buffered write and indlen fixes

index e0bc290..e2fbed4 100644
@@ -1079,6 +1079,8 @@ xfs_file_iomap_end_delalloc(
 	int			error = 0;
 
 	start_fsb = XFS_B_TO_FSB(mp, offset + written);
+	if (unlikely(!written))
+		start_fsb = XFS_B_TO_FSBT(mp, offset);
 	end_fsb = XFS_B_TO_FSB(mp, offset + length);
 
 	/*

@@ -1090,6 +1092,9 @@ xfs_file_iomap_end_delalloc(
 	 * blocks in the range, they are ours.
 	 */
 	if (start_fsb < end_fsb) {
+		truncate_pagecache_range(VFS_I(ip), XFS_FSB_TO_B(mp, start_fsb),
+					 XFS_FSB_TO_B(mp, end_fsb) - 1);
+
 		xfs_ilock(ip, XFS_ILOCK_EXCL);
 		error = xfs_bmap_punch_delalloc_range(ip, start_fsb,
 					       end_fsb - start_fsb);

index e0bc290..4009e7c 100644
@@ -1078,7 +1078,15 @@ xfs_file_iomap_end_delalloc(
 	xfs_fileoff_t		end_fsb;
 	int			error = 0;
 
-	start_fsb = XFS_B_TO_FSB(mp, offset + written);
+	/*
+	 * start_fsb refers to the first unused block after a short write. If
+	 * nothing was written, round offset down to point at the first block in
+	 * the range.
+	 */
+	if (unlikely(!written))
+		start_fsb = XFS_B_TO_FSBT(mp, offset);
+	else
+		start_fsb = XFS_B_TO_FSB(mp, offset + written);
 	end_fsb = XFS_B_TO_FSB(mp, offset + length);
 
 	/*

@@ -1090,6 +1098,9 @@ xfs_file_iomap_end_delalloc(
 	 * blocks in the range, they are ours.
 	 */
 	if (start_fsb < end_fsb) {
+		truncate_pagecache_range(VFS_I(ip), XFS_FSB_TO_B(mp, start_fsb),
+					 XFS_FSB_TO_B(mp, end_fsb) - 1);
+
 		xfs_ilock(ip, XFS_ILOCK_EXCL);
 		error = xfs_bmap_punch_delalloc_range(ip, start_fsb,
 					       end_fsb - start_fsb);

index e2fbed4..01e8780 100644
@@ -1078,6 +1078,10 @@ xfs_file_iomap_end_delalloc(
 	xfs_fileoff_t		end_fsb;
 	int			error = 0;
 
+	/* behave as if the write failed if drop writes is enabled */
+	if (xfs_mp_drop_writes(mp))
+		written = 0;
+
 	start_fsb = XFS_B_TO_FSB(mp, offset + written);
 	if (unlikely(!written))
 		start_fsb = XFS_B_TO_FSBT(mp, offset);

index 20e2981..6db6fd6 100644
@@ -200,11 +200,12 @@ typedef struct xfs_mount {
 	/*
 	 * DEBUG mode instrumentation to test and/or trigger delayed allocation
 	 * block killing in the event of failed writes. When enabled, all
-	 * buffered writes are forced to fail. All delalloc blocks in the range
-	 * of the write (including pre-existing delalloc blocks!) are tossed as
-	 * part of the write failure error handling sequence.
+	 * buffered writes are silenty dropped and handled as if they failed.
+	 * All delalloc blocks in the range of the write (including pre-existing
+	 * delalloc blocks!) are tossed as part of the write failure error
+	 * handling sequence.
 	 */
-	bool			m_fail_writes;
+	bool			m_drop_writes;
 #endif
 } xfs_mount_t;
 

@@ -325,13 +326,13 @@ xfs_daddr_to_agbno(struct xfs_mount *mp, xfs_daddr_t d)
 
 #ifdef DEBUG
 static inline bool
-xfs_mp_fail_writes(struct xfs_mount *mp)
+xfs_mp_drop_writes(struct xfs_mount *mp)
 {
-	return mp->m_fail_writes;
+	return mp->m_drop_writes;
 }
 #else
 static inline bool
-xfs_mp_fail_writes(struct xfs_mount *mp)
+xfs_mp_drop_writes(struct xfs_mount *mp)
 {
 	return 0;
 }

index de6195e..80ac15f 100644
@@ -93,7 +93,7 @@ to_mp(struct kobject *kobject)
 #ifdef DEBUG
 
 STATIC ssize_t
-fail_writes_store(
+drop_writes_store(
 	struct kobject		*kobject,
 	const char		*buf,
 	size_t			count)

@@ -107,9 +107,9 @@ fail_writes_store(
 		return ret;
 
 	if (val == 1)
-		mp->m_fail_writes = true;
+		mp->m_drop_writes = true;
 	else if (val == 0)
-		mp->m_fail_writes = false;
+		mp->m_drop_writes = false;
 	else
 		return -EINVAL;
 

@@ -117,21 +117,21 @@ fail_writes_store(
 }
 
 STATIC ssize_t
-fail_writes_show(
+drop_writes_show(
 	struct kobject		*kobject,
 	char			*buf)
 {
 	struct xfs_mount	*mp = to_mp(kobject);
 
-	return snprintf(buf, PAGE_SIZE, "%d\n", mp->m_fail_writes ? 1 : 0);
+	return snprintf(buf, PAGE_SIZE, "%d\n", mp->m_drop_writes ? 1 : 0);
 }
-XFS_SYSFS_ATTR_RW(fail_writes);
+XFS_SYSFS_ATTR_RW(drop_writes);
 
 #endif /* DEBUG */
 
 static struct attribute *xfs_mp_attrs[] = {
 #ifdef DEBUG
-	ATTR_LIST(fail_writes),
+	ATTR_LIST(drop_writes),
 #endif
 	NULL,
 };

index 1cee514..73c9546 100644
@@ -2805,7 +2805,8 @@ xfs_bmap_add_extent_hole_delay(
 		oldlen = startblockval(left.br_startblock) +
 			startblockval(new->br_startblock) +
 			startblockval(right.br_startblock);
-		newlen = xfs_bmap_worst_indlen(ip, temp);
+		newlen = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
+					 oldlen);
 		xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx),
 			nullstartblock((int)newlen));
 		trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);

@@ -2826,7 +2827,8 @@ xfs_bmap_add_extent_hole_delay(
 		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, *idx), temp);
 		oldlen = startblockval(left.br_startblock) +
 			startblockval(new->br_startblock);
-		newlen = xfs_bmap_worst_indlen(ip, temp);
+		newlen = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
+					 oldlen);
 		xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, *idx),
 			nullstartblock((int)newlen));
 		trace_xfs_bmap_post_update(ip, *idx, state, _THIS_IP_);

@@ -2842,7 +2844,8 @@ xfs_bmap_add_extent_hole_delay(
 		temp = new->br_blockcount + right.br_blockcount;
 		oldlen = startblockval(new->br_startblock) +
 			startblockval(right.br_startblock);
-		newlen = xfs_bmap_worst_indlen(ip, temp);
+		newlen = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
+					 oldlen);
 		xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, *idx),
 			new->br_startoff,
 			nullstartblock((int)newlen), temp, right.br_state);

index 73c9546..2ae55db 100644
@@ -4795,34 +4795,59 @@ xfs_bmap_split_indlen(
 	xfs_filblks_t			len2 = *indlen2;
 	xfs_filblks_t			nres = len1 + len2; /* new total res. */
 	xfs_filblks_t			stolen = 0;
+	xfs_filblks_t			resfactor;
 
 	/*
 	 * Steal as many blocks as we can to try and satisfy the worst case
 	 * indlen for both new extents.
 	 */
-	while (nres > ores && avail) {
-		nres--;
-		avail--;
-		stolen++;
-	}
+	if (ores < nres && avail)
+		stolen = XFS_FILBLKS_MIN(nres - ores, avail);
+	ores += stolen;
+
+	 /* nothing else to do if we've satisfied the new reservation */
+	if (ores >= nres)
+		return stolen;
+
+	/*
+	 * We can't meet the total required reservation for the two extents.
+	 * Calculate the percent of the overall shortage between both extents
+	 * and apply this percentage to each of the requested indlen values.
+	 * This distributes the shortage fairly and reduces the chances that one
+	 * of the two extents is left with nothing when extents are repeatedly
+	 * split.
+	 */
+	resfactor = (ores * 100);
+	do_div(resfactor, nres);
+	len1 *= resfactor;
+	do_div(len1, 100);
+	len2 *= resfactor;
+	do_div(len2, 100);
+	ASSERT(len1 + len2 <= ores);
+	ASSERT(len1 < *indlen1 && len2 < *indlen2);
 
 	/*
-	 * The only blocks available are those reserved for the original
-	 * extent and what we can steal from the extent being removed.
-	 * If this still isn't enough to satisfy the combined
-	 * requirements for the two new extents, skim blocks off of each
-	 * of the new reservations until they match what is available.
+	 * Hand out the remainder to each extent. If one of the two reservations
+	 * is zero, we want to make sure that one gets a block first. The loop
+	 * below starts with len1, so hand len2 a block right off the bat if it
+	 * is zero.
 	 */
-	while (nres > ores) {
-		if (len1) {
-			len1--;
-			nres--;
+	ores -= (len1 + len2);
+	ASSERT((*indlen1 - len1) + (*indlen2 - len2) >= ores);
+	if (ores && !len2 && *indlen2) {
+		len2++;
+		ores--;
+	}
+	while (ores) {
+		if (len1 < *indlen1) {
+			len1++;
+			ores--;
 		}
-		if (nres == ores)
+		if (!ores)
 			break;
-		if (len2) {
-			len2--;
-			nres--;
+		if (len2 < *indlen2) {
+			len2++;
+			ores--;
 		}
 	}