btrfs: handle empty block_group removal for async discard

block_group removal is a little tricky. It can race with the extent
allocator, the cleaner thread, and balancing. The current path is for a
block_group to be added to the unused_bgs list. Then, when the cleaner
thread comes around, it starts a transaction and then proceeds with
removing the block_group. Extents that are pinned are subsequently
removed from the pinned trees and then eventually a discard is issued
for the entire block_group.

Async discard introduces another player into the game, the discard
workqueue. While it has none of the racing issues, the new problem is
ensuring we don't leave free space untrimmed prior to forgetting the
block_group.  This is handled by placing fully free block_groups on a
separate discard queue. This is necessary to maintain discarding order
as in the future we will slowly trim even fully free block_groups. The
ordering helps us make progress on the same block_group rather than say
the last fully freed block_group or needing to search through the fully
freed block groups at the beginning of a list and insert after.

The new order of events is a fully freed block group gets placed on the
unused discard queue first. Once it's processed, it will be placed on
the unusued_bgs list and then the original sequence of events will
happen, just without the final whole block_group discard.

The mount flags can change when processing unused_bgs, so when flipping
from DISCARD to DISCARD_ASYNC, the unused_bgs must be punted to the
discard_list to be trimmed. If we flip off DISCARD_ASYNC, we punt
free block groups on the discard_list to the unused_bg queue which will
do the final discard for us.

Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Dennis Zhou <dennis@kernel.org>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>

1 files changed, 44 insertions, 3 deletions

diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index 6ba15c45e779..ba71a84fd1ee 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -1246,6 +1246,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 	struct btrfs_block_group *block_group;
 	struct btrfs_space_info *space_info;
 	struct btrfs_trans_handle *trans;
+	const bool async_trim_enabled = btrfs_test_opt(fs_info, DISCARD_ASYNC);
 	int ret = 0;
 
 	if (!test_bit(BTRFS_FS_OPEN, &fs_info->flags))
@@ -1275,6 +1276,22 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 
 		/* Don't want to race with allocators so take the groups_sem */
 		down_write(&space_info->groups_sem);
+
+		/*
+		 * Async discard moves the final block group discard to be prior
+		 * to the unused_bgs code path.  Therefore, if it's not fully
+		 * trimmed, punt it back to the async discard lists.
+		 */
+		if (btrfs_test_opt(fs_info, DISCARD_ASYNC) &&
+		    !btrfs_is_free_space_trimmed(block_group)) {
+			trace_btrfs_skip_unused_block_group(block_group);
+			up_write(&space_info->groups_sem);
+			/* Requeue if we failed because of async discard */
+			btrfs_discard_queue_work(&fs_info->discard_ctl,
+						 block_group);
+			goto next;
+		}
+
 		spin_lock(&block_group->lock);
 		if (block_group->reserved || block_group->pinned ||
 		    block_group->used || block_group->ro ||
@@ -1378,6 +1395,16 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info)
 		spin_unlock(&block_group->lock);
 		spin_unlock(&space_info->lock);
 
+		/*
+		 * The normal path here is an unused block group is passed here,
+		 * then trimming is handled in the transaction commit path.
+		 * Async discard interposes before this to do the trimming
+		 * before coming down the unused block group path as trimming
+		 * will no longer be done later in the transaction commit path.
+		 */
+		if (!async_trim_enabled && btrfs_test_opt(fs_info, DISCARD_ASYNC))
+			goto flip_async;
+
 		/* DISCARD can flip during remount */
 		trimming = btrfs_test_opt(fs_info, DISCARD_SYNC);
 
@@ -1422,6 +1449,13 @@ next:
 		spin_lock(&fs_info->unused_bgs_lock);
 	}
 	spin_unlock(&fs_info->unused_bgs_lock);
+	return;
+
+flip_async:
+	btrfs_end_transaction(trans);
+	mutex_unlock(&fs_info->delete_unused_bgs_mutex);
+	btrfs_put_block_group(block_group);
+	btrfs_discard_punt_unused_bgs_list(fs_info);
 }
 
 void btrfs_mark_bg_unused(struct btrfs_block_group *bg)
@@ -1626,6 +1660,8 @@ static struct btrfs_block_group *btrfs_create_block_group_cache(
 	cache->full_stripe_len = btrfs_full_stripe_len(fs_info, start);
 	set_free_space_tree_thresholds(cache);
 
+	cache->discard_index = BTRFS_DISCARD_INDEX_UNUSED;
+
 	atomic_set(&cache->count, 1);
 	spin_lock_init(&cache->lock);
 	init_rwsem(&cache->data_rwsem);
@@ -1792,7 +1828,10 @@ static int read_one_block_group(struct btrfs_fs_info *info,
 		inc_block_group_ro(cache, 1);
 	} else if (cache->used == 0) {
 		ASSERT(list_empty(&cache->bg_list));
-		btrfs_mark_bg_unused(cache);
+		if (btrfs_test_opt(info, DISCARD_ASYNC))
+			btrfs_discard_queue_work(&info->discard_ctl, cache);
+		else
+			btrfs_mark_bg_unused(cache);
 	}
 	return 0;
 error:
@@ -2755,8 +2794,10 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans,
 		 * dirty list to avoid races between cleaner kthread and space
 		 * cache writeout.
 		 */
-		if (!alloc && old_val == 0)
-			btrfs_mark_bg_unused(cache);
+		if (!alloc && old_val == 0) {
+			if (!btrfs_test_opt(info, DISCARD_ASYNC))
+				btrfs_mark_bg_unused(cache);
+		}
 
 		btrfs_put_block_group(cache);
 		total -= num_bytes;