[PATCHv2] cpufreq/schedutil: Only bind threads if needed

[PATCHv2] cpufreq/schedutil: Only bind threads if needed

[Christian Loehle]

Remove the unconditional binding of sugov kthreads to the affected CPUs
if the cpufreq driver indicates that updates can happen from any CPU.
This allows userspace to set affinities to either save power (waking up
bigger CPUs on HMP can be expensive) or increasing performance (by
letting the utilized CPUs run without preemption of the sugov kthread).

Signed-off-by: Christian Loehle <christian.loehle@arm.com>
---
- v2: Add comment for the dl_task_check_affinity return (Juri)
v1: https://lore.kernel.org/lkml/480f2140-ea59-4e1d-a68d-18cbcec10941@arm.com/

 kernel/sched/cpufreq_schedutil.c | 6 +++++-
 kernel/sched/syscalls.c          | 7 +++++++
 2 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
index c6ba15388ea7..10faab849a3e 100644
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -691,7 +691,11 @@ static int sugov_kthread_create(struct sugov_policy *sg_policy)
 	}
 
 	sg_policy->thread = thread;
-	kthread_bind_mask(thread, policy->related_cpus);
+	if (policy->dvfs_possible_from_any_cpu)
+		set_cpus_allowed_ptr(thread, policy->related_cpus);
+	else
+		kthread_bind_mask(thread, policy->related_cpus);
+
 	init_irq_work(&sg_policy->irq_work, sugov_irq_work);
 	mutex_init(&sg_policy->work_lock);
 
diff --git a/kernel/sched/syscalls.c b/kernel/sched/syscalls.c
index aa70beee9895..2ef1cb8626bc 100644
--- a/kernel/sched/syscalls.c
+++ b/kernel/sched/syscalls.c
@@ -1172,6 +1172,13 @@ int dl_task_check_affinity(struct task_struct *p, const struct cpumask *mask)
 	if (!task_has_dl_policy(p) || !dl_bandwidth_enabled())
 		return 0;
 
+	/*
+	 * The special/sugov task isn't part of regular bandwidth/admission
+	 * control so let userspace change affinities.
+	 */
+	if (dl_entity_is_special(&p->dl))
+		return 0;
+
 	/*
 	 * Since bandwidth control happens on root_domain basis,
 	 * if admission test is enabled, we only admit -deadline
-- 
2.34.1

Re: [PATCHv2] cpufreq/schedutil: Only bind threads if needed

[Viresh Kumar]

On 27-09-24, 09:59, Christian Loehle wrote:
> Remove the unconditional binding of sugov kthreads to the affected CPUs
> if the cpufreq driver indicates that updates can happen from any CPU.
> This allows userspace to set affinities to either save power (waking up
> bigger CPUs on HMP can be expensive) or increasing performance (by
> letting the utilized CPUs run without preemption of the sugov kthread).
> 
> Signed-off-by: Christian Loehle <christian.loehle@arm.com>
> ---
> - v2: Add comment for the dl_task_check_affinity return (Juri)
> v1: https://lore.kernel.org/lkml/480f2140-ea59-4e1d-a68d-18cbcec10941@arm.com/
> 
>  kernel/sched/cpufreq_schedutil.c | 6 +++++-
>  kernel/sched/syscalls.c          | 7 +++++++
>  2 files changed, 12 insertions(+), 1 deletion(-)

Acked-by: Viresh Kumar <viresh.kumar@linaro.org>

-- 
viresh

Re: [PATCHv2] cpufreq/schedutil: Only bind threads if needed

[Vincent Guittot]

On Fri, 27 Sept 2024 at 10:59, Christian Loehle
<christian.loehle@arm.com> wrote:
>
> Remove the unconditional binding of sugov kthreads to the affected CPUs
> if the cpufreq driver indicates that updates can happen from any CPU.
> This allows userspace to set affinities to either save power (waking up
> bigger CPUs on HMP can be expensive) or increasing performance (by
> letting the utilized CPUs run without preemption of the sugov kthread).
>
> Signed-off-by: Christian Loehle <christian.loehle@arm.com>

Acked-by: Vincent Guittot <vincent.guittot@linaro.org>

> ---
> - v2: Add comment for the dl_task_check_affinity return (Juri)
> v1: https://lore.kernel.org/lkml/480f2140-ea59-4e1d-a68d-18cbcec10941@arm.com/
>
>  kernel/sched/cpufreq_schedutil.c | 6 +++++-
>  kernel/sched/syscalls.c          | 7 +++++++
>  2 files changed, 12 insertions(+), 1 deletion(-)
>
> diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
> index c6ba15388ea7..10faab849a3e 100644
> --- a/kernel/sched/cpufreq_schedutil.c
> +++ b/kernel/sched/cpufreq_schedutil.c
> @@ -691,7 +691,11 @@ static int sugov_kthread_create(struct sugov_policy *sg_policy)
>         }
>
>         sg_policy->thread = thread;
> -       kthread_bind_mask(thread, policy->related_cpus);
> +       if (policy->dvfs_possible_from_any_cpu)
> +               set_cpus_allowed_ptr(thread, policy->related_cpus);
> +       else
> +               kthread_bind_mask(thread, policy->related_cpus);
> +
>         init_irq_work(&sg_policy->irq_work, sugov_irq_work);
>         mutex_init(&sg_policy->work_lock);
>
> diff --git a/kernel/sched/syscalls.c b/kernel/sched/syscalls.c
> index aa70beee9895..2ef1cb8626bc 100644
> --- a/kernel/sched/syscalls.c
> +++ b/kernel/sched/syscalls.c
> @@ -1172,6 +1172,13 @@ int dl_task_check_affinity(struct task_struct *p, const struct cpumask *mask)
>         if (!task_has_dl_policy(p) || !dl_bandwidth_enabled())
>                 return 0;
>
> +       /*
> +        * The special/sugov task isn't part of regular bandwidth/admission
> +        * control so let userspace change affinities.
> +        */
> +       if (dl_entity_is_special(&p->dl))
> +               return 0;
> +
>         /*
>          * Since bandwidth control happens on root_domain basis,
>          * if admission test is enabled, we only admit -deadline
> --
> 2.34.1

Re: [PATCHv2] cpufreq/schedutil: Only bind threads if needed

[Rafael J. Wysocki]

On Fri, Sep 27, 2024 at 10:59 AM Christian Loehle
<christian.loehle@arm.com> wrote:
>
> Remove the unconditional binding of sugov kthreads to the affected CPUs
> if the cpufreq driver indicates that updates can happen from any CPU.
> This allows userspace to set affinities to either save power (waking up
> bigger CPUs on HMP can be expensive) or increasing performance (by
> letting the utilized CPUs run without preemption of the sugov kthread).
>
> Signed-off-by: Christian Loehle <christian.loehle@arm.com>

Acked-by: Rafael J. Wysocki <rafael@kernel.org>

and I'm assuming that this will go in via tip.

> ---
> - v2: Add comment for the dl_task_check_affinity return (Juri)
> v1: https://lore.kernel.org/lkml/480f2140-ea59-4e1d-a68d-18cbcec10941@arm.com/
>
>  kernel/sched/cpufreq_schedutil.c | 6 +++++-
>  kernel/sched/syscalls.c          | 7 +++++++
>  2 files changed, 12 insertions(+), 1 deletion(-)
>
> diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
> index c6ba15388ea7..10faab849a3e 100644
> --- a/kernel/sched/cpufreq_schedutil.c
> +++ b/kernel/sched/cpufreq_schedutil.c
> @@ -691,7 +691,11 @@ static int sugov_kthread_create(struct sugov_policy *sg_policy)
>         }
>
>         sg_policy->thread = thread;
> -       kthread_bind_mask(thread, policy->related_cpus);
> +       if (policy->dvfs_possible_from_any_cpu)
> +               set_cpus_allowed_ptr(thread, policy->related_cpus);
> +       else
> +               kthread_bind_mask(thread, policy->related_cpus);
> +
>         init_irq_work(&sg_policy->irq_work, sugov_irq_work);
>         mutex_init(&sg_policy->work_lock);
>
> diff --git a/kernel/sched/syscalls.c b/kernel/sched/syscalls.c
> index aa70beee9895..2ef1cb8626bc 100644
> --- a/kernel/sched/syscalls.c
> +++ b/kernel/sched/syscalls.c
> @@ -1172,6 +1172,13 @@ int dl_task_check_affinity(struct task_struct *p, const struct cpumask *mask)
>         if (!task_has_dl_policy(p) || !dl_bandwidth_enabled())
>                 return 0;
>
> +       /*
> +        * The special/sugov task isn't part of regular bandwidth/admission
> +        * control so let userspace change affinities.
> +        */
> +       if (dl_entity_is_special(&p->dl))
> +               return 0;
> +
>         /*
>          * Since bandwidth control happens on root_domain basis,
>          * if admission test is enabled, we only admit -deadline
> --
> 2.34.1

Re: [PATCHv2] cpufreq/schedutil: Only bind threads if needed

[Christian Loehle]

On 10/1/24 19:31, Rafael J. Wysocki wrote:
> On Fri, Sep 27, 2024 at 10:59 AM Christian Loehle
> <christian.loehle@arm.com> wrote:
>>
>> Remove the unconditional binding of sugov kthreads to the affected CPUs
>> if the cpufreq driver indicates that updates can happen from any CPU.
>> This allows userspace to set affinities to either save power (waking up
>> bigger CPUs on HMP can be expensive) or increasing performance (by
>> letting the utilized CPUs run without preemption of the sugov kthread).
>>
>> Signed-off-by: Christian Loehle <christian.loehle@arm.com>
> 
> Acked-by: Rafael J. Wysocki <rafael@kernel.org>
> 
> and I'm assuming that this will go in via tip.

Peter, is that fine with you?

@Juri: I didn't add your (somewhat implied?) ACK on v2, so I'd be happy to
get it on the dl_task_check_affinity() part.


Regards,
Christian

Re: [PATCHv2] cpufreq/schedutil: Only bind threads if needed

[Christian Loehle]

On 10/7/24 11:42, Christian Loehle wrote:
> On 10/1/24 19:31, Rafael J. Wysocki wrote:
>> On Fri, Sep 27, 2024 at 10:59 AM Christian Loehle
>> <christian.loehle@arm.com> wrote:
>>>
>>> Remove the unconditional binding of sugov kthreads to the affected CPUs
>>> if the cpufreq driver indicates that updates can happen from any CPU.
>>> This allows userspace to set affinities to either save power (waking up
>>> bigger CPUs on HMP can be expensive) or increasing performance (by
>>> letting the utilized CPUs run without preemption of the sugov kthread).
>>>
>>> Signed-off-by: Christian Loehle <christian.loehle@arm.com>
>>
>> Acked-by: Rafael J. Wysocki <rafael@kernel.org>
>>
>> and I'm assuming that this will go in via tip.
> 
> Peter, is that fine with you?
> 
> @Juri: I didn't add your (somewhat implied?) ACK on v2, so I'd be happy to
> get it on the dl_task_check_affinity() part.
> 

Peter,
gentle ping on this.
Thank you.

Regards,
Christian

Re: [PATCHv2] cpufreq/schedutil: Only bind threads if needed

[Christian Loehle]

On 10/24/24 10:13, Christian Loehle wrote:
> On 10/7/24 11:42, Christian Loehle wrote:
>> On 10/1/24 19:31, Rafael J. Wysocki wrote:
>>> On Fri, Sep 27, 2024 at 10:59 AM Christian Loehle
>>> <christian.loehle@arm.com> wrote:
>>>>
>>>> Remove the unconditional binding of sugov kthreads to the affected CPUs
>>>> if the cpufreq driver indicates that updates can happen from any CPU.
>>>> This allows userspace to set affinities to either save power (waking up
>>>> bigger CPUs on HMP can be expensive) or increasing performance (by
>>>> letting the utilized CPUs run without preemption of the sugov kthread).
>>>>
>>>> Signed-off-by: Christian Loehle <christian.loehle@arm.com>
>>>
>>> Acked-by: Rafael J. Wysocki <rafael@kernel.org>
>>>
>>> and I'm assuming that this will go in via tip.
>>
>> Peter, is that fine with you?
>>
>> @Juri: I didn't add your (somewhat implied?) ACK on v2, so I'd be happy to
>> get it on the dl_task_check_affinity() part.
>>
> 
> Peter,
> gentle ping on this.
> Thank you.

Hi Peter,
Another gentle ping.

The accumulated ACKs are:

Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
Acked-by: Juri Lelli <juri.lelli@redhat.com>

Regards,
Christian

Re: [PATCHv2] cpufreq/schedutil: Only bind threads if needed

[Juri Lelli]

Hi,

On 27/09/24 09:59, Christian Loehle wrote:
> Remove the unconditional binding of sugov kthreads to the affected CPUs
> if the cpufreq driver indicates that updates can happen from any CPU.
> This allows userspace to set affinities to either save power (waking up
> bigger CPUs on HMP can be expensive) or increasing performance (by
> letting the utilized CPUs run without preemption of the sugov kthread).
> 
> Signed-off-by: Christian Loehle <christian.loehle@arm.com>
> ---
> - v2: Add comment for the dl_task_check_affinity return (Juri)
> v1: https://lore.kernel.org/lkml/480f2140-ea59-4e1d-a68d-18cbcec10941@arm.com/
> 
>  kernel/sched/cpufreq_schedutil.c | 6 +++++-
>  kernel/sched/syscalls.c          | 7 +++++++
>  2 files changed, 12 insertions(+), 1 deletion(-)
> 
> diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
> index c6ba15388ea7..10faab849a3e 100644
> --- a/kernel/sched/cpufreq_schedutil.c
> +++ b/kernel/sched/cpufreq_schedutil.c
> @@ -691,7 +691,11 @@ static int sugov_kthread_create(struct sugov_policy *sg_policy)
>  	}
>  
>  	sg_policy->thread = thread;
> -	kthread_bind_mask(thread, policy->related_cpus);
> +	if (policy->dvfs_possible_from_any_cpu)
> +		set_cpus_allowed_ptr(thread, policy->related_cpus);
> +	else
> +		kthread_bind_mask(thread, policy->related_cpus);
> +
>  	init_irq_work(&sg_policy->irq_work, sugov_irq_work);
>  	mutex_init(&sg_policy->work_lock);
>  
> diff --git a/kernel/sched/syscalls.c b/kernel/sched/syscalls.c
> index aa70beee9895..2ef1cb8626bc 100644
> --- a/kernel/sched/syscalls.c
> +++ b/kernel/sched/syscalls.c
> @@ -1172,6 +1172,13 @@ int dl_task_check_affinity(struct task_struct *p, const struct cpumask *mask)
>  	if (!task_has_dl_policy(p) || !dl_bandwidth_enabled())
>  		return 0;
>  
> +	/*
> +	 * The special/sugov task isn't part of regular bandwidth/admission
> +	 * control so let userspace change affinities.
> +	 */
> +	if (dl_entity_is_special(&p->dl))
> +		return 0;
> +
>  	/*
>  	 * Since bandwidth control happens on root_domain basis,
>  	 * if admission test is enabled, we only admit -deadline

For the DEADLINE related bits,

Acked-by: Juri Lelli <juri.lelli@redhat.com>

Thanks,
Juri