linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Mark Rutland <mark.rutland@arm.com>
To: Yicong Yang <yangyicong@huawei.com>
Cc: will@kernel.org, linux-arm-kernel@lists.infradead.org,
	james.clark@linaro.org, robh@kernel.org,
	anshuman.khandual@arm.com, jonathan.cameron@huawei.com,
	hejunhao3@huawei.com, linuxarm@huawei.com,
	prime.zeng@hisilicon.com, xuwei5@huawei.com,
	wangyushan12@huawei.com, yangyicong@hisilicon.com
Subject: Re: [PATCH 2/2] perf: arm_pmuv3: Don't use PMCCNTR_EL0 on SMT cores
Date: Tue, 12 Aug 2025 11:33:14 +0100	[thread overview]
Message-ID: <aJsYah_DGPIQ5v3-@J2N7QTR9R3> (raw)
In-Reply-To: <20250812080830.20796-3-yangyicong@huawei.com>

On Tue, Aug 12, 2025 at 04:08:30PM +0800, Yicong Yang wrote:
> From: Yicong Yang <yangyicong@hisilicon.com>
> 
> CPU_CYCLES is expected to count the logical CPU (PE) clock. Currently it's
> preferred to use PMCCNTR_EL0 for counting CPU_CYCLES, but it'll count
> processor clock rather than the PE clock (ARM DDI0487 L.b D13.1.3) if
> one of the SMT siblings is not idle on a multi-threaded implementation.
>
> So don't use it on SMT cores.

This is rather unfortunate.

When does this actually matter?

Per ARM DDI 0487 L.b, page D14-6918:

| If FEAT_PMUv3p9 is implemented, then CPU_CYCLES does not increment
| when the clocks are stopped by WFI and WFE instructions. Otherwise, it
| is CONSTRAINED UNPREDICTABLE whether or not CPU_CYCLES continues to
| increment when the clocks are stopped by WFI and WFE instructions.

... so prior to FEAT_PMUv3p9, no-one could rely on the difference
anyway.

> When counting cycles on SMT CPU 2-3 and CPU 3 is idle, without this
> patch we'll get:
> [root@client1 tmp]# perf stat -e cycles -A -C 2-3 -- stress-ng -c 1
> --taskset 2 --timeout 1
> [...]
>  Performance counter stats for 'CPU(s) 2-3':
> 
> CPU2           2880457316      cycles
> CPU3           2880459810      cycles
>        1.254688470 seconds time elapsed
> 
> With this patch the idle state of CPU3 is observed as expected:
> [root@client1 ~]#  perf stat -e cycles -A -C 2-3 -- stress-ng -c 1
> --taskset 2 --timeout 1
> [...]
>  Performance counter stats for 'CPU(s) 2-3':
> 
> CPU2           2558580492      cycles
> CPU3               305749      cycles
>        1.113626410 seconds time elapsed
> 
> Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
> ---
>  drivers/perf/arm_pmuv3.c | 9 +++++++++
>  1 file changed, 9 insertions(+)
> 
> diff --git a/drivers/perf/arm_pmuv3.c b/drivers/perf/arm_pmuv3.c
> index 95c899d07df5..ed3149632b71 100644
> --- a/drivers/perf/arm_pmuv3.c
> +++ b/drivers/perf/arm_pmuv3.c
> @@ -1002,6 +1002,15 @@ static bool armv8pmu_can_use_pmccntr(struct pmu_hw_events *cpuc,
>  	if (has_branch_stack(event))
>  		return false;
>  
> +	/*
> +	 * The PMCCNTR_EL0 increments from the processor clock rather than
> +	 * the PE clock (ARM DDI0487 L.b D13.1.3) which means it'll continue
> +	 * counting on a WFI PE if one of its SMT silbing is not idle on a
> +	 * multi-threaded implementation. So don't use it on SMT cores.
> +	 */
> +	if (cpumask_weight(topology_sibling_cpumask(smp_processor_id())) > 1)
> +		return false;

This effectively forbids use of PMCCNTR_EL0 for any events.

Is there any existing event that it is useful for?

Mark.

> +
>  	return true;
>  }
>  
> -- 
> 2.24.0
> 


  parent reply	other threads:[~2025-08-12 16:19 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-08-12  8:08 [PATCH 0/2] perf: arm_pmuv3: Don't use PMCCNTR_EL0 on SMT cores Yicong Yang
2025-08-12  8:08 ` [PATCH 1/2] perf: arm_pmuv3: Factor out PMCCNTR_EL0 use conditions Yicong Yang
2025-08-12 10:02   ` James Clark
2025-08-12 10:25   ` Mark Rutland
2025-08-12  8:08 ` [PATCH 2/2] perf: arm_pmuv3: Don't use PMCCNTR_EL0 on SMT cores Yicong Yang
2025-08-12 10:00   ` James Clark
2025-08-12 10:14     ` Yicong Yang
2025-08-12 10:22       ` Mark Rutland
2025-08-13  8:17         ` Yicong Yang
2025-08-12 10:31       ` James Clark
2025-08-13  8:32         ` Yicong Yang
2025-08-12 10:33   ` Mark Rutland [this message]
2025-08-13  8:03     ` Yicong Yang

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=aJsYah_DGPIQ5v3-@J2N7QTR9R3 \
    --to=mark.rutland@arm.com \
    --cc=anshuman.khandual@arm.com \
    --cc=hejunhao3@huawei.com \
    --cc=james.clark@linaro.org \
    --cc=jonathan.cameron@huawei.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linuxarm@huawei.com \
    --cc=prime.zeng@hisilicon.com \
    --cc=robh@kernel.org \
    --cc=wangyushan12@huawei.com \
    --cc=will@kernel.org \
    --cc=xuwei5@huawei.com \
    --cc=yangyicong@hisilicon.com \
    --cc=yangyicong@huawei.com \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).