linux-kernel - Re: [PATCH] perf/aux: Properly launch pending disable flow

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: <d87ddbbb-862d-4cf7-b2c4-a5eb0d072a33@linaro.org>
Date: Tue, 24 Jun 2025 14:11:38 +0100
From: James Clark <james.clark@...aro.org>
To: Leo Yan <leo.yan@....com>
Cc: Peter Zijlstra <peterz@...radead.org>, Ingo Molnar <mingo@...hat.com>,
 Arnaldo Carvalho de Melo <acme@...nel.org>,
 Namhyung Kim <namhyung@...nel.org>, Mark Rutland <mark.rutland@....com>,
 Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
 Jiri Olsa <jolsa@...nel.org>, Ian Rogers <irogers@...gle.com>,
 Adrian Hunter <adrian.hunter@...el.com>,
 "Liang, Kan" <kan.liang@...ux.intel.com>, Marco Elver <elver@...gle.com>,
 Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
 linux-perf-users@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] perf/aux: Properly launch pending disable flow



On 22/05/2025 4:05 pm, Leo Yan wrote:
> If an AUX event overruns, the event core layer intends to disable the
> event by setting the 'pending_disable' flag. Unfortunately, the event
> is not actually disabled afterwards.
> 
> Since commit ca6c21327c6a ("perf: Fix missing SIGTRAPs"), the
> 'pending_disable' flag was changed to a boolean toggles. However, the
> AUX event code was not updated accordingly. The flag ends up holding a
> CPU number. If this number is zero, the flag is taken as false and the
> IRQ work is never triggered.
> 
> Later, with commit 2b84def990d3 ("perf: Split __perf_pending_irq() out
> of perf_pending_irq()"), a new IRQ work 'pending_disable_irq' was
> introduced to handle event disabling. The AUX event path was not updated
> to kick off the work queue.
> 
> To fix this issue, when an AUX ring buffer overrun is detected, call
> perf_event_disable_inatomic() to initiate the pending disable flow.
> 
> Fixes: ca6c21327c6a ("perf: Fix missing SIGTRAPs")
> Fixes: 2b84def990d3 ("perf: Split __perf_pending_irq() out of perf_pending_irq()")
> Signed-off-by: Leo Yan <leo.yan@....com>
> ---
>   kernel/events/ring_buffer.c | 4 ++--
>   1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c
> index d2aef87c7e9f..aa9a759e824f 100644
> --- a/kernel/events/ring_buffer.c
> +++ b/kernel/events/ring_buffer.c
> @@ -441,7 +441,7 @@ void *perf_aux_output_begin(struct perf_output_handle *handle,
>   		 * store that will be enabled on successful return
>   		 */
>   		if (!handle->size) { /* A, matches D */
> -			event->pending_disable = smp_processor_id();
> +			perf_event_disable_inatomic(handle->event);
>   			perf_output_wakeup(handle);
>   			WRITE_ONCE(rb->aux_nest, 0);
>   			goto err_put;
> @@ -526,7 +526,7 @@ void perf_aux_output_end(struct perf_output_handle *handle, unsigned long size)
>   
>   	if (wakeup) {
>   		if (handle->aux_flags & PERF_AUX_FLAG_TRUNCATED)
> -			handle->event->pending_disable = smp_processor_id();
> +			perf_event_disable_inatomic(handle->event);
>   		perf_output_wakeup(handle);
>   	}
>   

The types are now a bit misleading and pending_wakeup and 
pending_disable could be bool types. The other pending_*s do use their 
types properly though.

__perf_pending_disable() also still contains a big comment that 
describes use of CPU ID and -1 values.

Other than that it makes sense.

Reviewed-by: James Clark <james.clark@...aro.org>