lkml.org 
[lkml]   [2020]   [Jan]   [7]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
SubjectRe: [PATCH] sched/rt: Add a new sysctl to control uclamp_util_min
Hi Qais,

On Friday 20 Dec 2019 at 16:48:38 (+0000), Qais Yousef wrote:
> +/*
> + * By default RT tasks run at the maximum performance point/capacity of the
> + * system. Uclamp enforces this by always setting UCLAMP_MIN of RT tasks to
> + * SCHED_CAPACITY_SCALE.
> + *
> + * This knob allows admins to change the default behavior when uclamp is being
> + * used. In battery powered devices particularly running at the maximum
> + * capacity will increase energy consumption and shorten the battery life.
> + *
> + * This knob only affects the default value RT uses when a new RT task is
> + * forked or has just changed policy to RT and no uclamp user settings were
> + * applied (ie: the task didn't modify the default value to a new value.
> + *
> + * This knob will not override the system default values defined above.
> + */

I suppose this comment could go in the sysctl doc file instead ?

> +unsigned int sysctl_sched_rt_uclamp_util_min = SCHED_CAPACITY_SCALE;

I would suggest renaming the knob as 'sysctl_sched_rt_default_uclamp_min'
or something along those lines to make it clear it's a default value.

And for consistency with the existing code, perhaps set the default to
uclamp_none(UCLAMP_MAX) instead of an explicit SCHED_CAPACITY_SCALE?

> /* All clamps are required to be less or equal than these values */
> static struct uclamp_se uclamp_default[UCLAMP_CNT];
>
> @@ -919,6 +936,14 @@ uclamp_eff_get(struct task_struct *p, enum uclamp_id clamp_id)
> return uc_req;
> }
>
> +void uclamp_rt_sync_default_util_min(struct task_struct *p)
> +{
> + struct uclamp_se *uc_se = &p->uclamp_req[UCLAMP_MIN];
> +
> + if (!uc_se->user_defined)
> + uclamp_se_set(uc_se, sysctl_sched_rt_uclamp_util_min, false);
> +}
> +
> unsigned int uclamp_eff_value(struct task_struct *p, enum uclamp_id clamp_id)
> {
> struct uclamp_se uc_eff;
> @@ -1116,12 +1141,13 @@ int sysctl_sched_uclamp_handler(struct ctl_table *table, int write,
> loff_t *ppos)
> {
> bool update_root_tg = false;
> - int old_min, old_max;
> + int old_min, old_max, old_rt_min;
> int result;
>
> mutex_lock(&uclamp_mutex);
> old_min = sysctl_sched_uclamp_util_min;
> old_max = sysctl_sched_uclamp_util_max;
> + old_rt_min = sysctl_sched_rt_uclamp_util_min;
>
> result = proc_dointvec(table, write, buffer, lenp, ppos);
> if (result)
> @@ -1129,12 +1155,23 @@ int sysctl_sched_uclamp_handler(struct ctl_table *table, int write,
> if (!write)
> goto done;
>
> + /*
> + * The new value will be applied to all RT tasks the next time they
> + * wakeup, assuming the task is using the system default and not a user
> + * specified value. In the latter we shall leave the value as the user
> + * requested.
> + */
> if (sysctl_sched_uclamp_util_min > sysctl_sched_uclamp_util_max ||
> sysctl_sched_uclamp_util_max > SCHED_CAPACITY_SCALE) {
> result = -EINVAL;
> goto undo;
> }
>
> + if (sysctl_sched_rt_uclamp_util_min > SCHED_CAPACITY_SCALE) {
> + result = -EINVAL;
> + goto undo;
> + }
> +
> if (old_min != sysctl_sched_uclamp_util_min) {
> uclamp_se_set(&uclamp_default[UCLAMP_MIN],
> sysctl_sched_uclamp_util_min, false);
> @@ -1160,6 +1197,7 @@ int sysctl_sched_uclamp_handler(struct ctl_table *table, int write,
> undo:
> sysctl_sched_uclamp_util_min = old_min;
> sysctl_sched_uclamp_util_max = old_max;
> + sysctl_sched_rt_uclamp_util_min = old_rt_min;
> done:
> mutex_unlock(&uclamp_mutex);
>
> @@ -1202,9 +1240,12 @@ static void __setscheduler_uclamp(struct task_struct *p,
> if (uc_se->user_defined)
> continue;
>
> - /* By default, RT tasks always get 100% boost */
> + /*
> + * By default, RT tasks always get 100% boost, which the admins
> + * are allowed change via sysctl_sched_rt_uclamp_util_min knob.
> + */
> if (unlikely(rt_task(p) && clamp_id == UCLAMP_MIN))
> - clamp_value = uclamp_none(UCLAMP_MAX);
> + clamp_value = sysctl_sched_rt_uclamp_util_min;
>
> uclamp_se_set(uc_se, clamp_value, false);
> }
> @@ -1236,9 +1277,12 @@ static void uclamp_fork(struct task_struct *p)
> for_each_clamp_id(clamp_id) {
> unsigned int clamp_value = uclamp_none(clamp_id);
>
> - /* By default, RT tasks always get 100% boost */
> + /*
> + * By default, RT tasks always get 100% boost, which the admins
> + * are allowed change via sysctl_sched_rt_uclamp_util_min knob.
> + */
> if (unlikely(rt_task(p) && clamp_id == UCLAMP_MIN))
> - clamp_value = uclamp_none(UCLAMP_MAX);
> + clamp_value = sysctl_sched_rt_uclamp_util_min;
>
> uclamp_se_set(&p->uclamp_req[clamp_id], clamp_value, false);
> }
> diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
> index e591d40fd645..19572dfc175b 100644
> --- a/kernel/sched/rt.c
> +++ b/kernel/sched/rt.c
> @@ -2147,6 +2147,12 @@ static void pull_rt_task(struct rq *this_rq)
> */
> static void task_woken_rt(struct rq *rq, struct task_struct *p)
> {
> + /*
> + * When sysctl_sched_rt_uclamp_util_min value is changed by the user,
> + * we apply any new value on the next wakeup, which is here.
> + */
> + uclamp_rt_sync_default_util_min(p);

The task has already been enqueued and sugov has been called by then I
think, so this is a bit late. You could do that in uclamp_rq_inc() maybe?

> +
> if (!task_running(rq, p) &&
> !test_tsk_need_resched(rq->curr) &&
> p->nr_cpus_allowed > 1 &&
> diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
> index 280a3c735935..337bf17b1a9d 100644
> --- a/kernel/sched/sched.h
> +++ b/kernel/sched/sched.h
> @@ -2300,6 +2300,8 @@ static inline void cpufreq_update_util(struct rq *rq, unsigned int flags) {}
> #endif /* CONFIG_CPU_FREQ */
>
> #ifdef CONFIG_UCLAMP_TASK
> +void uclamp_rt_sync_default_util_min(struct task_struct *p);
> +
> unsigned int uclamp_eff_value(struct task_struct *p, enum uclamp_id clamp_id);
>
> static __always_inline
> @@ -2330,6 +2332,8 @@ static inline unsigned int uclamp_util(struct rq *rq, unsigned int util)
> return uclamp_util_with(rq, util, NULL);
> }
> #else /* CONFIG_UCLAMP_TASK */
> +void uclamp_rt_sync_default_util_min(struct task_struct *p) {}
> +
> static inline unsigned int uclamp_util_with(struct rq *rq, unsigned int util,
> struct task_struct *p)
> {
> diff --git a/kernel/sysctl.c b/kernel/sysctl.c
> index 70665934d53e..06183762daac 100644
> --- a/kernel/sysctl.c
> +++ b/kernel/sysctl.c
> @@ -465,6 +465,13 @@ static struct ctl_table kern_table[] = {
> .mode = 0644,
> .proc_handler = sysctl_sched_uclamp_handler,
> },
> + {
> + .procname = "sched_rt_util_clamp_min",
> + .data = &sysctl_sched_rt_uclamp_util_min,
> + .maxlen = sizeof(unsigned int),
> + .mode = 0644,
> + .proc_handler = sysctl_sched_uclamp_handler,
> + },
> #endif
> #ifdef CONFIG_SCHED_AUTOGROUP
> {
> --
> 2.17.1
>

Apart from the small things above, this seems like a sensible idea and
would indeed be useful, so thanks for the patch!

Quentin

\
 
 \ /
  Last update: 2020-01-07 14:43    [W:1.724 / U:0.020 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site