btrfs: add the beginning of async discard, discard workqueue

When discard is enabled, everytime a pinned extent is released back to
the block_group's free space cache, a discard is issued for the extent.
This is an overeager approach when it comes to discarding and helping
the SSD maintain enough free space to prevent severe garbage collection
situations.

This adds the beginning of async discard. Instead of issuing a discard
prior to returning it to the free space, it is just marked as untrimmed.
The block_group is then added to a LRU which then feeds into a workqueue
to issue discards at a much slower rate. Full discarding of unused block
groups is still done and will be addressed in a future patch of the
series.

For now, we don't persist the discard state of extents and bitmaps.
Therefore, our failure recovery mode will be to consider extents
untrimmed. This lets us handle failure and unmounting as one in the
same.

On a number of Facebook webservers, I collected data every minute
accounting the time we spent in btrfs_finish_extent_commit() (col. 1)
and in btrfs_commit_transaction() (col. 2). btrfs_finish_extent_commit()
is where we discard extents synchronously before returning them to the
free space cache.

discard=sync:
                 p99 total per minute       p99 total per minute
      Drive   |   extent_commit() (ms)  |    commit_trans() (ms)
    ---------------------------------------------------------------
     Drive A  |           434           |          1170
     Drive B  |           880           |          2330
     Drive C  |          2943           |          3920
     Drive D  |          4763           |          5701

discard=async:
                 p99 total per minute       p99 total per minute
      Drive   |   extent_commit() (ms)  |    commit_trans() (ms)
    --------------------------------------------------------------
     Drive A  |           134           |           956
     Drive B  |            64           |          1972
     Drive C  |            59           |          1032
     Drive D  |            62           |          1200

While it's not great that the stats are cumulative over 1m, all of these
servers are running the same workload and and the delta between the two
are substantial. We are spending significantly less time in
btrfs_finish_extent_commit() which is responsible for discarding.

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, 33 insertions, 2 deletions

diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 08ac6a7a67f0..a906315efd19 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -46,6 +46,7 @@
 #include "sysfs.h"
 #include "tests/btrfs-tests.h"
 #include "block-group.h"
+#include "discard.h"
 
 #include "qgroup.h"
 #define CREATE_TRACE_POINTS
@@ -146,6 +147,8 @@ void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function
 	if (sb_rdonly(sb))
 		return;
 
+	btrfs_discard_stop(fs_info);
+
 	/* btrfs handle error by forcing the filesystem readonly */
 	sb->s_flags |= SB_RDONLY;
 	btrfs_info(fs_info, "forced readonly");
@@ -313,6 +316,7 @@ enum {
 	Opt_datasum, Opt_nodatasum,
 	Opt_defrag, Opt_nodefrag,
 	Opt_discard, Opt_nodiscard,
+	Opt_discard_mode,
 	Opt_nologreplay,
 	Opt_norecovery,
 	Opt_ratio,
@@ -375,6 +379,7 @@ static const match_table_t tokens = {
 	{Opt_defrag, "autodefrag"},
 	{Opt_nodefrag, "noautodefrag"},
 	{Opt_discard, "discard"},
+	{Opt_discard_mode, "discard=%s"},
 	{Opt_nodiscard, "nodiscard"},
 	{Opt_nologreplay, "nologreplay"},
 	{Opt_norecovery, "norecovery"},
@@ -695,12 +700,26 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options,
 				   info->metadata_ratio);
 			break;
 		case Opt_discard:
-			btrfs_set_and_info(info, DISCARD_SYNC,
-					   "turning on sync discard");
+		case Opt_discard_mode:
+			if (token == Opt_discard ||
+			    strcmp(args[0].from, "sync") == 0) {
+				btrfs_clear_opt(info->mount_opt, DISCARD_ASYNC);
+				btrfs_set_and_info(info, DISCARD_SYNC,
+						   "turning on sync discard");
+			} else if (strcmp(args[0].from, "async") == 0) {
+				btrfs_clear_opt(info->mount_opt, DISCARD_SYNC);
+				btrfs_set_and_info(info, DISCARD_ASYNC,
+						   "turning on async discard");
+			} else {
+				ret = -EINVAL;
+				goto out;
+			}
 			break;
 		case Opt_nodiscard:
 			btrfs_clear_and_info(info, DISCARD_SYNC,
 					     "turning off discard");
+			btrfs_clear_and_info(info, DISCARD_ASYNC,
+					     "turning off async discard");
 			break;
 		case Opt_space_cache:
 		case Opt_space_cache_version:
@@ -1324,6 +1343,8 @@ static int btrfs_show_options(struct seq_file *seq, struct dentry *dentry)
 		seq_puts(seq, ",flushoncommit");
 	if (btrfs_test_opt(info, DISCARD_SYNC))
 		seq_puts(seq, ",discard");
+	if (btrfs_test_opt(info, DISCARD_ASYNC))
+		seq_puts(seq, ",discard=async");
 	if (!(info->sb->s_flags & SB_POSIXACL))
 		seq_puts(seq, ",noacl");
 	if (btrfs_test_opt(info, SPACE_CACHE))
@@ -1713,6 +1734,14 @@ static inline void btrfs_remount_cleanup(struct btrfs_fs_info *fs_info,
 		btrfs_cleanup_defrag_inodes(fs_info);
 	}
 
+	/* If we toggled discard async */
+	if (!btrfs_raw_test_opt(old_opts, DISCARD_ASYNC) &&
+	    btrfs_test_opt(fs_info, DISCARD_ASYNC))
+		btrfs_discard_resume(fs_info);
+	else if (btrfs_raw_test_opt(old_opts, DISCARD_ASYNC) &&
+		 !btrfs_test_opt(fs_info, DISCARD_ASYNC))
+		btrfs_discard_cleanup(fs_info);
+
 	clear_bit(BTRFS_FS_STATE_REMOUNTING, &fs_info->fs_state);
 }
 
@@ -1760,6 +1789,8 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
 		 */
 		cancel_work_sync(&fs_info->async_reclaim_work);
 
+		btrfs_discard_cleanup(fs_info);
+
 		/* wait for the uuid_scan task to finish */
 		down(&fs_info->uuid_tree_rescan_sem);
 		/* avoid complains from lockdep et al. */