diff mbox series

[v2,2/2] cpufreq: schedutil: consolidate capacity margin calculation

Message ID 1506904212-15354-2-git-send-email-leo.yan@linaro.org
State Superseded
Headers show
Series [v2,1/2] sched/fair: make capacity_margin __read_mostly | expand

Commit Message

Leo Yan Oct. 2, 2017, 12:30 a.m. UTC
Scheduler CFS class has variable 'capacity_margin' to calculate the
capacity margin, and schedutil governor also needs to compensate the
same margin for frequency tipping point. Below are formulas used in
CFS class and schedutil governor separately:

CFS:       U` = U * capacity_margin / 1024 = U * 1.25
Schedutil: U` = U + U >> 2 = U + U * 0.25  = U * 1.25

This patch consolidates the capacity margin calculation so let
schedutil to use same formula with CFS class. As result this can avoid
the mismatch issue between schedutil and CFS class after change
'capacity_margin' to other values.

Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: Chris Redpath <Chris.Redpath@arm.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Patrick Bellasi <patrick.bellasi@arm.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Leo Yan <leo.yan@linaro.org>

---
 kernel/sched/cpufreq_schedutil.c | 6 ++++--
 kernel/sched/sched.h             | 1 +
 2 files changed, 5 insertions(+), 2 deletions(-)

-- 
2.7.4

Comments

Joel Fernandes Oct. 2, 2017, 1:02 a.m. UTC | #1
Hi Leo,

On Sun, Oct 1, 2017 at 5:30 PM, Leo Yan <leo.yan@linaro.org> wrote:
> Scheduler CFS class has variable 'capacity_margin' to calculate the


s/calculate/represent/ ?

> capacity margin, and schedutil governor also needs to compensate the

> same margin for frequency tipping point. Below are formulas used in

> CFS class and schedutil governor separately:

>

> CFS:       U` = U * capacity_margin / 1024 = U * 1.25


You should mention in the commit message, at the moment
capacity_margin is 1280 which makes U` = 1.25.

> Schedutil: U` = U + U >> 2 = U + U * 0.25  = U * 1.25

>

> This patch consolidates the capacity margin calculation so let

> schedutil to use same formula with CFS class. As result this can avoid


As a result.

> the mismatch issue between schedutil and CFS class after change

> 'capacity_margin' to other values.


This didn't make sense to me. May be you meant:

This patch consolidates the usage of the capacity margin value and
lets schedutil use the same formula as the CFS class. Thus we can
avoid the mismatch between schedutil and CFS class if
'capacity_margin' is changed to other values in the future.

> Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>

> Cc: Morten Rasmussen <morten.rasmussen@arm.com>

> Cc: Chris Redpath <Chris.Redpath@arm.com>

> Cc: Joel Fernandes <joelaf@google.com>

> Cc: Vincent Guittot <vincent.guittot@linaro.org>

> Cc: Patrick Bellasi <patrick.bellasi@arm.com>

> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

> Signed-off-by: Leo Yan <leo.yan@linaro.org>

> ---

>  kernel/sched/cpufreq_schedutil.c | 6 ++++--

>  kernel/sched/sched.h             | 1 +

>  2 files changed, 5 insertions(+), 2 deletions(-)

>

> diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c

> index 9209d83..13cc243 100644

> --- a/kernel/sched/cpufreq_schedutil.c

> +++ b/kernel/sched/cpufreq_schedutil.c

> @@ -155,7 +155,8 @@ static void sugov_update_commit(struct sugov_policy *sg_policy, u64 time,

>   *

>   * next_freq = C * curr_freq * util_raw / max

>   *

> - * Take C = 1.25 for the frequency tipping point at (util / max) = 0.8.

> + * Take C = capacity_margin / 1024 = 1.25, so it's for the frequency tipping

> + * point at (util / max) = 0.8.


The above comment assumes capacity_margin is 1280. If for any reason
the capacity_margin is changed to something else, then the comment
wont make sense anymore.

thanks,

- Joel
Leo Yan Oct. 2, 2017, 11:13 a.m. UTC | #2
Hi Joel,

On Sun, Oct 01, 2017 at 06:02:24PM -0700, Joel Fernandes wrote:
> Hi Leo,

> 

> On Sun, Oct 1, 2017 at 5:30 PM, Leo Yan <leo.yan@linaro.org> wrote:

> > Scheduler CFS class has variable 'capacity_margin' to calculate the

> 

> s/calculate/represent/ ?

> 

> > capacity margin, and schedutil governor also needs to compensate the

> > same margin for frequency tipping point. Below are formulas used in

> > CFS class and schedutil governor separately:

> >

> > CFS:       U` = U * capacity_margin / 1024 = U * 1.25

> 

> You should mention in the commit message, at the moment

> capacity_margin is 1280 which makes U` = 1.25.

> 

> > Schedutil: U` = U + U >> 2 = U + U * 0.25  = U * 1.25

> >

> > This patch consolidates the capacity margin calculation so let

> > schedutil to use same formula with CFS class. As result this can avoid

> 

> As a result.

> 

> > the mismatch issue between schedutil and CFS class after change

> > 'capacity_margin' to other values.

> 

> This didn't make sense to me. May be you meant:

> 

> This patch consolidates the usage of the capacity margin value and

> lets schedutil use the same formula as the CFS class. Thus we can

> avoid the mismatch between schedutil and CFS class if

> 'capacity_margin' is changed to other values in the future.


Yeah, thanks for the suggestion. And accept all upper comment.
Will refine for new patch.

Thanks,
Leo Yan

> > Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>

> > Cc: Morten Rasmussen <morten.rasmussen@arm.com>

> > Cc: Chris Redpath <Chris.Redpath@arm.com>

> > Cc: Joel Fernandes <joelaf@google.com>

> > Cc: Vincent Guittot <vincent.guittot@linaro.org>

> > Cc: Patrick Bellasi <patrick.bellasi@arm.com>

> > Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

> > Signed-off-by: Leo Yan <leo.yan@linaro.org>

> > ---

> >  kernel/sched/cpufreq_schedutil.c | 6 ++++--

> >  kernel/sched/sched.h             | 1 +

> >  2 files changed, 5 insertions(+), 2 deletions(-)

> >

> > diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c

> > index 9209d83..13cc243 100644

> > --- a/kernel/sched/cpufreq_schedutil.c

> > +++ b/kernel/sched/cpufreq_schedutil.c

> > @@ -155,7 +155,8 @@ static void sugov_update_commit(struct sugov_policy *sg_policy, u64 time,

> >   *

> >   * next_freq = C * curr_freq * util_raw / max

> >   *

> > - * Take C = 1.25 for the frequency tipping point at (util / max) = 0.8.

> > + * Take C = capacity_margin / 1024 = 1.25, so it's for the frequency tipping

> > + * point at (util / max) = 0.8.

> 

> The above comment assumes capacity_margin is 1280. If for any reason

> the capacity_margin is changed to something else, then the comment

> wont make sense anymore.

> 

> thanks,

> 

> - Joel
diff mbox series

Patch

diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
index 9209d83..13cc243 100644
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -155,7 +155,8 @@  static void sugov_update_commit(struct sugov_policy *sg_policy, u64 time,
  *
  * next_freq = C * curr_freq * util_raw / max
  *
- * Take C = 1.25 for the frequency tipping point at (util / max) = 0.8.
+ * Take C = capacity_margin / 1024 = 1.25, so it's for the frequency tipping
+ * point at (util / max) = 0.8.
  *
  * The lowest driver-supported frequency which is equal or greater than the raw
  * next_freq (as calculated above) is returned, subject to policy min/max and
@@ -168,7 +169,8 @@  static unsigned int get_next_freq(struct sugov_policy *sg_policy,
 	unsigned int freq = arch_scale_freq_invariant() ?
 				policy->cpuinfo.max_freq : policy->cur;
 
-	freq = (freq + (freq >> 2)) * util / max;
+	freq = freq * capacity_margin >> SCHED_CAPACITY_SHIFT;
+	freq = freq * util / max;
 
 	if (freq == sg_policy->cached_raw_freq && sg_policy->next_freq != UINT_MAX)
 		return sg_policy->next_freq;
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 14db76c..cf75bdc 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -52,6 +52,7 @@  struct cpuidle_state;
 #define TASK_ON_RQ_MIGRATING	2
 
 extern __read_mostly int scheduler_running;
+extern unsigned int capacity_margin __read_mostly;
 
 extern unsigned long calc_load_update;
 extern atomic_long_t calc_load_tasks;