1 files changed, 285 insertions, 82 deletions

diff --git a/fs/btrfs/space-info.c b/fs/btrfs/space-info.c
index 64099565ab8f..2da6177f4b0b 100644
--- a/fs/btrfs/space-info.c
+++ b/fs/btrfs/space-info.c
@@ -140,6 +140,12 @@
  *     be freed, plus any delayed work we may not have gotten rid of in the case
  *     of metadata.
  *
+ *   FORCE_COMMIT_TRANS
+ *     For use by the preemptive flusher.  We use this to bypass the ticketing
+ *     checks in may_commit_transaction, as we have more information about the
+ *     overall state of the system and may want to commit the transaction ahead
+ *     of actual ENOSPC conditions.
+ *
  * OVERCOMMIT
  *
  *   Because we hold so many reservations for metadata we will allow you to
@@ -163,6 +169,7 @@ u64 __pure btrfs_space_info_used(struct btrfs_space_info *s_info,
 	ASSERT(s_info);
 	return s_info->bytes_used + s_info->bytes_reserved +
 		s_info->bytes_pinned + s_info->bytes_readonly +
+		s_info->bytes_zone_unusable +
 		(may_use_included ? s_info->bytes_may_use : 0);
 }
 
@@ -206,6 +213,7 @@ static int create_space_info(struct btrfs_fs_info *info, u64 flags)
 	INIT_LIST_HEAD(&space_info->ro_bgs);
 	INIT_LIST_HEAD(&space_info->tickets);
 	INIT_LIST_HEAD(&space_info->priority_tickets);
+	space_info->clamp = 1;
 
 	ret = btrfs_sysfs_add_space_info_type(info, space_info);
 	if (ret)
@@ -257,7 +265,7 @@ out:
 
 void btrfs_update_space_info(struct btrfs_fs_info *info, u64 flags,
 			     u64 total_bytes, u64 bytes_used,
-			     u64 bytes_readonly,
+			     u64 bytes_readonly, u64 bytes_zone_unusable,
 			     struct btrfs_space_info **space_info)
 {
 	struct btrfs_space_info *found;
@@ -273,6 +281,7 @@ void btrfs_update_space_info(struct btrfs_fs_info *info, u64 flags,
 	found->bytes_used += bytes_used;
 	found->disk_used += bytes_used * factor;
 	found->bytes_readonly += bytes_readonly;
+	found->bytes_zone_unusable += bytes_zone_unusable;
 	if (total_bytes > 0)
 		found->full = 0;
 	btrfs_try_granting_tickets(info, found);
@@ -422,10 +431,10 @@ static void __btrfs_dump_space_info(struct btrfs_fs_info *fs_info,
 		   info->total_bytes - btrfs_space_info_used(info, true),
 		   info->full ? "" : "not ");
 	btrfs_info(fs_info,
-		"space_info total=%llu, used=%llu, pinned=%llu, reserved=%llu, may_use=%llu, readonly=%llu",
+		"space_info total=%llu, used=%llu, pinned=%llu, reserved=%llu, may_use=%llu, readonly=%llu zone_unusable=%llu",
 		info->total_bytes, info->bytes_used, info->bytes_pinned,
 		info->bytes_reserved, info->bytes_may_use,
-		info->bytes_readonly);
+		info->bytes_readonly, info->bytes_zone_unusable);
 
 	DUMP_BLOCK_RSV(fs_info, global_block_rsv);
 	DUMP_BLOCK_RSV(fs_info, trans_block_rsv);
@@ -454,9 +463,10 @@ again:
 	list_for_each_entry(cache, &info->block_groups[index], list) {
 		spin_lock(&cache->lock);
 		btrfs_info(fs_info,
-			"block group %llu has %llu bytes, %llu used %llu pinned %llu reserved %s",
+			"block group %llu has %llu bytes, %llu used %llu pinned %llu reserved %llu zone_unusable %s",
 			cache->start, cache->length, cache->used, cache->pinned,
-			cache->reserved, cache->ro ? "[readonly]" : "");
+			cache->reserved, cache->zone_unusable,
+			cache->ro ? "[readonly]" : "");
 		spin_unlock(&cache->lock);
 		btrfs_dump_free_space(cache, bytes);
 	}
@@ -489,7 +499,7 @@ static void shrink_delalloc(struct btrfs_fs_info *fs_info,
 {
 	struct btrfs_trans_handle *trans;
 	u64 delalloc_bytes;
-	u64 dio_bytes;
+	u64 ordered_bytes;
 	u64 items;
 	long time_left;
 	int loops;
@@ -513,26 +523,24 @@ static void shrink_delalloc(struct btrfs_fs_info *fs_info,
 
 	delalloc_bytes = percpu_counter_sum_positive(
 						&fs_info->delalloc_bytes);
-	dio_bytes = percpu_counter_sum_positive(&fs_info->dio_bytes);
-	if (delalloc_bytes == 0 && dio_bytes == 0) {
-		if (trans)
-			return;
-		if (wait_ordered)
-			btrfs_wait_ordered_roots(fs_info, items, 0, (u64)-1);
+	ordered_bytes = percpu_counter_sum_positive(&fs_info->ordered_bytes);
+	if (delalloc_bytes == 0 && ordered_bytes == 0)
 		return;
-	}
 
 	/*
 	 * If we are doing more ordered than delalloc we need to just wait on
 	 * ordered extents, otherwise we'll waste time trying to flush delalloc
 	 * that likely won't give us the space back we need.
 	 */
-	if (dio_bytes > delalloc_bytes)
+	if (ordered_bytes > delalloc_bytes)
 		wait_ordered = true;
 
 	loops = 0;
-	while ((delalloc_bytes || dio_bytes) && loops < 3) {
-		btrfs_start_delalloc_roots(fs_info, items);
+	while ((delalloc_bytes || ordered_bytes) && loops < 3) {
+		u64 temp = min(delalloc_bytes, to_reclaim) >> PAGE_SHIFT;
+		long nr_pages = min_t(u64, temp, LONG_MAX);
+
+		btrfs_start_delalloc_roots(fs_info, nr_pages, true);
 
 		loops++;
 		if (wait_ordered && !trans) {
@@ -553,15 +561,16 @@ static void shrink_delalloc(struct btrfs_fs_info *fs_info,
 
 		delalloc_bytes = percpu_counter_sum_positive(
 						&fs_info->delalloc_bytes);
-		dio_bytes = percpu_counter_sum_positive(&fs_info->dio_bytes);
+		ordered_bytes = percpu_counter_sum_positive(
+						&fs_info->ordered_bytes);
 	}
 }
 
 /**
- * maybe_commit_transaction - possibly commit the transaction if its ok to
- * @root - the root we're allocating for
- * @bytes - the number of bytes we want to reserve
- * @force - force the commit
+ * Possibly commit the transaction if its ok to
+ *
+ * @fs_info:    the filesystem
+ * @space_info: space_info we are checking for commit, either data or metadata
  *
  * This will check to make sure that committing the transaction will actually
  * get us somewhere and then commit the transaction if it does.  Otherwise it
@@ -667,7 +676,7 @@ enospc:
  */
 static void flush_space(struct btrfs_fs_info *fs_info,
 		       struct btrfs_space_info *space_info, u64 num_bytes,
-		       int state)
+		       enum btrfs_flush_state state, bool for_preempt)
 {
 	struct btrfs_root *root = fs_info->extent_root;
 	struct btrfs_trans_handle *trans;
@@ -736,13 +745,21 @@ static void flush_space(struct btrfs_fs_info *fs_info,
 	case COMMIT_TRANS:
 		ret = may_commit_transaction(fs_info, space_info);
 		break;
+	case FORCE_COMMIT_TRANS:
+		trans = btrfs_join_transaction(root);
+		if (IS_ERR(trans)) {
+			ret = PTR_ERR(trans);
+			break;
+		}
+		ret = btrfs_commit_transaction(trans);
+		break;
 	default:
 		ret = -ENOSPC;
 		break;
 	}
 
 	trace_btrfs_flush_space(fs_info, space_info->flags, num_bytes, state,
-				ret);
+				ret, for_preempt);
 	return;
 }
 
@@ -752,7 +769,6 @@ btrfs_calc_reclaim_metadata_size(struct btrfs_fs_info *fs_info,
 {
 	u64 used;
 	u64 avail;
-	u64 expected;
 	u64 to_reclaim = space_info->reclaim_size;
 
 	lockdep_assert_held(&space_info->lock);
@@ -770,43 +786,88 @@ btrfs_calc_reclaim_metadata_size(struct btrfs_fs_info *fs_info,
 	if (space_info->total_bytes + avail < used)
 		to_reclaim += used - (space_info->total_bytes + avail);
 
-	if (to_reclaim)
-		return to_reclaim;
-
-	to_reclaim = min_t(u64, num_online_cpus() * SZ_1M, SZ_16M);
-	if (btrfs_can_overcommit(fs_info, space_info, to_reclaim,
-				 BTRFS_RESERVE_FLUSH_ALL))
-		return 0;
-
-	used = btrfs_space_info_used(space_info, true);
-
-	if (btrfs_can_overcommit(fs_info, space_info, SZ_1M,
-				 BTRFS_RESERVE_FLUSH_ALL))
-		expected = div_factor_fine(space_info->total_bytes, 95);
-	else
-		expected = div_factor_fine(space_info->total_bytes, 90);
-
-	if (used > expected)
-		to_reclaim = used - expected;
-	else
-		to_reclaim = 0;
-	to_reclaim = min(to_reclaim, space_info->bytes_may_use +
-				     space_info->bytes_reserved);
 	return to_reclaim;
 }
 
-static inline int need_do_async_reclaim(struct btrfs_fs_info *fs_info,
-					struct btrfs_space_info *space_info,
-					u64 used)
+static bool need_preemptive_reclaim(struct btrfs_fs_info *fs_info,
+				    struct btrfs_space_info *space_info)
 {
+	u64 ordered, delalloc;
 	u64 thresh = div_factor_fine(space_info->total_bytes, 98);
+	u64 used;
 
 	/* If we're just plain full then async reclaim just slows us down. */
 	if ((space_info->bytes_used + space_info->bytes_reserved) >= thresh)
-		return 0;
+		return false;
 
-	if (!btrfs_calc_reclaim_metadata_size(fs_info, space_info))
-		return 0;
+	/*
+	 * We have tickets queued, bail so we don't compete with the async
+	 * flushers.
+	 */
+	if (space_info->reclaim_size)
+		return false;
+
+	/*
+	 * If we have over half of the free space occupied by reservations or
+	 * pinned then we want to start flushing.
+	 *
+	 * We do not do the traditional thing here, which is to say
+	 *
+	 *   if (used >= ((total_bytes + avail) / 2))
+	 *     return 1;
+	 *
+	 * because this doesn't quite work how we want.  If we had more than 50%
+	 * of the space_info used by bytes_used and we had 0 available we'd just
+	 * constantly run the background flusher.  Instead we want it to kick in
+	 * if our reclaimable space exceeds our clamped free space.
+	 *
+	 * Our clamping range is 2^1 -> 2^8.  Practically speaking that means
+	 * the following:
+	 *
+	 * Amount of RAM        Minimum threshold       Maximum threshold
+	 *
+	 *        256GiB                     1GiB                  128GiB
+	 *        128GiB                   512MiB                   64GiB
+	 *         64GiB                   256MiB                   32GiB
+	 *         32GiB                   128MiB                   16GiB
+	 *         16GiB                    64MiB                    8GiB
+	 *
+	 * These are the range our thresholds will fall in, corresponding to how
+	 * much delalloc we need for the background flusher to kick in.
+	 */
+
+	thresh = calc_available_free_space(fs_info, space_info,
+					   BTRFS_RESERVE_FLUSH_ALL);
+	thresh += (space_info->total_bytes - space_info->bytes_used -
+		   space_info->bytes_reserved - space_info->bytes_readonly);
+	thresh >>= space_info->clamp;
+
+	used = space_info->bytes_pinned;
+
+	/*
+	 * If we have more ordered bytes than delalloc bytes then we're either
+	 * doing a lot of DIO, or we simply don't have a lot of delalloc waiting
+	 * around.  Preemptive flushing is only useful in that it can free up
+	 * space before tickets need to wait for things to finish.  In the case
+	 * of ordered extents, preemptively waiting on ordered extents gets us
+	 * nothing, if our reservations are tied up in ordered extents we'll
+	 * simply have to slow down writers by forcing them to wait on ordered
+	 * extents.
+	 *
+	 * In the case that ordered is larger than delalloc, only include the
+	 * block reserves that we would actually be able to directly reclaim
+	 * from.  In this case if we're heavy on metadata operations this will
+	 * clearly be heavy enough to warrant preemptive flushing.  In the case
+	 * of heavy DIO or ordered reservations, preemptive flushing will just
+	 * waste time and cause us to slow down.
+	 */
+	ordered = percpu_counter_sum_positive(&fs_info->ordered_bytes);
+	delalloc = percpu_counter_sum_positive(&fs_info->delalloc_bytes);
+	if (ordered >= delalloc)
+		used += fs_info->delayed_refs_rsv.reserved +
+			fs_info->delayed_block_rsv.reserved;
+	else
+		used += space_info->bytes_may_use;
 
 	return (used >= thresh && !btrfs_fs_closing(fs_info) &&
 		!test_bit(BTRFS_FS_STATE_REMOUNTING, &fs_info->fs_state));
@@ -920,7 +981,7 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work)
 	struct btrfs_fs_info *fs_info;
 	struct btrfs_space_info *space_info;
 	u64 to_reclaim;
-	int flush_state;
+	enum btrfs_flush_state flush_state;
 	int commit_cycles = 0;
 	u64 last_tickets_id;
 
@@ -939,7 +1000,7 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work)
 
 	flush_state = FLUSH_DELAYED_ITEMS_NR;
 	do {
-		flush_space(fs_info, space_info, to_reclaim, flush_state);
+		flush_space(fs_info, space_info, to_reclaim, flush_state, false);
 		spin_lock(&space_info->lock);
 		if (list_empty(&space_info->tickets)) {
 			space_info->flush = 0;
@@ -988,6 +1049,105 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work)
 }
 
 /*
+ * This handles pre-flushing of metadata space before we get to the point that
+ * we need to start blocking threads on tickets.  The logic here is different
+ * from the other flush paths because it doesn't rely on tickets to tell us how
+ * much we need to flush, instead it attempts to keep us below the 80% full
+ * watermark of space by flushing whichever reservation pool is currently the
+ * largest.
+ */
+static void btrfs_preempt_reclaim_metadata_space(struct work_struct *work)
+{
+	struct btrfs_fs_info *fs_info;
+	struct btrfs_space_info *space_info;
+	struct btrfs_block_rsv *delayed_block_rsv;
+	struct btrfs_block_rsv *delayed_refs_rsv;
+	struct btrfs_block_rsv *global_rsv;
+	struct btrfs_block_rsv *trans_rsv;
+	int loops = 0;
+
+	fs_info = container_of(work, struct btrfs_fs_info,
+			       preempt_reclaim_work);
+	space_info = btrfs_find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA);
+	delayed_block_rsv = &fs_info->delayed_block_rsv;
+	delayed_refs_rsv = &fs_info->delayed_refs_rsv;
+	global_rsv = &fs_info->global_block_rsv;
+	trans_rsv = &fs_info->trans_block_rsv;
+
+	spin_lock(&space_info->lock);
+	while (need_preemptive_reclaim(fs_info, space_info)) {
+		enum btrfs_flush_state flush;
+		u64 delalloc_size = 0;
+		u64 to_reclaim, block_rsv_size;
+		u64 global_rsv_size = global_rsv->reserved;
+
+		loops++;
+
+		/*
+		 * We don't have a precise counter for the metadata being
+		 * reserved for delalloc, so we'll approximate it by subtracting
+		 * out the block rsv's space from the bytes_may_use.  If that
+		 * amount is higher than the individual reserves, then we can
+		 * assume it's tied up in delalloc reservations.
+		 */
+		block_rsv_size = global_rsv_size +
+			delayed_block_rsv->reserved +
+			delayed_refs_rsv->reserved +
+			trans_rsv->reserved;
+		if (block_rsv_size < space_info->bytes_may_use)
+			delalloc_size = space_info->bytes_may_use - block_rsv_size;
+		spin_unlock(&space_info->lock);
+
+		/*
+		 * We don't want to include the global_rsv in our calculation,
+		 * because that's space we can't touch.  Subtract it from the
+		 * block_rsv_size for the next checks.
+		 */
+		block_rsv_size -= global_rsv_size;
+
+		/*
+		 * We really want to avoid flushing delalloc too much, as it
+		 * could result in poor allocation patterns, so only flush it if
+		 * it's larger than the rest of the pools combined.
+		 */
+		if (delalloc_size > block_rsv_size) {
+			to_reclaim = delalloc_size;
+			flush = FLUSH_DELALLOC;
+		} else if (space_info->bytes_pinned >
+			   (delayed_block_rsv->reserved +
+			    delayed_refs_rsv->reserved)) {
+			to_reclaim = space_info->bytes_pinned;
+			flush = FORCE_COMMIT_TRANS;
+		} else if (delayed_block_rsv->reserved >
+			   delayed_refs_rsv->reserved) {
+			to_reclaim = delayed_block_rsv->reserved;
+			flush = FLUSH_DELAYED_ITEMS_NR;
+		} else {
+			to_reclaim = delayed_refs_rsv->reserved;
+			flush = FLUSH_DELAYED_REFS_NR;
+		}
+
+		/*
+		 * We don't want to reclaim everything, just a portion, so scale
+		 * down the to_reclaim by 1/4.  If it takes us down to 0,
+		 * reclaim 1 items worth.
+		 */
+		to_reclaim >>= 2;
+		if (!to_reclaim)
+			to_reclaim = btrfs_calc_insert_metadata_size(fs_info, 1);
+		flush_space(fs_info, space_info, to_reclaim, flush, true);
+		cond_resched();
+		spin_lock(&space_info->lock);
+	}
+
+	/* We only went through once, back off our clamping. */
+	if (loops == 1 && !space_info->reclaim_size)
+		space_info->clamp = max(1, space_info->clamp - 1);
+	trace_btrfs_done_preemptive_reclaim(fs_info, space_info);
+	spin_unlock(&space_info->lock);
+}
+
+/*
  * FLUSH_DELALLOC_WAIT:
  *   Space is freed from flushing delalloc in one of two ways.
  *
@@ -1052,7 +1212,7 @@ static void btrfs_async_reclaim_data_space(struct work_struct *work)
 	struct btrfs_fs_info *fs_info;
 	struct btrfs_space_info *space_info;
 	u64 last_tickets_id;
-	int flush_state = 0;
+	enum btrfs_flush_state flush_state = 0;
 
 	fs_info = container_of(work, struct btrfs_fs_info, async_data_reclaim_work);
 	space_info = fs_info->data_sinfo;
@@ -1067,7 +1227,7 @@ static void btrfs_async_reclaim_data_space(struct work_struct *work)
 	spin_unlock(&space_info->lock);
 
 	while (!space_info->full) {
-		flush_space(fs_info, space_info, U64_MAX, ALLOC_CHUNK_FORCE);
+		flush_space(fs_info, space_info, U64_MAX, ALLOC_CHUNK_FORCE, false);
 		spin_lock(&space_info->lock);
 		if (list_empty(&space_info->tickets)) {
 			space_info->flush = 0;
@@ -1080,7 +1240,7 @@ static void btrfs_async_reclaim_data_space(struct work_struct *work)
 
 	while (flush_state < ARRAY_SIZE(data_flush_states)) {
 		flush_space(fs_info, space_info, U64_MAX,
-			    data_flush_states[flush_state]);
+			    data_flush_states[flush_state], false);
 		spin_lock(&space_info->lock);
 		if (list_empty(&space_info->tickets)) {
 			space_info->flush = 0;
@@ -1113,6 +1273,8 @@ void btrfs_init_async_reclaim_work(struct btrfs_fs_info *fs_info)
 {
 	INIT_WORK(&fs_info->async_reclaim_work, btrfs_async_reclaim_metadata_space);
 	INIT_WORK(&fs_info->async_data_reclaim_work, btrfs_async_reclaim_data_space);
+	INIT_WORK(&fs_info->preempt_reclaim_work,
+		  btrfs_preempt_reclaim_metadata_space);
 }
 
 static const enum btrfs_flush_state priority_flush_states[] = {
@@ -1151,7 +1313,8 @@ static void priority_reclaim_metadata_space(struct btrfs_fs_info *fs_info,
 
 	flush_state = 0;
 	do {
-		flush_space(fs_info, space_info, to_reclaim, states[flush_state]);
+		flush_space(fs_info, space_info, to_reclaim, states[flush_state],
+			    false);
 		flush_state++;
 		spin_lock(&space_info->lock);
 		if (ticket->bytes == 0) {
@@ -1167,7 +1330,7 @@ static void priority_reclaim_data_space(struct btrfs_fs_info *fs_info,
 					struct reserve_ticket *ticket)
 {
 	while (!space_info->full) {
-		flush_space(fs_info, space_info, U64_MAX, ALLOC_CHUNK_FORCE);
+		flush_space(fs_info, space_info, U64_MAX, ALLOC_CHUNK_FORCE, false);
 		spin_lock(&space_info->lock);
 		if (ticket->bytes == 0) {
 			spin_unlock(&space_info->lock);
@@ -1212,11 +1375,14 @@ static void wait_reserve_ticket(struct btrfs_fs_info *fs_info,
 }
 
 /**
- * handle_reserve_ticket - do the appropriate flushing and waiting for a ticket
- * @fs_info - the fs
- * @space_info - the space_info for the reservation
- * @ticket - the ticket for the reservation
- * @flush - how much we can flush
+ * Do the appropriate flushing and waiting for a ticket
+ *
+ * @fs_info:    the filesystem
+ * @space_info: space info for the reservation
+ * @ticket:     ticket for the reservation
+ * @start_ns:   timestamp when the reservation started
+ * @orig_bytes: amount of bytes originally reserved
+ * @flush:      how much we can flush
  *
  * This does the work of figuring out how to flush for the ticket, waiting for
  * the reservation, and returning the appropriate error if there is one.
@@ -1224,6 +1390,7 @@ static void wait_reserve_ticket(struct btrfs_fs_info *fs_info,
 static int handle_reserve_ticket(struct btrfs_fs_info *fs_info,
 				 struct btrfs_space_info *space_info,
 				 struct reserve_ticket *ticket,
+				 u64 start_ns, u64 orig_bytes,
 				 enum btrfs_reserve_flush_enum flush)
 {
 	int ret;
@@ -1279,6 +1446,8 @@ static int handle_reserve_ticket(struct btrfs_fs_info *fs_info,
 	 * space wasn't reserved at all).
 	 */
 	ASSERT(!(ticket->bytes == 0 && ticket->error));
+	trace_btrfs_reserve_ticket(fs_info, space_info->flags, orig_bytes,
+				   start_ns, flush, ticket->error);
 	return ret;
 }
 
@@ -1292,12 +1461,31 @@ static inline bool is_normal_flushing(enum btrfs_reserve_flush_enum flush)
 		(flush == BTRFS_RESERVE_FLUSH_ALL_STEAL);
 }
 
+static inline void maybe_clamp_preempt(struct btrfs_fs_info *fs_info,
+				       struct btrfs_space_info *space_info)
+{
+	u64 ordered = percpu_counter_sum_positive(&fs_info->ordered_bytes);
+	u64 delalloc = percpu_counter_sum_positive(&fs_info->delalloc_bytes);
+
+	/*
+	 * If we're heavy on ordered operations then clamping won't help us.  We
+	 * need to clamp specifically to keep up with dirty'ing buffered
+	 * writers, because there's not a 1:1 correlation of writing delalloc
+	 * and freeing space, like there is with flushing delayed refs or
+	 * delayed nodes.  If we're already more ordered than delalloc then
+	 * we're keeping up, otherwise we aren't and should probably clamp.
+	 */
+	if (ordered < delalloc)
+		space_info->clamp = min(space_info->clamp + 1, 8);
+}
+
 /**
- * reserve_metadata_bytes - try to reserve bytes from the block_rsv's space
- * @root - the root we're allocating for
- * @space_info - the space info we want to allocate from
- * @orig_bytes - the number of bytes we want
- * @flush - whether or not we can flush to make our reservation
+ * Try to reserve bytes from the block_rsv's space
+ *
+ * @fs_info:    the filesystem
+ * @space_info: space info we want to allocate from
+ * @orig_bytes: number of bytes we want
+ * @flush:      whether or not we can flush to make our reservation
  *
  * This will reserve orig_bytes number of bytes from the space info associated
  * with the block_rsv.  If there is not enough space it will make an attempt to
@@ -1312,6 +1500,7 @@ static int __reserve_bytes(struct btrfs_fs_info *fs_info,
 {
 	struct work_struct *async_work;
 	struct reserve_ticket ticket;
+	u64 start_ns = 0;
 	u64 used;
 	int ret = 0;
 	bool pending_tickets;
@@ -1364,6 +1553,9 @@ static int __reserve_bytes(struct btrfs_fs_info *fs_info,
 		space_info->reclaim_size += ticket.bytes;
 		init_waitqueue_head(&ticket.wait);
 		ticket.steal = (flush == BTRFS_RESERVE_FLUSH_ALL_STEAL);
+		if (trace_btrfs_reserve_ticket_enabled())
+			start_ns = ktime_get_ns();
+
 		if (flush == BTRFS_RESERVE_FLUSH_ALL ||
 		    flush == BTRFS_RESERVE_FLUSH_ALL_STEAL ||
 		    flush == BTRFS_RESERVE_FLUSH_DATA) {
@@ -1380,6 +1572,14 @@ static int __reserve_bytes(struct btrfs_fs_info *fs_info,
 			list_add_tail(&ticket.list,
 				      &space_info->priority_tickets);
 		}
+
+		/*
+		 * We were forced to add a reserve ticket, so our preemptive
+		 * flushing is unable to keep up.  Clamp down on the threshold
+		 * for the preemptive flushing in order to keep up with the
+		 * workload.
+		 */
+		maybe_clamp_preempt(fs_info, space_info);
 	} else if (!ret && space_info->flags & BTRFS_BLOCK_GROUP_METADATA) {
 		used += orig_bytes;
 		/*
@@ -1388,27 +1588,29 @@ static int __reserve_bytes(struct btrfs_fs_info *fs_info,
 		 * the async reclaim as we will panic.
 		 */
 		if (!test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags) &&
-		    need_do_async_reclaim(fs_info, space_info, used) &&
-		    !work_busy(&fs_info->async_reclaim_work)) {
+		    need_preemptive_reclaim(fs_info, space_info) &&
+		    !work_busy(&fs_info->preempt_reclaim_work)) {
 			trace_btrfs_trigger_flush(fs_info, space_info->flags,
 						  orig_bytes, flush, "preempt");
 			queue_work(system_unbound_wq,
-				   &fs_info->async_reclaim_work);
+				   &fs_info->preempt_reclaim_work);
 		}
 	}
 	spin_unlock(&space_info->lock);
 	if (!ret || flush == BTRFS_RESERVE_NO_FLUSH)
 		return ret;
 
-	return handle_reserve_ticket(fs_info, space_info, &ticket, flush);
+	return handle_reserve_ticket(fs_info, space_info, &ticket, start_ns,
+				     orig_bytes, flush);
 }
 
 /**
- * reserve_metadata_bytes - try to reserve bytes from the block_rsv's space
- * @root - the root we're allocating for
- * @block_rsv - the block_rsv we're allocating for
- * @orig_bytes - the number of bytes we want
- * @flush - whether or not we can flush to make our reservation
+ * Trye to reserve metadata bytes from the block_rsv's space
+ *
+ * @root:       the root we're allocating for
+ * @block_rsv:  block_rsv we're allocating for
+ * @orig_bytes: number of bytes we want
+ * @flush:      whether or not we can flush to make our reservation
  *
  * This will reserve orig_bytes number of bytes from the space info associated
  * with the block_rsv.  If there is not enough space it will make an attempt to
@@ -1446,10 +1648,11 @@ int btrfs_reserve_metadata_bytes(struct btrfs_root *root,
 }
 
 /**
- * btrfs_reserve_data_bytes - try to reserve data bytes for an allocation
- * @fs_info - the filesystem
- * @bytes - the number of bytes we need
- * @flush - how we are allowed to flush
+ * Try to reserve data bytes for an allocation
+ *
+ * @fs_info: the filesystem
+ * @bytes:   number of bytes we need
+ * @flush:   how we are allowed to flush
  *
  * This will reserve bytes from the data space info.  If there is not enough
  * space then we will attempt to flush space as specified by flush.