linux-kernel - Re: [PATCH] perf/core: Fix endless multiplex timer

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <5c5b7fd1-4b85-8cce-b263-4a421b00f6df@linux.intel.com>
Date:   Thu, 5 Mar 2020 12:56:49 -0500
From:   "Liang, Kan" <kan.liang@...ux.intel.com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     mingo@...hat.com, linux-kernel@...r.kernel.org, irogers@...gle.com,
        eranian@...gle.com, ak@...ux.intel.com
Subject: Re: [PATCH] perf/core: Fix endless multiplex timer



On 3/5/2020 7:38 AM, Peter Zijlstra wrote:
> On Wed, Mar 04, 2020 at 09:20:42AM -0500, Liang, Kan wrote:
>>
>> NMI watchdog is pinned event.
>> ctx_event_to_rotate() will only pick an event from the flexible_groups.
>> So the cpu_ctx_sched_out() in perf_rotate_context() will never be called.
> 
> Surely that's fixable; same principle.
> 

Yes, after applying this patch, the issue has gone.

Tested-by: Kan Liang <kan.liang@...ux.intel.com>

Thanks,
Kan
> ---
> diff --git a/kernel/events/core.c b/kernel/events/core.c
> index 3f1f77de7247..595fb3decd43 100644
> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -2182,6 +2182,7 @@ __perf_remove_from_context(struct perf_event *event,
>   
>   	if (!ctx->nr_events && ctx->is_active) {
>   		ctx->is_active = 0;
> +		ctx->rotate_necessary = 0;
>   		if (ctx->task) {
>   			WARN_ON_ONCE(cpuctx->task_ctx != ctx);
>   			cpuctx->task_ctx = NULL;
> @@ -3077,12 +3078,6 @@ static void ctx_sched_out(struct perf_event_context *ctx,
>   	if (!ctx->nr_active || !(is_active & EVENT_ALL))
>   		return;
>   
> -	/*
> -	 * If we had been multiplexing, no rotations are necessary, now no events
> -	 * are active.
> -	 */
> -	ctx->rotate_necessary = 0;
> -
>   	perf_pmu_disable(ctx->pmu);
>   	if (is_active & EVENT_PINNED) {
>   		list_for_each_entry_safe(event, tmp, &ctx->pinned_active, active_list)
> @@ -3092,6 +3087,13 @@ static void ctx_sched_out(struct perf_event_context *ctx,
>   	if (is_active & EVENT_FLEXIBLE) {
>   		list_for_each_entry_safe(event, tmp, &ctx->flexible_active, active_list)
>   			group_sched_out(event, cpuctx, ctx);
> +
> +		/*
> +		 * Since we cleared EVENT_FLEXIBLE, also clear
> +		 * rotate_necessary, is will be reset by
> +		 * ctx_flexible_sched_in() when needed.
> +		 */
> +		ctx->rotate_necessary = 0;
>   	}
>   	perf_pmu_enable(ctx->pmu);
>   }
> @@ -3841,6 +3843,12 @@ ctx_event_to_rotate(struct perf_event_context *ctx)
>   				      typeof(*event), group_node);
>   	}
>   
> +	/*
> +	 * Unconditionally clear rotate_necessary; if ctx_flexible_sched_in()
> +	 * finds there are unschedulable events, it will set it again.
> +	 */
> +	ctx->rotate_necessary = 0;
> +
>   	return event;
>   }
>   
>