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

Commit f2121086 authored by Joonwoo Park's avatar Joonwoo Park
Browse files

sched: eliminate sched_freq_account_wait_time knob 



Kill unused scheduler knob sched_freq_account_wait_time.

Change-Id: Ib74123ebd69dfa3f86cf7335099f50c12a6e93c3
Signed-off-by: default avatarJoonwoo Park <joonwoop@codeaurora.org>
parent 08f5e5b5

Documentation/scheduler/sched-zone.txt

+16 −27
Original line number Diff line number Diff line
@@ -920,7 +920,7 @@ The above counters are resued for nt_curr_runnable_sum and 
nt_prev_runnable_sum.



"cpu demand" of a task includes its execution time and can also include its

wait time. 'sched_freq_account_wait_time' tunable controls whether task's wait

wait time. 'SCHED_FREQ_ACCOUNT_WAIT_TIME' controls whether task's wait

time is included in its 'curr_window' and 'prev_window' counters or not.



Needless to say, curr_runnable_sum counter of a cpu is derived from curr_window
@@ -934,7 +934,7 @@ PICK_NEXT_TASK 
	This represents beginning of execution for a task. Provided the task

	refers to a non-idle task, a portion of task's wait time that

	corresponds to the current window being tracked on a cpu is added to

	task's curr_window counter, provided sched_freq_account_wait_time is

	task's curr_window counter, provided SCHED_FREQ_ACCOUNT_WAIT_TIME is

	set. The same quantum is also added to cpu's curr_runnable_sum counter.

	The remaining portion, which corresponds to task's wait time in previous

	window is added to task's prev_window and cpu's prev_runnable_sum
@@ -969,7 +969,7 @@ TASK_MIGRATE 
	this event reflects actions taken under PICK_NEXT_TASK (i.e its

	wait time is added to task's curr/prev_window counters as well

	as src_cpu's curr/prev_runnable_sum counters, provided

	sched_freq_account_wait_time tunable is non-zero). After that update,

	SCHED_FREQ_ACCOUNT_WAIT_TIME is non-zero). After that update,

	src_cpu's curr_runnable_sum is reduced by task's curr_window value

	and dst_cpu's curr_runnable_sum is increased by task's curr_window

	value, provided sched_migration_fixup = 1. Similarly, src_cpu's
@@ -1113,17 +1113,7 @@ tracking mechanism maintains per task. If default values are used for 
both this and sched_ravg_window then a total of 50ms of task history

would be maintained in 5 10ms windows.



*** 7.11 sched_freq_account_wait_time



Appears at: /proc/sys/kernel/sched_freq_account_wait_time



Default value: 0



This controls whether a task's wait time is accounted in its curr_window and

prev_window attributes and thus in a cpu's curr_runnable_sum and

prev_runnable_sum counters.



*** 7.12 sched_migration_fixup

*** 7.11 sched_migration_fixup



Appears at: /proc/sys/kernel/sched_migration_fixup
@@ -1132,7 +1122,7 @@ Default value: 1 
This controls whether a cpu's busy time counters are adjusted during task

migration.



*** 7.13 sched_freq_inc_notify

*** 7.12 sched_freq_inc_notify



Appears at: /proc/sys/kernel/sched_freq_inc_notify
@@ -1144,7 +1134,7 @@ exceeds sched_freq_inc_notify, where freq_required is the frequency calculated 
by scheduler to meet current task demand. Note that sched_freq_inc_notify is

specified in kHz units.



*** 7.14 sched_freq_dec_notify

*** 7.13 sched_freq_dec_notify



Appears at: /proc/sys/kernel/sched_freq_dec_notify
@@ -1157,7 +1147,7 @@ exceeds sched_freq_dec_notify, where freq_required is the frequency calculated 
by scheduler to meet current task demand. Note that sched_freq_dec_notify is

specified in kHz units.



** 7.15 sched_heavy_task

*** 7.14 sched_heavy_task



Appears at: /proc/sys/kernel/sched_heavy_task
@@ -1169,7 +1159,7 @@ comparison. Scheduler will request a raise in cpu frequency when heavy tasks 
wakeup after at least one window of sleep, where window size is defined by

sched_ravg_window. Value 0 will disable this feature.



*** 7.16 sched_cpu_high_irqload

*** 7.15 sched_cpu_high_irqload



Appears at: /proc/sys/kernel/sched_cpu_high_irqload
@@ -1187,7 +1177,7 @@ longer eligible for placement. This will affect the task placement logic 
described above, causing the scheduler to try and steer tasks away from

the CPU.



** 7.17 cpu.upmigrate_discourage

*** 7.16 cpu.upmigrate_discourage



Default value : 0
@@ -1203,7 +1193,7 @@ overcommitted scenario. See notes on sched_spill_nr_run and sched_spill_load for 
how overcommitment threshold is defined and also notes on

'sched_upmigrate_min_nice' tunable.



*** 7.18 sched_static_cpu_pwr_cost

*** 7.17 sched_static_cpu_pwr_cost



Default value: 0
@@ -1218,7 +1208,7 @@ within a cluster and possibly have differing value between clusters as 
needed.





*** 7.19 sched_static_cluster_pwr_cost

*** 7.18 sched_static_cluster_pwr_cost



Default value: 0
@@ -1229,7 +1219,7 @@ power mode. It ignores the actual D-state that a cluster may be in and assumes 
the worst case power cost of the highest D-state. It is means of biasing task

placement away from idle clusters when necessary.



*** 7.20 sched_early_detection_duration

*** 7.19 sched_early_detection_duration



Default value: 9500000
@@ -1240,7 +1230,7 @@ tick for it to be eligible for the scheduler's early detection feature 
under scheduler boost. For more information on the feature itself please

refer to section 5.2.1.



*** 7.21 sched_restrict_cluster_spill

*** 7.20 sched_restrict_cluster_spill



Default value: 0
@@ -1259,7 +1249,7 @@ CPU across all clusters. When this tunable is enabled, the RT tasks are 
restricted to the lowest possible power cluster.





*** 7.22 sched_downmigrate

*** 7.21 sched_downmigrate



Appears at: /proc/sys/kernel/sched_downmigrate
@@ -1272,7 +1262,7 @@ its demand *in reference to the power-efficient cpu* drops less than 60% 
(sched_downmigrate).





*** 7.23 sched_small_wakee_task_load

*** 7.22 sched_small_wakee_task_load



Appears at: /proc/sys/kernel/sched_small_wakee_task_load
@@ -1284,7 +1274,7 @@ categorized as small wakee tasks. Scheduler places small wakee tasks on the 
waker's cluster.





*** 7.24 sched_big_waker_task_load

*** 7.23 sched_big_waker_task_load



Appears at: /proc/sys/kernel/sched_big_waker_task_load
@@ -1460,7 +1450,6 @@ cpus are being reset. Changes to below attributes result in such a reset: 
* sched_window_stats_policy (See Sec 2.4)

* sched_ravg_hist_size (See Sec 7.11)

* sched_migration_fixup (See Sec 7.12)

* sched_freq_account_wait_time (See Sec 7.11)



<task>-0     [004] d.h4 12700.711489: sched_reset_all_windows_stats: time_taken 1123 window_start 0 window_size 0 reason POLICY_CHANGE old_val 0 new_val 1

include/linux/sched/sysctl.h

+1 −1
Original line number Diff line number Diff line
@@ -44,7 +44,6 @@ extern unsigned int sysctl_sched_wakeup_load_threshold; 
extern unsigned int sysctl_sched_window_stats_policy;

extern unsigned int sysctl_sched_ravg_hist_size;

extern unsigned int sysctl_sched_cpu_high_irqload;

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_enable_power_aware;
@@ -75,6 +74,7 @@ extern unsigned int sysctl_sched_min_runtime; 
extern unsigned int sysctl_sched_small_task_pct;

extern unsigned int sysctl_sched_restrict_tasks_spread;

extern unsigned int sysctl_sched_account_wait_time;

extern unsigned int sysctl_sched_freq_account_wait_time;

#else

extern unsigned int sysctl_sched_select_prev_cpu_us;

extern unsigned int sysctl_sched_enable_colocation;

kernel/sched/core.c

+7 −16
Original line number Diff line number Diff line
@@ -1685,7 +1685,7 @@ struct cpu_cycle { 


/*

 * sched_window_stats_policy, sched_ravg_hist_size,

 * sched_migration_fixup, sched_freq_account_wait_time have a 'sysctl' copy

 * sched_migration_fixup have a 'sysctl' copy

 * associated with them. This is required for atomic update of those variables

 * when being modifed via sysctl interface.

 *
@@ -1726,8 +1726,7 @@ __read_mostly unsigned int sysctl_sched_new_task_windows = 5; 
static __read_mostly unsigned int sched_migration_fixup = 1;

__read_mostly unsigned int sysctl_sched_migration_fixup = 1;



static __read_mostly unsigned int sched_freq_account_wait_time;

__read_mostly unsigned int sysctl_sched_freq_account_wait_time;

#define SCHED_FREQ_ACCOUNT_WAIT_TIME 0



/*

 * For increase, send notification if
@@ -2093,11 +2092,11 @@ static int account_busy_for_cpu_time(struct rq *rq, struct task_struct *p, 
		if (rq->curr == p)

			return 1;



		return p->on_rq ? sched_freq_account_wait_time : 0;

		return p->on_rq ? SCHED_FREQ_ACCOUNT_WAIT_TIME : 0;

	}



	/* TASK_MIGRATE, PICK_NEXT_TASK left */

	return sched_freq_account_wait_time;

	return SCHED_FREQ_ACCOUNT_WAIT_TIME;

}



static inline int
@@ -2318,7 +2317,7 @@ void update_task_pred_demand(struct rq *rq, struct task_struct *p, int event) 
		return;



	if (event != PUT_PREV_TASK && event != TASK_UPDATE &&

			(!sched_freq_account_wait_time ||

			(!SCHED_FREQ_ACCOUNT_WAIT_TIME ||

			 (event != TASK_MIGRATE &&

			 event != PICK_NEXT_TASK)))

		return;
@@ -2328,7 +2327,7 @@ void update_task_pred_demand(struct rq *rq, struct task_struct *p, int event) 
	 * related groups

	 */

	if (event == TASK_UPDATE) {

		if (!p->on_rq && !sched_freq_account_wait_time)

		if (!p->on_rq && !SCHED_FREQ_ACCOUNT_WAIT_TIME)

			return;

	}
@@ -3204,7 +3203,6 @@ enum reset_reason_code { 
	POLICY_CHANGE,

	HIST_SIZE_CHANGE,

	MIGRATION_FIXUP_CHANGE,

	FREQ_ACCOUNT_WAIT_TIME_CHANGE,

	FREQ_AGGREGATE_CHANGE,

};
@@ -3213,7 +3211,7 @@ const char *sched_window_reset_reasons[] = { 
	"POLICY_CHANGE",

	"HIST_SIZE_CHANGE",

	"MIGRATION_FIXUP_CHANGE",

	"FREQ_ACCOUNT_WAIT_TIME_CHANGE"};

};



/* Called with IRQs enabled */

void reset_all_window_stats(u64 window_start, unsigned int window_size)
@@ -3287,13 +3285,6 @@ void reset_all_window_stats(u64 window_start, unsigned int window_size) 
		old = sched_migration_fixup;

		new = sysctl_sched_migration_fixup;

		sched_migration_fixup = sysctl_sched_migration_fixup;

	} else if (sched_freq_account_wait_time !=

					sysctl_sched_freq_account_wait_time) {

		reason = FREQ_ACCOUNT_WAIT_TIME_CHANGE;

		old = sched_freq_account_wait_time;

		new = sysctl_sched_freq_account_wait_time;

		sched_freq_account_wait_time =

				 sysctl_sched_freq_account_wait_time;

	} else if (sched_freq_aggregate !=

					sysctl_sched_freq_aggregate) {

		reason = FREQ_AGGREGATE_CHANGE;

kernel/sched/fair.c

+0 −3
Original line number Diff line number Diff line
@@ -3818,9 +3818,6 @@ static inline int invalid_value_freq_input(unsigned int *data) 
	if (data == &sysctl_sched_migration_fixup)

		return !(*data == 0 || *data == 1);



	if (data == &sysctl_sched_freq_account_wait_time)

		return !(*data == 0 || *data == 1);



	if (data == &sysctl_sched_freq_aggregate)

		return !(*data == 0 || *data == 1);

kernel/sysctl.c

+9 −7
Original line number Diff line number Diff line
@@ -324,13 +324,6 @@ static struct ctl_table kern_table[] = { 
		.mode           = 0644,

		.proc_handler   = sched_window_update_handler,

	},

	{

		.procname       = "sched_freq_account_wait_time",

		.data           = &sysctl_sched_freq_account_wait_time,

		.maxlen         = sizeof(unsigned int),

		.mode           = 0644,

		.proc_handler   = sched_window_update_handler,

	},

	{

		.procname       = "sched_heavy_task",

		.data           = &sysctl_sched_heavy_task_pct,
@@ -392,6 +385,15 @@ static struct ctl_table kern_table[] = { 
		.mode           = 0644,

		.proc_handler   = sched_window_update_handler,

	},

#ifdef CONFIG_SCHED_FREQ_INPUT

	{

		.procname       = "sched_freq_account_wait_time",

		.data           = &sysctl_sched_freq_account_wait_time,

		.maxlen         = sizeof(unsigned int),

		.mode           = 0644,

		.proc_handler   = sched_window_update_handler,

	},

#endif

#endif

	{

		.procname	= "sched_spill_load",