Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 7e13b27b authored by Olav Haugan's avatar Olav Haugan
Browse files

sched: Add sysctl to enable power aware scheduling 



Add sysctl to enable energy awareness at runtime. This is useful for
performance/power tuning/measurements and debugging. In addition this
will match up with the Documentation/scheduler/sched-hmp.txt documentation.

Change-Id: I0a9185498640d66917b38bf5d55f6c59fc60ad5c
Signed-off-by: default avatarOlav Haugan <ohaugan@codeaurora.org>
parent 294b88dc

Documentation/scheduler/sched-hmp.txt

+1 −1
Original line number Diff line number Diff line
@@ -1086,7 +1086,7 @@ high-performance CPU). 


Appears at: /proc/sys/kernel/sched_enable_power_aware



Default value: 1

Default value: 0



Controls whether or not per-CPU power values are used in determining

task placement. If this is disabled, tasks are simply placed on the

include/linux/sched/sysctl.h

+1 −0
Original line number Diff line number Diff line
@@ -45,6 +45,7 @@ extern unsigned int sysctl_sched_freq_account_wait_time; 
extern unsigned int sysctl_sched_migration_fixup;

extern unsigned int sysctl_sched_heavy_task_pct;

extern unsigned int sysctl_sched_min_runtime;

extern unsigned int sysctl_sched_enable_power_aware;



#if defined(CONFIG_SCHED_FREQ_INPUT) || defined(CONFIG_SCHED_HMP)

extern unsigned int sysctl_sched_init_task_load_pct;

kernel/sched/core.c

+1 −1
Original line number Diff line number Diff line
@@ -1078,7 +1078,7 @@ static int __init set_sched_enable_power_aware(char *str) 


	get_option(&str, &enable_power_aware);



	sched_enable_power_aware = !!enable_power_aware;

	sysctl_sched_enable_power_aware = !!enable_power_aware;



	return 0;

}

kernel/sched/fair.c

+9 −8
Original line number Diff line number Diff line
@@ -1265,8 +1265,9 @@ unsigned int __read_mostly sysctl_sched_spill_nr_run = 10; 
/*

 * Control whether or not individual CPU power consumption is used to

 * guide task placement.

 * This sysctl can be set to a default value using boot command line arguments.

 */

unsigned int __read_mostly sched_enable_power_aware = 0;

unsigned int __read_mostly sysctl_sched_enable_power_aware = 0;



/*

 * This specifies the maximum percent power difference between 2
@@ -1714,7 +1715,7 @@ unsigned int power_cost_at_freq(int cpu, unsigned int freq) 
	struct cpu_pstate_pwr *costs;



	if (!per_cpu_info || !per_cpu_info[cpu].ptable ||

	    !sched_enable_power_aware)

	    !sysctl_sched_enable_power_aware)

		/* When power aware scheduling is not in use, or CPU

		 * power data is not available, just use the CPU

		 * capacity as a rough stand-in for real CPU power
@@ -1745,7 +1746,7 @@ static unsigned int power_cost(struct task_struct *p, int cpu) 
	unsigned int task_freq;

	unsigned int cur_freq = cpu_rq(cpu)->cur_freq;



	if (!sched_enable_power_aware)

	if (!sysctl_sched_enable_power_aware)

		return cpu_rq(cpu)->max_possible_capacity;



	/* calculate % of max freq needed */
@@ -2375,7 +2376,7 @@ static inline int migration_needed(struct rq *rq, struct task_struct *p) 
	if (!task_will_fit(p, cpu_of(rq)))

		return MOVE_TO_BIG_CPU;



	if (sched_enable_power_aware &&

	if (sysctl_sched_enable_power_aware &&

	    !is_task_migration_throttled(p) &&

	    is_cpu_throttling_imminent(cpu_of(rq)) &&

	    lower_power_cpu_available(p, cpu_of(rq)))
@@ -2470,7 +2471,7 @@ unsigned int cpu_temp(int cpu) 


#else	/* CONFIG_SCHED_HMP */



#define sched_enable_power_aware 0

#define sysctl_sched_enable_power_aware 0



static inline int task_will_fit(struct task_struct *p, int cpu)

{
@@ -6166,7 +6167,7 @@ static bool update_sd_pick_busiest(struct lb_env *env, 
	 * prioritize spreading work over power optimization.

	 */

	cpu = cpumask_first(sched_group_cpus(sg));

	if (sched_enable_power_aware &&

	if (sysctl_sched_enable_power_aware &&

	    !sds->busiest && (capacity(env->dst_rq) == group_rq_capacity(sg)) &&

	    sgs->sum_nr_running && (env->idle != CPU_NOT_IDLE) &&

	    power_cost_at_freq(env->dst_cpu, 0) <
@@ -6963,7 +6964,7 @@ void idle_balance(int this_cpu, struct rq *this_rq) 
	 * most power-efficient idle CPU. */

	rcu_read_lock();

	sd = rcu_dereference_check_sched_domain(this_rq->sd);

	if (sd && sched_enable_power_aware) {

	if (sd && sysctl_sched_enable_power_aware) {

		for_each_cpu(i, sched_domain_span(sd)) {

			if (i == this_cpu || idle_cpu(i)) {

				cost = power_cost_at_freq(i, 0);
@@ -7372,7 +7373,7 @@ static int select_lowest_power_cpu(struct cpumask *cpus) 
	int lowest_power_cpu = -1;

	int lowest_power = INT_MAX;



	if (sched_enable_power_aware) {

	if (sysctl_sched_enable_power_aware) {

		for_each_cpu(i, cpus) {

			cost = power_cost_at_freq(i, 0);

			if (cost < lowest_power) {

kernel/sched/sched.h

+0 −1
Original line number Diff line number Diff line
@@ -886,7 +886,6 @@ static inline void clear_reserved(int cpu) 
}



extern unsigned int sched_enable_hmp;

extern unsigned int sched_enable_power_aware;



int mostly_idle_cpu(int cpu);

extern void check_for_migration(struct rq *rq, struct task_struct *p);