All of lore.kernel.org
 help / color / mirror / Atom feed
From: will.deacon@arm.com (Will Deacon)
To: linux-arm-kernel@lists.infradead.org
Subject: [RFCv4 5/7] drivers/perf: arm_pmu: expose a cpumask in sysfs
Date: Fri, 9 Sep 2016 11:24:43 +0100	[thread overview]
Message-ID: <20160909102443.GE20192@arm.com> (raw)
In-Reply-To: <1473330112-28528-6-git-send-email-mark.rutland@arm.com>

On Thu, Sep 08, 2016 at 11:21:50AM +0100, Mark Rutland wrote:
> In systems with heterogeneous CPUs, there are multiple logical CPU PMUs,
> each of which covers a subset of CPUs in the system. In some cases
> userspace needs to know which CPUs a given logical PMU covers, so we'd
> like to expose a cpumask under sysfs, similar to what is done for uncore
> PMUs.
> 
> Unfortunately, prior to commit 00e727bb389359c8 ("perf stat: Balance
> opening and reading events"), perf stat only correctly handled a cpumask
> holding a single CPU, and only when profiling in system-wide mode. In
> other cases, the presence of a cpumask file could cause perf stat to
> behave erratically.
> 
> Thus, exposing a cpumask file would break older perf binaries in cases
> where they would otherwise work.
> 
> To avoid this issue while still providing userspace with the information
> it needs, this patch exposes a differently-named file (cpus) under
> sysfs. New tools can look for this and operate correctly, while older
> tools will not be adversely affected by its presence.
> 
> Signed-off-by: Mark Rutland <mark.rutland@arm.com>
> Cc: Will Deacon <will.deacon@arm.com>
> ---
>  drivers/perf/arm_pmu.c       | 21 +++++++++++++++++++++
>  include/linux/perf/arm_pmu.h |  1 +
>  2 files changed, 22 insertions(+)
> 
> diff --git a/drivers/perf/arm_pmu.c b/drivers/perf/arm_pmu.c
> index ac83e1e..09e9944 100644
> --- a/drivers/perf/arm_pmu.c
> +++ b/drivers/perf/arm_pmu.c
> @@ -534,6 +534,25 @@ static int armpmu_filter_match(struct perf_event *event)
>  	return cpumask_test_cpu(cpu, &armpmu->supported_cpus);
>  }
>  
> +static ssize_t armpmu_cpumask_show(struct device *dev,
> +				   struct device_attribute *attr, char *buf)
> +{
> +	struct arm_pmu *armpmu = to_arm_pmu(dev_get_drvdata(dev));
> +	return cpumap_print_to_pagebuf(true, buf, &armpmu->supported_cpus);
> +}
> +
> +static struct device_attribute armpmu_cpumask_attr =
> +		__ATTR(cpus, S_IRUGO, armpmu_cpumask_show, NULL);

You can use the DEVICE_ATTR macro for this.

Will

WARNING: multiple messages have this Message-ID (diff)
From: Will Deacon <will.deacon@arm.com>
To: Mark Rutland <mark.rutland@arm.com>
Cc: linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org, acme@kernel.org,
	alexander.shishkin@linux.intel.com, jolsa@kernel.org,
	mingo@redhat.com, peterz@infradead.org
Subject: Re: [RFCv4 5/7] drivers/perf: arm_pmu: expose a cpumask in sysfs
Date: Fri, 9 Sep 2016 11:24:43 +0100	[thread overview]
Message-ID: <20160909102443.GE20192@arm.com> (raw)
In-Reply-To: <1473330112-28528-6-git-send-email-mark.rutland@arm.com>

On Thu, Sep 08, 2016 at 11:21:50AM +0100, Mark Rutland wrote:
> In systems with heterogeneous CPUs, there are multiple logical CPU PMUs,
> each of which covers a subset of CPUs in the system. In some cases
> userspace needs to know which CPUs a given logical PMU covers, so we'd
> like to expose a cpumask under sysfs, similar to what is done for uncore
> PMUs.
> 
> Unfortunately, prior to commit 00e727bb389359c8 ("perf stat: Balance
> opening and reading events"), perf stat only correctly handled a cpumask
> holding a single CPU, and only when profiling in system-wide mode. In
> other cases, the presence of a cpumask file could cause perf stat to
> behave erratically.
> 
> Thus, exposing a cpumask file would break older perf binaries in cases
> where they would otherwise work.
> 
> To avoid this issue while still providing userspace with the information
> it needs, this patch exposes a differently-named file (cpus) under
> sysfs. New tools can look for this and operate correctly, while older
> tools will not be adversely affected by its presence.
> 
> Signed-off-by: Mark Rutland <mark.rutland@arm.com>
> Cc: Will Deacon <will.deacon@arm.com>
> ---
>  drivers/perf/arm_pmu.c       | 21 +++++++++++++++++++++
>  include/linux/perf/arm_pmu.h |  1 +
>  2 files changed, 22 insertions(+)
> 
> diff --git a/drivers/perf/arm_pmu.c b/drivers/perf/arm_pmu.c
> index ac83e1e..09e9944 100644
> --- a/drivers/perf/arm_pmu.c
> +++ b/drivers/perf/arm_pmu.c
> @@ -534,6 +534,25 @@ static int armpmu_filter_match(struct perf_event *event)
>  	return cpumask_test_cpu(cpu, &armpmu->supported_cpus);
>  }
>  
> +static ssize_t armpmu_cpumask_show(struct device *dev,
> +				   struct device_attribute *attr, char *buf)
> +{
> +	struct arm_pmu *armpmu = to_arm_pmu(dev_get_drvdata(dev));
> +	return cpumap_print_to_pagebuf(true, buf, &armpmu->supported_cpus);
> +}
> +
> +static struct device_attribute armpmu_cpumask_attr =
> +		__ATTR(cpus, S_IRUGO, armpmu_cpumask_show, NULL);

You can use the DEVICE_ATTR macro for this.

Will

  reply	other threads:[~2016-09-09 10:24 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-09-08 10:21 [RFCv4 0/7] arm_pmu/perf tools: play nicely with CPU PMU cpumasks Mark Rutland
2016-09-08 10:21 ` Mark Rutland
2016-09-08 10:21 ` [RFCv4 1/7] drivers/perf: arm_pmu: add common attr group fields Mark Rutland
2016-09-08 10:21   ` Mark Rutland
2016-09-09 10:25   ` Will Deacon
2016-09-09 10:25     ` Will Deacon
2016-09-09 11:05     ` Mark Rutland
2016-09-09 11:05       ` Mark Rutland
2016-09-08 10:21 ` [RFCv4 2/7] arm64: perf: move to common attr_group fields Mark Rutland
2016-09-08 10:21   ` Mark Rutland
2016-09-08 10:21 ` [RFCv4 3/7] arm: " Mark Rutland
2016-09-08 10:21   ` Mark Rutland
2016-09-08 10:21 ` [RFCv4 4/7] drivers/perf: arm_pmu: only use common attr_groups Mark Rutland
2016-09-08 10:21   ` Mark Rutland
2016-09-08 10:21 ` [RFCv4 5/7] drivers/perf: arm_pmu: expose a cpumask in sysfs Mark Rutland
2016-09-08 10:21   ` Mark Rutland
2016-09-09 10:24   ` Will Deacon [this message]
2016-09-09 10:24     ` Will Deacon
2016-09-09 11:04     ` Mark Rutland
2016-09-09 11:04       ` Mark Rutland
2016-09-08 10:21 ` [RFCv4 6/7] perf: util: only open events on CPUs an evsel permits Mark Rutland
2016-09-08 10:21   ` Mark Rutland
2016-09-09  5:53   ` [tip:perf/core] perf evlist: Only " tip-bot for Mark Rutland
2016-09-08 10:21 ` [RFCv4 7/7] perf: util: support alternative sysfs cpumask Mark Rutland
2016-09-08 10:21   ` Mark Rutland
2016-09-09  5:54   ` [tip:perf/core] perf pmu: Support " tip-bot for Mark Rutland
2016-09-08 10:38 ` [RFCv4 0/7] arm_pmu/perf tools: play nicely with CPU PMU cpumasks Will Deacon
2016-09-08 10:38   ` Will Deacon
2016-09-08 11:29   ` Mark Rutland
2016-09-08 11:29     ` Mark Rutland
2016-09-08 16:25 ` Arnaldo Carvalho de Melo
2016-09-08 16:25   ` Arnaldo Carvalho de Melo
2016-09-08 18:16   ` Peter Zijlstra
2016-09-08 18:16     ` Peter Zijlstra
2016-09-09  9:31     ` Will Deacon
2016-09-09  9:31       ` Will Deacon

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20160909102443.GE20192@arm.com \
    --to=will.deacon@arm.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.