From: John Garry <john.g.garry@oracle.com>
To: Jing Zhang <renyu.zj@linux.alibaba.com>,
linux-arm-kernel@lists.infradead.org,
linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org,
Will Deacon <will@kernel.org>, James Clark <james.clark@arm.com>,
Mike Leach <mike.leach@linaro.org>, Leo Yan <leo.yan@linaro.org>,
Ian Rogers <irogers@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@redhat.com>,
Arnaldo Carvalho de Melo <acme@kernel.org>,
Mark Rutland <mark.rutland@arm.com>,
Alexander Shishkin <alexander.shishkin@linux.intel.com>,
Jiri Olsa <jolsa@kernel.org>, Namhyung Kim <namhyung@kernel.org>,
Andrew Kilroy <andrew.kilroy@arm.com>,
Shuai Xue <xueshuai@linux.alibaba.com>,
Zhuo Song <zhuo.song@linux.alibaba.com>
Subject: Re: [External] : [RFC PATCH v2 1/6] perf vendor events arm64: Add topdown L1 metrics for neoverse-n2
Date: Mon, 21 Nov 2022 17:55:20 +0000 [thread overview]
Message-ID: <75c4f0e6-3f28-a748-e891-7be6016ca28e@oracle.com> (raw)
In-Reply-To: <d904734a-e7c1-ca8e-7705-63fc4864ac4f@linux.alibaba.com>
On 21/11/2022 15:17, Jing Zhang wrote:
> I'm sorry that I misunderstood the purpose of putting metric as arch_std_event at first,
> and now it works after the modification over your suggestion.
>
> But there are also a few questions:
>
> 1. The value of the slot in the topdownL1 is various in different architectures, for example,
> the slot is 5 on neoverse-n2. If I put topdownL1 metric as arch_std_event, then I need to
> specify the slot to 5 in n2. I can specify slot values in metric like below, but is there any
> other concise way to do this?
>
> diff --git a/tools/perf/pmu-events/arch/arm64/arm/neoverse-n2/metrics.json b/tools/perf/pmu-events/arch/arm64/arm/neoverse-n2/metrics.json
> index 8ff1dfe..b473baf 100644
> --- a/tools/perf/pmu-events/arch/arm64/arm/neoverse-n2/metrics.json
> +++ b/tools/perf/pmu-events/arch/arm64/arm/neoverse-n2/metrics.json
> @@ -1,4 +1,23 @@
> [
> + {
> + "MetricExpr": "5",
> + "PublicDescription": "A pipeline slot represents the hardware resources needed to process one uOp",
> + "BriefDescription": "A pipeline slot represents the hardware resources needed to process one uOp",
> + "MetricName": "slot"
Ehhh....I'm not sure if that is a good idea. Ian or anyone else have an
opinion on this? It is possible to reuse metrics, so it should work, but...
One problem is that "slot" would show up as a metric, which you would
not want.
Alternatively I was going to suggest that you can overwrite specific std
arch event attributes. So for example of frontend_bound, you could have:
+ b/tools/perf/pmu-events/arch/arm64/arm/neoverse-n2/metrics.json
@@ -0,0 +1,30 @@
[
{
"ArchStdEvent": "FRONTEND_BOUND",
"MetricExpr": "(stall_slot_frontend - cpu_cycles) / (5 *
cpu_cycles)",
},
> + }
> + {
> + "ArchStdEvent": "FRONTEND_BOUND"
> + },
> + {
> + "ArchStdEvent": "BACKEND_BOUND"
> + },
> + {
> + "ArchStdEvent": "WASTED"
> + },
> + {
> + "ArchStdEvent": "RETIRING"
> + },
>
>
> 2. Should I add the topdownL1 metric to tools/perf/pmu-event/recommended.json,
> or create a new json file to place the general metric?
It would not belong in recommended.json as that is specifically for
arch-recommended events. It would really just depend on where the value
comes from, i.e. arm arm or sbsa.
>
> Looking forward to your reply.
next prev parent reply other threads:[~2022-11-21 17:56 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-10-31 11:11 [PATCH RFC 0/6] Add metrics for neoverse-n2 Jing Zhang
2022-10-31 11:11 ` [PATCH RFC 1/6] perf vendor events arm64: Add topdown L1 " Jing Zhang
2022-10-31 11:11 ` [PATCH RFC 2/6] perf vendor events arm64: Add TLB " Jing Zhang
2022-10-31 11:11 ` [PATCH RFC 3/6] perf vendor events arm64: Add cache " Jing Zhang
2022-10-31 11:11 ` [PATCH RFC 4/6] perf vendor events arm64: Add branch " Jing Zhang
2022-10-31 11:11 ` [PATCH RFC 5/6] perf vendor events arm64: Add PE utilization " Jing Zhang
2022-10-31 11:11 ` [PATCH RFC 6/6] perf vendor events arm64: Add instruction mix " Jing Zhang
2022-11-14 7:41 ` [RFC PATCH v2 0/6] Add " Jing Zhang
2022-11-24 17:14 ` [PATCH v3 " Jing Zhang
2022-11-24 17:14 ` [PATCH v3 1/6] perf vendor events arm64: Add topdown L1 " Jing Zhang
2022-11-24 17:14 ` [PATCH v3 2/6] perf vendor events arm64: Add TLB " Jing Zhang
2022-11-24 17:14 ` [PATCH v3 3/6] perf vendor events arm64: Add cache " Jing Zhang
2022-11-24 17:14 ` [PATCH v3 4/6] perf vendor events arm64: Add branch " Jing Zhang
2022-11-24 17:14 ` [PATCH v3 5/6] perf vendor events arm64: Add PE utilization " Jing Zhang
2022-11-30 18:58 ` Ian Rogers
2022-12-01 11:08 ` Jing Zhang
2022-12-02 20:05 ` Ian Rogers
2022-12-04 7:10 ` Jing Zhang
2022-11-24 17:14 ` [PATCH v3 6/6] perf vendor events arm64: Add instruction mix " Jing Zhang
2022-11-14 7:41 ` [RFC PATCH v2 1/6] perf vendor events arm64: Add topdown L1 " Jing Zhang
2022-11-14 12:59 ` [External] : " John Garry
2022-11-15 8:43 ` Jing Zhang
2022-11-15 11:19 ` John Garry
2022-11-21 9:53 ` Jing Zhang
2022-11-21 10:22 ` John Garry
2022-11-21 15:17 ` Jing Zhang
2022-11-21 17:55 ` John Garry [this message]
2022-11-22 9:24 ` Jing Zhang
2022-11-22 14:00 ` James Clark
2022-11-22 15:41 ` Jing Zhang
2022-11-23 14:26 ` James Clark
2022-11-24 16:32 ` Jing Zhang
2022-11-24 16:51 ` James Clark
2022-11-14 7:41 ` [RFC PATCH v2 2/6] perf vendor events arm64: Add TLB " Jing Zhang
2022-11-14 7:41 ` [RFC PATCH v2 3/6] perf vendor events arm64: Add cache " Jing Zhang
2022-11-14 8:35 ` Xing Zhengjun
2022-11-15 6:28 ` Jing Zhang
2022-11-14 7:41 ` [RFC PATCH v2 4/6] perf vendor events arm64: Add branch " Jing Zhang
2022-11-14 7:41 ` [RFC PATCH v2 5/6] perf vendor events arm64: Add PE utilization " Jing Zhang
2022-11-14 7:42 ` [RFC PATCH v2 6/6] perf vendor events arm64: Add instruction mix " Jing Zhang
2022-11-16 11:19 ` [PATCH RFC 0/6] Add " James Clark
2022-11-16 15:26 ` Jing Zhang
2022-11-21 11:51 ` James Clark
2022-11-22 7:11 ` Jing Zhang
2022-11-22 11:53 ` James Clark
2022-11-19 3:30 ` Jing Zhang
[not found] ` <CAP-5=fW+Z_Tc3BfK1bRKUeKWfxtPfoZXL9D2BhcU1SzNOruSsg@mail.gmail.com>
2022-11-20 3:49 ` Jing Zhang
2022-11-21 11:55 ` James Clark
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=75c4f0e6-3f28-a748-e891-7be6016ca28e@oracle.com \
--to=john.g.garry@oracle.com \
--cc=acme@kernel.org \
--cc=alexander.shishkin@linux.intel.com \
--cc=andrew.kilroy@arm.com \
--cc=irogers@google.com \
--cc=james.clark@arm.com \
--cc=jolsa@kernel.org \
--cc=leo.yan@linaro.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=mike.leach@linaro.org \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.org \
--cc=renyu.zj@linux.alibaba.com \
--cc=will@kernel.org \
--cc=xueshuai@linux.alibaba.com \
--cc=zhuo.song@linux.alibaba.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).