linux-kernel - Re: [PATCH v2 2/2] cpufreq: CPPC: Dont read counters for idle CPUs

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: <CAFivqm+D9mbGku-nZKSUEMcQV5XK_ayarxL9gpV5JyfmhirsPw@mail.gmail.com>
Date: Wed, 2 Jul 2025 11:38:11 -0700
From: Prashant Malani <pmalani@...gle.com>
To: Jie Zhan <zhanjie9@...ilicon.com>, Ionela Voinescu <ionela.voinescu@....com>, 
	Beata Michalska <beata.michalska@....com>
Cc: Ben Segall <bsegall@...gle.com>, Dietmar Eggemann <dietmar.eggemann@....com>, 
	Ingo Molnar <mingo@...hat.com>, Juri Lelli <juri.lelli@...hat.com>, 
	open list <linux-kernel@...r.kernel.org>, 
	"open list:CPU FREQUENCY SCALING FRAMEWORK" <linux-pm@...r.kernel.org>, Mel Gorman <mgorman@...e.de>, 
	Peter Zijlstra <peterz@...radead.org>, "Rafael J. Wysocki" <rafael@...nel.org>, 
	Steven Rostedt <rostedt@...dmis.org>, Valentin Schneider <vschneid@...hat.com>, 
	Vincent Guittot <vincent.guittot@...aro.org>, Viresh Kumar <viresh.kumar@...aro.org>, 
	z00813676 <zhenglifeng1@...wei.com>
Subject: Re: [PATCH v2 2/2] cpufreq: CPPC: Dont read counters for idle CPUs

Hi All,

Ionela, Beata, could you kindly review ?

On Fri, 27 Jun 2025 at 10:07, Prashant Malani <pmalani@...gle.com> wrote:
>
> Hi Jie,
>
> On Fri, 27 Jun 2025 at 00:55, Jie Zhan <zhanjie9@...ilicon.com> wrote:
> >
> >
> > Hi Prashant,
> >
> > Sorry for a late reply as I'm busy on other stuff and this doesn't seem to
> > be an easy issue to solve.
> >
>
> No worries, the ping was in general to all the people in the thread :)
>
> > For the latest kernel, [1] provides a new 'cpuinfo_avg_freq' sysfs file to
> > reflect the frequency base on AMUs, which is supposed to be more stable.
> > Though it usually shows 'Resource temporarily unavailable' on my platform
> > at the moment and looks a bit buggy.
> >
> > Most of the related discussions can be found in the reference links in [1].
> > [1] https://lore.kernel.org/linux-pm/20250131162439.3843071-1-beata.michalska@arm.com/
> >
> > As reported, the current frequency sampling method may show an large error
> > on 1) 100% load, 2) high memory access pressure, 3) idle cpus in your case.
> >
> > AFAICS, they may all come from the unstable latency accessing remote AMUs
> > for 4 times but delaying a fixed 2us sampling window.
>
> I tried applying [1] which consolidates the ref and del register reads
> into 1 IPI, but that did not make a difference. The values still
> fluctuate wildly.
>
> >
> > Increase the sampling windows seems to help but also increase the time
> > overhead, so that's not favoured by people.
> >
>
> This experiment did not appear to help in our case. It's a point in
> the direction that this method is inherently inaccurate during idle
> situations.
>
> > On 20/06/2025 13:07, Prashant Malani wrote:
> > > Hi Jie,
> > > On Thu, 19 Jun 2025 at 20:53, Jie Zhan <zhanjie9@...ilicon.com> wrote:
> > >> On 19/06/2025 08:09, Prashant Malani wrote:
> > >>> t0: ref=899127636, del=3012458473
> > >>> t1: ref=899129626, del=3012466509
> > >>> perf=40
> > >>
> > >> In this case, the target cpu is mostly idle but not fully idle during the
> > >> sampling window since the counter grows a little bit.
> > >> Perhaps some interrupts happen to run on the cpu shortly.
> >
> > Check back here again, I don't think it 'mostly idle'.
> > Diff of ref counters is around 2000, and I guess the ref counter freq is
> > 1GHz on your platform?  That's exactly 2us, so the target cpu is mostly
> > busy.

I think it is pertinent to note: the actual act of reading the CPPC counters
will (at least for ACPI_ADR_SPACE_FIXED_HARDWARE counters)
wake the CPU up, so even if a CPU *was* idle, the reading of the counters
calls cpc_read_ffh() [1] which does an IPI on the target CPU [2] thus waking
it up from WFI.

And that brings us back to the original assertion made in this patch:
the counter values are quite unreliable when the CPU is in this
idle (or rather I should correct that to, waking from WFI) state.

This work around probably hits more types of implementations, but
I can't see another way to limit it to only ARM FFH. Open to suggestions!

[1] https://elixir.bootlin.com/linux/v6.15.4/source/arch/arm64/kernel/topology.c#L482
[2] https://elixir.bootlin.com/linux/v6.15.4/source/arch/arm64/kernel/topology.c#L453

Best regards,

-Prashant


-- 
-Prashant