[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <abeaafbe-2290-d272-ddd1-f358f7271edc@oracle.com>
Date: Fri, 28 Jul 2023 09:30:00 +0100
From: John Garry <john.g.garry@...cle.com>
To: Jing Zhang <renyu.zj@...ux.alibaba.com>,
Ian Rogers <irogers@...gle.com>
Cc: Will Deacon <will@...nel.org>, Mark Rutland <mark.rutland@....com>,
Robin Murphy <robin.murphy@....com>,
Ilkka Koskinen <ilkka@...amperecomputing.com>,
Namhyung Kim <namhyung@...nel.org>,
Arnaldo Carvalho de Melo <acme@...nel.org>,
linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
linux-perf-users@...r.kernel.org, linux-doc@...r.kernel.org,
Zhuo Song <zhuo.song@...ux.alibaba.com>,
Shuai Xue <xueshuai@...ux.alibaba.com>
Subject: Re: [PATCH v5 3/5] perf test: Add pmu-event test for "Compat" and new
event_field.
On 28/07/2023 07:17, Jing Zhang wrote:
> Add new event test for uncore system event which is used to verify the
> functionality of "Compat" matching multiple identifiers and the new event
> fields "EventIdCode" and "Type".
>
Thanks for doing this. It looks ok. I just have a comment, below.
> Signed-off-by: Jing Zhang <renyu.zj@...ux.alibaba.com>
> ---
> .../pmu-events/arch/test/test_soc/sys/uncore.json | 8 ++++
> tools/perf/tests/pmu-events.c | 55 ++++++++++++++++++++++
> 2 files changed, 63 insertions(+)
>
> diff --git a/tools/perf/pmu-events/arch/test/test_soc/sys/uncore.json b/tools/perf/pmu-events/arch/test/test_soc/sys/uncore.json
> index c7e7528..879a0ae 100644
> --- a/tools/perf/pmu-events/arch/test/test_soc/sys/uncore.json
> +++ b/tools/perf/pmu-events/arch/test/test_soc/sys/uncore.json
> @@ -12,5 +12,13 @@
> "EventName": "sys_ccn_pmu.read_cycles",
> "Unit": "sys_ccn_pmu",
> "Compat": "0x01"
> + },
> + {
> + "BriefDescription": "Counts total cache misses in first lookup result (high priority).",
> + "Type": "0x05",
> + "EventIdCode": "0x01",
> + "EventName": "sys_cmn_pmu.hnf_cache_miss",
> + "Unit": "arm_cmn",
> + "Compat": "434*;436*;43c*;43a01"
> }
> ]
> diff --git a/tools/perf/tests/pmu-events.c b/tools/perf/tests/pmu-events.c
> index 1dff863b..e227dcd 100644
> --- a/tools/perf/tests/pmu-events.c
> +++ b/tools/perf/tests/pmu-events.c
> @@ -255,9 +255,24 @@ struct perf_pmu_test_pmu {
> .matching_pmu = "uncore_sys_ccn_pmu",
> };
>
> +static const struct perf_pmu_test_event sys_cmn_pmu_hnf_cache_miss = {
> + .event = {
> + .name = "sys_cmn_pmu.hnf_cache_miss",
> + .event = "type=0x05,eventid=0x01",
> + .desc = "Counts total cache misses in first lookup result (high priority). Unit: uncore_arm_cmn ",
> + .topic = "uncore",
> + .pmu = "uncore_arm_cmn",
> + .compat = "434*;436*;43c*;43a01",
> + },
> + .alias_str = "type=0x5,eventid=0x1",
> + .alias_long_desc = "Counts total cache misses in first lookup result (high priority). Unit: uncore_arm_cmn ",
> + .matching_pmu = "uncore_arm_cmn_0",
> +};
> +
> static const struct perf_pmu_test_event *sys_events[] = {
> &sys_ddr_pmu_write_cycles,
> &sys_ccn_pmu_read_cycles,
> + &sys_cmn_pmu_hnf_cache_miss,
> NULL
> };
>
> @@ -699,6 +714,46 @@ static int __test_uncore_pmu_event_aliases(struct perf_pmu_test_pmu *test_pmu)
> &sys_ccn_pmu_read_cycles,
> },
> },
> + {
> + .pmu = {
> + .name = (char *)"uncore_arm_cmn_0",
> + .is_uncore = 1,
> + .id = (char *)"43401",
> + },
> + .aliases = {
> + &sys_cmn_pmu_hnf_cache_miss,
> + },
> + },
> + {
> + .pmu = {
> + .name = (char *)"uncore_arm_cmn_0",
> + .is_uncore = 1,
> + .id = (char *)"43602",
> + },
> + .aliases = {
> + &sys_cmn_pmu_hnf_cache_miss,
> + },
> + },
> + {
> + .pmu = {
> + .name = (char *)"uncore_arm_cmn_1",
Shouldn't this match some perf_pmu_test_event entry with same
matching_pmu member? But is perf_pmu_test_event.matching_pmu member ever
used for any checking???
Thanks,
John
> + .is_uncore = 1,
> + .id = (char *)"43c03",
> + },
> + .aliases = {
> + &sys_cmn_pmu_hnf_cache_miss,
> + },
> + },
> + {
> + .pmu = {
> + .name = (char *)"uncore_arm_cmn_1",
> + .is_uncore = 1,
> + .id = (char *)"43a01",
> + },
> + .aliases = {
> + &sys_cmn_pmu_hnf_cache_miss,
> + },
> + }
> };
>
> /* Test that aliases generated are as expected */
Powered by blists - more mailing lists