(xfs_inode_free_eofblocks_invalid);
 DEFINE_INODE_EVENT(xfs_filemap_fault);
 DEFINE_INODE_EVENT(xfs_filemap_pmd_fault);
 DEFINE_INODE_EVENT(xfs_filemap_page_mkwrite);
+DEFINE_INODE_EVENT(xfs_filemap_pfn_mkwrite);
 
 DECLARE_EVENT_CLASS(xfs_iref_class,
 	TP_PROTO(struct xfs_inode *ip, unsigned long caller_ip),
@@ -1221,6 +1222,32 @@ DEFINE_PAGE_EVENT(xfs_writepage);
 DEFINE_PAGE_EVENT(xfs_releasepage);
 DEFINE_PAGE_EVENT(xfs_invalidatepage);
 
+DECLARE_EVENT_CLASS(xfs_readpage_class,
+	TP_PROTO(struct inode *inode, int nr_pages),
+	TP_ARGS(inode, nr_pages),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(int, nr_pages)
+	),
+	TP_fast_assign(
+		__entry->dev = inode->i_sb->s_dev;
+		__entry->ino = inode->i_ino;
+		__entry->nr_pages = nr_pages;
+	),
+	TP_printk("dev %d:%d ino 0x%llx nr_pages %d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __entry->nr_pages)
+)
+
+#define DEFINE_READPAGE_EVENT(name)		\
+DEFINE_EVENT(xfs_readpage_class, name,	\
+	TP_PROTO(struct inode *inode, int nr_pages), \
+	TP_ARGS(inode, nr_pages))
+DEFINE_READPAGE_EVENT(xfs_vm_readpage);
+DEFINE_READPAGE_EVENT(xfs_vm_readpages);
+
 DECLARE_EVENT_CLASS(xfs_imap_class,
 	TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count,
 		 int type, struct xfs_bmbt_irec *irec),
@@ -1269,11 +1296,7 @@ DEFINE_IOMAP_EVENT(xfs_map_blocks_found);
 DEFINE_IOMAP_EVENT(xfs_map_blocks_alloc);
 DEFINE_IOMAP_EVENT(xfs_get_blocks_found);
 DEFINE_IOMAP_EVENT(xfs_get_blocks_alloc);
-DEFINE_IOMAP_EVENT(xfs_gbmap_direct);
-DEFINE_IOMAP_EVENT(xfs_gbmap_direct_new);
-DEFINE_IOMAP_EVENT(xfs_gbmap_direct_update);
-DEFINE_IOMAP_EVENT(xfs_gbmap_direct_none);
-DEFINE_IOMAP_EVENT(xfs_gbmap_direct_endio);
+DEFINE_IOMAP_EVENT(xfs_get_blocks_map_direct);
 
 DECLARE_EVENT_CLASS(xfs_simple_io_class,
 	TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count),
@@ -1312,6 +1335,10 @@ DEFINE_SIMPLE_IO_EVENT(xfs_delalloc_enospc);
 DEFINE_SIMPLE_IO_EVENT(xfs_unwritten_convert);
 DEFINE_SIMPLE_IO_EVENT(xfs_get_blocks_notfound);
 DEFINE_SIMPLE_IO_EVENT(xfs_setfilesize);
+DEFINE_SIMPLE_IO_EVENT(xfs_zero_eof);
+DEFINE_SIMPLE_IO_EVENT(xfs_end_io_direct_write);
+DEFINE_SIMPLE_IO_EVENT(xfs_end_io_direct_write_unwritten);
+DEFINE_SIMPLE_IO_EVENT(xfs_end_io_direct_write_append);
 
 DECLARE_EVENT_CLASS(xfs_itrunc_class,
 	TP_PROTO(struct xfs_inode *ip, xfs_fsize_t new_size),
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index a0ab1dae9c31..20c53666cb4b 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -930,9 +930,9 @@ __xfs_trans_commit(
 	 */
 	if (sync) {
 		error = _xfs_log_force_lsn(mp, commit_lsn, XFS_LOG_SYNC, NULL);
-		XFS_STATS_INC(xs_trans_sync);
+		XFS_STATS_INC(mp, xs_trans_sync);
 	} else {
-		XFS_STATS_INC(xs_trans_async);
+		XFS_STATS_INC(mp, xs_trans_async);
 	}
 
 	return error;
@@ -955,7 +955,7 @@ out_unreserve:
 	xfs_trans_free_items(tp, NULLCOMMITLSN, !!error);
 	xfs_trans_free(tp);
 
-	XFS_STATS_INC(xs_trans_empty);
+	XFS_STATS_INC(mp, xs_trans_empty);
 	return error;
 }
 
@@ -1028,6 +1028,8 @@ __xfs_trans_roll(
 	struct xfs_trans_res	tres;
 	int			error;
 
+	*committed = 0;
+
 	/*
 	 * Ensure that the inode is always logged.
 	 */
@@ -1082,6 +1084,6 @@ xfs_trans_roll(
 	struct xfs_trans	**tpp,
 	struct xfs_inode	*dp)
 {
-	int			committed = 0;
+	int			committed;
 	return __xfs_trans_roll(tpp, dp, &committed);
 }
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index 4643070d7cae..e7c49cf43fbc 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -133,7 +133,6 @@ typedef struct xfs_trans {
  * XFS transaction mechanism exported interfaces that are
  * actually macros.
  */
-#define	xfs_trans_get_block_res(tp)	((tp)->t_blk_res)
 #define	xfs_trans_set_sync(tp)		((tp)->t_flags |= XFS_TRANS_SYNC)
 
 #if defined(DEBUG) || defined(XFS_WARN)
diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c
index 1098cf490189..d6c9c3e9e02b 100644
--- a/fs/xfs/xfs_trans_ail.c
+++ b/fs/xfs/xfs_trans_ail.c
@@ -349,7 +349,7 @@ xfsaild_push(
 	     xfs_ail_min_lsn(ailp))) {
 		ailp->xa_log_flush = 0;
 
-		XFS_STATS_INC(xs_push_ail_flush);
+		XFS_STATS_INC(mp, xs_push_ail_flush);
 		xfs_log_force(mp, XFS_LOG_SYNC);
 	}
 
@@ -371,7 +371,7 @@ xfsaild_push(
 		goto out_done;
 	}
 
-	XFS_STATS_INC(xs_push_ail);
+	XFS_STATS_INC(mp, xs_push_ail);
 
 	lsn = lip->li_lsn;
 	while ((XFS_LSN_CMP(lip->li_lsn, target) <= 0)) {
@@ -385,7 +385,7 @@ xfsaild_push(
 		lock_result = lip->li_ops->iop_push(lip, &ailp->xa_buf_list);
 		switch (lock_result) {
 		case XFS_ITEM_SUCCESS:
-			XFS_STATS_INC(xs_push_ail_success);
+			XFS_STATS_INC(mp, xs_push_ail_success);
 			trace_xfs_ail_push(lip);
 
 			ailp->xa_last_pushed_lsn = lsn;
@@ -403,7 +403,7 @@ xfsaild_push(
 			 * re-try the flushing relatively soon if most of the
 			 * AIL is beeing flushed.
 			 */
-			XFS_STATS_INC(xs_push_ail_flushing);
+			XFS_STATS_INC(mp, xs_push_ail_flushing);
 			trace_xfs_ail_flushing(lip);
 
 			flushing++;
@@ -411,14 +411,14 @@ xfsaild_push(
 			break;
 
 		case XFS_ITEM_PINNED:
-			XFS_STATS_INC(xs_push_ail_pinned);
+			XFS_STATS_INC(mp, xs_push_ail_pinned);
 			trace_xfs_ail_pinned(lip);
 
 			stuck++;
 			ailp->xa_log_flush++;
 			break;
 		case XFS_ITEM_LOCKED:
-			XFS_STATS_INC(xs_push_ail_locked);
+			XFS_STATS_INC(mp, xs_push_ail_locked);
 			trace_xfs_ail_locked(lip);
 
 			stuck++;
@@ -497,6 +497,7 @@ xfsaild(
 	long		tout = 0;	/* milliseconds */
 
 	current->flags |= PF_MEMALLOC;
+	set_freezable();
 
 	while (!kthread_should_stop()) {
 		if (tout && tout <= 20)
@@ -519,14 +520,14 @@ xfsaild(
 		if (!xfs_ail_min(ailp) &&
 		    ailp->xa_target == ailp->xa_target_prev) {
 			spin_unlock(&ailp->xa_lock);
-			schedule();
+			freezable_schedule();
 			tout = 0;
 			continue;
 		}
 		spin_unlock(&ailp->xa_lock);
 
 		if (tout)
-			schedule_timeout(msecs_to_jiffies(tout));
+			freezable_schedule_timeout(msecs_to_jiffies(tout));
 
 		__set_current_state(TASK_RUNNING);
 
diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c
index 75798412859a..8ee29ca132dc 100644
--- a/fs/xfs/xfs_trans_buf.c
+++ b/fs/xfs/xfs_trans_buf.c
@@ -155,7 +155,7 @@ xfs_trans_get_buf_map(
 		ASSERT(xfs_buf_islocked(bp));
 		if (XFS_FORCED_SHUTDOWN(tp->t_mountp)) {
 			xfs_buf_stale(bp);
-			XFS_BUF_DONE(bp);
+			bp->b_flags |= XBF_DONE;
 		}
 
 		ASSERT(bp->b_transp == tp);
@@ -518,7 +518,7 @@ xfs_trans_log_buf(xfs_trans_t	*tp,
 	 * inside the b_bdstrat callback so that this won't get written to
 	 * disk.
 	 */
-	XFS_BUF_DONE(bp);
+	bp->b_flags |= XBF_DONE;
 
 	ASSERT(atomic_read(&bip->bli_refcount) > 0);
 	bp->b_iodone = xfs_buf_iodone_callbacks;
@@ -534,8 +534,8 @@ xfs_trans_log_buf(xfs_trans_t	*tp,
 	 */
 	if (bip->bli_flags & XFS_BLI_STALE) {
 		bip->bli_flags &= ~XFS_BLI_STALE;
-		ASSERT(XFS_BUF_ISSTALE(bp));
-		XFS_BUF_UNSTALE(bp);
+		ASSERT(bp->b_flags & XBF_STALE);
+		bp->b_flags &= ~XBF_STALE;
 		bip->__bli_format.blf_flags &= ~XFS_BLF_CANCEL;
 	}
 
@@ -600,7 +600,7 @@ xfs_trans_binval(
 		 * If the buffer is already invalidated, then
 		 * just return.
 		 */
-		ASSERT(XFS_BUF_ISSTALE(bp));
+		ASSERT(bp->b_flags & XBF_STALE);
 		ASSERT(!(bip->bli_flags & (XFS_BLI_LOGGED | XFS_BLI_DIRTY)));
 		ASSERT(!(bip->__bli_format.blf_flags & XFS_BLF_INODE_BUF));
 		ASSERT(!(bip->__bli_format.blf_flags & XFS_BLFT_MASK));
diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c
index ce78534a047e..c3d547211d16 100644
--- a/fs/xfs/xfs_trans_dquot.c
+++ b/fs/xfs/xfs_trans_dquot.c
@@ -572,12 +572,16 @@ xfs_quota_warn(
 	struct xfs_dquot	*dqp,
 	int			type)
 {
-	/* no warnings for project quotas - we just return ENOSPC later */
+	enum quota_type qtype;
+
 	if (dqp->dq_flags & XFS_DQ_PROJ)
-		return;
-	quota_send_warning(make_kqid(&init_user_ns,
-				     (dqp->dq_flags & XFS_DQ_USER) ?
-				     USRQUOTA : GRPQUOTA,
+		qtype = PRJQUOTA;
+	else if (dqp->dq_flags & XFS_DQ_USER)
+		qtype = USRQUOTA;
+	else
+		qtype = GRPQUOTA;
+
+	quota_send_warning(make_kqid(&init_user_ns, qtype,
 				     be32_to_cpu(dqp->q_core.d_id)),
 			   mp->m_super->s_dev, type);
 }
@@ -605,17 +609,20 @@ xfs_trans_dqresv(
 	xfs_qcnt_t	total_count;
 	xfs_qcnt_t	*resbcountp;
 	xfs_quotainfo_t	*q = mp->m_quotainfo;
+	struct xfs_def_quota	*defq;
 
 
 	xfs_dqlock(dqp);
 
+	defq = xfs_get_defquota(dqp, q);
+
 	if (flags & XFS_TRANS_DQ_RES_BLKS) {
 		hardlimit = be64_to_cpu(dqp->q_core.d_blk_hardlimit);
 		if (!hardlimit)
-			hardlimit = q->qi_bhardlimit;
+			hardlimit = defq->bhardlimit;
 		softlimit = be64_to_cpu(dqp->q_core.d_blk_softlimit);
 		if (!softlimit)
-			softlimit = q->qi_bsoftlimit;
+			softlimit = defq->bsoftlimit;
 		timer = be32_to_cpu(dqp->q_core.d_btimer);
 		warns = be16_to_cpu(dqp->q_core.d_bwarns);
 		warnlimit = dqp->q_mount->m_quotainfo->qi_bwarnlimit;
@@ -624,10 +631,10 @@ xfs_trans_dqresv(
 		ASSERT(flags & XFS_TRANS_DQ_RES_RTBLKS);
 		hardlimit = be64_to_cpu(dqp->q_core.d_rtb_hardlimit);
 		if (!hardlimit)
-			hardlimit = q->qi_rtbhardlimit;
+			hardlimit = defq->rtbhardlimit;
 		softlimit = be64_to_cpu(dqp->q_core.d_rtb_softlimit);
 		if (!softlimit)
-			softlimit = q->qi_rtbsoftlimit;
+			softlimit = defq->rtbsoftlimit;
 		timer = be32_to_cpu(dqp->q_core.d_rtbtimer);
 		warns = be16_to_cpu(dqp->q_core.d_rtbwarns);
 		warnlimit = dqp->q_mount->m_quotainfo->qi_rtbwarnlimit;
@@ -668,10 +675,10 @@ xfs_trans_dqresv(
 			warnlimit = dqp->q_mount->m_quotainfo->qi_iwarnlimit;
 			hardlimit = be64_to_cpu(dqp->q_core.d_ino_hardlimit);
 			if (!hardlimit)
-				hardlimit = q->qi_ihardlimit;
+				hardlimit = defq->ihardlimit;
 			softlimit = be64_to_cpu(dqp->q_core.d_ino_softlimit);
 			if (!softlimit)
-				softlimit = q->qi_isoftlimit;
+				softlimit = defq->isoftlimit;
 
 			if (hardlimit && total_count > hardlimit) {
 				xfs_quota_warn(mp, dqp, QUOTA_NL_IHARDWARN);
diff --git a/fs/xfs/xfs_trans_inode.c b/fs/xfs/xfs_trans_inode.c
index 17280cd71934..11a3af08b5c7 100644
--- a/fs/xfs/xfs_trans_inode.c
+++ b/fs/xfs/xfs_trans_inode.c
@@ -75,18 +75,10 @@ xfs_trans_ichgtime(
 
 	tv = current_fs_time(inode->i_sb);
 
-	if ((flags & XFS_ICHGTIME_MOD) &&
-	    !timespec_equal(&inode->i_mtime, &tv)) {
+	if (flags & XFS_ICHGTIME_MOD)
 		inode->i_mtime = tv;
-		ip->i_d.di_mtime.t_sec = tv.tv_sec;
-		ip->i_d.di_mtime.t_nsec = tv.tv_nsec;
-	}
-	if ((flags & XFS_ICHGTIME_CHG) &&
-	    !timespec_equal(&inode->i_ctime, &tv)) {
+	if (flags & XFS_ICHGTIME_CHG)
 		inode->i_ctime = tv;
-		ip->i_d.di_ctime.t_sec = tv.tv_sec;
-		ip->i_d.di_ctime.t_nsec = tv.tv_nsec;
-	}
 }
 
 /*
@@ -108,6 +100,15 @@ xfs_trans_log_inode(
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
 	/*
+	 * Record the specific change for fdatasync optimisation. This
+	 * allows fdatasync to skip log forces for inodes that are only
+	 * timestamp dirty. We do this before the change count so that
+	 * the core being logged in this case does not impact on fdatasync
+	 * behaviour.
+	 */
+	ip->i_itemp->ili_fsync_fields |= flags;
+
+	/*
 	 * First time we log the inode in a transaction, bump the inode change
 	 * counter if it is configured for this to occur. We don't use
 	 * inode_inc_version() because there is no need for extra locking around
@@ -116,7 +117,7 @@ xfs_trans_log_inode(
 	 */
 	if (!(ip->i_itemp->ili_item.li_desc->lid_flags & XFS_LID_DIRTY) &&
 	    IS_I_VERSION(VFS_I(ip))) {
-		ip->i_d.di_changecount = ++VFS_I(ip)->i_version;
+		VFS_I(ip)->i_version++;
 		flags |= XFS_ILOG_CORE;
 	}
 
diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c
index c036815183cb..110f1d7d86b0 100644
--- a/fs/xfs/xfs_xattr.c
+++ b/fs/xfs/xfs_xattr.c
@@ -32,15 +32,13 @@
 
 
 static int
-xfs_xattr_get(struct dentry *dentry, const char *name,
-		void *value, size_t size, int xflags)
+xfs_xattr_get(const struct xattr_handler *handler, struct dentry *dentry,
+		const char *name, void *value, size_t size)
 {
+	int xflags = handler->flags;
 	struct xfs_inode *ip = XFS_I(d_inode(dentry));
 	int error, asize = size;
 
-	if (strcmp(name, "") == 0)
-		return -EINVAL;
-
 	/* Convert Linux syscall to XFS internal ATTR flags */
 	if (!size) {
 		xflags |= ATTR_KERNOVAL;
@@ -53,14 +51,35 @@ xfs_xattr_get(struct dentry *dentry, const char *name,
 	return asize;
 }
 
-static int
-xfs_xattr_set(struct dentry *dentry, const char *name, const void *value,
-		size_t size, int flags, int xflags)
+void
+xfs_forget_acl(
+	struct inode		*inode,
+	const char		*name,
+	int			xflags)
 {
-	struct xfs_inode *ip = XFS_I(d_inode(dentry));
+	/*
+	 * Invalidate any cached ACLs if the user has bypassed the ACL
+	 * interface. We don't validate the content whatsoever so it is caller
+	 * responsibility to provide data in valid format and ensure i_mode is
+	 * consistent.
+	 */
+	if (xflags & ATTR_ROOT) {
+#ifdef CONFIG_XFS_POSIX_ACL
+		if (!strcmp(name, SGI_ACL_FILE))
+			forget_cached_acl(inode, ACL_TYPE_ACCESS);
+		else if (!strcmp(name, SGI_ACL_DEFAULT))
+			forget_cached_acl(inode, ACL_TYPE_DEFAULT);
+#endif
+	}
+}
 
-	if (strcmp(name, "") == 0)
-		return -EINVAL;
+static int
+xfs_xattr_set(const struct xattr_handler *handler, struct dentry *dentry,
+		const char *name, const void *value, size_t size, int flags)
+{
+	int			xflags = handler->flags;
+	struct xfs_inode	*ip = XFS_I(d_inode(dentry));
+	int			error;
 
 	/* Convert Linux syscall to XFS internal ATTR flags */
 	if (flags & XATTR_CREATE)
@@ -70,8 +89,12 @@ xfs_xattr_set(struct dentry *dentry, const char *name, const void *value,
 
 	if (!value)
 		return xfs_attr_remove(ip, (unsigned char *)name, xflags);
-	return xfs_attr_set(ip, (unsigned char *)name,
+	error = xfs_attr_set(ip, (unsigned char *)name,
 				(void *)value, size, xflags);
+	if (!error)
+		xfs_forget_acl(d_inode(dentry), name, xflags);
+
+	return error;
 }
 
 static const struct xattr_handler xfs_xattr_user_handler = {
@@ -106,47 +129,19 @@ const struct xattr_handler *xfs_xattr_handlers[] = {
 	NULL
 };
 
-static unsigned int xfs_xattr_prefix_len(int flags)
-{
-	if (flags & XFS_ATTR_SECURE)
-		return sizeof("security");
-	else if (flags & XFS_ATTR_ROOT)
-		return sizeof("trusted");
-	else
-		return sizeof("user");
-}
-
-static const char *xfs_xattr_prefix(int flags)
-{
-	if (flags & XFS_ATTR_SECURE)
-		return xfs_xattr_security_handler.prefix;
-	else if (flags & XFS_ATTR_ROOT)
-		return xfs_xattr_trusted_handler.prefix;
-	else
-		return xfs_xattr_user_handler.prefix;
-}
-
 static int
-xfs_xattr_put_listent(
+__xfs_xattr_put_listent(
 	struct xfs_attr_list_context *context,
-	int		flags,
-	unsigned char	*name,
-	int		namelen,
-	int		valuelen,
-	unsigned char	*value)
+	char *prefix,
+	int prefix_len,
+	unsigned char *name,
+	int namelen)
 {
-	unsigned int prefix_len = xfs_xattr_prefix_len(flags);
 	char *offset;
 	int arraytop;
 
-	ASSERT(context->count >= 0);
-
-	/*
-	 * Only show root namespace entries if we are actually allowed to
-	 * see them.
-	 */
-	if ((flags & XFS_ATTR_ROOT) && !capable(CAP_SYS_ADMIN))
-		return 0;
+	if (!context->alist)
+		goto compute_size;
 
 	arraytop = context->count + prefix_len + namelen + 1;
 	if (arraytop > context->firstu) {
@@ -154,17 +149,19 @@ xfs_xattr_put_listent(
 		return 1;
 	}
 	offset = (char *)context->alist + context->count;
-	strncpy(offset, xfs_xattr_prefix(flags), prefix_len);
+	strncpy(offset, prefix, prefix_len);
 	offset += prefix_len;
 	strncpy(offset, (char *)name, namelen);			/* real name */
 	offset += namelen;
 	*offset = '\0';
+
+compute_size:
 	context->count += prefix_len + namelen + 1;
 	return 0;
 }
 
 static int
-xfs_xattr_put_listent_sizes(
+xfs_xattr_put_listent(
 	struct xfs_attr_list_context *context,
 	int		flags,
 	unsigned char	*name,
@@ -172,24 +169,55 @@ xfs_xattr_put_listent_sizes(
 	int		valuelen,
 	unsigned char	*value)
 {
-	context->count += xfs_xattr_prefix_len(flags) + namelen + 1;
-	return 0;
-}
+	char *prefix;
+	int prefix_len;
 
-static int
-list_one_attr(const char *name, const size_t len, void *data,
-		size_t size, ssize_t *result)
-{
-	char *p = data + *result;
+	ASSERT(context->count >= 0);
 
-	*result += len;
-	if (!size)
-		return 0;
-	if (*result > size)
-		return -ERANGE;
+	if (flags & XFS_ATTR_ROOT) {
+#ifdef CONFIG_XFS_POSIX_ACL
+		if (namelen == SGI_ACL_FILE_SIZE &&
+		    strncmp(name, SGI_ACL_FILE,
+			    SGI_ACL_FILE_SIZE) == 0) {
+			int ret = __xfs_xattr_put_listent(
+					context, XATTR_SYSTEM_PREFIX,
+					XATTR_SYSTEM_PREFIX_LEN,
+					XATTR_POSIX_ACL_ACCESS,
+					strlen(XATTR_POSIX_ACL_ACCESS));
+			if (ret)
+				return ret;
+		} else if (namelen == SGI_ACL_DEFAULT_SIZE &&
+			 strncmp(name, SGI_ACL_DEFAULT,
+				 SGI_ACL_DEFAULT_SIZE) == 0) {
+			int ret = __xfs_xattr_put_listent(
+					context, XATTR_SYSTEM_PREFIX,
+					XATTR_SYSTEM_PREFIX_LEN,
+					XATTR_POSIX_ACL_DEFAULT,
+					strlen(XATTR_POSIX_ACL_DEFAULT));
+			if (ret)
+				return ret;
+		}
+#endif
 
-	strcpy(p, name);
-	return 0;
+		/*
+		 * Only show root namespace entries if we are actually allowed to
+		 * see them.
+		 */
+		if (!capable(CAP_SYS_ADMIN))
+			return 0;
+
+		prefix = XATTR_TRUSTED_PREFIX;
+		prefix_len = XATTR_TRUSTED_PREFIX_LEN;
+	} else if (flags & XFS_ATTR_SECURE) {
+		prefix = XATTR_SECURITY_PREFIX;
+		prefix_len = XATTR_SECURITY_PREFIX_LEN;
+	} else {
+		prefix = XATTR_USER_PREFIX;
+		prefix_len = XATTR_USER_PREFIX_LEN;
+	}
+
+	return __xfs_xattr_put_listent(context, prefix, prefix_len, name,
+				       namelen);
 }
 
 ssize_t
@@ -198,7 +226,6 @@ xfs_vn_listxattr(struct dentry *dentry, char *data, size_t size)
 	struct xfs_attr_list_context context;
 	struct attrlist_cursor_kern cursor = { 0 };
 	struct inode		*inode = d_inode(dentry);
-	int			error;
 
 	/*
 	 * First read the regular on-disk attributes.
@@ -207,37 +234,14 @@ xfs_vn_listxattr(struct dentry *dentry, char *data, size_t size)
 	context.dp = XFS_I(inode);
 	context.cursor = &cursor;
 	context.resynch = 1;
-	context.alist = data;
+	context.alist = size ? data : NULL;
 	context.bufsize = size;
 	context.firstu = context.bufsize;
-
-	if (size)
-		context.put_listent = xfs_xattr_put_listent;
-	else
-		context.put_listent = xfs_xattr_put_listent_sizes;
+	context.put_listent = xfs_xattr_put_listent;
 
 	xfs_attr_list_int(&context);
 	if (context.count < 0)
 		return -ERANGE;
 
-	/*
-	 * Then add the two synthetic ACL attributes.
-	 */
-	if (posix_acl_access_exists(inode)) {
-		error = list_one_attr(POSIX_ACL_XATTR_ACCESS,
-				strlen(POSIX_ACL_XATTR_ACCESS) + 1,
-				data, size, &context.count);
-		if (error)
-			return error;
-	}
-
-	if (posix_acl_default_exists(inode)) {
-		error = list_one_attr(POSIX_ACL_XATTR_DEFAULT,
-				strlen(POSIX_ACL_XATTR_DEFAULT) + 1,
-				data, size, &context.count);
-		if (error)
-			return error;
-	}
-
 	return context.count;
 }