Merge tag 'v3.5' of 192.168.0.154:/repos/git/linux-2.6

Conflicts:
	drivers/Kconfig

Signed-off-by: Patrick McHardy <kaber@trash.net>

97 files changed, 4985 insertions, 7248 deletions

diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index 427a4e82a58..d2bf974b1a2 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -33,6 +33,7 @@ xfs-y				+= xfs_aops.o \
 				   xfs_discard.o \
 				   xfs_error.o \
 				   xfs_export.o \
+				   xfs_extent_busy.o \
 				   xfs_file.o \
 				   xfs_filestream.o \
 				   xfs_fsops.o \
@@ -49,7 +50,6 @@ xfs-y				+= xfs_aops.o \
 				   xfs_sync.o \
 				   xfs_xattr.o \
 				   xfs_rename.o \
-				   xfs_rw.o \
 				   xfs_utils.o \
 				   xfs_vnodeops.o \
 				   kmem.o \
@@ -96,9 +96,6 @@ xfs-$(CONFIG_XFS_QUOTA)		+= xfs_dquot.o \
 				   xfs_qm_bhv.o \
 				   xfs_qm.o \
 				   xfs_quotaops.o
-ifeq ($(CONFIG_XFS_QUOTA),y)
-xfs-$(CONFIG_PROC_FS)		+= xfs_qm_stats.o
-endif
 xfs-$(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 a907de565db..4a7286c1dc8 100644
--- a/fs/xfs/kmem.c
+++ b/fs/xfs/kmem.c
@@ -46,7 +46,7 @@ kmem_zalloc_greedy(size_t *size, size_t minsize, size_t maxsize)
 }
 
 void *
-kmem_alloc(size_t size, unsigned int __nocast flags)
+kmem_alloc(size_t size, xfs_km_flags_t flags)
 {
 	int	retries = 0;
 	gfp_t	lflags = kmem_flags_convert(flags);
@@ -65,7 +65,7 @@ kmem_alloc(size_t size, unsigned int __nocast flags)
 }
 
 void *
-kmem_zalloc(size_t size, unsigned int __nocast flags)
+kmem_zalloc(size_t size, xfs_km_flags_t flags)
 {
 	void	*ptr;
 
@@ -87,7 +87,7 @@ kmem_free(const void *ptr)
 
 void *
 kmem_realloc(const void *ptr, size_t newsize, size_t oldsize,
-	     unsigned int __nocast flags)
+	     xfs_km_flags_t flags)
 {
 	void	*new;
 
@@ -102,7 +102,7 @@ kmem_realloc(const void *ptr, size_t newsize, size_t oldsize,
 }
 
 void *
-kmem_zone_alloc(kmem_zone_t *zone, unsigned int __nocast flags)
+kmem_zone_alloc(kmem_zone_t *zone, xfs_km_flags_t flags)
 {
 	int	retries = 0;
 	gfp_t	lflags = kmem_flags_convert(flags);
@@ -121,7 +121,7 @@ kmem_zone_alloc(kmem_zone_t *zone, unsigned int __nocast flags)
 }
 
 void *
-kmem_zone_zalloc(kmem_zone_t *zone, unsigned int __nocast flags)
+kmem_zone_zalloc(kmem_zone_t *zone, xfs_km_flags_t flags)
 {
 	void	*ptr;
 
diff --git a/fs/xfs/kmem.h b/fs/xfs/kmem.h
index 292eff19803..b2f2620f9a8 100644
--- a/fs/xfs/kmem.h
+++ b/fs/xfs/kmem.h
@@ -27,10 +27,11 @@
  * General memory allocation interfaces
  */
 
-#define KM_SLEEP	0x0001u
-#define KM_NOSLEEP	0x0002u
-#define KM_NOFS		0x0004u
-#define KM_MAYFAIL	0x0008u
+typedef unsigned __bitwise xfs_km_flags_t;
+#define KM_SLEEP	((__force xfs_km_flags_t)0x0001u)
+#define KM_NOSLEEP	((__force xfs_km_flags_t)0x0002u)
+#define KM_NOFS		((__force xfs_km_flags_t)0x0004u)
+#define KM_MAYFAIL	((__force xfs_km_flags_t)0x0008u)
 
 /*
  * We use a special process flag to avoid recursive callbacks into
@@ -38,7 +39,7 @@
  * warnings, so we explicitly skip any generic ones (silly of us).
  */
 static inline gfp_t
-kmem_flags_convert(unsigned int __nocast flags)
+kmem_flags_convert(xfs_km_flags_t flags)
 {
 	gfp_t	lflags;
 
@@ -54,9 +55,9 @@ kmem_flags_convert(unsigned int __nocast flags)
 	return lflags;
 }
 
-extern void *kmem_alloc(size_t, unsigned int __nocast);
-extern void *kmem_zalloc(size_t, unsigned int __nocast);
-extern void *kmem_realloc(const void *, size_t, size_t, unsigned int __nocast);
+extern void *kmem_alloc(size_t, xfs_km_flags_t);
+extern void *kmem_zalloc(size_t, xfs_km_flags_t);
+extern void *kmem_realloc(const void *, size_t, size_t, xfs_km_flags_t);
 extern void  kmem_free(const void *);
 
 static inline void *kmem_zalloc_large(size_t size)
@@ -107,13 +108,7 @@ kmem_zone_destroy(kmem_zone_t *zone)
 		kmem_cache_destroy(zone);
 }
 
-extern void *kmem_zone_alloc(kmem_zone_t *, unsigned int __nocast);
-extern void *kmem_zone_zalloc(kmem_zone_t *, unsigned int __nocast);
-
-static inline int
-kmem_shake_allow(gfp_t gfp_mask)
-{
-	return ((gfp_mask & __GFP_WAIT) && (gfp_mask & __GFP_FS));
-}
+extern void *kmem_zone_alloc(kmem_zone_t *, xfs_km_flags_t);
+extern void *kmem_zone_zalloc(kmem_zone_t *, xfs_km_flags_t);
 
 #endif /* __XFS_SUPPORT_KMEM_H__ */
diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c
index 76e4266d2e7..ac702a6eab9 100644
--- a/fs/xfs/xfs_acl.c
+++ b/fs/xfs/xfs_acl.c
@@ -39,7 +39,7 @@ xfs_acl_from_disk(struct xfs_acl *aclp)
 	struct posix_acl_entry *acl_e;
 	struct posix_acl *acl;
 	struct xfs_acl_entry *ace;
-	int count, i;
+	unsigned int count, i;
 
 	count = be32_to_cpu(aclp->acl_cnt);
 	if (count > XFS_ACL_MAX_ENTRIES)
diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h
index 4805f009f92..44d65c1533c 100644
--- a/fs/xfs/xfs_ag.h
+++ b/fs/xfs/xfs_ag.h
@@ -175,24 +175,6 @@ typedef struct xfs_agfl {
 } xfs_agfl_t;
 
 /*
- * Busy block/extent entry.  Indexed by a rbtree in perag to mark blocks that
- * have been freed but whose transactions aren't committed to disk yet.
- *
- * Note that we use the transaction ID to record the transaction, not the
- * transaction structure itself. See xfs_alloc_busy_insert() for details.
- */
-struct xfs_busy_extent {
-	struct rb_node	rb_node;	/* ag by-bno indexed search tree */
-	struct list_head list;		/* transaction busy extent list */
-	xfs_agnumber_t	agno;
-	xfs_agblock_t	bno;
-	xfs_extlen_t	length;
-	unsigned int	flags;
-#define XFS_ALLOC_BUSY_DISCARDED	0x01	/* undergoing a discard op. */
-#define XFS_ALLOC_BUSY_SKIP_DISCARD	0x02	/* do not discard */
-};
-
-/*
  * Per-ag incore structure, copies of information in agf and agi,
  * to improve the performance of allocation group selection.
  */
diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c
index ce84ffd0264..4f33c32affe 100644
--- a/fs/xfs/xfs_alloc.c
+++ b/fs/xfs/xfs_alloc.c
@@ -20,7 +20,6 @@
 #include "xfs_types.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -32,9 +31,11 @@
 #include "xfs_inode.h"
 #include "xfs_btree.h"
 #include "xfs_alloc.h"
+#include "xfs_extent_busy.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
 
+struct workqueue_struct *xfs_alloc_wq;
 
 #define XFS_ABSDIFF(a,b)	(((a) <= (b)) ? ((b) - (a)) : ((a) - (b)))
 
@@ -46,8 +47,6 @@ STATIC int xfs_alloc_ag_vextent_near(xfs_alloc_arg_t *);
 STATIC int xfs_alloc_ag_vextent_size(xfs_alloc_arg_t *);
 STATIC int xfs_alloc_ag_vextent_small(xfs_alloc_arg_t *,
 		xfs_btree_cur_t *, xfs_agblock_t *, xfs_extlen_t *, int *);
-STATIC void xfs_alloc_busy_trim(struct xfs_alloc_arg *,
-		xfs_agblock_t, xfs_extlen_t, xfs_agblock_t *, xfs_extlen_t *);
 
 /*
  * Lookup the record equal to [bno, len] in the btree given by cur.
@@ -68,7 +67,7 @@ xfs_alloc_lookup_eq(
  * Lookup the first record greater than or equal to [bno, len]
  * in the btree given by cur.
  */
-STATIC int				/* error */
+int				/* error */
 xfs_alloc_lookup_ge(
 	struct xfs_btree_cur	*cur,	/* btree cursor */
 	xfs_agblock_t		bno,	/* starting block of extent */
@@ -151,7 +150,7 @@ xfs_alloc_compute_aligned(
 	xfs_extlen_t	len;
 
 	/* Trim busy sections out of found extent */
-	xfs_alloc_busy_trim(args, foundbno, foundlen, &bno, &len);
+	xfs_extent_busy_trim(args, foundbno, foundlen, &bno, &len);
 
 	if (args->alignment > 1 && len >= args->minlen) {
 		xfs_agblock_t	aligned_bno = roundup(bno, args->alignment);
@@ -535,7 +534,7 @@ xfs_alloc_ag_vextent(
 		if (error)
 			return error;
 
-		ASSERT(!xfs_alloc_busy_search(args->mp, args->agno,
+		ASSERT(!xfs_extent_busy_search(args->mp, args->agno,
 					      args->agbno, args->len));
 	}
 
@@ -602,7 +601,7 @@ xfs_alloc_ag_vextent_exact(
 	/*
 	 * Check for overlapping busy extents.
 	 */
-	xfs_alloc_busy_trim(args, fbno, flen, &tbno, &tlen);
+	xfs_extent_busy_trim(args, fbno, flen, &tbno, &tlen);
 
 	/*
 	 * Give up if the start of the extent is busy, or the freespace isn't
@@ -1075,12 +1074,13 @@ restart:
 	 * If we couldn't get anything, give up.
 	 */
 	if (bno_cur_lt == NULL && bno_cur_gt == NULL) {
+		xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
+
 		if (!forced++) {
 			trace_xfs_alloc_near_busy(args);
 			xfs_log_force(args->mp, XFS_LOG_SYNC);
 			goto restart;
 		}
-
 		trace_xfs_alloc_size_neither(args);
 		args->agbno = NULLAGBLOCK;
 		return 0;
@@ -1390,7 +1390,7 @@ xfs_alloc_ag_vextent_small(
 		if (error)
 			goto error0;
 		if (fbno != NULLAGBLOCK) {
-			xfs_alloc_busy_reuse(args->mp, args->agno, fbno, 1,
+			xfs_extent_busy_reuse(args->mp, args->agno, fbno, 1,
 					     args->userdata);
 
 			if (args->userdata) {
@@ -2207,7 +2207,7 @@ xfs_alloc_read_agf(
  * group or loop over the allocation groups to find the result.
  */
 int				/* error */
-xfs_alloc_vextent(
+__xfs_alloc_vextent(
 	xfs_alloc_arg_t	*args)	/* allocation argument structure */
 {
 	xfs_agblock_t	agsize;	/* allocation group size */
@@ -2417,6 +2417,46 @@ error0:
 	return error;
 }
 
+static void
+xfs_alloc_vextent_worker(
+	struct work_struct	*work)
+{
+	struct xfs_alloc_arg	*args = container_of(work,
+						struct xfs_alloc_arg, work);
+	unsigned long		pflags;
+
+	/* we are in a transaction context here */
+	current_set_flags_nested(&pflags, PF_FSTRANS);
+
+	args->result = __xfs_alloc_vextent(args);
+	complete(args->done);
+
+	current_restore_flags_nested(&pflags, PF_FSTRANS);
+}
+
+/*
+ * Data allocation requests often come in with little stack to work on. Push
+ * them off to a worker thread so there is lots of stack to use. Metadata
+ * requests, OTOH, are generally from low stack usage paths, so avoid the
+ * context switch overhead here.
+ */
+int
+xfs_alloc_vextent(
+	struct xfs_alloc_arg	*args)
+{
+	DECLARE_COMPLETION_ONSTACK(done);
+
+	if (!args->userdata)
+		return __xfs_alloc_vextent(args);
+
+
+	args->done = &done;
+	INIT_WORK_ONSTACK(&args->work, xfs_alloc_vextent_worker);
+	queue_work(xfs_alloc_wq, &args->work);
+	wait_for_completion(&done);
+	return args->result;
+}
+
 /*
  * Free an extent.
  * Just break up the extent address and hand off to xfs_free_ag_extent
@@ -2464,579 +2504,8 @@ xfs_free_extent(
 
 	error = xfs_free_ag_extent(tp, args.agbp, args.agno, args.agbno, len, 0);
 	if (!error)
-		xfs_alloc_busy_insert(tp, args.agno, args.agbno, len, 0);
+		xfs_extent_busy_insert(tp, args.agno, args.agbno, len, 0);
 error0:
 	xfs_perag_put(args.pag);
 	return error;
 }
-
-void
-xfs_alloc_busy_insert(
-	struct xfs_trans	*tp,
-	xfs_agnumber_t		agno,
-	xfs_agblock_t		bno,
-	xfs_extlen_t		len,
-	unsigned int		flags)
-{
-	struct xfs_busy_extent	*new;
-	struct xfs_busy_extent	*busyp;
-	struct xfs_perag	*pag;
-	struct rb_node		**rbp;
-	struct rb_node		*parent = NULL;
-
-	new = kmem_zalloc(sizeof(struct xfs_busy_extent), KM_MAYFAIL);
-	if (!new) {
-		/*
-		 * No Memory!  Since it is now not possible to track the free
-		 * block, make this a synchronous transaction to insure that
-		 * the block is not reused before this transaction commits.
-		 */
-		trace_xfs_alloc_busy_enomem(tp->t_mountp, agno, bno, len);
-		xfs_trans_set_sync(tp);
-		return;
-	}
-
-	new->agno = agno;
-	new->bno = bno;
-	new->length = len;
-	INIT_LIST_HEAD(&new->list);
-	new->flags = flags;
-
-	/* trace before insert to be able to see failed inserts */
-	trace_xfs_alloc_busy(tp->t_mountp, agno, bno, len);
-
-	pag = xfs_perag_get(tp->t_mountp, new->agno);
-	spin_lock(&pag->pagb_lock);
-	rbp = &pag->pagb_tree.rb_node;
-	while (*rbp) {
-		parent = *rbp;
-		busyp = rb_entry(parent, struct xfs_busy_extent, rb_node);
-
-		if (new->bno < busyp->bno) {
-			rbp = &(*rbp)->rb_left;
-			ASSERT(new->bno + new->length <= busyp->bno);
-		} else if (new->bno > busyp->bno) {
-			rbp = &(*rbp)->rb_right;
-			ASSERT(bno >= busyp->bno + busyp->length);
-		} else {
-			ASSERT(0);
-		}
-	}
-
-	rb_link_node(&new->rb_node, parent, rbp);
-	rb_insert_color(&new->rb_node, &pag->pagb_tree);
-
-	list_add(&new->list, &tp->t_busy);
-	spin_unlock(&pag->pagb_lock);
-	xfs_perag_put(pag);
-}
-
-/*
- * Search for a busy extent within the range of the extent we are about to
- * allocate.  You need to be holding the busy extent tree lock when calling
- * xfs_alloc_busy_search(). This function returns 0 for no overlapping busy
- * extent, -1 for an overlapping but not exact busy extent, and 1 for an exact
- * match. This is done so that a non-zero return indicates an overlap that
- * will require a synchronous transaction, but it can still be
- * used to distinguish between a partial or exact match.
- */
-int
-xfs_alloc_busy_search(
-	struct xfs_mount	*mp,
-	xfs_agnumber_t		agno,
-	xfs_agblock_t		bno,
-	xfs_extlen_t		len)
-{
-	struct xfs_perag	*pag;
-	struct rb_node		*rbp;
-	struct xfs_busy_extent	*busyp;
-	int			match = 0;
-
-	pag = xfs_perag_get(mp, agno);
-	spin_lock(&pag->pagb_lock);
-
-	rbp = pag->pagb_tree.rb_node;
-
-	/* find closest start bno overlap */
-	while (rbp) {
-		busyp = rb_entry(rbp, struct xfs_busy_extent, rb_node);
-		if (bno < busyp->bno) {
-			/* may overlap, but exact start block is lower */
-			if (bno + len > busyp->bno)
-				match = -1;
-			rbp = rbp->rb_left;
-		} else if (bno > busyp->bno) {
-			/* may overlap, but exact start block is higher */
-			if (bno < busyp->bno + busyp->length)
-				match = -1;
-			rbp = rbp->rb_right;
-		} else {
-			/* bno matches busyp, length determines exact match */
-			match = (busyp->length == len) ? 1 : -1;
-			break;
-		}
-	}
-	spin_unlock(&pag->pagb_lock);
-	xfs_perag_put(pag);
-	return match;
-}
-
-/*
- * The found free extent [fbno, fend] overlaps part or all of the given busy
- * extent.  If the overlap covers the beginning, the end, or all of the busy
- * extent, the overlapping portion can be made unbusy and used for the
- * allocation.  We can't split a busy extent because we can't modify a
- * transaction/CIL context busy list, but we can update an entries block
- * number or length.
- *
- * Returns true if the extent can safely be reused, or false if the search
- * needs to be restarted.
- */
-STATIC bool
-xfs_alloc_busy_update_extent(
-	struct xfs_mount	*mp,
-	struct xfs_perag	*pag,
-	struct xfs_busy_extent	*busyp,
-	xfs_agblock_t		fbno,
-	xfs_extlen_t		flen,
-	bool			userdata)
-{
-	xfs_agblock_t		fend = fbno + flen;
-	xfs_agblock_t		bbno = busyp->bno;
-	xfs_agblock_t		bend = bbno + busyp->length;
-
-	/*
-	 * This extent is currently being discarded.  Give the thread
-	 * performing the discard a chance to mark the extent unbusy
-	 * and retry.
-	 */
-	if (busyp->flags & XFS_ALLOC_BUSY_DISCARDED) {
-		spin_unlock(&pag->pagb_lock);
-		delay(1);
-		spin_lock(&pag->pagb_lock);
-		return false;
-	}
-
-	/*
-	 * If there is a busy extent overlapping a user allocation, we have
-	 * no choice but to force the log and retry the search.
-	 *
-	 * Fortunately this does not happen during normal operation, but
-	 * only if the filesystem is very low on space and has to dip into
-	 * the AGFL for normal allocations.
-	 */
-	if (userdata)
-		goto out_force_log;
-
-	if (bbno < fbno && bend > fend) {
-		/*
-		 * Case 1:
-		 *    bbno           bend
-		 *    +BBBBBBBBBBBBBBBBB+
-		 *        +---------+
-		 *        fbno   fend
-		 */
-
-		/*
-		 * We would have to split the busy extent to be able to track
-		 * it correct, which we cannot do because we would have to
-		 * modify the list of busy extents attached to the transaction
-		 * or CIL context, which is immutable.
-		 *
-		 * Force out the log to clear the busy extent and retry the
-		 * search.
-		 */
-		goto out_force_log;
-	} else if (bbno >= fbno && bend <= fend) {
-		/*
-		 * Case 2:
-		 *    bbno           bend
-		 *    +BBBBBBBBBBBBBBBBB+
-		 *    +-----------------+
-		 *    fbno           fend
-		 *
-		 * Case 3:
-		 *    bbno           bend
-		 *    +BBBBBBBBBBBBBBBBB+
-		 *    +--------------------------+
-		 *    fbno                    fend
-		 *
-		 * Case 4:
-		 *             bbno           bend
-		 *             +BBBBBBBBBBBBBBBBB+
-		 *    +--------------------------+
-		 *    fbno                    fend
-		 *
-		 * Case 5:
-		 *             bbno           bend
-		 *             +BBBBBBBBBBBBBBBBB+
-		 *    +-----------------------------------+
-		 *    fbno                             fend
-		 *
-		 */
-
-		/*
-		 * The busy extent is fully covered by the extent we are
-		 * allocating, and can simply be removed from the rbtree.
-		 * However we cannot remove it from the immutable list
-		 * tracking busy extents in the transaction or CIL context,
-		 * so set the length to zero to mark it invalid.
-		 *
-		 * We also need to restart the busy extent search from the
-		 * tree root, because erasing the node can rearrange the
-		 * tree topology.
-		 */
-		rb_erase(&busyp->rb_node, &pag->pagb_tree);
-		busyp->length = 0;
-		return false;
-	} else if (fend < bend) {
-		/*
-		 * Case 6:
-		 *              bbno           bend
-		 *             +BBBBBBBBBBBBBBBBB+
-		 *             +---------+
-		 *             fbno   fend
-		 *
-		 * Case 7:
-		 *             bbno           bend
-		 *             +BBBBBBBBBBBBBBBBB+
-		 *    +------------------+
-		 *    fbno            fend
-		 *
-		 */
-		busyp->bno = fend;
-	} else if (bbno < fbno) {
-		/*
-		 * Case 8:
-		 *    bbno           bend
-		 *    +BBBBBBBBBBBBBBBBB+
-		 *        +-------------+
-		 *        fbno       fend
-		 *
-		 * Case 9:
-		 *    bbno           bend
-		 *    +BBBBBBBBBBBBBBBBB+
-		 *        +----------------------+
-		 *        fbno                fend
-		 */
-		busyp->length = fbno - busyp->bno;
-	} else {
-		ASSERT(0);
-	}
-
-	trace_xfs_alloc_busy_reuse(mp, pag->pag_agno, fbno, flen);
-	return true;
-
-out_force_log:
-	spin_unlock(&pag->pagb_lock);
-	xfs_log_force(mp, XFS_LOG_SYNC);
-	trace_xfs_alloc_busy_force(mp, pag->pag_agno, fbno, flen);
-	spin_lock(&pag->pagb_lock);
-	return false;
-}
-
-
-/*
- * For a given extent [fbno, flen], make sure we can reuse it safely.
- */
-void
-xfs_alloc_busy_reuse(
-	struct xfs_mount	*mp,
-	xfs_agnumber_t		agno,
-	xfs_agblock_t		fbno,
-	xfs_extlen_t		flen,
-	bool			userdata)
-{
-	struct xfs_perag	*pag;
-	struct rb_node		*rbp;
-
-	ASSERT(flen > 0);
-
-	pag = xfs_perag_get(mp, agno);
-	spin_lock(&pag->pagb_lock);
-restart:
-	rbp = pag->pagb_tree.rb_node;
-	while (rbp) {
-		struct xfs_busy_extent *busyp =
-			rb_entry(rbp, struct xfs_busy_extent, rb_node);
-		xfs_agblock_t	bbno = busyp->bno;
-		xfs_agblock_t	bend = bbno + busyp->length;
-
-		if (fbno + flen <= bbno) {
-			rbp = rbp->rb_left;
-			continue;
-		} else if (fbno >= bend) {
-			rbp = rbp->rb_right;
-			continue;
-		}
-
-		if (!xfs_alloc_busy_update_extent(mp, pag, busyp, fbno, flen,
-						  userdata))
-			goto restart;
-	}
-	spin_unlock(&pag->pagb_lock);
-	xfs_perag_put(pag);
-}
-
-/*
- * For a given extent [fbno, flen], search the busy extent list to find a
- * subset of the extent that is not busy.  If *rlen is smaller than
- * args->minlen no suitable extent could be found, and the higher level
- * code needs to force out the log and retry the allocation.
- */
-STATIC void
-xfs_alloc_busy_trim(
-	struct xfs_alloc_arg	*args,
-	xfs_agblock_t		bno,
-	xfs_extlen_t		len,
-	xfs_agblock_t		*rbno,
-	xfs_extlen_t		*rlen)
-{
-	xfs_agblock_t		fbno;
-	xfs_extlen_t		flen;
-	struct rb_node		*rbp;
-
-	ASSERT(len > 0);
-
-	spin_lock(&args->pag->pagb_lock);
-restart:
-	fbno = bno;
-	flen = len;
-	rbp = args->pag->pagb_tree.rb_node;
-	while (rbp && flen >= args->minlen) {
-		struct xfs_busy_extent *busyp =
-			rb_entry(rbp, struct xfs_busy_extent, rb_node);
-		xfs_agblock_t	fend = fbno + flen;
-		xfs_agblock_t	bbno = busyp->bno;
-		xfs_agblock_t	bend = bbno + busyp->length;
-
-		if (fend <= bbno) {
-			rbp = rbp->rb_left;
-			continue;
-		} else if (fbno >= bend) {
-			rbp = rbp->rb_right;
-			continue;
-		}
-
-		/*
-		 * If this is a metadata allocation, try to reuse the busy
-		 * extent instead of trimming the allocation.
-		 */
-		if (!args->userdata &&
-		    !(busyp->flags & XFS_ALLOC_BUSY_DISCARDED)) {
-			if (!xfs_alloc_busy_update_extent(args->mp, args->pag,
-							  busyp, fbno, flen,
-							  false))
-				goto restart;
-			continue;
-		}
-
-		if (bbno <= fbno) {
-			/* start overlap */
-
-			/*
-			 * Case 1:
-			 *    bbno           bend
-			 *    +BBBBBBBBBBBBBBBBB+
-			 *        +---------+
-			 *        fbno   fend
-			 *
-			 * Case 2:
-			 *    bbno           bend
-			 *    +BBBBBBBBBBBBBBBBB+
-			 *    +-------------+
-			 *    fbno       fend
-			 *
-			 * Case 3:
-			 *    bbno           bend
-			 *    +BBBBBBBBBBBBBBBBB+
-			 *        +-------------+
-			 *        fbno       fend
-			 *
-			 * Case 4:
-			 *    bbno           bend
-			 *    +BBBBBBBBBBBBBBBBB+
-			 *    +-----------------+
-			 *    fbno           fend
-			 *
-			 * No unbusy region in extent, return failure.
-			 */
-			if (fend <= bend)
-				goto fail;
-
-			/*
-			 * Case 5:
-			 *    bbno           bend
-			 *    +BBBBBBBBBBBBBBBBB+
-			 *        +----------------------+
-			 *        fbno                fend
-			 *
-			 * Case 6:
-			 *    bbno           bend
-			 *    +BBBBBBBBBBBBBBBBB+
-			 *    +--------------------------+
-			 *    fbno                    fend
-			 *
-			 * Needs to be trimmed to:
-			 *                       +-------+
-			 *                       fbno fend
-			 */
-			fbno = bend;
-		} else if (bend >= fend) {
-			/* end overlap */
-
-			/*
-			 * Case 7:
-			 *             bbno           bend
-			 *             +BBBBBBBBBBBBBBBBB+
-			 *    +------------------+
-			 *    fbno            fend
-			 *
-			 * Case 8:
-			 *             bbno           bend
-			 *             +BBBBBBBBBBBBBBBBB+
-			 *    +--------------------------+
-			 *    fbno                    fend
-			 *
-			 * Needs to be trimmed to:
-			 *    +-------+
-			 *    fbno fend
-			 */
-			fend = bbno;
-		} else {
-			/* middle overlap */
-
-			/*
-			 * Case 9:
-			 *             bbno           bend
-			 *             +BBBBBBBBBBBBBBBBB+
-			 *    +-----------------------------------+
-			 *    fbno                             fend
-			 *
-			 * Can be trimmed to:
-			 *    +-------+        OR         +-------+
-			 *    fbno fend                   fbno fend
-			 *
-			 * Backward allocation leads to significant
-			 * fragmentation of directories, which degrades
-			 * directory performance, therefore we always want to
-			 * choose the option that produces forward allocation
-			 * patterns.
-			 * Preferring the lower bno extent will make the next
-			 * request use "fend" as the start of the next
-			 * allocation;  if the segment is no longer busy at
-			 * that point, we'll get a contiguous allocation, but
-			 * even if it is still busy, we will get a forward
-			 * allocation.
-			 * We try to avoid choosing the segment at "bend",
-			 * because that can lead to the next allocation
-			 * taking the segment at "fbno", which would be a
-			 * backward allocation.  We only use the segment at
-			 * "fbno" if it is much larger than the current
-			 * requested size, because in that case there's a
-			 * good chance subsequent allocations will be
-			 * contiguous.
-			 */
-			if (bbno - fbno >= args->maxlen) {
-				/* left candidate fits perfect */
-				fend = bbno;
-			} else if (fend - bend >= args->maxlen * 4) {
-				/* right candidate has enough free space */
-				fbno = bend;
-			} else if (bbno - fbno >= args->minlen) {
-				/* left candidate fits minimum requirement */
-				fend = bbno;
-			} else {
-				goto fail;
-			}
-		}
-
-		flen = fend - fbno;
-	}
-	spin_unlock(&args->pag->pagb_lock);
-
-	if (fbno != bno || flen != len) {
-		trace_xfs_alloc_busy_trim(args->mp, args->agno, bno, len,
-					  fbno, flen);
-	}
-	*rbno = fbno;
-	*rlen = flen;
-	return;
-fail:
-	/*
-	 * Return a zero extent length as failure indications.  All callers
-	 * re-check if the trimmed extent satisfies the minlen requirement.
-	 */
-	spin_unlock(&args->pag->pagb_lock);
-	trace_xfs_alloc_busy_trim(args->mp, args->agno, bno, len, fbno, 0);
-	*rbno = fbno;
-	*rlen = 0;
-}
-
-static void
-xfs_alloc_busy_clear_one(
-	struct xfs_mount	*mp,
-	struct xfs_perag	*pag,
-	struct xfs_busy_extent	*busyp)
-{
-	if (busyp->length) {
-		trace_xfs_alloc_busy_clear(mp, busyp->agno, busyp->bno,
-						busyp->length);
-		rb_erase(&busyp->rb_node, &pag->pagb_tree);
-	}
-
-	list_del_init(&busyp->list);
-	kmem_free(busyp);
-}
-
-/*
- * Remove all extents on the passed in list from the busy extents tree.
- * If do_discard is set skip extents that need to be discarded, and mark
- * these as undergoing a discard operation instead.
- */
-void
-xfs_alloc_busy_clear(
-	struct xfs_mount	*mp,
-	struct list_head	*list,
-	bool			do_discard)
-{
-	struct xfs_busy_extent	*busyp, *n;
-	struct xfs_perag	*pag = NULL;
-	xfs_agnumber_t		agno = NULLAGNUMBER;
-
-	list_for_each_entry_safe(busyp, n, list, list) {
-		if (busyp->agno != agno) {
-			if (pag) {
-				spin_unlock(&pag->pagb_lock);
-				xfs_perag_put(pag);
-			}
-			pag = xfs_perag_get(mp, busyp->agno);
-			spin_lock(&pag->pagb_lock);
-			agno = busyp->agno;
-		}
-
-		if (do_discard && busyp->length &&
-		    !(busyp->flags & XFS_ALLOC_BUSY_SKIP_DISCARD))
-			busyp->flags = XFS_ALLOC_BUSY_DISCARDED;
-		else
-			xfs_alloc_busy_clear_one(mp, pag, busyp);
-	}
-
-	if (pag) {
-		spin_unlock(&pag->pagb_lock);
-		xfs_perag_put(pag);
-	}
-}
-
-/*
- * Callback for list_sort to sort busy extents by the AG they reside in.
- */
-int
-xfs_busy_extent_ag_cmp(
-	void			*priv,
-	struct list_head	*a,
-	struct list_head	*b)
-{
-	return container_of(a, struct xfs_busy_extent, list)->agno -
-		container_of(b, struct xfs_busy_extent, list)->agno;
-}
diff --git a/fs/xfs/xfs_alloc.h b/fs/xfs/xfs_alloc.h
index 2f52b924be7..93be4a667ca 100644
--- a/fs/xfs/xfs_alloc.h
+++ b/fs/xfs/xfs_alloc.h
@@ -23,7 +23,8 @@ struct xfs_btree_cur;
 struct xfs_mount;
 struct xfs_perag;
 struct xfs_trans;
-struct xfs_busy_extent;
+
+extern struct workqueue_struct *xfs_alloc_wq;
 
 /*
  * Freespace allocation types.  Argument to xfs_alloc_[v]extent.
@@ -119,6 +120,9 @@ typedef struct xfs_alloc_arg {
 	char		isfl;		/* set if is freelist blocks - !acctg */
 	char		userdata;	/* set if this is user data */
 	xfs_fsblock_t	firstblock;	/* io first block allocated */
+	struct completion *done;
+	struct work_struct work;
+	int		result;
 } xfs_alloc_arg_t;
 
 /*
@@ -134,33 +138,6 @@ xfs_extlen_t
 xfs_alloc_longest_free_extent(struct xfs_mount *mp,
 		struct xfs_perag *pag);
 
-#ifdef __KERNEL__
-void
-xfs_alloc_busy_insert(struct xfs_trans *tp, xfs_agnumber_t agno,
-	xfs_agblock_t bno, xfs_extlen_t len, unsigned int flags);
-
-void
-xfs_alloc_busy_clear(struct xfs_mount *mp, struct list_head *list,
-	bool do_discard);
-
-int
-xfs_alloc_busy_search(struct xfs_mount *mp, xfs_agnumber_t agno,
-	xfs_agblock_t bno, xfs_extlen_t len);
-
-void
-xfs_alloc_busy_reuse(struct xfs_mount *mp, xfs_agnumber_t agno,
-	xfs_agblock_t fbno, xfs_extlen_t flen, bool userdata);
-
-int
-xfs_busy_extent_ag_cmp(void *priv, struct list_head *a, struct list_head *b);
-
-static inline void xfs_alloc_busy_sort(struct list_head *list)
-{
-	list_sort(NULL, list, xfs_busy_extent_ag_cmp);
-}
-
-#endif	/* __KERNEL__ */
-
 /*
  * Compute and fill in value of m_ag_maxlevels.
  */
@@ -243,6 +220,13 @@ xfs_alloc_lookup_le(
 	xfs_extlen_t		len,	/* length of extent */
 	int			*stat);	/* success/failure */
 
+int				/* error */
+xfs_alloc_lookup_ge(
+	struct xfs_btree_cur	*cur,	/* btree cursor */
+	xfs_agblock_t		bno,	/* starting block of extent */
+	xfs_extlen_t		len,	/* length of extent */
+	int			*stat);	/* success/failure */
+
 int					/* error */
 xfs_alloc_get_rec(
 	struct xfs_btree_cur	*cur,	/* btree cursor */
diff --git a/fs/xfs/xfs_alloc_btree.c b/fs/xfs/xfs_alloc_btree.c
index ffb3386e45c..f1647caace8 100644
--- a/fs/xfs/xfs_alloc_btree.c
+++ b/fs/xfs/xfs_alloc_btree.c
@@ -18,9 +18,7 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_types.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -32,6 +30,7 @@
 #include "xfs_inode.h"
 #include "xfs_btree.h"
 #include "xfs_alloc.h"
+#include "xfs_extent_busy.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
 
@@ -94,7 +93,7 @@ xfs_allocbt_alloc_block(
 		return 0;
 	}
 
-	xfs_alloc_busy_reuse(cur->bc_mp, cur->bc_private.a.agno, bno, 1, false);
+	xfs_extent_busy_reuse(cur->bc_mp, cur->bc_private.a.agno, bno, 1, false);
 
 	xfs_trans_agbtree_delta(cur->bc_tp, 1);
 	new->s = cpu_to_be32(bno);
@@ -119,8 +118,8 @@ xfs_allocbt_free_block(
 	if (error)
 		return error;
 
-	xfs_alloc_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1,
-			      XFS_ALLOC_BUSY_SKIP_DISCARD);
+	xfs_extent_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1,
+			      XFS_EXTENT_BUSY_SKIP_DISCARD);
 	xfs_trans_agbtree_delta(cur->bc_tp, -1);
 	return 0;
 }
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index 574d4ee9b62..8dad722c004 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -16,9 +16,7 @@
  * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */
 #include "xfs.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_trans.h"
@@ -26,9 +24,9 @@
 #include "xfs_bmap_btree.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_inode_item.h"
 #include "xfs_alloc.h"
 #include "xfs_error.h"
-#include "xfs_rw.h"
 #include "xfs_iomap.h"
 #include "xfs_vnodeops.h"
 #include "xfs_trace.h"
@@ -99,24 +97,6 @@ xfs_destroy_ioend(
 }
 
 /*
- * If the end of the current ioend is beyond the current EOF,
- * return the new EOF value, otherwise zero.
- */
-STATIC xfs_fsize_t
-xfs_ioend_new_eof(
-	xfs_ioend_t		*ioend)
-{
-	xfs_inode_t		*ip = XFS_I(ioend->io_inode);
-	xfs_fsize_t		isize;
-	xfs_fsize_t		bsize;
-
-	bsize = ioend->io_offset + ioend->io_size;
-	isize = MAX(ip->i_size, ip->i_new_size);
-	isize = MIN(isize, bsize);
-	return isize > ip->i_d.di_size ? isize : 0;
-}
-
-/*
  * Fast and loose check if this write could update the on-disk inode size.
  */
 static inline bool xfs_ioend_is_append(struct xfs_ioend *ioend)
@@ -125,36 +105,65 @@ static inline bool xfs_ioend_is_append(struct xfs_ioend *ioend)
 		XFS_I(ioend->io_inode)->i_d.di_size;
 }
 
+STATIC int
+xfs_setfilesize_trans_alloc(
+	struct xfs_ioend	*ioend)
+{
+	struct xfs_mount	*mp = XFS_I(ioend->io_inode)->i_mount;
+	struct xfs_trans	*tp;
+	int			error;
+
+	tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS);
+
+	error = xfs_trans_reserve(tp, 0, XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0);
+	if (error) {
+		xfs_trans_cancel(tp, 0);
+		return error;
+	}
+
+	ioend->io_append_trans = tp;
+
+	/*
+	 * We hand off the transaction to the completion thread now, so
+	 * clear the flag here.
+	 */
+	current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
+	return 0;
+}
+
 /*
- * Update on-disk file size now that data has been written to disk.  The
- * current in-memory file size is i_size.  If a write is beyond eof i_new_size
- * will be the intended file size until i_size is updated.  If this write does
- * not extend all the way to the valid file size then restrict this update to
- * the end of the write.
- *
- * This function does not block as blocking on the inode lock in IO completion
- * can lead to IO completion order dependency deadlocks.. If it can't get the
- * inode ilock it will return EAGAIN. Callers must handle this.
+ * Update on-disk file size now that data has been written to disk.
  */
 STATIC int
 xfs_setfilesize(
-	xfs_ioend_t		*ioend)
+	struct xfs_ioend	*ioend)
 {
-	xfs_inode_t		*ip = XFS_I(ioend->io_inode);
+	struct xfs_inode	*ip = XFS_I(ioend->io_inode);
+	struct xfs_trans	*tp = ioend->io_append_trans;
 	xfs_fsize_t		isize;
 
-	if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL))
-		return EAGAIN;
+	/*
+	 * The transaction was allocated in the I/O submission thread,
+	 * thus we need to mark ourselves as beeing in a transaction
+	 * manually.
+	 */
+	current_set_flags_nested(&tp->t_pflags, PF_FSTRANS);
 
-	isize = xfs_ioend_new_eof(ioend);
-	if (isize) {
-		trace_xfs_setfilesize(ip, ioend->io_offset, ioend->io_size);
-		ip->i_d.di_size = isize;
-		xfs_mark_inode_dirty(ip);
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	isize = xfs_new_eof(ip, ioend->io_offset + ioend->io_size);
+	if (!isize) {
+		xfs_iunlock(ip, XFS_ILOCK_EXCL);
+		xfs_trans_cancel(tp, 0);
+		return 0;
 	}
 
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	return 0;
+	trace_xfs_setfilesize(ip, ioend->io_offset, ioend->io_size);
+
+	ip->i_d.di_size = isize;
+	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+
+	return xfs_trans_commit(tp, 0);
 }
 
 /*
@@ -168,10 +177,12 @@ xfs_finish_ioend(
 	struct xfs_ioend	*ioend)
 {
 	if (atomic_dec_and_test(&ioend->io_remaining)) {
+		struct xfs_mount	*mp = XFS_I(ioend->io_inode)->i_mount;
+
 		if (ioend->io_type == IO_UNWRITTEN)
-			queue_work(xfsconvertd_workqueue, &ioend->io_work);
-		else if (xfs_ioend_is_append(ioend))
-			queue_work(xfsdatad_workqueue, &ioend->io_work);
+			queue_work(mp->m_unwritten_workqueue, &ioend->io_work);
+		else if (ioend->io_append_trans)
+			queue_work(mp->m_data_workqueue, &ioend->io_work);
 		else
 			xfs_destroy_ioend(ioend);
 	}
@@ -200,35 +211,36 @@ xfs_end_io(
 	 * range to normal written extens after the data I/O has finished.
 	 */
 	if (ioend->io_type == IO_UNWRITTEN) {
+		/*
+		 * For buffered I/O we never preallocate a transaction when
+		 * doing the unwritten extent conversion, but for direct I/O
+		 * we do not know if we are converting an unwritten extent
+		 * or not at the point where we preallocate the transaction.
+		 */
+		if (ioend->io_append_trans) {
+			ASSERT(ioend->io_isdirect);
+
+			current_set_flags_nested(
+				&ioend->io_append_trans->t_pflags, PF_FSTRANS);
+			xfs_trans_cancel(ioend->io_append_trans, 0);
+		}
+
 		error = xfs_iomap_write_unwritten(ip, ioend->io_offset,
 						 ioend->io_size);
 		if (error) {
 			ioend->io_error = -error;
 			goto done;
 		}
+	} else if (ioend->io_append_trans) {
+		error = xfs_setfilesize(ioend);
+		if (error)
+			ioend->io_error = -error;
+	} else {
+		ASSERT(!xfs_ioend_is_append(ioend));
 	}
 
-	/*
-	 * We might have to update the on-disk file size after extending
-	 * writes.
-	 */
-	error = xfs_setfilesize(ioend);
-	ASSERT(!error || error == EAGAIN);
-
 done:
-	/*
-	 * If we didn't complete processing of the ioend, requeue it to the
-	 * tail of the workqueue for another attempt later. Otherwise destroy
-	 * it.
-	 */
-	if (error == EAGAIN) {
-		atomic_inc(&ioend->io_remaining);
-		xfs_finish_ioend(ioend);
-		/* ensure we don't spin on blocked ioends */
-		delay(1);
-	} else {
-		xfs_destroy_ioend(ioend);
-	}
+	xfs_destroy_ioend(ioend);
 }
 
 /*
@@ -264,6 +276,7 @@ xfs_alloc_ioend(
 	 */
 	atomic_set(&ioend->io_remaining, 1);
 	ioend->io_isasync = 0;
+	ioend->io_isdirect = 0;
 	ioend->io_error = 0;
 	ioend->io_list = NULL;
 	ioend->io_type = type;
@@ -274,6 +287,7 @@ xfs_alloc_ioend(
 	ioend->io_size = 0;
 	ioend->io_iocb = NULL;
 	ioend->io_result = 0;
+	ioend->io_append_trans = NULL;
 
 	INIT_WORK(&ioend->io_work, xfs_end_io);
 	return ioend;
@@ -384,14 +398,6 @@ xfs_submit_ioend_bio(
 	atomic_inc(&ioend->io_remaining);
 	bio->bi_private = ioend;
 	bio->bi_end_io = xfs_end_bio;
-
-	/*
-	 * If the I/O is beyond EOF we mark the inode dirty immediately
-	 * but don't update the inode size until I/O completion.
-	 */
-	if (xfs_ioend_new_eof(ioend))
-		xfs_mark_inode_dirty(XFS_I(ioend->io_inode));
-
 	submit_bio(wbc->sync_mode == WB_SYNC_ALL ? WRITE_SYNC : WRITE, bio);
 }
 
@@ -614,7 +620,7 @@ xfs_map_at_offset(
  * or delayed allocate extent.
  */
 STATIC int
-xfs_is_delayed_page(
+xfs_check_page_type(
 	struct page		*page,
 	unsigned int		type)
 {
@@ -628,11 +634,11 @@ xfs_is_delayed_page(
 		bh = head = page_buffers(page);
 		do {
 			if (buffer_unwritten(bh))
-				acceptable = (type == IO_UNWRITTEN);
+				acceptable += (type == IO_UNWRITTEN);
 			else if (buffer_delay(bh))
-				acceptable = (type == IO_DELALLOC);
+				acceptable += (type == IO_DELALLOC);
 			else if (buffer_dirty(bh) && buffer_mapped(bh))
-				acceptable = (type == IO_OVERWRITE);
+				acceptable += (type == IO_OVERWRITE);
 			else
 				break;
 		} while ((bh = bh->b_this_page) != head);
@@ -675,7 +681,7 @@ xfs_convert_page(
 		goto fail_unlock_page;
 	if (page->mapping != inode->i_mapping)
 		goto fail_unlock_page;
-	if (!xfs_is_delayed_page(page, (*ioendp)->io_type))
+	if (!xfs_check_page_type(page, (*ioendp)->io_type))
 		goto fail_unlock_page;
 
 	/*
@@ -825,7 +831,7 @@ xfs_aops_discard_page(
 	struct buffer_head	*bh, *head;
 	loff_t			offset = page_offset(page);
 
-	if (!xfs_is_delayed_page(page, IO_DELALLOC))
+	if (!xfs_check_page_type(page, IO_DELALLOC))
 		goto out_invalidate;
 
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
@@ -975,10 +981,15 @@ xfs_vm_writepage(
 				imap_valid = 0;
 			}
 		} else {
-			if (PageUptodate(page)) {
+			if (PageUptodate(page))
 				ASSERT(buffer_mapped(bh));
-				imap_valid = 0;
-			}
+			/*
+			 * This buffer is not uptodate and will not be
+			 * written to disk.  Ensure that we will put any
+			 * subsequent writeable buffers into a new
+			 * ioend.
+			 */
+			imap_valid = 0;
 			continue;
 		}
 
@@ -1038,8 +1049,20 @@ xfs_vm_writepage(
 				  wbc, end_index);
 	}
 
-	if (iohead)
+	if (iohead) {
+		/*
+		 * Reserve log space if we might write beyond the on-disk
+		 * inode size.
+		 */
+		if (ioend->io_type != IO_UNWRITTEN &&
+		    xfs_ioend_is_append(ioend)) {
+			err = xfs_setfilesize_trans_alloc(ioend);
+			if (err)
+				goto error;
+		}
+
 		xfs_submit_ioend(wbc, iohead);
+	}
 
 	return 0;
 
@@ -1125,7 +1148,14 @@ __xfs_get_blocks(
 	if (!create && direct && offset >= i_size_read(inode))
 		return 0;
 
-	if (create) {
+	/*
+	 * Direct I/O is usually done on preallocated files, so try getting
+	 * a block mapping without an exclusive lock first.  For buffered
+	 * writes we already have the exclusive iolock anyway, so avoiding
+	 * a lock roundtrip here by taking the ilock exclusive from the
+	 * beginning is a useful micro optimization.
+	 */
+	if (create && !direct) {
 		lockmode = XFS_ILOCK_EXCL;
 		xfs_ilock(ip, lockmode);
 	} else {
@@ -1147,23 +1177,45 @@ __xfs_get_blocks(
 	    (!nimaps ||
 	     (imap.br_startblock == HOLESTARTBLOCK ||
 	      imap.br_startblock == DELAYSTARTBLOCK))) {
-		if (direct) {
+		if (direct || xfs_get_extsz_hint(ip)) {
+			/*
+			 * Drop the ilock in preparation for starting the block
+			 * allocation transaction.  It will be retaken
+			 * exclusively inside xfs_iomap_write_direct for the
+			 * actual allocation.
+			 */
+			xfs_iunlock(ip, lockmode);
 			error = xfs_iomap_write_direct(ip, offset, size,
 						       &imap, nimaps);
+			if (error)
+				return -error;
+			new = 1;
 		} else {
+			/*
+			 * Delalloc reservations do not require a transaction,
+			 * we can go on without dropping the lock here. If we
+			 * are allocating a new delalloc block, make sure that
+			 * we set the new flag so that we mark the buffer new so
+			 * that we know that it is newly allocated if the write
+			 * fails.
+			 */
+			if (nimaps && imap.br_startblock == HOLESTARTBLOCK)
+				new = 1;
 			error = xfs_iomap_write_delay(ip, offset, size, &imap);
+			if (error)
+				goto out_unlock;
+
+			xfs_iunlock(ip, lockmode);
 		}
-		if (error)
-			goto out_unlock;
 
 		trace_xfs_get_blocks_alloc(ip, offset, size, 0, &imap);
 	} else if (nimaps) {
 		trace_xfs_get_blocks_found(ip, offset, size, 0, &imap);
+		xfs_iunlock(ip, lockmode);
 	} else {
 		trace_xfs_get_blocks_notfound(ip, offset, size);
 		goto out_unlock;
 	}
-	xfs_iunlock(ip, lockmode);
 
 	if (imap.br_startblock != HOLESTARTBLOCK &&
 	    imap.br_startblock != DELAYSTARTBLOCK) {
@@ -1279,6 +1331,15 @@ xfs_end_io_direct_write(
 	struct xfs_ioend	*ioend = iocb->private;
 
 	/*
+	 * While the generic direct I/O code updates the inode size, it does
+	 * so only after the end_io handler is called, which means our
+	 * end_io handler thinks the on-disk size is outside the in-core
+	 * size.  To prevent this just update it a little bit earlier here.
+	 */
+	if (offset + size > i_size_read(ioend->io_inode))
+		i_size_write(ioend->io_inode, offset + size);
+
+	/*
 	 * blockdev_direct_IO can return an error even after the I/O
 	 * completion handler was called.  Thus we need to protect
 	 * against double-freeing.
@@ -1310,17 +1371,32 @@ xfs_vm_direct_IO(
 {
 	struct inode		*inode = iocb->ki_filp->f_mapping->host;
 	struct block_device	*bdev = xfs_find_bdev_for_inode(inode);
+	struct xfs_ioend	*ioend = NULL;
 	ssize_t			ret;
 
 	if (rw & WRITE) {
-		iocb->private = xfs_alloc_ioend(inode, IO_DIRECT);
+		size_t size = iov_length(iov, nr_segs);
+
+		/*
+		 * We need to preallocate a transaction for a size update
+		 * here.  In the case that this write both updates the size
+		 * and converts at least on unwritten extent we will cancel
+		 * the still clean transaction after the I/O has finished.
+		 */
+		iocb->private = ioend = xfs_alloc_ioend(inode, IO_DIRECT);
+		if (offset + size > XFS_I(inode)->i_d.di_size) {
+			ret = xfs_setfilesize_trans_alloc(ioend);
+			if (ret)
+				goto out_destroy_ioend;
+			ioend->io_isdirect = 1;
+		}
 
 		ret = __blockdev_direct_IO(rw, iocb, inode, bdev, iov,
 					    offset, nr_segs,
 					    xfs_get_blocks_direct,
 					    xfs_end_io_direct_write, NULL, 0);
 		if (ret != -EIOCBQUEUED && iocb->private)
-			xfs_destroy_ioend(iocb->private);
+			goto out_trans_cancel;
 	} else {
 		ret = __blockdev_direct_IO(rw, iocb, inode, bdev, iov,
 					    offset, nr_segs,
@@ -1329,55 +1405,103 @@ xfs_vm_direct_IO(
 	}
 
 	return ret;
+
+out_trans_cancel:
+	if (ioend->io_append_trans) {
+		current_set_flags_nested(&ioend->io_append_trans->t_pflags,
+					 PF_FSTRANS);
+		xfs_trans_cancel(ioend->io_append_trans, 0);
+	}
+out_destroy_ioend:
+	xfs_destroy_ioend(ioend);
+	return ret;
+}
+
+/*
+ * Punch out the delalloc blocks we have already allocated.
+ *
+ * Don't bother with xfs_setattr given that nothing can have made it to disk yet
+ * as the page is still locked at this point.
+ */
+STATIC void
+xfs_vm_kill_delalloc_range(
+	struct inode		*inode,
+	loff_t			start,
+	loff_t			end)
+{
+	struct xfs_inode	*ip = XFS_I(inode);
+	xfs_fileoff_t		start_fsb;
+	xfs_fileoff_t		end_fsb;
+	int			error;
+
+	start_fsb = XFS_B_TO_FSB(ip->i_mount, start);
+	end_fsb = XFS_B_TO_FSB(ip->i_mount, end);
+	if (end_fsb <= start_fsb)
+		return;
+
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	error = xfs_bmap_punch_delalloc_range(ip, start_fsb,
+						end_fsb - start_fsb);
+	if (error) {
+		/* something screwed, just bail */
+		if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) {
+			xfs_alert(ip->i_mount,
+		"xfs_vm_write_failed: unable to clean up ino %lld",
+					ip->i_ino);
+		}
+	}
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 }
 
 STATIC void
 xfs_vm_write_failed(
-	struct address_space	*mapping,
-	loff_t			to)
+	struct inode		*inode,
+	struct page		*page,
+	loff_t			pos,
+	unsigned		len)
 {
-	struct inode		*inode = mapping->host;
+	loff_t			block_offset = pos & PAGE_MASK;
+	loff_t			block_start;
+	loff_t			block_end;
+	loff_t			from = pos & (PAGE_CACHE_SIZE - 1);
+	loff_t			to = from + len;
+	struct buffer_head	*bh, *head;
 
-	if (to > inode->i_size) {
-		/*
-		 * punch out the delalloc blocks we have already allocated. We
-		 * don't call xfs_setattr() to do this as we may be in the
-		 * middle of a multi-iovec write and so the vfs inode->i_size
-		 * will not match the xfs ip->i_size and so it will zero too
-		 * much. Hence we jus truncate the page cache to zero what is
-		 * necessary and punch the delalloc blocks directly.
-		 */
-		struct xfs_inode	*ip = XFS_I(inode);
-		xfs_fileoff_t		start_fsb;
-		xfs_fileoff_t		end_fsb;
-		int			error;
+	ASSERT(block_offset + from == pos);
 
-		truncate_pagecache(inode, to, inode->i_size);
+	head = page_buffers(page);
+	block_start = 0;
+	for (bh = head; bh != head || !block_start;
+	     bh = bh->b_this_page, block_start = block_end,
+				   block_offset += bh->b_size) {
+		block_end = block_start + bh->b_size;
 
-		/*
-		 * Check if there are any blocks that are outside of i_size
-		 * that need to be trimmed back.
-		 */
-		start_fsb = XFS_B_TO_FSB(ip->i_mount, inode->i_size) + 1;
-		end_fsb = XFS_B_TO_FSB(ip->i_mount, to);
-		if (end_fsb <= start_fsb)
-			return;
-
-		xfs_ilock(ip, XFS_ILOCK_EXCL);
-		error = xfs_bmap_punch_delalloc_range(ip, start_fsb,
-							end_fsb - start_fsb);
-		if (error) {
-			/* something screwed, just bail */
-			if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) {
-				xfs_alert(ip->i_mount,
-			"xfs_vm_write_failed: unable to clean up ino %lld",
-						ip->i_ino);
-			}
-		}
-		xfs_iunlock(ip, XFS_ILOCK_EXCL);
+		/* skip buffers before the write */
+		if (block_end <= from)
+			continue;
+
+		/* if the buffer is after the write, we're done */
+		if (block_start >= to)
+			break;
+
+		if (!buffer_delay(bh))
+			continue;
+
+		if (!buffer_new(bh) && block_offset < i_size_read(inode))
+			continue;
+
+		xfs_vm_kill_delalloc_range(inode, block_offset,
+					   block_offset + bh->b_size);
 	}
+
 }
 
+/*
+ * This used to call block_write_begin(), but it unlocks and releases the page
+ * on error, and we need that page to be able to punch stale delalloc blocks out
+ * on failure. hence we copy-n-waste it here and call xfs_vm_write_failed() at
+ * the appropriate point.
+ */
 STATIC int
 xfs_vm_write_begin(
 	struct file		*file,
@@ -1388,15 +1512,40 @@ xfs_vm_write_begin(
 	struct page		**pagep,
 	void			**fsdata)
 {
-	int			ret;
+	pgoff_t			index = pos >> PAGE_CACHE_SHIFT;
+	struct page		*page;
+	int			status;
 
-	ret = block_write_begin(mapping, pos, len, flags | AOP_FLAG_NOFS,
-				pagep, xfs_get_blocks);
-	if (unlikely(ret))
-		xfs_vm_write_failed(mapping, pos + len);
-	return ret;
+	ASSERT(len <= PAGE_CACHE_SIZE);
+
+	page = grab_cache_page_write_begin(mapping, index,
+					   flags | AOP_FLAG_NOFS);
+	if (!page)
+		return -ENOMEM;
+
+	status = __block_write_begin(page, pos, len, xfs_get_blocks);
+	if (unlikely(status)) {
+		struct inode	*inode = mapping->host;
+
+		xfs_vm_write_failed(inode, page, pos, len);
+		unlock_page(page);
+
+		if (pos + len > i_size_read(inode))
+			truncate_pagecache(inode, pos + len, i_size_read(inode));
+
+		page_cache_release(page);
+		page = NULL;
+	}
+
+	*pagep = page;
+	return status;
 }
 
+/*
+ * On failure, we only need to kill delalloc blocks beyond EOF because they
+ * will never be written. For blocks within EOF, generic_write_end() zeros them
+ * so they are safe to leave alone and be written with all the other valid data.
+ */
 STATIC int
 xfs_vm_write_end(
 	struct file		*file,
@@ -1409,9 +1558,19 @@ xfs_vm_write_end(
 {
 	int			ret;
 
+	ASSERT(len <= PAGE_CACHE_SIZE);
+
 	ret = generic_write_end(file, mapping, pos, len, copied, page, fsdata);
-	if (unlikely(ret < len))
-		xfs_vm_write_failed(mapping, pos + len);
+	if (unlikely(ret < len)) {
+		struct inode	*inode = mapping->host;
+		size_t		isize = i_size_read(inode);
+		loff_t		to = pos + len;
+
+		if (to > isize) {
+			truncate_pagecache(inode, to, isize);
+			xfs_vm_kill_delalloc_range(inode, isize, to);
+		}
+	}
 	return ret;
 }
 
diff --git a/fs/xfs/xfs_aops.h b/fs/xfs/xfs_aops.h
index 116dd5c3703..84eafbcb0d9 100644
--- a/fs/xfs/xfs_aops.h
+++ b/fs/xfs/xfs_aops.h
@@ -18,8 +18,6 @@
 #ifndef __XFS_AOPS_H__
 #define __XFS_AOPS_H__
 
-extern struct workqueue_struct *xfsdatad_workqueue;
-extern struct workqueue_struct *xfsconvertd_workqueue;
 extern mempool_t *xfs_ioend_pool;
 
 /*
@@ -48,12 +46,14 @@ typedef struct xfs_ioend {
 	int			io_error;	/* I/O error code */
 	atomic_t		io_remaining;	/* hold count */
 	unsigned int		io_isasync : 1;	/* needs aio_complete */
+	unsigned int		io_isdirect : 1;/* direct I/O */
 	struct inode		*io_inode;	/* file being written to */
 	struct buffer_head	*io_buffer_head;/* buffer linked list head */
 	struct buffer_head	*io_buffer_tail;/* buffer linked list tail */
 	size_t			io_size;	/* size of the extent */
 	xfs_off_t		io_offset;	/* offset in the file */
 	struct work_struct	io_work;	/* xfsdatad work queue */
+	struct xfs_trans	*io_append_trans;/* xact. for size update */
 	struct kiocb		*io_iocb;
 	int			io_result;
 } xfs_ioend_t;
diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c
index 1e5d97f86ea..a17ff01b5ad 100644
--- a/fs/xfs/xfs_attr.c
+++ b/fs/xfs/xfs_attr.c
@@ -21,7 +21,6 @@
 #include "xfs_types.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -39,7 +38,6 @@
 #include "xfs_error.h"
 #include "xfs_quota.h"
 #include "xfs_trans_space.h"
-#include "xfs_rw.h"
 #include "xfs_vnodeops.h"
 #include "xfs_trace.h"
 
@@ -827,10 +825,6 @@ xfs_attr_inactive(xfs_inode_t *dp)
 	if (error)
 		goto out;
 
-	/*
-	 * Commit the last in the sequence of transactions.
-	 */
-	xfs_trans_log_inode(trans, dp, XFS_ILOG_CORE);
 	error = xfs_trans_commit(trans, XFS_TRANS_RELEASE_LOG_RES);
 	xfs_iunlock(dp, XFS_ILOCK_EXCL);
 
@@ -857,6 +851,8 @@ xfs_attr_shortform_addname(xfs_da_args_t *args)
 {
 	int newsize, forkoff, retval;
 
+	trace_xfs_attr_sf_addname(args);
+
 	retval = xfs_attr_shortform_lookup(args);
 	if ((args->flags & ATTR_REPLACE) && (retval == ENOATTR)) {
 		return(retval);
@@ -900,6 +896,8 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 	xfs_dabuf_t *bp;
 	int retval, error, committed, forkoff;
 
+	trace_xfs_attr_leaf_addname(args);
+
 	/*
 	 * Read the (only) block in the attribute list in.
 	 */
@@ -924,6 +922,9 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
 			xfs_da_brelse(args->trans, bp);
 			return(retval);
 		}
+
+		trace_xfs_attr_leaf_replace(args);
+
 		args->op_flags |= XFS_DA_OP_RENAME;	/* an atomic rename */
 		args->blkno2 = args->blkno;		/* set 2nd entry info*/
 		args->index2 = args->index;
@@ -1094,6 +1095,8 @@ xfs_attr_leaf_removename(xfs_da_args_t *args)
 	xfs_dabuf_t *bp;
 	int error, committed, forkoff;
 
+	trace_xfs_attr_leaf_removename(args);
+
 	/*
 	 * Remove the attribute.
 	 */
@@ -1227,6 +1230,8 @@ xfs_attr_node_addname(xfs_da_args_t *args)
 	xfs_mount_t *mp;
 	int committed, retval, error;
 
+	trace_xfs_attr_node_addname(args);
+
 	/*
 	 * Fill in bucket of arguments/results/context to carry around.
 	 */
@@ -1253,6 +1258,9 @@ restart:
 	} else if (retval == EEXIST) {
 		if (args->flags & ATTR_CREATE)
 			goto out;
+
+		trace_xfs_attr_node_replace(args);
+
 		args->op_flags |= XFS_DA_OP_RENAME;	/* atomic rename op */
 		args->blkno2 = args->blkno;		/* set 2nd entry info*/
 		args->index2 = args->index;
@@ -1484,6 +1492,8 @@ xfs_attr_node_removename(xfs_da_args_t *args)
 	xfs_dabuf_t *bp;
 	int retval, error, committed, forkoff;
 
+	trace_xfs_attr_node_removename(args);
+
 	/*
 	 * Tie a string around our finger to remind us where we are.
 	 */
@@ -1975,14 +1985,12 @@ xfs_attr_rmtval_get(xfs_da_args_t *args)
 			       (map[i].br_startblock != HOLESTARTBLOCK));
 			dblkno = XFS_FSB_TO_DADDR(mp, map[i].br_startblock);
 			blkcnt = XFS_FSB_TO_BB(mp, map[i].br_blockcount);
-			error = xfs_read_buf(mp, mp->m_ddev_targp, dblkno,
-					     blkcnt, XBF_LOCK | XBF_DONT_BLOCK,
-					     &bp);
+			error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp,
+						   dblkno, blkcnt, 0, &bp);
 			if (error)
 				return(error);
 
-			tmp = (valuelen < XFS_BUF_SIZE(bp))
-				? valuelen : XFS_BUF_SIZE(bp);
+			tmp = min_t(int, valuelen, BBTOB(bp->b_length));
 			xfs_buf_iomove(bp, 0, tmp, dst, XBRW_READ);
 			xfs_buf_relse(bp);
 			dst += tmp;
@@ -2085,6 +2093,8 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
 	lblkno = args->rmtblkno;
 	valuelen = args->valuelen;
 	while (valuelen > 0) {
+		int buflen;
+
 		/*
 		 * Try to remember where we decided to put the value.
 		 */
@@ -2102,15 +2112,16 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
 		dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock),
 		blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount);
 
-		bp = xfs_buf_get(mp->m_ddev_targp, dblkno, blkcnt,
-				 XBF_LOCK | XBF_DONT_BLOCK);
+		bp = xfs_buf_get(mp->m_ddev_targp, dblkno, blkcnt, 0);
 		if (!bp)
 			return ENOMEM;
-		tmp = (valuelen < XFS_BUF_SIZE(bp)) ? valuelen :
-							XFS_BUF_SIZE(bp);
+
+		buflen = BBTOB(bp->b_length);
+		tmp = min_t(int, valuelen, buflen);
 		xfs_buf_iomove(bp, 0, tmp, src, XBRW_WRITE);
-		if (tmp < XFS_BUF_SIZE(bp))
-			xfs_buf_zero(bp, tmp, XFS_BUF_SIZE(bp) - tmp);
+		if (tmp < buflen)
+			xfs_buf_zero(bp, tmp, buflen - tmp);
+
 		error = xfs_bwrite(bp);	/* GROT: NOTE: synchronous write */
 		xfs_buf_relse(bp);
 		if (error)
diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c
index c1b55e59655..7d89d800f51 100644
--- a/fs/xfs/xfs_attr_leaf.c
+++ b/fs/xfs/xfs_attr_leaf.c
@@ -20,7 +20,6 @@
 #include "xfs_types.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -235,6 +234,8 @@ xfs_attr_shortform_create(xfs_da_args_t *args)
 	xfs_inode_t *dp;
 	xfs_ifork_t *ifp;
 
+	trace_xfs_attr_sf_create(args);
+
 	dp = args->dp;
 	ASSERT(dp != NULL);
 	ifp = dp->i_afp;
@@ -268,13 +269,11 @@ xfs_attr_shortform_add(xfs_da_args_t *args, int forkoff)
 	xfs_inode_t *dp;
 	xfs_ifork_t *ifp;
 
+	trace_xfs_attr_sf_add(args);
+
 	dp = args->dp;
 	mp = dp->i_mount;
 	dp->i_d.di_forkoff = forkoff;
-	dp->i_df.if_ext_max =
-		XFS_IFORK_DSIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t);
-	dp->i_afp->if_ext_max =
-		XFS_IFORK_ASIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t);
 
 	ifp = dp->i_afp;
 	ASSERT(ifp->if_flags & XFS_IFINLINE);
@@ -326,7 +325,6 @@ xfs_attr_fork_reset(
 	ASSERT(ip->i_d.di_anextents == 0);
 	ASSERT(ip->i_afp == NULL);
 
-	ip->i_df.if_ext_max = XFS_IFORK_DSIZE(ip) / sizeof(xfs_bmbt_rec_t);
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 }
 
@@ -342,6 +340,8 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
 	xfs_mount_t *mp;
 	xfs_inode_t *dp;
 
+	trace_xfs_attr_sf_remove(args);
+
 	dp = args->dp;
 	mp = dp->i_mount;
 	base = sizeof(xfs_attr_sf_hdr_t);
@@ -389,10 +389,6 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
 				(args->op_flags & XFS_DA_OP_ADDNAME) ||
 				!(mp->m_flags & XFS_MOUNT_ATTR2) ||
 				dp->i_d.di_format == XFS_DINODE_FMT_BTREE);
-		dp->i_afp->if_ext_max =
-			XFS_IFORK_ASIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t);
-		dp->i_df.if_ext_max =
-			XFS_IFORK_DSIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t);
 		xfs_trans_log_inode(args->trans, dp,
 					XFS_ILOG_CORE | XFS_ILOG_ADATA);
 	}
@@ -414,6 +410,8 @@ xfs_attr_shortform_lookup(xfs_da_args_t *args)
 	int i;
 	xfs_ifork_t *ifp;
 
+	trace_xfs_attr_sf_lookup(args);
+
 	ifp = args->dp->i_afp;
 	ASSERT(ifp->if_flags & XFS_IFINLINE);
 	sf = (xfs_attr_shortform_t *)ifp->if_u1.if_data;
@@ -485,6 +483,8 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
 	xfs_dabuf_t *bp;
 	xfs_ifork_t *ifp;
 
+	trace_xfs_attr_sf_to_leaf(args);
+
 	dp = args->dp;
 	ifp = dp->i_afp;
 	sf = (xfs_attr_shortform_t *)ifp->if_u1.if_data;
@@ -784,6 +784,8 @@ xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args, int forkoff)
 	char *tmpbuffer;
 	int error, i;
 
+	trace_xfs_attr_leaf_to_sf(args);
+
 	dp = args->dp;
 	tmpbuffer = kmem_alloc(XFS_LBSIZE(dp->i_mount), KM_SLEEP);
 	ASSERT(tmpbuffer != NULL);
@@ -857,6 +859,8 @@ xfs_attr_leaf_to_node(xfs_da_args_t *args)
 	xfs_dablk_t blkno;
 	int error;
 
+	trace_xfs_attr_leaf_to_node(args);
+
 	dp = args->dp;
 	bp1 = bp2 = NULL;
 	error = xfs_da_grow_inode(args, &blkno);
@@ -920,6 +924,8 @@ xfs_attr_leaf_create(xfs_da_args_t *args, xfs_dablk_t blkno, xfs_dabuf_t **bpp)
 	xfs_dabuf_t *bp;
 	int error;
 
+	trace_xfs_attr_leaf_create(args);
+
 	dp = args->dp;
 	ASSERT(dp != NULL);
 	error = xfs_da_get_buf(args->trans, args->dp, blkno, -1, &bp,
@@ -957,6 +963,8 @@ xfs_attr_leaf_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
 	xfs_dablk_t blkno;
 	int error;
 
+	trace_xfs_attr_leaf_split(state->args);
+
 	/*
 	 * Allocate space for a new leaf node.
 	 */
@@ -986,10 +994,13 @@ xfs_attr_leaf_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
 	 *
 	 * Insert the "new" entry in the correct block.
 	 */
-	if (state->inleaf)
+	if (state->inleaf) {
+		trace_xfs_attr_leaf_add_old(state->args);
 		error = xfs_attr_leaf_add(oldblk->bp, state->args);
-	else
+	} else {
+		trace_xfs_attr_leaf_add_new(state->args);
 		error = xfs_attr_leaf_add(newblk->bp, state->args);
+	}
 
 	/*
 	 * Update last hashval in each block since we added the name.
@@ -1010,6 +1021,8 @@ xfs_attr_leaf_add(xfs_dabuf_t *bp, xfs_da_args_t *args)
 	xfs_attr_leaf_map_t *map;
 	int tablesize, entsize, sum, tmp, i;
 
+	trace_xfs_attr_leaf_add(args);
+
 	leaf = bp->data;
 	ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
 	ASSERT((args->index >= 0)
@@ -1137,8 +1150,6 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex)
 	       (be32_to_cpu(entry->hashval) <= be32_to_cpu((entry+1)->hashval)));
 
 	/*
-	 * Copy the attribute name and value into the new space.
-	 *
 	 * For "remote" attribute values, simply note that we need to
 	 * allocate space for the "remote" value.  We can't actually
 	 * allocate the extents in this transaction, and we can't decide
@@ -1274,6 +1285,8 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
 	ASSERT(leaf2->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
 	args = state->args;
 
+	trace_xfs_attr_leaf_rebalance(args);
+
 	/*
 	 * Check ordering of blocks, reverse if it makes things simpler.
 	 *
@@ -1819,6 +1832,8 @@ xfs_attr_leaf_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
 	xfs_mount_t *mp;
 	char *tmpbuffer;
 
+	trace_xfs_attr_leaf_unbalance(state->args);
+
 	/*
 	 * Set up environment.
 	 */
@@ -1928,6 +1943,8 @@ xfs_attr_leaf_lookup_int(xfs_dabuf_t *bp, xfs_da_args_t *args)
 	int probe, span;
 	xfs_dahash_t hashval;
 
+	trace_xfs_attr_leaf_lookup(args);
+
 	leaf = bp->data;
 	ASSERT(leaf->hdr.info.magic == cpu_to_be16(XFS_ATTR_LEAF_MAGIC));
 	ASSERT(be16_to_cpu(leaf->hdr.count)
@@ -2454,6 +2471,7 @@ xfs_attr_leaf_clearflag(xfs_da_args_t *args)
 	char *name;
 #endif /* DEBUG */
 
+	trace_xfs_attr_leaf_clearflag(args);
 	/*
 	 * Set up the operation.
 	 */
@@ -2518,6 +2536,8 @@ xfs_attr_leaf_setflag(xfs_da_args_t *args)
 	xfs_dabuf_t *bp;
 	int error;
 
+	trace_xfs_attr_leaf_setflag(args);
+
 	/*
 	 * Set up the operation.
 	 */
@@ -2574,6 +2594,8 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args)
 	char *name1, *name2;
 #endif /* DEBUG */
 
+	trace_xfs_attr_leaf_flipflags(args);
+
 	/*
 	 * Read the block containing the "old" attr
 	 */
@@ -2960,7 +2982,7 @@ xfs_attr_leaf_freextent(xfs_trans_t **trans, xfs_inode_t *dp,
 						map.br_blockcount);
 			bp = xfs_trans_get_buf(*trans,
 					dp->i_mount->m_ddev_targp,
-					dblkno, dblkcnt, XBF_LOCK);
+					dblkno, dblkcnt, 0);
 			if (!bp)
 				return ENOMEM;
 			xfs_trans_binval(*trans, bp);
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c
index d0ab7883705..58b815ec8c9 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/xfs_bmap.c
@@ -41,7 +41,6 @@
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
 #include "xfs_attr_leaf.h"
-#include "xfs_rw.h"
 #include "xfs_quota.h"
 #include "xfs_trans_space.h"
 #include "xfs_buf_item.h"
@@ -249,7 +248,27 @@ xfs_bmbt_lookup_ge(
 }
 
 /*
-* Update the record referred to by cur to the value given
+ * Check if the inode needs to be converted to btree format.
+ */
+static inline bool xfs_bmap_needs_btree(struct xfs_inode *ip, int whichfork)
+{
+	return XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS &&
+		XFS_IFORK_NEXTENTS(ip, whichfork) >
+			XFS_IFORK_MAXEXT(ip, whichfork);
+}
+
+/*
+ * Check if the inode should be converted to extent format.
+ */
+static inline bool xfs_bmap_wants_extents(struct xfs_inode *ip, int whichfork)
+{
+	return XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE &&
+		XFS_IFORK_NEXTENTS(ip, whichfork) <=
+			XFS_IFORK_MAXEXT(ip, whichfork);
+}
+
+/*
+ * Update the record referred to by cur to the value given
  * by [off, bno, len, state].
  * This either works (return 0) or gets an EFSCORRUPTED error.
  */
@@ -683,8 +702,8 @@ xfs_bmap_add_extent_delay_real(
 				goto done;
 			XFS_WANT_CORRUPTED_GOTO(i == 1, done);
 		}
-		if (bma->ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
-		    bma->ip->i_d.di_nextents > bma->ip->i_df.if_ext_max) {
+
+		if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
 			error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
 					bma->firstblock, bma->flist,
 					&bma->cur, 1, &tmp_rval, XFS_DATA_FORK);
@@ -767,8 +786,8 @@ xfs_bmap_add_extent_delay_real(
 				goto done;
 			XFS_WANT_CORRUPTED_GOTO(i == 1, done);
 		}
-		if (bma->ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
-		    bma->ip->i_d.di_nextents > bma->ip->i_df.if_ext_max) {
+
+		if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
 			error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
 				bma->firstblock, bma->flist, &bma->cur, 1,
 				&tmp_rval, XFS_DATA_FORK);
@@ -836,8 +855,8 @@ xfs_bmap_add_extent_delay_real(
 				goto done;
 			XFS_WANT_CORRUPTED_GOTO(i == 1, done);
 		}
-		if (bma->ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
-		    bma->ip->i_d.di_nextents > bma->ip->i_df.if_ext_max) {
+
+		if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
 			error = xfs_bmap_extents_to_btree(bma->tp, bma->ip,
 					bma->firstblock, bma->flist, &bma->cur,
 					1, &tmp_rval, XFS_DATA_FORK);
@@ -884,8 +903,7 @@ xfs_bmap_add_extent_delay_real(
 	}
 
 	/* convert to a btree if necessary */
-	if (XFS_IFORK_FORMAT(bma->ip, XFS_DATA_FORK) == XFS_DINODE_FMT_EXTENTS &&
-	    XFS_IFORK_NEXTENTS(bma->ip, XFS_DATA_FORK) > ifp->if_ext_max) {
+	if (xfs_bmap_needs_btree(bma->ip, XFS_DATA_FORK)) {
 		int	tmp_logflags;	/* partial log flag return val */
 
 		ASSERT(bma->cur == NULL);
@@ -1421,8 +1439,7 @@ xfs_bmap_add_extent_unwritten_real(
 	}
 
 	/* convert to a btree if necessary */
-	if (XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) == XFS_DINODE_FMT_EXTENTS &&
-	    XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) > ifp->if_ext_max) {
+	if (xfs_bmap_needs_btree(ip, XFS_DATA_FORK)) {
 		int	tmp_logflags;	/* partial log flag return val */
 
 		ASSERT(cur == NULL);
@@ -1812,8 +1829,7 @@ xfs_bmap_add_extent_hole_real(
 	}
 
 	/* convert to a btree if necessary */
-	if (XFS_IFORK_FORMAT(bma->ip, whichfork) == XFS_DINODE_FMT_EXTENTS &&
-	    XFS_IFORK_NEXTENTS(bma->ip, whichfork) > ifp->if_ext_max) {
+	if (xfs_bmap_needs_btree(bma->ip, whichfork)) {
 		int	tmp_logflags;	/* partial log flag return val */
 
 		ASSERT(bma->cur == NULL);
@@ -3037,8 +3053,7 @@ xfs_bmap_extents_to_btree(
 
 	ifp = XFS_IFORK_PTR(ip, whichfork);
 	ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS);
-	ASSERT(ifp->if_ext_max ==
-	       XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t));
+
 	/*
 	 * Make space in the inode incore.
 	 */
@@ -3184,13 +3199,8 @@ xfs_bmap_forkoff_reset(
 	    ip->i_d.di_format != XFS_DINODE_FMT_BTREE) {
 		uint	dfl_forkoff = xfs_default_attroffset(ip) >> 3;
 
-		if (dfl_forkoff > ip->i_d.di_forkoff) {
+		if (dfl_forkoff > ip->i_d.di_forkoff)
 			ip->i_d.di_forkoff = dfl_forkoff;
-			ip->i_df.if_ext_max =
-				XFS_IFORK_DSIZE(ip) / sizeof(xfs_bmbt_rec_t);
-			ip->i_afp->if_ext_max =
-				XFS_IFORK_ASIZE(ip) / sizeof(xfs_bmbt_rec_t);
-		}
 	}
 }
 
@@ -3430,8 +3440,6 @@ xfs_bmap_add_attrfork(
 	int			error;		/* error return value */
 
 	ASSERT(XFS_IFORK_Q(ip) == 0);
-	ASSERT(ip->i_df.if_ext_max ==
-	       XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t));
 
 	mp = ip->i_mount;
 	ASSERT(!XFS_NOT_DQATTACHED(mp, ip));
@@ -3486,12 +3494,9 @@ xfs_bmap_add_attrfork(
 		error = XFS_ERROR(EINVAL);
 		goto error1;
 	}
-	ip->i_df.if_ext_max =
-		XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
+
 	ASSERT(ip->i_afp == NULL);
 	ip->i_afp = kmem_zone_zalloc(xfs_ifork_zone, KM_SLEEP);
-	ip->i_afp->if_ext_max =
-		XFS_IFORK_ASIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
 	ip->i_afp->if_flags = XFS_IFEXTENTS;
 	logflags = 0;
 	xfs_bmap_init(&flist, &firstblock);
@@ -3535,20 +3540,17 @@ xfs_bmap_add_attrfork(
 		} else
 			spin_unlock(&mp->m_sb_lock);
 	}
-	if ((error = xfs_bmap_finish(&tp, &flist, &committed)))
+
+	error = xfs_bmap_finish(&tp, &flist, &committed);
+	if (error)
 		goto error2;
-	error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
-	ASSERT(ip->i_df.if_ext_max ==
-	       XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t));
-	return error;
+	return xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
 error2:
 	xfs_bmap_cancel(&flist);
 error1:
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 error0:
 	xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES|XFS_TRANS_ABORT);
-	ASSERT(ip->i_df.if_ext_max ==
-	       XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t));
 	return error;
 }
 
@@ -3994,11 +3996,8 @@ xfs_bmap_one_block(
 	xfs_bmbt_irec_t	s;		/* internal version of extent */
 
 #ifndef DEBUG
-	if (whichfork == XFS_DATA_FORK) {
-		return S_ISREG(ip->i_d.di_mode) ?
-			(ip->i_size == ip->i_mount->m_sb.sb_blocksize) :
-			(ip->i_d.di_size == ip->i_mount->m_sb.sb_blocksize);
-	}
+	if (whichfork == XFS_DATA_FORK)
+		return XFS_ISIZE(ip) == ip->i_mount->m_sb.sb_blocksize;
 #endif	/* !DEBUG */
 	if (XFS_IFORK_NEXTENTS(ip, whichfork) != 1)
 		return 0;
@@ -4010,7 +4009,7 @@ xfs_bmap_one_block(
 	xfs_bmbt_get_all(ep, &s);
 	rval = s.br_startoff == 0 && s.br_blockcount == 1;
 	if (rval && whichfork == XFS_DATA_FORK)
-		ASSERT(ip->i_size == ip->i_mount->m_sb.sb_blocksize);
+		ASSERT(XFS_ISIZE(ip) == ip->i_mount->m_sb.sb_blocksize);
 	return rval;
 }
 
@@ -4379,8 +4378,6 @@ xfs_bmapi_read(
 	XFS_STATS_INC(xs_blk_mapr);
 
 	ifp = XFS_IFORK_PTR(ip, whichfork);
-	ASSERT(ifp->if_ext_max ==
-	       XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t));
 
 	if (!(ifp->if_flags & XFS_IFEXTENTS)) {
 		error = xfs_iread_extents(NULL, ip, whichfork);
@@ -4529,7 +4526,7 @@ out_unreserve_blocks:
 		xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS, alen, 0);
 out_unreserve_quota:
 	if (XFS_IS_QUOTA_ON(mp))
-		xfs_trans_unreserve_quota_nblks(NULL, ip, alen, 0, rt ?
+		xfs_trans_unreserve_quota_nblks(NULL, ip, (long)alen, 0, rt ?
 				XFS_QMOPT_RES_RTBLKS : XFS_QMOPT_RES_REGBLKS);
 	return error;
 }
@@ -4871,8 +4868,6 @@ xfs_bmapi_write(
 		return XFS_ERROR(EIO);
 
 	ifp = XFS_IFORK_PTR(ip, whichfork);
-	ASSERT(ifp->if_ext_max ==
-	       XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t));
 
 	XFS_STATS_INC(xs_blk_mapw);
 
@@ -4981,8 +4976,7 @@ xfs_bmapi_write(
 	/*
 	 * Transform from btree to extents, give it cur.
 	 */
-	if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE &&
-	    XFS_IFORK_NEXTENTS(ip, whichfork) <= ifp->if_ext_max) {
+	if (xfs_bmap_wants_extents(ip, whichfork)) {
 		int		tmp_logflags = 0;
 
 		ASSERT(bma.cur);
@@ -4992,10 +4986,10 @@ xfs_bmapi_write(
 		if (error)
 			goto error0;
 	}
-	ASSERT(ifp->if_ext_max ==
-	       XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t));
+
 	ASSERT(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE ||
-	       XFS_IFORK_NEXTENTS(ip, whichfork) > ifp->if_ext_max);
+	       XFS_IFORK_NEXTENTS(ip, whichfork) >
+		XFS_IFORK_MAXEXT(ip, whichfork));
 	error = 0;
 error0:
 	/*
@@ -5095,8 +5089,7 @@ xfs_bunmapi(
 
 	ASSERT(len > 0);
 	ASSERT(nexts >= 0);
-	ASSERT(ifp->if_ext_max ==
-	       XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t));
+
 	if (!(ifp->if_flags & XFS_IFEXTENTS) &&
 	    (error = xfs_iread_extents(tp, ip, whichfork)))
 		return error;
@@ -5130,6 +5123,15 @@ xfs_bunmapi(
 		cur->bc_private.b.flags = 0;
 	} else
 		cur = NULL;
+
+	if (isrt) {
+		/*
+		 * Synchronize by locking the bitmap inode.
+		 */
+		xfs_ilock(mp->m_rbmip, XFS_ILOCK_EXCL);
+		xfs_trans_ijoin(tp, mp->m_rbmip, XFS_ILOCK_EXCL);
+	}
+
 	extno = 0;
 	while (bno != (xfs_fileoff_t)-1 && bno >= start && lastx >= 0 &&
 	       (nexts == 0 || extno < nexts)) {
@@ -5322,7 +5324,8 @@ xfs_bunmapi(
 		 */
 		if (!wasdel && xfs_trans_get_block_res(tp) == 0 &&
 		    XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS &&
-		    XFS_IFORK_NEXTENTS(ip, whichfork) >= ifp->if_ext_max &&
+		    XFS_IFORK_NEXTENTS(ip, whichfork) >= /* Note the >= */
+			XFS_IFORK_MAXEXT(ip, whichfork) &&
 		    del.br_startoff > got.br_startoff &&
 		    del.br_startoff + del.br_blockcount <
 		    got.br_startoff + got.br_blockcount) {
@@ -5353,13 +5356,11 @@ nodelete:
 		}
 	}
 	*done = bno == (xfs_fileoff_t)-1 || bno < start || lastx < 0;
-	ASSERT(ifp->if_ext_max ==
-	       XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t));
+
 	/*
 	 * Convert to a btree if necessary.
 	 */
-	if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS &&
-	    XFS_IFORK_NEXTENTS(ip, whichfork) > ifp->if_ext_max) {
+	if (xfs_bmap_needs_btree(ip, whichfork)) {
 		ASSERT(cur == NULL);
 		error = xfs_bmap_extents_to_btree(tp, ip, firstblock, flist,
 			&cur, 0, &tmp_logflags, whichfork);
@@ -5370,8 +5371,7 @@ nodelete:
 	/*
 	 * transform from btree to extents, give it cur
 	 */
-	else if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE &&
-		 XFS_IFORK_NEXTENTS(ip, whichfork) <= ifp->if_ext_max) {
+	else if (xfs_bmap_wants_extents(ip, whichfork)) {
 		ASSERT(cur != NULL);
 		error = xfs_bmap_btree_to_extents(tp, ip, cur, &tmp_logflags,
 			whichfork);
@@ -5382,8 +5382,6 @@ nodelete:
 	/*
 	 * transform from extents to local?
 	 */
-	ASSERT(ifp->if_ext_max ==
-	       XFS_IFORK_SIZE(ip, whichfork) / (uint)sizeof(xfs_bmbt_rec_t));
 	error = 0;
 error0:
 	/*
@@ -5434,7 +5432,7 @@ xfs_getbmapx_fix_eof_hole(
 	if (startblock == HOLESTARTBLOCK) {
 		mp = ip->i_mount;
 		out->bmv_block = -1;
-		fixlen = XFS_FSB_TO_BB(mp, XFS_B_TO_FSB(mp, ip->i_size));
+		fixlen = XFS_FSB_TO_BB(mp, XFS_B_TO_FSB(mp, XFS_ISIZE(ip)));
 		fixlen -= out->bmv_offset;
 		if (prealloced && out->bmv_offset + out->bmv_length == end) {
 			/* Came to hole at EOF. Trim it. */
@@ -5522,7 +5520,7 @@ xfs_getbmap(
 			fixlen = XFS_MAXIOFFSET(mp);
 		} else {
 			prealloced = 0;
-			fixlen = ip->i_size;
+			fixlen = XFS_ISIZE(ip);
 		}
 	}
 
@@ -5546,12 +5544,16 @@ xfs_getbmap(
 	if (bmv->bmv_count > ULONG_MAX / sizeof(struct getbmapx))
 		return XFS_ERROR(ENOMEM);
 	out = kmem_zalloc(bmv->bmv_count * sizeof(struct getbmapx), KM_MAYFAIL);
-	if (!out)
-		return XFS_ERROR(ENOMEM);
+	if (!out) {
+		out = kmem_zalloc_large(bmv->bmv_count *
+					sizeof(struct getbmapx));
+		if (!out)
+			return XFS_ERROR(ENOMEM);
+	}
 
 	xfs_ilock(ip, XFS_IOLOCK_SHARED);
 	if (whichfork == XFS_DATA_FORK && !(iflags & BMV_IF_DELALLOC)) {
-		if (ip->i_delayed_blks || ip->i_size > ip->i_d.di_size) {
+		if (ip->i_delayed_blks || XFS_ISIZE(ip) > ip->i_d.di_size) {
 			error = xfs_flush_pages(ip, 0, -1, 0, FI_REMAPF);
 			if (error)
 				goto out_unlock_iolock;
@@ -5618,8 +5620,20 @@ xfs_getbmap(
 				XFS_FSB_TO_BB(mp, map[i].br_blockcount);
 			out[cur_ext].bmv_unused1 = 0;
 			out[cur_ext].bmv_unused2 = 0;
-			ASSERT(((iflags & BMV_IF_DELALLOC) != 0) ||
-			      (map[i].br_startblock != DELAYSTARTBLOCK));
+
+			/*
+			 * delayed allocation extents that start beyond EOF can
+			 * occur due to speculative EOF allocation when the
+			 * delalloc extent is larger than the largest freespace
+			 * extent at conversion time. These extents cannot be
+			 * converted by data writeback, so can exist here even
+			 * if we are not supposed to be finding delalloc
+			 * extents.
+			 */
+			if (map[i].br_startblock == DELAYSTARTBLOCK &&
+			    map[i].br_startoff <= XFS_B_TO_FSB(mp, XFS_ISIZE(ip)))
+				ASSERT((iflags & BMV_IF_DELALLOC) != 0);
+
                         if (map[i].br_startblock == HOLESTARTBLOCK &&
 			    whichfork == XFS_ATTR_FORK) {
 				/* came to the end of attribute fork */
@@ -5671,7 +5685,10 @@ xfs_getbmap(
 			break;
 	}
 
-	kmem_free(out);
+	if (is_vmalloc_addr(out))
+		kmem_free_large(out);
+	else
+		kmem_free(out);
 	return error;
 }
 
@@ -6151,3 +6168,16 @@ next_block:
 
 	return error;
 }
+
+/*
+ * Convert the given file system block to a disk block.  We have to treat it
+ * differently based on whether the file is a real time file or not, because the
+ * bmap code does.
+ */
+xfs_daddr_t
+xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb)
+{
+	return (XFS_IS_REALTIME_INODE(ip) ? \
+		 (xfs_daddr_t)XFS_FSB_TO_BB((ip)->i_mount, (fsb)) : \
+		 XFS_FSB_TO_DADDR((ip)->i_mount, (fsb)));
+}
diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/xfs_bmap.h
index 89ee672d378..803b56d7ce1 100644
--- a/fs/xfs/xfs_bmap.h
+++ b/fs/xfs/xfs_bmap.h
@@ -211,6 +211,9 @@ int	xfs_bmap_count_blocks(struct xfs_trans *tp, struct xfs_inode *ip,
 		int whichfork, int *count);
 int	xfs_bmap_punch_delalloc_range(struct xfs_inode *ip,
 		xfs_fileoff_t start_fsb, xfs_fileoff_t length);
+
+xfs_daddr_t xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb);
+
 #endif	/* __KERNEL__ */
 
 #endif	/* __XFS_BMAP_H__ */
diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c
index e2f5d59cbea..862084a47a7 100644
--- a/fs/xfs/xfs_bmap_btree.c
+++ b/fs/xfs/xfs_bmap_btree.c
@@ -20,7 +20,6 @@
 #include "xfs_types.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c
index 1f19f03af9d..e53e317b158 100644
--- a/fs/xfs/xfs_btree.c
+++ b/fs/xfs/xfs_btree.c
@@ -20,7 +20,6 @@
 #include "xfs_types.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c
index cf0ac056815..269b35c084d 100644
--- a/fs/xfs/xfs_buf.c
+++ b/fs/xfs/xfs_buf.c
@@ -35,18 +35,14 @@
 #include <linux/freezer.h>
 
 #include "xfs_sb.h"
-#include "xfs_inum.h"
 #include "xfs_log.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
 #include "xfs_trace.h"
 
 static kmem_zone_t *xfs_buf_zone;
-STATIC int xfsbufd(void *);
 
 static struct workqueue_struct *xfslogd_workqueue;
-struct workqueue_struct *xfsdatad_workqueue;
-struct workqueue_struct *xfsconvertd_workqueue;
 
 #ifdef XFS_BUF_LOCK_TRACKING
 # define XB_SET_OWNER(bp)	((bp)->b_last_holder = current->pid)
@@ -59,11 +55,7 @@ struct workqueue_struct *xfsconvertd_workqueue;
 #endif
 
 #define xb_to_gfp(flags) \
-	((((flags) & XBF_READ_AHEAD) ? __GFP_NORETRY : \
-	  ((flags) & XBF_DONT_BLOCK) ? GFP_NOFS : GFP_KERNEL) | __GFP_NOWARN)
-
-#define xb_to_km(flags) \
-	 (((flags) & XBF_DONT_BLOCK) ? KM_NOFS : KM_SLEEP)
+	((((flags) & XBF_READ_AHEAD) ? __GFP_NORETRY : GFP_NOFS) | __GFP_NOWARN)
 
 
 static inline int
@@ -73,11 +65,11 @@ xfs_buf_is_vmapped(
 	/*
 	 * Return true if the buffer is vmapped.
 	 *
-	 * The XBF_MAPPED flag is set if the buffer should be mapped, but the
-	 * code is clever enough to know it doesn't have to map a single page,
-	 * so the check has to be both for XBF_MAPPED and bp->b_page_count > 1.
+	 * b_addr is null if the buffer is not mapped, but the code is clever
+	 * enough to know it doesn't have to map a single page, so the check has
+	 * to be both for b_addr and bp->b_page_count > 1.
 	 */
-	return (bp->b_flags & XBF_MAPPED) && bp->b_page_count > 1;
+	return bp->b_addr && bp->b_page_count > 1;
 }
 
 static inline int
@@ -146,8 +138,17 @@ void
 xfs_buf_stale(
 	struct xfs_buf	*bp)
 {
+	ASSERT(xfs_buf_islocked(bp));
+
 	bp->b_flags |= XBF_STALE;
-	xfs_buf_delwri_dequeue(bp);
+
+	/*
+	 * Clear the delwri status so that a delwri queue walker will not
+	 * flush this buffer to disk now that it is stale. The delwri queue has
+	 * a reference to the buffer, so this is safe to do.
+	 */
+	bp->b_flags &= ~_XBF_DELWRI_Q;
+
 	atomic_set(&(bp)->b_lru_ref, 0);
 	if (!list_empty(&bp->b_lru)) {
 		struct xfs_buftarg *btp = bp->b_target;
@@ -166,22 +167,22 @@ xfs_buf_stale(
 struct xfs_buf *
 xfs_buf_alloc(
 	struct xfs_buftarg	*target,
-	xfs_off_t		range_base,
-	size_t			range_length,
+	xfs_daddr_t		blkno,
+	size_t			numblks,
 	xfs_buf_flags_t		flags)
 {
 	struct xfs_buf		*bp;
 
-	bp = kmem_zone_alloc(xfs_buf_zone, xb_to_km(flags));
+	bp = kmem_zone_zalloc(xfs_buf_zone, KM_NOFS);
 	if (unlikely(!bp))
 		return NULL;
 
 	/*
-	 * We don't want certain flags to appear in b_flags.
+	 * We don't want certain flags to appear in b_flags unless they are
+	 * specifically set by later operations on the buffer.
 	 */
-	flags &= ~(XBF_LOCK|XBF_MAPPED|XBF_DONT_BLOCK|XBF_READ_AHEAD);
+	flags &= ~(XBF_UNMAPPED | XBF_TRYLOCK | XBF_ASYNC | XBF_READ_AHEAD);
 
-	memset(bp, 0, sizeof(xfs_buf_t));
 	atomic_set(&bp->b_hold, 1);
 	atomic_set(&bp->b_lru_ref, 1);
 	init_completion(&bp->b_iowait);
@@ -191,15 +192,16 @@ xfs_buf_alloc(
 	sema_init(&bp->b_sema, 0); /* held, no waiters */
 	XB_SET_OWNER(bp);
 	bp->b_target = target;
-	bp->b_file_offset = range_base;
+
 	/*
-	 * Set buffer_length and count_desired to the same value initially.
-	 * I/O routines should use count_desired, which will be the same in
+	 * Set length and io_length to the same value initially.
+	 * I/O routines should use io_length, which will be the same in
 	 * most cases but may be reset (e.g. XFS recovery).
 	 */
-	bp->b_buffer_length = bp->b_count_desired = range_length;
+	bp->b_length = numblks;
+	bp->b_io_length = numblks;
 	bp->b_flags = flags;
-	bp->b_bn = XFS_BUF_DADDR_NULL;
+	bp->b_bn = blkno;
 	atomic_set(&bp->b_pin_count, 0);
 	init_waitqueue_head(&bp->b_waiters);
 
@@ -221,13 +223,12 @@ _xfs_buf_get_pages(
 {
 	/* Make sure that we have a page list */
 	if (bp->b_pages == NULL) {
-		bp->b_offset = xfs_buf_poff(bp->b_file_offset);
 		bp->b_page_count = page_count;
 		if (page_count <= XB_PAGES) {
 			bp->b_pages = bp->b_page_array;
 		} else {
 			bp->b_pages = kmem_alloc(sizeof(struct page *) *
-					page_count, xb_to_km(flags));
+						 page_count, KM_NOFS);
 			if (bp->b_pages == NULL)
 				return -ENOMEM;
 		}
@@ -290,11 +291,11 @@ xfs_buf_allocate_memory(
 	xfs_buf_t		*bp,
 	uint			flags)
 {
-	size_t			size = bp->b_count_desired;
+	size_t			size;
 	size_t			nbytes, offset;
 	gfp_t			gfp_mask = xb_to_gfp(flags);
 	unsigned short		page_count, i;
-	xfs_off_t		end;
+	xfs_off_t		start, end;
 	int			error;
 
 	/*
@@ -302,15 +303,15 @@ xfs_buf_allocate_memory(
 	 * the memory from the heap - there's no need for the complexity of
 	 * page arrays to keep allocation down to order 0.
 	 */
-	if (bp->b_buffer_length < PAGE_SIZE) {
-		bp->b_addr = kmem_alloc(bp->b_buffer_length, xb_to_km(flags));
+	size = BBTOB(bp->b_length);
+	if (size < PAGE_SIZE) {
+		bp->b_addr = kmem_alloc(size, KM_NOFS);
 		if (!bp->b_addr) {
 			/* low memory - use alloc_page loop instead */
 			goto use_alloc_page;
 		}
 
-		if (((unsigned long)(bp->b_addr + bp->b_buffer_length - 1) &
-								PAGE_MASK) !=
+		if (((unsigned long)(bp->b_addr + size - 1) & PAGE_MASK) !=
 		    ((unsigned long)bp->b_addr & PAGE_MASK)) {
 			/* b_addr spans two pages - use alloc_page instead */
 			kmem_free(bp->b_addr);
@@ -321,13 +322,14 @@ xfs_buf_allocate_memory(
 		bp->b_pages = bp->b_page_array;
 		bp->b_pages[0] = virt_to_page(bp->b_addr);
 		bp->b_page_count = 1;
-		bp->b_flags |= XBF_MAPPED | _XBF_KMEM;
+		bp->b_flags |= _XBF_KMEM;
 		return 0;
 	}
 
 use_alloc_page:
-	end = bp->b_file_offset + bp->b_buffer_length;
-	page_count = xfs_buf_btoc(end) - xfs_buf_btoct(bp->b_file_offset);
+	start = BBTOB(bp->b_bn) >> PAGE_SHIFT;
+	end = (BBTOB(bp->b_bn + bp->b_length) + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	page_count = end - start;
 	error = _xfs_buf_get_pages(bp, page_count, flags);
 	if (unlikely(error))
 		return error;
@@ -390,8 +392,9 @@ _xfs_buf_map_pages(
 	if (bp->b_page_count == 1) {
 		/* A single page buffer is always mappable */
 		bp->b_addr = page_address(bp->b_pages[0]) + bp->b_offset;
-		bp->b_flags |= XBF_MAPPED;
-	} else if (flags & XBF_MAPPED) {
+	} else if (flags & XBF_UNMAPPED) {
+		bp->b_addr = NULL;
+	} else {
 		int retried = 0;
 
 		do {
@@ -405,7 +408,6 @@ _xfs_buf_map_pages(
 		if (!bp->b_addr)
 			return -ENOMEM;
 		bp->b_addr += bp->b_offset;
-		bp->b_flags |= XBF_MAPPED;
 	}
 
 	return 0;
@@ -422,29 +424,27 @@ _xfs_buf_map_pages(
  */
 xfs_buf_t *
 _xfs_buf_find(
-	xfs_buftarg_t		*btp,	/* block device target		*/
-	xfs_off_t		ioff,	/* starting offset of range	*/
-	size_t			isize,	/* length of range		*/
+	struct xfs_buftarg	*btp,
+	xfs_daddr_t		blkno,
+	size_t			numblks,
 	xfs_buf_flags_t		flags,
 	xfs_buf_t		*new_bp)
 {
-	xfs_off_t		range_base;
-	size_t			range_length;
+	size_t			numbytes;
 	struct xfs_perag	*pag;
 	struct rb_node		**rbp;
 	struct rb_node		*parent;
 	xfs_buf_t		*bp;
 
-	range_base = (ioff << BBSHIFT);
-	range_length = (isize << BBSHIFT);
+	numbytes = BBTOB(numblks);
 
 	/* Check for IOs smaller than the sector size / not sector aligned */
-	ASSERT(!(range_length < (1 << btp->bt_sshift)));
-	ASSERT(!(range_base & (xfs_off_t)btp->bt_smask));
+	ASSERT(!(numbytes < (1 << btp->bt_sshift)));
+	ASSERT(!(BBTOB(blkno) & (xfs_off_t)btp->bt_smask));
 
 	/* get tree root */
 	pag = xfs_perag_get(btp->bt_mount,
-				xfs_daddr_to_agno(btp->bt_mount, ioff));
+				xfs_daddr_to_agno(btp->bt_mount, blkno));
 
 	/* walk tree */
 	spin_lock(&pag->pag_buf_lock);
@@ -455,20 +455,20 @@ _xfs_buf_find(
 		parent = *rbp;
 		bp = rb_entry(parent, struct xfs_buf, b_rbnode);
 
-		if (range_base < bp->b_file_offset)
+		if (blkno < bp->b_bn)
 			rbp = &(*rbp)->rb_left;
-		else if (range_base > bp->b_file_offset)
+		else if (blkno > bp->b_bn)
 			rbp = &(*rbp)->rb_right;
 		else {
 			/*
-			 * found a block offset match. If the range doesn't
+			 * found a block number match. If the range doesn't
 			 * match, the only way this is allowed is if the buffer
 			 * in the cache is stale and the transaction that made
 			 * it stale has not yet committed. i.e. we are
 			 * reallocating a busy extent. Skip this buffer and
 			 * continue searching to the right for an exact match.
 			 */
-			if (bp->b_buffer_length != range_length) {
+			if (bp->b_length != numblks) {
 				ASSERT(bp->b_flags & XBF_STALE);
 				rbp = &(*rbp)->rb_right;
 				continue;
@@ -513,7 +513,7 @@ found:
 	 */
 	if (bp->b_flags & XBF_STALE) {
 		ASSERT((bp->b_flags & _XBF_DELWRI_Q) == 0);
-		bp->b_flags &= XBF_MAPPED | _XBF_KMEM | _XBF_PAGES;
+		bp->b_flags &= _XBF_KMEM | _XBF_PAGES;
 	}
 
 	trace_xfs_buf_find(bp, flags, _RET_IP_);
@@ -528,63 +528,54 @@ found:
  */
 struct xfs_buf *
 xfs_buf_get(
-	xfs_buftarg_t		*target,/* target for buffer		*/
-	xfs_off_t		ioff,	/* starting offset of range	*/
-	size_t			isize,	/* length of range		*/
+	xfs_buftarg_t		*target,
+	xfs_daddr_t		blkno,
+	size_t			numblks,
 	xfs_buf_flags_t		flags)
 {
 	struct xfs_buf		*bp;
 	struct xfs_buf		*new_bp;
 	int			error = 0;
 
-	bp = _xfs_buf_find(target, ioff, isize, flags, NULL);
+	bp = _xfs_buf_find(target, blkno, numblks, flags, NULL);
 	if (likely(bp))
 		goto found;
 
-	new_bp = xfs_buf_alloc(target, ioff << BBSHIFT, isize << BBSHIFT,
-			       flags);
+	new_bp = xfs_buf_alloc(target, blkno, numblks, flags);
 	if (unlikely(!new_bp))
 		return NULL;
 
-	bp = _xfs_buf_find(target, ioff, isize, flags, new_bp);
-	if (!bp) {
+	error = xfs_buf_allocate_memory(new_bp, flags);
+	if (error) {
 		kmem_zone_free(xfs_buf_zone, new_bp);
 		return NULL;
 	}
 
-	if (bp == new_bp) {
-		error = xfs_buf_allocate_memory(bp, flags);
-		if (error)
-			goto no_buffer;
-	} else
-		kmem_zone_free(xfs_buf_zone, new_bp);
+	bp = _xfs_buf_find(target, blkno, numblks, flags, new_bp);
+	if (!bp) {
+		xfs_buf_free(new_bp);
+		return NULL;
+	}
 
-	/*
-	 * Now we have a workable buffer, fill in the block number so
-	 * that we can do IO on it.
-	 */
-	bp->b_bn = ioff;
-	bp->b_count_desired = bp->b_buffer_length;
+	if (bp != new_bp)
+		xfs_buf_free(new_bp);
+
+	bp->b_io_length = bp->b_length;
 
 found:
-	if (!(bp->b_flags & XBF_MAPPED)) {
+	if (!bp->b_addr) {
 		error = _xfs_buf_map_pages(bp, flags);
 		if (unlikely(error)) {
 			xfs_warn(target->bt_mount,
 				"%s: failed to map pages\n", __func__);
-			goto no_buffer;
+			xfs_buf_relse(bp);
+			return NULL;
 		}
 	}
 
 	XFS_STATS_INC(xb_get);
 	trace_xfs_buf_get(bp, flags, _RET_IP_);
 	return bp;
-
-no_buffer:
-	if (flags & (XBF_LOCK | XBF_TRYLOCK))
-		xfs_buf_unlock(bp);
-	xfs_buf_rele(bp);
-	return NULL;
 }
 
 STATIC int
@@ -592,32 +583,30 @@ _xfs_buf_read(
 	xfs_buf_t		*bp,
 	xfs_buf_flags_t		flags)
 {
-	int			status;
-
-	ASSERT(!(flags & (XBF_DELWRI|XBF_WRITE)));
+	ASSERT(!(flags & XBF_WRITE));
 	ASSERT(bp->b_bn != XFS_BUF_DADDR_NULL);
 
-	bp->b_flags &= ~(XBF_WRITE | XBF_ASYNC | XBF_DELWRI | XBF_READ_AHEAD);
+	bp->b_flags &= ~(XBF_WRITE | XBF_ASYNC | XBF_READ_AHEAD);
 	bp->b_flags |= flags & (XBF_READ | XBF_ASYNC | XBF_READ_AHEAD);
 
-	status = xfs_buf_iorequest(bp);
-	if (status || bp->b_error || (flags & XBF_ASYNC))
-		return status;
+	xfs_buf_iorequest(bp);
+	if (flags & XBF_ASYNC)
+		return 0;
 	return xfs_buf_iowait(bp);
 }
 
 xfs_buf_t *
 xfs_buf_read(
 	xfs_buftarg_t		*target,
-	xfs_off_t		ioff,
-	size_t			isize,
+	xfs_daddr_t		blkno,
+	size_t			numblks,
 	xfs_buf_flags_t		flags)
 {
 	xfs_buf_t		*bp;
 
 	flags |= XBF_READ;
 
-	bp = xfs_buf_get(target, ioff, isize, flags);
+	bp = xfs_buf_get(target, blkno, numblks, flags);
 	if (bp) {
 		trace_xfs_buf_read(bp, flags, _RET_IP_);
 
@@ -629,7 +618,8 @@ xfs_buf_read(
 			 * Read ahead call which is already satisfied,
 			 * drop the buffer
 			 */
-			goto no_buffer;
+			xfs_buf_relse(bp);
+			return NULL;
 		} else {
 			/* We do not want read in the flags */
 			bp->b_flags &= ~XBF_READ;
@@ -637,12 +627,6 @@ xfs_buf_read(
 	}
 
 	return bp;
-
- no_buffer:
-	if (flags & (XBF_LOCK | XBF_TRYLOCK))
-		xfs_buf_unlock(bp);
-	xfs_buf_rele(bp);
-	return NULL;
 }
 
 /*
@@ -652,14 +636,14 @@ xfs_buf_read(
 void
 xfs_buf_readahead(
 	xfs_buftarg_t		*target,
-	xfs_off_t		ioff,
-	size_t			isize)
+	xfs_daddr_t		blkno,
+	size_t			numblks)
 {
 	if (bdi_read_congested(target->bt_bdi))
 		return;
 
-	xfs_buf_read(target, ioff, isize,
-		     XBF_TRYLOCK|XBF_ASYNC|XBF_READ_AHEAD|XBF_DONT_BLOCK);
+	xfs_buf_read(target, blkno, numblks,
+		     XBF_TRYLOCK|XBF_ASYNC|XBF_READ_AHEAD);
 }
 
 /*
@@ -668,16 +652,15 @@ xfs_buf_readahead(
  */
 struct xfs_buf *
 xfs_buf_read_uncached(
-	struct xfs_mount	*mp,
 	struct xfs_buftarg	*target,
 	xfs_daddr_t		daddr,
-	size_t			length,
+	size_t			numblks,
 	int			flags)
 {
 	xfs_buf_t		*bp;
 	int			error;
 
-	bp = xfs_buf_get_uncached(target, length, flags);
+	bp = xfs_buf_get_uncached(target, numblks, flags);
 	if (!bp)
 		return NULL;
 
@@ -685,9 +668,9 @@ xfs_buf_read_uncached(
 	XFS_BUF_SET_ADDR(bp, daddr);
 	XFS_BUF_READ(bp);
 
-	xfsbdstrat(mp, bp);
+	xfsbdstrat(target->bt_mount, bp);
 	error = xfs_buf_iowait(bp);
-	if (error || bp->b_error) {
+	if (error) {
 		xfs_buf_relse(bp);
 		return NULL;
 	}
@@ -701,7 +684,7 @@ xfs_buf_read_uncached(
 void
 xfs_buf_set_empty(
 	struct xfs_buf		*bp,
-	size_t			len)
+	size_t			numblks)
 {
 	if (bp->b_pages)
 		_xfs_buf_free_pages(bp);
@@ -709,10 +692,9 @@ xfs_buf_set_empty(
 	bp->b_pages = NULL;
 	bp->b_page_count = 0;
 	bp->b_addr = NULL;
-	bp->b_file_offset = 0;
-	bp->b_buffer_length = bp->b_count_desired = len;
+	bp->b_length = numblks;
+	bp->b_io_length = numblks;
 	bp->b_bn = XFS_BUF_DADDR_NULL;
-	bp->b_flags &= ~XBF_MAPPED;
 }
 
 static inline struct page *
@@ -751,7 +733,7 @@ xfs_buf_associate_memory(
 	bp->b_pages = NULL;
 	bp->b_addr = mem;
 
-	rval = _xfs_buf_get_pages(bp, page_count, XBF_DONT_BLOCK);
+	rval = _xfs_buf_get_pages(bp, page_count, 0);
 	if (rval)
 		return rval;
 
@@ -762,9 +744,8 @@ xfs_buf_associate_memory(
 		pageaddr += PAGE_SIZE;
 	}
 
-	bp->b_count_desired = len;
-	bp->b_buffer_length = buflen;
-	bp->b_flags |= XBF_MAPPED;
+	bp->b_io_length = BTOBB(len);
+	bp->b_length = BTOBB(buflen);
 
 	return 0;
 }
@@ -772,17 +753,18 @@ xfs_buf_associate_memory(
 xfs_buf_t *
 xfs_buf_get_uncached(
 	struct xfs_buftarg	*target,
-	size_t			len,
+	size_t			numblks,
 	int			flags)
 {
-	unsigned long		page_count = PAGE_ALIGN(len) >> PAGE_SHIFT;
+	unsigned long		page_count;
 	int			error, i;
 	xfs_buf_t		*bp;
 
-	bp = xfs_buf_alloc(target, 0, len, 0);
+	bp = xfs_buf_alloc(target, XFS_BUF_DADDR_NULL, numblks, 0);
 	if (unlikely(bp == NULL))
 		goto fail;
 
+	page_count = PAGE_ALIGN(numblks << BBSHIFT) >> PAGE_SHIFT;
 	error = _xfs_buf_get_pages(bp, page_count, 0);
 	if (error)
 		goto fail_free_buf;
@@ -794,7 +776,7 @@ xfs_buf_get_uncached(
 	}
 	bp->b_flags |= _XBF_PAGES;
 
-	error = _xfs_buf_map_pages(bp, XBF_MAPPED);
+	error = _xfs_buf_map_pages(bp, 0);
 	if (unlikely(error)) {
 		xfs_warn(target->bt_mount,
 			"%s: failed to map pages\n", __func__);
@@ -857,7 +839,7 @@ xfs_buf_rele(
 			spin_unlock(&pag->pag_buf_lock);
 		} else {
 			xfs_buf_lru_del(bp);
-			ASSERT(!(bp->b_flags & (XBF_DELWRI|_XBF_DELWRI_Q)));
+			ASSERT(!(bp->b_flags & _XBF_DELWRI_Q));
 			rb_erase(&bp->b_rbnode, &pag->pag_buf_tree);
 			spin_unlock(&pag->pag_buf_lock);
 			xfs_perag_put(pag);
@@ -917,13 +899,6 @@ xfs_buf_lock(
 	trace_xfs_buf_lock_done(bp, _RET_IP_);
 }
 
-/*
- *	Releases the lock on the buffer object.
- *	If the buffer is marked delwri but is not queued, do so before we
- *	unlock the buffer as we need to set flags correctly.  We also need to
- *	take a reference for the delwri queue because the unlocker is going to
- *	drop their's and they don't know we just queued it.
- */
 void
 xfs_buf_unlock(
 	struct xfs_buf		*bp)
@@ -1010,29 +985,8 @@ xfs_buf_ioerror_alert(
 	const char		*func)
 {
 	xfs_alert(bp->b_target->bt_mount,
-"metadata I/O error: block 0x%llx (\"%s\") error %d buf count %zd",
-		(__uint64_t)XFS_BUF_ADDR(bp), func,
-		bp->b_error, XFS_BUF_COUNT(bp));
-}
-
-int
-xfs_bwrite(
-	struct xfs_buf		*bp)
-{
-	int			error;
-
-	bp->b_flags |= XBF_WRITE;
-	bp->b_flags &= ~(XBF_ASYNC | XBF_READ);
-
-	xfs_buf_delwri_dequeue(bp);
-	xfs_bdstrat_cb(bp);
-
-	error = xfs_buf_iowait(bp);
-	if (error) {
-		xfs_force_shutdown(bp->b_target->bt_mount,
-				   SHUTDOWN_META_IO_ERROR);
-	}
-	return error;
+"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);
 }
 
 /*
@@ -1104,14 +1058,7 @@ xfs_bioerror_relse(
 	return EIO;
 }
 
-
-/*
- * All xfs metadata buffers except log state machine buffers
- * get this attached as their b_bdstrat callback function.
- * This is so that we can catch a buffer
- * after prematurely unpinning it to forcibly shutdown the filesystem.
- */
-int
+STATIC int
 xfs_bdstrat_cb(
 	struct xfs_buf	*bp)
 {
@@ -1132,6 +1079,27 @@ xfs_bdstrat_cb(
 	return 0;
 }
 
+int
+xfs_bwrite(
+	struct xfs_buf		*bp)
+{
+	int			error;
+
+	ASSERT(xfs_buf_islocked(bp));
+
+	bp->b_flags |= XBF_WRITE;
+	bp->b_flags &= ~(XBF_ASYNC | XBF_READ | _XBF_DELWRI_Q);
+
+	xfs_bdstrat_cb(bp);
+
+	error = xfs_buf_iowait(bp);
+	if (error) {
+		xfs_force_shutdown(bp->b_target->bt_mount,
+				   SHUTDOWN_META_IO_ERROR);
+	}
+	return error;
+}
+
 /*
  * Wrapper around bdstrat so that we can stop data from going to disk in case
  * we are shutting down the filesystem.  Typically user data goes thru this
@@ -1183,7 +1151,7 @@ _xfs_buf_ioapply(
 	int			rw, map_i, total_nr_pages, nr_pages;
 	struct bio		*bio;
 	int			offset = bp->b_offset;
-	int			size = bp->b_count_desired;
+	int			size = BBTOB(bp->b_io_length);
 	sector_t		sector = bp->b_bn;
 
 	total_nr_pages = bp->b_page_count;
@@ -1231,7 +1199,7 @@ next_chunk:
 			break;
 
 		offset = 0;
-		sector += nbytes >> BBSHIFT;
+		sector += BTOBB(nbytes);
 		size -= nbytes;
 		total_nr_pages--;
 	}
@@ -1250,13 +1218,13 @@ next_chunk:
 	}
 }
 
-int
+void
 xfs_buf_iorequest(
 	xfs_buf_t		*bp)
 {
 	trace_xfs_buf_iorequest(bp, _RET_IP_);
 
-	ASSERT(!(bp->b_flags & XBF_DELWRI));
+	ASSERT(!(bp->b_flags & _XBF_DELWRI_Q));
 
 	if (bp->b_flags & XBF_WRITE)
 		xfs_buf_wait_unpin(bp);
@@ -1268,16 +1236,15 @@ xfs_buf_iorequest(
 	 */
 	atomic_set(&bp->b_io_remaining, 1);
 	_xfs_buf_ioapply(bp);
-	_xfs_buf_ioend(bp, 0);
+	_xfs_buf_ioend(bp, 1);
 
 	xfs_buf_rele(bp);
-	return 0;
 }
 
 /*
- *	Waits for I/O to complete on the buffer supplied.
- *	It returns immediately if no I/O is pending.
- *	It returns the I/O error code, if any, or 0 if there was no error.
+ * Waits for I/O to complete on the buffer supplied.  It returns immediately if
+ * no I/O is pending or there is already a pending error on the buffer.  It
+ * returns the I/O error code, if any, or 0 if there was no error.
  */
 int
 xfs_buf_iowait(
@@ -1285,7 +1252,8 @@ xfs_buf_iowait(
 {
 	trace_xfs_buf_iowait(bp, _RET_IP_);
 
-	wait_for_completion(&bp->b_iowait);
+	if (!bp->b_error)
+		wait_for_completion(&bp->b_iowait);
 
 	trace_xfs_buf_iowait_done(bp, _RET_IP_);
 	return bp->b_error;
@@ -1298,7 +1266,7 @@ xfs_buf_offset(
 {
 	struct page		*page;
 
-	if (bp->b_flags & XBF_MAPPED)
+	if (bp->b_addr)
 		return bp->b_addr + offset;
 
 	offset += bp->b_offset;
@@ -1317,27 +1285,30 @@ xfs_buf_iomove(
 	void			*data,	/* data address			*/
 	xfs_buf_rw_t		mode)	/* read/write/zero flag		*/
 {
-	size_t			bend, cpoff, csize;
-	struct page		*page;
+	size_t			bend;
 
 	bend = boff + bsize;
 	while (boff < bend) {
-		page = bp->b_pages[xfs_buf_btoct(boff + bp->b_offset)];
-		cpoff = xfs_buf_poff(boff + bp->b_offset);
-		csize = min_t(size_t,
-			      PAGE_SIZE-cpoff, bp->b_count_desired-boff);
+		struct page	*page;
+		int		page_index, page_offset, csize;
+
+		page_index = (boff + bp->b_offset) >> PAGE_SHIFT;
+		page_offset = (boff + bp->b_offset) & ~PAGE_MASK;
+		page = bp->b_pages[page_index];
+		csize = min_t(size_t, PAGE_SIZE - page_offset,
+				      BBTOB(bp->b_io_length) - boff);
 
-		ASSERT(((csize + cpoff) <= PAGE_SIZE));
+		ASSERT((csize + page_offset) <= PAGE_SIZE);
 
 		switch (mode) {
 		case XBRW_ZERO:
-			memset(page_address(page) + cpoff, 0, csize);
+			memset(page_address(page) + page_offset, 0, csize);
 			break;
 		case XBRW_READ:
-			memcpy(data, page_address(page) + cpoff, csize);
+			memcpy(data, page_address(page) + page_offset, csize);
 			break;
 		case XBRW_WRITE:
-			memcpy(page_address(page) + cpoff, data, csize);
+			memcpy(page_address(page) + page_offset, data, csize);
 		}
 
 		boff += csize;
@@ -1370,7 +1341,7 @@ restart:
 			goto restart;
 		}
 		/*
-		 * clear the LRU reference count so the bufer doesn't get
+		 * clear the LRU reference count so the buffer doesn't get
 		 * ignored in xfs_buf_rele().
 		 */
 		atomic_set(&bp->b_lru_ref, 0);
@@ -1437,11 +1408,9 @@ xfs_free_buftarg(
 {
 	unregister_shrinker(&btp->bt_shrinker);
 
-	xfs_flush_buftarg(btp, 1);
 	if (mp->m_flags & XFS_MOUNT_BARRIER)
 		xfs_blkdev_issue_flush(btp);
 
-	kthread_stop(btp->bt_task);
 	kmem_free(btp);
 }
 
@@ -1493,20 +1462,6 @@ xfs_setsize_buftarg(
 	return xfs_setsize_buftarg_flags(btp, blocksize, sectorsize, 1);
 }
 
-STATIC int
-xfs_alloc_delwri_queue(
-	xfs_buftarg_t		*btp,
-	const char		*fsname)
-{
-	INIT_LIST_HEAD(&btp->bt_delwri_queue);
-	spin_lock_init(&btp->bt_delwri_lock);
-	btp->bt_flags = 0;
-	btp->bt_task = kthread_run(xfsbufd, btp, "xfsbufd/%s", fsname);
-	if (IS_ERR(btp->bt_task))
-		return PTR_ERR(btp->bt_task);
-	return 0;
-}
-
 xfs_buftarg_t *
 xfs_alloc_buftarg(
 	struct xfs_mount	*mp,
@@ -1529,8 +1484,6 @@ xfs_alloc_buftarg(
 	spin_lock_init(&btp->bt_lru_lock);
 	if (xfs_setsize_buftarg_early(btp, bdev))
 		goto error;
-	if (xfs_alloc_delwri_queue(btp, fsname))
-		goto error;
 	btp->bt_shrinker.shrink = xfs_buftarg_shrink;
 	btp->bt_shrinker.seeks = DEFAULT_SEEKS;
 	register_shrinker(&btp->bt_shrinker);
@@ -1541,125 +1494,52 @@ error:
 	return NULL;
 }
 
-
 /*
- *	Delayed write buffer handling
+ * Add a buffer to the delayed write list.
+ *
+ * This queues a buffer for writeout if it hasn't already been.  Note that
+ * neither this routine nor the buffer list submission functions perform
+ * any internal synchronization.  It is expected that the lists are thread-local
+ * to the callers.
+ *
+ * Returns true if we queued up the buffer, or false if it already had
+ * been on the buffer list.
  */
-void
+bool
 xfs_buf_delwri_queue(
-	xfs_buf_t		*bp)
+	struct xfs_buf		*bp,
+	struct list_head	*list)
 {
-	struct xfs_buftarg	*btp = bp->b_target;
-
-	trace_xfs_buf_delwri_queue(bp, _RET_IP_);
-
+	ASSERT(xfs_buf_islocked(bp));
 	ASSERT(!(bp->b_flags & XBF_READ));
 
-	spin_lock(&btp->bt_delwri_lock);
-	if (!list_empty(&bp->b_list)) {
-		/* if already in the queue, move it to the tail */
-		ASSERT(bp->b_flags & _XBF_DELWRI_Q);
-		list_move_tail(&bp->b_list, &btp->bt_delwri_queue);
-	} else {
-		/* start xfsbufd as it is about to have something to do */
-		if (list_empty(&btp->bt_delwri_queue))
-			wake_up_process(bp->b_target->bt_task);
-
-		atomic_inc(&bp->b_hold);
-		bp->b_flags |= XBF_DELWRI | _XBF_DELWRI_Q | XBF_ASYNC;
-		list_add_tail(&bp->b_list, &btp->bt_delwri_queue);
-	}
-	bp->b_queuetime = jiffies;
-	spin_unlock(&btp->bt_delwri_lock);
-}
-
-void
-xfs_buf_delwri_dequeue(
-	xfs_buf_t		*bp)
-{
-	int			dequeued = 0;
-
-	spin_lock(&bp->b_target->bt_delwri_lock);
-	if ((bp->b_flags & XBF_DELWRI) && !list_empty(&bp->b_list)) {
-		ASSERT(bp->b_flags & _XBF_DELWRI_Q);
-		list_del_init(&bp->b_list);
-		dequeued = 1;
+	/*
+	 * If the buffer is already marked delwri it already is queued up
+	 * by someone else for imediate writeout.  Just ignore it in that
+	 * case.
+	 */
+	if (bp->b_flags & _XBF_DELWRI_Q) {
+		trace_xfs_buf_delwri_queued(bp, _RET_IP_);
+		return false;
 	}
-	bp->b_flags &= ~(XBF_DELWRI|_XBF_DELWRI_Q);
-	spin_unlock(&bp->b_target->bt_delwri_lock);
-
-	if (dequeued)
-		xfs_buf_rele(bp);
 
-	trace_xfs_buf_delwri_dequeue(bp, _RET_IP_);
-}
-
-/*
- * If a delwri buffer needs to be pushed before it has aged out, then promote
- * it to the head of the delwri queue so that it will be flushed on the next
- * xfsbufd run. We do this by resetting the queuetime of the buffer to be older
- * than the age currently needed to flush the buffer. Hence the next time the
- * xfsbufd sees it is guaranteed to be considered old enough to flush.
- */
-void
-xfs_buf_delwri_promote(
-	struct xfs_buf	*bp)
-{
-	struct xfs_buftarg *btp = bp->b_target;
-	long		age = xfs_buf_age_centisecs * msecs_to_jiffies(10) + 1;
-
-	ASSERT(bp->b_flags & XBF_DELWRI);
-	ASSERT(bp->b_flags & _XBF_DELWRI_Q);
+	trace_xfs_buf_delwri_queue(bp, _RET_IP_);
 
 	/*
-	 * Check the buffer age before locking the delayed write queue as we
-	 * don't need to promote buffers that are already past the flush age.
+	 * If a buffer gets written out synchronously or marked stale while it
+	 * is on a delwri list we lazily remove it. To do this, the other party
+	 * clears the  _XBF_DELWRI_Q flag but otherwise leaves the buffer alone.
+	 * It remains referenced and on the list.  In a rare corner case it
+	 * might get readded to a delwri list after the synchronous writeout, in
+	 * which case we need just need to re-add the flag here.
 	 */
-	if (bp->b_queuetime < jiffies - age)
-		return;
-	bp->b_queuetime = jiffies - age;
-	spin_lock(&btp->bt_delwri_lock);
-	list_move(&bp->b_list, &btp->bt_delwri_queue);
-	spin_unlock(&btp->bt_delwri_lock);
-}
-
-/*
- * Move as many buffers as specified to the supplied list
- * idicating if we skipped any buffers to prevent deadlocks.
- */
-STATIC int
-xfs_buf_delwri_split(
-	xfs_buftarg_t	*target,
-	struct list_head *list,
-	unsigned long	age)
-{
-	xfs_buf_t	*bp, *n;
-	int		skipped = 0;
-	int		force;
-
-	force = test_and_clear_bit(XBT_FORCE_FLUSH, &target->bt_flags);
-	INIT_LIST_HEAD(list);
-	spin_lock(&target->bt_delwri_lock);
-	list_for_each_entry_safe(bp, n, &target->bt_delwri_queue, b_list) {
-		ASSERT(bp->b_flags & XBF_DELWRI);
-
-		if (!xfs_buf_ispinned(bp) && xfs_buf_trylock(bp)) {
-			if (!force &&
-			    time_before(jiffies, bp->b_queuetime + age)) {
-				xfs_buf_unlock(bp);
-				break;
-			}
-
-			bp->b_flags &= ~(XBF_DELWRI | _XBF_DELWRI_Q);
-			bp->b_flags |= XBF_WRITE;
-			list_move_tail(&bp->b_list, list);
-			trace_xfs_buf_delwri_split(bp, _RET_IP_);
-		} else
-			skipped++;
+	bp->b_flags |= _XBF_DELWRI_Q;
+	if (list_empty(&bp->b_list)) {
+		atomic_inc(&bp->b_hold);
+		list_add_tail(&bp->b_list, list);
 	}
 
-	spin_unlock(&target->bt_delwri_lock);
-	return skipped;
+	return true;
 }
 
 /*
@@ -1685,103 +1565,109 @@ xfs_buf_cmp(
 	return 0;
 }
 
-STATIC int
-xfsbufd(
-	void		*data)
+static int
+__xfs_buf_delwri_submit(
+	struct list_head	*buffer_list,
+	struct list_head	*io_list,
+	bool			wait)
 {
-	xfs_buftarg_t   *target = (xfs_buftarg_t *)data;
-
-	current->flags |= PF_MEMALLOC;
-
-	set_freezable();
-
-	do {
-		long	age = xfs_buf_age_centisecs * msecs_to_jiffies(10);
-		long	tout = xfs_buf_timer_centisecs * msecs_to_jiffies(10);
-		struct list_head tmp;
-		struct blk_plug plug;
-
-		if (unlikely(freezing(current))) {
-			set_bit(XBT_FORCE_SLEEP, &target->bt_flags);
-			refrigerator();
+	struct blk_plug		plug;
+	struct xfs_buf		*bp, *n;
+	int			pinned = 0;
+
+	list_for_each_entry_safe(bp, n, buffer_list, b_list) {
+		if (!wait) {
+			if (xfs_buf_ispinned(bp)) {
+				pinned++;
+				continue;
+			}
+			if (!xfs_buf_trylock(bp))
+				continue;
 		} else {
-			clear_bit(XBT_FORCE_SLEEP, &target->bt_flags);
+			xfs_buf_lock(bp);
 		}
 
-		/* sleep for a long time if there is nothing to do. */
-		if (list_empty(&target->bt_delwri_queue))
-			tout = MAX_SCHEDULE_TIMEOUT;
-		schedule_timeout_interruptible(tout);
+		/*
+		 * Someone else might have written the buffer synchronously or
+		 * marked it stale in the meantime.  In that case only the
+		 * _XBF_DELWRI_Q flag got cleared, and we have to drop the
+		 * reference and remove it from the list here.
+		 */
+		if (!(bp->b_flags & _XBF_DELWRI_Q)) {
+			list_del_init(&bp->b_list);
+			xfs_buf_relse(bp);
+			continue;
+		}
+
+		list_move_tail(&bp->b_list, io_list);
+		trace_xfs_buf_delwri_split(bp, _RET_IP_);
+	}
 
-		xfs_buf_delwri_split(target, &tmp, age);
-		list_sort(NULL, &tmp, xfs_buf_cmp);
+	list_sort(NULL, io_list, xfs_buf_cmp);
 
-		blk_start_plug(&plug);
-		while (!list_empty(&tmp)) {
-			struct xfs_buf *bp;
-			bp = list_first_entry(&tmp, struct xfs_buf, b_list);
+	blk_start_plug(&plug);
+	list_for_each_entry_safe(bp, n, io_list, b_list) {
+		bp->b_flags &= ~(_XBF_DELWRI_Q | XBF_ASYNC);
+		bp->b_flags |= XBF_WRITE;
+
+		if (!wait) {
+			bp->b_flags |= XBF_ASYNC;
 			list_del_init(&bp->b_list);
-			xfs_bdstrat_cb(bp);
 		}
-		blk_finish_plug(&plug);
-	} while (!kthread_should_stop());
+		xfs_bdstrat_cb(bp);
+	}
+	blk_finish_plug(&plug);
 
-	return 0;
+	return pinned;
 }
 
 /*
- *	Go through all incore buffers, and release buffers if they belong to
- *	the given device. This is used in filesystem error handling to
- *	preserve the consistency of its metadata.
+ * Write out a buffer list asynchronously.
+ *
+ * This will take the @buffer_list, write all non-locked and non-pinned buffers
+ * out and not wait for I/O completion on any of the buffers.  This interface
+ * is only safely useable for callers that can track I/O completion by higher
+ * level means, e.g. AIL pushing as the @buffer_list is consumed in this
+ * function.
  */
 int
-xfs_flush_buftarg(
-	xfs_buftarg_t	*target,
-	int		wait)
+xfs_buf_delwri_submit_nowait(
+	struct list_head	*buffer_list)
 {
-	xfs_buf_t	*bp;
-	int		pincount = 0;
-	LIST_HEAD(tmp_list);
-	LIST_HEAD(wait_list);
-	struct blk_plug plug;
+	LIST_HEAD		(io_list);
+	return __xfs_buf_delwri_submit(buffer_list, &io_list, false);
+}
 
-	flush_workqueue(xfslogd_workqueue);
+/*
+ * Write out a buffer list synchronously.
+ *
+ * This will take the @buffer_list, write all buffers out and wait for I/O
+ * completion on all of the buffers. @buffer_list is consumed by the function,
+ * so callers must have some other way of tracking buffers if they require such
+ * functionality.
+ */
+int
+xfs_buf_delwri_submit(
+	struct list_head	*buffer_list)
+{
+	LIST_HEAD		(io_list);
+	int			error = 0, error2;
+	struct xfs_buf		*bp;
 
-	set_bit(XBT_FORCE_FLUSH, &target->bt_flags);
-	pincount = xfs_buf_delwri_split(target, &tmp_list, 0);
+	__xfs_buf_delwri_submit(buffer_list, &io_list, true);
 
-	/*
-	 * Dropped the delayed write list lock, now walk the temporary list.
-	 * All I/O is issued async and then if we need to wait for completion
-	 * we do that after issuing all the IO.
-	 */
-	list_sort(NULL, &tmp_list, xfs_buf_cmp);
+	/* Wait for IO to complete. */
+	while (!list_empty(&io_list)) {
+		bp = list_first_entry(&io_list, struct xfs_buf, b_list);
 
-	blk_start_plug(&plug);
-	while (!list_empty(&tmp_list)) {
-		bp = list_first_entry(&tmp_list, struct xfs_buf, b_list);
-		ASSERT(target == bp->b_target);
 		list_del_init(&bp->b_list);
-		if (wait) {
-			bp->b_flags &= ~XBF_ASYNC;
-			list_add(&bp->b_list, &wait_list);
-		}
-		xfs_bdstrat_cb(bp);
-	}
-	blk_finish_plug(&plug);
-
-	if (wait) {
-		/* Wait for IO to complete. */
-		while (!list_empty(&wait_list)) {
-			bp = list_first_entry(&wait_list, struct xfs_buf, b_list);
-
-			list_del_init(&bp->b_list);
-			xfs_buf_iowait(bp);
-			xfs_buf_relse(bp);
-		}
+		error2 = xfs_buf_iowait(bp);
+		xfs_buf_relse(bp);
+		if (!error)
+			error = error2;
 	}
 
-	return pincount;
+	return error;
 }
 
 int __init
@@ -1797,21 +1683,8 @@ xfs_buf_init(void)
 	if (!xfslogd_workqueue)
 		goto out_free_buf_zone;
 
-	xfsdatad_workqueue = alloc_workqueue("xfsdatad", WQ_MEM_RECLAIM, 1);
-	if (!xfsdatad_workqueue)
-		goto out_destroy_xfslogd_workqueue;
-
-	xfsconvertd_workqueue = alloc_workqueue("xfsconvertd",
-						WQ_MEM_RECLAIM, 1);
-	if (!xfsconvertd_workqueue)
-		goto out_destroy_xfsdatad_workqueue;
-
 	return 0;
 
- out_destroy_xfsdatad_workqueue:
-	destroy_workqueue(xfsdatad_workqueue);
- out_destroy_xfslogd_workqueue:
-	destroy_workqueue(xfslogd_workqueue);
  out_free_buf_zone:
 	kmem_zone_destroy(xfs_buf_zone);
  out:
@@ -1821,8 +1694,6 @@ xfs_buf_init(void)
 void
 xfs_buf_terminate(void)
 {
-	destroy_workqueue(xfsconvertd_workqueue);
-	destroy_workqueue(xfsdatad_workqueue);
 	destroy_workqueue(xfslogd_workqueue);
 	kmem_zone_destroy(xfs_buf_zone);
 }
diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h
index 5bab046e859..79344c48008 100644
--- a/fs/xfs/xfs_buf.h
+++ b/fs/xfs/xfs_buf.h
@@ -21,7 +21,6 @@
 #include <linux/list.h>
 #include <linux/types.h>
 #include <linux/spinlock.h>
-#include <asm/system.h>
 #include <linux/mm.h>
 #include <linux/fs.h>
 #include <linux/buffer_head.h>
@@ -33,11 +32,6 @@
 
 #define XFS_BUF_DADDR_NULL	((xfs_daddr_t) (-1LL))
 
-#define xfs_buf_ctob(pp)	((pp) * PAGE_CACHE_SIZE)
-#define xfs_buf_btoc(dd)	(((dd) + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT)
-#define xfs_buf_btoct(dd)	((dd) >> PAGE_CACHE_SHIFT)
-#define xfs_buf_poff(aa)	((aa) & ~PAGE_CACHE_MASK)
-
 typedef enum {
 	XBRW_READ = 1,			/* transfer into target memory */
 	XBRW_WRITE = 2,			/* transfer from target memory */
@@ -47,11 +41,9 @@ typedef enum {
 #define XBF_READ	(1 << 0) /* buffer intended for reading from device */
 #define XBF_WRITE	(1 << 1) /* buffer intended for writing to device */
 #define XBF_READ_AHEAD	(1 << 2) /* asynchronous read-ahead */
-#define XBF_MAPPED	(1 << 3) /* buffer mapped (b_addr valid) */
 #define XBF_ASYNC	(1 << 4) /* initiator will not wait for completion */
 #define XBF_DONE	(1 << 5) /* all pages in the buffer uptodate */
-#define XBF_DELWRI	(1 << 6) /* buffer has dirty pages */
-#define XBF_STALE	(1 << 7) /* buffer has been staled, do not find it */
+#define XBF_STALE	(1 << 6) /* buffer has been staled, do not find it */
 
 /* I/O hints for the BIO layer */
 #define XBF_SYNCIO	(1 << 10)/* treat this buffer as synchronous I/O */
@@ -59,14 +51,13 @@ typedef enum {
 #define XBF_FLUSH	(1 << 12)/* flush the disk cache before a write */
 
 /* flags used only as arguments to access routines */
-#define XBF_LOCK	(1 << 15)/* lock requested */
 #define XBF_TRYLOCK	(1 << 16)/* lock requested, but do not wait */
-#define XBF_DONT_BLOCK	(1 << 17)/* do not block in current thread */
+#define XBF_UNMAPPED	(1 << 17)/* do not map the buffer */
 
 /* flags used only internally */
 #define _XBF_PAGES	(1 << 20)/* backed by refcounted pages */
 #define _XBF_KMEM	(1 << 21)/* backed by heap memory */
-#define _XBF_DELWRI_Q	(1 << 22)/* buffer on delwri queue */
+#define _XBF_DELWRI_Q	(1 << 22)/* buffer on a delwri queue */
 
 typedef unsigned int xfs_buf_flags_t;
 
@@ -74,26 +65,18 @@ typedef unsigned int xfs_buf_flags_t;
 	{ XBF_READ,		"READ" }, \
 	{ XBF_WRITE,		"WRITE" }, \
 	{ XBF_READ_AHEAD,	"READ_AHEAD" }, \
-	{ XBF_MAPPED,		"MAPPED" }, \
 	{ XBF_ASYNC,		"ASYNC" }, \
 	{ XBF_DONE,		"DONE" }, \
-	{ XBF_DELWRI,		"DELWRI" }, \
 	{ XBF_STALE,		"STALE" }, \
 	{ XBF_SYNCIO,		"SYNCIO" }, \
 	{ XBF_FUA,		"FUA" }, \
 	{ XBF_FLUSH,		"FLUSH" }, \
-	{ XBF_LOCK,		"LOCK" },  	/* should never be set */\
-	{ XBF_TRYLOCK,		"TRYLOCK" }, 	/* ditto */\
-	{ XBF_DONT_BLOCK,	"DONT_BLOCK" },	/* ditto */\
+	{ XBF_TRYLOCK,		"TRYLOCK" }, 	/* should never be set */\
+	{ XBF_UNMAPPED,		"UNMAPPED" },	/* ditto */\
 	{ _XBF_PAGES,		"PAGES" }, \
 	{ _XBF_KMEM,		"KMEM" }, \
 	{ _XBF_DELWRI_Q,	"DELWRI_Q" }
 
-typedef enum {
-	XBT_FORCE_SLEEP = 0,
-	XBT_FORCE_FLUSH = 1,
-} xfs_buftarg_flags_t;
-
 typedef struct xfs_buftarg {
 	dev_t			bt_dev;
 	struct block_device	*bt_bdev;
@@ -103,12 +86,6 @@ typedef struct xfs_buftarg {
 	unsigned int		bt_sshift;
 	size_t			bt_smask;
 
-	/* per device delwri queue */
-	struct task_struct	*bt_task;
-	struct list_head	bt_delwri_queue;
-	spinlock_t		bt_delwri_lock;
-	unsigned long		bt_flags;
-
 	/* LRU control structures */
 	struct shrinker		bt_shrinker;
 	struct list_head	bt_lru;
@@ -130,8 +107,8 @@ typedef struct xfs_buf {
 	 * fast-path on locking.
 	 */
 	struct rb_node		b_rbnode;	/* rbtree node */
-	xfs_off_t		b_file_offset;	/* offset in file */
-	size_t			b_buffer_length;/* size of buffer in bytes */
+	xfs_daddr_t		b_bn;		/* block number for I/O */
+	int			b_length;	/* size of buffer in BBs */
 	atomic_t		b_hold;		/* reference count */
 	atomic_t		b_lru_ref;	/* lru reclaim ref count */
 	xfs_buf_flags_t		b_flags;	/* status flags */
@@ -142,8 +119,6 @@ typedef struct xfs_buf {
 	struct list_head	b_list;
 	struct xfs_perag	*b_pag;		/* contains rbtree root */
 	xfs_buftarg_t		*b_target;	/* buffer target (device) */
-	xfs_daddr_t		b_bn;		/* block number for I/O */
-	size_t			b_count_desired;/* desired transfer size */
 	void			*b_addr;	/* virtual address of buffer */
 	struct work_struct	b_iodone_work;
 	xfs_buf_iodone_t	b_iodone;	/* I/O completion function */
@@ -152,7 +127,7 @@ typedef struct xfs_buf {
 	struct xfs_trans	*b_transp;
 	struct page		**b_pages;	/* array of page pointers */
 	struct page		*b_page_array[XB_PAGES]; /* inline pages */
-	unsigned long		b_queuetime;	/* time buffer was queued */
+	int			b_io_length;	/* IO size in BBs */
 	atomic_t		b_pin_count;	/* pin count */
 	atomic_t		b_io_remaining;	/* #outstanding I/O requests */
 	unsigned int		b_page_count;	/* size of page array */
@@ -165,26 +140,30 @@ typedef struct xfs_buf {
 
 
 /* Finding and Reading Buffers */
-extern xfs_buf_t *_xfs_buf_find(xfs_buftarg_t *, xfs_off_t, size_t,
-				xfs_buf_flags_t, xfs_buf_t *);
+struct xfs_buf *_xfs_buf_find(struct xfs_buftarg *target, xfs_daddr_t blkno,
+				size_t numblks, xfs_buf_flags_t flags,
+				struct xfs_buf *new_bp);
 #define xfs_incore(buftarg,blkno,len,lockit) \
 	_xfs_buf_find(buftarg, blkno ,len, lockit, NULL)
 
-extern xfs_buf_t *xfs_buf_get(xfs_buftarg_t *, xfs_off_t, size_t,
-				xfs_buf_flags_t);
-extern xfs_buf_t *xfs_buf_read(xfs_buftarg_t *, xfs_off_t, size_t,
-				xfs_buf_flags_t);
-
-struct xfs_buf *xfs_buf_alloc(struct xfs_buftarg *, xfs_off_t, size_t,
-			      xfs_buf_flags_t);
-extern void xfs_buf_set_empty(struct xfs_buf *bp, size_t len);
-extern xfs_buf_t *xfs_buf_get_uncached(struct xfs_buftarg *, size_t, int);
-extern int xfs_buf_associate_memory(xfs_buf_t *, void *, size_t);
-extern void xfs_buf_hold(xfs_buf_t *);
-extern void xfs_buf_readahead(xfs_buftarg_t *, xfs_off_t, size_t);
-struct xfs_buf *xfs_buf_read_uncached(struct xfs_mount *mp,
-				struct xfs_buftarg *target,
-				xfs_daddr_t daddr, size_t length, int flags);
+struct xfs_buf *xfs_buf_get(struct xfs_buftarg *target, xfs_daddr_t blkno,
+				size_t numblks, xfs_buf_flags_t flags);
+struct xfs_buf *xfs_buf_read(struct xfs_buftarg *target, xfs_daddr_t blkno,
+				size_t numblks, xfs_buf_flags_t flags);
+void xfs_buf_readahead(struct xfs_buftarg *target, xfs_daddr_t blkno,
+				size_t numblks);
+
+struct xfs_buf *xfs_buf_get_empty(struct xfs_buftarg *target, size_t numblks);
+struct xfs_buf *xfs_buf_alloc(struct xfs_buftarg *target, xfs_daddr_t blkno,
+				size_t numblks, xfs_buf_flags_t flags);
+void xfs_buf_set_empty(struct xfs_buf *bp, size_t numblks);
+int xfs_buf_associate_memory(struct xfs_buf *bp, void *mem, size_t length);
+
+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);
+void xfs_buf_hold(struct xfs_buf *bp);
 
 /* Releasing Buffers */
 extern void xfs_buf_free(xfs_buf_t *);
@@ -201,12 +180,11 @@ extern void xfs_buf_unlock(xfs_buf_t *);
 extern int xfs_bwrite(struct xfs_buf *bp);
 
 extern void xfsbdstrat(struct xfs_mount *, struct xfs_buf *);
-extern int xfs_bdstrat_cb(struct xfs_buf *);
 
 extern void xfs_buf_ioend(xfs_buf_t *,	int);
 extern void xfs_buf_ioerror(xfs_buf_t *, int);
 extern void xfs_buf_ioerror_alert(struct xfs_buf *, const char *func);
-extern int xfs_buf_iorequest(xfs_buf_t *);
+extern void xfs_buf_iorequest(xfs_buf_t *);
 extern int xfs_buf_iowait(xfs_buf_t *);
 extern void xfs_buf_iomove(xfs_buf_t *, size_t, size_t, void *,
 				xfs_buf_rw_t);
@@ -222,24 +200,22 @@ static inline int xfs_buf_geterror(xfs_buf_t *bp)
 extern xfs_caddr_t xfs_buf_offset(xfs_buf_t *, size_t);
 
 /* Delayed Write Buffer Routines */
-extern void xfs_buf_delwri_queue(struct xfs_buf *);
-extern void xfs_buf_delwri_dequeue(struct xfs_buf *);
-extern void xfs_buf_delwri_promote(struct xfs_buf *);
+extern bool xfs_buf_delwri_queue(struct xfs_buf *, struct list_head *);
+extern int xfs_buf_delwri_submit(struct list_head *);
+extern int xfs_buf_delwri_submit_nowait(struct list_head *);
 
 /* Buffer Daemon Setup Routines */
 extern int xfs_buf_init(void);
 extern void xfs_buf_terminate(void);
 
 #define XFS_BUF_ZEROFLAGS(bp) \
-	((bp)->b_flags &= ~(XBF_READ|XBF_WRITE|XBF_ASYNC|XBF_DELWRI| \
+	((bp)->b_flags &= ~(XBF_READ|XBF_WRITE|XBF_ASYNC| \
 			    XBF_SYNCIO|XBF_FUA|XBF_FLUSH))
 
 void xfs_buf_stale(struct xfs_buf *bp);
 #define XFS_BUF_UNSTALE(bp)	((bp)->b_flags &= ~XBF_STALE)
 #define XFS_BUF_ISSTALE(bp)	((bp)->b_flags & XBF_STALE)
 
-#define XFS_BUF_ISDELAYWRITE(bp)	((bp)->b_flags & XBF_DELWRI)
-
 #define XFS_BUF_DONE(bp)	((bp)->b_flags |= XBF_DONE)
 #define XFS_BUF_UNDONE(bp)	((bp)->b_flags &= ~XBF_DONE)
 #define XFS_BUF_ISDONE(bp)	((bp)->b_flags & XBF_DONE)
@@ -258,12 +234,6 @@ void xfs_buf_stale(struct xfs_buf *bp);
 
 #define XFS_BUF_ADDR(bp)		((bp)->b_bn)
 #define XFS_BUF_SET_ADDR(bp, bno)	((bp)->b_bn = (xfs_daddr_t)(bno))
-#define XFS_BUF_OFFSET(bp)		((bp)->b_file_offset)
-#define XFS_BUF_SET_OFFSET(bp, off)	((bp)->b_file_offset = (off))
-#define XFS_BUF_COUNT(bp)		((bp)->b_count_desired)
-#define XFS_BUF_SET_COUNT(bp, cnt)	((bp)->b_count_desired = (cnt))
-#define XFS_BUF_SIZE(bp)		((bp)->b_buffer_length)
-#define XFS_BUF_SET_SIZE(bp, cnt)	((bp)->b_buffer_length = (cnt))
 
 static inline void xfs_buf_set_ref(struct xfs_buf *bp, int lru_ref)
 {
@@ -289,7 +259,6 @@ extern xfs_buftarg_t *xfs_alloc_buftarg(struct xfs_mount *,
 extern void xfs_free_buftarg(struct xfs_mount *, struct xfs_buftarg *);
 extern void xfs_wait_buftarg(xfs_buftarg_t *);
 extern int xfs_setsize_buftarg(xfs_buftarg_t *, unsigned int, unsigned int);
-extern int xfs_flush_buftarg(xfs_buftarg_t *, int);
 
 #define xfs_getsize_buftarg(buftarg)	block_size((buftarg)->bt_bdev)
 #define xfs_readonly_buftarg(buftarg)	bdev_read_only((buftarg)->bt_bdev)
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index eac97ef81e2..d9e451115f9 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -20,7 +20,6 @@
 #include "xfs_types.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -123,11 +122,11 @@ xfs_buf_item_log_check(
 	ASSERT(bip->bli_logged != NULL);
 
 	bp = bip->bli_buf;
-	ASSERT(XFS_BUF_COUNT(bp) > 0);
+	ASSERT(bp->b_length > 0);
 	ASSERT(bp->b_addr != NULL);
 	orig = bip->bli_orig;
 	buffer = bp->b_addr;
-	for (x = 0; x < XFS_BUF_COUNT(bp); x++) {
+	for (x = 0; x < BBTOB(bp->b_length); x++) {
 		if (orig[x] != buffer[x] && !btst(bip->bli_logged, x)) {
 			xfs_emerg(bp->b_mount,
 				"%s: bip %x buffer %x orig %x index %d",
@@ -418,7 +417,6 @@ xfs_buf_item_unpin(
 	if (freed && stale) {
 		ASSERT(bip->bli_flags & XFS_BLI_STALE);
 		ASSERT(xfs_buf_islocked(bp));
-		ASSERT(!(XFS_BUF_ISDELAYWRITE(bp)));
 		ASSERT(XFS_BUF_ISSTALE(bp));
 		ASSERT(bip->bli_format.blf_flags & XFS_BLF_CANCEL);
 
@@ -455,42 +453,42 @@ xfs_buf_item_unpin(
 			bp->b_iodone = NULL;
 		} else {
 			spin_lock(&ailp->xa_lock);
-			xfs_trans_ail_delete(ailp, (xfs_log_item_t *)bip);
+			xfs_trans_ail_delete(ailp, lip, SHUTDOWN_LOG_IO_ERROR);
 			xfs_buf_item_relse(bp);
 			ASSERT(bp->b_fspriv == NULL);
 		}
 		xfs_buf_relse(bp);
+	} else if (freed && remove) {
+		xfs_buf_lock(bp);
+		xfs_buf_ioerror(bp, EIO);
+		XFS_BUF_UNDONE(bp);
+		xfs_buf_stale(bp);
+		xfs_buf_ioend(bp, 0);
 	}
 }
 
-/*
- * This is called to attempt to lock the buffer associated with this
- * buf log item.  Don't sleep on the buffer lock.  If we can't get
- * the lock right away, return 0.  If we can get the lock, take a
- * reference to the buffer. If this is a delayed write buffer that
- * needs AIL help to be written back, invoke the pushbuf routine
- * rather than the normal success path.
- */
 STATIC uint
-xfs_buf_item_trylock(
-	struct xfs_log_item	*lip)
+xfs_buf_item_push(
+	struct xfs_log_item	*lip,
+	struct list_head	*buffer_list)
 {
 	struct xfs_buf_log_item	*bip = BUF_ITEM(lip);
 	struct xfs_buf		*bp = bip->bli_buf;
+	uint			rval = XFS_ITEM_SUCCESS;
 
 	if (xfs_buf_ispinned(bp))
 		return XFS_ITEM_PINNED;
 	if (!xfs_buf_trylock(bp))
 		return XFS_ITEM_LOCKED;
 
-	/* take a reference to the buffer.  */
-	xfs_buf_hold(bp);
-
 	ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
-	trace_xfs_buf_item_trylock(bip);
-	if (XFS_BUF_ISDELAYWRITE(bp))
-		return XFS_ITEM_PUSHBUF;
-	return XFS_ITEM_SUCCESS;
+
+	trace_xfs_buf_item_push(bip);
+
+	if (!xfs_buf_delwri_queue(bp, buffer_list))
+		rval = XFS_ITEM_FLUSHING;
+	xfs_buf_unlock(bp);
+	return rval;
 }
 
 /*
@@ -603,49 +601,6 @@ xfs_buf_item_committed(
 	return lsn;
 }
 
-/*
- * The buffer is locked, but is not a delayed write buffer. This happens
- * if we race with IO completion and hence we don't want to try to write it
- * again. Just release the buffer.
- */
-STATIC void
-xfs_buf_item_push(
-	struct xfs_log_item	*lip)
-{
-	struct xfs_buf_log_item	*bip = BUF_ITEM(lip);
-	struct xfs_buf		*bp = bip->bli_buf;
-
-	ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
-	ASSERT(!XFS_BUF_ISDELAYWRITE(bp));
-
-	trace_xfs_buf_item_push(bip);
-
-	xfs_buf_relse(bp);
-}
-
-/*
- * The buffer is locked and is a delayed write buffer. Promote the buffer
- * in the delayed write queue as the caller knows that they must invoke
- * the xfsbufd to get this buffer written. We have to unlock the buffer
- * to allow the xfsbufd to write it, too.
- */
-STATIC bool
-xfs_buf_item_pushbuf(
-	struct xfs_log_item	*lip)
-{
-	struct xfs_buf_log_item	*bip = BUF_ITEM(lip);
-	struct xfs_buf		*bp = bip->bli_buf;
-
-	ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
-	ASSERT(XFS_BUF_ISDELAYWRITE(bp));
-
-	trace_xfs_buf_item_pushbuf(bip);
-
-	xfs_buf_delwri_promote(bp);
-	xfs_buf_relse(bp);
-	return true;
-}
-
 STATIC void
 xfs_buf_item_committing(
 	struct xfs_log_item	*lip,
@@ -661,11 +616,9 @@ static const struct xfs_item_ops xfs_buf_item_ops = {
 	.iop_format	= xfs_buf_item_format,
 	.iop_pin	= xfs_buf_item_pin,
 	.iop_unpin	= xfs_buf_item_unpin,
-	.iop_trylock	= xfs_buf_item_trylock,
 	.iop_unlock	= xfs_buf_item_unlock,
 	.iop_committed	= xfs_buf_item_committed,
 	.iop_push	= xfs_buf_item_push,
-	.iop_pushbuf	= xfs_buf_item_pushbuf,
 	.iop_committing = xfs_buf_item_committing
 };
 
@@ -703,7 +656,8 @@ xfs_buf_item_init(
 	 * truncate any pieces.  map_size is the size of the
 	 * bitmap needed to describe the chunks of the buffer.
 	 */
-	chunks = (int)((XFS_BUF_COUNT(bp) + (XFS_BLF_CHUNK - 1)) >> XFS_BLF_SHIFT);
+	chunks = (int)((BBTOB(bp->b_length) + (XFS_BLF_CHUNK - 1)) >>
+								XFS_BLF_SHIFT);
 	map_size = (int)((chunks + NBWORD) >> BIT_TO_WORD_SHIFT);
 
 	bip = (xfs_buf_log_item_t*)kmem_zone_zalloc(xfs_buf_item_zone,
@@ -713,7 +667,7 @@ xfs_buf_item_init(
 	xfs_buf_hold(bp);
 	bip->bli_format.blf_type = XFS_LI_BUF;
 	bip->bli_format.blf_blkno = (__int64_t)XFS_BUF_ADDR(bp);
-	bip->bli_format.blf_len = (ushort)BTOBB(XFS_BUF_COUNT(bp));
+	bip->bli_format.blf_len = (ushort)bp->b_length;
 	bip->bli_format.blf_map_size = map_size;
 
 #ifdef XFS_TRANS_DEBUG
@@ -725,9 +679,9 @@ xfs_buf_item_init(
 	 * the buffer to indicate which bytes the callers have asked
 	 * to have logged.
 	 */
-	bip->bli_orig = (char *)kmem_alloc(XFS_BUF_COUNT(bp), KM_SLEEP);
-	memcpy(bip->bli_orig, bp->b_addr, XFS_BUF_COUNT(bp));
-	bip->bli_logged = (char *)kmem_zalloc(XFS_BUF_COUNT(bp) / NBBY, KM_SLEEP);
+	bip->bli_orig = kmem_alloc(BBTOB(bp->b_length), KM_SLEEP);
+	memcpy(bip->bli_orig, bp->b_addr, BBTOB(bp->b_length));
+	bip->bli_logged = kmem_zalloc(BBTOB(bp->b_length) / NBBY, KM_SLEEP);
 #endif
 
 	/*
@@ -984,20 +938,27 @@ xfs_buf_iodone_callbacks(
 	 * If the write was asynchronous then no one will be looking for the
 	 * error.  Clear the error state and write the buffer out again.
 	 *
-	 * During sync or umount we'll write all pending buffers again
-	 * synchronous, which will catch these errors if they keep hanging
-	 * around.
+	 * XXX: This helps against transient write errors, but we need to find
+	 * 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
+	 * will make us give up.  But we really ought to do better here.
 	 */
 	if (XFS_BUF_ISASYNC(bp)) {
+		ASSERT(bp->b_iodone != NULL);
+
+		trace_xfs_buf_item_iodone_async(bp, _RET_IP_);
+
 		xfs_buf_ioerror(bp, 0); /* errno of 0 unsets the flag */
 
 		if (!XFS_BUF_ISSTALE(bp)) {
-			xfs_buf_delwri_queue(bp);
-			XFS_BUF_DONE(bp);
+			bp->b_flags |= XBF_WRITE | XBF_ASYNC | XBF_DONE;
+			xfs_buf_iorequest(bp);
+		} else {
+			xfs_buf_relse(bp);
 		}
-		ASSERT(bp->b_iodone != NULL);
-		trace_xfs_buf_item_iodone_async(bp, _RET_IP_);
-		xfs_buf_relse(bp);
+
 		return;
 	}
 
@@ -1045,6 +1006,6 @@ xfs_buf_iodone(
 	 * Either way, AIL is useless if we're forcing a shutdown.
 	 */
 	spin_lock(&ailp->xa_lock);
-	xfs_trans_ail_delete(ailp, lip);
+	xfs_trans_ail_delete(ailp, lip, SHUTDOWN_CORRUPT_INCORE);
 	xfs_buf_item_free(BUF_ITEM(lip));
 }
diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c
index 77c74257c2a..015b946c580 100644
--- a/fs/xfs/xfs_da_btree.c
+++ b/fs/xfs/xfs_da_btree.c
@@ -20,7 +20,6 @@
 #include "xfs_types.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -108,6 +107,8 @@ xfs_da_node_create(xfs_da_args_t *args, xfs_dablk_t blkno, int level,
 	int error;
 	xfs_trans_t *tp;
 
+	trace_xfs_da_node_create(args);
+
 	tp = args->trans;
 	error = xfs_da_get_buf(tp, args->dp, blkno, -1, &bp, whichfork);
 	if (error)
@@ -140,6 +141,8 @@ xfs_da_split(xfs_da_state_t *state)
 	xfs_dabuf_t *bp;
 	int max, action, error, i;
 
+	trace_xfs_da_split(state->args);
+
 	/*
 	 * Walk back up the tree splitting/inserting/adjusting as necessary.
 	 * If we need to insert and there isn't room, split the node, then
@@ -178,10 +181,12 @@ xfs_da_split(xfs_da_state_t *state)
 			state->extravalid = 1;
 			if (state->inleaf) {
 				state->extraafter = 0;	/* before newblk */
+				trace_xfs_attr_leaf_split_before(state->args);
 				error = xfs_attr_leaf_split(state, oldblk,
 							    &state->extrablk);
 			} else {
 				state->extraafter = 1;	/* after newblk */
+				trace_xfs_attr_leaf_split_after(state->args);
 				error = xfs_attr_leaf_split(state, newblk,
 							    &state->extrablk);
 			}
@@ -300,6 +305,8 @@ xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
 	xfs_mount_t *mp;
 	xfs_dir2_leaf_t *leaf;
 
+	trace_xfs_da_root_split(state->args);
+
 	/*
 	 * Copy the existing (incorrect) block from the root node position
 	 * to a free space somewhere.
@@ -380,6 +387,8 @@ xfs_da_node_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
 	int newcount, error;
 	int useextra;
 
+	trace_xfs_da_node_split(state->args);
+
 	node = oldblk->bp->data;
 	ASSERT(node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
 
@@ -466,6 +475,8 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
 	int count, tmp;
 	xfs_trans_t *tp;
 
+	trace_xfs_da_node_rebalance(state->args);
+
 	node1 = blk1->bp->data;
 	node2 = blk2->bp->data;
 	/*
@@ -574,6 +585,8 @@ xfs_da_node_add(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
 	xfs_da_node_entry_t *btree;
 	int tmp;
 
+	trace_xfs_da_node_add(state->args);
+
 	node = oldblk->bp->data;
 	ASSERT(node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
 	ASSERT((oldblk->index >= 0) && (oldblk->index <= be16_to_cpu(node->hdr.count)));
@@ -619,6 +632,8 @@ xfs_da_join(xfs_da_state_t *state)
 	xfs_da_state_blk_t *drop_blk, *save_blk;
 	int action, error;
 
+	trace_xfs_da_join(state->args);
+
 	action = 0;
 	drop_blk = &state->path.blk[ state->path.active-1 ];
 	save_blk = &state->altpath.blk[ state->path.active-1 ];
@@ -723,6 +738,8 @@ xfs_da_root_join(xfs_da_state_t *state, xfs_da_state_blk_t *root_blk)
 	xfs_dabuf_t *bp;
 	int error;
 
+	trace_xfs_da_root_join(state->args);
+
 	args = state->args;
 	ASSERT(args != NULL);
 	ASSERT(root_blk->magic == XFS_DA_NODE_MAGIC);
@@ -941,6 +958,8 @@ xfs_da_node_remove(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk)
 	xfs_da_node_entry_t *btree;
 	int tmp;
 
+	trace_xfs_da_node_remove(state->args);
+
 	node = drop_blk->bp->data;
 	ASSERT(drop_blk->index < be16_to_cpu(node->hdr.count));
 	ASSERT(drop_blk->index >= 0);
@@ -984,6 +1003,8 @@ xfs_da_node_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
 	int tmp;
 	xfs_trans_t *tp;
 
+	trace_xfs_da_node_unbalance(state->args);
+
 	drop_node = drop_blk->bp->data;
 	save_node = save_blk->bp->data;
 	ASSERT(drop_node->hdr.info.magic == cpu_to_be16(XFS_DA_NODE_MAGIC));
@@ -1230,6 +1251,7 @@ xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
 		/*
 		 * Link new block in before existing block.
 		 */
+		trace_xfs_da_link_before(args);
 		new_info->forw = cpu_to_be32(old_blk->blkno);
 		new_info->back = old_info->back;
 		if (old_info->back) {
@@ -1251,6 +1273,7 @@ xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
 		/*
 		 * Link new block in after existing block.
 		 */
+		trace_xfs_da_link_after(args);
 		new_info->forw = old_info->forw;
 		new_info->back = cpu_to_be32(old_blk->blkno);
 		if (old_info->forw) {
@@ -1348,6 +1371,7 @@ xfs_da_blk_unlink(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
 	 * Unlink the leaf block from the doubly linked chain of leaves.
 	 */
 	if (be32_to_cpu(save_info->back) == drop_blk->blkno) {
+		trace_xfs_da_unlink_back(args);
 		save_info->back = drop_info->back;
 		if (drop_info->back) {
 			error = xfs_da_read_buf(args->trans, args->dp,
@@ -1365,6 +1389,7 @@ xfs_da_blk_unlink(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
 			xfs_da_buf_done(bp);
 		}
 	} else {
+		trace_xfs_da_unlink_forward(args);
 		save_info->forw = drop_info->forw;
 		if (drop_info->forw) {
 			error = xfs_da_read_buf(args->trans, args->dp,
@@ -1652,6 +1677,8 @@ xfs_da_grow_inode(
 	int			count;
 	int			error;
 
+	trace_xfs_da_grow_inode(args);
+
 	if (args->whichfork == XFS_DATA_FORK) {
 		bno = args->dp->i_mount->m_dirleafblk;
 		count = args->dp->i_mount->m_dirblkfsbs;
@@ -1690,6 +1717,8 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
 	xfs_dir2_leaf_t *dead_leaf2;
 	xfs_dahash_t dead_hash;
 
+	trace_xfs_da_swap_lastblock(args);
+
 	dead_buf = *dead_bufp;
 	dead_blkno = *dead_blknop;
 	tp = args->trans;
@@ -1878,6 +1907,8 @@ xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno,
 	xfs_trans_t *tp;
 	xfs_mount_t *mp;
 
+	trace_xfs_da_shrink_inode(args);
+
 	dp = args->dp;
 	w = args->whichfork;
 	tp = args->trans;
@@ -2245,20 +2276,20 @@ xfs_da_buf_make(int nbuf, xfs_buf_t **bps)
 	if (nbuf == 1) {
 		dabuf->nbuf = 1;
 		bp = bps[0];
-		dabuf->bbcount = (short)BTOBB(XFS_BUF_COUNT(bp));
+		dabuf->bbcount = bp->b_length;
 		dabuf->data = bp->b_addr;
 		dabuf->bps[0] = bp;
 	} else {
 		dabuf->nbuf = nbuf;
 		for (i = 0, dabuf->bbcount = 0; i < nbuf; i++) {
 			dabuf->bps[i] = bp = bps[i];
-			dabuf->bbcount += BTOBB(XFS_BUF_COUNT(bp));
+			dabuf->bbcount += bp->b_length;
 		}
 		dabuf->data = kmem_alloc(BBTOB(dabuf->bbcount), KM_SLEEP);
-		for (i = off = 0; i < nbuf; i++, off += XFS_BUF_COUNT(bp)) {
+		for (i = off = 0; i < nbuf; i++, off += BBTOB(bp->b_length)) {
 			bp = bps[i];
 			memcpy((char *)dabuf->data + off, bp->b_addr,
-				XFS_BUF_COUNT(bp));
+				BBTOB(bp->b_length));
 		}
 	}
 	return dabuf;
@@ -2278,10 +2309,10 @@ xfs_da_buf_clean(xfs_dabuf_t *dabuf)
 		ASSERT(dabuf->nbuf > 1);
 		dabuf->dirty = 0;
 		for (i = off = 0; i < dabuf->nbuf;
-				i++, off += XFS_BUF_COUNT(bp)) {
+				i++, off += BBTOB(bp->b_length)) {
 			bp = dabuf->bps[i];
 			memcpy(bp->b_addr, dabuf->data + off,
-						XFS_BUF_COUNT(bp));
+						BBTOB(bp->b_length));
 		}
 	}
 }
@@ -2324,10 +2355,10 @@ xfs_da_log_buf(xfs_trans_t *tp, xfs_dabuf_t *dabuf, uint first, uint last)
 	}
 	dabuf->dirty = 1;
 	ASSERT(first <= last);
-	for (i = off = 0; i < dabuf->nbuf; i++, off += XFS_BUF_COUNT(bp)) {
+	for (i = off = 0; i < dabuf->nbuf; i++, off += BBTOB(bp->b_length)) {
 		bp = dabuf->bps[i];
 		f = off;
-		l = f + XFS_BUF_COUNT(bp) - 1;
+		l = f + BBTOB(bp->b_length) - 1;
 		if (f < first)
 			f = first;
 		if (l > last)
diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c
index 654dc6f05ba..e00de08dc8a 100644
--- a/fs/xfs/xfs_dfrag.c
+++ b/fs/xfs/xfs_dfrag.c
@@ -18,9 +18,7 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_types.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -163,12 +161,14 @@ xfs_swap_extents_check_format(
 
 	/* 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) > ip->i_df.if_ext_max)
+	    XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) >
+			XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK))
 		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) > tip->i_df.if_ext_max)
+	    XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) >
+			XFS_IFORK_MAXEXT(tip, XFS_DATA_FORK))
 		return EINVAL;
 
 	/*
@@ -180,18 +180,25 @@ xfs_swap_extents_check_format(
 	 * (a common defrag case) which will occur when the temp inode is in
 	 * extent format...
 	 */
-	if (tip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
-	    ((XFS_IFORK_BOFF(ip) &&
-	      tip->i_df.if_broot_bytes > XFS_IFORK_BOFF(ip)) ||
-	     XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) <= ip->i_df.if_ext_max))
-		return EINVAL;
+	if (tip->i_d.di_format == XFS_DINODE_FMT_BTREE) {
+		if (XFS_IFORK_BOFF(ip) &&
+		    tip->i_df.if_broot_bytes > XFS_IFORK_BOFF(ip))
+			return EINVAL;
+		if (XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) <=
+		    XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK))
+			return EINVAL;
+	}
 
 	/* Reciprocal target->temp btree format checks */
-	if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
-	    ((XFS_IFORK_BOFF(tip) &&
-	      ip->i_df.if_broot_bytes > XFS_IFORK_BOFF(tip)) ||
-	     XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) <= tip->i_df.if_ext_max))
-		return EINVAL;
+	if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE) {
+		if (XFS_IFORK_BOFF(tip) &&
+		    ip->i_df.if_broot_bytes > XFS_IFORK_BOFF(tip))
+			return EINVAL;
+
+		if (XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) <=
+		    XFS_IFORK_MAXEXT(tip, XFS_DATA_FORK))
+			return EINVAL;
+	}
 
 	return 0;
 }
@@ -206,7 +213,7 @@ xfs_swap_extents(
 	xfs_trans_t	*tp;
 	xfs_bstat_t	*sbp = &sxp->sx_stat;
 	xfs_ifork_t	*tempifp, *ifp, *tifp;
-	int		ilf_fields, tilf_fields;
+	int		src_log_flags, target_log_flags;
 	int		error = 0;
 	int		aforkblks = 0;
 	int		taforkblks = 0;
@@ -349,16 +356,6 @@ xfs_swap_extents(
 	*tifp = *tempifp;	/* struct copy */
 
 	/*
-	 * Fix the in-memory data fork values that are dependent on the fork
-	 * offset in the inode. We can't assume they remain the same as attr2
-	 * has dynamic fork offsets.
-	 */
-	ifp->if_ext_max = XFS_IFORK_SIZE(ip, XFS_DATA_FORK) /
-					(uint)sizeof(xfs_bmbt_rec_t);
-	tifp->if_ext_max = XFS_IFORK_SIZE(tip, XFS_DATA_FORK) /
-					(uint)sizeof(xfs_bmbt_rec_t);
-
-	/*
 	 * Fix the on-disk inode values
 	 */
 	tmp = (__uint64_t)ip->i_d.di_nblocks;
@@ -386,9 +383,8 @@ xfs_swap_extents(
 	tip->i_delayed_blks = ip->i_delayed_blks;
 	ip->i_delayed_blks = 0;
 
-	ilf_fields = XFS_ILOG_CORE;
-
-	switch(ip->i_d.di_format) {
+	src_log_flags = XFS_ILOG_CORE;
+	switch (ip->i_d.di_format) {
 	case XFS_DINODE_FMT_EXTENTS:
 		/* If the extents fit in the inode, fix the
 		 * pointer.  Otherwise it's already NULL or
@@ -398,16 +394,15 @@ xfs_swap_extents(
 			ifp->if_u1.if_extents =
 				ifp->if_u2.if_inline_ext;
 		}
-		ilf_fields |= XFS_ILOG_DEXT;
+		src_log_flags |= XFS_ILOG_DEXT;
 		break;
 	case XFS_DINODE_FMT_BTREE:
-		ilf_fields |= XFS_ILOG_DBROOT;
+		src_log_flags |= XFS_ILOG_DBROOT;
 		break;
 	}
 
-	tilf_fields = XFS_ILOG_CORE;
-
-	switch(tip->i_d.di_format) {
+	target_log_flags = XFS_ILOG_CORE;
+	switch (tip->i_d.di_format) {
 	case XFS_DINODE_FMT_EXTENTS:
 		/* If the extents fit in the inode, fix the
 		 * pointer.  Otherwise it's already NULL or
@@ -417,10 +412,10 @@ xfs_swap_extents(
 			tifp->if_u1.if_extents =
 				tifp->if_u2.if_inline_ext;
 		}
-		tilf_fields |= XFS_ILOG_DEXT;
+		target_log_flags |= XFS_ILOG_DEXT;
 		break;
 	case XFS_DINODE_FMT_BTREE:
-		tilf_fields |= XFS_ILOG_DBROOT;
+		target_log_flags |= XFS_ILOG_DBROOT;
 		break;
 	}
 
@@ -428,8 +423,8 @@ xfs_swap_extents(
 	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
 	xfs_trans_ijoin(tp, tip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
 
-	xfs_trans_log_inode(tp, ip,  ilf_fields);
-	xfs_trans_log_inode(tp, tip, tilf_fields);
+	xfs_trans_log_inode(tp, ip,  src_log_flags);
+	xfs_trans_log_inode(tp, tip, target_log_flags);
 
 	/*
 	 * If this is a synchronous mount, make sure that the
diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c
index a2e27010c7f..67a250c36d4 100644
--- a/fs/xfs/xfs_dir2.c
+++ b/fs/xfs/xfs_dir2.c
@@ -18,7 +18,6 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_types.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
 #include "xfs_inum.h"
 #include "xfs_trans.h"
diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/xfs_dir2_block.c
index 9245e029b8e..586732f2d80 100644
--- a/fs/xfs/xfs_dir2_block.c
+++ b/fs/xfs/xfs_dir2_block.c
@@ -19,7 +19,6 @@
 #include "xfs_fs.h"
 #include "xfs_types.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -29,6 +28,7 @@
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_inode_item.h"
+#include "xfs_dir2.h"
 #include "xfs_dir2_format.h"
 #include "xfs_dir2_priv.h"
 #include "xfs_error.h"
diff --git a/fs/xfs/xfs_dir2_data.c b/fs/xfs/xfs_dir2_data.c
index 5bbe2a8a023..2046988e9eb 100644
--- a/fs/xfs/xfs_dir2_data.c
+++ b/fs/xfs/xfs_dir2_data.c
@@ -19,7 +19,6 @@
 #include "xfs_fs.h"
 #include "xfs_types.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/xfs_dir2_leaf.c
index 66e108f561a..397ffbcbab1 100644
--- a/fs/xfs/xfs_dir2_leaf.c
+++ b/fs/xfs/xfs_dir2_leaf.c
@@ -20,7 +20,6 @@
 #include "xfs_types.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c
index 0179a41d9e5..b0f26780449 100644
--- a/fs/xfs/xfs_dir2_node.c
+++ b/fs/xfs/xfs_dir2_node.c
@@ -19,7 +19,6 @@
 #include "xfs_fs.h"
 #include "xfs_types.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/xfs_dir2_sf.c
index 79d05e84e29..19bf0c5e38f 100644
--- a/fs/xfs/xfs_dir2_sf.c
+++ b/fs/xfs/xfs_dir2_sf.c
@@ -19,7 +19,6 @@
 #include "xfs_fs.h"
 #include "xfs_types.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c
index 8a24f0c6c86..f9c3fe304a1 100644
--- a/fs/xfs/xfs_discard.c
+++ b/fs/xfs/xfs_discard.c
@@ -17,7 +17,6 @@
  */
 #include "xfs.h"
 #include "xfs_sb.h"
-#include "xfs_inum.h"
 #include "xfs_log.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
@@ -30,6 +29,7 @@
 #include "xfs_inode.h"
 #include "xfs_alloc.h"
 #include "xfs_error.h"
+#include "xfs_extent_busy.h"
 #include "xfs_discard.h"
 #include "xfs_trace.h"
 
@@ -37,9 +37,9 @@ STATIC int
 xfs_trim_extents(
 	struct xfs_mount	*mp,
 	xfs_agnumber_t		agno,
-	xfs_fsblock_t		start,
-	xfs_fsblock_t		end,
-	xfs_fsblock_t		minlen,
+	xfs_daddr_t		start,
+	xfs_daddr_t		end,
+	xfs_daddr_t		minlen,
 	__uint64_t		*blocks_trimmed)
 {
 	struct block_device	*bdev = mp->m_ddev_targp->bt_bdev;
@@ -67,8 +67,8 @@ xfs_trim_extents(
 	/*
 	 * Look up the longest btree in the AGF and start with it.
 	 */
-	error = xfs_alloc_lookup_le(cur, 0,
-				    XFS_BUF_TO_AGF(agbp)->agf_longest, &i);
+	error = xfs_alloc_lookup_ge(cur, 0,
+			    be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_longest), &i);
 	if (error)
 		goto out_del_cursor;
 
@@ -77,19 +77,29 @@ xfs_trim_extents(
 	 * enough to be worth discarding.
 	 */
 	while (i) {
-		xfs_agblock_t fbno;
-		xfs_extlen_t flen;
+		xfs_agblock_t	fbno;
+		xfs_extlen_t	flen;
+		xfs_daddr_t	dbno;
+		xfs_extlen_t	dlen;
 
 		error = xfs_alloc_get_rec(cur, &fbno, &flen, &i);
 		if (error)
 			goto out_del_cursor;
 		XFS_WANT_CORRUPTED_GOTO(i == 1, out_del_cursor);
-		ASSERT(flen <= XFS_BUF_TO_AGF(agbp)->agf_longest);
+		ASSERT(flen <= be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_longest));
+
+		/*
+		 * use daddr format for all range/len calculations as that is
+		 * the format the range/len variables are supplied in by
+		 * userspace.
+		 */
+		dbno = XFS_AGB_TO_DADDR(mp, agno, fbno);
+		dlen = XFS_FSB_TO_BB(mp, flen);
 
 		/*
 		 * Too small?  Give up.
 		 */
-		if (flen < minlen) {
+		if (dlen < minlen) {
 			trace_xfs_discard_toosmall(mp, agno, fbno, flen);
 			goto out_del_cursor;
 		}
@@ -99,8 +109,7 @@ xfs_trim_extents(
 		 * supposed to discard skip it.  Do not bother to trim
 		 * down partially overlapping ranges for now.
 		 */
-		if (XFS_AGB_TO_FSB(mp, agno, fbno) + flen < start ||
-		    XFS_AGB_TO_FSB(mp, agno, fbno) > end) {
+		if (dbno + dlen < start || dbno > end) {
 			trace_xfs_discard_exclude(mp, agno, fbno, flen);
 			goto next_extent;
 		}
@@ -109,16 +118,13 @@ xfs_trim_extents(
 		 * If any blocks in the range are still busy, skip the
 		 * discard and try again the next time.
 		 */
-		if (xfs_alloc_busy_search(mp, agno, fbno, flen)) {
+		if (xfs_extent_busy_search(mp, agno, fbno, flen)) {
 			trace_xfs_discard_busy(mp, agno, fbno, flen);
 			goto next_extent;
 		}
 
 		trace_xfs_discard_extent(mp, agno, fbno, flen);
-		error = -blkdev_issue_discard(bdev,
-				XFS_AGB_TO_DADDR(mp, agno, fbno),
-				XFS_FSB_TO_BB(mp, flen),
-				GFP_NOFS, 0);
+		error = -blkdev_issue_discard(bdev, dbno, dlen, GFP_NOFS, 0);
 		if (error)
 			goto out_del_cursor;
 		*blocks_trimmed += flen;
@@ -137,6 +143,15 @@ out_put_perag:
 	return error;
 }
 
+/*
+ * trim a range of the filesystem.
+ *
+ * Note: the parameters passed from userspace are byte ranges into the
+ * filesystem which does not match to the format we use for filesystem block
+ * addressing. FSB addressing is sparse (AGNO|AGBNO), while the incoming format
+ * is a linear address range. Hence we need to use DADDR based conversions and
+ * comparisons for determining the correct offset and regions to trim.
+ */
 int
 xfs_ioc_trim(
 	struct xfs_mount		*mp,
@@ -145,7 +160,7 @@ xfs_ioc_trim(
 	struct request_queue	*q = mp->m_ddev_targp->bt_bdev->bd_disk->queue;
 	unsigned int		granularity = q->limits.discard_granularity;
 	struct fstrim_range	range;
-	xfs_fsblock_t		start, end, minlen;
+	xfs_daddr_t		start, end, minlen;
 	xfs_agnumber_t		start_agno, end_agno, agno;
 	__uint64_t		blocks_trimmed = 0;
 	int			error, last_error = 0;
@@ -159,22 +174,22 @@ xfs_ioc_trim(
 
 	/*
 	 * Truncating down the len isn't actually quite correct, but using
-	 * XFS_B_TO_FSB would mean we trivially get overflows for values
+	 * BBTOB would mean we trivially get overflows for values
 	 * of ULLONG_MAX or slightly lower.  And ULLONG_MAX is the default
 	 * used by the fstrim application.  In the end it really doesn't
 	 * matter as trimming blocks is an advisory interface.
 	 */
-	start = XFS_B_TO_FSBT(mp, range.start);
-	end = start + XFS_B_TO_FSBT(mp, range.len) - 1;
-	minlen = XFS_B_TO_FSB(mp, max_t(u64, granularity, range.minlen));
+	start = BTOBB(range.start);
+	end = start + BTOBBT(range.len) - 1;
+	minlen = BTOBB(max_t(u64, granularity, range.minlen));
 
-	if (start >= mp->m_sb.sb_dblocks)
+	if (XFS_BB_TO_FSB(mp, start) >= mp->m_sb.sb_dblocks)
 		return -XFS_ERROR(EINVAL);
-	if (end > mp->m_sb.sb_dblocks - 1)
-		end = mp->m_sb.sb_dblocks - 1;
+	if (end > XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks) - 1)
+		end = XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks)- 1;
 
-	start_agno = XFS_FSB_TO_AGNO(mp, start);
-	end_agno = XFS_FSB_TO_AGNO(mp, end);
+	start_agno = xfs_daddr_to_agno(mp, start);
+	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,
@@ -197,7 +212,7 @@ xfs_discard_extents(
 	struct xfs_mount	*mp,
 	struct list_head	*list)
 {
-	struct xfs_busy_extent	*busyp;
+	struct xfs_extent_busy	*busyp;
 	int			error = 0;
 
 	list_for_each_entry(busyp, list, list) {
diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c
index 25d7280e9f6..bf27fcca484 100644
--- a/fs/xfs/xfs_dquot.c
+++ b/fs/xfs/xfs_dquot.c
@@ -19,7 +19,6 @@
 #include "xfs_fs.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -39,20 +38,18 @@
 #include "xfs_qm.h"
 #include "xfs_trace.h"
 
-
 /*
-   LOCK ORDER
-
-   inode lock		    (ilock)
-   dquot hash-chain lock    (hashlock)
-   xqm dquot freelist lock  (freelistlock
-   mount's dquot list lock  (mplistlock)
-   user dquot lock - lock ordering among dquots is based on the uid or gid
-   group dquot lock - similar to udquots. Between the two dquots, the udquot
-		      has to be locked first.
-   pin lock - the dquot lock must be held to take this lock.
-   flush lock - ditto.
-*/
+ * Lock order:
+ *
+ * ip->i_lock
+ *   qi->qi_tree_lock
+ *     dquot->q_qlock (xfs_dqlock() and friends)
+ *       dquot->q_flush (xfs_dqflock() and friends)
+ *       qi->qi_lru_lock
+ *
+ * If two dquots need to be locked the order is user before group/project,
+ * otherwise by the lowest id first, see xfs_dqlock2.
+ */
 
 #ifdef DEBUG
 xfs_buftarg_t *xfs_dqerror_target;
@@ -61,83 +58,10 @@ int xfs_dqreq_num;
 int xfs_dqerror_mod = 33;
 #endif
 
-static struct lock_class_key xfs_dquot_other_class;
-
-/*
- * Allocate and initialize a dquot. We don't always allocate fresh memory;
- * we try to reclaim a free dquot if the number of incore dquots are above
- * a threshold.
- * The only field inside the core that gets initialized at this point
- * is the d_id field. The idea is to fill in the entire q_core
- * when we read in the on disk dquot.
- */
-STATIC xfs_dquot_t *
-xfs_qm_dqinit(
-	xfs_mount_t  *mp,
-	xfs_dqid_t   id,
-	uint	     type)
-{
-	xfs_dquot_t	*dqp;
-	boolean_t	brandnewdquot;
-
-	brandnewdquot = xfs_qm_dqalloc_incore(&dqp);
-	dqp->dq_flags = type;
-	dqp->q_core.d_id = cpu_to_be32(id);
-	dqp->q_mount = mp;
-
-	/*
-	 * No need to re-initialize these if this is a reclaimed dquot.
-	 */
-	if (brandnewdquot) {
-		INIT_LIST_HEAD(&dqp->q_freelist);
-		mutex_init(&dqp->q_qlock);
-		init_waitqueue_head(&dqp->q_pinwait);
-
-		/*
-		 * Because we want to use a counting completion, complete
-		 * the flush completion once to allow a single access to
-		 * the flush completion without blocking.
-		 */
-		init_completion(&dqp->q_flush);
-		complete(&dqp->q_flush);
-
-		trace_xfs_dqinit(dqp);
-	} else {
-		/*
-		 * Only the q_core portion was zeroed in dqreclaim_one().
-		 * So, we need to reset others.
-		 */
-		dqp->q_nrefs = 0;
-		dqp->q_blkno = 0;
-		INIT_LIST_HEAD(&dqp->q_mplist);
-		INIT_LIST_HEAD(&dqp->q_hashlist);
-		dqp->q_bufoffset = 0;
-		dqp->q_fileoffset = 0;
-		dqp->q_transp = NULL;
-		dqp->q_gdquot = NULL;
-		dqp->q_res_bcount = 0;
-		dqp->q_res_icount = 0;
-		dqp->q_res_rtbcount = 0;
-		atomic_set(&dqp->q_pincount, 0);
-		dqp->q_hash = NULL;
-		ASSERT(list_empty(&dqp->q_freelist));
-
-		trace_xfs_dqreuse(dqp);
-	}
-
-	/*
-	 * In either case we need to make sure group quotas have a different
-	 * lock class than user quotas, to make sure lockdep knows we can
-	 * locks of one of each at the same time.
-	 */
-	if (!(type & XFS_DQ_USER))
-		lockdep_set_class(&dqp->q_qlock, &xfs_dquot_other_class);
+struct kmem_zone		*xfs_qm_dqtrxzone;
+static struct kmem_zone		*xfs_qm_dqzone;
 
-	/*
-	 * log item gets initialized later
-	 */
-	return (dqp);
-}
+static struct lock_class_key xfs_dquot_other_class;
 
 /*
  * This is called to free all the memory associated with a dquot
@@ -146,30 +70,12 @@ void
 xfs_qm_dqdestroy(
 	xfs_dquot_t	*dqp)
 {
-	ASSERT(list_empty(&dqp->q_freelist));
+	ASSERT(list_empty(&dqp->q_lru));
 
 	mutex_destroy(&dqp->q_qlock);
-	kmem_zone_free(xfs_Gqm->qm_dqzone, dqp);
+	kmem_zone_free(xfs_qm_dqzone, dqp);
 
-	atomic_dec(&xfs_Gqm->qm_totaldquots);
-}
-
-/*
- * This is what a 'fresh' dquot inside a dquot chunk looks like on disk.
- */
-STATIC void
-xfs_qm_dqinit_core(
-	xfs_dqid_t	id,
-	uint		type,
-	xfs_dqblk_t	*d)
-{
-	/*
-	 * Caller has zero'd the entire dquot 'chunk' already.
-	 */
-	d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC);
-	d->dd_diskdq.d_version = XFS_DQUOT_VERSION;
-	d->dd_diskdq.d_id = cpu_to_be32(id);
-	d->dd_diskdq.d_flags = type;
+	XFS_STATS_DEC(xs_qm_dquot);
 }
 
 /*
@@ -234,10 +140,10 @@ xfs_qm_adjust_dqtimers(
 
 	if (!d->d_btimer) {
 		if ((d->d_blk_softlimit &&
-		     (be64_to_cpu(d->d_bcount) >=
+		     (be64_to_cpu(d->d_bcount) >
 		      be64_to_cpu(d->d_blk_softlimit))) ||
 		    (d->d_blk_hardlimit &&
-		     (be64_to_cpu(d->d_bcount) >=
+		     (be64_to_cpu(d->d_bcount) >
 		      be64_to_cpu(d->d_blk_hardlimit)))) {
 			d->d_btimer = cpu_to_be32(get_seconds() +
 					mp->m_quotainfo->qi_btimelimit);
@@ -246,10 +152,10 @@ xfs_qm_adjust_dqtimers(
 		}
 	} else {
 		if ((!d->d_blk_softlimit ||
-		     (be64_to_cpu(d->d_bcount) <
+		     (be64_to_cpu(d->d_bcount) <=
 		      be64_to_cpu(d->d_blk_softlimit))) &&
 		    (!d->d_blk_hardlimit ||
-		    (be64_to_cpu(d->d_bcount) <
+		    (be64_to_cpu(d->d_bcount) <=
 		     be64_to_cpu(d->d_blk_hardlimit)))) {
 			d->d_btimer = 0;
 		}
@@ -257,10 +163,10 @@ xfs_qm_adjust_dqtimers(
 
 	if (!d->d_itimer) {
 		if ((d->d_ino_softlimit &&
-		     (be64_to_cpu(d->d_icount) >=
+		     (be64_to_cpu(d->d_icount) >
 		      be64_to_cpu(d->d_ino_softlimit))) ||
 		    (d->d_ino_hardlimit &&
-		     (be64_to_cpu(d->d_icount) >=
+		     (be64_to_cpu(d->d_icount) >
 		      be64_to_cpu(d->d_ino_hardlimit)))) {
 			d->d_itimer = cpu_to_be32(get_seconds() +
 					mp->m_quotainfo->qi_itimelimit);
@@ -269,10 +175,10 @@ xfs_qm_adjust_dqtimers(
 		}
 	} else {
 		if ((!d->d_ino_softlimit ||
-		     (be64_to_cpu(d->d_icount) <
+		     (be64_to_cpu(d->d_icount) <=
 		      be64_to_cpu(d->d_ino_softlimit)))  &&
 		    (!d->d_ino_hardlimit ||
-		     (be64_to_cpu(d->d_icount) <
+		     (be64_to_cpu(d->d_icount) <=
 		      be64_to_cpu(d->d_ino_hardlimit)))) {
 			d->d_itimer = 0;
 		}
@@ -280,10 +186,10 @@ xfs_qm_adjust_dqtimers(
 
 	if (!d->d_rtbtimer) {
 		if ((d->d_rtb_softlimit &&
-		     (be64_to_cpu(d->d_rtbcount) >=
+		     (be64_to_cpu(d->d_rtbcount) >
 		      be64_to_cpu(d->d_rtb_softlimit))) ||
 		    (d->d_rtb_hardlimit &&
-		     (be64_to_cpu(d->d_rtbcount) >=
+		     (be64_to_cpu(d->d_rtbcount) >
 		      be64_to_cpu(d->d_rtb_hardlimit)))) {
 			d->d_rtbtimer = cpu_to_be32(get_seconds() +
 					mp->m_quotainfo->qi_rtbtimelimit);
@@ -292,10 +198,10 @@ xfs_qm_adjust_dqtimers(
 		}
 	} else {
 		if ((!d->d_rtb_softlimit ||
-		     (be64_to_cpu(d->d_rtbcount) <
+		     (be64_to_cpu(d->d_rtbcount) <=
 		      be64_to_cpu(d->d_rtb_softlimit))) &&
 		    (!d->d_rtb_hardlimit ||
-		     (be64_to_cpu(d->d_rtbcount) <
+		     (be64_to_cpu(d->d_rtbcount) <=
 		      be64_to_cpu(d->d_rtb_hardlimit)))) {
 			d->d_rtbtimer = 0;
 		}
@@ -328,8 +234,13 @@ xfs_qm_init_dquot_blk(
 	curid = id - (id % q->qi_dqperchunk);
 	ASSERT(curid >= 0);
 	memset(d, 0, BBTOB(q->qi_dqchunklen));
-	for (i = 0; i < q->qi_dqperchunk; i++, d++, curid++)
-		xfs_qm_dqinit_core(curid, type, d);
+	for (i = 0; i < q->qi_dqperchunk; i++, d++, curid++) {
+		d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC);
+		d->dd_diskdq.d_version = XFS_DQUOT_VERSION;
+		d->dd_diskdq.d_id = cpu_to_be32(curid);
+		d->dd_diskdq.d_flags = type;
+	}
+
 	xfs_trans_dquot_buf(tp, bp,
 			    (type & XFS_DQ_USER ? XFS_BLF_UDQUOT_BUF :
 			    ((type & XFS_DQ_PROJ) ? XFS_BLF_PDQUOT_BUF :
@@ -372,7 +283,7 @@ xfs_qm_dqalloc(
 	 * Return if this type of quotas is turned off while we didn't
 	 * have an inode lock
 	 */
-	if (XFS_IS_THIS_QUOTA_OFF(dqp)) {
+	if (!xfs_this_quota_on(dqp->q_mount, dqp->dq_flags)) {
 		xfs_iunlock(quotip, XFS_ILOCK_EXCL);
 		return (ESRCH);
 	}
@@ -474,7 +385,7 @@ xfs_qm_dqtobp(
 	dqp->q_fileoffset = (xfs_fileoff_t)id / mp->m_quotainfo->qi_dqperchunk;
 
 	xfs_ilock(quotip, XFS_ILOCK_SHARED);
-	if (XFS_IS_THIS_QUOTA_OFF(dqp)) {
+	if (!xfs_this_quota_on(dqp->q_mount, dqp->dq_flags)) {
 		/*
 		 * Return if this type of quotas is turned off while we
 		 * didn't have the quota inode lock.
@@ -564,36 +475,87 @@ xfs_qm_dqtobp(
  * Read in the ondisk dquot using dqtobp() then copy it to an incore version,
  * and release the buffer immediately.
  *
+ * If XFS_QMOPT_DQALLOC is set, allocate a dquot on disk if it needed.
  */
-/* ARGSUSED */
-STATIC int
+int
 xfs_qm_dqread(
-	xfs_trans_t	**tpp,
-	xfs_dqid_t	id,
-	xfs_dquot_t	*dqp,	/* dquot to get filled in */
-	uint		flags)
+	struct xfs_mount	*mp,
+	xfs_dqid_t		id,
+	uint			type,
+	uint			flags,
+	struct xfs_dquot	**O_dqpp)
 {
-	xfs_disk_dquot_t *ddqp;
-	xfs_buf_t	 *bp;
-	int		 error;
-	xfs_trans_t	 *tp;
+	struct xfs_dquot	*dqp;
+	struct xfs_disk_dquot	*ddqp;
+	struct xfs_buf		*bp;
+	struct xfs_trans	*tp = NULL;
+	int			error;
+	int			cancelflags = 0;
+
+
+	dqp = kmem_zone_zalloc(xfs_qm_dqzone, KM_SLEEP);
+
+	dqp->dq_flags = type;
+	dqp->q_core.d_id = cpu_to_be32(id);
+	dqp->q_mount = mp;
+	INIT_LIST_HEAD(&dqp->q_lru);
+	mutex_init(&dqp->q_qlock);
+	init_waitqueue_head(&dqp->q_pinwait);
 
-	ASSERT(tpp);
+	/*
+	 * Because we want to use a counting completion, complete
+	 * the flush completion once to allow a single access to
+	 * the flush completion without blocking.
+	 */
+	init_completion(&dqp->q_flush);
+	complete(&dqp->q_flush);
+
+	/*
+	 * Make sure group quotas have a different lock class than user
+	 * quotas.
+	 */
+	if (!(type & XFS_DQ_USER))
+		lockdep_set_class(&dqp->q_qlock, &xfs_dquot_other_class);
+
+	XFS_STATS_INC(xs_qm_dquot);
 
 	trace_xfs_dqread(dqp);
 
+	if (flags & XFS_QMOPT_DQALLOC) {
+		tp = xfs_trans_alloc(mp, XFS_TRANS_QM_DQALLOC);
+		error = xfs_trans_reserve(tp, XFS_QM_DQALLOC_SPACE_RES(mp),
+				XFS_WRITE_LOG_RES(mp) +
+				/*
+				 * Round the chunklen up to the next multiple
+				 * of 128 (buf log item chunk size)).
+				 */
+				BBTOB(mp->m_quotainfo->qi_dqchunklen) - 1 + 128,
+				0,
+				XFS_TRANS_PERM_LOG_RES,
+				XFS_WRITE_LOG_COUNT);
+		if (error)
+			goto error1;
+		cancelflags = XFS_TRANS_RELEASE_LOG_RES;
+	}
+
 	/*
 	 * get a pointer to the on-disk dquot and the buffer containing it
 	 * dqp already knows its own type (GROUP/USER).
 	 */
-	if ((error = xfs_qm_dqtobp(tpp, dqp, &ddqp, &bp, flags))) {
-		return (error);
+	error = xfs_qm_dqtobp(&tp, dqp, &ddqp, &bp, flags);
+	if (error) {
+		/*
+		 * This can happen if quotas got turned off (ESRCH),
+		 * or if the dquot didn't exist on disk and we ask to
+		 * allocate (ENOENT).
+		 */
+		trace_xfs_dqread_fail(dqp);
+		cancelflags |= XFS_TRANS_ABORT;
+		goto error1;
 	}
-	tp = *tpp;
 
 	/* copy everything from disk dquot to the incore dquot */
 	memcpy(&dqp->q_core, ddqp, sizeof(xfs_disk_dquot_t));
-	ASSERT(be32_to_cpu(dqp->q_core.d_id) == id);
 	xfs_qm_dquot_logitem_init(dqp);
 
 	/*
@@ -622,173 +584,22 @@ xfs_qm_dqread(
 	ASSERT(xfs_buf_islocked(bp));
 	xfs_trans_brelse(tp, bp);
 
-	return (error);
-}
-
-
-/*
- * allocate an incore dquot from the kernel heap,
- * and fill its core with quota information kept on disk.
- * If XFS_QMOPT_DQALLOC is set, it'll allocate a dquot on disk
- * if it wasn't already allocated.
- */
-STATIC int
-xfs_qm_idtodq(
-	xfs_mount_t	*mp,
-	xfs_dqid_t	id,	 /* gid or uid, depending on type */
-	uint		type,	 /* UDQUOT or GDQUOT */
-	uint		flags,	 /* DQALLOC, DQREPAIR */
-	xfs_dquot_t	**O_dqpp)/* OUT : incore dquot, not locked */
-{
-	xfs_dquot_t	*dqp;
-	int		error;
-	xfs_trans_t	*tp;
-	int		cancelflags=0;
-
-	dqp = xfs_qm_dqinit(mp, id, type);
-	tp = NULL;
-	if (flags & XFS_QMOPT_DQALLOC) {
-		tp = xfs_trans_alloc(mp, XFS_TRANS_QM_DQALLOC);
-		error = xfs_trans_reserve(tp, XFS_QM_DQALLOC_SPACE_RES(mp),
-				XFS_WRITE_LOG_RES(mp) +
-				BBTOB(mp->m_quotainfo->qi_dqchunklen) - 1 +
-				128,
-				0,
-				XFS_TRANS_PERM_LOG_RES,
-				XFS_WRITE_LOG_COUNT);
-		if (error) {
-			cancelflags = 0;
-			goto error0;
-		}
-		cancelflags = XFS_TRANS_RELEASE_LOG_RES;
-	}
-
-	/*
-	 * Read it from disk; xfs_dqread() takes care of
-	 * all the necessary initialization of dquot's fields (locks, etc)
-	 */
-	if ((error = xfs_qm_dqread(&tp, id, dqp, flags))) {
-		/*
-		 * This can happen if quotas got turned off (ESRCH),
-		 * or if the dquot didn't exist on disk and we ask to
-		 * allocate (ENOENT).
-		 */
-		trace_xfs_dqread_fail(dqp);
-		cancelflags |= XFS_TRANS_ABORT;
-		goto error0;
-	}
 	if (tp) {
-		if ((error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES)))
-			goto error1;
+		error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
+		if (error)
+			goto error0;
 	}
 
 	*O_dqpp = dqp;
-	return (0);
+	return error;
 
- error0:
-	ASSERT(error);
+error1:
 	if (tp)
 		xfs_trans_cancel(tp, cancelflags);
- error1:
+error0:
 	xfs_qm_dqdestroy(dqp);
 	*O_dqpp = NULL;
-	return (error);
-}
-
-/*
- * Lookup a dquot in the incore dquot hashtable. We keep two separate
- * hashtables for user and group dquots; and, these are global tables
- * inside the XQM, not per-filesystem tables.
- * The hash chain must be locked by caller, and it is left locked
- * on return. Returning dquot is locked.
- */
-STATIC int
-xfs_qm_dqlookup(
-	xfs_mount_t		*mp,
-	xfs_dqid_t		id,
-	xfs_dqhash_t		*qh,
-	xfs_dquot_t		**O_dqpp)
-{
-	xfs_dquot_t		*dqp;
-	uint			flist_locked;
-
-	ASSERT(mutex_is_locked(&qh->qh_lock));
-
-	flist_locked = B_FALSE;
-
-	/*
-	 * Traverse the hashchain looking for a match
-	 */
-	list_for_each_entry(dqp, &qh->qh_list, q_hashlist) {
-		/*
-		 * We already have the hashlock. We don't need the
-		 * dqlock to look at the id field of the dquot, since the
-		 * id can't be modified without the hashlock anyway.
-		 */
-		if (be32_to_cpu(dqp->q_core.d_id) == id && dqp->q_mount == mp) {
-			trace_xfs_dqlookup_found(dqp);
-
-			/*
-			 * All in core dquots must be on the dqlist of mp
-			 */
-			ASSERT(!list_empty(&dqp->q_mplist));
-
-			xfs_dqlock(dqp);
-			if (dqp->q_nrefs == 0) {
-				ASSERT(!list_empty(&dqp->q_freelist));
-				if (!mutex_trylock(&xfs_Gqm->qm_dqfrlist_lock)) {
-					trace_xfs_dqlookup_want(dqp);
-
-					/*
-					 * We may have raced with dqreclaim_one()
-					 * (and lost). So, flag that we don't
-					 * want the dquot to be reclaimed.
-					 */
-					dqp->dq_flags |= XFS_DQ_WANT;
-					xfs_dqunlock(dqp);
-					mutex_lock(&xfs_Gqm->qm_dqfrlist_lock);
-					xfs_dqlock(dqp);
-					dqp->dq_flags &= ~(XFS_DQ_WANT);
-				}
-				flist_locked = B_TRUE;
-			}
-
-			/*
-			 * id couldn't have changed; we had the hashlock all
-			 * along
-			 */
-			ASSERT(be32_to_cpu(dqp->q_core.d_id) == id);
-
-			if (flist_locked) {
-				if (dqp->q_nrefs != 0) {
-					mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
-					flist_locked = B_FALSE;
-				} else {
-					/* take it off the freelist */
-					trace_xfs_dqlookup_freelist(dqp);
-					list_del_init(&dqp->q_freelist);
-					xfs_Gqm->qm_dqfrlist_cnt--;
-				}
-			}
-
-			XFS_DQHOLD(dqp);
-
-			if (flist_locked)
-				mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
-			/*
-			 * move the dquot to the front of the hashchain
-			 */
-			ASSERT(mutex_is_locked(&qh->qh_lock));
-			list_move(&dqp->q_hashlist, &qh->qh_list);
-			trace_xfs_dqlookup_done(dqp);
-			*O_dqpp = dqp;
-			return 0;
-		}
-	}
-
-	*O_dqpp = NULL;
-	ASSERT(mutex_is_locked(&qh->qh_lock));
-	return (1);
+	return error;
 }
 
 /*
@@ -808,10 +619,10 @@ xfs_qm_dqget(
 	uint		flags,	  /* DQALLOC, DQSUSER, DQREPAIR, DOWARN */
 	xfs_dquot_t	**O_dqpp) /* OUT : locked incore dquot */
 {
-	xfs_dquot_t	*dqp;
-	xfs_dqhash_t	*h;
-	uint		version;
-	int		error;
+	struct xfs_quotainfo	*qi = mp->m_quotainfo;
+	struct radix_tree_root *tree = XFS_DQUOT_TREE(qi, type);
+	struct xfs_dquot	*dqp;
+	int			error;
 
 	ASSERT(XFS_IS_QUOTA_RUNNING(mp));
 	if ((! XFS_IS_UQUOTA_ON(mp) && type == XFS_DQ_USER) ||
@@ -819,7 +630,6 @@ xfs_qm_dqget(
 	    (! XFS_IS_GQUOTA_ON(mp) && type == XFS_DQ_GROUP)) {
 		return (ESRCH);
 	}
-	h = XFS_DQ_HASH(mp, id, type);
 
 #ifdef DEBUG
 	if (xfs_do_dqerror) {
@@ -829,42 +639,39 @@ xfs_qm_dqget(
 			return (EIO);
 		}
 	}
-#endif
 
- again:
-
-#ifdef DEBUG
 	ASSERT(type == XFS_DQ_USER ||
 	       type == XFS_DQ_PROJ ||
 	       type == XFS_DQ_GROUP);
 	if (ip) {
 		ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-		if (type == XFS_DQ_USER)
-			ASSERT(ip->i_udquot == NULL);
-		else
-			ASSERT(ip->i_gdquot == NULL);
+		ASSERT(xfs_inode_dquot(ip, type) == NULL);
 	}
 #endif
-	mutex_lock(&h->qh_lock);
 
-	/*
-	 * Look in the cache (hashtable).
-	 * The chain is kept locked during lookup.
-	 */
-	if (xfs_qm_dqlookup(mp, id, h, O_dqpp) == 0) {
-		XQM_STATS_INC(xqmstats.xs_qm_dqcachehits);
-		/*
-		 * The dquot was found, moved to the front of the chain,
-		 * taken off the freelist if it was on it, and locked
-		 * at this point. Just unlock the hashchain and return.
-		 */
-		ASSERT(*O_dqpp);
-		ASSERT(XFS_DQ_IS_LOCKED(*O_dqpp));
-		mutex_unlock(&h->qh_lock);
-		trace_xfs_dqget_hit(*O_dqpp);
-		return (0);	/* success */
+restart:
+	mutex_lock(&qi->qi_tree_lock);
+	dqp = radix_tree_lookup(tree, id);
+	if (dqp) {
+		xfs_dqlock(dqp);
+		if (dqp->dq_flags & XFS_DQ_FREEING) {
+			xfs_dqunlock(dqp);
+			mutex_unlock(&qi->qi_tree_lock);
+			trace_xfs_dqget_freeing(dqp);
+			delay(1);
+			goto restart;
+		}
+
+		dqp->q_nrefs++;
+		mutex_unlock(&qi->qi_tree_lock);
+
+		trace_xfs_dqget_hit(dqp);
+		XFS_STATS_INC(xs_qm_dqcachehits);
+		*O_dqpp = dqp;
+		return 0;
 	}
-	XQM_STATS_INC(xqmstats.xs_qm_dqcachemisses);
+	mutex_unlock(&qi->qi_tree_lock);
+	XFS_STATS_INC(xs_qm_dqcachemisses);
 
 	/*
 	 * Dquot cache miss. We don't want to keep the inode lock across
@@ -875,130 +682,62 @@ xfs_qm_dqget(
 	 */
 	if (ip)
 		xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	/*
-	 * Save the hashchain version stamp, and unlock the chain, so that
-	 * we don't keep the lock across a disk read
-	 */
-	version = h->qh_version;
-	mutex_unlock(&h->qh_lock);
 
-	/*
-	 * Allocate the dquot on the kernel heap, and read the ondisk
-	 * portion off the disk. Also, do all the necessary initialization
-	 * This can return ENOENT if dquot didn't exist on disk and we didn't
-	 * ask it to allocate; ESRCH if quotas got turned off suddenly.
-	 */
-	if ((error = xfs_qm_idtodq(mp, id, type,
-				  flags & (XFS_QMOPT_DQALLOC|XFS_QMOPT_DQREPAIR|
-					   XFS_QMOPT_DOWARN),
-				  &dqp))) {
-		if (ip)
-			xfs_ilock(ip, XFS_ILOCK_EXCL);
-		return (error);
-	}
-
-	/*
-	 * See if this is mount code calling to look at the overall quota limits
-	 * which are stored in the id == 0 user or group's dquot.
-	 * Since we may not have done a quotacheck by this point, just return
-	 * the dquot without attaching it to any hashtables, lists, etc, or even
-	 * taking a reference.
-	 * The caller must dqdestroy this once done.
-	 */
-	if (flags & XFS_QMOPT_DQSUSER) {
-		ASSERT(id == 0);
-		ASSERT(! ip);
-		goto dqret;
-	}
+	error = xfs_qm_dqread(mp, id, type, flags, &dqp);
 
-	/*
-	 * Dquot lock comes after hashlock in the lock ordering
-	 */
-	if (ip) {
+	if (ip)
 		xfs_ilock(ip, XFS_ILOCK_EXCL);
 
+	if (error)
+		return error;
+
+	if (ip) {
 		/*
 		 * A dquot could be attached to this inode by now, since
 		 * we had dropped the ilock.
 		 */
-		if (type == XFS_DQ_USER) {
-			if (!XFS_IS_UQUOTA_ON(mp)) {
-				/* inode stays locked on return */
-				xfs_qm_dqdestroy(dqp);
-				return XFS_ERROR(ESRCH);
-			}
-			if (ip->i_udquot) {
+		if (xfs_this_quota_on(mp, type)) {
+			struct xfs_dquot	*dqp1;
+
+			dqp1 = xfs_inode_dquot(ip, type);
+			if (dqp1) {
 				xfs_qm_dqdestroy(dqp);
-				dqp = ip->i_udquot;
+				dqp = dqp1;
 				xfs_dqlock(dqp);
 				goto dqret;
 			}
 		} else {
-			if (!XFS_IS_OQUOTA_ON(mp)) {
-				/* inode stays locked on return */
-				xfs_qm_dqdestroy(dqp);
-				return XFS_ERROR(ESRCH);
-			}
-			if (ip->i_gdquot) {
-				xfs_qm_dqdestroy(dqp);
-				dqp = ip->i_gdquot;
-				xfs_dqlock(dqp);
-				goto dqret;
-			}
+			/* inode stays locked on return */
+			xfs_qm_dqdestroy(dqp);
+			return XFS_ERROR(ESRCH);
 		}
 	}
 
-	/*
-	 * Hashlock comes after ilock in lock order
-	 */
-	mutex_lock(&h->qh_lock);
-	if (version != h->qh_version) {
-		xfs_dquot_t *tmpdqp;
+	mutex_lock(&qi->qi_tree_lock);
+	error = -radix_tree_insert(tree, id, dqp);
+	if (unlikely(error)) {
+		WARN_ON(error != EEXIST);
+
 		/*
-		 * Now, see if somebody else put the dquot in the
-		 * hashtable before us. This can happen because we didn't
-		 * keep the hashchain lock. We don't have to worry about
-		 * lock order between the two dquots here since dqp isn't
-		 * on any findable lists yet.
+		 * Duplicate found. Just throw away the new dquot and start
+		 * over.
 		 */
-		if (xfs_qm_dqlookup(mp, id, h, &tmpdqp) == 0) {
-			/*
-			 * Duplicate found. Just throw away the new dquot
-			 * and start over.
-			 */
-			xfs_qm_dqput(tmpdqp);
-			mutex_unlock(&h->qh_lock);
-			xfs_qm_dqdestroy(dqp);
-			XQM_STATS_INC(xqmstats.xs_qm_dquot_dups);
-			goto again;
-		}
+		mutex_unlock(&qi->qi_tree_lock);
+		trace_xfs_dqget_dup(dqp);
+		xfs_qm_dqdestroy(dqp);
+		XFS_STATS_INC(xs_qm_dquot_dups);
+		goto restart;
 	}
 
 	/*
-	 * Put the dquot at the beginning of the hash-chain and mp's list
-	 * LOCK ORDER: hashlock, freelistlock, mplistlock, udqlock, gdqlock ..
-	 */
-	ASSERT(mutex_is_locked(&h->qh_lock));
-	dqp->q_hash = h;
-	list_add(&dqp->q_hashlist, &h->qh_list);
-	h->qh_version++;
-
-	/*
-	 * Attach this dquot to this filesystem's list of all dquots,
-	 * kept inside the mount structure in m_quotainfo field
-	 */
-	mutex_lock(&mp->m_quotainfo->qi_dqlist_lock);
-
-	/*
 	 * We return a locked dquot to the caller, with a reference taken
 	 */
 	xfs_dqlock(dqp);
 	dqp->q_nrefs = 1;
 
-	list_add(&dqp->q_mplist, &mp->m_quotainfo->qi_dqlist);
-	mp->m_quotainfo->qi_dquots++;
-	mutex_unlock(&mp->m_quotainfo->qi_dqlist_lock);
-	mutex_unlock(&h->qh_lock);
+	qi->qi_dquots++;
+	mutex_unlock(&qi->qi_tree_lock);
+
  dqret:
 	ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL));
 	trace_xfs_dqget_miss(dqp);
@@ -1007,75 +746,61 @@ xfs_qm_dqget(
 }
 
 
-/*
- * Release a reference to the dquot (decrement ref-count)
- * and unlock it. If there is a group quota attached to this
- * dquot, carefully release that too without tripping over
- * deadlocks'n'stuff.
- */
-void
-xfs_qm_dqput(
-	xfs_dquot_t	*dqp)
+STATIC void
+xfs_qm_dqput_final(
+	struct xfs_dquot	*dqp)
 {
-	xfs_dquot_t	*gdqp;
+	struct xfs_quotainfo	*qi = dqp->q_mount->m_quotainfo;
+	struct xfs_dquot	*gdqp;
 
-	ASSERT(dqp->q_nrefs > 0);
-	ASSERT(XFS_DQ_IS_LOCKED(dqp));
-
-	trace_xfs_dqput(dqp);
+	trace_xfs_dqput_free(dqp);
 
-	if (dqp->q_nrefs != 1) {
-		dqp->q_nrefs--;
-		xfs_dqunlock(dqp);
-		return;
+	mutex_lock(&qi->qi_lru_lock);
+	if (list_empty(&dqp->q_lru)) {
+		list_add_tail(&dqp->q_lru, &qi->qi_lru_list);
+		qi->qi_lru_count++;
+		XFS_STATS_INC(xs_qm_dquot_unused);
 	}
+	mutex_unlock(&qi->qi_lru_lock);
 
 	/*
-	 * drop the dqlock and acquire the freelist and dqlock
-	 * in the right order; but try to get it out-of-order first
+	 * If we just added a udquot to the freelist, then we want to release
+	 * the gdquot reference that it (probably) has. Otherwise it'll keep
+	 * the gdquot from getting reclaimed.
 	 */
-	if (!mutex_trylock(&xfs_Gqm->qm_dqfrlist_lock)) {
-		trace_xfs_dqput_wait(dqp);
-		xfs_dqunlock(dqp);
-		mutex_lock(&xfs_Gqm->qm_dqfrlist_lock);
-		xfs_dqlock(dqp);
+	gdqp = dqp->q_gdquot;
+	if (gdqp) {
+		xfs_dqlock(gdqp);
+		dqp->q_gdquot = NULL;
 	}
+	xfs_dqunlock(dqp);
 
-	while (1) {
-		gdqp = NULL;
+	/*
+	 * If we had a group quota hint, release it now.
+	 */
+	if (gdqp)
+		xfs_qm_dqput(gdqp);
+}
 
-		/* We can't depend on nrefs being == 1 here */
-		if (--dqp->q_nrefs == 0) {
-			trace_xfs_dqput_free(dqp);
+/*
+ * Release a reference to the dquot (decrement ref-count) and unlock it.
+ *
+ * If there is a group quota attached to this dquot, carefully release that
+ * too without tripping over deadlocks'n'stuff.
+ */
+void
+xfs_qm_dqput(
+	struct xfs_dquot	*dqp)
+{
+	ASSERT(dqp->q_nrefs > 0);
+	ASSERT(XFS_DQ_IS_LOCKED(dqp));
 
-			list_add_tail(&dqp->q_freelist, &xfs_Gqm->qm_dqfrlist);
-			xfs_Gqm->qm_dqfrlist_cnt++;
+	trace_xfs_dqput(dqp);
 
-			/*
-			 * If we just added a udquot to the freelist, then
-			 * we want to release the gdquot reference that
-			 * it (probably) has. Otherwise it'll keep the
-			 * gdquot from getting reclaimed.
-			 */
-			if ((gdqp = dqp->q_gdquot)) {
-				/*
-				 * Avoid a recursive dqput call
-				 */
-				xfs_dqlock(gdqp);
-				dqp->q_gdquot = NULL;
-			}
-		}
+	if (--dqp->q_nrefs > 0)
 		xfs_dqunlock(dqp);
-
-		/*
-		 * If we had a group quota inside the user quota as a hint,
-		 * release it now.
-		 */
-		if (! gdqp)
-			break;
-		dqp = gdqp;
-	}
-	mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
+	else
+		xfs_qm_dqput_final(dqp);
 }
 
 /*
@@ -1131,7 +856,7 @@ xfs_qm_dqflush_done(
 		/* xfs_trans_ail_delete() drops the AIL lock. */
 		spin_lock(&ailp->xa_lock);
 		if (lip->li_lsn == qip->qli_flush_lsn)
-			xfs_trans_ail_delete(ailp, lip);
+			xfs_trans_ail_delete(ailp, lip, SHUTDOWN_CORRUPT_INCORE);
 		else
 			spin_unlock(&ailp->xa_lock);
 	}
@@ -1152,8 +877,8 @@ xfs_qm_dqflush_done(
  */
 int
 xfs_qm_dqflush(
-	xfs_dquot_t		*dqp,
-	uint			flags)
+	struct xfs_dquot	*dqp,
+	struct xfs_buf		**bpp)
 {
 	struct xfs_mount	*mp = dqp->q_mount;
 	struct xfs_buf		*bp;
@@ -1165,25 +890,30 @@ xfs_qm_dqflush(
 
 	trace_xfs_dqflush(dqp);
 
-	/*
-	 * If not dirty, or it's pinned and we are not supposed to block, nada.
-	 */
-	if (!XFS_DQ_IS_DIRTY(dqp) ||
-	    (!(flags & SYNC_WAIT) && atomic_read(&dqp->q_pincount) > 0)) {
-		xfs_dqfunlock(dqp);
-		return 0;
-	}
+	*bpp = NULL;
+
 	xfs_qm_dqunpin_wait(dqp);
 
 	/*
 	 * This may have been unpinned because the filesystem is shutting
 	 * down forcibly. If that's the case we must not write this dquot
-	 * to disk, because the log record didn't make it to disk!
+	 * to disk, because the log record didn't make it to disk.
+	 *
+	 * We also have to remove the log item from the AIL in this case,
+	 * as we wait for an emptry AIL as part of the unmount process.
 	 */
 	if (XFS_FORCED_SHUTDOWN(mp)) {
+		struct xfs_log_item	*lip = &dqp->q_logitem.qli_item;
 		dqp->dq_flags &= ~XFS_DQ_DIRTY;
-		xfs_dqfunlock(dqp);
-		return XFS_ERROR(EIO);
+
+		spin_lock(&mp->m_ail->xa_lock);
+		if (lip->li_flags & XFS_LI_IN_AIL)
+			xfs_trans_ail_delete(mp->m_ail, lip,
+					     SHUTDOWN_CORRUPT_INCORE);
+		else
+			spin_unlock(&mp->m_ail->xa_lock);
+		error = XFS_ERROR(EIO);
+		goto out_unlock;
 	}
 
 	/*
@@ -1191,11 +921,8 @@ xfs_qm_dqflush(
 	 */
 	error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, dqp->q_blkno,
 				   mp->m_quotainfo->qi_dqchunklen, 0, &bp);
-	if (error) {
-		ASSERT(error != ENOENT);
-		xfs_dqfunlock(dqp);
-		return error;
-	}
+	if (error)
+		goto out_unlock;
 
 	/*
 	 * Calculate the location of the dquot inside the buffer.
@@ -1241,54 +968,13 @@ xfs_qm_dqflush(
 		xfs_log_force(mp, 0);
 	}
 
-	if (flags & SYNC_WAIT)
-		error = xfs_bwrite(bp);
-	else
-		xfs_buf_delwri_queue(bp);
-
-	xfs_buf_relse(bp);
-
 	trace_xfs_dqflush_done(dqp);
+	*bpp = bp;
+	return 0;
 
-	/*
-	 * dqp is still locked, but caller is free to unlock it now.
-	 */
-	return error;
-
-}
-
-int
-xfs_qm_dqlock_nowait(
-	xfs_dquot_t *dqp)
-{
-	return mutex_trylock(&dqp->q_qlock);
-}
-
-void
-xfs_dqlock(
-	xfs_dquot_t *dqp)
-{
-	mutex_lock(&dqp->q_qlock);
-}
-
-void
-xfs_dqunlock(
-	xfs_dquot_t *dqp)
-{
-	mutex_unlock(&(dqp->q_qlock));
-	if (dqp->q_logitem.qli_dquot == dqp) {
-		/* Once was dqp->q_mount, but might just have been cleared */
-		xfs_trans_unlocked_item(dqp->q_logitem.qli_item.li_ailp,
-					(xfs_log_item_t*)&(dqp->q_logitem));
-	}
-}
-
-
-void
-xfs_dqunlock_nonotify(
-	xfs_dquot_t *dqp)
-{
-	mutex_unlock(&(dqp->q_qlock));
+out_unlock:
+	xfs_dqfunlock(dqp);
+	return XFS_ERROR(EIO);
 }
 
 /*
@@ -1319,138 +1005,30 @@ xfs_dqlock2(
 	}
 }
 
-
-/*
- * Take a dquot out of the mount's dqlist as well as the hashlist.
- * This is called via unmount as well as quotaoff, and the purge
- * will always succeed unless there are soft (temp) references
- * outstanding.
- *
- * This returns 0 if it was purged, 1 if it wasn't. It's not an error code
- * that we're returning! XXXsup - not cool.
- */
-/* ARGSUSED */
-int
-xfs_qm_dqpurge(
-	xfs_dquot_t	*dqp)
+int __init
+xfs_qm_init(void)
 {
-	xfs_dqhash_t	*qh = dqp->q_hash;
-	xfs_mount_t	*mp = dqp->q_mount;
-
-	ASSERT(mutex_is_locked(&mp->m_quotainfo->qi_dqlist_lock));
-	ASSERT(mutex_is_locked(&dqp->q_hash->qh_lock));
+	xfs_qm_dqzone =
+		kmem_zone_init(sizeof(struct xfs_dquot), "xfs_dquot");
+	if (!xfs_qm_dqzone)
+		goto out;
 
-	xfs_dqlock(dqp);
-	/*
-	 * We really can't afford to purge a dquot that is
-	 * referenced, because these are hard refs.
-	 * It shouldn't happen in general because we went thru _all_ inodes in
-	 * dqrele_all_inodes before calling this and didn't let the mountlock go.
-	 * However it is possible that we have dquots with temporary
-	 * references that are not attached to an inode. e.g. see xfs_setattr().
-	 */
-	if (dqp->q_nrefs != 0) {
-		xfs_dqunlock(dqp);
-		mutex_unlock(&dqp->q_hash->qh_lock);
-		return (1);
-	}
-
-	ASSERT(!list_empty(&dqp->q_freelist));
-
-	/*
-	 * If we're turning off quotas, we have to make sure that, for
-	 * example, we don't delete quota disk blocks while dquots are
-	 * in the process of getting written to those disk blocks.
-	 * This dquot might well be on AIL, and we can't leave it there
-	 * if we're turning off quotas. Basically, we need this flush
-	 * lock, and are willing to block on it.
-	 */
-	if (!xfs_dqflock_nowait(dqp)) {
-		/*
-		 * Block on the flush lock after nudging dquot buffer,
-		 * if it is incore.
-		 */
-		xfs_qm_dqflock_pushbuf_wait(dqp);
-	}
-
-	/*
-	 * XXXIf we're turning this type of quotas off, we don't care
-	 * about the dirty metadata sitting in this dquot. OTOH, if
-	 * we're unmounting, we do care, so we flush it and wait.
-	 */
-	if (XFS_DQ_IS_DIRTY(dqp)) {
-		int	error;
+	xfs_qm_dqtrxzone =
+		kmem_zone_init(sizeof(struct xfs_dquot_acct), "xfs_dqtrx");
+	if (!xfs_qm_dqtrxzone)
+		goto out_free_dqzone;
 
-		/* dqflush unlocks dqflock */
-		/*
-		 * Given that dqpurge is a very rare occurrence, it is OK
-		 * that we're holding the hashlist and mplist locks
-		 * across the disk write. But, ... XXXsup
-		 *
-		 * We don't care about getting disk errors here. We need
-		 * to purge this dquot anyway, so we go ahead regardless.
-		 */
-		error = xfs_qm_dqflush(dqp, SYNC_WAIT);
-		if (error)
-			xfs_warn(mp, "%s: dquot %p flush failed",
-				__func__, dqp);
-		xfs_dqflock(dqp);
-	}
-	ASSERT(atomic_read(&dqp->q_pincount) == 0);
-	ASSERT(XFS_FORCED_SHUTDOWN(mp) ||
-	       !(dqp->q_logitem.qli_item.li_flags & XFS_LI_IN_AIL));
-
-	list_del_init(&dqp->q_hashlist);
-	qh->qh_version++;
-	list_del_init(&dqp->q_mplist);
-	mp->m_quotainfo->qi_dqreclaims++;
-	mp->m_quotainfo->qi_dquots--;
-	/*
-	 * XXX Move this to the front of the freelist, if we can get the
-	 * freelist lock.
-	 */
-	ASSERT(!list_empty(&dqp->q_freelist));
+	return 0;
 
-	dqp->q_mount = NULL;
-	dqp->q_hash = NULL;
-	dqp->dq_flags = XFS_DQ_INACTIVE;
-	memset(&dqp->q_core, 0, sizeof(dqp->q_core));
-	xfs_dqfunlock(dqp);
-	xfs_dqunlock(dqp);
-	mutex_unlock(&qh->qh_lock);
-	return (0);
+out_free_dqzone:
+	kmem_zone_destroy(xfs_qm_dqzone);
+out:
+	return -ENOMEM;
 }
 
-
-/*
- * Give the buffer a little push if it is incore and
- * wait on the flush lock.
- */
 void
-xfs_qm_dqflock_pushbuf_wait(
-	xfs_dquot_t	*dqp)
+xfs_qm_exit(void)
 {
-	xfs_mount_t	*mp = dqp->q_mount;
-	xfs_buf_t	*bp;
-
-	/*
-	 * Check to see if the dquot has been flushed delayed
-	 * write.  If so, grab its buffer and send it
-	 * out immediately.  We'll be able to acquire
-	 * the flush lock when the I/O completes.
-	 */
-	bp = xfs_incore(mp->m_ddev_targp, dqp->q_blkno,
-			mp->m_quotainfo->qi_dqchunklen, XBF_TRYLOCK);
-	if (!bp)
-		goto out_lock;
-
-	if (XFS_BUF_ISDELAYWRITE(bp)) {
-		if (xfs_buf_ispinned(bp))
-			xfs_log_force(mp, 0);
-		xfs_buf_delwri_promote(bp);
-		wake_up_process(bp->b_target->bt_task);
-	}
-	xfs_buf_relse(bp);
-out_lock:
-	xfs_dqflock(dqp);
+	kmem_zone_destroy(xfs_qm_dqtrxzone);
+	kmem_zone_destroy(xfs_qm_dqzone);
 }
diff --git a/fs/xfs/xfs_dquot.h b/fs/xfs/xfs_dquot.h
index 34b7e945dbf..7d20af27346 100644
--- a/fs/xfs/xfs_dquot.h
+++ b/fs/xfs/xfs_dquot.h
@@ -29,16 +29,6 @@
  * when quotas are off.
  */
 
-/*
- * The hash chain headers (hash buckets)
- */
-typedef struct xfs_dqhash {
-	struct list_head  qh_list;
-	struct mutex	  qh_lock;
-	uint		  qh_version;	/* ever increasing version */
-	uint		  qh_nelems;	/* number of dquots on the list */
-} xfs_dqhash_t;
-
 struct xfs_mount;
 struct xfs_trans;
 
@@ -47,10 +37,7 @@ struct xfs_trans;
  */
 typedef struct xfs_dquot {
 	uint		 dq_flags;	/* various flags (XFS_DQ_*) */
-	struct list_head q_freelist;	/* global free list of dquots */
-	struct list_head q_mplist;	/* mount's list of dquots */
-	struct list_head q_hashlist;	/* gloabl hash list of dquots */
-	xfs_dqhash_t	*q_hash;	/* the hashchain header */
+	struct list_head q_lru;		/* global free list of dquots */
 	struct xfs_mount*q_mount;	/* filesystem this relates to */
 	struct xfs_trans*q_transp;	/* trans this belongs to currently */
 	uint		 q_nrefs;	/* # active refs from inodes */
@@ -80,8 +67,6 @@ enum {
 	XFS_QLOCK_NESTED,
 };
 
-#define XFS_DQHOLD(dqp)		((dqp)->q_nrefs++)
-
 /*
  * Manage the q_flush completion queue embedded in the dquot.  This completion
  * queue synchronizes processes attempting to flush the in-core dquot back to
@@ -102,6 +87,47 @@ static inline void xfs_dqfunlock(xfs_dquot_t *dqp)
 	complete(&dqp->q_flush);
 }
 
+static inline int xfs_dqlock_nowait(struct xfs_dquot *dqp)
+{
+	return mutex_trylock(&dqp->q_qlock);
+}
+
+static inline void xfs_dqlock(struct xfs_dquot *dqp)
+{
+	mutex_lock(&dqp->q_qlock);
+}
+
+static inline void xfs_dqunlock(struct xfs_dquot *dqp)
+{
+	mutex_unlock(&dqp->q_qlock);
+}
+
+static inline int xfs_this_quota_on(struct xfs_mount *mp, int type)
+{
+	switch (type & XFS_DQ_ALLTYPES) {
+	case XFS_DQ_USER:
+		return XFS_IS_UQUOTA_ON(mp);
+	case XFS_DQ_GROUP:
+	case XFS_DQ_PROJ:
+		return XFS_IS_OQUOTA_ON(mp);
+	default:
+		return 0;
+	}
+}
+
+static inline xfs_dquot_t *xfs_inode_dquot(struct xfs_inode *ip, int type)
+{
+	switch (type & XFS_DQ_ALLTYPES) {
+	case XFS_DQ_USER:
+		return ip->i_udquot;
+	case XFS_DQ_GROUP:
+	case XFS_DQ_PROJ:
+		return ip->i_gdquot;
+	default:
+		return NULL;
+	}
+}
+
 #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)
@@ -112,16 +138,11 @@ static inline void xfs_dqfunlock(xfs_dquot_t *dqp)
 				 XFS_DQ_TO_QINF(dqp)->qi_uquotaip : \
 				 XFS_DQ_TO_QINF(dqp)->qi_gquotaip)
 
-#define XFS_IS_THIS_QUOTA_OFF(d) (! (XFS_QM_ISUDQ(d) ? \
-				     (XFS_IS_UQUOTA_ON((d)->q_mount)) : \
-				     (XFS_IS_OQUOTA_ON((d)->q_mount))))
-
+extern int		xfs_qm_dqread(struct xfs_mount *, xfs_dqid_t, uint,
+					uint, struct xfs_dquot	**);
 extern void		xfs_qm_dqdestroy(xfs_dquot_t *);
-extern int		xfs_qm_dqflush(xfs_dquot_t *, uint);
-extern int		xfs_qm_dqpurge(xfs_dquot_t *);
+extern int		xfs_qm_dqflush(struct xfs_dquot *, struct xfs_buf **);
 extern void		xfs_qm_dqunpin_wait(xfs_dquot_t *);
-extern int		xfs_qm_dqlock_nowait(xfs_dquot_t *);
-extern void		xfs_qm_dqflock_pushbuf_wait(xfs_dquot_t *dqp);
 extern void		xfs_qm_adjust_dqtimers(xfs_mount_t *,
 					xfs_disk_dquot_t *);
 extern void		xfs_qm_adjust_dqlimits(xfs_mount_t *,
@@ -129,9 +150,15 @@ extern void		xfs_qm_adjust_dqlimits(xfs_mount_t *,
 extern int		xfs_qm_dqget(xfs_mount_t *, xfs_inode_t *,
 					xfs_dqid_t, uint, uint, xfs_dquot_t **);
 extern void		xfs_qm_dqput(xfs_dquot_t *);
-extern void		xfs_dqlock(xfs_dquot_t *);
-extern void		xfs_dqlock2(xfs_dquot_t *, xfs_dquot_t *);
-extern void		xfs_dqunlock(xfs_dquot_t *);
-extern void		xfs_dqunlock_nonotify(xfs_dquot_t *);
+
+extern void		xfs_dqlock2(struct xfs_dquot *, struct xfs_dquot *);
+
+static inline struct xfs_dquot *xfs_qm_dqhold(struct xfs_dquot *dqp)
+{
+	xfs_dqlock(dqp);
+	dqp->q_nrefs++;
+	xfs_dqunlock(dqp);
+	return dqp;
+}
 
 #endif /* __XFS_DQUOT_H__ */
diff --git a/fs/xfs/xfs_dquot_item.c b/fs/xfs/xfs_dquot_item.c
index 0dee0b71029..57aa4b03720 100644
--- a/fs/xfs/xfs_dquot_item.c
+++ b/fs/xfs/xfs_dquot_item.c
@@ -17,9 +17,7 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -73,7 +71,6 @@ xfs_qm_dquot_logitem_format(
 	logvec->i_len  = sizeof(xfs_disk_dquot_t);
 	logvec->i_type = XLOG_REG_TYPE_DQUOT;
 
-	ASSERT(2 == lip->li_desc->lid_size);
 	qlip->qli_format.qlf_size = 2;
 
 }
@@ -109,38 +106,6 @@ xfs_qm_dquot_logitem_unpin(
 		wake_up(&dqp->q_pinwait);
 }
 
-/*
- * Given the logitem, this writes the corresponding dquot entry to disk
- * asynchronously. This is called with the dquot entry securely locked;
- * we simply get xfs_qm_dqflush() to do the work, and unlock the dquot
- * at the end.
- */
-STATIC void
-xfs_qm_dquot_logitem_push(
-	struct xfs_log_item	*lip)
-{
-	struct xfs_dquot	*dqp = DQUOT_ITEM(lip)->qli_dquot;
-	int			error;
-
-	ASSERT(XFS_DQ_IS_LOCKED(dqp));
-	ASSERT(!completion_done(&dqp->q_flush));
-
-	/*
-	 * Since we were able to lock the dquot's flush lock and
-	 * we found it on the AIL, the dquot must be dirty.  This
-	 * is because the dquot is removed from the AIL while still
-	 * holding the flush lock in xfs_dqflush_done().  Thus, if
-	 * we found it in the AIL and were able to obtain the flush
-	 * lock without sleeping, then there must not have been
-	 * anyone in the process of flushing the dquot.
-	 */
-	error = xfs_qm_dqflush(dqp, 0);
-	if (error)
-		xfs_warn(dqp->q_mount, "%s: push error %d on dqp %p",
-			__func__, error, dqp);
-	xfs_dqunlock(dqp);
-}
-
 STATIC xfs_lsn_t
 xfs_qm_dquot_logitem_committed(
 	struct xfs_log_item	*lip,
@@ -172,84 +137,57 @@ xfs_qm_dqunpin_wait(
 	wait_event(dqp->q_pinwait, (atomic_read(&dqp->q_pincount) == 0));
 }
 
-/*
- * This is called when IOP_TRYLOCK returns XFS_ITEM_PUSHBUF to indicate that
- * the dquot is locked by us, but the flush lock isn't. So, here we are
- * going to see if the relevant dquot buffer is incore, waiting on DELWRI.
- * If so, we want to push it out to help us take this item off the AIL as soon
- * as possible.
- *
- * We must not be holding the AIL lock at this point. Calling incore() to
- * search the buffer cache can be a time consuming thing, and AIL lock is a
- * spinlock.
- */
-STATIC bool
-xfs_qm_dquot_logitem_pushbuf(
-	struct xfs_log_item	*lip)
-{
-	struct xfs_dq_logitem	*qlip = DQUOT_ITEM(lip);
-	struct xfs_dquot	*dqp = qlip->qli_dquot;
-	struct xfs_buf		*bp;
-	bool			ret = true;
-
-	ASSERT(XFS_DQ_IS_LOCKED(dqp));
-
-	/*
-	 * If flushlock isn't locked anymore, chances are that the
-	 * inode flush completed and the inode was taken off the AIL.
-	 * So, just get out.
-	 */
-	if (completion_done(&dqp->q_flush) ||
-	    !(lip->li_flags & XFS_LI_IN_AIL)) {
-		xfs_dqunlock(dqp);
-		return true;
-	}
-
-	bp = xfs_incore(dqp->q_mount->m_ddev_targp, qlip->qli_format.qlf_blkno,
-			dqp->q_mount->m_quotainfo->qi_dqchunklen, XBF_TRYLOCK);
-	xfs_dqunlock(dqp);
-	if (!bp)
-		return true;
-	if (XFS_BUF_ISDELAYWRITE(bp))
-		xfs_buf_delwri_promote(bp);
-	if (xfs_buf_ispinned(bp))
-		ret = false;
-	xfs_buf_relse(bp);
-	return ret;
-}
-
-/*
- * This is called to attempt to lock the dquot associated with this
- * dquot log item.  Don't sleep on the dquot lock or the flush lock.
- * If the flush lock is already held, indicating that the dquot has
- * been or is in the process of being flushed, then see if we can
- * find the dquot's buffer in the buffer cache without sleeping.  If
- * we can and it is marked delayed write, then we want to send it out.
- * We delay doing so until the push routine, though, to avoid sleeping
- * in any device strategy routines.
- */
 STATIC uint
-xfs_qm_dquot_logitem_trylock(
-	struct xfs_log_item	*lip)
+xfs_qm_dquot_logitem_push(
+	struct xfs_log_item	*lip,
+	struct list_head	*buffer_list)
 {
 	struct xfs_dquot	*dqp = DQUOT_ITEM(lip)->qli_dquot;
+	struct xfs_buf		*bp = NULL;
+	uint			rval = XFS_ITEM_SUCCESS;
+	int			error;
 
 	if (atomic_read(&dqp->q_pincount) > 0)
 		return XFS_ITEM_PINNED;
 
-	if (!xfs_qm_dqlock_nowait(dqp))
+	if (!xfs_dqlock_nowait(dqp))
 		return XFS_ITEM_LOCKED;
 
+	/*
+	 * Re-check the pincount now that we stabilized the value by
+	 * taking the quota lock.
+	 */
+	if (atomic_read(&dqp->q_pincount) > 0) {
+		rval = XFS_ITEM_PINNED;
+		goto out_unlock;
+	}
+
+	/*
+	 * Someone else is already flushing the dquot.  Nothing we can do
+	 * here but wait for the flush to finish and remove the item from
+	 * the AIL.
+	 */
 	if (!xfs_dqflock_nowait(dqp)) {
-		/*
-		 * dquot has already been flushed to the backing buffer,
-		 * leave it locked, pushbuf routine will unlock it.
-		 */
-		return XFS_ITEM_PUSHBUF;
+		rval = XFS_ITEM_FLUSHING;
+		goto out_unlock;
 	}
 
-	ASSERT(lip->li_flags & XFS_LI_IN_AIL);
-	return XFS_ITEM_SUCCESS;
+	spin_unlock(&lip->li_ailp->xa_lock);
+
+	error = xfs_qm_dqflush(dqp, &bp);
+	if (error) {
+		xfs_warn(dqp->q_mount, "%s: push error %d on dqp %p",
+			__func__, error, dqp);
+	} else {
+		if (!xfs_buf_delwri_queue(bp, buffer_list))
+			rval = XFS_ITEM_FLUSHING;
+		xfs_buf_relse(bp);
+	}
+
+	spin_lock(&lip->li_ailp->xa_lock);
+out_unlock:
+	xfs_dqunlock(dqp);
+	return rval;
 }
 
 /*
@@ -300,11 +238,9 @@ static const struct xfs_item_ops xfs_dquot_item_ops = {
 	.iop_format	= xfs_qm_dquot_logitem_format,
 	.iop_pin	= xfs_qm_dquot_logitem_pin,
 	.iop_unpin	= xfs_qm_dquot_logitem_unpin,
-	.iop_trylock	= xfs_qm_dquot_logitem_trylock,
 	.iop_unlock	= xfs_qm_dquot_logitem_unlock,
 	.iop_committed	= xfs_qm_dquot_logitem_committed,
 	.iop_push	= xfs_qm_dquot_logitem_push,
-	.iop_pushbuf	= xfs_qm_dquot_logitem_pushbuf,
 	.iop_committing = xfs_qm_dquot_logitem_committing
 };
 
@@ -399,11 +335,13 @@ xfs_qm_qoff_logitem_unpin(
 }
 
 /*
- * Quotaoff items have no locking, so just return success.
+ * There isn't much you can do to push a quotaoff item.  It is simply
+ * stuck waiting for the log to be flushed to disk.
  */
 STATIC uint
-xfs_qm_qoff_logitem_trylock(
-	struct xfs_log_item	*lip)
+xfs_qm_qoff_logitem_push(
+	struct xfs_log_item	*lip,
+	struct list_head	*buffer_list)
 {
 	return XFS_ITEM_LOCKED;
 }
@@ -430,17 +368,6 @@ xfs_qm_qoff_logitem_committed(
 	return lsn;
 }
 
-/*
- * There isn't much you can do to push on an quotaoff item.  It is simply
- * stuck waiting for the log to be flushed to disk.
- */
-STATIC void
-xfs_qm_qoff_logitem_push(
-	struct xfs_log_item	*lip)
-{
-}
-
-
 STATIC xfs_lsn_t
 xfs_qm_qoffend_logitem_committed(
 	struct xfs_log_item	*lip,
@@ -455,7 +382,7 @@ xfs_qm_qoffend_logitem_committed(
 	 * xfs_trans_ail_delete() drops the AIL lock.
 	 */
 	spin_lock(&ailp->xa_lock);
-	xfs_trans_ail_delete(ailp, (xfs_log_item_t *)qfs);
+	xfs_trans_ail_delete(ailp, &qfs->qql_item, SHUTDOWN_LOG_IO_ERROR);
 
 	kmem_free(qfs);
 	kmem_free(qfe);
@@ -488,7 +415,6 @@ static const struct xfs_item_ops xfs_qm_qoffend_logitem_ops = {
 	.iop_format	= xfs_qm_qoff_logitem_format,
 	.iop_pin	= xfs_qm_qoff_logitem_pin,
 	.iop_unpin	= xfs_qm_qoff_logitem_unpin,
-	.iop_trylock	= xfs_qm_qoff_logitem_trylock,
 	.iop_unlock	= xfs_qm_qoff_logitem_unlock,
 	.iop_committed	= xfs_qm_qoffend_logitem_committed,
 	.iop_push	= xfs_qm_qoff_logitem_push,
@@ -503,7 +429,6 @@ static const struct xfs_item_ops xfs_qm_qoff_logitem_ops = {
 	.iop_format	= xfs_qm_qoff_logitem_format,
 	.iop_pin	= xfs_qm_qoff_logitem_pin,
 	.iop_unpin	= xfs_qm_qoff_logitem_unpin,
-	.iop_trylock	= xfs_qm_qoff_logitem_trylock,
 	.iop_unlock	= xfs_qm_qoff_logitem_unlock,
 	.iop_committed	= xfs_qm_qoff_logitem_committed,
 	.iop_push	= xfs_qm_qoff_logitem_push,
diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c
index 39f06336b99..610456054dc 100644
--- a/fs/xfs/xfs_error.c
+++ b/fs/xfs/xfs_error.c
@@ -19,7 +19,6 @@
 #include "xfs_fs.h"
 #include "xfs_types.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
diff --git a/fs/xfs/xfs_export.c b/fs/xfs/xfs_export.c
index 558910f5e3c..42679223a0f 100644
--- a/fs/xfs/xfs_export.c
+++ b/fs/xfs/xfs_export.c
@@ -17,7 +17,6 @@
  */
 #include "xfs.h"
 #include "xfs_types.h"
-#include "xfs_inum.h"
 #include "xfs_log.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
@@ -53,19 +52,18 @@ static int xfs_fileid_length(int fileid_type)
 
 STATIC int
 xfs_fs_encode_fh(
-	struct dentry		*dentry,
-	__u32			*fh,
-	int			*max_len,
-	int			connectable)
+	struct inode	*inode,
+	__u32		*fh,
+	int		*max_len,
+	struct inode	*parent)
 {
 	struct fid		*fid = (struct fid *)fh;
 	struct xfs_fid64	*fid64 = (struct xfs_fid64 *)fh;
-	struct inode		*inode = dentry->d_inode;
 	int			fileid_type;
 	int			len;
 
 	/* Directories don't need their parent encoded, they have ".." */
-	if (S_ISDIR(inode->i_mode) || !connectable)
+	if (!parent)
 		fileid_type = FILEID_INO32_GEN;
 	else
 		fileid_type = FILEID_INO32_GEN_PARENT;
@@ -97,20 +95,16 @@ xfs_fs_encode_fh(
 
 	switch (fileid_type) {
 	case FILEID_INO32_GEN_PARENT:
-		spin_lock(&dentry->d_lock);
-		fid->i32.parent_ino = XFS_I(dentry->d_parent->d_inode)->i_ino;
-		fid->i32.parent_gen = dentry->d_parent->d_inode->i_generation;
-		spin_unlock(&dentry->d_lock);
+		fid->i32.parent_ino = XFS_I(parent)->i_ino;
+		fid->i32.parent_gen = parent->i_generation;
 		/*FALLTHRU*/
 	case FILEID_INO32_GEN:
 		fid->i32.ino = XFS_I(inode)->i_ino;
 		fid->i32.gen = inode->i_generation;
 		break;
 	case FILEID_INO32_GEN_PARENT | XFS_FILEID_TYPE_64FLAG:
-		spin_lock(&dentry->d_lock);
-		fid64->parent_ino = XFS_I(dentry->d_parent->d_inode)->i_ino;
-		fid64->parent_gen = dentry->d_parent->d_inode->i_generation;
-		spin_unlock(&dentry->d_lock);
+		fid64->parent_ino = XFS_I(parent)->i_ino;
+		fid64->parent_gen = parent->i_generation;
 		/*FALLTHRU*/
 	case FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG:
 		fid64->ino = XFS_I(inode)->i_ino;
diff --git a/fs/xfs/xfs_extent_busy.c b/fs/xfs/xfs_extent_busy.c
new file mode 100644
index 00000000000..85e9f87a1a7
--- /dev/null
+++ b/fs/xfs/xfs_extent_busy.c
@@ -0,0 +1,603 @@
+/*
+ * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc.
+ * Copyright (c) 2010 David Chinner.
+ * Copyright (c) 2011 Christoph Hellwig.
+ * 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_fs.h"
+#include "xfs_types.h"
+#include "xfs_log.h"
+#include "xfs_trans.h"
+#include "xfs_sb.h"
+#include "xfs_ag.h"
+#include "xfs_mount.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_alloc.h"
+#include "xfs_inode.h"
+#include "xfs_extent_busy.h"
+#include "xfs_trace.h"
+
+void
+xfs_extent_busy_insert(
+	struct xfs_trans	*tp,
+	xfs_agnumber_t		agno,
+	xfs_agblock_t		bno,
+	xfs_extlen_t		len,
+	unsigned int		flags)
+{
+	struct xfs_extent_busy	*new;
+	struct xfs_extent_busy	*busyp;
+	struct xfs_perag	*pag;
+	struct rb_node		**rbp;
+	struct rb_node		*parent = NULL;
+
+	new = kmem_zalloc(sizeof(struct xfs_extent_busy), KM_MAYFAIL);
+	if (!new) {
+		/*
+		 * No Memory!  Since it is now not possible to track the free
+		 * block, make this a synchronous transaction to insure that
+		 * the block is not reused before this transaction commits.
+		 */
+		trace_xfs_extent_busy_enomem(tp->t_mountp, agno, bno, len);
+		xfs_trans_set_sync(tp);
+		return;
+	}
+
+	new->agno = agno;
+	new->bno = bno;
+	new->length = len;
+	INIT_LIST_HEAD(&new->list);
+	new->flags = flags;
+
+	/* trace before insert to be able to see failed inserts */
+	trace_xfs_extent_busy(tp->t_mountp, agno, bno, len);
+
+	pag = xfs_perag_get(tp->t_mountp, new->agno);
+	spin_lock(&pag->pagb_lock);
+	rbp = &pag->pagb_tree.rb_node;
+	while (*rbp) {
+		parent = *rbp;
+		busyp = rb_entry(parent, struct xfs_extent_busy, rb_node);
+
+		if (new->bno < busyp->bno) {
+			rbp = &(*rbp)->rb_left;
+			ASSERT(new->bno + new->length <= busyp->bno);
+		} else if (new->bno > busyp->bno) {
+			rbp = &(*rbp)->rb_right;
+			ASSERT(bno >= busyp->bno + busyp->length);
+		} else {
+			ASSERT(0);
+		}
+	}
+
+	rb_link_node(&new->rb_node, parent, rbp);
+	rb_insert_color(&new->rb_node, &pag->pagb_tree);
+
+	list_add(&new->list, &tp->t_busy);
+	spin_unlock(&pag->pagb_lock);
+	xfs_perag_put(pag);
+}
+
+/*
+ * Search for a busy extent within the range of the extent we are about to
+ * allocate.  You need to be holding the busy extent tree lock when calling
+ * xfs_extent_busy_search(). This function returns 0 for no overlapping busy
+ * extent, -1 for an overlapping but not exact busy extent, and 1 for an exact
+ * match. This is done so that a non-zero return indicates an overlap that
+ * will require a synchronous transaction, but it can still be
+ * used to distinguish between a partial or exact match.
+ */
+int
+xfs_extent_busy_search(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		agno,
+	xfs_agblock_t		bno,
+	xfs_extlen_t		len)
+{
+	struct xfs_perag	*pag;
+	struct rb_node		*rbp;
+	struct xfs_extent_busy	*busyp;
+	int			match = 0;
+
+	pag = xfs_perag_get(mp, agno);
+	spin_lock(&pag->pagb_lock);
+
+	rbp = pag->pagb_tree.rb_node;
+
+	/* find closest start bno overlap */
+	while (rbp) {
+		busyp = rb_entry(rbp, struct xfs_extent_busy, rb_node);
+		if (bno < busyp->bno) {
+			/* may overlap, but exact start block is lower */
+			if (bno + len > busyp->bno)
+				match = -1;
+			rbp = rbp->rb_left;
+		} else if (bno > busyp->bno) {
+			/* may overlap, but exact start block is higher */
+			if (bno < busyp->bno + busyp->length)
+				match = -1;
+			rbp = rbp->rb_right;
+		} else {
+			/* bno matches busyp, length determines exact match */
+			match = (busyp->length == len) ? 1 : -1;
+			break;
+		}
+	}
+	spin_unlock(&pag->pagb_lock);
+	xfs_perag_put(pag);
+	return match;
+}
+
+/*
+ * The found free extent [fbno, fend] overlaps part or all of the given busy
+ * extent.  If the overlap covers the beginning, the end, or all of the busy
+ * extent, the overlapping portion can be made unbusy and used for the
+ * allocation.  We can't split a busy extent because we can't modify a
+ * transaction/CIL context busy list, but we can update an entries block
+ * number or length.
+ *
+ * Returns true if the extent can safely be reused, or false if the search
+ * needs to be restarted.
+ */
+STATIC bool
+xfs_extent_busy_update_extent(
+	struct xfs_mount	*mp,
+	struct xfs_perag	*pag,
+	struct xfs_extent_busy	*busyp,
+	xfs_agblock_t		fbno,
+	xfs_extlen_t		flen,
+	bool			userdata)
+{
+	xfs_agblock_t		fend = fbno + flen;
+	xfs_agblock_t		bbno = busyp->bno;
+	xfs_agblock_t		bend = bbno + busyp->length;
+
+	/*
+	 * This extent is currently being discarded.  Give the thread
+	 * performing the discard a chance to mark the extent unbusy
+	 * and retry.
+	 */
+	if (busyp->flags & XFS_EXTENT_BUSY_DISCARDED) {
+		spin_unlock(&pag->pagb_lock);
+		delay(1);
+		spin_lock(&pag->pagb_lock);
+		return false;
+	}
+
+	/*
+	 * If there is a busy extent overlapping a user allocation, we have
+	 * no choice but to force the log and retry the search.
+	 *
+	 * Fortunately this does not happen during normal operation, but
+	 * only if the filesystem is very low on space and has to dip into
+	 * the AGFL for normal allocations.
+	 */
+	if (userdata)
+		goto out_force_log;
+
+	if (bbno < fbno && bend > fend) {
+		/*
+		 * Case 1:
+		 *    bbno           bend
+		 *    +BBBBBBBBBBBBBBBBB+
+		 *        +---------+
+		 *        fbno   fend
+		 */
+
+		/*
+		 * We would have to split the busy extent to be able to track
+		 * it correct, which we cannot do because we would have to
+		 * modify the list of busy extents attached to the transaction
+		 * or CIL context, which is immutable.
+		 *
+		 * Force out the log to clear the busy extent and retry the
+		 * search.
+		 */
+		goto out_force_log;
+	} else if (bbno >= fbno && bend <= fend) {
+		/*
+		 * Case 2:
+		 *    bbno           bend
+		 *    +BBBBBBBBBBBBBBBBB+
+		 *    +-----------------+
+		 *    fbno           fend
+		 *
+		 * Case 3:
+		 *    bbno           bend
+		 *    +BBBBBBBBBBBBBBBBB+
+		 *    +--------------------------+
+		 *    fbno                    fend
+		 *
+		 * Case 4:
+		 *             bbno           bend
+		 *             +BBBBBBBBBBBBBBBBB+
+		 *    +--------------------------+
+		 *    fbno                    fend
+		 *
+		 * Case 5:
+		 *             bbno           bend
+		 *             +BBBBBBBBBBBBBBBBB+
+		 *    +-----------------------------------+
+		 *    fbno                             fend
+		 *
+		 */
+
+		/*
+		 * The busy extent is fully covered by the extent we are
+		 * allocating, and can simply be removed from the rbtree.
+		 * However we cannot remove it from the immutable list
+		 * tracking busy extents in the transaction or CIL context,
+		 * so set the length to zero to mark it invalid.
+		 *
+		 * We also need to restart the busy extent search from the
+		 * tree root, because erasing the node can rearrange the
+		 * tree topology.
+		 */
+		rb_erase(&busyp->rb_node, &pag->pagb_tree);
+		busyp->length = 0;
+		return false;
+	} else if (fend < bend) {
+		/*
+		 * Case 6:
+		 *              bbno           bend
+		 *             +BBBBBBBBBBBBBBBBB+
+		 *             +---------+
+		 *             fbno   fend
+		 *
+		 * Case 7:
+		 *             bbno           bend
+		 *             +BBBBBBBBBBBBBBBBB+
+		 *    +------------------+
+		 *    fbno            fend
+		 *
+		 */
+		busyp->bno = fend;
+	} else if (bbno < fbno) {
+		/*
+		 * Case 8:
+		 *    bbno           bend
+		 *    +BBBBBBBBBBBBBBBBB+
+		 *        +-------------+
+		 *        fbno       fend
+		 *
+		 * Case 9:
+		 *    bbno           bend
+		 *    +BBBBBBBBBBBBBBBBB+
+		 *        +----------------------+
+		 *        fbno                fend
+		 */
+		busyp->length = fbno - busyp->bno;
+	} else {
+		ASSERT(0);
+	}
+
+	trace_xfs_extent_busy_reuse(mp, pag->pag_agno, fbno, flen);
+	return true;
+
+out_force_log:
+	spin_unlock(&pag->pagb_lock);
+	xfs_log_force(mp, XFS_LOG_SYNC);
+	trace_xfs_extent_busy_force(mp, pag->pag_agno, fbno, flen);
+	spin_lock(&pag->pagb_lock);
+	return false;
+}
+
+
+/*
+ * For a given extent [fbno, flen], make sure we can reuse it safely.
+ */
+void
+xfs_extent_busy_reuse(
+	struct xfs_mount	*mp,
+	xfs_agnumber_t		agno,
+	xfs_agblock_t		fbno,
+	xfs_extlen_t		flen,
+	bool			userdata)
+{
+	struct xfs_perag	*pag;
+	struct rb_node		*rbp;
+
+	ASSERT(flen > 0);
+
+	pag = xfs_perag_get(mp, agno);
+	spin_lock(&pag->pagb_lock);
+restart:
+	rbp = pag->pagb_tree.rb_node;
+	while (rbp) {
+		struct xfs_extent_busy *busyp =
+			rb_entry(rbp, struct xfs_extent_busy, rb_node);
+		xfs_agblock_t	bbno = busyp->bno;
+		xfs_agblock_t	bend = bbno + busyp->length;
+
+		if (fbno + flen <= bbno) {
+			rbp = rbp->rb_left;
+			continue;
+		} else if (fbno >= bend) {
+			rbp = rbp->rb_right;
+			continue;
+		}
+
+		if (!xfs_extent_busy_update_extent(mp, pag, busyp, fbno, flen,
+						  userdata))
+			goto restart;
+	}
+	spin_unlock(&pag->pagb_lock);
+	xfs_perag_put(pag);
+}
+
+/*
+ * For a given extent [fbno, flen], search the busy extent list to find a
+ * subset of the extent that is not busy.  If *rlen is smaller than
+ * args->minlen no suitable extent could be found, and the higher level
+ * code needs to force out the log and retry the allocation.
+ */
+void
+xfs_extent_busy_trim(
+	struct xfs_alloc_arg	*args,
+	xfs_agblock_t		bno,
+	xfs_extlen_t		len,
+	xfs_agblock_t		*rbno,
+	xfs_extlen_t		*rlen)
+{
+	xfs_agblock_t		fbno;
+	xfs_extlen_t		flen;
+	struct rb_node		*rbp;
+
+	ASSERT(len > 0);
+
+	spin_lock(&args->pag->pagb_lock);
+restart:
+	fbno = bno;
+	flen = len;
+	rbp = args->pag->pagb_tree.rb_node;
+	while (rbp && flen >= args->minlen) {
+		struct xfs_extent_busy *busyp =
+			rb_entry(rbp, struct xfs_extent_busy, rb_node);
+		xfs_agblock_t	fend = fbno + flen;
+		xfs_agblock_t	bbno = busyp->bno;
+		xfs_agblock_t	bend = bbno + busyp->length;
+
+		if (fend <= bbno) {
+			rbp = rbp->rb_left;
+			continue;
+		} else if (fbno >= bend) {
+			rbp = rbp->rb_right;
+			continue;
+		}
+
+		/*
+		 * If this is a metadata allocation, try to reuse the busy
+		 * extent instead of trimming the allocation.
+		 */
+		if (!args->userdata &&
+		    !(busyp->flags & XFS_EXTENT_BUSY_DISCARDED)) {
+			if (!xfs_extent_busy_update_extent(args->mp, args->pag,
+							  busyp, fbno, flen,
+							  false))
+				goto restart;
+			continue;
+		}
+
+		if (bbno <= fbno) {
+			/* start overlap */
+
+			/*
+			 * Case 1:
+			 *    bbno           bend
+			 *    +BBBBBBBBBBBBBBBBB+
+			 *        +---------+
+			 *        fbno   fend
+			 *
+			 * Case 2:
+			 *    bbno           bend
+			 *    +BBBBBBBBBBBBBBBBB+
+			 *    +-------------+
+			 *    fbno       fend
+			 *
+			 * Case 3:
+			 *    bbno           bend
+			 *    +BBBBBBBBBBBBBBBBB+
+			 *        +-------------+
+			 *        fbno       fend
+			 *
+			 * Case 4:
+			 *    bbno           bend
+			 *    +BBBBBBBBBBBBBBBBB+
+			 *    +-----------------+
+			 *    fbno           fend
+			 *
+			 * No unbusy region in extent, return failure.
+			 */
+			if (fend <= bend)
+				goto fail;
+
+			/*
+			 * Case 5:
+			 *    bbno           bend
+			 *    +BBBBBBBBBBBBBBBBB+
+			 *        +----------------------+
+			 *        fbno                fend
+			 *
+			 * Case 6:
+			 *    bbno           bend
+			 *    +BBBBBBBBBBBBBBBBB+
+			 *    +--------------------------+
+			 *    fbno                    fend
+			 *
+			 * Needs to be trimmed to:
+			 *                       +-------+
+			 *                       fbno fend
+			 */
+			fbno = bend;
+		} else if (bend >= fend) {
+			/* end overlap */
+
+			/*
+			 * Case 7:
+			 *             bbno           bend
+			 *             +BBBBBBBBBBBBBBBBB+
+			 *    +------------------+
+			 *    fbno            fend
+			 *
+			 * Case 8:
+			 *             bbno           bend
+			 *             +BBBBBBBBBBBBBBBBB+
+			 *    +--------------------------+
+			 *    fbno                    fend
+			 *
+			 * Needs to be trimmed to:
+			 *    +-------+
+			 *    fbno fend
+			 */
+			fend = bbno;
+		} else {
+			/* middle overlap */
+
+			/*
+			 * Case 9:
+			 *             bbno           bend
+			 *             +BBBBBBBBBBBBBBBBB+
+			 *    +-----------------------------------+
+			 *    fbno                             fend
+			 *
+			 * Can be trimmed to:
+			 *    +-------+        OR         +-------+
+			 *    fbno fend                   fbno fend
+			 *
+			 * Backward allocation leads to significant
+			 * fragmentation of directories, which degrades
+			 * directory performance, therefore we always want to
+			 * choose the option that produces forward allocation
+			 * patterns.
+			 * Preferring the lower bno extent will make the next
+			 * request use "fend" as the start of the next
+			 * allocation;  if the segment is no longer busy at
+			 * that point, we'll get a contiguous allocation, but
+			 * even if it is still busy, we will get a forward
+			 * allocation.
+			 * We try to avoid choosing the segment at "bend",
+			 * because that can lead to the next allocation
+			 * taking the segment at "fbno", which would be a
+			 * backward allocation.  We only use the segment at
+			 * "fbno" if it is much larger than the current
+			 * requested size, because in that case there's a
+			 * good chance subsequent allocations will be
+			 * contiguous.
+			 */
+			if (bbno - fbno >= args->maxlen) {
+				/* left candidate fits perfect */
+				fend = bbno;
+			} else if (fend - bend >= args->maxlen * 4) {
+				/* right candidate has enough free space */
+				fbno = bend;
+			} else if (bbno - fbno >= args->minlen) {
+				/* left candidate fits minimum requirement */
+				fend = bbno;
+			} else {
+				goto fail;
+			}
+		}
+
+		flen = fend - fbno;
+	}
+	spin_unlock(&args->pag->pagb_lock);
+
+	if (fbno != bno || flen != len) {
+		trace_xfs_extent_busy_trim(args->mp, args->agno, bno, len,
+					  fbno, flen);
+	}
+	*rbno = fbno;
+	*rlen = flen;
+	return;
+fail:
+	/*
+	 * Return a zero extent length as failure indications.  All callers
+	 * re-check if the trimmed extent satisfies the minlen requirement.
+	 */
+	spin_unlock(&args->pag->pagb_lock);
+	trace_xfs_extent_busy_trim(args->mp, args->agno, bno, len, fbno, 0);
+	*rbno = fbno;
+	*rlen = 0;
+}
+
+STATIC void
+xfs_extent_busy_clear_one(
+	struct xfs_mount	*mp,
+	struct xfs_perag	*pag,
+	struct xfs_extent_busy	*busyp)
+{
+	if (busyp->length) {
+		trace_xfs_extent_busy_clear(mp, busyp->agno, busyp->bno,
+						busyp->length);
+		rb_erase(&busyp->rb_node, &pag->pagb_tree);
+	}
+
+	list_del_init(&busyp->list);
+	kmem_free(busyp);
+}
+
+/*
+ * Remove all extents on the passed in list from the busy extents tree.
+ * If do_discard is set skip extents that need to be discarded, and mark
+ * these as undergoing a discard operation instead.
+ */
+void
+xfs_extent_busy_clear(
+	struct xfs_mount	*mp,
+	struct list_head	*list,
+	bool			do_discard)
+{
+	struct xfs_extent_busy	*busyp, *n;
+	struct xfs_perag	*pag = NULL;
+	xfs_agnumber_t		agno = NULLAGNUMBER;
+
+	list_for_each_entry_safe(busyp, n, list, list) {
+		if (busyp->agno != agno) {
+			if (pag) {
+				spin_unlock(&pag->pagb_lock);
+				xfs_perag_put(pag);
+			}
+			pag = xfs_perag_get(mp, busyp->agno);
+			spin_lock(&pag->pagb_lock);
+			agno = busyp->agno;
+		}
+
+		if (do_discard && busyp->length &&
+		    !(busyp->flags & XFS_EXTENT_BUSY_SKIP_DISCARD))
+			busyp->flags = XFS_EXTENT_BUSY_DISCARDED;
+		else
+			xfs_extent_busy_clear_one(mp, pag, busyp);
+	}
+
+	if (pag) {
+		spin_unlock(&pag->pagb_lock);
+		xfs_perag_put(pag);
+	}
+}
+
+/*
+ * Callback for list_sort to sort busy extents by the AG they reside in.
+ */
+int
+xfs_extent_busy_ag_cmp(
+	void			*priv,
+	struct list_head	*a,
+	struct list_head	*b)
+{
+	return container_of(a, struct xfs_extent_busy, list)->agno -
+		container_of(b, struct xfs_extent_busy, list)->agno;
+}
diff --git a/fs/xfs/xfs_extent_busy.h b/fs/xfs/xfs_extent_busy.h
new file mode 100644
index 00000000000..985412d65ba
--- /dev/null
+++ b/fs/xfs/xfs_extent_busy.h
@@ -0,0 +1,69 @@
+/*
+ * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc.
+ * Copyright (c) 2010 David Chinner.
+ * Copyright (c) 2011 Christoph Hellwig.
+ * 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_EXTENT_BUSY_H__
+#define	__XFS_EXTENT_BUSY_H__
+
+/*
+ * Busy block/extent entry.  Indexed by a rbtree in perag to mark blocks that
+ * have been freed but whose transactions aren't committed to disk yet.
+ *
+ * Note that we use the transaction ID to record the transaction, not the
+ * transaction structure itself. See xfs_extent_busy_insert() for details.
+ */
+struct xfs_extent_busy {
+	struct rb_node	rb_node;	/* ag by-bno indexed search tree */
+	struct list_head list;		/* transaction busy extent list */
+	xfs_agnumber_t	agno;
+	xfs_agblock_t	bno;
+	xfs_extlen_t	length;
+	unsigned int	flags;
+#define XFS_EXTENT_BUSY_DISCARDED	0x01	/* undergoing a discard op. */
+#define XFS_EXTENT_BUSY_SKIP_DISCARD	0x02	/* do not discard */
+};
+
+void
+xfs_extent_busy_insert(struct xfs_trans *tp, xfs_agnumber_t agno,
+	xfs_agblock_t bno, xfs_extlen_t len, unsigned int flags);
+
+void
+xfs_extent_busy_clear(struct xfs_mount *mp, struct list_head *list,
+	bool do_discard);
+
+int
+xfs_extent_busy_search(struct xfs_mount *mp, xfs_agnumber_t agno,
+	xfs_agblock_t bno, xfs_extlen_t len);
+
+void
+xfs_extent_busy_reuse(struct xfs_mount *mp, xfs_agnumber_t agno,
+	xfs_agblock_t fbno, xfs_extlen_t flen, bool userdata);
+
+void
+xfs_extent_busy_trim(struct xfs_alloc_arg *args, xfs_agblock_t bno,
+	xfs_extlen_t len, xfs_agblock_t *rbno, xfs_extlen_t *rlen);
+
+int
+xfs_extent_busy_ag_cmp(void *priv, struct list_head *a, struct list_head *b);
+
+static inline void xfs_extent_busy_sort(struct list_head *list)
+{
+	list_sort(NULL, list, xfs_extent_busy_ag_cmp);
+}
+
+#endif /* __XFS_EXTENT_BUSY_H__ */
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index 35c2aff38b2..feb36d7551a 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -19,7 +19,6 @@
 #include "xfs_fs.h"
 #include "xfs_types.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_sb.h"
@@ -64,7 +63,8 @@ __xfs_efi_release(
 	if (!test_and_clear_bit(XFS_EFI_COMMITTED, &efip->efi_flags)) {
 		spin_lock(&ailp->xa_lock);
 		/* xfs_trans_ail_delete() drops the AIL lock. */
-		xfs_trans_ail_delete(ailp, &efip->efi_item);
+		xfs_trans_ail_delete(ailp, &efip->efi_item,
+				     SHUTDOWN_LOG_IO_ERROR);
 		xfs_efi_item_free(efip);
 	}
 }
@@ -147,22 +147,20 @@ xfs_efi_item_unpin(
 }
 
 /*
- * Efi items have no locking or pushing.  However, since EFIs are
- * pulled from the AIL when their corresponding EFDs are committed
- * to disk, their situation is very similar to being pinned.  Return
- * XFS_ITEM_PINNED so that the caller will eventually flush the log.
- * This should help in getting the EFI out of the AIL.
+ * Efi items have no locking or pushing.  However, since EFIs are pulled from
+ * the AIL when their corresponding EFDs are committed to disk, their situation
+ * is very similar to being pinned.  Return XFS_ITEM_PINNED so that the caller
+ * will eventually flush the log.  This should help in getting the EFI out of
+ * the AIL.
  */
 STATIC uint
-xfs_efi_item_trylock(
-	struct xfs_log_item	*lip)
+xfs_efi_item_push(
+	struct xfs_log_item	*lip,
+	struct list_head	*buffer_list)
 {
 	return XFS_ITEM_PINNED;
 }
 
-/*
- * Efi items have no locking, so just return.
- */
 STATIC void
 xfs_efi_item_unlock(
 	struct xfs_log_item	*lip)
@@ -190,17 +188,6 @@ xfs_efi_item_committed(
 }
 
 /*
- * There isn't much you can do to push on an efi item.  It is simply
- * stuck waiting for all of its corresponding efd items to be
- * committed to disk.
- */
-STATIC void
-xfs_efi_item_push(
-	struct xfs_log_item	*lip)
-{
-}
-
-/*
  * The EFI dependency tracking op doesn't do squat.  It can't because
  * it doesn't know where the free extent is coming from.  The dependency
  * tracking has to be handled by the "enclosing" metadata object.  For
@@ -222,7 +209,6 @@ static const struct xfs_item_ops xfs_efi_item_ops = {
 	.iop_format	= xfs_efi_item_format,
 	.iop_pin	= xfs_efi_item_pin,
 	.iop_unpin	= xfs_efi_item_unpin,
-	.iop_trylock	= xfs_efi_item_trylock,
 	.iop_unlock	= xfs_efi_item_unlock,
 	.iop_committed	= xfs_efi_item_committed,
 	.iop_push	= xfs_efi_item_push,
@@ -404,19 +390,17 @@ xfs_efd_item_unpin(
 }
 
 /*
- * Efd items have no locking, so just return success.
+ * There isn't much you can do to push on an efd item.  It is simply stuck
+ * waiting for the log to be flushed to disk.
  */
 STATIC uint
-xfs_efd_item_trylock(
-	struct xfs_log_item	*lip)
+xfs_efd_item_push(
+	struct xfs_log_item	*lip,
+	struct list_head	*buffer_list)
 {
-	return XFS_ITEM_LOCKED;
+	return XFS_ITEM_PINNED;
 }
 
-/*
- * Efd items have no locking or pushing, so return failure
- * so that the caller doesn't bother with us.
- */
 STATIC void
 xfs_efd_item_unlock(
 	struct xfs_log_item	*lip)
@@ -451,16 +435,6 @@ xfs_efd_item_committed(
 }
 
 /*
- * There isn't much you can do to push on an efd item.  It is simply
- * stuck waiting for the log to be flushed to disk.
- */
-STATIC void
-xfs_efd_item_push(
-	struct xfs_log_item	*lip)
-{
-}
-
-/*
  * The EFD dependency tracking op doesn't do squat.  It can't because
  * it doesn't know where the free extent is coming from.  The dependency
  * tracking has to be handled by the "enclosing" metadata object.  For
@@ -482,7 +456,6 @@ static const struct xfs_item_ops xfs_efd_item_ops = {
 	.iop_format	= xfs_efd_item_format,
 	.iop_pin	= xfs_efd_item_pin,
 	.iop_unpin	= xfs_efd_item_unpin,
-	.iop_trylock	= xfs_efd_item_trylock,
 	.iop_unlock	= xfs_efd_item_unlock,
 	.iop_committed	= xfs_efd_item_committed,
 	.iop_push	= xfs_efd_item_push,
diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index 753ed9b5c70..9f7ec15a652 100644
--- a/fs/xfs/xfs_file.c
+++ b/fs/xfs/xfs_file.c
@@ -17,9 +17,7 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_trans.h"
@@ -163,7 +161,6 @@ xfs_file_fsync(
 	struct inode		*inode = file->f_mapping->host;
 	struct xfs_inode	*ip = XFS_I(inode);
 	struct xfs_mount	*mp = ip->i_mount;
-	struct xfs_trans	*tp;
 	int			error = 0;
 	int			log_flushed = 0;
 	xfs_lsn_t		lsn = 0;
@@ -194,75 +191,18 @@ xfs_file_fsync(
 	}
 
 	/*
-	 * We always need to make sure that the required inode state is safe on
-	 * disk.  The inode might be clean but we still might need to force the
-	 * log because of committed transactions that haven't hit the disk yet.
-	 * Likewise, there could be unflushed non-transactional changes to the
-	 * inode core that have to go to disk and this requires us to issue
-	 * a synchronous transaction to capture these changes correctly.
-	 *
-	 * This code relies on the assumption that if the i_update_core field
-	 * of the inode is clear and the inode is unpinned then it is clean
-	 * and no action is required.
+	 * All metadata updates are logged, which means that we just have
+	 * to flush the log up to the latest LSN that touched the inode.
 	 */
 	xfs_ilock(ip, XFS_ILOCK_SHARED);
-
-	/*
-	 * First check if the VFS inode is marked dirty.  All the dirtying
-	 * of non-transactional updates no goes through mark_inode_dirty*,
-	 * which allows us to distinguish beteeen pure timestamp updates
-	 * and i_size updates which need to be caught for fdatasync.
-	 * After that also theck for the dirty state in the XFS inode, which
-	 * might gets cleared when the inode gets written out via the AIL
-	 * or xfs_iflush_cluster.
-	 */
-	if (((inode->i_state & I_DIRTY_DATASYNC) ||
-	    ((inode->i_state & I_DIRTY_SYNC) && !datasync)) &&
-	    ip->i_update_core) {
-		/*
-		 * Kick off a transaction to log the inode core to get the
-		 * updates.  The sync transaction will also force the log.
-		 */
-		xfs_iunlock(ip, XFS_ILOCK_SHARED);
-		tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS);
-		error = xfs_trans_reserve(tp, 0,
-				XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0);
-		if (error) {
-			xfs_trans_cancel(tp, 0);
-			return -error;
-		}
-		xfs_ilock(ip, XFS_ILOCK_EXCL);
-
-		/*
-		 * Note - it's possible that we might have pushed ourselves out
-		 * of the way during trans_reserve which would flush the inode.
-		 * But there's no guarantee that the inode buffer has actually
-		 * gone out yet (it's delwri).	Plus the buffer could be pinned
-		 * anyway if it's part of an inode in another recent
-		 * transaction.	 So we play it safe and fire off the
-		 * transaction anyway.
-		 */
-		xfs_trans_ijoin(tp, ip, 0);
-		xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-		error = xfs_trans_commit(tp, 0);
-
-		lsn = ip->i_itemp->ili_last_lsn;
-		xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	} else {
-		/*
-		 * Timestamps/size haven't changed since last inode flush or
-		 * inode transaction commit.  That means either nothing got
-		 * written or a transaction committed which caught the updates.
-		 * If the latter happened and the transaction hasn't hit the
-		 * disk yet, the inode will be still be pinned.  If it is,
-		 * force the log.
-		 */
-		if (xfs_ipincount(ip))
+	if (xfs_ipincount(ip)) {
+		if (!datasync ||
+		    (ip->i_itemp->ili_fields & ~XFS_ILOG_TIMESTAMP))
 			lsn = ip->i_itemp->ili_last_lsn;
-		xfs_iunlock(ip, XFS_ILOCK_SHARED);
 	}
+	xfs_iunlock(ip, XFS_ILOCK_SHARED);
 
-	if (!error && lsn)
+	if (lsn)
 		error = _xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, &log_flushed);
 
 	/*
@@ -327,7 +267,7 @@ xfs_file_aio_read(
 				mp->m_rtdev_targp : mp->m_ddev_targp;
 		if ((iocb->ki_pos & target->bt_smask) ||
 		    (size & target->bt_smask)) {
-			if (iocb->ki_pos == ip->i_size)
+			if (iocb->ki_pos == i_size_read(inode))
 				return 0;
 			return -XFS_ERROR(EINVAL);
 		}
@@ -412,51 +352,6 @@ xfs_file_splice_read(
 	return ret;
 }
 
-STATIC void
-xfs_aio_write_isize_update(
-	struct inode	*inode,
-	loff_t		*ppos,
-	ssize_t		bytes_written)
-{
-	struct xfs_inode	*ip = XFS_I(inode);
-	xfs_fsize_t		isize = i_size_read(inode);
-
-	if (bytes_written > 0)
-		XFS_STATS_ADD(xs_write_bytes, bytes_written);
-
-	if (unlikely(bytes_written < 0 && bytes_written != -EFAULT &&
-					*ppos > isize))
-		*ppos = isize;
-
-	if (*ppos > ip->i_size) {
-		xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
-		if (*ppos > ip->i_size)
-			ip->i_size = *ppos;
-		xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
-	}
-}
-
-/*
- * If this was a direct or synchronous I/O that failed (such as ENOSPC) then
- * part of the I/O may have been written to disk before the error occurred.  In
- * this case the on-disk file size may have been adjusted beyond the in-memory
- * file size and now needs to be truncated back.
- */
-STATIC void
-xfs_aio_write_newsize_update(
-	struct xfs_inode	*ip,
-	xfs_fsize_t		new_size)
-{
-	if (new_size == ip->i_new_size) {
-		xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
-		if (new_size == ip->i_new_size)
-			ip->i_new_size = 0;
-		if (ip->i_d.di_size > ip->i_size)
-			ip->i_d.di_size = ip->i_size;
-		xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
-	}
-}
-
 /*
  * xfs_file_splice_write() does not use xfs_rw_ilock() because
  * generic_file_splice_write() takes the i_mutex itself. This, in theory,
@@ -475,7 +370,6 @@ xfs_file_splice_write(
 {
 	struct inode		*inode = outfilp->f_mapping->host;
 	struct xfs_inode	*ip = XFS_I(inode);
-	xfs_fsize_t		new_size;
 	int			ioflags = 0;
 	ssize_t			ret;
 
@@ -489,132 +383,107 @@ xfs_file_splice_write(
 
 	xfs_ilock(ip, XFS_IOLOCK_EXCL);
 
-	new_size = *ppos + count;
-
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	if (new_size > ip->i_size)
-		ip->i_new_size = new_size;
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-
 	trace_xfs_file_splice_write(ip, count, *ppos, ioflags);
 
 	ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags);
+	if (ret > 0)
+		XFS_STATS_ADD(xs_write_bytes, ret);
 
-	xfs_aio_write_isize_update(inode, ppos, ret);
-	xfs_aio_write_newsize_update(ip, new_size);
 	xfs_iunlock(ip, XFS_IOLOCK_EXCL);
 	return ret;
 }
 
 /*
- * This routine is called to handle zeroing any space in the last
- * block of the file that is beyond the EOF.  We do this since the
- * size is being increased without writing anything to that block
- * and we don't want anyone to read the garbage on the disk.
+ * This routine is called to handle zeroing any space in the last block of the
+ * file that is beyond the EOF.  We do this since the size is being increased
+ * without writing anything to that block and we don't want to read the
+ * garbage on the disk.
  */
 STATIC int				/* error (positive) */
 xfs_zero_last_block(
-	xfs_inode_t	*ip,
-	xfs_fsize_t	offset,
-	xfs_fsize_t	isize)
+	struct xfs_inode	*ip,
+	xfs_fsize_t		offset,
+	xfs_fsize_t		isize)
 {
-	xfs_fileoff_t	last_fsb;
-	xfs_mount_t	*mp = ip->i_mount;
-	int		nimaps;
-	int		zero_offset;
-	int		zero_len;
-	int		error = 0;
-	xfs_bmbt_irec_t	imap;
-
-	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-
-	zero_offset = XFS_B_FSB_OFFSET(mp, isize);
-	if (zero_offset == 0) {
-		/*
-		 * There are no extra bytes in the last block on disk to
-		 * zero, so return.
-		 */
-		return 0;
-	}
+	struct xfs_mount	*mp = ip->i_mount;
+	xfs_fileoff_t		last_fsb = XFS_B_TO_FSBT(mp, isize);
+	int			zero_offset = XFS_B_FSB_OFFSET(mp, isize);
+	int			zero_len;
+	int			nimaps = 1;
+	int			error = 0;
+	struct xfs_bmbt_irec	imap;
 
-	last_fsb = XFS_B_TO_FSBT(mp, isize);
-	nimaps = 1;
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
 	error = xfs_bmapi_read(ip, last_fsb, 1, &imap, &nimaps, 0);
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 	if (error)
 		return error;
+
 	ASSERT(nimaps > 0);
+
 	/*
 	 * If the block underlying isize is just a hole, then there
 	 * is nothing to zero.
 	 */
-	if (imap.br_startblock == HOLESTARTBLOCK) {
+	if (imap.br_startblock == HOLESTARTBLOCK)
 		return 0;
-	}
-	/*
-	 * Zero the part of the last block beyond the EOF, and write it
-	 * out sync.  We need to drop the ilock while we do this so we
-	 * don't deadlock when the buffer cache calls back to us.
-	 */
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 
 	zero_len = mp->m_sb.sb_blocksize - zero_offset;
 	if (isize + zero_len > offset)
 		zero_len = offset - isize;
-	error = xfs_iozero(ip, isize, zero_len);
-
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	ASSERT(error >= 0);
-	return error;
+	return xfs_iozero(ip, isize, zero_len);
 }
 
 /*
- * Zero any on disk space between the current EOF and the new,
- * larger EOF.  This handles the normal case of zeroing the remainder
- * of the last block in the file and the unusual case of zeroing blocks
- * out beyond the size of the file.  This second case only happens
- * with fixed size extents and when the system crashes before the inode
- * size was updated but after blocks were allocated.  If fill is set,
- * then any holes in the range are filled and zeroed.  If not, the holes
- * are left alone as holes.
+ * Zero any on disk space between the current EOF and the new, larger EOF.
+ *
+ * This handles the normal case of zeroing the remainder of the last block in
+ * the file and the unusual case of zeroing blocks out beyond the size of the
+ * file.  This second case only happens with fixed size extents and when the
+ * system crashes before the inode size was updated but after blocks were
+ * allocated.
+ *
+ * Expects the iolock to be held exclusive, and will take the ilock internally.
  */
-
 int					/* error (positive) */
 xfs_zero_eof(
-	xfs_inode_t	*ip,
-	xfs_off_t	offset,		/* starting I/O offset */
-	xfs_fsize_t	isize)		/* current inode size */
+	struct xfs_inode	*ip,
+	xfs_off_t		offset,		/* starting I/O offset */
+	xfs_fsize_t		isize)		/* current inode size */
 {
-	xfs_mount_t	*mp = ip->i_mount;
-	xfs_fileoff_t	start_zero_fsb;
-	xfs_fileoff_t	end_zero_fsb;
-	xfs_fileoff_t	zero_count_fsb;
-	xfs_fileoff_t	last_fsb;
-	xfs_fileoff_t	zero_off;
-	xfs_fsize_t	zero_len;
-	int		nimaps;
-	int		error = 0;
-	xfs_bmbt_irec_t	imap;
-
-	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
+	struct xfs_mount	*mp = ip->i_mount;
+	xfs_fileoff_t		start_zero_fsb;
+	xfs_fileoff_t		end_zero_fsb;
+	xfs_fileoff_t		zero_count_fsb;
+	xfs_fileoff_t		last_fsb;
+	xfs_fileoff_t		zero_off;
+	xfs_fsize_t		zero_len;
+	int			nimaps;
+	int			error = 0;
+	struct xfs_bmbt_irec	imap;
+
+	ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
 	ASSERT(offset > isize);
 
 	/*
 	 * First handle zeroing the block on which isize resides.
+	 *
 	 * We only zero a part of that block so it is handled specially.
 	 */
-	error = xfs_zero_last_block(ip, offset, isize);
-	if (error) {
-		ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
-		return error;
+	if (XFS_B_FSB_OFFSET(mp, isize) != 0) {
+		error = xfs_zero_last_block(ip, offset, isize);
+		if (error)
+			return error;
 	}
 
 	/*
-	 * Calculate the range between the new size and the old
-	 * where blocks needing to be zeroed may exist.  To get the
-	 * block where the last byte in the file currently resides,
-	 * we need to subtract one from the size and truncate back
-	 * to a block boundary.  We subtract 1 in case the size is
-	 * exactly on a block boundary.
+	 * Calculate the range between the new size and the old where blocks
+	 * needing to be zeroed may exist.
+	 *
+	 * To get the block where the last byte in the file currently resides,
+	 * we need to subtract one from the size and truncate back to a block
+	 * boundary.  We subtract 1 in case the size is exactly on a block
+	 * boundary.
 	 */
 	last_fsb = isize ? XFS_B_TO_FSBT(mp, isize - 1) : (xfs_fileoff_t)-1;
 	start_zero_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)isize);
@@ -632,23 +501,18 @@ xfs_zero_eof(
 	while (start_zero_fsb <= end_zero_fsb) {
 		nimaps = 1;
 		zero_count_fsb = end_zero_fsb - start_zero_fsb + 1;
+
+		xfs_ilock(ip, XFS_ILOCK_EXCL);
 		error = xfs_bmapi_read(ip, start_zero_fsb, zero_count_fsb,
 					  &imap, &nimaps, 0);
-		if (error) {
-			ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
+		xfs_iunlock(ip, XFS_ILOCK_EXCL);
+		if (error)
 			return error;
-		}
+
 		ASSERT(nimaps > 0);
 
 		if (imap.br_state == XFS_EXT_UNWRITTEN ||
 		    imap.br_startblock == HOLESTARTBLOCK) {
-			/*
-			 * This loop handles initializing pages that were
-			 * partially initialized by the code below this
-			 * loop. It basically zeroes the part of the page
-			 * that sits on a hole and sets the page as P_HOLE
-			 * and calls remapf if it is a mapped file.
-			 */
 			start_zero_fsb = imap.br_startoff + imap.br_blockcount;
 			ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
 			continue;
@@ -656,11 +520,7 @@ xfs_zero_eof(
 
 		/*
 		 * There are blocks we need to zero.
-		 * Drop the inode lock while we're doing the I/O.
-		 * We'll still have the iolock to protect us.
 		 */
-		xfs_iunlock(ip, XFS_ILOCK_EXCL);
-
 		zero_off = XFS_FSB_TO_B(mp, start_zero_fsb);
 		zero_len = XFS_FSB_TO_B(mp, imap.br_blockcount);
 
@@ -668,99 +528,76 @@ xfs_zero_eof(
 			zero_len = offset - zero_off;
 
 		error = xfs_iozero(ip, zero_off, zero_len);
-		if (error) {
-			goto out_lock;
-		}
+		if (error)
+			return error;
 
 		start_zero_fsb = imap.br_startoff + imap.br_blockcount;
 		ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
-
-		xfs_ilock(ip, XFS_ILOCK_EXCL);
 	}
 
 	return 0;
-
-out_lock:
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	ASSERT(error >= 0);
-	return error;
 }
 
 /*
  * Common pre-write limit and setup checks.
  *
- * Returns with iolock held according to @iolock.
+ * Called with the iolocked held either shared and exclusive according to
+ * @iolock, and returns with it held.  Might upgrade the iolock to exclusive
+ * if called for a direct write beyond i_size.
  */
 STATIC ssize_t
 xfs_file_aio_write_checks(
 	struct file		*file,
 	loff_t			*pos,
 	size_t			*count,
-	xfs_fsize_t		*new_sizep,
 	int			*iolock)
 {
 	struct inode		*inode = file->f_mapping->host;
 	struct xfs_inode	*ip = XFS_I(inode);
-	xfs_fsize_t		new_size;
 	int			error = 0;
 
-	xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
-	*new_sizep = 0;
 restart:
 	error = generic_write_checks(file, pos, count, S_ISBLK(inode->i_mode));
-	if (error) {
-		xfs_rw_iunlock(ip, XFS_ILOCK_EXCL | *iolock);
-		*iolock = 0;
+	if (error)
 		return error;
-	}
-
-	if (likely(!(file->f_mode & FMODE_NOCMTIME)))
-		file_update_time(file);
 
 	/*
 	 * If the offset is beyond the size of the file, we need to zero any
 	 * blocks that fall between the existing EOF and the start of this
-	 * write. There is no need to issue zeroing if another in-flght IO ends
-	 * at or before this one If zeronig is needed and we are currently
-	 * holding the iolock shared, we need to update it to exclusive which
-	 * involves dropping all locks and relocking to maintain correct locking
-	 * order. If we do this, restart the function to ensure all checks and
-	 * values are still valid.
+	 * write.  If zeroing is needed and we are currently holding the
+	 * iolock shared, we need to update it to exclusive which implies
+	 * having to redo all checks before.
 	 */
-	if ((ip->i_new_size && *pos > ip->i_new_size) ||
-	    (!ip->i_new_size && *pos > ip->i_size)) {
+	if (*pos > i_size_read(inode)) {
 		if (*iolock == XFS_IOLOCK_SHARED) {
-			xfs_rw_iunlock(ip, XFS_ILOCK_EXCL | *iolock);
+			xfs_rw_iunlock(ip, *iolock);
 			*iolock = XFS_IOLOCK_EXCL;
-			xfs_rw_ilock(ip, XFS_ILOCK_EXCL | *iolock);
+			xfs_rw_ilock(ip, *iolock);
 			goto restart;
 		}
-		error = -xfs_zero_eof(ip, *pos, ip->i_size);
+		error = -xfs_zero_eof(ip, *pos, i_size_read(inode));
+		if (error)
+			return error;
 	}
 
 	/*
-	 * If this IO extends beyond EOF, we may need to update ip->i_new_size.
-	 * We have already zeroed space beyond EOF (if necessary).  Only update
-	 * ip->i_new_size if this IO ends beyond any other in-flight writes.
+	 * Updating the timestamps will grab the ilock again from
+	 * xfs_fs_dirty_inode, so we have to call it after dropping the
+	 * lock above.  Eventually we should look into a way to avoid
+	 * the pointless lock roundtrip.
 	 */
-	new_size = *pos + *count;
-	if (new_size > ip->i_size) {
-		if (new_size > ip->i_new_size)
-			ip->i_new_size = new_size;
-		*new_sizep = new_size;
+	if (likely(!(file->f_mode & FMODE_NOCMTIME))) {
+		error = file_update_time(file);
+		if (error)
+			return error;
 	}
 
-	xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
-	if (error)
-		return error;
-
 	/*
 	 * If we're writing the file then make sure to clear the setuid and
 	 * setgid bits if the process is not being run by root.  This keeps
 	 * people from modifying setuid and setgid binaries.
 	 */
 	return file_remove_suid(file);
-
 }
 
 /*
@@ -794,9 +631,7 @@ xfs_file_dio_aio_write(
 	const struct iovec	*iovp,
 	unsigned long		nr_segs,
 	loff_t			pos,
-	size_t			ocount,
-	xfs_fsize_t		*new_size,
-	int			*iolock)
+	size_t			ocount)
 {
 	struct file		*file = iocb->ki_filp;
 	struct address_space	*mapping = file->f_mapping;
@@ -806,10 +641,10 @@ xfs_file_dio_aio_write(
 	ssize_t			ret = 0;
 	size_t			count = ocount;
 	int			unaligned_io = 0;
+	int			iolock;
 	struct xfs_buftarg	*target = XFS_IS_REALTIME_INODE(ip) ?
 					mp->m_rtdev_targp : mp->m_ddev_targp;
 
-	*iolock = 0;
 	if ((pos & target->bt_smask) || (count & target->bt_smask))
 		return -XFS_ERROR(EINVAL);
 
@@ -824,31 +659,31 @@ xfs_file_dio_aio_write(
 	 * EOF zeroing cases and fill out the new inode size as appropriate.
 	 */
 	if (unaligned_io || mapping->nrpages)
-		*iolock = XFS_IOLOCK_EXCL;
+		iolock = XFS_IOLOCK_EXCL;
 	else
-		*iolock = XFS_IOLOCK_SHARED;
-	xfs_rw_ilock(ip, *iolock);
+		iolock = XFS_IOLOCK_SHARED;
+	xfs_rw_ilock(ip, iolock);
 
 	/*
 	 * Recheck if there are cached pages that need invalidate after we got
 	 * the iolock to protect against other threads adding new pages while
 	 * we were waiting for the iolock.
 	 */
-	if (mapping->nrpages && *iolock == XFS_IOLOCK_SHARED) {
-		xfs_rw_iunlock(ip, *iolock);
-		*iolock = XFS_IOLOCK_EXCL;
-		xfs_rw_ilock(ip, *iolock);
+	if (mapping->nrpages && iolock == XFS_IOLOCK_SHARED) {
+		xfs_rw_iunlock(ip, iolock);
+		iolock = XFS_IOLOCK_EXCL;
+		xfs_rw_ilock(ip, iolock);
 	}
 
-	ret = xfs_file_aio_write_checks(file, &pos, &count, new_size, iolock);
+	ret = xfs_file_aio_write_checks(file, &pos, &count, &iolock);
 	if (ret)
-		return ret;
+		goto out;
 
 	if (mapping->nrpages) {
 		ret = -xfs_flushinval_pages(ip, (pos & PAGE_CACHE_MASK), -1,
 							FI_REMAPF_LOCKED);
 		if (ret)
-			return ret;
+			goto out;
 	}
 
 	/*
@@ -857,15 +692,18 @@ xfs_file_dio_aio_write(
 	 */
 	if (unaligned_io)
 		inode_dio_wait(inode);
-	else if (*iolock == XFS_IOLOCK_EXCL) {
+	else if (iolock == XFS_IOLOCK_EXCL) {
 		xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
-		*iolock = XFS_IOLOCK_SHARED;
+		iolock = XFS_IOLOCK_SHARED;
 	}
 
 	trace_xfs_file_direct_write(ip, count, iocb->ki_pos, 0);
 	ret = generic_file_direct_write(iocb, iovp,
 			&nr_segs, pos, &iocb->ki_pos, count, ocount);
 
+out:
+	xfs_rw_iunlock(ip, iolock);
+
 	/* No fallback to buffered IO on errors for XFS. */
 	ASSERT(ret < 0 || ret == count);
 	return ret;
@@ -877,9 +715,7 @@ xfs_file_buffered_aio_write(
 	const struct iovec	*iovp,
 	unsigned long		nr_segs,
 	loff_t			pos,
-	size_t			ocount,
-	xfs_fsize_t		*new_size,
-	int			*iolock)
+	size_t			ocount)
 {
 	struct file		*file = iocb->ki_filp;
 	struct address_space	*mapping = file->f_mapping;
@@ -887,14 +723,14 @@ xfs_file_buffered_aio_write(
 	struct xfs_inode	*ip = XFS_I(inode);
 	ssize_t			ret;
 	int			enospc = 0;
+	int			iolock = XFS_IOLOCK_EXCL;
 	size_t			count = ocount;
 
-	*iolock = XFS_IOLOCK_EXCL;
-	xfs_rw_ilock(ip, *iolock);
+	xfs_rw_ilock(ip, iolock);
 
-	ret = xfs_file_aio_write_checks(file, &pos, &count, new_size, iolock);
+	ret = xfs_file_aio_write_checks(file, &pos, &count, &iolock);
 	if (ret)
-		return ret;
+		goto out;
 
 	/* We can write back this queue in page reclaim */
 	current->backing_dev_info = mapping->backing_dev_info;
@@ -908,13 +744,15 @@ write_retry:
 	 * page locks and retry *once*
 	 */
 	if (ret == -ENOSPC && !enospc) {
-		ret = -xfs_flush_pages(ip, 0, -1, 0, FI_NONE);
-		if (ret)
-			return ret;
 		enospc = 1;
-		goto write_retry;
+		ret = -xfs_flush_pages(ip, 0, -1, 0, FI_NONE);
+		if (!ret)
+			goto write_retry;
 	}
+
 	current->backing_dev_info = NULL;
+out:
+	xfs_rw_iunlock(ip, iolock);
 	return ret;
 }
 
@@ -930,9 +768,7 @@ xfs_file_aio_write(
 	struct inode		*inode = mapping->host;
 	struct xfs_inode	*ip = XFS_I(inode);
 	ssize_t			ret;
-	int			iolock;
 	size_t			ocount = 0;
-	xfs_fsize_t		new_size = 0;
 
 	XFS_STATS_INC(xs_write_calls);
 
@@ -951,33 +787,22 @@ xfs_file_aio_write(
 		return -EIO;
 
 	if (unlikely(file->f_flags & O_DIRECT))
-		ret = xfs_file_dio_aio_write(iocb, iovp, nr_segs, pos,
-						ocount, &new_size, &iolock);
+		ret = xfs_file_dio_aio_write(iocb, iovp, nr_segs, pos, ocount);
 	else
 		ret = xfs_file_buffered_aio_write(iocb, iovp, nr_segs, pos,
-						ocount, &new_size, &iolock);
-
-	xfs_aio_write_isize_update(inode, &iocb->ki_pos, ret);
+						  ocount);
 
-	if (ret <= 0)
-		goto out_unlock;
+	if (ret > 0) {
+		ssize_t err;
 
-	/* Handle various SYNC-type writes */
-	if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) {
-		loff_t end = pos + ret - 1;
-		int error;
+		XFS_STATS_ADD(xs_write_bytes, ret);
 
-		xfs_rw_iunlock(ip, iolock);
-		error = xfs_file_fsync(file, pos, end,
-				      (file->f_flags & __O_SYNC) ? 0 : 1);
-		xfs_rw_ilock(ip, iolock);
-		if (error)
-			ret = error;
+		/* Handle various SYNC-type writes */
+		err = generic_write_sync(file, pos, ret);
+		if (err < 0)
+			ret = err;
 	}
 
-out_unlock:
-	xfs_aio_write_newsize_update(ip, new_size);
-	xfs_rw_iunlock(ip, iolock);
 	return ret;
 }
 
@@ -1141,8 +966,149 @@ xfs_vm_page_mkwrite(
 	return block_page_mkwrite(vma, vmf, xfs_get_blocks);
 }
 
+STATIC loff_t
+xfs_seek_data(
+	struct file		*file,
+	loff_t			start,
+	u32			type)
+{
+	struct inode		*inode = file->f_mapping->host;
+	struct xfs_inode	*ip = XFS_I(inode);
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_bmbt_irec	map[2];
+	int			nmap = 2;
+	loff_t			uninitialized_var(offset);
+	xfs_fsize_t		isize;
+	xfs_fileoff_t		fsbno;
+	xfs_filblks_t		end;
+	uint			lock;
+	int			error;
+
+	lock = xfs_ilock_map_shared(ip);
+
+	isize = i_size_read(inode);
+	if (start >= isize) {
+		error = ENXIO;
+		goto out_unlock;
+	}
+
+	fsbno = XFS_B_TO_FSBT(mp, start);
+
+	/*
+	 * Try to read extents from the first block indicated
+	 * by fsbno to the end block of the file.
+	 */
+	end = XFS_B_TO_FSB(mp, isize);
+
+	error = xfs_bmapi_read(ip, fsbno, end - fsbno, map, &nmap,
+			       XFS_BMAPI_ENTIRE);
+	if (error)
+		goto out_unlock;
+
+	/*
+	 * Treat unwritten extent as data extent since it might
+	 * contains dirty data in page cache.
+	 */
+	if (map[0].br_startblock != HOLESTARTBLOCK) {
+		offset = max_t(loff_t, start,
+			       XFS_FSB_TO_B(mp, map[0].br_startoff));
+	} else {
+		if (nmap == 1) {
+			error = ENXIO;
+			goto out_unlock;
+		}
+
+		offset = max_t(loff_t, start,
+			       XFS_FSB_TO_B(mp, map[1].br_startoff));
+	}
+
+	if (offset != file->f_pos)
+		file->f_pos = offset;
+
+out_unlock:
+	xfs_iunlock_map_shared(ip, lock);
+
+	if (error)
+		return -error;
+	return offset;
+}
+
+STATIC loff_t
+xfs_seek_hole(
+	struct file		*file,
+	loff_t			start,
+	u32			type)
+{
+	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);
+	loff_t			holeoff;
+	xfs_fsize_t		isize;
+	xfs_fileoff_t		fsbno;
+	uint			lock;
+	int			error;
+
+	if (XFS_FORCED_SHUTDOWN(mp))
+		return -XFS_ERROR(EIO);
+
+	lock = xfs_ilock_map_shared(ip);
+
+	isize = i_size_read(inode);
+	if (start >= isize) {
+		error = ENXIO;
+		goto out_unlock;
+	}
+
+	fsbno = XFS_B_TO_FSBT(mp, start);
+	error = xfs_bmap_first_unused(NULL, ip, 1, &fsbno, XFS_DATA_FORK);
+	if (error)
+		goto out_unlock;
+
+	holeoff = XFS_FSB_TO_B(mp, fsbno);
+	if (holeoff <= start)
+		offset = start;
+	else {
+		/*
+		 * xfs_bmap_first_unused() could return a value bigger than
+		 * isize if there are no more holes past the supplied offset.
+		 */
+		offset = min_t(loff_t, holeoff, isize);
+	}
+
+	if (offset != file->f_pos)
+		file->f_pos = offset;
+
+out_unlock:
+	xfs_iunlock_map_shared(ip, lock);
+
+	if (error)
+		return -error;
+	return offset;
+}
+
+STATIC loff_t
+xfs_file_llseek(
+	struct file	*file,
+	loff_t		offset,
+	int		origin)
+{
+	switch (origin) {
+	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, origin);
+	case SEEK_HOLE:
+		return xfs_seek_hole(file, offset, origin);
+	default:
+		return -EINVAL;
+	}
+}
+
 const struct file_operations xfs_file_operations = {
-	.llseek		= generic_file_llseek,
+	.llseek		= xfs_file_llseek,
 	.read		= do_sync_read,
 	.write		= do_sync_write,
 	.aio_read	= xfs_file_aio_read,
diff --git a/fs/xfs/xfs_fs_subr.c b/fs/xfs/xfs_fs_subr.c
index ed88ed16811..652b875a9d4 100644
--- a/fs/xfs/xfs_fs_subr.c
+++ b/fs/xfs/xfs_fs_subr.c
@@ -90,7 +90,7 @@ xfs_wait_on_pages(
 
 	if (mapping_tagged(mapping, PAGECACHE_TAG_WRITEBACK)) {
 		return -filemap_fdatawait_range(mapping, first,
-					last == -1 ? ip->i_size - 1 : last);
+					last == -1 ? XFS_ISIZE(ip) - 1 : last);
 	}
 	return 0;
 }
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index 1c6fdeb702f..c25b094efbf 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -18,8 +18,6 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_types.h"
-#include "xfs_bit.h"
-#include "xfs_inum.h"
 #include "xfs_log.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
@@ -39,7 +37,6 @@
 #include "xfs_itable.h"
 #include "xfs_trans_space.h"
 #include "xfs_rtalloc.h"
-#include "xfs_rw.h"
 #include "xfs_filestream.h"
 #include "xfs_trace.h"
 
@@ -147,9 +144,9 @@ xfs_growfs_data_private(
 	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, mp->m_ddev_targp,
+	bp = xfs_buf_read_uncached(mp->m_ddev_targp,
 				XFS_FSB_TO_BB(mp, nb) - XFS_FSS_TO_BB(mp, 1),
-				BBTOB(XFS_FSS_TO_BB(mp, 1)), 0);
+				XFS_FSS_TO_BB(mp, 1), 0);
 	if (!bp)
 		return EIO;
 	xfs_buf_relse(bp);
@@ -193,7 +190,7 @@ xfs_growfs_data_private(
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
 				 XFS_AG_DADDR(mp, agno, XFS_AGF_DADDR(mp)),
-				 XFS_FSS_TO_BB(mp, 1), XBF_LOCK | XBF_MAPPED);
+				 XFS_FSS_TO_BB(mp, 1), 0);
 		if (!bp) {
 			error = ENOMEM;
 			goto error0;
@@ -230,7 +227,7 @@ xfs_growfs_data_private(
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
 				 XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)),
-				 XFS_FSS_TO_BB(mp, 1), XBF_LOCK | XBF_MAPPED);
+				 XFS_FSS_TO_BB(mp, 1), 0);
 		if (!bp) {
 			error = ENOMEM;
 			goto error0;
@@ -259,8 +256,7 @@ xfs_growfs_data_private(
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
 				 XFS_AGB_TO_DADDR(mp, agno, XFS_BNO_BLOCK(mp)),
-				 BTOBB(mp->m_sb.sb_blocksize),
-				 XBF_LOCK | XBF_MAPPED);
+				 BTOBB(mp->m_sb.sb_blocksize), 0);
 		if (!bp) {
 			error = ENOMEM;
 			goto error0;
@@ -286,8 +282,7 @@ xfs_growfs_data_private(
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
 				 XFS_AGB_TO_DADDR(mp, agno, XFS_CNT_BLOCK(mp)),
-				 BTOBB(mp->m_sb.sb_blocksize),
-				 XBF_LOCK | XBF_MAPPED);
+				 BTOBB(mp->m_sb.sb_blocksize), 0);
 		if (!bp) {
 			error = ENOMEM;
 			goto error0;
@@ -314,8 +309,7 @@ xfs_growfs_data_private(
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
 				 XFS_AGB_TO_DADDR(mp, agno, XFS_IBT_BLOCK(mp)),
-				 BTOBB(mp->m_sb.sb_blocksize),
-				 XBF_LOCK | XBF_MAPPED);
+				 BTOBB(mp->m_sb.sb_blocksize), 0);
 		if (!bp) {
 			error = ENOMEM;
 			goto error0;
@@ -405,7 +399,7 @@ xfs_growfs_data_private(
 
 	/* update secondary superblocks. */
 	for (agno = 1; agno < nagcount; agno++) {
-		error = xfs_read_buf(mp, mp->m_ddev_targp,
+		error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp,
 				  XFS_AGB_TO_DADDR(mp, agno, XFS_SB_BLOCK(mp)),
 				  XFS_FSS_TO_BB(mp, 1), 0, &bp);
 		if (error) {
@@ -693,3 +687,63 @@ xfs_fs_goingdown(
 
 	return 0;
 }
+
+/*
+ * Force a shutdown of the filesystem instantly while keeping the filesystem
+ * consistent. We don't do an unmount here; just shutdown the shop, make sure
+ * that absolutely nothing persistent happens to this filesystem after this
+ * point.
+ */
+void
+xfs_do_force_shutdown(
+	xfs_mount_t	*mp,
+	int		flags,
+	char		*fname,
+	int		lnnum)
+{
+	int		logerror;
+
+	logerror = flags & SHUTDOWN_LOG_IO_ERROR;
+
+	if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
+		xfs_notice(mp,
+	"%s(0x%x) called from line %d of file %s.  Return address = 0x%p",
+			__func__, flags, lnnum, fname, __return_address);
+	}
+	/*
+	 * No need to duplicate efforts.
+	 */
+	if (XFS_FORCED_SHUTDOWN(mp) && !logerror)
+		return;
+
+	/*
+	 * This flags XFS_MOUNT_FS_SHUTDOWN, makes sure that we don't
+	 * queue up anybody new on the log reservations, and wakes up
+	 * everybody who's sleeping on log reservations to tell them
+	 * the bad news.
+	 */
+	if (xfs_log_force_umount(mp, logerror))
+		return;
+
+	if (flags & SHUTDOWN_CORRUPT_INCORE) {
+		xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_CORRUPT,
+    "Corruption of in-memory data detected.  Shutting down filesystem");
+		if (XFS_ERRLEVEL_HIGH <= xfs_error_level)
+			xfs_stack_trace();
+	} else if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
+		if (logerror) {
+			xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_LOGERROR,
+		"Log I/O Error Detected.  Shutting down filesystem");
+		} else if (flags & SHUTDOWN_DEVICE_REQ) {
+			xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_IOERROR,
+		"All device paths lost.  Shutting down filesystem");
+		} else if (!(flags & SHUTDOWN_REMOTE_REQ)) {
+			xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_IOERROR,
+		"I/O Error Detected. Shutting down filesystem");
+		}
+	}
+	if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
+		xfs_alert(mp,
+	"Please umount the filesystem and rectify the problem(s)");
+	}
+}
diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c
index 169380e6605..177a21a7ac4 100644
--- a/fs/xfs/xfs_ialloc.c
+++ b/fs/xfs/xfs_ialloc.c
@@ -200,8 +200,7 @@ xfs_ialloc_inode_init(
 		 */
 		d = XFS_AGB_TO_DADDR(mp, agno, agbno + (j * blks_per_cluster));
 		fbuf = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
-					 mp->m_bsize * blks_per_cluster,
-					 XBF_LOCK);
+					 mp->m_bsize * blks_per_cluster, 0);
 		if (!fbuf)
 			return ENOMEM;
 		/*
@@ -447,7 +446,7 @@ STATIC xfs_buf_t *			/* allocation group buffer */
 xfs_ialloc_ag_select(
 	xfs_trans_t	*tp,		/* transaction pointer */
 	xfs_ino_t	parent,		/* parent directory inode number */
-	mode_t		mode,		/* bits set to indicate file type */
+	umode_t		mode,		/* bits set to indicate file type */
 	int		okalloc)	/* ok to allocate more space */
 {
 	xfs_buf_t	*agbp;		/* allocation group header buffer */
@@ -610,6 +609,13 @@ xfs_ialloc_get_rec(
 /*
  * Visible inode allocation functions.
  */
+/*
+ * Find a free (set) bit in the inode bitmask.
+ */
+static inline int xfs_ialloc_find_free(xfs_inofree_t *fp)
+{
+	return xfs_lowbit64(*fp);
+}
 
 /*
  * Allocate an inode on disk.
@@ -640,7 +646,7 @@ int
 xfs_dialloc(
 	xfs_trans_t	*tp,		/* transaction pointer */
 	xfs_ino_t	parent,		/* parent inode (directory) */
-	mode_t		mode,		/* mode bits for new inode */
+	umode_t		mode,		/* mode bits for new inode */
 	int		okalloc,	/* ok to allocate more space */
 	xfs_buf_t	**IO_agbp,	/* in/out ag header's buffer */
 	boolean_t	*alloc_done,	/* true if we needed to replenish
diff --git a/fs/xfs/xfs_ialloc.h b/fs/xfs/xfs_ialloc.h
index bb5385475e1..65ac57c8063 100644
--- a/fs/xfs/xfs_ialloc.h
+++ b/fs/xfs/xfs_ialloc.h
@@ -47,15 +47,6 @@ xfs_make_iptr(struct xfs_mount *mp, struct xfs_buf *b, int o)
 }
 
 /*
- * Find a free (set) bit in the inode bitmask.
- */
-static inline int xfs_ialloc_find_free(xfs_inofree_t *fp)
-{
-	return xfs_lowbit64(*fp);
-}
-
-
-/*
  * Allocate an inode on disk.
  * Mode is used to tell whether the new inode will need space, and whether
  * it is a directory.
@@ -81,7 +72,7 @@ int					/* error */
 xfs_dialloc(
 	struct xfs_trans *tp,		/* transaction pointer */
 	xfs_ino_t	parent,		/* parent inode (directory) */
-	mode_t		mode,		/* mode bits for new inode */
+	umode_t		mode,		/* mode bits for new inode */
 	int		okalloc,	/* ok to allocate more space */
 	struct xfs_buf	**agbp,		/* buf for a.g. inode header */
 	boolean_t	*alloc_done,	/* an allocation was done to replenish
diff --git a/fs/xfs/xfs_ialloc_btree.c b/fs/xfs/xfs_ialloc_btree.c
index c6a75815aea..2b8b7a37aa1 100644
--- a/fs/xfs/xfs_ialloc_btree.c
+++ b/fs/xfs/xfs_ialloc_btree.c
@@ -20,7 +20,6 @@
 #include "xfs_types.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c
index 0fa98b1c70e..1bb4365e8c2 100644
--- a/fs/xfs/xfs_iget.c
+++ b/fs/xfs/xfs_iget.c
@@ -19,7 +19,6 @@
 #include "xfs_fs.h"
 #include "xfs_types.h"
 #include "xfs_acl.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
 #include "xfs_inum.h"
 #include "xfs_trans.h"
@@ -77,7 +76,7 @@ xfs_inode_alloc(
 
 	ASSERT(atomic_read(&ip->i_pincount) == 0);
 	ASSERT(!spin_is_locked(&ip->i_flags_lock));
-	ASSERT(completion_done(&ip->i_flush));
+	ASSERT(!xfs_isiflocked(ip));
 	ASSERT(ip->i_ino == 0);
 
 	mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino);
@@ -91,11 +90,8 @@ xfs_inode_alloc(
 	ip->i_afp = NULL;
 	memset(&ip->i_df, 0, sizeof(xfs_ifork_t));
 	ip->i_flags = 0;
-	ip->i_update_core = 0;
 	ip->i_delayed_blks = 0;
 	memset(&ip->i_d, 0, sizeof(xfs_icdinode_t));
-	ip->i_size = 0;
-	ip->i_new_size = 0;
 
 	return ip;
 }
@@ -107,7 +103,6 @@ xfs_inode_free_callback(
 	struct inode		*inode = container_of(head, struct inode, i_rcu);
 	struct xfs_inode	*ip = XFS_I(inode);
 
-	INIT_LIST_HEAD(&inode->i_dentry);
 	kmem_zone_free(xfs_inode_zone, ip);
 }
 
@@ -127,23 +122,7 @@ xfs_inode_free(
 		xfs_idestroy_fork(ip, XFS_ATTR_FORK);
 
 	if (ip->i_itemp) {
-		/*
-		 * Only if we are shutting down the fs will we see an
-		 * inode still in the AIL. If it is there, we should remove
-		 * it to prevent a use-after-free from occurring.
-		 */
-		xfs_log_item_t	*lip = &ip->i_itemp->ili_item;
-		struct xfs_ail	*ailp = lip->li_ailp;
-
-		ASSERT(((lip->li_flags & XFS_LI_IN_AIL) == 0) ||
-				       XFS_FORCED_SHUTDOWN(ip->i_mount));
-		if (lip->li_flags & XFS_LI_IN_AIL) {
-			spin_lock(&ailp->xa_lock);
-			if (lip->li_flags & XFS_LI_IN_AIL)
-				xfs_trans_ail_delete(ailp, lip);
-			else
-				spin_unlock(&ailp->xa_lock);
-		}
+		ASSERT(!(ip->i_itemp->ili_item.li_flags & XFS_LI_IN_AIL));
 		xfs_inode_item_destroy(ip);
 		ip->i_itemp = NULL;
 	}
@@ -151,7 +130,7 @@ xfs_inode_free(
 	/* asserts to verify all state is correct here */
 	ASSERT(atomic_read(&ip->i_pincount) == 0);
 	ASSERT(!spin_is_locked(&ip->i_flags_lock));
-	ASSERT(completion_done(&ip->i_flush));
+	ASSERT(!xfs_isiflocked(ip));
 
 	/*
 	 * Because we use RCU freeing we need to ensure the inode always
@@ -293,7 +272,7 @@ xfs_iget_cache_hit(
 	if (lock_flags != 0)
 		xfs_ilock(ip, lock_flags);
 
-	xfs_iflags_clear(ip, XFS_ISTALE);
+	xfs_iflags_clear(ip, XFS_ISTALE | XFS_IDONTCACHE);
 	XFS_STATS_INC(xs_ig_found);
 
 	return 0;
@@ -318,6 +297,7 @@ xfs_iget_cache_miss(
 	struct xfs_inode	*ip;
 	int			error;
 	xfs_agino_t		agino = XFS_INO_TO_AGINO(mp, ino);
+	int			iflags;
 
 	ip = xfs_inode_alloc(mp, ino);
 	if (!ip)
@@ -337,9 +317,10 @@ xfs_iget_cache_miss(
 	/*
 	 * Preload the radix tree so we can insert safely under the
 	 * write spinlock. Note that we cannot sleep inside the preload
-	 * region.
+	 * region. Since we can be called from transaction context, don't
+	 * recurse into the file system.
 	 */
-	if (radix_tree_preload(GFP_KERNEL)) {
+	if (radix_tree_preload(GFP_NOFS)) {
 		error = EAGAIN;
 		goto out_destroy;
 	}
@@ -353,9 +334,23 @@ xfs_iget_cache_miss(
 			BUG();
 	}
 
-	spin_lock(&pag->pag_ici_lock);
+	/*
+	 * These values must be set before inserting the inode into the radix
+	 * tree as the moment it is inserted a concurrent lookup (allowed by the
+	 * RCU locking mechanism) can find it and that lookup must see that this
+	 * is an inode currently under construction (i.e. that XFS_INEW is set).
+	 * The ip->i_flags_lock that protects the XFS_INEW flag forms the
+	 * memory barrier that ensures this detection works correctly at lookup
+	 * time.
+	 */
+	iflags = XFS_INEW;
+	if (flags & XFS_IGET_DONTCACHE)
+		iflags |= XFS_IDONTCACHE;
+	ip->i_udquot = ip->i_gdquot = NULL;
+	xfs_iflags_set(ip, iflags);
 
 	/* insert the new inode */
+	spin_lock(&pag->pag_ici_lock);
 	error = radix_tree_insert(&pag->pag_ici_root, agino, ip);
 	if (unlikely(error)) {
 		WARN_ON(error != -EEXIST);
@@ -363,11 +358,6 @@ xfs_iget_cache_miss(
 		error = EAGAIN;
 		goto out_preload_end;
 	}
-
-	/* These values _must_ be set before releasing the radix tree lock! */
-	ip->i_udquot = ip->i_gdquot = NULL;
-	xfs_iflags_set(ip, XFS_INEW);
-
 	spin_unlock(&pag->pag_ici_lock);
 	radix_tree_preload_end();
 
@@ -421,6 +411,15 @@ xfs_iget(
 	xfs_perag_t	*pag;
 	xfs_agino_t	agino;
 
+	/*
+	 * xfs_reclaim_inode() uses the ILOCK to ensure an inode
+	 * doesn't get freed while it's being referenced during a
+	 * radix tree traversal here.  It assumes this function
+	 * aqcuires only the ILOCK (and therefore it has no need to
+	 * involve the IOLOCK in this synchronization).
+	 */
+	ASSERT((lock_flags & (XFS_IOLOCK_EXCL | XFS_IOLOCK_SHARED)) == 0);
+
 	/* reject inode numbers outside existing AGs */
 	if (!ino || XFS_INO_TO_AGNO(mp, ino) >= mp->m_sb.sb_agcount)
 		return EINVAL;
@@ -451,8 +450,6 @@ again:
 
 	*ipp = ip;
 
-	ASSERT(ip->i_df.if_ext_max ==
-	       XFS_IFORK_DSIZE(ip) / sizeof(xfs_bmbt_rec_t));
 	/*
 	 * If we have a real type for an on-disk inode, we can set ops(&unlock)
 	 * now.	 If it's a new inode being created, xfs_ialloc will handle it.
@@ -647,8 +644,7 @@ xfs_iunlock(
 	       (XFS_IOLOCK_SHARED | XFS_IOLOCK_EXCL));
 	ASSERT((lock_flags & (XFS_ILOCK_SHARED | XFS_ILOCK_EXCL)) !=
 	       (XFS_ILOCK_SHARED | XFS_ILOCK_EXCL));
-	ASSERT((lock_flags & ~(XFS_LOCK_MASK | XFS_IUNLOCK_NONOTIFY |
-			XFS_LOCK_DEP_MASK)) == 0);
+	ASSERT((lock_flags & ~(XFS_LOCK_MASK | XFS_LOCK_DEP_MASK)) == 0);
 	ASSERT(lock_flags != 0);
 
 	if (lock_flags & XFS_IOLOCK_EXCL)
@@ -661,16 +657,6 @@ xfs_iunlock(
 	else if (lock_flags & XFS_ILOCK_SHARED)
 		mrunlock_shared(&ip->i_lock);
 
-	if ((lock_flags & (XFS_ILOCK_SHARED | XFS_ILOCK_EXCL)) &&
-	    !(lock_flags & XFS_IUNLOCK_NONOTIFY) && ip->i_itemp) {
-		/*
-		 * Let the AIL know that this item has been unlocked in case
-		 * it is in the AIL and anyone is waiting on it.  Don't do
-		 * this if the caller has asked us not to.
-		 */
-		xfs_trans_unlocked_item(ip->i_itemp->ili_item.li_ailp,
-					(xfs_log_item_t*)(ip->i_itemp));
-	}
 	trace_xfs_iunlock(ip, lock_flags, _RET_IP_);
 }
 
@@ -716,3 +702,19 @@ xfs_isilocked(
 	return 0;
 }
 #endif
+
+void
+__xfs_iflock(
+	struct xfs_inode	*ip)
+{
+	wait_queue_head_t *wq = bit_waitqueue(&ip->i_flags, __XFS_IFLOCK_BIT);
+	DEFINE_WAIT_BIT(wait, &ip->i_flags, __XFS_IFLOCK_BIT);
+
+	do {
+		prepare_to_wait_exclusive(wq, &wait.wait, TASK_UNINTERRUPTIBLE);
+		if (xfs_isiflocked(ip))
+			io_schedule();
+	} while (!xfs_iflock_nowait(ip));
+
+	finish_wait(wq, &wait.wait);
+}
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 755ee816488..a59eea09930 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -20,7 +20,6 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_types.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
 #include "xfs_inum.h"
 #include "xfs_trans.h"
@@ -61,6 +60,20 @@ STATIC int xfs_iformat_local(xfs_inode_t *, xfs_dinode_t *, int, int);
 STATIC int xfs_iformat_extents(xfs_inode_t *, xfs_dinode_t *, int);
 STATIC int xfs_iformat_btree(xfs_inode_t *, xfs_dinode_t *, int);
 
+/*
+ * helper function to extract extent size hint from inode
+ */
+xfs_extlen_t
+xfs_get_extsz_hint(
+	struct xfs_inode	*ip)
+{
+	if ((ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE) && ip->i_d.di_extsize)
+		return ip->i_d.di_extsize;
+	if (XFS_IS_REALTIME_INODE(ip))
+		return ip->i_mount->m_sb.sb_rextsize;
+	return 0;
+}
+
 #ifdef DEBUG
 /*
  * Make sure that the extents in the given memory buffer
@@ -137,6 +150,7 @@ xfs_imap_to_bp(
 	int		ni;
 	xfs_buf_t	*bp;
 
+	buf_flags |= XBF_UNMAPPED;
 	error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, imap->im_blkno,
 				   (int)imap->im_len, buf_flags, &bp);
 	if (error) {
@@ -226,7 +240,7 @@ xfs_inotobp(
 	if (error)
 		return error;
 
-	error = xfs_imap_to_bp(mp, tp, &imap, &bp, XBF_LOCK, imap_flags);
+	error = xfs_imap_to_bp(mp, tp, &imap, &bp, 0, imap_flags);
 	if (error)
 		return error;
 
@@ -299,11 +313,8 @@ xfs_iformat(
 {
 	xfs_attr_shortform_t	*atp;
 	int			size;
-	int			error;
+	int			error = 0;
 	xfs_fsize_t             di_size;
-	ip->i_df.if_ext_max =
-		XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
-	error = 0;
 
 	if (unlikely(be32_to_cpu(dip->di_nextents) +
 		     be16_to_cpu(dip->di_anextents) >
@@ -350,7 +361,6 @@ xfs_iformat(
 			return XFS_ERROR(EFSCORRUPTED);
 		}
 		ip->i_d.di_size = 0;
-		ip->i_size = 0;
 		ip->i_df.if_u2.if_rdev = xfs_dinode_get_rdev(dip);
 		break;
 
@@ -409,10 +419,10 @@ xfs_iformat(
 	}
 	if (!XFS_DFORK_Q(dip))
 		return 0;
+
 	ASSERT(ip->i_afp == NULL);
 	ip->i_afp = kmem_zone_zalloc(xfs_ifork_zone, KM_SLEEP | KM_NOFS);
-	ip->i_afp->if_ext_max =
-		XFS_IFORK_ASIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
+
 	switch (dip->di_aformat) {
 	case XFS_DINODE_FMT_LOCAL:
 		atp = (xfs_attr_shortform_t *)XFS_DFORK_APTR(dip);
@@ -604,10 +614,11 @@ xfs_iformat_btree(
 	 * or the number of extents is greater than the number of
 	 * blocks.
 	 */
-	if (unlikely(XFS_IFORK_NEXTENTS(ip, whichfork) <= ifp->if_ext_max
-	    || XFS_BMDR_SPACE_CALC(nrecs) >
-			XFS_DFORK_SIZE(dip, ip->i_mount, whichfork)
-	    || XFS_IFORK_NEXTENTS(ip, whichfork) > ip->i_d.di_nblocks)) {
+	if (unlikely(XFS_IFORK_NEXTENTS(ip, whichfork) <=
+			XFS_IFORK_MAXEXT(ip, whichfork) ||
+		     XFS_BMDR_SPACE_CALC(nrecs) >
+			XFS_DFORK_SIZE(dip, ip->i_mount, whichfork) ||
+		     XFS_IFORK_NEXTENTS(ip, whichfork) > ip->i_d.di_nblocks)) {
 		xfs_warn(ip->i_mount, "corrupt inode %Lu (btree).",
 			(unsigned long long) ip->i_ino);
 		XFS_CORRUPTION_ERROR("xfs_iformat_btree", XFS_ERRLEVEL_LOW,
@@ -785,8 +796,7 @@ xfs_iread(
 	/*
 	 * Get pointers to the on-disk inode and the buffer containing it.
 	 */
-	error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &bp,
-			       XBF_LOCK, iget_flags);
+	error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &bp, 0, iget_flags);
 	if (error)
 		return error;
 	dip = (xfs_dinode_t *)xfs_buf_offset(bp, ip->i_imap.im_boffset);
@@ -835,12 +845,6 @@ xfs_iread(
 		 * with the uninitialized part of it.
 		 */
 		ip->i_d.di_mode = 0;
-		/*
-		 * Initialize the per-fork minima and maxima for a new
-		 * inode here.  xfs_iformat will do it for old inodes.
-		 */
-		ip->i_df.if_ext_max =
-			XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
 	}
 
 	/*
@@ -861,7 +865,6 @@ xfs_iread(
 	}
 
 	ip->i_delayed_blks = 0;
-	ip->i_size = ip->i_d.di_size;
 
 	/*
 	 * Mark the buffer containing the inode as something to keep
@@ -961,7 +964,7 @@ int
 xfs_ialloc(
 	xfs_trans_t	*tp,
 	xfs_inode_t	*pip,
-	mode_t		mode,
+	umode_t		mode,
 	xfs_nlink_t	nlink,
 	xfs_dev_t	rdev,
 	prid_t		prid,
@@ -1002,7 +1005,7 @@ xfs_ialloc(
 		return error;
 	ASSERT(ip != NULL);
 
-	ip->i_d.di_mode = (__uint16_t)mode;
+	ip->i_d.di_mode = mode;
 	ip->i_d.di_onlink = 0;
 	ip->i_d.di_nlink = nlink;
 	ASSERT(ip->i_d.di_nlink == nlink);
@@ -1051,7 +1054,6 @@ xfs_ialloc(
 	}
 
 	ip->i_d.di_size = 0;
-	ip->i_size = 0;
 	ip->i_d.di_nextents = 0;
 	ASSERT(ip->i_d.di_nblocks == 0);
 
@@ -1166,52 +1168,6 @@ xfs_ialloc(
 }
 
 /*
- * Check to make sure that there are no blocks allocated to the
- * file beyond the size of the file.  We don't check this for
- * files with fixed size extents or real time extents, but we
- * at least do it for regular files.
- */
-#ifdef DEBUG
-STATIC void
-xfs_isize_check(
-	struct xfs_inode	*ip,
-	xfs_fsize_t		isize)
-{
-	struct xfs_mount	*mp = ip->i_mount;
-	xfs_fileoff_t		map_first;
-	int			nimaps;
-	xfs_bmbt_irec_t		imaps[2];
-	int			error;
-
-	if (!S_ISREG(ip->i_d.di_mode))
-		return;
-
-	if (XFS_IS_REALTIME_INODE(ip))
-		return;
-
-	if (ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE)
-		return;
-
-	nimaps = 2;
-	map_first = XFS_B_TO_FSB(mp, (xfs_ufsize_t)isize);
-	/*
-	 * The filesystem could be shutting down, so bmapi may return
-	 * an error.
-	 */
-	error = xfs_bmapi_read(ip, map_first,
-			 (XFS_B_TO_FSB(mp,
-			       (xfs_ufsize_t)XFS_MAXIOFFSET(mp)) - map_first),
-			 imaps, &nimaps, XFS_BMAPI_ENTIRE);
-	if (error)
-		return;
-	ASSERT(nimaps == 1);
-	ASSERT(imaps[0].br_startblock == HOLESTARTBLOCK);
-}
-#else	/* DEBUG */
-#define xfs_isize_check(ip, isize)
-#endif	/* DEBUG */
-
-/*
  * Free up the underlying blocks past new_size.  The new size must be smaller
  * than the current size.  This routine can be used both for the attribute and
  * data fork, and does not modify the inode size, which is left to the caller.
@@ -1252,12 +1208,14 @@ xfs_itruncate_extents(
 	int			done = 0;
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
-	ASSERT(new_size <= ip->i_size);
+	ASSERT(new_size <= XFS_ISIZE(ip));
 	ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
 	ASSERT(ip->i_itemp != NULL);
 	ASSERT(ip->i_itemp->ili_lock_flags == 0);
 	ASSERT(!XFS_NOT_DQATTACHED(mp, ip));
 
+	trace_xfs_itruncate_extents_start(ip, new_size);
+
 	/*
 	 * Since it is possible for space to become allocated beyond
 	 * the end of the file (in a crash where the space is allocated
@@ -1325,6 +1283,14 @@ xfs_itruncate_extents(
 			goto out;
 	}
 
+	/*
+	 * Always re-log the inode so that our permanent transaction can keep
+	 * on rolling it forward in the log.
+	 */
+	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+
+	trace_xfs_itruncate_extents_end(ip, new_size);
+
 out:
 	*tpp = tp;
 	return error;
@@ -1338,74 +1304,6 @@ out_bmap_cancel:
 	goto out;
 }
 
-int
-xfs_itruncate_data(
-	struct xfs_trans	**tpp,
-	struct xfs_inode	*ip,
-	xfs_fsize_t		new_size)
-{
-	int			error;
-
-	trace_xfs_itruncate_data_start(ip, new_size);
-
-	/*
-	 * The first thing we do is set the size to new_size permanently on
-	 * disk.  This way we don't have to worry about anyone ever being able
-	 * to look at the data being freed even in the face of a crash.
-	 * What we're getting around here is the case where we free a block, it
-	 * is allocated to another file, it is written to, and then we crash.
-	 * If the new data gets written to the file but the log buffers
-	 * containing the free and reallocation don't, then we'd end up with
-	 * garbage in the blocks being freed.  As long as we make the new_size
-	 * permanent before actually freeing any blocks it doesn't matter if
-	 * they get written to.
-	 */
-	if (ip->i_d.di_nextents > 0) {
-		/*
-		 * If we are not changing the file size then do not update
-		 * the on-disk file size - we may be called from
-		 * xfs_inactive_free_eofblocks().  If we update the on-disk
-		 * file size and then the system crashes before the contents
-		 * of the file are flushed to disk then the files may be
-		 * full of holes (ie NULL files bug).
-		 */
-		if (ip->i_size != new_size) {
-			ip->i_d.di_size = new_size;
-			ip->i_size = new_size;
-			xfs_trans_log_inode(*tpp, ip, XFS_ILOG_CORE);
-		}
-	}
-
-	error = xfs_itruncate_extents(tpp, ip, XFS_DATA_FORK, new_size);
-	if (error)
-		return error;
-
-	/*
-	 * If we are not changing the file size then do not update the on-disk
-	 * file size - we may be called from xfs_inactive_free_eofblocks().
-	 * If we update the on-disk file size and then the system crashes
-	 * before the contents of the file are flushed to disk then the files
-	 * may be full of holes (ie NULL files bug).
-	 */
-	xfs_isize_check(ip, new_size);
-	if (ip->i_size != new_size) {
-		ip->i_d.di_size = new_size;
-		ip->i_size = new_size;
-	}
-
-	ASSERT(new_size != 0 || ip->i_delayed_blks == 0);
-	ASSERT(new_size != 0 || ip->i_d.di_nextents == 0);
-
-	/*
-	 * Always re-log the inode so that our permanent transaction can keep
-	 * on rolling it forward in the log.
-	 */
-	xfs_trans_log_inode(*tpp, ip, XFS_ILOG_CORE);
-
-	trace_xfs_itruncate_data_end(ip, new_size);
-	return 0;
-}
-
 /*
  * This is called when the inode's link count goes to 0.
  * We place the on-disk inode on a list in the AGI.  It
@@ -1457,7 +1355,7 @@ xfs_iunlink(
 		 * Here we put the head pointer into our next pointer,
 		 * and then we fall through to point the head at us.
 		 */
-		error = xfs_itobp(mp, tp, ip, &dip, &ibp, XBF_LOCK);
+		error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0);
 		if (error)
 			return error;
 
@@ -1538,7 +1436,7 @@ xfs_iunlink_remove(
 		 * of dealing with the buffer when there is no need to
 		 * change it.
 		 */
-		error = xfs_itobp(mp, tp, ip, &dip, &ibp, XBF_LOCK);
+		error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0);
 		if (error) {
 			xfs_warn(mp, "%s: xfs_itobp() returned error %d.",
 				__func__, error);
@@ -1599,7 +1497,7 @@ xfs_iunlink_remove(
 		 * Now last_ibp points to the buffer previous to us on
 		 * the unlinked list.  Pull us from the list.
 		 */
-		error = xfs_itobp(mp, tp, ip, &dip, &ibp, XBF_LOCK);
+		error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0);
 		if (error) {
 			xfs_warn(mp, "%s: xfs_itobp(2) returned error %d.",
 				__func__, error);
@@ -1681,8 +1579,7 @@ xfs_ifree_cluster(
 		 * to mark all the active inodes on the buffer stale.
 		 */
 		bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, blkno,
-					mp->m_bsize * blks_per_cluster,
-					XBF_LOCK);
+					mp->m_bsize * blks_per_cluster, 0);
 
 		if (!bp)
 			return ENOMEM;
@@ -1771,14 +1668,13 @@ retry:
 			iip = ip->i_itemp;
 			if (!iip || xfs_inode_clean(ip)) {
 				ASSERT(ip != free_ip);
-				ip->i_update_core = 0;
 				xfs_ifunlock(ip);
 				xfs_iunlock(ip, XFS_ILOCK_EXCL);
 				continue;
 			}
 
-			iip->ili_last_fields = iip->ili_format.ilf_fields;
-			iip->ili_format.ilf_fields = 0;
+			iip->ili_last_fields = iip->ili_fields;
+			iip->ili_fields = 0;
 			iip->ili_logged = 1;
 			xfs_trans_ail_copy_lsn(mp->m_ail, &iip->ili_flush_lsn,
 						&iip->ili_item.li_lsn);
@@ -1824,8 +1720,7 @@ xfs_ifree(
 	ASSERT(ip->i_d.di_nlink == 0);
 	ASSERT(ip->i_d.di_nextents == 0);
 	ASSERT(ip->i_d.di_anextents == 0);
-	ASSERT((ip->i_d.di_size == 0 && ip->i_size == 0) ||
-	       (!S_ISREG(ip->i_d.di_mode)));
+	ASSERT(ip->i_d.di_size == 0 || !S_ISREG(ip->i_d.di_mode));
 	ASSERT(ip->i_d.di_nblocks == 0);
 
 	/*
@@ -1844,8 +1739,6 @@ xfs_ifree(
 	ip->i_d.di_flags = 0;
 	ip->i_d.di_dmevmask = 0;
 	ip->i_d.di_forkoff = 0;		/* mark the attr fork not in use */
-	ip->i_df.if_ext_max =
-		XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t);
 	ip->i_d.di_format = XFS_DINODE_FMT_EXTENTS;
 	ip->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS;
 	/*
@@ -1856,7 +1749,7 @@ xfs_ifree(
 
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 
-	error = xfs_itobp(ip->i_mount, tp, ip, &dip, &ibp, XBF_LOCK);
+	error = xfs_itobp(ip->i_mount, tp, ip, &dip, &ibp, 0);
 	if (error)
 		return error;
 
@@ -2151,7 +2044,7 @@ xfs_idestroy_fork(
  * once someone is waiting for it to be unpinned.
  */
 static void
-xfs_iunpin_nowait(
+xfs_iunpin(
 	struct xfs_inode	*ip)
 {
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
@@ -2163,14 +2056,29 @@ xfs_iunpin_nowait(
 
 }
 
+static void
+__xfs_iunpin_wait(
+	struct xfs_inode	*ip)
+{
+	wait_queue_head_t *wq = bit_waitqueue(&ip->i_flags, __XFS_IPINNED_BIT);
+	DEFINE_WAIT_BIT(wait, &ip->i_flags, __XFS_IPINNED_BIT);
+
+	xfs_iunpin(ip);
+
+	do {
+		prepare_to_wait(wq, &wait.wait, TASK_UNINTERRUPTIBLE);
+		if (xfs_ipincount(ip))
+			io_schedule();
+	} while (xfs_ipincount(ip));
+	finish_wait(wq, &wait.wait);
+}
+
 void
 xfs_iunpin_wait(
 	struct xfs_inode	*ip)
 {
-	if (xfs_ipincount(ip)) {
-		xfs_iunpin_nowait(ip);
-		wait_event(ip->i_ipin_wait, (xfs_ipincount(ip) == 0));
-	}
+	if (xfs_ipincount(ip))
+		__xfs_iunpin_wait(ip);
 }
 
 /*
@@ -2280,7 +2188,7 @@ xfs_iflush_fork(
 	mp = ip->i_mount;
 	switch (XFS_IFORK_FORMAT(ip, whichfork)) {
 	case XFS_DINODE_FMT_LOCAL:
-		if ((iip->ili_format.ilf_fields & dataflag[whichfork]) &&
+		if ((iip->ili_fields & dataflag[whichfork]) &&
 		    (ifp->if_bytes > 0)) {
 			ASSERT(ifp->if_u1.if_data != NULL);
 			ASSERT(ifp->if_bytes <= XFS_IFORK_SIZE(ip, whichfork));
@@ -2290,8 +2198,8 @@ xfs_iflush_fork(
 
 	case XFS_DINODE_FMT_EXTENTS:
 		ASSERT((ifp->if_flags & XFS_IFEXTENTS) ||
-		       !(iip->ili_format.ilf_fields & extflag[whichfork]));
-		if ((iip->ili_format.ilf_fields & extflag[whichfork]) &&
+		       !(iip->ili_fields & extflag[whichfork]));
+		if ((iip->ili_fields & extflag[whichfork]) &&
 		    (ifp->if_bytes > 0)) {
 			ASSERT(xfs_iext_get_ext(ifp, 0));
 			ASSERT(XFS_IFORK_NEXTENTS(ip, whichfork) > 0);
@@ -2301,7 +2209,7 @@ xfs_iflush_fork(
 		break;
 
 	case XFS_DINODE_FMT_BTREE:
-		if ((iip->ili_format.ilf_fields & brootflag[whichfork]) &&
+		if ((iip->ili_fields & brootflag[whichfork]) &&
 		    (ifp->if_broot_bytes > 0)) {
 			ASSERT(ifp->if_broot != NULL);
 			ASSERT(ifp->if_broot_bytes <=
@@ -2314,14 +2222,14 @@ xfs_iflush_fork(
 		break;
 
 	case XFS_DINODE_FMT_DEV:
-		if (iip->ili_format.ilf_fields & XFS_ILOG_DEV) {
+		if (iip->ili_fields & XFS_ILOG_DEV) {
 			ASSERT(whichfork == XFS_DATA_FORK);
 			xfs_dinode_put_rdev(dip, ip->i_df.if_u2.if_rdev);
 		}
 		break;
 
 	case XFS_DINODE_FMT_UUID:
-		if (iip->ili_format.ilf_fields & XFS_ILOG_UUID) {
+		if (iip->ili_fields & XFS_ILOG_UUID) {
 			ASSERT(whichfork == XFS_DATA_FORK);
 			memcpy(XFS_DFORK_DPTR(dip),
 			       &ip->i_df.if_u2.if_uuid,
@@ -2451,11 +2359,11 @@ cluster_corrupt_out:
 	 */
 	rcu_read_unlock();
 	/*
-	 * Clean up the buffer.  If it was B_DELWRI, just release it --
+	 * Clean up the buffer.  If it was delwri, just release it --
 	 * brelse can handle it with no problems.  If not, shut down the
 	 * filesystem before releasing the buffer.
 	 */
-	bufwasdelwri = XFS_BUF_ISDELAYWRITE(bp);
+	bufwasdelwri = (bp->b_flags & _XBF_DELWRI_Q);
 	if (bufwasdelwri)
 		xfs_buf_relse(bp);
 
@@ -2481,58 +2389,40 @@ cluster_corrupt_out:
 	/*
 	 * Unlocks the flush lock
 	 */
-	xfs_iflush_abort(iq);
+	xfs_iflush_abort(iq, false);
 	kmem_free(ilist);
 	xfs_perag_put(pag);
 	return XFS_ERROR(EFSCORRUPTED);
 }
 
 /*
- * xfs_iflush() will write a modified inode's changes out to the
- * inode's on disk home.  The caller must have the inode lock held
- * in at least shared mode and the inode flush completion must be
- * active as well.  The inode lock will still be held upon return from
- * the call and the caller is free to unlock it.
- * The inode flush will be completed when the inode reaches the disk.
- * The flags indicate how the inode's buffer should be written out.
+ * Flush dirty inode metadata into the backing buffer.
+ *
+ * The caller must have the inode lock and the inode flush lock held.  The
+ * inode lock will still be held upon return to the caller, and the inode
+ * flush lock will be released after the inode has reached the disk.
+ *
+ * The caller must write out the buffer returned in *bpp and release it.
  */
 int
 xfs_iflush(
-	xfs_inode_t		*ip,
-	uint			flags)
+	struct xfs_inode	*ip,
+	struct xfs_buf		**bpp)
 {
-	xfs_inode_log_item_t	*iip;
-	xfs_buf_t		*bp;
-	xfs_dinode_t		*dip;
-	xfs_mount_t		*mp;
+	struct xfs_mount	*mp = ip->i_mount;
+	struct xfs_buf		*bp;
+	struct xfs_dinode	*dip;
 	int			error;
 
 	XFS_STATS_INC(xs_iflush_count);
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
-	ASSERT(!completion_done(&ip->i_flush));
+	ASSERT(xfs_isiflocked(ip));
 	ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
-	       ip->i_d.di_nextents > ip->i_df.if_ext_max);
+	       ip->i_d.di_nextents > XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK));
 
-	iip = ip->i_itemp;
-	mp = ip->i_mount;
+	*bpp = NULL;
 
-	/*
-	 * We can't flush the inode until it is unpinned, so wait for it if we
-	 * are allowed to block.  We know no one new can pin it, because we are
-	 * holding the inode lock shared and you need to hold it exclusively to
-	 * pin the inode.
-	 *
-	 * If we are not allowed to block, force the log out asynchronously so
-	 * that when we come back the inode will be unpinned. If other inodes
-	 * in the same cluster are dirty, they will probably write the inode
-	 * out for us if they occur after the log force completes.
-	 */
-	if (!(flags & SYNC_WAIT) && xfs_ipincount(ip)) {
-		xfs_iunpin_nowait(ip);
-		xfs_ifunlock(ip);
-		return EAGAIN;
-	}
 	xfs_iunpin_wait(ip);
 
 	/*
@@ -2551,21 +2441,20 @@ xfs_iflush(
 	/*
 	 * This may have been unpinned because the filesystem is shutting
 	 * down forcibly. If that's the case we must not write this inode
-	 * to disk, because the log record didn't make it to disk!
+	 * to disk, because the log record didn't make it to disk.
+	 *
+	 * We also have to remove the log item from the AIL in this case,
+	 * as we wait for an empty AIL as part of the unmount process.
 	 */
 	if (XFS_FORCED_SHUTDOWN(mp)) {
-		ip->i_update_core = 0;
-		if (iip)
-			iip->ili_format.ilf_fields = 0;
-		xfs_ifunlock(ip);
-		return XFS_ERROR(EIO);
+		error = XFS_ERROR(EIO);
+		goto abort_out;
 	}
 
 	/*
 	 * Get the buffer containing the on-disk inode.
 	 */
-	error = xfs_itobp(mp, NULL, ip, &dip, &bp,
-				(flags & SYNC_TRYLOCK) ? XBF_TRYLOCK : XBF_LOCK);
+	error = xfs_itobp(mp, NULL, ip, &dip, &bp, XBF_TRYLOCK);
 	if (error || !bp) {
 		xfs_ifunlock(ip);
 		return error;
@@ -2593,23 +2482,20 @@ xfs_iflush(
 	if (error)
 		goto cluster_corrupt_out;
 
-	if (flags & SYNC_WAIT)
-		error = xfs_bwrite(bp);
-	else
-		xfs_buf_delwri_queue(bp);
-
-	xfs_buf_relse(bp);
-	return error;
+	*bpp = bp;
+	return 0;
 
 corrupt_out:
 	xfs_buf_relse(bp);
 	xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
 cluster_corrupt_out:
+	error = XFS_ERROR(EFSCORRUPTED);
+abort_out:
 	/*
 	 * Unlocks the flush lock
 	 */
-	xfs_iflush_abort(ip);
-	return XFS_ERROR(EFSCORRUPTED);
+	xfs_iflush_abort(ip, false);
+	return error;
 }
 
 
@@ -2626,9 +2512,9 @@ xfs_iflush_int(
 #endif
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
-	ASSERT(!completion_done(&ip->i_flush));
+	ASSERT(xfs_isiflocked(ip));
 	ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
-	       ip->i_d.di_nextents > ip->i_df.if_ext_max);
+	       ip->i_d.di_nextents > XFS_IFORK_MAXEXT(ip, XFS_DATA_FORK));
 
 	iip = ip->i_itemp;
 	mp = ip->i_mount;
@@ -2636,26 +2522,6 @@ xfs_iflush_int(
 	/* set *dip = inode's place in the buffer */
 	dip = (xfs_dinode_t *)xfs_buf_offset(bp, ip->i_imap.im_boffset);
 
-	/*
-	 * Clear i_update_core before copying out the data.
-	 * This is for coordination with our timestamp updates
-	 * that don't hold the inode lock. They will always
-	 * update the timestamps BEFORE setting i_update_core,
-	 * so if we clear i_update_core after they set it we
-	 * are guaranteed to see their updates to the timestamps.
-	 * I believe that this depends on strongly ordered memory
-	 * semantics, but we have that.  We use the SYNCHRONIZE
-	 * macro to make sure that the compiler does not reorder
-	 * the i_update_core access below the data copy below.
-	 */
-	ip->i_update_core = 0;
-	SYNCHRONIZE();
-
-	/*
-	 * Make sure to get the latest timestamps from the Linux inode.
-	 */
-	xfs_synchronize_times(ip);
-
 	if (XFS_TEST_ERROR(dip->di_magic != cpu_to_be16(XFS_DINODE_MAGIC),
 			       mp, XFS_ERRTAG_IFLUSH_1, XFS_RANDOM_IFLUSH_1)) {
 		xfs_alert_tag(mp, XFS_PTAG_IFLUSH,
@@ -2766,36 +2632,33 @@ xfs_iflush_int(
 	xfs_inobp_check(mp, bp);
 
 	/*
-	 * We've recorded everything logged in the inode, so we'd
-	 * like to clear the ilf_fields bits so we don't log and
-	 * flush things unnecessarily.  However, we can't stop
-	 * logging all this information until the data we've copied
-	 * into the disk buffer is written to disk.  If we did we might
-	 * overwrite the copy of the inode in the log with all the
-	 * data after re-logging only part of it, and in the face of
-	 * a crash we wouldn't have all the data we need to recover.
+	 * We've recorded everything logged in the inode, so we'd like to clear
+	 * the ili_fields bits so we don't log and flush things unnecessarily.
+	 * However, we can't stop logging all this information until the data
+	 * we've copied into the disk buffer is written to disk.  If we did we
+	 * might overwrite the copy of the inode in the log with all the data
+	 * after re-logging only part of it, and in the face of a crash we
+	 * wouldn't have all the data we need to recover.
 	 *
-	 * What we do is move the bits to the ili_last_fields field.
-	 * When logging the inode, these bits are moved back to the
-	 * ilf_fields field.  In the xfs_iflush_done() routine we
-	 * clear ili_last_fields, since we know that the information
-	 * those bits represent is permanently on disk.  As long as
-	 * the flush completes before the inode is logged again, then
-	 * both ilf_fields and ili_last_fields will be cleared.
+	 * What we do is move the bits to the ili_last_fields field.  When
+	 * logging the inode, these bits are moved back to the ili_fields field.
+	 * In the xfs_iflush_done() routine we clear ili_last_fields, since we
+	 * know that the information those bits represent is permanently on
+	 * disk.  As long as the flush completes before the inode is logged
+	 * again, then both ili_fields and ili_last_fields will be cleared.
 	 *
-	 * We can play with the ilf_fields bits here, because the inode
-	 * lock must be held exclusively in order to set bits there
-	 * and the flush lock protects the ili_last_fields bits.
-	 * Set ili_logged so the flush done
-	 * routine can tell whether or not to look in the AIL.
-	 * Also, store the current LSN of the inode so that we can tell
-	 * whether the item has moved in the AIL from xfs_iflush_done().
-	 * In order to read the lsn we need the AIL lock, because
-	 * it is a 64 bit value that cannot be read atomically.
-	 */
-	if (iip != NULL && iip->ili_format.ilf_fields != 0) {
-		iip->ili_last_fields = iip->ili_format.ilf_fields;
-		iip->ili_format.ilf_fields = 0;
+	 * We can play with the ili_fields bits here, because the inode lock
+	 * must be held exclusively in order to set bits there and the flush
+	 * lock protects the ili_last_fields bits.  Set ili_logged so the flush
+	 * done routine can tell whether or not to look in the AIL.  Also, store
+	 * the current LSN of the inode so that we can tell whether the item has
+	 * moved in the AIL from xfs_iflush_done().  In order to read the lsn we
+	 * need the AIL lock, because it is a 64 bit value that cannot be read
+	 * atomically.
+	 */
+	if (iip != NULL && iip->ili_fields != 0) {
+		iip->ili_last_fields = iip->ili_fields;
+		iip->ili_fields = 0;
 		iip->ili_logged = 1;
 
 		xfs_trans_ail_copy_lsn(mp->m_ail, &iip->ili_flush_lsn,
@@ -2814,8 +2677,7 @@ xfs_iflush_int(
 	} else {
 		/*
 		 * We're flushing an inode which is not in the AIL and has
-		 * not been logged but has i_update_core set.  For this
-		 * case we can use a B_DELWRI flush and immediately drop
+		 * not been logged.  For this case we can immediately drop
 		 * the inode flush lock because we can avoid the whole
 		 * AIL state thing.  It's OK to drop the flush lock now,
 		 * because we've already locked the buffer and to do anything
@@ -2835,27 +2697,6 @@ corrupt_out:
 	return XFS_ERROR(EFSCORRUPTED);
 }
 
-void
-xfs_promote_inode(
-	struct xfs_inode	*ip)
-{
-	struct xfs_buf		*bp;
-
-	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
-
-	bp = xfs_incore(ip->i_mount->m_ddev_targp, ip->i_imap.im_blkno,
-			ip->i_imap.im_len, XBF_TRYLOCK);
-	if (!bp)
-		return;
-
-	if (XFS_BUF_ISDELAYWRITE(bp)) {
-		xfs_buf_delwri_promote(bp);
-		wake_up_process(ip->i_mount->m_ddev_targp->bt_task);
-	}
-
-	xfs_buf_relse(bp);
-}
-
 /*
  * Return a pointer to the extent record at file index idx.
  */
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index b4cd4739f98..1efff36a75b 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -66,7 +66,6 @@ typedef struct xfs_ifork {
 	struct xfs_btree_block	*if_broot;	/* file's incore btree root */
 	short			if_broot_bytes;	/* bytes allocated for root */
 	unsigned char		if_flags;	/* per-fork flags */
-	unsigned char		if_ext_max;	/* max # of extent records */
 	union {
 		xfs_bmbt_rec_host_t *if_extents;/* linear map file exts */
 		xfs_ext_irec_t	*if_ext_irec;	/* irec map file exts */
@@ -206,12 +205,12 @@ typedef struct xfs_icdinode {
 	((w) == XFS_DATA_FORK ? \
 		((ip)->i_d.di_nextents = (n)) : \
 		((ip)->i_d.di_anextents = (n)))
-
+#define XFS_IFORK_MAXEXT(ip, w) \
+	(XFS_IFORK_SIZE(ip, w) / sizeof(xfs_bmbt_rec_t))
 
 
 #ifdef __KERNEL__
 
-struct bhv_desc;
 struct xfs_buf;
 struct xfs_bmap_free;
 struct xfs_bmbt_irec;
@@ -220,12 +219,6 @@ struct xfs_mount;
 struct xfs_trans;
 struct xfs_dquot;
 
-typedef struct dm_attrs_s {
-	__uint32_t	da_dmevmask;	/* DMIG event mask */
-	__uint16_t	da_dmstate;	/* DMIG state info */
-	__uint16_t	da_pad;		/* DMIG extra padding */
-} dm_attrs_t;
-
 typedef struct xfs_inode {
 	/* Inode linking and identification information. */
 	struct xfs_mount	*i_mount;	/* fs mount struct ptr */
@@ -244,27 +237,18 @@ typedef struct xfs_inode {
 	struct xfs_inode_log_item *i_itemp;	/* logging information */
 	mrlock_t		i_lock;		/* inode lock */
 	mrlock_t		i_iolock;	/* inode IO lock */
-	struct completion	i_flush;	/* inode flush completion q */
 	atomic_t		i_pincount;	/* inode pin count */
-	wait_queue_head_t	i_ipin_wait;	/* inode pinning wait queue */
 	spinlock_t		i_flags_lock;	/* inode i_flags lock */
 	/* Miscellaneous state. */
-	unsigned short		i_flags;	/* see defined flags below */
-	unsigned char		i_update_core;	/* timestamps/size is dirty */
+	unsigned long		i_flags;	/* see defined flags below */
 	unsigned int		i_delayed_blks;	/* count of delay alloc blks */
 
 	xfs_icdinode_t		i_d;		/* most of ondisk inode */
 
-	xfs_fsize_t		i_size;		/* in-memory size */
-	xfs_fsize_t		i_new_size;	/* size when write completes */
-
 	/* VFS inode */
 	struct inode		i_vnode;	/* embedded VFS inode */
 } xfs_inode_t;
 
-#define XFS_ISIZE(ip)	S_ISREG((ip)->i_d.di_mode) ? \
-				(ip)->i_size : (ip)->i_d.di_size;
-
 /* Convert from vfs inode to xfs inode */
 static inline struct xfs_inode *XFS_I(struct inode *inode)
 {
@@ -278,6 +262,32 @@ static inline struct inode *VFS_I(struct xfs_inode *ip)
 }
 
 /*
+ * For regular files we only update the on-disk filesize when actually
+ * writing data back to disk.  Until then only the copy in the VFS inode
+ * is uptodate.
+ */
+static inline xfs_fsize_t XFS_ISIZE(struct xfs_inode *ip)
+{
+	if (S_ISREG(ip->i_d.di_mode))
+		return i_size_read(VFS_I(ip));
+	return ip->i_d.di_size;
+}
+
+/*
+ * If this I/O goes past the on-disk inode size update it unless it would
+ * be past the current in-core inode size.
+ */
+static inline xfs_fsize_t
+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)
+		new_size = i_size;
+	return new_size > ip->i_d.di_size ? new_size : 0;
+}
+
+/*
  * i_flags helper functions
  */
 static inline void
@@ -331,6 +341,19 @@ xfs_iflags_test_and_clear(xfs_inode_t *ip, unsigned short flags)
 	return ret;
 }
 
+static inline int
+xfs_iflags_test_and_set(xfs_inode_t *ip, unsigned short flags)
+{
+	int ret;
+
+	spin_lock(&ip->i_flags_lock);
+	ret = ip->i_flags & flags;
+	if (!ret)
+		ip->i_flags |= flags;
+	spin_unlock(&ip->i_flags_lock);
+	return ret;
+}
+
 /*
  * Project quota id helpers (previously projid was 16bit only
  * and using two 16bit values to hold new 32bit projid was chosen
@@ -351,39 +374,24 @@ xfs_set_projid(struct xfs_inode *ip,
 }
 
 /*
- * Manage the i_flush queue embedded in the inode.  This completion
- * queue synchronizes processes attempting to flush the in-core
- * inode back to disk.
- */
-static inline void xfs_iflock(xfs_inode_t *ip)
-{
-	wait_for_completion(&ip->i_flush);
-}
-
-static inline int xfs_iflock_nowait(xfs_inode_t *ip)
-{
-	return try_wait_for_completion(&ip->i_flush);
-}
-
-static inline void xfs_ifunlock(xfs_inode_t *ip)
-{
-	complete(&ip->i_flush);
-}
-
-/*
  * In-core inode flags.
  */
-#define XFS_IRECLAIM		0x0001  /* started reclaiming this inode */
-#define XFS_ISTALE		0x0002	/* inode has been staled */
-#define XFS_IRECLAIMABLE	0x0004	/* inode can be reclaimed */
-#define XFS_INEW		0x0008	/* inode has just been allocated */
-#define XFS_IFILESTREAM		0x0010	/* inode is in a filestream directory */
-#define XFS_ITRUNCATED		0x0020	/* truncated down so flush-on-close */
-#define XFS_IDIRTY_RELEASE	0x0040	/* dirty release already seen */
+#define XFS_IRECLAIM		(1 << 0) /* started reclaiming this inode */
+#define XFS_ISTALE		(1 << 1) /* inode has been staled */
+#define XFS_IRECLAIMABLE	(1 << 2) /* inode can be reclaimed */
+#define XFS_INEW		(1 << 3) /* inode has just been allocated */
+#define XFS_IFILESTREAM		(1 << 4) /* inode is in a filestream dir. */
+#define XFS_ITRUNCATED		(1 << 5) /* truncated down so flush-on-close */
+#define XFS_IDIRTY_RELEASE	(1 << 6) /* dirty release already seen */
+#define __XFS_IFLOCK_BIT	7	 /* inode is being flushed right now */
+#define XFS_IFLOCK		(1 << __XFS_IFLOCK_BIT)
+#define __XFS_IPINNED_BIT	8	 /* wakeup key for zero pin count */
+#define XFS_IPINNED		(1 << __XFS_IPINNED_BIT)
+#define XFS_IDONTCACHE		(1 << 9) /* don't cache the inode long term */
 
 /*
  * Per-lifetime flags need to be reset when re-using a reclaimable inode during
- * inode lookup. Thi prevents unintended behaviour on the new inode from
+ * inode lookup. This prevents unintended behaviour on the new inode from
  * ocurring.
  */
 #define XFS_IRECLAIM_RESET_FLAGS	\
@@ -392,6 +400,34 @@ static inline void xfs_ifunlock(xfs_inode_t *ip)
 	 XFS_IFILESTREAM);
 
 /*
+ * Synchronize processes attempting to flush the in-core inode back to disk.
+ */
+
+extern void __xfs_iflock(struct xfs_inode *ip);
+
+static inline int xfs_iflock_nowait(struct xfs_inode *ip)
+{
+	return !xfs_iflags_test_and_set(ip, XFS_IFLOCK);
+}
+
+static inline void xfs_iflock(struct xfs_inode *ip)
+{
+	if (!xfs_iflock_nowait(ip))
+		__xfs_iflock(ip);
+}
+
+static inline void xfs_ifunlock(struct xfs_inode *ip)
+{
+	xfs_iflags_clear(ip, XFS_IFLOCK);
+	wake_up_bit(&ip->i_flags, __XFS_IFLOCK_BIT);
+}
+
+static inline int xfs_isiflocked(struct xfs_inode *ip)
+{
+	return xfs_iflags_test(ip, XFS_IFLOCK);
+}
+
+/*
  * Flags for inode locking.
  * Bit ranges:	1<<1  - 1<<16-1 -- iolock/ilock modes (bitfield)
  *		1<<16 - 1<<32-1 -- lockdep annotation (integers)
@@ -400,7 +436,6 @@ static inline void xfs_ifunlock(xfs_inode_t *ip)
 #define	XFS_IOLOCK_SHARED	(1<<1)
 #define	XFS_ILOCK_EXCL		(1<<2)
 #define	XFS_ILOCK_SHARED	(1<<3)
-#define	XFS_IUNLOCK_NONOTIFY	(1<<4)
 
 #define XFS_LOCK_MASK		(XFS_IOLOCK_EXCL | XFS_IOLOCK_SHARED \
 				| XFS_ILOCK_EXCL | XFS_ILOCK_SHARED)
@@ -409,8 +444,7 @@ static inline void xfs_ifunlock(xfs_inode_t *ip)
 	{ XFS_IOLOCK_EXCL,	"IOLOCK_EXCL" }, \
 	{ XFS_IOLOCK_SHARED,	"IOLOCK_SHARED" }, \
 	{ XFS_ILOCK_EXCL,	"ILOCK_EXCL" }, \
-	{ XFS_ILOCK_SHARED,	"ILOCK_SHARED" }, \
-	{ XFS_IUNLOCK_NONOTIFY,	"IUNLOCK_NONOTIFY" }
+	{ XFS_ILOCK_SHARED,	"ILOCK_SHARED" }
 
 
 /*
@@ -481,7 +515,7 @@ void		xfs_inode_free(struct xfs_inode *ip);
 /*
  * xfs_inode.c prototypes.
  */
-int		xfs_ialloc(struct xfs_trans *, xfs_inode_t *, mode_t,
+int		xfs_ialloc(struct xfs_trans *, xfs_inode_t *, umode_t,
 			   xfs_nlink_t, xfs_dev_t, prid_t, int,
 			   struct xfs_buf **, boolean_t *, xfs_inode_t **);
 
@@ -491,20 +525,15 @@ int		xfs_ifree(struct xfs_trans *, xfs_inode_t *,
 			   struct xfs_bmap_free *);
 int		xfs_itruncate_extents(struct xfs_trans **, struct xfs_inode *,
 				      int, xfs_fsize_t);
-int		xfs_itruncate_data(struct xfs_trans **, struct xfs_inode *,
-				   xfs_fsize_t);
 int		xfs_iunlink(struct xfs_trans *, xfs_inode_t *);
 
 void		xfs_iext_realloc(xfs_inode_t *, int, int);
 void		xfs_iunpin_wait(xfs_inode_t *);
-int		xfs_iflush(xfs_inode_t *, uint);
-void		xfs_promote_inode(struct xfs_inode *);
+int		xfs_iflush(struct xfs_inode *, struct xfs_buf **);
 void		xfs_lock_inodes(xfs_inode_t **, int, uint);
 void		xfs_lock_two_inodes(xfs_inode_t *, xfs_inode_t *, uint);
 
-void		xfs_synchronize_times(xfs_inode_t *);
-void		xfs_mark_inode_dirty(xfs_inode_t *);
-void		xfs_mark_inode_dirty_sync(xfs_inode_t *);
+xfs_extlen_t	xfs_get_extsz_hint(struct xfs_inode *ip);
 
 #define IHOLD(ip) \
 do { \
@@ -526,6 +555,7 @@ do { \
  */
 #define XFS_IGET_CREATE		0x1
 #define XFS_IGET_UNTRUSTED	0x2
+#define XFS_IGET_DONTCACHE	0x4
 
 int		xfs_inotobp(struct xfs_mount *, struct xfs_trans *,
 			    xfs_ino_t, struct xfs_dinode **,
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index abaafdbb3e6..d041d47d9d8 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -18,9 +18,7 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_types.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -57,79 +55,28 @@ xfs_inode_item_size(
 	struct xfs_inode	*ip = iip->ili_inode;
 	uint			nvecs = 2;
 
-	/*
-	 * Only log the data/extents/b-tree root if there is something
-	 * left to log.
-	 */
-	iip->ili_format.ilf_fields |= XFS_ILOG_CORE;
-
 	switch (ip->i_d.di_format) {
 	case XFS_DINODE_FMT_EXTENTS:
-		iip->ili_format.ilf_fields &=
-			~(XFS_ILOG_DDATA | XFS_ILOG_DBROOT |
-			  XFS_ILOG_DEV | XFS_ILOG_UUID);
-		if ((iip->ili_format.ilf_fields & XFS_ILOG_DEXT) &&
-		    (ip->i_d.di_nextents > 0) &&
-		    (ip->i_df.if_bytes > 0)) {
-			ASSERT(ip->i_df.if_u1.if_extents != NULL);
+		if ((iip->ili_fields & XFS_ILOG_DEXT) &&
+		    ip->i_d.di_nextents > 0 &&
+		    ip->i_df.if_bytes > 0)
 			nvecs++;
-		} else {
-			iip->ili_format.ilf_fields &= ~XFS_ILOG_DEXT;
-		}
 		break;
 
 	case XFS_DINODE_FMT_BTREE:
-		ASSERT(ip->i_df.if_ext_max ==
-		       XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t));
-		iip->ili_format.ilf_fields &=
-			~(XFS_ILOG_DDATA | XFS_ILOG_DEXT |
-			  XFS_ILOG_DEV | XFS_ILOG_UUID);
-		if ((iip->ili_format.ilf_fields & XFS_ILOG_DBROOT) &&
-		    (ip->i_df.if_broot_bytes > 0)) {
-			ASSERT(ip->i_df.if_broot != NULL);
+		if ((iip->ili_fields & XFS_ILOG_DBROOT) &&
+		    ip->i_df.if_broot_bytes > 0)
 			nvecs++;
-		} else {
-			ASSERT(!(iip->ili_format.ilf_fields &
-				 XFS_ILOG_DBROOT));
-#ifdef XFS_TRANS_DEBUG
-			if (iip->ili_root_size > 0) {
-				ASSERT(iip->ili_root_size ==
-				       ip->i_df.if_broot_bytes);
-				ASSERT(memcmp(iip->ili_orig_root,
-					    ip->i_df.if_broot,
-					    iip->ili_root_size) == 0);
-			} else {
-				ASSERT(ip->i_df.if_broot_bytes == 0);
-			}
-#endif
-			iip->ili_format.ilf_fields &= ~XFS_ILOG_DBROOT;
-		}
 		break;
 
 	case XFS_DINODE_FMT_LOCAL:
-		iip->ili_format.ilf_fields &=
-			~(XFS_ILOG_DEXT | XFS_ILOG_DBROOT |
-			  XFS_ILOG_DEV | XFS_ILOG_UUID);
-		if ((iip->ili_format.ilf_fields & XFS_ILOG_DDATA) &&
-		    (ip->i_df.if_bytes > 0)) {
-			ASSERT(ip->i_df.if_u1.if_data != NULL);
-			ASSERT(ip->i_d.di_size > 0);
+		if ((iip->ili_fields & XFS_ILOG_DDATA) &&
+		    ip->i_df.if_bytes > 0)
 			nvecs++;
-		} else {
-			iip->ili_format.ilf_fields &= ~XFS_ILOG_DDATA;
-		}
 		break;
 
 	case XFS_DINODE_FMT_DEV:
-		iip->ili_format.ilf_fields &=
-			~(XFS_ILOG_DDATA | XFS_ILOG_DBROOT |
-			  XFS_ILOG_DEXT | XFS_ILOG_UUID);
-		break;
-
 	case XFS_DINODE_FMT_UUID:
-		iip->ili_format.ilf_fields &=
-			~(XFS_ILOG_DDATA | XFS_ILOG_DBROOT |
-			  XFS_ILOG_DEXT | XFS_ILOG_DEV);
 		break;
 
 	default:
@@ -137,56 +84,31 @@ xfs_inode_item_size(
 		break;
 	}
 
-	/*
-	 * If there are no attributes associated with this file,
-	 * then there cannot be anything more to log.
-	 * Clear all attribute-related log flags.
-	 */
-	if (!XFS_IFORK_Q(ip)) {
-		iip->ili_format.ilf_fields &=
-			~(XFS_ILOG_ADATA | XFS_ILOG_ABROOT | XFS_ILOG_AEXT);
+	if (!XFS_IFORK_Q(ip))
 		return nvecs;
-	}
+
 
 	/*
 	 * Log any necessary attribute data.
 	 */
 	switch (ip->i_d.di_aformat) {
 	case XFS_DINODE_FMT_EXTENTS:
-		iip->ili_format.ilf_fields &=
-			~(XFS_ILOG_ADATA | XFS_ILOG_ABROOT);
-		if ((iip->ili_format.ilf_fields & XFS_ILOG_AEXT) &&
-		    (ip->i_d.di_anextents > 0) &&
-		    (ip->i_afp->if_bytes > 0)) {
-			ASSERT(ip->i_afp->if_u1.if_extents != NULL);
+		if ((iip->ili_fields & XFS_ILOG_AEXT) &&
+		    ip->i_d.di_anextents > 0 &&
+		    ip->i_afp->if_bytes > 0)
 			nvecs++;
-		} else {
-			iip->ili_format.ilf_fields &= ~XFS_ILOG_AEXT;
-		}
 		break;
 
 	case XFS_DINODE_FMT_BTREE:
-		iip->ili_format.ilf_fields &=
-			~(XFS_ILOG_ADATA | XFS_ILOG_AEXT);
-		if ((iip->ili_format.ilf_fields & XFS_ILOG_ABROOT) &&
-		    (ip->i_afp->if_broot_bytes > 0)) {
-			ASSERT(ip->i_afp->if_broot != NULL);
+		if ((iip->ili_fields & XFS_ILOG_ABROOT) &&
+		    ip->i_afp->if_broot_bytes > 0)
 			nvecs++;
-		} else {
-			iip->ili_format.ilf_fields &= ~XFS_ILOG_ABROOT;
-		}
 		break;
 
 	case XFS_DINODE_FMT_LOCAL:
-		iip->ili_format.ilf_fields &=
-			~(XFS_ILOG_AEXT | XFS_ILOG_ABROOT);
-		if ((iip->ili_format.ilf_fields & XFS_ILOG_ADATA) &&
-		    (ip->i_afp->if_bytes > 0)) {
-			ASSERT(ip->i_afp->if_u1.if_data != NULL);
+		if ((iip->ili_fields & XFS_ILOG_ADATA) &&
+		    ip->i_afp->if_bytes > 0)
 			nvecs++;
-		} else {
-			iip->ili_format.ilf_fields &= ~XFS_ILOG_ADATA;
-		}
 		break;
 
 	default:
@@ -256,48 +178,11 @@ xfs_inode_item_format(
 	vecp++;
 	nvecs	     = 1;
 
-	/*
-	 * Clear i_update_core if the timestamps (or any other
-	 * non-transactional modification) need flushing/logging
-	 * and we're about to log them with the rest of the core.
-	 *
-	 * This is the same logic as xfs_iflush() but this code can't
-	 * run at the same time as xfs_iflush because we're in commit
-	 * processing here and so we have the inode lock held in
-	 * exclusive mode.  Although it doesn't really matter
-	 * for the timestamps if both routines were to grab the
-	 * timestamps or not.  That would be ok.
-	 *
-	 * We clear i_update_core before copying out the data.
-	 * This is for coordination with our timestamp updates
-	 * that don't hold the inode lock. They will always
-	 * update the timestamps BEFORE setting i_update_core,
-	 * so if we clear i_update_core after they set it we
-	 * are guaranteed to see their updates to the timestamps
-	 * either here.  Likewise, if they set it after we clear it
-	 * here, we'll see it either on the next commit of this
-	 * inode or the next time the inode gets flushed via
-	 * xfs_iflush().  This depends on strongly ordered memory
-	 * semantics, but we have that.  We use the SYNCHRONIZE
-	 * macro to make sure that the compiler does not reorder
-	 * the i_update_core access below the data copy below.
-	 */
-	if (ip->i_update_core)  {
-		ip->i_update_core = 0;
-		SYNCHRONIZE();
-	}
-
-	/*
-	 * Make sure to get the latest timestamps from the Linux inode.
-	 */
-	xfs_synchronize_times(ip);
-
 	vecp->i_addr = &ip->i_d;
 	vecp->i_len  = sizeof(struct xfs_icdinode);
 	vecp->i_type = XLOG_REG_TYPE_ICORE;
 	vecp++;
 	nvecs++;
-	iip->ili_format.ilf_fields |= XFS_ILOG_CORE;
 
 	/*
 	 * If this is really an old format inode, then we need to
@@ -330,16 +215,17 @@ xfs_inode_item_format(
 
 	switch (ip->i_d.di_format) {
 	case XFS_DINODE_FMT_EXTENTS:
-		ASSERT(!(iip->ili_format.ilf_fields &
-			 (XFS_ILOG_DDATA | XFS_ILOG_DBROOT |
-			  XFS_ILOG_DEV | XFS_ILOG_UUID)));
-		if (iip->ili_format.ilf_fields & XFS_ILOG_DEXT) {
-			ASSERT(ip->i_df.if_bytes > 0);
+		iip->ili_fields &=
+			~(XFS_ILOG_DDATA | XFS_ILOG_DBROOT |
+			  XFS_ILOG_DEV | XFS_ILOG_UUID);
+
+		if ((iip->ili_fields & XFS_ILOG_DEXT) &&
+		    ip->i_d.di_nextents > 0 &&
+		    ip->i_df.if_bytes > 0) {
 			ASSERT(ip->i_df.if_u1.if_extents != NULL);
-			ASSERT(ip->i_d.di_nextents > 0);
+			ASSERT(ip->i_df.if_bytes / sizeof(xfs_bmbt_rec_t) > 0);
 			ASSERT(iip->ili_extents_buf == NULL);
-			ASSERT((ip->i_df.if_bytes /
-				(uint)sizeof(xfs_bmbt_rec_t)) > 0);
+
 #ifdef XFS_NATIVE_HOST
                        if (ip->i_d.di_nextents == ip->i_df.if_bytes /
                                                (uint)sizeof(xfs_bmbt_rec_t)) {
@@ -361,15 +247,18 @@ xfs_inode_item_format(
 			iip->ili_format.ilf_dsize = vecp->i_len;
 			vecp++;
 			nvecs++;
+		} else {
+			iip->ili_fields &= ~XFS_ILOG_DEXT;
 		}
 		break;
 
 	case XFS_DINODE_FMT_BTREE:
-		ASSERT(!(iip->ili_format.ilf_fields &
-			 (XFS_ILOG_DDATA | XFS_ILOG_DEXT |
-			  XFS_ILOG_DEV | XFS_ILOG_UUID)));
-		if (iip->ili_format.ilf_fields & XFS_ILOG_DBROOT) {
-			ASSERT(ip->i_df.if_broot_bytes > 0);
+		iip->ili_fields &=
+			~(XFS_ILOG_DDATA | XFS_ILOG_DEXT |
+			  XFS_ILOG_DEV | XFS_ILOG_UUID);
+
+		if ((iip->ili_fields & XFS_ILOG_DBROOT) &&
+		    ip->i_df.if_broot_bytes > 0) {
 			ASSERT(ip->i_df.if_broot != NULL);
 			vecp->i_addr = ip->i_df.if_broot;
 			vecp->i_len = ip->i_df.if_broot_bytes;
@@ -377,15 +266,30 @@ xfs_inode_item_format(
 			vecp++;
 			nvecs++;
 			iip->ili_format.ilf_dsize = ip->i_df.if_broot_bytes;
+		} else {
+			ASSERT(!(iip->ili_fields &
+				 XFS_ILOG_DBROOT));
+#ifdef XFS_TRANS_DEBUG
+			if (iip->ili_root_size > 0) {
+				ASSERT(iip->ili_root_size ==
+				       ip->i_df.if_broot_bytes);
+				ASSERT(memcmp(iip->ili_orig_root,
+					    ip->i_df.if_broot,
+					    iip->ili_root_size) == 0);
+			} else {
+				ASSERT(ip->i_df.if_broot_bytes == 0);
+			}
+#endif
+			iip->ili_fields &= ~XFS_ILOG_DBROOT;
 		}
 		break;
 
 	case XFS_DINODE_FMT_LOCAL:
-		ASSERT(!(iip->ili_format.ilf_fields &
-			 (XFS_ILOG_DBROOT | XFS_ILOG_DEXT |
-			  XFS_ILOG_DEV | XFS_ILOG_UUID)));
-		if (iip->ili_format.ilf_fields & XFS_ILOG_DDATA) {
-			ASSERT(ip->i_df.if_bytes > 0);
+		iip->ili_fields &=
+			~(XFS_ILOG_DEXT | XFS_ILOG_DBROOT |
+			  XFS_ILOG_DEV | XFS_ILOG_UUID);
+		if ((iip->ili_fields & XFS_ILOG_DDATA) &&
+		    ip->i_df.if_bytes > 0) {
 			ASSERT(ip->i_df.if_u1.if_data != NULL);
 			ASSERT(ip->i_d.di_size > 0);
 
@@ -403,24 +307,26 @@ xfs_inode_item_format(
 			vecp++;
 			nvecs++;
 			iip->ili_format.ilf_dsize = (unsigned)data_bytes;
+		} else {
+			iip->ili_fields &= ~XFS_ILOG_DDATA;
 		}
 		break;
 
 	case XFS_DINODE_FMT_DEV:
-		ASSERT(!(iip->ili_format.ilf_fields &
-			 (XFS_ILOG_DBROOT | XFS_ILOG_DEXT |
-			  XFS_ILOG_DDATA | XFS_ILOG_UUID)));
-		if (iip->ili_format.ilf_fields & XFS_ILOG_DEV) {
+		iip->ili_fields &=
+			~(XFS_ILOG_DDATA | XFS_ILOG_DBROOT |
+			  XFS_ILOG_DEXT | XFS_ILOG_UUID);
+		if (iip->ili_fields & XFS_ILOG_DEV) {
 			iip->ili_format.ilf_u.ilfu_rdev =
 				ip->i_df.if_u2.if_rdev;
 		}
 		break;
 
 	case XFS_DINODE_FMT_UUID:
-		ASSERT(!(iip->ili_format.ilf_fields &
-			 (XFS_ILOG_DBROOT | XFS_ILOG_DEXT |
-			  XFS_ILOG_DDATA | XFS_ILOG_DEV)));
-		if (iip->ili_format.ilf_fields & XFS_ILOG_UUID) {
+		iip->ili_fields &=
+			~(XFS_ILOG_DDATA | XFS_ILOG_DBROOT |
+			  XFS_ILOG_DEXT | XFS_ILOG_DEV);
+		if (iip->ili_fields & XFS_ILOG_UUID) {
 			iip->ili_format.ilf_u.ilfu_uuid =
 				ip->i_df.if_u2.if_uuid;
 		}
@@ -432,32 +338,25 @@ xfs_inode_item_format(
 	}
 
 	/*
-	 * If there are no attributes associated with the file,
-	 * then we're done.
-	 * Assert that no attribute-related log flags are set.
+	 * If there are no attributes associated with the file, then we're done.
 	 */
 	if (!XFS_IFORK_Q(ip)) {
-		ASSERT(nvecs == lip->li_desc->lid_size);
-		iip->ili_format.ilf_size = nvecs;
-		ASSERT(!(iip->ili_format.ilf_fields &
-			 (XFS_ILOG_ADATA | XFS_ILOG_ABROOT | XFS_ILOG_AEXT)));
-		return;
+		iip->ili_fields &=
+			~(XFS_ILOG_ADATA | XFS_ILOG_ABROOT | XFS_ILOG_AEXT);
+		goto out;
 	}
 
 	switch (ip->i_d.di_aformat) {
 	case XFS_DINODE_FMT_EXTENTS:
-		ASSERT(!(iip->ili_format.ilf_fields &
-			 (XFS_ILOG_ADATA | XFS_ILOG_ABROOT)));
-		if (iip->ili_format.ilf_fields & XFS_ILOG_AEXT) {
-#ifdef DEBUG
-			int nrecs = ip->i_afp->if_bytes /
-				(uint)sizeof(xfs_bmbt_rec_t);
-			ASSERT(nrecs > 0);
-			ASSERT(nrecs == ip->i_d.di_anextents);
-			ASSERT(ip->i_afp->if_bytes > 0);
+		iip->ili_fields &=
+			~(XFS_ILOG_ADATA | XFS_ILOG_ABROOT);
+
+		if ((iip->ili_fields & XFS_ILOG_AEXT) &&
+		    ip->i_d.di_anextents > 0 &&
+		    ip->i_afp->if_bytes > 0) {
+			ASSERT(ip->i_afp->if_bytes / sizeof(xfs_bmbt_rec_t) ==
+				ip->i_d.di_anextents);
 			ASSERT(ip->i_afp->if_u1.if_extents != NULL);
-			ASSERT(ip->i_d.di_anextents > 0);
-#endif
 #ifdef XFS_NATIVE_HOST
 			/*
 			 * There are not delayed allocation extents
@@ -474,29 +373,36 @@ xfs_inode_item_format(
 			iip->ili_format.ilf_asize = vecp->i_len;
 			vecp++;
 			nvecs++;
+		} else {
+			iip->ili_fields &= ~XFS_ILOG_AEXT;
 		}
 		break;
 
 	case XFS_DINODE_FMT_BTREE:
-		ASSERT(!(iip->ili_format.ilf_fields &
-			 (XFS_ILOG_ADATA | XFS_ILOG_AEXT)));
-		if (iip->ili_format.ilf_fields & XFS_ILOG_ABROOT) {
-			ASSERT(ip->i_afp->if_broot_bytes > 0);
+		iip->ili_fields &=
+			~(XFS_ILOG_ADATA | XFS_ILOG_AEXT);
+
+		if ((iip->ili_fields & XFS_ILOG_ABROOT) &&
+		    ip->i_afp->if_broot_bytes > 0) {
 			ASSERT(ip->i_afp->if_broot != NULL);
+
 			vecp->i_addr = ip->i_afp->if_broot;
 			vecp->i_len = ip->i_afp->if_broot_bytes;
 			vecp->i_type = XLOG_REG_TYPE_IATTR_BROOT;
 			vecp++;
 			nvecs++;
 			iip->ili_format.ilf_asize = ip->i_afp->if_broot_bytes;
+		} else {
+			iip->ili_fields &= ~XFS_ILOG_ABROOT;
 		}
 		break;
 
 	case XFS_DINODE_FMT_LOCAL:
-		ASSERT(!(iip->ili_format.ilf_fields &
-			 (XFS_ILOG_ABROOT | XFS_ILOG_AEXT)));
-		if (iip->ili_format.ilf_fields & XFS_ILOG_ADATA) {
-			ASSERT(ip->i_afp->if_bytes > 0);
+		iip->ili_fields &=
+			~(XFS_ILOG_AEXT | XFS_ILOG_ABROOT);
+
+		if ((iip->ili_fields & XFS_ILOG_ADATA) &&
+		    ip->i_afp->if_bytes > 0) {
 			ASSERT(ip->i_afp->if_u1.if_data != NULL);
 
 			vecp->i_addr = ip->i_afp->if_u1.if_data;
@@ -513,6 +419,8 @@ xfs_inode_item_format(
 			vecp++;
 			nvecs++;
 			iip->ili_format.ilf_asize = (unsigned)data_bytes;
+		} else {
+			iip->ili_fields &= ~XFS_ILOG_ADATA;
 		}
 		break;
 
@@ -521,7 +429,15 @@ xfs_inode_item_format(
 		break;
 	}
 
-	ASSERT(nvecs == lip->li_desc->lid_size);
+out:
+	/*
+	 * Now update the log format that goes out to disk from the in-core
+	 * values.  We always write the inode core to make the arithmetic
+	 * games in recovery easier, which isn't a big deal as just about any
+	 * transaction would dirty it anyway.
+	 */
+	iip->ili_format.ilf_fields = XFS_ILOG_CORE |
+		(iip->ili_fields & ~XFS_ILOG_TIMESTAMP);
 	iip->ili_format.ilf_size = nvecs;
 }
 
@@ -559,28 +475,19 @@ xfs_inode_item_unpin(
 	trace_xfs_inode_unpin(ip, _RET_IP_);
 	ASSERT(atomic_read(&ip->i_pincount) > 0);
 	if (atomic_dec_and_test(&ip->i_pincount))
-		wake_up(&ip->i_ipin_wait);
+		wake_up_bit(&ip->i_flags, __XFS_IPINNED_BIT);
 }
 
-/*
- * This is called to attempt to lock the inode associated with this
- * inode log item, in preparation for the push routine which does the actual
- * iflush.  Don't sleep on the inode lock or the flush lock.
- *
- * If the flush lock is already held, indicating that the inode has
- * been or is in the process of being flushed, then (ideally) we'd like to
- * see if the inode's buffer is still incore, and if so give it a nudge.
- * We delay doing so until the pushbuf routine, though, to avoid holding
- * the AIL lock across a call to the blackhole which is the buffer cache.
- * Also we don't want to sleep in any device strategy routines, which can happen
- * if we do the subsequent bawrite in here.
- */
 STATIC uint
-xfs_inode_item_trylock(
-	struct xfs_log_item	*lip)
+xfs_inode_item_push(
+	struct xfs_log_item	*lip,
+	struct list_head	*buffer_list)
 {
 	struct xfs_inode_log_item *iip = INODE_ITEM(lip);
 	struct xfs_inode	*ip = iip->ili_inode;
+	struct xfs_buf		*bp = NULL;
+	uint			rval = XFS_ITEM_SUCCESS;
+	int			error;
 
 	if (xfs_ipincount(ip) > 0)
 		return XFS_ITEM_PINNED;
@@ -588,34 +495,49 @@ xfs_inode_item_trylock(
 	if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED))
 		return XFS_ITEM_LOCKED;
 
-	if (!xfs_iflock_nowait(ip)) {
-		/*
-		 * inode has already been flushed to the backing buffer,
-		 * leave it locked in shared mode, pushbuf routine will
-		 * unlock it.
-		 */
-		return XFS_ITEM_PUSHBUF;
+	/*
+	 * Re-check the pincount now that we stabilized the value by
+	 * taking the ilock.
+	 */
+	if (xfs_ipincount(ip) > 0) {
+		rval = XFS_ITEM_PINNED;
+		goto out_unlock;
 	}
 
-	/* Stale items should force out the iclog */
+	/*
+	 * Stale inode items should force out the iclog.
+	 */
 	if (ip->i_flags & XFS_ISTALE) {
-		xfs_ifunlock(ip);
-		/*
-		 * we hold the AIL lock - notify the unlock routine of this
-		 * so it doesn't try to get the lock again.
-		 */
-		xfs_iunlock(ip, XFS_ILOCK_SHARED|XFS_IUNLOCK_NONOTIFY);
-		return XFS_ITEM_PINNED;
+		rval = XFS_ITEM_PINNED;
+		goto out_unlock;
 	}
 
-#ifdef DEBUG
-	if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) {
-		ASSERT(iip->ili_format.ilf_fields != 0);
-		ASSERT(iip->ili_logged == 0);
-		ASSERT(lip->li_flags & XFS_LI_IN_AIL);
+	/*
+	 * Someone else is already flushing the inode.  Nothing we can do
+	 * here but wait for the flush to finish and remove the item from
+	 * the AIL.
+	 */
+	if (!xfs_iflock_nowait(ip)) {
+		rval = XFS_ITEM_FLUSHING;
+		goto out_unlock;
 	}
-#endif
-	return XFS_ITEM_SUCCESS;
+
+	ASSERT(iip->ili_fields != 0 || XFS_FORCED_SHUTDOWN(ip->i_mount));
+	ASSERT(iip->ili_logged == 0 || XFS_FORCED_SHUTDOWN(ip->i_mount));
+
+	spin_unlock(&lip->li_ailp->xa_lock);
+
+	error = xfs_iflush(ip, &bp);
+	if (!error) {
+		if (!xfs_buf_delwri_queue(bp, buffer_list))
+			rval = XFS_ITEM_FLUSHING;
+		xfs_buf_relse(bp);
+	}
+
+	spin_lock(&lip->li_ailp->xa_lock);
+out_unlock:
+	xfs_iunlock(ip, XFS_ILOCK_SHARED);
+	return rval;
 }
 
 /*
@@ -642,7 +564,7 @@ xfs_inode_item_unlock(
 	if (iip->ili_extents_buf != NULL) {
 		ASSERT(ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS);
 		ASSERT(ip->i_d.di_nextents > 0);
-		ASSERT(iip->ili_format.ilf_fields & XFS_ILOG_DEXT);
+		ASSERT(iip->ili_fields & XFS_ILOG_DEXT);
 		ASSERT(ip->i_df.if_bytes > 0);
 		kmem_free(iip->ili_extents_buf);
 		iip->ili_extents_buf = NULL;
@@ -650,7 +572,7 @@ xfs_inode_item_unlock(
 	if (iip->ili_aextents_buf != NULL) {
 		ASSERT(ip->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS);
 		ASSERT(ip->i_d.di_anextents > 0);
-		ASSERT(iip->ili_format.ilf_fields & XFS_ILOG_AEXT);
+		ASSERT(iip->ili_fields & XFS_ILOG_AEXT);
 		ASSERT(ip->i_afp->if_bytes > 0);
 		kmem_free(iip->ili_aextents_buf);
 		iip->ili_aextents_buf = NULL;
@@ -700,87 +622,6 @@ xfs_inode_item_committed(
 }
 
 /*
- * This gets called by xfs_trans_push_ail(), when IOP_TRYLOCK
- * failed to get the inode flush lock but did get the inode locked SHARED.
- * Here we're trying to see if the inode buffer is incore, and if so whether it's
- * marked delayed write. If that's the case, we'll promote it and that will
- * allow the caller to write the buffer by triggering the xfsbufd to run.
- */
-STATIC bool
-xfs_inode_item_pushbuf(
-	struct xfs_log_item	*lip)
-{
-	struct xfs_inode_log_item *iip = INODE_ITEM(lip);
-	struct xfs_inode	*ip = iip->ili_inode;
-	struct xfs_buf		*bp;
-	bool			ret = true;
-
-	ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED));
-
-	/*
-	 * If a flush is not in progress anymore, chances are that the
-	 * inode was taken off the AIL. So, just get out.
-	 */
-	if (completion_done(&ip->i_flush) ||
-	    !(lip->li_flags & XFS_LI_IN_AIL)) {
-		xfs_iunlock(ip, XFS_ILOCK_SHARED);
-		return true;
-	}
-
-	bp = xfs_incore(ip->i_mount->m_ddev_targp, iip->ili_format.ilf_blkno,
-			iip->ili_format.ilf_len, XBF_TRYLOCK);
-
-	xfs_iunlock(ip, XFS_ILOCK_SHARED);
-	if (!bp)
-		return true;
-	if (XFS_BUF_ISDELAYWRITE(bp))
-		xfs_buf_delwri_promote(bp);
-	if (xfs_buf_ispinned(bp))
-		ret = false;
-	xfs_buf_relse(bp);
-	return ret;
-}
-
-/*
- * This is called to asynchronously write the inode associated with this
- * inode log item out to disk. The inode will already have been locked by
- * a successful call to xfs_inode_item_trylock().
- */
-STATIC void
-xfs_inode_item_push(
-	struct xfs_log_item	*lip)
-{
-	struct xfs_inode_log_item *iip = INODE_ITEM(lip);
-	struct xfs_inode	*ip = iip->ili_inode;
-
-	ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED));
-	ASSERT(!completion_done(&ip->i_flush));
-
-	/*
-	 * Since we were able to lock the inode's flush lock and
-	 * we found it on the AIL, the inode must be dirty.  This
-	 * is because the inode is removed from the AIL while still
-	 * holding the flush lock in xfs_iflush_done().  Thus, if
-	 * we found it in the AIL and were able to obtain the flush
-	 * lock without sleeping, then there must not have been
-	 * anyone in the process of flushing the inode.
-	 */
-	ASSERT(XFS_FORCED_SHUTDOWN(ip->i_mount) ||
-	       iip->ili_format.ilf_fields != 0);
-
-	/*
-	 * Push the inode to it's backing buffer. This will not remove the
-	 * inode from the AIL - a further push will be required to trigger a
-	 * buffer push. However, this allows all the dirty inodes to be pushed
-	 * to the buffer before it is pushed to disk. The buffer IO completion
-	 * will pull the inode from the AIL, mark it clean and unlock the flush
-	 * lock.
-	 */
-	(void) xfs_iflush(ip, SYNC_TRYLOCK);
-	xfs_iunlock(ip, XFS_ILOCK_SHARED);
-}
-
-/*
  * XXX rcc - this one really has to do something.  Probably needs
  * to stamp in a new field in the incore inode.
  */
@@ -800,11 +641,9 @@ static const struct xfs_item_ops xfs_inode_item_ops = {
 	.iop_format	= xfs_inode_item_format,
 	.iop_pin	= xfs_inode_item_pin,
 	.iop_unpin	= xfs_inode_item_unpin,
-	.iop_trylock	= xfs_inode_item_trylock,
 	.iop_unlock	= xfs_inode_item_unlock,
 	.iop_committed	= xfs_inode_item_committed,
 	.iop_push	= xfs_inode_item_push,
-	.iop_pushbuf	= xfs_inode_item_pushbuf,
 	.iop_committing = xfs_inode_item_committing
 };
 
@@ -935,7 +774,8 @@ xfs_iflush_done(
 			ASSERT(i <= need_ail);
 		}
 		/* xfs_trans_ail_delete_bulk() drops the AIL lock. */
-		xfs_trans_ail_delete_bulk(ailp, log_items, i);
+		xfs_trans_ail_delete_bulk(ailp, log_items, i,
+					  SHUTDOWN_CORRUPT_INCORE);
 	}
 
 
@@ -956,16 +796,15 @@ xfs_iflush_done(
 }
 
 /*
- * This is the inode flushing abort routine.  It is called
- * from xfs_iflush when the filesystem is shutting down to clean
- * up the inode state.
- * It is responsible for removing the inode item
- * from the AIL if it has not been re-logged, and unlocking the inode's
- * flush lock.
+ * This is the inode flushing abort routine.  It is called from xfs_iflush when
+ * the filesystem is shutting down to clean up the inode state.  It is
+ * responsible for removing the inode item from the AIL if it has not been
+ * re-logged, and unlocking the inode's flush lock.
  */
 void
 xfs_iflush_abort(
-	xfs_inode_t		*ip)
+	xfs_inode_t		*ip,
+	bool			stale)
 {
 	xfs_inode_log_item_t	*iip = ip->i_itemp;
 
@@ -975,7 +814,10 @@ xfs_iflush_abort(
 			spin_lock(&ailp->xa_lock);
 			if (iip->ili_item.li_flags & XFS_LI_IN_AIL) {
 				/* xfs_trans_ail_delete() drops the AIL lock. */
-				xfs_trans_ail_delete(ailp, (xfs_log_item_t *)iip);
+				xfs_trans_ail_delete(ailp, &iip->ili_item,
+						stale ?
+						     SHUTDOWN_LOG_IO_ERROR :
+						     SHUTDOWN_CORRUPT_INCORE);
 			} else
 				spin_unlock(&ailp->xa_lock);
 		}
@@ -989,7 +831,7 @@ xfs_iflush_abort(
 		 * Clear the inode logging fields so no more flushes are
 		 * attempted.
 		 */
-		iip->ili_format.ilf_fields = 0;
+		iip->ili_fields = 0;
 	}
 	/*
 	 * Release the inode's flush lock since we're done with it.
@@ -1002,7 +844,7 @@ xfs_istale_done(
 	struct xfs_buf		*bp,
 	struct xfs_log_item	*lip)
 {
-	xfs_iflush_abort(INODE_ITEM(lip)->ili_inode);
+	xfs_iflush_abort(INODE_ITEM(lip)->ili_inode, true);
 }
 
 /*
diff --git a/fs/xfs/xfs_inode_item.h b/fs/xfs/xfs_inode_item.h
index d3dee61e6d9..376d4d0b263 100644
--- a/fs/xfs/xfs_inode_item.h
+++ b/fs/xfs/xfs_inode_item.h
@@ -86,6 +86,15 @@ typedef struct xfs_inode_log_format_64 {
 #define	XFS_ILOG_AEXT	0x080	/* log i_af.if_extents */
 #define	XFS_ILOG_ABROOT	0x100	/* log i_af.i_broot */
 
+
+/*
+ * The timestamps are dirty, but not necessarily anything else in the inode
+ * core.  Unlike the other fields above this one must never make it to disk
+ * in the ilf_fields of the inode_log_format, but is purely store in-memory in
+ * ili_fields in the inode_log_item.
+ */
+#define XFS_ILOG_TIMESTAMP	0x4000
+
 #define	XFS_ILOG_NONCORE	(XFS_ILOG_DDATA | XFS_ILOG_DEXT | \
 				 XFS_ILOG_DBROOT | XFS_ILOG_DEV | \
 				 XFS_ILOG_UUID | XFS_ILOG_ADATA | \
@@ -101,7 +110,7 @@ typedef struct xfs_inode_log_format_64 {
 				 XFS_ILOG_DEXT | XFS_ILOG_DBROOT | \
 				 XFS_ILOG_DEV | XFS_ILOG_UUID | \
 				 XFS_ILOG_ADATA | XFS_ILOG_AEXT | \
-				 XFS_ILOG_ABROOT)
+				 XFS_ILOG_ABROOT | XFS_ILOG_TIMESTAMP)
 
 static inline int xfs_ilog_fbroot(int w)
 {
@@ -134,6 +143,7 @@ typedef struct xfs_inode_log_item {
 	unsigned short		ili_lock_flags;	   /* lock flags */
 	unsigned short		ili_logged;	   /* flushed logged data */
 	unsigned int		ili_last_fields;   /* fields when flushed */
+	unsigned int		ili_fields;	   /* fields to be logged */
 	struct xfs_bmbt_rec	*ili_extents_buf;  /* array of logged
 						      data exts */
 	struct xfs_bmbt_rec	*ili_aextents_buf; /* array of logged
@@ -148,16 +158,14 @@ typedef struct xfs_inode_log_item {
 
 static inline int xfs_inode_clean(xfs_inode_t *ip)
 {
-	return (!ip->i_itemp ||
-		!(ip->i_itemp->ili_format.ilf_fields & XFS_ILOG_ALL)) &&
-	       !ip->i_update_core;
+	return !ip->i_itemp || !(ip->i_itemp->ili_fields & XFS_ILOG_ALL);
 }
 
 extern void xfs_inode_item_init(struct xfs_inode *, struct xfs_mount *);
 extern void xfs_inode_item_destroy(struct xfs_inode *);
 extern void xfs_iflush_done(struct xfs_buf *, struct xfs_log_item *);
 extern void xfs_istale_done(struct xfs_buf *, struct xfs_log_item *);
-extern void xfs_iflush_abort(struct xfs_inode *);
+extern void xfs_iflush_abort(struct xfs_inode *, bool);
 extern int xfs_inode_item_format_convert(xfs_log_iovec_t *,
 					 xfs_inode_log_format_t *);
 
diff --git a/fs/xfs/xfs_inum.h b/fs/xfs/xfs_inum.h
index b253c0ea5be..90efdaf1706 100644
--- a/fs/xfs/xfs_inum.h
+++ b/fs/xfs/xfs_inum.h
@@ -26,11 +26,6 @@
  * high agno_log-agblklog-inopblog bits - 0
  */
 
-typedef	__uint32_t	xfs_agino_t;	/* within allocation grp inode number */
-
-#define	NULLFSINO	((xfs_ino_t)-1)
-#define	NULLAGINO	((xfs_agino_t)-1)
-
 struct xfs_mount;
 
 #define	XFS_INO_MASK(k)			(__uint32_t)((1ULL << (k)) - 1)
diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c
index d99a9051890..3a05a41b5d7 100644
--- a/fs/xfs/xfs_ioctl.c
+++ b/fs/xfs/xfs_ioctl.c
@@ -17,9 +17,7 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -209,6 +207,7 @@ xfs_open_by_handle(
 	struct file		*filp;
 	struct inode		*inode;
 	struct dentry		*dentry;
+	fmode_t			fmode;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -XFS_ERROR(EPERM);
@@ -228,26 +227,21 @@ xfs_open_by_handle(
 	hreq->oflags |= O_LARGEFILE;
 #endif
 
-	/* Put open permission in namei format. */
 	permflag = hreq->oflags;
-	if ((permflag+1) & O_ACCMODE)
-		permflag++;
-	if (permflag & O_TRUNC)
-		permflag |= 2;
-
+	fmode = OPEN_FMODE(permflag);
 	if ((!(permflag & O_APPEND) || (permflag & O_TRUNC)) &&
-	    (permflag & FMODE_WRITE) && IS_APPEND(inode)) {
+	    (fmode & FMODE_WRITE) && IS_APPEND(inode)) {
 		error = -XFS_ERROR(EPERM);
 		goto out_dput;
 	}
 
-	if ((permflag & FMODE_WRITE) && IS_IMMUTABLE(inode)) {
+	if ((fmode & FMODE_WRITE) && IS_IMMUTABLE(inode)) {
 		error = -XFS_ERROR(EACCES);
 		goto out_dput;
 	}
 
 	/* Can't write directories. */
-	if (S_ISDIR(inode->i_mode) && (permflag & FMODE_WRITE)) {
+	if (S_ISDIR(inode->i_mode) && (fmode & FMODE_WRITE)) {
 		error = -XFS_ERROR(EISDIR);
 		goto out_dput;
 	}
@@ -450,9 +444,12 @@ xfs_attrmulti_attr_get(
 
 	if (*len > XATTR_SIZE_MAX)
 		return EINVAL;
-	kbuf = kmalloc(*len, GFP_KERNEL);
-	if (!kbuf)
-		return ENOMEM;
+	kbuf = kmem_zalloc(*len, KM_SLEEP | KM_MAYFAIL);
+	if (!kbuf) {
+		kbuf = kmem_zalloc_large(*len);
+		if (!kbuf)
+			return ENOMEM;
+	}
 
 	error = xfs_attr_get(XFS_I(inode), name, kbuf, (int *)len, flags);
 	if (error)
@@ -462,7 +459,10 @@ xfs_attrmulti_attr_get(
 		error = EFAULT;
 
  out_kfree:
-	kfree(kbuf);
+	if (is_vmalloc_addr(kbuf))
+		kmem_free_large(kbuf);
+	else
+		kmem_free(kbuf);
 	return error;
 }
 
@@ -559,23 +559,23 @@ xfs_attrmulti_by_handle(
 					ops[i].am_flags);
 			break;
 		case ATTR_OP_SET:
-			ops[i].am_error = mnt_want_write(parfilp->f_path.mnt);
+			ops[i].am_error = mnt_want_write_file(parfilp);
 			if (ops[i].am_error)
 				break;
 			ops[i].am_error = xfs_attrmulti_attr_set(
 					dentry->d_inode, attr_name,
 					ops[i].am_attrvalue, ops[i].am_length,
 					ops[i].am_flags);
-			mnt_drop_write(parfilp->f_path.mnt);
+			mnt_drop_write_file(parfilp);
 			break;
 		case ATTR_OP_REMOVE:
-			ops[i].am_error = mnt_want_write(parfilp->f_path.mnt);
+			ops[i].am_error = mnt_want_write_file(parfilp);
 			if (ops[i].am_error)
 				break;
 			ops[i].am_error = xfs_attrmulti_attr_remove(
 					dentry->d_inode, attr_name,
 					ops[i].am_flags);
-			mnt_drop_write(parfilp->f_path.mnt);
+			mnt_drop_write_file(parfilp);
 			break;
 		default:
 			ops[i].am_error = EINVAL;
diff --git a/fs/xfs/xfs_ioctl32.c b/fs/xfs/xfs_ioctl32.c
index 54e623bfbb8..c4f2da0d2bf 100644
--- a/fs/xfs/xfs_ioctl32.c
+++ b/fs/xfs/xfs_ioctl32.c
@@ -22,9 +22,7 @@
 #include <asm/uaccess.h>
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -293,7 +291,7 @@ xfs_compat_ioc_bulkstat(
 		int res;
 
 		error = xfs_bulkstat_one_compat(mp, inlast, bulkreq.ubuffer,
-				sizeof(compat_xfs_bstat_t), 0, &res);
+				sizeof(compat_xfs_bstat_t), NULL, &res);
 	} else if (cmd == XFS_IOC_FSBULKSTAT_32) {
 		error = xfs_bulkstat(mp, &inlast, &count,
 			xfs_bulkstat_one_compat, sizeof(compat_xfs_bstat_t),
@@ -454,23 +452,23 @@ xfs_compat_attrmulti_by_handle(
 					&ops[i].am_length, ops[i].am_flags);
 			break;
 		case ATTR_OP_SET:
-			ops[i].am_error = mnt_want_write(parfilp->f_path.mnt);
+			ops[i].am_error = mnt_want_write_file(parfilp);
 			if (ops[i].am_error)
 				break;
 			ops[i].am_error = xfs_attrmulti_attr_set(
 					dentry->d_inode, attr_name,
 					compat_ptr(ops[i].am_attrvalue),
 					ops[i].am_length, ops[i].am_flags);
-			mnt_drop_write(parfilp->f_path.mnt);
+			mnt_drop_write_file(parfilp);
 			break;
 		case ATTR_OP_REMOVE:
-			ops[i].am_error = mnt_want_write(parfilp->f_path.mnt);
+			ops[i].am_error = mnt_want_write_file(parfilp);
 			if (ops[i].am_error)
 				break;
 			ops[i].am_error = xfs_attrmulti_attr_remove(
 					dentry->d_inode, attr_name,
 					ops[i].am_flags);
-			mnt_drop_write(parfilp->f_path.mnt);
+			mnt_drop_write_file(parfilp);
 			break;
 		default:
 			ops[i].am_error = EINVAL;
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 9afa282aa93..aadfce6681e 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -17,9 +17,7 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -31,12 +29,12 @@
 #include "xfs_ialloc_btree.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
+#include "xfs_inode_item.h"
 #include "xfs_btree.h"
 #include "xfs_bmap.h"
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
 #include "xfs_itable.h"
-#include "xfs_rw.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_space.h"
@@ -57,26 +55,26 @@ xfs_iomap_eof_align_last_fsb(
 	xfs_fileoff_t	*last_fsb)
 {
 	xfs_fileoff_t	new_last_fsb = 0;
-	xfs_extlen_t	align;
+	xfs_extlen_t	align = 0;
 	int		eof, error;
 
-	if (XFS_IS_REALTIME_INODE(ip))
-		;
-	/*
-	 * If mounted with the "-o swalloc" option, roundup the allocation
-	 * request to a stripe width boundary if the file size is >=
-	 * stripe width and we are allocating past the allocation eof.
-	 */
-	else if (mp->m_swidth && (mp->m_flags & XFS_MOUNT_SWALLOC) &&
-	        (ip->i_size >= XFS_FSB_TO_B(mp, mp->m_swidth)))
-		new_last_fsb = roundup_64(*last_fsb, mp->m_swidth);
-	/*
-	 * Roundup the allocation request to a stripe unit (m_dalign) boundary
-	 * if the file size is >= stripe unit size, and we are allocating past
-	 * the allocation eof.
-	 */
-	else if (mp->m_dalign && (ip->i_size >= XFS_FSB_TO_B(mp, mp->m_dalign)))
-		new_last_fsb = roundup_64(*last_fsb, mp->m_dalign);
+	if (!XFS_IS_REALTIME_INODE(ip)) {
+		/*
+		 * Round up the allocation request to a stripe unit
+		 * (m_dalign) boundary if the file size is >= stripe unit
+		 * size, and we are allocating past the allocation eof.
+		 *
+		 * If mounted with the "-o swalloc" option the alignment is
+		 * increased from the strip unit size to the stripe width.
+		 */
+		if (mp->m_swidth && (mp->m_flags & XFS_MOUNT_SWALLOC))
+			align = mp->m_swidth;
+		else if (mp->m_dalign)
+			align = mp->m_dalign;
+
+		if (align && XFS_ISIZE(ip) >= XFS_FSB_TO_B(mp, align))
+			new_last_fsb = roundup_64(*last_fsb, align);
+	}
 
 	/*
 	 * Always round up the allocation request to an extent boundary
@@ -141,11 +139,7 @@ xfs_iomap_write_direct(
 	int		committed;
 	int		error;
 
-	/*
-	 * Make sure that the dquots are there. This doesn't hold
-	 * the ilock across a disk read.
-	 */
-	error = xfs_qm_dqattach_locked(ip, 0);
+	error = xfs_qm_dqattach(ip, 0);
 	if (error)
 		return XFS_ERROR(error);
 
@@ -154,10 +148,10 @@ xfs_iomap_write_direct(
 
 	offset_fsb = XFS_B_TO_FSBT(mp, offset);
 	last_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)(offset + count)));
-	if ((offset + count) > ip->i_size) {
+	if ((offset + count) > XFS_ISIZE(ip)) {
 		error = xfs_iomap_eof_align_last_fsb(mp, ip, extsz, &last_fsb);
 		if (error)
-			goto error_out;
+			return XFS_ERROR(error);
 	} else {
 		if (nmaps && (imap->br_startblock == HOLESTARTBLOCK))
 			last_fsb = MIN(last_fsb, (xfs_fileoff_t)
@@ -189,7 +183,6 @@ xfs_iomap_write_direct(
 	/*
 	 * Allocate and setup the transaction
 	 */
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 	tp = xfs_trans_alloc(mp, XFS_TRANS_DIOSTRAT);
 	error = xfs_trans_reserve(tp, resblks,
 			XFS_WRITE_LOG_RES(mp), resrtextents,
@@ -198,20 +191,21 @@ xfs_iomap_write_direct(
 	/*
 	 * Check for running out of space, note: need lock to return
 	 */
-	if (error)
+	if (error) {
 		xfs_trans_cancel(tp, 0);
+		return XFS_ERROR(error);
+	}
+
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	if (error)
-		goto error_out;
 
 	error = xfs_trans_reserve_quota_nblks(tp, ip, qblocks, 0, quota_flag);
 	if (error)
-		goto error1;
+		goto out_trans_cancel;
 
 	xfs_trans_ijoin(tp, ip, 0);
 
 	bmapi_flag = 0;
-	if (offset < ip->i_size || extsz)
+	if (offset < XFS_ISIZE(ip) || extsz)
 		bmapi_flag |= XFS_BMAPI_PREALLOC;
 
 	/*
@@ -223,42 +217,39 @@ xfs_iomap_write_direct(
 	error = xfs_bmapi_write(tp, ip, offset_fsb, count_fsb, bmapi_flag,
 				&firstfsb, 0, imap, &nimaps, &free_list);
 	if (error)
-		goto error0;
+		goto out_bmap_cancel;
 
 	/*
 	 * Complete the transaction
 	 */
 	error = xfs_bmap_finish(&tp, &free_list, &committed);
 	if (error)
-		goto error0;
+		goto out_bmap_cancel;
 	error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
 	if (error)
-		goto error_out;
+		goto out_unlock;
 
 	/*
 	 * Copy any maps to caller's array and return any error.
 	 */
 	if (nimaps == 0) {
-		error = ENOSPC;
-		goto error_out;
+		error = XFS_ERROR(ENOSPC);
+		goto out_unlock;
 	}
 
-	if (!(imap->br_startblock || XFS_IS_REALTIME_INODE(ip))) {
+	if (!(imap->br_startblock || XFS_IS_REALTIME_INODE(ip)))
 		error = xfs_alert_fsblock_zero(ip, imap);
-		goto error_out;
-	}
 
-	return 0;
+out_unlock:
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	return error;
 
-error0:	/* Cancel bmap, unlock inode, unreserve quota blocks, cancel trans */
+out_bmap_cancel:
 	xfs_bmap_cancel(&free_list);
-	xfs_trans_unreserve_quota_nblks(tp, ip, qblocks, 0, quota_flag);
-
-error1:	/* Just cancel transaction */
+	xfs_trans_unreserve_quota_nblks(tp, ip, (long)qblocks, 0, quota_flag);
+out_trans_cancel:
 	xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | XFS_TRANS_ABORT);
-
-error_out:
-	return XFS_ERROR(error);
+	goto out_unlock;
 }
 
 /*
@@ -286,7 +277,7 @@ xfs_iomap_eof_want_preallocate(
 	int		found_delalloc = 0;
 
 	*prealloc = 0;
-	if ((offset + count) <= ip->i_size)
+	if (offset + count <= XFS_ISIZE(ip))
 		return 0;
 
 	/*
@@ -340,7 +331,7 @@ xfs_iomap_prealloc_size(
 		 * if we pass in alloc_blocks = 0. Hence the "+ 1" to
 		 * ensure we always pass in a non-zero value.
 		 */
-		alloc_blocks = XFS_B_TO_FSB(mp, ip->i_size) + 1;
+		alloc_blocks = XFS_B_TO_FSB(mp, XFS_ISIZE(ip)) + 1;
 		alloc_blocks = XFS_FILEOFF_MIN(MAXEXTLEN,
 					rounddown_pow_of_two(alloc_blocks));
 
@@ -421,6 +412,15 @@ retry:
 			return error;
 	}
 
+	/*
+	 * Make sure preallocation does not create extents beyond the range we
+	 * actually support in this filesystem.
+	 */
+	if (last_fsb > XFS_B_TO_FSB(mp, mp->m_maxioffset))
+		last_fsb = XFS_B_TO_FSB(mp, mp->m_maxioffset);
+
+	ASSERT(last_fsb > offset_fsb);
+
 	nimaps = XFS_WRITE_IMAPS;
 	error = xfs_bmapi_delay(ip, offset_fsb, last_fsb - offset_fsb,
 				imap, &nimaps, XFS_BMAPI_ENTIRE);
@@ -564,7 +564,7 @@ xfs_iomap_write_allocate(
 			 * back....
 			 */
 			nimaps = 1;
-			end_fsb = XFS_B_TO_FSB(mp, ip->i_size);
+			end_fsb = XFS_B_TO_FSB(mp, XFS_ISIZE(ip));
 			error = xfs_bmap_last_offset(NULL, ip, &last_block,
 							XFS_DATA_FORK);
 			if (error)
@@ -645,6 +645,7 @@ xfs_iomap_write_unwritten(
 	xfs_trans_t	*tp;
 	xfs_bmbt_irec_t imap;
 	xfs_bmap_free_t free_list;
+	xfs_fsize_t	i_size;
 	uint		resblks;
 	int		committed;
 	int		error;
@@ -705,7 +706,22 @@ xfs_iomap_write_unwritten(
 		if (error)
 			goto error_on_bmapi_transaction;
 
-		error = xfs_bmap_finish(&(tp), &(free_list), &committed);
+		/*
+		 * Log the updated inode size as we go.  We have to be careful
+		 * to only log it up to the actual write offset if it is
+		 * halfway into a block.
+		 */
+		i_size = XFS_FSB_TO_B(mp, offset_fsb + count_fsb);
+		if (i_size > offset + count)
+			i_size = offset + count;
+
+		i_size = xfs_new_eof(ip, i_size);
+		if (i_size) {
+			ip->i_d.di_size = i_size;
+			xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+		}
+
+		error = xfs_bmap_finish(&tp, &free_list, &committed);
 		if (error)
 			goto error_on_bmapi_transaction;
 
diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c
index 23ce927973a..1a25fd80279 100644
--- a/fs/xfs/xfs_iops.c
+++ b/fs/xfs/xfs_iops.c
@@ -18,9 +18,7 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_acl.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -34,7 +32,6 @@
 #include "xfs_rtalloc.h"
 #include "xfs_error.h"
 #include "xfs_itable.h"
-#include "xfs_rw.h"
 #include "xfs_attr.h"
 #include "xfs_buf_item.h"
 #include "xfs_utils.h"
@@ -50,65 +47,15 @@
 #include <linux/fiemap.h>
 #include <linux/slab.h>
 
-/*
- * Bring the timestamps in the XFS inode uptodate.
- *
- * Used before writing the inode to disk.
- */
-void
-xfs_synchronize_times(
-	xfs_inode_t	*ip)
-{
-	struct inode	*inode = VFS_I(ip);
-
-	ip->i_d.di_atime.t_sec = (__int32_t)inode->i_atime.tv_sec;
-	ip->i_d.di_atime.t_nsec = (__int32_t)inode->i_atime.tv_nsec;
-	ip->i_d.di_ctime.t_sec = (__int32_t)inode->i_ctime.tv_sec;
-	ip->i_d.di_ctime.t_nsec = (__int32_t)inode->i_ctime.tv_nsec;
-	ip->i_d.di_mtime.t_sec = (__int32_t)inode->i_mtime.tv_sec;
-	ip->i_d.di_mtime.t_nsec = (__int32_t)inode->i_mtime.tv_nsec;
-}
-
-/*
- * If the linux inode is valid, mark it dirty, else mark the dirty state
- * in the XFS inode to make sure we pick it up when reclaiming the inode.
- */
-void
-xfs_mark_inode_dirty_sync(
-	xfs_inode_t	*ip)
-{
-	struct inode	*inode = VFS_I(ip);
-
-	if (!(inode->i_state & (I_WILL_FREE|I_FREEING)))
-		mark_inode_dirty_sync(inode);
-	else {
-		barrier();
-		ip->i_update_core = 1;
-	}
-}
-
-void
-xfs_mark_inode_dirty(
-	xfs_inode_t	*ip)
-{
-	struct inode	*inode = VFS_I(ip);
-
-	if (!(inode->i_state & (I_WILL_FREE|I_FREEING)))
-		mark_inode_dirty(inode);
-	else {
-		barrier();
-		ip->i_update_core = 1;
-	}
-
-}
-
-
-int xfs_initxattrs(struct inode *inode, const struct xattr *xattr_array,
-		   void *fs_info)
+static int
+xfs_initxattrs(
+	struct inode		*inode,
+	const struct xattr	*xattr_array,
+	void			*fs_info)
 {
-	const struct xattr *xattr;
-	struct xfs_inode *ip = XFS_I(inode);
-	int error = 0;
+	const struct xattr	*xattr;
+	struct xfs_inode	*ip = XFS_I(inode);
+	int			error = 0;
 
 	for (xattr = xattr_array; xattr->name != NULL; xattr++) {
 		error = xfs_attr_set(ip, xattr->name, xattr->value,
@@ -168,7 +115,7 @@ STATIC int
 xfs_vn_mknod(
 	struct inode	*dir,
 	struct dentry	*dentry,
-	int		mode,
+	umode_t		mode,
 	dev_t		rdev)
 {
 	struct inode	*inode;
@@ -231,7 +178,7 @@ STATIC int
 xfs_vn_create(
 	struct inode	*dir,
 	struct dentry	*dentry,
-	int		mode,
+	umode_t		mode,
 	struct nameidata *nd)
 {
 	return xfs_vn_mknod(dir, dentry, mode, 0);
@@ -241,7 +188,7 @@ STATIC int
 xfs_vn_mkdir(
 	struct inode	*dir,
 	struct dentry	*dentry,
-	int		mode)
+	umode_t		mode)
 {
 	return xfs_vn_mknod(dir, dentry, mode|S_IFDIR, 0);
 }
@@ -366,7 +313,7 @@ xfs_vn_symlink(
 	struct xfs_inode *cip = NULL;
 	struct xfs_name	name;
 	int		error;
-	mode_t		mode;
+	umode_t		mode;
 
 	mode = S_IFLNK |
 		(irix_symlink_mode ? 0777 & ~current_umask() : S_IRWXUGO);
@@ -678,19 +625,16 @@ xfs_setattr_nonsize(
 		inode->i_atime = iattr->ia_atime;
 		ip->i_d.di_atime.t_sec = iattr->ia_atime.tv_sec;
 		ip->i_d.di_atime.t_nsec = iattr->ia_atime.tv_nsec;
-		ip->i_update_core = 1;
 	}
 	if (mask & ATTR_CTIME) {
 		inode->i_ctime = iattr->ia_ctime;
 		ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec;
 		ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec;
-		ip->i_update_core = 1;
 	}
 	if (mask & ATTR_MTIME) {
 		inode->i_mtime = iattr->ia_mtime;
 		ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec;
 		ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec;
-		ip->i_update_core = 1;
 	}
 
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
@@ -750,9 +694,10 @@ xfs_setattr_size(
 	struct xfs_mount	*mp = ip->i_mount;
 	struct inode		*inode = VFS_I(ip);
 	int			mask = iattr->ia_valid;
+	xfs_off_t		oldsize, newsize;
 	struct xfs_trans	*tp;
 	int			error;
-	uint			lock_flags;
+	uint			lock_flags = 0;
 	uint			commit_flags = 0;
 
 	trace_xfs_setattr(ip);
@@ -772,16 +717,18 @@ xfs_setattr_size(
 			ATTR_MTIME_SET|ATTR_KILL_SUID|ATTR_KILL_SGID|
 			ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0);
 
-	lock_flags = XFS_ILOCK_EXCL;
-	if (!(flags & XFS_ATTR_NOLOCK))
+	if (!(flags & XFS_ATTR_NOLOCK)) {
 		lock_flags |= XFS_IOLOCK_EXCL;
-	xfs_ilock(ip, lock_flags);
+		xfs_ilock(ip, lock_flags);
+	}
+
+	oldsize = inode->i_size;
+	newsize = iattr->ia_size;
 
 	/*
 	 * Short circuit the truncate case for zero length files.
 	 */
-	if (iattr->ia_size == 0 &&
-	    ip->i_size == 0 && ip->i_d.di_nextents == 0) {
+	if (newsize == 0 && oldsize == 0 && ip->i_d.di_nextents == 0) {
 		if (!(mask & (ATTR_CTIME|ATTR_MTIME)))
 			goto out_unlock;
 
@@ -796,7 +743,7 @@ xfs_setattr_size(
 	/*
 	 * Make sure that the dquots are attached to the inode.
 	 */
-	error = xfs_qm_dqattach_locked(ip, 0);
+	error = xfs_qm_dqattach(ip, 0);
 	if (error)
 		goto out_unlock;
 
@@ -807,19 +754,17 @@ xfs_setattr_size(
 	 * the inode to the transaction, because the inode cannot be unlocked
 	 * once it is a part of the transaction.
 	 */
-	if (iattr->ia_size > ip->i_size) {
+	if (newsize > oldsize) {
 		/*
 		 * Do the first part of growing a file: zero any data in the
 		 * last block that is beyond the old EOF.  We need to do this
 		 * before the inode is joined to the transaction to modify
 		 * i_size.
 		 */
-		error = xfs_zero_eof(ip, iattr->ia_size, ip->i_size);
+		error = xfs_zero_eof(ip, newsize, oldsize);
 		if (error)
 			goto out_unlock;
 	}
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	lock_flags &= ~XFS_ILOCK_EXCL;
 
 	/*
 	 * We are going to log the inode size change in this transaction so
@@ -833,8 +778,8 @@ xfs_setattr_size(
 	 * here and prevents waiting for other data not within the range we
 	 * care about here.
 	 */
-	if (ip->i_size != ip->i_d.di_size && iattr->ia_size > ip->i_d.di_size) {
-		error = xfs_flush_pages(ip, ip->i_d.di_size, iattr->ia_size, 0,
+	if (oldsize != ip->i_d.di_size && newsize > ip->i_d.di_size) {
+		error = xfs_flush_pages(ip, ip->i_d.di_size, newsize, 0,
 					FI_NONE);
 		if (error)
 			goto out_unlock;
@@ -845,8 +790,7 @@ xfs_setattr_size(
 	 */
 	inode_dio_wait(inode);
 
-	error = -block_truncate_page(inode->i_mapping, iattr->ia_size,
-				     xfs_get_blocks);
+	error = -block_truncate_page(inode->i_mapping, newsize, xfs_get_blocks);
 	if (error)
 		goto out_unlock;
 
@@ -857,7 +801,7 @@ xfs_setattr_size(
 	if (error)
 		goto out_trans_cancel;
 
-	truncate_setsize(inode, iattr->ia_size);
+	truncate_setsize(inode, newsize);
 
 	commit_flags = XFS_TRANS_RELEASE_LOG_RES;
 	lock_flags |= XFS_ILOCK_EXCL;
@@ -876,19 +820,29 @@ xfs_setattr_size(
 	 * these flags set.  For all other operations the VFS set these flags
 	 * explicitly if it wants a timestamp update.
 	 */
-	if (iattr->ia_size != ip->i_size &&
-	    (!(mask & (ATTR_CTIME | ATTR_MTIME)))) {
+	if (newsize != oldsize && (!(mask & (ATTR_CTIME | ATTR_MTIME)))) {
 		iattr->ia_ctime = iattr->ia_mtime =
 			current_fs_time(inode->i_sb);
 		mask |= ATTR_CTIME | ATTR_MTIME;
 	}
 
-	if (iattr->ia_size > ip->i_size) {
-		ip->i_d.di_size = iattr->ia_size;
-		ip->i_size = iattr->ia_size;
-	} else if (iattr->ia_size <= ip->i_size ||
-		   (iattr->ia_size == 0 && ip->i_d.di_nextents)) {
-		error = xfs_itruncate_data(&tp, ip, iattr->ia_size);
+	/*
+	 * The first thing we do is set the size to new_size permanently on
+	 * disk.  This way we don't have to worry about anyone ever being able
+	 * to look at the data being freed even in the face of a crash.
+	 * What we're getting around here is the case where we free a block, it
+	 * is allocated to another file, it is written to, and then we crash.
+	 * If the new data gets written to the file but the log buffers
+	 * containing the free and reallocation don't, then we'd end up with
+	 * garbage in the blocks being freed.  As long as we make the new size
+	 * permanent before actually freeing any blocks it doesn't matter if
+	 * they get written to.
+	 */
+	ip->i_d.di_size = newsize;
+	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+
+	if (newsize <= oldsize) {
+		error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK, newsize);
 		if (error)
 			goto out_trans_abort;
 
@@ -906,13 +860,11 @@ xfs_setattr_size(
 		inode->i_ctime = iattr->ia_ctime;
 		ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec;
 		ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec;
-		ip->i_update_core = 1;
 	}
 	if (mask & ATTR_MTIME) {
 		inode->i_mtime = iattr->ia_mtime;
 		ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec;
 		ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec;
-		ip->i_update_core = 1;
 	}
 
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c
index 751e94fe1f7..eff577a9b67 100644
--- a/fs/xfs/xfs_itable.c
+++ b/fs/xfs/xfs_itable.c
@@ -18,7 +18,6 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_types.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
 #include "xfs_inum.h"
 #include "xfs_trans.h"
@@ -62,7 +61,6 @@ xfs_bulkstat_one_int(
 {
 	struct xfs_icdinode	*dic;		/* dinode core info pointer */
 	struct xfs_inode	*ip;		/* incore inode pointer */
-	struct inode		*inode;
 	struct xfs_bstat	*buf;		/* return buffer */
 	int			error = 0;	/* error value */
 
@@ -76,7 +74,8 @@ xfs_bulkstat_one_int(
 		return XFS_ERROR(ENOMEM);
 
 	error = xfs_iget(mp, NULL, ino,
-			 XFS_IGET_UNTRUSTED, XFS_ILOCK_SHARED, &ip);
+			 (XFS_IGET_DONTCACHE | XFS_IGET_UNTRUSTED),
+			 XFS_ILOCK_SHARED, &ip);
 	if (error) {
 		*stat = BULKSTAT_RV_NOTHING;
 		goto out_free;
@@ -86,7 +85,6 @@ xfs_bulkstat_one_int(
 	ASSERT(ip->i_imap.im_blkno != 0);
 
 	dic = &ip->i_d;
-	inode = VFS_I(ip);
 
 	/* xfs_iget returns the following without needing
 	 * further change.
@@ -99,19 +97,12 @@ xfs_bulkstat_one_int(
 	buf->bs_uid = dic->di_uid;
 	buf->bs_gid = dic->di_gid;
 	buf->bs_size = dic->di_size;
-
-	/*
-	 * We need to read the timestamps from the Linux inode because
-	 * the VFS keeps writing directly into the inode structure instead
-	 * of telling us about the updates.
-	 */
-	buf->bs_atime.tv_sec = inode->i_atime.tv_sec;
-	buf->bs_atime.tv_nsec = inode->i_atime.tv_nsec;
-	buf->bs_mtime.tv_sec = inode->i_mtime.tv_sec;
-	buf->bs_mtime.tv_nsec = inode->i_mtime.tv_nsec;
-	buf->bs_ctime.tv_sec = inode->i_ctime.tv_sec;
-	buf->bs_ctime.tv_nsec = inode->i_ctime.tv_nsec;
-
+	buf->bs_atime.tv_sec = dic->di_atime.t_sec;
+	buf->bs_atime.tv_nsec = dic->di_atime.t_nsec;
+	buf->bs_mtime.tv_sec = dic->di_mtime.t_sec;
+	buf->bs_mtime.tv_nsec = dic->di_mtime.t_nsec;
+	buf->bs_ctime.tv_sec = dic->di_ctime.t_sec;
+	buf->bs_ctime.tv_nsec = dic->di_ctime.t_nsec;
 	buf->bs_xflags = xfs_ip2xflags(ip);
 	buf->bs_extsize = dic->di_extsize << mp->m_sb.sb_blocklog;
 	buf->bs_extents = dic->di_nextents;
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 34817adf4b9..d90d4a38860 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -18,9 +18,7 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_types.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -35,19 +33,26 @@
 #include "xfs_trans_priv.h"
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
-#include "xfs_rw.h"
 #include "xfs_trace.h"
 
 kmem_zone_t	*xfs_log_ticket_zone;
 
 /* Local miscellaneous function prototypes */
-STATIC int	 xlog_commit_record(struct log *log, struct xlog_ticket *ticket,
-				    xlog_in_core_t **, xfs_lsn_t *);
+STATIC int
+xlog_commit_record(
+	struct xlog		*log,
+	struct xlog_ticket	*ticket,
+	struct xlog_in_core	**iclog,
+	xfs_lsn_t		*commitlsnp);
+
 STATIC xlog_t *  xlog_alloc_log(xfs_mount_t	*mp,
 				xfs_buftarg_t	*log_target,
 				xfs_daddr_t	blk_offset,
 				int		num_bblks);
-STATIC int	 xlog_space_left(struct log *log, atomic64_t *head);
+STATIC int
+xlog_space_left(
+	struct xlog		*log,
+	atomic64_t		*head);
 STATIC int	 xlog_sync(xlog_t *log, xlog_in_core_t *iclog);
 STATIC void	 xlog_dealloc_log(xlog_t *log);
 
@@ -67,21 +72,20 @@ STATIC void xlog_state_switch_iclogs(xlog_t		*log,
 				     int		eventual_size);
 STATIC void xlog_state_want_sync(xlog_t	*log, xlog_in_core_t *iclog);
 
-/* local functions to manipulate grant head */
-STATIC int  xlog_grant_log_space(xlog_t		*log,
-				 xlog_ticket_t	*xtic);
-STATIC void xlog_grant_push_ail(struct log	*log,
-				int		need_bytes);
+STATIC void
+xlog_grant_push_ail(
+	struct xlog	*log,
+	int		need_bytes);
 STATIC void xlog_regrant_reserve_log_space(xlog_t	 *log,
 					   xlog_ticket_t *ticket);
-STATIC int xlog_regrant_write_log_space(xlog_t		*log,
-					 xlog_ticket_t  *ticket);
 STATIC void xlog_ungrant_log_space(xlog_t	 *log,
 				   xlog_ticket_t *ticket);
 
 #if defined(DEBUG)
 STATIC void	xlog_verify_dest_ptr(xlog_t *log, char *ptr);
-STATIC void	xlog_verify_grant_tail(struct log *log);
+STATIC void
+xlog_verify_grant_tail(
+	struct xlog	*log);
 STATIC void	xlog_verify_iclog(xlog_t *log, xlog_in_core_t *iclog,
 				  int count, boolean_t syncing);
 STATIC void	xlog_verify_tail_lsn(xlog_t *log, xlog_in_core_t *iclog,
@@ -97,9 +101,9 @@ STATIC int	xlog_iclogs_empty(xlog_t *log);
 
 static void
 xlog_grant_sub_space(
-	struct log	*log,
-	atomic64_t	*head,
-	int		bytes)
+	struct xlog		*log,
+	atomic64_t		*head,
+	int			bytes)
 {
 	int64_t	head_val = atomic64_read(head);
 	int64_t new, old;
@@ -123,9 +127,9 @@ xlog_grant_sub_space(
 
 static void
 xlog_grant_add_space(
-	struct log	*log,
-	atomic64_t	*head,
-	int		bytes)
+	struct xlog		*log,
+	atomic64_t		*head,
+	int			bytes)
 {
 	int64_t	head_val = atomic64_read(head);
 	int64_t new, old;
@@ -150,78 +154,93 @@ xlog_grant_add_space(
 	} while (head_val != old);
 }
 
-STATIC bool
-xlog_reserveq_wake(
-	struct log		*log,
-	int			*free_bytes)
+STATIC void
+xlog_grant_head_init(
+	struct xlog_grant_head	*head)
+{
+	xlog_assign_grant_head(&head->grant, 1, 0);
+	INIT_LIST_HEAD(&head->waiters);
+	spin_lock_init(&head->lock);
+}
+
+STATIC void
+xlog_grant_head_wake_all(
+	struct xlog_grant_head	*head)
 {
 	struct xlog_ticket	*tic;
-	int			need_bytes;
 
-	list_for_each_entry(tic, &log->l_reserveq, t_queue) {
+	spin_lock(&head->lock);
+	list_for_each_entry(tic, &head->waiters, t_queue)
+		wake_up_process(tic->t_task);
+	spin_unlock(&head->lock);
+}
+
+static inline int
+xlog_ticket_reservation(
+	struct xlog		*log,
+	struct xlog_grant_head	*head,
+	struct xlog_ticket	*tic)
+{
+	if (head == &log->l_write_head) {
+		ASSERT(tic->t_flags & XLOG_TIC_PERM_RESERV);
+		return tic->t_unit_res;
+	} else {
 		if (tic->t_flags & XLOG_TIC_PERM_RESERV)
-			need_bytes = tic->t_unit_res * tic->t_cnt;
+			return tic->t_unit_res * tic->t_cnt;
 		else
-			need_bytes = tic->t_unit_res;
-
-		if (*free_bytes < need_bytes)
-			return false;
-		*free_bytes -= need_bytes;
-
-		trace_xfs_log_grant_wake_up(log, tic);
-		wake_up(&tic->t_wait);
+			return tic->t_unit_res;
 	}
-
-	return true;
 }
 
 STATIC bool
-xlog_writeq_wake(
-	struct log		*log,
+xlog_grant_head_wake(
+	struct xlog		*log,
+	struct xlog_grant_head	*head,
 	int			*free_bytes)
 {
 	struct xlog_ticket	*tic;
 	int			need_bytes;
 
-	list_for_each_entry(tic, &log->l_writeq, t_queue) {
-		ASSERT(tic->t_flags & XLOG_TIC_PERM_RESERV);
-
-		need_bytes = tic->t_unit_res;
-
+	list_for_each_entry(tic, &head->waiters, t_queue) {
+		need_bytes = xlog_ticket_reservation(log, head, tic);
 		if (*free_bytes < need_bytes)
 			return false;
-		*free_bytes -= need_bytes;
 
-		trace_xfs_log_regrant_write_wake_up(log, tic);
-		wake_up(&tic->t_wait);
+		*free_bytes -= need_bytes;
+		trace_xfs_log_grant_wake_up(log, tic);
+		wake_up_process(tic->t_task);
 	}
 
 	return true;
 }
 
 STATIC int
-xlog_reserveq_wait(
-	struct log		*log,
+xlog_grant_head_wait(
+	struct xlog		*log,
+	struct xlog_grant_head	*head,
 	struct xlog_ticket	*tic,
 	int			need_bytes)
 {
-	list_add_tail(&tic->t_queue, &log->l_reserveq);
+	list_add_tail(&tic->t_queue, &head->waiters);
 
 	do {
 		if (XLOG_FORCED_SHUTDOWN(log))
 			goto shutdown;
 		xlog_grant_push_ail(log, need_bytes);
 
+		__set_current_state(TASK_UNINTERRUPTIBLE);
+		spin_unlock(&head->lock);
+
 		XFS_STATS_INC(xs_sleep_logspace);
-		trace_xfs_log_grant_sleep(log, tic);
 
-		xlog_wait(&tic->t_wait, &log->l_grant_reserve_lock);
+		trace_xfs_log_grant_sleep(log, tic);
+		schedule();
 		trace_xfs_log_grant_wake(log, tic);
 
-		spin_lock(&log->l_grant_reserve_lock);
+		spin_lock(&head->lock);
 		if (XLOG_FORCED_SHUTDOWN(log))
 			goto shutdown;
-	} while (xlog_space_left(log, &log->l_grant_reserve_head) < need_bytes);
+	} while (xlog_space_left(log, &head->grant) < need_bytes);
 
 	list_del_init(&tic->t_queue);
 	return 0;
@@ -230,35 +249,58 @@ shutdown:
 	return XFS_ERROR(EIO);
 }
 
+/*
+ * Atomically get the log space required for a log ticket.
+ *
+ * Once a ticket gets put onto head->waiters, it will only return after the
+ * needed reservation is satisfied.
+ *
+ * This function is structured so that it has a lock free fast path. This is
+ * necessary because every new transaction reservation will come through this
+ * path. Hence any lock will be globally hot if we take it unconditionally on
+ * every pass.
+ *
+ * As tickets are only ever moved on and off head->waiters under head->lock, we
+ * only need to take that lock if we are going to add the ticket to the queue
+ * and sleep. We can avoid taking the lock if the ticket was never added to
+ * head->waiters because the t_queue list head will be empty and we hold the
+ * only reference to it so it can safely be checked unlocked.
+ */
 STATIC int
-xlog_writeq_wait(
-	struct log		*log,
+xlog_grant_head_check(
+	struct xlog		*log,
+	struct xlog_grant_head	*head,
 	struct xlog_ticket	*tic,
-	int			need_bytes)
+	int			*need_bytes)
 {
-	list_add_tail(&tic->t_queue, &log->l_writeq);
-
-	do {
-		if (XLOG_FORCED_SHUTDOWN(log))
-			goto shutdown;
-		xlog_grant_push_ail(log, need_bytes);
-
-		XFS_STATS_INC(xs_sleep_logspace);
-		trace_xfs_log_regrant_write_sleep(log, tic);
+	int			free_bytes;
+	int			error = 0;
 
-		xlog_wait(&tic->t_wait, &log->l_grant_write_lock);
-		trace_xfs_log_regrant_write_wake(log, tic);
+	ASSERT(!(log->l_flags & XLOG_ACTIVE_RECOVERY));
 
-		spin_lock(&log->l_grant_write_lock);
-		if (XLOG_FORCED_SHUTDOWN(log))
-			goto shutdown;
-	} while (xlog_space_left(log, &log->l_grant_write_head) < need_bytes);
+	/*
+	 * If there are other waiters on the queue then give them a chance at
+	 * logspace before us.  Wake up the first waiters, if we do not wake
+	 * up all the waiters then go to sleep waiting for more free space,
+	 * otherwise try to get some space for this transaction.
+	 */
+	*need_bytes = xlog_ticket_reservation(log, head, tic);
+	free_bytes = xlog_space_left(log, &head->grant);
+	if (!list_empty_careful(&head->waiters)) {
+		spin_lock(&head->lock);
+		if (!xlog_grant_head_wake(log, head, &free_bytes) ||
+		    free_bytes < *need_bytes) {
+			error = xlog_grant_head_wait(log, head, tic,
+						     *need_bytes);
+		}
+		spin_unlock(&head->lock);
+	} else if (free_bytes < *need_bytes) {
+		spin_lock(&head->lock);
+		error = xlog_grant_head_wait(log, head, tic, *need_bytes);
+		spin_unlock(&head->lock);
+	}
 
-	list_del_init(&tic->t_queue);
-	return 0;
-shutdown:
-	list_del_init(&tic->t_queue);
-	return XFS_ERROR(EIO);
+	return error;
 }
 
 static void
@@ -286,6 +328,128 @@ xlog_tic_add_region(xlog_ticket_t *tic, uint len, uint type)
 }
 
 /*
+ * Replenish the byte reservation required by moving the grant write head.
+ */
+int
+xfs_log_regrant(
+	struct xfs_mount	*mp,
+	struct xlog_ticket	*tic)
+{
+	struct xlog		*log = mp->m_log;
+	int			need_bytes;
+	int			error = 0;
+
+	if (XLOG_FORCED_SHUTDOWN(log))
+		return XFS_ERROR(EIO);
+
+	XFS_STATS_INC(xs_try_logspace);
+
+	/*
+	 * This is a new transaction on the ticket, so we need to change the
+	 * transaction ID so that the next transaction has a different TID in
+	 * the log. Just add one to the existing tid so that we can see chains
+	 * of rolling transactions in the log easily.
+	 */
+	tic->t_tid++;
+
+	xlog_grant_push_ail(log, tic->t_unit_res);
+
+	tic->t_curr_res = tic->t_unit_res;
+	xlog_tic_reset_res(tic);
+
+	if (tic->t_cnt > 0)
+		return 0;
+
+	trace_xfs_log_regrant(log, tic);
+
+	error = xlog_grant_head_check(log, &log->l_write_head, tic,
+				      &need_bytes);
+	if (error)
+		goto out_error;
+
+	xlog_grant_add_space(log, &log->l_write_head.grant, need_bytes);
+	trace_xfs_log_regrant_exit(log, tic);
+	xlog_verify_grant_tail(log);
+	return 0;
+
+out_error:
+	/*
+	 * If we are failing, make sure the ticket doesn't have any current
+	 * reservations.  We don't want to add this back when the ticket/
+	 * transaction gets cancelled.
+	 */
+	tic->t_curr_res = 0;
+	tic->t_cnt = 0;	/* ungrant will give back unit_res * t_cnt. */
+	return error;
+}
+
+/*
+ * Reserve log space and return a ticket corresponding the reservation.
+ *
+ * Each reservation is going to reserve extra space for a log record header.
+ * When writes happen to the on-disk log, we don't subtract the length of the
+ * log record header from any reservation.  By wasting space in each
+ * reservation, we prevent over allocation problems.
+ */
+int
+xfs_log_reserve(
+	struct xfs_mount	*mp,
+	int		 	unit_bytes,
+	int		 	cnt,
+	struct xlog_ticket	**ticp,
+	__uint8_t	 	client,
+	bool			permanent,
+	uint		 	t_type)
+{
+	struct xlog		*log = mp->m_log;
+	struct xlog_ticket	*tic;
+	int			need_bytes;
+	int			error = 0;
+
+	ASSERT(client == XFS_TRANSACTION || client == XFS_LOG);
+
+	if (XLOG_FORCED_SHUTDOWN(log))
+		return XFS_ERROR(EIO);
+
+	XFS_STATS_INC(xs_try_logspace);
+
+	ASSERT(*ticp == NULL);
+	tic = xlog_ticket_alloc(log, unit_bytes, cnt, client, permanent,
+				KM_SLEEP | KM_MAYFAIL);
+	if (!tic)
+		return XFS_ERROR(ENOMEM);
+
+	tic->t_trans_type = t_type;
+	*ticp = tic;
+
+	xlog_grant_push_ail(log, tic->t_unit_res * tic->t_cnt);
+
+	trace_xfs_log_reserve(log, tic);
+
+	error = xlog_grant_head_check(log, &log->l_reserve_head, tic,
+				      &need_bytes);
+	if (error)
+		goto out_error;
+
+	xlog_grant_add_space(log, &log->l_reserve_head.grant, need_bytes);
+	xlog_grant_add_space(log, &log->l_write_head.grant, need_bytes);
+	trace_xfs_log_reserve_exit(log, tic);
+	xlog_verify_grant_tail(log);
+	return 0;
+
+out_error:
+	/*
+	 * If we are failing, make sure the ticket doesn't have any current
+	 * reservations.  We don't want to add this back when the ticket/
+	 * transaction gets cancelled.
+	 */
+	tic->t_curr_res = 0;
+	tic->t_cnt = 0;	/* ungrant will give back unit_res * t_cnt. */
+	return error;
+}
+
+
+/*
  * NOTES:
  *
  *	1. currblock field gets updated at startup and after in-core logs
@@ -313,7 +477,7 @@ xfs_log_done(
 	struct xlog_in_core	**iclog,
 	uint			flags)
 {
-	struct log		*log = mp->m_log;
+	struct xlog		*log = mp->m_log;
 	xfs_lsn_t		lsn = 0;
 
 	if (XLOG_FORCED_SHUTDOWN(log) ||
@@ -395,88 +559,6 @@ xfs_log_release_iclog(
 }
 
 /*
- *  1. Reserve an amount of on-disk log space and return a ticket corresponding
- *	to the reservation.
- *  2. Potentially, push buffers at tail of log to disk.
- *
- * Each reservation is going to reserve extra space for a log record header.
- * When writes happen to the on-disk log, we don't subtract the length of the
- * log record header from any reservation.  By wasting space in each
- * reservation, we prevent over allocation problems.
- */
-int
-xfs_log_reserve(
-	struct xfs_mount	*mp,
-	int		 	unit_bytes,
-	int		 	cnt,
-	struct xlog_ticket	**ticket,
-	__uint8_t	 	client,
-	uint		 	flags,
-	uint		 	t_type)
-{
-	struct log		*log = mp->m_log;
-	struct xlog_ticket	*internal_ticket;
-	int			retval = 0;
-
-	ASSERT(client == XFS_TRANSACTION || client == XFS_LOG);
-
-	if (XLOG_FORCED_SHUTDOWN(log))
-		return XFS_ERROR(EIO);
-
-	XFS_STATS_INC(xs_try_logspace);
-
-
-	if (*ticket != NULL) {
-		ASSERT(flags & XFS_LOG_PERM_RESERV);
-		internal_ticket = *ticket;
-
-		/*
-		 * this is a new transaction on the ticket, so we need to
-		 * change the transaction ID so that the next transaction has a
-		 * different TID in the log. Just add one to the existing tid
-		 * so that we can see chains of rolling transactions in the log
-		 * easily.
-		 */
-		internal_ticket->t_tid++;
-
-		trace_xfs_log_reserve(log, internal_ticket);
-
-		xlog_grant_push_ail(log, internal_ticket->t_unit_res);
-		retval = xlog_regrant_write_log_space(log, internal_ticket);
-	} else {
-		/* may sleep if need to allocate more tickets */
-		internal_ticket = xlog_ticket_alloc(log, unit_bytes, cnt,
-						  client, flags,
-						  KM_SLEEP|KM_MAYFAIL);
-		if (!internal_ticket)
-			return XFS_ERROR(ENOMEM);
-		internal_ticket->t_trans_type = t_type;
-		*ticket = internal_ticket;
-
-		trace_xfs_log_reserve(log, internal_ticket);
-
-		xlog_grant_push_ail(log,
-				    (internal_ticket->t_unit_res *
-				     internal_ticket->t_cnt));
-		retval = xlog_grant_log_space(log, internal_ticket);
-	}
-
-	if (unlikely(retval)) {
-		/*
-		 * If we are failing, make sure the ticket doesn't have any
-		 * current reservations.  We don't want to add this back
-		 * when the ticket/ transaction gets cancelled.
-		 */
-		internal_ticket->t_curr_res = 0;
-		/* ungrant will give back unit_res * t_cnt. */
-		internal_ticket->t_cnt = 0;
-	}
-
-	return retval;
-}
-
-
-/*
  * Mount a log filesystem
  *
  * mp		- ubiquitous xfs mount point structure
@@ -653,8 +735,9 @@ xfs_log_unmount_write(xfs_mount_t *mp)
 				.lv_iovecp = &reg,
 			};
 
-			/* remove inited flag */
+			/* remove inited flag, and account for space used */
 			tic->t_flags = 0;
+			tic->t_curr_res -= sizeof(magic);
 			error = xlog_write(log, &vec, tic, &lsn,
 					   NULL, XLOG_UNMOUNT_TRANS);
 			/*
@@ -739,6 +822,7 @@ xfs_log_unmount_write(xfs_mount_t *mp)
 void
 xfs_log_unmount(xfs_mount_t *mp)
 {
+	cancel_delayed_work_sync(&mp->m_sync_work);
 	xfs_trans_ail_destroy(mp);
 	xlog_dealloc_log(mp->m_log);
 }
@@ -761,95 +845,34 @@ xfs_log_item_init(
 }
 
 /*
- * Write region vectors to log.  The write happens using the space reservation
- * of the ticket (tic).  It is not a requirement that all writes for a given
- * transaction occur with one call to xfs_log_write(). However, it is important
- * to note that the transaction reservation code makes an assumption about the
- * number of log headers a transaction requires that may be violated if you
- * don't pass all the transaction vectors in one call....
+ * Wake up processes waiting for log space after we have moved the log tail.
  */
-int
-xfs_log_write(
-	struct xfs_mount	*mp,
-	struct xfs_log_iovec	reg[],
-	int			nentries,
-	struct xlog_ticket	*tic,
-	xfs_lsn_t		*start_lsn)
-{
-	struct log		*log = mp->m_log;
-	int			error;
-	struct xfs_log_vec	vec = {
-		.lv_niovecs = nentries,
-		.lv_iovecp = reg,
-	};
-
-	if (XLOG_FORCED_SHUTDOWN(log))
-		return XFS_ERROR(EIO);
-
-	error = xlog_write(log, &vec, tic, start_lsn, NULL, 0);
-	if (error)
-		xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
-	return error;
-}
-
 void
-xfs_log_move_tail(xfs_mount_t	*mp,
-		  xfs_lsn_t	tail_lsn)
+xfs_log_space_wake(
+	struct xfs_mount	*mp)
 {
-	xlog_ticket_t	*tic;
-	xlog_t		*log = mp->m_log;
-	int		need_bytes, free_bytes;
+	struct xlog		*log = mp->m_log;
+	int			free_bytes;
 
 	if (XLOG_FORCED_SHUTDOWN(log))
 		return;
 
-	if (tail_lsn == 0)
-		tail_lsn = atomic64_read(&log->l_last_sync_lsn);
+	if (!list_empty_careful(&log->l_write_head.waiters)) {
+		ASSERT(!(log->l_flags & XLOG_ACTIVE_RECOVERY));
 
-	/* tail_lsn == 1 implies that we weren't passed a valid value.  */
-	if (tail_lsn != 1)
-		atomic64_set(&log->l_tail_lsn, tail_lsn);
-
-	if (!list_empty_careful(&log->l_writeq)) {
-#ifdef DEBUG
-		if (log->l_flags & XLOG_ACTIVE_RECOVERY)
-			panic("Recovery problem");
-#endif
-		spin_lock(&log->l_grant_write_lock);
-		free_bytes = xlog_space_left(log, &log->l_grant_write_head);
-		list_for_each_entry(tic, &log->l_writeq, t_queue) {
-			ASSERT(tic->t_flags & XLOG_TIC_PERM_RESERV);
-
-			if (free_bytes < tic->t_unit_res && tail_lsn != 1)
-				break;
-			tail_lsn = 0;
-			free_bytes -= tic->t_unit_res;
-			trace_xfs_log_regrant_write_wake_up(log, tic);
-			wake_up(&tic->t_wait);
-		}
-		spin_unlock(&log->l_grant_write_lock);
+		spin_lock(&log->l_write_head.lock);
+		free_bytes = xlog_space_left(log, &log->l_write_head.grant);
+		xlog_grant_head_wake(log, &log->l_write_head, &free_bytes);
+		spin_unlock(&log->l_write_head.lock);
 	}
 
-	if (!list_empty_careful(&log->l_reserveq)) {
-#ifdef DEBUG
-		if (log->l_flags & XLOG_ACTIVE_RECOVERY)
-			panic("Recovery problem");
-#endif
-		spin_lock(&log->l_grant_reserve_lock);
-		free_bytes = xlog_space_left(log, &log->l_grant_reserve_head);
-		list_for_each_entry(tic, &log->l_reserveq, t_queue) {
-			if (tic->t_flags & XLOG_TIC_PERM_RESERV)
-				need_bytes = tic->t_unit_res*tic->t_cnt;
-			else
-				need_bytes = tic->t_unit_res;
-			if (free_bytes < need_bytes && tail_lsn != 1)
-				break;
-			tail_lsn = 0;
-			free_bytes -= need_bytes;
-			trace_xfs_log_grant_wake_up(log, tic);
-			wake_up(&tic->t_wait);
-		}
-		spin_unlock(&log->l_grant_reserve_lock);
+	if (!list_empty_careful(&log->l_reserve_head.waiters)) {
+		ASSERT(!(log->l_flags & XLOG_ACTIVE_RECOVERY));
+
+		spin_lock(&log->l_reserve_head.lock);
+		free_bytes = xlog_space_left(log, &log->l_reserve_head.grant);
+		xlog_grant_head_wake(log, &log->l_reserve_head, &free_bytes);
+		spin_unlock(&log->l_reserve_head.lock);
 	}
 }
 
@@ -899,38 +922,46 @@ xfs_log_need_covered(xfs_mount_t *mp)
 	return needed;
 }
 
-/******************************************************************************
- *
- *	local routines
- *
- ******************************************************************************
- */
-
-/* xfs_trans_tail_ail returns 0 when there is nothing in the list.
- * The log manager must keep track of the last LR which was committed
- * to disk.  The lsn of this LR will become the new tail_lsn whenever
- * xfs_trans_tail_ail returns 0.  If we don't do this, we run into
- * the situation where stuff could be written into the log but nothing
- * was ever in the AIL when asked.  Eventually, we panic since the
- * tail hits the head.
- *
+/*
  * We may be holding the log iclog lock upon entering this routine.
  */
 xfs_lsn_t
-xlog_assign_tail_lsn(
+xlog_assign_tail_lsn_locked(
 	struct xfs_mount	*mp)
 {
+	struct xlog		*log = mp->m_log;
+	struct xfs_log_item	*lip;
 	xfs_lsn_t		tail_lsn;
-	struct log		*log = mp->m_log;
 
-	tail_lsn = xfs_ail_min_lsn(mp->m_ail);
-	if (!tail_lsn)
-		tail_lsn = atomic64_read(&log->l_last_sync_lsn);
+	assert_spin_locked(&mp->m_ail->xa_lock);
 
+	/*
+	 * To make sure we always have a valid LSN for the log tail we keep
+	 * track of the last LSN which was committed in log->l_last_sync_lsn,
+	 * and use that when the AIL was empty.
+	 */
+	lip = xfs_ail_min(mp->m_ail);
+	if (lip)
+		tail_lsn = lip->li_lsn;
+	else
+		tail_lsn = atomic64_read(&log->l_last_sync_lsn);
 	atomic64_set(&log->l_tail_lsn, tail_lsn);
 	return tail_lsn;
 }
 
+xfs_lsn_t
+xlog_assign_tail_lsn(
+	struct xfs_mount	*mp)
+{
+	xfs_lsn_t		tail_lsn;
+
+	spin_lock(&mp->m_ail->xa_lock);
+	tail_lsn = xlog_assign_tail_lsn_locked(mp);
+	spin_unlock(&mp->m_ail->xa_lock);
+
+	return tail_lsn;
+}
+
 /*
  * Return the space in the log between the tail and the head.  The head
  * is passed in the cycle/bytes formal parms.  In the special case where
@@ -947,7 +978,7 @@ xlog_assign_tail_lsn(
  */
 STATIC int
 xlog_space_left(
-	struct log	*log,
+	struct xlog	*log,
 	atomic64_t	*head)
 {
 	int		free_bytes;
@@ -1132,12 +1163,9 @@ xlog_alloc_log(xfs_mount_t	*mp,
 	xlog_assign_atomic_lsn(&log->l_tail_lsn, 1, 0);
 	xlog_assign_atomic_lsn(&log->l_last_sync_lsn, 1, 0);
 	log->l_curr_cycle  = 1;	    /* 0 is bad since this is initial value */
-	xlog_assign_grant_head(&log->l_grant_reserve_head, 1, 0);
-	xlog_assign_grant_head(&log->l_grant_write_head, 1, 0);
-	INIT_LIST_HEAD(&log->l_reserveq);
-	INIT_LIST_HEAD(&log->l_writeq);
-	spin_lock_init(&log->l_grant_reserve_lock);
-	spin_lock_init(&log->l_grant_write_lock);
+
+	xlog_grant_head_init(&log->l_reserve_head);
+	xlog_grant_head_init(&log->l_write_head);
 
 	error = EFSCORRUPTED;
 	if (xfs_sb_version_hassector(&mp->m_sb)) {
@@ -1169,7 +1197,7 @@ xlog_alloc_log(xfs_mount_t	*mp,
 	xlog_get_iclog_buffer_size(mp, log);
 
 	error = ENOMEM;
-	bp = xfs_buf_alloc(mp->m_logdev_targp, 0, log->l_iclog_size, 0);
+	bp = xfs_buf_alloc(mp->m_logdev_targp, 0, BTOBB(log->l_iclog_size), 0);
 	if (!bp)
 		goto out_free_log;
 	bp->b_iodone = xlog_iodone;
@@ -1179,9 +1207,6 @@ xlog_alloc_log(xfs_mount_t	*mp,
 	spin_lock_init(&log->l_icloglock);
 	init_waitqueue_head(&log->l_flush_wait);
 
-	/* log record size must be multiple of BBSIZE; see xlog_rec_header_t */
-	ASSERT((XFS_BUF_SIZE(bp) & BBMASK) == 0);
-
 	iclogp = &log->l_iclog;
 	/*
 	 * The amount of memory to allocate for the iclog structure is
@@ -1201,7 +1226,7 @@ xlog_alloc_log(xfs_mount_t	*mp,
 		prev_iclog = iclog;
 
 		bp = xfs_buf_get_uncached(mp->m_logdev_targp,
-						log->l_iclog_size, 0);
+						BTOBB(log->l_iclog_size), 0);
 		if (!bp)
 			goto out_free_iclog;
 
@@ -1221,7 +1246,7 @@ xlog_alloc_log(xfs_mount_t	*mp,
 		head->h_fmt = cpu_to_be32(XLOG_FMT);
 		memcpy(&head->h_fs_uuid, &mp->m_sb.sb_uuid, sizeof(uuid_t));
 
-		iclog->ic_size = XFS_BUF_SIZE(bp) - log->l_iclog_hsize;
+		iclog->ic_size = BBTOB(bp->b_length) - log->l_iclog_hsize;
 		iclog->ic_state = XLOG_STATE_ACTIVE;
 		iclog->ic_log = log;
 		atomic_set(&iclog->ic_refcnt, 0);
@@ -1265,7 +1290,7 @@ out:
  */
 STATIC int
 xlog_commit_record(
-	struct log		*log,
+	struct xlog		*log,
 	struct xlog_ticket	*ticket,
 	struct xlog_in_core	**iclog,
 	xfs_lsn_t		*commitlsnp)
@@ -1299,7 +1324,7 @@ xlog_commit_record(
  */
 STATIC void
 xlog_grant_push_ail(
-	struct log	*log,
+	struct xlog	*log,
 	int		need_bytes)
 {
 	xfs_lsn_t	threshold_lsn = 0;
@@ -1312,7 +1337,7 @@ xlog_grant_push_ail(
 
 	ASSERT(BTOBB(need_bytes) < log->l_logBBsize);
 
-	free_bytes = xlog_space_left(log, &log->l_grant_reserve_head);
+	free_bytes = xlog_space_left(log, &log->l_reserve_head.grant);
 	free_blocks = BTOBBT(free_bytes);
 
 	/*
@@ -1444,8 +1469,8 @@ xlog_sync(xlog_t		*log,
 		 roundoff < BBTOB(1)));
 
 	/* move grant heads by roundoff in sync */
-	xlog_grant_add_space(log, &log->l_grant_reserve_head, roundoff);
-	xlog_grant_add_space(log, &log->l_grant_write_head, roundoff);
+	xlog_grant_add_space(log, &log->l_reserve_head.grant, roundoff);
+	xlog_grant_add_space(log, &log->l_write_head.grant, roundoff);
 
 	/* put cycle number in every block */
 	xlog_pack_data(log, iclog, roundoff); 
@@ -1472,7 +1497,7 @@ xlog_sync(xlog_t		*log,
 	} else {
 		iclog->ic_bwritecnt = 1;
 	}
-	XFS_BUF_SET_COUNT(bp, count);
+	bp->b_io_length = BTOBB(count);
 	bp->b_fspriv = iclog;
 	XFS_BUF_ZEROFLAGS(bp);
 	XFS_BUF_ASYNC(bp);
@@ -1570,7 +1595,7 @@ xlog_dealloc_log(xlog_t *log)
 	 * always need to ensure that the extra buffer does not point to memory
 	 * owned by another log buffer before we free it.
 	 */
-	xfs_buf_set_empty(log->l_xbuf, log->l_iclog_size);
+	xfs_buf_set_empty(log->l_xbuf, BTOBB(log->l_iclog_size));
 	xfs_buf_free(log->l_xbuf);
 
 	iclog = log->l_iclog;
@@ -1685,7 +1710,7 @@ xlog_print_tic_res(
 	};
 
 	xfs_warn(mp,
-		"xfs_log_write: reservation summary:\n"
+		"xlog_write: reservation summary:\n"
 		"  trans type  = %s (%u)\n"
 		"  unit res    = %d bytes\n"
 		"  current res = %d bytes\n"
@@ -1714,7 +1739,7 @@ xlog_print_tic_res(
 	}
 
 	xfs_alert_tag(mp, XFS_PTAG_LOGRES,
-		"xfs_log_write: reservation ran out. Need to up reservation");
+		"xlog_write: reservation ran out. Need to up reservation");
 	xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
 }
 
@@ -1778,7 +1803,7 @@ xlog_write_start_rec(
 
 static xlog_op_header_t *
 xlog_write_setup_ophdr(
-	struct log		*log,
+	struct xlog		*log,
 	struct xlog_op_header	*ophdr,
 	struct xlog_ticket	*ticket,
 	uint			flags)
@@ -1861,7 +1886,7 @@ xlog_write_setup_copy(
 
 static int
 xlog_write_copy_finish(
-	struct log		*log,
+	struct xlog		*log,
 	struct xlog_in_core	*iclog,
 	uint			flags,
 	int			*record_cnt,
@@ -1946,7 +1971,7 @@ xlog_write_copy_finish(
  */
 int
 xlog_write(
-	struct log		*log,
+	struct xlog		*log,
 	struct xfs_log_vec	*log_vector,
 	struct xlog_ticket	*ticket,
 	xfs_lsn_t		*start_lsn,
@@ -1968,23 +1993,21 @@ xlog_write(
 	*start_lsn = 0;
 
 	len = xlog_write_calc_vec_length(ticket, log_vector);
-	if (log->l_cilp) {
-		/*
-		 * Region headers and bytes are already accounted for.
-		 * We only need to take into account start records and
-		 * split regions in this function.
-		 */
-		if (ticket->t_flags & XLOG_TIC_INITED)
-			ticket->t_curr_res -= sizeof(xlog_op_header_t);
 
-		/*
-		 * Commit record headers need to be accounted for. These
-		 * come in as separate writes so are easy to detect.
-		 */
-		if (flags & (XLOG_COMMIT_TRANS | XLOG_UNMOUNT_TRANS))
-			ticket->t_curr_res -= sizeof(xlog_op_header_t);
-	} else
-		ticket->t_curr_res -= len;
+	/*
+	 * Region headers and bytes are already accounted for.
+	 * We only need to take into account start records and
+	 * split regions in this function.
+	 */
+	if (ticket->t_flags & XLOG_TIC_INITED)
+		ticket->t_curr_res -= sizeof(xlog_op_header_t);
+
+	/*
+	 * Commit record headers need to be accounted for. These
+	 * come in as separate writes so are easy to detect.
+	 */
+	if (flags & (XLOG_COMMIT_TRANS | XLOG_UNMOUNT_TRANS))
+		ticket->t_curr_res -= sizeof(xlog_op_header_t);
 
 	if (ticket->t_curr_res < 0)
 		xlog_print_tic_res(log->l_mp, ticket);
@@ -2600,119 +2623,6 @@ restart:
 	return 0;
 }	/* xlog_state_get_iclog_space */
 
-/*
- * Atomically get the log space required for a log ticket.
- *
- * Once a ticket gets put onto the reserveq, it will only return after the
- * needed reservation is satisfied.
- *
- * This function is structured so that it has a lock free fast path. This is
- * necessary because every new transaction reservation will come through this
- * path. Hence any lock will be globally hot if we take it unconditionally on
- * every pass.
- *
- * As tickets are only ever moved on and off the reserveq under the
- * l_grant_reserve_lock, we only need to take that lock if we are going to add
- * the ticket to the queue and sleep. We can avoid taking the lock if the ticket
- * was never added to the reserveq because the t_queue list head will be empty
- * and we hold the only reference to it so it can safely be checked unlocked.
- */
-STATIC int
-xlog_grant_log_space(
-	struct log		*log,
-	struct xlog_ticket	*tic)
-{
-	int			free_bytes, need_bytes;
-	int			error = 0;
-
-	ASSERT(!(log->l_flags & XLOG_ACTIVE_RECOVERY));
-
-	trace_xfs_log_grant_enter(log, tic);
-
-	/*
-	 * If there are other waiters on the queue then give them a chance at
-	 * logspace before us.  Wake up the first waiters, if we do not wake
-	 * up all the waiters then go to sleep waiting for more free space,
-	 * otherwise try to get some space for this transaction.
-	 */
-	need_bytes = tic->t_unit_res;
-	if (tic->t_flags & XFS_LOG_PERM_RESERV)
-		need_bytes *= tic->t_ocnt;
-	free_bytes = xlog_space_left(log, &log->l_grant_reserve_head);
-	if (!list_empty_careful(&log->l_reserveq)) {
-		spin_lock(&log->l_grant_reserve_lock);
-		if (!xlog_reserveq_wake(log, &free_bytes) ||
-		    free_bytes < need_bytes)
-			error = xlog_reserveq_wait(log, tic, need_bytes);
-		spin_unlock(&log->l_grant_reserve_lock);
-	} else if (free_bytes < need_bytes) {
-		spin_lock(&log->l_grant_reserve_lock);
-		error = xlog_reserveq_wait(log, tic, need_bytes);
-		spin_unlock(&log->l_grant_reserve_lock);
-	}
-	if (error)
-		return error;
-
-	xlog_grant_add_space(log, &log->l_grant_reserve_head, need_bytes);
-	xlog_grant_add_space(log, &log->l_grant_write_head, need_bytes);
-	trace_xfs_log_grant_exit(log, tic);
-	xlog_verify_grant_tail(log);
-	return 0;
-}
-
-/*
- * Replenish the byte reservation required by moving the grant write head.
- *
- * Similar to xlog_grant_log_space, the function is structured to have a lock
- * free fast path.
- */
-STATIC int
-xlog_regrant_write_log_space(
-	struct log		*log,
-	struct xlog_ticket	*tic)
-{
-	int			free_bytes, need_bytes;
-	int			error = 0;
-
-	tic->t_curr_res = tic->t_unit_res;
-	xlog_tic_reset_res(tic);
-
-	if (tic->t_cnt > 0)
-		return 0;
-
-	ASSERT(!(log->l_flags & XLOG_ACTIVE_RECOVERY));
-
-	trace_xfs_log_regrant_write_enter(log, tic);
-
-	/*
-	 * If there are other waiters on the queue then give them a chance at
-	 * logspace before us.  Wake up the first waiters, if we do not wake
-	 * up all the waiters then go to sleep waiting for more free space,
-	 * otherwise try to get some space for this transaction.
-	 */
-	need_bytes = tic->t_unit_res;
-	free_bytes = xlog_space_left(log, &log->l_grant_write_head);
-	if (!list_empty_careful(&log->l_writeq)) {
-		spin_lock(&log->l_grant_write_lock);
-		if (!xlog_writeq_wake(log, &free_bytes) ||
-		    free_bytes < need_bytes)
-			error = xlog_writeq_wait(log, tic, need_bytes);
-		spin_unlock(&log->l_grant_write_lock);
-	} else if (free_bytes < need_bytes) {
-		spin_lock(&log->l_grant_write_lock);
-		error = xlog_writeq_wait(log, tic, need_bytes);
-		spin_unlock(&log->l_grant_write_lock);
-	}
-
-	if (error)
-		return error;
-
-	xlog_grant_add_space(log, &log->l_grant_write_head, need_bytes);
-	trace_xfs_log_regrant_write_exit(log, tic);
-	xlog_verify_grant_tail(log);
-	return 0;
-}
-
 /* The first cnt-1 times through here we don't need to
  * move the grant write head because the permanent
  * reservation has reserved cnt times the unit amount.
@@ -2729,9 +2639,9 @@ xlog_regrant_reserve_log_space(xlog_t	     *log,
 	if (ticket->t_cnt > 0)
 		ticket->t_cnt--;
 
-	xlog_grant_sub_space(log, &log->l_grant_reserve_head,
+	xlog_grant_sub_space(log, &log->l_reserve_head.grant,
 					ticket->t_curr_res);
-	xlog_grant_sub_space(log, &log->l_grant_write_head,
+	xlog_grant_sub_space(log, &log->l_write_head.grant,
 					ticket->t_curr_res);
 	ticket->t_curr_res = ticket->t_unit_res;
 	xlog_tic_reset_res(ticket);
@@ -2742,7 +2652,7 @@ xlog_regrant_reserve_log_space(xlog_t	     *log,
 	if (ticket->t_cnt > 0)
 		return;
 
-	xlog_grant_add_space(log, &log->l_grant_reserve_head,
+	xlog_grant_add_space(log, &log->l_reserve_head.grant,
 					ticket->t_unit_res);
 
 	trace_xfs_log_regrant_reserve_exit(log, ticket);
@@ -2788,14 +2698,13 @@ xlog_ungrant_log_space(xlog_t	     *log,
 		bytes += ticket->t_unit_res*ticket->t_cnt;
 	}
 
-	xlog_grant_sub_space(log, &log->l_grant_reserve_head, bytes);
-	xlog_grant_sub_space(log, &log->l_grant_write_head, bytes);
+	xlog_grant_sub_space(log, &log->l_reserve_head.grant, bytes);
+	xlog_grant_sub_space(log, &log->l_write_head.grant, bytes);
 
 	trace_xfs_log_ungrant_exit(log, ticket);
 
-	xfs_log_move_tail(log->l_mp, 1);
-}	/* xlog_ungrant_log_space */
-
+	xfs_log_space_wake(log->l_mp);
+}
 
 /*
  * Flush iclog to disk if this is the last reference to the given iclog and
@@ -2925,14 +2834,13 @@ _xfs_log_force(
 	uint			flags,
 	int			*log_flushed)
 {
-	struct log		*log = mp->m_log;
+	struct xlog		*log = mp->m_log;
 	struct xlog_in_core	*iclog;
 	xfs_lsn_t		lsn;
 
 	XFS_STATS_INC(xs_log_force);
 
-	if (log->l_cilp)
-		xlog_cil_force(log);
+	xlog_cil_force(log);
 
 	spin_lock(&log->l_icloglock);
 
@@ -3046,6 +2954,7 @@ xfs_log_force(
 {
 	int	error;
 
+	trace_xfs_log_force(mp, 0);
 	error = _xfs_log_force(mp, flags, NULL);
 	if (error)
 		xfs_warn(mp, "%s: error %d returned.", __func__, error);
@@ -3073,7 +2982,7 @@ _xfs_log_force_lsn(
 	uint			flags,
 	int			*log_flushed)
 {
-	struct log		*log = mp->m_log;
+	struct xlog		*log = mp->m_log;
 	struct xlog_in_core	*iclog;
 	int			already_slept = 0;
 
@@ -3081,11 +2990,9 @@ _xfs_log_force_lsn(
 
 	XFS_STATS_INC(xs_log_force);
 
-	if (log->l_cilp) {
-		lsn = xlog_cil_force_lsn(log, lsn);
-		if (lsn == NULLCOMMITLSN)
-			return 0;
-	}
+	lsn = xlog_cil_force_lsn(log, lsn);
+	if (lsn == NULLCOMMITLSN)
+		return 0;
 
 try_again:
 	spin_lock(&log->l_icloglock);
@@ -3196,6 +3103,7 @@ xfs_log_force_lsn(
 {
 	int	error;
 
+	trace_xfs_log_force(mp, lsn);
 	error = _xfs_log_force_lsn(mp, lsn, flags, NULL);
 	if (error)
 		xfs_warn(mp, "%s: error %d returned.", __func__, error);
@@ -3252,12 +3160,12 @@ xfs_log_ticket_get(
  */
 xlog_ticket_t *
 xlog_ticket_alloc(
-	struct log	*log,
+	struct xlog	*log,
 	int		unit_bytes,
 	int		cnt,
 	char		client,
-	uint		xflags,
-	int		alloc_flags)
+	bool		permanent,
+	xfs_km_flags_t	alloc_flags)
 {
 	struct xlog_ticket *tic;
 	uint		num_headers;
@@ -3350,6 +3258,7 @@ xlog_ticket_alloc(
         }
 
 	atomic_set(&tic->t_ref, 1);
+	tic->t_task		= current;
 	INIT_LIST_HEAD(&tic->t_queue);
 	tic->t_unit_res		= unit_bytes;
 	tic->t_curr_res		= unit_bytes;
@@ -3359,9 +3268,8 @@ xlog_ticket_alloc(
 	tic->t_clientid		= client;
 	tic->t_flags		= XLOG_TIC_INITED;
 	tic->t_trans_type	= 0;
-	if (xflags & XFS_LOG_PERM_RESERV)
+	if (permanent)
 		tic->t_flags |= XLOG_TIC_PERM_RESERV;
-	init_waitqueue_head(&tic->t_wait);
 
 	xlog_tic_reset_res(tic);
 
@@ -3383,7 +3291,7 @@ xlog_ticket_alloc(
  */
 void
 xlog_verify_dest_ptr(
-	struct log	*log,
+	struct xlog	*log,
 	char		*ptr)
 {
 	int i;
@@ -3412,12 +3320,12 @@ xlog_verify_dest_ptr(
  */
 STATIC void
 xlog_verify_grant_tail(
-	struct log	*log)
+	struct xlog	*log)
 {
 	int		tail_cycle, tail_blocks;
 	int		cycle, space;
 
-	xlog_crack_grant_head(&log->l_grant_write_head, &cycle, &space);
+	xlog_crack_grant_head(&log->l_write_head.grant, &cycle, &space);
 	xlog_crack_atomic_lsn(&log->l_tail_lsn, &tail_cycle, &tail_blocks);
 	if (tail_cycle != cycle) {
 		if (cycle - 1 != tail_cycle &&
@@ -3619,7 +3527,6 @@ xfs_log_force_umount(
 	struct xfs_mount	*mp,
 	int			logerror)
 {
-	xlog_ticket_t	*tic;
 	xlog_t		*log;
 	int		retval;
 
@@ -3653,7 +3560,7 @@ xfs_log_force_umount(
 	 * completed transactions are flushed to disk with the xfs_log_force()
 	 * call below.
 	 */
-	if (!logerror && (mp->m_flags & XFS_MOUNT_DELAYLOG))
+	if (!logerror)
 		xlog_cil_force(log);
 
 	/*
@@ -3687,15 +3594,8 @@ xfs_log_force_umount(
 	 * we don't enqueue anything once the SHUTDOWN flag is set, and this
 	 * action is protected by the grant locks.
 	 */
-	spin_lock(&log->l_grant_reserve_lock);
-	list_for_each_entry(tic, &log->l_reserveq, t_queue)
-		wake_up(&tic->t_wait);
-	spin_unlock(&log->l_grant_reserve_lock);
-
-	spin_lock(&log->l_grant_write_lock);
-	list_for_each_entry(tic, &log->l_writeq, t_queue)
-		wake_up(&tic->t_wait);
-	spin_unlock(&log->l_grant_write_lock);
+	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);
diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h
index 3f7bf451c03..748d312850e 100644
--- a/fs/xfs/xfs_log.h
+++ b/fs/xfs/xfs_log.h
@@ -53,15 +53,6 @@ static inline xfs_lsn_t	_lsn_cmp(xfs_lsn_t lsn1, xfs_lsn_t lsn2)
 #define XFS_LOG_REL_PERM_RESERV	0x1
 
 /*
- * Flags to xfs_log_reserve()
- *
- *	XFS_LOG_PERM_RESERV: Permanent reservation.  When writes are
- *		performed against this type of reservation, the reservation
- *		is not decreased.  Long running transactions should use this.
- */
-#define XFS_LOG_PERM_RESERV	0x2
-
-/*
  * Flags to xfs_log_force()
  *
  *	XFS_LOG_SYNC:	Synchronous force in-core log to disk
@@ -160,8 +151,9 @@ int	  xfs_log_mount(struct xfs_mount	*mp,
 			xfs_daddr_t		start_block,
 			int		 	num_bblocks);
 int	  xfs_log_mount_finish(struct xfs_mount *mp);
-void	  xfs_log_move_tail(struct xfs_mount	*mp,
-			    xfs_lsn_t		tail_lsn);
+xfs_lsn_t xlog_assign_tail_lsn(struct xfs_mount *mp);
+xfs_lsn_t xlog_assign_tail_lsn_locked(struct xfs_mount *mp);
+void	  xfs_log_space_wake(struct xfs_mount *mp);
 int	  xfs_log_notify(struct xfs_mount	*mp,
 			 struct xlog_in_core	*iclog,
 			 xfs_log_callback_t	*callback_entry);
@@ -172,13 +164,9 @@ int	  xfs_log_reserve(struct xfs_mount *mp,
 			  int		   count,
 			  struct xlog_ticket **ticket,
 			  __uint8_t	   clientid,
-			  uint		   flags,
+			  bool		   permanent,
 			  uint		   t_type);
-int	  xfs_log_write(struct xfs_mount *mp,
-			xfs_log_iovec_t  region[],
-			int		 nentries,
-			struct xlog_ticket *ticket,
-			xfs_lsn_t	 *start_lsn);
+int	  xfs_log_regrant(struct xfs_mount *mp, struct xlog_ticket *tic);
 int	  xfs_log_unmount_write(struct xfs_mount *mp);
 void      xfs_log_unmount(struct xfs_mount *mp);
 int	  xfs_log_force_umount(struct xfs_mount *mp, int logerror);
@@ -189,8 +177,7 @@ void	  xlog_iodone(struct xfs_buf *);
 struct xlog_ticket *xfs_log_ticket_get(struct xlog_ticket *ticket);
 void	  xfs_log_ticket_put(struct xlog_ticket *ticket);
 
-void	xfs_log_commit_cil(struct xfs_mount *mp, struct xfs_trans *tp,
-				struct xfs_log_vec *log_vector,
+int	xfs_log_commit_cil(struct xfs_mount *mp, struct xfs_trans *tp,
 				xfs_lsn_t *commit_lsn, int flags);
 bool	xfs_log_item_in_current_chkpt(struct xfs_log_item *lip);
 
diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c
index c7755d5a5fb..ddc4529d07d 100644
--- a/fs/xfs/xfs_log_cil.c
+++ b/fs/xfs/xfs_log_cil.c
@@ -18,9 +18,7 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_types.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_trans_priv.h"
 #include "xfs_log_priv.h"
@@ -29,71 +27,10 @@
 #include "xfs_mount.h"
 #include "xfs_error.h"
 #include "xfs_alloc.h"
+#include "xfs_extent_busy.h"
 #include "xfs_discard.h"
 
 /*
- * Perform initial CIL structure initialisation. If the CIL is not
- * enabled in this filesystem, ensure the log->l_cilp is null so
- * we can check this conditional to determine if we are doing delayed
- * logging or not.
- */
-int
-xlog_cil_init(
-	struct log	*log)
-{
-	struct xfs_cil	*cil;
-	struct xfs_cil_ctx *ctx;
-
-	log->l_cilp = NULL;
-	if (!(log->l_mp->m_flags & XFS_MOUNT_DELAYLOG))
-		return 0;
-
-	cil = kmem_zalloc(sizeof(*cil), KM_SLEEP|KM_MAYFAIL);
-	if (!cil)
-		return ENOMEM;
-
-	ctx = kmem_zalloc(sizeof(*ctx), KM_SLEEP|KM_MAYFAIL);
-	if (!ctx) {
-		kmem_free(cil);
-		return ENOMEM;
-	}
-
-	INIT_LIST_HEAD(&cil->xc_cil);
-	INIT_LIST_HEAD(&cil->xc_committing);
-	spin_lock_init(&cil->xc_cil_lock);
-	init_rwsem(&cil->xc_ctx_lock);
-	init_waitqueue_head(&cil->xc_commit_wait);
-
-	INIT_LIST_HEAD(&ctx->committing);
-	INIT_LIST_HEAD(&ctx->busy_extents);
-	ctx->sequence = 1;
-	ctx->cil = cil;
-	cil->xc_ctx = ctx;
-	cil->xc_current_sequence = ctx->sequence;
-
-	cil->xc_log = log;
-	log->l_cilp = cil;
-	return 0;
-}
-
-void
-xlog_cil_destroy(
-	struct log	*log)
-{
-	if (!log->l_cilp)
-		return;
-
-	if (log->l_cilp->xc_ctx) {
-		if (log->l_cilp->xc_ctx->ticket)
-			xfs_log_ticket_put(log->l_cilp->xc_ctx->ticket);
-		kmem_free(log->l_cilp->xc_ctx);
-	}
-
-	ASSERT(list_empty(&log->l_cilp->xc_cil));
-	kmem_free(log->l_cilp);
-}
-
-/*
  * Allocate a new ticket. Failing to get a new ticket makes it really hard to
  * recover, so we don't allow failure here. Also, we allocate in a context that
  * we don't want to be issuing transactions from, so we need to tell the
@@ -107,7 +44,7 @@ xlog_cil_destroy(
  */
 static struct xlog_ticket *
 xlog_cil_ticket_alloc(
-	struct log	*log)
+	struct xlog	*log)
 {
 	struct xlog_ticket *tic;
 
@@ -135,11 +72,8 @@ xlog_cil_ticket_alloc(
  */
 void
 xlog_cil_init_post_recovery(
-	struct log	*log)
+	struct xlog	*log)
 {
-	if (!log->l_cilp)
-		return;
-
 	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,
@@ -172,37 +106,73 @@ xlog_cil_init_post_recovery(
  * format the regions into the iclog as though they are being formatted
  * directly out of the objects themselves.
  */
-static void
-xlog_cil_format_items(
-	struct log		*log,
-	struct xfs_log_vec	*log_vector)
+static struct xfs_log_vec *
+xlog_cil_prepare_log_vecs(
+	struct xfs_trans	*tp)
 {
-	struct xfs_log_vec *lv;
+	struct xfs_log_item_desc *lidp;
+	struct xfs_log_vec	*lv = NULL;
+	struct xfs_log_vec	*ret_lv = NULL;
 
-	ASSERT(log_vector);
-	for (lv = log_vector; lv; lv = lv->lv_next) {
+
+	/* Bail out if we didn't find a log item.  */
+	if (list_empty(&tp->t_items)) {
+		ASSERT(0);
+		return NULL;
+	}
+
+	list_for_each_entry(lidp, &tp->t_items, lid_trans) {
+		struct xfs_log_vec *new_lv;
 		void	*ptr;
 		int	index;
 		int	len = 0;
+		uint	niovecs;
+
+		/* Skip items which aren't dirty in this transaction. */
+		if (!(lidp->lid_flags & XFS_LID_DIRTY))
+			continue;
+
+		/* Skip items that do not have any vectors for writing */
+		niovecs = IOP_SIZE(lidp->lid_item);
+		if (!niovecs)
+			continue;
+
+		new_lv = kmem_zalloc(sizeof(*new_lv) +
+				niovecs * sizeof(struct xfs_log_iovec),
+				KM_SLEEP);
+
+		/* The allocated iovec region lies beyond the log vector. */
+		new_lv->lv_iovecp = (struct xfs_log_iovec *)&new_lv[1];
+		new_lv->lv_niovecs = niovecs;
+		new_lv->lv_item = lidp->lid_item;
 
 		/* build the vector array and calculate it's length */
-		IOP_FORMAT(lv->lv_item, lv->lv_iovecp);
-		for (index = 0; index < lv->lv_niovecs; index++)
-			len += lv->lv_iovecp[index].i_len;
+		IOP_FORMAT(new_lv->lv_item, new_lv->lv_iovecp);
+		for (index = 0; index < new_lv->lv_niovecs; index++)
+			len += new_lv->lv_iovecp[index].i_len;
 
-		lv->lv_buf_len = len;
-		lv->lv_buf = kmem_alloc(lv->lv_buf_len, KM_SLEEP|KM_NOFS);
-		ptr = lv->lv_buf;
+		new_lv->lv_buf_len = len;
+		new_lv->lv_buf = kmem_alloc(new_lv->lv_buf_len,
+				KM_SLEEP|KM_NOFS);
+		ptr = new_lv->lv_buf;
 
-		for (index = 0; index < lv->lv_niovecs; index++) {
-			struct xfs_log_iovec *vec = &lv->lv_iovecp[index];
+		for (index = 0; index < new_lv->lv_niovecs; index++) {
+			struct xfs_log_iovec *vec = &new_lv->lv_iovecp[index];
 
 			memcpy(ptr, vec->i_addr, vec->i_len);
 			vec->i_addr = ptr;
 			ptr += vec->i_len;
 		}
-		ASSERT(ptr == lv->lv_buf + lv->lv_buf_len);
+		ASSERT(ptr == new_lv->lv_buf + new_lv->lv_buf_len);
+
+		if (!ret_lv)
+			ret_lv = new_lv;
+		else
+			lv->lv_next = new_lv;
+		lv = new_lv;
 	}
+
+	return ret_lv;
 }
 
 /*
@@ -212,7 +182,7 @@ xlog_cil_format_items(
  */
 STATIC void
 xfs_cil_prepare_item(
-	struct log		*log,
+	struct xlog		*log,
 	struct xfs_log_vec	*lv,
 	int			*len,
 	int			*diff_iovecs)
@@ -256,12 +226,12 @@ xfs_cil_prepare_item(
  * Insert the log items into the CIL and calculate the difference in space
  * consumed by the item. Add the space to the checkpoint ticket and calculate
  * if the change requires additional log metadata. If it does, take that space
- * as well. Remove the amount of space we addded to the checkpoint ticket from
+ * as well. Remove the amount of space we added to the checkpoint ticket from
  * the current transaction ticket so that the accounting works out correctly.
  */
 static void
 xlog_cil_insert_items(
-	struct log		*log,
+	struct xlog		*log,
 	struct xfs_log_vec	*log_vector,
 	struct xlog_ticket	*ticket)
 {
@@ -367,8 +337,8 @@ xlog_cil_committed(
 	xfs_trans_committed_bulk(ctx->cil->xc_log->l_ailp, ctx->lv_chain,
 					ctx->start_lsn, abort);
 
-	xfs_alloc_busy_sort(&ctx->busy_extents);
-	xfs_alloc_busy_clear(mp, &ctx->busy_extents,
+	xfs_extent_busy_sort(&ctx->busy_extents);
+	xfs_extent_busy_clear(mp, &ctx->busy_extents,
 			     (mp->m_flags & XFS_MOUNT_DISCARD) && !abort);
 
 	spin_lock(&ctx->cil->xc_cil_lock);
@@ -381,7 +351,7 @@ xlog_cil_committed(
 		ASSERT(mp->m_flags & XFS_MOUNT_DISCARD);
 
 		xfs_discard_extents(mp, &ctx->busy_extents);
-		xfs_alloc_busy_clear(mp, &ctx->busy_extents, false);
+		xfs_extent_busy_clear(mp, &ctx->busy_extents, false);
 	}
 
 	kmem_free(ctx);
@@ -403,8 +373,7 @@ xlog_cil_committed(
  */
 STATIC int
 xlog_cil_push(
-	struct log		*log,
-	xfs_lsn_t		push_seq)
+	struct xlog		*log)
 {
 	struct xfs_cil		*cil = log->l_cilp;
 	struct xfs_log_vec	*lv;
@@ -420,39 +389,36 @@ xlog_cil_push(
 	struct xfs_log_iovec	lhdr;
 	struct xfs_log_vec	lvhdr = { NULL };
 	xfs_lsn_t		commit_lsn;
+	xfs_lsn_t		push_seq;
 
 	if (!cil)
 		return 0;
 
-	ASSERT(!push_seq || push_seq <= cil->xc_ctx->sequence);
-
 	new_ctx = kmem_zalloc(sizeof(*new_ctx), KM_SLEEP|KM_NOFS);
 	new_ctx->ticket = xlog_cil_ticket_alloc(log);
 
-	/*
-	 * Lock out transaction commit, but don't block for background pushes
-	 * unless we are well over the CIL space limit. See the definition of
-	 * XLOG_CIL_HARD_SPACE_LIMIT() for the full explanation of the logic
-	 * used here.
-	 */
-	if (!down_write_trylock(&cil->xc_ctx_lock)) {
-		if (!push_seq &&
-		    cil->xc_ctx->space_used < XLOG_CIL_HARD_SPACE_LIMIT(log))
-			goto out_free_ticket;
-		down_write(&cil->xc_ctx_lock);
-	}
+	down_write(&cil->xc_ctx_lock);
 	ctx = cil->xc_ctx;
 
-	/* check if we've anything to push */
-	if (list_empty(&cil->xc_cil))
-		goto out_skip;
+	spin_lock(&cil->xc_cil_lock);
+	push_seq = cil->xc_push_seq;
+	ASSERT(push_seq <= ctx->sequence);
 
-	/* check for spurious background flush */
-	if (!push_seq && cil->xc_ctx->space_used < XLOG_CIL_SPACE_LIMIT(log))
+	/*
+	 * Check if we've anything to push. If there is nothing, then we don't
+	 * move on to a new sequence number and so we have to be able to push
+	 * this sequence again later.
+	 */
+	if (list_empty(&cil->xc_cil)) {
+		cil->xc_push_seq = 0;
+		spin_unlock(&cil->xc_cil_lock);
 		goto out_skip;
+	}
+	spin_unlock(&cil->xc_cil_lock);
+
 
 	/* check for a previously pushed seqeunce */
-	if (push_seq && push_seq < cil->xc_ctx->sequence)
+	if (push_seq < cil->xc_ctx->sequence)
 		goto out_skip;
 
 	/*
@@ -606,7 +572,6 @@ restart:
 
 out_skip:
 	up_write(&cil->xc_ctx_lock);
-out_free_ticket:
 	xfs_log_ticket_put(new_ctx->ticket);
 	kmem_free(new_ctx);
 	return 0;
@@ -618,6 +583,82 @@ out_abort:
 	return XFS_ERROR(EIO);
 }
 
+static void
+xlog_cil_push_work(
+	struct work_struct	*work)
+{
+	struct xfs_cil		*cil = container_of(work, struct xfs_cil,
+							xc_push_work);
+	xlog_cil_push(cil->xc_log);
+}
+
+/*
+ * We need to push CIL every so often so we don't cache more than we can fit in
+ * the log. The limit really is that a checkpoint can't be more than half the
+ * log (the current checkpoint is not allowed to overwrite the previous
+ * checkpoint), but commit latency and memory usage limit this to a smaller
+ * size.
+ */
+static void
+xlog_cil_push_background(
+	struct xlog	*log)
+{
+	struct xfs_cil	*cil = log->l_cilp;
+
+	/*
+	 * The cil won't be empty because we are called while holding the
+	 * context lock so whatever we added to the CIL will still be there
+	 */
+	ASSERT(!list_empty(&cil->xc_cil));
+
+	/*
+	 * don't do a background push if we haven't used up all the
+	 * space available yet.
+	 */
+	if (cil->xc_ctx->space_used < XLOG_CIL_SPACE_LIMIT(log))
+		return;
+
+	spin_lock(&cil->xc_cil_lock);
+	if (cil->xc_push_seq < cil->xc_current_sequence) {
+		cil->xc_push_seq = cil->xc_current_sequence;
+		queue_work(log->l_mp->m_cil_workqueue, &cil->xc_push_work);
+	}
+	spin_unlock(&cil->xc_cil_lock);
+
+}
+
+static void
+xlog_cil_push_foreground(
+	struct xlog	*log,
+	xfs_lsn_t	push_seq)
+{
+	struct xfs_cil	*cil = log->l_cilp;
+
+	if (!cil)
+		return;
+
+	ASSERT(push_seq && push_seq <= cil->xc_current_sequence);
+
+	/* start on any pending background push to minimise wait time on it */
+	flush_work(&cil->xc_push_work);
+
+	/*
+	 * If the CIL is empty or we've already pushed the sequence then
+	 * there's no work we need to do.
+	 */
+	spin_lock(&cil->xc_cil_lock);
+	if (list_empty(&cil->xc_cil) || push_seq <= cil->xc_push_seq) {
+		spin_unlock(&cil->xc_cil_lock);
+		return;
+	}
+
+	cil->xc_push_seq = push_seq;
+	spin_unlock(&cil->xc_cil_lock);
+
+	/* do the push now */
+	xlog_cil_push(log);
+}
+
 /*
  * Commit a transaction with the given vector to the Committed Item List.
  *
@@ -635,28 +676,29 @@ out_abort:
  * background commit, returns without it held once background commits are
  * allowed again.
  */
-void
+int
 xfs_log_commit_cil(
 	struct xfs_mount	*mp,
 	struct xfs_trans	*tp,
-	struct xfs_log_vec	*log_vector,
 	xfs_lsn_t		*commit_lsn,
 	int			flags)
 {
-	struct log		*log = mp->m_log;
+	struct xlog		*log = mp->m_log;
 	int			log_flags = 0;
-	int			push = 0;
+	struct xfs_log_vec	*log_vector;
 
 	if (flags & XFS_TRANS_RELEASE_LOG_RES)
 		log_flags = XFS_LOG_REL_PERM_RESERV;
 
 	/*
-	 * do all the hard work of formatting items (including memory
+	 * Do all the hard work of formatting items (including memory
 	 * allocation) outside the CIL context lock. This prevents stalling CIL
 	 * pushes when we are low on memory and a transaction commit spends a
 	 * lot of time in memory reclaim.
 	 */
-	xlog_cil_format_items(log, log_vector);
+	log_vector = xlog_cil_prepare_log_vecs(tp);
+	if (!log_vector)
+		return ENOMEM;
 
 	/* lock out background commit */
 	down_read(&log->l_cilp->xc_ctx_lock);
@@ -694,21 +736,10 @@ xfs_log_commit_cil(
 	 */
 	xfs_trans_free_items(tp, *commit_lsn, 0);
 
-	/* check for background commit before unlock */
-	if (log->l_cilp->xc_ctx->space_used > XLOG_CIL_SPACE_LIMIT(log))
-		push = 1;
+	xlog_cil_push_background(log);
 
 	up_read(&log->l_cilp->xc_ctx_lock);
-
-	/*
-	 * We need to push CIL every so often so we don't cache more than we
-	 * can fit in the log. The limit really is that a checkpoint can't be
-	 * more than half the log (the current checkpoint is not allowed to
-	 * overwrite the previous checkpoint), but commit latency and memory
-	 * usage limit this to a smaller size in most cases.
-	 */
-	if (push)
-		xlog_cil_push(log, 0);
+	return 0;
 }
 
 /*
@@ -720,13 +751,10 @@ xfs_log_commit_cil(
  *
  * We return the current commit lsn to allow the callers to determine if a
  * iclog flush is necessary following this call.
- *
- * XXX: Initially, just push the CIL unconditionally and return whatever
- * commit lsn is there. It'll be empty, so this is broken for now.
  */
 xfs_lsn_t
 xlog_cil_force_lsn(
-	struct log	*log,
+	struct xlog	*log,
 	xfs_lsn_t	sequence)
 {
 	struct xfs_cil		*cil = log->l_cilp;
@@ -740,8 +768,7 @@ xlog_cil_force_lsn(
 	 * xlog_cil_push() handles racing pushes for the same sequence,
 	 * so no need to deal with it here.
 	 */
-	if (sequence == cil->xc_current_sequence)
-		xlog_cil_push(log, sequence);
+	xlog_cil_push_foreground(log, sequence);
 
 	/*
 	 * See if we can find a previous sequence still committing.
@@ -786,8 +813,6 @@ xfs_log_item_in_current_chkpt(
 {
 	struct xfs_cil_ctx *ctx;
 
-	if (!(lip->li_mountp->m_flags & XFS_MOUNT_DELAYLOG))
-		return false;
 	if (list_empty(&lip->li_cil))
 		return false;
 
@@ -802,3 +827,57 @@ xfs_log_item_in_current_chkpt(
 		return false;
 	return true;
 }
+
+/*
+ * Perform initial CIL structure initialisation.
+ */
+int
+xlog_cil_init(
+	struct xlog	*log)
+{
+	struct xfs_cil	*cil;
+	struct xfs_cil_ctx *ctx;
+
+	cil = kmem_zalloc(sizeof(*cil), KM_SLEEP|KM_MAYFAIL);
+	if (!cil)
+		return ENOMEM;
+
+	ctx = kmem_zalloc(sizeof(*ctx), KM_SLEEP|KM_MAYFAIL);
+	if (!ctx) {
+		kmem_free(cil);
+		return ENOMEM;
+	}
+
+	INIT_WORK(&cil->xc_push_work, xlog_cil_push_work);
+	INIT_LIST_HEAD(&cil->xc_cil);
+	INIT_LIST_HEAD(&cil->xc_committing);
+	spin_lock_init(&cil->xc_cil_lock);
+	init_rwsem(&cil->xc_ctx_lock);
+	init_waitqueue_head(&cil->xc_commit_wait);
+
+	INIT_LIST_HEAD(&ctx->committing);
+	INIT_LIST_HEAD(&ctx->busy_extents);
+	ctx->sequence = 1;
+	ctx->cil = cil;
+	cil->xc_ctx = ctx;
+	cil->xc_current_sequence = ctx->sequence;
+
+	cil->xc_log = log;
+	log->l_cilp = cil;
+	return 0;
+}
+
+void
+xlog_cil_destroy(
+	struct xlog	*log)
+{
+	if (log->l_cilp->xc_ctx) {
+		if (log->l_cilp->xc_ctx->ticket)
+			xfs_log_ticket_put(log->l_cilp->xc_ctx->ticket);
+		kmem_free(log->l_cilp->xc_ctx);
+	}
+
+	ASSERT(list_empty(&log->l_cilp->xc_cil));
+	kmem_free(log->l_cilp);
+}
+
diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h
index 2d3b6a498d6..72eba2201b1 100644
--- a/fs/xfs/xfs_log_priv.h
+++ b/fs/xfs/xfs_log_priv.h
@@ -19,7 +19,7 @@
 #define __XFS_LOG_PRIV_H__
 
 struct xfs_buf;
-struct log;
+struct xlog;
 struct xlog_ticket;
 struct xfs_mount;
 
@@ -239,8 +239,8 @@ typedef struct xlog_res {
 } xlog_res_t;
 
 typedef struct xlog_ticket {
-	wait_queue_head_t  t_wait;	 /* ticket wait queue */
 	struct list_head   t_queue;	 /* reserve/write queue */
+	struct task_struct *t_task;	 /* task that owns this ticket */
 	xlog_tid_t	   t_tid;	 /* transaction identifier	 : 4  */
 	atomic_t	   t_ref;	 /* ticket reference count       : 4  */
 	int		   t_curr_res;	 /* current reservation in bytes : 4  */
@@ -352,7 +352,7 @@ typedef struct xlog_in_core {
 	struct xlog_in_core	*ic_next;
 	struct xlog_in_core	*ic_prev;
 	struct xfs_buf		*ic_bp;
-	struct log		*ic_log;
+	struct xlog		*ic_log;
 	int			ic_size;
 	int			ic_offset;
 	int			ic_bwritecnt;
@@ -409,7 +409,7 @@ struct xfs_cil_ctx {
  * operations almost as efficient as the old logging methods.
  */
 struct xfs_cil {
-	struct log		*xc_log;
+	struct xlog		*xc_log;
 	struct list_head	xc_cil;
 	spinlock_t		xc_cil_lock;
 	struct xfs_cil_ctx	*xc_ctx;
@@ -417,6 +417,8 @@ struct xfs_cil {
 	struct list_head	xc_committing;
 	wait_queue_head_t	xc_commit_wait;
 	xfs_lsn_t		xc_current_sequence;
+	struct work_struct	xc_push_work;
+	xfs_lsn_t		xc_push_seq;
 };
 
 /*
@@ -470,12 +472,22 @@ struct xfs_cil {
 #define XLOG_CIL_HARD_SPACE_LIMIT(log)	(3 * (log->l_logsize >> 4))
 
 /*
+ * ticket grant locks, queues and accounting have their own cachlines
+ * as these are quite hot and can be operated on concurrently.
+ */
+struct xlog_grant_head {
+	spinlock_t		lock ____cacheline_aligned_in_smp;
+	struct list_head	waiters;
+	atomic64_t		grant;
+};
+
+/*
  * The reservation head lsn is not made up of a cycle number and block number.
  * Instead, it uses a cycle number and byte number.  Logs don't expect to
  * overflow 31 bits worth of byte offset, so using a byte number will mean
  * that round off problems won't occur when releasing partial reservations.
  */
-typedef struct log {
+typedef struct xlog {
 	/* The following fields don't need locking */
 	struct xfs_mount	*l_mp;	        /* mount point */
 	struct xfs_ail		*l_ailp;	/* AIL log is working with */
@@ -520,17 +532,8 @@ typedef struct log {
 	/* lsn of 1st LR with unflushed * buffers */
 	atomic64_t		l_tail_lsn ____cacheline_aligned_in_smp;
 
-	/*
-	 * ticket grant locks, queues and accounting have their own cachlines
-	 * as these are quite hot and can be operated on concurrently.
-	 */
-	spinlock_t		l_grant_reserve_lock ____cacheline_aligned_in_smp;
-	struct list_head	l_reserveq;
-	atomic64_t		l_grant_reserve_head;
-
-	spinlock_t		l_grant_write_lock ____cacheline_aligned_in_smp;
-	struct list_head	l_writeq;
-	atomic64_t		l_grant_write_head;
+	struct xlog_grant_head	l_reserve_head;
+	struct xlog_grant_head	l_write_head;
 
 	/* The following field are used for debugging; need to hold icloglock */
 #ifdef DEBUG
@@ -545,15 +548,19 @@ typedef struct log {
 #define XLOG_FORCED_SHUTDOWN(log)	((log)->l_flags & XLOG_IO_ERROR)
 
 /* common routines */
-extern xfs_lsn_t xlog_assign_tail_lsn(struct xfs_mount *mp);
 extern int	 xlog_recover(xlog_t *log);
 extern int	 xlog_recover_finish(xlog_t *log);
 extern void	 xlog_pack_data(xlog_t *log, xlog_in_core_t *iclog, int);
 
 extern kmem_zone_t *xfs_log_ticket_zone;
-struct xlog_ticket *xlog_ticket_alloc(struct log *log, int unit_bytes,
-				int count, char client, uint xflags,
-				int alloc_flags);
+struct xlog_ticket *
+xlog_ticket_alloc(
+	struct xlog	*log,
+	int		unit_bytes,
+	int		count,
+	char		client,
+	bool		permanent,
+	xfs_km_flags_t	alloc_flags);
 
 
 static inline void
@@ -565,9 +572,14 @@ xlog_write_adv_cnt(void **ptr, int *len, int *off, size_t bytes)
 }
 
 void	xlog_print_tic_res(struct xfs_mount *mp, struct xlog_ticket *ticket);
-int	xlog_write(struct log *log, struct xfs_log_vec *log_vector,
-				struct xlog_ticket *tic, xfs_lsn_t *start_lsn,
-				xlog_in_core_t **commit_iclog, uint flags);
+int
+xlog_write(
+	struct xlog		*log,
+	struct xfs_log_vec	*log_vector,
+	struct xlog_ticket	*tic,
+	xfs_lsn_t		*start_lsn,
+	struct xlog_in_core	**commit_iclog,
+	uint			flags);
 
 /*
  * When we crack an atomic LSN, we sample it first so that the value will not
@@ -627,17 +639,23 @@ xlog_assign_grant_head(atomic64_t *head, int cycle, int space)
 /*
  * Committed Item List interfaces
  */
-int	xlog_cil_init(struct log *log);
-void	xlog_cil_init_post_recovery(struct log *log);
-void	xlog_cil_destroy(struct log *log);
+int
+xlog_cil_init(struct xlog *log);
+void
+xlog_cil_init_post_recovery(struct xlog *log);
+void
+xlog_cil_destroy(struct xlog *log);
 
 /*
  * CIL force routines
  */
-xfs_lsn_t xlog_cil_force_lsn(struct log *log, xfs_lsn_t sequence);
+xfs_lsn_t
+xlog_cil_force_lsn(
+	struct xlog *log,
+	xfs_lsn_t sequence);
 
 static inline void
-xlog_cil_force(struct log *log)
+xlog_cil_force(struct xlog *log)
 {
 	xlog_cil_force_lsn(log, log->l_cilp->xc_current_sequence);
 }
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 541a508adea..a7be98abd6a 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -40,7 +40,6 @@
 #include "xfs_extfree_item.h"
 #include "xfs_trans_priv.h"
 #include "xfs_quota.h"
-#include "xfs_rw.h"
 #include "xfs_utils.h"
 #include "xfs_trace.h"
 
@@ -120,7 +119,7 @@ xlog_get_bp(
 		nbblks += log->l_sectBBsize;
 	nbblks = round_up(nbblks, log->l_sectBBsize);
 
-	bp = xfs_buf_get_uncached(log->l_mp->m_logdev_targp, BBTOB(nbblks), 0);
+	bp = xfs_buf_get_uncached(log->l_mp->m_logdev_targp, nbblks, 0);
 	if (bp)
 		xfs_buf_unlock(bp);
 	return bp;
@@ -146,7 +145,7 @@ xlog_align(
 {
 	xfs_daddr_t	offset = blk_no & ((xfs_daddr_t)log->l_sectBBsize - 1);
 
-	ASSERT(BBTOB(offset + nbblks) <= XFS_BUF_SIZE(bp));
+	ASSERT(offset + nbblks <= bp->b_length);
 	return bp->b_addr + BBTOB(offset);
 }
 
@@ -174,11 +173,12 @@ xlog_bread_noalign(
 	nbblks = round_up(nbblks, log->l_sectBBsize);
 
 	ASSERT(nbblks > 0);
-	ASSERT(BBTOB(nbblks) <= XFS_BUF_SIZE(bp));
+	ASSERT(nbblks <= bp->b_length);
 
 	XFS_BUF_SET_ADDR(bp, log->l_logBBstart + blk_no);
 	XFS_BUF_READ(bp);
-	XFS_BUF_SET_COUNT(bp, BBTOB(nbblks));
+	bp->b_io_length = nbblks;
+	bp->b_error = 0;
 
 	xfsbdstrat(log->l_mp, bp);
 	error = xfs_buf_iowait(bp);
@@ -218,7 +218,7 @@ xlog_bread_offset(
 	xfs_caddr_t	offset)
 {
 	xfs_caddr_t	orig_offset = bp->b_addr;
-	int		orig_len = bp->b_buffer_length;
+	int		orig_len = BBTOB(bp->b_length);
 	int		error, error2;
 
 	error = xfs_buf_associate_memory(bp, offset, BBTOB(nbblks));
@@ -259,13 +259,14 @@ xlog_bwrite(
 	nbblks = round_up(nbblks, log->l_sectBBsize);
 
 	ASSERT(nbblks > 0);
-	ASSERT(BBTOB(nbblks) <= XFS_BUF_SIZE(bp));
+	ASSERT(nbblks <= bp->b_length);
 
 	XFS_BUF_SET_ADDR(bp, log->l_logBBstart + blk_no);
 	XFS_BUF_ZEROFLAGS(bp);
 	xfs_buf_hold(bp);
 	xfs_buf_lock(bp);
-	XFS_BUF_SET_COUNT(bp, BBTOB(nbblks));
+	bp->b_io_length = nbblks;
+	bp->b_error = 0;
 
 	error = xfs_bwrite(bp);
 	if (error)
@@ -440,6 +441,8 @@ xlog_find_verify_cycle(
 	 * a log sector, or we're out of luck.
 	 */
 	bufblks = 1 << ffs(nbblks);
+	while (bufblks > log->l_logBBsize)
+		bufblks >>= 1;
 	while (!(bp = xlog_get_bp(log, bufblks))) {
 		bufblks >>= 1;
 		if (bufblks < log->l_sectBBsize)
@@ -965,9 +968,9 @@ xlog_find_tail(
 		log->l_curr_cycle++;
 	atomic64_set(&log->l_tail_lsn, be64_to_cpu(rhead->h_tail_lsn));
 	atomic64_set(&log->l_last_sync_lsn, be64_to_cpu(rhead->h_lsn));
-	xlog_assign_grant_head(&log->l_grant_reserve_head, log->l_curr_cycle,
+	xlog_assign_grant_head(&log->l_reserve_head.grant, log->l_curr_cycle,
 					BBTOB(log->l_curr_block));
-	xlog_assign_grant_head(&log->l_grant_write_head, log->l_curr_cycle,
+	xlog_assign_grant_head(&log->l_write_head.grant, log->l_curr_cycle,
 					BBTOB(log->l_curr_block));
 
 	/*
@@ -1225,6 +1228,8 @@ xlog_write_log_records(
 	 * log sector, or we're out of luck.
 	 */
 	bufblks = 1 << ffs(blocks);
+	while (bufblks > log->l_logBBsize)
+		bufblks >>= 1;
 	while (!(bp = xlog_get_bp(log, bufblks))) {
 		bufblks >>= 1;
 		if (bufblks < sectbb)
@@ -1466,8 +1471,8 @@ xlog_recover_add_item(
 
 STATIC int
 xlog_recover_add_to_cont_trans(
-	struct log		*log,
-	xlog_recover_t		*trans,
+	struct xlog		*log,
+	struct xlog_recover	*trans,
 	xfs_caddr_t		dp,
 	int			len)
 {
@@ -1489,7 +1494,7 @@ xlog_recover_add_to_cont_trans(
 	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, 0u);
+	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;
@@ -1512,8 +1517,8 @@ xlog_recover_add_to_cont_trans(
  */
 STATIC int
 xlog_recover_add_to_trans(
-	struct log		*log,
-	xlog_recover_t		*trans,
+	struct xlog		*log,
+	struct xlog_recover	*trans,
 	xfs_caddr_t		dp,
 	int			len)
 {
@@ -1583,8 +1588,8 @@ xlog_recover_add_to_trans(
  */
 STATIC int
 xlog_recover_reorder_trans(
-	struct log		*log,
-	xlog_recover_t		*trans,
+	struct xlog		*log,
+	struct xlog_recover	*trans,
 	int			pass)
 {
 	xlog_recover_item_t	*item, *n;
@@ -1637,8 +1642,8 @@ xlog_recover_reorder_trans(
  */
 STATIC int
 xlog_recover_buffer_pass1(
-	struct log		*log,
-	xlog_recover_item_t	*item)
+	struct xlog			*log,
+	struct xlog_recover_item	*item)
 {
 	xfs_buf_log_format_t	*buf_f = item->ri_buf[0].i_addr;
 	struct list_head	*bucket;
@@ -1691,7 +1696,7 @@ xlog_recover_buffer_pass1(
  */
 STATIC int
 xlog_check_buffer_cancelled(
-	struct log		*log,
+	struct xlog		*log,
 	xfs_daddr_t		blkno,
 	uint			len,
 	ushort			flags)
@@ -1772,7 +1777,7 @@ xlog_recover_do_inode_buffer(
 
 	trace_xfs_log_recover_buf_inode_buf(mp->m_log, buf_f);
 
-	inodes_per_buf = XFS_BUF_COUNT(bp) >> mp->m_sb.sb_inodelog;
+	inodes_per_buf = BBTOB(bp->b_io_length) >> mp->m_sb.sb_inodelog;
 	for (i = 0; i < inodes_per_buf; i++) {
 		next_unlinked_offset = (i * mp->m_sb.sb_inodesize) +
 			offsetof(xfs_dinode_t, di_next_unlinked);
@@ -1814,7 +1819,8 @@ xlog_recover_do_inode_buffer(
 
 		ASSERT(item->ri_buf[item_index].i_addr != NULL);
 		ASSERT((item->ri_buf[item_index].i_len % XFS_BLF_CHUNK) == 0);
-		ASSERT((reg_buf_offset + reg_buf_bytes) <= XFS_BUF_COUNT(bp));
+		ASSERT((reg_buf_offset + reg_buf_bytes) <=
+							BBTOB(bp->b_io_length));
 
 		/*
 		 * The current logged region contains a copy of the
@@ -1873,8 +1879,8 @@ xlog_recover_do_reg_buffer(
 		ASSERT(nbits > 0);
 		ASSERT(item->ri_buf[i].i_addr != NULL);
 		ASSERT(item->ri_buf[i].i_len % XFS_BLF_CHUNK == 0);
-		ASSERT(XFS_BUF_COUNT(bp) >=
-		       ((uint)bit << XFS_BLF_SHIFT)+(nbits<<XFS_BLF_SHIFT));
+		ASSERT(BBTOB(bp->b_io_length) >=
+		       ((uint)bit << XFS_BLF_SHIFT) + (nbits << XFS_BLF_SHIFT));
 
 		/*
 		 * Do a sanity check if this is a dquot buffer. Just checking
@@ -1981,7 +1987,7 @@ xfs_qm_dqcheck(
 
 	if (!errs && ddq->d_id) {
 		if (ddq->d_blk_softlimit &&
-		    be64_to_cpu(ddq->d_bcount) >=
+		    be64_to_cpu(ddq->d_bcount) >
 				be64_to_cpu(ddq->d_blk_softlimit)) {
 			if (!ddq->d_btimer) {
 				if (flags & XFS_QMOPT_DOWARN)
@@ -1992,7 +1998,7 @@ xfs_qm_dqcheck(
 			}
 		}
 		if (ddq->d_ino_softlimit &&
-		    be64_to_cpu(ddq->d_icount) >=
+		    be64_to_cpu(ddq->d_icount) >
 				be64_to_cpu(ddq->d_ino_softlimit)) {
 			if (!ddq->d_itimer) {
 				if (flags & XFS_QMOPT_DOWARN)
@@ -2003,7 +2009,7 @@ xfs_qm_dqcheck(
 			}
 		}
 		if (ddq->d_rtb_softlimit &&
-		    be64_to_cpu(ddq->d_rtbcount) >=
+		    be64_to_cpu(ddq->d_rtbcount) >
 				be64_to_cpu(ddq->d_rtb_softlimit)) {
 			if (!ddq->d_rtbtimer) {
 				if (flags & XFS_QMOPT_DOWARN)
@@ -2103,6 +2109,7 @@ xlog_recover_do_dquot_buffer(
 STATIC int
 xlog_recover_buffer_pass2(
 	xlog_t			*log,
+	struct list_head	*buffer_list,
 	xlog_recover_item_t	*item)
 {
 	xfs_buf_log_format_t	*buf_f = item->ri_buf[0].i_addr;
@@ -2123,9 +2130,9 @@ xlog_recover_buffer_pass2(
 
 	trace_xfs_log_recover_buf_recover(log, buf_f);
 
-	buf_flags = XBF_LOCK;
-	if (!(buf_f->blf_flags & XFS_BLF_INODE_BUF))
-		buf_flags |= XBF_MAPPED;
+	buf_flags = 0;
+	if (buf_f->blf_flags & XFS_BLF_INODE_BUF)
+		buf_flags |= XBF_UNMAPPED;
 
 	bp = xfs_buf_read(mp->m_ddev_targp, buf_f->blf_blkno, buf_f->blf_len,
 			  buf_flags);
@@ -2166,14 +2173,14 @@ xlog_recover_buffer_pass2(
 	 */
 	if (XFS_DINODE_MAGIC ==
 	    be16_to_cpu(*((__be16 *)xfs_buf_offset(bp, 0))) &&
-	    (XFS_BUF_COUNT(bp) != MAX(log->l_mp->m_sb.sb_blocksize,
+	    (BBTOB(bp->b_io_length) != MAX(log->l_mp->m_sb.sb_blocksize,
 			(__uint32_t)XFS_INODE_CLUSTER_SIZE(log->l_mp)))) {
 		xfs_buf_stale(bp);
 		error = xfs_bwrite(bp);
 	} else {
 		ASSERT(bp->b_target->bt_mount == mp);
 		bp->b_iodone = xlog_recover_iodone;
-		xfs_buf_delwri_queue(bp);
+		xfs_buf_delwri_queue(bp, buffer_list);
 	}
 
 	xfs_buf_relse(bp);
@@ -2183,6 +2190,7 @@ xlog_recover_buffer_pass2(
 STATIC int
 xlog_recover_inode_pass2(
 	xlog_t			*log,
+	struct list_head	*buffer_list,
 	xlog_recover_item_t	*item)
 {
 	xfs_inode_log_format_t	*in_f;
@@ -2220,8 +2228,7 @@ xlog_recover_inode_pass2(
 	}
 	trace_xfs_log_recover_inode_recover(log, in_f);
 
-	bp = xfs_buf_read(mp->m_ddev_targp, in_f->ilf_blkno, in_f->ilf_len,
-			  XBF_LOCK);
+	bp = xfs_buf_read(mp->m_ddev_targp, in_f->ilf_blkno, in_f->ilf_len, 0);
 	if (!bp) {
 		error = ENOMEM;
 		goto error;
@@ -2436,7 +2443,7 @@ xlog_recover_inode_pass2(
 write_inode_buffer:
 	ASSERT(bp->b_target->bt_mount == mp);
 	bp->b_iodone = xlog_recover_iodone;
-	xfs_buf_delwri_queue(bp);
+	xfs_buf_delwri_queue(bp, buffer_list);
 	xfs_buf_relse(bp);
 error:
 	if (need_free)
@@ -2477,6 +2484,7 @@ xlog_recover_quotaoff_pass1(
 STATIC int
 xlog_recover_dquot_pass2(
 	xlog_t			*log,
+	struct list_head	*buffer_list,
 	xlog_recover_item_t	*item)
 {
 	xfs_mount_t		*mp = log->l_mp;
@@ -2530,14 +2538,11 @@ xlog_recover_dquot_pass2(
 		return XFS_ERROR(EIO);
 	ASSERT(dq_f->qlf_len == 1);
 
-	error = xfs_read_buf(mp, mp->m_ddev_targp,
-			     dq_f->qlf_blkno,
-			     XFS_FSB_TO_BB(mp, dq_f->qlf_len),
-			     0, &bp);
-	if (error) {
-		xfs_buf_ioerror_alert(bp, "xlog_recover_do..(read#3)");
+	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);
+	if (error)
 		return error;
-	}
+
 	ASSERT(bp);
 	ddq = (xfs_disk_dquot_t *)xfs_buf_offset(bp, dq_f->qlf_boffset);
 
@@ -2558,7 +2563,7 @@ xlog_recover_dquot_pass2(
 	ASSERT(dq_f->qlf_size == 2);
 	ASSERT(bp->b_target->bt_mount == mp);
 	bp->b_iodone = xlog_recover_iodone;
-	xfs_buf_delwri_queue(bp);
+	xfs_buf_delwri_queue(bp, buffer_list);
 	xfs_buf_relse(bp);
 
 	return (0);
@@ -2642,7 +2647,8 @@ xlog_recover_efd_pass2(
 				 * xfs_trans_ail_delete() drops the
 				 * AIL lock.
 				 */
-				xfs_trans_ail_delete(ailp, lip);
+				xfs_trans_ail_delete(ailp, lip,
+						     SHUTDOWN_CORRUPT_INCORE);
 				xfs_efi_item_free(efip);
 				spin_lock(&ailp->xa_lock);
 				break;
@@ -2683,9 +2689,9 @@ xlog_recover_free_trans(
 
 STATIC int
 xlog_recover_commit_pass1(
-	struct log		*log,
-	struct xlog_recover	*trans,
-	xlog_recover_item_t	*item)
+	struct xlog			*log,
+	struct xlog_recover		*trans,
+	struct xlog_recover_item	*item)
 {
 	trace_xfs_log_recover_item_recover(log, trans, item, XLOG_RECOVER_PASS1);
 
@@ -2710,23 +2716,24 @@ xlog_recover_commit_pass1(
 
 STATIC int
 xlog_recover_commit_pass2(
-	struct log		*log,
-	struct xlog_recover	*trans,
-	xlog_recover_item_t	*item)
+	struct xlog			*log,
+	struct xlog_recover		*trans,
+	struct list_head		*buffer_list,
+	struct xlog_recover_item	*item)
 {
 	trace_xfs_log_recover_item_recover(log, trans, item, XLOG_RECOVER_PASS2);
 
 	switch (ITEM_TYPE(item)) {
 	case XFS_LI_BUF:
-		return xlog_recover_buffer_pass2(log, item);
+		return xlog_recover_buffer_pass2(log, buffer_list, item);
 	case XFS_LI_INODE:
-		return xlog_recover_inode_pass2(log, item);
+		return xlog_recover_inode_pass2(log, buffer_list, item);
 	case XFS_LI_EFI:
 		return xlog_recover_efi_pass2(log, item, trans->r_lsn);
 	case XFS_LI_EFD:
 		return xlog_recover_efd_pass2(log, item);
 	case XFS_LI_DQUOT:
-		return xlog_recover_dquot_pass2(log, item);
+		return xlog_recover_dquot_pass2(log, buffer_list, item);
 	case XFS_LI_QUOTAOFF:
 		/* nothing to do in pass2 */
 		return 0;
@@ -2746,12 +2753,13 @@ xlog_recover_commit_pass2(
  */
 STATIC int
 xlog_recover_commit_trans(
-	struct log		*log,
+	struct xlog		*log,
 	struct xlog_recover	*trans,
 	int			pass)
 {
-	int			error = 0;
+	int			error = 0, error2;
 	xlog_recover_item_t	*item;
+	LIST_HEAD		(buffer_list);
 
 	hlist_del(&trans->r_list);
 
@@ -2760,22 +2768,33 @@ xlog_recover_commit_trans(
 		return error;
 
 	list_for_each_entry(item, &trans->r_itemq, ri_list) {
-		if (pass == XLOG_RECOVER_PASS1)
+		switch (pass) {
+		case XLOG_RECOVER_PASS1:
 			error = xlog_recover_commit_pass1(log, trans, item);
-		else
-			error = xlog_recover_commit_pass2(log, trans, item);
+			break;
+		case XLOG_RECOVER_PASS2:
+			error = xlog_recover_commit_pass2(log, trans,
+							  &buffer_list, item);
+			break;
+		default:
+			ASSERT(0);
+		}
+
 		if (error)
-			return error;
+			goto out;
 	}
 
 	xlog_recover_free_trans(trans);
-	return 0;
+
+out:
+	error2 = xfs_buf_delwri_submit(&buffer_list);
+	return error ? error : error2;
 }
 
 STATIC int
 xlog_recover_unmount_trans(
-	struct log		*log,
-	xlog_recover_t		*trans)
+	struct xlog		*log,
+	struct xlog_recover	*trans)
 {
 	/* Do nothing now */
 	xfs_warn(log->l_mp, "%s: Unmount LR", __func__);
@@ -3079,7 +3098,7 @@ xlog_recover_process_one_iunlink(
 	/*
 	 * Get the on disk inode to find the next inode in the bucket.
 	 */
-	error = xfs_itobp(mp, NULL, ip, &dip, &ibp, XBF_LOCK);
+	error = xfs_itobp(mp, NULL, ip, &dip, &ibp, 0);
 	if (error)
 		goto fail_iput;
 
@@ -3161,37 +3180,26 @@ xlog_recover_process_iunlinks(
 			 */
 			continue;
 		}
+		/*
+		 * Unlock the buffer so that it can be acquired in the normal
+		 * course of the transaction to truncate and free each inode.
+		 * Because we are not racing with anyone else here for the AGI
+		 * buffer, we don't even need to hold it locked to read the
+		 * initial unlinked bucket entries out of the buffer. We keep
+		 * buffer reference though, so that it stays pinned in memory
+		 * while we need the buffer.
+		 */
 		agi = XFS_BUF_TO_AGI(agibp);
+		xfs_buf_unlock(agibp);
 
 		for (bucket = 0; bucket < XFS_AGI_UNLINKED_BUCKETS; bucket++) {
 			agino = be32_to_cpu(agi->agi_unlinked[bucket]);
 			while (agino != NULLAGINO) {
-				/*
-				 * Release the agi buffer so that it can
-				 * be acquired in the normal course of the
-				 * transaction to truncate and free the inode.
-				 */
-				xfs_buf_relse(agibp);
-
 				agino = xlog_recover_process_one_iunlink(mp,
 							agno, agino, bucket);
-
-				/*
-				 * Reacquire the agibuffer and continue around
-				 * the loop. This should never fail as we know
-				 * the buffer was good earlier on.
-				 */
-				error = xfs_read_agi(mp, NULL, agno, &agibp);
-				ASSERT(error == 0);
-				agi = XFS_BUF_TO_AGI(agibp);
 			}
 		}
-
-		/*
-		 * Release the buffer for the current agi so we can
-		 * go on to the next one.
-		 */
-		xfs_buf_relse(agibp);
+		xfs_buf_rele(agibp);
 	}
 
 	mp->m_dmevmask = mp_dmevmask;
@@ -3650,11 +3658,8 @@ xlog_do_recover(
 	 * First replay the images in the log.
 	 */
 	error = xlog_do_log_recovery(log, head_blk, tail_blk);
-	if (error) {
+	if (error)
 		return error;
-	}
-
-	xfs_flush_buftarg(log->l_mp->m_ddev_targp, 1);
 
 	/*
 	 * If IO errors happened during recovery, bail out.
@@ -3681,7 +3686,6 @@ xlog_do_recover(
 	bp = xfs_getsb(log->l_mp, 0);
 	XFS_BUF_UNDONE(bp);
 	ASSERT(!(XFS_BUF_ISWRITE(bp)));
-	ASSERT(!(XFS_BUF_ISDELAYWRITE(bp)));
 	XFS_BUF_READ(bp);
 	XFS_BUF_UNASYNC(bp);
 	xfsbdstrat(log->l_mp, bp);
@@ -3695,7 +3699,7 @@ xlog_do_recover(
 
 	/* Convert superblock from on-disk format */
 	sbp = &log->l_mp->m_sb;
-	xfs_sb_from_disk(sbp, XFS_BUF_TO_SBP(bp));
+	xfs_sb_from_disk(log->l_mp, XFS_BUF_TO_SBP(bp));
 	ASSERT(sbp->sb_magicnum == XFS_SB_MAGIC);
 	ASSERT(xfs_sb_good_version(sbp));
 	xfs_buf_relse(bp);
diff --git a/fs/xfs/xfs_message.c b/fs/xfs/xfs_message.c
index bd672def95a..331cd9f83a7 100644
--- a/fs/xfs/xfs_message.c
+++ b/fs/xfs/xfs_message.c
@@ -19,7 +19,6 @@
 #include "xfs_fs.h"
 #include "xfs_types.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index d06afbc3540..536021fb3d4 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -22,6 +22,7 @@
 #include "xfs_log.h"
 #include "xfs_inum.h"
 #include "xfs_trans.h"
+#include "xfs_trans_priv.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_dir2.h"
@@ -37,7 +38,6 @@
 #include "xfs_rtalloc.h"
 #include "xfs_bmap.h"
 #include "xfs_error.h"
-#include "xfs_rw.h"
 #include "xfs_quota.h"
 #include "xfs_fsops.h"
 #include "xfs_utils.h"
@@ -158,7 +158,7 @@ xfs_uuid_mount(
 
  out_duplicate:
 	mutex_unlock(&xfs_uuid_table_mutex);
-	xfs_warn(mp, "Filesystem has duplicate UUID - can't mount");
+	xfs_warn(mp, "Filesystem has duplicate UUID %pU - can't mount", uuid);
 	return XFS_ERROR(EINVAL);
 }
 
@@ -553,9 +553,11 @@ out_unwind:
 
 void
 xfs_sb_from_disk(
-	xfs_sb_t	*to,
+	struct xfs_mount	*mp,
 	xfs_dsb_t	*from)
 {
+	struct xfs_sb *to = &mp->m_sb;
+
 	to->sb_magicnum = be32_to_cpu(from->sb_magicnum);
 	to->sb_blocksize = be32_to_cpu(from->sb_blocksize);
 	to->sb_dblocks = be64_to_cpu(from->sb_dblocks);
@@ -681,8 +683,8 @@ xfs_readsb(xfs_mount_t *mp, int flags)
 	sector_size = xfs_getsize_buftarg(mp->m_ddev_targp);
 
 reread:
-	bp = xfs_buf_read_uncached(mp, mp->m_ddev_targp,
-					XFS_SB_DADDR, sector_size, 0);
+	bp = xfs_buf_read_uncached(mp->m_ddev_targp, XFS_SB_DADDR,
+					BTOBB(sector_size), 0);
 	if (!bp) {
 		if (loud)
 			xfs_warn(mp, "SB buffer read failed");
@@ -693,7 +695,7 @@ reread:
 	 * Initialize the mount structure from the superblock.
 	 * But first do some basic consistency checking.
 	 */
-	xfs_sb_from_disk(&mp->m_sb, XFS_BUF_TO_SBP(bp));
+	xfs_sb_from_disk(mp, XFS_BUF_TO_SBP(bp));
 	error = xfs_mount_validate_sb(mp, &(mp->m_sb), flags);
 	if (error) {
 		if (loud)
@@ -1030,9 +1032,9 @@ xfs_check_sizes(xfs_mount_t *mp)
 		xfs_warn(mp, "filesystem size mismatch detected");
 		return XFS_ERROR(EFBIG);
 	}
-	bp = xfs_buf_read_uncached(mp, mp->m_ddev_targp,
+	bp = xfs_buf_read_uncached(mp->m_ddev_targp,
 					d - XFS_FSS_TO_BB(mp, 1),
-					BBTOB(XFS_FSS_TO_BB(mp, 1)), 0);
+					XFS_FSS_TO_BB(mp, 1), 0);
 	if (!bp) {
 		xfs_warn(mp, "last sector read failed");
 		return EIO;
@@ -1045,9 +1047,9 @@ xfs_check_sizes(xfs_mount_t *mp)
 			xfs_warn(mp, "log size mismatch detected");
 			return XFS_ERROR(EFBIG);
 		}
-		bp = xfs_buf_read_uncached(mp, mp->m_logdev_targp,
+		bp = xfs_buf_read_uncached(mp->m_logdev_targp,
 					d - XFS_FSB_TO_BB(mp, 1),
-					XFS_FSB_TO_B(mp, 1), 0);
+					XFS_FSB_TO_BB(mp, 1), 0);
 		if (!bp) {
 			xfs_warn(mp, "log device read failed");
 			return EIO;
@@ -1286,7 +1288,7 @@ xfs_mountfs(
 			      XFS_FSB_TO_BB(mp, sbp->sb_logblocks));
 	if (error) {
 		xfs_warn(mp, "log mount failed");
-		goto out_free_perag;
+		goto out_fail_wait;
 	}
 
 	/*
@@ -1313,7 +1315,7 @@ xfs_mountfs(
 	     !mp->m_sb.sb_inprogress) {
 		error = xfs_initialize_perag_data(mp, sbp->sb_agcount);
 		if (error)
-			goto out_free_perag;
+			goto out_fail_wait;
 	}
 
 	/*
@@ -1437,6 +1439,10 @@ xfs_mountfs(
 	IRELE(rip);
  out_log_dealloc:
 	xfs_log_unmount(mp);
+ out_fail_wait:
+	if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp)
+		xfs_wait_buftarg(mp->m_logdev_targp);
+	xfs_wait_buftarg(mp->m_ddev_targp);
  out_free_perag:
 	xfs_free_perag(mp);
  out_remove_uuid:
@@ -1473,15 +1479,15 @@ xfs_unmountfs(
 	xfs_log_force(mp, XFS_LOG_SYNC);
 
 	/*
-	 * Do a delwri reclaim pass first so that as many dirty inodes are
-	 * queued up for IO as possible. Then flush the buffers before making
-	 * a synchronous path to catch all the remaining inodes are reclaimed.
-	 * This makes the reclaim process as quick as possible by avoiding
-	 * synchronous writeout and blocking on inodes already in the delwri
-	 * state as much as possible.
+	 * Flush all pending changes from the AIL.
+	 */
+	xfs_ail_push_all_sync(mp->m_ail);
+
+	/*
+	 * And reclaim all inodes.  At this point there should be no dirty
+	 * inode, and none should be pinned or locked, but use synchronous
+	 * reclaim just to be sure.
 	 */
-	xfs_reclaim_inodes(mp, 0);
-	xfs_flush_buftarg(mp->m_ddev_targp, 1);
 	xfs_reclaim_inodes(mp, SYNC_WAIT);
 
 	xfs_qm_unmount(mp);
@@ -1517,15 +1523,12 @@ xfs_unmountfs(
 	if (error)
 		xfs_warn(mp, "Unable to update superblock counters. "
 				"Freespace may not be correct on next mount.");
-	xfs_unmountfs_writesb(mp);
 
 	/*
-	 * Make sure all buffers have been flushed and completed before
-	 * unmounting the log.
+	 * At this point we might have modified the superblock again and thus
+	 * added an item to the AIL, thus flush it again.
 	 */
-	error = xfs_flush_buftarg(mp->m_ddev_targp, 1);
-	if (error)
-		xfs_warn(mp, "%d busy buffers during unmount.", error);
+	xfs_ail_push_all_sync(mp->m_ail);
 	xfs_wait_buftarg(mp->m_ddev_targp);
 
 	xfs_log_unmount_write(mp);
@@ -1586,36 +1589,6 @@ xfs_log_sbcount(xfs_mount_t *mp)
 	return error;
 }
 
-int
-xfs_unmountfs_writesb(xfs_mount_t *mp)
-{
-	xfs_buf_t	*sbp;
-	int		error = 0;
-
-	/*
-	 * skip superblock write if fs is read-only, or
-	 * if we are doing a forced umount.
-	 */
-	if (!((mp->m_flags & XFS_MOUNT_RDONLY) ||
-		XFS_FORCED_SHUTDOWN(mp))) {
-
-		sbp = xfs_getsb(mp, 0);
-
-		XFS_BUF_UNDONE(sbp);
-		XFS_BUF_UNREAD(sbp);
-		xfs_buf_delwri_dequeue(sbp);
-		XFS_BUF_WRITE(sbp);
-		XFS_BUF_UNASYNC(sbp);
-		ASSERT(sbp->b_target == mp->m_ddev_targp);
-		xfsbdstrat(mp, sbp);
-		error = xfs_buf_iowait(sbp);
-		if (error)
-			xfs_buf_ioerror_alert(sbp, __func__);
-		xfs_buf_relse(sbp);
-	}
-	return error;
-}
-
 /*
  * xfs_mod_sb() can be used to copy arbitrary changes to the
  * in-core superblock into the superblock buffer to be logged.
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index bb24dac42a2..90c1fc9eaea 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -53,7 +53,7 @@ typedef struct xfs_trans_reservations {
 
 #include "xfs_sync.h"
 
-struct log;
+struct xlog;
 struct xfs_mount_args;
 struct xfs_inode;
 struct xfs_bmbt_irec;
@@ -133,7 +133,7 @@ typedef struct xfs_mount {
 	uint			m_readio_blocks; /* min read size blocks */
 	uint			m_writeio_log;	/* min write size log bytes */
 	uint			m_writeio_blocks; /* min write size blocks */
-	struct log		*m_log;		/* log specific stuff */
+	struct xlog		*m_log;		/* log specific stuff */
 	int			m_logbufs;	/* number of log buffers */
 	int			m_logbsize;	/* size of each log buffer */
 	uint			m_rsumlevels;	/* rt summary levels */
@@ -211,6 +211,10 @@ typedef struct xfs_mount {
 	struct shrinker		m_inode_shrink;	/* inode reclaim shrinker */
 	int64_t			m_low_space[XFS_LOWSP_MAX];
 						/* low free space thresholds */
+
+	struct workqueue_struct	*m_data_workqueue;
+	struct workqueue_struct	*m_unwritten_workqueue;
+	struct workqueue_struct	*m_cil_workqueue;
 } xfs_mount_t;
 
 /*
@@ -219,7 +223,6 @@ typedef struct xfs_mount {
 #define XFS_MOUNT_WSYNC		(1ULL << 0)	/* for nfs - all metadata ops
 						   must be synchronous except
 						   for space allocations */
-#define XFS_MOUNT_DELAYLOG	(1ULL << 1)	/* delayed logging is enabled */
 #define XFS_MOUNT_WAS_CLEAN	(1ULL << 3)
 #define XFS_MOUNT_FS_SHUTDOWN	(1ULL << 4)	/* atomic stop of all filesystem
 						   operations, typically for
@@ -376,7 +379,6 @@ extern __uint64_t xfs_default_resblks(xfs_mount_t *mp);
 extern int	xfs_mountfs(xfs_mount_t *mp);
 
 extern void	xfs_unmountfs(xfs_mount_t *);
-extern int	xfs_unmountfs_writesb(xfs_mount_t *);
 extern int	xfs_mod_incore_sb(xfs_mount_t *, xfs_sb_field_t, int64_t, int);
 extern int	xfs_mod_incore_sb_batch(xfs_mount_t *, xfs_mod_sb_t *,
 			uint, int);
@@ -396,7 +398,7 @@ extern void	xfs_set_low_space_thresholds(struct xfs_mount *);
 extern void	xfs_mod_sb(struct xfs_trans *, __int64_t);
 extern int	xfs_initialize_perag(struct xfs_mount *, xfs_agnumber_t,
 					xfs_agnumber_t *);
-extern void	xfs_sb_from_disk(struct xfs_sb *, struct xfs_dsb *);
+extern void	xfs_sb_from_disk(struct xfs_mount *, struct xfs_dsb *);
 extern void	xfs_sb_to_disk(struct xfs_dsb *, struct xfs_sb *, __int64_t);
 
 #endif	/* __XFS_MOUNT_H__ */
diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c
index 0bbb1a41998..249db198776 100644
--- a/fs/xfs/xfs_qm.c
+++ b/fs/xfs/xfs_qm.c
@@ -19,7 +19,6 @@
 #include "xfs_fs.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -48,227 +47,182 @@
  * quota functionality, including maintaining the freelist and hash
  * tables of dquots.
  */
-struct mutex	xfs_Gqm_lock;
-struct xfs_qm	*xfs_Gqm;
-uint		ndquot;
-
-kmem_zone_t	*qm_dqzone;
-kmem_zone_t	*qm_dqtrxzone;
-
-STATIC void	xfs_qm_list_init(xfs_dqlist_t *, char *, int);
-STATIC void	xfs_qm_list_destroy(xfs_dqlist_t *);
-
 STATIC int	xfs_qm_init_quotainos(xfs_mount_t *);
 STATIC int	xfs_qm_init_quotainfo(xfs_mount_t *);
 STATIC int	xfs_qm_shake(struct shrinker *, struct shrink_control *);
 
-static struct shrinker xfs_qm_shaker = {
-	.shrink = xfs_qm_shake,
-	.seeks = DEFAULT_SEEKS,
-};
-
 /*
- * Initialize the XQM structure.
- * Note that there is not one quota manager per file system.
+ * We use the batch lookup interface to iterate over the dquots as it
+ * currently is the only interface into the radix tree code that allows
+ * fuzzy lookups instead of exact matches.  Holding the lock over multiple
+ * operations is fine as all callers are used either during mount/umount
+ * or quotaoff.
  */
-STATIC struct xfs_qm *
-xfs_Gqm_init(void)
+#define XFS_DQ_LOOKUP_BATCH	32
+
+STATIC int
+xfs_qm_dquot_walk(
+	struct xfs_mount	*mp,
+	int			type,
+	int			(*execute)(struct xfs_dquot *dqp, void *data),
+	void			*data)
 {
-	xfs_dqhash_t	*udqhash, *gdqhash;
-	xfs_qm_t	*xqm;
-	size_t		hsize;
-	uint		i;
+	struct xfs_quotainfo	*qi = mp->m_quotainfo;
+	struct radix_tree_root	*tree = XFS_DQUOT_TREE(qi, type);
+	uint32_t		next_index;
+	int			last_error = 0;
+	int			skipped;
+	int			nr_found;
+
+restart:
+	skipped = 0;
+	next_index = 0;
+	nr_found = 0;
+
+	while (1) {
+		struct xfs_dquot *batch[XFS_DQ_LOOKUP_BATCH];
+		int		error = 0;
+		int		i;
+
+		mutex_lock(&qi->qi_tree_lock);
+		nr_found = radix_tree_gang_lookup(tree, (void **)batch,
+					next_index, XFS_DQ_LOOKUP_BATCH);
+		if (!nr_found) {
+			mutex_unlock(&qi->qi_tree_lock);
+			break;
+		}
 
-	/*
-	 * Initialize the dquot hash tables.
-	 */
-	udqhash = kmem_zalloc_greedy(&hsize,
-				     XFS_QM_HASHSIZE_LOW * sizeof(xfs_dqhash_t),
-				     XFS_QM_HASHSIZE_HIGH * sizeof(xfs_dqhash_t));
-	if (!udqhash)
-		goto out;
+		for (i = 0; i < nr_found; i++) {
+			struct xfs_dquot *dqp = batch[i];
 
-	gdqhash = kmem_zalloc_large(hsize);
-	if (!gdqhash)
-		goto out_free_udqhash;
+			next_index = be32_to_cpu(dqp->q_core.d_id) + 1;
 
-	hsize /= sizeof(xfs_dqhash_t);
-	ndquot = hsize << 8;
+			error = execute(batch[i], data);
+			if (error == EAGAIN) {
+				skipped++;
+				continue;
+			}
+			if (error && last_error != EFSCORRUPTED)
+				last_error = error;
+		}
 
-	xqm = kmem_zalloc(sizeof(xfs_qm_t), KM_SLEEP);
-	xqm->qm_dqhashmask = hsize - 1;
-	xqm->qm_usr_dqhtable = udqhash;
-	xqm->qm_grp_dqhtable = gdqhash;
-	ASSERT(xqm->qm_usr_dqhtable != NULL);
-	ASSERT(xqm->qm_grp_dqhtable != NULL);
+		mutex_unlock(&qi->qi_tree_lock);
 
-	for (i = 0; i < hsize; i++) {
-		xfs_qm_list_init(&(xqm->qm_usr_dqhtable[i]), "uxdqh", i);
-		xfs_qm_list_init(&(xqm->qm_grp_dqhtable[i]), "gxdqh", i);
+		/* bail out if the filesystem is corrupted.  */
+		if (last_error == EFSCORRUPTED) {
+			skipped = 0;
+			break;
+		}
 	}
 
-	/*
-	 * Freelist of all dquots of all file systems
-	 */
-	INIT_LIST_HEAD(&xqm->qm_dqfrlist);
-	xqm->qm_dqfrlist_cnt = 0;
-	mutex_init(&xqm->qm_dqfrlist_lock);
-
-	/*
-	 * dquot zone. we register our own low-memory callback.
-	 */
-	if (!qm_dqzone) {
-		xqm->qm_dqzone = kmem_zone_init(sizeof(xfs_dquot_t),
-						"xfs_dquots");
-		qm_dqzone = xqm->qm_dqzone;
-	} else
-		xqm->qm_dqzone = qm_dqzone;
-
-	register_shrinker(&xfs_qm_shaker);
-
-	/*
-	 * The t_dqinfo portion of transactions.
-	 */
-	if (!qm_dqtrxzone) {
-		xqm->qm_dqtrxzone = kmem_zone_init(sizeof(xfs_dquot_acct_t),
-						   "xfs_dqtrx");
-		qm_dqtrxzone = xqm->qm_dqtrxzone;
-	} else
-		xqm->qm_dqtrxzone = qm_dqtrxzone;
-
-	atomic_set(&xqm->qm_totaldquots, 0);
-	xqm->qm_dqfree_ratio = XFS_QM_DQFREE_RATIO;
-	xqm->qm_nrefs = 0;
-	return xqm;
+	if (skipped) {
+		delay(1);
+		goto restart;
+	}
 
- out_free_udqhash:
-	kmem_free_large(udqhash);
- out:
-	return NULL;
+	return last_error;
 }
 
-/*
- * Destroy the global quota manager when its reference count goes to zero.
- */
-STATIC void
-xfs_qm_destroy(
-	struct xfs_qm	*xqm)
-{
-	struct xfs_dquot *dqp, *n;
-	int		hsize, i;
-
-	ASSERT(xqm != NULL);
-	ASSERT(xqm->qm_nrefs == 0);
-	unregister_shrinker(&xfs_qm_shaker);
-	hsize = xqm->qm_dqhashmask + 1;
-	for (i = 0; i < hsize; i++) {
-		xfs_qm_list_destroy(&(xqm->qm_usr_dqhtable[i]));
-		xfs_qm_list_destroy(&(xqm->qm_grp_dqhtable[i]));
-	}
-	kmem_free_large(xqm->qm_usr_dqhtable);
-	kmem_free_large(xqm->qm_grp_dqhtable);
-	xqm->qm_usr_dqhtable = NULL;
-	xqm->qm_grp_dqhtable = NULL;
-	xqm->qm_dqhashmask = 0;
-
-	/* frlist cleanup */
-	mutex_lock(&xqm->qm_dqfrlist_lock);
-	list_for_each_entry_safe(dqp, n, &xqm->qm_dqfrlist, q_freelist) {
-		xfs_dqlock(dqp);
-		list_del_init(&dqp->q_freelist);
-		xfs_Gqm->qm_dqfrlist_cnt--;
-		xfs_dqunlock(dqp);
-		xfs_qm_dqdestroy(dqp);
-	}
-	mutex_unlock(&xqm->qm_dqfrlist_lock);
-	mutex_destroy(&xqm->qm_dqfrlist_lock);
-	kmem_free(xqm);
-}
 
 /*
- * Called at mount time to let XQM know that another file system is
- * starting quotas. This isn't crucial information as the individual mount
- * structures are pretty independent, but it helps the XQM keep a
- * global view of what's going on.
+ * Purge a dquot from all tracking data structures and free it.
  */
-/* ARGSUSED */
 STATIC int
-xfs_qm_hold_quotafs_ref(
-	struct xfs_mount *mp)
+xfs_qm_dqpurge(
+	struct xfs_dquot	*dqp,
+	void			*data)
 {
-	/*
-	 * Need to lock the xfs_Gqm structure for things like this. For example,
-	 * the structure could disappear between the entry to this routine and
-	 * a HOLD operation if not locked.
-	 */
-	mutex_lock(&xfs_Gqm_lock);
+	struct xfs_mount	*mp = dqp->q_mount;
+	struct xfs_quotainfo	*qi = mp->m_quotainfo;
+	struct xfs_dquot	*gdqp = NULL;
 
-	if (!xfs_Gqm) {
-		xfs_Gqm = xfs_Gqm_init();
-		if (!xfs_Gqm) {
-			mutex_unlock(&xfs_Gqm_lock);
-			return ENOMEM;
-		}
+	xfs_dqlock(dqp);
+	if ((dqp->dq_flags & XFS_DQ_FREEING) || dqp->q_nrefs != 0) {
+		xfs_dqunlock(dqp);
+		return EAGAIN;
 	}
 
 	/*
-	 * We can keep a list of all filesystems with quotas mounted for
-	 * debugging and statistical purposes, but ...
-	 * Just take a reference and get out.
+	 * If this quota has a group hint attached, prepare for releasing it
+	 * now.
 	 */
-	xfs_Gqm->qm_nrefs++;
-	mutex_unlock(&xfs_Gqm_lock);
+	gdqp = dqp->q_gdquot;
+	if (gdqp) {
+		xfs_dqlock(gdqp);
+		dqp->q_gdquot = NULL;
+	}
 
-	return 0;
-}
+	dqp->dq_flags |= XFS_DQ_FREEING;
 
-
-/*
- * Release the reference that a filesystem took at mount time,
- * so that we know when we need to destroy the entire quota manager.
- */
-/* ARGSUSED */
-STATIC void
-xfs_qm_rele_quotafs_ref(
-	struct xfs_mount *mp)
-{
-	xfs_dquot_t	*dqp, *n;
-
-	ASSERT(xfs_Gqm);
-	ASSERT(xfs_Gqm->qm_nrefs > 0);
+	xfs_dqflock(dqp);
 
 	/*
-	 * Go thru the freelist and destroy all inactive dquots.
+	 * If we are turning this type of quotas off, we don't care
+	 * about the dirty metadata sitting in this dquot. OTOH, if
+	 * we're unmounting, we do care, so we flush it and wait.
 	 */
-	mutex_lock(&xfs_Gqm->qm_dqfrlist_lock);
-
-	list_for_each_entry_safe(dqp, n, &xfs_Gqm->qm_dqfrlist, q_freelist) {
-		xfs_dqlock(dqp);
-		if (dqp->dq_flags & XFS_DQ_INACTIVE) {
-			ASSERT(dqp->q_mount == NULL);
-			ASSERT(! XFS_DQ_IS_DIRTY(dqp));
-			ASSERT(list_empty(&dqp->q_hashlist));
-			ASSERT(list_empty(&dqp->q_mplist));
-			list_del_init(&dqp->q_freelist);
-			xfs_Gqm->qm_dqfrlist_cnt--;
-			xfs_dqunlock(dqp);
-			xfs_qm_dqdestroy(dqp);
+	if (XFS_DQ_IS_DIRTY(dqp)) {
+		struct xfs_buf	*bp = NULL;
+		int		error;
+
+		/*
+		 * We don't care about getting disk errors here. We need
+		 * to purge this dquot anyway, so we go ahead regardless.
+		 */
+		error = xfs_qm_dqflush(dqp, &bp);
+		if (error) {
+			xfs_warn(mp, "%s: dquot %p flush failed",
+				__func__, dqp);
 		} else {
-			xfs_dqunlock(dqp);
+			error = xfs_bwrite(bp);
+			xfs_buf_relse(bp);
 		}
+		xfs_dqflock(dqp);
 	}
-	mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
+
+	ASSERT(atomic_read(&dqp->q_pincount) == 0);
+	ASSERT(XFS_FORCED_SHUTDOWN(mp) ||
+	       !(dqp->q_logitem.qli_item.li_flags & XFS_LI_IN_AIL));
+
+	xfs_dqfunlock(dqp);
+	xfs_dqunlock(dqp);
+
+	radix_tree_delete(XFS_DQUOT_TREE(qi, dqp->q_core.d_flags),
+			  be32_to_cpu(dqp->q_core.d_id));
+	qi->qi_dquots--;
 
 	/*
-	 * Destroy the entire XQM. If somebody mounts with quotaon, this'll
-	 * be restarted.
+	 * We move dquots to the freelist as soon as their reference count
+	 * hits zero, so it really should be on the freelist here.
 	 */
-	mutex_lock(&xfs_Gqm_lock);
-	if (--xfs_Gqm->qm_nrefs == 0) {
-		xfs_qm_destroy(xfs_Gqm);
-		xfs_Gqm = NULL;
-	}
-	mutex_unlock(&xfs_Gqm_lock);
+	mutex_lock(&qi->qi_lru_lock);
+	ASSERT(!list_empty(&dqp->q_lru));
+	list_del_init(&dqp->q_lru);
+	qi->qi_lru_count--;
+	XFS_STATS_DEC(xs_qm_dquot_unused);
+	mutex_unlock(&qi->qi_lru_lock);
+
+	xfs_qm_dqdestroy(dqp);
+
+	if (gdqp)
+		xfs_qm_dqput(gdqp);
+	return 0;
+}
+
+/*
+ * Purge the dquot cache.
+ */
+void
+xfs_qm_dqpurge_all(
+	struct xfs_mount	*mp,
+	uint			flags)
+{
+	if (flags & XFS_QMOPT_UQUOTA)
+		xfs_qm_dquot_walk(mp, XFS_DQ_USER, xfs_qm_dqpurge, NULL);
+	if (flags & XFS_QMOPT_GQUOTA)
+		xfs_qm_dquot_walk(mp, XFS_DQ_GROUP, xfs_qm_dqpurge, NULL);
+	if (flags & XFS_QMOPT_PQUOTA)
+		xfs_qm_dquot_walk(mp, XFS_DQ_PROJ, xfs_qm_dqpurge, NULL);
 }
 
 /*
@@ -409,200 +363,6 @@ xfs_qm_unmount_quotas(
 	}
 }
 
-/*
- * Flush all dquots of the given file system to disk. The dquots are
- * _not_ purged from memory here, just their data written to disk.
- */
-STATIC int
-xfs_qm_dqflush_all(
-	struct xfs_mount	*mp,
-	int			sync_mode)
-{
-	struct xfs_quotainfo	*q = mp->m_quotainfo;
-	int			recl;
-	struct xfs_dquot	*dqp;
-	int			error;
-
-	if (!q)
-		return 0;
-again:
-	mutex_lock(&q->qi_dqlist_lock);
-	list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) {
-		xfs_dqlock(dqp);
-		if (! XFS_DQ_IS_DIRTY(dqp)) {
-			xfs_dqunlock(dqp);
-			continue;
-		}
-
-		/* XXX a sentinel would be better */
-		recl = q->qi_dqreclaims;
-		if (!xfs_dqflock_nowait(dqp)) {
-			/*
-			 * If we can't grab the flush lock then check
-			 * to see if the dquot has been flushed delayed
-			 * write.  If so, grab its buffer and send it
-			 * out immediately.  We'll be able to acquire
-			 * the flush lock when the I/O completes.
-			 */
-			xfs_qm_dqflock_pushbuf_wait(dqp);
-		}
-		/*
-		 * Let go of the mplist lock. We don't want to hold it
-		 * across a disk write.
-		 */
-		mutex_unlock(&q->qi_dqlist_lock);
-		error = xfs_qm_dqflush(dqp, sync_mode);
-		xfs_dqunlock(dqp);
-		if (error)
-			return error;
-
-		mutex_lock(&q->qi_dqlist_lock);
-		if (recl != q->qi_dqreclaims) {
-			mutex_unlock(&q->qi_dqlist_lock);
-			/* XXX restart limit */
-			goto again;
-		}
-	}
-
-	mutex_unlock(&q->qi_dqlist_lock);
-	/* return ! busy */
-	return 0;
-}
-/*
- * Release the group dquot pointers the user dquots may be
- * carrying around as a hint. mplist is locked on entry and exit.
- */
-STATIC void
-xfs_qm_detach_gdquots(
-	struct xfs_mount	*mp)
-{
-	struct xfs_quotainfo	*q = mp->m_quotainfo;
-	struct xfs_dquot	*dqp, *gdqp;
-	int			nrecl;
-
- again:
-	ASSERT(mutex_is_locked(&q->qi_dqlist_lock));
-	list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) {
-		xfs_dqlock(dqp);
-		if ((gdqp = dqp->q_gdquot)) {
-			xfs_dqlock(gdqp);
-			dqp->q_gdquot = NULL;
-		}
-		xfs_dqunlock(dqp);
-
-		if (gdqp) {
-			/*
-			 * Can't hold the mplist lock across a dqput.
-			 * XXXmust convert to marker based iterations here.
-			 */
-			nrecl = q->qi_dqreclaims;
-			mutex_unlock(&q->qi_dqlist_lock);
-			xfs_qm_dqput(gdqp);
-
-			mutex_lock(&q->qi_dqlist_lock);
-			if (nrecl != q->qi_dqreclaims)
-				goto again;
-		}
-	}
-}
-
-/*
- * Go through all the incore dquots of this file system and take them
- * off the mplist and hashlist, if the dquot type matches the dqtype
- * parameter. This is used when turning off quota accounting for
- * users and/or groups, as well as when the filesystem is unmounting.
- */
-STATIC int
-xfs_qm_dqpurge_int(
-	struct xfs_mount	*mp,
-	uint			flags)
-{
-	struct xfs_quotainfo	*q = mp->m_quotainfo;
-	struct xfs_dquot	*dqp, *n;
-	uint			dqtype;
-	int			nrecl;
-	int			nmisses;
-
-	if (!q)
-		return 0;
-
-	dqtype = (flags & XFS_QMOPT_UQUOTA) ? XFS_DQ_USER : 0;
-	dqtype |= (flags & XFS_QMOPT_PQUOTA) ? XFS_DQ_PROJ : 0;
-	dqtype |= (flags & XFS_QMOPT_GQUOTA) ? XFS_DQ_GROUP : 0;
-
-	mutex_lock(&q->qi_dqlist_lock);
-
-	/*
-	 * In the first pass through all incore dquots of this filesystem,
-	 * we release the group dquot pointers the user dquots may be
-	 * carrying around as a hint. We need to do this irrespective of
-	 * what's being turned off.
-	 */
-	xfs_qm_detach_gdquots(mp);
-
-      again:
-	nmisses = 0;
-	ASSERT(mutex_is_locked(&q->qi_dqlist_lock));
-	/*
-	 * Try to get rid of all of the unwanted dquots. The idea is to
-	 * get them off mplist and hashlist, but leave them on freelist.
-	 */
-	list_for_each_entry_safe(dqp, n, &q->qi_dqlist, q_mplist) {
-		/*
-		 * It's OK to look at the type without taking dqlock here.
-		 * We're holding the mplist lock here, and that's needed for
-		 * a dqreclaim.
-		 */
-		if ((dqp->dq_flags & dqtype) == 0)
-			continue;
-
-		if (!mutex_trylock(&dqp->q_hash->qh_lock)) {
-			nrecl = q->qi_dqreclaims;
-			mutex_unlock(&q->qi_dqlist_lock);
-			mutex_lock(&dqp->q_hash->qh_lock);
-			mutex_lock(&q->qi_dqlist_lock);
-
-			/*
-			 * XXXTheoretically, we can get into a very long
-			 * ping pong game here.
-			 * No one can be adding dquots to the mplist at
-			 * this point, but somebody might be taking things off.
-			 */
-			if (nrecl != q->qi_dqreclaims) {
-				mutex_unlock(&dqp->q_hash->qh_lock);
-				goto again;
-			}
-		}
-
-		/*
-		 * Take the dquot off the mplist and hashlist. It may remain on
-		 * freelist in INACTIVE state.
-		 */
-		nmisses += xfs_qm_dqpurge(dqp);
-	}
-	mutex_unlock(&q->qi_dqlist_lock);
-	return nmisses;
-}
-
-int
-xfs_qm_dqpurge_all(
-	xfs_mount_t	*mp,
-	uint		flags)
-{
-	int		ndquots;
-
-	/*
-	 * Purge the dquot cache.
-	 * None of the dquots should really be busy at this point.
-	 */
-	if (mp->m_quotainfo) {
-		while ((ndquots = xfs_qm_dqpurge_int(mp, flags))) {
-			delay(ndquots * 10);
-		}
-	}
-	return 0;
-}
-
 STATIC int
 xfs_qm_dqattach_one(
 	xfs_inode_t	*ip,
@@ -648,12 +408,9 @@ xfs_qm_dqattach_one(
 		 */
 		dqp = udqhint->q_gdquot;
 		if (dqp && be32_to_cpu(dqp->q_core.d_id) == id) {
-			xfs_dqlock(dqp);
-			XFS_DQHOLD(dqp);
 			ASSERT(*IO_idqpp == NULL);
-			*IO_idqpp = dqp;
 
-			xfs_dqunlock(dqp);
+			*IO_idqpp = xfs_qm_dqhold(dqp);
 			xfs_dqunlock(udqhint);
 			return 0;
 		}
@@ -693,11 +450,7 @@ xfs_qm_dqattach_one(
 
 /*
  * Given a udquot and gdquot, attach a ptr to the group dquot in the
- * udquot as a hint for future lookups. The idea sounds simple, but the
- * execution isn't, because the udquot might have a group dquot attached
- * already and getting rid of that gets us into lock ordering constraints.
- * The process is complicated more by the fact that the dquots may or may not
- * be locked on entry.
+ * udquot as a hint for future lookups.
  */
 STATIC void
 xfs_qm_dqattach_grouphint(
@@ -708,48 +461,37 @@ xfs_qm_dqattach_grouphint(
 
 	xfs_dqlock(udq);
 
-	if ((tmp = udq->q_gdquot)) {
-		if (tmp == gdq) {
-			xfs_dqunlock(udq);
-			return;
-		}
+	tmp = udq->q_gdquot;
+	if (tmp) {
+		if (tmp == gdq)
+			goto done;
 
 		udq->q_gdquot = NULL;
-		/*
-		 * We can't keep any dqlocks when calling dqrele,
-		 * because the freelist lock comes before dqlocks.
-		 */
-		xfs_dqunlock(udq);
-		/*
-		 * we took a hard reference once upon a time in dqget,
-		 * so give it back when the udquot no longer points at it
-		 * dqput() does the unlocking of the dquot.
-		 */
 		xfs_qm_dqrele(tmp);
-
-		xfs_dqlock(udq);
-		xfs_dqlock(gdq);
-
-	} else {
-		ASSERT(XFS_DQ_IS_LOCKED(udq));
-		xfs_dqlock(gdq);
 	}
 
-	ASSERT(XFS_DQ_IS_LOCKED(udq));
-	ASSERT(XFS_DQ_IS_LOCKED(gdq));
-	/*
-	 * Somebody could have attached a gdquot here,
-	 * when we dropped the uqlock. If so, just do nothing.
-	 */
-	if (udq->q_gdquot == NULL) {
-		XFS_DQHOLD(gdq);
-		udq->q_gdquot = gdq;
-	}
-
-	xfs_dqunlock(gdq);
+	udq->q_gdquot = xfs_qm_dqhold(gdq);
+done:
 	xfs_dqunlock(udq);
 }
 
+static bool
+xfs_qm_need_dqattach(
+	struct xfs_inode	*ip)
+{
+	struct xfs_mount	*mp = ip->i_mount;
+
+	if (!XFS_IS_QUOTA_RUNNING(mp))
+		return false;
+	if (!XFS_IS_QUOTA_ON(mp))
+		return false;
+	if (!XFS_NOT_DQATTACHED(mp, ip))
+		return false;
+	if (ip->i_ino == mp->m_sb.sb_uquotino ||
+	    ip->i_ino == mp->m_sb.sb_gquotino)
+		return false;
+	return true;
+}
 
 /*
  * Given a locked inode, attach dquot(s) to it, taking U/G/P-QUOTAON
@@ -767,11 +509,7 @@ xfs_qm_dqattach_locked(
 	uint		nquotas = 0;
 	int		error = 0;
 
-	if (!XFS_IS_QUOTA_RUNNING(mp) ||
-	    !XFS_IS_QUOTA_ON(mp) ||
-	    !XFS_NOT_DQATTACHED(mp, ip) ||
-	    ip->i_ino == mp->m_sb.sb_uquotino ||
-	    ip->i_ino == mp->m_sb.sb_gquotino)
+	if (!xfs_qm_need_dqattach(ip))
 		return 0;
 
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
@@ -813,17 +551,13 @@ xfs_qm_dqattach_locked(
 		ASSERT(ip->i_gdquot);
 
 		/*
-		 * We may or may not have the i_udquot locked at this point,
-		 * but this check is OK since we don't depend on the i_gdquot to
-		 * be accurate 100% all the time. It is just a hint, and this
-		 * will succeed in general.
-		 */
-		if (ip->i_udquot->q_gdquot == ip->i_gdquot)
-			goto done;
-		/*
-		 * Attach i_gdquot to the gdquot hint inside the i_udquot.
+		 * We do not have i_udquot locked at this point, but this check
+		 * is OK since we don't depend on the i_gdquot to be accurate
+		 * 100% all the time. It is just a hint, and this will
+		 * succeed in general.
 		 */
-		xfs_qm_dqattach_grouphint(ip->i_udquot, ip->i_gdquot);
+		if (ip->i_udquot->q_gdquot != ip->i_gdquot)
+			xfs_qm_dqattach_grouphint(ip->i_udquot, ip->i_gdquot);
 	}
 
  done:
@@ -846,6 +580,9 @@ xfs_qm_dqattach(
 {
 	int			error;
 
+	if (!xfs_qm_need_dqattach(ip))
+		return 0;
+
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
 	error = xfs_qm_dqattach_locked(ip, flags);
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
@@ -879,108 +616,6 @@ xfs_qm_dqdetach(
 	}
 }
 
-int
-xfs_qm_sync(
-	struct xfs_mount	*mp,
-	int			flags)
-{
-	struct xfs_quotainfo	*q = mp->m_quotainfo;
-	int			recl, restarts;
-	struct xfs_dquot	*dqp;
-	int			error;
-
-	if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp))
-		return 0;
-
-	restarts = 0;
-
-  again:
-	mutex_lock(&q->qi_dqlist_lock);
-	/*
-	 * dqpurge_all() also takes the mplist lock and iterate thru all dquots
-	 * in quotaoff. However, if the QUOTA_ACTIVE bits are not cleared
-	 * when we have the mplist lock, we know that dquots will be consistent
-	 * as long as we have it locked.
-	 */
-	if (!XFS_IS_QUOTA_ON(mp)) {
-		mutex_unlock(&q->qi_dqlist_lock);
-		return 0;
-	}
-	ASSERT(mutex_is_locked(&q->qi_dqlist_lock));
-	list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) {
-		/*
-		 * If this is vfs_sync calling, then skip the dquots that
-		 * don't 'seem' to be dirty. ie. don't acquire dqlock.
-		 * This is very similar to what xfs_sync does with inodes.
-		 */
-		if (flags & SYNC_TRYLOCK) {
-			if (!XFS_DQ_IS_DIRTY(dqp))
-				continue;
-			if (!xfs_qm_dqlock_nowait(dqp))
-				continue;
-		} else {
-			xfs_dqlock(dqp);
-		}
-
-		/*
-		 * Now, find out for sure if this dquot is dirty or not.
-		 */
-		if (! XFS_DQ_IS_DIRTY(dqp)) {
-			xfs_dqunlock(dqp);
-			continue;
-		}
-
-		/* XXX a sentinel would be better */
-		recl = q->qi_dqreclaims;
-		if (!xfs_dqflock_nowait(dqp)) {
-			if (flags & SYNC_TRYLOCK) {
-				xfs_dqunlock(dqp);
-				continue;
-			}
-			/*
-			 * If we can't grab the flush lock then if the caller
-			 * really wanted us to give this our best shot, so
-			 * see if we can give a push to the buffer before we wait
-			 * on the flush lock. At this point, we know that
-			 * even though the dquot is being flushed,
-			 * it has (new) dirty data.
-			 */
-			xfs_qm_dqflock_pushbuf_wait(dqp);
-		}
-		/*
-		 * Let go of the mplist lock. We don't want to hold it
-		 * across a disk write
-		 */
-		mutex_unlock(&q->qi_dqlist_lock);
-		error = xfs_qm_dqflush(dqp, flags);
-		xfs_dqunlock(dqp);
-		if (error && XFS_FORCED_SHUTDOWN(mp))
-			return 0;	/* Need to prevent umount failure */
-		else if (error)
-			return error;
-
-		mutex_lock(&q->qi_dqlist_lock);
-		if (recl != q->qi_dqreclaims) {
-			if (++restarts >= XFS_QM_SYNC_MAX_RESTARTS)
-				break;
-
-			mutex_unlock(&q->qi_dqlist_lock);
-			goto again;
-		}
-	}
-
-	mutex_unlock(&q->qi_dqlist_lock);
-	return 0;
-}
-
-/*
- * The hash chains and the mplist use the same xfs_dqhash structure as
- * their list head, but we can take the mplist qh_lock and one of the
- * hash qh_locks at the same time without any problem as they aren't
- * related.
- */
-static struct lock_class_key xfs_quota_mplist_class;
-
 /*
  * This initializes all the quota information that's kept in the
  * mount structure
@@ -995,13 +630,6 @@ xfs_qm_init_quotainfo(
 
 	ASSERT(XFS_IS_QUOTA_RUNNING(mp));
 
-	/*
-	 * Tell XQM that we exist as soon as possible.
-	 */
-	if ((error = xfs_qm_hold_quotafs_ref(mp))) {
-		return error;
-	}
-
 	qinf = mp->m_quotainfo = kmem_zalloc(sizeof(xfs_quotainfo_t), KM_SLEEP);
 
 	/*
@@ -1014,11 +642,13 @@ xfs_qm_init_quotainfo(
 		return error;
 	}
 
-	INIT_LIST_HEAD(&qinf->qi_dqlist);
-	mutex_init(&qinf->qi_dqlist_lock);
-	lockdep_set_class(&qinf->qi_dqlist_lock, &xfs_quota_mplist_class);
+	INIT_RADIX_TREE(&qinf->qi_uquota_tree, GFP_NOFS);
+	INIT_RADIX_TREE(&qinf->qi_gquota_tree, GFP_NOFS);
+	mutex_init(&qinf->qi_tree_lock);
 
-	qinf->qi_dqreclaims = 0;
+	INIT_LIST_HEAD(&qinf->qi_lru_list);
+	qinf->qi_lru_count = 0;
+	mutex_init(&qinf->qi_lru_lock);
 
 	/* mutex used to serialize quotaoffs */
 	mutex_init(&qinf->qi_quotaofflock);
@@ -1034,18 +664,21 @@ xfs_qm_init_quotainfo(
 	/*
 	 * We try to get the limits from the superuser's limits fields.
 	 * This is quite hacky, but it is standard quota practice.
+	 *
 	 * We look at the USR dquot with id == 0 first, but if user quotas
 	 * are not enabled we goto the GRP dquot with id == 0.
 	 * We don't really care to keep separate default limits for user
 	 * and group quotas, at least not at this point.
+	 *
+	 * Since we may not have done a quotacheck by this point, just read
+	 * the dquot without attaching it to any hashtables or lists.
 	 */
-	error = xfs_qm_dqget(mp, NULL, (xfs_dqid_t)0,
-			     XFS_IS_UQUOTA_RUNNING(mp) ? XFS_DQ_USER : 
-			     (XFS_IS_GQUOTA_RUNNING(mp) ? XFS_DQ_GROUP :
-				XFS_DQ_PROJ),
-			     XFS_QMOPT_DQSUSER|XFS_QMOPT_DOWARN,
-			     &dqp);
-	if (! error) {
+	error = xfs_qm_dqread(mp, 0,
+			XFS_IS_UQUOTA_RUNNING(mp) ? XFS_DQ_USER :
+			 (XFS_IS_GQUOTA_RUNNING(mp) ? XFS_DQ_GROUP :
+			  XFS_DQ_PROJ),
+			XFS_QMOPT_DOWARN, &dqp);
+	if (!error) {
 		xfs_disk_dquot_t	*ddqp = &dqp->q_core;
 
 		/*
@@ -1072,11 +705,6 @@ xfs_qm_init_quotainfo(
 		qinf->qi_rtbhardlimit = be64_to_cpu(ddqp->d_rtb_hardlimit);
 		qinf->qi_rtbsoftlimit = be64_to_cpu(ddqp->d_rtb_softlimit);
  
-		/*
-		 * We sent the XFS_QMOPT_DQSUSER flag to dqget because
-		 * we don't want this dquot cached. We haven't done a
-		 * quotacheck yet, and quotacheck doesn't like incore dquots.
-		 */
 		xfs_qm_dqdestroy(dqp);
 	} else {
 		qinf->qi_btimelimit = XFS_QM_BTIMELIMIT;
@@ -1087,6 +715,9 @@ xfs_qm_init_quotainfo(
 		qinf->qi_rtbwarnlimit = XFS_QM_RTBWARNLIMIT;
 	}
 
+	qinf->qi_shrinker.shrink = xfs_qm_shake;
+	qinf->qi_shrinker.seeks = DEFAULT_SEEKS;
+	register_shrinker(&qinf->qi_shrinker);
 	return 0;
 }
 
@@ -1104,17 +735,8 @@ xfs_qm_destroy_quotainfo(
 
 	qi = mp->m_quotainfo;
 	ASSERT(qi != NULL);
-	ASSERT(xfs_Gqm != NULL);
-
-	/*
-	 * Release the reference that XQM kept, so that we know
-	 * when the XQM structure should be freed. We cannot assume
-	 * that xfs_Gqm is non-null after this point.
-	 */
-	xfs_qm_rele_quotafs_ref(mp);
 
-	ASSERT(list_empty(&qi->qi_dqlist));
-	mutex_destroy(&qi->qi_dqlist_lock);
+	unregister_shrinker(&qi->qi_shrinker);
 
 	if (qi->qi_uquotaip) {
 		IRELE(qi->qi_uquotaip);
@@ -1129,30 +751,6 @@ xfs_qm_destroy_quotainfo(
 	mp->m_quotainfo = NULL;
 }
 
-
-
-/* ------------------- PRIVATE STATIC FUNCTIONS ----------------------- */
-
-/* ARGSUSED */
-STATIC void
-xfs_qm_list_init(
-	xfs_dqlist_t	*list,
-	char		*str,
-	int		n)
-{
-	mutex_init(&list->qh_lock);
-	INIT_LIST_HEAD(&list->qh_list);
-	list->qh_version = 0;
-	list->qh_nelems = 0;
-}
-
-STATIC void
-xfs_qm_list_destroy(
-	xfs_dqlist_t	*list)
-{
-	mutex_destroy(&(list->qh_lock));
-}
-
 /*
  * Create an inode and return with a reference already taken, but unlocked
  * This is how we create quota inodes
@@ -1265,15 +863,16 @@ xfs_qm_reset_dqcounts(
 
 STATIC int
 xfs_qm_dqiter_bufs(
-	xfs_mount_t	*mp,
-	xfs_dqid_t	firstid,
-	xfs_fsblock_t	bno,
-	xfs_filblks_t	blkcnt,
-	uint		flags)
+	struct xfs_mount	*mp,
+	xfs_dqid_t		firstid,
+	xfs_fsblock_t		bno,
+	xfs_filblks_t		blkcnt,
+	uint			flags,
+	struct list_head	*buffer_list)
 {
-	xfs_buf_t	*bp;
-	int		error;
-	int		type;
+	struct xfs_buf		*bp;
+	int			error;
+	int			type;
 
 	ASSERT(blkcnt > 0);
 	type = flags & XFS_QMOPT_UQUOTA ? XFS_DQ_USER :
@@ -1297,7 +896,7 @@ xfs_qm_dqiter_bufs(
 			break;
 
 		xfs_qm_reset_dqcounts(mp, bp, firstid, type);
-		xfs_buf_delwri_queue(bp);
+		xfs_buf_delwri_queue(bp, buffer_list);
 		xfs_buf_relse(bp);
 		/*
 		 * goto the next block.
@@ -1305,6 +904,7 @@ xfs_qm_dqiter_bufs(
 		bno++;
 		firstid += mp->m_quotainfo->qi_dqperchunk;
 	}
+
 	return error;
 }
 
@@ -1314,11 +914,12 @@ xfs_qm_dqiter_bufs(
  */
 STATIC int
 xfs_qm_dqiterate(
-	xfs_mount_t	*mp,
-	xfs_inode_t	*qip,
-	uint		flags)
+	struct xfs_mount	*mp,
+	struct xfs_inode	*qip,
+	uint			flags,
+	struct list_head	*buffer_list)
 {
-	xfs_bmbt_irec_t		*map;
+	struct xfs_bmbt_irec	*map;
 	int			i, nmaps;	/* number of map entries */
 	int			error;		/* return value */
 	xfs_fileoff_t		lblkno;
@@ -1385,21 +986,17 @@ xfs_qm_dqiterate(
 			 * Iterate thru all the blks in the extent and
 			 * reset the counters of all the dquots inside them.
 			 */
-			if ((error = xfs_qm_dqiter_bufs(mp,
-						       firstid,
-						       map[i].br_startblock,
-						       map[i].br_blockcount,
-						       flags))) {
-				break;
-			}
+			error = xfs_qm_dqiter_bufs(mp, firstid,
+						   map[i].br_startblock,
+						   map[i].br_blockcount,
+						   flags, buffer_list);
+			if (error)
+				goto out;
 		}
-
-		if (error)
-			break;
 	} while (nmaps > 0);
 
+out:
 	kmem_free(map);
-
 	return error;
 }
 
@@ -1590,6 +1187,33 @@ error0:
 	return error;
 }
 
+STATIC int
+xfs_qm_flush_one(
+	struct xfs_dquot	*dqp,
+	void			*data)
+{
+	struct list_head	*buffer_list = data;
+	struct xfs_buf		*bp = NULL;
+	int			error = 0;
+
+	xfs_dqlock(dqp);
+	if (dqp->dq_flags & XFS_DQ_FREEING)
+		goto out_unlock;
+	if (!XFS_DQ_IS_DIRTY(dqp))
+		goto out_unlock;
+
+	xfs_dqflock(dqp);
+	error = xfs_qm_dqflush(dqp, &bp);
+	if (error)
+		goto out_unlock;
+
+	xfs_buf_delwri_queue(bp, buffer_list);
+	xfs_buf_relse(bp);
+out_unlock:
+	xfs_dqunlock(dqp);
+	return error;
+}
+
 /*
  * Walk thru all the filesystem inodes and construct a consistent view
  * of the disk quota world. If the quotacheck fails, disable quotas.
@@ -1598,11 +1222,12 @@ int
 xfs_qm_quotacheck(
 	xfs_mount_t	*mp)
 {
-	int		done, count, error;
+	int		done, count, error, error2;
 	xfs_ino_t	lastino;
 	size_t		structsz;
 	xfs_inode_t	*uip, *gip;
 	uint		flags;
+	LIST_HEAD	(buffer_list);
 
 	count = INT_MAX;
 	structsz = 1;
@@ -1612,12 +1237,6 @@ xfs_qm_quotacheck(
 	ASSERT(mp->m_quotainfo->qi_uquotaip || mp->m_quotainfo->qi_gquotaip);
 	ASSERT(XFS_IS_QUOTA_RUNNING(mp));
 
-	/*
-	 * There should be no cached dquots. The (simplistic) quotacheck
-	 * algorithm doesn't like that.
-	 */
-	ASSERT(list_empty(&mp->m_quotainfo->qi_dqlist));
-
 	xfs_notice(mp, "Quotacheck needed: Please wait.");
 
 	/*
@@ -1627,7 +1246,8 @@ xfs_qm_quotacheck(
 	 */
 	uip = mp->m_quotainfo->qi_uquotaip;
 	if (uip) {
-		error = xfs_qm_dqiterate(mp, uip, XFS_QMOPT_UQUOTA);
+		error = xfs_qm_dqiterate(mp, uip, XFS_QMOPT_UQUOTA,
+					 &buffer_list);
 		if (error)
 			goto error_return;
 		flags |= XFS_UQUOTA_CHKD;
@@ -1636,7 +1256,8 @@ xfs_qm_quotacheck(
 	gip = mp->m_quotainfo->qi_gquotaip;
 	if (gip) {
 		error = xfs_qm_dqiterate(mp, gip, XFS_IS_GQUOTA_ON(mp) ?
-					XFS_QMOPT_GQUOTA : XFS_QMOPT_PQUOTA);
+					 XFS_QMOPT_GQUOTA : XFS_QMOPT_PQUOTA,
+					 &buffer_list);
 		if (error)
 			goto error_return;
 		flags |= XFS_OQUOTA_CHKD;
@@ -1656,12 +1277,29 @@ xfs_qm_quotacheck(
 	} while (!done);
 
 	/*
-	 * We've made all the changes that we need to make incore.
-	 * Flush them down to disk buffers if everything was updated
-	 * successfully.
+	 * We've made all the changes that we need to make incore.  Flush them
+	 * down to disk buffers if everything was updated successfully.
 	 */
+	if (XFS_IS_UQUOTA_ON(mp)) {
+		error = xfs_qm_dquot_walk(mp, XFS_DQ_USER, xfs_qm_flush_one,
+					  &buffer_list);
+	}
+	if (XFS_IS_GQUOTA_ON(mp)) {
+		error2 = xfs_qm_dquot_walk(mp, XFS_DQ_GROUP, xfs_qm_flush_one,
+					   &buffer_list);
+		if (!error)
+			error = error2;
+	}
+	if (XFS_IS_PQUOTA_ON(mp)) {
+		error2 = xfs_qm_dquot_walk(mp, XFS_DQ_PROJ, xfs_qm_flush_one,
+					   &buffer_list);
+		if (!error)
+			error = error2;
+	}
+
+	error2 = xfs_buf_delwri_submit(&buffer_list);
 	if (!error)
-		error = xfs_qm_dqflush_all(mp, 0);
+		error = error2;
 
 	/*
 	 * We can get this error if we couldn't do a dquot allocation inside
@@ -1676,23 +1314,21 @@ xfs_qm_quotacheck(
 	}
 
 	/*
-	 * We didn't log anything, because if we crashed, we'll have to
-	 * start the quotacheck from scratch anyway. However, we must make
-	 * sure that our dquot changes are secure before we put the
-	 * quotacheck'd stamp on the superblock. So, here we do a synchronous
-	 * flush.
-	 */
-	xfs_flush_buftarg(mp->m_ddev_targp, 1);
-
-	/*
 	 * 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.
 	 */
-	mp->m_qflags &= ~(XFS_OQUOTA_CHKD | XFS_UQUOTA_CHKD);
+	mp->m_qflags &= ~XFS_ALL_QUOTA_CHKD;
 	mp->m_qflags |= flags;
 
  error_return:
+	while (!list_empty(&buffer_list)) {
+		struct xfs_buf *bp =
+			list_first_entry(&buffer_list, struct xfs_buf, b_list);
+		list_del_init(&bp->b_list);
+		xfs_buf_relse(bp);
+	}
+
 	if (error) {
 		xfs_warn(mp,
 	"Quotacheck: Unsuccessful (Error %d): Disabling quotas.",
@@ -1701,7 +1337,6 @@ xfs_qm_quotacheck(
 		 * We must turn off quotas.
 		 */
 		ASSERT(mp->m_quotainfo != NULL);
-		ASSERT(xfs_Gqm != NULL);
 		xfs_qm_destroy_quotainfo(mp);
 		if (xfs_mount_reset_sbqflags(mp)) {
 			xfs_warn(mp,
@@ -1790,257 +1425,150 @@ xfs_qm_init_quotainos(
 	return 0;
 }
 
+STATIC void
+xfs_qm_dqfree_one(
+	struct xfs_dquot	*dqp)
+{
+	struct xfs_mount	*mp = dqp->q_mount;
+	struct xfs_quotainfo	*qi = mp->m_quotainfo;
 
+	mutex_lock(&qi->qi_tree_lock);
+	radix_tree_delete(XFS_DQUOT_TREE(qi, dqp->q_core.d_flags),
+			  be32_to_cpu(dqp->q_core.d_id));
 
-/*
- * Just pop the least recently used dquot off the freelist and
- * recycle it. The returned dquot is locked.
- */
-STATIC xfs_dquot_t *
-xfs_qm_dqreclaim_one(void)
-{
-	xfs_dquot_t	*dqpout;
-	xfs_dquot_t	*dqp;
-	int		restarts;
-	int		startagain;
+	qi->qi_dquots--;
+	mutex_unlock(&qi->qi_tree_lock);
 
-	restarts = 0;
-	dqpout = NULL;
+	xfs_qm_dqdestroy(dqp);
+}
 
-	/* lockorder: hashchainlock, freelistlock, mplistlock, dqlock, dqflock */
-again:
-	startagain = 0;
-	mutex_lock(&xfs_Gqm->qm_dqfrlist_lock);
+STATIC void
+xfs_qm_dqreclaim_one(
+	struct xfs_dquot	*dqp,
+	struct list_head	*buffer_list,
+	struct list_head	*dispose_list)
+{
+	struct xfs_mount	*mp = dqp->q_mount;
+	struct xfs_quotainfo	*qi = mp->m_quotainfo;
+	int			error;
 
-	list_for_each_entry(dqp, &xfs_Gqm->qm_dqfrlist, q_freelist) {
-		struct xfs_mount *mp = dqp->q_mount;
-		xfs_dqlock(dqp);
+	if (!xfs_dqlock_nowait(dqp))
+		goto out_busy;
 
-		/*
-		 * We are racing with dqlookup here. Naturally we don't
-		 * want to reclaim a dquot that lookup wants. We release the
-		 * freelist lock and start over, so that lookup will grab
-		 * both the dquot and the freelistlock.
-		 */
-		if (dqp->dq_flags & XFS_DQ_WANT) {
-			ASSERT(! (dqp->dq_flags & XFS_DQ_INACTIVE));
-
-			trace_xfs_dqreclaim_want(dqp);
-			XQM_STATS_INC(xqmstats.xs_qm_dqwants);
-			restarts++;
-			startagain = 1;
-			goto dqunlock;
-		}
+	/*
+	 * This dquot has acquired a reference in the meantime remove it from
+	 * the freelist and try again.
+	 */
+	if (dqp->q_nrefs) {
+		xfs_dqunlock(dqp);
 
-		/*
-		 * If the dquot is inactive, we are assured that it is
-		 * not on the mplist or the hashlist, and that makes our
-		 * life easier.
-		 */
-		if (dqp->dq_flags & XFS_DQ_INACTIVE) {
-			ASSERT(mp == NULL);
-			ASSERT(! XFS_DQ_IS_DIRTY(dqp));
-			ASSERT(list_empty(&dqp->q_hashlist));
-			ASSERT(list_empty(&dqp->q_mplist));
-			list_del_init(&dqp->q_freelist);
-			xfs_Gqm->qm_dqfrlist_cnt--;
-			dqpout = dqp;
-			XQM_STATS_INC(xqmstats.xs_qm_dqinact_reclaims);
-			goto dqunlock;
-		}
+		trace_xfs_dqreclaim_want(dqp);
+		XFS_STATS_INC(xs_qm_dqwants);
 
-		ASSERT(dqp->q_hash);
-		ASSERT(!list_empty(&dqp->q_mplist));
+		list_del_init(&dqp->q_lru);
+		qi->qi_lru_count--;
+		XFS_STATS_DEC(xs_qm_dquot_unused);
+		return;
+	}
 
-		/*
-		 * Try to grab the flush lock. If this dquot is in the process
-		 * of getting flushed to disk, we don't want to reclaim it.
-		 */
-		if (!xfs_dqflock_nowait(dqp))
-			goto dqunlock;
+	/*
+	 * Try to grab the flush lock. If this dquot is in the process of
+	 * getting flushed to disk, we don't want to reclaim it.
+	 */
+	if (!xfs_dqflock_nowait(dqp))
+		goto out_busy;
 
-		/*
-		 * We have the flush lock so we know that this is not in the
-		 * process of being flushed. So, if this is dirty, flush it
-		 * DELWRI so that we don't get a freelist infested with
-		 * dirty dquots.
-		 */
-		if (XFS_DQ_IS_DIRTY(dqp)) {
-			int	error;
+	if (XFS_DQ_IS_DIRTY(dqp)) {
+		struct xfs_buf	*bp = NULL;
 
-			trace_xfs_dqreclaim_dirty(dqp);
+		trace_xfs_dqreclaim_dirty(dqp);
 
-			/*
-			 * We flush it delayed write, so don't bother
-			 * releasing the freelist lock.
-			 */
-			error = xfs_qm_dqflush(dqp, 0);
-			if (error) {
-				xfs_warn(mp, "%s: dquot %p flush failed",
-					__func__, dqp);
-			}
-			goto dqunlock;
+		error = xfs_qm_dqflush(dqp, &bp);
+		if (error) {
+			xfs_warn(mp, "%s: dquot %p flush failed",
+				 __func__, dqp);
+			goto out_busy;
 		}
 
+		xfs_buf_delwri_queue(bp, buffer_list);
+		xfs_buf_relse(bp);
 		/*
-		 * We're trying to get the hashlock out of order. This races
-		 * with dqlookup; so, we giveup and goto the next dquot if
-		 * we couldn't get the hashlock. This way, we won't starve
-		 * a dqlookup process that holds the hashlock that is
-		 * waiting for the freelist lock.
+		 * Give the dquot another try on the freelist, as the
+		 * flushing will take some time.
 		 */
-		if (!mutex_trylock(&dqp->q_hash->qh_lock)) {
-			restarts++;
-			goto dqfunlock;
-		}
+		goto out_busy;
+	}
+	xfs_dqfunlock(dqp);
 
-		/*
-		 * This races with dquot allocation code as well as dqflush_all
-		 * and reclaim code. So, if we failed to grab the mplist lock,
-		 * giveup everything and start over.
-		 */
-		if (!mutex_trylock(&mp->m_quotainfo->qi_dqlist_lock)) {
-			restarts++;
-			startagain = 1;
-			goto qhunlock;
-		}
+	/*
+	 * Prevent lookups now that we are past the point of no return.
+	 */
+	dqp->dq_flags |= XFS_DQ_FREEING;
+	xfs_dqunlock(dqp);
 
-		ASSERT(dqp->q_nrefs == 0);
-		list_del_init(&dqp->q_mplist);
-		mp->m_quotainfo->qi_dquots--;
-		mp->m_quotainfo->qi_dqreclaims++;
-		list_del_init(&dqp->q_hashlist);
-		dqp->q_hash->qh_version++;
-		list_del_init(&dqp->q_freelist);
-		xfs_Gqm->qm_dqfrlist_cnt--;
-		dqpout = dqp;
-		mutex_unlock(&mp->m_quotainfo->qi_dqlist_lock);
-qhunlock:
-		mutex_unlock(&dqp->q_hash->qh_lock);
-dqfunlock:
-		xfs_dqfunlock(dqp);
-dqunlock:
-		xfs_dqunlock(dqp);
-		if (dqpout)
-			break;
-		if (restarts >= XFS_QM_RECLAIM_MAX_RESTARTS)
-			break;
-		if (startagain) {
-			mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
-			goto again;
-		}
-	}
-	mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
-	return dqpout;
-}
+	ASSERT(dqp->q_nrefs == 0);
+	list_move_tail(&dqp->q_lru, dispose_list);
+	qi->qi_lru_count--;
+	XFS_STATS_DEC(xs_qm_dquot_unused);
 
-/*
- * Traverse the freelist of dquots and attempt to reclaim a maximum of
- * 'howmany' dquots. This operation races with dqlookup(), and attempts to
- * favor the lookup function ...
- */
-STATIC int
-xfs_qm_shake_freelist(
-	int	howmany)
-{
-	int		nreclaimed = 0;
-	xfs_dquot_t	*dqp;
+	trace_xfs_dqreclaim_done(dqp);
+	XFS_STATS_INC(xs_qm_dqreclaims);
+	return;
 
-	if (howmany <= 0)
-		return 0;
+out_busy:
+	xfs_dqunlock(dqp);
 
-	while (nreclaimed < howmany) {
-		dqp = xfs_qm_dqreclaim_one();
-		if (!dqp)
-			return nreclaimed;
-		xfs_qm_dqdestroy(dqp);
-		nreclaimed++;
-	}
-	return nreclaimed;
+	/*
+	 * Move the dquot to the tail of the list so that we don't spin on it.
+	 */
+	list_move_tail(&dqp->q_lru, &qi->qi_lru_list);
+
+	trace_xfs_dqreclaim_busy(dqp);
+	XFS_STATS_INC(xs_qm_dqreclaim_misses);
 }
 
-/*
- * The kmem_shake interface is invoked when memory is running low.
- */
-/* ARGSUSED */
 STATIC int
 xfs_qm_shake(
-	struct shrinker	*shrink,
-	struct shrink_control *sc)
+	struct shrinker		*shrink,
+	struct shrink_control	*sc)
 {
-	int	ndqused, nfree, n;
-	gfp_t gfp_mask = sc->gfp_mask;
-
-	if (!kmem_shake_allow(gfp_mask))
-		return 0;
-	if (!xfs_Gqm)
-		return 0;
-
-	nfree = xfs_Gqm->qm_dqfrlist_cnt; /* free dquots */
-	/* incore dquots in all f/s's */
-	ndqused = atomic_read(&xfs_Gqm->qm_totaldquots) - nfree;
-
-	ASSERT(ndqused >= 0);
+	struct xfs_quotainfo	*qi =
+		container_of(shrink, struct xfs_quotainfo, qi_shrinker);
+	int			nr_to_scan = sc->nr_to_scan;
+	LIST_HEAD		(buffer_list);
+	LIST_HEAD		(dispose_list);
+	struct xfs_dquot	*dqp;
+	int			error;
 
-	if (nfree <= ndqused && nfree < ndquot)
+	if ((sc->gfp_mask & (__GFP_FS|__GFP_WAIT)) != (__GFP_FS|__GFP_WAIT))
 		return 0;
+	if (!nr_to_scan)
+		goto out;
 
-	ndqused *= xfs_Gqm->qm_dqfree_ratio;	/* target # of free dquots */
-	n = nfree - ndqused - ndquot;		/* # over target */
-
-	return xfs_qm_shake_freelist(MAX(nfree, n));
-}
-
-
-/*------------------------------------------------------------------*/
+	mutex_lock(&qi->qi_lru_lock);
+	while (!list_empty(&qi->qi_lru_list)) {
+		if (nr_to_scan-- <= 0)
+			break;
+		dqp = list_first_entry(&qi->qi_lru_list, struct xfs_dquot,
+				       q_lru);
+		xfs_qm_dqreclaim_one(dqp, &buffer_list, &dispose_list);
+	}
+	mutex_unlock(&qi->qi_lru_lock);
 
-/*
- * Return a new incore dquot. Depending on the number of
- * dquots in the system, we either allocate a new one on the kernel heap,
- * or reclaim a free one.
- * Return value is B_TRUE if we allocated a new dquot, B_FALSE if we managed
- * to reclaim an existing one from the freelist.
- */
-boolean_t
-xfs_qm_dqalloc_incore(
-	xfs_dquot_t **O_dqpp)
-{
-	xfs_dquot_t	*dqp;
+	error = xfs_buf_delwri_submit(&buffer_list);
+	if (error)
+		xfs_warn(NULL, "%s: dquot reclaim failed", __func__);
 
-	/*
-	 * Check against high water mark to see if we want to pop
-	 * a nincompoop dquot off the freelist.
-	 */
-	if (atomic_read(&xfs_Gqm->qm_totaldquots) >= ndquot) {
-		/*
-		 * Try to recycle a dquot from the freelist.
-		 */
-		if ((dqp = xfs_qm_dqreclaim_one())) {
-			XQM_STATS_INC(xqmstats.xs_qm_dqreclaims);
-			/*
-			 * Just zero the core here. The rest will get
-			 * reinitialized by caller. XXX we shouldn't even
-			 * do this zero ...
-			 */
-			memset(&dqp->q_core, 0, sizeof(dqp->q_core));
-			*O_dqpp = dqp;
-			return B_FALSE;
-		}
-		XQM_STATS_INC(xqmstats.xs_qm_dqreclaim_misses);
+	while (!list_empty(&dispose_list)) {
+		dqp = list_first_entry(&dispose_list, struct xfs_dquot, q_lru);
+		list_del_init(&dqp->q_lru);
+		xfs_qm_dqfree_one(dqp);
 	}
 
-	/*
-	 * Allocate a brand new dquot on the kernel heap and return it
-	 * to the caller to initialize.
-	 */
-	ASSERT(xfs_Gqm->qm_dqzone != NULL);
-	*O_dqpp = kmem_zone_zalloc(xfs_Gqm->qm_dqzone, KM_SLEEP);
-	atomic_inc(&xfs_Gqm->qm_totaldquots);
-
-	return B_TRUE;
+out:
+	return (qi->qi_lru_count / 100) * sysctl_vfs_cache_pressure;
 }
 
-
 /*
  * Start a transaction and write the incore superblock changes to
  * disk. flags parameter indicates which fields have changed.
@@ -2151,10 +1679,7 @@ xfs_qm_vop_dqalloc(
 			 * this to caller
 			 */
 			ASSERT(ip->i_udquot);
-			uq = ip->i_udquot;
-			xfs_dqlock(uq);
-			XFS_DQHOLD(uq);
-			xfs_dqunlock(uq);
+			uq = xfs_qm_dqhold(ip->i_udquot);
 		}
 	}
 	if ((flags & XFS_QMOPT_GQUOTA) && XFS_IS_GQUOTA_ON(mp)) {
@@ -2175,10 +1700,7 @@ xfs_qm_vop_dqalloc(
 			xfs_ilock(ip, lockflags);
 		} else {
 			ASSERT(ip->i_gdquot);
-			gq = ip->i_gdquot;
-			xfs_dqlock(gq);
-			XFS_DQHOLD(gq);
-			xfs_dqunlock(gq);
+			gq = xfs_qm_dqhold(ip->i_gdquot);
 		}
 	} else if ((flags & XFS_QMOPT_PQUOTA) && XFS_IS_PQUOTA_ON(mp)) {
 		if (xfs_get_projid(ip) != prid) {
@@ -2198,10 +1720,7 @@ xfs_qm_vop_dqalloc(
 			xfs_ilock(ip, lockflags);
 		} else {
 			ASSERT(ip->i_gdquot);
-			gq = ip->i_gdquot;
-			xfs_dqlock(gq);
-			XFS_DQHOLD(gq);
-			xfs_dqunlock(gq);
+			gq = xfs_qm_dqhold(ip->i_gdquot);
 		}
 	}
 	if (uq)
@@ -2251,14 +1770,10 @@ xfs_qm_vop_chown(
 	xfs_trans_mod_dquot(tp, newdq, XFS_TRANS_DQ_ICOUNT, 1);
 
 	/*
-	 * Take an extra reference, because the inode
-	 * is going to keep this dquot pointer even
-	 * after the trans_commit.
+	 * Take an extra reference, because the inode is going to keep
+	 * this dquot pointer even after the trans_commit.
 	 */
-	xfs_dqlock(newdq);
-	XFS_DQHOLD(newdq);
-	xfs_dqunlock(newdq);
-	*IO_olddq = newdq;
+	*IO_olddq = xfs_qm_dqhold(newdq);
 
 	return prevdq;
 }
@@ -2390,25 +1905,21 @@ xfs_qm_vop_create_dqattach(
 	ASSERT(XFS_IS_QUOTA_RUNNING(mp));
 
 	if (udqp) {
-		xfs_dqlock(udqp);
-		XFS_DQHOLD(udqp);
-		xfs_dqunlock(udqp);
 		ASSERT(ip->i_udquot == NULL);
-		ip->i_udquot = udqp;
 		ASSERT(XFS_IS_UQUOTA_ON(mp));
 		ASSERT(ip->i_d.di_uid == be32_to_cpu(udqp->q_core.d_id));
+
+		ip->i_udquot = xfs_qm_dqhold(udqp);
 		xfs_trans_mod_dquot(tp, udqp, XFS_TRANS_DQ_ICOUNT, 1);
 	}
 	if (gdqp) {
-		xfs_dqlock(gdqp);
-		XFS_DQHOLD(gdqp);
-		xfs_dqunlock(gdqp);
 		ASSERT(ip->i_gdquot == NULL);
-		ip->i_gdquot = gdqp;
 		ASSERT(XFS_IS_OQUOTA_ON(mp));
 		ASSERT((XFS_IS_GQUOTA_ON(mp) ?
 			ip->i_d.di_gid : xfs_get_projid(ip)) ==
 				be32_to_cpu(gdqp->q_core.d_id));
+
+		ip->i_gdquot = xfs_qm_dqhold(gdqp);
 		xfs_trans_mod_dquot(tp, gdqp, XFS_TRANS_DQ_ICOUNT, 1);
 	}
 }
diff --git a/fs/xfs/xfs_qm.h b/fs/xfs/xfs_qm.h
index 43b9abe1052..44b858b79d7 100644
--- a/fs/xfs/xfs_qm.h
+++ b/fs/xfs/xfs_qm.h
@@ -21,39 +21,10 @@
 #include "xfs_dquot_item.h"
 #include "xfs_dquot.h"
 #include "xfs_quota_priv.h"
-#include "xfs_qm_stats.h"
 
-struct xfs_qm;
 struct xfs_inode;
 
-extern uint		ndquot;
-extern struct mutex	xfs_Gqm_lock;
-extern struct xfs_qm	*xfs_Gqm;
-extern kmem_zone_t	*qm_dqzone;
-extern kmem_zone_t	*qm_dqtrxzone;
-
-/*
- * Used in xfs_qm_sync called by xfs_sync to count the max times that it can
- * iterate over the mountpt's dquot list in one call.
- */
-#define XFS_QM_SYNC_MAX_RESTARTS	7
-
-/*
- * Ditto, for xfs_qm_dqreclaim_one.
- */
-#define XFS_QM_RECLAIM_MAX_RESTARTS	4
-
-/*
- * Ideal ratio of free to in use dquots. Quota manager makes an attempt
- * to keep this balance.
- */
-#define XFS_QM_DQFREE_RATIO		2
-
-/*
- * Dquot hashtable constants/threshold values.
- */
-#define XFS_QM_HASHSIZE_LOW		(PAGE_SIZE / sizeof(xfs_dqhash_t))
-#define XFS_QM_HASHSIZE_HIGH		((PAGE_SIZE * 4) / sizeof(xfs_dqhash_t))
+extern struct kmem_zone	*xfs_qm_dqtrxzone;
 
 /*
  * This defines the unit of allocation of dquots.
@@ -66,37 +37,20 @@ extern kmem_zone_t	*qm_dqtrxzone;
  */
 #define XFS_DQUOT_CLUSTER_SIZE_FSB	(xfs_filblks_t)1
 
-typedef xfs_dqhash_t	xfs_dqlist_t;
-
-/*
- * Quota Manager (global) structure. Lives only in core.
- */
-typedef struct xfs_qm {
-	xfs_dqlist_t	*qm_usr_dqhtable;/* udquot hash table */
-	xfs_dqlist_t	*qm_grp_dqhtable;/* gdquot hash table */
-	uint		 qm_dqhashmask;	 /* # buckets in dq hashtab - 1 */
-	struct list_head qm_dqfrlist;	 /* freelist of dquots */
-	struct mutex	 qm_dqfrlist_lock;
-	int		 qm_dqfrlist_cnt;
-	atomic_t	 qm_totaldquots; /* total incore dquots */
-	uint		 qm_nrefs;	 /* file systems with quota on */
-	int		 qm_dqfree_ratio;/* ratio of free to inuse dquots */
-	kmem_zone_t	*qm_dqzone;	 /* dquot mem-alloc zone */
-	kmem_zone_t	*qm_dqtrxzone;	 /* t_dqinfo of transactions */
-} xfs_qm_t;
-
 /*
  * Various quota information for individual filesystems.
  * The mount structure keeps a pointer to this.
  */
 typedef struct xfs_quotainfo {
+	struct radix_tree_root qi_uquota_tree;
+	struct radix_tree_root qi_gquota_tree;
+	struct mutex qi_tree_lock;
 	xfs_inode_t	*qi_uquotaip;	 /* user quota inode */
 	xfs_inode_t	*qi_gquotaip;	 /* group quota inode */
-	struct list_head qi_dqlist;	 /* all dquots in filesys */
-	struct mutex	 qi_dqlist_lock;
+	struct list_head qi_lru_list;
+	struct mutex	 qi_lru_lock;
+	int		 qi_lru_count;
 	int		 qi_dquots;
-	int		 qi_dqreclaims;	 /* a change here indicates
-					    a removal in the dqlist */
 	time_t		 qi_btimelimit;	 /* limit for blks timer */
 	time_t		 qi_itimelimit;	 /* limit for inodes timer */
 	time_t		 qi_rtbtimelimit;/* limit for rt blks timer */
@@ -112,8 +66,14 @@ typedef struct xfs_quotainfo {
 	xfs_qcnt_t	 qi_isoftlimit;	 /* default inode count soft limit */
 	xfs_qcnt_t	 qi_rtbhardlimit;/* default realtime blk hard limit */
 	xfs_qcnt_t	 qi_rtbsoftlimit;/* default realtime blk soft limit */
+	struct shrinker  qi_shrinker;
 } xfs_quotainfo_t;
 
+#define XFS_DQUOT_TREE(qi, type) \
+	((type & XFS_DQ_USER) ? \
+	 &((qi)->qi_uquota_tree) : \
+	 &((qi)->qi_gquota_tree))
+
 
 extern void	xfs_trans_mod_dquot(xfs_trans_t *, xfs_dquot_t *, uint, long);
 extern int	xfs_trans_reserve_quota_bydquots(xfs_trans_t *, xfs_mount_t *,
@@ -149,8 +109,7 @@ extern int		xfs_qm_quotacheck(xfs_mount_t *);
 extern int		xfs_qm_write_sb_changes(xfs_mount_t *, __int64_t);
 
 /* dquot stuff */
-extern boolean_t	xfs_qm_dqalloc_incore(xfs_dquot_t **);
-extern int		xfs_qm_dqpurge_all(xfs_mount_t *, uint);
+extern void		xfs_qm_dqpurge_all(xfs_mount_t *, uint);
 extern void		xfs_qm_dqrele_all_inodes(xfs_mount_t *, uint);
 
 /* quota ops */
diff --git a/fs/xfs/xfs_qm_bhv.c b/fs/xfs/xfs_qm_bhv.c
index a0a829addca..6b39115bf14 100644
--- a/fs/xfs/xfs_qm_bhv.c
+++ b/fs/xfs/xfs_qm_bhv.c
@@ -17,9 +17,7 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -40,28 +38,28 @@
 STATIC void
 xfs_fill_statvfs_from_dquot(
 	struct kstatfs		*statp,
-	xfs_disk_dquot_t	*dp)
+	struct xfs_dquot	*dqp)
 {
 	__uint64_t		limit;
 
-	limit = dp->d_blk_softlimit ?
-		be64_to_cpu(dp->d_blk_softlimit) :
-		be64_to_cpu(dp->d_blk_hardlimit);
+	limit = dqp->q_core.d_blk_softlimit ?
+		be64_to_cpu(dqp->q_core.d_blk_softlimit) :
+		be64_to_cpu(dqp->q_core.d_blk_hardlimit);
 	if (limit && statp->f_blocks > limit) {
 		statp->f_blocks = limit;
 		statp->f_bfree = statp->f_bavail =
-			(statp->f_blocks > be64_to_cpu(dp->d_bcount)) ?
-			 (statp->f_blocks - be64_to_cpu(dp->d_bcount)) : 0;
+			(statp->f_blocks > dqp->q_res_bcount) ?
+			 (statp->f_blocks - dqp->q_res_bcount) : 0;
 	}
 
-	limit = dp->d_ino_softlimit ?
-		be64_to_cpu(dp->d_ino_softlimit) :
-		be64_to_cpu(dp->d_ino_hardlimit);
+	limit = dqp->q_core.d_ino_softlimit ?
+		be64_to_cpu(dqp->q_core.d_ino_softlimit) :
+		be64_to_cpu(dqp->q_core.d_ino_hardlimit);
 	if (limit && statp->f_files > limit) {
 		statp->f_files = limit;
 		statp->f_ffree =
-			(statp->f_files > be64_to_cpu(dp->d_icount)) ?
-			 (statp->f_ffree - be64_to_cpu(dp->d_icount)) : 0;
+			(statp->f_files > dqp->q_res_icount) ?
+			 (statp->f_ffree - dqp->q_res_icount) : 0;
 	}
 }
 
@@ -82,7 +80,7 @@ xfs_qm_statvfs(
 	xfs_dquot_t		*dqp;
 
 	if (!xfs_qm_dqget(mp, NULL, xfs_get_projid(ip), XFS_DQ_PROJ, 0, &dqp)) {
-		xfs_fill_statvfs_from_dquot(statp, &dqp->q_core);
+		xfs_fill_statvfs_from_dquot(statp, dqp);
 		xfs_qm_dqput(dqp);
 	}
 }
@@ -156,21 +154,3 @@ xfs_qm_newmount(
 
 	return 0;
 }
-
-void __init
-xfs_qm_init(void)
-{
-	printk(KERN_INFO "SGI XFS Quota Management subsystem\n");
-	mutex_init(&xfs_Gqm_lock);
-	xfs_qm_init_procfs();
-}
-
-void __exit
-xfs_qm_exit(void)
-{
-	xfs_qm_cleanup_procfs();
-	if (qm_dqzone)
-		kmem_zone_destroy(qm_dqzone);
-	if (qm_dqtrxzone)
-		kmem_zone_destroy(qm_dqtrxzone);
-}
diff --git a/fs/xfs/xfs_qm_stats.c b/fs/xfs/xfs_qm_stats.c
deleted file mode 100644
index 8671a0b3264..00000000000
--- a/fs/xfs/xfs_qm_stats.c
+++ /dev/null
@@ -1,105 +0,0 @@
-/*
- * Copyright (c) 2000-2003 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write the Free Software Foundation,
- * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
- */
-#include "xfs.h"
-#include "xfs_fs.h"
-#include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_inum.h"
-#include "xfs_trans.h"
-#include "xfs_sb.h"
-#include "xfs_ag.h"
-#include "xfs_alloc.h"
-#include "xfs_quota.h"
-#include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_inode.h"
-#include "xfs_itable.h"
-#include "xfs_bmap.h"
-#include "xfs_rtalloc.h"
-#include "xfs_error.h"
-#include "xfs_attr.h"
-#include "xfs_buf_item.h"
-#include "xfs_qm.h"
-
-struct xqmstats xqmstats;
-
-static int xqm_proc_show(struct seq_file *m, void *v)
-{
-	/* maximum; incore; ratio free to inuse; freelist */
-	seq_printf(m, "%d\t%d\t%d\t%u\n",
-			ndquot,
-			xfs_Gqm? atomic_read(&xfs_Gqm->qm_totaldquots) : 0,
-			xfs_Gqm? xfs_Gqm->qm_dqfree_ratio : 0,
-			xfs_Gqm? xfs_Gqm->qm_dqfrlist_cnt : 0);
-	return 0;
-}
-
-static int xqm_proc_open(struct inode *inode, struct file *file)
-{
-	return single_open(file, xqm_proc_show, NULL);
-}
-
-static const struct file_operations xqm_proc_fops = {
-	.owner		= THIS_MODULE,
-	.open		= xqm_proc_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-};
-
-static int xqmstat_proc_show(struct seq_file *m, void *v)
-{
-	/* quota performance statistics */
-	seq_printf(m, "qm %u %u %u %u %u %u %u %u\n",
-			xqmstats.xs_qm_dqreclaims,
-			xqmstats.xs_qm_dqreclaim_misses,
-			xqmstats.xs_qm_dquot_dups,
-			xqmstats.xs_qm_dqcachemisses,
-			xqmstats.xs_qm_dqcachehits,
-			xqmstats.xs_qm_dqwants,
-			xqmstats.xs_qm_dqshake_reclaims,
-			xqmstats.xs_qm_dqinact_reclaims);
-	return 0;
-}
-
-static int xqmstat_proc_open(struct inode *inode, struct file *file)
-{
-	return single_open(file, xqmstat_proc_show, NULL);
-}
-
-static const struct file_operations xqmstat_proc_fops = {
-	.owner		= THIS_MODULE,
-	.open		= xqmstat_proc_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-};
-
-void
-xfs_qm_init_procfs(void)
-{
-	proc_create("fs/xfs/xqmstat", 0, NULL, &xqmstat_proc_fops);
-	proc_create("fs/xfs/xqm", 0, NULL, &xqm_proc_fops);
-}
-
-void
-xfs_qm_cleanup_procfs(void)
-{
-	remove_proc_entry("fs/xfs/xqm", NULL);
-	remove_proc_entry("fs/xfs/xqmstat", NULL);
-}
diff --git a/fs/xfs/xfs_qm_stats.h b/fs/xfs/xfs_qm_stats.h
deleted file mode 100644
index 5b964fc0dc0..00000000000
--- a/fs/xfs/xfs_qm_stats.h
+++ /dev/null
@@ -1,53 +0,0 @@
-/*
- * Copyright (c) 2002 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_QM_STATS_H__
-#define __XFS_QM_STATS_H__
-
-#if defined(CONFIG_PROC_FS) && !defined(XFS_STATS_OFF)
-
-/*
- * XQM global statistics
- */
-struct xqmstats {
-	__uint32_t		xs_qm_dqreclaims;
-	__uint32_t		xs_qm_dqreclaim_misses;
-	__uint32_t		xs_qm_dquot_dups;
-	__uint32_t		xs_qm_dqcachemisses;
-	__uint32_t		xs_qm_dqcachehits;
-	__uint32_t		xs_qm_dqwants;
-	__uint32_t		xs_qm_dqshake_reclaims;
-	__uint32_t		xs_qm_dqinact_reclaims;
-};
-
-extern struct xqmstats xqmstats;
-
-# define XQM_STATS_INC(count)	( (count)++ )
-
-extern void xfs_qm_init_procfs(void);
-extern void xfs_qm_cleanup_procfs(void);
-
-#else
-
-# define XQM_STATS_INC(count)	do { } while (0)
-
-static inline void xfs_qm_init_procfs(void) { };
-static inline void xfs_qm_cleanup_procfs(void) { };
-
-#endif
-
-#endif	/* __XFS_QM_STATS_H__ */
diff --git a/fs/xfs/xfs_qm_syscalls.c b/fs/xfs/xfs_qm_syscalls.c
index 5cc3dde1bc9..858a3b18611 100644
--- a/fs/xfs/xfs_qm_syscalls.c
+++ b/fs/xfs/xfs_qm_syscalls.c
@@ -22,7 +22,6 @@
 #include "xfs_fs.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -31,6 +30,7 @@
 #include "xfs_mount.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_inode.h"
+#include "xfs_inode_item.h"
 #include "xfs_itable.h"
 #include "xfs_bmap.h"
 #include "xfs_rtalloc.h"
@@ -46,9 +46,6 @@ STATIC int	xfs_qm_log_quotaoff_end(xfs_mount_t *, xfs_qoff_logitem_t *,
 					uint);
 STATIC uint	xfs_qm_export_flags(uint);
 STATIC uint	xfs_qm_export_qtype_flags(uint);
-STATIC void	xfs_qm_export_dquot(xfs_mount_t *, xfs_disk_dquot_t *,
-					fs_disk_quota_t *);
-
 
 /*
  * Turn off quota accounting and/or enforcement for all udquots and/or
@@ -68,7 +65,6 @@ xfs_qm_scall_quotaoff(
 	int			error;
 	uint			inactivate_flags;
 	xfs_qoff_logitem_t	*qoffstart;
-	int			nculprits;
 
 	/*
 	 * No file system can have quotas enabled on disk but not in core.
@@ -174,18 +170,13 @@ xfs_qm_scall_quotaoff(
 	 * This isn't protected by a particular lock directly, because we
 	 * don't want to take a mrlock every time we depend on quotas being on.
 	 */
-	mp->m_qflags &= ~(flags);
+	mp->m_qflags &= ~flags;
 
 	/*
 	 * Go through all the dquots of this file system and purge them,
-	 * according to what was turned off. We may not be able to get rid
-	 * of all dquots, because dquots can have temporary references that
-	 * are not attached to inodes. eg. xfs_setattr, xfs_create.
-	 * So, if we couldn't purge all the dquots from the filesystem,
-	 * we can't get rid of the incore data structures.
+	 * according to what was turned off.
 	 */
-	while ((nculprits = xfs_qm_dqpurge_all(mp, dqtype)))
-		delay(10 * nculprits);
+	xfs_qm_dqpurge_all(mp, dqtype);
 
 	/*
 	 * Transactions that had started before ACTIVE state bit was cleared
@@ -263,13 +254,18 @@ xfs_qm_scall_trunc_qfile(
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
 	xfs_trans_ijoin(tp, ip, 0);
 
-	error = xfs_itruncate_data(&tp, ip, 0);
+	ip->i_d.di_size = 0;
+	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+
+	error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK, 0);
 	if (error) {
 		xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES |
 				     XFS_TRANS_ABORT);
 		goto out_unlock;
 	}
 
+	ASSERT(ip->i_d.di_nextents == 0);
+
 	xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
 	error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
 
@@ -629,42 +625,6 @@ xfs_qm_scall_setqlim(
 	return error;
 }
 
-int
-xfs_qm_scall_getquota(
-	xfs_mount_t	*mp,
-	xfs_dqid_t	id,
-	uint		type,
-	fs_disk_quota_t *out)
-{
-	xfs_dquot_t	*dqp;
-	int		error;
-
-	/*
-	 * Try to get the dquot. We don't want it allocated on disk, so
-	 * we aren't passing the XFS_QMOPT_DOALLOC flag. If it doesn't
-	 * exist, we'll get ENOENT back.
-	 */
-	if ((error = xfs_qm_dqget(mp, NULL, id, type, 0, &dqp))) {
-		return (error);
-	}
-
-	/*
-	 * If everything's NULL, this dquot doesn't quite exist as far as
-	 * our utility programs are concerned.
-	 */
-	if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) {
-		xfs_qm_dqput(dqp);
-		return XFS_ERROR(ENOENT);
-	}
-	/*
-	 * Convert the disk dquot to the exportable format
-	 */
-	xfs_qm_export_dquot(mp, &dqp->q_core, out);
-	xfs_qm_dqput(dqp);
-	return (error ? XFS_ERROR(EFAULT) : 0);
-}
-
-
 STATIC int
 xfs_qm_log_quotaoff_end(
 	xfs_mount_t		*mp,
@@ -753,50 +713,66 @@ error0:
 }
 
 
-/*
- * Translate an internal style on-disk-dquot to the exportable format.
- * The main differences are that the counters/limits are all in Basic
- * Blocks (BBs) instead of the internal FSBs, and all on-disk data has
- * to be converted to the native endianness.
- */
-STATIC void
-xfs_qm_export_dquot(
-	xfs_mount_t		*mp,
-	xfs_disk_dquot_t	*src,
+int
+xfs_qm_scall_getquota(
+	struct xfs_mount	*mp,
+	xfs_dqid_t		id,
+	uint			type,
 	struct fs_disk_quota	*dst)
 {
+	struct xfs_dquot	*dqp;
+	int			error;
+
+	/*
+	 * Try to get the dquot. We don't want it allocated on disk, so
+	 * we aren't passing the XFS_QMOPT_DOALLOC flag. If it doesn't
+	 * exist, we'll get ENOENT back.
+	 */
+	error = xfs_qm_dqget(mp, NULL, id, type, 0, &dqp);
+	if (error)
+		return error;
+
+	/*
+	 * If everything's NULL, this dquot doesn't quite exist as far as
+	 * our utility programs are concerned.
+	 */
+	if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) {
+		error = XFS_ERROR(ENOENT);
+		goto out_put;
+	}
+
 	memset(dst, 0, sizeof(*dst));
-	dst->d_version = FS_DQUOT_VERSION;  /* different from src->d_version */
-	dst->d_flags = xfs_qm_export_qtype_flags(src->d_flags);
-	dst->d_id = be32_to_cpu(src->d_id);
+	dst->d_version = FS_DQUOT_VERSION;
+	dst->d_flags = xfs_qm_export_qtype_flags(dqp->q_core.d_flags);
+	dst->d_id = be32_to_cpu(dqp->q_core.d_id);
 	dst->d_blk_hardlimit =
-		XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_blk_hardlimit));
+		XFS_FSB_TO_BB(mp, be64_to_cpu(dqp->q_core.d_blk_hardlimit));
 	dst->d_blk_softlimit =
-		XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_blk_softlimit));
-	dst->d_ino_hardlimit = be64_to_cpu(src->d_ino_hardlimit);
-	dst->d_ino_softlimit = be64_to_cpu(src->d_ino_softlimit);
-	dst->d_bcount = XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_bcount));
-	dst->d_icount = be64_to_cpu(src->d_icount);
-	dst->d_btimer = be32_to_cpu(src->d_btimer);
-	dst->d_itimer = be32_to_cpu(src->d_itimer);
-	dst->d_iwarns = be16_to_cpu(src->d_iwarns);
-	dst->d_bwarns = be16_to_cpu(src->d_bwarns);
+		XFS_FSB_TO_BB(mp, be64_to_cpu(dqp->q_core.d_blk_softlimit));
+	dst->d_ino_hardlimit = be64_to_cpu(dqp->q_core.d_ino_hardlimit);
+	dst->d_ino_softlimit = be64_to_cpu(dqp->q_core.d_ino_softlimit);
+	dst->d_bcount = XFS_FSB_TO_BB(mp, dqp->q_res_bcount);
+	dst->d_icount = dqp->q_res_icount;
+	dst->d_btimer = be32_to_cpu(dqp->q_core.d_btimer);
+	dst->d_itimer = be32_to_cpu(dqp->q_core.d_itimer);
+	dst->d_iwarns = be16_to_cpu(dqp->q_core.d_iwarns);
+	dst->d_bwarns = be16_to_cpu(dqp->q_core.d_bwarns);
 	dst->d_rtb_hardlimit =
-		XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_rtb_hardlimit));
+		XFS_FSB_TO_BB(mp, be64_to_cpu(dqp->q_core.d_rtb_hardlimit));
 	dst->d_rtb_softlimit =
-		XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_rtb_softlimit));
-	dst->d_rtbcount = XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_rtbcount));
-	dst->d_rtbtimer = be32_to_cpu(src->d_rtbtimer);
-	dst->d_rtbwarns = be16_to_cpu(src->d_rtbwarns);
+		XFS_FSB_TO_BB(mp, be64_to_cpu(dqp->q_core.d_rtb_softlimit));
+	dst->d_rtbcount = XFS_FSB_TO_BB(mp, dqp->q_res_rtbcount);
+	dst->d_rtbtimer = be32_to_cpu(dqp->q_core.d_rtbtimer);
+	dst->d_rtbwarns = be16_to_cpu(dqp->q_core.d_rtbwarns);
 
 	/*
 	 * Internally, we don't reset all the timers when quota enforcement
 	 * gets turned off. No need to confuse the user level code,
 	 * so return zeroes in that case.
 	 */
-	if ((!XFS_IS_UQUOTA_ENFORCED(mp) && src->d_flags == XFS_DQ_USER) ||
+	if ((!XFS_IS_UQUOTA_ENFORCED(mp) && dqp->q_core.d_flags == XFS_DQ_USER) ||
 	    (!XFS_IS_OQUOTA_ENFORCED(mp) &&
-			(src->d_flags & (XFS_DQ_PROJ | XFS_DQ_GROUP)))) {
+			(dqp->q_core.d_flags & (XFS_DQ_PROJ | XFS_DQ_GROUP)))) {
 		dst->d_btimer = 0;
 		dst->d_itimer = 0;
 		dst->d_rtbtimer = 0;
@@ -807,16 +783,19 @@ xfs_qm_export_dquot(
 	     (XFS_IS_OQUOTA_ENFORCED(mp) &&
 			(dst->d_flags & (FS_PROJ_QUOTA | FS_GROUP_QUOTA)))) &&
 	    dst->d_id != 0) {
-		if (((int) dst->d_bcount >= (int) dst->d_blk_softlimit) &&
+		if (((int) dst->d_bcount > (int) dst->d_blk_softlimit) &&
 		    (dst->d_blk_softlimit > 0)) {
 			ASSERT(dst->d_btimer != 0);
 		}
-		if (((int) dst->d_icount >= (int) dst->d_ino_softlimit) &&
+		if (((int) dst->d_icount > (int) dst->d_ino_softlimit) &&
 		    (dst->d_ino_softlimit > 0)) {
 			ASSERT(dst->d_itimer != 0);
 		}
 	}
 #endif
+out_put:
+	xfs_qm_dqput(dqp);
+	return error;
 }
 
 STATIC uint
diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h
index a595f29567f..b50ec5b95d5 100644
--- a/fs/xfs/xfs_quota.h
+++ b/fs/xfs/xfs_quota.h
@@ -87,8 +87,7 @@ typedef struct xfs_dqblk {
 #define XFS_DQ_PROJ		0x0002		/* project quota */
 #define XFS_DQ_GROUP		0x0004		/* a group quota */
 #define XFS_DQ_DIRTY		0x0008		/* dquot is dirty */
-#define XFS_DQ_WANT		0x0010		/* for lookup/reclaim race */
-#define XFS_DQ_INACTIVE		0x0020		/* dq off mplist & hashlist */
+#define XFS_DQ_FREEING		0x0010		/* dquot is beeing torn down */
 
 #define XFS_DQ_ALLTYPES		(XFS_DQ_USER|XFS_DQ_PROJ|XFS_DQ_GROUP)
 
@@ -97,8 +96,7 @@ typedef struct xfs_dqblk {
 	{ XFS_DQ_PROJ,		"PROJ" }, \
 	{ XFS_DQ_GROUP,		"GROUP" }, \
 	{ XFS_DQ_DIRTY,		"DIRTY" }, \
-	{ XFS_DQ_WANT,		"WANT" }, \
-	{ XFS_DQ_INACTIVE,	"INACTIVE" }
+	{ XFS_DQ_FREEING,	"FREEING" }
 
 /*
  * In the worst case, when both user and group quotas are on,
@@ -176,6 +174,8 @@ typedef struct xfs_qoff_logformat {
 #define XFS_UQUOTA_ACTIVE	0x0100  /* uquotas are being turned off */
 #define XFS_PQUOTA_ACTIVE	0x0200  /* pquotas are being turned off */
 #define XFS_GQUOTA_ACTIVE	0x0400  /* gquotas are being turned off */
+#define XFS_ALL_QUOTA_ACTIVE	\
+	(XFS_UQUOTA_ACTIVE | XFS_PQUOTA_ACTIVE | XFS_GQUOTA_ACTIVE)
 
 /*
  * Checking XFS_IS_*QUOTA_ON() while holding any inode lock guarantees
@@ -199,7 +199,6 @@ typedef struct xfs_qoff_logformat {
 #define XFS_QMOPT_UQUOTA	0x0000004 /* user dquot requested */
 #define XFS_QMOPT_PQUOTA	0x0000008 /* project dquot requested */
 #define XFS_QMOPT_FORCE_RES	0x0000010 /* ignore quota limits */
-#define XFS_QMOPT_DQSUSER	0x0000020 /* don't cache super users dquot */
 #define XFS_QMOPT_SBVERSION	0x0000040 /* change superblock version num */
 #define XFS_QMOPT_DOWARN        0x0000400 /* increase warning cnt if needed */
 #define XFS_QMOPT_DQREPAIR	0x0001000 /* repair dquot if damaged */
@@ -326,7 +325,6 @@ extern int xfs_qm_dqattach_locked(struct xfs_inode *, uint);
 extern void xfs_qm_dqdetach(struct xfs_inode *);
 extern void xfs_qm_dqrele(struct xfs_dquot *);
 extern void xfs_qm_statvfs(struct xfs_inode *, struct kstatfs *);
-extern int xfs_qm_sync(struct xfs_mount *, int);
 extern int xfs_qm_newmount(struct xfs_mount *, uint *, uint *);
 extern void xfs_qm_mount_quotas(struct xfs_mount *);
 extern void xfs_qm_unmount(struct xfs_mount *);
@@ -366,10 +364,6 @@ static inline int xfs_trans_reserve_quota_bydquots(struct xfs_trans *tp,
 #define xfs_qm_dqdetach(ip)
 #define xfs_qm_dqrele(d)
 #define xfs_qm_statvfs(ip, s)
-static inline int xfs_qm_sync(struct xfs_mount *mp, int flags)
-{
-	return 0;
-}
 #define xfs_qm_newmount(mp, a, b)					(0)
 #define xfs_qm_mount_quotas(mp)
 #define xfs_qm_unmount(mp)
diff --git a/fs/xfs/xfs_quota_priv.h b/fs/xfs/xfs_quota_priv.h
index 94a3d927d71..6d86219d93d 100644
--- a/fs/xfs/xfs_quota_priv.h
+++ b/fs/xfs/xfs_quota_priv.h
@@ -24,17 +24,6 @@
  */
 #define XFS_DQITER_MAP_SIZE	10
 
-/*
- * Hash into a bucket in the dquot hash table, based on <mp, id>.
- */
-#define XFS_DQ_HASHVAL(mp, id) (((__psunsigned_t)(mp) + \
-				 (__psunsigned_t)(id)) & \
-				(xfs_Gqm->qm_dqhashmask - 1))
-#define XFS_DQ_HASH(mp, id, type)   (type == XFS_DQ_USER ? \
-				     (xfs_Gqm->qm_usr_dqhtable + \
-				      XFS_DQ_HASHVAL(mp, id)) : \
-				     (xfs_Gqm->qm_grp_dqhtable + \
-				      XFS_DQ_HASHVAL(mp, id)))
 #define XFS_IS_DQUOT_UNINITIALIZED(dqp) ( \
 	!dqp->q_core.d_blk_hardlimit && \
 	!dqp->q_core.d_blk_softlimit && \
diff --git a/fs/xfs/xfs_quotaops.c b/fs/xfs/xfs_quotaops.c
index 7e76f537abb..fed504fc299 100644
--- a/fs/xfs/xfs_quotaops.c
+++ b/fs/xfs/xfs_quotaops.c
@@ -17,7 +17,6 @@
  */
 #include "xfs.h"
 #include "xfs_sb.h"
-#include "xfs_inum.h"
 #include "xfs_log.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
diff --git a/fs/xfs/xfs_rename.c b/fs/xfs/xfs_rename.c
index 866de277079..30ff5f401d2 100644
--- a/fs/xfs/xfs_rename.c
+++ b/fs/xfs/xfs_rename.c
@@ -19,7 +19,6 @@
 #include "xfs_fs.h"
 #include "xfs_types.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -118,17 +117,6 @@ xfs_rename(
 	new_parent = (src_dp != target_dp);
 	src_is_directory = S_ISDIR(src_ip->i_d.di_mode);
 
-	if (src_is_directory) {
-		/*
-		 * Check for link count overflow on target_dp
-		 */
-		if (target_ip == NULL && new_parent &&
-		    target_dp->i_d.di_nlink >= XFS_MAXLINK) {
-			error = XFS_ERROR(EMLINK);
-			goto std_return;
-		}
-	}
-
 	xfs_sort_for_rename(src_dp, target_dp, src_ip, target_ip,
 				inodes, &num_inodes);
 
diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c
index 87323f1ded6..92d4331cd4f 100644
--- a/fs/xfs/xfs_rtalloc.c
+++ b/fs/xfs/xfs_rtalloc.c
@@ -20,7 +20,6 @@
 #include "xfs_types.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -34,7 +33,6 @@
 #include "xfs_rtalloc.h"
 #include "xfs_fsops.h"
 #include "xfs_error.h"
-#include "xfs_rw.h"
 #include "xfs_inode_item.h"
 #include "xfs_trans_space.h"
 #include "xfs_utils.h"
@@ -183,6 +181,7 @@ error_cancel:
 		oblocks = map.br_startoff + map.br_blockcount;
 	}
 	return 0;
+
 error:
 	return error;
 }
@@ -1871,9 +1870,9 @@ xfs_growfs_rt(
 	/*
 	 * Read in the last block of the device, make sure it exists.
 	 */
-	bp = xfs_buf_read_uncached(mp, mp->m_rtdev_targp,
+	bp = xfs_buf_read_uncached(mp->m_rtdev_targp,
 				XFS_FSB_TO_BB(mp, nrblocks - 1),
-				XFS_FSB_TO_B(mp, 1), 0);
+				XFS_FSB_TO_BB(mp, 1), 0);
 	if (!bp)
 		return EIO;
 	xfs_buf_relse(bp);
@@ -2139,11 +2138,9 @@ xfs_rtfree_extent(
 	xfs_buf_t	*sumbp;		/* summary file block buffer */
 
 	mp = tp->t_mountp;
-	/*
-	 * Synchronize by locking the bitmap inode.
-	 */
-	xfs_ilock(mp->m_rbmip, XFS_ILOCK_EXCL);
-	xfs_trans_ijoin(tp, mp->m_rbmip, XFS_ILOCK_EXCL);
+
+	ASSERT(mp->m_rbmip->i_itemp != NULL);
+	ASSERT(xfs_isilocked(mp->m_rbmip, XFS_ILOCK_EXCL));
 
 #if defined(__KERNEL__) && defined(DEBUG)
 	/*
@@ -2220,9 +2217,9 @@ xfs_rtmount_init(
 			(unsigned long long) mp->m_sb.sb_rblocks);
 		return XFS_ERROR(EFBIG);
 	}
-	bp = xfs_buf_read_uncached(mp, mp->m_rtdev_targp,
+	bp = xfs_buf_read_uncached(mp->m_rtdev_targp,
 					d - XFS_FSB_TO_BB(mp, 1),
-					XFS_FSB_TO_B(mp, 1), 0);
+					XFS_FSB_TO_BB(mp, 1), 0);
 	if (!bp) {
 		xfs_warn(mp, "realtime device size check failed");
 		return EIO;
diff --git a/fs/xfs/xfs_rw.c b/fs/xfs/xfs_rw.c
deleted file mode 100644
index 597d044a09a..00000000000
--- a/fs/xfs/xfs_rw.c
+++ /dev/null
@@ -1,156 +0,0 @@
-/*
- * Copyright (c) 2000-2006 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
- */
-#include "xfs.h"
-#include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_inum.h"
-#include "xfs_trans.h"
-#include "xfs_sb.h"
-#include "xfs_ag.h"
-#include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dinode.h"
-#include "xfs_inode.h"
-#include "xfs_error.h"
-#include "xfs_rw.h"
-
-/*
- * Force a shutdown of the filesystem instantly while keeping
- * the filesystem consistent. We don't do an unmount here; just shutdown
- * the shop, make sure that absolutely nothing persistent happens to
- * this filesystem after this point.
- */
-void
-xfs_do_force_shutdown(
-	xfs_mount_t	*mp,
-	int		flags,
-	char		*fname,
-	int		lnnum)
-{
-	int		logerror;
-
-	logerror = flags & SHUTDOWN_LOG_IO_ERROR;
-
-	if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
-		xfs_notice(mp,
-	"%s(0x%x) called from line %d of file %s.  Return address = 0x%p",
-			__func__, flags, lnnum, fname, __return_address);
-	}
-	/*
-	 * No need to duplicate efforts.
-	 */
-	if (XFS_FORCED_SHUTDOWN(mp) && !logerror)
-		return;
-
-	/*
-	 * This flags XFS_MOUNT_FS_SHUTDOWN, makes sure that we don't
-	 * queue up anybody new on the log reservations, and wakes up
-	 * everybody who's sleeping on log reservations to tell them
-	 * the bad news.
-	 */
-	if (xfs_log_force_umount(mp, logerror))
-		return;
-
-	if (flags & SHUTDOWN_CORRUPT_INCORE) {
-		xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_CORRUPT,
-    "Corruption of in-memory data detected.  Shutting down filesystem");
-		if (XFS_ERRLEVEL_HIGH <= xfs_error_level)
-			xfs_stack_trace();
-	} else if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
-		if (logerror) {
-			xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_LOGERROR,
-		"Log I/O Error Detected.  Shutting down filesystem");
-		} else if (flags & SHUTDOWN_DEVICE_REQ) {
-			xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_IOERROR,
-		"All device paths lost.  Shutting down filesystem");
-		} else if (!(flags & SHUTDOWN_REMOTE_REQ)) {
-			xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_IOERROR,
-		"I/O Error Detected. Shutting down filesystem");
-		}
-	}
-	if (!(flags & SHUTDOWN_FORCE_UMOUNT)) {
-		xfs_alert(mp,
-	"Please umount the filesystem and rectify the problem(s)");
-	}
-}
-
-/*
- * This isn't an absolute requirement, but it is
- * just a good idea to call xfs_read_buf instead of
- * directly doing a read_buf call. For one, we shouldn't
- * be doing this disk read if we are in SHUTDOWN state anyway,
- * so this stops that from happening. Secondly, this does all
- * the error checking stuff and the brelse if appropriate for
- * the caller, so the code can be a little leaner.
- */
-
-int
-xfs_read_buf(
-	struct xfs_mount *mp,
-	xfs_buftarg_t	 *target,
-	xfs_daddr_t	 blkno,
-	int              len,
-	uint             flags,
-	xfs_buf_t	 **bpp)
-{
-	xfs_buf_t	 *bp;
-	int		 error;
-
-	if (!flags)
-		flags = XBF_LOCK | XBF_MAPPED;
-
-	bp = xfs_buf_read(target, blkno, len, flags);
-	if (!bp)
-		return XFS_ERROR(EIO);
-	error = bp->b_error;
-	if (!error && !XFS_FORCED_SHUTDOWN(mp)) {
-		*bpp = bp;
-	} else {
-		*bpp = NULL;
-		if (error) {
-			xfs_buf_ioerror_alert(bp, __func__);
-		} else {
-			error = XFS_ERROR(EIO);
-		}
-		if (bp) {
-			XFS_BUF_UNDONE(bp);
-			xfs_buf_stale(bp);
-			/*
-			 * brelse clears B_ERROR and b_error
-			 */
-			xfs_buf_relse(bp);
-		}
-	}
-	return (error);
-}
-
-/*
- * helper function to extract extent size hint from inode
- */
-xfs_extlen_t
-xfs_get_extsz_hint(
-	struct xfs_inode	*ip)
-{
-	if ((ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE) && ip->i_d.di_extsize)
-		return ip->i_d.di_extsize;
-	if (XFS_IS_REALTIME_INODE(ip))
-		return ip->i_mount->m_sb.sb_rextsize;
-	return 0;
-}
diff --git a/fs/xfs/xfs_rw.h b/fs/xfs/xfs_rw.h
deleted file mode 100644
index bbdb9ad6a4b..00000000000
--- a/fs/xfs/xfs_rw.h
+++ /dev/null
@@ -1,47 +0,0 @@
-/*
- * Copyright (c) 2000-2006 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_RW_H__
-#define	__XFS_RW_H__
-
-struct xfs_buf;
-struct xfs_inode;
-struct xfs_mount;
-
-/*
- * Convert the given file system block to a disk block.
- * We have to treat it differently based on whether the
- * file is a real time file or not, because the bmap code
- * does.
- */
-static inline xfs_daddr_t
-xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb)
-{
-	return (XFS_IS_REALTIME_INODE(ip) ? \
-		 (xfs_daddr_t)XFS_FSB_TO_BB((ip)->i_mount, (fsb)) : \
-		 XFS_FSB_TO_DADDR((ip)->i_mount, (fsb)));
-}
-
-/*
- * Prototypes for functions in xfs_rw.c.
- */
-extern int xfs_read_buf(struct xfs_mount *mp, xfs_buftarg_t *btp,
-			xfs_daddr_t blkno, int len, uint flags,
-			struct xfs_buf **bpp);
-extern xfs_extlen_t xfs_get_extsz_hint(struct xfs_inode *ip);
-
-#endif /* __XFS_RW_H__ */
diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/xfs_sb.h
index cb6ae715814..f429d9d5d32 100644
--- a/fs/xfs/xfs_sb.h
+++ b/fs/xfs/xfs_sb.h
@@ -529,7 +529,6 @@ static inline int xfs_sb_version_hasprojid32bit(xfs_sb_t *sbp)
 #define	XFS_BB_TO_FSB(mp,bb)	\
 	(((bb) + (XFS_FSB_TO_BB(mp,1) - 1)) >> (mp)->m_blkbb_log)
 #define	XFS_BB_TO_FSBT(mp,bb)	((bb) >> (mp)->m_blkbb_log)
-#define	XFS_BB_FSB_OFFSET(mp,bb) ((bb) & ((mp)->m_bsize - 1))
 
 /*
  * File system block to byte conversions.
diff --git a/fs/xfs/xfs_stats.c b/fs/xfs/xfs_stats.c
index 76fdc586193..ce372b7d564 100644
--- a/fs/xfs/xfs_stats.c
+++ b/fs/xfs/xfs_stats.c
@@ -20,9 +20,18 @@
 
 DEFINE_PER_CPU(struct xfsstats, xfsstats);
 
+static int counter_val(int idx)
+{
+	int val = 0, cpu;
+
+	for_each_possible_cpu(cpu)
+		val += *(((__u32 *)&per_cpu(xfsstats, cpu) + idx));
+	return val;
+}
+
 static int xfs_stat_proc_show(struct seq_file *m, void *v)
 {
-	int		c, i, j, val;
+	int		i, j;
 	__uint64_t	xs_xstrat_bytes = 0;
 	__uint64_t	xs_write_bytes = 0;
 	__uint64_t	xs_read_bytes = 0;
@@ -50,20 +59,16 @@ static int xfs_stat_proc_show(struct seq_file *m, void *v)
 		{ "abtc2",		XFSSTAT_END_ABTC_V2		},
 		{ "bmbt2",		XFSSTAT_END_BMBT_V2		},
 		{ "ibt2",		XFSSTAT_END_IBT_V2		},
+		/* we print both series of quota information together */
+		{ "qm",			XFSSTAT_END_QM			},
 	};
 
 	/* Loop over all stats groups */
-	for (i=j = 0; i < ARRAY_SIZE(xstats); i++) {
+	for (i = j = 0; i < ARRAY_SIZE(xstats); i++) {
 		seq_printf(m, "%s", xstats[i].desc);
 		/* inner loop does each group */
-		while (j < xstats[i].endpoint) {
-			val = 0;
-			/* sum over all cpus */
-			for_each_possible_cpu(c)
-				val += *(((__u32*)&per_cpu(xfsstats, c) + j));
-			seq_printf(m, " %u", val);
-			j++;
-		}
+		for (; j < xstats[i].endpoint; j++)
+			seq_printf(m, " %u", counter_val(j));
 		seq_putc(m, '\n');
 	}
 	/* extra precision counters */
@@ -97,6 +102,58 @@ static const struct file_operations xfs_stat_proc_fops = {
 	.release	= single_release,
 };
 
+/* legacy quota interfaces */
+#ifdef CONFIG_XFS_QUOTA
+static int xqm_proc_show(struct seq_file *m, void *v)
+{
+	/* maximum; incore; ratio free to inuse; freelist */
+	seq_printf(m, "%d\t%d\t%d\t%u\n",
+			0,
+			counter_val(XFSSTAT_END_XQMSTAT),
+			0,
+			counter_val(XFSSTAT_END_XQMSTAT + 1));
+	return 0;
+}
+
+static int xqm_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, xqm_proc_show, NULL);
+}
+
+static const struct file_operations xqm_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= xqm_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+/* legacy quota stats interface no 2 */
+static int xqmstat_proc_show(struct seq_file *m, void *v)
+{
+	int j;
+
+	seq_printf(m, "qm");
+	for (j = XFSSTAT_END_IBT_V2; j < XFSSTAT_END_XQMSTAT; j++)
+		seq_printf(m, " %u", counter_val(j));
+	seq_putc(m, '\n');
+	return 0;
+}
+
+static int xqmstat_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, xqmstat_proc_show, NULL);
+}
+
+static const struct file_operations xqmstat_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= xqmstat_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+#endif /* CONFIG_XFS_QUOTA */
+
 int
 xfs_init_procfs(void)
 {
@@ -105,10 +162,24 @@ xfs_init_procfs(void)
 
 	if (!proc_create("fs/xfs/stat", 0, NULL,
 			 &xfs_stat_proc_fops))
-		goto out_remove_entry;
+		goto out_remove_xfs_dir;
+#ifdef CONFIG_XFS_QUOTA
+	if (!proc_create("fs/xfs/xqmstat", 0, NULL,
+			 &xqmstat_proc_fops))
+		goto out_remove_stat_file;
+	if (!proc_create("fs/xfs/xqm", 0, NULL,
+			 &xqm_proc_fops))
+		goto out_remove_xqmstat_file;
+#endif
 	return 0;
 
- out_remove_entry:
+#ifdef CONFIG_XFS_QUOTA
+ out_remove_xqmstat_file:
+	remove_proc_entry("fs/xfs/xqmstat", NULL);
+ out_remove_stat_file:
+	remove_proc_entry("fs/xfs/stat", NULL);
+#endif
+ out_remove_xfs_dir:
 	remove_proc_entry("fs/xfs", NULL);
  out:
 	return -ENOMEM;
@@ -117,6 +188,10 @@ xfs_init_procfs(void)
 void
 xfs_cleanup_procfs(void)
 {
+#ifdef CONFIG_XFS_QUOTA
+	remove_proc_entry("fs/xfs/xqm", NULL);
+	remove_proc_entry("fs/xfs/xqmstat", NULL);
+#endif
 	remove_proc_entry("fs/xfs/stat", NULL);
 	remove_proc_entry("fs/xfs", NULL);
 }
diff --git a/fs/xfs/xfs_stats.h b/fs/xfs/xfs_stats.h
index 736854b1ca1..c03ad38ceae 100644
--- a/fs/xfs/xfs_stats.h
+++ b/fs/xfs/xfs_stats.h
@@ -183,6 +183,16 @@ struct xfsstats {
 	__uint32_t		xs_ibt_2_alloc;
 	__uint32_t		xs_ibt_2_free;
 	__uint32_t		xs_ibt_2_moves;
+#define XFSSTAT_END_XQMSTAT		(XFSSTAT_END_IBT_V2+6)
+	__uint32_t		xs_qm_dqreclaims;
+	__uint32_t		xs_qm_dqreclaim_misses;
+	__uint32_t		xs_qm_dquot_dups;
+	__uint32_t		xs_qm_dqcachemisses;
+	__uint32_t		xs_qm_dqcachehits;
+	__uint32_t		xs_qm_dqwants;
+#define XFSSTAT_END_QM			(XFSSTAT_END_XQMSTAT+2)
+	__uint32_t		xs_qm_dquot;
+	__uint32_t		xs_qm_dquot_unused;
 /* Extra precision counters */
 	__uint64_t		xs_xstrat_bytes;
 	__uint64_t		xs_write_bytes;
diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
index 8a899496fd5..0d9de41a715 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -17,7 +17,6 @@
  */
 
 #include "xfs.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
 #include "xfs_inum.h"
 #include "xfs_trans.h"
@@ -199,7 +198,6 @@ xfs_parseargs(
 	mp->m_flags |= XFS_MOUNT_BARRIER;
 	mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
 	mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
-	mp->m_flags |= XFS_MOUNT_DELAYLOG;
 
 	/*
 	 * These can be overridden by the mount option parsing.
@@ -325,10 +323,9 @@ xfs_parseargs(
 		} else if (!strcmp(this_char, MNTOPT_FILESTREAM)) {
 			mp->m_flags |= XFS_MOUNT_FILESTREAMS;
 		} else if (!strcmp(this_char, MNTOPT_NOQUOTA)) {
-			mp->m_qflags &= ~(XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE |
-					  XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE |
-					  XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE |
-					  XFS_UQUOTA_ENFD | XFS_OQUOTA_ENFD);
+			mp->m_qflags &= ~XFS_ALL_QUOTA_ACCT;
+			mp->m_qflags &= ~XFS_ALL_QUOTA_ENFD;
+			mp->m_qflags &= ~XFS_ALL_QUOTA_ACTIVE;
 		} else if (!strcmp(this_char, MNTOPT_QUOTA) ||
 			   !strcmp(this_char, MNTOPT_UQUOTA) ||
 			   !strcmp(this_char, MNTOPT_USRQUOTA)) {
@@ -353,11 +350,11 @@ xfs_parseargs(
 			mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE);
 			mp->m_qflags &= ~XFS_OQUOTA_ENFD;
 		} else if (!strcmp(this_char, MNTOPT_DELAYLOG)) {
-			mp->m_flags |= XFS_MOUNT_DELAYLOG;
+			xfs_warn(mp,
+	"delaylog is the default now, option is deprecated.");
 		} else if (!strcmp(this_char, MNTOPT_NODELAYLOG)) {
-			mp->m_flags &= ~XFS_MOUNT_DELAYLOG;
 			xfs_warn(mp,
-	"nodelaylog is deprecated and will be removed in Linux 3.3");
+	"nodelaylog support has been removed, option is deprecated.");
 		} else if (!strcmp(this_char, MNTOPT_DISCARD)) {
 			mp->m_flags |= XFS_MOUNT_DISCARD;
 		} else if (!strcmp(this_char, MNTOPT_NODISCARD)) {
@@ -395,13 +392,6 @@ xfs_parseargs(
 		return EINVAL;
 	}
 
-	if ((mp->m_flags & XFS_MOUNT_DISCARD) &&
-	    !(mp->m_flags & XFS_MOUNT_DELAYLOG)) {
-		xfs_warn(mp,
-	"the discard option is incompatible with the nodelaylog option");
-		return EINVAL;
-	}
-
 #ifndef CONFIG_XFS_QUOTA
 	if (XFS_IS_QUOTA_RUNNING(mp)) {
 		xfs_warn(mp, "quota support not available in this kernel.");
@@ -501,7 +491,6 @@ xfs_showargs(
 		{ XFS_MOUNT_ATTR2,		"," MNTOPT_ATTR2 },
 		{ XFS_MOUNT_FILESTREAMS,	"," MNTOPT_FILESTREAM },
 		{ XFS_MOUNT_GRPID,		"," MNTOPT_GRPID },
-		{ XFS_MOUNT_DELAYLOG,		"," MNTOPT_DELAYLOG },
 		{ XFS_MOUNT_DISCARD,		"," MNTOPT_DISCARD },
 		{ 0, NULL }
 	};
@@ -632,7 +621,7 @@ void
 xfs_blkdev_issue_flush(
 	xfs_buftarg_t		*buftarg)
 {
-	blkdev_issue_flush(buftarg->bt_bdev, GFP_KERNEL, NULL);
+	blkdev_issue_flush(buftarg->bt_bdev, GFP_NOFS, NULL);
 }
 
 STATIC void
@@ -769,6 +758,43 @@ xfs_setup_devices(
 	return 0;
 }
 
+STATIC int
+xfs_init_mount_workqueues(
+	struct xfs_mount	*mp)
+{
+	mp->m_data_workqueue = alloc_workqueue("xfs-data/%s",
+			WQ_MEM_RECLAIM, 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);
+	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);
+	if (!mp->m_cil_workqueue)
+		goto out_destroy_unwritten;
+	return 0;
+
+out_destroy_unwritten:
+	destroy_workqueue(mp->m_unwritten_workqueue);
+out_destroy_data_iodone_queue:
+	destroy_workqueue(mp->m_data_workqueue);
+out:
+	return -ENOMEM;
+}
+
+STATIC void
+xfs_destroy_mount_workqueues(
+	struct xfs_mount	*mp)
+{
+	destroy_workqueue(mp->m_cil_workqueue);
+	destroy_workqueue(mp->m_data_workqueue);
+	destroy_workqueue(mp->m_unwritten_workqueue);
+}
+
 /* Catch misguided souls that try to use this interface on XFS */
 STATIC struct inode *
 xfs_fs_alloc_inode(
@@ -837,105 +863,64 @@ xfs_fs_inode_init_once(
 	/* xfs inode */
 	atomic_set(&ip->i_pincount, 0);
 	spin_lock_init(&ip->i_flags_lock);
-	init_waitqueue_head(&ip->i_ipin_wait);
-	/*
-	 * Because we want to use a counting completion, complete
-	 * the flush completion once to allow a single access to
-	 * the flush completion without blocking.
-	 */
-	init_completion(&ip->i_flush);
-	complete(&ip->i_flush);
 
 	mrlock_init(&ip->i_lock, MRLOCK_ALLOW_EQUAL_PRI|MRLOCK_BARRIER,
 		     "xfsino", ip->i_ino);
 }
 
 /*
- * Dirty the XFS inode when mark_inode_dirty_sync() is called so that
- * we catch unlogged VFS level updates to the inode.
+ * This is called by the VFS when dirtying inode metadata.  This can happen
+ * for a few reasons, but we only care about timestamp updates, given that
+ * we handled the rest ourselves.  In theory no other calls should happen,
+ * but for example generic_write_end() keeps dirtying the inode after
+ * updating i_size.  Thus we check that the flags are exactly I_DIRTY_SYNC,
+ * and skip this call otherwise.
  *
- * We need the barrier() to maintain correct ordering between unlogged
- * updates and the transaction commit code that clears the i_update_core
- * field. This requires all updates to be completed before marking the
- * inode dirty.
+ * We'll hopefull get a different method just for updating timestamps soon,
+ * at which point this hack can go away, and maybe we'll also get real
+ * error handling here.
  */
 STATIC void
 xfs_fs_dirty_inode(
-	struct inode	*inode,
-	int		flags)
-{
-	barrier();
-	XFS_I(inode)->i_update_core = 1;
-}
-
-STATIC int
-xfs_fs_write_inode(
 	struct inode		*inode,
-	struct writeback_control *wbc)
+	int			flags)
 {
 	struct xfs_inode	*ip = XFS_I(inode);
 	struct xfs_mount	*mp = ip->i_mount;
-	int			error = EAGAIN;
+	struct xfs_trans	*tp;
+	int			error;
 
-	trace_xfs_write_inode(ip);
+	if (flags != I_DIRTY_SYNC)
+		return;
 
-	if (XFS_FORCED_SHUTDOWN(mp))
-		return -XFS_ERROR(EIO);
+	trace_xfs_dirty_inode(ip);
 
-	if (wbc->sync_mode == WB_SYNC_ALL || wbc->for_kupdate) {
-		/*
-		 * Make sure the inode has made it it into the log.  Instead
-		 * of forcing it all the way to stable storage using a
-		 * synchronous transaction we let the log force inside the
-		 * ->sync_fs call do that for thus, which reduces the number
-		 * of synchronous log forces dramatically.
-		 */
-		error = xfs_log_dirty_inode(ip, NULL, 0);
-		if (error)
-			goto out;
-		return 0;
-	} else {
-		if (!ip->i_update_core)
-			return 0;
-
-		/*
-		 * We make this non-blocking if the inode is contended, return
-		 * EAGAIN to indicate to the caller that they did not succeed.
-		 * This prevents the flush path from blocking on inodes inside
-		 * another operation right now, they get caught later by
-		 * xfs_sync.
-		 */
-		if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED))
-			goto out;
-
-		if (xfs_ipincount(ip) || !xfs_iflock_nowait(ip))
-			goto out_unlock;
-
-		/*
-		 * Now we have the flush lock and the inode is not pinned, we
-		 * can check if the inode is really clean as we know that
-		 * there are no pending transaction completions, it is not
-		 * waiting on the delayed write queue and there is no IO in
-		 * progress.
-		 */
-		if (xfs_inode_clean(ip)) {
-			xfs_ifunlock(ip);
-			error = 0;
-			goto out_unlock;
-		}
-		error = xfs_iflush(ip, SYNC_TRYLOCK);
+	tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS);
+	error = xfs_trans_reserve(tp, 0, XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0);
+	if (error) {
+		xfs_trans_cancel(tp, 0);
+		goto trouble;
 	}
-
- out_unlock:
-	xfs_iunlock(ip, XFS_ILOCK_SHARED);
- out:
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
 	/*
-	 * if we failed to write out the inode then mark
-	 * it dirty again so we'll try again later.
+	 * Grab all the latest timestamps from the Linux inode.
 	 */
+	ip->i_d.di_atime.t_sec = (__int32_t)inode->i_atime.tv_sec;
+	ip->i_d.di_atime.t_nsec = (__int32_t)inode->i_atime.tv_nsec;
+	ip->i_d.di_ctime.t_sec = (__int32_t)inode->i_ctime.tv_sec;
+	ip->i_d.di_ctime.t_nsec = (__int32_t)inode->i_ctime.tv_nsec;
+	ip->i_d.di_mtime.t_sec = (__int32_t)inode->i_mtime.tv_sec;
+	ip->i_d.di_mtime.t_nsec = (__int32_t)inode->i_mtime.tv_nsec;
+
+	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+	xfs_trans_log_inode(tp, ip, XFS_ILOG_TIMESTAMP);
+	error = xfs_trans_commit(tp, 0);
 	if (error)
-		xfs_mark_inode_dirty_sync(ip);
-	return -error;
+		goto trouble;
+	return;
+
+trouble:
+	xfs_warn(mp, "failed to update timestamps for inode 0x%llx", ip->i_ino);
 }
 
 STATIC void
@@ -947,7 +932,7 @@ xfs_fs_evict_inode(
 	trace_xfs_evict_inode(ip);
 
 	truncate_inode_pages(&inode->i_data, 0);
-	end_writeback(inode);
+	clear_inode(inode);
 	XFS_STATS_INC(vn_rele);
 	XFS_STATS_INC(vn_remove);
 	XFS_STATS_DEC(vn_active);
@@ -971,6 +956,22 @@ xfs_fs_evict_inode(
 	xfs_inactive(ip);
 }
 
+/*
+ * We do an unlocked check for XFS_IDONTCACHE here because we are already
+ * serialised against cache hits here via the inode->i_lock and igrab() in
+ * xfs_iget_cache_hit(). Hence a lookup that might clear this flag will not be
+ * racing with us, and it avoids needing to grab a spinlock here for every inode
+ * we drop the final reference on.
+ */
+STATIC int
+xfs_fs_drop_inode(
+	struct inode		*inode)
+{
+	struct xfs_inode	*ip = XFS_I(inode);
+
+	return generic_drop_inode(inode) || (ip->i_flags & XFS_IDONTCACHE);
+}
+
 STATIC void
 xfs_free_fsname(
 	struct xfs_mount	*mp)
@@ -986,20 +987,12 @@ xfs_fs_put_super(
 {
 	struct xfs_mount	*mp = XFS_M(sb);
 
-	xfs_syncd_stop(mp);
-
-	/*
-	 * Blow away any referenced inode in the filestreams cache.
-	 * This can and will cause log traffic as inodes go inactive
-	 * here.
-	 */
 	xfs_filestream_unmount(mp);
-
-	xfs_flush_buftarg(mp->m_ddev_targp, 1);
-
 	xfs_unmountfs(mp);
+	xfs_syncd_stop(mp);
 	xfs_freesb(mp);
 	xfs_icsb_destroy_counters(mp);
+	xfs_destroy_mount_workqueues(mp);
 	xfs_close_devices(mp);
 	xfs_free_fsname(mp);
 	kfree(mp);
@@ -1014,17 +1007,10 @@ xfs_fs_sync_fs(
 	int			error;
 
 	/*
-	 * Not much we can do for the first async pass.  Writing out the
-	 * superblock would be counter-productive as we are going to redirty
-	 * when writing out other data and metadata (and writing out a single
-	 * block is quite fast anyway).
-	 *
-	 * Try to asynchronously kick off quota syncing at least.
+	 * Doing anything during the async pass would be counterproductive.
 	 */
-	if (!wait) {
-		xfs_qm_sync(mp, SYNC_TRYLOCK);
+	if (!wait)
 		return 0;
-	}
 
 	error = xfs_quiesce_data(mp);
 	if (error)
@@ -1083,7 +1069,7 @@ xfs_fs_statfs(
 
 	spin_unlock(&mp->m_sb_lock);
 
-	if ((ip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) ||
+	if ((ip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) &&
 	    ((mp->m_qflags & (XFS_PQUOTA_ACCT|XFS_OQUOTA_ENFD))) ==
 			      (XFS_PQUOTA_ACCT|XFS_OQUOTA_ENFD))
 		xfs_qm_statvfs(ip, statp);
@@ -1238,9 +1224,9 @@ xfs_fs_unfreeze(
 STATIC int
 xfs_fs_show_options(
 	struct seq_file		*m,
-	struct vfsmount		*mnt)
+	struct dentry		*root)
 {
-	return -xfs_showargs(XFS_M(mnt->mnt_sb), m);
+	return -xfs_showargs(XFS_M(root->d_sb), m);
 }
 
 /*
@@ -1333,10 +1319,14 @@ xfs_fs_fill_super(
 	if (error)
 		goto out_free_fsname;
 
-	error = xfs_icsb_init_counters(mp);
+	error = xfs_init_mount_workqueues(mp);
 	if (error)
 		goto out_close_devices;
 
+	error = xfs_icsb_init_counters(mp);
+	if (error)
+		goto out_destroy_workqueues;
+
 	error = xfs_readsb(mp, flags);
 	if (error)
 		goto out_destroy_counters;
@@ -1365,40 +1355,44 @@ xfs_fs_fill_super(
 	sb->s_blocksize = mp->m_sb.sb_blocksize;
 	sb->s_blocksize_bits = ffs(sb->s_blocksize) - 1;
 	sb->s_maxbytes = xfs_max_file_offset(sb->s_blocksize_bits);
+	sb->s_max_links = XFS_MAXLINK;
 	sb->s_time_gran = 1;
 	set_posix_acl_flag(sb);
 
-	error = xfs_mountfs(mp);
+	error = xfs_syncd_init(mp);
 	if (error)
 		goto out_filestream_unmount;
 
-	error = xfs_syncd_init(mp);
+	error = xfs_mountfs(mp);
 	if (error)
-		goto out_unmount;
+		goto out_syncd_stop;
 
 	root = igrab(VFS_I(mp->m_rootip));
 	if (!root) {
 		error = ENOENT;
-		goto out_syncd_stop;
+		goto out_unmount;
 	}
 	if (is_bad_inode(root)) {
 		error = EINVAL;
-		goto out_syncd_stop;
+		goto out_unmount;
 	}
-	sb->s_root = d_alloc_root(root);
+	sb->s_root = d_make_root(root);
 	if (!sb->s_root) {
 		error = ENOMEM;
-		goto out_iput;
+		goto out_unmount;
 	}
 
 	return 0;
-
+ out_syncd_stop:
+	xfs_syncd_stop(mp);
  out_filestream_unmount:
 	xfs_filestream_unmount(mp);
  out_free_sb:
 	xfs_freesb(mp);
  out_destroy_counters:
 	xfs_icsb_destroy_counters(mp);
+out_destroy_workqueues:
+	xfs_destroy_mount_workqueues(mp);
  out_close_devices:
 	xfs_close_devices(mp);
  out_free_fsname:
@@ -1407,21 +1401,10 @@ xfs_fs_fill_super(
  out:
 	return -error;
 
- out_iput:
-	iput(root);
- out_syncd_stop:
-	xfs_syncd_stop(mp);
  out_unmount:
-	/*
-	 * Blow away any referenced inode in the filestreams cache.
-	 * This can and will cause log traffic as inodes go inactive
-	 * here.
-	 */
 	xfs_filestream_unmount(mp);
-
-	xfs_flush_buftarg(mp->m_ddev_targp, 1);
-
 	xfs_unmountfs(mp);
+	xfs_syncd_stop(mp);
 	goto out_free_sb;
 }
 
@@ -1454,8 +1437,8 @@ static const struct super_operations xfs_super_operations = {
 	.alloc_inode		= xfs_fs_alloc_inode,
 	.destroy_inode		= xfs_fs_destroy_inode,
 	.dirty_inode		= xfs_fs_dirty_inode,
-	.write_inode		= xfs_fs_write_inode,
 	.evict_inode		= xfs_fs_evict_inode,
+	.drop_inode		= xfs_fs_drop_inode,
 	.put_super		= xfs_fs_put_super,
 	.sync_fs		= xfs_fs_sync_fs,
 	.freeze_fs		= xfs_fs_freeze,
@@ -1621,20 +1604,36 @@ STATIC int __init
 xfs_init_workqueues(void)
 {
 	/*
-	 * max_active is set to 8 to give enough concurency to allow
-	 * multiple work operations on each CPU to run. This allows multiple
-	 * filesystems to be running sync work concurrently, and scales with
-	 * the number of CPUs in the system.
+	 * We never want to the same work item to run twice, reclaiming inodes
+	 * or idling the log is not going to get any faster by multiple CPUs
+	 * competing for ressources.  Use the default large max_active value
+	 * so that even lots of filesystems can perform these task in parallel.
 	 */
-	xfs_syncd_wq = alloc_workqueue("xfssyncd", WQ_CPU_INTENSIVE, 8);
+	xfs_syncd_wq = alloc_workqueue("xfssyncd", WQ_NON_REENTRANT, 0);
 	if (!xfs_syncd_wq)
 		return -ENOMEM;
+
+	/*
+	 * The allocation workqueue can be used in memory reclaim situations
+	 * (writepage path), and parallelism is only limited by the number of
+	 * 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);
+	if (!xfs_alloc_wq)
+		goto out_destroy_syncd;
+
 	return 0;
+
+out_destroy_syncd:
+	destroy_workqueue(xfs_syncd_wq);
+	return -ENOMEM;
 }
 
 STATIC void
 xfs_destroy_workqueues(void)
 {
+	destroy_workqueue(xfs_alloc_wq);
 	destroy_workqueue(xfs_syncd_wq);
 }
 
@@ -1676,13 +1675,17 @@ init_xfs_fs(void)
 	if (error)
 		goto out_cleanup_procfs;
 
-	vfs_initquota();
+	error = xfs_qm_init();
+	if (error)
+		goto out_sysctl_unregister;
 
 	error = register_filesystem(&xfs_fs_type);
 	if (error)
-		goto out_sysctl_unregister;
+		goto out_qm_exit;
 	return 0;
 
+ out_qm_exit:
+	xfs_qm_exit();
  out_sysctl_unregister:
 	xfs_sysctl_unregister();
  out_cleanup_procfs:
@@ -1704,7 +1707,7 @@ init_xfs_fs(void)
 STATIC void __exit
 exit_xfs_fs(void)
 {
-	vfs_exitquota();
+	xfs_qm_exit();
 	unregister_filesystem(&xfs_fs_type);
 	xfs_sysctl_unregister();
 	xfs_cleanup_procfs();
diff --git a/fs/xfs/xfs_super.h b/fs/xfs/xfs_super.h
index 50a3266c999..09b0c26b224 100644
--- a/fs/xfs/xfs_super.h
+++ b/fs/xfs/xfs_super.h
@@ -21,13 +21,11 @@
 #include <linux/exportfs.h>
 
 #ifdef CONFIG_XFS_QUOTA
-extern void xfs_qm_init(void);
+extern int xfs_qm_init(void);
 extern void xfs_qm_exit(void);
-# define vfs_initquota()	xfs_qm_init()
-# define vfs_exitquota()	xfs_qm_exit()
 #else
-# define vfs_initquota()	do { } while (0)
-# define vfs_exitquota()	do { } while (0)
+# define xfs_qm_init()	(0)
+# define xfs_qm_exit()	do { } while (0)
 #endif
 
 #ifdef CONFIG_XFS_POSIX_ACL
diff --git a/fs/xfs/xfs_sync.c b/fs/xfs/xfs_sync.c
index f0994aedcd1..1e9ee064dbb 100644
--- a/fs/xfs/xfs_sync.c
+++ b/fs/xfs/xfs_sync.c
@@ -18,7 +18,6 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_types.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
 #include "xfs_inum.h"
 #include "xfs_trans.h"
@@ -241,45 +240,6 @@ xfs_sync_inode_data(
 	return error;
 }
 
-STATIC int
-xfs_sync_inode_attr(
-	struct xfs_inode	*ip,
-	struct xfs_perag	*pag,
-	int			flags)
-{
-	int			error = 0;
-
-	xfs_ilock(ip, XFS_ILOCK_SHARED);
-	if (xfs_inode_clean(ip))
-		goto out_unlock;
-	if (!xfs_iflock_nowait(ip)) {
-		if (!(flags & SYNC_WAIT))
-			goto out_unlock;
-		xfs_iflock(ip);
-	}
-
-	if (xfs_inode_clean(ip)) {
-		xfs_ifunlock(ip);
-		goto out_unlock;
-	}
-
-	error = xfs_iflush(ip, flags);
-
-	/*
-	 * We don't want to try again on non-blocking flushes that can't run
-	 * again immediately. If an inode really must be written, then that's
-	 * what the SYNC_WAIT flag is for.
-	 */
-	if (error == EAGAIN) {
-		ASSERT(!(flags & SYNC_WAIT));
-		error = 0;
-	}
-
- out_unlock:
-	xfs_iunlock(ip, XFS_ILOCK_SHARED);
-	return error;
-}
-
 /*
  * Write out pagecache data for the whole filesystem.
  */
@@ -300,19 +260,6 @@ xfs_sync_data(
 	return 0;
 }
 
-/*
- * Write out inode metadata (attributes) for the whole filesystem.
- */
-STATIC int
-xfs_sync_attr(
-	struct xfs_mount	*mp,
-	int			flags)
-{
-	ASSERT((flags & ~SYNC_WAIT) == 0);
-
-	return xfs_inode_ag_iterator(mp, xfs_sync_inode_attr, flags);
-}
-
 STATIC int
 xfs_sync_fsdata(
 	struct xfs_mount	*mp)
@@ -336,32 +283,6 @@ xfs_sync_fsdata(
 	return error;
 }
 
-int
-xfs_log_dirty_inode(
-	struct xfs_inode	*ip,
-	struct xfs_perag	*pag,
-	int			flags)
-{
-	struct xfs_mount	*mp = ip->i_mount;
-	struct xfs_trans	*tp;
-	int			error;
-
-	if (!ip->i_update_core)
-		return 0;
-
-	tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS);
-	error = xfs_trans_reserve(tp, 0, XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0);
-	if (error) {
-		xfs_trans_cancel(tp, 0);
-		return error;
-	}
-
-	xfs_ilock(ip, XFS_ILOCK_EXCL);
-	xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-	return xfs_trans_commit(tp, 0);
-}
-
 /*
  * When remounting a filesystem read-only or freezing the filesystem, we have
  * two phases to execute. This first phase is syncing the data before we
@@ -376,7 +297,7 @@ xfs_log_dirty_inode(
  * First stage of freeze - no writers will make progress now we are here,
  * so we flush delwri and delalloc buffers here, then wait for all I/O to
  * complete.  Data is frozen at that point. Metadata is not frozen,
- * transactions can still occur here so don't bother flushing the buftarg
+ * transactions can still occur here so don't bother emptying the AIL
  * because it'll just get dirty again.
  */
 int
@@ -385,66 +306,19 @@ xfs_quiesce_data(
 {
 	int			error, error2 = 0;
 
-	/*
-	 * Log all pending size and timestamp updates.  The vfs writeback
-	 * code is supposed to do this, but due to its overagressive
-	 * livelock detection it will skip inodes where appending writes
-	 * were written out in the first non-blocking sync phase if their
-	 * completion took long enough that it happened after taking the
-	 * timestamp for the cut-off in the blocking phase.
-	 */
-	xfs_inode_ag_iterator(mp, xfs_log_dirty_inode, 0);
-
-	xfs_qm_sync(mp, SYNC_TRYLOCK);
-	xfs_qm_sync(mp, SYNC_WAIT);
-
-	/* force out the newly dirtied log buffers */
+	/* force out the log */
 	xfs_log_force(mp, XFS_LOG_SYNC);
 
 	/* write superblock and hoover up shutdown errors */
 	error = xfs_sync_fsdata(mp);
 
-	/* make sure all delwri buffers are written out */
-	xfs_flush_buftarg(mp->m_ddev_targp, 1);
-
 	/* mark the log as covered if needed */
 	if (xfs_log_need_covered(mp))
 		error2 = xfs_fs_log_dummy(mp);
 
-	/* flush data-only devices */
-	if (mp->m_rtdev_targp)
-		xfs_flush_buftarg(mp->m_rtdev_targp, 1);
-
 	return error ? error : error2;
 }
 
-STATIC void
-xfs_quiesce_fs(
-	struct xfs_mount	*mp)
-{
-	int	count = 0, pincount;
-
-	xfs_reclaim_inodes(mp, 0);
-	xfs_flush_buftarg(mp->m_ddev_targp, 0);
-
-	/*
-	 * This loop must run at least twice.  The first instance of the loop
-	 * will flush most meta data but that will generate more meta data
-	 * (typically directory updates).  Which then must be flushed and
-	 * logged before we can write the unmount record. We also so sync
-	 * reclaim of inodes to catch any that the above delwri flush skipped.
-	 */
-	do {
-		xfs_reclaim_inodes(mp, SYNC_WAIT);
-		xfs_sync_attr(mp, SYNC_WAIT);
-		pincount = xfs_flush_buftarg(mp->m_ddev_targp, 1);
-		if (!pincount) {
-			delay(50);
-			count++;
-		}
-	} while (count < 2);
-}
-
 /*
  * Second stage of a quiesce. The data is already synced, now we have to take
  * care of the metadata. New transactions are already blocked, so we need to
@@ -460,8 +334,12 @@ xfs_quiesce_attr(
 	while (atomic_read(&mp->m_active_trans) > 0)
 		delay(100);
 
-	/* flush inodes and push all remaining buffers out to disk */
-	xfs_quiesce_fs(mp);
+	/* reclaim inodes to do any IO before the freeze completes */
+	xfs_reclaim_inodes(mp, 0);
+	xfs_reclaim_inodes(mp, SYNC_WAIT);
+
+	/* flush all pending changes from the AIL */
+	xfs_ail_push_all_sync(mp->m_ail);
 
 	/*
 	 * Just warn here till VFS can correctly support
@@ -475,7 +353,12 @@ xfs_quiesce_attr(
 		xfs_warn(mp, "xfs_attr_quiesce: failed to log sb changes. "
 				"Frozen image may not be consistent.");
 	xfs_log_unmount_write(mp);
-	xfs_unmountfs_writesb(mp);
+
+	/*
+	 * At this point we might have modified the superblock again and thus
+	 * added an item to the AIL, thus flush it again.
+	 */
+	xfs_ail_push_all_sync(mp->m_ail);
 }
 
 static void
@@ -499,16 +382,26 @@ xfs_sync_worker(
 					struct xfs_mount, m_sync_work);
 	int		error;
 
-	if (!(mp->m_flags & XFS_MOUNT_RDONLY)) {
+	/*
+	 * We shouldn't write/force the log if we are in the mount/unmount
+	 * process or on a read only filesystem. The workqueue still needs to be
+	 * active in both cases, however, because it is used for inode reclaim
+	 * during these times.  Use the MS_ACTIVE flag to avoid doing anything
+	 * during mount.  Doing work during unmount is avoided by calling
+	 * cancel_delayed_work_sync on this work queue before tearing down
+	 * the ail and the log in xfs_log_unmount.
+	 */
+	if (!(mp->m_super->s_flags & MS_ACTIVE) &&
+	    !(mp->m_flags & XFS_MOUNT_RDONLY)) {
 		/* dgc: errors ignored here */
 		if (mp->m_super->s_frozen == SB_UNFROZEN &&
 		    xfs_log_need_covered(mp))
 			error = xfs_fs_log_dummy(mp);
 		else
 			xfs_log_force(mp, 0);
-		error = xfs_qm_sync(mp, SYNC_TRYLOCK);
 
-		/* start pushing all the metadata that is currently dirty */
+		/* start pushing all the metadata that is currently
+		 * dirty */
 		xfs_ail_push_all(mp->m_ail);
 	}
 
@@ -528,14 +421,6 @@ xfs_syncd_queue_reclaim(
 	struct xfs_mount        *mp)
 {
 
-	/*
-	 * We can have inodes enter reclaim after we've shut down the syncd
-	 * workqueue during unmount, so don't allow reclaim work to be queued
-	 * during unmount.
-	 */
-	if (!(mp->m_super->s_flags & MS_ACTIVE))
-		return;
-
 	rcu_read_lock();
 	if (radix_tree_tagged(&mp->m_perag_tree, XFS_ICI_RECLAIM_TAG)) {
 		queue_delayed_work(xfs_syncd_wq, &mp->m_reclaim_work,
@@ -604,7 +489,6 @@ xfs_syncd_init(
 	INIT_DELAYED_WORK(&mp->m_reclaim_work, xfs_reclaim_worker);
 
 	xfs_syncd_queue_sync(mp);
-	xfs_syncd_queue_reclaim(mp);
 
 	return 0;
 }
@@ -711,14 +595,13 @@ xfs_reclaim_inode_grab(
 		return 1;
 
 	/*
-	 * do some unlocked checks first to avoid unnecessary lock traffic.
-	 * The first is a flush lock check, the second is a already in reclaim
-	 * check. Only do these checks if we are not going to block on locks.
+	 * If we are asked for non-blocking operation, do unlocked checks to
+	 * see if the inode already is being flushed or in reclaim to avoid
+	 * lock traffic.
 	 */
 	if ((flags & SYNC_TRYLOCK) &&
-	    (!ip->i_flush.done || __xfs_iflags_test(ip, XFS_IRECLAIM))) {
+	    __xfs_iflags_test(ip, XFS_IFLOCK | XFS_IRECLAIM))
 		return 1;
-	}
 
 	/*
 	 * The radix tree lock here protects a thread in xfs_iget from racing
@@ -743,11 +626,8 @@ xfs_reclaim_inode_grab(
 }
 
 /*
- * Inodes in different states need to be treated differently, and the return
- * value of xfs_iflush is not sufficient to get this right. The following table
- * lists the inode states and the reclaim actions necessary for non-blocking
- * reclaim:
- *
+ * Inodes in different states need to be treated differently. The following
+ * table lists the inode states and the reclaim actions necessary:
  *
  *	inode state	     iflush ret		required action
  *      ---------------      ----------         ---------------
@@ -757,39 +637,31 @@ xfs_reclaim_inode_grab(
  *	stale, unpinned		0		reclaim
  *	clean, pinned(*)	0		requeue
  *	stale, pinned		EAGAIN		requeue
- *	dirty, delwri ok	0		requeue
- *	dirty, delwri blocked	EAGAIN		requeue
- *	dirty, sync flush	0		reclaim
+ *	dirty, async		-		requeue
+ *	dirty, sync		0		reclaim
  *
  * (*) dgc: I don't think the clean, pinned state is possible but it gets
  * handled anyway given the order of checks implemented.
  *
- * As can be seen from the table, the return value of xfs_iflush() is not
- * sufficient to correctly decide the reclaim action here. The checks in
- * xfs_iflush() might look like duplicates, but they are not.
- *
  * Also, because we get the flush lock first, we know that any inode that has
  * been flushed delwri has had the flush completed by the time we check that
- * the inode is clean. The clean inode check needs to be done before flushing
- * the inode delwri otherwise we would loop forever requeuing clean inodes as
- * we cannot tell apart a successful delwri flush and a clean inode from the
- * return value of xfs_iflush().
+ * the inode is clean.
  *
- * Note that because the inode is flushed delayed write by background
- * writeback, the flush lock may already be held here and waiting on it can
- * result in very long latencies. Hence for sync reclaims, where we wait on the
- * flush lock, the caller should push out delayed write inodes first before
- * trying to reclaim them to minimise the amount of time spent waiting. For
- * background relaim, we just requeue the inode for the next pass.
+ * Note that because the inode is flushed delayed write by AIL pushing, the
+ * flush lock may already be held here and waiting on it can result in very
+ * long latencies.  Hence for sync reclaims, where we wait on the flush lock,
+ * the caller should push the AIL first before trying to reclaim inodes to
+ * minimise the amount of time spent waiting.  For background relaim, we only
+ * bother to reclaim clean inodes anyway.
  *
  * Hence the order of actions after gaining the locks should be:
  *	bad		=> reclaim
  *	shutdown	=> unpin and reclaim
- *	pinned, delwri	=> requeue
+ *	pinned, async	=> requeue
  *	pinned, sync	=> unpin
  *	stale		=> reclaim
  *	clean		=> reclaim
- *	dirty, delwri	=> flush and requeue
+ *	dirty, async	=> requeue
  *	dirty, sync	=> flush, wait and reclaim
  */
 STATIC int
@@ -798,7 +670,8 @@ xfs_reclaim_inode(
 	struct xfs_perag	*pag,
 	int			sync_mode)
 {
-	int	error;
+	struct xfs_buf		*bp = NULL;
+	int			error;
 
 restart:
 	error = 0;
@@ -806,17 +679,6 @@ restart:
 	if (!xfs_iflock_nowait(ip)) {
 		if (!(sync_mode & SYNC_WAIT))
 			goto out;
-
-		/*
-		 * If we only have a single dirty inode in a cluster there is
-		 * a fair chance that the AIL push may have pushed it into
-		 * the buffer, but xfsbufd won't touch it until 30 seconds
-		 * from now, and thus we will lock up here.
-		 *
-		 * Promote the inode buffer to the front of the delwri list
-		 * and wake up xfsbufd now.
-		 */
-		xfs_promote_inode(ip);
 		xfs_iflock(ip);
 	}
 
@@ -824,13 +686,12 @@ restart:
 		goto reclaim;
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
 		xfs_iunpin_wait(ip);
+		xfs_iflush_abort(ip, false);
 		goto reclaim;
 	}
 	if (xfs_ipincount(ip)) {
-		if (!(sync_mode & SYNC_WAIT)) {
-			xfs_ifunlock(ip);
-			goto out;
-		}
+		if (!(sync_mode & SYNC_WAIT))
+			goto out_ifunlock;
 		xfs_iunpin_wait(ip);
 	}
 	if (xfs_iflags_test(ip, XFS_ISTALE))
@@ -839,60 +700,42 @@ restart:
 		goto reclaim;
 
 	/*
+	 * Never flush out dirty data during non-blocking reclaim, as it would
+	 * just contend with AIL pushing trying to do the same job.
+	 */
+	if (!(sync_mode & SYNC_WAIT))
+		goto out_ifunlock;
+
+	/*
 	 * Now we have an inode that needs flushing.
 	 *
-	 * We do a nonblocking flush here even if we are doing a SYNC_WAIT
-	 * reclaim as we can deadlock with inode cluster removal.
+	 * Note that xfs_iflush will never block on the inode buffer lock, as
 	 * xfs_ifree_cluster() can lock the inode buffer before it locks the
-	 * ip->i_lock, and we are doing the exact opposite here. As a result,
-	 * doing a blocking xfs_itobp() to get the cluster buffer will result
+	 * ip->i_lock, and we are doing the exact opposite here.  As a result,
+	 * doing a blocking xfs_itobp() to get the cluster buffer would result
 	 * in an ABBA deadlock with xfs_ifree_cluster().
 	 *
 	 * As xfs_ifree_cluser() must gather all inodes that are active in the
 	 * cache to mark them stale, if we hit this case we don't actually want
 	 * to do IO here - we want the inode marked stale so we can simply
-	 * reclaim it. Hence if we get an EAGAIN error on a SYNC_WAIT flush,
-	 * just unlock the inode, back off and try again. Hopefully the next
-	 * pass through will see the stale flag set on the inode.
+	 * reclaim it.  Hence if we get an EAGAIN error here,  just unlock the
+	 * inode, back off and try again.  Hopefully the next pass through will
+	 * see the stale flag set on the inode.
 	 */
-	error = xfs_iflush(ip, SYNC_TRYLOCK | sync_mode);
-	if (sync_mode & SYNC_WAIT) {
-		if (error == EAGAIN) {
-			xfs_iunlock(ip, XFS_ILOCK_EXCL);
-			/* backoff longer than in xfs_ifree_cluster */
-			delay(2);
-			goto restart;
-		}
-		xfs_iflock(ip);
-		goto reclaim;
+	error = xfs_iflush(ip, &bp);
+	if (error == EAGAIN) {
+		xfs_iunlock(ip, XFS_ILOCK_EXCL);
+		/* backoff longer than in xfs_ifree_cluster */
+		delay(2);
+		goto restart;
 	}
 
-	/*
-	 * When we have to flush an inode but don't have SYNC_WAIT set, we
-	 * flush the inode out using a delwri buffer and wait for the next
-	 * call into reclaim to find it in a clean state instead of waiting for
-	 * it now. We also don't return errors here - if the error is transient
-	 * then the next reclaim pass will flush the inode, and if the error
-	 * is permanent then the next sync reclaim will reclaim the inode and
-	 * pass on the error.
-	 */
-	if (error && error != EAGAIN && !XFS_FORCED_SHUTDOWN(ip->i_mount)) {
-		xfs_warn(ip->i_mount,
-			"inode 0x%llx background reclaim flush failed with %d",
-			(long long)ip->i_ino, error);
+	if (!error) {
+		error = xfs_bwrite(bp);
+		xfs_buf_relse(bp);
 	}
-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
-	 * a short while. However, this just burns CPU time scanning the tree
-	 * waiting for IO to complete and xfssyncd never goes back to the idle
-	 * state. Instead, return 0 to let the next scheduled background reclaim
-	 * attempt to reclaim the inode again.
-	 */
-	return 0;
 
+	xfs_iflock(ip);
 reclaim:
 	xfs_ifunlock(ip);
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
@@ -918,17 +761,28 @@ reclaim:
 	 * can reference the inodes in the cache without taking references.
 	 *
 	 * We make that OK here by ensuring that we wait until the inode is
-	 * unlocked after the lookup before we go ahead and free it.  We get
-	 * both the ilock and the iolock because the code may need to drop the
-	 * ilock one but will still hold the iolock.
+	 * unlocked after the lookup before we go ahead and free it.
 	 */
-	xfs_ilock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
 	xfs_qm_dqdetach(ip);
-	xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL);
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 
 	xfs_inode_free(ip);
 	return error;
 
+out_ifunlock:
+	xfs_ifunlock(ip);
+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
+	 * a short while. However, this just burns CPU time scanning the tree
+	 * waiting for IO to complete and xfssyncd never goes back to the idle
+	 * state. Instead, return 0 to let the next scheduled background reclaim
+	 * attempt to reclaim the inode again.
+	 */
+	return 0;
 }
 
 /*
diff --git a/fs/xfs/xfs_sync.h b/fs/xfs/xfs_sync.h
index fa965479d78..941202e7ac6 100644
--- a/fs/xfs/xfs_sync.h
+++ b/fs/xfs/xfs_sync.h
@@ -34,8 +34,6 @@ void xfs_quiesce_attr(struct xfs_mount *mp);
 
 void xfs_flush_inodes(struct xfs_inode *ip);
 
-int xfs_log_dirty_inode(struct xfs_inode *ip, struct xfs_perag *pag, int flags);
-
 int xfs_reclaim_inodes(struct xfs_mount *mp, int mode);
 int xfs_reclaim_inodes_count(struct xfs_mount *mp);
 void xfs_reclaim_inodes_nr(struct xfs_mount *mp, int nr_to_scan);
diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c
index 9010ce885e6..624bedd8135 100644
--- a/fs/xfs/xfs_trace.c
+++ b/fs/xfs/xfs_trace.c
@@ -18,9 +18,7 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_types.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h
index 49403579887..caf5dabfd55 100644
--- a/fs/xfs/xfs_trace.h
+++ b/fs/xfs/xfs_trace.h
@@ -32,7 +32,7 @@ struct xfs_da_node_entry;
 struct xfs_dquot;
 struct xfs_log_item;
 struct xlog_ticket;
-struct log;
+struct xlog;
 struct xlog_recover;
 struct xlog_recover_item;
 struct xfs_buf_log_format;
@@ -281,7 +281,7 @@ DECLARE_EVENT_CLASS(xfs_buf_class,
 	TP_STRUCT__entry(
 		__field(dev_t, dev)
 		__field(xfs_daddr_t, bno)
-		__field(size_t, buffer_length)
+		__field(int, nblks)
 		__field(int, hold)
 		__field(int, pincount)
 		__field(unsigned, lockval)
@@ -291,18 +291,18 @@ DECLARE_EVENT_CLASS(xfs_buf_class,
 	TP_fast_assign(
 		__entry->dev = bp->b_target->bt_dev;
 		__entry->bno = bp->b_bn;
-		__entry->buffer_length = bp->b_buffer_length;
+		__entry->nblks = bp->b_length;
 		__entry->hold = atomic_read(&bp->b_hold);
 		__entry->pincount = atomic_read(&bp->b_pin_count);
 		__entry->lockval = bp->b_sema.count;
 		__entry->flags = bp->b_flags;
 		__entry->caller_ip = caller_ip;
 	),
-	TP_printk("dev %d:%d bno 0x%llx len 0x%zx hold %d pincount %d "
+	TP_printk("dev %d:%d bno 0x%llx nblks 0x%x hold %d pincount %d "
 		  "lock %d flags %s caller %pf",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  (unsigned long long)__entry->bno,
-		  __entry->buffer_length,
+		  __entry->nblks,
 		  __entry->hold,
 		  __entry->pincount,
 		  __entry->lockval,
@@ -328,7 +328,7 @@ DEFINE_BUF_EVENT(xfs_buf_unlock);
 DEFINE_BUF_EVENT(xfs_buf_iowait);
 DEFINE_BUF_EVENT(xfs_buf_iowait_done);
 DEFINE_BUF_EVENT(xfs_buf_delwri_queue);
-DEFINE_BUF_EVENT(xfs_buf_delwri_dequeue);
+DEFINE_BUF_EVENT(xfs_buf_delwri_queued);
 DEFINE_BUF_EVENT(xfs_buf_delwri_split);
 DEFINE_BUF_EVENT(xfs_buf_get_uncached);
 DEFINE_BUF_EVENT(xfs_bdstrat_shut);
@@ -362,7 +362,7 @@ DECLARE_EVENT_CLASS(xfs_buf_flags_class,
 	TP_fast_assign(
 		__entry->dev = bp->b_target->bt_dev;
 		__entry->bno = bp->b_bn;
-		__entry->buffer_length = bp->b_buffer_length;
+		__entry->buffer_length = BBTOB(bp->b_length);
 		__entry->flags = flags;
 		__entry->hold = atomic_read(&bp->b_hold);
 		__entry->pincount = atomic_read(&bp->b_pin_count);
@@ -406,7 +406,7 @@ TRACE_EVENT(xfs_buf_ioerror,
 	TP_fast_assign(
 		__entry->dev = bp->b_target->bt_dev;
 		__entry->bno = bp->b_bn;
-		__entry->buffer_length = bp->b_buffer_length;
+		__entry->buffer_length = BBTOB(bp->b_length);
 		__entry->hold = atomic_read(&bp->b_hold);
 		__entry->pincount = atomic_read(&bp->b_pin_count);
 		__entry->lockval = bp->b_sema.count;
@@ -450,7 +450,7 @@ DECLARE_EVENT_CLASS(xfs_buf_item_class,
 		__entry->bli_recur = bip->bli_recur;
 		__entry->bli_refcount = atomic_read(&bip->bli_refcount);
 		__entry->buf_bno = bip->bli_buf->b_bn;
-		__entry->buf_len = bip->bli_buf->b_buffer_length;
+		__entry->buf_len = BBTOB(bip->bli_buf->b_length);
 		__entry->buf_flags = bip->bli_buf->b_flags;
 		__entry->buf_hold = atomic_read(&bip->bli_buf->b_hold);
 		__entry->buf_pincount = atomic_read(&bip->bli_buf->b_pin_count);
@@ -486,12 +486,10 @@ DEFINE_BUF_ITEM_EVENT(xfs_buf_item_format_stale);
 DEFINE_BUF_ITEM_EVENT(xfs_buf_item_pin);
 DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unpin);
 DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unpin_stale);
-DEFINE_BUF_ITEM_EVENT(xfs_buf_item_trylock);
 DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unlock);
 DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unlock_stale);
 DEFINE_BUF_ITEM_EVENT(xfs_buf_item_committed);
 DEFINE_BUF_ITEM_EVENT(xfs_buf_item_push);
-DEFINE_BUF_ITEM_EVENT(xfs_buf_item_pushbuf);
 DEFINE_BUF_ITEM_EVENT(xfs_trans_get_buf);
 DEFINE_BUF_ITEM_EVENT(xfs_trans_get_buf_recur);
 DEFINE_BUF_ITEM_EVENT(xfs_trans_getsb);
@@ -580,7 +578,7 @@ DEFINE_INODE_EVENT(xfs_ioctl_setattr);
 DEFINE_INODE_EVENT(xfs_dir_fsync);
 DEFINE_INODE_EVENT(xfs_file_fsync);
 DEFINE_INODE_EVENT(xfs_destroy_inode);
-DEFINE_INODE_EVENT(xfs_write_inode);
+DEFINE_INODE_EVENT(xfs_dirty_inode);
 DEFINE_INODE_EVENT(xfs_evict_inode);
 
 DEFINE_INODE_EVENT(xfs_dquot_dqalloc);
@@ -627,16 +625,19 @@ DECLARE_EVENT_CLASS(xfs_namespace_class,
 	TP_STRUCT__entry(
 		__field(dev_t, dev)
 		__field(xfs_ino_t, dp_ino)
+		__field(int, namelen)
 		__dynamic_array(char, name, name->len)
 	),
 	TP_fast_assign(
 		__entry->dev = VFS_I(dp)->i_sb->s_dev;
 		__entry->dp_ino = dp->i_ino;
+		__entry->namelen = name->len;
 		memcpy(__get_str(name), name->name, name->len);
 	),
-	TP_printk("dev %d:%d dp ino 0x%llx name %s",
+	TP_printk("dev %d:%d dp ino 0x%llx name %.*s",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->dp_ino,
+		  __entry->namelen,
 		  __get_str(name))
 )
 
@@ -658,6 +659,8 @@ TRACE_EVENT(xfs_rename,
 		__field(dev_t, dev)
 		__field(xfs_ino_t, src_dp_ino)
 		__field(xfs_ino_t, target_dp_ino)
+		__field(int, src_namelen)
+		__field(int, target_namelen)
 		__dynamic_array(char, src_name, src_name->len)
 		__dynamic_array(char, target_name, target_name->len)
 	),
@@ -665,15 +668,20 @@ TRACE_EVENT(xfs_rename,
 		__entry->dev = VFS_I(src_dp)->i_sb->s_dev;
 		__entry->src_dp_ino = src_dp->i_ino;
 		__entry->target_dp_ino = target_dp->i_ino;
+		__entry->src_namelen = src_name->len;
+		__entry->target_namelen = target_name->len;
 		memcpy(__get_str(src_name), src_name->name, src_name->len);
-		memcpy(__get_str(target_name), target_name->name, target_name->len);
+		memcpy(__get_str(target_name), target_name->name,
+			target_name->len);
 	),
 	TP_printk("dev %d:%d src dp ino 0x%llx target dp ino 0x%llx"
-		  " src name %s target name %s",
+		  " src name %.*s target name %.*s",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->src_dp_ino,
 		  __entry->target_dp_ino,
+		  __entry->src_namelen,
 		  __get_str(src_name),
+		  __entry->target_namelen,
 		  __get_str(target_name))
 )
 
@@ -733,21 +741,18 @@ DEFINE_EVENT(xfs_dquot_class, name, \
 DEFINE_DQUOT_EVENT(xfs_dqadjust);
 DEFINE_DQUOT_EVENT(xfs_dqreclaim_want);
 DEFINE_DQUOT_EVENT(xfs_dqreclaim_dirty);
-DEFINE_DQUOT_EVENT(xfs_dqreclaim_unlink);
+DEFINE_DQUOT_EVENT(xfs_dqreclaim_busy);
+DEFINE_DQUOT_EVENT(xfs_dqreclaim_done);
 DEFINE_DQUOT_EVENT(xfs_dqattach_found);
 DEFINE_DQUOT_EVENT(xfs_dqattach_get);
-DEFINE_DQUOT_EVENT(xfs_dqinit);
-DEFINE_DQUOT_EVENT(xfs_dqreuse);
 DEFINE_DQUOT_EVENT(xfs_dqalloc);
 DEFINE_DQUOT_EVENT(xfs_dqtobp_read);
 DEFINE_DQUOT_EVENT(xfs_dqread);
 DEFINE_DQUOT_EVENT(xfs_dqread_fail);
-DEFINE_DQUOT_EVENT(xfs_dqlookup_found);
-DEFINE_DQUOT_EVENT(xfs_dqlookup_want);
-DEFINE_DQUOT_EVENT(xfs_dqlookup_freelist);
-DEFINE_DQUOT_EVENT(xfs_dqlookup_done);
 DEFINE_DQUOT_EVENT(xfs_dqget_hit);
 DEFINE_DQUOT_EVENT(xfs_dqget_miss);
+DEFINE_DQUOT_EVENT(xfs_dqget_freeing);
+DEFINE_DQUOT_EVENT(xfs_dqget_dup);
 DEFINE_DQUOT_EVENT(xfs_dqput);
 DEFINE_DQUOT_EVENT(xfs_dqput_wait);
 DEFINE_DQUOT_EVENT(xfs_dqput_free);
@@ -757,7 +762,7 @@ DEFINE_DQUOT_EVENT(xfs_dqflush_force);
 DEFINE_DQUOT_EVENT(xfs_dqflush_done);
 
 DECLARE_EVENT_CLASS(xfs_loggrant_class,
-	TP_PROTO(struct log *log, struct xlog_ticket *tic),
+	TP_PROTO(struct xlog *log, struct xlog_ticket *tic),
 	TP_ARGS(log, tic),
 	TP_STRUCT__entry(
 		__field(dev_t, dev)
@@ -785,12 +790,12 @@ DECLARE_EVENT_CLASS(xfs_loggrant_class,
 		__entry->curr_res = tic->t_curr_res;
 		__entry->unit_res = tic->t_unit_res;
 		__entry->flags = tic->t_flags;
-		__entry->reserveq = list_empty(&log->l_reserveq);
-		__entry->writeq = list_empty(&log->l_writeq);
-		xlog_crack_grant_head(&log->l_grant_reserve_head,
+		__entry->reserveq = list_empty(&log->l_reserve_head.waiters);
+		__entry->writeq = list_empty(&log->l_write_head.waiters);
+		xlog_crack_grant_head(&log->l_reserve_head.grant,
 				&__entry->grant_reserve_cycle,
 				&__entry->grant_reserve_bytes);
-		xlog_crack_grant_head(&log->l_grant_write_head,
+		xlog_crack_grant_head(&log->l_write_head.grant,
 				&__entry->grant_write_cycle,
 				&__entry->grant_write_bytes);
 		__entry->curr_cycle = log->l_curr_cycle;
@@ -825,24 +830,18 @@ DECLARE_EVENT_CLASS(xfs_loggrant_class,
 
 #define DEFINE_LOGGRANT_EVENT(name) \
 DEFINE_EVENT(xfs_loggrant_class, name, \
-	TP_PROTO(struct log *log, struct xlog_ticket *tic), \
+	TP_PROTO(struct xlog *log, struct xlog_ticket *tic), \
 	TP_ARGS(log, tic))
 DEFINE_LOGGRANT_EVENT(xfs_log_done_nonperm);
 DEFINE_LOGGRANT_EVENT(xfs_log_done_perm);
-DEFINE_LOGGRANT_EVENT(xfs_log_reserve);
 DEFINE_LOGGRANT_EVENT(xfs_log_umount_write);
-DEFINE_LOGGRANT_EVENT(xfs_log_grant_enter);
-DEFINE_LOGGRANT_EVENT(xfs_log_grant_exit);
-DEFINE_LOGGRANT_EVENT(xfs_log_grant_error);
 DEFINE_LOGGRANT_EVENT(xfs_log_grant_sleep);
 DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake);
 DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake_up);
-DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_enter);
-DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_exit);
-DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_error);
-DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_sleep);
-DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake);
-DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake_up);
+DEFINE_LOGGRANT_EVENT(xfs_log_reserve);
+DEFINE_LOGGRANT_EVENT(xfs_log_reserve_exit);
+DEFINE_LOGGRANT_EVENT(xfs_log_regrant);
+DEFINE_LOGGRANT_EVENT(xfs_log_regrant_exit);
 DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_enter);
 DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_exit);
 DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_sub);
@@ -875,15 +874,30 @@ DECLARE_EVENT_CLASS(xfs_log_item_class,
 		  __print_flags(__entry->flags, "|", XFS_LI_FLAGS))
 )
 
+TRACE_EVENT(xfs_log_force,
+	TP_PROTO(struct xfs_mount *mp, xfs_lsn_t lsn),
+	TP_ARGS(mp, lsn),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_lsn_t, lsn)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->lsn = lsn;
+	),
+	TP_printk("dev %d:%d lsn 0x%llx",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->lsn)
+)
+
 #define DEFINE_LOG_ITEM_EVENT(name) \
 DEFINE_EVENT(xfs_log_item_class, name, \
 	TP_PROTO(struct xfs_log_item *lip), \
 	TP_ARGS(lip))
 DEFINE_LOG_ITEM_EVENT(xfs_ail_push);
-DEFINE_LOG_ITEM_EVENT(xfs_ail_pushbuf);
-DEFINE_LOG_ITEM_EVENT(xfs_ail_pushbuf_pinned);
 DEFINE_LOG_ITEM_EVENT(xfs_ail_pinned);
 DEFINE_LOG_ITEM_EVENT(xfs_ail_locked);
+DEFINE_LOG_ITEM_EVENT(xfs_ail_flushing);
 
 
 DECLARE_EVENT_CLASS(xfs_file_class,
@@ -893,7 +907,6 @@ DECLARE_EVENT_CLASS(xfs_file_class,
 		__field(dev_t, dev)
 		__field(xfs_ino_t, ino)
 		__field(xfs_fsize_t, size)
-		__field(xfs_fsize_t, new_size)
 		__field(loff_t, offset)
 		__field(size_t, count)
 		__field(int, flags)
@@ -902,17 +915,15 @@ DECLARE_EVENT_CLASS(xfs_file_class,
 		__entry->dev = VFS_I(ip)->i_sb->s_dev;
 		__entry->ino = ip->i_ino;
 		__entry->size = ip->i_d.di_size;
-		__entry->new_size = ip->i_new_size;
 		__entry->offset = offset;
 		__entry->count = count;
 		__entry->flags = flags;
 	),
-	TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx "
+	TP_printk("dev %d:%d ino 0x%llx size 0x%llx "
 		  "offset 0x%llx count 0x%zx ioflags %s",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->ino,
 		  __entry->size,
-		  __entry->new_size,
 		  __entry->offset,
 		  __entry->count,
 		  __print_flags(__entry->flags, "|", XFS_IO_FLAGS))
@@ -980,7 +991,6 @@ DECLARE_EVENT_CLASS(xfs_imap_class,
 		__field(dev_t, dev)
 		__field(xfs_ino_t, ino)
 		__field(loff_t, size)
-		__field(loff_t, new_size)
 		__field(loff_t, offset)
 		__field(size_t, count)
 		__field(int, type)
@@ -992,7 +1002,6 @@ DECLARE_EVENT_CLASS(xfs_imap_class,
 		__entry->dev = VFS_I(ip)->i_sb->s_dev;
 		__entry->ino = ip->i_ino;
 		__entry->size = ip->i_d.di_size;
-		__entry->new_size = ip->i_new_size;
 		__entry->offset = offset;
 		__entry->count = count;
 		__entry->type = type;
@@ -1000,13 +1009,11 @@ DECLARE_EVENT_CLASS(xfs_imap_class,
 		__entry->startblock = irec ? irec->br_startblock : 0;
 		__entry->blockcount = irec ? irec->br_blockcount : 0;
 	),
-	TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx "
-		  "offset 0x%llx count %zd type %s "
-		  "startoff 0x%llx startblock %lld blockcount 0x%llx",
+	TP_printk("dev %d:%d ino 0x%llx size 0x%llx offset 0x%llx count %zd "
+		  "type %s startoff 0x%llx startblock %lld blockcount 0x%llx",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->ino,
 		  __entry->size,
-		  __entry->new_size,
 		  __entry->offset,
 		  __entry->count,
 		  __print_symbolic(__entry->type, XFS_IO_TYPES),
@@ -1033,26 +1040,23 @@ DECLARE_EVENT_CLASS(xfs_simple_io_class,
 		__field(xfs_ino_t, ino)
 		__field(loff_t, isize)
 		__field(loff_t, disize)
-		__field(loff_t, new_size)
 		__field(loff_t, offset)
 		__field(size_t, count)
 	),
 	TP_fast_assign(
 		__entry->dev = VFS_I(ip)->i_sb->s_dev;
 		__entry->ino = ip->i_ino;
-		__entry->isize = ip->i_size;
+		__entry->isize = VFS_I(ip)->i_size;
 		__entry->disize = ip->i_d.di_size;
-		__entry->new_size = ip->i_new_size;
 		__entry->offset = offset;
 		__entry->count = count;
 	),
-	TP_printk("dev %d:%d ino 0x%llx isize 0x%llx disize 0x%llx new_size 0x%llx "
+	TP_printk("dev %d:%d ino 0x%llx isize 0x%llx disize 0x%llx "
 		  "offset 0x%llx count %zd",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->ino,
 		  __entry->isize,
 		  __entry->disize,
-		  __entry->new_size,
 		  __entry->offset,
 		  __entry->count)
 );
@@ -1092,8 +1096,8 @@ DECLARE_EVENT_CLASS(xfs_itrunc_class,
 DEFINE_EVENT(xfs_itrunc_class, name, \
 	TP_PROTO(struct xfs_inode *ip, xfs_fsize_t new_size), \
 	TP_ARGS(ip, new_size))
-DEFINE_ITRUNC_EVENT(xfs_itruncate_data_start);
-DEFINE_ITRUNC_EVENT(xfs_itruncate_data_end);
+DEFINE_ITRUNC_EVENT(xfs_itruncate_extents_start);
+DEFINE_ITRUNC_EVENT(xfs_itruncate_extents_end);
 
 TRACE_EVENT(xfs_pagecache_inval,
 	TP_PROTO(struct xfs_inode *ip, xfs_off_t start, xfs_off_t finish),
@@ -1154,7 +1158,7 @@ TRACE_EVENT(xfs_bunmap,
 
 );
 
-DECLARE_EVENT_CLASS(xfs_busy_class,
+DECLARE_EVENT_CLASS(xfs_extent_busy_class,
 	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno,
 		 xfs_agblock_t agbno, xfs_extlen_t len),
 	TP_ARGS(mp, agno, agbno, len),
@@ -1177,17 +1181,17 @@ DECLARE_EVENT_CLASS(xfs_busy_class,
 		  __entry->len)
 );
 #define DEFINE_BUSY_EVENT(name) \
-DEFINE_EVENT(xfs_busy_class, name, \
+DEFINE_EVENT(xfs_extent_busy_class, name, \
 	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, \
 		 xfs_agblock_t agbno, xfs_extlen_t len), \
 	TP_ARGS(mp, agno, agbno, len))
-DEFINE_BUSY_EVENT(xfs_alloc_busy);
-DEFINE_BUSY_EVENT(xfs_alloc_busy_enomem);
-DEFINE_BUSY_EVENT(xfs_alloc_busy_force);
-DEFINE_BUSY_EVENT(xfs_alloc_busy_reuse);
-DEFINE_BUSY_EVENT(xfs_alloc_busy_clear);
+DEFINE_BUSY_EVENT(xfs_extent_busy);
+DEFINE_BUSY_EVENT(xfs_extent_busy_enomem);
+DEFINE_BUSY_EVENT(xfs_extent_busy_force);
+DEFINE_BUSY_EVENT(xfs_extent_busy_reuse);
+DEFINE_BUSY_EVENT(xfs_extent_busy_clear);
 
-TRACE_EVENT(xfs_alloc_busy_trim,
+TRACE_EVENT(xfs_extent_busy_trim,
 	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno,
 		 xfs_agblock_t agbno, xfs_extlen_t len,
 		 xfs_agblock_t tbno, xfs_extlen_t tlen),
@@ -1427,7 +1431,7 @@ DEFINE_ALLOC_EVENT(xfs_alloc_vextent_noagbp);
 DEFINE_ALLOC_EVENT(xfs_alloc_vextent_loopfailed);
 DEFINE_ALLOC_EVENT(xfs_alloc_vextent_allfailed);
 
-DECLARE_EVENT_CLASS(xfs_dir2_class,
+DECLARE_EVENT_CLASS(xfs_da_class,
 	TP_PROTO(struct xfs_da_args *args),
 	TP_ARGS(args),
 	TP_STRUCT__entry(
@@ -1462,7 +1466,7 @@ DECLARE_EVENT_CLASS(xfs_dir2_class,
 )
 
 #define DEFINE_DIR2_EVENT(name) \
-DEFINE_EVENT(xfs_dir2_class, name, \
+DEFINE_EVENT(xfs_da_class, name, \
 	TP_PROTO(struct xfs_da_args *args), \
 	TP_ARGS(args))
 DEFINE_DIR2_EVENT(xfs_dir2_sf_addname);
@@ -1491,6 +1495,64 @@ DEFINE_DIR2_EVENT(xfs_dir2_node_replace);
 DEFINE_DIR2_EVENT(xfs_dir2_node_removename);
 DEFINE_DIR2_EVENT(xfs_dir2_node_to_leaf);
 
+#define DEFINE_ATTR_EVENT(name) \
+DEFINE_EVENT(xfs_da_class, name, \
+	TP_PROTO(struct xfs_da_args *args), \
+	TP_ARGS(args))
+DEFINE_ATTR_EVENT(xfs_attr_sf_add);
+DEFINE_ATTR_EVENT(xfs_attr_sf_addname);
+DEFINE_ATTR_EVENT(xfs_attr_sf_create);
+DEFINE_ATTR_EVENT(xfs_attr_sf_lookup);
+DEFINE_ATTR_EVENT(xfs_attr_sf_remove);
+DEFINE_ATTR_EVENT(xfs_attr_sf_removename);
+DEFINE_ATTR_EVENT(xfs_attr_sf_to_leaf);
+
+DEFINE_ATTR_EVENT(xfs_attr_leaf_add);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_add_old);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_add_new);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_addname);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_create);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_lookup);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_replace);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_removename);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_split);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_split_before);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_split_after);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_clearflag);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_setflag);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_flipflags);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_to_sf);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_to_node);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_rebalance);
+DEFINE_ATTR_EVENT(xfs_attr_leaf_unbalance);
+
+DEFINE_ATTR_EVENT(xfs_attr_node_addname);
+DEFINE_ATTR_EVENT(xfs_attr_node_lookup);
+DEFINE_ATTR_EVENT(xfs_attr_node_replace);
+DEFINE_ATTR_EVENT(xfs_attr_node_removename);
+
+#define DEFINE_DA_EVENT(name) \
+DEFINE_EVENT(xfs_da_class, name, \
+	TP_PROTO(struct xfs_da_args *args), \
+	TP_ARGS(args))
+DEFINE_DA_EVENT(xfs_da_split);
+DEFINE_DA_EVENT(xfs_da_join);
+DEFINE_DA_EVENT(xfs_da_link_before);
+DEFINE_DA_EVENT(xfs_da_link_after);
+DEFINE_DA_EVENT(xfs_da_unlink_back);
+DEFINE_DA_EVENT(xfs_da_unlink_forward);
+DEFINE_DA_EVENT(xfs_da_root_split);
+DEFINE_DA_EVENT(xfs_da_root_join);
+DEFINE_DA_EVENT(xfs_da_node_add);
+DEFINE_DA_EVENT(xfs_da_node_create);
+DEFINE_DA_EVENT(xfs_da_node_split);
+DEFINE_DA_EVENT(xfs_da_node_remove);
+DEFINE_DA_EVENT(xfs_da_node_rebalance);
+DEFINE_DA_EVENT(xfs_da_node_unbalance);
+DEFINE_DA_EVENT(xfs_da_swap_lastblock);
+DEFINE_DA_EVENT(xfs_da_grow_inode);
+DEFINE_DA_EVENT(xfs_da_shrink_inode);
+
 DECLARE_EVENT_CLASS(xfs_dir2_space_class,
 	TP_PROTO(struct xfs_da_args *args, int idx),
 	TP_ARGS(args, idx),
@@ -1570,7 +1632,6 @@ DECLARE_EVENT_CLASS(xfs_swap_extent_class,
 		__field(xfs_ino_t, ino)
 		__field(int, format)
 		__field(int, nex)
-		__field(int, max_nex)
 		__field(int, broot_size)
 		__field(int, fork_off)
 	),
@@ -1580,18 +1641,16 @@ DECLARE_EVENT_CLASS(xfs_swap_extent_class,
 		__entry->ino = ip->i_ino;
 		__entry->format = ip->i_d.di_format;
 		__entry->nex = ip->i_d.di_nextents;
-		__entry->max_nex = ip->i_df.if_ext_max;
 		__entry->broot_size = ip->i_df.if_broot_bytes;
 		__entry->fork_off = XFS_IFORK_BOFF(ip);
 	),
 	TP_printk("dev %d:%d ino 0x%llx (%s), %s format, num_extents %d, "
-		  "Max in-fork extents %d, broot size %d, fork offset %d",
+		  "broot size %d, fork offset %d",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->ino,
 		  __print_symbolic(__entry->which, XFS_SWAPEXT_INODES),
 		  __print_symbolic(__entry->format, XFS_INODE_FORMAT_STR),
 		  __entry->nex,
-		  __entry->max_nex,
 		  __entry->broot_size,
 		  __entry->fork_off)
 )
@@ -1605,7 +1664,7 @@ DEFINE_SWAPEXT_EVENT(xfs_swap_extent_before);
 DEFINE_SWAPEXT_EVENT(xfs_swap_extent_after);
 
 DECLARE_EVENT_CLASS(xfs_log_recover_item_class,
-	TP_PROTO(struct log *log, struct xlog_recover *trans,
+	TP_PROTO(struct xlog *log, struct xlog_recover *trans,
 		struct xlog_recover_item *item, int pass),
 	TP_ARGS(log, trans, item, pass),
 	TP_STRUCT__entry(
@@ -1639,7 +1698,7 @@ DECLARE_EVENT_CLASS(xfs_log_recover_item_class,
 
 #define DEFINE_LOG_RECOVER_ITEM(name) \
 DEFINE_EVENT(xfs_log_recover_item_class, name, \
-	TP_PROTO(struct log *log, struct xlog_recover *trans, \
+	TP_PROTO(struct xlog *log, struct xlog_recover *trans, \
 		struct xlog_recover_item *item, int pass), \
 	TP_ARGS(log, trans, item, pass))
 
@@ -1650,7 +1709,7 @@ DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_reorder_tail);
 DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_recover);
 
 DECLARE_EVENT_CLASS(xfs_log_recover_buf_item_class,
-	TP_PROTO(struct log *log, struct xfs_buf_log_format *buf_f),
+	TP_PROTO(struct xlog *log, struct xfs_buf_log_format *buf_f),
 	TP_ARGS(log, buf_f),
 	TP_STRUCT__entry(
 		__field(dev_t, dev)
@@ -1680,7 +1739,7 @@ DECLARE_EVENT_CLASS(xfs_log_recover_buf_item_class,
 
 #define DEFINE_LOG_RECOVER_BUF_ITEM(name) \
 DEFINE_EVENT(xfs_log_recover_buf_item_class, name, \
-	TP_PROTO(struct log *log, struct xfs_buf_log_format *buf_f), \
+	TP_PROTO(struct xlog *log, struct xfs_buf_log_format *buf_f), \
 	TP_ARGS(log, buf_f))
 
 DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_not_cancel);
@@ -1693,7 +1752,7 @@ DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_reg_buf);
 DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_dquot_buf);
 
 DECLARE_EVENT_CLASS(xfs_log_recover_ino_item_class,
-	TP_PROTO(struct log *log, struct xfs_inode_log_format *in_f),
+	TP_PROTO(struct xlog *log, struct xfs_inode_log_format *in_f),
 	TP_ARGS(log, in_f),
 	TP_STRUCT__entry(
 		__field(dev_t, dev)
@@ -1731,7 +1790,7 @@ DECLARE_EVENT_CLASS(xfs_log_recover_ino_item_class,
 )
 #define DEFINE_LOG_RECOVER_INO_ITEM(name) \
 DEFINE_EVENT(xfs_log_recover_ino_item_class, name, \
-	TP_PROTO(struct log *log, struct xfs_inode_log_format *in_f), \
+	TP_PROTO(struct xlog *log, struct xfs_inode_log_format *in_f), \
 	TP_ARGS(log, in_f))
 
 DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_recover);
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index 1f35b2feca9..fdf324508c5 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -19,9 +19,7 @@
 #include "xfs.h"
 #include "xfs_fs.h"
 #include "xfs_types.h"
-#include "xfs_bit.h"
 #include "xfs_log.h"
-#include "xfs_inum.h"
 #include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
@@ -36,6 +34,7 @@
 #include "xfs_btree.h"
 #include "xfs_ialloc.h"
 #include "xfs_alloc.h"
+#include "xfs_extent_busy.h"
 #include "xfs_bmap.h"
 #include "xfs_quota.h"
 #include "xfs_trans_priv.h"
@@ -585,7 +584,7 @@ xfs_trans_t *
 _xfs_trans_alloc(
 	xfs_mount_t	*mp,
 	uint		type,
-	uint		memflags)
+	xfs_km_flags_t	memflags)
 {
 	xfs_trans_t	*tp;
 
@@ -608,8 +607,8 @@ STATIC void
 xfs_trans_free(
 	struct xfs_trans	*tp)
 {
-	xfs_alloc_busy_sort(&tp->t_busy);
-	xfs_alloc_busy_clear(tp->t_mountp, &tp->t_busy, false);
+	xfs_extent_busy_sort(&tp->t_busy);
+	xfs_extent_busy_clear(tp->t_mountp, &tp->t_busy, false);
 
 	atomic_dec(&tp->t_mountp->m_active_trans);
 	xfs_trans_free_dqinfo(tp);
@@ -681,7 +680,6 @@ xfs_trans_reserve(
 	uint		flags,
 	uint		logcount)
 {
-	int		log_flags;
 	int		error = 0;
 	int		rsvd = (tp->t_flags & XFS_TRANS_RESERVE) != 0;
 
@@ -707,24 +705,32 @@ xfs_trans_reserve(
 	 * Reserve the log space needed for this transaction.
 	 */
 	if (logspace > 0) {
-		ASSERT((tp->t_log_res == 0) || (tp->t_log_res == logspace));
-		ASSERT((tp->t_log_count == 0) ||
-			(tp->t_log_count == logcount));
+		bool	permanent = false;
+
+		ASSERT(tp->t_log_res == 0 || tp->t_log_res == logspace);
+		ASSERT(tp->t_log_count == 0 || tp->t_log_count == logcount);
+
 		if (flags & XFS_TRANS_PERM_LOG_RES) {
-			log_flags = XFS_LOG_PERM_RESERV;
 			tp->t_flags |= XFS_TRANS_PERM_LOG_RES;
+			permanent = true;
 		} else {
 			ASSERT(tp->t_ticket == NULL);
 			ASSERT(!(tp->t_flags & XFS_TRANS_PERM_LOG_RES));
-			log_flags = 0;
 		}
 
-		error = xfs_log_reserve(tp->t_mountp, logspace, logcount,
-					&tp->t_ticket,
-					XFS_TRANSACTION, log_flags, tp->t_type);
-		if (error) {
-			goto undo_blocks;
+		if (tp->t_ticket != NULL) {
+			ASSERT(flags & XFS_TRANS_PERM_LOG_RES);
+			error = xfs_log_regrant(tp->t_mountp, tp->t_ticket);
+		} else {
+			error = xfs_log_reserve(tp->t_mountp, logspace,
+						logcount, &tp->t_ticket,
+						XFS_TRANSACTION, permanent,
+						tp->t_type);
 		}
+
+		if (error)
+			goto undo_blocks;
+
 		tp->t_log_res = logspace;
 		tp->t_log_count = logcount;
 	}
@@ -752,6 +758,8 @@ xfs_trans_reserve(
 	 */
 undo_log:
 	if (logspace > 0) {
+		int		log_flags;
+
 		if (flags & XFS_TRANS_PERM_LOG_RES) {
 			log_flags = XFS_LOG_REL_PERM_RESERV;
 		} else {
@@ -1151,14 +1159,13 @@ xfs_trans_add_item(
 {
 	struct xfs_log_item_desc *lidp;
 
-	ASSERT(lip->li_mountp = tp->t_mountp);
-	ASSERT(lip->li_ailp = tp->t_mountp->m_ail);
+	ASSERT(lip->li_mountp == tp->t_mountp);
+	ASSERT(lip->li_ailp == tp->t_mountp->m_ail);
 
 	lidp = kmem_zone_zalloc(xfs_log_item_desc_zone, KM_SLEEP | KM_NOFS);
 
 	lidp->lid_item = lip;
 	lidp->lid_flags = 0;
-	lidp->lid_size = 0;
 	list_add_tail(&lidp->lid_trans, &tp->t_items);
 
 	lip->li_desc = lidp;
@@ -1210,219 +1217,6 @@ xfs_trans_free_items(
 	}
 }
 
-/*
- * Unlock the items associated with a transaction.
- *
- * Items which were not logged should be freed.  Those which were logged must
- * still be tracked so they can be unpinned when the transaction commits.
- */
-STATIC void
-xfs_trans_unlock_items(
-	struct xfs_trans	*tp,
-	xfs_lsn_t		commit_lsn)
-{
-	struct xfs_log_item_desc *lidp, *next;
-
-	list_for_each_entry_safe(lidp, next, &tp->t_items, lid_trans) {
-		struct xfs_log_item	*lip = lidp->lid_item;
-
-		lip->li_desc = NULL;
-
-		if (commit_lsn != NULLCOMMITLSN)
-			IOP_COMMITTING(lip, commit_lsn);
-		IOP_UNLOCK(lip);
-
-		/*
-		 * Free the descriptor if the item is not dirty
-		 * within this transaction.
-		 */
-		if (!(lidp->lid_flags & XFS_LID_DIRTY))
-			xfs_trans_free_item_desc(lidp);
-	}
-}
-
-/*
- * Total up the number of log iovecs needed to commit this
- * transaction.  The transaction itself needs one for the
- * transaction header.  Ask each dirty item in turn how many
- * it needs to get the total.
- */
-static uint
-xfs_trans_count_vecs(
-	struct xfs_trans	*tp)
-{
-	int			nvecs;
-	struct xfs_log_item_desc *lidp;
-
-	nvecs = 1;
-
-	/* In the non-debug case we need to start bailing out if we
-	 * didn't find a log_item here, return zero and let trans_commit
-	 * deal with it.
-	 */
-	if (list_empty(&tp->t_items)) {
-		ASSERT(0);
-		return 0;
-	}
-
-	list_for_each_entry(lidp, &tp->t_items, lid_trans) {
-		/*
-		 * Skip items which aren't dirty in this transaction.
-		 */
-		if (!(lidp->lid_flags & XFS_LID_DIRTY))
-			continue;
-		lidp->lid_size = IOP_SIZE(lidp->lid_item);
-		nvecs += lidp->lid_size;
-	}
-
-	return nvecs;
-}
-
-/*
- * Fill in the vector with pointers to data to be logged
- * by this transaction.  The transaction header takes
- * the first vector, and then each dirty item takes the
- * number of vectors it indicated it needed in xfs_trans_count_vecs().
- *
- * As each item fills in the entries it needs, also pin the item
- * so that it cannot be flushed out until the log write completes.
- */
-static void
-xfs_trans_fill_vecs(
-	struct xfs_trans	*tp,
-	struct xfs_log_iovec	*log_vector)
-{
-	struct xfs_log_item_desc *lidp;
-	struct xfs_log_iovec	*vecp;
-	uint			nitems;
-
-	/*
-	 * Skip over the entry for the transaction header, we'll
-	 * fill that in at the end.
-	 */
-	vecp = log_vector + 1;
-
-	nitems = 0;
-	ASSERT(!list_empty(&tp->t_items));
-	list_for_each_entry(lidp, &tp->t_items, lid_trans) {
-		/* Skip items which aren't dirty in this transaction. */
-		if (!(lidp->lid_flags & XFS_LID_DIRTY))
-			continue;
-
-		/*
-		 * The item may be marked dirty but not log anything.  This can
-		 * be used to get called when a transaction is committed.
-		 */
-		if (lidp->lid_size)
-			nitems++;
-		IOP_FORMAT(lidp->lid_item, vecp);
-		vecp += lidp->lid_size;
-		IOP_PIN(lidp->lid_item);
-	}
-
-	/*
-	 * Now that we've counted the number of items in this transaction, fill
-	 * in the transaction header. Note that the transaction header does not
-	 * have a log item.
-	 */
-	tp->t_header.th_magic = XFS_TRANS_HEADER_MAGIC;
-	tp->t_header.th_type = tp->t_type;
-	tp->t_header.th_num_items = nitems;
-	log_vector->i_addr = (xfs_caddr_t)&tp->t_header;
-	log_vector->i_len = sizeof(xfs_trans_header_t);
-	log_vector->i_type = XLOG_REG_TYPE_TRANSHDR;
-}
-
-/*
- * The committed item processing consists of calling the committed routine of
- * each logged item, updating the item's position in the AIL if necessary, and
- * unpinning each item.  If the committed routine returns -1, then do nothing
- * further with the item because it may have been freed.
- *
- * Since items are unlocked when they are copied to the incore log, it is
- * possible for two transactions to be completing and manipulating the same
- * item simultaneously.  The AIL lock will protect the lsn field of each item.
- * The value of this field can never go backwards.
- *
- * We unpin the items after repositioning them in the AIL, because otherwise
- * they could be immediately flushed and we'd have to race with the flusher
- * trying to pull the item from the AIL as we add it.
- */
-static void
-xfs_trans_item_committed(
-	struct xfs_log_item	*lip,
-	xfs_lsn_t		commit_lsn,
-	int			aborted)
-{
-	xfs_lsn_t		item_lsn;
-	struct xfs_ail		*ailp;
-
-	if (aborted)
-		lip->li_flags |= XFS_LI_ABORTED;
-	item_lsn = IOP_COMMITTED(lip, commit_lsn);
-
-	/* item_lsn of -1 means the item needs no further processing */
-	if (XFS_LSN_CMP(item_lsn, (xfs_lsn_t)-1) == 0)
-		return;
-
-	/*
-	 * If the returned lsn is greater than what it contained before, update
-	 * the location of the item in the AIL.  If it is not, then do nothing.
-	 * Items can never move backwards in the AIL.
-	 *
-	 * While the new lsn should usually be greater, it is possible that a
-	 * later transaction completing simultaneously with an earlier one
-	 * using the same item could complete first with a higher lsn.  This
-	 * would cause the earlier transaction to fail the test below.
-	 */
-	ailp = lip->li_ailp;
-	spin_lock(&ailp->xa_lock);
-	if (XFS_LSN_CMP(item_lsn, lip->li_lsn) > 0) {
-		/*
-		 * This will set the item's lsn to item_lsn and update the
-		 * position of the item in the AIL.
-		 *
-		 * xfs_trans_ail_update() drops the AIL lock.
-