Re: [PATCH][v3] cpufreq: intel_pstate: Get percpu max freq via HWP MSR register if available

[Chen Yu]

On Tue, Jan 12, 2021 at 02:52:50PM +0100, Rafael J. Wysocki wrote:
> On Tue, Jan 12, 2021 at 6:19 AM Chen Yu <yu.c.chen@xxxxxxxxx> wrote:
> >
> > Currently when turbo is disabled(either by BIOS or by the user), the
> > intel_pstate driver reads the max non-turbo frequency from the package-wide
> > MSR_PLATFORM_INFO(0xce) register. However on asymmetric platforms it is
> > possible in theory that small and big core with HWP enabled might have
> > different max non-turbo cpu frequency, because the MSR_HWP_CAPABILITIES
> > is percpu scope according to Intel Software Developer Manual.
> >
> > The turbo max freq is already percpu basis in current code, thus make
> > similar change to the max non-turbo frequency as well.
> >
> > Reported-by: Wendy Wang <wendy.wang@xxxxxxxxx>
> > Signed-off-by: Chen Yu <yu.c.chen@xxxxxxxxx>
> > ---
> > v2: Per Srinivas' suggestion, avoid duplicated assignment of max_pstate.
> > v3: Per Rafael's suggestion, do not add new argument in intel_pstate_get_hwp_max()
> >     to avoid redundant local vars.
> >     Per Srinivas' suggestion, refined the commit log to reflect the 'non-turbo'
> >     max frequency.
> 
> Looks good now, thanks!
> 
> Is it needed in -stable and if so, which -stable series should it go into?
>
Yes, I think so, it should be 
Cc: stable@xxxxxxxxxxxxxxx # 4.18+
as the HWP reading turbo frequency was firstly introduced in v4.18-rc2 and it
was easier to be applied.
BTW, this patch is on top of your previous patch set on intel_pstate clean up:
https://patchwork.kernel.org/project/linux-pm/list/?series=410797

thanks,
Chenyu