[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAKohpokFUpQyHYO017kOn-Jbt0CFZ1GuxoG3N-fenWJ_poW=4Q@mail.gmail.com>
Date: Fri, 8 Nov 2024 11:58:37 +0530
From: Viresh Kumar <viresh.kumar@...aro.org>
To: Beata Michalska <beata.michalska@....com>
Cc: linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
linux-pm@...r.kernel.org, ionela.voinescu@....com, sudeep.holla@....com,
will@...nel.org, catalin.marinas@....com, rafael@...nel.org,
sumitg@...dia.com, yang@...amperecomputing.com,
vanshikonda@...amperecomputing.com, lihuisong@...wei.com,
zhanjie9@...ilicon.com
Subject: Re: [PATCH v7 1/4] cpufreq: Introduce an optional cpuinfo_avg_freq
sysfs entry
On Mon, 4 Nov 2024 at 13:30, Beata Michalska <beata.michalska@....com> wrote:
> We need a way to discover whether the new sysfs attrib is to be enabled or not.
Ahh, this is too much trouble to make it optional then :)
> I guess I could change the signature for arch_freq_get_on_cpu to return an error
> if that functionality is not supported for given policy ?
That would be more controversial I guess as we may want an unsigned int return
type for frequency.
I think we can do two things here:
- Just enable the sysfs attribute all the time ?
- Make it optional based on some static cpufreq driver flag ?
I would just enable it for everyone I guess.
--
Viresh
Powered by blists - more mailing lists