From: "Liang, Kan" <kan.liang@linux.intel.com>
To: Leo Yan <leo.yan@linaro.org>
Cc: acme@kernel.org, irogers@google.com, peterz@infradead.org,
mingo@redhat.com, namhyung@kernel.org, jolsa@kernel.org,
adrian.hunter@intel.com, john.g.garry@oracle.com,
will@kernel.org, james.clark@arm.com, mike.leach@linaro.org,
yuhaixin.yhx@linux.alibaba.com, renyu.zj@linux.alibaba.com,
tmricht@linux.ibm.com, ravi.bangoria@amd.com,
linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH V2 3/5] perf mem: Clean up perf_mem_events__name()
Date: Wed, 13 Dec 2023 11:17:47 -0500 [thread overview]
Message-ID: <64c2808e-ac2b-41c0-9e98-61b22e3031c1@linux.intel.com> (raw)
In-Reply-To: <20231213133336.GA3895246@leoy-yangtze.lan>
On 2023-12-13 8:33 a.m., Leo Yan wrote:
>>> I am a bit suspect if we really need the field '.aux_event', the
>>> '.aux_event' field is only used for generating event string.
>> No, it stores the event encoding for the extra event.
>> ARM doesn't need it, so it's 0.
> I searched a bit and confirmed '.aux_event' is only used in
> util/mem-events.c and for 'perf record'.
>
> I failed to connect the code with "it stores the event encoding for the
> extra event". Could you elaborate a bit for this?
The details of the reason of introducing the mem_load aux event can be
found here.
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=61b985e3e775a3a75fda04ce7ef1b1aefc4758bc
A mem_load_aux event is a new requirement for SPR. For the other Intel
platforms, a single mem_load event is good enough to collect the data
source information. But for SPR, we have to group both the mem_load
event and the mem_load_aux event when collecting the data source
information. In the group, the mem_load_aux event must be the leader
event. But for the sample read case, only the sampling of the mem_load
make sense. So the is_mem_loads_aux_event() is introduced to switch the
sampling event to the mem_load event. Here is the perf tool patch.
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=2a57d40832dc8366bc517bcbbfdb1d7fb583735b
The .aux_event is to store the event encoding of the mem_load_aux event.
If it's the leader of a sampling read group, we should use the second
event as a sampling event.
Thanks,
Kan
next prev parent reply other threads:[~2023-12-13 16:17 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-12-07 19:23 [PATCH V2 0/5] Clean up perf mem kan.liang
2023-12-07 19:23 ` [PATCH V2 1/5] perf mem: Add mem_events into the supported perf_pmu kan.liang
2023-12-08 10:29 ` Leo Yan
2023-12-08 18:14 ` Liang, Kan
2023-12-09 6:34 ` Leo Yan
2023-12-11 19:01 ` Liang, Kan
2023-12-13 14:24 ` Leo Yan
2023-12-13 16:19 ` Liang, Kan
2023-12-07 19:23 ` [PATCH V2 2/5] perf mem: Clean up perf_mem_events__ptr() kan.liang
2023-12-09 4:31 ` Leo Yan
2023-12-11 18:09 ` Liang, Kan
2023-12-07 19:23 ` [PATCH V2 3/5] perf mem: Clean up perf_mem_events__name() kan.liang
2023-12-08 0:01 ` Ian Rogers
2023-12-09 5:48 ` Leo Yan
2023-12-11 18:39 ` Liang, Kan
2023-12-13 13:33 ` Leo Yan
2023-12-13 16:17 ` Liang, Kan [this message]
2023-12-13 17:33 ` Ian Rogers
2023-12-18 3:21 ` Leo Yan
2023-12-07 19:23 ` [PATCH V2 4/5] perf mem: Clean up perf_mem_event__supported() kan.liang
2023-12-09 6:17 ` Leo Yan
2023-12-11 18:44 ` Liang, Kan
2023-12-13 13:51 ` Leo Yan
2023-12-13 13:55 ` Ravi Bangoria
2023-12-07 19:23 ` [PATCH V2 5/5] perf mem: Clean up is_mem_loads_aux_event() kan.liang
2023-12-09 6:27 ` Leo Yan
2023-12-11 18:45 ` Liang, Kan
2023-12-07 20:31 ` [PATCH V2 0/5] Clean up perf mem Arnaldo Carvalho de Melo
2023-12-13 9:51 ` Athira Rajeev
2023-12-13 19:54 ` Liang, Kan
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=64c2808e-ac2b-41c0-9e98-61b22e3031c1@linux.intel.com \
--to=kan.liang@linux.intel.com \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=irogers@google.com \
--cc=james.clark@arm.com \
--cc=john.g.garry@oracle.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=mike.leach@linaro.org \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.org \
--cc=ravi.bangoria@amd.com \
--cc=renyu.zj@linux.alibaba.com \
--cc=tmricht@linux.ibm.com \
--cc=will@kernel.org \
--cc=yuhaixin.yhx@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).