* [PATCH] sched/fair: document the slow path and fast path in select_task_rq_fair
@ 2021-10-15 4:14 Barry Song
2021-10-15 13:26 ` Steven Rostedt
0 siblings, 1 reply; 2+ messages in thread
From: Barry Song @ 2021-10-15 4:14 UTC (permalink / raw)
To: mingo, peterz, juri.lelli, vincent.guittot, dietmar.eggemann,
rostedt, bsegall, mgorman, bristot
Cc: linux-kernel, linuxarm, yangyicong, Barry Song
From: Barry Song <song.bao.hua@hisilicon.com>
All people I know including myself took a long time to understand typical
wake_up will always go to fast path and never go to slow path except
FORK and EXEC.
Vincent reminded me once in a linaro meeting and it made me understand
slow path won't happen for WF_TTWU. But my other friends repeatedly
wasted a lot of time on testing this path like me before I reminded
them.
So obviously the code needs some document.
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
---
kernel/sched/fair.c | 11 ++++++++++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index f6a05d9b5443..ccf8f73dd4d1 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6951,6 +6951,11 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
break;
}
+ /*
+ * This is usually true only for WF_EXEC and WF_FORK, for WF_TTWU
+ * it is almost always false as sched_domain hasn't SD_BALANCE_WAKE
+ * in default
+ */
if (tmp->flags & sd_flag)
sd = tmp;
else if (!want_affine)
@@ -6958,7 +6963,11 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
}
if (unlikely(sd)) {
- /* Slow path */
+ /*
+ * Slow path, usually only for WF_EXEC and WF_FORK; WF_TTWU almost
+ * always goes to fast path as sched_domain hasn't SD_BALANCE_WAKE
+ * in default
+ */
new_cpu = find_idlest_cpu(sd, p, cpu, prev_cpu, sd_flag);
} else if (wake_flags & WF_TTWU) { /* XXX always ? */
/* Fast path */
--
2.25.1
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] sched/fair: document the slow path and fast path in select_task_rq_fair
2021-10-15 4:14 [PATCH] sched/fair: document the slow path and fast path in select_task_rq_fair Barry Song
@ 2021-10-15 13:26 ` Steven Rostedt
0 siblings, 0 replies; 2+ messages in thread
From: Steven Rostedt @ 2021-10-15 13:26 UTC (permalink / raw)
To: Barry Song
Cc: mingo, peterz, juri.lelli, vincent.guittot, dietmar.eggemann,
bsegall, mgorman, bristot, linux-kernel, linuxarm, yangyicong,
Barry Song
On Fri, 15 Oct 2021 12:14:12 +0800
Barry Song <21cnbao@gmail.com> wrote:
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -6951,6 +6951,11 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
> break;
> }
>
> + /*
> + * This is usually true only for WF_EXEC and WF_FORK, for WF_TTWU
> + * it is almost always false as sched_domain hasn't SD_BALANCE_WAKE
> + * in default
Comments are fine, but the above needs some grammar work.
/*
* Usually only true for WF_EXEC and WF_FORK, as
* sched_domains usually do not have SD_BALANCE_WAKE set.
*/
?
> + */
> if (tmp->flags & sd_flag)
> sd = tmp;
> else if (!want_affine)
> @@ -6958,7 +6963,11 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
> }
>
> if (unlikely(sd)) {
> - /* Slow path */
> + /*
> + * Slow path, usually only for WF_EXEC and WF_FORK; WF_TTWU almost
> + * always goes to fast path as sched_domain hasn't SD_BALANCE_WAKE
> + * in default
> + */
As the only way to have sd set, is from the above if statement, I believe
this comment is redundant.
-- Steve
> new_cpu = find_idlest_cpu(sd, p, cpu, prev_cpu, sd_flag);
> } else if (wake_flags & WF_TTWU) { /* XXX always ? */
> /* Fast path */
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2021-10-15 13:26 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-10-15 4:14 [PATCH] sched/fair: document the slow path and fast path in select_task_rq_fair Barry Song
2021-10-15 13:26 ` Steven Rostedt
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).