bcachefs: Run btree updates after write out of write_point

In the write path, after the write to the block device(s) complete we
have to punt to process context to do the btree update.

Instead of using the work item embedded in op->cl, this patch switches
to a per write-point work item. This helps with two different issues:

 - lock contention: btree updates to the same writepoint will (usually)
   be updating the same alloc keys
 - context switch overhead: when we're bottlenecked on btree updates,
   having a thread (running out of a work item) checking the write point
   for completed ops is cheaper than queueing up a new work item and
   waking up a kworker.

In an arbitrary benchmark, 4k random writes with fio running inside a
VM, this patch resulted in a 10% improvement in total iops.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>

1 files changed, 4 insertions, 2 deletions

diff --git a/fs/bcachefs/io.h b/fs/bcachefs/io.h
index f8ce9543c9e3..b484d3387968 100644
--- a/fs/bcachefs/io.h
+++ b/fs/bcachefs/io.h
@@ -41,7 +41,7 @@ enum bch_write_flags {
 	__BCH_WRITE_CHECK_ENOSPC,
 	__BCH_WRITE_MOVE,
 	__BCH_WRITE_JOURNAL_SEQ_PTR,
-	__BCH_WRITE_SKIP_CLOSURE_PUT,
+	__BCH_WRITE_IN_WORKER,
 	__BCH_WRITE_DONE,
 };
 
@@ -59,7 +59,7 @@ enum bch_write_flags {
 
 /* Internal: */
 #define BCH_WRITE_JOURNAL_SEQ_PTR	(1U << __BCH_WRITE_JOURNAL_SEQ_PTR)
-#define BCH_WRITE_SKIP_CLOSURE_PUT	(1U << __BCH_WRITE_SKIP_CLOSURE_PUT)
+#define BCH_WRITE_IN_WORKER		(1U << __BCH_WRITE_IN_WORKER)
 #define BCH_WRITE_DONE			(1U << __BCH_WRITE_DONE)
 
 static inline u64 *op_journal_seq(struct bch_write_op *op)
@@ -115,6 +115,8 @@ static inline void bch2_write_op_init(struct bch_write_op *op, struct bch_fs *c,
 
 void bch2_write(struct closure *);
 
+void bch2_write_point_do_index_updates(struct work_struct *);
+
 static inline struct bch_write_bio *wbio_init(struct bio *bio)
 {
 	struct bch_write_bio *wbio = to_wbio(bio);