Re: [PATCH v1 1/2] perf auxtrace: Add 'T' itrace option for timestamp trace

From: Adrian Hunter <adrian.hunter@intel.com>
To: Leo Yan <leo.yan@linaro.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>,
	James Clark <james.clark@arm.com>,
	Suzuki K Poulose <suzuki.poulose@arm.com>,
	Mike Leach <mike.leach@linaro.org>,
	John Garry <john.g.garry@oracle.com>,
	Will Deacon <will@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@redhat.com>,
	Mark Rutland <mark.rutland@arm.com>,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Jiri Olsa <jolsa@kernel.org>, Namhyung Kim <namhyung@kernel.org>,
	Ian Rogers <irogers@google.com>,
	coresight@lists.linaro.org, linux-arm-kernel@lists.infradead.org,
	linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v1 1/2] perf auxtrace: Add 'T' itrace option for timestamp trace
Date: Tue, 7 Nov 2023 12:16:25 +0200	[thread overview]
Message-ID: <6fc8bc4c-abc3-45d5-a8c8-045f91adfebf@intel.com> (raw)
In-Reply-To: <20231107094852.GA656098@leoy-yangtze.lan>

On 7/11/23 11:48, Leo Yan wrote:
> Hi Adrian,
> 
> On Tue, Nov 07, 2023 at 09:19:10AM +0200, Adrian Hunter wrote:
>> On 6/11/23 23:52, Arnaldo Carvalho de Melo wrote:
>>> Em Thu, Oct 19, 2023 at 01:47:15PM +0300, Adrian Hunter escreveu:
>>>> On 14/10/23 10:45, Leo Yan wrote:
>>>>> An AUX trace can contain timestamp, but in some situations, the hardware
>>>>> trace module (e.g. Arm CoreSight) cannot decide the traced timestamp is
>>>>> the same source with CPU's time, thus the decoder can not use the
>>>>> timestamp trace for samples.
>>>>>
>>>>> This patch introduces 'T' itrace option. If users know the platforms
>>>>
>>>> "If users know" <- how would users know?  Could the kernel
>>>> or tools also figure it out?
>>>
>>> Adrian, I'm trying to go all the outstanding patches, do you still have
>>> any issues with this series?
>>
>> No, although the question wasn't actually answered.  I presume users
>> just have to try the 'T' option and see if it helps.
> 
> Sometimes, users are software developers in SoC companies, they can
> know well for the hardware design but are confused why current
> implementation cannot use timestamp trace.  This is the main reason
> I sent this patch set.
> 
> An example hardware platform is DB410c [1], we know its CoreSight can
> support timestamp trace, but if without this adding option 'T', we
> have no chance to use it due to it its CPU arch is prior to Armv8.4.

perf config might be better than an itrace option, but you decide.

> 
> @Arnaldo, since James gave comments in his replying, I will respin new
> patch set and send out.  Thanks for popping up this patch set!
> 
> Leo
> 
> [1] https://developer.qualcomm.com/hardware/dragonboard-410c