bcachefs: Use try_cmpxchg() family of functions instead of cmpxchg()

Use try_cmpxchg() family of functions instead of
cmpxchg (*ptr, old, new) == old. x86 CMPXCHG instruction returns
success in ZF flag, so this change saves a compare after cmpxchg
(and related move instruction in front of cmpxchg).

Also, try_cmpxchg() implicitly assigns old *ptr value to "old" when
cmpxchg fails. There is no need to re-read the value in the loop.

No functional change intended.

Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Cc: Kent Overstreet <kent.overstreet@linux.dev>
Cc: Brian Foster <bfoster@redhat.com>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>

1 files changed, 3 insertions, 4 deletions

diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c
index 05e0cbef420b..c6197e6aa0b8 100644
--- a/fs/bcachefs/io_write.c
+++ b/fs/bcachefs/io_write.c
@@ -69,11 +69,10 @@ void bch2_latency_acct(struct bch_dev *ca, u64 submit_time, int rw)
 	u64 io_latency = time_after64(now, submit_time)
 		? now - submit_time
 		: 0;
-	u64 old, new, v = atomic64_read(latency);
+	u64 old, new;
 
+	old = atomic64_read(latency);
 	do {
-		old = v;
-
 		/*
 		 * If the io latency was reasonably close to the current
 		 * latency, skip doing the update and atomic operation - most of
@@ -84,7 +83,7 @@ void bch2_latency_acct(struct bch_dev *ca, u64 submit_time, int rw)
 			break;
 
 		new = ewma_add(old, io_latency, 5);
-	} while ((v = atomic64_cmpxchg(latency, old, new)) != old);
+	} while (!atomic64_try_cmpxchg(latency, &old, new));
 
 	bch2_congested_acct(ca, io_latency, now, rw);