Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Nadav Amit <nadav.amit@gmail.com>
Content-Type: text/plain;
        charset=utf-8
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0 (Mac OS X Mail 14.0 \(3654.120.0.1.13\))
Subject: Using perf_event_open() to sample multiple events of a process
Message-Id: <92645262-D319-4068-9C44-2409EF44888E@gmail.com>
Date:   Thu, 4 Nov 2021 22:57:50 -0700
Cc:     LKML <linux-kernel@vger.kernel.org>,
        linux-perf-users@vger.kernel.org
To:     kan.liang@linux.intel.com, Peter Zijlstra <peterz@infradead.org>
Precedence: bulk

Hello Ken, Peter,

I would appreciate some help regarding the use of perf_event_open()
to have multiple samples getting into the same mmap=E2=80=99d 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 =3D=3D -1 && output_event->ctx !=3D =
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=E2=80=A6

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():


#define _GNU_SOURCE 1

#include <asm/unistd.h>
#include <linux/perf_event.h>
#include <linux/hw_breakpoint.h>
#include <sys/mman.h>
#include <unistd.h>
#include <stdlib.h>
#include <stdio.h>

long perf_event_open(struct perf_event_attr* event_attr, pid_t pid, int =
cpu,
		     int group_fd, unsigned long flags)
{
    return syscall(__NR_perf_event_open, event_attr, pid, cpu, group_fd, =
flags);
}

int main(void)
{
	pid_t pid =3D getpid();
	int group_fd, fd;
	void *p;
	struct perf_event_attr pe =3D {
		.type =3D 4,
		.size =3D sizeof(struct perf_event_attr),
		.config =3D 0x11d0,
		.sample_type =3D 0x8,
		.sample_period =3D 1000,
		.precise_ip =3D 2,
	};

	group_fd =3D perf_event_open(&pe, pid, -1, -1, =
PERF_FLAG_FD_CLOEXEC |=20
						     =
PERF_FLAG_FD_NO_GROUP |
						     =
PERF_FLAG_FD_OUTPUT);

	if (group_fd < 0) {
		perror("first perf_event_open");
		exit(-1);
	}

	p =3D mmap(NULL, 3 * 4096, PROT_READ|PROT_WRITE, MAP_SHARED, =
group_fd, 0);

	if (p =3D=3D MAP_FAILED) {
		perror("MAP_FAILED");
		exit(-1);
	}
=09
	pe.config =3D 0x12d0;
=09
	fd =3D perf_event_open(&pe, pid, -1, group_fd, =
PERF_FLAG_FD_CLOEXEC |=20
						     =
PERF_FLAG_FD_NO_GROUP |
						     =
PERF_FLAG_FD_OUTPUT);

	if (fd < 0) {
		perror("second perf_event_open");
		exit(-1);
	}

	printf("success\n");
}