Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
CC:     <yangyicong@hisilicon.com>, <alexander.shishkin@linux.intel.com>,
        <helgaas@kernel.org>, <linux-pci@vger.kernel.org>,
        <prime.zeng@huawei.com>
Subject: Re: [PATCH v4 4/5] hwtracing: hisi_ptt: Advertise
 PERF_PMU_CAP_NO_EXCLUDE for PTT PMU
To:     hejunhao <hejunhao3@huawei.com>, <mathieu.poirier@linaro.org>,
        <suzuki.poulose@arm.com>, <jonathan.cameron@huawei.com>,
        <corbet@lwn.net>, <linux-kernel@vger.kernel.org>,
        <linux-doc@vger.kernel.org>
References: <20230606142244.10939-1-yangyicong@huawei.com>
 <20230606142244.10939-5-yangyicong@huawei.com>
 <01c39cb9-e56a-862b-2d99-024b7ba5a769@huawei.com>
From:   Yicong Yang <yangyicong@huawei.com>
Message-ID: <e3424908-a6a3-63f7-f3b7-e10b00cdfd13@huawei.com>
Date:   Tue, 20 Jun 2023 17:09:46 +0800
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:78.0) Gecko/20100101
 Thunderbird/78.5.1
MIME-Version: 1.0
In-Reply-To: <01c39cb9-e56a-862b-2d99-024b7ba5a769@huawei.com>
Content-Type: text/plain; charset="windows-1252"
Content-Transfer-Encoding: 8bit
Precedence: bulk

Hi Junhao,

On 2023/6/19 21:00, hejunhao wrote:
> Hi, Yicong
> 
> 
> On 2023/6/6 22:22, Yicong Yang wrote:
>> From: Yicong Yang <yangyicong@hisilicon.com>
>>
>> The PTT trace collects PCIe TLP headers from the PCIe link and don't
>> have the ability to exclude certain context. It doesn't support itrace
>> as well. So only advertise PERF_PMU_CAP_NO_EXCLUDE. This will greatly
>> save the storage of final data. Tested tracing idle link for ~15s,
>> without this patch we'll collect ~28.682MB data for context related
>> information and with this patch it reduced to ~0.226MB.
>>
>> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
>> Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
>> ---
>> ? drivers/hwtracing/ptt/hisi_ptt.c | 2 +-
>> ? 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/hwtracing/ptt/hisi_ptt.c b/drivers/hwtracing/ptt/hisi_ptt.c
>> index 5c7e93e7705f..ff2c16efe5b1 100644
>> --- a/drivers/hwtracing/ptt/hisi_ptt.c
>> +++ b/drivers/hwtracing/ptt/hisi_ptt.c
>> @@ -1210,7 +1210,7 @@ static int hisi_ptt_register_pmu(struct hisi_ptt *hisi_ptt)
>> ? ????? hisi_ptt->hisi_ptt_pmu = (struct pmu) {
>> ????????? .module??????? = THIS_MODULE,
>> -??????? .capabilities??? = PERF_PMU_CAP_EXCLUSIVE | PERF_PMU_CAP_ITRACE,
>> +??????? .capabilities??? = PERF_PMU_CAP_NO_EXCLUDE,
> 
> According to the error prompt of function hisi_ptt_trace_start().
> https://elixir.bootlin.com/linux/v6.4-rc7/source/drivers/hwtracing/ptt/hisi_ptt.c#L197
> 
> You may be need to keep advertise PERF_PMU_CAP_EXCLUSIVE?
> Such pmus can only have one event scheduled at a time, if not the perf
> tool will report device busy.
> 

Thanks for point it out. Checking [1] I think we still need this flag here, will add
it back.

[1] https://github.com/torvalds/linux/blob/v6.4-rc6/kernel/events/core.c#L5054

Thanks.