linux-kernel - Re: [PATCH v2 2/5] perf vendor events: Add JSON metrics for CMN-700

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <75c36607-9cdd-42fe-b3f3-bb049b3eb223@oracle.com>
Date:   Mon, 1 May 2023 13:27:31 +0100
From:   John Garry <john.g.garry@...cle.com>
To:     Jing Zhang <renyu.zj@...ux.alibaba.com>,
        Ian Rogers <irogers@...gle.com>, Will Deacon <will@...nel.org>,
        Shuai Xue <xueshuai@...ux.alibaba.com>
Cc:     James Clark <james.clark@....com>,
        Mike Leach <mike.leach@...aro.org>,
        Leo Yan <leo.yan@...aro.org>,
        Mark Rutland <mark.rutland@....com>,
        Ilkka Koskinen <ilkka@...amperecomputing.com>,
        Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        Jiri Olsa <jolsa@...nel.org>,
        Namhyung Kim <namhyung@...nel.org>,
        Adrian Hunter <adrian.hunter@...el.com>,
        linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
        linux-perf-users@...r.kernel.org,
        Zhuo Song <zhuo.song@...ux.alibaba.com>
Subject: Re: [PATCH v2 2/5] perf vendor events: Add JSON metrics for CMN-700

On 24/04/2023 10:44, Jing Zhang wrote:
> Add JSON metrics for CMN-700. Currently just add part of CMN-700 PMU
> metrics which are general and compatible for any SoC.
> 
> Signed-off-by: Jing Zhang <renyu.zj@...ux.alibaba.com>
> ---
>   .../arch/arm64/arm/cmn700/sys/metrics.json         | 74 ++++++++++++++++++++++
>   tools/perf/pmu-events/jevents.py                   |  1 +
>   2 files changed, 75 insertions(+)
>   create mode 100644 tools/perf/pmu-events/arch/arm64/arm/cmn700/sys/metrics.json
> 
> diff --git a/tools/perf/pmu-events/arch/arm64/arm/cmn700/sys/metrics.json b/tools/perf/pmu-events/arch/arm64/arm/cmn700/sys/metrics.json
> new file mode 100644
> index 0000000..1577d86
> --- /dev/null
> +++ b/tools/perf/pmu-events/arch/arm64/arm/cmn700/sys/metrics.json

I suppose putting these here is ok. Are all these events a must for a 
CMN-700 implementation? Is there potential for impdef event in some 
implementations?

Previously we considered ARM PMCG events would be put in a per-SoC sys 
folder only (and not in perf/pmu-events/arch/arm64/arm)

> @@ -0,0 +1,74 @@
> +[
> +	{
> +		"MetricName": "slc_miss_rate",
> +		"BriefDescription": "The system level cache miss rate include.",
> +		"MetricGroup": "cmn700",
> +		"MetricExpr": "hnf_cache_miss / hnf_slc_sf_cache_access",
> +		"ScaleUnit": "100%",
> +		"Unit": "arm_cmn",
> +		"Compat": "cmn700"
> +	},
> +	{
> +		"MetricName": "hnf_message_retry_rate",
> +		"BriefDescription": "HN-F message retry rate indicates whether a lack of credits is causing the bottlenecks.",
> +		"MetricGroup": "cmn700",
> +		"MetricExpr": "hnf_pocq_retry / hnf_pocq_reqs_recvd",
> +		"ScaleUnit": "100%",
> +		"Unit": "arm_cmn",
> +		"Compat": "cmn700"
> +	},
> +	{
> +		"MetricName": "sf_hit_rate",
> +		"BriefDescription": "Snoop filter hit rate can be used to measure the Snoop Filter efficiency.",
> +		"MetricGroup": "cmn700",
> +		"MetricExpr": "hnf_sf_hit / hnf_slc_sf_cache_access",
> +		"ScaleUnit": "100%",
> +		"Unit": "arm_cmn",
> +		"Compat": "cmn700"
> +	},
> +	{
> +		"MetricName": "mc_message_retry_rate",
> +		"BriefDescription": "The memory controller request retries rate indicates whether the memory controller is the bottleneck.",
> +		"MetricGroup": "cmn700",
> +		"MetricExpr": "hnf_mc_retries / hnf_mc_reqs",
> +		"ScaleUnit": "100%",
> +		"Unit": "arm_cmn",
> +		"Compat": "cmn700"
> +	},
> +	{
> +		"MetricName": "rni_actual_read_bandwidth.all",
> +		"BriefDescription": "This event measure the actual bandwidth(MB/sec) that RN-I bridge sends to the interconnect.",
> +		"MetricGroup": "cmn700",
> +		"MetricExpr": "rnid_rxdat_flits * 32 / 1e6 / duration_time",
> +		"ScaleUnit": "1MB/s",
> +		"Unit": "arm_cmn",
> +		"Compat": "cmn700"
> +	},
> +	{
> +		"MetricName": "rni_actual_write_bandwidth.all",
> +		"BriefDescription": "This event measures the actual write bandwidth(MB/sec) at RN-I bridges.",
> +		"MetricGroup": "cmn700",
> +		"MetricExpr": "rnid_txdat_flits * 32 / 1e6 / duration_time",
> +		"ScaleUnit": "1MB/s",
> +		"Unit": "arm_cmn",
> +		"Compat": "cmn700"
> +	},
> +	{
> +		"MetricName": "rni_retry_rate",
> +		"BriefDescription": "RN-I bridge retry rate indicates whether the memory controller is the bottleneck.",
> +		"MetricGroup": "cmn700",
> +		"MetricExpr": "rnid_txreq_flits_retried / rnid_txreq_flits_total",
> +		"ScaleUnit": "100%",
> +		"Unit": "arm_cmn",
> +		"Compat": "cmn700"
> +	},
> +	{
> +		"MetricName": "sbsx_actual_write_bandwidth.all",
> +		"BriefDescription": "sbsx actual write bandwidth(MB/sec).",
> +		"MetricGroup": "cmn700",
> +		"MetricExpr": "sbsx_txdat_flitv * 32 / 1e6 / duration_time",
> +		"ScaleUnit": "1MB/s",
> +		"Unit": "arm_cmn",
> +		"Compat": "cmn700"
> +	}
> +]
> diff --git a/tools/perf/pmu-events/jevents.py b/tools/perf/pmu-events/jevents.py
> index 2bcd07c..7cff2c6 100755
> --- a/tools/perf/pmu-events/jevents.py
> +++ b/tools/perf/pmu-events/jevents.py
> @@ -256,6 +256,7 @@ class JsonEvent:
>             'DFPMC': 'amd_df',
>             'cpu_core': 'cpu_core',
>             'cpu_atom': 'cpu_atom',
> +          'arm_cmn': 'arm_cmn',
>         }
>         return table[unit] if unit in table else f'uncore_{unit.lower()}'
>