sched: do not keep current in the tree and get rid of sched_entity::fair_key
Get rid of 'sched_entity::fair_key'.
As a side effect, 'current' is not kept withing the tree for
SCHED_NORMAL/BATCH tasks anymore. This simplifies some parts of code
(e.g. entity_tick() and yield_task_fair()) and also somewhat optimizes
them (e.g. a single update_curr() now vs. dequeue/enqueue() before in
entity_tick()).
Signed-off-by: Dmitry Adamushko <dmitry.adamushko@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 572df1b..f776a30 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -891,7 +891,6 @@
* 6 se->load.weight
*/
struct sched_entity {
- s64 fair_key;
struct load_weight load; /* for load-balancing */
struct rb_node run_node;
unsigned int on_rq;
diff --git a/kernel/sched.c b/kernel/sched.c
index 6d18921..3b10463 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6610,7 +6610,6 @@
read_lock_irq(&tasklist_lock);
do_each_thread(g, p) {
- p->se.fair_key = 0;
p->se.exec_start = 0;
#ifdef CONFIG_SCHEDSTATS
p->se.wait_start = 0;
diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c
index e3b6232..bb34b81 100644
--- a/kernel/sched_debug.c
+++ b/kernel/sched_debug.c
@@ -38,7 +38,7 @@
SEQ_printf(m, "%15s %5d %15Ld %13Ld %5d ",
p->comm, p->pid,
- (long long)p->se.fair_key,
+ (long long)p->se.vruntime,
(long long)(p->nvcsw + p->nivcsw),
p->prio);
#ifdef CONFIG_SCHEDSTATS
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 144f3ef..b9e426a 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -139,7 +139,7 @@
static inline s64
entity_key(struct cfs_rq *cfs_rq, struct sched_entity *se)
{
- return se->fair_key - cfs_rq->min_vruntime;
+ return se->vruntime - cfs_rq->min_vruntime;
}
/*
@@ -181,9 +181,6 @@
rb_link_node(&se->run_node, parent, link);
rb_insert_color(&se->run_node, &cfs_rq->tasks_timeline);
- update_load_add(&cfs_rq->load, se->load.weight);
- cfs_rq->nr_running++;
- se->on_rq = 1;
}
static void
@@ -193,9 +190,6 @@
set_leftmost(cfs_rq, rb_next(&se->run_node));
rb_erase(&se->run_node, &cfs_rq->tasks_timeline);
- update_load_sub(&cfs_rq->load, se->load.weight);
- cfs_rq->nr_running--;
- se->on_rq = 0;
}
static inline struct rb_node *first_fair(struct cfs_rq *cfs_rq)
@@ -341,10 +335,6 @@
*/
if (se != cfs_rq->curr)
update_stats_wait_start(cfs_rq, se);
- /*
- * Update the key:
- */
- se->fair_key = se->vruntime;
}
static void
@@ -392,6 +382,22 @@
* Scheduling class queueing methods:
*/
+static void
+account_entity_enqueue(struct cfs_rq *cfs_rq, struct sched_entity *se)
+{
+ update_load_add(&cfs_rq->load, se->load.weight);
+ cfs_rq->nr_running++;
+ se->on_rq = 1;
+}
+
+static void
+account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se)
+{
+ update_load_sub(&cfs_rq->load, se->load.weight);
+ cfs_rq->nr_running--;
+ se->on_rq = 0;
+}
+
static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se)
{
#ifdef CONFIG_SCHEDSTATS
@@ -479,7 +485,9 @@
}
update_stats_enqueue(cfs_rq, se);
- __enqueue_entity(cfs_rq, se);
+ if (se != cfs_rq->curr)
+ __enqueue_entity(cfs_rq, se);
+ account_entity_enqueue(cfs_rq, se);
}
static void
@@ -498,7 +506,9 @@
}
}
#endif
- __dequeue_entity(cfs_rq, se);
+ if (se != cfs_rq->curr)
+ __dequeue_entity(cfs_rq, se);
+ account_entity_dequeue(cfs_rq, se);
}
/*
@@ -544,6 +554,10 @@
{
struct sched_entity *se = __pick_next_entity(cfs_rq);
+ /* 'current' is not kept within the tree. */
+ if (se)
+ __dequeue_entity(cfs_rq, se);
+
set_next_entity(cfs_rq, se);
return se;
@@ -560,19 +574,20 @@
update_stats_curr_end(cfs_rq, prev);
- if (prev->on_rq)
+ if (prev->on_rq) {
update_stats_wait_start(cfs_rq, prev);
+ /* Put 'current' back into the tree. */
+ __enqueue_entity(cfs_rq, prev);
+ }
cfs_rq->curr = NULL;
}
static void entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr)
{
/*
- * Dequeue and enqueue the task to update its
- * position within the tree:
+ * Update run-time statistics of the 'current'.
*/
- dequeue_entity(cfs_rq, curr, 0);
- enqueue_entity(cfs_rq, curr, 0);
+ update_curr(cfs_rq);
if (cfs_rq->nr_running > 1)
check_preempt_tick(cfs_rq, curr);
@@ -749,7 +764,7 @@
/*
* Minimally necessary key value to be last in the tree:
*/
- se->fair_key = rightmost->fair_key + 1;
+ se->vruntime = rightmost->vruntime + 1;
if (cfs_rq->rb_leftmost == &se->run_node)
cfs_rq->rb_leftmost = rb_next(&se->run_node);
@@ -965,6 +980,7 @@
update_stats_enqueue(cfs_rq, se);
__enqueue_entity(cfs_rq, se);
+ account_entity_enqueue(cfs_rq, se);
resched_task(rq->curr);
}