Re: [PATCH] perf: Explain perf_sched_mutex

[Alexander Shishkin]

Alexander Shishkin <alexander.shishkin@xxxxxxxxxxxxxxx> writes:

> To clarify why atomic_inc_return(&perf_sched_events) is not sufficient and
> a mutex is needed to order static branch enabling vs the atomic counter
> increment, this adds a comment with an short explanation.

"with a short explanation", sigh

>
> Signed-off-by: Alexander Shishkin <alexander.shishkin@xxxxxxxxxxxxxxx>
> ---
>  kernel/events/core.c | 5 +++++
>  1 file changed, 5 insertions(+)
>
> diff --git a/kernel/events/core.c b/kernel/events/core.c
> index efe09b8c38..2c8719b635 100644
> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -9569,6 +9569,11 @@ static void account_event(struct perf_event *event)
>  		inc = true;
>  
>  	if (inc) {
> +		/*
> +		 * We need the mutex here because static_branch_enable()
> +		 * must complete *before* the perf_sched_count increment
> +		 * becomes visible.
> +		 */
>  		if (atomic_inc_not_zero(&perf_sched_count))
>  			goto enabled;
>  
> -- 
> 2.14.1