Re: [PATCH v3 5/6] cpufreq: schedutil: relax rate-limiting while running RT/DL tasks

[Juri Lelli]

On 30/11/17 15:54, Patrick Bellasi wrote:
> On 30-Nov 14:36, Juri Lelli wrote:

[...]

> > I wonder if we would also need some way to trigger a back to back update
> > as soon as a currently running one finishes and an RT/DL task asked for
> > an update (without waiting for the next tick).
> 
> Good point, I think that would actually be an interesting
> optimization. We already discussed that in the past, but refrained by
> adding more on top of this already "substantial" set of changes.
> 
> Can we think about that once we decided about some patches of this
> series?

Definitely yes for me. One bit at a time. :)

Best,

Juri

Re: [PATCH v3 5/6] cpufreq: schedutil: relax rate-limiting while running RT/DL tasks

[Juri Lelli]

On 30/11/17 15:54, Patrick Bellasi wrote:
> On 30-Nov 14:36, Juri Lelli wrote:

[...]

> > I wonder if we would also need some way to trigger a back to back update
> > as soon as a currently running one finishes and an RT/DL task asked for
> > an update (without waiting for the next tick).
> 
> Good point, I think that would actually be an interesting
> optimization. We already discussed that in the past, but refrained by
> adding more on top of this already "substantial" set of changes.
> 
> Can we think about that once we decided about some patches of this
> series?

Definitely yes for me. One bit at a time. :)

Best,

Juri

Re: [PATCH v3 5/6] cpufreq: schedutil: relax rate-limiting while running RT/DL tasks

[Patrick Bellasi]

On 30-Nov 14:36, Juri Lelli wrote:
> Hi,
> 
> On 30/11/17 11:47, Patrick Bellasi wrote:
> > The policy in use for RT/DL tasks sets the maximum frequency when a task
> > in these classes calls for a cpufreq_update_util().  However, the
> > current implementation is still enforcing a frequency switch rate
> > limiting when these tasks are running.
> > This is potentially working against the goal to switch to the maximum OPP
> > when RT tasks are running. In certain unfortunate cases it can also happen
> > that a RT task almost completes its activation at a lower OPP.
> > 
> > This patch overrides on purpose the rate limiting configuration
> > to better serve RT/DL tasks. As long as a frequency scaling operation
> > is not in progress, a frequency switch is always authorized when
> > running in "rt_mode", i.e. the current task in a CPU belongs to the
> > RT/DL class.
> > 
> > Signed-off-by: Patrick Bellasi 
> > Reviewed-by: Dietmar Eggemann 
> > Cc: Ingo Molnar 
> > Cc: Peter Zijlstra 
> > Cc: Rafael J. Wysocki 
> > Cc: Viresh Kumar 
> > Cc: linux-kernel@vger.kernel.org
> > Cc: linux...@vger.kernel.org
> > 
> > ---
> > Changes from v2:
> > - rebased on v4.15-rc1
> > 
> > Change-Id: I733d47b9e265cebb2e3e5e71a3cd468e9be002d1
> 
> Luckily this gets ignored... :)

Indeed, and it was intended... just to verify if ML folks can tolerate
the idea to have Change-Ids in the "notes section".

This would help some internal review workflow, which sometimes are
based... yes... ehm... on gerrit :-]

> > ---
> >  kernel/sched/cpufreq_schedutil.c | 19 ---
> >  1 file changed, 12 insertions(+), 7 deletions(-)
> > 
> > diff --git a/kernel/sched/cpufreq_schedutil.c 
> > b/kernel/sched/cpufreq_schedutil.c
> > index 40521d59630b..3eea8884e61b 100644
> > --- a/kernel/sched/cpufreq_schedutil.c
> > +++ b/kernel/sched/cpufreq_schedutil.c
> > @@ -74,7 +74,8 @@ static DEFINE_PER_CPU(struct sugov_cpu, sugov_cpu);
> >  
> >  / Governor internals ***/
> >  
> > -static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 
> > time)
> > +static bool sugov_should_update_freq(struct sugov_policy *sg_policy,
> > +u64 time, bool rt_mode)
> >  {
> > s64 delta_ns;
> >  
> > @@ -111,6 +112,10 @@ static bool sugov_should_update_freq(struct 
> > sugov_policy *sg_policy, u64 time)
> > return true;
> > }
> >  
> > +   /* Always update if a RT/DL task is running */
> > +   if (rt_mode)
> > +   return true;
> > +
> > delta_ns = time - sg_policy->last_freq_update_time;
> > return delta_ns >= sg_policy->freq_update_delay_ns;
> >  }
> > @@ -268,11 +273,6 @@ static void sugov_update_single(struct 
> > update_util_data *hook, u64 time,
> > sugov_set_iowait_boost(sg_cpu, time, flags);
> > sg_cpu->last_update = time;
> >  
> > -   if (!sugov_should_update_freq(sg_policy, time))
> > -   return;
> > -
> > -   busy = sugov_cpu_is_busy(sg_cpu);
> > -
> > /*
> >  * While RT/DL tasks are running we do not want FAIR tasks to
> >  * overvrite this CPU's flags, still we can update utilization and
> > @@ -281,6 +281,11 @@ static void sugov_update_single(struct 
> > update_util_data *hook, u64 time,
> > rt_mode = task_has_dl_policy(current) ||
> >   task_has_rt_policy(current) ||
> >   (flags & SCHED_CPUFREQ_RT_DL);
> > +   if (!sugov_should_update_freq(sg_policy, time, rt_mode))
> > +   return;
> > +
> > +   busy = sugov_cpu_is_busy(sg_cpu);
> > +
> > if (rt_mode) {
> > next_f = policy->cpuinfo.max_freq;
> > } else {
> > @@ -379,7 +384,7 @@ static void sugov_update_shared(struct update_util_data 
> > *hook, u64 time,
> > sugov_set_iowait_boost(sg_cpu, time, flags);
> > sg_cpu->last_update = time;
> >  
> > -   if (sugov_should_update_freq(sg_policy, time)) {
> > +   if (sugov_should_update_freq(sg_policy, time, rt_mode)) {
> > next_f = rt_mode
> > ? sg_policy->policy->cpuinfo.max_freq
> > : sugov_next_freq_shared(sg_cpu, time);
> 
> Reviewed-by: Juri Lelli 

Tks ;-)
 
> I wonder if we would also need some way to trigger a back to back update
> as soon as a currently running one finishes and an RT/DL task asked for
> an update (without waiting for the next tick).

Good point, I think that would actually be an interesting
optimization. We already discussed that in the past, but refrained by
adding more on top of this already "substantial" set of changes.

Can we think about that once we decided about some patches of this
series?

> Best,
> 
> Juri

Cheers Patrick

-- 
#include 

Patrick Bellasi

Re: [PATCH v3 5/6] cpufreq: schedutil: relax rate-limiting while running RT/DL tasks

[Patrick Bellasi]

On 30-Nov 14:36, Juri Lelli wrote:
> Hi,
> 
> On 30/11/17 11:47, Patrick Bellasi wrote:
> > The policy in use for RT/DL tasks sets the maximum frequency when a task
> > in these classes calls for a cpufreq_update_util().  However, the
> > current implementation is still enforcing a frequency switch rate
> > limiting when these tasks are running.
> > This is potentially working against the goal to switch to the maximum OPP
> > when RT tasks are running. In certain unfortunate cases it can also happen
> > that a RT task almost completes its activation at a lower OPP.
> > 
> > This patch overrides on purpose the rate limiting configuration
> > to better serve RT/DL tasks. As long as a frequency scaling operation
> > is not in progress, a frequency switch is always authorized when
> > running in "rt_mode", i.e. the current task in a CPU belongs to the
> > RT/DL class.
> > 
> > Signed-off-by: Patrick Bellasi 
> > Reviewed-by: Dietmar Eggemann 
> > Cc: Ingo Molnar 
> > Cc: Peter Zijlstra 
> > Cc: Rafael J. Wysocki 
> > Cc: Viresh Kumar 
> > Cc: linux-kernel@vger.kernel.org
> > Cc: linux...@vger.kernel.org
> > 
> > ---
> > Changes from v2:
> > - rebased on v4.15-rc1
> > 
> > Change-Id: I733d47b9e265cebb2e3e5e71a3cd468e9be002d1
> 
> Luckily this gets ignored... :)

Indeed, and it was intended... just to verify if ML folks can tolerate
the idea to have Change-Ids in the "notes section".

This would help some internal review workflow, which sometimes are
based... yes... ehm... on gerrit :-]

> > ---
> >  kernel/sched/cpufreq_schedutil.c | 19 ---
> >  1 file changed, 12 insertions(+), 7 deletions(-)
> > 
> > diff --git a/kernel/sched/cpufreq_schedutil.c 
> > b/kernel/sched/cpufreq_schedutil.c
> > index 40521d59630b..3eea8884e61b 100644
> > --- a/kernel/sched/cpufreq_schedutil.c
> > +++ b/kernel/sched/cpufreq_schedutil.c
> > @@ -74,7 +74,8 @@ static DEFINE_PER_CPU(struct sugov_cpu, sugov_cpu);
> >  
> >  / Governor internals ***/
> >  
> > -static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 
> > time)
> > +static bool sugov_should_update_freq(struct sugov_policy *sg_policy,
> > +u64 time, bool rt_mode)
> >  {
> > s64 delta_ns;
> >  
> > @@ -111,6 +112,10 @@ static bool sugov_should_update_freq(struct 
> > sugov_policy *sg_policy, u64 time)
> > return true;
> > }
> >  
> > +   /* Always update if a RT/DL task is running */
> > +   if (rt_mode)
> > +   return true;
> > +
> > delta_ns = time - sg_policy->last_freq_update_time;
> > return delta_ns >= sg_policy->freq_update_delay_ns;
> >  }
> > @@ -268,11 +273,6 @@ static void sugov_update_single(struct 
> > update_util_data *hook, u64 time,
> > sugov_set_iowait_boost(sg_cpu, time, flags);
> > sg_cpu->last_update = time;
> >  
> > -   if (!sugov_should_update_freq(sg_policy, time))
> > -   return;
> > -
> > -   busy = sugov_cpu_is_busy(sg_cpu);
> > -
> > /*
> >  * While RT/DL tasks are running we do not want FAIR tasks to
> >  * overvrite this CPU's flags, still we can update utilization and
> > @@ -281,6 +281,11 @@ static void sugov_update_single(struct 
> > update_util_data *hook, u64 time,
> > rt_mode = task_has_dl_policy(current) ||
> >   task_has_rt_policy(current) ||
> >   (flags & SCHED_CPUFREQ_RT_DL);
> > +   if (!sugov_should_update_freq(sg_policy, time, rt_mode))
> > +   return;
> > +
> > +   busy = sugov_cpu_is_busy(sg_cpu);
> > +
> > if (rt_mode) {
> > next_f = policy->cpuinfo.max_freq;
> > } else {
> > @@ -379,7 +384,7 @@ static void sugov_update_shared(struct update_util_data 
> > *hook, u64 time,
> > sugov_set_iowait_boost(sg_cpu, time, flags);
> > sg_cpu->last_update = time;
> >  
> > -   if (sugov_should_update_freq(sg_policy, time)) {
> > +   if (sugov_should_update_freq(sg_policy, time, rt_mode)) {
> > next_f = rt_mode
> > ? sg_policy->policy->cpuinfo.max_freq
> > : sugov_next_freq_shared(sg_cpu, time);
> 
> Reviewed-by: Juri Lelli 

Tks ;-)
 
> I wonder if we would also need some way to trigger a back to back update
> as soon as a currently running one finishes and an RT/DL task asked for
> an update (without waiting for the next tick).

Good point, I think that would actually be an interesting
optimization. We already discussed that in the past, but refrained by
adding more on top of this already "substantial" set of changes.

Can we think about that once we decided about some patches of this
series?

> Best,
> 
> Juri

Cheers Patrick

-- 
#include 

Patrick Bellasi

Re: [PATCH v3 5/6] cpufreq: schedutil: relax rate-limiting while running RT/DL tasks

[Juri Lelli]

Hi,

On 30/11/17 11:47, Patrick Bellasi wrote:
> The policy in use for RT/DL tasks sets the maximum frequency when a task
> in these classes calls for a cpufreq_update_util().  However, the
> current implementation is still enforcing a frequency switch rate
> limiting when these tasks are running.
> This is potentially working against the goal to switch to the maximum OPP
> when RT tasks are running. In certain unfortunate cases it can also happen
> that a RT task almost completes its activation at a lower OPP.
> 
> This patch overrides on purpose the rate limiting configuration
> to better serve RT/DL tasks. As long as a frequency scaling operation
> is not in progress, a frequency switch is always authorized when
> running in "rt_mode", i.e. the current task in a CPU belongs to the
> RT/DL class.
> 
> Signed-off-by: Patrick Bellasi 
> Reviewed-by: Dietmar Eggemann 
> Cc: Ingo Molnar 
> Cc: Peter Zijlstra 
> Cc: Rafael J. Wysocki 
> Cc: Viresh Kumar 
> Cc: linux-kernel@vger.kernel.org
> Cc: linux...@vger.kernel.org
> 
> ---
> Changes from v2:
> - rebased on v4.15-rc1
> 
> Change-Id: I733d47b9e265cebb2e3e5e71a3cd468e9be002d1

Luckily this gets ignored... :)

> ---
>  kernel/sched/cpufreq_schedutil.c | 19 ---
>  1 file changed, 12 insertions(+), 7 deletions(-)
> 
> diff --git a/kernel/sched/cpufreq_schedutil.c 
> b/kernel/sched/cpufreq_schedutil.c
> index 40521d59630b..3eea8884e61b 100644
> --- a/kernel/sched/cpufreq_schedutil.c
> +++ b/kernel/sched/cpufreq_schedutil.c
> @@ -74,7 +74,8 @@ static DEFINE_PER_CPU(struct sugov_cpu, sugov_cpu);
>  
>  / Governor internals ***/
>  
> -static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 
> time)
> +static bool sugov_should_update_freq(struct sugov_policy *sg_policy,
> +  u64 time, bool rt_mode)
>  {
>   s64 delta_ns;
>  
> @@ -111,6 +112,10 @@ static bool sugov_should_update_freq(struct sugov_policy 
> *sg_policy, u64 time)
>   return true;
>   }
>  
> + /* Always update if a RT/DL task is running */
> + if (rt_mode)
> + return true;
> +
>   delta_ns = time - sg_policy->last_freq_update_time;
>   return delta_ns >= sg_policy->freq_update_delay_ns;
>  }
> @@ -268,11 +273,6 @@ static void sugov_update_single(struct update_util_data 
> *hook, u64 time,
>   sugov_set_iowait_boost(sg_cpu, time, flags);
>   sg_cpu->last_update = time;
>  
> - if (!sugov_should_update_freq(sg_policy, time))
> - return;
> -
> - busy = sugov_cpu_is_busy(sg_cpu);
> -
>   /*
>* While RT/DL tasks are running we do not want FAIR tasks to
>* overvrite this CPU's flags, still we can update utilization and
> @@ -281,6 +281,11 @@ static void sugov_update_single(struct update_util_data 
> *hook, u64 time,
>   rt_mode = task_has_dl_policy(current) ||
> task_has_rt_policy(current) ||
> (flags & SCHED_CPUFREQ_RT_DL);
> + if (!sugov_should_update_freq(sg_policy, time, rt_mode))
> + return;
> +
> + busy = sugov_cpu_is_busy(sg_cpu);
> +
>   if (rt_mode) {
>   next_f = policy->cpuinfo.max_freq;
>   } else {
> @@ -379,7 +384,7 @@ static void sugov_update_shared(struct update_util_data 
> *hook, u64 time,
>   sugov_set_iowait_boost(sg_cpu, time, flags);
>   sg_cpu->last_update = time;
>  
> - if (sugov_should_update_freq(sg_policy, time)) {
> + if (sugov_should_update_freq(sg_policy, time, rt_mode)) {
>   next_f = rt_mode
>   ? sg_policy->policy->cpuinfo.max_freq
>   : sugov_next_freq_shared(sg_cpu, time);

Reviewed-by: Juri Lelli 

I wonder if we would also need some way to trigger a back to back update
as soon as a currently running one finishes and an RT/DL task asked for
an update (without waiting for the next tick).

Best,

Juri

Re: [PATCH v3 5/6] cpufreq: schedutil: relax rate-limiting while running RT/DL tasks

[Juri Lelli]

Hi,

On 30/11/17 11:47, Patrick Bellasi wrote:
> The policy in use for RT/DL tasks sets the maximum frequency when a task
> in these classes calls for a cpufreq_update_util().  However, the
> current implementation is still enforcing a frequency switch rate
> limiting when these tasks are running.
> This is potentially working against the goal to switch to the maximum OPP
> when RT tasks are running. In certain unfortunate cases it can also happen
> that a RT task almost completes its activation at a lower OPP.
> 
> This patch overrides on purpose the rate limiting configuration
> to better serve RT/DL tasks. As long as a frequency scaling operation
> is not in progress, a frequency switch is always authorized when
> running in "rt_mode", i.e. the current task in a CPU belongs to the
> RT/DL class.
> 
> Signed-off-by: Patrick Bellasi 
> Reviewed-by: Dietmar Eggemann 
> Cc: Ingo Molnar 
> Cc: Peter Zijlstra 
> Cc: Rafael J. Wysocki 
> Cc: Viresh Kumar 
> Cc: linux-kernel@vger.kernel.org
> Cc: linux...@vger.kernel.org
> 
> ---
> Changes from v2:
> - rebased on v4.15-rc1
> 
> Change-Id: I733d47b9e265cebb2e3e5e71a3cd468e9be002d1

Luckily this gets ignored... :)

> ---
>  kernel/sched/cpufreq_schedutil.c | 19 ---
>  1 file changed, 12 insertions(+), 7 deletions(-)
> 
> diff --git a/kernel/sched/cpufreq_schedutil.c 
> b/kernel/sched/cpufreq_schedutil.c
> index 40521d59630b..3eea8884e61b 100644
> --- a/kernel/sched/cpufreq_schedutil.c
> +++ b/kernel/sched/cpufreq_schedutil.c
> @@ -74,7 +74,8 @@ static DEFINE_PER_CPU(struct sugov_cpu, sugov_cpu);
>  
>  / Governor internals ***/
>  
> -static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 
> time)
> +static bool sugov_should_update_freq(struct sugov_policy *sg_policy,
> +  u64 time, bool rt_mode)
>  {
>   s64 delta_ns;
>  
> @@ -111,6 +112,10 @@ static bool sugov_should_update_freq(struct sugov_policy 
> *sg_policy, u64 time)
>   return true;
>   }
>  
> + /* Always update if a RT/DL task is running */
> + if (rt_mode)
> + return true;
> +
>   delta_ns = time - sg_policy->last_freq_update_time;
>   return delta_ns >= sg_policy->freq_update_delay_ns;
>  }
> @@ -268,11 +273,6 @@ static void sugov_update_single(struct update_util_data 
> *hook, u64 time,
>   sugov_set_iowait_boost(sg_cpu, time, flags);
>   sg_cpu->last_update = time;
>  
> - if (!sugov_should_update_freq(sg_policy, time))
> - return;
> -
> - busy = sugov_cpu_is_busy(sg_cpu);
> -
>   /*
>* While RT/DL tasks are running we do not want FAIR tasks to
>* overvrite this CPU's flags, still we can update utilization and
> @@ -281,6 +281,11 @@ static void sugov_update_single(struct update_util_data 
> *hook, u64 time,
>   rt_mode = task_has_dl_policy(current) ||
> task_has_rt_policy(current) ||
> (flags & SCHED_CPUFREQ_RT_DL);
> + if (!sugov_should_update_freq(sg_policy, time, rt_mode))
> + return;
> +
> + busy = sugov_cpu_is_busy(sg_cpu);
> +
>   if (rt_mode) {
>   next_f = policy->cpuinfo.max_freq;
>   } else {
> @@ -379,7 +384,7 @@ static void sugov_update_shared(struct update_util_data 
> *hook, u64 time,
>   sugov_set_iowait_boost(sg_cpu, time, flags);
>   sg_cpu->last_update = time;
>  
> - if (sugov_should_update_freq(sg_policy, time)) {
> + if (sugov_should_update_freq(sg_policy, time, rt_mode)) {
>   next_f = rt_mode
>   ? sg_policy->policy->cpuinfo.max_freq
>   : sugov_next_freq_shared(sg_cpu, time);

Reviewed-by: Juri Lelli 

I wonder if we would also need some way to trigger a back to back update
as soon as a currently running one finishes and an RT/DL task asked for
an update (without waiting for the next tick).

Best,

Juri

[PATCH v3 5/6] cpufreq: schedutil: relax rate-limiting while running RT/DL tasks

[Patrick Bellasi]

The policy in use for RT/DL tasks sets the maximum frequency when a task
in these classes calls for a cpufreq_update_util().  However, the
current implementation is still enforcing a frequency switch rate
limiting when these tasks are running.
This is potentially working against the goal to switch to the maximum OPP
when RT tasks are running. In certain unfortunate cases it can also happen
that a RT task almost completes its activation at a lower OPP.

This patch overrides on purpose the rate limiting configuration
to better serve RT/DL tasks. As long as a frequency scaling operation
is not in progress, a frequency switch is always authorized when
running in "rt_mode", i.e. the current task in a CPU belongs to the
RT/DL class.

Signed-off-by: Patrick Bellasi 
Reviewed-by: Dietmar Eggemann 
Cc: Ingo Molnar 
Cc: Peter Zijlstra 
Cc: Rafael J. Wysocki 
Cc: Viresh Kumar 
Cc: linux-kernel@vger.kernel.org
Cc: linux...@vger.kernel.org

---
Changes from v2:
- rebased on v4.15-rc1

Change-Id: I733d47b9e265cebb2e3e5e71a3cd468e9be002d1
---
 kernel/sched/cpufreq_schedutil.c | 19 ---
 1 file changed, 12 insertions(+), 7 deletions(-)

diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
index 40521d59630b..3eea8884e61b 100644
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -74,7 +74,8 @@ static DEFINE_PER_CPU(struct sugov_cpu, sugov_cpu);
 
 / Governor internals ***/
 
-static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 time)
+static bool sugov_should_update_freq(struct sugov_policy *sg_policy,
+u64 time, bool rt_mode)
 {
s64 delta_ns;
 
@@ -111,6 +112,10 @@ static bool sugov_should_update_freq(struct sugov_policy 
*sg_policy, u64 time)
return true;
}
 
+   /* Always update if a RT/DL task is running */
+   if (rt_mode)
+   return true;
+
delta_ns = time - sg_policy->last_freq_update_time;
return delta_ns >= sg_policy->freq_update_delay_ns;
 }
@@ -268,11 +273,6 @@ static void sugov_update_single(struct update_util_data 
*hook, u64 time,
sugov_set_iowait_boost(sg_cpu, time, flags);
sg_cpu->last_update = time;
 
-   if (!sugov_should_update_freq(sg_policy, time))
-   return;
-
-   busy = sugov_cpu_is_busy(sg_cpu);
-
/*
 * While RT/DL tasks are running we do not want FAIR tasks to
 * overvrite this CPU's flags, still we can update utilization and
@@ -281,6 +281,11 @@ static void sugov_update_single(struct update_util_data 
*hook, u64 time,
rt_mode = task_has_dl_policy(current) ||
  task_has_rt_policy(current) ||
  (flags & SCHED_CPUFREQ_RT_DL);
+   if (!sugov_should_update_freq(sg_policy, time, rt_mode))
+   return;
+
+   busy = sugov_cpu_is_busy(sg_cpu);
+
if (rt_mode) {
next_f = policy->cpuinfo.max_freq;
} else {
@@ -379,7 +384,7 @@ static void sugov_update_shared(struct update_util_data 
*hook, u64 time,
sugov_set_iowait_boost(sg_cpu, time, flags);
sg_cpu->last_update = time;
 
-   if (sugov_should_update_freq(sg_policy, time)) {
+   if (sugov_should_update_freq(sg_policy, time, rt_mode)) {
next_f = rt_mode
? sg_policy->policy->cpuinfo.max_freq
: sugov_next_freq_shared(sg_cpu, time);
-- 
2.14.1

[PATCH v3 5/6] cpufreq: schedutil: relax rate-limiting while running RT/DL tasks

[Patrick Bellasi]

The policy in use for RT/DL tasks sets the maximum frequency when a task
in these classes calls for a cpufreq_update_util().  However, the
current implementation is still enforcing a frequency switch rate
limiting when these tasks are running.
This is potentially working against the goal to switch to the maximum OPP
when RT tasks are running. In certain unfortunate cases it can also happen
that a RT task almost completes its activation at a lower OPP.

This patch overrides on purpose the rate limiting configuration
to better serve RT/DL tasks. As long as a frequency scaling operation
is not in progress, a frequency switch is always authorized when
running in "rt_mode", i.e. the current task in a CPU belongs to the
RT/DL class.

Signed-off-by: Patrick Bellasi 
Reviewed-by: Dietmar Eggemann 
Cc: Ingo Molnar 
Cc: Peter Zijlstra 
Cc: Rafael J. Wysocki 
Cc: Viresh Kumar 
Cc: linux-kernel@vger.kernel.org
Cc: linux...@vger.kernel.org

---
Changes from v2:
- rebased on v4.15-rc1

Change-Id: I733d47b9e265cebb2e3e5e71a3cd468e9be002d1
---
 kernel/sched/cpufreq_schedutil.c | 19 ---
 1 file changed, 12 insertions(+), 7 deletions(-)

diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
index 40521d59630b..3eea8884e61b 100644
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -74,7 +74,8 @@ static DEFINE_PER_CPU(struct sugov_cpu, sugov_cpu);
 
 / Governor internals ***/
 
-static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 time)
+static bool sugov_should_update_freq(struct sugov_policy *sg_policy,
+u64 time, bool rt_mode)
 {
s64 delta_ns;
 
@@ -111,6 +112,10 @@ static bool sugov_should_update_freq(struct sugov_policy 
*sg_policy, u64 time)
return true;
}
 
+   /* Always update if a RT/DL task is running */
+   if (rt_mode)
+   return true;
+
delta_ns = time - sg_policy->last_freq_update_time;
return delta_ns >= sg_policy->freq_update_delay_ns;
 }
@@ -268,11 +273,6 @@ static void sugov_update_single(struct update_util_data 
*hook, u64 time,
sugov_set_iowait_boost(sg_cpu, time, flags);
sg_cpu->last_update = time;
 
-   if (!sugov_should_update_freq(sg_policy, time))
-   return;
-
-   busy = sugov_cpu_is_busy(sg_cpu);
-
/*
 * While RT/DL tasks are running we do not want FAIR tasks to
 * overvrite this CPU's flags, still we can update utilization and
@@ -281,6 +281,11 @@ static void sugov_update_single(struct update_util_data 
*hook, u64 time,
rt_mode = task_has_dl_policy(current) ||
  task_has_rt_policy(current) ||
  (flags & SCHED_CPUFREQ_RT_DL);
+   if (!sugov_should_update_freq(sg_policy, time, rt_mode))
+   return;
+
+   busy = sugov_cpu_is_busy(sg_cpu);
+
if (rt_mode) {
next_f = policy->cpuinfo.max_freq;
} else {
@@ -379,7 +384,7 @@ static void sugov_update_shared(struct update_util_data 
*hook, u64 time,
sugov_set_iowait_boost(sg_cpu, time, flags);
sg_cpu->last_update = time;
 
-   if (sugov_should_update_freq(sg_policy, time)) {
+   if (sugov_should_update_freq(sg_policy, time, rt_mode)) {
next_f = rt_mode
? sg_policy->policy->cpuinfo.max_freq
: sugov_next_freq_shared(sg_cpu, time);
-- 
2.14.1