Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.37 as permitted sender) client-ip=23.128.96.37;
Message-ID: <07677ab2-c29b-499b-b473-f7535fb27a8c@linux.intel.com>
Date:   Tue, 12 Dec 2023 10:56:15 -0500
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH] perf top: Use evsel's cpus to replace user_requested_cpus
To:     Arnaldo Carvalho de Melo <acme@kernel.org>
Cc:     irogers@google.com, namhyung@kernel.org, mark.rutland@arm.com,
        maz@kernel.org, marcan@marcan.st, linux-kernel@vger.kernel.org,
        linux-perf-users@vger.kernel.org
References: <20231208210855.407580-1-kan.liang@linux.intel.com>
 <ZXd7ZuxbNNsjAyqm@kernel.org>
Content-Language: en-US
From:   "Liang, Kan" <kan.liang@linux.intel.com>
In-Reply-To: <ZXd7ZuxbNNsjAyqm@kernel.org>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk


On 2023-12-11 4:13 p.m., Arnaldo Carvalho de Melo wrote:
> Em Fri, Dec 08, 2023 at 01:08:55PM -0800, kan.liang@linux.intel.com escreveu:
>> From: Kan Liang <kan.liang@linux.intel.com>
>>
>> perf top errors out on a hybrid machine
>>  $perf top
>>
>>  Error:
>>  The cycles:P event is not supported.
>>
>> The user_requested_cpus may contain CPUs that are invalid for a hybrid
>> PMU. It causes perf_event_open to fail.
> 
> ?
> 
> All perf top expects is that the "cycles", the most basic one, be
> collected, on all CPUs in the system.
>

Yes, but for hybrid there is no single "cycles" event which can cover
all CPUs. Perf has to split it into two cycles events, cpu_core/cycles/
and cpu_atom/cycles/. Each event has its own CPU mask. If a event is
opened on the unsupported CPU. The open fails. That's the reason perf
top fails. So perf should only open the cycles event on the
corresponding CPU.

>> The commit ef91871c960e ("perf evlist: Propagate user CPU maps
>> intersecting core PMU maps") already intersect the requested CPU map
>> with the CPU map of the PMU. Use the evsel's cpus to replace
>> user_requested_cpus.
>  
>> The evlist's threads is also propagated to evsel's threads in
>> __perf_evlist__propagate_maps(). Replace it as well.
> 
> Thanks, but please try to add more detail to the fix so as to help
> others to consider looking at the patch.

OK. For the threads, the same as other tools, e.g., perf record, perf
appends a dummy for the system wide event. For a per-thread event, the
evlist's thread_map is assigned to the evsel. So using the evsel's
threads is appropriate and also be consistent with other tools.

I will update the description and send a V2.

Thanks,
Kan

> 
> - Arnaldo
>  
>> Reported-by: Arnaldo Carvalho de Melo <acme@kernel.org>
>> Closes: https://lore.kernel.org/linux-perf-users/ZXNnDrGKXbEELMXV@kernel.org/
>> Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
>> ---
>>  tools/perf/builtin-top.c | 4 ++--
>>  1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
>> index ea8c7eca5eee..cce9350177e2 100644
>> --- a/tools/perf/builtin-top.c
>> +++ b/tools/perf/builtin-top.c
>> @@ -1027,8 +1027,8 @@ static int perf_top__start_counters(struct perf_top *top)
>>  
>>  	evlist__for_each_entry(evlist, counter) {
>>  try_again:
>> -		if (evsel__open(counter, top->evlist->core.user_requested_cpus,
>> -				     top->evlist->core.threads) < 0) {
>> +		if (evsel__open(counter, counter->core.cpus,
>> +				counter->core.threads) < 0) {
>>  
>>  			/*
>>  			 * Specially handle overwrite fall back.
>> -- 
>> 2.35.1
>>
>