/*
* Targeted preemption latency for CPU-bound tasks:
- * (default: 20ms * (1 + ilog(ncpus)), units: nanoseconds)
+ * (default: 5ms * (1 + ilog(ncpus)), units: nanoseconds)
*
* NOTE: this latency value is not the same as the concept of
* 'timeslice length' - timeslices in CFS are of variable length
* (to see the precise effective timeslice length of your workload,
* run vmstat and monitor the context-switches (cs) field)
*/
-unsigned int sysctl_sched_latency = 20000000ULL;
+unsigned int sysctl_sched_latency = 5000000ULL;
/*
* Minimal preemption granularity for CPU-bound tasks:
- * (default: 4 msec * (1 + ilog(ncpus)), units: nanoseconds)
+ * (default: 1 msec * (1 + ilog(ncpus)), units: nanoseconds)
*/
-unsigned int sysctl_sched_min_granularity = 4000000ULL;
+unsigned int sysctl_sched_min_granularity = 1000000ULL;
/*
* is kept at sysctl_sched_latency / sysctl_sched_min_granularity
static unsigned int sched_nr_latency = 5;
/*
- * After fork, child runs first. (default) If set to 0 then
+ * After fork, child runs first. If set to 0 (default) then
* parent will (try to) run first.
*/
-const_debug unsigned int sysctl_sched_child_runs_first = 1;
+unsigned int sysctl_sched_child_runs_first __read_mostly;
/*
* sys_sched_yield() compat mode
/*
* SCHED_OTHER wake-up granularity.
- * (default: 5 msec * (1 + ilog(ncpus)), units: nanoseconds)
+ * (default: 1 msec * (1 + ilog(ncpus)), units: nanoseconds)
*
* This option delays the preemption effects of decoupled workloads
* and reduces their over-scheduling. Synchronous workloads will still
* have immediate wakeup/sleep latencies.
*/
-unsigned int sysctl_sched_wakeup_granularity = 5000000UL;
+unsigned int sysctl_sched_wakeup_granularity = 1000000UL;
const_debug unsigned int sysctl_sched_migration_cost = 500000UL;
* CFS operations on generic schedulable entities:
*/
-static inline struct task_struct *task_of(struct sched_entity *se)
-{
- return container_of(se, struct task_struct, se);
-}
-
#ifdef CONFIG_FAIR_GROUP_SCHED
/* cpu runqueue to which this cfs_rq is attached */
/* An entity is a task if it doesn't "own" a runqueue */
#define entity_is_task(se) (!se->my_q)
+static inline struct task_struct *task_of(struct sched_entity *se)
+{
+#ifdef CONFIG_SCHED_DEBUG
+ WARN_ON_ONCE(!entity_is_task(se));
+#endif
+ return container_of(se, struct task_struct, se);
+}
+
/* Walk up scheduling entities hierarchy */
#define for_each_sched_entity(se) \
for (; se; se = se->parent)
}
}
-#else /* CONFIG_FAIR_GROUP_SCHED */
+#else /* !CONFIG_FAIR_GROUP_SCHED */
+
+static inline struct task_struct *task_of(struct sched_entity *se)
+{
+ return container_of(se, struct task_struct, se);
+}
static inline struct rq *rq_of(struct cfs_rq *cfs_rq)
{
return min_vruntime;
}
+static inline int entity_before(struct sched_entity *a,
+ struct sched_entity *b)
+{
+ return (s64)(a->vruntime - b->vruntime) < 0;
+}
+
static inline s64 entity_key(struct cfs_rq *cfs_rq, struct sched_entity *se)
{
return se->vruntime - cfs_rq->min_vruntime;
for_each_sched_entity(se) {
struct load_weight *load;
+ struct load_weight lw;
cfs_rq = cfs_rq_of(se);
load = &cfs_rq->load;
if (unlikely(!se->on_rq)) {
- struct load_weight lw = cfs_rq->load;
+ lw = cfs_rq->load;
update_load_add(&lw, se->load.weight);
load = &lw;
schedstat_set(se->wait_count, se->wait_count + 1);
schedstat_set(se->wait_sum, se->wait_sum +
rq_of(cfs_rq)->clock - se->wait_start);
+#ifdef CONFIG_SCHEDSTATS
+ if (entity_is_task(se)) {
+ trace_sched_stat_wait(task_of(se),
+ rq_of(cfs_rq)->clock - se->wait_start);
+ }
+#endif
schedstat_set(se->wait_start, 0);
}
static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se)
{
#ifdef CONFIG_SCHEDSTATS
+ struct task_struct *tsk = NULL;
+
+ if (entity_is_task(se))
+ tsk = task_of(se);
+
if (se->sleep_start) {
u64 delta = rq_of(cfs_rq)->clock - se->sleep_start;
- struct task_struct *tsk = task_of(se);
if ((s64)delta < 0)
delta = 0;
se->sleep_start = 0;
se->sum_sleep_runtime += delta;
- account_scheduler_latency(tsk, delta >> 10, 1);
+ if (tsk) {
+ account_scheduler_latency(tsk, delta >> 10, 1);
+ trace_sched_stat_sleep(tsk, delta);
+ }
}
if (se->block_start) {
u64 delta = rq_of(cfs_rq)->clock - se->block_start;
- struct task_struct *tsk = task_of(se);
if ((s64)delta < 0)
delta = 0;
se->block_start = 0;
se->sum_sleep_runtime += delta;
- /*
- * Blocking time is in units of nanosecs, so shift by 20 to
- * get a milliseconds-range estimation of the amount of
- * time that the task spent sleeping:
- */
- if (unlikely(prof_on == SLEEP_PROFILING)) {
+ if (tsk) {
+ if (tsk->in_iowait) {
+ se->iowait_sum += delta;
+ se->iowait_count++;
+ trace_sched_stat_iowait(tsk, delta);
+ }
- profile_hits(SLEEP_PROFILING, (void *)get_wchan(tsk),
- delta >> 20);
+ /*
+ * Blocking time is in units of nanosecs, so shift by
+ * 20 to get a milliseconds-range estimation of the
+ * amount of time that the task spent sleeping:
+ */
+ if (unlikely(prof_on == SLEEP_PROFILING)) {
+ profile_hits(SLEEP_PROFILING,
+ (void *)get_wchan(tsk),
+ delta >> 20);
+ }
+ account_scheduler_latency(tsk, delta >> 10, 0);
}
- account_scheduler_latency(tsk, delta >> 10, 0);
}
#endif
}
* all of which have the same weight.
*/
if (sched_feat(NORMALIZED_SLEEPER) &&
- task_of(se)->policy != SCHED_IDLE)
+ (!entity_is_task(se) ||
+ task_of(se)->policy != SCHED_IDLE))
thresh = calc_delta_fair(thresh, se);
vruntime -= thresh;
}
-
- /* ensure we never gain time by being placed backwards. */
- vruntime = max_vruntime(se->vruntime, vruntime);
}
+ /* ensure we never gain time by being placed backwards. */
+ vruntime = max_vruntime(se->vruntime, vruntime);
+
se->vruntime = vruntime;
}
/*
* Already in the rightmost position?
*/
- if (unlikely(!rightmost || rightmost->vruntime < se->vruntime))
+ if (unlikely(!rightmost || entity_before(rightmost, se)))
return;
/*
* search starts with cpus closest then further out as needed,
* so we always favor a closer, idle cpu.
* Domains may include CPUs that are not usable for migration,
- * hence we need to mask them out (cpu_active_mask)
+ * hence we need to mask them out (rq->rd->online)
*
* Returns the CPU we should wake onto.
*/
#if defined(ARCH_HAS_SCHED_WAKE_IDLE)
+
+#define cpu_rd_active(cpu, rq) cpumask_test_cpu(cpu, rq->rd->online)
+
static int wake_idle(int cpu, struct task_struct *p)
{
struct sched_domain *sd;
int i;
unsigned int chosen_wakeup_cpu;
int this_cpu;
+ struct rq *task_rq = task_rq(p);
/*
* At POWERSAVINGS_BALANCE_WAKEUP level, if both this_cpu and prev_cpu
for_each_domain(cpu, sd) {
if ((sd->flags & SD_WAKE_IDLE)
|| ((sd->flags & SD_WAKE_IDLE_FAR)
- && !task_hot(p, task_rq(p)->clock, sd))) {
+ && !task_hot(p, task_rq->clock, sd))) {
for_each_cpu_and(i, sched_domain_span(sd),
&p->cpus_allowed) {
- if (cpu_active(i) && idle_cpu(i)) {
+ if (cpu_rd_active(i, task_rq) && idle_cpu(i)) {
if (i != task_cpu(p)) {
schedstat_inc(p,
se.nr_wakeups_idle);
int idx, unsigned long load, unsigned long this_load,
unsigned int imbalance)
{
+ struct task_struct *curr = this_rq->curr;
+ struct task_group *tg;
unsigned long tl = this_load;
unsigned long tl_per_task;
- struct task_group *tg;
unsigned long weight;
int balanced;
if (!(this_sd->flags & SD_WAKE_AFFINE) || !sched_feat(AFFINE_WAKEUPS))
return 0;
+ if (sync && (curr->se.avg_overlap > sysctl_sched_migration_cost ||
+ p->se.avg_overlap > sysctl_sched_migration_cost))
+ sync = 0;
+
/*
* If sync wakeup then subtract the (maximum possible)
* effect of the currently running task from the load
tg = task_group(p);
weight = p->se.load.weight;
- balanced = 100*(tl + effective_load(tg, this_cpu, weight, weight)) <=
+ /*
+ * In low-load situations, where prev_cpu is idle and this_cpu is idle
+ * due to the sync cause above having dropped tl to 0, we'll always have
+ * an imbalance, but there's really nothing you can do about that, so
+ * that's good too.
+ *
+ * Otherwise check if either cpus are near enough in load to allow this
+ * task to be woken on this_cpu.
+ */
+ balanced = !tl ||
+ 100*(tl + effective_load(tg, this_cpu, weight, weight)) <=
imbalance*(load + effective_load(tg, prev_cpu, 0, weight));
/*
this_rq = cpu_rq(this_cpu);
new_cpu = prev_cpu;
- if (prev_cpu == this_cpu)
- goto out;
/*
* 'this_sd' is the first domain that both
* this_cpu and prev_cpu are present in:
}
#endif /* CONFIG_SMP */
-static unsigned long wakeup_gran(struct sched_entity *se)
+/*
+ * Adaptive granularity
+ *
+ * se->avg_wakeup gives the average time a task runs until it does a wakeup,
+ * with the limit of wakeup_gran -- when it never does a wakeup.
+ *
+ * So the smaller avg_wakeup is the faster we want this task to preempt,
+ * but we don't want to treat the preemptee unfairly and therefore allow it
+ * to run for at least the amount of time we'd like to run.
+ *
+ * NOTE: we use 2*avg_wakeup to increase the probability of actually doing one
+ *
+ * NOTE: we use *nr_running to scale with load, this nicely matches the
+ * degrading latency on load.
+ */
+static unsigned long
+adaptive_gran(struct sched_entity *curr, struct sched_entity *se)
+{
+ u64 this_run = curr->sum_exec_runtime - curr->prev_sum_exec_runtime;
+ u64 expected_wakeup = 2*se->avg_wakeup * cfs_rq_of(se)->nr_running;
+ u64 gran = 0;
+
+ if (this_run < expected_wakeup)
+ gran = expected_wakeup - this_run;
+
+ return min_t(s64, gran, sysctl_sched_wakeup_granularity);
+}
+
+static unsigned long
+wakeup_gran(struct sched_entity *curr, struct sched_entity *se)
{
unsigned long gran = sysctl_sched_wakeup_granularity;
+ if (cfs_rq_of(curr)->curr && sched_feat(ADAPTIVE_GRAN))
+ gran = adaptive_gran(curr, se);
+
/*
- * More easily preempt - nice tasks, while not making it harder for
- * + nice tasks.
+ * Since its curr running now, convert the gran from real-time
+ * to virtual-time in his units.
*/
- if (!sched_feat(ASYM_GRAN) || se->load.weight > NICE_0_LOAD)
- gran = calc_delta_fair(sysctl_sched_wakeup_granularity, se);
+ if (sched_feat(ASYM_GRAN)) {
+ /*
+ * By using 'se' instead of 'curr' we penalize light tasks, so
+ * they get preempted easier. That is, if 'se' < 'curr' then
+ * the resulting gran will be larger, therefore penalizing the
+ * lighter, if otoh 'se' > 'curr' then the resulting gran will
+ * be smaller, again penalizing the lighter task.
+ *
+ * This is especially important for buddies when the leftmost
+ * task is higher priority than the buddy.
+ */
+ if (unlikely(se->load.weight != NICE_0_LOAD))
+ gran = calc_delta_fair(gran, se);
+ } else {
+ if (unlikely(curr->load.weight != NICE_0_LOAD))
+ gran = calc_delta_fair(gran, curr);
+ }
return gran;
}
if (vdiff <= 0)
return -1;
- gran = wakeup_gran(curr);
+ gran = wakeup_gran(curr, se);
if (vdiff > gran)
return 1;
*/
if (sched_feat(LAST_BUDDY) && likely(se->on_rq && curr != rq->idle))
set_last_buddy(se);
- set_next_buddy(pse);
+ if (sched_feat(NEXT_BUDDY))
+ set_next_buddy(pse);
/*
* We can come here with TIF_NEED_RESCHED already set from new task
if (!sched_feat(WAKEUP_PREEMPT))
return;
- if (sched_feat(WAKEUP_OVERLAP) && sync) {
+ if ((sched_feat(WAKEUP_SYNC) && sync) ||
+ (sched_feat(WAKEUP_OVERLAP) &&
+ (se->avg_overlap < sysctl_sched_migration_cost &&
+ pse->avg_overlap < sysctl_sched_migration_cost))) {
resched_task(curr);
return;
}
find_matching_se(&se, &pse);
- while (se) {
- BUG_ON(!pse);
-
- if (wakeup_preempt_entity(se, pse) == 1) {
- resched_task(curr);
- break;
- }
+ BUG_ON(!pse);
- se = parent_entity(se);
- pse = parent_entity(pse);
- }
+ if (wakeup_preempt_entity(se, pse) == 1)
+ resched_task(curr);
}
static struct task_struct *pick_next_task_fair(struct rq *rq)
sched_info_queued(p);
update_curr(cfs_rq);
+ if (curr)
+ se->vruntime = curr->vruntime;
place_entity(cfs_rq, se, 1);
/* 'curr' will be NULL if the child belongs to a different group */
if (sysctl_sched_child_runs_first && this_cpu == task_cpu(p) &&
- curr && curr->vruntime < se->vruntime) {
+ curr && entity_before(curr, se)) {
/*
* Upon rescheduling, sched_class::put_prev_task() will place
* 'current' within the tree based on its new key value.