| Message ID | 1445043101-19079-1-git-send-email-alexandra.yates@linux.intel.com (mailing list archive) |
|---|---|
| State | Not Applicable, archived |
| Headers | show |
Hi, On Sat, Oct 17, 2015 at 2:51 AM, Alexandra Yates <alexandra.yates@linux.intel.com> wrote: > From: root <root@skl-s-fedora22.jf.intel.com> > > When changing from powersave to performance governors > Intel_pstate fails to update the MSR values that reflect the > max_perf_pct to 100%. For instance: > > Governor MSR max_perf_pct > ========= ==== ============ > Powersave 2828 100% > Powersave 2028 80% > Performance 2028 100% <- > > The arrow shows the culprit. At this point the MSR should > reflect the max_perf_pct that is 100% that corresponds MSR 2828 > the maximum performance for the Performance governor. > Instead it holds back the MSR value previously set > by the Powersave governor 2028. > > Signed-off-by: root <root@skl-s-fedora22.jf.intel.com> It looks like you need to update your git configuration. More to the point, I wonder if this commit addresses the problem your seeing: http://git.kernel.org/cgit/linux/kernel/git/rafael/linux-pm.git/commit/?h=bleeding-edge&id=51443fbf3d2cde16011b994252c8004ebcd66fb0 > --- > drivers/cpufreq/intel_pstate.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c > index cde38c8..992dbb5 100644 > --- a/drivers/cpufreq/intel_pstate.c > +++ b/drivers/cpufreq/intel_pstate.c > @@ -1239,6 +1239,8 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy) > policy->max >= policy->cpuinfo.max_freq) { > pr_debug("intel_pstate: set performance\n"); > limits = &performance_limits; > + if (hwp_active) > + intel_pstate_hwp_set(); > return 0; > } > > -- Thanks, Rafael -- To unsubscribe from this list: send the line "unsubscribe linux-pm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 10/16/2015 06:15 PM, Rafael J. Wysocki wrote: > Hi, > > On Sat, Oct 17, 2015 at 2:51 AM, Alexandra Yates > <alexandra.yates@linux.intel.com> wrote: >> From: root <root@skl-s-fedora22.jf.intel.com> >> >> When changing from powersave to performance governors >> Intel_pstate fails to update the MSR values that reflect the >> max_perf_pct to 100%. For instance: >> >> Governor MSR max_perf_pct >> ========= ==== ============ >> Powersave 2828 100% >> Powersave 2028 80% >> Performance 2028 100% <- >> >> The arrow shows the culprit. At this point the MSR should >> reflect the max_perf_pct that is 100% that corresponds MSR 2828 >> the maximum performance for the Performance governor. >> Instead it holds back the MSR value previously set >> by the Powersave governor 2028. >> >> Signed-off-by: root <root@skl-s-fedora22.jf.intel.com> > Sorry about this I sent the patch against an internal tree. Won't happen again. > It looks like you need to update your git configuration. > > More to the point, I wonder if this commit addresses the problem your seeing: > I tested only with the following patch in question and it doesn't fix the error for the MSR values on SKL. I'm resending my patch rebased to linux-pm, including a better explanation on what the error is and what is fixing. > http://git.kernel.org/cgit/linux/kernel/git/rafael/linux-pm.git/commit/?h=bleeding-edge&id=51443fbf3d2cde16011b994252c8004ebcd66fb0 > >> --- >> drivers/cpufreq/intel_pstate.c | 2 ++ >> 1 file changed, 2 insertions(+) >> >> diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c >> index cde38c8..992dbb5 100644 >> --- a/drivers/cpufreq/intel_pstate.c >> +++ b/drivers/cpufreq/intel_pstate.c >> @@ -1239,6 +1239,8 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy) >> policy->max >= policy->cpuinfo.max_freq) { >> pr_debug("intel_pstate: set performance\n"); >> limits = &performance_limits; >> + if (hwp_active) >> + intel_pstate_hwp_set(); >> return 0; >> } >> >> -- > > Thanks, > Rafael >
========= ==== ============ Powersave 2828 100% Powersave 2028 80% Performance 2028 100% <- The arrow shows the culprit. At this point the MSR should reflect the max_perf_pct that is 100% that corresponds MSR 2828 the maximum performance for the Performance governor. Instead it holds back the MSR value previously set by the Powersave governor 2028. Signed-off-by: root <root@skl-s-fedora22.jf.intel.com> --- drivers/cpufreq/intel_pstate.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index cde38c8..992dbb5 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -1239,6 +1239,8 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy) policy->max >= policy->cpuinfo.max_freq) { pr_debug("intel_pstate: set performance\n"); limits = &performance_limits; + if (hwp_active) + intel_pstate_hwp_set(); return 0; }
From: root <root@skl-s-fedora22.jf.intel.com> When changing from powersave to performance governors Intel_pstate fails to update the MSR values that reflect the max_perf_pct to 100%. For instance: Governor MSR max_perf_pct