lkml.org 
[lkml]   [2016]   [Feb]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRE: [PATCH] intel-pstate: Update frequencies of policy->cpus only from ->set_policy()
Date
Hi Kumar,
> -----Original Message-----
> From: linux-pm-owner@vger.kernel.org [mailto:linux-pm-
> owner@vger.kernel.org] On Behalf Of Viresh Kumar
> Sent: Monday, February 22, 2016 12:58 PM
> To: Rafael Wysocki; Srinivas Pandruvada; Len Brown; Viresh Kumar
> Cc: linaro-kernel@lists.linaro.org; linux-pm@vger.kernel.org; Joonas Lahtinen;
> linux-kernel@vger.kernel.org
> Subject: [PATCH] intel-pstate: Update frequencies of policy->cpus only from
> ->set_policy()
>
> The intel-pstate driver is using intel_pstate_hwp_set() from two separate
> paths, i.e. ->set_policy() callback and sysfs update path for the files present
> in /sys/devices/system/cpu/intel_pstate/ directory.
>
> While an update to the sysfs path applies to all the CPUs being managed by
> the driver (which essentially means all the online CPUs), the update via the -
> >set_policy() callback applies to a smaller group of CPUs managed by the
> policy for which ->set_policy() is called.
>
> And so, intel_pstate_hwp_set() should update frequencies of only the CPUs
> that are part of policy->cpus mask, while it is called from
> ->set_policy() callback.
>
> In order to do that, add a parameter (cpumask) to intel_pstate_hwp_set()
> and apply the frequency changes only to the concerned CPUs.
>
> For ->set_policy() path, we are only concerned about policy->cpus, and so
> policy->rwsem lock taken by the core prior to calling ->set_policy() is enough
> to take care of any races. The larger lock acquired by
> get_online_cpus() is required only for the updates to sysfs files.
>
IIRC,
1.HWP is hardwarely per-package, CPUs inside one package have one shared HWP.
2.Currently all the CPUs share the same HWP settings according to intel_pstate design.
3. The policy is per-cpu in intel_pstate driver.(policy->cpus only contains one cpu)

So with this patch applied, it is likely CPUs may have different HWP settings?
For example:
CPU 0 belongs to package A with policy 0, and CPU 1 belongs to package B with policy 1,
If you change the policy 0 from powersave to performance, then only CPU0 will update its
min/max freq in HWP, however we should also update CPU 2's min/max in HWP settings?
Plz correct me if I'm wrong..

thanks,
yu

\
 
 \ /
  Last update: 2016-02-22 11:41    [W:0.084 / U:0.516 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site