[PATCH RFC v2 4/4] sched: cpufreq_cfs: pelt-based cpu frequency scaling
From: Michael Turquette
Date: Mon May 11 2015 - 22:14:51 EST
Scheduler-driven cpu frequency selection is desirable as part of the
on-going effort to make the scheduler better aware of energy
consumption. No piece of the Linux kernel has a better view of the
factors that affect a cpu frequency selection policy than the
scheduler[0], and this patch is an attempt to converge on an initial
solution.
This patch implements a cpufreq governor that directly accesses
scheduler statistics, in particular per-runqueue capacity utilization
data from cfs via cfs.utilization_load_avg.
Put plainly, this governor selects the lowest cpu frequency that will
prevent a runqueue from being over-utilized (until we hit the highest
frequency of course). This is accomplished by requesting a frequency
that matches the current capacity utilization, plus a margin.
Unlike the previous posting from 2014[1] this governor implements a
"follow the utilization" method, where utilization is defined as the
frequency-invariant product of cfs.utilization_load_avg and
cpu_capacity_orig.
This governor is event-driven. There is no polling loop to check cpu
idle time nor any other method which is unsynchronized with the
scheduler. The entry points for this policy are in fair.c:
enqueue_task_fair, dequeue_task_fair and task_tick_fair.
This policy is implemented using the cpufreq governor interface for two
main reasons:
1) re-using the cpufreq machine drivers without using the governor
interface is hard.
2) using the cpufreq interface allows us to switch between the
scheduler-driven policy and legacy cpufreq governors such as ondemand at
run-time. This is very useful for comparative testing and tuning.
Finally, it is worth mentioning that this approach neglects all
scheduling classes except for cfs. It is possible to add support for
deadline and other other classes here, but I also wonder if a
multi-governor approach would be a more maintainable solution, where the
cpufreq core aggregates the constraints set by multiple governors.
Supporting such an approach in the cpufreq core would also allow for
peripheral devices to place constraint on cpu frequency without having
to hack such behavior in at the governor level.
Thanks to Juri Lelli <juri.lelli@xxxxxxx> for contributing design ideas,
code and test results.
[0] http://article.gmane.org/gmane.linux.kernel/1499836
[1] https://lkml.org/lkml/2014/10/22/22
Signed-off-by: Juri Lelli <juri.lelli@xxxxxxx>
Signed-off-by: Michael Turquette <mturquette@xxxxxxxxxx>
---
Changes in v2:
Folded in Abel's patch to fix builds for non-SMP. Thanks!
Dropped use of get_cpu_usage. Instead pass in
cfs.utilization_load_avg from fair.c
Added two additional conditions to quickly bail from _update_cpu
Return requested capacity from cpufreq_cfs_update_cpu
Handle frequency-table based systems more gooder
Internal data structures and the way data is shared with the
thread are changed considerably
Food for thought: in cpufreq_cfs_update_cpu we could break out
all of the code preceeding the call to cpufreq_cpu_get into
fair.c. The interface would change from,
unsigned long cpufreq_cfs_update_cpu(int cpu, unsigned long util);
to,
unsigned long cpufreq_cfs_update_cpu(int cpu, unsigned long cap_target);
This would give fair.c more control over the capacity it wants
to target, and makes the governor interface a bit more flexible
and useful.
drivers/cpufreq/Kconfig | 24 ++++
include/linux/cpufreq.h | 3 +
kernel/sched/Makefile | 1 +
kernel/sched/cpufreq_cfs.c | 343 +++++++++++++++++++++++++++++++++++++++++++++
kernel/sched/fair.c | 14 ++
kernel/sched/sched.h | 8 ++
6 files changed, 393 insertions(+)
create mode 100644 kernel/sched/cpufreq_cfs.c
diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig
index a171fef..83d51b4 100644
--- a/drivers/cpufreq/Kconfig
+++ b/drivers/cpufreq/Kconfig
@@ -102,6 +102,15 @@ config CPU_FREQ_DEFAULT_GOV_CONSERVATIVE
Be aware that not all cpufreq drivers support the conservative
governor. If unsure have a look at the help section of the
driver. Fallback governor will be the performance governor.
+
+config CPU_FREQ_DEFAULT_GOV_CFS
+ bool "cfs"
+ select CPU_FREQ_GOV_CFS
+ select CPU_FREQ_GOV_PERFORMANCE
+ help
+ Use the CPUfreq governor 'cfs' as default. This scales
+ cpu frequency from the scheduler as per-entity load tracking
+ statistics are updated.
endchoice
config CPU_FREQ_GOV_PERFORMANCE
@@ -183,6 +192,21 @@ config CPU_FREQ_GOV_CONSERVATIVE
If in doubt, say N.
+config CPU_FREQ_GOV_CFS
+ tristate "'cfs' cpufreq governor"
+ depends on CPU_FREQ
+ select CPU_FREQ_GOV_COMMON
+ help
+ 'cfs' - this governor scales cpu frequency from the
+ scheduler as a function of cpu capacity utilization. It does
+ not evaluate utilization on a periodic basis (as ondemand
+ does) but instead is invoked from the completely fair
+ scheduler when updating per-entity load tracking statistics.
+ Latency to respond to changes in load is improved over polling
+ governors due to its event-driven design.
+
+ If in doubt, say N.
+
comment "CPU frequency scaling drivers"
config CPUFREQ_DT
diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
index 2ee4888..62e8152 100644
--- a/include/linux/cpufreq.h
+++ b/include/linux/cpufreq.h
@@ -485,6 +485,9 @@ extern struct cpufreq_governor cpufreq_gov_ondemand;
#elif defined(CONFIG_CPU_FREQ_DEFAULT_GOV_CONSERVATIVE)
extern struct cpufreq_governor cpufreq_gov_conservative;
#define CPUFREQ_DEFAULT_GOVERNOR (&cpufreq_gov_conservative)
+#elif defined(CONFIG_CPU_FREQ_DEFAULT_GOV_CAP_GOV)
+extern struct cpufreq_governor cpufreq_gov_cap_gov;
+#define CPUFREQ_DEFAULT_GOVERNOR (&cpufreq_gov_cap_gov)
#endif
/*********************************************************************
diff --git a/kernel/sched/Makefile b/kernel/sched/Makefile
index 46be870..466960d 100644
--- a/kernel/sched/Makefile
+++ b/kernel/sched/Makefile
@@ -19,3 +19,4 @@ obj-$(CONFIG_SCHED_AUTOGROUP) += auto_group.o
obj-$(CONFIG_SCHEDSTATS) += stats.o
obj-$(CONFIG_SCHED_DEBUG) += debug.o
obj-$(CONFIG_CGROUP_CPUACCT) += cpuacct.o
+obj-$(CONFIG_CPU_FREQ_GOV_CFS) += cpufreq_cfs.o
diff --git a/kernel/sched/cpufreq_cfs.c b/kernel/sched/cpufreq_cfs.c
new file mode 100644
index 0000000..bcb63b6
--- /dev/null
+++ b/kernel/sched/cpufreq_cfs.c
@@ -0,0 +1,343 @@
+/*
+ * Copyright (C) 2015 Michael Turquette <mturquette@xxxxxxxxxx>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/cpufreq.h>
+#include <linux/module.h>
+#include <linux/kthread.h>
+#include <linux/percpu.h>
+#include <linux/irq_work.h>
+
+#include "sched.h"
+
+#define MARGIN_PCT 125 /* taken from imbalance_pct = 125 */
+#define THROTTLE_NSEC 50000000 /* 50ms default */
+
+static DEFINE_PER_CPU(unsigned long, pcpu_util);
+static DEFINE_PER_CPU(struct cpufreq_policy *, pcpu_policy);
+
+/**
+ * gov_data - per-policy data internal to the governor
+ * @throttle: next throttling period expiry. Derived from throttle_nsec
+ * @throttle_nsec: throttle period length in nanoseconds
+ * @task: worker thread for dvfs transition that may block/sleep
+ * @irq_work: callback used to wake up worker thread
+ * @freq: new frequency stored in *_cfs_update_cpu and used in *_cfs_thread
+ *
+ * struct gov_data is the per-policy cpufreq_cfs-specific data structure. A
+ * per-policy instance of it is created when the cpufreq_cfs governor receives
+ * the CPUFREQ_GOV_START condition and a pointer to it exists in the gov_data
+ * member of struct cpufreq_policy.
+ *
+ * Readers of this data must call down_read(policy->rwsem). Writers must
+ * call down_write(policy->rwsem).
+ */
+struct gov_data {
+ ktime_t throttle;
+ unsigned int throttle_nsec;
+ struct task_struct *task;
+ struct irq_work irq_work;
+ struct cpufreq_policy *policy;
+ unsigned int freq;
+};
+
+/*
+ * we pass in struct cpufreq_policy. This is safe because changing out the
+ * policy requires a call to __cpufreq_governor(policy, CPUFREQ_GOV_STOP),
+ * which tears down all of the data structures and __cpufreq_governor(policy,
+ * CPUFREQ_GOV_START) will do a full rebuild, including this kthread with the
+ * new policy pointer
+ */
+static int cpufreq_cfs_thread(void *data)
+{
+ struct sched_param param;
+ struct cpufreq_policy *policy;
+ struct gov_data *gd;
+ int ret;
+
+ policy = (struct cpufreq_policy *) data;
+ if (!policy) {
+ pr_warn("%s: missing policy\n", __func__);
+ do_exit(-EINVAL);
+ }
+
+ gd = policy->governor_data;
+ if (!gd) {
+ pr_warn("%s: missing governor data\n", __func__);
+ do_exit(-EINVAL);
+ }
+
+ param.sched_priority = 50;
+ ret = sched_setscheduler_nocheck(gd->task, SCHED_FIFO, ¶m);
+ if (ret) {
+ pr_warn("%s: failed to set SCHED_FIFO\n", __func__);
+ do_exit(-EINVAL);
+ } else {
+ pr_debug("%s: kthread (%d) set to SCHED_FIFO\n",
+ __func__, gd->task->pid);
+ }
+
+ ret = set_cpus_allowed_ptr(gd->task, policy->related_cpus);
+ if (ret) {
+ pr_warn("%s: failed to set allowed ptr\n", __func__);
+ do_exit(-EINVAL);
+ }
+
+ /* main loop of the per-policy kthread */
+ do {
+ set_current_state(TASK_INTERRUPTIBLE);
+ schedule();
+ if (kthread_should_stop())
+ break;
+
+ /* avoid race with cpufreq_cfs_stop */
+ if (!down_write_trylock(&policy->rwsem))
+ continue;
+
+ ret = __cpufreq_driver_target(policy, gd->freq,
+ CPUFREQ_RELATION_L);
+ if (ret)
+ pr_debug("%s: __cpufreq_driver_target returned %d\n",
+ __func__, ret);
+
+ gd->throttle = ktime_add_ns(ktime_get(), gd->throttle_nsec);
+ up_write(&policy->rwsem);
+ } while (!kthread_should_stop());
+
+ do_exit(0);
+}
+
+static void cpufreq_cfs_irq_work(struct irq_work *irq_work)
+{
+ struct gov_data *gd;
+
+ gd = container_of(irq_work, struct gov_data, irq_work);
+ if (!gd) {
+ return;
+ }
+
+ wake_up_process(gd->task);
+}
+
+/**
+ * cpufreq_cfs_update_cpu - interface to scheduler for changing capacity values
+ * @cpu: cpu whose capacity utilization has recently changed
+ *
+ * cpufreq_cfs_update_cpu is an interface exposed to the scheduler so that the
+ * scheduler may inform the governor of updates to capacity utilization and
+ * make changes to cpu frequency. Currently this interface is designed around
+ * PELT values in CFS. It can be expanded to other scheduling classes in the
+ * future if needed.
+ *
+ * cpufreq_cfs_update_cpu raises an IPI. The irq_work handler for that IPI wakes up
+ * the thread that does the actual work, cpufreq_cfs_thread.
+ *
+ * This functions bails out early if either condition is true:
+ * 1) this cpu is not the new maximum utilization for its frequency domain
+ * 2) no change in cpu frequency is necessary to meet the new capacity request
+ *
+ * Returns the newly chosen capacity. Note that this may not reflect reality if
+ * the hardware fails to transition to this new capacity state.
+ */
+unsigned long cpufreq_cfs_update_cpu(int cpu, unsigned long util)
+{
+ unsigned long util_new, util_old, util_max, capacity_new;
+ unsigned int freq_new, freq_tmp, cpu_tmp;
+ struct cpufreq_policy *policy;
+ struct gov_data *gd;
+ struct cpufreq_frequency_table *pos;
+
+ /* handle rounding errors */
+ util_new = util > SCHED_LOAD_SCALE ? SCHED_LOAD_SCALE : util;
+
+ /* update per-cpu utilization */
+ util_old = __this_cpu_read(pcpu_util);
+ __this_cpu_write(pcpu_util, util_new);
+
+ /* avoid locking policy for now; accessing .cpus only */
+ policy = per_cpu(pcpu_policy, cpu);
+
+ /* find max utilization of cpus in this policy */
+ util_max = 0;
+ for_each_cpu(cpu_tmp, policy->cpus)
+ util_max = max(util_max, per_cpu(pcpu_util, cpu_tmp));
+
+ /*
+ * We only change frequency if this cpu's utilization represents a new
+ * max. If another cpu has increased its utilization beyond the
+ * previous max then we rely on that cpu to hit this code path and make
+ * the change. IOW, the cpu with the new max utilization is responsible
+ * for setting the new capacity/frequency.
+ *
+ * If this cpu is not the new maximum then bail, returning the current
+ * capacity.
+ */
+ if (util_max > util_new)
+ return capacity_of(cpu);
+
+ /*
+ * We are going to request a new capacity, which might result in a new
+ * cpu frequency. From here on we need to serialize access to the
+ * policy and the governor private data.
+ */
+ policy = cpufreq_cpu_get(cpu);
+ if (IS_ERR_OR_NULL(policy)) {
+ return capacity_of(cpu);
+ }
+
+ capacity_new = capacity_of(cpu);
+ if (!policy->governor_data) {
+ goto out;
+ }
+
+ gd = policy->governor_data;
+
+ /* bail early if we are throttled */
+ if (ktime_before(ktime_get(), gd->throttle)) {
+ goto out;
+ }
+
+ /*
+ * Convert the new maximum capacity utilization into a cpu frequency
+ *
+ * It is possible to convert capacity utilization directly into a
+ * frequency, but that implies that we would be 100% utilized. Instead,
+ * first add a margin (default 25% capacity increase) to the new
+ * capacity request. This provides some head room if load increases.
+ */
+ capacity_new = util_new + (SCHED_CAPACITY_SCALE >> 2);
+ freq_new = capacity_new * policy->max >> SCHED_CAPACITY_SHIFT;
+
+ /*
+ * If a frequency table is available then find the frequency
+ * corresponding to freq_new.
+ *
+ * For cpufreq drivers without a frequency table, use the frequency
+ * directly computed from capacity_new + 25% margin.
+ */
+ if (policy->freq_table) {
+ freq_tmp = policy->max;
+ cpufreq_for_each_entry(pos, policy->freq_table) {
+ if (pos->frequency >= freq_new &&
+ pos->frequency < freq_tmp)
+ freq_tmp = pos->frequency;
+ }
+ freq_new = freq_tmp;
+ capacity_new = (freq_new << SCHED_CAPACITY_SHIFT) / policy->max;
+ }
+
+ /* No change in frequency? Bail and return current capacity. */
+ if (freq_new == policy->cur) {
+ capacity_new = capacity_of(cpu);
+ goto out;
+ }
+
+ /* store the new frequency and kick the thread */
+ gd->freq = freq_new;
+
+ /* XXX can we use something like try_to_wake_up_local here instead? */
+ irq_work_queue_on(&gd->irq_work, cpu);
+
+out:
+ cpufreq_cpu_put(policy);
+ return capacity_new;
+}
+
+static void cpufreq_cfs_start(struct cpufreq_policy *policy)
+{
+ struct gov_data *gd;
+ int cpu;
+
+ /* prepare per-policy private data */
+ gd = kzalloc(sizeof(*gd), GFP_KERNEL);
+ if (!gd) {
+ pr_debug("%s: failed to allocate private data\n", __func__);
+ return;
+ }
+
+ /* initialize per-cpu data */
+ for_each_cpu(cpu, policy->cpus) {
+ per_cpu(pcpu_util, cpu) = 0;
+ per_cpu(pcpu_policy, cpu) = policy;
+ }
+
+ /*
+ * Don't ask for freq changes at an higher rate than what
+ * the driver advertises as transition latency.
+ */
+ gd->throttle_nsec = policy->cpuinfo.transition_latency ?
+ policy->cpuinfo.transition_latency :
+ THROTTLE_NSEC;
+ pr_debug("%s: throttle threshold = %u [ns]\n",
+ __func__, gd->throttle_nsec);
+
+ /* init per-policy kthread */
+ gd->task = kthread_run(cpufreq_cfs_thread, policy, "kcpufreq_cfs_task");
+ if (IS_ERR_OR_NULL(gd->task))
+ pr_err("%s: failed to create kcpufreq_cfs_task thread\n", __func__);
+
+ init_irq_work(&gd->irq_work, cpufreq_cfs_irq_work);
+ policy->governor_data = gd;
+ gd->policy = policy;
+}
+
+static void cpufreq_cfs_stop(struct cpufreq_policy *policy)
+{
+ struct gov_data *gd;
+
+ gd = policy->governor_data;
+ kthread_stop(gd->task);
+
+ policy->governor_data = NULL;
+
+ /* FIXME replace with devm counterparts? */
+ kfree(gd);
+}
+
+static int cpufreq_cfs_setup(struct cpufreq_policy *policy, unsigned int event)
+{
+ switch (event) {
+ case CPUFREQ_GOV_START:
+ /* Start managing the frequency */
+ cpufreq_cfs_start(policy);
+ return 0;
+
+ case CPUFREQ_GOV_STOP:
+ cpufreq_cfs_stop(policy);
+ return 0;
+
+ case CPUFREQ_GOV_LIMITS: /* unused */
+ case CPUFREQ_GOV_POLICY_INIT: /* unused */
+ case CPUFREQ_GOV_POLICY_EXIT: /* unused */
+ break;
+ }
+ return 0;
+}
+
+#ifndef CONFIG_CPU_FREQ_DEFAULT_GOV_SCHED_CFS
+static
+#endif
+struct cpufreq_governor cpufreq_cfs = {
+ .name = "cfs",
+ .governor = cpufreq_cfs_setup,
+ .owner = THIS_MODULE,
+};
+
+static int __init cpufreq_cfs_init(void)
+{
+ return cpufreq_register_governor(&cpufreq_cfs);
+}
+
+static void __exit cpufreq_cfs_exit(void)
+{
+ cpufreq_unregister_governor(&cpufreq_cfs);
+}
+
+/* Try to make this the default governor */
+fs_initcall(cpufreq_cfs_init);
+
+MODULE_LICENSE("GPL");
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index d27ded9..f3c93b9 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -4257,6 +4257,11 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
update_rq_runnable_avg(rq, rq->nr_running);
add_nr_running(rq, 1);
}
+
+ if(sched_energy_freq())
+ cpufreq_cfs_update_cpu(cpu_of(rq),
+ rq->cfs.utilization_load_avg);
+
hrtick_update(rq);
}
@@ -4318,6 +4323,11 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
sub_nr_running(rq, 1);
update_rq_runnable_avg(rq, 1);
}
+
+ if(sched_energy_freq())
+ cpufreq_cfs_update_cpu(cpu_of(rq),
+ rq->cfs.utilization_load_avg);
+
hrtick_update(rq);
}
@@ -7816,6 +7826,10 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued)
task_tick_numa(rq, curr);
update_rq_runnable_avg(rq, 1);
+
+ if(sched_energy_freq())
+ cpufreq_cfs_update_cpu(cpu_of(rq),
+ rq->cfs.utilization_load_avg);
}
/*
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 4925bc4..a8585e1 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1401,6 +1401,13 @@ static inline unsigned long capacity_of(int cpu)
return cpu_rq(cpu)->cpu_capacity;
}
+#ifdef CONFIG_CPU_FREQ_GOV_CFS
+unsigned long cpufreq_cfs_update_cpu(int cpu, unsigned long util);
+#else
+static inline unsigned long cpufreq_cfs_update_cpu(int cpu, unsigned long util)
+{ }
+#endif
+
static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta)
{
rq->rt_avg += rt_delta * arch_scale_freq_capacity(NULL, cpu_of(rq));
@@ -1409,6 +1416,7 @@ static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta)
#else
static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta) { }
static inline void sched_avg_update(struct rq *rq) { }
+static inline void gov_cfs_update_cpu(int cpu) {}
#endif
extern void start_bandwidth_timer(struct hrtimer *period_timer, ktime_t period);
--
1.9.1
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/