CFS-devel, performance improvements
Hi,
please find a couple of minor cleanups below (on top of sched-cfs-v2.6.23-rc6-v21-combo-3.patch):
(1)
Better placement of #ifdef CONFIG_SCHEDSTAT block in dequeue_entity().
Signed-off-by Dmitry Adamushko <[email protected]>
---
kernel/sched_fair.c +2 -2
@@ -453,8 +453,8 @@ static void dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int sleep) {update_stats_dequeue (cfs_rq, se); -if (sleep) { #ifdef CONFIG_SCHEDSTATS + if (sleep) { if (entity_is_task(se)) { struct task_struct *tsk = task_of(se);
@@ -463,8 +463,8 @@ dequeue_entity(struct cfs_rq *cfs_rq, stif (tsk->state &TASK_UNINTERRUPTIBLE ) se->block_start = rq_of(cfs_rq)->clock; } -#endif } +#endif __dequeue_entity (cfs_rq, se); }
---
(2)
unless we are very eager to keep an additional layer of abstraction, 'struct load_stat' is redundant now so let's get rid of it.
Signed-off-by Dmitry Adamushko <[email protected]>
---
kernel/sched.c +7 -11
@@ -170,10 +170,6 @@ struct rt_prio_array {struct list_head queue[MAX_RT_PRIO ]; }; -struct load_stat { -struct load_weight load; -}; -/* CFS-related fields in a runqueue */ struct cfs_rq {struct load_weight load;
@@ -232,7 +228,7 @@ struct rq {#ifdef CONFIG_NO_HZ unsigned char in_nohz_recently;#endif -struct load_stat ls;/* capture load from *all* tasks on this cpu */ +struct load_weight load;/* capture load from *all* tasks on this cpu */ unsigned long nr_load_updates;u64 nr_switches;
@@ -804,7 +800,7 @@ static int balance_tasks(struct rq *this* Update delta_exec, delta_fair fields for rq. * * delta_fair clock advances at a rate inversely proportional to -* total load (rq->ls.load.weight) on the runqueue, while +* total load (rq->load.weight) on the runqueue, while * delta_exec advances at the same rate as wall-clock (provided * cpu is not idle). *
@@ -812,17 +808,17 @@ static int balance_tasks(struct rq *this* runqueue over any given interval. This (smoothened) load is used * during load balance. * -* This function is called /before/ updating rq->ls.load +* This function is called /before/ updating rq->load * and when switching tasks. */ static inline void inc_load (struct rq *rq,const struct task_struct *p) { -update_load_add (&rq->ls.load, p->se.load.weight); +update_load_add (&rq->load, p->se.load.weight); }static inline void dec_load (struct rq *rq,const struct task_struct *p) { -update_load_sub (&rq->ls.load, p->se.load.weight); +update_load_sub (&rq->load, p->se.load.weight); }static void inc_nr_running (struct task_struct *p,struct rq *rq)
@@ -967,7 +963,7 @@ inline int task_curr(const struct task_s/* Used instead of source_load when we know the type == 0 */ unsigned long weighted_cpuload (const int cpu) { -return cpu_rq (cpu)->ls.load.weight; +return cpu_rq (cpu)->load.weight; }static inline void __set_task_cpu (struct task_struct *p,unsigned int cpu)
@@ -1933,7 +1929,7 @@ unsigned long nr_active(void)*/ static void update_cpu_load (struct rq *this_rq) { -unsigned long this_load = this_rq->ls.load.weight; +unsigned long this_load = this_rq->load.weight;int i, scale; this_rq->nr_load_updates++;
kernel/sched_debug.c +1 -1
@@ -137,7 +137,7 @@ static void print_cpu(struct seq_file *mP (nr_running);SEQ_printf (m," .%-30s: %lu\n" ,"load" , - rq->ls.load.weight); + rq->load.weight);P (nr_switches);P (nr_load_updates);P (nr_uninterruptible);
kernel/sched_fair.c +1 -1
@@ -499,7 +499,7 @@ set_next_entity(struct cfs_rq *cfs_rq, s* least twice that of our own weight (i.e. dont track it * when there are only lesser-weight tasks around): */ -if (rq_of (cfs_rq)->ls.load.weight >=2 *se->load.weight) { +if (rq_of (cfs_rq)->load.weight >=2 *se->load.weight) { se->slice_max =max (se->slice_max, se->sum_exec_runtime - se->prev_sum_exec_runtime); }
---
Ingo Molnar
AppliedQuoted 1 line · dimm
Better placement of #ifdef
CONFIG_SCHEDSTATblock indequeue_entity().
thanks, applied.
Ingo
Quoted 2 lines · dimm
unless we are very eager to keep an additional layer of abstraction, 'struct
load_stat' is redundant now so let's get rid of it.
yeah - good one, it's indeed redundant. Applied.
Ingo