Re: Using perf_event_open() to sample multiple events of a process

From: Nadav Amit
Date: Fri Nov 05 2021 - 20:59:36 EST




> On Nov 5, 2021, at 5:45 PM, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
>
> On Thu, Nov 04, 2021 at 10:57:50PM -0700, Nadav Amit wrote:
>> Hello Ken, Peter,
>>
>> I would appreciate some help regarding the use of perf_event_open()
>> to have multiple samples getting into the same mmap’d memory when they
>> are both attached to the same process.
>>
>> I am doing so (using both PERF_FLAG_FD_NO_GROUP and PERF_FLAG_FD_OUTPUT),
>> but it results in -EINVAL. Debugging the code shows that
>> perf_event_set_output() fails due to the following check:
>>
>> /*
>> * If its not a per-cpu rb, it must be the same task.
>> */
>> if (output_event->cpu == -1 && output_event->ctx != event->ctx)
>> goto out;
>>
>> However, it appears that at this point, event->ctx is not initialized
>> (it is null) so the test fails and the whole perf_event_open() syscall
>> fails.
>>
>> Am I missing something? If not, I am unsure, unfortunately, what the
>> proper way to fix it is…
>>
>> I include a small test that fails on my system. The second
>> perf_event_open fails due to the check in perf_event_set_output():
>>
>
> Works when you use the SET_OUTPUT ioctl()...
>
> I think something went sideways in the syscall path and things went out
> of order :/ I'll try and have a look.

Highly appreciated.

While at it, I would note that there is a mistake in Intel SDM 31.4.2.27,
Table 31-47 “Block Item Packet Definition”. The ID[5:0] takes 6 bits,
when in fact I presume it is only 5 bits (according to the table).

Perhaps you can forward it to the relevant people.

Thanks again,
Nadav