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 572df1bbaeecfcfbee36e0ab2d9aa98010f2771f..f776a30b403e7d5ceb15c05352a2ab36b7d7c01e 100644 (file)
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -891,7 +891,6 @@ struct load_weight {
  *     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 6d1892192e216424c47af594054ea9cce114c98b..3b104635a8eafb4b9ef6664e8898606595405010 100644 (file)
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6610,7 +6610,6 @@ void normalize_rt_tasks(void)
 
        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 e3b62324ac3130dce15eeeaa7b4e47a7388ed2ab..bb34b8188f61b67d69d781f099ab3ff937e2657b 100644 (file)
--- a/kernel/sched_debug.c
+++ b/kernel/sched_debug.c
@@ -38,7 +38,7 @@ print_task(struct seq_file *m, struct rq *rq, struct task_struct *p)
 
        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 144f3ef973805e83ecf24b3026452c9e9dd76777..b9e426a8a09fe57976e0cd1d9789c4ed2fb7adc3 100644 (file)
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -139,7 +139,7 @@ set_leftmost(struct cfs_rq *cfs_rq, struct rb_node *leftmost)
 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 @@ __enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se)
 
        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 @@ __dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se)
                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 @@ static void update_stats_enqueue(struct cfs_rq *cfs_rq, struct sched_entity *se)
         */
        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 @@ update_stats_curr_end(struct cfs_rq *cfs_rq, struct sched_entity *se)
  * 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 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int wakeup)
        }
 
        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 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int sleep)
                }
        }
 #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 @@ static struct sched_entity *pick_next_entity(struct cfs_rq *cfs_rq)
 {
        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 @@ static void put_prev_entity(struct cfs_rq *cfs_rq, struct sched_entity *prev)
 
        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 @@ static void yield_task_fair(struct rq *rq, struct task_struct *p)
        /*
         * 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 @@ static void task_new_fair(struct rq *rq, struct task_struct *p)
 
        update_stats_enqueue(cfs_rq, se);
        __enqueue_entity(cfs_rq, se);
+       account_entity_enqueue(cfs_rq, se);
        resched_task(rq->curr);
 }