133 files changed, 7101 insertions, 6284 deletions

diff --git a/fs/xfs/Kconfig b/fs/xfs/Kconfig
index 399e8cec6e60..5d47b4df61ea 100644
--- a/fs/xfs/Kconfig
+++ b/fs/xfs/Kconfig
@@ -1,6 +1,7 @@
 config XFS_FS
 	tristate "XFS filesystem support"
 	depends on BLOCK
+	depends on (64BIT || LBDAF)
 	select EXPORTFS
 	select LIBCRC32C
 	help
diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index c21f43506661..d61799949580 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -17,6 +17,7 @@
 #
 
 ccflags-y += -I$(src)			# needed for trace events
+ccflags-y += -I$(src)/libxfs
 
 ccflags-$(CONFIG_XFS_DEBUG) += -g
 
@@ -25,6 +26,39 @@ obj-$(CONFIG_XFS_FS)		+= xfs.o
 # this one should be compiled first, as the tracing macros can easily blow up
 xfs-y				+= xfs_trace.o
 
+# build the libxfs code first
+xfs-y				+= $(addprefix libxfs/, \
+				   xfs_alloc.o \
+				   xfs_alloc_btree.o \
+				   xfs_attr.o \
+				   xfs_attr_leaf.o \
+				   xfs_attr_remote.o \
+				   xfs_bmap.o \
+				   xfs_bmap_btree.o \
+				   xfs_btree.o \
+				   xfs_da_btree.o \
+				   xfs_da_format.o \
+				   xfs_dir2.o \
+				   xfs_dir2_block.o \
+				   xfs_dir2_data.o \
+				   xfs_dir2_leaf.o \
+				   xfs_dir2_node.o \
+				   xfs_dir2_sf.o \
+				   xfs_dquot_buf.o \
+				   xfs_ialloc.o \
+				   xfs_ialloc_btree.o \
+				   xfs_inode_fork.o \
+				   xfs_inode_buf.o \
+				   xfs_log_rlimit.o \
+				   xfs_sb.o \
+				   xfs_symlink_remote.o \
+				   xfs_trans_resv.o \
+				   )
+# xfs_rtbitmap is shared with libxfs
+xfs-$(CONFIG_XFS_RT)		+= $(addprefix libxfs/, \
+				   xfs_rtbitmap.o \
+				   )
+
 # highlevel code
 xfs-y				+= xfs_aops.o \
 				   xfs_attr_inactive.o \
@@ -45,53 +79,27 @@ xfs-y				+= xfs_aops.o \
 				   xfs_ioctl.o \
 				   xfs_iomap.o \
 				   xfs_iops.o \
+				   xfs_inode.o \
 				   xfs_itable.o \
 				   xfs_message.o \
 				   xfs_mount.o \
 				   xfs_mru_cache.o \
 				   xfs_super.o \
 				   xfs_symlink.o \
+				   xfs_sysfs.o \
 				   xfs_trans.o \
 				   xfs_xattr.o \
 				   kmem.o \
 				   uuid.o
 
-# code shared with libxfs
-xfs-y				+= xfs_alloc.o \
-				   xfs_alloc_btree.o \
-				   xfs_attr.o \
-				   xfs_attr_leaf.o \
-				   xfs_attr_remote.o \
-				   xfs_bmap.o \
-				   xfs_bmap_btree.o \
-				   xfs_btree.o \
-				   xfs_da_btree.o \
-				   xfs_da_format.o \
-				   xfs_dir2.o \
-				   xfs_dir2_block.o \
-				   xfs_dir2_data.o \
-				   xfs_dir2_leaf.o \
-				   xfs_dir2_node.o \
-				   xfs_dir2_sf.o \
-				   xfs_dquot_buf.o \
-				   xfs_ialloc.o \
-				   xfs_ialloc_btree.o \
-				   xfs_icreate_item.o \
-				   xfs_inode.o \
-				   xfs_inode_fork.o \
-				   xfs_inode_buf.o \
-				   xfs_log_recover.o \
-				   xfs_log_rlimit.o \
-				   xfs_sb.o \
-				   xfs_symlink_remote.o \
-				   xfs_trans_resv.o
-
 # low-level transaction/log code
 xfs-y				+= xfs_log.o \
 				   xfs_log_cil.o \
 				   xfs_buf_item.o \
 				   xfs_extfree_item.o \
+				   xfs_icreate_item.o \
 				   xfs_inode_item.o \
+				   xfs_log_recover.o \
 				   xfs_trans_ail.o \
 				   xfs_trans_buf.o \
 				   xfs_trans_extfree.o \
@@ -107,8 +115,7 @@ xfs-$(CONFIG_XFS_QUOTA)		+= xfs_dquot.o \
 				   xfs_quotaops.o
 
 # xfs_rtbitmap is shared with libxfs
-xfs-$(CONFIG_XFS_RT)		+= xfs_rtalloc.o \
-				   xfs_rtbitmap.o
+xfs-$(CONFIG_XFS_RT)		+= xfs_rtalloc.o
 
 xfs-$(CONFIG_XFS_POSIX_ACL)	+= xfs_acl.o
 xfs-$(CONFIG_PROC_FS)		+= xfs_stats.o
diff --git a/fs/xfs/kmem.c b/fs/xfs/kmem.c
index 66a36befc5c0..844e288b9576 100644
--- a/fs/xfs/kmem.c
+++ b/fs/xfs/kmem.c
@@ -65,12 +65,31 @@ kmem_alloc(size_t size, xfs_km_flags_t flags)
 void *
 kmem_zalloc_large(size_t size, xfs_km_flags_t flags)
 {
+	unsigned noio_flag = 0;
 	void	*ptr;
+	gfp_t	lflags;
 
 	ptr = kmem_zalloc(size, flags | KM_MAYFAIL);
 	if (ptr)
 		return ptr;
-	return vzalloc(size);
+
+	/*
+	 * __vmalloc() will allocate data pages and auxillary structures (e.g.
+	 * pagetables) with GFP_KERNEL, yet we may be under GFP_NOFS context
+	 * here. Hence we need to tell memory reclaim that we are in such a
+	 * context via PF_MEMALLOC_NOIO to prevent memory reclaim re-entering
+	 * the filesystem here and potentially deadlocking.
+	 */
+	if ((current->flags & PF_FSTRANS) || (flags & KM_NOFS))
+		noio_flag = memalloc_noio_save();
+
+	lflags = kmem_flags_convert(flags);
+	ptr = __vmalloc(size, lflags | __GFP_HIGHMEM | __GFP_ZERO, PAGE_KERNEL);
+
+	if ((current->flags & PF_FSTRANS) || (flags & KM_NOFS))
+		memalloc_noio_restore(noio_flag);
+
+	return ptr;
 }
 
 void
diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/libxfs/xfs_ag.h
index 3fc109819c34..6e247a99f5db 100644
--- a/fs/xfs/xfs_ag.h
+++ b/fs/xfs/libxfs/xfs_ag.h
@@ -89,6 +89,8 @@ typedef struct xfs_agf {
 	/* structure must be padded to 64 bit alignment */
 } xfs_agf_t;
 
+#define XFS_AGF_CRC_OFF		offsetof(struct xfs_agf, agf_crc)
+
 #define	XFS_AGF_MAGICNUM	0x00000001
 #define	XFS_AGF_VERSIONNUM	0x00000002
 #define	XFS_AGF_SEQNO		0x00000004
@@ -158,28 +160,38 @@ typedef struct xfs_agi {
 	 * still being referenced.
 	 */
 	__be32		agi_unlinked[XFS_AGI_UNLINKED_BUCKETS];
-
+	/*
+	 * This marks the end of logging region 1 and start of logging region 2.
+	 */
 	uuid_t		agi_uuid;	/* uuid of filesystem */
 	__be32		agi_crc;	/* crc of agi sector */
 	__be32		agi_pad32;
 	__be64		agi_lsn;	/* last write sequence */
 
+	__be32		agi_free_root; /* root of the free inode btree */
+	__be32		agi_free_level;/* levels in free inode btree */
+
 	/* structure must be padded to 64 bit alignment */
 } xfs_agi_t;
 
-#define	XFS_AGI_MAGICNUM	0x00000001
-#define	XFS_AGI_VERSIONNUM	0x00000002
-#define	XFS_AGI_SEQNO		0x00000004
-#define	XFS_AGI_LENGTH		0x00000008
-#define	XFS_AGI_COUNT		0x00000010
-#define	XFS_AGI_ROOT		0x00000020
-#define	XFS_AGI_LEVEL		0x00000040
-#define	XFS_AGI_FREECOUNT	0x00000080
-#define	XFS_AGI_NEWINO		0x00000100
-#define	XFS_AGI_DIRINO		0x00000200
-#define	XFS_AGI_UNLINKED	0x00000400
-#define	XFS_AGI_NUM_BITS	11
-#define	XFS_AGI_ALL_BITS	((1 << XFS_AGI_NUM_BITS) - 1)
+#define XFS_AGI_CRC_OFF		offsetof(struct xfs_agi, agi_crc)
+
+#define	XFS_AGI_MAGICNUM	(1 << 0)
+#define	XFS_AGI_VERSIONNUM	(1 << 1)
+#define	XFS_AGI_SEQNO		(1 << 2)
+#define	XFS_AGI_LENGTH		(1 << 3)
+#define	XFS_AGI_COUNT		(1 << 4)
+#define	XFS_AGI_ROOT		(1 << 5)
+#define	XFS_AGI_LEVEL		(1 << 6)
+#define	XFS_AGI_FREECOUNT	(1 << 7)
+#define	XFS_AGI_NEWINO		(1 << 8)
+#define	XFS_AGI_DIRINO		(1 << 9)
+#define	XFS_AGI_UNLINKED	(1 << 10)
+#define	XFS_AGI_NUM_BITS_R1	11	/* end of the 1st agi logging region */
+#define	XFS_AGI_ALL_BITS_R1	((1 << XFS_AGI_NUM_BITS_R1) - 1)
+#define	XFS_AGI_FREE_ROOT	(1 << 11)
+#define	XFS_AGI_FREE_LEVEL	(1 << 12)
+#define	XFS_AGI_NUM_BITS_R2	13
 
 /* disk block (xfs_daddr_t) in the AG */
 #define XFS_AGI_DADDR(mp)	((xfs_daddr_t)(2 << (mp)->m_sectbb_log))
@@ -222,6 +234,8 @@ typedef struct xfs_agfl {
 	__be32		agfl_bno[];	/* actually XFS_AGFL_SIZE(mp) */
 } xfs_agfl_t;
 
+#define XFS_AGFL_CRC_OFF	offsetof(struct xfs_agfl, agfl_crc)
+
 /*
  * tags for inode radix tree
  */
diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 9eab2dfdcbb5..4bffffe038a1 100644
--- a/fs/xfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -257,16 +257,14 @@ xfs_alloc_fix_len(
 	k = rlen % args->prod;
 	if (k == args->mod)
 		return;
-	if (k > args->mod) {
-		if ((int)(rlen = rlen - k - args->mod) < (int)args->minlen)
-			return;
-	} else {
-		if ((int)(rlen = rlen - args->prod - (args->mod - k)) <
-		    (int)args->minlen)
-			return;
-	}
-	ASSERT(rlen >= args->minlen);
-	ASSERT(rlen <= args->maxlen);
+	if (k > args->mod)
+		rlen = rlen - (k - args->mod);
+	else
+		rlen = rlen - args->prod + (args->mod - k);
+	if ((int)rlen < (int)args->minlen)
+		return;
+	ASSERT(rlen >= args->minlen && rlen <= args->maxlen);
+	ASSERT(rlen % args->prod == args->mod);
 	args->len = rlen;
 }
 
@@ -474,7 +472,6 @@ xfs_agfl_read_verify(
 	struct xfs_buf	*bp)
 {
 	struct xfs_mount *mp = bp->b_target->bt_mount;
-	int		agfl_ok = 1;
 
 	/*
 	 * There is no verification of non-crc AGFLs because mkfs does not
@@ -485,15 +482,13 @@ xfs_agfl_read_verify(
 	if (!xfs_sb_version_hascrc(&mp->m_sb))
 		return;
 
-	agfl_ok = xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
-				   offsetof(struct xfs_agfl, agfl_crc));
+	if (!xfs_buf_verify_cksum(bp, XFS_AGFL_CRC_OFF))
+		xfs_buf_ioerror(bp, -EFSBADCRC);
+	else if (!xfs_agfl_verify(bp))
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
-	agfl_ok = agfl_ok && xfs_agfl_verify(bp);
-
-	if (!agfl_ok) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
-	}
+	if (bp->b_error)
+		xfs_verifier_error(bp);
 }
 
 static void
@@ -508,16 +503,15 @@ xfs_agfl_write_verify(
 		return;
 
 	if (!xfs_agfl_verify(bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
 		return;
 	}
 
 	if (bip)
 		XFS_BUF_TO_AGFL(bp)->agfl_lsn = cpu_to_be64(bip->bli_item.li_lsn);
 
-	xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
-			 offsetof(struct xfs_agfl, agfl_crc));
+	xfs_buf_update_cksum(bp, XFS_AGFL_CRC_OFF);
 }
 
 const struct xfs_buf_ops xfs_agfl_buf_ops = {
@@ -545,7 +539,6 @@ xfs_alloc_read_agfl(
 			XFS_FSS_TO_BB(mp, 1), 0, &bp, &xfs_agfl_buf_ops);
 	if (error)
 		return error;
-	ASSERT(!xfs_buf_geterror(bp));
 	xfs_buf_set_ref(bp, XFS_AGFL_REF);
 	*bpp = bp;
 	return 0;
@@ -566,7 +559,7 @@ xfs_alloc_update_counters(
 	xfs_trans_agblocks_delta(tp, len);
 	if (unlikely(be32_to_cpu(agf->agf_freeblks) >
 		     be32_to_cpu(agf->agf_length)))
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 
 	xfs_alloc_log_agf(tp, agbp, XFS_AGF_FREEBLKS);
 	return 0;
@@ -2238,19 +2231,17 @@ xfs_agf_read_verify(
 	struct xfs_buf	*bp)
 {
 	struct xfs_mount *mp = bp->b_target->bt_mount;
-	int		agf_ok = 1;
-
-	if (xfs_sb_version_hascrc(&mp->m_sb))
-		agf_ok = xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
-					  offsetof(struct xfs_agf, agf_crc));
 
-	agf_ok = agf_ok && xfs_agf_verify(mp, bp);
-
-	if (unlikely(XFS_TEST_ERROR(!agf_ok, mp, XFS_ERRTAG_ALLOC_READ_AGF,
-			XFS_RANDOM_ALLOC_READ_AGF))) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
-	}
+	if (xfs_sb_version_hascrc(&mp->m_sb) &&
+	    !xfs_buf_verify_cksum(bp, XFS_AGF_CRC_OFF))
+		xfs_buf_ioerror(bp, -EFSBADCRC);
+	else if (XFS_TEST_ERROR(!xfs_agf_verify(mp, bp), mp,
+				XFS_ERRTAG_ALLOC_READ_AGF,
+				XFS_RANDOM_ALLOC_READ_AGF))
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+
+	if (bp->b_error)
+		xfs_verifier_error(bp);
 }
 
 static void
@@ -2261,8 +2252,8 @@ xfs_agf_write_verify(
 	struct xfs_buf_log_item	*bip = bp->b_fspriv;
 
 	if (!xfs_agf_verify(mp, bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
 		return;
 	}
 
@@ -2272,8 +2263,7 @@ xfs_agf_write_verify(
 	if (bip)
 		XFS_BUF_TO_AGF(bp)->agf_lsn = cpu_to_be64(bip->bli_item.li_lsn);
 
-	xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
-			 offsetof(struct xfs_agf, agf_crc));
+	xfs_buf_update_cksum(bp, XFS_AGF_CRC_OFF);
 }
 
 const struct xfs_buf_ops xfs_agf_buf_ops = {
@@ -2611,11 +2601,11 @@ xfs_free_extent(
 	 */
 	args.agno = XFS_FSB_TO_AGNO(args.mp, bno);
 	if (args.agno >= args.mp->m_sb.sb_agcount)
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 
 	args.agbno = XFS_FSB_TO_AGBNO(args.mp, bno);
 	if (args.agbno >= args.mp->m_sb.sb_agblocks)
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 
 	args.pag = xfs_perag_get(args.mp, args.agno);
 	ASSERT(args.pag);
@@ -2627,7 +2617,7 @@ xfs_free_extent(
 	/* validate the extent size is legal now we have the agf locked */
 	if (args.agbno + len >
 			be32_to_cpu(XFS_BUF_TO_AGF(args.agbp)->agf_length)) {
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto error0;
 	}
 
diff --git a/fs/xfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h
index feacb061bab7..feacb061bab7 100644
--- a/fs/xfs/xfs_alloc.h
+++ b/fs/xfs/libxfs/xfs_alloc.h
diff --git a/fs/xfs/xfs_alloc_btree.c b/fs/xfs/libxfs/xfs_alloc_btree.c
index 13085429e523..e0e83e24d3ef 100644
--- a/fs/xfs/xfs_alloc_btree.c
+++ b/fs/xfs/libxfs/xfs_alloc_btree.c
@@ -70,7 +70,6 @@ xfs_allocbt_alloc_block(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_ptr	*start,
 	union xfs_btree_ptr	*new,
-	int			length,
 	int			*stat)
 {
 	int			error;
@@ -355,12 +354,14 @@ static void
 xfs_allocbt_read_verify(
 	struct xfs_buf	*bp)
 {
-	if (!(xfs_btree_sblock_verify_crc(bp) &&
-	      xfs_allocbt_verify(bp))) {
+	if (!xfs_btree_sblock_verify_crc(bp))
+		xfs_buf_ioerror(bp, -EFSBADCRC);
+	else if (!xfs_allocbt_verify(bp))
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+
+	if (bp->b_error) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
-				     bp->b_target->bt_mount, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_verifier_error(bp);
 	}
 }
 
@@ -370,9 +371,9 @@ xfs_allocbt_write_verify(
 {
 	if (!xfs_allocbt_verify(bp)) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
-				     bp->b_target->bt_mount, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
+		return;
 	}
 	xfs_btree_sblock_calc_crc(bp);
 
diff --git a/fs/xfs/xfs_alloc_btree.h b/fs/xfs/libxfs/xfs_alloc_btree.h
index 45e189e7e81c..45e189e7e81c 100644
--- a/fs/xfs/xfs_alloc_btree.h
+++ b/fs/xfs/libxfs/xfs_alloc_btree.h
diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c
index b86127072ac3..353fb425faef 100644
--- a/fs/xfs/xfs_attr.c
+++ b/fs/xfs/libxfs/xfs_attr.c
@@ -77,17 +77,27 @@ STATIC int xfs_attr_refillstate(xfs_da_state_t *state);
 
 
 STATIC int
-xfs_attr_name_to_xname(
-	struct xfs_name	*xname,
-	const unsigned char *aname)
+xfs_attr_args_init(
+	struct xfs_da_args	*args,
+	struct xfs_inode	*dp,
+	const unsigned char	*name,
+	int			flags)
 {
-	if (!aname)
-		return EINVAL;
-	xname->name = aname;
-	xname->len = strlen((char *)aname);
-	if (xname->len >= MAXNAMELEN)
-		return EFAULT;		/* match IRIX behaviour */
 
+	if (!name)
+		return -EINVAL;
+
+	memset(args, 0, sizeof(*args));
+	args->geo = dp->i_mount->m_attr_geo;
+	args->whichfork = XFS_ATTR_FORK;
+	args->dp = dp;
+	args->flags = flags;
+	args->name = name;
+	args->namelen = strlen((const char *)name);
+	if (args->namelen >= MAXNAMELEN)
+		return -EFAULT;		/* match IRIX behaviour */
+
+	args->hashval = xfs_da_hashname(args->name, args->namelen);
 	return 0;
 }
 
@@ -106,78 +116,46 @@ xfs_inode_hasattr(
  * Overall external interface routines.
  *========================================================================*/
 
-STATIC int
-xfs_attr_get_int(
+int
+xfs_attr_get(
 	struct xfs_inode	*ip,
-	struct xfs_name		*name,
+	const unsigned char	*name,
 	unsigned char		*value,
 	int			*valuelenp,
 	int			flags)
 {
-	xfs_da_args_t   args;
-	int             error;
+	struct xfs_da_args	args;
+	uint			lock_mode;
+	int			error;
+
+	XFS_STATS_INC(xs_attr_get);
+
+	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+		return -EIO;
 
 	if (!xfs_inode_hasattr(ip))
-		return ENOATTR;
+		return -ENOATTR;
+
+	error = xfs_attr_args_init(&args, ip, name, flags);
+	if (error)
+		return error;
 
-	/*
-	 * Fill in the arg structure for this request.
-	 */
-	memset((char *)&args, 0, sizeof(args));
-	args.name = name->name;
-	args.namelen = name->len;
 	args.value = value;
 	args.valuelen = *valuelenp;
-	args.flags = flags;
-	args.hashval = xfs_da_hashname(args.name, args.namelen);
-	args.dp = ip;
-	args.whichfork = XFS_ATTR_FORK;
 
-	/*
-	 * Decide on what work routines to call based on the inode size.
-	 */
-	if (ip->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
+	lock_mode = xfs_ilock_attr_map_shared(ip);
+	if (!xfs_inode_hasattr(ip))
+		error = -ENOATTR;
+	else if (ip->i_d.di_aformat == XFS_DINODE_FMT_LOCAL)
 		error = xfs_attr_shortform_getvalue(&args);
-	} else if (xfs_bmap_one_block(ip, XFS_ATTR_FORK)) {
+	else if (xfs_bmap_one_block(ip, XFS_ATTR_FORK))
 		error = xfs_attr_leaf_get(&args);
-	} else {
+	else
 		error = xfs_attr_node_get(&args);
-	}
+	xfs_iunlock(ip, lock_mode);
 
-	/*
-	 * Return the number of bytes in the value to the caller.
-	 */
 	*valuelenp = args.valuelen;
-
-	if (error == EEXIST)
-		error = 0;
-	return(error);
-}
-
-int
-xfs_attr_get(
-	xfs_inode_t	*ip,
-	const unsigned char *name,
-	unsigned char	*value,
-	int		*valuelenp,
-	int		flags)
-{
-	int		error;
-	struct xfs_name	xname;
-
-	XFS_STATS_INC(xs_attr_get);
-
-	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
-		return(EIO);
-
-	error = xfs_attr_name_to_xname(&xname, name);
-	if (error)
-		return error;
-
-	xfs_ilock(ip, XFS_ILOCK_SHARED);
-	error = xfs_attr_get_int(ip, &xname, value, valuelenp, flags);
-	xfs_iunlock(ip, XFS_ILOCK_SHARED);
-	return(error);
+	return error == -EEXIST ? 0 : error;
 }
 
 /*
@@ -185,12 +163,10 @@ xfs_attr_get(
  */
 STATIC int
 xfs_attr_calc_size(
-	struct xfs_inode 	*ip,
-	int			namelen,
-	int			valuelen,
+	struct xfs_da_args	*args,
 	int			*local)
 {
-	struct xfs_mount 	*mp = ip->i_mount;
+	struct xfs_mount	*mp = args->dp->i_mount;
 	int			size;
 	int			nblks;
 
@@ -198,12 +174,10 @@ xfs_attr_calc_size(
 	 * Determine space new attribute will use, and if it would be
 	 * "local" or "remote" (note: local != inline).
 	 */
-	size = xfs_attr_leaf_newentsize(namelen, valuelen,
-					mp->m_sb.sb_blocksize, local);
-
+	size = xfs_attr_leaf_newentsize(args, local);
 	nblks = XFS_DAENTER_SPACE_RES(mp, XFS_ATTR_FORK);
 	if (*local) {
-		if (size > (mp->m_sb.sb_blocksize >> 1)) {
+		if (size > (args->geo->blksize / 2)) {
 			/* Double split possible */
 			nblks *= 2;
 		}
@@ -212,7 +186,7 @@ xfs_attr_calc_size(
 		 * Out of line attribute, cannot double split, but
 		 * make room for the attribute value itself.
 		 */
-		uint	dblocks = XFS_B_TO_FSB(mp, valuelen);
+		uint	dblocks = xfs_attr3_rmt_blocks(mp, args->valuelen);
 		nblks += dblocks;
 		nblks += XFS_NEXTENTADD_SPACE_RES(mp, dblocks, XFS_ATTR_FORK);
 	}
@@ -220,26 +194,38 @@ xfs_attr_calc_size(
 	return nblks;
 }
 
-STATIC int
-xfs_attr_set_int(
-	struct xfs_inode *dp,
-	struct xfs_name	*name,
-	unsigned char	*value,
-	int		valuelen,
-	int		flags)
+int
+xfs_attr_set(
+	struct xfs_inode	*dp,
+	const unsigned char	*name,
+	unsigned char		*value,
+	int			valuelen,
+	int			flags)
 {
-	xfs_da_args_t		args;
-	xfs_fsblock_t		firstblock;
-	xfs_bmap_free_t		flist;
-	int			error, err2, committed;
 	struct xfs_mount	*mp = dp->i_mount;
+	struct xfs_da_args	args;
+	struct xfs_bmap_free	flist;
 	struct xfs_trans_res	tres;
+	xfs_fsblock_t		firstblock;
 	int			rsvd = (flags & ATTR_ROOT) != 0;
-	int			local;
+	int			error, err2, committed, local;
+
+	XFS_STATS_INC(xs_attr_set);
+
+	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
+		return -EIO;
+
+	error = xfs_attr_args_init(&args, dp, name, flags);
+	if (error)
+		return error;
+
+	args.value = value;
+	args.valuelen = valuelen;
+	args.firstblock = &firstblock;
+	args.flist = &flist;
+	args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT;
+	args.total = xfs_attr_calc_size(&args, &local);
 
-	/*
-	 * Attach the dquots to the inode.
-	 */
 	error = xfs_qm_dqattach(dp, 0);
 	if (error)
 		return error;
@@ -250,32 +236,14 @@ xfs_attr_set_int(
 	 */
 	if (XFS_IFORK_Q(dp) == 0) {
 		int sf_size = sizeof(xfs_attr_sf_hdr_t) +
-			      XFS_ATTR_SF_ENTSIZE_BYNAME(name->len, valuelen);
+			XFS_ATTR_SF_ENTSIZE_BYNAME(args.namelen, valuelen);
 
-		if ((error = xfs_bmap_add_attrfork(dp, sf_size, rsvd)))
-			return(error);
+		error = xfs_bmap_add_attrfork(dp, sf_size, rsvd);
+		if (error)
+			return error;
 	}
 
 	/*
-	 * Fill in the arg structure for this request.
-	 */
-	memset((char *)&args, 0, sizeof(args));
-	args.name = name->name;
-	args.namelen = name->len;
-	args.value = value;
-	args.valuelen = valuelen;
-	args.flags = flags;
-	args.hashval = xfs_da_hashname(args.name, args.namelen);
-	args.dp = dp;
-	args.firstblock = &firstblock;
-	args.flist = &flist;
-	args.whichfork = XFS_ATTR_FORK;
-	args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT;
-
-	/* Size is now blocks for attribute data */
-	args.total = xfs_attr_calc_size(dp, name->len, valuelen, &local);
-
-	/*
 	 * Start our first transaction of the day.
 	 *
 	 * All future transactions during this code must be "chained" off
@@ -302,7 +270,7 @@ xfs_attr_set_int(
 	error = xfs_trans_reserve(args.trans, &tres, args.total, 0);
 	if (error) {
 		xfs_trans_cancel(args.trans, 0);
-		return(error);
+		return error;
 	}
 	xfs_ilock(dp, XFS_ILOCK_EXCL);
 
@@ -312,7 +280,7 @@ xfs_attr_set_int(
 	if (error) {
 		xfs_iunlock(dp, XFS_ILOCK_EXCL);
 		xfs_trans_cancel(args.trans, XFS_TRANS_RELEASE_LOG_RES);
-		return (error);
+		return error;
 	}
 
 	xfs_trans_ijoin(args.trans, dp, 0);
@@ -321,9 +289,9 @@ xfs_attr_set_int(
 	 * If the attribute list is non-existent or a shortform list,
 	 * upgrade it to a single-leaf-block attribute list.
 	 */
-	if ((dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) ||
-	    ((dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS) &&
-	     (dp->i_d.di_anextents == 0))) {
+	if (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL ||
+	    (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS &&
+	     dp->i_d.di_anextents == 0)) {
 
 		/*
 		 * Build initial attribute list (if required).
@@ -336,7 +304,7 @@ xfs_attr_set_int(
 		 * the inode.
 		 */
 		error = xfs_attr_shortform_addname(&args);
-		if (error != ENOSPC) {
+		if (error != -ENOSPC) {
 			/*
 			 * Commit the shortform mods, and we're done.
 			 * NOTE: this is also the error path (EEXIST, etc).
@@ -348,9 +316,8 @@ xfs_attr_set_int(
 			 * the transaction goes to disk before returning
 			 * to the user.
 			 */
-			if (mp->m_flags & XFS_MOUNT_WSYNC) {
+			if (mp->m_flags & XFS_MOUNT_WSYNC)
 				xfs_trans_set_sync(args.trans);
-			}
 
 			if (!error && (flags & ATTR_KERNOTIME) == 0) {
 				xfs_trans_ichgtime(args.trans, dp,
@@ -360,7 +327,7 @@ xfs_attr_set_int(
 						 XFS_TRANS_RELEASE_LOG_RES);
 			xfs_iunlock(dp, XFS_ILOCK_EXCL);
 
-			return(error == 0 ? err2 : error);
+			return error ? error : err2;
 		}
 
 		/*
@@ -398,22 +365,19 @@ xfs_attr_set_int(
 
 	}
 
-	if (xfs_bmap_one_block(dp, XFS_ATTR_FORK)) {
+	if (xfs_bmap_one_block(dp, XFS_ATTR_FORK))
 		error = xfs_attr_leaf_addname(&args);
-	} else {
+	else
 		error = xfs_attr_node_addname(&args);
-	}
-	if (error) {
+	if (error)
 		goto out;
-	}
 
 	/*
 	 * If this is a synchronous mount, make sure that the
 	 * transaction goes to disk before returning to the user.
 	 */
-	if (mp->m_flags & XFS_MOUNT_WSYNC) {
+	if (mp->m_flags & XFS_MOUNT_WSYNC)
 		xfs_trans_set_sync(args.trans);
-	}
 
 	if ((flags & ATTR_KERNOTIME) == 0)
 		xfs_trans_ichgtime(args.trans, dp, XFS_ICHGTIME_CHG);
@@ -425,65 +389,47 @@ xfs_attr_set_int(
 	error = xfs_trans_commit(args.trans, XFS_TRANS_RELEASE_LOG_RES);
 	xfs_iunlock(dp, XFS_ILOCK_EXCL);
 
-	return(error);
+	return error;
 
 out:
-	if (args.trans)
+	if (args.trans) {
 		xfs_trans_cancel(args.trans,
 			XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
+	}
 	xfs_iunlock(dp, XFS_ILOCK_EXCL);
-	return(error);
+	return error;
 }
 
+/*
+ * Generic handler routine to remove a name from an attribute list.
+ * Transitions attribute list from Btree to shortform as necessary.
+ */
 int
-xfs_attr_set(
-	xfs_inode_t	*dp,
-	const unsigned char *name,
-	unsigned char	*value,
-	int		valuelen,
-	int		flags)
+xfs_attr_remove(
+	struct xfs_inode	*dp,
+	const unsigned char	*name,
+	int			flags)
 {
-	int             error;
-	struct xfs_name	xname;
+	struct xfs_mount	*mp = dp->i_mount;
+	struct xfs_da_args	args;
+	struct xfs_bmap_free	flist;
+	xfs_fsblock_t		firstblock;
+	int			error;
 
-	XFS_STATS_INC(xs_attr_set);
+	XFS_STATS_INC(xs_attr_remove);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return (EIO);
+		return -EIO;
+
+	if (!xfs_inode_hasattr(dp))
+		return -ENOATTR;
 
-	error = xfs_attr_name_to_xname(&xname, name);
+	error = xfs_attr_args_init(&args, dp, name, flags);
 	if (error)
 		return error;
 
-	return xfs_attr_set_int(dp, &xname, value, valuelen, flags);
-}
-
-/*
- * Generic handler routine to remove a name from an attribute list.
- * Transitions attribute list from Btree to shortform as necessary.
- */
-STATIC int
-xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags)
-{
-	xfs_da_args_t	args;
-	xfs_fsblock_t	firstblock;
-	xfs_bmap_free_t	flist;
-	int		error;
-	xfs_mount_t	*mp = dp->i_mount;
-
-	/*
-	 * Fill in the arg structure for this request.
-	 */
-	memset((char *)&args, 0, sizeof(args));
-	args.name = name->name;
-	args.namelen = name->len;
-	args.flags = flags;
-	args.hashval = xfs_da_hashname(args.name, args.namelen);
-	args.dp = dp;
 	args.firstblock = &firstblock;
 	args.flist = &flist;
-	args.total = 0;
-	args.whichfork = XFS_ATTR_FORK;
 
 	/*
 	 * we have no control over the attribute names that userspace passes us
@@ -492,9 +438,6 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags)
 	 */
 	args.op_flags = XFS_DA_OP_OKNOENT;
 
-	/*
-	 * Attach the dquots to the inode.
-	 */
 	error = xfs_qm_dqattach(dp, 0);
 	if (error)
 		return error;
@@ -523,7 +466,7 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags)
 				  XFS_ATTRRM_SPACE_RES(mp), 0);
 	if (error) {
 		xfs_trans_cancel(args.trans, 0);
-		return(error);
+		return error;
 	}
 
 	xfs_ilock(dp, XFS_ILOCK_EXCL);
@@ -533,35 +476,26 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags)
 	 */
 	xfs_trans_ijoin(args.trans, dp, 0);
 
-	/*
-	 * Decide on what work routines to call based on the inode size.
-	 */
 	if (!xfs_inode_hasattr(dp)) {
-		error = XFS_ERROR(ENOATTR);
-		goto out;
-	}
-	if (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
+		error = -ENOATTR;
+	} else if (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
 		ASSERT(dp->i_afp->if_flags & XFS_IFINLINE);
 		error = xfs_attr_shortform_remove(&args);
-		if (error) {
-			goto out;
-		}
 	} else if (xfs_bmap_one_block(dp, XFS_ATTR_FORK)) {
 		error = xfs_attr_leaf_removename(&args);
 	} else {
 		error = xfs_attr_node_removename(&args);
 	}
-	if (error) {
+
+	if (error)
 		goto out;
-	}
 
 	/*
 	 * If this is a synchronous mount, make sure that the
 	 * transaction goes to disk before returning to the user.
 	 */
-	if (mp->m_flags & XFS_MOUNT_WSYNC) {
+	if (mp->m_flags & XFS_MOUNT_WSYNC)
 		xfs_trans_set_sync(args.trans);
-	}
 
 	if ((flags & ATTR_KERNOTIME) == 0)
 		xfs_trans_ichgtime(args.trans, dp, XFS_ICHGTIME_CHG);
@@ -573,45 +507,17 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags)
 	error = xfs_trans_commit(args.trans, XFS_TRANS_RELEASE_LOG_RES);
 	xfs_iunlock(dp, XFS_ILOCK_EXCL);
 
-	return(error);
+	return error;
 
 out:
-	if (args.trans)
+	if (args.trans) {
 		xfs_trans_cancel(args.trans,
 			XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
-	xfs_iunlock(dp, XFS_ILOCK_EXCL);
-	return(error);
-}
-
-int
-xfs_attr_remove(
-	xfs_inode_t	*dp,
-	const unsigned char *name,
-	int		flags)
-{
-	int		error;
-	struct xfs_name	xname;
-
-	XFS_STATS_INC(xs_attr_remove);
-
-	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return (EIO);
-
-	error = xfs_attr_name_to_xname(&xname, name);
-	if (error)
-		return error;
-
-	xfs_ilock(dp, XFS_ILOCK_SHARED);
-	if (!xfs_inode_hasattr(dp)) {
-		xfs_iunlock(dp, XFS_ILOCK_SHARED);
-		return XFS_ERROR(ENOATTR);
 	}
-	xfs_iunlock(dp, XFS_ILOCK_SHARED);
-
-	return xfs_attr_remove_int(dp, &xname, flags);
+	xfs_iunlock(dp, XFS_ILOCK_EXCL);
+	return error;
 }
 
-
 /*========================================================================
  * External routines when attribute list is inside the inode
  *========================================================================*/
@@ -628,28 +534,28 @@ xfs_attr_shortform_addname(xfs_da_args_t *args)
 	trace_xfs_attr_sf_addname(args);
 
 	retval = xfs_attr_shortform_lookup(args);
-	if ((args->flags & ATTR_REPLACE) && (retval == ENOATTR)) {
-		return(retval);
-	} else if (retval == EEXIST) {
+	if ((args->flags & ATTR_REPLACE) && (retval == -ENOATTR)) {
+		return retval;
+	} else if (retval == -EEXIST) {
 		if (args->flags & ATTR_CREATE)
-			return(retval);
+			return retval;
 		retval = xfs_attr_shortform_remove(args);
 		ASSERT(retval == 0);
 	}
 
 	if (args->namelen >= XFS_ATTR_SF_ENTSIZE_MAX ||
 	    args->valuelen >= XFS_ATTR_SF_ENTSIZE_MAX)
-		return(XFS_ERROR(ENOSPC));
+		return -ENOSPC;
 
 	newsize = XFS_ATTR_SF_TOTSIZE(args->dp);
 	newsize += XFS_ATTR_SF_ENTSIZE_BYNAME(args->namelen, args->valuelen);
 
 	forkoff = xfs_attr_shortform_bytesfit(args->dp, newsize);
 	if (!forkoff)
-		return(XFS_ERROR(ENOSPC));
+		return -ENOSPC;
 
 	xfs_attr_shortform_add(args, forkoff);
-	return(0);
+	return 0;
 }
 
 
@@ -686,10 +592,10 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 	 * the given flags produce an error or call for an atomic rename.
 	 */
 	retval = xfs_attr3_leaf_lookup_int(bp, args);
-	if ((args->flags & ATTR_REPLACE) && (retval == ENOATTR)) {
+	if ((args->flags & ATTR_REPLACE) && (retval == -ENOATTR)) {
 		xfs_trans_brelse(args->trans, bp);
 		return retval;
-	} else if (retval == EEXIST) {
+	} else if (retval == -EEXIST) {
 		if (args->flags & ATTR_CREATE) {	/* pure create op */
 			xfs_trans_brelse(args->trans, bp);
 			return retval;
@@ -697,11 +603,22 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 
 		trace_xfs_attr_leaf_replace(args);
 
+		/* save the attribute state for later removal*/
 		args->op_flags |= XFS_DA_OP_RENAME;	/* an atomic rename */
 		args->blkno2 = args->blkno;		/* set 2nd entry info*/
 		args->index2 = args->index;
 		args->rmtblkno2 = args->rmtblkno;
 		args->rmtblkcnt2 = args->rmtblkcnt;
+		args->rmtvaluelen2 = args->rmtvaluelen;
+
+		/*
+		 * clear the remote attr state now that it is saved so that the
+		 * values reflect the state of the attribute we are about to
+		 * add, not the attribute we just found and will remove later.
+		 */
+		args->rmtblkno = 0;
+		args->rmtblkcnt = 0;
+		args->rmtvaluelen = 0;
 	}
 
 	/*
@@ -709,7 +626,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 	 * if required.
 	 */
 	retval = xfs_attr3_leaf_add(bp, args);
-	if (retval == ENOSPC) {
+	if (retval == -ENOSPC) {
 		/*
 		 * Promote the attribute list to the Btree format, then
 		 * Commit that transaction so that the node_addname() call
@@ -725,7 +642,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 			ASSERT(committed);
 			args->trans = NULL;
 			xfs_bmap_cancel(args->flist);
-			return(error);
+			return error;
 		}
 
 		/*
@@ -741,13 +658,13 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 		 */
 		error = xfs_trans_roll(&args->trans, dp);
 		if (error)
-			return (error);
+			return error;
 
 		/*
 		 * Fob the whole rest of the problem off on the Btree code.
 		 */
 		error = xfs_attr_node_addname(args);
-		return(error);
+		return error;
 	}
 
 	/*
@@ -756,7 +673,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 	 */
 	error = xfs_trans_roll(&args->trans, dp);
 	if (error)
-		return (error);
+		return error;
 
 	/*
 	 * If there was an out-of-line value, allocate the blocks we
@@ -767,7 +684,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 	if (args->rmtblkno > 0) {
 		error = xfs_attr_rmtval_set(args);
 		if (error)
-			return(error);
+			return error;
 	}
 
 	/*
@@ -783,7 +700,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 		 */
 		error = xfs_attr3_leaf_flipflags(args);
 		if (error)
-			return(error);
+			return error;
 
 		/*
 		 * Dismantle the "old" attribute/value pair by removing
@@ -793,10 +710,11 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 		args->blkno = args->blkno2;
 		args->rmtblkno = args->rmtblkno2;
 		args->rmtblkcnt = args->rmtblkcnt2;
+		args->rmtvaluelen = args->rmtvaluelen2;
 		if (args->rmtblkno) {
 			error = xfs_attr_rmtval_remove(args);
 			if (error)
-				return(error);
+				return error;
 		}
 
 		/*
@@ -826,7 +744,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 				ASSERT(committed);
 				args->trans = NULL;
 				xfs_bmap_cancel(args->flist);
-				return(error);
+				return error;
 			}
 
 			/*
@@ -877,7 +795,7 @@ xfs_attr_leaf_removename(xfs_da_args_t *args)
 		return error;
 
 	error = xfs_attr3_leaf_lookup_int(bp, args);
-	if (error == ENOATTR) {
+	if (error == -ENOATTR) {
 		xfs_trans_brelse(args->trans, bp);
 		return error;
 	}
@@ -932,7 +850,7 @@ xfs_attr_leaf_get(xfs_da_args_t *args)
 		return error;
 
 	error = xfs_attr3_leaf_lookup_int(bp, args);
-	if (error != EEXIST)  {
+	if (error != -EEXIST)  {
 		xfs_trans_brelse(args->trans, bp);
 		return error;
 	}
@@ -945,7 +863,7 @@ xfs_attr_leaf_get(xfs_da_args_t *args)
 }
 
 /*========================================================================
- * External routines when attribute list size > XFS_LBSIZE(mp).
+ * External routines when attribute list size > geo->blksize
  *========================================================================*/
 
 /*
@@ -978,8 +896,6 @@ restart:
 	state = xfs_da_state_alloc();
 	state->args = args;
 	state->mp = mp;
-	state->blocksize = state->mp->m_sb.sb_blocksize;
-	state->node_ents = state->mp->m_attr_node_ents;
 
 	/*
 	 * Search to see if name already exists, and get back a pointer
@@ -990,25 +906,34 @@ restart:
 		goto out;
 	blk = &state->path.blk[ state->path.active-1 ];
 	ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
-	if ((args->flags & ATTR_REPLACE) && (retval == ENOATTR)) {
+	if ((args->flags & ATTR_REPLACE) && (retval == -ENOATTR)) {
 		goto out;
-	} else if (retval == EEXIST) {
+	} else if (retval == -EEXIST) {
 		if (args->flags & ATTR_CREATE)
 			goto out;
 
 		trace_xfs_attr_node_replace(args);
 
+		/* save the attribute state for later removal*/
 		args->op_flags |= XFS_DA_OP_RENAME;	/* atomic rename op */
 		args->blkno2 = args->blkno;		/* set 2nd entry info*/
 		args->index2 = args->index;
 		args->rmtblkno2 = args->rmtblkno;
 		args->rmtblkcnt2 = args->rmtblkcnt;
+		args->rmtvaluelen2 = args->rmtvaluelen;
+
+		/*
+		 * clear the remote attr state now that it is saved so that the
+		 * values reflect the state of the attribute we are about to
+		 * add, not the attribute we just found and will remove later.
+		 */
 		args->rmtblkno = 0;
 		args->rmtblkcnt = 0;
+		args->rmtvaluelen = 0;
 	}
 
 	retval = xfs_attr3_leaf_add(blk->bp, state->args);
-	if (retval == ENOSPC) {
+	if (retval == -ENOSPC) {
 		if (state->path.active == 1) {
 			/*
 			 * Its really a single leaf node, but it had
@@ -1106,7 +1031,7 @@ restart:
 	if (args->rmtblkno > 0) {
 		error = xfs_attr_rmtval_set(args);
 		if (error)
-			return(error);
+			return error;
 	}
 
 	/*
@@ -1132,10 +1057,11 @@ restart:
 		args->blkno = args->blkno2;
 		args->rmtblkno = args->rmtblkno2;
 		args->rmtblkcnt = args->rmtblkcnt2;
+		args->rmtvaluelen = args->rmtvaluelen2;
 		if (args->rmtblkno) {
 			error = xfs_attr_rmtval_remove(args);
 			if (error)
-				return(error);
+				return error;
 		}
 
 		/*
@@ -1147,8 +1073,6 @@ restart:
 		state = xfs_da_state_alloc();
 		state->args = args;
 		state->mp = mp;
-		state->blocksize = state->mp->m_sb.sb_blocksize;
-		state->node_ents = state->mp->m_attr_node_ents;
 		state->inleaf = 0;
 		error = xfs_da3_node_lookup_int(state, &retval);
 		if (error)
@@ -1210,8 +1134,8 @@ out:
 	if (state)
 		xfs_da_state_free(state);
 	if (error)
-		return(error);
-	return(retval);
+		return error;
+	return retval;
 }
 
 /*
@@ -1239,14 +1163,12 @@ xfs_attr_node_removename(xfs_da_args_t *args)
 	state = xfs_da_state_alloc();
 	state->args = args;
 	state->mp = dp->i_mount;
-	state->blocksize = state->mp->m_sb.sb_blocksize;
-	state->node_ents = state->mp->m_attr_node_ents;
 
 	/*
 	 * Search to see if name exists, and get back a pointer to it.
 	 */
 	error = xfs_da3_node_lookup_int(state, &retval);
-	if (error || (retval != EEXIST)) {
+	if (error || (retval != -EEXIST)) {
 		if (error == 0)
 			error = retval;
 		goto out;
@@ -1375,7 +1297,7 @@ xfs_attr_node_removename(xfs_da_args_t *args)
 
 out:
 	xfs_da_state_free(state);
-	return(error);
+	return error;
 }
 
 /*
@@ -1423,7 +1345,7 @@ xfs_attr_fillstate(xfs_da_state_t *state)
 		}
 	}
 
-	return(0);
+	return 0;
 }
 
 /*
@@ -1454,7 +1376,7 @@ xfs_attr_refillstate(xfs_da_state_t *state)
 						blk->blkno, blk->disk_blkno,
 						&blk->bp, XFS_ATTR_FORK);
 			if (error)
-				return(error);
+				return error;
 		} else {
 			blk->bp = NULL;
 		}
@@ -1473,13 +1395,13 @@ xfs_attr_refillstate(xfs_da_state_t *state)
 						blk->blkno, blk->disk_blkno,
 						&blk->bp, XFS_ATTR_FORK);
 			if (error)
-				return(error);
+				return error;
 		} else {
 			blk->bp = NULL;
 		}
 	}
 
-	return(0);
+	return 0;
 }
 
 /*
@@ -1502,8 +1424,6 @@ xfs_attr_node_get(xfs_da_args_t *args)
 	state = xfs_da_state_alloc();
 	state->args = args;
 	state->mp = args->dp->i_mount;
-	state->blocksize = state->mp->m_sb.sb_blocksize;
-	state->node_ents = state->mp->m_attr_node_ents;
 
 	/*
 	 * Search to see if name exists, and get back a pointer to it.
@@ -1511,7 +1431,7 @@ xfs_attr_node_get(xfs_da_args_t *args)
 	error = xfs_da3_node_lookup_int(state, &retval);
 	if (error) {
 		retval = error;
-	} else if (retval == EEXIST) {
+	} else if (retval == -EEXIST) {
 		blk = &state->path.blk[ state->path.active-1 ];
 		ASSERT(blk->bp != NULL);
 		ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC);
@@ -1535,5 +1455,5 @@ xfs_attr_node_get(xfs_da_args_t *args)
 	}
 
 	xfs_da_state_free(state);
-	return(retval);
+	return retval;
 }
diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c
index 7b126f46a2f9..b1f73dbbf3d8 100644
--- a/fs/xfs/xfs_attr_leaf.c
+++ b/fs/xfs/libxfs/xfs_attr_leaf.c
@@ -80,11 +80,12 @@ STATIC int xfs_attr3_leaf_figure_balance(xfs_da_state_t *state,
 /*
  * Utility routines.
  */
-STATIC void xfs_attr3_leaf_moveents(struct xfs_attr_leafblock *src_leaf,
+STATIC void xfs_attr3_leaf_moveents(struct xfs_da_args *args,
+			struct xfs_attr_leafblock *src_leaf,
 			struct xfs_attr3_icleaf_hdr *src_ichdr, int src_start,
 			struct xfs_attr_leafblock *dst_leaf,
 			struct xfs_attr3_icleaf_hdr *dst_ichdr, int dst_start,
-			int move_count, struct xfs_mount *mp);
+			int move_count);
 STATIC int xfs_attr_leaf_entsize(xfs_attr_leafblock_t *leaf, int index);
 
 void
@@ -213,8 +214,8 @@ xfs_attr3_leaf_write_verify(
 	struct xfs_attr3_leaf_hdr *hdr3 = bp->b_addr;
 
 	if (!xfs_attr3_leaf_verify(bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
 		return;
 	}
 
@@ -224,7 +225,7 @@ xfs_attr3_leaf_write_verify(
 	if (bip)
 		hdr3->info.lsn = cpu_to_be64(bip->bli_item.li_lsn);
 
-	xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), XFS_ATTR3_LEAF_CRC_OFF);
+	xfs_buf_update_cksum(bp, XFS_ATTR3_LEAF_CRC_OFF);
 }
 
 /*
@@ -239,13 +240,14 @@ xfs_attr3_leaf_read_verify(
 {
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 
-	if ((xfs_sb_version_hascrc(&mp->m_sb) &&
-	     !xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
-					  XFS_ATTR3_LEAF_CRC_OFF)) ||
-	    !xfs_attr3_leaf_verify(bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
-	}
+	if (xfs_sb_version_hascrc(&mp->m_sb) &&
+	     !xfs_buf_verify_cksum(bp, XFS_ATTR3_LEAF_CRC_OFF))
+		xfs_buf_ioerror(bp, -EFSBADCRC);
+	else if (!xfs_attr3_leaf_verify(bp))
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+
+	if (bp->b_error)
+		xfs_verifier_error(bp);
 }
 
 const struct xfs_buf_ops xfs_attr3_leaf_buf_ops = {
@@ -545,7 +547,7 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
 		break;
 	}
 	if (i == end)
-		return(XFS_ERROR(ENOATTR));
+		return -ENOATTR;
 
 	/*
 	 * Fix up the attribute fork data, covering the hole
@@ -580,7 +582,7 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
 
 	xfs_sbversion_add_attr2(mp, args->trans);
 
-	return(0);
+	return 0;
 }
 
 /*
@@ -609,9 +611,9 @@ xfs_attr_shortform_lookup(xfs_da_args_t *args)
 			continue;
 		if (!xfs_attr_namesp_match(args->flags, sfe->flags))
 			continue;
-		return(XFS_ERROR(EEXIST));
+		return -EEXIST;
 	}
-	return(XFS_ERROR(ENOATTR));
+	return -ENOATTR;
 }
 
 /*
@@ -638,18 +640,18 @@ xfs_attr_shortform_getvalue(xfs_da_args_t *args)
 			continue;
 		if (args->flags & ATTR_KERNOVAL) {
 			args->valuelen = sfe->valuelen;
-			return(XFS_ERROR(EEXIST));
+			return -EEXIST;
 		}
 		if (args->valuelen < sfe->valuelen) {
 			args->valuelen = sfe->valuelen;
-			return(XFS_ERROR(ERANGE));
+			return -ERANGE;
 		}
 		args->valuelen = sfe->valuelen;
 		memcpy(args->value, &sfe->nameval[args->namelen],
 						    args->valuelen);
-		return(XFS_ERROR(EEXIST));
+		return -EEXIST;
 	}
-	return(XFS_ERROR(ENOATTR));
+	return -ENOATTR;
 }
 
 /*
@@ -689,7 +691,7 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
 		 * If we hit an IO error middle of the transaction inside
 		 * grow_inode(), we may have inconsistent data. Bail out.
 		 */
-		if (error == EIO)
+		if (error == -EIO)
 			goto out;
 		xfs_idata_realloc(dp, size, XFS_ATTR_FORK);	/* try to put */
 		memcpy(ifp->if_u1.if_data, tmpbuffer, size);	/* it back */
@@ -710,6 +712,7 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
 
 	memset((char *)&nargs, 0, sizeof(nargs));
 	nargs.dp = dp;
+	nargs.geo = args->geo;
 	nargs.firstblock = args->firstblock;
 	nargs.flist = args->flist;
 	nargs.total = args->total;
@@ -727,9 +730,9 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
 						sfe->namelen);
 		nargs.flags = XFS_ATTR_NSP_ONDISK_TO_ARGS(sfe->flags);
 		error = xfs_attr3_leaf_lookup_int(bp, &nargs); /* set a->index */
-		ASSERT(error == ENOATTR);
+		ASSERT(error == -ENOATTR);
 		error = xfs_attr3_leaf_add(bp, &nargs);
-		ASSERT(error != ENOSPC);
+		ASSERT(error != -ENOSPC);
 		if (error)
 			goto out;
 		sfe = XFS_ATTR_SF_NEXTENTRY(sfe);
@@ -738,7 +741,7 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
 
 out:
 	kmem_free(tmpbuffer);
-	return(error);
+	return error;
 }
 
 /*
@@ -766,12 +769,12 @@ xfs_attr_shortform_allfit(
 		if (entry->flags & XFS_ATTR_INCOMPLETE)
 			continue;		/* don't copy partial entries */
 		if (!(entry->flags & XFS_ATTR_LOCAL))
-			return(0);
+			return 0;
 		name_loc = xfs_attr3_leaf_name_local(leaf, i);
 		if (name_loc->namelen >= XFS_ATTR_SF_ENTSIZE_MAX)
-			return(0);
+			return 0;
 		if (be16_to_cpu(name_loc->valuelen) >= XFS_ATTR_SF_ENTSIZE_MAX)
-			return(0);
+			return 0;
 		bytes += sizeof(struct xfs_attr_sf_entry) - 1
 				+ name_loc->namelen
 				+ be16_to_cpu(name_loc->valuelen);
@@ -804,18 +807,18 @@ xfs_attr3_leaf_to_shortform(
 
 	trace_xfs_attr_leaf_to_sf(args);
 
-	tmpbuffer = kmem_alloc(XFS_LBSIZE(dp->i_mount), KM_SLEEP);
+	tmpbuffer = kmem_alloc(args->geo->blksize, KM_SLEEP);
 	if (!tmpbuffer)
-		return ENOMEM;
+		return -ENOMEM;
 
-	memcpy(tmpbuffer, bp->b_addr, XFS_LBSIZE(dp->i_mount));
+	memcpy(tmpbuffer, bp->b_addr, args->geo->blksize);
 
 	leaf = (xfs_attr_leafblock_t *)tmpbuffer;
 	xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf);
 	entry = xfs_attr3_leaf_entryp(leaf);
 
 	/* XXX (dgc): buffer is about to be marked stale - why zero it? */
-	memset(bp->b_addr, 0, XFS_LBSIZE(dp->i_mount));
+	memset(bp->b_addr, 0, args->geo->blksize);
 
 	/*
 	 * Clean out the prior contents of the attribute list.
@@ -837,6 +840,7 @@ xfs_attr3_leaf_to_shortform(
 	 * Copy the attributes
 	 */
 	memset((char *)&nargs, 0, sizeof(nargs));
+	nargs.geo = args->geo;
 	nargs.dp = dp;
 	nargs.firstblock = args->firstblock;
 	nargs.flist = args->flist;
@@ -903,12 +907,12 @@ xfs_attr3_leaf_to_node(
 	/* copy leaf to new buffer, update identifiers */
 	xfs_trans_buf_set_type(args->trans, bp2, XFS_BLFT_ATTR_LEAF_BUF);
 	bp2->b_ops = bp1->b_ops;
-	memcpy(bp2->b_addr, bp1->b_addr, XFS_LBSIZE(mp));
+	memcpy(bp2->b_addr, bp1->b_addr, args->geo->blksize);
 	if (xfs_sb_version_hascrc(&mp->m_sb)) {
 		struct xfs_da3_blkinfo *hdr3 = bp2->b_addr;
 		hdr3->blkno = cpu_to_be64(bp2->b_bn);
 	}
-	xfs_trans_log_buf(args->trans, bp2, 0, XFS_LBSIZE(mp) - 1);
+	xfs_trans_log_buf(args->trans, bp2, 0, args->geo->blksize - 1);
 
 	/*
 	 * Set up the new root node.
@@ -929,7 +933,7 @@ xfs_attr3_leaf_to_node(
 	btree[0].before = cpu_to_be32(blkno);
 	icnodehdr.count = 1;
 	dp->d_ops->node_hdr_to_disk(node, &icnodehdr);
-	xfs_trans_log_buf(args->trans, bp1, 0, XFS_LBSIZE(mp) - 1);
+	xfs_trans_log_buf(args->trans, bp1, 0, args->geo->blksize - 1);
 	error = 0;
 out:
 	return error;
@@ -965,10 +969,10 @@ xfs_attr3_leaf_create(
 	bp->b_ops = &xfs_attr3_leaf_buf_ops;
 	xfs_trans_buf_set_type(args->trans, bp, XFS_BLFT_ATTR_LEAF_BUF);
 	leaf = bp->b_addr;
-	memset(leaf, 0, XFS_LBSIZE(mp));
+	memset(leaf, 0, args->geo->blksize);
 
 	memset(&ichdr, 0, sizeof(ichdr));
-	ichdr.firstused = XFS_LBSIZE(mp);
+	ichdr.firstused = args->geo->blksize;
 
 	if (xfs_sb_version_hascrc(&mp->m_sb)) {
 		struct xfs_da3_blkinfo *hdr3 = bp->b_addr;
@@ -987,7 +991,7 @@ xfs_attr3_leaf_create(
 	ichdr.freemap[0].size = ichdr.firstused - ichdr.freemap[0].base;
 
 	xfs_attr3_leaf_hdr_to_disk(leaf, &ichdr);
-	xfs_trans_log_buf(args->trans, bp, 0, XFS_LBSIZE(mp) - 1);
+	xfs_trans_log_buf(args->trans, bp, 0, args->geo->blksize - 1);
 
 	*bpp = bp;
 	return 0;
@@ -1013,10 +1017,10 @@ xfs_attr3_leaf_split(
 	ASSERT(oldblk->magic == XFS_ATTR_LEAF_MAGIC);
 	error = xfs_da_grow_inode(state->args, &blkno);
 	if (error)
-		return(error);
+		return error;
 	error = xfs_attr3_leaf_create(state->args, blkno, &newblk->bp);
 	if (error)
-		return(error);
+		return error;
 	newblk->blkno = blkno;
 	newblk->magic = XFS_ATTR_LEAF_MAGIC;
 
@@ -1027,7 +1031,7 @@ xfs_attr3_leaf_split(
 	xfs_attr3_leaf_rebalance(state, oldblk, newblk);
 	error = xfs_da3_blk_link(state, oldblk, newblk);
 	if (error)
-		return(error);
+		return error;
 
 	/*
 	 * Save info on "old" attribute for "atomic rename" ops, leaf_add()
@@ -1049,7 +1053,7 @@ xfs_attr3_leaf_split(
 	 */
 	oldblk->hashval = xfs_attr_leaf_lasthash(oldblk->bp, NULL);
 	newblk->hashval = xfs_attr_leaf_lasthash(newblk->bp, NULL);
-	return(error);
+	return error;
 }
 
 /*
@@ -1073,8 +1077,7 @@ xfs_attr3_leaf_add(
 	leaf = bp->b_addr;
 	xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf);
 	ASSERT(args->index >= 0 && args->index <= ichdr.count);
-	entsize = xfs_attr_leaf_newentsize(args->namelen, args->valuelen,
-			   args->trans->t_mountp->m_sb.sb_blocksize, NULL);
+	entsize = xfs_attr_leaf_newentsize(args, NULL);
 
 	/*
 	 * Search through freemap for first-fit on new name length.
@@ -1105,7 +1108,7 @@ xfs_attr3_leaf_add(
 	 * no good and we should just give up.
 	 */
 	if (!ichdr.holes && sum < entsize)
-		return XFS_ERROR(ENOSPC);
+		return -ENOSPC;
 
 	/*
 	 * Compact the entries to coalesce free space.
@@ -1118,7 +1121,7 @@ xfs_attr3_leaf_add(
 	 * free region, in freemap[0].  If it is not big enough, give up.
 	 */
 	if (ichdr.freemap[0].size < (entsize + sizeof(xfs_attr_leaf_entry_t))) {
-		tmp = ENOSPC;
+		tmp = -ENOSPC;
 		goto out_log_hdr;
 	}
 
@@ -1173,17 +1176,14 @@ xfs_attr3_leaf_add_work(
 	 * Allocate space for the new string (at the end of the run).
 	 */
 	mp = args->trans->t_mountp;
-	ASSERT(ichdr->freemap[mapindex].base < XFS_LBSIZE(mp));
+	ASSERT(ichdr->freemap[mapindex].base < args->geo->blksize);
 	ASSERT((ichdr->freemap[mapindex].base & 0x3) == 0);
 	ASSERT(ichdr->freemap[mapindex].size >=
-		xfs_attr_leaf_newentsize(args->namelen, args->valuelen,
-					 mp->m_sb.sb_blocksize, NULL));
-	ASSERT(ichdr->freemap[mapindex].size < XFS_LBSIZE(mp));
+		xfs_attr_leaf_newentsize(args, NULL));
+	ASSERT(ichdr->freemap[mapindex].size < args->geo->blksize);
 	ASSERT((ichdr->freemap[mapindex].size & 0x3) == 0);
 
-	ichdr->freemap[mapindex].size -=
-			xfs_attr_leaf_newentsize(args->namelen, args->valuelen,
-						 mp->m_sb.sb_blocksize, &tmp);
+	ichdr->freemap[mapindex].size -= xfs_attr_leaf_newentsize(args, &tmp);
 
 	entry->nameidx = cpu_to_be16(ichdr->freemap[mapindex].base +
 				     ichdr->freemap[mapindex].size);
@@ -1228,6 +1228,7 @@ xfs_attr3_leaf_add_work(
 		name_rmt->valueblk = 0;
 		args->rmtblkno = 1;
 		args->rmtblkcnt = xfs_attr3_rmt_blocks(mp, args->valuelen);
+		args->rmtvaluelen = args->valuelen;
 	}
 	xfs_trans_log_buf(args->trans, bp,
 	     XFS_DA_LOGRANGE(leaf, xfs_attr3_leaf_name(leaf, args->index),
@@ -1267,14 +1268,13 @@ xfs_attr3_leaf_compact(
 	struct xfs_attr_leafblock *leaf_dst;
 	struct xfs_attr3_icleaf_hdr ichdr_src;
 	struct xfs_trans	*trans = args->trans;
-	struct xfs_mount	*mp = trans->t_mountp;
 	char			*tmpbuffer;
 
 	trace_xfs_attr_leaf_compact(args);
 
-	tmpbuffer = kmem_alloc(XFS_LBSIZE(mp), KM_SLEEP);
-	memcpy(tmpbuffer, bp->b_addr, XFS_LBSIZE(mp));
-	memset(bp->b_addr, 0, XFS_LBSIZE(mp));
+	tmpbuffer = kmem_alloc(args->geo->blksize, KM_SLEEP);
+	memcpy(tmpbuffer, bp->b_addr, args->geo->blksize);
+	memset(bp->b_addr, 0, args->geo->blksize);
 	leaf_src = (xfs_attr_leafblock_t *)tmpbuffer;
 	leaf_dst = bp->b_addr;
 
@@ -1287,7 +1287,7 @@ xfs_attr3_leaf_compact(
 
 	/* Initialise the incore headers */
 	ichdr_src = *ichdr_dst;	/* struct copy */
-	ichdr_dst->firstused = XFS_LBSIZE(mp);
+	ichdr_dst->firstused = args->geo->blksize;
 	ichdr_dst->usedbytes = 0;
 	ichdr_dst->count = 0;
 	ichdr_dst->holes = 0;
@@ -1302,13 +1302,13 @@ xfs_attr3_leaf_compact(
 	 * Copy all entry's in the same (sorted) order,
 	 * but allocate name/value pairs packed and in sequence.
 	 */
-	xfs_attr3_leaf_moveents(leaf_src, &ichdr_src, 0, leaf_dst, ichdr_dst, 0,
-				ichdr_src.count, mp);
+	xfs_attr3_leaf_moveents(args, leaf_src, &ichdr_src, 0,
+				leaf_dst, ichdr_dst, 0, ichdr_src.count);
 	/*
 	 * this logs the entire buffer, but the caller must write the header
 	 * back to the buffer when it is finished modifying it.
 	 */
-	xfs_trans_log_buf(trans, bp, 0, XFS_LBSIZE(mp) - 1);
+	xfs_trans_log_buf(trans, bp, 0, args->geo->blksize - 1);
 
 	kmem_free(tmpbuffer);
 }
@@ -1459,8 +1459,8 @@ xfs_attr3_leaf_rebalance(
 		/*
 		 * Move high entries from leaf1 to low end of leaf2.
 		 */
-		xfs_attr3_leaf_moveents(leaf1, &ichdr1, ichdr1.count - count,
-				leaf2, &ichdr2, 0, count, state->mp);
+		xfs_attr3_leaf_moveents(args, leaf1, &ichdr1,
+				ichdr1.count - count, leaf2, &ichdr2, 0, count);
 
 	} else if (count > ichdr1.count) {
 		/*
@@ -1488,14 +1488,14 @@ xfs_attr3_leaf_rebalance(
 		/*
 		 * Move low entries from leaf2 to high end of leaf1.
 		 */
-		xfs_attr3_leaf_moveents(leaf2, &ichdr2, 0, leaf1, &ichdr1,
-					ichdr1.count, count, state->mp);
+		xfs_attr3_leaf_moveents(args, leaf2, &ichdr2, 0, leaf1, &ichdr1,
+					ichdr1.count, count);
 	}
 
 	xfs_attr3_leaf_hdr_to_disk(leaf1, &ichdr1);
 	xfs_attr3_leaf_hdr_to_disk(leaf2, &ichdr2);
-	xfs_trans_log_buf(args->trans, blk1->bp, 0, state->blocksize-1);
-	xfs_trans_log_buf(args->trans, blk2->bp, 0, state->blocksize-1);
+	xfs_trans_log_buf(args->trans, blk1->bp, 0, args->geo->blksize - 1);
+	xfs_trans_log_buf(args->trans, blk2->bp, 0, args->geo->blksize - 1);
 
 	/*
 	 * Copy out last hashval in each block for B-tree code.
@@ -1590,11 +1590,9 @@ xfs_attr3_leaf_figure_balance(
 	max = ichdr1->count + ichdr2->count;
 	half = (max + 1) * sizeof(*entry);
 	half += ichdr1->usedbytes + ichdr2->usedbytes +
-			xfs_attr_leaf_newentsize(state->args->namelen,
-						 state->args->valuelen,
-						 state->blocksize, NULL);
+			xfs_attr_leaf_newentsize(state->args, NULL);
 	half /= 2;
-	lastdelta = state->blocksize;
+	lastdelta = state->args->geo->blksize;
 	entry = xfs_attr3_leaf_entryp(leaf1);
 	for (count = index = 0; count < max; entry++, index++, count++) {
 
@@ -1604,10 +1602,7 @@ xfs_attr3_leaf_figure_balance(
 		 */
 		if (count == blk1->index) {
 			tmp = totallen + sizeof(*entry) +
-				xfs_attr_leaf_newentsize(
-						state->args->namelen,
-						state->args->valuelen,
-						state->blocksize, NULL);
+				xfs_attr_leaf_newentsize(state->args, NULL);
 			if (XFS_ATTR_ABS(half - tmp) > lastdelta)
 				break;
 			lastdelta = XFS_ATTR_ABS(half - tmp);
@@ -1643,10 +1638,7 @@ xfs_attr3_leaf_figure_balance(
 	totallen -= count * sizeof(*entry);
 	if (foundit) {
 		totallen -= sizeof(*entry) +
-				xfs_attr_leaf_newentsize(
-						state->args->namelen,
-						state->args->valuelen,
-						state->blocksize, NULL);
+				xfs_attr_leaf_newentsize(state->args, NULL);
 	}
 
 	*countarg = count;
@@ -1698,9 +1690,9 @@ xfs_attr3_leaf_toosmall(
 	bytes = xfs_attr3_leaf_hdr_size(leaf) +
 		ichdr.count * sizeof(xfs_attr_leaf_entry_t) +
 		ichdr.usedbytes;
-	if (bytes > (state->blocksize >> 1)) {
+	if (bytes > (state->args->geo->blksize >> 1)) {
 		*action = 0;	/* blk over 50%, don't try to join */
-		return(0);
+		return 0;
 	}
 
 	/*
@@ -1719,7 +1711,7 @@ xfs_attr3_leaf_toosmall(
 		error = xfs_da3_path_shift(state, &state->altpath, forward,
 						 0, &retval);
 		if (error)
-			return(error);
+			return error;
 		if (retval) {
 			*action = 0;
 		} else {
@@ -1748,11 +1740,12 @@ xfs_attr3_leaf_toosmall(
 		error = xfs_attr3_leaf_read(state->args->trans, state->args->dp,
 					blkno, -1, &bp);
 		if (error)
-			return(error);
+			return error;
 
 		xfs_attr3_leaf_hdr_from_disk(&ichdr2, bp->b_addr);
 
-		bytes = state->blocksize - (state->blocksize >> 2) -
+		bytes = state->args->geo->blksize -
+			(state->args->geo->blksize >> 2) -
 			ichdr.usedbytes - ichdr2.usedbytes -
 			((ichdr.count + ichdr2.count) *
 					sizeof(xfs_attr_leaf_entry_t)) -
@@ -1764,7 +1757,7 @@ xfs_attr3_leaf_toosmall(
 	}
 	if (i >= 2) {
 		*action = 0;
-		return(0);
+		return 0;
 	}
 
 	/*
@@ -1780,13 +1773,13 @@ xfs_attr3_leaf_toosmall(
 						 0, &retval);
 	}
 	if (error)
-		return(error);
+		return error;
 	if (retval) {
 		*action = 0;
 	} else {
 		*action = 1;
 	}
-	return(0);
+	return 0;
 }
 
 /*
@@ -1803,7 +1796,6 @@ xfs_attr3_leaf_remove(
 	struct xfs_attr_leafblock *leaf;
 	struct xfs_attr3_icleaf_hdr ichdr;
 	struct xfs_attr_leaf_entry *entry;
-	struct xfs_mount	*mp = args->trans->t_mountp;
 	int			before;
 	int			after;
 	int			smallest;
@@ -1817,7 +1809,7 @@ xfs_attr3_leaf_remove(
 	leaf = bp->b_addr;
 	xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf);
 
-	ASSERT(ichdr.count > 0 && ichdr.count < XFS_LBSIZE(mp) / 8);
+	ASSERT(ichdr.count > 0 && ichdr.count < args->geo->blksize / 8);
 	ASSERT(args->index >= 0 && args->index < ichdr.count);
 	ASSERT(ichdr.firstused >= ichdr.count * sizeof(*entry) +
 					xfs_attr3_leaf_hdr_size(leaf));
@@ -1825,7 +1817,7 @@ xfs_attr3_leaf_remove(
 	entry = &xfs_attr3_leaf_entryp(leaf)[args->index];
 
 	ASSERT(be16_to_cpu(entry->nameidx) >= ichdr.firstused);
-	ASSERT(be16_to_cpu(entry->nameidx) < XFS_LBSIZE(mp));
+	ASSERT(be16_to_cpu(entry->nameidx) < args->geo->blksize);
 
 	/*
 	 * Scan through free region table:
@@ -1840,8 +1832,8 @@ xfs_attr3_leaf_remove(
 	smallest = XFS_ATTR_LEAF_MAPSIZE - 1;
 	entsize = xfs_attr_leaf_entsize(leaf, args->index);
 	for (i = 0; i < XFS_ATTR_LEAF_MAPSIZE; i++) {
-		ASSERT(ichdr.freemap[i].base < XFS_LBSIZE(mp));
-		ASSERT(ichdr.freemap[i].size < XFS_LBSIZE(mp));
+		ASSERT(ichdr.freemap[i].base < args->geo->blksize);
+		ASSERT(ichdr.freemap[i].size < args->geo->blksize);
 		if (ichdr.freemap[i].base == tablesize) {
 			ichdr.freemap[i].base -= sizeof(xfs_attr_leaf_entry_t);
 			ichdr.freemap[i].size += sizeof(xfs_attr_leaf_entry_t);
@@ -1918,11 +1910,11 @@ xfs_attr3_leaf_remove(
 	 * removing the name.
 	 */
 	if (smallest) {
-		tmp = XFS_LBSIZE(mp);
+		tmp = args->geo->blksize;
 		entry = xfs_attr3_leaf_entryp(leaf);
 		for (i = ichdr.count - 1; i >= 0; entry++, i--) {
 			ASSERT(be16_to_cpu(entry->nameidx) >= ichdr.firstused);
-			ASSERT(be16_to_cpu(entry->nameidx) < XFS_LBSIZE(mp));
+			ASSERT(be16_to_cpu(entry->nameidx) < args->geo->blksize);
 
 			if (be16_to_cpu(entry->nameidx) < tmp)
 				tmp = be16_to_cpu(entry->nameidx);
@@ -1945,7 +1937,7 @@ xfs_attr3_leaf_remove(
 	tmp = ichdr.usedbytes + xfs_attr3_leaf_hdr_size(leaf) +
 	      ichdr.count * sizeof(xfs_attr_leaf_entry_t);
 
-	return tmp < mp->m_attr_magicpct; /* leaf is < 37% full */
+	return tmp < args->geo->magicpct; /* leaf is < 37% full */
 }
 
 /*
@@ -1962,7 +1954,6 @@ xfs_attr3_leaf_unbalance(
 	struct xfs_attr3_icleaf_hdr drophdr;
 	struct xfs_attr3_icleaf_hdr savehdr;
 	struct xfs_attr_leaf_entry *entry;
-	struct xfs_mount	*mp = state->mp;
 
 	trace_xfs_attr_leaf_unbalance(state->args);
 
@@ -1989,13 +1980,15 @@ xfs_attr3_leaf_unbalance(
 		 */
 		if (xfs_attr3_leaf_order(save_blk->bp, &savehdr,
 					 drop_blk->bp, &drophdr)) {
-			xfs_attr3_leaf_moveents(drop_leaf, &drophdr, 0,
+			xfs_attr3_leaf_moveents(state->args,
+						drop_leaf, &drophdr, 0,
 						save_leaf, &savehdr, 0,
-						drophdr.count, mp);
+						drophdr.count);
 		} else {
-			xfs_attr3_leaf_moveents(drop_leaf, &drophdr, 0,
+			xfs_attr3_leaf_moveents(state->args,
+						drop_leaf, &drophdr, 0,
 						save_leaf, &savehdr,
-						savehdr.count, drophdr.count, mp);
+						savehdr.count, drophdr.count);
 		}
 	} else {
 		/*
@@ -2005,7 +1998,7 @@ xfs_attr3_leaf_unbalance(
 		struct xfs_attr_leafblock *tmp_leaf;
 		struct xfs_attr3_icleaf_hdr tmphdr;
 
-		tmp_leaf = kmem_zalloc(state->blocksize, KM_SLEEP);
+		tmp_leaf = kmem_zalloc(state->args->geo->blksize, KM_SLEEP);
 
 		/*
 		 * Copy the header into the temp leaf so that all the stuff
@@ -2018,35 +2011,39 @@ xfs_attr3_leaf_unbalance(
 		tmphdr.magic = savehdr.magic;
 		tmphdr.forw = savehdr.forw;
 		tmphdr.back = savehdr.back;
-		tmphdr.firstused = state->blocksize;
+		tmphdr.firstused = state->args->geo->blksize;
 
 		/* write the header to the temp buffer to initialise it */
 		xfs_attr3_leaf_hdr_to_disk(tmp_leaf, &tmphdr);
 
 		if (xfs_attr3_leaf_order(save_blk->bp, &savehdr,
 					 drop_blk->bp, &drophdr)) {
-			xfs_attr3_leaf_moveents(drop_leaf, &drophdr, 0,
+			xfs_attr3_leaf_moveents(state->args,
+						drop_leaf, &drophdr, 0,
 						tmp_leaf, &tmphdr, 0,
-						drophdr.count, mp);
-			xfs_attr3_leaf_moveents(save_leaf, &savehdr, 0,
+						drophdr.count);
+			xfs_attr3_leaf_moveents(state->args,
+						save_leaf, &savehdr, 0,
 						tmp_leaf, &tmphdr, tmphdr.count,
-						savehdr.count, mp);
+						savehdr.count);
 		} else {
-			xfs_attr3_leaf_moveents(save_leaf, &savehdr, 0,
+			xfs_attr3_leaf_moveents(state->args,
+						save_leaf, &savehdr, 0,
 						tmp_leaf, &tmphdr, 0,
-						savehdr.count, mp);
-			xfs_attr3_leaf_moveents(drop_leaf, &drophdr, 0,
+						savehdr.count);
+			xfs_attr3_leaf_moveents(state->args,
+						drop_leaf, &drophdr, 0,
 						tmp_leaf, &tmphdr, tmphdr.count,
-						drophdr.count, mp);
+						drophdr.count);
 		}
-		memcpy(save_leaf, tmp_leaf, state->blocksize);
+		memcpy(save_leaf, tmp_leaf, state->args->geo->blksize);
 		savehdr = tmphdr; /* struct copy */
 		kmem_free(tmp_leaf);
 	}
 
 	xfs_attr3_leaf_hdr_to_disk(save_leaf, &savehdr);
 	xfs_trans_log_buf(state->args->trans, save_blk->bp, 0,
-					   state->blocksize - 1);
+					   state->args->geo->blksize - 1);
 
 	/*
 	 * Copy out last hashval in each block for B-tree code.
@@ -2092,7 +2089,7 @@ xfs_attr3_leaf_lookup_int(
 	leaf = bp->b_addr;
 	xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf);
 	entries = xfs_attr3_leaf_entryp(leaf);
-	ASSERT(ichdr.count < XFS_LBSIZE(args->dp->i_mount) / 8);
+	ASSERT(ichdr.count < args->geo->blksize / 8);
 
 	/*
 	 * Binary search.  (note: small blocks will skip this loop)
@@ -2126,7 +2123,7 @@ xfs_attr3_leaf_lookup_int(
 	}
 	if (probe == ichdr.count || be32_to_cpu(entry->hashval) != hashval) {
 		args->index = probe;
-		return XFS_ERROR(ENOATTR);
+		return -ENOATTR;
 	}
 
 	/*
@@ -2155,7 +2152,7 @@ xfs_attr3_leaf_lookup_int(
 			if (!xfs_attr_namesp_match(args->flags, entry->flags))
 				continue;
 			args->index = probe;
-			return XFS_ERROR(EEXIST);
+			return -EEXIST;
 		} else {
 			name_rmt = xfs_attr3_leaf_name_remote(leaf, probe);
 			if (name_rmt->namelen != args->namelen)
@@ -2166,16 +2163,16 @@ xfs_attr3_leaf_lookup_int(
 			if (!xfs_attr_namesp_match(args->flags, entry->flags))
 				continue;
 			args->index = probe;
-			args->valuelen = be32_to_cpu(name_rmt->valuelen);
+			args->rmtvaluelen = be32_to_cpu(name_rmt->valuelen);
 			args->rmtblkno = be32_to_cpu(name_rmt->valueblk);
 			args->rmtblkcnt = xfs_attr3_rmt_blocks(
 							args->dp->i_mount,
-							args->valuelen);
-			return XFS_ERROR(EEXIST);
+							args->rmtvaluelen);
+			return -EEXIST;
 		}
 	}
 	args->index = probe;
-	return XFS_ERROR(ENOATTR);
+	return -ENOATTR;
 }
 
 /*
@@ -2196,7 +2193,7 @@ xfs_attr3_leaf_getvalue(
 
 	leaf = bp->b_addr;
 	xfs_attr3_leaf_hdr_from_disk(&ichdr, leaf);
-	ASSERT(ichdr.count < XFS_LBSIZE(args->dp->i_mount) / 8);
+	ASSERT(ichdr.count < args->geo->blksize / 8);
 	ASSERT(args->index < ichdr.count);
 
 	entry = &xfs_attr3_leaf_entryp(leaf)[args->index];
@@ -2211,7 +2208,7 @@ xfs_attr3_leaf_getvalue(
 		}
 		if (args->valuelen < valuelen) {
 			args->valuelen = valuelen;
-			return XFS_ERROR(ERANGE);
+			return -ERANGE;
 		}
 		args->valuelen = valuelen;
 		memcpy(args->value, &name_loc->nameval[args->namelen], valuelen);
@@ -2219,19 +2216,19 @@ xfs_attr3_leaf_getvalue(
 		name_rmt = xfs_attr3_leaf_name_remote(leaf, args->index);
 		ASSERT(name_rmt->namelen == args->namelen);
 		ASSERT(memcmp(args->name, name_rmt->name, args->namelen) == 0);
-		valuelen = be32_to_cpu(name_rmt->valuelen);
+		args->rmtvaluelen = be32_to_cpu(name_rmt->valuelen);
 		args->rmtblkno = be32_to_cpu(name_rmt->valueblk);
 		args->rmtblkcnt = xfs_attr3_rmt_blocks(args->dp->i_mount,
-						       valuelen);
+						       args->rmtvaluelen);
 		if (args->flags & ATTR_KERNOVAL) {
-			args->valuelen = valuelen;
+			args->valuelen = args->rmtvaluelen;
 			return 0;
 		}
-		if (args->valuelen < valuelen) {
-			args->valuelen = valuelen;
-			return XFS_ERROR(ERANGE);
+		if (args->valuelen < args->rmtvaluelen) {
+			args->valuelen = args->rmtvaluelen;
+			return -ERANGE;
 		}
-		args->valuelen = valuelen;
+		args->valuelen = args->rmtvaluelen;
 	}
 	return 0;
 }
@@ -2247,14 +2244,14 @@ xfs_attr3_leaf_getvalue(
 /*ARGSUSED*/
 STATIC void
 xfs_attr3_leaf_moveents(
+	struct xfs_da_args		*args,
 	struct xfs_attr_leafblock	*leaf_s,
 	struct xfs_attr3_icleaf_hdr	*ichdr_s,
 	int				start_s,
 	struct xfs_attr_leafblock	*leaf_d,
 	struct xfs_attr3_icleaf_hdr	*ichdr_d,
 	int				start_d,
-	int				count,
-	struct xfs_mount		*mp)
+	int				count)
 {
 	struct xfs_attr_leaf_entry	*entry_s;
 	struct xfs_attr_leaf_entry	*entry_d;
@@ -2274,10 +2271,10 @@ xfs_attr3_leaf_moveents(
 	ASSERT(ichdr_s->magic == XFS_ATTR_LEAF_MAGIC ||
 	       ichdr_s->magic == XFS_ATTR3_LEAF_MAGIC);
 	ASSERT(ichdr_s->magic == ichdr_d->magic);
-	ASSERT(ichdr_s->count > 0 && ichdr_s->count < XFS_LBSIZE(mp) / 8);
+	ASSERT(ichdr_s->count > 0 && ichdr_s->count < args->geo->blksize / 8);
 	ASSERT(ichdr_s->firstused >= (ichdr_s->count * sizeof(*entry_s))
 					+ xfs_attr3_leaf_hdr_size(leaf_s));
-	ASSERT(ichdr_d->count < XFS_LBSIZE(mp) / 8);
+	ASSERT(ichdr_d->count < args->geo->blksize / 8);
 	ASSERT(ichdr_d->firstused >= (ichdr_d->count * sizeof(*entry_d))
 					+ xfs_attr3_leaf_hdr_size(leaf_d));
 
@@ -2329,11 +2326,11 @@ xfs_attr3_leaf_moveents(
 			entry_d->nameidx = cpu_to_be16(ichdr_d->firstused);
 			entry_d->flags = entry_s->flags;
 			ASSERT(be16_to_cpu(entry_d->nameidx) + tmp
-							<= XFS_LBSIZE(mp));
+							<= args->geo->blksize);
 			memmove(xfs_attr3_leaf_name(leaf_d, desti),
 				xfs_attr3_leaf_name(leaf_s, start_s + i), tmp);
 			ASSERT(be16_to_cpu(entry_s->nameidx) + tmp
-							<= XFS_LBSIZE(mp));
+							<= args->geo->blksize);
 			memset(xfs_attr3_leaf_name(leaf_s, start_s + i), 0, tmp);
 			ichdr_s->usedbytes -= tmp;
 			ichdr_d->usedbytes += tmp;
@@ -2354,7 +2351,7 @@ xfs_attr3_leaf_moveents(
 		tmp = count * sizeof(xfs_attr_leaf_entry_t);
 		entry_s = &xfs_attr3_leaf_entryp(leaf_s)[start_s];
 		ASSERT(((char *)entry_s + tmp) <=
-		       ((char *)leaf_s + XFS_LBSIZE(mp)));
+		       ((char *)leaf_s + args->geo->blksize));
 		memset(entry_s, 0, tmp);
 	} else {
 		/*
@@ -2369,7 +2366,7 @@ xfs_attr3_leaf_moveents(
 		tmp = count * sizeof(xfs_attr_leaf_entry_t);
 		entry_s = &xfs_attr3_leaf_entryp(leaf_s)[ichdr_s->count];
 		ASSERT(((char *)entry_s + tmp) <=
-		       ((char *)leaf_s + XFS_LBSIZE(mp)));
+		       ((char *)leaf_s + args->geo->blksize));
 		memset(entry_s, 0, tmp);
 	}
 
@@ -2437,22 +2434,21 @@ xfs_attr_leaf_entsize(xfs_attr_leafblock_t *leaf, int index)
  * a "local" or a "remote" attribute.
  */
 int
-xfs_attr_leaf_newentsize(int namelen, int valuelen, int blocksize, int *local)
+xfs_attr_leaf_newentsize(
+	struct xfs_da_args	*args,
+	int			*local)
 {
-	int size;
+	int			size;
 
-	size = xfs_attr_leaf_entsize_local(namelen, valuelen);
-	if (size < xfs_attr_leaf_entsize_local_max(blocksize)) {
-		if (local) {
+	size = xfs_attr_leaf_entsize_local(args->namelen, args->valuelen);
+	if (size < xfs_attr_leaf_entsize_local_max(args->geo->blksize)) {
+		if (local)
 			*local = 1;
-		}
-	} else {
-		size = xfs_attr_leaf_entsize_remote(namelen);
-		if (local) {
-			*local = 0;
-		}
+		return size;
 	}
-	return size;
+	if (local)
+		*local = 0;
+	return xfs_attr_leaf_entsize_remote(args->namelen);
 }
 
 
@@ -2485,7 +2481,7 @@ xfs_attr3_leaf_clearflag(
 	 */
 	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
 	if (error)
-		return(error);
+		return error;
 
 	leaf = bp->b_addr;
 	entry = &xfs_attr3_leaf_entryp(leaf)[args->index];
@@ -2518,7 +2514,7 @@ xfs_attr3_leaf_clearflag(
 		ASSERT((entry->flags & XFS_ATTR_LOCAL) == 0);
 		name_rmt = xfs_attr3_leaf_name_remote(leaf, args->index);
 		name_rmt->valueblk = cpu_to_be32(args->rmtblkno);
-		name_rmt->valuelen = cpu_to_be32(args->valuelen);
+		name_rmt->valuelen = cpu_to_be32(args->rmtvaluelen);
 		xfs_trans_log_buf(args->trans, bp,
 			 XFS_DA_LOGRANGE(leaf, name_rmt, sizeof(*name_rmt)));
 	}
@@ -2552,7 +2548,7 @@ xfs_attr3_leaf_setflag(
 	 */
 	error = xfs_attr3_leaf_read(args->trans, args->dp, args->blkno, -1, &bp);
 	if (error)
-		return(error);
+		return error;
 
 	leaf = bp->b_addr;
 #ifdef DEBUG
@@ -2676,7 +2672,7 @@ xfs_attr3_leaf_flipflags(
 		ASSERT((entry1->flags & XFS_ATTR_LOCAL) == 0);
 		name_rmt = xfs_attr3_leaf_name_remote(leaf1, args->index);
 		name_rmt->valueblk = cpu_to_be32(args->rmtblkno);
-		name_rmt->valuelen = cpu_to_be32(args->valuelen);
+		name_rmt->valuelen = cpu_to_be32(args->rmtvaluelen);
 		xfs_trans_log_buf(args->trans, bp1,
 			 XFS_DA_LOGRANGE(leaf1, name_rmt, sizeof(*name_rmt)));
 	}
diff --git a/fs/xfs/xfs_attr_leaf.h b/fs/xfs/libxfs/xfs_attr_leaf.h
index 3ec5ec0b8678..e2929da7c3ba 100644
--- a/fs/xfs/xfs_attr_leaf.h
+++ b/fs/xfs/libxfs/xfs_attr_leaf.h
@@ -96,8 +96,7 @@ int	xfs_attr3_root_inactive(struct xfs_trans **trans, struct xfs_inode *dp);
 xfs_dahash_t	xfs_attr_leaf_lasthash(struct xfs_buf *bp, int *count);
 int	xfs_attr_leaf_order(struct xfs_buf *leaf1_bp,
 				   struct xfs_buf *leaf2_bp);
-int	xfs_attr_leaf_newentsize(int namelen, int valuelen, int blocksize,
-					int *local);
+int	xfs_attr_leaf_newentsize(struct xfs_da_args *args, int *local);
 int	xfs_attr3_leaf_read(struct xfs_trans *tp, struct xfs_inode *dp,
 			xfs_dablk_t bno, xfs_daddr_t mappedbno,
 			struct xfs_buf **bpp);
diff --git a/fs/xfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c
index 739e0a52deda..7510ab8058a4 100644
--- a/fs/xfs/xfs_attr_remote.c
+++ b/fs/xfs/libxfs/xfs_attr_remote.c
@@ -68,7 +68,6 @@ xfs_attr3_rmt_blocks(
  */
 static bool
 xfs_attr3_rmt_hdr_ok(
-	struct xfs_mount	*mp,
 	void			*ptr,
 	xfs_ino_t		ino,
 	uint32_t		offset,
@@ -110,7 +109,7 @@ 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) > XATTR_SIZE_MAX)
 		return false;
 	if (rmt->rm_owner == 0)
 		return false;
@@ -125,8 +124,8 @@ xfs_attr3_rmt_read_verify(
 	struct xfs_mount *mp = bp->b_target->bt_mount;
 	char		*ptr;
 	int		len;
-	bool		corrupt = false;
 	xfs_daddr_t	bno;
+	int		blksize = mp->m_attr_geo->blksize;
 
 	/* no verification of non-crc buffers */
 	if (!xfs_sb_version_hascrc(&mp->m_sb))
@@ -135,27 +134,25 @@ xfs_attr3_rmt_read_verify(
 	ptr = bp->b_addr;
 	bno = bp->b_bn;
 	len = BBTOB(bp->b_length);
-	ASSERT(len >= XFS_LBSIZE(mp));
+	ASSERT(len >= blksize);
 
 	while (len > 0) {
-		if (!xfs_verify_cksum(ptr, XFS_LBSIZE(mp),
-				      XFS_ATTR3_RMT_CRC_OFF)) {
-			corrupt = true;
+		if (!xfs_verify_cksum(ptr, blksize, XFS_ATTR3_RMT_CRC_OFF)) {
+			xfs_buf_ioerror(bp, -EFSBADCRC);
 			break;
 		}
-		if (!xfs_attr3_rmt_verify(mp, ptr, XFS_LBSIZE(mp), bno)) {
-			corrupt = true;
+		if (!xfs_attr3_rmt_verify(mp, ptr, blksize, bno)) {
+			xfs_buf_ioerror(bp, -EFSCORRUPTED);
 			break;
 		}
-		len -= XFS_LBSIZE(mp);
-		ptr += XFS_LBSIZE(mp);
-		bno += mp->m_bsize;
+		len -= blksize;
+		ptr += blksize;
+		bno += BTOBB(blksize);
 	}
 
-	if (corrupt) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
-	} else
+	if (bp->b_error)
+		xfs_verifier_error(bp);
+	else
 		ASSERT(len == 0);
 }
 
@@ -168,6 +165,7 @@ xfs_attr3_rmt_write_verify(
 	char		*ptr;
 	int		len;
 	xfs_daddr_t	bno;
+	int		blksize = mp->m_attr_geo->blksize;
 
 	/* no verification of non-crc buffers */
 	if (!xfs_sb_version_hascrc(&mp->m_sb))
@@ -176,13 +174,12 @@ xfs_attr3_rmt_write_verify(
 	ptr = bp->b_addr;
 	bno = bp->b_bn;
 	len = BBTOB(bp->b_length);
-	ASSERT(len >= XFS_LBSIZE(mp));
+	ASSERT(len >= blksize);
 
 	while (len > 0) {
-		if (!xfs_attr3_rmt_verify(mp, ptr, XFS_LBSIZE(mp), bno)) {
-			XFS_CORRUPTION_ERROR(__func__,
-					    XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-			xfs_buf_ioerror(bp, EFSCORRUPTED);
+		if (!xfs_attr3_rmt_verify(mp, ptr, blksize, bno)) {
+			xfs_buf_ioerror(bp, -EFSCORRUPTED);
+			xfs_verifier_error(bp);
 			return;
 		}
 		if (bip) {
@@ -191,11 +188,11 @@ xfs_attr3_rmt_write_verify(
 			rmt = (struct xfs_attr3_rmt_hdr *)ptr;
 			rmt->rm_lsn = cpu_to_be64(bip->bli_item.li_lsn);
 		}
-		xfs_update_cksum(ptr, XFS_LBSIZE(mp), XFS_ATTR3_RMT_CRC_OFF);
+		xfs_update_cksum(ptr, blksize, XFS_ATTR3_RMT_CRC_OFF);
 
-		len -= XFS_LBSIZE(mp);
-		ptr += XFS_LBSIZE(mp);
-		bno += mp->m_bsize;
+		len -= blksize;
+		ptr += blksize;
+		bno += BTOBB(blksize);
 	}
 	ASSERT(len == 0);
 }
@@ -244,22 +241,23 @@ xfs_attr_rmtval_copyout(
 	char		*src = bp->b_addr;
 	xfs_daddr_t	bno = bp->b_bn;
 	int		len = BBTOB(bp->b_length);
+	int		blksize = mp->m_attr_geo->blksize;
 
-	ASSERT(len >= XFS_LBSIZE(mp));
+	ASSERT(len >= blksize);
 
 	while (len > 0 && *valuelen > 0) {
 		int hdr_size = 0;
-		int byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, XFS_LBSIZE(mp));
+		int byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, blksize);
 
 		byte_cnt = min(*valuelen, byte_cnt);
 
 		if (xfs_sb_version_hascrc(&mp->m_sb)) {
-			if (!xfs_attr3_rmt_hdr_ok(mp, src, ino, *offset,
+			if (!xfs_attr3_rmt_hdr_ok(src, ino, *offset,
 						  byte_cnt, bno)) {
 				xfs_alert(mp,
 "remote attribute header mismatch bno/off/len/owner (0x%llx/0x%x/Ox%x/0x%llx)",
 					bno, *offset, byte_cnt, ino);
-				return EFSCORRUPTED;
+				return -EFSCORRUPTED;
 			}
 			hdr_size = sizeof(struct xfs_attr3_rmt_hdr);
 		}
@@ -267,9 +265,9 @@ xfs_attr_rmtval_copyout(
 		memcpy(*dst, src + hdr_size, byte_cnt);
 
 		/* roll buffer forwards */
-		len -= XFS_LBSIZE(mp);
-		src += XFS_LBSIZE(mp);
-		bno += mp->m_bsize;
+		len -= blksize;
+		src += blksize;
+		bno += BTOBB(blksize);
 
 		/* roll attribute data forwards */
 		*valuelen -= byte_cnt;
@@ -291,12 +289,13 @@ xfs_attr_rmtval_copyin(
 	char		*dst = bp->b_addr;
 	xfs_daddr_t	bno = bp->b_bn;
 	int		len = BBTOB(bp->b_length);
+	int		blksize = mp->m_attr_geo->blksize;
 
-	ASSERT(len >= XFS_LBSIZE(mp));
+	ASSERT(len >= blksize);
 
 	while (len > 0 && *valuelen > 0) {
 		int hdr_size;
-		int byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, XFS_LBSIZE(mp));
+		int byte_cnt = XFS_ATTR3_RMT_BUF_SPACE(mp, blksize);
 
 		byte_cnt = min(*valuelen, byte_cnt);
 		hdr_size = xfs_attr3_rmt_hdr_set(mp, dst, ino, *offset,
@@ -308,17 +307,17 @@ xfs_attr_rmtval_copyin(
 		 * If this is the last block, zero the remainder of it.
 		 * Check that we are actually the last block, too.
 		 */
-		if (byte_cnt + hdr_size < XFS_LBSIZE(mp)) {
+		if (byte_cnt + hdr_size < blksize) {
 			ASSERT(*valuelen - byte_cnt == 0);
-			ASSERT(len == XFS_LBSIZE(mp));
+			ASSERT(len == blksize);
 			memset(dst + hdr_size + byte_cnt, 0,
-					XFS_LBSIZE(mp) - hdr_size - byte_cnt);
+					blksize - hdr_size - byte_cnt);
 		}
 
 		/* roll buffer forwards */
-		len -= XFS_LBSIZE(mp);
-		dst += XFS_LBSIZE(mp);
-		bno += mp->m_bsize;
+		len -= blksize;
+		dst += blksize;
+		bno += BTOBB(blksize);
 
 		/* roll attribute data forwards */
 		*valuelen -= byte_cnt;
@@ -340,7 +339,7 @@ xfs_attr_rmtval_get(
 	struct xfs_buf		*bp;
 	xfs_dablk_t		lblkno = args->rmtblkno;
 	__uint8_t		*dst = args->value;
-	int			valuelen = args->valuelen;
+	int			valuelen;
 	int			nmap;
 	int			error;
 	int			blkcnt = args->rmtblkcnt;
@@ -350,7 +349,9 @@ xfs_attr_rmtval_get(
 	trace_xfs_attr_rmtval_get(args);
 
 	ASSERT(!(args->flags & ATTR_KERNOVAL));
+	ASSERT(args->rmtvaluelen == args->valuelen);
 
+	valuelen = args->rmtvaluelen;
 	while (valuelen > 0) {
 		nmap = ATTR_RMTVALUE_MAPSIZE;
 		error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno,
@@ -418,7 +419,7 @@ xfs_attr_rmtval_set(
 	 * attributes have headers, we can't just do a straight byte to FSB
 	 * conversion and have to take the header space into account.
 	 */
-	blkcnt = xfs_attr3_rmt_blocks(mp, args->valuelen);
+	blkcnt = xfs_attr3_rmt_blocks(mp, args->rmtvaluelen);
 	error = xfs_bmap_first_unused(args->trans, args->dp, blkcnt, &lfileoff,
 						   XFS_ATTR_FORK);
 	if (error)
@@ -451,7 +452,7 @@ xfs_attr_rmtval_set(
 			ASSERT(committed);
 			args->trans = NULL;
 			xfs_bmap_cancel(args->flist);
-			return(error);
+			return error;
 		}
 
 		/*
@@ -472,7 +473,7 @@ xfs_attr_rmtval_set(
 		 */
 		error = xfs_trans_roll(&args->trans, dp);
 		if (error)
-			return (error);
+			return error;
 	}
 
 	/*
@@ -483,7 +484,7 @@ xfs_attr_rmtval_set(
 	 */
 	lblkno = args->rmtblkno;
 	blkcnt = args->rmtblkcnt;
-	valuelen = args->valuelen;
+	valuelen = args->rmtvaluelen;
 	while (valuelen > 0) {
 		struct xfs_buf	*bp;
 		xfs_daddr_t	dblkno;
@@ -497,7 +498,7 @@ xfs_attr_rmtval_set(
 				       blkcnt, &map, &nmap,
 				       XFS_BMAPI_ATTRFORK);
 		if (error)
-			return(error);
+			return error;
 		ASSERT(nmap == 1);
 		ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
 		       (map.br_startblock != HOLESTARTBLOCK));
@@ -507,7 +508,7 @@ xfs_attr_rmtval_set(
 
 		bp = xfs_buf_get(mp->m_ddev_targp, dblkno, dblkcnt, 0);
 		if (!bp)
-			return ENOMEM;
+			return -ENOMEM;
 		bp->b_ops = &xfs_attr3_rmt_buf_ops;
 
 		xfs_attr_rmtval_copyin(mp, bp, args->dp->i_ino, &offset,
@@ -562,7 +563,7 @@ xfs_attr_rmtval_remove(
 		error = xfs_bmapi_read(args->dp, (xfs_fileoff_t)lblkno,
 				       blkcnt, &map, &nmap, XFS_BMAPI_ATTRFORK);
 		if (error)
-			return(error);
+			return error;
 		ASSERT(nmap == 1);
 		ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
 		       (map.br_startblock != HOLESTARTBLOCK));
@@ -621,7 +622,7 @@ xfs_attr_rmtval_remove(
 		 */
 		error = xfs_trans_roll(&args->trans, args->dp);
 		if (error)
-			return (error);
+			return error;
 	}
-	return(0);
+	return 0;
 }
diff --git a/fs/xfs/xfs_attr_remote.h b/fs/xfs/libxfs/xfs_attr_remote.h
index 5a9acfa156d7..5a9acfa156d7 100644
--- a/fs/xfs/xfs_attr_remote.h
+++ b/fs/xfs/libxfs/xfs_attr_remote.h
diff --git a/fs/xfs/xfs_attr_sf.h b/fs/xfs/libxfs/xfs_attr_sf.h
index 919756e3ba53..919756e3ba53 100644
--- a/fs/xfs/xfs_attr_sf.h
+++ b/fs/xfs/libxfs/xfs_attr_sf.h
diff --git a/fs/xfs/xfs_bit.h b/fs/xfs/libxfs/xfs_bit.h
index f1e3c907044d..e1649c0d3e02 100644
--- a/fs/xfs/xfs_bit.h
+++ b/fs/xfs/libxfs/xfs_bit.h
@@ -66,8 +66,11 @@ static inline int xfs_lowbit64(__uint64_t v)
 		n = ffs(w);
 	} else {	/* upper bits */
 		w = (__uint32_t)(v >> 32);
-		if (w && (n = ffs(w)))
-		n += 32;
+		if (w) {
+			n = ffs(w);
+			if (n)
+				n += 32;
+		}
 	}
 	return n - 1;
 }
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 3ef11b22e750..86df952d3e24 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -94,7 +94,7 @@ xfs_bmap_compute_maxlevels(
 		maxleafents = MAXAEXTNUM;
 		sz = XFS_BMDR_SPACE_CALC(MINABTPTRS);
 	}
-	maxrootrecs = xfs_bmdr_maxrecs(mp, sz, 0);
+	maxrootrecs = xfs_bmdr_maxrecs(sz, 0);
 	minleafrecs = mp->m_bmap_dmnr[0];
 	minnoderecs = mp->m_bmap_dmnr[1];
 	maxblocks = (maxleafents + minleafrecs - 1) / minleafrecs;
@@ -233,7 +233,6 @@ xfs_default_attroffset(
  */
 STATIC void
 xfs_bmap_forkoff_reset(
-	xfs_mount_t	*mp,
 	xfs_inode_t	*ip,
 	int		whichfork)
 {
@@ -393,7 +392,7 @@ xfs_bmap_check_leaf_extents(
 	pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
 	bno = be64_to_cpu(*pp);
 
-	ASSERT(bno != NULLDFSBNO);
+	ASSERT(bno != NULLFSBLOCK);
 	ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
 	ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
 
@@ -905,7 +904,7 @@ xfs_bmap_local_to_extents_empty(
 	ASSERT(ifp->if_bytes == 0);
 	ASSERT(XFS_IFORK_NEXTENTS(ip, whichfork) == 0);
 
-	xfs_bmap_forkoff_reset(ip->i_mount, ip, whichfork);
+	xfs_bmap_forkoff_reset(ip, whichfork);
 	ifp->if_flags &= ~XFS_IFINLINE;
 	ifp->if_flags |= XFS_IFEXTENTS;
 	XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS);
@@ -1034,7 +1033,7 @@ xfs_bmap_add_attrfork_btree(
 			goto error0;
 		if (stat == 0) {
 			xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		}
 		*firstblock = cur->bc_private.b.firstblock;
 		cur->bc_private.b.allocated = 0;
@@ -1099,10 +1098,11 @@ xfs_bmap_add_attrfork_local(
 
 	if (S_ISDIR(ip->i_d.di_mode)) {
 		memset(&dargs, 0, sizeof(dargs));
+		dargs.geo = ip->i_mount->m_dir_geo;
 		dargs.dp = ip;
 		dargs.firstblock = firstblock;
 		dargs.flist = flist;
-		dargs.total = ip->i_mount->m_dirblkfsbs;
+		dargs.total = dargs.geo->fsbcount;
 		dargs.whichfork = XFS_DATA_FORK;
 		dargs.trans = tp;
 		return xfs_dir2_sf_to_block(&dargs);
@@ -1115,7 +1115,7 @@ xfs_bmap_add_attrfork_local(
 
 	/* should only be called for types that support local format data */
 	ASSERT(0);
-	return EFSCORRUPTED;
+	return -EFSCORRUPTED;
 }
 
 /*
@@ -1192,7 +1192,7 @@ xfs_bmap_add_attrfork(
 		break;
 	default:
 		ASSERT(0);
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto trans_cancel;
 	}
 
@@ -1299,7 +1299,7 @@ xfs_bmap_read_extents(
 	ASSERT(level > 0);
 	pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
 	bno = be64_to_cpu(*pp);
-	ASSERT(bno != NULLDFSBNO);
+	ASSERT(bno != NULLFSBLOCK);
 	ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
 	ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
 	/*
@@ -1399,7 +1399,7 @@ xfs_bmap_read_extents(
 	return 0;
 error0:
 	xfs_trans_brelse(tp, bp);
-	return XFS_ERROR(EFSCORRUPTED);
+	return -EFSCORRUPTED;
 }
 
 
@@ -1429,11 +1429,7 @@ xfs_bmap_search_multi_extents(
 	gotp->br_startoff = 0xffa5a5a5a5a5a5a5LL;
 	gotp->br_blockcount = 0xa55a5a5a5a5a5a5aLL;
 	gotp->br_state = XFS_EXT_INVALID;
-#if XFS_BIG_BLKNOS
 	gotp->br_startblock = 0xffffa5a5a5a5a5a5LL;
-#else
-	gotp->br_startblock = 0xffffa5a5;
-#endif
 	prevp->br_startoff = NULLFILEOFF;
 
 	ep = xfs_iext_bno_to_ext(ifp, bno, &lastx);
@@ -1576,7 +1572,7 @@ xfs_bmap_last_before(
 	if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE &&
 	    XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
 	    XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_LOCAL)
-	       return XFS_ERROR(EIO);
+	       return -EIO;
 	if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_LOCAL) {
 		*last_block = 0;
 		return 0;
@@ -1635,7 +1631,7 @@ xfs_bmap_last_extent(
  * blocks at the end of the file which do not start at the previous data block,
  * we will try to align the new blocks at stripe unit boundaries.
  *
- * Returns 0 in bma->aeof if the file (fork) is empty as any new write will be
+ * Returns 1 in bma->aeof if the file (fork) is empty as any new write will be
  * at, or past the EOF.
  */
 STATIC int
@@ -1650,9 +1646,14 @@ xfs_bmap_isaeof(
 	bma->aeof = 0;
 	error = xfs_bmap_last_extent(NULL, bma->ip, whichfork, &rec,
 				     &is_empty);
-	if (error || is_empty)
+	if (error)
 		return error;
 
+	if (is_empty) {
+		bma->aeof = 1;
+		return 0;
+	}
+
 	/*
 	 * Check if we are allocation or past the last extent, or at least into
 	 * the last delayed allocated extent.
@@ -1670,7 +1671,6 @@ xfs_bmap_isaeof(
  */
 int
 xfs_bmap_last_offset(
-	struct xfs_trans	*tp,
 	struct xfs_inode	*ip,
 	xfs_fileoff_t		*last_block,
 	int			whichfork)
@@ -1686,7 +1686,7 @@ xfs_bmap_last_offset(
 
 	if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE &&
 	    XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)
-	       return XFS_ERROR(EIO);
+	       return -EIO;
 
 	error = xfs_bmap_last_extent(NULL, ip, whichfork, &rec, &is_empty);
 	if (error || is_empty)
@@ -3319,7 +3319,7 @@ xfs_bmap_extsize_align(
 		if (orig_off < align_off ||
 		    orig_end > align_off + align_alen ||
 		    align_alen - temp < orig_alen)
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		/*
 		 * Try to fix it by moving the start up.
 		 */
@@ -3344,7 +3344,7 @@ xfs_bmap_extsize_align(
 		 * Result doesn't cover the request, fail it.
 		 */
 		if (orig_off < align_off || orig_end > align_off + align_alen)
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 	} else {
 		ASSERT(orig_off >= align_off);
 		ASSERT(orig_end <= align_off + align_alen);
@@ -3512,6 +3512,67 @@ xfs_bmap_adjacent(
 #undef ISVALID
 }
 
+static int
+xfs_bmap_longest_free_extent(
+	struct xfs_trans	*tp,
+	xfs_agnumber_t		ag,
+	xfs_extlen_t		*blen,
+	int			*notinit)
+{
+	struct xfs_mount	*mp = tp->t_mountp;
+	struct xfs_perag	*pag;
+	xfs_extlen_t		longest;
+	int			error = 0;
+
+	pag = xfs_perag_get(mp, ag);
+	if (!pag->pagf_init) {
+		error = xfs_alloc_pagf_init(mp, tp, ag, XFS_ALLOC_FLAG_TRYLOCK);
+		if (error)
+			goto out;
+
+		if (!pag->pagf_init) {
+			*notinit = 1;
+			goto out;
+		}
+	}
+
+	longest = xfs_alloc_longest_free_extent(mp, pag);
+	if (*blen < longest)
+		*blen = longest;
+
+out:
+	xfs_perag_put(pag);
+	return error;
+}
+
+static void
+xfs_bmap_select_minlen(
+	struct xfs_bmalloca	*ap,
+	struct xfs_alloc_arg	*args,
+	xfs_extlen_t		*blen,
+	int			notinit)
+{
+	if (notinit || *blen < ap->minlen) {
+		/*
+		 * Since we did a BUF_TRYLOCK above, it is possible that
+		 * there is space for this request.
+		 */
+		args->minlen = ap->minlen;
+	} else if (*blen < args->maxlen) {
+		/*
+		 * If the best seen length is less than the request length,
+		 * use the best as the minimum.
+		 */
+		args->minlen = *blen;
+	} else {
+		/*
+		 * Otherwise we've seen an extent as big as maxlen, use that
+		 * as the minimum.
+		 */
+		args->minlen = args->maxlen;
+	}
+}
+
 STATIC int
 xfs_bmap_btalloc_nullfb(
 	struct xfs_bmalloca	*ap,
@@ -3519,111 +3580,74 @@ xfs_bmap_btalloc_nullfb(
 	xfs_extlen_t		*blen)
 {
 	struct xfs_mount	*mp = ap->ip->i_mount;
-	struct xfs_perag	*pag;
 	xfs_agnumber_t		ag, startag;
 	int			notinit = 0;
 	int			error;
 
-	if (ap->userdata && xfs_inode_is_filestream(ap->ip))
-		args->type = XFS_ALLOCTYPE_NEAR_BNO;
-	else
-		args->type = XFS_ALLOCTYPE_START_BNO;
+	args->type = XFS_ALLOCTYPE_START_BNO;
 	args->total = ap->total;
 
-	/*
-	 * Search for an allocation group with a single extent large enough
-	 * for the request.  If one isn't found, then adjust the minimum
-	 * allocation size to the largest space found.
-	 */
 	startag = ag = XFS_FSB_TO_AGNO(mp, args->fsbno);
 	if (startag == NULLAGNUMBER)
 		startag = ag = 0;
 
-	pag = xfs_perag_get(mp, ag);
 	while (*blen < args->maxlen) {
-		if (!pag->pagf_init) {
-			error = xfs_alloc_pagf_init(mp, args->tp, ag,
-						    XFS_ALLOC_FLAG_TRYLOCK);
-			if (error) {
-				xfs_perag_put(pag);
-				return error;
-			}
-		}
-
-		/*
-		 * See xfs_alloc_fix_freelist...
-		 */
-		if (pag->pagf_init) {
-			xfs_extlen_t	longest;
-			longest = xfs_alloc_longest_free_extent(mp, pag);
-			if (*blen < longest)
-				*blen = longest;
-		} else
-			notinit = 1;
-
-		if (xfs_inode_is_filestream(ap->ip)) {
-			if (*blen >= args->maxlen)
-				break;
-
-			if (ap->userdata) {
-				/*
-				 * If startag is an invalid AG, we've
-				 * come here once before and
-				 * xfs_filestream_new_ag picked the
-				 * best currently available.
-				 *
-				 * Don't continue looping, since we
-				 * could loop forever.
-				 */
-				if (startag == NULLAGNUMBER)
-					break;
-
-				error = xfs_filestream_new_ag(ap, &ag);
-				xfs_perag_put(pag);
-				if (error)
-					return error;
+		error = xfs_bmap_longest_free_extent(args->tp, ag, blen,
+						     &notinit);
+		if (error)
+			return error;
 
-				/* loop again to set 'blen'*/
-				startag = NULLAGNUMBER;
-				pag = xfs_perag_get(mp, ag);
-				continue;
-			}
-		}
 		if (++ag == mp->m_sb.sb_agcount)
 			ag = 0;
 		if (ag == startag)
 			break;
-		xfs_perag_put(pag);
-		pag = xfs_perag_get(mp, ag);
 	}
-	xfs_perag_put(pag);
 
-	/*
-	 * Since the above loop did a BUF_TRYLOCK, it is
-	 * possible that there is space for this request.
-	 */
-	if (notinit || *blen < ap->minlen)
-		args->minlen = ap->minlen;
-	/*
-	 * If the best seen length is less than the request
-	 * length, use the best as the minimum.
-	 */
-	else if (*blen < args->maxlen)
-		args->minlen = *blen;
-	/*
-	 * Otherwise we've seen an extent as big as maxlen,
-	 * use that as the minimum.
-	 */
-	else
-		args->minlen = args->maxlen;
+	xfs_bmap_select_minlen(ap, args, blen, notinit);
+	return 0;
+}
+
+STATIC int
+xfs_bmap_btalloc_filestreams(
+	struct xfs_bmalloca	*ap,
+	struct xfs_alloc_arg	*args,
+	xfs_extlen_t		*blen)
+{
+	struct xfs_mount	*mp = ap->ip->i_mount;
+	xfs_agnumber_t		ag;
+	int			notinit = 0;
+	int			error;
+
+	args->type = XFS_ALLOCTYPE_NEAR_BNO;
+	args->total = ap->total;
+
+	ag = XFS_FSB_TO_AGNO(mp, args->fsbno);
+	if (ag == NULLAGNUMBER)
+		ag = 0;
+
+	error = xfs_bmap_longest_free_extent(args->tp, ag, blen, &notinit);
+	if (error)
+		return error;
+
+	if (*blen < args->maxlen) {
+		error = xfs_filestream_new_ag(ap, &ag);
+		if (error)
+			return error;
+
+		error = xfs_bmap_longest_free_extent(args->tp, ag, blen,
+						     &notinit);
+		if (error)
+			return error;
+
+	}
+
+	xfs_bmap_select_minlen(ap, args, blen, notinit);
 
 	/*
-	 * set the failure fallback case to look in the selected
-	 * AG as the stream may have moved.
+	 * Set the failure fallback case to look in the selected AG as stream
+	 * may have moved.
 	 */
-	if (xfs_inode_is_filestream(ap->ip))
-		ap->blkno = args->fsbno = XFS_AGB_TO_FSB(mp, ag, 0);
-
+	ap->blkno = args->fsbno = XFS_AGB_TO_FSB(mp, ag, 0);
 	return 0;
 }
 
@@ -3643,10 +3667,19 @@ xfs_bmap_btalloc(
 	int		isaligned;
 	int		tryagain;
 	int		error;
+	int		stripe_align;
 
 	ASSERT(ap->length);
 
 	mp = ap->ip->i_mount;
+
+	/* stripe alignment for allocation is determined by mount parameters */
+	stripe_align = 0;
+	if (mp->m_swidth && (mp->m_flags & XFS_MOUNT_SWALLOC))
+		stripe_align = mp->m_swidth;
+	else if (mp->m_dalign)
+		stripe_align = mp->m_dalign;
+
 	align = ap->userdata ? xfs_get_extsz_hint(ap->ip) : 0;
 	if (unlikely(align)) {
 		error = xfs_bmap_extsize_align(mp, &ap->got, &ap->prev,
@@ -3655,6 +3688,8 @@ xfs_bmap_btalloc(
 		ASSERT(!error);
 		ASSERT(ap->length);
 	}
+
+
 	nullfb = *ap->firstblock == NULLFSBLOCK;
 	fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, *ap->firstblock);
 	if (nullfb) {
@@ -3692,7 +3727,15 @@ xfs_bmap_btalloc(
 	args.firstblock = *ap->firstblock;
 	blen = 0;
 	if (nullfb) {
-		error = xfs_bmap_btalloc_nullfb(ap, &args, &blen);
+		/*
+		 * Search for an allocation group with a single extent large
+		 * enough for the request.  If one isn't found, then adjust
+		 * the minimum allocation size to the largest space found.
+		 */
+		if (ap->userdata && xfs_inode_is_filestream(ap->ip))
+			error = xfs_bmap_btalloc_filestreams(ap, &args, &blen);
+		else
+			error = xfs_bmap_btalloc_nullfb(ap, &args, &blen);
 		if (error)
 			return error;
 	} else if (ap->flist->xbf_low) {
@@ -3730,7 +3773,7 @@ xfs_bmap_btalloc(
 	 */
 	if (!ap->flist->xbf_low && ap->aeof) {
 		if (!ap->offset) {
-			args.alignment = mp->m_dalign;
+			args.alignment = stripe_align;
 			atype = args.type;
 			isaligned = 1;
 			/*
@@ -3755,13 +3798,13 @@ xfs_bmap_btalloc(
 			 * of minlen+alignment+slop doesn't go up
 			 * between the calls.
 			 */
-			if (blen > mp->m_dalign && blen <= args.maxlen)
-				nextminlen = blen - mp->m_dalign;
+			if (blen > stripe_align && blen <= args.maxlen)
+				nextminlen = blen - stripe_align;
 			else
 				nextminlen = args.minlen;
-			if (nextminlen + mp->m_dalign > args.minlen + 1)
+			if (nextminlen + stripe_align > args.minlen + 1)
 				args.minalignslop =
-					nextminlen + mp->m_dalign -
+					nextminlen + stripe_align -
 					args.minlen - 1;
 			else
 				args.minalignslop = 0;
@@ -3783,7 +3826,7 @@ xfs_bmap_btalloc(
 		 */
 		args.type = atype;
 		args.fsbno = ap->blkno;
-		args.alignment = mp->m_dalign;
+		args.alignment = stripe_align;
 		args.minlen = nextminlen;
 		args.minalignslop = 0;
 		isaligned = 1;
@@ -3997,17 +4040,18 @@ xfs_bmapi_read(
 	ASSERT(*nmap >= 1);
 	ASSERT(!(flags & ~(XFS_BMAPI_ATTRFORK|XFS_BMAPI_ENTIRE|
 			   XFS_BMAPI_IGSTATE)));
+	ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED|XFS_ILOCK_EXCL));
 
 	if (unlikely(XFS_TEST_ERROR(
 	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
 	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
 	     mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) {
 		XFS_ERROR_REPORT("xfs_bmapi_read", XFS_ERRLEVEL_LOW, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	XFS_STATS_INC(xs_blk_mapr);
 
@@ -4191,17 +4235,18 @@ xfs_bmapi_delay(
 	ASSERT(*nmap >= 1);
 	ASSERT(*nmap <= XFS_BMAP_MAX_NMAP);
 	ASSERT(!(flags & ~XFS_BMAPI_ENTIRE));
+	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
 	if (unlikely(XFS_TEST_ERROR(
 	    (XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_EXTENTS &&
 	     XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_BTREE),
 	     mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) {
 		XFS_ERROR_REPORT("xfs_bmapi_delay", XFS_ERRLEVEL_LOW, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	XFS_STATS_INC(xs_blk_mapw);
 
@@ -4249,8 +4294,8 @@ xfs_bmapi_delay(
 }
 
 
-int
-__xfs_bmapi_allocate(
+static int
+xfs_bmapi_allocate(
 	struct xfs_bmalloca	*bma)
 {
 	struct xfs_mount	*mp = bma->ip->i_mount;
@@ -4420,7 +4465,7 @@ xfs_bmapi_convert_unwritten(
 	 * so generate another request.
 	 */
 	if (mval->br_blockcount < len)
-		return EAGAIN;
+		return -EAGAIN;
 	return 0;
 }
 
@@ -4484,17 +4529,18 @@ xfs_bmapi_write(
 	ASSERT(tp != NULL);
 	ASSERT(len > 0);
 	ASSERT(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_LOCAL);
+	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
 	if (unlikely(XFS_TEST_ERROR(
 	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
 	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
 	     mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) {
 		XFS_ERROR_REPORT("xfs_bmapi_write", XFS_ERRLEVEL_LOW, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	ifp = XFS_IFORK_PTR(ip, whichfork);
 
@@ -4528,9 +4574,6 @@ xfs_bmapi_write(
 	bma.flist = flist;
 	bma.firstblock = firstblock;
 
-	if (flags & XFS_BMAPI_STACK_SWITCH)
-		bma.stack_switch = 1;
-
 	while (bno < end && n < *nmap) {
 		inhole = eof || bma.got.br_startoff > bno;
 		wasdelay = !inhole && isnullstartblock(bma.got.br_startblock);
@@ -4573,7 +4616,7 @@ xfs_bmapi_write(
 
 		/* Execute unwritten extent conversion if necessary */
 		error = xfs_bmapi_convert_unwritten(&bma, mval, len, flags);
-		if (error == EAGAIN)
+		if (error == -EAGAIN)
 			continue;
 		if (error)
 			goto error0;
@@ -4875,7 +4918,7 @@ xfs_bmap_del_extent(
 					goto done;
 				cur->bc_rec.b = new;
 				error = xfs_btree_insert(cur, &i);
-				if (error && error != ENOSPC)
+				if (error && error != -ENOSPC)
 					goto done;
 				/*
 				 * If get no-space back from btree insert,
@@ -4883,7 +4926,7 @@ xfs_bmap_del_extent(
 				 * block reservation.
 				 * Fix up our state and return the error.
 				 */
-				if (error == ENOSPC) {
+				if (error == -ENOSPC) {
 					/*
 					 * Reset the cursor, don't trust
 					 * it after any insert operation.
@@ -4911,7 +4954,7 @@ xfs_bmap_del_extent(
 					xfs_bmbt_set_blockcount(ep,
 						got.br_blockcount);
 					flags = 0;
-					error = XFS_ERROR(ENOSPC);
+					error = -ENOSPC;
 					goto done;
 				}
 				XFS_WANT_CORRUPTED_GOTO(i == 1, done);
@@ -5029,12 +5072,13 @@ xfs_bunmapi(
 	    XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) {
 		XFS_ERROR_REPORT("xfs_bunmapi", XFS_ERRLEVEL_LOW,
 				 ip->i_mount);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	mp = ip->i_mount;
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
+	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	ASSERT(len > 0);
 	ASSERT(nexts >= 0);
 
@@ -5277,7 +5321,7 @@ xfs_bunmapi(
 		    del.br_startoff > got.br_startoff &&
 		    del.br_startoff + del.br_blockcount <
 		    got.br_startoff + got.br_blockcount) {
-			error = XFS_ERROR(ENOSPC);
+			error = -ENOSPC;
 			goto error0;
 		}
 		error = xfs_bmap_del_extent(ip, tp, &lastx, flist, cur, &del,
@@ -5358,3 +5402,203 @@ error0:
 	}
 	return error;
 }
+
+/*
+ * Shift extent records to the left to cover a hole.
+ *
+ * The maximum number of extents to be shifted in a single operation
+ * is @num_exts, and @current_ext keeps track of the current extent
+ * index we have shifted. @offset_shift_fsb is the length by which each
+ * extent is shifted. If there is no hole to shift the extents
+ * into, this will be considered invalid operation and we abort immediately.
+ */
+int
+xfs_bmap_shift_extents(
+	struct xfs_trans	*tp,
+	struct xfs_inode	*ip,
+	int			*done,
+	xfs_fileoff_t		start_fsb,
+	xfs_fileoff_t		offset_shift_fsb,
+	xfs_extnum_t		*current_ext,
+	xfs_fsblock_t		*firstblock,
+	struct xfs_bmap_free	*flist,
+	int			num_exts)
+{
+	struct xfs_btree_cur		*cur = NULL;
+	struct xfs_bmbt_rec_host	*gotp;
+	struct xfs_bmbt_irec            got;
+	struct xfs_bmbt_irec		left;
+	struct xfs_mount		*mp = ip->i_mount;
+	struct xfs_ifork		*ifp;
+	xfs_extnum_t			nexts = 0;
+	xfs_fileoff_t			startoff;
+	int				error = 0;
+	int				i;
+	int				whichfork = XFS_DATA_FORK;
+	int				logflags = 0;
+	xfs_filblks_t			blockcount = 0;
+	int				total_extents;
+
+	if (unlikely(XFS_TEST_ERROR(
+	    (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS &&
+	     XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE),
+	     mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) {
+		XFS_ERROR_REPORT("xfs_bmap_shift_extents",
+				 XFS_ERRLEVEL_LOW, mp);
+		return -EFSCORRUPTED;
+	}
+
+	if (XFS_FORCED_SHUTDOWN(mp))
+		return -EIO;
+
+	ASSERT(current_ext != NULL);
+
+	ifp = XFS_IFORK_PTR(ip, whichfork);
+	if (!(ifp->if_flags & XFS_IFEXTENTS)) {
+		/* Read in all the extents */
+		error = xfs_iread_extents(tp, ip, whichfork);
+		if (error)
+			return error;
+	}
+
+	/*
+	 * If *current_ext is 0, we would need to lookup the extent
+	 * from where we would start shifting and store it in gotp.
+	 */
+	if (!*current_ext) {
+		gotp = xfs_iext_bno_to_ext(ifp, start_fsb, current_ext);
+		/*
+		 * gotp can be null in 2 cases: 1) if there are no extents
+		 * or 2) start_fsb lies in a hole beyond which there are
+		 * no extents. Either way, we are done.
+		 */
+		if (!gotp) {
+			*done = 1;
+			return 0;
+		}
+	}
+
+	if (ifp->if_flags & XFS_IFBROOT) {
+		cur = xfs_bmbt_init_cursor(mp, tp, ip, whichfork);
+		cur->bc_private.b.firstblock = *firstblock;
+		cur->bc_private.b.flist = flist;
+		cur->bc_private.b.flags = 0;
+	}
+
+	/*
+	 * There may be delalloc extents in the data fork before the range we
+	 * are collapsing out, so we cannot
+	 * use the count of real extents here. Instead we have to calculate it
+	 * from the incore fork.
+	 */
+	total_extents = ifp->if_bytes / sizeof(xfs_bmbt_rec_t);
+	while (nexts++ < num_exts && *current_ext < total_extents) {
+
+		gotp = xfs_iext_get_ext(ifp, *current_ext);
+		xfs_bmbt_get_all(gotp, &got);
+		startoff = got.br_startoff - offset_shift_fsb;
+
+		/*
+		 * Before shifting extent into hole, make sure that the hole
+		 * is large enough to accomodate the shift.
+		 */
+		if (*current_ext) {
+			xfs_bmbt_get_all(xfs_iext_get_ext(ifp,
+						*current_ext - 1), &left);
+
+			if (startoff < left.br_startoff + left.br_blockcount)
+				error = -EINVAL;
+		} else if (offset_shift_fsb > got.br_startoff) {
+			/*
+			 * When first extent is shifted, offset_shift_fsb
+			 * should be less than the stating offset of
+			 * the first extent.
+			 */
+			error = -EINVAL;
+		}
+
+		if (error)
+			goto del_cursor;
+
+		if (cur) {
+			error = xfs_bmbt_lookup_eq(cur, got.br_startoff,
+						   got.br_startblock,
+						   got.br_blockcount,
+						   &i);
+			if (error)
+				goto del_cursor;
+			XFS_WANT_CORRUPTED_GOTO(i == 1, del_cursor);
+		}
+
+		/* Check if we can merge 2 adjacent extents */
+		if (*current_ext &&
+		    left.br_startoff + left.br_blockcount == startoff &&
+		    left.br_startblock + left.br_blockcount ==
+				got.br_startblock &&
+		    left.br_state == got.br_state &&
+		    left.br_blockcount + got.br_blockcount <= MAXEXTLEN) {
+			blockcount = left.br_blockcount +
+				got.br_blockcount;
+			xfs_iext_remove(ip, *current_ext, 1, 0);
+			logflags |= XFS_ILOG_CORE;
+			if (cur) {
+				error = xfs_btree_delete(cur, &i);
+				if (error)
+					goto del_cursor;
+				XFS_WANT_CORRUPTED_GOTO(i == 1, del_cursor);
+			} else {
+				logflags |= XFS_ILOG_DEXT;
+			}
+			XFS_IFORK_NEXT_SET(ip, whichfork,
+				XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
+			gotp = xfs_iext_get_ext(ifp, --*current_ext);
+			xfs_bmbt_get_all(gotp, &got);
+
+			/* Make cursor point to the extent we will update */
+			if (cur) {
+				error = xfs_bmbt_lookup_eq(cur, got.br_startoff,
+							   got.br_startblock,
+							   got.br_blockcount,
+							   &i);
+				if (error)
+					goto del_cursor;
+				XFS_WANT_CORRUPTED_GOTO(i == 1, del_cursor);
+			}
+
+			xfs_bmbt_set_blockcount(gotp, blockcount);
+			got.br_blockcount = blockcount;
+		} else {
+			/* We have to update the startoff */
+			xfs_bmbt_set_startoff(gotp, startoff);
+			got.br_startoff = startoff;
+		}
+
+		logflags |= XFS_ILOG_CORE;
+		if (cur) {
+			error = xfs_bmbt_update(cur, got.br_startoff,
+						got.br_startblock,
+						got.br_blockcount,
+						got.br_state);
+			if (error)
+				goto del_cursor;
+		} else {
+			logflags |= XFS_ILOG_DEXT;
+		}
+
+		(*current_ext)++;
+		total_extents = ifp->if_bytes / sizeof(xfs_bmbt_rec_t);
+	}
+
+	/* Check if we are done */
+	if (*current_ext == total_extents)
+		*done = 1;
+
+del_cursor:
+	if (cur)
+		xfs_btree_del_cursor(cur,
+			error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
+
+	if (logflags)
+		xfs_trans_log_inode(tp, ip, logflags);
+	return error;
+}
diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h
index 33b41f351225..b879ca56a64c 100644
--- a/fs/xfs/xfs_bmap.h
+++ b/fs/xfs/libxfs/xfs_bmap.h
@@ -77,7 +77,6 @@ typedef	struct xfs_bmap_free
  * from written to unwritten, otherwise convert from unwritten to written.
  */
 #define XFS_BMAPI_CONVERT	0x040
-#define XFS_BMAPI_STACK_SWITCH	0x080
 
 #define XFS_BMAPI_FLAGS \
 	{ XFS_BMAPI_ENTIRE,	"ENTIRE" }, \
@@ -86,8 +85,7 @@ typedef	struct xfs_bmap_free
 	{ XFS_BMAPI_PREALLOC,	"PREALLOC" }, \
 	{ XFS_BMAPI_IGSTATE,	"IGSTATE" }, \
 	{ XFS_BMAPI_CONTIG,	"CONTIG" }, \
-	{ XFS_BMAPI_CONVERT,	"CONVERT" }, \
-	{ XFS_BMAPI_STACK_SWITCH, "STACK_SWITCH" }
+	{ XFS_BMAPI_CONVERT,	"CONVERT" }
 
 
 static inline int xfs_bmapi_aflag(int w)
@@ -127,6 +125,16 @@ static inline void xfs_bmap_init(xfs_bmap_free_t *flp, xfs_fsblock_t *fbp)
 	{ BMAP_RIGHT_FILLING,	"RF" }, \
 	{ BMAP_ATTRFORK,	"ATTR" }
 
+
+/*
+ * This macro is used to determine how many extents will be shifted
+ * in one write transaction. We could require two splits,
+ * an extent move on the first and an extent merge on the second,
+ * So it is proper that one extent is shifted inside write transaction
+ * at a time.
+ */
+#define XFS_BMAP_MAX_SHIFT_EXTENTS	1
+
 #ifdef DEBUG
 void	xfs_bmap_trace_exlist(struct xfs_inode *ip, xfs_extnum_t cnt,
 		int whichfork, unsigned long caller_ip);
@@ -146,8 +154,8 @@ int	xfs_bmap_first_unused(struct xfs_trans *tp, struct xfs_inode *ip,
 		xfs_extlen_t len, xfs_fileoff_t *unused, int whichfork);
 int	xfs_bmap_last_before(struct xfs_trans *tp, struct xfs_inode *ip,
 		xfs_fileoff_t *last_block, int whichfork);
-int	xfs_bmap_last_offset(struct xfs_trans *tp, struct xfs_inode *ip,
-		xfs_fileoff_t *unused, int whichfork);
+int	xfs_bmap_last_offset(struct xfs_inode *ip, xfs_fileoff_t *unused,
+		int whichfork);
 int	xfs_bmap_one_block(struct xfs_inode *ip, int whichfork);
 int	xfs_bmap_read_extents(struct xfs_trans *tp, struct xfs_inode *ip,
 		int whichfork);
@@ -169,5 +177,10 @@ int	xfs_bunmapi(struct xfs_trans *tp, struct xfs_inode *ip,
 int	xfs_check_nostate_extents(struct xfs_ifork *ifp, xfs_extnum_t idx,
 		xfs_extnum_t num);
 uint	xfs_default_attroffset(struct xfs_inode *ip);
+int	xfs_bmap_shift_extents(struct xfs_trans *tp, struct xfs_inode *ip,
+		int *done, xfs_fileoff_t start_fsb,
+		xfs_fileoff_t offset_shift_fsb, xfs_extnum_t *current_ext,
+		xfs_fsblock_t *firstblock, struct xfs_bmap_free	*flist,
+		int num_exts);
 
 #endif	/* __XFS_BMAP_H__ */
diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/libxfs/xfs_bmap_btree.c
index 706bc3f777cb..fba753308f31 100644
--- a/fs/xfs/xfs_bmap_btree.c
+++ b/fs/xfs/libxfs/xfs_bmap_btree.c
@@ -84,7 +84,7 @@ xfs_bmdr_to_bmbt(
 	rblock->bb_level = dblock->bb_level;
 	ASSERT(be16_to_cpu(rblock->bb_level) > 0);
 	rblock->bb_numrecs = dblock->bb_numrecs;
-	dmxr = xfs_bmdr_maxrecs(mp, dblocklen, 0);
+	dmxr = xfs_bmdr_maxrecs(dblocklen, 0);
 	fkp = XFS_BMDR_KEY_ADDR(dblock, 1);
 	tkp = XFS_BMBT_KEY_ADDR(mp, rblock, 1);
 	fpp = XFS_BMDR_PTR_ADDR(dblock, 1, dmxr);
@@ -111,23 +111,8 @@ __xfs_bmbt_get_all(
 	ext_flag = (int)(l0 >> (64 - BMBT_EXNTFLAG_BITLEN));
 	s->br_startoff = ((xfs_fileoff_t)l0 &
 			   xfs_mask64lo(64 - BMBT_EXNTFLAG_BITLEN)) >> 9;
-#if XFS_BIG_BLKNOS
 	s->br_startblock = (((xfs_fsblock_t)l0 & xfs_mask64lo(9)) << 43) |
 			   (((xfs_fsblock_t)l1) >> 21);
-#else
-#ifdef DEBUG
-	{
-		xfs_dfsbno_t	b;
-
-		b = (((xfs_dfsbno_t)l0 & xfs_mask64lo(9)) << 43) |
-		    (((xfs_dfsbno_t)l1) >> 21);
-		ASSERT((b >> 32) == 0 || isnulldstartblock(b));
-		s->br_startblock = (xfs_fsblock_t)b;
-	}
-#else	/* !DEBUG */
-	s->br_startblock = (xfs_fsblock_t)(((xfs_dfsbno_t)l1) >> 21);
-#endif	/* DEBUG */
-#endif	/* XFS_BIG_BLKNOS */
 	s->br_blockcount = (xfs_filblks_t)(l1 & xfs_mask64lo(21));
 	/* This is xfs_extent_state() in-line */
 	if (ext_flag) {
@@ -163,21 +148,8 @@ xfs_fsblock_t
 xfs_bmbt_get_startblock(
 	xfs_bmbt_rec_host_t	*r)
 {
-#if XFS_BIG_BLKNOS
 	return (((xfs_fsblock_t)r->l0 & xfs_mask64lo(9)) << 43) |
 	       (((xfs_fsblock_t)r->l1) >> 21);
-#else
-#ifdef DEBUG
-	xfs_dfsbno_t	b;
-
-	b = (((xfs_dfsbno_t)r->l0 & xfs_mask64lo(9)) << 43) |
-	    (((xfs_dfsbno_t)r->l1) >> 21);
-	ASSERT((b >> 32) == 0 || isnulldstartblock(b));
-	return (xfs_fsblock_t)b;
-#else	/* !DEBUG */
-	return (xfs_fsblock_t)(((xfs_dfsbno_t)r->l1) >> 21);
-#endif	/* DEBUG */
-#endif	/* XFS_BIG_BLKNOS */
 }
 
 /*
@@ -241,7 +213,6 @@ xfs_bmbt_set_allf(
 	ASSERT((startoff & xfs_mask64hi(64-BMBT_STARTOFF_BITLEN)) == 0);
 	ASSERT((blockcount & xfs_mask64hi(64-BMBT_BLOCKCOUNT_BITLEN)) == 0);
 
-#if XFS_BIG_BLKNOS
 	ASSERT((startblock & xfs_mask64hi(64-BMBT_STARTBLOCK_BITLEN)) == 0);
 
 	r->l0 = ((xfs_bmbt_rec_base_t)extent_flag << 63) |
@@ -250,23 +221,6 @@ xfs_bmbt_set_allf(
 	r->l1 = ((xfs_bmbt_rec_base_t)startblock << 21) |
 		((xfs_bmbt_rec_base_t)blockcount &
 		(xfs_bmbt_rec_base_t)xfs_mask64lo(21));
-#else	/* !XFS_BIG_BLKNOS */
-	if (isnullstartblock(startblock)) {
-		r->l0 = ((xfs_bmbt_rec_base_t)extent_flag << 63) |
-			((xfs_bmbt_rec_base_t)startoff << 9) |
-			 (xfs_bmbt_rec_base_t)xfs_mask64lo(9);
-		r->l1 = xfs_mask64hi(11) |
-			  ((xfs_bmbt_rec_base_t)startblock << 21) |
-			  ((xfs_bmbt_rec_base_t)blockcount &
-			   (xfs_bmbt_rec_base_t)xfs_mask64lo(21));
-	} else {
-		r->l0 = ((xfs_bmbt_rec_base_t)extent_flag << 63) |
-			((xfs_bmbt_rec_base_t)startoff << 9);
-		r->l1 = ((xfs_bmbt_rec_base_t)startblock << 21) |
-			 ((xfs_bmbt_rec_base_t)blockcount &
-			 (xfs_bmbt_rec_base_t)xfs_mask64lo(21));
-	}
-#endif	/* XFS_BIG_BLKNOS */
 }
 
 /*
@@ -298,8 +252,6 @@ xfs_bmbt_disk_set_allf(
 	ASSERT(state == XFS_EXT_NORM || state == XFS_EXT_UNWRITTEN);
 	ASSERT((startoff & xfs_mask64hi(64-BMBT_STARTOFF_BITLEN)) == 0);
 	ASSERT((blockcount & xfs_mask64hi(64-BMBT_BLOCKCOUNT_BITLEN)) == 0);
-
-#if XFS_BIG_BLKNOS
 	ASSERT((startblock & xfs_mask64hi(64-BMBT_STARTBLOCK_BITLEN)) == 0);
 
 	r->l0 = cpu_to_be64(
@@ -310,26 +262,6 @@ xfs_bmbt_disk_set_allf(
 		((xfs_bmbt_rec_base_t)startblock << 21) |
 		 ((xfs_bmbt_rec_base_t)blockcount &
 		  (xfs_bmbt_rec_base_t)xfs_mask64lo(21)));
-#else	/* !XFS_BIG_BLKNOS */
-	if (isnullstartblock(startblock)) {
-		r->l0 = cpu_to_be64(
-			((xfs_bmbt_rec_base_t)extent_flag << 63) |
-			 ((xfs_bmbt_rec_base_t)startoff << 9) |
-			  (xfs_bmbt_rec_base_t)xfs_mask64lo(9));
-		r->l1 = cpu_to_be64(xfs_mask64hi(11) |
-			  ((xfs_bmbt_rec_base_t)startblock << 21) |
-			  ((xfs_bmbt_rec_base_t)blockcount &
-			   (xfs_bmbt_rec_base_t)xfs_mask64lo(21)));
-	} else {
-		r->l0 = cpu_to_be64(
-			((xfs_bmbt_rec_base_t)extent_flag << 63) |
-			 ((xfs_bmbt_rec_base_t)startoff << 9));
-		r->l1 = cpu_to_be64(
-			((xfs_bmbt_rec_base_t)startblock << 21) |
-			 ((xfs_bmbt_rec_base_t)blockcount &
-			  (xfs_bmbt_rec_base_t)xfs_mask64lo(21)));
-	}
-#endif	/* XFS_BIG_BLKNOS */
 }
 
 /*
@@ -365,24 +297,11 @@ xfs_bmbt_set_startblock(
 	xfs_bmbt_rec_host_t *r,
 	xfs_fsblock_t	v)
 {
-#if XFS_BIG_BLKNOS
 	ASSERT((v & xfs_mask64hi(12)) == 0);
 	r->l0 = (r->l0 & (xfs_bmbt_rec_base_t)xfs_mask64hi(55)) |
 		  (xfs_bmbt_rec_base_t)(v >> 43);
 	r->l1 = (r->l1 & (xfs_bmbt_rec_base_t)xfs_mask64lo(21)) |
 		  (xfs_bmbt_rec_base_t)(v << 21);
-#else	/* !XFS_BIG_BLKNOS */
-	if (isnullstartblock(v)) {
-		r->l0 |= (xfs_bmbt_rec_base_t)xfs_mask64lo(9);
-		r->l1 = (xfs_bmbt_rec_base_t)xfs_mask64hi(11) |
-			  ((xfs_bmbt_rec_base_t)v << 21) |
-			  (r->l1 & (xfs_bmbt_rec_base_t)xfs_mask64lo(21));
-	} else {
-		r->l0 &= ~(xfs_bmbt_rec_base_t)xfs_mask64lo(9);
-		r->l1 = ((xfs_bmbt_rec_base_t)v << 21) |
-			  (r->l1 & (xfs_bmbt_rec_base_t)xfs_mask64lo(21));
-	}
-#endif	/* XFS_BIG_BLKNOS */
 }
 
 /*
@@ -438,12 +357,12 @@ xfs_bmbt_to_bmdr(
 		       cpu_to_be64(XFS_BUF_DADDR_NULL));
 	} else
 		ASSERT(rblock->bb_magic == cpu_to_be32(XFS_BMAP_MAGIC));
-	ASSERT(rblock->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO));
-	ASSERT(rblock->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO));
+	ASSERT(rblock->bb_u.l.bb_leftsib == cpu_to_be64(NULLFSBLOCK));
+	ASSERT(rblock->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK));
 	ASSERT(rblock->bb_level != 0);
 	dblock->bb_level = rblock->bb_level;
 	dblock->bb_numrecs = rblock->bb_numrecs;
-	dmxr = xfs_bmdr_maxrecs(mp, dblocklen, 0);
+	dmxr = xfs_bmdr_maxrecs(dblocklen, 0);
 	fkp = XFS_BMBT_KEY_ADDR(mp, rblock, 1);
 	tkp = XFS_BMDR_KEY_ADDR(dblock, 1);
 	fpp = XFS_BMAP_BROOT_PTR_ADDR(mp, rblock, 1, rblocklen);
@@ -519,7 +438,6 @@ xfs_bmbt_alloc_block(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_ptr	*start,
 	union xfs_btree_ptr	*new,
-	int			length,
 	int			*stat)
 {
 	xfs_alloc_arg_t		args;		/* block allocation args */
@@ -555,7 +473,7 @@ 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) {
-		error = XFS_ERROR(ENOSPC);
+		error = -ENOSPC;
 		goto error0;
 	}
 	error = xfs_alloc_vextent(&args);
@@ -672,8 +590,7 @@ xfs_bmbt_get_dmaxrecs(
 {
 	if (level != cur->bc_nlevels - 1)
 		return cur->bc_mp->m_bmap_dmxr[level != 0];
-	return xfs_bmdr_maxrecs(cur->bc_mp, cur->bc_private.b.forksize,
-				level == 0);
+	return xfs_bmdr_maxrecs(cur->bc_private.b.forksize, level == 0);
 }
 
 STATIC void
@@ -765,11 +682,11 @@ xfs_bmbt_verify(
 
 	/* sibling pointer verification */
 	if (!block->bb_u.l.bb_leftsib ||
-	    (block->bb_u.l.bb_leftsib != cpu_to_be64(NULLDFSBNO) &&
+	    (block->bb_u.l.bb_leftsib != cpu_to_be64(NULLFSBLOCK) &&
 	     !XFS_FSB_SANITY_CHECK(mp, be64_to_cpu(block->bb_u.l.bb_leftsib))))
 		return false;
 	if (!block->bb_u.l.bb_rightsib ||
-	    (block->bb_u.l.bb_rightsib != cpu_to_be64(NULLDFSBNO) &&
+	    (block->bb_u.l.bb_rightsib != cpu_to_be64(NULLFSBLOCK) &&
 	     !XFS_FSB_SANITY_CHECK(mp, be64_to_cpu(block->bb_u.l.bb_rightsib))))
 		return false;
 
@@ -780,12 +697,14 @@ static void
 xfs_bmbt_read_verify(
 	struct xfs_buf	*bp)
 {
-	if (!(xfs_btree_lblock_verify_crc(bp) &&
-	      xfs_bmbt_verify(bp))) {
+	if (!xfs_btree_lblock_verify_crc(bp))
+		xfs_buf_ioerror(bp, -EFSBADCRC);
+	else if (!xfs_bmbt_verify(bp))
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+
+	if (bp->b_error) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
-				     bp->b_target->bt_mount, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_verifier_error(bp);
 	}
 }
 
@@ -794,11 +713,9 @@ xfs_bmbt_write_verify(
 	struct xfs_buf	*bp)
 {
 	if (!xfs_bmbt_verify(bp)) {
-		xfs_warn(bp->b_target->bt_mount, "bmbt daddr 0x%llx failed", bp->b_bn);
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
-				     bp->b_target->bt_mount, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
 		return;
 	}
 	xfs_btree_lblock_calc_crc(bp);
@@ -914,7 +831,6 @@ xfs_bmbt_maxrecs(
  */
 int
 xfs_bmdr_maxrecs(
-	struct xfs_mount	*mp,
 	int			blocklen,
 	int			leaf)
 {
@@ -962,7 +878,7 @@ xfs_bmbt_change_owner(
 
 	cur = xfs_bmbt_init_cursor(ip->i_mount, tp, ip, whichfork);
 	if (!cur)
-		return ENOMEM;
+		return -ENOMEM;
 
 	error = xfs_btree_change_owner(cur, new_owner, buffer_list);
 	xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
diff --git a/fs/xfs/xfs_bmap_btree.h b/fs/xfs/libxfs/xfs_bmap_btree.h
index 6e42e1e50b89..819a8a4dee95 100644
--- a/fs/xfs/xfs_bmap_btree.h
+++ b/fs/xfs/libxfs/xfs_bmap_btree.h
@@ -130,7 +130,7 @@ extern void xfs_bmbt_to_bmdr(struct xfs_mount *, struct xfs_btree_block *, int,
 			xfs_bmdr_block_t *, int);
 
 extern int xfs_bmbt_get_maxrecs(struct xfs_btree_cur *, int level);
-extern int xfs_bmdr_maxrecs(struct xfs_mount *, int blocklen, int leaf);
+extern int xfs_bmdr_maxrecs(int blocklen, int leaf);
 extern int xfs_bmbt_maxrecs(struct xfs_mount *, int blocklen, int leaf);
 
 extern int xfs_bmbt_change_owner(struct xfs_trans *tp, struct xfs_inode *ip,
diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
index 9adaae4f3e2f..8fe6a93ff473 100644
--- a/fs/xfs/xfs_btree.c
+++ b/fs/xfs/libxfs/xfs_btree.c
@@ -33,6 +33,7 @@
 #include "xfs_error.h"
 #include "xfs_trace.h"
 #include "xfs_cksum.h"
+#include "xfs_alloc.h"
 
 /*
  * Cursor allocation zone.
@@ -43,9 +44,10 @@ kmem_zone_t	*xfs_btree_cur_zone;
  * Btree magic numbers.
  */
 static const __uint32_t xfs_magics[2][XFS_BTNUM_MAX] = {
-	{ XFS_ABTB_MAGIC, XFS_ABTC_MAGIC, XFS_BMAP_MAGIC, XFS_IBT_MAGIC },
+	{ XFS_ABTB_MAGIC, XFS_ABTC_MAGIC, XFS_BMAP_MAGIC, XFS_IBT_MAGIC,
+	  XFS_FIBT_MAGIC },
 	{ XFS_ABTB_CRC_MAGIC, XFS_ABTC_CRC_MAGIC,
-	  XFS_BMAP_CRC_MAGIC, XFS_IBT_CRC_MAGIC }
+	  XFS_BMAP_CRC_MAGIC, XFS_IBT_CRC_MAGIC, XFS_FIBT_CRC_MAGIC }
 };
 #define xfs_btree_magic(cur) \
 	xfs_magics[!!((cur)->bc_flags & XFS_BTREE_CRC_BLOCKS)][cur->bc_btnum]
@@ -76,11 +78,11 @@ xfs_btree_check_lblock(
 		be16_to_cpu(block->bb_numrecs) <=
 			cur->bc_ops->get_maxrecs(cur, level) &&
 		block->bb_u.l.bb_leftsib &&
-		(block->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO) ||
+		(block->bb_u.l.bb_leftsib == cpu_to_be64(NULLFSBLOCK) ||
 		 XFS_FSB_SANITY_CHECK(mp,
 			be64_to_cpu(block->bb_u.l.bb_leftsib))) &&
 		block->bb_u.l.bb_rightsib &&
-		(block->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO) ||
+		(block->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK) ||
 		 XFS_FSB_SANITY_CHECK(mp,
 			be64_to_cpu(block->bb_u.l.bb_rightsib)));
 
@@ -90,7 +92,7 @@ xfs_btree_check_lblock(
 		if (bp)
 			trace_xfs_btree_corrupt(bp, _RET_IP_);
 		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -138,7 +140,7 @@ xfs_btree_check_sblock(
 		if (bp)
 			trace_xfs_btree_corrupt(bp, _RET_IP_);
 		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -165,12 +167,12 @@ xfs_btree_check_block(
 int					/* error (0 or EFSCORRUPTED) */
 xfs_btree_check_lptr(
 	struct xfs_btree_cur	*cur,	/* btree cursor */
-	xfs_dfsbno_t		bno,	/* btree block disk address */
+	xfs_fsblock_t		bno,	/* btree block disk address */
 	int			level)	/* btree block level */
 {
 	XFS_WANT_CORRUPTED_RETURN(
 		level > 0 &&
-		bno != NULLDFSBNO &&
+		bno != NULLFSBLOCK &&
 		XFS_FSB_SANITY_CHECK(cur->bc_mp, bno));
 	return 0;
 }
@@ -234,8 +236,7 @@ xfs_btree_lblock_calc_crc(
 		return;
 	if (bip)
 		block->bb_u.l.bb_lsn = cpu_to_be64(bip->bli_item.li_lsn);
-	xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
-			 XFS_BTREE_LBLOCK_CRC_OFF);
+	xfs_buf_update_cksum(bp, XFS_BTREE_LBLOCK_CRC_OFF);
 }
 
 bool
@@ -243,8 +244,8 @@ xfs_btree_lblock_verify_crc(
 	struct xfs_buf		*bp)
 {
 	if (xfs_sb_version_hascrc(&bp->b_target->bt_mount->m_sb))
-		return xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
-					XFS_BTREE_LBLOCK_CRC_OFF);
+		return xfs_buf_verify_cksum(bp, XFS_BTREE_LBLOCK_CRC_OFF);
+
 	return true;
 }
 
@@ -267,8 +268,7 @@ xfs_btree_sblock_calc_crc(
 		return;
 	if (bip)
 		block->bb_u.s.bb_lsn = cpu_to_be64(bip->bli_item.li_lsn);
-	xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
-			 XFS_BTREE_SBLOCK_CRC_OFF);
+	xfs_buf_update_cksum(bp, XFS_BTREE_SBLOCK_CRC_OFF);
 }
 
 bool
@@ -276,8 +276,8 @@ xfs_btree_sblock_verify_crc(
 	struct xfs_buf		*bp)
 {
 	if (xfs_sb_version_hascrc(&bp->b_target->bt_mount->m_sb))
-		return xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
-					XFS_BTREE_SBLOCK_CRC_OFF);
+		return xfs_buf_verify_cksum(bp, XFS_BTREE_SBLOCK_CRC_OFF);
+
 	return true;
 }
 
@@ -554,14 +554,11 @@ xfs_btree_get_bufl(
 	xfs_fsblock_t	fsbno,		/* file system block number */
 	uint		lock)		/* lock flags for get_buf */
 {
-	xfs_buf_t	*bp;		/* buffer pointer (return value) */
 	xfs_daddr_t		d;		/* real disk block address */
 
 	ASSERT(fsbno != NULLFSBLOCK);
 	d = XFS_FSB_TO_DADDR(mp, fsbno);
-	bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, d, mp->m_bsize, lock);
-	ASSERT(!xfs_buf_geterror(bp));
-	return bp;
+	return xfs_trans_get_buf(tp, mp->m_ddev_targp, d, mp->m_bsize, lock);
 }
 
 /*
@@ -576,15 +573,12 @@ xfs_btree_get_bufs(
 	xfs_agblock_t	agbno,		/* allocation group block number */
 	uint		lock)		/* lock flags for get_buf */
 {
-	xfs_buf_t	*bp;		/* buffer pointer (return value) */
 	xfs_daddr_t		d;		/* real disk block address */
 
 	ASSERT(agno != NULLAGNUMBER);
 	ASSERT(agbno != NULLAGBLOCK);
 	d = XFS_AGB_TO_DADDR(mp, agno, agbno);
-	bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, d, mp->m_bsize, lock);
-	ASSERT(!xfs_buf_geterror(bp));
-	return bp;
+	return xfs_trans_get_buf(tp, mp->m_ddev_targp, d, mp->m_bsize, lock);
 }
 
 /*
@@ -601,7 +595,7 @@ xfs_btree_islastblock(
 	block = xfs_btree_get_block(cur, level, &bp);
 	xfs_btree_check_block(cur, block, level, bp);
 	if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
-		return block->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO);
+		return block->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK);
 	else
 		return block->bb_u.s.bb_rightsib == cpu_to_be32(NULLAGBLOCK);
 }
@@ -724,7 +718,6 @@ xfs_btree_read_bufl(
 				   mp->m_bsize, lock, &bp, ops);
 	if (error)
 		return error;
-	ASSERT(!xfs_buf_geterror(bp));
 	if (bp)
 		xfs_buf_set_ref(bp, refval);
 	*bpp = bp;
@@ -778,16 +771,16 @@ xfs_btree_readahead_lblock(
 	struct xfs_btree_block	*block)
 {
 	int			rval = 0;
-	xfs_dfsbno_t		left = be64_to_cpu(block->bb_u.l.bb_leftsib);
-	xfs_dfsbno_t		right = be64_to_cpu(block->bb_u.l.bb_rightsib);
+	xfs_fsblock_t		left = be64_to_cpu(block->bb_u.l.bb_leftsib);
+	xfs_fsblock_t		right = be64_to_cpu(block->bb_u.l.bb_rightsib);
 
-	if ((lr & XFS_BTCUR_LEFTRA) && left != NULLDFSBNO) {
+	if ((lr & XFS_BTCUR_LEFTRA) && left != NULLFSBLOCK) {
 		xfs_btree_reada_bufl(cur->bc_mp, left, 1,
 				     cur->bc_ops->buf_ops);
 		rval++;
 	}
 
-	if ((lr & XFS_BTCUR_RIGHTRA) && right != NULLDFSBNO) {
+	if ((lr & XFS_BTCUR_RIGHTRA) && right != NULLFSBLOCK) {
 		xfs_btree_reada_bufl(cur->bc_mp, right, 1,
 				     cur->bc_ops->buf_ops);
 		rval++;
@@ -859,7 +852,7 @@ xfs_btree_ptr_to_daddr(
 	union xfs_btree_ptr	*ptr)
 {
 	if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
-		ASSERT(ptr->l != cpu_to_be64(NULLDFSBNO));
+		ASSERT(ptr->l != cpu_to_be64(NULLFSBLOCK));
 
 		return XFS_FSB_TO_DADDR(cur->bc_mp, be64_to_cpu(ptr->l));
 	} else {
@@ -907,9 +900,9 @@ xfs_btree_setbuf(
 
 	b = XFS_BUF_TO_BLOCK(bp);
 	if (cur->bc_flags & XFS_BTREE_LONG_PTRS) {
-		if (b->bb_u.l.bb_leftsib == cpu_to_be64(NULLDFSBNO))
+		if (b->bb_u.l.bb_leftsib == cpu_to_be64(NULLFSBLOCK))
 			cur->bc_ra[lev] |= XFS_BTCUR_LEFTRA;
-		if (b->bb_u.l.bb_rightsib == cpu_to_be64(NULLDFSBNO))
+		if (b->bb_u.l.bb_rightsib == cpu_to_be64(NULLFSBLOCK))
 			cur->bc_ra[lev] |= XFS_BTCUR_RIGHTRA;
 	} else {
 		if (b->bb_u.s.bb_leftsib == cpu_to_be32(NULLAGBLOCK))
@@ -925,7 +918,7 @@ xfs_btree_ptr_is_null(
 	union xfs_btree_ptr	*ptr)
 {
 	if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
-		return ptr->l == cpu_to_be64(NULLDFSBNO);
+		return ptr->l == cpu_to_be64(NULLFSBLOCK);
 	else
 		return ptr->s == cpu_to_be32(NULLAGBLOCK);
 }
@@ -936,7 +929,7 @@ xfs_btree_set_ptr_null(
 	union xfs_btree_ptr	*ptr)
 {
 	if (cur->bc_flags & XFS_BTREE_LONG_PTRS)
-		ptr->l = cpu_to_be64(NULLDFSBNO);
+		ptr->l = cpu_to_be64(NULLFSBLOCK);
 	else
 		ptr->s = cpu_to_be32(NULLAGBLOCK);
 }
@@ -1004,8 +997,8 @@ xfs_btree_init_block_int(
 	buf->bb_numrecs = cpu_to_be16(numrecs);
 
 	if (flags & XFS_BTREE_LONG_PTRS) {
-		buf->bb_u.l.bb_leftsib = cpu_to_be64(NULLDFSBNO);
-		buf->bb_u.l.bb_rightsib = cpu_to_be64(NULLDFSBNO);
+		buf->bb_u.l.bb_leftsib = cpu_to_be64(NULLFSBLOCK);
+		buf->bb_u.l.bb_rightsib = cpu_to_be64(NULLFSBLOCK);
 		if (flags & XFS_BTREE_CRC_BLOCKS) {
 			buf->bb_u.l.bb_blkno = cpu_to_be64(blkno);
 			buf->bb_u.l.bb_owner = cpu_to_be64(owner);
@@ -1117,6 +1110,7 @@ xfs_btree_set_refs(
 		xfs_buf_set_ref(bp, XFS_ALLOC_BTREE_REF);
 		break;
 	case XFS_BTNUM_INO:
+	case XFS_BTNUM_FINO:
 		xfs_buf_set_ref(bp, XFS_INO_BTREE_REF);
 		break;
 	case XFS_BTNUM_BMAP:
@@ -1146,7 +1140,7 @@ xfs_btree_get_buf_block(
 				 mp->m_bsize, flags);
 
 	if (!*bpp)
-		return ENOMEM;
+		return -ENOMEM;
 
 	(*bpp)->b_ops = cur->bc_ops->buf_ops;
 	*block = XFS_BUF_TO_BLOCK(*bpp);
@@ -1161,7 +1155,6 @@ STATIC int
 xfs_btree_read_buf_block(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_ptr	*ptr,
-	int			level,
 	int			flags,
 	struct xfs_btree_block	**block,
 	struct xfs_buf		**bpp)
@@ -1180,7 +1173,6 @@ xfs_btree_read_buf_block(
 	if (error)
 		return error;
 
-	ASSERT(!xfs_buf_geterror(*bpp));
 	xfs_btree_set_refs(cur, *bpp);
 	*block = XFS_BUF_TO_BLOCK(*bpp);
 	return 0;
@@ -1506,7 +1498,7 @@ xfs_btree_increment(
 		if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE)
 			goto out0;
 		ASSERT(0);
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto error0;
 	}
 	ASSERT(lev < cur->bc_nlevels);
@@ -1519,8 +1511,8 @@ xfs_btree_increment(
 		union xfs_btree_ptr	*ptrp;
 
 		ptrp = xfs_btree_ptr_addr(cur, cur->bc_ptrs[lev], block);
-		error = xfs_btree_read_buf_block(cur, ptrp, --lev,
-							0, &block, &bp);
+		--lev;
+		error = xfs_btree_read_buf_block(cur, ptrp, 0, &block, &bp);
 		if (error)
 			goto error0;
 
@@ -1605,7 +1597,7 @@ xfs_btree_decrement(
 		if (cur->bc_flags & XFS_BTREE_ROOT_IN_INODE)
 			goto out0;
 		ASSERT(0);
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto error0;
 	}
 	ASSERT(lev < cur->bc_nlevels);
@@ -1618,8 +1610,8 @@ xfs_btree_decrement(
 		union xfs_btree_ptr	*ptrp;
 
 		ptrp = xfs_btree_ptr_addr(cur, cur->bc_ptrs[lev], block);
-		error = xfs_btree_read_buf_block(cur, ptrp, --lev,
-							0, &block, &bp);
+		--lev;
+		error = xfs_btree_read_buf_block(cur, ptrp, 0, &block, &bp);
 		if (error)
 			goto error0;
 		xfs_btree_setbuf(cur, lev, bp);
@@ -1669,7 +1661,7 @@ xfs_btree_lookup_get_block(
 		return 0;
 	}
 
-	error = xfs_btree_read_buf_block(cur, pp, level, 0, blkp, &bp);
+	error = xfs_btree_read_buf_block(cur, pp, 0, blkp, &bp);
 	if (error)
 		return error;
 
@@ -2020,7 +2012,7 @@ xfs_btree_lshift(
 		goto out0;
 
 	/* Set up the left neighbor as "left". */
-	error = xfs_btree_read_buf_block(cur, &lptr, level, 0, &left, &lbp);
+	error = xfs_btree_read_buf_block(cur, &lptr, 0, &left, &lbp);
 	if (error)
 		goto error0;
 
@@ -2204,7 +2196,7 @@ xfs_btree_rshift(
 		goto out0;
 
 	/* Set up the right neighbor as "right". */
-	error = xfs_btree_read_buf_block(cur, &rptr, level, 0, &right, &rbp);
+	error = xfs_btree_read_buf_block(cur, &rptr, 0, &right, &rbp);
 	if (error)
 		goto error0;
 
@@ -2332,7 +2324,7 @@ error1:
  * record (to be inserted into parent).
  */
 STATIC int					/* error */
-xfs_btree_split(
+__xfs_btree_split(
 	struct xfs_btree_cur	*cur,
 	int			level,
 	union xfs_btree_ptr	*ptrp,
@@ -2374,7 +2366,7 @@ xfs_btree_split(
 	xfs_btree_buf_to_ptr(cur, lbp, &lptr);
 
 	/* Allocate the new block. If we can't do it, we're toast. Give up. */
-	error = cur->bc_ops->alloc_block(cur, &lptr, &rptr, 1, stat);
+	error = cur->bc_ops->alloc_block(cur, &lptr, &rptr, stat);
 	if (error)
 		goto error0;
 	if (*stat == 0)
@@ -2472,7 +2464,7 @@ xfs_btree_split(
 	 * point back to right instead of to left.
 	 */
 	if (!xfs_btree_ptr_is_null(cur, &rrptr)) {
-		error = xfs_btree_read_buf_block(cur, &rrptr, level,
+		error = xfs_btree_read_buf_block(cur, &rrptr,
 							0, &rrblock, &rrbp);
 		if (error)
 			goto error0;
@@ -2512,6 +2504,85 @@ error0:
 	return error;
 }
 
+struct xfs_btree_split_args {
+	struct xfs_btree_cur	*cur;
+	int			level;
+	union xfs_btree_ptr	*ptrp;
+	union xfs_btree_key	*key;
+	struct xfs_btree_cur	**curp;
+	int			*stat;		/* success/failure */
+	int			result;
+	bool			kswapd;	/* allocation in kswapd context */
+	struct completion	*done;
+	struct work_struct	work;
+};
+
+/*
+ * Stack switching interfaces for allocation
+ */
+static void
+xfs_btree_split_worker(
+	struct work_struct	*work)
+{
+	struct xfs_btree_split_args	*args = container_of(work,
+						struct xfs_btree_split_args, work);
+	unsigned long		pflags;
+	unsigned long		new_pflags = PF_FSTRANS;
+
+	/*
+	 * we are in a transaction context here, but may also be doing work
+	 * in kswapd context, and hence we may need to inherit that state
+	 * temporarily to ensure that we don't block waiting for memory reclaim
+	 * in any way.
+	 */
+	if (args->kswapd)
+		new_pflags |= PF_MEMALLOC | PF_SWAPWRITE | PF_KSWAPD;
+
+	current_set_flags_nested(&pflags, new_pflags);
+
+	args->result = __xfs_btree_split(args->cur, args->level, args->ptrp,
+					 args->key, args->curp, args->stat);
+	complete(args->done);
+
+	current_restore_flags_nested(&pflags, new_pflags);
+}
+
+/*
+ * BMBT split requests often come in with little stack to work on. Push
+ * them off to a worker thread so there is lots of stack to use. For the other
+ * btree types, just call directly to avoid the context switch overhead here.
+ */
+STATIC int					/* error */
+xfs_btree_split(
+	struct xfs_btree_cur	*cur,
+	int			level,
+	union xfs_btree_ptr	*ptrp,
+	union xfs_btree_key	*key,
+	struct xfs_btree_cur	**curp,
+	int			*stat)		/* success/failure */
+{
+	struct xfs_btree_split_args	args;
+	DECLARE_COMPLETION_ONSTACK(done);
+
+	if (cur->bc_btnum != XFS_BTNUM_BMAP)
+		return __xfs_btree_split(cur, level, ptrp, key, curp, stat);
+
+	args.cur = cur;
+	args.level = level;
+	args.ptrp = ptrp;
+	args.key = key;
+	args.curp = curp;
+	args.stat = stat;
+	args.done = &done;
+	args.kswapd = current_is_kswapd();
+	INIT_WORK_ONSTACK(&args.work, xfs_btree_split_worker);
+	queue_work(xfs_alloc_wq, &args.work);
+	wait_for_completion(&done);
+	destroy_work_on_stack(&args.work);
+	return args.result;
+}
+
+
 /*
  * Copy the old inode root contents into a real block and make the
  * broot point to it.
@@ -2547,7 +2618,7 @@ xfs_btree_new_iroot(
 	pp = xfs_btree_ptr_addr(cur, 1, block);
 
 	/* Allocate the new block. If we can't do it, we're toast. Give up. */
-	error = cur->bc_ops->alloc_block(cur, pp, &nptr, 1, stat);
+	error = cur->bc_ops->alloc_block(cur, pp, &nptr, stat);
 	if (error)
 		goto error0;
 	if (*stat == 0) {
@@ -2651,7 +2722,7 @@ xfs_btree_new_root(
 	cur->bc_ops->init_ptr_from_cur(cur, &rptr);
 
 	/* Allocate the new block. If we can't do it, we're toast. Give up. */
-	error = cur->bc_ops->alloc_block(cur, &rptr, &lptr, 1, stat);
+	error = cur->bc_ops->alloc_block(cur, &rptr, &lptr, stat);
 	if (error)
 		goto error0;
 	if (*stat == 0)
@@ -2686,8 +2757,7 @@ xfs_btree_new_root(
 		lbp = bp;
 		xfs_btree_buf_to_ptr(cur, lbp, &lptr);
 		left = block;
-		error = xfs_btree_read_buf_block(cur, &rptr,
-					cur->bc_nlevels - 1, 0, &right, &rbp);
+		error = xfs_btree_read_buf_block(cur, &rptr, 0, &right, &rbp);
 		if (error)
 			goto error0;
 		bp = rbp;
@@ -2698,8 +2768,7 @@ xfs_btree_new_root(
 		xfs_btree_buf_to_ptr(cur, rbp, &rptr);
 		right = block;
 		xfs_btree_get_sibling(cur, right, &lptr, XFS_BB_LEFTSIB);
-		error = xfs_btree_read_buf_block(cur, &lptr,
-					cur->bc_nlevels - 1, 0, &left, &lbp);
+		error = xfs_btree_read_buf_block(cur, &lptr, 0, &left, &lbp);
 		if (error)
 			goto error0;
 		bp = lbp;
@@ -3651,8 +3720,7 @@ xfs_btree_delrec(
 		rptr = cptr;
 		right = block;
 		rbp = bp;
-		error = xfs_btree_read_buf_block(cur, &lptr, level,
-							0, &left, &lbp);
+		error = xfs_btree_read_buf_block(cur, &lptr, 0, &left, &lbp);
 		if (error)
 			goto error0;
 
@@ -3669,8 +3737,7 @@ xfs_btree_delrec(
 		lptr = cptr;
 		left = block;
 		lbp = bp;
-		error = xfs_btree_read_buf_block(cur, &rptr, level,
-							0, &right, &rbp);
+		error = xfs_btree_read_buf_block(cur, &rptr, 0, &right, &rbp);
 		if (error)
 			goto error0;
 
@@ -3742,8 +3809,7 @@ xfs_btree_delrec(
 	/* If there is a right sibling, point it to the remaining block. */
 	xfs_btree_get_sibling(cur, left, &cptr, XFS_BB_RIGHTSIB);
 	if (!xfs_btree_ptr_is_null(cur, &cptr)) {
-		error = xfs_btree_read_buf_block(cur, &cptr, level,
-							0, &rrblock, &rrbp);
+		error = xfs_btree_read_buf_block(cur, &cptr, 0, &rrblock, &rrbp);
 		if (error)
 			goto error0;
 		xfs_btree_set_sibling(cur, rrblock, &lptr, XFS_BB_LEFTSIB);
@@ -3952,7 +4018,7 @@ xfs_btree_block_change_owner(
 	/* now read rh sibling block for next iteration */
 	xfs_btree_get_sibling(cur, block, &rptr, XFS_BB_RIGHTSIB);
 	if (xfs_btree_ptr_is_null(cur, &rptr))
-		return ENOENT;
+		return -ENOENT;
 
 	return xfs_btree_lookup_get_block(cur, level, &rptr, &block);
 }
@@ -3995,7 +4061,7 @@ xfs_btree_change_owner(
 							     buffer_list);
 		} while (!error);
 
-		if (error != ENOENT)
+		if (error != -ENOENT)
 			return error;
 	}
 
diff --git a/fs/xfs/xfs_btree.h b/fs/xfs/libxfs/xfs_btree.h
index 91e34f21bace..8f18bab73ea5 100644
--- a/fs/xfs/xfs_btree.h
+++ b/fs/xfs/libxfs/xfs_btree.h
@@ -62,6 +62,7 @@ union xfs_btree_rec {
 #define	XFS_BTNUM_CNT	((xfs_btnum_t)XFS_BTNUM_CNTi)
 #define	XFS_BTNUM_BMAP	((xfs_btnum_t)XFS_BTNUM_BMAPi)
 #define	XFS_BTNUM_INO	((xfs_btnum_t)XFS_BTNUM_INOi)
+#define	XFS_BTNUM_FINO	((xfs_btnum_t)XFS_BTNUM_FINOi)
 
 /*
  * For logging record fields.
@@ -92,6 +93,7 @@ do {    \
 	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_MAX: ASSERT(0); /* fucking gcc */ ; break;	\
 	}       \
 } while (0)
@@ -105,6 +107,7 @@ do {    \
 	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;	\
 	}       \
 } while (0)
@@ -129,7 +132,7 @@ struct xfs_btree_ops {
 	int	(*alloc_block)(struct xfs_btree_cur *cur,
 			       union xfs_btree_ptr *start_bno,
 			       union xfs_btree_ptr *new_bno,
-			       int length, int *stat);
+			       int *stat);
 	int	(*free_block)(struct xfs_btree_cur *cur, struct xfs_buf *bp);
 
 	/* update last record information */
@@ -255,7 +258,7 @@ xfs_btree_check_block(
 int					/* error (0 or EFSCORRUPTED) */
 xfs_btree_check_lptr(
 	struct xfs_btree_cur	*cur,	/* btree cursor */
-	xfs_dfsbno_t		ptr,	/* btree block disk address */
+	xfs_fsblock_t		ptr,	/* btree block disk address */
 	int			level);	/* btree block level */
 
 /*
diff --git a/fs/xfs/xfs_cksum.h b/fs/xfs/libxfs/xfs_cksum.h
index fad1676ad8cd..fad1676ad8cd 100644
--- a/fs/xfs/xfs_cksum.h
+++ b/fs/xfs/libxfs/xfs_cksum.h
diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/libxfs/xfs_da_btree.c
index 796272a2e129..2c42ae28d027 100644
--- a/fs/xfs/xfs_da_btree.c
+++ b/fs/xfs/libxfs/xfs_da_btree.c
@@ -167,8 +167,8 @@ xfs_da3_node_verify(
 	 * we don't know if the node is for and attribute or directory tree,
 	 * so only fail if the count is outside both bounds
 	 */
-	if (ichdr.count > mp->m_dir_node_ents &&
-	    ichdr.count > mp->m_attr_node_ents)
+	if (ichdr.count > mp->m_dir_geo->node_ents &&
+	    ichdr.count > mp->m_attr_geo->node_ents)
 		return false;
 
 	/* XXX: hash order check? */
@@ -185,8 +185,8 @@ xfs_da3_node_write_verify(
 	struct xfs_da3_node_hdr *hdr3 = bp->b_addr;
 
 	if (!xfs_da3_node_verify(bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
 		return;
 	}
 
@@ -196,7 +196,7 @@ xfs_da3_node_write_verify(
 	if (bip)
 		hdr3->info.lsn = cpu_to_be64(bip->bli_item.li_lsn);
 
-	xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), XFS_DA3_NODE_CRC_OFF);
+	xfs_buf_update_cksum(bp, XFS_DA3_NODE_CRC_OFF);
 }
 
 /*
@@ -209,18 +209,20 @@ static void
 xfs_da3_node_read_verify(
 	struct xfs_buf		*bp)
 {
-	struct xfs_mount	*mp = bp->b_target->bt_mount;
 	struct xfs_da_blkinfo	*info = bp->b_addr;
 
 	switch (be16_to_cpu(info->magic)) {
 		case XFS_DA3_NODE_MAGIC:
-			if (!xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
-					      XFS_DA3_NODE_CRC_OFF))
+			if (!xfs_buf_verify_cksum(bp, XFS_DA3_NODE_CRC_OFF)) {
+				xfs_buf_ioerror(bp, -EFSBADCRC);
 				break;
+			}
 			/* fall through */
 		case XFS_DA_NODE_MAGIC:
-			if (!xfs_da3_node_verify(bp))
+			if (!xfs_da3_node_verify(bp)) {
+				xfs_buf_ioerror(bp, -EFSCORRUPTED);
 				break;
+			}
 			return;
 		case XFS_ATTR_LEAF_MAGIC:
 		case XFS_ATTR3_LEAF_MAGIC:
@@ -237,8 +239,7 @@ xfs_da3_node_read_verify(
 	}
 
 	/* corrupt block */
-	XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-	xfs_buf_ioerror(bp, EFSCORRUPTED);
+	xfs_verifier_error(bp);
 }
 
 const struct xfs_buf_ops xfs_da3_node_buf_ops = {
@@ -314,7 +315,7 @@ xfs_da3_node_create(
 
 	error = xfs_da_get_buf(tp, dp, blkno, -1, &bp, whichfork);
 	if (error)
-		return(error);
+		return error;
 	bp->b_ops = &xfs_da3_node_buf_ops;
 	xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DA_NODE_BUF);
 	node = bp->b_addr;
@@ -336,7 +337,7 @@ xfs_da3_node_create(
 		XFS_DA_LOGRANGE(node, &node->hdr, dp->d_ops->node_hdr_size));
 
 	*bpp = bp;
-	return(0);
+	return 0;
 }
 
 /*
@@ -384,8 +385,8 @@ xfs_da3_split(
 		switch (oldblk->magic) {
 		case XFS_ATTR_LEAF_MAGIC:
 			error = xfs_attr3_leaf_split(state, oldblk, newblk);
-			if ((error != 0) && (error != ENOSPC)) {
-				return(error);	/* GROT: attr is inconsistent */
+			if ((error != 0) && (error != -ENOSPC)) {
+				return error;	/* GROT: attr is inconsistent */
 			}
 			if (!error) {
 				addblk = newblk;
@@ -407,7 +408,7 @@ xfs_da3_split(
 							    &state->extrablk);
 			}
 			if (error)
-				return(error);	/* GROT: attr inconsistent */
+				return error;	/* GROT: attr inconsistent */
 			addblk = newblk;
 			break;
 		case XFS_DIR2_LEAFN_MAGIC:
@@ -421,7 +422,7 @@ xfs_da3_split(
 							 max - i, &action);
 			addblk->bp = NULL;
 			if (error)
-				return(error);	/* GROT: dir is inconsistent */
+				return error;	/* GROT: dir is inconsistent */
 			/*
 			 * Record the newly split block for the next time thru?
 			 */
@@ -438,7 +439,7 @@ xfs_da3_split(
 		xfs_da3_fixhashpath(state, &state->path);
 	}
 	if (!addblk)
-		return(0);
+		return 0;
 
 	/*
 	 * Split the root node.
@@ -448,7 +449,7 @@ xfs_da3_split(
 	error = xfs_da3_root_split(state, oldblk, addblk);
 	if (error) {
 		addblk->bp = NULL;
-		return(error);	/* GROT: dir is inconsistent */
+		return error;	/* GROT: dir is inconsistent */
 	}
 
 	/*
@@ -491,7 +492,7 @@ xfs_da3_split(
 		    sizeof(node->hdr.info)));
 	}
 	addblk->bp = NULL;
-	return(0);
+	return 0;
 }
 
 /*
@@ -597,7 +598,7 @@ xfs_da3_root_split(
 	 * Set up the new root node.
 	 */
 	error = xfs_da3_node_create(args,
-		(args->whichfork == XFS_DATA_FORK) ? mp->m_dirleafblk : 0,
+		(args->whichfork == XFS_DATA_FORK) ? args->geo->leafblk : 0,
 		level + 1, &bp, args->whichfork);
 	if (error)
 		return error;
@@ -615,10 +616,10 @@ xfs_da3_root_split(
 #ifdef DEBUG
 	if (oldroot->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
 	    oldroot->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC)) {
-		ASSERT(blk1->blkno >= mp->m_dirleafblk &&
-		       blk1->blkno < mp->m_dirfreeblk);
-		ASSERT(blk2->blkno >= mp->m_dirleafblk &&
-		       blk2->blkno < mp->m_dirfreeblk);
+		ASSERT(blk1->blkno >= args->geo->leafblk &&
+		       blk1->blkno < args->geo->freeblk);
+		ASSERT(blk2->blkno >= args->geo->leafblk &&
+		       blk2->blkno < args->geo->freeblk);
 	}
 #endif
 
@@ -662,25 +663,25 @@ xfs_da3_node_split(
 	/*
 	 * Do we have to split the node?
 	 */
-	if (nodehdr.count + newcount > state->node_ents) {
+	if (nodehdr.count + newcount > state->args->geo->node_ents) {
 		/*
 		 * Allocate a new node, add to the doubly linked chain of
 		 * nodes, then move some of our excess entries into it.
 		 */
 		error = xfs_da_grow_inode(state->args, &blkno);
 		if (error)
-			return(error);	/* GROT: dir is inconsistent */
+			return error;	/* GROT: dir is inconsistent */
 
 		error = xfs_da3_node_create(state->args, blkno, treelevel,
 					   &newblk->bp, state->args->whichfork);
 		if (error)
-			return(error);	/* GROT: dir is inconsistent */
+			return error;	/* GROT: dir is inconsistent */
 		newblk->blkno = blkno;
 		newblk->magic = XFS_DA_NODE_MAGIC;
 		xfs_da3_node_rebalance(state, oldblk, newblk);
 		error = xfs_da3_blk_link(state, oldblk, newblk);
 		if (error)
-			return(error);
+			return error;
 		*result = 1;
 	} else {
 		*result = 0;
@@ -720,7 +721,7 @@ xfs_da3_node_split(
 		}
 	}
 
-	return(0);
+	return 0;
 }
 
 /*
@@ -893,8 +894,8 @@ xfs_da3_node_add(
 	ASSERT(oldblk->index >= 0 && oldblk->index <= nodehdr.count);
 	ASSERT(newblk->blkno != 0);
 	if (state->args->whichfork == XFS_DATA_FORK)
-		ASSERT(newblk->blkno >= state->mp->m_dirleafblk &&
-		       newblk->blkno < state->mp->m_dirfreeblk);
+		ASSERT(newblk->blkno >= state->args->geo->leafblk &&
+		       newblk->blkno < state->args->geo->freeblk);
 
 	/*
 	 * We may need to make some room before we insert the new node.
@@ -962,9 +963,9 @@ xfs_da3_join(
 		case XFS_ATTR_LEAF_MAGIC:
 			error = xfs_attr3_leaf_toosmall(state, &action);
 			if (error)
-				return(error);
+				return error;
 			if (action == 0)
-				return(0);
+				return 0;
 			xfs_attr3_leaf_unbalance(state, drop_blk, save_blk);
 			break;
 		case XFS_DIR2_LEAFN_MAGIC:
@@ -984,7 +985,7 @@ xfs_da3_join(
 			xfs_da3_fixhashpath(state, &state->path);
 			error = xfs_da3_node_toosmall(state, &action);
 			if (error)
-				return(error);
+				return error;
 			if (action == 0)
 				return 0;
 			xfs_da3_node_unbalance(state, drop_blk, save_blk);
@@ -994,12 +995,12 @@ xfs_da3_join(
 		error = xfs_da3_blk_unlink(state, drop_blk, save_blk);
 		xfs_da_state_kill_altpath(state);
 		if (error)
-			return(error);
+			return error;
 		error = xfs_da_shrink_inode(state->args, drop_blk->blkno,
 							 drop_blk->bp);
 		drop_blk->bp = NULL;
 		if (error)
-			return(error);
+			return error;
 	}
 	/*
 	 * We joined all the way to the top.  If it turns out that
@@ -1009,7 +1010,7 @@ xfs_da3_join(
 	xfs_da3_node_remove(state, drop_blk);
 	xfs_da3_fixhashpath(state, &state->path);
 	error = xfs_da3_root_join(state, &state->path.blk[0]);
-	return(error);
+	return error;
 }
 
 #ifdef	DEBUG
@@ -1088,16 +1089,17 @@ xfs_da3_root_join(
 	 * that could occur. For dir3 blocks we also need to update the block
 	 * number in the buffer header.
 	 */
-	memcpy(root_blk->bp->b_addr, bp->b_addr, state->blocksize);
+	memcpy(root_blk->bp->b_addr, bp->b_addr, args->geo->blksize);
 	root_blk->bp->b_ops = bp->b_ops;
 	xfs_trans_buf_copy_type(root_blk->bp, bp);
 	if (oldroothdr.magic == XFS_DA3_NODE_MAGIC) {
 		struct xfs_da3_blkinfo *da3 = root_blk->bp->b_addr;
 		da3->blkno = cpu_to_be64(root_blk->bp->b_bn);
 	}
-	xfs_trans_log_buf(args->trans, root_blk->bp, 0, state->blocksize - 1);
+	xfs_trans_log_buf(args->trans, root_blk->bp, 0,
+			  args->geo->blksize - 1);
 	error = xfs_da_shrink_inode(args, child, bp);
-	return(error);
+	return error;
 }
 
 /*
@@ -1138,9 +1140,9 @@ xfs_da3_node_toosmall(
 	info = blk->bp->b_addr;
 	node = (xfs_da_intnode_t *)info;
 	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
-	if (nodehdr.count > (state->node_ents >> 1)) {
+	if (nodehdr.count > (state->args->geo->node_ents >> 1)) {
 		*action = 0;	/* blk over 50%, don't try to join */
-		return(0);	/* blk over 50%, don't try to join */
+		return 0;	/* blk over 50%, don't try to join */
 	}
 
 	/*
@@ -1159,13 +1161,13 @@ xfs_da3_node_toosmall(
 		error = xfs_da3_path_shift(state, &state->altpath, forward,
 						 0, &retval);
 		if (error)
-			return(error);
+			return error;
 		if (retval) {
 			*action = 0;
 		} else {
 			*action = 2;
 		}
-		return(0);
+		return 0;
 	}
 
 	/*
@@ -1175,8 +1177,8 @@ xfs_da3_node_toosmall(
 	 * We prefer coalescing with the lower numbered sibling so as
 	 * to shrink a directory over time.
 	 */
-	count  = state->node_ents;
-	count -= state->node_ents >> 2;
+	count  = state->args->geo->node_ents;
+	count -= state->args->geo->node_ents >> 2;
 	count -= nodehdr.count;
 
 	/* start with smaller blk num */
@@ -1192,7 +1194,7 @@ xfs_da3_node_toosmall(
 		error = xfs_da3_node_read(state->args->trans, dp,
 					blkno, -1, &bp, state->args->whichfork);
 		if (error)
-			return(error);
+			return error;
 
 		node = bp->b_addr;
 		dp->d_ops->node_hdr_from_disk(&thdr, node);
@@ -1295,7 +1297,7 @@ xfs_da3_fixhashpath(
 		node = blk->bp->b_addr;
 		dp->d_ops->node_hdr_from_disk(&nodehdr, node);
 		btree = dp->d_ops->node_tree_p(node);
-		if (be32_to_cpu(btree->hashval) == lasthash)
+		if (be32_to_cpu(btree[blk->index].hashval) == lasthash)
 			break;
 		blk->hashval = lasthash;
 		btree[blk->index].hashval = cpu_to_be32(lasthash);
@@ -1471,7 +1473,7 @@ xfs_da3_node_lookup_int(
 	 * Descend thru the B-tree searching each level for the right
 	 * node to use, until the right hashval is found.
 	 */
-	blkno = (args->whichfork == XFS_DATA_FORK)? state->mp->m_dirleafblk : 0;
+	blkno = (args->whichfork == XFS_DATA_FORK)? args->geo->leafblk : 0;
 	for (blk = &state->path.blk[0], state->path.active = 1;
 			 state->path.active <= XFS_DA_NODE_MAXDEPTH;
 			 blk++, state->path.active++) {
@@ -1484,7 +1486,7 @@ xfs_da3_node_lookup_int(
 		if (error) {
 			blk->blkno = 0;
 			state->path.active--;
-			return(error);
+			return error;
 		}
 		curr = blk->bp->b_addr;
 		blk->magic = be16_to_cpu(curr->magic);
@@ -1577,25 +1579,25 @@ xfs_da3_node_lookup_int(
 			args->blkno = blk->blkno;
 		} else {
 			ASSERT(0);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
-		if (((retval == ENOENT) || (retval == ENOATTR)) &&
+		if (((retval == -ENOENT) || (retval == -ENOATTR)) &&
 		    (blk->hashval == args->hashval)) {
 			error = xfs_da3_path_shift(state, &state->path, 1, 1,
 							 &retval);
 			if (error)
-				return(error);
+				return error;
 			if (retval == 0) {
 				continue;
 			} else if (blk->magic == XFS_ATTR_LEAF_MAGIC) {
 				/* path_shift() gives ENOENT */
-				retval = XFS_ERROR(ENOATTR);
+				retval = -ENOATTR;
 			}
 		}
 		break;
 	}
 	*result = retval;
-	return(0);
+	return 0;
 }
 
 /*========================================================================
@@ -1690,7 +1692,7 @@ xfs_da3_blk_link(
 						be32_to_cpu(old_info->back),
 						-1, &bp, args->whichfork);
 			if (error)
-				return(error);
+				return error;
 			ASSERT(bp != NULL);
 			tmp_info = bp->b_addr;
 			ASSERT(tmp_info->magic == old_info->magic);
@@ -1711,7 +1713,7 @@ xfs_da3_blk_link(
 						be32_to_cpu(old_info->forw),
 						-1, &bp, args->whichfork);
 			if (error)
-				return(error);
+				return error;
 			ASSERT(bp != NULL);
 			tmp_info = bp->b_addr;
 			ASSERT(tmp_info->magic == old_info->magic);
@@ -1724,7 +1726,7 @@ xfs_da3_blk_link(
 
 	xfs_trans_log_buf(args->trans, old_blk->bp, 0, sizeof(*tmp_info) - 1);
 	xfs_trans_log_buf(args->trans, new_blk->bp, 0, sizeof(*tmp_info) - 1);
-	return(0);
+	return 0;
 }
 
 /*
@@ -1770,7 +1772,7 @@ xfs_da3_blk_unlink(
 						be32_to_cpu(drop_info->back),
 						-1, &bp, args->whichfork);
 			if (error)
-				return(error);
+				return error;
 			ASSERT(bp != NULL);
 			tmp_info = bp->b_addr;
 			ASSERT(tmp_info->magic == save_info->magic);
@@ -1787,7 +1789,7 @@ xfs_da3_blk_unlink(
 						be32_to_cpu(drop_info->forw),
 						-1, &bp, args->whichfork);
 			if (error)
-				return(error);
+				return error;
 			ASSERT(bp != NULL);
 			tmp_info = bp->b_addr;
 			ASSERT(tmp_info->magic == save_info->magic);
@@ -1799,7 +1801,7 @@ xfs_da3_blk_unlink(
 	}
 
 	xfs_trans_log_buf(args->trans, save_blk->bp, 0, sizeof(*save_info) - 1);
-	return(0);
+	return 0;
 }
 
 /*
@@ -1857,9 +1859,9 @@ xfs_da3_path_shift(
 		}
 	}
 	if (level < 0) {
-		*result = XFS_ERROR(ENOENT);	/* we're out of our tree */
+		*result = -ENOENT;	/* we're out of our tree */
 		ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
-		return(0);
+		return 0;
 	}
 
 	/*
@@ -1881,7 +1883,7 @@ xfs_da3_path_shift(
 		error = xfs_da3_node_read(args->trans, dp, blkno, -1,
 					&blk->bp, args->whichfork);
 		if (error)
-			return(error);
+			return error;
 		info = blk->bp->b_addr;
 		ASSERT(info->magic == cpu_to_be16(XFS_DA_NODE_MAGIC) ||
 		       info->magic == cpu_to_be16(XFS_DA3_NODE_MAGIC) ||
@@ -2002,7 +2004,7 @@ xfs_da_grow_inode_int(
 	struct xfs_trans	*tp = args->trans;
 	struct xfs_inode	*dp = args->dp;
 	int			w = args->whichfork;
-	xfs_drfsbno_t		nblks = dp->i_d.di_nblocks;
+	xfs_rfsblock_t		nblks = dp->i_d.di_nblocks;
 	struct xfs_bmbt_irec	map, *mapp;
 	int			nmap, error, got, i, mapi;
 
@@ -2066,7 +2068,7 @@ xfs_da_grow_inode_int(
 	if (got != count || mapp[0].br_startoff != *bno ||
 	    mapp[mapi - 1].br_startoff + mapp[mapi - 1].br_blockcount !=
 	    *bno + count) {
-		error = XFS_ERROR(ENOSPC);
+		error = -ENOSPC;
 		goto out_free_map;
 	}
 
@@ -2089,20 +2091,12 @@ xfs_da_grow_inode(
 	xfs_dablk_t		*new_blkno)
 {
 	xfs_fileoff_t		bno;
-	int			count;
 	int			error;
 
 	trace_xfs_da_grow_inode(args);
 
-	if (args->whichfork == XFS_DATA_FORK) {
-		bno = args->dp->i_mount->m_dirleafblk;
-		count = args->dp->i_mount->m_dirblkfsbs;
-	} else {
-		bno = 0;
-		count = 1;
-	}
-
-	error = xfs_da_grow_inode_int(args, &bno, count);
+	bno = args->geo->leafblk;
+	error = xfs_da_grow_inode_int(args, &bno, args->geo->fsbcount);
 	if (!error)
 		*new_blkno = (xfs_dablk_t)bno;
 	return error;
@@ -2157,27 +2151,27 @@ xfs_da3_swap_lastblock(
 	w = args->whichfork;
 	ASSERT(w == XFS_DATA_FORK);
 	mp = dp->i_mount;
-	lastoff = mp->m_dirfreeblk;
+	lastoff = args->geo->freeblk;
 	error = xfs_bmap_last_before(tp, dp, &lastoff, w);
 	if (error)
 		return error;
 	if (unlikely(lastoff == 0)) {
 		XFS_ERROR_REPORT("xfs_da_swap_lastblock(1)", XFS_ERRLEVEL_LOW,
 				 mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	/*
 	 * Read the last block in the btree space.
 	 */
-	last_blkno = (xfs_dablk_t)lastoff - mp->m_dirblkfsbs;
+	last_blkno = (xfs_dablk_t)lastoff - args->geo->fsbcount;
 	error = xfs_da3_node_read(tp, dp, last_blkno, -1, &last_buf, w);
 	if (error)
 		return error;
 	/*
 	 * Copy the last block into the dead buffer and log it.
 	 */
-	memcpy(dead_buf->b_addr, last_buf->b_addr, mp->m_dirblksize);
-	xfs_trans_log_buf(tp, dead_buf, 0, mp->m_dirblksize - 1);
+	memcpy(dead_buf->b_addr, last_buf->b_addr, args->geo->blksize);
+	xfs_trans_log_buf(tp, dead_buf, 0, args->geo->blksize - 1);
 	dead_info = dead_buf->b_addr;
 	/*
 	 * Get values from the moved block.
@@ -2215,7 +2209,7 @@ xfs_da3_swap_lastblock(
 		    sib_info->magic != dead_info->magic)) {
 			XFS_ERROR_REPORT("xfs_da_swap_lastblock(2)",
 					 XFS_ERRLEVEL_LOW, mp);
-			error = XFS_ERROR(EFSCORRUPTED);
+			error = -EFSCORRUPTED;
 			goto done;
 		}
 		sib_info->forw = cpu_to_be32(dead_blkno);
@@ -2237,7 +2231,7 @@ xfs_da3_swap_lastblock(
 		       sib_info->magic != dead_info->magic)) {
 			XFS_ERROR_REPORT("xfs_da_swap_lastblock(3)",
 					 XFS_ERRLEVEL_LOW, mp);
-			error = XFS_ERROR(EFSCORRUPTED);
+			error = -EFSCORRUPTED;
 			goto done;
 		}
 		sib_info->back = cpu_to_be32(dead_blkno);
@@ -2246,7 +2240,7 @@ xfs_da3_swap_lastblock(
 					sizeof(sib_info->back)));
 		sib_buf = NULL;
 	}
-	par_blkno = mp->m_dirleafblk;
+	par_blkno = args->geo->leafblk;
 	level = -1;
 	/*
 	 * Walk down the tree looking for the parent of the moved block.
@@ -2260,7 +2254,7 @@ xfs_da3_swap_lastblock(
 		if (level >= 0 && level != par_hdr.level + 1) {
 			XFS_ERROR_REPORT("xfs_da_swap_lastblock(4)",
 					 XFS_ERRLEVEL_LOW, mp);
-			error = XFS_ERROR(EFSCORRUPTED);
+			error = -EFSCORRUPTED;
 			goto done;
 		}
 		level = par_hdr.level;
@@ -2273,7 +2267,7 @@ xfs_da3_swap_lastblock(
 		if (entno == par_hdr.count) {
 			XFS_ERROR_REPORT("xfs_da_swap_lastblock(5)",
 					 XFS_ERRLEVEL_LOW, mp);
-			error = XFS_ERROR(EFSCORRUPTED);
+			error = -EFSCORRUPTED;
 			goto done;
 		}
 		par_blkno = be32_to_cpu(btree[entno].before);
@@ -2300,7 +2294,7 @@ xfs_da3_swap_lastblock(
 		if (unlikely(par_blkno == 0)) {
 			XFS_ERROR_REPORT("xfs_da_swap_lastblock(6)",
 					 XFS_ERRLEVEL_LOW, mp);
-			error = XFS_ERROR(EFSCORRUPTED);
+			error = -EFSCORRUPTED;
 			goto done;
 		}
 		error = xfs_da3_node_read(tp, dp, par_blkno, -1, &par_buf, w);
@@ -2311,7 +2305,7 @@ xfs_da3_swap_lastblock(
 		if (par_hdr.level != level) {
 			XFS_ERROR_REPORT("xfs_da_swap_lastblock(7)",
 					 XFS_ERRLEVEL_LOW, mp);
-			error = XFS_ERROR(EFSCORRUPTED);
+			error = -EFSCORRUPTED;
 			goto done;
 		}
 		btree = dp->d_ops->node_tree_p(par_node);
@@ -2356,10 +2350,7 @@ xfs_da_shrink_inode(
 	w = args->whichfork;
 	tp = args->trans;
 	mp = dp->i_mount;
-	if (w == XFS_DATA_FORK)
-		count = mp->m_dirblkfsbs;
-	else
-		count = 1;
+	count = args->geo->fsbcount;
 	for (;;) {
 		/*
 		 * Remove extents.  If we get ENOSPC for a dir we have to move
@@ -2368,7 +2359,7 @@ xfs_da_shrink_inode(
 		error = xfs_bunmapi(tp, dp, dead_blkno, count,
 				    xfs_bmapi_aflag(w)|XFS_BMAPI_METADATA,
 				    0, args->firstblock, args->flist, &done);
-		if (error == ENOSPC) {
+		if (error == -ENOSPC) {
 			if (w != XFS_DATA_FORK)
 				break;
 			error = xfs_da3_swap_lastblock(args, &dead_blkno,
@@ -2436,7 +2427,7 @@ xfs_buf_map_from_irec(
 		map = kmem_zalloc(nirecs * sizeof(struct xfs_buf_map),
 				  KM_SLEEP | KM_NOFS);
 		if (!map)
-			return ENOMEM;
+			return -ENOMEM;
 		*mapp = map;
 	}
 
@@ -2461,7 +2452,6 @@ xfs_buf_map_from_irec(
  */
 static int
 xfs_dabuf_map(
-	struct xfs_trans	*trans,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
 	xfs_daddr_t		mappedbno,
@@ -2479,7 +2469,10 @@ xfs_dabuf_map(
 	ASSERT(map && *map);
 	ASSERT(*nmaps == 1);
 
-	nfsb = (whichfork == XFS_DATA_FORK) ? mp->m_dirblkfsbs : 1;
+	if (whichfork == XFS_DATA_FORK)
+		nfsb = mp->m_dir_geo->fsbcount;
+	else
+		nfsb = mp->m_attr_geo->fsbcount;
 
 	/*
 	 * Caller doesn't have a mapping.  -2 means don't complain
@@ -2507,8 +2500,8 @@ xfs_dabuf_map(
 	}
 
 	if (!xfs_da_map_covers_blocks(nirecs, irecs, bno, nfsb)) {
-		error = mappedbno == -2 ? -1 : XFS_ERROR(EFSCORRUPTED);
-		if (unlikely(error == EFSCORRUPTED)) {
+		error = mappedbno == -2 ? -1 : -EFSCORRUPTED;
+		if (unlikely(error == -EFSCORRUPTED)) {
 			if (xfs_error_level >= XFS_ERRLEVEL_LOW) {
 				int i;
 				xfs_alert(mp, "%s: bno %lld dir: inode %lld",
@@ -2557,7 +2550,7 @@ xfs_da_get_buf(
 	*bpp = NULL;
 	mapp = &map;
 	nmap = 1;
-	error = xfs_dabuf_map(trans, dp, bno, mappedbno, whichfork,
+	error = xfs_dabuf_map(dp, bno, mappedbno, whichfork,
 				&mapp, &nmap);
 	if (error) {
 		/* mapping a hole is not an error, but we don't continue */
@@ -2568,7 +2561,7 @@ xfs_da_get_buf(
 
 	bp = xfs_trans_get_buf_map(trans, dp->i_mount->m_ddev_targp,
 				    mapp, nmap, 0);
-	error = bp ? bp->b_error : XFS_ERROR(EIO);
+	error = bp ? bp->b_error : -EIO;
 	if (error) {
 		xfs_trans_brelse(trans, bp);
 		goto out_free;
@@ -2605,7 +2598,7 @@ xfs_da_read_buf(
 	*bpp = NULL;
 	mapp = &map;
 	nmap = 1;
-	error = xfs_dabuf_map(trans, dp, bno, mappedbno, whichfork,
+	error = xfs_dabuf_map(dp, bno, mappedbno, whichfork,
 				&mapp, &nmap);
 	if (error) {
 		/* mapping a hole is not an error, but we don't continue */
@@ -2624,47 +2617,6 @@ xfs_da_read_buf(
 		xfs_buf_set_ref(bp, XFS_ATTR_BTREE_REF);
 	else
 		xfs_buf_set_ref(bp, XFS_DIR_BTREE_REF);
-
-	/*
-	 * This verification code will be moved to a CRC verification callback
-	 * function so just leave it here unchanged until then.
-	 */
-	{
-		xfs_dir2_data_hdr_t	*hdr = bp->b_addr;
-		xfs_dir2_free_t		*free = bp->b_addr;
-		xfs_da_blkinfo_t	*info = bp->b_addr;
-		uint			magic, magic1;
-		struct xfs_mount	*mp = dp->i_mount;
-
-		magic = be16_to_cpu(info->magic);
-		magic1 = be32_to_cpu(hdr->magic);
-		if (unlikely(
-		    XFS_TEST_ERROR((magic != XFS_DA_NODE_MAGIC) &&
-				   (magic != XFS_DA3_NODE_MAGIC) &&
-				   (magic != XFS_ATTR_LEAF_MAGIC) &&
-				   (magic != XFS_ATTR3_LEAF_MAGIC) &&
-				   (magic != XFS_DIR2_LEAF1_MAGIC) &&
-				   (magic != XFS_DIR3_LEAF1_MAGIC) &&
-				   (magic != XFS_DIR2_LEAFN_MAGIC) &&
-				   (magic != XFS_DIR3_LEAFN_MAGIC) &&
-				   (magic1 != XFS_DIR2_BLOCK_MAGIC) &&
-				   (magic1 != XFS_DIR3_BLOCK_MAGIC) &&
-				   (magic1 != XFS_DIR2_DATA_MAGIC) &&
-				   (magic1 != XFS_DIR3_DATA_MAGIC) &&
-				   (free->hdr.magic !=
-					cpu_to_be32(XFS_DIR2_FREE_MAGIC)) &&
-				   (free->hdr.magic !=
-					cpu_to_be32(XFS_DIR3_FREE_MAGIC)),
-				mp, XFS_ERRTAG_DA_READ_BUF,
-				XFS_RANDOM_DA_READ_BUF))) {
-			trace_xfs_da_btree_corrupt(bp, _RET_IP_);
-			XFS_CORRUPTION_ERROR("xfs_da_do_buf(2)",
-					     XFS_ERRLEVEL_LOW, mp, info);
-			error = XFS_ERROR(EFSCORRUPTED);
-			xfs_trans_brelse(trans, bp);
-			goto out_free;
-		}
-	}
 	*bpp = bp;
 out_free:
 	if (mapp != &map)
@@ -2678,7 +2630,6 @@ out_free:
  */
 xfs_daddr_t
 xfs_da_reada_buf(
-	struct xfs_trans	*trans,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
 	xfs_daddr_t		mappedbno,
@@ -2692,7 +2643,7 @@ xfs_da_reada_buf(
 
 	mapp = &map;
 	nmap = 1;
-	error = xfs_dabuf_map(trans, dp, bno, mappedbno, whichfork,
+	error = xfs_dabuf_map(dp, bno, mappedbno, whichfork,
 				&mapp, &nmap);
 	if (error) {
 		/* mapping a hole is not an error, but we don't continue */
diff --git a/fs/xfs/xfs_da_btree.h b/fs/xfs/libxfs/xfs_da_btree.h
index 6e95ea79f5d7..6e153e399a77 100644
--- a/fs/xfs/xfs_da_btree.h
+++ b/fs/xfs/libxfs/xfs_da_btree.h
@@ -25,6 +25,23 @@ struct xfs_trans;
 struct zone;
 struct xfs_dir_ops;
 
+/*
+ * Directory/attribute geometry information. There will be one of these for each
+ * data fork type, and it will be passed around via the xfs_da_args. Global
+ * structures will be attached to the xfs_mount.
+ */
+struct xfs_da_geometry {
+	int		blksize;	/* da block size in bytes */
+	int		fsbcount;	/* da block size in filesystem blocks */
+	uint8_t		fsblog;		/* log2 of _filesystem_ block size */
+	uint8_t		blklog;		/* log2 of da block size */
+	uint		node_ents;	/* # of entries in a danode */
+	int		magicpct;	/* 37% of block size in bytes */
+	xfs_dablk_t	datablk;	/* blockno of dir data v2 */
+	xfs_dablk_t	leafblk;	/* blockno of leaf data v2 */
+	xfs_dablk_t	freeblk;	/* blockno of free data v2 */
+};
+
 /*========================================================================
  * Btree searching and modification structure definitions.
  *========================================================================*/
@@ -42,6 +59,7 @@ enum xfs_dacmp {
  * Structure to ease passing around component names.
  */
 typedef struct xfs_da_args {
+	struct xfs_da_geometry *geo;	/* da block geometry */
 	const __uint8_t	*name;		/* string (maybe not NULL terminated) */
 	int		namelen;	/* length of string (maybe no NULL) */
 	__uint8_t	filetype;	/* filetype of inode for directories */
@@ -60,10 +78,12 @@ typedef struct xfs_da_args {
 	int		index;		/* index of attr of interest in blk */
 	xfs_dablk_t	rmtblkno;	/* remote attr value starting blkno */
 	int		rmtblkcnt;	/* remote attr value block count */
+	int		rmtvaluelen;	/* remote attr value length in bytes */
 	xfs_dablk_t	blkno2;		/* blkno of 2nd attr leaf of interest */
 	int		index2;		/* index of 2nd attr in blk */
 	xfs_dablk_t	rmtblkno2;	/* remote attr value starting blkno */
 	int		rmtblkcnt2;	/* remote attr value block count */
+	int		rmtvaluelen2;	/* remote attr value length in bytes */
 	int		op_flags;	/* operation flags */
 	enum xfs_dacmp	cmpresult;	/* name compare result for lookups */
 } xfs_da_args_t;
@@ -108,8 +128,6 @@ typedef struct xfs_da_state_path {
 typedef struct xfs_da_state {
 	xfs_da_args_t		*args;		/* filename arguments */
 	struct xfs_mount	*mp;		/* filesystem mount point */
-	unsigned int		blocksize;	/* logical block size */
-	unsigned int		node_ents;	/* how many entries in danode */
 	xfs_da_state_path_t	path;		/* search/split paths */
 	xfs_da_state_path_t	altpath;	/* alternate path for join */
 	unsigned char		inleaf;		/* insert into 1->lf, 0->splf */
@@ -183,9 +201,9 @@ int	xfs_da_read_buf(struct xfs_trans *trans, struct xfs_inode *dp,
 			       xfs_dablk_t bno, xfs_daddr_t mappedbno,
 			       struct xfs_buf **bpp, int whichfork,
 			       const struct xfs_buf_ops *ops);
-xfs_daddr_t	xfs_da_reada_buf(struct xfs_trans *trans, struct xfs_inode *dp,
-				xfs_dablk_t bno, xfs_daddr_t mapped_bno,
-				int whichfork, const struct xfs_buf_ops *ops);
+xfs_daddr_t	xfs_da_reada_buf(struct xfs_inode *dp, xfs_dablk_t bno,
+				xfs_daddr_t mapped_bno, int whichfork,
+				const struct xfs_buf_ops *ops);
 int	xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno,
 					  struct xfs_buf *dead_buf);
 
diff --git a/fs/xfs/xfs_da_format.c b/fs/xfs/libxfs/xfs_da_format.c
index e6c83e1fbc8a..c9aee52a37e2 100644
--- a/fs/xfs/xfs_da_format.c
+++ b/fs/xfs/libxfs/xfs_da_format.c
@@ -26,8 +26,10 @@
 #include "xfs_ag.h"
 #include "xfs_mount.h"
 #include "xfs_da_format.h"
+#include "xfs_da_btree.h"
 #include "xfs_inode.h"
 #include "xfs_dir2.h"
+#include "xfs_dir2_priv.h"
 
 /*
  * Shortform directory ops
@@ -425,9 +427,9 @@ xfs_dir3_data_unused_p(struct xfs_dir2_data_hdr *hdr)
  * Directory Leaf block operations
  */
 static int
-xfs_dir2_max_leaf_ents(struct xfs_mount *mp)
+xfs_dir2_max_leaf_ents(struct xfs_da_geometry *geo)
 {
-	return (mp->m_dirblksize - sizeof(struct xfs_dir2_leaf_hdr)) /
+	return (geo->blksize - sizeof(struct xfs_dir2_leaf_hdr)) /
 		(uint)sizeof(struct xfs_dir2_leaf_entry);
 }
 
@@ -438,9 +440,9 @@ xfs_dir2_leaf_ents_p(struct xfs_dir2_leaf *lp)
 }
 
 static int
-xfs_dir3_max_leaf_ents(struct xfs_mount *mp)
+xfs_dir3_max_leaf_ents(struct xfs_da_geometry *geo)
 {
-	return (mp->m_dirblksize - sizeof(struct xfs_dir3_leaf_hdr)) /
+	return (geo->blksize - sizeof(struct xfs_dir3_leaf_hdr)) /
 		(uint)sizeof(struct xfs_dir2_leaf_entry);
 }
 
@@ -591,9 +593,9 @@ xfs_da3_node_hdr_to_disk(
  * Directory free space block operations
  */
 static int
-xfs_dir2_free_max_bests(struct xfs_mount *mp)
+xfs_dir2_free_max_bests(struct xfs_da_geometry *geo)
 {
-	return (mp->m_dirblksize - sizeof(struct xfs_dir2_free_hdr)) /
+	return (geo->blksize - sizeof(struct xfs_dir2_free_hdr)) /
 		sizeof(xfs_dir2_data_off_t);
 }
 
@@ -607,24 +609,25 @@ xfs_dir2_free_bests_p(struct xfs_dir2_free *free)
  * Convert data space db to the corresponding free db.
  */
 static xfs_dir2_db_t
-xfs_dir2_db_to_fdb(struct xfs_mount *mp, xfs_dir2_db_t db)
+xfs_dir2_db_to_fdb(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
 {
-	return XFS_DIR2_FREE_FIRSTDB(mp) + db / xfs_dir2_free_max_bests(mp);
+	return xfs_dir2_byte_to_db(geo, XFS_DIR2_FREE_OFFSET) +
+			(db / xfs_dir2_free_max_bests(geo));
 }
 
 /*
  * Convert data space db to the corresponding index in a free db.
  */
 static int
-xfs_dir2_db_to_fdindex(struct xfs_mount *mp, xfs_dir2_db_t db)
+xfs_dir2_db_to_fdindex(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
 {
-	return db % xfs_dir2_free_max_bests(mp);
+	return db % xfs_dir2_free_max_bests(geo);
 }
 
 static int
-xfs_dir3_free_max_bests(struct xfs_mount *mp)
+xfs_dir3_free_max_bests(struct xfs_da_geometry *geo)
 {
-	return (mp->m_dirblksize - sizeof(struct xfs_dir3_free_hdr)) /
+	return (geo->blksize - sizeof(struct xfs_dir3_free_hdr)) /
 		sizeof(xfs_dir2_data_off_t);
 }
 
@@ -638,18 +641,19 @@ xfs_dir3_free_bests_p(struct xfs_dir2_free *free)
  * Convert data space db to the corresponding free db.
  */
 static xfs_dir2_db_t
-xfs_dir3_db_to_fdb(struct xfs_mount *mp, xfs_dir2_db_t db)
+xfs_dir3_db_to_fdb(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
 {
-	return XFS_DIR2_FREE_FIRSTDB(mp) + db / xfs_dir3_free_max_bests(mp);
+	return xfs_dir2_byte_to_db(geo, XFS_DIR2_FREE_OFFSET) +
+			(db / xfs_dir3_free_max_bests(geo));
 }
 
 /*
  * Convert data space db to the corresponding index in a free db.
  */
 static int
-xfs_dir3_db_to_fdindex(struct xfs_mount *mp, xfs_dir2_db_t db)
+xfs_dir3_db_to_fdindex(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
 {
-	return db % xfs_dir3_free_max_bests(mp);
+	return db % xfs_dir3_free_max_bests(geo);
 }
 
 static void
diff --git a/fs/xfs/xfs_da_format.h b/fs/xfs/libxfs/xfs_da_format.h
index a19d3f8f639c..0a49b0286372 100644
--- a/fs/xfs/xfs_da_format.h
+++ b/fs/xfs/libxfs/xfs_da_format.h
@@ -19,10 +19,6 @@
 #ifndef __XFS_DA_FORMAT_H__
 #define __XFS_DA_FORMAT_H__
 
-/*========================================================================
- * Directory Structure when greater than XFS_LBSIZE(mp) bytes.
- *========================================================================*/
-
 /*
  * This structure is common to both leaf nodes and non-leaf nodes in the Btree.
  *
@@ -122,8 +118,6 @@ struct xfs_da3_icnode_hdr {
 	__uint16_t	level;
 };
 
-#define	XFS_LBSIZE(mp)	(mp)->m_sb.sb_blocksize
-
 /*
  * Directory version 2.
  *
@@ -330,8 +324,6 @@ xfs_dir2_sf_firstentry(struct xfs_dir2_sf_hdr *hdr)
 #define	XFS_DIR2_SPACE_SIZE	(1ULL << (32 + XFS_DIR2_DATA_ALIGN_LOG))
 #define	XFS_DIR2_DATA_SPACE	0
 #define	XFS_DIR2_DATA_OFFSET	(XFS_DIR2_DATA_SPACE * XFS_DIR2_SPACE_SIZE)
-#define	XFS_DIR2_DATA_FIRSTDB(mp)	\
-	xfs_dir2_byte_to_db(mp, XFS_DIR2_DATA_OFFSET)
 
 /*
  * Describe a free area in the data block.
@@ -456,8 +448,6 @@ xfs_dir2_data_unused_tag_p(struct xfs_dir2_data_unused *dup)
  */
 #define	XFS_DIR2_LEAF_SPACE	1
 #define	XFS_DIR2_LEAF_OFFSET	(XFS_DIR2_LEAF_SPACE * XFS_DIR2_SPACE_SIZE)
-#define	XFS_DIR2_LEAF_FIRSTDB(mp)	\
-	xfs_dir2_byte_to_db(mp, XFS_DIR2_LEAF_OFFSET)
 
 /*
  * Leaf block header.
@@ -514,17 +504,6 @@ struct xfs_dir3_leaf {
 #define XFS_DIR3_LEAF_CRC_OFF  offsetof(struct xfs_dir3_leaf_hdr, info.crc)
 
 /*
- * Get address of the bestcount field in the single-leaf block.
- */
-static inline struct xfs_dir2_leaf_tail *
-xfs_dir2_leaf_tail_p(struct xfs_mount *mp, struct xfs_dir2_leaf *lp)
-{
-	return (struct xfs_dir2_leaf_tail *)
-		((char *)lp + mp->m_dirblksize -
-		  sizeof(struct xfs_dir2_leaf_tail));
-}
-
-/*
  * Get address of the bests array in the single-leaf block.
  */
 static inline __be16 *
@@ -534,123 +513,6 @@ xfs_dir2_leaf_bests_p(struct xfs_dir2_leaf_tail *ltp)
 }
 
 /*
- * DB blocks here are logical directory block numbers, not filesystem blocks.
- */
-
-/*
- * Convert dataptr to byte in file space
- */
-static inline xfs_dir2_off_t
-xfs_dir2_dataptr_to_byte(struct xfs_mount *mp, xfs_dir2_dataptr_t dp)
-{
-	return (xfs_dir2_off_t)dp << XFS_DIR2_DATA_ALIGN_LOG;
-}
-
-/*
- * Convert byte in file space to dataptr.  It had better be aligned.
- */
-static inline xfs_dir2_dataptr_t
-xfs_dir2_byte_to_dataptr(struct xfs_mount *mp, xfs_dir2_off_t by)
-{
-	return (xfs_dir2_dataptr_t)(by >> XFS_DIR2_DATA_ALIGN_LOG);
-}
-
-/*
- * Convert byte in space to (DB) block
- */
-static inline xfs_dir2_db_t
-xfs_dir2_byte_to_db(struct xfs_mount *mp, xfs_dir2_off_t by)
-{
-	return (xfs_dir2_db_t)
-		(by >> (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog));
-}
-
-/*
- * Convert dataptr to a block number
- */
-static inline xfs_dir2_db_t
-xfs_dir2_dataptr_to_db(struct xfs_mount *mp, xfs_dir2_dataptr_t dp)
-{
-	return xfs_dir2_byte_to_db(mp, xfs_dir2_dataptr_to_byte(mp, dp));
-}
-
-/*
- * Convert byte in space to offset in a block
- */
-static inline xfs_dir2_data_aoff_t
-xfs_dir2_byte_to_off(struct xfs_mount *mp, xfs_dir2_off_t by)
-{
-	return (xfs_dir2_data_aoff_t)(by &
-		((1 << (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog)) - 1));
-}
-
-/*
- * Convert dataptr to a byte offset in a block
- */
-static inline xfs_dir2_data_aoff_t
-xfs_dir2_dataptr_to_off(struct xfs_mount *mp, xfs_dir2_dataptr_t dp)
-{
-	return xfs_dir2_byte_to_off(mp, xfs_dir2_dataptr_to_byte(mp, dp));
-}
-
-/*
- * Convert block and offset to byte in space
- */
-static inline xfs_dir2_off_t
-xfs_dir2_db_off_to_byte(struct xfs_mount *mp, xfs_dir2_db_t db,
-			xfs_dir2_data_aoff_t o)
-{
-	return ((xfs_dir2_off_t)db <<
-		(mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog)) + o;
-}
-
-/*
- * Convert block (DB) to block (dablk)
- */
-static inline xfs_dablk_t
-xfs_dir2_db_to_da(struct xfs_mount *mp, xfs_dir2_db_t db)
-{
-	return (xfs_dablk_t)(db << mp->m_sb.sb_dirblklog);
-}
-
-/*
- * Convert byte in space to (DA) block
- */
-static inline xfs_dablk_t
-xfs_dir2_byte_to_da(struct xfs_mount *mp, xfs_dir2_off_t by)
-{
-	return xfs_dir2_db_to_da(mp, xfs_dir2_byte_to_db(mp, by));
-}
-
-/*
- * Convert block and offset to dataptr
- */
-static inline xfs_dir2_dataptr_t
-xfs_dir2_db_off_to_dataptr(struct xfs_mount *mp, xfs_dir2_db_t db,
-			   xfs_dir2_data_aoff_t o)
-{
-	return xfs_dir2_byte_to_dataptr(mp, xfs_dir2_db_off_to_byte(mp, db, o));
-}
-
-/*
- * Convert block (dablk) to block (DB)
- */
-static inline xfs_dir2_db_t
-xfs_dir2_da_to_db(struct xfs_mount *mp, xfs_dablk_t da)
-{
-	return (xfs_dir2_db_t)(da >> mp->m_sb.sb_dirblklog);
-}
-
-/*
- * Convert block (dablk) to byte offset in space
- */
-static inline xfs_dir2_off_t
-xfs_dir2_da_to_byte(struct xfs_mount *mp, xfs_dablk_t da)
-{
-	return xfs_dir2_db_off_to_byte(mp, xfs_dir2_da_to_db(mp, da), 0);
-}
-
-/*
  * Free space block defintions for the node format.
  */
 
@@ -659,8 +521,6 @@ xfs_dir2_da_to_byte(struct xfs_mount *mp, xfs_dablk_t da)
  */
 #define	XFS_DIR2_FREE_SPACE	2
 #define	XFS_DIR2_FREE_OFFSET	(XFS_DIR2_FREE_SPACE * XFS_DIR2_SPACE_SIZE)
-#define	XFS_DIR2_FREE_FIRSTDB(mp)	\
-	xfs_dir2_byte_to_db(mp, XFS_DIR2_FREE_OFFSET)
 
 typedef	struct xfs_dir2_free_hdr {
 	__be32			magic;		/* XFS_DIR2_FREE_MAGIC */
@@ -736,16 +596,6 @@ typedef struct xfs_dir2_block_tail {
 } xfs_dir2_block_tail_t;
 
 /*
- * Pointer to the leaf header embedded in a data block (1-block format)
- */
-static inline struct xfs_dir2_block_tail *
-xfs_dir2_block_tail_p(struct xfs_mount *mp, struct xfs_dir2_data_hdr *hdr)
-{
-	return ((struct xfs_dir2_block_tail *)
-		((char *)hdr + mp->m_dirblksize)) - 1;
-}
-
-/*
  * Pointer to the leaf entries embedded in a data block (1-block format)
  */
 static inline struct xfs_dir2_leaf_entry *
@@ -764,10 +614,6 @@ xfs_dir2_block_leaf_p(struct xfs_dir2_block_tail *btp)
  * of an attribute name may not be unique, we may have duplicate keys.  The
  * internal links in the Btree are logical block offsets into the file.
  *
- *========================================================================
- * Attribute structure when equal to XFS_LBSIZE(mp) bytes.
- *========================================================================
- *
  * Struct leaf_entry's are packed from the top.  Name/values grow from the
  * bottom but are not packed.  The freemap contains run-length-encoded entries
  * for the free bytes after the leaf_entry's, but only the N largest such,
diff --git a/fs/xfs/xfs_dinode.h b/fs/xfs/libxfs/xfs_dinode.h
index e5869b50dc41..623bbe8fd921 100644
--- a/fs/xfs/xfs_dinode.h
+++ b/fs/xfs/libxfs/xfs_dinode.h
@@ -89,6 +89,8 @@ typedef struct xfs_dinode {
 	/* structure must be padded to 64 bit alignment */
 } xfs_dinode_t;
 
+#define XFS_DINODE_CRC_OFF	offsetof(struct xfs_dinode, di_crc)
+
 #define DI_MAX_FLUSH 0xffff
 
 /*
diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c
index ce16ef02997a..6cef22152fd6 100644
--- a/fs/xfs/xfs_dir2.c
+++ b/fs/xfs/libxfs/xfs_dir2.c
@@ -85,38 +85,74 @@ static struct xfs_nameops xfs_ascii_ci_nameops = {
 	.compname	= xfs_ascii_ci_compname,
 };
 
-void
-xfs_dir_mount(
-	xfs_mount_t	*mp)
+int
+xfs_da_mount(
+	struct xfs_mount	*mp)
 {
-	int	nodehdr_size;
+	struct xfs_da_geometry	*dageo;
+	int			nodehdr_size;
 
 
-	ASSERT(xfs_sb_version_hasdirv2(&mp->m_sb));
+	ASSERT(mp->m_sb.sb_versionnum & XFS_SB_VERSION_DIRV2BIT);
 	ASSERT((1 << (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog)) <=
 	       XFS_MAX_BLOCKSIZE);
 
 	mp->m_dir_inode_ops = xfs_dir_get_ops(mp, NULL);
 	mp->m_nondir_inode_ops = xfs_nondir_get_ops(mp, NULL);
 
-	mp->m_dirblksize = 1 << (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog);
-	mp->m_dirblkfsbs = 1 << mp->m_sb.sb_dirblklog;
-	mp->m_dirdatablk = xfs_dir2_db_to_da(mp, XFS_DIR2_DATA_FIRSTDB(mp));
-	mp->m_dirleafblk = xfs_dir2_db_to_da(mp, XFS_DIR2_LEAF_FIRSTDB(mp));
-	mp->m_dirfreeblk = xfs_dir2_db_to_da(mp, XFS_DIR2_FREE_FIRSTDB(mp));
-
 	nodehdr_size = mp->m_dir_inode_ops->node_hdr_size;
-	mp->m_attr_node_ents = (mp->m_sb.sb_blocksize - nodehdr_size) /
+	mp->m_dir_geo = kmem_zalloc(sizeof(struct xfs_da_geometry),
+				    KM_SLEEP | KM_MAYFAIL);
+	mp->m_attr_geo = kmem_zalloc(sizeof(struct xfs_da_geometry),
+				     KM_SLEEP | KM_MAYFAIL);
+	if (!mp->m_dir_geo || !mp->m_attr_geo) {
+		kmem_free(mp->m_dir_geo);
+		kmem_free(mp->m_attr_geo);
+		return -ENOMEM;
+	}
+
+	/* set up directory geometry */
+	dageo = mp->m_dir_geo;
+	dageo->blklog = mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog;
+	dageo->fsblog = mp->m_sb.sb_blocklog;
+	dageo->blksize = 1 << dageo->blklog;
+	dageo->fsbcount = 1 << mp->m_sb.sb_dirblklog;
+
+	/*
+	 * Now we've set up the block conversion variables, we can calculate the
+	 * segment block constants using the geometry structure.
+	 */
+	dageo->datablk = xfs_dir2_byte_to_da(dageo, XFS_DIR2_DATA_OFFSET);
+	dageo->leafblk = xfs_dir2_byte_to_da(dageo, XFS_DIR2_LEAF_OFFSET);
+	dageo->freeblk = xfs_dir2_byte_to_da(dageo, XFS_DIR2_FREE_OFFSET);
+	dageo->node_ents = (dageo->blksize - nodehdr_size) /
 				(uint)sizeof(xfs_da_node_entry_t);
-	mp->m_dir_node_ents = (mp->m_dirblksize - nodehdr_size) /
+	dageo->magicpct = (dageo->blksize * 37) / 100;
+
+	/* set up attribute geometry - single fsb only */
+	dageo = mp->m_attr_geo;
+	dageo->blklog = mp->m_sb.sb_blocklog;
+	dageo->fsblog = mp->m_sb.sb_blocklog;
+	dageo->blksize = 1 << dageo->blklog;
+	dageo->fsbcount = 1;
+	dageo->node_ents = (dageo->blksize - nodehdr_size) /
 				(uint)sizeof(xfs_da_node_entry_t);
+	dageo->magicpct = (dageo->blksize * 37) / 100;
 
-	mp->m_dir_magicpct = (mp->m_dirblksize * 37) / 100;
 	if (xfs_sb_version_hasasciici(&mp->m_sb))
 		mp->m_dirnameops = &xfs_ascii_ci_nameops;
 	else
 		mp->m_dirnameops = &xfs_default_nameops;
 
+	return 0;
+}
+
+void
+xfs_da_unmount(
+	struct xfs_mount	*mp)
+{
+	kmem_free(mp->m_dir_geo);
+	kmem_free(mp->m_attr_geo);
 }
 
 /*
@@ -166,7 +202,7 @@ xfs_dir_ino_validate(
 		xfs_warn(mp, "Invalid inode number 0x%Lx",
 				(unsigned long long) ino);
 		XFS_ERROR_REPORT("xfs_dir_ino_validate", XFS_ERRLEVEL_LOW, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -180,16 +216,24 @@ xfs_dir_init(
 	xfs_inode_t	*dp,
 	xfs_inode_t	*pdp)
 {
-	xfs_da_args_t	args;
+	struct xfs_da_args *args;
 	int		error;
 
-	memset((char *)&args, 0, sizeof(args));
-	args.dp = dp;
-	args.trans = tp;
 	ASSERT(S_ISDIR(dp->i_d.di_mode));
-	if ((error = xfs_dir_ino_validate(tp->t_mountp, pdp->i_ino)))
+	error = xfs_dir_ino_validate(tp->t_mountp, pdp->i_ino);
+	if (error)
 		return error;
-	return xfs_dir2_sf_create(&args, pdp->i_ino);
+
+	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
+	if (!args)
+		return -ENOMEM;
+
+	args->geo = dp->i_mount->m_dir_geo;
+	args->dp = dp;
+	args->trans = tp;
+	error = xfs_dir2_sf_create(args, pdp->i_ino);
+	kmem_free(args);
+	return error;
 }
 
 /*
@@ -205,41 +249,57 @@ xfs_dir_createname(
 	xfs_bmap_free_t		*flist,		/* bmap's freeblock list */
 	xfs_extlen_t		total)		/* bmap's total block count */
 {
-	xfs_da_args_t		args;
+	struct xfs_da_args	*args;
 	int			rval;
 	int			v;		/* type-checking value */
 
 	ASSERT(S_ISDIR(dp->i_d.di_mode));
-	if ((rval = xfs_dir_ino_validate(tp->t_mountp, inum)))
+	rval = xfs_dir_ino_validate(tp->t_mountp, inum);
+	if (rval)
 		return rval;
 	XFS_STATS_INC(xs_dir_create);
 
-	memset(&args, 0, sizeof(xfs_da_args_t));
-	args.name = name->name;
-	args.namelen = name->len;
-	args.filetype = name->type;
-	args.hashval = dp->i_mount->m_dirnameops->hashname(name);
-	args.inumber = inum;
-	args.dp = dp;
-	args.firstblock = first;
-	args.flist = flist;
-	args.total = total;
-	args.whichfork = XFS_DATA_FORK;
-	args.trans = tp;
-	args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT;
-
-	if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
-		rval = xfs_dir2_sf_addname(&args);
-	else if ((rval = xfs_dir2_isblock(tp, dp, &v)))
-		return rval;
-	else if (v)
-		rval = xfs_dir2_block_addname(&args);
-	else if ((rval = xfs_dir2_isleaf(tp, dp, &v)))
-		return rval;
-	else if (v)
-		rval = xfs_dir2_leaf_addname(&args);
+	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
+	if (!args)
+		return -ENOMEM;
+
+	args->geo = dp->i_mount->m_dir_geo;
+	args->name = name->name;
+	args->namelen = name->len;
+	args->filetype = name->type;
+	args->hashval = dp->i_mount->m_dirnameops->hashname(name);
+	args->inumber = inum;
+	args->dp = dp;
+	args->firstblock = first;
+	args->flist = flist;
+	args->total = total;
+	args->whichfork = XFS_DATA_FORK;
+	args->trans = tp;
+	args->op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT;
+
+	if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
+		rval = xfs_dir2_sf_addname(args);
+		goto out_free;
+	}
+
+	rval = xfs_dir2_isblock(args, &v);
+	if (rval)
+		goto out_free;
+	if (v) {
+		rval = xfs_dir2_block_addname(args);
+		goto out_free;
+	}
+
+	rval = xfs_dir2_isleaf(args, &v);
+	if (rval)
+		goto out_free;
+	if (v)
+		rval = xfs_dir2_leaf_addname(args);
 	else
-		rval = xfs_dir2_node_addname(&args);
+		rval = xfs_dir2_node_addname(args);
+
+out_free:
+	kmem_free(args);
 	return rval;
 }
 
@@ -254,18 +314,18 @@ xfs_dir_cilookup_result(
 	int		len)
 {
 	if (args->cmpresult == XFS_CMP_DIFFERENT)
-		return ENOENT;
+		return -ENOENT;
 	if (args->cmpresult != XFS_CMP_CASE ||
 					!(args->op_flags & XFS_DA_OP_CILOOKUP))
-		return EEXIST;
+		return -EEXIST;
 
 	args->value = kmem_alloc(len, KM_NOFS | KM_MAYFAIL);
 	if (!args->value)
-		return ENOMEM;
+		return -ENOMEM;
 
 	memcpy(args->value, name, len);
 	args->valuelen = len;
-	return EEXIST;
+	return -EEXIST;
 }
 
 /*
@@ -282,46 +342,67 @@ xfs_dir_lookup(
 	xfs_ino_t	*inum,		/* out: inode number */
 	struct xfs_name *ci_name)	/* out: actual name if CI match */
 {
-	xfs_da_args_t	args;
+	struct xfs_da_args *args;
 	int		rval;
 	int		v;		/* type-checking value */
 
 	ASSERT(S_ISDIR(dp->i_d.di_mode));
 	XFS_STATS_INC(xs_dir_lookup);
 
-	memset(&args, 0, sizeof(xfs_da_args_t));
-	args.name = name->name;
-	args.namelen = name->len;
-	args.filetype = name->type;
-	args.hashval = dp->i_mount->m_dirnameops->hashname(name);
-	args.dp = dp;
-	args.whichfork = XFS_DATA_FORK;
-	args.trans = tp;
-	args.op_flags = XFS_DA_OP_OKNOENT;
+	/*
+	 * We need to use KM_NOFS here so that lockdep will not throw false
+	 * positive deadlock warnings on a non-transactional lookup path. It is
+	 * safe to recurse into inode recalim in that case, but lockdep can't
+	 * easily be taught about it. Hence KM_NOFS avoids having to add more
+	 * lockdep Doing this avoids having to add a bunch of lockdep class
+	 * annotations into the reclaim path for the ilock.
+	 */
+	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
+	args->geo = dp->i_mount->m_dir_geo;
+	args->name = name->name;
+	args->namelen = name->len;
+	args->filetype = name->type;
+	args->hashval = dp->i_mount->m_dirnameops->hashname(name);
+	args->dp = dp;
+	args->whichfork = XFS_DATA_FORK;
+	args->trans = tp;
+	args->op_flags = XFS_DA_OP_OKNOENT;
 	if (ci_name)
-		args.op_flags |= XFS_DA_OP_CILOOKUP;
+		args->op_flags |= XFS_DA_OP_CILOOKUP;
 
-	if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
-		rval = xfs_dir2_sf_lookup(&args);
-	else if ((rval = xfs_dir2_isblock(tp, dp, &v)))
-		return rval;
-	else if (v)
-		rval = xfs_dir2_block_lookup(&args);
-	else if ((rval = xfs_dir2_isleaf(tp, dp, &v)))
-		return rval;
-	else if (v)
-		rval = xfs_dir2_leaf_lookup(&args);
+	if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
+		rval = xfs_dir2_sf_lookup(args);
+		goto out_check_rval;
+	}
+
+	rval = xfs_dir2_isblock(args, &v);
+	if (rval)
+		goto out_free;
+	if (v) {
+		rval = xfs_dir2_block_lookup(args);
+		goto out_check_rval;
+	}
+
+	rval = xfs_dir2_isleaf(args, &v);
+	if (rval)
+		goto out_free;
+	if (v)
+		rval = xfs_dir2_leaf_lookup(args);
 	else
-		rval = xfs_dir2_node_lookup(&args);
-	if (rval == EEXIST)
+		rval = xfs_dir2_node_lookup(args);
+
+out_check_rval:
+	if (rval == -EEXIST)
 		rval = 0;
 	if (!rval) {
-		*inum = args.inumber;
+		*inum = args->inumber;
 		if (ci_name) {
-			ci_name->name = args.value;
-			ci_name->len = args.valuelen;
+			ci_name->name = args->value;
+			ci_name->len = args->valuelen;
 		}
 	}
+out_free:
+	kmem_free(args);
 	return rval;
 }
 
@@ -338,38 +419,52 @@ xfs_dir_removename(
 	xfs_bmap_free_t	*flist,		/* bmap's freeblock list */
 	xfs_extlen_t	total)		/* bmap's total block count */
 {
-	xfs_da_args_t	args;
+	struct xfs_da_args *args;
 	int		rval;
 	int		v;		/* type-checking value */
 
 	ASSERT(S_ISDIR(dp->i_d.di_mode));
 	XFS_STATS_INC(xs_dir_remove);
 
-	memset(&args, 0, sizeof(xfs_da_args_t));
-	args.name = name->name;
-	args.namelen = name->len;
-	args.filetype = name->type;
-	args.hashval = dp->i_mount->m_dirnameops->hashname(name);
-	args.inumber = ino;
-	args.dp = dp;
-	args.firstblock = first;
-	args.flist = flist;
-	args.total = total;
-	args.whichfork = XFS_DATA_FORK;
-	args.trans = tp;
-
-	if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
-		rval = xfs_dir2_sf_removename(&args);
-	else if ((rval = xfs_dir2_isblock(tp, dp, &v)))
-		return rval;
-	else if (v)
-		rval = xfs_dir2_block_removename(&args);
-	else if ((rval = xfs_dir2_isleaf(tp, dp, &v)))
-		return rval;
-	else if (v)
-		rval = xfs_dir2_leaf_removename(&args);
+	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
+	if (!args)
+		return -ENOMEM;
+
+	args->geo = dp->i_mount->m_dir_geo;
+	args->name = name->name;
+	args->namelen = name->len;
+	args->filetype = name->type;
+	args->hashval = dp->i_mount->m_dirnameops->hashname(name);
+	args->inumber = ino;
+	args->dp = dp;
+	args->firstblock = first;
+	args->flist = flist;
+	args->total = total;
+	args->whichfork = XFS_DATA_FORK;
+	args->trans = tp;
+
+	if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
+		rval = xfs_dir2_sf_removename(args);
+		goto out_free;
+	}
+
+	rval = xfs_dir2_isblock(args, &v);
+	if (rval)
+		goto out_free;
+	if (v) {
+		rval = xfs_dir2_block_removename(args);
+		goto out_free;
+	}
+
+	rval = xfs_dir2_isleaf(args, &v);
+	if (rval)
+		goto out_free;
+	if (v)
+		rval = xfs_dir2_leaf_removename(args);
 	else
-		rval = xfs_dir2_node_removename(&args);
+		rval = xfs_dir2_node_removename(args);
+out_free:
+	kmem_free(args);
 	return rval;
 }
 
@@ -386,40 +481,55 @@ xfs_dir_replace(
 	xfs_bmap_free_t	*flist,		/* bmap's freeblock list */
 	xfs_extlen_t	total)		/* bmap's total block count */
 {
-	xfs_da_args_t	args;
+	struct xfs_da_args *args;
 	int		rval;
 	int		v;		/* type-checking value */
 
 	ASSERT(S_ISDIR(dp->i_d.di_mode));
 
-	if ((rval = xfs_dir_ino_validate(tp->t_mountp, inum)))
+	rval = xfs_dir_ino_validate(tp->t_mountp, inum);
+	if (rval)
 		return rval;
 
-	memset(&args, 0, sizeof(xfs_da_args_t));
-	args.name = name->name;
-	args.namelen = name->len;
-	args.filetype = name->type;
-	args.hashval = dp->i_mount->m_dirnameops->hashname(name);
-	args.inumber = inum;
-	args.dp = dp;
-	args.firstblock = first;
-	args.flist = flist;
-	args.total = total;
-	args.whichfork = XFS_DATA_FORK;
-	args.trans = tp;
-
-	if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
-		rval = xfs_dir2_sf_replace(&args);
-	else if ((rval = xfs_dir2_isblock(tp, dp, &v)))
-		return rval;
-	else if (v)
-		rval = xfs_dir2_block_replace(&args);
-	else if ((rval = xfs_dir2_isleaf(tp, dp, &v)))
-		return rval;
-	else if (v)
-		rval = xfs_dir2_leaf_replace(&args);
+	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
+	if (!args)
+		return -ENOMEM;
+
+	args->geo = dp->i_mount->m_dir_geo;
+	args->name = name->name;
+	args->namelen = name->len;
+	args->filetype = name->type;
+	args->hashval = dp->i_mount->m_dirnameops->hashname(name);
+	args->inumber = inum;
+	args->dp = dp;
+	args->firstblock = first;
+	args->flist = flist;
+	args->total = total;
+	args->whichfork = XFS_DATA_FORK;
+	args->trans = tp;
+
+	if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
+		rval = xfs_dir2_sf_replace(args);
+		goto out_free;
+	}
+
+	rval = xfs_dir2_isblock(args, &v);
+	if (rval)
+		goto out_free;
+	if (v) {
+		rval = xfs_dir2_block_replace(args);
+		goto out_free;
+	}
+
+	rval = xfs_dir2_isleaf(args, &v);
+	if (rval)
+		goto out_free;
+	if (v)
+		rval = xfs_dir2_leaf_replace(args);
 	else
-		rval = xfs_dir2_node_replace(&args);
+		rval = xfs_dir2_node_replace(args);
+out_free:
+	kmem_free(args);
 	return rval;
 }
 
@@ -434,7 +544,7 @@ xfs_dir_canenter(
 	struct xfs_name	*name,		/* name of entry to add */
 	uint		resblks)
 {
-	xfs_da_args_t	args;
+	struct xfs_da_args *args;
 	int		rval;
 	int		v;		/* type-checking value */
 
@@ -443,29 +553,43 @@ xfs_dir_canenter(
 
 	ASSERT(S_ISDIR(dp->i_d.di_mode));
 
-	memset(&args, 0, sizeof(xfs_da_args_t));
-	args.name = name->name;
-	args.namelen = name->len;
-	args.filetype = name->type;
-	args.hashval = dp->i_mount->m_dirnameops->hashname(name);
-	args.dp = dp;
-	args.whichfork = XFS_DATA_FORK;
-	args.trans = tp;
-	args.op_flags = XFS_DA_OP_JUSTCHECK | XFS_DA_OP_ADDNAME |
+	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
+	if (!args)
+		return -ENOMEM;
+
+	args->geo = dp->i_mount->m_dir_geo;
+	args->name = name->name;
+	args->namelen = name->len;
+	args->filetype = name->type;
+	args->hashval = dp->i_mount->m_dirnameops->hashname(name);
+	args->dp = dp;
+	args->whichfork = XFS_DATA_FORK;
+	args->trans = tp;
+	args->op_flags = XFS_DA_OP_JUSTCHECK | XFS_DA_OP_ADDNAME |
 							XFS_DA_OP_OKNOENT;
 
-	if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
-		rval = xfs_dir2_sf_addname(&args);
-	else if ((rval = xfs_dir2_isblock(tp, dp, &v)))
-		return rval;
-	else if (v)
-		rval = xfs_dir2_block_addname(&args);
-	else if ((rval = xfs_dir2_isleaf(tp, dp, &v)))
-		return rval;
-	else if (v)
-		rval = xfs_dir2_leaf_addname(&args);
+	if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
+		rval = xfs_dir2_sf_addname(args);
+		goto out_free;
+	}
+
+	rval = xfs_dir2_isblock(args, &v);
+	if (rval)
+		goto out_free;
+	if (v) {
+		rval = xfs_dir2_block_addname(args);
+		goto out_free;
+	}
+
+	rval = xfs_dir2_isleaf(args, &v);
+	if (rval)
+		goto out_free;
+	if (v)
+		rval = xfs_dir2_leaf_addname(args);
 	else
-		rval = xfs_dir2_node_addname(&args);
+		rval = xfs_dir2_node_addname(args);
+out_free:
+	kmem_free(args);
 	return rval;
 }
 
@@ -497,13 +621,13 @@ xfs_dir2_grow_inode(
 	 * Set lowest possible block in the space requested.
 	 */
 	bno = XFS_B_TO_FSBT(mp, space * XFS_DIR2_SPACE_SIZE);
-	count = mp->m_dirblkfsbs;
+	count = args->geo->fsbcount;
 
 	error = xfs_da_grow_inode_int(args, &bno, count);
 	if (error)
 		return error;
 
-	*dbp = xfs_dir2_da_to_db(mp, (xfs_dablk_t)bno);
+	*dbp = xfs_dir2_da_to_db(args->geo, (xfs_dablk_t)bno);
 
 	/*
 	 * Update file's size if this is the data space and it grew.
@@ -525,19 +649,16 @@ xfs_dir2_grow_inode(
  */
 int
 xfs_dir2_isblock(
-	xfs_trans_t	*tp,
-	xfs_inode_t	*dp,
-	int		*vp)		/* out: 1 is block, 0 is not block */
+	struct xfs_da_args	*args,
+	int			*vp)	/* out: 1 is block, 0 is not block */
 {
-	xfs_fileoff_t	last;		/* last file offset */
-	xfs_mount_t	*mp;
-	int		rval;
+	xfs_fileoff_t		last;	/* last file offset */
+	int			rval;
 
-	mp = dp->i_mount;
-	if ((rval = xfs_bmap_last_offset(tp, dp, &last, XFS_DATA_FORK)))
+	if ((rval = xfs_bmap_last_offset(args->dp, &last, XFS_DATA_FORK)))
 		return rval;
-	rval = XFS_FSB_TO_B(mp, last) == mp->m_dirblksize;
-	ASSERT(rval == 0 || dp->i_d.di_size == mp->m_dirblksize);
+	rval = XFS_FSB_TO_B(args->dp->i_mount, last) == args->geo->blksize;
+	ASSERT(rval == 0 || args->dp->i_d.di_size == args->geo->blksize);
 	*vp = rval;
 	return 0;
 }
@@ -547,18 +668,15 @@ xfs_dir2_isblock(
  */
 int
 xfs_dir2_isleaf(
-	xfs_trans_t	*tp,
-	xfs_inode_t	*dp,
-	int		*vp)		/* out: 1 is leaf, 0 is not leaf */
+	struct xfs_da_args	*args,
+	int			*vp)	/* out: 1 is block, 0 is not block */
 {
-	xfs_fileoff_t	last;		/* last file offset */
-	xfs_mount_t	*mp;
-	int		rval;
+	xfs_fileoff_t		last;	/* last file offset */
+	int			rval;
 
-	mp = dp->i_mount;
-	if ((rval = xfs_bmap_last_offset(tp, dp, &last, XFS_DATA_FORK)))
+	if ((rval = xfs_bmap_last_offset(args->dp, &last, XFS_DATA_FORK)))
 		return rval;
-	*vp = last == mp->m_dirleafblk + (1 << mp->m_sb.sb_dirblklog);
+	*vp = last == args->geo->leafblk + args->geo->fsbcount;
 	return 0;
 }
 
@@ -586,11 +704,11 @@ xfs_dir2_shrink_inode(
 	dp = args->dp;
 	mp = dp->i_mount;
 	tp = args->trans;
-	da = xfs_dir2_db_to_da(mp, db);
+	da = xfs_dir2_db_to_da(args->geo, db);
 	/*
 	 * Unmap the fsblock(s).
 	 */
-	if ((error = xfs_bunmapi(tp, dp, da, mp->m_dirblkfsbs,
+	if ((error = xfs_bunmapi(tp, dp, da, args->geo->fsbcount,
 			XFS_BMAPI_METADATA, 0, args->firstblock, args->flist,
 			&done))) {
 		/*
@@ -617,12 +735,12 @@ xfs_dir2_shrink_inode(
 	/*
 	 * If it's not a data block, we're done.
 	 */
-	if (db >= XFS_DIR2_LEAF_FIRSTDB(mp))
+	if (db >= xfs_dir2_byte_to_db(args->geo, XFS_DIR2_LEAF_OFFSET))
 		return 0;
 	/*
 	 * If the block isn't the last one in the directory, we're done.
 	 */
-	if (dp->i_d.di_size > xfs_dir2_db_off_to_byte(mp, db + 1, 0))
+	if (dp->i_d.di_size > xfs_dir2_db_off_to_byte(args->geo, db + 1, 0))
 		return 0;
 	bno = da;
 	if ((error = xfs_bmap_last_before(tp, dp, &bno, XFS_DATA_FORK))) {
@@ -631,7 +749,7 @@ xfs_dir2_shrink_inode(
 		 */
 		return error;
 	}
-	if (db == mp->m_dirdatablk)
+	if (db == args->geo->datablk)
 		ASSERT(bno == 0);
 	else
 		ASSERT(bno > 0);
diff --git a/fs/xfs/xfs_dir2.h b/fs/xfs/libxfs/xfs_dir2.h
index cec70e0781ab..c8e86b0b5e99 100644
--- a/fs/xfs/xfs_dir2.h
+++ b/fs/xfs/libxfs/xfs_dir2.h
@@ -80,7 +80,7 @@ struct xfs_dir_ops {
 				    struct xfs_dir3_icleaf_hdr *from);
 	void	(*leaf_hdr_from_disk)(struct xfs_dir3_icleaf_hdr *to,
 				      struct xfs_dir2_leaf *from);
-	int	(*leaf_max_ents)(struct xfs_mount *mp);
+	int	(*leaf_max_ents)(struct xfs_da_geometry *geo);
 	struct xfs_dir2_leaf_entry *
 		(*leaf_ents_p)(struct xfs_dir2_leaf *lp);
 
@@ -97,10 +97,12 @@ struct xfs_dir_ops {
 				    struct xfs_dir3_icfree_hdr *from);
 	void	(*free_hdr_from_disk)(struct xfs_dir3_icfree_hdr *to,
 				      struct xfs_dir2_free *from);
-	int	(*free_max_bests)(struct xfs_mount *mp);
+	int	(*free_max_bests)(struct xfs_da_geometry *geo);
 	__be16 * (*free_bests_p)(struct xfs_dir2_free *free);
-	xfs_dir2_db_t (*db_to_fdb)(struct xfs_mount *mp, xfs_dir2_db_t db);
-	int	(*db_to_fdindex)(struct xfs_mount *mp, xfs_dir2_db_t db);
+	xfs_dir2_db_t (*db_to_fdb)(struct xfs_da_geometry *geo,
+				   xfs_dir2_db_t db);
+	int	(*db_to_fdindex)(struct xfs_da_geometry *geo,
+				 xfs_dir2_db_t db);
 };
 
 extern const struct xfs_dir_ops *
@@ -112,7 +114,9 @@ extern const struct xfs_dir_ops *
  * Generic directory interface routines
  */
 extern void xfs_dir_startup(void);
-extern void xfs_dir_mount(struct xfs_mount *mp);
+extern int xfs_da_mount(struct xfs_mount *mp);
+extern void xfs_da_unmount(struct xfs_mount *mp);
+
 extern int xfs_dir_isempty(struct xfs_inode *dp);
 extern int xfs_dir_init(struct xfs_trans *tp, struct xfs_inode *dp,
 				struct xfs_inode *pdp);
@@ -142,23 +146,23 @@ extern int xfs_dir2_sf_to_block(struct xfs_da_args *args);
 /*
  * Interface routines used by userspace utilities
  */
-extern int xfs_dir2_isblock(struct xfs_trans *tp, struct xfs_inode *dp, int *r);
-extern int xfs_dir2_isleaf(struct xfs_trans *tp, struct xfs_inode *dp, int *r);
+extern int xfs_dir2_isblock(struct xfs_da_args *args, int *r);
+extern int xfs_dir2_isleaf(struct xfs_da_args *args, int *r);
 extern int xfs_dir2_shrink_inode(struct xfs_da_args *args, xfs_dir2_db_t db,
 				struct xfs_buf *bp);
 
 extern void xfs_dir2_data_freescan(struct xfs_inode *dp,
 		struct xfs_dir2_data_hdr *hdr, int *loghead);
-extern void xfs_dir2_data_log_entry(struct xfs_trans *tp, struct xfs_inode *dp,
+extern void xfs_dir2_data_log_entry(struct xfs_da_args *args,
 		struct xfs_buf *bp, struct xfs_dir2_data_entry *dep);
-extern void xfs_dir2_data_log_header(struct xfs_trans *tp, struct xfs_inode *dp,
+extern void xfs_dir2_data_log_header(struct xfs_da_args *args,
 		struct xfs_buf *bp);
-extern void xfs_dir2_data_log_unused(struct xfs_trans *tp, struct xfs_buf *bp,
-		struct xfs_dir2_data_unused *dup);
-extern void xfs_dir2_data_make_free(struct xfs_trans *tp, struct xfs_inode *dp,
+extern void xfs_dir2_data_log_unused(struct xfs_da_args *args,
+		struct xfs_buf *bp, struct xfs_dir2_data_unused *dup);
+extern void xfs_dir2_data_make_free(struct xfs_da_args *args,
 		struct xfs_buf *bp, xfs_dir2_data_aoff_t offset,
 		xfs_dir2_data_aoff_t len, int *needlogp, int *needscanp);
-extern void xfs_dir2_data_use_free(struct xfs_trans *tp, struct xfs_inode *dp,
+extern void xfs_dir2_data_use_free(struct xfs_da_args *args,
 		struct xfs_buf *bp, struct xfs_dir2_data_unused *dup,
 		xfs_dir2_data_aoff_t offset, xfs_dir2_data_aoff_t len,
 		int *needlogp, int *needscanp);
diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/libxfs/xfs_dir2_block.c
index 90cdbf4b5f19..9628ceccfa02 100644
--- a/fs/xfs/xfs_dir2_block.c
+++ b/fs/xfs/libxfs/xfs_dir2_block.c
@@ -89,13 +89,14 @@ xfs_dir3_block_read_verify(
 {
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 
-	if ((xfs_sb_version_hascrc(&mp->m_sb) &&
-	     !xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
-					  XFS_DIR3_DATA_CRC_OFF)) ||
-	    !xfs_dir3_block_verify(bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
-	}
+	if (xfs_sb_version_hascrc(&mp->m_sb) &&
+	     !xfs_buf_verify_cksum(bp, XFS_DIR3_DATA_CRC_OFF))
+		xfs_buf_ioerror(bp, -EFSBADCRC);
+	else if (!xfs_dir3_block_verify(bp))
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+
+	if (bp->b_error)
+		xfs_verifier_error(bp);
 }
 
 static void
@@ -107,8 +108,8 @@ xfs_dir3_block_write_verify(
 	struct xfs_dir3_blk_hdr	*hdr3 = bp->b_addr;
 
 	if (!xfs_dir3_block_verify(bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
 		return;
 	}
 
@@ -118,7 +119,7 @@ xfs_dir3_block_write_verify(
 	if (bip)
 		hdr3->lsn = cpu_to_be64(bip->bli_item.li_lsn);
 
-	xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), XFS_DIR3_DATA_CRC_OFF);
+	xfs_buf_update_cksum(bp, XFS_DIR3_DATA_CRC_OFF);
 }
 
 const struct xfs_buf_ops xfs_dir3_block_buf_ops = {
@@ -135,7 +136,7 @@ xfs_dir3_block_read(
 	struct xfs_mount	*mp = dp->i_mount;
 	int			err;
 
-	err = xfs_da_read_buf(tp, dp, mp->m_dirdatablk, -1, bpp,
+	err = xfs_da_read_buf(tp, dp, mp->m_dir_geo->datablk, -1, bpp,
 				XFS_DATA_FORK, &xfs_dir3_block_buf_ops);
 	if (!err && tp)
 		xfs_trans_buf_set_type(tp, *bpp, XFS_BLFT_DIR_BLOCK_BUF);
@@ -280,8 +281,7 @@ out:
  */
 static void
 xfs_dir2_block_compact(
-	struct xfs_trans		*tp,
-	struct xfs_inode		*dp,
+	struct xfs_da_args		*args,
 	struct xfs_buf			*bp,
 	struct xfs_dir2_data_hdr	*hdr,
 	struct xfs_dir2_block_tail	*btp,
@@ -314,18 +314,17 @@ xfs_dir2_block_compact(
 	*lfloglow = toidx + 1 - (be32_to_cpu(btp->stale) - 1);
 	*lfloghigh -= be32_to_cpu(btp->stale) - 1;
 	be32_add_cpu(&btp->count, -(be32_to_cpu(btp->stale) - 1));
-	xfs_dir2_data_make_free(tp, dp, bp,
+	xfs_dir2_data_make_free(args, bp,
 		(xfs_dir2_data_aoff_t)((char *)blp - (char *)hdr),
 		(xfs_dir2_data_aoff_t)((be32_to_cpu(btp->stale) - 1) * sizeof(*blp)),
 		needlog, &needscan);
-	blp += be32_to_cpu(btp->stale) - 1;
 	btp->stale = cpu_to_be32(1);
 	/*
 	 * If we now need to rebuild the bestfree map, do so.
 	 * This needs to happen before the next call to use_free.
 	 */
 	if (needscan)
-		xfs_dir2_data_freescan(dp, hdr, needlog);
+		xfs_dir2_data_freescan(args->dp, hdr, needlog);
 }
 
 /*
@@ -377,7 +376,7 @@ xfs_dir2_block_addname(
 	 * Set up pointers to parts of the block.
 	 */
 	hdr = bp->b_addr;
-	btp = xfs_dir2_block_tail_p(mp, hdr);
+	btp = xfs_dir2_block_tail_p(args->geo, hdr);
 	blp = xfs_dir2_block_leaf_p(btp);
 
 	/*
@@ -393,7 +392,7 @@ xfs_dir2_block_addname(
 	if (args->op_flags & XFS_DA_OP_JUSTCHECK) {
 		xfs_trans_brelse(tp, bp);
 		if (!dup)
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		return 0;
 	}
 
@@ -403,7 +402,7 @@ xfs_dir2_block_addname(
 	if (!dup) {
 		/* Don't have a space reservation: return no-space.  */
 		if (args->total == 0)
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		/*
 		 * Convert to the next larger format.
 		 * Then add the new entry in that format.
@@ -420,7 +419,7 @@ xfs_dir2_block_addname(
 	 * If need to compact the leaf entries, do it now.
 	 */
 	if (compact) {
-		xfs_dir2_block_compact(tp, dp, bp, hdr, btp, blp, &needlog,
+		xfs_dir2_block_compact(args, bp, hdr, btp, blp, &needlog,
 				      &lfloghigh, &lfloglow);
 		/* recalculate blp post-compaction */
 		blp = xfs_dir2_block_leaf_p(btp);
@@ -455,7 +454,7 @@ xfs_dir2_block_addname(
 		/*
 		 * Mark the space needed for the new leaf entry, now in use.
 		 */
-		xfs_dir2_data_use_free(tp, dp, bp, enddup,
+		xfs_dir2_data_use_free(args, bp, enddup,
 			(xfs_dir2_data_aoff_t)
 			((char *)enddup - (char *)hdr + be16_to_cpu(enddup->length) -
 			 sizeof(*blp)),
@@ -536,13 +535,13 @@ xfs_dir2_block_addname(
 	 * Fill in the leaf entry.
 	 */
 	blp[mid].hashval = cpu_to_be32(args->hashval);
-	blp[mid].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp,
+	blp[mid].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(
 				(char *)dep - (char *)hdr));
 	xfs_dir2_block_log_leaf(tp, bp, lfloglow, lfloghigh);
 	/*
 	 * Mark space for the data entry used.
 	 */
-	xfs_dir2_data_use_free(tp, dp, bp, dup,
+	xfs_dir2_data_use_free(args, bp, dup,
 		(xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr),
 		(xfs_dir2_data_aoff_t)len, &needlog, &needscan);
 	/*
@@ -560,9 +559,9 @@ xfs_dir2_block_addname(
 	if (needscan)
 		xfs_dir2_data_freescan(dp, hdr, &needlog);
 	if (needlog)
-		xfs_dir2_data_log_header(tp, dp, bp);
+		xfs_dir2_data_log_header(args, bp);
 	xfs_dir2_block_log_tail(tp, bp);
-	xfs_dir2_data_log_entry(tp, dp, bp, dep);
+	xfs_dir2_data_log_entry(args, bp, dep);
 	xfs_dir3_data_check(dp, bp);
 	return 0;
 }
@@ -581,7 +580,7 @@ xfs_dir2_block_log_leaf(
 	xfs_dir2_leaf_entry_t	*blp;
 	xfs_dir2_block_tail_t	*btp;
 
-	btp = xfs_dir2_block_tail_p(tp->t_mountp, hdr);
+	btp = xfs_dir2_block_tail_p(tp->t_mountp->m_dir_geo, hdr);
 	blp = xfs_dir2_block_leaf_p(btp);
 	xfs_trans_log_buf(tp, bp, (uint)((char *)&blp[first] - (char *)hdr),
 		(uint)((char *)&blp[last + 1] - (char *)hdr - 1));
@@ -598,7 +597,7 @@ xfs_dir2_block_log_tail(
 	xfs_dir2_data_hdr_t	*hdr = bp->b_addr;
 	xfs_dir2_block_tail_t	*btp;
 
-	btp = xfs_dir2_block_tail_p(tp->t_mountp, hdr);
+	btp = xfs_dir2_block_tail_p(tp->t_mountp->m_dir_geo, hdr);
 	xfs_trans_log_buf(tp, bp, (uint)((char *)btp - (char *)hdr),
 		(uint)((char *)(btp + 1) - (char *)hdr - 1));
 }
@@ -633,13 +632,14 @@ xfs_dir2_block_lookup(
 	mp = dp->i_mount;
 	hdr = bp->b_addr;
 	xfs_dir3_data_check(dp, bp);
-	btp = xfs_dir2_block_tail_p(mp, hdr);
+	btp = xfs_dir2_block_tail_p(args->geo, hdr);
 	blp = xfs_dir2_block_leaf_p(btp);
 	/*
 	 * Get the offset from the leaf entry, to point to the data.
 	 */
 	dep = (xfs_dir2_data_entry_t *)((char *)hdr +
-		xfs_dir2_dataptr_to_off(mp, be32_to_cpu(blp[ent].address)));
+			xfs_dir2_dataptr_to_off(args->geo,
+						be32_to_cpu(blp[ent].address)));
 	/*
 	 * Fill in inode number, CI name if appropriate, release the block.
 	 */
@@ -647,7 +647,7 @@ xfs_dir2_block_lookup(
 	args->filetype = dp->d_ops->data_get_ftype(dep);
 	error = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
 	xfs_trans_brelse(args->trans, bp);
-	return XFS_ERROR(error);
+	return error;
 }
 
 /*
@@ -685,7 +685,7 @@ xfs_dir2_block_lookup_int(
 
 	hdr = bp->b_addr;
 	xfs_dir3_data_check(dp, bp);
-	btp = xfs_dir2_block_tail_p(mp, hdr);
+	btp = xfs_dir2_block_tail_p(args->geo, hdr);
 	blp = xfs_dir2_block_leaf_p(btp);
 	/*
 	 * Loop doing a binary search for our hash value.
@@ -703,7 +703,7 @@ xfs_dir2_block_lookup_int(
 		if (low > high) {
 			ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
 			xfs_trans_brelse(tp, bp);
-			return XFS_ERROR(ENOENT);
+			return -ENOENT;
 		}
 	}
 	/*
@@ -723,7 +723,7 @@ xfs_dir2_block_lookup_int(
 		 * Get pointer to the entry from the leaf.
 		 */
 		dep = (xfs_dir2_data_entry_t *)
-			((char *)hdr + xfs_dir2_dataptr_to_off(mp, addr));
+			((char *)hdr + xfs_dir2_dataptr_to_off(args->geo, addr));
 		/*
 		 * Compare name and if it's an exact match, return the index
 		 * and buffer. If it's the first case-insensitive match, store
@@ -751,7 +751,7 @@ xfs_dir2_block_lookup_int(
 	 * No match, release the buffer and return ENOENT.
 	 */
 	xfs_trans_brelse(tp, bp);
-	return XFS_ERROR(ENOENT);
+	return -ENOENT;
 }
 
 /*
@@ -790,18 +790,19 @@ xfs_dir2_block_removename(
 	tp = args->trans;
 	mp = dp->i_mount;
 	hdr = bp->b_addr;
-	btp = xfs_dir2_block_tail_p(mp, hdr);
+	btp = xfs_dir2_block_tail_p(args->geo, hdr);
 	blp = xfs_dir2_block_leaf_p(btp);
 	/*
 	 * Point to the data entry using the leaf entry.
 	 */
-	dep = (xfs_dir2_data_entry_t *)
-	      ((char *)hdr + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(blp[ent].address)));
+	dep = (xfs_dir2_data_entry_t *)((char *)hdr +
+			xfs_dir2_dataptr_to_off(args->geo,
+						be32_to_cpu(blp[ent].address)));
 	/*
 	 * Mark the data entry's space free.
 	 */
 	needlog = needscan = 0;
-	xfs_dir2_data_make_free(tp, dp, bp,
+	xfs_dir2_data_make_free(args, bp,
 		(xfs_dir2_data_aoff_t)((char *)dep - (char *)hdr),
 		dp->d_ops->data_entsize(dep->namelen), &needlog, &needscan);
 	/*
@@ -820,7 +821,7 @@ xfs_dir2_block_removename(
 	if (needscan)
 		xfs_dir2_data_freescan(dp, hdr, &needlog);
 	if (needlog)
-		xfs_dir2_data_log_header(tp, dp, bp);
+		xfs_dir2_data_log_header(args, bp);
 	xfs_dir3_data_check(dp, bp);
 	/*
 	 * See if the size as a shortform is good enough.
@@ -865,20 +866,21 @@ xfs_dir2_block_replace(
 	dp = args->dp;
 	mp = dp->i_mount;
 	hdr = bp->b_addr;
-	btp = xfs_dir2_block_tail_p(mp, hdr);
+	btp = xfs_dir2_block_tail_p(args->geo, hdr);
 	blp = xfs_dir2_block_leaf_p(btp);
 	/*
 	 * Point to the data entry we need to change.
 	 */
-	dep = (xfs_dir2_data_entry_t *)
-	      ((char *)hdr + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(blp[ent].address)));
+	dep = (xfs_dir2_data_entry_t *)((char *)hdr +
+			xfs_dir2_dataptr_to_off(args->geo,
+						be32_to_cpu(blp[ent].address)));
 	ASSERT(be64_to_cpu(dep->inumber) != args->inumber);
 	/*
 	 * Change the inode number to the new value.
 	 */
 	dep->inumber = cpu_to_be64(args->inumber);
 	dp->d_ops->data_put_ftype(dep, args->filetype);
-	xfs_dir2_data_log_entry(args->trans, dp, bp, dep);
+	xfs_dir2_data_log_entry(args, bp, dep);
 	xfs_dir3_data_check(dp, bp);
 	return 0;
 }
@@ -938,7 +940,7 @@ xfs_dir2_leaf_to_block(
 	leaf = lbp->b_addr;
 	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
 	ents = dp->d_ops->leaf_ents_p(leaf);
-	ltp = xfs_dir2_leaf_tail_p(mp, leaf);
+	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
 
 	ASSERT(leafhdr.magic == XFS_DIR2_LEAF1_MAGIC ||
 	       leafhdr.magic == XFS_DIR3_LEAF1_MAGIC);
@@ -948,13 +950,13 @@ xfs_dir2_leaf_to_block(
 	 * been left behind during no-space-reservation operations.
 	 * These will show up in the leaf bests table.
 	 */
-	while (dp->i_d.di_size > mp->m_dirblksize) {
+	while (dp->i_d.di_size > args->geo->blksize) {
 		int hdrsz;
 
 		hdrsz = dp->d_ops->data_entry_offset;
 		bestsp = xfs_dir2_leaf_bests_p(ltp);
 		if (be16_to_cpu(bestsp[be32_to_cpu(ltp->bestcount) - 1]) ==
-					    mp->m_dirblksize - hdrsz) {
+					    args->geo->blksize - hdrsz) {
 			if ((error =
 			    xfs_dir2_leaf_trim_data(args, lbp,
 				    (xfs_dir2_db_t)(be32_to_cpu(ltp->bestcount) - 1))))
@@ -966,7 +968,7 @@ xfs_dir2_leaf_to_block(
 	 * Read the data block if we don't already have it, give up if it fails.
 	 */
 	if (!dbp) {
-		error = xfs_dir3_data_read(tp, dp, mp->m_dirdatablk, -1, &dbp);
+		error = xfs_dir3_data_read(tp, dp, args->geo->datablk, -1, &dbp);
 		if (error)
 			return error;
 	}
@@ -982,7 +984,7 @@ xfs_dir2_leaf_to_block(
 	/*
 	 * Look at the last data entry.
 	 */
-	tagp = (__be16 *)((char *)hdr + mp->m_dirblksize) - 1;
+	tagp = (__be16 *)((char *)hdr + args->geo->blksize) - 1;
 	dup = (xfs_dir2_data_unused_t *)((char *)hdr + be16_to_cpu(*tagp));
 	/*
 	 * If it's not free or is too short we can't do it.
@@ -1001,12 +1003,12 @@ xfs_dir2_leaf_to_block(
 	/*
 	 * Use up the space at the end of the block (blp/btp).
 	 */
-	xfs_dir2_data_use_free(tp, dp, dbp, dup, mp->m_dirblksize - size, size,
+	xfs_dir2_data_use_free(args, dbp, dup, args->geo->blksize - size, size,
 		&needlog, &needscan);
 	/*
 	 * Initialize the block tail.
 	 */
-	btp = xfs_dir2_block_tail_p(mp, hdr);
+	btp = xfs_dir2_block_tail_p(args->geo, hdr);
 	btp->count = cpu_to_be32(leafhdr.count - leafhdr.stale);
 	btp->stale = 0;
 	xfs_dir2_block_log_tail(tp, dbp);
@@ -1027,11 +1029,11 @@ xfs_dir2_leaf_to_block(
 	if (needscan)
 		xfs_dir2_data_freescan(dp, hdr, &needlog);
 	if (needlog)
-		xfs_dir2_data_log_header(tp, dp, dbp);
+		xfs_dir2_data_log_header(args, dbp);
 	/*
 	 * Pitch the old leaf block.
 	 */
-	error = xfs_da_shrink_inode(args, mp->m_dirleafblk, lbp);
+	error = xfs_da_shrink_inode(args, args->geo->leafblk, lbp);
 	if (error)
 		return error;
 
@@ -1089,7 +1091,7 @@ xfs_dir2_sf_to_block(
 	 */
 	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
 		ASSERT(XFS_FORCED_SHUTDOWN(mp));
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	oldsfp = (xfs_dir2_sf_hdr_t *)ifp->if_u1.if_data;
@@ -1140,13 +1142,13 @@ xfs_dir2_sf_to_block(
 	 */
 	dup = dp->d_ops->data_unused_p(hdr);
 	needlog = needscan = 0;
-	xfs_dir2_data_use_free(tp, dp, bp, dup, mp->m_dirblksize - i, i, &needlog,
-		&needscan);
+	xfs_dir2_data_use_free(args, bp, dup, args->geo->blksize - i,
+			       i, &needlog, &needscan);
 	ASSERT(needscan == 0);
 	/*
 	 * Fill in the tail.
 	 */
-	btp = xfs_dir2_block_tail_p(mp, hdr);
+	btp = xfs_dir2_block_tail_p(args->geo, hdr);
 	btp->count = cpu_to_be32(sfp->count + 2);	/* ., .. */
 	btp->stale = 0;
 	blp = xfs_dir2_block_leaf_p(btp);
@@ -1154,7 +1156,7 @@ xfs_dir2_sf_to_block(
 	/*
 	 * Remove the freespace, we'll manage it.
 	 */
-	xfs_dir2_data_use_free(tp, dp, bp, dup,
+	xfs_dir2_data_use_free(args, bp, dup,
 		(xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr),
 		be16_to_cpu(dup->length), &needlog, &needscan);
 	/*
@@ -1167,9 +1169,9 @@ xfs_dir2_sf_to_block(
 	dp->d_ops->data_put_ftype(dep, XFS_DIR3_FT_DIR);
 	tagp = dp->d_ops->data_entry_tag_p(dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
-	xfs_dir2_data_log_entry(tp, dp, bp, dep);
+	xfs_dir2_data_log_entry(args, bp, dep);
 	blp[0].hashval = cpu_to_be32(xfs_dir_hash_dot);
-	blp[0].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp,
+	blp[0].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(
 				(char *)dep - (char *)hdr));
 	/*
 	 * Create entry for ..
@@ -1181,9 +1183,9 @@ xfs_dir2_sf_to_block(
 	dp->d_ops->data_put_ftype(dep, XFS_DIR3_FT_DIR);
 	tagp = dp->d_ops->data_entry_tag_p(dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
-	xfs_dir2_data_log_entry(tp, dp, bp, dep);
+	xfs_dir2_data_log_entry(args, bp, dep);
 	blp[1].hashval = cpu_to_be32(xfs_dir_hash_dotdot);
-	blp[1].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp,
+	blp[1].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(
 				(char *)dep - (char *)hdr));
 	offset = dp->d_ops->data_first_offset;
 	/*
@@ -1215,7 +1217,7 @@ xfs_dir2_sf_to_block(
 			dup->length = cpu_to_be16(newoffset - offset);
 			*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16(
 				((char *)dup - (char *)hdr));
-			xfs_dir2_data_log_unused(tp, bp, dup);
+			xfs_dir2_data_log_unused(args, bp, dup);
 			xfs_dir2_data_freeinsert(hdr,
 						 dp->d_ops->data_bestfree_p(hdr),
 						 dup, &dummy);
@@ -1232,12 +1234,12 @@ xfs_dir2_sf_to_block(
 		memcpy(dep->name, sfep->name, dep->namelen);
 		tagp = dp->d_ops->data_entry_tag_p(dep);
 		*tagp = cpu_to_be16((char *)dep - (char *)hdr);
-		xfs_dir2_data_log_entry(tp, dp, bp, dep);
+		xfs_dir2_data_log_entry(args, bp, dep);
 		name.name = sfep->name;
 		name.len = sfep->namelen;
 		blp[2 + i].hashval = cpu_to_be32(mp->m_dirnameops->
 							hashname(&name));
-		blp[2 + i].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp,
+		blp[2 + i].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(
 						 (char *)dep - (char *)hdr));
 		offset = (int)((char *)(tagp + 1) - (char *)hdr);
 		if (++i == sfp->count)
diff --git a/fs/xfs/xfs_dir2_data.c b/fs/xfs/libxfs/xfs_dir2_data.c
index 70acff4ee173..fdd803fecb8e 100644
--- a/fs/xfs/xfs_dir2_data.c
+++ b/fs/xfs/libxfs/xfs_dir2_data.c
@@ -63,8 +63,10 @@ __xfs_dir3_data_check(
 	int			stale;		/* count of stale leaves */
 	struct xfs_name		name;
 	const struct xfs_dir_ops *ops;
+	struct xfs_da_geometry	*geo;
 
 	mp = bp->b_target->bt_mount;
+	geo = mp->m_dir_geo;
 
 	/*
 	 * We can be passed a null dp here from a verifier, so we need to go the
@@ -78,7 +80,7 @@ __xfs_dir3_data_check(
 	switch (hdr->magic) {
 	case cpu_to_be32(XFS_DIR3_BLOCK_MAGIC):
 	case cpu_to_be32(XFS_DIR2_BLOCK_MAGIC):
-		btp = xfs_dir2_block_tail_p(mp, hdr);
+		btp = xfs_dir2_block_tail_p(geo, hdr);
 		lep = xfs_dir2_block_leaf_p(btp);
 		endp = (char *)lep;
 
@@ -94,11 +96,11 @@ __xfs_dir3_data_check(
 		break;
 	case cpu_to_be32(XFS_DIR3_DATA_MAGIC):
 	case cpu_to_be32(XFS_DIR2_DATA_MAGIC):
-		endp = (char *)hdr + mp->m_dirblksize;
+		endp = (char *)hdr + geo->blksize;
 		break;
 	default:
 		XFS_ERROR_REPORT("Bad Magic", XFS_ERRLEVEL_LOW, mp);
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 	}
 
 	/*
@@ -172,9 +174,9 @@ __xfs_dir3_data_check(
 		lastfree = 0;
 		if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
 		    hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC)) {
-			addr = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk,
-				(xfs_dir2_data_aoff_t)
-				((char *)dep - (char *)hdr));
+			addr = xfs_dir2_db_off_to_dataptr(geo, geo->datablk,
+						(xfs_dir2_data_aoff_t)
+						((char *)dep - (char *)hdr));
 			name.name = dep->name;
 			name.len = dep->namelen;
 			hash = mp->m_dirnameops->hashname(&name);
@@ -241,7 +243,6 @@ static void
 xfs_dir3_data_reada_verify(
 	struct xfs_buf		*bp)
 {
-	struct xfs_mount	*mp = bp->b_target->bt_mount;
 	struct xfs_dir2_data_hdr *hdr = bp->b_addr;
 
 	switch (hdr->magic) {
@@ -255,8 +256,8 @@ xfs_dir3_data_reada_verify(
 		xfs_dir3_data_verify(bp);
 		return;
 	default:
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, hdr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
 		break;
 	}
 }
@@ -267,13 +268,14 @@ xfs_dir3_data_read_verify(
 {
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 
-	if ((xfs_sb_version_hascrc(&mp->m_sb) &&
-	     !xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
-					  XFS_DIR3_DATA_CRC_OFF)) ||
-	    !xfs_dir3_data_verify(bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
-	}
+	if (xfs_sb_version_hascrc(&mp->m_sb) &&
+	     !xfs_buf_verify_cksum(bp, XFS_DIR3_DATA_CRC_OFF))
+		 xfs_buf_ioerror(bp, -EFSBADCRC);
+	else if (!xfs_dir3_data_verify(bp))
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+
+	if (bp->b_error)
+		xfs_verifier_error(bp);
 }
 
 static void
@@ -285,8 +287,8 @@ xfs_dir3_data_write_verify(
 	struct xfs_dir3_blk_hdr	*hdr3 = bp->b_addr;
 
 	if (!xfs_dir3_data_verify(bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
 		return;
 	}
 
@@ -296,7 +298,7 @@ xfs_dir3_data_write_verify(
 	if (bip)
 		hdr3->lsn = cpu_to_be64(bip->bli_item.li_lsn);
 
-	xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), XFS_DIR3_DATA_CRC_OFF);
+	xfs_buf_update_cksum(bp, XFS_DIR3_DATA_CRC_OFF);
 }
 
 const struct xfs_buf_ops xfs_dir3_data_buf_ops = {
@@ -329,12 +331,11 @@ xfs_dir3_data_read(
 
 int
 xfs_dir3_data_readahead(
-	struct xfs_trans	*tp,
 	struct xfs_inode	*dp,
 	xfs_dablk_t		bno,
 	xfs_daddr_t		mapped_bno)
 {
-	return xfs_da_reada_buf(tp, dp, bno, mapped_bno,
+	return xfs_da_reada_buf(dp, bno, mapped_bno,
 				XFS_DATA_FORK, &xfs_dir3_data_reada_buf_ops);
 }
 
@@ -510,6 +511,7 @@ xfs_dir2_data_freescan(
 	struct xfs_dir2_data_free *bf;
 	char			*endp;		/* end of block's data */
 	char			*p;		/* current entry pointer */
+	struct xfs_da_geometry	*geo = dp->i_mount->m_dir_geo;
 
 	ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC) ||
@@ -528,10 +530,10 @@ xfs_dir2_data_freescan(
 	p = (char *)dp->d_ops->data_entry_p(hdr);
 	if (hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
 	    hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC)) {
-		btp = xfs_dir2_block_tail_p(dp->i_mount, hdr);
+		btp = xfs_dir2_block_tail_p(geo, hdr);
 		endp = (char *)xfs_dir2_block_leaf_p(btp);
 	} else
-		endp = (char *)hdr + dp->i_mount->m_dirblksize;
+		endp = (char *)hdr + geo->blksize;
 	/*
 	 * Loop over the block's entries.
 	 */
@@ -585,8 +587,8 @@ xfs_dir3_data_init(
 	/*
 	 * Get the buffer set up for the block.
 	 */
-	error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, blkno), -1, &bp,
-		XFS_DATA_FORK);
+	error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(args->geo, blkno),
+			       -1, &bp, XFS_DATA_FORK);
 	if (error)
 		return error;
 	bp->b_ops = &xfs_dir3_data_buf_ops;
@@ -621,15 +623,15 @@ xfs_dir3_data_init(
 	dup = dp->d_ops->data_unused_p(hdr);
 	dup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
 
-	t = mp->m_dirblksize - (uint)dp->d_ops->data_entry_offset;
+	t = args->geo->blksize - (uint)dp->d_ops->data_entry_offset;
 	bf[0].length = cpu_to_be16(t);
 	dup->length = cpu_to_be16(t);
 	*xfs_dir2_data_unused_tag_p(dup) = cpu_to_be16((char *)dup - (char *)hdr);
 	/*
 	 * Log it and return it.
 	 */
-	xfs_dir2_data_log_header(tp, dp, bp);
-	xfs_dir2_data_log_unused(tp, bp, dup);
+	xfs_dir2_data_log_header(args, bp);
+	xfs_dir2_data_log_unused(args, bp, dup);
 	*bpp = bp;
 	return 0;
 }
@@ -639,8 +641,7 @@ xfs_dir3_data_init(
  */
 void
 xfs_dir2_data_log_entry(
-	struct xfs_trans	*tp,
-	struct xfs_inode	*dp,
+	struct xfs_da_args	*args,
 	struct xfs_buf		*bp,
 	xfs_dir2_data_entry_t	*dep)		/* data entry pointer */
 {
@@ -651,8 +652,8 @@ xfs_dir2_data_log_entry(
 	       hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
 
-	xfs_trans_log_buf(tp, bp, (uint)((char *)dep - (char *)hdr),
-		(uint)((char *)(dp->d_ops->data_entry_tag_p(dep) + 1) -
+	xfs_trans_log_buf(args->trans, bp, (uint)((char *)dep - (char *)hdr),
+		(uint)((char *)(args->dp->d_ops->data_entry_tag_p(dep) + 1) -
 		       (char *)hdr - 1));
 }
 
@@ -661,8 +662,7 @@ xfs_dir2_data_log_entry(
  */
 void
 xfs_dir2_data_log_header(
-	struct xfs_trans	*tp,
-	struct xfs_inode	*dp,
+	struct xfs_da_args	*args,
 	struct xfs_buf		*bp)
 {
 #ifdef DEBUG
@@ -674,7 +674,8 @@ xfs_dir2_data_log_header(
 	       hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
 #endif
 
-	xfs_trans_log_buf(tp, bp, 0, dp->d_ops->data_entry_offset - 1);
+	xfs_trans_log_buf(args->trans, bp, 0,
+			  args->dp->d_ops->data_entry_offset - 1);
 }
 
 /*
@@ -682,7 +683,7 @@ xfs_dir2_data_log_header(
  */
 void
 xfs_dir2_data_log_unused(
-	struct xfs_trans	*tp,
+	struct xfs_da_args	*args,
 	struct xfs_buf		*bp,
 	xfs_dir2_data_unused_t	*dup)		/* data unused pointer */
 {
@@ -696,13 +697,13 @@ xfs_dir2_data_log_unused(
 	/*
 	 * Log the first part of the unused entry.
 	 */
-	xfs_trans_log_buf(tp, bp, (uint)((char *)dup - (char *)hdr),
+	xfs_trans_log_buf(args->trans, bp, (uint)((char *)dup - (char *)hdr),
 		(uint)((char *)&dup->length + sizeof(dup->length) -
 		       1 - (char *)hdr));
 	/*
 	 * Log the end (tag) of the unused entry.
 	 */
-	xfs_trans_log_buf(tp, bp,
+	xfs_trans_log_buf(args->trans, bp,
 		(uint)((char *)xfs_dir2_data_unused_tag_p(dup) - (char *)hdr),
 		(uint)((char *)xfs_dir2_data_unused_tag_p(dup) - (char *)hdr +
 		       sizeof(xfs_dir2_data_off_t) - 1));
@@ -714,8 +715,7 @@ xfs_dir2_data_log_unused(
  */
 void
 xfs_dir2_data_make_free(
-	struct xfs_trans	*tp,
-	struct xfs_inode	*dp,
+	struct xfs_da_args	*args,
 	struct xfs_buf		*bp,
 	xfs_dir2_data_aoff_t	offset,		/* starting byte offset */
 	xfs_dir2_data_aoff_t	len,		/* length in bytes */
@@ -725,14 +725,12 @@ xfs_dir2_data_make_free(
 	xfs_dir2_data_hdr_t	*hdr;		/* data block pointer */
 	xfs_dir2_data_free_t	*dfp;		/* bestfree pointer */
 	char			*endptr;	/* end of data area */
-	xfs_mount_t		*mp;		/* filesystem mount point */
 	int			needscan;	/* need to regen bestfree */
 	xfs_dir2_data_unused_t	*newdup;	/* new unused entry */
 	xfs_dir2_data_unused_t	*postdup;	/* unused entry after us */
 	xfs_dir2_data_unused_t	*prevdup;	/* unused entry before us */
 	struct xfs_dir2_data_free *bf;
 
-	mp = tp->t_mountp;
 	hdr = bp->b_addr;
 
 	/*
@@ -740,20 +738,20 @@ xfs_dir2_data_make_free(
 	 */
 	if (hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
 	    hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC))
-		endptr = (char *)hdr + mp->m_dirblksize;
+		endptr = (char *)hdr + args->geo->blksize;
 	else {
 		xfs_dir2_block_tail_t	*btp;	/* block tail */
 
 		ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_BLOCK_MAGIC) ||
 			hdr->magic == cpu_to_be32(XFS_DIR3_BLOCK_MAGIC));
-		btp = xfs_dir2_block_tail_p(mp, hdr);
+		btp = xfs_dir2_block_tail_p(args->geo, hdr);
 		endptr = (char *)xfs_dir2_block_leaf_p(btp);
 	}
 	/*
 	 * If this isn't the start of the block, then back up to
 	 * the previous entry and see if it's free.
 	 */
-	if (offset > dp->d_ops->data_entry_offset) {
+	if (offset > args->dp->d_ops->data_entry_offset) {
 		__be16			*tagp;	/* tag just before us */
 
 		tagp = (__be16 *)((char *)hdr + offset) - 1;
@@ -779,7 +777,7 @@ xfs_dir2_data_make_free(
 	 * Previous and following entries are both free,
 	 * merge everything into a single free entry.
 	 */
-	bf = dp->d_ops->data_bestfree_p(hdr);
+	bf = args->dp->d_ops->data_bestfree_p(hdr);
 	if (prevdup && postdup) {
 		xfs_dir2_data_free_t	*dfp2;	/* another bestfree pointer */
 
@@ -801,7 +799,7 @@ xfs_dir2_data_make_free(
 		be16_add_cpu(&prevdup->length, len + be16_to_cpu(postdup->length));
 		*xfs_dir2_data_unused_tag_p(prevdup) =
 			cpu_to_be16((char *)prevdup - (char *)hdr);
-		xfs_dir2_data_log_unused(tp, bp, prevdup);
+		xfs_dir2_data_log_unused(args, bp, prevdup);
 		if (!needscan) {
 			/*
 			 * Has to be the case that entries 0 and 1 are
@@ -836,7 +834,7 @@ xfs_dir2_data_make_free(
 		be16_add_cpu(&prevdup->length, len);
 		*xfs_dir2_data_unused_tag_p(prevdup) =
 			cpu_to_be16((char *)prevdup - (char *)hdr);
-		xfs_dir2_data_log_unused(tp, bp, prevdup);
+		xfs_dir2_data_log_unused(args, bp, prevdup);
 		/*
 		 * If the previous entry was in the table, the new entry
 		 * is longer, so it will be in the table too.  Remove
@@ -864,7 +862,7 @@ xfs_dir2_data_make_free(
 		newdup->length = cpu_to_be16(len + be16_to_cpu(postdup->length));
 		*xfs_dir2_data_unused_tag_p(newdup) =
 			cpu_to_be16((char *)newdup - (char *)hdr);
-		xfs_dir2_data_log_unused(tp, bp, newdup);
+		xfs_dir2_data_log_unused(args, bp, newdup);
 		/*
 		 * If the following entry was in the table, the new entry
 		 * is longer, so it will be in the table too.  Remove
@@ -891,7 +889,7 @@ xfs_dir2_data_make_free(
 		newdup->length = cpu_to_be16(len);
 		*xfs_dir2_data_unused_tag_p(newdup) =
 			cpu_to_be16((char *)newdup - (char *)hdr);
-		xfs_dir2_data_log_unused(tp, bp, newdup);
+		xfs_dir2_data_log_unused(args, bp, newdup);
 		xfs_dir2_data_freeinsert(hdr, bf, newdup, needlogp);
 	}
 	*needscanp = needscan;
@@ -902,8 +900,7 @@ xfs_dir2_data_make_free(
  */
 void
 xfs_dir2_data_use_free(
-	struct xfs_trans	*tp,
-	struct xfs_inode	*dp,
+	struct xfs_da_args	*args,
 	struct xfs_buf		*bp,
 	xfs_dir2_data_unused_t	*dup,		/* unused entry */
 	xfs_dir2_data_aoff_t	offset,		/* starting offset to use */
@@ -934,7 +931,7 @@ xfs_dir2_data_use_free(
 	 * Look up the entry in the bestfree table.
 	 */
 	oldlen = be16_to_cpu(dup->length);
-	bf = dp->d_ops->data_bestfree_p(hdr);
+	bf = args->dp->d_ops->data_bestfree_p(hdr);
 	dfp = xfs_dir2_data_freefind(hdr, bf, dup);
 	ASSERT(dfp || oldlen <= be16_to_cpu(bf[2].length));
 	/*
@@ -966,7 +963,7 @@ xfs_dir2_data_use_free(
 		newdup->length = cpu_to_be16(oldlen - len);
 		*xfs_dir2_data_unused_tag_p(newdup) =
 			cpu_to_be16((char *)newdup - (char *)hdr);
-		xfs_dir2_data_log_unused(tp, bp, newdup);
+		xfs_dir2_data_log_unused(args, bp, newdup);
 		/*
 		 * If it was in the table, remove it and add the new one.
 		 */
@@ -994,7 +991,7 @@ xfs_dir2_data_use_free(
 		newdup->length = cpu_to_be16(((char *)hdr + offset) - (char *)newdup);
 		*xfs_dir2_data_unused_tag_p(newdup) =
 			cpu_to_be16((char *)newdup - (char *)hdr);
-		xfs_dir2_data_log_unused(tp, bp, newdup);
+		xfs_dir2_data_log_unused(args, bp, newdup);
 		/*
 		 * If it was in the table, remove it and add the new one.
 		 */
@@ -1022,13 +1019,13 @@ xfs_dir2_data_use_free(
 		newdup->length = cpu_to_be16(((char *)hdr + offset) - (char *)newdup);
 		*xfs_dir2_data_unused_tag_p(newdup) =
 			cpu_to_be16((char *)newdup - (char *)hdr);
-		xfs_dir2_data_log_unused(tp, bp, newdup);
+		xfs_dir2_data_log_unused(args, bp, newdup);
 		newdup2 = (xfs_dir2_data_unused_t *)((char *)hdr + offset + len);
 		newdup2->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
 		newdup2->length = cpu_to_be16(oldlen - len - be16_to_cpu(newdup->length));
 		*xfs_dir2_data_unused_tag_p(newdup2) =
 			cpu_to_be16((char *)newdup2 - (char *)hdr);
-		xfs_dir2_data_log_unused(tp, bp, newdup2);
+		xfs_dir2_data_log_unused(args, bp, newdup2);
 		/*
 		 * If the old entry was in the table, we need to scan
 		 * if the 3rd entry was valid, since these entries
diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/libxfs/xfs_dir2_leaf.c
index ae47ec6e16c4..a19174eb3cb2 100644
--- a/fs/xfs/xfs_dir2_leaf.c
+++ b/fs/xfs/libxfs/xfs_dir2_leaf.c
@@ -41,9 +41,10 @@
  */
 static int xfs_dir2_leaf_lookup_int(xfs_da_args_t *args, struct xfs_buf **lbpp,
 				    int *indexp, struct xfs_buf **dbpp);
-static void xfs_dir3_leaf_log_bests(struct xfs_trans *tp, struct xfs_buf *bp,
-				    int first, int last);
-static void xfs_dir3_leaf_log_tail(struct xfs_trans *tp, struct xfs_buf *bp);
+static void xfs_dir3_leaf_log_bests(struct xfs_da_args *args,
+				    struct xfs_buf *bp, int first, int last);
+static void xfs_dir3_leaf_log_tail(struct xfs_da_args *args,
+				   struct xfs_buf *bp);
 
 /*
  * Check the internal consistency of a leaf1 block.
@@ -92,6 +93,7 @@ xfs_dir3_leaf_check_int(
 	int			i;
 	const struct xfs_dir_ops *ops;
 	struct xfs_dir3_icleaf_hdr leafhdr;
+	struct xfs_da_geometry	*geo = mp->m_dir_geo;
 
 	/*
 	 * we can be passed a null dp here from a verifier, so we need to go the
@@ -105,14 +107,14 @@ xfs_dir3_leaf_check_int(
 	}
 
 	ents = ops->leaf_ents_p(leaf);
-	ltp = xfs_dir2_leaf_tail_p(mp, leaf);
+	ltp = xfs_dir2_leaf_tail_p(geo, leaf);
 
 	/*
 	 * XXX (dgc): This value is not restrictive enough.
 	 * Should factor in the size of the bests table as well.
 	 * We can deduce a value for that from di_size.
 	 */
-	if (hdr->count > ops->leaf_max_ents(mp))
+	if (hdr->count > ops->leaf_max_ents(geo))
 		return false;
 
 	/* Leaves and bests don't overlap in leaf format. */
@@ -179,13 +181,14 @@ __read_verify(
 {
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 
-	if ((xfs_sb_version_hascrc(&mp->m_sb) &&
-	     !xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
-					  XFS_DIR3_LEAF_CRC_OFF)) ||
-	    !xfs_dir3_leaf_verify(bp, magic)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
-	}
+	if (xfs_sb_version_hascrc(&mp->m_sb) &&
+	     !xfs_buf_verify_cksum(bp, XFS_DIR3_LEAF_CRC_OFF))
+		xfs_buf_ioerror(bp, -EFSBADCRC);
+	else if (!xfs_dir3_leaf_verify(bp, magic))
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+
+	if (bp->b_error)
+		xfs_verifier_error(bp);
 }
 
 static void
@@ -198,8 +201,8 @@ __write_verify(
 	struct xfs_dir3_leaf_hdr *hdr3 = bp->b_addr;
 
 	if (!xfs_dir3_leaf_verify(bp, magic)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
 		return;
 	}
 
@@ -209,7 +212,7 @@ __write_verify(
 	if (bip)
 		hdr3->info.lsn = cpu_to_be64(bip->bli_item.li_lsn);
 
-	xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), XFS_DIR3_LEAF_CRC_OFF);
+	xfs_buf_update_cksum(bp, XFS_DIR3_LEAF_CRC_OFF);
 }
 
 static void
@@ -322,7 +325,7 @@ xfs_dir3_leaf_init(
 	if (type == XFS_DIR2_LEAF1_MAGIC) {
 		struct xfs_dir2_leaf_tail *ltp;
 
-		ltp = xfs_dir2_leaf_tail_p(mp, leaf);
+		ltp = xfs_dir2_leaf_tail_p(mp->m_dir_geo, leaf);
 		ltp->bestcount = 0;
 		bp->b_ops = &xfs_dir3_leaf1_buf_ops;
 		xfs_trans_buf_set_type(tp, bp, XFS_BLFT_DIR_LEAF1_BUF);
@@ -346,18 +349,18 @@ xfs_dir3_leaf_get_buf(
 	int			error;
 
 	ASSERT(magic == XFS_DIR2_LEAF1_MAGIC || magic == XFS_DIR2_LEAFN_MAGIC);
-	ASSERT(bno >= XFS_DIR2_LEAF_FIRSTDB(mp) &&
-	       bno < XFS_DIR2_FREE_FIRSTDB(mp));
+	ASSERT(bno >= xfs_dir2_byte_to_db(args->geo, XFS_DIR2_LEAF_OFFSET) &&
+	       bno < xfs_dir2_byte_to_db(args->geo, XFS_DIR2_FREE_OFFSET));
 
-	error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, bno), -1, &bp,
-			       XFS_DATA_FORK);
+	error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(args->geo, bno),
+			       -1, &bp, XFS_DATA_FORK);
 	if (error)
 		return error;
 
 	xfs_dir3_leaf_init(mp, tp, bp, dp->i_ino, magic);
-	xfs_dir3_leaf_log_header(tp, dp, bp);
+	xfs_dir3_leaf_log_header(args, bp);
 	if (magic == XFS_DIR2_LEAF1_MAGIC)
-		xfs_dir3_leaf_log_tail(tp, bp);
+		xfs_dir3_leaf_log_tail(args, bp);
 	*bpp = bp;
 	return 0;
 }
@@ -402,8 +405,8 @@ xfs_dir2_block_to_leaf(
 	if ((error = xfs_da_grow_inode(args, &blkno))) {
 		return error;
 	}
-	ldb = xfs_dir2_da_to_db(mp, blkno);
-	ASSERT(ldb == XFS_DIR2_LEAF_FIRSTDB(mp));
+	ldb = xfs_dir2_da_to_db(args->geo, blkno);
+	ASSERT(ldb == xfs_dir2_byte_to_db(args->geo, XFS_DIR2_LEAF_OFFSET));
 	/*
 	 * Initialize the leaf block, get a buffer for it.
 	 */
@@ -414,7 +417,7 @@ xfs_dir2_block_to_leaf(
 	leaf = lbp->b_addr;
 	hdr = dbp->b_addr;
 	xfs_dir3_data_check(dp, dbp);
-	btp = xfs_dir2_block_tail_p(mp, hdr);
+	btp = xfs_dir2_block_tail_p(args->geo, hdr);
 	blp = xfs_dir2_block_leaf_p(btp);
 	bf = dp->d_ops->data_bestfree_p(hdr);
 	ents = dp->d_ops->leaf_ents_p(leaf);
@@ -426,23 +429,23 @@ xfs_dir2_block_to_leaf(
 	leafhdr.count = be32_to_cpu(btp->count);
 	leafhdr.stale = be32_to_cpu(btp->stale);
 	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
-	xfs_dir3_leaf_log_header(tp, dp, lbp);
+	xfs_dir3_leaf_log_header(args, lbp);
 
 	/*
 	 * Could compact these but I think we always do the conversion
 	 * after squeezing out stale entries.
 	 */
 	memcpy(ents, blp, be32_to_cpu(btp->count) * sizeof(xfs_dir2_leaf_entry_t));
-	xfs_dir3_leaf_log_ents(tp, dp, lbp, 0, leafhdr.count - 1);
+	xfs_dir3_leaf_log_ents(args, lbp, 0, leafhdr.count - 1);
 	needscan = 0;
 	needlog = 1;
 	/*
 	 * Make the space formerly occupied by the leaf entries and block
 	 * tail be free.
 	 */
-	xfs_dir2_data_make_free(tp, dp, dbp,
+	xfs_dir2_data_make_free(args, dbp,
 		(xfs_dir2_data_aoff_t)((char *)blp - (char *)hdr),
-		(xfs_dir2_data_aoff_t)((char *)hdr + mp->m_dirblksize -
+		(xfs_dir2_data_aoff_t)((char *)hdr + args->geo->blksize -
 				       (char *)blp),
 		&needlog, &needscan);
 	/*
@@ -460,7 +463,7 @@ xfs_dir2_block_to_leaf(
 	/*
 	 * Set up leaf tail and bests table.
 	 */
-	ltp = xfs_dir2_leaf_tail_p(mp, leaf);
+	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
 	ltp->bestcount = cpu_to_be32(1);
 	bestsp = xfs_dir2_leaf_bests_p(ltp);
 	bestsp[0] =  bf[0].length;
@@ -468,10 +471,10 @@ xfs_dir2_block_to_leaf(
 	 * Log the data header and leaf bests table.
 	 */
 	if (needlog)
-		xfs_dir2_data_log_header(tp, dp, dbp);
+		xfs_dir2_data_log_header(args, dbp);
 	xfs_dir3_leaf_check(dp, lbp);
 	xfs_dir3_data_check(dp, dbp);
-	xfs_dir3_leaf_log_bests(tp, lbp, 0, 0);
+	xfs_dir3_leaf_log_bests(args, lbp, 0, 0);
 	return 0;
 }
 
@@ -640,7 +643,7 @@ xfs_dir2_leaf_addname(
 	tp = args->trans;
 	mp = dp->i_mount;
 
-	error = xfs_dir3_leaf_read(tp, dp, mp->m_dirleafblk, -1, &lbp);
+	error = xfs_dir3_leaf_read(tp, dp, args->geo->leafblk, -1, &lbp);
 	if (error)
 		return error;
 
@@ -652,7 +655,7 @@ xfs_dir2_leaf_addname(
 	 */
 	index = xfs_dir2_leaf_search_hash(args, lbp);
 	leaf = lbp->b_addr;
-	ltp = xfs_dir2_leaf_tail_p(mp, leaf);
+	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
 	ents = dp->d_ops->leaf_ents_p(leaf);
 	dp->d_ops->leaf_hdr_from_disk(&leafhdr, leaf);
 	bestsp = xfs_dir2_leaf_bests_p(ltp);
@@ -669,7 +672,7 @@ xfs_dir2_leaf_addname(
 	     index++, lep++) {
 		if (be32_to_cpu(lep->address) == XFS_DIR2_NULL_DATAPTR)
 			continue;
-		i = xfs_dir2_dataptr_to_db(mp, be32_to_cpu(lep->address));
+		i = xfs_dir2_dataptr_to_db(args->geo, be32_to_cpu(lep->address));
 		ASSERT(i < be32_to_cpu(ltp->bestcount));
 		ASSERT(bestsp[i] != cpu_to_be16(NULLDATAOFF));
 		if (be16_to_cpu(bestsp[i]) >= length) {
@@ -728,7 +731,7 @@ xfs_dir2_leaf_addname(
 		if ((args->op_flags & XFS_DA_OP_JUSTCHECK) ||
 							args->total == 0) {
 			xfs_trans_brelse(tp, lbp);
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		}
 		/*
 		 * Convert to node form.
@@ -752,7 +755,7 @@ xfs_dir2_leaf_addname(
 	 */
 	if (args->op_flags & XFS_DA_OP_JUSTCHECK) {
 		xfs_trans_brelse(tp, lbp);
-		return use_block == -1 ? XFS_ERROR(ENOSPC) : 0;
+		return use_block == -1 ? -ENOSPC : 0;
 	}
 	/*
 	 * If no allocations are allowed, return now before we've
@@ -760,7 +763,7 @@ xfs_dir2_leaf_addname(
 	 */
 	if (args->total == 0 && use_block == -1) {
 		xfs_trans_brelse(tp, lbp);
-		return XFS_ERROR(ENOSPC);
+		return -ENOSPC;
 	}
 	/*
 	 * Need to compact the leaf entries, removing stale ones.
@@ -809,14 +812,15 @@ xfs_dir2_leaf_addname(
 			memmove(&bestsp[0], &bestsp[1],
 				be32_to_cpu(ltp->bestcount) * sizeof(bestsp[0]));
 			be32_add_cpu(&ltp->bestcount, 1);
-			xfs_dir3_leaf_log_tail(tp, lbp);
-			xfs_dir3_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
+			xfs_dir3_leaf_log_tail(args, lbp);
+			xfs_dir3_leaf_log_bests(args, lbp, 0,
+						be32_to_cpu(ltp->bestcount) - 1);
 		}
 		/*
 		 * If we're filling in a previously empty block just log it.
 		 */
 		else
-			xfs_dir3_leaf_log_bests(tp, lbp, use_block, use_block);
+			xfs_dir3_leaf_log_bests(args, lbp, use_block, use_block);
 		hdr = dbp->b_addr;
 		bf = dp->d_ops->data_bestfree_p(hdr);
 		bestsp[use_block] = bf[0].length;
@@ -827,8 +831,8 @@ xfs_dir2_leaf_addname(
 		 * Just read that one in.
 		 */
 		error = xfs_dir3_data_read(tp, dp,
-					   xfs_dir2_db_to_da(mp, use_block),
-					   -1, &dbp);
+				   xfs_dir2_db_to_da(args->geo, use_block),
+				   -1, &dbp);
 		if (error) {
 			xfs_trans_brelse(tp, lbp);
 			return error;
@@ -847,7 +851,7 @@ xfs_dir2_leaf_addname(
 	/*
 	 * Mark the initial part of our freespace in use for the new entry.
 	 */
-	xfs_dir2_data_use_free(tp, dp, dbp, dup,
+	xfs_dir2_data_use_free(args, dbp, dup,
 		(xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr), length,
 		&needlog, &needscan);
 	/*
@@ -869,8 +873,8 @@ xfs_dir2_leaf_addname(
 	 * Need to log the data block's header.
 	 */
 	if (needlog)
-		xfs_dir2_data_log_header(tp, dp, dbp);
-	xfs_dir2_data_log_entry(tp, dp, dbp, dep);
+		xfs_dir2_data_log_header(args, dbp);
+	xfs_dir2_data_log_entry(args, dbp, dep);
 	/*
 	 * If the bests table needs to be changed, do it.
 	 * Log the change unless we've already done that.
@@ -878,7 +882,7 @@ xfs_dir2_leaf_addname(
 	if (be16_to_cpu(bestsp[use_block]) != be16_to_cpu(bf[0].length)) {
 		bestsp[use_block] = bf[0].length;
 		if (!grown)
-			xfs_dir3_leaf_log_bests(tp, lbp, use_block, use_block);
+			xfs_dir3_leaf_log_bests(args, lbp, use_block, use_block);
 	}
 
 	lep = xfs_dir3_leaf_find_entry(&leafhdr, ents, index, compact, lowstale,
@@ -888,14 +892,15 @@ xfs_dir2_leaf_addname(
 	 * Fill in the new leaf entry.
 	 */
 	lep->hashval = cpu_to_be32(args->hashval);
-	lep->address = cpu_to_be32(xfs_dir2_db_off_to_dataptr(mp, use_block,
+	lep->address = cpu_to_be32(
+				xfs_dir2_db_off_to_dataptr(args->geo, use_block,
 				be16_to_cpu(*tagp)));
 	/*
 	 * Log the leaf fields and give up the buffers.
 	 */
 	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
-	xfs_dir3_leaf_log_header(tp, dp, lbp);
-	xfs_dir3_leaf_log_ents(tp, dp, lbp, lfloglow, lfloghigh);
+	xfs_dir3_leaf_log_header(args, lbp);
+	xfs_dir3_leaf_log_ents(args, lbp, lfloglow, lfloghigh);
 	xfs_dir3_leaf_check(dp, lbp);
 	xfs_dir3_data_check(dp, dbp);
 	return 0;
@@ -947,9 +952,9 @@ xfs_dir3_leaf_compact(
 	leafhdr->stale = 0;
 
 	dp->d_ops->leaf_hdr_to_disk(leaf, leafhdr);
-	xfs_dir3_leaf_log_header(args->trans, dp, bp);
+	xfs_dir3_leaf_log_header(args, bp);
 	if (loglow != -1)
-		xfs_dir3_leaf_log_ents(args->trans, dp, bp, loglow, to - 1);
+		xfs_dir3_leaf_log_ents(args, bp, loglow, to - 1);
 }
 
 /*
@@ -1051,7 +1056,7 @@ xfs_dir3_leaf_compact_x1(
  */
 static void
 xfs_dir3_leaf_log_bests(
-	xfs_trans_t		*tp,		/* transaction pointer */
+	struct xfs_da_args	*args,
 	struct xfs_buf		*bp,		/* leaf buffer */
 	int			first,		/* first entry to log */
 	int			last)		/* last entry to log */
@@ -1064,10 +1069,11 @@ xfs_dir3_leaf_log_bests(
 	ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC) ||
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAF1_MAGIC));
 
-	ltp = xfs_dir2_leaf_tail_p(tp->t_mountp, leaf);
+	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
 	firstb = xfs_dir2_leaf_bests_p(ltp) + first;
 	lastb = xfs_dir2_leaf_bests_p(ltp) + last;
-	xfs_trans_log_buf(tp, bp, (uint)((char *)firstb - (char *)leaf),
+	xfs_trans_log_buf(args->trans, bp,
+		(uint)((char *)firstb - (char *)leaf),
 		(uint)((char *)lastb - (char *)leaf + sizeof(*lastb) - 1));
 }
 
@@ -1076,8 +1082,7 @@ xfs_dir3_leaf_log_bests(
  */
 void
 xfs_dir3_leaf_log_ents(
-	struct xfs_trans	*tp,
-	struct xfs_inode	*dp,
+	struct xfs_da_args	*args,
 	struct xfs_buf		*bp,
 	int			first,
 	int			last)
@@ -1092,10 +1097,11 @@ xfs_dir3_leaf_log_ents(
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC));
 
-	ents = dp->d_ops->leaf_ents_p(leaf);
+	ents = args->dp->d_ops->leaf_ents_p(leaf);
 	firstlep = &ents[first];
 	lastlep = &ents[last];
-	xfs_trans_log_buf(tp, bp, (uint)((char *)firstlep - (char *)leaf),
+	xfs_trans_log_buf(args->trans, bp,
+		(uint)((char *)firstlep - (char *)leaf),
 		(uint)((char *)lastlep - (char *)leaf + sizeof(*lastlep) - 1));
 }
 
@@ -1104,8 +1110,7 @@ xfs_dir3_leaf_log_ents(
  */
 void
 xfs_dir3_leaf_log_header(
-	struct xfs_trans	*tp,
-	struct xfs_inode	*dp,
+	struct xfs_da_args	*args,
 	struct xfs_buf		*bp)
 {
 	struct xfs_dir2_leaf	*leaf = bp->b_addr;
@@ -1115,8 +1120,9 @@ xfs_dir3_leaf_log_header(
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC));
 
-	xfs_trans_log_buf(tp, bp, (uint)((char *)&leaf->hdr - (char *)leaf),
-			  dp->d_ops->leaf_hdr_size - 1);
+	xfs_trans_log_buf(args->trans, bp,
+			  (uint)((char *)&leaf->hdr - (char *)leaf),
+			  args->dp->d_ops->leaf_hdr_size - 1);
 }
 
 /*
@@ -1124,21 +1130,20 @@ xfs_dir3_leaf_log_header(
  */
 STATIC void
 xfs_dir3_leaf_log_tail(
-	struct xfs_trans	*tp,
+	struct xfs_da_args	*args,
 	struct xfs_buf		*bp)
 {
 	struct xfs_dir2_leaf	*leaf = bp->b_addr;
 	xfs_dir2_leaf_tail_t	*ltp;		/* leaf tail structure */
-	struct xfs_mount	*mp = tp->t_mountp;
 
 	ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAF1_MAGIC) ||
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAF1_MAGIC) ||
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR2_LEAFN_MAGIC) ||
 	       leaf->hdr.info.magic == cpu_to_be16(XFS_DIR3_LEAFN_MAGIC));
 
-	ltp = xfs_dir2_leaf_tail_p(mp, leaf);
-	xfs_trans_log_buf(tp, bp, (uint)((char *)ltp - (char *)leaf),
-		(uint)(mp->m_dirblksize - 1));
+	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
+	xfs_trans_log_buf(args->trans, bp, (uint)((char *)ltp - (char *)leaf),
+		(uint)(args->geo->blksize - 1));
 }
 
 /*
@@ -1184,7 +1189,7 @@ xfs_dir2_leaf_lookup(
 	 */
 	dep = (xfs_dir2_data_entry_t *)
 	      ((char *)dbp->b_addr +
-	       xfs_dir2_dataptr_to_off(dp->i_mount, be32_to_cpu(lep->address)));
+	       xfs_dir2_dataptr_to_off(args->geo, be32_to_cpu(lep->address)));
 	/*
 	 * Return the found inode number & CI name if appropriate
 	 */
@@ -1193,7 +1198,7 @@ xfs_dir2_leaf_lookup(
 	error = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
 	xfs_trans_brelse(tp, dbp);
 	xfs_trans_brelse(tp, lbp);
-	return XFS_ERROR(error);
+	return error;
 }
 
 /*
@@ -1230,7 +1235,7 @@ xfs_dir2_leaf_lookup_int(
 	tp = args->trans;
 	mp = dp->i_mount;
 
-	error = xfs_dir3_leaf_read(tp, dp, mp->m_dirleafblk, -1, &lbp);
+	error = xfs_dir3_leaf_read(tp, dp, args->geo->leafblk, -1, &lbp);
 	if (error)
 		return error;
 
@@ -1259,7 +1264,8 @@ xfs_dir2_leaf_lookup_int(
 		/*
 		 * Get the new data block number.
 		 */
-		newdb = xfs_dir2_dataptr_to_db(mp, be32_to_cpu(lep->address));
+		newdb = xfs_dir2_dataptr_to_db(args->geo,
+					       be32_to_cpu(lep->address));
 		/*
 		 * If it's not the same as the old data block number,
 		 * need to pitch the old one and read the new one.
@@ -1268,8 +1274,8 @@ xfs_dir2_leaf_lookup_int(
 			if (dbp)
 				xfs_trans_brelse(tp, dbp);
 			error = xfs_dir3_data_read(tp, dp,
-						   xfs_dir2_db_to_da(mp, newdb),
-						   -1, &dbp);
+					   xfs_dir2_db_to_da(args->geo, newdb),
+					   -1, &dbp);
 			if (error) {
 				xfs_trans_brelse(tp, lbp);
 				return error;
@@ -1280,7 +1286,8 @@ xfs_dir2_leaf_lookup_int(
 		 * Point to the data entry.
 		 */
 		dep = (xfs_dir2_data_entry_t *)((char *)dbp->b_addr +
-			xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)));
+			xfs_dir2_dataptr_to_off(args->geo,
+						be32_to_cpu(lep->address)));
 		/*
 		 * Compare name and if it's an exact match, return the index
 		 * and buffer. If it's the first case-insensitive match, store
@@ -1309,8 +1316,8 @@ xfs_dir2_leaf_lookup_int(
 		if (cidb != curdb) {
 			xfs_trans_brelse(tp, dbp);
 			error = xfs_dir3_data_read(tp, dp,
-						   xfs_dir2_db_to_da(mp, cidb),
-						   -1, &dbp);
+					   xfs_dir2_db_to_da(args->geo, cidb),
+					   -1, &dbp);
 			if (error) {
 				xfs_trans_brelse(tp, lbp);
 				return error;
@@ -1320,13 +1327,13 @@ xfs_dir2_leaf_lookup_int(
 		return 0;
 	}
 	/*
-	 * No match found, return ENOENT.
+	 * No match found, return -ENOENT.
 	 */
 	ASSERT(cidb == -1);
 	if (dbp)
 		xfs_trans_brelse(tp, dbp);
 	xfs_trans_brelse(tp, lbp);
-	return XFS_ERROR(ENOENT);
+	return -ENOENT;
 }
 
 /*
@@ -1379,18 +1386,18 @@ xfs_dir2_leaf_removename(
 	 * Point to the leaf entry, use that to point to the data entry.
 	 */
 	lep = &ents[index];
-	db = xfs_dir2_dataptr_to_db(mp, be32_to_cpu(lep->address));
-	dep = (xfs_dir2_data_entry_t *)
-	      ((char *)hdr + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)));
+	db = xfs_dir2_dataptr_to_db(args->geo, be32_to_cpu(lep->address));
+	dep = (xfs_dir2_data_entry_t *)((char *)hdr +
+		xfs_dir2_dataptr_to_off(args->geo, be32_to_cpu(lep->address)));
 	needscan = needlog = 0;
 	oldbest = be16_to_cpu(bf[0].length);
-	ltp = xfs_dir2_leaf_tail_p(mp, leaf);
+	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
 	bestsp = xfs_dir2_leaf_bests_p(ltp);
 	ASSERT(be16_to_cpu(bestsp[db]) == oldbest);
 	/*
 	 * Mark the former data entry unused.
 	 */
-	xfs_dir2_data_make_free(tp, dp, dbp,
+	xfs_dir2_data_make_free(args, dbp,
 		(xfs_dir2_data_aoff_t)((char *)dep - (char *)hdr),
 		dp->d_ops->data_entsize(dep->namelen), &needlog, &needscan);
 	/*
@@ -1398,10 +1405,10 @@ xfs_dir2_leaf_removename(
 	 */
 	leafhdr.stale++;
 	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
-	xfs_dir3_leaf_log_header(tp, dp, lbp);
+	xfs_dir3_leaf_log_header(args, lbp);
 
 	lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
-	xfs_dir3_leaf_log_ents(tp, dp, lbp, index, index);
+	xfs_dir3_leaf_log_ents(args, lbp, index, index);
 
 	/*
 	 * Scan the freespace in the data block again if necessary,
@@ -1410,22 +1417,22 @@ xfs_dir2_leaf_removename(
 	if (needscan)
 		xfs_dir2_data_freescan(dp, hdr, &needlog);
 	if (needlog)
-		xfs_dir2_data_log_header(tp, dp, dbp);
+		xfs_dir2_data_log_header(args, dbp);
 	/*
 	 * If the longest freespace in the data block has changed,
 	 * put the new value in the bests table and log that.
 	 */
 	if (be16_to_cpu(bf[0].length) != oldbest) {
 		bestsp[db] = bf[0].length;
-		xfs_dir3_leaf_log_bests(tp, lbp, db, db);
+		xfs_dir3_leaf_log_bests(args, lbp, db, db);
 	}
 	xfs_dir3_data_check(dp, dbp);
 	/*
 	 * If the data block is now empty then get rid of the data block.
 	 */
 	if (be16_to_cpu(bf[0].length) ==
-			mp->m_dirblksize - dp->d_ops->data_entry_offset) {
-		ASSERT(db != mp->m_dirdatablk);
+			args->geo->blksize - dp->d_ops->data_entry_offset) {
+		ASSERT(db != args->geo->datablk);
 		if ((error = xfs_dir2_shrink_inode(args, db, dbp))) {
 			/*
 			 * Nope, can't get rid of it because it caused
@@ -1433,7 +1440,7 @@ xfs_dir2_leaf_removename(
 			 * Just go on, returning success, leaving the
 			 * empty block in place.
 			 */
-			if (error == ENOSPC && args->total == 0)
+			if (error == -ENOSPC && args->total == 0)
 				error = 0;
 			xfs_dir3_leaf_check(dp, lbp);
 			return error;
@@ -1458,15 +1465,16 @@ xfs_dir2_leaf_removename(
 			memmove(&bestsp[db - i], bestsp,
 				(be32_to_cpu(ltp->bestcount) - (db - i)) * sizeof(*bestsp));
 			be32_add_cpu(&ltp->bestcount, -(db - i));
-			xfs_dir3_leaf_log_tail(tp, lbp);
-			xfs_dir3_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
+			xfs_dir3_leaf_log_tail(args, lbp);
+			xfs_dir3_leaf_log_bests(args, lbp, 0,
+						be32_to_cpu(ltp->bestcount) - 1);
 		} else
 			bestsp[db] = cpu_to_be16(NULLDATAOFF);
 	}
 	/*
 	 * If the data block was not the first one, drop it.
 	 */
-	else if (db != mp->m_dirdatablk)
+	else if (db != args->geo->datablk)
 		dbp = NULL;
 
 	xfs_dir3_leaf_check(dp, lbp);
@@ -1514,7 +1522,7 @@ xfs_dir2_leaf_replace(
 	 */
 	dep = (xfs_dir2_data_entry_t *)
 	      ((char *)dbp->b_addr +
-	       xfs_dir2_dataptr_to_off(dp->i_mount, be32_to_cpu(lep->address)));
+	       xfs_dir2_dataptr_to_off(args->geo, be32_to_cpu(lep->address)));
 	ASSERT(args->inumber != be64_to_cpu(dep->inumber));
 	/*
 	 * Put the new inode number in, log it.
@@ -1522,7 +1530,7 @@ xfs_dir2_leaf_replace(
 	dep->inumber = cpu_to_be64(args->inumber);
 	dp->d_ops->data_put_ftype(dep, args->filetype);
 	tp = args->trans;
-	xfs_dir2_data_log_entry(tp, dp, dbp, dep);
+	xfs_dir2_data_log_entry(args, dbp, dep);
 	xfs_dir3_leaf_check(dp, lbp);
 	xfs_trans_brelse(tp, lbp);
 	return 0;
@@ -1608,12 +1616,13 @@ xfs_dir2_leaf_trim_data(
 	/*
 	 * Read the offending data block.  We need its buffer.
 	 */
-	error = xfs_dir3_data_read(tp, dp, xfs_dir2_db_to_da(mp, db), -1, &dbp);
+	error = xfs_dir3_data_read(tp, dp, xfs_dir2_db_to_da(args->geo, db),
+				   -1, &dbp);
 	if (error)
 		return error;
 
 	leaf = lbp->b_addr;
-	ltp = xfs_dir2_leaf_tail_p(mp, leaf);
+	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
 
 #ifdef DEBUG
 {
@@ -1623,7 +1632,7 @@ xfs_dir2_leaf_trim_data(
 	ASSERT(hdr->magic == cpu_to_be32(XFS_DIR2_DATA_MAGIC) ||
 	       hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC));
 	ASSERT(be16_to_cpu(bf[0].length) ==
-	       mp->m_dirblksize - dp->d_ops->data_entry_offset);
+	       args->geo->blksize - dp->d_ops->data_entry_offset);
 	ASSERT(db == be32_to_cpu(ltp->bestcount) - 1);
 }
 #endif
@@ -1632,7 +1641,7 @@ xfs_dir2_leaf_trim_data(
 	 * Get rid of the data block.
 	 */
 	if ((error = xfs_dir2_shrink_inode(args, db, dbp))) {
-		ASSERT(error != ENOSPC);
+		ASSERT(error != -ENOSPC);
 		xfs_trans_brelse(tp, dbp);
 		return error;
 	}
@@ -1642,8 +1651,8 @@ xfs_dir2_leaf_trim_data(
 	bestsp = xfs_dir2_leaf_bests_p(ltp);
 	be32_add_cpu(&ltp->bestcount, -1);
 	memmove(&bestsp[1], &bestsp[0], be32_to_cpu(ltp->bestcount) * sizeof(*bestsp));
-	xfs_dir3_leaf_log_tail(tp, lbp);
-	xfs_dir3_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
+	xfs_dir3_leaf_log_tail(args, lbp);
+	xfs_dir3_leaf_log_bests(args, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
 	return 0;
 }
 
@@ -1707,22 +1716,22 @@ xfs_dir2_node_to_leaf(
 	/*
 	 * Get the last offset in the file.
 	 */
-	if ((error = xfs_bmap_last_offset(tp, dp, &fo, XFS_DATA_FORK))) {
+	if ((error = xfs_bmap_last_offset(dp, &fo, XFS_DATA_FORK))) {
 		return error;
 	}
-	fo -= mp->m_dirblkfsbs;
+	fo -= args->geo->fsbcount;
 	/*
 	 * If there are freespace blocks other than the first one,
 	 * take this opportunity to remove trailing empty freespace blocks
 	 * that may have been left behind during no-space-reservation
 	 * operations.
 	 */
-	while (fo > mp->m_dirfreeblk) {
+	while (fo > args->geo->freeblk) {
 		if ((error = xfs_dir2_node_trim_free(args, fo, &rval))) {
 			return error;
 		}
 		if (rval)
-			fo -= mp->m_dirblkfsbs;
+			fo -= args->geo->fsbcount;
 		else
 			return 0;
 	}
@@ -1735,7 +1744,7 @@ xfs_dir2_node_to_leaf(
 	/*
 	 * If it's not the single leaf block, give up.
 	 */
-	if (XFS_FSB_TO_B(mp, fo) > XFS_DIR2_LEAF_OFFSET + mp->m_dirblksize)
+	if (XFS_FSB_TO_B(mp, fo) > XFS_DIR2_LEAF_OFFSET + args->geo->blksize)
 		return 0;
 	lbp = state->path.blk[0].bp;
 	leaf = lbp->b_addr;
@@ -1747,7 +1756,7 @@ xfs_dir2_node_to_leaf(
 	/*
 	 * Read the freespace block.
 	 */
-	error = xfs_dir2_free_read(tp, dp,  mp->m_dirfreeblk, &fbp);
+	error = xfs_dir2_free_read(tp, dp,  args->geo->freeblk, &fbp);
 	if (error)
 		return error;
 	free = fbp->b_addr;
@@ -1759,7 +1768,7 @@ xfs_dir2_node_to_leaf(
 	 * Now see if the leafn and free data will fit in a leaf1.
 	 * If not, release the buffer and give up.
 	 */
-	if (xfs_dir3_leaf_size(&leafhdr, freehdr.nvalid) > mp->m_dirblksize) {
+	if (xfs_dir3_leaf_size(&leafhdr, freehdr.nvalid) > args->geo->blksize) {
 		xfs_trans_brelse(tp, fbp);
 		return 0;
 	}
@@ -1779,7 +1788,7 @@ xfs_dir2_node_to_leaf(
 	/*
 	 * Set up the leaf tail from the freespace block.
 	 */
-	ltp = xfs_dir2_leaf_tail_p(mp, leaf);
+	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
 	ltp->bestcount = cpu_to_be32(freehdr.nvalid);
 
 	/*
@@ -1789,22 +1798,24 @@ xfs_dir2_node_to_leaf(
 		freehdr.nvalid * sizeof(xfs_dir2_data_off_t));
 
 	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
-	xfs_dir3_leaf_log_header(tp, dp, lbp);
-	xfs_dir3_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
-	xfs_dir3_leaf_log_tail(tp, lbp);
+	xfs_dir3_leaf_log_header(args, lbp);
+	xfs_dir3_leaf_log_bests(args, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
+	xfs_dir3_leaf_log_tail(args, lbp);
 	xfs_dir3_leaf_check(dp, lbp);
 
 	/*
 	 * Get rid of the freespace block.
 	 */
-	error = xfs_dir2_shrink_inode(args, XFS_DIR2_FREE_FIRSTDB(mp), fbp);
+	error = xfs_dir2_shrink_inode(args,
+			xfs_dir2_byte_to_db(args->geo, XFS_DIR2_FREE_OFFSET),
+			fbp);
 	if (error) {
 		/*
 		 * This can't fail here because it can only happen when
 		 * punching out the middle of an extent, and this is an
 		 * isolated block.
 		 */
-		ASSERT(error != ENOSPC);
+		ASSERT(error != -ENOSPC);
 		return error;
 	}
 	fbp = NULL;
diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/libxfs/xfs_dir2_node.c
index 56369d4509d5..2ae6ac2c11ae 100644
--- a/fs/xfs/xfs_dir2_node.c
+++ b/fs/xfs/libxfs/xfs_dir2_node.c
@@ -115,13 +115,14 @@ xfs_dir3_free_read_verify(
 {
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 
-	if ((xfs_sb_version_hascrc(&mp->m_sb) &&
-	     !xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
-					  XFS_DIR3_FREE_CRC_OFF)) ||
-	    !xfs_dir3_free_verify(bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
-	}
+	if (xfs_sb_version_hascrc(&mp->m_sb) &&
+	    !xfs_buf_verify_cksum(bp, XFS_DIR3_FREE_CRC_OFF))
+		xfs_buf_ioerror(bp, -EFSBADCRC);
+	else if (!xfs_dir3_free_verify(bp))
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+
+	if (bp->b_error)
+		xfs_verifier_error(bp);
 }
 
 static void
@@ -133,8 +134,8 @@ xfs_dir3_free_write_verify(
 	struct xfs_dir3_blk_hdr	*hdr3 = bp->b_addr;
 
 	if (!xfs_dir3_free_verify(bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
 		return;
 	}
 
@@ -144,7 +145,7 @@ xfs_dir3_free_write_verify(
 	if (bip)
 		hdr3->lsn = cpu_to_be64(bip->bli_item.li_lsn);
 
-	xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length), XFS_DIR3_FREE_CRC_OFF);
+	xfs_buf_update_cksum(bp, XFS_DIR3_FREE_CRC_OFF);
 }
 
 const struct xfs_buf_ops xfs_dir3_free_buf_ops = {
@@ -194,17 +195,18 @@ xfs_dir2_free_try_read(
 
 static int
 xfs_dir3_free_get_buf(
-	struct xfs_trans	*tp,
-	struct xfs_inode	*dp,
+	xfs_da_args_t		*args,
 	xfs_dir2_db_t		fbno,
 	struct xfs_buf		**bpp)
 {
+	struct xfs_trans	*tp = args->trans;
+	struct xfs_inode	*dp = args->dp;
 	struct xfs_mount	*mp = dp->i_mount;
 	struct xfs_buf		*bp;
 	int			error;
 	struct xfs_dir3_icfree_hdr hdr;
 
-	error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(mp, fbno),
+	error = xfs_da_get_buf(tp, dp, xfs_dir2_db_to_da(args->geo, fbno),
 				   -1, &bp, XFS_DATA_FORK);
 	if (error)
 		return error;
@@ -239,8 +241,7 @@ xfs_dir3_free_get_buf(
  */
 STATIC void
 xfs_dir2_free_log_bests(
-	struct xfs_trans	*tp,
-	struct xfs_inode	*dp,
+	struct xfs_da_args	*args,
 	struct xfs_buf		*bp,
 	int			first,		/* first entry to log */
 	int			last)		/* last entry to log */
@@ -249,10 +250,10 @@ xfs_dir2_free_log_bests(
 	__be16			*bests;
 
 	free = bp->b_addr;
-	bests = dp->d_ops->free_bests_p(free);
+	bests = args->dp->d_ops->free_bests_p(free);
 	ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) ||
 	       free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC));
-	xfs_trans_log_buf(tp, bp,
+	xfs_trans_log_buf(args->trans, bp,
 		(uint)((char *)&bests[first] - (char *)free),
 		(uint)((char *)&bests[last] - (char *)free +
 		       sizeof(bests[0]) - 1));
@@ -263,8 +264,7 @@ xfs_dir2_free_log_bests(
  */
 static void
 xfs_dir2_free_log_header(
-	struct xfs_trans	*tp,
-	struct xfs_inode	*dp,
+	struct xfs_da_args	*args,
 	struct xfs_buf		*bp)
 {
 #ifdef DEBUG
@@ -274,7 +274,8 @@ xfs_dir2_free_log_header(
 	ASSERT(free->hdr.magic == cpu_to_be32(XFS_DIR2_FREE_MAGIC) ||
 	       free->hdr.magic == cpu_to_be32(XFS_DIR3_FREE_MAGIC));
 #endif
-	xfs_trans_log_buf(tp, bp, 0, dp->d_ops->free_hdr_size - 1);
+	xfs_trans_log_buf(args->trans, bp, 0,
+			  args->dp->d_ops->free_hdr_size - 1);
 }
 
 /*
@@ -314,20 +315,20 @@ xfs_dir2_leaf_to_node(
 	if ((error = xfs_dir2_grow_inode(args, XFS_DIR2_FREE_SPACE, &fdb))) {
 		return error;
 	}
-	ASSERT(fdb == XFS_DIR2_FREE_FIRSTDB(mp));
+	ASSERT(fdb == xfs_dir2_byte_to_db(args->geo, XFS_DIR2_FREE_OFFSET));
 	/*
 	 * Get the buffer for the new freespace block.
 	 */
-	error = xfs_dir3_free_get_buf(tp, dp, fdb, &fbp);
+	error = xfs_dir3_free_get_buf(args, fdb, &fbp);
 	if (error)
 		return error;
 
 	free = fbp->b_addr;
 	dp->d_ops->free_hdr_from_disk(&freehdr, free);
 	leaf = lbp->b_addr;
-	ltp = xfs_dir2_leaf_tail_p(mp, leaf);
+	ltp = xfs_dir2_leaf_tail_p(args->geo, leaf);
 	ASSERT(be32_to_cpu(ltp->bestcount) <=
-				(uint)dp->i_d.di_size / mp->m_dirblksize);
+				(uint)dp->i_d.di_size / args->geo->blksize);
 
 	/*
 	 * Copy freespace entries from the leaf block to the new block.
@@ -348,8 +349,8 @@ xfs_dir2_leaf_to_node(
 	freehdr.nvalid = be32_to_cpu(ltp->bestcount);
 
 	dp->d_ops->free_hdr_to_disk(fbp->b_addr, &freehdr);
-	xfs_dir2_free_log_bests(tp, dp, fbp, 0, freehdr.nvalid - 1);
-	xfs_dir2_free_log_header(tp, dp, fbp);
+	xfs_dir2_free_log_bests(args, fbp, 0, freehdr.nvalid - 1);
+	xfs_dir2_free_log_header(args, fbp);
 
 	/*
 	 * Converting the leaf to a leafnode is just a matter of changing the
@@ -363,7 +364,7 @@ xfs_dir2_leaf_to_node(
 		leaf->hdr.info.magic = cpu_to_be16(XFS_DIR3_LEAFN_MAGIC);
 	lbp->b_ops = &xfs_dir3_leafn_buf_ops;
 	xfs_trans_buf_set_type(tp, lbp, XFS_BLFT_DIR_LEAFN_BUF);
-	xfs_dir3_leaf_log_header(tp, dp, lbp);
+	xfs_dir3_leaf_log_header(args, lbp);
 	xfs_dir3_leaf_check(dp, lbp);
 	return 0;
 }
@@ -405,7 +406,7 @@ xfs_dir2_leafn_add(
 	 * into other peoples memory
 	 */
 	if (index < 0)
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 
 	/*
 	 * If there are already the maximum number of leaf entries in
@@ -414,9 +415,9 @@ xfs_dir2_leafn_add(
 	 * a compact.
 	 */
 
-	if (leafhdr.count == dp->d_ops->leaf_max_ents(mp)) {
+	if (leafhdr.count == dp->d_ops->leaf_max_ents(args->geo)) {
 		if (!leafhdr.stale)
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		compact = leafhdr.stale > 1;
 	} else
 		compact = 0;
@@ -449,12 +450,12 @@ xfs_dir2_leafn_add(
 				       highstale, &lfloglow, &lfloghigh);
 
 	lep->hashval = cpu_to_be32(args->hashval);
-	lep->address = cpu_to_be32(xfs_dir2_db_off_to_dataptr(mp,
+	lep->address = cpu_to_be32(xfs_dir2_db_off_to_dataptr(args->geo,
 				args->blkno, args->index));
 
 	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
-	xfs_dir3_leaf_log_header(tp, dp, bp);
-	xfs_dir3_leaf_log_ents(tp, dp, bp, lfloglow, lfloghigh);
+	xfs_dir3_leaf_log_header(args, bp);
+	xfs_dir3_leaf_log_ents(args, bp, lfloglow, lfloghigh);
 	xfs_dir3_leaf_check(dp, bp);
 	return 0;
 }
@@ -470,7 +471,8 @@ xfs_dir2_free_hdr_check(
 
 	dp->d_ops->free_hdr_from_disk(&hdr, bp->b_addr);
 
-	ASSERT((hdr.firstdb % dp->d_ops->free_max_bests(dp->i_mount)) == 0);
+	ASSERT((hdr.firstdb %
+		dp->d_ops->free_max_bests(dp->i_mount->m_dir_geo)) == 0);
 	ASSERT(hdr.firstdb <= db);
 	ASSERT(db < hdr.firstdb + hdr.nvalid);
 }
@@ -575,7 +577,8 @@ xfs_dir2_leafn_lookup_for_addname(
 		/*
 		 * Pull the data block number from the entry.
 		 */
-		newdb = xfs_dir2_dataptr_to_db(mp, be32_to_cpu(lep->address));
+		newdb = xfs_dir2_dataptr_to_db(args->geo,
+					       be32_to_cpu(lep->address));
 		/*
 		 * For addname, we're looking for a place to put the new entry.
 		 * We want to use a data block with an entry of equal
@@ -592,7 +595,7 @@ xfs_dir2_leafn_lookup_for_addname(
 			 * Convert the data block to the free block
 			 * holding its freespace information.
 			 */
-			newfdb = dp->d_ops->db_to_fdb(mp, newdb);
+			newfdb = dp->d_ops->db_to_fdb(args->geo, newdb);
 			/*
 			 * If it's not the one we have in hand, read it in.
 			 */
@@ -604,7 +607,8 @@ xfs_dir2_leafn_lookup_for_addname(
 					xfs_trans_brelse(tp, curbp);
 
 				error = xfs_dir2_free_read(tp, dp,
-						xfs_dir2_db_to_da(mp, newfdb),
+						xfs_dir2_db_to_da(args->geo,
+								  newfdb),
 						&curbp);
 				if (error)
 					return error;
@@ -615,7 +619,7 @@ xfs_dir2_leafn_lookup_for_addname(
 			/*
 			 * Get the index for our entry.
 			 */
-			fi = dp->d_ops->db_to_fdindex(mp, curdb);
+			fi = dp->d_ops->db_to_fdindex(args->geo, curdb);
 			/*
 			 * If it has room, return it.
 			 */
@@ -625,7 +629,7 @@ xfs_dir2_leafn_lookup_for_addname(
 							XFS_ERRLEVEL_LOW, mp);
 				if (curfdb != newfdb)
 					xfs_trans_brelse(tp, curbp);
-				return XFS_ERROR(EFSCORRUPTED);
+				return -EFSCORRUPTED;
 			}
 			curfdb = newfdb;
 			if (be16_to_cpu(bests[fi]) >= length)
@@ -656,7 +660,7 @@ out:
 	 * Return the index, that will be the insertion point.
 	 */
 	*indexp = index;
-	return XFS_ERROR(ENOENT);
+	return -ENOENT;
 }
 
 /*
@@ -720,7 +724,8 @@ xfs_dir2_leafn_lookup_for_entry(
 		/*
 		 * Pull the data block number from the entry.
 		 */
-		newdb = xfs_dir2_dataptr_to_db(mp, be32_to_cpu(lep->address));
+		newdb = xfs_dir2_dataptr_to_db(args->geo,
+					       be32_to_cpu(lep->address));
 		/*
 		 * Not adding a new entry, so we really want to find
 		 * the name given to us.
@@ -745,7 +750,8 @@ xfs_dir2_leafn_lookup_for_entry(
 				curbp = state->extrablk.bp;
 			} else {
 				error = xfs_dir3_data_read(tp, dp,
-						xfs_dir2_db_to_da(mp, newdb),
+						xfs_dir2_db_to_da(args->geo,
+								  newdb),
 						-1, &curbp);
 				if (error)
 					return error;
@@ -757,7 +763,8 @@ xfs_dir2_leafn_lookup_for_entry(
 		 * Point to the data entry.
 		 */
 		dep = (xfs_dir2_data_entry_t *)((char *)curbp->b_addr +
-			xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)));
+			xfs_dir2_dataptr_to_off(args->geo,
+						be32_to_cpu(lep->address)));
 		/*
 		 * Compare the entry and if it's an exact match, return
 		 * EEXIST immediately. If it's the first case-insensitive
@@ -782,7 +789,7 @@ xfs_dir2_leafn_lookup_for_entry(
 			curbp->b_ops = &xfs_dir3_data_buf_ops;
 			xfs_trans_buf_set_type(tp, curbp, XFS_BLFT_DIR_DATA_BUF);
 			if (cmp == XFS_CMP_EXACT)
-				return XFS_ERROR(EEXIST);
+				return -EEXIST;
 		}
 	}
 	ASSERT(index == leafhdr.count || (args->op_flags & XFS_DA_OP_OKNOENT));
@@ -805,7 +812,7 @@ xfs_dir2_leafn_lookup_for_entry(
 		state->extravalid = 0;
 	}
 	*indexp = index;
-	return XFS_ERROR(ENOENT);
+	return -ENOENT;
 }
 
 /*
@@ -843,7 +850,6 @@ xfs_dir3_leafn_moveents(
 	int				start_d,/* destination leaf index */
 	int				count)	/* count of leaves to copy */
 {
-	struct xfs_trans		*tp = args->trans;
 	int				stale;	/* count stale leaves copied */
 
 	trace_xfs_dir2_leafn_moveents(args, start_s, start_d, count);
@@ -862,7 +868,7 @@ xfs_dir3_leafn_moveents(
 	if (start_d < dhdr->count) {
 		memmove(&dents[start_d + count], &dents[start_d],
 			(dhdr->count - start_d) * sizeof(xfs_dir2_leaf_entry_t));
-		xfs_dir3_leaf_log_ents(tp, args->dp, bp_d, start_d + count,
+		xfs_dir3_leaf_log_ents(args, bp_d, start_d + count,
 				       count + dhdr->count - 1);
 	}
 	/*
@@ -884,8 +890,7 @@ xfs_dir3_leafn_moveents(
 	 */
 	memcpy(&dents[start_d], &sents[start_s],
 		count * sizeof(xfs_dir2_leaf_entry_t));
-	xfs_dir3_leaf_log_ents(tp, args->dp, bp_d,
-			       start_d, start_d + count - 1);
+	xfs_dir3_leaf_log_ents(args, bp_d, start_d, start_d + count - 1);
 
 	/*
 	 * If there are source entries after the ones we copied,
@@ -894,8 +899,7 @@ xfs_dir3_leafn_moveents(
 	if (start_s + count < shdr->count) {
 		memmove(&sents[start_s], &sents[start_s + count],
 			count * sizeof(xfs_dir2_leaf_entry_t));
-		xfs_dir3_leaf_log_ents(tp, args->dp, bp_s,
-				       start_s, start_s + count - 1);
+		xfs_dir3_leaf_log_ents(args, bp_s, start_s, start_s + count - 1);
 	}
 
 	/*
@@ -1031,8 +1035,8 @@ xfs_dir2_leafn_rebalance(
 	/* log the changes made when moving the entries */
 	dp->d_ops->leaf_hdr_to_disk(leaf1, &hdr1);
 	dp->d_ops->leaf_hdr_to_disk(leaf2, &hdr2);
-	xfs_dir3_leaf_log_header(args->trans, dp, blk1->bp);
-	xfs_dir3_leaf_log_header(args->trans, dp, blk2->bp);
+	xfs_dir3_leaf_log_header(args, blk1->bp);
+	xfs_dir3_leaf_log_header(args, blk2->bp);
 
 	xfs_dir3_leaf_check(dp, blk1->bp);
 	xfs_dir3_leaf_check(dp, blk2->bp);
@@ -1075,7 +1079,6 @@ xfs_dir3_data_block_free(
 	struct xfs_buf		*fbp,
 	int			longest)
 {
-	struct xfs_trans	*tp = args->trans;
 	int			logfree = 0;
 	__be16			*bests;
 	struct xfs_dir3_icfree_hdr freehdr;
@@ -1089,7 +1092,7 @@ xfs_dir3_data_block_free(
 		 * value.
 		 */
 		bests[findex] = cpu_to_be16(longest);
-		xfs_dir2_free_log_bests(tp, dp, fbp, findex, findex);
+		xfs_dir2_free_log_bests(args, fbp, findex, findex);
 		return 0;
 	}
 
@@ -1117,7 +1120,7 @@ xfs_dir3_data_block_free(
 	}
 
 	dp->d_ops->free_hdr_to_disk(free, &freehdr);
-	xfs_dir2_free_log_header(tp, dp, fbp);
+	xfs_dir2_free_log_header(args, fbp);
 
 	/*
 	 * If there are no useful entries left in the block, get rid of the
@@ -1130,7 +1133,7 @@ xfs_dir3_data_block_free(
 		if (error == 0) {
 			fbp = NULL;
 			logfree = 0;
-		} else if (error != ENOSPC || args->total != 0)
+		} else if (error != -ENOSPC || args->total != 0)
 			return error;
 		/*
 		 * It's possible to get ENOSPC if there is no
@@ -1141,7 +1144,7 @@ xfs_dir3_data_block_free(
 
 	/* Log the free entry that changed, unless we got rid of it.  */
 	if (logfree)
-		xfs_dir2_free_log_bests(tp, dp, fbp, findex, findex);
+		xfs_dir2_free_log_bests(args, fbp, findex, findex);
 	return 0;
 }
 
@@ -1192,9 +1195,9 @@ xfs_dir2_leafn_remove(
 	/*
 	 * Extract the data block and offset from the entry.
 	 */
-	db = xfs_dir2_dataptr_to_db(mp, be32_to_cpu(lep->address));
+	db = xfs_dir2_dataptr_to_db(args->geo, be32_to_cpu(lep->address));
 	ASSERT(dblk->blkno == db);
-	off = xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address));
+	off = xfs_dir2_dataptr_to_off(args->geo, be32_to_cpu(lep->address));
 	ASSERT(dblk->index == off);
 
 	/*
@@ -1203,10 +1206,10 @@ xfs_dir2_leafn_remove(
 	 */
 	leafhdr.stale++;
 	dp->d_ops->leaf_hdr_to_disk(leaf, &leafhdr);
-	xfs_dir3_leaf_log_header(tp, dp, bp);
+	xfs_dir3_leaf_log_header(args, bp);
 
 	lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
-	xfs_dir3_leaf_log_ents(tp, dp, bp, index, index);
+	xfs_dir3_leaf_log_ents(args, bp, index, index);
 
 	/*
 	 * Make the data entry free.  Keep track of the longest freespace
@@ -1218,7 +1221,7 @@ xfs_dir2_leafn_remove(
 	bf = dp->d_ops->data_bestfree_p(hdr);
 	longest = be16_to_cpu(bf[0].length);
 	needlog = needscan = 0;
-	xfs_dir2_data_make_free(tp, dp, dbp, off,
+	xfs_dir2_data_make_free(args, dbp, off,
 		dp->d_ops->data_entsize(dep->namelen), &needlog, &needscan);
 	/*
 	 * Rescan the data block freespaces for bestfree.
@@ -1227,7 +1230,7 @@ xfs_dir2_leafn_remove(
 	if (needscan)
 		xfs_dir2_data_freescan(dp, hdr, &needlog);
 	if (needlog)
-		xfs_dir2_data_log_header(tp, dp, dbp);
+		xfs_dir2_data_log_header(args, dbp);
 	xfs_dir3_data_check(dp, dbp);
 	/*
 	 * If the longest data block freespace changes, need to update
@@ -1244,8 +1247,9 @@ xfs_dir2_leafn_remove(
 		 * Convert the data block number to a free block,
 		 * read in the free block.
 		 */
-		fdb = dp->d_ops->db_to_fdb(mp, db);
-		error = xfs_dir2_free_read(tp, dp, xfs_dir2_db_to_da(mp, fdb),
+		fdb = dp->d_ops->db_to_fdb(args->geo, db);
+		error = xfs_dir2_free_read(tp, dp,
+					   xfs_dir2_db_to_da(args->geo, fdb),
 					   &fbp);
 		if (error)
 			return error;
@@ -1254,20 +1258,21 @@ xfs_dir2_leafn_remove(
 	{
 		struct xfs_dir3_icfree_hdr freehdr;
 		dp->d_ops->free_hdr_from_disk(&freehdr, free);
-		ASSERT(freehdr.firstdb == dp->d_ops->free_max_bests(mp) *
-					  (fdb - XFS_DIR2_FREE_FIRSTDB(mp)));
+		ASSERT(freehdr.firstdb == dp->d_ops->free_max_bests(args->geo) *
+			(fdb - xfs_dir2_byte_to_db(args->geo,
+						   XFS_DIR2_FREE_OFFSET)));
 	}
 #endif
 		/*
 		 * Calculate which entry we need to fix.
 		 */
-		findex = dp->d_ops->db_to_fdindex(mp, db);
+		findex = dp->d_ops->db_to_fdindex(args->geo, db);
 		longest = be16_to_cpu(bf[0].length);
 		/*
 		 * If the data block is now empty we can get rid of it
 		 * (usually).
 		 */
-		if (longest == mp->m_dirblksize -
+		if (longest == args->geo->blksize -
 			       dp->d_ops->data_entry_offset) {
 			/*
 			 * Try to punch out the data block.
@@ -1282,7 +1287,7 @@ xfs_dir2_leafn_remove(
 			 * In this case just drop the buffer and some one else
 			 * will eventually get rid of the empty block.
 			 */
-			else if (!(error == ENOSPC && args->total == 0))
+			else if (!(error == -ENOSPC && args->total == 0))
 				return error;
 		}
 		/*
@@ -1302,7 +1307,7 @@ xfs_dir2_leafn_remove(
 	 */
 	*rval = (dp->d_ops->leaf_hdr_size +
 		 (uint)sizeof(ents[0]) * (leafhdr.count - leafhdr.stale)) <
-		mp->m_dir_magicpct;
+		args->geo->magicpct;
 	return 0;
 }
 
@@ -1335,7 +1340,7 @@ xfs_dir2_leafn_split(
 	/*
 	 * Initialize the new leaf block.
 	 */
-	error = xfs_dir3_leaf_get_buf(args, xfs_dir2_da_to_db(mp, blkno),
+	error = xfs_dir3_leaf_get_buf(args, xfs_dir2_da_to_db(args->geo, blkno),
 				      &newblk->bp, XFS_DIR2_LEAFN_MAGIC);
 	if (error)
 		return error;
@@ -1409,7 +1414,7 @@ xfs_dir2_leafn_toosmall(
 
 	count = leafhdr.count - leafhdr.stale;
 	bytes = dp->d_ops->leaf_hdr_size + count * sizeof(ents[0]);
-	if (bytes > (state->blocksize >> 1)) {
+	if (bytes > (state->args->geo->blksize >> 1)) {
 		/*
 		 * Blk over 50%, don't try to join.
 		 */
@@ -1462,7 +1467,8 @@ xfs_dir2_leafn_toosmall(
 		 * Count bytes in the two blocks combined.
 		 */
 		count = leafhdr.count - leafhdr.stale;
-		bytes = state->blocksize - (state->blocksize >> 2);
+		bytes = state->args->geo->blksize -
+			(state->args->geo->blksize >> 2);
 
 		leaf = bp->b_addr;
 		dp->d_ops->leaf_hdr_from_disk(&hdr2, leaf);
@@ -1559,8 +1565,8 @@ xfs_dir2_leafn_unbalance(
 	/* log the changes made when moving the entries */
 	dp->d_ops->leaf_hdr_to_disk(save_leaf, &savehdr);
 	dp->d_ops->leaf_hdr_to_disk(drop_leaf, &drophdr);
-	xfs_dir3_leaf_log_header(args->trans, dp, save_blk->bp);
-	xfs_dir3_leaf_log_header(args->trans, dp, drop_blk->bp);
+	xfs_dir3_leaf_log_header(args, save_blk->bp);
+	xfs_dir3_leaf_log_header(args, drop_blk->bp);
 
 	xfs_dir3_leaf_check(dp, save_blk->bp);
 	xfs_dir3_leaf_check(dp, drop_blk->bp);
@@ -1586,8 +1592,6 @@ xfs_dir2_node_addname(
 	state = xfs_da_state_alloc();
 	state->args = args;
 	state->mp = args->dp->i_mount;
-	state->blocksize = state->mp->m_dirblksize;
-	state->node_ents = state->mp->m_dir_node_ents;
 	/*
 	 * Look up the name.  We're not supposed to find it, but
 	 * this gives us the insertion point.
@@ -1595,7 +1599,7 @@ xfs_dir2_node_addname(
 	error = xfs_da3_node_lookup_int(state, &rval);
 	if (error)
 		rval = error;
-	if (rval != ENOENT) {
+	if (rval != -ENOENT) {
 		goto done;
 	}
 	/*
@@ -1624,7 +1628,7 @@ xfs_dir2_node_addname(
 		 * It didn't work, we need to split the leaf block.
 		 */
 		if (args->total == 0) {
-			ASSERT(rval == ENOSPC);
+			ASSERT(rval == -ENOSPC);
 			goto done;
 		}
 		/*
@@ -1726,9 +1730,9 @@ xfs_dir2_node_addname_int(
 	if (dbno == -1) {
 		xfs_fileoff_t	fo;		/* freespace block number */
 
-		if ((error = xfs_bmap_last_offset(tp, dp, &fo, XFS_DATA_FORK)))
+		if ((error = xfs_bmap_last_offset(dp, &fo, XFS_DATA_FORK)))
 			return error;
-		lastfbno = xfs_dir2_da_to_db(mp, (xfs_dablk_t)fo);
+		lastfbno = xfs_dir2_da_to_db(args->geo, (xfs_dablk_t)fo);
 		fbno = ifbno;
 	}
 	/*
@@ -1746,7 +1750,8 @@ xfs_dir2_node_addname_int(
 			 * us a freespace block to start with.
 			 */
 			if (++fbno == 0)
-				fbno = XFS_DIR2_FREE_FIRSTDB(mp);
+				fbno = xfs_dir2_byte_to_db(args->geo,
+							XFS_DIR2_FREE_OFFSET);
 			/*
 			 * If it's ifbno we already looked at it.
 			 */
@@ -1764,8 +1769,8 @@ xfs_dir2_node_addname_int(
 			 * to avoid it.
 			 */
 			error = xfs_dir2_free_try_read(tp, dp,
-						xfs_dir2_db_to_da(mp, fbno),
-						&fbp);
+					xfs_dir2_db_to_da(args->geo, fbno),
+					&fbp);
 			if (error)
 				return error;
 			if (!fbp)
@@ -1810,7 +1815,7 @@ xfs_dir2_node_addname_int(
 		 * Not allowed to allocate, return failure.
 		 */
 		if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || args->total == 0)
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 
 		/*
 		 * Allocate and initialize the new data block.
@@ -1833,10 +1838,10 @@ xfs_dir2_node_addname_int(
 		 * Get the freespace block corresponding to the data block
 		 * that was just allocated.
 		 */
-		fbno = dp->d_ops->db_to_fdb(mp, dbno);
+		fbno = dp->d_ops->db_to_fdb(args->geo, dbno);
 		error = xfs_dir2_free_try_read(tp, dp,
-					       xfs_dir2_db_to_da(mp, fbno),
-					       &fbp);
+				       xfs_dir2_db_to_da(args->geo, fbno),
+				       &fbp);
 		if (error)
 			return error;
 
@@ -1850,12 +1855,13 @@ xfs_dir2_node_addname_int(
 			if (error)
 				return error;
 
-			if (unlikely(dp->d_ops->db_to_fdb(mp, dbno) != fbno)) {
+			if (dp->d_ops->db_to_fdb(args->geo, dbno) != fbno) {
 				xfs_alert(mp,
 			"%s: dir ino %llu needed freesp block %lld for\n"
 			"  data block %lld, got %lld ifbno %llu lastfbno %d",
 					__func__, (unsigned long long)dp->i_ino,
-					(long long)dp->d_ops->db_to_fdb(mp, dbno),
+					(long long)dp->d_ops->db_to_fdb(
+								args->geo, dbno),
 					(long long)dbno, (long long)fbno,
 					(unsigned long long)ifbno, lastfbno);
 				if (fblk) {
@@ -1870,13 +1876,13 @@ xfs_dir2_node_addname_int(
 				}
 				XFS_ERROR_REPORT("xfs_dir2_node_addname_int",
 						 XFS_ERRLEVEL_LOW, mp);
-				return XFS_ERROR(EFSCORRUPTED);
+				return -EFSCORRUPTED;
 			}
 
 			/*
 			 * Get a buffer for the new block.
 			 */
-			error = xfs_dir3_free_get_buf(tp, dp, fbno, &fbp);
+			error = xfs_dir3_free_get_buf(args, fbno, &fbp);
 			if (error)
 				return error;
 			free = fbp->b_addr;
@@ -1886,8 +1892,10 @@ xfs_dir2_node_addname_int(
 			/*
 			 * Remember the first slot as our empty slot.
 			 */
-			freehdr.firstdb = (fbno - XFS_DIR2_FREE_FIRSTDB(mp)) *
-					dp->d_ops->free_max_bests(mp);
+			freehdr.firstdb =
+				(fbno - xfs_dir2_byte_to_db(args->geo,
+							XFS_DIR2_FREE_OFFSET)) *
+					dp->d_ops->free_max_bests(args->geo);
 		} else {
 			free = fbp->b_addr;
 			bests = dp->d_ops->free_bests_p(free);
@@ -1897,13 +1905,13 @@ xfs_dir2_node_addname_int(
 		/*
 		 * Set the freespace block index from the data block number.
 		 */
-		findex = dp->d_ops->db_to_fdindex(mp, dbno);
+		findex = dp->d_ops->db_to_fdindex(args->geo, dbno);
 		/*
 		 * If it's after the end of the current entries in the
 		 * freespace block, extend that table.
 		 */
 		if (findex >= freehdr.nvalid) {
-			ASSERT(findex < dp->d_ops->free_max_bests(mp));
+			ASSERT(findex < dp->d_ops->free_max_bests(args->geo));
 			freehdr.nvalid = findex + 1;
 			/*
 			 * Tag new entry so nused will go up.
@@ -1917,7 +1925,7 @@ xfs_dir2_node_addname_int(
 		if (bests[findex] == cpu_to_be16(NULLDATAOFF)) {
 			freehdr.nused++;
 			dp->d_ops->free_hdr_to_disk(fbp->b_addr, &freehdr);
-			xfs_dir2_free_log_header(tp, dp, fbp);
+			xfs_dir2_free_log_header(args, fbp);
 		}
 		/*
 		 * Update the real value in the table.
@@ -1942,7 +1950,8 @@ xfs_dir2_node_addname_int(
 		/*
 		 * Read the data block in.
 		 */
-		error = xfs_dir3_data_read(tp, dp, xfs_dir2_db_to_da(mp, dbno),
+		error = xfs_dir3_data_read(tp, dp,
+					   xfs_dir2_db_to_da(args->geo, dbno),
 					   -1, &dbp);
 		if (error)
 			return error;
@@ -1960,7 +1969,7 @@ xfs_dir2_node_addname_int(
 	/*
 	 * Mark the first part of the unused space, inuse for us.
 	 */
-	xfs_dir2_data_use_free(tp, dp, dbp, dup,
+	xfs_dir2_data_use_free(args, dbp, dup,
 		(xfs_dir2_data_aoff_t)((char *)dup - (char *)hdr), length,
 		&needlog, &needscan);
 	/*
@@ -1973,7 +1982,7 @@ xfs_dir2_node_addname_int(
 	dp->d_ops->data_put_ftype(dep, args->filetype);
 	tagp = dp->d_ops->data_entry_tag_p(dep);
 	*tagp = cpu_to_be16((char *)dep - (char *)hdr);
-	xfs_dir2_data_log_entry(tp, dp, dbp, dep);
+	xfs_dir2_data_log_entry(args, dbp, dep);
 	/*
 	 * Rescan the block for bestfree if needed.
 	 */
@@ -1983,7 +1992,7 @@ xfs_dir2_node_addname_int(
 	 * Log the data block header if needed.
 	 */
 	if (needlog)
-		xfs_dir2_data_log_header(tp, dp, dbp);
+		xfs_dir2_data_log_header(args, dbp);
 	/*
 	 * If the freespace entry is now wrong, update it.
 	 */
@@ -1996,7 +2005,7 @@ xfs_dir2_node_addname_int(
 	 * Log the freespace entry if needed.
 	 */
 	if (logfree)
-		xfs_dir2_free_log_bests(tp, dp, fbp, findex, findex);
+		xfs_dir2_free_log_bests(args, fbp, findex, findex);
 	/*
 	 * Return the data block and offset in args, then drop the data block.
 	 */
@@ -2027,16 +2036,14 @@ xfs_dir2_node_lookup(
 	state = xfs_da_state_alloc();
 	state->args = args;
 	state->mp = args->dp->i_mount;
-	state->blocksize = state->mp->m_dirblksize;
-	state->node_ents = state->mp->m_dir_node_ents;
 	/*
 	 * Fill in the path to the entry in the cursor.
 	 */
 	error = xfs_da3_node_lookup_int(state, &rval);
 	if (error)
 		rval = error;
-	else if (rval == ENOENT && args->cmpresult == XFS_CMP_CASE) {
-		/* If a CI match, dup the actual name and return EEXIST */
+	else if (rval == -ENOENT && args->cmpresult == XFS_CMP_CASE) {
+		/* If a CI match, dup the actual name and return -EEXIST */
 		xfs_dir2_data_entry_t	*dep;
 
 		dep = (xfs_dir2_data_entry_t *)
@@ -2067,12 +2074,12 @@ xfs_dir2_node_lookup(
  */
 int						/* error */
 xfs_dir2_node_removename(
-	xfs_da_args_t		*args)		/* operation arguments */
+	struct xfs_da_args	*args)		/* operation arguments */
 {
-	xfs_da_state_blk_t	*blk;		/* leaf block */
+	struct xfs_da_state_blk	*blk;		/* leaf block */
 	int			error;		/* error return value */
 	int			rval;		/* operation return value */
-	xfs_da_state_t		*state;		/* btree cursor */
+	struct xfs_da_state	*state;		/* btree cursor */
 
 	trace_xfs_dir2_node_removename(args);
 
@@ -2082,21 +2089,18 @@ xfs_dir2_node_removename(
 	state = xfs_da_state_alloc();
 	state->args = args;
 	state->mp = args->dp->i_mount;
-	state->blocksize = state->mp->m_dirblksize;
-	state->node_ents = state->mp->m_dir_node_ents;
-	/*
-	 * Look up the entry we're deleting, set up the cursor.
-	 */
+
+	/* Look up the entry we're deleting, set up the cursor. */
 	error = xfs_da3_node_lookup_int(state, &rval);
 	if (error)
-		rval = error;
-	/*
-	 * Didn't find it, upper layer screwed up.
-	 */
-	if (rval != EEXIST) {
-		xfs_da_state_free(state);
-		return rval;
+		goto out_free;
+
+	/* Didn't find it, upper layer screwed up. */
+	if (rval != -EEXIST) {
+		error = rval;
+		goto out_free;
 	}
+
 	blk = &state->path.blk[state->path.active - 1];
 	ASSERT(blk->magic == XFS_DIR2_LEAFN_MAGIC);
 	ASSERT(state->extravalid);
@@ -2107,7 +2111,7 @@ xfs_dir2_node_removename(
 	error = xfs_dir2_leafn_remove(args, blk->bp, blk->index,
 		&state->extrablk, &rval);
 	if (error)
-		return error;
+		goto out_free;
 	/*
 	 * Fix the hash values up the btree.
 	 */
@@ -2122,6 +2126,7 @@ xfs_dir2_node_removename(
 	 */
 	if (!error)
 		error = xfs_dir2_node_to_leaf(state);
+out_free:
 	xfs_da_state_free(state);
 	return error;
 }
@@ -2152,8 +2157,6 @@ xfs_dir2_node_replace(
 	state = xfs_da_state_alloc();
 	state->args = args;
 	state->mp = args->dp->i_mount;
-	state->blocksize = state->mp->m_dirblksize;
-	state->node_ents = state->mp->m_dir_node_ents;
 	inum = args->inumber;
 	/*
 	 * Lookup the entry to change in the btree.
@@ -2166,7 +2169,7 @@ xfs_dir2_node_replace(
 	 * It should be found, since the vnodeops layer has looked it up
 	 * and locked it.  But paranoia is good.
 	 */
-	if (rval == EEXIST) {
+	if (rval == -EEXIST) {
 		struct xfs_dir2_leaf_entry *ents;
 		/*
 		 * Find the leaf entry.
@@ -2185,15 +2188,15 @@ xfs_dir2_node_replace(
 		       hdr->magic == cpu_to_be32(XFS_DIR3_DATA_MAGIC));
 		dep = (xfs_dir2_data_entry_t *)
 		      ((char *)hdr +
-		       xfs_dir2_dataptr_to_off(state->mp, be32_to_cpu(lep->address)));
+		       xfs_dir2_dataptr_to_off(args->geo,
+					       be32_to_cpu(lep->address)));
 		ASSERT(inum != be64_to_cpu(dep->inumber));
 		/*
 		 * Fill in the new inode number and log the entry.
 		 */
 		dep->inumber = cpu_to_be64(inum);
 		args->dp->d_ops->data_put_ftype(dep, args->filetype);
-		xfs_dir2_data_log_entry(args->trans, args->dp,
-					state->extrablk.bp, dep);
+		xfs_dir2_data_log_entry(args, state->extrablk.bp, dep);
 		rval = 0;
 	}
 	/*
@@ -2261,15 +2264,15 @@ xfs_dir2_node_trim_free(
 	/*
 	 * Blow the block away.
 	 */
-	if ((error =
-	    xfs_dir2_shrink_inode(args, xfs_dir2_da_to_db(mp, (xfs_dablk_t)fo),
-		    bp))) {
+	error = xfs_dir2_shrink_inode(args,
+			xfs_dir2_da_to_db(args->geo, (xfs_dablk_t)fo), bp);
+	if (error) {
 		/*
 		 * Can't fail with ENOSPC since that only happens with no
 		 * space reservation, when breaking up an extent into two
 		 * pieces.  This is the last block of an extent.
 		 */
-		ASSERT(error != ENOSPC);
+		ASSERT(error != -ENOSPC);
 		xfs_trans_brelse(tp, bp);
 		return error;
 	}
diff --git a/fs/xfs/xfs_dir2_priv.h b/fs/xfs/libxfs/xfs_dir2_priv.h
index 8b9d2281f85b..27ce0794d196 100644
--- a/fs/xfs/xfs_dir2_priv.h
+++ b/fs/xfs/libxfs/xfs_dir2_priv.h
@@ -20,6 +20,140 @@
 
 struct dir_context;
 
+/*
+ * Directory offset/block conversion functions.
+ *
+ * DB blocks here are logical directory block numbers, not filesystem blocks.
+ */
+
+/*
+ * Convert dataptr to byte in file space
+ */
+static inline xfs_dir2_off_t
+xfs_dir2_dataptr_to_byte(xfs_dir2_dataptr_t dp)
+{
+	return (xfs_dir2_off_t)dp << XFS_DIR2_DATA_ALIGN_LOG;
+}
+
+/*
+ * Convert byte in file space to dataptr.  It had better be aligned.
+ */
+static inline xfs_dir2_dataptr_t
+xfs_dir2_byte_to_dataptr(xfs_dir2_off_t by)
+{
+	return (xfs_dir2_dataptr_t)(by >> XFS_DIR2_DATA_ALIGN_LOG);
+}
+
+/*
+ * Convert byte in space to (DB) block
+ */
+static inline xfs_dir2_db_t
+xfs_dir2_byte_to_db(struct xfs_da_geometry *geo, xfs_dir2_off_t by)
+{
+	return (xfs_dir2_db_t)(by >> geo->blklog);
+}
+
+/*
+ * Convert dataptr to a block number
+ */
+static inline xfs_dir2_db_t
+xfs_dir2_dataptr_to_db(struct xfs_da_geometry *geo, xfs_dir2_dataptr_t dp)
+{
+	return xfs_dir2_byte_to_db(geo, xfs_dir2_dataptr_to_byte(dp));
+}
+
+/*
+ * Convert byte in space to offset in a block
+ */
+static inline xfs_dir2_data_aoff_t
+xfs_dir2_byte_to_off(struct xfs_da_geometry *geo, xfs_dir2_off_t by)
+{
+	return (xfs_dir2_data_aoff_t)(by & (geo->blksize - 1));
+}
+
+/*
+ * Convert dataptr to a byte offset in a block
+ */
+static inline xfs_dir2_data_aoff_t
+xfs_dir2_dataptr_to_off(struct xfs_da_geometry *geo, xfs_dir2_dataptr_t dp)
+{
+	return xfs_dir2_byte_to_off(geo, xfs_dir2_dataptr_to_byte(dp));
+}
+
+/*
+ * Convert block and offset to byte in space
+ */
+static inline xfs_dir2_off_t
+xfs_dir2_db_off_to_byte(struct xfs_da_geometry *geo, xfs_dir2_db_t db,
+			xfs_dir2_data_aoff_t o)
+{
+	return ((xfs_dir2_off_t)db << geo->blklog) + o;
+}
+
+/*
+ * Convert block (DB) to block (dablk)
+ */
+static inline xfs_dablk_t
+xfs_dir2_db_to_da(struct xfs_da_geometry *geo, xfs_dir2_db_t db)
+{
+	return (xfs_dablk_t)(db << (geo->blklog - geo->fsblog));
+}
+
+/*
+ * Convert byte in space to (DA) block
+ */
+static inline xfs_dablk_t
+xfs_dir2_byte_to_da(struct xfs_da_geometry *geo, xfs_dir2_off_t by)
+{
+	return xfs_dir2_db_to_da(geo, xfs_dir2_byte_to_db(geo, by));
+}
+
+/*
+ * Convert block and offset to dataptr
+ */
+static inline xfs_dir2_dataptr_t
+xfs_dir2_db_off_to_dataptr(struct xfs_da_geometry *geo, xfs_dir2_db_t db,
+			   xfs_dir2_data_aoff_t o)
+{
+	return xfs_dir2_byte_to_dataptr(xfs_dir2_db_off_to_byte(geo, db, o));
+}
+
+/*
+ * Convert block (dablk) to block (DB)
+ */
+static inline xfs_dir2_db_t
+xfs_dir2_da_to_db(struct xfs_da_geometry *geo, xfs_dablk_t da)
+{
+	return (xfs_dir2_db_t)(da >> (geo->blklog - geo->fsblog));
+}
+
+/*
+ * Convert block (dablk) to byte offset in space
+ */
+static inline xfs_dir2_off_t
+xfs_dir2_da_to_byte(struct xfs_da_geometry *geo, xfs_dablk_t da)
+{
+	return xfs_dir2_db_off_to_byte(geo, xfs_dir2_da_to_db(geo, da), 0);
+}
+
+/*
+ * Directory tail pointer accessor functions. Based on block geometry.
+ */
+static inline struct xfs_dir2_block_tail *
+xfs_dir2_block_tail_p(struct xfs_da_geometry *geo, struct xfs_dir2_data_hdr *hdr)
+{
+	return ((struct xfs_dir2_block_tail *)
+		((char *)hdr + geo->blksize)) - 1;
+}
+
+static inline struct xfs_dir2_leaf_tail *
+xfs_dir2_leaf_tail_p(struct xfs_da_geometry *geo, struct xfs_dir2_leaf *lp)
+{
+	return (struct xfs_dir2_leaf_tail *)
+		((char *)lp + geo->blksize -
+		  sizeof(struct xfs_dir2_leaf_tail));
+}
+
 /* xfs_dir2.c */
 extern int xfs_dir_ino_validate(struct xfs_mount *mp, xfs_ino_t ino);
 extern int xfs_dir2_grow_inode(struct xfs_da_args *args, int space,
@@ -54,8 +188,8 @@ extern int xfs_dir2_leaf_to_block(struct xfs_da_args *args,
 extern int __xfs_dir3_data_check(struct xfs_inode *dp, struct xfs_buf *bp);
 extern int xfs_dir3_data_read(struct xfs_trans *tp, struct xfs_inode *dp,
 		xfs_dablk_t bno, xfs_daddr_t mapped_bno, struct xfs_buf **bpp);
-extern int xfs_dir3_data_readahead(struct xfs_trans *tp, struct xfs_inode *dp,
-		xfs_dablk_t bno, xfs_daddr_t mapped_bno);
+extern int xfs_dir3_data_readahead(struct xfs_inode *dp, xfs_dablk_t bno,
+		xfs_daddr_t mapped_bno);
 
 extern struct xfs_dir2_data_free *
 xfs_dir2_data_freeinsert(struct xfs_dir2_data_hdr *hdr,
@@ -77,9 +211,9 @@ extern void xfs_dir3_leaf_compact_x1(struct xfs_dir3_icleaf_hdr *leafhdr,
 		int *lowstalep, int *highstalep, int *lowlogp, int *highlogp);
 extern int xfs_dir3_leaf_get_buf(struct xfs_da_args *args, xfs_dir2_db_t bno,
 		struct xfs_buf **bpp, __uint16_t magic);
-extern void xfs_dir3_leaf_log_ents(struct xfs_trans *tp, struct xfs_inode *dp,
+extern void xfs_dir3_leaf_log_ents(struct xfs_da_args *args,
 		struct xfs_buf *bp, int first, int last);
-extern void xfs_dir3_leaf_log_header(struct xfs_trans *tp, struct xfs_inode *dp,
+extern void xfs_dir3_leaf_log_header(struct xfs_da_args *args,
 		struct xfs_buf *bp);
 extern int xfs_dir2_leaf_lookup(struct xfs_da_args *args);
 extern int xfs_dir2_leaf_removename(struct xfs_da_args *args);
diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/libxfs/xfs_dir2_sf.c
index aafc6e46cb58..5079e051ef08 100644
--- a/fs/xfs/xfs_dir2_sf.c
+++ b/fs/xfs/libxfs/xfs_dir2_sf.c
@@ -51,10 +51,9 @@ static void xfs_dir2_sf_check(xfs_da_args_t *args);
 #else
 #define	xfs_dir2_sf_check(args)
 #endif /* DEBUG */
-#if XFS_BIG_INUMS
+
 static void xfs_dir2_sf_toino4(xfs_da_args_t *args);
 static void xfs_dir2_sf_toino8(xfs_da_args_t *args);
-#endif /* XFS_BIG_INUMS */
 
 /*
  * Given a block directory (dp/block), calculate its size as a shortform (sf)
@@ -82,8 +81,10 @@ xfs_dir2_block_sfsize(
 	xfs_ino_t		parent = 0;	/* parent inode number */
 	int			size=0;		/* total computed size */
 	int			has_ftype;
+	struct xfs_da_geometry	*geo;
 
 	mp = dp->i_mount;
+	geo = mp->m_dir_geo;
 
 	/*
 	 * if there is a filetype field, add the extra byte to the namelen
@@ -92,7 +93,7 @@ xfs_dir2_block_sfsize(
 	has_ftype = xfs_sb_version_hasftype(&mp->m_sb) ? 1 : 0;
 
 	count = i8count = namelen = 0;
-	btp = xfs_dir2_block_tail_p(mp, hdr);
+	btp = xfs_dir2_block_tail_p(geo, hdr);
 	blp = xfs_dir2_block_leaf_p(btp);
 
 	/*
@@ -104,8 +105,8 @@ xfs_dir2_block_sfsize(
 		/*
 		 * Calculate the pointer to the entry at hand.
 		 */
-		dep = (xfs_dir2_data_entry_t *)
-		      ((char *)hdr + xfs_dir2_dataptr_to_off(mp, addr));
+		dep = (xfs_dir2_data_entry_t *)((char *)hdr +
+				xfs_dir2_dataptr_to_off(geo, addr));
 		/*
 		 * Detect . and .., so we can special-case them.
 		 * . is not included in sf directories.
@@ -115,10 +116,10 @@ xfs_dir2_block_sfsize(
 		isdotdot =
 			dep->namelen == 2 &&
 			dep->name[0] == '.' && dep->name[1] == '.';
-#if XFS_BIG_INUMS
+
 		if (!isdot)
 			i8count += be64_to_cpu(dep->inumber) > XFS_DIR2_MAX_SHORT_INUM;
-#endif
+
 		/* take into account the file type field */
 		if (!isdot && !isdotdot) {
 			count++;
@@ -170,6 +171,7 @@ xfs_dir2_block_to_sf(
 	char			*ptr;		/* current data pointer */
 	xfs_dir2_sf_entry_t	*sfep;		/* shortform entry */
 	xfs_dir2_sf_hdr_t	*sfp;		/* shortform directory header */
+	xfs_dir2_sf_hdr_t	*dst;		/* temporary data buffer */
 
 	trace_xfs_dir2_block_to_sf(args);
 
@@ -177,39 +179,24 @@ xfs_dir2_block_to_sf(
 	mp = dp->i_mount;
 
 	/*
-	 * Make a copy of the block data, so we can shrink the inode
-	 * and add local data.
+	 * allocate a temporary destination buffer the size of the inode
+	 * to format the data into. Once we have formatted the data, we
+	 * can free the block and copy the formatted data into the inode literal
+	 * area.
 	 */
-	hdr = kmem_alloc(mp->m_dirblksize, KM_SLEEP);
-	memcpy(hdr, bp->b_addr, mp->m_dirblksize);
-	logflags = XFS_ILOG_CORE;
-	if ((error = xfs_dir2_shrink_inode(args, mp->m_dirdatablk, bp))) {
-		ASSERT(error != ENOSPC);
-		goto out;
-	}
+	dst = kmem_alloc(mp->m_sb.sb_inodesize, KM_SLEEP);
+	hdr = bp->b_addr;
 
 	/*
-	 * The buffer is now unconditionally gone, whether
-	 * xfs_dir2_shrink_inode worked or not.
-	 *
-	 * Convert the inode to local format.
-	 */
-	dp->i_df.if_flags &= ~XFS_IFEXTENTS;
-	dp->i_df.if_flags |= XFS_IFINLINE;
-	dp->i_d.di_format = XFS_DINODE_FMT_LOCAL;
-	ASSERT(dp->i_df.if_bytes == 0);
-	xfs_idata_realloc(dp, size, XFS_DATA_FORK);
-	logflags |= XFS_ILOG_DDATA;
-	/*
 	 * Copy the header into the newly allocate local space.
 	 */
-	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
+	sfp = (xfs_dir2_sf_hdr_t *)dst;
 	memcpy(sfp, sfhp, xfs_dir2_sf_hdr_size(sfhp->i8count));
-	dp->i_d.di_size = size;
+
 	/*
 	 * Set up to loop over the block's entries.
 	 */
-	btp = xfs_dir2_block_tail_p(mp, hdr);
+	btp = xfs_dir2_block_tail_p(args->geo, hdr);
 	ptr = (char *)dp->d_ops->data_entry_p(hdr);
 	endptr = (char *)xfs_dir2_block_leaf_p(btp);
 	sfep = xfs_dir2_sf_firstentry(sfp);
@@ -258,10 +245,34 @@ xfs_dir2_block_to_sf(
 		ptr += dp->d_ops->data_entsize(dep->namelen);
 	}
 	ASSERT((char *)sfep - (char *)sfp == size);
+
+	/* now we are done with the block, we can shrink the inode */
+	logflags = XFS_ILOG_CORE;
+	error = xfs_dir2_shrink_inode(args, args->geo->datablk, bp);
+	if (error) {
+		ASSERT(error != -ENOSPC);
+		goto out;
+	}
+
+	/*
+	 * The buffer is now unconditionally gone, whether
+	 * xfs_dir2_shrink_inode worked or not.
+	 *
+	 * Convert the inode to local format and copy the data in.
+	 */
+	dp->i_df.if_flags &= ~XFS_IFEXTENTS;
+	dp->i_df.if_flags |= XFS_IFINLINE;
+	dp->i_d.di_format = XFS_DINODE_FMT_LOCAL;
+	ASSERT(dp->i_df.if_bytes == 0);
+	xfs_idata_realloc(dp, size, XFS_DATA_FORK);
+
+	logflags |= XFS_ILOG_DDATA;
+	memcpy(dp->i_df.if_u1.if_data, dst, size);
+	dp->i_d.di_size = size;
 	xfs_dir2_sf_check(args);
 out:
 	xfs_trans_log_inode(args->trans, dp, logflags);
-	kmem_free(hdr);
+	kmem_free(dst);
 	return error;
 }
 
@@ -275,21 +286,19 @@ int						/* error */
 xfs_dir2_sf_addname(
 	xfs_da_args_t		*args)		/* operation arguments */
 {
-	int			add_entsize;	/* size of the new entry */
 	xfs_inode_t		*dp;		/* incore directory inode */
 	int			error;		/* error return value */
 	int			incr_isize;	/* total change in size */
 	int			new_isize;	/* di_size after adding name */
 	int			objchange;	/* changing to 8-byte inodes */
 	xfs_dir2_data_aoff_t	offset = 0;	/* offset for new entry */
-	int			old_isize;	/* di_size before adding name */
 	int			pick;		/* which algorithm to use */
 	xfs_dir2_sf_hdr_t	*sfp;		/* shortform structure */
 	xfs_dir2_sf_entry_t	*sfep = NULL;	/* shortform entry */
 
 	trace_xfs_dir2_sf_addname(args);
 
-	ASSERT(xfs_dir2_sf_lookup(args) == ENOENT);
+	ASSERT(xfs_dir2_sf_lookup(args) == -ENOENT);
 	dp = args->dp;
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
 	/*
@@ -297,7 +306,7 @@ xfs_dir2_sf_addname(
 	 */
 	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
 		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 	ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
@@ -306,29 +315,24 @@ xfs_dir2_sf_addname(
 	/*
 	 * Compute entry (and change in) size.
 	 */
-	add_entsize = dp->d_ops->sf_entsize(sfp, args->namelen);
-	incr_isize = add_entsize;
+	incr_isize = dp->d_ops->sf_entsize(sfp, args->namelen);
 	objchange = 0;
-#if XFS_BIG_INUMS
+
 	/*
 	 * Do we have to change to 8 byte inodes?
 	 */
 	if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && sfp->i8count == 0) {
 		/*
-		 * Yes, adjust the entry size and the total size.
+		 * Yes, adjust the inode size.  old count + (parent + new)
 		 */
-		add_entsize +=
-			(uint)sizeof(xfs_dir2_ino8_t) -
-			(uint)sizeof(xfs_dir2_ino4_t);
 		incr_isize +=
 			(sfp->count + 2) *
 			((uint)sizeof(xfs_dir2_ino8_t) -
 			 (uint)sizeof(xfs_dir2_ino4_t));
 		objchange = 1;
 	}
-#endif
-	old_isize = (int)dp->i_d.di_size;
-	new_isize = old_isize + incr_isize;
+
+	new_isize = (int)dp->i_d.di_size + incr_isize;
 	/*
 	 * Won't fit as shortform any more (due to size),
 	 * or the pick routine says it won't (due to offset values).
@@ -340,7 +344,7 @@ xfs_dir2_sf_addname(
 		 * Just checking or no space reservation, it doesn't fit.
 		 */
 		if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || args->total == 0)
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		/*
 		 * Convert to block form then add the name.
 		 */
@@ -365,10 +369,8 @@ xfs_dir2_sf_addname(
 	 */
 	else {
 		ASSERT(pick == 2);
-#if XFS_BIG_INUMS
 		if (objchange)
 			xfs_dir2_sf_toino8(args);
-#endif
 		xfs_dir2_sf_addname_hard(args, objchange, new_isize);
 	}
 	xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_DDATA);
@@ -420,10 +422,8 @@ xfs_dir2_sf_addname_easy(
 	 * Update the header and inode.
 	 */
 	sfp->count++;
-#if XFS_BIG_INUMS
 	if (args->inumber > XFS_DIR2_MAX_SHORT_INUM)
 		sfp->i8count++;
-#endif
 	dp->i_d.di_size = new_isize;
 	xfs_dir2_sf_check(args);
 }
@@ -511,10 +511,8 @@ xfs_dir2_sf_addname_hard(
 	dp->d_ops->sf_put_ino(sfp, sfep, args->inumber);
 	dp->d_ops->sf_put_ftype(sfep, args->filetype);
 	sfp->count++;
-#if XFS_BIG_INUMS
 	if (args->inumber > XFS_DIR2_MAX_SHORT_INUM && !objchange)
 		sfp->i8count++;
-#endif
 	/*
 	 * If there's more left to copy, do that.
 	 */
@@ -583,22 +581,17 @@ xfs_dir2_sf_addname_pick(
 	 * we'll go back, convert to block, then try the insert and convert
 	 * to leaf.
 	 */
-	if (used + (holefit ? 0 : size) > mp->m_dirblksize)
+	if (used + (holefit ? 0 : size) > args->geo->blksize)
 		return 0;
 	/*
 	 * If changing the inode number size, do it the hard way.
 	 */
-#if XFS_BIG_INUMS
-	if (objchange) {
+	if (objchange)
 		return 2;
-	}
-#else
-	ASSERT(objchange == 0);
-#endif
 	/*
 	 * If it won't fit at the end then do it the hard way (use the hole).
 	 */
-	if (used + size > mp->m_dirblksize)
+	if (used + size > args->geo->blksize)
 		return 2;
 	/*
 	 * Do it the easy way.
@@ -645,11 +638,10 @@ xfs_dir2_sf_check(
 		ASSERT(dp->d_ops->sf_get_ftype(sfep) < XFS_DIR3_FT_MAX);
 	}
 	ASSERT(i8count == sfp->i8count);
-	ASSERT(XFS_BIG_INUMS || i8count == 0);
 	ASSERT((char *)sfep - (char *)sfp == dp->i_d.di_size);
 	ASSERT(offset +
 	       (sfp->count + 2) * (uint)sizeof(xfs_dir2_leaf_entry_t) +
-	       (uint)sizeof(xfs_dir2_block_tail_t) <= mp->m_dirblksize);
+	       (uint)sizeof(xfs_dir2_block_tail_t) <= args->geo->blksize);
 }
 #endif	/* DEBUG */
 
@@ -733,7 +725,7 @@ xfs_dir2_sf_lookup(
 	 */
 	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
 		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 	ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
@@ -746,7 +738,7 @@ xfs_dir2_sf_lookup(
 		args->inumber = dp->i_ino;
 		args->cmpresult = XFS_CMP_EXACT;
 		args->filetype = XFS_DIR3_FT_DIR;
-		return XFS_ERROR(EEXIST);
+		return -EEXIST;
 	}
 	/*
 	 * Special case for ..
@@ -756,7 +748,7 @@ xfs_dir2_sf_lookup(
 		args->inumber = dp->d_ops->sf_get_parent_ino(sfp);
 		args->cmpresult = XFS_CMP_EXACT;
 		args->filetype = XFS_DIR3_FT_DIR;
-		return XFS_ERROR(EEXIST);
+		return -EEXIST;
 	}
 	/*
 	 * Loop over all the entries trying to match ours.
@@ -776,20 +768,20 @@ xfs_dir2_sf_lookup(
 			args->inumber = dp->d_ops->sf_get_ino(sfp, sfep);
 			args->filetype = dp->d_ops->sf_get_ftype(sfep);
 			if (cmp == XFS_CMP_EXACT)
-				return XFS_ERROR(EEXIST);
+				return -EEXIST;
 			ci_sfep = sfep;
 		}
 	}
 	ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
 	/*
 	 * Here, we can only be doing a lookup (not a rename or replace).
-	 * If a case-insensitive match was not found, return ENOENT.
+	 * If a case-insensitive match was not found, return -ENOENT.
 	 */
 	if (!ci_sfep)
-		return XFS_ERROR(ENOENT);
+		return -ENOENT;
 	/* otherwise process the CI match as required by the caller */
 	error = xfs_dir_cilookup_result(args, ci_sfep->name, ci_sfep->namelen);
-	return XFS_ERROR(error);
+	return error;
 }
 
 /*
@@ -819,7 +811,7 @@ xfs_dir2_sf_removename(
 	 */
 	if (oldsize < offsetof(xfs_dir2_sf_hdr_t, parent)) {
 		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 	ASSERT(dp->i_df.if_bytes == oldsize);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
@@ -842,7 +834,7 @@ xfs_dir2_sf_removename(
 	 * Didn't find it.
 	 */
 	if (i == sfp->count)
-		return XFS_ERROR(ENOENT);
+		return -ENOENT;
 	/*
 	 * Calculate sizes.
 	 */
@@ -865,7 +857,6 @@ xfs_dir2_sf_removename(
 	 */
 	xfs_idata_realloc(dp, newsize - oldsize, XFS_DATA_FORK);
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
-#if XFS_BIG_INUMS
 	/*
 	 * Are we changing inode number size?
 	 */
@@ -875,7 +866,6 @@ xfs_dir2_sf_removename(
 		else
 			sfp->i8count--;
 	}
-#endif
 	xfs_dir2_sf_check(args);
 	xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_DDATA);
 	return 0;
@@ -890,12 +880,8 @@ xfs_dir2_sf_replace(
 {
 	xfs_inode_t		*dp;		/* incore directory inode */
 	int			i;		/* entry index */
-#if XFS_BIG_INUMS || defined(DEBUG)
 	xfs_ino_t		ino=0;		/* entry old inode number */
-#endif
-#if XFS_BIG_INUMS
 	int			i8elevated;	/* sf_toino8 set i8count=1 */
-#endif
 	xfs_dir2_sf_entry_t	*sfep;		/* shortform directory entry */
 	xfs_dir2_sf_hdr_t	*sfp;		/* shortform structure */
 
@@ -909,13 +895,13 @@ xfs_dir2_sf_replace(
 	 */
 	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
 		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 	ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
 	ASSERT(dp->i_df.if_u1.if_data != NULL);
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
 	ASSERT(dp->i_d.di_size >= xfs_dir2_sf_hdr_size(sfp->i8count));
-#if XFS_BIG_INUMS
+
 	/*
 	 * New inode number is large, and need to convert to 8-byte inodes.
 	 */
@@ -946,17 +932,15 @@ xfs_dir2_sf_replace(
 		sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
 	} else
 		i8elevated = 0;
-#endif
+
 	ASSERT(args->namelen != 1 || args->name[0] != '.');
 	/*
 	 * Replace ..'s entry.
 	 */
 	if (args->namelen == 2 &&
 	    args->name[0] == '.' && args->name[1] == '.') {
-#if XFS_BIG_INUMS || defined(DEBUG)
 		ino = dp->d_ops->sf_get_parent_ino(sfp);
 		ASSERT(args->inumber != ino);
-#endif
 		dp->d_ops->sf_put_parent_ino(sfp, args->inumber);
 	}
 	/*
@@ -967,10 +951,8 @@ xfs_dir2_sf_replace(
 		     i++, sfep = dp->d_ops->sf_nextentry(sfp, sfep)) {
 			if (xfs_da_compname(args, sfep->name, sfep->namelen) ==
 								XFS_CMP_EXACT) {
-#if XFS_BIG_INUMS || defined(DEBUG)
 				ino = dp->d_ops->sf_get_ino(sfp, sfep);
 				ASSERT(args->inumber != ino);
-#endif
 				dp->d_ops->sf_put_ino(sfp, sfep, args->inumber);
 				dp->d_ops->sf_put_ftype(sfep, args->filetype);
 				break;
@@ -981,14 +963,11 @@ xfs_dir2_sf_replace(
 		 */
 		if (i == sfp->count) {
 			ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
-#if XFS_BIG_INUMS
 			if (i8elevated)
 				xfs_dir2_sf_toino4(args);
-#endif
-			return XFS_ERROR(ENOENT);
+			return -ENOENT;
 		}
 	}
-#if XFS_BIG_INUMS
 	/*
 	 * See if the old number was large, the new number is small.
 	 */
@@ -1015,13 +994,11 @@ xfs_dir2_sf_replace(
 		if (!i8elevated)
 			sfp->i8count++;
 	}
-#endif
 	xfs_dir2_sf_check(args);
 	xfs_trans_log_inode(args->trans, dp, XFS_ILOG_DDATA);
 	return 0;
 }
 
-#if XFS_BIG_INUMS
 /*
  * Convert from 8-byte inode numbers to 4-byte inode numbers.
  * The last 8-byte inode number is gone, but the count is still 1.
@@ -1100,9 +1077,9 @@ xfs_dir2_sf_toino4(
 }
 
 /*
- * Convert from 4-byte inode numbers to 8-byte inode numbers.
- * The new 8-byte inode number is not there yet, we leave with the
- * count 1 but no corresponding entry.
+ * Convert existing entries from 4-byte inode numbers to 8-byte inode numbers.
+ * The new entry w/ an 8-byte inode number is not there yet; we leave with
+ * i8count set to 1, but no corresponding 8-byte entry.
  */
 static void
 xfs_dir2_sf_toino8(
@@ -1135,7 +1112,7 @@ xfs_dir2_sf_toino8(
 	ASSERT(oldsfp->i8count == 0);
 	memcpy(buf, oldsfp, oldsize);
 	/*
-	 * Compute the new inode size.
+	 * Compute the new inode size (nb: entry count + 1 for parent)
 	 */
 	newsize =
 		oldsize +
@@ -1176,4 +1153,3 @@ xfs_dir2_sf_toino8(
 	dp->i_d.di_size = newsize;
 	xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_DDATA);
 }
-#endif	/* XFS_BIG_INUMS */
diff --git a/fs/xfs/xfs_dquot_buf.c b/fs/xfs/libxfs/xfs_dquot_buf.c
index d401457d2f25..bb969337efc8 100644
--- a/fs/xfs/xfs_dquot_buf.c
+++ b/fs/xfs/libxfs/xfs_dquot_buf.c
@@ -35,7 +35,6 @@
 
 int
 xfs_calc_dquots_per_chunk(
-	struct xfs_mount	*mp,
 	unsigned int		nbblks)	/* basic block units */
 {
 	unsigned int	ndquots;
@@ -194,7 +193,7 @@ xfs_dquot_buf_verify_crc(
 	if (mp->m_quotainfo)
 		ndquots = mp->m_quotainfo->qi_dqperchunk;
 	else
-		ndquots = xfs_calc_dquots_per_chunk(mp,
+		ndquots = xfs_calc_dquots_per_chunk(
 					XFS_BB_TO_FSB(mp, bp->b_length));
 
 	for (i = 0; i < ndquots; i++, d++) {
@@ -225,7 +224,7 @@ xfs_dquot_buf_verify(
 	if (mp->m_quotainfo)
 		ndquots = mp->m_quotainfo->qi_dqperchunk;
 	else
-		ndquots = xfs_calc_dquots_per_chunk(mp, bp->b_length);
+		ndquots = xfs_calc_dquots_per_chunk(bp->b_length);
 
 	/*
 	 * On the first read of the buffer, verify that each dquot is valid.
@@ -257,10 +256,13 @@ xfs_dquot_buf_read_verify(
 {
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 
-	if (!xfs_dquot_buf_verify_crc(mp, bp) || !xfs_dquot_buf_verify(mp, bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
-	}
+	if (!xfs_dquot_buf_verify_crc(mp, bp))
+		xfs_buf_ioerror(bp, -EFSBADCRC);
+	else if (!xfs_dquot_buf_verify(mp, bp))
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+
+	if (bp->b_error)
+		xfs_verifier_error(bp);
 }
 
 /*
@@ -275,8 +277,8 @@ xfs_dquot_buf_write_verify(
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 
 	if (!xfs_dquot_buf_verify(mp, bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
 		return;
 	}
 }
diff --git a/fs/xfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h
index b6ab5a3cfa12..7e42bba9a420 100644
--- a/fs/xfs/xfs_format.h
+++ b/fs/xfs/libxfs/xfs_format.h
@@ -68,11 +68,7 @@ struct xfs_ifork;
 #define	XFS_RTLOBIT(w)	xfs_lowbit32(w)
 #define	XFS_RTHIBIT(w)	xfs_highbit32(w)
 
-#if XFS_BIG_BLKNOS
 #define	XFS_RTBLOCKLOG(b)	xfs_highbit64(b)
-#else
-#define	XFS_RTBLOCKLOG(b)	xfs_highbit32(b)
-#endif
 
 /*
  * Dquot and dquot block format definitions
@@ -145,6 +141,8 @@ struct xfs_dsymlink_hdr {
 	__be64	sl_lsn;
 };
 
+#define XFS_SYMLINK_CRC_OFF	offsetof(struct xfs_dsymlink_hdr, sl_crc)
+
 /*
  * The maximum pathlen is 1024 bytes. Since the minimum file system
  * blocksize is 512 bytes, we can get a max of 3 extents back from
@@ -200,6 +198,8 @@ typedef __be32 xfs_alloc_ptr_t;
  */
 #define	XFS_IBT_MAGIC		0x49414254	/* 'IABT' */
 #define	XFS_IBT_CRC_MAGIC	0x49414233	/* 'IAB3' */
+#define	XFS_FIBT_MAGIC		0x46494254	/* 'FIBT' */
+#define	XFS_FIBT_CRC_MAGIC	0x46494233	/* 'FIB3' */
 
 typedef	__uint64_t	xfs_inofree_t;
 #define	XFS_INODES_PER_CHUNK		(NBBY * sizeof(xfs_inofree_t))
@@ -242,7 +242,17 @@ typedef __be32 xfs_inobt_ptr_t;
  * block numbers in the AG.
  */
 #define	XFS_IBT_BLOCK(mp)		((xfs_agblock_t)(XFS_CNT_BLOCK(mp) + 1))
-#define	XFS_PREALLOC_BLOCKS(mp)		((xfs_agblock_t)(XFS_IBT_BLOCK(mp) + 1))
+#define	XFS_FIBT_BLOCK(mp)		((xfs_agblock_t)(XFS_IBT_BLOCK(mp) + 1))
+
+/*
+ * The first data block of an AG depends on whether the filesystem was formatted
+ * with the finobt feature. If so, account for the finobt reserved root btree
+ * block.
+ */
+#define XFS_PREALLOC_BLOCKS(mp) \
+	(xfs_sb_version_hasfinobt(&((mp)->m_sb)) ? \
+	 XFS_FIBT_BLOCK(mp) + 1 : \
+	 XFS_IBT_BLOCK(mp) + 1)
 
 
 
@@ -290,23 +300,15 @@ typedef struct xfs_bmbt_rec_host {
  * Values and macros for delayed-allocation startblock fields.
  */
 #define STARTBLOCKVALBITS	17
-#define STARTBLOCKMASKBITS	(15 + XFS_BIG_BLKNOS * 20)
-#define DSTARTBLOCKMASKBITS	(15 + 20)
+#define STARTBLOCKMASKBITS	(15 + 20)
 #define STARTBLOCKMASK		\
 	(((((xfs_fsblock_t)1) << STARTBLOCKMASKBITS) - 1) << STARTBLOCKVALBITS)
-#define DSTARTBLOCKMASK		\
-	(((((xfs_dfsbno_t)1) << DSTARTBLOCKMASKBITS) - 1) << STARTBLOCKVALBITS)
 
 static inline int isnullstartblock(xfs_fsblock_t x)
 {
 	return ((x) & STARTBLOCKMASK) == STARTBLOCKMASK;
 }
 
-static inline int isnulldstartblock(xfs_dfsbno_t x)
-{
-	return ((x) & DSTARTBLOCKMASK) == DSTARTBLOCKMASK;
-}
-
 static inline xfs_fsblock_t nullstartblock(int k)
 {
 	ASSERT(k < (1 << STARTBLOCKVALBITS));
diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/libxfs/xfs_ialloc.c
index e87719c5bebe..b62771f1f4b5 100644
--- a/fs/xfs/xfs_ialloc.c
+++ b/fs/xfs/libxfs/xfs_ialloc.c
@@ -52,7 +52,7 @@ xfs_ialloc_cluster_alignment(
 {
 	if (xfs_sb_version_hasalign(&args->mp->m_sb) &&
 	    args->mp->m_sb.sb_inoalignmt >=
-	     XFS_B_TO_FSBT(args->mp, XFS_INODE_CLUSTER_SIZE(args->mp)))
+	     XFS_B_TO_FSBT(args->mp, args->mp->m_inode_cluster_size))
 		return args->mp->m_sb.sb_inoalignmt;
 	return 1;
 }
@@ -112,6 +112,66 @@ xfs_inobt_get_rec(
 }
 
 /*
+ * Insert a single inobt record. Cursor must already point to desired location.
+ */
+STATIC int
+xfs_inobt_insert_rec(
+	struct xfs_btree_cur	*cur,
+	__int32_t		freecount,
+	xfs_inofree_t		free,
+	int			*stat)
+{
+	cur->bc_rec.i.ir_freecount = freecount;
+	cur->bc_rec.i.ir_free = free;
+	return xfs_btree_insert(cur, stat);
+}
+
+/*
+ * Insert records describing a newly allocated inode chunk into the inobt.
+ */
+STATIC int
+xfs_inobt_insert(
+	struct xfs_mount	*mp,
+	struct xfs_trans	*tp,
+	struct xfs_buf		*agbp,
+	xfs_agino_t		newino,
+	xfs_agino_t		newlen,
+	xfs_btnum_t		btnum)
+{
+	struct xfs_btree_cur	*cur;
+	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agbp);
+	xfs_agnumber_t		agno = be32_to_cpu(agi->agi_seqno);
+	xfs_agino_t		thisino;
+	int			i;
+	int			error;
+
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, btnum);
+
+	for (thisino = newino;
+	     thisino < newino + newlen;
+	     thisino += XFS_INODES_PER_CHUNK) {
+		error = xfs_inobt_lookup(cur, thisino, XFS_LOOKUP_EQ, &i);
+		if (error) {
+			xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
+			return error;
+		}
+		ASSERT(i == 0);
+
+		error = xfs_inobt_insert_rec(cur, XFS_INODES_PER_CHUNK,
+					     XFS_INOBT_ALL_FREE, &i);
+		if (error) {
+			xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
+			return error;
+		}
+		ASSERT(i == 1);
+	}
+
+	xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
+
+	return 0;
+}
+
+/*
  * Verify that the number of free inodes in the AGI is correct.
  */
 #ifdef DEBUG
@@ -170,27 +230,20 @@ xfs_ialloc_inode_init(
 {
 	struct xfs_buf		*fbuf;
 	struct xfs_dinode	*free;
-	int			blks_per_cluster, nbufs, ninodes;
+	int			nbufs, blks_per_cluster, inodes_per_cluster;
 	int			version;
 	int			i, j;
 	xfs_daddr_t		d;
 	xfs_ino_t		ino = 0;
 
 	/*
-	 * Loop over the new block(s), filling in the inodes.
-	 * For small block sizes, manipulate the inodes in buffers
-	 * which are multiples of the blocks size.
+	 * Loop over the new block(s), filling in the inodes.  For small block
+	 * sizes, manipulate the inodes in buffers  which are multiples of the
+	 * blocks size.
 	 */
-	if (mp->m_sb.sb_blocksize >= XFS_INODE_CLUSTER_SIZE(mp)) {
-		blks_per_cluster = 1;
-		nbufs = length;
-		ninodes = mp->m_sb.sb_inopblock;
-	} else {
-		blks_per_cluster = XFS_INODE_CLUSTER_SIZE(mp) /
-				   mp->m_sb.sb_blocksize;
-		nbufs = length / blks_per_cluster;
-		ninodes = blks_per_cluster * mp->m_sb.sb_inopblock;
-	}
+	blks_per_cluster = xfs_icluster_size_fsb(mp);
+	inodes_per_cluster = blks_per_cluster << mp->m_sb.sb_inopblog;
+	nbufs = length / blks_per_cluster;
 
 	/*
 	 * Figure out what version number to use in the inodes we create.  If
@@ -225,12 +278,10 @@ xfs_ialloc_inode_init(
 		 * they track in the AIL as if they were physically logged.
 		 */
 		if (tp)
-			xfs_icreate_log(tp, agno, agbno, XFS_IALLOC_INODES(mp),
+			xfs_icreate_log(tp, agno, agbno, mp->m_ialloc_inos,
 					mp->m_sb.sb_inodesize, length, gen);
-	} else if (xfs_sb_version_hasnlink(&mp->m_sb))
+	} else
 		version = 2;
-	else
-		version = 1;
 
 	for (j = 0; j < nbufs; j++) {
 		/*
@@ -241,12 +292,12 @@ xfs_ialloc_inode_init(
 					 mp->m_bsize * blks_per_cluster,
 					 XBF_UNMAPPED);
 		if (!fbuf)
-			return ENOMEM;
+			return -ENOMEM;
 
 		/* Initialize the inode buffers and log them appropriately. */
 		fbuf->b_ops = &xfs_inode_buf_ops;
 		xfs_buf_zero(fbuf, 0, BBTOB(fbuf->b_length));
-		for (i = 0; i < ninodes; i++) {
+		for (i = 0; i < inodes_per_cluster; i++) {
 			int	ioffset = i << mp->m_sb.sb_inodelog;
 			uint	isize = xfs_dinode_size(version);
 
@@ -310,13 +361,10 @@ xfs_ialloc_ag_alloc(
 {
 	xfs_agi_t	*agi;		/* allocation group header */
 	xfs_alloc_arg_t	args;		/* allocation argument structure */
-	xfs_btree_cur_t	*cur;		/* inode btree cursor */
 	xfs_agnumber_t	agno;
 	int		error;
-	int		i;
 	xfs_agino_t	newino;		/* new first inode's number */
 	xfs_agino_t	newlen;		/* new number of inodes */
-	xfs_agino_t	thisino;	/* current inode number, for loop */
 	int		isaligned = 0;	/* inode allocation at stripe unit */
 					/* boundary */
 	struct xfs_perag *pag;
@@ -329,11 +377,11 @@ xfs_ialloc_ag_alloc(
 	 * Locking will ensure that we don't have two callers in here
 	 * at one time.
 	 */
-	newlen = XFS_IALLOC_INODES(args.mp);
+	newlen = args.mp->m_ialloc_inos;
 	if (args.mp->m_maxicount &&
 	    args.mp->m_sb.sb_icount + newlen > args.mp->m_maxicount)
-		return XFS_ERROR(ENOSPC);
-	args.minlen = args.maxlen = XFS_IALLOC_BLOCKS(args.mp);
+		return -ENOSPC;
+	args.minlen = args.maxlen = args.mp->m_ialloc_blks;
 	/*
 	 * First try to allocate inodes contiguous with the last-allocated
 	 * chunk of inodes.  If the filesystem is striped, this will fill
@@ -343,7 +391,7 @@ xfs_ialloc_ag_alloc(
 	newino = be32_to_cpu(agi->agi_newino);
 	agno = be32_to_cpu(agi->agi_seqno);
 	args.agbno = XFS_AGINO_TO_AGBNO(args.mp, newino) +
-			XFS_IALLOC_BLOCKS(args.mp);
+		     args.mp->m_ialloc_blks;
 	if (likely(newino != NULLAGINO &&
 		  (args.agbno < be32_to_cpu(agi->agi_length)))) {
 		args.fsbno = XFS_AGB_TO_FSB(args.mp, agno, args.agbno);
@@ -370,6 +418,18 @@ xfs_ialloc_ag_alloc(
 		args.minleft = args.mp->m_in_maxlevels - 1;
 		if ((error = xfs_alloc_vextent(&args)))
 			return error;
+
+		/*
+		 * This request might have dirtied the transaction if the AG can
+		 * satisfy the request, but the exact block was not available.
+		 * If the allocation did fail, subsequent requests will relax
+		 * the exact agbno requirement and increase the alignment
+		 * instead. It is critical that the total size of the request
+		 * (len + alignment + slop) does not increase from this point
+		 * on, so reset minalignslop to ensure it is not included in
+		 * subsequent requests.
+		 */
+		args.minalignslop = 0;
 	} else
 		args.fsbno = NULLFSBLOCK;
 
@@ -454,29 +514,19 @@ xfs_ialloc_ag_alloc(
 	agi->agi_newino = cpu_to_be32(newino);
 
 	/*
-	 * Insert records describing the new inode chunk into the btree.
+	 * Insert records describing the new inode chunk into the btrees.
 	 */
-	cur = xfs_inobt_init_cursor(args.mp, tp, agbp, agno);
-	for (thisino = newino;
-	     thisino < newino + newlen;
-	     thisino += XFS_INODES_PER_CHUNK) {
-		cur->bc_rec.i.ir_startino = thisino;
-		cur->bc_rec.i.ir_freecount = XFS_INODES_PER_CHUNK;
-		cur->bc_rec.i.ir_free = XFS_INOBT_ALL_FREE;
-		error = xfs_btree_lookup(cur, XFS_LOOKUP_EQ, &i);
-		if (error) {
-			xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
-			return error;
-		}
-		ASSERT(i == 0);
-		error = xfs_btree_insert(cur, &i);
-		if (error) {
-			xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
+	error = xfs_inobt_insert(args.mp, tp, agbp, newino, newlen,
+				 XFS_BTNUM_INO);
+	if (error)
+		return error;
+
+	if (xfs_sb_version_hasfinobt(&args.mp->m_sb)) {
+		error = xfs_inobt_insert(args.mp, tp, agbp, newino, newlen,
+					 XFS_BTNUM_FINO);
+		if (error)
 			return error;
-		}
-		ASSERT(i == 1);
 	}
-	xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
 	/*
 	 * Log allocation group header fields
 	 */
@@ -585,7 +635,7 @@ xfs_ialloc_ag_select(
 		 * Is there enough free space for the file plus a block of
 		 * inodes? (if we need to allocate some)?
 		 */
-		ineed = XFS_IALLOC_BLOCKS(mp);
+		ineed = mp->m_ialloc_blks;
 		longest = pag->pagf_longest;
 		if (!longest)
 			longest = pag->pagf_flcount > 0;
@@ -670,13 +720,10 @@ xfs_ialloc_get_rec(
 }
 
 /*
- * Allocate an inode.
- *
- * The caller selected an AG for us, and made sure that free inodes are
- * available.
+ * Allocate an inode using the inobt-only algorithm.
  */
 STATIC int
-xfs_dialloc_ag(
+xfs_dialloc_ag_inobt(
 	struct xfs_trans	*tp,
 	struct xfs_buf		*agbp,
 	xfs_ino_t		parent,
@@ -702,7 +749,7 @@ xfs_dialloc_ag(
 	ASSERT(pag->pagi_freecount > 0);
 
  restart_pagno:
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
 	/*
 	 * If pagino is 0 (this is the root inode allocation) use newino.
 	 * This must work because we've just allocated some.
@@ -935,6 +982,294 @@ error0:
 }
 
 /*
+ * Use the free inode btree to allocate an inode based on distance from the
+ * parent. Note that the provided cursor may be deleted and replaced.
+ */
+STATIC int
+xfs_dialloc_ag_finobt_near(
+	xfs_agino_t			pagino,
+	struct xfs_btree_cur		**ocur,
+	struct xfs_inobt_rec_incore	*rec)
+{
+	struct xfs_btree_cur		*lcur = *ocur;	/* left search cursor */
+	struct xfs_btree_cur		*rcur;	/* right search cursor */
+	struct xfs_inobt_rec_incore	rrec;
+	int				error;
+	int				i, j;
+
+	error = xfs_inobt_lookup(lcur, pagino, XFS_LOOKUP_LE, &i);
+	if (error)
+		return error;
+
+	if (i == 1) {
+		error = xfs_inobt_get_rec(lcur, rec, &i);
+		if (error)
+			return error;
+		XFS_WANT_CORRUPTED_RETURN(i == 1);
+
+		/*
+		 * See if we've landed in the parent inode record. The finobt
+		 * only tracks chunks with at least one free inode, so record
+		 * existence is enough.
+		 */
+		if (pagino >= rec->ir_startino &&
+		    pagino < (rec->ir_startino + XFS_INODES_PER_CHUNK))
+			return 0;
+	}
+
+	error = xfs_btree_dup_cursor(lcur, &rcur);
+	if (error)
+		return error;
+
+	error = xfs_inobt_lookup(rcur, pagino, XFS_LOOKUP_GE, &j);
+	if (error)
+		goto error_rcur;
+	if (j == 1) {
+		error = xfs_inobt_get_rec(rcur, &rrec, &j);
+		if (error)
+			goto error_rcur;
+		XFS_WANT_CORRUPTED_GOTO(j == 1, error_rcur);
+	}
+
+	XFS_WANT_CORRUPTED_GOTO(i == 1 || j == 1, error_rcur);
+	if (i == 1 && j == 1) {
+		/*
+		 * Both the left and right records are valid. Choose the closer
+		 * inode chunk to the target.
+		 */
+		if ((pagino - rec->ir_startino + XFS_INODES_PER_CHUNK - 1) >
+		    (rrec.ir_startino - pagino)) {
+			*rec = rrec;
+			xfs_btree_del_cursor(lcur, XFS_BTREE_NOERROR);
+			*ocur = rcur;
+		} else {
+			xfs_btree_del_cursor(rcur, XFS_BTREE_NOERROR);
+		}
+	} else if (j == 1) {
+		/* only the right record is valid */
+		*rec = rrec;
+		xfs_btree_del_cursor(lcur, XFS_BTREE_NOERROR);
+		*ocur = rcur;
+	} else if (i == 1) {
+		/* only the left record is valid */
+		xfs_btree_del_cursor(rcur, XFS_BTREE_NOERROR);
+	}
+
+	return 0;
+
+error_rcur:
+	xfs_btree_del_cursor(rcur, XFS_BTREE_ERROR);
+	return error;
+}
+
+/*
+ * Use the free inode btree to find a free inode based on a newino hint. If
+ * the hint is NULL, find the first free inode in the AG.
+ */
+STATIC int
+xfs_dialloc_ag_finobt_newino(
+	struct xfs_agi			*agi,
+	struct xfs_btree_cur		*cur,
+	struct xfs_inobt_rec_incore	*rec)
+{
+	int error;
+	int i;
+
+	if (agi->agi_newino != cpu_to_be32(NULLAGINO)) {
+		error = xfs_inobt_lookup(cur, agi->agi_newino, XFS_LOOKUP_EQ,
+					 &i);
+		if (error)
+			return error;
+		if (i == 1) {
+			error = xfs_inobt_get_rec(cur, rec, &i);
+			if (error)
+				return error;
+			XFS_WANT_CORRUPTED_RETURN(i == 1);
+
+			return 0;
+		}
+	}
+
+	/*
+	 * Find the first inode available in the AG.
+	 */
+	error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &i);
+	if (error)
+		return error;
+	XFS_WANT_CORRUPTED_RETURN(i == 1);
+
+	error = xfs_inobt_get_rec(cur, rec, &i);
+	if (error)
+		return error;
+	XFS_WANT_CORRUPTED_RETURN(i == 1);
+
+	return 0;
+}
+
+/*
+ * Update the inobt based on a modification made to the finobt. Also ensure that
+ * the records from both trees are equivalent post-modification.
+ */
+STATIC int
+xfs_dialloc_ag_update_inobt(
+	struct xfs_btree_cur		*cur,	/* inobt cursor */
+	struct xfs_inobt_rec_incore	*frec,	/* finobt record */
+	int				offset) /* inode offset */
+{
+	struct xfs_inobt_rec_incore	rec;
+	int				error;
+	int				i;
+
+	error = xfs_inobt_lookup(cur, frec->ir_startino, XFS_LOOKUP_EQ, &i);
+	if (error)
+		return error;
+	XFS_WANT_CORRUPTED_RETURN(i == 1);
+
+	error = xfs_inobt_get_rec(cur, &rec, &i);
+	if (error)
+		return error;
+	XFS_WANT_CORRUPTED_RETURN(i == 1);
+	ASSERT((XFS_AGINO_TO_OFFSET(cur->bc_mp, rec.ir_startino) %
+				   XFS_INODES_PER_CHUNK) == 0);
+
+	rec.ir_free &= ~XFS_INOBT_MASK(offset);
+	rec.ir_freecount--;
+
+	XFS_WANT_CORRUPTED_RETURN((rec.ir_free == frec->ir_free) &&
+				  (rec.ir_freecount == frec->ir_freecount));
+
+	error = xfs_inobt_update(cur, &rec);
+	if (error)
+		return error;
+
+	return 0;
+}
+
+/*
+ * Allocate an inode using the free inode btree, if available. Otherwise, fall
+ * back to the inobt search algorithm.
+ *
+ * The caller selected an AG for us, and made sure that free inodes are
+ * available.
+ */
+STATIC int
+xfs_dialloc_ag(
+	struct xfs_trans	*tp,
+	struct xfs_buf		*agbp,
+	xfs_ino_t		parent,
+	xfs_ino_t		*inop)
+{
+	struct xfs_mount		*mp = tp->t_mountp;
+	struct xfs_agi			*agi = XFS_BUF_TO_AGI(agbp);
+	xfs_agnumber_t			agno = be32_to_cpu(agi->agi_seqno);
+	xfs_agnumber_t			pagno = XFS_INO_TO_AGNO(mp, parent);
+	xfs_agino_t			pagino = XFS_INO_TO_AGINO(mp, parent);
+	struct xfs_perag		*pag;
+	struct xfs_btree_cur		*cur;	/* finobt cursor */
+	struct xfs_btree_cur		*icur;	/* inobt cursor */
+	struct xfs_inobt_rec_incore	rec;
+	xfs_ino_t			ino;
+	int				error;
+	int				offset;
+	int				i;
+
+	if (!xfs_sb_version_hasfinobt(&mp->m_sb))
+		return xfs_dialloc_ag_inobt(tp, agbp, parent, inop);
+
+	pag = xfs_perag_get(mp, agno);
+
+	/*
+	 * If pagino is 0 (this is the root inode allocation) use newino.
+	 * This must work because we've just allocated some.
+	 */
+	if (!pagino)
+		pagino = be32_to_cpu(agi->agi_newino);
+
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_FINO);
+
+	error = xfs_check_agi_freecount(cur, agi);
+	if (error)
+		goto error_cur;
+
+	/*
+	 * The search algorithm depends on whether we're in the same AG as the
+	 * parent. If so, find the closest available inode to the parent. If
+	 * not, consider the agi hint or find the first free inode in the AG.
+	 */
+	if (agno == pagno)
+		error = xfs_dialloc_ag_finobt_near(pagino, &cur, &rec);
+	else
+		error = xfs_dialloc_ag_finobt_newino(agi, cur, &rec);
+	if (error)
+		goto error_cur;
+
+	offset = xfs_lowbit64(rec.ir_free);
+	ASSERT(offset >= 0);
+	ASSERT(offset < XFS_INODES_PER_CHUNK);
+	ASSERT((XFS_AGINO_TO_OFFSET(mp, rec.ir_startino) %
+				   XFS_INODES_PER_CHUNK) == 0);
+	ino = XFS_AGINO_TO_INO(mp, agno, rec.ir_startino + offset);
+
+	/*
+	 * Modify or remove the finobt record.
+	 */
+	rec.ir_free &= ~XFS_INOBT_MASK(offset);
+	rec.ir_freecount--;
+	if (rec.ir_freecount)
+		error = xfs_inobt_update(cur, &rec);
+	else
+		error = xfs_btree_delete(cur, &i);
+	if (error)
+		goto error_cur;
+
+	/*
+	 * The finobt has now been updated appropriately. We haven't updated the
+	 * agi and superblock yet, so we can create an inobt cursor and validate
+	 * the original freecount. If all is well, make the equivalent update to
+	 * the inobt using the finobt record and offset information.
+	 */
+	icur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
+
+	error = xfs_check_agi_freecount(icur, agi);
+	if (error)
+		goto error_icur;
+
+	error = xfs_dialloc_ag_update_inobt(icur, &rec, offset);
+	if (error)
+		goto error_icur;
+
+	/*
+	 * Both trees have now been updated. We must update the perag and
+	 * superblock before we can check the freecount for each btree.
+	 */
+	be32_add_cpu(&agi->agi_freecount, -1);
+	xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREECOUNT);
+	pag->pagi_freecount--;
+
+	xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, -1);
+
+	error = xfs_check_agi_freecount(icur, agi);
+	if (error)
+		goto error_icur;
+	error = xfs_check_agi_freecount(cur, agi);
+	if (error)
+		goto error_icur;
+
+	xfs_btree_del_cursor(icur, XFS_BTREE_NOERROR);
+	xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
+	xfs_perag_put(pag);
+	*inop = ino;
+	return 0;
+
+error_icur:
+	xfs_btree_del_cursor(icur, XFS_BTREE_ERROR);
+error_cur:
+	xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
+	xfs_perag_put(pag);
+	return error;
+}
+
+/*
  * Allocate an inode on disk.
  *
  * Mode is used to tell whether the new inode will need space, and whether it
@@ -999,7 +1334,7 @@ xfs_dialloc(
 	 * inode.
 	 */
 	if (mp->m_maxicount &&
-	    mp->m_sb.sb_icount + XFS_IALLOC_INODES(mp) > mp->m_maxicount) {
+	    mp->m_sb.sb_icount + mp->m_ialloc_inos > mp->m_maxicount) {
 		noroom = 1;
 		okalloc = 0;
 	}
@@ -1050,7 +1385,7 @@ xfs_dialloc(
 		if (error) {
 			xfs_trans_brelse(tp, agbp);
 
-			if (error != ENOSPC)
+			if (error != -ENOSPC)
 				goto out_error;
 
 			xfs_perag_put(pag);
@@ -1081,7 +1416,7 @@ nextag:
 			agno = 0;
 		if (agno == start_agno) {
 			*inop = NULLFSINO;
-			return noroom ? ENOSPC : 0;
+			return noroom ? -ENOSPC : 0;
 		}
 	}
 
@@ -1090,81 +1425,37 @@ out_alloc:
 	return xfs_dialloc_ag(tp, agbp, parent, inop);
 out_error:
 	xfs_perag_put(pag);
-	return XFS_ERROR(error);
+	return error;
 }
 
-/*
- * Free disk inode.  Carefully avoids touching the incore inode, all
- * manipulations incore are the caller's responsibility.
- * The on-disk inode is not changed by this operation, only the
- * btree (free inode mask) is changed.
- */
-int
-xfs_difree(
-	xfs_trans_t	*tp,		/* transaction pointer */
-	xfs_ino_t	inode,		/* inode to be freed */
-	xfs_bmap_free_t	*flist,		/* extents to free */
-	int		*delete,	/* set if inode cluster was deleted */
-	xfs_ino_t	*first_ino)	/* first inode in deleted cluster */
+STATIC int
+xfs_difree_inobt(
+	struct xfs_mount		*mp,
+	struct xfs_trans		*tp,
+	struct xfs_buf			*agbp,
+	xfs_agino_t			agino,
+	struct xfs_bmap_free		*flist,
+	int				*deleted,
+	xfs_ino_t			*first_ino,
+	struct xfs_inobt_rec_incore	*orec)
 {
-	/* REFERENCED */
-	xfs_agblock_t	agbno;	/* block number containing inode */
-	xfs_buf_t	*agbp;	/* buffer containing allocation group header */
-	xfs_agino_t	agino;	/* inode number relative to allocation group */
-	xfs_agnumber_t	agno;	/* allocation group number */
-	xfs_agi_t	*agi;	/* allocation group header */
-	xfs_btree_cur_t	*cur;	/* inode btree cursor */
-	int		error;	/* error return value */
-	int		i;	/* result code */
-	int		ilen;	/* inodes in an inode cluster */
-	xfs_mount_t	*mp;	/* mount structure for filesystem */
-	int		off;	/* offset of inode in inode chunk */
-	xfs_inobt_rec_incore_t rec;	/* btree record */
-	struct xfs_perag *pag;
-
-	mp = tp->t_mountp;
+	struct xfs_agi			*agi = XFS_BUF_TO_AGI(agbp);
+	xfs_agnumber_t			agno = be32_to_cpu(agi->agi_seqno);
+	struct xfs_perag		*pag;
+	struct xfs_btree_cur		*cur;
+	struct xfs_inobt_rec_incore	rec;
+	int				ilen;
+	int				error;
+	int				i;
+	int				off;
 
-	/*
-	 * Break up inode number into its components.
-	 */
-	agno = XFS_INO_TO_AGNO(mp, inode);
-	if (agno >= mp->m_sb.sb_agcount)  {
-		xfs_warn(mp, "%s: agno >= mp->m_sb.sb_agcount (%d >= %d).",
-			__func__, agno, mp->m_sb.sb_agcount);
-		ASSERT(0);
-		return XFS_ERROR(EINVAL);
-	}
-	agino = XFS_INO_TO_AGINO(mp, inode);
-	if (inode != XFS_AGINO_TO_INO(mp, agno, agino))  {
-		xfs_warn(mp, "%s: inode != XFS_AGINO_TO_INO() (%llu != %llu).",
-			__func__, (unsigned long long)inode,
-			(unsigned long long)XFS_AGINO_TO_INO(mp, agno, agino));
-		ASSERT(0);
-		return XFS_ERROR(EINVAL);
-	}
-	agbno = XFS_AGINO_TO_AGBNO(mp, agino);
-	if (agbno >= mp->m_sb.sb_agblocks)  {
-		xfs_warn(mp, "%s: agbno >= mp->m_sb.sb_agblocks (%d >= %d).",
-			__func__, agbno, mp->m_sb.sb_agblocks);
-		ASSERT(0);
-		return XFS_ERROR(EINVAL);
-	}
-	/*
-	 * Get the allocation group header.
-	 */
-	error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
-	if (error) {
-		xfs_warn(mp, "%s: xfs_ialloc_read_agi() returned error %d.",
-			__func__, error);
-		return error;
-	}
-	agi = XFS_BUF_TO_AGI(agbp);
 	ASSERT(agi->agi_magicnum == cpu_to_be32(XFS_AGI_MAGIC));
-	ASSERT(agbno < be32_to_cpu(agi->agi_length));
+	ASSERT(XFS_AGINO_TO_AGBNO(mp, agino) < be32_to_cpu(agi->agi_length));
+
 	/*
 	 * Initialize the cursor.
 	 */
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
 
 	error = xfs_check_agi_freecount(cur, agi);
 	if (error)
@@ -1202,9 +1493,9 @@ xfs_difree(
 	 * When an inode cluster is free, it becomes eligible for removal
 	 */
 	if (!(mp->m_flags & XFS_MOUNT_IKEEP) &&
-	    (rec.ir_freecount == XFS_IALLOC_INODES(mp))) {
+	    (rec.ir_freecount == mp->m_ialloc_inos)) {
 
-		*delete = 1;
+		*deleted = 1;
 		*first_ino = XFS_AGINO_TO_INO(mp, agno, rec.ir_startino);
 
 		/*
@@ -1212,7 +1503,7 @@ xfs_difree(
 		 * AGI and Superblock inode counts, and mark the disk space
 		 * to be freed when the transaction is committed.
 		 */
-		ilen = XFS_IALLOC_INODES(mp);
+		ilen = mp->m_ialloc_inos;
 		be32_add_cpu(&agi->agi_count, -ilen);
 		be32_add_cpu(&agi->agi_freecount, -(ilen - 1));
 		xfs_ialloc_log_agi(tp, agbp, XFS_AGI_COUNT | XFS_AGI_FREECOUNT);
@@ -1228,11 +1519,11 @@ xfs_difree(
 			goto error0;
 		}
 
-		xfs_bmap_add_free(XFS_AGB_TO_FSB(mp,
-				agno, XFS_INO_TO_AGBNO(mp,rec.ir_startino)),
-				XFS_IALLOC_BLOCKS(mp), flist, mp);
+		xfs_bmap_add_free(XFS_AGB_TO_FSB(mp, agno,
+				  XFS_AGINO_TO_AGBNO(mp, rec.ir_startino)),
+				  mp->m_ialloc_blks, flist, mp);
 	} else {
-		*delete = 0;
+		*deleted = 0;
 
 		error = xfs_inobt_update(cur, &rec);
 		if (error) {
@@ -1256,6 +1547,7 @@ xfs_difree(
 	if (error)
 		goto error0;
 
+	*orec = rec;
 	xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
 	return 0;
 
@@ -1264,6 +1556,182 @@ error0:
 	return error;
 }
 
+/*
+ * Free an inode in the free inode btree.
+ */
+STATIC int
+xfs_difree_finobt(
+	struct xfs_mount		*mp,
+	struct xfs_trans		*tp,
+	struct xfs_buf			*agbp,
+	xfs_agino_t			agino,
+	struct xfs_inobt_rec_incore	*ibtrec) /* inobt record */
+{
+	struct xfs_agi			*agi = XFS_BUF_TO_AGI(agbp);
+	xfs_agnumber_t			agno = be32_to_cpu(agi->agi_seqno);
+	struct xfs_btree_cur		*cur;
+	struct xfs_inobt_rec_incore	rec;
+	int				offset = agino - ibtrec->ir_startino;
+	int				error;
+	int				i;
+
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_FINO);
+
+	error = xfs_inobt_lookup(cur, ibtrec->ir_startino, XFS_LOOKUP_EQ, &i);
+	if (error)
+		goto error;
+	if (i == 0) {
+		/*
+		 * If the record does not exist in the finobt, we must have just
+		 * freed an inode in a previously fully allocated chunk. If not,
+		 * something is out of sync.
+		 */
+		XFS_WANT_CORRUPTED_GOTO(ibtrec->ir_freecount == 1, error);
+
+		error = xfs_inobt_insert_rec(cur, ibtrec->ir_freecount,
+					     ibtrec->ir_free, &i);
+		if (error)
+			goto error;
+		ASSERT(i == 1);
+
+		goto out;
+	}
+
+	/*
+	 * Read and update the existing record. We could just copy the ibtrec
+	 * across here, but that would defeat the purpose of having redundant
+	 * metadata. By making the modifications independently, we can catch
+	 * corruptions that we wouldn't see if we just copied from one record
+	 * to another.
+	 */
+	error = xfs_inobt_get_rec(cur, &rec, &i);
+	if (error)
+		goto error;
+	XFS_WANT_CORRUPTED_GOTO(i == 1, error);
+
+	rec.ir_free |= XFS_INOBT_MASK(offset);
+	rec.ir_freecount++;
+
+	XFS_WANT_CORRUPTED_GOTO((rec.ir_free == ibtrec->ir_free) &&
+				(rec.ir_freecount == ibtrec->ir_freecount),
+				error);
+
+	/*
+	 * The content of inobt records should always match between the inobt
+	 * and finobt. The lifecycle of records in the finobt is different from
+	 * the inobt in that the finobt only tracks records with at least one
+	 * free inode. Hence, if all of the inodes are free and we aren't
+	 * keeping inode chunks permanently on disk, remove the record.
+	 * Otherwise, update the record with the new information.
+	 */
+	if (rec.ir_freecount == mp->m_ialloc_inos &&
+	    !(mp->m_flags & XFS_MOUNT_IKEEP)) {
+		error = xfs_btree_delete(cur, &i);
+		if (error)
+			goto error;
+		ASSERT(i == 1);
+	} else {
+		error = xfs_inobt_update(cur, &rec);
+		if (error)
+			goto error;
+	}
+
+out:
+	error = xfs_check_agi_freecount(cur, agi);
+	if (error)
+		goto error;
+
+	xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
+	return 0;
+
+error:
+	xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
+	return error;
+}
+
+/*
+ * Free disk inode.  Carefully avoids touching the incore inode, all
+ * manipulations incore are the caller's responsibility.
+ * The on-disk inode is not changed by this operation, only the
+ * btree (free inode mask) is changed.
+ */
+int
+xfs_difree(
+	struct xfs_trans	*tp,		/* transaction pointer */
+	xfs_ino_t		inode,		/* inode to be freed */
+	struct xfs_bmap_free	*flist,		/* extents to free */
+	int			*deleted,/* set if inode cluster was deleted */
+	xfs_ino_t		*first_ino)/* first inode in deleted cluster */
+{
+	/* REFERENCED */
+	xfs_agblock_t		agbno;	/* block number containing inode */
+	struct xfs_buf		*agbp;	/* buffer for allocation group header */
+	xfs_agino_t		agino;	/* allocation group inode number */
+	xfs_agnumber_t		agno;	/* allocation group number */
+	int			error;	/* error return value */
+	struct xfs_mount	*mp;	/* mount structure for filesystem */
+	struct xfs_inobt_rec_incore rec;/* btree record */
+
+	mp = tp->t_mountp;
+
+	/*
+	 * Break up inode number into its components.
+	 */
+	agno = XFS_INO_TO_AGNO(mp, inode);
+	if (agno >= mp->m_sb.sb_agcount)  {
+		xfs_warn(mp, "%s: agno >= mp->m_sb.sb_agcount (%d >= %d).",
+			__func__, agno, mp->m_sb.sb_agcount);
+		ASSERT(0);
+		return -EINVAL;
+	}
+	agino = XFS_INO_TO_AGINO(mp, inode);
+	if (inode != XFS_AGINO_TO_INO(mp, agno, agino))  {
+		xfs_warn(mp, "%s: inode != XFS_AGINO_TO_INO() (%llu != %llu).",
+			__func__, (unsigned long long)inode,
+			(unsigned long long)XFS_AGINO_TO_INO(mp, agno, agino));
+		ASSERT(0);
+		return -EINVAL;
+	}
+	agbno = XFS_AGINO_TO_AGBNO(mp, agino);
+	if (agbno >= mp->m_sb.sb_agblocks)  {
+		xfs_warn(mp, "%s: agbno >= mp->m_sb.sb_agblocks (%d >= %d).",
+			__func__, agbno, mp->m_sb.sb_agblocks);
+		ASSERT(0);
+		return -EINVAL;
+	}
+	/*
+	 * Get the allocation group header.
+	 */
+	error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
+	if (error) {
+		xfs_warn(mp, "%s: xfs_ialloc_read_agi() returned error %d.",
+			__func__, error);
+		return error;
+	}
+
+	/*
+	 * Fix up the inode allocation btree.
+	 */
+	error = xfs_difree_inobt(mp, tp, agbp, agino, flist, deleted, first_ino,
+				 &rec);
+	if (error)
+		goto error0;
+
+	/*
+	 * Fix up the free inode btree.
+	 */
+	if (xfs_sb_version_hasfinobt(&mp->m_sb)) {
+		error = xfs_difree_finobt(mp, tp, agbp, agino, &rec);
+		if (error)
+			goto error0;
+	}
+
+	return 0;
+
+error0:
+	return error;
+}
+
 STATIC int
 xfs_imap_lookup(
 	struct xfs_mount	*mp,
@@ -1295,13 +1763,13 @@ xfs_imap_lookup(
 	 * we have a record, we need to ensure it contains the inode number
 	 * we are looking up.
 	 */
-	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno);
+	cur = xfs_inobt_init_cursor(mp, tp, agbp, agno, XFS_BTNUM_INO);
 	error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, &i);
 	if (!error) {
 		if (i)
 			error = xfs_inobt_get_rec(cur, &rec, &i);
 		if (!error && i == 0)
-			error = EINVAL;
+			error = -EINVAL;
 	}
 
 	xfs_trans_brelse(tp, agbp);
@@ -1311,13 +1779,13 @@ xfs_imap_lookup(
 
 	/* check that the returned record contains the required inode */
 	if (rec.ir_startino > agino ||
-	    rec.ir_startino + XFS_IALLOC_INODES(mp) <= agino)
-		return EINVAL;
+	    rec.ir_startino + mp->m_ialloc_inos <= agino)
+		return -EINVAL;
 
 	/* for untrusted inodes check it is allocated first */
 	if ((flags & XFS_IGET_UNTRUSTED) &&
 	    (rec.ir_free & XFS_INOBT_MASK(agino - rec.ir_startino)))
-		return EINVAL;
+		return -EINVAL;
 
 	*chunk_agbno = XFS_AGINO_TO_AGBNO(mp, rec.ir_startino);
 	*offset_agbno = agbno - *chunk_agbno;
@@ -1361,7 +1829,7 @@ xfs_imap(
 		 * as they can be invalid without implying corruption.
 		 */
 		if (flags & XFS_IGET_UNTRUSTED)
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		if (agno >= mp->m_sb.sb_agcount) {
 			xfs_alert(mp,
 				"%s: agno (%d) >= mp->m_sb.sb_agcount (%d)",
@@ -1381,10 +1849,10 @@ xfs_imap(
 		}
 		xfs_stack_trace();
 #endif /* DEBUG */
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
-	blks_per_cluster = XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_blocklog;
+	blks_per_cluster = xfs_icluster_size_fsb(mp);
 
 	/*
 	 * For bulkstat and handle lookups, we have an untrusted inode number
@@ -1405,7 +1873,7 @@ xfs_imap(
 	 * If the inode cluster size is the same as the blocksize or
 	 * smaller we get to the buffer by simple arithmetics.
 	 */
-	if (XFS_INODE_CLUSTER_SIZE(mp) <= mp->m_sb.sb_blocksize) {
+	if (blks_per_cluster == 1) {
 		offset = XFS_INO_TO_OFFSET(mp, ino);
 		ASSERT(offset < mp->m_sb.sb_inopblock);
 
@@ -1454,7 +1922,7 @@ out_map:
 			__func__, (unsigned long long) imap->im_blkno,
 			(unsigned long long) imap->im_len,
 			XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks));
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 	return 0;
 }
@@ -1483,7 +1951,16 @@ xfs_ialloc_compute_maxlevels(
 }
 
 /*
- * Log specified fields for the ag hdr (inode section)
+ * Log specified fields for the ag hdr (inode section). The growth of the agi
+ * structure over time requires that we interpret the buffer as two logical
+ * regions delineated by the end of the unlinked list. This is due to the size
+ * of the hash table and its location in the middle of the agi.
+ *
+ * For example, a request to log a field before agi_unlinked and a field after
+ * agi_unlinked could cause us to log the entire hash table and use an excessive
+ * amount of log space. To avoid this behavior, log the region up through
+ * agi_unlinked in one call and the region after agi_unlinked through the end of
+ * the structure in another.
  */
 void
 xfs_ialloc_log_agi(
@@ -1506,6 +1983,8 @@ xfs_ialloc_log_agi(
 		offsetof(xfs_agi_t, agi_newino),
 		offsetof(xfs_agi_t, agi_dirino),
 		offsetof(xfs_agi_t, agi_unlinked),
+		offsetof(xfs_agi_t, agi_free_root),
+		offsetof(xfs_agi_t, agi_free_level),
 		sizeof(xfs_agi_t)
 	};
 #ifdef DEBUG
@@ -1514,15 +1993,30 @@ xfs_ialloc_log_agi(
 	agi = XFS_BUF_TO_AGI(bp);
 	ASSERT(agi->agi_magicnum == cpu_to_be32(XFS_AGI_MAGIC));
 #endif
+
+	xfs_trans_buf_set_type(tp, bp, XFS_BLFT_AGI_BUF);
+
 	/*
-	 * Compute byte offsets for the first and last fields.
+	 * Compute byte offsets for the first and last fields in the first
+	 * region and log the agi buffer. This only logs up through
+	 * agi_unlinked.
 	 */
-	xfs_btree_offsets(fields, offsets, XFS_AGI_NUM_BITS, &first, &last);
+	if (fields & XFS_AGI_ALL_BITS_R1) {
+		xfs_btree_offsets(fields, offsets, XFS_AGI_NUM_BITS_R1,
+				  &first, &last);
+		xfs_trans_log_buf(tp, bp, first, last);
+	}
+
 	/*
-	 * Log the allocation group inode header buffer.
+	 * Mask off the bits in the first region and calculate the first and
+	 * last field offsets for any bits in the second region.
 	 */
-	xfs_trans_buf_set_type(tp, bp, XFS_BLFT_AGI_BUF);
-	xfs_trans_log_buf(tp, bp, first, last);
+	fields &= ~XFS_AGI_ALL_BITS_R1;
+	if (fields) {
+		xfs_btree_offsets(fields, offsets, XFS_AGI_NUM_BITS_R2,
+				  &first, &last);
+		xfs_trans_log_buf(tp, bp, first, last);
+	}
 }
 
 #ifdef DEBUG
@@ -1575,18 +2069,17 @@ xfs_agi_read_verify(
 	struct xfs_buf	*bp)
 {
 	struct xfs_mount *mp = bp->b_target->bt_mount;
-	int		agi_ok = 1;
 
-	if (xfs_sb_version_hascrc(&mp->m_sb))
-		agi_ok = xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
-					  offsetof(struct xfs_agi, agi_crc));
-	agi_ok = agi_ok && xfs_agi_verify(bp);
-
-	if (unlikely(XFS_TEST_ERROR(!agi_ok, mp, XFS_ERRTAG_IALLOC_READ_AGI,
-			XFS_RANDOM_IALLOC_READ_AGI))) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
-	}
+	if (xfs_sb_version_hascrc(&mp->m_sb) &&
+	    !xfs_buf_verify_cksum(bp, XFS_AGI_CRC_OFF))
+		xfs_buf_ioerror(bp, -EFSBADCRC);
+	else if (XFS_TEST_ERROR(!xfs_agi_verify(bp), mp,
+				XFS_ERRTAG_IALLOC_READ_AGI,
+				XFS_RANDOM_IALLOC_READ_AGI))
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+
+	if (bp->b_error)
+		xfs_verifier_error(bp);
 }
 
 static void
@@ -1597,8 +2090,8 @@ xfs_agi_write_verify(
 	struct xfs_buf_log_item	*bip = bp->b_fspriv;
 
 	if (!xfs_agi_verify(bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
 		return;
 	}
 
@@ -1607,8 +2100,7 @@ xfs_agi_write_verify(
 
 	if (bip)
 		XFS_BUF_TO_AGI(bp)->agi_lsn = cpu_to_be64(bip->bli_item.li_lsn);
-	xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
-			 offsetof(struct xfs_agi, agi_crc));
+	xfs_buf_update_cksum(bp, XFS_AGI_CRC_OFF);
 }
 
 const struct xfs_buf_ops xfs_agi_buf_ops = {
@@ -1637,7 +2129,6 @@ xfs_read_agi(
 	if (error)
 		return error;
 
-	ASSERT(!xfs_buf_geterror(*bpp));
 	xfs_buf_set_ref(*bpp, XFS_AGI_REF);
 	return 0;
 }
diff --git a/fs/xfs/xfs_ialloc.h b/fs/xfs/libxfs/xfs_ialloc.h
index a8f76a5ff418..95ad1c002d60 100644
--- a/fs/xfs/xfs_ialloc.h
+++ b/fs/xfs/libxfs/xfs_ialloc.h
@@ -25,17 +25,18 @@ struct xfs_mount;
 struct xfs_trans;
 struct xfs_btree_cur;
 
-/*
- * Allocation parameters for inode allocation.
- */
-#define	XFS_IALLOC_INODES(mp)	(mp)->m_ialloc_inos
-#define	XFS_IALLOC_BLOCKS(mp)	(mp)->m_ialloc_blks
-
-/*
- * Move inodes in clusters of this size.
- */
+/* Move inodes in clusters of this size */
 #define	XFS_INODE_BIG_CLUSTER_SIZE	8192
-#define	XFS_INODE_CLUSTER_SIZE(mp)	(mp)->m_inode_cluster_size
+
+/* Calculate and return the number of filesystem blocks per inode cluster */
+static inline int
+xfs_icluster_size_fsb(
+	struct xfs_mount	*mp)
+{
+	if (mp->m_sb.sb_blocksize >= mp->m_inode_cluster_size)
+		return 1;
+	return mp->m_inode_cluster_size >> mp->m_sb.sb_blocklog;
+}
 
 /*
  * Make an inode pointer out of the buffer/offset.
@@ -89,7 +90,7 @@ xfs_difree(
 	struct xfs_trans *tp,		/* transaction pointer */
 	xfs_ino_t	inode,		/* inode to be freed */
 	struct xfs_bmap_free *flist,	/* extents to free */
-	int		*delete,	/* set if inode cluster was deleted */
+	int		*deleted,	/* set if inode cluster was deleted */
 	xfs_ino_t	*first_ino);	/* first inode in deleted cluster */
 
 /*
diff --git a/fs/xfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c
index c8fa5bbb36de..c9b06f30fe86 100644
--- a/fs/xfs/xfs_ialloc_btree.c
+++ b/fs/xfs/libxfs/xfs_ialloc_btree.c
@@ -49,7 +49,8 @@ xfs_inobt_dup_cursor(
 	struct xfs_btree_cur	*cur)
 {
 	return xfs_inobt_init_cursor(cur->bc_mp, cur->bc_tp,
-			cur->bc_private.a.agbp, cur->bc_private.a.agno);
+			cur->bc_private.a.agbp, cur->bc_private.a.agno,
+			cur->bc_btnum);
 }
 
 STATIC void
@@ -66,12 +67,26 @@ xfs_inobt_set_root(
 	xfs_ialloc_log_agi(cur->bc_tp, agbp, XFS_AGI_ROOT | XFS_AGI_LEVEL);
 }
 
+STATIC void
+xfs_finobt_set_root(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_ptr	*nptr,
+	int			inc)	/* level change */
+{
+	struct xfs_buf		*agbp = cur->bc_private.a.agbp;
+	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agbp);
+
+	agi->agi_free_root = nptr->s;
+	be32_add_cpu(&agi->agi_free_level, inc);
+	xfs_ialloc_log_agi(cur->bc_tp, agbp,
+			   XFS_AGI_FREE_ROOT | XFS_AGI_FREE_LEVEL);
+}
+
 STATIC int
 xfs_inobt_alloc_block(
 	struct xfs_btree_cur	*cur,
 	union xfs_btree_ptr	*start,
 	union xfs_btree_ptr	*new,
-	int			length,
 	int			*stat)
 {
 	xfs_alloc_arg_t		args;		/* block allocation args */
@@ -173,6 +188,17 @@ xfs_inobt_init_ptr_from_cur(
 	ptr->s = agi->agi_root;
 }
 
+STATIC void
+xfs_finobt_init_ptr_from_cur(
+	struct xfs_btree_cur	*cur,
+	union xfs_btree_ptr	*ptr)
+{
+	struct xfs_agi		*agi = XFS_BUF_TO_AGI(cur->bc_private.a.agbp);
+
+	ASSERT(cur->bc_private.a.agno == be32_to_cpu(agi->agi_seqno));
+	ptr->s = agi->agi_free_root;
+}
+
 STATIC __int64_t
 xfs_inobt_key_diff(
 	struct xfs_btree_cur	*cur,
@@ -203,6 +229,7 @@ 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_uuid))
@@ -214,6 +241,7 @@ xfs_inobt_verify(
 			return false;
 		/* fall through */
 	case cpu_to_be32(XFS_IBT_MAGIC):
+	case cpu_to_be32(XFS_FIBT_MAGIC):
 		break;
 	default:
 		return 0;
@@ -243,12 +271,14 @@ static void
 xfs_inobt_read_verify(
 	struct xfs_buf	*bp)
 {
-	if (!(xfs_btree_sblock_verify_crc(bp) &&
-	      xfs_inobt_verify(bp))) {
+	if (!xfs_btree_sblock_verify_crc(bp))
+		xfs_buf_ioerror(bp, -EFSBADCRC);
+	else if (!xfs_inobt_verify(bp))
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+
+	if (bp->b_error) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
-				     bp->b_target->bt_mount, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_verifier_error(bp);
 	}
 }
 
@@ -258,9 +288,9 @@ xfs_inobt_write_verify(
 {
 	if (!xfs_inobt_verify(bp)) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
-				     bp->b_target->bt_mount, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
+		return;
 	}
 	xfs_btree_sblock_calc_crc(bp);
 
@@ -315,6 +345,28 @@ static const struct xfs_btree_ops xfs_inobt_ops = {
 #endif
 };
 
+static const struct xfs_btree_ops xfs_finobt_ops = {
+	.rec_len		= sizeof(xfs_inobt_rec_t),
+	.key_len		= sizeof(xfs_inobt_key_t),
+
+	.dup_cursor		= xfs_inobt_dup_cursor,
+	.set_root		= xfs_finobt_set_root,
+	.alloc_block		= xfs_inobt_alloc_block,
+	.free_block		= xfs_inobt_free_block,
+	.get_minrecs		= xfs_inobt_get_minrecs,
+	.get_maxrecs		= xfs_inobt_get_maxrecs,
+	.init_key_from_rec	= xfs_inobt_init_key_from_rec,
+	.init_rec_from_key	= xfs_inobt_init_rec_from_key,
+	.init_rec_from_cur	= xfs_inobt_init_rec_from_cur,
+	.init_ptr_from_cur	= xfs_finobt_init_ptr_from_cur,
+	.key_diff		= xfs_inobt_key_diff,
+	.buf_ops		= &xfs_inobt_buf_ops,
+#if defined(DEBUG) || defined(XFS_WARN)
+	.keys_inorder		= xfs_inobt_keys_inorder,
+	.recs_inorder		= xfs_inobt_recs_inorder,
+#endif
+};
+
 /*
  * Allocate a new inode btree cursor.
  */
@@ -323,7 +375,8 @@ xfs_inobt_init_cursor(
 	struct xfs_mount	*mp,		/* file system mount point */
 	struct xfs_trans	*tp,		/* transaction pointer */
 	struct xfs_buf		*agbp,		/* buffer for agi structure */
-	xfs_agnumber_t		agno)		/* allocation group number */
+	xfs_agnumber_t		agno,		/* allocation group number */
+	xfs_btnum_t		btnum)		/* ialloc or free ino btree */
 {
 	struct xfs_agi		*agi = XFS_BUF_TO_AGI(agbp);
 	struct xfs_btree_cur	*cur;
@@ -332,11 +385,17 @@ xfs_inobt_init_cursor(
 
 	cur->bc_tp = tp;
 	cur->bc_mp = mp;
-	cur->bc_nlevels = be32_to_cpu(agi->agi_level);
-	cur->bc_btnum = XFS_BTNUM_INO;
+	cur->bc_btnum = btnum;
+	if (btnum == XFS_BTNUM_INO) {
+		cur->bc_nlevels = be32_to_cpu(agi->agi_level);
+		cur->bc_ops = &xfs_inobt_ops;
+	} else {
+		cur->bc_nlevels = be32_to_cpu(agi->agi_free_level);
+		cur->bc_ops = &xfs_finobt_ops;
+	}
+
 	cur->bc_blocklog = mp->m_sb.sb_blocklog;
 
-	cur->bc_ops = &xfs_inobt_ops;
 	if (xfs_sb_version_hascrc(&mp->m_sb))
 		cur->bc_flags |= XFS_BTREE_CRC_BLOCKS;
 
diff --git a/fs/xfs/xfs_ialloc_btree.h b/fs/xfs/libxfs/xfs_ialloc_btree.h
index f38b22011c4e..d7ebea72c2d0 100644
--- a/fs/xfs/xfs_ialloc_btree.h
+++ b/fs/xfs/libxfs/xfs_ialloc_btree.h
@@ -58,7 +58,8 @@ struct xfs_mount;
 		 ((index) - 1) * sizeof(xfs_inobt_ptr_t)))
 
 extern struct xfs_btree_cur *xfs_inobt_init_cursor(struct xfs_mount *,
-		struct xfs_trans *, struct xfs_buf *, xfs_agnumber_t);
+		struct xfs_trans *, struct xfs_buf *, xfs_agnumber_t,
+		xfs_btnum_t);
 extern int xfs_inobt_maxrecs(struct xfs_mount *, int, int);
 
 #endif	/* __XFS_IALLOC_BTREE_H__ */
diff --git a/fs/xfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c
index 4fc9f39dd89e..f18fd2da49f7 100644
--- a/fs/xfs/xfs_inode_buf.c
+++ b/fs/xfs/libxfs/xfs_inode_buf.c
@@ -101,9 +101,8 @@ xfs_inode_buf_verify(
 				return;
 			}
 
-			xfs_buf_ioerror(bp, EFSCORRUPTED);
-			XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_HIGH,
-					     mp, dip);
+			xfs_buf_ioerror(bp, -EFSCORRUPTED);
+			xfs_verifier_error(bp);
 #ifdef DEBUG
 			xfs_alert(mp,
 				"bad inode magic/vsn daddr %lld #%d (magic=%x)",
@@ -175,14 +174,14 @@ xfs_imap_to_bp(
 				   (int)imap->im_len, buf_flags, &bp,
 				   &xfs_inode_buf_ops);
 	if (error) {
-		if (error == EAGAIN) {
+		if (error == -EAGAIN) {
 			ASSERT(buf_flags & XBF_TRYLOCK);
 			return error;
 		}
 
-		if (error == EFSCORRUPTED &&
+		if (error == -EFSCORRUPTED &&
 		    (iget_flags & XFS_IGET_UNTRUSTED))
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 
 		xfs_warn(mp, "%s: xfs_trans_read_buf() returned error %d.",
 			__func__, error);
@@ -306,7 +305,7 @@ xfs_dinode_verify(
 	if (!xfs_sb_version_hascrc(&mp->m_sb))
 		return false;
 	if (!xfs_verify_cksum((char *)dip, mp->m_sb.sb_inodesize,
-			      offsetof(struct xfs_dinode, di_crc)))
+			      XFS_DINODE_CRC_OFF))
 		return false;
 	if (be64_to_cpu(dip->di_ino) != ip->i_ino)
 		return false;
@@ -327,7 +326,7 @@ xfs_dinode_calc_crc(
 
 	ASSERT(xfs_sb_version_hascrc(&mp->m_sb));
 	crc = xfs_start_cksum((char *)dip, mp->m_sb.sb_inodesize,
-			      offsetof(struct xfs_dinode, di_crc));
+			      XFS_DINODE_CRC_OFF);
 	dip->di_crc = xfs_end_cksum(crc);
 }
 
@@ -391,7 +390,7 @@ xfs_iread(
 				__func__, ip->i_ino);
 
 		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, dip);
-		error = XFS_ERROR(EFSCORRUPTED);
+		error = -EFSCORRUPTED;
 		goto out_brelse;
 	}
 
@@ -438,17 +437,16 @@ xfs_iread(
 	}
 
 	/*
-	 * The inode format changed when we moved the link count and
-	 * made it 32 bits long.  If this is an old format inode,
-	 * convert it in memory to look like a new one.  If it gets
-	 * flushed to disk we will convert back before flushing or
-	 * logging it.  We zero out the new projid field and the old link
-	 * count field.  We'll handle clearing the pad field (the remains
-	 * of the old uuid field) when we actually convert the inode to
-	 * the new format. We don't change the version number so that we
-	 * can distinguish this from a real new format inode.
+	 * 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);
diff --git a/fs/xfs/xfs_inode_buf.h b/fs/xfs/libxfs/xfs_inode_buf.h
index 9308c47f2a52..9308c47f2a52 100644
--- a/fs/xfs/xfs_inode_buf.h
+++ b/fs/xfs/libxfs/xfs_inode_buf.h
diff --git a/fs/xfs/xfs_inode_fork.c b/fs/xfs/libxfs/xfs_inode_fork.c
index cfee14a83cfe..6a00f7fed69d 100644
--- a/fs/xfs/xfs_inode_fork.c
+++ b/fs/xfs/libxfs/xfs_inode_fork.c
@@ -102,7 +102,7 @@ xfs_iformat_fork(
 				be64_to_cpu(dip->di_nblocks));
 		XFS_CORRUPTION_ERROR("xfs_iformat(1)", XFS_ERRLEVEL_LOW,
 				     ip->i_mount, dip);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	if (unlikely(dip->di_forkoff > ip->i_mount->m_sb.sb_inodesize)) {
@@ -111,7 +111,7 @@ xfs_iformat_fork(
 			dip->di_forkoff);
 		XFS_CORRUPTION_ERROR("xfs_iformat(2)", XFS_ERRLEVEL_LOW,
 				     ip->i_mount, dip);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	if (unlikely((ip->i_d.di_flags & XFS_DIFLAG_REALTIME) &&
@@ -121,7 +121,7 @@ xfs_iformat_fork(
 			ip->i_ino);
 		XFS_CORRUPTION_ERROR("xfs_iformat(realtime)",
 				     XFS_ERRLEVEL_LOW, ip->i_mount, dip);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	switch (ip->i_d.di_mode & S_IFMT) {
@@ -132,7 +132,7 @@ xfs_iformat_fork(
 		if (unlikely(dip->di_format != XFS_DINODE_FMT_DEV)) {
 			XFS_CORRUPTION_ERROR("xfs_iformat(3)", XFS_ERRLEVEL_LOW,
 					      ip->i_mount, dip);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 		ip->i_d.di_size = 0;
 		ip->i_df.if_u2.if_rdev = xfs_dinode_get_rdev(dip);
@@ -153,7 +153,7 @@ xfs_iformat_fork(
 				XFS_CORRUPTION_ERROR("xfs_iformat(4)",
 						     XFS_ERRLEVEL_LOW,
 						     ip->i_mount, dip);
-				return XFS_ERROR(EFSCORRUPTED);
+				return -EFSCORRUPTED;
 			}
 
 			di_size = be64_to_cpu(dip->di_size);
@@ -166,7 +166,7 @@ xfs_iformat_fork(
 				XFS_CORRUPTION_ERROR("xfs_iformat(5)",
 						     XFS_ERRLEVEL_LOW,
 						     ip->i_mount, dip);
-				return XFS_ERROR(EFSCORRUPTED);
+				return -EFSCORRUPTED;
 			}
 
 			size = (int)di_size;
@@ -181,13 +181,13 @@ xfs_iformat_fork(
 		default:
 			XFS_ERROR_REPORT("xfs_iformat(6)", XFS_ERRLEVEL_LOW,
 					 ip->i_mount);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 		break;
 
 	default:
 		XFS_ERROR_REPORT("xfs_iformat(7)", XFS_ERRLEVEL_LOW, ip->i_mount);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	if (error) {
 		return error;
@@ -211,7 +211,7 @@ xfs_iformat_fork(
 			XFS_CORRUPTION_ERROR("xfs_iformat(8)",
 					     XFS_ERRLEVEL_LOW,
 					     ip->i_mount, dip);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 
 		error = xfs_iformat_local(ip, dip, XFS_ATTR_FORK, size);
@@ -223,7 +223,7 @@ xfs_iformat_fork(
 		error = xfs_iformat_btree(ip, dip, XFS_ATTR_FORK);
 		break;
 	default:
-		error = XFS_ERROR(EFSCORRUPTED);
+		error = -EFSCORRUPTED;
 		break;
 	}
 	if (error) {
@@ -266,7 +266,7 @@ xfs_iformat_local(
 			XFS_DFORK_SIZE(dip, ip->i_mount, whichfork));
 		XFS_CORRUPTION_ERROR("xfs_iformat_local", XFS_ERRLEVEL_LOW,
 				     ip->i_mount, dip);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	ifp = XFS_IFORK_PTR(ip, whichfork);
 	real_size = 0;
@@ -322,7 +322,7 @@ xfs_iformat_extents(
 			(unsigned long long) ip->i_ino, nex);
 		XFS_CORRUPTION_ERROR("xfs_iformat_extents(1)", XFS_ERRLEVEL_LOW,
 				     ip->i_mount, dip);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	ifp->if_real_bytes = 0;
@@ -350,7 +350,7 @@ xfs_iformat_extents(
 					XFS_ERROR_REPORT("xfs_iformat_extents(2)",
 							 XFS_ERRLEVEL_LOW,
 							 ip->i_mount);
-					return XFS_ERROR(EFSCORRUPTED);
+					return -EFSCORRUPTED;
 				}
 	}
 	ifp->if_flags |= XFS_IFEXTENTS;
@@ -399,7 +399,7 @@ xfs_iformat_btree(
 					(unsigned long long) ip->i_ino);
 		XFS_CORRUPTION_ERROR("xfs_iformat_btree", XFS_ERRLEVEL_LOW,
 					 mp, dip);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	ifp->if_broot_bytes = size;
@@ -431,10 +431,12 @@ xfs_iread_extents(
 	xfs_ifork_t	*ifp;
 	xfs_extnum_t	nextents;
 
+	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
+
 	if (unlikely(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) {
 		XFS_ERROR_REPORT("xfs_iread_extents", XFS_ERRLEVEL_LOW,
 				 ip->i_mount);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	nextents = XFS_IFORK_NEXTENTS(ip, whichfork);
 	ifp = XFS_IFORK_PTR(ip, whichfork);
@@ -526,7 +528,7 @@ xfs_iroot_realloc(
 		ifp->if_broot_bytes = (int)new_size;
 		ASSERT(XFS_BMAP_BMDR_SPACE(ifp->if_broot) <=
 			XFS_IFORK_SIZE(ip, whichfork));
-		memmove(np, op, cur_max * (uint)sizeof(xfs_dfsbno_t));
+		memmove(np, op, cur_max * (uint)sizeof(xfs_fsblock_t));
 		return;
 	}
 
@@ -573,7 +575,7 @@ xfs_iroot_realloc(
 						     ifp->if_broot_bytes);
 		np = (char *)XFS_BMAP_BROOT_PTR_ADDR(mp, new_broot, 1,
 						     (int)new_size);
-		memcpy(np, op, new_max * (uint)sizeof(xfs_dfsbno_t));
+		memcpy(np, op, new_max * (uint)sizeof(xfs_fsblock_t));
 	}
 	kmem_free(ifp->if_broot);
 	ifp->if_broot = new_broot;
@@ -721,15 +723,16 @@ xfs_idestroy_fork(
 }
 
 /*
- * xfs_iextents_copy()
+ * Convert in-core extents to on-disk form
  *
- * This is called to copy the REAL extents (as opposed to the delayed
- * allocation extents) from the inode into the given buffer.  It
- * returns the number of bytes copied into the buffer.
+ * For either the data or attr fork in extent format, we need to endian convert
+ * the in-core extent as we place them into the on-disk inode.
  *
- * If there are no delayed allocation extents, then we can just
- * memcpy() the extents into the buffer.  Otherwise, we need to
- * examine each extent in turn and skip those which are delayed.
+ * In the case of the data fork, the in-core and on-disk fork sizes can be
+ * different due to delayed allocation extents. We only copy on-disk extents
+ * here, so callers must always use the physical fork size to determine the
+ * size of the buffer passed to this routine.  We will return the size actually
+ * used.
  */
 int
 xfs_iextents_copy(
@@ -795,8 +798,7 @@ xfs_iflush_fork(
 	xfs_inode_t		*ip,
 	xfs_dinode_t		*dip,
 	xfs_inode_log_item_t	*iip,
-	int			whichfork,
-	xfs_buf_t		*bp)
+	int			whichfork)
 {
 	char			*cp;
 	xfs_ifork_t		*ifp;
@@ -1690,7 +1692,7 @@ xfs_iext_idx_to_irec(
 	}
 	*idxp = page_idx;
 	*erp_idxp = erp_idx;
-	return(erp);
+	return erp;
 }
 
 /*
diff --git a/fs/xfs/xfs_inode_fork.h b/fs/xfs/libxfs/xfs_inode_fork.h
index eb329a1ea888..7d3b1ed6dcbe 100644
--- a/fs/xfs/xfs_inode_fork.h
+++ b/fs/xfs/libxfs/xfs_inode_fork.h
@@ -127,8 +127,7 @@ typedef struct xfs_ifork {
 
 int		xfs_iformat_fork(struct xfs_inode *, struct xfs_dinode *);
 void		xfs_iflush_fork(struct xfs_inode *, struct xfs_dinode *,
-				struct xfs_inode_log_item *, int,
-				struct xfs_buf *);
+				struct xfs_inode_log_item *, int);
 void		xfs_idestroy_fork(struct xfs_inode *, int);
 void		xfs_idata_realloc(struct xfs_inode *, int, int);
 void		xfs_iroot_realloc(struct xfs_inode *, int, int);
diff --git a/fs/xfs/xfs_inum.h b/fs/xfs/libxfs/xfs_inum.h
index 90efdaf1706f..4ff2278e147a 100644
--- a/fs/xfs/xfs_inum.h
+++ b/fs/xfs/libxfs/xfs_inum.h
@@ -54,11 +54,7 @@ struct xfs_mount;
 #define	XFS_OFFBNO_TO_AGINO(mp,b,o)	\
 	((xfs_agino_t)(((b) << XFS_INO_OFFSET_BITS(mp)) | (o)))
 
-#if XFS_BIG_INUMS
 #define	XFS_MAXINUMBER		((xfs_ino_t)((1ULL << 56) - 1ULL))
-#else
-#define	XFS_MAXINUMBER		((xfs_ino_t)((1ULL << 32) - 1ULL))
-#endif
 #define	XFS_MAXINUMBER_32	((xfs_ino_t)((1ULL << 32) - 1ULL))
 
 #endif	/* __XFS_INUM_H__ */
diff --git a/fs/xfs/xfs_log_format.h b/fs/xfs/libxfs/xfs_log_format.h
index f0969c77bdbe..aff12f2d4428 100644
--- a/fs/xfs/xfs_log_format.h
+++ b/fs/xfs/libxfs/xfs_log_format.h
@@ -380,7 +380,7 @@ typedef struct xfs_icdinode {
 	xfs_ictimestamp_t di_mtime;	/* time last modified */
 	xfs_ictimestamp_t di_ctime;	/* time created/inode modified */
 	xfs_fsize_t	di_size;	/* number of bytes in file */
-	xfs_drfsbno_t	di_nblocks;	/* # of direct & btree blocks used */
+	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*/
@@ -516,7 +516,7 @@ xfs_blft_from_flags(struct xfs_buf_log_format *blf)
  * EFI/EFD log format definitions
  */
 typedef struct xfs_extent {
-	xfs_dfsbno_t	ext_start;
+	xfs_fsblock_t	ext_start;
 	xfs_extlen_t	ext_len;
 } xfs_extent_t;
 
diff --git a/fs/xfs/xfs_log_recover.h b/fs/xfs/libxfs/xfs_log_recover.h
index 1c55ccbb379d..1c55ccbb379d 100644
--- a/fs/xfs/xfs_log_recover.h
+++ b/fs/xfs/libxfs/xfs_log_recover.h
diff --git a/fs/xfs/xfs_log_rlimit.c b/fs/xfs/libxfs/xfs_log_rlimit.c
index 2af1a0a4d0f1..ee7e0e80246b 100644
--- a/fs/xfs/xfs_log_rlimit.c
+++ b/fs/xfs/libxfs/xfs_log_rlimit.c
@@ -42,7 +42,7 @@ xfs_log_calc_max_attrsetm_res(
 	int			size;
 	int			nblks;
 
-	size = xfs_attr_leaf_entsize_local_max(mp->m_sb.sb_blocksize) -
+	size = xfs_attr_leaf_entsize_local_max(mp->m_attr_geo->blksize) -
 	       MAXNAMELEN - 1;
 	nblks = XFS_DAENTER_SPACE_RES(mp, XFS_ATTR_FORK);
 	nblks += XFS_B_TO_FSB(mp, size);
diff --git a/fs/xfs/xfs_quota_defs.h b/fs/xfs/libxfs/xfs_quota_defs.h
index b3b2b1065c0f..1b0a08379759 100644
--- a/fs/xfs/xfs_quota_defs.h
+++ b/fs/xfs/libxfs/xfs_quota_defs.h
@@ -98,8 +98,6 @@ typedef __uint16_t	xfs_qwarncnt_t;
 #define XFS_IS_QUOTA_ON(mp)	((mp)->m_qflags & (XFS_UQUOTA_ACTIVE | \
 						   XFS_GQUOTA_ACTIVE | \
 						   XFS_PQUOTA_ACTIVE))
-#define XFS_IS_OQUOTA_ON(mp)	((mp)->m_qflags & (XFS_GQUOTA_ACTIVE | \
-						   XFS_PQUOTA_ACTIVE))
 #define XFS_IS_UQUOTA_ON(mp)	((mp)->m_qflags & XFS_UQUOTA_ACTIVE)
 #define XFS_IS_GQUOTA_ON(mp)	((mp)->m_qflags & XFS_GQUOTA_ACTIVE)
 #define XFS_IS_PQUOTA_ON(mp)	((mp)->m_qflags & XFS_PQUOTA_ACTIVE)
@@ -156,6 +154,6 @@ typedef __uint16_t	xfs_qwarncnt_t;
 
 extern int xfs_dqcheck(struct xfs_mount *mp, xfs_disk_dquot_t *ddq,
 		       xfs_dqid_t id, uint type, uint flags, char *str);
-extern int xfs_calc_dquots_per_chunk(struct xfs_mount *mp, unsigned int nbblks);
+extern int xfs_calc_dquots_per_chunk(unsigned int nbblks);
 
 #endif	/* __XFS_QUOTA_H__ */
diff --git a/fs/xfs/xfs_rtbitmap.c b/fs/xfs/libxfs/xfs_rtbitmap.c
index b1f2fe8af4a8..f4dd697cac08 100644
--- a/fs/xfs/xfs_rtbitmap.c
+++ b/fs/xfs/libxfs/xfs_rtbitmap.c
@@ -74,7 +74,6 @@ xfs_rtbuf_get(
 				   mp->m_bsize, 0, &bp, NULL);
 	if (error)
 		return error;
-	ASSERT(!xfs_buf_geterror(bp));
 	*bpp = bp;
 	return 0;
 }
diff --git a/fs/xfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c
index b7c9aea77f8f..ad525a5623a4 100644
--- a/fs/xfs/xfs_sb.c
+++ b/fs/xfs/libxfs/xfs_sb.c
@@ -186,13 +186,13 @@ xfs_mount_validate_sb(
 	 */
 	if (sbp->sb_magicnum != XFS_SB_MAGIC) {
 		xfs_warn(mp, "bad magic number");
-		return XFS_ERROR(EWRONGFS);
+		return -EWRONGFS;
 	}
 
 
 	if (!xfs_sb_good_version(sbp)) {
 		xfs_warn(mp, "bad version");
-		return XFS_ERROR(EWRONGFS);
+		return -EWRONGFS;
 	}
 
 	/*
@@ -201,10 +201,6 @@ xfs_mount_validate_sb(
 	 * write validation, we don't need to check feature masks.
 	 */
 	if (check_version && XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) {
-		xfs_alert(mp,
-"Version 5 superblock detected. This kernel has EXPERIMENTAL support enabled!\n"
-"Use of these features in this kernel is at your own risk!");
-
 		if (xfs_sb_has_compat_feature(sbp,
 					XFS_SB_FEAT_COMPAT_UNKNOWN)) {
 			xfs_warn(mp,
@@ -224,7 +220,7 @@ xfs_mount_validate_sb(
 				xfs_warn(mp,
 "Attempted to mount read-only compatible filesystem read-write.\n"
 "Filesystem can only be safely mounted read only.");
-				return XFS_ERROR(EINVAL);
+				return -EINVAL;
 			}
 		}
 		if (xfs_sb_has_incompat_feature(sbp,
@@ -234,7 +230,7 @@ xfs_mount_validate_sb(
 "Filesystem can not be safely mounted by this kernel.",
 				(sbp->sb_features_incompat &
 						XFS_SB_FEAT_INCOMPAT_UNKNOWN));
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 	}
 
@@ -242,13 +238,13 @@ xfs_mount_validate_sb(
 		if (sbp->sb_qflags & (XFS_OQUOTA_ENFD | XFS_OQUOTA_CHKD)) {
 			xfs_notice(mp,
 			   "Version 5 of Super block has XFS_OQUOTA bits.");
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 	} else if (sbp->sb_qflags & (XFS_PQUOTA_ENFD | XFS_GQUOTA_ENFD |
 				XFS_PQUOTA_CHKD | XFS_GQUOTA_CHKD)) {
 			xfs_notice(mp,
 "Superblock earlier than Version 5 has XFS_[PQ]UOTA_{ENFD|CHKD} bits.");
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 	}
 
 	if (unlikely(
@@ -256,7 +252,7 @@ xfs_mount_validate_sb(
 		xfs_warn(mp,
 		"filesystem is marked as having an external log; "
 		"specify logdev on the mount command line.");
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	if (unlikely(
@@ -264,7 +260,7 @@ xfs_mount_validate_sb(
 		xfs_warn(mp,
 		"filesystem is marked as having an internal log; "
 		"do not specify logdev on the mount command line.");
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	/*
@@ -288,16 +284,17 @@ xfs_mount_validate_sb(
 	    sbp->sb_inodelog < XFS_DINODE_MIN_LOG			||
 	    sbp->sb_inodelog > XFS_DINODE_MAX_LOG			||
 	    sbp->sb_inodesize != (1 << sbp->sb_inodelog)		||
+	    sbp->sb_inopblock != howmany(sbp->sb_blocksize,sbp->sb_inodesize) ||
 	    (sbp->sb_blocklog - sbp->sb_inodelog != sbp->sb_inopblog)	||
 	    (sbp->sb_rextsize * sbp->sb_blocksize > XFS_MAX_RTEXTSIZE)	||
 	    (sbp->sb_rextsize * sbp->sb_blocksize < XFS_MIN_RTEXTSIZE)	||
 	    (sbp->sb_imax_pct > 100 /* zero sb_imax_pct is valid */)	||
 	    sbp->sb_dblocks == 0					||
 	    sbp->sb_dblocks > XFS_MAX_DBLOCKS(sbp)			||
-	    sbp->sb_dblocks < XFS_MIN_DBLOCKS(sbp))) {
-		XFS_CORRUPTION_ERROR("SB sanity check failed",
-				XFS_ERRLEVEL_LOW, mp, sbp);
-		return XFS_ERROR(EFSCORRUPTED);
+	    sbp->sb_dblocks < XFS_MIN_DBLOCKS(sbp)			||
+	    sbp->sb_shared_vn != 0)) {
+		xfs_notice(mp, "SB sanity check failed");
+		return -EFSCORRUPTED;
 	}
 
 	/*
@@ -308,7 +305,7 @@ xfs_mount_validate_sb(
 		"File system with blocksize %d bytes. "
 		"Only pagesize (%ld) or less will currently work.",
 				sbp->sb_blocksize, PAGE_SIZE);
-		return XFS_ERROR(ENOSYS);
+		return -ENOSYS;
 	}
 
 	/*
@@ -323,29 +320,20 @@ xfs_mount_validate_sb(
 	default:
 		xfs_warn(mp, "inode size of %d bytes not supported",
 				sbp->sb_inodesize);
-		return XFS_ERROR(ENOSYS);
+		return -ENOSYS;
 	}
 
 	if (xfs_sb_validate_fsb_count(sbp, sbp->sb_dblocks) ||
 	    xfs_sb_validate_fsb_count(sbp, sbp->sb_rblocks)) {
 		xfs_warn(mp,
 		"file system too large to be mounted on this system.");
-		return XFS_ERROR(EFBIG);
+		return -EFBIG;
 	}
 
 	if (check_inprogress && sbp->sb_inprogress) {
 		xfs_warn(mp, "Offline file system operation in progress!");
-		return XFS_ERROR(EFSCORRUPTED);
-	}
-
-	/*
-	 * Version 1 directory format has never worked on Linux.
-	 */
-	if (unlikely(!xfs_sb_version_hasdirv2(sbp))) {
-		xfs_warn(mp, "file system using version 1 directory format");
-		return XFS_ERROR(ENOSYS);
+		return -EFSCORRUPTED;
 	}
-
 	return 0;
 }
 
@@ -398,10 +386,11 @@ xfs_sb_quota_from_disk(struct xfs_sb *sbp)
 	}
 }
 
-void
-xfs_sb_from_disk(
+static void
+__xfs_sb_from_disk(
 	struct xfs_sb	*to,
-	xfs_dsb_t	*from)
+	xfs_dsb_t	*from,
+	bool		convert_xquota)
 {
 	to->sb_magicnum = be32_to_cpu(from->sb_magicnum);
 	to->sb_blocksize = be32_to_cpu(from->sb_blocksize);
@@ -457,6 +446,17 @@ xfs_sb_from_disk(
 	to->sb_pad = 0;
 	to->sb_pquotino = be64_to_cpu(from->sb_pquotino);
 	to->sb_lsn = be64_to_cpu(from->sb_lsn);
+	/* Convert on-disk flags to in-memory flags? */
+	if (convert_xquota)
+		xfs_sb_quota_from_disk(to);
+}
+
+void
+xfs_sb_from_disk(
+	struct xfs_sb	*to,
+	xfs_dsb_t	*from)
+{
+	__xfs_sb_from_disk(to, from, true);
 }
 
 static inline void
@@ -495,10 +495,16 @@ xfs_sb_quota_to_disk(
 	}
 
 	/*
-	 * GQUOTINO and PQUOTINO cannot be used together in versions
-	 * of superblock that do not have pquotino. from->sb_flags
-	 * tells us which quota is active and should be copied to
-	 * disk.
+	 * GQUOTINO and PQUOTINO cannot be used together in versions of
+	 * superblock that do not have pquotino. from->sb_flags tells us which
+	 * quota is active and should be copied to disk. If neither are active,
+	 * make sure we write NULLFSINO to the sb_gquotino field as a quota
+	 * inode value of "0" is invalid when the XFS_SB_VERSION_QUOTA feature
+	 * bit is set.
+	 *
+	 * Note that we don't need to handle the sb_uquotino or sb_pquotino here
+	 * as they do not require any translation. Hence the main sb field loop
+	 * will write them appropriately from the in-core superblock.
 	 */
 	if ((*fields & XFS_SB_GQUOTINO) &&
 				(from->sb_qflags & XFS_GQUOTA_ACCT))
@@ -506,6 +512,17 @@ xfs_sb_quota_to_disk(
 	else if ((*fields & XFS_SB_PQUOTINO) &&
 				(from->sb_qflags & XFS_PQUOTA_ACCT))
 		to->sb_gquotino = cpu_to_be64(from->sb_pquotino);
+	else {
+		/*
+		 * We can't rely on just the fields being logged to tell us
+		 * that it is safe to write NULLFSINO - we should only do that
+		 * if quotas are not actually enabled. Hence only write
+		 * NULLFSINO if both in-core quota inodes are NULL.
+		 */
+		if (from->sb_gquotino == NULLFSINO &&
+		    from->sb_pquotino == NULLFSINO)
+			to->sb_gquotino = cpu_to_be64(NULLFSINO);
+	}
 
 	*fields &= ~(XFS_SB_PQUOTINO | XFS_SB_GQUOTINO);
 }
@@ -572,7 +589,11 @@ xfs_sb_verify(
 	struct xfs_mount *mp = bp->b_target->bt_mount;
 	struct xfs_sb	sb;
 
-	xfs_sb_from_disk(&sb, XFS_BUF_TO_SBP(bp));
+	/*
+	 * Use call variant which doesn't convert quota flags from disk 
+	 * format, because xfs_mount_validate_sb checks the on-disk flags.
+	 */
+	__xfs_sb_from_disk(&sb, XFS_BUF_TO_SBP(bp), false);
 
 	/*
 	 * Only check the in progress field for the primary superblock as
@@ -611,12 +632,11 @@ xfs_sb_read_verify(
 						XFS_SB_VERSION_5) ||
 	     dsb->sb_crc != 0)) {
 
-		if (!xfs_verify_cksum(bp->b_addr, be16_to_cpu(dsb->sb_sectsize),
-				      offsetof(struct xfs_sb, sb_crc))) {
+		if (!xfs_buf_verify_cksum(bp, XFS_SB_CRC_OFF)) {
 			/* Only fail bad secondaries on a known V5 filesystem */
-			if (bp->b_bn != XFS_SB_DADDR &&
+			if (bp->b_bn == XFS_SB_DADDR ||
 			    xfs_sb_version_hascrc(&mp->m_sb)) {
-				error = EFSCORRUPTED;
+				error = -EFSBADCRC;
 				goto out_error;
 			}
 		}
@@ -625,10 +645,9 @@ xfs_sb_read_verify(
 
 out_error:
 	if (error) {
-		if (error != EWRONGFS)
-			XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
-					     mp, bp->b_addr);
 		xfs_buf_ioerror(bp, error);
+		if (error == -EFSCORRUPTED || error == -EFSBADCRC)
+			xfs_verifier_error(bp);
 	}
 }
 
@@ -644,14 +663,13 @@ xfs_sb_quiet_read_verify(
 {
 	struct xfs_dsb	*dsb = XFS_BUF_TO_SBP(bp);
 
-
 	if (dsb->sb_magicnum == cpu_to_be32(XFS_SB_MAGIC)) {
 		/* XFS filesystem, verify noisily! */
 		xfs_sb_read_verify(bp);
 		return;
 	}
 	/* quietly fail */
-	xfs_buf_ioerror(bp, EWRONGFS);
+	xfs_buf_ioerror(bp, -EWRONGFS);
 }
 
 static void
@@ -664,9 +682,8 @@ xfs_sb_write_verify(
 
 	error = xfs_sb_verify(bp, false);
 	if (error) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW,
-				     mp, bp->b_addr);
 		xfs_buf_ioerror(bp, error);
+		xfs_verifier_error(bp);
 		return;
 	}
 
@@ -676,8 +693,7 @@ xfs_sb_write_verify(
 	if (bip)
 		XFS_BUF_TO_SBP(bp)->sb_lsn = cpu_to_be64(bip->bli_item.li_lsn);
 
-	xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
-			 offsetof(struct xfs_sb, sb_crc));
+	xfs_buf_update_cksum(bp, XFS_SB_CRC_OFF);
 }
 
 const struct xfs_buf_ops xfs_sb_buf_ops = {
diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/libxfs/xfs_sb.h
index 35061d4b614c..2e739708afd3 100644
--- a/fs/xfs/xfs_sb.h
+++ b/fs/xfs/libxfs/xfs_sb.h
@@ -36,8 +36,6 @@ struct xfs_trans;
 #define	XFS_SB_VERSION_5	5		/* CRC enabled filesystem */
 #define	XFS_SB_VERSION_NUMBITS		0x000f
 #define	XFS_SB_VERSION_ALLFBITS		0xfff0
-#define	XFS_SB_VERSION_SASHFBITS	0xf000
-#define	XFS_SB_VERSION_REALFBITS	0x0ff0
 #define	XFS_SB_VERSION_ATTRBIT		0x0010
 #define	XFS_SB_VERSION_NLINKBIT		0x0020
 #define	XFS_SB_VERSION_QUOTABIT		0x0040
@@ -50,24 +48,15 @@ struct xfs_trans;
 #define	XFS_SB_VERSION_DIRV2BIT		0x2000
 #define	XFS_SB_VERSION_BORGBIT		0x4000	/* ASCII only case-insens. */
 #define	XFS_SB_VERSION_MOREBITSBIT	0x8000
-#define	XFS_SB_VERSION_OKSASHFBITS	\
-	(XFS_SB_VERSION_EXTFLGBIT | \
-	 XFS_SB_VERSION_DIRV2BIT | \
-	 XFS_SB_VERSION_BORGBIT)
-#define	XFS_SB_VERSION_OKREALFBITS	\
-	(XFS_SB_VERSION_ATTRBIT | \
-	 XFS_SB_VERSION_NLINKBIT | \
-	 XFS_SB_VERSION_QUOTABIT | \
-	 XFS_SB_VERSION_ALIGNBIT | \
-	 XFS_SB_VERSION_DALIGNBIT | \
-	 XFS_SB_VERSION_SHAREDBIT | \
-	 XFS_SB_VERSION_LOGV2BIT | \
-	 XFS_SB_VERSION_SECTORBIT | \
-	 XFS_SB_VERSION_MOREBITSBIT)
-#define	XFS_SB_VERSION_OKREALBITS	\
-	(XFS_SB_VERSION_NUMBITS | \
-	 XFS_SB_VERSION_OKREALFBITS | \
-	 XFS_SB_VERSION_OKSASHFBITS)
+
+/*
+ * 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.
+ */
+#define	XFS_SB_VERSION_OKBITS		\
+	((XFS_SB_VERSION_NUMBITS | XFS_SB_VERSION_ALLFBITS) & \
+		~XFS_SB_VERSION_SHAREDBIT)
 
 /*
  * There are two words to hold XFS "feature" bits: the original
@@ -76,7 +65,6 @@ struct xfs_trans;
  *
  * These defines represent bits in sb_features2.
  */
-#define XFS_SB_VERSION2_REALFBITS	0x00ffffff	/* Mask: features */
 #define XFS_SB_VERSION2_RESERVED1BIT	0x00000001
 #define XFS_SB_VERSION2_LAZYSBCOUNTBIT	0x00000002	/* Superblk counters */
 #define XFS_SB_VERSION2_RESERVED4BIT	0x00000004
@@ -86,16 +74,11 @@ struct xfs_trans;
 #define XFS_SB_VERSION2_CRCBIT		0x00000100	/* metadata CRCs */
 #define XFS_SB_VERSION2_FTYPE		0x00000200	/* inode type in dir */
 
-#define	XFS_SB_VERSION2_OKREALFBITS	\
+#define	XFS_SB_VERSION2_OKBITS		\
 	(XFS_SB_VERSION2_LAZYSBCOUNTBIT	| \
 	 XFS_SB_VERSION2_ATTR2BIT	| \
 	 XFS_SB_VERSION2_PROJID32BIT	| \
 	 XFS_SB_VERSION2_FTYPE)
-#define	XFS_SB_VERSION2_OKSASHFBITS	\
-	(0)
-#define XFS_SB_VERSION2_OKREALBITS	\
-	(XFS_SB_VERSION2_OKREALFBITS |	\
-	 XFS_SB_VERSION2_OKSASHFBITS )
 
 /*
  * Superblock - in core version.  Must match the ondisk version below.
@@ -104,11 +87,11 @@ struct xfs_trans;
 typedef struct xfs_sb {
 	__uint32_t	sb_magicnum;	/* magic number == XFS_SB_MAGIC */
 	__uint32_t	sb_blocksize;	/* logical block size, bytes */
-	xfs_drfsbno_t	sb_dblocks;	/* number of data blocks */
-	xfs_drfsbno_t	sb_rblocks;	/* number of realtime blocks */
-	xfs_drtbno_t	sb_rextents;	/* number of realtime extents */
+	xfs_rfsblock_t	sb_dblocks;	/* number of data blocks */
+	xfs_rfsblock_t	sb_rblocks;	/* number of realtime blocks */
+	xfs_rtblock_t	sb_rextents;	/* number of realtime extents */
 	uuid_t		sb_uuid;	/* file system unique id */
-	xfs_dfsbno_t	sb_logstart;	/* starting block of log if internal */
+	xfs_fsblock_t	sb_logstart;	/* starting block of log if internal */
 	xfs_ino_t	sb_rootino;	/* root inode number */
 	xfs_ino_t	sb_rbmino;	/* bitmap inode for realtime extents */
 	xfs_ino_t	sb_rsumino;	/* summary inode for rt bitmap */
@@ -182,6 +165,8 @@ typedef struct xfs_sb {
 	/* must be padded to 64 bit alignment */
 } xfs_sb_t;
 
+#define XFS_SB_CRC_OFF		offsetof(struct xfs_sb, sb_crc)
+
 /*
  * Superblock - on disk version.  Must match the in core version above.
  * Must be padded to 64 bit alignment.
@@ -343,214 +328,140 @@ typedef enum {
 
 #define	XFS_SB_VERSION_NUM(sbp)	((sbp)->sb_versionnum & XFS_SB_VERSION_NUMBITS)
 
-static inline int xfs_sb_good_version(xfs_sb_t *sbp)
-{
-	/* We always support version 1-3 */
-	if (sbp->sb_versionnum >= XFS_SB_VERSION_1 &&
-	    sbp->sb_versionnum <= XFS_SB_VERSION_3)
-		return 1;
-
-	/* We support version 4 if all feature bits are supported */
-	if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) {
-		if ((sbp->sb_versionnum & ~XFS_SB_VERSION_OKREALBITS) ||
-		    ((sbp->sb_versionnum & XFS_SB_VERSION_MOREBITSBIT) &&
-		     (sbp->sb_features2 & ~XFS_SB_VERSION2_OKREALBITS)))
-			return 0;
-
-		if (sbp->sb_shared_vn > XFS_SB_MAX_SHARED_VN)
-			return 0;
-		return 1;
-	}
-	if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5)
-		return 1;
-
-	return 0;
-}
-
 /*
- * Detect a mismatched features2 field.  Older kernels read/wrote
- * this into the wrong slot, so to be safe we keep them in sync.
+ * The first XFS version we support is a v4 superblock with V2 directories.
  */
-static inline int xfs_sb_has_mismatched_features2(xfs_sb_t *sbp)
+static inline bool xfs_sb_good_v4_features(struct xfs_sb *sbp)
 {
-	return (sbp->sb_bad_features2 != sbp->sb_features2);
-}
-
-static inline unsigned xfs_sb_version_tonew(unsigned v)
-{
-	if (v == XFS_SB_VERSION_1)
-		return XFS_SB_VERSION_4;
-
-	if (v == XFS_SB_VERSION_2)
-		return XFS_SB_VERSION_4 | XFS_SB_VERSION_ATTRBIT;
+	if (!(sbp->sb_versionnum & XFS_SB_VERSION_DIRV2BIT))
+		return false;
 
-	return XFS_SB_VERSION_4 | XFS_SB_VERSION_ATTRBIT |
-		XFS_SB_VERSION_NLINKBIT;
-}
+	/* check for unknown features in the fs */
+	if ((sbp->sb_versionnum & ~XFS_SB_VERSION_OKBITS) ||
+	    ((sbp->sb_versionnum & XFS_SB_VERSION_MOREBITSBIT) &&
+	     (sbp->sb_features2 & ~XFS_SB_VERSION2_OKBITS)))
+		return false;
 
-static inline unsigned xfs_sb_version_toold(unsigned v)
-{
-	if (v & (XFS_SB_VERSION_QUOTABIT | XFS_SB_VERSION_ALIGNBIT))
-		return 0;
-	if (v & XFS_SB_VERSION_NLINKBIT)
-		return XFS_SB_VERSION_3;
-	if (v & XFS_SB_VERSION_ATTRBIT)
-		return XFS_SB_VERSION_2;
-	return XFS_SB_VERSION_1;
+	return true;
 }
 
-static inline int xfs_sb_version_hasattr(xfs_sb_t *sbp)
+static inline bool xfs_sb_good_version(struct xfs_sb *sbp)
 {
-	return sbp->sb_versionnum == XFS_SB_VERSION_2 ||
-		sbp->sb_versionnum == XFS_SB_VERSION_3 ||
-		(XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
-		 (sbp->sb_versionnum & XFS_SB_VERSION_ATTRBIT));
+	if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5)
+		return true;
+	if (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4)
+		return xfs_sb_good_v4_features(sbp);
+	return false;
 }
 
-static inline void xfs_sb_version_addattr(xfs_sb_t *sbp)
+/*
+ * Detect a mismatched features2 field.  Older kernels read/wrote
+ * this into the wrong slot, so to be safe we keep them in sync.
+ */
+static inline bool xfs_sb_has_mismatched_features2(struct xfs_sb *sbp)
 {
-	if (sbp->sb_versionnum == XFS_SB_VERSION_1)
-		sbp->sb_versionnum = XFS_SB_VERSION_2;
-	else if (XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4)
-		sbp->sb_versionnum |= XFS_SB_VERSION_ATTRBIT;
-	else
-		sbp->sb_versionnum = XFS_SB_VERSION_4 | XFS_SB_VERSION_ATTRBIT;
+	return sbp->sb_bad_features2 != sbp->sb_features2;
 }
 
-static inline int xfs_sb_version_hasnlink(xfs_sb_t *sbp)
+static inline bool xfs_sb_version_hasattr(struct xfs_sb *sbp)
 {
-	return sbp->sb_versionnum == XFS_SB_VERSION_3 ||
-		 (XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
-		  (sbp->sb_versionnum & XFS_SB_VERSION_NLINKBIT));
+	return (sbp->sb_versionnum & XFS_SB_VERSION_ATTRBIT);
 }
 
-static inline void xfs_sb_version_addnlink(xfs_sb_t *sbp)
+static inline void xfs_sb_version_addattr(struct xfs_sb *sbp)
 {
-	if (sbp->sb_versionnum <= XFS_SB_VERSION_2)
-		sbp->sb_versionnum = XFS_SB_VERSION_3;
-	else
-		sbp->sb_versionnum |= XFS_SB_VERSION_NLINKBIT;
+	sbp->sb_versionnum |= XFS_SB_VERSION_ATTRBIT;
 }
 
-static inline int xfs_sb_version_hasquota(xfs_sb_t *sbp)
+static inline bool xfs_sb_version_hasquota(struct xfs_sb *sbp)
 {
-	return XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
-		(sbp->sb_versionnum & XFS_SB_VERSION_QUOTABIT);
+	return (sbp->sb_versionnum & XFS_SB_VERSION_QUOTABIT);
 }
 
-static inline void xfs_sb_version_addquota(xfs_sb_t *sbp)
+static inline void xfs_sb_version_addquota(struct xfs_sb *sbp)
 {
-	if (XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4)
-		sbp->sb_versionnum |= XFS_SB_VERSION_QUOTABIT;
-	else
-		sbp->sb_versionnum = xfs_sb_version_tonew(sbp->sb_versionnum) |
-					XFS_SB_VERSION_QUOTABIT;
+	sbp->sb_versionnum |= XFS_SB_VERSION_QUOTABIT;
 }
 
-static inline int xfs_sb_version_hasalign(xfs_sb_t *sbp)
+static inline bool xfs_sb_version_hasalign(struct xfs_sb *sbp)
 {
-	return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
-	       (XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
+	return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5 ||
 		(sbp->sb_versionnum & XFS_SB_VERSION_ALIGNBIT));
 }
 
-static inline int xfs_sb_version_hasdalign(xfs_sb_t *sbp)
-{
-	return XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
-		(sbp->sb_versionnum & XFS_SB_VERSION_DALIGNBIT);
-}
-
-static inline int xfs_sb_version_hasshared(xfs_sb_t *sbp)
+static inline bool xfs_sb_version_hasdalign(struct xfs_sb *sbp)
 {
-	return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
-		(sbp->sb_versionnum & XFS_SB_VERSION_SHAREDBIT);
+	return (sbp->sb_versionnum & XFS_SB_VERSION_DALIGNBIT);
 }
 
-static inline int xfs_sb_version_hasdirv2(xfs_sb_t *sbp)
+static inline bool xfs_sb_version_haslogv2(struct xfs_sb *sbp)
 {
-	return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
-	       (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
-		(sbp->sb_versionnum & XFS_SB_VERSION_DIRV2BIT));
+	return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5 ||
+	       (sbp->sb_versionnum & XFS_SB_VERSION_LOGV2BIT);
 }
 
-static inline int xfs_sb_version_haslogv2(xfs_sb_t *sbp)
+static inline bool xfs_sb_version_hasextflgbit(struct xfs_sb *sbp)
 {
-	return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
-	       (XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
-		(sbp->sb_versionnum & XFS_SB_VERSION_LOGV2BIT));
+	return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5 ||
+	       (sbp->sb_versionnum & XFS_SB_VERSION_EXTFLGBIT);
 }
 
-static inline int xfs_sb_version_hasextflgbit(xfs_sb_t *sbp)
+static inline bool xfs_sb_version_hassector(struct xfs_sb *sbp)
 {
-	return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
-	       (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
-		(sbp->sb_versionnum & XFS_SB_VERSION_EXTFLGBIT));
+	return (sbp->sb_versionnum & XFS_SB_VERSION_SECTORBIT);
 }
 
-static inline int xfs_sb_version_hassector(xfs_sb_t *sbp)
+static inline bool xfs_sb_version_hasasciici(struct xfs_sb *sbp)
 {
-	return XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
-		(sbp->sb_versionnum & XFS_SB_VERSION_SECTORBIT);
+	return (sbp->sb_versionnum & XFS_SB_VERSION_BORGBIT);
 }
 
-static inline int xfs_sb_version_hasasciici(xfs_sb_t *sbp)
+static inline bool xfs_sb_version_hasmorebits(struct xfs_sb *sbp)
 {
-	return XFS_SB_VERSION_NUM(sbp) >= XFS_SB_VERSION_4 &&
-		(sbp->sb_versionnum & XFS_SB_VERSION_BORGBIT);
-}
-
-static inline int xfs_sb_version_hasmorebits(xfs_sb_t *sbp)
-{
-	return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
-	       (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 &&
-		(sbp->sb_versionnum & XFS_SB_VERSION_MOREBITSBIT));
+	return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5 ||
+	       (sbp->sb_versionnum & XFS_SB_VERSION_MOREBITSBIT);
 }
 
 /*
  * sb_features2 bit version macros.
- *
- * For example, for a bit defined as XFS_SB_VERSION2_FUNBIT, has a macro:
- *
- * SB_VERSION_HASFUNBIT(xfs_sb_t *sbp)
- *	((xfs_sb_version_hasmorebits(sbp) &&
- *	 ((sbp)->sb_features2 & XFS_SB_VERSION2_FUNBIT)
  */
-
-static inline int xfs_sb_version_haslazysbcount(xfs_sb_t *sbp)
+static inline bool xfs_sb_version_haslazysbcount(struct xfs_sb *sbp)
 {
 	return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
 	       (xfs_sb_version_hasmorebits(sbp) &&
 		(sbp->sb_features2 & XFS_SB_VERSION2_LAZYSBCOUNTBIT));
 }
 
-static inline int xfs_sb_version_hasattr2(xfs_sb_t *sbp)
+static inline bool xfs_sb_version_hasattr2(struct xfs_sb *sbp)
 {
 	return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
 	       (xfs_sb_version_hasmorebits(sbp) &&
 		(sbp->sb_features2 & XFS_SB_VERSION2_ATTR2BIT));
 }
 
-static inline void xfs_sb_version_addattr2(xfs_sb_t *sbp)
+static inline void xfs_sb_version_addattr2(struct xfs_sb *sbp)
 {
 	sbp->sb_versionnum |= XFS_SB_VERSION_MOREBITSBIT;
 	sbp->sb_features2 |= XFS_SB_VERSION2_ATTR2BIT;
+	sbp->sb_bad_features2 |= XFS_SB_VERSION2_ATTR2BIT;
 }
 
-static inline void xfs_sb_version_removeattr2(xfs_sb_t *sbp)
+static inline void xfs_sb_version_removeattr2(struct xfs_sb *sbp)
 {
 	sbp->sb_features2 &= ~XFS_SB_VERSION2_ATTR2BIT;
+	sbp->sb_bad_features2 &= ~XFS_SB_VERSION2_ATTR2BIT;
 	if (!sbp->sb_features2)
 		sbp->sb_versionnum &= ~XFS_SB_VERSION_MOREBITSBIT;
 }
 
-static inline int xfs_sb_version_hasprojid32bit(xfs_sb_t *sbp)
+static inline bool xfs_sb_version_hasprojid32bit(struct xfs_sb *sbp)
 {
 	return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) ||
 	       (xfs_sb_version_hasmorebits(sbp) &&
 		(sbp->sb_features2 & XFS_SB_VERSION2_PROJID32BIT));
 }
 
-static inline void xfs_sb_version_addprojid32bit(xfs_sb_t *sbp)
+static inline void xfs_sb_version_addprojid32bit(struct xfs_sb *sbp)
 {
 	sbp->sb_versionnum |= XFS_SB_VERSION_MOREBITSBIT;
 	sbp->sb_features2 |= XFS_SB_VERSION2_PROJID32BIT;
@@ -585,7 +496,9 @@ xfs_sb_has_compat_feature(
 	return (sbp->sb_features_compat & feature) != 0;
 }
 
-#define XFS_SB_FEAT_RO_COMPAT_ALL 0
+#define XFS_SB_FEAT_RO_COMPAT_FINOBT   (1 << 0)		/* free inode btree */
+#define XFS_SB_FEAT_RO_COMPAT_ALL \
+		(XFS_SB_FEAT_RO_COMPAT_FINOBT)
 #define XFS_SB_FEAT_RO_COMPAT_UNKNOWN	~XFS_SB_FEAT_RO_COMPAT_ALL
 static inline bool
 xfs_sb_has_ro_compat_feature(
@@ -621,12 +534,12 @@ xfs_sb_has_incompat_log_feature(
 /*
  * V5 superblock specific feature checks
  */
-static inline int xfs_sb_version_hascrc(xfs_sb_t *sbp)
+static inline int xfs_sb_version_hascrc(struct xfs_sb *sbp)
 {
 	return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5;
 }
 
-static inline int xfs_sb_version_has_pquotino(xfs_sb_t *sbp)
+static inline int xfs_sb_version_has_pquotino(struct xfs_sb *sbp)
 {
 	return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5;
 }
@@ -639,6 +552,12 @@ static inline int xfs_sb_version_hasftype(struct xfs_sb *sbp)
 		 (sbp->sb_features2 & XFS_SB_VERSION2_FTYPE));
 }
 
+static inline int xfs_sb_version_hasfinobt(xfs_sb_t *sbp)
+{
+	return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5) &&
+		(sbp->sb_features_ro_compat & XFS_SB_FEAT_RO_COMPAT_FINOBT);
+}
+
 /*
  * end of superblock version macros
  */
diff --git a/fs/xfs/xfs_shared.h b/fs/xfs/libxfs/xfs_shared.h
index 8c5035a13df1..82404da2ca67 100644
--- a/fs/xfs/xfs_shared.h
+++ b/fs/xfs/libxfs/xfs_shared.h
@@ -104,7 +104,8 @@ extern const struct xfs_buf_ops xfs_symlink_buf_ops;
 #define	XFS_TRANS_SB_COUNT		41
 #define	XFS_TRANS_CHECKPOINT		42
 #define	XFS_TRANS_ICREATE		43
-#define	XFS_TRANS_TYPE_MAX		43
+#define	XFS_TRANS_CREATE_TMPFILE	44
+#define	XFS_TRANS_TYPE_MAX		44
 /* new transaction types need to be reflected in xfs_logprint(8) */
 
 #define XFS_TRANS_TYPES \
@@ -112,6 +113,7 @@ extern const struct xfs_buf_ops xfs_symlink_buf_ops;
 	{ XFS_TRANS_SETATTR_SIZE,	"SETATTR_SIZE" }, \
 	{ XFS_TRANS_INACTIVE,		"INACTIVE" }, \
 	{ XFS_TRANS_CREATE,		"CREATE" }, \
+	{ XFS_TRANS_CREATE_TMPFILE,	"CREATE_TMPFILE" }, \
 	{ XFS_TRANS_CREATE_TRUNC,	"CREATE_TRUNC" }, \
 	{ XFS_TRANS_TRUNCATE_FILE,	"TRUNCATE_FILE" }, \
 	{ XFS_TRANS_REMOVE,		"REMOVE" }, \
@@ -236,7 +238,7 @@ int	xfs_log_calc_minimum_size(struct xfs_mount *);
 int xfs_symlink_blocks(struct xfs_mount *mp, int pathlen);
 int xfs_symlink_hdr_set(struct xfs_mount *mp, xfs_ino_t ino, uint32_t offset,
 			uint32_t size, struct xfs_buf *bp);
-bool xfs_symlink_hdr_ok(struct xfs_mount *mp, xfs_ino_t ino, uint32_t offset,
+bool xfs_symlink_hdr_ok(xfs_ino_t ino, uint32_t offset,
 			uint32_t size, struct xfs_buf *bp);
 void xfs_symlink_local_to_remote(struct xfs_trans *tp, struct xfs_buf *bp,
 				 struct xfs_inode *ip, struct xfs_ifork *ifp);
diff --git a/fs/xfs/xfs_symlink_remote.c b/fs/xfs/libxfs/xfs_symlink_remote.c
index bf59a2b45f8c..5782f037eab4 100644
--- a/fs/xfs/xfs_symlink_remote.c
+++ b/fs/xfs/libxfs/xfs_symlink_remote.c
@@ -80,7 +80,6 @@ xfs_symlink_hdr_set(
  */
 bool
 xfs_symlink_hdr_ok(
-	struct xfs_mount	*mp,
 	xfs_ino_t		ino,
 	uint32_t		offset,
 	uint32_t		size,
@@ -133,12 +132,13 @@ xfs_symlink_read_verify(
 	if (!xfs_sb_version_hascrc(&mp->m_sb))
 		return;
 
-	if (!xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
-				  offsetof(struct xfs_dsymlink_hdr, sl_crc)) ||
-	    !xfs_symlink_verify(bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
-	}
+	if (!xfs_buf_verify_cksum(bp, XFS_SYMLINK_CRC_OFF))
+		xfs_buf_ioerror(bp, -EFSBADCRC);
+	else if (!xfs_symlink_verify(bp))
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+
+	if (bp->b_error)
+		xfs_verifier_error(bp);
 }
 
 static void
@@ -153,8 +153,8 @@ xfs_symlink_write_verify(
 		return;
 
 	if (!xfs_symlink_verify(bp)) {
-		XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, bp->b_addr);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
+		xfs_verifier_error(bp);
 		return;
 	}
 
@@ -162,8 +162,7 @@ xfs_symlink_write_verify(
 		struct xfs_dsymlink_hdr *dsl = bp->b_addr;
 		dsl->sl_lsn = cpu_to_be64(bip->bli_item.li_lsn);
 	}
-	xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
-			 offsetof(struct xfs_dsymlink_hdr, sl_crc));
+	xfs_buf_update_cksum(bp, XFS_SYMLINK_CRC_OFF);
 }
 
 const struct xfs_buf_ops xfs_symlink_buf_ops = {
diff --git a/fs/xfs/xfs_trans_resv.c b/fs/xfs/libxfs/xfs_trans_resv.c
index 2fd59c0dae66..f2bda7c76b8a 100644
--- a/fs/xfs/xfs_trans_resv.c
+++ b/fs/xfs/libxfs/xfs_trans_resv.c
@@ -26,6 +26,7 @@
 #include "xfs_ag.h"
 #include "xfs_mount.h"
 #include "xfs_da_format.h"
+#include "xfs_da_btree.h"
 #include "xfs_inode.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_ialloc.h"
@@ -81,20 +82,69 @@ xfs_calc_buf_res(
  * on disk. Hence we need an inode reservation function that calculates all this
  * correctly. So, we log:
  *
- * - log op headers for object
+ * - 4 log op headers for object
+ *	- for the ilf, the inode core and 2 forks
  * - inode log format object
- * - the entire inode contents (core + 2 forks)
- * - two bmap btree block headers
+ * - the inode core
+ * - two inode forks containing bmap btree root blocks.
+ *	- the btree data contained by both forks will fit into the inode size,
+ *	  hence when combined with the inode core above, we have a total of the
+ *	  actual inode size.
+ *	- the BMBT headers need to be accounted separately, as they are
+ *	  additional to the records and pointers that fit inside the inode
+ *	  forks.
  */
 STATIC uint
 xfs_calc_inode_res(
 	struct xfs_mount	*mp,
 	uint			ninodes)
 {
-	return ninodes * (sizeof(struct xlog_op_header) +
-			  sizeof(struct xfs_inode_log_format) +
-			  mp->m_sb.sb_inodesize +
-			  2 * XFS_BMBT_BLOCK_LEN(mp));
+	return ninodes *
+		(4 * sizeof(struct xlog_op_header) +
+		 sizeof(struct xfs_inode_log_format) +
+		 mp->m_sb.sb_inodesize +
+		 2 * XFS_BMBT_BLOCK_LEN(mp));
+}
+
+/*
+ * The free inode btree is a conditional feature and the log reservation
+ * requirements differ slightly from that of the traditional inode allocation
+ * btree. The finobt tracks records for inode chunks with at least one free
+ * inode. A record can be removed from the tree for an inode allocation
+ * or free and thus the finobt reservation is unconditional across:
+ *
+ * 	- inode allocation
+ * 	- inode free
+ * 	- inode chunk allocation
+ *
+ * The 'modify' param indicates to include the record modification scenario. The
+ * 'alloc' param indicates to include the reservation for free space btree
+ * modifications on behalf of finobt modifications. This is required only for
+ * transactions that do not already account for free space btree modifications.
+ *
+ * the free inode btree: max depth * block size
+ * the allocation btrees: 2 trees * (max depth - 1) * block size
+ * the free inode btree entry: block size
+ */
+STATIC uint
+xfs_calc_finobt_res(
+	struct xfs_mount 	*mp,
+	int			alloc,
+	int			modify)
+{
+	uint res;
+
+	if (!xfs_sb_version_hasfinobt(&mp->m_sb))
+		return 0;
+
+	res = xfs_calc_buf_res(mp->m_in_maxlevels, XFS_FSB_TO_B(mp, 1));
+	if (alloc)
+		res += xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1), 
+					XFS_FSB_TO_B(mp, 1));
+	if (modify)
+		res += (uint)XFS_FSB_TO_B(mp, 1);
+
+	return res;
 }
 
 /*
@@ -174,7 +224,7 @@ xfs_calc_itruncate_reservation(
 		    xfs_calc_buf_res(5, 0) +
 		    xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1),
 				     XFS_FSB_TO_B(mp, 1)) +
-		    xfs_calc_buf_res(2 + XFS_IALLOC_BLOCKS(mp) +
+		    xfs_calc_buf_res(2 + mp->m_ialloc_blks +
 				     mp->m_in_maxlevels, 0)));
 }
 
@@ -204,6 +254,19 @@ xfs_calc_rename_reservation(
 }
 
 /*
+ * For removing an inode from unlinked list at first, we can modify:
+ *    the agi hash list and counters: sector size
+ *    the on disk inode before ours in the agi hash list: inode cluster size
+ */
+STATIC uint
+xfs_calc_iunlink_remove_reservation(
+	struct xfs_mount        *mp)
+{
+	return xfs_calc_buf_res(1, mp->m_sb.sb_sectsize) +
+	       max_t(uint, XFS_FSB_TO_B(mp, 1), mp->m_inode_cluster_size);
+}
+
+/*
  * For creating a link to an inode:
  *    the parent directory inode: inode size
  *    the linked inode: inode size
@@ -220,6 +283,7 @@ xfs_calc_link_reservation(
 	struct xfs_mount	*mp)
 {
 	return XFS_DQUOT_LOGRES(mp) +
+		xfs_calc_iunlink_remove_reservation(mp) +
 		MAX((xfs_calc_inode_res(mp, 2) +
 		     xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp),
 				      XFS_FSB_TO_B(mp, 1))),
@@ -229,6 +293,18 @@ xfs_calc_link_reservation(
 }
 
 /*
+ * For adding an inode to unlinked list we can modify:
+ *    the agi hash list: sector size
+ *    the unlinked inode: inode size
+ */
+STATIC uint
+xfs_calc_iunlink_add_reservation(xfs_mount_t *mp)
+{
+	return xfs_calc_buf_res(1, mp->m_sb.sb_sectsize) +
+		xfs_calc_inode_res(mp, 1);
+}
+
+/*
  * For removing a directory entry we can modify:
  *    the parent directory inode: inode size
  *    the removed inode: inode size
@@ -245,10 +321,11 @@ xfs_calc_remove_reservation(
 	struct xfs_mount	*mp)
 {
 	return XFS_DQUOT_LOGRES(mp) +
-		MAX((xfs_calc_inode_res(mp, 2) +
+		xfs_calc_iunlink_add_reservation(mp) +
+		MAX((xfs_calc_inode_res(mp, 1) +
 		     xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp),
 				      XFS_FSB_TO_B(mp, 1))),
-		    (xfs_calc_buf_res(5, mp->m_sb.sb_sectsize) +
+		    (xfs_calc_buf_res(4, mp->m_sb.sb_sectsize) +
 		     xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 2),
 				      XFS_FSB_TO_B(mp, 1))));
 }
@@ -267,6 +344,7 @@ xfs_calc_remove_reservation(
  *    the superblock for the nlink flag: sector size
  *    the directory btree: (max depth + v2) * dir block size
  *    the directory inode's bmap btree: (max depth + v2) * block size
+ *    the finobt (record modification and allocation btrees)
  */
 STATIC uint
 xfs_calc_create_resv_modify(
@@ -275,14 +353,15 @@ xfs_calc_create_resv_modify(
 	return xfs_calc_inode_res(mp, 2) +
 		xfs_calc_buf_res(1, mp->m_sb.sb_sectsize) +
 		(uint)XFS_FSB_TO_B(mp, 1) +
-		xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp), XFS_FSB_TO_B(mp, 1));
+		xfs_calc_buf_res(XFS_DIROP_LOG_COUNT(mp), XFS_FSB_TO_B(mp, 1)) +
+		xfs_calc_finobt_res(mp, 1, 1);
 }
 
 /*
  * For create we can allocate some inodes giving:
  *    the agi and agf of the ag getting the new inodes: 2 * sectorsize
  *    the superblock for the nlink flag: sector size
- *    the inode blocks allocated: XFS_IALLOC_BLOCKS * blocksize
+ *    the inode blocks allocated: mp->m_ialloc_blks * blocksize
  *    the inode btree: max depth * blocksize
  *    the allocation btrees: 2 trees * (max depth - 1) * block size
  */
@@ -292,7 +371,7 @@ xfs_calc_create_resv_alloc(
 {
 	return xfs_calc_buf_res(2, mp->m_sb.sb_sectsize) +
 		mp->m_sb.sb_sectsize +
-		xfs_calc_buf_res(XFS_IALLOC_BLOCKS(mp), XFS_FSB_TO_B(mp, 1)) +
+		xfs_calc_buf_res(mp->m_ialloc_blks, XFS_FSB_TO_B(mp, 1)) +
 		xfs_calc_buf_res(mp->m_in_maxlevels, XFS_FSB_TO_B(mp, 1)) +
 		xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1),
 				 XFS_FSB_TO_B(mp, 1));
@@ -313,6 +392,7 @@ __xfs_calc_create_reservation(
  *    the superblock for the nlink flag: sector size
  *    the inode btree: max depth * blocksize
  *    the allocation btrees: 2 trees * (max depth - 1) * block size
+ *    the finobt (record insertion)
  */
 STATIC uint
 xfs_calc_icreate_resv_alloc(
@@ -322,7 +402,8 @@ xfs_calc_icreate_resv_alloc(
 		mp->m_sb.sb_sectsize +
 		xfs_calc_buf_res(mp->m_in_maxlevels, XFS_FSB_TO_B(mp, 1)) +
 		xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1),
-				 XFS_FSB_TO_B(mp, 1));
+				 XFS_FSB_TO_B(mp, 1)) +
+		xfs_calc_finobt_res(mp, 0, 0);
 }
 
 STATIC uint
@@ -343,6 +424,20 @@ xfs_calc_create_reservation(
 
 }
 
+STATIC uint
+xfs_calc_create_tmpfile_reservation(
+	struct xfs_mount        *mp)
+{
+	uint	res = XFS_DQUOT_LOGRES(mp);
+
+	if (xfs_sb_version_hascrc(&mp->m_sb))
+		res += xfs_calc_icreate_resv_alloc(mp);
+	else
+		res += xfs_calc_create_resv_alloc(mp);
+
+	return res + xfs_calc_iunlink_add_reservation(mp);
+}
+
 /*
  * Making a new directory is the same as creating a new file.
  */
@@ -376,6 +471,7 @@ xfs_calc_symlink_reservation(
  *    the on disk inode before ours in the agi hash list: inode cluster size
  *    the inode btree: max depth * blocksize
  *    the allocation btrees: 2 trees * (max depth - 1) * block size
+ *    the finobt (record insertion, removal or modification)
  */
 STATIC uint
 xfs_calc_ifree_reservation(
@@ -383,14 +479,15 @@ xfs_calc_ifree_reservation(
 {
 	return XFS_DQUOT_LOGRES(mp) +
 		xfs_calc_inode_res(mp, 1) +
-		xfs_calc_buf_res(2, mp->m_sb.sb_sectsize) +
+		xfs_calc_buf_res(1, mp->m_sb.sb_sectsize) +
 		xfs_calc_buf_res(1, XFS_FSB_TO_B(mp, 1)) +
-		max_t(uint, XFS_FSB_TO_B(mp, 1), XFS_INODE_CLUSTER_SIZE(mp)) +
+		xfs_calc_iunlink_remove_reservation(mp) +
 		xfs_calc_buf_res(1, 0) +
-		xfs_calc_buf_res(2 + XFS_IALLOC_BLOCKS(mp) +
+		xfs_calc_buf_res(2 + mp->m_ialloc_blks +
 				 mp->m_in_maxlevels, 0) +
 		xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1),
-				 XFS_FSB_TO_B(mp, 1));
+				 XFS_FSB_TO_B(mp, 1)) +
+		xfs_calc_finobt_res(mp, 0, 1);
 }
 
 /*
@@ -513,7 +610,7 @@ xfs_calc_addafork_reservation(
 	return XFS_DQUOT_LOGRES(mp) +
 		xfs_calc_inode_res(mp, 1) +
 		xfs_calc_buf_res(2, mp->m_sb.sb_sectsize) +
-		xfs_calc_buf_res(1, mp->m_dirblksize) +
+		xfs_calc_buf_res(1, mp->m_dir_geo->blksize) +
 		xfs_calc_buf_res(XFS_DAENTER_BMAP1B(mp, XFS_DATA_FORK) + 1,
 				 XFS_FSB_TO_B(mp, 1)) +
 		xfs_calc_buf_res(XFS_ALLOCFREE_LOG_COUNT(mp, 1),
@@ -644,15 +741,14 @@ xfs_calc_qm_setqlim_reservation(
 
 /*
  * Allocating quota on disk if needed.
- *	the write transaction log space: M_RES(mp)->tr_write.tr_logres
+ *	the write transaction log space for quota file extent allocation
  *	the unit of quota allocation: one system block size
  */
 STATIC uint
 xfs_calc_qm_dqalloc_reservation(
 	struct xfs_mount	*mp)
 {
-	ASSERT(M_RES(mp)->tr_write.tr_logres);
-	return M_RES(mp)->tr_write.tr_logres +
+	return xfs_calc_write_reservation(mp) +
 		xfs_calc_buf_res(1,
 			XFS_FSB_TO_B(mp, XFS_DQUOT_CLUSTER_SIZE_FSB) - 1);
 }
@@ -729,6 +825,11 @@ xfs_trans_resv_calc(
 	resp->tr_create.tr_logcount = XFS_CREATE_LOG_COUNT;
 	resp->tr_create.tr_logflags |= XFS_TRANS_PERM_LOG_RES;
 
+	resp->tr_create_tmpfile.tr_logres =
+			xfs_calc_create_tmpfile_reservation(mp);
+	resp->tr_create_tmpfile.tr_logcount = XFS_CREATE_TMPFILE_LOG_COUNT;
+	resp->tr_create_tmpfile.tr_logflags |= XFS_TRANS_PERM_LOG_RES;
+
 	resp->tr_mkdir.tr_logres = xfs_calc_mkdir_reservation(mp);
 	resp->tr_mkdir.tr_logcount = XFS_MKDIR_LOG_COUNT;
 	resp->tr_mkdir.tr_logflags |= XFS_TRANS_PERM_LOG_RES;
@@ -784,7 +885,6 @@ xfs_trans_resv_calc(
 	/* The following transaction are logged in logical format */
 	resp->tr_ichange.tr_logres = xfs_calc_ichange_reservation(mp);
 	resp->tr_growdata.tr_logres = xfs_calc_growdata_reservation(mp);
-	resp->tr_swrite.tr_logres = xfs_calc_swrite_reservation(mp);
 	resp->tr_fsyncts.tr_logres = xfs_calc_swrite_reservation(mp);
 	resp->tr_writeid.tr_logres = xfs_calc_writeid_reservation(mp);
 	resp->tr_attrsetrt.tr_logres = xfs_calc_attrsetrt_reservation(mp);
diff --git a/fs/xfs/xfs_trans_resv.h b/fs/xfs/libxfs/xfs_trans_resv.h
index de7de9aaad8a..1097d14cd583 100644
--- a/fs/xfs/xfs_trans_resv.h
+++ b/fs/xfs/libxfs/xfs_trans_resv.h
@@ -38,11 +38,11 @@ struct xfs_trans_resv {
 	struct xfs_trans_res	tr_remove;	/* unlink trans */
 	struct xfs_trans_res	tr_symlink;	/* symlink trans */
 	struct xfs_trans_res	tr_create;	/* create trans */
+	struct xfs_trans_res	tr_create_tmpfile; /* create O_TMPFILE trans */
 	struct xfs_trans_res	tr_mkdir;	/* mkdir trans */
 	struct xfs_trans_res	tr_ifree;	/* inode free trans */
 	struct xfs_trans_res	tr_ichange;	/* inode update trans */
 	struct xfs_trans_res	tr_growdata;	/* fs data section grow trans */
-	struct xfs_trans_res	tr_swrite;	/* sync write inode trans */
 	struct xfs_trans_res	tr_addafork;	/* add inode attr fork trans */
 	struct xfs_trans_res	tr_writeid;	/* write setuid/setgid file */
 	struct xfs_trans_res	tr_attrinval;	/* attr fork buffer
@@ -100,6 +100,7 @@ struct xfs_trans_resv {
 #define	XFS_ITRUNCATE_LOG_COUNT		2
 #define XFS_INACTIVE_LOG_COUNT		2
 #define	XFS_CREATE_LOG_COUNT		2
+#define	XFS_CREATE_TMPFILE_LOG_COUNT	2
 #define	XFS_MKDIR_LOG_COUNT		3
 #define	XFS_SYMLINK_LOG_COUNT		3
 #define	XFS_REMOVE_LOG_COUNT		2
diff --git a/fs/xfs/xfs_trans_space.h b/fs/xfs/libxfs/xfs_trans_space.h
index 7d2c920dfb9c..bf9c4579334d 100644
--- a/fs/xfs/xfs_trans_space.h
+++ b/fs/xfs/libxfs/xfs_trans_space.h
@@ -28,7 +28,8 @@
 	(((b + XFS_MAX_CONTIG_EXTENTS_PER_BLOCK(mp) - 1) / \
 	  XFS_MAX_CONTIG_EXTENTS_PER_BLOCK(mp)) * \
 	  XFS_EXTENTADD_SPACE_RES(mp,w))
-#define	XFS_DAENTER_1B(mp,w)	((w) == XFS_DATA_FORK ? (mp)->m_dirblkfsbs : 1)
+#define	XFS_DAENTER_1B(mp,w)	\
+	((w) == XFS_DATA_FORK ? (mp)->m_dir_geo->fsbcount : 1)
 #define	XFS_DAENTER_DBS(mp,w)	\
 	(XFS_DA_NODE_MAXDEPTH + (((w) == XFS_DATA_FORK) ? 2 : 0))
 #define	XFS_DAENTER_BLOCKS(mp,w)	\
@@ -47,13 +48,15 @@
 #define	XFS_DIRREMOVE_SPACE_RES(mp)	\
 	XFS_DAREMOVE_SPACE_RES(mp, XFS_DATA_FORK)
 #define	XFS_IALLOC_SPACE_RES(mp)	\
-	(XFS_IALLOC_BLOCKS(mp) + (mp)->m_in_maxlevels - 1)
+	((mp)->m_ialloc_blks + \
+	 (xfs_sb_version_hasfinobt(&mp->m_sb) ? 2 : 1 * \
+	  ((mp)->m_in_maxlevels - 1)))
 
 /*
  * Space reservation values for various transactions.
  */
 #define	XFS_ADDAFORK_SPACE_RES(mp)	\
-	((mp)->m_dirblkfsbs + XFS_DAENTER_BMAP1B(mp, XFS_DATA_FORK))
+	((mp)->m_dir_geo->fsbcount + XFS_DAENTER_BMAP1B(mp, XFS_DATA_FORK))
 #define	XFS_ATTRRM_SPACE_RES(mp)	\
 	XFS_DAREMOVE_SPACE_RES(mp, XFS_ATTR_FORK)
 /* This macro is not used - see inline code in xfs_attr_set */
@@ -82,5 +85,8 @@
 	(XFS_DIRREMOVE_SPACE_RES(mp) + XFS_DIRENTER_SPACE_RES(mp,nl))
 #define	XFS_SYMLINK_SPACE_RES(mp,nl,b)	\
 	(XFS_IALLOC_SPACE_RES(mp) + XFS_DIRENTER_SPACE_RES(mp,nl) + (b))
+#define XFS_IFREE_SPACE_RES(mp)		\
+	(xfs_sb_version_hasfinobt(&mp->m_sb) ? (mp)->m_in_maxlevels : 0)
+
 
 #endif	/* __XFS_TRANS_SPACE_H__ */
diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c
index 370eb3e121d1..a65fa5dde6e9 100644
--- a/fs/xfs/xfs_acl.c
+++ b/fs/xfs/xfs_acl.c
@@ -124,16 +124,12 @@ struct posix_acl *
 xfs_get_acl(struct inode *inode, int type)
 {
 	struct xfs_inode *ip = XFS_I(inode);
-	struct posix_acl *acl;
+	struct posix_acl *acl = NULL;
 	struct xfs_acl *xfs_acl;
 	unsigned char *ea_name;
 	int error;
 	int len;
 
-	acl = get_cached_acl(inode, type);
-	if (acl != ACL_NOT_CACHED)
-		return acl;
-
 	trace_xfs_get_acl(ip);
 
 	switch (type) {
@@ -156,7 +152,7 @@ xfs_get_acl(struct inode *inode, int type)
 	if (!xfs_acl)
 		return ERR_PTR(-ENOMEM);
 
-	error = -xfs_attr_get(ip, ea_name, (unsigned char *)xfs_acl,
+	error = xfs_attr_get(ip, ea_name, (unsigned char *)xfs_acl,
 							&len, ATTR_ROOT);
 	if (error) {
 		/*
@@ -164,10 +160,8 @@ xfs_get_acl(struct inode *inode, int type)
 		 * cache entry, for any other error assume it is transient and
 		 * leave the cache entry as ACL_NOT_CACHED.
 		 */
-		if (error == -ENOATTR) {
-			acl = NULL;
+		if (error == -ENOATTR)
 			goto out_update_cache;
-		}
 		goto out;
 	}
 
@@ -183,15 +177,12 @@ out:
 }
 
 STATIC int
-xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
+__xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 {
 	struct xfs_inode *ip = XFS_I(inode);
 	unsigned char *ea_name;
 	int error;
 
-	if (S_ISLNK(inode->i_mode))
-		return -EOPNOTSUPP;
-
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		ea_name = SGI_ACL_FILE;
@@ -219,7 +210,7 @@ xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 		len -= sizeof(struct xfs_acl_entry) *
 			 (XFS_ACL_MAX_ENTRIES(ip->i_mount) - acl->a_count);
 
-		error = -xfs_attr_set(ip, ea_name, (unsigned char *)xfs_acl,
+		error = xfs_attr_set(ip, ea_name, (unsigned char *)xfs_acl,
 				len, ATTR_ROOT);
 
 		kmem_free(xfs_acl);
@@ -227,7 +218,7 @@ xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 		/*
 		 * A NULL ACL argument means we want to remove the ACL.
 		 */
-		error = -xfs_attr_remove(ip, ea_name, ATTR_ROOT);
+		error = xfs_attr_remove(ip, ea_name, ATTR_ROOT);
 
 		/*
 		 * If the attribute didn't exist to start with that's fine.
@@ -253,7 +244,7 @@ xfs_set_mode(struct inode *inode, umode_t mode)
 		iattr.ia_mode = mode;
 		iattr.ia_ctime = current_fs_time(inode->i_sb);
 
-		error = -xfs_setattr_nonsize(XFS_I(inode), &iattr, XFS_ATTR_NOACL);
+		error = xfs_setattr_nonsize(XFS_I(inode), &iattr, XFS_ATTR_NOACL);
 	}
 
 	return error;
@@ -282,131 +273,23 @@ posix_acl_default_exists(struct inode *inode)
 	return xfs_acl_exists(inode, SGI_ACL_DEFAULT);
 }
 
-/*
- * No need for i_mutex because the inode is not yet exposed to the VFS.
- */
 int
-xfs_inherit_acl(struct inode *inode, struct posix_acl *acl)
+xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 {
-	umode_t mode = inode->i_mode;
-	int error = 0, inherit = 0;
-
-	if (S_ISDIR(inode->i_mode)) {
-		error = xfs_set_acl(inode, ACL_TYPE_DEFAULT, acl);
-		if (error)
-			goto out;
-	}
-
-	error = posix_acl_create(&acl, GFP_KERNEL, &mode);
-	if (error < 0)
-		return error;
-
-	/*
-	 * If posix_acl_create returns a positive value we need to
-	 * inherit a permission that can't be represented using the Unix
-	 * mode bits and we actually need to set an ACL.
-	 */
-	if (error > 0)
-		inherit = 1;
-
-	error = xfs_set_mode(inode, mode);
-	if (error)
-		goto out;
-
-	if (inherit)
-		error = xfs_set_acl(inode, ACL_TYPE_ACCESS, acl);
-
-out:
-	posix_acl_release(acl);
-	return error;
-}
-
-int
-xfs_acl_chmod(struct inode *inode)
-{
-	struct posix_acl *acl;
-	int error;
-
-	if (S_ISLNK(inode->i_mode))
-		return -EOPNOTSUPP;
-
-	acl = xfs_get_acl(inode, ACL_TYPE_ACCESS);
-	if (IS_ERR(acl) || !acl)
-		return PTR_ERR(acl);
-
-	error = posix_acl_chmod(&acl, GFP_KERNEL, inode->i_mode);
-	if (error)
-		return error;
-
-	error = xfs_set_acl(inode, ACL_TYPE_ACCESS, acl);
-	posix_acl_release(acl);
-	return error;
-}
-
-static int
-xfs_xattr_acl_get(struct dentry *dentry, const char *name,
-		void *value, size_t size, int type)
-{
-	struct posix_acl *acl;
-	int error;
-
-	acl = xfs_get_acl(dentry->d_inode, type);
-	if (IS_ERR(acl))
-		return PTR_ERR(acl);
-	if (acl == NULL)
-		return -ENODATA;
-
-	error = posix_acl_to_xattr(&init_user_ns, acl, value, size);
-	posix_acl_release(acl);
-
-	return error;
-}
-
-static int
-xfs_xattr_acl_set(struct dentry *dentry, const char *name,
-		const void *value, size_t size, int flags, int type)
-{
-	struct inode *inode = dentry->d_inode;
-	struct posix_acl *acl = NULL;
 	int error = 0;
 
-	if (flags & XATTR_CREATE)
-		return -EINVAL;
-	if (type == ACL_TYPE_DEFAULT && !S_ISDIR(inode->i_mode))
-		return value ? -EACCES : 0;
-	if (!inode_owner_or_capable(inode))
-		return -EPERM;
-
-	if (!value)
+	if (!acl)
 		goto set_acl;
 
-	acl = posix_acl_from_xattr(&init_user_ns, value, size);
-	if (!acl) {
-		/*
-		 * acl_set_file(3) may request that we set default ACLs with
-		 * zero length -- defend (gracefully) against that here.
-		 */
-		goto out;
-	}
-	if (IS_ERR(acl)) {
-		error = PTR_ERR(acl);
-		goto out;
-	}
-
-	error = posix_acl_valid(acl);
-	if (error)
-		goto out_release;
-
-	error = -EINVAL;
+	error = -E2BIG;
 	if (acl->a_count > XFS_ACL_MAX_ENTRIES(XFS_M(inode->i_sb)))
-		goto out_release;
+		return error;
 
 	if (type == ACL_TYPE_ACCESS) {
 		umode_t mode = inode->i_mode;
 		error = posix_acl_equiv_mode(acl, &mode);
 
 		if (error <= 0) {
-			posix_acl_release(acl);
 			acl = NULL;
 
 			if (error < 0)
@@ -415,27 +298,9 @@ xfs_xattr_acl_set(struct dentry *dentry, const char *name,
 
 		error = xfs_set_mode(inode, mode);
 		if (error)
-			goto out_release;
+			return error;
 	}
 
  set_acl:
-	error = xfs_set_acl(inode, type, acl);
- out_release:
-	posix_acl_release(acl);
- out:
-	return error;
+	return __xfs_set_acl(inode, type, acl);
 }
-
-const struct xattr_handler xfs_xattr_acl_access_handler = {
-	.prefix	= POSIX_ACL_XATTR_ACCESS,
-	.flags	= ACL_TYPE_ACCESS,
-	.get	= xfs_xattr_acl_get,
-	.set	= xfs_xattr_acl_set,
-};
-
-const struct xattr_handler xfs_xattr_acl_default_handler = {
-	.prefix	= POSIX_ACL_XATTR_DEFAULT,
-	.flags	= ACL_TYPE_DEFAULT,
-	.get	= xfs_xattr_acl_get,
-	.set	= xfs_xattr_acl_set,
-};
diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h
index 4016a567b83c..5dc163744511 100644
--- a/fs/xfs/xfs_acl.h
+++ b/fs/xfs/xfs_acl.h
@@ -60,20 +60,15 @@ struct xfs_acl {
 
 #ifdef CONFIG_XFS_POSIX_ACL
 extern struct posix_acl *xfs_get_acl(struct inode *inode, int type);
-extern int xfs_inherit_acl(struct inode *inode, struct posix_acl *default_acl);
-extern int xfs_acl_chmod(struct inode *inode);
+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);
-
-extern const struct xattr_handler xfs_xattr_acl_access_handler;
-extern const struct xattr_handler xfs_xattr_acl_default_handler;
 #else
 static inline struct posix_acl *xfs_get_acl(struct inode *inode, int type)
 {
 	return NULL;
 }
-# define xfs_inherit_acl(inode, default_acl)		0
-# define xfs_acl_chmod(inode)				0
+# define xfs_set_acl					NULL
 # define posix_acl_access_exists(inode)			0
 # define posix_acl_default_exists(inode)		0
 #endif /* CONFIG_XFS_POSIX_ACL */
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index 71c8c9d2b882..b984647c24db 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -240,7 +240,7 @@ xfs_end_io(
 
 done:
 	if (error)
-		ioend->io_error = -error;
+		ioend->io_error = error;
 	xfs_destroy_ioend(ioend);
 }
 
@@ -308,14 +308,14 @@ xfs_map_blocks(
 	int			nimaps = 1;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+		return -EIO;
 
 	if (type == XFS_IO_UNWRITTEN)
 		bmapi_flags |= XFS_BMAPI_IGSTATE;
 
 	if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED)) {
 		if (nonblocking)
-			return -XFS_ERROR(EAGAIN);
+			return -EAGAIN;
 		xfs_ilock(ip, XFS_ILOCK_SHARED);
 	}
 
@@ -332,14 +332,14 @@ xfs_map_blocks(
 	xfs_iunlock(ip, XFS_ILOCK_SHARED);
 
 	if (error)
-		return -XFS_ERROR(error);
+		return error;
 
 	if (type == XFS_IO_DELALLOC &&
 	    (!nimaps || isnullstartblock(imap->br_startblock))) {
 		error = xfs_iomap_write_allocate(ip, offset, imap);
 		if (!error)
 			trace_xfs_map_blocks_alloc(ip, offset, count, type, imap);
-		return -XFS_ERROR(error);
+		return error;
 	}
 
 #ifdef DEBUG
@@ -407,7 +407,7 @@ xfs_alloc_ioend_bio(
 	struct bio		*bio = bio_alloc(GFP_NOIO, nvecs);
 
 	ASSERT(bio->bi_private == NULL);
-	bio->bi_sector = bh->b_blocknr * (bh->b_size >> 9);
+	bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9);
 	bio->bi_bdev = bh->b_bdev;
 	return bio;
 }
@@ -502,7 +502,7 @@ xfs_submit_ioend(
 		 * time.
 		 */
 		if (fail) {
-			ioend->io_error = -fail;
+			ioend->io_error = fail;
 			xfs_finish_ioend(ioend);
 			continue;
 		}
@@ -632,38 +632,46 @@ xfs_map_at_offset(
 }
 
 /*
- * Test if a given page is suitable for writing as part of an unwritten
- * or delayed allocate extent.
+ * Test if a given page contains at least one buffer of a given @type.
+ * If @check_all_buffers is true, then we walk all the buffers in the page to
+ * try to find one of the type passed in. If it is not set, then the caller only
+ * needs to check the first buffer on the page for a match.
  */
-STATIC int
+STATIC bool
 xfs_check_page_type(
 	struct page		*page,
-	unsigned int		type)
+	unsigned int		type,
+	bool			check_all_buffers)
 {
-	if (PageWriteback(page))
-		return 0;
+	struct buffer_head	*bh;
+	struct buffer_head	*head;
 
-	if (page->mapping && page_has_buffers(page)) {
-		struct buffer_head	*bh, *head;
-		int			acceptable = 0;
+	if (PageWriteback(page))
+		return false;
+	if (!page->mapping)
+		return false;
+	if (!page_has_buffers(page))
+		return false;
 
-		bh = head = page_buffers(page);
-		do {
-			if (buffer_unwritten(bh))
-				acceptable += (type == XFS_IO_UNWRITTEN);
-			else if (buffer_delay(bh))
-				acceptable += (type == XFS_IO_DELALLOC);
-			else if (buffer_dirty(bh) && buffer_mapped(bh))
-				acceptable += (type == XFS_IO_OVERWRITE);
-			else
-				break;
-		} while ((bh = bh->b_this_page) != head);
+	bh = head = page_buffers(page);
+	do {
+		if (buffer_unwritten(bh)) {
+			if (type == XFS_IO_UNWRITTEN)
+				return true;
+		} else if (buffer_delay(bh)) {
+			if (type == XFS_IO_DELALLOC)
+				return true;
+		} else if (buffer_dirty(bh) && buffer_mapped(bh)) {
+			if (type == XFS_IO_OVERWRITE)
+				return true;
+		}
 
-		if (acceptable)
-			return 1;
-	}
+		/* If we are only checking the first buffer, we are done now. */
+		if (!check_all_buffers)
+			break;
+	} while ((bh = bh->b_this_page) != head);
 
-	return 0;
+	return false;
 }
 
 /*
@@ -697,7 +705,7 @@ xfs_convert_page(
 		goto fail_unlock_page;
 	if (page->mapping != inode->i_mapping)
 		goto fail_unlock_page;
-	if (!xfs_check_page_type(page, (*ioendp)->io_type))
+	if (!xfs_check_page_type(page, (*ioendp)->io_type, false))
 		goto fail_unlock_page;
 
 	/*
@@ -742,6 +750,15 @@ xfs_convert_page(
 	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)
@@ -750,7 +767,7 @@ xfs_convert_page(
 			uptodate = 0;
 		if (!(PageUptodate(page) || buffer_uptodate(bh))) {
 			done = 1;
-			continue;
+			break;
 		}
 
 		if (buffer_unwritten(bh) || buffer_delay(bh) ||
@@ -762,10 +779,11 @@ xfs_convert_page(
 			else
 				type = XFS_IO_OVERWRITE;
 
-			if (!xfs_imap_valid(inode, imap, offset)) {
-				done = 1;
-				continue;
-			}
+			/*
+			 * 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)
@@ -777,6 +795,7 @@ xfs_convert_page(
 			count++;
 		} else {
 			done = 1;
+			break;
 		}
 	} while (offset += len, (bh = bh->b_this_page) != head);
 
@@ -868,7 +887,7 @@ xfs_aops_discard_page(
 	struct buffer_head	*bh, *head;
 	loff_t			offset = page_offset(page);
 
-	if (!xfs_check_page_type(page, XFS_IO_DELALLOC))
+	if (!xfs_check_page_type(page, XFS_IO_DELALLOC, true))
 		goto out_invalidate;
 
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
@@ -956,14 +975,39 @@ xfs_vm_writepage(
 	 * Given that we do not allow direct reclaim to call us, we should
 	 * never be called while in a filesystem transaction.
 	 */
-	if (WARN_ON(current->flags & PF_FSTRANS))
+	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;
-	if (page->index >= end_index) {
+
+	/*
+	 * The page index is less than the end_index, adjust the end_offset
+	 * to the highest offset that this page should represent.
+	 * -----------------------------------------------------
+	 * |			file mapping	       | <EOF> |
+	 * -----------------------------------------------------
+	 * | Page ... | Page N-2 | Page N-1 |  Page N  |       |
+	 * ^--------------------------------^----------|--------
+	 * |     desired writeback range    |      see else    |
+	 * ---------------------------------^------------------|
+	 */
+	if (page->index < end_index)
+		end_offset = (xfs_off_t)(page->index + 1) << PAGE_CACHE_SHIFT;
+	else {
+		/*
+		 * Check whether the page to write out is beyond or straddles
+		 * i_size or not.
+		 * -------------------------------------------------------
+		 * |		file mapping		        | <EOF>  |
+		 * -------------------------------------------------------
+		 * | Page ... | Page N-2 | Page N-1 |  Page N   | Beyond |
+		 * ^--------------------------------^-----------|---------
+		 * |				    |      Straddles     |
+		 * ---------------------------------^-----------|--------|
+		 */
 		unsigned offset_into_page = offset & (PAGE_CACHE_SIZE - 1);
 
 		/*
@@ -971,24 +1015,36 @@ xfs_vm_writepage(
 		 * truncate operation that is in progress. We must redirty the
 		 * page so that reclaim stops reclaiming it. Otherwise
 		 * xfs_vm_releasepage() is called on it and gets confused.
+		 *
+		 * Note that the end_index is unsigned long, it would overflow
+		 * if the given offset is greater than 16TB on 32-bit system
+		 * and if we do check the page is fully outside i_size or not
+		 * via "if (page->index >= end_index + 1)" as "end_index + 1"
+		 * will be evaluated to 0.  Hence this page will be redirtied
+		 * and be written out repeatedly which would result in an
+		 * infinite loop, the user program that perform this operation
+		 * will hang.  Instead, we can verify this situation by checking
+		 * if the page to write is totally beyond the i_size or if it's
+		 * offset is just equal to the EOF.
 		 */
-		if (page->index >= end_index + 1 || offset_into_page == 0)
+		if (page->index > end_index ||
+		    (page->index == end_index && offset_into_page == 0))
 			goto redirty;
 
 		/*
 		 * The page straddles i_size.  It must be zeroed out on each
 		 * and every writepage invocation because it may be mmapped.
 		 * "A file is mapped in multiples of the page size.  For a file
-		 * that is not a multiple of the  page size, the remaining
+		 * that is not a multiple of the page size, the remaining
 		 * 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);
+
+		/* Adjust the end_offset to the end of file */
+		end_offset = offset;
 	}
 
-	end_offset = min_t(unsigned long long,
-			(xfs_off_t)(page->index + 1) << PAGE_CACHE_SHIFT,
-			offset);
 	len = 1 << inode->i_blkbits;
 
 	bh = head = page_buffers(page);
@@ -1169,9 +1225,9 @@ xfs_vm_releasepage(
 
 	xfs_count_page_state(page, &delalloc, &unwritten);
 
-	if (WARN_ON(delalloc))
+	if (WARN_ON_ONCE(delalloc))
 		return 0;
-	if (WARN_ON(unwritten))
+	if (WARN_ON_ONCE(unwritten))
 		return 0;
 
 	return try_to_free_buffers(page);
@@ -1197,7 +1253,7 @@ __xfs_get_blocks(
 	int			new = 0;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+		return -EIO;
 
 	offset = (xfs_off_t)iblock << inode->i_blkbits;
 	ASSERT(bh_result->b_size >= (1 << inode->i_blkbits));
@@ -1217,7 +1273,7 @@ __xfs_get_blocks(
 		lockmode = XFS_ILOCK_EXCL;
 		xfs_ilock(ip, lockmode);
 	} else {
-		lockmode = xfs_ilock_map_shared(ip);
+		lockmode = xfs_ilock_data_map_shared(ip);
 	}
 
 	ASSERT(offset <= mp->m_super->s_maxbytes);
@@ -1246,7 +1302,7 @@ __xfs_get_blocks(
 			error = xfs_iomap_write_direct(ip, offset, size,
 						       &imap, nimaps);
 			if (error)
-				return -error;
+				return error;
 			new = 1;
 		} else {
 			/*
@@ -1325,6 +1381,14 @@ __xfs_get_blocks(
 	/*
 	 * If this is O_DIRECT or the mpage code calling tell them how large
 	 * the mapping is, so that we can avoid repeated get_blocks calls.
+	 *
+	 * If the mapping spans EOF, then we have to break the mapping up as the
+	 * mapping for blocks beyond EOF must be marked new so that sub block
+	 * regions can be correctly zeroed. We can't do this for mappings within
+	 * EOF unless the mapping was just allocated or is unwritten, otherwise
+	 * the callers would overwrite existing data with zeros. Hence we have
+	 * to split the mapping into a range up to and including EOF, and a
+	 * second mapping for beyond EOF.
 	 */
 	if (direct || size > (1 << inode->i_blkbits)) {
 		xfs_off_t		mapping_size;
@@ -1335,6 +1399,12 @@ __xfs_get_blocks(
 		ASSERT(mapping_size > 0);
 		if (mapping_size > size)
 			mapping_size = size;
+		if (offset < i_size_read(inode) &&
+		    offset + mapping_size >= i_size_read(inode)) {
+			/* limit mapping to block that spans EOF */
+			mapping_size = roundup_64(i_size_read(inode) - offset,
+						  1 << inode->i_blkbits);
+		}
 		if (mapping_size > LONG_MAX)
 			mapping_size = LONG_MAX;
 
@@ -1345,7 +1415,7 @@ __xfs_get_blocks(
 
 out_unlock:
 	xfs_iunlock(ip, lockmode);
-	return -error;
+	return error;
 }
 
 int
@@ -1416,9 +1486,8 @@ STATIC ssize_t
 xfs_vm_direct_IO(
 	int			rw,
 	struct kiocb		*iocb,
-	const struct iovec	*iov,
-	loff_t			offset,
-	unsigned long		nr_segs)
+	struct iov_iter		*iter,
+	loff_t			offset)
 {
 	struct inode		*inode = iocb->ki_filp->f_mapping->host;
 	struct block_device	*bdev = xfs_find_bdev_for_inode(inode);
@@ -1426,7 +1495,7 @@ xfs_vm_direct_IO(
 	ssize_t			ret;
 
 	if (rw & WRITE) {
-		size_t size = iov_length(iov, nr_segs);
+		size_t size = iov_iter_count(iter);
 
 		/*
 		 * We cannot preallocate a size update transaction here as we
@@ -1438,16 +1507,15 @@ xfs_vm_direct_IO(
 		if (offset + size > XFS_I(inode)->i_d.di_size)
 			ioend->io_isdirect = 1;
 
-		ret = __blockdev_direct_IO(rw, iocb, inode, bdev, iov,
-					    offset, nr_segs,
-					    xfs_get_blocks_direct,
-					    xfs_end_io_direct_write, NULL, 0);
+		ret = __blockdev_direct_IO(rw, iocb, inode, bdev, iter,
+					    offset, xfs_get_blocks_direct,
+					    xfs_end_io_direct_write, NULL,
+					    DIO_ASYNC_EXTEND);
 		if (ret != -EIOCBQUEUED && iocb->private)
 			goto out_destroy_ioend;
 	} else {
-		ret = __blockdev_direct_IO(rw, iocb, inode, bdev, iov,
-					    offset, nr_segs,
-					    xfs_get_blocks_direct,
+		ret = __blockdev_direct_IO(rw, iocb, inode, bdev, iter,
+					    offset, xfs_get_blocks_direct,
 					    NULL, NULL, 0);
 	}
 
@@ -1546,6 +1614,16 @@ xfs_vm_write_failed(
 
 		xfs_vm_kill_delalloc_range(inode, block_offset,
 					   block_offset + bh->b_size);
+
+		/*
+		 * This buffer does not contain data anymore. make sure anyone
+		 * who finds it knows that for certain.
+		 */
+		clear_buffer_delay(bh);
+		clear_buffer_uptodate(bh);
+		clear_buffer_mapped(bh);
+		clear_buffer_new(bh);
+		clear_buffer_dirty(bh);
 	}
 
 }
@@ -1579,12 +1657,21 @@ xfs_vm_write_begin(
 	status = __block_write_begin(page, pos, len, xfs_get_blocks);
 	if (unlikely(status)) {
 		struct inode	*inode = mapping->host;
+		size_t		isize = i_size_read(inode);
 
 		xfs_vm_write_failed(inode, page, pos, len);
 		unlock_page(page);
 
-		if (pos + len > i_size_read(inode))
-			truncate_pagecache(inode, i_size_read(inode));
+		/*
+		 * If the write is beyond EOF, we only want to kill blocks
+		 * allocated in this write, not blocks that were previously
+		 * written successfully.
+		 */
+		if (pos + len > isize) {
+			ssize_t start = max_t(ssize_t, pos, isize);
+
+			truncate_pagecache_range(inode, start, pos + len);
+		}
 
 		page_cache_release(page);
 		page = NULL;
@@ -1595,9 +1682,12 @@ xfs_vm_write_begin(
 }
 
 /*
- * On failure, we only need to kill delalloc blocks beyond EOF because they
- * will never be written. For blocks within EOF, generic_write_end() zeros them
- * so they are safe to leave alone and be written with all the other valid data.
+ * On failure, we only need to kill delalloc blocks beyond EOF in the range of
+ * this specific write because they will never be written. Previous writes
+ * beyond EOF where block allocation succeeded do not need to be trashed, so
+ * only new blocks from this write should be trashed. For blocks within
+ * EOF, generic_write_end() zeros them so they are safe to leave alone and be
+ * written with all the other valid data.
  */
 STATIC int
 xfs_vm_write_end(
@@ -1620,8 +1710,11 @@ xfs_vm_write_end(
 		loff_t		to = pos + len;
 
 		if (to > isize) {
-			truncate_pagecache(inode, isize);
+			/* only kill blocks in this write beyond EOF */
+			if (pos > isize)
+				isize = pos;
 			xfs_vm_kill_delalloc_range(inode, isize, to);
+			truncate_pagecache_range(inode, isize, to);
 		}
 	}
 	return ret;
@@ -1660,11 +1753,72 @@ xfs_vm_readpages(
 	return mpage_readpages(mapping, pages, nr_pages, xfs_get_blocks);
 }
 
+/*
+ * This is basically a copy of __set_page_dirty_buffers() with one
+ * small tweak: buffers beyond EOF do not get marked dirty. If we mark them
+ * dirty, we'll never be able to clean them because we don't write buffers
+ * beyond EOF, and that means we can't invalidate pages that span EOF
+ * that have been marked dirty. Further, the dirty state can leak into
+ * the file interior if the file is extended, resulting in all sorts of
+ * bad things happening as the state does not match the underlying data.
+ *
+ * XXX: this really indicates that bufferheads in XFS need to die. Warts like
+ * this only exist because of bufferheads and how the generic code manages them.
+ */
+STATIC int
+xfs_vm_set_page_dirty(
+	struct page		*page)
+{
+	struct address_space	*mapping = page->mapping;
+	struct inode		*inode = mapping->host;
+	loff_t			end_offset;
+	loff_t			offset;
+	int			newly_dirty;
+
+	if (unlikely(!mapping))
+		return !TestSetPageDirty(page);
+
+	end_offset = i_size_read(inode);
+	offset = page_offset(page);
+
+	spin_lock(&mapping->private_lock);
+	if (page_has_buffers(page)) {
+		struct buffer_head *head = page_buffers(page);
+		struct buffer_head *bh = head;
+
+		do {
+			if (offset < end_offset)
+				set_buffer_dirty(bh);
+			bh = bh->b_this_page;
+			offset += 1 << inode->i_blkbits;
+		} while (bh != head);
+	}
+	newly_dirty = !TestSetPageDirty(page);
+	spin_unlock(&mapping->private_lock);
+
+	if (newly_dirty) {
+		/* sigh - __set_page_dirty() is static, so copy it here, too */
+		unsigned long flags;
+
+		spin_lock_irqsave(&mapping->tree_lock, flags);
+		if (page->mapping) {	/* Race with truncate? */
+			WARN_ON_ONCE(!PageUptodate(page));
+			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);
+		__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
+	}
+	return newly_dirty;
+}
+
 const struct address_space_operations xfs_address_space_operations = {
 	.readpage		= xfs_vm_readpage,
 	.readpages		= xfs_vm_readpages,
 	.writepage		= xfs_vm_writepage,
 	.writepages		= xfs_vm_writepages,
+	.set_page_dirty		= xfs_vm_set_page_dirty,
 	.releasepage		= xfs_vm_releasepage,
 	.invalidatepage		= xfs_vm_invalidatepage,
 	.write_begin		= xfs_vm_write_begin,
diff --git a/fs/xfs/xfs_attr_inactive.c b/fs/xfs/xfs_attr_inactive.c
index 09480c57f069..aa2a8b1838a2 100644
--- a/fs/xfs/xfs_attr_inactive.c
+++ b/fs/xfs/xfs_attr_inactive.c
@@ -76,7 +76,7 @@ xfs_attr3_leaf_freextent(
 		error = xfs_bmapi_read(dp, (xfs_fileoff_t)tblkno, tblkcnt,
 				       &map, &nmap, XFS_BMAPI_ATTRFORK);
 		if (error) {
-			return(error);
+			return error;
 		}
 		ASSERT(nmap == 1);
 		ASSERT(map.br_startblock != DELAYSTARTBLOCK);
@@ -95,21 +95,21 @@ xfs_attr3_leaf_freextent(
 					dp->i_mount->m_ddev_targp,
 					dblkno, dblkcnt, 0);
 			if (!bp)
-				return ENOMEM;
+				return -ENOMEM;
 			xfs_trans_binval(*trans, bp);
 			/*
 			 * Roll to next transaction.
 			 */
 			error = xfs_trans_roll(trans, dp);
 			if (error)
-				return (error);
+				return error;
 		}
 
 		tblkno += map.br_blockcount;
 		tblkcnt -= map.br_blockcount;
 	}
 
-	return(0);
+	return 0;
 }
 
 /*
@@ -227,7 +227,7 @@ xfs_attr3_node_inactive(
 	 */
 	if (level > XFS_DA_NODE_MAXDEPTH) {
 		xfs_trans_brelse(*trans, bp);	/* no locks for later trans */
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	node = bp->b_addr;
@@ -256,7 +256,7 @@ xfs_attr3_node_inactive(
 		error = xfs_da3_node_read(*trans, dp, child_fsb, -2, &child_bp,
 						XFS_ATTR_FORK);
 		if (error)
-			return(error);
+			return error;
 		if (child_bp) {
 						/* save for re-read later */
 			child_blkno = XFS_BUF_ADDR(child_bp);
@@ -277,7 +277,7 @@ xfs_attr3_node_inactive(
 							child_bp);
 				break;
 			default:
-				error = XFS_ERROR(EIO);
+				error = -EIO;
 				xfs_trans_brelse(*trans, child_bp);
 				break;
 			}
@@ -360,7 +360,7 @@ xfs_attr3_root_inactive(
 		error = xfs_attr3_leaf_inactive(trans, dp, bp);
 		break;
 	default:
-		error = XFS_ERROR(EIO);
+		error = -EIO;
 		xfs_trans_brelse(*trans, bp);
 		break;
 	}
@@ -414,7 +414,7 @@ xfs_attr_inactive(xfs_inode_t *dp)
 	error = xfs_trans_reserve(trans, &M_RES(mp)->tr_attrinval, 0, 0);
 	if (error) {
 		xfs_trans_cancel(trans, 0);
-		return(error);
+		return error;
 	}
 	xfs_ilock(dp, XFS_ILOCK_EXCL);
 
@@ -443,10 +443,10 @@ xfs_attr_inactive(xfs_inode_t *dp)
 	error = xfs_trans_commit(trans, XFS_TRANS_RELEASE_LOG_RES);
 	xfs_iunlock(dp, XFS_ILOCK_EXCL);
 
-	return(error);
+	return error;
 
 out:
 	xfs_trans_cancel(trans, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
 	xfs_iunlock(dp, XFS_ILOCK_EXCL);
-	return(error);
+	return error;
 }
diff --git a/fs/xfs/xfs_attr_list.c b/fs/xfs/xfs_attr_list.c
index 2d174b128153..62db83ab6cbc 100644
--- a/fs/xfs/xfs_attr_list.c
+++ b/fs/xfs/xfs_attr_list.c
@@ -50,11 +50,11 @@ xfs_attr_shortform_compare(const void *a, const void *b)
 	sa = (xfs_attr_sf_sort_t *)a;
 	sb = (xfs_attr_sf_sort_t *)b;
 	if (sa->hash < sb->hash) {
-		return(-1);
+		return -1;
 	} else if (sa->hash > sb->hash) {
-		return(1);
+		return 1;
 	} else {
-		return(sa->entno - sb->entno);
+		return sa->entno - sb->entno;
 	}
 }
 
@@ -86,7 +86,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 	sf = (xfs_attr_shortform_t *)dp->i_afp->if_u1.if_data;
 	ASSERT(sf != NULL);
 	if (!sf->hdr.count)
-		return(0);
+		return 0;
 	cursor = context->cursor;
 	ASSERT(cursor != NULL);
 
@@ -124,7 +124,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 			sfe = XFS_ATTR_SF_NEXTENTRY(sfe);
 		}
 		trace_xfs_attr_list_sf_all(context);
-		return(0);
+		return 0;
 	}
 
 	/* do no more for a search callback */
@@ -150,7 +150,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 					     XFS_ERRLEVEL_LOW,
 					     context->dp->i_mount, sfe);
 			kmem_free(sbuf);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 
 		sbp->entno = i;
@@ -188,7 +188,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 	}
 	if (i == nsbuf) {
 		kmem_free(sbuf);
-		return(0);
+		return 0;
 	}
 
 	/*
@@ -213,7 +213,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 	}
 
 	kmem_free(sbuf);
-	return(0);
+	return 0;
 }
 
 STATIC int
@@ -243,8 +243,8 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
 	if (cursor->blkno > 0) {
 		error = xfs_da3_node_read(NULL, dp, cursor->blkno, -1,
 					      &bp, XFS_ATTR_FORK);
-		if ((error != 0) && (error != EFSCORRUPTED))
-			return(error);
+		if ((error != 0) && (error != -EFSCORRUPTED))
+			return error;
 		if (bp) {
 			struct xfs_attr_leaf_entry *entries;
 
@@ -295,7 +295,7 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
 						      cursor->blkno, -1, &bp,
 						      XFS_ATTR_FORK);
 			if (error)
-				return(error);
+				return error;
 			node = bp->b_addr;
 			magic = be16_to_cpu(node->hdr.info.magic);
 			if (magic == XFS_ATTR_LEAF_MAGIC ||
@@ -308,7 +308,7 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
 						     context->dp->i_mount,
 						     node);
 				xfs_trans_brelse(NULL, bp);
-				return XFS_ERROR(EFSCORRUPTED);
+				return -EFSCORRUPTED;
 			}
 
 			dp->d_ops->node_hdr_from_disk(&nodehdr, node);
@@ -444,9 +444,11 @@ xfs_attr3_leaf_list_int(
 				xfs_da_args_t args;
 
 				memset((char *)&args, 0, sizeof(args));
+				args.geo = context->dp->i_mount->m_attr_geo;
 				args.dp = context->dp;
 				args.whichfork = XFS_ATTR_FORK;
 				args.valuelen = valuelen;
+				args.rmtvaluelen = valuelen;
 				args.value = kmem_alloc(valuelen, KM_SLEEP | KM_NOFS);
 				args.rmtblkno = be32_to_cpu(name_rmt->valueblk);
 				args.rmtblkcnt = xfs_attr3_rmt_blocks(
@@ -494,11 +496,11 @@ xfs_attr_leaf_list(xfs_attr_list_context_t *context)
 	context->cursor->blkno = 0;
 	error = xfs_attr3_leaf_read(NULL, context->dp, 0, -1, &bp);
 	if (error)
-		return XFS_ERROR(error);
+		return error;
 
 	error = xfs_attr3_leaf_list_int(bp, context);
 	xfs_trans_brelse(NULL, bp);
-	return XFS_ERROR(error);
+	return error;
 }
 
 int
@@ -507,17 +509,17 @@ xfs_attr_list_int(
 {
 	int error;
 	xfs_inode_t *dp = context->dp;
+	uint		lock_mode;
 
 	XFS_STATS_INC(xs_attr_list);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return EIO;
-
-	xfs_ilock(dp, XFS_ILOCK_SHARED);
+		return -EIO;
 
 	/*
 	 * Decide on what work routines to call based on the inode size.
 	 */
+	lock_mode = xfs_ilock_attr_map_shared(dp);
 	if (!xfs_inode_hasattr(dp)) {
 		error = 0;
 	} else if (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
@@ -527,9 +529,7 @@ xfs_attr_list_int(
 	} else {
 		error = xfs_attr_node_list(context);
 	}
-
-	xfs_iunlock(dp, XFS_ILOCK_SHARED);
-
+	xfs_iunlock(dp, lock_mode);
 	return error;
 }
 
@@ -616,16 +616,16 @@ xfs_attr_list(
 	 * Validate the cursor.
 	 */
 	if (cursor->pad1 || cursor->pad2)
-		return(XFS_ERROR(EINVAL));
+		return -EINVAL;
 	if ((cursor->initted == 0) &&
 	    (cursor->hashval || cursor->blkno || cursor->offset))
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	/*
 	 * Check for a properly aligned buffer.
 	 */
 	if (((long)buffer) & (sizeof(int)-1))
-		return XFS_ERROR(EFAULT);
+		return -EFAULT;
 	if (flags & ATTR_KERNOVAL)
 		bufsize = 0;
 
@@ -648,6 +648,6 @@ xfs_attr_list(
 	alist->al_offset[0] = context.bufsize;
 
 	error = xfs_attr_list_int(&context);
-	ASSERT(error >= 0);
+	ASSERT(error <= 0);
 	return error;
 }
diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
index 5887e41c0323..1707980f9a4b 100644
--- a/fs/xfs/xfs_bmap_util.c
+++ b/fs/xfs/xfs_bmap_util.c
@@ -133,7 +133,7 @@ xfs_bmap_finish(
 			mp = ntp->t_mountp;
 			if (!XFS_FORCED_SHUTDOWN(mp))
 				xfs_force_shutdown(mp,
-						   (error == EFSCORRUPTED) ?
+						   (error == -EFSCORRUPTED) ?
 						   SHUTDOWN_CORRUPT_INCORE :
 						   SHUTDOWN_META_IO_ERROR);
 			return error;
@@ -249,48 +249,6 @@ xfs_bmap_rtalloc(
 }
 
 /*
- * Stack switching interfaces for allocation
- */
-static void
-xfs_bmapi_allocate_worker(
-	struct work_struct	*work)
-{
-	struct xfs_bmalloca	*args = container_of(work,
-						struct xfs_bmalloca, work);
-	unsigned long		pflags;
-
-	/* we are in a transaction context here */
-	current_set_flags_nested(&pflags, PF_FSTRANS);
-
-	args->result = __xfs_bmapi_allocate(args);
-	complete(args->done);
-
-	current_restore_flags_nested(&pflags, PF_FSTRANS);
-}
-
-/*
- * Some allocation requests often come in with little stack to work on. Push
- * them off to a worker thread so there is lots of stack to use. Otherwise just
- * call directly to avoid the context switch overhead here.
- */
-int
-xfs_bmapi_allocate(
-	struct xfs_bmalloca	*args)
-{
-	DECLARE_COMPLETION_ONSTACK(done);
-
-	if (!args->stack_switch)
-		return __xfs_bmapi_allocate(args);
-
-
-	args->done = &done;
-	INIT_WORK_ONSTACK(&args->work, xfs_bmapi_allocate_worker);
-	queue_work(xfs_alloc_wq, &args->work);
-	wait_for_completion(&done);
-	return args->result;
-}
-
-/*
  * Check if the endoff is outside the last extent. If so the caller will grow
  * the allocation to a stripe unit boundary.  All offsets are considered outside
  * the end of file for an empty fork, so 1 is returned in *eof in that case.
@@ -407,7 +365,7 @@ xfs_bmap_count_tree(
 			xfs_trans_brelse(tp, bp);
 			XFS_ERROR_REPORT("xfs_bmap_count_tree(1)",
 					 XFS_ERRLEVEL_LOW, mp);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 		xfs_trans_brelse(tp, bp);
 	} else {
@@ -467,14 +425,14 @@ xfs_bmap_count_blocks(
 	ASSERT(level > 0);
 	pp = XFS_BMAP_BROOT_PTR_ADDR(mp, block, 1, ifp->if_broot_bytes);
 	bno = be64_to_cpu(*pp);
-	ASSERT(bno != NULLDFSBNO);
+	ASSERT(bno != NULLFSBLOCK);
 	ASSERT(XFS_FSB_TO_AGNO(mp, bno) < mp->m_sb.sb_agcount);
 	ASSERT(XFS_FSB_TO_AGBNO(mp, bno) < mp->m_sb.sb_agblocks);
 
 	if (unlikely(xfs_bmap_count_tree(mp, tp, ifp, bno, level, count) < 0)) {
 		XFS_ERROR_REPORT("xfs_bmap_count_blocks(2)", XFS_ERRLEVEL_LOW,
 				 mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	return 0;
@@ -566,13 +524,13 @@ xfs_getbmap(
 			if (ip->i_d.di_aformat != XFS_DINODE_FMT_EXTENTS &&
 			    ip->i_d.di_aformat != XFS_DINODE_FMT_BTREE &&
 			    ip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)
-				return XFS_ERROR(EINVAL);
+				return -EINVAL;
 		} else if (unlikely(
 			   ip->i_d.di_aformat != 0 &&
 			   ip->i_d.di_aformat != XFS_DINODE_FMT_EXTENTS)) {
 			XFS_ERROR_REPORT("xfs_getbmap", XFS_ERRLEVEL_LOW,
 					 ip->i_mount);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 
 		prealloced = 0;
@@ -581,7 +539,7 @@ xfs_getbmap(
 		if (ip->i_d.di_format != XFS_DINODE_FMT_EXTENTS &&
 		    ip->i_d.di_format != XFS_DINODE_FMT_BTREE &&
 		    ip->i_d.di_format != XFS_DINODE_FMT_LOCAL)
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 
 		if (xfs_get_extsz_hint(ip) ||
 		    ip->i_d.di_flags & (XFS_DIFLAG_PREALLOC|XFS_DIFLAG_APPEND)){
@@ -601,38 +559,43 @@ xfs_getbmap(
 		bmv->bmv_entries = 0;
 		return 0;
 	} else if (bmv->bmv_length < 0) {
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	nex = bmv->bmv_count - 1;
 	if (nex <= 0)
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	bmvend = bmv->bmv_offset + bmv->bmv_length;
 
 
 	if (bmv->bmv_count > ULONG_MAX / sizeof(struct getbmapx))
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 	out = kmem_zalloc_large(bmv->bmv_count * sizeof(struct getbmapx), 0);
 	if (!out)
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 
 	xfs_ilock(ip, XFS_IOLOCK_SHARED);
-	if (whichfork == XFS_DATA_FORK && !(iflags & BMV_IF_DELALLOC)) {
-		if (ip->i_delayed_blks || XFS_ISIZE(ip) > ip->i_d.di_size) {
-			error = -filemap_write_and_wait(VFS_I(ip)->i_mapping);
+	if (whichfork == XFS_DATA_FORK) {
+		if (!(iflags & BMV_IF_DELALLOC) &&
+		    (ip->i_delayed_blks || XFS_ISIZE(ip) > ip->i_d.di_size)) {
+			error = filemap_write_and_wait(VFS_I(ip)->i_mapping);
 			if (error)
 				goto out_unlock_iolock;
+
+			/*
+			 * Even after flushing the inode, there can still be
+			 * delalloc blocks on the inode beyond EOF due to
+			 * speculative preallocation.  These are not removed
+			 * until the release function is called or the inode
+			 * is inactivated.  Hence we cannot assert here that
+			 * ip->i_delayed_blks == 0.
+			 */
 		}
-		/*
-		 * even after flushing the inode, there can still be delalloc
-		 * blocks on the inode beyond EOF due to speculative
-		 * preallocation. These are not removed until the release
-		 * function is called or the inode is inactivated. Hence we
-		 * cannot assert here that ip->i_delayed_blks == 0.
-		 */
-	}
 
-	lock = xfs_ilock_map_shared(ip);
+		lock = xfs_ilock_data_map_shared(ip);
+	} else {
+		lock = xfs_ilock_attr_map_shared(ip);
+	}
 
 	/*
 	 * Don't let nex be bigger than the number of extents
@@ -648,7 +611,7 @@ xfs_getbmap(
 	/*
 	 * Allocate enough space to handle "subnex" maps at a time.
 	 */
-	error = ENOMEM;
+	error = -ENOMEM;
 	subnex = 16;
 	map = kmem_alloc(subnex * sizeof(*map), KM_MAYFAIL | KM_NOFS);
 	if (!map)
@@ -737,7 +700,7 @@ xfs_getbmap(
  out_free_map:
 	kmem_free(map);
  out_unlock_ilock:
-	xfs_iunlock_map_shared(ip, lock);
+	xfs_iunlock(ip, lock);
  out_unlock_iolock:
 	xfs_iunlock(ip, XFS_IOLOCK_SHARED);
 
@@ -846,7 +809,7 @@ xfs_can_free_eofblocks(struct xfs_inode *ip, bool force)
 	 * have speculative prealloc/delalloc blocks to remove.
 	 */
 	if (VFS_I(ip)->i_size == 0 &&
-	    VN_CACHED(VFS_I(ip)) == 0 &&
+	    VFS_I(ip)->i_mapping->nrpages == 0 &&
 	    ip->i_delayed_blks == 0)
 		return false;
 
@@ -919,7 +882,7 @@ xfs_free_eofblocks(
 		if (need_iolock) {
 			if (!xfs_ilock_nowait(ip, XFS_IOLOCK_EXCL)) {
 				xfs_trans_cancel(tp, 0);
-				return EAGAIN;
+				return -EAGAIN;
 			}
 		}
 
@@ -992,14 +955,14 @@ xfs_alloc_file_space(
 	trace_xfs_alloc_file_space(ip);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	error = xfs_qm_dqattach(ip, 0);
 	if (error)
 		return error;
 
 	if (len <= 0)
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	rt = XFS_IS_REALTIME_INODE(ip);
 	extsz = xfs_get_extsz_hint(ip);
@@ -1065,7 +1028,7 @@ xfs_alloc_file_space(
 			/*
 			 * Free the transaction structure.
 			 */
-			ASSERT(error == ENOSPC || XFS_FORCED_SHUTDOWN(mp));
+			ASSERT(error == -ENOSPC || XFS_FORCED_SHUTDOWN(mp));
 			xfs_trans_cancel(tp, 0);
 			break;
 		}
@@ -1102,7 +1065,7 @@ xfs_alloc_file_space(
 		allocated_fsb = imapp->br_blockcount;
 
 		if (nimaps == 0) {
-			error = XFS_ERROR(ENOSPC);
+			error = -ENOSPC;
 			break;
 		}
 
@@ -1163,14 +1126,20 @@ xfs_zero_remaining_bytes(
 					mp->m_rtdev_targp : mp->m_ddev_targp,
 				  BTOBB(mp->m_sb.sb_blocksize), 0);
 	if (!bp)
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 
 	xfs_buf_unlock(bp);
 
 	for (offset = startoff; offset <= endoff; offset = lastoffset + 1) {
+		uint lock_mode;
+
 		offset_fsb = XFS_B_TO_FSBT(mp, offset);
 		nimap = 1;
+
+		lock_mode = xfs_ilock_data_map_shared(ip);
 		error = xfs_bmapi_read(ip, offset_fsb, 1, &imap, &nimap, 0);
+		xfs_iunlock(ip, lock_mode);
+
 		if (error || nimap < 1)
 			break;
 		ASSERT(imap.br_blockcount >= 1);
@@ -1187,7 +1156,12 @@ xfs_zero_remaining_bytes(
 		XFS_BUF_UNWRITE(bp);
 		XFS_BUF_READ(bp);
 		XFS_BUF_SET_ADDR(bp, xfs_fsb_to_db(ip, imap.br_startblock));
-		xfsbdstrat(mp, bp);
+
+		if (XFS_FORCED_SHUTDOWN(mp)) {
+			error = -EIO;
+			break;
+		}
+		xfs_buf_iorequest(bp);
 		error = xfs_buf_iowait(bp);
 		if (error) {
 			xfs_buf_ioerror_alert(bp,
@@ -1200,7 +1174,12 @@ xfs_zero_remaining_bytes(
 		XFS_BUF_UNDONE(bp);
 		XFS_BUF_UNREAD(bp);
 		XFS_BUF_WRITE(bp);
-		xfsbdstrat(mp, bp);
+
+		if (XFS_FORCED_SHUTDOWN(mp)) {
+			error = -EIO;
+			break;
+		}
+		xfs_buf_iorequest(bp);
 		error = xfs_buf_iowait(bp);
 		if (error) {
 			xfs_buf_ioerror_alert(bp,
@@ -1255,7 +1234,7 @@ xfs_free_file_space(
 
 	rounding = max_t(xfs_off_t, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE);
 	ioffset = offset & ~(rounding - 1);
-	error = -filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
+	error = filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
 					      ioffset, -1);
 	if (error)
 		goto out;
@@ -1327,7 +1306,6 @@ xfs_free_file_space(
 		 * the freeing of the space succeeds at ENOSPC.
 		 */
 		tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT);
-		tp->t_flags |= XFS_TRANS_RESERVE;
 		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_write, resblks, 0);
 
 		/*
@@ -1337,7 +1315,7 @@ xfs_free_file_space(
 			/*
 			 * Free the transaction structure.
 			 */
-			ASSERT(error == ENOSPC || XFS_FORCED_SHUTDOWN(mp));
+			ASSERT(error == -ENOSPC || XFS_FORCED_SHUTDOWN(mp));
 			xfs_trans_cancel(tp, 0);
 			break;
 		}
@@ -1397,6 +1375,8 @@ xfs_zero_file_space(
 	xfs_off_t		end_boundary;
 	int			error;
 
+	trace_xfs_zero_file_space(ip);
+
 	granularity = max_t(uint, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE);
 
 	/*
@@ -1411,9 +1391,18 @@ xfs_zero_file_space(
 	ASSERT(end_boundary <= offset + len);
 
 	if (start_boundary < end_boundary - 1) {
-		/* punch out the page cache over the conversion range */
+		/*
+		 * punch out delayed allocation blocks and the page cache over
+		 * the conversion range
+		 */
+		xfs_ilock(ip, XFS_ILOCK_EXCL);
+		error = xfs_bmap_punch_delalloc_range(ip,
+				XFS_B_TO_FSBT(mp, start_boundary),
+				XFS_B_TO_FSB(mp, end_boundary - start_boundary));
+		xfs_iunlock(ip, XFS_ILOCK_EXCL);
 		truncate_pagecache_range(VFS_I(ip), start_boundary,
 					 end_boundary - 1);
+
 		/* convert the blocks */
 		error = xfs_alloc_file_space(ip, start_boundary,
 					end_boundary - start_boundary - 1,
@@ -1446,6 +1435,120 @@ out:
 }
 
 /*
+ * xfs_collapse_file_space()
+ *	This routine frees disk space and shift extent for the given file.
+ *	The first thing we do is to free data blocks in the specified range
+ *	by calling xfs_free_file_space(). It would also sync dirty data
+ *	and invalidate page cache over the region on which collapse range
+ *	is working. And Shift extent records to the left to cover a hole.
+ * RETURNS:
+ *	0 on success
+ *	errno on error
+ *
+ */
+int
+xfs_collapse_file_space(
+	struct xfs_inode	*ip,
+	xfs_off_t		offset,
+	xfs_off_t		len)
+{
+	int			done = 0;
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_trans	*tp;
+	int			error;
+	xfs_extnum_t		current_ext = 0;
+	struct xfs_bmap_free	free_list;
+	xfs_fsblock_t		first_block;
+	int			committed;
+	xfs_fileoff_t		start_fsb;
+	xfs_fileoff_t		shift_fsb;
+
+	ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
+
+	trace_xfs_collapse_file_space(ip);
+
+	start_fsb = XFS_B_TO_FSB(mp, offset + len);
+	shift_fsb = XFS_B_TO_FSB(mp, len);
+
+	/*
+	 * Writeback the entire file and force remove any post-eof blocks. The
+	 * writeback prevents changes to the extent list via concurrent
+	 * writeback and the eofblocks trim prevents the extent shift algorithm
+	 * from running into a post-eof delalloc extent.
+	 *
+	 * XXX: This is a temporary fix until the extent shift loop below is
+	 * converted to use offsets and lookups within the ILOCK rather than
+	 * carrying around the index into the extent list for the next
+	 * iteration.
+	 */
+	error = filemap_write_and_wait(VFS_I(ip)->i_mapping);
+	if (error)
+		return error;
+	if (xfs_can_free_eofblocks(ip, true)) {
+		error = xfs_free_eofblocks(mp, ip, false);
+		if (error)
+			return error;
+	}
+
+	error = xfs_free_file_space(ip, offset, len);
+	if (error)
+		return error;
+
+	while (!error && !done) {
+		tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT);
+		/*
+		 * We would need to reserve permanent block for transaction.
+		 * This will come into picture when after shifting extent into
+		 * hole we found that adjacent extents can be merged which
+		 * may lead to freeing of a block during record update.
+		 */
+		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_write,
+				XFS_DIOSTRAT_SPACE_RES(mp, 0), 0);
+		if (error) {
+			xfs_trans_cancel(tp, 0);
+			break;
+		}
+
+		xfs_ilock(ip, XFS_ILOCK_EXCL);
+		error = xfs_trans_reserve_quota(tp, mp, ip->i_udquot,
+				ip->i_gdquot, ip->i_pdquot,
+				XFS_DIOSTRAT_SPACE_RES(mp, 0), 0,
+				XFS_QMOPT_RES_REGBLKS);
+		if (error)
+			goto out;
+
+		xfs_trans_ijoin(tp, ip, 0);
+
+		xfs_bmap_init(&free_list, &first_block);
+
+		/*
+		 * We are using the write transaction in which max 2 bmbt
+		 * updates are allowed
+		 */
+		error = xfs_bmap_shift_extents(tp, ip, &done, start_fsb,
+					       shift_fsb, &current_ext,
+					       &first_block, &free_list,
+					       XFS_BMAP_MAX_SHIFT_EXTENTS);
+		if (error)
+			goto out;
+
+		error = xfs_bmap_finish(&tp, &free_list, &committed);
+		if (error)
+			goto out;
+
+		error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
+		xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	}
+
+	return error;
+
+out:
+	xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	return error;
+}
+
+/*
  * We need to check that the format of the data fork in the temporary inode is
  * valid for the target inode before doing the swap. This is not a problem with
  * attr1 because of the fixed fork offset, but attr2 has a dynamically sized
@@ -1474,14 +1577,14 @@ xfs_swap_extents_check_format(
 	/* Should never get a local format */
 	if (ip->i_d.di_format == XFS_DINODE_FMT_LOCAL ||
 	    tip->i_d.di_format == XFS_DINODE_FMT_LOCAL)
-		return EINVAL;
+		return -EINVAL;
 
 	/*
 	 * if the target inode has less extents that then temporary inode then
 	 * why did userspace call us?
 	 */
 	if (ip->i_d.di_nextents < tip->i_d.di_nextents)
-		return EINVAL;
+		return -EINVAL;
 
 	/*
 	 * if the target inode is in extent form and the temp inode is in btree
@@ -1490,19 +1593,19 @@ xfs_swap_extents_check_format(
 	 */
 	if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
 	    tip->i_d.di_format == XFS_DINODE_FMT_BTREE)
-		return EINVAL;
+		return -EINVAL;
 
 	/* Check temp in extent form to max in target */
 	if (tip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
 	    XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) >
 			XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK))
-		return EINVAL;
+		return -EINVAL;
 
 	/* Check target in extent form to max in temp */
 	if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
 	    XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) >
 			XFS_IFORK_MAXEXT(tip, XFS_DATA_FORK))
-		return EINVAL;
+		return -EINVAL;
 
 	/*
 	 * If we are in a btree format, check that the temp root block will fit
@@ -1516,26 +1619,50 @@ xfs_swap_extents_check_format(
 	if (tip->i_d.di_format == XFS_DINODE_FMT_BTREE) {
 		if (XFS_IFORK_BOFF(ip) &&
 		    XFS_BMAP_BMDR_SPACE(tip->i_df.if_broot) > XFS_IFORK_BOFF(ip))
-			return EINVAL;
+			return -EINVAL;
 		if (XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) <=
 		    XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK))
-			return EINVAL;
+			return -EINVAL;
 	}
 
 	/* Reciprocal target->temp btree format checks */
 	if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE) {
 		if (XFS_IFORK_BOFF(tip) &&
 		    XFS_BMAP_BMDR_SPACE(ip->i_df.if_broot) > XFS_IFORK_BOFF(tip))
-			return EINVAL;
+			return -EINVAL;
 		if (XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) <=
 		    XFS_IFORK_MAXEXT(tip, XFS_DATA_FORK))
-			return EINVAL;
+			return -EINVAL;
 	}
 
 	return 0;
 }
 
 int
+xfs_swap_extent_flush(
+	struct xfs_inode	*ip)
+{
+	int	error;
+
+	error = filemap_write_and_wait(VFS_I(ip)->i_mapping);
+	if (error)
+		return error;
+	truncate_pagecache_range(VFS_I(ip), 0, -1);
+
+	/* Verify O_DIRECT for ftmp */
+	if (VFS_I(ip)->i_mapping->nrpages)
+		return -EINVAL;
+
+	/*
+	 * Don't try to swap extents on mmap()d files because we can't lock
+	 * out races against page faults safely.
+	 */
+	if (mapping_mapped(VFS_I(ip)->i_mapping))
+		return -EBUSY;
+	return 0;
+}
+
+int
 xfs_swap_extents(
 	xfs_inode_t	*ip,	/* target inode */
 	xfs_inode_t	*tip,	/* tmp inode */
@@ -1550,51 +1677,57 @@ xfs_swap_extents(
 	int		aforkblks = 0;
 	int		taforkblks = 0;
 	__uint64_t	tmp;
+	int		lock_flags;
 
 	tempifp = kmem_alloc(sizeof(xfs_ifork_t), KM_MAYFAIL);
 	if (!tempifp) {
-		error = XFS_ERROR(ENOMEM);
+		error = -ENOMEM;
 		goto out;
 	}
 
 	/*
-	 * we have to do two separate lock calls here to keep lockdep
-	 * happy. If we try to get all the locks in one call, lock will
-	 * report false positives when we drop the ILOCK and regain them
-	 * below.
+	 * Lock up the inodes against other IO and truncate to begin with.
+	 * Then we can ensure the inodes are flushed and have no page cache
+	 * safely. Once we have done this we can take the ilocks and do the rest
+	 * of the checks.
 	 */
+	lock_flags = XFS_IOLOCK_EXCL;
 	xfs_lock_two_inodes(ip, tip, XFS_IOLOCK_EXCL);
-	xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL);
 
 	/* Verify that both files have the same format */
 	if ((ip->i_d.di_mode & S_IFMT) != (tip->i_d.di_mode & S_IFMT)) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_unlock;
 	}
 
 	/* Verify both files are either real-time or non-realtime */
 	if (XFS_IS_REALTIME_INODE(ip) != XFS_IS_REALTIME_INODE(tip)) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_unlock;
 	}
 
-	error = -filemap_write_and_wait(VFS_I(tip)->i_mapping);
+	error = xfs_swap_extent_flush(ip);
+	if (error)
+		goto out_unlock;
+	error = xfs_swap_extent_flush(tip);
 	if (error)
 		goto out_unlock;
-	truncate_pagecache_range(VFS_I(tip), 0, -1);
 
-	/* Verify O_DIRECT for ftmp */
-	if (VN_CACHED(VFS_I(tip)) != 0) {
-		error = XFS_ERROR(EINVAL);
+	tp = xfs_trans_alloc(mp, XFS_TRANS_SWAPEXT);
+	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ichange, 0, 0);
+	if (error) {
+		xfs_trans_cancel(tp, 0);
 		goto out_unlock;
 	}
+	xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL);
+	lock_flags |= XFS_ILOCK_EXCL;
 
 	/* Verify all data are being swapped */
 	if (sxp->sx_offset != 0 ||
 	    sxp->sx_length != ip->i_d.di_size ||
 	    sxp->sx_length != tip->i_d.di_size) {
-		error = XFS_ERROR(EFAULT);
-		goto out_unlock;
+		error = -EFAULT;
+		goto out_trans_cancel;
 	}
 
 	trace_xfs_swap_extent_before(ip, 0);
@@ -1606,7 +1739,7 @@ xfs_swap_extents(
 		xfs_notice(mp,
 		    "%s: inode 0x%llx format is incompatible for exchanging.",
 				__func__, ip->i_ino);
-		goto out_unlock;
+		goto out_trans_cancel;
 	}
 
 	/*
@@ -1620,43 +1753,9 @@ xfs_swap_extents(
 	    (sbp->bs_ctime.tv_nsec != VFS_I(ip)->i_ctime.tv_nsec) ||
 	    (sbp->bs_mtime.tv_sec != VFS_I(ip)->i_mtime.tv_sec) ||
 	    (sbp->bs_mtime.tv_nsec != VFS_I(ip)->i_mtime.tv_nsec)) {
-		error = XFS_ERROR(EBUSY);
-		goto out_unlock;
+		error = -EBUSY;
+		goto out_trans_cancel;
 	}
-
-	/* We need to fail if the file is memory mapped.  Once we have tossed
-	 * all existing pages, the page fault will have no option
-	 * but to go to the filesystem for pages. By making the page fault call
-	 * vop_read (or write in the case of autogrow) they block on the iolock
-	 * until we have switched the extents.
-	 */
-	if (VN_MAPPED(VFS_I(ip))) {
-		error = XFS_ERROR(EBUSY);
-		goto out_unlock;
-	}
-
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	xfs_iunlock(tip, XFS_ILOCK_EXCL);
-
-	/*
-	 * There is a race condition here since we gave up the
-	 * ilock.  However, the data fork will not change since
-	 * we have the iolock (locked for truncation too) so we
-	 * are safe.  We don't really care if non-io related
-	 * fields change.
-	 */
-	truncate_pagecache_range(VFS_I(ip), 0, -1);
-
-	tp = xfs_trans_alloc(mp, XFS_TRANS_SWAPEXT);
-	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ichange, 0, 0);
-	if (error) {
-		xfs_iunlock(ip,  XFS_IOLOCK_EXCL);
-		xfs_iunlock(tip, XFS_IOLOCK_EXCL);
-		xfs_trans_cancel(tp, 0);
-		goto out;
-	}
-	xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL);
-
 	/*
 	 * Count the number of extended attribute blocks
 	 */
@@ -1674,8 +1773,8 @@ xfs_swap_extents(
 			goto out_trans_cancel;
 	}
 
-	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-	xfs_trans_ijoin(tp, tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
+	xfs_trans_ijoin(tp, ip, lock_flags);
+	xfs_trans_ijoin(tp, tip, lock_flags);
 
 	/*
 	 * Before we've swapped the forks, lets set the owners of the forks
@@ -1804,8 +1903,8 @@ out:
 	return error;
 
 out_unlock:
-	xfs_iunlock(ip,  XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
-	xfs_iunlock(tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
+	xfs_iunlock(ip, lock_flags);
+	xfs_iunlock(tip, lock_flags);
 	goto out;
 
 out_trans_cancel:
diff --git a/fs/xfs/xfs_bmap_util.h b/fs/xfs/xfs_bmap_util.h
index 900747b25772..2fdb72d2c908 100644
--- a/fs/xfs/xfs_bmap_util.h
+++ b/fs/xfs/xfs_bmap_util.h
@@ -50,12 +50,11 @@ struct xfs_bmalloca {
 	xfs_extlen_t		total;	/* total blocks needed for xaction */
 	xfs_extlen_t		minlen;	/* minimum allocation size (blocks) */
 	xfs_extlen_t		minleft; /* amount must be left after alloc */
-	char			eof;	/* set if allocating past last extent */
-	char			wasdel;	/* replacing a delayed allocation */
-	char			userdata;/* set if is user data */
-	char			aeof;	/* allocated space at eof */
-	char			conv;	/* overwriting unwritten extents */
-	char			stack_switch;
+	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 */
 	int			flags;
 	struct completion	*done;
 	struct work_struct	work;
@@ -65,8 +64,6 @@ struct xfs_bmalloca {
 int	xfs_bmap_finish(struct xfs_trans **tp, struct xfs_bmap_free *flist,
 			int *committed);
 int	xfs_bmap_rtalloc(struct xfs_bmalloca *ap);
-int	xfs_bmapi_allocate(struct xfs_bmalloca *args);
-int	__xfs_bmapi_allocate(struct xfs_bmalloca *args);
 int	xfs_bmap_eof(struct xfs_inode *ip, xfs_fileoff_t endoff,
 		     int whichfork, int *eof);
 int	xfs_bmap_count_blocks(struct xfs_trans *tp, struct xfs_inode *ip,
@@ -99,6 +96,8 @@ int	xfs_free_file_space(struct xfs_inode *ip, xfs_off_t offset,
 			    xfs_off_t len);
 int	xfs_zero_file_space(struct xfs_inode *ip, xfs_off_t offset,
 			    xfs_off_t len);
+int	xfs_collapse_file_space(struct xfs_inode *, xfs_off_t offset,
+				xfs_off_t len);
 
 /* EOF block manipulation functions */
 bool	xfs_can_free_eofblocks(struct xfs_inode *ip, bool force);
diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c
index c7f0b77dcb00..cd7b8ca9b064 100644
--- a/fs/xfs/xfs_buf.c
+++ b/fs/xfs/xfs_buf.c
@@ -130,7 +130,7 @@ xfs_buf_get_maps(
 	bp->b_maps = kmem_zalloc(map_count * sizeof(struct xfs_buf_map),
 				KM_NOFS);
 	if (!bp->b_maps)
-		return ENOMEM;
+		return -ENOMEM;
 	return 0;
 }
 
@@ -216,8 +216,7 @@ _xfs_buf_alloc(
 STATIC int
 _xfs_buf_get_pages(
 	xfs_buf_t		*bp,
-	int			page_count,
-	xfs_buf_flags_t		flags)
+	int			page_count)
 {
 	/* Make sure that we have a page list */
 	if (bp->b_pages == NULL) {
@@ -330,7 +329,7 @@ use_alloc_page:
 	end = (BBTOB(bp->b_maps[0].bm_bn + bp->b_length) + PAGE_SIZE - 1)
 								>> PAGE_SHIFT;
 	page_count = end - start;
-	error = _xfs_buf_get_pages(bp, page_count, flags);
+	error = _xfs_buf_get_pages(bp, page_count);
 	if (unlikely(error))
 		return error;
 
@@ -345,7 +344,7 @@ retry:
 		if (unlikely(page == NULL)) {
 			if (flags & XBF_READ_AHEAD) {
 				bp->b_page_count = i;
-				error = ENOMEM;
+				error = -ENOMEM;
 				goto out_free_pages;
 			}
 
@@ -396,7 +395,17 @@ _xfs_buf_map_pages(
 		bp->b_addr = NULL;
 	} else {
 		int retried = 0;
+		unsigned noio_flag;
 
+		/*
+		 * vm_map_ram() will allocate auxillary structures (e.g.
+		 * pagetables) with GFP_KERNEL, yet we are likely to be under
+		 * GFP_NOFS context here. Hence we need to tell memory reclaim
+		 * that we are in such a context via PF_MEMALLOC_NOIO to prevent
+		 * memory reclaim re-entering the filesystem here and
+		 * potentially deadlocking.
+		 */
+		noio_flag = memalloc_noio_save();
 		do {
 			bp->b_addr = vm_map_ram(bp->b_pages, bp->b_page_count,
 						-1, PAGE_KERNEL);
@@ -404,6 +413,7 @@ _xfs_buf_map_pages(
 				break;
 			vm_unmap_aliases();
 		} while (retried++ <= 1);
+		memalloc_noio_restore(noio_flag);
 
 		if (!bp->b_addr)
 			return -ENOMEM;
@@ -445,8 +455,8 @@ _xfs_buf_find(
 	numbytes = BBTOB(numblks);
 
 	/* Check for IOs smaller than the sector size / not sector aligned */
-	ASSERT(!(numbytes < (1 << btp->bt_sshift)));
-	ASSERT(!(BBTOB(blkno) & (xfs_off_t)btp->bt_smask));
+	ASSERT(!(numbytes < btp->bt_meta_sectorsize));
+	ASSERT(!(BBTOB(blkno) & (xfs_off_t)btp->bt_meta_sectormask));
 
 	/*
 	 * Corrupted block numbers can get through to here, unfortunately, so we
@@ -455,7 +465,7 @@ _xfs_buf_find(
 	eofs = XFS_FSB_TO_BB(btp->bt_mount, btp->bt_mount->m_sb.sb_dblocks);
 	if (blkno >= eofs) {
 		/*
-		 * XXX (dgc): we should really be returning EFSCORRUPTED here,
+		 * XXX (dgc): we should really be returning -EFSCORRUPTED here,
 		 * but none of the higher level infrastructure supports
 		 * returning a specific error on buffer lookup failures.
 		 */
@@ -698,7 +708,11 @@ xfs_buf_read_uncached(
 	bp->b_flags |= XBF_READ;
 	bp->b_ops = ops;
 
-	xfsbdstrat(target->bt_mount, bp);
+	if (XFS_FORCED_SHUTDOWN(target->bt_mount)) {
+		xfs_buf_relse(bp);
+		return NULL;
+	}
+	xfs_buf_iorequest(bp);
 	xfs_buf_iowait(bp);
 	return bp;
 }
@@ -763,7 +777,7 @@ xfs_buf_associate_memory(
 	bp->b_pages = NULL;
 	bp->b_addr = mem;
 
-	rval = _xfs_buf_get_pages(bp, page_count, 0);
+	rval = _xfs_buf_get_pages(bp, page_count);
 	if (rval)
 		return rval;
 
@@ -796,7 +810,7 @@ xfs_buf_get_uncached(
 		goto fail;
 
 	page_count = PAGE_ALIGN(numblks << BBSHIFT) >> PAGE_SHIFT;
-	error = _xfs_buf_get_pages(bp, page_count, 0);
+	error = _xfs_buf_get_pages(bp, page_count);
 	if (error)
 		goto fail_free_buf;
 
@@ -1038,8 +1052,8 @@ xfs_buf_ioerror(
 	xfs_buf_t		*bp,
 	int			error)
 {
-	ASSERT(error >= 0 && error <= 0xffff);
-	bp->b_error = (unsigned short)error;
+	ASSERT(error <= 0 && error >= -1000);
+	bp->b_error = error;
 	trace_xfs_buf_ioerror(bp, error, _RET_IP_);
 }
 
@@ -1050,7 +1064,7 @@ xfs_buf_ioerror_alert(
 {
 	xfs_alert(bp->b_target->bt_mount,
 "metadata I/O error: block 0x%llx (\"%s\") error %d numblks %d",
-		(__uint64_t)XFS_BUF_ADDR(bp), func, bp->b_error, bp->b_length);
+		(__uint64_t)XFS_BUF_ADDR(bp), func, -bp->b_error, bp->b_length);
 }
 
 /*
@@ -1069,7 +1083,7 @@ xfs_bioerror(
 	/*
 	 * No need to wait until the buffer is unpinned, we aren't flushing it.
 	 */
-	xfs_buf_ioerror(bp, EIO);
+	xfs_buf_ioerror(bp, -EIO);
 
 	/*
 	 * We're calling xfs_buf_ioend, so delete XBF_DONE flag.
@@ -1080,7 +1094,7 @@ xfs_bioerror(
 
 	xfs_buf_ioend(bp, 0);
 
-	return EIO;
+	return -EIO;
 }
 
 /*
@@ -1089,7 +1103,7 @@ xfs_bioerror(
  * This is meant for userdata errors; metadata bufs come with
  * iodone functions attached, so that we can track down errors.
  */
-STATIC int
+int
 xfs_bioerror_relse(
 	struct xfs_buf	*bp)
 {
@@ -1113,13 +1127,13 @@ xfs_bioerror_relse(
 		 * There's no reason to mark error for
 		 * ASYNC buffers.
 		 */
-		xfs_buf_ioerror(bp, EIO);
+		xfs_buf_ioerror(bp, -EIO);
 		complete(&bp->b_iowait);
 	} else {
 		xfs_buf_relse(bp);
 	}
 
-	return EIO;
+	return -EIO;
 }
 
 STATIC int
@@ -1152,7 +1166,7 @@ xfs_bwrite(
 	ASSERT(xfs_buf_islocked(bp));
 
 	bp->b_flags |= XBF_WRITE;
-	bp->b_flags &= ~(XBF_ASYNC | XBF_READ | _XBF_DELWRI_Q);
+	bp->b_flags &= ~(XBF_ASYNC | XBF_READ | _XBF_DELWRI_Q | XBF_WRITE_FAIL);
 
 	xfs_bdstrat_cb(bp);
 
@@ -1164,25 +1178,6 @@ xfs_bwrite(
 	return error;
 }
 
-/*
- * Wrapper around bdstrat so that we can stop data from going to disk in case
- * we are shutting down the filesystem.  Typically user data goes thru this
- * path; one of the exceptions is the superblock.
- */
-void
-xfsbdstrat(
-	struct xfs_mount	*mp,
-	struct xfs_buf		*bp)
-{
-	if (XFS_FORCED_SHUTDOWN(mp)) {
-		trace_xfs_bdstrat_shut(bp, _RET_IP_);
-		xfs_bioerror_relse(bp);
-		return;
-	}
-
-	xfs_buf_iorequest(bp);
-}
-
 STATIC void
 _xfs_buf_ioend(
 	xfs_buf_t		*bp,
@@ -1204,7 +1199,7 @@ xfs_buf_bio_end_io(
 	 * buffers that require multiple bios to complete.
 	 */
 	if (!bp->b_error)
-		xfs_buf_ioerror(bp, -error);
+		xfs_buf_ioerror(bp, error);
 
 	if (!bp->b_error && xfs_buf_is_vmapped(bp) && (bp->b_flags & XBF_READ))
 		invalidate_kernel_vmap_range(bp->b_addr, xfs_buf_vmap_len(bp));
@@ -1255,7 +1250,7 @@ next_chunk:
 
 	bio = bio_alloc(GFP_NOIO, nr_pages);
 	bio->bi_bdev = bp->b_target->bt_bdev;
-	bio->bi_sector = sector;
+	bio->bi_iter.bi_sector = sector;
 	bio->bi_end_io = xfs_buf_bio_end_io;
 	bio->bi_private = bp;
 
@@ -1277,7 +1272,7 @@ next_chunk:
 		total_nr_pages--;
 	}
 
-	if (likely(bio->bi_size)) {
+	if (likely(bio->bi_iter.bi_size)) {
 		if (xfs_buf_is_vmapped(bp)) {
 			flush_kernel_vmap_range(bp->b_addr,
 						xfs_buf_vmap_len(bp));
@@ -1291,7 +1286,7 @@ next_chunk:
 		 * because the caller (xfs_buf_iorequest) holds a count itself.
 		 */
 		atomic_dec(&bp->b_io_remaining);
-		xfs_buf_ioerror(bp, EIO);
+		xfs_buf_ioerror(bp, -EIO);
 		bio_put(bio);
 	}
 
@@ -1335,6 +1330,20 @@ _xfs_buf_ioapply(
 						   SHUTDOWN_CORRUPT_INCORE);
 				return;
 			}
+		} else if (bp->b_bn != XFS_BUF_DADDR_NULL) {
+			struct xfs_mount *mp = bp->b_target->bt_mount;
+
+			/*
+			 * non-crc filesystems don't attach verifiers during
+			 * log recovery, so don't warn for such filesystems.
+			 */
+			if (xfs_sb_version_hascrc(&mp->m_sb)) {
+				xfs_warn(mp,
+					"%s: no ops on block 0x%llx/0x%x",
+					__func__, bp->b_bn, bp->b_length);
+				xfs_hex_dump(bp->b_addr, 64);
+				dump_stack();
+			}
 		}
 	} else if (bp->b_flags & XBF_READ_AHEAD) {
 		rw = READA;
@@ -1376,21 +1385,29 @@ xfs_buf_iorequest(
 		xfs_buf_wait_unpin(bp);
 	xfs_buf_hold(bp);
 
-	/* Set the count to 1 initially, this will stop an I/O
+	/*
+	 * Set the count to 1 initially, this will stop an I/O
 	 * completion callout which happens before we have started
 	 * all the I/O from calling xfs_buf_ioend too early.
 	 */
 	atomic_set(&bp->b_io_remaining, 1);
 	_xfs_buf_ioapply(bp);
-	_xfs_buf_ioend(bp, 1);
+	/*
+	 * If _xfs_buf_ioapply failed, we'll get back here with
+	 * only the reference we took above.  _xfs_buf_ioend will
+	 * drop it to zero, so we'd better not queue it for later,
+	 * or we'll free it before it's done.
+	 */
+	_xfs_buf_ioend(bp, bp->b_error ? 0 : 1);
 
 	xfs_buf_rele(bp);
 }
 
 /*
  * Waits for I/O to complete on the buffer supplied.  It returns immediately if
- * no I/O is pending or there is already a pending error on the buffer.  It
- * returns the I/O error code, if any, or 0 if there was no error.
+ * no I/O is pending or there is already a pending error on the buffer, in which
+ * case nothing will ever complete.  It returns the I/O error code, if any, or
+ * 0 if there was no error.
  */
 int
 xfs_buf_iowait(
@@ -1516,6 +1533,12 @@ xfs_wait_buftarg(
 			struct xfs_buf *bp;
 			bp = list_first_entry(&dispose, struct xfs_buf, b_lru);
 			list_del_init(&bp->b_lru);
+			if (bp->b_flags & XBF_WRITE_FAIL) {
+				xfs_alert(btp->bt_mount,
+"Corruption Alert: Buffer at block 0x%llx had permanent write failures!\n"
+"Please run xfs_repair to determine the extent of the problem.",
+					(long long)bp->b_bn);
+			}
 			xfs_buf_rele(bp);
 		}
 		if (loop++ != 0)
@@ -1602,16 +1625,14 @@ xfs_free_buftarg(
 	kmem_free(btp);
 }
 
-STATIC int
-xfs_setsize_buftarg_flags(
+int
+xfs_setsize_buftarg(
 	xfs_buftarg_t		*btp,
-	unsigned int		blocksize,
-	unsigned int		sectorsize,
-	int			verbose)
+	unsigned int		sectorsize)
 {
-	btp->bt_bsize = blocksize;
-	btp->bt_sshift = ffs(sectorsize) - 1;
-	btp->bt_smask = sectorsize - 1;
+	/* Set up metadata sector size info */
+	btp->bt_meta_sectorsize = sectorsize;
+	btp->bt_meta_sectormask = sectorsize - 1;
 
 	if (set_blocksize(btp->bt_bdev, sectorsize)) {
 		char name[BDEVNAME_SIZE];
@@ -1621,41 +1642,33 @@ xfs_setsize_buftarg_flags(
 		xfs_warn(btp->bt_mount,
 			"Cannot set_blocksize to %u on device %s",
 			sectorsize, name);
-		return EINVAL;
+		return -EINVAL;
 	}
 
+	/* Set up device logical sector size mask */
+	btp->bt_logical_sectorsize = bdev_logical_block_size(btp->bt_bdev);
+	btp->bt_logical_sectormask = bdev_logical_block_size(btp->bt_bdev) - 1;
+
 	return 0;
 }
 
 /*
- *	When allocating the initial buffer target we have not yet
- *	read in the superblock, so don't know what sized sectors
- *	are being used at this early stage.  Play safe.
+ * When allocating the initial buffer target we have not yet
+ * read in the superblock, so don't know what sized sectors
+ * are being used at this early stage.  Play safe.
  */
 STATIC int
 xfs_setsize_buftarg_early(
 	xfs_buftarg_t		*btp,
 	struct block_device	*bdev)
 {
-	return xfs_setsize_buftarg_flags(btp,
-			PAGE_SIZE, bdev_logical_block_size(bdev), 0);
-}
-
-int
-xfs_setsize_buftarg(
-	xfs_buftarg_t		*btp,
-	unsigned int		blocksize,
-	unsigned int		sectorsize)
-{
-	return xfs_setsize_buftarg_flags(btp, blocksize, sectorsize, 1);
+	return xfs_setsize_buftarg(btp, bdev_logical_block_size(bdev));
 }
 
 xfs_buftarg_t *
 xfs_alloc_buftarg(
 	struct xfs_mount	*mp,
-	struct block_device	*bdev,
-	int			external,
-	const char		*fsname)
+	struct block_device	*bdev)
 {
 	xfs_buftarg_t		*btp;
 
@@ -1799,7 +1812,7 @@ __xfs_buf_delwri_submit(
 
 	blk_start_plug(&plug);
 	list_for_each_entry_safe(bp, n, io_list, b_list) {
-		bp->b_flags &= ~(_XBF_DELWRI_Q | XBF_ASYNC);
+		bp->b_flags &= ~(_XBF_DELWRI_Q | XBF_ASYNC | XBF_WRITE_FAIL);
 		bp->b_flags |= XBF_WRITE;
 
 		if (!wait) {
diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h
index e65683361017..c753183900b3 100644
--- a/fs/xfs/xfs_buf.h
+++ b/fs/xfs/xfs_buf.h
@@ -45,6 +45,7 @@ typedef enum {
 #define XBF_ASYNC	 (1 << 4) /* initiator will not wait for completion */
 #define XBF_DONE	 (1 << 5) /* all pages in the buffer uptodate */
 #define XBF_STALE	 (1 << 6) /* buffer has been staled, do not find it */
+#define XBF_WRITE_FAIL	 (1 << 24)/* async writes have failed on this buffer */
 
 /* I/O hints for the BIO layer */
 #define XBF_SYNCIO	 (1 << 10)/* treat this buffer as synchronous I/O */
@@ -70,6 +71,7 @@ typedef unsigned int xfs_buf_flags_t;
 	{ XBF_ASYNC,		"ASYNC" }, \
 	{ XBF_DONE,		"DONE" }, \
 	{ XBF_STALE,		"STALE" }, \
+	{ XBF_WRITE_FAIL,	"WRITE_FAIL" }, \
 	{ XBF_SYNCIO,		"SYNCIO" }, \
 	{ XBF_FUA,		"FUA" }, \
 	{ XBF_FLUSH,		"FLUSH" }, \
@@ -80,19 +82,34 @@ typedef unsigned int xfs_buf_flags_t;
 	{ _XBF_DELWRI_Q,	"DELWRI_Q" }, \
 	{ _XBF_COMPOUND,	"COMPOUND" }
 
+
 /*
  * Internal state flags.
  */
 #define XFS_BSTATE_DISPOSE	 (1 << 0)	/* buffer being discarded */
 
+/*
+ * The xfs_buftarg contains 2 notions of "sector size" -
+ *
+ * 1) The metadata sector size, which is the minimum unit and
+ *    alignment of IO which will be performed by metadata operations.
+ * 2) The device logical sector size
+ *
+ * The first is specified at mkfs time, and is stored on-disk in the
+ * superblock's sb_sectsize.
+ *
+ * The latter is derived from the underlying device, and controls direct IO
+ * alignment constraints.
+ */
 typedef struct xfs_buftarg {
 	dev_t			bt_dev;
 	struct block_device	*bt_bdev;
 	struct backing_dev_info	*bt_bdi;
 	struct xfs_mount	*bt_mount;
-	unsigned int		bt_bsize;
-	unsigned int		bt_sshift;
-	size_t			bt_smask;
+	unsigned int		bt_meta_sectorsize;
+	size_t			bt_meta_sectormask;
+	size_t			bt_logical_sectorsize;
+	size_t			bt_logical_sectormask;
 
 	/* LRU control structures */
 	struct shrinker		bt_shrinker;
@@ -161,7 +178,7 @@ typedef struct xfs_buf {
 	atomic_t		b_io_remaining;	/* #outstanding I/O requests */
 	unsigned int		b_page_count;	/* size of page array */
 	unsigned int		b_offset;	/* page offset in first page */
-	unsigned short		b_error;	/* error code on I/O */
+	int			b_error;	/* error code on I/O */
 	const struct xfs_buf_ops	*b_ops;
 
 #ifdef XFS_BUF_LOCK_TRACKING
@@ -269,9 +286,6 @@ extern void xfs_buf_unlock(xfs_buf_t *);
 
 /* Buffer Read and Write Routines */
 extern int xfs_bwrite(struct xfs_buf *bp);
-
-extern void xfsbdstrat(struct xfs_mount *, struct xfs_buf *);
-
 extern void xfs_buf_ioend(xfs_buf_t *,	int);
 extern void xfs_buf_ioerror(xfs_buf_t *, int);
 extern void xfs_buf_ioerror_alert(struct xfs_buf *, const char *func);
@@ -282,10 +296,7 @@ extern void xfs_buf_iomove(xfs_buf_t *, size_t, size_t, void *,
 #define xfs_buf_zero(bp, off, len) \
 	    xfs_buf_iomove((bp), (off), (len), NULL, XBRW_ZERO)
 
-static inline int xfs_buf_geterror(xfs_buf_t *bp)
-{
-	return bp ? bp->b_error : ENOMEM;
-}
+extern int xfs_bioerror_relse(struct xfs_buf *);
 
 /* Buffer Utility Routines */
 extern xfs_caddr_t xfs_buf_offset(xfs_buf_t *, size_t);
@@ -301,7 +312,8 @@ 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_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)
@@ -352,14 +364,28 @@ static inline void xfs_buf_relse(xfs_buf_t *bp)
 	xfs_buf_rele(bp);
 }
 
+static inline int
+xfs_buf_verify_cksum(struct xfs_buf *bp, unsigned long cksum_offset)
+{
+	return xfs_verify_cksum(bp->b_addr, BBTOB(bp->b_length),
+				cksum_offset);
+}
+
+static inline void
+xfs_buf_update_cksum(struct xfs_buf *bp, unsigned long cksum_offset)
+{
+	xfs_update_cksum(bp->b_addr, BBTOB(bp->b_length),
+			 cksum_offset);
+}
+
 /*
  *	Handling of buftargs.
  */
 extern xfs_buftarg_t *xfs_alloc_buftarg(struct xfs_mount *,
-			struct block_device *, int, const char *);
+			struct block_device *);
 extern void xfs_free_buftarg(struct xfs_mount *, struct xfs_buftarg *);
 extern void xfs_wait_buftarg(xfs_buftarg_t *);
-extern int xfs_setsize_buftarg(xfs_buftarg_t *, unsigned int, unsigned int);
+extern int xfs_setsize_buftarg(xfs_buftarg_t *, unsigned int);
 
 #define xfs_getsize_buftarg(buftarg)	block_size((buftarg)->bt_bdev)
 #define xfs_readonly_buftarg(buftarg)	bdev_read_only((buftarg)->bt_bdev)
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index a64f67ba25d3..76007deed31f 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -182,21 +182,47 @@ xfs_buf_item_size(
 	trace_xfs_buf_item_size(bip);
 }
 
-static struct xfs_log_iovec *
+static inline void
+xfs_buf_item_copy_iovec(
+	struct xfs_log_vec	*lv,
+	struct xfs_log_iovec	**vecp,
+	struct xfs_buf		*bp,
+	uint			offset,
+	int			first_bit,
+	uint			nbits)
+{
+	offset += first_bit * XFS_BLF_CHUNK;
+	xlog_copy_iovec(lv, vecp, XLOG_REG_TYPE_BCHUNK,
+			xfs_buf_offset(bp, offset),
+			nbits * XFS_BLF_CHUNK);
+}
+
+static inline bool
+xfs_buf_item_straddle(
+	struct xfs_buf		*bp,
+	uint			offset,
+	int			next_bit,
+	int			last_bit)
+{
+	return xfs_buf_offset(bp, offset + (next_bit << XFS_BLF_SHIFT)) !=
+		(xfs_buf_offset(bp, offset + (last_bit << XFS_BLF_SHIFT)) +
+		 XFS_BLF_CHUNK);
+}
+
+static void
 xfs_buf_item_format_segment(
 	struct xfs_buf_log_item	*bip,
-	struct xfs_log_iovec	*vecp,
+	struct xfs_log_vec	*lv,
+	struct xfs_log_iovec	**vecp,
 	uint			offset,
 	struct xfs_buf_log_format *blfp)
 {
 	struct xfs_buf	*bp = bip->bli_buf;
 	uint		base_size;
-	uint		nvecs;
 	int		first_bit;
 	int		last_bit;
 	int		next_bit;
 	uint		nbits;
-	uint		buffer_offset;
 
 	/* copy the flags across from the base format item */
 	blfp->blf_flags = bip->__bli_format.blf_flags;
@@ -208,21 +234,17 @@ xfs_buf_item_format_segment(
 	 */
 	base_size = xfs_buf_log_format_size(blfp);
 
-	nvecs = 0;
 	first_bit = xfs_next_bit(blfp->blf_data_map, blfp->blf_map_size, 0);
 	if (!(bip->bli_flags & XFS_BLI_STALE) && first_bit == -1) {
 		/*
 		 * If the map is not be dirty in the transaction, mark
 		 * the size as zero and do not advance the vector pointer.
 		 */
-		goto out;
+		return;
 	}
 
-	vecp->i_addr = blfp;
-	vecp->i_len = base_size;
-	vecp->i_type = XLOG_REG_TYPE_BFORMAT;
-	vecp++;
-	nvecs = 1;
+	blfp = xlog_copy_iovec(lv, vecp, XLOG_REG_TYPE_BFORMAT, blfp, base_size);
+	blfp->blf_size = 1;
 
 	if (bip->bli_flags & XFS_BLI_STALE) {
 		/*
@@ -232,14 +254,13 @@ xfs_buf_item_format_segment(
 		 */
 		trace_xfs_buf_item_format_stale(bip);
 		ASSERT(blfp->blf_flags & XFS_BLF_CANCEL);
-		goto out;
+		return;
 	}
 
 
 	/*
 	 * Fill in an iovec for each set of contiguous chunks.
 	 */
-
 	last_bit = first_bit;
 	nbits = 1;
 	for (;;) {
@@ -252,42 +273,22 @@ xfs_buf_item_format_segment(
 		next_bit = xfs_next_bit(blfp->blf_data_map, blfp->blf_map_size,
 					(uint)last_bit + 1);
 		/*
-		 * If we run out of bits fill in the last iovec and get
-		 * out of the loop.
-		 * Else if we start a new set of bits then fill in the
-		 * iovec for the series we were looking at and start
-		 * counting the bits in the new one.
-		 * Else we're still in the same set of bits so just
-		 * keep counting and scanning.
+		 * If we run out of bits fill in the last iovec and get out of
+		 * the loop.  Else if we start a new set of bits then fill in
+		 * the iovec for the series we were looking at and start
+		 * counting the bits in the new one.  Else we're still in the
+		 * same set of bits so just keep counting and scanning.
 		 */
 		if (next_bit == -1) {
-			buffer_offset = offset + first_bit * XFS_BLF_CHUNK;
-			vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
-			vecp->i_len = nbits * XFS_BLF_CHUNK;
-			vecp->i_type = XLOG_REG_TYPE_BCHUNK;
-			nvecs++;
+			xfs_buf_item_copy_iovec(lv, vecp, bp, offset,
+						first_bit, nbits);
+			blfp->blf_size++;
 			break;
-		} else if (next_bit != last_bit + 1) {
-			buffer_offset = offset + first_bit * XFS_BLF_CHUNK;
-			vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
-			vecp->i_len = nbits * XFS_BLF_CHUNK;
-			vecp->i_type = XLOG_REG_TYPE_BCHUNK;
-			nvecs++;
-			vecp++;
-			first_bit = next_bit;
-			last_bit = next_bit;
-			nbits = 1;
-		} else if (xfs_buf_offset(bp, offset +
-					      (next_bit << XFS_BLF_SHIFT)) !=
-			   (xfs_buf_offset(bp, offset +
-					       (last_bit << XFS_BLF_SHIFT)) +
-			    XFS_BLF_CHUNK)) {
-			buffer_offset = offset + first_bit * XFS_BLF_CHUNK;
-			vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
-			vecp->i_len = nbits * XFS_BLF_CHUNK;
-			vecp->i_type = XLOG_REG_TYPE_BCHUNK;
-			nvecs++;
-			vecp++;
+		} else if (next_bit != last_bit + 1 ||
+		           xfs_buf_item_straddle(bp, offset, next_bit, last_bit)) {
+			xfs_buf_item_copy_iovec(lv, vecp, bp, offset,
+						first_bit, nbits);
+			blfp->blf_size++;
 			first_bit = next_bit;
 			last_bit = next_bit;
 			nbits = 1;
@@ -296,9 +297,6 @@ xfs_buf_item_format_segment(
 			nbits++;
 		}
 	}
-out:
-	blfp->blf_size = nvecs;
-	return vecp;
 }
 
 /*
@@ -310,10 +308,11 @@ out:
 STATIC void
 xfs_buf_item_format(
 	struct xfs_log_item	*lip,
-	struct xfs_log_iovec	*vecp)
+	struct xfs_log_vec	*lv)
 {
 	struct xfs_buf_log_item	*bip = BUF_ITEM(lip);
 	struct xfs_buf		*bp = bip->bli_buf;
+	struct xfs_log_iovec	*vecp = NULL;
 	uint			offset = 0;
 	int			i;
 
@@ -354,8 +353,8 @@ xfs_buf_item_format(
 	}
 
 	for (i = 0; i < bip->bli_format_count; i++) {
-		vecp = xfs_buf_item_format_segment(bip, vecp, offset,
-						&bip->bli_formats[i]);
+		xfs_buf_item_format_segment(bip, lv, &vecp, offset,
+					    &bip->bli_formats[i]);
 		offset += bp->b_maps[i].bm_len;
 	}
 
@@ -489,13 +488,21 @@ xfs_buf_item_unpin(
 		xfs_buf_lock(bp);
 		xfs_buf_hold(bp);
 		bp->b_flags |= XBF_ASYNC;
-		xfs_buf_ioerror(bp, EIO);
+		xfs_buf_ioerror(bp, -EIO);
 		XFS_BUF_UNDONE(bp);
 		xfs_buf_stale(bp);
 		xfs_buf_ioend(bp, 0);
 	}
 }
 
+/*
+ * Buffer IO error rate limiting. Limit it to no more than 10 messages per 30
+ * seconds so as to not spam logs too much on repeated detection of the same
+ * buffer being bad..
+ */
+
+DEFINE_RATELIMIT_STATE(xfs_buf_write_fail_rl_state, 30 * HZ, 10);
+
 STATIC uint
 xfs_buf_item_push(
 	struct xfs_log_item	*lip,
@@ -524,6 +531,14 @@ xfs_buf_item_push(
 
 	trace_xfs_buf_item_push(bip);
 
+	/* has a previous flush failed due to IO errors? */
+	if ((bp->b_flags & XBF_WRITE_FAIL) &&
+	    ___ratelimit(&xfs_buf_write_fail_rl_state, "XFS:")) {
+		xfs_warn(bp->b_target->bt_mount,
+"Detected failing async write on buffer block 0x%llx. Retrying async write.\n",
+			 (long long)bp->b_bn);
+	}
+
 	if (!xfs_buf_delwri_queue(bp, buffer_list))
 		rval = XFS_ITEM_FLUSHING;
 	xfs_buf_unlock(bp);
@@ -710,7 +725,7 @@ xfs_buf_item_get_format(
 	bip->bli_formats = kmem_zalloc(count * sizeof(struct xfs_buf_log_format),
 				KM_SLEEP);
 	if (!bip->bli_formats)
-		return ENOMEM;
+		return -ENOMEM;
 	return 0;
 }
 
@@ -781,20 +796,6 @@ xfs_buf_item_init(
 		bip->bli_formats[i].blf_map_size = map_size;
 	}
 
-#ifdef XFS_TRANS_DEBUG
-	/*
-	 * Allocate the arrays for tracking what needs to be logged
-	 * and what our callers request to be logged.  bli_orig
-	 * holds a copy of the original, clean buffer for comparison
-	 * against, and bli_logged keeps a 1 bit flag per byte in
-	 * the buffer to indicate which bytes the callers have asked
-	 * to have logged.
-	 */
-	bip->bli_orig = kmem_alloc(BBTOB(bp->b_length), KM_SLEEP);
-	memcpy(bip->bli_orig, bp->b_addr, BBTOB(bp->b_length));
-	bip->bli_logged = kmem_zalloc(BBTOB(bp->b_length) / NBBY, KM_SLEEP);
-#endif
-
 	/*
 	 * Put the buf item into the list of items attached to the
 	 * buffer at the front.
@@ -811,7 +812,6 @@ xfs_buf_item_init(
  */
 static void
 xfs_buf_item_log_segment(
-	struct xfs_buf_log_item	*bip,
 	uint			first,
 	uint			last,
 	uint			*map)
@@ -919,7 +919,7 @@ xfs_buf_item_log(
 		if (end > last)
 			end = last;
 
-		xfs_buf_item_log_segment(bip, first, end,
+		xfs_buf_item_log_segment(first, end,
 					 &bip->bli_formats[i].blf_data_map[0]);
 
 		start += bp->b_maps[i].bm_len;
@@ -942,11 +942,6 @@ STATIC void
 xfs_buf_item_free(
 	xfs_buf_log_item_t	*bip)
 {
-#ifdef XFS_TRANS_DEBUG
-	kmem_free(bip->bli_orig);
-	kmem_free(bip->bli_logged);
-#endif /* XFS_TRANS_DEBUG */
-
 	xfs_buf_item_free_format(bip);
 	kmem_zone_free(xfs_buf_item_zone, bip);
 }
@@ -1057,7 +1052,7 @@ xfs_buf_iodone_callbacks(
 	static ulong		lasttime;
 	static xfs_buftarg_t	*lasttarg;
 
-	if (likely(!xfs_buf_geterror(bp)))
+	if (likely(!bp->b_error))
 		goto do_callbacks;
 
 	/*
@@ -1096,8 +1091,9 @@ xfs_buf_iodone_callbacks(
 
 		xfs_buf_ioerror(bp, 0); /* errno of 0 unsets the flag */
 
-		if (!XFS_BUF_ISSTALE(bp)) {
-			bp->b_flags |= XBF_WRITE | XBF_ASYNC | XBF_DONE;
+		if (!(bp->b_flags & (XBF_STALE|XBF_WRITE_FAIL))) {
+			bp->b_flags |= XBF_WRITE | XBF_ASYNC |
+				       XBF_DONE | XBF_WRITE_FAIL;
 			xfs_buf_iorequest(bp);
 		} else {
 			xfs_buf_relse(bp);
diff --git a/fs/xfs/xfs_dir2_readdir.c b/fs/xfs/xfs_dir2_readdir.c
index c4e50c6ed584..f1b69edcdf31 100644
--- a/fs/xfs/xfs_dir2_readdir.c
+++ b/fs/xfs/xfs_dir2_readdir.c
@@ -76,27 +76,26 @@ const unsigned char xfs_mode_to_ftype[S_IFMT >> S_SHIFT] = {
 
 STATIC int
 xfs_dir2_sf_getdents(
-	xfs_inode_t		*dp,		/* incore directory inode */
+	struct xfs_da_args	*args,
 	struct dir_context	*ctx)
 {
 	int			i;		/* shortform entry number */
-	xfs_mount_t		*mp;		/* filesystem mount point */
+	struct xfs_inode	*dp = args->dp;	/* incore directory inode */
 	xfs_dir2_dataptr_t	off;		/* current entry's offset */
 	xfs_dir2_sf_entry_t	*sfep;		/* shortform directory entry */
 	xfs_dir2_sf_hdr_t	*sfp;		/* shortform structure */
 	xfs_dir2_dataptr_t	dot_offset;
 	xfs_dir2_dataptr_t	dotdot_offset;
 	xfs_ino_t		ino;
-
-	mp = dp->i_mount;
+	struct xfs_da_geometry	*geo = args->geo;
 
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
 	/*
 	 * Give up if the directory is way too short.
 	 */
 	if (dp->i_d.di_size < offsetof(xfs_dir2_sf_hdr_t, parent)) {
-		ASSERT(XFS_FORCED_SHUTDOWN(mp));
-		return XFS_ERROR(EIO);
+		ASSERT(XFS_FORCED_SHUTDOWN(dp->i_mount));
+		return -EIO;
 	}
 
 	ASSERT(dp->i_df.if_bytes == dp->i_d.di_size);
@@ -109,18 +108,18 @@ xfs_dir2_sf_getdents(
 	/*
 	 * If the block number in the offset is out of range, we're done.
 	 */
-	if (xfs_dir2_dataptr_to_db(mp, ctx->pos) > mp->m_dirdatablk)
+	if (xfs_dir2_dataptr_to_db(geo, ctx->pos) > geo->datablk)
 		return 0;
 
 	/*
 	 * Precalculate offsets for . and .. as we will always need them.
 	 *
 	 * XXX(hch): the second argument is sometimes 0 and sometimes
-	 * mp->m_dirdatablk.
+	 * geo->datablk
 	 */
-	dot_offset = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk,
+	dot_offset = xfs_dir2_db_off_to_dataptr(geo, geo->datablk,
 						dp->d_ops->data_dot_offset);
-	dotdot_offset = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk,
+	dotdot_offset = xfs_dir2_db_off_to_dataptr(geo, geo->datablk,
 						dp->d_ops->data_dotdot_offset);
 
 	/*
@@ -149,7 +148,7 @@ xfs_dir2_sf_getdents(
 	for (i = 0; i < sfp->count; i++) {
 		__uint8_t filetype;
 
-		off = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk,
+		off = xfs_dir2_db_off_to_dataptr(geo, geo->datablk,
 				xfs_dir2_sf_get_offset(sfep));
 
 		if (ctx->pos > off) {
@@ -161,13 +160,13 @@ xfs_dir2_sf_getdents(
 		filetype = dp->d_ops->sf_get_ftype(sfep);
 		ctx->pos = off & 0x7fffffff;
 		if (!dir_emit(ctx, (char *)sfep->name, sfep->namelen, ino,
-			    xfs_dir3_get_dtype(mp, filetype)))
+			    xfs_dir3_get_dtype(dp->i_mount, filetype)))
 			return 0;
 		sfep = dp->d_ops->sf_nextentry(sfp, sfep);
 	}
 
-	ctx->pos = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk + 1, 0) &
-			0x7fffffff;
+	ctx->pos = xfs_dir2_db_off_to_dataptr(geo, geo->datablk + 1, 0) &
+								0x7fffffff;
 	return 0;
 }
 
@@ -176,9 +175,10 @@ xfs_dir2_sf_getdents(
  */
 STATIC int
 xfs_dir2_block_getdents(
-	xfs_inode_t		*dp,		/* incore inode */
+	struct xfs_da_args	*args,
 	struct dir_context	*ctx)
 {
+	struct xfs_inode	*dp = args->dp;	/* incore directory inode */
 	xfs_dir2_data_hdr_t	*hdr;		/* block header */
 	struct xfs_buf		*bp;		/* buffer for block */
 	xfs_dir2_block_tail_t	*btp;		/* block tail */
@@ -186,16 +186,15 @@ xfs_dir2_block_getdents(
 	xfs_dir2_data_unused_t	*dup;		/* block unused entry */
 	char			*endptr;	/* end of the data entries */
 	int			error;		/* error return value */
-	xfs_mount_t		*mp;		/* filesystem mount point */
 	char			*ptr;		/* current data entry */
 	int			wantoff;	/* starting block offset */
 	xfs_off_t		cook;
+	struct xfs_da_geometry	*geo = args->geo;
 
-	mp = dp->i_mount;
 	/*
 	 * If the block number in the offset is out of range, we're done.
 	 */
-	if (xfs_dir2_dataptr_to_db(mp, ctx->pos) > mp->m_dirdatablk)
+	if (xfs_dir2_dataptr_to_db(geo, ctx->pos) > geo->datablk)
 		return 0;
 
 	error = xfs_dir3_block_read(NULL, dp, &bp);
@@ -206,13 +205,13 @@ xfs_dir2_block_getdents(
 	 * Extract the byte offset we start at from the seek pointer.
 	 * We'll skip entries before this.
 	 */
-	wantoff = xfs_dir2_dataptr_to_off(mp, ctx->pos);
+	wantoff = xfs_dir2_dataptr_to_off(geo, ctx->pos);
 	hdr = bp->b_addr;
 	xfs_dir3_data_check(dp, bp);
 	/*
 	 * Set up values for the loop.
 	 */
-	btp = xfs_dir2_block_tail_p(mp, hdr);
+	btp = xfs_dir2_block_tail_p(geo, hdr);
 	ptr = (char *)dp->d_ops->data_entry_p(hdr);
 	endptr = (char *)xfs_dir2_block_leaf_p(btp);
 
@@ -244,7 +243,7 @@ xfs_dir2_block_getdents(
 		if ((char *)dep - (char *)hdr < wantoff)
 			continue;
 
-		cook = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk,
+		cook = xfs_dir2_db_off_to_dataptr(geo, geo->datablk,
 					    (char *)dep - (char *)hdr);
 
 		ctx->pos = cook & 0x7fffffff;
@@ -254,7 +253,7 @@ xfs_dir2_block_getdents(
 		 */
 		if (!dir_emit(ctx, (char *)dep->name, dep->namelen,
 			    be64_to_cpu(dep->inumber),
-			    xfs_dir3_get_dtype(mp, filetype))) {
+			    xfs_dir3_get_dtype(dp->i_mount, filetype))) {
 			xfs_trans_brelse(NULL, bp);
 			return 0;
 		}
@@ -264,8 +263,8 @@ xfs_dir2_block_getdents(
 	 * Reached the end of the block.
 	 * Set the offset to a non-existent block 1 and return.
 	 */
-	ctx->pos = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk + 1, 0) &
-			0x7fffffff;
+	ctx->pos = xfs_dir2_db_off_to_dataptr(geo, geo->datablk + 1, 0) &
+								0x7fffffff;
 	xfs_trans_brelse(NULL, bp);
 	return 0;
 }
@@ -286,13 +285,13 @@ struct xfs_dir2_leaf_map_info {
 
 STATIC int
 xfs_dir2_leaf_readbuf(
-	struct xfs_inode	*dp,
+	struct xfs_da_args	*args,
 	size_t			bufsize,
 	struct xfs_dir2_leaf_map_info *mip,
 	xfs_dir2_off_t		*curoff,
 	struct xfs_buf		**bpp)
 {
-	struct xfs_mount	*mp = dp->i_mount;
+	struct xfs_inode	*dp = args->dp;
 	struct xfs_buf		*bp = *bpp;
 	struct xfs_bmbt_irec	*map = mip->map;
 	struct blk_plug		plug;
@@ -300,6 +299,7 @@ xfs_dir2_leaf_readbuf(
 	int			length;
 	int			i;
 	int			j;
+	struct xfs_da_geometry	*geo = args->geo;
 
 	/*
 	 * If we have a buffer, we need to release it and
@@ -309,12 +309,12 @@ xfs_dir2_leaf_readbuf(
 	if (bp) {
 		xfs_trans_brelse(NULL, bp);
 		bp = NULL;
-		mip->map_blocks -= mp->m_dirblkfsbs;
+		mip->map_blocks -= geo->fsbcount;
 		/*
 		 * Loop to get rid of the extents for the
 		 * directory block.
 		 */
-		for (i = mp->m_dirblkfsbs; i > 0; ) {
+		for (i = geo->fsbcount; i > 0; ) {
 			j = min_t(int, map->br_blockcount, i);
 			map->br_blockcount -= j;
 			map->br_startblock += j;
@@ -333,8 +333,7 @@ xfs_dir2_leaf_readbuf(
 	/*
 	 * Recalculate the readahead blocks wanted.
 	 */
-	mip->ra_want = howmany(bufsize + mp->m_dirblksize,
-			       mp->m_sb.sb_blocksize) - 1;
+	mip->ra_want = howmany(bufsize + geo->blksize, (1 << geo->fsblog)) - 1;
 	ASSERT(mip->ra_want >= 0);
 
 	/*
@@ -342,14 +341,14 @@ xfs_dir2_leaf_readbuf(
 	 * run out of data blocks, get some more mappings.
 	 */
 	if (1 + mip->ra_want > mip->map_blocks &&
-	    mip->map_off < xfs_dir2_byte_to_da(mp, XFS_DIR2_LEAF_OFFSET)) {
+	    mip->map_off < xfs_dir2_byte_to_da(geo, XFS_DIR2_LEAF_OFFSET)) {
 		/*
 		 * Get more bmaps, fill in after the ones
 		 * we already have in the table.
 		 */
 		mip->nmap = mip->map_size - mip->map_valid;
 		error = xfs_bmapi_read(dp, mip->map_off,
-				xfs_dir2_byte_to_da(mp, XFS_DIR2_LEAF_OFFSET) -
+				xfs_dir2_byte_to_da(geo, XFS_DIR2_LEAF_OFFSET) -
 								mip->map_off,
 				&map[mip->map_valid], &mip->nmap, 0);
 
@@ -370,7 +369,7 @@ xfs_dir2_leaf_readbuf(
 			i = mip->map_valid + mip->nmap - 1;
 			mip->map_off = map[i].br_startoff + map[i].br_blockcount;
 		} else
-			mip->map_off = xfs_dir2_byte_to_da(mp,
+			mip->map_off = xfs_dir2_byte_to_da(geo,
 							XFS_DIR2_LEAF_OFFSET);
 
 		/*
@@ -396,18 +395,18 @@ xfs_dir2_leaf_readbuf(
 	 * No valid mappings, so no more data blocks.
 	 */
 	if (!mip->map_valid) {
-		*curoff = xfs_dir2_da_to_byte(mp, mip->map_off);
+		*curoff = xfs_dir2_da_to_byte(geo, mip->map_off);
 		goto out;
 	}
 
 	/*
 	 * Read the directory block starting at the first mapping.
 	 */
-	mip->curdb = xfs_dir2_da_to_db(mp, map->br_startoff);
+	mip->curdb = xfs_dir2_da_to_db(geo, map->br_startoff);
 	error = xfs_dir3_data_read(NULL, dp, map->br_startoff,
-			map->br_blockcount >= mp->m_dirblkfsbs ?
-			    XFS_FSB_TO_DADDR(mp, map->br_startblock) : -1, &bp);
-
+			map->br_blockcount >= geo->fsbcount ?
+			    XFS_FSB_TO_DADDR(dp->i_mount, map->br_startblock) :
+			    -1, &bp);
 	/*
 	 * Should just skip over the data block instead of giving up.
 	 */
@@ -419,7 +418,7 @@ xfs_dir2_leaf_readbuf(
 	 * was previously ra.
 	 */
 	if (mip->ra_current)
-		mip->ra_current -= mp->m_dirblkfsbs;
+		mip->ra_current -= geo->fsbcount;
 
 	/*
 	 * Do we need more readahead?
@@ -427,16 +426,16 @@ xfs_dir2_leaf_readbuf(
 	blk_start_plug(&plug);
 	for (mip->ra_index = mip->ra_offset = i = 0;
 	     mip->ra_want > mip->ra_current && i < mip->map_blocks;
-	     i += mp->m_dirblkfsbs) {
+	     i += geo->fsbcount) {
 		ASSERT(mip->ra_index < mip->map_valid);
 		/*
 		 * Read-ahead a contiguous directory block.
 		 */
 		if (i > mip->ra_current &&
-		    map[mip->ra_index].br_blockcount >= mp->m_dirblkfsbs) {
-			xfs_dir3_data_readahead(NULL, dp,
+		    map[mip->ra_index].br_blockcount >= geo->fsbcount) {
+			xfs_dir3_data_readahead(dp,
 				map[mip->ra_index].br_startoff + mip->ra_offset,
-				XFS_FSB_TO_DADDR(mp,
+				XFS_FSB_TO_DADDR(dp->i_mount,
 					map[mip->ra_index].br_startblock +
 							mip->ra_offset));
 			mip->ra_current = i;
@@ -447,7 +446,7 @@ xfs_dir2_leaf_readbuf(
 		 * use our mapping, but this is a very rare case.
 		 */
 		else if (i > mip->ra_current) {
-			xfs_dir3_data_readahead(NULL, dp,
+			xfs_dir3_data_readahead(dp,
 					map[mip->ra_index].br_startoff +
 							mip->ra_offset, -1);
 			mip->ra_current = i;
@@ -456,15 +455,14 @@ xfs_dir2_leaf_readbuf(
 		/*
 		 * Advance offset through the mapping table.
 		 */
-		for (j = 0; j < mp->m_dirblkfsbs; j++) {
+		for (j = 0; j < geo->fsbcount; j += length ) {
 			/*
 			 * The rest of this extent but not more than a dir
 			 * block.
 			 */
-			length = min_t(int, mp->m_dirblkfsbs,
+			length = min_t(int, geo->fsbcount,
 					map[mip->ra_index].br_blockcount -
 							mip->ra_offset);
-			j += length;
 			mip->ra_offset += length;
 
 			/*
@@ -489,22 +487,23 @@ out:
  */
 STATIC int
 xfs_dir2_leaf_getdents(
-	xfs_inode_t		*dp,		/* incore directory inode */
+	struct xfs_da_args	*args,
 	struct dir_context	*ctx,
 	size_t			bufsize)
 {
+	struct xfs_inode	*dp = args->dp;
 	struct xfs_buf		*bp = NULL;	/* data block buffer */
 	xfs_dir2_data_hdr_t	*hdr;		/* data block header */
 	xfs_dir2_data_entry_t	*dep;		/* data entry */
 	xfs_dir2_data_unused_t	*dup;		/* unused entry */
 	int			error = 0;	/* error return value */
 	int			length;		/* temporary length value */
-	xfs_mount_t		*mp;		/* filesystem mount point */
 	int			byteoff;	/* offset in current block */
 	xfs_dir2_off_t		curoff;		/* current overall offset */
 	xfs_dir2_off_t		newoff;		/* new curoff after new blk */
 	char			*ptr = NULL;	/* pointer to current data */
 	struct xfs_dir2_leaf_map_info *map_info;
+	struct xfs_da_geometry	*geo = args->geo;
 
 	/*
 	 * If the offset is at or past the largest allowed value,
@@ -513,15 +512,12 @@ xfs_dir2_leaf_getdents(
 	if (ctx->pos >= XFS_DIR2_MAX_DATAPTR)
 		return 0;
 
-	mp = dp->i_mount;
-
 	/*
 	 * Set up to bmap a number of blocks based on the caller's
 	 * buffer size, the directory block size, and the filesystem
 	 * block size.
 	 */
-	length = howmany(bufsize + mp->m_dirblksize,
-				     mp->m_sb.sb_blocksize);
+	length = howmany(bufsize + geo->blksize, (1 << geo->fsblog));
 	map_info = kmem_zalloc(offsetof(struct xfs_dir2_leaf_map_info, map) +
 				(length * sizeof(struct xfs_bmbt_irec)),
 			       KM_SLEEP | KM_NOFS);
@@ -531,14 +527,14 @@ xfs_dir2_leaf_getdents(
 	 * Inside the loop we keep the main offset value as a byte offset
 	 * in the directory file.
 	 */
-	curoff = xfs_dir2_dataptr_to_byte(mp, ctx->pos);
+	curoff = xfs_dir2_dataptr_to_byte(ctx->pos);
 
 	/*
 	 * Force this conversion through db so we truncate the offset
 	 * down to get the start of the data block.
 	 */
-	map_info->map_off = xfs_dir2_db_to_da(mp,
-					      xfs_dir2_byte_to_db(mp, curoff));
+	map_info->map_off = xfs_dir2_db_to_da(geo,
+					      xfs_dir2_byte_to_db(geo, curoff));
 
 	/*
 	 * Loop over directory entries until we reach the end offset.
@@ -551,9 +547,9 @@ xfs_dir2_leaf_getdents(
 		 * If we have no buffer, or we're off the end of the
 		 * current buffer, need to get another one.
 		 */
-		if (!bp || ptr >= (char *)bp->b_addr + mp->m_dirblksize) {
+		if (!bp || ptr >= (char *)bp->b_addr + geo->blksize) {
 
-			error = xfs_dir2_leaf_readbuf(dp, bufsize, map_info,
+			error = xfs_dir2_leaf_readbuf(args, bufsize, map_info,
 						      &curoff, &bp);
 			if (error || !map_info->map_valid)
 				break;
@@ -561,7 +557,8 @@ xfs_dir2_leaf_getdents(
 			/*
 			 * Having done a read, we need to set a new offset.
 			 */
-			newoff = xfs_dir2_db_off_to_byte(mp, map_info->curdb, 0);
+			newoff = xfs_dir2_db_off_to_byte(geo,
+							 map_info->curdb, 0);
 			/*
 			 * Start of the current block.
 			 */
@@ -571,7 +568,7 @@ xfs_dir2_leaf_getdents(
 			 * Make sure we're in the right block.
 			 */
 			else if (curoff > newoff)
-				ASSERT(xfs_dir2_byte_to_db(mp, curoff) ==
+				ASSERT(xfs_dir2_byte_to_db(geo, curoff) ==
 				       map_info->curdb);
 			hdr = bp->b_addr;
 			xfs_dir3_data_check(dp, bp);
@@ -579,7 +576,7 @@ xfs_dir2_leaf_getdents(
 			 * Find our position in the block.
 			 */
 			ptr = (char *)dp->d_ops->data_entry_p(hdr);
-			byteoff = xfs_dir2_byte_to_off(mp, curoff);
+			byteoff = xfs_dir2_byte_to_off(geo, curoff);
 			/*
 			 * Skip past the header.
 			 */
@@ -608,10 +605,10 @@ xfs_dir2_leaf_getdents(
 				 * Now set our real offset.
 				 */
 				curoff =
-					xfs_dir2_db_off_to_byte(mp,
-					    xfs_dir2_byte_to_db(mp, curoff),
+					xfs_dir2_db_off_to_byte(geo,
+					    xfs_dir2_byte_to_db(geo, curoff),
 					    (char *)ptr - (char *)hdr);
-				if (ptr >= (char *)hdr + mp->m_dirblksize) {
+				if (ptr >= (char *)hdr + geo->blksize) {
 					continue;
 				}
 			}
@@ -635,10 +632,10 @@ xfs_dir2_leaf_getdents(
 		length = dp->d_ops->data_entsize(dep->namelen);
 		filetype = dp->d_ops->data_get_ftype(dep);
 
-		ctx->pos = xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff;
+		ctx->pos = xfs_dir2_byte_to_dataptr(curoff) & 0x7fffffff;
 		if (!dir_emit(ctx, (char *)dep->name, dep->namelen,
 			    be64_to_cpu(dep->inumber),
-			    xfs_dir3_get_dtype(mp, filetype)))
+			    xfs_dir3_get_dtype(dp->i_mount, filetype)))
 			break;
 
 		/*
@@ -653,10 +650,10 @@ xfs_dir2_leaf_getdents(
 	/*
 	 * All done.  Set output offset value to current offset.
 	 */
-	if (curoff > xfs_dir2_dataptr_to_byte(mp, XFS_DIR2_MAX_DATAPTR))
+	if (curoff > xfs_dir2_dataptr_to_byte(XFS_DIR2_MAX_DATAPTR))
 		ctx->pos = XFS_DIR2_MAX_DATAPTR & 0x7fffffff;
 	else
-		ctx->pos = xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff;
+		ctx->pos = xfs_dir2_byte_to_dataptr(curoff) & 0x7fffffff;
 	kmem_free(map_info);
 	if (bp)
 		xfs_trans_brelse(NULL, bp);
@@ -668,28 +665,36 @@ xfs_dir2_leaf_getdents(
  */
 int
 xfs_readdir(
-	xfs_inode_t	*dp,
-	struct dir_context *ctx,
-	size_t		bufsize)
+	struct xfs_inode	*dp,
+	struct dir_context	*ctx,
+	size_t			bufsize)
 {
-	int		rval;		/* return value */
-	int		v;		/* type-checking value */
+	struct xfs_da_args	args = { NULL };
+	int			rval;
+	int			v;
+	uint			lock_mode;
 
 	trace_xfs_readdir(dp);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	ASSERT(S_ISDIR(dp->i_d.di_mode));
 	XFS_STATS_INC(xs_dir_getdents);
 
+	args.dp = dp;
+	args.geo = dp->i_mount->m_dir_geo;
+
+	lock_mode = xfs_ilock_data_map_shared(dp);
 	if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
-		rval = xfs_dir2_sf_getdents(dp, ctx);
-	else if ((rval = xfs_dir2_isblock(NULL, dp, &v)))
+		rval = xfs_dir2_sf_getdents(&args, ctx);
+	else if ((rval = xfs_dir2_isblock(&args, &v)))
 		;
 	else if (v)
-		rval = xfs_dir2_block_getdents(dp, ctx);
+		rval = xfs_dir2_block_getdents(&args, ctx);
 	else
-		rval = xfs_dir2_leaf_getdents(dp, ctx, bufsize);
+		rval = xfs_dir2_leaf_getdents(&args, ctx, bufsize);
+	xfs_iunlock(dp, lock_mode);
+
 	return rval;
 }
diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c
index 4f11ef011139..13d08a1b390e 100644
--- a/fs/xfs/xfs_discard.c
+++ b/fs/xfs/xfs_discard.c
@@ -124,7 +124,7 @@ xfs_trim_extents(
 		}
 
 		trace_xfs_discard_extent(mp, agno, fbno, flen);
-		error = -blkdev_issue_discard(bdev, dbno, dlen, GFP_NOFS, 0);
+		error = blkdev_issue_discard(bdev, dbno, dlen, GFP_NOFS, 0);
 		if (error)
 			goto out_del_cursor;
 		*blocks_trimmed += flen;
@@ -166,11 +166,11 @@ xfs_ioc_trim(
 	int			error, last_error = 0;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 	if (!blk_queue_discard(q))
-		return -XFS_ERROR(EOPNOTSUPP);
+		return -EOPNOTSUPP;
 	if (copy_from_user(&range, urange, sizeof(range)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	/*
 	 * Truncating down the len isn't actually quite correct, but using
@@ -182,7 +182,7 @@ xfs_ioc_trim(
 	if (range.start >= XFS_FSB_TO_B(mp, mp->m_sb.sb_dblocks) ||
 	    range.minlen > XFS_FSB_TO_B(mp, XFS_ALLOC_AG_MAX_USABLE(mp)) ||
 	    range.len < mp->m_sb.sb_blocksize)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	start = BTOBB(range.start);
 	end = start + BTOBBT(range.len) - 1;
@@ -195,7 +195,7 @@ xfs_ioc_trim(
 	end_agno = xfs_daddr_to_agno(mp, end);
 
 	for (agno = start_agno; agno <= end_agno; agno++) {
-		error = -xfs_trim_extents(mp, agno, start, end, minlen,
+		error = xfs_trim_extents(mp, agno, start, end, minlen,
 					  &blocks_trimmed);
 		if (error)
 			last_error = error;
@@ -206,7 +206,7 @@ xfs_ioc_trim(
 
 	range.len = XFS_FSB_TO_B(mp, blocks_trimmed);
 	if (copy_to_user(urange, &range, sizeof(range)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -222,11 +222,11 @@ xfs_discard_extents(
 		trace_xfs_discard_extent(mp, busyp->agno, busyp->bno,
 					 busyp->length);
 
-		error = -blkdev_issue_discard(mp->m_ddev_targp->bt_bdev,
+		error = blkdev_issue_discard(mp->m_ddev_targp->bt_bdev,
 				XFS_AGB_TO_DADDR(mp, busyp->agno, busyp->bno),
 				XFS_FSB_TO_BB(mp, busyp->length),
 				GFP_NOFS, 0);
-		if (error && error != EOPNOTSUPP) {
+		if (error && error != -EOPNOTSUPP) {
 			xfs_info(mp,
 	 "discard failed for extent [0x%llu,%u], error %d",
 				 (unsigned long long)busyp->bno,
diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c
index 6b1e695caf0e..63c2de49f61d 100644
--- a/fs/xfs/xfs_dquot.c
+++ b/fs/xfs/xfs_dquot.c
@@ -327,7 +327,7 @@ xfs_qm_dqalloc(
 	 */
 	if (!xfs_this_quota_on(dqp->q_mount, dqp->dq_flags)) {
 		xfs_iunlock(quotip, XFS_ILOCK_EXCL);
-		return (ESRCH);
+		return -ESRCH;
 	}
 
 	xfs_trans_ijoin(tp, quotip, XFS_ILOCK_EXCL);
@@ -353,10 +353,10 @@ xfs_qm_dqalloc(
 			       dqp->q_blkno,
 			       mp->m_quotainfo->qi_dqchunklen,
 			       0);
-
-	error = xfs_buf_geterror(bp);
-	if (error)
+	if (!bp) {
+		error = -ENOMEM;
 		goto error1;
+	}
 	bp->b_ops = &xfs_dquot_buf_ops;
 
 	/*
@@ -400,7 +400,7 @@ xfs_qm_dqalloc(
       error0:
 	xfs_iunlock(quotip, XFS_ILOCK_EXCL);
 
-	return (error);
+	return error;
 }
 
 STATIC int
@@ -426,7 +426,7 @@ xfs_qm_dqrepair(
 
 	if (error) {
 		ASSERT(*bpp == NULL);
-		return XFS_ERROR(error);
+		return error;
 	}
 	(*bpp)->b_ops = &xfs_dquot_buf_ops;
 
@@ -442,7 +442,7 @@ xfs_qm_dqrepair(
 		if (error) {
 			/* repair failed, we're screwed */
 			xfs_trans_brelse(tp, *bpp);
-			return XFS_ERROR(EIO);
+			return -EIO;
 		}
 	}
 
@@ -469,17 +469,18 @@ xfs_qm_dqtobp(
 	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;
 
 	dqp->q_fileoffset = (xfs_fileoff_t)id / mp->m_quotainfo->qi_dqperchunk;
 
-	xfs_ilock(quotip, XFS_ILOCK_SHARED);
+	lock_mode = xfs_ilock_data_map_shared(quotip);
 	if (!xfs_this_quota_on(dqp->q_mount, dqp->dq_flags)) {
 		/*
 		 * Return if this type of quotas is turned off while we
 		 * didn't have the quota inode lock.
 		 */
-		xfs_iunlock(quotip, XFS_ILOCK_SHARED);
-		return ESRCH;
+		xfs_iunlock(quotip, lock_mode);
+		return -ESRCH;
 	}
 
 	/*
@@ -488,7 +489,7 @@ xfs_qm_dqtobp(
 	error = xfs_bmapi_read(quotip, dqp->q_fileoffset,
 			       XFS_DQUOT_CLUSTER_SIZE_FSB, &map, &nmaps, 0);
 
-	xfs_iunlock(quotip, XFS_ILOCK_SHARED);
+	xfs_iunlock(quotip, lock_mode);
 	if (error)
 		return error;
 
@@ -507,7 +508,7 @@ xfs_qm_dqtobp(
 		 * We don't allocate unless we're asked to
 		 */
 		if (!(flags & XFS_QMOPT_DQALLOC))
-			return ENOENT;
+			return -ENOENT;
 
 		ASSERT(tp);
 		error = xfs_qm_dqalloc(tpp, mp, dqp, quotip,
@@ -529,7 +530,7 @@ xfs_qm_dqtobp(
 					   mp->m_quotainfo->qi_dqchunklen,
 					   0, &bp, &xfs_dquot_buf_ops);
 
-		if (error == EFSCORRUPTED && (flags & XFS_QMOPT_DQREPAIR)) {
+		if (error == -EFSCORRUPTED && (flags & XFS_QMOPT_DQREPAIR)) {
 			xfs_dqid_t firstid = (xfs_dqid_t)map.br_startoff *
 						mp->m_quotainfo->qi_dqperchunk;
 			ASSERT(bp == NULL);
@@ -538,7 +539,7 @@ xfs_qm_dqtobp(
 
 		if (error) {
 			ASSERT(bp == NULL);
-			return XFS_ERROR(error);
+			return error;
 		}
 	}
 
@@ -546,7 +547,7 @@ xfs_qm_dqtobp(
 	*O_bpp = bp;
 	*O_ddpp = bp->b_addr + dqp->q_bufoffset;
 
-	return (0);
+	return 0;
 }
 
 
@@ -614,7 +615,7 @@ xfs_qm_dqread(
 
 	if (flags & XFS_QMOPT_DQALLOC) {
 		tp = xfs_trans_alloc(mp, XFS_TRANS_QM_DQALLOC);
-		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_attrsetm,
+		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_qm_dqalloc,
 					  XFS_QM_DQALLOC_SPACE_RES(mp), 0);
 		if (error)
 			goto error1;
@@ -714,7 +715,7 @@ xfs_qm_dqget(
 	if ((! XFS_IS_UQUOTA_ON(mp) && type == XFS_DQ_USER) ||
 	    (! XFS_IS_PQUOTA_ON(mp) && type == XFS_DQ_PROJ) ||
 	    (! XFS_IS_GQUOTA_ON(mp) && type == XFS_DQ_GROUP)) {
-		return (ESRCH);
+		return -ESRCH;
 	}
 
 #ifdef DEBUG
@@ -722,7 +723,7 @@ xfs_qm_dqget(
 		if ((xfs_dqerror_target == mp->m_ddev_targp) &&
 		    (xfs_dqreq_num++ % xfs_dqerror_mod) == 0) {
 			xfs_debug(mp, "Returning error in dqget");
-			return (EIO);
+			return -EIO;
 		}
 	}
 
@@ -795,14 +796,14 @@ restart:
 		} else {
 			/* inode stays locked on return */
 			xfs_qm_dqdestroy(dqp);
-			return XFS_ERROR(ESRCH);
+			return -ESRCH;
 		}
 	}
 
 	mutex_lock(&qi->qi_tree_lock);
-	error = -radix_tree_insert(tree, id, dqp);
+	error = radix_tree_insert(tree, id, dqp);
 	if (unlikely(error)) {
-		WARN_ON(error != EEXIST);
+		WARN_ON(error != -EEXIST);
 
 		/*
 		 * Duplicate found. Just throw away the new dquot and start
@@ -828,48 +829,7 @@ restart:
 	ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	trace_xfs_dqget_miss(dqp);
 	*O_dqpp = dqp;
-	return (0);
-}
-
-
-STATIC void
-xfs_qm_dqput_final(
-	struct xfs_dquot	*dqp)
-{
-	struct xfs_quotainfo	*qi = dqp->q_mount->m_quotainfo;
-	struct xfs_dquot	*gdqp;
-	struct xfs_dquot	*pdqp;
-
-	trace_xfs_dqput_free(dqp);
-
-	if (list_lru_add(&qi->qi_lru, &dqp->q_lru))
-		XFS_STATS_INC(xs_qm_dquot_unused);
-
-	/*
-	 * If we just added a udquot to the freelist, then we want to release
-	 * the gdquot/pdquot reference that it (probably) has. Otherwise it'll
-	 * keep the gdquot/pdquot from getting reclaimed.
-	 */
-	gdqp = dqp->q_gdquot;
-	if (gdqp) {
-		xfs_dqlock(gdqp);
-		dqp->q_gdquot = NULL;
-	}
-
-	pdqp = dqp->q_pdquot;
-	if (pdqp) {
-		xfs_dqlock(pdqp);
-		dqp->q_pdquot = NULL;
-	}
-	xfs_dqunlock(dqp);
-
-	/*
-	 * If we had a group/project quota hint, release it now.
-	 */
-	if (gdqp)
-		xfs_qm_dqput(gdqp);
-	if (pdqp)
-		xfs_qm_dqput(pdqp);
+	return 0;
 }
 
 /*
@@ -887,10 +847,14 @@ xfs_qm_dqput(
 
 	trace_xfs_dqput(dqp);
 
-	if (--dqp->q_nrefs > 0)
-		xfs_dqunlock(dqp);
-	else
-		xfs_qm_dqput_final(dqp);
+	if (--dqp->q_nrefs == 0) {
+		struct xfs_quotainfo	*qi = dqp->q_mount->m_quotainfo;
+		trace_xfs_dqput_free(dqp);
+
+		if (list_lru_add(&qi->qi_lru, &dqp->q_lru))
+			XFS_STATS_INC(xs_qm_dquot_unused);
+	}
+	xfs_dqunlock(dqp);
 }
 
 /*
@@ -1002,7 +966,7 @@ xfs_qm_dqflush(
 					     SHUTDOWN_CORRUPT_INCORE);
 		else
 			spin_unlock(&mp->m_ail->xa_lock);
-		error = XFS_ERROR(EIO);
+		error = -EIO;
 		goto out_unlock;
 	}
 
@@ -1010,7 +974,8 @@ xfs_qm_dqflush(
 	 * Get the buffer containing the on-disk dquot
 	 */
 	error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, dqp->q_blkno,
-				   mp->m_quotainfo->qi_dqchunklen, 0, &bp, NULL);
+				   mp->m_quotainfo->qi_dqchunklen, 0, &bp,
+				   &xfs_dquot_buf_ops);
 	if (error)
 		goto out_unlock;
 
@@ -1028,7 +993,7 @@ xfs_qm_dqflush(
 		xfs_buf_relse(bp);
 		xfs_dqfunlock(dqp);
 		xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	/* This is the only portion of data that needs to persist */
@@ -1081,7 +1046,7 @@ xfs_qm_dqflush(
 
 out_unlock:
 	xfs_dqfunlock(dqp);
-	return XFS_ERROR(EIO);
+	return -EIO;
 }
 
 /*
diff --git a/fs/xfs/xfs_dquot.h b/fs/xfs/xfs_dquot.h
index d22ed0053c32..c24c67e22a2a 100644
--- a/fs/xfs/xfs_dquot.h
+++ b/fs/xfs/xfs_dquot.h
@@ -52,8 +52,6 @@ typedef struct xfs_dquot {
 	int		 q_bufoffset;	/* off of dq in buffer (# dquots) */
 	xfs_fileoff_t	 q_fileoffset;	/* offset in quotas file */
 
-	struct xfs_dquot*q_gdquot;	/* group dquot, hint only */
-	struct xfs_dquot*q_pdquot;	/* project dquot, hint only */
 	xfs_disk_dquot_t q_core;	/* actual usage & quotas */
 	xfs_dq_logitem_t q_logitem;	/* dquot log item */
 	xfs_qcnt_t	 q_res_bcount;	/* total regular nblks used+reserved */
@@ -141,6 +139,21 @@ static inline xfs_dquot_t *xfs_inode_dquot(struct xfs_inode *ip, int type)
 	}
 }
 
+/*
+ * Check whether a dquot is under low free space conditions. We assume the quota
+ * is enabled and enforced.
+ */
+static inline bool xfs_dquot_lowsp(struct xfs_dquot *dqp)
+{
+	int64_t freesp;
+
+	freesp = be64_to_cpu(dqp->q_core.d_blk_hardlimit) - dqp->q_res_bcount;
+	if (freesp < dqp->q_low_space[XFS_QLOWSP_1_PCNT])
+		return true;
+
+	return false;
+}
+
 #define XFS_DQ_IS_LOCKED(dqp)	(mutex_is_locked(&((dqp)->q_qlock)))
 #define XFS_DQ_IS_DIRTY(dqp)	((dqp)->dq_flags & XFS_DQ_DIRTY)
 #define XFS_QM_ISUDQ(dqp)	((dqp)->dq_flags & XFS_DQ_USER)
diff --git a/fs/xfs/xfs_dquot_item.c b/fs/xfs/xfs_dquot_item.c
index 92e5f62eefc6..f33fbaaa4d8a 100644
--- a/fs/xfs/xfs_dquot_item.c
+++ b/fs/xfs/xfs_dquot_item.c
@@ -57,20 +57,24 @@ xfs_qm_dquot_logitem_size(
 STATIC void
 xfs_qm_dquot_logitem_format(
 	struct xfs_log_item	*lip,
-	struct xfs_log_iovec	*logvec)
+	struct xfs_log_vec	*lv)
 {
 	struct xfs_dq_logitem	*qlip = DQUOT_ITEM(lip);
-
-	logvec->i_addr = &qlip->qli_format;
-	logvec->i_len  = sizeof(xfs_dq_logformat_t);
-	logvec->i_type = XLOG_REG_TYPE_QFORMAT;
-	logvec++;
-	logvec->i_addr = &qlip->qli_dquot->q_core;
-	logvec->i_len  = sizeof(xfs_disk_dquot_t);
-	logvec->i_type = XLOG_REG_TYPE_DQUOT;
-
-	qlip->qli_format.qlf_size = 2;
-
+	struct xfs_log_iovec	*vecp = NULL;
+	struct xfs_dq_logformat	*qlf;
+
+	qlf = xlog_prepare_iovec(lv, &vecp, XLOG_REG_TYPE_QFORMAT);
+	qlf->qlf_type = XFS_LI_DQUOT;
+	qlf->qlf_size = 2;
+	qlf->qlf_id = be32_to_cpu(qlip->qli_dquot->q_core.d_id);
+	qlf->qlf_blkno = qlip->qli_dquot->q_blkno;
+	qlf->qlf_len = 1;
+	qlf->qlf_boffset = qlip->qli_dquot->q_bufoffset;
+	xlog_finish_iovec(lv, vecp, sizeof(struct xfs_dq_logformat));
+
+	xlog_copy_iovec(lv, &vecp, XLOG_REG_TYPE_DQUOT,
+			&qlip->qli_dquot->q_core,
+			sizeof(struct xfs_disk_dquot));
 }
 
 /*
@@ -257,18 +261,6 @@ xfs_qm_dquot_logitem_init(
 	xfs_log_item_init(dqp->q_mount, &lp->qli_item, XFS_LI_DQUOT,
 					&xfs_dquot_item_ops);
 	lp->qli_dquot = dqp;
-	lp->qli_format.qlf_type = XFS_LI_DQUOT;
-	lp->qli_format.qlf_id = be32_to_cpu(dqp->q_core.d_id);
-	lp->qli_format.qlf_blkno = dqp->q_blkno;
-	lp->qli_format.qlf_len = 1;
-	/*
-	 * This is just the offset of this dquot within its buffer
-	 * (which is currently 1 FSB and probably won't change).
-	 * Hence 32 bits for this offset should be just fine.
-	 * Alternatively, we can store (bufoffset / sizeof(xfs_dqblk_t))
-	 * here, and recompute it at recovery time.
-	 */
-	lp->qli_format.qlf_boffset = (__uint32_t)dqp->q_bufoffset;
 }
 
 /*------------------  QUOTAOFF LOG ITEMS  -------------------*/
@@ -294,26 +286,20 @@ xfs_qm_qoff_logitem_size(
 	*nbytes += sizeof(struct xfs_qoff_logitem);
 }
 
-/*
- * This is called to fill in the vector of log iovecs for the
- * given quotaoff log item. We use only 1 iovec, and we point that
- * at the quotaoff_log_format structure embedded in the quotaoff item.
- * It is at this point that we assert that all of the extent
- * slots in the quotaoff item have been filled.
- */
 STATIC void
 xfs_qm_qoff_logitem_format(
 	struct xfs_log_item	*lip,
-	struct xfs_log_iovec	*log_vector)
+	struct xfs_log_vec	*lv)
 {
 	struct xfs_qoff_logitem	*qflip = QOFF_ITEM(lip);
-
-	ASSERT(qflip->qql_format.qf_type == XFS_LI_QUOTAOFF);
-
-	log_vector->i_addr = &qflip->qql_format;
-	log_vector->i_len = sizeof(xfs_qoff_logitem_t);
-	log_vector->i_type = XLOG_REG_TYPE_QUOTAOFF;
-	qflip->qql_format.qf_size = 1;
+	struct xfs_log_iovec	*vecp = NULL;
+	struct xfs_qoff_logformat *qlf;
+
+	qlf = xlog_prepare_iovec(lv, &vecp, XLOG_REG_TYPE_QUOTAOFF);
+	qlf->qf_type = XFS_LI_QUOTAOFF;
+	qlf->qf_size = 1;
+	qlf->qf_flags = qflip->qql_flags;
+	xlog_finish_iovec(lv, vecp, sizeof(struct xfs_qoff_logitem));
 }
 
 /*
@@ -453,8 +439,7 @@ xfs_qm_qoff_logitem_init(
 	xfs_log_item_init(mp, &qf->qql_item, XFS_LI_QUOTAOFF, start ?
 			&xfs_qm_qoffend_logitem_ops : &xfs_qm_qoff_logitem_ops);
 	qf->qql_item.li_mountp = mp;
-	qf->qql_format.qf_type = XFS_LI_QUOTAOFF;
-	qf->qql_format.qf_flags = flags;
 	qf->qql_start_lip = start;
+	qf->qql_flags = flags;
 	return qf;
 }
diff --git a/fs/xfs/xfs_dquot_item.h b/fs/xfs/xfs_dquot_item.h
index 5acae2ada70b..502e9464634a 100644
--- a/fs/xfs/xfs_dquot_item.h
+++ b/fs/xfs/xfs_dquot_item.h
@@ -27,13 +27,12 @@ typedef struct xfs_dq_logitem {
 	xfs_log_item_t		 qli_item;	   /* common portion */
 	struct xfs_dquot	*qli_dquot;	   /* dquot ptr */
 	xfs_lsn_t		 qli_flush_lsn;	   /* lsn at last flush */
-	xfs_dq_logformat_t	 qli_format;	   /* logged structure */
 } xfs_dq_logitem_t;
 
 typedef struct xfs_qoff_logitem {
 	xfs_log_item_t		 qql_item;	/* common portion */
 	struct xfs_qoff_logitem *qql_start_lip; /* qoff-start logitem, if any */
-	xfs_qoff_logformat_t	 qql_format;	/* logged structure */
+	unsigned int		qql_flags;
 } xfs_qoff_logitem_t;
 
 
diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c
index 9995b807d627..b92fd7bc49e3 100644
--- a/fs/xfs/xfs_error.c
+++ b/fs/xfs/xfs_error.c
@@ -27,29 +27,6 @@
 
 #ifdef DEBUG
 
-int	xfs_etrap[XFS_ERROR_NTRAP] = {
-	0,
-};
-
-int
-xfs_error_trap(int e)
-{
-	int i;
-
-	if (!e)
-		return 0;
-	for (i = 0; i < XFS_ERROR_NTRAP; i++) {
-		if (xfs_etrap[i] == 0)
-			break;
-		if (e != xfs_etrap[i])
-			continue;
-		xfs_notice(NULL, "%s: error %d", __func__, e);
-		BUG();
-		break;
-	}
-	return e;
-}
-
 int	xfs_etest[XFS_NUM_INJECT_ERROR];
 int64_t	xfs_etest_fsid[XFS_NUM_INJECT_ERROR];
 char *	xfs_etest_fsname[XFS_NUM_INJECT_ERROR];
@@ -156,7 +133,7 @@ xfs_error_report(
 {
 	if (level <= xfs_error_level) {
 		xfs_alert_tag(mp, XFS_PTAG_ERROR_REPORT,
-		"Internal error %s at line %d of file %s.  Caller 0x%p",
+		"Internal error %s at line %d of file %s.  Caller %pF",
 			    tag, linenum, filename, ra);
 
 		xfs_stack_trace();
@@ -178,3 +155,28 @@ xfs_corruption_error(
 	xfs_error_report(tag, level, mp, filename, linenum, ra);
 	xfs_alert(mp, "Corruption detected. Unmount and run xfs_repair");
 }
+
+/*
+ * Warnings specifically for verifier errors.  Differentiate CRC vs. invalid
+ * values, and omit the stack trace unless the error level is tuned high.
+ */
+void
+xfs_verifier_error(
+	struct xfs_buf		*bp)
+{
+	struct xfs_mount *mp = bp->b_target->bt_mount;
+
+	xfs_alert(mp, "Metadata %s detected at %pF, block 0x%llx",
+		  bp->b_error == -EFSBADCRC ? "CRC error" : "corruption",
+		  __return_address, bp->b_bn);
+
+	xfs_alert(mp, "Unmount and run xfs_repair");
+
+	if (xfs_error_level >= XFS_ERRLEVEL_LOW) {
+		xfs_alert(mp, "First 64 bytes of corrupted metadata buffer:");
+		xfs_hex_dump(xfs_buf_offset(bp, 0), 64);
+	}
+
+	if (xfs_error_level >= XFS_ERRLEVEL_HIGH)
+		xfs_stack_trace();
+}
diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h
index 079a367f44ee..279a76e52791 100644
--- a/fs/xfs/xfs_error.h
+++ b/fs/xfs/xfs_error.h
@@ -18,15 +18,6 @@
 #ifndef	__XFS_ERROR_H__
 #define	__XFS_ERROR_H__
 
-#ifdef DEBUG
-#define	XFS_ERROR_NTRAP	10
-extern int	xfs_etrap[XFS_ERROR_NTRAP];
-extern int	xfs_error_trap(int);
-#define	XFS_ERROR(e)	xfs_error_trap(e)
-#else
-#define	XFS_ERROR(e)	(e)
-#endif
-
 struct xfs_mount;
 
 extern void xfs_error_report(const char *tag, int level, struct xfs_mount *mp,
@@ -34,6 +25,7 @@ extern void xfs_error_report(const char *tag, int level, struct xfs_mount *mp,
 extern void xfs_corruption_error(const char *tag, int level,
 			struct xfs_mount *mp, void *p, const char *filename,
 			int linenum, inst_t *ra);
+extern void xfs_verifier_error(struct xfs_buf *bp);
 
 #define	XFS_ERROR_REPORT(e, lvl, mp)	\
 	xfs_error_report(e, lvl, mp, __FILE__, __LINE__, __return_address)
@@ -55,7 +47,7 @@ extern void xfs_corruption_error(const char *tag, int level,
 		if (unlikely(!fs_is_ok)) { \
 			XFS_ERROR_REPORT("XFS_WANT_CORRUPTED_GOTO", \
 					 XFS_ERRLEVEL_LOW, NULL); \
-			error = XFS_ERROR(EFSCORRUPTED); \
+			error = -EFSCORRUPTED; \
 			goto l; \
 		} \
 	}
@@ -67,7 +59,7 @@ extern void xfs_corruption_error(const char *tag, int level,
 		if (unlikely(!fs_is_ok)) { \
 			XFS_ERROR_REPORT("XFS_WANT_CORRUPTED_RETURN", \
 					 XFS_ERRLEVEL_LOW, NULL); \
-			return XFS_ERROR(EFSCORRUPTED); \
+			return -EFSCORRUPTED; \
 		} \
 	}
 
diff --git a/fs/xfs/xfs_export.c b/fs/xfs/xfs_export.c
index 1399e187d425..5a6bd5d8779a 100644
--- a/fs/xfs/xfs_export.c
+++ b/fs/xfs/xfs_export.c
@@ -147,9 +147,9 @@ xfs_nfs_get_inode(
 		 * We don't use ESTALE directly down the chain to not
 		 * confuse applications using bulkstat that expect EINVAL.
 		 */
-		if (error == EINVAL || error == ENOENT)
-			error = ESTALE;
-		return ERR_PTR(-error);
+		if (error == -EINVAL || error == -ENOENT)
+			error = -ESTALE;
+		return ERR_PTR(error);
 	}
 
 	if (ip->i_d.di_gen != generation) {
@@ -217,7 +217,7 @@ xfs_fs_get_parent(
 
 	error = xfs_lookup(XFS_I(child->d_inode), &xfs_name_dotdot, &cip, NULL);
 	if (unlikely(error))
-		return ERR_PTR(-error);
+		return ERR_PTR(error);
 
 	return d_obtain_alias(VFS_I(cip));
 }
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index 3680d04f973f..c4327419dc5c 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -26,6 +26,7 @@
 #include "xfs_trans_priv.h"
 #include "xfs_buf_item.h"
 #include "xfs_extfree_item.h"
+#include "xfs_log.h"
 
 
 kmem_zone_t	*xfs_efi_zone;
@@ -101,9 +102,10 @@ xfs_efi_item_size(
 STATIC void
 xfs_efi_item_format(
 	struct xfs_log_item	*lip,
-	struct xfs_log_iovec	*log_vector)
+	struct xfs_log_vec	*lv)
 {
 	struct xfs_efi_log_item	*efip = EFI_ITEM(lip);
+	struct xfs_log_iovec	*vecp = NULL;
 
 	ASSERT(atomic_read(&efip->efi_next_extent) ==
 				efip->efi_format.efi_nextents);
@@ -111,10 +113,9 @@ xfs_efi_item_format(
 	efip->efi_format.efi_type = XFS_LI_EFI;
 	efip->efi_format.efi_size = 1;
 
-	log_vector->i_addr = &efip->efi_format;
-	log_vector->i_len = xfs_efi_item_sizeof(efip);
-	log_vector->i_type = XLOG_REG_TYPE_EFI_FORMAT;
-	ASSERT(log_vector->i_len >= sizeof(xfs_efi_log_format_t));
+	xlog_copy_iovec(lv, &vecp, XLOG_REG_TYPE_EFI_FORMAT,
+			&efip->efi_format,
+			xfs_efi_item_sizeof(efip));
 }
 
 
@@ -297,7 +298,7 @@ xfs_efi_copy_format(xfs_log_iovec_t *buf, xfs_efi_log_format_t *dst_efi_fmt)
 		}
 		return 0;
 	}
-	return EFSCORRUPTED;
+	return -EFSCORRUPTED;
 }
 
 /*
@@ -368,19 +369,19 @@ xfs_efd_item_size(
 STATIC void
 xfs_efd_item_format(
 	struct xfs_log_item	*lip,
-	struct xfs_log_iovec	*log_vector)
+	struct xfs_log_vec	*lv)
 {
 	struct xfs_efd_log_item	*efdp = EFD_ITEM(lip);
+	struct xfs_log_iovec	*vecp = NULL;
 
 	ASSERT(efdp->efd_next_extent == efdp->efd_format.efd_nextents);
 
 	efdp->efd_format.efd_type = XFS_LI_EFD;
 	efdp->efd_format.efd_size = 1;
 
-	log_vector->i_addr = &efdp->efd_format;
-	log_vector->i_len = xfs_efd_item_sizeof(efdp);
-	log_vector->i_type = XLOG_REG_TYPE_EFD_FORMAT;
-	ASSERT(log_vector->i_len >= sizeof(xfs_efd_log_format_t));
+	xlog_copy_iovec(lv, &vecp, XLOG_REG_TYPE_EFD_FORMAT,
+			&efdp->efd_format,
+			xfs_efd_item_sizeof(efdp));
 }
 
 /*
diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index 52c91e143725..de5368c803f9 100644
--- a/fs/xfs/xfs_file.c
+++ b/fs/xfs/xfs_file.c
@@ -38,6 +38,7 @@
 #include "xfs_trace.h"
 #include "xfs_log.h"
 #include "xfs_dinode.h"
+#include "xfs_icache.h"
 
 #include <linux/aio.h>
 #include <linux/dcache.h>
@@ -179,7 +180,7 @@ xfs_file_fsync(
 		return error;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+		return -EIO;
 
 	xfs_iflags_clear(ip, XFS_ITRUNCATED);
 
@@ -225,46 +226,40 @@ xfs_file_fsync(
 	    !log_flushed)
 		xfs_blkdev_issue_flush(mp->m_ddev_targp);
 
-	return -error;
+	return error;
 }
 
 STATIC ssize_t
-xfs_file_aio_read(
+xfs_file_read_iter(
 	struct kiocb		*iocb,
-	const struct iovec	*iovp,
-	unsigned long		nr_segs,
-	loff_t			pos)
+	struct iov_iter		*to)
 {
 	struct file		*file = iocb->ki_filp;
 	struct inode		*inode = file->f_mapping->host;
 	struct xfs_inode	*ip = XFS_I(inode);
 	struct xfs_mount	*mp = ip->i_mount;
-	size_t			size = 0;
+	size_t			size = iov_iter_count(to);
 	ssize_t			ret = 0;
 	int			ioflags = 0;
 	xfs_fsize_t		n;
+	loff_t			pos = iocb->ki_pos;
 
 	XFS_STATS_INC(xs_read_calls);
 
-	BUG_ON(iocb->ki_pos != pos);
-
 	if (unlikely(file->f_flags & O_DIRECT))
-		ioflags |= IO_ISDIRECT;
+		ioflags |= XFS_IO_ISDIRECT;
 	if (file->f_mode & FMODE_NOCMTIME)
-		ioflags |= IO_INVIS;
+		ioflags |= XFS_IO_INVIS;
 
-	ret = generic_segment_checks(iovp, &nr_segs, &size, VERIFY_WRITE);
-	if (ret < 0)
-		return ret;
-
-	if (unlikely(ioflags & IO_ISDIRECT)) {
+	if (unlikely(ioflags & XFS_IO_ISDIRECT)) {
 		xfs_buftarg_t	*target =
 			XFS_IS_REALTIME_INODE(ip) ?
 				mp->m_rtdev_targp : mp->m_ddev_targp;
-		if ((pos & target->bt_smask) || (size & target->bt_smask)) {
+		/* DIO must be aligned to device logical sector size */
+		if ((pos | size) & target->bt_logical_sectormask) {
 			if (pos == i_size_read(inode))
 				return 0;
-			return -XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 	}
 
@@ -289,26 +284,36 @@ xfs_file_aio_read(
 	 * proceeed concurrently without serialisation.
 	 */
 	xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
-	if ((ioflags & IO_ISDIRECT) && inode->i_mapping->nrpages) {
+	if ((ioflags & XFS_IO_ISDIRECT) && inode->i_mapping->nrpages) {
 		xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
 		xfs_rw_ilock(ip, XFS_IOLOCK_EXCL);
 
 		if (inode->i_mapping->nrpages) {
-			ret = -filemap_write_and_wait_range(
+			ret = filemap_write_and_wait_range(
 							VFS_I(ip)->i_mapping,
-							pos, -1);
+							pos, pos + size - 1);
 			if (ret) {
 				xfs_rw_iunlock(ip, XFS_IOLOCK_EXCL);
 				return ret;
 			}
-			truncate_pagecache_range(VFS_I(ip), pos, -1);
+
+			/*
+			 * Invalidate whole pages. This can return an error if
+			 * we fail to invalidate a page, but this should never
+			 * happen on XFS. Warn if it does fail.
+			 */
+			ret = invalidate_inode_pages2_range(VFS_I(ip)->i_mapping,
+					pos >> PAGE_CACHE_SHIFT,
+					(pos + size - 1) >> PAGE_CACHE_SHIFT);
+			WARN_ON_ONCE(ret);
+			ret = 0;
 		}
 		xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
 	}
 
 	trace_xfs_file_read(ip, size, pos, ioflags);
 
-	ret = generic_file_aio_read(iocb, iovp, nr_segs, pos);
+	ret = generic_file_read_iter(iocb, to);
 	if (ret > 0)
 		XFS_STATS_ADD(xs_read_bytes, ret);
 
@@ -331,7 +336,7 @@ xfs_file_splice_read(
 	XFS_STATS_INC(xs_read_calls);
 
 	if (infilp->f_mode & FMODE_NOCMTIME)
-		ioflags |= IO_INVIS;
+		ioflags |= XFS_IO_INVIS;
 
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
 		return -EIO;
@@ -349,47 +354,6 @@ xfs_file_splice_read(
 }
 
 /*
- * xfs_file_splice_write() does not use xfs_rw_ilock() because
- * generic_file_splice_write() takes the i_mutex itself. This, in theory,
- * couuld cause lock inversions between the aio_write path and the splice path
- * if someone is doing concurrent splice(2) based writes and write(2) based
- * writes to the same inode. The only real way to fix this is to re-implement
- * the generic code here with correct locking orders.
- */
-STATIC ssize_t
-xfs_file_splice_write(
-	struct pipe_inode_info	*pipe,
-	struct file		*outfilp,
-	loff_t			*ppos,
-	size_t			count,
-	unsigned int		flags)
-{
-	struct inode		*inode = outfilp->f_mapping->host;
-	struct xfs_inode	*ip = XFS_I(inode);
-	int			ioflags = 0;
-	ssize_t			ret;
-
-	XFS_STATS_INC(xs_write_calls);
-
-	if (outfilp->f_mode & FMODE_NOCMTIME)
-		ioflags |= IO_INVIS;
-
-	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
-		return -EIO;
-
-	xfs_ilock(ip, XFS_IOLOCK_EXCL);
-
-	trace_xfs_file_splice_write(ip, count, *ppos, ioflags);
-
-	ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags);
-	if (ret > 0)
-		XFS_STATS_ADD(xs_write_bytes, ret);
-
-	xfs_iunlock(ip, XFS_IOLOCK_EXCL);
-	return ret;
-}
-
-/*
  * This routine is called to handle zeroing any space in the last block of the
  * file that is beyond the EOF.  We do this since the size is being increased
  * without writing anything to that block and we don't want to read the
@@ -571,7 +535,7 @@ restart:
 			xfs_rw_ilock(ip, *iolock);
 			goto restart;
 		}
-		error = -xfs_zero_eof(ip, *pos, i_size_read(inode));
+		error = xfs_zero_eof(ip, *pos, i_size_read(inode));
 		if (error)
 			return error;
 	}
@@ -624,10 +588,7 @@ restart:
 STATIC ssize_t
 xfs_file_dio_aio_write(
 	struct kiocb		*iocb,
-	const struct iovec	*iovp,
-	unsigned long		nr_segs,
-	loff_t			pos,
-	size_t			ocount)
+	struct iov_iter		*from)
 {
 	struct file		*file = iocb->ki_filp;
 	struct address_space	*mapping = file->f_mapping;
@@ -635,15 +596,18 @@ xfs_file_dio_aio_write(
 	struct xfs_inode	*ip = XFS_I(inode);
 	struct xfs_mount	*mp = ip->i_mount;
 	ssize_t			ret = 0;
-	size_t			count = ocount;
 	int			unaligned_io = 0;
 	int			iolock;
+	size_t			count = iov_iter_count(from);
+	loff_t			pos = iocb->ki_pos;
 	struct xfs_buftarg	*target = XFS_IS_REALTIME_INODE(ip) ?
 					mp->m_rtdev_targp : mp->m_ddev_targp;
 
-	if ((pos & target->bt_smask) || (count & target->bt_smask))
-		return -XFS_ERROR(EINVAL);
+	/* DIO must be aligned to device logical sector size */
+	if ((pos | count) & target->bt_logical_sectormask)
+		return -EINVAL;
 
+	/* "unaligned" here means not aligned to a filesystem block */
 	if ((pos & mp->m_blockmask) || ((pos + count) & mp->m_blockmask))
 		unaligned_io = 1;
 
@@ -674,13 +638,23 @@ xfs_file_dio_aio_write(
 	ret = xfs_file_aio_write_checks(file, &pos, &count, &iolock);
 	if (ret)
 		goto out;
+	iov_iter_truncate(from, count);
 
 	if (mapping->nrpages) {
-		ret = -filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
-						    pos, -1);
+		ret = filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
+						    pos, pos + count - 1);
 		if (ret)
 			goto out;
-		truncate_pagecache_range(VFS_I(ip), pos, -1);
+		/*
+		 * Invalidate whole pages. This can return an error if
+		 * we fail to invalidate a page, but this should never
+		 * happen on XFS. Warn if it does fail.
+		 */
+		ret = invalidate_inode_pages2_range(VFS_I(ip)->i_mapping,
+					pos >> PAGE_CACHE_SHIFT,
+					(pos + count - 1) >> PAGE_CACHE_SHIFT);
+		WARN_ON_ONCE(ret);
+		ret = 0;
 	}
 
 	/*
@@ -695,8 +669,7 @@ xfs_file_dio_aio_write(
 	}
 
 	trace_xfs_file_direct_write(ip, count, iocb->ki_pos, 0);
-	ret = generic_file_direct_write(iocb, iovp,
-			&nr_segs, pos, &iocb->ki_pos, count, ocount);
+	ret = generic_file_direct_write(iocb, from, pos);
 
 out:
 	xfs_rw_iunlock(ip, iolock);
@@ -709,10 +682,7 @@ out:
 STATIC ssize_t
 xfs_file_buffered_aio_write(
 	struct kiocb		*iocb,
-	const struct iovec	*iovp,
-	unsigned long		nr_segs,
-	loff_t			pos,
-	size_t			ocount)
+	struct iov_iter		*from)
 {
 	struct file		*file = iocb->ki_filp;
 	struct address_space	*mapping = file->f_mapping;
@@ -721,7 +691,8 @@ xfs_file_buffered_aio_write(
 	ssize_t			ret;
 	int			enospc = 0;
 	int			iolock = XFS_IOLOCK_EXCL;
-	size_t			count = ocount;
+	loff_t			pos = iocb->ki_pos;
+	size_t			count = iov_iter_count(from);
 
 	xfs_rw_ilock(ip, iolock);
 
@@ -729,22 +700,37 @@ xfs_file_buffered_aio_write(
 	if (ret)
 		goto out;
 
+	iov_iter_truncate(from, count);
 	/* We can write back this queue in page reclaim */
 	current->backing_dev_info = mapping->backing_dev_info;
 
 write_retry:
 	trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, 0);
-	ret = generic_file_buffered_write(iocb, iovp, nr_segs,
-			pos, &iocb->ki_pos, count, 0);
+	ret = generic_perform_write(file, from, pos);
+	if (likely(ret >= 0))
+		iocb->ki_pos = pos + ret;
 
 	/*
-	 * If we just got an ENOSPC, try to write back all dirty inodes to
-	 * convert delalloc space to free up some of the excess reserved
-	 * metadata space.
+	 * If we hit a space limit, try to free up some lingering preallocated
+	 * space before returning an error. In the case of ENOSPC, first try to
+	 * write back all dirty inodes to free up some of the excess reserved
+	 * metadata space. This reduces the chances that the eofblocks scan
+	 * waits on dirty mappings. Since xfs_flush_inodes() is serialized, this
+	 * also behaves as a filter to prevent too many eofblocks scans from
+	 * running at the same time.
 	 */
-	if (ret == -ENOSPC && !enospc) {
+	if (ret == -EDQUOT && !enospc) {
+		enospc = xfs_inode_free_quota_eofblocks(ip);
+		if (enospc)
+			goto write_retry;
+	} else if (ret == -ENOSPC && !enospc) {
+		struct xfs_eofblocks eofb = {0};
+
 		enospc = 1;
 		xfs_flush_inodes(ip->i_mount);
+		eofb.eof_scan_owner = ip->i_ino; /* for locking */
+		eofb.eof_flags = XFS_EOF_FLAGS_SYNC;
+		xfs_icache_free_eofblocks(ip->i_mount, &eofb);
 		goto write_retry;
 	}
 
@@ -755,40 +741,29 @@ out:
 }
 
 STATIC ssize_t
-xfs_file_aio_write(
+xfs_file_write_iter(
 	struct kiocb		*iocb,
-	const struct iovec	*iovp,
-	unsigned long		nr_segs,
-	loff_t			pos)
+	struct iov_iter		*from)
 {
 	struct file		*file = iocb->ki_filp;
 	struct address_space	*mapping = file->f_mapping;
 	struct inode		*inode = mapping->host;
 	struct xfs_inode	*ip = XFS_I(inode);
 	ssize_t			ret;
-	size_t			ocount = 0;
+	size_t			ocount = iov_iter_count(from);
 
 	XFS_STATS_INC(xs_write_calls);
 
-	BUG_ON(iocb->ki_pos != pos);
-
-	ret = generic_segment_checks(iovp, &nr_segs, &ocount, VERIFY_READ);
-	if (ret)
-		return ret;
-
 	if (ocount == 0)
 		return 0;
 
-	if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
-		ret = -EIO;
-		goto out;
-	}
+	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+		return -EIO;
 
 	if (unlikely(file->f_flags & O_DIRECT))
-		ret = xfs_file_dio_aio_write(iocb, iovp, nr_segs, pos, ocount);
+		ret = xfs_file_dio_aio_write(iocb, from);
 	else
-		ret = xfs_file_buffered_aio_write(iocb, iovp, nr_segs, pos,
-						  ocount);
+		ret = xfs_file_buffered_aio_write(iocb, from);
 
 	if (ret > 0) {
 		ssize_t err;
@@ -796,12 +771,10 @@ xfs_file_aio_write(
 		XFS_STATS_ADD(xs_write_bytes, ret);
 
 		/* Handle various SYNC-type writes */
-		err = generic_write_sync(file, pos, ret);
+		err = generic_write_sync(file, iocb->ki_pos - ret, ret);
 		if (err < 0)
 			ret = err;
 	}
-
-out:
 	return ret;
 }
 
@@ -820,7 +793,8 @@ xfs_file_fallocate(
 
 	if (!S_ISREG(inode->i_mode))
 		return -EINVAL;
-	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
+	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE |
+		     FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_ZERO_RANGE))
 		return -EOPNOTSUPP;
 
 	xfs_ilock(ip, XFS_IOLOCK_EXCL);
@@ -828,17 +802,42 @@ xfs_file_fallocate(
 		error = xfs_free_file_space(ip, offset, len);
 		if (error)
 			goto out_unlock;
+	} else if (mode & FALLOC_FL_COLLAPSE_RANGE) {
+		unsigned blksize_mask = (1 << inode->i_blkbits) - 1;
+
+		if (offset & blksize_mask || len & blksize_mask) {
+			error = -EINVAL;
+			goto out_unlock;
+		}
+
+		/*
+		 * There is no need to overlap collapse range with EOF,
+		 * in which case it is effectively a truncate operation
+		 */
+		if (offset + len >= i_size_read(inode)) {
+			error = -EINVAL;
+			goto out_unlock;
+		}
+
+		new_size = i_size_read(inode) - len;
+
+		error = xfs_collapse_file_space(ip, offset, len);
+		if (error)
+			goto out_unlock;
 	} else {
 		if (!(mode & FALLOC_FL_KEEP_SIZE) &&
 		    offset + len > i_size_read(inode)) {
 			new_size = offset + len;
-			error = -inode_newsize_ok(inode, new_size);
+			error = inode_newsize_ok(inode, new_size);
 			if (error)
 				goto out_unlock;
 		}
 
-		error = xfs_alloc_file_space(ip, offset, len,
-					     XFS_BMAPI_PREALLOC);
+		if (mode & FALLOC_FL_ZERO_RANGE)
+			error = xfs_zero_file_space(ip, offset, len);
+		else
+			error = xfs_alloc_file_space(ip, offset, len,
+						     XFS_BMAPI_PREALLOC);
 		if (error)
 			goto out_unlock;
 	}
@@ -856,7 +855,7 @@ xfs_file_fallocate(
 	if (ip->i_d.di_mode & S_IXGRP)
 		ip->i_d.di_mode &= ~S_ISGID;
 
-	if (!(mode & FALLOC_FL_PUNCH_HOLE))
+	if (!(mode & (FALLOC_FL_PUNCH_HOLE | FALLOC_FL_COLLAPSE_RANGE)))
 		ip->i_d.di_flags |= XFS_DIFLAG_PREALLOC;
 
 	xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
@@ -879,7 +878,7 @@ xfs_file_fallocate(
 
 out_unlock:
 	xfs_iunlock(ip, XFS_IOLOCK_EXCL);
-	return -error;
+	return error;
 }
 
 
@@ -912,9 +911,9 @@ xfs_dir_open(
 	 * If there are any blocks, read-ahead block 0 as we're almost
 	 * certain to have the next operation be a read there.
 	 */
-	mode = xfs_ilock_map_shared(ip);
+	mode = xfs_ilock_data_map_shared(ip);
 	if (ip->i_d.di_nextents > 0)
-		xfs_dir3_data_readahead(NULL, ip, 0, -1);
+		xfs_dir3_data_readahead(ip, 0, -1);
 	xfs_iunlock(ip, mode);
 	return 0;
 }
@@ -924,7 +923,7 @@ xfs_file_release(
 	struct inode	*inode,
 	struct file	*filp)
 {
-	return -xfs_release(XFS_I(inode));
+	return xfs_release(XFS_I(inode));
 }
 
 STATIC int
@@ -953,7 +952,7 @@ xfs_file_readdir(
 
 	error = xfs_readdir(ip, ctx, bufsize);
 	if (error)
-		return -error;
+		return error;
 	return 0;
 }
 
@@ -1215,11 +1214,11 @@ xfs_seek_data(
 	uint			lock;
 	int			error;
 
-	lock = xfs_ilock_map_shared(ip);
+	lock = xfs_ilock_data_map_shared(ip);
 
 	isize = i_size_read(inode);
 	if (start >= isize) {
-		error = ENXIO;
+		error = -ENXIO;
 		goto out_unlock;
 	}
 
@@ -1241,7 +1240,7 @@ xfs_seek_data(
 
 		/* No extents at given offset, must be beyond EOF */
 		if (nmap == 0) {
-			error = ENXIO;
+			error = -ENXIO;
 			goto out_unlock;
 		}
 
@@ -1272,7 +1271,7 @@ xfs_seek_data(
 		 * we are reading after EOF if nothing in map[1].
 		 */
 		if (nmap == 1) {
-			error = ENXIO;
+			error = -ENXIO;
 			goto out_unlock;
 		}
 
@@ -1285,7 +1284,7 @@ xfs_seek_data(
 		fsbno = map[i - 1].br_startoff + map[i - 1].br_blockcount;
 		start = XFS_FSB_TO_B(mp, fsbno);
 		if (start >= isize) {
-			error = ENXIO;
+			error = -ENXIO;
 			goto out_unlock;
 		}
 	}
@@ -1294,10 +1293,10 @@ out:
 	offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes);
 
 out_unlock:
-	xfs_iunlock_map_shared(ip, lock);
+	xfs_iunlock(ip, lock);
 
 	if (error)
-		return -error;
+		return error;
 	return offset;
 }
 
@@ -1317,13 +1316,13 @@ xfs_seek_hole(
 	int			error;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+		return -EIO;
 
-	lock = xfs_ilock_map_shared(ip);
+	lock = xfs_ilock_data_map_shared(ip);
 
 	isize = i_size_read(inode);
 	if (start >= isize) {
-		error = ENXIO;
+		error = -ENXIO;
 		goto out_unlock;
 	}
 
@@ -1342,7 +1341,7 @@ xfs_seek_hole(
 
 		/* No extents at given offset, must be beyond EOF */
 		if (nmap == 0) {
-			error = ENXIO;
+			error = -ENXIO;
 			goto out_unlock;
 		}
 
@@ -1402,10 +1401,10 @@ out:
 	offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes);
 
 out_unlock:
-	xfs_iunlock_map_shared(ip, lock);
+	xfs_iunlock(ip, lock);
 
 	if (error)
-		return -error;
+		return error;
 	return offset;
 }
 
@@ -1431,12 +1430,12 @@ xfs_file_llseek(
 
 const struct file_operations xfs_file_operations = {
 	.llseek		= xfs_file_llseek,
-	.read		= do_sync_read,
-	.write		= do_sync_write,
-	.aio_read	= xfs_file_aio_read,
-	.aio_write	= xfs_file_aio_write,
+	.read		= new_sync_read,
+	.write		= new_sync_write,
+	.read_iter	= xfs_file_read_iter,
+	.write_iter	= xfs_file_write_iter,
 	.splice_read	= xfs_file_splice_read,
-	.splice_write	= xfs_file_splice_write,
+	.splice_write	= iter_file_splice_write,
 	.unlocked_ioctl	= xfs_file_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl	= xfs_file_compat_ioctl,
@@ -1462,6 +1461,7 @@ const struct file_operations xfs_dir_file_operations = {
 
 static const struct vm_operations_struct xfs_file_vm_ops = {
 	.fault		= filemap_fault,
+	.map_pages	= filemap_map_pages,
 	.page_mkwrite	= xfs_vm_page_mkwrite,
 	.remap_pages	= generic_file_remap_pages,
 };
diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c
index 12b6e7701985..e92730c1d3ca 100644
--- a/fs/xfs/xfs_filestream.c
+++ b/fs/xfs/xfs_filestream.c
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2006-2007 Silicon Graphics, Inc.
+ * Copyright (c) 2014 Christoph Hellwig.
  * All Rights Reserved.
  *
  * This program is free software; you can redistribute it and/or
@@ -32,100 +33,20 @@
 #include "xfs_filestream.h"
 #include "xfs_trace.h"
 
-#ifdef XFS_FILESTREAMS_TRACE
-
-ktrace_t *xfs_filestreams_trace_buf;
-
-STATIC void
-xfs_filestreams_trace(
-	xfs_mount_t	*mp,	/* mount point */
-	int		type,	/* type of trace */
-	const char	*func,	/* source function */
-	int		line,	/* source line number */
-	__psunsigned_t	arg0,
-	__psunsigned_t	arg1,
-	__psunsigned_t	arg2,
-	__psunsigned_t	arg3,
-	__psunsigned_t	arg4,
-	__psunsigned_t	arg5)
-{
-	ktrace_enter(xfs_filestreams_trace_buf,
-		(void *)(__psint_t)(type | (line << 16)),
-		(void *)func,
-		(void *)(__psunsigned_t)current_pid(),
-		(void *)mp,
-		(void *)(__psunsigned_t)arg0,
-		(void *)(__psunsigned_t)arg1,
-		(void *)(__psunsigned_t)arg2,
-		(void *)(__psunsigned_t)arg3,
-		(void *)(__psunsigned_t)arg4,
-		(void *)(__psunsigned_t)arg5,
-		NULL, NULL, NULL, NULL, NULL, NULL);
-}
-
-#define TRACE0(mp,t)			TRACE6(mp,t,0,0,0,0,0,0)
-#define TRACE1(mp,t,a0)			TRACE6(mp,t,a0,0,0,0,0,0)
-#define TRACE2(mp,t,a0,a1)		TRACE6(mp,t,a0,a1,0,0,0,0)
-#define TRACE3(mp,t,a0,a1,a2)		TRACE6(mp,t,a0,a1,a2,0,0,0)
-#define TRACE4(mp,t,a0,a1,a2,a3)	TRACE6(mp,t,a0,a1,a2,a3,0,0)
-#define TRACE5(mp,t,a0,a1,a2,a3,a4)	TRACE6(mp,t,a0,a1,a2,a3,a4,0)
-#define TRACE6(mp,t,a0,a1,a2,a3,a4,a5) \
-	xfs_filestreams_trace(mp, t, __func__, __LINE__, \
-				(__psunsigned_t)a0, (__psunsigned_t)a1, \
-				(__psunsigned_t)a2, (__psunsigned_t)a3, \
-				(__psunsigned_t)a4, (__psunsigned_t)a5)
-
-#define TRACE_AG_SCAN(mp, ag, ag2) \
-		TRACE2(mp, XFS_FSTRM_KTRACE_AGSCAN, ag, ag2);
-#define TRACE_AG_PICK1(mp, max_ag, maxfree) \
-		TRACE2(mp, XFS_FSTRM_KTRACE_AGPICK1, max_ag, maxfree);
-#define TRACE_AG_PICK2(mp, ag, ag2, cnt, free, scan, flag) \
-		TRACE6(mp, XFS_FSTRM_KTRACE_AGPICK2, ag, ag2, \
-			 cnt, free, scan, flag)
-#define TRACE_UPDATE(mp, ip, ag, cnt, ag2, cnt2) \
-		TRACE5(mp, XFS_FSTRM_KTRACE_UPDATE, ip, ag, cnt, ag2, cnt2)
-#define TRACE_FREE(mp, ip, pip, ag, cnt) \
-		TRACE4(mp, XFS_FSTRM_KTRACE_FREE, ip, pip, ag, cnt)
-#define TRACE_LOOKUP(mp, ip, pip, ag, cnt) \
-		TRACE4(mp, XFS_FSTRM_KTRACE_ITEM_LOOKUP, ip, pip, ag, cnt)
-#define TRACE_ASSOCIATE(mp, ip, pip, ag, cnt) \
-		TRACE4(mp, XFS_FSTRM_KTRACE_ASSOCIATE, ip, pip, ag, cnt)
-#define TRACE_MOVEAG(mp, ip, pip, oag, ocnt, nag, ncnt) \
-		TRACE6(mp, XFS_FSTRM_KTRACE_MOVEAG, ip, pip, oag, ocnt, nag, ncnt)
-#define TRACE_ORPHAN(mp, ip, ag) \
-		TRACE2(mp, XFS_FSTRM_KTRACE_ORPHAN, ip, ag);
-
-
-#else
-#define TRACE_AG_SCAN(mp, ag, ag2)
-#define TRACE_AG_PICK1(mp, max_ag, maxfree)
-#define TRACE_AG_PICK2(mp, ag, ag2, cnt, free, scan, flag)
-#define TRACE_UPDATE(mp, ip, ag, cnt, ag2, cnt2)
-#define TRACE_FREE(mp, ip, pip, ag, cnt)
-#define TRACE_LOOKUP(mp, ip, pip, ag, cnt)
-#define TRACE_ASSOCIATE(mp, ip, pip, ag, cnt)
-#define TRACE_MOVEAG(mp, ip, pip, oag, ocnt, nag, ncnt)
-#define TRACE_ORPHAN(mp, ip, ag)
-#endif
-
-static kmem_zone_t *item_zone;
+struct xfs_fstrm_item {
+	struct xfs_mru_cache_elem	mru;
+	struct xfs_inode		*ip;
+	xfs_agnumber_t			ag; /* AG in use for this directory */
+};
 
-/*
- * Structure for associating a file or a directory with an allocation group.
- * The parent directory pointer is only needed for files, but since there will
- * generally be vastly more files than directories in the cache, using the same
- * data structure simplifies the code with very little memory overhead.
- */
-typedef struct fstrm_item
-{
-	xfs_agnumber_t	ag;	/* AG currently in use for the file/directory. */
-	xfs_inode_t	*ip;	/* inode self-pointer. */
-	xfs_inode_t	*pip;	/* Parent directory inode pointer. */
-} fstrm_item_t;
+enum xfs_fstrm_alloc {
+	XFS_PICK_USERDATA = 1,
+	XFS_PICK_LOWSPACE = 2,
+};
 
 /*
  * Allocation group filestream associations are tracked with per-ag atomic
- * counters.  These counters allow _xfs_filestream_pick_ag() to tell whether a
+ * counters.  These counters allow xfs_filestream_pick_ag() to tell whether a
  * particular AG already has active filestreams associated with it. The mount
  * point's m_peraglock is used to protect these counters from per-ag array
  * re-allocation during a growfs operation.  When xfs_growfs_data_private() is
@@ -160,7 +81,7 @@ typedef struct fstrm_item
  * the cache that reference per-ag array elements that have since been
  * reallocated.
  */
-static int
+int
 xfs_filestream_peek_ag(
 	xfs_mount_t	*mp,
 	xfs_agnumber_t	agno)
@@ -200,23 +121,40 @@ xfs_filestream_put_ag(
 	xfs_perag_put(pag);
 }
 
+static void
+xfs_fstrm_free_func(
+	struct xfs_mru_cache_elem *mru)
+{
+	struct xfs_fstrm_item	*item =
+		container_of(mru, struct xfs_fstrm_item, mru);
+
+	xfs_filestream_put_ag(item->ip->i_mount, item->ag);
+
+	trace_xfs_filestream_free(item->ip, item->ag);
+
+	kmem_free(item);
+}
+
 /*
  * Scan the AGs starting at startag looking for an AG that isn't in use and has
  * at least minlen blocks free.
  */
 static int
-_xfs_filestream_pick_ag(
-	xfs_mount_t	*mp,
-	xfs_agnumber_t	startag,
-	xfs_agnumber_t	*agp,
-	int		flags,
-	xfs_extlen_t	minlen)
+xfs_filestream_pick_ag(
+	struct xfs_inode	*ip,
+	xfs_agnumber_t		startag,
+	xfs_agnumber_t		*agp,
+	int			flags,
+	xfs_extlen_t		minlen)
 {
-	int		streams, max_streams;
-	int		err, trylock, nscan;
-	xfs_extlen_t	longest, free, minfree, maxfree = 0;
-	xfs_agnumber_t	ag, max_ag = NULLAGNUMBER;
-	struct xfs_perag *pag;
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_fstrm_item	*item;
+	struct xfs_perag	*pag;
+	xfs_extlen_t		longest, free = 0, minfree, maxfree = 0;
+	xfs_agnumber_t		ag, max_ag = NULLAGNUMBER;
+	int			err, trylock, nscan;
+
+	ASSERT(S_ISDIR(ip->i_d.di_mode));
 
 	/* 2% of an AG's blocks must be free for it to be chosen. */
 	minfree = mp->m_sb.sb_agblocks / 50;
@@ -228,8 +166,9 @@ _xfs_filestream_pick_ag(
 	trylock = XFS_ALLOC_FLAG_TRYLOCK;
 
 	for (nscan = 0; 1; nscan++) {
+		trace_xfs_filestream_scan(ip, ag);
+
 		pag = xfs_perag_get(mp, ag);
-		TRACE_AG_SCAN(mp, ag, atomic_read(&pag->pagf_fstrms));
 
 		if (!pag->pagf_init) {
 			err = xfs_alloc_pagf_init(mp, NULL, ag, trylock);
@@ -246,7 +185,6 @@ _xfs_filestream_pick_ag(
 		/* Keep track of the AG with the most free blocks. */
 		if (pag->pagf_freeblks > maxfree) {
 			maxfree = pag->pagf_freeblks;
-			max_streams = atomic_read(&pag->pagf_fstrms);
 			max_ag = ag;
 		}
 
@@ -269,7 +207,6 @@ _xfs_filestream_pick_ag(
 
 			/* Break out, retaining the reference on the AG. */
 			free = pag->pagf_freeblks;
-			streams = atomic_read(&pag->pagf_fstrms);
 			xfs_perag_put(pag);
 			*agp = ag;
 			break;
@@ -305,317 +242,98 @@ next_ag:
 		 */
 		if (max_ag != NULLAGNUMBER) {
 			xfs_filestream_get_ag(mp, max_ag);
-			TRACE_AG_PICK1(mp, max_ag, maxfree);
-			streams = max_streams;
 			free = maxfree;
 			*agp = max_ag;
 			break;
 		}
 
 		/* take AG 0 if none matched */
-		TRACE_AG_PICK1(mp, max_ag, maxfree);
+		trace_xfs_filestream_pick(ip, *agp, free, nscan);
 		*agp = 0;
 		return 0;
 	}
 
-	TRACE_AG_PICK2(mp, startag, *agp, streams, free, nscan, flags);
-
-	return 0;
-}
+	trace_xfs_filestream_pick(ip, *agp, free, nscan);
 
-/*
- * Set the allocation group number for a file or a directory, updating inode
- * references and per-AG references as appropriate.
- */
-static int
-_xfs_filestream_update_ag(
-	xfs_inode_t	*ip,
-	xfs_inode_t	*pip,
-	xfs_agnumber_t	ag)
-{
-	int		err = 0;
-	xfs_mount_t	*mp;
-	xfs_mru_cache_t	*cache;
-	fstrm_item_t	*item;
-	xfs_agnumber_t	old_ag;
-	xfs_inode_t	*old_pip;
-
-	/*
-	 * Either ip is a regular file and pip is a directory, or ip is a
-	 * directory and pip is NULL.
-	 */
-	ASSERT(ip && ((S_ISREG(ip->i_d.di_mode) && pip &&
-	               S_ISDIR(pip->i_d.di_mode)) ||
-	              (S_ISDIR(ip->i_d.di_mode) && !pip)));
-
-	mp = ip->i_mount;
-	cache = mp->m_filestream;
-
-	item = xfs_mru_cache_lookup(cache, ip->i_ino);
-	if (item) {
-		ASSERT(item->ip == ip);
-		old_ag = item->ag;
-		item->ag = ag;
-		old_pip = item->pip;
-		item->pip = pip;
-		xfs_mru_cache_done(cache);
-
-		/*
-		 * If the AG has changed, drop the old ref and take a new one,
-		 * effectively transferring the reference from old to new AG.
-		 */
-		if (ag != old_ag) {
-			xfs_filestream_put_ag(mp, old_ag);
-			xfs_filestream_get_ag(mp, ag);
-		}
-
-		/*
-		 * If ip is a file and its pip has changed, drop the old ref and
-		 * take a new one.
-		 */
-		if (pip && pip != old_pip) {
-			IRELE(old_pip);
-			IHOLD(pip);
-		}
-
-		TRACE_UPDATE(mp, ip, old_ag, xfs_filestream_peek_ag(mp, old_ag),
-				ag, xfs_filestream_peek_ag(mp, ag));
+	if (*agp == NULLAGNUMBER)
 		return 0;
-	}
 
-	item = kmem_zone_zalloc(item_zone, KM_MAYFAIL);
+	err = -ENOMEM;
+	item = kmem_alloc(sizeof(*item), KM_MAYFAIL);
 	if (!item)
-		return ENOMEM;
+		goto out_put_ag;
 
-	item->ag = ag;
+	item->ag = *agp;
 	item->ip = ip;
-	item->pip = pip;
 
-	err = xfs_mru_cache_insert(cache, ip->i_ino, item);
+	err = xfs_mru_cache_insert(mp->m_filestream, ip->i_ino, &item->mru);
 	if (err) {
-		kmem_zone_free(item_zone, item);
-		return err;
+		if (err == -EEXIST)
+			err = 0;
+		goto out_free_item;
 	}
 
-	/* Take a reference on the AG. */
-	xfs_filestream_get_ag(mp, ag);
-
-	/*
-	 * Take a reference on the inode itself regardless of whether it's a
-	 * regular file or a directory.
-	 */
-	IHOLD(ip);
-
-	/*
-	 * In the case of a regular file, take a reference on the parent inode
-	 * as well to ensure it remains in-core.
-	 */
-	if (pip)
-		IHOLD(pip);
-
-	TRACE_UPDATE(mp, ip, ag, xfs_filestream_peek_ag(mp, ag),
-			ag, xfs_filestream_peek_ag(mp, ag));
-
 	return 0;
-}
-
-/* xfs_fstrm_free_func(): callback for freeing cached stream items. */
-STATIC void
-xfs_fstrm_free_func(
-	unsigned long	ino,
-	void		*data)
-{
-	fstrm_item_t	*item  = (fstrm_item_t *)data;
-	xfs_inode_t	*ip = item->ip;
-
-	ASSERT(ip->i_ino == ino);
-
-	xfs_iflags_clear(ip, XFS_IFILESTREAM);
-
-	/* Drop the reference taken on the AG when the item was added. */
-	xfs_filestream_put_ag(ip->i_mount, item->ag);
-
-	TRACE_FREE(ip->i_mount, ip, item->pip, item->ag,
-		xfs_filestream_peek_ag(ip->i_mount, item->ag));
-
-	/*
-	 * _xfs_filestream_update_ag() always takes a reference on the inode
-	 * itself, whether it's a file or a directory.  Release it here.
-	 * This can result in the inode being freed and so we must
-	 * not hold any inode locks when freeing filesstreams objects
-	 * otherwise we can deadlock here.
-	 */
-	IRELE(ip);
-
-	/*
-	 * In the case of a regular file, _xfs_filestream_update_ag() also
-	 * takes a ref on the parent inode to keep it in-core.  Release that
-	 * too.
-	 */
-	if (item->pip)
-		IRELE(item->pip);
-
-	/* Finally, free the memory allocated for the item. */
-	kmem_zone_free(item_zone, item);
-}
-
-/*
- * xfs_filestream_init() is called at xfs initialisation time to set up the
- * memory zone that will be used for filestream data structure allocation.
- */
-int
-xfs_filestream_init(void)
-{
-	item_zone = kmem_zone_init(sizeof(fstrm_item_t), "fstrm_item");
-	if (!item_zone)
-		return -ENOMEM;
-
-	return 0;
-}
-
-/*
- * xfs_filestream_uninit() is called at xfs termination time to destroy the
- * memory zone that was used for filestream data structure allocation.
- */
-void
-xfs_filestream_uninit(void)
-{
-	kmem_zone_destroy(item_zone);
-}
-
-/*
- * xfs_filestream_mount() is called when a file system is mounted with the
- * filestream option.  It is responsible for allocating the data structures
- * needed to track the new file system's file streams.
- */
-int
-xfs_filestream_mount(
-	xfs_mount_t	*mp)
-{
-	int		err;
-	unsigned int	lifetime, grp_count;
-
-	/*
-	 * The filestream timer tunable is currently fixed within the range of
-	 * one second to four minutes, with five seconds being the default.  The
-	 * group count is somewhat arbitrary, but it'd be nice to adhere to the
-	 * timer tunable to within about 10 percent.  This requires at least 10
-	 * groups.
-	 */
-	lifetime  = xfs_fstrm_centisecs * 10;
-	grp_count = 10;
-
-	err = xfs_mru_cache_create(&mp->m_filestream, lifetime, grp_count,
-	                     xfs_fstrm_free_func);
 
+out_free_item:
+	kmem_free(item);
+out_put_ag:
+	xfs_filestream_put_ag(mp, *agp);
 	return err;
 }
 
-/*
- * xfs_filestream_unmount() is called when a file system that was mounted with
- * the filestream option is unmounted.  It drains the data structures created
- * to track the file system's file streams and frees all the memory that was
- * allocated.
- */
-void
-xfs_filestream_unmount(
-	xfs_mount_t	*mp)
+static struct xfs_inode *
+xfs_filestream_get_parent(
+	struct xfs_inode	*ip)
 {
-	xfs_mru_cache_destroy(mp->m_filestream);
-}
+	struct inode		*inode = VFS_I(ip), *dir = NULL;
+	struct dentry		*dentry, *parent;
 
-/*
- * Return the AG of the filestream the file or directory belongs to, or
- * NULLAGNUMBER otherwise.
- */
-xfs_agnumber_t
-xfs_filestream_lookup_ag(
-	xfs_inode_t	*ip)
-{
-	xfs_mru_cache_t	*cache;
-	fstrm_item_t	*item;
-	xfs_agnumber_t	ag;
-	int		ref;
-
-	if (!S_ISREG(ip->i_d.di_mode) && !S_ISDIR(ip->i_d.di_mode)) {
-		ASSERT(0);
-		return NULLAGNUMBER;
-	}
+	dentry = d_find_alias(inode);
+	if (!dentry)
+		goto out;
 
-	cache = ip->i_mount->m_filestream;
-	item = xfs_mru_cache_lookup(cache, ip->i_ino);
-	if (!item) {
-		TRACE_LOOKUP(ip->i_mount, ip, NULL, NULLAGNUMBER, 0);
-		return NULLAGNUMBER;
-	}
+	parent = dget_parent(dentry);
+	if (!parent)
+		goto out_dput;
 
-	ASSERT(ip == item->ip);
-	ag = item->ag;
-	ref = xfs_filestream_peek_ag(ip->i_mount, ag);
-	xfs_mru_cache_done(cache);
+	dir = igrab(parent->d_inode);
+	dput(parent);
 
-	TRACE_LOOKUP(ip->i_mount, ip, item->pip, ag, ref);
-	return ag;
+out_dput:
+	dput(dentry);
+out:
+	return dir ? XFS_I(dir) : NULL;
 }
 
 /*
- * xfs_filestream_associate() should only be called to associate a regular file
- * with its parent directory.  Calling it with a child directory isn't
- * appropriate because filestreams don't apply to entire directory hierarchies.
- * Creating a file in a child directory of an existing filestream directory
- * starts a new filestream with its own allocation group association.
+ * Find the right allocation group for a file, either by finding an
+ * existing file stream or creating a new one.
  *
- * Returns < 0 on error, 0 if successful association occurred, > 0 if
- * we failed to get an association because of locking issues.
+ * Returns NULLAGNUMBER in case of an error.
  */
-int
-xfs_filestream_associate(
-	xfs_inode_t	*pip,
-	xfs_inode_t	*ip)
+xfs_agnumber_t
+xfs_filestream_lookup_ag(
+	struct xfs_inode	*ip)
 {
-	xfs_mount_t	*mp;
-	xfs_mru_cache_t	*cache;
-	fstrm_item_t	*item;
-	xfs_agnumber_t	ag, rotorstep, startag;
-	int		err = 0;
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_inode	*pip = NULL;
+	xfs_agnumber_t		startag, ag = NULLAGNUMBER;
+	struct xfs_mru_cache_elem *mru;
 
-	ASSERT(S_ISDIR(pip->i_d.di_mode));
 	ASSERT(S_ISREG(ip->i_d.di_mode));
-	if (!S_ISDIR(pip->i_d.di_mode) || !S_ISREG(ip->i_d.di_mode))
-		return -EINVAL;
 
-	mp = pip->i_mount;
-	cache = mp->m_filestream;
+	pip = xfs_filestream_get_parent(ip);
+	if (!pip)
+		goto out;
 
-	/*
-	 * We have a problem, Houston.
-	 *
-	 * Taking the iolock here violates inode locking order - we already
-	 * hold the ilock. Hence if we block getting this lock we may never
-	 * wake. Unfortunately, that means if we can't get the lock, we're
-	 * screwed in terms of getting a stream association - we can't spin
-	 * waiting for the lock because someone else is waiting on the lock we
-	 * hold and we cannot drop that as we are in a transaction here.
-	 *
-	 * Lucky for us, this inversion is not a problem because it's a
-	 * directory inode that we are trying to lock here.
-	 *
-	 * So, if we can't get the iolock without sleeping then just give up
-	 */
-	if (!xfs_ilock_nowait(pip, XFS_IOLOCK_EXCL))
-		return 1;
-
-	/* If the parent directory is already in the cache, use its AG. */
-	item = xfs_mru_cache_lookup(cache, pip->i_ino);
-	if (item) {
-		ASSERT(item->ip == pip);
-		ag = item->ag;
-		xfs_mru_cache_done(cache);
-
-		TRACE_LOOKUP(mp, pip, pip, ag, xfs_filestream_peek_ag(mp, ag));
-		err = _xfs_filestream_update_ag(ip, pip, ag);
+	mru = xfs_mru_cache_lookup(mp->m_filestream, pip->i_ino);
+	if (mru) {
+		ag = container_of(mru, struct xfs_fstrm_item, mru)->ag;
+		xfs_mru_cache_done(mp->m_filestream);
 
-		goto exit;
+		trace_xfs_filestream_lookup(ip, ag);
+		goto out;
 	}
 
 	/*
@@ -623,202 +341,94 @@ xfs_filestream_associate(
 	 * use the directory inode's AG.
 	 */
 	if (mp->m_flags & XFS_MOUNT_32BITINODES) {
-		rotorstep = xfs_rotorstep;
+		xfs_agnumber_t	 rotorstep = xfs_rotorstep;
 		startag = (mp->m_agfrotor / rotorstep) % mp->m_sb.sb_agcount;
 		mp->m_agfrotor = (mp->m_agfrotor + 1) %
 		                 (mp->m_sb.sb_agcount * rotorstep);
 	} else
 		startag = XFS_INO_TO_AGNO(mp, pip->i_ino);
 
-	/* Pick a new AG for the parent inode starting at startag. */
-	err = _xfs_filestream_pick_ag(mp, startag, &ag, 0, 0);
-	if (err || ag == NULLAGNUMBER)
-		goto exit_did_pick;
-
-	/* Associate the parent inode with the AG. */
-	err = _xfs_filestream_update_ag(pip, NULL, ag);
-	if (err)
-		goto exit_did_pick;
-
-	/* Associate the file inode with the AG. */
-	err = _xfs_filestream_update_ag(ip, pip, ag);
-	if (err)
-		goto exit_did_pick;
-
-	TRACE_ASSOCIATE(mp, ip, pip, ag, xfs_filestream_peek_ag(mp, ag));
-
-exit_did_pick:
-	/*
-	 * If _xfs_filestream_pick_ag() returned a valid AG, remove the
-	 * reference it took on it, since the file and directory will have taken
-	 * their own now if they were successfully cached.
-	 */
-	if (ag != NULLAGNUMBER)
-		xfs_filestream_put_ag(mp, ag);
-
-exit:
-	xfs_iunlock(pip, XFS_IOLOCK_EXCL);
-	return -err;
+	if (xfs_filestream_pick_ag(pip, startag, &ag, 0, 0))
+		ag = NULLAGNUMBER;
+out:
+	IRELE(pip);
+	return ag;
 }
 
 /*
- * Pick a new allocation group for the current file and its file stream.  This
- * function is called by xfs_bmap_filestreams() with the mount point's per-ag
- * lock held.
+ * Pick a new allocation group for the current file and its file stream.
+ *
+ * This is called when the allocator can't find a suitable extent in the
+ * current AG, and we have to move the stream into a new AG with more space.
  */
 int
 xfs_filestream_new_ag(
 	struct xfs_bmalloca	*ap,
 	xfs_agnumber_t		*agp)
 {
-	int		flags, err;
-	xfs_inode_t	*ip, *pip = NULL;
-	xfs_mount_t	*mp;
-	xfs_mru_cache_t	*cache;
-	xfs_extlen_t	minlen;
-	fstrm_item_t	*dir, *file;
-	xfs_agnumber_t	ag = NULLAGNUMBER;
-
-	ip = ap->ip;
-	mp = ip->i_mount;
-	cache = mp->m_filestream;
-	minlen = ap->length;
-	*agp = NULLAGNUMBER;
+	struct xfs_inode	*ip = ap->ip, *pip;
+	struct xfs_mount	*mp = ip->i_mount;
+	xfs_extlen_t		minlen = ap->length;
+	xfs_agnumber_t		startag = 0;
+	int			flags, err = 0;
+	struct xfs_mru_cache_elem *mru;
 
-	/*
-	 * Look for the file in the cache, removing it if it's found.  Doing
-	 * this allows it to be held across the dir lookup that follows.
-	 */
-	file = xfs_mru_cache_remove(cache, ip->i_ino);
-	if (file) {
-		ASSERT(ip == file->ip);
-
-		/* Save the file's parent inode and old AG number for later. */
-		pip = file->pip;
-		ag = file->ag;
-
-		/* Look for the file's directory in the cache. */
-		dir = xfs_mru_cache_lookup(cache, pip->i_ino);
-		if (dir) {
-			ASSERT(pip == dir->ip);
-
-			/*
-			 * If the directory has already moved on to a new AG,
-			 * use that AG as the new AG for the file. Don't
-			 * forget to twiddle the AG refcounts to match the
-			 * movement.
-			 */
-			if (dir->ag != file->ag) {
-				xfs_filestream_put_ag(mp, file->ag);
-				xfs_filestream_get_ag(mp, dir->ag);
-				*agp = file->ag = dir->ag;
-			}
-
-			xfs_mru_cache_done(cache);
-		}
+	*agp = NULLAGNUMBER;
 
-		/*
-		 * Put the file back in the cache.  If this fails, the free
-		 * function needs to be called to tidy up in the same way as if
-		 * the item had simply expired from the cache.
-		 */
-		err = xfs_mru_cache_insert(cache, ip->i_ino, file);
-		if (err) {
-			xfs_fstrm_free_func(ip->i_ino, file);
-			return err;
-		}
+	pip = xfs_filestream_get_parent(ip);
+	if (!pip)
+		goto exit;
 
-		/*
-		 * If the file's AG was moved to the directory's new AG, there's
-		 * nothing more to be done.
-		 */
-		if (*agp != NULLAGNUMBER) {
-			TRACE_MOVEAG(mp, ip, pip,
-					ag, xfs_filestream_peek_ag(mp, ag),
-					*agp, xfs_filestream_peek_ag(mp, *agp));
-			return 0;
-		}
+	mru = xfs_mru_cache_remove(mp->m_filestream, pip->i_ino);
+	if (mru) {
+		struct xfs_fstrm_item *item =
+			container_of(mru, struct xfs_fstrm_item, mru);
+		startag = (item->ag + 1) % mp->m_sb.sb_agcount;
 	}
 
-	/*
-	 * If the file's parent directory is known, take its iolock in exclusive
-	 * mode to prevent two sibling files from racing each other to migrate
-	 * themselves and their parent to different AGs.
-	 *
-	 * Note that we lock the parent directory iolock inside the child
-	 * iolock here.  That's fine as we never hold both parent and child
-	 * iolock in any other place.  This is different from the ilock,
-	 * which requires locking of the child after the parent for namespace
-	 * operations.
-	 */
-	if (pip)
-		xfs_ilock(pip, XFS_IOLOCK_EXCL | XFS_IOLOCK_PARENT);
-
-	/*
-	 * A new AG needs to be found for the file.  If the file's parent
-	 * directory is also known, it will be moved to the new AG as well to
-	 * ensure that files created inside it in future use the new AG.
-	 */
-	ag = (ag == NULLAGNUMBER) ? 0 : (ag + 1) % mp->m_sb.sb_agcount;
 	flags = (ap->userdata ? XFS_PICK_USERDATA : 0) |
 	        (ap->flist->xbf_low ? XFS_PICK_LOWSPACE : 0);
 
-	err = _xfs_filestream_pick_ag(mp, ag, agp, flags, minlen);
-	if (err || *agp == NULLAGNUMBER)
-		goto exit;
+	err = xfs_filestream_pick_ag(pip, startag, agp, flags, minlen);
 
 	/*
-	 * If the file wasn't found in the file cache, then its parent directory
-	 * inode isn't known.  For this to have happened, the file must either
-	 * be pre-existing, or it was created long enough ago that its cache
-	 * entry has expired.  This isn't the sort of usage that the filestreams
-	 * allocator is trying to optimise, so there's no point trying to track
-	 * its new AG somehow in the filestream data structures.
+	 * Only free the item here so we skip over the old AG earlier.
 	 */
-	if (!pip) {
-		TRACE_ORPHAN(mp, ip, *agp);
-		goto exit;
-	}
-
-	/* Associate the parent inode with the AG. */
-	err = _xfs_filestream_update_ag(pip, NULL, *agp);
-	if (err)
-		goto exit;
-
-	/* Associate the file inode with the AG. */
-	err = _xfs_filestream_update_ag(ip, pip, *agp);
-	if (err)
-		goto exit;
-
-	TRACE_MOVEAG(mp, ip, pip, NULLAGNUMBER, 0,
-			*agp, xfs_filestream_peek_ag(mp, *agp));
+	if (mru)
+		xfs_fstrm_free_func(mru);
 
+	IRELE(pip);
 exit:
-	/*
-	 * If _xfs_filestream_pick_ag() returned a valid AG, remove the
-	 * reference it took on it, since the file and directory will have taken
-	 * their own now if they were successfully cached.
-	 */
-	if (*agp != NULLAGNUMBER)
-		xfs_filestream_put_ag(mp, *agp);
-	else
+	if (*agp == NULLAGNUMBER)
 		*agp = 0;
-
-	if (pip)
-		xfs_iunlock(pip, XFS_IOLOCK_EXCL);
-
 	return err;
 }
 
-/*
- * Remove an association between an inode and a filestream object.
- * Typically this is done on last close of an unlinked file.
- */
 void
 xfs_filestream_deassociate(
-	xfs_inode_t	*ip)
+	struct xfs_inode	*ip)
 {
-	xfs_mru_cache_t	*cache = ip->i_mount->m_filestream;
+	xfs_mru_cache_delete(ip->i_mount->m_filestream, ip->i_ino);
+}
+
+int
+xfs_filestream_mount(
+	xfs_mount_t	*mp)
+{
+	/*
+	 * The filestream timer tunable is currently fixed within the range of
+	 * one second to four minutes, with five seconds being the default.  The
+	 * group count is somewhat arbitrary, but it'd be nice to adhere to the
+	 * timer tunable to within about 10 percent.  This requires at least 10
+	 * groups.
+	 */
+	return xfs_mru_cache_create(&mp->m_filestream, xfs_fstrm_centisecs * 10,
+				    10, xfs_fstrm_free_func);
+}
 
-	xfs_mru_cache_delete(cache, ip->i_ino);
+void
+xfs_filestream_unmount(
+	xfs_mount_t	*mp)
+{
+	xfs_mru_cache_destroy(mp->m_filestream);
 }
diff --git a/fs/xfs/xfs_filestream.h b/fs/xfs/xfs_filestream.h
index 6d61dbee8564..2ef43406e53b 100644
--- a/fs/xfs/xfs_filestream.h
+++ b/fs/xfs/xfs_filestream.h
@@ -20,50 +20,20 @@
 
 struct xfs_mount;
 struct xfs_inode;
-struct xfs_perag;
 struct xfs_bmalloca;
 
-#ifdef XFS_FILESTREAMS_TRACE
-#define XFS_FSTRM_KTRACE_INFO		1
-#define XFS_FSTRM_KTRACE_AGSCAN		2
-#define XFS_FSTRM_KTRACE_AGPICK1	3
-#define XFS_FSTRM_KTRACE_AGPICK2	4
-#define XFS_FSTRM_KTRACE_UPDATE		5
-#define XFS_FSTRM_KTRACE_FREE		6
-#define	XFS_FSTRM_KTRACE_ITEM_LOOKUP	7
-#define	XFS_FSTRM_KTRACE_ASSOCIATE	8
-#define	XFS_FSTRM_KTRACE_MOVEAG		9
-#define	XFS_FSTRM_KTRACE_ORPHAN		10
-
-#define XFS_FSTRM_KTRACE_SIZE	16384
-extern ktrace_t *xfs_filestreams_trace_buf;
-
-#endif
-
-/* allocation selection flags */
-typedef enum xfs_fstrm_alloc {
-	XFS_PICK_USERDATA = 1,
-	XFS_PICK_LOWSPACE = 2,
-} xfs_fstrm_alloc_t;
-
-/* prototypes for filestream.c */
-int xfs_filestream_init(void);
-void xfs_filestream_uninit(void);
 int xfs_filestream_mount(struct xfs_mount *mp);
 void xfs_filestream_unmount(struct xfs_mount *mp);
-xfs_agnumber_t xfs_filestream_lookup_ag(struct xfs_inode *ip);
-int xfs_filestream_associate(struct xfs_inode *dip, struct xfs_inode *ip);
 void xfs_filestream_deassociate(struct xfs_inode *ip);
+xfs_agnumber_t xfs_filestream_lookup_ag(struct xfs_inode *ip);
 int xfs_filestream_new_ag(struct xfs_bmalloca *ap, xfs_agnumber_t *agp);
+int xfs_filestream_peek_ag(struct xfs_mount *mp, xfs_agnumber_t agno);
 
-
-/* filestreams for the inode? */
 static inline int
 xfs_inode_is_filestream(
 	struct xfs_inode	*ip)
 {
 	return (ip->i_mount->m_flags & XFS_MOUNT_FILESTREAMS) ||
-		xfs_iflags_test(ip, XFS_IFILESTREAM) ||
 		(ip->i_d.di_flags & XFS_DIFLAG_FILESTREAM);
 }
 
diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h
index c5fc116dfaa3..18dc721ca19f 100644
--- a/fs/xfs/xfs_fs.h
+++ b/fs/xfs/xfs_fs.h
@@ -238,6 +238,7 @@ typedef struct xfs_fsop_resblks {
 #define XFS_FSOP_GEOM_FLAGS_LAZYSB	0x4000	/* lazy superblock counters */
 #define XFS_FSOP_GEOM_FLAGS_V5SB	0x8000	/* version 5 superblock */
 #define XFS_FSOP_GEOM_FLAGS_FTYPE	0x10000	/* inode directory types */
+#define XFS_FSOP_GEOM_FLAGS_FINOBT	0x20000	/* free inode btree */
 
 /*
  * Minimum and maximum sizes need for growth checks.
@@ -254,8 +255,8 @@ typedef struct xfs_fsop_resblks {
 	((2 * 1024 * 1024 * 1024ULL) - XFS_MIN_LOG_BYTES)
 
 /* Used for sanity checks on superblock */
-#define XFS_MAX_DBLOCKS(s) ((xfs_drfsbno_t)(s)->sb_agcount * (s)->sb_agblocks)
-#define XFS_MIN_DBLOCKS(s) ((xfs_drfsbno_t)((s)->sb_agcount - 1) *	\
+#define XFS_MAX_DBLOCKS(s) ((xfs_rfsblock_t)(s)->sb_agcount * (s)->sb_agblocks)
+#define XFS_MIN_DBLOCKS(s) ((xfs_rfsblock_t)((s)->sb_agcount - 1) *	\
 			 (s)->sb_agblocks + XFS_MIN_AG_BLOCKS)
 
 /*
@@ -374,6 +375,9 @@ struct xfs_fs_eofblocks {
 #define XFS_EOF_FLAGS_GID		(1 << 2) /* filter by gid */
 #define XFS_EOF_FLAGS_PRID		(1 << 3) /* filter by project id */
 #define XFS_EOF_FLAGS_MINFILESIZE	(1 << 4) /* filter by min file size */
+#define XFS_EOF_FLAGS_UNION		(1 << 5) /* union filter algorithm;
+						  * kernel only, not included in
+						  * valid mask */
 #define XFS_EOF_FLAGS_VALID	\
 	(XFS_EOF_FLAGS_SYNC |	\
 	 XFS_EOF_FLAGS_UID |	\
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index 02fb943cbf22..f91de1ef05e1 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -24,6 +24,8 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_da_format.h"
+#include "xfs_da_btree.h"
 #include "xfs_inode.h"
 #include "xfs_trans.h"
 #include "xfs_inode_item.h"
@@ -74,23 +76,18 @@ xfs_fs_geometry(
 	}
 	if (new_version >= 3) {
 		geo->version = XFS_FSOP_GEOM_VERSION;
-		geo->flags =
+		geo->flags = XFS_FSOP_GEOM_FLAGS_NLINK |
+			     XFS_FSOP_GEOM_FLAGS_DIRV2 |
 			(xfs_sb_version_hasattr(&mp->m_sb) ?
 				XFS_FSOP_GEOM_FLAGS_ATTR : 0) |
-			(xfs_sb_version_hasnlink(&mp->m_sb) ?
-				XFS_FSOP_GEOM_FLAGS_NLINK : 0) |
 			(xfs_sb_version_hasquota(&mp->m_sb) ?
 				XFS_FSOP_GEOM_FLAGS_QUOTA : 0) |
 			(xfs_sb_version_hasalign(&mp->m_sb) ?
 				XFS_FSOP_GEOM_FLAGS_IALIGN : 0) |
 			(xfs_sb_version_hasdalign(&mp->m_sb) ?
 				XFS_FSOP_GEOM_FLAGS_DALIGN : 0) |
-			(xfs_sb_version_hasshared(&mp->m_sb) ?
-				XFS_FSOP_GEOM_FLAGS_SHARED : 0) |
 			(xfs_sb_version_hasextflgbit(&mp->m_sb) ?
 				XFS_FSOP_GEOM_FLAGS_EXTFLG : 0) |
-			(xfs_sb_version_hasdirv2(&mp->m_sb) ?
-				XFS_FSOP_GEOM_FLAGS_DIRV2 : 0) |
 			(xfs_sb_version_hassector(&mp->m_sb) ?
 				XFS_FSOP_GEOM_FLAGS_SECTOR : 0) |
 			(xfs_sb_version_hasasciici(&mp->m_sb) ?
@@ -104,11 +101,13 @@ xfs_fs_geometry(
 			(xfs_sb_version_hascrc(&mp->m_sb) ?
 				XFS_FSOP_GEOM_FLAGS_V5SB : 0) |
 			(xfs_sb_version_hasftype(&mp->m_sb) ?
-				XFS_FSOP_GEOM_FLAGS_FTYPE : 0);
+				XFS_FSOP_GEOM_FLAGS_FTYPE : 0) |
+			(xfs_sb_version_hasfinobt(&mp->m_sb) ?
+				XFS_FSOP_GEOM_FLAGS_FINOBT : 0);
 		geo->logsectsize = xfs_sb_version_hassector(&mp->m_sb) ?
 				mp->m_sb.sb_logsectsize : BBSIZE;
 		geo->rtsectsize = mp->m_sb.sb_blocksize;
-		geo->dirblocksize = mp->m_dirblksize;
+		geo->dirblocksize = mp->m_dir_geo->blksize;
 	}
 	if (new_version >= 4) {
 		geo->flags |=
@@ -169,7 +168,7 @@ xfs_growfs_data_private(
 	nb = in->newblocks;
 	pct = in->imaxpct;
 	if (nb < mp->m_sb.sb_dblocks || pct < 0 || pct > 100)
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	if ((error = xfs_sb_validate_fsb_count(&mp->m_sb, nb)))
 		return error;
 	dpct = pct - mp->m_sb.sb_imax_pct;
@@ -177,7 +176,7 @@ xfs_growfs_data_private(
 				XFS_FSB_TO_BB(mp, nb) - XFS_FSS_TO_BB(mp, 1),
 				XFS_FSS_TO_BB(mp, 1), 0, NULL);
 	if (!bp)
-		return EIO;
+		return -EIO;
 	if (bp->b_error) {
 		error = bp->b_error;
 		xfs_buf_relse(bp);
@@ -192,7 +191,7 @@ xfs_growfs_data_private(
 		nagcount--;
 		nb = (xfs_rfsblock_t)nagcount * mp->m_sb.sb_agblocks;
 		if (nb < mp->m_sb.sb_dblocks)
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 	}
 	new = nb - mp->m_sb.sb_dblocks;
 	oagcount = mp->m_sb.sb_agcount;
@@ -230,7 +229,7 @@ xfs_growfs_data_private(
 				XFS_FSS_TO_BB(mp, 1), 0,
 				&xfs_agf_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -271,7 +270,7 @@ xfs_growfs_data_private(
 				XFS_FSS_TO_BB(mp, 1), 0,
 				&xfs_agfl_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -299,7 +298,7 @@ xfs_growfs_data_private(
 				XFS_FSS_TO_BB(mp, 1), 0,
 				&xfs_agi_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -316,6 +315,10 @@ xfs_growfs_data_private(
 		agi->agi_dirino = cpu_to_be32(NULLAGINO);
 		if (xfs_sb_version_hascrc(&mp->m_sb))
 			uuid_copy(&agi->agi_uuid, &mp->m_sb.sb_uuid);
+		if (xfs_sb_version_hasfinobt(&mp->m_sb)) {
+			agi->agi_free_root = cpu_to_be32(XFS_FIBT_BLOCK(mp));
+			agi->agi_free_level = cpu_to_be32(1);
+		}
 		for (bucket = 0; bucket < XFS_AGI_UNLINKED_BUCKETS; bucket++)
 			agi->agi_unlinked[bucket] = cpu_to_be32(NULLAGINO);
 
@@ -333,7 +336,7 @@ xfs_growfs_data_private(
 				&xfs_allocbt_buf_ops);
 
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -362,7 +365,7 @@ xfs_growfs_data_private(
 				BTOBB(mp->m_sb.sb_blocksize), 0,
 				&xfs_allocbt_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -392,7 +395,7 @@ xfs_growfs_data_private(
 				BTOBB(mp->m_sb.sb_blocksize), 0,
 				&xfs_inobt_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -407,6 +410,34 @@ xfs_growfs_data_private(
 		xfs_buf_relse(bp);
 		if (error)
 			goto error0;
+
+		/*
+		 * FINO btree root block
+		 */
+		if (xfs_sb_version_hasfinobt(&mp->m_sb)) {
+			bp = xfs_growfs_get_hdr_buf(mp,
+				XFS_AGB_TO_DADDR(mp, agno, XFS_FIBT_BLOCK(mp)),
+				BTOBB(mp->m_sb.sb_blocksize), 0,
+				&xfs_inobt_buf_ops);
+			if (!bp) {
+				error = -ENOMEM;
+				goto error0;
+			}
+
+			if (xfs_sb_version_hascrc(&mp->m_sb))
+				xfs_btree_init_block(mp, bp, XFS_FIBT_CRC_MAGIC,
+						     0, 0, agno,
+						     XFS_BTREE_CRC_BLOCKS);
+			else
+				xfs_btree_init_block(mp, bp, XFS_FIBT_MAGIC, 0,
+						     0, agno, 0);
+
+			error = xfs_bwrite(bp);
+			xfs_buf_relse(bp);
+			if (error)
+				goto error0;
+		}
+
 	}
 	xfs_trans_agblocks_delta(tp, nfree);
 	/*
@@ -500,7 +531,7 @@ xfs_growfs_data_private(
 				bp->b_ops = &xfs_sb_buf_ops;
 				xfs_buf_zero(bp, 0, BBTOB(bp->b_length));
 			} else
-				error = ENOMEM;
+				error = -ENOMEM;
 		}
 
 		/*
@@ -545,17 +576,17 @@ xfs_growfs_log_private(
 
 	nb = in->newblocks;
 	if (nb < XFS_MIN_LOG_BLOCKS || nb < XFS_B_TO_FSB(mp, XFS_MIN_LOG_BYTES))
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	if (nb == mp->m_sb.sb_logblocks &&
 	    in->isint == (mp->m_sb.sb_logstart != 0))
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	/*
 	 * Moving the log is hard, need new interfaces to sync
 	 * the log first, hold off all activity while moving it.
 	 * Can have shorter or longer log in the same space,
 	 * or transform internal to external log or vice versa.
 	 */
-	return XFS_ERROR(ENOSYS);
+	return -ENOSYS;
 }
 
 /*
@@ -573,9 +604,9 @@ xfs_growfs_data(
 	int error;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return XFS_ERROR(EPERM);
+		return -EPERM;
 	if (!mutex_trylock(&mp->m_growlock))
-		return XFS_ERROR(EWOULDBLOCK);
+		return -EWOULDBLOCK;
 	error = xfs_growfs_data_private(mp, in);
 	mutex_unlock(&mp->m_growlock);
 	return error;
@@ -589,9 +620,9 @@ xfs_growfs_log(
 	int error;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return XFS_ERROR(EPERM);
+		return -EPERM;
 	if (!mutex_trylock(&mp->m_growlock))
-		return XFS_ERROR(EWOULDBLOCK);
+		return -EWOULDBLOCK;
 	error = xfs_growfs_log_private(mp, in);
 	mutex_unlock(&mp->m_growlock);
 	return error;
@@ -643,7 +674,7 @@ xfs_reserve_blocks(
 	/* If inval is null, report current values and return */
 	if (inval == (__uint64_t *)NULL) {
 		if (!outval)
-			return EINVAL;
+			return -EINVAL;
 		outval->resblks = mp->m_resblks;
 		outval->resblks_avail = mp->m_resblks_avail;
 		return 0;
@@ -726,7 +757,7 @@ out:
 		int error;
 		error = xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS,
 						 fdblks_delta, 0);
-		if (error == ENOSPC)
+		if (error == -ENOSPC)
 			goto retry;
 	}
 	return 0;
@@ -787,7 +818,7 @@ xfs_fs_goingdown(
 				SHUTDOWN_FORCE_UMOUNT | SHUTDOWN_LOG_IO_ERROR);
 		break;
 	default:
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	return 0;
diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c
index 98d35244eecc..981b2cf51985 100644
--- a/fs/xfs/xfs_icache.c
+++ b/fs/xfs/xfs_icache.c
@@ -33,6 +33,9 @@
 #include "xfs_trace.h"
 #include "xfs_icache.h"
 #include "xfs_bmap_util.h"
+#include "xfs_quota.h"
+#include "xfs_dquot_item.h"
+#include "xfs_dquot.h"
 
 #include <linux/kthread.h>
 #include <linux/freezer.h>
@@ -158,7 +161,7 @@ xfs_iget_cache_hit(
 	if (ip->i_ino != ino) {
 		trace_xfs_iget_skip(ip);
 		XFS_STATS_INC(xs_ig_frecycle);
-		error = EAGAIN;
+		error = -EAGAIN;
 		goto out_error;
 	}
 
@@ -176,7 +179,7 @@ xfs_iget_cache_hit(
 	if (ip->i_flags & (XFS_INEW|XFS_IRECLAIM)) {
 		trace_xfs_iget_skip(ip);
 		XFS_STATS_INC(xs_ig_frecycle);
-		error = EAGAIN;
+		error = -EAGAIN;
 		goto out_error;
 	}
 
@@ -184,7 +187,7 @@ 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)) {
-		error = ENOENT;
+		error = -ENOENT;
 		goto out_error;
 	}
 
@@ -206,7 +209,7 @@ xfs_iget_cache_hit(
 		spin_unlock(&ip->i_flags_lock);
 		rcu_read_unlock();
 
-		error = -inode_init_always(mp->m_super, inode);
+		error = inode_init_always(mp->m_super, inode);
 		if (error) {
 			/*
 			 * Re-initializing the inode failed, and we are in deep
@@ -243,7 +246,7 @@ xfs_iget_cache_hit(
 		/* If the VFS inode is being torn down, pause and try again. */
 		if (!igrab(inode)) {
 			trace_xfs_iget_skip(ip);
-			error = EAGAIN;
+			error = -EAGAIN;
 			goto out_error;
 		}
 
@@ -285,7 +288,7 @@ xfs_iget_cache_miss(
 
 	ip = xfs_inode_alloc(mp, ino);
 	if (!ip)
-		return ENOMEM;
+		return -ENOMEM;
 
 	error = xfs_iread(mp, tp, ip, flags);
 	if (error)
@@ -294,7 +297,7 @@ xfs_iget_cache_miss(
 	trace_xfs_iget_miss(ip);
 
 	if ((ip->i_d.di_mode == 0) && !(flags & XFS_IGET_CREATE)) {
-		error = ENOENT;
+		error = -ENOENT;
 		goto out_destroy;
 	}
 
@@ -305,7 +308,7 @@ xfs_iget_cache_miss(
 	 * recurse into the file system.
 	 */
 	if (radix_tree_preload(GFP_NOFS)) {
-		error = EAGAIN;
+		error = -EAGAIN;
 		goto out_destroy;
 	}
 
@@ -341,7 +344,7 @@ xfs_iget_cache_miss(
 	if (unlikely(error)) {
 		WARN_ON(error != -EEXIST);
 		XFS_STATS_INC(xs_ig_dup);
-		error = EAGAIN;
+		error = -EAGAIN;
 		goto out_preload_end;
 	}
 	spin_unlock(&pag->pag_ici_lock);
@@ -408,7 +411,7 @@ xfs_iget(
 
 	/* reject inode numbers outside existing AGs */
 	if (!ino || XFS_INO_TO_AGNO(mp, ino) >= mp->m_sb.sb_agcount)
-		return EINVAL;
+		return -EINVAL;
 
 	/* get the perag structure and ensure that it's inode capable */
 	pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ino));
@@ -445,7 +448,7 @@ again:
 	return 0;
 
 out_error_or_again:
-	if (error == EAGAIN) {
+	if (error == -EAGAIN) {
 		delay(1);
 		goto again;
 	}
@@ -489,26 +492,25 @@ xfs_inode_ag_walk_grab(
 
 	/* nothing to sync during shutdown */
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 
 	/* If we can't grab the inode, it must on it's way to reclaim. */
 	if (!igrab(inode))
-		return ENOENT;
+		return -ENOENT;
 
 	/* inode is valid */
 	return 0;
 
 out_unlock_noent:
 	spin_unlock(&ip->i_flags_lock);
-	return ENOENT;
+	return -ENOENT;
 }
 
 STATIC int
 xfs_inode_ag_walk(
 	struct xfs_mount	*mp,
 	struct xfs_perag	*pag,
-	int			(*execute)(struct xfs_inode *ip,
-					   struct xfs_perag *pag, int flags,
+	int			(*execute)(struct xfs_inode *ip, int flags,
 					   void *args),
 	int			flags,
 	void			*args,
@@ -582,18 +584,18 @@ restart:
 		for (i = 0; i < nr_found; i++) {
 			if (!batch[i])
 				continue;
-			error = execute(batch[i], pag, flags, args);
+			error = execute(batch[i], flags, args);
 			IRELE(batch[i]);
-			if (error == EAGAIN) {
+			if (error == -EAGAIN) {
 				skipped++;
 				continue;
 			}
-			if (error && last_error != EFSCORRUPTED)
+			if (error && last_error != -EFSCORRUPTED)
 				last_error = error;
 		}
 
 		/* bail out if the filesystem is corrupted.  */
-		if (error == EFSCORRUPTED)
+		if (error == -EFSCORRUPTED)
 			break;
 
 		cond_resched();
@@ -636,8 +638,7 @@ xfs_eofblocks_worker(
 int
 xfs_inode_ag_iterator(
 	struct xfs_mount	*mp,
-	int			(*execute)(struct xfs_inode *ip,
-					   struct xfs_perag *pag, int flags,
+	int			(*execute)(struct xfs_inode *ip, int flags,
 					   void *args),
 	int			flags,
 	void			*args)
@@ -654,18 +655,17 @@ xfs_inode_ag_iterator(
 		xfs_perag_put(pag);
 		if (error) {
 			last_error = error;
-			if (error == EFSCORRUPTED)
+			if (error == -EFSCORRUPTED)
 				break;
 		}
 	}
-	return XFS_ERROR(last_error);
+	return last_error;
 }
 
 int
 xfs_inode_ag_iterator_tag(
 	struct xfs_mount	*mp,
-	int			(*execute)(struct xfs_inode *ip,
-					   struct xfs_perag *pag, int flags,
+	int			(*execute)(struct xfs_inode *ip, int flags,
 					   void *args),
 	int			flags,
 	void			*args,
@@ -683,11 +683,11 @@ xfs_inode_ag_iterator_tag(
 		xfs_perag_put(pag);
 		if (error) {
 			last_error = error;
-			if (error == EFSCORRUPTED)
+			if (error == -EFSCORRUPTED)
 				break;
 		}
 	}
-	return XFS_ERROR(last_error);
+	return last_error;
 }
 
 /*
@@ -947,7 +947,7 @@ restart:
 	 * see the stale flag set on the inode.
 	 */
 	error = xfs_iflush(ip, &bp);
-	if (error == EAGAIN) {
+	if (error == -EAGAIN) {
 		xfs_iunlock(ip, XFS_ILOCK_EXCL);
 		/* backoff longer than in xfs_ifree_cluster */
 		delay(2);
@@ -1000,7 +1000,7 @@ out:
 	xfs_iflags_clear(ip, XFS_IRECLAIM);
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 	/*
-	 * We could return EAGAIN here to make reclaim rescan the inode tree in
+	 * We could return -EAGAIN here to make reclaim rescan the inode tree in
 	 * a short while. However, this just burns CPU time scanning the tree
 	 * waiting for IO to complete and the reclaim work never goes back to
 	 * the idle state. Instead, return 0 to let the next scheduled
@@ -1103,7 +1103,7 @@ restart:
 				if (!batch[i])
 					continue;
 				error = xfs_reclaim_inode(batch[i], pag, flags);
-				if (error && last_error != EFSCORRUPTED)
+				if (error && last_error != -EFSCORRUPTED)
 					last_error = error;
 			}
 
@@ -1132,7 +1132,7 @@ restart:
 		trylock = 0;
 		goto restart;
 	}
-	return XFS_ERROR(last_error);
+	return last_error;
 }
 
 int
@@ -1206,15 +1206,42 @@ xfs_inode_match_id(
 	return 1;
 }
 
+/*
+ * A union-based inode filtering algorithm. Process the inode if any of the
+ * criteria match. This is for global/internal scans only.
+ */
+STATIC int
+xfs_inode_match_id_union(
+	struct xfs_inode	*ip,
+	struct xfs_eofblocks	*eofb)
+{
+	if ((eofb->eof_flags & XFS_EOF_FLAGS_UID) &&
+	    uid_eq(VFS_I(ip)->i_uid, eofb->eof_uid))
+		return 1;
+
+	if ((eofb->eof_flags & XFS_EOF_FLAGS_GID) &&
+	    gid_eq(VFS_I(ip)->i_gid, eofb->eof_gid))
+		return 1;
+
+	if ((eofb->eof_flags & XFS_EOF_FLAGS_PRID) &&
+	    xfs_get_projid(ip) == eofb->eof_prid)
+		return 1;
+
+	return 0;
+}
+
 STATIC int
 xfs_inode_free_eofblocks(
 	struct xfs_inode	*ip,
-	struct xfs_perag	*pag,
 	int			flags,
 	void			*args)
 {
 	int ret;
 	struct xfs_eofblocks *eofb = args;
+	bool need_iolock = true;
+	int match;
+
+	ASSERT(!eofb || (eofb && eofb->eof_scan_owner != 0));
 
 	if (!xfs_can_free_eofblocks(ip, false)) {
 		/* inode could be preallocated or append-only */
@@ -1232,19 +1259,31 @@ xfs_inode_free_eofblocks(
 		return 0;
 
 	if (eofb) {
-		if (!xfs_inode_match_id(ip, eofb))
+		if (eofb->eof_flags & XFS_EOF_FLAGS_UNION)
+			match = xfs_inode_match_id_union(ip, eofb);
+		else
+			match = xfs_inode_match_id(ip, eofb);
+		if (!match)
 			return 0;
 
 		/* skip the inode if the file size is too small */
 		if (eofb->eof_flags & XFS_EOF_FLAGS_MINFILESIZE &&
 		    XFS_ISIZE(ip) < eofb->eof_min_file_size)
 			return 0;
+
+		/*
+		 * A scan owner implies we already hold the iolock. Skip it in
+		 * xfs_free_eofblocks() to avoid deadlock. This also eliminates
+		 * the possibility of EAGAIN being returned.
+		 */
+		if (eofb->eof_scan_owner == ip->i_ino)
+			need_iolock = false;
 	}
 
-	ret = xfs_free_eofblocks(ip->i_mount, ip, true);
+	ret = xfs_free_eofblocks(ip->i_mount, ip, need_iolock);
 
 	/* don't revisit the inode if we're not waiting */
-	if (ret == EAGAIN && !(flags & SYNC_WAIT))
+	if (ret == -EAGAIN && !(flags & SYNC_WAIT))
 		ret = 0;
 
 	return ret;
@@ -1264,6 +1303,55 @@ xfs_icache_free_eofblocks(
 					 eofb, XFS_ICI_EOFBLOCKS_TAG);
 }
 
+/*
+ * Run eofblocks scans on the quotas applicable to the inode. For inodes with
+ * multiple quotas, we don't know exactly which quota caused an allocation
+ * failure. We make a best effort by including each quota under low free space
+ * conditions (less than 1% free space) in the scan.
+ */
+int
+xfs_inode_free_quota_eofblocks(
+	struct xfs_inode *ip)
+{
+	int scan = 0;
+	struct xfs_eofblocks eofb = {0};
+	struct xfs_dquot *dq;
+
+	ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
+
+	/*
+	 * Set the scan owner to avoid a potential livelock. Otherwise, the scan
+	 * can repeatedly trylock on the inode we're currently processing. We
+	 * run a sync scan to increase effectiveness and use the union filter to
+	 * cover all applicable quotas in a single scan.
+	 */
+	eofb.eof_scan_owner = ip->i_ino;
+	eofb.eof_flags = XFS_EOF_FLAGS_UNION|XFS_EOF_FLAGS_SYNC;
+
+	if (XFS_IS_UQUOTA_ENFORCED(ip->i_mount)) {
+		dq = xfs_inode_dquot(ip, XFS_DQ_USER);
+		if (dq && xfs_dquot_lowsp(dq)) {
+			eofb.eof_uid = VFS_I(ip)->i_uid;
+			eofb.eof_flags |= XFS_EOF_FLAGS_UID;
+			scan = 1;
+		}
+	}
+
+	if (XFS_IS_GQUOTA_ENFORCED(ip->i_mount)) {
+		dq = xfs_inode_dquot(ip, XFS_DQ_GROUP);
+		if (dq && xfs_dquot_lowsp(dq)) {
+			eofb.eof_gid = VFS_I(ip)->i_gid;
+			eofb.eof_flags |= XFS_EOF_FLAGS_GID;
+			scan = 1;
+		}
+	}
+
+	if (scan)
+		xfs_icache_free_eofblocks(ip->i_mount, &eofb);
+
+	return scan;
+}
+
 void
 xfs_inode_set_eofblocks_tag(
 	xfs_inode_t	*ip)
diff --git a/fs/xfs/xfs_icache.h b/fs/xfs/xfs_icache.h
index 9ed68bb750f5..46748b86b12f 100644
--- a/fs/xfs/xfs_icache.h
+++ b/fs/xfs/xfs_icache.h
@@ -27,6 +27,7 @@ struct xfs_eofblocks {
 	kgid_t		eof_gid;
 	prid_t		eof_prid;
 	__u64		eof_min_file_size;
+	xfs_ino_t	eof_scan_owner;
 };
 
 #define SYNC_WAIT		0x0001	/* wait for i/o to complete */
@@ -57,15 +58,14 @@ void xfs_inode_set_reclaim_tag(struct xfs_inode *ip);
 void xfs_inode_set_eofblocks_tag(struct xfs_inode *ip);
 void xfs_inode_clear_eofblocks_tag(struct xfs_inode *ip);
 int xfs_icache_free_eofblocks(struct xfs_mount *, struct xfs_eofblocks *);
+int xfs_inode_free_quota_eofblocks(struct xfs_inode *ip);
 void xfs_eofblocks_worker(struct work_struct *);
 
 int xfs_inode_ag_iterator(struct xfs_mount *mp,
-	int (*execute)(struct xfs_inode *ip, struct xfs_perag *pag,
-		int flags, void *args),
+	int (*execute)(struct xfs_inode *ip, int flags, void *args),
 	int flags, void *args);
 int xfs_inode_ag_iterator_tag(struct xfs_mount *mp,
-	int (*execute)(struct xfs_inode *ip, struct xfs_perag *pag,
-		int flags, void *args),
+	int (*execute)(struct xfs_inode *ip, int flags, void *args),
 	int flags, void *args, int tag);
 
 static inline int
@@ -74,31 +74,32 @@ xfs_fs_eofblocks_from_user(
 	struct xfs_eofblocks		*dst)
 {
 	if (src->eof_version != XFS_EOFBLOCKS_VERSION)
-		return EINVAL;
+		return -EINVAL;
 
 	if (src->eof_flags & ~XFS_EOF_FLAGS_VALID)
-		return EINVAL;
+		return -EINVAL;
 
 	if (memchr_inv(&src->pad32, 0, sizeof(src->pad32)) ||
 	    memchr_inv(src->pad64, 0, sizeof(src->pad64)))
-		return EINVAL;
+		return -EINVAL;
 
 	dst->eof_flags = src->eof_flags;
 	dst->eof_prid = src->eof_prid;
 	dst->eof_min_file_size = src->eof_min_file_size;
+	dst->eof_scan_owner = NULLFSINO;
 
 	dst->eof_uid = INVALID_UID;
 	if (src->eof_flags & XFS_EOF_FLAGS_UID) {
 		dst->eof_uid = make_kuid(current_user_ns(), src->eof_uid);
 		if (!uid_valid(dst->eof_uid))
-			return EINVAL;
+			return -EINVAL;
 	}
 
 	dst->eof_gid = INVALID_GID;
 	if (src->eof_flags & XFS_EOF_FLAGS_GID) {
 		dst->eof_gid = make_kgid(current_user_ns(), src->eof_gid);
 		if (!gid_valid(dst->eof_gid))
-			return EINVAL;
+			return -EINVAL;
 	}
 	return 0;
 }
diff --git a/fs/xfs/xfs_icreate_item.c b/fs/xfs/xfs_icreate_item.c
index d2eaccfa73f4..7e4549233251 100644
--- a/fs/xfs/xfs_icreate_item.c
+++ b/fs/xfs/xfs_icreate_item.c
@@ -28,6 +28,7 @@
 #include "xfs_trans_priv.h"
 #include "xfs_error.h"
 #include "xfs_icreate_item.h"
+#include "xfs_log.h"
 
 kmem_zone_t	*xfs_icreate_zone;		/* inode create item zone */
 
@@ -58,13 +59,14 @@ xfs_icreate_item_size(
 STATIC void
 xfs_icreate_item_format(
 	struct xfs_log_item	*lip,
-	struct xfs_log_iovec	*log_vector)
+	struct xfs_log_vec	*lv)
 {
 	struct xfs_icreate_item	*icp = ICR_ITEM(lip);
+	struct xfs_log_iovec	*vecp = NULL;
 
-	log_vector->i_addr = (xfs_caddr_t)&icp->ic_format;
-	log_vector->i_len  = sizeof(struct xfs_icreate_log);
-	log_vector->i_type = XLOG_REG_TYPE_ICREATE;
+	xlog_copy_iovec(lv, &vecp, XLOG_REG_TYPE_ICREATE,
+			&icp->ic_format,
+			sizeof(struct xfs_icreate_log));
 }
 
 
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 001aa893ed59..fea3c92fb3f0 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -42,7 +42,6 @@
 #include "xfs_bmap_util.h"
 #include "xfs_error.h"
 #include "xfs_quota.h"
-#include "xfs_dinode.h"
 #include "xfs_filestream.h"
 #include "xfs_cksum.h"
 #include "xfs_trace.h"
@@ -62,6 +61,8 @@ kmem_zone_t *xfs_inode_zone;
 
 STATIC int xfs_iflush_int(xfs_inode_t *, xfs_buf_t *);
 
+STATIC int xfs_iunlink_remove(xfs_trans_t *, xfs_inode_t *);
+
 /*
  * helper function to extract extent size hint from inode
  */
@@ -77,48 +78,44 @@ xfs_get_extsz_hint(
 }
 
 /*
- * This is a wrapper routine around the xfs_ilock() routine used to centralize
- * some grungy code.  It is used in places that wish to lock the inode solely
- * for reading the extents.  The reason these places can't just call
- * xfs_ilock(SHARED) is that the inode lock also guards to bringing in of the
- * extents from disk for a file in b-tree format.  If the inode is in b-tree
- * format, then we need to lock the inode exclusively until the extents are read
- * in.  Locking it exclusively all the time would limit our parallelism
- * unnecessarily, though.  What we do instead is check to see if the extents
- * have been read in yet, and only lock the inode exclusively if they have not.
+ * These two are wrapper routines around the xfs_ilock() routine used to
+ * centralize some grungy code.  They are used in places that wish to lock the
+ * inode solely for reading the extents.  The reason these places can't just
+ * call xfs_ilock(ip, XFS_ILOCK_SHARED) is that the inode lock also guards to
+ * bringing in of the extents from disk for a file in b-tree format.  If the
+ * inode is in b-tree format, then we need to lock the inode exclusively until
+ * the extents are read in.  Locking it exclusively all the time would limit
+ * our parallelism unnecessarily, though.  What we do instead is check to see
+ * if the extents have been read in yet, and only lock the inode exclusively
+ * if they have not.
  *
- * The function returns a value which should be given to the corresponding
- * xfs_iunlock_map_shared().  This value is the mode in which the lock was
- * actually taken.
+ * The functions return a value which should be given to the corresponding
+ * xfs_iunlock() call.
  */
 uint
-xfs_ilock_map_shared(
-	xfs_inode_t	*ip)
+xfs_ilock_data_map_shared(
+	struct xfs_inode	*ip)
 {
-	uint	lock_mode;
+	uint			lock_mode = XFS_ILOCK_SHARED;
 
-	if ((ip->i_d.di_format == XFS_DINODE_FMT_BTREE) &&
-	    ((ip->i_df.if_flags & XFS_IFEXTENTS) == 0)) {
+	if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
+	    (ip->i_df.if_flags & XFS_IFEXTENTS) == 0)
 		lock_mode = XFS_ILOCK_EXCL;
-	} else {
-		lock_mode = XFS_ILOCK_SHARED;
-	}
-
 	xfs_ilock(ip, lock_mode);
-
 	return lock_mode;
 }
 
-/*
- * This is simply the unlock routine to go with xfs_ilock_map_shared().
- * All it does is call xfs_iunlock() with the given lock_mode.
- */
-void
-xfs_iunlock_map_shared(
-	xfs_inode_t	*ip,
-	unsigned int	lock_mode)
+uint
+xfs_ilock_attr_map_shared(
+	struct xfs_inode	*ip)
 {
-	xfs_iunlock(ip, lock_mode);
+	uint			lock_mode = XFS_ILOCK_SHARED;
+
+	if (ip->i_d.di_aformat == XFS_DINODE_FMT_BTREE &&
+	    (ip->i_afp->if_flags & XFS_IFEXTENTS) == 0)
+		lock_mode = XFS_ILOCK_EXCL;
+	xfs_ilock(ip, lock_mode);
+	return lock_mode;
 }
 
 /*
@@ -586,11 +583,11 @@ xfs_lookup(
 	trace_xfs_lookup(dp, name);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
-	lock_mode = xfs_ilock_map_shared(dp);
+	lock_mode = xfs_ilock_data_map_shared(dp);
 	error = xfs_dir_lookup(NULL, dp, name, &inum, ci_name);
-	xfs_iunlock_map_shared(dp, lock_mode);
+	xfs_iunlock(dp, lock_mode);
 
 	if (error)
 		goto out;
@@ -658,7 +655,6 @@ xfs_ialloc(
 	uint		flags;
 	int		error;
 	timespec_t	tv;
-	int		filestreams = 0;
 
 	/*
 	 * Call the space management code to pick
@@ -685,6 +681,14 @@ xfs_ialloc(
 		return error;
 	ASSERT(ip != NULL);
 
+	/*
+	 * We always convert v1 inodes to v2 now - we only support filesystems
+	 * with >= v2 inode capability, so there is no reason for ever leaving
+	 * an inode in v1 format.
+	 */
+	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;
@@ -694,27 +698,6 @@ xfs_ialloc(
 	xfs_set_projid(ip, prid);
 	memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad));
 
-	/*
-	 * If the superblock version is up to where we support new format
-	 * inodes and this is currently an old format inode, then change
-	 * the inode version number now.  This way we only do the conversion
-	 * here rather than here and in the flush/logging code.
-	 */
-	if (xfs_sb_version_hasnlink(&mp->m_sb) &&
-	    ip->i_d.di_version == 1) {
-		ip->i_d.di_version = 2;
-		/*
-		 * We've already zeroed the old link count, the projid field,
-		 * and the pad field.
-		 */
-	}
-
-	/*
-	 * Project ids won't be stored on disk if we are using a version 1 inode.
-	 */
-	if ((prid != 0) && (ip->i_d.di_version == 1))
-		xfs_bump_ino_vers2(tp, ip);
-
 	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)) {
@@ -775,13 +758,6 @@ xfs_ialloc(
 		flags |= XFS_ILOG_DEV;
 		break;
 	case S_IFREG:
-		/*
-		 * we can't set up filestreams until after the VFS inode
-		 * is set up properly.
-		 */
-		if (pip && xfs_inode_is_filestream(pip))
-			filestreams = 1;
-		/* fall through */
 	case S_IFDIR:
 		if (pip && (pip->i_d.di_flags & XFS_DIFLAG_ANY)) {
 			uint	di_flags = 0;
@@ -847,15 +823,6 @@ xfs_ialloc(
 	/* now that we have an i_mode we can setup inode ops and unlock */
 	xfs_setup_inode(ip);
 
-	/* now we have set up the vfs inode we can associate the filestream */
-	if (filestreams) {
-		error = xfs_filestream_associate(pip, ip);
-		if (error < 0)
-			return -error;
-		if (!error)
-			xfs_iflags_set(ip, XFS_IFILESTREAM);
-	}
-
 	*ipp = ip;
 	return 0;
 }
@@ -926,7 +893,7 @@ xfs_dir_ialloc(
 	}
 	if (!ialloc_context && !ip) {
 		*ipp = NULL;
-		return XFS_ERROR(ENOSPC);
+		return -ENOSPC;
 	}
 
 	/*
@@ -1076,40 +1043,6 @@ xfs_droplink(
 }
 
 /*
- * This gets called when the inode's version needs to be changed from 1 to 2.
- * Currently this happens when the nlink field overflows the old 16-bit value
- * or when chproj is called to change the project for the first time.
- * As a side effect the superblock version will also get rev'd
- * to contain the NLINK bit.
- */
-void
-xfs_bump_ino_vers2(
-	xfs_trans_t	*tp,
-	xfs_inode_t	*ip)
-{
-	xfs_mount_t	*mp;
-
-	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-	ASSERT(ip->i_d.di_version == 1);
-
-	ip->i_d.di_version = 2;
-	ip->i_d.di_onlink = 0;
-	memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad));
-	mp = tp->t_mountp;
-	if (!xfs_sb_version_hasnlink(&mp->m_sb)) {
-		spin_lock(&mp->m_sb_lock);
-		if (!xfs_sb_version_hasnlink(&mp->m_sb)) {
-			xfs_sb_version_addnlink(&mp->m_sb);
-			spin_unlock(&mp->m_sb_lock);
-			xfs_mod_sb(tp, XFS_SB_VERSIONNUM);
-		} else {
-			spin_unlock(&mp->m_sb_lock);
-		}
-	}
-	/* Caller must log the inode */
-}
-
-/*
  * Increment the link count on an inode & log the change.
  */
 int
@@ -1119,22 +1052,10 @@ xfs_bumplink(
 {
 	xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG);
 
-	ASSERT(ip->i_d.di_nlink > 0);
+	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));
-	if ((ip->i_d.di_version == 1) &&
-	    (ip->i_d.di_nlink > XFS_MAXLINK_1)) {
-		/*
-		 * The inode has increased its number of links beyond
-		 * what can fit in an old format inode.  It now needs
-		 * to be converted to a version 2 inode with a 32 bit
-		 * link count.  If this is the first inode in the file
-		 * system to do this, then we need to bump the superblock
-		 * version number as well.
-		 */
-		xfs_bump_ino_vers2(tp, ip);
-	}
-
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 	return 0;
 }
@@ -1167,12 +1088,9 @@ xfs_create(
 	trace_xfs_create(dp, name);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
-	if (dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)
-		prid = xfs_get_projid(dp);
-	else
-		prid = XFS_PROJID_DEFAULT;
+	prid = xfs_get_initial_prid(dp);
 
 	/*
 	 * Make sure that we have allocated dquot(s) on disk.
@@ -1207,12 +1125,12 @@ xfs_create(
 	 */
 	tres.tr_logflags = XFS_TRANS_PERM_LOG_RES;
 	error = xfs_trans_reserve(tp, &tres, resblks, 0);
-	if (error == ENOSPC) {
+	if (error == -ENOSPC) {
 		/* flush outstanding delalloc blocks and retry */
 		xfs_flush_inodes(mp);
 		error = xfs_trans_reserve(tp, &tres, resblks, 0);
 	}
-	if (error == ENOSPC) {
+	if (error == -ENOSPC) {
 		/* No space at all so try a "no-allocation" reservation */
 		resblks = 0;
 		error = xfs_trans_reserve(tp, &tres, 0, 0);
@@ -1247,7 +1165,7 @@ xfs_create(
 	error = xfs_dir_ialloc(&tp, dp, mode, is_dir ? 2 : 1, rdev,
 			       prid, resblks > 0, &ip, &committed);
 	if (error) {
-		if (error == ENOSPC)
+		if (error == -ENOSPC)
 			goto out_trans_cancel;
 		goto out_trans_abort;
 	}
@@ -1266,7 +1184,7 @@ xfs_create(
 					&first_block, &free_list, resblks ?
 					resblks - XFS_IALLOC_SPACE_RES(mp) : 0);
 	if (error) {
-		ASSERT(error != ENOSPC);
+		ASSERT(error != -ENOSPC);
 		goto out_trans_abort;
 	}
 	xfs_trans_ichgtime(tp, dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
@@ -1337,6 +1255,114 @@ xfs_create(
 }
 
 int
+xfs_create_tmpfile(
+	struct xfs_inode	*dp,
+	struct dentry		*dentry,
+	umode_t			mode,
+	struct xfs_inode	**ipp)
+{
+	struct xfs_mount	*mp = dp->i_mount;
+	struct xfs_inode	*ip = NULL;
+	struct xfs_trans	*tp = NULL;
+	int			error;
+	uint			cancel_flags = XFS_TRANS_RELEASE_LOG_RES;
+	prid_t                  prid;
+	struct xfs_dquot	*udqp = NULL;
+	struct xfs_dquot	*gdqp = NULL;
+	struct xfs_dquot	*pdqp = NULL;
+	struct xfs_trans_res	*tres;
+	uint			resblks;
+
+	if (XFS_FORCED_SHUTDOWN(mp))
+		return -EIO;
+
+	prid = xfs_get_initial_prid(dp);
+
+	/*
+	 * Make sure that we have allocated dquot(s) on disk.
+	 */
+	error = xfs_qm_vop_dqalloc(dp, xfs_kuid_to_uid(current_fsuid()),
+				xfs_kgid_to_gid(current_fsgid()), prid,
+				XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT,
+				&udqp, &gdqp, &pdqp);
+	if (error)
+		return error;
+
+	resblks = XFS_IALLOC_SPACE_RES(mp);
+	tp = xfs_trans_alloc(mp, XFS_TRANS_CREATE_TMPFILE);
+
+	tres = &M_RES(mp)->tr_create_tmpfile;
+	error = xfs_trans_reserve(tp, tres, resblks, 0);
+	if (error == -ENOSPC) {
+		/* No space at all so try a "no-allocation" reservation */
+		resblks = 0;
+		error = xfs_trans_reserve(tp, tres, 0, 0);
+	}
+	if (error) {
+		cancel_flags = 0;
+		goto out_trans_cancel;
+	}
+
+	error = xfs_trans_reserve_quota(tp, mp, udqp, gdqp,
+						pdqp, resblks, 1, 0);
+	if (error)
+		goto out_trans_cancel;
+
+	error = xfs_dir_ialloc(&tp, dp, mode, 1, 0,
+				prid, resblks > 0, &ip, NULL);
+	if (error) {
+		if (error == -ENOSPC)
+			goto out_trans_cancel;
+		goto out_trans_abort;
+	}
+
+	if (mp->m_flags & XFS_MOUNT_WSYNC)
+		xfs_trans_set_sync(tp);
+
+	/*
+	 * Attach the dquot(s) to the inodes and modify them incore.
+	 * These ids of the inode couldn't have changed since the new
+	 * inode has been locked ever since it was created.
+	 */
+	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_abort;
+
+	error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
+	if (error)
+		goto out_release_inode;
+
+	xfs_qm_dqrele(udqp);
+	xfs_qm_dqrele(gdqp);
+	xfs_qm_dqrele(pdqp);
+
+	*ipp = ip;
+	return 0;
+
+ out_trans_abort:
+	cancel_flags |= XFS_TRANS_ABORT;
+ out_trans_cancel:
+	xfs_trans_cancel(tp, cancel_flags);
+ out_release_inode:
+	/*
+	 * Wait until after the current transaction is aborted to
+	 * release the inode.  This prevents recursive transactions
+	 * and deadlocks from xfs_inactive.
+	 */
+	if (ip)
+		IRELE(ip);
+
+	xfs_qm_dqrele(udqp);
+	xfs_qm_dqrele(gdqp);
+	xfs_qm_dqrele(pdqp);
+
+	return error;
+}
+
+int
 xfs_link(
 	xfs_inode_t		*tdp,
 	xfs_inode_t		*sip,
@@ -1356,7 +1382,7 @@ xfs_link(
 	ASSERT(!S_ISDIR(sip->i_d.di_mode));
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	error = xfs_qm_dqattach(sip, 0);
 	if (error)
@@ -1370,7 +1396,7 @@ xfs_link(
 	cancel_flags = XFS_TRANS_RELEASE_LOG_RES;
 	resblks = XFS_LINK_SPACE_RES(mp, target_name->len);
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_link, resblks, 0);
-	if (error == ENOSPC) {
+	if (error == -ENOSPC) {
 		resblks = 0;
 		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_link, 0, 0);
 	}
@@ -1391,7 +1417,7 @@ xfs_link(
 	 */
 	if (unlikely((tdp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) &&
 		     (xfs_get_projid(tdp) != xfs_get_projid(sip)))) {
-		error = XFS_ERROR(EXDEV);
+		error = -EXDEV;
 		goto error_return;
 	}
 
@@ -1401,6 +1427,12 @@ xfs_link(
 
 	xfs_bmap_init(&free_list, &first_block);
 
+	if (sip->i_d.di_nlink == 0) {
+		error = xfs_iunlink_remove(tp, sip);
+		if (error)
+			goto abort_return;
+	}
+
 	error = xfs_dir_createname(tp, tdp, target_name, sip->i_ino,
 					&first_block, &free_list, resblks);
 	if (error)
@@ -1591,16 +1623,6 @@ xfs_release(
 		int truncated;
 
 		/*
-		 * If we are using filestreams, and we have an unlinked
-		 * file that we are processing the last close on, then nothing
-		 * will be able to reopen and write to this file. Purge this
-		 * inode from the filestreams cache so that it doesn't delay
-		 * teardown of the inode.
-		 */
-		if ((ip->i_d.di_nlink == 0) && xfs_inode_is_filestream(ip))
-			xfs_filestream_deassociate(ip);
-
-		/*
 		 * If we previously truncated this file and removed old data
 		 * in the process, we want to initiate "early" writeout on
 		 * the last close.  This is an attempt to combat the notorious
@@ -1613,8 +1635,8 @@ xfs_release(
 		truncated = xfs_iflags_test_and_clear(ip, XFS_ITRUNCATED);
 		if (truncated) {
 			xfs_iflags_clear(ip, XFS_IDIRTY_RELEASE);
-			if (VN_DIRTY(VFS_I(ip)) && ip->i_delayed_blks > 0) {
-				error = -filemap_flush(VFS_I(ip)->i_mapping);
+			if (ip->i_delayed_blks > 0) {
+				error = filemap_flush(VFS_I(ip)->i_mapping);
 				if (error)
 					return error;
 			}
@@ -1651,7 +1673,7 @@ xfs_release(
 			return 0;
 
 		error = xfs_free_eofblocks(mp, ip, true);
-		if (error && error != EAGAIN)
+		if (error && error != -EAGAIN)
 			return error;
 
 		/* delalloc blocks after truncation means it really is dirty */
@@ -1730,9 +1752,33 @@ xfs_inactive_ifree(
 	int			error;
 
 	tp = xfs_trans_alloc(mp, XFS_TRANS_INACTIVE);
-	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ifree, 0, 0);
+
+	/*
+	 * The ifree transaction might need to allocate blocks for record
+	 * insertion to the finobt. We don't want to fail here at ENOSPC, so
+	 * allow ifree to dip into the reserved block pool if necessary.
+	 *
+	 * Freeing large sets of inodes generally means freeing inode chunks,
+	 * directory and file data blocks, so this should be relatively safe.
+	 * Only under severe circumstances should it be possible to free enough
+	 * inodes to exhaust the reserve block pool via finobt expansion while
+	 * at the same time not creating free space in the filesystem.
+	 *
+	 * Send a warning if the reservation does happen to fail, as the inode
+	 * now remains allocated and sits on the unlinked list until the fs is
+	 * repaired.
+	 */
+	tp->t_flags |= XFS_TRANS_RESERVE;
+	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ifree,
+				  XFS_IFREE_SPACE_RES(mp), 0);
 	if (error) {
-		ASSERT(XFS_FORCED_SHUTDOWN(mp));
+		if (error == -ENOSPC) {
+			xfs_warn_ratelimited(mp,
+			"Failed to remove inode(s) from unlinked list. "
+			"Please free space, unmount and run xfs_repair.");
+		} else {
+			ASSERT(XFS_FORCED_SHUTDOWN(mp));
+		}
 		xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES);
 		return error;
 	}
@@ -2141,8 +2187,8 @@ xfs_ifree_cluster(
 {
 	xfs_mount_t		*mp = free_ip->i_mount;
 	int			blks_per_cluster;
+	int			inodes_per_cluster;
 	int			nbufs;
-	int			ninodes;
 	int			i, j;
 	xfs_daddr_t		blkno;
 	xfs_buf_t		*bp;
@@ -2152,18 +2198,11 @@ xfs_ifree_cluster(
 	struct xfs_perag	*pag;
 
 	pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, inum));
-	if (mp->m_sb.sb_blocksize >= XFS_INODE_CLUSTER_SIZE(mp)) {
-		blks_per_cluster = 1;
-		ninodes = mp->m_sb.sb_inopblock;
-		nbufs = XFS_IALLOC_BLOCKS(mp);
-	} else {
-		blks_per_cluster = XFS_INODE_CLUSTER_SIZE(mp) /
-					mp->m_sb.sb_blocksize;
-		ninodes = blks_per_cluster * mp->m_sb.sb_inopblock;
-		nbufs = XFS_IALLOC_BLOCKS(mp) / blks_per_cluster;
-	}
+	blks_per_cluster = xfs_icluster_size_fsb(mp);
+	inodes_per_cluster = blks_per_cluster << mp->m_sb.sb_inopblog;
+	nbufs = mp->m_ialloc_blks / blks_per_cluster;
 
-	for (j = 0; j < nbufs; j++, inum += ninodes) {
+	for (j = 0; j < nbufs; j++, inum += inodes_per_cluster) {
 		blkno = XFS_AGB_TO_DADDR(mp, XFS_INO_TO_AGNO(mp, inum),
 					 XFS_INO_TO_AGBNO(mp, inum));
 
@@ -2180,7 +2219,7 @@ xfs_ifree_cluster(
 					XBF_UNMAPPED);
 
 		if (!bp)
-			return ENOMEM;
+			return -ENOMEM;
 
 		/*
 		 * This buffer may not have been correctly initialised as we
@@ -2225,7 +2264,7 @@ xfs_ifree_cluster(
 		 * transaction stale above, which means there is no point in
 		 * even trying to lock them.
 		 */
-		for (i = 0; i < ninodes; i++) {
+		for (i = 0; i < inodes_per_cluster; i++) {
 retry:
 			rcu_read_lock();
 			ip = radix_tree_lookup(&pag->pag_ici_root,
@@ -2452,7 +2491,7 @@ xfs_remove(
 	trace_xfs_remove(dp, name);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	error = xfs_qm_dqattach(dp, 0);
 	if (error)
@@ -2482,12 +2521,12 @@ xfs_remove(
 	 */
 	resblks = XFS_REMOVE_SPACE_RES(mp);
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_remove, resblks, 0);
-	if (error == ENOSPC) {
+	if (error == -ENOSPC) {
 		resblks = 0;
 		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_remove, 0, 0);
 	}
 	if (error) {
-		ASSERT(error != ENOSPC);
+		ASSERT(error != -ENOSPC);
 		cancel_flags = 0;
 		goto out_trans_cancel;
 	}
@@ -2504,11 +2543,11 @@ xfs_remove(
 	if (is_dir) {
 		ASSERT(ip->i_d.di_nlink >= 2);
 		if (ip->i_d.di_nlink != 2) {
-			error = XFS_ERROR(ENOTEMPTY);
+			error = -ENOTEMPTY;
 			goto out_trans_cancel;
 		}
 		if (!xfs_dir_isempty(ip)) {
-			error = XFS_ERROR(ENOTEMPTY);
+			error = -ENOTEMPTY;
 			goto out_trans_cancel;
 		}
 
@@ -2543,7 +2582,7 @@ xfs_remove(
 	error = xfs_dir_removename(tp, dp, name, ip->i_ino,
 					&first_block, &free_list, resblks);
 	if (error) {
-		ASSERT(error != ENOENT);
+		ASSERT(error != -ENOENT);
 		goto out_bmap_cancel;
 	}
 
@@ -2563,13 +2602,7 @@ xfs_remove(
 	if (error)
 		goto std_return;
 
-	/*
-	 * If we are using filestreams, kill the stream association.
-	 * If the file is still open it may get a new one but that
-	 * will get killed on last close in xfs_close() so we don't
-	 * have to worry about that.
-	 */
-	if (!is_dir && link_zero && xfs_inode_is_filestream(ip))
+	if (is_dir && xfs_inode_is_filestream(ip))
 		xfs_filestream_deassociate(ip);
 
 	return 0;
@@ -2669,7 +2702,7 @@ xfs_rename(
 	cancel_flags = XFS_TRANS_RELEASE_LOG_RES;
 	spaceres = XFS_RENAME_SPACE_RES(mp, target_name->len);
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_rename, spaceres, 0);
-	if (error == ENOSPC) {
+	if (error == -ENOSPC) {
 		spaceres = 0;
 		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_rename, 0, 0);
 	}
@@ -2714,7 +2747,7 @@ xfs_rename(
 	 */
 	if (unlikely((target_dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) &&
 		     (xfs_get_projid(target_dp) != xfs_get_projid(src_ip)))) {
-		error = XFS_ERROR(EXDEV);
+		error = -EXDEV;
 		goto error_return;
 	}
 
@@ -2737,7 +2770,7 @@ xfs_rename(
 		error = xfs_dir_createname(tp, target_dp, target_name,
 						src_ip->i_ino, &first_block,
 						&free_list, spaceres);
-		if (error == ENOSPC)
+		if (error == -ENOSPC)
 			goto error_return;
 		if (error)
 			goto abort_return;
@@ -2762,7 +2795,7 @@ xfs_rename(
 			 */
 			if (!(xfs_dir_isempty(target_ip)) ||
 			    (target_ip->i_d.di_nlink > 2)) {
-				error = XFS_ERROR(EEXIST);
+				error = -EEXIST;
 				goto error_return;
 			}
 		}
@@ -2814,7 +2847,7 @@ xfs_rename(
 		error = xfs_dir_replace(tp, src_ip, &xfs_name_dotdot,
 					target_dp->i_ino,
 					&first_block, &free_list, spaceres);
-		ASSERT(error != EEXIST);
+		ASSERT(error != -EEXIST);
 		if (error)
 			goto abort_return;
 	}
@@ -2906,13 +2939,13 @@ xfs_iflush_cluster(
 
 	pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino));
 
-	inodes_per_cluster = XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_inodelog;
+	inodes_per_cluster = mp->m_inode_cluster_size >> mp->m_sb.sb_inodelog;
 	ilist_size = inodes_per_cluster * sizeof(xfs_inode_t *);
 	ilist = kmem_alloc(ilist_size, KM_MAYFAIL|KM_NOFS);
 	if (!ilist)
 		goto out_put;
 
-	mask = ~(((XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_inodelog)) - 1);
+	mask = ~(((mp->m_inode_cluster_size >> mp->m_sb.sb_inodelog)) - 1);
 	first_index = XFS_INO_TO_AGINO(mp, ip->i_ino) & mask;
 	rcu_read_lock();
 	/* really need a gang lookup range call here */
@@ -3022,7 +3055,7 @@ cluster_corrupt_out:
 		if (bp->b_iodone) {
 			XFS_BUF_UNDONE(bp);
 			xfs_buf_stale(bp);
-			xfs_buf_ioerror(bp, EIO);
+			xfs_buf_ioerror(bp, -EIO);
 			xfs_buf_ioend(bp, 0);
 		} else {
 			xfs_buf_stale(bp);
@@ -3036,7 +3069,7 @@ cluster_corrupt_out:
 	xfs_iflush_abort(iq, false);
 	kmem_free(ilist);
 	xfs_perag_put(pag);
-	return XFS_ERROR(EFSCORRUPTED);
+	return -EFSCORRUPTED;
 }
 
 /*
@@ -3091,7 +3124,7 @@ xfs_iflush(
 	 * as we wait for an empty AIL as part of the unmount process.
 	 */
 	if (XFS_FORCED_SHUTDOWN(mp)) {
-		error = XFS_ERROR(EIO);
+		error = -EIO;
 		goto abort_out;
 	}
 
@@ -3134,7 +3167,7 @@ corrupt_out:
 	xfs_buf_relse(bp);
 	xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
 cluster_corrupt_out:
-	error = XFS_ERROR(EFSCORRUPTED);
+	error = -EFSCORRUPTED;
 abort_out:
 	/*
 	 * Unlocks the flush lock
@@ -3157,6 +3190,7 @@ xfs_iflush_int(
 	ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
 	       ip->i_d.di_nextents > XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK));
 	ASSERT(iip != NULL && iip->ili_fields != 0);
+	ASSERT(ip->i_d.di_version > 1);
 
 	/* set *dip = inode's place in the buffer */
 	dip = (xfs_dinode_t *)xfs_buf_offset(bp, ip->i_imap.im_boffset);
@@ -3217,7 +3251,7 @@ xfs_iflush_int(
 	}
 
 	/*
-	 * Inode item log recovery for v1/v2 inodes are dependent on the
+	 * Inode item log recovery for v2 inodes are dependent on the
 	 * di_flushiter count for correct sequencing. We bump the flush
 	 * iteration count so we can detect flushes which postdate a log record
 	 * during recovery. This is redundant as we now log every change and
@@ -3240,40 +3274,9 @@ xfs_iflush_int(
 	if (ip->i_d.di_flushiter == DI_MAX_FLUSH)
 		ip->i_d.di_flushiter = 0;
 
-	/*
-	 * If this is really an old format inode and the superblock version
-	 * has not been updated to support only new format inodes, then
-	 * convert back to the old inode format.  If the superblock version
-	 * has been updated, then make the conversion permanent.
-	 */
-	ASSERT(ip->i_d.di_version == 1 || xfs_sb_version_hasnlink(&mp->m_sb));
-	if (ip->i_d.di_version == 1) {
-		if (!xfs_sb_version_hasnlink(&mp->m_sb)) {
-			/*
-			 * Convert it back.
-			 */
-			ASSERT(ip->i_d.di_nlink <= XFS_MAXLINK_1);
-			dip->di_onlink = cpu_to_be16(ip->i_d.di_nlink);
-		} else {
-			/*
-			 * The superblock version has already been bumped,
-			 * so just make the conversion to the new inode
-			 * format permanent.
-			 */
-			ip->i_d.di_version = 2;
-			dip->di_version = 2;
-			ip->i_d.di_onlink = 0;
-			dip->di_onlink = 0;
-			memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad));
-			memset(&(dip->di_pad[0]), 0,
-			      sizeof(dip->di_pad));
-			ASSERT(xfs_get_projid(ip) == 0);
-		}
-	}
-
-	xfs_iflush_fork(ip, dip, iip, XFS_DATA_FORK, bp);
+	xfs_iflush_fork(ip, dip, iip, XFS_DATA_FORK);
 	if (XFS_IFORK_Q(ip))
-		xfs_iflush_fork(ip, dip, iip, XFS_ATTR_FORK, bp);
+		xfs_iflush_fork(ip, dip, iip, XFS_ATTR_FORK);
 	xfs_inobp_check(mp, bp);
 
 	/*
@@ -3328,5 +3331,5 @@ xfs_iflush_int(
 	return 0;
 
 corrupt_out:
-	return XFS_ERROR(EFSCORRUPTED);
+	return -EFSCORRUPTED;
 }
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 9e6efccbae04..c10e3fadd9af 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -20,6 +20,7 @@
 
 #include "xfs_inode_buf.h"
 #include "xfs_inode_fork.h"
+#include "xfs_dinode.h"
 
 /*
  * Kernel only inode definitions
@@ -192,6 +193,15 @@ xfs_set_projid(struct xfs_inode *ip,
 	ip->i_d.di_projid_lo = (__uint16_t) (projid & 0xffff);
 }
 
+static inline prid_t
+xfs_get_initial_prid(struct xfs_inode *dp)
+{
+	if (dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)
+		return xfs_get_projid(dp);
+
+	return XFS_PROJID_DEFAULT;
+}
+
 /*
  * In-core inode flags.
  */
@@ -199,7 +209,6 @@ xfs_set_projid(struct xfs_inode *ip,
 #define XFS_ISTALE		(1 << 1) /* inode has been staled */
 #define XFS_IRECLAIMABLE	(1 << 2) /* inode can be reclaimed */
 #define XFS_INEW		(1 << 3) /* inode has just been allocated */
-#define XFS_IFILESTREAM		(1 << 4) /* inode is in a filestream dir. */
 #define XFS_ITRUNCATED		(1 << 5) /* truncated down so flush-on-close */
 #define XFS_IDIRTY_RELEASE	(1 << 6) /* dirty release already seen */
 #define __XFS_IFLOCK_BIT	7	 /* inode is being flushed right now */
@@ -215,8 +224,7 @@ xfs_set_projid(struct xfs_inode *ip,
  */
 #define XFS_IRECLAIM_RESET_FLAGS	\
 	(XFS_IRECLAIMABLE | XFS_IRECLAIM | \
-	 XFS_IDIRTY_RELEASE | XFS_ITRUNCATED | \
-	 XFS_IFILESTREAM);
+	 XFS_IDIRTY_RELEASE | XFS_ITRUNCATED)
 
 /*
  * Synchronize processes attempting to flush the in-core inode back to disk.
@@ -323,6 +331,8 @@ int		xfs_lookup(struct xfs_inode *dp, struct xfs_name *name,
 			   struct xfs_inode **ipp, struct xfs_name *ci_name);
 int		xfs_create(struct xfs_inode *dp, struct xfs_name *name,
 			   umode_t mode, xfs_dev_t rdev, struct xfs_inode **ipp);
+int		xfs_create_tmpfile(struct xfs_inode *dp, struct dentry *dentry,
+			   umode_t mode, struct xfs_inode **ipp);
 int		xfs_remove(struct xfs_inode *dp, struct xfs_name *name,
 			   struct xfs_inode *ip);
 int		xfs_link(struct xfs_inode *tdp, struct xfs_inode *sip,
@@ -337,8 +347,8 @@ int		xfs_ilock_nowait(xfs_inode_t *, uint);
 void		xfs_iunlock(xfs_inode_t *, uint);
 void		xfs_ilock_demote(xfs_inode_t *, uint);
 int		xfs_isilocked(xfs_inode_t *, uint);
-uint		xfs_ilock_map_shared(xfs_inode_t *);
-void		xfs_iunlock_map_shared(xfs_inode_t *, uint);
+uint		xfs_ilock_data_map_shared(struct xfs_inode *);
+uint		xfs_ilock_attr_map_shared(struct xfs_inode *);
 int		xfs_ialloc(struct xfs_trans *, xfs_inode_t *, umode_t,
 			   xfs_nlink_t, xfs_dev_t, prid_t, int,
 			   struct xfs_buf **, xfs_inode_t **);
@@ -367,7 +377,6 @@ int		xfs_dir_ialloc(struct xfs_trans **, struct xfs_inode *, umode_t,
 			       struct xfs_inode **, int *);
 int		xfs_droplink(struct xfs_trans *, struct xfs_inode *);
 int		xfs_bumplink(struct xfs_trans *, struct xfs_inode *);
-void		xfs_bump_ino_vers2(struct xfs_trans *, struct xfs_inode *);
 
 /* from xfs_file.c */
 int		xfs_zero_eof(struct xfs_inode *, xfs_off_t, xfs_fsize_t);
@@ -389,4 +398,14 @@ do { \
 
 extern struct kmem_zone	*xfs_inode_zone;
 
+/*
+ * Flags for read/write calls
+ */
+#define XFS_IO_ISDIRECT	0x00001		/* bypass page cache */
+#define XFS_IO_INVIS	0x00002		/* don't update inode timestamps */
+
+#define XFS_IO_FLAGS \
+	{ XFS_IO_ISDIRECT,	"DIRECT" }, \
+	{ XFS_IO_INVIS,		"INVIS"}
+
 #endif	/* __XFS_INODE_H__ */
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index 7c0d391f9a6e..de5a7be36e60 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -30,6 +30,7 @@
 #include "xfs_trace.h"
 #include "xfs_trans_priv.h"
 #include "xfs_dinode.h"
+#include "xfs_log.h"
 
 
 kmem_zone_t	*xfs_ili_zone;		/* inode log item zone */
@@ -39,27 +40,14 @@ static inline struct xfs_inode_log_item *INODE_ITEM(struct xfs_log_item *lip)
 	return container_of(lip, struct xfs_inode_log_item, ili_item);
 }
 
-
-/*
- * This returns the number of iovecs needed to log the given inode item.
- *
- * We need one iovec for the inode log format structure, one for the
- * inode core, and possibly one for the inode data/extents/b-tree root
- * and one for the inode attribute data/extents/b-tree root.
- */
 STATIC void
-xfs_inode_item_size(
-	struct xfs_log_item	*lip,
+xfs_inode_item_data_fork_size(
+	struct xfs_inode_log_item *iip,
 	int			*nvecs,
 	int			*nbytes)
 {
-	struct xfs_inode_log_item *iip = INODE_ITEM(lip);
 	struct xfs_inode	*ip = iip->ili_inode;
 
-	*nvecs += 2;
-	*nbytes += sizeof(struct xfs_inode_log_format) +
-		   xfs_icdinode_size(ip->i_d.di_version);
-
 	switch (ip->i_d.di_format) {
 	case XFS_DINODE_FMT_EXTENTS:
 		if ((iip->ili_fields & XFS_ILOG_DEXT) &&
@@ -70,7 +58,6 @@ xfs_inode_item_size(
 			*nvecs += 1;
 		}
 		break;
-
 	case XFS_DINODE_FMT_BTREE:
 		if ((iip->ili_fields & XFS_ILOG_DBROOT) &&
 		    ip->i_df.if_broot_bytes > 0) {
@@ -78,7 +65,6 @@ xfs_inode_item_size(
 			*nvecs += 1;
 		}
 		break;
-
 	case XFS_DINODE_FMT_LOCAL:
 		if ((iip->ili_fields & XFS_ILOG_DDATA) &&
 		    ip->i_df.if_bytes > 0) {
@@ -90,19 +76,20 @@ xfs_inode_item_size(
 	case XFS_DINODE_FMT_DEV:
 	case XFS_DINODE_FMT_UUID:
 		break;
-
 	default:
 		ASSERT(0);
 		break;
 	}
+}
 
-	if (!XFS_IFORK_Q(ip))
-		return;
-
+STATIC void
+xfs_inode_item_attr_fork_size(
+	struct xfs_inode_log_item *iip,
+	int			*nvecs,
+	int			*nbytes)
+{
+	struct xfs_inode	*ip = iip->ili_inode;
 
-	/*
-	 * Log any necessary attribute data.
-	 */
 	switch (ip->i_d.di_aformat) {
 	case XFS_DINODE_FMT_EXTENTS:
 		if ((iip->ili_fields & XFS_ILOG_AEXT) &&
@@ -113,7 +100,6 @@ xfs_inode_item_size(
 			*nvecs += 1;
 		}
 		break;
-
 	case XFS_DINODE_FMT_BTREE:
 		if ((iip->ili_fields & XFS_ILOG_ABROOT) &&
 		    ip->i_afp->if_broot_bytes > 0) {
@@ -121,7 +107,6 @@ xfs_inode_item_size(
 			*nvecs += 1;
 		}
 		break;
-
 	case XFS_DINODE_FMT_LOCAL:
 		if ((iip->ili_fields & XFS_ILOG_ADATA) &&
 		    ip->i_afp->if_bytes > 0) {
@@ -129,7 +114,6 @@ xfs_inode_item_size(
 			*nvecs += 1;
 		}
 		break;
-
 	default:
 		ASSERT(0);
 		break;
@@ -137,98 +121,39 @@ xfs_inode_item_size(
 }
 
 /*
- * xfs_inode_item_format_extents - convert in-core extents to on-disk form
- *
- * For either the data or attr fork in extent format, we need to endian convert
- * the in-core extent as we place them into the on-disk inode. In this case, we
- * need to do this conversion before we write the extents into the log. Because
- * we don't have the disk inode to write into here, we allocate a buffer and
- * format the extents into it via xfs_iextents_copy(). We free the buffer in
- * the unlock routine after the copy for the log has been made.
+ * This returns the number of iovecs needed to log the given inode item.
  *
- * In the case of the data fork, the in-core and on-disk fork sizes can be
- * different due to delayed allocation extents. We only log on-disk extents
- * here, so always use the physical fork size to determine the size of the
- * buffer we need to allocate.
+ * We need one iovec for the inode log format structure, one for the
+ * inode core, and possibly one for the inode data/extents/b-tree root
+ * and one for the inode attribute data/extents/b-tree root.
  */
 STATIC void
-xfs_inode_item_format_extents(
-	struct xfs_inode	*ip,
-	struct xfs_log_iovec	*vecp,
-	int			whichfork,
-	int			type)
+xfs_inode_item_size(
+	struct xfs_log_item	*lip,
+	int			*nvecs,
+	int			*nbytes)
 {
-	xfs_bmbt_rec_t		*ext_buffer;
+	struct xfs_inode_log_item *iip = INODE_ITEM(lip);
+	struct xfs_inode	*ip = iip->ili_inode;
 
-	ext_buffer = kmem_alloc(XFS_IFORK_SIZE(ip, whichfork), KM_SLEEP);
-	if (whichfork == XFS_DATA_FORK)
-		ip->i_itemp->ili_extents_buf = ext_buffer;
-	else
-		ip->i_itemp->ili_aextents_buf = ext_buffer;
+	*nvecs += 2;
+	*nbytes += sizeof(struct xfs_inode_log_format) +
+		   xfs_icdinode_size(ip->i_d.di_version);
 
-	vecp->i_addr = ext_buffer;
-	vecp->i_len = xfs_iextents_copy(ip, ext_buffer, whichfork);
-	vecp->i_type = type;
+	xfs_inode_item_data_fork_size(iip, nvecs, nbytes);
+	if (XFS_IFORK_Q(ip))
+		xfs_inode_item_attr_fork_size(iip, nvecs, nbytes);
 }
 
-/*
- * 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, the second with the on-disk inode structure,
- * and a possible third and/or fourth with the inode data/extents/b-tree
- * root and inode attributes data/extents/b-tree root.
- */
 STATIC void
-xfs_inode_item_format(
-	struct xfs_log_item	*lip,
-	struct xfs_log_iovec	*vecp)
+xfs_inode_item_format_data_fork(
+	struct xfs_inode_log_item *iip,
+	struct xfs_inode_log_format *ilf,
+	struct xfs_log_vec	*lv,
+	struct xfs_log_iovec	**vecp)
 {
-	struct xfs_inode_log_item *iip = INODE_ITEM(lip);
 	struct xfs_inode	*ip = iip->ili_inode;
-	uint			nvecs;
 	size_t			data_bytes;
-	xfs_mount_t		*mp;
-
-	vecp->i_addr = &iip->ili_format;
-	vecp->i_len  = sizeof(xfs_inode_log_format_t);
-	vecp->i_type = XLOG_REG_TYPE_IFORMAT;
-	vecp++;
-	nvecs	     = 1;
-
-	vecp->i_addr = &ip->i_d;
-	vecp->i_len  = xfs_icdinode_size(ip->i_d.di_version);
-	vecp->i_type = XLOG_REG_TYPE_ICORE;
-	vecp++;
-	nvecs++;
-
-	/*
-	 * If this is really an old format inode, then we need to
-	 * log it as such.  This means that we have to copy the link
-	 * count from the new field to the old.  We don't have to worry
-	 * about the new fields, because nothing trusts them as long as
-	 * the old inode version number is there.  If the superblock already
-	 * has a new version number, then we don't bother converting back.
-	 */
-	mp = ip->i_mount;
-	ASSERT(ip->i_d.di_version == 1 || xfs_sb_version_hasnlink(&mp->m_sb));
-	if (ip->i_d.di_version == 1) {
-		if (!xfs_sb_version_hasnlink(&mp->m_sb)) {
-			/*
-			 * Convert it back.
-			 */
-			ASSERT(ip->i_d.di_nlink <= XFS_MAXLINK_1);
-			ip->i_d.di_onlink = ip->i_d.di_nlink;
-		} else {
-			/*
-			 * The superblock version has already been bumped,
-			 * so just make the conversion to the new inode
-			 * format permanent.
-			 */
-			ip->i_d.di_version = 2;
-			ip->i_d.di_onlink = 0;
-			memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad));
-		}
-	}
 
 	switch (ip->i_d.di_format) {
 	case XFS_DINODE_FMT_EXTENTS:
@@ -239,36 +164,23 @@ xfs_inode_item_format(
 		if ((iip->ili_fields & XFS_ILOG_DEXT) &&
 		    ip->i_d.di_nextents > 0 &&
 		    ip->i_df.if_bytes > 0) {
+			struct xfs_bmbt_rec *p;
+
 			ASSERT(ip->i_df.if_u1.if_extents != NULL);
 			ASSERT(ip->i_df.if_bytes / sizeof(xfs_bmbt_rec_t) > 0);
-			ASSERT(iip->ili_extents_buf == NULL);
-
-#ifdef XFS_NATIVE_HOST
-                       if (ip->i_d.di_nextents == ip->i_df.if_bytes /
-                                               (uint)sizeof(xfs_bmbt_rec_t)) {
-				/*
-				 * There are no delayed allocation
-				 * extents, so just point to the
-				 * real extents array.
-				 */
-				vecp->i_addr = ip->i_df.if_u1.if_extents;
-				vecp->i_len = ip->i_df.if_bytes;
-				vecp->i_type = XLOG_REG_TYPE_IEXT;
-			} else
-#endif
-			{
-				xfs_inode_item_format_extents(ip, vecp,
-					XFS_DATA_FORK, XLOG_REG_TYPE_IEXT);
-			}
-			ASSERT(vecp->i_len <= ip->i_df.if_bytes);
-			iip->ili_format.ilf_dsize = vecp->i_len;
-			vecp++;
-			nvecs++;
+
+			p = xlog_prepare_iovec(lv, vecp, XLOG_REG_TYPE_IEXT);
+			data_bytes = xfs_iextents_copy(ip, p, XFS_DATA_FORK);
+			xlog_finish_iovec(lv, *vecp, data_bytes);
+
+			ASSERT(data_bytes <= ip->i_df.if_bytes);
+
+			ilf->ilf_dsize = data_bytes;
+			ilf->ilf_size++;
 		} else {
 			iip->ili_fields &= ~XFS_ILOG_DEXT;
 		}
 		break;
-
 	case XFS_DINODE_FMT_BTREE:
 		iip->ili_fields &=
 			~(XFS_ILOG_DDATA | XFS_ILOG_DEXT |
@@ -277,80 +189,70 @@ xfs_inode_item_format(
 		if ((iip->ili_fields & XFS_ILOG_DBROOT) &&
 		    ip->i_df.if_broot_bytes > 0) {
 			ASSERT(ip->i_df.if_broot != NULL);
-			vecp->i_addr = ip->i_df.if_broot;
-			vecp->i_len = ip->i_df.if_broot_bytes;
-			vecp->i_type = XLOG_REG_TYPE_IBROOT;
-			vecp++;
-			nvecs++;
-			iip->ili_format.ilf_dsize = ip->i_df.if_broot_bytes;
+			xlog_copy_iovec(lv, vecp, XLOG_REG_TYPE_IBROOT,
+					ip->i_df.if_broot,
+					ip->i_df.if_broot_bytes);
+			ilf->ilf_dsize = ip->i_df.if_broot_bytes;
+			ilf->ilf_size++;
 		} else {
 			ASSERT(!(iip->ili_fields &
 				 XFS_ILOG_DBROOT));
 			iip->ili_fields &= ~XFS_ILOG_DBROOT;
 		}
 		break;
-
 	case XFS_DINODE_FMT_LOCAL:
 		iip->ili_fields &=
 			~(XFS_ILOG_DEXT | XFS_ILOG_DBROOT |
 			  XFS_ILOG_DEV | XFS_ILOG_UUID);
 		if ((iip->ili_fields & XFS_ILOG_DDATA) &&
 		    ip->i_df.if_bytes > 0) {
-			ASSERT(ip->i_df.if_u1.if_data != NULL);
-			ASSERT(ip->i_d.di_size > 0);
-
-			vecp->i_addr = ip->i_df.if_u1.if_data;
 			/*
 			 * Round i_bytes up to a word boundary.
 			 * The underlying memory is guaranteed to
 			 * to be there by xfs_idata_realloc().
 			 */
 			data_bytes = roundup(ip->i_df.if_bytes, 4);
-			ASSERT((ip->i_df.if_real_bytes == 0) ||
-			       (ip->i_df.if_real_bytes == data_bytes));
-			vecp->i_len = (int)data_bytes;
-			vecp->i_type = XLOG_REG_TYPE_ILOCAL;
-			vecp++;
-			nvecs++;
-			iip->ili_format.ilf_dsize = (unsigned)data_bytes;
+			ASSERT(ip->i_df.if_real_bytes == 0 ||
+			       ip->i_df.if_real_bytes == data_bytes);
+			ASSERT(ip->i_df.if_u1.if_data != NULL);
+			ASSERT(ip->i_d.di_size > 0);
+			xlog_copy_iovec(lv, vecp, XLOG_REG_TYPE_ILOCAL,
+					ip->i_df.if_u1.if_data, data_bytes);
+			ilf->ilf_dsize = (unsigned)data_bytes;
+			ilf->ilf_size++;
 		} else {
 			iip->ili_fields &= ~XFS_ILOG_DDATA;
 		}
 		break;
-
 	case XFS_DINODE_FMT_DEV:
 		iip->ili_fields &=
 			~(XFS_ILOG_DDATA | XFS_ILOG_DBROOT |
 			  XFS_ILOG_DEXT | XFS_ILOG_UUID);
-		if (iip->ili_fields & XFS_ILOG_DEV) {
-			iip->ili_format.ilf_u.ilfu_rdev =
-				ip->i_df.if_u2.if_rdev;
-		}
+		if (iip->ili_fields & XFS_ILOG_DEV)
+			ilf->ilf_u.ilfu_rdev = ip->i_df.if_u2.if_rdev;
 		break;
-
 	case XFS_DINODE_FMT_UUID:
 		iip->ili_fields &=
 			~(XFS_ILOG_DDATA | XFS_ILOG_DBROOT |
 			  XFS_ILOG_DEXT | XFS_ILOG_DEV);
-		if (iip->ili_fields & XFS_ILOG_UUID) {
-			iip->ili_format.ilf_u.ilfu_uuid =
-				ip->i_df.if_u2.if_uuid;
-		}
+		if (iip->ili_fields & XFS_ILOG_UUID)
+			ilf->ilf_u.ilfu_uuid = ip->i_df.if_u2.if_uuid;
 		break;
-
 	default:
 		ASSERT(0);
 		break;
 	}
+}
 
-	/*
-	 * If there are no attributes associated with the file, then we're done.
-	 */
-	if (!XFS_IFORK_Q(ip)) {
-		iip->ili_fields &=
-			~(XFS_ILOG_ADATA | XFS_ILOG_ABROOT | XFS_ILOG_AEXT);
-		goto out;
-	}
+STATIC void
+xfs_inode_item_format_attr_fork(
+	struct xfs_inode_log_item *iip,
+	struct xfs_inode_log_format *ilf,
+	struct xfs_log_vec	*lv,
+	struct xfs_log_iovec	**vecp)
+{
+	struct xfs_inode	*ip = iip->ili_inode;
+	size_t			data_bytes;
 
 	switch (ip->i_d.di_aformat) {
 	case XFS_DINODE_FMT_EXTENTS:
@@ -360,30 +262,22 @@ xfs_inode_item_format(
 		if ((iip->ili_fields & XFS_ILOG_AEXT) &&
 		    ip->i_d.di_anextents > 0 &&
 		    ip->i_afp->if_bytes > 0) {
+			struct xfs_bmbt_rec *p;
+
 			ASSERT(ip->i_afp->if_bytes / sizeof(xfs_bmbt_rec_t) ==
 				ip->i_d.di_anextents);
 			ASSERT(ip->i_afp->if_u1.if_extents != NULL);
-#ifdef XFS_NATIVE_HOST
-			/*
-			 * There are not delayed allocation extents
-			 * for attributes, so just point at the array.
-			 */
-			vecp->i_addr = ip->i_afp->if_u1.if_extents;
-			vecp->i_len = ip->i_afp->if_bytes;
-			vecp->i_type = XLOG_REG_TYPE_IATTR_EXT;
-#else
-			ASSERT(iip->ili_aextents_buf == NULL);
-			xfs_inode_item_format_extents(ip, vecp,
-					XFS_ATTR_FORK, XLOG_REG_TYPE_IATTR_EXT);
-#endif
-			iip->ili_format.ilf_asize = vecp->i_len;
-			vecp++;
-			nvecs++;
+
+			p = xlog_prepare_iovec(lv, vecp, XLOG_REG_TYPE_IATTR_EXT);
+			data_bytes = xfs_iextents_copy(ip, p, XFS_ATTR_FORK);
+			xlog_finish_iovec(lv, *vecp, data_bytes);
+
+			ilf->ilf_asize = data_bytes;
+			ilf->ilf_size++;
 		} else {
 			iip->ili_fields &= ~XFS_ILOG_AEXT;
 		}
 		break;
-
 	case XFS_DINODE_FMT_BTREE:
 		iip->ili_fields &=
 			~(XFS_ILOG_ADATA | XFS_ILOG_AEXT);
@@ -392,61 +286,89 @@ xfs_inode_item_format(
 		    ip->i_afp->if_broot_bytes > 0) {
 			ASSERT(ip->i_afp->if_broot != NULL);
 
-			vecp->i_addr = ip->i_afp->if_broot;
-			vecp->i_len = ip->i_afp->if_broot_bytes;
-			vecp->i_type = XLOG_REG_TYPE_IATTR_BROOT;
-			vecp++;
-			nvecs++;
-			iip->ili_format.ilf_asize = ip->i_afp->if_broot_bytes;
+			xlog_copy_iovec(lv, vecp, XLOG_REG_TYPE_IATTR_BROOT,
+					ip->i_afp->if_broot,
+					ip->i_afp->if_broot_bytes);
+			ilf->ilf_asize = ip->i_afp->if_broot_bytes;
+			ilf->ilf_size++;
 		} else {
 			iip->ili_fields &= ~XFS_ILOG_ABROOT;
 		}
 		break;
-
 	case XFS_DINODE_FMT_LOCAL:
 		iip->ili_fields &=
 			~(XFS_ILOG_AEXT | XFS_ILOG_ABROOT);
 
 		if ((iip->ili_fields & XFS_ILOG_ADATA) &&
 		    ip->i_afp->if_bytes > 0) {
-			ASSERT(ip->i_afp->if_u1.if_data != NULL);
-
-			vecp->i_addr = ip->i_afp->if_u1.if_data;
 			/*
 			 * Round i_bytes up to a word boundary.
 			 * The underlying memory is guaranteed to
 			 * to be there by xfs_idata_realloc().
 			 */
 			data_bytes = roundup(ip->i_afp->if_bytes, 4);
-			ASSERT((ip->i_afp->if_real_bytes == 0) ||
-			       (ip->i_afp->if_real_bytes == data_bytes));
-			vecp->i_len = (int)data_bytes;
-			vecp->i_type = XLOG_REG_TYPE_IATTR_LOCAL;
-			vecp++;
-			nvecs++;
-			iip->ili_format.ilf_asize = (unsigned)data_bytes;
+			ASSERT(ip->i_afp->if_real_bytes == 0 ||
+			       ip->i_afp->if_real_bytes == data_bytes);
+			ASSERT(ip->i_afp->if_u1.if_data != NULL);
+			xlog_copy_iovec(lv, vecp, XLOG_REG_TYPE_IATTR_LOCAL,
+					ip->i_afp->if_u1.if_data,
+					data_bytes);
+			ilf->ilf_asize = (unsigned)data_bytes;
+			ilf->ilf_size++;
 		} else {
 			iip->ili_fields &= ~XFS_ILOG_ADATA;
 		}
 		break;
-
 	default:
 		ASSERT(0);
 		break;
 	}
-
-out:
-	/*
-	 * Now update the log format that goes out to disk from the in-core
-	 * values.  We always write the inode core to make the arithmetic
-	 * games in recovery easier, which isn't a big deal as just about any
-	 * transaction would dirty it anyway.
-	 */
-	iip->ili_format.ilf_fields = XFS_ILOG_CORE |
-		(iip->ili_fields & ~XFS_ILOG_TIMESTAMP);
-	iip->ili_format.ilf_size = nvecs;
 }
 
+/*
+ * 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,
+ * the second with the on-disk inode structure, and a possible third and/or
+ * fourth with the inode data/extents/b-tree root and inode attributes
+ * data/extents/b-tree root.
+ */
+STATIC void
+xfs_inode_item_format(
+	struct xfs_log_item	*lip,
+	struct xfs_log_vec	*lv)
+{
+	struct xfs_inode_log_item *iip = INODE_ITEM(lip);
+	struct xfs_inode	*ip = iip->ili_inode;
+	struct xfs_inode_log_format *ilf;
+	struct xfs_log_iovec	*vecp = NULL;
+
+	ASSERT(ip->i_d.di_version > 1);
+
+	ilf = xlog_prepare_iovec(lv, &vecp, XLOG_REG_TYPE_IFORMAT);
+	ilf->ilf_type = XFS_LI_INODE;
+	ilf->ilf_ino = ip->i_ino;
+	ilf->ilf_blkno = ip->i_imap.im_blkno;
+	ilf->ilf_len = ip->i_imap.im_len;
+	ilf->ilf_boffset = ip->i_imap.im_boffset;
+	ilf->ilf_fields = XFS_ILOG_CORE;
+	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_data_fork(iip, ilf, lv, &vecp);
+	if (XFS_IFORK_Q(ip)) {
+		xfs_inode_item_format_attr_fork(iip, ilf, lv, &vecp);
+	} else {
+		iip->ili_fields &=
+			~(XFS_ILOG_ADATA | XFS_ILOG_ABROOT | XFS_ILOG_AEXT);
+	}
+
+	/* update the format with the exact fields we actually logged */
+	ilf->ilf_fields |= (iip->ili_fields & ~XFS_ILOG_TIMESTAMP);
+}
 
 /*
  * This is called to pin the inode associated with the inode log
@@ -563,27 +485,6 @@ xfs_inode_item_unlock(
 	ASSERT(ip->i_itemp != NULL);
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
-	/*
-	 * If the inode needed a separate buffer with which to log
-	 * its extents, then free it now.
-	 */
-	if (iip->ili_extents_buf != NULL) {
-		ASSERT(ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS);
-		ASSERT(ip->i_d.di_nextents > 0);
-		ASSERT(iip->ili_fields & XFS_ILOG_DEXT);
-		ASSERT(ip->i_df.if_bytes > 0);
-		kmem_free(iip->ili_extents_buf);
-		iip->ili_extents_buf = NULL;
-	}
-	if (iip->ili_aextents_buf != NULL) {
-		ASSERT(ip->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS);
-		ASSERT(ip->i_d.di_anextents > 0);
-		ASSERT(iip->ili_fields & XFS_ILOG_AEXT);
-		ASSERT(ip->i_afp->if_bytes > 0);
-		kmem_free(iip->ili_aextents_buf);
-		iip->ili_aextents_buf = NULL;
-	}
-
 	lock_flags = iip->ili_lock_flags;
 	iip->ili_lock_flags = 0;
 	if (lock_flags)
@@ -670,11 +571,6 @@ xfs_inode_item_init(
 	iip->ili_inode = ip;
 	xfs_log_item_init(mp, &iip->ili_item, XFS_LI_INODE,
 						&xfs_inode_item_ops);
-	iip->ili_format.ilf_type = XFS_LI_INODE;
-	iip->ili_format.ilf_ino = ip->i_ino;
-	iip->ili_format.ilf_blkno = ip->i_imap.im_blkno;
-	iip->ili_format.ilf_len = ip->i_imap.im_len;
-	iip->ili_format.ilf_boffset = ip->i_imap.im_boffset;
 }
 
 /*
@@ -892,5 +788,5 @@ xfs_inode_item_format_convert(
 		in_f->ilf_boffset = in_f64->ilf_boffset;
 		return 0;
 	}
-	return EFSCORRUPTED;
+	return -EFSCORRUPTED;
 }
diff --git a/fs/xfs/xfs_inode_item.h b/fs/xfs/xfs_inode_item.h
index dce4d656768c..488d81254e28 100644
--- a/fs/xfs/xfs_inode_item.h
+++ b/fs/xfs/xfs_inode_item.h
@@ -34,11 +34,6 @@ 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 */
-	struct xfs_bmbt_rec	*ili_extents_buf;  /* array of logged
-						      data exts */
-	struct xfs_bmbt_rec	*ili_aextents_buf; /* array of logged
-						      attr exts */
-	xfs_inode_log_format_t	ili_format;	   /* logged structure */
 } 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 33ad9a77791f..3799695b9249 100644
--- a/fs/xfs/xfs_ioctl.c
+++ b/fs/xfs/xfs_ioctl.c
@@ -112,15 +112,11 @@ xfs_find_handle(
 		memset(&handle.ha_fid, 0, sizeof(handle.ha_fid));
 		hsize = sizeof(xfs_fsid_t);
 	} else {
-		int		lock_mode;
-
-		lock_mode = xfs_ilock_map_shared(ip);
 		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_ino = ip->i_ino;
-		xfs_iunlock_map_shared(ip, lock_mode);
 
 		hsize = XFS_HSIZE(handle);
 	}
@@ -211,7 +207,7 @@ xfs_open_by_handle(
 	struct path		path;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 
 	dentry = xfs_handlereq_to_dentry(parfilp, hreq);
 	if (IS_ERR(dentry))
@@ -220,7 +216,7 @@ xfs_open_by_handle(
 
 	/* Restrict xfs_open_by_handle to directories & regular files. */
 	if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode))) {
-		error = -XFS_ERROR(EPERM);
+		error = -EPERM;
 		goto out_dput;
 	}
 
@@ -232,18 +228,18 @@ xfs_open_by_handle(
 	fmode = OPEN_FMODE(permflag);
 	if ((!(permflag & O_APPEND) || (permflag & O_TRUNC)) &&
 	    (fmode & FMODE_WRITE) && IS_APPEND(inode)) {
-		error = -XFS_ERROR(EPERM);
+		error = -EPERM;
 		goto out_dput;
 	}
 
 	if ((fmode & FMODE_WRITE) && IS_IMMUTABLE(inode)) {
-		error = -XFS_ERROR(EACCES);
+		error = -EACCES;
 		goto out_dput;
 	}
 
 	/* Can't write directories. */
 	if (S_ISDIR(inode->i_mode) && (fmode & FMODE_WRITE)) {
-		error = -XFS_ERROR(EISDIR);
+		error = -EISDIR;
 		goto out_dput;
 	}
 
@@ -275,32 +271,6 @@ xfs_open_by_handle(
 	return error;
 }
 
-/*
- * This is a copy from fs/namei.c:vfs_readlink(), except for removing it's
- * unused first argument.
- */
-STATIC int
-do_readlink(
-	char __user		*buffer,
-	int			buflen,
-	const char		*link)
-{
-        int len;
-
-	len = PTR_ERR(link);
-	if (IS_ERR(link))
-		goto out;
-
-	len = strlen(link);
-	if (len > (unsigned) buflen)
-		len = buflen;
-	if (copy_to_user(buffer, link, len))
-		len = -EFAULT;
- out:
-	return len;
-}
-
-
 int
 xfs_readlink_by_handle(
 	struct file		*parfilp,
@@ -312,7 +282,7 @@ xfs_readlink_by_handle(
 	int			error;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 
 	dentry = xfs_handlereq_to_dentry(parfilp, hreq);
 	if (IS_ERR(dentry))
@@ -320,25 +290,25 @@ xfs_readlink_by_handle(
 
 	/* Restrict this handle operation to symlinks only. */
 	if (!S_ISLNK(dentry->d_inode->i_mode)) {
-		error = -XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_dput;
 	}
 
 	if (copy_from_user(&olen, hreq->ohandlen, sizeof(__u32))) {
-		error = -XFS_ERROR(EFAULT);
+		error = -EFAULT;
 		goto out_dput;
 	}
 
 	link = kmalloc(MAXPATHLEN+1, GFP_KERNEL);
 	if (!link) {
-		error = -XFS_ERROR(ENOMEM);
+		error = -ENOMEM;
 		goto out_dput;
 	}
 
-	error = -xfs_readlink(XFS_I(dentry->d_inode), link);
+	error = xfs_readlink(XFS_I(dentry->d_inode), link);
 	if (error)
 		goto out_kfree;
-	error = do_readlink(hreq->ohandle, olen, link);
+	error = readlink_copy(hreq->ohandle, olen, link);
 	if (error)
 		goto out_kfree;
 
@@ -360,10 +330,10 @@ xfs_set_dmattrs(
 	int		error;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return XFS_ERROR(EPERM);
+		return -EPERM;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	tp = xfs_trans_alloc(mp, XFS_TRANS_SET_DMATTRS);
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ichange, 0, 0);
@@ -394,9 +364,9 @@ xfs_fssetdm_by_handle(
 	struct dentry		*dentry;
 
 	if (!capable(CAP_MKNOD))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 	if (copy_from_user(&dmhreq, arg, sizeof(xfs_fsop_setdm_handlereq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	error = mnt_want_write_file(parfilp);
 	if (error)
@@ -409,16 +379,16 @@ xfs_fssetdm_by_handle(
 	}
 
 	if (IS_IMMUTABLE(dentry->d_inode) || IS_APPEND(dentry->d_inode)) {
-		error = -XFS_ERROR(EPERM);
+		error = -EPERM;
 		goto out;
 	}
 
 	if (copy_from_user(&fsd, dmhreq.data, sizeof(fsd))) {
-		error = -XFS_ERROR(EFAULT);
+		error = -EFAULT;
 		goto out;
 	}
 
-	error = -xfs_set_dmattrs(XFS_I(dentry->d_inode), fsd.fsd_dmevmask,
+	error = xfs_set_dmattrs(XFS_I(dentry->d_inode), fsd.fsd_dmevmask,
 				 fsd.fsd_dmstate);
 
  out:
@@ -439,18 +409,18 @@ xfs_attrlist_by_handle(
 	char			*kbuf;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 	if (copy_from_user(&al_hreq, arg, sizeof(xfs_fsop_attrlist_handlereq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	if (al_hreq.buflen < sizeof(struct attrlist) ||
 	    al_hreq.buflen > XATTR_LIST_MAX)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	/*
 	 * Reject flags, only allow namespaces.
 	 */
 	if (al_hreq.flags & ~(ATTR_ROOT | ATTR_SECURE))
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	dentry = xfs_handlereq_to_dentry(parfilp, &al_hreq.hreq);
 	if (IS_ERR(dentry))
@@ -461,7 +431,7 @@ xfs_attrlist_by_handle(
 		goto out_dput;
 
 	cursor = (attrlist_cursor_kern_t *)&al_hreq.pos;
-	error = -xfs_attr_list(XFS_I(dentry->d_inode), kbuf, al_hreq.buflen,
+	error = xfs_attr_list(XFS_I(dentry->d_inode), kbuf, al_hreq.buflen,
 					al_hreq.flags, cursor);
 	if (error)
 		goto out_kfree;
@@ -485,20 +455,20 @@ xfs_attrmulti_attr_get(
 	__uint32_t		flags)
 {
 	unsigned char		*kbuf;
-	int			error = EFAULT;
+	int			error = -EFAULT;
 
 	if (*len > XATTR_SIZE_MAX)
-		return EINVAL;
+		return -EINVAL;
 	kbuf = kmem_zalloc_large(*len, KM_SLEEP);
 	if (!kbuf)
-		return ENOMEM;
+		return -ENOMEM;
 
 	error = xfs_attr_get(XFS_I(inode), name, kbuf, (int *)len, flags);
 	if (error)
 		goto out_kfree;
 
 	if (copy_to_user(ubuf, kbuf, *len))
-		error = EFAULT;
+		error = -EFAULT;
 
 out_kfree:
 	kmem_free(kbuf);
@@ -514,20 +484,17 @@ xfs_attrmulti_attr_set(
 	__uint32_t		flags)
 {
 	unsigned char		*kbuf;
-	int			error = EFAULT;
 
 	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
-		return EPERM;
+		return -EPERM;
 	if (len > XATTR_SIZE_MAX)
-		return EINVAL;
+		return -EINVAL;
 
 	kbuf = memdup_user(ubuf, len);
 	if (IS_ERR(kbuf))
 		return PTR_ERR(kbuf);
 
-	error = xfs_attr_set(XFS_I(inode), name, kbuf, len, flags);
-
-	return error;
+	return xfs_attr_set(XFS_I(inode), name, kbuf, len, flags);
 }
 
 int
@@ -537,7 +504,7 @@ xfs_attrmulti_attr_remove(
 	__uint32_t		flags)
 {
 	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
-		return EPERM;
+		return -EPERM;
 	return xfs_attr_remove(XFS_I(inode), name, flags);
 }
 
@@ -554,9 +521,9 @@ xfs_attrmulti_by_handle(
 	unsigned char		*attr_name;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 	if (copy_from_user(&am_hreq, arg, sizeof(xfs_fsop_attrmulti_handlereq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	/* overflow check */
 	if (am_hreq.opcount >= INT_MAX / sizeof(xfs_attr_multiop_t))
@@ -566,7 +533,7 @@ xfs_attrmulti_by_handle(
 	if (IS_ERR(dentry))
 		return PTR_ERR(dentry);
 
-	error = E2BIG;
+	error = -E2BIG;
 	size = am_hreq.opcount * sizeof(xfs_attr_multiop_t);
 	if (!size || size > 16 * PAGE_SIZE)
 		goto out_dput;
@@ -577,6 +544,7 @@ xfs_attrmulti_by_handle(
 		goto out_dput;
 	}
 
+	error = -ENOMEM;
 	attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL);
 	if (!attr_name)
 		goto out_kfree_ops;
@@ -617,19 +585,19 @@ xfs_attrmulti_by_handle(
 			mnt_drop_write_file(parfilp);
 			break;
 		default:
-			ops[i].am_error = EINVAL;
+			ops[i].am_error = -EINVAL;
 		}
 	}
 
 	if (copy_to_user(am_hreq.ops, ops, size))
-		error = XFS_ERROR(EFAULT);
+		error = -EFAULT;
 
 	kfree(attr_name);
  out_kfree_ops:
 	kfree(ops);
  out_dput:
 	dput(dentry);
-	return -error;
+	return error;
 }
 
 int
@@ -654,16 +622,16 @@ xfs_ioc_space(
 	 */
 	if (!xfs_sb_version_hasextflgbit(&ip->i_mount->m_sb) &&
 	    !capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 
 	if (inode->i_flags & (S_IMMUTABLE|S_APPEND))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 
 	if (!(filp->f_mode & FMODE_WRITE))
-		return -XFS_ERROR(EBADF);
+		return -EBADF;
 
 	if (!S_ISREG(inode->i_mode))
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	error = mnt_want_write_file(filp);
 	if (error)
@@ -681,7 +649,7 @@ xfs_ioc_space(
 		bf->l_start += XFS_ISIZE(ip);
 		break;
 	default:
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_unlock;
 	}
 
@@ -698,7 +666,7 @@ xfs_ioc_space(
 	case XFS_IOC_UNRESVSP:
 	case XFS_IOC_UNRESVSP64:
 		if (bf->l_len <= 0) {
-			error = XFS_ERROR(EINVAL);
+			error = -EINVAL;
 			goto out_unlock;
 		}
 		break;
@@ -711,7 +679,7 @@ xfs_ioc_space(
 	    bf->l_start > mp->m_super->s_maxbytes ||
 	    bf->l_start + bf->l_len < 0 ||
 	    bf->l_start + bf->l_len >= mp->m_super->s_maxbytes) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_unlock;
 	}
 
@@ -752,7 +720,7 @@ xfs_ioc_space(
 		break;
 	default:
 		ASSERT(0);
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 	}
 
 	if (error)
@@ -768,7 +736,7 @@ xfs_ioc_space(
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
 	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
 
-	if (!(ioflags & IO_INVIS)) {
+	if (!(ioflags & XFS_IO_INVIS)) {
 		ip->i_d.di_mode &= ~S_ISUID;
 		if (ip->i_d.di_mode & S_IXGRP)
 			ip->i_d.di_mode &= ~S_ISGID;
@@ -788,7 +756,7 @@ xfs_ioc_space(
 out_unlock:
 	xfs_iunlock(ip, XFS_IOLOCK_EXCL);
 	mnt_drop_write_file(filp);
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -810,41 +778,41 @@ xfs_ioc_bulkstat(
 		return -EPERM;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+		return -EIO;
 
 	if (copy_from_user(&bulkreq, arg, sizeof(xfs_fsop_bulkreq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	if (copy_from_user(&inlast, bulkreq.lastip, sizeof(__s64)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	if ((count = bulkreq.icount) <= 0)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	if (bulkreq.ubuffer == NULL)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	if (cmd == XFS_IOC_FSINUMBERS)
 		error = xfs_inumbers(mp, &inlast, &count,
 					bulkreq.ubuffer, xfs_inumbers_fmt);
 	else if (cmd == XFS_IOC_FSBULKSTAT_SINGLE)
-		error = xfs_bulkstat_single(mp, &inlast,
-						bulkreq.ubuffer, &done);
+		error = xfs_bulkstat_one(mp, inlast, bulkreq.ubuffer,
+					sizeof(xfs_bstat_t), NULL, &done);
 	else	/* XFS_IOC_FSBULKSTAT */
 		error = xfs_bulkstat(mp, &inlast, &count, xfs_bulkstat_one,
 				     sizeof(xfs_bstat_t), bulkreq.ubuffer,
 				     &done);
 
 	if (error)
-		return -error;
+		return error;
 
 	if (bulkreq.ocount != NULL) {
 		if (copy_to_user(bulkreq.lastip, &inlast,
 						sizeof(xfs_ino_t)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		if (copy_to_user(bulkreq.ocount, &count, sizeof(count)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 	}
 
 	return 0;
@@ -860,7 +828,7 @@ xfs_ioc_fsgeometry_v1(
 
 	error = xfs_fs_geometry(mp, &fsgeo, 3);
 	if (error)
-		return -error;
+		return error;
 
 	/*
 	 * Caller should have passed an argument of type
@@ -868,7 +836,7 @@ xfs_ioc_fsgeometry_v1(
 	 * xfs_fsop_geom_t that xfs_fs_geometry() fills in.
 	 */
 	if (copy_to_user(arg, &fsgeo, sizeof(xfs_fsop_geom_v1_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -882,10 +850,10 @@ xfs_ioc_fsgeometry(
 
 	error = xfs_fs_geometry(mp, &fsgeo, 4);
 	if (error)
-		return -error;
+		return error;
 
 	if (copy_to_user(arg, &fsgeo, sizeof(fsgeo)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -1070,16 +1038,16 @@ xfs_ioctl_setattr(
 	trace_xfs_ioctl_setattr(ip);
 
 	if (mp->m_flags & XFS_MOUNT_RDONLY)
-		return XFS_ERROR(EROFS);
+		return -EROFS;
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	/*
 	 * Disallow 32bit project ids when projid32bit feature is not enabled.
 	 */
 	if ((mask & FSX_PROJID) && (fa->fsx_projid > (__uint16_t)-1) &&
 			!xfs_sb_version_hasprojid32bit(&ip->i_mount->m_sb))
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	/*
 	 * If disk quotas is on, we make sure that the dquots do exist on disk,
@@ -1117,7 +1085,7 @@ xfs_ioctl_setattr(
 	 * CAP_FSETID capability is applicable.
 	 */
 	if (!inode_owner_or_capable(VFS_I(ip))) {
-		code = XFS_ERROR(EPERM);
+		code = -EPERM;
 		goto error_return;
 	}
 
@@ -1128,7 +1096,7 @@ xfs_ioctl_setattr(
 	 */
 	if (mask & FSX_PROJID) {
 		if (current_user_ns() != &init_user_ns) {
-			code = XFS_ERROR(EINVAL);
+			code = -EINVAL;
 			goto error_return;
 		}
 
@@ -1151,7 +1119,7 @@ xfs_ioctl_setattr(
 		if (ip->i_d.di_nextents &&
 		    ((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) !=
 		     fa->fsx_extsize)) {
-			code = XFS_ERROR(EINVAL);	/* EFBIG? */
+			code = -EINVAL;	/* EFBIG? */
 			goto error_return;
 		}
 
@@ -1170,7 +1138,7 @@ xfs_ioctl_setattr(
 
 			extsize_fsb = XFS_B_TO_FSB(mp, fa->fsx_extsize);
 			if (extsize_fsb > MAXEXTLEN) {
-				code = XFS_ERROR(EINVAL);
+				code = -EINVAL;
 				goto error_return;
 			}
 
@@ -1182,13 +1150,13 @@ xfs_ioctl_setattr(
 			} else {
 				size = mp->m_sb.sb_blocksize;
 				if (extsize_fsb > mp->m_sb.sb_agblocks / 2) {
-					code = XFS_ERROR(EINVAL);
+					code = -EINVAL;
 					goto error_return;
 				}
 			}
 
 			if (fa->fsx_extsize % size) {
-				code = XFS_ERROR(EINVAL);
+				code = -EINVAL;
 				goto error_return;
 			}
 		}
@@ -1202,7 +1170,7 @@ xfs_ioctl_setattr(
 		if ((ip->i_d.di_nextents || ip->i_delayed_blks) &&
 		    (XFS_IS_REALTIME_INODE(ip)) !=
 		    (fa->fsx_xflags & XFS_XFLAG_REALTIME)) {
-			code = XFS_ERROR(EINVAL);	/* EFBIG? */
+			code = -EINVAL;	/* EFBIG? */
 			goto error_return;
 		}
 
@@ -1213,7 +1181,7 @@ xfs_ioctl_setattr(
 			if ((mp->m_sb.sb_rblocks == 0) ||
 			    (mp->m_sb.sb_rextsize == 0) ||
 			    (ip->i_d.di_extsize % mp->m_sb.sb_rextsize)) {
-				code = XFS_ERROR(EINVAL);
+				code = -EINVAL;
 				goto error_return;
 			}
 		}
@@ -1227,7 +1195,7 @@ xfs_ioctl_setattr(
 		     (fa->fsx_xflags &
 				(XFS_XFLAG_IMMUTABLE | XFS_XFLAG_APPEND))) &&
 		    !capable(CAP_LINUX_IMMUTABLE)) {
-			code = XFS_ERROR(EPERM);
+			code = -EPERM;
 			goto error_return;
 		}
 	}
@@ -1245,7 +1213,7 @@ xfs_ioctl_setattr(
 		 * cleared upon successful return from chown()
 		 */
 		if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) &&
-		    !inode_capable(VFS_I(ip), CAP_FSETID))
+		    !capable_wrt_inode_uidgid(VFS_I(ip), CAP_FSETID))
 			ip->i_d.di_mode &= ~(S_ISUID|S_ISGID);
 
 		/*
@@ -1257,15 +1225,8 @@ xfs_ioctl_setattr(
 				olddquot = xfs_qm_vop_chown(tp, ip,
 							&ip->i_pdquot, pdqp);
 			}
+			ASSERT(ip->i_d.di_version > 1);
 			xfs_set_projid(ip, fa->fsx_projid);
-
-			/*
-			 * We may have to rev the inode as well as
-			 * the superblock version number since projids didn't
-			 * exist before DINODE_VERSION_2 and SB_VERSION_NLINK.
-			 */
-			if (ip->i_d.di_version == 1)
-				xfs_bump_ino_vers2(tp, ip);
 		}
 
 	}
@@ -1337,7 +1298,7 @@ xfs_ioc_fssetxattr(
 		return error;
 	error = xfs_ioctl_setattr(ip, &fa, mask);
 	mnt_drop_write_file(filp);
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -1382,7 +1343,7 @@ xfs_ioc_setxflags(
 		return error;
 	error = xfs_ioctl_setattr(ip, &fa, mask);
 	mnt_drop_write_file(filp);
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -1392,7 +1353,7 @@ xfs_getbmap_format(void **ap, struct getbmapx *bmv, int *full)
 
 	/* copy only getbmap portion (not getbmapx) */
 	if (copy_to_user(base, bmv, sizeof(struct getbmap)))
-		return XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	*ap += sizeof(struct getbmap);
 	return 0;
@@ -1409,23 +1370,23 @@ xfs_ioc_getbmap(
 	int			error;
 
 	if (copy_from_user(&bmx, arg, sizeof(struct getbmapx)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	if (bmx.bmv_count < 2)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	bmx.bmv_iflags = (cmd == XFS_IOC_GETBMAPA ? BMV_IF_ATTRFORK : 0);
-	if (ioflags & IO_INVIS)
+	if (ioflags & XFS_IO_INVIS)
 		bmx.bmv_iflags |= BMV_IF_NO_DMAPI_READ;
 
 	error = xfs_getbmap(ip, &bmx, xfs_getbmap_format,
 			    (struct getbmap *)arg+1);
 	if (error)
-		return -error;
+		return error;
 
 	/* copy back header - only size of getbmap */
 	if (copy_to_user(arg, &bmx, sizeof(struct getbmap)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -1435,7 +1396,7 @@ xfs_getbmapx_format(void **ap, struct getbmapx *bmv, int *full)
 	struct getbmapx __user	*base = *ap;
 
 	if (copy_to_user(base, bmv, sizeof(struct getbmapx)))
-		return XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	*ap += sizeof(struct getbmapx);
 	return 0;
@@ -1450,22 +1411,22 @@ xfs_ioc_getbmapx(
 	int			error;
 
 	if (copy_from_user(&bmx, arg, sizeof(bmx)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	if (bmx.bmv_count < 2)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	if (bmx.bmv_iflags & (~BMV_IF_VALID))
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	error = xfs_getbmap(ip, &bmx, xfs_getbmapx_format,
 			    (struct getbmapx *)arg+1);
 	if (error)
-		return -error;
+		return error;
 
 	/* copy back header */
 	if (copy_to_user(arg, &bmx, sizeof(struct getbmapx)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	return 0;
 }
@@ -1481,33 +1442,33 @@ xfs_ioc_swapext(
 	/* Pull information for the target fd */
 	f = fdget((int)sxp->sx_fdtarget);
 	if (!f.file) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out;
 	}
 
 	if (!(f.file->f_mode & FMODE_WRITE) ||
 	    !(f.file->f_mode & FMODE_READ) ||
 	    (f.file->f_flags & O_APPEND)) {
-		error = XFS_ERROR(EBADF);
+		error = -EBADF;
 		goto out_put_file;
 	}
 
 	tmp = fdget((int)sxp->sx_fdtmp);
 	if (!tmp.file) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_put_file;
 	}
 
 	if (!(tmp.file->f_mode & FMODE_WRITE) ||
 	    !(tmp.file->f_mode & FMODE_READ) ||
 	    (tmp.file->f_flags & O_APPEND)) {
-		error = XFS_ERROR(EBADF);
+		error = -EBADF;
 		goto out_put_tmp_file;
 	}
 
 	if (IS_SWAPFILE(file_inode(f.file)) ||
 	    IS_SWAPFILE(file_inode(tmp.file))) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_put_tmp_file;
 	}
 
@@ -1515,17 +1476,17 @@ xfs_ioc_swapext(
 	tip = XFS_I(file_inode(tmp.file));
 
 	if (ip->i_mount != tip->i_mount) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_put_tmp_file;
 	}
 
 	if (ip->i_ino == tip->i_ino) {
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_put_tmp_file;
 	}
 
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
-		error = XFS_ERROR(EIO);
+		error = -EIO;
 		goto out_put_tmp_file;
 	}
 
@@ -1559,7 +1520,7 @@ xfs_file_ioctl(
 	int			error;
 
 	if (filp->f_mode & FMODE_NOCMTIME)
-		ioflags |= IO_INVIS;
+		ioflags |= XFS_IO_INVIS;
 
 	trace_xfs_file_ioctl(ip);
 
@@ -1578,7 +1539,7 @@ xfs_file_ioctl(
 		xfs_flock64_t		bf;
 
 		if (copy_from_user(&bf, arg, sizeof(bf)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return xfs_ioc_space(ip, inode, filp, ioflags, cmd, &bf);
 	}
 	case XFS_IOC_DIOINFO: {
@@ -1587,11 +1548,11 @@ xfs_file_ioctl(
 			XFS_IS_REALTIME_INODE(ip) ?
 			mp->m_rtdev_targp : mp->m_ddev_targp;
 
-		da.d_mem = da.d_miniosz = 1 << target->bt_sshift;
+		da.d_mem =  da.d_miniosz = target->bt_logical_sectorsize;
 		da.d_maxiosz = INT_MAX & ~(da.d_miniosz - 1);
 
 		if (copy_to_user(arg, &da, sizeof(da)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return 0;
 	}
 
@@ -1624,7 +1585,7 @@ xfs_file_ioctl(
 		struct fsdmidata	dmi;
 
 		if (copy_from_user(&dmi, arg, sizeof(dmi)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		error = mnt_want_write_file(filp);
 		if (error)
@@ -1633,7 +1594,7 @@ xfs_file_ioctl(
 		error = xfs_set_dmattrs(ip, dmi.fsd_dmevmask,
 				dmi.fsd_dmstate);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 
 	case XFS_IOC_GETBMAP:
@@ -1649,14 +1610,14 @@ xfs_file_ioctl(
 		xfs_fsop_handlereq_t	hreq;
 
 		if (copy_from_user(&hreq, arg, sizeof(hreq)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return xfs_find_handle(cmd, &hreq);
 	}
 	case XFS_IOC_OPEN_BY_HANDLE: {
 		xfs_fsop_handlereq_t	hreq;
 
 		if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return xfs_open_by_handle(filp, &hreq);
 	}
 	case XFS_IOC_FSSETDM_BY_HANDLE:
@@ -1666,7 +1627,7 @@ xfs_file_ioctl(
 		xfs_fsop_handlereq_t	hreq;
 
 		if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return xfs_readlink_by_handle(filp, &hreq);
 	}
 	case XFS_IOC_ATTRLIST_BY_HANDLE:
@@ -1679,13 +1640,13 @@ xfs_file_ioctl(
 		struct xfs_swapext	sxp;
 
 		if (copy_from_user(&sxp, arg, sizeof(xfs_swapext_t)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		error = mnt_want_write_file(filp);
 		if (error)
 			return error;
 		error = xfs_ioc_swapext(&sxp);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 
 	case XFS_IOC_FSCOUNTS: {
@@ -1693,10 +1654,10 @@ xfs_file_ioctl(
 
 		error = xfs_fs_counts(mp, &out);
 		if (error)
-			return -error;
+			return error;
 
 		if (copy_to_user(arg, &out, sizeof(out)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return 0;
 	}
 
@@ -1708,10 +1669,10 @@ xfs_file_ioctl(
 			return -EPERM;
 
 		if (mp->m_flags & XFS_MOUNT_RDONLY)
-			return -XFS_ERROR(EROFS);
+			return -EROFS;
 
 		if (copy_from_user(&inout, arg, sizeof(inout)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		error = mnt_want_write_file(filp);
 		if (error)
@@ -1722,10 +1683,10 @@ xfs_file_ioctl(
 		error = xfs_reserve_blocks(mp, &in, &inout);
 		mnt_drop_write_file(filp);
 		if (error)
-			return -error;
+			return error;
 
 		if (copy_to_user(arg, &inout, sizeof(inout)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return 0;
 	}
 
@@ -1737,10 +1698,10 @@ xfs_file_ioctl(
 
 		error = xfs_reserve_blocks(mp, NULL, &out);
 		if (error)
-			return -error;
+			return error;
 
 		if (copy_to_user(arg, &out, sizeof(out)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		return 0;
 	}
@@ -1749,42 +1710,42 @@ xfs_file_ioctl(
 		xfs_growfs_data_t in;
 
 		if (copy_from_user(&in, arg, sizeof(in)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		error = mnt_want_write_file(filp);
 		if (error)
 			return error;
 		error = xfs_growfs_data(mp, &in);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 
 	case XFS_IOC_FSGROWFSLOG: {
 		xfs_growfs_log_t in;
 
 		if (copy_from_user(&in, arg, sizeof(in)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		error = mnt_want_write_file(filp);
 		if (error)
 			return error;
 		error = xfs_growfs_log(mp, &in);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 
 	case XFS_IOC_FSGROWFSRT: {
 		xfs_growfs_rt_t in;
 
 		if (copy_from_user(&in, arg, sizeof(in)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		error = mnt_want_write_file(filp);
 		if (error)
 			return error;
 		error = xfs_growfs_rt(mp, &in);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 
 	case XFS_IOC_GOINGDOWN: {
@@ -1794,10 +1755,9 @@ xfs_file_ioctl(
 			return -EPERM;
 
 		if (get_user(in, (__uint32_t __user *)arg))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
-		error = xfs_fs_goingdown(mp, in);
-		return -error;
+		return xfs_fs_goingdown(mp, in);
 	}
 
 	case XFS_IOC_ERROR_INJECTION: {
@@ -1807,18 +1767,16 @@ xfs_file_ioctl(
 			return -EPERM;
 
 		if (copy_from_user(&in, arg, sizeof(in)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
-		error = xfs_errortag_add(in.errtag, mp);
-		return -error;
+		return xfs_errortag_add(in.errtag, mp);
 	}
 
 	case XFS_IOC_ERROR_CLEARALL:
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
 
-		error = xfs_errortag_clearall(mp, 1);
-		return -error;
+		return xfs_errortag_clearall(mp, 1);
 
 	case XFS_IOC_FREE_EOFBLOCKS: {
 		struct xfs_fs_eofblocks eofb;
@@ -1828,16 +1786,16 @@ xfs_file_ioctl(
 			return -EPERM;
 
 		if (mp->m_flags & XFS_MOUNT_RDONLY)
-			return -XFS_ERROR(EROFS);
+			return -EROFS;
 
 		if (copy_from_user(&eofb, arg, sizeof(eofb)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		error = xfs_fs_eofblocks_from_user(&eofb, &keofb);
 		if (error)
-			return -error;
+			return error;
 
-		return -xfs_icache_free_eofblocks(mp, &keofb);
+		return xfs_icache_free_eofblocks(mp, &keofb);
 	}
 
 	default:
diff --git a/fs/xfs/xfs_ioctl32.c b/fs/xfs/xfs_ioctl32.c
index a7992f8de9d3..a554646ff141 100644
--- a/fs/xfs/xfs_ioctl32.c
+++ b/fs/xfs/xfs_ioctl32.c
@@ -28,7 +28,6 @@
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
-#include "xfs_vnode.h"
 #include "xfs_inode.h"
 #include "xfs_itable.h"
 #include "xfs_error.h"
@@ -56,7 +55,7 @@ xfs_compat_flock64_copyin(
 	    get_user(bf->l_sysid,	&arg32->l_sysid) ||
 	    get_user(bf->l_pid,		&arg32->l_pid) ||
 	    copy_from_user(bf->l_pad,	&arg32->l_pad,	4*sizeof(u32)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -70,10 +69,10 @@ xfs_compat_ioc_fsgeometry_v1(
 
 	error = xfs_fs_geometry(mp, &fsgeo, 3);
 	if (error)
-		return -error;
+		return error;
 	/* The 32-bit variant simply has some padding at the end */
 	if (copy_to_user(arg32, &fsgeo, sizeof(struct compat_xfs_fsop_geom_v1)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -84,7 +83,7 @@ xfs_compat_growfs_data_copyin(
 {
 	if (get_user(in->newblocks, &arg32->newblocks) ||
 	    get_user(in->imaxpct,   &arg32->imaxpct))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -95,14 +94,14 @@ xfs_compat_growfs_rt_copyin(
 {
 	if (get_user(in->newblocks, &arg32->newblocks) ||
 	    get_user(in->extsize,   &arg32->extsize))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
 STATIC int
 xfs_inumbers_fmt_compat(
 	void			__user *ubuffer,
-	const xfs_inogrp_t	*buffer,
+	const struct xfs_inogrp	*buffer,
 	long			count,
 	long			*written)
 {
@@ -113,7 +112,7 @@ xfs_inumbers_fmt_compat(
 		if (put_user(buffer[i].xi_startino,   &p32[i].xi_startino) ||
 		    put_user(buffer[i].xi_alloccount, &p32[i].xi_alloccount) ||
 		    put_user(buffer[i].xi_allocmask,  &p32[i].xi_allocmask))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 	}
 	*written = count * sizeof(*p32);
 	return 0;
@@ -132,7 +131,7 @@ xfs_ioctl32_bstime_copyin(
 
 	if (get_user(sec32,		&bstime32->tv_sec)	||
 	    get_user(bstime->tv_nsec,	&bstime32->tv_nsec))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	bstime->tv_sec = sec32;
 	return 0;
 }
@@ -164,7 +163,7 @@ xfs_ioctl32_bstat_copyin(
 	    get_user(bstat->bs_dmevmask, &bstat32->bs_dmevmask)	||
 	    get_user(bstat->bs_dmstate,	&bstat32->bs_dmstate)	||
 	    get_user(bstat->bs_aextents, &bstat32->bs_aextents))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -180,7 +179,7 @@ xfs_bstime_store_compat(
 	sec32 = p->tv_sec;
 	if (put_user(sec32, &p32->tv_sec) ||
 	    put_user(p->tv_nsec, &p32->tv_nsec))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	return 0;
 }
 
@@ -195,7 +194,7 @@ xfs_bulkstat_one_fmt_compat(
 	compat_xfs_bstat_t	__user *p32 = ubuffer;
 
 	if (ubsize < sizeof(*p32))
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 
 	if (put_user(buffer->bs_ino,	  &p32->bs_ino)		||
 	    put_user(buffer->bs_mode,	  &p32->bs_mode)	||
@@ -218,7 +217,7 @@ xfs_bulkstat_one_fmt_compat(
 	    put_user(buffer->bs_dmevmask, &p32->bs_dmevmask)	||
 	    put_user(buffer->bs_dmstate,  &p32->bs_dmstate)	||
 	    put_user(buffer->bs_aextents, &p32->bs_aextents))
-		return XFS_ERROR(EFAULT);
+		return -EFAULT;
 	if (ubused)
 		*ubused = sizeof(*p32);
 	return 0;
@@ -256,30 +255,30 @@ xfs_compat_ioc_bulkstat(
 	/* should be called again (unused here, but used in dmapi) */
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+		return -EIO;
 
 	if (get_user(addr, &p32->lastip))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	bulkreq.lastip = compat_ptr(addr);
 	if (get_user(bulkreq.icount, &p32->icount) ||
 	    get_user(addr, &p32->ubuffer))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	bulkreq.ubuffer = compat_ptr(addr);
 	if (get_user(addr, &p32->ocount))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	bulkreq.ocount = compat_ptr(addr);
 
 	if (copy_from_user(&inlast, bulkreq.lastip, sizeof(__s64)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	if ((count = bulkreq.icount) <= 0)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	if (bulkreq.ubuffer == NULL)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	if (cmd == XFS_IOC_FSINUMBERS_32) {
 		error = xfs_inumbers(mp, &inlast, &count,
@@ -294,17 +293,17 @@ xfs_compat_ioc_bulkstat(
 			xfs_bulkstat_one_compat, sizeof(compat_xfs_bstat_t),
 			bulkreq.ubuffer, &done);
 	} else
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 	if (error)
-		return -error;
+		return error;
 
 	if (bulkreq.ocount != NULL) {
 		if (copy_to_user(bulkreq.lastip, &inlast,
 						sizeof(xfs_ino_t)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 
 		if (copy_to_user(bulkreq.ocount, &count, sizeof(count)))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 	}
 
 	return 0;
@@ -318,7 +317,7 @@ xfs_compat_handlereq_copyin(
 	compat_xfs_fsop_handlereq_t	hreq32;
 
 	if (copy_from_user(&hreq32, arg32, sizeof(compat_xfs_fsop_handlereq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	hreq->fd = hreq32.fd;
 	hreq->path = compat_ptr(hreq32.path);
@@ -352,19 +351,19 @@ xfs_compat_attrlist_by_handle(
 	char			*kbuf;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 	if (copy_from_user(&al_hreq, arg,
 			   sizeof(compat_xfs_fsop_attrlist_handlereq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 	if (al_hreq.buflen < sizeof(struct attrlist) ||
 	    al_hreq.buflen > XATTR_LIST_MAX)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	/*
 	 * Reject flags, only allow namespaces.
 	 */
 	if (al_hreq.flags & ~(ATTR_ROOT | ATTR_SECURE))
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	dentry = xfs_compat_handlereq_to_dentry(parfilp, &al_hreq.hreq);
 	if (IS_ERR(dentry))
@@ -376,7 +375,7 @@ xfs_compat_attrlist_by_handle(
 		goto out_dput;
 
 	cursor = (attrlist_cursor_kern_t *)&al_hreq.pos;
-	error = -xfs_attr_list(XFS_I(dentry->d_inode), kbuf, al_hreq.buflen,
+	error = xfs_attr_list(XFS_I(dentry->d_inode), kbuf, al_hreq.buflen,
 					al_hreq.flags, cursor);
 	if (error)
 		goto out_kfree;
@@ -404,10 +403,10 @@ xfs_compat_attrmulti_by_handle(
 	unsigned char				*attr_name;
 
 	if (!capable(CAP_SYS_ADMIN))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 	if (copy_from_user(&am_hreq, arg,
 			   sizeof(compat_xfs_fsop_attrmulti_handlereq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	/* overflow check */
 	if (am_hreq.opcount >= INT_MAX / sizeof(compat_xfs_attr_multiop_t))
@@ -417,17 +416,18 @@ xfs_compat_attrmulti_by_handle(
 	if (IS_ERR(dentry))
 		return PTR_ERR(dentry);
 
-	error = E2BIG;
+	error = -E2BIG;
 	size = am_hreq.opcount * sizeof(compat_xfs_attr_multiop_t);
 	if (!size || size > 16 * PAGE_SIZE)
 		goto out_dput;
 
 	ops = memdup_user(compat_ptr(am_hreq.ops), size);
 	if (IS_ERR(ops)) {
-		error = PTR_ERR(ops);
+		error = -PTR_ERR(ops);
 		goto out_dput;
 	}
 
+	error = -ENOMEM;
 	attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL);
 	if (!attr_name)
 		goto out_kfree_ops;
@@ -469,19 +469,19 @@ xfs_compat_attrmulti_by_handle(
 			mnt_drop_write_file(parfilp);
 			break;
 		default:
-			ops[i].am_error = EINVAL;
+			ops[i].am_error = -EINVAL;
 		}
 	}
 
 	if (copy_to_user(compat_ptr(am_hreq.ops), ops, size))
-		error = XFS_ERROR(EFAULT);
+		error = -EFAULT;
 
 	kfree(attr_name);
  out_kfree_ops:
 	kfree(ops);
  out_dput:
 	dput(dentry);
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -495,26 +495,26 @@ xfs_compat_fssetdm_by_handle(
 	struct dentry		*dentry;
 
 	if (!capable(CAP_MKNOD))
-		return -XFS_ERROR(EPERM);
+		return -EPERM;
 	if (copy_from_user(&dmhreq, arg,
 			   sizeof(compat_xfs_fsop_setdm_handlereq_t)))
-		return -XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	dentry = xfs_compat_handlereq_to_dentry(parfilp, &dmhreq.hreq);
 	if (IS_ERR(dentry))
 		return PTR_ERR(dentry);
 
 	if (IS_IMMUTABLE(dentry->d_inode) || IS_APPEND(dentry->d_inode)) {
-		error = -XFS_ERROR(EPERM);
+		error = -EPERM;
 		goto out;
 	}
 
 	if (copy_from_user(&fsd, compat_ptr(dmhreq.data), sizeof(fsd))) {
-		error = -XFS_ERROR(EFAULT);
+		error = -EFAULT;
 		goto out;
 	}
 
-	error = -xfs_set_dmattrs(XFS_I(dentry->d_inode), fsd.fsd_dmevmask,
+	error = xfs_set_dmattrs(XFS_I(dentry->d_inode), fsd.fsd_dmevmask,
 				 fsd.fsd_dmstate);
 
 out:
@@ -536,7 +536,7 @@ xfs_file_compat_ioctl(
 	int			error;
 
 	if (filp->f_mode & FMODE_NOCMTIME)
-		ioflags |= IO_INVIS;
+		ioflags |= XFS_IO_INVIS;
 
 	trace_xfs_file_compat_ioctl(ip);
 
@@ -587,7 +587,7 @@ xfs_file_compat_ioctl(
 		struct xfs_flock64	bf;
 
 		if (xfs_compat_flock64_copyin(&bf, arg))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		cmd = _NATIVE_IOC(cmd, struct xfs_flock64);
 		return xfs_ioc_space(ip, inode, filp, ioflags, cmd, &bf);
 	}
@@ -597,25 +597,25 @@ xfs_file_compat_ioctl(
 		struct xfs_growfs_data	in;
 
 		if (xfs_compat_growfs_data_copyin(&in, arg))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		error = mnt_want_write_file(filp);
 		if (error)
 			return error;
 		error = xfs_growfs_data(mp, &in);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 	case XFS_IOC_FSGROWFSRT_32: {
 		struct xfs_growfs_rt	in;
 
 		if (xfs_compat_growfs_rt_copyin(&in, arg))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		error = mnt_want_write_file(filp);
 		if (error)
 			return error;
 		error = xfs_growfs_rt(mp, &in);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 #endif
 	/* long changes size, but xfs only copiese out 32 bits */
@@ -632,13 +632,13 @@ xfs_file_compat_ioctl(
 		if (copy_from_user(&sxp, sxu,
 				   offsetof(struct xfs_swapext, sx_stat)) ||
 		    xfs_ioctl32_bstat_copyin(&sxp.sx_stat, &sxu->sx_stat))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		error = mnt_want_write_file(filp);
 		if (error)
 			return error;
 		error = xfs_ioc_swapext(&sxp);
 		mnt_drop_write_file(filp);
-		return -error;
+		return error;
 	}
 	case XFS_IOC_FSBULKSTAT_32:
 	case XFS_IOC_FSBULKSTAT_SINGLE_32:
@@ -650,7 +650,7 @@ xfs_file_compat_ioctl(
 		struct xfs_fsop_handlereq	hreq;
 
 		if (xfs_compat_handlereq_copyin(&hreq, arg))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		cmd = _NATIVE_IOC(cmd, struct xfs_fsop_handlereq);
 		return xfs_find_handle(cmd, &hreq);
 	}
@@ -658,14 +658,14 @@ xfs_file_compat_ioctl(
 		struct xfs_fsop_handlereq	hreq;
 
 		if (xfs_compat_handlereq_copyin(&hreq, arg))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return xfs_open_by_handle(filp, &hreq);
 	}
 	case XFS_IOC_READLINK_BY_HANDLE_32: {
 		struct xfs_fsop_handlereq	hreq;
 
 		if (xfs_compat_handlereq_copyin(&hreq, arg))
-			return -XFS_ERROR(EFAULT);
+			return -EFAULT;
 		return xfs_readlink_by_handle(filp, &hreq);
 	}
 	case XFS_IOC_ATTRLIST_BY_HANDLE_32:
@@ -675,6 +675,6 @@ xfs_file_compat_ioctl(
 	case XFS_IOC_FSSETDM_BY_HANDLE_32:
 		return xfs_compat_fssetdm_by_handle(filp, arg);
 	default:
-		return -XFS_ERROR(ENOIOCTLCMD);
+		return -ENOIOCTLCMD;
 	}
 }
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 22d1cbea283d..e9c47b6f5e5a 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -110,7 +110,7 @@ xfs_alert_fsblock_zero(
 		(unsigned long long)imap->br_startoff,
 		(unsigned long long)imap->br_blockcount,
 		imap->br_state);
-	return EFSCORRUPTED;
+	return -EFSCORRUPTED;
 }
 
 int
@@ -128,7 +128,6 @@ xfs_iomap_write_direct(
 	xfs_fsblock_t	firstfsb;
 	xfs_extlen_t	extsz, temp;
 	int		nimaps;
-	int		bmapi_flag;
 	int		quota_flag;
 	int		rt;
 	xfs_trans_t	*tp;
@@ -139,7 +138,7 @@ xfs_iomap_write_direct(
 
 	error = xfs_qm_dqattach(ip, 0);
 	if (error)
-		return XFS_ERROR(error);
+		return error;
 
 	rt = XFS_IS_REALTIME_INODE(ip);
 	extsz = xfs_get_extsz_hint(ip);
@@ -149,7 +148,7 @@ xfs_iomap_write_direct(
 	if ((offset + count) > XFS_ISIZE(ip)) {
 		error = xfs_iomap_eof_align_last_fsb(mp, ip, extsz, &last_fsb);
 		if (error)
-			return XFS_ERROR(error);
+			return error;
 	} else {
 		if (nmaps && (imap->br_startblock == HOLESTARTBLOCK))
 			last_fsb = MIN(last_fsb, (xfs_fileoff_t)
@@ -189,7 +188,7 @@ xfs_iomap_write_direct(
 	 */
 	if (error) {
 		xfs_trans_cancel(tp, 0);
-		return XFS_ERROR(error);
+		return error;
 	}
 
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
@@ -200,18 +199,15 @@ xfs_iomap_write_direct(
 
 	xfs_trans_ijoin(tp, ip, 0);
 
-	bmapi_flag = 0;
-	if (offset < XFS_ISIZE(ip) || extsz)
-		bmapi_flag |= XFS_BMAPI_PREALLOC;
-
 	/*
 	 * From this point onwards we overwrite the imap pointer that the
 	 * caller gave to us.
 	 */
 	xfs_bmap_init(&free_list, &firstfsb);
 	nimaps = 1;
-	error = xfs_bmapi_write(tp, ip, offset_fsb, count_fsb, bmapi_flag,
-				&firstfsb, 0, imap, &nimaps, &free_list);
+	error = xfs_bmapi_write(tp, ip, offset_fsb, count_fsb,
+				XFS_BMAPI_PREALLOC, &firstfsb, 0,
+				imap, &nimaps, &free_list);
 	if (error)
 		goto out_bmap_cancel;
 
@@ -229,7 +225,7 @@ xfs_iomap_write_direct(
 	 * Copy any maps to caller's array and return any error.
 	 */
 	if (nimaps == 0) {
-		error = XFS_ERROR(ENOSPC);
+		error = -ENOSPC;
 		goto out_unlock;
 	}
 
@@ -401,7 +397,8 @@ xfs_quota_calc_throttle(
 	struct xfs_inode *ip,
 	int type,
 	xfs_fsblock_t *qblocks,
-	int *qshift)
+	int *qshift,
+	int64_t	*qfreesp)
 {
 	int64_t freesp;
 	int shift = 0;
@@ -410,6 +407,7 @@ xfs_quota_calc_throttle(
 	/* over hi wmark, squash the prealloc completely */
 	if (dq->q_res_bcount >= dq->q_prealloc_hi_wmark) {
 		*qblocks = 0;
+		*qfreesp = 0;
 		return;
 	}
 
@@ -422,6 +420,9 @@ xfs_quota_calc_throttle(
 			shift += 2;
 	}
 
+	if (freesp < *qfreesp)
+		*qfreesp = freesp;
+
 	/* only overwrite the throttle values if we are more aggressive */
 	if ((freesp >> shift) < (*qblocks >> *qshift)) {
 		*qblocks = freesp;
@@ -480,15 +481,18 @@ xfs_iomap_prealloc_size(
 	}
 
 	/*
-	 * Check each quota to cap the prealloc size and provide a shift
-	 * value to throttle with.
+	 * Check each quota to cap the prealloc size, provide a shift value to
+	 * throttle with and adjust amount of available space.
 	 */
 	if (xfs_quota_need_throttle(ip, XFS_DQ_USER, alloc_blocks))
-		xfs_quota_calc_throttle(ip, XFS_DQ_USER, &qblocks, &qshift);
+		xfs_quota_calc_throttle(ip, XFS_DQ_USER, &qblocks, &qshift,
+					&freesp);
 	if (xfs_quota_need_throttle(ip, XFS_DQ_GROUP, alloc_blocks))
-		xfs_quota_calc_throttle(ip, XFS_DQ_GROUP, &qblocks, &qshift);
+		xfs_quota_calc_throttle(ip, XFS_DQ_GROUP, &qblocks, &qshift,
+					&freesp);
 	if (xfs_quota_need_throttle(ip, XFS_DQ_PROJ, alloc_blocks))
-		xfs_quota_calc_throttle(ip, XFS_DQ_PROJ, &qblocks, &qshift);
+		xfs_quota_calc_throttle(ip, XFS_DQ_PROJ, &qblocks, &qshift,
+					&freesp);
 
 	/*
 	 * The final prealloc size is set to the minimum of free space available
@@ -556,7 +560,7 @@ xfs_iomap_write_delay(
 	 */
 	error = xfs_qm_dqattach_locked(ip, 0);
 	if (error)
-		return XFS_ERROR(error);
+		return error;
 
 	extsz = xfs_get_extsz_hint(ip);
 	offset_fsb = XFS_B_TO_FSBT(mp, offset);
@@ -600,11 +604,11 @@ retry:
 				imap, &nimaps, XFS_BMAPI_ENTIRE);
 	switch (error) {
 	case 0:
-	case ENOSPC:
-	case EDQUOT:
+	case -ENOSPC:
+	case -EDQUOT:
 		break;
 	default:
-		return XFS_ERROR(error);
+		return error;
 	}
 
 	/*
@@ -618,7 +622,7 @@ retry:
 			error = 0;
 			goto retry;
 		}
-		return XFS_ERROR(error ? error : ENOSPC);
+		return error ? error : -ENOSPC;
 	}
 
 	if (!(imap[0].br_startblock || XFS_IS_REALTIME_INODE(ip)))
@@ -667,7 +671,7 @@ xfs_iomap_write_allocate(
 	 */
 	error = xfs_qm_dqattach(ip, 0);
 	if (error)
-		return XFS_ERROR(error);
+		return error;
 
 	offset_fsb = XFS_B_TO_FSBT(mp, offset);
 	count_fsb = imap->br_blockcount;
@@ -694,7 +698,7 @@ xfs_iomap_write_allocate(
 						  nres, 0);
 			if (error) {
 				xfs_trans_cancel(tp, 0);
-				return XFS_ERROR(error);
+				return error;
 			}
 			xfs_ilock(ip, XFS_ILOCK_EXCL);
 			xfs_trans_ijoin(tp, ip, 0);
@@ -734,7 +738,7 @@ xfs_iomap_write_allocate(
 			 */
 			nimaps = 1;
 			end_fsb = XFS_B_TO_FSB(mp, XFS_ISIZE(ip));
-			error = xfs_bmap_last_offset(NULL, ip, &last_block,
+			error = xfs_bmap_last_offset(ip, &last_block,
 							XFS_DATA_FORK);
 			if (error)
 				goto trans_cancel;
@@ -743,7 +747,7 @@ xfs_iomap_write_allocate(
 			if ((map_start_fsb + count_fsb) > last_block) {
 				count_fsb = last_block - map_start_fsb;
 				if (count_fsb == 0) {
-					error = EAGAIN;
+					error = -EAGAIN;
 					goto trans_cancel;
 				}
 			}
@@ -753,8 +757,7 @@ xfs_iomap_write_allocate(
 			 * pointer that the caller gave to us.
 			 */
 			error = xfs_bmapi_write(tp, ip, map_start_fsb,
-						count_fsb,
-						XFS_BMAPI_STACK_SWITCH,
+						count_fsb, 0,
 						&first_block, 1,
 						imap, &nimaps, &free_list);
 			if (error)
@@ -798,7 +801,7 @@ trans_cancel:
 	xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
 error0:
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	return XFS_ERROR(error);
+	return error;
 }
 
 int
@@ -858,7 +861,7 @@ xfs_iomap_write_unwritten(
 					  resblks, 0);
 		if (error) {
 			xfs_trans_cancel(tp, 0);
-			return XFS_ERROR(error);
+			return error;
 		}
 
 		xfs_ilock(ip, XFS_ILOCK_EXCL);
@@ -897,7 +900,7 @@ xfs_iomap_write_unwritten(
 		error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
 		xfs_iunlock(ip, XFS_ILOCK_EXCL);
 		if (error)
-			return XFS_ERROR(error);
+			return error;
 
 		if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip)))
 			return xfs_alert_fsblock_zero(ip, &imap);
@@ -920,5 +923,5 @@ error_on_bmapi_transaction:
 	xfs_bmap_cancel(&free_list);
 	xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT));
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	return XFS_ERROR(error);
+	return error;
 }
diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c
index 27e0e544e963..72129493e9d3 100644
--- a/fs/xfs/xfs_iops.c
+++ b/fs/xfs/xfs_iops.c
@@ -39,6 +39,7 @@
 #include "xfs_da_btree.h"
 #include "xfs_dir2_priv.h"
 #include "xfs_dinode.h"
+#include "xfs_trans_space.h"
 
 #include <linux/capability.h>
 #include <linux/xattr.h>
@@ -48,6 +49,18 @@
 #include <linux/fiemap.h>
 #include <linux/slab.h>
 
+/*
+ * Directories have different lock order w.r.t. mmap_sem compared to regular
+ * files. This is due to readdir potentially triggering page faults on a user
+ * buffer inside filldir(), and this happens with the ilock on the directory
+ * held. For regular files, the lock order is the other way around - the
+ * mmap_sem is taken during the page fault, and then we lock the ilock to do
+ * block mapping. Hence we need a different class for the directory ilock so
+ * that lockdep can tell them apart.
+ */
+static struct lock_class_key xfs_nondir_ilock_class;
+static struct lock_class_key xfs_dir_ilock_class;
+
 static int
 xfs_initxattrs(
 	struct inode		*inode,
@@ -60,7 +73,7 @@ xfs_initxattrs(
 
 	for (xattr = xattr_array; xattr->name != NULL; xattr++) {
 		error = xfs_attr_set(ip, xattr->name, xattr->value,
-				     xattr->value_len, ATTR_SECURE);
+				      xattr->value_len, ATTR_SECURE);
 		if (error < 0)
 			break;
 	}
@@ -81,7 +94,7 @@ xfs_init_security(
 	const struct qstr *qstr)
 {
 	return security_inode_init_security(inode, dir, qstr,
-					    &xfs_initxattrs, NULL);
+					     &xfs_initxattrs, NULL);
 }
 
 static void
@@ -111,19 +124,19 @@ xfs_cleanup_inode(
 	xfs_dentry_to_name(&teardown, dentry, 0);
 
 	xfs_remove(XFS_I(dir), &teardown, XFS_I(inode));
-	iput(inode);
 }
 
 STATIC int
-xfs_vn_mknod(
+xfs_generic_create(
 	struct inode	*dir,
 	struct dentry	*dentry,
 	umode_t		mode,
-	dev_t		rdev)
+	dev_t		rdev,
+	bool		tmpfile)	/* unnamed file */
 {
 	struct inode	*inode;
 	struct xfs_inode *ip = NULL;
-	struct posix_acl *default_acl = NULL;
+	struct posix_acl *default_acl, *acl;
 	struct xfs_name	name;
 	int		error;
 
@@ -139,17 +152,16 @@ xfs_vn_mknod(
 		rdev = 0;
 	}
 
-	if (IS_POSIXACL(dir)) {
-		default_acl = xfs_get_acl(dir, ACL_TYPE_DEFAULT);
-		if (IS_ERR(default_acl))
-			return PTR_ERR(default_acl);
+	error = posix_acl_create(dir, &mode, &default_acl, &acl);
+	if (error)
+		return error;
 
-		if (!default_acl)
-			mode &= ~current_umask();
+	if (!tmpfile) {
+		xfs_dentry_to_name(&name, dentry, mode);
+		error = xfs_create(XFS_I(dir), &name, mode, rdev, &ip);
+	} else {
+		error = xfs_create_tmpfile(XFS_I(dir), dentry, mode, &ip);
 	}
-
-	xfs_dentry_to_name(&name, dentry, mode);
-	error = xfs_create(XFS_I(dir), &name, mode, rdev, &ip);
 	if (unlikely(error))
 		goto out_free_acl;
 
@@ -159,22 +171,46 @@ xfs_vn_mknod(
 	if (unlikely(error))
 		goto out_cleanup_inode;
 
+#ifdef CONFIG_XFS_POSIX_ACL
 	if (default_acl) {
-		error = -xfs_inherit_acl(inode, default_acl);
-		default_acl = NULL;
-		if (unlikely(error))
+		error = xfs_set_acl(inode, default_acl, ACL_TYPE_DEFAULT);
+		if (error)
+			goto out_cleanup_inode;
+	}
+	if (acl) {
+		error = xfs_set_acl(inode, acl, ACL_TYPE_ACCESS);
+		if (error)
 			goto out_cleanup_inode;
 	}
+#endif
 
+	if (tmpfile)
+		d_tmpfile(dentry, inode);
+	else
+		d_instantiate(dentry, inode);
 
-	d_instantiate(dentry, inode);
-	return -error;
+ out_free_acl:
+	if (default_acl)
+		posix_acl_release(default_acl);
+	if (acl)
+		posix_acl_release(acl);
+	return error;
 
  out_cleanup_inode:
-	xfs_cleanup_inode(dir, inode, dentry);
- out_free_acl:
-	posix_acl_release(default_acl);
-	return -error;
+	if (!tmpfile)
+		xfs_cleanup_inode(dir, inode, dentry);
+	iput(inode);
+	goto out_free_acl;
+}
+
+STATIC int
+xfs_vn_mknod(
+	struct inode	*dir,
+	struct dentry	*dentry,
+	umode_t		mode,
+	dev_t		rdev)
+{
+	return xfs_generic_create(dir, dentry, mode, rdev, false);
 }
 
 STATIC int
@@ -212,8 +248,8 @@ xfs_vn_lookup(
 	xfs_dentry_to_name(&name, dentry, 0);
 	error = xfs_lookup(XFS_I(dir), &name, &cip, NULL);
 	if (unlikely(error)) {
-		if (unlikely(error != ENOENT))
-			return ERR_PTR(-error);
+		if (unlikely(error != -ENOENT))
+			return ERR_PTR(error);
 		d_add(dentry, NULL);
 		return NULL;
 	}
@@ -239,8 +275,8 @@ xfs_vn_ci_lookup(
 	xfs_dentry_to_name(&xname, dentry, 0);
 	error = xfs_lookup(XFS_I(dir), &xname, &ip, &ci_name);
 	if (unlikely(error)) {
-		if (unlikely(error != ENOENT))
-			return ERR_PTR(-error);
+		if (unlikely(error != -ENOENT))
+			return ERR_PTR(error);
 		/*
 		 * call d_add(dentry, NULL) here when d_drop_negative_children
 		 * is called in xfs_vn_mknod (ie. allow negative dentries
@@ -275,7 +311,7 @@ xfs_vn_link(
 
 	error = xfs_link(XFS_I(dir), XFS_I(inode), &name);
 	if (unlikely(error))
-		return -error;
+		return error;
 
 	ihold(inode);
 	d_instantiate(dentry, inode);
@@ -292,7 +328,7 @@ xfs_vn_unlink(
 
 	xfs_dentry_to_name(&name, dentry, 0);
 
-	error = -xfs_remove(XFS_I(dir), &name, XFS_I(dentry->d_inode));
+	error = xfs_remove(XFS_I(dir), &name, XFS_I(dentry->d_inode));
 	if (error)
 		return error;
 
@@ -337,8 +373,9 @@ xfs_vn_symlink(
 
  out_cleanup_inode:
 	xfs_cleanup_inode(dir, inode, dentry);
+	iput(inode);
  out:
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -355,8 +392,8 @@ xfs_vn_rename(
 	xfs_dentry_to_name(&oname, odentry, 0);
 	xfs_dentry_to_name(&nname, ndentry, odentry->d_inode->i_mode);
 
-	return -xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode),
-			   XFS_I(ndir), &nname, new_inode ?
+	return xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode),
+			  XFS_I(ndir), &nname, new_inode ?
 						XFS_I(new_inode) : NULL);
 }
 
@@ -377,7 +414,7 @@ xfs_vn_follow_link(
 	if (!link)
 		goto out_err;
 
-	error = -xfs_readlink(XFS_I(dentry->d_inode), link);
+	error = xfs_readlink(XFS_I(dentry->d_inode), link);
 	if (unlikely(error))
 		goto out_kfree;
 
@@ -391,18 +428,6 @@ xfs_vn_follow_link(
 	return NULL;
 }
 
-STATIC void
-xfs_vn_put_link(
-	struct dentry	*dentry,
-	struct nameidata *nd,
-	void		*p)
-{
-	char		*s = nd_get_link(nd);
-
-	if (!IS_ERR(s))
-		kfree(s);
-}
-
 STATIC int
 xfs_vn_getattr(
 	struct vfsmount		*mnt,
@@ -416,7 +441,7 @@ xfs_vn_getattr(
 	trace_xfs_getattr(ip);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+		return -EIO;
 
 	stat->size = XFS_ISIZE(ip);
 	stat->dev = inode->i_sb->s_dev;
@@ -459,14 +484,12 @@ xfs_vn_getattr(
 
 static void
 xfs_setattr_mode(
-	struct xfs_trans	*tp,
 	struct xfs_inode	*ip,
 	struct iattr		*iattr)
 {
-	struct inode	*inode = VFS_I(ip);
-	umode_t		mode = iattr->ia_mode;
+	struct inode		*inode = VFS_I(ip);
+	umode_t			mode = iattr->ia_mode;
 
-	ASSERT(tp);
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
 	ip->i_d.di_mode &= S_IFMT;
@@ -476,6 +499,32 @@ xfs_setattr_mode(
 	inode->i_mode |= mode & ~S_IFMT;
 }
 
+static void
+xfs_setattr_time(
+	struct xfs_inode	*ip,
+	struct iattr		*iattr)
+{
+	struct inode		*inode = VFS_I(ip);
+
+	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
+
+	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) {
+		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) {
+		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
 xfs_setattr_nonsize(
 	struct xfs_inode	*ip,
@@ -497,14 +546,14 @@ xfs_setattr_nonsize(
 	/* If acls are being inherited, we already have this checked */
 	if (!(flags & XFS_ATTR_NOACL)) {
 		if (mp->m_flags & XFS_MOUNT_RDONLY)
-			return XFS_ERROR(EROFS);
+			return -EROFS;
 
 		if (XFS_FORCED_SHUTDOWN(mp))
-			return XFS_ERROR(EIO);
+			return -EIO;
 
-		error = -inode_change_ok(inode, iattr);
+		error = inode_change_ok(inode, iattr);
 		if (error)
-			return XFS_ERROR(error);
+			return error;
 	}
 
 	ASSERT((mask & ATTR_SIZE) == 0);
@@ -618,7 +667,8 @@ xfs_setattr_nonsize(
 		}
 		if (!gid_eq(igid, gid)) {
 			if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_GQUOTA_ON(mp)) {
-				ASSERT(!XFS_IS_PQUOTA_ON(mp));
+				ASSERT(xfs_sb_version_has_pquotino(&mp->m_sb) ||
+				       !XFS_IS_PQUOTA_ON(mp));
 				ASSERT(mask & ATTR_GID);
 				ASSERT(gdqp);
 				olddquot2 = xfs_qm_vop_chown(tp, ip,
@@ -629,30 +679,10 @@ xfs_setattr_nonsize(
 		}
 	}
 
-	/*
-	 * Change file access modes.
-	 */
 	if (mask & ATTR_MODE)
-		xfs_setattr_mode(tp, ip, iattr);
-
-	/*
-	 * Change file access or modified times.
-	 */
-	if (mask & 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 (mask & 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 (mask & 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;
-	}
+		xfs_setattr_mode(ip, iattr);
+	if (mask & (ATTR_ATIME|ATTR_CTIME|ATTR_MTIME))
+		xfs_setattr_time(ip, iattr);
 
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 
@@ -673,7 +703,7 @@ xfs_setattr_nonsize(
 	xfs_qm_dqrele(gdqp);
 
 	if (error)
-		return XFS_ERROR(error);
+		return error;
 
 	/*
 	 * XXX(hch): Updating the ACL entries is not atomic vs the i_mode
@@ -683,9 +713,9 @@ xfs_setattr_nonsize(
 	 * 	     Posix ACL code seems to care about this issue either.
 	 */
 	if ((mask & ATTR_MODE) && !(flags & XFS_ATTR_NOACL)) {
-		error = -xfs_acl_chmod(inode);
+		error = posix_acl_chmod(inode, inode->i_mode);
 		if (error)
-			return XFS_ERROR(error);
+			return error;
 	}
 
 	return 0;
@@ -709,7 +739,6 @@ xfs_setattr_size(
 {
 	struct xfs_mount	*mp = ip->i_mount;
 	struct inode		*inode = VFS_I(ip);
-	int			mask = iattr->ia_valid;
 	xfs_off_t		oldsize, newsize;
 	struct xfs_trans	*tp;
 	int			error;
@@ -719,19 +748,19 @@ xfs_setattr_size(
 	trace_xfs_setattr(ip);
 
 	if (mp->m_flags & XFS_MOUNT_RDONLY)
-		return XFS_ERROR(EROFS);
+		return -EROFS;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
-	error = -inode_change_ok(inode, iattr);
+	error = inode_change_ok(inode, iattr);
 	if (error)
-		return XFS_ERROR(error);
+		return error;
 
 	ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
 	ASSERT(S_ISREG(ip->i_d.di_mode));
-	ASSERT((mask & (ATTR_UID|ATTR_GID|ATTR_ATIME|ATTR_ATIME_SET|
-			ATTR_MTIME_SET|ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0);
+	ASSERT((iattr->ia_valid & (ATTR_UID|ATTR_GID|ATTR_ATIME|ATTR_ATIME_SET|
+		ATTR_MTIME_SET|ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0);
 
 	oldsize = inode->i_size;
 	newsize = iattr->ia_size;
@@ -740,7 +769,7 @@ xfs_setattr_size(
 	 * Short circuit the truncate case for zero length files.
 	 */
 	if (newsize == 0 && oldsize == 0 && ip->i_d.di_nextents == 0) {
-		if (!(mask & (ATTR_CTIME|ATTR_MTIME)))
+		if (!(iattr->ia_valid & (ATTR_CTIME|ATTR_MTIME)))
 			return 0;
 
 		/*
@@ -789,7 +818,7 @@ xfs_setattr_size(
 	 * care about here.
 	 */
 	if (oldsize != ip->i_d.di_size && newsize > ip->i_d.di_size) {
-		error = -filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
+		error = filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
 						      ip->i_d.di_size, newsize);
 		if (error)
 			return error;
@@ -800,22 +829,34 @@ xfs_setattr_size(
 	 */
 	inode_dio_wait(inode);
 
-	error = -block_truncate_page(inode->i_mapping, newsize, xfs_get_blocks);
+	/*
+	 * Do all the page cache truncate work outside the transaction context
+	 * as the "lock" order is page lock->log space reservation.  i.e.
+	 * locking pages inside the transaction can ABBA deadlock with
+	 * writeback. We have to do the VFS inode size update before we truncate
+	 * the pagecache, however, to avoid racing with page faults beyond the
+	 * new EOF they are not serialised against truncate operations except by
+	 * page locks and size updates.
+	 *
+	 * Hence we are in a situation where a truncate can fail with ENOMEM
+	 * from xfs_trans_reserve(), but having already truncated the in-memory
+	 * version of the file (i.e. made user visible changes). There's not
+	 * much we can do about this, except to hope that the caller sees ENOMEM
+	 * and retries the truncate operation.
+	 */
+	error = block_truncate_page(inode->i_mapping, newsize, xfs_get_blocks);
 	if (error)
 		return error;
+	truncate_setsize(inode, newsize);
 
 	tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_SIZE);
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0);
 	if (error)
 		goto out_trans_cancel;
 
-	truncate_setsize(inode, newsize);
-
 	commit_flags = XFS_TRANS_RELEASE_LOG_RES;
 	lock_flags |= XFS_ILOCK_EXCL;
-
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
-
 	xfs_trans_ijoin(tp, ip, 0);
 
 	/*
@@ -828,10 +869,11 @@ xfs_setattr_size(
 	 * these flags set.  For all other operations the VFS set these flags
 	 * explicitly if it wants a timestamp update.
 	 */
-	if (newsize != oldsize && (!(mask & (ATTR_CTIME | ATTR_MTIME)))) {
+	if (newsize != oldsize &&
+	    !(iattr->ia_valid & (ATTR_CTIME | ATTR_MTIME))) {
 		iattr->ia_ctime = iattr->ia_mtime =
 			current_fs_time(inode->i_sb);
-		mask |= ATTR_CTIME | ATTR_MTIME;
+		iattr->ia_valid |= ATTR_CTIME | ATTR_MTIME;
 	}
 
 	/*
@@ -867,22 +909,10 @@ xfs_setattr_size(
 		xfs_inode_clear_eofblocks_tag(ip);
 	}
 
-	/*
-	 * Change file access modes.
-	 */
-	if (mask & ATTR_MODE)
-		xfs_setattr_mode(tp, ip, iattr);
-
-	if (mask & 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 (mask & 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;
-	}
+	if (iattr->ia_valid & ATTR_MODE)
+		xfs_setattr_mode(ip, iattr);
+	if (iattr->ia_valid & (ATTR_ATIME|ATTR_CTIME|ATTR_MTIME))
+		xfs_setattr_time(ip, iattr);
 
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 
@@ -920,7 +950,7 @@ xfs_vn_setattr(
 		error = xfs_setattr_nonsize(ip, iattr, 0);
 	}
 
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -940,7 +970,7 @@ xfs_vn_update_time(
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_fsyncts, 0, 0);
 	if (error) {
 		xfs_trans_cancel(tp, 0);
-		return -error;
+		return error;
 	}
 
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
@@ -961,7 +991,7 @@ xfs_vn_update_time(
 	}
 	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_TIMESTAMP);
-	return -xfs_trans_commit(tp, 0);
+	return xfs_trans_commit(tp, 0);
 }
 
 #define XFS_FIEMAP_FLAGS	(FIEMAP_FLAG_SYNC|FIEMAP_FLAG_XATTR)
@@ -1006,7 +1036,7 @@ xfs_fiemap_format(
 		*full = 1;	/* user array now full */
 	}
 
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -1025,12 +1055,12 @@ xfs_vn_fiemap(
 		return error;
 
 	/* Set up bmap header for xfs internal routine */
-	bm.bmv_offset = BTOBB(start);
+	bm.bmv_offset = BTOBBT(start);
 	/* Special case for whole file */
 	if (length == FIEMAP_MAX_OFFSET)
 		bm.bmv_length = -1LL;
 	else
-		bm.bmv_length = BTOBB(length);
+		bm.bmv_length = BTOBB(start + length) - bm.bmv_offset;
 
 	/* We add one because in getbmap world count includes the header */
 	bm.bmv_count = !fieinfo->fi_extents_max ? MAXEXTNUM :
@@ -1045,13 +1075,23 @@ xfs_vn_fiemap(
 
 	error = xfs_getbmap(ip, &bm, xfs_fiemap_format, fieinfo);
 	if (error)
-		return -error;
+		return error;
 
 	return 0;
 }
 
+STATIC int
+xfs_vn_tmpfile(
+	struct inode	*dir,
+	struct dentry	*dentry,
+	umode_t		mode)
+{
+	return xfs_generic_create(dir, dentry, mode, 0, true);
+}
+
 static const struct inode_operations xfs_inode_operations = {
 	.get_acl		= xfs_get_acl,
+	.set_acl		= xfs_set_acl,
 	.getattr		= xfs_vn_getattr,
 	.setattr		= xfs_vn_setattr,
 	.setxattr		= generic_setxattr,
@@ -1079,6 +1119,7 @@ static const struct inode_operations xfs_dir_inode_operations = {
 	.mknod			= xfs_vn_mknod,
 	.rename			= xfs_vn_rename,
 	.get_acl		= xfs_get_acl,
+	.set_acl		= xfs_set_acl,
 	.getattr		= xfs_vn_getattr,
 	.setattr		= xfs_vn_setattr,
 	.setxattr		= generic_setxattr,
@@ -1086,6 +1127,7 @@ static const struct inode_operations xfs_dir_inode_operations = {
 	.removexattr		= generic_removexattr,
 	.listxattr		= xfs_vn_listxattr,
 	.update_time		= xfs_vn_update_time,
+	.tmpfile		= xfs_vn_tmpfile,
 };
 
 static const struct inode_operations xfs_dir_ci_inode_operations = {
@@ -1105,6 +1147,7 @@ static const struct inode_operations xfs_dir_ci_inode_operations = {
 	.mknod			= xfs_vn_mknod,
 	.rename			= xfs_vn_rename,
 	.get_acl		= xfs_get_acl,
+	.set_acl		= xfs_set_acl,
 	.getattr		= xfs_vn_getattr,
 	.setattr		= xfs_vn_setattr,
 	.setxattr		= generic_setxattr,
@@ -1112,13 +1155,13 @@ static const struct inode_operations xfs_dir_ci_inode_operations = {
 	.removexattr		= generic_removexattr,
 	.listxattr		= xfs_vn_listxattr,
 	.update_time		= xfs_vn_update_time,
+	.tmpfile		= xfs_vn_tmpfile,
 };
 
 static const struct inode_operations xfs_symlink_inode_operations = {
 	.readlink		= generic_readlink,
 	.follow_link		= xfs_vn_follow_link,
-	.put_link		= xfs_vn_put_link,
-	.get_acl		= xfs_get_acl,
+	.put_link		= kfree_put_link,
 	.getattr		= xfs_vn_getattr,
 	.setattr		= xfs_vn_setattr,
 	.setxattr		= generic_setxattr,
@@ -1205,6 +1248,7 @@ xfs_setup_inode(
 	xfs_diflags_to_iflags(inode, ip);
 
 	ip->d_ops = ip->i_mount->m_nondir_inode_ops;
+	lockdep_set_class(&ip->i_lock.mr_lock, &xfs_nondir_ilock_class);
 	switch (inode->i_mode & S_IFMT) {
 	case S_IFREG:
 		inode->i_op = &xfs_inode_operations;
@@ -1212,6 +1256,7 @@ xfs_setup_inode(
 		inode->i_mapping->a_ops = &xfs_address_space_operations;
 		break;
 	case S_IFDIR:
+		lockdep_set_class(&ip->i_lock.mr_lock, &xfs_dir_ilock_class);
 		if (xfs_sb_version_hasasciici(&XFS_M(inode->i_sb)->m_sb))
 			inode->i_op = &xfs_dir_ci_inode_operations;
 		else
diff --git a/fs/xfs/xfs_iops.h b/fs/xfs/xfs_iops.h
index d2c5057b5cc4..1c34e4335920 100644
--- a/fs/xfs/xfs_iops.h
+++ b/fs/xfs/xfs_iops.h
@@ -30,7 +30,7 @@ extern void xfs_setup_inode(struct xfs_inode *);
 /*
  * Internal setattr interfaces.
  */
-#define XFS_ATTR_NOACL		0x01	/* Don't call xfs_acl_chmod */
+#define XFS_ATTR_NOACL		0x01	/* Don't call posix_acl_chmod */
 
 extern int xfs_setattr_nonsize(struct xfs_inode *ip, struct iattr *vap,
 			       int flags);
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c
index c237ad15d500..f71be9c68017 100644
--- a/fs/xfs/xfs_itable.c
+++ b/fs/xfs/xfs_itable.c
@@ -67,19 +67,17 @@ xfs_bulkstat_one_int(
 	*stat = BULKSTAT_RV_NOTHING;
 
 	if (!buffer || xfs_internal_inum(mp, ino))
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	buf = kmem_alloc(sizeof(*buf), KM_SLEEP | KM_MAYFAIL);
 	if (!buf)
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 
 	error = xfs_iget(mp, NULL, ino,
 			 (XFS_IGET_DONTCACHE | XFS_IGET_UNTRUSTED),
 			 XFS_ILOCK_SHARED, &ip);
-	if (error) {
-		*stat = BULKSTAT_RV_NOTHING;
+	if (error)
 		goto out_free;
-	}
 
 	ASSERT(ip != NULL);
 	ASSERT(ip->i_imap.im_blkno != 0);
@@ -136,7 +134,6 @@ xfs_bulkstat_one_int(
 	IRELE(ip);
 
 	error = formatter(buffer, ubsize, ubused, buf);
-
 	if (!error)
 		*stat = BULKSTAT_RV_DIDONE;
 
@@ -154,9 +151,9 @@ xfs_bulkstat_one_fmt(
 	const xfs_bstat_t	*buffer)
 {
 	if (ubsize < sizeof(*buffer))
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 	if (copy_to_user(ubuffer, buffer, sizeof(*buffer)))
-		return XFS_ERROR(EFAULT);
+		return -EFAULT;
 	if (ubused)
 		*ubused = sizeof(*buffer);
 	return 0;
@@ -175,9 +172,170 @@ xfs_bulkstat_one(
 				    xfs_bulkstat_one_fmt, ubused, stat);
 }
 
+/*
+ * Loop over all clusters in a chunk for a given incore inode allocation btree
+ * record.  Do a readahead if there are any allocated inodes in that cluster.
+ */
+STATIC void
+xfs_bulkstat_ichunk_ra(
+	struct xfs_mount		*mp,
+	xfs_agnumber_t			agno,
+	struct xfs_inobt_rec_incore	*irec)
+{
+	xfs_agblock_t			agbno;
+	struct blk_plug			plug;
+	int				blks_per_cluster;
+	int				inodes_per_cluster;
+	int				i;	/* inode chunk index */
+
+	agbno = XFS_AGINO_TO_AGBNO(mp, irec->ir_startino);
+	blks_per_cluster = xfs_icluster_size_fsb(mp);
+	inodes_per_cluster = blks_per_cluster << mp->m_sb.sb_inopblog;
+
+	blk_start_plug(&plug);
+	for (i = 0; i < XFS_INODES_PER_CHUNK;
+	     i += inodes_per_cluster, agbno += blks_per_cluster) {
+		if (xfs_inobt_maskn(i, inodes_per_cluster) & ~irec->ir_free) {
+			xfs_btree_reada_bufs(mp, agno, agbno, blks_per_cluster,
+					     &xfs_inode_buf_ops);
+		}
+	}
+	blk_finish_plug(&plug);
+}
+
+/*
+ * Lookup the inode chunk that the given inode lives in and then get the record
+ * if we found the chunk.  If the inode was not the last in the chunk and there
+ * are some left allocated, update the data for the pointed-to record as well as
+ * return the count of grabbed inodes.
+ */
+STATIC int
+xfs_bulkstat_grab_ichunk(
+	struct xfs_btree_cur		*cur,	/* btree cursor */
+	xfs_agino_t			agino,	/* starting inode of chunk */
+	int				*icount,/* return # of inodes grabbed */
+	struct xfs_inobt_rec_incore	*irec)	/* btree record */
+{
+	int				idx;	/* index into inode chunk */
+	int				stat;
+	int				error = 0;
+
+	/* Lookup the inode chunk that this inode lives in */
+	error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, &stat);
+	if (error)
+		return error;
+	if (!stat) {
+		*icount = 0;
+		return error;
+	}
+
+	/* Get the record, should always work */
+	error = xfs_inobt_get_rec(cur, irec, &stat);
+	if (error)
+		return error;
+	XFS_WANT_CORRUPTED_RETURN(stat == 1);
+
+	/* Check if the record contains the inode in request */
+	if (irec->ir_startino + XFS_INODES_PER_CHUNK <= agino)
+		return -EINVAL;
+
+	idx = agino - irec->ir_startino + 1;
+	if (idx < XFS_INODES_PER_CHUNK &&
+	    (xfs_inobt_maskn(idx, XFS_INODES_PER_CHUNK - idx) & ~irec->ir_free)) {
+		int	i;
+
+		/* We got a right chunk with some left inodes allocated at it.
+		 * Grab the chunk record.  Mark all the uninteresting inodes
+		 * free -- because they're before our start point.
+		 */
+		for (i = 0; i < idx; i++) {
+			if (XFS_INOBT_MASK(i) & ~irec->ir_free)
+				irec->ir_freecount++;
+		}
+
+		irec->ir_free |= xfs_inobt_maskn(0, idx);
+		*icount = XFS_INODES_PER_CHUNK - irec->ir_freecount;
+	}
+
+	return 0;
+}
+
 #define XFS_BULKSTAT_UBLEFT(ubleft)	((ubleft) >= statstruct_size)
 
 /*
+ * Process inodes in chunk with a pointer to a formatter function
+ * that will iget the inode and fill in the appropriate structure.
+ */
+int
+xfs_bulkstat_ag_ichunk(
+	struct xfs_mount		*mp,
+	xfs_agnumber_t			agno,
+	struct xfs_inobt_rec_incore	*irbp,
+	bulkstat_one_pf			formatter,
+	size_t				statstruct_size,
+	struct xfs_bulkstat_agichunk	*acp)
+{
+	xfs_ino_t			lastino = acp->ac_lastino;
+	char				__user **ubufp = acp->ac_ubuffer;
+	int				ubleft = acp->ac_ubleft;
+	int				ubelem = acp->ac_ubelem;
+	int				chunkidx, clustidx;
+	int				error = 0;
+	xfs_agino_t			agino;
+
+	for (agino = irbp->ir_startino, chunkidx = clustidx = 0;
+	     XFS_BULKSTAT_UBLEFT(ubleft) &&
+	     irbp->ir_freecount < XFS_INODES_PER_CHUNK;
+	     chunkidx++, clustidx++, agino++) {
+		int		fmterror;	/* bulkstat formatter result */
+		int		ubused;
+		xfs_ino_t	ino = XFS_AGINO_TO_INO(mp, agno, agino);
+
+		ASSERT(chunkidx < XFS_INODES_PER_CHUNK);
+
+		/* Skip if this inode is free */
+		if (XFS_INOBT_MASK(chunkidx) & irbp->ir_free) {
+			lastino = ino;
+			continue;
+		}
+
+		/*
+		 * Count used inodes as free so we can tell when the
+		 * chunk is used up.
+		 */
+		irbp->ir_freecount++;
+
+		/* Get the inode and fill in a single buffer */
+		ubused = statstruct_size;
+		error = formatter(mp, ino, *ubufp, ubleft, &ubused, &fmterror);
+		if (fmterror == BULKSTAT_RV_NOTHING) {
+			if (error && error != -ENOENT && error != -EINVAL) {
+				ubleft = 0;
+				break;
+			}
+			lastino = ino;
+			continue;
+		}
+		if (fmterror == BULKSTAT_RV_GIVEUP) {
+			ubleft = 0;
+			ASSERT(error);
+			break;
+		}
+		if (*ubufp)
+			*ubufp += ubused;
+		ubleft -= ubused;
+		ubelem++;
+		lastino = ino;
+	}
+
+	acp->ac_lastino = lastino;
+	acp->ac_ubleft = ubleft;
+	acp->ac_ubelem = ubelem;
+
+	return error;
+}
+
+/*
  * Return stat information in bulk (by-inode) for the filesystem.
  */
 int					/* error status */
@@ -190,13 +348,10 @@ xfs_bulkstat(
 	char			__user *ubuffer, /* buffer with inode stats */
 	int			*done)	/* 1 if there are more stats to get */
 {
-	xfs_agblock_t		agbno=0;/* allocation group block number */
 	xfs_buf_t		*agbp;	/* agi header buffer */
 	xfs_agi_t		*agi;	/* agi header data */
 	xfs_agino_t		agino;	/* inode # in allocation group */
 	xfs_agnumber_t		agno;	/* allocation group number */
-	int			chunkidx; /* current index into inode chunk */
-	int			clustidx; /* current index into inode cluster */
 	xfs_btree_cur_t		*cur;	/* btree cursor for ialloc btree */
 	int			end_of_ag; /* set if we've seen the ag end */
 	int			error;	/* error code */
@@ -209,9 +364,6 @@ xfs_bulkstat(
 	xfs_inobt_rec_incore_t	*irbuf;	/* start of irec buffer */
 	xfs_inobt_rec_incore_t	*irbufend; /* end of good irec buffer entries */
 	xfs_ino_t		lastino; /* last inode number returned */
-	int			nbcluster; /* # of blocks in a cluster */
-	int			nicluster; /* # of inodes in a cluster */
-	int			nimask;	/* mask for inode clusters */
 	int			nirbuf;	/* size of irbuf */
 	int			rval;	/* return value error code */
 	int			tmp;	/* result value from btree calls */
@@ -219,7 +371,6 @@ xfs_bulkstat(
 	int			ubleft;	/* bytes left in user's buffer */
 	char			__user *ubufp;	/* pointer into user's buffer */
 	int			ubelem;	/* spaces used in user's buffer */
-	int			ubused;	/* bytes used by formatter */
 
 	/*
 	 * Get the last inode value, see if there's nothing to do.
@@ -234,23 +385,16 @@ xfs_bulkstat(
 		*ubcountp = 0;
 		return 0;
 	}
-	if (!ubcountp || *ubcountp <= 0) {
-		return EINVAL;
-	}
+
 	ubcount = *ubcountp; /* statstruct's */
 	ubleft = ubcount * statstruct_size; /* bytes */
 	*ubcountp = ubelem = 0;
 	*done = 0;
 	fmterror = 0;
 	ubufp = ubuffer;
-	nicluster = mp->m_sb.sb_blocksize >= XFS_INODE_CLUSTER_SIZE(mp) ?
-		mp->m_sb.sb_inopblock :
-		(XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_inodelog);
-	nimask = ~(nicluster - 1);
-	nbcluster = nicluster >> mp->m_sb.sb_inopblog;
 	irbuf = kmem_zalloc_greedy(&irbsize, PAGE_SIZE, PAGE_SIZE * 4);
 	if (!irbuf)
-		return ENOMEM;
+		return -ENOMEM;
 
 	nirbuf = irbsize / sizeof(*irbuf);
 
@@ -262,112 +406,50 @@ xfs_bulkstat(
 	while (XFS_BULKSTAT_UBLEFT(ubleft) && agno < mp->m_sb.sb_agcount) {
 		cond_resched();
 		error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp);
-		if (error) {
-			/*
-			 * Skip this allocation group and go to the next one.
-			 */
-			agno++;
-			agino = 0;
-			continue;
-		}
+		if (error)
+			break;
 		agi = XFS_BUF_TO_AGI(agbp);
 		/*
 		 * Allocate and initialize a btree cursor for ialloc btree.
 		 */
-		cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno);
+		cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno,
+					    XFS_BTNUM_INO);
 		irbp = irbuf;
 		irbufend = irbuf + nirbuf;
 		end_of_ag = 0;
-		/*
-		 * If we're returning in the middle of an allocation group,
-		 * we need to get the remainder of the chunk we're in.
-		 */
+		icount = 0;
 		if (agino > 0) {
-			xfs_inobt_rec_incore_t r;
-
 			/*
-			 * Lookup the inode chunk that this inode lives in.
+			 * In the middle of an allocation group, we need to get
+			 * the remainder of the chunk we're in.
 			 */
-			error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE,
-						 &tmp);
-			if (!error &&	/* no I/O error */
-			    tmp &&	/* lookup succeeded */
-					/* got the record, should always work */
-			    !(error = xfs_inobt_get_rec(cur, &r, &i)) &&
-			    i == 1 &&
-					/* this is the right chunk */
-			    agino < r.ir_startino + XFS_INODES_PER_CHUNK &&
-					/* lastino was not last in chunk */
-			    (chunkidx = agino - r.ir_startino + 1) <
-				    XFS_INODES_PER_CHUNK &&
-					/* there are some left allocated */
-			    xfs_inobt_maskn(chunkidx,
-				    XFS_INODES_PER_CHUNK - chunkidx) &
-				    ~r.ir_free) {
-				/*
-				 * Grab the chunk record.  Mark all the
-				 * uninteresting inodes (because they're
-				 * before our start point) free.
-				 */
-				for (i = 0; i < chunkidx; i++) {
-					if (XFS_INOBT_MASK(i) & ~r.ir_free)
-						r.ir_freecount++;
-				}
-				r.ir_free |= xfs_inobt_maskn(0, chunkidx);
+			struct xfs_inobt_rec_incore	r;
+
+			error = xfs_bulkstat_grab_ichunk(cur, agino, &icount, &r);
+			if (error)
+				break;
+			if (icount) {
 				irbp->ir_startino = r.ir_startino;
 				irbp->ir_freecount = r.ir_freecount;
 				irbp->ir_free = r.ir_free;
 				irbp++;
 				agino = r.ir_startino + XFS_INODES_PER_CHUNK;
-				icount = XFS_INODES_PER_CHUNK - r.ir_freecount;
-			} else {
-				/*
-				 * If any of those tests failed, bump the
-				 * inode number (just in case).
-				 */
-				agino++;
-				icount = 0;
 			}
-			/*
-			 * In any case, increment to the next record.
-			 */
-			if (!error)
-				error = xfs_btree_increment(cur, 0, &tmp);
+			/* Increment to the next record */
+			error = xfs_btree_increment(cur, 0, &tmp);
 		} else {
-			/*
-			 * Start of ag.  Lookup the first inode chunk.
-			 */
+			/* Start of ag.  Lookup the first inode chunk */
 			error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &tmp);
-			icount = 0;
 		}
+		if (error)
+			break;
+
 		/*
 		 * Loop through inode btree records in this ag,
 		 * until we run out of inodes or space in the buffer.
 		 */
 		while (irbp < irbufend && icount < ubcount) {
-			xfs_inobt_rec_incore_t r;
-
-			/*
-			 * Loop as long as we're unable to read the
-			 * inode btree.
-			 */
-			while (error) {
-				agino += XFS_INODES_PER_CHUNK;
-				if (XFS_AGINO_TO_AGBNO(mp, agino) >=
-						be32_to_cpu(agi->agi_length))
-					break;
-				error = xfs_inobt_lookup(cur, agino,
-							 XFS_LOOKUP_GE, &tmp);
-				cond_resched();
-			}
-			/*
-			 * If ran off the end of the ag either with an error,
-			 * or the normal way, set end and stop collecting.
-			 */
-			if (error) {
-				end_of_ag = 1;
-				break;
-			}
+			struct xfs_inobt_rec_incore	r;
 
 			error = xfs_inobt_get_rec(cur, &r, &i);
 			if (error || i == 0) {
@@ -380,25 +462,7 @@ xfs_bulkstat(
 			 * Also start read-ahead now for this chunk.
 			 */
 			if (r.ir_freecount < XFS_INODES_PER_CHUNK) {
-				struct blk_plug	plug;
-				/*
-				 * Loop over all clusters in the next chunk.
-				 * Do a readahead if there are any allocated
-				 * inodes in that cluster.
-				 */
-				blk_start_plug(&plug);
-				agbno = XFS_AGINO_TO_AGBNO(mp, r.ir_startino);
-				for (chunkidx = 0;
-				     chunkidx < XFS_INODES_PER_CHUNK;
-				     chunkidx += nicluster,
-				     agbno += nbcluster) {
-					if (xfs_inobt_maskn(chunkidx, nicluster)
-							& ~r.ir_free)
-						xfs_btree_reada_bufs(mp, agno,
-							agbno, nbcluster,
-							&xfs_inode_buf_ops);
-				}
-				blk_finish_plug(&plug);
+				xfs_bulkstat_ichunk_ra(mp, agno, &r);
 				irbp->ir_startino = r.ir_startino;
 				irbp->ir_freecount = r.ir_freecount;
 				irbp->ir_free = r.ir_free;
@@ -425,57 +489,20 @@ xfs_bulkstat(
 		irbufend = irbp;
 		for (irbp = irbuf;
 		     irbp < irbufend && XFS_BULKSTAT_UBLEFT(ubleft); irbp++) {
-			/*
-			 * Now process this chunk of inodes.
-			 */
-			for (agino = irbp->ir_startino, chunkidx = clustidx = 0;
-			     XFS_BULKSTAT_UBLEFT(ubleft) &&
-				irbp->ir_freecount < XFS_INODES_PER_CHUNK;
-			     chunkidx++, clustidx++, agino++) {
-				ASSERT(chunkidx < XFS_INODES_PER_CHUNK);
-
-				ino = XFS_AGINO_TO_INO(mp, agno, agino);
-				/*
-				 * Skip if this inode is free.
-				 */
-				if (XFS_INOBT_MASK(chunkidx) & irbp->ir_free) {
-					lastino = ino;
-					continue;
-				}
-				/*
-				 * Count used inodes as free so we can tell
-				 * when the chunk is used up.
-				 */
-				irbp->ir_freecount++;
-
-				/*
-				 * Get the inode and fill in a single buffer.
-				 */
-				ubused = statstruct_size;
-				error = formatter(mp, ino, ubufp, ubleft,
-						  &ubused, &fmterror);
-				if (fmterror == BULKSTAT_RV_NOTHING) {
-					if (error && error != ENOENT &&
-						error != EINVAL) {
-						ubleft = 0;
-						rval = error;
-						break;
-					}
-					lastino = ino;
-					continue;
-				}
-				if (fmterror == BULKSTAT_RV_GIVEUP) {
-					ubleft = 0;
-					ASSERT(error);
-					rval = error;
-					break;
-				}
-				if (ubufp)
-					ubufp += ubused;
-				ubleft -= ubused;
-				ubelem++;
-				lastino = ino;
-			}
+			struct xfs_bulkstat_agichunk ac;
+
+			ac.ac_lastino = lastino;
+			ac.ac_ubuffer = &ubuffer;
+			ac.ac_ubleft = ubleft;
+			ac.ac_ubelem = ubelem;
+			error = xfs_bulkstat_ag_ichunk(mp, agno, irbp,
+					formatter, statstruct_size, &ac);
+			if (error)
+				rval = error;
+
+			lastino = ac.ac_lastino;
+			ubleft = ac.ac_ubleft;
+			ubelem = ac.ac_ubelem;
 
 			cond_resched();
 		}
@@ -515,58 +542,10 @@ xfs_bulkstat(
 	return rval;
 }
 
-/*
- * Return stat information in bulk (by-inode) for the filesystem.
- * Special case for non-sequential one inode bulkstat.
- */
-int					/* error status */
-xfs_bulkstat_single(
-	xfs_mount_t		*mp,	/* mount point for filesystem */
-	xfs_ino_t		*lastinop, /* inode to return */
-	char			__user *buffer, /* buffer with inode stats */
-	int			*done)	/* 1 if there are more stats to get */
-{
-	int			count;	/* count value for bulkstat call */
-	int			error;	/* return value */
-	xfs_ino_t		ino;	/* filesystem inode number */
-	int			res;	/* result from bs1 */
-
-	/*
-	 * note that requesting valid inode numbers which are not allocated
-	 * to inodes will most likely cause xfs_imap_to_bp to generate warning
-	 * messages about bad magic numbers. This is ok. The fact that
-	 * the inode isn't actually an inode is handled by the
-	 * error check below. Done this way to make the usual case faster
-	 * at the expense of the error case.
-	 */
-
-	ino = *lastinop;
-	error = xfs_bulkstat_one(mp, ino, buffer, sizeof(xfs_bstat_t),
-				 NULL, &res);
-	if (error) {
-		/*
-		 * Special case way failed, do it the "long" way
-		 * to see if that works.
-		 */
-		(*lastinop)--;
-		count = 1;
-		if (xfs_bulkstat(mp, lastinop, &count, xfs_bulkstat_one,
-				sizeof(xfs_bstat_t), buffer, done))
-			return error;
-		if (count == 0 || (xfs_ino_t)*lastinop != ino)
-			return error == EFSCORRUPTED ?
-				XFS_ERROR(EINVAL) : error;
-		else
-			return 0;
-	}
-	*done = 0;
-	return 0;
-}
-
 int
 xfs_inumbers_fmt(
 	void			__user *ubuffer, /* buffer to write to */
-	const xfs_inogrp_t	*buffer,	/* buffer to read from */
+	const struct xfs_inogrp	*buffer,	/* buffer to read from */
 	long			count,		/* # of elements to read */
 	long			*written)	/* # of bytes written */
 {
@@ -581,126 +560,104 @@ xfs_inumbers_fmt(
  */
 int					/* error status */
 xfs_inumbers(
-	xfs_mount_t	*mp,		/* mount point for filesystem */
-	xfs_ino_t	*lastino,	/* last inode returned */
-	int		*count,		/* size of buffer/count returned */
-	void		__user *ubuffer,/* buffer with inode descriptions */
-	inumbers_fmt_pf	formatter)
+	struct xfs_mount	*mp,/* mount point for filesystem */
+	xfs_ino_t		*lastino,/* last inode returned */
+	int			*count,/* size of buffer/count returned */
+	void			__user *ubuffer,/* buffer with inode descriptions */
+	inumbers_fmt_pf		formatter)
 {
-	xfs_buf_t	*agbp;
-	xfs_agino_t	agino;
-	xfs_agnumber_t	agno;
-	int		bcount;
-	xfs_inogrp_t	*buffer;
-	int		bufidx;
-	xfs_btree_cur_t	*cur;
-	int		error;
-	xfs_inobt_rec_incore_t r;
-	int		i;
-	xfs_ino_t	ino;
-	int		left;
-	int		tmp;
-
-	ino = (xfs_ino_t)*lastino;
-	agno = XFS_INO_TO_AGNO(mp, ino);
-	agino = XFS_INO_TO_AGINO(mp, ino);
-	left = *count;
+	xfs_agnumber_t		agno = XFS_INO_TO_AGNO(mp, *lastino);
+	xfs_agino_t		agino = XFS_INO_TO_AGINO(mp, *lastino);
+	struct xfs_btree_cur	*cur = NULL;
+	struct xfs_buf		*agbp = NULL;
+	struct xfs_inogrp	*buffer;
+	int			bcount;
+	int			left = *count;
+	int			bufidx = 0;
+	int			error = 0;
+
 	*count = 0;
+	if (agno >= mp->m_sb.sb_agcount ||
+	    *lastino != XFS_AGINO_TO_INO(mp, agno, agino))
+		return error;
+
 	bcount = MIN(left, (int)(PAGE_SIZE / sizeof(*buffer)));
 	buffer = kmem_alloc(bcount * sizeof(*buffer), KM_SLEEP);
-	error = bufidx = 0;
-	cur = NULL;
-	agbp = NULL;
-	while (left > 0 && agno < mp->m_sb.sb_agcount) {
-		if (agbp == NULL) {
+	do {
+		struct xfs_inobt_rec_incore	r;
+		int				stat;
+
+		if (!agbp) {
 			error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp);
-			if (error) {
-				/*
-				 * If we can't read the AGI of this ag,
-				 * then just skip to the next one.
-				 */
-				ASSERT(cur == NULL);
-				agbp = NULL;
-				agno++;
-				agino = 0;
-				continue;
-			}
-			cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno);
+			if (error)
+				break;
+
+			cur = xfs_inobt_init_cursor(mp, NULL, agbp, agno,
+						    XFS_BTNUM_INO);
 			error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_GE,
-						 &tmp);
-			if (error) {
-				xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
-				cur = NULL;
-				xfs_buf_relse(agbp);
-				agbp = NULL;
-				/*
-				 * Move up the last inode in the current
-				 * chunk.  The lookup_ge will always get
-				 * us the first inode in the next chunk.
-				 */
-				agino += XFS_INODES_PER_CHUNK - 1;
-				continue;
-			}
-		}
-		error = xfs_inobt_get_rec(cur, &r, &i);
-		if (error || i == 0) {
-			xfs_buf_relse(agbp);
-			agbp = NULL;
-			xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
-			cur = NULL;
-			agno++;
-			agino = 0;
-			continue;
+						 &stat);
+			if (error)
+				break;
+			if (!stat)
+				goto next_ag;
 		}
+
+		error = xfs_inobt_get_rec(cur, &r, &stat);
+		if (error)
+			break;
+		if (!stat)
+			goto next_ag;
+
 		agino = r.ir_startino + XFS_INODES_PER_CHUNK - 1;
 		buffer[bufidx].xi_startino =
 			XFS_AGINO_TO_INO(mp, agno, r.ir_startino);
 		buffer[bufidx].xi_alloccount =
 			XFS_INODES_PER_CHUNK - r.ir_freecount;
 		buffer[bufidx].xi_allocmask = ~r.ir_free;
-		bufidx++;
-		left--;
-		if (bufidx == bcount) {
-			long written;
-			if (formatter(ubuffer, buffer, bufidx, &written)) {
-				error = XFS_ERROR(EFAULT);
+		if (++bufidx == bcount) {
+			long	written;
+
+			error = formatter(ubuffer, buffer, bufidx, &written);
+			if (error)
 				break;
-			}
 			ubuffer += written;
 			*count += bufidx;
 			bufidx = 0;
 		}
-		if (left) {
-			error = xfs_btree_increment(cur, 0, &tmp);
-			if (error) {
-				xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
-				cur = NULL;
-				xfs_buf_relse(agbp);
-				agbp = NULL;
-				/*
-				 * The agino value has already been bumped.
-				 * Just try to skip up to it.
-				 */
-				agino += XFS_INODES_PER_CHUNK;
-				continue;
-			}
-		}
-	}
+		if (!--left)
+			break;
+
+		error = xfs_btree_increment(cur, 0, &stat);
+		if (error)
+			break;
+		if (stat)
+			continue;
+
+next_ag:
+		xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
+		cur = NULL;
+		xfs_buf_relse(agbp);
+		agbp = NULL;
+		agino = 0;
+	} while (++agno < mp->m_sb.sb_agcount);
+
 	if (!error) {
 		if (bufidx) {
-			long written;
-			if (formatter(ubuffer, buffer, bufidx, &written))
-				error = XFS_ERROR(EFAULT);
-			else
+			long	written;
+
+			error = formatter(ubuffer, buffer, bufidx, &written);
+			if (!error)
 				*count += bufidx;
 		}
 		*lastino = XFS_AGINO_TO_INO(mp, agno, agino);
 	}
+
 	kmem_free(buffer);
 	if (cur)
 		xfs_btree_del_cursor(cur, (error ? XFS_BTREE_ERROR :
 					   XFS_BTREE_NOERROR));
 	if (agbp)
 		xfs_buf_relse(agbp);
+
 	return error;
 }
diff --git a/fs/xfs/xfs_itable.h b/fs/xfs/xfs_itable.h
index 97295d91d170..aaed08022eb9 100644
--- a/fs/xfs/xfs_itable.h
+++ b/fs/xfs/xfs_itable.h
@@ -30,6 +30,22 @@ typedef int (*bulkstat_one_pf)(struct xfs_mount	*mp,
 			       int		*ubused,
 			       int		*stat);
 
+struct xfs_bulkstat_agichunk {
+	xfs_ino_t	ac_lastino;	/* last inode returned */
+	char		__user **ac_ubuffer;/* pointer into user's buffer */
+	int		ac_ubleft;	/* bytes left in user's buffer */
+	int		ac_ubelem;	/* spaces used in user's buffer */
+};
+
+int
+xfs_bulkstat_ag_ichunk(
+	struct xfs_mount		*mp,
+	xfs_agnumber_t			agno,
+	struct xfs_inobt_rec_incore	*irbp,
+	bulkstat_one_pf			formatter,
+	size_t				statstruct_size,
+	struct xfs_bulkstat_agichunk	*acp);
+
 /*
  * Values for stat return value.
  */
@@ -50,13 +66,6 @@ xfs_bulkstat(
 	char		__user *ubuffer,/* buffer with inode stats */
 	int		*done);		/* 1 if there are more stats to get */
 
-int
-xfs_bulkstat_single(
-	xfs_mount_t		*mp,
-	xfs_ino_t		*lastinop,
-	char			__user *buffer,
-	int			*done);
-
 typedef int (*bulkstat_one_fmt_pf)(  /* used size in bytes or negative error */
 	void			__user *ubuffer, /* buffer to write to */
 	int			ubsize,		 /* remaining user buffer sz */
diff --git a/fs/xfs/xfs_linux.h b/fs/xfs/xfs_linux.h
index f9bb590acc0e..d10dc8f397c9 100644
--- a/fs/xfs/xfs_linux.h
+++ b/fs/xfs/xfs_linux.h
@@ -21,18 +21,6 @@
 #include <linux/types.h>
 
 /*
- * XFS_BIG_BLKNOS needs block layer disk addresses to be 64 bits.
- * XFS_BIG_INUMS requires XFS_BIG_BLKNOS to be set.
- */
-#if defined(CONFIG_LBDAF) || (BITS_PER_LONG == 64)
-# define XFS_BIG_BLKNOS	1
-# define XFS_BIG_INUMS	1
-#else
-# define XFS_BIG_BLKNOS	0
-# define XFS_BIG_INUMS	0
-#endif
-
-/*
  * Kernel specific type declarations for XFS
  */
 typedef signed char		__int8_t;
@@ -113,12 +101,13 @@ typedef __uint64_t __psunsigned_t;
 #include <asm/byteorder.h>
 #include <asm/unaligned.h>
 
-#include "xfs_vnode.h"
+#include "xfs_fs.h"
 #include "xfs_stats.h"
 #include "xfs_sysctl.h"
 #include "xfs_iops.h"
 #include "xfs_aops.h"
 #include "xfs_super.h"
+#include "xfs_cksum.h"
 #include "xfs_buf.h"
 #include "xfs_message.h"
 
@@ -178,6 +167,7 @@ typedef __uint64_t __psunsigned_t;
 #define ENOATTR		ENODATA		/* Attribute not found */
 #define EWRONGFS	EINVAL		/* Mount with wrong filesystem type */
 #define EFSCORRUPTED	EUCLEAN		/* Filesystem is corrupted */
+#define EFSBADCRC	EBADMSG		/* Bad CRC detected */
 
 #define SYNCHRONIZE()	barrier()
 #define __return_address __builtin_return_address(0)
@@ -189,6 +179,17 @@ typedef __uint64_t __psunsigned_t;
 #define MAX(a,b)	(max(a,b))
 #define howmany(x, y)	(((x)+((y)-1))/(y))
 
+/*
+ * XFS wrapper structure for sysfs support. It depends on external data
+ * structures and is embedded in various internal data structures to implement
+ * the XFS sysfs object heirarchy. Define it here for broad access throughout
+ * the codebase.
+ */
+struct xfs_kobj {
+	struct kobject		kobject;
+	struct completion	complete;
+};
+
 /* 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
@@ -329,7 +330,7 @@ static inline __uint64_t roundup_64(__uint64_t x, __uint32_t y)
 {
 	x += y - 1;
 	do_div(x, y);
-	return(x * y);
+	return x * y;
 }
 
 static inline __uint64_t howmany_64(__uint64_t x, __uint32_t y)
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 8497a00e399d..ca4fd5bd8522 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -34,6 +34,7 @@
 #include "xfs_trace.h"
 #include "xfs_fsops.h"
 #include "xfs_cksum.h"
+#include "xfs_sysfs.h"
 
 kmem_zone_t	*xfs_log_ticket_zone;
 
@@ -283,7 +284,7 @@ xlog_grant_head_wait(
 	return 0;
 shutdown:
 	list_del_init(&tic->t_queue);
-	return XFS_ERROR(EIO);
+	return -EIO;
 }
 
 /*
@@ -377,7 +378,7 @@ xfs_log_regrant(
 	int			error = 0;
 
 	if (XLOG_FORCED_SHUTDOWN(log))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	XFS_STATS_INC(xs_try_logspace);
 
@@ -446,7 +447,7 @@ xfs_log_reserve(
 	ASSERT(client == XFS_TRANSACTION || client == XFS_LOG);
 
 	if (XLOG_FORCED_SHUTDOWN(log))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	XFS_STATS_INC(xs_try_logspace);
 
@@ -454,7 +455,7 @@ xfs_log_reserve(
 	tic = xlog_ticket_alloc(log, unit_bytes, cnt, client, permanent,
 				KM_SLEEP | KM_MAYFAIL);
 	if (!tic)
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 
 	tic->t_trans_type = t_type;
 	*ticp = tic;
@@ -590,7 +591,7 @@ xfs_log_release_iclog(
 {
 	if (xlog_state_release_iclog(mp->m_log, iclog)) {
 		xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
-		return EIO;
+		return -EIO;
 	}
 
 	return 0;
@@ -616,17 +617,19 @@ xfs_log_mount(
 	int		error = 0;
 	int		min_logfsbs;
 
-	if (!(mp->m_flags & XFS_MOUNT_NORECOVERY))
-		xfs_notice(mp, "Mounting Filesystem");
-	else {
+	if (!(mp->m_flags & XFS_MOUNT_NORECOVERY)) {
+		xfs_notice(mp, "Mounting V%d Filesystem",
+			   XFS_SB_VERSION_NUM(&mp->m_sb));
+	} else {
 		xfs_notice(mp,
-"Mounting filesystem in no-recovery mode.  Filesystem will be inconsistent.");
+"Mounting V%d filesystem in no-recovery mode. Filesystem will be inconsistent.",
+			   XFS_SB_VERSION_NUM(&mp->m_sb));
 		ASSERT(mp->m_flags & XFS_MOUNT_RDONLY);
 	}
 
 	mp->m_log = xlog_alloc_log(mp, log_target, blk_offset, num_bblks);
 	if (IS_ERR(mp->m_log)) {
-		error = -PTR_ERR(mp->m_log);
+		error = PTR_ERR(mp->m_log);
 		goto out;
 	}
 
@@ -650,18 +653,18 @@ xfs_log_mount(
 		xfs_warn(mp,
 		"Log size %d blocks too small, minimum size is %d blocks",
 			 mp->m_sb.sb_logblocks, min_logfsbs);
-		error = EINVAL;
+		error = -EINVAL;
 	} else if (mp->m_sb.sb_logblocks > XFS_MAX_LOG_BLOCKS) {
 		xfs_warn(mp,
 		"Log size %d blocks too large, maximum size is %lld blocks",
 			 mp->m_sb.sb_logblocks, XFS_MAX_LOG_BLOCKS);
-		error = EINVAL;
+		error = -EINVAL;
 	} else if (XFS_FSB_TO_B(mp, mp->m_sb.sb_logblocks) > XFS_MAX_LOG_BYTES) {
 		xfs_warn(mp,
 		"log size %lld bytes too large, maximum size is %lld bytes",
 			 XFS_FSB_TO_B(mp, mp->m_sb.sb_logblocks),
 			 XFS_MAX_LOG_BYTES);
-		error = EINVAL;
+		error = -EINVAL;
 	}
 	if (error) {
 		if (xfs_sb_version_hascrc(&mp->m_sb)) {
@@ -705,6 +708,11 @@ xfs_log_mount(
 		}
 	}
 
+	error = xfs_sysfs_init(&mp->m_log->l_kobj, &xfs_log_ktype, &mp->m_kobj,
+			       "log");
+	if (error)
+		goto out_destroy_ail;
+
 	/* Normal transactions can now occur */
 	mp->m_log->l_flags &= ~XLOG_ACTIVE_RECOVERY;
 
@@ -945,6 +953,9 @@ xfs_log_unmount(
 	xfs_log_quiesce(mp);
 
 	xfs_trans_ail_destroy(mp);
+
+	xfs_sysfs_del(&mp->m_log->l_kobj);
+
 	xlog_dealloc_log(mp->m_log);
 }
 
@@ -1163,7 +1174,7 @@ xlog_iodone(xfs_buf_t *bp)
 	/*
 	 * Race to shutdown the filesystem if we see an error.
 	 */
-	if (XFS_TEST_ERROR((xfs_buf_geterror(bp)), l->l_mp,
+	if (XFS_TEST_ERROR(bp->b_error, l->l_mp,
 			XFS_ERRTAG_IODONE_IOERR, XFS_RANDOM_IODONE_IOERR)) {
 		xfs_buf_ioerror_alert(bp, __func__);
 		xfs_buf_stale(bp);
@@ -1181,11 +1192,14 @@ xlog_iodone(xfs_buf_t *bp)
 	/* log I/O is always issued ASYNC */
 	ASSERT(XFS_BUF_ISASYNC(bp));
 	xlog_state_done_syncing(iclog, aborted);
+
 	/*
-	 * do not reference the buffer (bp) here as we could race
-	 * with it being freed after writing the unmount record to the
-	 * log.
+	 * drop the buffer lock now that we are done. Nothing references
+	 * the buffer after this, so an unmount waiting on this lock can now
+	 * tear it down safely. As such, it is unsafe to reference the buffer
+	 * (bp) after the unlock as we could race with it being freed.
 	 */
+	xfs_buf_unlock(bp);
 }
 
 /*
@@ -1308,7 +1322,7 @@ xlog_alloc_log(
 	xlog_in_core_t		*iclog, *prev_iclog=NULL;
 	xfs_buf_t		*bp;
 	int			i;
-	int			error = ENOMEM;
+	int			error = -ENOMEM;
 	uint			log2_size = 0;
 
 	log = kmem_zalloc(sizeof(struct xlog), KM_MAYFAIL);
@@ -1335,7 +1349,7 @@ xlog_alloc_log(
 	xlog_grant_head_init(&log->l_reserve_head);
 	xlog_grant_head_init(&log->l_write_head);
 
-	error = EFSCORRUPTED;
+	error = -EFSCORRUPTED;
 	if (xfs_sb_version_hassector(&mp->m_sb)) {
 	        log2_size = mp->m_sb.sb_logsectlog;
 		if (log2_size < BBSHIFT) {
@@ -1364,12 +1378,26 @@ xlog_alloc_log(
 
 	xlog_get_iclog_buffer_size(mp, log);
 
-	error = ENOMEM;
-	bp = xfs_buf_alloc(mp->m_logdev_targp, 0, BTOBB(log->l_iclog_size), 0);
+	/*
+	 * Use a NULL block for the extra log buffer used during splits so that
+	 * it will trigger errors if we ever try to do IO on it without first
+	 * having set it up properly.
+	 */
+	error = -ENOMEM;
+	bp = xfs_buf_alloc(mp->m_logdev_targp, XFS_BUF_DADDR_NULL,
+			   BTOBB(log->l_iclog_size), 0);
 	if (!bp)
 		goto out_free_log;
-	bp->b_iodone = xlog_iodone;
+
+	/*
+	 * The iclogbuf buffer locks are held over IO but we are not going to do
+	 * IO yet.  Hence unlock the buffer so that the log IO path can grab it
+	 * when appropriately.
+	 */
 	ASSERT(xfs_buf_islocked(bp));
+	xfs_buf_unlock(bp);
+
+	bp->b_iodone = xlog_iodone;
 	log->l_xbuf = bp;
 
 	spin_lock_init(&log->l_icloglock);
@@ -1398,6 +1426,9 @@ xlog_alloc_log(
 		if (!bp)
 			goto out_free_iclog;
 
+		ASSERT(xfs_buf_islocked(bp));
+		xfs_buf_unlock(bp);
+
 		bp->b_iodone = xlog_iodone;
 		iclog->ic_bp = bp;
 		iclog->ic_data = bp->b_addr;
@@ -1422,7 +1453,6 @@ xlog_alloc_log(
 		iclog->ic_callback_tail = &(iclog->ic_callback);
 		iclog->ic_datap = (char *)iclog->ic_data + log->l_iclog_hsize;
 
-		ASSERT(xfs_buf_islocked(iclog->ic_bp));
 		init_waitqueue_head(&iclog->ic_force_wait);
 		init_waitqueue_head(&iclog->ic_write_wait);
 
@@ -1448,7 +1478,7 @@ out_free_iclog:
 out_free_log:
 	kmem_free(log);
 out:
-	return ERR_PTR(-error);
+	return ERR_PTR(error);
 }	/* xlog_alloc_log */
 
 
@@ -1631,6 +1661,12 @@ xlog_cksum(
  * we transition the iclogs to IOERROR state *after* flushing all existing
  * iclogs to disk. This is because we don't want anymore new transactions to be
  * started or completed afterwards.
+ *
+ * We lock the iclogbufs here so that we can serialise against IO completion
+ * during unmount. We might be processing a shutdown triggered during unmount,
+ * and that can occur asynchronously to the unmount thread, and hence we need to
+ * ensure that completes before tearing down the iclogbufs. Hence we need to
+ * hold the buffer lock across the log IO to acheive that.
  */
 STATIC int
 xlog_bdstrat(
@@ -1638,14 +1674,16 @@ xlog_bdstrat(
 {
 	struct xlog_in_core	*iclog = bp->b_fspriv;
 
+	xfs_buf_lock(bp);
 	if (iclog->ic_state & XLOG_STATE_IOERROR) {
-		xfs_buf_ioerror(bp, EIO);
+		xfs_buf_ioerror(bp, -EIO);
 		xfs_buf_stale(bp);
 		xfs_buf_ioend(bp, 0);
 		/*
 		 * It would seem logical to return EIO here, but we rely on
 		 * the log state machine to propagate I/O errors instead of
-		 * doing it here.
+		 * doing it here. Similarly, IO completion will unlock the
+		 * buffer, so we don't do it here.
 		 */
 		return 0;
 	}
@@ -1847,14 +1885,28 @@ xlog_dealloc_log(
 	xlog_cil_destroy(log);
 
 	/*
-	 * always need to ensure that the extra buffer does not point to memory
-	 * owned by another log buffer before we free it.
+	 * Cycle all the iclogbuf locks to make sure all log IO completion
+	 * is done before we tear down these buffers.
 	 */
+	iclog = log->l_iclog;
+	for (i = 0; i < log->l_iclog_bufs; i++) {
+		xfs_buf_lock(iclog->ic_bp);
+		xfs_buf_unlock(iclog->ic_bp);
+		iclog = iclog->ic_next;
+	}
+
+	/*
+	 * Always need to ensure that the extra buffer does not point to memory
+	 * owned by another log buffer before we free it. Also, cycle the lock
+	 * first to ensure we've completed IO on it.
+	 */
+	xfs_buf_lock(log->l_xbuf);
+	xfs_buf_unlock(log->l_xbuf);
 	xfs_buf_set_empty(log->l_xbuf, BTOBB(log->l_iclog_size));
 	xfs_buf_free(log->l_xbuf);
 
 	iclog = log->l_iclog;
-	for (i=0; i<log->l_iclog_bufs; i++) {
+	for (i = 0; i < log->l_iclog_bufs; i++) {
 		xfs_buf_free(iclog->ic_bp);
 		next_iclog = iclog->ic_next;
 		kmem_free(iclog);
@@ -2323,7 +2375,7 @@ xlog_write(
 
 			ophdr = xlog_write_setup_ophdr(log, ptr, ticket, flags);
 			if (!ophdr)
-				return XFS_ERROR(EIO);
+				return -EIO;
 
 			xlog_write_adv_cnt(&ptr, &len, &log_offset,
 					   sizeof(struct xlog_op_header));
@@ -2822,7 +2874,7 @@ restart:
 	spin_lock(&log->l_icloglock);
 	if (XLOG_FORCED_SHUTDOWN(log)) {
 		spin_unlock(&log->l_icloglock);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	iclog = log->l_iclog;
@@ -3010,7 +3062,7 @@ xlog_state_release_iclog(
 	int		sync = 0;	/* do we sync? */
 
 	if (iclog->ic_state & XLOG_STATE_IOERROR)
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	ASSERT(atomic_read(&iclog->ic_refcnt) > 0);
 	if (!atomic_dec_and_lock(&iclog->ic_refcnt, &log->l_icloglock))
@@ -3018,7 +3070,7 @@ xlog_state_release_iclog(
 
 	if (iclog->ic_state & XLOG_STATE_IOERROR) {
 		spin_unlock(&log->l_icloglock);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 	ASSERT(iclog->ic_state == XLOG_STATE_ACTIVE ||
 	       iclog->ic_state == XLOG_STATE_WANT_SYNC);
@@ -3135,7 +3187,7 @@ _xfs_log_force(
 	iclog = log->l_iclog;
 	if (iclog->ic_state & XLOG_STATE_IOERROR) {
 		spin_unlock(&log->l_icloglock);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	/* If the head iclog is not active nor dirty, we just attach
@@ -3173,7 +3225,7 @@ _xfs_log_force(
 				spin_unlock(&log->l_icloglock);
 
 				if (xlog_state_release_iclog(log, iclog))
-					return XFS_ERROR(EIO);
+					return -EIO;
 
 				if (log_flushed)
 					*log_flushed = 1;
@@ -3209,7 +3261,7 @@ maybe_sleep:
 		 */
 		if (iclog->ic_state & XLOG_STATE_IOERROR) {
 			spin_unlock(&log->l_icloglock);
-			return XFS_ERROR(EIO);
+			return -EIO;
 		}
 		XFS_STATS_INC(xs_log_force_sleep);
 		xlog_wait(&iclog->ic_force_wait, &log->l_icloglock);
@@ -3219,7 +3271,7 @@ maybe_sleep:
 		 * and the memory read should be atomic.
 		 */
 		if (iclog->ic_state & XLOG_STATE_IOERROR)
-			return XFS_ERROR(EIO);
+			return -EIO;
 		if (log_flushed)
 			*log_flushed = 1;
 	} else {
@@ -3287,7 +3339,7 @@ try_again:
 	iclog = log->l_iclog;
 	if (iclog->ic_state & XLOG_STATE_IOERROR) {
 		spin_unlock(&log->l_icloglock);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	do {
@@ -3338,7 +3390,7 @@ try_again:
 			xlog_state_switch_iclogs(log, iclog, 0);
 			spin_unlock(&log->l_icloglock);
 			if (xlog_state_release_iclog(log, iclog))
-				return XFS_ERROR(EIO);
+				return -EIO;
 			if (log_flushed)
 				*log_flushed = 1;
 			spin_lock(&log->l_icloglock);
@@ -3353,7 +3405,7 @@ try_again:
 			 */
 			if (iclog->ic_state & XLOG_STATE_IOERROR) {
 				spin_unlock(&log->l_icloglock);
-				return XFS_ERROR(EIO);
+				return -EIO;
 			}
 			XFS_STATS_INC(xs_log_force_sleep);
 			xlog_wait(&iclog->ic_force_wait, &log->l_icloglock);
@@ -3363,7 +3415,7 @@ try_again:
 			 * and the memory read should be atomic.
 			 */
 			if (iclog->ic_state & XLOG_STATE_IOERROR)
-				return XFS_ERROR(EIO);
+				return -EIO;
 
 			if (log_flushed)
 				*log_flushed = 1;
@@ -3915,11 +3967,14 @@ xfs_log_force_umount(
 		retval = xlog_state_ioerror(log);
 		spin_unlock(&log->l_icloglock);
 	}
+
 	/*
-	 * Wake up everybody waiting on xfs_log_force.
-	 * Callback all log item committed functions as if the
-	 * log writes were completed.
+	 * Wake up everybody waiting on xfs_log_force. Wake the CIL push first
+	 * as if the log writes were completed. The abort handling in the log
+	 * item committed callback functions will do this again under lock to
+	 * avoid races.
 	 */
+	wake_up_all(&log->l_cilp->xc_commit_wait);
 	xlog_state_do_callback(log, XFS_LI_ABORTED, NULL);
 
 #ifdef XFSERRORDEBUG
diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h
index e148719e0a5d..84e0deb95abd 100644
--- a/fs/xfs/xfs_log.h
+++ b/fs/xfs/xfs_log.h
@@ -24,12 +24,65 @@ struct xfs_log_vec {
 	struct xfs_log_iovec	*lv_iovecp;	/* iovec array */
 	struct xfs_log_item	*lv_item;	/* owner */
 	char			*lv_buf;	/* formatted buffer */
-	int			lv_buf_len;	/* size of formatted buffer */
+	int			lv_bytes;	/* accounted space in buffer */
+	int			lv_buf_len;	/* aligned size of buffer */
 	int			lv_size;	/* size of allocated lv */
 };
 
 #define XFS_LOG_VEC_ORDERED	(-1)
 
+static inline void *
+xlog_prepare_iovec(struct xfs_log_vec *lv, struct xfs_log_iovec **vecp,
+		uint type)
+{
+	struct xfs_log_iovec *vec = *vecp;
+
+	if (vec) {
+		ASSERT(vec - lv->lv_iovecp < lv->lv_niovecs);
+		vec++;
+	} else {
+		vec = &lv->lv_iovecp[0];
+	}
+
+	vec->i_type = type;
+	vec->i_addr = lv->lv_buf + lv->lv_buf_len;
+
+	ASSERT(IS_ALIGNED((unsigned long)vec->i_addr, sizeof(uint64_t)));
+
+	*vecp = vec;
+	return vec->i_addr;
+}
+
+/*
+ * We need to make sure the next buffer is naturally aligned for the biggest
+ * basic data type we put into it.  We already accounted for this padding when
+ * sizing the buffer.
+ *
+ * However, this padding does not get written into the log, and hence we have to
+ * track the space used by the log vectors separately to prevent log space hangs
+ * due to inaccurate accounting (i.e. a leak) of the used log space through the
+ * CIL context ticket.
+ */
+static inline void
+xlog_finish_iovec(struct xfs_log_vec *lv, struct xfs_log_iovec *vec, int len)
+{
+	lv->lv_buf_len += round_up(len, sizeof(uint64_t));
+	lv->lv_bytes += len;
+	vec->i_len = len;
+}
+
+static inline void *
+xlog_copy_iovec(struct xfs_log_vec *lv, struct xfs_log_iovec **vecp,
+		uint type, void *data, int len)
+{
+	void *buf;
+
+	buf = xlog_prepare_iovec(lv, vecp, type);
+	memcpy(buf, data, len);
+	xlog_finish_iovec(lv, *vecp, len);
+	return buf;
+}
+
 /*
  * Structure used to pass callback function and the function's argument
  * to the log manager.
@@ -129,7 +182,7 @@ void	  xlog_iodone(struct xfs_buf *);
 struct xlog_ticket *xfs_log_ticket_get(struct xlog_ticket *ticket);
 void	  xfs_log_ticket_put(struct xlog_ticket *ticket);
 
-int	xfs_log_commit_cil(struct xfs_mount *mp, struct xfs_trans *tp,
+void	xfs_log_commit_cil(struct xfs_mount *mp, struct xfs_trans *tp,
 				xfs_lsn_t *commit_lsn, int flags);
 bool	xfs_log_item_in_current_chkpt(struct xfs_log_item *lip);
 
diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c
index 5eb51fc5eb84..f6b79e5325dd 100644
--- a/fs/xfs/xfs_log_cil.c
+++ b/fs/xfs/xfs_log_cil.c
@@ -78,38 +78,6 @@ xlog_cil_init_post_recovery(
 {
 	log->l_cilp->xc_ctx->ticket = xlog_cil_ticket_alloc(log);
 	log->l_cilp->xc_ctx->sequence = 1;
-	log->l_cilp->xc_ctx->commit_lsn = xlog_assign_lsn(log->l_curr_cycle,
-								log->l_curr_block);
-}
-
-STATIC int
-xlog_cil_lv_item_format(
-	struct xfs_log_item	*lip,
-	struct xfs_log_vec	*lv)
-{
-	int	index;
-	char	*ptr;
-
-	/* format new vectors into array */
-	lip->li_ops->iop_format(lip, lv->lv_iovecp);
-
-	/* copy data into existing array */
-	ptr = lv->lv_buf;
-	for (index = 0; index < lv->lv_niovecs; index++) {
-		struct xfs_log_iovec *vec = &lv->lv_iovecp[index];
-
-		memcpy(ptr, vec->i_addr, vec->i_len);
-		vec->i_addr = ptr;
-		ptr += vec->i_len;
-	}
-
-	/*
-	 * some size calculations for log vectors over-estimate, so the caller
-	 * doesn't know the amount of space actually used by the item. Return
-	 * the byte count to the caller so they can check and store it
-	 * appropriately.
-	 */
-	return ptr - lv->lv_buf;
 }
 
 /*
@@ -127,7 +95,7 @@ xfs_cil_prepare_item(
 {
 	/* Account for the new LV being passed in */
 	if (lv->lv_buf_len != XFS_LOG_VEC_ORDERED) {
-		*diff_len += lv->lv_buf_len;
+		*diff_len += lv->lv_bytes;
 		*diff_iovecs += lv->lv_niovecs;
 	}
 
@@ -141,7 +109,7 @@ xfs_cil_prepare_item(
 	else if (old_lv != lv) {
 		ASSERT(lv->lv_buf_len != XFS_LOG_VEC_ORDERED);
 
-		*diff_len -= old_lv->lv_buf_len;
+		*diff_len -= old_lv->lv_bytes;
 		*diff_iovecs -= old_lv->lv_niovecs;
 		kmem_free(old_lv);
 	}
@@ -232,12 +200,28 @@ xlog_cil_insert_format_items(
 			nbytes = 0;
 		}
 
+		/*
+		 * We 64-bit align the length of each iovec so that the start
+		 * of the next one is naturally aligned.  We'll need to
+		 * account for that slack space here. Then round nbytes up
+		 * to 64-bit alignment so that the initial buffer alignment is
+		 * easy to calculate and verify.
+		 */
+		nbytes += niovecs * sizeof(uint64_t);
+		nbytes = round_up(nbytes, sizeof(uint64_t));
+
 		/* grab the old item if it exists for reservation accounting */
 		old_lv = lip->li_lv;
 
-		/* calc buffer size */
-		buf_size = sizeof(struct xfs_log_vec) + nbytes +
-				niovecs * sizeof(struct xfs_log_iovec);
+		/*
+		 * The data buffer needs to start 64-bit aligned, so round up
+		 * that space to ensure we can align it appropriately and not
+		 * overrun the buffer.
+		 */
+		buf_size = nbytes +
+			   round_up((sizeof(struct xfs_log_vec) +
+				     niovecs * sizeof(struct xfs_log_iovec)),
+				    sizeof(uint64_t));
 
 		/* compare to existing item size */
 		if (lip->li_lv && buf_size <= lip->li_lv->lv_size) {
@@ -253,35 +237,31 @@ xlog_cil_insert_format_items(
 			 * that the space reservation accounting is correct.
 			 */
 			*diff_iovecs -= lv->lv_niovecs;
-			*diff_len -= lv->lv_buf_len;
-
-			/* Ensure the lv is set up according to ->iop_size */
-			lv->lv_niovecs = niovecs;
-			lv->lv_buf = (char *)lv + buf_size - nbytes;
-
-			lv->lv_buf_len = xlog_cil_lv_item_format(lip, lv);
-			goto insert;
+			*diff_len -= lv->lv_bytes;
+		} else {
+			/* allocate new data chunk */
+			lv = kmem_zalloc(buf_size, KM_SLEEP|KM_NOFS);
+			lv->lv_item = lip;
+			lv->lv_size = buf_size;
+			if (ordered) {
+				/* track as an ordered logvec */
+				ASSERT(lip->li_lv == NULL);
+				lv->lv_buf_len = XFS_LOG_VEC_ORDERED;
+				goto insert;
+			}
+			lv->lv_iovecp = (struct xfs_log_iovec *)&lv[1];
 		}
 
-		/* allocate new data chunk */
-		lv = kmem_zalloc(buf_size, KM_SLEEP|KM_NOFS);
-		lv->lv_item = lip;
-		lv->lv_size = buf_size;
+		/* Ensure the lv is set up according to ->iop_size */
 		lv->lv_niovecs = niovecs;
-		if (ordered) {
-			/* track as an ordered logvec */
-			ASSERT(lip->li_lv == NULL);
-			lv->lv_buf_len = XFS_LOG_VEC_ORDERED;
-			goto insert;
-		}
-
-		/* The allocated iovec region lies beyond the log vector. */
-		lv->lv_iovecp = (struct xfs_log_iovec *)&lv[1];
 
 		/* The allocated data region lies beyond the iovec region */
+		lv->lv_buf_len = 0;
+		lv->lv_bytes = 0;
 		lv->lv_buf = (char *)lv + buf_size - nbytes;
+		ASSERT(IS_ALIGNED((unsigned long)lv->lv_buf, sizeof(uint64_t)));
 
-		lv->lv_buf_len = xlog_cil_lv_item_format(lip, lv);
+		lip->li_ops->iop_format(lip, lv);
 insert:
 		ASSERT(lv->lv_buf_len <= nbytes);
 		xfs_cil_prepare_item(log, lv, old_lv, diff_len, diff_iovecs);
@@ -404,7 +384,15 @@ xlog_cil_committed(
 	xfs_extent_busy_clear(mp, &ctx->busy_extents,
 			     (mp->m_flags & XFS_MOUNT_DISCARD) && !abort);
 
+	/*
+	 * If we are aborting the commit, wake up anyone waiting on the
+	 * committing list.  If we don't, then a shutdown we can leave processes
+	 * waiting in xlog_cil_force_lsn() waiting on a sequence commit that
+	 * will never happen because we aborted it.
+	 */
 	spin_lock(&ctx->cil->xc_push_lock);
+	if (abort)
+		wake_up_all(&ctx->cil->xc_commit_wait);
 	list_del(&ctx->committing);
 	spin_unlock(&ctx->cil->xc_push_lock);
 
@@ -518,13 +506,6 @@ xlog_cil_push(
 	cil->xc_ctx = new_ctx;
 
 	/*
-	 * mirror the new sequence into the cil structure so that we can do
-	 * unlocked checks against the current sequence in log forces without
-	 * risking deferencing a freed context pointer.
-	 */
-	cil->xc_current_sequence = new_ctx->sequence;
-
-	/*
 	 * The switch is now done, so we can drop the context lock and move out
 	 * of a shared context. We can't just go straight to the commit record,
 	 * though - we need to synchronise with previous and future commits so
@@ -542,8 +523,15 @@ xlog_cil_push(
 	 * Hence we need to add this context to the committing context list so
 	 * that higher sequences will wait for us to write out a commit record
 	 * before they do.
+	 *
+	 * xfs_log_force_lsn requires us to mirror the new sequence into the cil
+	 * structure atomically with the addition of this sequence to the
+	 * committing list. This also ensures that we can do unlocked checks
+	 * against the current sequence in log forces without risking
+	 * deferencing a freed context pointer.
 	 */
 	spin_lock(&cil->xc_push_lock);
+	cil->xc_current_sequence = new_ctx->sequence;
 	list_add(&ctx->committing, &cil->xc_committing);
 	spin_unlock(&cil->xc_push_lock);
 	up_write(&cil->xc_ctx_lock);
@@ -583,8 +571,18 @@ restart:
 	spin_lock(&cil->xc_push_lock);
 	list_for_each_entry(new_ctx, &cil->xc_committing, committing) {
 		/*
+		 * Avoid getting stuck in this loop because we were woken by the
+		 * shutdown, but then went back to sleep once already in the
+		 * shutdown state.
+		 */
+		if (XLOG_FORCED_SHUTDOWN(log)) {
+			spin_unlock(&cil->xc_push_lock);
+			goto out_abort_free_ticket;
+		}
+
+		/*
 		 * Higher sequences will wait for this one so skip them.
-		 * Don't wait for own own sequence, either.
+		 * Don't wait for our own sequence, either.
 		 */
 		if (new_ctx->sequence >= ctx->sequence)
 			continue;
@@ -634,7 +632,7 @@ out_abort_free_ticket:
 	xfs_log_ticket_put(tic);
 out_abort:
 	xlog_cil_committed(ctx, XFS_LI_ABORTED);
-	return XFS_ERROR(EIO);
+	return -EIO;
 }
 
 static void
@@ -681,8 +679,14 @@ xlog_cil_push_background(
 
 }
 
+/*
+ * xlog_cil_push_now() is used to trigger an immediate CIL push to the sequence
+ * number that is passed. When it returns, the work will be queued for
+ * @push_seq, but it won't be completed. The caller is expected to do any
+ * waiting for push_seq to complete if it is required.
+ */
 static void
-xlog_cil_push_foreground(
+xlog_cil_push_now(
 	struct xlog	*log,
 	xfs_lsn_t	push_seq)
 {
@@ -707,10 +711,8 @@ xlog_cil_push_foreground(
 	}
 
 	cil->xc_push_seq = push_seq;
+	queue_work(log->l_mp->m_cil_workqueue, &cil->xc_push_work);
 	spin_unlock(&cil->xc_push_lock);
-
-	/* do the push now */
-	xlog_cil_push(log);
 }
 
 bool
@@ -740,7 +742,7 @@ xlog_cil_empty(
  * background commit, returns without it held once background commits are
  * allowed again.
  */
-int
+void
 xfs_log_commit_cil(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
@@ -786,7 +788,6 @@ xfs_log_commit_cil(
 	xlog_cil_push_background(log);
 
 	up_read(&cil->xc_ctx_lock);
-	return 0;
 }
 
 /*
@@ -815,7 +816,8 @@ xlog_cil_force_lsn(
 	 * xlog_cil_push() handles racing pushes for the same sequence,
 	 * so no need to deal with it here.
 	 */
-	xlog_cil_push_foreground(log, sequence);
+restart:
+	xlog_cil_push_now(log, sequence);
 
 	/*
 	 * See if we can find a previous sequence still committing.
@@ -823,9 +825,15 @@ xlog_cil_force_lsn(
 	 * before allowing the force of push_seq to go ahead. Hence block
 	 * on commits for those as well.
 	 */
-restart:
 	spin_lock(&cil->xc_push_lock);
 	list_for_each_entry(ctx, &cil->xc_committing, committing) {
+		/*
+		 * Avoid getting stuck in this loop because we were woken by the
+		 * shutdown, but then went back to sleep once already in the
+		 * shutdown state.
+		 */
+		if (XLOG_FORCED_SHUTDOWN(log))
+			goto out_shutdown;
 		if (ctx->sequence > sequence)
 			continue;
 		if (!ctx->commit_lsn) {
@@ -841,8 +849,39 @@ restart:
 		/* found it! */
 		commit_lsn = ctx->commit_lsn;
 	}
+
+	/*
+	 * The call to xlog_cil_push_now() executes the push in the background.
+	 * Hence by the time we have got here it our sequence may not have been
+	 * pushed yet. This is true if the current sequence still matches the
+	 * push sequence after the above wait loop and the CIL still contains
+	 * dirty objects.
+	 *
+	 * When the push occurs, it will empty the CIL and atomically increment
+	 * the currect sequence past the push sequence and move it into the
+	 * committing list. Of course, if the CIL is clean at the time of the
+	 * push, it won't have pushed the CIL at all, so in that case we should
+	 * try the push for this sequence again from the start just in case.
+	 */
+	if (sequence == cil->xc_current_sequence &&
+	    !list_empty(&cil->xc_cil)) {
+		spin_unlock(&cil->xc_push_lock);
+		goto restart;
+	}
+
 	spin_unlock(&cil->xc_push_lock);
 	return commit_lsn;
+
+	/*
+	 * We detected a shutdown in progress. We need to trigger the log force
+	 * to pass through it's iclog state machine error handling, even though
+	 * we are already in a shutdown state. Hence we can't return
+	 * NULLCOMMITLSN here as that has special meaning to log forces (i.e.
+	 * LSN is already stable), so we return a zero LSN instead.
+	 */
+out_shutdown:
+	spin_unlock(&cil->xc_push_lock);
+	return 0;
 }
 
 /*
@@ -887,12 +926,12 @@ xlog_cil_init(
 
 	cil = kmem_zalloc(sizeof(*cil), KM_SLEEP|KM_MAYFAIL);
 	if (!cil)
-		return ENOMEM;
+		return -ENOMEM;
 
 	ctx = kmem_zalloc(sizeof(*ctx), KM_SLEEP|KM_MAYFAIL);
 	if (!ctx) {
 		kmem_free(cil);
-		return ENOMEM;
+		return -ENOMEM;
 	}
 
 	INIT_WORK(&cil->xc_push_work, xlog_cil_push_work);
diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h
index 9bc403a9e54f..db7cbdeb2b42 100644
--- a/fs/xfs/xfs_log_priv.h
+++ b/fs/xfs/xfs_log_priv.h
@@ -405,6 +405,8 @@ struct xlog {
 	struct xlog_grant_head	l_reserve_head;
 	struct xlog_grant_head	l_write_head;
 
+	struct xfs_kobj		l_kobj;
+
 	/* The following field are used for debugging; need to hold icloglock */
 #ifdef DEBUG
 	char			*l_iclog_bak[XLOG_MAX_ICLOGS];
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index b6b669df40f3..1fd5787add99 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -179,7 +179,7 @@ xlog_bread_noalign(
 		xfs_warn(log->l_mp, "Invalid block length (0x%x) for buffer",
 			nbblks);
 		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 	}
 
 	blk_no = round_down(blk_no, log->l_sectBBsize);
@@ -193,7 +193,10 @@ xlog_bread_noalign(
 	bp->b_io_length = nbblks;
 	bp->b_error = 0;
 
-	xfsbdstrat(log->l_mp, bp);
+	if (XFS_FORCED_SHUTDOWN(log->l_mp))
+		return -EIO;
+
+	xfs_buf_iorequest(bp);
 	error = xfs_buf_iowait(bp);
 	if (error)
 		xfs_buf_ioerror_alert(bp, __func__);
@@ -265,7 +268,7 @@ xlog_bwrite(
 		xfs_warn(log->l_mp, "Invalid block length (0x%x) for buffer",
 			nbblks);
 		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 	}
 
 	blk_no = round_down(blk_no, log->l_sectBBsize);
@@ -327,14 +330,14 @@ xlog_header_check_recover(
 		xlog_header_check_dump(mp, head);
 		XFS_ERROR_REPORT("xlog_header_check_recover(1)",
 				 XFS_ERRLEVEL_HIGH, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	} else if (unlikely(!uuid_equal(&mp->m_sb.sb_uuid, &head->h_fs_uuid))) {
 		xfs_warn(mp,
 	"dirty log entry has mismatched uuid - can't recover");
 		xlog_header_check_dump(mp, head);
 		XFS_ERROR_REPORT("xlog_header_check_recover(2)",
 				 XFS_ERRLEVEL_HIGH, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -361,7 +364,7 @@ xlog_header_check_mount(
 		xlog_header_check_dump(mp, head);
 		XFS_ERROR_REPORT("xlog_header_check_mount",
 				 XFS_ERRLEVEL_HIGH, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -459,7 +462,7 @@ xlog_find_verify_cycle(
 	while (!(bp = xlog_get_bp(log, bufblks))) {
 		bufblks >>= 1;
 		if (bufblks < log->l_sectBBsize)
-			return ENOMEM;
+			return -ENOMEM;
 	}
 
 	for (i = start_blk; i < start_blk + nbblks; i += bufblks) {
@@ -521,7 +524,7 @@ xlog_find_verify_log_record(
 
 	if (!(bp = xlog_get_bp(log, num_blks))) {
 		if (!(bp = xlog_get_bp(log, 1)))
-			return ENOMEM;
+			return -ENOMEM;
 		smallmem = 1;
 	} else {
 		error = xlog_bread(log, start_blk, num_blks, bp, &offset);
@@ -536,7 +539,7 @@ xlog_find_verify_log_record(
 			xfs_warn(log->l_mp,
 		"Log inconsistent (didn't find previous header)");
 			ASSERT(0);
-			error = XFS_ERROR(EIO);
+			error = -EIO;
 			goto out;
 		}
 
@@ -561,7 +564,7 @@ xlog_find_verify_log_record(
 	 * will be called again for the end of the physical log.
 	 */
 	if (i == -1) {
-		error = -1;
+		error = 1;
 		goto out;
 	}
 
@@ -625,7 +628,12 @@ xlog_find_head(
 	int		error, log_bbnum = log->l_logBBsize;
 
 	/* Is the end of the log device zeroed? */
-	if ((error = xlog_find_zeroed(log, &first_blk)) == -1) {
+	error = xlog_find_zeroed(log, &first_blk);
+	if (error < 0) {
+		xfs_warn(log->l_mp, "empty log check failed");
+		return error;
+	}
+	if (error == 1) {
 		*return_head_blk = first_blk;
 
 		/* Is the whole lot zeroed? */
@@ -638,15 +646,12 @@ xlog_find_head(
 		}
 
 		return 0;
-	} else if (error) {
-		xfs_warn(log->l_mp, "empty log check failed");
-		return error;
 	}
 
 	first_blk = 0;			/* get cycle # of 1st block */
 	bp = xlog_get_bp(log, 1);
 	if (!bp)
-		return ENOMEM;
+		return -ENOMEM;
 
 	error = xlog_bread(log, 0, 1, bp, &offset);
 	if (error)
@@ -815,29 +820,29 @@ validate_head:
 		start_blk = head_blk - num_scan_bblks; /* don't read head_blk */
 
 		/* start ptr at last block ptr before head_blk */
-		if ((error = xlog_find_verify_log_record(log, start_blk,
-							&head_blk, 0)) == -1) {
-			error = XFS_ERROR(EIO);
-			goto bp_err;
-		} else if (error)
+		error = xlog_find_verify_log_record(log, start_blk, &head_blk, 0);
+		if (error == 1)
+			error = -EIO;
+		if (error)
 			goto bp_err;
 	} else {
 		start_blk = 0;
 		ASSERT(head_blk <= INT_MAX);
-		if ((error = xlog_find_verify_log_record(log, start_blk,
-							&head_blk, 0)) == -1) {
+		error = xlog_find_verify_log_record(log, start_blk, &head_blk, 0);
+		if (error < 0)
+			goto bp_err;
+		if (error == 1) {
 			/* We hit the beginning of the log during our search */
 			start_blk = log_bbnum - (num_scan_bblks - head_blk);
 			new_blk = log_bbnum;
 			ASSERT(start_blk <= INT_MAX &&
 				(xfs_daddr_t) log_bbnum-start_blk >= 0);
 			ASSERT(head_blk <= INT_MAX);
-			if ((error = xlog_find_verify_log_record(log,
-							start_blk, &new_blk,
-							(int)head_blk)) == -1) {
-				error = XFS_ERROR(EIO);
-				goto bp_err;
-			} else if (error)
+			error = xlog_find_verify_log_record(log, start_blk,
+							&new_blk, (int)head_blk);
+			if (error == 1)
+				error = -EIO;
+			if (error)
 				goto bp_err;
 			if (new_blk != log_bbnum)
 				head_blk = new_blk;
@@ -908,7 +913,7 @@ xlog_find_tail(
 
 	bp = xlog_get_bp(log, 1);
 	if (!bp)
-		return ENOMEM;
+		return -ENOMEM;
 	if (*head_blk == 0) {				/* special case */
 		error = xlog_bread(log, 0, 1, bp, &offset);
 		if (error)
@@ -958,7 +963,7 @@ xlog_find_tail(
 		xfs_warn(log->l_mp, "%s: couldn't find sync record", __func__);
 		xlog_put_bp(bp);
 		ASSERT(0);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	/* find blk_no of tail of log */
@@ -1089,8 +1094,8 @@ done:
  *
  * Return:
  *	0  => the log is completely written to
- *	-1 => use *blk_no as the first block of the log
- *	>0 => error has occurred
+ *	1 => use *blk_no as the first block of the log
+ *	<0 => error has occurred
  */
 STATIC int
 xlog_find_zeroed(
@@ -1109,7 +1114,7 @@ xlog_find_zeroed(
 	/* check totally zeroed log */
 	bp = xlog_get_bp(log, 1);
 	if (!bp)
-		return ENOMEM;
+		return -ENOMEM;
 	error = xlog_bread(log, 0, 1, bp, &offset);
 	if (error)
 		goto bp_err;
@@ -1118,7 +1123,7 @@ xlog_find_zeroed(
 	if (first_cycle == 0) {		/* completely zeroed log */
 		*blk_no = 0;
 		xlog_put_bp(bp);
-		return -1;
+		return 1;
 	}
 
 	/* check partially zeroed log */
@@ -1138,7 +1143,7 @@ xlog_find_zeroed(
 		 */
 		xfs_warn(log->l_mp,
 			"Log inconsistent or not a log (last==0, first!=1)");
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto bp_err;
 	}
 
@@ -1176,19 +1181,18 @@ xlog_find_zeroed(
 	 * Potentially backup over partial log record write.  We don't need
 	 * to search the end of the log because we know it is zero.
 	 */
-	if ((error = xlog_find_verify_log_record(log, start_blk,
-				&last_blk, 0)) == -1) {
-	    error = XFS_ERROR(EIO);
-	    goto bp_err;
-	} else if (error)
-	    goto bp_err;
+	error = xlog_find_verify_log_record(log, start_blk, &last_blk, 0);
+	if (error == 1)
+		error = -EIO;
+	if (error)
+		goto bp_err;
 
 	*blk_no = last_blk;
 bp_err:
 	xlog_put_bp(bp);
 	if (error)
 		return error;
-	return -1;
+	return 1;
 }
 
 /*
@@ -1248,7 +1252,7 @@ xlog_write_log_records(
 	while (!(bp = xlog_get_bp(log, bufblks))) {
 		bufblks >>= 1;
 		if (bufblks < sectbb)
-			return ENOMEM;
+			return -ENOMEM;
 	}
 
 	/* We may need to do a read at the start to fill in part of
@@ -1351,7 +1355,7 @@ xlog_clear_stale_blocks(
 		if (unlikely(head_block < tail_block || head_block >= log->l_logBBsize)) {
 			XFS_ERROR_REPORT("xlog_clear_stale_blocks(1)",
 					 XFS_ERRLEVEL_LOW, log->l_mp);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 		tail_distance = tail_block + (log->l_logBBsize - head_block);
 	} else {
@@ -1363,7 +1367,7 @@ xlog_clear_stale_blocks(
 		if (unlikely(head_block >= tail_block || head_cycle != (tail_cycle + 1))){
 			XFS_ERROR_REPORT("xlog_clear_stale_blocks(2)",
 					 XFS_ERRLEVEL_LOW, log->l_mp);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 		tail_distance = tail_block - head_block;
 	}
@@ -1548,7 +1552,7 @@ xlog_recover_add_to_trans(
 			xfs_warn(log->l_mp, "%s: bad header magic number",
 				__func__);
 			ASSERT(0);
-			return XFS_ERROR(EIO);
+			return -EIO;
 		}
 		if (len == sizeof(xfs_trans_header_t))
 			xlog_recover_add_item(&trans->r_itemq);
@@ -1578,7 +1582,7 @@ xlog_recover_add_to_trans(
 				  in_f->ilf_size);
 			ASSERT(0);
 			kmem_free(ptr);
-			return XFS_ERROR(EIO);
+			return -EIO;
 		}
 
 		item->ri_total = in_f->ilf_size;
@@ -1651,6 +1655,7 @@ xlog_recover_reorder_trans(
 	int			pass)
 {
 	xlog_recover_item_t	*item, *n;
+	int			error = 0;
 	LIST_HEAD(sort_list);
 	LIST_HEAD(cancel_list);
 	LIST_HEAD(buffer_list);
@@ -1692,9 +1697,17 @@ xlog_recover_reorder_trans(
 				"%s: unrecognized type of log operation",
 				__func__);
 			ASSERT(0);
-			return XFS_ERROR(EIO);
+			/*
+			 * return the remaining items back to the transaction
+			 * item list so they can be freed in caller.
+			 */
+			if (!list_empty(&sort_list))
+				list_splice_init(&sort_list, &trans->r_itemq);
+			error = -EIO;
+			goto out;
 		}
 	}
+out:
 	ASSERT(list_empty(&sort_list));
 	if (!list_empty(&buffer_list))
 		list_splice(&buffer_list, &trans->r_itemq);
@@ -1704,7 +1717,7 @@ xlog_recover_reorder_trans(
 		list_splice_tail(&inode_buffer_list, &trans->r_itemq);
 	if (!list_empty(&cancel_list))
 		list_splice_tail(&cancel_list, &trans->r_itemq);
-	return 0;
+	return error;
 }
 
 /*
@@ -1931,7 +1944,7 @@ xlog_recover_do_inode_buffer(
 				item, bp);
 			XFS_ERROR_REPORT("xlog_recover_do_inode_buf",
 					 XFS_ERRLEVEL_LOW, mp);
-			return XFS_ERROR(EFSCORRUPTED);
+			return -EFSCORRUPTED;
 		}
 
 		buffer_nextp = (xfs_agino_t *)xfs_buf_offset(bp,
@@ -2113,6 +2126,17 @@ xlog_recover_validate_buf_type(
 	__uint16_t		magic16;
 	__uint16_t		magicda;
 
+	/*
+	 * We can only do post recovery validation on items on CRC enabled
+	 * fielsystems as we need to know when the buffer was written to be able
+	 * to determine if we should have replayed the item. If we replay old
+	 * metadata over a newer buffer, then it will enter a temporarily
+	 * inconsistent state resulting in verification failures. Hence for now
+	 * just avoid the verification stage for non-crc filesystems
+	 */
+	if (!xfs_sb_version_hascrc(&mp->m_sb))
+		return;
+
 	magic32 = be32_to_cpu(*(__be32 *)bp->b_addr);
 	magic16 = be16_to_cpu(*(__be16*)bp->b_addr);
 	magicda = be16_to_cpu(info->magic);
@@ -2126,7 +2150,9 @@ xlog_recover_validate_buf_type(
 			bp->b_ops = &xfs_allocbt_buf_ops;
 			break;
 		case XFS_IBT_CRC_MAGIC:
+		case XFS_FIBT_CRC_MAGIC:
 		case XFS_IBT_MAGIC:
+		case XFS_FIBT_MAGIC:
 			bp->b_ops = &xfs_inobt_buf_ops;
 			break;
 		case XFS_BMAP_CRC_MAGIC:
@@ -2148,8 +2174,6 @@ xlog_recover_validate_buf_type(
 		bp->b_ops = &xfs_agf_buf_ops;
 		break;
 	case XFS_BLFT_AGFL_BUF:
-		if (!xfs_sb_version_hascrc(&mp->m_sb))
-			break;
 		if (magic32 != XFS_AGFL_MAGIC) {
 			xfs_warn(mp, "Bad AGFL block magic!");
 			ASSERT(0);
@@ -2182,10 +2206,6 @@ xlog_recover_validate_buf_type(
 #endif
 		break;
 	case XFS_BLFT_DINO_BUF:
-		/*
-		 * we get here with inode allocation buffers, not buffers that
-		 * track unlinked list changes.
-		 */
 		if (magic16 != XFS_DINODE_MAGIC) {
 			xfs_warn(mp, "Bad INODE block magic!");
 			ASSERT(0);
@@ -2265,8 +2285,6 @@ xlog_recover_validate_buf_type(
 		bp->b_ops = &xfs_attr3_leaf_buf_ops;
 		break;
 	case XFS_BLFT_ATTR_RMT_BUF:
-		if (!xfs_sb_version_hascrc(&mp->m_sb))
-			break;
 		if (magic32 != XFS_ATTR3_RMT_MAGIC) {
 			xfs_warn(mp, "Bad attr remote magic!");
 			ASSERT(0);
@@ -2373,16 +2391,7 @@ xlog_recover_do_reg_buffer(
 	/* Shouldn't be any more regions */
 	ASSERT(i == item->ri_total);
 
-	/*
-	 * We can only do post recovery validation on items on CRC enabled
-	 * fielsystems as we need to know when the buffer was written to be able
-	 * to determine if we should have replayed the item. If we replay old
-	 * metadata over a newer buffer, then it will enter a temporarily
-	 * inconsistent state resulting in verification failures. Hence for now
-	 * just avoid the verification stage for non-crc filesystems
-	 */
-	if (xfs_sb_version_hascrc(&mp->m_sb))
-		xlog_recover_validate_buf_type(mp, bp, buf_f);
+	xlog_recover_validate_buf_type(mp, bp, buf_f);
 }
 
 /*
@@ -2390,8 +2399,11 @@ xlog_recover_do_reg_buffer(
  * Simple algorithm: if we have found a QUOTAOFF log item of the same type
  * (ie. USR or GRP), then just toss this buffer away; don't recover it.
  * Else, treat it as a regular buffer and do recovery.
+ *
+ * Return false if the buffer was tossed and true if we recovered the buffer to
+ * indicate to the caller if the buffer needs writing.
  */
-STATIC void
+STATIC bool
 xlog_recover_do_dquot_buffer(
 	struct xfs_mount		*mp,
 	struct xlog			*log,
@@ -2406,9 +2418,8 @@ xlog_recover_do_dquot_buffer(
 	/*
 	 * Filesystems are required to send in quota flags at mount time.
 	 */
-	if (mp->m_qflags == 0) {
-		return;
-	}
+	if (!mp->m_qflags)
+		return false;
 
 	type = 0;
 	if (buf_f->blf_flags & XFS_BLF_UDQUOT_BUF)
@@ -2421,9 +2432,10 @@ xlog_recover_do_dquot_buffer(
 	 * This type of quotas was turned off, so ignore this buffer
 	 */
 	if (log->l_quotaoffs_flag & type)
-		return;
+		return false;
 
 	xlog_recover_do_reg_buffer(mp, item, bp, buf_f);
+	return true;
 }
 
 /*
@@ -2482,7 +2494,7 @@ xlog_recover_buffer_pass2(
 	bp = xfs_buf_read(mp->m_ddev_targp, buf_f->blf_blkno, buf_f->blf_len,
 			  buf_flags, NULL);
 	if (!bp)
-		return XFS_ERROR(ENOMEM);
+		return -ENOMEM;
 	error = bp->b_error;
 	if (error) {
 		xfs_buf_ioerror_alert(bp, "xlog_recover_do..(read#1)");
@@ -2490,23 +2502,44 @@ xlog_recover_buffer_pass2(
 	}
 
 	/*
-	 * recover the buffer only if we get an LSN from it and it's less than
+	 * Recover the buffer only if we get an LSN from it and it's less than
 	 * the lsn of the transaction we are replaying.
+	 *
+	 * Note that we have to be extremely careful of readahead here.
+	 * Readahead does not attach verfiers to the buffers so if we don't
+	 * actually do any replay after readahead because of the LSN we found
+	 * in the buffer if more recent than that current transaction then we
+	 * need to attach the verifier directly. Failure to do so can lead to
+	 * future recovery actions (e.g. EFI and unlinked list recovery) can
+	 * operate on the buffers and they won't get the verifier attached. This
+	 * can lead to blocks on disk having the correct content but a stale
+	 * CRC.
+	 *
+	 * It is safe to assume these clean buffers are currently up to date.
+	 * If the buffer is dirtied by a later transaction being replayed, then
+	 * the verifier will be reset to match whatever recover turns that
+	 * buffer into.
 	 */
 	lsn = xlog_recover_get_buf_lsn(mp, bp);
-	if (lsn && lsn != -1 && XFS_LSN_CMP(lsn, current_lsn) >= 0)
+	if (lsn && lsn != -1 && XFS_LSN_CMP(lsn, current_lsn) >= 0) {
+		xlog_recover_validate_buf_type(mp, bp, buf_f);
 		goto out_release;
+	}
 
 	if (buf_f->blf_flags & XFS_BLF_INODE_BUF) {
 		error = xlog_recover_do_inode_buffer(mp, item, bp, buf_f);
+		if (error)
+			goto out_release;
 	} else if (buf_f->blf_flags &
 		  (XFS_BLF_UDQUOT_BUF|XFS_BLF_PDQUOT_BUF|XFS_BLF_GDQUOT_BUF)) {
-		xlog_recover_do_dquot_buffer(mp, log, item, bp, buf_f);
+		bool	dirty;
+
+		dirty = xlog_recover_do_dquot_buffer(mp, log, item, bp, buf_f);
+		if (!dirty)
+			goto out_release;
 	} else {
 		xlog_recover_do_reg_buffer(mp, item, bp, buf_f);
 	}
-	if (error)
-		goto out_release;
 
 	/*
 	 * Perform delayed write on the buffer.  Asynchronous writes will be
@@ -2514,19 +2547,19 @@ xlog_recover_buffer_pass2(
 	 *
 	 * Also make sure that only inode buffers with good sizes stay in
 	 * the buffer cache.  The kernel moves inodes in buffers of 1 block
-	 * or XFS_INODE_CLUSTER_SIZE bytes, whichever is bigger.  The inode
+	 * or mp->m_inode_cluster_size bytes, whichever is bigger.  The inode
 	 * buffers in the log can be a different size if the log was generated
 	 * by an older kernel using unclustered inode buffers or a newer kernel
 	 * running with a different inode cluster size.  Regardless, if the
-	 * the inode buffer size isn't MAX(blocksize, XFS_INODE_CLUSTER_SIZE)
-	 * for *our* value of XFS_INODE_CLUSTER_SIZE, then we need to keep
+	 * the inode buffer size isn't MAX(blocksize, mp->m_inode_cluster_size)
+	 * for *our* value of mp->m_inode_cluster_size, then we need to keep
 	 * the buffer out of the buffer cache so that the buffer won't
 	 * overlap with future reads of those inodes.
 	 */
 	if (XFS_DINODE_MAGIC ==
 	    be16_to_cpu(*((__be16 *)xfs_buf_offset(bp, 0))) &&
 	    (BBTOB(bp->b_io_length) != MAX(log->l_mp->m_sb.sb_blocksize,
-			(__uint32_t)XFS_INODE_CLUSTER_SIZE(log->l_mp)))) {
+			(__uint32_t)log->l_mp->m_inode_cluster_size))) {
 		xfs_buf_stale(bp);
 		error = xfs_bwrite(bp);
 	} else {
@@ -2584,7 +2617,7 @@ xfs_recover_inode_owner_change(
 
 	ip = xfs_inode_alloc(mp, in_f->ilf_ino);
 	if (!ip)
-		return ENOMEM;
+		return -ENOMEM;
 
 	/* instantiate the inode */
 	xfs_dinode_from_disk(&ip->i_d, dip);
@@ -2662,7 +2695,7 @@ xlog_recover_inode_pass2(
 	bp = xfs_buf_read(mp->m_ddev_targp, in_f->ilf_blkno, in_f->ilf_len, 0,
 			  &xfs_inode_buf_ops);
 	if (!bp) {
-		error = ENOMEM;
+		error = -ENOMEM;
 		goto error;
 	}
 	error = bp->b_error;
@@ -2683,7 +2716,7 @@ xlog_recover_inode_pass2(
 			__func__, dip, bp, in_f->ilf_ino);
 		XFS_ERROR_REPORT("xlog_recover_inode_pass2(1)",
 				 XFS_ERRLEVEL_LOW, mp);
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto out_release;
 	}
 	dicp = item->ri_buf[1].i_addr;
@@ -2693,7 +2726,7 @@ xlog_recover_inode_pass2(
 			__func__, item, in_f->ilf_ino);
 		XFS_ERROR_REPORT("xlog_recover_inode_pass2(2)",
 				 XFS_ERRLEVEL_LOW, mp);
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto out_release;
 	}
 
@@ -2750,7 +2783,7 @@ xlog_recover_inode_pass2(
 		"%s: Bad regular inode log record, rec ptr 0x%p, "
 		"ino ptr = 0x%p, ino bp = 0x%p, ino %Ld",
 				__func__, item, dip, bp, in_f->ilf_ino);
-			error = EFSCORRUPTED;
+			error = -EFSCORRUPTED;
 			goto out_release;
 		}
 	} else if (unlikely(S_ISDIR(dicp->di_mode))) {
@@ -2763,7 +2796,7 @@ xlog_recover_inode_pass2(
 		"%s: Bad dir inode log record, rec ptr 0x%p, "
 		"ino ptr = 0x%p, ino bp = 0x%p, ino %Ld",
 				__func__, item, dip, bp, in_f->ilf_ino);
-			error = EFSCORRUPTED;
+			error = -EFSCORRUPTED;
 			goto out_release;
 		}
 	}
@@ -2776,7 +2809,7 @@ xlog_recover_inode_pass2(
 			__func__, item, dip, bp, in_f->ilf_ino,
 			dicp->di_nextents + dicp->di_anextents,
 			dicp->di_nblocks);
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto out_release;
 	}
 	if (unlikely(dicp->di_forkoff > mp->m_sb.sb_inodesize)) {
@@ -2786,7 +2819,7 @@ xlog_recover_inode_pass2(
 	"%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);
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto out_release;
 	}
 	isize = xfs_icdinode_size(dicp->di_version);
@@ -2796,7 +2829,7 @@ xlog_recover_inode_pass2(
 		xfs_alert(mp,
 			"%s: Bad inode log record length %d, rec ptr 0x%p",
 			__func__, item->ri_buf[1].i_len, item);
-		error = EFSCORRUPTED;
+		error = -EFSCORRUPTED;
 		goto out_release;
 	}
 
@@ -2884,7 +2917,7 @@ xlog_recover_inode_pass2(
 		default:
 			xfs_warn(log->l_mp, "%s: Invalid flag", __func__);
 			ASSERT(0);
-			error = EIO;
+			error = -EIO;
 			goto out_release;
 		}
 	}
@@ -2905,7 +2938,7 @@ out_release:
 error:
 	if (need_free)
 		kmem_free(in_f);
-	return XFS_ERROR(error);
+	return error;
 }
 
 /*
@@ -2932,7 +2965,7 @@ xlog_recover_quotaoff_pass1(
 	if (qoff_f->qf_flags & XFS_GQUOTA_ACCT)
 		log->l_quotaoffs_flag |= XFS_DQ_GROUP;
 
-	return (0);
+	return 0;
 }
 
 /*
@@ -2957,17 +2990,17 @@ xlog_recover_dquot_pass2(
 	 * Filesystems are required to send in quota flags at mount time.
 	 */
 	if (mp->m_qflags == 0)
-		return (0);
+		return 0;
 
 	recddq = item->ri_buf[1].i_addr;
 	if (recddq == NULL) {
 		xfs_alert(log->l_mp, "NULL dquot in %s.", __func__);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 	if (item->ri_buf[1].i_len < sizeof(xfs_disk_dquot_t)) {
 		xfs_alert(log->l_mp, "dquot too small (%d) in %s.",
 			item->ri_buf[1].i_len, __func__);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	/*
@@ -2976,7 +3009,7 @@ xlog_recover_dquot_pass2(
 	type = recddq->d_flags & (XFS_DQ_USER | XFS_DQ_PROJ | XFS_DQ_GROUP);
 	ASSERT(type);
 	if (log->l_quotaoffs_flag & type)
-		return (0);
+		return 0;
 
 	/*
 	 * At this point we know that quota was _not_ turned off.
@@ -2993,12 +3026,19 @@ xlog_recover_dquot_pass2(
 	error = xfs_dqcheck(mp, recddq, dq_f->qlf_id, 0, XFS_QMOPT_DOWARN,
 			   "xlog_recover_dquot_pass2 (log copy)");
 	if (error)
-		return XFS_ERROR(EIO);
+		return -EIO;
 	ASSERT(dq_f->qlf_len == 1);
 
+	/*
+	 * At this point we are assuming that the dquots have been allocated
+	 * and hence the buffer has valid dquots stamped in it. It should,
+	 * therefore, pass verifier validation. If the dquot is bad, then the
+	 * we'll return an error here, so we don't need to specifically check
+	 * the dquot in the buffer after the verifier has run.
+	 */
 	error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, dq_f->qlf_blkno,
 				   XFS_FSB_TO_BB(mp, dq_f->qlf_len), 0, &bp,
-				   NULL);
+				   &xfs_dquot_buf_ops);
 	if (error)
 		return error;
 
@@ -3006,18 +3046,6 @@ xlog_recover_dquot_pass2(
 	ddq = (xfs_disk_dquot_t *)xfs_buf_offset(bp, dq_f->qlf_boffset);
 
 	/*
-	 * At least the magic num portion should be on disk because this
-	 * was among a chunk of dquots created earlier, and we did some
-	 * minimal initialization then.
-	 */
-	error = xfs_dqcheck(mp, ddq, dq_f->qlf_id, 0, XFS_QMOPT_DOWARN,
-			   "xlog_recover_dquot_pass2");
-	if (error) {
-		xfs_buf_relse(bp);
-		return XFS_ERROR(EIO);
-	}
-
-	/*
 	 * If the dquot has an LSN in it, recover the dquot only if it's less
 	 * than the lsn of the transaction we are replaying.
 	 */
@@ -3133,7 +3161,7 @@ xlog_recover_efd_pass2(
 		}
 		lip = xfs_trans_ail_cursor_next(ailp, &cur);
 	}
-	xfs_trans_ail_cursor_done(ailp, &cur);
+	xfs_trans_ail_cursor_done(&cur);
 	spin_unlock(&ailp->xa_lock);
 
 	return 0;
@@ -3164,47 +3192,47 @@ xlog_recover_do_icreate_pass2(
 	icl = (struct xfs_icreate_log *)item->ri_buf[0].i_addr;
 	if (icl->icl_type != XFS_LI_ICREATE) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad type");
-		return EINVAL;
+		return -EINVAL;
 	}
 
 	if (icl->icl_size != 1) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad icl size");
-		return EINVAL;
+		return -EINVAL;
 	}
 
 	agno = be32_to_cpu(icl->icl_ag);
 	if (agno >= mp->m_sb.sb_agcount) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad agno");
-		return EINVAL;
+		return -EINVAL;
 	}
 	agbno = be32_to_cpu(icl->icl_agbno);
 	if (!agbno || agbno == NULLAGBLOCK || agbno >= mp->m_sb.sb_agblocks) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad agbno");
-		return EINVAL;
+		return -EINVAL;
 	}
 	isize = be32_to_cpu(icl->icl_isize);
 	if (isize != mp->m_sb.sb_inodesize) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad isize");
-		return EINVAL;
+		return -EINVAL;
 	}
 	count = be32_to_cpu(icl->icl_count);
 	if (!count) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad count");
-		return EINVAL;
+		return -EINVAL;
 	}
 	length = be32_to_cpu(icl->icl_length);
 	if (!length || length >= mp->m_sb.sb_agblocks) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad length");
-		return EINVAL;
+		return -EINVAL;
 	}
 
 	/* existing allocation is fixed value */
-	ASSERT(count == XFS_IALLOC_INODES(mp));
-	ASSERT(length == XFS_IALLOC_BLOCKS(mp));
-	if (count != XFS_IALLOC_INODES(mp) ||
-	     length != XFS_IALLOC_BLOCKS(mp)) {
+	ASSERT(count == mp->m_ialloc_inos);
+	ASSERT(length == mp->m_ialloc_blks);
+	if (count != mp->m_ialloc_inos ||
+	     length != mp->m_ialloc_blks) {
 		xfs_warn(log->l_mp, "xlog_recover_do_icreate_trans: bad count 2");
-		return EINVAL;
+		return -EINVAL;
 	}
 
 	/*
@@ -3375,7 +3403,7 @@ xlog_recover_commit_pass1(
 		xfs_warn(log->l_mp, "%s: invalid item type (%d)",
 			__func__, ITEM_TYPE(item));
 		ASSERT(0);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 }
 
@@ -3411,7 +3439,7 @@ xlog_recover_commit_pass2(
 		xfs_warn(log->l_mp, "%s: invalid item type (%d)",
 			__func__, ITEM_TYPE(item));
 		ASSERT(0);
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 }
 
@@ -3508,8 +3536,7 @@ out:
 
 STATIC int
 xlog_recover_unmount_trans(
-	struct xlog		*log,
-	struct xlog_recover	*trans)
+	struct xlog		*log)
 {
 	/* Do nothing now */
 	xfs_warn(log->l_mp, "%s: Unmount LR", __func__);
@@ -3547,7 +3574,7 @@ xlog_recover_process_data(
 
 	/* check the log format matches our own - else we can't recover */
 	if (xlog_header_check_recover(log->l_mp, rhead))
-		return (XFS_ERROR(EIO));
+		return -EIO;
 
 	while ((dp < lp) && num_logops) {
 		ASSERT(dp + sizeof(xlog_op_header_t) <= lp);
@@ -3558,7 +3585,7 @@ xlog_recover_process_data(
 			xfs_warn(log->l_mp, "%s: bad clientid 0x%x",
 					__func__, ohead->oh_clientid);
 			ASSERT(0);
-			return (XFS_ERROR(EIO));
+			return -EIO;
 		}
 		tid = be32_to_cpu(ohead->oh_tid);
 		hash = XLOG_RHASH(tid);
@@ -3572,7 +3599,7 @@ xlog_recover_process_data(
 				xfs_warn(log->l_mp, "%s: bad length 0x%x",
 					__func__, be32_to_cpu(ohead->oh_len));
 				WARN_ON(1);
-				return (XFS_ERROR(EIO));
+				return -EIO;
 			}
 			flags = ohead->oh_flags & ~XLOG_END_TRANS;
 			if (flags & XLOG_WAS_CONT_TRANS)
@@ -3583,7 +3610,7 @@ xlog_recover_process_data(
 								trans, pass);
 				break;
 			case XLOG_UNMOUNT_TRANS:
-				error = xlog_recover_unmount_trans(log, trans);
+				error = xlog_recover_unmount_trans(log);
 				break;
 			case XLOG_WAS_CONT_TRANS:
 				error = xlog_recover_add_to_cont_trans(log,
@@ -3594,7 +3621,7 @@ xlog_recover_process_data(
 				xfs_warn(log->l_mp, "%s: bad transaction",
 					__func__);
 				ASSERT(0);
-				error = XFS_ERROR(EIO);
+				error = -EIO;
 				break;
 			case 0:
 			case XLOG_CONTINUE_TRANS:
@@ -3605,11 +3632,13 @@ xlog_recover_process_data(
 				xfs_warn(log->l_mp, "%s: bad flag 0x%x",
 					__func__, flags);
 				ASSERT(0);
-				error = XFS_ERROR(EIO);
+				error = -EIO;
 				break;
 			}
-			if (error)
+			if (error) {
+				xlog_recover_free_trans(trans);
 				return error;
+			}
 		}
 		dp += be32_to_cpu(ohead->oh_len);
 		num_logops--;
@@ -3654,7 +3683,7 @@ xlog_recover_process_efi(
 			 */
 			set_bit(XFS_EFI_RECOVERED, &efip->efi_flags);
 			xfs_efi_release(efip, efip->efi_format.efi_nextents);
-			return XFS_ERROR(EIO);
+			return -EIO;
 		}
 	}
 
@@ -3743,7 +3772,7 @@ xlog_recover_process_efis(
 		lip = xfs_trans_ail_cursor_next(ailp, &cur);
 	}
 out:
-	xfs_trans_ail_cursor_done(ailp, &cur);
+	xfs_trans_ail_cursor_done(&cur);
 	spin_unlock(&ailp->xa_lock);
 	return error;
 }
@@ -3954,7 +3983,7 @@ xlog_unpack_data_crc(
 		 * CRC protection by punting an error back up the stack.
 		 */
 		if (xfs_sb_version_hascrc(&log->l_mp->m_sb))
-			return EFSCORRUPTED;
+			return -EFSCORRUPTED;
 	}
 
 	return 0;
@@ -4003,14 +4032,14 @@ xlog_valid_rec_header(
 	if (unlikely(rhead->h_magicno != cpu_to_be32(XLOG_HEADER_MAGIC_NUM))) {
 		XFS_ERROR_REPORT("xlog_valid_rec_header(1)",
 				XFS_ERRLEVEL_LOW, log->l_mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	if (unlikely(
 	    (!rhead->h_version ||
 	    (be32_to_cpu(rhead->h_version) & (~XLOG_VERSION_OKBITS))))) {
 		xfs_warn(log->l_mp, "%s: unrecognised log version (%d).",
 			__func__, be32_to_cpu(rhead->h_version));
-		return XFS_ERROR(EIO);
+		return -EIO;
 	}
 
 	/* LR body must have data or it wouldn't have been written */
@@ -4018,12 +4047,12 @@ xlog_valid_rec_header(
 	if (unlikely( hlen <= 0 || hlen > INT_MAX )) {
 		XFS_ERROR_REPORT("xlog_valid_rec_header(2)",
 				XFS_ERRLEVEL_LOW, log->l_mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	if (unlikely( blkno > log->l_logBBsize || blkno > INT_MAX )) {
 		XFS_ERROR_REPORT("xlog_valid_rec_header(3)",
 				XFS_ERRLEVEL_LOW, log->l_mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -4066,7 +4095,7 @@ xlog_do_recovery_pass(
 		 */
 		hbp = xlog_get_bp(log, 1);
 		if (!hbp)
-			return ENOMEM;
+			return -ENOMEM;
 
 		error = xlog_bread(log, tail_blk, 1, hbp, &offset);
 		if (error)
@@ -4095,11 +4124,11 @@ xlog_do_recovery_pass(
 	}
 
 	if (!hbp)
-		return ENOMEM;
+		return -ENOMEM;
 	dbp = xlog_get_bp(log, BTOBB(h_size));
 	if (!dbp) {
 		xlog_put_bp(hbp);
-		return ENOMEM;
+		return -ENOMEM;
 	}
 
 	memset(rhash, 0, sizeof(rhash));
@@ -4373,7 +4402,7 @@ xlog_do_recover(
 	 * If IO errors happened during recovery, bail out.
 	 */
 	if (XFS_FORCED_SHUTDOWN(log->l_mp)) {
-		return (EIO);
+		return -EIO;
 	}
 
 	/*
@@ -4397,7 +4426,13 @@ xlog_do_recover(
 	XFS_BUF_READ(bp);
 	XFS_BUF_UNASYNC(bp);
 	bp->b_ops = &xfs_sb_buf_ops;
-	xfsbdstrat(log->l_mp, bp);
+
+	if (XFS_FORCED_SHUTDOWN(log->l_mp)) {
+		xfs_buf_relse(bp);
+		return -EIO;
+	}
+
+	xfs_buf_iorequest(bp);
 	error = xfs_buf_iowait(bp);
 	if (error) {
 		xfs_buf_ioerror_alert(bp, __func__);
@@ -4471,7 +4506,7 @@ xlog_recover(
 "Please recover the log on a kernel that supports the unknown features.",
 				(log->l_mp->m_sb.sb_features_log_incompat &
 					XFS_SB_FEAT_INCOMPAT_LOG_UNKNOWN));
-			return EINVAL;
+			return -EINVAL;
 		}
 
 		xfs_notice(log->l_mp, "Starting recovery (logdev: %s)",
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index 02df7b408a26..fbf0384a466f 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -42,6 +42,7 @@
 #include "xfs_trace.h"
 #include "xfs_icache.h"
 #include "xfs_dinode.h"
+#include "xfs_sysfs.h"
 
 
 #ifdef HAVE_PERCPU_SB
@@ -60,6 +61,8 @@ static DEFINE_MUTEX(xfs_uuid_table_mutex);
 static int xfs_uuid_table_size;
 static uuid_t *xfs_uuid_table;
 
+extern struct kset *xfs_kset;
+
 /*
  * 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.
@@ -76,7 +79,7 @@ xfs_uuid_mount(
 
 	if (uuid_is_nil(uuid)) {
 		xfs_warn(mp, "Filesystem has nil UUID - can't mount");
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	mutex_lock(&xfs_uuid_table_mutex);
@@ -104,7 +107,7 @@ xfs_uuid_mount(
  out_duplicate:
 	mutex_unlock(&xfs_uuid_table_mutex);
 	xfs_warn(mp, "Filesystem has duplicate UUID %pU - can't mount", uuid);
-	return XFS_ERROR(EINVAL);
+	return -EINVAL;
 }
 
 STATIC void
@@ -173,13 +176,9 @@ xfs_sb_validate_fsb_count(
 	ASSERT(PAGE_SHIFT >= sbp->sb_blocklog);
 	ASSERT(sbp->sb_blocklog >= BBSHIFT);
 
-#if XFS_BIG_BLKNOS     /* Limited by ULONG_MAX of page cache index */
+	/* Limited by ULONG_MAX of page cache index */
 	if (nblocks >> (PAGE_CACHE_SHIFT - sbp->sb_blocklog) > ULONG_MAX)
-		return EFBIG;
-#else                  /* Limited by UINT_MAX of sectors */
-	if (nblocks << (sbp->sb_blocklog - BBSHIFT) > UINT_MAX)
-		return EFBIG;
-#endif
+		return -EFBIG;
 	return 0;
 }
 
@@ -250,9 +249,9 @@ xfs_initialize_perag(
 		mp->m_flags &= ~XFS_MOUNT_32BITINODES;
 
 	if (mp->m_flags & XFS_MOUNT_32BITINODES)
-		index = xfs_set_inode32(mp);
+		index = xfs_set_inode32(mp, agcount);
 	else
-		index = xfs_set_inode64(mp);
+		index = xfs_set_inode64(mp, agcount);
 
 	if (maxagi)
 		*maxagi = index;
@@ -282,39 +281,59 @@ xfs_readsb(
 	struct xfs_sb	*sbp = &mp->m_sb;
 	int		error;
 	int		loud = !(flags & XFS_MFSI_QUIET);
+	const struct xfs_buf_ops *buf_ops;
 
 	ASSERT(mp->m_sb_bp == NULL);
 	ASSERT(mp->m_ddev_targp != NULL);
 
 	/*
+	 * For the initial read, we must guess at the sector
+	 * size based on the block device.  It's enough to
+	 * get the sb_sectsize out of the superblock and
+	 * then reread with the proper length.
+	 * We don't verify it yet, because it may not be complete.
+	 */
+	sector_size = xfs_getsize_buftarg(mp->m_ddev_targp);
+	buf_ops = NULL;
+
+	/*
 	 * Allocate a (locked) buffer to hold the superblock.
 	 * This will be kept around at all times to optimize
 	 * access to the superblock.
 	 */
-	sector_size = xfs_getsize_buftarg(mp->m_ddev_targp);
-
 reread:
 	bp = xfs_buf_read_uncached(mp->m_ddev_targp, XFS_SB_DADDR,
-				   BTOBB(sector_size), 0,
-				   loud ? &xfs_sb_buf_ops
-				        : &xfs_sb_quiet_buf_ops);
+				   BTOBB(sector_size), 0, buf_ops);
 	if (!bp) {
 		if (loud)
 			xfs_warn(mp, "SB buffer read failed");
-		return EIO;
+		return -EIO;
 	}
 	if (bp->b_error) {
 		error = bp->b_error;
 		if (loud)
 			xfs_warn(mp, "SB validate failed with error %d.", error);
+		/* bad CRC means corrupted metadata */
+		if (error == -EFSBADCRC)
+			error = -EFSCORRUPTED;
 		goto release_buf;
 	}
 
 	/*
 	 * Initialize the mount structure from the superblock.
 	 */
-	xfs_sb_from_disk(&mp->m_sb, XFS_BUF_TO_SBP(bp));
-	xfs_sb_quota_from_disk(&mp->m_sb);
+	xfs_sb_from_disk(sbp, XFS_BUF_TO_SBP(bp));
+
+	/*
+	 * If we haven't validated the superblock, do so now before we try
+	 * to check the sector size and reread the superblock appropriately.
+	 */
+	if (sbp->sb_magicnum != XFS_SB_MAGIC) {
+		if (loud)
+			xfs_warn(mp, "Invalid superblock magic number");
+		error = -EINVAL;
+		goto release_buf;
+	}
 
 	/*
 	 * We must be able to do sector-sized and sector-aligned IO.
@@ -323,17 +342,18 @@ reread:
 		if (loud)
 			xfs_warn(mp, "device supports %u byte sectors (not %u)",
 				sector_size, sbp->sb_sectsize);
-		error = ENOSYS;
+		error = -ENOSYS;
 		goto release_buf;
 	}
 
-	/*
-	 * If device sector size is smaller than the superblock size,
-	 * re-read the superblock so the buffer is correctly sized.
-	 */
-	if (sector_size < sbp->sb_sectsize) {
+	if (buf_ops == NULL) {
+		/*
+		 * Re-read the superblock so the buffer is correctly sized,
+		 * and properly verified.
+		 */
 		xfs_buf_relse(bp);
 		sector_size = sbp->sb_sectsize;
+		buf_ops = loud ? &xfs_sb_buf_ops : &xfs_sb_quiet_buf_ops;
 		goto reread;
 	}
 
@@ -370,7 +390,7 @@ xfs_update_alignment(xfs_mount_t *mp)
 			xfs_warn(mp,
 		"alignment check failed: sunit/swidth vs. blocksize(%d)",
 				sbp->sb_blocksize);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		} else {
 			/*
 			 * Convert the stripe unit and width to FSBs.
@@ -380,14 +400,14 @@ xfs_update_alignment(xfs_mount_t *mp)
 				xfs_warn(mp,
 			"alignment check failed: sunit/swidth vs. agsize(%d)",
 					 sbp->sb_agblocks);
-				return XFS_ERROR(EINVAL);
+				return -EINVAL;
 			} else if (mp->m_dalign) {
 				mp->m_swidth = XFS_BB_TO_FSBT(mp, mp->m_swidth);
 			} else {
 				xfs_warn(mp,
 			"alignment check failed: sunit(%d) less than bsize(%d)",
 					 mp->m_dalign, sbp->sb_blocksize);
-				return XFS_ERROR(EINVAL);
+				return -EINVAL;
 			}
 		}
 
@@ -407,7 +427,7 @@ xfs_update_alignment(xfs_mount_t *mp)
 		} else {
 			xfs_warn(mp,
 	"cannot change alignment: superblock does not support data alignment");
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 	} else if ((mp->m_flags & XFS_MOUNT_NOALIGN) != XFS_MOUNT_NOALIGN &&
 		    xfs_sb_version_hasdalign(&mp->m_sb)) {
@@ -534,14 +554,14 @@ xfs_check_sizes(xfs_mount_t *mp)
 	d = (xfs_daddr_t)XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks);
 	if (XFS_BB_TO_FSB(mp, d) != mp->m_sb.sb_dblocks) {
 		xfs_warn(mp, "filesystem size mismatch detected");
-		return XFS_ERROR(EFBIG);
+		return -EFBIG;
 	}
 	bp = xfs_buf_read_uncached(mp->m_ddev_targp,
 					d - XFS_FSS_TO_BB(mp, 1),
 					XFS_FSS_TO_BB(mp, 1), 0, NULL);
 	if (!bp) {
 		xfs_warn(mp, "last sector read failed");
-		return EIO;
+		return -EIO;
 	}
 	xfs_buf_relse(bp);
 
@@ -549,14 +569,14 @@ xfs_check_sizes(xfs_mount_t *mp)
 		d = (xfs_daddr_t)XFS_FSB_TO_BB(mp, mp->m_sb.sb_logblocks);
 		if (XFS_BB_TO_FSB(mp, d) != mp->m_sb.sb_logblocks) {
 			xfs_warn(mp, "log size mismatch detected");
-			return XFS_ERROR(EFBIG);
+			return -EFBIG;
 		}
 		bp = xfs_buf_read_uncached(mp->m_logdev_targp,
 					d - XFS_FSB_TO_BB(mp, 1),
 					XFS_FSB_TO_BB(mp, 1), 0, NULL);
 		if (!bp) {
 			xfs_warn(mp, "log device read failed");
-			return EIO;
+			return -EIO;
 		}
 		xfs_buf_relse(bp);
 	}
@@ -686,6 +706,12 @@ xfs_mountfs(
 			mp->m_update_flags |= XFS_SB_VERSIONNUM;
 	}
 
+	/* always use v2 inodes by default now */
+	if (!(mp->m_sb.sb_versionnum & XFS_SB_VERSION_NLINKBIT)) {
+		mp->m_sb.sb_versionnum |= XFS_SB_VERSION_NLINKBIT;
+		mp->m_update_flags |= XFS_SB_VERSIONNUM;
+	}
+
 	/*
 	 * Check if sb_agblocks is aligned at stripe boundary
 	 * If sb_agblocks is NOT aligned turn off m_dalign since
@@ -703,10 +729,15 @@ xfs_mountfs(
 
 	xfs_set_maxicount(mp);
 
-	error = xfs_uuid_mount(mp);
+	mp->m_kobj.kobject.kset = xfs_kset;
+	error = xfs_sysfs_init(&mp->m_kobj, &xfs_mp_ktype, NULL, mp->m_fsname);
 	if (error)
 		goto out;
 
+	error = xfs_uuid_mount(mp);
+	if (error)
+		goto out_remove_sysfs;
+
 	/*
 	 * Set the minimum read and write sizes
 	 */
@@ -732,8 +763,6 @@ xfs_mountfs(
 		new_size *= mp->m_sb.sb_inodesize / XFS_DINODE_MIN_SIZE;
 		if (mp->m_sb.sb_inoalignmt >= XFS_B_TO_FSBT(mp, new_size))
 			mp->m_inode_cluster_size = new_size;
-		xfs_info(mp, "Using inode cluster size of %d bytes",
-			 mp->m_inode_cluster_size);
 	}
 
 	/*
@@ -765,12 +794,11 @@ xfs_mountfs(
 
 	mp->m_dmevmask = 0;	/* not persistent; set after each mount */
 
-	xfs_dir_mount(mp);
-
-	/*
-	 * Initialize the attribute manager's entries.
-	 */
-	mp->m_attr_magicpct = (mp->m_sb.sb_blocksize * 37) / 100;
+	error = xfs_da_mount(mp);
+	if (error) {
+		xfs_warn(mp, "Failed dir/attr init: %d", error);
+		goto out_remove_uuid;
+	}
 
 	/*
 	 * Initialize the precomputed transaction reservations values.
@@ -785,13 +813,13 @@ xfs_mountfs(
 	error = xfs_initialize_perag(mp, sbp->sb_agcount, &mp->m_maxagi);
 	if (error) {
 		xfs_warn(mp, "Failed per-ag init: %d", error);
-		goto out_remove_uuid;
+		goto out_free_dir;
 	}
 
 	if (!sbp->sb_logblocks) {
 		xfs_warn(mp, "no log defined");
 		XFS_ERROR_REPORT("xfs_mountfs", XFS_ERRLEVEL_LOW, mp);
-		error = XFS_ERROR(EFSCORRUPTED);
+		error = -EFSCORRUPTED;
 		goto out_free_perag;
 	}
 
@@ -830,7 +858,7 @@ xfs_mountfs(
 	     !mp->m_sb.sb_inprogress) {
 		error = xfs_initialize_perag_data(mp, sbp->sb_agcount);
 		if (error)
-			goto out_fail_wait;
+			goto out_log_dealloc;
 	}
 
 	/*
@@ -851,7 +879,7 @@ xfs_mountfs(
 		xfs_iunlock(rip, XFS_ILOCK_EXCL);
 		XFS_ERROR_REPORT("xfs_mountfs_int(2)", XFS_ERRLEVEL_LOW,
 				 mp);
-		error = XFS_ERROR(EFSCORRUPTED);
+		error = -EFSCORRUPTED;
 		goto out_rele_rip;
 	}
 	mp->m_rootip = rip;	/* save it */
@@ -902,7 +930,7 @@ xfs_mountfs(
 			xfs_notice(mp, "resetting quota flags");
 			error = xfs_mount_reset_sbqflags(mp);
 			if (error)
-				return error;
+				goto out_rtunmount;
 		}
 	}
 
@@ -960,8 +988,12 @@ xfs_mountfs(
 	xfs_wait_buftarg(mp->m_ddev_targp);
  out_free_perag:
 	xfs_free_perag(mp);
+ out_free_dir:
+	xfs_da_unmount(mp);
  out_remove_uuid:
 	xfs_uuid_unmount(mp);
+ out_remove_sysfs:
+	xfs_sysfs_del(&mp->m_kobj);
  out:
 	return error;
 }
@@ -1037,12 +1069,15 @@ xfs_unmountfs(
 				"Freespace may not be correct on next mount.");
 
 	xfs_log_unmount(mp);
+	xfs_da_unmount(mp);
 	xfs_uuid_unmount(mp);
 
 #if defined(DEBUG)
 	xfs_errortag_clearall(mp, 0);
 #endif
 	xfs_free_perag(mp);
+
+	xfs_sysfs_del(&mp->m_kobj);
 }
 
 int
@@ -1124,7 +1159,7 @@ xfs_mod_incore_sb_unlocked(
 		lcounter += delta;
 		if (lcounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_icount = lcounter;
 		return 0;
@@ -1133,7 +1168,7 @@ xfs_mod_incore_sb_unlocked(
 		lcounter += delta;
 		if (lcounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_ifree = lcounter;
 		return 0;
@@ -1163,7 +1198,7 @@ xfs_mod_incore_sb_unlocked(
 			 * blocks if were allowed to.
 			 */
 			if (!rsvd)
-				return XFS_ERROR(ENOSPC);
+				return -ENOSPC;
 
 			lcounter = (long long)mp->m_resblks_avail + delta;
 			if (lcounter >= 0) {
@@ -1174,7 +1209,7 @@ xfs_mod_incore_sb_unlocked(
 				"Filesystem \"%s\": reserve blocks depleted! "
 				"Consider increasing reserve pool size.",
 				mp->m_fsname);
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		}
 
 		mp->m_sb.sb_fdblocks = lcounter + XFS_ALLOC_SET_ASIDE(mp);
@@ -1183,7 +1218,7 @@ xfs_mod_incore_sb_unlocked(
 		lcounter = (long long)mp->m_sb.sb_frextents;
 		lcounter += delta;
 		if (lcounter < 0) {
-			return XFS_ERROR(ENOSPC);
+			return -ENOSPC;
 		}
 		mp->m_sb.sb_frextents = lcounter;
 		return 0;
@@ -1192,7 +1227,7 @@ xfs_mod_incore_sb_unlocked(
 		lcounter += delta;
 		if (lcounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_dblocks = lcounter;
 		return 0;
@@ -1201,7 +1236,7 @@ xfs_mod_incore_sb_unlocked(
 		scounter += delta;
 		if (scounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_agcount = scounter;
 		return 0;
@@ -1210,7 +1245,7 @@ xfs_mod_incore_sb_unlocked(
 		scounter += delta;
 		if (scounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_imax_pct = scounter;
 		return 0;
@@ -1219,7 +1254,7 @@ xfs_mod_incore_sb_unlocked(
 		scounter += delta;
 		if (scounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_rextsize = scounter;
 		return 0;
@@ -1228,7 +1263,7 @@ xfs_mod_incore_sb_unlocked(
 		scounter += delta;
 		if (scounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_rbmblocks = scounter;
 		return 0;
@@ -1237,7 +1272,7 @@ xfs_mod_incore_sb_unlocked(
 		lcounter += delta;
 		if (lcounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_rblocks = lcounter;
 		return 0;
@@ -1246,7 +1281,7 @@ xfs_mod_incore_sb_unlocked(
 		lcounter += delta;
 		if (lcounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_rextents = lcounter;
 		return 0;
@@ -1255,13 +1290,13 @@ xfs_mod_incore_sb_unlocked(
 		scounter += delta;
 		if (scounter < 0) {
 			ASSERT(0);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 		mp->m_sb.sb_rextslog = scounter;
 		return 0;
 	default:
 		ASSERT(0);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 }
 
@@ -1424,7 +1459,7 @@ xfs_dev_is_read_only(
 	    (mp->m_rtdev_targp && xfs_readonly_buftarg(mp->m_rtdev_targp))) {
 		xfs_notice(mp, "%s required on read-only device.", message);
 		xfs_notice(mp, "write access unavailable, cannot proceed.");
-		return EROFS;
+		return -EROFS;
 	}
 	return 0;
 }
@@ -1967,7 +2002,7 @@ slow_path:
 	 * (e.g. lots of space just got freed). After that
 	 * we are done.
 	 */
-	if (ret != ENOSPC)
+	if (ret != -ENOSPC)
 		xfs_icsb_balance_counter(mp, field, 0);
 	xfs_icsb_unlock(mp);
 	return ret;
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index a466c5e5826e..b0447c86e7e2 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -27,6 +27,7 @@ struct xfs_nameops;
 struct xfs_ail;
 struct xfs_quotainfo;
 struct xfs_dir_ops;
+struct xfs_da_geometry;
 
 #ifdef HAVE_PERCPU_SB
 
@@ -96,6 +97,8 @@ typedef struct xfs_mount {
 	uint			m_readio_blocks; /* min read size blocks */
 	uint			m_writeio_log;	/* min write size log bytes */
 	uint			m_writeio_blocks; /* min write size blocks */
+	struct xfs_da_geometry	*m_dir_geo;	/* directory block geometry */
+	struct xfs_da_geometry	*m_attr_geo;	/* attribute block geometry */
 	struct xlog		*m_log;		/* log specific stuff */
 	int			m_logbufs;	/* number of log buffers */
 	int			m_logbsize;	/* size of each log buffer */
@@ -131,8 +134,6 @@ typedef struct xfs_mount {
 	int			m_fixedfsid[2];	/* unchanged for life of FS */
 	uint			m_dmevmask;	/* DMI events for this FS */
 	__uint64_t		m_flags;	/* global mount flags */
-	uint			m_dir_node_ents; /* #entries in a dir danode */
-	uint			m_attr_node_ents; /* #entries in attr danode */
 	int			m_ialloc_inos;	/* inodes in inode allocation */
 	int			m_ialloc_blks;	/* blocks in inode allocation */
 	int			m_inoalign_mask;/* mask sb_inoalignmt if used */
@@ -145,17 +146,10 @@ typedef struct xfs_mount {
 	int			m_dalign;	/* stripe unit */
 	int			m_swidth;	/* stripe width */
 	int			m_sinoalign;	/* stripe unit inode alignment */
-	int			m_attr_magicpct;/* 37% of the blocksize */
-	int			m_dir_magicpct;	/* 37% of the dir blocksize */
 	__uint8_t		m_sectbb_log;	/* sectlog - BBSHIFT */
 	const struct xfs_nameops *m_dirnameops;	/* vector of dir name ops */
 	const struct xfs_dir_ops *m_dir_inode_ops; /* vector of dir inode ops */
 	const struct xfs_dir_ops *m_nondir_inode_ops; /* !dir inode ops */
-	int			m_dirblksize;	/* directory block sz--bytes */
-	int			m_dirblkfsbs;	/* directory block sz--fsbs */
-	xfs_dablk_t		m_dirdatablk;	/* blockno of dir data v2 */
-	xfs_dablk_t		m_dirleafblk;	/* blockno of dir non-data v2 */
-	xfs_dablk_t		m_dirfreeblk;	/* blockno of dirfreeindex v2 */
 	uint			m_chsize;	/* size of next field */
 	atomic_t		m_active_trans;	/* number trans frozen */
 #ifdef HAVE_PERCPU_SB
@@ -172,6 +166,7 @@ typedef struct xfs_mount {
 						   on the next remount,rw */
 	int64_t			m_low_space[XFS_LOWSP_MAX];
 						/* low free space thresholds */
+	struct xfs_kobj		m_kobj;
 
 	struct workqueue_struct	*m_data_workqueue;
 	struct workqueue_struct	*m_unwritten_workqueue;
diff --git a/fs/xfs/xfs_mru_cache.c b/fs/xfs/xfs_mru_cache.c
index 4aff56395732..1eb6f3df698c 100644
--- a/fs/xfs/xfs_mru_cache.c
+++ b/fs/xfs/xfs_mru_cache.c
@@ -100,14 +100,20 @@
  * likely result in a loop in one of the lists.  That's a sure-fire recipe for
  * an infinite loop in the code.
  */
-typedef struct xfs_mru_cache_elem
-{
-	struct list_head list_node;
-	unsigned long	key;
-	void		*value;
-} xfs_mru_cache_elem_t;
+struct xfs_mru_cache {
+	struct radix_tree_root	store;     /* Core storage data structure.  */
+	struct list_head	*lists;    /* Array of lists, one per grp.  */
+	struct list_head	reap_list; /* Elements overdue for reaping. */
+	spinlock_t		lock;      /* Lock to protect this struct.  */
+	unsigned int		grp_count; /* Number of discrete groups.    */
+	unsigned int		grp_time;  /* Time period spanned by grps.  */
+	unsigned int		lru_grp;   /* Group containing time zero.   */
+	unsigned long		time_zero; /* Time first element was added. */
+	xfs_mru_cache_free_func_t free_func; /* Function pointer for freeing. */
+	struct delayed_work	work;      /* Workqueue data for reaping.   */
+	unsigned int		queued;	   /* work has been queued */
+};
 
-static kmem_zone_t		*xfs_mru_elem_zone;
 static struct workqueue_struct	*xfs_mru_reap_wq;
 
 /*
@@ -129,12 +135,12 @@ static struct workqueue_struct	*xfs_mru_reap_wq;
  */
 STATIC unsigned long
 _xfs_mru_cache_migrate(
-	xfs_mru_cache_t	*mru,
-	unsigned long	now)
+	struct xfs_mru_cache	*mru,
+	unsigned long		now)
 {
-	unsigned int	grp;
-	unsigned int	migrated = 0;
-	struct list_head *lru_list;
+	unsigned int		grp;
+	unsigned int		migrated = 0;
+	struct list_head	*lru_list;
 
 	/* Nothing to do if the data store is empty. */
 	if (!mru->time_zero)
@@ -193,11 +199,11 @@ _xfs_mru_cache_migrate(
  */
 STATIC void
 _xfs_mru_cache_list_insert(
-	xfs_mru_cache_t		*mru,
-	xfs_mru_cache_elem_t	*elem)
+	struct xfs_mru_cache	*mru,
+	struct xfs_mru_cache_elem *elem)
 {
-	unsigned int	grp = 0;
-	unsigned long	now = jiffies;
+	unsigned int		grp = 0;
+	unsigned long		now = jiffies;
 
 	/*
 	 * If the data store is empty, initialise time zero, leave grp set to
@@ -231,10 +237,10 @@ _xfs_mru_cache_list_insert(
  */
 STATIC void
 _xfs_mru_cache_clear_reap_list(
-	xfs_mru_cache_t		*mru) __releases(mru->lock) __acquires(mru->lock)
-
+	struct xfs_mru_cache	*mru)
+		__releases(mru->lock) __acquires(mru->lock)
 {
-	xfs_mru_cache_elem_t	*elem, *next;
+	struct xfs_mru_cache_elem *elem, *next;
 	struct list_head	tmp;
 
 	INIT_LIST_HEAD(&tmp);
@@ -252,15 +258,8 @@ _xfs_mru_cache_clear_reap_list(
 	spin_unlock(&mru->lock);
 
 	list_for_each_entry_safe(elem, next, &tmp, list_node) {
-
-		/* Remove the element from the reap list. */
 		list_del_init(&elem->list_node);
-
-		/* Call the client's free function with the key and value pointer. */
-		mru->free_func(elem->key, elem->value);
-
-		/* Free the element structure. */
-		kmem_zone_free(xfs_mru_elem_zone, elem);
+		mru->free_func(elem);
 	}
 
 	spin_lock(&mru->lock);
@@ -277,7 +276,8 @@ STATIC void
 _xfs_mru_cache_reap(
 	struct work_struct	*work)
 {
-	xfs_mru_cache_t		*mru = container_of(work, xfs_mru_cache_t, work.work);
+	struct xfs_mru_cache	*mru =
+		container_of(work, struct xfs_mru_cache, work.work);
 	unsigned long		now, next;
 
 	ASSERT(mru && mru->lists);
@@ -304,28 +304,16 @@ _xfs_mru_cache_reap(
 int
 xfs_mru_cache_init(void)
 {
-	xfs_mru_elem_zone = kmem_zone_init(sizeof(xfs_mru_cache_elem_t),
-	                                 "xfs_mru_cache_elem");
-	if (!xfs_mru_elem_zone)
-		goto out;
-
 	xfs_mru_reap_wq = alloc_workqueue("xfs_mru_cache", WQ_MEM_RECLAIM, 1);
 	if (!xfs_mru_reap_wq)
-		goto out_destroy_mru_elem_zone;
-
+		return -ENOMEM;
 	return 0;
-
- out_destroy_mru_elem_zone:
-	kmem_zone_destroy(xfs_mru_elem_zone);
- out:
-	return -ENOMEM;
 }
 
 void
 xfs_mru_cache_uninit(void)
 {
 	destroy_workqueue(xfs_mru_reap_wq);
-	kmem_zone_destroy(xfs_mru_elem_zone);
 }
 
 /*
@@ -336,33 +324,33 @@ xfs_mru_cache_uninit(void)
  */
 int
 xfs_mru_cache_create(
-	xfs_mru_cache_t		**mrup,
+	struct xfs_mru_cache	**mrup,
 	unsigned int		lifetime_ms,
 	unsigned int		grp_count,
 	xfs_mru_cache_free_func_t free_func)
 {
-	xfs_mru_cache_t	*mru = NULL;
-	int		err = 0, grp;
-	unsigned int	grp_time;
+	struct xfs_mru_cache	*mru = NULL;
+	int			err = 0, grp;
+	unsigned int		grp_time;
 
 	if (mrup)
 		*mrup = NULL;
 
 	if (!mrup || !grp_count || !lifetime_ms || !free_func)
-		return EINVAL;
+		return -EINVAL;
 
 	if (!(grp_time = msecs_to_jiffies(lifetime_ms) / grp_count))
-		return EINVAL;
+		return -EINVAL;
 
 	if (!(mru = kmem_zalloc(sizeof(*mru), KM_SLEEP)))
-		return ENOMEM;
+		return -ENOMEM;
 
 	/* An extra list is needed to avoid reaping up to a grp_time early. */
 	mru->grp_count = grp_count + 1;
 	mru->lists = kmem_zalloc(mru->grp_count * sizeof(*mru->lists), KM_SLEEP);
 
 	if (!mru->lists) {
-		err = ENOMEM;
+		err = -ENOMEM;
 		goto exit;
 	}
 
@@ -400,7 +388,7 @@ exit:
  */
 static void
 xfs_mru_cache_flush(
-	xfs_mru_cache_t		*mru)
+	struct xfs_mru_cache	*mru)
 {
 	if (!mru || !mru->lists)
 		return;
@@ -420,7 +408,7 @@ xfs_mru_cache_flush(
 
 void
 xfs_mru_cache_destroy(
-	xfs_mru_cache_t		*mru)
+	struct xfs_mru_cache	*mru)
 {
 	if (!mru || !mru->lists)
 		return;
@@ -438,38 +426,30 @@ xfs_mru_cache_destroy(
  */
 int
 xfs_mru_cache_insert(
-	xfs_mru_cache_t	*mru,
-	unsigned long	key,
-	void		*value)
+	struct xfs_mru_cache	*mru,
+	unsigned long		key,
+	struct xfs_mru_cache_elem *elem)
 {
-	xfs_mru_cache_elem_t *elem;
+	int			error;
 
 	ASSERT(mru && mru->lists);
 	if (!mru || !mru->lists)
-		return EINVAL;
+		return -EINVAL;
 
-	elem = kmem_zone_zalloc(xfs_mru_elem_zone, KM_SLEEP);
-	if (!elem)
-		return ENOMEM;
-
-	if (radix_tree_preload(GFP_KERNEL)) {
-		kmem_zone_free(xfs_mru_elem_zone, elem);
-		return ENOMEM;
-	}
+	if (radix_tree_preload(GFP_KERNEL))
+		return -ENOMEM;
 
 	INIT_LIST_HEAD(&elem->list_node);
 	elem->key = key;
-	elem->value = value;
 
 	spin_lock(&mru->lock);
-
-	radix_tree_insert(&mru->store, key, elem);
+	error = radix_tree_insert(&mru->store, key, elem);
 	radix_tree_preload_end();
-	_xfs_mru_cache_list_insert(mru, elem);
-
+	if (!error)
+		_xfs_mru_cache_list_insert(mru, elem);
 	spin_unlock(&mru->lock);
 
-	return 0;
+	return error;
 }
 
 /*
@@ -478,13 +458,12 @@ xfs_mru_cache_insert(
  * the client data pointer for the removed element is returned, otherwise this
  * function will return a NULL pointer.
  */
-void *
+struct xfs_mru_cache_elem *
 xfs_mru_cache_remove(
-	xfs_mru_cache_t	*mru,
-	unsigned long	key)
+	struct xfs_mru_cache	*mru,
+	unsigned long		key)
 {
-	xfs_mru_cache_elem_t *elem;
-	void		*value = NULL;
+	struct xfs_mru_cache_elem *elem;
 
 	ASSERT(mru && mru->lists);
 	if (!mru || !mru->lists)
@@ -492,17 +471,11 @@ xfs_mru_cache_remove(
 
 	spin_lock(&mru->lock);
 	elem = radix_tree_delete(&mru->store, key);
-	if (elem) {
-		value = elem->value;
+	if (elem)
 		list_del(&elem->list_node);
-	}
-
 	spin_unlock(&mru->lock);
 
-	if (elem)
-		kmem_zone_free(xfs_mru_elem_zone, elem);
-
-	return value;
+	return elem;
 }
 
 /*
@@ -511,13 +484,14 @@ xfs_mru_cache_remove(
  */
 void
 xfs_mru_cache_delete(
-	xfs_mru_cache_t	*mru,
-	unsigned long	key)
+	struct xfs_mru_cache	*mru,
+	unsigned long		key)
 {
-	void		*value = xfs_mru_cache_remove(mru, key);
+	struct xfs_mru_cache_elem *elem;
 
-	if (value)
-		mru->free_func(key, value);
+	elem = xfs_mru_cache_remove(mru, key);
+	if (elem)
+		mru->free_func(elem);
 }
 
 /*
@@ -540,12 +514,12 @@ xfs_mru_cache_delete(
  * status, we need to help it get it right by annotating the path that does
  * not release the lock.
  */
-void *
+struct xfs_mru_cache_elem *
 xfs_mru_cache_lookup(
-	xfs_mru_cache_t	*mru,
-	unsigned long	key)
+	struct xfs_mru_cache	*mru,
+	unsigned long		key)
 {
-	xfs_mru_cache_elem_t *elem;
+	struct xfs_mru_cache_elem *elem;
 
 	ASSERT(mru && mru->lists);
 	if (!mru || !mru->lists)
@@ -560,7 +534,7 @@ xfs_mru_cache_lookup(
 	} else
 		spin_unlock(&mru->lock);
 
-	return elem ? elem->value : NULL;
+	return elem;
 }
 
 /*
@@ -570,7 +544,8 @@ xfs_mru_cache_lookup(
  */
 void
 xfs_mru_cache_done(
-	xfs_mru_cache_t	*mru) __releases(mru->lock)
+	struct xfs_mru_cache	*mru)
+		__releases(mru->lock)
 {
 	spin_unlock(&mru->lock);
 }
diff --git a/fs/xfs/xfs_mru_cache.h b/fs/xfs/xfs_mru_cache.h
index 36dd3ec8b4eb..fb5245ba5ff7 100644
--- a/fs/xfs/xfs_mru_cache.h
+++ b/fs/xfs/xfs_mru_cache.h
@@ -18,24 +18,15 @@
 #ifndef __XFS_MRU_CACHE_H__
 #define __XFS_MRU_CACHE_H__
 
+struct xfs_mru_cache;
 
-/* Function pointer type for callback to free a client's data pointer. */
-typedef void (*xfs_mru_cache_free_func_t)(unsigned long, void*);
+struct xfs_mru_cache_elem {
+	struct list_head list_node;
+	unsigned long	key;
+};
 
-typedef struct xfs_mru_cache
-{
-	struct radix_tree_root	store;     /* Core storage data structure.  */
-	struct list_head	*lists;    /* Array of lists, one per grp.  */
-	struct list_head	reap_list; /* Elements overdue for reaping. */
-	spinlock_t		lock;      /* Lock to protect this struct.  */
-	unsigned int		grp_count; /* Number of discrete groups.    */
-	unsigned int		grp_time;  /* Time period spanned by grps.  */
-	unsigned int		lru_grp;   /* Group containing time zero.   */
-	unsigned long		time_zero; /* Time first element was added. */
-	xfs_mru_cache_free_func_t free_func; /* Function pointer for freeing. */
-	struct delayed_work	work;      /* Workqueue data for reaping.   */
-	unsigned int		queued;	   /* work has been queued */
-} xfs_mru_cache_t;
+/* Function pointer type for callback to free a client's data pointer. */
+typedef void (*xfs_mru_cache_free_func_t)(struct xfs_mru_cache_elem *elem);
 
 int xfs_mru_cache_init(void);
 void xfs_mru_cache_uninit(void);
@@ -44,10 +35,12 @@ int xfs_mru_cache_create(struct xfs_mru_cache **mrup, unsigned int lifetime_ms,
 			     xfs_mru_cache_free_func_t free_func);
 void xfs_mru_cache_destroy(struct xfs_mru_cache *mru);
 int xfs_mru_cache_insert(struct xfs_mru_cache *mru, unsigned long key,
-				void *value);
-void * xfs_mru_cache_remove(struct xfs_mru_cache *mru, unsigned long key);
+		struct xfs_mru_cache_elem *elem);
+struct xfs_mru_cache_elem *
+xfs_mru_cache_remove(struct xfs_mru_cache *mru, unsigned long key);
 void xfs_mru_cache_delete(struct xfs_mru_cache *mru, unsigned long key);
-void *xfs_mru_cache_lookup(struct xfs_mru_cache *mru, unsigned long key);
+struct xfs_mru_cache_elem *
+xfs_mru_cache_lookup(struct xfs_mru_cache *mru, unsigned long key);
 void xfs_mru_cache_done(struct xfs_mru_cache *mru);
 
 #endif /* __XFS_MRU_CACHE_H__ */
diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c
index 14a4996cfec6..10232102b4a6 100644
--- a/fs/xfs/xfs_qm.c
+++ b/fs/xfs/xfs_qm.c
@@ -98,18 +98,18 @@ restart:
 			next_index = be32_to_cpu(dqp->q_core.d_id) + 1;
 
 			error = execute(batch[i], data);
-			if (error == EAGAIN) {
+			if (error == -EAGAIN) {
 				skipped++;
 				continue;
 			}
-			if (error && last_error != EFSCORRUPTED)
+			if (error && last_error != -EFSCORRUPTED)
 				last_error = error;
 		}
 
 		mutex_unlock(&qi->qi_tree_lock);
 
 		/* bail out if the filesystem is corrupted.  */
-		if (last_error == EFSCORRUPTED) {
+		if (last_error == -EFSCORRUPTED) {
 			skipped = 0;
 			break;
 		}
@@ -134,28 +134,11 @@ xfs_qm_dqpurge(
 {
 	struct xfs_mount	*mp = dqp->q_mount;
 	struct xfs_quotainfo	*qi = mp->m_quotainfo;
-	struct xfs_dquot	*gdqp = NULL;
-	struct xfs_dquot	*pdqp = NULL;
 
 	xfs_dqlock(dqp);
 	if ((dqp->dq_flags & XFS_DQ_FREEING) || dqp->q_nrefs != 0) {
 		xfs_dqunlock(dqp);
-		return EAGAIN;
-	}
-
-	/*
-	 * If this quota has a hint attached, prepare for releasing it now.
-	 */
-	gdqp = dqp->q_gdquot;
-	if (gdqp) {
-		xfs_dqlock(gdqp);
-		dqp->q_gdquot = NULL;
-	}
-
-	pdqp = dqp->q_pdquot;
-	if (pdqp) {
-		xfs_dqlock(pdqp);
-		dqp->q_pdquot = NULL;
+		return -EAGAIN;
 	}
 
 	dqp->dq_flags |= XFS_DQ_FREEING;
@@ -206,11 +189,6 @@ xfs_qm_dqpurge(
 	XFS_STATS_DEC(xs_qm_dquot_unused);
 
 	xfs_qm_dqdestroy(dqp);
-
-	if (gdqp)
-		xfs_qm_dqput(gdqp);
-	if (pdqp)
-		xfs_qm_dqput(pdqp);
 	return 0;
 }
 
@@ -243,100 +221,6 @@ xfs_qm_unmount(
 	}
 }
 
-
-/*
- * This is called from xfs_mountfs to start quotas and initialize all
- * necessary data structures like quotainfo.  This is also responsible for
- * running a quotacheck as necessary.  We are guaranteed that the superblock
- * is consistently read in at this point.
- *
- * If we fail here, the mount will continue with quota turned off. We don't
- * need to inidicate success or failure at all.
- */
-void
-xfs_qm_mount_quotas(
-	xfs_mount_t	*mp)
-{
-	int		error = 0;
-	uint		sbf;
-
-	/*
-	 * If quotas on realtime volumes is not supported, we disable
-	 * quotas immediately.
-	 */
-	if (mp->m_sb.sb_rextents) {
-		xfs_notice(mp, "Cannot turn on quotas for realtime filesystem");
-		mp->m_qflags = 0;
-		goto write_changes;
-	}
-
-	ASSERT(XFS_IS_QUOTA_RUNNING(mp));
-
-	/*
-	 * Allocate the quotainfo structure inside the mount struct, and
-	 * create quotainode(s), and change/rev superblock if necessary.
-	 */
-	error = xfs_qm_init_quotainfo(mp);
-	if (error) {
-		/*
-		 * We must turn off quotas.
-		 */
-		ASSERT(mp->m_quotainfo == NULL);
-		mp->m_qflags = 0;
-		goto write_changes;
-	}
-	/*
-	 * If any of the quotas are not consistent, do a quotacheck.
-	 */
-	if (XFS_QM_NEED_QUOTACHECK(mp)) {
-		error = xfs_qm_quotacheck(mp);
-		if (error) {
-			/* Quotacheck failed and disabled quotas. */
-			return;
-		}
-	}
-	/* 
-	 * If one type of quotas is off, then it will lose its
-	 * quotachecked status, since we won't be doing accounting for
-	 * that type anymore.
-	 */
-	if (!XFS_IS_UQUOTA_ON(mp))
-		mp->m_qflags &= ~XFS_UQUOTA_CHKD;
-	if (!XFS_IS_GQUOTA_ON(mp))
-		mp->m_qflags &= ~XFS_GQUOTA_CHKD;
-	if (!XFS_IS_PQUOTA_ON(mp))
-		mp->m_qflags &= ~XFS_PQUOTA_CHKD;
-
- write_changes:
-	/*
-	 * We actually don't have to acquire the m_sb_lock at all.
-	 * This can only be called from mount, and that's single threaded. XXX
-	 */
-	spin_lock(&mp->m_sb_lock);
-	sbf = mp->m_sb.sb_qflags;
-	mp->m_sb.sb_qflags = mp->m_qflags & XFS_MOUNT_QUOTA_ALL;
-	spin_unlock(&mp->m_sb_lock);
-
-	if (sbf != (mp->m_qflags & XFS_MOUNT_QUOTA_ALL)) {
-		if (xfs_qm_write_sb_changes(mp, XFS_SB_QFLAGS)) {
-			/*
-			 * We could only have been turning quotas off.
-			 * We aren't in very good shape actually because
-			 * the incore structures are convinced that quotas are
-			 * off, but the on disk superblock doesn't know that !
-			 */
-			ASSERT(!(XFS_IS_QUOTA_RUNNING(mp)));
-			xfs_alert(mp, "%s: Superblock update failed!",
-				__func__);
-		}
-	}
-
-	if (error) {
-		xfs_warn(mp, "Failed to initialize disk quotas.");
-		return;
-	}
-}
-
 /*
  * Called from the vfsops layer.
  */
@@ -380,7 +264,6 @@ xfs_qm_dqattach_one(
 	xfs_dqid_t	id,
 	uint		type,
 	uint		doalloc,
-	xfs_dquot_t	*udqhint, /* hint */
 	xfs_dquot_t	**IO_idqpp)
 {
 	xfs_dquot_t	*dqp;
@@ -390,9 +273,9 @@ xfs_qm_dqattach_one(
 	error = 0;
 
 	/*
-	 * See if we already have it in the inode itself. IO_idqpp is
-	 * &i_udquot or &i_gdquot. This made the code look weird, but
-	 * made the logic a lot simpler.
+	 * See if we already have it in the inode itself. IO_idqpp is &i_udquot
+	 * or &i_gdquot. This made the code look weird, but made the logic a lot
+	 * simpler.
 	 */
 	dqp = *IO_idqpp;
 	if (dqp) {
@@ -401,49 +284,10 @@ xfs_qm_dqattach_one(
 	}
 
 	/*
-	 * udqhint is the i_udquot field in inode, and is non-NULL only
-	 * when the type arg is group/project. Its purpose is to save a
-	 * lookup by dqid (xfs_qm_dqget) by caching a group dquot inside
-	 * the user dquot.
-	 */
-	if (udqhint) {
-		ASSERT(type == XFS_DQ_GROUP || type == XFS_DQ_PROJ);
-		xfs_dqlock(udqhint);
-
-		/*
-		 * No need to take dqlock to look at the id.
-		 *
-		 * The ID can't change until it gets reclaimed, and it won't
-		 * be reclaimed as long as we have a ref from inode and we
-		 * hold the ilock.
-		 */
-		if (type == XFS_DQ_GROUP)
-			dqp = udqhint->q_gdquot;
-		else
-			dqp = udqhint->q_pdquot;
-		if (dqp && be32_to_cpu(dqp->q_core.d_id) == id) {
-			ASSERT(*IO_idqpp == NULL);
-
-			*IO_idqpp = xfs_qm_dqhold(dqp);
-			xfs_dqunlock(udqhint);
-			return 0;
-		}
-
-		/*
-		 * We can't hold a dquot lock when we call the dqget code.
-		 * We'll deadlock in no time, because of (not conforming to)
-		 * lock ordering - the inodelock comes before any dquot lock,
-		 * and we may drop and reacquire the ilock in xfs_qm_dqget().
-		 */
-		xfs_dqunlock(udqhint);
-	}
-
-	/*
-	 * Find the dquot from somewhere. This bumps the
-	 * reference count of dquot and returns it locked.
-	 * This can return ENOENT if dquot didn't exist on
-	 * disk and we didn't ask it to allocate;
-	 * ESRCH if quotas got turned off suddenly.
+	 * Find the dquot from somewhere. This bumps the reference count of
+	 * dquot and returns it locked.  This can return ENOENT if dquot didn't
+	 * exist on disk and we didn't ask it to allocate; ESRCH if quotas got
+	 * turned off suddenly.
 	 */
 	error = xfs_qm_dqget(ip->i_mount, ip, id, type,
 			     doalloc | XFS_QMOPT_DOWARN, &dqp);
@@ -461,48 +305,6 @@ xfs_qm_dqattach_one(
 	return 0;
 }
 
-
-/*
- * Given a udquot and group/project type, attach the group/project
- * dquot pointer to the udquot as a hint for future lookups.
- */
-STATIC void
-xfs_qm_dqattach_hint(
-	struct xfs_inode	*ip,
-	int			type)
-{
-	struct xfs_dquot **dqhintp;
-	struct xfs_dquot *dqp;
-	struct xfs_dquot *udq = ip->i_udquot;
-
-	ASSERT(type == XFS_DQ_GROUP || type == XFS_DQ_PROJ);
-
-	xfs_dqlock(udq);
-
-	if (type == XFS_DQ_GROUP) {
-		dqp = ip->i_gdquot;
-		dqhintp = &udq->q_gdquot;
-	} else {
-		dqp = ip->i_pdquot;
-		dqhintp = &udq->q_pdquot;
-	}
-
-	if (*dqhintp) {
-		struct xfs_dquot *tmp;
-
-		if (*dqhintp == dqp)
-			goto done;
-
-		tmp = *dqhintp;
-		*dqhintp = NULL;
-		xfs_qm_dqrele(tmp);
-	}
-
-	*dqhintp = xfs_qm_dqhold(dqp);
-done:
-	xfs_dqunlock(udq);
-}
-
 static bool
 xfs_qm_need_dqattach(
 	struct xfs_inode	*ip)
@@ -533,7 +335,6 @@ xfs_qm_dqattach_locked(
 	uint		flags)
 {
 	xfs_mount_t	*mp = ip->i_mount;
-	uint		nquotas = 0;
 	int		error = 0;
 
 	if (!xfs_qm_need_dqattach(ip))
@@ -541,77 +342,39 @@ xfs_qm_dqattach_locked(
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
-	if (XFS_IS_UQUOTA_ON(mp)) {
+	if (XFS_IS_UQUOTA_ON(mp) && !ip->i_udquot) {
 		error = xfs_qm_dqattach_one(ip, ip->i_d.di_uid, XFS_DQ_USER,
 						flags & XFS_QMOPT_DQALLOC,
-						NULL, &ip->i_udquot);
+						&ip->i_udquot);
 		if (error)
 			goto done;
-		nquotas++;
+		ASSERT(ip->i_udquot);
 	}
 
-	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-	if (XFS_IS_GQUOTA_ON(mp)) {
+	if (XFS_IS_GQUOTA_ON(mp) && !ip->i_gdquot) {
 		error = xfs_qm_dqattach_one(ip, ip->i_d.di_gid, XFS_DQ_GROUP,
 						flags & XFS_QMOPT_DQALLOC,
-						ip->i_udquot, &ip->i_gdquot);
-		/*
-		 * Don't worry about the udquot that we may have
-		 * attached above. It'll get detached, if not already.
-		 */
+						&ip->i_gdquot);
 		if (error)
 			goto done;
-		nquotas++;
+		ASSERT(ip->i_gdquot);
 	}
 
-	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-	if (XFS_IS_PQUOTA_ON(mp)) {
+	if (XFS_IS_PQUOTA_ON(mp) && !ip->i_pdquot) {
 		error = xfs_qm_dqattach_one(ip, xfs_get_projid(ip), XFS_DQ_PROJ,
 						flags & XFS_QMOPT_DQALLOC,
-						ip->i_udquot, &ip->i_pdquot);
-		/*
-		 * Don't worry about the udquot that we may have
-		 * attached above. It'll get detached, if not already.
-		 */
+						&ip->i_pdquot);
 		if (error)
 			goto done;
-		nquotas++;
+		ASSERT(ip->i_pdquot);
 	}
 
+done:
 	/*
-	 * Attach this group/project quota to the user quota as a hint.
-	 * This WON'T, in general, result in a thrash.
+	 * Don't worry about the dquots that we may have attached before any
+	 * error - they'll get detached later if it has not already been done.
 	 */
-	if (nquotas > 1 && ip->i_udquot) {
-		ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-		ASSERT(ip->i_gdquot || !XFS_IS_GQUOTA_ON(mp));
-		ASSERT(ip->i_pdquot || !XFS_IS_PQUOTA_ON(mp));
-
-		/*
-		 * We do not have i_udquot locked at this point, but this check
-		 * is OK since we don't depend on the i_gdquot to be accurate
-		 * 100% all the time. It is just a hint, and this will
-		 * succeed in general.
-		 */
-		if (ip->i_udquot->q_gdquot != ip->i_gdquot)
-			xfs_qm_dqattach_hint(ip, XFS_DQ_GROUP);
-
-		if (ip->i_udquot->q_pdquot != ip->i_pdquot)
-			xfs_qm_dqattach_hint(ip, XFS_DQ_PROJ);
-	}
-
- done:
-#ifdef DEBUG
-	if (!error) {
-		if (XFS_IS_UQUOTA_ON(mp))
-			ASSERT(ip->i_udquot);
-		if (XFS_IS_GQUOTA_ON(mp))
-			ASSERT(ip->i_gdquot);
-		if (XFS_IS_PQUOTA_ON(mp))
-			ASSERT(ip->i_pdquot);
-	}
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-#endif
 	return error;
 }
 
@@ -814,22 +577,17 @@ xfs_qm_init_quotainfo(
 
 	qinf = mp->m_quotainfo = kmem_zalloc(sizeof(xfs_quotainfo_t), KM_SLEEP);
 
-	if ((error = list_lru_init(&qinf->qi_lru))) {
-		kmem_free(qinf);
-		mp->m_quotainfo = NULL;
-		return error;
-	}
+	error = list_lru_init(&qinf->qi_lru);
+	if (error)
+		goto out_free_qinf;
 
 	/*
 	 * See if quotainodes are setup, and if not, allocate them,
 	 * and change the superblock accordingly.
 	 */
-	if ((error = xfs_qm_init_quotainos(mp))) {
-		list_lru_destroy(&qinf->qi_lru);
-		kmem_free(qinf);
-		mp->m_quotainfo = NULL;
-		return error;
-	}
+	error = xfs_qm_init_quotainos(mp);
+	if (error)
+		goto out_free_lru;
 
 	INIT_RADIX_TREE(&qinf->qi_uquota_tree, GFP_NOFS);
 	INIT_RADIX_TREE(&qinf->qi_gquota_tree, GFP_NOFS);
@@ -841,8 +599,7 @@ xfs_qm_init_quotainfo(
 
 	/* Precalc some constants */
 	qinf->qi_dqchunklen = XFS_FSB_TO_BB(mp, XFS_DQUOT_CLUSTER_SIZE_FSB);
-	qinf->qi_dqperchunk = xfs_calc_dquots_per_chunk(mp,
-							qinf->qi_dqchunklen);
+	qinf->qi_dqperchunk = xfs_calc_dquots_per_chunk(qinf->qi_dqchunklen);
 
 	mp->m_qflags |= (mp->m_sb.sb_qflags & XFS_ALL_QUOTA_CHKD);
 
@@ -889,7 +646,7 @@ xfs_qm_init_quotainfo(
 		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;
@@ -906,6 +663,13 @@ xfs_qm_init_quotainfo(
 	qinf->qi_shrinker.flags = SHRINKER_NUMA_AWARE;
 	register_shrinker(&qinf->qi_shrinker);
 	return 0;
+
+out_free_lru:
+	list_lru_destroy(&qinf->qi_lru);
+out_free_qinf:
+	kmem_free(qinf);
+	mp->m_quotainfo = NULL;
+	return error;
 }
 
 
@@ -1137,7 +901,7 @@ xfs_qm_dqiter_bufs(
 		 * will leave a trace in the log indicating corruption has
 		 * been detected.
 		 */
-		if (error == EFSCORRUPTED) {
+		if (error == -EFSCORRUPTED) {
 			error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp,
 				      XFS_FSB_TO_DADDR(mp, bno),
 				      mp->m_quotainfo->qi_dqchunklen, 0, &bp,
@@ -1147,6 +911,12 @@ xfs_qm_dqiter_bufs(
 		if (error)
 			break;
 
+		/*
+		 * A corrupt buffer might not have a verifier attached, so
+		 * make sure we have the correct one attached before writeback
+		 * occurs.
+		 */
+		bp->b_ops = &xfs_dquot_buf_ops;
 		xfs_qm_reset_dqcounts(mp, bp, firstid, type);
 		xfs_buf_delwri_queue(bp, buffer_list);
 		xfs_buf_relse(bp);
@@ -1193,16 +963,18 @@ xfs_qm_dqiterate(
 	lblkno = 0;
 	maxlblkcnt = XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes);
 	do {
+		uint		lock_mode;
+
 		nmaps = XFS_DQITER_MAP_SIZE;
 		/*
 		 * We aren't changing the inode itself. Just changing
 		 * some of its data. No new blocks are added here, and
 		 * the inode is never added to the transaction.
 		 */
-		xfs_ilock(qip, XFS_ILOCK_SHARED);
+		lock_mode = xfs_ilock_data_map_shared(qip);
 		error = xfs_bmapi_read(qip, lblkno, maxlblkcnt - lblkno,
 				       map, &nmaps, 0);
-		xfs_iunlock(qip, XFS_ILOCK_SHARED);
+		xfs_iunlock(qip, lock_mode);
 		if (error)
 			break;
 
@@ -1230,7 +1002,7 @@ xfs_qm_dqiterate(
 					xfs_buf_readahead(mp->m_ddev_targp,
 					       XFS_FSB_TO_DADDR(mp, rablkno),
 					       mp->m_quotainfo->qi_dqchunklen,
-					       NULL);
+					       &xfs_dquot_buf_ops);
 					rablkno++;
 				}
 			}
@@ -1278,8 +1050,8 @@ xfs_qm_quotacheck_dqadjust(
 		/*
 		 * Shouldn't be able to turn off quotas here.
 		 */
-		ASSERT(error != ESRCH);
-		ASSERT(error != ENOENT);
+		ASSERT(error != -ESRCH);
+		ASSERT(error != -ENOENT);
 		return error;
 	}
 
@@ -1366,7 +1138,7 @@ xfs_qm_dqusage_adjust(
 	 */
 	if (xfs_is_quota_inode(&mp->m_sb, ino)) {
 		*res = BULKSTAT_RV_NOTHING;
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	/*
@@ -1470,7 +1242,7 @@ out_unlock:
  * Walk thru all the filesystem inodes and construct a consistent view
  * of the disk quota world. If the quotacheck fails, disable quotas.
  */
-int
+STATIC int
 xfs_qm_quotacheck(
 	xfs_mount_t	*mp)
 {
@@ -1603,7 +1375,100 @@ xfs_qm_quotacheck(
 		}
 	} else
 		xfs_notice(mp, "Quotacheck: Done.");
-	return (error);
+	return error;
+}
+
+/*
+ * This is called from xfs_mountfs to start quotas and initialize all
+ * necessary data structures like quotainfo.  This is also responsible for
+ * running a quotacheck as necessary.  We are guaranteed that the superblock
+ * is consistently read in at this point.
+ *
+ * If we fail here, the mount will continue with quota turned off. We don't
+ * need to inidicate success or failure at all.
+ */
+void
+xfs_qm_mount_quotas(
+	struct xfs_mount	*mp)
+{
+	int			error = 0;
+	uint			sbf;
+
+	/*
+	 * If quotas on realtime volumes is not supported, we disable
+	 * quotas immediately.
+	 */
+	if (mp->m_sb.sb_rextents) {
+		xfs_notice(mp, "Cannot turn on quotas for realtime filesystem");
+		mp->m_qflags = 0;
+		goto write_changes;
+	}
+
+	ASSERT(XFS_IS_QUOTA_RUNNING(mp));
+
+	/*
+	 * Allocate the quotainfo structure inside the mount struct, and
+	 * create quotainode(s), and change/rev superblock if necessary.
+	 */
+	error = xfs_qm_init_quotainfo(mp);
+	if (error) {
+		/*
+		 * We must turn off quotas.
+		 */
+		ASSERT(mp->m_quotainfo == NULL);
+		mp->m_qflags = 0;
+		goto write_changes;
+	}
+	/*
+	 * If any of the quotas are not consistent, do a quotacheck.
+	 */
+	if (XFS_QM_NEED_QUOTACHECK(mp)) {
+		error = xfs_qm_quotacheck(mp);
+		if (error) {
+			/* Quotacheck failed and disabled quotas. */
+			return;
+		}
+	}
+	/*
+	 * If one type of quotas is off, then it will lose its
+	 * quotachecked status, since we won't be doing accounting for
+	 * that type anymore.
+	 */
+	if (!XFS_IS_UQUOTA_ON(mp))
+		mp->m_qflags &= ~XFS_UQUOTA_CHKD;
+	if (!XFS_IS_GQUOTA_ON(mp))
+		mp->m_qflags &= ~XFS_GQUOTA_CHKD;
+	if (!XFS_IS_PQUOTA_ON(mp))
+		mp->m_qflags &= ~XFS_PQUOTA_CHKD;
+
+ write_changes:
+	/*
+	 * We actually don't have to acquire the m_sb_lock at all.
+	 * This can only be called from mount, and that's single threaded. XXX
+	 */
+	spin_lock(&mp->m_sb_lock);
+	sbf = mp->m_sb.sb_qflags;
+	mp->m_sb.sb_qflags = mp->m_qflags & XFS_MOUNT_QUOTA_ALL;
+	spin_unlock(&mp->m_sb_lock);
+
+	if (sbf != (mp->m_qflags & XFS_MOUNT_QUOTA_ALL)) {
+		if (xfs_qm_write_sb_changes(mp, XFS_SB_QFLAGS)) {
+			/*
+			 * We could only have been turning quotas off.
+			 * We aren't in very good shape actually because
+			 * the incore structures are convinced that quotas are
+			 * off, but the on disk superblock doesn't know that !
+			 */
+			ASSERT(!(XFS_IS_QUOTA_RUNNING(mp)));
+			xfs_alert(mp, "%s: Superblock update failed!",
+				__func__);
+		}
+	}
+
+	if (error) {
+		xfs_warn(mp, "Failed to initialize disk quotas.");
+		return;
+	}
 }
 
 /*
@@ -1633,7 +1498,7 @@ xfs_qm_init_quotainos(
 			error = xfs_iget(mp, NULL, mp->m_sb.sb_uquotino,
 					     0, 0, &uip);
 			if (error)
-				return XFS_ERROR(error);
+				return error;
 		}
 		if (XFS_IS_GQUOTA_ON(mp) &&
 		    mp->m_sb.sb_gquotino != NULLFSINO) {
@@ -1703,7 +1568,7 @@ error_rele:
 		IRELE(gip);
 	if (pip)
 		IRELE(pip);
-	return XFS_ERROR(error);
+	return error;
 }
 
 STATIC void
@@ -1819,7 +1684,7 @@ xfs_qm_vop_dqalloc(
 						 XFS_QMOPT_DOWARN,
 						 &uq);
 			if (error) {
-				ASSERT(error != ENOENT);
+				ASSERT(error != -ENOENT);
 				return error;
 			}
 			/*
@@ -1846,7 +1711,7 @@ xfs_qm_vop_dqalloc(
 						 XFS_QMOPT_DOWARN,
 						 &gq);
 			if (error) {
-				ASSERT(error != ENOENT);
+				ASSERT(error != -ENOENT);
 				goto error_rele;
 			}
 			xfs_dqunlock(gq);
@@ -1866,7 +1731,7 @@ xfs_qm_vop_dqalloc(
 						 XFS_QMOPT_DOWARN,
 						 &pq);
 			if (error) {
-				ASSERT(error != ENOENT);
+				ASSERT(error != -ENOENT);
 				goto error_rele;
 			}
 			xfs_dqunlock(pq);
@@ -2035,7 +1900,7 @@ xfs_qm_vop_chown_reserve(
 				-((xfs_qcnt_t)delblks), 0, blkflags);
 	}
 
-	return (0);
+	return 0;
 }
 
 int
@@ -2082,24 +1947,21 @@ xfs_qm_vop_create_dqattach(
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	ASSERT(XFS_IS_QUOTA_RUNNING(mp));
 
-	if (udqp) {
+	if (udqp && XFS_IS_UQUOTA_ON(mp)) {
 		ASSERT(ip->i_udquot == NULL);
-		ASSERT(XFS_IS_UQUOTA_ON(mp));
 		ASSERT(ip->i_d.di_uid == be32_to_cpu(udqp->q_core.d_id));
 
 		ip->i_udquot = xfs_qm_dqhold(udqp);
 		xfs_trans_mod_dquot(tp, udqp, XFS_TRANS_DQ_ICOUNT, 1);
 	}
-	if (gdqp) {
+	if (gdqp && XFS_IS_GQUOTA_ON(mp)) {
 		ASSERT(ip->i_gdquot == NULL);
-		ASSERT(XFS_IS_GQUOTA_ON(mp));
 		ASSERT(ip->i_d.di_gid == be32_to_cpu(gdqp->q_core.d_id));
 		ip->i_gdquot = xfs_qm_dqhold(gdqp);
 		xfs_trans_mod_dquot(tp, gdqp, XFS_TRANS_DQ_ICOUNT, 1);
 	}
-	if (pdqp) {
+	if (pdqp && XFS_IS_PQUOTA_ON(mp)) {
 		ASSERT(ip->i_pdquot == NULL);
-		ASSERT(XFS_IS_PQUOTA_ON(mp));
 		ASSERT(xfs_get_projid(ip) == be32_to_cpu(pdqp->q_core.d_id));
 
 		ip->i_pdquot = xfs_qm_dqhold(pdqp);
diff --git a/fs/xfs/xfs_qm.h b/fs/xfs/xfs_qm.h
index a788b66a5cb1..3a07a937e232 100644
--- a/fs/xfs/xfs_qm.h
+++ b/fs/xfs/xfs_qm.h
@@ -20,13 +20,29 @@
 
 #include "xfs_dquot_item.h"
 #include "xfs_dquot.h"
-#include "xfs_quota_priv.h"
 
 struct xfs_inode;
 
 extern struct kmem_zone	*xfs_qm_dqtrxzone;
 
 /*
+ * Number of bmaps that we ask from bmapi when doing a quotacheck.
+ * We make this restriction to keep the memory usage to a minimum.
+ */
+#define XFS_DQITER_MAP_SIZE	10
+
+#define XFS_IS_DQUOT_UNINITIALIZED(dqp) ( \
+	!dqp->q_core.d_blk_hardlimit && \
+	!dqp->q_core.d_blk_softlimit && \
+	!dqp->q_core.d_rtb_hardlimit && \
+	!dqp->q_core.d_rtb_softlimit && \
+	!dqp->q_core.d_ino_hardlimit && \
+	!dqp->q_core.d_ino_softlimit && \
+	!dqp->q_core.d_bcount && \
+	!dqp->q_core.d_rtbcount && \
+	!dqp->q_core.d_icount)
+
+/*
  * This defines the unit of allocation of dquots.
  * Currently, it is just one file system block, and a 4K blk contains 30
  * (136 * 30 = 4080) dquots. It's probably not worth trying to make
@@ -141,7 +157,6 @@ struct xfs_dquot_acct {
 #define XFS_QM_RTBWARNLIMIT	5
 
 extern void		xfs_qm_destroy_quotainfo(struct xfs_mount *);
-extern int		xfs_qm_quotacheck(struct xfs_mount *);
 extern int		xfs_qm_write_sb_changes(struct xfs_mount *, __int64_t);
 
 /* dquot stuff */
diff --git a/fs/xfs/xfs_qm_bhv.c b/fs/xfs/xfs_qm_bhv.c
index e9be63abd8d2..2c61e61b0205 100644
--- a/fs/xfs/xfs_qm_bhv.c
+++ b/fs/xfs/xfs_qm_bhv.c
@@ -117,7 +117,7 @@ xfs_qm_newmount(
 			(uquotaondisk ? " usrquota" : ""),
 			(gquotaondisk ? " grpquota" : ""),
 			(pquotaondisk ? " prjquota" : ""));
-		return XFS_ERROR(EPERM);
+		return -EPERM;
 	}
 
 	if (XFS_IS_QUOTA_ON(mp) || quotaondisk) {
diff --git a/fs/xfs/xfs_qm_syscalls.c b/fs/xfs/xfs_qm_syscalls.c
index 437c9198031a..80f2d77d929a 100644
--- a/fs/xfs/xfs_qm_syscalls.c
+++ b/fs/xfs/xfs_qm_syscalls.c
@@ -64,10 +64,10 @@ xfs_qm_scall_quotaoff(
 	/*
 	 * No file system can have quotas enabled on disk but not in core.
 	 * Note that quota utilities (like quotaoff) _expect_
-	 * errno == EEXIST here.
+	 * errno == -EEXIST here.
 	 */
 	if ((mp->m_qflags & flags) == 0)
-		return XFS_ERROR(EEXIST);
+		return -EEXIST;
 	error = 0;
 
 	flags &= (XFS_ALL_QUOTA_ACCT | XFS_ALL_QUOTA_ENFD);
@@ -94,7 +94,7 @@ xfs_qm_scall_quotaoff(
 
 		/* XXX what to do if error ? Revert back to old vals incore ? */
 		error = xfs_qm_write_sb_changes(mp, XFS_SB_QFLAGS);
-		return (error);
+		return error;
 	}
 
 	dqtype = 0;
@@ -198,7 +198,7 @@ xfs_qm_scall_quotaoff(
 	if (mp->m_qflags == 0) {
 		mutex_unlock(&q->qi_quotaofflock);
 		xfs_qm_destroy_quotainfo(mp);
-		return (0);
+		return 0;
 	}
 
 	/*
@@ -278,22 +278,29 @@ xfs_qm_scall_trunc_qfiles(
 	xfs_mount_t	*mp,
 	uint		flags)
 {
-	int		error = 0, error2 = 0;
+	int		error = -EINVAL;
 
-	if (!xfs_sb_version_hasquota(&mp->m_sb) || flags == 0) {
+	if (!xfs_sb_version_hasquota(&mp->m_sb) || flags == 0 ||
+	    (flags & ~XFS_DQ_ALLTYPES)) {
 		xfs_debug(mp, "%s: flags=%x m_qflags=%x",
 			__func__, flags, mp->m_qflags);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
-	if (flags & XFS_DQ_USER)
+	if (flags & XFS_DQ_USER) {
 		error = xfs_qm_scall_trunc_qfile(mp, mp->m_sb.sb_uquotino);
-	if (flags & XFS_DQ_GROUP)
-		error2 = xfs_qm_scall_trunc_qfile(mp, mp->m_sb.sb_gquotino);
+		if (error)
+			return error;
+	}
+	if (flags & XFS_DQ_GROUP) {
+		error = xfs_qm_scall_trunc_qfile(mp, mp->m_sb.sb_gquotino);
+		if (error)
+			return error;
+	}
 	if (flags & XFS_DQ_PROJ)
-		error2 = xfs_qm_scall_trunc_qfile(mp, mp->m_sb.sb_pquotino);
+		error = xfs_qm_scall_trunc_qfile(mp, mp->m_sb.sb_pquotino);
 
-	return error ? error : error2;
+	return error;
 }
 
 /*
@@ -321,7 +328,7 @@ xfs_qm_scall_quotaon(
 	if (flags == 0) {
 		xfs_debug(mp, "%s: zero flags, m_qflags=%x",
 			__func__, mp->m_qflags);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	/* No fs can turn on quotas with a delayed effect */
@@ -344,13 +351,13 @@ xfs_qm_scall_quotaon(
 		xfs_debug(mp,
 			"%s: Can't enforce without acct, flags=%x sbflags=%x",
 			__func__, flags, mp->m_sb.sb_qflags);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 	/*
 	 * If everything's up to-date incore, then don't waste time.
 	 */
 	if ((mp->m_qflags & flags) == flags)
-		return XFS_ERROR(EEXIST);
+		return -EEXIST;
 
 	/*
 	 * Change sb_qflags on disk but not incore mp->qflags
@@ -365,11 +372,11 @@ xfs_qm_scall_quotaon(
 	 * There's nothing to change if it's the same.
 	 */
 	if ((qf & flags) == flags && sbflags == 0)
-		return XFS_ERROR(EEXIST);
+		return -EEXIST;
 	sbflags |= XFS_SB_QFLAGS;
 
 	if ((error = xfs_qm_write_sb_changes(mp, sbflags)))
-		return (error);
+		return error;
 	/*
 	 * If we aren't trying to switch on quota enforcement, we are done.
 	 */
@@ -380,10 +387,10 @@ xfs_qm_scall_quotaon(
 	     ((mp->m_sb.sb_qflags & XFS_GQUOTA_ACCT) !=
 	     (mp->m_qflags & XFS_GQUOTA_ACCT)) ||
 	    (flags & XFS_ALL_QUOTA_ENFD) == 0)
-		return (0);
+		return 0;
 
 	if (! XFS_IS_QUOTA_RUNNING(mp))
-		return XFS_ERROR(ESRCH);
+		return -ESRCH;
 
 	/*
 	 * Switch on quota enforcement in core.
@@ -392,7 +399,7 @@ xfs_qm_scall_quotaon(
 	mp->m_qflags |= (flags & XFS_ALL_QUOTA_ENFD);
 	mutex_unlock(&mp->m_quotainfo->qi_quotaofflock);
 
-	return (0);
+	return 0;
 }
 
 
@@ -419,7 +426,7 @@ xfs_qm_scall_getqstat(
 	if (!xfs_sb_version_hasquota(&mp->m_sb)) {
 		out->qs_uquota.qfs_ino = NULLFSINO;
 		out->qs_gquota.qfs_ino = NULLFSINO;
-		return (0);
+		return 0;
 	}
 
 	out->qs_flags = (__uint16_t) xfs_qm_export_flags(mp->m_qflags &
@@ -507,7 +514,7 @@ xfs_qm_scall_getqstatv(
 		out->qs_uquota.qfs_ino = NULLFSINO;
 		out->qs_gquota.qfs_ino = NULLFSINO;
 		out->qs_pquota.qfs_ino = NULLFSINO;
-		return (0);
+		return 0;
 	}
 
 	out->qs_flags = (__uint16_t) xfs_qm_export_flags(mp->m_qflags &
@@ -588,7 +595,7 @@ xfs_qm_scall_setqlim(
 	xfs_qcnt_t		hard, soft;
 
 	if (newlim->d_fieldmask & ~XFS_DQ_MASK)
-		return EINVAL;
+		return -EINVAL;
 	if ((newlim->d_fieldmask & XFS_DQ_MASK) == 0)
 		return 0;
 
@@ -608,7 +615,7 @@ xfs_qm_scall_setqlim(
 	 */
 	error = xfs_qm_dqget(mp, NULL, id, type, XFS_QMOPT_DQALLOC, &dqp);
 	if (error) {
-		ASSERT(error != ENOENT);
+		ASSERT(error != -ENOENT);
 		goto out_unlock;
 	}
 	xfs_dqunlock(dqp);
@@ -751,7 +758,7 @@ xfs_qm_log_quotaoff_end(
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_qm_equotaoff, 0, 0);
 	if (error) {
 		xfs_trans_cancel(tp, 0);
-		return (error);
+		return error;
 	}
 
 	qoffi = xfs_trans_get_qoff_item(tp, startqoff,
@@ -765,7 +772,7 @@ xfs_qm_log_quotaoff_end(
 	 */
 	xfs_trans_set_sync(tp);
 	error = xfs_trans_commit(tp, 0);
-	return (error);
+	return error;
 }
 
 
@@ -815,7 +822,7 @@ error0:
 		spin_unlock(&mp->m_sb_lock);
 	}
 	*qoffstartp = qoffi;
-	return (error);
+	return error;
 }
 
 
@@ -843,7 +850,7 @@ xfs_qm_scall_getquota(
 	 * our utility programs are concerned.
 	 */
 	if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) {
-		error = XFS_ERROR(ENOENT);
+		error = -ENOENT;
 		goto out_put;
 	}
 
@@ -946,14 +953,13 @@ xfs_qm_export_flags(
 		uflags |= FS_QUOTA_GDQ_ENFD;
 	if (flags & XFS_PQUOTA_ENFD)
 		uflags |= FS_QUOTA_PDQ_ENFD;
-	return (uflags);
+	return uflags;
 }
 
 
 STATIC int
 xfs_dqrele_inode(
 	struct xfs_inode	*ip,
-	struct xfs_perag	*pag,
 	int			flags,
 	void			*args)
 {
diff --git a/fs/xfs/xfs_quota_priv.h b/fs/xfs/xfs_quota_priv.h
deleted file mode 100644
index 6d86219d93da..000000000000
--- a/fs/xfs/xfs_quota_priv.h
+++ /dev/null
@@ -1,42 +0,0 @@
-/*
- * Copyright (c) 2000-2003 Silicon Graphics, Inc.
- * 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_QUOTA_PRIV_H__
-#define __XFS_QUOTA_PRIV_H__
-
-/*
- * Number of bmaps that we ask from bmapi when doing a quotacheck.
- * We make this restriction to keep the memory usage to a minimum.
- */
-#define XFS_DQITER_MAP_SIZE	10
-
-#define XFS_IS_DQUOT_UNINITIALIZED(dqp) ( \
-	!dqp->q_core.d_blk_hardlimit && \
-	!dqp->q_core.d_blk_softlimit && \
-	!dqp->q_core.d_rtb_hardlimit && \
-	!dqp->q_core.d_rtb_softlimit && \
-	!dqp->q_core.d_ino_hardlimit && \
-	!dqp->q_core.d_ino_softlimit && \
-	!dqp->q_core.d_bcount && \
-	!dqp->q_core.d_rtbcount && \
-	!dqp->q_core.d_icount)
-
-#define DQFLAGTO_TYPESTR(d)	(((d)->dq_flags & XFS_DQ_USER) ? "USR" : \
-				 (((d)->dq_flags & XFS_DQ_GROUP) ? "GRP" : \
-				 (((d)->dq_flags & XFS_DQ_PROJ) ? "PRJ":"???")))
-
-#endif	/* __XFS_QUOTA_PRIV_H__ */
diff --git a/fs/xfs/xfs_quotaops.c b/fs/xfs/xfs_quotaops.c
index af33cafe69b6..b238027df987 100644
--- a/fs/xfs/xfs_quotaops.c
+++ b/fs/xfs/xfs_quotaops.c
@@ -51,7 +51,7 @@ xfs_fs_get_xstate(
 
 	if (!XFS_IS_QUOTA_RUNNING(mp))
 		return -ENOSYS;
-	return -xfs_qm_scall_getqstat(mp, fqs);
+	return xfs_qm_scall_getqstat(mp, fqs);
 }
 
 STATIC int
@@ -63,7 +63,7 @@ xfs_fs_get_xstatev(
 
 	if (!XFS_IS_QUOTA_RUNNING(mp))
 		return -ENOSYS;
-	return -xfs_qm_scall_getqstatv(mp, fqs);
+	return xfs_qm_scall_getqstatv(mp, fqs);
 }
 
 STATIC int
@@ -95,21 +95,41 @@ xfs_fs_set_xstate(
 
 	switch (op) {
 	case Q_XQUOTAON:
-		return -xfs_qm_scall_quotaon(mp, flags);
+		return xfs_qm_scall_quotaon(mp, flags);
 	case Q_XQUOTAOFF:
 		if (!XFS_IS_QUOTA_ON(mp))
 			return -EINVAL;
-		return -xfs_qm_scall_quotaoff(mp, flags);
-	case Q_XQUOTARM:
-		if (XFS_IS_QUOTA_ON(mp))
-			return -EINVAL;
-		return -xfs_qm_scall_trunc_qfiles(mp, flags);
+		return xfs_qm_scall_quotaoff(mp, flags);
 	}
 
 	return -EINVAL;
 }
 
 STATIC int
+xfs_fs_rm_xquota(
+	struct super_block	*sb,
+	unsigned int		uflags)
+{
+	struct xfs_mount	*mp = XFS_M(sb);
+	unsigned int		flags = 0;
+
+	if (sb->s_flags & MS_RDONLY)
+		return -EROFS;
+
+	if (XFS_IS_QUOTA_ON(mp))
+		return -EINVAL;
+
+	if (uflags & FS_USER_QUOTA)
+		flags |= XFS_DQ_USER;
+	if (uflags & FS_GROUP_QUOTA)
+		flags |= XFS_DQ_GROUP;
+	if (uflags & FS_PROJ_QUOTA)
+		flags |= XFS_DQ_PROJ;
+
+	return xfs_qm_scall_trunc_qfiles(mp, flags);
+}
+
+STATIC int
 xfs_fs_get_dqblk(
 	struct super_block	*sb,
 	struct kqid		qid,
@@ -122,7 +142,7 @@ xfs_fs_get_dqblk(
 	if (!XFS_IS_QUOTA_ON(mp))
 		return -ESRCH;
 
-	return -xfs_qm_scall_getquota(mp, from_kqid(&init_user_ns, qid),
+	return xfs_qm_scall_getquota(mp, from_kqid(&init_user_ns, qid),
 				      xfs_quota_type(qid.type), fdq);
 }
 
@@ -141,7 +161,7 @@ xfs_fs_set_dqblk(
 	if (!XFS_IS_QUOTA_ON(mp))
 		return -ESRCH;
 
-	return -xfs_qm_scall_setqlim(mp, from_kqid(&init_user_ns, qid),
+	return xfs_qm_scall_setqlim(mp, from_kqid(&init_user_ns, qid),
 				     xfs_quota_type(qid.type), fdq);
 }
 
@@ -149,6 +169,7 @@ const struct quotactl_ops xfs_quotactl_operations = {
 	.get_xstatev		= xfs_fs_get_xstatev,
 	.get_xstate		= xfs_fs_get_xstate,
 	.set_xstate		= xfs_fs_set_xstate,
+	.rm_xquota		= xfs_fs_rm_xquota,
 	.get_dqblk		= xfs_fs_get_dqblk,
 	.set_dqblk		= xfs_fs_set_dqblk,
 };
diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c
index a6a76b2b6a85..909e143b87ae 100644
--- a/fs/xfs/xfs_rtalloc.c
+++ b/fs/xfs/xfs_rtalloc.c
@@ -842,7 +842,7 @@ xfs_growfs_rt_alloc(
 		/*
 		 * Reserve space & log for one extent added to the file.
 		 */
-		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_growdata,
+		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_growrtalloc,
 					  resblks, 0);
 		if (error)
 			goto error_cancel;
@@ -863,7 +863,7 @@ xfs_growfs_rt_alloc(
 					XFS_BMAPI_METADATA, &firstblock,
 					resblks, &map, &nmap, &flist);
 		if (!error && nmap < 1)
-			error = XFS_ERROR(ENOSPC);
+			error = -ENOSPC;
 		if (error)
 			goto error_cancel;
 		/*
@@ -903,7 +903,7 @@ xfs_growfs_rt_alloc(
 			bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
 				mp->m_bsize, 0);
 			if (bp == NULL) {
-				error = XFS_ERROR(EIO);
+				error = -EIO;
 error_cancel:
 				xfs_trans_cancel(tp, cancelflags);
 				goto error;
@@ -944,9 +944,9 @@ xfs_growfs_rt(
 	xfs_buf_t	*bp;		/* temporary buffer */
 	int		error;		/* error return value */
 	xfs_mount_t	*nmp;		/* new (fake) mount structure */
-	xfs_drfsbno_t	nrblocks;	/* new number of realtime blocks */
+	xfs_rfsblock_t	nrblocks;	/* new number of realtime blocks */
 	xfs_extlen_t	nrbmblocks;	/* new number of rt bitmap blocks */
-	xfs_drtbno_t	nrextents;	/* new number of realtime extents */
+	xfs_rtblock_t	nrextents;	/* new number of realtime extents */
 	uint8_t		nrextslog;	/* new log2 of sb_rextents */
 	xfs_extlen_t	nrsumblocks;	/* new number of summary blocks */
 	uint		nrsumlevels;	/* new rt summary levels */
@@ -962,11 +962,11 @@ xfs_growfs_rt(
 	 * Initial error checking.
 	 */
 	if (!capable(CAP_SYS_ADMIN))
-		return XFS_ERROR(EPERM);
+		return -EPERM;
 	if (mp->m_rtdev_targp == NULL || mp->m_rbmip == NULL ||
 	    (nrblocks = in->newblocks) <= sbp->sb_rblocks ||
 	    (sbp->sb_rblocks && (in->extsize != sbp->sb_rextsize)))
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	if ((error = xfs_sb_validate_fsb_count(sbp, nrblocks)))
 		return error;
 	/*
@@ -976,7 +976,7 @@ xfs_growfs_rt(
 				XFS_FSB_TO_BB(mp, nrblocks - 1),
 				XFS_FSB_TO_BB(mp, 1), 0, NULL);
 	if (!bp)
-		return EIO;
+		return -EIO;
 	if (bp->b_error) {
 		error = bp->b_error;
 		xfs_buf_relse(bp);
@@ -1001,7 +1001,7 @@ xfs_growfs_rt(
 	 * since we'll log basically the whole summary file at once.
 	 */
 	if (nrsumblocks > (mp->m_sb.sb_logblocks >> 1))
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	/*
 	 * Get the old block counts for bitmap and summary inodes.
 	 * These can't change since other growfs callers are locked out.
@@ -1208,7 +1208,7 @@ xfs_rtallocate_extent(
 				len, &sumbp, &sb, prod, &r);
 		break;
 	default:
-		error = EIO;
+		error = -EIO;
 		ASSERT(0);
 	}
 	if (error)
@@ -1247,7 +1247,7 @@ xfs_rtmount_init(
 	if (mp->m_rtdev_targp == NULL) {
 		xfs_warn(mp,
 	"Filesystem has a realtime volume, use rtdev=device option");
-		return XFS_ERROR(ENODEV);
+		return -ENODEV;
 	}
 	mp->m_rsumlevels = sbp->sb_rextslog + 1;
 	mp->m_rsumsize =
@@ -1263,7 +1263,7 @@ xfs_rtmount_init(
 		xfs_warn(mp, "realtime mount -- %llu != %llu",
 			(unsigned long long) XFS_BB_TO_FSB(mp, d),
 			(unsigned long long) mp->m_sb.sb_rblocks);
-		return XFS_ERROR(EFBIG);
+		return -EFBIG;
 	}
 	bp = xfs_buf_read_uncached(mp->m_rtdev_targp,
 					d - XFS_FSB_TO_BB(mp, 1),
@@ -1272,7 +1272,7 @@ xfs_rtmount_init(
 		xfs_warn(mp, "realtime device size check failed");
 		if (bp)
 			xfs_buf_relse(bp);
-		return EIO;
+		return -EIO;
 	}
 	xfs_buf_relse(bp);
 	return 0;
diff --git a/fs/xfs/xfs_rtalloc.h b/fs/xfs/xfs_rtalloc.h
index 752b63d10300..c642795324af 100644
--- a/fs/xfs/xfs_rtalloc.h
+++ b/fs/xfs/xfs_rtalloc.h
@@ -132,7 +132,7 @@ xfs_rtmount_init(
 		return 0;
 
 	xfs_warn(mp, "Not built with CONFIG_XFS_RT");
-	return ENOSYS;
+	return -ENOSYS;
 }
 # define xfs_rtmount_inodes(m)  (((mp)->m_sb.sb_rblocks == 0)? 0 : (ENOSYS))
 # define xfs_rtunmount_inodes(m)
diff --git a/fs/xfs/xfs_stats.c b/fs/xfs/xfs_stats.c
index ce372b7d5644..f2240383d4bb 100644
--- a/fs/xfs/xfs_stats.c
+++ b/fs/xfs/xfs_stats.c
@@ -59,6 +59,7 @@ static int xfs_stat_proc_show(struct seq_file *m, void *v)
 		{ "abtc2",		XFSSTAT_END_ABTC_V2		},
 		{ "bmbt2",		XFSSTAT_END_BMBT_V2		},
 		{ "ibt2",		XFSSTAT_END_IBT_V2		},
+		{ "fibt2",		XFSSTAT_END_FIBT_V2		},
 		/* we print both series of quota information together */
 		{ "qm",			XFSSTAT_END_QM			},
 	};
diff --git a/fs/xfs/xfs_stats.h b/fs/xfs/xfs_stats.h
index c03ad38ceaeb..c8f238b8299a 100644
--- a/fs/xfs/xfs_stats.h
+++ b/fs/xfs/xfs_stats.h
@@ -183,7 +183,23 @@ struct xfsstats {
 	__uint32_t		xs_ibt_2_alloc;
 	__uint32_t		xs_ibt_2_free;
 	__uint32_t		xs_ibt_2_moves;
-#define XFSSTAT_END_XQMSTAT		(XFSSTAT_END_IBT_V2+6)
+#define XFSSTAT_END_FIBT_V2		(XFSSTAT_END_IBT_V2+15)
+	__uint32_t		xs_fibt_2_lookup;
+	__uint32_t		xs_fibt_2_compare;
+	__uint32_t		xs_fibt_2_insrec;
+	__uint32_t		xs_fibt_2_delrec;
+	__uint32_t		xs_fibt_2_newroot;
+	__uint32_t		xs_fibt_2_killroot;
+	__uint32_t		xs_fibt_2_increment;
+	__uint32_t		xs_fibt_2_decrement;
+	__uint32_t		xs_fibt_2_lshift;
+	__uint32_t		xs_fibt_2_rshift;
+	__uint32_t		xs_fibt_2_split;
+	__uint32_t		xs_fibt_2_join;
+	__uint32_t		xs_fibt_2_alloc;
+	__uint32_t		xs_fibt_2_free;
+	__uint32_t		xs_fibt_2_moves;
+#define XFSSTAT_END_XQMSTAT		(XFSSTAT_END_FIBT_V2+6)
 	__uint32_t		xs_qm_dqreclaims;
 	__uint32_t		xs_qm_dqreclaim_misses;
 	__uint32_t		xs_qm_dquot_dups;
diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
index f317488263dd..b194652033cd 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -61,6 +61,7 @@
 static const struct super_operations xfs_super_operations;
 static kmem_zone_t *xfs_ioend_zone;
 mempool_t *xfs_ioend_pool;
+struct kset *xfs_kset;
 
 #define MNTOPT_LOGBUFS	"logbufs"	/* number of XFS log buffers */
 #define MNTOPT_LOGBSIZE	"logbsize"	/* size of XFS log buffers */
@@ -185,7 +186,7 @@ xfs_parseargs(
 	 */
 	mp->m_fsname = kstrndup(sb->s_id, MAXNAMELEN, GFP_KERNEL);
 	if (!mp->m_fsname)
-		return ENOMEM;
+		return -ENOMEM;
 	mp->m_fsname_len = strlen(mp->m_fsname) + 1;
 
 	/*
@@ -204,9 +205,6 @@ xfs_parseargs(
 	 */
 	mp->m_flags |= XFS_MOUNT_BARRIER;
 	mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
-#if !XFS_BIG_INUMS
-	mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
-#endif
 
 	/*
 	 * These can be overridden by the mount option parsing.
@@ -227,57 +225,57 @@ xfs_parseargs(
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			if (kstrtoint(value, 10, &mp->m_logbufs))
-				return EINVAL;
+				return -EINVAL;
 		} else if (!strcmp(this_char, MNTOPT_LOGBSIZE)) {
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			if (suffix_kstrtoint(value, 10, &mp->m_logbsize))
-				return EINVAL;
+				return -EINVAL;
 		} else if (!strcmp(this_char, MNTOPT_LOGDEV)) {
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			mp->m_logname = kstrndup(value, MAXNAMELEN, GFP_KERNEL);
 			if (!mp->m_logname)
-				return ENOMEM;
+				return -ENOMEM;
 		} else if (!strcmp(this_char, MNTOPT_MTPT)) {
 			xfs_warn(mp, "%s option not allowed on this system",
 				this_char);
-			return EINVAL;
+			return -EINVAL;
 		} else if (!strcmp(this_char, MNTOPT_RTDEV)) {
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			mp->m_rtname = kstrndup(value, MAXNAMELEN, GFP_KERNEL);
 			if (!mp->m_rtname)
-				return ENOMEM;
+				return -ENOMEM;
 		} else if (!strcmp(this_char, MNTOPT_BIOSIZE)) {
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			if (kstrtoint(value, 10, &iosize))
-				return EINVAL;
+				return -EINVAL;
 			iosizelog = ffs(iosize) - 1;
 		} else if (!strcmp(this_char, MNTOPT_ALLOCSIZE)) {
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			if (suffix_kstrtoint(value, 10, &iosize))
-				return EINVAL;
+				return -EINVAL;
 			iosizelog = ffs(iosize) - 1;
 		} else if (!strcmp(this_char, MNTOPT_GRPID) ||
 			   !strcmp(this_char, MNTOPT_BSDGROUPS)) {
@@ -297,27 +295,22 @@ xfs_parseargs(
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			if (kstrtoint(value, 10, &dsunit))
-				return EINVAL;
+				return -EINVAL;
 		} else if (!strcmp(this_char, MNTOPT_SWIDTH)) {
 			if (!value || !*value) {
 				xfs_warn(mp, "%s option requires an argument",
 					this_char);
-				return EINVAL;
+				return -EINVAL;
 			}
 			if (kstrtoint(value, 10, &dswidth))
-				return EINVAL;
+				return -EINVAL;
 		} else if (!strcmp(this_char, MNTOPT_32BITINODE)) {
 			mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
 		} else if (!strcmp(this_char, MNTOPT_64BITINODE)) {
 			mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
-#if !XFS_BIG_INUMS
-			xfs_warn(mp, "%s option not allowed on this system",
-				this_char);
-			return EINVAL;
-#endif
 		} else if (!strcmp(this_char, MNTOPT_NOUUID)) {
 			mp->m_flags |= XFS_MOUNT_NOUUID;
 		} else if (!strcmp(this_char, MNTOPT_BARRIER)) {
@@ -390,7 +383,7 @@ xfs_parseargs(
 	"irixsgid is now a sysctl(2) variable, option is deprecated.");
 		} else {
 			xfs_warn(mp, "unknown mount option [%s].", this_char);
-			return EINVAL;
+			return -EINVAL;
 		}
 	}
 
@@ -400,32 +393,32 @@ xfs_parseargs(
 	if ((mp->m_flags & XFS_MOUNT_NORECOVERY) &&
 	    !(mp->m_flags & XFS_MOUNT_RDONLY)) {
 		xfs_warn(mp, "no-recovery mounts must be read-only.");
-		return EINVAL;
+		return -EINVAL;
 	}
 
 	if ((mp->m_flags & XFS_MOUNT_NOALIGN) && (dsunit || dswidth)) {
 		xfs_warn(mp,
 	"sunit and swidth options incompatible with the noalign option");
-		return EINVAL;
+		return -EINVAL;
 	}
 
 #ifndef CONFIG_XFS_QUOTA
 	if (XFS_IS_QUOTA_RUNNING(mp)) {
 		xfs_warn(mp, "quota support not available in this kernel.");
-		return EINVAL;
+		return -EINVAL;
 	}
 #endif
 
 	if ((dsunit && !dswidth) || (!dsunit && dswidth)) {
 		xfs_warn(mp, "sunit and swidth must be specified together");
-		return EINVAL;
+		return -EINVAL;
 	}
 
 	if (dsunit && (dswidth % dsunit != 0)) {
 		xfs_warn(mp,
 	"stripe width (%d) must be a multiple of the stripe unit (%d)",
 			dswidth, dsunit);
-		return EINVAL;
+		return -EINVAL;
 	}
 
 done:
@@ -446,7 +439,7 @@ done:
 	     mp->m_logbufs > XLOG_MAX_ICLOGS)) {
 		xfs_warn(mp, "invalid logbufs value: %d [not %d-%d]",
 			mp->m_logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 	if (mp->m_logbsize != -1 &&
 	    mp->m_logbsize !=  0 &&
@@ -456,7 +449,7 @@ done:
 		xfs_warn(mp,
 			"invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]",
 			mp->m_logbsize);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	if (iosizelog) {
@@ -465,7 +458,7 @@ done:
 			xfs_warn(mp, "invalid log iosize: %d [not %d-%d]",
 				iosizelog, XFS_MIN_IO_LOG,
 				XFS_MAX_IO_LOG);
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 
 		mp->m_flags |= XFS_MOUNT_DFLT_IOSIZE;
@@ -597,15 +590,20 @@ xfs_max_file_offset(
 	return (((__uint64_t)pagefactor) << bitshift) - 1;
 }
 
+/*
+ * 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.
+ */
 xfs_agnumber_t
-xfs_set_inode32(struct xfs_mount *mp)
+xfs_set_inode32(struct xfs_mount *mp, xfs_agnumber_t agcount)
 {
 	xfs_agnumber_t	index = 0;
 	xfs_agnumber_t	maxagi = 0;
 	xfs_sb_t	*sbp = &mp->m_sb;
 	xfs_agnumber_t	max_metadata;
-	xfs_agino_t	agino =	XFS_OFFBNO_TO_AGINO(mp, sbp->sb_agblocks -1, 0);
-	xfs_ino_t	ino = XFS_AGINO_TO_INO(mp, sbp->sb_agcount -1, agino);
+	xfs_agino_t	agino;
+	xfs_ino_t	ino;
 	xfs_perag_t	*pag;
 
 	/* Calculate how much should be reserved for inodes to meet
@@ -620,10 +618,12 @@ xfs_set_inode32(struct xfs_mount *mp)
 		do_div(icount, sbp->sb_agblocks);
 		max_metadata = icount;
 	} else {
-		max_metadata = sbp->sb_agcount;
+		max_metadata = agcount;
 	}
 
-	for (index = 0; index < sbp->sb_agcount; index++) {
+	agino =	XFS_OFFBNO_TO_AGINO(mp, sbp->sb_agblocks - 1, 0);
+
+	for (index = 0; index < agcount; index++) {
 		ino = XFS_AGINO_TO_INO(mp, index, agino);
 
 		if (ino > XFS_MAXINUMBER_32) {
@@ -648,11 +648,11 @@ xfs_set_inode32(struct xfs_mount *mp)
 }
 
 xfs_agnumber_t
-xfs_set_inode64(struct xfs_mount *mp)
+xfs_set_inode64(struct xfs_mount *mp, xfs_agnumber_t agcount)
 {
 	xfs_agnumber_t index = 0;
 
-	for (index = 0; index < mp->m_sb.sb_agcount; index++) {
+	for (index = 0; index < agcount; index++) {
 		struct xfs_perag	*pag;
 
 		pag = xfs_perag_get(mp, index);
@@ -686,7 +686,7 @@ xfs_blkdev_get(
 		xfs_warn(mp, "Invalid device [%s], error=%d\n", name, error);
 	}
 
-	return -error;
+	return error;
 }
 
 STATIC void
@@ -756,7 +756,7 @@ xfs_open_devices(
 		if (rtdev == ddev || rtdev == logdev) {
 			xfs_warn(mp,
 	"Cannot mount filesystem with identical rtdev and ddev/logdev.");
-			error = EINVAL;
+			error = -EINVAL;
 			goto out_close_rtdev;
 		}
 	}
@@ -764,21 +764,19 @@ xfs_open_devices(
 	/*
 	 * Setup xfs_mount buffer target pointers
 	 */
-	error = ENOMEM;
-	mp->m_ddev_targp = xfs_alloc_buftarg(mp, ddev, 0, mp->m_fsname);
+	error = -ENOMEM;
+	mp->m_ddev_targp = xfs_alloc_buftarg(mp, ddev);
 	if (!mp->m_ddev_targp)
 		goto out_close_rtdev;
 
 	if (rtdev) {
-		mp->m_rtdev_targp = xfs_alloc_buftarg(mp, rtdev, 1,
-							mp->m_fsname);
+		mp->m_rtdev_targp = xfs_alloc_buftarg(mp, rtdev);
 		if (!mp->m_rtdev_targp)
 			goto out_free_ddev_targ;
 	}
 
 	if (logdev && logdev != ddev) {
-		mp->m_logdev_targp = xfs_alloc_buftarg(mp, logdev, 1,
-							mp->m_fsname);
+		mp->m_logdev_targp = xfs_alloc_buftarg(mp, logdev);
 		if (!mp->m_logdev_targp)
 			goto out_free_rtdev_targ;
 	} else {
@@ -811,8 +809,7 @@ xfs_setup_devices(
 {
 	int			error;
 
-	error = xfs_setsize_buftarg(mp->m_ddev_targp, mp->m_sb.sb_blocksize,
-				    mp->m_sb.sb_sectsize);
+	error = xfs_setsize_buftarg(mp->m_ddev_targp, mp->m_sb.sb_sectsize);
 	if (error)
 		return error;
 
@@ -822,14 +819,12 @@ xfs_setup_devices(
 		if (xfs_sb_version_hassector(&mp->m_sb))
 			log_sector_size = mp->m_sb.sb_logsectsize;
 		error = xfs_setsize_buftarg(mp->m_logdev_targp,
-					    mp->m_sb.sb_blocksize,
 					    log_sector_size);
 		if (error)
 			return error;
 	}
 	if (mp->m_rtdev_targp) {
 		error = xfs_setsize_buftarg(mp->m_rtdev_targp,
-					    mp->m_sb.sb_blocksize,
 					    mp->m_sb.sb_sectsize);
 		if (error)
 			return error;
@@ -913,7 +908,7 @@ xfs_flush_inodes(
 	struct super_block	*sb = mp->m_super;
 
 	if (down_read_trylock(&sb->s_umount)) {
-		sync_inodes_sb(sb, jiffies);
+		sync_inodes_sb(sb);
 		up_read(&sb->s_umount);
 	}
 }
@@ -996,7 +991,7 @@ xfs_fs_evict_inode(
 
 	trace_xfs_evict_inode(ip);
 
-	truncate_inode_pages(&inode->i_data, 0);
+	truncate_inode_pages_final(&inode->i_data);
 	clear_inode(inode);
 	XFS_STATS_INC(vn_rele);
 	XFS_STATS_INC(vn_remove);
@@ -1193,10 +1188,12 @@ xfs_fs_remount(
 	char			*options)
 {
 	struct xfs_mount	*mp = XFS_M(sb);
+	xfs_sb_t		*sbp = &mp->m_sb;
 	substring_t		args[MAX_OPT_ARGS];
 	char			*p;
 	int			error;
 
+	sync_filesystem(sb);
 	while ((p = strsep(&options, ",")) != NULL) {
 		int token;
 
@@ -1212,10 +1209,10 @@ xfs_fs_remount(
 			mp->m_flags &= ~XFS_MOUNT_BARRIER;
 			break;
 		case Opt_inode64:
-			mp->m_maxagi = xfs_set_inode64(mp);
+			mp->m_maxagi = xfs_set_inode64(mp, sbp->sb_agcount);
 			break;
 		case Opt_inode32:
-			mp->m_maxagi = xfs_set_inode32(mp);
+			mp->m_maxagi = xfs_set_inode32(mp, sbp->sb_agcount);
 			break;
 		default:
 			/*
@@ -1299,7 +1296,7 @@ xfs_fs_freeze(
 
 	xfs_save_resvblks(mp);
 	xfs_quiesce_attr(mp);
-	return -xfs_fs_log_dummy(mp);
+	return xfs_fs_log_dummy(mp);
 }
 
 STATIC int
@@ -1318,7 +1315,7 @@ xfs_fs_show_options(
 	struct seq_file		*m,
 	struct dentry		*root)
 {
-	return -xfs_showargs(XFS_M(root->d_sb), m);
+	return xfs_showargs(XFS_M(root->d_sb), m);
 }
 
 /*
@@ -1340,14 +1337,14 @@ xfs_finish_flags(
 			   mp->m_logbsize < mp->m_sb.sb_logsunit) {
 			xfs_warn(mp,
 		"logbuf size must be greater than or equal to log stripe size");
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 	} else {
 		/* Fail a mount if the logbuf is larger than 32K */
 		if (mp->m_logbsize > XLOG_BIG_RECORD_BSIZE) {
 			xfs_warn(mp,
 		"logbuf size for version 1 logs must be 16K or 32K");
-			return XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 	}
 
@@ -1359,7 +1356,7 @@ xfs_finish_flags(
 		xfs_warn(mp,
 "Cannot mount a V5 filesystem as %s. %s is always enabled for V5 filesystems.",
 			MNTOPT_NOATTR2, MNTOPT_ATTR2);
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	/*
@@ -1376,7 +1373,7 @@ xfs_finish_flags(
 	if ((mp->m_sb.sb_flags & XFS_SBF_READONLY) && !ronly) {
 		xfs_warn(mp,
 			"cannot mount a read-only filesystem as read-write");
-		return XFS_ERROR(EROFS);
+		return -EROFS;
 	}
 
 	if ((mp->m_qflags & (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE)) &&
@@ -1384,7 +1381,7 @@ xfs_finish_flags(
 	    !xfs_sb_version_has_pquotino(&mp->m_sb)) {
 		xfs_warn(mp,
 		  "Super block does not support project and group quota together");
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	return 0;
@@ -1398,7 +1395,7 @@ xfs_fs_fill_super(
 {
 	struct inode		*root;
 	struct xfs_mount	*mp = NULL;
-	int			flags = 0, error = ENOMEM;
+	int			flags = 0, error = -ENOMEM;
 
 	mp = kzalloc(sizeof(struct xfs_mount), GFP_KERNEL);
 	if (!mp)
@@ -1478,12 +1475,12 @@ xfs_fs_fill_super(
 
 	root = igrab(VFS_I(mp->m_rootip));
 	if (!root) {
-		error = ENOENT;
+		error = -ENOENT;
 		goto out_unmount;
 	}
 	sb->s_root = d_make_root(root);
 	if (!sb->s_root) {
-		error = ENOMEM;
+		error = -ENOMEM;
 		goto out_unmount;
 	}
 
@@ -1503,7 +1500,7 @@ out_destroy_workqueues:
 	xfs_free_fsname(mp);
 	kfree(mp);
  out:
-	return -error;
+	return error;
 
  out_unmount:
 	xfs_filestream_unmount(mp);
@@ -1753,13 +1750,9 @@ init_xfs_fs(void)
 	if (error)
 		goto out_destroy_wq;
 
-	error = xfs_filestream_init();
-	if (error)
-		goto out_mru_cache_uninit;
-
 	error = xfs_buf_init();
 	if (error)
-		goto out_filestream_uninit;
+		goto out_mru_cache_uninit;
 
 	error = xfs_init_procfs();
 	if (error)
@@ -1769,9 +1762,15 @@ init_xfs_fs(void)
 	if (error)
 		goto out_cleanup_procfs;
 
+	xfs_kset = kset_create_and_add("xfs", NULL, fs_kobj);
+	if (!xfs_kset) {
+		error = -ENOMEM;
+		goto out_sysctl_unregister;;
+	}
+
 	error = xfs_qm_init();
 	if (error)
-		goto out_sysctl_unregister;
+		goto out_kset_unregister;
 
 	error = register_filesystem(&xfs_fs_type);
 	if (error)
@@ -1780,14 +1779,14 @@ init_xfs_fs(void)
 
  out_qm_exit:
 	xfs_qm_exit();
+ out_kset_unregister:
+	kset_unregister(xfs_kset);
  out_sysctl_unregister:
 	xfs_sysctl_unregister();
  out_cleanup_procfs:
 	xfs_cleanup_procfs();
  out_buf_terminate:
 	xfs_buf_terminate();
- out_filestream_uninit:
-	xfs_filestream_uninit();
  out_mru_cache_uninit:
 	xfs_mru_cache_uninit();
  out_destroy_wq:
@@ -1803,10 +1802,10 @@ exit_xfs_fs(void)
 {
 	xfs_qm_exit();
 	unregister_filesystem(&xfs_fs_type);
+	kset_unregister(xfs_kset);
 	xfs_sysctl_unregister();
 	xfs_cleanup_procfs();
 	xfs_buf_terminate();
-	xfs_filestream_uninit();
 	xfs_mru_cache_uninit();
 	xfs_destroy_workqueues();
 	xfs_destroy_zones();
diff --git a/fs/xfs/xfs_super.h b/fs/xfs/xfs_super.h
index bbe3d15a7904..2b830c2f322e 100644
--- a/fs/xfs/xfs_super.h
+++ b/fs/xfs/xfs_super.h
@@ -44,16 +44,6 @@ extern void xfs_qm_exit(void);
 # define XFS_REALTIME_STRING
 #endif
 
-#if XFS_BIG_BLKNOS
-# if XFS_BIG_INUMS
-#  define XFS_BIGFS_STRING	"large block/inode numbers, "
-# else
-#  define XFS_BIGFS_STRING	"large block numbers, "
-# endif
-#else
-# define XFS_BIGFS_STRING
-#endif
-
 #ifdef DEBUG
 # define XFS_DBG_STRING		"debug"
 #else
@@ -64,7 +54,6 @@ extern void xfs_qm_exit(void);
 #define XFS_BUILD_OPTIONS	XFS_ACL_STRING \
 				XFS_SECURITY_STRING \
 				XFS_REALTIME_STRING \
-				XFS_BIGFS_STRING \
 				XFS_DBG_STRING /* DBG must be last */
 
 struct xfs_inode;
@@ -76,8 +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 *);
-extern xfs_agnumber_t xfs_set_inode64(struct xfs_mount *);
+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 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 14e58f2c96bd..6a944a2cd36f 100644
--- a/fs/xfs/xfs_symlink.c
+++ b/fs/xfs/xfs_symlink.c
@@ -27,6 +27,7 @@
 #include "xfs_ag.h"
 #include "xfs_mount.h"
 #include "xfs_da_format.h"
+#include "xfs_da_btree.h"
 #include "xfs_dir2.h"
 #include "xfs_inode.h"
 #include "xfs_ialloc.h"
@@ -75,11 +76,15 @@ xfs_readlink_bmap(
 		bp = xfs_buf_read(mp->m_ddev_targp, d, BTOBB(byte_cnt), 0,
 				  &xfs_symlink_buf_ops);
 		if (!bp)
-			return XFS_ERROR(ENOMEM);
+			return -ENOMEM;
 		error = bp->b_error;
 		if (error) {
 			xfs_buf_ioerror_alert(bp, __func__);
 			xfs_buf_relse(bp);
+
+			/* bad CRC means corrupted metadata */
+			if (error == -EFSBADCRC)
+				error = -EFSCORRUPTED;
 			goto out;
 		}
 		byte_cnt = XFS_SYMLINK_BUF_SPACE(mp, byte_cnt);
@@ -88,9 +93,9 @@ xfs_readlink_bmap(
 
 		cur_chunk = bp->b_addr;
 		if (xfs_sb_version_hascrc(&mp->m_sb)) {
-			if (!xfs_symlink_hdr_ok(mp, ip->i_ino, offset,
+			if (!xfs_symlink_hdr_ok(ip->i_ino, offset,
 							byte_cnt, bp)) {
-				error = EFSCORRUPTED;
+				error = -EFSCORRUPTED;
 				xfs_alert(mp,
 "symlink header does not match required off/len/owner (0x%x/Ox%x,0x%llx)",
 					offset, byte_cnt, ip->i_ino);
@@ -130,7 +135,7 @@ xfs_readlink(
 	trace_xfs_readlink(ip);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	xfs_ilock(ip, XFS_ILOCK_SHARED);
 
@@ -143,7 +148,7 @@ xfs_readlink(
 			 __func__, (unsigned long long) ip->i_ino,
 			 (long long) pathlen);
 		ASSERT(0);
-		error = XFS_ERROR(EFSCORRUPTED);
+		error = -EFSCORRUPTED;
 		goto out;
 	}
 
@@ -198,20 +203,17 @@ xfs_symlink(
 	trace_xfs_symlink(dp, link_name);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	/*
 	 * Check component lengths of the target path name.
 	 */
 	pathlen = strlen(target_path);
 	if (pathlen >= MAXPATHLEN)      /* total string too long */
-		return XFS_ERROR(ENAMETOOLONG);
+		return -ENAMETOOLONG;
 
 	udqp = gdqp = NULL;
-	if (dp->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)
-		prid = xfs_get_projid(dp);
-	else
-		prid = XFS_PROJID_DEFAULT;
+	prid = xfs_get_initial_prid(dp);
 
 	/*
 	 * Make sure that we have allocated dquot(s) on disk.
@@ -236,7 +238,7 @@ xfs_symlink(
 		fs_blocks = xfs_symlink_blocks(mp, pathlen);
 	resblks = XFS_SYMLINK_SPACE_RES(mp, link_name->len, fs_blocks);
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_symlink, resblks, 0);
-	if (error == ENOSPC && fs_blocks == 0) {
+	if (error == -ENOSPC && fs_blocks == 0) {
 		resblks = 0;
 		error = xfs_trans_reserve(tp, &M_RES(mp)->tr_symlink, 0, 0);
 	}
@@ -252,7 +254,7 @@ xfs_symlink(
 	 * Check whether the directory allows new symlinks or not.
 	 */
 	if (dp->i_d.di_flags & XFS_DIFLAG_NOSYMLINKS) {
-		error = XFS_ERROR(EPERM);
+		error = -EPERM;
 		goto error_return;
 	}
 
@@ -282,7 +284,7 @@ xfs_symlink(
 	error = xfs_dir_ialloc(&tp, dp, S_IFLNK | (mode & ~S_IFMT), 1, 0,
 			       prid, resblks > 0, &ip, NULL);
 	if (error) {
-		if (error == ENOSPC)
+		if (error == -ENOSPC)
 			goto error_return;
 		goto error1;
 	}
@@ -346,7 +348,7 @@ xfs_symlink(
 			bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
 					       BTOBB(byte_cnt), 0);
 			if (!bp) {
-				error = ENOMEM;
+				error = -ENOMEM;
 				goto error2;
 			}
 			bp->b_ops = &xfs_symlink_buf_ops;
@@ -487,7 +489,7 @@ xfs_inactive_symlink_rmt(
 			XFS_FSB_TO_DADDR(mp, mval[i].br_startblock),
 			XFS_FSB_TO_BB(mp, mval[i].br_blockcount), 0);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error_bmap_cancel;
 		}
 		xfs_trans_binval(tp, bp);
@@ -560,7 +562,7 @@ xfs_inactive_symlink(
 	trace_xfs_inactive_symlink(ip);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
 
@@ -578,7 +580,7 @@ xfs_inactive_symlink(
 			 __func__, (unsigned long long)ip->i_ino, pathlen);
 		xfs_iunlock(ip, XFS_ILOCK_EXCL);
 		ASSERT(0);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	if (ip->i_df.if_flags & XFS_IFINLINE) {
diff --git a/fs/xfs/xfs_sysfs.c b/fs/xfs/xfs_sysfs.c
new file mode 100644
index 000000000000..9835139ce1ec
--- /dev/null
+++ b/fs/xfs/xfs_sysfs.c
@@ -0,0 +1,165 @@
+/*
+ * Copyright (c) 2014 Red Hat, Inc.
+ * 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
+ */
+
+#include "xfs.h"
+#include "xfs_sysfs.h"
+#include "xfs_log_format.h"
+#include "xfs_log.h"
+#include "xfs_log_priv.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);
+};
+
+static inline struct xfs_sysfs_attr *
+to_attr(struct attribute *attr)
+{
+	return container_of(attr, struct xfs_sysfs_attr, attr);
+}
+
+#define XFS_SYSFS_ATTR_RW(name) \
+	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 ATTR_LIST(name) &xfs_sysfs_attr_##name.attr
+
+/*
+ * 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.
+ */
+
+struct kobj_type xfs_mp_ktype = {
+	.release = xfs_sysfs_release,
+};
+
+/* xlog */
+
+STATIC ssize_t
+log_head_lsn_show(
+	char	*buf,
+	void	*data)
+{
+	struct xlog *log = data;
+	int cycle;
+	int block;
+
+	spin_lock(&log->l_icloglock);
+	cycle = log->l_curr_cycle;
+	block = log->l_curr_block;
+	spin_unlock(&log->l_icloglock);
+
+	return snprintf(buf, PAGE_SIZE, "%d:%d\n", cycle, block);
+}
+XFS_SYSFS_ATTR_RO(log_head_lsn);
+
+STATIC ssize_t
+log_tail_lsn_show(
+	char	*buf,
+	void	*data)
+{
+	struct xlog *log = data;
+	int cycle;
+	int block;
+
+	xlog_crack_atomic_lsn(&log->l_tail_lsn, &cycle, &block);
+	return snprintf(buf, PAGE_SIZE, "%d:%d\n", cycle, block);
+}
+XFS_SYSFS_ATTR_RO(log_tail_lsn);
+
+STATIC ssize_t
+reserve_grant_head_show(
+	char	*buf,
+	void	*data)
+{
+	struct xlog *log = data;
+	int cycle;
+	int bytes;
+
+	xlog_crack_grant_head(&log->l_reserve_head.grant, &cycle, &bytes);
+	return snprintf(buf, PAGE_SIZE, "%d:%d\n", cycle, bytes);
+}
+XFS_SYSFS_ATTR_RO(reserve_grant_head);
+
+STATIC ssize_t
+write_grant_head_show(
+	char	*buf,
+	void	*data)
+{
+	struct xlog *log = data;
+	int cycle;
+	int bytes;
+
+	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);
+}
+
+STATIC ssize_t
+xfs_log_show(
+	struct kobject		*kobject,
+	struct attribute	*attr,
+	char			*buf)
+{
+	struct xlog *log = to_xlog(kobject);
+	struct xfs_sysfs_attr *xfs_attr = to_attr(attr);
+
+	return xfs_attr->show ? xfs_attr->show(buf, log) : 0;
+}
+
+STATIC ssize_t
+xfs_log_store(
+	struct kobject		*kobject,
+	struct attribute	*attr,
+	const char		*buf,
+	size_t			count)
+{
+	struct xlog *log = to_xlog(kobject);
+	struct xfs_sysfs_attr *xfs_attr = to_attr(attr);
+
+	return xfs_attr->store ? xfs_attr->store(buf, count, log) : 0;
+}
+
+static struct sysfs_ops xfs_log_ops = {
+	.show = xfs_log_show,
+	.store = xfs_log_store,
+};
+
+struct kobj_type xfs_log_ktype = {
+	.release = xfs_sysfs_release,
+	.sysfs_ops = &xfs_log_ops,
+	.default_attrs = xfs_log_attrs,
+};
diff --git a/fs/xfs/xfs_sysfs.h b/fs/xfs/xfs_sysfs.h
new file mode 100644
index 000000000000..54a2091183c0
--- /dev/null
+++ b/fs/xfs/xfs_sysfs.h
@@ -0,0 +1,59 @@
+/*
+ * Copyright (c) 2014 Red Hat, Inc.
+ * 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_SYSFS_H__
+#define __XFS_SYSFS_H__
+
+extern struct kobj_type xfs_mp_ktype;	/* xfs_mount */
+extern struct kobj_type xfs_log_ktype;	/* xlog */
+
+static inline struct xfs_kobj *
+to_kobj(struct kobject *kobject)
+{
+	return container_of(kobject, struct xfs_kobj, kobject);
+}
+
+static inline void
+xfs_sysfs_release(struct kobject *kobject)
+{
+	struct xfs_kobj *kobj = to_kobj(kobject);
+	complete(&kobj->complete);
+}
+
+static inline int
+xfs_sysfs_init(
+	struct xfs_kobj		*kobj,
+	struct kobj_type	*ktype,
+	struct xfs_kobj		*parent_kobj,
+	const char		*name)
+{
+	init_completion(&kobj->complete);
+	return kobject_init_and_add(&kobj->kobject, ktype,
+				    &parent_kobj->kobject, "%s", name);
+}
+
+static inline void
+xfs_sysfs_del(
+	struct xfs_kobj	*kobj)
+{
+	kobject_del(&kobj->kobject);
+	kobject_put(&kobj->kobject);
+	wait_for_completion(&kobj->complete);
+}
+
+#endif	/* __XFS_SYSFS_H__ */
diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c
index dee3279c095e..1e85bcd0e418 100644
--- a/fs/xfs/xfs_trace.c
+++ b/fs/xfs/xfs_trace.c
@@ -46,6 +46,7 @@
 #include "xfs_log_recover.h"
 #include "xfs_inode_item.h"
 #include "xfs_bmap_btree.h"
+#include "xfs_filestream.h"
 
 /*
  * We include this last to have the helpers above available for the trace
diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
index 425dfa45b9a0..152f82782630 100644
--- a/fs/xfs/xfs_trace.h
+++ b/fs/xfs/xfs_trace.h
@@ -538,6 +538,64 @@ DEFINE_BUF_ITEM_EVENT(xfs_trans_bhold_release);
 DEFINE_BUF_ITEM_EVENT(xfs_trans_binval);
 DEFINE_BUF_ITEM_EVENT(xfs_trans_buf_ordered);
 
+DECLARE_EVENT_CLASS(xfs_filestream_class,
+	TP_PROTO(struct xfs_inode *ip, xfs_agnumber_t agno),
+	TP_ARGS(ip, agno),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(xfs_agnumber_t, agno)
+		__field(int, streams)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(ip)->i_sb->s_dev;
+		__entry->ino = ip->i_ino;
+		__entry->agno = agno;
+		__entry->streams = xfs_filestream_peek_ag(ip->i_mount, agno);
+	),
+	TP_printk("dev %d:%d ino 0x%llx agno %u streams %d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __entry->agno,
+		  __entry->streams)
+)
+#define DEFINE_FILESTREAM_EVENT(name) \
+DEFINE_EVENT(xfs_filestream_class, name, \
+	TP_PROTO(struct xfs_inode *ip, xfs_agnumber_t agno), \
+	TP_ARGS(ip, agno))
+DEFINE_FILESTREAM_EVENT(xfs_filestream_free);
+DEFINE_FILESTREAM_EVENT(xfs_filestream_lookup);
+DEFINE_FILESTREAM_EVENT(xfs_filestream_scan);
+
+TRACE_EVENT(xfs_filestream_pick,
+	TP_PROTO(struct xfs_inode *ip, xfs_agnumber_t agno,
+		 xfs_extlen_t free, int nscan),
+	TP_ARGS(ip, agno, free, nscan),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(xfs_agnumber_t, agno)
+		__field(int, streams)
+		__field(xfs_extlen_t, free)
+		__field(int, nscan)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(ip)->i_sb->s_dev;
+		__entry->ino = ip->i_ino;
+		__entry->agno = agno;
+		__entry->streams = xfs_filestream_peek_ag(ip->i_mount, agno);
+		__entry->free = free;
+		__entry->nscan = nscan;
+	),
+	TP_printk("dev %d:%d ino 0x%llx agno %u streams %d free %d nscan %d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __entry->agno,
+		  __entry->streams,
+		  __entry->free,
+		  __entry->nscan)
+);
+
 DECLARE_EVENT_CLASS(xfs_lock_class,
 	TP_PROTO(struct xfs_inode *ip, unsigned lock_flags,
 		 unsigned long caller_ip),
@@ -603,6 +661,8 @@ DEFINE_INODE_EVENT(xfs_readlink);
 DEFINE_INODE_EVENT(xfs_inactive_symlink);
 DEFINE_INODE_EVENT(xfs_alloc_file_space);
 DEFINE_INODE_EVENT(xfs_free_file_space);
+DEFINE_INODE_EVENT(xfs_zero_file_space);
+DEFINE_INODE_EVENT(xfs_collapse_file_space);
 DEFINE_INODE_EVENT(xfs_readdir);
 #ifdef CONFIG_XFS_POSIX_ACL
 DEFINE_INODE_EVENT(xfs_get_acl);
@@ -1058,7 +1118,6 @@ DEFINE_RW_EVENT(xfs_file_read);
 DEFINE_RW_EVENT(xfs_file_buffered_write);
 DEFINE_RW_EVENT(xfs_file_direct_write);
 DEFINE_RW_EVENT(xfs_file_splice_read);
-DEFINE_RW_EVENT(xfs_file_splice_write);
 
 DECLARE_EVENT_CLASS(xfs_page_class,
 	TP_PROTO(struct inode *inode, struct page *page, unsigned long off,
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index c812c5c060de..30e8e3410955 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -190,7 +190,7 @@ xfs_trans_reserve(
 					  -((int64_t)blocks), rsvd);
 		if (error != 0) {
 			current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
-			return (XFS_ERROR(ENOSPC));
+			return -ENOSPC;
 		}
 		tp->t_blk_res += blocks;
 	}
@@ -241,7 +241,7 @@ xfs_trans_reserve(
 		error = xfs_mod_incore_sb(tp->t_mountp, XFS_SBS_FREXTENTS,
 					  -((int64_t)rtextents), rsvd);
 		if (error) {
-			error = XFS_ERROR(ENOSPC);
+			error = -ENOSPC;
 			goto undo_log;
 		}
 		tp->t_rtx_res += rtextents;
@@ -827,7 +827,7 @@ xfs_trans_committed_bulk(
 		xfs_log_item_batch_insert(ailp, &cur, log_items, i, commit_lsn);
 
 	spin_lock(&ailp->xa_lock);
-	xfs_trans_ail_cursor_done(ailp, &cur);
+	xfs_trans_ail_cursor_done(&cur);
 	spin_unlock(&ailp->xa_lock);
 }
 
@@ -874,7 +874,7 @@ xfs_trans_commit(
 		goto out_unreserve;
 
 	if (XFS_FORCED_SHUTDOWN(mp)) {
-		error = XFS_ERROR(EIO);
+		error = -EIO;
 		goto out_unreserve;
 	}
 
@@ -887,12 +887,7 @@ xfs_trans_commit(
 		xfs_trans_apply_sb_deltas(tp);
 	xfs_trans_apply_dquot_deltas(tp);
 
-	error = xfs_log_commit_cil(mp, tp, &commit_lsn, flags);
-	if (error == ENOMEM) {
-		xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
-		error = XFS_ERROR(EIO);
-		goto out_unreserve;
-	}
+	xfs_log_commit_cil(mp, tp, &commit_lsn, flags);
 
 	current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
 	xfs_trans_free(tp);
@@ -902,10 +897,7 @@ xfs_trans_commit(
 	 * log out now and wait for it.
 	 */
 	if (sync) {
-		if (!error) {
-			error = _xfs_log_force_lsn(mp, commit_lsn,
-				      XFS_LOG_SYNC, NULL);
-		}
+		error = _xfs_log_force_lsn(mp, commit_lsn, XFS_LOG_SYNC, NULL);
 		XFS_STATS_INC(xs_trans_sync);
 	} else {
 		XFS_STATS_INC(xs_trans_async);
@@ -925,7 +917,7 @@ out_unreserve:
 	if (tp->t_ticket) {
 		commit_lsn = xfs_log_done(mp, tp->t_ticket, NULL, log_flags);
 		if (commit_lsn == -1 && !error)
-			error = XFS_ERROR(EIO);
+			error = -EIO;
 	}
 	current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
 	xfs_trans_free_items(tp, NULLCOMMITLSN, error ? XFS_TRANS_ABORT : 0);
@@ -1032,7 +1024,7 @@ xfs_trans_roll(
 	 */
 	error = xfs_trans_commit(trans, 0);
 	if (error)
-		return (error);
+		return error;
 
 	trans = *tpp;
 
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index 9b96d35e483d..b5bc1ab3c4da 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -64,7 +64,7 @@ typedef struct xfs_log_item {
 
 struct xfs_item_ops {
 	void (*iop_size)(xfs_log_item_t *, int *, int *);
-	void (*iop_format)(xfs_log_item_t *, struct xfs_log_iovec *);
+	void (*iop_format)(xfs_log_item_t *, struct xfs_log_vec *);
 	void (*iop_pin)(xfs_log_item_t *);
 	void (*iop_unpin)(xfs_log_item_t *, int remove);
 	uint (*iop_push)(struct xfs_log_item *, struct list_head *);
diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c
index a7287354e535..859482f53b5a 100644
--- a/fs/xfs/xfs_trans_ail.c
+++ b/fs/xfs/xfs_trans_ail.c
@@ -173,7 +173,6 @@ xfs_trans_ail_cursor_next(
  */
 void
 xfs_trans_ail_cursor_done(
-	struct xfs_ail		*ailp,
 	struct xfs_ail_cursor	*cur)
 {
 	cur->item = NULL;
@@ -368,7 +367,7 @@ xfsaild_push(
 		 * If the AIL is empty or our push has reached the end we are
 		 * done now.
 		 */
-		xfs_trans_ail_cursor_done(ailp, &cur);
+		xfs_trans_ail_cursor_done(&cur);
 		spin_unlock(&ailp->xa_lock);
 		goto out_done;
 	}
@@ -453,7 +452,7 @@ xfsaild_push(
 			break;
 		lsn = lip->li_lsn;
 	}
-	xfs_trans_ail_cursor_done(ailp, &cur);
+	xfs_trans_ail_cursor_done(&cur);
 	spin_unlock(&ailp->xa_lock);
 
 	if (xfs_buf_delwri_submit_nowait(&ailp->xa_buf_list))
@@ -763,7 +762,7 @@ xfs_trans_ail_init(
 
 	ailp = kmem_zalloc(sizeof(struct xfs_ail), KM_MAYFAIL);
 	if (!ailp)
-		return ENOMEM;
+		return -ENOMEM;
 
 	ailp->xa_mount = mp;
 	INIT_LIST_HEAD(&ailp->xa_ail);
@@ -782,7 +781,7 @@ xfs_trans_ail_init(
 
 out_free_ailp:
 	kmem_free(ailp);
-	return ENOMEM;
+	return -ENOMEM;
 }
 
 void
diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c
index c035d11b7734..96c898e7ac9a 100644
--- a/fs/xfs/xfs_trans_buf.c
+++ b/fs/xfs/xfs_trans_buf.c
@@ -166,7 +166,7 @@ xfs_trans_get_buf_map(
 		ASSERT(atomic_read(&bip->bli_refcount) > 0);
 		bip->bli_recur++;
 		trace_xfs_trans_get_buf_recur(bip);
-		return (bp);
+		return bp;
 	}
 
 	bp = xfs_buf_get_map(target, map, nmaps, flags);
@@ -178,7 +178,7 @@ xfs_trans_get_buf_map(
 
 	_xfs_trans_bjoin(tp, bp, 1);
 	trace_xfs_trans_get_buf(bp->b_fspriv);
-	return (bp);
+	return bp;
 }
 
 /*
@@ -201,9 +201,8 @@ xfs_trans_getsb(xfs_trans_t	*tp,
 	 * Default to just trying to lock the superblock buffer
 	 * if tp is NULL.
 	 */
-	if (tp == NULL) {
-		return (xfs_getsb(mp, flags));
-	}
+	if (tp == NULL)
+		return xfs_getsb(mp, flags);
 
 	/*
 	 * If the superblock buffer already has this transaction
@@ -218,7 +217,7 @@ xfs_trans_getsb(xfs_trans_t	*tp,
 		ASSERT(atomic_read(&bip->bli_refcount) > 0);
 		bip->bli_recur++;
 		trace_xfs_trans_getsb_recur(bip);
-		return (bp);
+		return bp;
 	}
 
 	bp = xfs_getsb(mp, flags);
@@ -227,7 +226,7 @@ xfs_trans_getsb(xfs_trans_t	*tp,
 
 	_xfs_trans_bjoin(tp, bp, 1);
 	trace_xfs_trans_getsb(bp->b_fspriv);
-	return (bp);
+	return bp;
 }
 
 #ifdef DEBUG
@@ -267,7 +266,7 @@ xfs_trans_read_buf_map(
 		bp = xfs_buf_read_map(target, map, nmaps, flags, ops);
 		if (!bp)
 			return (flags & XBF_TRYLOCK) ?
-					EAGAIN : XFS_ERROR(ENOMEM);
+					-EAGAIN : -ENOMEM;
 
 		if (bp->b_error) {
 			error = bp->b_error;
@@ -275,6 +274,10 @@ xfs_trans_read_buf_map(
 			XFS_BUF_UNDONE(bp);
 			xfs_buf_stale(bp);
 			xfs_buf_relse(bp);
+
+			/* bad CRC means corrupted metadata */
+			if (error == -EFSBADCRC)
+				error = -EFSCORRUPTED;
 			return error;
 		}
 #ifdef DEBUG
@@ -283,7 +286,7 @@ xfs_trans_read_buf_map(
 				if (((xfs_req_num++) % xfs_error_mod) == 0) {
 					xfs_buf_relse(bp);
 					xfs_debug(mp, "Returning error!");
-					return XFS_ERROR(EIO);
+					return -EIO;
 				}
 			}
 		}
@@ -314,7 +317,18 @@ xfs_trans_read_buf_map(
 			ASSERT(bp->b_iodone == NULL);
 			XFS_BUF_READ(bp);
 			bp->b_ops = ops;
-			xfsbdstrat(tp->t_mountp, bp);
+
+			/*
+			 * XXX(hch): clean up the error handling here to be less
+			 * of a mess..
+			 */
+			if (XFS_FORCED_SHUTDOWN(mp)) {
+				trace_xfs_bdstrat_shut(bp, _RET_IP_);
+				xfs_bioerror_relse(bp);
+			} else {
+				xfs_buf_iorequest(bp);
+			}
+
 			error = xfs_buf_iowait(bp);
 			if (error) {
 				xfs_buf_ioerror_alert(bp, __func__);
@@ -327,6 +341,9 @@ xfs_trans_read_buf_map(
 				if (tp->t_flags & XFS_TRANS_DIRTY)
 					xfs_force_shutdown(tp->t_mountp,
 							SHUTDOWN_META_IO_ERROR);
+				/* bad CRC means corrupted metadata */
+				if (error == -EFSBADCRC)
+					error = -EFSCORRUPTED;
 				return error;
 			}
 		}
@@ -337,7 +354,7 @@ xfs_trans_read_buf_map(
 		if (XFS_FORCED_SHUTDOWN(mp)) {
 			trace_xfs_trans_read_buf_shut(bp, _RET_IP_);
 			*bpp = NULL;
-			return XFS_ERROR(EIO);
+			return -EIO;
 		}
 
 
@@ -354,7 +371,7 @@ xfs_trans_read_buf_map(
 	if (bp == NULL) {
 		*bpp = NULL;
 		return (flags & XBF_TRYLOCK) ?
-					0 : XFS_ERROR(ENOMEM);
+					0 : -ENOMEM;
 	}
 	if (bp->b_error) {
 		error = bp->b_error;
@@ -364,6 +381,10 @@ xfs_trans_read_buf_map(
 		if (tp->t_flags & XFS_TRANS_DIRTY)
 			xfs_force_shutdown(tp->t_mountp, SHUTDOWN_META_IO_ERROR);
 		xfs_buf_relse(bp);
+
+		/* bad CRC means corrupted metadata */
+		if (error == -EFSBADCRC)
+			error = -EFSCORRUPTED;
 		return error;
 	}
 #ifdef DEBUG
@@ -374,7 +395,7 @@ xfs_trans_read_buf_map(
 						   SHUTDOWN_META_IO_ERROR);
 				xfs_buf_relse(bp);
 				xfs_debug(mp, "Returning trans error!");
-				return XFS_ERROR(EIO);
+				return -EIO;
 			}
 		}
 	}
@@ -392,7 +413,7 @@ shutdown_abort:
 	trace_xfs_trans_read_buf_shut(bp, _RET_IP_);
 	xfs_buf_relse(bp);
 	*bpp = NULL;
-	return XFS_ERROR(EIO);
+	return -EIO;
 }
 
 /*
diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c
index cd2a10e15d3a..846e061c2e98 100644
--- a/fs/xfs/xfs_trans_dquot.c
+++ b/fs/xfs/xfs_trans_dquot.c
@@ -295,8 +295,8 @@ xfs_trans_mod_dquot(
 /*
  * Given an array of dqtrx structures, lock all the dquots associated and join
  * them to the transaction, provided they have been modified.  We know that the
- * highest number of dquots of one type - usr, grp OR prj - involved in a
- * transaction is 2 so we don't need to make this very generic.
+ * highest number of dquots of one type - usr, grp and prj - involved in a
+ * transaction is 3 so we don't need to make this very generic.
  */
 STATIC void
 xfs_trans_dqlockedjoin(
@@ -722,8 +722,8 @@ xfs_trans_dqresv(
 error_return:
 	xfs_dqunlock(dqp);
 	if (flags & XFS_QMOPT_ENOSPC)
-		return ENOSPC;
-	return EDQUOT;
+		return -ENOSPC;
+	return -EDQUOT;
 }
 
 
diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h
index 12e86af9d9b9..bd1281862ad7 100644
--- a/fs/xfs/xfs_trans_priv.h
+++ b/fs/xfs/xfs_trans_priv.h
@@ -133,8 +133,7 @@ struct xfs_log_item *	xfs_trans_ail_cursor_last(struct xfs_ail *ailp,
 					xfs_lsn_t lsn);
 struct xfs_log_item *	xfs_trans_ail_cursor_next(struct xfs_ail *ailp,
 					struct xfs_ail_cursor *cur);
-void			xfs_trans_ail_cursor_done(struct xfs_ail *ailp,
-					struct xfs_ail_cursor *cur);
+void			xfs_trans_ail_cursor_done(struct xfs_ail_cursor *cur);
 
 #if BITS_PER_LONG != 64
 static inline void
diff --git a/fs/xfs/xfs_types.h b/fs/xfs/xfs_types.h
index 82bbc34d54a3..b79dc66b2ecd 100644
--- a/fs/xfs/xfs_types.h
+++ b/fs/xfs/xfs_types.h
@@ -38,43 +38,18 @@ typedef	__int32_t	xfs_tid_t;	/* transaction identifier */
 typedef	__uint32_t	xfs_dablk_t;	/* dir/attr block number (in file) */
 typedef	__uint32_t	xfs_dahash_t;	/* dir/attr hash value */
 
-/*
- * These types are 64 bits on disk but are either 32 or 64 bits in memory.
- * Disk based types:
- */
-typedef __uint64_t	xfs_dfsbno_t;	/* blockno in filesystem (agno|agbno) */
-typedef __uint64_t	xfs_drfsbno_t;	/* blockno in filesystem (raw) */
-typedef	__uint64_t	xfs_drtbno_t;	/* extent (block) in realtime area */
-typedef	__uint64_t	xfs_dfiloff_t;	/* block number in a file */
-typedef	__uint64_t	xfs_dfilblks_t;	/* number of blocks in a file */
-
-/*
- * Memory based types are conditional.
- */
-#if XFS_BIG_BLKNOS
 typedef	__uint64_t	xfs_fsblock_t;	/* blockno in filesystem (agno|agbno) */
 typedef __uint64_t	xfs_rfsblock_t;	/* blockno in filesystem (raw) */
 typedef __uint64_t	xfs_rtblock_t;	/* extent (block) in realtime area */
-typedef	__int64_t	xfs_srtblock_t;	/* signed version of xfs_rtblock_t */
-#else
-typedef	__uint32_t	xfs_fsblock_t;	/* blockno in filesystem (agno|agbno) */
-typedef __uint32_t	xfs_rfsblock_t;	/* blockno in filesystem (raw) */
-typedef __uint32_t	xfs_rtblock_t;	/* extent (block) in realtime area */
-typedef	__int32_t	xfs_srtblock_t;	/* signed version of xfs_rtblock_t */
-#endif
 typedef __uint64_t	xfs_fileoff_t;	/* block number in a file */
-typedef __int64_t	xfs_sfiloff_t;	/* signed block number in a file */
 typedef __uint64_t	xfs_filblks_t;	/* number of blocks in a file */
 
+typedef	__int64_t	xfs_srtblock_t;	/* signed version of xfs_rtblock_t */
+typedef __int64_t	xfs_sfiloff_t;	/* signed block number in a file */
 
 /*
  * Null values for the types.
  */
-#define	NULLDFSBNO	((xfs_dfsbno_t)-1)
-#define	NULLDRFSBNO	((xfs_drfsbno_t)-1)
-#define	NULLDRTBNO	((xfs_drtbno_t)-1)
-#define	NULLDFILOFF	((xfs_dfiloff_t)-1)
-
 #define	NULLFSBLOCK	((xfs_fsblock_t)-1)
 #define	NULLRFSBLOCK	((xfs_rfsblock_t)-1)
 #define	NULLRTBLOCK	((xfs_rtblock_t)-1)
@@ -134,7 +109,7 @@ typedef enum {
 
 typedef enum {
 	XFS_BTNUM_BNOi, XFS_BTNUM_CNTi, XFS_BTNUM_BMAPi, XFS_BTNUM_INOi,
-	XFS_BTNUM_MAX
+	XFS_BTNUM_FINOi, XFS_BTNUM_MAX
 } xfs_btnum_t;
 
 struct xfs_name {
diff --git a/fs/xfs/xfs_vnode.h b/fs/xfs/xfs_vnode.h
deleted file mode 100644
index 3e8e797c6d11..000000000000
--- a/fs/xfs/xfs_vnode.h
+++ /dev/null
@@ -1,55 +0,0 @@
-/*
- * Copyright (c) 2000-2005 Silicon Graphics, Inc.
- * 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_VNODE_H__
-#define __XFS_VNODE_H__
-
-#include "xfs_fs.h"
-
-struct file;
-struct xfs_inode;
-struct attrlist_cursor_kern;
-
-/*
- * Flags for read/write calls - same values as IRIX
- */
-#define IO_ISDIRECT	0x00004		/* bypass page cache */
-#define IO_INVIS	0x00020		/* don't update inode timestamps */
-
-#define XFS_IO_FLAGS \
-	{ IO_ISDIRECT,	"DIRECT" }, \
-	{ IO_INVIS,	"INVIS"}
-
-/*
- * Flush/Invalidate options for vop_toss/flush/flushinval_pages.
- */
-#define FI_NONE			0	/* none */
-#define FI_REMAPF		1	/* Do a remapf prior to the operation */
-#define FI_REMAPF_LOCKED	2	/* Do a remapf prior to the operation.
-					   Prevent VM access to the pages until
-					   the operation completes. */
-
-/*
- * Some useful predicates.
- */
-#define VN_MAPPED(vp)	mapping_mapped(vp->i_mapping)
-#define VN_CACHED(vp)	(vp->i_mapping->nrpages)
-#define VN_DIRTY(vp)	mapping_tagged(vp->i_mapping, \
-					PAGECACHE_TAG_DIRTY)
-
-
-#endif	/* __XFS_VNODE_H__ */
diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c
index 9d479073ba41..93455b998041 100644
--- a/fs/xfs/xfs_xattr.c
+++ b/fs/xfs/xfs_xattr.c
@@ -49,7 +49,7 @@ xfs_xattr_get(struct dentry *dentry, const char *name,
 		value = NULL;
 	}
 
-	error = -xfs_attr_get(ip, (unsigned char *)name, value, &asize, xflags);
+	error = xfs_attr_get(ip, (unsigned char *)name, value, &asize, xflags);
 	if (error)
 		return error;
 	return asize;
@@ -71,8 +71,8 @@ xfs_xattr_set(struct dentry *dentry, const char *name, const void *value,
 		xflags |= ATTR_REPLACE;
 
 	if (!value)
-		return -xfs_attr_remove(ip, (unsigned char *)name, xflags);
-	return -xfs_attr_set(ip, (unsigned char *)name,
+		return xfs_attr_remove(ip, (unsigned char *)name, xflags);
+	return xfs_attr_set(ip, (unsigned char *)name,
 				(void *)value, size, xflags);
 }
 
@@ -102,8 +102,8 @@ const struct xattr_handler *xfs_xattr_handlers[] = {
 	&xfs_xattr_trusted_handler,
 	&xfs_xattr_security_handler,
 #ifdef CONFIG_XFS_POSIX_ACL
-	&xfs_xattr_acl_access_handler,
-	&xfs_xattr_acl_default_handler,
+	&posix_acl_access_xattr_handler,
+	&posix_acl_default_xattr_handler,
 #endif
 	NULL
 };