Merge branch 'for-linus-4.8' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs

Pull btrfs updates from Chris Mason:
 "This pull is dedicated to Josef's enospc rework, which we've been
  testing for a few releases now.  It fixes some early enospc problems
  and is dramatically faster.

  This also includes an updated fix for the delalloc accounting that
  happens after a fault in copy_from_user.  My patch in v4.7 was almost
  but not quite enough"

* 'for-linus-4.8' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
  Btrfs: fix delalloc accounting after copy_from_user faults
  Btrfs: avoid deadlocks during reservations in btrfs_truncate_block
  Btrfs: use FLUSH_LIMIT for relocation in reserve_metadata_bytes
  Btrfs: fill relocation block rsv after allocation
  Btrfs: always use trans->block_rsv for orphans
  Btrfs: change how we calculate the global block rsv
  Btrfs: use root when checking need_async_flush
  Btrfs: don't bother kicking async if there's nothing to reclaim
  Btrfs: fix release reserved extents trace points
  Btrfs: add fsid to some tracepoints
  Btrfs: add tracepoints for flush events
  Btrfs: fix delalloc reservation amount tracepoint
  Btrfs: trace pinned extents
  Btrfs: introduce ticketed enospc infrastructure
  Btrfs: add tracepoint for adding block groups
  Btrfs: warn_on for unaccounted spaces
  Btrfs: change delayed reservation fallback behavior
  Btrfs: always reserve metadata for delalloc extents
  Btrfs: fix callers of btrfs_block_rsv_migrate
  Btrfs: add bytes_readonly to the spaceinfo at once

7 files changed, 677 insertions, 344 deletions

diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index b2620d1f883f..443fcc402114 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -439,6 +439,8 @@ struct btrfs_space_info {
 	struct list_head list;
 	/* Protected by the spinlock 'lock'. */
 	struct list_head ro_bgs;
+	struct list_head priority_tickets;
+	struct list_head tickets;
 
 	struct rw_semaphore groups_sem;
 	/* for block groups in our same type */
@@ -2624,6 +2626,15 @@ enum btrfs_reserve_flush_enum {
 	BTRFS_RESERVE_FLUSH_ALL,
 };
 
+enum btrfs_flush_state {
+	FLUSH_DELAYED_ITEMS_NR	=	1,
+	FLUSH_DELAYED_ITEMS	=	2,
+	FLUSH_DELALLOC		=	3,
+	FLUSH_DELALLOC_WAIT	=	4,
+	ALLOC_CHUNK		=	5,
+	COMMIT_TRANS		=	6,
+};
+
 int btrfs_check_data_free_space(struct inode *inode, u64 start, u64 len);
 int btrfs_alloc_data_chunk_ondemand(struct inode *inode, u64 bytes);
 void btrfs_free_reserved_data_space(struct inode *inode, u64 start, u64 len);
@@ -2661,8 +2672,8 @@ int btrfs_block_rsv_refill(struct btrfs_root *root,
 			   struct btrfs_block_rsv *block_rsv, u64 min_reserved,
 			   enum btrfs_reserve_flush_enum flush);
 int btrfs_block_rsv_migrate(struct btrfs_block_rsv *src_rsv,
-			    struct btrfs_block_rsv *dst_rsv,
-			    u64 num_bytes);
+			    struct btrfs_block_rsv *dst_rsv, u64 num_bytes,
+			    int update_size);
 int btrfs_cond_migrate_bytes(struct btrfs_fs_info *fs_info,
 			     struct btrfs_block_rsv *dest, u64 num_bytes,
 			     int min_factor);
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index d3aaabbfada0..dd3c040139a2 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -553,7 +553,7 @@ static int btrfs_delayed_item_reserve_metadata(struct btrfs_trans_handle *trans,
 	dst_rsv = &root->fs_info->delayed_block_rsv;
 
 	num_bytes = btrfs_calc_trans_metadata_size(root, 1);
-	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
+	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, 1);
 	if (!ret) {
 		trace_btrfs_space_reservation(root->fs_info, "delayed_item",
 					      item->key.objectid,
@@ -598,6 +598,29 @@ static int btrfs_delayed_inode_reserve_metadata(
 	num_bytes = btrfs_calc_trans_metadata_size(root, 1);
 
 	/*
+	 * If our block_rsv is the delalloc block reserve then check and see if
+	 * we have our extra reservation for updating the inode.  If not fall
+	 * through and try to reserve space quickly.
+	 *
+	 * We used to try and steal from the delalloc block rsv or the global
+	 * reserve, but we'd steal a full reservation, which isn't kind.  We are
+	 * here through delalloc which means we've likely just cowed down close
+	 * to the leaf that contains the inode, so we would steal less just
+	 * doing the fallback inode update, so if we do end up having to steal
+	 * from the global block rsv we hopefully only steal one or two blocks
+	 * worth which is less likely to hurt us.
+	 */
+	if (src_rsv && src_rsv->type == BTRFS_BLOCK_RSV_DELALLOC) {
+		spin_lock(&BTRFS_I(inode)->lock);
+		if (test_and_clear_bit(BTRFS_INODE_DELALLOC_META_RESERVED,
+				       &BTRFS_I(inode)->runtime_flags))
+			release = true;
+		else
+			src_rsv = NULL;
+		spin_unlock(&BTRFS_I(inode)->lock);
+	}
+
+	/*
 	 * btrfs_dirty_inode will update the inode under btrfs_join_transaction
 	 * which doesn't reserve space for speed.  This is a problem since we
 	 * still need to reserve space for this update, so try to reserve the
@@ -626,51 +649,10 @@ static int btrfs_delayed_inode_reserve_metadata(
 						      num_bytes, 1);
 		}
 		return ret;
-	} else if (src_rsv->type == BTRFS_BLOCK_RSV_DELALLOC) {
-		spin_lock(&BTRFS_I(inode)->lock);
-		if (test_and_clear_bit(BTRFS_INODE_DELALLOC_META_RESERVED,
-				       &BTRFS_I(inode)->runtime_flags)) {
-			spin_unlock(&BTRFS_I(inode)->lock);
-			release = true;
-			goto migrate;
-		}
-		spin_unlock(&BTRFS_I(inode)->lock);
-
-		/* Ok we didn't have space pre-reserved.  This shouldn't happen
-		 * too often but it can happen if we do delalloc to an existing
-		 * inode which gets dirtied because of the time update, and then
-		 * isn't touched again until after the transaction commits and
-		 * then we try to write out the data.  First try to be nice and
-		 * reserve something strictly for us.  If not be a pain and try
-		 * to steal from the delalloc block rsv.
-		 */
-		ret = btrfs_block_rsv_add(root, dst_rsv, num_bytes,
-					  BTRFS_RESERVE_NO_FLUSH);
-		if (!ret)
-			goto out;
-
-		ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
-		if (!ret)
-			goto out;
-
-		if (btrfs_test_opt(root, ENOSPC_DEBUG)) {
-			btrfs_debug(root->fs_info,
-				    "block rsv migrate returned %d", ret);
-			WARN_ON(1);
-		}
-		/*
-		 * Ok this is a problem, let's just steal from the global rsv
-		 * since this really shouldn't happen that often.
-		 */
-		ret = btrfs_block_rsv_migrate(&root->fs_info->global_block_rsv,
-					      dst_rsv, num_bytes);
-		goto out;
 	}
 
-migrate:
-	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes);
+	ret = btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, 1);
 
-out:
 	/*
 	 * Migrate only takes a reservation, it doesn't touch the size of the
 	 * block_rsv.  This is to simplify people who don't normally have things
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index b480fd555774..e9376b1657e2 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -111,6 +111,16 @@ static int block_rsv_use_bytes(struct btrfs_block_rsv *block_rsv,
 			       u64 num_bytes);
 int btrfs_pin_extent(struct btrfs_root *root,
 		     u64 bytenr, u64 num_bytes, int reserved);
+static int __reserve_metadata_bytes(struct btrfs_root *root,
+				    struct btrfs_space_info *space_info,
+				    u64 orig_bytes,
+				    enum btrfs_reserve_flush_enum flush);
+static void space_info_add_new_bytes(struct btrfs_fs_info *fs_info,
+				     struct btrfs_space_info *space_info,
+				     u64 num_bytes);
+static void space_info_add_old_bytes(struct btrfs_fs_info *fs_info,
+				     struct btrfs_space_info *space_info,
+				     u64 num_bytes);
 
 static noinline int
 block_group_cache_done(struct btrfs_block_group_cache *cache)
@@ -3913,6 +3923,7 @@ static const char *alloc_name(u64 flags)
 
 static int update_space_info(struct btrfs_fs_info *info, u64 flags,
 			     u64 total_bytes, u64 bytes_used,
+			     u64 bytes_readonly,
 			     struct btrfs_space_info **space_info)
 {
 	struct btrfs_space_info *found;
@@ -3933,8 +3944,11 @@ static int update_space_info(struct btrfs_fs_info *info, u64 flags,
 		found->disk_total += total_bytes * factor;
 		found->bytes_used += bytes_used;
 		found->disk_used += bytes_used * factor;
+		found->bytes_readonly += bytes_readonly;
 		if (total_bytes > 0)
 			found->full = 0;
+		space_info_add_new_bytes(info, found, total_bytes -
+					 bytes_used - bytes_readonly);
 		spin_unlock(&found->lock);
 		*space_info = found;
 		return 0;
@@ -3960,7 +3974,7 @@ static int update_space_info(struct btrfs_fs_info *info, u64 flags,
 	found->disk_used = bytes_used * factor;
 	found->bytes_pinned = 0;
 	found->bytes_reserved = 0;
-	found->bytes_readonly = 0;
+	found->bytes_readonly = bytes_readonly;
 	found->bytes_may_use = 0;
 	found->full = 0;
 	found->max_extent_size = 0;
@@ -3969,6 +3983,8 @@ static int update_space_info(struct btrfs_fs_info *info, u64 flags,
 	found->flush = 0;
 	init_waitqueue_head(&found->wait);
 	INIT_LIST_HEAD(&found->ro_bgs);
+	INIT_LIST_HEAD(&found->tickets);
+	INIT_LIST_HEAD(&found->priority_tickets);
 
 	ret = kobject_init_and_add(&found->kobj, &space_info_ktype,
 				    info->space_info_kobj, "%s",
@@ -4470,7 +4486,7 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans,
 	space_info = __find_space_info(extent_root->fs_info, flags);
 	if (!space_info) {
 		ret = update_space_info(extent_root->fs_info, flags,
-					0, 0, &space_info);
+					0, 0, 0, &space_info);
 		BUG_ON(ret); /* -ENOMEM */
 	}
 	BUG_ON(!space_info); /* Logic error */
@@ -4582,12 +4598,19 @@ static int can_overcommit(struct btrfs_root *root,
 			  struct btrfs_space_info *space_info, u64 bytes,
 			  enum btrfs_reserve_flush_enum flush)
 {
-	struct btrfs_block_rsv *global_rsv = &root->fs_info->global_block_rsv;
-	u64 profile = btrfs_get_alloc_profile(root, 0);
+	struct btrfs_block_rsv *global_rsv;
+	u64 profile;
 	u64 space_size;
 	u64 avail;
 	u64 used;
 
+	/* Don't overcommit when in mixed mode. */
+	if (space_info->flags & BTRFS_BLOCK_GROUP_DATA)
+		return 0;
+
+	BUG_ON(root->fs_info == NULL);
+	global_rsv = &root->fs_info->global_block_rsv;
+	profile = btrfs_get_alloc_profile(root, 0);
 	used = space_info->bytes_used + space_info->bytes_reserved +
 		space_info->bytes_pinned + space_info->bytes_readonly;
 
@@ -4739,6 +4762,11 @@ skip_async:
 			spin_unlock(&space_info->lock);
 			break;
 		}
+		if (list_empty(&space_info->tickets) &&
+		    list_empty(&space_info->priority_tickets)) {
+			spin_unlock(&space_info->lock);
+			break;
+		}
 		spin_unlock(&space_info->lock);
 
 		loops++;
@@ -4807,13 +4835,11 @@ commit:
 	return btrfs_commit_transaction(trans, root);
 }
 
-enum flush_state {
-	FLUSH_DELAYED_ITEMS_NR	=	1,
-	FLUSH_DELAYED_ITEMS	=	2,
-	FLUSH_DELALLOC		=	3,
-	FLUSH_DELALLOC_WAIT	=	4,
-	ALLOC_CHUNK		=	5,
-	COMMIT_TRANS		=	6,
+struct reserve_ticket {
+	u64 bytes;
+	int error;
+	struct list_head list;
+	wait_queue_head_t wait;
 };
 
 static int flush_space(struct btrfs_root *root,
@@ -4866,6 +4892,8 @@ static int flush_space(struct btrfs_root *root,
 		break;
 	}
 
+	trace_btrfs_flush_space(root->fs_info, space_info->flags, num_bytes,
+				orig_bytes, state, ret);
 	return ret;
 }
 
@@ -4873,17 +4901,22 @@ static inline u64
 btrfs_calc_reclaim_metadata_size(struct btrfs_root *root,
 				 struct btrfs_space_info *space_info)
 {
+	struct reserve_ticket *ticket;
 	u64 used;
 	u64 expected;
-	u64 to_reclaim;
+	u64 to_reclaim = 0;
 
 	to_reclaim = min_t(u64, num_online_cpus() * SZ_1M, SZ_16M);
-	spin_lock(&space_info->lock);
 	if (can_overcommit(root, space_info, to_reclaim,
-			   BTRFS_RESERVE_FLUSH_ALL)) {
-		to_reclaim = 0;
-		goto out;
-	}
+			   BTRFS_RESERVE_FLUSH_ALL))
+		return 0;
+
+	list_for_each_entry(ticket, &space_info->tickets, list)
+		to_reclaim += ticket->bytes;
+	list_for_each_entry(ticket, &space_info->priority_tickets, list)
+		to_reclaim += ticket->bytes;
+	if (to_reclaim)
+		return to_reclaim;
 
 	used = space_info->bytes_used + space_info->bytes_reserved +
 	       space_info->bytes_pinned + space_info->bytes_readonly +
@@ -4899,14 +4932,11 @@ btrfs_calc_reclaim_metadata_size(struct btrfs_root *root,
 		to_reclaim = 0;
 	to_reclaim = min(to_reclaim, space_info->bytes_may_use +
 				     space_info->bytes_reserved);
-out:
-	spin_unlock(&space_info->lock);
-
 	return to_reclaim;
 }
 
 static inline int need_do_async_reclaim(struct btrfs_space_info *space_info,
-					struct btrfs_fs_info *fs_info, u64 used)
+					struct btrfs_root *root, u64 used)
 {
 	u64 thresh = div_factor_fine(space_info->total_bytes, 98);
 
@@ -4914,73 +4944,177 @@ static inline int need_do_async_reclaim(struct btrfs_space_info *space_info,
 	if ((space_info->bytes_used + space_info->bytes_reserved) >= thresh)
 		return 0;
 
-	return (used >= thresh && !btrfs_fs_closing(fs_info) &&
-		!test_bit(BTRFS_FS_STATE_REMOUNTING, &fs_info->fs_state));
+	if (!btrfs_calc_reclaim_metadata_size(root, space_info))
+		return 0;
+
+	return (used >= thresh && !btrfs_fs_closing(root->fs_info) &&
+		!test_bit(BTRFS_FS_STATE_REMOUNTING,
+			  &root->fs_info->fs_state));
 }
 
-static int btrfs_need_do_async_reclaim(struct btrfs_space_info *space_info,
-				       struct btrfs_fs_info *fs_info,
-				       int flush_state)
+static void wake_all_tickets(struct list_head *head)
 {
-	u64 used;
-
-	spin_lock(&space_info->lock);
-	/*
-	 * We run out of space and have not got any free space via flush_space,
-	 * so don't bother doing async reclaim.
-	 */
-	if (flush_state > COMMIT_TRANS && space_info->full) {
-		spin_unlock(&space_info->lock);
-		return 0;
-	}
+	struct reserve_ticket *ticket;
 
-	used = space_info->bytes_used + space_info->bytes_reserved +
-	       space_info->bytes_pinned + space_info->bytes_readonly +
-	       space_info->bytes_may_use;
-	if (need_do_async_reclaim(space_info, fs_info, used)) {
-		spin_unlock(&space_info->lock);
-		return 1;
+	while (!list_empty(head)) {
+		ticket = list_first_entry(head, struct reserve_ticket, list);
+		list_del_init(&ticket->list);
+		ticket->error = -ENOSPC;
+		wake_up(&ticket->wait);
 	}
-	spin_unlock(&space_info->lock);
-
-	return 0;
 }
 
+/*
+ * This is for normal flushers, we can wait all goddamned day if we want to.  We
+ * will loop and continuously try to flush as long as we are making progress.
+ * We count progress as clearing off tickets each time we have to loop.
+ */
 static void btrfs_async_reclaim_metadata_space(struct work_struct *work)
 {
+	struct reserve_ticket *last_ticket = NULL;
 	struct btrfs_fs_info *fs_info;
 	struct btrfs_space_info *space_info;
 	u64 to_reclaim;
 	int flush_state;
+	int commit_cycles = 0;
 
 	fs_info = container_of(work, struct btrfs_fs_info, async_reclaim_work);
 	space_info = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA);
 
+	spin_lock(&space_info->lock);
 	to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info->fs_root,
 						      space_info);
-	if (!to_reclaim)
+	if (!to_reclaim) {
+		space_info->flush = 0;
+		spin_unlock(&space_info->lock);
 		return;
+	}
+	last_ticket = list_first_entry(&space_info->tickets,
+				       struct reserve_ticket, list);
+	spin_unlock(&space_info->lock);
 
 	flush_state = FLUSH_DELAYED_ITEMS_NR;
 	do {
+		struct reserve_ticket *ticket;
+		int ret;
+
+		ret = flush_space(fs_info->fs_root, space_info, to_reclaim,
+			    to_reclaim, flush_state);
+		spin_lock(&space_info->lock);
+		if (list_empty(&space_info->tickets)) {
+			space_info->flush = 0;
+			spin_unlock(&space_info->lock);
+			return;
+		}
+		to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info->fs_root,
+							      space_info);
+		ticket = list_first_entry(&space_info->tickets,
+					  struct reserve_ticket, list);
+		if (last_ticket == ticket) {
+			flush_state++;
+		} else {
+			last_ticket = ticket;
+			flush_state = FLUSH_DELAYED_ITEMS_NR;
+			if (commit_cycles)
+				commit_cycles--;
+		}
+
+		if (flush_state > COMMIT_TRANS) {
+			commit_cycles++;
+			if (commit_cycles > 2) {
+				wake_all_tickets(&space_info->tickets);
+				space_info->flush = 0;
+			} else {
+				flush_state = FLUSH_DELAYED_ITEMS_NR;
+			}
+		}
+		spin_unlock(&space_info->lock);
+	} while (flush_state <= COMMIT_TRANS);
+}
+
+void btrfs_init_async_reclaim_work(struct work_struct *work)
+{
+	INIT_WORK(work, btrfs_async_reclaim_metadata_space);
+}
+
+static void priority_reclaim_metadata_space(struct btrfs_fs_info *fs_info,
+					    struct btrfs_space_info *space_info,
+					    struct reserve_ticket *ticket)
+{
+	u64 to_reclaim;
+	int flush_state = FLUSH_DELAYED_ITEMS_NR;
+
+	spin_lock(&space_info->lock);
+	to_reclaim = btrfs_calc_reclaim_metadata_size(fs_info->fs_root,
+						      space_info);
+	if (!to_reclaim) {
+		spin_unlock(&space_info->lock);
+		return;
+	}
+	spin_unlock(&space_info->lock);
+
+	do {
 		flush_space(fs_info->fs_root, space_info, to_reclaim,
 			    to_reclaim, flush_state);
 		flush_state++;
-		if (!btrfs_need_do_async_reclaim(space_info, fs_info,
-						 flush_state))
+		spin_lock(&space_info->lock);
+		if (ticket->bytes == 0) {
+			spin_unlock(&space_info->lock);
 			return;
+		}
+		spin_unlock(&space_info->lock);
+
+		/*
+		 * Priority flushers can't wait on delalloc without
+		 * deadlocking.
+		 */
+		if (flush_state == FLUSH_DELALLOC ||
+		    flush_state == FLUSH_DELALLOC_WAIT)
+			flush_state = ALLOC_CHUNK;
 	} while (flush_state < COMMIT_TRANS);
 }
 
-void btrfs_init_async_reclaim_work(struct work_struct *work)
+static int wait_reserve_ticket(struct btrfs_fs_info *fs_info,
+			       struct btrfs_space_info *space_info,
+			       struct reserve_ticket *ticket, u64 orig_bytes)
+
 {
-	INIT_WORK(work, btrfs_async_reclaim_metadata_space);
+	DEFINE_WAIT(wait);
+	int ret = 0;
+
+	spin_lock(&space_info->lock);
+	while (ticket->bytes > 0 && ticket->error == 0) {
+		ret = prepare_to_wait_event(&ticket->wait, &wait, TASK_KILLABLE);
+		if (ret) {
+			ret = -EINTR;
+			break;
+		}
+		spin_unlock(&space_info->lock);
+
+		schedule();
+
+		finish_wait(&ticket->wait, &wait);
+		spin_lock(&space_info->lock);
+	}
+	if (!ret)
+		ret = ticket->error;
+	if (!list_empty(&ticket->list))
+		list_del_init(&ticket->list);
+	if (ticket->bytes && ticket->bytes < orig_bytes) {
+		u64 num_bytes = orig_bytes - ticket->bytes;
+		space_info->bytes_may_use -= num_bytes;
+		trace_btrfs_space_reservation(fs_info, "space_info",
+					      space_info->flags, num_bytes, 0);
+	}
+	spin_unlock(&space_info->lock);
+
+	return ret;
 }
 
 /**
  * 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
+ * @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
  *
@@ -4991,81 +5125,36 @@ void btrfs_init_async_reclaim_work(struct work_struct *work)
  * regain reservations will be made and this will fail if there is not enough
  * space already.
  */
-static int reserve_metadata_bytes(struct btrfs_root *root,
-				  struct btrfs_block_rsv *block_rsv,
-				  u64 orig_bytes,
-				  enum btrfs_reserve_flush_enum flush)
+static int __reserve_metadata_bytes(struct btrfs_root *root,
+				    struct btrfs_space_info *space_info,
+				    u64 orig_bytes,
+				    enum btrfs_reserve_flush_enum flush)
 {
-	struct btrfs_space_info *space_info = block_rsv->space_info;
+	struct reserve_ticket ticket;
 	u64 used;
-	u64 num_bytes = orig_bytes;
-	int flush_state = FLUSH_DELAYED_ITEMS_NR;
 	int ret = 0;
-	bool flushing = false;
 
-again:
-	ret = 0;
-	spin_lock(&space_info->lock);
-	/*
-	 * We only want to wait if somebody other than us is flushing and we
-	 * are actually allowed to flush all things.
-	 */
-	while (flush == BTRFS_RESERVE_FLUSH_ALL && !flushing &&
-	       space_info->flush) {
-		spin_unlock(&space_info->lock);
-		/*
-		 * If we have a trans handle we can't wait because the flusher
-		 * may have to commit the transaction, which would mean we would
-		 * deadlock since we are waiting for the flusher to finish, but
-		 * hold the current transaction open.
-		 */
-		if (current->journal_info)
-			return -EAGAIN;
-		ret = wait_event_killable(space_info->wait, !space_info->flush);
-		/* Must have been killed, return */
-		if (ret)
-			return -EINTR;
-
-		spin_lock(&space_info->lock);
-	}
+	ASSERT(orig_bytes);
+	ASSERT(!current->journal_info || flush != BTRFS_RESERVE_FLUSH_ALL);
 
+	spin_lock(&space_info->lock);
 	ret = -ENOSPC;
 	used = space_info->bytes_used + space_info->bytes_reserved +
 		space_info->bytes_pinned + space_info->bytes_readonly +
 		space_info->bytes_may_use;
 
 	/*
-	 * The idea here is that we've not already over-reserved the block group
-	 * then we can go ahead and save our reservation first and then start
-	 * flushing if we need to.  Otherwise if we've already overcommitted
-	 * lets start flushing stuff first and then come back and try to make
-	 * our reservation.
+	 * If we have enough space then hooray, make our reservation and carry
+	 * on.  If not see if we can overcommit, and if we can, hooray carry on.
+	 * If not things get more complicated.
 	 */
-	if (used <= space_info->total_bytes) {
-		if (used + orig_bytes <= space_info->total_bytes) {
-			space_info->bytes_may_use += orig_bytes;
-			trace_btrfs_space_reservation(root->fs_info,
-				"space_info", space_info->flags, orig_bytes, 1);
-			ret = 0;
-		} else {
-			/*
-			 * Ok set num_bytes to orig_bytes since we aren't
-			 * overocmmitted, this way we only try and reclaim what
-			 * we need.
-			 */
-			num_bytes = orig_bytes;
-		}
-	} else {
-		/*
-		 * Ok we're over committed, set num_bytes to the overcommitted
-		 * amount plus the amount of bytes that we need for this
-		 * reservation.
-		 */
-		num_bytes = used - space_info->total_bytes +
-			(orig_bytes * 2);
-	}
-
-	if (ret && can_overcommit(root, space_info, orig_bytes, flush)) {
+	if (used + orig_bytes <= space_info->total_bytes) {
+		space_info->bytes_may_use += orig_bytes;
+		trace_btrfs_space_reservation(root->fs_info, "space_info",
+					      space_info->flags, orig_bytes,
+					      1);
+		ret = 0;
+	} else if (can_overcommit(root, space_info, orig_bytes, flush)) {
 		space_info->bytes_may_use += orig_bytes;
 		trace_btrfs_space_reservation(root->fs_info, "space_info",
 					      space_info->flags, orig_bytes,
@@ -5074,16 +5163,31 @@ again:
 	}
 
 	/*
-	 * Couldn't make our reservation, save our place so while we're trying
-	 * to reclaim space we can actually use it instead of somebody else
-	 * stealing it from us.
+	 * If we couldn't make a reservation then setup our reservation ticket
+	 * and kick the async worker if it's not already running.
 	 *
-	 * We make the other tasks wait for the flush only when we can flush
-	 * all things.
+	 * If we are a priority flusher then we just need to add our ticket to
+	 * the list and we will do our own flushing further down.
 	 */
 	if (ret && flush != BTRFS_RESERVE_NO_FLUSH) {
-		flushing = true;
-		space_info->flush = 1;
+		ticket.bytes = orig_bytes;
+		ticket.error = 0;
+		init_waitqueue_head(&ticket.wait);
+		if (flush == BTRFS_RESERVE_FLUSH_ALL) {
+			list_add_tail(&ticket.list, &space_info->tickets);
+			if (!space_info->flush) {
+				space_info->flush = 1;
+				trace_btrfs_trigger_flush(root->fs_info,
+							  space_info->flags,
+							  orig_bytes, flush,
+							  "enospc");
+				queue_work(system_unbound_wq,
+					   &root->fs_info->async_reclaim_work);
+			}
+		} else {
+			list_add_tail(&ticket.list,
+				      &space_info->priority_tickets);
+		}
 	} else if (!ret && space_info->flags & BTRFS_BLOCK_GROUP_METADATA) {
 		used += orig_bytes;
 		/*
@@ -5092,39 +5196,67 @@ again:
 		 * the async reclaim as we will panic.
 		 */
 		if (!root->fs_info->log_root_recovering &&
-		    need_do_async_reclaim(space_info, root->fs_info, used) &&
-		    !work_busy(&root->fs_info->async_reclaim_work))
+		    need_do_async_reclaim(space_info, root, used) &&
+		    !work_busy(&root->fs_info->async_reclaim_work)) {
+			trace_btrfs_trigger_flush(root->fs_info,
+						  space_info->flags,
+						  orig_bytes, flush,
+						  "preempt");
 			queue_work(system_unbound_wq,
 				   &root->fs_info->async_reclaim_work);
+		}
 	}
 	spin_unlock(&space_info->lock);
-
 	if (!ret || flush == BTRFS_RESERVE_NO_FLUSH)
-		goto out;
+		return ret;
 
-	ret = flush_space(root, space_info, num_bytes, orig_bytes,
-			  flush_state);
-	flush_state++;
+	if (flush == BTRFS_RESERVE_FLUSH_ALL)
+		return wait_reserve_ticket(root->fs_info, space_info, &ticket,
+					   orig_bytes);
 
-	/*
-	 * If we are FLUSH_LIMIT, we can not flush delalloc, or the deadlock
-	 * would happen. So skip delalloc flush.
-	 */
-	if (flush == BTRFS_RESERVE_FLUSH_LIMIT &&
-	    (flush_state == FLUSH_DELALLOC ||
-	     flush_state == FLUSH_DELALLOC_WAIT))
-		flush_state = ALLOC_CHUNK;
+	ret = 0;
+	priority_reclaim_metadata_space(root->fs_info, space_info, &ticket);
+	spin_lock(&space_info->lock);
+	if (ticket.bytes) {
+		if (ticket.bytes < orig_bytes) {
+			u64 num_bytes = orig_bytes - ticket.bytes;
+			space_info->bytes_may_use -= num_bytes;
+			trace_btrfs_space_reservation(root->fs_info,
+					"space_info", space_info->flags,
+					num_bytes, 0);
 
-	if (!ret)
-		goto again;
-	else if (flush == BTRFS_RESERVE_FLUSH_LIMIT &&
-		 flush_state < COMMIT_TRANS)
-		goto again;
-	else if (flush == BTRFS_RESERVE_FLUSH_ALL &&
-		 flush_state <= COMMIT_TRANS)
-		goto again;
+		}
+		list_del_init(&ticket.list);
+		ret = -ENOSPC;
+	}
+	spin_unlock(&space_info->lock);
+	ASSERT(list_empty(&ticket.list));
+	return ret;
+}
 
-out:
+/**
+ * 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
+ *
+ * This will reserve orgi_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
+ * flush out space to make room.  It will do this by flushing delalloc if
+ * possible or committing the transaction.  If flush is 0 then no attempts to
+ * regain reservations will be made and this will fail if there is not enough
+ * space already.
+ */
+static int reserve_metadata_bytes(struct btrfs_root *root,
+				  struct btrfs_block_rsv *block_rsv,
+				  u64 orig_bytes,
+				  enum btrfs_reserve_flush_enum flush)
+{
+	int ret;
+
+	ret = __reserve_metadata_bytes(root, block_rsv->space_info, orig_bytes,
+				       flush);
 	if (ret == -ENOSPC &&
 	    unlikely(root->orphan_cleanup_state == ORPHAN_CLEANUP_STARTED)) {
 		struct btrfs_block_rsv *global_rsv =
@@ -5137,13 +5269,8 @@ out:
 	if (ret == -ENOSPC)
 		trace_btrfs_space_reservation(root->fs_info,
 					      "space_info:enospc",
-					      space_info->flags, orig_bytes, 1);
-	if (flushing) {
-		spin_lock(&space_info->lock);
-		space_info->flush = 0;
-		wake_up_all(&space_info->wait);
-		spin_unlock(&space_info->lock);
-	}
+					      block_rsv->space_info->flags,
+					      orig_bytes, 1);
 	return ret;
 }
 
@@ -5219,6 +5346,108 @@ int btrfs_cond_migrate_bytes(struct btrfs_fs_info *fs_info,
 	return 0;
 }
 
+/*
+ * This is for space we already have accounted in space_info->bytes_may_use, so
+ * basically when we're returning space from block_rsv's.
+ */
+static void space_info_add_old_bytes(struct btrfs_fs_info *fs_info,
+				     struct btrfs_space_info *space_info,
+				     u64 num_bytes)
+{
+	struct reserve_ticket *ticket;
+	struct list_head *head;
+	u64 used;
+	enum btrfs_reserve_flush_enum flush = BTRFS_RESERVE_NO_FLUSH;
+	bool check_overcommit = false;
+
+	spin_lock(&space_info->lock);
+	head = &space_info->priority_tickets;
+
+	/*
+	 * If we are over our limit then we need to check and see if we can
+	 * overcommit, and if we can't then we just need to free up our space
+	 * and not satisfy any requests.
+	 */
+	used = space_info->bytes_used + space_info->bytes_reserved +
+		space_info->bytes_pinned + space_info->bytes_readonly +
+		space_info->bytes_may_use;
+	if (used - num_bytes >= space_info->total_bytes)
+		check_overcommit = true;
+again:
+	while (!list_empty(head) && num_bytes) {
+		ticket = list_first_entry(head, struct reserve_ticket,
+					  list);
+		/*
+		 * We use 0 bytes because this space is already reserved, so
+		 * adding the ticket space would be a double count.
+		 */
+		if (check_overcommit &&
+		    !can_overcommit(fs_info->extent_root, space_info, 0,
+				    flush))
+			break;
+		if (num_bytes >= ticket->bytes) {
+			list_del_init(&ticket->list);
+			num_bytes -= ticket->bytes;
+			ticket->bytes = 0;
+			wake_up(&ticket->wait);
+		} else {
+			ticket->bytes -= num_bytes;
+			num_bytes = 0;
+		}
+	}
+
+	if (num_bytes && head == &space_info->priority_tickets) {
+		head = &space_info->tickets;
+		flush = BTRFS_RESERVE_FLUSH_ALL;
+		goto again;
+	}
+	space_info->bytes_may_use -= num_bytes;
+	trace_btrfs_space_reservation(fs_info, "space_info",
+				      space_info->flags, num_bytes, 0);
+	spin_unlock(&space_info->lock);
+}
+
+/*
+ * This is for newly allocated space that isn't accounted in
+ * space_info->bytes_may_use yet.  So if we allocate a chunk or unpin an extent
+ * we use this helper.
+ */
+static void space_info_add_new_bytes(struct btrfs_fs_info *fs_info,
+				     struct btrfs_space_info *space_info,
+				     u64 num_bytes)
+{
+	struct reserve_ticket *ticket;
+	struct list_head *head = &space_info->priority_tickets;
+
+again:
+	while (!list_empty(head) && num_bytes) {
+		ticket = list_first_entry(head, struct reserve_ticket,
+					  list);
+		if (num_bytes >= ticket->bytes) {
+			trace_btrfs_space_reservation(fs_info, "space_info",
+						      space_info->flags,
+						      ticket->bytes, 1);
+			list_del_init(&ticket->list);
+			num_bytes -= ticket->bytes;
+			space_info->bytes_may_use += ticket->bytes;
+			ticket->bytes = 0;
+			wake_up(&ticket->wait);
+		} else {
+			trace_btrfs_space_reservation(fs_info, "space_info",
+						      space_info->flags,
+						      num_bytes, 1);
+			space_info->bytes_may_use += num_bytes;
+			ticket->bytes -= num_bytes;
+			num_bytes = 0;
+		}
+	}
+
+	if (num_bytes && head == &space_info->priority_tickets) {
+		head = &space_info->tickets;
+		goto again;
+	}
+}
+
 static void block_rsv_release_bytes(struct btrfs_fs_info *fs_info,
 				    struct btrfs_block_rsv *block_rsv,
 				    struct btrfs_block_rsv *dest, u64 num_bytes)
@@ -5253,18 +5482,15 @@ static void block_rsv_release_bytes(struct btrfs_fs_info *fs_info,
 			}
 			spin_unlock(&dest->lock);
 		}
-		if (num_bytes) {
-			spin_lock(&space_info->lock);
-			space_info->bytes_may_use -= num_bytes;
-			trace_btrfs_space_reservation(fs_info, "space_info",
-					space_info->flags, num_bytes, 0);
-			spin_unlock(&space_info->lock);
-		}
+		if (num_bytes)
+			space_info_add_old_bytes(fs_info, space_info,
+						 num_bytes);
 	}
 }
 
-static int block_rsv_migrate_bytes(struct btrfs_block_rsv *src,
-				   struct btrfs_block_rsv *dst, u64 num_bytes)
+int btrfs_block_rsv_migrate(struct btrfs_block_rsv *src,
+			    struct btrfs_block_rsv *dst, u64 num_bytes,
+			    int update_size)
 {
 	int ret;
 
@@ -5272,7 +5498,7 @@ static int block_rsv_migrate_bytes(struct btrfs_block_rsv *src,
 	if (ret)
 		return ret;
 
-	block_rsv_add_bytes(dst, num_bytes, 1);
+	block_rsv_add_bytes(dst, num_bytes, update_size);
 	return 0;
 }
 
@@ -5379,13 +5605,6 @@ int btrfs_block_rsv_refill(struct btrfs_root *root,
 	return ret;
 }
 
-int btrfs_block_rsv_migrate(struct btrfs_block_rsv *src_rsv,
-			    struct btrfs_block_rsv *dst_rsv,
-			    u64 num_bytes)
-{
-	return block_rsv_migrate_bytes(src_rsv, dst_rsv, num_bytes);
-}
-
 void btrfs_block_rsv_release(struct btrfs_root *root,
 			     struct btrfs_block_rsv *block_rsv,
 			     u64 num_bytes)
@@ -5398,48 +5617,21 @@ void btrfs_block_rsv_release(struct btrfs_root *root,
 				num_bytes);
 }
 
-/*
- * helper to calculate size of global block reservation.
- * the desired value is sum of space used by extent tree,
- * checksum tree and root tree
- */
-static u64 calc_global_metadata_size(struct btrfs_fs_info *fs_info)
-{
-	struct btrfs_space_info *sinfo;
-	u64 num_bytes;
-	u64 meta_used;
-	u64 data_used;
-	int csum_size = btrfs_super_csum_size(fs_info->super_copy);
-
-	sinfo = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_DATA);
-	spin_lock(&sinfo->lock);
-	data_used = sinfo->bytes_used;
-	spin_unlock(&sinfo->lock);
-
-	sinfo = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA);
-	spin_lock(&sinfo->lock);
-	if (sinfo->flags & BTRFS_BLOCK_GROUP_DATA)
-		data_used = 0;
-	meta_used = sinfo->bytes_used;
-	spin_unlock(&sinfo->lock);
-
-	num_bytes = (data_used >> fs_info->sb->s_blocksize_bits) *
-		    csum_size * 2;
-	num_bytes += div_u64(data_used + meta_used, 50);
-
-	if (num_bytes * 3 > meta_used)
-		num_bytes = div_u64(meta_used, 3);
-
-	return ALIGN(num_bytes, fs_info->extent_root->nodesize << 10);
-}
-
 static void update_global_block_rsv(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_block_rsv *block_rsv = &fs_info->global_block_rsv;
 	struct btrfs_space_info *sinfo = block_rsv->space_info;
 	u64 num_bytes;
 
-	num_bytes = calc_global_metadata_size(fs_info);
+	/*
+	 * The global block rsv is based on the size of the extent tree, the
+	 * checksum tree and the root tree.  If the fs is empty we want to set
+	 * it to a minimal amount for safety.
+	 */
+	num_bytes = btrfs_root_used(&fs_info->extent_root->root_item) +
+		btrfs_root_used(&fs_info->csum_root->root_item) +
+		btrfs_root_used(&fs_info->tree_root->root_item);
+	num_bytes = max_t(u64, num_bytes, SZ_16M);
 
 	spin_lock(&sinfo->lock);
 	spin_lock(&block_rsv->lock);
@@ -5554,7 +5746,13 @@ int btrfs_orphan_reserve_metadata(struct btrfs_trans_handle *trans,
 				  struct inode *inode)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
-	struct btrfs_block_rsv *src_rsv = get_block_rsv(trans, root);
+	/*
+	 * We always use trans->block_rsv here as we will have reserved space
+	 * for our orphan when starting the transaction, using get_block_rsv()
+	 * here will sometimes make us choose the wrong block rsv as we could be
+	 * doing a reloc inode for a non refcounted root.
+	 */
+	struct btrfs_block_rsv *src_rsv = trans->block_rsv;
 	struct btrfs_block_rsv *dst_rsv = root->orphan_block_rsv;
 
 	/*
@@ -5565,7 +5763,7 @@ int btrfs_orphan_reserve_metadata(struct btrfs_trans_handle *trans,
 	u64 num_bytes = btrfs_calc_trans_metadata_size(root, 1);
 	trace_btrfs_space_reservation(root->fs_info, "orphan",
 				      btrfs_ino(inode), num_bytes, 1);
-	return block_rsv_migrate_bytes(src_rsv, dst_rsv, num_bytes);
+	return btrfs_block_rsv_migrate(src_rsv, dst_rsv, num_bytes, 1);
 }
 
 void btrfs_orphan_release_metadata(struct inode *inode)
@@ -5620,7 +5818,7 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root,
 				  BTRFS_RESERVE_FLUSH_ALL);
 
 	if (ret == -ENOSPC && use_global_rsv)
-		ret = btrfs_block_rsv_migrate(global_rsv, rsv, num_bytes);
+		ret = btrfs_block_rsv_migrate(global_rsv, rsv, num_bytes, 1);
 
 	if (ret && *qgroup_reserved)
 		btrfs_qgroup_free_meta(root, *qgroup_reserved);
@@ -5730,21 +5928,26 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
 	u64 to_reserve = 0;
 	u64 csum_bytes;
 	unsigned nr_extents = 0;
-	int extra_reserve = 0;
 	enum btrfs_reserve_flush_enum flush = BTRFS_RESERVE_FLUSH_ALL;
 	int ret = 0;
 	bool delalloc_lock = true;
 	u64 to_free = 0;
 	unsigned dropped;
+	bool release_extra = false;
 
 	/* If we are a free space inode we need to not flush since we will be in
 	 * the middle of a transaction commit.  We also don't need the delalloc
 	 * mutex since we won't race with anybody.  We need this mostly to make
 	 * lockdep shut its filthy mouth.
+	 *
+	 * If we have a transaction open (can happen if we call truncate_block
+	 * from truncate), then we need FLUSH_LIMIT so we don't deadlock.
 	 */
 	if (btrfs_is_free_space_inode(inode)) {
 		flush = BTRFS_RESERVE_NO_FLUSH;
 		delalloc_lock = false;
+	} else if (current->journal_info) {
+		flush = BTRFS_RESERVE_FLUSH_LIMIT;
 	}
 
 	if (flush != BTRFS_RESERVE_NO_FLUSH &&
@@ -5761,24 +5964,15 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
 					 BTRFS_MAX_EXTENT_SIZE - 1,
 					 BTRFS_MAX_EXTENT_SIZE);
 	BTRFS_I(inode)->outstanding_extents += nr_extents;
-	nr_extents = 0;
 
+	nr_extents = 0;
 	if (BTRFS_I(inode)->outstanding_extents >
 	    BTRFS_I(inode)->reserved_extents)
-		nr_extents = BTRFS_I(inode)->outstanding_extents -
+		nr_extents += BTRFS_I(inode)->outstanding_extents -
 			BTRFS_I(inode)->reserved_extents;
 
-	/*
-	 * Add an item to reserve for updating the inode when we complete the
-	 * delalloc io.
-	 */
-	if (!test_bit(BTRFS_INODE_DELALLOC_META_RESERVED,
-		      &BTRFS_I(inode)->runtime_flags)) {
-		nr_extents++;
-		extra_reserve = 1;
-	}
-
-	to_reserve = btrfs_calc_trans_metadata_size(root, nr_extents);
+	/* We always want to reserve a slot for updating the inode. */
+	to_reserve = btrfs_calc_trans_metadata_size(root, nr_extents + 1);
 	to_reserve += calc_csum_metadata_size(inode, num_bytes, 1);
 	csum_bytes = BTRFS_I(inode)->csum_bytes;
 	spin_unlock(&BTRFS_I(inode)->lock);
@@ -5790,17 +5984,17 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
 			goto out_fail;
 	}
 
-	ret = reserve_metadata_bytes(root, block_rsv, to_reserve, flush);
+	ret = btrfs_block_rsv_add(root, block_rsv, to_reserve, flush);
 	if (unlikely(ret)) {
 		btrfs_qgroup_free_meta(root, nr_extents * root->nodesize);
 		goto out_fail;
 	}
 
 	spin_lock(&BTRFS_I(inode)->lock);
-	if (extra_reserve) {
-		set_bit(BTRFS_INODE_DELALLOC_META_RESERVED,
-			&BTRFS_I(inode)->runtime_flags);
-		nr_extents--;
+	if (test_and_set_bit(BTRFS_INODE_DELALLOC_META_RESERVED,
+			     &BTRFS_I(inode)->runtime_flags)) {
+		to_reserve -= btrfs_calc_trans_metadata_size(root, 1);
+		release_extra = true;
 	}
 	BTRFS_I(inode)->reserved_extents += nr_extents;
 	spin_unlock(&BTRFS_I(inode)->lock);
@@ -5811,8 +6005,10 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
 	if (to_reserve)
 		trace_btrfs_space_reservation(root->fs_info, "delalloc",
 					      btrfs_ino(inode), to_reserve, 1);
-	block_rsv_add_bytes(block_rsv, to_reserve, 1);
-
+	if (release_extra)
+		btrfs_block_rsv_release(root, block_rsv,
+					btrfs_calc_trans_metadata_size(root,
+								       1));
 	return 0;
 
 out_fail:
@@ -6044,6 +6240,9 @@ static int update_block_group(struct btrfs_trans_handle *trans,
 			spin_unlock(&cache->lock);
 			spin_unlock(&cache->space_info->lock);
 
+			trace_btrfs_space_reservation(root->fs_info, "pinned",
+						      cache->space_info->flags,
+						      num_bytes, 1);
 			set_extent_dirty(info->pinned_extents,
 					 bytenr, bytenr + num_bytes - 1,
 					 GFP_NOFS | __GFP_NOFAIL);
@@ -6118,10 +6317,10 @@ static int pin_down_extent(struct btrfs_root *root,
 	spin_unlock(&cache->lock);
 	spin_unlock(&cache->space_info->lock);
 
+	trace_btrfs_space_reservation(root->fs_info, "pinned",
+				      cache->space_info->flags, num_bytes, 1);
 	set_extent_dirty(root->fs_info->pinned_extents, bytenr,
 			 bytenr + num_bytes - 1, GFP_NOFS | __GFP_NOFAIL);
-	if (reserved)
-		trace_btrfs_reserved_extent_free(root, bytenr, num_bytes);
 	return 0;
 }
 
@@ -6476,6 +6675,9 @@ static int unpin_extent_range(struct btrfs_root *root, u64 start, u64 end,
 		spin_lock(&cache->lock);
 		cache->pinned -= len;
 		space_info->bytes_pinned -= len;
+
+		trace_btrfs_space_reservation(fs_info, "pinned",
+					      space_info->flags, len, 0);
 		space_info->max_extent_size = 0;
 		percpu_counter_add(&space_info->total_bytes_pinned, -len);
 		if (cache->ro) {
@@ -6483,17 +6685,29 @@ static int unpin_extent_range(struct btrfs_root *root, u64 start, u64 end,
 			readonly = true;
 		}
 		spin_unlock(&cache->lock);
-		if (!readonly && global_rsv->space_info == space_info) {
+		if (!readonly && return_free_space &&
+		    global_rsv->space_info == space_info) {
+			u64 to_add = len;
+			WARN_ON(!return_free_space);
 			spin_lock(&global_rsv->lock);
 			if (!global_rsv->full) {
-				len = min(len, global_rsv->size -
-					  global_rsv->reserved);
-				global_rsv->reserved += len;
-				space_info->bytes_may_use += len;
+				to_add = min(len, global_rsv->size -
+					     global_rsv->reserved);
+				global_rsv->reserved += to_add;
+				space_info->bytes_may_use += to_add;
 				if (global_rsv->reserved >= global_rsv->size)
 					global_rsv->full = 1;
+				trace_btrfs_space_reservation(fs_info,
+							      "space_info",
+							      space_info->flags,
+							      to_add, 1);
+				len -= to_add;
 			}
 			spin_unlock(&global_rsv->lock);
+			/* Add to any tickets we may have */
+			if (len)
+				space_info_add_new_bytes(fs_info, space_info,
+							 len);
 		}
 		spin_unlock(&space_info->lock);
 	}
@@ -7782,12 +7996,10 @@ static int __btrfs_free_reserved_extent(struct btrfs_root *root,
 			ret = btrfs_discard_extent(root, start, len, NULL);
 		btrfs_add_free_space(cache, start, len);
 		btrfs_update_reserved_bytes(cache, len, RESERVE_FREE, delalloc);
+		trace_btrfs_reserved_extent_free(root, start, len);
 	}
 
 	btrfs_put_block_group(cache);
-
-	trace_btrfs_reserved_extent_free(root, start, len);
-
 	return ret;
 }
 
@@ -9791,13 +10003,15 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 		space_info = list_entry(info->space_info.next,
 					struct btrfs_space_info,
 					list);
-		if (btrfs_test_opt(info->tree_root, ENOSPC_DEBUG)) {
-			if (WARN_ON(space_info->bytes_pinned > 0 ||
+
+		/*
+		 * Do not hide this behind enospc_debug, this is actually
+		 * important and indicates a real bug if this happens.
+		 */
+		if (WARN_ON(space_info->bytes_pinned > 0 ||
 			    space_info->bytes_reserved > 0 ||
-			    space_info->bytes_may_use > 0)) {
-				dump_space_info(space_info, 0, 0);
-			}
-		}
+			    space_info->bytes_may_use > 0))
+			dump_space_info(space_info, 0, 0);
 		list_del(&space_info->list);
 		for (i = 0; i < BTRFS_NR_RAID_TYPES; i++) {
 			struct kobject *kobj;
@@ -10005,9 +10219,10 @@ int btrfs_read_block_groups(struct btrfs_root *root)
 			goto error;
 		}
 
+		trace_btrfs_add_block_group(root->fs_info, cache, 0);
 		ret = update_space_info(info, cache->flags, found_key.offset,
 					btrfs_block_group_used(&cache->item),
-					&space_info);
+					cache->bytes_super, &space_info);
 		if (ret) {
 			btrfs_remove_free_space_cache(cache);
 			spin_lock(&info->block_group_cache_lock);
@@ -10020,9 +10235,6 @@ int btrfs_read_block_groups(struct btrfs_root *root)
 		}
 
 		cache->space_info = space_info;
-		spin_lock(&cache->space_info->lock);
-		cache->space_info->bytes_readonly += cache->bytes_super;
-		spin_unlock(&cache->space_info->lock);
 
 		__link_block_group(space_info, cache);
 
@@ -10114,7 +10326,6 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
 	int ret;
 	struct btrfs_root *extent_root;
 	struct btrfs_block_group_cache *cache;
-
 	extent_root = root->fs_info->extent_root;
 
 	btrfs_set_log_full_commit(root->fs_info, trans);
@@ -10160,7 +10371,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
 	 * assigned to our block group, but don't update its counters just yet.
 	 * We want our bg to be added to the rbtree with its ->space_info set.
 	 */
-	ret = update_space_info(root->fs_info, cache->flags, 0, 0,
+	ret = update_space_info(root->fs_info, cache->flags, 0, 0, 0,
 				&cache->space_info);
 	if (ret) {
 		btrfs_remove_free_space_cache(cache);
@@ -10179,8 +10390,9 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
 	 * Now that our block group has its ->space_info set and is inserted in
 	 * the rbtree, update the space info's counters.
 	 */
+	trace_btrfs_add_block_group(root->fs_info, cache, 1);
 	ret = update_space_info(root->fs_info, cache->flags, size, bytes_used,
-				&cache->space_info);
+				cache->bytes_super, &cache->space_info);
 	if (ret) {
 		btrfs_remove_free_space_cache(cache);
 		spin_lock(&root->fs_info->block_group_cache_lock);
@@ -10193,16 +10405,11 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans,
 	}
 	update_global_block_rsv(root->fs_info);
 
-	spin_lock(&cache->space_info->lock);
-	cache->space_info->bytes_readonly += cache->bytes_super;
-	spin_unlock(&cache->space_info->lock);
-
 	__link_block_group(cache->space_info, cache);
 
 	list_add_tail(&cache->bg_list, &trans->new_bgs);
 
 	set_avail_alloc_bits(extent_root->fs_info, type);
-
 	return 0;
 }
 
@@ -10747,21 +10954,21 @@ int btrfs_init_space_info(struct btrfs_fs_info *fs_info)
 		mixed = 1;
 
 	flags = BTRFS_BLOCK_GROUP_SYSTEM;
-	ret = update_space_info(fs_info, flags, 0, 0, &space_info);
+	ret = update_space_info(fs_info, flags, 0, 0, 0, &space_info);
 	if (ret)
 		goto out;
 
 	if (mixed) {
 		flags = BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_DATA;
-		ret = update_space_info(fs_info, flags, 0, 0, &space_info);
+		ret = update_space_info(fs_info, flags, 0, 0, 0, &space_info);
 	} else {
 		flags = BTRFS_BLOCK_GROUP_METADATA;
-		ret = update_space_info(fs_info, flags, 0, 0, &space_info);
+		ret = update_space_info(fs_info, flags, 0, 0, 0, &space_info);
 		if (ret)
 			goto out;
 
 		flags = BTRFS_BLOCK_GROUP_DATA;
-		ret = update_space_info(fs_info, flags, 0, 0, &space_info);
+		ret = update_space_info(fs_info, flags, 0, 0, 0, &space_info);
 	}
 out:
 	return ret;
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 2234e88cf674..bcfb4a27ddd4 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1629,13 +1629,11 @@ again:
 		 * managed to copy.
 		 */
 		if (num_sectors > dirty_sectors) {
-			/*
-			 * we round down because we don't want to count
-			 * any partial blocks actually sent through the
-			 * IO machines
-			 */
-			release_bytes = round_down(release_bytes - copied,
-				      root->sectorsize);
+
+			/* release everything except the sectors we dirtied */
+			release_bytes -= dirty_sectors <<
+				root->fs_info->sb->s_blocksize_bits;
+
 			if (copied > 0) {
 				spin_lock(&BTRFS_I(inode)->lock);
 				BTRFS_I(inode)->outstanding_extents++;
@@ -2479,7 +2477,7 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 	}
 
 	ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv, rsv,
-				      min_size);
+				      min_size, 0);
 	BUG_ON(ret);
 	trans->block_rsv = rsv;
 
@@ -2522,7 +2520,7 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 		}
 
 		ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv,
-					      rsv, min_size);
+					      rsv, min_size, 0);
 		BUG_ON(ret);	/* shouldn't happen */
 		trans->block_rsv = rsv;
 
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index df731c0ebec7..8078077d1090 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5263,7 +5263,7 @@ void btrfs_evict_inode(struct inode *inode)
 		if (steal_from_global) {
 			if (!btrfs_check_space_for_delayed_refs(trans, root))
 				ret = btrfs_block_rsv_migrate(global_rsv, rsv,
-							      min_size);
+							      min_size, 0);
 			else
 				ret = -ENOSPC;
 		}
@@ -9116,7 +9116,7 @@ static int btrfs_truncate(struct inode *inode)
 
 	/* Migrate the slack space for the truncate to our reserve */
 	ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv, rsv,
-				      min_size);
+				      min_size, 0);
 	BUG_ON(ret);
 
 	/*
@@ -9156,7 +9156,7 @@ static int btrfs_truncate(struct inode *inode)
 		}
 
 		ret = btrfs_block_rsv_migrate(&root->fs_info->trans_block_rsv,
-					      rsv, min_size);
+					      rsv, min_size, 0);
 		BUG_ON(ret);	/* shouldn't happen */
 		trans->block_rsv = rsv;
 	}
@@ -9177,7 +9177,6 @@ static int btrfs_truncate(struct inode *inode)
 		ret = btrfs_end_transaction(trans, root);
 		btrfs_btree_balance_dirty(root);
 	}
-
 out:
 	btrfs_free_block_rsv(root, rsv);
 
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 0477dca154ed..fc067b07e31f 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -2604,25 +2604,28 @@ static int reserve_metadata_space(struct btrfs_trans_handle *trans,
 
 	trans->block_rsv = rc->block_rsv;
 	rc->reserved_bytes += num_bytes;
+
+	/*
+	 * We are under a transaction here so we can only do limited flushing.
+	 * If we get an enospc just kick back -EAGAIN so we know to drop the
+	 * transaction and try to refill when we can flush all the things.
+	 */
 	ret = btrfs_block_rsv_refill(root, rc->block_rsv, num_bytes,
-				BTRFS_RESERVE_FLUSH_ALL);
+				BTRFS_RESERVE_FLUSH_LIMIT);
 	if (ret) {
-		if (ret == -EAGAIN) {
-			tmp = rc->extent_root->nodesize *
-				RELOCATION_RESERVED_NODES;
-			while (tmp <= rc->reserved_bytes)
-				tmp <<= 1;
-			/*
-			 * only one thread can access block_rsv at this point,
-			 * so we don't need hold lock to protect block_rsv.
-			 * we expand more reservation size here to allow enough
-			 * space for relocation and we will return earlier in
-			 * enospc case.
-			 */
-			rc->block_rsv->size = tmp + rc->extent_root->nodesize *
-					      RELOCATION_RESERVED_NODES;
-		}
-		return ret;
+		tmp = rc->extent_root->nodesize * RELOCATION_RESERVED_NODES;
+		while (tmp <= rc->reserved_bytes)
+			tmp <<= 1;
+		/*
+		 * only one thread can access block_rsv at this point,
+		 * so we don't need hold lock to protect block_rsv.
+		 * we expand more reservation size here to allow enough
+		 * space for relocation and we will return eailer in
+		 * enospc case.
+		 */
+		rc->block_rsv->size = tmp + rc->extent_root->nodesize *
+			RELOCATION_RESERVED_NODES;
+		return -EAGAIN;
 	}
 
 	return 0;
@@ -3871,6 +3874,7 @@ static noinline_for_stack
 int prepare_to_relocate(struct reloc_control *rc)
 {
 	struct btrfs_trans_handle *trans;
+	int ret;
 
 	rc->block_rsv = btrfs_alloc_block_rsv(rc->extent_root,
 					      BTRFS_BLOCK_RSV_TEMP);
@@ -3885,6 +3889,11 @@ int prepare_to_relocate(struct reloc_control *rc)
 	rc->reserved_bytes = 0;
 	rc->block_rsv->size = rc->extent_root->nodesize *
 			      RELOCATION_RESERVED_NODES;
+	ret = btrfs_block_rsv_refill(rc->extent_root,
+				     rc->block_rsv, rc->block_rsv->size,
+				     BTRFS_RESERVE_FLUSH_ALL);
+	if (ret)
+		return ret;
 
 	rc->create_reloc_tree = 1;
 	set_reloc_control(rc);
@@ -4643,7 +4652,7 @@ int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans,
 	if (rc->merge_reloc_tree) {
 		ret = btrfs_block_rsv_migrate(&pending->block_rsv,
 					      rc->block_rsv,
-					      rc->nodes_relocated);
+					      rc->nodes_relocated, 1);
 		if (ret)
 			return ret;
 	}
diff --git a/include/trace/events/btrfs.h b/include/trace/events/btrfs.h
index e90e82ad6875..5b81ef304388 100644
--- a/include/trace/events/btrfs.h
+++ b/include/trace/events/btrfs.h
@@ -440,6 +440,46 @@ TRACE_EVENT(btrfs_sync_fs,
 	TP_printk("wait = %d", __entry->wait)
 );
 
+TRACE_EVENT(btrfs_add_block_group,
+
+	TP_PROTO(struct btrfs_fs_info *fs_info,
+		 struct btrfs_block_group_cache *block_group, int create),
+
+	TP_ARGS(fs_info, block_group, create),
+
+	TP_STRUCT__entry(
+		__array(	u8,	fsid,	BTRFS_UUID_SIZE	)
+		__field(	u64,	offset			)
+		__field(	u64,	size			)
+		__field(	u64,	flags			)
+		__field(	u64,	bytes_used		)
+		__field(	u64,	bytes_super		)
+		__field(	int,	create			)
+	),
+
+	TP_fast_assign(
+		memcpy(__entry->fsid, fs_info->fsid, BTRFS_UUID_SIZE);
+		__entry->offset		= block_group->key.objectid;
+		__entry->size		= block_group->key.offset;
+		__entry->flags		= block_group->flags;
+		__entry->bytes_used	=
+			btrfs_block_group_used(&block_group->item);
+		__entry->bytes_super	= block_group->bytes_super;
+		__entry->create		= create;
+	),
+
+	TP_printk("%pU: block_group offset = %llu, size = %llu, "
+		  "flags = %llu(%s), bytes_used = %llu, bytes_super = %llu, "
+		  "create = %d", __entry->fsid,
+		  (unsigned long long)__entry->offset,
+		  (unsigned long long)__entry->size,
+		  (unsigned long long)__entry->flags,
+		  __print_flags((unsigned long)__entry->flags, "|",
+				BTRFS_GROUP_FLAGS),
+		  (unsigned long long)__entry->bytes_used,
+		  (unsigned long long)__entry->bytes_super, __entry->create)
+);
+
 #define show_ref_action(action)						\
 	__print_symbolic(action,					\
 		{ BTRFS_ADD_DELAYED_REF,    "ADD_DELAYED_REF" },	\
@@ -744,6 +784,88 @@ TRACE_EVENT(btrfs_space_reservation,
 		  __entry->bytes)
 );
 
+#define show_flush_action(action)						\
+	__print_symbolic(action,						\
+		{ BTRFS_RESERVE_NO_FLUSH,	"BTRFS_RESERVE_NO_FLUSH"},	\
+		{ BTRFS_RESERVE_FLUSH_LIMIT,	"BTRFS_RESERVE_FLUSH_LIMIT"},	\
+		{ BTRFS_RESERVE_FLUSH_ALL,	"BTRFS_RESERVE_FLUSH_ALL"})
+
+TRACE_EVENT(btrfs_trigger_flush,
+
+	TP_PROTO(struct btrfs_fs_info *fs_info, u64 flags, u64 bytes,
+		 int flush, char *reason),
+
+	TP_ARGS(fs_info, flags, bytes, flush, reason),
+
+	TP_STRUCT__entry(
+		__array(	u8,	fsid,	BTRFS_UUID_SIZE	)
+		__field(	u64,	flags			)
+		__field(	u64,	bytes			)
+		__field(	int,	flush			)
+		__string(	reason,	reason			)
+	),
+
+	TP_fast_assign(
+		memcpy(__entry->fsid, fs_info->fsid, BTRFS_UUID_SIZE);
+		__entry->flags	= flags;
+		__entry->bytes	= bytes;
+		__entry->flush	= flush;
+		__assign_str(reason, reason)
+	),
+
+	TP_printk("%pU: %s: flush = %d(%s), flags = %llu(%s), bytes = %llu",
+		  __entry->fsid, __get_str(reason), __entry->flush,
+		  show_flush_action(__entry->flush),
+		  (unsigned long long)__entry->flags,
+		  __print_flags((unsigned long)__entry->flags, "|",
+				BTRFS_GROUP_FLAGS),
+		  (unsigned long long)__entry->bytes)
+);
+
+#define show_flush_state(state)							\
+	__print_symbolic(state,							\
+		{ FLUSH_DELAYED_ITEMS_NR,	"FLUSH_DELAYED_ITEMS_NR"},	\
+		{ FLUSH_DELAYED_ITEMS,		"FLUSH_DELAYED_ITEMS"},		\
+		{ FLUSH_DELALLOC,		"FLUSH_DELALLOC"},		\
+		{ FLUSH_DELALLOC_WAIT,		"FLUSH_DELALLOC_WAIT"},		\
+		{ ALLOC_CHUNK,			"ALLOC_CHUNK"},			\
+		{ COMMIT_TRANS,			"COMMIT_TRANS"})
+
+TRACE_EVENT(btrfs_flush_space,
+
+	TP_PROTO(struct btrfs_fs_info *fs_info, u64 flags, u64 num_bytes,
+		 u64 orig_bytes, int state, int ret),
+
+	TP_ARGS(fs_info, flags, num_bytes, orig_bytes, state, ret),
+
+	TP_STRUCT__entry(
+		__array(	u8,	fsid,	BTRFS_UUID_SIZE	)
+		__field(	u64,	flags			)
+		__field(	u64,	num_bytes		)
+		__field(	u64,	orig_bytes		)
+		__field(	int,	state			)
+		__field(	int,	ret			)
+	),
+
+	TP_fast_assign(
+		memcpy(__entry->fsid, fs_info->fsid, BTRFS_UUID_SIZE);
+		__entry->flags		=	flags;
+		__entry->num_bytes	=	num_bytes;
+		__entry->orig_bytes	=	orig_bytes;
+		__entry->state		=	state;
+		__entry->ret		=	ret;
+	),
+
+	TP_printk("%pU: state = %d(%s), flags = %llu(%s), num_bytes = %llu, "
+		  "orig_bytes = %llu, ret = %d", __entry->fsid, __entry->state,
+		  show_flush_state(__entry->state),
+		  (unsigned long long)__entry->flags,
+		  __print_flags((unsigned long)__entry->flags, "|",
+				BTRFS_GROUP_FLAGS),
+		  (unsigned long long)__entry->num_bytes,
+		  (unsigned long long)__entry->orig_bytes, __entry->ret)
+);
+
 DECLARE_EVENT_CLASS(btrfs__reserved_extent,
 
 	TP_PROTO(struct btrfs_root *root, u64 start, u64 len),
@@ -751,18 +873,21 @@ DECLARE_EVENT_CLASS(btrfs__reserved_extent,
 	TP_ARGS(root, start, len),
 
 	TP_STRUCT__entry(
-		__field(	u64,  root_objectid		)
-		__field(	u64,  start			)
-		__field(	u64,  len			)
+		__array(	u8,	fsid,	BTRFS_UUID_SIZE	)
+		__field(	u64,	root_objectid		)
+		__field(	u64,	start			)
+		__field(	u64,	len			)
 	),
 
 	TP_fast_assign(
+		memcpy(__entry->fsid, root->fs_info->fsid, BTRFS_UUID_SIZE);
 		__entry->root_objectid	= root->root_key.objectid;
 		__entry->start		= start;
 		__entry->len		= len;
 	),
 
-	TP_printk("root = %llu(%s), start = %llu, len = %llu",
+	TP_printk("%pU: root = %llu(%s), start = %llu, len = %llu",
+		  __entry->fsid,
 		  show_root_type(__entry->root_objectid),
 		  (unsigned long long)__entry->start,
 		  (unsigned long long)__entry->len)
@@ -819,6 +944,7 @@ DECLARE_EVENT_CLASS(btrfs__reserve_extent,
 	TP_ARGS(root, block_group, start, len),
 
 	TP_STRUCT__entry(
+		__array(	u8,	fsid,	BTRFS_UUID_SIZE	)
 		__field(	u64,	root_objectid		)
 		__field(	u64,	bg_objectid		)
 		__field(	u64,	flags			)
@@ -827,6 +953,7 @@ DECLARE_EVENT_CLASS(btrfs__reserve_extent,
 	),
 
 	TP_fast_assign(
+		memcpy(__entry->fsid, root->fs_info->fsid, BTRFS_UUID_SIZE);
 		__entry->root_objectid	= root->root_key.objectid;
 		__entry->bg_objectid	= block_group->key.objectid;
 		__entry->flags		= block_group->flags;
@@ -834,8 +961,8 @@ DECLARE_EVENT_CLASS(btrfs__reserve_extent,
 		__entry->len		= len;
 	),
 
-	TP_printk("root = %Lu(%s), block_group = %Lu, flags = %Lu(%s), "
-		  "start = %Lu, len = %Lu",
+	TP_printk("%pU: root = %Lu(%s), block_group = %Lu, flags = %Lu(%s), "
+		  "start = %Lu, len = %Lu", __entry->fsid,
 		  show_root_type(__entry->root_objectid), __entry->bg_objectid,
 		  __entry->flags, __print_flags((unsigned long)__entry->flags,
 						"|", BTRFS_GROUP_FLAGS),