linux-kernel - Re: [RFC PATCH v2 0/2] Enable Core RAPL for AMD

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] [day] [month] [year] [list]

Message-ID: <2670154.mvXUDI8C0e@natalenko.name>
Date:   Thu, 02 Mar 2023 20:45:47 +0100
From:   Oleksandr Natalenko <oleksandr@...alenko.name>
To:     Peter Zijlstra <peterz@...radead.org>,
        Ingo Molnar <mingo@...hat.com>,
        Arnaldo Carvalho de Melo <acme@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        Jiri Olsa <jolsa@...nel.org>,
        Namhyung Kim <namhyung@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Borislav Petkov <bp@...en8.de>,
        Dave Hansen <dave.hansen@...ux.intel.com>, hpa@...or.com,
        Stephane Eranian <eranian@...gle.com>,
        Wyes Karny <wyes.karny@....com>
Cc:     x86@...nel.org, linux-perf-users@...r.kernel.org,
        linux-kernel@...r.kernel.org, gautham.shenoy@....com,
        ananth.narayan@....com,
        Muhammad Usama Anjum <usama.anjum@...labora.com>,
        Wyes Karny <wyes.karny@....com>
Subject: Re: [RFC PATCH v2 0/2] Enable Core RAPL for AMD

Hello.

On středa 1. března 2023 19:14:47 CET Wyes Karny wrote:
> This is v2 patch set for enabling per-core RAPL counters through
> energy-cores event on RAPL PMU. Added RFC tag because a new patch is
> added and this additional patch may need clarification from the
> community.
> 
> While enabling this support with v1 patch [1], Stephane reported an
> issue where for the first reading of energy-cores event for every core was
> invalid [2]. This issue is addressed with patch 1 of this series. There
> is no change in patch 2.
> 
> [1]: https://lore.kernel.org/lkml/20230217161354.129442-1-wyes.karny@amd.com/
> [2]: https://lore.kernel.org/lkml/CABPqkBQ_bSTC-OEe_LrgUrpj2VsseX1ThvO-yLcEtF8vb4+AAw@mail.gmail.com/
> 
> Change log:
> v1 -> v2:
> - Added new patch to fix the issue reported by Stephane
> - Added RFC tag
> 
> Wyes Karny (2):
>   perf/x86/rapl: Fix energy-cores event
>   perf/x86/rapl: Enable Core RAPL for AMD
> 
>  arch/x86/events/rapl.c | 20 +++++++++++++++-----
>  1 file changed, 15 insertions(+), 5 deletions(-)

Since I was asked to re-test v2:

```
$ sudo perf stat -a --per-core -C 0-15 -e power/energy-cores/ -- dd if=/dev/zero of=/dev/null bs=1M count=1000000
1000000+0 records in
1000000+0 records out
1048576000000 bytes (1,0 TB, 977 GiB) copied, 17,0321 s, 61,6 GB/s

Performance counter stats for 'system wide':

S0-D0-C0              1               6,67 Joules power/energy-cores/
S0-D0-C1              1             197,93 Joules power/energy-cores/
S0-D0-C2              1              11,32 Joules power/energy-cores/
S0-D0-C3              1              13,88 Joules power/energy-cores/
S0-D0-C4              1              12,40 Joules power/energy-cores/
S0-D0-C5              1               5,10 Joules power/energy-cores/
S0-D0-C6              1               3,92 Joules power/energy-cores/
S0-D0-C7              1               5,65 Joules power/energy-cores/
S0-D0-C8              1               4,91 Joules power/energy-cores/
S0-D0-C9              1              57,18 Joules power/energy-cores/
S0-D0-C10             1             174,95 Joules power/energy-cores/
S0-D0-C11             1               3,58 Joules power/energy-cores/
S0-D0-C12             1               5,41 Joules power/energy-cores/
S0-D0-C13             1               5,33 Joules power/energy-cores/
S0-D0-C14             1               2,89 Joules power/energy-cores/
S0-D0-C15             1               5,61 Joules power/energy-cores/

17,034519241 seconds time elapsed
```

Tested-by: Oleksandr Natalenko <oleksandr@...alenko.name>

Thanks.

-- 
Oleksandr Natalenko (post-factum)