Received-SPF: pass (google.com: domain of linux-kernel+bounces-199456-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Precedence: bulk
MIME-Version: 1.0
References: <20240603092812.46616-1-yangyicong@huawei.com>
In-Reply-To: <20240603092812.46616-1-yangyicong@huawei.com>
From: Ian Rogers <irogers@google.com>
Date: Mon, 3 Jun 2024 09:42:10 -0700
Message-ID: <CAP-5=fXw0=dHORC-kObYFiAco64PGP4_uFr__f9YmVctWtGSfA@mail.gmail.com>
Subject: Re: [PATCH 0/3] Perf avoid opening events on offline CPUs
To: Yicong Yang <yangyicong@huawei.com>
Cc: will@kernel.org, mark.rutland@arm.com, acme@kernel.org, 
	namhyung@kernel.org, linux-arm-kernel@lists.infradead.org, 
	linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org, 
	peterz@infradead.org, mingo@redhat.com, alexander.shishkin@linux.intel.com, 
	jolsa@kernel.org, james.clark@arm.com, dongli.zhang@oracle.com, 
	jonathan.cameron@huawei.com, prime.zeng@hisilicon.com, linuxarm@huawei.com, 
	yangyicong@hisilicon.com
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Mon, Jun 3, 2024 at 2:33=E2=80=AFAM Yicong Yang <yangyicong@huawei.com> =
wrote:
>
> From: Yicong Yang <yangyicong@hisilicon.com>
>
> If user doesn't specify the CPUs, perf will try to open events on CPUs
> of the PMU which is initialized from the PMU's "cpumask" or "cpus" sysfs
> attributes if provided. But we doesn't check whether the CPUs provided
> by the PMU are all online. So we may open events on offline CPUs if PMU
> driver provide offline CPUs and then we'll be rejected by the kernel:
>
> [root@localhost yang]# echo 0 > /sys/devices/system/cpu/cpu0/online

Generally Linux won't let you take CPU0 off line, I'm not able to
follow this step on x86 Linux. Fwiw, I routinely run perf with the
core hyperthread siblings offline.

Thanks,
Ian

> [root@localhost yang]# ./perf_static stat -e armv8_pmuv3_0/cycles/ --time=
out 100
> Error:
> The sys_perf_event_open() syscall returned with 19 (No such device) for e=
vent (cpu-clock).
> /bin/dmesg | grep -i perf may provide additional information.
>
> This patchset tries to avoid this case by:
> - Double check the PMU's cpumask in the perf tool and only include online=
 CPUs
> - Trying to make the PMU drivers only export online CPUs in its "cpus" or=
 "cpumask"
>   attributes
>
> Previously discussion can be found at [1]. Will suggested to do it in use=
rspace.
> I think it makes sense to do a double check in the perf tool in case the =
driver
> doesn't do this. So PATCH 1/3 is added in this version.
>
> [1] https://lore.kernel.org/linux-arm-kernel/20240410095833.63934-1-yangy=
icong@huawei.com/
>
> Yicong Yang (3):
>   perf pmu: Limit PMU cpumask to online CPUs
>   perf: arm_pmu: Only show online CPUs in device's "cpus" attribute
>   perf: arm_spe: Only show online CPUs in device's "cpumask" attribute
>
>  drivers/perf/arm_pmu.c     | 24 +++++++++++++++++++++++-
>  drivers/perf/arm_spe_pmu.c | 22 +++++++++++++++++++++-
>  tools/perf/util/pmu.c      | 13 +++++++++++--
>  3 files changed, 55 insertions(+), 4 deletions(-)
>
> --
> 2.24.0
>