122 files changed, 3630 insertions, 3287 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 844e288b9576..53e95b2a1369 100644
--- a/fs/xfs/kmem.c
+++ b/fs/xfs/kmem.c
@@ -21,7 +21,6 @@
 #include <linux/swap.h>
 #include <linux/blkdev.h>
 #include <linux/backing-dev.h>
-#include "time.h"
 #include "kmem.h"
 #include "xfs_message.h"
 
diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/libxfs/xfs_ag.h
index 6e247a99f5db..6e247a99f5db 100644
--- a/fs/xfs/xfs_ag.h
+++ b/fs/xfs/libxfs/xfs_ag.h
diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index d43813267a80..eff34218f405 100644
--- a/fs/xfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -483,9 +483,9 @@ xfs_agfl_read_verify(
 		return;
 
 	if (!xfs_buf_verify_cksum(bp, XFS_AGFL_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_agfl_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -503,7 +503,7 @@ xfs_agfl_write_verify(
 		return;
 
 	if (!xfs_agfl_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -559,7 +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;
@@ -2209,6 +2209,10 @@ xfs_agf_verify(
 	      be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp)))
 		return false;
 
+	if (be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]) > XFS_BTREE_MAXLEVELS ||
+	    be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]) > XFS_BTREE_MAXLEVELS)
+		return false;
+
 	/*
 	 * during growfs operations, the perag is not fully initialised,
 	 * so we can't use it for any useful checking. growfs ensures we can't
@@ -2234,11 +2238,11 @@ xfs_agf_read_verify(
 
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	    !xfs_buf_verify_cksum(bp, XFS_AGF_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		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);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -2252,7 +2256,7 @@ xfs_agf_write_verify(
 	struct xfs_buf_log_item	*bip = bp->b_fspriv;
 
 	if (!xfs_agf_verify(mp, bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -2601,11 +2605,11 @@ 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);
@@ -2617,7 +2621,7 @@ 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 8358f1ded94d..e0e83e24d3ef 100644
--- a/fs/xfs/xfs_alloc_btree.c
+++ b/fs/xfs/libxfs/xfs_alloc_btree.c
@@ -355,9 +355,9 @@ xfs_allocbt_read_verify(
 	struct xfs_buf	*bp)
 {
 	if (!xfs_btree_sblock_verify_crc(bp))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_allocbt_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
@@ -371,7 +371,7 @@ xfs_allocbt_write_verify(
 {
 	if (!xfs_allocbt_verify(bp)) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
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 bfe36fc2cdc2..353fb425faef 100644
--- a/fs/xfs/xfs_attr.c
+++ b/fs/xfs/libxfs/xfs_attr.c
@@ -85,7 +85,7 @@ xfs_attr_args_init(
 {
 
 	if (!name)
-		return EINVAL;
+		return -EINVAL;
 
 	memset(args, 0, sizeof(*args));
 	args->geo = dp->i_mount->m_attr_geo;
@@ -95,7 +95,7 @@ xfs_attr_args_init(
 	args->name = name;
 	args->namelen = strlen((const char *)name);
 	if (args->namelen >= MAXNAMELEN)
-		return EFAULT;		/* match IRIX behaviour */
+		return -EFAULT;		/* match IRIX behaviour */
 
 	args->hashval = xfs_da_hashname(args->name, args->namelen);
 	return 0;
@@ -131,10 +131,10 @@ xfs_attr_get(
 	XFS_STATS_INC(xs_attr_get);
 
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
-		return EIO;
+		return -EIO;
 
 	if (!xfs_inode_hasattr(ip))
-		return ENOATTR;
+		return -ENOATTR;
 
 	error = xfs_attr_args_init(&args, ip, name, flags);
 	if (error)
@@ -145,7 +145,7 @@ xfs_attr_get(
 
 	lock_mode = xfs_ilock_attr_map_shared(ip);
 	if (!xfs_inode_hasattr(ip))
-		error = ENOATTR;
+		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))
@@ -155,7 +155,7 @@ xfs_attr_get(
 	xfs_iunlock(ip, lock_mode);
 
 	*valuelenp = args.valuelen;
-	return error == EEXIST ? 0 : error;
+	return error == -EEXIST ? 0 : error;
 }
 
 /*
@@ -213,7 +213,7 @@ xfs_attr_set(
 	XFS_STATS_INC(xs_attr_set);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return EIO;
+		return -EIO;
 
 	error = xfs_attr_args_init(&args, dp, name, flags);
 	if (error)
@@ -304,7 +304,7 @@ xfs_attr_set(
 		 * 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).
@@ -419,10 +419,10 @@ xfs_attr_remove(
 	XFS_STATS_INC(xs_attr_remove);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return EIO;
+		return -EIO;
 
 	if (!xfs_inode_hasattr(dp))
-		return ENOATTR;
+		return -ENOATTR;
 
 	error = xfs_attr_args_init(&args, dp, name, flags);
 	if (error)
@@ -477,7 +477,7 @@ xfs_attr_remove(
 	xfs_trans_ijoin(args.trans, dp, 0);
 
 	if (!xfs_inode_hasattr(dp)) {
-		error = XFS_ERROR(ENOATTR);
+		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);
@@ -534,28 +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;
 }
 
 
@@ -592,10 +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;
@@ -626,7 +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
@@ -642,7 +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;
 		}
 
 		/*
@@ -658,13 +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;
 	}
 
 	/*
@@ -673,7 +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
@@ -684,7 +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;
 	}
 
 	/*
@@ -700,7 +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
@@ -714,7 +714,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 		if (args->rmtblkno) {
 			error = xfs_attr_rmtval_remove(args);
 			if (error)
-				return(error);
+				return error;
 		}
 
 		/*
@@ -744,7 +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;
 			}
 
 			/*
@@ -795,7 +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;
 	}
@@ -850,7 +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;
 	}
@@ -906,9 +906,9 @@ 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;
 
@@ -933,7 +933,7 @@ restart:
 	}
 
 	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
@@ -1031,7 +1031,7 @@ restart:
 	if (args->rmtblkno > 0) {
 		error = xfs_attr_rmtval_set(args);
 		if (error)
-			return(error);
+			return error;
 	}
 
 	/*
@@ -1061,7 +1061,7 @@ restart:
 		if (args->rmtblkno) {
 			error = xfs_attr_rmtval_remove(args);
 			if (error)
-				return(error);
+				return error;
 		}
 
 		/*
@@ -1134,8 +1134,8 @@ out:
 	if (state)
 		xfs_da_state_free(state);
 	if (error)
-		return(error);
-	return(retval);
+		return error;
+	return retval;
 }
 
 /*
@@ -1168,7 +1168,7 @@ xfs_attr_node_removename(xfs_da_args_t *args)
 	 * 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;
@@ -1297,7 +1297,7 @@ xfs_attr_node_removename(xfs_da_args_t *args)
 
 out:
 	xfs_da_state_free(state);
-	return(error);
+	return error;
 }
 
 /*
@@ -1345,7 +1345,7 @@ xfs_attr_fillstate(xfs_da_state_t *state)
 		}
 	}
 
-	return(0);
+	return 0;
 }
 
 /*
@@ -1376,7 +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;
 		}
@@ -1395,13 +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;
 }
 
 /*
@@ -1431,7 +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);
@@ -1455,5 +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 28712d29e43c..b1f73dbbf3d8 100644
--- a/fs/xfs/xfs_attr_leaf.c
+++ b/fs/xfs/libxfs/xfs_attr_leaf.c
@@ -214,7 +214,7 @@ xfs_attr3_leaf_write_verify(
 	struct xfs_attr3_leaf_hdr *hdr3 = bp->b_addr;
 
 	if (!xfs_attr3_leaf_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -242,9 +242,9 @@ xfs_attr3_leaf_read_verify(
 
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	     !xfs_buf_verify_cksum(bp, XFS_ATTR3_LEAF_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_attr3_leaf_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -547,7 +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
@@ -582,7 +582,7 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
 
 	xfs_sbversion_add_attr2(mp, args->trans);
 
-	return(0);
+	return 0;
 }
 
 /*
@@ -611,9 +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;
 }
 
 /*
@@ -640,18 +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;
 }
 
 /*
@@ -691,7 +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 */
@@ -730,9 +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);
@@ -741,7 +741,7 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
 
 out:
 	kmem_free(tmpbuffer);
-	return(error);
+	return error;
 }
 
 /*
@@ -769,12 +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);
@@ -809,7 +809,7 @@ xfs_attr3_leaf_to_shortform(
 
 	tmpbuffer = kmem_alloc(args->geo->blksize, KM_SLEEP);
 	if (!tmpbuffer)
-		return ENOMEM;
+		return -ENOMEM;
 
 	memcpy(tmpbuffer, bp->b_addr, args->geo->blksize);
 
@@ -1017,10 +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;
 
@@ -1031,7 +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()
@@ -1053,7 +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;
 }
 
 /*
@@ -1108,7 +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.
@@ -1121,7 +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;
 	}
 
@@ -1692,7 +1692,7 @@ xfs_attr3_leaf_toosmall(
 		ichdr.usedbytes;
 	if (bytes > (state->args->geo->blksize >> 1)) {
 		*action = 0;	/* blk over 50%, don't try to join */
-		return(0);
+		return 0;
 	}
 
 	/*
@@ -1711,7 +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 {
@@ -1740,7 +1740,7 @@ 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);
 
@@ -1757,7 +1757,7 @@ xfs_attr3_leaf_toosmall(
 	}
 	if (i >= 2) {
 		*action = 0;
-		return(0);
+		return 0;
 	}
 
 	/*
@@ -1773,13 +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;
 }
 
 /*
@@ -2123,7 +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;
 	}
 
 	/*
@@ -2152,7 +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)
@@ -2168,11 +2168,11 @@ xfs_attr3_leaf_lookup_int(
 			args->rmtblkcnt = xfs_attr3_rmt_blocks(
 							args->dp->i_mount,
 							args->rmtvaluelen);
-			return XFS_ERROR(EEXIST);
+			return -EEXIST;
 		}
 	}
 	args->index = probe;
-	return XFS_ERROR(ENOATTR);
+	return -ENOATTR;
 }
 
 /*
@@ -2208,7 +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);
@@ -2226,7 +2226,7 @@ xfs_attr3_leaf_getvalue(
 		}
 		if (args->valuelen < args->rmtvaluelen) {
 			args->valuelen = args->rmtvaluelen;
-			return XFS_ERROR(ERANGE);
+			return -ERANGE;
 		}
 		args->valuelen = args->rmtvaluelen;
 	}
@@ -2481,7 +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];
@@ -2548,7 +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
diff --git a/fs/xfs/xfs_attr_leaf.h b/fs/xfs/libxfs/xfs_attr_leaf.h
index e2929da7c3ba..e2929da7c3ba 100644
--- a/fs/xfs/xfs_attr_leaf.h
+++ b/fs/xfs/libxfs/xfs_attr_leaf.h
diff --git a/fs/xfs/xfs_attr_remote.c b/fs/xfs/libxfs/xfs_attr_remote.c
index b5adfecbb8ee..7510ab8058a4 100644
--- a/fs/xfs/xfs_attr_remote.c
+++ b/fs/xfs/libxfs/xfs_attr_remote.c
@@ -138,11 +138,11 @@ xfs_attr3_rmt_read_verify(
 
 	while (len > 0) {
 		if (!xfs_verify_cksum(ptr, blksize, XFS_ATTR3_RMT_CRC_OFF)) {
-			xfs_buf_ioerror(bp, EFSBADCRC);
+			xfs_buf_ioerror(bp, -EFSBADCRC);
 			break;
 		}
 		if (!xfs_attr3_rmt_verify(mp, ptr, blksize, bno)) {
-			xfs_buf_ioerror(bp, EFSCORRUPTED);
+			xfs_buf_ioerror(bp, -EFSCORRUPTED);
 			break;
 		}
 		len -= blksize;
@@ -178,7 +178,7 @@ xfs_attr3_rmt_write_verify(
 
 	while (len > 0) {
 		if (!xfs_attr3_rmt_verify(mp, ptr, blksize, bno)) {
-			xfs_buf_ioerror(bp, EFSCORRUPTED);
+			xfs_buf_ioerror(bp, -EFSCORRUPTED);
 			xfs_verifier_error(bp);
 			return;
 		}
@@ -257,7 +257,7 @@ xfs_attr_rmtval_copyout(
 				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);
 		}
@@ -452,7 +452,7 @@ xfs_attr_rmtval_set(
 			ASSERT(committed);
 			args->trans = NULL;
 			xfs_bmap_cancel(args->flist);
-			return(error);
+			return error;
 		}
 
 		/*
@@ -473,7 +473,7 @@ xfs_attr_rmtval_set(
 		 */
 		error = xfs_trans_roll(&args->trans, dp);
 		if (error)
-			return (error);
+			return error;
 	}
 
 	/*
@@ -498,7 +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));
@@ -508,7 +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,
@@ -563,7 +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));
@@ -622,7 +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 e1649c0d3e02..e1649c0d3e02 100644
--- a/fs/xfs/xfs_bit.h
+++ b/fs/xfs/libxfs/xfs_bit.h
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 75c3fe5f3d9d..79c981984dca 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -392,7 +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);
 
@@ -1033,7 +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;
@@ -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;
@@ -1690,7 +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)
@@ -3323,7 +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.
 		 */
@@ -3348,7 +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);
@@ -4051,11 +4047,11 @@ xfs_bmapi_read(
 	     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);
 
@@ -4246,11 +4242,11 @@ xfs_bmapi_delay(
 	     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);
 
@@ -4469,7 +4465,7 @@ xfs_bmapi_convert_unwritten(
 	 * so generate another request.
 	 */
 	if (mval->br_blockcount < len)
-		return EAGAIN;
+		return -EAGAIN;
 	return 0;
 }
 
@@ -4540,11 +4536,11 @@ xfs_bmapi_write(
 	     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);
 
@@ -4620,7 +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;
@@ -4922,7 +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,
@@ -4930,7 +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.
@@ -4958,7 +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);
@@ -5076,11 +5072,11 @@ 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);
@@ -5325,7 +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,
@@ -5408,39 +5404,237 @@ error0:
 }
 
 /*
+ * Determine whether an extent shift can be accomplished by a merge with the
+ * extent that precedes the target hole of the shift.
+ */
+STATIC bool
+xfs_bmse_can_merge(
+	struct xfs_bmbt_irec	*left,	/* preceding extent */
+	struct xfs_bmbt_irec	*got,	/* current extent to shift */
+	xfs_fileoff_t		shift)	/* shift fsb */
+{
+	xfs_fileoff_t		startoff;
+
+	startoff = got->br_startoff - shift;
+
+	/*
+	 * The extent, once shifted, must be adjacent in-file and on-disk with
+	 * the preceding extent.
+	 */
+	if ((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))
+		return false;
+
+	return true;
+}
+
+/*
+ * A bmap extent shift adjusts the file offset of an extent to fill a preceding
+ * hole in the file. If an extent shift would result in the extent being fully
+ * adjacent to the extent that currently precedes the hole, we can merge with
+ * the preceding extent rather than do the shift.
+ *
+ * This function assumes the caller has verified a shift-by-merge is possible
+ * with the provided extents via xfs_bmse_can_merge().
+ */
+STATIC int
+xfs_bmse_merge(
+	struct xfs_inode		*ip,
+	int				whichfork,
+	xfs_fileoff_t			shift,		/* shift fsb */
+	int				current_ext,	/* idx of gotp */
+	struct xfs_bmbt_rec_host	*gotp,		/* extent to shift */
+	struct xfs_bmbt_rec_host	*leftp,		/* preceding extent */
+	struct xfs_btree_cur		*cur,
+	int				*logflags)	/* output */
+{
+	struct xfs_ifork		*ifp;
+	struct xfs_bmbt_irec		got;
+	struct xfs_bmbt_irec		left;
+	xfs_filblks_t			blockcount;
+	int				error, i;
+
+	ifp = XFS_IFORK_PTR(ip, whichfork);
+	xfs_bmbt_get_all(gotp, &got);
+	xfs_bmbt_get_all(leftp, &left);
+	blockcount = left.br_blockcount + got.br_blockcount;
+
+	ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
+	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
+	ASSERT(xfs_bmse_can_merge(&left, &got, shift));
+
+	/*
+	 * Merge the in-core extents. Note that the host record pointers and
+	 * current_ext index are invalid once the extent has been removed via
+	 * xfs_iext_remove().
+	 */
+	xfs_bmbt_set_blockcount(leftp, blockcount);
+	xfs_iext_remove(ip, current_ext, 1, 0);
+
+	/*
+	 * Update the on-disk extent count, the btree if necessary and log the
+	 * inode.
+	 */
+	XFS_IFORK_NEXT_SET(ip, whichfork,
+			   XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
+	*logflags |= XFS_ILOG_CORE;
+	if (!cur) {
+		*logflags |= XFS_ILOG_DEXT;
+		return 0;
+	}
+
+	/* lookup and remove the extent to merge */
+	error = xfs_bmbt_lookup_eq(cur, got.br_startoff, got.br_startblock,
+				   got.br_blockcount, &i);
+	if (error)
+		goto out_error;
+	XFS_WANT_CORRUPTED_GOTO(i == 1, out_error);
+
+	error = xfs_btree_delete(cur, &i);
+	if (error)
+		goto out_error;
+	XFS_WANT_CORRUPTED_GOTO(i == 1, out_error);
+
+	/* lookup and update size of the previous extent */
+	error = xfs_bmbt_lookup_eq(cur, left.br_startoff, left.br_startblock,
+				   left.br_blockcount, &i);
+	if (error)
+		goto out_error;
+	XFS_WANT_CORRUPTED_GOTO(i == 1, out_error);
+
+	left.br_blockcount = blockcount;
+
+	error = xfs_bmbt_update(cur, left.br_startoff, left.br_startblock,
+				left.br_blockcount, left.br_state);
+	if (error)
+		goto out_error;
+
+	return 0;
+
+out_error:
+	return error;
+}
+
+/*
+ * Shift a single extent.
+ */
+STATIC int
+xfs_bmse_shift_one(
+	struct xfs_inode		*ip,
+	int				whichfork,
+	xfs_fileoff_t			offset_shift_fsb,
+	int				*current_ext,
+	struct xfs_bmbt_rec_host	*gotp,
+	struct xfs_btree_cur		*cur,
+	int				*logflags)
+{
+	struct xfs_ifork		*ifp;
+	xfs_fileoff_t			startoff;
+	struct xfs_bmbt_rec_host	*leftp;
+	struct xfs_bmbt_irec		got;
+	struct xfs_bmbt_irec		left;
+	int				error;
+	int				i;
+
+	ifp = XFS_IFORK_PTR(ip, whichfork);
+
+	xfs_bmbt_get_all(gotp, &got);
+	startoff = got.br_startoff - offset_shift_fsb;
+
+	/* delalloc extents should be prevented by caller */
+	XFS_WANT_CORRUPTED_GOTO(!isnullstartblock(got.br_startblock),
+				out_error);
+
+	/*
+	 * If this is the first extent in the file, make sure there's enough
+	 * room at the start of the file and jump right to the shift as there's
+	 * no left extent to merge.
+	 */
+	if (*current_ext == 0) {
+		if (got.br_startoff < offset_shift_fsb)
+			return -EINVAL;
+		goto shift_extent;
+	}
+
+	/* grab the left extent and check for a large enough hole */
+	leftp = xfs_iext_get_ext(ifp, *current_ext - 1);
+	xfs_bmbt_get_all(leftp, &left);
+
+	if (startoff < left.br_startoff + left.br_blockcount)
+		return -EINVAL;
+
+	/* check whether to merge the extent or shift it down */
+	if (!xfs_bmse_can_merge(&left, &got, offset_shift_fsb))
+		goto shift_extent;
+
+	return xfs_bmse_merge(ip, whichfork, offset_shift_fsb, *current_ext,
+			      gotp, leftp, cur, logflags);
+
+shift_extent:
+	/*
+	 * Increment the extent index for the next iteration, update the start
+	 * offset of the in-core extent and update the btree if applicable.
+	 */
+	(*current_ext)++;
+	xfs_bmbt_set_startoff(gotp, startoff);
+	*logflags |= XFS_ILOG_CORE;
+	if (!cur) {
+		*logflags |= XFS_ILOG_DEXT;
+		return 0;
+	}
+
+	error = xfs_bmbt_lookup_eq(cur, got.br_startoff, got.br_startblock,
+				   got.br_blockcount, &i);
+	if (error)
+		return error;
+	XFS_WANT_CORRUPTED_GOTO(i == 1, out_error);
+
+	got.br_startoff = startoff;
+	error = xfs_bmbt_update(cur, got.br_startoff, got.br_startblock,
+				got.br_blockcount, got.br_state);
+	if (error)
+		return error;
+
+	return 0;
+
+out_error:
+	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.
+ * The maximum number of extents to be shifted in a single operation is
+ * @num_exts. @start_fsb specifies the file offset to start the shift and the
+ * file offset where we've left off is returned in @next_fsb. @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,
+	int			*done,
+	xfs_fileoff_t		*next_fsb,
 	xfs_fsblock_t		*firstblock,
 	struct xfs_bmap_free	*flist,
 	int			num_exts)
 {
-	struct xfs_btree_cur		*cur;
+	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;
+	xfs_extnum_t			current_ext;
 	int				error = 0;
-	int				i;
 	int				whichfork = XFS_DATA_FORK;
-	int				logflags;
-	xfs_filblks_t			blockcount = 0;
+	int				logflags = 0;
 	int				total_extents;
 
 	if (unlikely(XFS_TEST_ERROR(
@@ -5449,13 +5643,14 @@ xfs_bmap_shift_extents(
 	     mp, XFS_ERRTAG_BMAPIFORMAT, XFS_RANDOM_BMAPIFORMAT))) {
 		XFS_ERROR_REPORT("xfs_bmap_shift_extents",
 				 XFS_ERRLEVEL_LOW, mp);
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
-	ASSERT(current_ext != NULL);
+	ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
+	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
 	ifp = XFS_IFORK_PTR(ip, whichfork);
 	if (!(ifp->if_flags & XFS_IFEXTENTS)) {
@@ -5465,142 +5660,62 @@ xfs_bmap_shift_extents(
 			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;
-		}
-	}
-
-	/* We are going to change core inode */
-	logflags = XFS_ILOG_CORE;
 	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;
-	} else {
-		cur = NULL;
-		logflags |= XFS_ILOG_DEXT;
+	}
+
+	/*
+	 * Look up the extent index for the fsb where we start shifting. We can
+	 * henceforth iterate with current_ext as extent list changes are locked
+	 * out via ilock.
+	 *
+	 * 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.
+	 */
+	gotp = xfs_iext_bno_to_ext(ifp, start_fsb, &current_ext);
+	if (!gotp) {
+		*done = 1;
+		goto del_cursor;
 	}
 
 	/*
 	 * 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.
+	 * 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 = XFS_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 = XFS_ERROR(EINVAL);
-		}
-
+	while (nexts++ < num_exts && current_ext < total_extents) {
+		error = xfs_bmse_shift_one(ip, whichfork, offset_shift_fsb,
+					&current_ext, gotp, cur, &logflags);
 		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);
-			if (cur) {
-				error = xfs_btree_delete(cur, &i);
-				if (error)
-					goto del_cursor;
-				XFS_WANT_CORRUPTED_GOTO(i == 1, del_cursor);
-			}
-			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;
-		}
-
-		if (cur) {
-			error = xfs_bmbt_update(cur, got.br_startoff,
-						got.br_startblock,
-						got.br_blockcount,
-						got.br_state);
-			if (error)
-				goto del_cursor;
-		}
-
-		(*current_ext)++;
+		/* update total extent count and grab the next record */
 		total_extents = ifp->if_bytes / sizeof(xfs_bmbt_rec_t);
+		if (current_ext >= total_extents)
+			break;
+		gotp = xfs_iext_get_ext(ifp, current_ext);
 	}
 
 	/* Check if we are done */
-	if (*current_ext == total_extents)
+	if (current_ext == total_extents) {
 		*done = 1;
+	} else if (next_fsb) {
+		xfs_bmbt_get_all(gotp, &got);
+		*next_fsb = got.br_startoff;
+	}
 
 del_cursor:
 	if (cur)
 		xfs_btree_del_cursor(cur,
 			error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
 
-	xfs_trans_log_inode(tp, ip, logflags);
+	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 b879ca56a64c..44db6db86402 100644
--- a/fs/xfs/xfs_bmap.h
+++ b/fs/xfs/libxfs/xfs_bmap.h
@@ -178,9 +178,8 @@ 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);
+		xfs_fileoff_t start_fsb, xfs_fileoff_t offset_shift_fsb,
+		int *done, xfs_fileoff_t *next_fsb, 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 948836c4fd90..fba753308f31 100644
--- a/fs/xfs/xfs_bmap_btree.c
+++ b/fs/xfs/libxfs/xfs_bmap_btree.c
@@ -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 */
 }
 
 /*
@@ -438,8 +357,8 @@ 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;
@@ -554,7 +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);
@@ -763,11 +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;
 
@@ -779,9 +698,9 @@ xfs_bmbt_read_verify(
 	struct xfs_buf	*bp)
 {
 	if (!xfs_btree_lblock_verify_crc(bp))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_bmbt_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
@@ -795,7 +714,7 @@ xfs_bmbt_write_verify(
 {
 	if (!xfs_bmbt_verify(bp)) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -959,7 +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 819a8a4dee95..819a8a4dee95 100644
--- a/fs/xfs/xfs_bmap_btree.h
+++ b/fs/xfs/libxfs/xfs_bmap_btree.h
diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c
index cf893bc1e373..8fe6a93ff473 100644
--- a/fs/xfs/xfs_btree.c
+++ b/fs/xfs/libxfs/xfs_btree.c
@@ -78,11 +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)));
 
@@ -92,7 +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;
 }
@@ -140,7 +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;
 }
@@ -167,12 +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;
 }
@@ -595,7 +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);
 }
@@ -771,16 +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++;
@@ -852,7 +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 {
@@ -900,9 +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))
@@ -918,7 +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);
 }
@@ -929,7 +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);
 }
@@ -997,8 +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);
@@ -1140,7 +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);
@@ -1498,7 +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);
@@ -1597,7 +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);
@@ -4018,7 +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);
 }
@@ -4061,7 +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 a04b69422f67..8f18bab73ea5 100644
--- a/fs/xfs/xfs_btree.h
+++ b/fs/xfs/libxfs/xfs_btree.h
@@ -258,7 +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 a514ab616650..fd827530afec 100644
--- a/fs/xfs/xfs_da_btree.c
+++ b/fs/xfs/libxfs/xfs_da_btree.c
@@ -185,7 +185,7 @@ xfs_da3_node_write_verify(
 	struct xfs_da3_node_hdr *hdr3 = bp->b_addr;
 
 	if (!xfs_da3_node_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -214,13 +214,13 @@ xfs_da3_node_read_verify(
 	switch (be16_to_cpu(info->magic)) {
 		case XFS_DA3_NODE_MAGIC:
 			if (!xfs_buf_verify_cksum(bp, XFS_DA3_NODE_CRC_OFF)) {
-				xfs_buf_ioerror(bp, EFSBADCRC);
+				xfs_buf_ioerror(bp, -EFSBADCRC);
 				break;
 			}
 			/* fall through */
 		case XFS_DA_NODE_MAGIC:
 			if (!xfs_da3_node_verify(bp)) {
-				xfs_buf_ioerror(bp, EFSCORRUPTED);
+				xfs_buf_ioerror(bp, -EFSCORRUPTED);
 				break;
 			}
 			return;
@@ -315,7 +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;
@@ -337,7 +337,7 @@ xfs_da3_node_create(
 		XFS_DA_LOGRANGE(node, &node->hdr, dp->d_ops->node_hdr_size));
 
 	*bpp = bp;
-	return(0);
+	return 0;
 }
 
 /*
@@ -385,8 +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;
@@ -408,7 +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:
@@ -422,7 +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?
 			 */
@@ -439,7 +439,7 @@ xfs_da3_split(
 		xfs_da3_fixhashpath(state, &state->path);
 	}
 	if (!addblk)
-		return(0);
+		return 0;
 
 	/*
 	 * Split the root node.
@@ -449,7 +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 */
 	}
 
 	/*
@@ -492,7 +492,7 @@ xfs_da3_split(
 		    sizeof(node->hdr.info)));
 	}
 	addblk->bp = NULL;
-	return(0);
+	return 0;
 }
 
 /*
@@ -670,18 +670,18 @@ xfs_da3_node_split(
 		 */
 		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;
@@ -721,7 +721,7 @@ xfs_da3_node_split(
 		}
 	}
 
-	return(0);
+	return 0;
 }
 
 /*
@@ -963,9 +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:
@@ -985,7 +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);
@@ -995,12 +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
@@ -1010,7 +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
@@ -1099,7 +1099,7 @@ xfs_da3_root_join(
 	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;
 }
 
 /*
@@ -1142,7 +1142,7 @@ xfs_da3_node_toosmall(
 	dp->d_ops->node_hdr_from_disk(&nodehdr, node);
 	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 */
 	}
 
 	/*
@@ -1161,13 +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;
 	}
 
 	/*
@@ -1194,7 +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);
@@ -1486,7 +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);
@@ -1579,25 +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;
 }
 
 /*========================================================================
@@ -1692,7 +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);
@@ -1713,7 +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);
@@ -1726,7 +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;
 }
 
 /*
@@ -1772,7 +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);
@@ -1789,7 +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);
@@ -1801,7 +1801,7 @@ xfs_da3_blk_unlink(
 	}
 
 	xfs_trans_log_buf(args->trans, save_blk->bp, 0, sizeof(*save_info) - 1);
-	return(0);
+	return 0;
 }
 
 /*
@@ -1859,9 +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;
 	}
 
 	/*
@@ -1883,7 +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) ||
@@ -2004,7 +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;
 
@@ -2068,7 +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;
 	}
 
@@ -2158,7 +2158,7 @@ xfs_da3_swap_lastblock(
 	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.
@@ -2209,7 +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);
@@ -2231,7 +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);
@@ -2254,7 +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;
@@ -2267,7 +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);
@@ -2294,7 +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);
@@ -2305,7 +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);
@@ -2359,7 +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,
@@ -2427,7 +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;
 	}
 
@@ -2500,8 +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",
@@ -2561,9 +2561,10 @@ 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);
+		if (bp)
+			xfs_trans_brelse(trans, bp);
 		goto out_free;
 	}
 
diff --git a/fs/xfs/xfs_da_btree.h b/fs/xfs/libxfs/xfs_da_btree.h
index 6e153e399a77..6e153e399a77 100644
--- a/fs/xfs/xfs_da_btree.h
+++ b/fs/xfs/libxfs/xfs_da_btree.h
diff --git a/fs/xfs/xfs_da_format.c b/fs/xfs/libxfs/xfs_da_format.c
index c9aee52a37e2..7e42fdfd2f1d 100644
--- a/fs/xfs/xfs_da_format.c
+++ b/fs/xfs/libxfs/xfs_da_format.c
@@ -270,7 +270,6 @@ xfs_dir3_data_get_ftype(
 {
 	__uint8_t	ftype = dep->name[dep->namelen];
 
-	ASSERT(ftype < XFS_DIR3_FT_MAX);
 	if (ftype >= XFS_DIR3_FT_MAX)
 		return XFS_DIR3_FT_UNKNOWN;
 	return ftype;
diff --git a/fs/xfs/xfs_da_format.h b/fs/xfs/libxfs/xfs_da_format.h
index 0a49b0286372..0a49b0286372 100644
--- a/fs/xfs/xfs_da_format.h
+++ b/fs/xfs/libxfs/xfs_da_format.h
diff --git a/fs/xfs/xfs_dinode.h b/fs/xfs/libxfs/xfs_dinode.h
index 623bbe8fd921..623bbe8fd921 100644
--- a/fs/xfs/xfs_dinode.h
+++ b/fs/xfs/libxfs/xfs_dinode.h
diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/libxfs/xfs_dir2.c
index 79670cda48ae..7075aaf131f4 100644
--- a/fs/xfs/xfs_dir2.c
+++ b/fs/xfs/libxfs/xfs_dir2.c
@@ -108,7 +108,7 @@ xfs_da_mount(
 	if (!mp->m_dir_geo || !mp->m_attr_geo) {
 		kmem_free(mp->m_dir_geo);
 		kmem_free(mp->m_attr_geo);
-		return ENOMEM;
+		return -ENOMEM;
 	}
 
 	/* set up directory geometry */
@@ -202,7 +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;
 }
@@ -226,7 +226,7 @@ xfs_dir_init(
 
 	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
 	if (!args)
-		return ENOMEM;
+		return -ENOMEM;
 
 	args->geo = dp->i_mount->m_dir_geo;
 	args->dp = dp;
@@ -237,7 +237,8 @@ xfs_dir_init(
 }
 
 /*
-  Enter a name in a directory.
+ * Enter a name in a directory, or check for available space.
+ * If inum is 0, only the available space test is performed.
  */
 int
 xfs_dir_createname(
@@ -254,14 +255,16 @@ xfs_dir_createname(
 	int			v;		/* type-checking value */
 
 	ASSERT(S_ISDIR(dp->i_d.di_mode));
-	rval = xfs_dir_ino_validate(tp->t_mountp, inum);
-	if (rval)
-		return rval;
-	XFS_STATS_INC(xs_dir_create);
+	if (inum) {
+		rval = xfs_dir_ino_validate(tp->t_mountp, inum);
+		if (rval)
+			return rval;
+		XFS_STATS_INC(xs_dir_create);
+	}
 
 	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
 	if (!args)
-		return ENOMEM;
+		return -ENOMEM;
 
 	args->geo = dp->i_mount->m_dir_geo;
 	args->name = name->name;
@@ -276,6 +279,8 @@ xfs_dir_createname(
 	args->whichfork = XFS_DATA_FORK;
 	args->trans = tp;
 	args->op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT;
+	if (!inum)
+		args->op_flags |= XFS_DA_OP_JUSTCHECK;
 
 	if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
 		rval = xfs_dir2_sf_addname(args);
@@ -314,18 +319,18 @@ 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;
 }
 
 /*
@@ -392,7 +397,7 @@ xfs_dir_lookup(
 		rval = xfs_dir2_node_lookup(args);
 
 out_check_rval:
-	if (rval == EEXIST)
+	if (rval == -EEXIST)
 		rval = 0;
 	if (!rval) {
 		*inum = args->inumber;
@@ -428,7 +433,7 @@ xfs_dir_removename(
 
 	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
 	if (!args)
-		return ENOMEM;
+		return -ENOMEM;
 
 	args->geo = dp->i_mount->m_dir_geo;
 	args->name = name->name;
@@ -493,7 +498,7 @@ xfs_dir_replace(
 
 	args = kmem_zalloc(sizeof(*args), KM_SLEEP | KM_NOFS);
 	if (!args)
-		return ENOMEM;
+		return -ENOMEM;
 
 	args->geo = dp->i_mount->m_dir_geo;
 	args->name = name->name;
@@ -535,62 +540,14 @@ out_free:
 
 /*
  * See if this entry can be added to the directory without allocating space.
- * First checks that the caller couldn't reserve enough space (resblks = 0).
  */
 int
 xfs_dir_canenter(
 	xfs_trans_t	*tp,
 	xfs_inode_t	*dp,
-	struct xfs_name	*name,		/* name of entry to add */
-	uint		resblks)
+	struct xfs_name	*name)		/* name of entry to add */
 {
-	struct xfs_da_args *args;
-	int		rval;
-	int		v;		/* type-checking value */
-
-	if (resblks)
-		return 0;
-
-	ASSERT(S_ISDIR(dp->i_d.di_mode));
-
-	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);
-		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);
-out_free:
-	kmem_free(args);
-	return rval;
+	return xfs_dir_createname(tp, dp, name, 0, NULL, NULL, 0);
 }
 
 /*
diff --git a/fs/xfs/xfs_dir2.h b/fs/xfs/libxfs/xfs_dir2.h
index c8e86b0b5e99..4dff261e6ed5 100644
--- a/fs/xfs/xfs_dir2.h
+++ b/fs/xfs/libxfs/xfs_dir2.h
@@ -136,7 +136,7 @@ extern int xfs_dir_replace(struct xfs_trans *tp, struct xfs_inode *dp,
 				xfs_fsblock_t *first,
 				struct xfs_bmap_free *flist, xfs_extlen_t tot);
 extern int xfs_dir_canenter(struct xfs_trans *tp, struct xfs_inode *dp,
-				struct xfs_name *name, uint resblks);
+				struct xfs_name *name);
 
 /*
  * Direct call from the bmap code, bypassing the generic directory layer.
diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/libxfs/xfs_dir2_block.c
index c7cd3154026a..9628ceccfa02 100644
--- a/fs/xfs/xfs_dir2_block.c
+++ b/fs/xfs/libxfs/xfs_dir2_block.c
@@ -91,9 +91,9 @@ xfs_dir3_block_read_verify(
 
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	     !xfs_buf_verify_cksum(bp, XFS_DIR3_DATA_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_dir3_block_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -108,7 +108,7 @@ xfs_dir3_block_write_verify(
 	struct xfs_dir3_blk_hdr	*hdr3 = bp->b_addr;
 
 	if (!xfs_dir3_block_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -392,7 +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;
 	}
 
@@ -402,7 +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.
@@ -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;
 }
 
 /*
@@ -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;
 		}
 	}
 	/*
@@ -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;
 }
 
 /*
@@ -1091,7 +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;
diff --git a/fs/xfs/xfs_dir2_data.c b/fs/xfs/libxfs/xfs_dir2_data.c
index 8c2f6422648e..fdd803fecb8e 100644
--- a/fs/xfs/xfs_dir2_data.c
+++ b/fs/xfs/libxfs/xfs_dir2_data.c
@@ -100,7 +100,7 @@ __xfs_dir3_data_check(
 		break;
 	default:
 		XFS_ERROR_REPORT("Bad Magic", XFS_ERRLEVEL_LOW, mp);
-		return EFSCORRUPTED;
+		return -EFSCORRUPTED;
 	}
 
 	/*
@@ -256,7 +256,7 @@ xfs_dir3_data_reada_verify(
 		xfs_dir3_data_verify(bp);
 		return;
 	default:
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		break;
 	}
@@ -270,9 +270,9 @@ xfs_dir3_data_read_verify(
 
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	     !xfs_buf_verify_cksum(bp, XFS_DIR3_DATA_CRC_OFF))
-		 xfs_buf_ioerror(bp, EFSBADCRC);
+		 xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_dir3_data_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -287,7 +287,7 @@ xfs_dir3_data_write_verify(
 	struct xfs_dir3_blk_hdr	*hdr3 = bp->b_addr;
 
 	if (!xfs_dir3_data_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/libxfs/xfs_dir2_leaf.c
index fb0aad4440c1..a19174eb3cb2 100644
--- a/fs/xfs/xfs_dir2_leaf.c
+++ b/fs/xfs/libxfs/xfs_dir2_leaf.c
@@ -183,9 +183,9 @@ __read_verify(
 
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	     !xfs_buf_verify_cksum(bp, XFS_DIR3_LEAF_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_dir3_leaf_verify(bp, magic))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -201,7 +201,7 @@ __write_verify(
 	struct xfs_dir3_leaf_hdr *hdr3 = bp->b_addr;
 
 	if (!xfs_dir3_leaf_verify(bp, magic)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -731,7 +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.
@@ -755,7 +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
@@ -763,7 +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.
@@ -1198,7 +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;
 }
 
 /*
@@ -1327,13 +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;
 }
 
 /*
@@ -1440,7 +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;
@@ -1641,7 +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;
 	}
@@ -1815,7 +1815,7 @@ xfs_dir2_node_to_leaf(
 		 * 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 da43d304fca2..2ae6ac2c11ae 100644
--- a/fs/xfs/xfs_dir2_node.c
+++ b/fs/xfs/libxfs/xfs_dir2_node.c
@@ -117,9 +117,9 @@ xfs_dir3_free_read_verify(
 
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	    !xfs_buf_verify_cksum(bp, XFS_DIR3_FREE_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_dir3_free_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -134,7 +134,7 @@ xfs_dir3_free_write_verify(
 	struct xfs_dir3_blk_hdr	*hdr3 = bp->b_addr;
 
 	if (!xfs_dir3_free_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
@@ -406,7 +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
@@ -417,7 +417,7 @@ xfs_dir2_leafn_add(
 
 	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;
@@ -629,7 +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)
@@ -660,7 +660,7 @@ out:
 	 * Return the index, that will be the insertion point.
 	 */
 	*indexp = index;
-	return XFS_ERROR(ENOENT);
+	return -ENOENT;
 }
 
 /*
@@ -789,7 +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));
@@ -812,7 +812,7 @@ xfs_dir2_leafn_lookup_for_entry(
 		state->extravalid = 0;
 	}
 	*indexp = index;
-	return XFS_ERROR(ENOENT);
+	return -ENOENT;
 }
 
 /*
@@ -1133,7 +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
@@ -1287,7 +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;
 		}
 		/*
@@ -1599,7 +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;
 	}
 	/*
@@ -1628,7 +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;
 		}
 		/*
@@ -1815,7 +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.
@@ -1876,7 +1876,7 @@ xfs_dir2_node_addname_int(
 				}
 				XFS_ERROR_REPORT("xfs_dir2_node_addname_int",
 						 XFS_ERRLEVEL_LOW, mp);
-				return XFS_ERROR(EFSCORRUPTED);
+				return -EFSCORRUPTED;
 			}
 
 			/*
@@ -2042,8 +2042,8 @@ xfs_dir2_node_lookup(
 	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 *)
@@ -2096,7 +2096,7 @@ xfs_dir2_node_removename(
 		goto out_free;
 
 	/* Didn't find it, upper layer screwed up. */
-	if (rval != EEXIST) {
+	if (rval != -EEXIST) {
 		error = rval;
 		goto out_free;
 	}
@@ -2169,7 +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.
@@ -2272,7 +2272,7 @@ xfs_dir2_node_trim_free(
 		 * 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 27ce0794d196..27ce0794d196 100644
--- a/fs/xfs/xfs_dir2_priv.h
+++ b/fs/xfs/libxfs/xfs_dir2_priv.h
diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/libxfs/xfs_dir2_sf.c
index 53c3be619db5..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)
@@ -117,10 +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++;
@@ -251,7 +250,7 @@ xfs_dir2_block_to_sf(
 	logflags = XFS_ILOG_CORE;
 	error = xfs_dir2_shrink_inode(args, args->geo->datablk, bp);
 	if (error) {
-		ASSERT(error != ENOSPC);
+		ASSERT(error != -ENOSPC);
 		goto out;
 	}
 
@@ -299,7 +298,7 @@ xfs_dir2_sf_addname(
 
 	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);
 	/*
@@ -307,7 +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);
@@ -318,7 +317,7 @@ xfs_dir2_sf_addname(
 	 */
 	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?
 	 */
@@ -332,7 +331,7 @@ xfs_dir2_sf_addname(
 			 (uint)sizeof(xfs_dir2_ino4_t));
 		objchange = 1;
 	}
-#endif
+
 	new_isize = (int)dp->i_d.di_size + incr_isize;
 	/*
 	 * Won't fit as shortform any more (due to size),
@@ -345,7 +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.
 		 */
@@ -370,10 +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);
@@ -425,10 +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);
 }
@@ -516,10 +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.
 	 */
@@ -593,13 +586,8 @@ xfs_dir2_sf_addname_pick(
 	/*
 	 * 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).
 	 */
@@ -650,7 +638,6 @@ 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) +
@@ -738,7 +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);
@@ -751,7 +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 ..
@@ -761,7 +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.
@@ -781,20 +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;
 }
 
 /*
@@ -824,7 +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);
@@ -847,7 +834,7 @@ xfs_dir2_sf_removename(
 	 * Didn't find it.
 	 */
 	if (i == sfp->count)
-		return XFS_ERROR(ENOENT);
+		return -ENOENT;
 	/*
 	 * Calculate sizes.
 	 */
@@ -870,7 +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?
 	 */
@@ -880,7 +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;
@@ -895,12 +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 */
 
@@ -914,13 +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.
 	 */
@@ -951,17 +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);
 	}
 	/*
@@ -972,10 +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;
@@ -986,14 +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.
 	 */
@@ -1020,13 +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.
@@ -1181,4 +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 c2ac0c611ad8..bb969337efc8 100644
--- a/fs/xfs/xfs_dquot_buf.c
+++ b/fs/xfs/libxfs/xfs_dquot_buf.c
@@ -257,9 +257,9 @@ xfs_dquot_buf_read_verify(
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 
 	if (!xfs_dquot_buf_verify_crc(mp, bp))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_dquot_buf_verify(mp, bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -277,7 +277,7 @@ xfs_dquot_buf_write_verify(
 	struct xfs_mount	*mp = bp->b_target->bt_mount;
 
 	if (!xfs_dquot_buf_verify(mp, bp)) {
-		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 34d85aca3058..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
@@ -304,23 +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 5960e5593fe0..23dcb72fc5e6 100644
--- a/fs/xfs/xfs_ialloc.c
+++ b/fs/xfs/libxfs/xfs_ialloc.c
@@ -292,7 +292,7 @@ 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;
@@ -380,7 +380,7 @@ xfs_ialloc_ag_alloc(
 	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);
+		return -ENOSPC;
 	args.minlen = args.maxlen = args.mp->m_ialloc_blks;
 	/*
 	 * First try to allocate inodes contiguous with the last-allocated
@@ -1076,8 +1076,8 @@ xfs_dialloc_ag_finobt_newino(
 	int i;
 
 	if (agi->agi_newino != cpu_to_be32(NULLAGINO)) {
-		error = xfs_inobt_lookup(cur, agi->agi_newino, XFS_LOOKUP_EQ,
-					 &i);
+		error = xfs_inobt_lookup(cur, be32_to_cpu(agi->agi_newino),
+					 XFS_LOOKUP_EQ, &i);
 		if (error)
 			return error;
 		if (i == 1) {
@@ -1085,7 +1085,6 @@ xfs_dialloc_ag_finobt_newino(
 			if (error)
 				return error;
 			XFS_WANT_CORRUPTED_RETURN(i == 1);
-
 			return 0;
 		}
 	}
@@ -1385,7 +1384,7 @@ xfs_dialloc(
 		if (error) {
 			xfs_trans_brelse(tp, agbp);
 
-			if (error != ENOSPC)
+			if (error != -ENOSPC)
 				goto out_error;
 
 			xfs_perag_put(pag);
@@ -1416,7 +1415,7 @@ nextag:
 			agno = 0;
 		if (agno == start_agno) {
 			*inop = NULLFSINO;
-			return noroom ? ENOSPC : 0;
+			return noroom ? -ENOSPC : 0;
 		}
 	}
 
@@ -1425,7 +1424,7 @@ out_alloc:
 	return xfs_dialloc_ag(tp, agbp, parent, inop);
 out_error:
 	xfs_perag_put(pag);
-	return XFS_ERROR(error);
+	return error;
 }
 
 STATIC int
@@ -1682,7 +1681,7 @@ xfs_difree(
 		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);
+		return -EINVAL;
 	}
 	agino = XFS_INO_TO_AGINO(mp, inode);
 	if (inode != XFS_AGINO_TO_INO(mp, agno, agino))  {
@@ -1690,14 +1689,14 @@ xfs_difree(
 			__func__, (unsigned long long)inode,
 			(unsigned long long)XFS_AGINO_TO_INO(mp, agno, agino));
 		ASSERT(0);
-		return XFS_ERROR(EINVAL);
+		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 XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 	/*
 	 * Get the allocation group header.
@@ -1769,7 +1768,7 @@ xfs_imap_lookup(
 		if (i)
 			error = xfs_inobt_get_rec(cur, &rec, &i);
 		if (!error && i == 0)
-			error = EINVAL;
+			error = -EINVAL;
 	}
 
 	xfs_trans_brelse(tp, agbp);
@@ -1780,12 +1779,12 @@ xfs_imap_lookup(
 	/* check that the returned record contains the required inode */
 	if (rec.ir_startino > agino ||
 	    rec.ir_startino + mp->m_ialloc_inos <= agino)
-		return EINVAL;
+		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;
@@ -1829,7 +1828,7 @@ 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)",
@@ -1849,7 +1848,7 @@ xfs_imap(
 		}
 		xfs_stack_trace();
 #endif /* DEBUG */
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	blks_per_cluster = xfs_icluster_size_fsb(mp);
@@ -1922,7 +1921,7 @@ 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;
 }
@@ -2051,6 +2050,8 @@ xfs_agi_verify(
 	if (!XFS_AGI_GOOD_VERSION(be32_to_cpu(agi->agi_versionnum)))
 		return false;
 
+	if (be32_to_cpu(agi->agi_level) > XFS_BTREE_MAXLEVELS)
+		return false;
 	/*
 	 * during growfs operations, the perag is not fully initialised,
 	 * so we can't use it for any useful checking. growfs ensures we can't
@@ -2072,11 +2073,11 @@ xfs_agi_read_verify(
 
 	if (xfs_sb_version_hascrc(&mp->m_sb) &&
 	    !xfs_buf_verify_cksum(bp, XFS_AGI_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		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);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -2090,7 +2091,7 @@ xfs_agi_write_verify(
 	struct xfs_buf_log_item	*bip = bp->b_fspriv;
 
 	if (!xfs_agi_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
diff --git a/fs/xfs/xfs_ialloc.h b/fs/xfs/libxfs/xfs_ialloc.h
index 95ad1c002d60..95ad1c002d60 100644
--- a/fs/xfs/xfs_ialloc.h
+++ b/fs/xfs/libxfs/xfs_ialloc.h
diff --git a/fs/xfs/xfs_ialloc_btree.c b/fs/xfs/libxfs/xfs_ialloc_btree.c
index 726f83a681a5..c9b06f30fe86 100644
--- a/fs/xfs/xfs_ialloc_btree.c
+++ b/fs/xfs/libxfs/xfs_ialloc_btree.c
@@ -272,9 +272,9 @@ xfs_inobt_read_verify(
 	struct xfs_buf	*bp)
 {
 	if (!xfs_btree_sblock_verify_crc(bp))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_inobt_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
@@ -288,7 +288,7 @@ xfs_inobt_write_verify(
 {
 	if (!xfs_inobt_verify(bp)) {
 		trace_xfs_btree_corrupt(bp, _RET_IP_);
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
diff --git a/fs/xfs/xfs_ialloc_btree.h b/fs/xfs/libxfs/xfs_ialloc_btree.h
index d7ebea72c2d0..d7ebea72c2d0 100644
--- a/fs/xfs/xfs_ialloc_btree.h
+++ b/fs/xfs/libxfs/xfs_ialloc_btree.h
diff --git a/fs/xfs/xfs_inode_buf.c b/fs/xfs/libxfs/xfs_inode_buf.c
index cb35ae41d4a1..f18fd2da49f7 100644
--- a/fs/xfs/xfs_inode_buf.c
+++ b/fs/xfs/libxfs/xfs_inode_buf.c
@@ -101,7 +101,7 @@ xfs_inode_buf_verify(
 				return;
 			}
 
-			xfs_buf_ioerror(bp, EFSCORRUPTED);
+			xfs_buf_ioerror(bp, -EFSCORRUPTED);
 			xfs_verifier_error(bp);
 #ifdef DEBUG
 			xfs_alert(mp,
@@ -174,14 +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);
@@ -390,7 +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;
 	}
 
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 b031e8d0d928..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;
@@ -436,7 +436,7 @@ xfs_iread_extents(
 	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);
@@ -528,7 +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;
 	}
 
@@ -575,7 +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;
@@ -1692,7 +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 7d3b1ed6dcbe..7d3b1ed6dcbe 100644
--- a/fs/xfs/xfs_inode_fork.h
+++ b/fs/xfs/libxfs/xfs_inode_fork.h
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 ee7e0e80246b..ee7e0e80246b 100644
--- a/fs/xfs/xfs_log_rlimit.c
+++ b/fs/xfs/libxfs/xfs_log_rlimit.c
diff --git a/fs/xfs/xfs_quota_defs.h b/fs/xfs/libxfs/xfs_quota_defs.h
index 137e20937077..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)
diff --git a/fs/xfs/xfs_rtbitmap.c b/fs/xfs/libxfs/xfs_rtbitmap.c
index f4dd697cac08..7c818f1e4484 100644
--- a/fs/xfs/xfs_rtbitmap.c
+++ b/fs/xfs/libxfs/xfs_rtbitmap.c
@@ -424,20 +424,24 @@ xfs_rtfind_forw(
 }
 
 /*
- * Read and modify the summary information for a given extent size,
+ * Read and/or modify the summary information for a given extent size,
  * bitmap block combination.
  * Keeps track of a current summary block, so we don't keep reading
  * it from the buffer cache.
+ *
+ * Summary information is returned in *sum if specified.
+ * If no delta is specified, returns summary only.
  */
 int
-xfs_rtmodify_summary(
-	xfs_mount_t	*mp,		/* file system mount point */
+xfs_rtmodify_summary_int(
+	xfs_mount_t	*mp,		/* file system mount structure */
 	xfs_trans_t	*tp,		/* transaction pointer */
 	int		log,		/* log2 of extent size */
 	xfs_rtblock_t	bbno,		/* bitmap block number */
 	int		delta,		/* change to make to summary info */
 	xfs_buf_t	**rbpp,		/* in/out: summary block buffer */
-	xfs_fsblock_t	*rsb)		/* in/out: summary block number */
+	xfs_fsblock_t	*rsb,		/* in/out: summary block number */
+	xfs_suminfo_t	*sum)		/* out: summary info for this block */
 {
 	xfs_buf_t	*bp;		/* buffer for the summary block */
 	int		error;		/* error value */
@@ -456,7 +460,7 @@ xfs_rtmodify_summary(
 	/*
 	 * If we have an old buffer, and the block number matches, use that.
 	 */
-	if (rbpp && *rbpp && *rsb == sb)
+	if (*rbpp && *rsb == sb)
 		bp = *rbpp;
 	/*
 	 * Otherwise we have to get the buffer.
@@ -465,7 +469,7 @@ xfs_rtmodify_summary(
 		/*
 		 * If there was an old one, get rid of it first.
 		 */
-		if (rbpp && *rbpp)
+		if (*rbpp)
 			xfs_trans_brelse(tp, *rbpp);
 		error = xfs_rtbuf_get(mp, tp, sb, 1, &bp);
 		if (error) {
@@ -474,21 +478,38 @@ xfs_rtmodify_summary(
 		/*
 		 * Remember this buffer and block for the next call.
 		 */
-		if (rbpp) {
-			*rbpp = bp;
-			*rsb = sb;
-		}
+		*rbpp = bp;
+		*rsb = sb;
 	}
 	/*
-	 * Point to the summary information, modify and log it.
+	 * Point to the summary information, modify/log it, and/or copy it out.
 	 */
 	sp = XFS_SUMPTR(mp, bp, so);
-	*sp += delta;
-	xfs_trans_log_buf(tp, bp, (uint)((char *)sp - (char *)bp->b_addr),
-		(uint)((char *)sp - (char *)bp->b_addr + sizeof(*sp) - 1));
+	if (delta) {
+		uint first = (uint)((char *)sp - (char *)bp->b_addr);
+
+		*sp += delta;
+		xfs_trans_log_buf(tp, bp, first, first + sizeof(*sp) - 1);
+	}
+	if (sum)
+		*sum = *sp;
 	return 0;
 }
 
+int
+xfs_rtmodify_summary(
+	xfs_mount_t	*mp,		/* file system mount structure */
+	xfs_trans_t	*tp,		/* transaction pointer */
+	int		log,		/* log2 of extent size */
+	xfs_rtblock_t	bbno,		/* bitmap block number */
+	int		delta,		/* change to make to summary info */
+	xfs_buf_t	**rbpp,		/* in/out: summary block buffer */
+	xfs_fsblock_t	*rsb)		/* in/out: summary block number */
+{
+	return xfs_rtmodify_summary_int(mp, tp, log, bbno,
+					delta, rbpp, rsb, NULL);
+}
+
 /*
  * Set the given range of bitmap bits to the given value.
  * Do whatever I/O and logging is required.
diff --git a/fs/xfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c
index 7703fa6770ff..5f902fa7913f 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;
 	}
 
 	/*
@@ -220,7 +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,
@@ -230,7 +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;
 		}
 	}
 
@@ -238,13 +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(
@@ -252,7 +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(
@@ -260,7 +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;
 	}
 
 	/*
@@ -279,11 +279,13 @@ xfs_mount_validate_sb(
 	    sbp->sb_blocklog < XFS_MIN_BLOCKSIZE_LOG			||
 	    sbp->sb_blocklog > XFS_MAX_BLOCKSIZE_LOG			||
 	    sbp->sb_blocksize != (1 << sbp->sb_blocklog)		||
+	    sbp->sb_dirblklog > XFS_MAX_BLOCKSIZE_LOG			||
 	    sbp->sb_inodesize < XFS_DINODE_MIN_SIZE			||
 	    sbp->sb_inodesize > XFS_DINODE_MAX_SIZE			||
 	    sbp->sb_inodelog < XFS_DINODE_MIN_LOG			||
 	    sbp->sb_inodelog > XFS_DINODE_MAX_LOG			||
 	    sbp->sb_inodesize != (1 << sbp->sb_inodelog)		||
+	    sbp->sb_logsunit > XLOG_MAX_RECORD_BSIZE			||
 	    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)	||
@@ -294,7 +296,7 @@ xfs_mount_validate_sb(
 	    sbp->sb_dblocks < XFS_MIN_DBLOCKS(sbp)			||
 	    sbp->sb_shared_vn != 0)) {
 		xfs_notice(mp, "SB sanity check failed");
-		return XFS_ERROR(EFSCORRUPTED);
+		return -EFSCORRUPTED;
 	}
 
 	/*
@@ -305,7 +307,7 @@ 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;
 	}
 
 	/*
@@ -320,19 +322,19 @@ 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);
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -386,10 +388,11 @@ 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);
@@ -442,9 +445,22 @@ xfs_sb_from_disk(
 	to->sb_features_incompat = be32_to_cpu(from->sb_features_incompat);
 	to->sb_features_log_incompat =
 				be32_to_cpu(from->sb_features_log_incompat);
+	/* crc is only used on disk, not in memory; just init to 0 here. */
+	to->sb_crc = 0;
 	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
@@ -536,6 +552,9 @@ xfs_sb_to_disk(
 	if (!fields)
 		return;
 
+	/* We should never write the crc here, it's updated in the IO path */
+	fields &= ~XFS_SB_CRC;
+
 	xfs_sb_quota_to_disk(to, from, &fields);
 	while (fields) {
 		f = (xfs_sb_field_t)xfs_lowbit64((__uint64_t)fields);
@@ -577,7 +596,11 @@ 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
@@ -620,7 +643,7 @@ xfs_sb_read_verify(
 			/* Only fail bad secondaries on a known V5 filesystem */
 			if (bp->b_bn == XFS_SB_DADDR ||
 			    xfs_sb_version_hascrc(&mp->m_sb)) {
-				error = EFSBADCRC;
+				error = -EFSBADCRC;
 				goto out_error;
 			}
 		}
@@ -630,7 +653,7 @@ xfs_sb_read_verify(
 out_error:
 	if (error) {
 		xfs_buf_ioerror(bp, error);
-		if (error == EFSCORRUPTED || error == EFSBADCRC)
+		if (error == -EFSCORRUPTED || error == -EFSBADCRC)
 			xfs_verifier_error(bp);
 	}
 }
@@ -653,7 +676,7 @@ xfs_sb_quiet_read_verify(
 		return;
 	}
 	/* quietly fail */
-	xfs_buf_ioerror(bp, EWRONGFS);
+	xfs_buf_ioerror(bp, -EWRONGFS);
 }
 
 static void
diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/libxfs/xfs_sb.h
index c43c2d609a24..2e739708afd3 100644
--- a/fs/xfs/xfs_sb.h
+++ b/fs/xfs/libxfs/xfs_sb.h
@@ -87,11 +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 */
diff --git a/fs/xfs/xfs_shared.h b/fs/xfs/libxfs/xfs_shared.h
index 82404da2ca67..82404da2ca67 100644
--- a/fs/xfs/xfs_shared.h
+++ b/fs/xfs/libxfs/xfs_shared.h
diff --git a/fs/xfs/xfs_symlink_remote.c b/fs/xfs/libxfs/xfs_symlink_remote.c
index 23c2f2577c8d..5782f037eab4 100644
--- a/fs/xfs/xfs_symlink_remote.c
+++ b/fs/xfs/libxfs/xfs_symlink_remote.c
@@ -133,9 +133,9 @@ xfs_symlink_read_verify(
 		return;
 
 	if (!xfs_buf_verify_cksum(bp, XFS_SYMLINK_CRC_OFF))
-		xfs_buf_ioerror(bp, EFSBADCRC);
+		xfs_buf_ioerror(bp, -EFSBADCRC);
 	else if (!xfs_symlink_verify(bp))
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 
 	if (bp->b_error)
 		xfs_verifier_error(bp);
@@ -153,7 +153,7 @@ xfs_symlink_write_verify(
 		return;
 
 	if (!xfs_symlink_verify(bp)) {
-		xfs_buf_ioerror(bp, EFSCORRUPTED);
+		xfs_buf_ioerror(bp, -EFSCORRUPTED);
 		xfs_verifier_error(bp);
 		return;
 	}
diff --git a/fs/xfs/xfs_trans_resv.c b/fs/xfs/libxfs/xfs_trans_resv.c
index f2bda7c76b8a..f2bda7c76b8a 100644
--- a/fs/xfs/xfs_trans_resv.c
+++ b/fs/xfs/libxfs/xfs_trans_resv.c
diff --git a/fs/xfs/xfs_trans_resv.h b/fs/xfs/libxfs/xfs_trans_resv.h
index 1097d14cd583..1097d14cd583 100644
--- a/fs/xfs/xfs_trans_resv.h
+++ b/fs/xfs/libxfs/xfs_trans_resv.h
diff --git a/fs/xfs/xfs_trans_space.h b/fs/xfs/libxfs/xfs_trans_space.h
index bf9c4579334d..bf9c4579334d 100644
--- a/fs/xfs/xfs_trans_space.h
+++ b/fs/xfs/libxfs/xfs_trans_space.h
diff --git a/fs/xfs/time.h b/fs/xfs/time.h
deleted file mode 100644
index 387e695a184c..000000000000
--- a/fs/xfs/time.h
+++ /dev/null
@@ -1,36 +0,0 @@
-/*
- * Copyright (c) 2000-2003,2005 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_SUPPORT_TIME_H__
-#define __XFS_SUPPORT_TIME_H__
-
-#include <linux/sched.h>
-#include <linux/time.h>
-
-typedef struct timespec timespec_t;
-
-static inline void delay(long ticks)
-{
-	schedule_timeout_uninterruptible(ticks);
-}
-
-static inline void nanotime(struct timespec *tvp)
-{
-	*tvp = CURRENT_TIME;
-}
-
-#endif /* __XFS_SUPPORT_TIME_H__ */
diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c
index 6888ad886ff6..a65fa5dde6e9 100644
--- a/fs/xfs/xfs_acl.c
+++ b/fs/xfs/xfs_acl.c
@@ -152,7 +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) {
 		/*
@@ -210,7 +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);
@@ -218,7 +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.
@@ -244,7 +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;
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index faaf716e2080..f5b2453a43b2 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
@@ -434,10 +434,22 @@ xfs_start_page_writeback(
 {
 	ASSERT(PageLocked(page));
 	ASSERT(!PageWriteback(page));
-	if (clear_dirty)
+
+	/*
+	 * if the page was not fully cleaned, we need to ensure that the higher
+	 * layers come back to it correctly. That means we need to keep the page
+	 * dirty, and for WB_SYNC_ALL writeback we need to ensure the
+	 * PAGECACHE_TAG_TOWRITE index mark is not removed so another attempt to
+	 * write this page in this writeback sweep will be made.
+	 */
+	if (clear_dirty) {
 		clear_page_dirty_for_io(page);
-	set_page_writeback(page);
+		set_page_writeback(page);
+	} else
+		set_page_writeback_keepwrite(page);
+
 	unlock_page(page);
+
 	/* If no buffers on the page are to be written, finish it here */
 	if (!buffers)
 		end_page_writeback(page);
@@ -502,7 +514,7 @@ xfs_submit_ioend(
 		 * time.
 		 */
 		if (fail) {
-			ioend->io_error = -fail;
+			ioend->io_error = fail;
 			xfs_finish_ioend(ioend);
 			continue;
 		}
@@ -548,6 +560,13 @@ xfs_cancel_ioend(
 		do {
 			next_bh = bh->b_private;
 			clear_buffer_async_write(bh);
+			/*
+			 * The unwritten flag is cleared when added to the
+			 * ioend. We're not submitting for I/O so mark the
+			 * buffer unwritten again for next time around.
+			 */
+			if (ioend->io_type == XFS_IO_UNWRITTEN)
+				set_buffer_unwritten(bh);
 			unlock_buffer(bh);
 		} while ((bh = next_bh) != NULL);
 
@@ -1253,7 +1272,7 @@ __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));
@@ -1302,7 +1321,7 @@ __xfs_get_blocks(
 			error = xfs_iomap_write_direct(ip, offset, size,
 						       &imap, nimaps);
 			if (error)
-				return -error;
+				return error;
 			new = 1;
 		} else {
 			/*
@@ -1415,7 +1434,7 @@ __xfs_get_blocks(
 
 out_unlock:
 	xfs_iunlock(ip, lockmode);
-	return -error;
+	return error;
 }
 
 int
@@ -1753,11 +1772,72 @@ 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 90e2eeb21207..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);
@@ -496,11 +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
@@ -514,7 +514,7 @@ xfs_attr_list_int(
 	XFS_STATS_INC(xs_attr_list);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return EIO;
+		return -EIO;
 
 	/*
 	 * Decide on what work routines to call based on the inode size.
@@ -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 64731ef3324d..281002689d64 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;
@@ -365,7 +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 {
@@ -425,14 +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;
@@ -524,13 +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;
@@ -539,7 +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)){
@@ -559,26 +559,26 @@ 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) {
 		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);
+			error = filemap_write_and_wait(VFS_I(ip)->i_mapping);
 			if (error)
 				goto out_unlock_iolock;
 
@@ -611,7 +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)
@@ -809,7 +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;
 
@@ -882,7 +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;
 			}
 		}
 
@@ -955,14 +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);
@@ -1028,7 +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;
 		}
@@ -1065,7 +1065,7 @@ xfs_alloc_file_space(
 		allocated_fsb = imapp->br_blockcount;
 
 		if (nimaps == 0) {
-			error = XFS_ERROR(ENOSPC);
+			error = -ENOSPC;
 			break;
 		}
 
@@ -1122,14 +1122,6 @@ xfs_zero_remaining_bytes(
 	if (endoff > XFS_ISIZE(ip))
 		endoff = XFS_ISIZE(ip);
 
-	bp = xfs_buf_get_uncached(XFS_IS_REALTIME_INODE(ip) ?
-					mp->m_rtdev_targp : mp->m_ddev_targp,
-				  BTOBB(mp->m_sb.sb_blocksize), 0);
-	if (!bp)
-		return XFS_ERROR(ENOMEM);
-
-	xfs_buf_unlock(bp);
-
 	for (offset = startoff; offset <= endoff; offset = lastoffset + 1) {
 		uint lock_mode;
 
@@ -1152,42 +1144,24 @@ xfs_zero_remaining_bytes(
 		ASSERT(imap.br_startblock != DELAYSTARTBLOCK);
 		if (imap.br_state == XFS_EXT_UNWRITTEN)
 			continue;
-		XFS_BUF_UNDONE(bp);
-		XFS_BUF_UNWRITE(bp);
-		XFS_BUF_READ(bp);
-		XFS_BUF_SET_ADDR(bp, xfs_fsb_to_db(ip, imap.br_startblock));
 
-		if (XFS_FORCED_SHUTDOWN(mp)) {
-			error = XFS_ERROR(EIO);
-			break;
-		}
-		xfs_buf_iorequest(bp);
-		error = xfs_buf_iowait(bp);
-		if (error) {
-			xfs_buf_ioerror_alert(bp,
-					"xfs_zero_remaining_bytes(read)");
-			break;
-		}
+		error = xfs_buf_read_uncached(XFS_IS_REALTIME_INODE(ip) ?
+				mp->m_rtdev_targp : mp->m_ddev_targp,
+				xfs_fsb_to_db(ip, imap.br_startblock),
+				BTOBB(mp->m_sb.sb_blocksize),
+				0, &bp, NULL);
+		if (error)
+			return error;
+
 		memset(bp->b_addr +
-			(offset - XFS_FSB_TO_B(mp, imap.br_startoff)),
-		      0, lastoffset - offset + 1);
-		XFS_BUF_UNDONE(bp);
-		XFS_BUF_UNREAD(bp);
-		XFS_BUF_WRITE(bp);
-
-		if (XFS_FORCED_SHUTDOWN(mp)) {
-			error = XFS_ERROR(EIO);
-			break;
-		}
-		xfs_buf_iorequest(bp);
-		error = xfs_buf_iowait(bp);
-		if (error) {
-			xfs_buf_ioerror_alert(bp,
-					"xfs_zero_remaining_bytes(write)");
-			break;
-		}
+				(offset - XFS_FSB_TO_B(mp, imap.br_startoff)),
+		       0, lastoffset - offset + 1);
+
+		error = xfs_bwrite(bp);
+		xfs_buf_relse(bp);
+		if (error)
+			return error;
 	}
-	xfs_buf_free(bp);
 	return error;
 }
 
@@ -1205,6 +1179,7 @@ xfs_free_file_space(
 	xfs_bmap_free_t		free_list;
 	xfs_bmbt_irec_t		imap;
 	xfs_off_t		ioffset;
+	xfs_off_t		iendoffset;
 	xfs_extlen_t		mod=0;
 	xfs_mount_t		*mp;
 	int			nimap;
@@ -1233,12 +1208,13 @@ xfs_free_file_space(
 	inode_dio_wait(VFS_I(ip));
 
 	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,
-					      ioffset, -1);
+	ioffset = round_down(offset, rounding);
+	iendoffset = round_up(offset + len, rounding) - 1;
+	error = filemap_write_and_wait_range(VFS_I(ip)->i_mapping, ioffset,
+					     iendoffset);
 	if (error)
 		goto out;
-	truncate_pagecache_range(VFS_I(ip), ioffset, -1);
+	truncate_pagecache_range(VFS_I(ip), ioffset, iendoffset);
 
 	/*
 	 * Need to zero the stuff we're not freeing, on disk.
@@ -1315,7 +1291,7 @@ 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;
 		}
@@ -1362,7 +1338,10 @@ xfs_free_file_space(
 	goto out;
 }
 
-
+/*
+ * Preallocate and zero a range of a file. This mechanism has the allocation
+ * semantics of fallocate and in addition converts data in the range to zeroes.
+ */
 int
 xfs_zero_file_space(
 	struct xfs_inode	*ip,
@@ -1370,65 +1349,30 @@ xfs_zero_file_space(
 	xfs_off_t		len)
 {
 	struct xfs_mount	*mp = ip->i_mount;
-	uint			granularity;
-	xfs_off_t		start_boundary;
-	xfs_off_t		end_boundary;
+	uint			blksize;
 	int			error;
 
 	trace_xfs_zero_file_space(ip);
 
-	granularity = max_t(uint, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE);
+	blksize = 1 << mp->m_sb.sb_blocklog;
 
 	/*
-	 * Round the range of extents we are going to convert inwards.  If the
-	 * offset is aligned, then it doesn't get changed so we zero from the
-	 * start of the block offset points to.
+	 * Punch a hole and prealloc the range. We use hole punch rather than
+	 * unwritten extent conversion for two reasons:
+	 *
+	 * 1.) Hole punch handles partial block zeroing for us.
+	 *
+	 * 2.) If prealloc returns ENOSPC, the file range is still zero-valued
+	 * by virtue of the hole punch.
 	 */
-	start_boundary = round_up(offset, granularity);
-	end_boundary = round_down(offset + len, granularity);
-
-	ASSERT(start_boundary >= offset);
-	ASSERT(end_boundary <= offset + len);
-
-	if (start_boundary < end_boundary - 1) {
-		/*
-		 * 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,
-					XFS_BMAPI_PREALLOC | XFS_BMAPI_CONVERT);
-		if (error)
-			goto out;
-
-		/* We've handled the interior of the range, now for the edges */
-		if (start_boundary != offset) {
-			error = xfs_iozero(ip, offset, start_boundary - offset);
-			if (error)
-				goto out;
-		}
-
-		if (end_boundary != offset + len)
-			error = xfs_iozero(ip, end_boundary,
-					   offset + len - end_boundary);
-
-	} else {
-		/*
-		 * It's either a sub-granularity range or the range spanned lies
-		 * partially across two adjacent blocks.
-		 */
-		error = xfs_iozero(ip, offset, len);
-	}
+	error = xfs_free_file_space(ip, offset, len);
+	if (error)
+		goto out;
 
+	error = xfs_alloc_file_space(ip, round_down(offset, blksize),
+				     round_up(offset + len, blksize) -
+				     round_down(offset, blksize),
+				     XFS_BMAPI_PREALLOC);
 out:
 	return error;
 
@@ -1456,24 +1400,50 @@ xfs_collapse_file_space(
 	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		next_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);
+	next_fsb = XFS_B_TO_FSB(mp, offset + len);
 	shift_fsb = XFS_B_TO_FSB(mp, len);
 
 	error = xfs_free_file_space(ip, offset, len);
 	if (error)
 		return error;
 
+	/*
+	 * Trim eofblocks to avoid shifting uninitialized post-eof preallocation
+	 * into the accessible region of the file.
+	 */
+	if (xfs_can_free_eofblocks(ip, true)) {
+		error = xfs_free_eofblocks(mp, ip, false);
+		if (error)
+			return error;
+	}
+
+	/*
+	 * Writeback and invalidate cache for the remainder of the file as we're
+	 * about to shift down every extent from the collapse range to EOF. The
+	 * free of the collapse range above might have already done some of
+	 * this, but we shouldn't rely on it to do anything outside of the range
+	 * that was freed.
+	 */
+	error = filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
+					     offset + len, -1);
+	if (error)
+		return error;
+	error = invalidate_inode_pages2_range(VFS_I(ip)->i_mapping,
+					(offset + len) >> PAGE_CACHE_SHIFT, -1);
+	if (error)
+		return error;
+
 	while (!error && !done) {
 		tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT);
 		/*
@@ -1505,10 +1475,10 @@ xfs_collapse_file_space(
 		 * 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);
+		start_fsb = next_fsb;
+		error = xfs_bmap_shift_extents(tp, ip, start_fsb, shift_fsb,
+				&done, &next_fsb, &first_block, &free_list,
+				XFS_BMAP_MAX_SHIFT_EXTENTS);
 		if (error)
 			goto out;
 
@@ -1557,14 +1527,14 @@ 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
@@ -1573,19 +1543,19 @@ 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
@@ -1599,25 +1569,49 @@ 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;
 }
 
+static 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 */
@@ -1633,51 +1627,57 @@ 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);
@@ -1689,7 +1689,7 @@ 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;
 	}
 
 	/*
@@ -1703,43 +1703,9 @@ 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
 	 */
@@ -1757,8 +1723,8 @@ 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
@@ -1887,8 +1853,8 @@ 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_buf.c b/fs/xfs/xfs_buf.c
index 7a34a1ae6552..24b4ebea0d4d 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;
 }
 
@@ -344,7 +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;
 			}
 
@@ -465,7 +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.
 		 */
@@ -623,10 +623,11 @@ _xfs_buf_read(
 	bp->b_flags &= ~(XBF_WRITE | XBF_ASYNC | XBF_READ_AHEAD);
 	bp->b_flags |= flags & (XBF_READ | XBF_ASYNC | XBF_READ_AHEAD);
 
-	xfs_buf_iorequest(bp);
-	if (flags & XBF_ASYNC)
+	if (flags & XBF_ASYNC) {
+		xfs_buf_submit(bp);
 		return 0;
-	return xfs_buf_iowait(bp);
+	}
+	return xfs_buf_submit_wait(bp);
 }
 
 xfs_buf_t *
@@ -687,34 +688,39 @@ xfs_buf_readahead_map(
  * Read an uncached buffer from disk. Allocates and returns a locked
  * buffer containing the disk contents or nothing.
  */
-struct xfs_buf *
+int
 xfs_buf_read_uncached(
 	struct xfs_buftarg	*target,
 	xfs_daddr_t		daddr,
 	size_t			numblks,
 	int			flags,
+	struct xfs_buf		**bpp,
 	const struct xfs_buf_ops *ops)
 {
 	struct xfs_buf		*bp;
 
+	*bpp = NULL;
+
 	bp = xfs_buf_get_uncached(target, numblks, flags);
 	if (!bp)
-		return NULL;
+		return -ENOMEM;
 
 	/* set up the buffer for a read IO */
 	ASSERT(bp->b_map_count == 1);
-	bp->b_bn = daddr;
+	bp->b_bn = XFS_BUF_DADDR_NULL;  /* always null for uncached buffers */
 	bp->b_maps[0].bm_bn = daddr;
 	bp->b_flags |= XBF_READ;
 	bp->b_ops = ops;
 
-	if (XFS_FORCED_SHUTDOWN(target->bt_mount)) {
+	xfs_buf_submit_wait(bp);
+	if (bp->b_error) {
+		int	error = bp->b_error;
 		xfs_buf_relse(bp);
-		return NULL;
+		return error;
 	}
-	xfs_buf_iorequest(bp);
-	xfs_buf_iowait(bp);
-	return bp;
+
+	*bpp = bp;
+	return 0;
 }
 
 /*
@@ -998,53 +1004,56 @@ xfs_buf_wait_unpin(
  *	Buffer Utility Routines
  */
 
-STATIC void
-xfs_buf_iodone_work(
-	struct work_struct	*work)
+void
+xfs_buf_ioend(
+	struct xfs_buf	*bp)
 {
-	struct xfs_buf		*bp =
-		container_of(work, xfs_buf_t, b_iodone_work);
-	bool			read = !!(bp->b_flags & XBF_READ);
+	bool		read = bp->b_flags & XBF_READ;
+
+	trace_xfs_buf_iodone(bp, _RET_IP_);
 
 	bp->b_flags &= ~(XBF_READ | XBF_WRITE | XBF_READ_AHEAD);
 
-	/* only validate buffers that were read without errors */
-	if (read && bp->b_ops && !bp->b_error && (bp->b_flags & XBF_DONE))
+	/*
+	 * Pull in IO completion errors now. We are guaranteed to be running
+	 * single threaded, so we don't need the lock to read b_io_error.
+	 */
+	if (!bp->b_error && bp->b_io_error)
+		xfs_buf_ioerror(bp, bp->b_io_error);
+
+	/* Only validate buffers that were read without errors */
+	if (read && !bp->b_error && bp->b_ops) {
+		ASSERT(!bp->b_iodone);
 		bp->b_ops->verify_read(bp);
+	}
+
+	if (!bp->b_error)
+		bp->b_flags |= XBF_DONE;
 
 	if (bp->b_iodone)
 		(*(bp->b_iodone))(bp);
 	else if (bp->b_flags & XBF_ASYNC)
 		xfs_buf_relse(bp);
-	else {
-		ASSERT(read && bp->b_ops);
+	else
 		complete(&bp->b_iowait);
-	}
 }
 
-void
-xfs_buf_ioend(
-	struct xfs_buf	*bp,
-	int		schedule)
+static void
+xfs_buf_ioend_work(
+	struct work_struct	*work)
 {
-	bool		read = !!(bp->b_flags & XBF_READ);
-
-	trace_xfs_buf_iodone(bp, _RET_IP_);
+	struct xfs_buf		*bp =
+		container_of(work, xfs_buf_t, b_iodone_work);
 
-	if (bp->b_error == 0)
-		bp->b_flags |= XBF_DONE;
+	xfs_buf_ioend(bp);
+}
 
-	if (bp->b_iodone || (read && bp->b_ops) || (bp->b_flags & XBF_ASYNC)) {
-		if (schedule) {
-			INIT_WORK(&bp->b_iodone_work, xfs_buf_iodone_work);
-			queue_work(xfslogd_workqueue, &bp->b_iodone_work);
-		} else {
-			xfs_buf_iodone_work(&bp->b_iodone_work);
-		}
-	} else {
-		bp->b_flags &= ~(XBF_READ | XBF_WRITE | XBF_READ_AHEAD);
-		complete(&bp->b_iowait);
-	}
+void
+xfs_buf_ioend_async(
+	struct xfs_buf	*bp)
+{
+	INIT_WORK(&bp->b_iodone_work, xfs_buf_ioend_work);
+	queue_work(xfslogd_workqueue, &bp->b_iodone_work);
 }
 
 void
@@ -1052,8 +1061,8 @@ 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_);
 }
 
@@ -1064,97 +1073,7 @@ 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);
-}
-
-/*
- * Called when we want to stop a buffer from getting written or read.
- * We attach the EIO error, muck with its flags, and call xfs_buf_ioend
- * so that the proper iodone callbacks get called.
- */
-STATIC int
-xfs_bioerror(
-	xfs_buf_t *bp)
-{
-#ifdef XFSERRORDEBUG
-	ASSERT(XFS_BUF_ISREAD(bp) || bp->b_iodone);
-#endif
-
-	/*
-	 * No need to wait until the buffer is unpinned, we aren't flushing it.
-	 */
-	xfs_buf_ioerror(bp, EIO);
-
-	/*
-	 * We're calling xfs_buf_ioend, so delete XBF_DONE flag.
-	 */
-	XFS_BUF_UNREAD(bp);
-	XFS_BUF_UNDONE(bp);
-	xfs_buf_stale(bp);
-
-	xfs_buf_ioend(bp, 0);
-
-	return EIO;
-}
-
-/*
- * Same as xfs_bioerror, except that we are releasing the buffer
- * here ourselves, and avoiding the xfs_buf_ioend call.
- * This is meant for userdata errors; metadata bufs come with
- * iodone functions attached, so that we can track down errors.
- */
-int
-xfs_bioerror_relse(
-	struct xfs_buf	*bp)
-{
-	int64_t		fl = bp->b_flags;
-	/*
-	 * No need to wait until the buffer is unpinned.
-	 * We aren't flushing it.
-	 *
-	 * chunkhold expects B_DONE to be set, whether
-	 * we actually finish the I/O or not. We don't want to
-	 * change that interface.
-	 */
-	XFS_BUF_UNREAD(bp);
-	XFS_BUF_DONE(bp);
-	xfs_buf_stale(bp);
-	bp->b_iodone = NULL;
-	if (!(fl & XBF_ASYNC)) {
-		/*
-		 * Mark b_error and B_ERROR _both_.
-		 * Lot's of chunkcache code assumes that.
-		 * There's no reason to mark error for
-		 * ASYNC buffers.
-		 */
-		xfs_buf_ioerror(bp, EIO);
-		complete(&bp->b_iowait);
-	} else {
-		xfs_buf_relse(bp);
-	}
-
-	return EIO;
-}
-
-STATIC int
-xfs_bdstrat_cb(
-	struct xfs_buf	*bp)
-{
-	if (XFS_FORCED_SHUTDOWN(bp->b_target->bt_mount)) {
-		trace_xfs_bdstrat_shut(bp, _RET_IP_);
-		/*
-		 * Metadata write that didn't get logged but
-		 * written delayed anyway. These aren't associated
-		 * with a transaction, and can be ignored.
-		 */
-		if (!bp->b_iodone && !XFS_BUF_ISREAD(bp))
-			return xfs_bioerror_relse(bp);
-		else
-			return xfs_bioerror(bp);
-	}
-
-	xfs_buf_iorequest(bp);
-	return 0;
+		(__uint64_t)XFS_BUF_ADDR(bp), func, -bp->b_error, bp->b_length);
 }
 
 int
@@ -1166,11 +1085,10 @@ xfs_bwrite(
 	ASSERT(xfs_buf_islocked(bp));
 
 	bp->b_flags |= XBF_WRITE;
-	bp->b_flags &= ~(XBF_ASYNC | XBF_READ | _XBF_DELWRI_Q | XBF_WRITE_FAIL);
-
-	xfs_bdstrat_cb(bp);
+	bp->b_flags &= ~(XBF_ASYNC | XBF_READ | _XBF_DELWRI_Q |
+			 XBF_WRITE_FAIL | XBF_DONE);
 
-	error = xfs_buf_iowait(bp);
+	error = xfs_buf_submit_wait(bp);
 	if (error) {
 		xfs_force_shutdown(bp->b_target->bt_mount,
 				   SHUTDOWN_META_IO_ERROR);
@@ -1179,15 +1097,6 @@ xfs_bwrite(
 }
 
 STATIC void
-_xfs_buf_ioend(
-	xfs_buf_t		*bp,
-	int			schedule)
-{
-	if (atomic_dec_and_test(&bp->b_io_remaining) == 1)
-		xfs_buf_ioend(bp, schedule);
-}
-
-STATIC void
 xfs_buf_bio_end_io(
 	struct bio		*bio,
 	int			error)
@@ -1198,13 +1107,18 @@ xfs_buf_bio_end_io(
 	 * don't overwrite existing errors - otherwise we can lose errors on
 	 * buffers that require multiple bios to complete.
 	 */
-	if (!bp->b_error)
-		xfs_buf_ioerror(bp, -error);
+	if (error) {
+		spin_lock(&bp->b_lock);
+		if (!bp->b_io_error)
+			bp->b_io_error = error;
+		spin_unlock(&bp->b_lock);
+	}
 
 	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));
 
-	_xfs_buf_ioend(bp, 1);
+	if (atomic_dec_and_test(&bp->b_io_remaining) == 1)
+		xfs_buf_ioend_async(bp);
 	bio_put(bio);
 }
 
@@ -1283,10 +1197,10 @@ next_chunk:
 	} else {
 		/*
 		 * This is guaranteed not to be the last io reference count
-		 * because the caller (xfs_buf_iorequest) holds a count itself.
+		 * because the caller (xfs_buf_submit) holds a count itself.
 		 */
 		atomic_dec(&bp->b_io_remaining);
-		xfs_buf_ioerror(bp, EIO);
+		xfs_buf_ioerror(bp, -EIO);
 		bio_put(bio);
 	}
 
@@ -1330,6 +1244,20 @@ _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;
@@ -1359,53 +1287,131 @@ _xfs_buf_ioapply(
 	blk_finish_plug(&plug);
 }
 
+/*
+ * Asynchronous IO submission path. This transfers the buffer lock ownership and
+ * the current reference to the IO. It is not safe to reference the buffer after
+ * a call to this function unless the caller holds an additional reference
+ * itself.
+ */
 void
-xfs_buf_iorequest(
-	xfs_buf_t		*bp)
+xfs_buf_submit(
+	struct xfs_buf	*bp)
 {
-	trace_xfs_buf_iorequest(bp, _RET_IP_);
+	trace_xfs_buf_submit(bp, _RET_IP_);
 
 	ASSERT(!(bp->b_flags & _XBF_DELWRI_Q));
+	ASSERT(bp->b_flags & XBF_ASYNC);
+
+	/* on shutdown we stale and complete the buffer immediately */
+	if (XFS_FORCED_SHUTDOWN(bp->b_target->bt_mount)) {
+		xfs_buf_ioerror(bp, -EIO);
+		bp->b_flags &= ~XBF_DONE;
+		xfs_buf_stale(bp);
+		xfs_buf_ioend(bp);
+		return;
+	}
 
 	if (bp->b_flags & XBF_WRITE)
 		xfs_buf_wait_unpin(bp);
+
+	/* clear the internal error state to avoid spurious errors */
+	bp->b_io_error = 0;
+
+	/*
+	 * The caller's reference is released during I/O completion.
+	 * This occurs some time after the last b_io_remaining reference is
+	 * released, so after we drop our Io reference we have to have some
+	 * other reference to ensure the buffer doesn't go away from underneath
+	 * us. Take a direct reference to ensure we have safe access to the
+	 * buffer until we are finished with it.
+	 */
 	xfs_buf_hold(bp);
 
 	/*
-	 * 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.
+	 * 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);
+
 	/*
-	 * 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.
+	 * If _xfs_buf_ioapply failed, we can get back here with only the IO
+	 * reference we took above. If we drop it to zero, run completion so
+	 * that we don't return to the caller with completion still pending.
 	 */
-	_xfs_buf_ioend(bp, bp->b_error ? 0 : 1);
+	if (atomic_dec_and_test(&bp->b_io_remaining) == 1) {
+		if (bp->b_error)
+			xfs_buf_ioend(bp);
+		else
+			xfs_buf_ioend_async(bp);
+	}
 
 	xfs_buf_rele(bp);
+	/* Note: it is not safe to reference bp now we've dropped our ref */
 }
 
 /*
- * 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, in which
- * case nothing will ever complete.  It returns the I/O error code, if any, or
- * 0 if there was no error.
+ * Synchronous buffer IO submission path, read or write.
  */
 int
-xfs_buf_iowait(
-	xfs_buf_t		*bp)
+xfs_buf_submit_wait(
+	struct xfs_buf	*bp)
 {
-	trace_xfs_buf_iowait(bp, _RET_IP_);
+	int		error;
 
-	if (!bp->b_error)
-		wait_for_completion(&bp->b_iowait);
+	trace_xfs_buf_submit_wait(bp, _RET_IP_);
+
+	ASSERT(!(bp->b_flags & (_XBF_DELWRI_Q | XBF_ASYNC)));
+
+	if (XFS_FORCED_SHUTDOWN(bp->b_target->bt_mount)) {
+		xfs_buf_ioerror(bp, -EIO);
+		xfs_buf_stale(bp);
+		bp->b_flags &= ~XBF_DONE;
+		return -EIO;
+	}
+
+	if (bp->b_flags & XBF_WRITE)
+		xfs_buf_wait_unpin(bp);
+
+	/* clear the internal error state to avoid spurious errors */
+	bp->b_io_error = 0;
+
+	/*
+	 * For synchronous IO, the IO does not inherit the submitters reference
+	 * count, nor the buffer lock. Hence we cannot release the reference we
+	 * are about to take until we've waited for all IO completion to occur,
+	 * including any xfs_buf_ioend_async() work that may be pending.
+	 */
+	xfs_buf_hold(bp);
+
+	/*
+	 * 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);
+
+	/*
+	 * make sure we run completion synchronously if it raced with us and is
+	 * already complete.
+	 */
+	if (atomic_dec_and_test(&bp->b_io_remaining) == 1)
+		xfs_buf_ioend(bp);
 
+	/* wait for completion before gathering the error from the buffer */
+	trace_xfs_buf_iowait(bp, _RET_IP_);
+	wait_for_completion(&bp->b_iowait);
 	trace_xfs_buf_iowait_done(bp, _RET_IP_);
-	return bp->b_error;
+	error = bp->b_error;
+
+	/*
+	 * all done now, we can release the hold that keeps the buffer
+	 * referenced for the entire IO.
+	 */
+	xfs_buf_rele(bp);
+	return error;
 }
 
 xfs_caddr_t
@@ -1628,7 +1634,7 @@ xfs_setsize_buftarg(
 		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 */
@@ -1664,8 +1670,6 @@ xfs_alloc_buftarg(
 	btp->bt_dev =  bdev->bd_dev;
 	btp->bt_bdev = bdev;
 	btp->bt_bdi = blk_get_backing_dev_info(bdev);
-	if (!btp->bt_bdi)
-		goto error;
 
 	if (xfs_setsize_buftarg_early(btp, bdev))
 		goto error;
@@ -1799,13 +1803,19 @@ __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 | XBF_WRITE_FAIL);
-		bp->b_flags |= XBF_WRITE;
+		bp->b_flags |= XBF_WRITE | XBF_ASYNC;
 
-		if (!wait) {
-			bp->b_flags |= XBF_ASYNC;
+		/*
+		 * we do all Io submission async. This means if we need to wait
+		 * for IO completion we need to take an extra reference so the
+		 * buffer is still valid on the other side.
+		 */
+		if (wait)
+			xfs_buf_hold(bp);
+		else
 			list_del_init(&bp->b_list);
-		}
-		xfs_bdstrat_cb(bp);
+
+		xfs_buf_submit(bp);
 	}
 	blk_finish_plug(&plug);
 
@@ -1852,7 +1862,10 @@ xfs_buf_delwri_submit(
 		bp = list_first_entry(&io_list, struct xfs_buf, b_list);
 
 		list_del_init(&bp->b_list);
-		error2 = xfs_buf_iowait(bp);
+
+		/* locking the buffer will wait for async IO completion. */
+		xfs_buf_lock(bp);
+		error2 = bp->b_error;
 		xfs_buf_relse(bp);
 		if (!error)
 			error = error2;
@@ -1870,7 +1883,7 @@ xfs_buf_init(void)
 		goto out;
 
 	xfslogd_workqueue = alloc_workqueue("xfslogd",
-					WQ_MEM_RECLAIM | WQ_HIGHPRI, 1);
+				WQ_MEM_RECLAIM | WQ_HIGHPRI | WQ_FREEZABLE, 1);
 	if (!xfslogd_workqueue)
 		goto out_free_buf_zone;
 
diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h
index 3a7a5523d3dc..82002c00af90 100644
--- a/fs/xfs/xfs_buf.h
+++ b/fs/xfs/xfs_buf.h
@@ -158,6 +158,7 @@ typedef struct xfs_buf {
 	struct list_head	b_lru;		/* lru list */
 	spinlock_t		b_lock;		/* internal state lock */
 	unsigned int		b_state;	/* internal state flags */
+	int			b_io_error;	/* internal IO error state */
 	wait_queue_head_t	b_waiters;	/* unpin waiters */
 	struct list_head	b_list;
 	struct xfs_perag	*b_pag;		/* contains rbtree root */
@@ -178,7 +179,7 @@ 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
@@ -268,9 +269,9 @@ int xfs_buf_associate_memory(struct xfs_buf *bp, void *mem, size_t length);
 
 struct xfs_buf *xfs_buf_get_uncached(struct xfs_buftarg *target, size_t numblks,
 				int flags);
-struct xfs_buf *xfs_buf_read_uncached(struct xfs_buftarg *target,
-				xfs_daddr_t daddr, size_t numblks, int flags,
-				const struct xfs_buf_ops *ops);
+int xfs_buf_read_uncached(struct xfs_buftarg *target, xfs_daddr_t daddr,
+			  size_t numblks, int flags, struct xfs_buf **bpp,
+			  const struct xfs_buf_ops *ops);
 void xfs_buf_hold(struct xfs_buf *bp);
 
 /* Releasing Buffers */
@@ -286,18 +287,16 @@ extern void xfs_buf_unlock(xfs_buf_t *);
 
 /* Buffer Read and Write Routines */
 extern int xfs_bwrite(struct xfs_buf *bp);
-extern void xfs_buf_ioend(xfs_buf_t *,	int);
+extern void xfs_buf_ioend(struct xfs_buf *bp);
 extern void xfs_buf_ioerror(xfs_buf_t *, int);
 extern void xfs_buf_ioerror_alert(struct xfs_buf *, const char *func);
-extern void xfs_buf_iorequest(xfs_buf_t *);
-extern int xfs_buf_iowait(xfs_buf_t *);
+extern void xfs_buf_submit(struct xfs_buf *bp);
+extern int xfs_buf_submit_wait(struct xfs_buf *bp);
 extern void xfs_buf_iomove(xfs_buf_t *, size_t, size_t, void *,
 				xfs_buf_rw_t);
 #define xfs_buf_zero(bp, off, len) \
 	    xfs_buf_iomove((bp), (off), (len), NULL, XBRW_ZERO)
 
-extern int xfs_bioerror_relse(struct xfs_buf *);
-
 /* Buffer Utility Routines */
 extern xfs_caddr_t xfs_buf_offset(xfs_buf_t *, size_t);
 
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index 4654338b03fc..f15969543326 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -488,10 +488,10 @@ 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);
+		xfs_buf_ioend(bp);
 	}
 }
 
@@ -501,7 +501,7 @@ xfs_buf_item_unpin(
  * buffer being bad..
  */
 
-DEFINE_RATELIMIT_STATE(xfs_buf_write_fail_rl_state, 30 * HZ, 10);
+static DEFINE_RATELIMIT_STATE(xfs_buf_write_fail_rl_state, 30 * HZ, 10);
 
 STATIC uint
 xfs_buf_item_push(
@@ -725,7 +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;
 }
 
@@ -1081,7 +1081,7 @@ xfs_buf_iodone_callbacks(
 	 * a way to shut the filesystem down if the writes keep failing.
 	 *
 	 * In practice we'll shut the filesystem down soon as non-transient
-	 * erorrs tend to affect the whole device and a failing log write
+	 * errors tend to affect the whole device and a failing log write
 	 * will make us give up.  But we really ought to do better here.
 	 */
 	if (XFS_BUF_ISASYNC(bp)) {
@@ -1094,7 +1094,7 @@ xfs_buf_iodone_callbacks(
 		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);
+			xfs_buf_submit(bp);
 		} else {
 			xfs_buf_relse(bp);
 		}
@@ -1115,7 +1115,7 @@ do_callbacks:
 	xfs_buf_do_callbacks(bp);
 	bp->b_fspriv = NULL;
 	bp->b_iodone = NULL;
-	xfs_buf_ioend(bp, 0);
+	xfs_buf_ioend(bp);
 }
 
 /*
diff --git a/fs/xfs/xfs_dir2_readdir.c b/fs/xfs/xfs_dir2_readdir.c
index 48e99afb9cb0..f1b69edcdf31 100644
--- a/fs/xfs/xfs_dir2_readdir.c
+++ b/fs/xfs/xfs_dir2_readdir.c
@@ -95,7 +95,7 @@ xfs_dir2_sf_getdents(
 	 */
 	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);
@@ -677,7 +677,7 @@ xfs_readdir(
 	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);
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 3ee0cd43edc0..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);
@@ -354,7 +354,7 @@ xfs_qm_dqalloc(
 			       mp->m_quotainfo->qi_dqchunklen,
 			       0);
 	if (!bp) {
-		error = ENOMEM;
+		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;
 		}
 	}
 
@@ -480,7 +480,7 @@ xfs_qm_dqtobp(
 		 * didn't have the quota inode lock.
 		 */
 		xfs_iunlock(quotip, lock_mode);
-		return ESRCH;
+		return -ESRCH;
 	}
 
 	/*
@@ -508,7 +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,
@@ -530,7 +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);
@@ -539,7 +539,7 @@ xfs_qm_dqtobp(
 
 		if (error) {
 			ASSERT(bp == NULL);
-			return XFS_ERROR(error);
+			return error;
 		}
 	}
 
@@ -547,7 +547,7 @@ xfs_qm_dqtobp(
 	*O_bpp = bp;
 	*O_ddpp = bp->b_addr + dqp->q_bufoffset;
 
-	return (0);
+	return 0;
 }
 
 
@@ -715,7 +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
@@ -723,7 +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;
 		}
 	}
 
@@ -796,14 +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
@@ -829,7 +829,7 @@ restart:
 	ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	trace_xfs_dqget_miss(dqp);
 	*O_dqpp = dqp;
-	return (0);
+	return 0;
 }
 
 /*
@@ -966,7 +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;
 	}
 
@@ -974,7 +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;
 
@@ -992,7 +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 */
@@ -1045,7 +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 68a68f704837..c24c67e22a2a 100644
--- a/fs/xfs/xfs_dquot.h
+++ b/fs/xfs/xfs_dquot.h
@@ -139,6 +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_error.c b/fs/xfs/xfs_error.c
index edac5b057d28..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];
@@ -190,7 +167,7 @@ xfs_verifier_error(
 	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",
+		  bp->b_error == -EFSBADCRC ? "CRC error" : "corruption",
 		  __return_address, bp->b_bn);
 
 	xfs_alert(mp, "Unmount and run xfs_repair");
diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h
index c1c57d4a4b5d..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,
@@ -56,7 +47,7 @@ extern void xfs_verifier_error(struct xfs_buf *bp);
 		if (unlikely(!fs_is_ok)) { \
 			XFS_ERROR_REPORT("XFS_WANT_CORRUPTED_GOTO", \
 					 XFS_ERRLEVEL_LOW, NULL); \
-			error = XFS_ERROR(EFSCORRUPTED); \
+			error = -EFSCORRUPTED; \
 			goto l; \
 		} \
 	}
@@ -68,7 +59,7 @@ extern void xfs_verifier_error(struct xfs_buf *bp);
 		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 753e467aa1a5..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));
 }
@@ -237,7 +237,7 @@ xfs_fs_nfs_commit_metadata(
 
 	if (!lsn)
 		return 0;
-	return -_xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, NULL);
+	return _xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, NULL);
 }
 
 const struct export_operations xfs_export_operations = {
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index fb7a4c1ce1c5..c4327419dc5c 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -298,7 +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;
 }
 
 /*
diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index 1f66779d7a46..eb596b419942 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>
@@ -155,7 +156,7 @@ xfs_dir_fsync(
 
 	if (!lsn)
 		return 0;
-	return -_xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, NULL);
+	return _xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, NULL);
 }
 
 STATIC int
@@ -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);
 
@@ -225,7 +226,7 @@ xfs_file_fsync(
 	    !log_flushed)
 		xfs_blkdev_issue_flush(mp->m_ddev_targp);
 
-	return -error;
+	return error;
 }
 
 STATIC ssize_t
@@ -246,11 +247,11 @@ xfs_file_read_iter(
 	XFS_STATS_INC(xs_read_calls);
 
 	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;
 
-	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;
@@ -258,7 +259,7 @@ xfs_file_read_iter(
 		if ((pos | size) & target->bt_logical_sectormask) {
 			if (pos == i_size_read(inode))
 				return 0;
-			return -XFS_ERROR(EINVAL);
+			return -EINVAL;
 		}
 	}
 
@@ -283,19 +284,29 @@ xfs_file_read_iter(
 	 * 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(
 							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);
 	}
@@ -325,7 +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;
@@ -524,7 +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;
 	}
@@ -594,7 +605,7 @@ xfs_file_dio_aio_write(
 
 	/* DIO must be aligned to device logical sector size */
 	if ((pos | count) & target->bt_logical_sectormask)
-		return -XFS_ERROR(EINVAL);
+		return -EINVAL;
 
 	/* "unaligned" here means not aligned to a filesystem block */
 	if ((pos & mp->m_blockmask) || ((pos + count) & mp->m_blockmask))
@@ -631,10 +642,19 @@ xfs_file_dio_aio_write(
 
 	if (mapping->nrpages) {
 		ret = filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
-						    pos, -1);
+						    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;
 	}
 
 	/*
@@ -689,14 +709,28 @@ write_retry:
 	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;
 	}
 
@@ -772,7 +806,7 @@ xfs_file_fallocate(
 		unsigned blksize_mask = (1 << inode->i_blkbits) - 1;
 
 		if (offset & blksize_mask || len & blksize_mask) {
-			error = EINVAL;
+			error = -EINVAL;
 			goto out_unlock;
 		}
 
@@ -781,7 +815,7 @@ xfs_file_fallocate(
 		 * in which case it is effectively a truncate operation
 		 */
 		if (offset + len >= i_size_read(inode)) {
-			error = EINVAL;
+			error = -EINVAL;
 			goto out_unlock;
 		}
 
@@ -794,7 +828,7 @@ xfs_file_fallocate(
 		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;
 		}
@@ -844,7 +878,7 @@ xfs_file_fallocate(
 
 out_unlock:
 	xfs_iunlock(ip, XFS_IOLOCK_EXCL);
-	return -error;
+	return error;
 }
 
 
@@ -889,7 +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
@@ -918,7 +952,7 @@ xfs_file_readdir(
 
 	error = xfs_readdir(ip, ctx, bufsize);
 	if (error)
-		return -error;
+		return error;
 	return 0;
 }
 
@@ -949,7 +983,7 @@ xfs_vm_page_mkwrite(
 
 /*
  * This type is designed to indicate the type of offset we would like
- * to search from page cache for either xfs_seek_data() or xfs_seek_hole().
+ * to search from page cache for xfs_seek_hole_data().
  */
 enum {
 	HOLE_OFF = 0,
@@ -1006,7 +1040,7 @@ xfs_lookup_buffer_offset(
 /*
  * This routine is called to find out and return a data or hole offset
  * from the page cache for unwritten extents according to the desired
- * type for xfs_seek_data() or xfs_seek_hole().
+ * type for xfs_seek_hole_data().
  *
  * The argument offset is used to tell where we start to search from the
  * page cache.  Map is used to figure out the end points of the range to
@@ -1166,9 +1200,10 @@ out:
 }
 
 STATIC loff_t
-xfs_seek_data(
+xfs_seek_hole_data(
 	struct file		*file,
-	loff_t			start)
+	loff_t			start,
+	int			whence)
 {
 	struct inode		*inode = file->f_mapping->host;
 	struct xfs_inode	*ip = XFS_I(inode);
@@ -1180,11 +1215,14 @@ xfs_seek_data(
 	uint			lock;
 	int			error;
 
+	if (XFS_FORCED_SHUTDOWN(mp))
+		return -EIO;
+
 	lock = xfs_ilock_data_map_shared(ip);
 
 	isize = i_size_read(inode);
 	if (start >= isize) {
-		error = ENXIO;
+		error = -ENXIO;
 		goto out_unlock;
 	}
 
@@ -1194,6 +1232,7 @@ xfs_seek_data(
 	 */
 	fsbno = XFS_B_TO_FSBT(mp, start);
 	end = XFS_B_TO_FSB(mp, isize);
+
 	for (;;) {
 		struct xfs_bmbt_irec	map[2];
 		int			nmap = 2;
@@ -1206,7 +1245,7 @@ xfs_seek_data(
 
 		/* No extents at given offset, must be beyond EOF */
 		if (nmap == 0) {
-			error = ENXIO;
+			error = -ENXIO;
 			goto out_unlock;
 		}
 
@@ -1214,30 +1253,49 @@ xfs_seek_data(
 			offset = max_t(loff_t, start,
 				       XFS_FSB_TO_B(mp, map[i].br_startoff));
 
-			/* Landed in a data extent */
-			if (map[i].br_startblock == DELAYSTARTBLOCK ||
-			    (map[i].br_state == XFS_EXT_NORM &&
-			     !isnullstartblock(map[i].br_startblock)))
+			/* Landed in the hole we wanted? */
+			if (whence == SEEK_HOLE &&
+			    map[i].br_startblock == HOLESTARTBLOCK)
+				goto out;
+
+			/* Landed in the data extent we wanted? */
+			if (whence == SEEK_DATA &&
+			    (map[i].br_startblock == DELAYSTARTBLOCK ||
+			     (map[i].br_state == XFS_EXT_NORM &&
+			      !isnullstartblock(map[i].br_startblock))))
 				goto out;
 
 			/*
-			 * Landed in an unwritten extent, try to search data
-			 * from page cache.
+			 * Landed in an unwritten extent, try to search
+			 * for hole or data from page cache.
 			 */
 			if (map[i].br_state == XFS_EXT_UNWRITTEN) {
 				if (xfs_find_get_desired_pgoff(inode, &map[i],
-							DATA_OFF, &offset))
+				      whence == SEEK_HOLE ? HOLE_OFF : DATA_OFF,
+							&offset))
 					goto out;
 			}
 		}
 
 		/*
-		 * map[0] is hole or its an unwritten extent but
-		 * without data in page cache.  Probably means that
-		 * we are reading after EOF if nothing in map[1].
+		 * We only received one extent out of the two requested. This
+		 * means we've hit EOF and didn't find what we are looking for.
 		 */
 		if (nmap == 1) {
-			error = ENXIO;
+			/*
+			 * If we were looking for a hole, set offset to
+			 * the end of the file (i.e., there is an implicit
+			 * hole at the end of any file).
+		 	 */
+			if (whence == SEEK_HOLE) {
+				offset = isize;
+				break;
+			}
+			/*
+			 * If we were looking for data, it's nowhere to be found
+			 */
+			ASSERT(whence == SEEK_DATA);
+			error = -ENXIO;
 			goto out_unlock;
 		}
 
@@ -1245,132 +1303,37 @@ xfs_seek_data(
 
 		/*
 		 * Nothing was found, proceed to the next round of search
-		 * if reading offset not beyond or hit EOF.
+		 * if the next reading offset is not at or beyond EOF.
 		 */
 		fsbno = map[i - 1].br_startoff + map[i - 1].br_blockcount;
 		start = XFS_FSB_TO_B(mp, fsbno);
 		if (start >= isize) {
-			error = ENXIO;
-			goto out_unlock;
-		}
-	}
-
-out:
-	offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes);
-
-out_unlock:
-	xfs_iunlock(ip, lock);
-
-	if (error)
-		return -error;
-	return offset;
-}
-
-STATIC loff_t
-xfs_seek_hole(
-	struct file		*file,
-	loff_t			start)
-{
-	struct inode		*inode = file->f_mapping->host;
-	struct xfs_inode	*ip = XFS_I(inode);
-	struct xfs_mount	*mp = ip->i_mount;
-	loff_t			uninitialized_var(offset);
-	xfs_fsize_t		isize;
-	xfs_fileoff_t		fsbno;
-	xfs_filblks_t		end;
-	uint			lock;
-	int			error;
-
-	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
-
-	lock = xfs_ilock_data_map_shared(ip);
-
-	isize = i_size_read(inode);
-	if (start >= isize) {
-		error = ENXIO;
-		goto out_unlock;
-	}
-
-	fsbno = XFS_B_TO_FSBT(mp, start);
-	end = XFS_B_TO_FSB(mp, isize);
-
-	for (;;) {
-		struct xfs_bmbt_irec	map[2];
-		int			nmap = 2;
-		unsigned int		i;
-
-		error = xfs_bmapi_read(ip, fsbno, end - fsbno, map, &nmap,
-				       XFS_BMAPI_ENTIRE);
-		if (error)
-			goto out_unlock;
-
-		/* No extents at given offset, must be beyond EOF */
-		if (nmap == 0) {
-			error = ENXIO;
-			goto out_unlock;
-		}
-
-		for (i = 0; i < nmap; i++) {
-			offset = max_t(loff_t, start,
-				       XFS_FSB_TO_B(mp, map[i].br_startoff));
-
-			/* Landed in a hole */
-			if (map[i].br_startblock == HOLESTARTBLOCK)
-				goto out;
-
-			/*
-			 * Landed in an unwritten extent, try to search hole
-			 * from page cache.
-			 */
-			if (map[i].br_state == XFS_EXT_UNWRITTEN) {
-				if (xfs_find_get_desired_pgoff(inode, &map[i],
-							HOLE_OFF, &offset))
-					goto out;
+			if (whence == SEEK_HOLE) {
+				offset = isize;
+				break;
 			}
-		}
-
-		/*
-		 * map[0] contains data or its unwritten but contains
-		 * data in page cache, probably means that we are
-		 * reading after EOF.  We should fix offset to point
-		 * to the end of the file(i.e., there is an implicit
-		 * hole at the end of any file).
-		 */
-		if (nmap == 1) {
-			offset = isize;
-			break;
-		}
-
-		ASSERT(i > 1);
-
-		/*
-		 * Both mappings contains data, proceed to the next round of
-		 * search if the current reading offset not beyond or hit EOF.
-		 */
-		fsbno = map[i - 1].br_startoff + map[i - 1].br_blockcount;
-		start = XFS_FSB_TO_B(mp, fsbno);
-		if (start >= isize) {
-			offset = isize;
-			break;
+			ASSERT(whence == SEEK_DATA);
+			error = -ENXIO;
+			goto out_unlock;
 		}
 	}
 
 out:
 	/*
-	 * At this point, we must have found a hole.  However, the returned
+	 * If at this point we have found the hole we wanted, the returned
 	 * offset may be bigger than the file size as it may be aligned to
-	 * page boundary for unwritten extents, we need to deal with this
+	 * page boundary for unwritten extents.  We need to deal with this
 	 * situation in particular.
 	 */
-	offset = min_t(loff_t, offset, isize);
+	if (whence == SEEK_HOLE)
+		offset = min_t(loff_t, offset, isize);
 	offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes);
 
 out_unlock:
 	xfs_iunlock(ip, lock);
 
 	if (error)
-		return -error;
+		return error;
 	return offset;
 }
 
@@ -1378,17 +1341,16 @@ STATIC loff_t
 xfs_file_llseek(
 	struct file	*file,
 	loff_t		offset,
-	int		origin)
+	int		whence)
 {
-	switch (origin) {
+	switch (whence) {
 	case SEEK_END:
 	case SEEK_CUR:
 	case SEEK_SET:
-		return generic_file_llseek(file, offset, origin);
-	case SEEK_DATA:
-		return xfs_seek_data(file, offset);
+		return generic_file_llseek(file, offset, whence);
 	case SEEK_HOLE:
-		return xfs_seek_hole(file, offset);
+	case SEEK_DATA:
+		return xfs_seek_hole_data(file, offset, whence);
 	default:
 		return -EINVAL;
 	}
diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c
index 8ec81bed7992..e92730c1d3ca 100644
--- a/fs/xfs/xfs_filestream.c
+++ b/fs/xfs/xfs_filestream.c
@@ -258,7 +258,7 @@ next_ag:
 	if (*agp == NULLAGNUMBER)
 		return 0;
 
-	err = ENOMEM;
+	err = -ENOMEM;
 	item = kmem_alloc(sizeof(*item), KM_MAYFAIL);
 	if (!item)
 		goto out_put_ag;
@@ -268,7 +268,7 @@ next_ag:
 
 	err = xfs_mru_cache_insert(mp->m_filestream, ip->i_ino, &item->mru);
 	if (err) {
-		if (err == EEXIST)
+		if (err == -EEXIST)
 			err = 0;
 		goto out_free_item;
 	}
diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h
index d34703dbcb42..18dc721ca19f 100644
--- a/fs/xfs/xfs_fs.h
+++ b/fs/xfs/xfs_fs.h
@@ -255,8 +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)
 
 /*
@@ -375,6 +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 d2295561570a..c05ac8b70fa9 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -168,20 +168,15 @@ 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;
-	bp = xfs_buf_read_uncached(mp->m_ddev_targp,
+	error = xfs_buf_read_uncached(mp->m_ddev_targp,
 				XFS_FSB_TO_BB(mp, nb) - XFS_FSS_TO_BB(mp, 1),
-				XFS_FSS_TO_BB(mp, 1), 0, NULL);
-	if (!bp)
-		return EIO;
-	if (bp->b_error) {
-		error = bp->b_error;
-		xfs_buf_relse(bp);
+				XFS_FSS_TO_BB(mp, 1), 0, &bp, NULL);
+	if (error)
 		return error;
-	}
 	xfs_buf_relse(bp);
 
 	new = nb;	/* use new as a temporary here */
@@ -191,7 +186,7 @@ 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;
@@ -229,7 +224,7 @@ xfs_growfs_data_private(
 				XFS_FSS_TO_BB(mp, 1), 0,
 				&xfs_agf_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -270,7 +265,7 @@ xfs_growfs_data_private(
 				XFS_FSS_TO_BB(mp, 1), 0,
 				&xfs_agfl_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -298,7 +293,7 @@ xfs_growfs_data_private(
 				XFS_FSS_TO_BB(mp, 1), 0,
 				&xfs_agi_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -336,7 +331,7 @@ xfs_growfs_data_private(
 				&xfs_allocbt_buf_ops);
 
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -365,7 +360,7 @@ xfs_growfs_data_private(
 				BTOBB(mp->m_sb.sb_blocksize), 0,
 				&xfs_allocbt_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -395,7 +390,7 @@ xfs_growfs_data_private(
 				BTOBB(mp->m_sb.sb_blocksize), 0,
 				&xfs_inobt_buf_ops);
 		if (!bp) {
-			error = ENOMEM;
+			error = -ENOMEM;
 			goto error0;
 		}
 
@@ -420,7 +415,7 @@ xfs_growfs_data_private(
 				BTOBB(mp->m_sb.sb_blocksize), 0,
 				&xfs_inobt_buf_ops);
 			if (!bp) {
-				error = ENOMEM;
+				error = -ENOMEM;
 				goto error0;
 			}
 
@@ -531,7 +526,7 @@ 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;
 		}
 
 		/*
@@ -576,17 +571,17 @@ 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;
 }
 
 /*
@@ -604,9 +599,9 @@ 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;
@@ -620,9 +615,9 @@ 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;
@@ -674,7 +669,7 @@ 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;
@@ -757,7 +752,7 @@ 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;
@@ -818,7 +813,7 @@ 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_globals.c b/fs/xfs/xfs_globals.c
index 5399ef222dd7..4d41b241298f 100644
--- a/fs/xfs/xfs_globals.c
+++ b/fs/xfs/xfs_globals.c
@@ -43,3 +43,7 @@ xfs_param_t xfs_params = {
 	.fstrm_timer	= {	1,		30*100,		3600*100},
 	.eofb_timer	= {	1,		300,		3600*24},
 };
+
+struct xfs_globals xfs_globals = {
+	.log_recovery_delay	=	0,	/* no delay by default */
+};
diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c
index c48df5f25b9f..b45f7b27b5df 100644
--- a/fs/xfs/xfs_icache.c
+++ b/fs/xfs/xfs_icache.c
@@ -33,6 +33,8 @@
 #include "xfs_trace.h"
 #include "xfs_icache.h"
 #include "xfs_bmap_util.h"
+#include "xfs_dquot_item.h"
+#include "xfs_dquot.h"
 
 #include <linux/kthread.h>
 #include <linux/freezer.h>
@@ -158,7 +160,7 @@ 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 +178,7 @@ 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 +186,7 @@ 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 +208,7 @@ 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 +245,7 @@ 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 +287,7 @@ 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 +296,7 @@ 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 +307,7 @@ xfs_iget_cache_miss(
 	 * recurse into the file system.
 	 */
 	if (radix_tree_preload(GFP_NOFS)) {
-		error = EAGAIN;
+		error = -EAGAIN;
 		goto out_destroy;
 	}
 
@@ -341,7 +343,7 @@ 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 +410,7 @@ 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 +447,7 @@ again:
 	return 0;
 
 out_error_or_again:
-	if (error == EAGAIN) {
+	if (error == -EAGAIN) {
 		delay(1);
 		goto again;
 	}
@@ -489,18 +491,18 @@ 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
@@ -583,16 +585,16 @@ restart:
 				continue;
 			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();
@@ -652,11 +654,11 @@ 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
@@ -680,11 +682,11 @@ 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;
 }
 
 /*
@@ -944,7 +946,7 @@ 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);
@@ -997,7 +999,7 @@ 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
@@ -1100,7 +1102,7 @@ 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;
 			}
 
@@ -1129,7 +1131,7 @@ restart:
 		trylock = 0;
 		goto restart;
 	}
-	return XFS_ERROR(last_error);
+	return last_error;
 }
 
 int
@@ -1203,6 +1205,30 @@ 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,
@@ -1211,6 +1237,10 @@ xfs_inode_free_eofblocks(
 {
 	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 */
@@ -1228,19 +1258,31 @@ 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;
@@ -1260,6 +1302,55 @@ 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 9cf017b899be..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 */
@@ -57,6 +58,7 @@ 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,
@@ -72,31 +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_inode.c b/fs/xfs/xfs_inode.c
index a6115fe1ac94..8ed049d1e332 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -583,7 +583,7 @@ xfs_lookup(
 	trace_xfs_lookup(dp, name);
 
 	if (XFS_FORCED_SHUTDOWN(dp->i_mount))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	lock_mode = xfs_ilock_data_map_shared(dp);
 	error = xfs_dir_lookup(NULL, dp, name, &inum, ci_name);
@@ -654,7 +654,7 @@ xfs_ialloc(
 	xfs_inode_t	*ip;
 	uint		flags;
 	int		error;
-	timespec_t	tv;
+	struct timespec	tv;
 
 	/*
 	 * Call the space management code to pick
@@ -720,7 +720,7 @@ xfs_ialloc(
 	ip->i_d.di_nextents = 0;
 	ASSERT(ip->i_d.di_nblocks == 0);
 
-	nanotime(&tv);
+	tv = current_fs_time(mp->m_super);
 	ip->i_d.di_mtime.t_sec = (__int32_t)tv.tv_sec;
 	ip->i_d.di_mtime.t_nsec = (__int32_t)tv.tv_nsec;
 	ip->i_d.di_atime = ip->i_d.di_mtime;
@@ -769,6 +769,8 @@ xfs_ialloc(
 					di_flags |= XFS_DIFLAG_EXTSZINHERIT;
 					ip->i_d.di_extsize = pip->i_d.di_extsize;
 				}
+				if (pip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)
+					di_flags |= XFS_DIFLAG_PROJINHERIT;
 			} else if (S_ISREG(mode)) {
 				if (pip->i_d.di_flags & XFS_DIFLAG_RTINHERIT)
 					di_flags |= XFS_DIFLAG_REALTIME;
@@ -789,8 +791,6 @@ xfs_ialloc(
 			if ((pip->i_d.di_flags & XFS_DIFLAG_NOSYMLINKS) &&
 			    xfs_inherit_nosymlinks)
 				di_flags |= XFS_DIFLAG_NOSYMLINKS;
-			if (pip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT)
-				di_flags |= XFS_DIFLAG_PROJINHERIT;
 			if ((pip->i_d.di_flags & XFS_DIFLAG_NODEFRAG) &&
 			    xfs_inherit_nodefrag)
 				di_flags |= XFS_DIFLAG_NODEFRAG;
@@ -893,7 +893,7 @@ xfs_dir_ialloc(
 	}
 	if (!ialloc_context && !ip) {
 		*ipp = NULL;
-		return XFS_ERROR(ENOSPC);
+		return -ENOSPC;
 	}
 
 	/*
@@ -1088,7 +1088,7 @@ xfs_create(
 	trace_xfs_create(dp, name);
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	prid = xfs_get_initial_prid(dp);
 
@@ -1125,12 +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);
@@ -1153,9 +1153,11 @@ xfs_create(
 	if (error)
 		goto out_trans_cancel;
 
-	error = xfs_dir_canenter(tp, dp, name, resblks);
-	if (error)
-		goto out_trans_cancel;
+	if (!resblks) {
+		error = xfs_dir_canenter(tp, dp, name);
+		if (error)
+			goto out_trans_cancel;
+	}
 
 	/*
 	 * A newly created regular or special file just has one directory
@@ -1165,7 +1167,7 @@ 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;
 	}
@@ -1184,7 +1186,7 @@ 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);
@@ -1274,7 +1276,7 @@ xfs_create_tmpfile(
 	uint			resblks;
 
 	if (XFS_FORCED_SHUTDOWN(mp))
-		return XFS_ERROR(EIO);
+		return -EIO;
 
 	prid = xfs_get_initial_prid(dp);
 
@@ -1293,7 +1295,7 @@ xfs_create_tmpfile(
 
 	tres = &M_RES(mp)->tr_create_tmpfile;
 	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);
@@ -1311,7 +1313,7 @@ xfs_create_tmpfile(
 	error = xfs_dir_ialloc(&tp, dp, mode, 1, 0,
 				prid, resblks > 0, &ip, NULL);
 	if (error) {
-		if (error == ENOSPC)
+		if (error == -ENOSPC)
 			goto out_trans_cancel;
 		goto out_trans_abort;
 	}
@@ -1382,7 +1384,7 @@ 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)
@@ -1396,7 +1398,7 @@ 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);
 	}
@@ -1417,13 +1419,15 @@ 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;
 	}
 
-	error = xfs_dir_canenter(tp, tdp, target_name, resblks);
-	if (error)
-		goto error_return;
+	if (!resblks) {
+		error = xfs_dir_canenter(tp, tdp, target_name);
+		if (error)
+			goto error_return;
+	}
 
 	xfs_bmap_init(&free_list, &first_block);
 
@@ -1635,8 +1639,8 @@ 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;
 			}
@@ -1673,7 +1677,7 @@ 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 */
@@ -1772,7 +1776,7 @@ xfs_inactive_ifree(
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ifree,
 				  XFS_IFREE_SPACE_RES(mp), 0);
 	if (error) {
-		if (error == ENOSPC) {
+		if (error == -ENOSPC) {
 			xfs_warn_ratelimited(mp,
 			"Failed to remove inode(s) from unlinked list. "
 			"Please free space, unmount and run xfs_repair.");
@@ -2219,7 +2223,7 @@ xfs_ifree_cluster(
 					XBF_UNMAPPED);
 
 		if (!bp)
-			return ENOMEM;
+			return -ENOMEM;
 
 		/*
 		 * This buffer may not have been correctly initialised as we
@@ -2491,7 +2495,7 @@ 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)
@@ -2521,12 +2525,12 @@ 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;
 	}
@@ -2543,11 +2547,11 @@ 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;
 		}
 
@@ -2582,7 +2586,7 @@ 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;
 	}
 
@@ -2702,7 +2706,7 @@ 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);
 	}
@@ -2747,7 +2751,7 @@ 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;
 	}
 
@@ -2759,9 +2763,11 @@ xfs_rename(
 		 * If there's no space reservation, check the entry will
 		 * fit before actually inserting it.
 		 */
-		error = xfs_dir_canenter(tp, target_dp, target_name, spaceres);
-		if (error)
-			goto error_return;
+		if (!spaceres) {
+			error = xfs_dir_canenter(tp, target_dp, target_name);
+			if (error)
+				goto error_return;
+		}
 		/*
 		 * If target does not exist and the rename crosses
 		 * directories, adjust the target directory link count
@@ -2770,7 +2776,7 @@ 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;
@@ -2795,7 +2801,7 @@ xfs_rename(
 			 */
 			if (!(xfs_dir_isempty(target_ip)) ||
 			    (target_ip->i_d.di_nlink > 2)) {
-				error = XFS_ERROR(EEXIST);
+				error = -EEXIST;
 				goto error_return;
 			}
 		}
@@ -2847,7 +2853,7 @@ 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;
 	}
@@ -3055,8 +3061,8 @@ cluster_corrupt_out:
 		if (bp->b_iodone) {
 			XFS_BUF_UNDONE(bp);
 			xfs_buf_stale(bp);
-			xfs_buf_ioerror(bp, EIO);
-			xfs_buf_ioend(bp, 0);
+			xfs_buf_ioerror(bp, -EIO);
+			xfs_buf_ioend(bp);
 		} else {
 			xfs_buf_stale(bp);
 			xfs_buf_relse(bp);
@@ -3069,7 +3075,7 @@ cluster_corrupt_out:
 	xfs_iflush_abort(iq, false);
 	kmem_free(ilist);
 	xfs_perag_put(pag);
-	return XFS_ERROR(EFSCORRUPTED);
+	return -EFSCORRUPTED;
 }
 
 /*
@@ -3124,7 +3130,7 @@ 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;
 	}
 
@@ -3167,7 +3173,7 @@ 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
@@ -3331,5 +3337,5 @@ 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 f72bffa67266..9af2882e1f4c 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -102,7 +102,7 @@ xfs_new_eof(struct xfs_inode *ip, xfs_fsize_t new_size)
 {
 	xfs_fsize_t i_size = i_size_read(VFS_I(ip));
 
-	if (new_size > i_size)
+	if (new_size > i_size || new_size < 0)
 		new_size = i_size;
 	return new_size > ip->i_d.di_size ? new_size : 0;
 }
@@ -398,4 +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 a640137b3573..63de0b0acc32 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -615,7 +615,7 @@ xfs_iflush_done(
 	blip = bp->b_fspriv;
 	prev = NULL;
 	while (blip != NULL) {
-		if (lip->li_cb != xfs_iflush_done) {
+		if (blip->li_cb != xfs_iflush_done) {
 			prev = blip;
 			blip = blip->li_bio_list;
 			continue;
@@ -788,5 +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_ioctl.c b/fs/xfs/xfs_ioctl.c
index 8bc1bbce7451..24c926b6fe85 100644
--- a/fs/xfs/xfs_ioctl.c
+++ b/fs/xfs/xfs_ioctl.c
@@ -207,7 +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))
@@ -216,7 +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;
 	}
 
@@ -228,18 +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;
 	}
 
@@ -282,7 +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))
@@ -290,22 +290,22 @@ 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 = readlink_copy(hreq->ohandle, olen, link);
@@ -330,10 +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);
@@ -364,9 +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)
@@ -379,16 +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:
@@ -409,18 +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))
@@ -431,7 +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;
@@ -455,20 +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);
@@ -484,20 +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
@@ -507,7 +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);
 }
 
@@ -524,9 +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))
@@ -536,18 +533,18 @@ 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;
 
 	ops = memdup_user(am_hreq.ops, size);
 	if (IS_ERR(ops)) {
-		error = -PTR_ERR(ops);
+		error = PTR_ERR(ops);
 		goto out_dput;
 	}
 
-	error = ENOMEM;
+	error = -ENOMEM;
 	attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL);
 	if (!attr_name)
 		goto out_kfree_ops;
@@ -557,7 +554,7 @@ xfs_attrmulti_by_handle(
 		ops[i].am_error = strncpy_from_user((char *)attr_name,
 				ops[i].am_attrname, MAXNAMELEN);
 		if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN)
-			error = ERANGE;
+			error = -ERANGE;
 		if (ops[i].am_error < 0)
 			break;
 
@@ -588,19 +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
@@ -625,16 +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)
@@ -652,7 +649,7 @@ xfs_ioc_space(
 		bf->l_start += XFS_ISIZE(ip);
 		break;
 	default:
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 		goto out_unlock;
 	}
 
@@ -669,7 +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;
@@ -682,7 +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;
 	}
 
@@ -723,7 +720,7 @@ xfs_ioc_space(
 		break;
 	default:
 		ASSERT(0);
-		error = XFS_ERROR(EINVAL);
+		error = -EINVAL;
 	}
 
 	if (error)
@@ -739,7 +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;
@@ -759,7 +756,7 @@ xfs_ioc_space(
 out_unlock:
 	xfs_iunlock(ip, XFS_IOLOCK_EXCL);
 	mnt_drop_write_file(filp);
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -781,41 +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;
@@ -831,7 +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
@@ -839,7 +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;
 }
 
@@ -853,10 +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;
 }
 
@@ -971,8 +968,6 @@ xfs_set_diflags(
 		di_flags |= XFS_DIFLAG_NOATIME;
 	if (xflags & XFS_XFLAG_NODUMP)
 		di_flags |= XFS_DIFLAG_NODUMP;
-	if (xflags & XFS_XFLAG_PROJINHERIT)
-		di_flags |= XFS_DIFLAG_PROJINHERIT;
 	if (xflags & XFS_XFLAG_NODEFRAG)
 		di_flags |= XFS_DIFLAG_NODEFRAG;
 	if (xflags & XFS_XFLAG_FILESTREAM)
@@ -984,6 +979,8 @@ xfs_set_diflags(
 			di_flags |= XFS_DIFLAG_NOSYMLINKS;
 		if (xflags & XFS_XFLAG_EXTSZINHERIT)
 			di_flags |= XFS_DIFLAG_EXTSZINHERIT;
+		if (xflags & XFS_XFLAG_PROJINHERIT)
+			di_flags |= XFS_DIFLAG_PROJINHERIT;
 	} else if (S_ISREG(ip->i_d.di_mode)) {
 		if (xflags & XFS_XFLAG_REALTIME)
 			di_flags |= XFS_DIFLAG_REALTIME;
@@ -1041,16 +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,
@@ -1088,7 +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;
 	}
 
@@ -1099,7 +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;
 		}
 
@@ -1122,7 +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;
 		}
 
@@ -1141,7 +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;
 			}
 
@@ -1153,13 +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;
 			}
 		}
@@ -1173,7 +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;
 		}
 
@@ -1184,7 +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;
 			}
 		}
@@ -1198,7 +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;
 		}
 	}
@@ -1234,13 +1231,25 @@ xfs_ioctl_setattr(
 
 	}
 
-	if (mask & FSX_EXTSIZE)
-		ip->i_d.di_extsize = fa->fsx_extsize >> mp->m_sb.sb_blocklog;
 	if (mask & FSX_XFLAGS) {
 		xfs_set_diflags(ip, fa->fsx_xflags);
 		xfs_diflags_to_linux(ip);
 	}
 
+	/*
+	 * Only set the extent size hint if we've already determined that the
+	 * extent size hint should be set on the inode. If no extent size flags
+	 * are set on the inode then unconditionally clear the extent size hint.
+	 */
+	if (mask & FSX_EXTSIZE) {
+		int	extsize = 0;
+
+		if (ip->i_d.di_flags &
+				(XFS_DIFLAG_EXTSIZE | XFS_DIFLAG_EXTSZINHERIT))
+			extsize = fa->fsx_extsize >> mp->m_sb.sb_blocklog;
+		ip->i_d.di_extsize = extsize;
+	}
+
 	xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG);
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 
@@ -1301,7 +1310,7 @@ xfs_ioc_fssetxattr(
 		return error;
 	error = xfs_ioctl_setattr(ip, &fa, mask);
 	mnt_drop_write_file(filp);
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -1346,17 +1355,17 @@ xfs_ioc_setxflags(
 		return error;
 	error = xfs_ioctl_setattr(ip, &fa, mask);
 	mnt_drop_write_file(filp);
-	return -error;
+	return error;
 }
 
 STATIC int
 xfs_getbmap_format(void **ap, struct getbmapx *bmv, int *full)
 {
-	struct getbmap __user	*base = *ap;
+	struct getbmap __user	*base = (struct getbmap __user *)*ap;
 
 	/* 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;
@@ -1373,33 +1382,33 @@ 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);
+			    (__force 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;
 }
 
 STATIC int
 xfs_getbmapx_format(void **ap, struct getbmapx *bmv, int *full)
 {
-	struct getbmapx __user	*base = *ap;
+	struct getbmapx __user	*base = (struct getbmapx __user *)*ap;
 
 	if (copy_to_user(base, bmv, sizeof(struct getbmapx)))
-		return XFS_ERROR(EFAULT);
+		return -EFAULT;
 
 	*ap += sizeof(struct getbmapx);
 	return 0;
@@ -1414,22 +1423,22 @@ 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);
+			    (__force 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;
 }
@@ -1445,33 +1454,33 @@ 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;
 	}
 
@@ -1479,17 +1488,17 @@ 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;
 	}
 
@@ -1523,7 +1532,7 @@ xfs_file_ioctl(
 	int			error;
 
 	if (filp->f_mode & FMODE_NOCMTIME)
-		ioflags |= IO_INVIS;
+		ioflags |= XFS_IO_INVIS;
 
 	trace_xfs_file_ioctl(ip);
 
@@ -1542,7 +1551,7 @@ 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: {
@@ -1555,7 +1564,7 @@ xfs_file_ioctl(
 		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;
 	}
 
@@ -1588,7 +1597,7 @@ 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)
@@ -1597,7 +1606,7 @@ 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:
@@ -1613,14 +1622,14 @@ 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:
@@ -1630,7 +1639,7 @@ 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:
@@ -1643,13 +1652,13 @@ 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: {
@@ -1657,10 +1666,10 @@ 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;
 	}
 
@@ -1672,10 +1681,10 @@ 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)
@@ -1686,10 +1695,10 @@ 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;
 	}
 
@@ -1701,10 +1710,10 @@ 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;
 	}
@@ -1713,42 +1722,42 @@ 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: {
@@ -1758,10 +1767,9 @@ 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: {
@@ -1771,18 +1779,16 @@ 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;
@@ -1792,16 +1798,16 @@ 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 944d5baa710a..94ce027e28e3 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;
 }
@@ -161,10 +160,11 @@ xfs_ioctl32_bstat_copyin(
 	    get_user(bstat->bs_gen,	&bstat32->bs_gen)	||
 	    get_user(bstat->bs_projid_lo, &bstat32->bs_projid_lo) ||
 	    get_user(bstat->bs_projid_hi, &bstat32->bs_projid_hi) ||
+	    get_user(bstat->bs_forkoff,	&bstat32->bs_forkoff)	||
 	    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 +180,7 @@ 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 +195,7 @@ 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)	||
@@ -215,10 +215,11 @@ xfs_bulkstat_one_fmt_compat(
 	    put_user(buffer->bs_gen,	  &p32->bs_gen)		||
 	    put_user(buffer->bs_projid,	  &p32->bs_projid)	||
 	    put_user(buffer->bs_projid_hi,	&p32->bs_projid_hi)	||
+	    put_user(buffer->bs_forkoff,  &p32->bs_forkoff)	||
 	    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 +257,30 @@ 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 +295,17 @@ 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 +319,7 @@ 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 +353,19 @@ 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 +377,7 @@ 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 +405,10 @@ 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))
@@ -417,7 +418,7 @@ 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;
@@ -428,7 +429,7 @@ xfs_compat_attrmulti_by_handle(
 		goto out_dput;
 	}
 
-	error = ENOMEM;
+	error = -ENOMEM;
 	attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL);
 	if (!attr_name)
 		goto out_kfree_ops;
@@ -439,7 +440,7 @@ xfs_compat_attrmulti_by_handle(
 				compat_ptr(ops[i].am_attrname),
 				MAXNAMELEN);
 		if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN)
-			error = ERANGE;
+			error = -ERANGE;
 		if (ops[i].am_error < 0)
 			break;
 
@@ -470,19 +471,19 @@ 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
@@ -496,26 +497,26 @@ 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:
@@ -537,7 +538,7 @@ 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);
 
@@ -588,7 +589,7 @@ 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);
 	}
@@ -598,25 +599,25 @@ 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 */
@@ -633,13 +634,13 @@ 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:
@@ -651,7 +652,7 @@ 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);
 	}
@@ -659,14 +660,14 @@ 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:
@@ -676,6 +677,6 @@ 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_ioctl32.h b/fs/xfs/xfs_ioctl32.h
index 80f4060e8970..b1bb45444df8 100644
--- a/fs/xfs/xfs_ioctl32.h
+++ b/fs/xfs/xfs_ioctl32.h
@@ -67,8 +67,9 @@ typedef struct compat_xfs_bstat {
 	__u32		bs_gen;		/* generation count		*/
 	__u16		bs_projid_lo;	/* lower part of project id	*/
 #define	bs_projid	bs_projid_lo	/* (previously just bs_projid)	*/
+	__u16		bs_forkoff;	/* inode fork offset in bytes	*/
 	__u16		bs_projid_hi;	/* high part of project id	*/
-	unsigned char	bs_pad[12];	/* pad space, unused		*/
+	unsigned char	bs_pad[10];	/* pad space, unused		*/
 	__u32		bs_dmevmask;	/* DMIG event mask		*/
 	__u16		bs_dmstate;	/* DMIG state info		*/
 	__u16		bs_aextents;	/* attribute number of extents	*/
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 6d3ec2b6ee29..afcf3c926565 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
@@ -138,7 +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);
@@ -148,7 +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)
@@ -188,7 +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);
@@ -225,7 +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;
 	}
 
@@ -397,15 +397,17 @@ 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;
 	struct xfs_dquot *dq = xfs_inode_dquot(ip, type);
 
-	/* over hi wmark, squash the prealloc completely */
-	if (dq->q_res_bcount >= dq->q_prealloc_hi_wmark) {
+	/* no dq, or over hi wmark, squash the prealloc completely */
+	if (!dq || dq->q_res_bcount >= dq->q_prealloc_hi_wmark) {
 		*qblocks = 0;
+		*qfreesp = 0;
 		return;
 	}
 
@@ -418,6 +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;
@@ -476,15 +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
@@ -552,7 +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);
@@ -596,11 +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;
 	}
 
 	/*
@@ -614,7 +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)))
@@ -663,7 +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;
@@ -690,7 +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);
@@ -739,7 +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;
 				}
 			}
@@ -793,7 +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
@@ -853,7 +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);
@@ -892,7 +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);
@@ -915,5 +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 205613a06068..ec6dcdc181ee 100644
--- a/fs/xfs/xfs_iops.c
+++ b/fs/xfs/xfs_iops.c
@@ -72,7 +72,7 @@ xfs_initxattrs(
 	int			error = 0;
 
 	for (xattr = xattr_array; xattr->name != NULL; xattr++) {
-		error = -xfs_attr_set(ip, xattr->name, xattr->value,
+		error = xfs_attr_set(ip, xattr->name, xattr->value,
 				      xattr->value_len, ATTR_SECURE);
 		if (error < 0)
 			break;
@@ -93,7 +93,7 @@ xfs_init_security(
 	struct inode	*dir,
 	const struct qstr *qstr)
 {
-	return -security_inode_init_security(inode, dir, qstr,
+	return security_inode_init_security(inode, dir, qstr,
 					     &xfs_initxattrs, NULL);
 }
 
@@ -173,12 +173,12 @@ xfs_generic_create(
 
 #ifdef CONFIG_XFS_POSIX_ACL
 	if (default_acl) {
-		error = -xfs_set_acl(inode, default_acl, ACL_TYPE_DEFAULT);
+		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);
+		error = xfs_set_acl(inode, acl, ACL_TYPE_ACCESS);
 		if (error)
 			goto out_cleanup_inode;
 	}
@@ -194,7 +194,7 @@ xfs_generic_create(
 		posix_acl_release(default_acl);
 	if (acl)
 		posix_acl_release(acl);
-	return -error;
+	return error;
 
  out_cleanup_inode:
 	if (!tmpfile)
@@ -248,8 +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;
 	}
@@ -275,8 +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
@@ -311,7 +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);
@@ -328,7 +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;
 
@@ -375,7 +375,7 @@ xfs_vn_symlink(
 	xfs_cleanup_inode(dir, inode, dentry);
 	iput(inode);
  out:
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -392,8 +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);
 }
 
@@ -414,7 +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;
 
@@ -441,7 +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;
@@ -546,14 +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);
@@ -703,7 +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
@@ -713,9 +713,9 @@ xfs_setattr_nonsize(
 	 * 	     Posix ACL code seems to care about this issue either.
 	 */
 	if ((mask & ATTR_MODE) && !(flags & XFS_ATTR_NOACL)) {
-		error = -posix_acl_chmod(inode, inode->i_mode);
+		error = posix_acl_chmod(inode, inode->i_mode);
 		if (error)
-			return XFS_ERROR(error);
+			return error;
 	}
 
 	return 0;
@@ -748,14 +748,14 @@ 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));
@@ -818,7 +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;
@@ -844,11 +844,41 @@ xfs_setattr_size(
 	 * 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);
+	error = block_truncate_page(inode->i_mapping, newsize, xfs_get_blocks);
 	if (error)
 		return error;
 	truncate_setsize(inode, newsize);
 
+	/*
+	 * The "we can't serialise against page faults" pain gets worse.
+	 *
+	 * If the file is mapped then we have to clean the page at the old EOF
+	 * when extending the file. Extending the file can expose changes the
+	 * underlying page mapping (e.g. from beyond EOF to a hole or
+	 * unwritten), and so on the next attempt to write to that page we need
+	 * to remap it for write. i.e. we need .page_mkwrite() to be called.
+	 * Hence we need to clean the page to clean the pte and so a new write
+	 * fault will be triggered appropriately.
+	 *
+	 * If we do it before we change the inode size, then we can race with a
+	 * page fault that maps the page with exactly the same problem. If we do
+	 * it after we change the file size, then a new page fault can come in
+	 * and allocate space before we've run the rest of the truncate
+	 * transaction. That's kinda grotesque, but it's better than have data
+	 * over a hole, and so that's the lesser evil that has been chosen here.
+	 *
+	 * The real solution, however, is to have some mechanism for locking out
+	 * page faults while a truncate is in progress.
+	 */
+	if (newsize > oldsize && mapping_mapped(VFS_I(ip)->i_mapping)) {
+		error = filemap_write_and_wait_range(
+				VFS_I(ip)->i_mapping,
+				round_down(oldsize, PAGE_CACHE_SIZE),
+				round_up(oldsize, PAGE_CACHE_SIZE) - 1);
+		if (error)
+			return error;
+	}
+
 	tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_SIZE);
 	error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0);
 	if (error)
@@ -950,7 +980,7 @@ xfs_vn_setattr(
 		error = xfs_setattr_nonsize(ip, iattr, 0);
 	}
 
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -970,7 +1000,7 @@ 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);
@@ -991,7 +1021,7 @@ 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)
@@ -1036,7 +1066,7 @@ xfs_fiemap_format(
 		*full = 1;	/* user array now full */
 	}
 
-	return -error;
+	return error;
 }
 
 STATIC int
@@ -1055,12 +1085,12 @@ 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 :
@@ -1075,7 +1105,7 @@ xfs_vn_fiemap(
 
 	error = xfs_getbmap(ip, &bm, xfs_fiemap_format, fieinfo);
 	if (error)
-		return -error;
+		return error;
 
 	return 0;
 }
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c
index cb64f222d607..894924a5129b 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;
@@ -175,8 +172,172 @@ 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) {
+		*icount = 0;
+		return 0;
+	}
+
+	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)
 
+struct xfs_bulkstat_agichunk {
+	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 */
+};
+
+/*
+ * Process inodes in chunk with a pointer to a formatter function
+ * that will iget the inode and fill in the appropriate structure.
+ */
+static 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_agino_t			*last_agino)
+{
+	char				__user **ubufp = acp->ac_ubuffer;
+	int				chunkidx;
+	int				error = 0;
+	xfs_agino_t			agino = irbp->ir_startino;
+
+	for (chunkidx = 0; chunkidx < XFS_INODES_PER_CHUNK;
+	     chunkidx++, agino++) {
+		int		fmterror;
+		int		ubused;
+
+		/* inode won't fit in buffer, we are done */
+		if (acp->ac_ubleft < statstruct_size)
+			break;
+
+		/* Skip if this inode is free */
+		if (XFS_INOBT_MASK(chunkidx) & irbp->ir_free)
+			continue;
+
+		/* Get the inode and fill in a single buffer */
+		ubused = statstruct_size;
+		error = formatter(mp, XFS_AGINO_TO_INO(mp, agno, agino),
+				  *ubufp, acp->ac_ubleft, &ubused, &fmterror);
+
+		if (fmterror == BULKSTAT_RV_GIVEUP ||
+		    (error && error != -ENOENT && error != -EINVAL)) {
+			acp->ac_ubleft = 0;
+			ASSERT(error);
+			break;
+		}
+
+		/* be careful not to leak error if at end of chunk */
+		if (fmterror == BULKSTAT_RV_NOTHING || error) {
+			error = 0;
+			continue;
+		}
+
+		*ubufp += ubused;
+		acp->ac_ubleft -= ubused;
+		acp->ac_ubelem++;
+	}
+
+	/*
+	 * Post-update *last_agino. At this point, agino will always point one
+	 * inode past the last inode we processed successfully. Hence we
+	 * substract that inode when setting the *last_agino cursor so that we
+	 * return the correct cookie to userspace. On the next bulkstat call,
+	 * the inode under the lastino cookie will be skipped as we have already
+	 * processed it here.
+	 */
+	*last_agino = agino - 1;
+
+	return error;
+}
+
 /*
  * Return stat information in bulk (by-inode) for the filesystem.
  */
@@ -190,63 +351,41 @@ 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 */
-	int                     fmterror;/* bulkstat formatter result */
-	int			i;	/* loop index */
-	int			icount;	/* count of inodes good in irbuf */
 	size_t			irbsize; /* size of irec buffer in bytes */
-	xfs_ino_t		ino;	/* inode number (filesystem) */
-	xfs_inobt_rec_incore_t	*irbp;	/* current irec buffer pointer */
 	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			blks_per_cluster; /* # of blocks per cluster */
-	int			inodes_per_cluster;/* # of inodes per cluster */
 	int			nirbuf;	/* size of irbuf */
-	int			rval;	/* return value error code */
-	int			tmp;	/* result value from btree calls */
 	int			ubcount; /* size of user's buffer */
-	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 */
+	struct xfs_bulkstat_agichunk ac;
+	int			error = 0;
 
 	/*
 	 * Get the last inode value, see if there's nothing to do.
 	 */
-	ino = (xfs_ino_t)*lastinop;
-	lastino = ino;
-	agno = XFS_INO_TO_AGNO(mp, ino);
-	agino = XFS_INO_TO_AGINO(mp, ino);
+	agno = XFS_INO_TO_AGNO(mp, *lastinop);
+	agino = XFS_INO_TO_AGINO(mp, *lastinop);
 	if (agno >= mp->m_sb.sb_agcount ||
-	    ino != XFS_AGINO_TO_INO(mp, agno, agino)) {
+	    *lastinop != XFS_AGINO_TO_INO(mp, agno, agino)) {
 		*done = 1;
 		*ubcountp = 0;
 		return 0;
 	}
-	if (!ubcountp || *ubcountp <= 0) {
-		return EINVAL;
-	}
+
 	ubcount = *ubcountp; /* statstruct's */
-	ubleft = ubcount * statstruct_size; /* bytes */
-	*ubcountp = ubelem = 0;
+	ac.ac_ubuffer = &ubuffer;
+	ac.ac_ubleft = ubcount * statstruct_size; /* bytes */;
+	ac.ac_ubelem = 0;
+
+	*ubcountp = 0;
 	*done = 0;
-	fmterror = 0;
-	ubufp = ubuffer;
-	blks_per_cluster = xfs_icluster_size_fsb(mp);
-	inodes_per_cluster = blks_per_cluster << 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);
 
@@ -254,122 +393,60 @@ xfs_bulkstat(
 	 * Loop over the allocation groups, starting from the last
 	 * inode returned; 0 means start of the allocation group.
 	 */
-	rval = 0;
-	while (XFS_BULKSTAT_UBLEFT(ubleft) && agno < mp->m_sb.sb_agcount) {
-		cond_resched();
+	while (agno < mp->m_sb.sb_agcount) {
+		struct xfs_inobt_rec_incore	*irbp = irbuf;
+		struct xfs_inobt_rec_incore	*irbufend = irbuf + nirbuf;
+		bool				end_of_ag = false;
+		int				icount = 0;
+		int				stat;
+
 		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,
 					    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.
-		 */
 		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)
+				goto del_cursor;
+			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, &stat);
 		} else {
-			/*
-			 * Start of ag.  Lookup the first inode chunk.
-			 */
-			error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &tmp);
-			icount = 0;
+			/* Start of ag.  Lookup the first inode chunk */
+			error = xfs_inobt_lookup(cur, 0, XFS_LOOKUP_GE, &stat);
 		}
+		if (error || stat == 0) {
+			end_of_ag = true;
+			goto del_cursor;
+		}
+
 		/*
 		 * 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) {
-				end_of_ag = 1;
-				break;
+			error = xfs_inobt_get_rec(cur, &r, &stat);
+			if (error || stat == 0) {
+				end_of_ag = true;
+				goto del_cursor;
 			}
 
 			/*
@@ -377,193 +454,92 @@ 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 += inodes_per_cluster,
-				     agbno += blks_per_cluster) {
-					if (xfs_inobt_maskn(chunkidx,
-					    inodes_per_cluster) & ~r.ir_free)
-						xfs_btree_reada_bufs(mp, agno,
-							agbno, blks_per_cluster,
-							&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;
 				irbp++;
 				icount += XFS_INODES_PER_CHUNK - r.ir_freecount;
 			}
-			/*
-			 * Set agino to after this chunk and bump the cursor.
-			 */
-			agino = r.ir_startino + XFS_INODES_PER_CHUNK;
-			error = xfs_btree_increment(cur, 0, &tmp);
+			error = xfs_btree_increment(cur, 0, &stat);
+			if (error || stat == 0) {
+				end_of_ag = true;
+				goto del_cursor;
+			}
 			cond_resched();
 		}
+
 		/*
-		 * Drop the btree buffers and the agi buffer.
-		 * We can't hold any of the locks these represent
-		 * when calling iget.
+		 * Drop the btree buffers and the agi buffer as we can't hold any
+		 * of the locks these represent when calling iget. If there is a
+		 * pending error, then we are done.
 		 */
+del_cursor:
 		xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
 		xfs_buf_relse(agbp);
+		if (error)
+			break;
 		/*
-		 * Now format all the good inodes into the user's buffer.
+		 * Now format all the good inodes into the user's buffer. The
+		 * call to xfs_bulkstat_ag_ichunk() sets up the agino pointer
+		 * for the next loop iteration.
 		 */
 		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;
-			}
+		     irbp < irbufend && ac.ac_ubleft >= statstruct_size;
+		     irbp++) {
+			error = xfs_bulkstat_ag_ichunk(mp, agno, irbp,
+					formatter, statstruct_size, &ac,
+					&agino);
+			if (error)
+				break;
 
 			cond_resched();
 		}
+
 		/*
-		 * Set up for the next loop iteration.
+		 * If we've run out of space or had a formatting error, we
+		 * are now done
 		 */
-		if (XFS_BULKSTAT_UBLEFT(ubleft)) {
-			if (end_of_ag) {
-				agno++;
-				agino = 0;
-			} else
-				agino = XFS_INO_TO_AGINO(mp, lastino);
-		} else
+		if (ac.ac_ubleft < statstruct_size || error)
 			break;
+
+		if (end_of_ag) {
+			agno++;
+			agino = 0;
+		}
 	}
 	/*
 	 * Done, we're either out of filesystem or space to put the data.
 	 */
 	kmem_free(irbuf);
-	*ubcountp = ubelem;
+	*ubcountp = ac.ac_ubelem;
+
 	/*
-	 * Found some inodes, return them now and return the error next time.
+	 * We found some inodes, so clear the error status and return them.
+	 * The lastino pointer will point directly at the inode that triggered
+	 * any error that occurred, so on the next call the error will be
+	 * triggered again and propagated to userspace as there will be no
+	 * formatted inodes in the buffer.
 	 */
-	if (ubelem)
-		rval = 0;
-	if (agno >= mp->m_sb.sb_agcount) {
-		/*
-		 * If we ran out of filesystem, mark lastino as off
-		 * the end of the filesystem, so the next call
-		 * will return immediately.
-		 */
-		*lastinop = (xfs_ino_t)XFS_AGINO_TO_INO(mp, agno, 0);
-		*done = 1;
-	} else
-		*lastinop = (xfs_ino_t)lastino;
-
-	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 */
+	if (ac.ac_ubelem)
+		error = 0;
 
 	/*
-	 * 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.
+	 * If we ran out of filesystem, lastino will point off the end of
+	 * the filesystem so the next call will return immediately.
 	 */
+	*lastinop = XFS_AGINO_TO_INO(mp, agno, agino);
+	if (agno >= mp->m_sb.sb_agcount)
+		*done = 1;
 
-	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;
+	return error;
 }
 
 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 */
 {
@@ -578,127 +554,105 @@ 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;
-			}
+			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;
+		agno++;
+	} 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..6ea8b3912fa4 100644
--- a/fs/xfs/xfs_itable.h
+++ b/fs/xfs/xfs_itable.h
@@ -50,13 +50,6 @@ 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 825249d2dfc1..6a51619d8690 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;
@@ -68,7 +56,6 @@ typedef __uint64_t __psunsigned_t;
 
 #include "kmem.h"
 #include "mrlock.h"
-#include "time.h"
 #include "uuid.h"
 
 #include <linux/semaphore.h>
@@ -113,7 +100,7 @@ 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"
@@ -191,6 +178,22 @@ typedef __uint64_t __psunsigned_t;
 #define MAX(a,b)	(max(a,b))
 #define howmany(x, y)	(((x)+((y)-1))/(y))
 
+static inline void delay(long ticks)
+{
+	schedule_timeout_uninterruptible(ticks);
+}
+
+/*
+ * 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
@@ -331,7 +334,7 @@ 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 292308dede6d..fe88ef67f93a 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;
@@ -628,7 +629,7 @@ xfs_log_mount(
 
 	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;
 	}
 
@@ -652,18 +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)) {
@@ -707,6 +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;
 
@@ -947,6 +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);
 }
 
@@ -1313,7 +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);
@@ -1340,7 +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) {
@@ -1369,8 +1378,14 @@ 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;
 
@@ -1463,7 +1478,7 @@ out_free_iclog:
 out_free_log:
 	kmem_free(log);
 out:
-	return ERR_PTR(-error);
+	return ERR_PTR(error);
 }	/* xlog_alloc_log */
 
 
@@ -1661,9 +1676,9 @@ xlog_bdstrat(
 
 	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);
+		xfs_buf_ioend(bp);
 		/*
 		 * It would seem logical to return EIO here, but we rely on
 		 * the log state machine to propagate I/O errors instead of
@@ -1673,7 +1688,7 @@ xlog_bdstrat(
 		return 0;
 	}
 
-	xfs_buf_iorequest(bp);
+	xfs_buf_submit(bp);
 	return 0;
 }
 
@@ -2360,7 +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));
@@ -2859,7 +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;
@@ -3047,7 +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))
@@ -3055,7 +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);
@@ -3172,7 +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
@@ -3210,7 +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;
@@ -3246,7 +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);
@@ -3256,7 +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 {
@@ -3324,7 +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 {
@@ -3375,7 +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);
@@ -3390,7 +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);
@@ -3400,7 +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;
@@ -3852,18 +3867,17 @@ xlog_state_ioerror(
  * This is called from xfs_force_shutdown, when we're forcibly
  * shutting down the filesystem, typically because of an IO error.
  * Our main objectives here are to make sure that:
- *	a. the filesystem gets marked 'SHUTDOWN' for all interested
+ *	a. if !logerror, flush the logs to disk. Anything modified
+ *	   after this is ignored.
+ *	b. the filesystem gets marked 'SHUTDOWN' for all interested
  *	   parties to find out, 'atomically'.
- *	b. those who're sleeping on log reservations, pinned objects and
+ *	c. those who're sleeping on log reservations, pinned objects and
  *	    other resources get woken up, and be told the bad news.
- *	c. nothing new gets queued up after (a) and (b) are done.
- *	d. if !logerror, flush the iclogs to disk, then seal them off
- *	   for business.
+ *	d. nothing new gets queued up after (b) and (c) are done.
  *
- * Note: for delayed logging the !logerror case needs to flush the regions
- * held in memory out to the iclogs before flushing them to disk. This needs
- * to be done before the log is marked as shutdown, otherwise the flush to the
- * iclogs will fail.
+ * Note: for the !logerror case we need to flush the regions held in memory out
+ * to disk first. This needs to be done before the log is marked as shutdown,
+ * otherwise the iclog writes will fail.
  */
 int
 xfs_log_force_umount(
@@ -3895,16 +3909,16 @@ xfs_log_force_umount(
 		ASSERT(XLOG_FORCED_SHUTDOWN(log));
 		return 1;
 	}
-	retval = 0;
 
 	/*
-	 * Flush the in memory commit item list before marking the log as
-	 * being shut down. We need to do it in this order to ensure all the
-	 * completed transactions are flushed to disk with the xfs_log_force()
-	 * call below.
+	 * Flush all the completed transactions to disk before marking the log
+	 * being shut down. We need to do it in this order to ensure that
+	 * completed operations are safely on disk before we shut down, and that
+	 * we don't have to issue any buffer IO after the shutdown flags are set
+	 * to guarantee this.
 	 */
 	if (!logerror)
-		xlog_cil_force(log);
+		_xfs_log_force(mp, XFS_LOG_SYNC, NULL);
 
 	/*
 	 * mark the filesystem and the as in a shutdown state and wake
@@ -3916,18 +3930,11 @@ xfs_log_force_umount(
 		XFS_BUF_DONE(mp->m_sb_bp);
 
 	/*
-	 * This flag is sort of redundant because of the mount flag, but
-	 * it's good to maintain the separation between the log and the rest
-	 * of XFS.
+	 * Mark the log and the iclogs with IO error flags to prevent any
+	 * further log IO from being issued or completed.
 	 */
 	log->l_flags |= XLOG_IO_ERROR;
-
-	/*
-	 * If we hit a log error, we want to mark all the iclogs IOERROR
-	 * while we're still holding the loglock.
-	 */
-	if (logerror)
-		retval = xlog_state_ioerror(log);
+	retval = xlog_state_ioerror(log);
 	spin_unlock(&log->l_icloglock);
 
 	/*
@@ -3940,19 +3947,6 @@ xfs_log_force_umount(
 	xlog_grant_head_wake_all(&log->l_reserve_head);
 	xlog_grant_head_wake_all(&log->l_write_head);
 
-	if (!(log->l_iclog->ic_state & XLOG_STATE_IOERROR)) {
-		ASSERT(!logerror);
-		/*
-		 * Force the incore logs to disk before shutting the
-		 * log down completely.
-		 */
-		_xfs_log_force(mp, XFS_LOG_SYNC, NULL);
-
-		spin_lock(&log->l_icloglock);
-		retval = xlog_state_ioerror(log);
-		spin_unlock(&log->l_icloglock);
-	}
-
 	/*
 	 * 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
diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c
index b3425b34e3d5..f506c457011e 100644
--- a/fs/xfs/xfs_log_cil.c
+++ b/fs/xfs/xfs_log_cil.c
@@ -78,8 +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);
 }
 
 /*
@@ -465,12 +463,40 @@ xlog_cil_push(
 		spin_unlock(&cil->xc_push_lock);
 		goto out_skip;
 	}
-	spin_unlock(&cil->xc_push_lock);
 
 
 	/* check for a previously pushed seqeunce */
-	if (push_seq < cil->xc_ctx->sequence)
+	if (push_seq < cil->xc_ctx->sequence) {
+		spin_unlock(&cil->xc_push_lock);
 		goto out_skip;
+	}
+
+	/*
+	 * We are now going to push this context, so add it to the committing
+	 * list before we do anything else. This ensures that anyone waiting on
+	 * this push can easily detect the difference between a "push in
+	 * progress" and "CIL is empty, nothing to do".
+	 *
+	 * IOWs, a wait loop can now check for:
+	 *	the current sequence not being found on the committing list;
+	 *	an empty CIL; and
+	 *	an unchanged sequence number
+	 * to detect a push that had nothing to do and therefore does not need
+	 * waiting on. If the CIL is not empty, we get put on the committing
+	 * list before emptying the CIL and bumping the sequence number. Hence
+	 * an empty CIL and an unchanged sequence number means we jumped out
+	 * above after doing nothing.
+	 *
+	 * Hence the waiter will either find the commit sequence on the
+	 * committing list or the sequence number will be unchanged and the CIL
+	 * still dirty. In that latter case, the push has not yet started, and
+	 * so the waiter will have to continue trying to check the CIL
+	 * committing list until it is found. In extreme cases of delay, the
+	 * sequence may fully commit between the attempts the wait makes to wait
+	 * on the commit sequence.
+	 */
+	list_add(&ctx->committing, &cil->xc_committing);
+	spin_unlock(&cil->xc_push_lock);
 
 	/*
 	 * pull all the log vectors off the items in the CIL, and
@@ -534,7 +560,6 @@ xlog_cil_push(
 	 */
 	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);
 
@@ -634,7 +659,7 @@ 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
@@ -857,13 +882,15 @@ restart:
 	 * 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.
+	 * dirty objects. This is guaranteed by the push code first adding the
+	 * context to the committing list before emptying the CIL.
 	 *
-	 * 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.
+	 * Hence if we don't find the context in the committing list and the
+	 * current sequence number is unchanged then the CIL contents are
+	 * significant.  If the CIL is empty, if means there was nothing to push
+	 * and that means there is nothing to wait for. If the CIL is not empty,
+	 * it means we haven't yet started the push, because if it had started
+	 * we would have found the context on the committing list.
 	 */
 	if (sequence == cil->xc_current_sequence &&
 	    !list_empty(&cil->xc_cil)) {
@@ -928,12 +955,12 @@ 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 981af0f6504b..00cd7f3a8f59 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);
@@ -193,12 +193,8 @@ xlog_bread_noalign(
 	bp->b_io_length = nbblks;
 	bp->b_error = 0;
 
-	if (XFS_FORCED_SHUTDOWN(log->l_mp))
-		return XFS_ERROR(EIO);
-
-	xfs_buf_iorequest(bp);
-	error = xfs_buf_iowait(bp);
-	if (error)
+	error = xfs_buf_submit_wait(bp);
+	if (error && !XFS_FORCED_SHUTDOWN(log->l_mp))
 		xfs_buf_ioerror_alert(bp, __func__);
 	return error;
 }
@@ -268,7 +264,7 @@ 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);
@@ -330,14 +326,14 @@ 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;
 }
@@ -364,7 +360,7 @@ 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;
 }
@@ -378,12 +374,14 @@ xlog_recover_iodone(
 		 * We're not going to bother about retrying
 		 * this during recovery. One strike!
 		 */
-		xfs_buf_ioerror_alert(bp, __func__);
-		xfs_force_shutdown(bp->b_target->bt_mount,
-					SHUTDOWN_META_IO_ERROR);
+		if (!XFS_FORCED_SHUTDOWN(bp->b_target->bt_mount)) {
+			xfs_buf_ioerror_alert(bp, __func__);
+			xfs_force_shutdown(bp->b_target->bt_mount,
+						SHUTDOWN_META_IO_ERROR);
+		}
 	}
 	bp->b_iodone = NULL;
-	xfs_buf_ioend(bp, 0);
+	xfs_buf_ioend(bp);
 }
 
 /*
@@ -462,7 +460,7 @@ 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) {
@@ -524,7 +522,7 @@ 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);
@@ -539,7 +537,7 @@ 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;
 		}
 
@@ -564,7 +562,7 @@ 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;
 	}
 
@@ -628,7 +626,12 @@ 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? */
@@ -641,15 +644,12 @@ 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)
@@ -818,29 +818,29 @@ 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;
@@ -911,7 +911,7 @@ 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)
@@ -961,7 +961,7 @@ 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 */
@@ -1092,8 +1092,8 @@ 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(
@@ -1112,7 +1112,7 @@ 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;
@@ -1121,7 +1121,7 @@ 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 */
@@ -1141,7 +1141,7 @@ 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;
 	}
 
@@ -1179,19 +1179,18 @@ 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;
 }
 
 /*
@@ -1251,7 +1250,7 @@ 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
@@ -1354,7 +1353,7 @@ 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 {
@@ -1366,7 +1365,7 @@ 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;
 	}
@@ -1444,160 +1443,6 @@ xlog_clear_stale_blocks(
  ******************************************************************************
  */
 
-STATIC xlog_recover_t *
-xlog_recover_find_tid(
-	struct hlist_head	*head,
-	xlog_tid_t		tid)
-{
-	xlog_recover_t		*trans;
-
-	hlist_for_each_entry(trans, head, r_list) {
-		if (trans->r_log_tid == tid)
-			return trans;
-	}
-	return NULL;
-}
-
-STATIC void
-xlog_recover_new_tid(
-	struct hlist_head	*head,
-	xlog_tid_t		tid,
-	xfs_lsn_t		lsn)
-{
-	xlog_recover_t		*trans;
-
-	trans = kmem_zalloc(sizeof(xlog_recover_t), KM_SLEEP);
-	trans->r_log_tid   = tid;
-	trans->r_lsn	   = lsn;
-	INIT_LIST_HEAD(&trans->r_itemq);
-
-	INIT_HLIST_NODE(&trans->r_list);
-	hlist_add_head(&trans->r_list, head);
-}
-
-STATIC void
-xlog_recover_add_item(
-	struct list_head	*head)
-{
-	xlog_recover_item_t	*item;
-
-	item = kmem_zalloc(sizeof(xlog_recover_item_t), KM_SLEEP);
-	INIT_LIST_HEAD(&item->ri_list);
-	list_add_tail(&item->ri_list, head);
-}
-
-STATIC int
-xlog_recover_add_to_cont_trans(
-	struct xlog		*log,
-	struct xlog_recover	*trans,
-	xfs_caddr_t		dp,
-	int			len)
-{
-	xlog_recover_item_t	*item;
-	xfs_caddr_t		ptr, old_ptr;
-	int			old_len;
-
-	if (list_empty(&trans->r_itemq)) {
-		/* finish copying rest of trans header */
-		xlog_recover_add_item(&trans->r_itemq);
-		ptr = (xfs_caddr_t) &trans->r_theader +
-				sizeof(xfs_trans_header_t) - len;
-		memcpy(ptr, dp, len); /* d, s, l */
-		return 0;
-	}
-	/* take the tail entry */
-	item = list_entry(trans->r_itemq.prev, xlog_recover_item_t, ri_list);
-
-	old_ptr = item->ri_buf[item->ri_cnt-1].i_addr;
-	old_len = item->ri_buf[item->ri_cnt-1].i_len;
-
-	ptr = kmem_realloc(old_ptr, len+old_len, old_len, KM_SLEEP);
-	memcpy(&ptr[old_len], dp, len); /* d, s, l */
-	item->ri_buf[item->ri_cnt-1].i_len += len;
-	item->ri_buf[item->ri_cnt-1].i_addr = ptr;
-	trace_xfs_log_recover_item_add_cont(log, trans, item, 0);
-	return 0;
-}
-
-/*
- * The next region to add is the start of a new region.  It could be
- * a whole region or it could be the first part of a new region.  Because
- * of this, the assumption here is that the type and size fields of all
- * format structures fit into the first 32 bits of the structure.
- *
- * This works because all regions must be 32 bit aligned.  Therefore, we
- * either have both fields or we have neither field.  In the case we have
- * neither field, the data part of the region is zero length.  We only have
- * a log_op_header and can throw away the header since a new one will appear
- * later.  If we have at least 4 bytes, then we can determine how many regions
- * will appear in the current log item.
- */
-STATIC int
-xlog_recover_add_to_trans(
-	struct xlog		*log,
-	struct xlog_recover	*trans,
-	xfs_caddr_t		dp,
-	int			len)
-{
-	xfs_inode_log_format_t	*in_f;			/* any will do */
-	xlog_recover_item_t	*item;
-	xfs_caddr_t		ptr;
-
-	if (!len)
-		return 0;
-	if (list_empty(&trans->r_itemq)) {
-		/* we need to catch log corruptions here */
-		if (*(uint *)dp != XFS_TRANS_HEADER_MAGIC) {
-			xfs_warn(log->l_mp, "%s: bad header magic number",
-				__func__);
-			ASSERT(0);
-			return XFS_ERROR(EIO);
-		}
-		if (len == sizeof(xfs_trans_header_t))
-			xlog_recover_add_item(&trans->r_itemq);
-		memcpy(&trans->r_theader, dp, len); /* d, s, l */
-		return 0;
-	}
-
-	ptr = kmem_alloc(len, KM_SLEEP);
-	memcpy(ptr, dp, len);
-	in_f = (xfs_inode_log_format_t *)ptr;
-
-	/* take the tail entry */
-	item = list_entry(trans->r_itemq.prev, xlog_recover_item_t, ri_list);
-	if (item->ri_total != 0 &&
-	     item->ri_total == item->ri_cnt) {
-		/* tail item is in use, get a new one */
-		xlog_recover_add_item(&trans->r_itemq);
-		item = list_entry(trans->r_itemq.prev,
-					xlog_recover_item_t, ri_list);
-	}
-
-	if (item->ri_total == 0) {		/* first region to be added */
-		if (in_f->ilf_size == 0 ||
-		    in_f->ilf_size > XLOG_MAX_REGIONS_IN_ITEM) {
-			xfs_warn(log->l_mp,
-		"bad number of regions (%d) in inode log format",
-				  in_f->ilf_size);
-			ASSERT(0);
-			kmem_free(ptr);
-			return XFS_ERROR(EIO);
-		}
-
-		item->ri_total = in_f->ilf_size;
-		item->ri_buf =
-			kmem_zalloc(item->ri_total * sizeof(xfs_log_iovec_t),
-				    KM_SLEEP);
-	}
-	ASSERT(item->ri_total > item->ri_cnt);
-	/* Description region is ri_buf[0] */
-	item->ri_buf[item->ri_cnt].i_addr = ptr;
-	item->ri_buf[item->ri_cnt].i_len  = len;
-	item->ri_cnt++;
-	trace_xfs_log_recover_item_add(log, trans, item, 0);
-	return 0;
-}
-
 /*
  * Sort the log items in the transaction.
  *
@@ -1702,7 +1547,7 @@ xlog_recover_reorder_trans(
 			 */
 			if (!list_empty(&sort_list))
 				list_splice_init(&sort_list, &trans->r_itemq);
-			error = XFS_ERROR(EIO);
+			error = -EIO;
 			goto out;
 		}
 	}
@@ -1943,7 +1788,7 @@ 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,
@@ -2125,6 +1970,17 @@ 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);
@@ -2162,8 +2018,6 @@ 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);
@@ -2196,10 +2050,6 @@ 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);
@@ -2279,8 +2129,6 @@ 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);
@@ -2387,16 +2235,7 @@ 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);
 }
 
 /*
@@ -2404,8 +2243,11 @@ 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,
@@ -2420,9 +2262,8 @@ 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)
@@ -2435,9 +2276,10 @@ 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;
 }
 
 /*
@@ -2496,7 +2338,7 @@ 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)");
@@ -2504,23 +2346,44 @@ 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
@@ -2598,7 +2461,7 @@ 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);
@@ -2676,7 +2539,7 @@ 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;
@@ -2697,7 +2560,7 @@ 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;
@@ -2707,7 +2570,7 @@ 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;
 	}
 
@@ -2764,7 +2627,7 @@ 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))) {
@@ -2777,7 +2640,7 @@ 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;
 		}
 	}
@@ -2790,7 +2653,7 @@ 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)) {
@@ -2800,7 +2663,7 @@ 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);
@@ -2810,7 +2673,7 @@ 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;
 	}
 
@@ -2898,7 +2761,7 @@ xlog_recover_inode_pass2(
 		default:
 			xfs_warn(log->l_mp, "%s: Invalid flag", __func__);
 			ASSERT(0);
-			error = EIO;
+			error = -EIO;
 			goto out_release;
 		}
 	}
@@ -2919,7 +2782,7 @@ out_release:
 error:
 	if (need_free)
 		kmem_free(in_f);
-	return XFS_ERROR(error);
+	return error;
 }
 
 /*
@@ -2946,7 +2809,7 @@ xlog_recover_quotaoff_pass1(
 	if (qoff_f->qf_flags & XFS_GQUOTA_ACCT)
 		log->l_quotaoffs_flag |= XFS_DQ_GROUP;
 
-	return (0);
+	return 0;
 }
 
 /*
@@ -2971,17 +2834,17 @@ 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;
 	}
 
 	/*
@@ -2990,7 +2853,7 @@ 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.
@@ -3007,12 +2870,19 @@ 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;
 
@@ -3020,18 +2890,6 @@ 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.
 	 */
@@ -3178,38 +3036,38 @@ 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 */
@@ -3218,7 +3076,7 @@ xlog_recover_do_icreate_pass2(
 	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;
 	}
 
 	/*
@@ -3240,31 +3098,6 @@ xlog_recover_do_icreate_pass2(
 	return 0;
 }
 
-/*
- * Free up any resources allocated by the transaction
- *
- * Remember that EFIs, EFDs, and IUNLINKs are handled later.
- */
-STATIC void
-xlog_recover_free_trans(
-	struct xlog_recover	*trans)
-{
-	xlog_recover_item_t	*item, *n;
-	int			i;
-
-	list_for_each_entry_safe(item, n, &trans->r_itemq, ri_list) {
-		/* Free the regions in the item. */
-		list_del(&item->ri_list);
-		for (i = 0; i < item->ri_cnt; i++)
-			kmem_free(item->ri_buf[i].i_addr);
-		/* Free the item itself */
-		kmem_free(item->ri_buf);
-		kmem_free(item);
-	}
-	/* Free the transaction recover structure */
-	kmem_free(trans);
-}
-
 STATIC void
 xlog_recover_buffer_ra_pass2(
 	struct xlog                     *log,
@@ -3389,7 +3222,7 @@ 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;
 	}
 }
 
@@ -3425,7 +3258,7 @@ 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;
 	}
 }
 
@@ -3514,22 +3347,309 @@ out:
 	if (!list_empty(&done_list))
 		list_splice_init(&done_list, &trans->r_itemq);
 
-	xlog_recover_free_trans(trans);
-
 	error2 = xfs_buf_delwri_submit(&buffer_list);
 	return error ? error : error2;
 }
 
+STATIC void
+xlog_recover_add_item(
+	struct list_head	*head)
+{
+	xlog_recover_item_t	*item;
+
+	item = kmem_zalloc(sizeof(xlog_recover_item_t), KM_SLEEP);
+	INIT_LIST_HEAD(&item->ri_list);
+	list_add_tail(&item->ri_list, head);
+}
+
 STATIC int
-xlog_recover_unmount_trans(
-	struct xlog		*log)
+xlog_recover_add_to_cont_trans(
+	struct xlog		*log,
+	struct xlog_recover	*trans,
+	xfs_caddr_t		dp,
+	int			len)
 {
-	/* Do nothing now */
-	xfs_warn(log->l_mp, "%s: Unmount LR", __func__);
+	xlog_recover_item_t	*item;
+	xfs_caddr_t		ptr, old_ptr;
+	int			old_len;
+
+	if (list_empty(&trans->r_itemq)) {
+		/* finish copying rest of trans header */
+		xlog_recover_add_item(&trans->r_itemq);
+		ptr = (xfs_caddr_t) &trans->r_theader +
+				sizeof(xfs_trans_header_t) - len;
+		memcpy(ptr, dp, len);
+		return 0;
+	}
+	/* take the tail entry */
+	item = list_entry(trans->r_itemq.prev, xlog_recover_item_t, ri_list);
+
+	old_ptr = item->ri_buf[item->ri_cnt-1].i_addr;
+	old_len = item->ri_buf[item->ri_cnt-1].i_len;
+
+	ptr = kmem_realloc(old_ptr, len+old_len, old_len, KM_SLEEP);
+	memcpy(&ptr[old_len], dp, len);
+	item->ri_buf[item->ri_cnt-1].i_len += len;
+	item->ri_buf[item->ri_cnt-1].i_addr = ptr;
+	trace_xfs_log_recover_item_add_cont(log, trans, item, 0);
+	return 0;
+}
+
+/*
+ * The next region to add is the start of a new region.  It could be
+ * a whole region or it could be the first part of a new region.  Because
+ * of this, the assumption here is that the type and size fields of all
+ * format structures fit into the first 32 bits of the structure.
+ *
+ * This works because all regions must be 32 bit aligned.  Therefore, we
+ * either have both fields or we have neither field.  In the case we have
+ * neither field, the data part of the region is zero length.  We only have
+ * a log_op_header and can throw away the header since a new one will appear
+ * later.  If we have at least 4 bytes, then we can determine how many regions
+ * will appear in the current log item.
+ */
+STATIC int
+xlog_recover_add_to_trans(
+	struct xlog		*log,
+	struct xlog_recover	*trans,
+	xfs_caddr_t		dp,
+	int			len)
+{
+	xfs_inode_log_format_t	*in_f;			/* any will do */
+	xlog_recover_item_t	*item;
+	xfs_caddr_t		ptr;
+
+	if (!len)
+		return 0;
+	if (list_empty(&trans->r_itemq)) {
+		/* we need to catch log corruptions here */
+		if (*(uint *)dp != XFS_TRANS_HEADER_MAGIC) {
+			xfs_warn(log->l_mp, "%s: bad header magic number",
+				__func__);
+			ASSERT(0);
+			return -EIO;
+		}
+		if (len == sizeof(xfs_trans_header_t))
+			xlog_recover_add_item(&trans->r_itemq);
+		memcpy(&trans->r_theader, dp, len);
+		return 0;
+	}
+
+	ptr = kmem_alloc(len, KM_SLEEP);
+	memcpy(ptr, dp, len);
+	in_f = (xfs_inode_log_format_t *)ptr;
+
+	/* take the tail entry */
+	item = list_entry(trans->r_itemq.prev, xlog_recover_item_t, ri_list);
+	if (item->ri_total != 0 &&
+	     item->ri_total == item->ri_cnt) {
+		/* tail item is in use, get a new one */
+		xlog_recover_add_item(&trans->r_itemq);
+		item = list_entry(trans->r_itemq.prev,
+					xlog_recover_item_t, ri_list);
+	}
+
+	if (item->ri_total == 0) {		/* first region to be added */
+		if (in_f->ilf_size == 0 ||
+		    in_f->ilf_size > XLOG_MAX_REGIONS_IN_ITEM) {
+			xfs_warn(log->l_mp,
+		"bad number of regions (%d) in inode log format",
+				  in_f->ilf_size);
+			ASSERT(0);
+			kmem_free(ptr);
+			return -EIO;
+		}
+
+		item->ri_total = in_f->ilf_size;
+		item->ri_buf =
+			kmem_zalloc(item->ri_total * sizeof(xfs_log_iovec_t),
+				    KM_SLEEP);
+	}
+	ASSERT(item->ri_total > item->ri_cnt);
+	/* Description region is ri_buf[0] */
+	item->ri_buf[item->ri_cnt].i_addr = ptr;
+	item->ri_buf[item->ri_cnt].i_len  = len;
+	item->ri_cnt++;
+	trace_xfs_log_recover_item_add(log, trans, item, 0);
 	return 0;
 }
 
 /*
+ * Free up any resources allocated by the transaction
+ *
+ * Remember that EFIs, EFDs, and IUNLINKs are handled later.
+ */
+STATIC void
+xlog_recover_free_trans(
+	struct xlog_recover	*trans)
+{
+	xlog_recover_item_t	*item, *n;
+	int			i;
+
+	list_for_each_entry_safe(item, n, &trans->r_itemq, ri_list) {
+		/* Free the regions in the item. */
+		list_del(&item->ri_list);
+		for (i = 0; i < item->ri_cnt; i++)
+			kmem_free(item->ri_buf[i].i_addr);
+		/* Free the item itself */
+		kmem_free(item->ri_buf);
+		kmem_free(item);
+	}
+	/* Free the transaction recover structure */
+	kmem_free(trans);
+}
+
+/*
+ * On error or completion, trans is freed.
+ */
+STATIC int
+xlog_recovery_process_trans(
+	struct xlog		*log,
+	struct xlog_recover	*trans,
+	xfs_caddr_t		dp,
+	unsigned int		len,
+	unsigned int		flags,
+	int			pass)
+{
+	int			error = 0;
+	bool			freeit = false;
+
+	/* mask off ophdr transaction container flags */
+	flags &= ~XLOG_END_TRANS;
+	if (flags & XLOG_WAS_CONT_TRANS)
+		flags &= ~XLOG_CONTINUE_TRANS;
+
+	/*
+	 * Callees must not free the trans structure. We'll decide if we need to
+	 * free it or not based on the operation being done and it's result.
+	 */
+	switch (flags) {
+	/* expected flag values */
+	case 0:
+	case XLOG_CONTINUE_TRANS:
+		error = xlog_recover_add_to_trans(log, trans, dp, len);
+		break;
+	case XLOG_WAS_CONT_TRANS:
+		error = xlog_recover_add_to_cont_trans(log, trans, dp, len);
+		break;
+	case XLOG_COMMIT_TRANS:
+		error = xlog_recover_commit_trans(log, trans, pass);
+		/* success or fail, we are now done with this transaction. */
+		freeit = true;
+		break;
+
+	/* unexpected flag values */
+	case XLOG_UNMOUNT_TRANS:
+		/* just skip trans */
+		xfs_warn(log->l_mp, "%s: Unmount LR", __func__);
+		freeit = true;
+		break;
+	case XLOG_START_TRANS:
+	default:
+		xfs_warn(log->l_mp, "%s: bad flag 0x%x", __func__, flags);
+		ASSERT(0);
+		error = -EIO;
+		break;
+	}
+	if (error || freeit)
+		xlog_recover_free_trans(trans);
+	return error;
+}
+
+/*
+ * Lookup the transaction recovery structure associated with the ID in the
+ * current ophdr. If the transaction doesn't exist and the start flag is set in
+ * the ophdr, then allocate a new transaction for future ID matches to find.
+ * Either way, return what we found during the lookup - an existing transaction
+ * or nothing.
+ */
+STATIC struct xlog_recover *
+xlog_recover_ophdr_to_trans(
+	struct hlist_head	rhash[],
+	struct xlog_rec_header	*rhead,
+	struct xlog_op_header	*ohead)
+{
+	struct xlog_recover	*trans;
+	xlog_tid_t		tid;
+	struct hlist_head	*rhp;
+
+	tid = be32_to_cpu(ohead->oh_tid);
+	rhp = &rhash[XLOG_RHASH(tid)];
+	hlist_for_each_entry(trans, rhp, r_list) {
+		if (trans->r_log_tid == tid)
+			return trans;
+	}
+
+	/*
+	 * skip over non-start transaction headers - we could be
+	 * processing slack space before the next transaction starts
+	 */
+	if (!(ohead->oh_flags & XLOG_START_TRANS))
+		return NULL;
+
+	ASSERT(be32_to_cpu(ohead->oh_len) == 0);
+
+	/*
+	 * This is a new transaction so allocate a new recovery container to
+	 * hold the recovery ops that will follow.
+	 */
+	trans = kmem_zalloc(sizeof(struct xlog_recover), KM_SLEEP);
+	trans->r_log_tid = tid;
+	trans->r_lsn = be64_to_cpu(rhead->h_lsn);
+	INIT_LIST_HEAD(&trans->r_itemq);
+	INIT_HLIST_NODE(&trans->r_list);
+	hlist_add_head(&trans->r_list, rhp);
+
+	/*
+	 * Nothing more to do for this ophdr. Items to be added to this new
+	 * transaction will be in subsequent ophdr containers.
+	 */
+	return NULL;
+}
+
+STATIC int
+xlog_recover_process_ophdr(
+	struct xlog		*log,
+	struct hlist_head	rhash[],
+	struct xlog_rec_header	*rhead,
+	struct xlog_op_header	*ohead,
+	xfs_caddr_t		dp,
+	xfs_caddr_t		end,
+	int			pass)
+{
+	struct xlog_recover	*trans;
+	unsigned int		len;
+
+	/* Do we understand who wrote this op? */
+	if (ohead->oh_clientid != XFS_TRANSACTION &&
+	    ohead->oh_clientid != XFS_LOG) {
+		xfs_warn(log->l_mp, "%s: bad clientid 0x%x",
+			__func__, ohead->oh_clientid);
+		ASSERT(0);
+		return -EIO;
+	}
+
+	/*
+	 * Check the ophdr contains all the data it is supposed to contain.
+	 */
+	len = be32_to_cpu(ohead->oh_len);
+	if (dp + len > end) {
+		xfs_warn(log->l_mp, "%s: bad length 0x%x", __func__, len);
+		WARN_ON(1);
+		return -EIO;
+	}
+
+	trans = xlog_recover_ophdr_to_trans(rhash, rhead, ohead);
+	if (!trans) {
+		/* nothing to do, so skip over this ophdr */
+		return 0;
+	}
+
+	return xlog_recovery_process_trans(log, trans, dp, len,
+					   ohead->oh_flags, pass);
+}
+
+/*
  * There are two valid states of the r_state field.  0 indicates that the
  * transaction structure is in a normal state.  We have either seen the
  * start of the transaction or the last operation we added was not a partial
@@ -3546,86 +3666,30 @@ xlog_recover_process_data(
 	xfs_caddr_t		dp,
 	int			pass)
 {
-	xfs_caddr_t		lp;
+	struct xlog_op_header	*ohead;
+	xfs_caddr_t		end;
 	int			num_logops;
-	xlog_op_header_t	*ohead;
-	xlog_recover_t		*trans;
-	xlog_tid_t		tid;
 	int			error;
-	unsigned long		hash;
-	uint			flags;
 
-	lp = dp + be32_to_cpu(rhead->h_len);
+	end = dp + be32_to_cpu(rhead->h_len);
 	num_logops = be32_to_cpu(rhead->h_num_logops);
 
 	/* 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));
-
-	while ((dp < lp) && num_logops) {
-		ASSERT(dp + sizeof(xlog_op_header_t) <= lp);
-		ohead = (xlog_op_header_t *)dp;
-		dp += sizeof(xlog_op_header_t);
-		if (ohead->oh_clientid != XFS_TRANSACTION &&
-		    ohead->oh_clientid != XFS_LOG) {
-			xfs_warn(log->l_mp, "%s: bad clientid 0x%x",
-					__func__, ohead->oh_clientid);
-			ASSERT(0);
-			return (XFS_ERROR(EIO));
-		}
-		tid = be32_to_cpu(ohead->oh_tid);
-		hash = XLOG_RHASH(tid);
-		trans = xlog_recover_find_tid(&rhash[hash], tid);
-		if (trans == NULL) {		   /* not found; add new tid */
-			if (ohead->oh_flags & XLOG_START_TRANS)
-				xlog_recover_new_tid(&rhash[hash], tid,
-					be64_to_cpu(rhead->h_lsn));
-		} else {
-			if (dp + be32_to_cpu(ohead->oh_len) > lp) {
-				xfs_warn(log->l_mp, "%s: bad length 0x%x",
-					__func__, be32_to_cpu(ohead->oh_len));
-				WARN_ON(1);
-				return (XFS_ERROR(EIO));
-			}
-			flags = ohead->oh_flags & ~XLOG_END_TRANS;
-			if (flags & XLOG_WAS_CONT_TRANS)
-				flags &= ~XLOG_CONTINUE_TRANS;
-			switch (flags) {
-			case XLOG_COMMIT_TRANS:
-				error = xlog_recover_commit_trans(log,
-								trans, pass);
-				break;
-			case XLOG_UNMOUNT_TRANS:
-				error = xlog_recover_unmount_trans(log);
-				break;
-			case XLOG_WAS_CONT_TRANS:
-				error = xlog_recover_add_to_cont_trans(log,
-						trans, dp,
-						be32_to_cpu(ohead->oh_len));
-				break;
-			case XLOG_START_TRANS:
-				xfs_warn(log->l_mp, "%s: bad transaction",
-					__func__);
-				ASSERT(0);
-				error = XFS_ERROR(EIO);
-				break;
-			case 0:
-			case XLOG_CONTINUE_TRANS:
-				error = xlog_recover_add_to_trans(log, trans,
-						dp, be32_to_cpu(ohead->oh_len));
-				break;
-			default:
-				xfs_warn(log->l_mp, "%s: bad flag 0x%x",
-					__func__, flags);
-				ASSERT(0);
-				error = XFS_ERROR(EIO);
-				break;
-			}
-			if (error) {
-				xlog_recover_free_trans(trans);
-				return error;
-			}
-		}
+		return -EIO;
+
+	while ((dp < end) && num_logops) {
+
+		ohead = (struct xlog_op_header *)dp;
+		dp += sizeof(*ohead);
+		ASSERT(dp <= end);
+
+		/* errors will abort recovery */
+		error = xlog_recover_process_ophdr(log, rhash, rhead, ohead,
+						    dp, end, pass);
+		if (error)
+			return error;
+
 		dp += be32_to_cpu(ohead->oh_len);
 		num_logops--;
 	}
@@ -3669,7 +3733,7 @@ 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;
 		}
 	}
 
@@ -3969,7 +4033,7 @@ 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;
@@ -4018,14 +4082,14 @@ 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 */
@@ -4033,12 +4097,12 @@ 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;
 }
@@ -4081,7 +4145,7 @@ 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)
@@ -4110,49 +4174,21 @@ 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));
-	if (tail_blk <= head_blk) {
-		for (blk_no = tail_blk; blk_no < head_blk; ) {
-			error = xlog_bread(log, blk_no, hblks, hbp, &offset);
-			if (error)
-				goto bread_err2;
-
-			rhead = (xlog_rec_header_t *)offset;
-			error = xlog_valid_rec_header(log, rhead, blk_no);
-			if (error)
-				goto bread_err2;
-
-			/* blocks in data section */
-			bblks = (int)BTOBB(be32_to_cpu(rhead->h_len));
-			error = xlog_bread(log, blk_no + hblks, bblks, dbp,
-					   &offset);
-			if (error)
-				goto bread_err2;
-
-			error = xlog_unpack_data(rhead, offset, log);
-			if (error)
-				goto bread_err2;
-
-			error = xlog_recover_process_data(log,
-						rhash, rhead, offset, pass);
-			if (error)
-				goto bread_err2;
-			blk_no += bblks + hblks;
-		}
-	} else {
+	blk_no = tail_blk;
+	if (tail_blk > head_blk) {
 		/*
 		 * Perform recovery around the end of the physical log.
 		 * When the head is not on the same cycle number as the tail,
-		 * we can't do a sequential recovery as above.
+		 * we can't do a sequential recovery.
 		 */
-		blk_no = tail_blk;
 		while (blk_no < log->l_logBBsize) {
 			/*
 			 * Check for header wrapping around physical end-of-log
@@ -4266,34 +4302,35 @@ xlog_do_recovery_pass(
 
 		ASSERT(blk_no >= log->l_logBBsize);
 		blk_no -= log->l_logBBsize;
+	}
 
-		/* read first part of physical log */
-		while (blk_no < head_blk) {
-			error = xlog_bread(log, blk_no, hblks, hbp, &offset);
-			if (error)
-				goto bread_err2;
+	/* read first part of physical log */
+	while (blk_no < head_blk) {
+		error = xlog_bread(log, blk_no, hblks, hbp, &offset);
+		if (error)
+			goto bread_err2;
 
-			rhead = (xlog_rec_header_t *)offset;
-			error = xlog_valid_rec_header(log, rhead, blk_no);
-			if (error)
-				goto bread_err2;
+		rhead = (xlog_rec_header_t *)offset;
+		error = xlog_valid_rec_header(log, rhead, blk_no);
+		if (error)
+			goto bread_err2;
 
-			bblks = (int)BTOBB(be32_to_cpu(rhead->h_len));
-			error = xlog_bread(log, blk_no+hblks, bblks, dbp,
-					   &offset);
-			if (error)
-				goto bread_err2;
+		/* blocks in data section */
+		bblks = (int)BTOBB(be32_to_cpu(rhead->h_len));
+		error = xlog_bread(log, blk_no+hblks, bblks, dbp,
+				   &offset);
+		if (error)
+			goto bread_err2;
 
-			error = xlog_unpack_data(rhead, offset, log);
-			if (error)
-				goto bread_err2;
+		error = xlog_unpack_data(rhead, offset, log);
+		if (error)
+			goto bread_err2;
 
-			error = xlog_recover_process_data(log, rhash,
-							rhead, offset, pass);
-			if (error)
-				goto bread_err2;
-			blk_no += bblks + hblks;
-		}
+		error = xlog_recover_process_data(log, rhash,
+						rhead, offset, pass);
+		if (error)
+			goto bread_err2;
+		blk_no += bblks + hblks;
 	}
 
  bread_err2:
@@ -4388,7 +4425,7 @@ xlog_do_recover(
 	 * If IO errors happened during recovery, bail out.
 	 */
 	if (XFS_FORCED_SHUTDOWN(log->l_mp)) {
-		return (EIO);
+		return -EIO;
 	}
 
 	/*
@@ -4413,16 +4450,12 @@ xlog_do_recover(
 	XFS_BUF_UNASYNC(bp);
 	bp->b_ops = &xfs_sb_buf_ops;
 
-	if (XFS_FORCED_SHUTDOWN(log->l_mp)) {
-		xfs_buf_relse(bp);
-		return XFS_ERROR(EIO);
-	}
-
-	xfs_buf_iorequest(bp);
-	error = xfs_buf_iowait(bp);
+	error = xfs_buf_submit_wait(bp);
 	if (error) {
-		xfs_buf_ioerror_alert(bp, __func__);
-		ASSERT(0);
+		if (!XFS_FORCED_SHUTDOWN(log->l_mp)) {
+			xfs_buf_ioerror_alert(bp, __func__);
+			ASSERT(0);
+		}
 		xfs_buf_relse(bp);
 		return error;
 	}
@@ -4492,7 +4525,19 @@ 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;
+		}
+
+		/*
+		 * Delay log recovery if the debug hook is set. This is debug
+		 * instrumention to coordinate simulation of I/O failures with
+		 * log recovery.
+		 */
+		if (xfs_globals.log_recovery_delay) {
+			xfs_notice(log->l_mp,
+				"Delaying log recovery for %d seconds.",
+				xfs_globals.log_recovery_delay);
+			msleep(xfs_globals.log_recovery_delay * 1000);
 		}
 
 		xfs_notice(log->l_mp, "Starting recovery (logdev: %s)",
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index 3507cd0ec400..51435dbce9c4 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
@@ -76,7 +77,7 @@ 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 +105,7 @@ 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 +174,9 @@ 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 +247,9 @@ 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;
@@ -303,28 +300,21 @@ xfs_readsb(
 	 * access to the superblock.
 	 */
 reread:
-	bp = xfs_buf_read_uncached(mp->m_ddev_targp, XFS_SB_DADDR,
-				   BTOBB(sector_size), 0, buf_ops);
-	if (!bp) {
-		if (loud)
-			xfs_warn(mp, "SB buffer read failed");
-		return EIO;
-	}
-	if (bp->b_error) {
-		error = bp->b_error;
+	error = xfs_buf_read_uncached(mp->m_ddev_targp, XFS_SB_DADDR,
+				   BTOBB(sector_size), 0, &bp, buf_ops);
+	if (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;
+		if (error == -EFSBADCRC)
+			error = -EFSCORRUPTED;
+		return error;
 	}
 
 	/*
 	 * Initialize the mount structure from the superblock.
 	 */
 	xfs_sb_from_disk(sbp, XFS_BUF_TO_SBP(bp));
-	xfs_sb_quota_from_disk(sbp);
 
 	/*
 	 * If we haven't validated the superblock, do so now before we try
@@ -333,7 +323,7 @@ reread:
 	if (sbp->sb_magicnum != XFS_SB_MAGIC) {
 		if (loud)
 			xfs_warn(mp, "Invalid superblock magic number");
-		error = EINVAL;
+		error = -EINVAL;
 		goto release_buf;
 	}
 
@@ -344,7 +334,7 @@ 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;
 	}
 
@@ -392,7 +382,7 @@ 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.
@@ -402,14 +392,14 @@ 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;
 			}
 		}
 
@@ -429,7 +419,7 @@ 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)) {
@@ -548,40 +538,43 @@ xfs_set_inoalignment(xfs_mount_t *mp)
  * Check that the data (and log if separate) is an ok size.
  */
 STATIC int
-xfs_check_sizes(xfs_mount_t *mp)
+xfs_check_sizes(
+	struct xfs_mount *mp)
 {
-	xfs_buf_t	*bp;
+	struct xfs_buf	*bp;
 	xfs_daddr_t	d;
+	int		error;
 
 	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,
+	error = 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_FSS_TO_BB(mp, 1), 0, &bp, NULL);
+	if (error) {
 		xfs_warn(mp, "last sector read failed");
-		return EIO;
+		return error;
 	}
 	xfs_buf_relse(bp);
 
-	if (mp->m_logdev_targp != mp->m_ddev_targp) {
-		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);
-		}
-		bp = xfs_buf_read_uncached(mp->m_logdev_targp,
+	if (mp->m_logdev_targp == mp->m_ddev_targp)
+		return 0;
+
+	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 -EFBIG;
+	}
+	error = 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;
-		}
-		xfs_buf_relse(bp);
+					XFS_FSB_TO_BB(mp, 1), 0, &bp, NULL);
+	if (error) {
+		xfs_warn(mp, "log device read failed");
+		return error;
 	}
+	xfs_buf_relse(bp);
 	return 0;
 }
 
@@ -731,10 +724,14 @@ xfs_mountfs(
 
 	xfs_set_maxicount(mp);
 
-	error = xfs_uuid_mount(mp);
+	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
 	 */
@@ -816,7 +813,7 @@ xfs_mountfs(
 	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;
 	}
 
@@ -855,7 +852,7 @@ 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;
 	}
 
 	/*
@@ -876,7 +873,7 @@ 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 */
@@ -927,7 +924,7 @@ xfs_mountfs(
 			xfs_notice(mp, "resetting quota flags");
 			error = xfs_mount_reset_sbqflags(mp);
 			if (error)
-				return error;
+				goto out_rtunmount;
 		}
 	}
 
@@ -989,6 +986,8 @@ xfs_mountfs(
 	xfs_da_unmount(mp);
  out_remove_uuid:
 	xfs_uuid_unmount(mp);
+ out_remove_sysfs:
+	xfs_sysfs_del(&mp->m_kobj);
  out:
 	return error;
 }
@@ -1071,6 +1070,8 @@ xfs_unmountfs(
 	xfs_errortag_clearall(mp, 0);
 #endif
 	xfs_free_perag(mp);
+
+	xfs_sysfs_del(&mp->m_kobj);
 }
 
 int
@@ -1152,7 +1153,7 @@ 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;
@@ -1161,7 +1162,7 @@ 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;
@@ -1191,7 +1192,7 @@ 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) {
@@ -1202,7 +1203,7 @@ 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);
@@ -1211,7 +1212,7 @@ 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;
@@ -1220,7 +1221,7 @@ 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;
@@ -1229,7 +1230,7 @@ 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;
@@ -1238,7 +1239,7 @@ 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;
@@ -1247,7 +1248,7 @@ 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;
@@ -1256,7 +1257,7 @@ 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;
@@ -1265,7 +1266,7 @@ 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;
@@ -1274,7 +1275,7 @@ 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;
@@ -1283,13 +1284,13 @@ 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;
 	}
 }
 
@@ -1452,7 +1453,7 @@ 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;
 }
@@ -1995,7 +1996,7 @@ 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 7295a0b7c343..b0447c86e7e2 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -166,6 +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 f99b4933dc22..30ecca3037e3 100644
--- a/fs/xfs/xfs_mru_cache.c
+++ b/fs/xfs/xfs_mru_cache.c
@@ -304,7 +304,8 @@ _xfs_mru_cache_reap(
 int
 xfs_mru_cache_init(void)
 {
-	xfs_mru_reap_wq = alloc_workqueue("xfs_mru_cache", WQ_MEM_RECLAIM, 1);
+	xfs_mru_reap_wq = alloc_workqueue("xfs_mru_cache",
+				WQ_MEM_RECLAIM|WQ_FREEZABLE, 1);
 	if (!xfs_mru_reap_wq)
 		return -ENOMEM;
 	return 0;
@@ -337,20 +338,20 @@ xfs_mru_cache_create(
 		*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;
 	}
 
@@ -434,16 +435,16 @@ xfs_mru_cache_insert(
 
 	ASSERT(mru && mru->lists);
 	if (!mru || !mru->lists)
-		return EINVAL;
+		return -EINVAL;
 
 	if (radix_tree_preload(GFP_KERNEL))
-		return ENOMEM;
+		return -ENOMEM;
 
 	INIT_LIST_HEAD(&elem->list_node);
 	elem->key = key;
 
 	spin_lock(&mru->lock);
-	error = -radix_tree_insert(&mru->store, key, elem);
+	error = radix_tree_insert(&mru->store, key, elem);
 	radix_tree_preload_end();
 	if (!error)
 		_xfs_mru_cache_list_insert(mru, elem);
diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c
index 6d26759c779a..d68f23021af3 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;
 		}
@@ -138,7 +138,7 @@ xfs_qm_dqpurge(
 	xfs_dqlock(dqp);
 	if ((dqp->dq_flags & XFS_DQ_FREEING) || dqp->q_nrefs != 0) {
 		xfs_dqunlock(dqp);
-		return EAGAIN;
+		return -EAGAIN;
 	}
 
 	dqp->dq_flags |= XFS_DQ_FREEING;
@@ -221,100 +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.
  */
@@ -528,6 +434,7 @@ xfs_qm_dquot_isolate(
 	struct list_head	*item,
 	spinlock_t		*lru_lock,
 	void			*arg)
+		__releases(lru_lock) __acquires(lru_lock)
 {
 	struct xfs_dquot	*dqp = container_of(item,
 						struct xfs_dquot, q_lru);
@@ -671,7 +578,7 @@ xfs_qm_init_quotainfo(
 
 	qinf = mp->m_quotainfo = kmem_zalloc(sizeof(xfs_quotainfo_t), KM_SLEEP);
 
-	error = -list_lru_init(&qinf->qi_lru);
+	error = list_lru_init(&qinf->qi_lru);
 	if (error)
 		goto out_free_qinf;
 
@@ -995,7 +902,7 @@ 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,
@@ -1005,6 +912,12 @@ 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);
@@ -1090,7 +1003,7 @@ 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++;
 				}
 			}
@@ -1138,8 +1051,8 @@ 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;
 	}
 
@@ -1226,7 +1139,7 @@ xfs_qm_dqusage_adjust(
 	 */
 	if (xfs_is_quota_inode(&mp->m_sb, ino)) {
 		*res = BULKSTAT_RV_NOTHING;
-		return XFS_ERROR(EINVAL);
+		return -EINVAL;
 	}
 
 	/*
@@ -1330,7 +1243,7 @@ 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)
 {
@@ -1463,7 +1376,100 @@ 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;
+	}
 }
 
 /*
@@ -1493,7 +1499,7 @@ 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) {
@@ -1563,7 +1569,7 @@ error_rele:
 		IRELE(gip);
 	if (pip)
 		IRELE(pip);
-	return XFS_ERROR(error);
+	return error;
 }
 
 STATIC void
@@ -1679,7 +1685,7 @@ xfs_qm_vop_dqalloc(
 						 XFS_QMOPT_DOWARN,
 						 &uq);
 			if (error) {
-				ASSERT(error != ENOENT);
+				ASSERT(error != -ENOENT);
 				return error;
 			}
 			/*
@@ -1706,7 +1712,7 @@ xfs_qm_vop_dqalloc(
 						 XFS_QMOPT_DOWARN,
 						 &gq);
 			if (error) {
-				ASSERT(error != ENOENT);
+				ASSERT(error != -ENOENT);
 				goto error_rele;
 			}
 			xfs_dqunlock(gq);
@@ -1726,7 +1732,7 @@ xfs_qm_vop_dqalloc(
 						 XFS_QMOPT_DOWARN,
 						 &pq);
 			if (error) {
-				ASSERT(error != ENOENT);
+				ASSERT(error != -ENOENT);
 				goto error_rele;
 			}
 			xfs_dqunlock(pq);
@@ -1895,7 +1901,7 @@ xfs_qm_vop_chown_reserve(
 				-((xfs_qcnt_t)delblks), 0, blkflags);
 	}
 
-	return (0);
+	return 0;
 }
 
 int
diff --git a/fs/xfs/xfs_qm.h b/fs/xfs/xfs_qm.h
index 797fd4636273..3a07a937e232 100644
--- a/fs/xfs/xfs_qm.h
+++ b/fs/xfs/xfs_qm.h
@@ -157,7 +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 bbc813caba4c..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;
 	}
 
 	/*
@@ -278,13 +278,13 @@ xfs_qm_scall_trunc_qfiles(
 	xfs_mount_t	*mp,
 	uint		flags)
 {
-	int		error = EINVAL;
+	int		error = -EINVAL;
 
 	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) {
@@ -328,7 +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 */
@@ -351,13 +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
@@ -372,11 +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.
 	 */
@@ -387,10 +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.
@@ -399,7 +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;
 }
 
 
@@ -426,7 +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 &
@@ -514,7 +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 &
@@ -595,7 +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;
 
@@ -615,7 +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);
@@ -758,7 +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,
@@ -772,7 +772,7 @@ xfs_qm_log_quotaoff_end(
 	 */
 	xfs_trans_set_sync(tp);
 	error = xfs_trans_commit(tp, 0);
-	return (error);
+	return error;
 }
 
 
@@ -822,7 +822,7 @@ error0:
 		spin_unlock(&mp->m_sb_lock);
 	}
 	*qoffstartp = qoffi;
-	return (error);
+	return error;
 }
 
 
@@ -850,7 +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;
 	}
 
@@ -953,7 +953,7 @@ xfs_qm_export_flags(
 		uflags |= FS_QUOTA_GDQ_ENFD;
 	if (flags & XFS_PQUOTA_ENFD)
 		uflags |= FS_QUOTA_PDQ_ENFD;
-	return (uflags);
+	return uflags;
 }
 
 
diff --git a/fs/xfs/xfs_quotaops.c b/fs/xfs/xfs_quotaops.c
index 2ad1b9822e92..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
@@ -95,11 +95,11 @@ 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);
+		return xfs_qm_scall_quotaoff(mp, flags);
 	}
 
 	return -EINVAL;
@@ -112,7 +112,7 @@ xfs_fs_rm_xquota(
 {
 	struct xfs_mount	*mp = XFS_M(sb);
 	unsigned int		flags = 0;
-	
+
 	if (sb->s_flags & MS_RDONLY)
 		return -EROFS;
 
@@ -123,11 +123,11 @@ xfs_fs_rm_xquota(
 		flags |= XFS_DQ_USER;
 	if (uflags & FS_GROUP_QUOTA)
 		flags |= XFS_DQ_GROUP;
-	if (uflags & FS_USER_QUOTA)
+	if (uflags & FS_PROJ_QUOTA)
 		flags |= XFS_DQ_PROJ;
 
-	return -xfs_qm_scall_trunc_qfiles(mp, flags);
-}	
+	return xfs_qm_scall_trunc_qfiles(mp, flags);
+}
 
 STATIC int
 xfs_fs_get_dqblk(
@@ -142,7 +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);
 }
 
@@ -161,7 +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);
 }
 
diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c
index ec5ca65c6211..e1175ea9b551 100644
--- a/fs/xfs/xfs_rtalloc.c
+++ b/fs/xfs/xfs_rtalloc.c
@@ -46,7 +46,7 @@
  * Keeps track of a current summary block, so we don't keep reading
  * it from the buffer cache.
  */
-STATIC int				/* error */
+static int
 xfs_rtget_summary(
 	xfs_mount_t	*mp,		/* file system mount structure */
 	xfs_trans_t	*tp,		/* transaction pointer */
@@ -56,60 +56,9 @@ xfs_rtget_summary(
 	xfs_fsblock_t	*rsb,		/* in/out: summary block number */
 	xfs_suminfo_t	*sum)		/* out: summary info for this block */
 {
-	xfs_buf_t	*bp;		/* buffer for summary block */
-	int		error;		/* error value */
-	xfs_fsblock_t	sb;		/* summary fsblock */
-	int		so;		/* index into the summary file */
-	xfs_suminfo_t	*sp;		/* pointer to returned data */
-
-	/*
-	 * Compute entry number in the summary file.
-	 */
-	so = XFS_SUMOFFS(mp, log, bbno);
-	/*
-	 * Compute the block number in the summary file.
-	 */
-	sb = XFS_SUMOFFSTOBLOCK(mp, so);
-	/*
-	 * If we have an old buffer, and the block number matches, use that.
-	 */
-	if (rbpp && *rbpp && *rsb == sb)
-		bp = *rbpp;
-	/*
-	 * Otherwise we have to get the buffer.
-	 */
-	else {
-		/*
-		 * If there was an old one, get rid of it first.
-		 */
-		if (rbpp && *rbpp)
-			xfs_trans_brelse(tp, *rbpp);
-		error = xfs_rtbuf_get(mp, tp, sb, 1, &bp);
-		if (error) {
-			return error;
-		}
-		/*
-		 * Remember this buffer and block for the next call.
-		 */
-		if (rbpp) {
-			*rbpp = bp;
-			*rsb = sb;
-		}
-	}
-	/*
-	 * Point to the summary information & copy it out.
-	 */
-	sp = XFS_SUMPTR(mp, bp, so);
-	*sum = *sp;
-	/*
-	 * Drop the buffer if we're not asked to remember it.
-	 */
-	if (!rbpp)
-		xfs_trans_brelse(tp, bp);
-	return 0;
+	return xfs_rtmodify_summary_int(mp, tp, log, bbno, 0, rbpp, rsb, sum);
 }
 
-
 /*
  * Return whether there are any free extents in the size range given
  * by low and high, for the bitmap block bbno.
@@ -863,7 +812,7 @@ 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 +852,7 @@ 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 +893,9 @@ 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 */
@@ -962,26 +911,21 @@ 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;
 	/*
 	 * Read in the last block of the device, make sure it exists.
 	 */
-	bp = xfs_buf_read_uncached(mp->m_rtdev_targp,
+	error = xfs_buf_read_uncached(mp->m_rtdev_targp,
 				XFS_FSB_TO_BB(mp, nrblocks - 1),
-				XFS_FSB_TO_BB(mp, 1), 0, NULL);
-	if (!bp)
-		return EIO;
-	if (bp->b_error) {
-		error = bp->b_error;
-		xfs_buf_relse(bp);
+				XFS_FSB_TO_BB(mp, 1), 0, &bp, NULL);
+	if (error)
 		return error;
-	}
 	xfs_buf_relse(bp);
 
 	/*
@@ -1001,7 +945,7 @@ 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 +1152,7 @@ xfs_rtallocate_extent(
 				len, &sumbp, &sb, prod, &r);
 		break;
 	default:
-		error = EIO;
+		error = -EIO;
 		ASSERT(0);
 	}
 	if (error)
@@ -1235,11 +1179,12 @@ xfs_rtallocate_extent(
  */
 int				/* error */
 xfs_rtmount_init(
-	xfs_mount_t	*mp)	/* file system mount structure */
+	struct xfs_mount	*mp)	/* file system mount structure */
 {
-	xfs_buf_t	*bp;	/* buffer for last block of subvolume */
-	xfs_daddr_t	d;	/* address of last block of subvolume */
-	xfs_sb_t	*sbp;	/* filesystem superblock copy in mount */
+	struct xfs_buf		*bp;	/* buffer for last block of subvolume */
+	struct xfs_sb		*sbp;	/* filesystem superblock copy in mount */
+	xfs_daddr_t		d;	/* address of last block of subvolume */
+	int			error;
 
 	sbp = &mp->m_sb;
 	if (sbp->sb_rblocks == 0)
@@ -1247,7 +1192,7 @@ 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 =
@@ -1263,16 +1208,14 @@ 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,
+	error = xfs_buf_read_uncached(mp->m_rtdev_targp,
 					d - XFS_FSB_TO_BB(mp, 1),
-					XFS_FSB_TO_BB(mp, 1), 0, NULL);
-	if (!bp || bp->b_error) {
+					XFS_FSB_TO_BB(mp, 1), 0, &bp, NULL);
+	if (error) {
 		xfs_warn(mp, "realtime device size check failed");
-		if (bp)
-			xfs_buf_relse(bp);
-		return EIO;
+		return error;
 	}
 	xfs_buf_relse(bp);
 	return 0;
diff --git a/fs/xfs/xfs_rtalloc.h b/fs/xfs/xfs_rtalloc.h
index 752b63d10300..76c0a4a9bb17 100644
--- a/fs/xfs/xfs_rtalloc.h
+++ b/fs/xfs/xfs_rtalloc.h
@@ -111,6 +111,10 @@ int xfs_rtfind_forw(struct xfs_mount *mp, struct xfs_trans *tp,
 		    xfs_rtblock_t *rtblock);
 int xfs_rtmodify_range(struct xfs_mount *mp, struct xfs_trans *tp,
 		       xfs_rtblock_t start, xfs_extlen_t len, int val);
+int xfs_rtmodify_summary_int(struct xfs_mount *mp, struct xfs_trans *tp,
+			     int log, xfs_rtblock_t bbno, int delta,
+			     xfs_buf_t **rbpp, xfs_fsblock_t *rsb,
+			     xfs_suminfo_t *sum);
 int xfs_rtmodify_summary(struct xfs_mount *mp, struct xfs_trans *tp, int log,
 			 xfs_rtblock_t bbno, int delta, xfs_buf_t **rbpp,
 			 xfs_fsblock_t *rsb);
@@ -132,7 +136,7 @@ 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_super.c b/fs/xfs/xfs_super.c
index 8f0333b3f7a0..9f622feda6a4 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -47,6 +47,7 @@
 #include "xfs_dinode.h"
 #include "xfs_filestream.h"
 #include "xfs_quota.h"
+#include "xfs_sysfs.h"
 
 #include <linux/namei.h>
 #include <linux/init.h>
@@ -62,6 +63,11 @@ static const struct super_operations xfs_super_operations;
 static kmem_zone_t *xfs_ioend_zone;
 mempool_t *xfs_ioend_pool;
 
+static struct kset *xfs_kset;		/* top-level xfs sysfs dir */
+#ifdef DEBUG
+static struct xfs_kobj xfs_dbg_kobj;	/* global debug sysfs attrs */
+#endif
+
 #define MNTOPT_LOGBUFS	"logbufs"	/* number of XFS log buffers */
 #define MNTOPT_LOGBSIZE	"logbsize"	/* size of XFS log buffers */
 #define MNTOPT_LOGDEV	"logdev"	/* log device */
@@ -185,7 +191,7 @@ 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 +210,6 @@ 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 +230,57 @@ 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 +300,22 @@ 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 +388,7 @@ 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 +398,32 @@ 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 +444,7 @@ 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 +454,7 @@ 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 +463,7 @@ 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 +595,20 @@ 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 +623,12 @@ 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 +653,11 @@ 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 +691,7 @@ xfs_blkdev_get(
 		xfs_warn(mp, "Invalid device [%s], error=%d\n", name, error);
 	}
 
-	return -error;
+	return error;
 }
 
 STATIC void
@@ -756,7 +761,7 @@ 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;
 		}
 	}
@@ -764,7 +769,7 @@ xfs_open_devices(
 	/*
 	 * Setup xfs_mount buffer target pointers
 	 */
-	error = ENOMEM;
+	error = -ENOMEM;
 	mp->m_ddev_targp = xfs_alloc_buftarg(mp, ddev);
 	if (!mp->m_ddev_targp)
 		goto out_close_rtdev;
@@ -838,32 +843,32 @@ xfs_init_mount_workqueues(
 	struct xfs_mount	*mp)
 {
 	mp->m_data_workqueue = alloc_workqueue("xfs-data/%s",
-			WQ_MEM_RECLAIM, 0, mp->m_fsname);
+			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_fsname);
 	if (!mp->m_data_workqueue)
 		goto out;
 
 	mp->m_unwritten_workqueue = alloc_workqueue("xfs-conv/%s",
-			WQ_MEM_RECLAIM, 0, mp->m_fsname);
+			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_fsname);
 	if (!mp->m_unwritten_workqueue)
 		goto out_destroy_data_iodone_queue;
 
 	mp->m_cil_workqueue = alloc_workqueue("xfs-cil/%s",
-			WQ_MEM_RECLAIM, 0, mp->m_fsname);
+			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_fsname);
 	if (!mp->m_cil_workqueue)
 		goto out_destroy_unwritten;
 
 	mp->m_reclaim_workqueue = alloc_workqueue("xfs-reclaim/%s",
-			0, 0, mp->m_fsname);
+			WQ_FREEZABLE, 0, mp->m_fsname);
 	if (!mp->m_reclaim_workqueue)
 		goto out_destroy_cil;
 
 	mp->m_log_workqueue = alloc_workqueue("xfs-log/%s",
-			0, 0, mp->m_fsname);
+			WQ_FREEZABLE, 0, mp->m_fsname);
 	if (!mp->m_log_workqueue)
 		goto out_destroy_reclaim;
 
 	mp->m_eofblocks_workqueue = alloc_workqueue("xfs-eofblocks/%s",
-			0, 0, mp->m_fsname);
+			WQ_FREEZABLE, 0, mp->m_fsname);
 	if (!mp->m_eofblocks_workqueue)
 		goto out_destroy_log;
 
@@ -1188,6 +1193,7 @@ 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;
@@ -1208,10 +1214,10 @@ 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:
 			/*
@@ -1295,7 +1301,7 @@ 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
@@ -1314,7 +1320,7 @@ 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);
 }
 
 /*
@@ -1336,14 +1342,14 @@ 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;
 		}
 	}
 
@@ -1355,7 +1361,7 @@ 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;
 	}
 
 	/*
@@ -1372,7 +1378,7 @@ 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)) &&
@@ -1380,7 +1386,7 @@ 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;
@@ -1394,7 +1400,7 @@ 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)
@@ -1405,6 +1411,7 @@ xfs_fs_fill_super(
 	atomic_set(&mp->m_active_trans, 0);
 	INIT_DELAYED_WORK(&mp->m_reclaim_work, xfs_reclaim_worker);
 	INIT_DELAYED_WORK(&mp->m_eofblocks_work, xfs_eofblocks_worker);
+	mp->m_kobj.kobject.kset = xfs_kset;
 
 	mp->m_super = sb;
 	sb->s_fs_info = mp;
@@ -1428,11 +1435,11 @@ xfs_fs_fill_super(
 	if (error)
 		goto out_free_fsname;
 
-	error = -xfs_init_mount_workqueues(mp);
+	error = xfs_init_mount_workqueues(mp);
 	if (error)
 		goto out_close_devices;
 
-	error = -xfs_icsb_init_counters(mp);
+	error = xfs_icsb_init_counters(mp);
 	if (error)
 		goto out_destroy_workqueues;
 
@@ -1474,12 +1481,12 @@ 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;
 	}
 
@@ -1499,7 +1506,7 @@ out_destroy_workqueues:
 	xfs_free_fsname(mp);
 	kfree(mp);
  out:
-	return -error;
+	return error;
 
  out_unmount:
 	xfs_filestream_unmount(mp);
@@ -1714,7 +1721,8 @@ xfs_init_workqueues(void)
 	 * AGs in all the filesystems mounted. Hence use the default large
 	 * max_active value for this workqueue.
 	 */
-	xfs_alloc_wq = alloc_workqueue("xfsalloc", WQ_MEM_RECLAIM, 0);
+	xfs_alloc_wq = alloc_workqueue("xfsalloc",
+			WQ_MEM_RECLAIM|WQ_FREEZABLE, 0);
 	if (!xfs_alloc_wq)
 		return -ENOMEM;
 
@@ -1761,9 +1769,22 @@ 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;;
+	}
+
+#ifdef DEBUG
+	xfs_dbg_kobj.kobject.kset = xfs_kset;
+	error = xfs_sysfs_init(&xfs_dbg_kobj, &xfs_dbg_ktype, NULL, "debug");
+	if (error)
+		goto out_kset_unregister;
+#endif
+
 	error = xfs_qm_init();
 	if (error)
-		goto out_sysctl_unregister;
+		goto out_remove_kobj;
 
 	error = register_filesystem(&xfs_fs_type);
 	if (error)
@@ -1772,6 +1793,12 @@ init_xfs_fs(void)
 
  out_qm_exit:
 	xfs_qm_exit();
+ out_remove_kobj:
+#ifdef DEBUG
+	xfs_sysfs_del(&xfs_dbg_kobj);
+ out_kset_unregister:
+#endif
+	kset_unregister(xfs_kset);
  out_sysctl_unregister:
 	xfs_sysctl_unregister();
  out_cleanup_procfs:
@@ -1793,6 +1820,10 @@ exit_xfs_fs(void)
 {
 	xfs_qm_exit();
 	unregister_filesystem(&xfs_fs_type);
+#ifdef DEBUG
+	xfs_sysfs_del(&xfs_dbg_kobj);
+#endif
+	kset_unregister(xfs_kset);
 	xfs_sysctl_unregister();
 	xfs_cleanup_procfs();
 	xfs_buf_terminate();
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 d69363c833e1..02ae62a998e0 100644
--- a/fs/xfs/xfs_symlink.c
+++ b/fs/xfs/xfs_symlink.c
@@ -76,15 +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;
+			if (error == -EFSBADCRC)
+				error = -EFSCORRUPTED;
 			goto out;
 		}
 		byte_cnt = XFS_SYMLINK_BUF_SPACE(mp, byte_cnt);
@@ -95,7 +95,7 @@ xfs_readlink_bmap(
 		if (xfs_sb_version_hascrc(&mp->m_sb)) {
 			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);
@@ -135,7 +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);
 
@@ -148,7 +148,7 @@ xfs_readlink(
 			 __func__, (unsigned long long) ip->i_ino,
 			 (long long) pathlen);
 		ASSERT(0);
-		error = XFS_ERROR(EFSCORRUPTED);
+		error = -EFSCORRUPTED;
 		goto out;
 	}
 
@@ -203,14 +203,14 @@ 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;
 	prid = xfs_get_initial_prid(dp);
@@ -238,7 +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);
 	}
@@ -254,7 +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;
 	}
 
@@ -269,9 +269,11 @@ xfs_symlink(
 	/*
 	 * Check for ability to enter directory entry, if no space reserved.
 	 */
-	error = xfs_dir_canenter(tp, dp, link_name, resblks);
-	if (error)
-		goto error_return;
+	if (!resblks) {
+		error = xfs_dir_canenter(tp, dp, link_name);
+		if (error)
+			goto error_return;
+	}
 	/*
 	 * Initialize the bmap freelist prior to calling either
 	 * bmapi or the directory create code.
@@ -284,7 +286,7 @@ 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;
 	}
@@ -348,7 +350,7 @@ 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;
@@ -489,7 +491,7 @@ 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);
@@ -562,7 +564,7 @@ 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);
 
@@ -580,7 +582,7 @@ 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_sysctl.h b/fs/xfs/xfs_sysctl.h
index bd8e157c20ef..ffef45375754 100644
--- a/fs/xfs/xfs_sysctl.h
+++ b/fs/xfs/xfs_sysctl.h
@@ -92,6 +92,11 @@ enum {
 
 extern xfs_param_t	xfs_params;
 
+struct xfs_globals {
+	int	log_recovery_delay;	/* log recovery delay (secs) */
+};
+extern struct xfs_globals	xfs_globals;
+
 #ifdef CONFIG_SYSCTL
 extern int xfs_sysctl_register(void);
 extern void xfs_sysctl_unregister(void);
diff --git a/fs/xfs/xfs_sysfs.c b/fs/xfs/xfs_sysfs.c
new file mode 100644
index 000000000000..aa03670851d8
--- /dev/null
+++ b/fs/xfs/xfs_sysfs.c
@@ -0,0 +1,239 @@
+/*
+ * 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,
+};
+
+#ifdef DEBUG
+/* debug */
+
+STATIC ssize_t
+log_recovery_delay_store(
+	const char	*buf,
+	size_t		count,
+	void		*data)
+{
+	int		ret;
+	int		val;
+
+	ret = kstrtoint(buf, 0, &val);
+	if (ret)
+		return ret;
+
+	if (val < 0 || val > 60)
+		return -EINVAL;
+
+	xfs_globals.log_recovery_delay = val;
+
+	return count;
+}
+
+STATIC ssize_t
+log_recovery_delay_show(
+	char	*buf,
+	void	*data)
+{
+	return snprintf(buf, PAGE_SIZE, "%d\n", xfs_globals.log_recovery_delay);
+}
+XFS_SYSFS_ATTR_RW(log_recovery_delay);
+
+static struct attribute *xfs_dbg_attrs[] = {
+	ATTR_LIST(log_recovery_delay),
+	NULL,
+};
+
+STATIC ssize_t
+xfs_dbg_show(
+	struct kobject		*kobject,
+	struct attribute	*attr,
+	char			*buf)
+{
+	struct xfs_sysfs_attr *xfs_attr = to_attr(attr);
+
+	return xfs_attr->show ? xfs_attr->show(buf, NULL) : 0;
+}
+
+STATIC ssize_t
+xfs_dbg_store(
+	struct kobject		*kobject,
+	struct attribute	*attr,
+	const char		*buf,
+	size_t			count)
+{
+	struct xfs_sysfs_attr *xfs_attr = to_attr(attr);
+
+	return xfs_attr->store ? xfs_attr->store(buf, count, NULL) : 0;
+}
+
+static struct sysfs_ops xfs_dbg_ops = {
+	.show = xfs_dbg_show,
+	.store = xfs_dbg_store,
+};
+
+struct kobj_type xfs_dbg_ktype = {
+	.release = xfs_sysfs_release,
+	.sysfs_ops = &xfs_dbg_ops,
+	.default_attrs = xfs_dbg_attrs,
+};
+
+#endif /* DEBUG */
+
+/* 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..240eee35f342
--- /dev/null
+++ b/fs/xfs/xfs_sysfs.h
@@ -0,0 +1,60 @@
+/*
+ * 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_dbg_ktype;	/* debug */
+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.h b/fs/xfs/xfs_trace.h
index 152f82782630..51372e34d988 100644
--- a/fs/xfs/xfs_trace.h
+++ b/fs/xfs/xfs_trace.h
@@ -349,7 +349,8 @@ DEFINE_BUF_EVENT(xfs_buf_free);
 DEFINE_BUF_EVENT(xfs_buf_hold);
 DEFINE_BUF_EVENT(xfs_buf_rele);
 DEFINE_BUF_EVENT(xfs_buf_iodone);
-DEFINE_BUF_EVENT(xfs_buf_iorequest);
+DEFINE_BUF_EVENT(xfs_buf_submit);
+DEFINE_BUF_EVENT(xfs_buf_submit_wait);
 DEFINE_BUF_EVENT(xfs_buf_bawrite);
 DEFINE_BUF_EVENT(xfs_buf_lock);
 DEFINE_BUF_EVENT(xfs_buf_lock_done);
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index d03932564ccb..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;
@@ -874,7 +874,7 @@ xfs_trans_commit(
 		goto out_unreserve;
 
 	if (XFS_FORCED_SHUTDOWN(mp)) {
-		error = XFS_ERROR(EIO);
+		error = -EIO;
 		goto out_unreserve;
 	}
 
@@ -917,7 +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);
@@ -1024,7 +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_ail.c b/fs/xfs/xfs_trans_ail.c
index cb0f3a84cc68..859482f53b5a 100644
--- a/fs/xfs/xfs_trans_ail.c
+++ b/fs/xfs/xfs_trans_ail.c
@@ -762,7 +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);
@@ -781,7 +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 b8eef0549f3f..e2b2216b1635 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;
@@ -277,8 +276,8 @@ xfs_trans_read_buf_map(
 			xfs_buf_relse(bp);
 
 			/* bad CRC means corrupted metadata */
-			if (error == EFSBADCRC)
-				error = EFSCORRUPTED;
+			if (error == -EFSBADCRC)
+				error = -EFSCORRUPTED;
 			return error;
 		}
 #ifdef DEBUG
@@ -287,7 +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;
 				}
 			}
 		}
@@ -319,20 +318,10 @@ xfs_trans_read_buf_map(
 			XFS_BUF_READ(bp);
 			bp->b_ops = ops;
 
-			/*
-			 * 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);
+			error = xfs_buf_submit_wait(bp);
 			if (error) {
-				xfs_buf_ioerror_alert(bp, __func__);
+				if (!XFS_FORCED_SHUTDOWN(mp))
+					xfs_buf_ioerror_alert(bp, __func__);
 				xfs_buf_relse(bp);
 				/*
 				 * We can gracefully recover from most read
@@ -343,8 +332,8 @@ xfs_trans_read_buf_map(
 					xfs_force_shutdown(tp->t_mountp,
 							SHUTDOWN_META_IO_ERROR);
 				/* bad CRC means corrupted metadata */
-				if (error == EFSBADCRC)
-					error = EFSCORRUPTED;
+				if (error == -EFSBADCRC)
+					error = -EFSCORRUPTED;
 				return error;
 			}
 		}
@@ -355,7 +344,7 @@ 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;
 		}
 
 
@@ -372,7 +361,7 @@ 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;
@@ -384,8 +373,8 @@ xfs_trans_read_buf_map(
 		xfs_buf_relse(bp);
 
 		/* bad CRC means corrupted metadata */
-		if (error == EFSBADCRC)
-			error = EFSCORRUPTED;
+		if (error == -EFSBADCRC)
+			error = -EFSCORRUPTED;
 		return error;
 	}
 #ifdef DEBUG
@@ -396,7 +385,7 @@ 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;
 			}
 		}
 	}
@@ -414,7 +403,7 @@ 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 41172861e857..846e061c2e98 100644
--- a/fs/xfs/xfs_trans_dquot.c
+++ b/fs/xfs/xfs_trans_dquot.c
@@ -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_inode.c b/fs/xfs/xfs_trans_inode.c
index 50c3f5614288..cdb4d86520e1 100644
--- a/fs/xfs/xfs_trans_inode.c
+++ b/fs/xfs/xfs_trans_inode.c
@@ -70,7 +70,7 @@ xfs_trans_ichgtime(
 	int			flags)
 {
 	struct inode		*inode = VFS_I(ip);
-	timespec_t		tv;
+	struct timespec		tv;
 
 	ASSERT(tp);
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
diff --git a/fs/xfs/xfs_types.h b/fs/xfs/xfs_types.h
index 65c6e6650b1a..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)
diff --git a/fs/xfs/xfs_vnode.h b/fs/xfs/xfs_vnode.h
deleted file mode 100644
index e8a77383c0d5..000000000000
--- a/fs/xfs/xfs_vnode.h
+++ /dev/null
@@ -1,46 +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"}
-
-/*
- * 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 78ed92a46fdd..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);
 }