1 files changed, 201 insertions, 243 deletions

diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index 77b09750e92..4654338b03f 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -17,17 +17,18 @@
  */
 #include "xfs.h"
 #include "xfs_fs.h"
-#include "xfs_types.h"
+#include "xfs_log_format.h"
+#include "xfs_trans_resv.h"
 #include "xfs_bit.h"
-#include "xfs_log.h"
-#include "xfs_trans.h"
 #include "xfs_sb.h"
 #include "xfs_ag.h"
 #include "xfs_mount.h"
+#include "xfs_trans.h"
 #include "xfs_buf_item.h"
 #include "xfs_trans_priv.h"
 #include "xfs_error.h"
 #include "xfs_trace.h"
+#include "xfs_log.h"
 
 
 kmem_zone_t	*xfs_buf_item_zone;
@@ -37,110 +38,15 @@ static inline struct xfs_buf_log_item *BUF_ITEM(struct xfs_log_item *lip)
 	return container_of(lip, struct xfs_buf_log_item, bli_item);
 }
 
+STATIC void	xfs_buf_do_callbacks(struct xfs_buf *bp);
 
-#ifdef XFS_TRANS_DEBUG
-/*
- * This function uses an alternate strategy for tracking the bytes
- * that the user requests to be logged.  This can then be used
- * in conjunction with the bli_orig array in the buf log item to
- * catch bugs in our callers' code.
- *
- * We also double check the bits set in xfs_buf_item_log using a
- * simple algorithm to check that every byte is accounted for.
- */
-STATIC void
-xfs_buf_item_log_debug(
-	xfs_buf_log_item_t	*bip,
-	uint			first,
-	uint			last)
-{
-	uint	x;
-	uint	byte;
-	uint	nbytes;
-	uint	chunk_num;
-	uint	word_num;
-	uint	bit_num;
-	uint	bit_set;
-	uint	*wordp;
-
-	ASSERT(bip->bli_logged != NULL);
-	byte = first;
-	nbytes = last - first + 1;
-	bfset(bip->bli_logged, first, nbytes);
-	for (x = 0; x < nbytes; x++) {
-		chunk_num = byte >> XFS_BLF_SHIFT;
-		word_num = chunk_num >> BIT_TO_WORD_SHIFT;
-		bit_num = chunk_num & (NBWORD - 1);
-		wordp = &(bip->__bli_format.blf_data_map[word_num]);
-		bit_set = *wordp & (1 << bit_num);
-		ASSERT(bit_set);
-		byte++;
-	}
-}
-
-/*
- * This function is called when we flush something into a buffer without
- * logging it.  This happens for things like inodes which are logged
- * separately from the buffer.
- */
-void
-xfs_buf_item_flush_log_debug(
-	xfs_buf_t	*bp,
-	uint		first,
-	uint		last)
-{
-	xfs_buf_log_item_t	*bip = bp->b_fspriv;
-	uint			nbytes;
-
-	if (bip == NULL || (bip->bli_item.li_type != XFS_LI_BUF))
-		return;
-
-	ASSERT(bip->bli_logged != NULL);
-	nbytes = last - first + 1;
-	bfset(bip->bli_logged, first, nbytes);
-}
-
-/*
- * This function is called to verify that our callers have logged
- * all the bytes that they changed.
- *
- * It does this by comparing the original copy of the buffer stored in
- * the buf log item's bli_orig array to the current copy of the buffer
- * and ensuring that all bytes which mismatch are set in the bli_logged
- * array of the buf log item.
- */
-STATIC void
-xfs_buf_item_log_check(
-	xfs_buf_log_item_t	*bip)
+static inline int
+xfs_buf_log_format_size(
+	struct xfs_buf_log_format *blfp)
 {
-	char		*orig;
-	char		*buffer;
-	int		x;
-	xfs_buf_t	*bp;
-
-	ASSERT(bip->bli_orig != NULL);
-	ASSERT(bip->bli_logged != NULL);
-
-	bp = bip->bli_buf;
-	ASSERT(bp->b_length > 0);
-	ASSERT(bp->b_addr != NULL);
-	orig = bip->bli_orig;
-	buffer = bp->b_addr;
-	for (x = 0; x < BBTOB(bp->b_length); x++) {
-		if (orig[x] != buffer[x] && !btst(bip->bli_logged, x)) {
-			xfs_emerg(bp->b_mount,
-				"%s: bip %x buffer %x orig %x index %d",
-				__func__, bip, bp, orig, x);
-			ASSERT(0);
-		}
-	}
+	return offsetof(struct xfs_buf_log_format, blf_data_map) +
+			(blfp->blf_map_size * sizeof(blfp->blf_data_map[0]));
 }
-#else
-#define		xfs_buf_item_log_debug(x,y,z)
-#define		xfs_buf_item_log_check(x)
-#endif
-
-STATIC void	xfs_buf_do_callbacks(struct xfs_buf *bp);
 
 /*
  * This returns the number of log iovecs needed to log the
@@ -152,25 +58,27 @@ STATIC void	xfs_buf_do_callbacks(struct xfs_buf *bp);
  *
  * If the XFS_BLI_STALE flag has been set, then log nothing.
  */
-STATIC uint
+STATIC void
 xfs_buf_item_size_segment(
 	struct xfs_buf_log_item	*bip,
-	struct xfs_buf_log_format *blfp)
+	struct xfs_buf_log_format *blfp,
+	int			*nvecs,
+	int			*nbytes)
 {
 	struct xfs_buf		*bp = bip->bli_buf;
-	uint			nvecs;
 	int			next_bit;
 	int			last_bit;
 
 	last_bit = xfs_next_bit(blfp->blf_data_map, blfp->blf_map_size, 0);
 	if (last_bit == -1)
-		return 0;
+		return;
 
 	/*
 	 * initial count for a dirty buffer is 2 vectors - the format structure
 	 * and the first dirty region.
 	 */
-	nvecs = 2;
+	*nvecs += 2;
+	*nbytes += xfs_buf_log_format_size(blfp) + XFS_BLF_CHUNK;
 
 	while (last_bit != -1) {
 		/*
@@ -190,18 +98,17 @@ xfs_buf_item_size_segment(
 			break;
 		} else if (next_bit != last_bit + 1) {
 			last_bit = next_bit;
-			nvecs++;
+			(*nvecs)++;
 		} else if (xfs_buf_offset(bp, next_bit * XFS_BLF_CHUNK) !=
 			   (xfs_buf_offset(bp, last_bit * XFS_BLF_CHUNK) +
 			    XFS_BLF_CHUNK)) {
 			last_bit = next_bit;
-			nvecs++;
+			(*nvecs)++;
 		} else {
 			last_bit++;
 		}
+		*nbytes += XFS_BLF_CHUNK;
 	}
-
-	return nvecs;
 }
 
 /*
@@ -221,12 +128,13 @@ xfs_buf_item_size_segment(
  * If the XFS_BLI_STALE flag has been set, then log nothing but the buf log
  * format structures.
  */
-STATIC uint
+STATIC void
 xfs_buf_item_size(
-	struct xfs_log_item	*lip)
+	struct xfs_log_item	*lip,
+	int			*nvecs,
+	int			*nbytes)
 {
 	struct xfs_buf_log_item	*bip = BUF_ITEM(lip);
-	uint			nvecs;
 	int			i;
 
 	ASSERT(atomic_read(&bip->bli_refcount) > 0);
@@ -238,11 +146,26 @@ xfs_buf_item_size(
 		 */
 		trace_xfs_buf_item_size_stale(bip);
 		ASSERT(bip->__bli_format.blf_flags & XFS_BLF_CANCEL);
-		return bip->bli_format_count;
+		*nvecs += bip->bli_format_count;
+		for (i = 0; i < bip->bli_format_count; i++) {
+			*nbytes += xfs_buf_log_format_size(&bip->bli_formats[i]);
+		}
+		return;
 	}
 
 	ASSERT(bip->bli_flags & XFS_BLI_LOGGED);
 
+	if (bip->bli_flags & XFS_BLI_ORDERED) {
+		/*
+		 * The buffer has been logged just to order it.
+		 * It is not being included in the transaction
+		 * commit, so no vectors are used at all.
+		 */
+		trace_xfs_buf_item_size_ordered(bip);
+		*nvecs = XFS_LOG_VEC_ORDERED;
+		return;
+	}
+
 	/*
 	 * the vector count is based on the number of buffer vectors we have
 	 * dirty bits in. This will only be greater than one when we have a
@@ -252,30 +175,54 @@ xfs_buf_item_size(
 	 * count for the extra buf log format structure that will need to be
 	 * written.
 	 */
-	nvecs = 0;
 	for (i = 0; i < bip->bli_format_count; i++) {
-		nvecs += xfs_buf_item_size_segment(bip, &bip->bli_formats[i]);
+		xfs_buf_item_size_segment(bip, &bip->bli_formats[i],
+					  nvecs, nbytes);
 	}
-
 	trace_xfs_buf_item_size(bip);
-	return nvecs;
 }
 
-static struct xfs_log_iovec *
+static inline void
+xfs_buf_item_copy_iovec(
+	struct xfs_log_vec	*lv,
+	struct xfs_log_iovec	**vecp,
+	struct xfs_buf		*bp,
+	uint			offset,
+	int			first_bit,
+	uint			nbits)
+{
+	offset += first_bit * XFS_BLF_CHUNK;
+	xlog_copy_iovec(lv, vecp, XLOG_REG_TYPE_BCHUNK,
+			xfs_buf_offset(bp, offset),
+			nbits * XFS_BLF_CHUNK);
+}
+
+static inline bool
+xfs_buf_item_straddle(
+	struct xfs_buf		*bp,
+	uint			offset,
+	int			next_bit,
+	int			last_bit)
+{
+	return xfs_buf_offset(bp, offset + (next_bit << XFS_BLF_SHIFT)) !=
+		(xfs_buf_offset(bp, offset + (last_bit << XFS_BLF_SHIFT)) +
+		 XFS_BLF_CHUNK);
+}
+
+static void
 xfs_buf_item_format_segment(
 	struct xfs_buf_log_item	*bip,
-	struct xfs_log_iovec	*vecp,
+	struct xfs_log_vec	*lv,
+	struct xfs_log_iovec	**vecp,
 	uint			offset,
 	struct xfs_buf_log_format *blfp)
 {
 	struct xfs_buf	*bp = bip->bli_buf;
 	uint		base_size;
-	uint		nvecs;
 	int		first_bit;
 	int		last_bit;
 	int		next_bit;
 	uint		nbits;
-	uint		buffer_offset;
 
 	/* copy the flags across from the base format item */
 	blfp->blf_flags = bip->__bli_format.blf_flags;
@@ -285,24 +232,19 @@ xfs_buf_item_format_segment(
 	 * the actual size of the dirty bitmap rather than the size of the in
 	 * memory structure.
 	 */
-	base_size = offsetof(struct xfs_buf_log_format, blf_data_map) +
-			(blfp->blf_map_size * sizeof(blfp->blf_data_map[0]));
+	base_size = xfs_buf_log_format_size(blfp);
 
-	nvecs = 0;
 	first_bit = xfs_next_bit(blfp->blf_data_map, blfp->blf_map_size, 0);
 	if (!(bip->bli_flags & XFS_BLI_STALE) && first_bit == -1) {
 		/*
 		 * If the map is not be dirty in the transaction, mark
 		 * the size as zero and do not advance the vector pointer.
 		 */
-		goto out;
+		return;
 	}
 
-	vecp->i_addr = blfp;
-	vecp->i_len = base_size;
-	vecp->i_type = XLOG_REG_TYPE_BFORMAT;
-	vecp++;
-	nvecs = 1;
+	blfp = xlog_copy_iovec(lv, vecp, XLOG_REG_TYPE_BFORMAT, blfp, base_size);
+	blfp->blf_size = 1;
 
 	if (bip->bli_flags & XFS_BLI_STALE) {
 		/*
@@ -312,13 +254,13 @@ xfs_buf_item_format_segment(
 		 */
 		trace_xfs_buf_item_format_stale(bip);
 		ASSERT(blfp->blf_flags & XFS_BLF_CANCEL);
-		goto out;
+		return;
 	}
 
+
 	/*
 	 * Fill in an iovec for each set of contiguous chunks.
 	 */
-
 	last_bit = first_bit;
 	nbits = 1;
 	for (;;) {
@@ -331,47 +273,22 @@ xfs_buf_item_format_segment(
 		next_bit = xfs_next_bit(blfp->blf_data_map, blfp->blf_map_size,
 					(uint)last_bit + 1);
 		/*
-		 * If we run out of bits fill in the last iovec and get
-		 * out of the loop.
-		 * Else if we start a new set of bits then fill in the
-		 * iovec for the series we were looking at and start
-		 * counting the bits in the new one.
-		 * Else we're still in the same set of bits so just
-		 * keep counting and scanning.
+		 * If we run out of bits fill in the last iovec and get out of
+		 * the loop.  Else if we start a new set of bits then fill in
+		 * the iovec for the series we were looking at and start
+		 * counting the bits in the new one.  Else we're still in the
+		 * same set of bits so just keep counting and scanning.
 		 */
 		if (next_bit == -1) {
-			buffer_offset = offset + first_bit * XFS_BLF_CHUNK;
-			vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
-			vecp->i_len = nbits * XFS_BLF_CHUNK;
-			vecp->i_type = XLOG_REG_TYPE_BCHUNK;
-			nvecs++;
+			xfs_buf_item_copy_iovec(lv, vecp, bp, offset,
+						first_bit, nbits);
+			blfp->blf_size++;
 			break;
-		} else if (next_bit != last_bit + 1) {
-			buffer_offset = offset + first_bit * XFS_BLF_CHUNK;
-			vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
-			vecp->i_len = nbits * XFS_BLF_CHUNK;
-			vecp->i_type = XLOG_REG_TYPE_BCHUNK;
-			nvecs++;
-			vecp++;
-			first_bit = next_bit;
-			last_bit = next_bit;
-			nbits = 1;
-		} else if (xfs_buf_offset(bp, offset +
-					      (next_bit << XFS_BLF_SHIFT)) !=
-			   (xfs_buf_offset(bp, offset +
-					       (last_bit << XFS_BLF_SHIFT)) +
-			    XFS_BLF_CHUNK)) {
-			buffer_offset = offset + first_bit * XFS_BLF_CHUNK;
-			vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
-			vecp->i_len = nbits * XFS_BLF_CHUNK;
-			vecp->i_type = XLOG_REG_TYPE_BCHUNK;
-/*
- * You would think we need to bump the nvecs here too, but we do not
- * this number is used by recovery, and it gets confused by the boundary
- * split here
- *			nvecs++;
- */
-			vecp++;
+		} else if (next_bit != last_bit + 1 ||
+		           xfs_buf_item_straddle(bp, offset, next_bit, last_bit)) {
+			xfs_buf_item_copy_iovec(lv, vecp, bp, offset,
+						first_bit, nbits);
+			blfp->blf_size++;
 			first_bit = next_bit;
 			last_bit = next_bit;
 			nbits = 1;
@@ -380,9 +297,6 @@ xfs_buf_item_format_segment(
 			nbits++;
 		}
 	}
-out:
-	blfp->blf_size = nvecs;
-	return vecp;
 }
 
 /*
@@ -394,10 +308,11 @@ out:
 STATIC void
 xfs_buf_item_format(
 	struct xfs_log_item	*lip,
-	struct xfs_log_iovec	*vecp)
+	struct xfs_log_vec	*lv)
 {
 	struct xfs_buf_log_item	*bip = BUF_ITEM(lip);
 	struct xfs_buf		*bp = bip->bli_buf;
+	struct xfs_log_iovec	*vecp = NULL;
 	uint			offset = 0;
 	int			i;
 
@@ -407,21 +322,39 @@ xfs_buf_item_format(
 
 	/*
 	 * If it is an inode buffer, transfer the in-memory state to the
-	 * format flags and clear the in-memory state. We do not transfer
+	 * format flags and clear the in-memory state.
+	 *
+	 * For buffer based inode allocation, we do not transfer
 	 * this state if the inode buffer allocation has not yet been committed
 	 * to the log as setting the XFS_BLI_INODE_BUF flag will prevent
 	 * correct replay of the inode allocation.
+	 *
+	 * For icreate item based inode allocation, the buffers aren't written
+	 * to the journal during allocation, and hence we should always tag the
+	 * buffer as an inode buffer so that the correct unlinked list replay
+	 * occurs during recovery.
 	 */
 	if (bip->bli_flags & XFS_BLI_INODE_BUF) {
-		if (!((bip->bli_flags & XFS_BLI_INODE_ALLOC_BUF) &&
+		if (xfs_sb_version_hascrc(&lip->li_mountp->m_sb) ||
+		    !((bip->bli_flags & XFS_BLI_INODE_ALLOC_BUF) &&
 		      xfs_log_item_in_current_chkpt(lip)))
 			bip->__bli_format.blf_flags |= XFS_BLF_INODE_BUF;
 		bip->bli_flags &= ~XFS_BLI_INODE_BUF;
 	}
 
+	if ((bip->bli_flags & (XFS_BLI_ORDERED|XFS_BLI_STALE)) ==
+							XFS_BLI_ORDERED) {
+		/*
+		 * The buffer has been logged just to order it.  It is not being
+		 * included in the transaction commit, so don't format it.
+		 */
+		trace_xfs_buf_item_format_ordered(bip);
+		return;
+	}
+
 	for (i = 0; i < bip->bli_format_count; i++) {
-		vecp = xfs_buf_item_format_segment(bip, vecp, offset,
-						&bip->bli_formats[i]);
+		xfs_buf_item_format_segment(bip, lv, &vecp, offset,
+					    &bip->bli_formats[i]);
 		offset += bp->b_maps[i].bm_len;
 	}
 
@@ -429,7 +362,6 @@ xfs_buf_item_format(
 	 * Check to make sure everything is consistent.
 	 */
 	trace_xfs_buf_item_format(bip);
-	xfs_buf_item_log_check(bip);
 }
 
 /*
@@ -449,6 +381,7 @@ xfs_buf_item_pin(
 
 	ASSERT(atomic_read(&bip->bli_refcount) > 0);
 	ASSERT((bip->bli_flags & XFS_BLI_LOGGED) ||
+	       (bip->bli_flags & XFS_BLI_ORDERED) ||
 	       (bip->bli_flags & XFS_BLI_STALE));
 
 	trace_xfs_buf_item_pin(bip);
@@ -562,6 +495,14 @@ xfs_buf_item_unpin(
 	}
 }
 
+/*
+ * Buffer IO error rate limiting. Limit it to no more than 10 messages per 30
+ * seconds so as to not spam logs too much on repeated detection of the same
+ * buffer being bad..
+ */
+
+DEFINE_RATELIMIT_STATE(xfs_buf_write_fail_rl_state, 30 * HZ, 10);
+
 STATIC uint
 xfs_buf_item_push(
 	struct xfs_log_item	*lip,
@@ -573,13 +514,31 @@ xfs_buf_item_push(
 
 	if (xfs_buf_ispinned(bp))
 		return XFS_ITEM_PINNED;
-	if (!xfs_buf_trylock(bp))
+	if (!xfs_buf_trylock(bp)) {
+		/*
+		 * If we have just raced with a buffer being pinned and it has
+		 * been marked stale, we could end up stalling until someone else
+		 * issues a log force to unpin the stale buffer. Check for the
+		 * race condition here so xfsaild recognizes the buffer is pinned
+		 * and queues a log force to move it along.
+		 */
+		if (xfs_buf_ispinned(bp))
+			return XFS_ITEM_PINNED;
 		return XFS_ITEM_LOCKED;
+	}
 
 	ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
 
 	trace_xfs_buf_item_push(bip);
 
+	/* has a previous flush failed due to IO errors? */
+	if ((bp->b_flags & XBF_WRITE_FAIL) &&
+	    ___ratelimit(&xfs_buf_write_fail_rl_state, "XFS:")) {
+		xfs_warn(bp->b_target->bt_mount,
+"Detected failing async write on buffer block 0x%llx. Retrying async write.\n",
+			 (long long)bp->b_bn);
+	}
+
 	if (!xfs_buf_delwri_queue(bp, buffer_list))
 		rval = XFS_ITEM_FLUSHING;
 	xfs_buf_unlock(bp);
@@ -611,8 +570,9 @@ xfs_buf_item_unlock(
 {
 	struct xfs_buf_log_item	*bip = BUF_ITEM(lip);
 	struct xfs_buf		*bp = bip->bli_buf;
-	int			aborted, clean, i;
-	uint			hold;
+	bool			clean;
+	bool			aborted;
+	int			flags;
 
 	/* Clear the buffer's association with this transaction. */
 	bp->b_transp = NULL;
@@ -623,23 +583,21 @@ xfs_buf_item_unlock(
 	 * (cancelled) buffers at unpin time, but we'll never go through the
 	 * pin/unpin cycle if we abort inside commit.
 	 */
-	aborted = (lip->li_flags & XFS_LI_ABORTED) != 0;
-
+	aborted = (lip->li_flags & XFS_LI_ABORTED) ? true : false;
 	/*
-	 * Before possibly freeing the buf item, determine if we should
-	 * release the buffer at the end of this routine.
+	 * Before possibly freeing the buf item, copy the per-transaction state
+	 * so we can reference it safely later after clearing it from the
+	 * buffer log item.
 	 */
-	hold = bip->bli_flags & XFS_BLI_HOLD;
-
-	/* Clear the per transaction state. */
-	bip->bli_flags &= ~(XFS_BLI_LOGGED | XFS_BLI_HOLD);
+	flags = bip->bli_flags;
+	bip->bli_flags &= ~(XFS_BLI_LOGGED | XFS_BLI_HOLD | XFS_BLI_ORDERED);
 
 	/*
 	 * If the buf item is marked stale, then don't do anything.  We'll
 	 * unlock the buffer and free the buf item when the buffer is unpinned
 	 * for the last time.
 	 */
-	if (bip->bli_flags & XFS_BLI_STALE) {
+	if (flags & XFS_BLI_STALE) {
 		trace_xfs_buf_item_unlock_stale(bip);
 		ASSERT(bip->__bli_format.blf_flags & XFS_BLF_CANCEL);
 		if (!aborted) {
@@ -652,22 +610,49 @@ xfs_buf_item_unlock(
 
 	/*
 	 * If the buf item isn't tracking any data, free it, otherwise drop the
-	 * reference we hold to it.
+	 * reference we hold to it. If we are aborting the transaction, this may
+	 * be the only reference to the buf item, so we free it anyway
+	 * regardless of whether it is dirty or not. A dirty abort implies a
+	 * shutdown, anyway.
+	 *
+	 * Ordered buffers are dirty but may have no recorded changes, so ensure
+	 * we only release clean items here.
 	 */
-	clean = 1;
-	for (i = 0; i < bip->bli_format_count; i++) {
-		if (!xfs_bitmap_empty(bip->bli_formats[i].blf_data_map,
-			     bip->bli_formats[i].blf_map_size)) {
-			clean = 0;
-			break;
+	clean = (flags & XFS_BLI_DIRTY) ? false : true;
+	if (clean) {
+		int i;
+		for (i = 0; i < bip->bli_format_count; i++) {
+			if (!xfs_bitmap_empty(bip->bli_formats[i].blf_data_map,
+				     bip->bli_formats[i].blf_map_size)) {
+				clean = false;
+				break;
+			}
 		}
 	}
-	if (clean)
-		xfs_buf_item_relse(bp);
-	else
-		atomic_dec(&bip->bli_refcount);
 
-	if (!hold)
+	/*
+	 * Clean buffers, by definition, cannot be in the AIL. However, aborted
+	 * buffers may be dirty and hence in the AIL. Therefore if we are
+	 * aborting a buffer and we've just taken the last refernce away, we
+	 * have to check if it is in the AIL before freeing it. We need to free
+	 * it in this case, because an aborted transaction has already shut the
+	 * filesystem down and this is the last chance we will have to do so.
+	 */
+	if (atomic_dec_and_test(&bip->bli_refcount)) {
+		if (clean)
+			xfs_buf_item_relse(bp);
+		else if (aborted) {
+			ASSERT(XFS_FORCED_SHUTDOWN(lip->li_mountp));
+			if (lip->li_flags & XFS_LI_IN_AIL) {
+				spin_lock(&lip->li_ailp->xa_lock);
+				xfs_trans_ail_delete(lip->li_ailp, lip,
+						     SHUTDOWN_LOG_IO_ERROR);
+			}
+			xfs_buf_item_relse(bp);
+		}
+	}
+
+	if (!(flags & XFS_BLI_HOLD))
 		xfs_buf_relse(bp);
 }
 
@@ -811,20 +796,6 @@ xfs_buf_item_init(
 		bip->bli_formats[i].blf_map_size = map_size;
 	}
 
-#ifdef XFS_TRANS_DEBUG
-	/*
-	 * Allocate the arrays for tracking what needs to be logged
-	 * and what our callers request to be logged.  bli_orig
-	 * holds a copy of the original, clean buffer for comparison
-	 * against, and bli_logged keeps a 1 bit flag per byte in
-	 * the buffer to indicate which bytes the callers have asked
-	 * to have logged.
-	 */
-	bip->bli_orig = kmem_alloc(BBTOB(bp->b_length), KM_SLEEP);
-	memcpy(bip->bli_orig, bp->b_addr, BBTOB(bp->b_length));
-	bip->bli_logged = kmem_zalloc(BBTOB(bp->b_length) / NBBY, KM_SLEEP);
-#endif
-
 	/*
 	 * Put the buf item into the list of items attached to the
 	 * buffer at the front.
@@ -839,9 +810,8 @@ xfs_buf_item_init(
  * Mark bytes first through last inclusive as dirty in the buf
  * item's bitmap.
  */
-void
+static void
 xfs_buf_item_log_segment(
-	struct xfs_buf_log_item	*bip,
 	uint			first,
 	uint			last,
 	uint			*map)
@@ -915,8 +885,6 @@ xfs_buf_item_log_segment(
 		mask = (1 << end_bit) - 1;
 		*wordp |= mask;
 	}
-
-	xfs_buf_item_log_debug(bip, first, last);
 }
 
 /*
@@ -935,12 +903,6 @@ xfs_buf_item_log(
 	struct xfs_buf		*bp = bip->bli_buf;
 
 	/*
-	 * Mark the item as having some dirty data for
-	 * quick reference in xfs_buf_item_dirty.
-	 */
-	bip->bli_flags |= XFS_BLI_DIRTY;
-
-	/*
 	 * walk each buffer segment and mark them dirty appropriately.
 	 */
 	start = 0;
@@ -957,7 +919,7 @@ xfs_buf_item_log(
 		if (end > last)
 			end = last;
 
-		xfs_buf_item_log_segment(bip, first, end,
+		xfs_buf_item_log_segment(first, end,
 					 &bip->bli_formats[i].blf_data_map[0]);
 
 		start += bp->b_maps[i].bm_len;
@@ -966,7 +928,7 @@ xfs_buf_item_log(
 
 
 /*
- * Return 1 if the buffer has some data that has been logged (at any
+ * Return 1 if the buffer has been logged or ordered in a transaction (at any
  * point, not just the current transaction) and 0 if not.
  */
 uint
@@ -980,11 +942,6 @@ STATIC void
 xfs_buf_item_free(
 	xfs_buf_log_item_t	*bip)
 {
-#ifdef XFS_TRANS_DEBUG
-	kmem_free(bip->bli_orig);
-	kmem_free(bip->bli_logged);
-#endif /* XFS_TRANS_DEBUG */
-
 	xfs_buf_item_free_format(bip);
 	kmem_zone_free(xfs_buf_item_zone, bip);
 }
@@ -1000,11 +957,11 @@ void
 xfs_buf_item_relse(
 	xfs_buf_t	*bp)
 {
-	xfs_buf_log_item_t	*bip;
+	xfs_buf_log_item_t	*bip = bp->b_fspriv;
 
 	trace_xfs_buf_item_relse(bp, _RET_IP_);
+	ASSERT(!(bip->bli_item.li_flags & XFS_LI_IN_AIL));
 
-	bip = bp->b_fspriv;
 	bp->b_fspriv = bip->bli_item.li_bio_list;
 	if (bp->b_fspriv == NULL)
 		bp->b_iodone = NULL;
@@ -1095,7 +1052,7 @@ xfs_buf_iodone_callbacks(
 	static ulong		lasttime;
 	static xfs_buftarg_t	*lasttarg;
 
-	if (likely(!xfs_buf_geterror(bp)))
+	if (likely(!bp->b_error))
 		goto do_callbacks;
 
 	/*
@@ -1134,8 +1091,9 @@ xfs_buf_iodone_callbacks(
 
 		xfs_buf_ioerror(bp, 0); /* errno of 0 unsets the flag */
 
-		if (!XFS_BUF_ISSTALE(bp)) {
-			bp->b_flags |= XBF_WRITE | XBF_ASYNC | XBF_DONE;
+		if (!(bp->b_flags & (XBF_STALE|XBF_WRITE_FAIL))) {
+			bp->b_flags |= XBF_WRITE | XBF_ASYNC |
+				       XBF_DONE | XBF_WRITE_FAIL;
 			xfs_buf_iorequest(bp);
 		} else {
 			xfs_buf_relse(bp);