Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Alexander Shishkin <alexander.shishkin@linux.intel.com>
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     Arnaldo Carvalho de Melo <acme@redhat.com>,
        Ingo Molnar <mingo@redhat.com>, linux-kernel@vger.kernel.org,
        jolsa@redhat.com, adrian.hunter@intel.com,
        mathieu.poirier@linaro.org, mark.rutland@arm.com,
        alexander.shishkin@linux.intel.com
Subject: Re: [PATCH v3 1/3] perf: Allow using AUX data in perf samples
In-Reply-To: <20191028162712.GH4097@hirez.programming.kicks-ass.net>
References: <20191025140835.53665-1-alexander.shishkin@linux.intel.com> <20191025140835.53665-2-alexander.shishkin@linux.intel.com> <20191028162712.GH4097@hirez.programming.kicks-ass.net>
Date:   Mon, 28 Oct 2019 19:08:18 +0200
Message-ID: <87tv7sg5ml.fsf@ashishki-desk.ger.corp.intel.com>
MIME-Version: 1.0
Content-Type: text/plain
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Peter Zijlstra <peterz@infradead.org> writes:

> I have the below delta on top of this patch.
>
> And while I get why we need recursion protection for pmu::snapshot_aux,
> I'm a little puzzled on why it is over the padding, that is, why isn't
> the whole of aux_in_sampling inside (the newly minted)
> perf_pmu_snapshot_aux() ?

No reason. Too long staring at that code by myself.

> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -6237,7 +6237,7 @@ perf_output_sample_ustack(struct perf_ou
>  	}
>  }
>  
> -static unsigned long perf_aux_sample_size(struct perf_event *event,
> +static unsigned long perf_prepare_sample_aux(struct perf_event *event,
>  					  struct perf_sample_data *data,
>  					  size_t size)
>  {
> @@ -6275,9 +6275,9 @@ static unsigned long perf_aux_sample_siz
>  	return data->aux_size;
>  }
>  
> -long perf_pmu_aux_sample_output(struct perf_event *event,
> -				struct perf_output_handle *handle,
> -				unsigned long size)
> +long perf_pmu_snapshot_aux(struct perf_event *event,
> +			   struct perf_output_handle *handle,
> +			   unsigned long size)

That makes more sense indeed.

>  {
>  	unsigned long flags;
>  	long ret;
> @@ -6318,11 +6318,12 @@ static void perf_aux_sample_output(struc
>  
>  	/*
>  	 * Guard against NMI hits inside the critical section;
> -	 * see also perf_aux_sample_size().
> +	 * see also perf_prepare_sample_aux().
>  	 */
>  	WRITE_ONCE(rb->aux_in_sampling, 1);
> +	barrier();

Isn't WRITE_ONCE() barrier enough on its own? My thinking was that we
only need a compiler barrier here, hence the WRITE_ONCE.

Thanks,
--
Alex