Merge branch 'master' into for-2.6.35

Conflicts:
	fs/ext3/fsync.c

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

43 files changed, 2036 insertions, 2134 deletions

diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c
index 0f8b9968a80..089eaca860b 100644
--- a/fs/xfs/linux-2.6/xfs_aops.c
+++ b/fs/xfs/linux-2.6/xfs_aops.c
@@ -45,6 +45,15 @@
 #include <linux/pagevec.h>
 #include <linux/writeback.h>
 
+/*
+ * Types of I/O for bmap clustering and I/O completion tracking.
+ */
+enum {
+	IO_READ,	/* mapping for a read */
+	IO_DELAY,	/* mapping covers delalloc region */
+	IO_UNWRITTEN,	/* mapping covers allocated but uninitialized data */
+	IO_NEW		/* just allocated */
+};
 
 /*
  * Prime number of hash buckets since address is used as the key.
@@ -103,8 +112,9 @@ xfs_count_page_state(
 
 STATIC struct block_device *
 xfs_find_bdev_for_inode(
-	struct xfs_inode	*ip)
+	struct inode		*inode)
 {
+	struct xfs_inode	*ip = XFS_I(inode);
 	struct xfs_mount	*mp = ip->i_mount;
 
 	if (XFS_IS_REALTIME_INODE(ip))
@@ -183,7 +193,7 @@ xfs_setfilesize(
 	xfs_fsize_t		isize;
 
 	ASSERT((ip->i_d.di_mode & S_IFMT) == S_IFREG);
-	ASSERT(ioend->io_type != IOMAP_READ);
+	ASSERT(ioend->io_type != IO_READ);
 
 	if (unlikely(ioend->io_error))
 		return 0;
@@ -214,7 +224,7 @@ xfs_finish_ioend(
 	if (atomic_dec_and_test(&ioend->io_remaining)) {
 		struct workqueue_struct *wq;
 
-		wq = (ioend->io_type == IOMAP_UNWRITTEN) ?
+		wq = (ioend->io_type == IO_UNWRITTEN) ?
 			xfsconvertd_workqueue : xfsdatad_workqueue;
 		queue_work(wq, &ioend->io_work);
 		if (wait)
@@ -237,7 +247,7 @@ xfs_end_io(
 	 * For unwritten extents we need to issue transactions to convert a
 	 * range to normal written extens after the data I/O has finished.
 	 */
-	if (ioend->io_type == IOMAP_UNWRITTEN &&
+	if (ioend->io_type == IO_UNWRITTEN &&
 	    likely(!ioend->io_error && !XFS_FORCED_SHUTDOWN(ip->i_mount))) {
 
 		error = xfs_iomap_write_unwritten(ip, ioend->io_offset,
@@ -250,7 +260,7 @@ xfs_end_io(
 	 * We might have to update the on-disk file size after extending
 	 * writes.
 	 */
-	if (ioend->io_type != IOMAP_READ) {
+	if (ioend->io_type != IO_READ) {
 		error = xfs_setfilesize(ioend);
 		ASSERT(!error || error == EAGAIN);
 	}
@@ -309,21 +319,25 @@ xfs_map_blocks(
 	struct inode		*inode,
 	loff_t			offset,
 	ssize_t			count,
-	xfs_iomap_t		*mapp,
+	struct xfs_bmbt_irec	*imap,
 	int			flags)
 {
 	int			nmaps = 1;
+	int			new = 0;
 
-	return -xfs_iomap(XFS_I(inode), offset, count, flags, mapp, &nmaps);
+	return -xfs_iomap(XFS_I(inode), offset, count, flags, imap, &nmaps, &new);
 }
 
 STATIC int
-xfs_iomap_valid(
-	xfs_iomap_t		*iomapp,
-	loff_t			offset)
+xfs_imap_valid(
+	struct inode		*inode,
+	struct xfs_bmbt_irec	*imap,
+	xfs_off_t		offset)
 {
-	return offset >= iomapp->iomap_offset &&
-		offset < iomapp->iomap_offset + iomapp->iomap_bsize;
+	offset >>= inode->i_blkbits;
+
+	return offset >= imap->br_startoff &&
+		offset < imap->br_startoff + imap->br_blockcount;
 }
 
 /*
@@ -554,19 +568,23 @@ xfs_add_to_ioend(
 
 STATIC void
 xfs_map_buffer(
+	struct inode		*inode,
 	struct buffer_head	*bh,
-	xfs_iomap_t		*mp,
-	xfs_off_t		offset,
-	uint			block_bits)
+	struct xfs_bmbt_irec	*imap,
+	xfs_off_t		offset)
 {
 	sector_t		bn;
+	struct xfs_mount	*m = XFS_I(inode)->i_mount;
+	xfs_off_t		iomap_offset = XFS_FSB_TO_B(m, imap->br_startoff);
+	xfs_daddr_t		iomap_bn = xfs_fsb_to_db(XFS_I(inode), imap->br_startblock);
 
-	ASSERT(mp->iomap_bn != IOMAP_DADDR_NULL);
+	ASSERT(imap->br_startblock != HOLESTARTBLOCK);
+	ASSERT(imap->br_startblock != DELAYSTARTBLOCK);
 
-	bn = (mp->iomap_bn >> (block_bits - BBSHIFT)) +
-	      ((offset - mp->iomap_offset) >> block_bits);
+	bn = (iomap_bn >> (inode->i_blkbits - BBSHIFT)) +
+	      ((offset - iomap_offset) >> inode->i_blkbits);
 
-	ASSERT(bn || (mp->iomap_flags & IOMAP_REALTIME));
+	ASSERT(bn || XFS_IS_REALTIME_INODE(XFS_I(inode)));
 
 	bh->b_blocknr = bn;
 	set_buffer_mapped(bh);
@@ -574,17 +592,17 @@ xfs_map_buffer(
 
 STATIC void
 xfs_map_at_offset(
+	struct inode		*inode,
 	struct buffer_head	*bh,
-	loff_t			offset,
-	int			block_bits,
-	xfs_iomap_t		*iomapp)
+	struct xfs_bmbt_irec	*imap,
+	xfs_off_t		offset)
 {
-	ASSERT(!(iomapp->iomap_flags & IOMAP_HOLE));
-	ASSERT(!(iomapp->iomap_flags & IOMAP_DELAY));
+	ASSERT(imap->br_startblock != HOLESTARTBLOCK);
+	ASSERT(imap->br_startblock != DELAYSTARTBLOCK);
 
 	lock_buffer(bh);
-	xfs_map_buffer(bh, iomapp, offset, block_bits);
-	bh->b_bdev = iomapp->iomap_target->bt_bdev;
+	xfs_map_buffer(inode, bh, imap, offset);
+	bh->b_bdev = xfs_find_bdev_for_inode(inode);
 	set_buffer_mapped(bh);
 	clear_buffer_delay(bh);
 	clear_buffer_unwritten(bh);
@@ -713,11 +731,11 @@ xfs_is_delayed_page(
 		bh = head = page_buffers(page);
 		do {
 			if (buffer_unwritten(bh))
-				acceptable = (type == IOMAP_UNWRITTEN);
+				acceptable = (type == IO_UNWRITTEN);
 			else if (buffer_delay(bh))
-				acceptable = (type == IOMAP_DELAY);
+				acceptable = (type == IO_DELAY);
 			else if (buffer_dirty(bh) && buffer_mapped(bh))
-				acceptable = (type == IOMAP_NEW);
+				acceptable = (type == IO_NEW);
 			else
 				break;
 		} while ((bh = bh->b_this_page) != head);
@@ -740,7 +758,7 @@ xfs_convert_page(
 	struct inode		*inode,
 	struct page		*page,
 	loff_t			tindex,
-	xfs_iomap_t		*mp,
+	struct xfs_bmbt_irec	*imap,
 	xfs_ioend_t		**ioendp,
 	struct writeback_control *wbc,
 	int			startio,
@@ -750,7 +768,6 @@ xfs_convert_page(
 	xfs_off_t		end_offset;
 	unsigned long		p_offset;
 	unsigned int		type;
-	int			bbits = inode->i_blkbits;
 	int			len, page_dirty;
 	int			count = 0, done = 0, uptodate = 1;
  	xfs_off_t		offset = page_offset(page);
@@ -802,19 +819,19 @@ xfs_convert_page(
 
 		if (buffer_unwritten(bh) || buffer_delay(bh)) {
 			if (buffer_unwritten(bh))
-				type = IOMAP_UNWRITTEN;
+				type = IO_UNWRITTEN;
 			else
-				type = IOMAP_DELAY;
+				type = IO_DELAY;
 
-			if (!xfs_iomap_valid(mp, offset)) {
+			if (!xfs_imap_valid(inode, imap, offset)) {
 				done = 1;
 				continue;
 			}
 
-			ASSERT(!(mp->iomap_flags & IOMAP_HOLE));
-			ASSERT(!(mp->iomap_flags & IOMAP_DELAY));
+			ASSERT(imap->br_startblock != HOLESTARTBLOCK);
+			ASSERT(imap->br_startblock != DELAYSTARTBLOCK);
 
-			xfs_map_at_offset(bh, offset, bbits, mp);
+			xfs_map_at_offset(inode, bh, imap, offset);
 			if (startio) {
 				xfs_add_to_ioend(inode, bh, offset,
 						type, ioendp, done);
@@ -826,7 +843,7 @@ xfs_convert_page(
 			page_dirty--;
 			count++;
 		} else {
-			type = IOMAP_NEW;
+			type = IO_NEW;
 			if (buffer_mapped(bh) && all_bh && startio) {
 				lock_buffer(bh);
 				xfs_add_to_ioend(inode, bh, offset,
@@ -866,7 +883,7 @@ STATIC void
 xfs_cluster_write(
 	struct inode		*inode,
 	pgoff_t			tindex,
-	xfs_iomap_t		*iomapp,
+	struct xfs_bmbt_irec	*imap,
 	xfs_ioend_t		**ioendp,
 	struct writeback_control *wbc,
 	int			startio,
@@ -885,7 +902,7 @@ xfs_cluster_write(
 
 		for (i = 0; i < pagevec_count(&pvec); i++) {
 			done = xfs_convert_page(inode, pvec.pages[i], tindex++,
-					iomapp, ioendp, wbc, startio, all_bh);
+					imap, ioendp, wbc, startio, all_bh);
 			if (done)
 				break;
 		}
@@ -930,7 +947,7 @@ xfs_aops_discard_page(
 	loff_t			offset = page_offset(page);
 	ssize_t			len = 1 << inode->i_blkbits;
 
-	if (!xfs_is_delayed_page(page, IOMAP_DELAY))
+	if (!xfs_is_delayed_page(page, IO_DELAY))
 		goto out_invalidate;
 
 	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
@@ -1042,15 +1059,15 @@ xfs_page_state_convert(
 	int		unmapped) /* also implies page uptodate */
 {
 	struct buffer_head	*bh, *head;
-	xfs_iomap_t		iomap;
+	struct xfs_bmbt_irec	imap;
 	xfs_ioend_t		*ioend = NULL, *iohead = NULL;
 	loff_t			offset;
 	unsigned long           p_offset = 0;
 	unsigned int		type;
 	__uint64_t              end_offset;
-	pgoff_t                 end_index, last_index, tlast;
+	pgoff_t                 end_index, last_index;
 	ssize_t			size, len;
-	int			flags, err, iomap_valid = 0, uptodate = 1;
+	int			flags, err, imap_valid = 0, uptodate = 1;
 	int			page_dirty, count = 0;
 	int			trylock = 0;
 	int			all_bh = unmapped;
@@ -1097,7 +1114,7 @@ xfs_page_state_convert(
 	bh = head = page_buffers(page);
 	offset = page_offset(page);
 	flags = BMAPI_READ;
-	type = IOMAP_NEW;
+	type = IO_NEW;
 
 	/* TODO: cleanup count and page_dirty */
 
@@ -1111,12 +1128,12 @@ xfs_page_state_convert(
 			 * the iomap is actually still valid, but the ioend
 			 * isn't.  shouldn't happen too often.
 			 */
-			iomap_valid = 0;
+			imap_valid = 0;
 			continue;
 		}
 
-		if (iomap_valid)
-			iomap_valid = xfs_iomap_valid(&iomap, offset);
+		if (imap_valid)
+			imap_valid = xfs_imap_valid(inode, &imap, offset);
 
 		/*
 		 * First case, map an unwritten extent and prepare for
@@ -1137,20 +1154,20 @@ xfs_page_state_convert(
 			 * Make sure we don't use a read-only iomap
 			 */
 			if (flags == BMAPI_READ)
-				iomap_valid = 0;
+				imap_valid = 0;
 
 			if (buffer_unwritten(bh)) {
-				type = IOMAP_UNWRITTEN;
+				type = IO_UNWRITTEN;
 				flags = BMAPI_WRITE | BMAPI_IGNSTATE;
 			} else if (buffer_delay(bh)) {
-				type = IOMAP_DELAY;
+				type = IO_DELAY;
 				flags = BMAPI_ALLOCATE | trylock;
 			} else {
-				type = IOMAP_NEW;
+				type = IO_NEW;
 				flags = BMAPI_WRITE | BMAPI_MMAP;
 			}
 
-			if (!iomap_valid) {
+			if (!imap_valid) {
 				/*
 				 * if we didn't have a valid mapping then we
 				 * need to ensure that we put the new mapping
@@ -1160,7 +1177,7 @@ xfs_page_state_convert(
 				 * for unwritten extent conversion.
 				 */
 				new_ioend = 1;
-				if (type == IOMAP_NEW) {
+				if (type == IO_NEW) {
 					size = xfs_probe_cluster(inode,
 							page, bh, head, 0);
 				} else {
@@ -1168,14 +1185,14 @@ xfs_page_state_convert(
 				}
 
 				err = xfs_map_blocks(inode, offset, size,
-						&iomap, flags);
+						&imap, flags);
 				if (err)
 					goto error;
-				iomap_valid = xfs_iomap_valid(&iomap, offset);
+				imap_valid = xfs_imap_valid(inode, &imap,
+							    offset);
 			}
-			if (iomap_valid) {
-				xfs_map_at_offset(bh, offset,
-						inode->i_blkbits, &iomap);
+			if (imap_valid) {
+				xfs_map_at_offset(inode, bh, &imap, offset);
 				if (startio) {
 					xfs_add_to_ioend(inode, bh, offset,
 							type, &ioend,
@@ -1194,40 +1211,41 @@ xfs_page_state_convert(
 			 * That means it must already have extents allocated
 			 * underneath it. Map the extent by reading it.
 			 */
-			if (!iomap_valid || flags != BMAPI_READ) {
+			if (!imap_valid || flags != BMAPI_READ) {
 				flags = BMAPI_READ;
 				size = xfs_probe_cluster(inode, page, bh,
 								head, 1);
 				err = xfs_map_blocks(inode, offset, size,
-						&iomap, flags);
+						&imap, flags);
 				if (err)
 					goto error;
-				iomap_valid = xfs_iomap_valid(&iomap, offset);
+				imap_valid = xfs_imap_valid(inode, &imap,
+							    offset);
 			}
 
 			/*
-			 * We set the type to IOMAP_NEW in case we are doing a
+			 * We set the type to IO_NEW in case we are doing a
 			 * small write at EOF that is extending the file but
 			 * without needing an allocation. We need to update the
 			 * file size on I/O completion in this case so it is
 			 * the same case as having just allocated a new extent
 			 * that we are writing into for the first time.
 			 */
-			type = IOMAP_NEW;
+			type = IO_NEW;
 			if (trylock_buffer(bh)) {
 				ASSERT(buffer_mapped(bh));
-				if (iomap_valid)
+				if (imap_valid)
 					all_bh = 1;
 				xfs_add_to_ioend(inode, bh, offset, type,
-						&ioend, !iomap_valid);
+						&ioend, !imap_valid);
 				page_dirty--;
 				count++;
 			} else {
-				iomap_valid = 0;
+				imap_valid = 0;
 			}
 		} else if ((buffer_uptodate(bh) || PageUptodate(page)) &&
 			   (unmapped || startio)) {
-			iomap_valid = 0;
+			imap_valid = 0;
 		}
 
 		if (!iohead)
@@ -1241,12 +1259,23 @@ xfs_page_state_convert(
 	if (startio)
 		xfs_start_page_writeback(page, 1, count);
 
-	if (ioend && iomap_valid) {
-		offset = (iomap.iomap_offset + iomap.iomap_bsize - 1) >>
-					PAGE_CACHE_SHIFT;
-		tlast = min_t(pgoff_t, offset, last_index);
-		xfs_cluster_write(inode, page->index + 1, &iomap, &ioend,
-					wbc, startio, all_bh, tlast);
+	if (ioend && imap_valid) {
+		xfs_off_t		end_index;
+
+		end_index = imap.br_startoff + imap.br_blockcount;
+
+		/* to bytes */
+		end_index <<= inode->i_blkbits;
+
+		/* to pages */
+		end_index = (end_index - 1) >> PAGE_CACHE_SHIFT;
+
+		/* check against file size */
+		if (end_index > last_index)
+			end_index = last_index;
+
+		xfs_cluster_write(inode, page->index + 1, &imap, &ioend,
+					wbc, startio, all_bh, end_index);
 	}
 
 	if (iohead)
@@ -1448,10 +1477,11 @@ __xfs_get_blocks(
 	int			direct,
 	bmapi_flags_t		flags)
 {
-	xfs_iomap_t		iomap;
+	struct xfs_bmbt_irec	imap;
 	xfs_off_t		offset;
 	ssize_t			size;
-	int			niomap = 1;
+	int			nimap = 1;
+	int			new = 0;
 	int			error;
 
 	offset = (xfs_off_t)iblock << inode->i_blkbits;
@@ -1462,22 +1492,21 @@ __xfs_get_blocks(
 		return 0;
 
 	error = xfs_iomap(XFS_I(inode), offset, size,
-			     create ? flags : BMAPI_READ, &iomap, &niomap);
+			     create ? flags : BMAPI_READ, &imap, &nimap, &new);
 	if (error)
 		return -error;
-	if (niomap == 0)
+	if (nimap == 0)
 		return 0;
 
-	if (iomap.iomap_bn != IOMAP_DADDR_NULL) {
+	if (imap.br_startblock != HOLESTARTBLOCK &&
+	    imap.br_startblock != DELAYSTARTBLOCK) {
 		/*
 		 * For unwritten extents do not report a disk address on
 		 * the read case (treat as if we're reading into a hole).
 		 */
-		if (create || !(iomap.iomap_flags & IOMAP_UNWRITTEN)) {
-			xfs_map_buffer(bh_result, &iomap, offset,
-				       inode->i_blkbits);
-		}
-		if (create && (iomap.iomap_flags & IOMAP_UNWRITTEN)) {
+		if (create || !ISUNWRITTEN(&imap))
+			xfs_map_buffer(inode, bh_result, &imap, offset);
+		if (create && ISUNWRITTEN(&imap)) {
 			if (direct)
 				bh_result->b_private = inode;
 			set_buffer_unwritten(bh_result);
@@ -1488,7 +1517,7 @@ __xfs_get_blocks(
 	 * If this is a realtime file, data may be on a different device.
 	 * to that pointed to from the buffer_head b_bdev currently.
 	 */
-	bh_result->b_bdev = iomap.iomap_target->bt_bdev;
+	bh_result->b_bdev = xfs_find_bdev_for_inode(inode);
 
 	/*
 	 * If we previously allocated a block out beyond eof and we are now
@@ -1502,10 +1531,10 @@ __xfs_get_blocks(
 	if (create &&
 	    ((!buffer_mapped(bh_result) && !buffer_uptodate(bh_result)) ||
 	     (offset >= i_size_read(inode)) ||
-	     (iomap.iomap_flags & (IOMAP_NEW|IOMAP_UNWRITTEN))))
+	     (new || ISUNWRITTEN(&imap))))
 		set_buffer_new(bh_result);
 
-	if (iomap.iomap_flags & IOMAP_DELAY) {
+	if (imap.br_startblock == DELAYSTARTBLOCK) {
 		BUG_ON(direct);
 		if (create) {
 			set_buffer_uptodate(bh_result);
@@ -1514,11 +1543,23 @@ __xfs_get_blocks(
 		}
 	}
 
+	/*
+	 * If this is O_DIRECT or the mpage code calling tell them how large
+	 * the mapping is, so that we can avoid repeated get_blocks calls.
+	 */
 	if (direct || size > (1 << inode->i_blkbits)) {
-		ASSERT(iomap.iomap_bsize - iomap.iomap_delta > 0);
-		offset = min_t(xfs_off_t,
-				iomap.iomap_bsize - iomap.iomap_delta, size);
-		bh_result->b_size = (ssize_t)min_t(xfs_off_t, LONG_MAX, offset);
+		xfs_off_t		mapping_size;
+
+		mapping_size = imap.br_startoff + imap.br_blockcount - iblock;
+		mapping_size <<= inode->i_blkbits;
+
+		ASSERT(mapping_size > 0);
+		if (mapping_size > size)
+			mapping_size = size;
+		if (mapping_size > LONG_MAX)
+			mapping_size = LONG_MAX;
+
+		bh_result->b_size = mapping_size;
 	}
 
 	return 0;
@@ -1576,7 +1617,7 @@ xfs_end_io_direct(
 	 */
 	ioend->io_offset = offset;
 	ioend->io_size = size;
-	if (ioend->io_type == IOMAP_READ) {
+	if (ioend->io_type == IO_READ) {
 		xfs_finish_ioend(ioend, 0);
 	} else if (private && size > 0) {
 		xfs_finish_ioend(ioend, is_sync_kiocb(iocb));
@@ -1587,7 +1628,7 @@ xfs_end_io_direct(
 		 * didn't map an unwritten extent so switch it's completion
 		 * handler.
 		 */
-		ioend->io_type = IOMAP_NEW;
+		ioend->io_type = IO_NEW;
 		xfs_finish_ioend(ioend, 0);
 	}
 
@@ -1612,10 +1653,10 @@ xfs_vm_direct_IO(
 	struct block_device *bdev;
 	ssize_t		ret;
 
-	bdev = xfs_find_bdev_for_inode(XFS_I(inode));
+	bdev = xfs_find_bdev_for_inode(inode);
 
 	iocb->private = xfs_alloc_ioend(inode, rw == WRITE ?
-					IOMAP_UNWRITTEN : IOMAP_READ);
+					IO_UNWRITTEN : IO_READ);
 
 	ret = blockdev_direct_IO_no_locking(rw, iocb, inode, bdev, iov,
 					    offset, nr_segs,
diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index 44c2b0ef9a4..f01de3c55c4 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -1007,25 +1007,20 @@ xfs_bwrite(
 	struct xfs_mount	*mp,
 	struct xfs_buf		*bp)
 {
-	int			iowait = (bp->b_flags & XBF_ASYNC) == 0;
-	int			error = 0;
+	int			error;
 
 	bp->b_strat = xfs_bdstrat_cb;
 	bp->b_mount = mp;
 	bp->b_flags |= XBF_WRITE;
-	if (!iowait)
-		bp->b_flags |= _XBF_RUN_QUEUES;
+	bp->b_flags &= ~(XBF_ASYNC | XBF_READ);
 
 	xfs_buf_delwri_dequeue(bp);
 	xfs_buf_iostrategy(bp);
 
-	if (iowait) {
-		error = xfs_buf_iowait(bp);
-		if (error)
-			xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR);
-		xfs_buf_relse(bp);
-	}
-
+	error = xfs_buf_iowait(bp);
+	if (error)
+		xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR);
+	xfs_buf_relse(bp);
 	return error;
 }
 
@@ -1614,7 +1609,8 @@ xfs_mapping_buftarg(
 
 STATIC int
 xfs_alloc_delwrite_queue(
-	xfs_buftarg_t		*btp)
+	xfs_buftarg_t		*btp,
+	const char		*fsname)
 {
 	int	error = 0;
 
@@ -1622,7 +1618,7 @@ xfs_alloc_delwrite_queue(
 	INIT_LIST_HEAD(&btp->bt_delwrite_queue);
 	spin_lock_init(&btp->bt_delwrite_lock);
 	btp->bt_flags = 0;
-	btp->bt_task = kthread_run(xfsbufd, btp, "xfsbufd");
+	btp->bt_task = kthread_run(xfsbufd, btp, "xfsbufd/%s", fsname);
 	if (IS_ERR(btp->bt_task)) {
 		error = PTR_ERR(btp->bt_task);
 		goto out_error;
@@ -1635,7 +1631,8 @@ out_error:
 xfs_buftarg_t *
 xfs_alloc_buftarg(
 	struct block_device	*bdev,
-	int			external)
+	int			external,
+	const char		*fsname)
 {
 	xfs_buftarg_t		*btp;
 
@@ -1647,7 +1644,7 @@ xfs_alloc_buftarg(
 		goto error;
 	if (xfs_mapping_buftarg(btp, bdev))
 		goto error;
-	if (xfs_alloc_delwrite_queue(btp))
+	if (xfs_alloc_delwrite_queue(btp, fsname))
 		goto error;
 	xfs_alloc_bufhash(btp, external);
 	return btp;
diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h
index 386e7361e50..5fbecefa5df 100644
--- a/fs/xfs/linux-2.6/xfs_buf.h
+++ b/fs/xfs/linux-2.6/xfs_buf.h
@@ -390,7 +390,7 @@ static inline void xfs_buf_relse(xfs_buf_t *bp)
 /*
  *	Handling of buftargs.
  */
-extern xfs_buftarg_t *xfs_alloc_buftarg(struct block_device *, int);
+extern xfs_buftarg_t *xfs_alloc_buftarg(struct block_device *, int, const char *);
 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);
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index 42dd3bcfba6..d8fb1b5d6cb 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -115,6 +115,8 @@ xfs_file_fsync(
 
 	xfs_iflags_clear(ip, XFS_ITRUNCATED);
 
+	xfs_ioend_wait(ip);
+
 	/*
 	 * 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
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c
index 7b26cc2fd28..699b60cbab9 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl.c
@@ -527,6 +527,10 @@ xfs_attrmulti_by_handle(
 	if (copy_from_user(&am_hreq, arg, sizeof(xfs_fsop_attrmulti_handlereq_t)))
 		return -XFS_ERROR(EFAULT);
 
+	/* overflow check */
+	if (am_hreq.opcount >= INT_MAX / sizeof(xfs_attr_multiop_t))
+		return -E2BIG;
+
 	dentry = xfs_handlereq_to_dentry(parfilp, &am_hreq.hreq);
 	if (IS_ERR(dentry))
 		return PTR_ERR(dentry);
diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.c b/fs/xfs/linux-2.6/xfs_ioctl32.c
index 593c05b4df8..9287135e9bf 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl32.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl32.c
@@ -420,6 +420,10 @@ xfs_compat_attrmulti_by_handle(
 			   sizeof(compat_xfs_fsop_attrmulti_handlereq_t)))
 		return -XFS_ERROR(EFAULT);
 
+	/* overflow check */
+	if (am_hreq.opcount >= INT_MAX / sizeof(compat_xfs_attr_multiop_t))
+		return -E2BIG;
+
 	dentry = xfs_compat_handlereq_to_dentry(parfilp, &am_hreq.hreq);
 	if (IS_ERR(dentry))
 		return PTR_ERR(dentry);
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index e65a7937f3a..9c8019c78c9 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -673,7 +673,10 @@ xfs_vn_fiemap(
 		bm.bmv_length = BTOBB(length);
 
 	/* We add one because in getbmap world count includes the header */
-	bm.bmv_count = fieinfo->fi_extents_max + 1;
+	bm.bmv_count = !fieinfo->fi_extents_max ? MAXEXTNUM :
+					fieinfo->fi_extents_max + 1;
+	bm.bmv_count = min_t(__s32, bm.bmv_count,
+			     (PAGE_SIZE * 16 / sizeof(struct getbmapx)));
 	bm.bmv_iflags = BMV_IF_PREALLOC;
 	if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR)
 		bm.bmv_iflags |= BMV_IF_ATTRFORK;
diff --git a/fs/xfs/linux-2.6/xfs_quotaops.c b/fs/xfs/linux-2.6/xfs_quotaops.c
index 1947514ce1a..e31bf21fe5d 100644
--- a/fs/xfs/linux-2.6/xfs_quotaops.c
+++ b/fs/xfs/linux-2.6/xfs_quotaops.c
@@ -97,7 +97,7 @@ xfs_fs_set_xstate(
 }
 
 STATIC int
-xfs_fs_get_xquota(
+xfs_fs_get_dqblk(
 	struct super_block	*sb,
 	int			type,
 	qid_t			id,
@@ -114,7 +114,7 @@ xfs_fs_get_xquota(
 }
 
 STATIC int
-xfs_fs_set_xquota(
+xfs_fs_set_dqblk(
 	struct super_block	*sb,
 	int			type,
 	qid_t			id,
@@ -135,6 +135,6 @@ xfs_fs_set_xquota(
 const struct quotactl_ops xfs_quotactl_operations = {
 	.get_xstate		= xfs_fs_get_xstate,
 	.set_xstate		= xfs_fs_set_xstate,
-	.get_xquota		= xfs_fs_get_xquota,
-	.set_xquota		= xfs_fs_set_xquota,
+	.get_dqblk		= xfs_fs_get_dqblk,
+	.set_dqblk		= xfs_fs_set_dqblk,
 };
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 2b177c778ba..f24dbe5efde 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -790,18 +790,18 @@ xfs_open_devices(
 	 * Setup xfs_mount buffer target pointers
 	 */
 	error = ENOMEM;
-	mp->m_ddev_targp = xfs_alloc_buftarg(ddev, 0);
+	mp->m_ddev_targp = xfs_alloc_buftarg(ddev, 0, mp->m_fsname);
 	if (!mp->m_ddev_targp)
 		goto out_close_rtdev;
 
 	if (rtdev) {
-		mp->m_rtdev_targp = xfs_alloc_buftarg(rtdev, 1);
+		mp->m_rtdev_targp = xfs_alloc_buftarg(rtdev, 1, mp->m_fsname);
 		if (!mp->m_rtdev_targp)
 			goto out_free_ddev_targ;
 	}
 
 	if (logdev && logdev != ddev) {
-		mp->m_logdev_targp = xfs_alloc_buftarg(logdev, 1);
+		mp->m_logdev_targp = xfs_alloc_buftarg(logdev, 1, mp->m_fsname);
 		if (!mp->m_logdev_targp)
 			goto out_free_rtdev_targ;
 	} else {
@@ -903,7 +903,8 @@ xfsaild_start(
 	struct xfs_ail	*ailp)
 {
 	ailp->xa_target = 0;
-	ailp->xa_task = kthread_run(xfsaild, ailp, "xfsaild");
+	ailp->xa_task = kthread_run(xfsaild, ailp, "xfsaild/%s",
+				    ailp->xa_mount->m_fsname);
 	if (IS_ERR(ailp->xa_task))
 		return -PTR_ERR(ailp->xa_task);
 	return 0;
@@ -1093,6 +1094,7 @@ xfs_fs_write_inode(
 		 * the code will only flush the inode if it isn't already
 		 * being flushed.
 		 */
+		xfs_ioend_wait(ip);
 		xfs_ilock(ip, XFS_ILOCK_SHARED);
 		if (ip->i_update_core) {
 			error = xfs_log_inode(ip);
@@ -1210,6 +1212,7 @@ xfs_fs_put_super(
 
 	xfs_unmountfs(mp);
 	xfs_freesb(mp);
+	xfs_inode_shrinker_unregister(mp);
 	xfs_icsb_destroy_counters(mp);
 	xfs_close_devices(mp);
 	xfs_dmops_put(mp);
@@ -1623,6 +1626,8 @@ xfs_fs_fill_super(
 	if (error)
 		goto fail_vnrele;
 
+	xfs_inode_shrinker_register(mp);
+
 	kfree(mtpt);
 	return 0;
 
@@ -1868,6 +1873,7 @@ init_xfs_fs(void)
 		goto out_cleanup_procfs;
 
 	vfs_initquota();
+	xfs_inode_shrinker_init();
 
 	error = register_filesystem(&xfs_fs_type);
 	if (error)
@@ -1895,6 +1901,7 @@ exit_xfs_fs(void)
 {
 	vfs_exitquota();
 	unregister_filesystem(&xfs_fs_type);
+	xfs_inode_shrinker_destroy();
 	xfs_sysctl_unregister();
 	xfs_cleanup_procfs();
 	xfs_buf_terminate();
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index fd969821575..3884e20bc14 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -95,7 +95,8 @@ xfs_inode_ag_walk(
 					   struct xfs_perag *pag, int flags),
 	int			flags,
 	int			tag,
-	int			exclusive)
+	int			exclusive,
+	int			*nr_to_scan)
 {
 	uint32_t		first_index;
 	int			last_error = 0;
@@ -134,7 +135,7 @@ restart:
 		if (error == EFSCORRUPTED)
 			break;
 
-	} while (1);
+	} while ((*nr_to_scan)--);
 
 	if (skipped) {
 		delay(1);
@@ -150,12 +151,15 @@ xfs_inode_ag_iterator(
 					   struct xfs_perag *pag, int flags),
 	int			flags,
 	int			tag,
-	int			exclusive)
+	int			exclusive,
+	int			*nr_to_scan)
 {
 	int			error = 0;
 	int			last_error = 0;
 	xfs_agnumber_t		ag;
+	int			nr;
 
+	nr = nr_to_scan ? *nr_to_scan : INT_MAX;
 	for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) {
 		struct xfs_perag	*pag;
 
@@ -165,14 +169,18 @@ xfs_inode_ag_iterator(
 			continue;
 		}
 		error = xfs_inode_ag_walk(mp, pag, execute, flags, tag,
-						exclusive);
+						exclusive, &nr);
 		xfs_perag_put(pag);
 		if (error) {
 			last_error = error;
 			if (error == EFSCORRUPTED)
 				break;
 		}
+		if (nr <= 0)
+			break;
 	}
+	if (nr_to_scan)
+		*nr_to_scan = nr;
 	return XFS_ERROR(last_error);
 }
 
@@ -291,7 +299,7 @@ xfs_sync_data(
 	ASSERT((flags & ~(SYNC_TRYLOCK|SYNC_WAIT)) == 0);
 
 	error = xfs_inode_ag_iterator(mp, xfs_sync_inode_data, flags,
-				      XFS_ICI_NO_TAG, 0);
+				      XFS_ICI_NO_TAG, 0, NULL);
 	if (error)
 		return XFS_ERROR(error);
 
@@ -310,7 +318,7 @@ xfs_sync_attr(
 	ASSERT((flags & ~SYNC_WAIT) == 0);
 
 	return xfs_inode_ag_iterator(mp, xfs_sync_inode_attr, flags,
-				     XFS_ICI_NO_TAG, 0);
+				     XFS_ICI_NO_TAG, 0, NULL);
 }
 
 STATIC int
@@ -348,68 +356,23 @@ xfs_commit_dummy_trans(
 
 STATIC int
 xfs_sync_fsdata(
-	struct xfs_mount	*mp,
-	int			flags)
+	struct xfs_mount	*mp)
 {
 	struct xfs_buf		*bp;
-	struct xfs_buf_log_item	*bip;
-	int			error = 0;
-
-	/*
-	 * If this is xfssyncd() then only sync the superblock if we can
-	 * lock it without sleeping and it is not pinned.
-	 */
-	if (flags & SYNC_TRYLOCK) {
-		ASSERT(!(flags & SYNC_WAIT));
-
-		bp = xfs_getsb(mp, XBF_TRYLOCK);
-		if (!bp)
-			goto out;
-
-		bip = XFS_BUF_FSPRIVATE(bp, struct xfs_buf_log_item *);
-		if (!bip || !xfs_buf_item_dirty(bip) || XFS_BUF_ISPINNED(bp))
-			goto out_brelse;
-	} else {
-		bp = xfs_getsb(mp, 0);
-
-		/*
-		 * If the buffer is pinned then push on the log so we won't
-		 * get stuck waiting in the write for someone, maybe
-		 * ourselves, to flush the log.
-		 *
-		 * Even though we just pushed the log above, we did not have
-		 * the superblock buffer locked at that point so it can
-		 * become pinned in between there and here.
-		 */
-		if (XFS_BUF_ISPINNED(bp))
-			xfs_log_force(mp, 0);
-	}
-
-
-	if (flags & SYNC_WAIT)
-		XFS_BUF_UNASYNC(bp);
-	else
-		XFS_BUF_ASYNC(bp);
-
-	error = xfs_bwrite(mp, bp);
-	if (error)
-		return error;
 
 	/*
-	 * If this is a data integrity sync make sure all pending buffers
-	 * are flushed out for the log coverage check below.
+	 * If the buffer is pinned then push on the log so we won't get stuck
+	 * waiting in the write for someone, maybe ourselves, to flush the log.
+	 *
+	 * Even though we just pushed the log above, we did not have the
+	 * superblock buffer locked at that point so it can become pinned in
+	 * between there and here.
 	 */
-	if (flags & SYNC_WAIT)
-		xfs_flush_buftarg(mp->m_ddev_targp, 1);
-
-	if (xfs_log_need_covered(mp))
-		error = xfs_commit_dummy_trans(mp, flags);
-	return error;
+	bp = xfs_getsb(mp, 0);
+	if (XFS_BUF_ISPINNED(bp))
+		xfs_log_force(mp, 0);
 
- out_brelse:
-	xfs_buf_relse(bp);
- out:
-	return error;
+	return xfs_bwrite(mp, bp);
 }
 
 /*
@@ -433,7 +396,7 @@ int
 xfs_quiesce_data(
 	struct xfs_mount	*mp)
 {
-	int error;
+	int			error, error2 = 0;
 
 	/* push non-blocking */
 	xfs_sync_data(mp, 0);
@@ -444,13 +407,20 @@ xfs_quiesce_data(
 	xfs_qm_sync(mp, SYNC_WAIT);
 
 	/* write superblock and hoover up shutdown errors */
-	error = xfs_sync_fsdata(mp, SYNC_WAIT);
+	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_commit_dummy_trans(mp, SYNC_WAIT);
 
 	/* flush data-only devices */
 	if (mp->m_rtdev_targp)
 		XFS_bflush(mp->m_rtdev_targp);
 
-	return error;
+	return error ? error : error2;
 }
 
 STATIC void
@@ -573,9 +543,9 @@ xfs_flush_inodes(
 }
 
 /*
- * Every sync period we need to unpin all items, reclaim inodes, sync
- * quota and write out the superblock. We might need to cover the log
- * to indicate it is idle.
+ * Every sync period we need to unpin all items, reclaim inodes and sync
+ * disk quotas.  We might need to cover the log to indicate that the
+ * filesystem is idle.
  */
 STATIC void
 xfs_sync_worker(
@@ -589,7 +559,8 @@ xfs_sync_worker(
 		xfs_reclaim_inodes(mp, 0);
 		/* dgc: errors ignored here */
 		error = xfs_qm_sync(mp, SYNC_TRYLOCK);
-		error = xfs_sync_fsdata(mp, SYNC_TRYLOCK);
+		if (xfs_log_need_covered(mp))
+			error = xfs_commit_dummy_trans(mp, 0);
 	}
 	mp->m_sync_seq++;
 	wake_up(&mp->m_wait_single_sync_task);
@@ -652,7 +623,7 @@ xfs_syncd_init(
 	mp->m_sync_work.w_syncer = xfs_sync_worker;
 	mp->m_sync_work.w_mount = mp;
 	mp->m_sync_work.w_completion = NULL;
-	mp->m_sync_task = kthread_run(xfssyncd, mp, "xfssyncd");
+	mp->m_sync_task = kthread_run(xfssyncd, mp, "xfssyncd/%s", mp->m_fsname);
 	if (IS_ERR(mp->m_sync_task))
 		return -PTR_ERR(mp->m_sync_task);
 	return 0;
@@ -673,6 +644,7 @@ __xfs_inode_set_reclaim_tag(
 	radix_tree_tag_set(&pag->pag_ici_root,
 			   XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino),
 			   XFS_ICI_RECLAIM_TAG);
+	pag->pag_ici_reclaimable++;
 }
 
 /*
@@ -705,6 +677,7 @@ __xfs_inode_clear_reclaim_tag(
 {
 	radix_tree_tag_clear(&pag->pag_ici_root,
 			XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG);
+	pag->pag_ici_reclaimable--;
 }
 
 /*
@@ -854,5 +827,93 @@ xfs_reclaim_inodes(
 	int		mode)
 {
 	return xfs_inode_ag_iterator(mp, xfs_reclaim_inode, mode,
-					XFS_ICI_RECLAIM_TAG, 1);
+					XFS_ICI_RECLAIM_TAG, 1, NULL);
+}
+
+/*
+ * Shrinker infrastructure.
+ *
+ * This is all far more complex than it needs to be. It adds a global list of
+ * mounts because the shrinkers can only call a global context. We need to make
+ * the shrinkers pass a context to avoid the need for global state.
+ */
+static LIST_HEAD(xfs_mount_list);
+static struct rw_semaphore xfs_mount_list_lock;
+
+static int
+xfs_reclaim_inode_shrink(
+	int		nr_to_scan,
+	gfp_t		gfp_mask)
+{
+	struct xfs_mount *mp;
+	struct xfs_perag *pag;
+	xfs_agnumber_t	ag;
+	int		reclaimable = 0;
+
+	if (nr_to_scan) {
+		if (!(gfp_mask & __GFP_FS))
+			return -1;
+
+		down_read(&xfs_mount_list_lock);
+		list_for_each_entry(mp, &xfs_mount_list, m_mplist) {
+			xfs_inode_ag_iterator(mp, xfs_reclaim_inode, 0,
+					XFS_ICI_RECLAIM_TAG, 1, &nr_to_scan);
+			if (nr_to_scan <= 0)
+				break;
+		}
+		up_read(&xfs_mount_list_lock);
+	}
+
+	down_read(&xfs_mount_list_lock);
+	list_for_each_entry(mp, &xfs_mount_list, m_mplist) {
+		for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) {
+
+			pag = xfs_perag_get(mp, ag);
+			if (!pag->pag_ici_init) {
+				xfs_perag_put(pag);
+				continue;
+			}
+			reclaimable += pag->pag_ici_reclaimable;
+			xfs_perag_put(pag);
+		}
+	}
+	up_read(&xfs_mount_list_lock);
+	return reclaimable;
+}
+
+static struct shrinker xfs_inode_shrinker = {
+	.shrink = xfs_reclaim_inode_shrink,
+	.seeks = DEFAULT_SEEKS,
+};
+
+void __init
+xfs_inode_shrinker_init(void)
+{
+	init_rwsem(&xfs_mount_list_lock);
+	register_shrinker(&xfs_inode_shrinker);
+}
+
+void
+xfs_inode_shrinker_destroy(void)
+{
+	ASSERT(list_empty(&xfs_mount_list));
+	unregister_shrinker(&xfs_inode_shrinker);
+}
+
+void
+xfs_inode_shrinker_register(
+	struct xfs_mount	*mp)
+{
+	down_write(&xfs_mount_list_lock);
+	list_add_tail(&mp->m_mplist, &xfs_mount_list);
+	up_write(&xfs_mount_list_lock);
+}
+
+void
+xfs_inode_shrinker_unregister(
+	struct xfs_mount	*mp)
+{
+	down_write(&xfs_mount_list_lock);
+	list_del(&mp->m_mplist);
+	up_write(&xfs_mount_list_lock);
 }
diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h
index d480c346cab..cdcbaaca988 100644
--- a/fs/xfs/linux-2.6/xfs_sync.h
+++ b/fs/xfs/linux-2.6/xfs_sync.h
@@ -53,6 +53,11 @@ void __xfs_inode_clear_reclaim_tag(struct xfs_mount *mp, struct xfs_perag *pag,
 int xfs_sync_inode_valid(struct xfs_inode *ip, struct xfs_perag *pag);
 int xfs_inode_ag_iterator(struct xfs_mount *mp,
 	int (*execute)(struct xfs_inode *ip, struct xfs_perag *pag, int flags),
-	int flags, int tag, int write_lock);
+	int flags, int tag, int write_lock, int *nr_to_scan);
+
+void xfs_inode_shrinker_init(void);
+void xfs_inode_shrinker_destroy(void);
+void xfs_inode_shrinker_register(struct xfs_mount *mp);
+void xfs_inode_shrinker_unregister(struct xfs_mount *mp);
 
 #endif
diff --git a/fs/xfs/linux-2.6/xfs_trace.c b/fs/xfs/linux-2.6/xfs_trace.c
index 5a107601e96..207fa77f63a 100644
--- a/fs/xfs/linux-2.6/xfs_trace.c
+++ b/fs/xfs/linux-2.6/xfs_trace.c
@@ -41,7 +41,6 @@
 #include "xfs_alloc.h"
 #include "xfs_bmap.h"
 #include "xfs_attr.h"
-#include "xfs_attr_sf.h"
 #include "xfs_attr_leaf.h"
 #include "xfs_log_priv.h"
 #include "xfs_buf_item.h"
@@ -50,6 +49,9 @@
 #include "xfs_aops.h"
 #include "quota/xfs_dquot_item.h"
 #include "quota/xfs_dquot.h"
+#include "xfs_log_recover.h"
+#include "xfs_buf_item.h"
+#include "xfs_inode_item.h"
 
 /*
  * We include this last to have the helpers above available for the trace
diff --git a/fs/xfs/linux-2.6/xfs_trace.h b/fs/xfs/linux-2.6/xfs_trace.h
index fcaa62f0799..8a319cfd290 100644
--- a/fs/xfs/linux-2.6/xfs_trace.h
+++ b/fs/xfs/linux-2.6/xfs_trace.h
@@ -32,6 +32,10 @@ struct xfs_da_node_entry;
 struct xfs_dquot;
 struct xlog_ticket;
 struct log;
+struct xlog_recover;
+struct xlog_recover_item;
+struct xfs_buf_log_format;
+struct xfs_inode_log_format;
 
 DECLARE_EVENT_CLASS(xfs_attr_list_class,
 	TP_PROTO(struct xfs_attr_list_context *ctx),
@@ -562,18 +566,21 @@ DECLARE_EVENT_CLASS(xfs_inode_class,
 		__field(dev_t, dev)
 		__field(xfs_ino_t, ino)
 		__field(int, count)
+		__field(int, pincount)
 		__field(unsigned long, caller_ip)
 	),
 	TP_fast_assign(
 		__entry->dev = VFS_I(ip)->i_sb->s_dev;
 		__entry->ino = ip->i_ino;
 		__entry->count = atomic_read(&VFS_I(ip)->i_count);
+		__entry->pincount = atomic_read(&ip->i_pincount);
 		__entry->caller_ip = caller_ip;
 	),
-	TP_printk("dev %d:%d ino 0x%llx count %d caller %pf",
+	TP_printk("dev %d:%d ino 0x%llx count %d pincount %d caller %pf",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->ino,
 		  __entry->count,
+		  __entry->pincount,
 		  (char *)__entry->caller_ip)
 )
 
@@ -583,6 +590,10 @@ DEFINE_EVENT(xfs_inode_class, name, \
 	TP_ARGS(ip, caller_ip))
 DEFINE_INODE_EVENT(xfs_ihold);
 DEFINE_INODE_EVENT(xfs_irele);
+DEFINE_INODE_EVENT(xfs_inode_pin);
+DEFINE_INODE_EVENT(xfs_inode_unpin);
+DEFINE_INODE_EVENT(xfs_inode_unpin_nowait);
+
 /* the old xfs_itrace_entry tracer - to be replaced by s.th. in the VFS */
 DEFINE_INODE_EVENT(xfs_inode);
 #define xfs_itrace_entry(ip)    \
@@ -642,8 +653,6 @@ DEFINE_EVENT(xfs_dquot_class, name, \
 	TP_PROTO(struct xfs_dquot *dqp), \
 	TP_ARGS(dqp))
 DEFINE_DQUOT_EVENT(xfs_dqadjust);
-DEFINE_DQUOT_EVENT(xfs_dqshake_dirty);
-DEFINE_DQUOT_EVENT(xfs_dqshake_unlink);
 DEFINE_DQUOT_EVENT(xfs_dqreclaim_want);
 DEFINE_DQUOT_EVENT(xfs_dqreclaim_dirty);
 DEFINE_DQUOT_EVENT(xfs_dqreclaim_unlink);
@@ -658,7 +667,6 @@ 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_move);
 DEFINE_DQUOT_EVENT(xfs_dqlookup_done);
 DEFINE_DQUOT_EVENT(xfs_dqget_hit);
 DEFINE_DQUOT_EVENT(xfs_dqget_miss);
@@ -1495,6 +1503,140 @@ DEFINE_EVENT(xfs_swap_extent_class, name, \
 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,
+		struct xlog_recover_item *item, int pass),
+	TP_ARGS(log, trans, item, pass),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(unsigned long, item)
+		__field(xlog_tid_t, tid)
+		__field(int, type)
+		__field(int, pass)
+		__field(int, count)
+		__field(int, total)
+	),
+	TP_fast_assign(
+		__entry->dev = log->l_mp->m_super->s_dev;
+		__entry->item = (unsigned long)item;
+		__entry->tid = trans->r_log_tid;
+		__entry->type = ITEM_TYPE(item);
+		__entry->pass = pass;
+		__entry->count = item->ri_cnt;
+		__entry->total = item->ri_total;
+	),
+	TP_printk("dev %d:%d trans 0x%x, pass %d, item 0x%p, item type %s "
+		  "item region count/total %d/%d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->tid,
+		  __entry->pass,
+		  (void *)__entry->item,
+		  __print_symbolic(__entry->type, XFS_LI_TYPE_DESC),
+		  __entry->count,
+		  __entry->total)
+)
+
+#define DEFINE_LOG_RECOVER_ITEM(name) \
+DEFINE_EVENT(xfs_log_recover_item_class, name, \
+	TP_PROTO(struct log *log, struct xlog_recover *trans, \
+		struct xlog_recover_item *item, int pass), \
+	TP_ARGS(log, trans, item, pass))
+
+DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_add);
+DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_add_cont);
+DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_reorder_head);
+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_ARGS(log, buf_f),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(__int64_t, blkno)
+		__field(unsigned short, len)
+		__field(unsigned short, flags)
+		__field(unsigned short, size)
+		__field(unsigned int, map_size)
+	),
+	TP_fast_assign(
+		__entry->dev = log->l_mp->m_super->s_dev;
+		__entry->blkno = buf_f->blf_blkno;
+		__entry->len = buf_f->blf_len;
+		__entry->flags = buf_f->blf_flags;
+		__entry->size = buf_f->blf_size;
+		__entry->map_size = buf_f->blf_map_size;
+	),
+	TP_printk("dev %d:%d blkno 0x%llx, len %u, flags 0x%x, size %d, "
+			"map_size %d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->blkno,
+		  __entry->len,
+		  __entry->flags,
+		  __entry->size,
+		  __entry->map_size)
+)
+
+#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_ARGS(log, buf_f))
+
+DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_not_cancel);
+DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_cancel);
+DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_cancel_add);
+DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_cancel_ref_inc);
+DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_recover);
+DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_inode_buf);
+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_ARGS(log, in_f),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(unsigned short, size)
+		__field(int, fields)
+		__field(unsigned short, asize)
+		__field(unsigned short, dsize)
+		__field(__int64_t, blkno)
+		__field(int, len)
+		__field(int, boffset)
+	),
+	TP_fast_assign(
+		__entry->dev = log->l_mp->m_super->s_dev;
+		__entry->ino = in_f->ilf_ino;
+		__entry->size = in_f->ilf_size;
+		__entry->fields = in_f->ilf_fields;
+		__entry->asize = in_f->ilf_asize;
+		__entry->dsize = in_f->ilf_dsize;
+		__entry->blkno = in_f->ilf_blkno;
+		__entry->len = in_f->ilf_len;
+		__entry->boffset = in_f->ilf_boffset;
+	),
+	TP_printk("dev %d:%d ino 0x%llx, size %u, fields 0x%x, asize %d, "
+			"dsize %d, blkno 0x%llx, len %d, boffset %d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __entry->size,
+		  __entry->fields,
+		  __entry->asize,
+		  __entry->dsize,
+		  __entry->blkno,
+		  __entry->len,
+		  __entry->boffset)
+)
+#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_ARGS(log, in_f))
+
+DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_recover);
+DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_cancel);
+DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_skip);
+
 #endif /* _TRACE_XFS_H */
 
 #undef TRACE_INCLUDE_PATH
diff --git a/fs/xfs/quota/xfs_dquot.c b/fs/xfs/quota/xfs_dquot.c
index 5f79dd78626..b89ec5df012 100644
--- a/fs/xfs/quota/xfs_dquot.c
+++ b/fs/xfs/quota/xfs_dquot.c
@@ -101,7 +101,7 @@ xfs_qm_dqinit(
 	 * No need to re-initialize these if this is a reclaimed dquot.
 	 */
 	if (brandnewdquot) {
-		dqp->dq_flnext = dqp->dq_flprev = dqp;
+		INIT_LIST_HEAD(&dqp->q_freelist);
 		mutex_init(&dqp->q_qlock);
 		init_waitqueue_head(&dqp->q_pinwait);
 
@@ -119,20 +119,20 @@ xfs_qm_dqinit(
 		 * Only the q_core portion was zeroed in dqreclaim_one().
 		 * So, we need to reset others.
 		 */
-		 dqp->q_nrefs = 0;
-		 dqp->q_blkno = 0;
-		 dqp->MPL_NEXT = dqp->HL_NEXT = NULL;
-		 dqp->HL_PREVP = dqp->MPL_PREVP = NULL;
-		 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(dqp->dq_flnext == dqp->dq_flprev);
+		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);
 	}
@@ -158,7 +158,7 @@ void
 xfs_qm_dqdestroy(
 	xfs_dquot_t	*dqp)
 {
-	ASSERT(! XFS_DQ_IS_ON_FREELIST(dqp));
+	ASSERT(list_empty(&dqp->q_freelist));
 
 	mutex_destroy(&dqp->q_qlock);
 	sv_destroy(&dqp->q_pinwait);
@@ -252,7 +252,7 @@ xfs_qm_adjust_dqtimers(
 		     (be64_to_cpu(d->d_bcount) >=
 		      be64_to_cpu(d->d_blk_hardlimit)))) {
 			d->d_btimer = cpu_to_be32(get_seconds() +
-					XFS_QI_BTIMELIMIT(mp));
+					mp->m_quotainfo->qi_btimelimit);
 		} else {
 			d->d_bwarns = 0;
 		}
@@ -275,7 +275,7 @@ xfs_qm_adjust_dqtimers(
 		     (be64_to_cpu(d->d_icount) >=
 		      be64_to_cpu(d->d_ino_hardlimit)))) {
 			d->d_itimer = cpu_to_be32(get_seconds() +
-					XFS_QI_ITIMELIMIT(mp));
+					mp->m_quotainfo->qi_itimelimit);
 		} else {
 			d->d_iwarns = 0;
 		}
@@ -298,7 +298,7 @@ xfs_qm_adjust_dqtimers(
 		     (be64_to_cpu(d->d_rtbcount) >=
 		      be64_to_cpu(d->d_rtb_hardlimit)))) {
 			d->d_rtbtimer = cpu_to_be32(get_seconds() +
-					XFS_QI_RTBTIMELIMIT(mp));
+					mp->m_quotainfo->qi_rtbtimelimit);
 		} else {
 			d->d_rtbwarns = 0;
 		}
@@ -325,6 +325,7 @@ xfs_qm_init_dquot_blk(
 	uint		type,
 	xfs_buf_t	*bp)
 {
+	struct xfs_quotainfo	*q = mp->m_quotainfo;
 	xfs_dqblk_t	*d;
 	int		curid, i;
 
@@ -337,16 +338,16 @@ xfs_qm_init_dquot_blk(
 	/*
 	 * ID of the first dquot in the block - id's are zero based.
 	 */
-	curid = id - (id % XFS_QM_DQPERBLK(mp));
+	curid = id - (id % q->qi_dqperchunk);
 	ASSERT(curid >= 0);
-	memset(d, 0, BBTOB(XFS_QI_DQCHUNKLEN(mp)));
-	for (i = 0; i < XFS_QM_DQPERBLK(mp); i++, d++, curid++)
+	memset(d, 0, BBTOB(q->qi_dqchunklen));
+	for (i = 0; i < q->qi_dqperchunk; i++, d++, curid++)
 		xfs_qm_dqinit_core(curid, type, d);
 	xfs_trans_dquot_buf(tp, bp,
 			    (type & XFS_DQ_USER ? XFS_BLI_UDQUOT_BUF :
 			    ((type & XFS_DQ_PROJ) ? XFS_BLI_PDQUOT_BUF :
 			     XFS_BLI_GDQUOT_BUF)));
-	xfs_trans_log_buf(tp, bp, 0, BBTOB(XFS_QI_DQCHUNKLEN(mp)) - 1);
+	xfs_trans_log_buf(tp, bp, 0, BBTOB(q->qi_dqchunklen) - 1);
 }
 
 
@@ -419,7 +420,7 @@ xfs_qm_dqalloc(
 	/* now we can just get the buffer (there's nothing to read yet) */
 	bp = xfs_trans_get_buf(tp, mp->m_ddev_targp,
 			       dqp->q_blkno,
-			       XFS_QI_DQCHUNKLEN(mp),
+			       mp->m_quotainfo->qi_dqchunklen,
 			       0);
 	if (!bp || (error = XFS_BUF_GETERROR(bp)))
 		goto error1;
@@ -500,7 +501,8 @@ xfs_qm_dqtobp(
 	 */
 	if (dqp->q_blkno == (xfs_daddr_t) 0) {
 		/* We use the id as an index */
-		dqp->q_fileoffset = (xfs_fileoff_t)id / XFS_QM_DQPERBLK(mp);
+		dqp->q_fileoffset = (xfs_fileoff_t)id /
+					mp->m_quotainfo->qi_dqperchunk;
 		nmaps = 1;
 		quotip = XFS_DQ_TO_QIP(dqp);
 		xfs_ilock(quotip, XFS_ILOCK_SHARED);
@@ -529,7 +531,7 @@ xfs_qm_dqtobp(
 		/*
 		 * offset of dquot in the (fixed sized) dquot chunk.
 		 */
-		dqp->q_bufoffset = (id % XFS_QM_DQPERBLK(mp)) *
+		dqp->q_bufoffset = (id % mp->m_quotainfo->qi_dqperchunk) *
 			sizeof(xfs_dqblk_t);
 		if (map.br_startblock == HOLESTARTBLOCK) {
 			/*
@@ -559,15 +561,13 @@ xfs_qm_dqtobp(
 	 * Read in the buffer, unless we've just done the allocation
 	 * (in which case we already have the buf).
 	 */
-	if (! newdquot) {
+	if (!newdquot) {
 		trace_xfs_dqtobp_read(dqp);
 
-		if ((error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
-					       dqp->q_blkno,
-					       XFS_QI_DQCHUNKLEN(mp),
-					       0, &bp))) {
-			return (error);
-		}
+		error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
+					   dqp->q_blkno,
+					   mp->m_quotainfo->qi_dqchunklen,
+					   0, &bp);
 		if (error || !bp)
 			return XFS_ERROR(error);
 	}
@@ -689,14 +689,14 @@ xfs_qm_idtodq(
 	tp = NULL;
 	if (flags & XFS_QMOPT_DQALLOC) {
 		tp = xfs_trans_alloc(mp, XFS_TRANS_QM_DQALLOC);
-		if ((error = xfs_trans_reserve(tp,
-				       XFS_QM_DQALLOC_SPACE_RES(mp),
-				       XFS_WRITE_LOG_RES(mp) +
-					      BBTOB(XFS_QI_DQCHUNKLEN(mp)) - 1 +
-					      128,
-				       0,
-				       XFS_TRANS_PERM_LOG_RES,
-				       XFS_WRITE_LOG_COUNT))) {
+		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;
 		}
@@ -751,7 +751,6 @@ xfs_qm_dqlookup(
 {
 	xfs_dquot_t		*dqp;
 	uint			flist_locked;
-	xfs_dquot_t		*d;
 
 	ASSERT(mutex_is_locked(&qh->qh_lock));
 
@@ -760,7 +759,7 @@ xfs_qm_dqlookup(
 	/*
 	 * Traverse the hashchain looking for a match
 	 */
-	for (dqp = qh->qh_next; dqp != NULL; dqp = dqp->HL_NEXT) {
+	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
@@ -772,12 +771,12 @@ xfs_qm_dqlookup(
 			/*
 			 * All in core dquots must be on the dqlist of mp
 			 */
-			ASSERT(dqp->MPL_PREVP != NULL);
+			ASSERT(!list_empty(&dqp->q_mplist));
 
 			xfs_dqlock(dqp);
 			if (dqp->q_nrefs == 0) {
-				ASSERT (XFS_DQ_IS_ON_FREELIST(dqp));
-				if (! xfs_qm_freelist_lock_nowait(xfs_Gqm)) {
+				ASSERT(!list_empty(&dqp->q_freelist));
+				if (!mutex_trylock(&xfs_Gqm->qm_dqfrlist_lock)) {
 					trace_xfs_dqlookup_want(dqp);
 
 					/*
@@ -787,7 +786,7 @@ xfs_qm_dqlookup(
 					 */
 					dqp->dq_flags |= XFS_DQ_WANT;
 					xfs_dqunlock(dqp);
-					xfs_qm_freelist_lock(xfs_Gqm);
+					mutex_lock(&xfs_Gqm->qm_dqfrlist_lock);
 					xfs_dqlock(dqp);
 					dqp->dq_flags &= ~(XFS_DQ_WANT);
 				}
@@ -802,46 +801,28 @@ xfs_qm_dqlookup(
 
 			if (flist_locked) {
 				if (dqp->q_nrefs != 0) {
-					xfs_qm_freelist_unlock(xfs_Gqm);
+					mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
 					flist_locked = B_FALSE;
 				} else {
-					/*
-					 * take it off the freelist
-					 */
+					/* take it off the freelist */
 					trace_xfs_dqlookup_freelist(dqp);
-					XQM_FREELIST_REMOVE(dqp);
-					/* xfs_qm_freelist_print(&(xfs_Gqm->
-							qm_dqfreelist),
-							"after removal"); */
+					list_del_init(&dqp->q_freelist);
+					xfs_Gqm->qm_dqfrlist_cnt--;
 				}
 			}
 
-			/*
-			 * grab a reference
-			 */
 			XFS_DQHOLD(dqp);
 
 			if (flist_locked)
-				xfs_qm_freelist_unlock(xfs_Gqm);
+				mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
 			/*
 			 * move the dquot to the front of the hashchain
 			 */
 			ASSERT(mutex_is_locked(&qh->qh_lock));
-			if (dqp->HL_PREVP != &qh->qh_next) {
-				trace_xfs_dqlookup_move(dqp);
-				if ((d = dqp->HL_NEXT))
-					d->HL_PREVP = dqp->HL_PREVP;
-				*(dqp->HL_PREVP) = d;
-				d = qh->qh_next;
-				d->HL_PREVP = &dqp->HL_NEXT;
-				dqp->HL_NEXT = d;
-				dqp->HL_PREVP = &qh->qh_next;
-				qh->qh_next = dqp;
-			}
+			list_move(&dqp->q_hashlist, &qh->qh_list);
 			trace_xfs_dqlookup_done(dqp);
 			*O_dqpp = dqp;
-			ASSERT(mutex_is_locked(&qh->qh_lock));
-			return (0);
+			return 0;
 		}
 	}
 
@@ -975,16 +956,17 @@ xfs_qm_dqget(
 	 */
 	if (ip) {
 		xfs_ilock(ip, XFS_ILOCK_EXCL);
-		if (! XFS_IS_DQTYPE_ON(mp, type)) {
-			/* inode stays locked on return */
-			xfs_qm_dqdestroy(dqp);
-			return XFS_ERROR(ESRCH);
-		}
+
 		/*
 		 * 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) {
 				xfs_qm_dqdestroy(dqp);
 				dqp = ip->i_udquot;
@@ -992,6 +974,11 @@ xfs_qm_dqget(
 				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;
@@ -1033,13 +1020,14 @@ xfs_qm_dqget(
 	 */
 	ASSERT(mutex_is_locked(&h->qh_lock));
 	dqp->q_hash = h;
-	XQM_HASHLIST_INSERT(h, dqp);
+	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
 	 */
-	xfs_qm_mplist_lock(mp);
+	mutex_lock(&mp->m_quotainfo->qi_dqlist_lock);
 
 	/*
 	 * We return a locked dquot to the caller, with a reference taken
@@ -1047,9 +1035,9 @@ xfs_qm_dqget(
 	xfs_dqlock(dqp);
 	dqp->q_nrefs = 1;
 
-	XQM_MPLIST_INSERT(&(XFS_QI_MPL_LIST(mp)), dqp);
-
-	xfs_qm_mplist_unlock(mp);
+	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);
  dqret:
 	ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL));
@@ -1086,10 +1074,10 @@ xfs_qm_dqput(
 	 * drop the dqlock and acquire the freelist and dqlock
 	 * in the right order; but try to get it out-of-order first
 	 */
-	if (! xfs_qm_freelist_lock_nowait(xfs_Gqm)) {
+	if (!mutex_trylock(&xfs_Gqm->qm_dqfrlist_lock)) {
 		trace_xfs_dqput_wait(dqp);
 		xfs_dqunlock(dqp);
-		xfs_qm_freelist_lock(xfs_Gqm);
+		mutex_lock(&xfs_Gqm->qm_dqfrlist_lock);
 		xfs_dqlock(dqp);
 	}
 
@@ -1100,10 +1088,8 @@ xfs_qm_dqput(
 		if (--dqp->q_nrefs == 0) {
 			trace_xfs_dqput_free(dqp);
 
-			/*
-			 * insert at end of the freelist.
-			 */
-			XQM_FREELIST_INSERT(&(xfs_Gqm->qm_dqfreelist), dqp);
+			list_add_tail(&dqp->q_freelist, &xfs_Gqm->qm_dqfrlist);
+			xfs_Gqm->qm_dqfrlist_cnt++;
 
 			/*
 			 * If we just added a udquot to the freelist, then
@@ -1118,10 +1104,6 @@ xfs_qm_dqput(
 				xfs_dqlock(gdqp);
 				dqp->q_gdquot = NULL;
 			}
-
-			/* xfs_qm_freelist_print(&(xfs_Gqm->qm_dqfreelist),
-			   "@@@@@++ Free list (after append) @@@@@+");
-			   */
 		}
 		xfs_dqunlock(dqp);
 
@@ -1133,7 +1115,7 @@ xfs_qm_dqput(
 			break;
 		dqp = gdqp;
 	}
-	xfs_qm_freelist_unlock(xfs_Gqm);
+	mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
 }
 
 /*
@@ -1386,10 +1368,10 @@ int
 xfs_qm_dqpurge(
 	xfs_dquot_t	*dqp)
 {
-	xfs_dqhash_t	*thishash;
+	xfs_dqhash_t	*qh = dqp->q_hash;
 	xfs_mount_t	*mp = dqp->q_mount;
 
-	ASSERT(XFS_QM_IS_MPLIST_LOCKED(mp));
+	ASSERT(mutex_is_locked(&mp->m_quotainfo->qi_dqlist_lock));
 	ASSERT(mutex_is_locked(&dqp->q_hash->qh_lock));
 
 	xfs_dqlock(dqp);
@@ -1407,7 +1389,7 @@ xfs_qm_dqpurge(
 		return (1);
 	}
 
-	ASSERT(XFS_DQ_IS_ON_FREELIST(dqp));
+	ASSERT(!list_empty(&dqp->q_freelist));
 
 	/*
 	 * If we're turning off quotas, we have to make sure that, for
@@ -1452,14 +1434,16 @@ xfs_qm_dqpurge(
 	ASSERT(XFS_FORCED_SHUTDOWN(mp) ||
 	       !(dqp->q_logitem.qli_item.li_flags & XFS_LI_IN_AIL));
 
-	thishash = dqp->q_hash;
-	XQM_HASHLIST_REMOVE(thishash, dqp);
-	XQM_MPLIST_REMOVE(&(XFS_QI_MPL_LIST(mp)), dqp);
+	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(XFS_DQ_IS_ON_FREELIST(dqp));
+	ASSERT(!list_empty(&dqp->q_freelist));
 
 	dqp->q_mount = NULL;
 	dqp->q_hash = NULL;
@@ -1467,7 +1451,7 @@ xfs_qm_dqpurge(
 	memset(&dqp->q_core, 0, sizeof(dqp->q_core));
 	xfs_dqfunlock(dqp);
 	xfs_dqunlock(dqp);
-	mutex_unlock(&thishash->qh_lock);
+	mutex_unlock(&qh->qh_lock);
 	return (0);
 }
 
@@ -1517,6 +1501,7 @@ void
 xfs_qm_dqflock_pushbuf_wait(
 	xfs_dquot_t	*dqp)
 {
+	xfs_mount_t	*mp = dqp->q_mount;
 	xfs_buf_t	*bp;
 
 	/*
@@ -1525,14 +1510,14 @@ xfs_qm_dqflock_pushbuf_wait(
 	 * out immediately.  We'll be able to acquire
 	 * the flush lock when the I/O completes.
 	 */
-	bp = xfs_incore(dqp->q_mount->m_ddev_targp, dqp->q_blkno,
-		    XFS_QI_DQCHUNKLEN(dqp->q_mount), XBF_TRYLOCK);
+	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(dqp->q_mount, 0);
+			xfs_log_force(mp, 0);
 		xfs_buf_delwri_promote(bp);
 		wake_up_process(bp->b_target->bt_task);
 	}
diff --git a/fs/xfs/quota/xfs_dquot.h b/fs/xfs/quota/xfs_dquot.h
index a0f7da586d1..5da3a23b820 100644
--- a/fs/xfs/quota/xfs_dquot.h
+++ b/fs/xfs/quota/xfs_dquot.h
@@ -33,40 +33,23 @@
  * The hash chain headers (hash buckets)
  */
 typedef struct xfs_dqhash {
-	struct xfs_dquot *qh_next;
+	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;
 
-typedef struct xfs_dqlink {
-	struct xfs_dquot  *ql_next;	/* forward link */
-	struct xfs_dquot **ql_prevp;	/* pointer to prev ql_next */
-} xfs_dqlink_t;
-
 struct xfs_mount;
 struct xfs_trans;
 
 /*
- * This is the marker which is designed to occupy the first few
- * bytes of the xfs_dquot_t structure. Even inside this, the freelist pointers
- * must come first.
- * This serves as the marker ("sentinel") when we have to restart list
- * iterations because of locking considerations.
- */
-typedef struct xfs_dqmarker {
-	struct xfs_dquot*dqm_flnext;	/* link to freelist: must be first */
-	struct xfs_dquot*dqm_flprev;
-	xfs_dqlink_t	 dqm_mplist;	/* link to mount's list of dquots */
-	xfs_dqlink_t	 dqm_hashlist;	/* link to the hash chain */
-	uint		 dqm_flags;	/* various flags (XFS_DQ_*) */
-} xfs_dqmarker_t;
-
-/*
  * The incore dquot structure
  */
 typedef struct xfs_dquot {
-	xfs_dqmarker_t	 q_lists;	/* list ptrs, q_flags (marker) */
+	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 xfs_mount*q_mount;	/* filesystem this relates to */
 	struct xfs_trans*q_transp;	/* trans this belongs to currently */
@@ -87,13 +70,6 @@ typedef struct xfs_dquot {
 	wait_queue_head_t q_pinwait;	/* dquot pinning wait queue */
 } xfs_dquot_t;
 
-
-#define dq_flnext	q_lists.dqm_flnext
-#define dq_flprev	q_lists.dqm_flprev
-#define dq_mplist	q_lists.dqm_mplist
-#define dq_hashlist	q_lists.dqm_hashlist
-#define dq_flags	q_lists.dqm_flags
-
 /*
  * Lock hierarchy for q_qlock:
  *	XFS_QLOCK_NORMAL is the implicit default,
@@ -127,7 +103,6 @@ static inline void xfs_dqfunlock(xfs_dquot_t *dqp)
 }
 
 #define XFS_DQ_IS_LOCKED(dqp)	(mutex_is_locked(&((dqp)->q_qlock)))
-#define XFS_DQ_IS_ON_FREELIST(dqp)  ((dqp)->dq_flnext != (dqp))
 #define XFS_DQ_IS_DIRTY(dqp)	((dqp)->dq_flags & XFS_DQ_DIRTY)
 #define XFS_QM_ISUDQ(dqp)	((dqp)->dq_flags & XFS_DQ_USER)
 #define XFS_QM_ISPDQ(dqp)	((dqp)->dq_flags & XFS_DQ_PROJ)
diff --git a/fs/xfs/quota/xfs_dquot_item.c b/fs/xfs/quota/xfs_dquot_item.c
index 4e4ee9a5719..8d89a24ae32 100644
--- a/fs/xfs/quota/xfs_dquot_item.c
+++ b/fs/xfs/quota/xfs_dquot_item.c
@@ -107,8 +107,7 @@ xfs_qm_dquot_logitem_pin(
 /* ARGSUSED */
 STATIC void
 xfs_qm_dquot_logitem_unpin(
-	xfs_dq_logitem_t *logitem,
-	int		  stale)
+	xfs_dq_logitem_t *logitem)
 {
 	xfs_dquot_t *dqp = logitem->qli_dquot;
 
@@ -123,7 +122,7 @@ xfs_qm_dquot_logitem_unpin_remove(
 	xfs_dq_logitem_t *logitem,
 	xfs_trans_t	 *tp)
 {
-	xfs_qm_dquot_logitem_unpin(logitem, 0);
+	xfs_qm_dquot_logitem_unpin(logitem);
 }
 
 /*
@@ -228,7 +227,7 @@ xfs_qm_dquot_logitem_pushbuf(
 	}
 	mp = dqp->q_mount;
 	bp = xfs_incore(mp->m_ddev_targp, qip->qli_format.qlf_blkno,
-		    XFS_QI_DQCHUNKLEN(mp), XBF_TRYLOCK);
+			mp->m_quotainfo->qi_dqchunklen, XBF_TRYLOCK);
 	xfs_dqunlock(dqp);
 	if (!bp)
 		return;
@@ -329,8 +328,7 @@ static struct xfs_item_ops xfs_dquot_item_ops = {
 	.iop_format	= (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
 					xfs_qm_dquot_logitem_format,
 	.iop_pin	= (void(*)(xfs_log_item_t*))xfs_qm_dquot_logitem_pin,
-	.iop_unpin	= (void(*)(xfs_log_item_t*, int))
-					xfs_qm_dquot_logitem_unpin,
+	.iop_unpin	= (void(*)(xfs_log_item_t*))xfs_qm_dquot_logitem_unpin,
 	.iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t*))
 					xfs_qm_dquot_logitem_unpin_remove,
 	.iop_trylock	= (uint(*)(xfs_log_item_t*))
@@ -357,9 +355,8 @@ xfs_qm_dquot_logitem_init(
 	xfs_dq_logitem_t  *lp;
 	lp = &dqp->q_logitem;
 
-	lp->qli_item.li_type = XFS_LI_DQUOT;
-	lp->qli_item.li_ops = &xfs_dquot_item_ops;
-	lp->qli_item.li_mountp = dqp->q_mount;
+	xfs_log_item_init(dqp->q_mount, &lp->qli_item, XFS_LI_DQUOT,
+					&xfs_dquot_item_ops);
 	lp->qli_dquot = dqp;
 	lp->qli_format.qlf_type = XFS_LI_DQUOT;
 	lp->qli_format.qlf_id = be32_to_cpu(dqp->q_core.d_id);
@@ -426,7 +423,7 @@ xfs_qm_qoff_logitem_pin(xfs_qoff_logitem_t *qf)
  */
 /*ARGSUSED*/
 STATIC void
-xfs_qm_qoff_logitem_unpin(xfs_qoff_logitem_t *qf, int stale)
+xfs_qm_qoff_logitem_unpin(xfs_qoff_logitem_t *qf)
 {
 	return;
 }
@@ -537,8 +534,7 @@ static struct xfs_item_ops xfs_qm_qoffend_logitem_ops = {
 	.iop_format	= (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
 					xfs_qm_qoff_logitem_format,
 	.iop_pin	= (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_pin,
-	.iop_unpin	= (void(*)(xfs_log_item_t* ,int))
-					xfs_qm_qoff_logitem_unpin,
+	.iop_unpin	= (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_unpin,
 	.iop_unpin_remove = (void(*)(xfs_log_item_t*,xfs_trans_t*))
 					xfs_qm_qoff_logitem_unpin_remove,
 	.iop_trylock	= (uint(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_trylock,
@@ -559,8 +555,7 @@ static struct xfs_item_ops xfs_qm_qoff_logitem_ops = {
 	.iop_format	= (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
 					xfs_qm_qoff_logitem_format,
 	.iop_pin	= (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_pin,
-	.iop_unpin	= (void(*)(xfs_log_item_t*, int))
-					xfs_qm_qoff_logitem_unpin,
+	.iop_unpin	= (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_unpin,
 	.iop_unpin_remove = (void(*)(xfs_log_item_t*,xfs_trans_t*))
 					xfs_qm_qoff_logitem_unpin_remove,
 	.iop_trylock	= (uint(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_trylock,
@@ -586,11 +581,8 @@ xfs_qm_qoff_logitem_init(
 
 	qf = (xfs_qoff_logitem_t*) kmem_zalloc(sizeof(xfs_qoff_logitem_t), KM_SLEEP);
 
-	qf->qql_item.li_type = XFS_LI_QUOTAOFF;
-	if (start)
-		qf->qql_item.li_ops = &xfs_qm_qoffend_logitem_ops;
-	else
-		qf->qql_item.li_ops = &xfs_qm_qoff_logitem_ops;
+	xfs_log_item_init(mp, &qf->qql_item, XFS_LI_QUOTAOFF, start ?
+			&xfs_qm_qoffend_logitem_ops : &xfs_qm_qoff_logitem_ops);
 	qf->qql_item.li_mountp = mp;
 	qf->qql_format.qf_type = XFS_LI_QUOTAOFF;
 	qf->qql_format.qf_flags = flags;
diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c
index 417e61e3d9d..38e76414664 100644
--- a/fs/xfs/quota/xfs_qm.c
+++ b/fs/xfs/quota/xfs_qm.c
@@ -67,9 +67,6 @@ static cred_t	xfs_zerocr;
 STATIC void	xfs_qm_list_init(xfs_dqlist_t *, char *, int);
 STATIC void	xfs_qm_list_destroy(xfs_dqlist_t *);
 
-STATIC void	xfs_qm_freelist_init(xfs_frlist_t *);
-STATIC void	xfs_qm_freelist_destroy(xfs_frlist_t *);
-
 STATIC int	xfs_qm_init_quotainos(xfs_mount_t *);
 STATIC int	xfs_qm_init_quotainfo(xfs_mount_t *);
 STATIC int	xfs_qm_shake(int, gfp_t);
@@ -84,21 +81,25 @@ extern struct mutex	qcheck_lock;
 #endif
 
 #ifdef QUOTADEBUG
-#define XQM_LIST_PRINT(l, NXT, title) \
-{ \
-	xfs_dquot_t	*dqp; int i = 0; \
-	cmn_err(CE_DEBUG, "%s (#%d)", title, (int) (l)->qh_nelems); \
-	for (dqp = (l)->qh_next; dqp != NULL; dqp = dqp->NXT) { \
-		cmn_err(CE_DEBUG, "   %d.  \"%d (%s)\"   " \
-				  "bcnt = %d, icnt = %d, refs = %d", \
-			++i, (int) be32_to_cpu(dqp->q_core.d_id), \
-			DQFLAGTO_TYPESTR(dqp),	     \
-			(int) be64_to_cpu(dqp->q_core.d_bcount), \
-			(int) be64_to_cpu(dqp->q_core.d_icount), \
-			(int) dqp->q_nrefs);  } \
+static void
+xfs_qm_dquot_list_print(
+	struct xfs_mount *mp)
+{
+	xfs_dquot_t	*dqp;
+	int		i = 0;
+
+	list_for_each_entry(dqp, &mp->m_quotainfo->qi_dqlist_lock, qi_mplist) {
+		cmn_err(CE_DEBUG, "   %d. \"%d (%s)\"   "
+				  "bcnt = %lld, icnt = %lld, refs = %d",
+			i++, be32_to_cpu(dqp->q_core.d_id),
+			DQFLAGTO_TYPESTR(dqp),
+			(long long)be64_to_cpu(dqp->q_core.d_bcount),
+			(long long)be64_to_cpu(dqp->q_core.d_icount),
+			dqp->q_nrefs);
+	}
 }
 #else
-#define XQM_LIST_PRINT(l, NXT, title) do { } while (0)
+static void xfs_qm_dquot_list_print(struct xfs_mount *mp) { }
 #endif
 
 /*
@@ -144,7 +145,9 @@ xfs_Gqm_init(void)
 	/*
 	 * Freelist of all dquots of all file systems
 	 */
-	xfs_qm_freelist_init(&(xqm->qm_dqfreelist));
+	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.
@@ -189,6 +192,7 @@ STATIC void
 xfs_qm_destroy(
 	struct xfs_qm	*xqm)
 {
+	struct xfs_dquot *dqp, *n;
 	int		hsize, i;
 
 	ASSERT(xqm != NULL);
@@ -204,7 +208,21 @@ xfs_qm_destroy(
 	xqm->qm_usr_dqhtable = NULL;
 	xqm->qm_grp_dqhtable = NULL;
 	xqm->qm_dqhashmask = 0;
-	xfs_qm_freelist_destroy(&(xqm->qm_dqfreelist));
+
+	/* frlist cleanup */
+	mutex_lock(&xqm->qm_dqfrlist_lock);
+	list_for_each_entry_safe(dqp, n, &xqm->qm_dqfrlist, q_freelist) {
+		xfs_dqlock(dqp);
+#ifdef QUOTADEBUG
+		cmn_err(CE_DEBUG, "FREELIST destroy 0x%p", dqp);
+#endif
+		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);
 #ifdef DEBUG
 	mutex_destroy(&qcheck_lock);
 #endif
@@ -256,7 +274,7 @@ STATIC void
 xfs_qm_rele_quotafs_ref(
 	struct xfs_mount *mp)
 {
-	xfs_dquot_t	*dqp, *nextdqp;
+	xfs_dquot_t	*dqp, *n;
 
 	ASSERT(xfs_Gqm);
 	ASSERT(xfs_Gqm->qm_nrefs > 0);
@@ -264,26 +282,24 @@ xfs_qm_rele_quotafs_ref(
 	/*
 	 * Go thru the freelist and destroy all inactive dquots.
 	 */
-	xfs_qm_freelist_lock(xfs_Gqm);
+	mutex_lock(&xfs_Gqm->qm_dqfrlist_lock);
 
-	for (dqp = xfs_Gqm->qm_dqfreelist.qh_next;
-	     dqp != (xfs_dquot_t *)&(xfs_Gqm->qm_dqfreelist); ) {
+	list_for_each_entry_safe(dqp, n, &xfs_Gqm->qm_dqfrlist, q_freelist) {
 		xfs_dqlock(dqp);
-		nextdqp = dqp->dq_flnext;
 		if (dqp->dq_flags & XFS_DQ_INACTIVE) {
 			ASSERT(dqp->q_mount == NULL);
 			ASSERT(! XFS_DQ_IS_DIRTY(dqp));
-			ASSERT(dqp->HL_PREVP == NULL);
-			ASSERT(dqp->MPL_PREVP == NULL);
-			XQM_FREELIST_REMOVE(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);
 		} else {
 			xfs_dqunlock(dqp);
 		}
-		dqp = nextdqp;
 	}
-	xfs_qm_freelist_unlock(xfs_Gqm);
+	mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
 
 	/*
 	 * Destroy the entire XQM. If somebody mounts with quotaon, this'll
@@ -305,7 +321,7 @@ xfs_qm_unmount(
 	struct xfs_mount	*mp)
 {
 	if (mp->m_quotainfo) {
-		xfs_qm_dqpurge_all(mp, XFS_QMOPT_QUOTALL | XFS_QMOPT_UMOUNTING);
+		xfs_qm_dqpurge_all(mp, XFS_QMOPT_QUOTALL);
 		xfs_qm_destroy_quotainfo(mp);
 	}
 }
@@ -449,20 +465,21 @@ xfs_qm_unmount_quotas(
  */
 STATIC int
 xfs_qm_dqflush_all(
-	xfs_mount_t	*mp,
-	int		sync_mode)
+	struct xfs_mount	*mp,
+	int			sync_mode)
 {
-	int		recl;
-	xfs_dquot_t	*dqp;
-	int		niters;
-	int		error;
+	struct xfs_quotainfo	*q = mp->m_quotainfo;
+	int			recl;
+	struct xfs_dquot	*dqp;
+	int			niters;
+	int			error;
 
-	if (mp->m_quotainfo == NULL)
+	if (!q)
 		return 0;
 	niters = 0;
 again:
-	xfs_qm_mplist_lock(mp);
-	FOREACH_DQUOT_IN_MP(dqp, mp) {
+	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);
@@ -470,7 +487,7 @@ again:
 		}
 
 		/* XXX a sentinel would be better */
-		recl = XFS_QI_MPLRECLAIMS(mp);
+		recl = q->qi_dqreclaims;
 		if (!xfs_dqflock_nowait(dqp)) {
 			/*
 			 * If we can't grab the flush lock then check
@@ -485,21 +502,21 @@ again:
 		 * Let go of the mplist lock. We don't want to hold it
 		 * across a disk write.
 		 */
-		xfs_qm_mplist_unlock(mp);
+		mutex_unlock(&q->qi_dqlist_lock);
 		error = xfs_qm_dqflush(dqp, sync_mode);
 		xfs_dqunlock(dqp);
 		if (error)
 			return error;
 
-		xfs_qm_mplist_lock(mp);
-		if (recl != XFS_QI_MPLRECLAIMS(mp)) {
-			xfs_qm_mplist_unlock(mp);
+		mutex_lock(&q->qi_dqlist_lock);
+		if (recl != q->qi_dqreclaims) {
+			mutex_unlock(&q->qi_dqlist_lock);
 			/* XXX restart limit */
 			goto again;
 		}
 	}
 
-	xfs_qm_mplist_unlock(mp);
+	mutex_unlock(&q->qi_dqlist_lock);
 	/* return ! busy */
 	return 0;
 }
@@ -509,15 +526,15 @@ again:
  */
 STATIC void
 xfs_qm_detach_gdquots(
-	xfs_mount_t	*mp)
+	struct xfs_mount	*mp)
 {
-	xfs_dquot_t	*dqp, *gdqp;
-	int		nrecl;
+	struct xfs_quotainfo	*q = mp->m_quotainfo;
+	struct xfs_dquot	*dqp, *gdqp;
+	int			nrecl;
 
  again:
-	ASSERT(XFS_QM_IS_MPLIST_LOCKED(mp));
-	dqp = XFS_QI_MPLNEXT(mp);
-	while (dqp) {
+	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);
@@ -530,15 +547,14 @@ xfs_qm_detach_gdquots(
 			 * Can't hold the mplist lock across a dqput.
 			 * XXXmust convert to marker based iterations here.
 			 */
-			nrecl = XFS_QI_MPLRECLAIMS(mp);
-			xfs_qm_mplist_unlock(mp);
+			nrecl = q->qi_dqreclaims;
+			mutex_unlock(&q->qi_dqlist_lock);
 			xfs_qm_dqput(gdqp);
 
-			xfs_qm_mplist_lock(mp);
-			if (nrecl != XFS_QI_MPLRECLAIMS(mp))
+			mutex_lock(&q->qi_dqlist_lock);
+			if (nrecl != q->qi_dqreclaims)
 				goto again;
 		}
-		dqp = dqp->MPL_NEXT;
 	}
 }
 
@@ -550,23 +566,23 @@ xfs_qm_detach_gdquots(
  */
 STATIC int
 xfs_qm_dqpurge_int(
-	xfs_mount_t	*mp,
-	uint		flags) /* QUOTAOFF/UMOUNTING/UQUOTA/PQUOTA/GQUOTA */
+	struct xfs_mount	*mp,
+	uint			flags)
 {
-	xfs_dquot_t	*dqp;
-	uint		dqtype;
-	int		nrecl;
-	xfs_dquot_t	*nextdqp;
-	int		nmisses;
+	struct xfs_quotainfo	*q = mp->m_quotainfo;
+	struct xfs_dquot	*dqp, *n;
+	uint			dqtype;
+	int			nrecl;
+	int			nmisses;
 
-	if (mp->m_quotainfo == NULL)
+	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;
 
-	xfs_qm_mplist_lock(mp);
+	mutex_lock(&q->qi_dqlist_lock);
 
 	/*
 	 * In the first pass through all incore dquots of this filesystem,
@@ -578,28 +594,25 @@ xfs_qm_dqpurge_int(
 
       again:
 	nmisses = 0;
-	ASSERT(XFS_QM_IS_MPLIST_LOCKED(mp));
+	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.
 	 */
-	dqp = XFS_QI_MPLNEXT(mp);
-	while (dqp) {
+	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) {
-			dqp = dqp->MPL_NEXT;
+		if ((dqp->dq_flags & dqtype) == 0)
 			continue;
-		}
 
 		if (!mutex_trylock(&dqp->q_hash->qh_lock)) {
-			nrecl = XFS_QI_MPLRECLAIMS(mp);
-			xfs_qm_mplist_unlock(mp);
+			nrecl = q->qi_dqreclaims;
+			mutex_unlock(&q->qi_dqlist_lock);
 			mutex_lock(&dqp->q_hash->qh_lock);
-			xfs_qm_mplist_lock(mp);
+			mutex_lock(&q->qi_dqlist_lock);
 
 			/*
 			 * XXXTheoretically, we can get into a very long
@@ -607,7 +620,7 @@ xfs_qm_dqpurge_int(
 			 * No one can be adding dquots to the mplist at
 			 * this point, but somebody might be taking things off.
 			 */
-			if (nrecl != XFS_QI_MPLRECLAIMS(mp)) {
+			if (nrecl != q->qi_dqreclaims) {
 				mutex_unlock(&dqp->q_hash->qh_lock);
 				goto again;
 			}
@@ -617,11 +630,9 @@ xfs_qm_dqpurge_int(
 		 * Take the dquot off the mplist and hashlist. It may remain on
 		 * freelist in INACTIVE state.
 		 */
-		nextdqp = dqp->MPL_NEXT;
 		nmisses += xfs_qm_dqpurge(dqp);
-		dqp = nextdqp;
 	}
-	xfs_qm_mplist_unlock(mp);
+	mutex_unlock(&q->qi_dqlist_lock);
 	return nmisses;
 }
 
@@ -921,12 +932,13 @@ xfs_qm_dqdetach(
 
 int
 xfs_qm_sync(
-	xfs_mount_t	*mp,
-	int		flags)
+	struct xfs_mount	*mp,
+	int			flags)
 {
-	int		recl, restarts;
-	xfs_dquot_t	*dqp;
-	int		error;
+	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;
@@ -934,18 +946,19 @@ xfs_qm_sync(
 	restarts = 0;
 
   again:
-	xfs_qm_mplist_lock(mp);
+	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)) {
-		xfs_qm_mplist_unlock(mp);
+	if (!XFS_IS_QUOTA_ON(mp)) {
+		mutex_unlock(&q->qi_dqlist_lock);
 		return 0;
 	}
-	FOREACH_DQUOT_IN_MP(dqp, mp) {
+	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.
@@ -969,7 +982,7 @@ xfs_qm_sync(
 		}
 
 		/* XXX a sentinel would be better */
-		recl = XFS_QI_MPLRECLAIMS(mp);
+		recl = q->qi_dqreclaims;
 		if (!xfs_dqflock_nowait(dqp)) {
 			if (flags & SYNC_TRYLOCK) {
 				xfs_dqunlock(dqp);
@@ -989,7 +1002,7 @@ xfs_qm_sync(
 		 * Let go of the mplist lock. We don't want to hold it
 		 * across a disk write
 		 */
-		xfs_qm_mplist_unlock(mp);
+		mutex_unlock(&q->qi_dqlist_lock);
 		error = xfs_qm_dqflush(dqp, flags);
 		xfs_dqunlock(dqp);
 		if (error && XFS_FORCED_SHUTDOWN(mp))
@@ -997,17 +1010,17 @@ xfs_qm_sync(
 		else if (error)
 			return error;
 
-		xfs_qm_mplist_lock(mp);
-		if (recl != XFS_QI_MPLRECLAIMS(mp)) {
+		mutex_lock(&q->qi_dqlist_lock);
+		if (recl != q->qi_dqreclaims) {
 			if (++restarts >= XFS_QM_SYNC_MAX_RESTARTS)
 				break;
 
-			xfs_qm_mplist_unlock(mp);
+			mutex_unlock(&q->qi_dqlist_lock);
 			goto again;
 		}
 	}
 
-	xfs_qm_mplist_unlock(mp);
+	mutex_unlock(&q->qi_dqlist_lock);
 	return 0;
 }
 
@@ -1052,8 +1065,9 @@ xfs_qm_init_quotainfo(
 		return error;
 	}
 
-	xfs_qm_list_init(&qinf->qi_dqlist, "mpdqlist", 0);
-	lockdep_set_class(&qinf->qi_dqlist.qh_lock, &xfs_quota_mplist_class);
+	INIT_LIST_HEAD(&qinf->qi_dqlist);
+	mutex_init(&qinf->qi_dqlist_lock);
+	lockdep_set_class(&qinf->qi_dqlist_lock, &xfs_quota_mplist_class);
 
 	qinf->qi_dqreclaims = 0;
 
@@ -1150,7 +1164,8 @@ xfs_qm_destroy_quotainfo(
 	 */
 	xfs_qm_rele_quotafs_ref(mp);
 
-	xfs_qm_list_destroy(&qi->qi_dqlist);
+	ASSERT(list_empty(&qi->qi_dqlist));
+	mutex_destroy(&qi->qi_dqlist_lock);
 
 	if (qi->qi_uquotaip) {
 		IRELE(qi->qi_uquotaip);
@@ -1177,7 +1192,7 @@ xfs_qm_list_init(
 	int		n)
 {
 	mutex_init(&list->qh_lock);
-	list->qh_next = NULL;
+	INIT_LIST_HEAD(&list->qh_list);
 	list->qh_version = 0;
 	list->qh_nelems = 0;
 }
@@ -1316,9 +1331,6 @@ xfs_qm_qino_alloc(
 	 */
 	spin_lock(&mp->m_sb_lock);
 	if (flags & XFS_QMOPT_SBVERSION) {
-#if defined(DEBUG) && defined(XFS_LOUD_RECOVERY)
-		unsigned oldv = mp->m_sb.sb_versionnum;
-#endif
 		ASSERT(!xfs_sb_version_hasquota(&mp->m_sb));
 		ASSERT((sbfields & (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO |
 				   XFS_SB_GQUOTINO | XFS_SB_QFLAGS)) ==
@@ -1331,11 +1343,6 @@ xfs_qm_qino_alloc(
 
 		/* qflags will get updated _after_ quotacheck */
 		mp->m_sb.sb_qflags = 0;
-#if defined(DEBUG) && defined(XFS_LOUD_RECOVERY)
-		cmn_err(CE_NOTE,
-			"Old superblock version %x, converting to %x.",
-			oldv, mp->m_sb.sb_versionnum);
-#endif
 	}
 	if (flags & XFS_QMOPT_UQUOTA)
 		mp->m_sb.sb_uquotino = (*ip)->i_ino;
@@ -1371,10 +1378,10 @@ xfs_qm_reset_dqcounts(
 #ifdef DEBUG
 	j = XFS_FSB_TO_B(mp, XFS_DQUOT_CLUSTER_SIZE_FSB);
 	do_div(j, sizeof(xfs_dqblk_t));
-	ASSERT(XFS_QM_DQPERBLK(mp) == j);
+	ASSERT(mp->m_quotainfo->qi_dqperchunk == j);
 #endif
 	ddq = (xfs_disk_dquot_t *)XFS_BUF_PTR(bp);
-	for (j = 0; j < XFS_QM_DQPERBLK(mp); j++) {
+	for (j = 0; j < mp->m_quotainfo->qi_dqperchunk; j++) {
 		/*
 		 * Do a sanity check, and if needed, repair the dqblk. Don't
 		 * output any warnings because it's perfectly possible to
@@ -1429,7 +1436,7 @@ xfs_qm_dqiter_bufs(
 	while (blkcnt--) {
 		error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp,
 			      XFS_FSB_TO_DADDR(mp, bno),
-			      (int)XFS_QI_DQCHUNKLEN(mp), 0, &bp);
+			      mp->m_quotainfo->qi_dqchunklen, 0, &bp);
 		if (error)
 			break;
 
@@ -1439,7 +1446,7 @@ xfs_qm_dqiter_bufs(
 		 * goto the next block.
 		 */
 		bno++;
-		firstid += XFS_QM_DQPERBLK(mp);
+		firstid += mp->m_quotainfo->qi_dqperchunk;
 	}
 	return error;
 }
@@ -1505,7 +1512,7 @@ xfs_qm_dqiterate(
 				continue;
 
 			firstid = (xfs_dqid_t) map[i].br_startoff *
-				XFS_QM_DQPERBLK(mp);
+				mp->m_quotainfo->qi_dqperchunk;
 			/*
 			 * Do a read-ahead on the next extent.
 			 */
@@ -1516,7 +1523,7 @@ xfs_qm_dqiterate(
 				while (rablkcnt--) {
 					xfs_baread(mp->m_ddev_targp,
 					       XFS_FSB_TO_DADDR(mp, rablkno),
-					       (int)XFS_QI_DQCHUNKLEN(mp));
+					       mp->m_quotainfo->qi_dqchunklen);
 					rablkno++;
 				}
 			}
@@ -1576,8 +1583,10 @@ xfs_qm_quotacheck_dqadjust(
 
 	/*
 	 * Set default limits, adjust timers (since we changed usages)
+	 *
+	 * There are no timers for the default values set in the root dquot.
 	 */
-	if (! XFS_IS_SUSER_DQUOT(dqp)) {
+	if (dqp->q_core.d_id) {
 		xfs_qm_adjust_dqlimits(dqp->q_mount, &dqp->q_core);
 		xfs_qm_adjust_dqtimers(dqp->q_mount, &dqp->q_core);
 	}
@@ -1747,14 +1756,14 @@ xfs_qm_quotacheck(
 	lastino = 0;
 	flags = 0;
 
-	ASSERT(XFS_QI_UQIP(mp) || XFS_QI_GQIP(mp));
+	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(XFS_QI_MPLNDQUOTS(mp) == 0);
+	ASSERT(list_empty(&mp->m_quotainfo->qi_dqlist));
 
 	cmn_err(CE_NOTE, "XFS quotacheck %s: Please wait.", mp->m_fsname);
 
@@ -1763,15 +1772,19 @@ xfs_qm_quotacheck(
 	 * their counters to zero. We need a clean slate.
 	 * We don't log our changes till later.
 	 */
-	if ((uip = XFS_QI_UQIP(mp))) {
-		if ((error = xfs_qm_dqiterate(mp, uip, XFS_QMOPT_UQUOTA)))
+	uip = mp->m_quotainfo->qi_uquotaip;
+	if (uip) {
+		error = xfs_qm_dqiterate(mp, uip, XFS_QMOPT_UQUOTA);
+		if (error)
 			goto error_return;
 		flags |= XFS_UQUOTA_CHKD;
 	}
 
-	if ((gip = XFS_QI_GQIP(mp))) {
-		if ((error = xfs_qm_dqiterate(mp, gip, XFS_IS_GQUOTA_ON(mp) ?
-					XFS_QMOPT_GQUOTA : XFS_QMOPT_PQUOTA)))
+	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);
+		if (error)
 			goto error_return;
 		flags |= XFS_OQUOTA_CHKD;
 	}
@@ -1804,7 +1817,7 @@ xfs_qm_quotacheck(
 	 * at this point (because we intentionally didn't in dqget_noattach).
 	 */
 	if (error) {
-		xfs_qm_dqpurge_all(mp, XFS_QMOPT_QUOTALL | XFS_QMOPT_QUOTAOFF);
+		xfs_qm_dqpurge_all(mp, XFS_QMOPT_QUOTALL);
 		goto error_return;
 	}
 
@@ -1825,7 +1838,7 @@ xfs_qm_quotacheck(
 	mp->m_qflags &= ~(XFS_OQUOTA_CHKD | XFS_UQUOTA_CHKD);
 	mp->m_qflags |= flags;
 
-	XQM_LIST_PRINT(&(XFS_QI_MPL_LIST(mp)), MPL_NEXT, "++++ Mp list +++");
+	xfs_qm_dquot_list_print(mp);
 
  error_return:
 	if (error) {
@@ -1920,59 +1933,53 @@ xfs_qm_init_quotainos(
 		}
 	}
 
-	XFS_QI_UQIP(mp) = uip;
-	XFS_QI_GQIP(mp) = gip;
+	mp->m_quotainfo->qi_uquotaip = uip;
+	mp->m_quotainfo->qi_gquotaip = gip;
 
 	return 0;
 }
 
 
+
 /*
- * 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 ...
- * XXXsup merge this with qm_reclaim_one().
+ * Just pop the least recently used dquot off the freelist and
+ * recycle it. The returned dquot is locked.
  */
-STATIC int
-xfs_qm_shake_freelist(
-	int howmany)
+STATIC xfs_dquot_t *
+xfs_qm_dqreclaim_one(void)
 {
-	int		nreclaimed;
-	xfs_dqhash_t	*hash;
-	xfs_dquot_t	*dqp, *nextdqp;
+	xfs_dquot_t	*dqpout;
+	xfs_dquot_t	*dqp;
 	int		restarts;
-	int		nflushes;
-
-	if (howmany <= 0)
-		return 0;
 
-	nreclaimed = 0;
 	restarts = 0;
-	nflushes = 0;
+	dqpout = NULL;
 
-#ifdef QUOTADEBUG
-	cmn_err(CE_DEBUG, "Shake free 0x%x", howmany);
-#endif
-	/* lock order is : hashchainlock, freelistlock, mplistlock */
- tryagain:
-	xfs_qm_freelist_lock(xfs_Gqm);
+	/* lockorder: hashchainlock, freelistlock, mplistlock, dqlock, dqflock */
+startagain:
+	mutex_lock(&xfs_Gqm->qm_dqfrlist_lock);
 
-	for (dqp = xfs_Gqm->qm_dqfreelist.qh_next;
-	     ((dqp != (xfs_dquot_t *) &xfs_Gqm->qm_dqfreelist) &&
-	      nreclaimed < howmany); ) {
+	list_for_each_entry(dqp, &xfs_Gqm->qm_dqfrlist, q_freelist) {
+		struct xfs_mount *mp = dqp->q_mount;
 		xfs_dqlock(dqp);
 
 		/*
 		 * We are racing with dqlookup here. Naturally we don't
-		 * want to reclaim a dquot that lookup wants.
+		 * 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);
+
 			xfs_dqunlock(dqp);
-			xfs_qm_freelist_unlock(xfs_Gqm);
+			mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
 			if (++restarts >= XFS_QM_RECLAIM_MAX_RESTARTS)
-				return nreclaimed;
+				return NULL;
 			XQM_STATS_INC(xqmstats.xs_qm_dqwants);
-			goto tryagain;
+			goto startagain;
 		}
 
 		/*
@@ -1981,23 +1988,27 @@ xfs_qm_shake_freelist(
 		 * life easier.
 		 */
 		if (dqp->dq_flags & XFS_DQ_INACTIVE) {
-			ASSERT(dqp->q_mount == NULL);
+			ASSERT(mp == NULL);
 			ASSERT(! XFS_DQ_IS_DIRTY(dqp));
-			ASSERT(dqp->HL_PREVP == NULL);
-			ASSERT(dqp->MPL_PREVP == NULL);
+			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);
+			dqpout = dqp;
 			XQM_STATS_INC(xqmstats.xs_qm_dqinact_reclaims);
-			nextdqp = dqp->dq_flnext;
-			goto off_freelist;
+			break;
 		}
 
-		ASSERT(dqp->MPL_PREVP);
+		ASSERT(dqp->q_hash);
+		ASSERT(!list_empty(&dqp->q_mplist));
+
 		/*
 		 * 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)) {
 			xfs_dqunlock(dqp);
-			dqp = dqp->dq_flnext;
 			continue;
 		}
 
@@ -2010,21 +2021,21 @@ xfs_qm_shake_freelist(
 		if (XFS_DQ_IS_DIRTY(dqp)) {
 			int	error;
 
-			trace_xfs_dqshake_dirty(dqp);
+			trace_xfs_dqreclaim_dirty(dqp);
 
 			/*
 			 * We flush it delayed write, so don't bother
-			 * releasing the mplock.
+			 * releasing the freelist lock.
 			 */
 			error = xfs_qm_dqflush(dqp, 0);
 			if (error) {
-				xfs_fs_cmn_err(CE_WARN, dqp->q_mount,
-			"xfs_qm_dqflush_all: dquot %p flush failed", dqp);
+				xfs_fs_cmn_err(CE_WARN, mp,
+			"xfs_qm_dqreclaim: dquot %p flush failed", dqp);
 			}
 			xfs_dqunlock(dqp); /* dqflush unlocks dqflock */
-			dqp = dqp->dq_flnext;
 			continue;
 		}
+
 		/*
 		 * We're trying to get the hashlock out of order. This races
 		 * with dqlookup; so, we giveup and goto the next dquot if
@@ -2033,56 +2044,74 @@ xfs_qm_shake_freelist(
 		 * waiting for the freelist lock.
 		 */
 		if (!mutex_trylock(&dqp->q_hash->qh_lock)) {
-			xfs_dqfunlock(dqp);
-			xfs_dqunlock(dqp);
-			dqp = dqp->dq_flnext;
-			continue;
+			restarts++;
+			goto dqfunlock;
 		}
+
 		/*
 		 * 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.
 		 */
-		hash = dqp->q_hash;
-		ASSERT(hash);
-		if (! xfs_qm_mplist_nowait(dqp->q_mount)) {
-			/* XXX put a sentinel so that we can come back here */
+		if (!mutex_trylock(&mp->m_quotainfo->qi_dqlist_lock)) {
+			restarts++;
+			mutex_unlock(&dqp->q_hash->qh_lock);
 			xfs_dqfunlock(dqp);
 			xfs_dqunlock(dqp);
-			mutex_unlock(&hash->qh_lock);
-			xfs_qm_freelist_unlock(xfs_Gqm);
-			if (++restarts >= XFS_QM_RECLAIM_MAX_RESTARTS)
-				return nreclaimed;
-			goto tryagain;
+			mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
+			if (restarts++ >= XFS_QM_RECLAIM_MAX_RESTARTS)
+				return NULL;
+			goto startagain;
 		}
 
-		trace_xfs_dqshake_unlink(dqp);
-
-#ifdef QUOTADEBUG
-		cmn_err(CE_DEBUG, "Shake 0x%p, ID 0x%x\n",
-			dqp, be32_to_cpu(dqp->q_core.d_id));
-#endif
 		ASSERT(dqp->q_nrefs == 0);
-		nextdqp = dqp->dq_flnext;
-		XQM_MPLIST_REMOVE(&(XFS_QI_MPL_LIST(dqp->q_mount)), dqp);
-		XQM_HASHLIST_REMOVE(hash, dqp);
+		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);
+		mutex_unlock(&dqp->q_hash->qh_lock);
+dqfunlock:
 		xfs_dqfunlock(dqp);
-		xfs_qm_mplist_unlock(dqp->q_mount);
-		mutex_unlock(&hash->qh_lock);
-
- off_freelist:
-		XQM_FREELIST_REMOVE(dqp);
 		xfs_dqunlock(dqp);
-		nreclaimed++;
-		XQM_STATS_INC(xqmstats.xs_qm_dqshake_reclaims);
+		if (dqpout)
+			break;
+		if (restarts >= XFS_QM_RECLAIM_MAX_RESTARTS)
+			return NULL;
+	}
+	mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
+	return dqpout;
+}
+
+/*
+ * 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;
+
+	if (howmany <= 0)
+		return 0;
+
+	while (nreclaimed < howmany) {
+		dqp = xfs_qm_dqreclaim_one();
+		if (!dqp)
+			return nreclaimed;
 		xfs_qm_dqdestroy(dqp);
-		dqp = nextdqp;
+		nreclaimed++;
 	}
-	xfs_qm_freelist_unlock(xfs_Gqm);
 	return nreclaimed;
 }
 
-
 /*
  * The kmem_shake interface is invoked when memory is running low.
  */
@@ -2097,7 +2126,7 @@ xfs_qm_shake(int nr_to_scan, gfp_t gfp_mask)
 	if (!xfs_Gqm)
 		return 0;
 
-	nfree = xfs_Gqm->qm_dqfreelist.qh_nelems; /* free dquots */
+	nfree = xfs_Gqm->qm_dqfrlist_cnt; /* free dquots */
 	/* incore dquots in all f/s's */
 	ndqused = atomic_read(&xfs_Gqm->qm_totaldquots) - nfree;
 
@@ -2113,131 +2142,6 @@ xfs_qm_shake(int nr_to_scan, gfp_t gfp_mask)
 }
 
 
-/*
- * 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		nflushes;
-
-	restarts = 0;
-	dqpout = NULL;
-	nflushes = 0;
-
-	/* lockorder: hashchainlock, freelistlock, mplistlock, dqlock, dqflock */
- startagain:
-	xfs_qm_freelist_lock(xfs_Gqm);
-
-	FOREACH_DQUOT_IN_FREELIST(dqp, &(xfs_Gqm->qm_dqfreelist)) {
-		xfs_dqlock(dqp);
-
-		/*
-		 * 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);
-
-			xfs_dqunlock(dqp);
-			xfs_qm_freelist_unlock(xfs_Gqm);
-			if (++restarts >= XFS_QM_RECLAIM_MAX_RESTARTS)
-				return NULL;
-			XQM_STATS_INC(xqmstats.xs_qm_dqwants);
-			goto startagain;
-		}
-
-		/*
-		 * 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(dqp->q_mount == NULL);
-			ASSERT(! XFS_DQ_IS_DIRTY(dqp));
-			ASSERT(dqp->HL_PREVP == NULL);
-			ASSERT(dqp->MPL_PREVP == NULL);
-			XQM_FREELIST_REMOVE(dqp);
-			xfs_dqunlock(dqp);
-			dqpout = dqp;
-			XQM_STATS_INC(xqmstats.xs_qm_dqinact_reclaims);
-			break;
-		}
-
-		ASSERT(dqp->q_hash);
-		ASSERT(dqp->MPL_PREVP);
-
-		/*
-		 * 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)) {
-			xfs_dqunlock(dqp);
-			continue;
-		}
-
-		/*
-		 * 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;
-
-			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_fs_cmn_err(CE_WARN, dqp->q_mount,
-			"xfs_qm_dqreclaim: dquot %p flush failed", dqp);
-			}
-			xfs_dqunlock(dqp); /* dqflush unlocks dqflock */
-			continue;
-		}
-
-		if (! xfs_qm_mplist_nowait(dqp->q_mount)) {
-			xfs_dqfunlock(dqp);
-			xfs_dqunlock(dqp);
-			continue;
-		}
-
-		if (!mutex_trylock(&dqp->q_hash->qh_lock))
-			goto mplistunlock;
-
-		trace_xfs_dqreclaim_unlink(dqp);
-
-		ASSERT(dqp->q_nrefs == 0);
-		XQM_MPLIST_REMOVE(&(XFS_QI_MPL_LIST(dqp->q_mount)), dqp);
-		XQM_HASHLIST_REMOVE(dqp->q_hash, dqp);
-		XQM_FREELIST_REMOVE(dqp);
-		dqpout = dqp;
-		mutex_unlock(&dqp->q_hash->qh_lock);
- mplistunlock:
-		xfs_qm_mplist_unlock(dqp->q_mount);
-		xfs_dqfunlock(dqp);
-		xfs_dqunlock(dqp);
-		if (dqpout)
-			break;
-	}
-
-	xfs_qm_freelist_unlock(xfs_Gqm);
-	return dqpout;
-}
-
-
 /*------------------------------------------------------------------*/
 
 /*
@@ -2662,66 +2566,3 @@ xfs_qm_vop_create_dqattach(
 	}
 }
 
-/* ------------- list stuff -----------------*/
-STATIC void
-xfs_qm_freelist_init(xfs_frlist_t *ql)
-{
-	ql->qh_next = ql->qh_prev = (xfs_dquot_t *) ql;
-	mutex_init(&ql->qh_lock);
-	ql->qh_version = 0;
-	ql->qh_nelems = 0;
-}
-
-STATIC void
-xfs_qm_freelist_destroy(xfs_frlist_t *ql)
-{
-	xfs_dquot_t	*dqp, *nextdqp;
-
-	mutex_lock(&ql->qh_lock);
-	for (dqp = ql->qh_next;
-	     dqp != (xfs_dquot_t *)ql; ) {
-		xfs_dqlock(dqp);
-		nextdqp = dqp->dq_flnext;
-#ifdef QUOTADEBUG
-		cmn_err(CE_DEBUG, "FREELIST destroy 0x%p", dqp);
-#endif
-		XQM_FREELIST_REMOVE(dqp);
-		xfs_dqunlock(dqp);
-		xfs_qm_dqdestroy(dqp);
-		dqp = nextdqp;
-	}
-	mutex_unlock(&ql->qh_lock);
-	mutex_destroy(&ql->qh_lock);
-
-	ASSERT(ql->qh_nelems == 0);
-}
-
-STATIC void
-xfs_qm_freelist_insert(xfs_frlist_t *ql, xfs_dquot_t *dq)
-{
-	dq->dq_flnext = ql->qh_next;
-	dq->dq_flprev = (xfs_dquot_t *)ql;
-	ql->qh_next = dq;
-	dq->dq_flnext->dq_flprev = dq;
-	xfs_Gqm->qm_dqfreelist.qh_nelems++;
-	xfs_Gqm->qm_dqfreelist.qh_version++;
-}
-
-void
-xfs_qm_freelist_unlink(xfs_dquot_t *dq)
-{
-	xfs_dquot_t *next = dq->dq_flnext;
-	xfs_dquot_t *prev = dq->dq_flprev;
-
-	next->dq_flprev = prev;
-	prev->dq_flnext = next;
-	dq->dq_flnext = dq->dq_flprev = dq;
-	xfs_Gqm->qm_dqfreelist.qh_nelems--;
-	xfs_Gqm->qm_dqfreelist.qh_version++;
-}
-
-void
-xfs_qm_freelist_append(xfs_frlist_t *ql, xfs_dquot_t *dq)
-{
-	xfs_qm_freelist_insert((xfs_frlist_t *)ql->qh_prev, dq);
-}
diff --git a/fs/xfs/quota/xfs_qm.h b/fs/xfs/quota/xfs_qm.h
index 495564b8af3..c9446f1c726 100644
--- a/fs/xfs/quota/xfs_qm.h
+++ b/fs/xfs/quota/xfs_qm.h
@@ -72,17 +72,6 @@ extern kmem_zone_t	*qm_dqtrxzone;
 #define XFS_QM_MAX_DQCLUSTER_LOGSZ	3
 
 typedef xfs_dqhash_t	xfs_dqlist_t;
-/*
- * The freelist head. The first two fields match the first two in the
- * xfs_dquot_t structure (in xfs_dqmarker_t)
- */
-typedef struct xfs_frlist {
-       struct xfs_dquot *qh_next;
-       struct xfs_dquot *qh_prev;
-       struct mutex	 qh_lock;
-       uint		 qh_version;
-       uint		 qh_nelems;
-} xfs_frlist_t;
 
 /*
  * Quota Manager (global) structure. Lives only in core.
@@ -91,7 +80,9 @@ 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 */
-	xfs_frlist_t	 qm_dqfreelist;	 /* freelist of dquots */
+	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 */
@@ -106,7 +97,9 @@ typedef struct xfs_qm {
 typedef struct xfs_quotainfo {
 	xfs_inode_t	*qi_uquotaip;	 /* user quota inode */
 	xfs_inode_t	*qi_gquotaip;	 /* group quota inode */
-	xfs_dqlist_t	 qi_dqlist;	 /* all dquots in filesys */
+	struct list_head qi_dqlist;	 /* all dquots in filesys */
+	struct mutex	 qi_dqlist_lock;
+	int		 qi_dquots;
 	int		 qi_dqreclaims;	 /* a change here indicates
 					    a removal in the dqlist */
 	time_t		 qi_btimelimit;	 /* limit for blks timer */
@@ -175,10 +168,6 @@ extern int		xfs_qm_scall_getqstat(xfs_mount_t *, fs_quota_stat_t *);
 extern int		xfs_qm_scall_quotaon(xfs_mount_t *, uint);
 extern int		xfs_qm_scall_quotaoff(xfs_mount_t *, uint);
 
-/* list stuff */
-extern void		xfs_qm_freelist_append(xfs_frlist_t *, xfs_dquot_t *);
-extern void		xfs_qm_freelist_unlink(xfs_dquot_t *);
-
 #ifdef DEBUG
 extern int		xfs_qm_internalqcheck(xfs_mount_t *);
 #else
diff --git a/fs/xfs/quota/xfs_qm_stats.c b/fs/xfs/quota/xfs_qm_stats.c
index 83e7ea3e25f..3d1fc79532e 100644
--- a/fs/xfs/quota/xfs_qm_stats.c
+++ b/fs/xfs/quota/xfs_qm_stats.c
@@ -55,7 +55,7 @@ static int xqm_proc_show(struct seq_file *m, void *v)
 			ndquot,
 			xfs_Gqm? atomic_read(&xfs_Gqm->qm_totaldquots) : 0,
 			xfs_Gqm? xfs_Gqm->qm_dqfree_ratio : 0,
-			xfs_Gqm? xfs_Gqm->qm_dqfreelist.qh_nelems : 0);
+			xfs_Gqm? xfs_Gqm->qm_dqfrlist_cnt : 0);
 	return 0;
 }
 
diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c
index 5d0ee8d492d..92b002f1805 100644
--- a/fs/xfs/quota/xfs_qm_syscalls.c
+++ b/fs/xfs/quota/xfs_qm_syscalls.c
@@ -79,6 +79,7 @@ xfs_qm_scall_quotaoff(
 	xfs_mount_t		*mp,
 	uint			flags)
 {
+	struct xfs_quotainfo	*q = mp->m_quotainfo;
 	uint			dqtype;
 	int			error;
 	uint			inactivate_flags;
@@ -102,11 +103,8 @@ xfs_qm_scall_quotaoff(
 	 * critical thing.
 	 * If quotaoff, then we must be dealing with the root filesystem.
 	 */
-	ASSERT(mp->m_quotainfo);
-	if (mp->m_quotainfo)
-		mutex_lock(&(XFS_QI_QOFFLOCK(mp)));
-
-	ASSERT(mp->m_quotainfo);
+	ASSERT(q);
+	mutex_lock(&q->qi_quotaofflock);
 
 	/*
 	 * If we're just turning off quota enforcement, change mp and go.
@@ -117,7 +115,7 @@ xfs_qm_scall_quotaoff(
 		spin_lock(&mp->m_sb_lock);
 		mp->m_sb.sb_qflags = mp->m_qflags;
 		spin_unlock(&mp->m_sb_lock);
-		mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
+		mutex_unlock(&q->qi_quotaofflock);
 
 		/* XXX what to do if error ? Revert back to old vals incore ? */
 		error = xfs_qm_write_sb_changes(mp, XFS_SB_QFLAGS);
@@ -150,10 +148,8 @@ xfs_qm_scall_quotaoff(
 	 * Nothing to do?  Don't complain. This happens when we're just
 	 * turning off quota enforcement.
 	 */
-	if ((mp->m_qflags & flags) == 0) {
-		mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
-		return (0);
-	}
+	if ((mp->m_qflags & flags) == 0)
+		goto out_unlock;
 
 	/*
 	 * Write the LI_QUOTAOFF log record, and do SB changes atomically,
@@ -162,7 +158,7 @@ xfs_qm_scall_quotaoff(
 	 */
 	error = xfs_qm_log_quotaoff(mp, &qoffstart, flags);
 	if (error)
-		goto out_error;
+		goto out_unlock;
 
 	/*
 	 * Next we clear the XFS_MOUNT_*DQ_ACTIVE bit(s) in the mount struct
@@ -204,7 +200,7 @@ xfs_qm_scall_quotaoff(
 	 * So, if we couldn't purge all the dquots from the filesystem,
 	 * we can't get rid of the incore data structures.
 	 */
-	while ((nculprits = xfs_qm_dqpurge_all(mp, dqtype|XFS_QMOPT_QUOTAOFF)))
+	while ((nculprits = xfs_qm_dqpurge_all(mp, dqtype)))
 		delay(10 * nculprits);
 
 	/*
@@ -222,7 +218,7 @@ xfs_qm_scall_quotaoff(
 	if (error) {
 		/* We're screwed now. Shutdown is the only option. */
 		xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
-		goto out_error;
+		goto out_unlock;
 	}
 
 	/*
@@ -230,27 +226,26 @@ xfs_qm_scall_quotaoff(
 	 */
 	if (((flags & XFS_MOUNT_QUOTA_ALL) == XFS_MOUNT_QUOTA_SET1) ||
 	    ((flags & XFS_MOUNT_QUOTA_ALL) == XFS_MOUNT_QUOTA_SET2)) {
-		mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
+		mutex_unlock(&q->qi_quotaofflock);
 		xfs_qm_destroy_quotainfo(mp);
 		return (0);
 	}
 
 	/*
-	 * Release our quotainode references, and vn_purge them,
-	 * if we don't need them anymore.
+	 * Release our quotainode references if we don't need them anymore.
 	 */
-	if ((dqtype & XFS_QMOPT_UQUOTA) && XFS_QI_UQIP(mp)) {
-		IRELE(XFS_QI_UQIP(mp));
-		XFS_QI_UQIP(mp) = NULL;
+	if ((dqtype & XFS_QMOPT_UQUOTA) && q->qi_uquotaip) {
+		IRELE(q->qi_uquotaip);
+		q->qi_uquotaip = NULL;
 	}
-	if ((dqtype & (XFS_QMOPT_GQUOTA|XFS_QMOPT_PQUOTA)) && XFS_QI_GQIP(mp)) {
-		IRELE(XFS_QI_GQIP(mp));
-		XFS_QI_GQIP(mp) = NULL;
+	if ((dqtype & (XFS_QMOPT_GQUOTA|XFS_QMOPT_PQUOTA)) && q->qi_gquotaip) {
+		IRELE(q->qi_gquotaip);
+		q->qi_gquotaip = NULL;
 	}
-out_error:
-	mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
 
-	return (error);
+out_unlock:
+	mutex_unlock(&q->qi_quotaofflock);
+	return error;
 }
 
 int
@@ -379,9 +374,9 @@ xfs_qm_scall_quotaon(
 	/*
 	 * Switch on quota enforcement in core.
 	 */
-	mutex_lock(&(XFS_QI_QOFFLOCK(mp)));
+	mutex_lock(&mp->m_quotainfo->qi_quotaofflock);
 	mp->m_qflags |= (flags & XFS_ALL_QUOTA_ENFD);
-	mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
+	mutex_unlock(&mp->m_quotainfo->qi_quotaofflock);
 
 	return (0);
 }
@@ -392,11 +387,12 @@ xfs_qm_scall_quotaon(
  */
 int
 xfs_qm_scall_getqstat(
-	xfs_mount_t	*mp,
-	fs_quota_stat_t *out)
+	struct xfs_mount	*mp,
+	struct fs_quota_stat	*out)
 {
-	xfs_inode_t	*uip, *gip;
-	boolean_t	tempuqip, tempgqip;
+	struct xfs_quotainfo	*q = mp->m_quotainfo;
+	struct xfs_inode	*uip, *gip;
+	boolean_t		tempuqip, tempgqip;
 
 	uip = gip = NULL;
 	tempuqip = tempgqip = B_FALSE;
@@ -415,9 +411,9 @@ xfs_qm_scall_getqstat(
 	out->qs_uquota.qfs_ino = mp->m_sb.sb_uquotino;
 	out->qs_gquota.qfs_ino = mp->m_sb.sb_gquotino;
 
-	if (mp->m_quotainfo) {
-		uip = mp->m_quotainfo->qi_uquotaip;
-		gip = mp->m_quotainfo->qi_gquotaip;
+	if (q) {
+		uip = q->qi_uquotaip;
+		gip = q->qi_gquotaip;
 	}
 	if (!uip && mp->m_sb.sb_uquotino != NULLFSINO) {
 		if (xfs_iget(mp, NULL, mp->m_sb.sb_uquotino,
@@ -441,17 +437,20 @@ xfs_qm_scall_getqstat(
 		if (tempgqip)
 			IRELE(gip);
 	}
-	if (mp->m_quotainfo) {
-		out->qs_incoredqs = XFS_QI_MPLNDQUOTS(mp);
-		out->qs_btimelimit = XFS_QI_BTIMELIMIT(mp);
-		out->qs_itimelimit = XFS_QI_ITIMELIMIT(mp);
-		out->qs_rtbtimelimit = XFS_QI_RTBTIMELIMIT(mp);
-		out->qs_bwarnlimit = XFS_QI_BWARNLIMIT(mp);
-		out->qs_iwarnlimit = XFS_QI_IWARNLIMIT(mp);
+	if (q) {
+		out->qs_incoredqs = q->qi_dquots;
+		out->qs_btimelimit = q->qi_btimelimit;
+		out->qs_itimelimit = q->qi_itimelimit;
+		out->qs_rtbtimelimit = q->qi_rtbtimelimit;
+		out->qs_bwarnlimit = q->qi_bwarnlimit;
+		out->qs_iwarnlimit = q->qi_iwarnlimit;
 	}
-	return (0);
+	return 0;
 }
 
+#define XFS_DQ_MASK \
+	(FS_DQ_LIMIT_MASK | FS_DQ_TIMER_MASK | FS_DQ_WARNS_MASK)
+
 /*
  * Adjust quota limits, and start/stop timers accordingly.
  */
@@ -462,15 +461,17 @@ xfs_qm_scall_setqlim(
 	uint			type,
 	fs_disk_quota_t		*newlim)
 {
+	struct xfs_quotainfo	*q = mp->m_quotainfo;
 	xfs_disk_dquot_t	*ddq;
 	xfs_dquot_t		*dqp;
 	xfs_trans_t		*tp;
 	int			error;
 	xfs_qcnt_t		hard, soft;
 
-	if ((newlim->d_fieldmask &
-	    (FS_DQ_LIMIT_MASK|FS_DQ_TIMER_MASK|FS_DQ_WARNS_MASK)) == 0)
-		return (0);
+	if (newlim->d_fieldmask & ~XFS_DQ_MASK)
+		return EINVAL;
+	if ((newlim->d_fieldmask & XFS_DQ_MASK) == 0)
+		return 0;
 
 	tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SETQLIM);
 	if ((error = xfs_trans_reserve(tp, 0, sizeof(xfs_disk_dquot_t) + 128,
@@ -485,7 +486,7 @@ xfs_qm_scall_setqlim(
 	 * a quotaoff from happening). (XXXThis doesn't currently happen
 	 * because we take the vfslock before calling xfs_qm_sysent).
 	 */
-	mutex_lock(&(XFS_QI_QOFFLOCK(mp)));
+	mutex_lock(&q->qi_quotaofflock);
 
 	/*
 	 * Get the dquot (locked), and join it to the transaction.
@@ -493,9 +494,8 @@ xfs_qm_scall_setqlim(
 	 */
 	if ((error = xfs_qm_dqget(mp, NULL, id, type, XFS_QMOPT_DQALLOC, &dqp))) {
 		xfs_trans_cancel(tp, XFS_TRANS_ABORT);
-		mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
 		ASSERT(error != ENOENT);
-		return (error);
+		goto out_unlock;
 	}
 	xfs_trans_dqjoin(tp, dqp);
 	ddq = &dqp->q_core;
@@ -513,8 +513,8 @@ xfs_qm_scall_setqlim(
 		ddq->d_blk_hardlimit = cpu_to_be64(hard);
 		ddq->d_blk_softlimit = cpu_to_be64(soft);
 		if (id == 0) {
-			mp->m_quotainfo->qi_bhardlimit = hard;
-			mp->m_quotainfo->qi_bsoftlimit = soft;
+			q->qi_bhardlimit = hard;
+			q->qi_bsoftlimit = soft;
 		}
 	} else {
 		qdprintk("blkhard %Ld < blksoft %Ld\n", hard, soft);
@@ -529,8 +529,8 @@ xfs_qm_scall_setqlim(
 		ddq->d_rtb_hardlimit = cpu_to_be64(hard);
 		ddq->d_rtb_softlimit = cpu_to_be64(soft);
 		if (id == 0) {
-			mp->m_quotainfo->qi_rtbhardlimit = hard;
-			mp->m_quotainfo->qi_rtbsoftlimit = soft;
+			q->qi_rtbhardlimit = hard;
+			q->qi_rtbsoftlimit = soft;
 		}
 	} else {
 		qdprintk("rtbhard %Ld < rtbsoft %Ld\n", hard, soft);
@@ -546,8 +546,8 @@ xfs_qm_scall_setqlim(
 		ddq->d_ino_hardlimit = cpu_to_be64(hard);
 		ddq->d_ino_softlimit = cpu_to_be64(soft);
 		if (id == 0) {
-			mp->m_quotainfo->qi_ihardlimit = hard;
-			mp->m_quotainfo->qi_isoftlimit = soft;
+			q->qi_ihardlimit = hard;
+			q->qi_isoftlimit = soft;
 		}
 	} else {
 		qdprintk("ihard %Ld < isoft %Ld\n", hard, soft);
@@ -572,23 +572,23 @@ xfs_qm_scall_setqlim(
 		 * for warnings.
 		 */
 		if (newlim->d_fieldmask & FS_DQ_BTIMER) {
-			mp->m_quotainfo->qi_btimelimit = newlim->d_btimer;
+			q->qi_btimelimit = newlim->d_btimer;
 			ddq->d_btimer = cpu_to_be32(newlim->d_btimer);
 		}
 		if (newlim->d_fieldmask & FS_DQ_ITIMER) {
-			mp->m_quotainfo->qi_itimelimit = newlim->d_itimer;
+			q->qi_itimelimit = newlim->d_itimer;
 			ddq->d_itimer = cpu_to_be32(newlim->d_itimer);
 		}
 		if (newlim->d_fieldmask & FS_DQ_RTBTIMER) {
-			mp->m_quotainfo->qi_rtbtimelimit = newlim->d_rtbtimer;
+			q->qi_rtbtimelimit = newlim->d_rtbtimer;
 			ddq->d_rtbtimer = cpu_to_be32(newlim->d_rtbtimer);
 		}
 		if (newlim->d_fieldmask & FS_DQ_BWARNS)
-			mp->m_quotainfo->qi_bwarnlimit = newlim->d_bwarns;
+			q->qi_bwarnlimit = newlim->d_bwarns;
 		if (newlim->d_fieldmask & FS_DQ_IWARNS)
-			mp->m_quotainfo->qi_iwarnlimit = newlim->d_iwarns;
+			q->qi_iwarnlimit = newlim->d_iwarns;
 		if (newlim->d_fieldmask & FS_DQ_RTBWARNS)
-			mp->m_quotainfo->qi_rtbwarnlimit = newlim->d_rtbwarns;
+			q->qi_rtbwarnlimit = newlim->d_rtbwarns;
 	} else {
 		/*
 		 * If the user is now over quota, start the timelimit.
@@ -605,8 +605,9 @@ xfs_qm_scall_setqlim(
 	error = xfs_trans_commit(tp, 0);
 	xfs_qm_dqprint(dqp);
 	xfs_qm_dqrele(dqp);
-	mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
 
+ out_unlock:
+	mutex_unlock(&q->qi_quotaofflock);
 	return error;
 }
 
@@ -853,7 +854,8 @@ xfs_dqrele_inode(
 	int			error;
 
 	/* skip quota inodes */
-	if (ip == XFS_QI_UQIP(ip->i_mount) || ip == XFS_QI_GQIP(ip->i_mount)) {
+	if (ip == ip->i_mount->m_quotainfo->qi_uquotaip ||
+	    ip == ip->i_mount->m_quotainfo->qi_gquotaip) {
 		ASSERT(ip->i_udquot == NULL);
 		ASSERT(ip->i_gdquot == NULL);
 		read_unlock(&pag->pag_ici_lock);
@@ -891,7 +893,8 @@ xfs_qm_dqrele_all_inodes(
 	uint		 flags)
 {
 	ASSERT(mp->m_quotainfo);
-	xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags, XFS_ICI_NO_TAG, 0);
+	xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags,
+				XFS_ICI_NO_TAG, 0, NULL);
 }
 
 /*------------------------------------------------------------------------*/
@@ -930,7 +933,8 @@ struct mutex  qcheck_lock;
 }
 
 typedef struct dqtest {
-	xfs_dqmarker_t	q_lists;
+	uint		 dq_flags;	/* various flags (XFS_DQ_*) */
+	struct list_head q_hashlist;
 	xfs_dqhash_t	*q_hash;	/* the hashchain header */
 	xfs_mount_t	*q_mount;	/* filesystem this relates to */
 	xfs_dqid_t	d_id;		/* user id or group id */
@@ -941,14 +945,9 @@ typedef struct dqtest {
 STATIC void
 xfs_qm_hashinsert(xfs_dqhash_t *h, xfs_dqtest_t *dqp)
 {
-	xfs_dquot_t *d;
-	if (((d) = (h)->qh_next))
-		(d)->HL_PREVP = &((dqp)->HL_NEXT);
-	(dqp)->HL_NEXT = d;
-	(dqp)->HL_PREVP = &((h)->qh_next);
-	(h)->qh_next = (xfs_dquot_t *)dqp;
-	(h)->qh_version++;
-	(h)->qh_nelems++;
+	list_add(&dqp->q_hashlist, &h->qh_list);
+	h->qh_version++;
+	h->qh_nelems++;
 }
 STATIC void
 xfs_qm_dqtest_print(
@@ -1060,9 +1059,7 @@ xfs_qm_internalqcheck_dqget(
 	xfs_dqhash_t	*h;
 
 	h = DQTEST_HASH(mp, id, type);
-	for (d = (xfs_dqtest_t *) h->qh_next; d != NULL;
-	     d = (xfs_dqtest_t *) d->HL_NEXT) {
-		/* DQTEST_LIST_PRINT(h, HL_NEXT, "@@@@@ dqtestlist @@@@@"); */
+	list_for_each_entry(d, &h->qh_list, q_hashlist) {
 		if (d->d_id == id && mp == d->q_mount) {
 			*O_dq = d;
 			return (0);
@@ -1073,6 +1070,7 @@ xfs_qm_internalqcheck_dqget(
 	d->d_id = id;
 	d->q_mount = mp;
 	d->q_hash = h;
+	INIT_LIST_HEAD(&d->q_hashlist);
 	xfs_qm_hashinsert(h, d);
 	*O_dq = d;
 	return (0);
@@ -1179,8 +1177,6 @@ xfs_qm_internalqcheck(
 	xfs_ino_t	lastino;
 	int		done, count;
 	int		i;
-	xfs_dqtest_t	*d, *e;
-	xfs_dqhash_t	*h1;
 	int		error;
 
 	lastino = 0;
@@ -1220,19 +1216,18 @@ xfs_qm_internalqcheck(
 	}
 	cmn_err(CE_DEBUG, "Checking results against system dquots");
 	for (i = 0; i < qmtest_hashmask; i++) {
-		h1 = &qmtest_udqtab[i];
-		for (d = (xfs_dqtest_t *) h1->qh_next; d != NULL; ) {
+		xfs_dqtest_t	*d, *n;
+		xfs_dqhash_t	*h;
+
+		h = &qmtest_udqtab[i];
+		list_for_each_entry_safe(d, n, &h->qh_list, q_hashlist) {
 			xfs_dqtest_cmp(d);
-			e = (xfs_dqtest_t *) d->HL_NEXT;
 			kmem_free(d);
-			d = e;
 		}
-		h1 = &qmtest_gdqtab[i];
-		for (d = (xfs_dqtest_t *) h1->qh_next; d != NULL; ) {
+		h = &qmtest_gdqtab[i];
+		list_for_each_entry_safe(d, n, &h->qh_list, q_hashlist) {
 			xfs_dqtest_cmp(d);
-			e = (xfs_dqtest_t *) d->HL_NEXT;
 			kmem_free(d);
-			d = e;
 		}
 	}
 
diff --git a/fs/xfs/quota/xfs_quota_priv.h b/fs/xfs/quota/xfs_quota_priv.h
index 8286b2842b6..94a3d927d71 100644
--- a/fs/xfs/quota/xfs_quota_priv.h
+++ b/fs/xfs/quota/xfs_quota_priv.h
@@ -24,43 +24,6 @@
  */
 #define XFS_DQITER_MAP_SIZE	10
 
-/* Number of dquots that fit in to a dquot block */
-#define XFS_QM_DQPERBLK(mp)	((mp)->m_quotainfo->qi_dqperchunk)
-
-#define XFS_DQ_IS_ADDEDTO_TRX(t, d)	((d)->q_transp == (t))
-
-#define XFS_QI_MPLRECLAIMS(mp)	((mp)->m_quotainfo->qi_dqreclaims)
-#define XFS_QI_UQIP(mp)		((mp)->m_quotainfo->qi_uquotaip)
-#define XFS_QI_GQIP(mp)		((mp)->m_quotainfo->qi_gquotaip)
-#define XFS_QI_DQCHUNKLEN(mp)	((mp)->m_quotainfo->qi_dqchunklen)
-#define XFS_QI_BTIMELIMIT(mp)	((mp)->m_quotainfo->qi_btimelimit)
-#define XFS_QI_RTBTIMELIMIT(mp) ((mp)->m_quotainfo->qi_rtbtimelimit)
-#define XFS_QI_ITIMELIMIT(mp)	((mp)->m_quotainfo->qi_itimelimit)
-#define XFS_QI_BWARNLIMIT(mp)	((mp)->m_quotainfo->qi_bwarnlimit)
-#define XFS_QI_RTBWARNLIMIT(mp)	((mp)->m_quotainfo->qi_rtbwarnlimit)
-#define XFS_QI_IWARNLIMIT(mp)	((mp)->m_quotainfo->qi_iwarnlimit)
-#define XFS_QI_QOFFLOCK(mp)	((mp)->m_quotainfo->qi_quotaofflock)
-
-#define XFS_QI_MPL_LIST(mp)	((mp)->m_quotainfo->qi_dqlist)
-#define XFS_QI_MPLNEXT(mp)	((mp)->m_quotainfo->qi_dqlist.qh_next)
-#define XFS_QI_MPLNDQUOTS(mp)	((mp)->m_quotainfo->qi_dqlist.qh_nelems)
-
-#define xfs_qm_mplist_lock(mp) \
-	mutex_lock(&(XFS_QI_MPL_LIST(mp).qh_lock))
-#define xfs_qm_mplist_nowait(mp) \
-	mutex_trylock(&(XFS_QI_MPL_LIST(mp).qh_lock))
-#define xfs_qm_mplist_unlock(mp) \
-	mutex_unlock(&(XFS_QI_MPL_LIST(mp).qh_lock))
-#define XFS_QM_IS_MPLIST_LOCKED(mp) \
-	mutex_is_locked(&(XFS_QI_MPL_LIST(mp).qh_lock))
-
-#define xfs_qm_freelist_lock(qm) \
-	mutex_lock(&((qm)->qm_dqfreelist.qh_lock))
-#define xfs_qm_freelist_lock_nowait(qm) \
-	mutex_trylock(&((qm)->qm_dqfreelist.qh_lock))
-#define xfs_qm_freelist_unlock(qm) \
-	mutex_unlock(&((qm)->qm_dqfreelist.qh_lock))
-
 /*
  * Hash into a bucket in the dquot hash table, based on <mp, id>.
  */
@@ -72,9 +35,6 @@
 				      XFS_DQ_HASHVAL(mp, id)) : \
 				     (xfs_Gqm->qm_grp_dqhtable + \
 				      XFS_DQ_HASHVAL(mp, id)))
-#define XFS_IS_DQTYPE_ON(mp, type)   (type == XFS_DQ_USER ? \
-					XFS_IS_UQUOTA_ON(mp) : \
-					XFS_IS_OQUOTA_ON(mp))
 #define XFS_IS_DQUOT_UNINITIALIZED(dqp) ( \
 	!dqp->q_core.d_blk_hardlimit && \
 	!dqp->q_core.d_blk_softlimit && \
@@ -86,68 +46,6 @@
 	!dqp->q_core.d_rtbcount && \
 	!dqp->q_core.d_icount)
 
-#define HL_PREVP	dq_hashlist.ql_prevp
-#define HL_NEXT		dq_hashlist.ql_next
-#define MPL_PREVP	dq_mplist.ql_prevp
-#define MPL_NEXT	dq_mplist.ql_next
-
-
-#define _LIST_REMOVE(h, dqp, PVP, NXT)				\
-	{							\
-		 xfs_dquot_t *d;				\
-		 if (((d) = (dqp)->NXT))				\
-			 (d)->PVP = (dqp)->PVP;			\
-		 *((dqp)->PVP) = d;				\
-		 (dqp)->NXT = NULL;				\
-		 (dqp)->PVP = NULL;				\
-		 (h)->qh_version++;				\
-		 (h)->qh_nelems--;				\
-	}
-
-#define _LIST_INSERT(h, dqp, PVP, NXT)				\
-	{							\
-		 xfs_dquot_t *d;				\
-		 if (((d) = (h)->qh_next))			\
-			 (d)->PVP = &((dqp)->NXT);		\
-		 (dqp)->NXT = d;				\
-		 (dqp)->PVP = &((h)->qh_next);			\
-		 (h)->qh_next = dqp;				\
-		 (h)->qh_version++;				\
-		 (h)->qh_nelems++;				\
-	 }
-
-#define FOREACH_DQUOT_IN_MP(dqp, mp) \
-	for ((dqp) = XFS_QI_MPLNEXT(mp); (dqp) != NULL; (dqp) = (dqp)->MPL_NEXT)
-
-#define FOREACH_DQUOT_IN_FREELIST(dqp, qlist)	\
-for ((dqp) = (qlist)->qh_next; (dqp) != (xfs_dquot_t *)(qlist); \
-     (dqp) = (dqp)->dq_flnext)
-
-#define XQM_HASHLIST_INSERT(h, dqp)	\
-	 _LIST_INSERT(h, dqp, HL_PREVP, HL_NEXT)
-
-#define XQM_FREELIST_INSERT(h, dqp)	\
-	 xfs_qm_freelist_append(h, dqp)
-
-#define XQM_MPLIST_INSERT(h, dqp)	\
-	 _LIST_INSERT(h, dqp, MPL_PREVP, MPL_NEXT)
-
-#define XQM_HASHLIST_REMOVE(h, dqp)	\
-	 _LIST_REMOVE(h, dqp, HL_PREVP, HL_NEXT)
-#define XQM_FREELIST_REMOVE(dqp)	\
-	 xfs_qm_freelist_unlink(dqp)
-#define XQM_MPLIST_REMOVE(h, dqp)	\
-	{ _LIST_REMOVE(h, dqp, MPL_PREVP, MPL_NEXT); \
-	  XFS_QI_MPLRECLAIMS((dqp)->q_mount)++; }
-
-#define XFS_DQ_IS_LOGITEM_INITD(dqp)	((dqp)->q_logitem.qli_dquot == (dqp))
-
-#define XFS_QM_DQP_TO_DQACCT(tp, dqp)	(XFS_QM_ISUDQ(dqp) ? \
-					 (tp)->t_dqinfo->dqa_usrdquots : \
-					 (tp)->t_dqinfo->dqa_grpdquots)
-#define XFS_IS_SUSER_DQUOT(dqp)		\
-	(!((dqp)->q_core.d_id))
-
 #define DQFLAGTO_TYPESTR(d)	(((d)->dq_flags & XFS_DQ_USER) ? "USR" : \
 				 (((d)->dq_flags & XFS_DQ_GROUP) ? "GRP" : \
 				 (((d)->dq_flags & XFS_DQ_PROJ) ? "PRJ":"???")))
diff --git a/fs/xfs/quota/xfs_trans_dquot.c b/fs/xfs/quota/xfs_trans_dquot.c
index c3ab75cb1d9..061d827da33 100644
--- a/fs/xfs/quota/xfs_trans_dquot.c
+++ b/fs/xfs/quota/xfs_trans_dquot.c
@@ -59,12 +59,11 @@ xfs_trans_dqjoin(
 	xfs_trans_t	*tp,
 	xfs_dquot_t	*dqp)
 {
-	xfs_dq_logitem_t    *lp;
+	xfs_dq_logitem_t    *lp = &dqp->q_logitem;
 
-	ASSERT(! XFS_DQ_IS_ADDEDTO_TRX(tp, dqp));
+	ASSERT(dqp->q_transp != tp);
 	ASSERT(XFS_DQ_IS_LOCKED(dqp));
-	ASSERT(XFS_DQ_IS_LOGITEM_INITD(dqp));
-	lp = &dqp->q_logitem;
+	ASSERT(lp->qli_dquot == dqp);
 
 	/*
 	 * Get a log_item_desc to point at the new item.
@@ -96,7 +95,7 @@ xfs_trans_log_dquot(
 {
 	xfs_log_item_desc_t	*lidp;
 
-	ASSERT(XFS_DQ_IS_ADDEDTO_TRX(tp, dqp));
+	ASSERT(dqp->q_transp == tp);
 	ASSERT(XFS_DQ_IS_LOCKED(dqp));
 
 	lidp = xfs_trans_find_item(tp, (xfs_log_item_t*)(&dqp->q_logitem));
@@ -198,16 +197,16 @@ xfs_trans_get_dqtrx(
 	int		i;
 	xfs_dqtrx_t	*qa;
 
-	for (i = 0; i < XFS_QM_TRANS_MAXDQS; i++) {
-		qa = XFS_QM_DQP_TO_DQACCT(tp, dqp);
+	qa = XFS_QM_ISUDQ(dqp) ?
+		tp->t_dqinfo->dqa_usrdquots : tp->t_dqinfo->dqa_grpdquots;
 
+	for (i = 0; i < XFS_QM_TRANS_MAXDQS; i++) {
 		if (qa[i].qt_dquot == NULL ||
-		    qa[i].qt_dquot == dqp) {
-			return (&qa[i]);
-		}
+		    qa[i].qt_dquot == dqp)
+			return &qa[i];
 	}
 
-	return (NULL);
+	return NULL;
 }
 
 /*
@@ -381,7 +380,7 @@ xfs_trans_apply_dquot_deltas(
 				break;
 
 			ASSERT(XFS_DQ_IS_LOCKED(dqp));
-			ASSERT(XFS_DQ_IS_ADDEDTO_TRX(tp, dqp));
+			ASSERT(dqp->q_transp == tp);
 
 			/*
 			 * adjust the actual number of blocks used
@@ -639,7 +638,7 @@ xfs_trans_dqresv(
 			softlimit = q->qi_bsoftlimit;
 		timer = be32_to_cpu(dqp->q_core.d_btimer);
 		warns = be16_to_cpu(dqp->q_core.d_bwarns);
-		warnlimit = XFS_QI_BWARNLIMIT(dqp->q_mount);
+		warnlimit = dqp->q_mount->m_quotainfo->qi_bwarnlimit;
 		resbcountp = &dqp->q_res_bcount;
 	} else {
 		ASSERT(flags & XFS_TRANS_DQ_RES_RTBLKS);
@@ -651,7 +650,7 @@ xfs_trans_dqresv(
 			softlimit = q->qi_rtbsoftlimit;
 		timer = be32_to_cpu(dqp->q_core.d_rtbtimer);
 		warns = be16_to_cpu(dqp->q_core.d_rtbwarns);
-		warnlimit = XFS_QI_RTBWARNLIMIT(dqp->q_mount);
+		warnlimit = dqp->q_mount->m_quotainfo->qi_rtbwarnlimit;
 		resbcountp = &dqp->q_res_rtbcount;
 	}
 
@@ -691,7 +690,7 @@ xfs_trans_dqresv(
 			count = be64_to_cpu(dqp->q_core.d_icount);
 			timer = be32_to_cpu(dqp->q_core.d_itimer);
 			warns = be16_to_cpu(dqp->q_core.d_iwarns);
-			warnlimit = XFS_QI_IWARNLIMIT(dqp->q_mount);
+			warnlimit = dqp->q_mount->m_quotainfo->qi_iwarnlimit;
 			hardlimit = be64_to_cpu(dqp->q_core.d_ino_hardlimit);
 			if (!hardlimit)
 				hardlimit = q->qi_ihardlimit;
diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h
index b1a5a1ff88e..abb8222b88c 100644
--- a/fs/xfs/xfs_ag.h
+++ b/fs/xfs/xfs_ag.h
@@ -223,6 +223,7 @@ typedef struct xfs_perag {
 	int		pag_ici_init;	/* incore inode cache initialised */
 	rwlock_t	pag_ici_lock;	/* incore inode lock */
 	struct radix_tree_root pag_ici_root;	/* incore inode cache root */
+	int		pag_ici_reclaimable;	/* reclaimable inodes */
 #endif
 	int		pagb_count;	/* pagb slots in use */
 	xfs_perag_busy_t pagb_list[XFS_PAGB_NUM_SLOTS];	/* unstable blocks */
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c
index 5c11e4d1701..99587ded043 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/xfs_bmap.c
@@ -3829,7 +3829,7 @@ xfs_bmap_add_attrfork(
 	}
 	if ((error = xfs_bmap_finish(&tp, &flist, &committed)))
 		goto error2;
-	error = xfs_trans_commit(tp, XFS_TRANS_PERM_LOG_RES);
+	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;
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index f3c49e69eab..240340a4727 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -372,12 +372,12 @@ xfs_buf_item_pin(
  */
 STATIC void
 xfs_buf_item_unpin(
-	xfs_buf_log_item_t	*bip,
-	int			stale)
+	xfs_buf_log_item_t	*bip)
 {
 	struct xfs_ail	*ailp;
 	xfs_buf_t	*bp;
 	int		freed;
+	int		stale = bip->bli_flags & XFS_BLI_STALE;
 
 	bp = bip->bli_buf;
 	ASSERT(bp != NULL);
@@ -428,40 +428,34 @@ xfs_buf_item_unpin_remove(
 	xfs_buf_log_item_t	*bip,
 	xfs_trans_t		*tp)
 {
-	xfs_buf_t		*bp;
-	xfs_log_item_desc_t	*lidp;
-	int			stale = 0;
-
-	bp = bip->bli_buf;
-	/*
-	 * will xfs_buf_item_unpin() call xfs_buf_item_relse()?
-	 */
+	/* will xfs_buf_item_unpin() call xfs_buf_item_relse()? */
 	if ((atomic_read(&bip->bli_refcount) == 1) &&
 	    (bip->bli_flags & XFS_BLI_STALE)) {
+		/*
+		 * yes -- We can safely do some work here and then call
+		 * buf_item_unpin to do the rest because we are
+		 * are holding the buffer locked so no one else will be
+		 * able to bump up the refcount. We have to remove the
+		 * log item from the transaction as we are about to release
+		 * our reference to the buffer. If we don't, the unlock that
+		 * occurs later in the xfs_trans_uncommit() will try to
+		 * reference the buffer which we no longer have a hold on.
+		 */
+		struct xfs_log_item_desc *lidp;
+
 		ASSERT(XFS_BUF_VALUSEMA(bip->bli_buf) <= 0);
 		trace_xfs_buf_item_unpin_stale(bip);
 
-		/*
-		 * yes -- clear the xaction descriptor in-use flag
-		 * and free the chunk if required.  We can safely
-		 * do some work here and then call buf_item_unpin
-		 * to do the rest because if the if is true, then
-		 * we are holding the buffer locked so no one else
-		 * will be able to bump up the refcount.
-		 */
-		lidp = xfs_trans_find_item(tp, (xfs_log_item_t *) bip);
-		stale = lidp->lid_flags & XFS_LID_BUF_STALE;
+		lidp = xfs_trans_find_item(tp, (xfs_log_item_t *)bip);
 		xfs_trans_free_item(tp, lidp);
+
 		/*
-		 * Since the transaction no longer refers to the buffer,
-		 * the buffer should no longer refer to the transaction.
+		 * Since the transaction no longer refers to the buffer, the
+		 * buffer should no longer refer to the transaction.
 		 */
-		XFS_BUF_SET_FSPRIVATE2(bp, NULL);
+		XFS_BUF_SET_FSPRIVATE2(bip->bli_buf, NULL);
 	}
-
-	xfs_buf_item_unpin(bip, stale);
-
-	return;
+	xfs_buf_item_unpin(bip);
 }
 
 /*
@@ -675,7 +669,7 @@ static struct xfs_item_ops xfs_buf_item_ops = {
 	.iop_format	= (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
 					xfs_buf_item_format,
 	.iop_pin	= (void(*)(xfs_log_item_t*))xfs_buf_item_pin,
-	.iop_unpin	= (void(*)(xfs_log_item_t*, int))xfs_buf_item_unpin,
+	.iop_unpin	= (void(*)(xfs_log_item_t*))xfs_buf_item_unpin,
 	.iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t *))
 					xfs_buf_item_unpin_remove,
 	.iop_trylock	= (uint(*)(xfs_log_item_t*))xfs_buf_item_trylock,
@@ -733,10 +727,7 @@ xfs_buf_item_init(
 
 	bip = (xfs_buf_log_item_t*)kmem_zone_zalloc(xfs_buf_item_zone,
 						    KM_SLEEP);
-	bip->bli_item.li_type = XFS_LI_BUF;
-	bip->bli_item.li_ops = &xfs_buf_item_ops;
-	bip->bli_item.li_mountp = mp;
-	bip->bli_item.li_ailp = mp->m_ail;
+	xfs_log_item_init(mp, &bip->bli_item, XFS_LI_BUF, &xfs_buf_item_ops);
 	bip->bli_buf = bp;
 	xfs_buf_hold(bp);
 	bip->bli_format.blf_type = XFS_LI_BUF;
diff --git a/fs/xfs/xfs_buf_item.h b/fs/xfs/xfs_buf_item.h
index 217f34af00c..df4454511f7 100644
--- a/fs/xfs/xfs_buf_item.h
+++ b/fs/xfs/xfs_buf_item.h
@@ -26,7 +26,7 @@ extern kmem_zone_t	*xfs_buf_item_zone;
  * have been logged.
  * For 6.2 and beyond, this is XFS_LI_BUF.  We use this to log everything.
  */
-typedef struct xfs_buf_log_format_t {
+typedef struct xfs_buf_log_format {
 	unsigned short	blf_type;	/* buf log item type indicator */
 	unsigned short	blf_size;	/* size of this item */
 	ushort		blf_flags;	/* misc state */
diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c
index 92d5cd5bf4f..ef96175c074 100644
--- a/fs/xfs/xfs_error.c
+++ b/fs/xfs/xfs_error.c
@@ -186,18 +186,18 @@ xfs_cmn_err(int panic_tag, int level, xfs_mount_t *mp, char *fmt, ...)
 
 void
 xfs_error_report(
-	char		*tag,
-	int		level,
-	xfs_mount_t	*mp,
-	char		*fname,
-	int		linenum,
-	inst_t		*ra)
+	const char		*tag,
+	int			level,
+	struct xfs_mount	*mp,
+	const char		*filename,
+	int			linenum,
+	inst_t			*ra)
 {
 	if (level <= xfs_error_level) {
 		xfs_cmn_err(XFS_PTAG_ERROR_REPORT,
 			    CE_ALERT, mp,
 		"XFS internal error %s at line %d of file %s.  Caller 0x%p\n",
-			    tag, linenum, fname, ra);
+			    tag, linenum, filename, ra);
 
 		xfs_stack_trace();
 	}
@@ -205,15 +205,15 @@ xfs_error_report(
 
 void
 xfs_corruption_error(
-	char		*tag,
-	int		level,
-	xfs_mount_t	*mp,
-	void		*p,
-	char		*fname,
-	int		linenum,
-	inst_t		*ra)
+	const char		*tag,
+	int			level,
+	struct xfs_mount	*mp,
+	void			*p,
+	const char		*filename,
+	int			linenum,
+	inst_t			*ra)
 {
 	if (level <= xfs_error_level)
 		xfs_hex_dump(p, 16);
-	xfs_error_report(tag, level, mp, fname, linenum, ra);
+	xfs_error_report(tag, level, mp, filename, linenum, ra);
 }
diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h
index 0c93051c465..c2c1a072bb8 100644
--- a/fs/xfs/xfs_error.h
+++ b/fs/xfs/xfs_error.h
@@ -29,10 +29,11 @@ extern int	xfs_error_trap(int);
 
 struct xfs_mount;
 
-extern void xfs_error_report(char *tag, int level, struct xfs_mount *mp,
-				char *fname, int linenum, inst_t *ra);
-extern void xfs_corruption_error(char *tag, int level, struct xfs_mount *mp,
-				void *p, char *fname, int linenum, inst_t *ra);
+extern void xfs_error_report(const char *tag, int level, struct xfs_mount *mp,
+			const char *filename, int linenum, inst_t *ra);
+extern void xfs_corruption_error(const char *tag, int level,
+			struct xfs_mount *mp, void *p, const char *filename,
+			int linenum, inst_t *ra);
 
 #define	XFS_ERROR_REPORT(e, lvl, mp)	\
 	xfs_error_report(e, lvl, mp, __FILE__, __LINE__, __return_address)
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index 6f35ed1b39b..409fe81585f 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -106,7 +106,7 @@ xfs_efi_item_pin(xfs_efi_log_item_t *efip)
  */
 /*ARGSUSED*/
 STATIC void
-xfs_efi_item_unpin(xfs_efi_log_item_t *efip, int stale)
+xfs_efi_item_unpin(xfs_efi_log_item_t *efip)
 {
 	struct xfs_ail		*ailp = efip->efi_item.li_ailp;
 
@@ -224,7 +224,7 @@ static struct xfs_item_ops xfs_efi_item_ops = {
 	.iop_format	= (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
 					xfs_efi_item_format,
 	.iop_pin	= (void(*)(xfs_log_item_t*))xfs_efi_item_pin,
-	.iop_unpin	= (void(*)(xfs_log_item_t*, int))xfs_efi_item_unpin,
+	.iop_unpin	= (void(*)(xfs_log_item_t*))xfs_efi_item_unpin,
 	.iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t *))
 					xfs_efi_item_unpin_remove,
 	.iop_trylock	= (uint(*)(xfs_log_item_t*))xfs_efi_item_trylock,
@@ -259,10 +259,7 @@ xfs_efi_init(xfs_mount_t	*mp,
 							     KM_SLEEP);
 	}
 
-	efip->efi_item.li_type = XFS_LI_EFI;
-	efip->efi_item.li_ops = &xfs_efi_item_ops;
-	efip->efi_item.li_mountp = mp;
-	efip->efi_item.li_ailp = mp->m_ail;
+	xfs_log_item_init(mp, &efip->efi_item, XFS_LI_EFI, &xfs_efi_item_ops);
 	efip->efi_format.efi_nextents = nextents;
 	efip->efi_format.efi_id = (__psint_t)(void*)efip;
 
@@ -428,7 +425,7 @@ xfs_efd_item_pin(xfs_efd_log_item_t *efdp)
  */
 /*ARGSUSED*/
 STATIC void
-xfs_efd_item_unpin(xfs_efd_log_item_t *efdp, int stale)
+xfs_efd_item_unpin(xfs_efd_log_item_t *efdp)
 {
 	return;
 }
@@ -518,7 +515,7 @@ static struct xfs_item_ops xfs_efd_item_ops = {
 	.iop_format	= (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
 					xfs_efd_item_format,
 	.iop_pin	= (void(*)(xfs_log_item_t*))xfs_efd_item_pin,
-	.iop_unpin	= (void(*)(xfs_log_item_t*, int))xfs_efd_item_unpin,
+	.iop_unpin	= (void(*)(xfs_log_item_t*))xfs_efd_item_unpin,
 	.iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t*))
 					xfs_efd_item_unpin_remove,
 	.iop_trylock	= (uint(*)(xfs_log_item_t*))xfs_efd_item_trylock,
@@ -554,10 +551,7 @@ xfs_efd_init(xfs_mount_t	*mp,
 							     KM_SLEEP);
 	}
 
-	efdp->efd_item.li_type = XFS_LI_EFD;
-	efdp->efd_item.li_ops = &xfs_efd_item_ops;
-	efdp->efd_item.li_mountp = mp;
-	efdp->efd_item.li_ailp = mp->m_ail;
+	xfs_log_item_init(mp, &efdp->efd_item, XFS_LI_EFD, &xfs_efd_item_ops);
 	efdp->efd_efip = efip;
 	efdp->efd_format.efd_nextents = nextents;
 	efdp->efd_format.efd_efi_id = efip->efi_format.efi_id;
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 0ffd5644704..8cd6e8d8fe9 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -2449,6 +2449,8 @@ xfs_iunpin_nowait(
 {
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
 
+	trace_xfs_inode_unpin_nowait(ip, _RET_IP_);
+
 	/* Give the log a push to start the unpinning I/O */
 	xfs_log_force_lsn(ip->i_mount, ip->i_itemp->ili_last_lsn, 0);
 
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index 7bfea854015..cf8249a6000 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -543,6 +543,7 @@ xfs_inode_item_pin(
 {
 	ASSERT(xfs_isilocked(iip->ili_inode, XFS_ILOCK_EXCL));
 
+	trace_xfs_inode_pin(iip->ili_inode, _RET_IP_);
 	atomic_inc(&iip->ili_inode->i_pincount);
 }
 
@@ -556,11 +557,11 @@ xfs_inode_item_pin(
 /* ARGSUSED */
 STATIC void
 xfs_inode_item_unpin(
-	xfs_inode_log_item_t	*iip,
-	int			stale)
+	xfs_inode_log_item_t	*iip)
 {
 	struct xfs_inode	*ip = iip->ili_inode;
 
+	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);
@@ -572,7 +573,7 @@ xfs_inode_item_unpin_remove(
 	xfs_inode_log_item_t	*iip,
 	xfs_trans_t		*tp)
 {
-	xfs_inode_item_unpin(iip, 0);
+	xfs_inode_item_unpin(iip);
 }
 
 /*
@@ -838,7 +839,7 @@ static struct xfs_item_ops xfs_inode_item_ops = {
 	.iop_format	= (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
 					xfs_inode_item_format,
 	.iop_pin	= (void(*)(xfs_log_item_t*))xfs_inode_item_pin,
-	.iop_unpin	= (void(*)(xfs_log_item_t*, int))xfs_inode_item_unpin,
+	.iop_unpin	= (void(*)(xfs_log_item_t*))xfs_inode_item_unpin,
 	.iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t*))
 					xfs_inode_item_unpin_remove,
 	.iop_trylock	= (uint(*)(xfs_log_item_t*))xfs_inode_item_trylock,
@@ -865,17 +866,9 @@ xfs_inode_item_init(
 	ASSERT(ip->i_itemp == NULL);
 	iip = ip->i_itemp = kmem_zone_zalloc(xfs_ili_zone, KM_SLEEP);
 
-	iip->ili_item.li_type = XFS_LI_INODE;
-	iip->ili_item.li_ops = &xfs_inode_item_ops;
-	iip->ili_item.li_mountp = mp;
-	iip->ili_item.li_ailp = mp->m_ail;
 	iip->ili_inode = ip;
-
-	/*
-	   We have zeroed memory. No need ...
-	   iip->ili_extents_buf = NULL;
-	 */
-
+	xfs_log_item_init(mp, &iip->ili_item, XFS_LI_INODE,
+						&xfs_inode_item_ops);
 	iip->ili_format.ilf_type = XFS_LI_INODE;
 	iip->ili_format.ilf_ino = ip->i_ino;
 	iip->ili_format.ilf_blkno = ip->i_imap.im_blkno;
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 0b65039951a..ef14943829d 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -55,71 +55,33 @@
 #define XFS_STRAT_WRITE_IMAPS	2
 #define XFS_WRITE_IMAPS		XFS_BMAP_MAX_NMAP
 
-STATIC int
-xfs_imap_to_bmap(
-	xfs_inode_t	*ip,
-	xfs_off_t	offset,
-	xfs_bmbt_irec_t *imap,
-	xfs_iomap_t	*iomapp,
-	int		imaps,			/* Number of imap entries */
-	int		iomaps,			/* Number of iomap entries */
-	int		flags)
-{
-	xfs_mount_t	*mp = ip->i_mount;
-	int		pbm;
-	xfs_fsblock_t	start_block;
-
-
-	for (pbm = 0; imaps && pbm < iomaps; imaps--, iomapp++, imap++, pbm++) {
-		iomapp->iomap_offset = XFS_FSB_TO_B(mp, imap->br_startoff);
-		iomapp->iomap_delta = offset - iomapp->iomap_offset;
-		iomapp->iomap_bsize = XFS_FSB_TO_B(mp, imap->br_blockcount);
-		iomapp->iomap_flags = flags;
-
-		if (XFS_IS_REALTIME_INODE(ip)) {
-			iomapp->iomap_flags |= IOMAP_REALTIME;
-			iomapp->iomap_target = mp->m_rtdev_targp;
-		} else {
-			iomapp->iomap_target = mp->m_ddev_targp;
-		}
-		start_block = imap->br_startblock;
-		if (start_block == HOLESTARTBLOCK) {
-			iomapp->iomap_bn = IOMAP_DADDR_NULL;
-			iomapp->iomap_flags |= IOMAP_HOLE;
-		} else if (start_block == DELAYSTARTBLOCK) {
-			iomapp->iomap_bn = IOMAP_DADDR_NULL;
-			iomapp->iomap_flags |= IOMAP_DELAY;
-		} else {
-			iomapp->iomap_bn = xfs_fsb_to_db(ip, start_block);
-			if (ISUNWRITTEN(imap))
-				iomapp->iomap_flags |= IOMAP_UNWRITTEN;
-		}
-
-		offset += iomapp->iomap_bsize - iomapp->iomap_delta;
-	}
-	return pbm;	/* Return the number filled */
-}
+STATIC int xfs_iomap_write_direct(struct xfs_inode *, xfs_off_t, size_t,
+				  int, struct xfs_bmbt_irec *, int *);
+STATIC int xfs_iomap_write_delay(struct xfs_inode *, xfs_off_t, size_t, int,
+				 struct xfs_bmbt_irec *, int *);
+STATIC int xfs_iomap_write_allocate(struct xfs_inode *, xfs_off_t, size_t,
+				struct xfs_bmbt_irec *, int *);
 
 int
 xfs_iomap(
-	xfs_inode_t	*ip,
-	xfs_off_t	offset,
-	ssize_t		count,
-	int		flags,
-	xfs_iomap_t	*iomapp,
-	int		*niomaps)
+	struct xfs_inode	*ip,
+	xfs_off_t		offset,
+	ssize_t			count,
+	int			flags,
+	struct xfs_bmbt_irec	*imap,
+	int			*nimaps,
+	int			*new)
 {
-	xfs_mount_t	*mp = ip->i_mount;
-	xfs_fileoff_t	offset_fsb, end_fsb;
-	int		error = 0;
-	int		lockmode = 0;
-	xfs_bmbt_irec_t	imap;
-	int		nimaps = 1;
-	int		bmapi_flags = 0;
-	int		iomap_flags = 0;
+	struct xfs_mount	*mp = ip->i_mount;
+	xfs_fileoff_t		offset_fsb, end_fsb;
+	int			error = 0;
+	int			lockmode = 0;
+	int			bmapi_flags = 0;
 
 	ASSERT((ip->i_d.di_mode & S_IFMT) == S_IFREG);
 
+	*new = 0;
+
 	if (XFS_FORCED_SHUTDOWN(mp))
 		return XFS_ERROR(EIO);
 
@@ -160,8 +122,8 @@ xfs_iomap(
 
 	error = xfs_bmapi(NULL, ip, offset_fsb,
 			(xfs_filblks_t)(end_fsb - offset_fsb),
-			bmapi_flags,  NULL, 0, &imap,
-			&nimaps, NULL, NULL);
+			bmapi_flags,  NULL, 0, imap,
+			nimaps, NULL, NULL);
 
 	if (error)
 		goto out;
@@ -169,46 +131,41 @@ xfs_iomap(
 	switch (flags & (BMAPI_WRITE|BMAPI_ALLOCATE)) {
 	case BMAPI_WRITE:
 		/* If we found an extent, return it */
-		if (nimaps &&
-		    (imap.br_startblock != HOLESTARTBLOCK) &&
-		    (imap.br_startblock != DELAYSTARTBLOCK)) {
-			trace_xfs_iomap_found(ip, offset, count, flags, &imap);
+		if (*nimaps &&
+		    (imap->br_startblock != HOLESTARTBLOCK) &&
+		    (imap->br_startblock != DELAYSTARTBLOCK)) {
+			trace_xfs_iomap_found(ip, offset, count, flags, imap);
 			break;
 		}
 
 		if (flags & (BMAPI_DIRECT|BMAPI_MMAP)) {
 			error = xfs_iomap_write_direct(ip, offset, count, flags,
-						       &imap, &nimaps, nimaps);
+						       imap, nimaps);
 		} else {
 			error = xfs_iomap_write_delay(ip, offset, count, flags,
-						      &imap, &nimaps);
+						      imap, nimaps);
 		}
 		if (!error) {
-			trace_xfs_iomap_alloc(ip, offset, count, flags, &imap);
+			trace_xfs_iomap_alloc(ip, offset, count, flags, imap);
 		}
-		iomap_flags = IOMAP_NEW;
+		*new = 1;
 		break;
 	case BMAPI_ALLOCATE:
 		/* If we found an extent, return it */
 		xfs_iunlock(ip, lockmode);
 		lockmode = 0;
 
-		if (nimaps && !isnullstartblock(imap.br_startblock)) {
-			trace_xfs_iomap_found(ip, offset, count, flags, &imap);
+		if (*nimaps && !isnullstartblock(imap->br_startblock)) {
+			trace_xfs_iomap_found(ip, offset, count, flags, imap);
 			break;
 		}
 
 		error = xfs_iomap_write_allocate(ip, offset, count,
-						 &imap, &nimaps);
+						 imap, nimaps);
 		break;
 	}
 
-	if (nimaps) {
-		*niomaps = xfs_imap_to_bmap(ip, offset, &imap,
-					    iomapp, nimaps, *niomaps, iomap_flags);
-	} else if (niomaps) {
-		*niomaps = 0;
-	}
+	ASSERT(*nimaps <= 1);
 
 out:
 	if (lockmode)
@@ -216,7 +173,6 @@ out:
 	return XFS_ERROR(error);
 }
 
-
 STATIC int
 xfs_iomap_eof_align_last_fsb(
 	xfs_mount_t	*mp,
@@ -285,15 +241,14 @@ xfs_cmn_err_fsblock_zero(
 	return EFSCORRUPTED;
 }
 
-int
+STATIC int
 xfs_iomap_write_direct(
 	xfs_inode_t	*ip,
 	xfs_off_t	offset,
 	size_t		count,
 	int		flags,
 	xfs_bmbt_irec_t *ret_imap,
-	int		*nmaps,
-	int		found)
+	int		*nmaps)
 {
 	xfs_mount_t	*mp = ip->i_mount;
 	xfs_fileoff_t	offset_fsb;
@@ -330,7 +285,7 @@ xfs_iomap_write_direct(
 		if (error)
 			goto error_out;
 	} else {
-		if (found && (ret_imap->br_startblock == HOLESTARTBLOCK))
+		if (*nmaps && (ret_imap->br_startblock == HOLESTARTBLOCK))
 			last_fsb = MIN(last_fsb, (xfs_fileoff_t)
 					ret_imap->br_blockcount +
 					ret_imap->br_startoff);
@@ -485,7 +440,7 @@ xfs_iomap_eof_want_preallocate(
 	return 0;
 }
 
-int
+STATIC int
 xfs_iomap_write_delay(
 	xfs_inode_t	*ip,
 	xfs_off_t	offset,
@@ -588,7 +543,7 @@ retry:
  * We no longer bother to look at the incoming map - all we have to
  * guarantee is that whatever we allocate fills the required range.
  */
-int
+STATIC int
 xfs_iomap_write_allocate(
 	xfs_inode_t	*ip,
 	xfs_off_t	offset,
diff --git a/fs/xfs/xfs_iomap.h b/fs/xfs/xfs_iomap.h
index 174f2999099..81ac4afd45b 100644
--- a/fs/xfs/xfs_iomap.h
+++ b/fs/xfs/xfs_iomap.h
@@ -18,19 +18,6 @@
 #ifndef __XFS_IOMAP_H__
 #define __XFS_IOMAP_H__
 
-#define IOMAP_DADDR_NULL ((xfs_daddr_t) (-1LL))
-
-
-typedef enum {				/* iomap_flags values */
-	IOMAP_READ =		0,	/* mapping for a read */
-	IOMAP_HOLE =		0x02,	/* mapping covers a hole  */
-	IOMAP_DELAY =		0x04,	/* mapping covers delalloc region  */
-	IOMAP_REALTIME =	0x10,	/* mapping on the realtime device  */
-	IOMAP_UNWRITTEN =	0x20,	/* mapping covers allocated */
-					/* but uninitialized file data  */
-	IOMAP_NEW =		0x40	/* just allocate */
-} iomap_flags_t;
-
 typedef enum {
 	/* base extent manipulation calls */
 	BMAPI_READ = (1 << 0),		/* read extents */
@@ -52,43 +39,11 @@ typedef enum {
 	{ BMAPI_MMAP,		"MMAP" }, \
 	{ BMAPI_TRYLOCK,	"TRYLOCK" }
 
-/*
- * xfs_iomap_t:  File system I/O map
- *
- * The iomap_bn field is expressed in 512-byte blocks, and is where the
- * mapping starts on disk.
- *
- * The iomap_offset, iomap_bsize and iomap_delta fields are in bytes.
- * iomap_offset is the offset of the mapping in the file itself.
- * iomap_bsize is the size of the mapping,  iomap_delta is the
- * desired data's offset into the mapping, given the offset supplied
- * to the file I/O map routine.
- *
- * When a request is made to read beyond the logical end of the object,
- * iomap_size may be set to 0, but iomap_offset and iomap_length should be set
- * to the actual amount of underlying storage that has been allocated, if any.
- */
-
-typedef struct xfs_iomap {
-	xfs_daddr_t		iomap_bn;	/* first 512B blk of mapping */
-	xfs_buftarg_t		*iomap_target;
-	xfs_off_t		iomap_offset;	/* offset of mapping, bytes */
-	xfs_off_t		iomap_bsize;	/* size of mapping, bytes */
-	xfs_off_t		iomap_delta;	/* offset into mapping, bytes */
-	iomap_flags_t		iomap_flags;
-} xfs_iomap_t;
-
 struct xfs_inode;
 struct xfs_bmbt_irec;
 
 extern int xfs_iomap(struct xfs_inode *, xfs_off_t, ssize_t, int,
-		     struct xfs_iomap *, int *);
-extern int xfs_iomap_write_direct(struct xfs_inode *, xfs_off_t, size_t,
-				  int, struct xfs_bmbt_irec *, int *, int);
-extern int xfs_iomap_write_delay(struct xfs_inode *, xfs_off_t, size_t, int,
-				 struct xfs_bmbt_irec *, int *);
-extern int xfs_iomap_write_allocate(struct xfs_inode *, xfs_off_t, size_t,
-				struct xfs_bmbt_irec *, int *);
+		     struct xfs_bmbt_irec *, int *, int *);
 extern int xfs_iomap_write_unwritten(struct xfs_inode *, xfs_off_t, size_t);
 
 #endif /* __XFS_IOMAP_H__*/
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 2be01913628..3038dd52c72 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -44,13 +44,8 @@
 
 kmem_zone_t	*xfs_log_ticket_zone;
 
-#define xlog_write_adv_cnt(ptr, len, off, bytes) \
-	{ (ptr) += (bytes); \
-	  (len) -= (bytes); \
-	  (off) += (bytes);}
-
 /* Local miscellaneous function prototypes */
-STATIC int	 xlog_commit_record(xfs_mount_t *mp, xlog_ticket_t *ticket,
+STATIC int	 xlog_commit_record(struct log *log, struct xlog_ticket *ticket,
 				    xlog_in_core_t **, xfs_lsn_t *);
 STATIC xlog_t *  xlog_alloc_log(xfs_mount_t	*mp,
 				xfs_buftarg_t	*log_target,
@@ -59,11 +54,9 @@ STATIC xlog_t *  xlog_alloc_log(xfs_mount_t	*mp,
 STATIC int	 xlog_space_left(xlog_t *log, int cycle, int bytes);
 STATIC int	 xlog_sync(xlog_t *log, xlog_in_core_t *iclog);
 STATIC void	 xlog_dealloc_log(xlog_t *log);
-STATIC int	 xlog_write(xfs_mount_t *mp, xfs_log_iovec_t region[],
-			    int nentries, struct xlog_ticket *tic,
-			    xfs_lsn_t *start_lsn,
-			    xlog_in_core_t **commit_iclog,
-			    uint flags);
+STATIC 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);
 
 /* local state machine functions */
 STATIC void xlog_state_done_syncing(xlog_in_core_t *iclog, int);
@@ -102,7 +95,7 @@ STATIC xlog_ticket_t	*xlog_ticket_alloc(xlog_t *log,
 					 uint	flags);
 
 #if defined(DEBUG)
-STATIC void	xlog_verify_dest_ptr(xlog_t *log, __psint_t ptr);
+STATIC void	xlog_verify_dest_ptr(xlog_t *log, char *ptr);
 STATIC void	xlog_verify_grant_head(xlog_t *log, int equals);
 STATIC void	xlog_verify_iclog(xlog_t *log, xlog_in_core_t *iclog,
 				  int count, boolean_t syncing);
@@ -258,7 +251,7 @@ xfs_log_done(
 	     * If we get an error, just continue and give back the log ticket.
 	     */
 	    (((ticket->t_flags & XLOG_TIC_INITED) == 0) &&
-	     (xlog_commit_record(mp, ticket, iclog, &lsn)))) {
+	     (xlog_commit_record(log, ticket, iclog, &lsn)))) {
 		lsn = (xfs_lsn_t) -1;
 		if (ticket->t_flags & XLOG_TIC_PERM_RESERV) {
 			flags |= XFS_LOG_REL_PERM_RESERV;
@@ -516,18 +509,10 @@ xfs_log_unmount_write(xfs_mount_t *mp)
 #ifdef DEBUG
 	xlog_in_core_t	 *first_iclog;
 #endif
-	xfs_log_iovec_t  reg[1];
 	xlog_ticket_t	*tic = NULL;
 	xfs_lsn_t	 lsn;
 	int		 error;
 
-	/* the data section must be 32 bit size aligned */
-	struct {
-	    __uint16_t magic;
-	    __uint16_t pad1;
-	    __uint32_t pad2; /* may as well make it 64 bits */
-	} magic = { XLOG_UNMOUNT_TYPE, 0, 0 };
-
 	/*
 	 * Don't write out unmount record on read-only mounts.
 	 * Or, if we are doing a forced umount (typically because of IO errors).
@@ -549,16 +534,30 @@ xfs_log_unmount_write(xfs_mount_t *mp)
 	} while (iclog != first_iclog);
 #endif
 	if (! (XLOG_FORCED_SHUTDOWN(log))) {
-		reg[0].i_addr = (void*)&magic;
-		reg[0].i_len  = sizeof(magic);
-		reg[0].i_type = XLOG_REG_TYPE_UNMOUNT;
-
 		error = xfs_log_reserve(mp, 600, 1, &tic,
 					XFS_LOG, 0, XLOG_UNMOUNT_REC_TYPE);
 		if (!error) {
+			/* the data section must be 32 bit size aligned */
+			struct {
+			    __uint16_t magic;
+			    __uint16_t pad1;
+			    __uint32_t pad2; /* may as well make it 64 bits */
+			} magic = {
+				.magic = XLOG_UNMOUNT_TYPE,
+			};
+			struct xfs_log_iovec reg = {
+				.i_addr = (void *)&magic,
+				.i_len = sizeof(magic),
+				.i_type = XLOG_REG_TYPE_UNMOUNT,
+			};
+			struct xfs_log_vec vec = {
+				.lv_niovecs = 1,
+				.lv_iovecp = &reg,
+			};
+
 			/* remove inited flag */
-			((xlog_ticket_t *)tic)->t_flags = 0;
-			error = xlog_write(mp, reg, 1, tic, &lsn,
+			tic->t_flags = 0;
+			error = xlog_write(log, &vec, tic, &lsn,
 					   NULL, XLOG_UNMOUNT_TRANS);
 			/*
 			 * At this point, we're umounting anyway,
@@ -648,10 +647,26 @@ xfs_log_unmount(xfs_mount_t *mp)
 	xlog_dealloc_log(mp->m_log);
 }
 
+void
+xfs_log_item_init(
+	struct xfs_mount	*mp,
+	struct xfs_log_item	*item,
+	int			type,
+	struct xfs_item_ops	*ops)
+{
+	item->li_mountp = mp;
+	item->li_ailp = mp->m_ail;
+	item->li_type = type;
+	item->li_ops = ops;
+}
+
 /*
  * 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().
+ * 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....
  */
 int
 xfs_log_write(
@@ -663,11 +678,15 @@ xfs_log_write(
 {
 	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(mp, reg, nentries, tic, start_lsn, NULL, 0);
+	error = xlog_write(log, &vec, tic, start_lsn, NULL, 0);
 	if (error)
 		xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
 	return error;
@@ -1020,6 +1039,7 @@ xlog_alloc_log(xfs_mount_t	*mp,
 	int			i;
 	int			iclogsize;
 	int			error = ENOMEM;
+	uint			log2_size = 0;
 
 	log = kmem_zalloc(sizeof(xlog_t), KM_MAYFAIL);
 	if (!log) {
@@ -1045,29 +1065,30 @@ xlog_alloc_log(xfs_mount_t	*mp,
 
 	error = EFSCORRUPTED;
 	if (xfs_sb_version_hassector(&mp->m_sb)) {
-		log->l_sectbb_log = mp->m_sb.sb_logsectlog - BBSHIFT;
-		if (log->l_sectbb_log < 0 ||
-		    log->l_sectbb_log > mp->m_sectbb_log) {
-			xlog_warn("XFS: Log sector size (0x%x) out of range.",
-						log->l_sectbb_log);
+	        log2_size = mp->m_sb.sb_logsectlog;
+		if (log2_size < BBSHIFT) {
+			xlog_warn("XFS: Log sector size too small "
+				"(0x%x < 0x%x)", log2_size, BBSHIFT);
 			goto out_free_log;
 		}
 
-		/* for larger sector sizes, must have v2 or external log */
-		if (log->l_sectbb_log != 0 &&
-		    (log->l_logBBstart != 0 &&
-		     !xfs_sb_version_haslogv2(&mp->m_sb))) {
-			xlog_warn("XFS: log sector size (0x%x) invalid "
-				  "for configuration.", log->l_sectbb_log);
+	        log2_size -= BBSHIFT;
+		if (log2_size > mp->m_sectbb_log) {
+			xlog_warn("XFS: Log sector size too large "
+				"(0x%x > 0x%x)", log2_size, mp->m_sectbb_log);
 			goto out_free_log;
 		}
-		if (mp->m_sb.sb_logsectlog < BBSHIFT) {
-			xlog_warn("XFS: Log sector log (0x%x) too small.",
-						mp->m_sb.sb_logsectlog);
+
+		/* for larger sector sizes, must have v2 or external log */
+		if (log2_size && log->l_logBBstart > 0 &&
+			    !xfs_sb_version_haslogv2(&mp->m_sb)) {
+
+			xlog_warn("XFS: log sector size (0x%x) invalid "
+				  "for configuration.", log2_size);
 			goto out_free_log;
 		}
 	}
-	log->l_sectbb_mask = (1 << log->l_sectbb_log) - 1;
+	log->l_sectBBsize = 1 << log2_size;
 
 	xlog_get_iclog_buffer_size(mp, log);
 
@@ -1174,26 +1195,31 @@ out:
  * ticket.  Return the lsn of the commit record.
  */
 STATIC int
-xlog_commit_record(xfs_mount_t  *mp,
-		   xlog_ticket_t *ticket,
-		   xlog_in_core_t **iclog,
-		   xfs_lsn_t	*commitlsnp)
+xlog_commit_record(
+	struct log		*log,
+	struct xlog_ticket	*ticket,
+	struct xlog_in_core	**iclog,
+	xfs_lsn_t		*commitlsnp)
 {
-	int		error;
-	xfs_log_iovec_t	reg[1];
-
-	reg[0].i_addr = NULL;
-	reg[0].i_len = 0;
-	reg[0].i_type = XLOG_REG_TYPE_COMMIT;
+	struct xfs_mount *mp = log->l_mp;
+	int	error;
+	struct xfs_log_iovec reg = {
+		.i_addr = NULL,
+		.i_len = 0,
+		.i_type = XLOG_REG_TYPE_COMMIT,
+	};
+	struct xfs_log_vec vec = {
+		.lv_niovecs = 1,
+		.lv_iovecp = &reg,
+	};
 
 	ASSERT_ALWAYS(iclog);
-	if ((error = xlog_write(mp, reg, 1, ticket, commitlsnp,
-			       iclog, XLOG_COMMIT_TRANS))) {
+	error = xlog_write(log, &vec, ticket, commitlsnp, iclog,
+					XLOG_COMMIT_TRANS);
+	if (error)
 		xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
-	}
 	return error;
-}	/* xlog_commit_record */
-
+}
 
 /*
  * Push on the buffer cache code if we ever use more than 75% of the on-disk
@@ -1614,6 +1640,192 @@ xlog_print_tic_res(xfs_mount_t *mp, xlog_ticket_t *ticket)
 }
 
 /*
+ * Calculate the potential space needed by the log vector.  Each region gets
+ * its own xlog_op_header_t and may need to be double word aligned.
+ */
+static int
+xlog_write_calc_vec_length(
+	struct xlog_ticket	*ticket,
+	struct xfs_log_vec	*log_vector)
+{
+	struct xfs_log_vec	*lv;
+	int			headers = 0;
+	int			len = 0;
+	int			i;
+
+	/* acct for start rec of xact */
+	if (ticket->t_flags & XLOG_TIC_INITED)
+		headers++;
+
+	for (lv = log_vector; lv; lv = lv->lv_next) {
+		headers += lv->lv_niovecs;
+
+		for (i = 0; i < lv->lv_niovecs; i++) {
+			struct xfs_log_iovec	*vecp = &lv->lv_iovecp[i];
+
+			len += vecp->i_len;
+			xlog_tic_add_region(ticket, vecp->i_len, vecp->i_type);
+		}
+	}
+
+	ticket->t_res_num_ophdrs += headers;
+	len += headers * sizeof(struct xlog_op_header);
+
+	return len;
+}
+
+/*
+ * If first write for transaction, insert start record  We can't be trying to
+ * commit if we are inited.  We can't have any "partial_copy" if we are inited.
+ */
+static int
+xlog_write_start_rec(
+	struct xlog_op_header	*ophdr,
+	struct xlog_ticket	*ticket)
+{
+	if (!(ticket->t_flags & XLOG_TIC_INITED))
+		return 0;
+
+	ophdr->oh_tid	= cpu_to_be32(ticket->t_tid);
+	ophdr->oh_clientid = ticket->t_clientid;
+	ophdr->oh_len = 0;
+	ophdr->oh_flags = XLOG_START_TRANS;
+	ophdr->oh_res2 = 0;
+
+	ticket->t_flags &= ~XLOG_TIC_INITED;
+
+	return sizeof(struct xlog_op_header);
+}
+
+static xlog_op_header_t *
+xlog_write_setup_ophdr(
+	struct log		*log,
+	struct xlog_op_header	*ophdr,
+	struct xlog_ticket	*ticket,
+	uint			flags)
+{
+	ophdr->oh_tid = cpu_to_be32(ticket->t_tid);
+	ophdr->oh_clientid = ticket->t_clientid;
+	ophdr->oh_res2 = 0;
+
+	/* are we copying a commit or unmount record? */
+	ophdr->oh_flags = flags;
+
+	/*
+	 * We've seen logs corrupted with bad transaction client ids.  This
+	 * makes sure that XFS doesn't generate them on.  Turn this into an EIO
+	 * and shut down the filesystem.
+	 */
+	switch (ophdr->oh_clientid)  {
+	case XFS_TRANSACTION:
+	case XFS_VOLUME:
+	case XFS_LOG:
+		break;
+	default:
+		xfs_fs_cmn_err(CE_WARN, log->l_mp,
+			"Bad XFS transaction clientid 0x%x in ticket 0x%p",
+			ophdr->oh_clientid, ticket);
+		return NULL;
+	}
+
+	return ophdr;
+}
+
+/*
+ * Set up the parameters of the region copy into the log. This has
+ * to handle region write split across multiple log buffers - this
+ * state is kept external to this function so that this code can
+ * can be written in an obvious, self documenting manner.
+ */
+static int
+xlog_write_setup_copy(
+	struct xlog_ticket	*ticket,
+	struct xlog_op_header	*ophdr,
+	int			space_available,
+	int			space_required,
+	int			*copy_off,
+	int			*copy_len,
+	int			*last_was_partial_copy,
+	int			*bytes_consumed)
+{
+	int			still_to_copy;
+
+	still_to_copy = space_required - *bytes_consumed;
+	*copy_off = *bytes_consumed;
+
+	if (still_to_copy <= space_available) {
+		/* write of region completes here */
+		*copy_len = still_to_copy;
+		ophdr->oh_len = cpu_to_be32(*copy_len);
+		if (*last_was_partial_copy)
+			ophdr->oh_flags |= (XLOG_END_TRANS|XLOG_WAS_CONT_TRANS);
+		*last_was_partial_copy = 0;
+		*bytes_consumed = 0;
+		return 0;
+	}
+
+	/* partial write of region, needs extra log op header reservation */
+	*copy_len = space_available;
+	ophdr->oh_len = cpu_to_be32(*copy_len);
+	ophdr->oh_flags |= XLOG_CONTINUE_TRANS;
+	if (*last_was_partial_copy)
+		ophdr->oh_flags |= XLOG_WAS_CONT_TRANS;
+	*bytes_consumed += *copy_len;
+	(*last_was_partial_copy)++;
+
+	/* account for new log op header */
+	ticket->t_curr_res -= sizeof(struct xlog_op_header);
+	ticket->t_res_num_ophdrs++;
+
+	return sizeof(struct xlog_op_header);
+}
+
+static int
+xlog_write_copy_finish(
+	struct log		*log,
+	struct xlog_in_core	*iclog,
+	uint			flags,
+	int			*record_cnt,
+	int			*data_cnt,
+	int			*partial_copy,
+	int			*partial_copy_len,
+	int			log_offset,
+	struct xlog_in_core	**commit_iclog)
+{
+	if (*partial_copy) {
+		/*
+		 * This iclog has already been marked WANT_SYNC by
+		 * xlog_state_get_iclog_space.
+		 */
+		xlog_state_finish_copy(log, iclog, *record_cnt, *data_cnt);
+		*record_cnt = 0;
+		*data_cnt = 0;
+		return xlog_state_release_iclog(log, iclog);
+	}
+
+	*partial_copy = 0;
+	*partial_copy_len = 0;
+
+	if (iclog->ic_size - log_offset <= sizeof(xlog_op_header_t)) {
+		/* no more space in this iclog - push it. */
+		xlog_state_finish_copy(log, iclog, *record_cnt, *data_cnt);
+		*record_cnt = 0;
+		*data_cnt = 0;
+
+		spin_lock(&log->l_icloglock);
+		xlog_state_want_sync(log, iclog);
+		spin_unlock(&log->l_icloglock);
+
+		if (!commit_iclog)
+			return xlog_state_release_iclog(log, iclog);
+		ASSERT(flags & XLOG_COMMIT_TRANS);
+		*commit_iclog = iclog;
+	}
+
+	return 0;
+}
+
+/*
  * Write some region out to in-core log
  *
  * This will be called when writing externally provided regions or when
@@ -1655,209 +1867,157 @@ xlog_print_tic_res(xfs_mount_t *mp, xlog_ticket_t *ticket)
  */
 STATIC int
 xlog_write(
-	struct xfs_mount	*mp,
-	struct xfs_log_iovec	reg[],
-	int			nentries,
+	struct log		*log,
+	struct xfs_log_vec	*log_vector,
 	struct xlog_ticket	*ticket,
 	xfs_lsn_t		*start_lsn,
 	struct xlog_in_core	**commit_iclog,
 	uint			flags)
 {
-    xlog_t	     *log = mp->m_log;
-    xlog_in_core_t   *iclog = NULL;  /* ptr to current in-core log */
-    xlog_op_header_t *logop_head;    /* ptr to log operation header */
-    __psint_t	     ptr;	     /* copy address into data region */
-    int		     len;	     /* # xlog_write() bytes 2 still copy */
-    int		     index;	     /* region index currently copying */
-    int		     log_offset;     /* offset (from 0) into data region */
-    int		     start_rec_copy; /* # bytes to copy for start record */
-    int		     partial_copy;   /* did we split a region? */
-    int		     partial_copy_len;/* # bytes copied if split region */
-    int		     need_copy;	     /* # bytes need to memcpy this region */
-    int		     copy_len;	     /* # bytes actually memcpy'ing */
-    int		     copy_off;	     /* # bytes from entry start */
-    int		     contwr;	     /* continued write of in-core log? */
-    int		     error;
-    int		     record_cnt = 0, data_cnt = 0;
-
-    partial_copy_len = partial_copy = 0;
-
-    /* Calculate potential maximum space.  Each region gets its own
-     * xlog_op_header_t and may need to be double word aligned.
-     */
-    len = 0;
-    if (ticket->t_flags & XLOG_TIC_INITED) {    /* acct for start rec of xact */
-	len += sizeof(xlog_op_header_t);
-	ticket->t_res_num_ophdrs++;
-    }
+	struct xlog_in_core	*iclog = NULL;
+	struct xfs_log_iovec	*vecp;
+	struct xfs_log_vec	*lv;
+	int			len;
+	int			index;
+	int			partial_copy = 0;
+	int			partial_copy_len = 0;
+	int			contwr = 0;
+	int			record_cnt = 0;
+	int			data_cnt = 0;
+	int			error;
 
-    for (index = 0; index < nentries; index++) {
-	len += sizeof(xlog_op_header_t);	    /* each region gets >= 1 */
-	ticket->t_res_num_ophdrs++;
-	len += reg[index].i_len;
-	xlog_tic_add_region(ticket, reg[index].i_len, reg[index].i_type);
-    }
-    contwr = *start_lsn = 0;
+	*start_lsn = 0;
 
-    if (ticket->t_curr_res < len) {
-	xlog_print_tic_res(mp, ticket);
+	len = xlog_write_calc_vec_length(ticket, log_vector);
+	if (ticket->t_curr_res < len) {
+		xlog_print_tic_res(log->l_mp, ticket);
 #ifdef DEBUG
-	xlog_panic(
-		"xfs_log_write: reservation ran out. Need to up reservation");
+		xlog_panic(
+	"xfs_log_write: reservation ran out. Need to up reservation");
 #else
-	/* Customer configurable panic */
-	xfs_cmn_err(XFS_PTAG_LOGRES, CE_ALERT, mp,
-		"xfs_log_write: reservation ran out. Need to up reservation");
-	/* If we did not panic, shutdown the filesystem */
-	xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
+		/* Customer configurable panic */
+		xfs_cmn_err(XFS_PTAG_LOGRES, CE_ALERT, log->l_mp,
+	"xfs_log_write: reservation ran out. Need to up reservation");
+
+		/* If we did not panic, shutdown the filesystem */
+		xfs_force_shutdown(log->l_mp, SHUTDOWN_CORRUPT_INCORE);
 #endif
-    } else
+	}
+
 	ticket->t_curr_res -= len;
 
-    for (index = 0; index < nentries; ) {
-	if ((error = xlog_state_get_iclog_space(log, len, &iclog, ticket,
-					       &contwr, &log_offset)))
-		return error;
+	index = 0;
+	lv = log_vector;
+	vecp = lv->lv_iovecp;
+	while (lv && index < lv->lv_niovecs) {
+		void		*ptr;
+		int		log_offset;
 
-	ASSERT(log_offset <= iclog->ic_size - 1);
-	ptr = (__psint_t) ((char *)iclog->ic_datap+log_offset);
+		error = xlog_state_get_iclog_space(log, len, &iclog, ticket,
+						   &contwr, &log_offset);
+		if (error)
+			return error;
 
-	/* start_lsn is the first lsn written to. That's all we need. */
-	if (! *start_lsn)
-	    *start_lsn = be64_to_cpu(iclog->ic_header.h_lsn);
+		ASSERT(log_offset <= iclog->ic_size - 1);
+		ptr = iclog->ic_datap + log_offset;
 
-	/* This loop writes out as many regions as can fit in the amount
-	 * of space which was allocated by xlog_state_get_iclog_space().
-	 */
-	while (index < nentries) {
-	    ASSERT(reg[index].i_len % sizeof(__int32_t) == 0);
-	    ASSERT((__psint_t)ptr % sizeof(__int32_t) == 0);
-	    start_rec_copy = 0;
-
-	    /* If first write for transaction, insert start record.
-	     * We can't be trying to commit if we are inited.  We can't
-	     * have any "partial_copy" if we are inited.
-	     */
-	    if (ticket->t_flags & XLOG_TIC_INITED) {
-		logop_head		= (xlog_op_header_t *)ptr;
-		logop_head->oh_tid	= cpu_to_be32(ticket->t_tid);
-		logop_head->oh_clientid = ticket->t_clientid;
-		logop_head->oh_len	= 0;
-		logop_head->oh_flags    = XLOG_START_TRANS;
-		logop_head->oh_res2	= 0;
-		ticket->t_flags		&= ~XLOG_TIC_INITED;	/* clear bit */
-		record_cnt++;
-
-		start_rec_copy = sizeof(xlog_op_header_t);
-		xlog_write_adv_cnt(ptr, len, log_offset, start_rec_copy);
-	    }
+		/* start_lsn is the first lsn written to. That's all we need. */
+		if (!*start_lsn)
+			*start_lsn = be64_to_cpu(iclog->ic_header.h_lsn);
 
-	    /* Copy log operation header directly into data section */
-	    logop_head			= (xlog_op_header_t *)ptr;
-	    logop_head->oh_tid		= cpu_to_be32(ticket->t_tid);
-	    logop_head->oh_clientid	= ticket->t_clientid;
-	    logop_head->oh_res2		= 0;
+		/*
+		 * This loop writes out as many regions as can fit in the amount
+		 * of space which was allocated by xlog_state_get_iclog_space().
+		 */
+		while (lv && index < lv->lv_niovecs) {
+			struct xfs_log_iovec	*reg = &vecp[index];
+			struct xlog_op_header	*ophdr;
+			int			start_rec_copy;
+			int			copy_len;
+			int			copy_off;
+
+			ASSERT(reg->i_len % sizeof(__int32_t) == 0);
+			ASSERT((unsigned long)ptr % sizeof(__int32_t) == 0);
+
+			start_rec_copy = xlog_write_start_rec(ptr, ticket);
+			if (start_rec_copy) {
+				record_cnt++;
+				xlog_write_adv_cnt(&ptr, &len, &log_offset,
+						   start_rec_copy);
+			}
 
-	    /* header copied directly */
-	    xlog_write_adv_cnt(ptr, len, log_offset, sizeof(xlog_op_header_t));
+			ophdr = xlog_write_setup_ophdr(log, ptr, ticket, flags);
+			if (!ophdr)
+				return XFS_ERROR(EIO);
 
-	    /* are we copying a commit or unmount record? */
-	    logop_head->oh_flags = flags;
+			xlog_write_adv_cnt(&ptr, &len, &log_offset,
+					   sizeof(struct xlog_op_header));
+
+			len += xlog_write_setup_copy(ticket, ophdr,
+						     iclog->ic_size-log_offset,
+						     reg->i_len,
+						     &copy_off, &copy_len,
+						     &partial_copy,
+						     &partial_copy_len);
+			xlog_verify_dest_ptr(log, ptr);
+
+			/* copy region */
+			ASSERT(copy_len >= 0);
+			memcpy(ptr, reg->i_addr + copy_off, copy_len);
+			xlog_write_adv_cnt(&ptr, &len, &log_offset, copy_len);
+
+			copy_len += start_rec_copy + sizeof(xlog_op_header_t);
+			record_cnt++;
+			data_cnt += contwr ? copy_len : 0;
+
+			error = xlog_write_copy_finish(log, iclog, flags,
+						       &record_cnt, &data_cnt,
+						       &partial_copy,
+						       &partial_copy_len,
+						       log_offset,
+						       commit_iclog);
+			if (error)
+				return error;
 
-	    /*
-	     * We've seen logs corrupted with bad transaction client
-	     * ids.  This makes sure that XFS doesn't generate them on.
-	     * Turn this into an EIO and shut down the filesystem.
-	     */
-	    switch (logop_head->oh_clientid)  {
-	    case XFS_TRANSACTION:
-	    case XFS_VOLUME:
-	    case XFS_LOG:
-		break;
-	    default:
-		xfs_fs_cmn_err(CE_WARN, mp,
-		    "Bad XFS transaction clientid 0x%x in ticket 0x%p",
-		    logop_head->oh_clientid, ticket);
-		return XFS_ERROR(EIO);
-	    }
+			/*
+			 * if we had a partial copy, we need to get more iclog
+			 * space but we don't want to increment the region
+			 * index because there is still more is this region to
+			 * write.
+			 *
+			 * If we completed writing this region, and we flushed
+			 * the iclog (indicated by resetting of the record
+			 * count), then we also need to get more log space. If
+			 * this was the last record, though, we are done and
+			 * can just return.
+			 */
+			if (partial_copy)
+				break;
 
-	    /* Partial write last time? => (partial_copy != 0)
-	     * need_copy is the amount we'd like to copy if everything could
-	     * fit in the current memcpy.
-	     */
-	    need_copy =	reg[index].i_len - partial_copy_len;
-
-	    copy_off = partial_copy_len;
-	    if (need_copy <= iclog->ic_size - log_offset) { /*complete write */
-	        copy_len = need_copy;
-		logop_head->oh_len = cpu_to_be32(copy_len);
-		if (partial_copy)
-		    logop_head->oh_flags|= (XLOG_END_TRANS|XLOG_WAS_CONT_TRANS);
-		partial_copy_len = partial_copy = 0;
-	    } else {					    /* partial write */
-		copy_len = iclog->ic_size - log_offset;
-		logop_head->oh_len = cpu_to_be32(copy_len);
-		logop_head->oh_flags |= XLOG_CONTINUE_TRANS;
-		if (partial_copy)
-			logop_head->oh_flags |= XLOG_WAS_CONT_TRANS;
-		partial_copy_len += copy_len;
-		partial_copy++;
-		len += sizeof(xlog_op_header_t); /* from splitting of region */
-		/* account for new log op header */
-		ticket->t_curr_res -= sizeof(xlog_op_header_t);
-		ticket->t_res_num_ophdrs++;
-	    }
-	    xlog_verify_dest_ptr(log, ptr);
-
-	    /* copy region */
-	    ASSERT(copy_len >= 0);
-	    memcpy((xfs_caddr_t)ptr, reg[index].i_addr + copy_off, copy_len);
-	    xlog_write_adv_cnt(ptr, len, log_offset, copy_len);
-
-	    /* make copy_len total bytes copied, including headers */
-	    copy_len += start_rec_copy + sizeof(xlog_op_header_t);
-	    record_cnt++;
-	    data_cnt += contwr ? copy_len : 0;
-	    if (partial_copy) {			/* copied partial region */
-		    /* already marked WANT_SYNC by xlog_state_get_iclog_space */
-		    xlog_state_finish_copy(log, iclog, record_cnt, data_cnt);
-		    record_cnt = data_cnt = 0;
-		    if ((error = xlog_state_release_iclog(log, iclog)))
-			    return error;
-		    break;			/* don't increment index */
-	    } else {				/* copied entire region */
-		index++;
-		partial_copy_len = partial_copy = 0;
-
-		if (iclog->ic_size - log_offset <= sizeof(xlog_op_header_t)) {
-		    xlog_state_finish_copy(log, iclog, record_cnt, data_cnt);
-		    record_cnt = data_cnt = 0;
-		    spin_lock(&log->l_icloglock);
-		    xlog_state_want_sync(log, iclog);
-		    spin_unlock(&log->l_icloglock);
-		    if (commit_iclog) {
-			ASSERT(flags & XLOG_COMMIT_TRANS);
-			*commit_iclog = iclog;
-		    } else if ((error = xlog_state_release_iclog(log, iclog)))
-			   return error;
-		    if (index == nentries)
-			    return 0;		/* we are done */
-		    else
-			    break;
+			if (++index == lv->lv_niovecs) {
+				lv = lv->lv_next;
+				index = 0;
+				if (lv)
+					vecp = lv->lv_iovecp;
+			}
+			if (record_cnt == 0) {
+				if (!lv)
+					return 0;
+				break;
+			}
 		}
-	    } /* if (partial_copy) */
-	} /* while (index < nentries) */
-    } /* for (index = 0; index < nentries; ) */
-    ASSERT(len == 0);
+	}
+
+	ASSERT(len == 0);
+
+	xlog_state_finish_copy(log, iclog, record_cnt, data_cnt);
+	if (!commit_iclog)
+		return xlog_state_release_iclog(log, iclog);
 
-    xlog_state_finish_copy(log, iclog, record_cnt, data_cnt);
-    if (commit_iclog) {
 	ASSERT(flags & XLOG_COMMIT_TRANS);
 	*commit_iclog = iclog;
 	return 0;
-    }
-    return xlog_state_release_iclog(log, iclog);
-}	/* xlog_write */
+}
 
 
 /*****************************************************************************
@@ -3157,14 +3317,16 @@ xfs_log_ticket_get(
  * Allocate and initialise a new log ticket.
  */
 STATIC xlog_ticket_t *
-xlog_ticket_alloc(xlog_t		*log,
-		int		unit_bytes,
-		int		cnt,
-		char		client,
-		uint		xflags)
+xlog_ticket_alloc(
+	struct log	*log,
+	int		unit_bytes,
+	int		cnt,
+	char		client,
+	uint		xflags)
 {
-	xlog_ticket_t	*tic;
+	struct xlog_ticket *tic;
 	uint		num_headers;
+	int		iclog_space;
 
 	tic = kmem_zone_zalloc(xfs_log_ticket_zone, KM_SLEEP|KM_MAYFAIL);
 	if (!tic)
@@ -3208,16 +3370,40 @@ xlog_ticket_alloc(xlog_t		*log,
 	/* for start-rec */
 	unit_bytes += sizeof(xlog_op_header_t);
 
-	/* for LR headers */
-	num_headers = ((unit_bytes + log->l_iclog_size-1) >> log->l_iclog_size_log);
+	/*
+	 * for LR headers - the space for data in an iclog is the size minus
+	 * the space used for the headers. If we use the iclog size, then we
+	 * undercalculate the number of headers required.
+	 *
+	 * Furthermore - the addition of op headers for split-recs might
+	 * increase the space required enough to require more log and op
+	 * headers, so take that into account too.
+	 *
+	 * IMPORTANT: This reservation makes the assumption that if this
+	 * transaction is the first in an iclog and hence has the LR headers
+	 * accounted to it, then the remaining space in the iclog is
+	 * exclusively for this transaction.  i.e. if the transaction is larger
+	 * than the iclog, it will be the only thing in that iclog.
+	 * Fundamentally, this means we must pass the entire log vector to
+	 * xlog_write to guarantee this.
+	 */
+	iclog_space = log->l_iclog_size - log->l_iclog_hsize;
+	num_headers = howmany(unit_bytes, iclog_space);
+
+	/* for split-recs - ophdrs added when data split over LRs */
+	unit_bytes += sizeof(xlog_op_header_t) * num_headers;
+
+	/* add extra header reservations if we overrun */
+	while (!num_headers ||
+	       howmany(unit_bytes, iclog_space) > num_headers) {
+		unit_bytes += sizeof(xlog_op_header_t);
+		num_headers++;
+	}
 	unit_bytes += log->l_iclog_hsize * num_headers;
 
 	/* for commit-rec LR header - note: padding will subsume the ophdr */
 	unit_bytes += log->l_iclog_hsize;
 
-	/* for split-recs - ophdrs added when data split over LRs */
-	unit_bytes += sizeof(xlog_op_header_t) * num_headers;
-
 	/* for roundoff padding for transaction data and one for commit record */
 	if (xfs_sb_version_haslogv2(&log->l_mp->m_sb) &&
 	    log->l_mp->m_sb.sb_logsunit > 1) {
@@ -3233,13 +3419,13 @@ xlog_ticket_alloc(xlog_t		*log,
 	tic->t_curr_res		= unit_bytes;
 	tic->t_cnt		= cnt;
 	tic->t_ocnt		= cnt;
-	tic->t_tid		= (xlog_tid_t)((__psint_t)tic & 0xffffffff);
+	tic->t_tid		= random32();
 	tic->t_clientid		= client;
 	tic->t_flags		= XLOG_TIC_INITED;
 	tic->t_trans_type	= 0;
 	if (xflags & XFS_LOG_PERM_RESERV)
 		tic->t_flags |= XLOG_TIC_PERM_RESERV;
-	sv_init(&(tic->t_wait), SV_DEFAULT, "logtick");
+	sv_init(&tic->t_wait, SV_DEFAULT, "logtick");
 
 	xlog_tic_reset_res(tic);
 
@@ -3260,20 +3446,22 @@ xlog_ticket_alloc(xlog_t		*log,
  * part of the log in case we trash the log structure.
  */
 void
-xlog_verify_dest_ptr(xlog_t     *log,
-		     __psint_t  ptr)
+xlog_verify_dest_ptr(
+	struct log	*log,
+	char		*ptr)
 {
 	int i;
 	int good_ptr = 0;
 
-	for (i=0; i < log->l_iclog_bufs; i++) {
-		if (ptr >= (__psint_t)log->l_iclog_bak[i] &&
-		    ptr <= (__psint_t)log->l_iclog_bak[i]+log->l_iclog_size)
+	for (i = 0; i < log->l_iclog_bufs; i++) {
+		if (ptr >= log->l_iclog_bak[i] &&
+		    ptr <= log->l_iclog_bak[i] + log->l_iclog_size)
 			good_ptr++;
 	}
-	if (! good_ptr)
+
+	if (!good_ptr)
 		xlog_panic("xlog_verify_dest_ptr: invalid ptr");
-}	/* xlog_verify_dest_ptr */
+}
 
 STATIC void
 xlog_verify_grant_head(xlog_t *log, int equals)
diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h
index 97a24c7795a..229d1f36ba9 100644
--- a/fs/xfs/xfs_log.h
+++ b/fs/xfs/xfs_log.h
@@ -110,6 +110,12 @@ typedef struct xfs_log_iovec {
 	uint		i_type;		/* type of region */
 } xfs_log_iovec_t;
 
+struct xfs_log_vec {
+	struct xfs_log_vec	*lv_next;	/* next lv in build list */
+	int			lv_niovecs;	/* number of iovecs in lv */
+	struct xfs_log_iovec	*lv_iovecp;	/* iovec array */
+};
+
 /*
  * Structure used to pass callback function and the function's argument
  * to the log manager.
@@ -126,6 +132,13 @@ typedef struct xfs_log_callback {
 struct xfs_mount;
 struct xlog_in_core;
 struct xlog_ticket;
+struct xfs_log_item;
+struct xfs_item_ops;
+
+void	xfs_log_item_init(struct xfs_mount	*mp,
+			struct xfs_log_item	*item,
+			int			type,
+			struct xfs_item_ops	*ops);
 
 xfs_lsn_t xfs_log_done(struct xfs_mount *mp,
 		       struct xlog_ticket *ticket,
diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h
index fd02a18facd..9cf69515445 100644
--- a/fs/xfs/xfs_log_priv.h
+++ b/fs/xfs/xfs_log_priv.h
@@ -396,9 +396,7 @@ typedef struct log {
 	struct xfs_buf_cancel	**l_buf_cancel_table;
 	int			l_iclog_hsize;  /* size of iclog header */
 	int			l_iclog_heads;  /* # of iclog header sectors */
-	uint			l_sectbb_log;   /* log2 of sector size in BBs */
-	uint			l_sectbb_mask;  /* sector size (in BBs)
-						 * alignment mask */
+	uint			l_sectBBsize;   /* sector size in BBs (2^n) */
 	int			l_iclog_size;	/* size of log in bytes */
 	int			l_iclog_size_log; /* log power size of log */
 	int			l_iclog_bufs;	/* number of iclog buffers */
@@ -449,6 +447,14 @@ extern void	 xlog_pack_data(xlog_t *log, xlog_in_core_t *iclog, int);
 
 extern kmem_zone_t	*xfs_log_ticket_zone;
 
+static inline void
+xlog_write_adv_cnt(void **ptr, int *len, int *off, size_t bytes)
+{
+	*ptr += bytes;
+	*len -= bytes;
+	*off += bytes;
+}
+
 /*
  * Unmount record type is used as a pseudo transaction type for the ticket.
  * It's value must be outside the range of XFS_TRANS_* values.
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 22e6efdc17e..0de08e36631 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -56,33 +56,61 @@ STATIC void	xlog_recover_check_summary(xlog_t *);
 #define	xlog_recover_check_summary(log)
 #endif
 
-
 /*
  * Sector aligned buffer routines for buffer create/read/write/access
  */
 
-#define XLOG_SECTOR_ROUNDUP_BBCOUNT(log, bbs)	\
-	( ((log)->l_sectbb_mask && (bbs & (log)->l_sectbb_mask)) ? \
-	((bbs + (log)->l_sectbb_mask + 1) & ~(log)->l_sectbb_mask) : (bbs) )
-#define XLOG_SECTOR_ROUNDDOWN_BLKNO(log, bno)	((bno) & ~(log)->l_sectbb_mask)
+/*
+ * Verify the given count of basic blocks is valid number of blocks
+ * to specify for an operation involving the given XFS log buffer.
+ * Returns nonzero if the count is valid, 0 otherwise.
+ */
 
+static inline int
+xlog_buf_bbcount_valid(
+	xlog_t		*log,
+	int		bbcount)
+{
+	return bbcount > 0 && bbcount <= log->l_logBBsize;
+}
+
+/*
+ * Allocate a buffer to hold log data.  The buffer needs to be able
+ * to map to a range of nbblks basic blocks at any valid (basic
+ * block) offset within the log.
+ */
 STATIC xfs_buf_t *
 xlog_get_bp(
 	xlog_t		*log,
 	int		nbblks)
 {
-	if (nbblks <= 0 || nbblks > log->l_logBBsize) {
-		xlog_warn("XFS: Invalid block length (0x%x) given for buffer", nbblks);
-		XFS_ERROR_REPORT("xlog_get_bp(1)",
-				 XFS_ERRLEVEL_HIGH, log->l_mp);
+	if (!xlog_buf_bbcount_valid(log, nbblks)) {
+		xlog_warn("XFS: Invalid block length (0x%x) given for buffer",
+			nbblks);
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
 		return NULL;
 	}
 
-	if (log->l_sectbb_log) {
-		if (nbblks > 1)
-			nbblks += XLOG_SECTOR_ROUNDUP_BBCOUNT(log, 1);
-		nbblks = XLOG_SECTOR_ROUNDUP_BBCOUNT(log, nbblks);
-	}
+	/*
+	 * We do log I/O in units of log sectors (a power-of-2
+	 * multiple of the basic block size), so we round up the
+	 * requested size to acommodate the basic blocks required
+	 * for complete log sectors.
+	 *
+	 * In addition, the buffer may be used for a non-sector-
+	 * aligned block offset, in which case an I/O of the
+	 * requested size could extend beyond the end of the
+	 * buffer.  If the requested size is only 1 basic block it
+	 * will never straddle a sector boundary, so this won't be
+	 * an issue.  Nor will this be a problem if the log I/O is
+	 * done in basic blocks (sector size 1).  But otherwise we
+	 * extend the buffer by one extra log sector to ensure
+	 * there's space to accomodate this possiblility.
+	 */
+	if (nbblks > 1 && log->l_sectBBsize > 1)
+		nbblks += log->l_sectBBsize;
+	nbblks = round_up(nbblks, log->l_sectBBsize);
+
 	return xfs_buf_get_noaddr(BBTOB(nbblks), log->l_mp->m_logdev_targp);
 }
 
@@ -93,6 +121,10 @@ xlog_put_bp(
 	xfs_buf_free(bp);
 }
 
+/*
+ * Return the address of the start of the given block number's data
+ * in a log buffer.  The buffer covers a log sector-aligned region.
+ */
 STATIC xfs_caddr_t
 xlog_align(
 	xlog_t		*log,
@@ -100,14 +132,14 @@ xlog_align(
 	int		nbblks,
 	xfs_buf_t	*bp)
 {
+	xfs_daddr_t	offset;
 	xfs_caddr_t	ptr;
 
-	if (!log->l_sectbb_log)
-		return XFS_BUF_PTR(bp);
+	offset = blk_no & ((xfs_daddr_t) log->l_sectBBsize - 1);
+	ptr = XFS_BUF_PTR(bp) + BBTOB(offset);
+
+	ASSERT(ptr + BBTOB(nbblks) <= XFS_BUF_PTR(bp) + XFS_BUF_SIZE(bp));
 
-	ptr = XFS_BUF_PTR(bp) + BBTOB((int)blk_no & log->l_sectbb_mask);
-	ASSERT(XFS_BUF_SIZE(bp) >=
-		BBTOB(nbblks + (blk_no & log->l_sectbb_mask)));
 	return ptr;
 }
 
@@ -124,21 +156,18 @@ xlog_bread_noalign(
 {
 	int		error;
 
-	if (nbblks <= 0 || nbblks > log->l_logBBsize) {
-		xlog_warn("XFS: Invalid block length (0x%x) given for buffer", nbblks);
-		XFS_ERROR_REPORT("xlog_bread(1)",
-				 XFS_ERRLEVEL_HIGH, log->l_mp);
+	if (!xlog_buf_bbcount_valid(log, nbblks)) {
+		xlog_warn("XFS: Invalid block length (0x%x) given for buffer",
+			nbblks);
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
 		return EFSCORRUPTED;
 	}
 
-	if (log->l_sectbb_log) {
-		blk_no = XLOG_SECTOR_ROUNDDOWN_BLKNO(log, blk_no);
-		nbblks = XLOG_SECTOR_ROUNDUP_BBCOUNT(log, nbblks);
-	}
+	blk_no = round_down(blk_no, log->l_sectBBsize);
+	nbblks = round_up(nbblks, log->l_sectBBsize);
 
 	ASSERT(nbblks > 0);
 	ASSERT(BBTOB(nbblks) <= XFS_BUF_SIZE(bp));
-	ASSERT(bp);
 
 	XFS_BUF_SET_ADDR(bp, log->l_logBBstart + blk_no);
 	XFS_BUF_READ(bp);
@@ -186,17 +215,15 @@ xlog_bwrite(
 {
 	int		error;
 
-	if (nbblks <= 0 || nbblks > log->l_logBBsize) {
-		xlog_warn("XFS: Invalid block length (0x%x) given for buffer", nbblks);
-		XFS_ERROR_REPORT("xlog_bwrite(1)",
-				 XFS_ERRLEVEL_HIGH, log->l_mp);
+	if (!xlog_buf_bbcount_valid(log, nbblks)) {
+		xlog_warn("XFS: Invalid block length (0x%x) given for buffer",
+			nbblks);
+		XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
 		return EFSCORRUPTED;
 	}
 
-	if (log->l_sectbb_log) {
-		blk_no = XLOG_SECTOR_ROUNDDOWN_BLKNO(log, blk_no);
-		nbblks = XLOG_SECTOR_ROUNDUP_BBCOUNT(log, nbblks);
-	}
+	blk_no = round_down(blk_no, log->l_sectBBsize);
+	nbblks = round_up(nbblks, log->l_sectBBsize);
 
 	ASSERT(nbblks > 0);
 	ASSERT(BBTOB(nbblks) <= XFS_BUF_SIZE(bp));
@@ -327,39 +354,38 @@ xlog_find_cycle_start(
 {
 	xfs_caddr_t	offset;
 	xfs_daddr_t	mid_blk;
+	xfs_daddr_t	end_blk;
 	uint		mid_cycle;
 	int		error;
 
-	mid_blk = BLK_AVG(first_blk, *last_blk);
-	while (mid_blk != first_blk && mid_blk != *last_blk) {
+	end_blk = *last_blk;
+	mid_blk = BLK_AVG(first_blk, end_blk);
+	while (mid_blk != first_blk && mid_blk != end_blk) {
 		error = xlog_bread(log, mid_blk, 1, bp, &offset);
 		if (error)
 			return error;
 		mid_cycle = xlog_get_cycle(offset);
-		if (mid_cycle == cycle) {
-			*last_blk = mid_blk;
-			/* last_half_cycle == mid_cycle */
-		} else {
-			first_blk = mid_blk;
-			/* first_half_cycle == mid_cycle */
-		}
-		mid_blk = BLK_AVG(first_blk, *last_blk);
+		if (mid_cycle == cycle)
+			end_blk = mid_blk;   /* last_half_cycle == mid_cycle */
+		else
+			first_blk = mid_blk; /* first_half_cycle == mid_cycle */
+		mid_blk = BLK_AVG(first_blk, end_blk);
 	}
-	ASSERT((mid_blk == first_blk && mid_blk+1 == *last_blk) ||
-	       (mid_blk == *last_blk && mid_blk-1 == first_blk));
+	ASSERT((mid_blk == first_blk && mid_blk+1 == end_blk) ||
+	       (mid_blk == end_blk && mid_blk-1 == first_blk));
+
+	*last_blk = end_blk;
 
 	return 0;
 }
 
 /*
- * Check that the range of blocks does not contain the cycle number
- * given.  The scan needs to occur from front to back and the ptr into the
- * region must be updated since a later routine will need to perform another
- * test.  If the region is completely good, we end up returning the same
- * last block number.
- *
- * Set blkno to -1 if we encounter no errors.  This is an invalid block number
- * since we don't ever expect logs to get this large.
+ * Check that a range of blocks does not contain stop_on_cycle_no.
+ * Fill in *new_blk with the block offset where such a block is
+ * found, or with -1 (an invalid block number) if there is no such
+ * block in the range.  The scan needs to occur from front to back
+ * and the pointer into the region must be updated since a later
+ * routine will need to perform another test.
  */
 STATIC int
 xlog_find_verify_cycle(
@@ -376,12 +402,16 @@ xlog_find_verify_cycle(
 	xfs_caddr_t	buf = NULL;
 	int		error = 0;
 
+	/*
+	 * Greedily allocate a buffer big enough to handle the full
+	 * range of basic blocks we'll be examining.  If that fails,
+	 * try a smaller size.  We need to be able to read at least
+	 * a log sector, or we're out of luck.
+	 */
 	bufblks = 1 << ffs(nbblks);
-
 	while (!(bp = xlog_get_bp(log, bufblks))) {
-		/* can't get enough memory to do everything in one big buffer */
 		bufblks >>= 1;
-		if (bufblks <= log->l_sectbb_log)
+		if (bufblks < log->l_sectBBsize)
 			return ENOMEM;
 	}
 
@@ -629,7 +659,7 @@ xlog_find_head(
 		 * In this case we want to find the first block with cycle
 		 * number matching last_half_cycle.  We expect the log to be
 		 * some variation on
-		 *        x + 1 ... | x ...
+		 *        x + 1 ... | x ... | x
 		 * The first block with cycle number x (last_half_cycle) will
 		 * be where the new head belongs.  First we do a binary search
 		 * for the first occurrence of last_half_cycle.  The binary
@@ -639,11 +669,13 @@ xlog_find_head(
 		 * the log, then we look for occurrences of last_half_cycle - 1
 		 * at the end of the log.  The cases we're looking for look
 		 * like
-		 *        x + 1 ... | x | x + 1 | x ...
-		 *                               ^ binary search stopped here
+		 *                               v binary search stopped here
+		 *        x + 1 ... | x | x + 1 | x ... | x
+		 *                   ^ but we want to locate this spot
 		 * or
-		 *        x + 1 ... | x ... | x - 1 | x
 		 *        <---------> less than scan distance
+		 *        x + 1 ... | x ... | x - 1 | x
+		 *                           ^ we want to locate this spot
 		 */
 		stop_on_cycle = last_half_cycle;
 		if ((error = xlog_find_cycle_start(log, bp, first_blk,
@@ -699,16 +731,16 @@ xlog_find_head(
 		 * certainly not the head of the log.  By searching for
 		 * last_half_cycle-1 we accomplish that.
 		 */
-		start_blk = log_bbnum - num_scan_bblks + head_blk;
 		ASSERT(head_blk <= INT_MAX &&
-			(xfs_daddr_t) num_scan_bblks - head_blk >= 0);
+			(xfs_daddr_t) num_scan_bblks >= head_blk);
+		start_blk = log_bbnum - (num_scan_bblks - head_blk);
 		if ((error = xlog_find_verify_cycle(log, start_blk,
 					num_scan_bblks - (int)head_blk,
 					(stop_on_cycle - 1), &new_blk)))
 			goto bp_err;
 		if (new_blk != -1) {
 			head_blk = new_blk;
-			goto bad_blk;
+			goto validate_head;
 		}
 
 		/*
@@ -726,7 +758,7 @@ xlog_find_head(
 			head_blk = new_blk;
 	}
 
- bad_blk:
+validate_head:
 	/*
 	 * Now we need to make sure head_blk is not pointing to a block in
 	 * the middle of a log record.
@@ -748,7 +780,7 @@ xlog_find_head(
 		if ((error = xlog_find_verify_log_record(log, start_blk,
 							&head_blk, 0)) == -1) {
 			/* We hit the beginning of the log during our search */
-			start_blk = log_bbnum - num_scan_bblks + head_blk;
+			start_blk = log_bbnum - (num_scan_bblks - head_blk);
 			new_blk = log_bbnum;
 			ASSERT(start_blk <= INT_MAX &&
 				(xfs_daddr_t) log_bbnum-start_blk >= 0);
@@ -833,12 +865,12 @@ xlog_find_tail(
 	if (*head_blk == 0) {				/* special case */
 		error = xlog_bread(log, 0, 1, bp, &offset);
 		if (error)
-			goto bread_err;
+			goto done;
 
 		if (xlog_get_cycle(offset) == 0) {
 			*tail_blk = 0;
 			/* leave all other log inited values alone */
-			goto exit;
+			goto done;
 		}
 	}
 
@@ -849,7 +881,7 @@ xlog_find_tail(
 	for (i = (int)(*head_blk) - 1; i >= 0; i--) {
 		error = xlog_bread(log, i, 1, bp, &offset);
 		if (error)
-			goto bread_err;
+			goto done;
 
 		if (XLOG_HEADER_MAGIC_NUM == be32_to_cpu(*(__be32 *)offset)) {
 			found = 1;
@@ -866,7 +898,7 @@ xlog_find_tail(
 		for (i = log->l_logBBsize - 1; i >= (int)(*head_blk); i--) {
 			error = xlog_bread(log, i, 1, bp, &offset);
 			if (error)
-				goto bread_err;
+				goto done;
 
 			if (XLOG_HEADER_MAGIC_NUM ==
 			    be32_to_cpu(*(__be32 *)offset)) {
@@ -941,7 +973,7 @@ xlog_find_tail(
 		umount_data_blk = (i + hblks) % log->l_logBBsize;
 		error = xlog_bread(log, umount_data_blk, 1, bp, &offset);
 		if (error)
-			goto bread_err;
+			goto done;
 
 		op_head = (xlog_op_header_t *)offset;
 		if (op_head->oh_flags & XLOG_UNMOUNT_TRANS) {
@@ -987,12 +1019,10 @@ xlog_find_tail(
 	 * But... if the -device- itself is readonly, just skip this.
 	 * We can't recover this device anyway, so it won't matter.
 	 */
-	if (!xfs_readonly_buftarg(log->l_mp->m_logdev_targp)) {
+	if (!xfs_readonly_buftarg(log->l_mp->m_logdev_targp))
 		error = xlog_clear_stale_blocks(log, tail_lsn);
-	}
 
-bread_err:
-exit:
+done:
 	xlog_put_bp(bp);
 
 	if (error)
@@ -1152,16 +1182,22 @@ xlog_write_log_records(
 	xfs_caddr_t	offset;
 	xfs_buf_t	*bp;
 	int		balign, ealign;
-	int		sectbb = XLOG_SECTOR_ROUNDUP_BBCOUNT(log, 1);
+	int		sectbb = log->l_sectBBsize;
 	int		end_block = start_block + blocks;
 	int		bufblks;
 	int		error = 0;
 	int		i, j = 0;
 
+	/*
+	 * Greedily allocate a buffer big enough to handle the full
+	 * range of basic blocks to be written.  If that fails, try
+	 * a smaller size.  We need to be able to write at least a
+	 * log sector, or we're out of luck.
+	 */
 	bufblks = 1 << ffs(blocks);
 	while (!(bp = xlog_get_bp(log, bufblks))) {
 		bufblks >>= 1;
-		if (bufblks <= log->l_sectbb_log)
+		if (bufblks < sectbb)
 			return ENOMEM;
 	}
 
@@ -1169,7 +1205,7 @@ xlog_write_log_records(
 	 * the buffer in the starting sector not covered by the first
 	 * write below.
 	 */
-	balign = XLOG_SECTOR_ROUNDDOWN_BLKNO(log, start_block);
+	balign = round_down(start_block, sectbb);
 	if (balign != start_block) {
 		error = xlog_bread_noalign(log, start_block, 1, bp);
 		if (error)
@@ -1188,7 +1224,7 @@ xlog_write_log_records(
 		 * the buffer in the final sector not covered by the write.
 		 * If this is the same sector as the above read, skip it.
 		 */
-		ealign = XLOG_SECTOR_ROUNDDOWN_BLKNO(log, end_block);
+		ealign = round_down(end_block, sectbb);
 		if (j == 0 && (start_block + endcount > ealign)) {
 			offset = XFS_BUF_PTR(bp);
 			balign = BBTOB(ealign - start_block);
@@ -1408,6 +1444,7 @@ xlog_recover_add_item(
 
 STATIC int
 xlog_recover_add_to_cont_trans(
+	struct log		*log,
 	xlog_recover_t		*trans,
 	xfs_caddr_t		dp,
 	int			len)
@@ -1434,6 +1471,7 @@ xlog_recover_add_to_cont_trans(
 	memcpy(&ptr[old_len], dp, len); /* d, s, l */
 	item->ri_buf[item->ri_cnt-1].i_len += len;
 	item->ri_buf[item->ri_cnt-1].i_addr = ptr;
+	trace_xfs_log_recover_item_add_cont(log, trans, item, 0);
 	return 0;
 }
 
@@ -1452,6 +1490,7 @@ xlog_recover_add_to_cont_trans(
  */
 STATIC int
 xlog_recover_add_to_trans(
+	struct log		*log,
 	xlog_recover_t		*trans,
 	xfs_caddr_t		dp,
 	int			len)
@@ -1510,6 +1549,7 @@ xlog_recover_add_to_trans(
 	item->ri_buf[item->ri_cnt].i_addr = ptr;
 	item->ri_buf[item->ri_cnt].i_len  = len;
 	item->ri_cnt++;
+	trace_xfs_log_recover_item_add(log, trans, item, 0);
 	return 0;
 }
 
@@ -1521,7 +1561,9 @@ xlog_recover_add_to_trans(
  */
 STATIC int
 xlog_recover_reorder_trans(
-	xlog_recover_t		*trans)
+	struct log		*log,
+	xlog_recover_t		*trans,
+	int			pass)
 {
 	xlog_recover_item_t	*item, *n;
 	LIST_HEAD(sort_list);
@@ -1535,6 +1577,8 @@ xlog_recover_reorder_trans(
 		switch (ITEM_TYPE(item)) {
 		case XFS_LI_BUF:
 			if (!(buf_f->blf_flags & XFS_BLI_CANCEL)) {
+				trace_xfs_log_recover_item_reorder_head(log,
+							trans, item, pass);
 				list_move(&item->ri_list, &trans->r_itemq);
 				break;
 			}
@@ -1543,6 +1587,8 @@ xlog_recover_reorder_trans(
 		case XFS_LI_QUOTAOFF:
 		case XFS_LI_EFD:
 		case XFS_LI_EFI:
+			trace_xfs_log_recover_item_reorder_tail(log,
+							trans, item, pass);
 			list_move_tail(&item->ri_list, &trans->r_itemq);
 			break;
 		default:
@@ -1592,8 +1638,10 @@ xlog_recover_do_buffer_pass1(
 	/*
 	 * If this isn't a cancel buffer item, then just return.
 	 */
-	if (!(flags & XFS_BLI_CANCEL))
+	if (!(flags & XFS_BLI_CANCEL)) {
+		trace_xfs_log_recover_buf_not_cancel(log, buf_f);
 		return;
+	}
 
 	/*
 	 * Insert an xfs_buf_cancel record into the hash table of
@@ -1627,6 +1675,7 @@ xlog_recover_do_buffer_pass1(
 	while (nextp != NULL) {
 		if (nextp->bc_blkno == blkno && nextp->bc_len == len) {
 			nextp->bc_refcount++;
+			trace_xfs_log_recover_buf_cancel_ref_inc(log, buf_f);
 			return;
 		}
 		prevp = nextp;
@@ -1640,6 +1689,7 @@ xlog_recover_do_buffer_pass1(
 	bcp->bc_refcount = 1;
 	bcp->bc_next = NULL;
 	prevp->bc_next = bcp;
+	trace_xfs_log_recover_buf_cancel_add(log, buf_f);
 }
 
 /*
@@ -1779,6 +1829,8 @@ xlog_recover_do_inode_buffer(
 	unsigned int		*data_map = NULL;
 	unsigned int		map_size = 0;
 
+	trace_xfs_log_recover_buf_inode_buf(mp->m_log, buf_f);
+
 	switch (buf_f->blf_type) {
 	case XFS_LI_BUF:
 		data_map = buf_f->blf_data_map;
@@ -1874,6 +1926,7 @@ xlog_recover_do_inode_buffer(
 /*ARGSUSED*/
 STATIC void
 xlog_recover_do_reg_buffer(
+	struct xfs_mount	*mp,
 	xlog_recover_item_t	*item,
 	xfs_buf_t		*bp,
 	xfs_buf_log_format_t	*buf_f)
@@ -1885,6 +1938,8 @@ xlog_recover_do_reg_buffer(
 	unsigned int		map_size = 0;
 	int                     error;
 
+	trace_xfs_log_recover_buf_reg_buf(mp->m_log, buf_f);
+
 	switch (buf_f->blf_type) {
 	case XFS_LI_BUF:
 		data_map = buf_f->blf_data_map;
@@ -2083,6 +2138,8 @@ xlog_recover_do_dquot_buffer(
 {
 	uint			type;
 
+	trace_xfs_log_recover_buf_dquot_buf(log, buf_f);
+
 	/*
 	 * Filesystems are required to send in quota flags at mount time.
 	 */
@@ -2103,7 +2160,7 @@ xlog_recover_do_dquot_buffer(
 	if (log->l_quotaoffs_flag & type)
 		return;
 
-	xlog_recover_do_reg_buffer(item, bp, buf_f);
+	xlog_recover_do_reg_buffer(mp, item, bp, buf_f);
 }
 
 /*
@@ -2164,9 +2221,11 @@ xlog_recover_do_buffer_trans(
 		 */
 		cancel = xlog_recover_do_buffer_pass2(log, buf_f);
 		if (cancel) {
+			trace_xfs_log_recover_buf_cancel(log, buf_f);
 			return 0;
 		}
 	}
+	trace_xfs_log_recover_buf_recover(log, buf_f);
 	switch (buf_f->blf_type) {
 	case XFS_LI_BUF:
 		blkno = buf_f->blf_blkno;
@@ -2204,7 +2263,7 @@ xlog_recover_do_buffer_trans(
 		  (XFS_BLI_UDQUOT_BUF|XFS_BLI_PDQUOT_BUF|XFS_BLI_GDQUOT_BUF)) {
 		xlog_recover_do_dquot_buffer(mp, log, item, bp, buf_f);
 	} else {
-		xlog_recover_do_reg_buffer(item, bp, buf_f);
+		xlog_recover_do_reg_buffer(mp, item, bp, buf_f);
 	}
 	if (error)
 		return XFS_ERROR(error);
@@ -2284,8 +2343,10 @@ xlog_recover_do_inode_trans(
 	if (xlog_check_buffer_cancelled(log, in_f->ilf_blkno,
 					in_f->ilf_len, 0)) {
 		error = 0;
+		trace_xfs_log_recover_inode_cancel(log, in_f);
 		goto error;
 	}
+	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);
@@ -2337,6 +2398,7 @@ xlog_recover_do_inode_trans(
 			/* do nothing */
 		} else {
 			xfs_buf_relse(bp);
+			trace_xfs_log_recover_inode_skip(log, in_f);
 			error = 0;
 			goto error;
 		}
@@ -2758,11 +2820,12 @@ xlog_recover_do_trans(
 	int			error = 0;
 	xlog_recover_item_t	*item;
 
-	error = xlog_recover_reorder_trans(trans);
+	error = xlog_recover_reorder_trans(log, trans, pass);
 	if (error)
 		return error;
 
 	list_for_each_entry(item, &trans->r_itemq, ri_list) {
+		trace_xfs_log_recover_item_recover(log, trans, item, pass);
 		switch (ITEM_TYPE(item)) {
 		case XFS_LI_BUF:
 			error = xlog_recover_do_buffer_trans(log, item, pass);
@@ -2919,8 +2982,9 @@ xlog_recover_process_data(
 				error = xlog_recover_unmount_trans(trans);
 				break;
 			case XLOG_WAS_CONT_TRANS:
-				error = xlog_recover_add_to_cont_trans(trans,
-						dp, be32_to_cpu(ohead->oh_len));
+				error = xlog_recover_add_to_cont_trans(log,
+						trans, dp,
+						be32_to_cpu(ohead->oh_len));
 				break;
 			case XLOG_START_TRANS:
 				xlog_warn(
@@ -2930,7 +2994,7 @@ xlog_recover_process_data(
 				break;
 			case 0:
 			case XLOG_CONTINUE_TRANS:
-				error = xlog_recover_add_to_trans(trans,
+				error = xlog_recover_add_to_trans(log, trans,
 						dp, be32_to_cpu(ohead->oh_len));
 				break;
 			default:
@@ -3331,42 +3395,6 @@ xlog_pack_data(
 	}
 }
 
-#if defined(DEBUG) && defined(XFS_LOUD_RECOVERY)
-STATIC void
-xlog_unpack_data_checksum(
-	xlog_rec_header_t	*rhead,
-	xfs_caddr_t		dp,
-	xlog_t			*log)
-{
-	__be32			*up = (__be32 *)dp;
-	uint			chksum = 0;
-	int			i;
-
-	/* divide length by 4 to get # words */
-	for (i=0; i < be32_to_cpu(rhead->h_len) >> 2; i++) {
-		chksum ^= be32_to_cpu(*up);
-		up++;
-	}
-	if (chksum != be32_to_cpu(rhead->h_chksum)) {
-	    if (rhead->h_chksum ||
-		((log->l_flags & XLOG_CHKSUM_MISMATCH) == 0)) {
-		    cmn_err(CE_DEBUG,
-			"XFS: LogR chksum mismatch: was (0x%x) is (0x%x)\n",
-			    be32_to_cpu(rhead->h_chksum), chksum);
-		    cmn_err(CE_DEBUG,
-"XFS: Disregard message if filesystem was created with non-DEBUG kernel");
-		    if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) {
-			    cmn_err(CE_DEBUG,
-				"XFS: LogR this is a LogV2 filesystem\n");
-		    }
-		    log->l_flags |= XLOG_CHKSUM_MISMATCH;
-	    }
-	}
-}
-#else
-#define xlog_unpack_data_checksum(rhead, dp, log)
-#endif
-
 STATIC void
 xlog_unpack_data(
 	xlog_rec_header_t	*rhead,
@@ -3390,8 +3418,6 @@ xlog_unpack_data(
 			dp += BBSIZE;
 		}
 	}
-
-	xlog_unpack_data_checksum(rhead, dp, log);
 }
 
 STATIC int
@@ -3490,7 +3516,7 @@ xlog_do_recovery_pass(
 			hblks = 1;
 		}
 	} else {
-		ASSERT(log->l_sectbb_log == 0);
+		ASSERT(log->l_sectBBsize == 1);
 		hblks = 1;
 		hbp = xlog_get_bp(log, 1);
 		h_size = XLOG_BIG_RECORD_BSIZE;
@@ -3946,10 +3972,6 @@ xlog_recover_check_summary(
 	xfs_agf_t	*agfp;
 	xfs_buf_t	*agfbp;
 	xfs_buf_t	*agibp;
-	xfs_buf_t	*sbbp;
-#ifdef XFS_LOUD_RECOVERY
-	xfs_sb_t	*sbp;
-#endif
 	xfs_agnumber_t	agno;
 	__uint64_t	freeblks;
 	__uint64_t	itotal;
@@ -3984,30 +4006,5 @@ xlog_recover_check_summary(
 			xfs_buf_relse(agibp);
 		}
 	}
-
-	sbbp = xfs_getsb(mp, 0);
-#ifdef XFS_LOUD_RECOVERY
-	sbp = &mp->m_sb;
-	xfs_sb_from_disk(sbp, XFS_BUF_TO_SBP(sbbp));
-	cmn_err(CE_NOTE,
-		"xlog_recover_check_summary: sb_icount %Lu itotal %Lu",
-		sbp->sb_icount, itotal);
-	cmn_err(CE_NOTE,
-		"xlog_recover_check_summary: sb_ifree %Lu itotal %Lu",
-		sbp->sb_ifree, ifree);
-	cmn_err(CE_NOTE,
-		"xlog_recover_check_summary: sb_fdblocks %Lu freeblks %Lu",
-		sbp->sb_fdblocks, freeblks);
-#if 0
-	/*
-	 * This is turned off until I account for the allocation
-	 * btree blocks which live in free space.
-	 */
-	ASSERT(sbp->sb_icount == itotal);
-	ASSERT(sbp->sb_ifree == ifree);
-	ASSERT(sbp->sb_fdblocks == freeblks);
-#endif
-#endif
-	xfs_buf_relse(sbbp);
 }
 #endif /* DEBUG */
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index e79b56b4bca..d7bf38c8cd1 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -1405,13 +1405,6 @@ xfs_mountfs(
 		xfs_qm_mount_quotas(mp);
 	}
 
-#if defined(DEBUG) && defined(XFS_LOUD_RECOVERY)
-	if (XFS_IS_QUOTA_ON(mp))
-		xfs_fs_cmn_err(CE_NOTE, mp, "Disk quotas turned on");
-	else
-		xfs_fs_cmn_err(CE_NOTE, mp, "Disk quotas not turned on");
-#endif
-
 	/*
 	 * Now we are mounted, reserve a small amount of unused space for
 	 * privileged transactions. This is needed so that transaction
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index 4fa0bc7b983..9ff48a16a7e 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -259,6 +259,7 @@ typedef struct xfs_mount {
 	wait_queue_head_t	m_wait_single_sync_task;
 	__int64_t		m_update_flags;	/* sb flags we need to update
 						   on the next remount,rw */
+	struct list_head	m_mplist;	/* inode shrinker mount list */
 } xfs_mount_t;
 
 /*
diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h
index fdcab3f81dd..e0e64b113bd 100644
--- a/fs/xfs/xfs_quota.h
+++ b/fs/xfs/xfs_quota.h
@@ -201,9 +201,6 @@ typedef struct xfs_qoff_logformat {
 #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_QUOTAOFF	0x0000080 /* quotas are being turned off */
-#define XFS_QMOPT_UMOUNTING	0x0000100 /* filesys is being unmounted */
-#define XFS_QMOPT_DOLOG		0x0000200 /* log buf changes (in quotacheck) */
 #define XFS_QMOPT_DOWARN        0x0000400 /* increase warning cnt if needed */
 #define XFS_QMOPT_DQREPAIR	0x0001000 /* repair dquot if damaged */
 #define XFS_QMOPT_GQUOTA	0x0002000 /* group dquot requested */
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index f73e358bae8..be578ecb4af 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -45,23 +45,12 @@
 #include "xfs_trans_space.h"
 #include "xfs_inode_item.h"
 
-
-STATIC void	xfs_trans_apply_sb_deltas(xfs_trans_t *);
-STATIC uint	xfs_trans_count_vecs(xfs_trans_t *);
-STATIC void	xfs_trans_fill_vecs(xfs_trans_t *, xfs_log_iovec_t *);
-STATIC void	xfs_trans_uncommit(xfs_trans_t *, uint);
-STATIC void	xfs_trans_committed(xfs_trans_t *, int);
-STATIC void	xfs_trans_chunk_committed(xfs_log_item_chunk_t *, xfs_lsn_t, int);
-STATIC void	xfs_trans_free(xfs_trans_t *);
-
 kmem_zone_t	*xfs_trans_zone;
 
-
 /*
  * Reservation functions here avoid a huge stack in xfs_trans_init
  * due to register overflow from temporaries in the calculations.
  */
-
 STATIC uint
 xfs_calc_write_reservation(xfs_mount_t *mp)
 {
@@ -261,6 +250,19 @@ _xfs_trans_alloc(
 }
 
 /*
+ * Free the transaction structure.  If there is more clean up
+ * to do when the structure is freed, add it here.
+ */
+STATIC void
+xfs_trans_free(
+	xfs_trans_t	*tp)
+{
+	atomic_dec(&tp->t_mountp->m_active_trans);
+	xfs_trans_free_dqinfo(tp);
+	kmem_zone_free(xfs_trans_zone, tp);
+}
+
+/*
  * This is called to create a new transaction which will share the
  * permanent log reservation of the given transaction.  The remaining
  * unused block and rt extent reservations are also inherited.  This
@@ -764,94 +766,278 @@ xfs_trans_unreserve_and_mod_sb(
 	}
 }
 
+/*
+ * 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;
+	xfs_log_item_desc_t	*lidp;
+
+	nvecs = 1;
+	lidp = xfs_trans_first_item(tp);
+	ASSERT(lidp != NULL);
+
+	/* 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 (lidp == NULL)
+		return 0;
+
+	while (lidp != NULL) {
+		/*
+		 * Skip items which aren't dirty in this transaction.
+		 */
+		if (!(lidp->lid_flags & XFS_LID_DIRTY)) {
+			lidp = xfs_trans_next_item(tp, lidp);
+			continue;
+		}
+		lidp->lid_size = IOP_SIZE(lidp->lid_item);
+		nvecs += lidp->lid_size;
+		lidp = xfs_trans_next_item(tp, lidp);
+	}
+
+	return nvecs;
+}
 
 /*
- * xfs_trans_commit
+ * 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().
  *
- * Commit the given transaction to the log a/synchronously.
+ * 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)
+{
+	xfs_log_item_desc_t	*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;
+	lidp = xfs_trans_first_item(tp);
+	ASSERT(lidp);
+	while (lidp) {
+		/* Skip items which aren't dirty in this transaction. */
+		if (!(lidp->lid_flags & XFS_LID_DIRTY)) {
+			lidp = xfs_trans_next_item(tp, lidp);
+			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);
+		lidp = xfs_trans_next_item(tp, lidp);
+	}
+
+	/*
+	 * 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.
  *
- * XFS disk error handling mechanism is not based on a typical
- * transaction abort mechanism. Logically after the filesystem
- * gets marked 'SHUTDOWN', we can't let any new transactions
- * be durable - ie. committed to disk - because some metadata might
- * be inconsistent. In such cases, this returns an error, and the
- * caller may assume that all locked objects joined to the transaction
- * have already been unlocked as if the commit had succeeded.
- * Do not reference the transaction structure after this call.
+ * 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.
  */
- /*ARGSUSED*/
-int
-_xfs_trans_commit(
-	xfs_trans_t	*tp,
-	uint		flags,
-	int		*log_flushed)
+static void
+xfs_trans_item_committed(
+	struct xfs_log_item	*lip,
+	xfs_lsn_t		commit_lsn,
+	int			aborted)
 {
-	xfs_log_iovec_t		*log_vector;
-	int			nvec;
-	xfs_mount_t		*mp;
-	xfs_lsn_t		commit_lsn;
-	/* REFERENCED */
-	int			error;
-	int			log_flags;
-	int			sync;
-#define	XFS_TRANS_LOGVEC_COUNT	16
-	xfs_log_iovec_t		log_vector_fast[XFS_TRANS_LOGVEC_COUNT];
-	struct xlog_in_core	*commit_iclog;
-	int			shutdown;
+	xfs_lsn_t		item_lsn;
+	struct xfs_ail		*ailp;
 
-	commit_lsn = -1;
+	if (aborted)
+		lip->li_flags |= XFS_LI_ABORTED;
+	item_lsn = IOP_COMMITTED(lip, commit_lsn);
+
+	/* If the committed routine returns -1, item has been freed. */
+	if (XFS_LSN_CMP(item_lsn, (xfs_lsn_t)-1) == 0)
+		return;
 
 	/*
-	 * Determine whether this commit is releasing a permanent
-	 * log reservation or not.
+	 * 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.
 	 */
-	if (flags & XFS_TRANS_RELEASE_LOG_RES) {
-		ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
-		log_flags = XFS_LOG_REL_PERM_RESERV;
+	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.
+		 */
+		xfs_trans_ail_update(ailp, lip, item_lsn);
 	} else {
-		log_flags = 0;
+		spin_unlock(&ailp->xa_lock);
 	}
-	mp = tp->t_mountp;
 
 	/*
-	 * If there is nothing to be logged by the transaction,
-	 * then unlock all of the items associated with the
-	 * transaction and free the transaction structure.
-	 * Also make sure to return any reserved blocks to
-	 * the free pool.
+	 * Now that we've repositioned the item in the AIL, unpin it so it can
+	 * be flushed. Pass information about buffer stale state down from the
+	 * log item flags, if anyone else stales the buffer we do not want to
+	 * pay any attention to it.
 	 */
-shut_us_down:
-	shutdown = XFS_FORCED_SHUTDOWN(mp) ? EIO : 0;
-	if (!(tp->t_flags & XFS_TRANS_DIRTY) || shutdown) {
-		xfs_trans_unreserve_and_mod_sb(tp);
+	IOP_UNPIN(lip);
+}
+
+/* Clear all the per-AG busy list items listed in this transaction */
+static void
+xfs_trans_clear_busy_extents(
+	struct xfs_trans	*tp)
+{
+	xfs_log_busy_chunk_t	*lbcp;
+	xfs_log_busy_slot_t	*lbsp;
+	int			i;
+
+	for (lbcp = &tp->t_busy; lbcp != NULL; lbcp = lbcp->lbc_next) {
+		i = 0;
+		for (lbsp = lbcp->lbc_busy; i < lbcp->lbc_unused; i++, lbsp++) {
+			if (XFS_LBC_ISFREE(lbcp, i))
+				continue;
+			xfs_alloc_clear_busy(tp, lbsp->lbc_ag, lbsp->lbc_idx);
+		}
+	}
+	xfs_trans_free_busy(tp);
+}
+
+/*
+ * This is typically called by the LM when a transaction has been fully
+ * committed to disk.  It needs to unpin the items which have
+ * been logged by the transaction and update their positions
+ * in the AIL if necessary.
+ *
+ * This also gets called when the transactions didn't get written out
+ * because of an I/O error. Abortflag & XFS_LI_ABORTED is set then.
+ */
+STATIC void
+xfs_trans_committed(
+	struct xfs_trans	*tp,
+	int			abortflag)
+{
+	xfs_log_item_desc_t	*lidp;
+	xfs_log_item_chunk_t	*licp;
+	xfs_log_item_chunk_t	*next_licp;
+
+	/* Call the transaction's completion callback if there is one. */
+	if (tp->t_callback != NULL)
+		tp->t_callback(tp, tp->t_callarg);
+
+	for (lidp = xfs_trans_first_item(tp);
+	     lidp != NULL;
+	     lidp = xfs_trans_next_item(tp, lidp)) {
+		xfs_trans_item_committed(lidp->lid_item, tp->t_lsn, abortflag);
+	}
+
+	/* free the item chunks, ignoring the embedded chunk */
+	for (licp = tp->t_items.lic_next; licp != NULL; licp = next_licp) {
+		next_licp = licp->lic_next;
+		kmem_free(licp);
+	}
+
+	xfs_trans_clear_busy_extents(tp);
+	xfs_trans_free(tp);
+}
+
+/*
+ * Called from the trans_commit code when we notice that
+ * the filesystem is in the middle of a forced shutdown.
+ */
+STATIC void
+xfs_trans_uncommit(
+	struct xfs_trans	*tp,
+	uint			flags)
+{
+	xfs_log_item_desc_t	*lidp;
+
+	for (lidp = xfs_trans_first_item(tp);
+	     lidp != NULL;
+	     lidp = xfs_trans_next_item(tp, lidp)) {
 		/*
-		 * It is indeed possible for the transaction to be
-		 * not dirty but the dqinfo portion to be. All that
-		 * means is that we have some (non-persistent) quota
-		 * reservations that need to be unreserved.
+		 * Unpin all but those that aren't dirty.
 		 */
-		xfs_trans_unreserve_and_mod_dquots(tp);
-		if (tp->t_ticket) {
-			commit_lsn = xfs_log_done(mp, tp->t_ticket,
-							NULL, log_flags);
-			if (commit_lsn == -1 && !shutdown)
-				shutdown = XFS_ERROR(EIO);
-		}
-		current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
-		xfs_trans_free_items(tp, shutdown? XFS_TRANS_ABORT : 0);
-		xfs_trans_free_busy(tp);
-		xfs_trans_free(tp);
-		XFS_STATS_INC(xs_trans_empty);
-		return (shutdown);
+		if (lidp->lid_flags & XFS_LID_DIRTY)
+			IOP_UNPIN_REMOVE(lidp->lid_item, tp);
 	}
-	ASSERT(tp->t_ticket != NULL);
 
-	/*
-	 * If we need to update the superblock, then do it now.
-	 */
-	if (tp->t_flags & XFS_TRANS_SB_DIRTY)
-		xfs_trans_apply_sb_deltas(tp);
-	xfs_trans_apply_dquot_deltas(tp);
+	xfs_trans_unreserve_and_mod_sb(tp);
+	xfs_trans_unreserve_and_mod_dquots(tp);
+
+	xfs_trans_free_items(tp, flags);
+	xfs_trans_free_busy(tp);
+	xfs_trans_free(tp);
+}
+
+/*
+ * Format the transaction direct to the iclog. This isolates the physical
+ * transaction commit operation from the logical operation and hence allows
+ * other methods to be introduced without affecting the existing commit path.
+ */
+static int
+xfs_trans_commit_iclog(
+	struct xfs_mount	*mp,
+	struct xfs_trans	*tp,
+	xfs_lsn_t		*commit_lsn,
+	int			flags)
+{
+	int			shutdown;
+	int			error;
+	int			log_flags = 0;
+	struct xlog_in_core	*commit_iclog;
+#define XFS_TRANS_LOGVEC_COUNT  16
+	struct xfs_log_iovec	log_vector_fast[XFS_TRANS_LOGVEC_COUNT];
+	struct xfs_log_iovec	*log_vector;
+	uint			nvec;
+
 
 	/*
 	 * Ask each log item how many log_vector entries it will
@@ -861,8 +1047,7 @@ shut_us_down:
 	 */
 	nvec = xfs_trans_count_vecs(tp);
 	if (nvec == 0) {
-		xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
-		goto shut_us_down;
+		return ENOMEM;	/* triggers a shutdown! */
 	} else if (nvec <= XFS_TRANS_LOGVEC_COUNT) {
 		log_vector = log_vector_fast;
 	} else {
@@ -877,6 +1062,9 @@ shut_us_down:
 	 */
 	xfs_trans_fill_vecs(tp, log_vector);
 
+	if (flags & XFS_TRANS_RELEASE_LOG_RES)
+		log_flags = XFS_LOG_REL_PERM_RESERV;
+
 	error = xfs_log_write(mp, log_vector, nvec, tp->t_ticket, &(tp->t_lsn));
 
 	/*
@@ -884,18 +1072,17 @@ shut_us_down:
 	 * at any time after this call.  However, all the items associated
 	 * with the transaction are still locked and pinned in memory.
 	 */
-	commit_lsn = xfs_log_done(mp, tp->t_ticket, &commit_iclog, log_flags);
+	*commit_lsn = xfs_log_done(mp, tp->t_ticket, &commit_iclog, log_flags);
 
-	tp->t_commit_lsn = commit_lsn;
-	if (nvec > XFS_TRANS_LOGVEC_COUNT) {
+	tp->t_commit_lsn = *commit_lsn;
+	if (nvec > XFS_TRANS_LOGVEC_COUNT)
 		kmem_free(log_vector);
-	}
 
 	/*
 	 * If we got a log write error. Unpin the logitems that we
 	 * had pinned, clean up, free trans structure, and return error.
 	 */
-	if (error || commit_lsn == -1) {
+	if (error || *commit_lsn == -1) {
 		current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
 		xfs_trans_uncommit(tp, flags|XFS_TRANS_ABORT);
 		return XFS_ERROR(EIO);
@@ -909,8 +1096,6 @@ shut_us_down:
 	 */
 	xfs_trans_unreserve_and_mod_sb(tp);
 
-	sync = tp->t_flags & XFS_TRANS_SYNC;
-
 	/*
 	 * Tell the LM to call the transaction completion routine
 	 * when the log write with LSN commit_lsn completes (e.g.
@@ -953,7 +1138,7 @@ shut_us_down:
 	 * the commit lsn of this transaction for dependency tracking
 	 * purposes.
 	 */
-	xfs_trans_unlock_items(tp, commit_lsn);
+	xfs_trans_unlock_items(tp, *commit_lsn);
 
 	/*
 	 * If we detected a log error earlier, finish committing
@@ -973,156 +1158,114 @@ shut_us_down:
 	 * and the items are released we can finally allow the iclog to
 	 * go to disk.
 	 */
-	error = xfs_log_release_iclog(mp, commit_iclog);
-
-	/*
-	 * If the transaction needs to be synchronous, then force the
-	 * log out now and wait for it.
-	 */
-	if (sync) {
-		if (!error) {
-			error = _xfs_log_force_lsn(mp, commit_lsn,
-				      XFS_LOG_SYNC, log_flushed);
-		}
-		XFS_STATS_INC(xs_trans_sync);
-	} else {
-		XFS_STATS_INC(xs_trans_async);
-	}
-
-	return (error);
+	return xfs_log_release_iclog(mp, commit_iclog);
 }
 
 
 /*
- * 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.
+ * xfs_trans_commit
+ *
+ * Commit the given transaction to the log a/synchronously.
+ *
+ * XFS disk error handling mechanism is not based on a typical
+ * transaction abort mechanism. Logically after the filesystem
+ * gets marked 'SHUTDOWN', we can't let any new transactions
+ * be durable - ie. committed to disk - because some metadata might
+ * be inconsistent. In such cases, this returns an error, and the
+ * caller may assume that all locked objects joined to the transaction
+ * have already been unlocked as if the commit had succeeded.
+ * Do not reference the transaction structure after this call.
  */
-STATIC uint
-xfs_trans_count_vecs(
-	xfs_trans_t	*tp)
+int
+_xfs_trans_commit(
+	struct xfs_trans	*tp,
+	uint			flags,
+	int			*log_flushed)
 {
-	int			nvecs;
-	xfs_log_item_desc_t	*lidp;
+	struct xfs_mount	*mp = tp->t_mountp;
+	xfs_lsn_t		commit_lsn = -1;
+	int			error = 0;
+	int			log_flags = 0;
+	int			sync = tp->t_flags & XFS_TRANS_SYNC;
 
-	nvecs = 1;
-	lidp = xfs_trans_first_item(tp);
-	ASSERT(lidp != NULL);
-
-	/* 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.
+	/*
+	 * Determine whether this commit is releasing a permanent
+	 * log reservation or not.
 	 */
-	if (lidp == NULL)
-		return 0;
-
-	while (lidp != NULL) {
-		/*
-		 * Skip items which aren't dirty in this transaction.
-		 */
-		if (!(lidp->lid_flags & XFS_LID_DIRTY)) {
-			lidp = xfs_trans_next_item(tp, lidp);
-			continue;
-		}
-		lidp->lid_size = IOP_SIZE(lidp->lid_item);
-		nvecs += lidp->lid_size;
-		lidp = xfs_trans_next_item(tp, lidp);
+	if (flags & XFS_TRANS_RELEASE_LOG_RES) {
+		ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
+		log_flags = XFS_LOG_REL_PERM_RESERV;
 	}
 
-	return nvecs;
-}
-
-/*
- * Called from the trans_commit code when we notice that
- * the filesystem is in the middle of a forced shutdown.
- */
-STATIC void
-xfs_trans_uncommit(
-	xfs_trans_t	*tp,
-	uint		flags)
-{
-	xfs_log_item_desc_t	*lidp;
+	/*
+	 * If there is nothing to be logged by the transaction,
+	 * then unlock all of the items associated with the
+	 * transaction and free the transaction structure.
+	 * Also make sure to return any reserved blocks to
+	 * the free pool.
+	 */
+	if (!(tp->t_flags & XFS_TRANS_DIRTY))
+		goto out_unreserve;
 
-	for (lidp = xfs_trans_first_item(tp);
-	     lidp != NULL;
-	     lidp = xfs_trans_next_item(tp, lidp)) {
-		/*
-		 * Unpin all but those that aren't dirty.
-		 */
-		if (lidp->lid_flags & XFS_LID_DIRTY)
-			IOP_UNPIN_REMOVE(lidp->lid_item, tp);
+	if (XFS_FORCED_SHUTDOWN(mp)) {
+		error = XFS_ERROR(EIO);
+		goto out_unreserve;
 	}
 
-	xfs_trans_unreserve_and_mod_sb(tp);
-	xfs_trans_unreserve_and_mod_dquots(tp);
+	ASSERT(tp->t_ticket != NULL);
 
-	xfs_trans_free_items(tp, flags);
-	xfs_trans_free_busy(tp);
-	xfs_trans_free(tp);
-}
+	/*
+	 * If we need to update the superblock, then do it now.
+	 */
+	if (tp->t_flags & XFS_TRANS_SB_DIRTY)
+		xfs_trans_apply_sb_deltas(tp);
+	xfs_trans_apply_dquot_deltas(tp);
 
-/*
- * 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(
-	xfs_trans_t		*tp,
-	xfs_log_iovec_t		*log_vector)
-{
-	xfs_log_item_desc_t	*lidp;
-	xfs_log_iovec_t		*vecp;
-	uint			nitems;
+	error = xfs_trans_commit_iclog(mp, tp, &commit_lsn, flags);
+	if (error == ENOMEM) {
+		xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
+		error = XFS_ERROR(EIO);
+		goto out_unreserve;
+	}
 
 	/*
-	 * Skip over the entry for the transaction header, we'll
-	 * fill that in at the end.
+	 * If the transaction needs to be synchronous, then force the
+	 * log out now and wait for it.
 	 */
-	vecp = log_vector + 1;		/* pointer arithmetic */
-
-	nitems = 0;
-	lidp = xfs_trans_first_item(tp);
-	ASSERT(lidp != NULL);
-	while (lidp != NULL) {
-		/*
-		 * Skip items which aren't dirty in this transaction.
-		 */
-		if (!(lidp->lid_flags & XFS_LID_DIRTY)) {
-			lidp = xfs_trans_next_item(tp, lidp);
-			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++;
+	if (sync) {
+		if (!error) {
+			error = _xfs_log_force_lsn(mp, commit_lsn,
+				      XFS_LOG_SYNC, log_flushed);
 		}
-		IOP_FORMAT(lidp->lid_item, vecp);
-		vecp += lidp->lid_size;		/* pointer arithmetic */
-		IOP_PIN(lidp->lid_item);
-		lidp = xfs_trans_next_item(tp, lidp);
+		XFS_STATS_INC(xs_trans_sync);
+	} else {
+		XFS_STATS_INC(xs_trans_async);
 	}
 
+	return error;
+
+out_unreserve:
+	xfs_trans_unreserve_and_mod_sb(tp);
+
 	/*
-	 * Now that we've counted the number of items in this
-	 * transaction, fill in the transaction header.
+	 * It is indeed possible for the transaction to be not dirty but
+	 * the dqinfo portion to be.  All that means is that we have some
+	 * (non-persistent) quota reservations that need to be unreserved.
 	 */
-	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;
-}
+	xfs_trans_unreserve_and_mod_dquots(tp);
+	if (tp->t_ticket) {
+		commit_lsn = xfs_log_done(mp, tp->t_ticket, NULL, log_flags);
+		if (commit_lsn == -1 && !error)
+			error = XFS_ERROR(EIO);
+	}
+	current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
+	xfs_trans_free_items(tp, error ? XFS_TRANS_ABORT : 0);
+	xfs_trans_free_busy(tp);
+	xfs_trans_free(tp);
 
+	XFS_STATS_INC(xs_trans_empty);
+	return error;
+}
 
 /*
  * Unlock all of the transaction's items and free the transaction.
@@ -1200,20 +1343,6 @@ xfs_trans_cancel(
 	xfs_trans_free(tp);
 }
 
-
-/*
- * Free the transaction structure.  If there is more clean up
- * to do when the structure is freed, add it here.
- */
-STATIC void
-xfs_trans_free(
-	xfs_trans_t	*tp)
-{
-	atomic_dec(&tp->t_mountp->m_active_trans);
-	xfs_trans_free_dqinfo(tp);
-	kmem_zone_free(xfs_trans_zone, tp);
-}
-
 /*
  * Roll from one trans in the sequence of PERMANENT transactions to
  * the next: permanent transactions are only flushed out when
@@ -1283,174 +1412,3 @@ xfs_trans_roll(
 	xfs_trans_ihold(trans, dp);
 	return 0;
 }
-
-/*
- * THIS SHOULD BE REWRITTEN TO USE xfs_trans_next_item().
- *
- * This is typically called by the LM when a transaction has been fully
- * committed to disk.  It needs to unpin the items which have
- * been logged by the transaction and update their positions
- * in the AIL if necessary.
- * This also gets called when the transactions didn't get written out
- * because of an I/O error. Abortflag & XFS_LI_ABORTED is set then.
- *
- * Call xfs_trans_chunk_committed() to process the items in
- * each chunk.
- */
-STATIC void
-xfs_trans_committed(
-	xfs_trans_t	*tp,
-	int		abortflag)
-{
-	xfs_log_item_chunk_t	*licp;
-	xfs_log_item_chunk_t	*next_licp;
-	xfs_log_busy_chunk_t	*lbcp;
-	xfs_log_busy_slot_t	*lbsp;
-	int			i;
-
-	/*
-	 * Call the transaction's completion callback if there
-	 * is one.
-	 */
-	if (tp->t_callback != NULL) {
-		tp->t_callback(tp, tp->t_callarg);
-	}
-
-	/*
-	 * Special case the chunk embedded in the transaction.
-	 */
-	licp = &(tp->t_items);
-	if (!(xfs_lic_are_all_free(licp))) {
-		xfs_trans_chunk_committed(licp, tp->t_lsn, abortflag);
-	}
-
-	/*
-	 * Process the items in each chunk in turn.
-	 */
-	licp = licp->lic_next;
-	while (licp != NULL) {
-		ASSERT(!xfs_lic_are_all_free(licp));
-		xfs_trans_chunk_committed(licp, tp->t_lsn, abortflag);
-		next_licp = licp->lic_next;
-		kmem_free(licp);
-		licp = next_licp;
-	}
-
-	/*
-	 * Clear all the per-AG busy list items listed in this transaction
-	 */
-	lbcp = &tp->t_busy;
-	while (lbcp != NULL) {
-		for (i = 0, lbsp = lbcp->lbc_busy; i < lbcp->lbc_unused; i++, lbsp++) {
-			if (!XFS_LBC_ISFREE(lbcp, i)) {
-				xfs_alloc_clear_busy(tp, lbsp->lbc_ag,
-						     lbsp->lbc_idx);
-			}
-		}
-		lbcp = lbcp->lbc_next;
-	}
-	xfs_trans_free_busy(tp);
-
-	/*
-	 * That's it for the transaction structure.  Free it.
-	 */
-	xfs_trans_free(tp);
-}
-
-/*
- * This is called to perform the commit processing for each
- * item described by the given chunk.
- *
- * The commit processing consists of unlocking items which were
- * held locked with the SYNC_UNLOCK attribute, 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_chunk_committed(
-	xfs_log_item_chunk_t	*licp,
-	xfs_lsn_t		lsn,
-	int			aborted)
-{
-	xfs_log_item_desc_t	*lidp;
-	xfs_log_item_t		*lip;
-	xfs_lsn_t		item_lsn;
-	int			i;
-
-	lidp = licp->lic_descs;
-	for (i = 0; i < licp->lic_unused; i++, lidp++) {
-		struct xfs_ail		*ailp;
-
-		if (xfs_lic_isfree(licp, i)) {
-			continue;
-		}
-
-		lip = lidp->lid_item;
-		if (aborted)
-			lip->li_flags |= XFS_LI_ABORTED;
-
-		/*
-		 * Send in the ABORTED flag to the COMMITTED routine
-		 * so that it knows whether the transaction was aborted
-		 * or not.
-		 */
-		item_lsn = IOP_COMMITTED(lip, lsn);
-
-		/*
-		 * If the committed routine returns -1, make
-		 * no more references to the item.
-		 */
-		if (XFS_LSN_CMP(item_lsn, (xfs_lsn_t)-1) == 0) {
-			continue;
-		}
-
-		/*
-		 * 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.
-			 */
-			xfs_trans_ail_update(ailp, lip, item_lsn);
-		} else {
-			spin_unlock(&ailp->xa_lock);
-		}
-
-		/*
-		 * Now that we've repositioned the item in the AIL,
-		 * unpin it so it can be flushed. Pass information
-		 * about buffer stale state down from the log item
-		 * flags, if anyone else stales the buffer we do not
-		 * want to pay any attention to it.
-		 */
-		IOP_UNPIN(lip, lidp->lid_flags & XFS_LID_BUF_STALE);
-	}
-}
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index 79c8bab9dff..c62beee0921 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -49,6 +49,15 @@ typedef struct xfs_trans_header {
 #define	XFS_LI_DQUOT		0x123d
 #define	XFS_LI_QUOTAOFF		0x123e
 
+#define XFS_LI_TYPE_DESC \
+	{ XFS_LI_EFI,		"XFS_LI_EFI" }, \
+	{ XFS_LI_EFD,		"XFS_LI_EFD" }, \
+	{ XFS_LI_IUNLINK,	"XFS_LI_IUNLINK" }, \
+	{ XFS_LI_INODE,		"XFS_LI_INODE" }, \
+	{ XFS_LI_BUF,		"XFS_LI_BUF" }, \
+	{ XFS_LI_DQUOT,		"XFS_LI_DQUOT" }, \
+	{ XFS_LI_QUOTAOFF,	"XFS_LI_QUOTAOFF" }
+
 /*
  * Transaction types.  Used to distinguish types of buffers.
  */
@@ -159,7 +168,6 @@ typedef struct xfs_log_item_desc {
 
 #define XFS_LID_DIRTY		0x1
 #define XFS_LID_PINNED		0x2
-#define XFS_LID_BUF_STALE	0x8
 
 /*
  * This structure is used to maintain a chunk list of log_item_desc
@@ -833,7 +841,7 @@ typedef struct xfs_item_ops {
 	uint (*iop_size)(xfs_log_item_t *);
 	void (*iop_format)(xfs_log_item_t *, struct xfs_log_iovec *);
 	void (*iop_pin)(xfs_log_item_t *);
-	void (*iop_unpin)(xfs_log_item_t *, int);
+	void (*iop_unpin)(xfs_log_item_t *);
 	void (*iop_unpin_remove)(xfs_log_item_t *, struct xfs_trans *);
 	uint (*iop_trylock)(xfs_log_item_t *);
 	void (*iop_unlock)(xfs_log_item_t *);
@@ -846,7 +854,7 @@ typedef struct xfs_item_ops {
 #define IOP_SIZE(ip)		(*(ip)->li_ops->iop_size)(ip)
 #define IOP_FORMAT(ip,vp)	(*(ip)->li_ops->iop_format)(ip, vp)
 #define IOP_PIN(ip)		(*(ip)->li_ops->iop_pin)(ip)
-#define IOP_UNPIN(ip, flags)	(*(ip)->li_ops->iop_unpin)(ip, flags)
+#define IOP_UNPIN(ip)		(*(ip)->li_ops->iop_unpin)(ip)
 #define IOP_UNPIN_REMOVE(ip,tp) (*(ip)->li_ops->iop_unpin_remove)(ip, tp)
 #define IOP_TRYLOCK(ip)		(*(ip)->li_ops->iop_trylock)(ip)
 #define IOP_UNLOCK(ip)		(*(ip)->li_ops->iop_unlock)(ip)
diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c
index fb586360d1c..9cd809025f3 100644
--- a/fs/xfs/xfs_trans_buf.c
+++ b/fs/xfs/xfs_trans_buf.c
@@ -40,11 +40,51 @@
 #include "xfs_rw.h"
 #include "xfs_trace.h"
 
+/*
+ * Check to see if a buffer matching the given parameters is already
+ * a part of the given transaction.
+ */
+STATIC struct xfs_buf *
+xfs_trans_buf_item_match(
+	struct xfs_trans	*tp,
+	struct xfs_buftarg	*target,
+	xfs_daddr_t		blkno,
+	int			len)
+{
+	xfs_log_item_chunk_t	*licp;
+	xfs_log_item_desc_t	*lidp;
+	xfs_buf_log_item_t	*blip;
+	int			i;
 
-STATIC xfs_buf_t *xfs_trans_buf_item_match(xfs_trans_t *, xfs_buftarg_t *,
-		xfs_daddr_t, int);
-STATIC xfs_buf_t *xfs_trans_buf_item_match_all(xfs_trans_t *, xfs_buftarg_t *,
-		xfs_daddr_t, int);
+	len = BBTOB(len);
+	for (licp = &tp->t_items; licp != NULL; licp = licp->lic_next) {
+		if (xfs_lic_are_all_free(licp)) {
+			ASSERT(licp == &tp->t_items);
+			ASSERT(licp->lic_next == NULL);
+			return NULL;
+		}
+
+		for (i = 0; i < licp->lic_unused; i++) {
+			/*
+			 * Skip unoccupied slots.
+			 */
+			if (xfs_lic_isfree(licp, i))
+				continue;
+
+			lidp = xfs_lic_slot(licp, i);
+			blip = (xfs_buf_log_item_t *)lidp->lid_item;
+			if (blip->bli_item.li_type != XFS_LI_BUF)
+				continue;
+
+			if (XFS_BUF_TARGET(blip->bli_buf) == target &&
+			    XFS_BUF_ADDR(blip->bli_buf) == blkno &&
+			    XFS_BUF_COUNT(blip->bli_buf) == len)
+				return blip->bli_buf;
+		}
+	}
+
+	return NULL;
+}
 
 /*
  * Add the locked buffer to the transaction.
@@ -112,14 +152,6 @@ xfs_trans_bjoin(
  * within the transaction, just increment its lock recursion count
  * and return a pointer to it.
  *
- * Use the fast path function xfs_trans_buf_item_match() or the buffer
- * cache routine incore_match() to find the buffer
- * if it is already owned by this transaction.
- *
- * If we don't already own the buffer, use get_buf() to get it.
- * If it doesn't yet have an associated xfs_buf_log_item structure,
- * then allocate one and add the item to this transaction.
- *
  * If the transaction pointer is NULL, make this just a normal
  * get_buf() call.
  */
@@ -149,11 +181,7 @@ xfs_trans_get_buf(xfs_trans_t	*tp,
 	 * have it locked.  In this case we just increment the lock
 	 * recursion count and return the buffer to the caller.
 	 */
-	if (tp->t_items.lic_next == NULL) {
-		bp = xfs_trans_buf_item_match(tp, target_dev, blkno, len);
-	} else {
-		bp  = xfs_trans_buf_item_match_all(tp, target_dev, blkno, len);
-	}
+	bp = xfs_trans_buf_item_match(tp, target_dev, blkno, len);
 	if (bp != NULL) {
 		ASSERT(XFS_BUF_VALUSEMA(bp) <= 0);
 		if (XFS_FORCED_SHUTDOWN(tp->t_mountp))
@@ -259,14 +287,6 @@ int	xfs_error_mod = 33;
  * within the transaction and already read in, just increment its
  * lock recursion count and return a pointer to it.
  *
- * Use the fast path function xfs_trans_buf_item_match() or the buffer
- * cache routine incore_match() to find the buffer
- * if it is already owned by this transaction.
- *
- * If we don't already own the buffer, use read_buf() to get it.
- * If it doesn't yet have an associated xfs_buf_log_item structure,
- * then allocate one and add the item to this transaction.
- *
  * If the transaction pointer is NULL, make this just a normal
  * read_buf() call.
  */
@@ -328,11 +348,7 @@ xfs_trans_read_buf(
 	 * If the buffer is not yet read in, then we read it in, increment
 	 * the lock recursion count, and return it to the caller.
 	 */
-	if (tp->t_items.lic_next == NULL) {
-		bp = xfs_trans_buf_item_match(tp, target, blkno, len);
-	} else {
-		bp = xfs_trans_buf_item_match_all(tp, target, blkno, len);
-	}
+	bp = xfs_trans_buf_item_match(tp, target, blkno, len);
 	if (bp != NULL) {
 		ASSERT(XFS_BUF_VALUSEMA(bp) <= 0);
 		ASSERT(XFS_BUF_FSPRIVATE2(bp, xfs_trans_t *) == tp);
@@ -696,7 +712,6 @@ xfs_trans_log_buf(xfs_trans_t	*tp,
 
 	tp->t_flags |= XFS_TRANS_DIRTY;
 	lidp->lid_flags |= XFS_LID_DIRTY;
-	lidp->lid_flags &= ~XFS_LID_BUF_STALE;
 	bip->bli_flags |= XFS_BLI_LOGGED;
 	xfs_buf_item_log(bip, first, last);
 }
@@ -782,7 +797,7 @@ xfs_trans_binval(
 	bip->bli_format.blf_flags |= XFS_BLI_CANCEL;
 	memset((char *)(bip->bli_format.blf_data_map), 0,
 	      (bip->bli_format.blf_map_size * sizeof(uint)));
-	lidp->lid_flags |= XFS_LID_DIRTY|XFS_LID_BUF_STALE;
+	lidp->lid_flags |= XFS_LID_DIRTY;
 	tp->t_flags |= XFS_TRANS_DIRTY;
 }
 
@@ -902,111 +917,3 @@ xfs_trans_dquot_buf(
 
 	bip->bli_format.blf_flags |= type;
 }
-
-/*
- * Check to see if a buffer matching the given parameters is already
- * a part of the given transaction.  Only check the first, embedded
- * chunk, since we don't want to spend all day scanning large transactions.
- */
-STATIC xfs_buf_t *
-xfs_trans_buf_item_match(
-	xfs_trans_t	*tp,
-	xfs_buftarg_t	*target,
-	xfs_daddr_t	blkno,
-	int		len)
-{
-	xfs_log_item_chunk_t	*licp;
-	xfs_log_item_desc_t	*lidp;
-	xfs_buf_log_item_t	*blip;
-	xfs_buf_t		*bp;
-	int			i;
-
-	bp = NULL;
-	len = BBTOB(len);
-	licp = &tp->t_items;
-	if (!xfs_lic_are_all_free(licp)) {
-		for (i = 0; i < licp->lic_unused; i++) {
-			/*
-			 * Skip unoccupied slots.
-			 */
-			if (xfs_lic_isfree(licp, i)) {
-				continue;
-			}
-
-			lidp = xfs_lic_slot(licp, i);
-			blip = (xfs_buf_log_item_t *)lidp->lid_item;
-			if (blip->bli_item.li_type != XFS_LI_BUF) {
-				continue;
-			}
-
-			bp = blip->bli_buf;
-			if ((XFS_BUF_TARGET(bp) == target) &&
-			    (XFS_BUF_ADDR(bp) == blkno) &&
-			    (XFS_BUF_COUNT(bp) == len)) {
-				/*
-				 * We found it.  Break out and
-				 * return the pointer to the buffer.
-				 */
-				break;
-			} else {
-				bp = NULL;
-			}
-		}
-	}
-	return bp;
-}
-
-/*
- * Check to see if a buffer matching the given parameters is already
- * a part of the given transaction.  Check all the chunks, we
- * want to be thorough.
- */
-STATIC xfs_buf_t *
-xfs_trans_buf_item_match_all(
-	xfs_trans_t	*tp,
-	xfs_buftarg_t	*target,
-	xfs_daddr_t	blkno,
-	int		len)
-{
-	xfs_log_item_chunk_t	*licp;
-	xfs_log_item_desc_t	*lidp;
-	xfs_buf_log_item_t	*blip;
-	xfs_buf_t		*bp;
-	int			i;
-
-	bp = NULL;
-	len = BBTOB(len);
-	for (licp = &tp->t_items; licp != NULL; licp = licp->lic_next) {
-		if (xfs_lic_are_all_free(licp)) {
-			ASSERT(licp == &tp->t_items);
-			ASSERT(licp->lic_next == NULL);
-			return NULL;
-		}
-		for (i = 0; i < licp->lic_unused; i++) {
-			/*
-			 * Skip unoccupied slots.
-			 */
-			if (xfs_lic_isfree(licp, i)) {
-				continue;
-			}
-
-			lidp = xfs_lic_slot(licp, i);
-			blip = (xfs_buf_log_item_t *)lidp->lid_item;
-			if (blip->bli_item.li_type != XFS_LI_BUF) {
-				continue;
-			}
-
-			bp = blip->bli_buf;
-			if ((XFS_BUF_TARGET(bp) == target) &&
-			    (XFS_BUF_ADDR(bp) == blkno) &&
-			    (XFS_BUF_COUNT(bp) == len)) {
-				/*
-				 * We found it.  Break out and
-				 * return the pointer to the buffer.
-				 */
-				return bp;
-			}
-		}
-	}
-	return NULL;
-}