linux-kernel - Re: [PATCH 2/2] perf intel-pt: Use aux

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: <e8097eb7-66cb-432a-a799-2514b28ae48e@intel.com>
Date:   Wed, 14 Apr 2021 09:00:59 +0300
From:   Adrian Hunter <adrian.hunter@...el.com>
To:     Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        Peter Zijlstra <a.p.zijlstra@...llo.nl>,
        Arnaldo Carvalho de Melo <acme@...hat.com>
Cc:     Ingo Molnar <mingo@...hat.com>, linux-kernel@...r.kernel.org,
        Jiri Olsa <jolsa@...nel.org>,
        Mathieu Poirier <mathieu.poirier@...aro.org>
Subject: Re: [PATCH 2/2] perf intel-pt: Use aux_watermark

On 8/04/21 6:31 pm, Alexander Shishkin wrote:
> Turns out, the default setting of attr.aux_watermark to half of the total
> buffer size is not very useful, especially with smaller buffers. The
> problem is that, after half of the buffer is filled up, the kernel updates
> ->aux_head and sets up the next "transaction", while observing that
> ->aux_tail is still zero (as userspace haven't had the chance to update
> it), meaning that the trace will have to stop at the end of this second
> "transaction". This means, for example, that the second PERF_RECORD_AUX in
> every trace comes with TRUNCATED flag set.
> 
> Setting attr.aux_watermark to quarter of the buffer gives enough space for
> the ->aux_tail update to be observed and prevents the data loss.
> 
> The obligatory before/after showcase:
> 
>> # perf_before record -e intel_pt//u -m,8 uname
>> Linux
>> [ perf record: Woken up 6 times to write data ]
>> Warning:
>> AUX data lost 4 times out of 10!
>>
>> [ perf record: Captured and wrote 0.099 MB perf.data ]
>> # perf record -e intel_pt//u -m,8 uname
>> Linux
>> [ perf record: Woken up 4 times to write data ]
>> [ perf record: Captured and wrote 0.039 MB perf.data ]
> 
> The effect is still visible with large workloads and large buffers,
> although less pronounced.
> 
> Signed-off-by: Alexander Shishkin <alexander.shishkin@...ux.intel.com>
> ---
>  tools/perf/arch/x86/util/intel-pt.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/tools/perf/arch/x86/util/intel-pt.c b/tools/perf/arch/x86/util/intel-pt.c
> index a6420c647959..d00707faf547 100644
> --- a/tools/perf/arch/x86/util/intel-pt.c
> +++ b/tools/perf/arch/x86/util/intel-pt.c
> @@ -776,6 +776,10 @@ static int intel_pt_recording_options(struct auxtrace_record *itr,
>  		}
>  	}
>  
> +	if (opts->full_auxtrace)
> +		intel_pt_evsel->core.attr.aux_watermark =
> +		       opts->auxtrace_mmap_pages / 4 * page_size;
> +

I would be explicit about the mode and put "/ 4" at the end
for the case auxtrace_mmap_pages is not a multiple of 4 (e.g. 2).
i.e.

	if (!opts->auxtrace_snapshot_mode && !opts->auxtrace_sample_mode) {
		u32 aux_watermark = opts->auxtrace_mmap_pages * page_size / 4;

		intel_pt_evsel->core.attr.aux_watermark = aux_watermark;
	}


>  	intel_pt_parse_terms(intel_pt_pmu->name, &intel_pt_pmu->format,
>  			     "tsc", &tsc_bit);
>  
>