posix-timers: Use atomic64_try_cmpxchg() in __update_gt_cputime()

Use atomic64_try_cmpxchg() instead of atomic64_cmpxchg() in
__update_gt_cputime(). The 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, atomic64_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>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20230116165337.5810-1-ubizjak@gmail.com
This commit is contained in:
Uros Bizjak 2023-01-16 17:53:37 +01:00 committed by Thomas Gleixner
parent cbdb1f163a
commit 915d4ad383
1 changed files with 6 additions and 7 deletions

View File

@ -243,13 +243,12 @@ static void proc_sample_cputime_atomic(struct task_cputime_atomic *at,
*/
static inline void __update_gt_cputime(atomic64_t *cputime, u64 sum_cputime)
{
u64 curr_cputime;
retry:
curr_cputime = atomic64_read(cputime);
if (sum_cputime > curr_cputime) {
if (atomic64_cmpxchg(cputime, curr_cputime, sum_cputime) != curr_cputime)
goto retry;
}
u64 curr_cputime = atomic64_read(cputime);
do {
if (sum_cputime <= curr_cputime)
return;
} while (!atomic64_try_cmpxchg(cputime, &curr_cputime, sum_cputime));
}
static void update_gt_cputime(struct task_cputime_atomic *cputime_atomic,