[v8,6/9] sched/fair: Add sched group latency support

Message ID 20221110175009.18458-7-vincent.guittot@linaro.org
State New
Headers
Series [v8,1/9] sched/fair: fix unfairness at wakeup |

Commit Message

Vincent Guittot Nov. 10, 2022, 5:50 p.m. UTC
  Task can set its latency priority with sched_setattr(), which is then used
to set the latency offset of its sched_enity, but sched group entities
still have the default latency offset value.

Add a latency.nice field in cpu cgroup controller to set the latency
priority of the group similarly to sched_setattr(). The latency priority
is then used to set the offset of the sched_entities of the group.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
---
 Documentation/admin-guide/cgroup-v2.rst | 10 +++++
 kernel/sched/core.c                     | 52 +++++++++++++++++++++++++
 kernel/sched/fair.c                     | 33 ++++++++++++++++
 kernel/sched/sched.h                    |  4 ++
 4 files changed, 99 insertions(+)
  

Comments

Patrick Bellasi Nov. 14, 2022, 4:20 p.m. UTC | #1
Hi Vincent,

On 10-Nov 18:50, Vincent Guittot wrote:

[...]

> diff --git a/Documentation/admin-guide/cgroup-v2.rst b/Documentation/admin-guide/cgroup-v2.rst
> index be4a77baf784..a4866cd4e58c 100644
> --- a/Documentation/admin-guide/cgroup-v2.rst
> +++ b/Documentation/admin-guide/cgroup-v2.rst
> @@ -1095,6 +1095,16 @@ All time durations are in microseconds.
>          values similar to the sched_setattr(2). This maximum utilization
>          value is used to clamp the task specific maximum utilization clamp.
>  
> +  cpu.latency.nice
> +	A read-write single value file which exists on non-root
> +	cgroups.  The default is "0".
> +
> +	The nice value is in the range [-20, 19].
> +
> +	This interface file allows reading and setting latency using the
> +	same values used by sched_setattr(2). The latency_nice of a group is
> + used to limit the impact of the latency_nice of a task outside the
> + group.

This control model is not clear to me.

It does not seem matching what we have for uclamp, where the cgroup values are
used to restrict how much a task can ask or give (in terms of latency here).

in the uclamp's requested-vs-effective values model:

A) a task can "request" (or give up) latency as much as it likes

B) the cgroup in which the task is in any moment limits wthe maximum
   latency a task can "request" (or give up)

C) the system wide knob set the "request" limit for the root cgroup an any task
   not in a cgroup.

This model seems to be what we should use here too.

IOW, for each task compute an "effective" latency_nice value which is defined
starting for a task "requested" latency value and by restricting this value
based on the (B) cgroup value and the (C) system wide value.

That's what we do in uclamp_eff_get():
   https://elixir.bootlin.com/linux/v6.0/source/kernel/sched/core.c#L1484

Why such a model cannot be used for latency_nice too?
Am I missing something?


Best,
patrick
  
Vincent Guittot Nov. 14, 2022, 4:57 p.m. UTC | #2
On Mon, 14 Nov 2022 at 17:20, Patrick Bellasi
<patrick.bellasi@matbug.net> wrote:
>
> Hi Vincent,
>
> On 10-Nov 18:50, Vincent Guittot wrote:
>
> [...]
>
> > diff --git a/Documentation/admin-guide/cgroup-v2.rst b/Documentation/admin-guide/cgroup-v2.rst
> > index be4a77baf784..a4866cd4e58c 100644
> > --- a/Documentation/admin-guide/cgroup-v2.rst
> > +++ b/Documentation/admin-guide/cgroup-v2.rst
> > @@ -1095,6 +1095,16 @@ All time durations are in microseconds.
> >          values similar to the sched_setattr(2). This maximum utilization
> >          value is used to clamp the task specific maximum utilization clamp.
> >
> > +  cpu.latency.nice
> > +     A read-write single value file which exists on non-root
> > +     cgroups.  The default is "0".
> > +
> > +     The nice value is in the range [-20, 19].
> > +
> > +     This interface file allows reading and setting latency using the
> > +     same values used by sched_setattr(2). The latency_nice of a group is
> > + used to limit the impact of the latency_nice of a task outside the
> > + group.
>
> This control model is not clear to me.
>
> It does not seem matching what we have for uclamp, where the cgroup values are
> used to restrict how much a task can ask or give (in terms of latency here).
>
> in the uclamp's requested-vs-effective values model:
>
> A) a task can "request" (or give up) latency as much as it likes
>
> B) the cgroup in which the task is in any moment limits wthe maximum
>    latency a task can "request" (or give up)
>
> C) the system wide knob set the "request" limit for the root cgroup an any task
>    not in a cgroup.
>
> This model seems to be what we should use here too.
>
> IOW, for each task compute an "effective" latency_nice value which is defined
> starting for a task "requested" latency value and by restricting this value
> based on the (B) cgroup value and the (C) system wide value.
>
> That's what we do in uclamp_eff_get():
>    https://elixir.bootlin.com/linux/v6.0/source/kernel/sched/core.c#L1484
>
> Why such a model cannot be used for latency_nice too?
> Am I missing something?

Have you read the previous email thread on the subject ?

As I mentioned previously we don't need an effective latency for this
patchset because the current use of cgroup latency_nice is done at
each scheduling level just like the cgroup weight is used at each
level.

Regards,
Vincent

>
>
> Best,
> patrick
>
> --
> #include <best/regards.h>
>
> Patrick Bellasi
  

Patch

diff --git a/Documentation/admin-guide/cgroup-v2.rst b/Documentation/admin-guide/cgroup-v2.rst
index be4a77baf784..a4866cd4e58c 100644
--- a/Documentation/admin-guide/cgroup-v2.rst
+++ b/Documentation/admin-guide/cgroup-v2.rst
@@ -1095,6 +1095,16 @@  All time durations are in microseconds.
         values similar to the sched_setattr(2). This maximum utilization
         value is used to clamp the task specific maximum utilization clamp.
 
+  cpu.latency.nice
+	A read-write single value file which exists on non-root
+	cgroups.  The default is "0".
+
+	The nice value is in the range [-20, 19].
+
+	This interface file allows reading and setting latency using the
+	same values used by sched_setattr(2). The latency_nice of a group is
+       used to limit the impact of the latency_nice of a task outside the
+       group.
 
 
 Memory
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index caf54e54a74f..3f42b1f61a7e 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -10890,6 +10890,47 @@  static int cpu_idle_write_s64(struct cgroup_subsys_state *css,
 {
 	return sched_group_set_idle(css_tg(css), idle);
 }
+
+static s64 cpu_latency_nice_read_s64(struct cgroup_subsys_state *css,
+				    struct cftype *cft)
+{
+	int prio, delta, last_delta = INT_MAX;
+	s64 weight;
+
+	weight = css_tg(css)->latency_offset * NICE_LATENCY_WEIGHT_MAX;
+	weight = div_s64(weight, get_sched_latency(false));
+
+	/* Find the closest nice value to the current weight */
+	for (prio = 0; prio < ARRAY_SIZE(sched_latency_to_weight); prio++) {
+		delta = abs(sched_latency_to_weight[prio] - weight);
+		if (delta >= last_delta)
+			break;
+		last_delta = delta;
+	}
+
+	return LATENCY_TO_NICE(prio-1);
+}
+
+static int cpu_latency_nice_write_s64(struct cgroup_subsys_state *css,
+				     struct cftype *cft, s64 nice)
+{
+	s64 latency_offset;
+	long weight;
+	int idx;
+
+	if (nice < MIN_LATENCY_NICE || nice > MAX_LATENCY_NICE)
+		return -ERANGE;
+
+	idx = NICE_TO_LATENCY(nice);
+	idx = array_index_nospec(idx, LATENCY_NICE_WIDTH);
+	weight = sched_latency_to_weight[idx];
+
+	latency_offset = weight * get_sched_latency(false);
+	latency_offset = div_s64(latency_offset, NICE_LATENCY_WEIGHT_MAX);
+
+	return sched_group_set_latency(css_tg(css), latency_offset);
+}
+
 #endif
 
 static struct cftype cpu_legacy_files[] = {
@@ -10904,6 +10945,11 @@  static struct cftype cpu_legacy_files[] = {
 		.read_s64 = cpu_idle_read_s64,
 		.write_s64 = cpu_idle_write_s64,
 	},
+	{
+		.name = "latency.nice",
+		.read_s64 = cpu_latency_nice_read_s64,
+		.write_s64 = cpu_latency_nice_write_s64,
+	},
 #endif
 #ifdef CONFIG_CFS_BANDWIDTH
 	{
@@ -11121,6 +11167,12 @@  static struct cftype cpu_files[] = {
 		.read_s64 = cpu_idle_read_s64,
 		.write_s64 = cpu_idle_write_s64,
 	},
+	{
+		.name = "latency.nice",
+		.flags = CFTYPE_NOT_ON_ROOT,
+		.read_s64 = cpu_latency_nice_read_s64,
+		.write_s64 = cpu_latency_nice_write_s64,
+	},
 #endif
 #ifdef CONFIG_CFS_BANDWIDTH
 	{
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 4299d5108dc7..9583936ce30c 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -11764,6 +11764,7 @@  int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent)
 		goto err;
 
 	tg->shares = NICE_0_LOAD;
+	tg->latency_offset = 0;
 
 	init_cfs_bandwidth(tg_cfs_bandwidth(tg));
 
@@ -11862,6 +11863,9 @@  void init_tg_cfs_entry(struct task_group *tg, struct cfs_rq *cfs_rq,
 	}
 
 	se->my_q = cfs_rq;
+
+	se->latency_offset = tg->latency_offset;
+
 	/* guarantee group entities always have weight */
 	update_load_set(&se->load, NICE_0_LOAD);
 	se->parent = parent;
@@ -11992,6 +11996,35 @@  int sched_group_set_idle(struct task_group *tg, long idle)
 	return 0;
 }
 
+int sched_group_set_latency(struct task_group *tg, s64 latency)
+{
+	int i;
+
+	if (tg == &root_task_group)
+		return -EINVAL;
+
+	if (abs(latency) > sysctl_sched_latency)
+		return -EINVAL;
+
+	mutex_lock(&shares_mutex);
+
+	if (tg->latency_offset == latency) {
+		mutex_unlock(&shares_mutex);
+		return 0;
+	}
+
+	tg->latency_offset = latency;
+
+	for_each_possible_cpu(i) {
+		struct sched_entity *se = tg->se[i];
+
+		WRITE_ONCE(se->latency_offset, latency);
+	}
+
+	mutex_unlock(&shares_mutex);
+	return 0;
+}
+
 #else /* CONFIG_FAIR_GROUP_SCHED */
 
 void free_fair_sched_group(struct task_group *tg) { }
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 99f10b4dc230..95d4be4f3af6 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -407,6 +407,8 @@  struct task_group {
 
 	/* A positive value indicates that this is a SCHED_IDLE group. */
 	int			idle;
+	/* latency constraint of the group. */
+	int			latency_offset;
 
 #ifdef	CONFIG_SMP
 	/*
@@ -517,6 +519,8 @@  extern int sched_group_set_shares(struct task_group *tg, unsigned long shares);
 
 extern int sched_group_set_idle(struct task_group *tg, long idle);
 
+extern int sched_group_set_latency(struct task_group *tg, s64 latency);
+
 #ifdef CONFIG_SMP
 extern void set_task_rq_fair(struct sched_entity *se,
 			     struct cfs_rq *prev, struct cfs_rq *next);