linux-kernel - Re: [RFC v3 6/6] arm64/amu: use capacity_ref

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]

Date:   Thu, 26 Oct 2023 13:19:49 +0200
From:   Dietmar Eggemann <dietmar.eggemann@....com>
To:     Vincent Guittot <vincent.guittot@...aro.org>,
        Ionela Voinescu <ionela.voinescu@....com>
Cc:     linux@...linux.org.uk, catalin.marinas@....com, will@...nel.org,
        paul.walmsley@...ive.com, palmer@...belt.com,
        aou@...s.berkeley.edu, sudeep.holla@....com,
        gregkh@...uxfoundation.org, rafael@...nel.org, mingo@...hat.com,
        peterz@...radead.org, juri.lelli@...hat.com, rostedt@...dmis.org,
        bsegall@...gle.com, mgorman@...e.de, bristot@...hat.com,
        vschneid@...hat.com, viresh.kumar@...aro.org, lenb@...nel.org,
        robert.moore@...el.com, lukasz.luba@....com,
        pierre.gondois@....com, linux-arm-kernel@...ts.infradead.org,
        linux-kernel@...r.kernel.org, linux-riscv@...ts.infradead.org,
        linux-pm@...r.kernel.org, linux-acpi@...r.kernel.org,
        acpica-devel@...ts.linuxfoundation.org, conor.dooley@...rochip.com,
        suagrfillet@...il.com, ajones@...tanamicro.com, lftan@...nel.org
Subject: Re: [RFC v3 6/6] arm64/amu: use capacity_ref_freq to set AMU ratio

On 24/10/2023 11:58, Vincent Guittot wrote:
> On Mon, 23 Oct 2023 at 22:58, Ionela Voinescu <ionela.voinescu@....com> wrote:
>>
>> Hi,
>>
>> On Wednesday 18 Oct 2023 at 18:25:40 (+0200), Vincent Guittot wrote:
>>> Use the new capacity_ref_freq to set the ratio that is used by AMU for
>>> computing the arch_scale_freq_capacity().
>>> This helps to keep everything aligned using the same reference for
>>> computing CPUs capacity.
>>>
>>> The default value of the ratio ensures that arch_scale_freq_capacity()
>>> returns max capacity until it is set to its correct value with the
>>> cpu capacity and capacity_ref_freq.

Nitpick: Could you mention that arch_max_freq_scale is the default value
for this ratio? Took me a while to recreate the (not so simple) story
for this change, i.e. make the connection between ratio and
arch_max_freq_scale.

init_cpu_capacity_callback()

  freq_inv_set_max_ratio()

    u64 ratio
    ...
    per_cpu(arch_max_freq_scale, cpu) = (unsigned long)ratio
            ^^^^^^^^^^^^^^^^^^^


static struct scale_freq_data amu_sfd = {
        .set_freq_scale = amu_scale_freq_tick,
}

#define arch_scale_freq_tick topology_scale_freq_tick

topology_scale_freq_tick()

  sfd->set_freq_scale()


amu_scale_freq_tick()

  ...
  scale *= this_cpu_read(arch_max_freq_scale)
                         ^^^^^^^^^^^^^^^^^^^
  ...
  this_cpu_write(arch_freq_scale, (unsigned long)scale);


#define arch_scale_freq_capacity topology_get_freq_scale

topology_get_freq_scale(cpu)

  return per_cpu(arch_freq_scale, cpu)

[...]