1 files changed, 156 insertions, 94 deletions

diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index 5a97db9888..aca24186d6 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -935,7 +935,7 @@ int btrfs_cache_block_group(struct btrfs_block_group *cache, bool wait)
 	caching_ctl->block_group = cache;
 	refcount_set(&caching_ctl->count, 2);
 	atomic_set(&caching_ctl->progress, 0);
-	btrfs_init_work(&caching_ctl->work, caching_thread, NULL, NULL);
+	btrfs_init_work(&caching_ctl->work, caching_thread, NULL);
 
 	spin_lock(&cache->lock);
 	if (cache->cached != BTRFS_CACHE_NO) {
@@ -1286,7 +1286,7 @@ out:
 	/* Once for the lookup reference */
 	btrfs_put_block_group(block_group);
 	if (remove_rsv)
-		btrfs_delayed_refs_rsv_release(fs_info, 1);
+		btrfs_dec_delayed_refs_rsv_bg_updates(fs_info);
 	btrfs_free_path(path);
 	return ret;
 }
@@ -1467,6 +1467,7 @@ out:
  */
 void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 {
+	LIST_HEAD(retry_list);
 	struct btrfs_block_group *block_group;
 	struct btrfs_space_info *space_info;
 	struct btrfs_trans_handle *trans;
@@ -1488,6 +1489,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 
 	spin_lock(&fs_info->unused_bgs_lock);
 	while (!list_empty(&fs_info->unused_bgs)) {
+		u64 used;
 		int trimming;
 
 		block_group = list_first_entry(&fs_info->unused_bgs,
@@ -1523,9 +1525,9 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 			goto next;
 		}
 
+		spin_lock(&space_info->lock);
 		spin_lock(&block_group->lock);
-		if (block_group->reserved || block_group->pinned ||
-		    block_group->used || block_group->ro ||
+		if (btrfs_is_block_group_used(block_group) || block_group->ro ||
 		    list_is_singular(&block_group->list)) {
 			/*
 			 * We want to bail if we made new allocations or have
@@ -1535,10 +1537,49 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 			 */
 			trace_btrfs_skip_unused_block_group(block_group);
 			spin_unlock(&block_group->lock);
+			spin_unlock(&space_info->lock);
+			up_write(&space_info->groups_sem);
+			goto next;
+		}
+
+		/*
+		 * The block group may be unused but there may be space reserved
+		 * accounting with the existence of that block group, that is,
+		 * space_info->bytes_may_use was incremented by a task but no
+		 * space was yet allocated from the block group by the task.
+		 * That space may or may not be allocated, as we are generally
+		 * pessimistic about space reservation for metadata as well as
+		 * for data when using compression (as we reserve space based on
+		 * the worst case, when data can't be compressed, and before
+		 * actually attempting compression, before starting writeback).
+		 *
+		 * So check if the total space of the space_info minus the size
+		 * of this block group is less than the used space of the
+		 * space_info - if that's the case, then it means we have tasks
+		 * that might be relying on the block group in order to allocate
+		 * extents, and add back the block group to the unused list when
+		 * we finish, so that we retry later in case no tasks ended up
+		 * needing to allocate extents from the block group.
+		 */
+		used = btrfs_space_info_used(space_info, true);
+		if (space_info->total_bytes - block_group->length < used) {
+			/*
+			 * Add a reference for the list, compensate for the ref
+			 * drop under the "next" label for the
+			 * fs_info->unused_bgs list.
+			 */
+			btrfs_get_block_group(block_group);
+			list_add_tail(&block_group->bg_list, &retry_list);
+
+			trace_btrfs_skip_unused_block_group(block_group);
+			spin_unlock(&block_group->lock);
+			spin_unlock(&space_info->lock);
 			up_write(&space_info->groups_sem);
 			goto next;
 		}
+
 		spin_unlock(&block_group->lock);
+		spin_unlock(&space_info->lock);
 
 		/* We don't want to force the issue, only flip if it's ok. */
 		ret = inc_block_group_ro(block_group, 0);
@@ -1662,12 +1703,16 @@ next:
 		btrfs_put_block_group(block_group);
 		spin_lock(&fs_info->unused_bgs_lock);
 	}
+	list_splice_tail(&retry_list, &fs_info->unused_bgs);
 	spin_unlock(&fs_info->unused_bgs_lock);
 	mutex_unlock(&fs_info->reclaim_bgs_lock);
 	return;
 
 flip_async:
 	btrfs_end_transaction(trans);
+	spin_lock(&fs_info->unused_bgs_lock);
+	list_splice_tail(&retry_list, &fs_info->unused_bgs);
+	spin_unlock(&fs_info->unused_bgs_lock);
 	mutex_unlock(&fs_info->reclaim_bgs_lock);
 	btrfs_put_block_group(block_group);
 	btrfs_discard_punt_unused_bgs_list(fs_info);
@@ -2709,9 +2754,40 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans)
 
 		/* Already aborted the transaction if it failed. */
 next:
-		btrfs_delayed_refs_rsv_release(fs_info, 1);
+		btrfs_dec_delayed_refs_rsv_bg_inserts(fs_info);
 		list_del_init(&block_group->bg_list);
 		clear_bit(BLOCK_GROUP_FLAG_NEW, &block_group->runtime_flags);
+
+		/*
+		 * If the block group is still unused, add it to the list of
+		 * unused block groups. The block group may have been created in
+		 * order to satisfy a space reservation, in which case the
+		 * extent allocation only happens later. But often we don't
+		 * actually need to allocate space that we previously reserved,
+		 * so the block group may become unused for a long time. For
+		 * example for metadata we generally reserve space for a worst
+		 * possible scenario, but then don't end up allocating all that
+		 * space or none at all (due to no need to COW, extent buffers
+		 * were already COWed in the current transaction and still
+		 * unwritten, tree heights lower than the maximum possible
+		 * height, etc). For data we generally reserve the axact amount
+		 * of space we are going to allocate later, the exception is
+		 * when using compression, as we must reserve space based on the
+		 * uncompressed data size, because the compression is only done
+		 * when writeback triggered and we don't know how much space we
+		 * are actually going to need, so we reserve the uncompressed
+		 * size because the data may be uncompressible in the worst case.
+		 */
+		if (ret == 0) {
+			bool used;
+
+			spin_lock(&block_group->lock);
+			used = btrfs_is_block_group_used(block_group);
+			spin_unlock(&block_group->lock);
+
+			if (!used)
+				btrfs_mark_bg_unused(block_group);
+		}
 	}
 	btrfs_trans_release_chunk_metadata(trans);
 }
@@ -2819,8 +2895,7 @@ struct btrfs_block_group *btrfs_make_block_group(struct btrfs_trans_handle *tran
 #endif
 
 	list_add_tail(&cache->bg_list, &trans->new_bgs);
-	trans->delayed_ref_updates++;
-	btrfs_update_delayed_refs_rsv(trans);
+	btrfs_inc_delayed_refs_rsv_bg_inserts(fs_info);
 
 	set_avail_alloc_bits(fs_info, type);
 	return cache;
@@ -3051,7 +3126,6 @@ static int cache_save_setup(struct btrfs_block_group *block_group,
 			    struct btrfs_path *path)
 {
 	struct btrfs_fs_info *fs_info = block_group->fs_info;
-	struct btrfs_root *root = fs_info->tree_root;
 	struct inode *inode = NULL;
 	struct extent_changeset *data_reserved = NULL;
 	u64 alloc_hint = 0;
@@ -3103,7 +3177,7 @@ again:
 	 * time.
 	 */
 	BTRFS_I(inode)->generation = 0;
-	ret = btrfs_update_inode(trans, root, BTRFS_I(inode));
+	ret = btrfs_update_inode(trans, BTRFS_I(inode));
 	if (ret) {
 		/*
 		 * So theoretically we could recover from this, simply set the
@@ -3370,7 +3444,7 @@ again:
 		if (should_put)
 			btrfs_put_block_group(cache);
 		if (drop_reserve)
-			btrfs_delayed_refs_rsv_release(fs_info, 1);
+			btrfs_dec_delayed_refs_rsv_bg_updates(fs_info);
 		/*
 		 * Avoid blocking other tasks for too long. It might even save
 		 * us from writing caches for block groups that are going to be
@@ -3474,8 +3548,7 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans)
 		cache_save_setup(cache, trans, path);
 
 		if (!ret)
-			ret = btrfs_run_delayed_refs(trans,
-						     (unsigned long) -1);
+			ret = btrfs_run_delayed_refs(trans, U64_MAX);
 
 		if (!ret && cache->disk_cache_state == BTRFS_DC_SETUP) {
 			cache->io_ctl.inode = NULL;
@@ -3518,7 +3591,7 @@ int btrfs_write_dirty_block_groups(struct btrfs_trans_handle *trans)
 		/* If its not on the io list, we need to put the block group */
 		if (should_put)
 			btrfs_put_block_group(cache);
-		btrfs_delayed_refs_rsv_release(fs_info, 1);
+		btrfs_dec_delayed_refs_rsv_bg_updates(fs_info);
 		spin_lock(&cur_trans->dirty_bgs_lock);
 	}
 	spin_unlock(&cur_trans->dirty_bgs_lock);
@@ -3543,12 +3616,12 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 			     u64 bytenr, u64 num_bytes, bool alloc)
 {
 	struct btrfs_fs_info *info = trans->fs_info;
-	struct btrfs_block_group *cache = NULL;
-	u64 total = num_bytes;
+	struct btrfs_space_info *space_info;
+	struct btrfs_block_group *cache;
 	u64 old_val;
-	u64 byte_in_group;
+	bool reclaim = false;
+	bool bg_already_dirty = true;
 	int factor;
-	int ret = 0;
 
 	/* Block accounting for super block */
 	spin_lock(&info->delalloc_root_lock);
@@ -3560,97 +3633,86 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 	btrfs_set_super_bytes_used(info->super_copy, old_val);
 	spin_unlock(&info->delalloc_root_lock);
 
-	while (total) {
-		struct btrfs_space_info *space_info;
-		bool reclaim = false;
-
-		cache = btrfs_lookup_block_group(info, bytenr);
-		if (!cache) {
-			ret = -ENOENT;
-			break;
-		}
-		space_info = cache->space_info;
-		factor = btrfs_bg_type_to_factor(cache->flags);
+	cache = btrfs_lookup_block_group(info, bytenr);
+	if (!cache)
+		return -ENOENT;
 
-		/*
-		 * If this block group has free space cache written out, we
-		 * need to make sure to load it if we are removing space.  This
-		 * is because we need the unpinning stage to actually add the
-		 * space back to the block group, otherwise we will leak space.
-		 */
-		if (!alloc && !btrfs_block_group_done(cache))
-			btrfs_cache_block_group(cache, true);
+	/* An extent can not span multiple block groups. */
+	ASSERT(bytenr + num_bytes <= cache->start + cache->length);
 
-		byte_in_group = bytenr - cache->start;
-		WARN_ON(byte_in_group > cache->length);
+	space_info = cache->space_info;
+	factor = btrfs_bg_type_to_factor(cache->flags);
 
-		spin_lock(&space_info->lock);
-		spin_lock(&cache->lock);
+	/*
+	 * If this block group has free space cache written out, we need to make
+	 * sure to load it if we are removing space.  This is because we need
+	 * the unpinning stage to actually add the space back to the block group,
+	 * otherwise we will leak space.
+	 */
+	if (!alloc && !btrfs_block_group_done(cache))
+		btrfs_cache_block_group(cache, true);
 
-		if (btrfs_test_opt(info, SPACE_CACHE) &&
-		    cache->disk_cache_state < BTRFS_DC_CLEAR)
-			cache->disk_cache_state = BTRFS_DC_CLEAR;
+	spin_lock(&space_info->lock);
+	spin_lock(&cache->lock);
 
-		old_val = cache->used;
-		num_bytes = min(total, cache->length - byte_in_group);
-		if (alloc) {
-			old_val += num_bytes;
-			cache->used = old_val;
-			cache->reserved -= num_bytes;
-			space_info->bytes_reserved -= num_bytes;
-			space_info->bytes_used += num_bytes;
-			space_info->disk_used += num_bytes * factor;
-			spin_unlock(&cache->lock);
-			spin_unlock(&space_info->lock);
-		} else {
-			old_val -= num_bytes;
-			cache->used = old_val;
-			cache->pinned += num_bytes;
-			btrfs_space_info_update_bytes_pinned(info, space_info,
-							     num_bytes);
-			space_info->bytes_used -= num_bytes;
-			space_info->disk_used -= num_bytes * factor;
+	if (btrfs_test_opt(info, SPACE_CACHE) &&
+	    cache->disk_cache_state < BTRFS_DC_CLEAR)
+		cache->disk_cache_state = BTRFS_DC_CLEAR;
 
-			reclaim = should_reclaim_block_group(cache, num_bytes);
+	old_val = cache->used;
+	if (alloc) {
+		old_val += num_bytes;
+		cache->used = old_val;
+		cache->reserved -= num_bytes;
+		space_info->bytes_reserved -= num_bytes;
+		space_info->bytes_used += num_bytes;
+		space_info->disk_used += num_bytes * factor;
+		spin_unlock(&cache->lock);
+		spin_unlock(&space_info->lock);
+	} else {
+		old_val -= num_bytes;
+		cache->used = old_val;
+		cache->pinned += num_bytes;
+		btrfs_space_info_update_bytes_pinned(info, space_info, num_bytes);
+		space_info->bytes_used -= num_bytes;
+		space_info->disk_used -= num_bytes * factor;
 
-			spin_unlock(&cache->lock);
-			spin_unlock(&space_info->lock);
+		reclaim = should_reclaim_block_group(cache, num_bytes);
 
-			set_extent_bit(&trans->transaction->pinned_extents,
-				       bytenr, bytenr + num_bytes - 1,
-				       EXTENT_DIRTY, NULL);
-		}
+		spin_unlock(&cache->lock);
+		spin_unlock(&space_info->lock);
 
-		spin_lock(&trans->transaction->dirty_bgs_lock);
-		if (list_empty(&cache->dirty_list)) {
-			list_add_tail(&cache->dirty_list,
-				      &trans->transaction->dirty_bgs);
-			trans->delayed_ref_updates++;
-			btrfs_get_block_group(cache);
-		}
-		spin_unlock(&trans->transaction->dirty_bgs_lock);
+		set_extent_bit(&trans->transaction->pinned_extents, bytenr,
+			       bytenr + num_bytes - 1, EXTENT_DIRTY, NULL);
+	}
 
-		/*
-		 * No longer have used bytes in this block group, queue it for
-		 * deletion. We do this after adding the block group to the
-		 * dirty list to avoid races between cleaner kthread and space
-		 * cache writeout.
-		 */
-		if (!alloc && old_val == 0) {
-			if (!btrfs_test_opt(info, DISCARD_ASYNC))
-				btrfs_mark_bg_unused(cache);
-		} else if (!alloc && reclaim) {
-			btrfs_mark_bg_to_reclaim(cache);
-		}
+	spin_lock(&trans->transaction->dirty_bgs_lock);
+	if (list_empty(&cache->dirty_list)) {
+		list_add_tail(&cache->dirty_list, &trans->transaction->dirty_bgs);
+		bg_already_dirty = false;
+		btrfs_get_block_group(cache);
+	}
+	spin_unlock(&trans->transaction->dirty_bgs_lock);
 
-		btrfs_put_block_group(cache);
-		total -= num_bytes;
-		bytenr += num_bytes;
+	/*
+	 * No longer have used bytes in this block group, queue it for deletion.
+	 * We do this after adding the block group to the dirty list to avoid
+	 * races between cleaner kthread and space cache writeout.
+	 */
+	if (!alloc && old_val == 0) {
+		if (!btrfs_test_opt(info, DISCARD_ASYNC))
+			btrfs_mark_bg_unused(cache);
+	} else if (!alloc && reclaim) {
+		btrfs_mark_bg_to_reclaim(cache);
 	}
 
+	btrfs_put_block_group(cache);
+
 	/* Modified block groups are accounted for in the delayed_refs_rsv. */
-	btrfs_update_delayed_refs_rsv(trans);
-	return ret;
+	if (!bg_already_dirty)
+		btrfs_inc_delayed_refs_rsv_bg_updates(info);
+
+	return 0;
 }
 
 /*