Messages in this thread | | | Date | Thu, 27 Jul 2017 08:53:27 +0530 | From | Viresh Kumar <> | Subject | Re: [PATCH V4 1/3] sched: cpufreq: Allow remote cpufreq callbacks |
| |
On 26-07-17, 19:42, Rafael J. Wysocki wrote: > On Wednesday, July 26, 2017 02:52:32 PM Viresh Kumar wrote:
> > + /* Don't allow remote callbacks */ > > + if (smp_processor_id() != data->cpu) > > + return; > > You can do this check against cpu->cpu, however. > > > + /* Don't allow remote callbacks */ > > + if (smp_processor_id() != data->cpu) > > + return; > > And same here. > > > + > > if (flags & SCHED_CPUFREQ_IOWAIT) { > > cpu->iowait_boost = int_tofp(1); > > } else if (cpu->iowait_boost) { > > diff --git a/include/linux/sched/cpufreq.h b/include/linux/sched/cpufreq.h > > index d2be2ccbb372..8256a8f35f22 100644 > > --- a/include/linux/sched/cpufreq.h > > +++ b/include/linux/sched/cpufreq.h > > @@ -16,6 +16,7 @@ > > #ifdef CONFIG_CPU_FREQ > > struct update_util_data { > > void (*func)(struct update_util_data *data, u64 time, unsigned int flags); > > + unsigned int cpu; > > So it looks like you don't need this. > > schedutil doesn't need it as per patch [2/3].
Hmm, so your comments are exactly same as what Peter suggested few days back.
sugov_get_util() uses it in 2/3, as we need to know the target CPU anyway.
But I think it would be better to add the cpu variable in sugov_cpu structure instead as only schedutil needs it. I will do that change and send V5.
-- viresh
| |