*/
#include <linux/latencytop.h>
+#include <linux/sched.h>
/*
* Targeted preemption latency for CPU-bound tasks:
* run vmstat and monitor the context-switches (cs) field)
*/
unsigned int sysctl_sched_latency = 5000000ULL;
+unsigned int normalized_sysctl_sched_latency = 5000000ULL;
+
+/*
+ * The initial- and re-scaling of tunables is configurable
+ * (default SCHED_TUNABLESCALING_LOG = *(1+ilog(ncpus))
+ *
+ * Options are:
+ * SCHED_TUNABLESCALING_NONE - unscaled, always *1
+ * SCHED_TUNABLESCALING_LOG - scaled logarithmical, *1+ilog(ncpus)
+ * SCHED_TUNABLESCALING_LINEAR - scaled linear, *ncpus
+ */
+enum sched_tunable_scaling sysctl_sched_tunable_scaling
+ = SCHED_TUNABLESCALING_LOG;
/*
* Minimal preemption granularity for CPU-bound tasks:
* (default: 1 msec * (1 + ilog(ncpus)), units: nanoseconds)
*/
unsigned int sysctl_sched_min_granularity = 1000000ULL;
+unsigned int normalized_sysctl_sched_min_granularity = 1000000ULL;
/*
* is kept at sysctl_sched_latency / sysctl_sched_min_granularity
* have immediate wakeup/sleep latencies.
*/
unsigned int sysctl_sched_wakeup_granularity = 1000000UL;
+unsigned int normalized_sysctl_sched_wakeup_granularity = 1000000UL;
const_debug unsigned int sysctl_sched_migration_cost = 500000UL;
*/
#ifdef CONFIG_SCHED_DEBUG
-int sched_nr_latency_handler(struct ctl_table *table, int write,
- struct file *filp, void __user *buffer, size_t *lenp,
+int sched_proc_update_handler(struct ctl_table *table, int write,
+ void __user *buffer, size_t *lenp,
loff_t *ppos)
{
- int ret = proc_dointvec_minmax(table, write, filp, buffer, lenp, ppos);
+ int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
+ int factor = get_update_sysctl_factor();
if (ret || !write)
return ret;
sched_nr_latency = DIV_ROUND_UP(sysctl_sched_latency,
sysctl_sched_min_granularity);
+#define WRT_SYSCTL(name) \
+ (normalized_sysctl_##name = sysctl_##name / (factor))
+ WRT_SYSCTL(sched_min_granularity);
+ WRT_SYSCTL(sched_latency);
+ WRT_SYSCTL(sched_wakeup_granularity);
+ WRT_SYSCTL(sched_shares_ratelimit);
+#undef WRT_SYSCTL
+
return 0;
}
#endif
curr->sum_exec_runtime += delta_exec;
schedstat_add(cfs_rq, exec_clock, delta_exec);
delta_exec_weighted = calc_delta_fair(delta_exec, curr);
+
curr->vruntime += delta_exec_weighted;
update_min_vruntime(cfs_rq);
}
if (entity_is_task(curr)) {
struct task_struct *curtask = task_of(curr);
+ trace_sched_stat_runtime(curtask, delta_exec, curr->vruntime);
cpuacct_charge(curtask, delta_exec);
account_group_exec_runtime(curtask, delta_exec);
}
if (initial && sched_feat(START_DEBIT))
vruntime += sched_vslice(cfs_rq, se);
- if (!initial) {
- /* sleeps upto a single latency don't count. */
- if (sched_feat(NEW_FAIR_SLEEPERS)) {
- unsigned long thresh = sysctl_sched_latency;
+ /* sleeps up to a single latency don't count. */
+ if (!initial && sched_feat(FAIR_SLEEPERS)) {
+ unsigned long thresh = sysctl_sched_latency;
- /*
- * Convert the sleeper threshold into virtual time.
- * SCHED_IDLE is a special sub-class. We care about
- * fairness only relative to other SCHED_IDLE tasks,
- * all of which have the same weight.
- */
- if (sched_feat(NORMALIZED_SLEEPER) &&
- (!entity_is_task(se) ||
- task_of(se)->policy != SCHED_IDLE))
- thresh = calc_delta_fair(thresh, se);
+ /*
+ * Convert the sleeper threshold into virtual time.
+ * SCHED_IDLE is a special sub-class. We care about
+ * fairness only relative to other SCHED_IDLE tasks,
+ * all of which have the same weight.
+ */
+ if (sched_feat(NORMALIZED_SLEEPER) && (!entity_is_task(se) ||
+ task_of(se)->policy != SCHED_IDLE))
+ thresh = calc_delta_fair(thresh, se);
- vruntime -= thresh;
- }
+ /*
+ * Halve their sleep time's effect, to allow
+ * for a gentler effect of sleepers:
+ */
+ if (sched_feat(GENTLE_FAIR_SLEEPERS))
+ thresh >>= 1;
+
+ vruntime -= thresh;
}
/* ensure we never gain time by being placed backwards. */
se->vruntime = vruntime;
}
+#define ENQUEUE_WAKEUP 1
+#define ENQUEUE_MIGRATE 2
+
static void
-enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int wakeup)
+enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags)
{
/*
+ * Update the normalized vruntime before updating min_vruntime
+ * through callig update_curr().
+ */
+ if (!(flags & ENQUEUE_WAKEUP) || (flags & ENQUEUE_MIGRATE))
+ se->vruntime += cfs_rq->min_vruntime;
+
+ /*
* Update run-time statistics of the 'current'.
*/
update_curr(cfs_rq);
account_entity_enqueue(cfs_rq, se);
- if (wakeup) {
+ if (flags & ENQUEUE_WAKEUP) {
place_entity(cfs_rq, se, 0);
enqueue_sleeper(cfs_rq, se);
}
static void __clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se)
{
- if (cfs_rq->last == se)
+ if (!se || cfs_rq->last == se)
cfs_rq->last = NULL;
- if (cfs_rq->next == se)
+ if (!se || cfs_rq->next == se)
cfs_rq->next = NULL;
}
__dequeue_entity(cfs_rq, se);
account_entity_dequeue(cfs_rq, se);
update_min_vruntime(cfs_rq);
+
+ /*
+ * Normalize the entity after updating the min_vruntime because the
+ * update can refer to the ->curr item and we need to reflect this
+ * movement in our normalized position.
+ */
+ if (!sleep)
+ se->vruntime -= cfs_rq->min_vruntime;
}
/*
* re-elected due to buddy favours.
*/
clear_buddies(cfs_rq, curr);
+ return;
+ }
+
+ /*
+ * Ensure that a task that missed wakeup preemption by a
+ * narrow margin doesn't have to wait for a full slice.
+ * This also mitigates buddy induced latencies under load.
+ */
+ if (!sched_feat(WAKEUP_PREEMPT))
+ return;
+
+ if (delta_exec < sysctl_sched_min_granularity)
+ return;
+
+ if (cfs_rq->nr_running > 1) {
+ struct sched_entity *se = __pick_next_entity(cfs_rq);
+ s64 delta = curr->vruntime - se->vruntime;
+
+ if (delta > ideal_runtime)
+ resched_task(rq_of(cfs_rq)->curr);
}
}
static struct sched_entity *pick_next_entity(struct cfs_rq *cfs_rq)
{
struct sched_entity *se = __pick_next_entity(cfs_rq);
+ struct sched_entity *left = se;
- if (cfs_rq->next && wakeup_preempt_entity(cfs_rq->next, se) < 1)
- return cfs_rq->next;
+ if (cfs_rq->next && wakeup_preempt_entity(cfs_rq->next, left) < 1)
+ se = cfs_rq->next;
- if (cfs_rq->last && wakeup_preempt_entity(cfs_rq->last, se) < 1)
- return cfs_rq->last;
+ /*
+ * Prefer last buddy, try to return the CPU to a preempted task.
+ */
+ if (cfs_rq->last && wakeup_preempt_entity(cfs_rq->last, left) < 1)
+ se = cfs_rq->last;
+
+ clear_buddies(cfs_rq, se);
return se;
}
{
struct cfs_rq *cfs_rq;
struct sched_entity *se = &p->se;
+ int flags = 0;
+
+ if (wakeup)
+ flags |= ENQUEUE_WAKEUP;
+ if (p->state == TASK_WAKING)
+ flags |= ENQUEUE_MIGRATE;
for_each_sched_entity(se) {
if (se->on_rq)
break;
cfs_rq = cfs_rq_of(se);
- enqueue_entity(cfs_rq, se, wakeup);
- wakeup = 1;
+ enqueue_entity(cfs_rq, se, flags);
+ flags = ENQUEUE_WAKEUP;
}
hrtick_update(rq);
#ifdef CONFIG_SMP
+static void task_waking_fair(struct rq *rq, struct task_struct *p)
+{
+ struct sched_entity *se = &p->se;
+ struct cfs_rq *cfs_rq = cfs_rq_of(se);
+
+ se->vruntime -= cfs_rq->min_vruntime;
+}
+
#ifdef CONFIG_FAIR_GROUP_SCHED
/*
* effective_load() calculates the load change as seen from the root_task_group
load = source_load(prev_cpu, idx);
this_load = target_load(this_cpu, idx);
- if (sync && (curr->se.avg_overlap > sysctl_sched_migration_cost ||
- p->se.avg_overlap > sysctl_sched_migration_cost))
- sync = 0;
+ if (sync) {
+ if (sched_feat(SYNC_LESS) &&
+ (curr->se.avg_overlap > sysctl_sched_migration_cost ||
+ p->se.avg_overlap > sysctl_sched_migration_cost))
+ sync = 0;
+ } else {
+ if (sched_feat(SYNC_MORE) &&
+ (curr->se.avg_overlap < sysctl_sched_migration_cost &&
+ p->se.avg_overlap < sysctl_sched_migration_cost))
+ sync = 1;
+ }
/*
* If sync wakeup then subtract the (maximum possible)
* domain.
*/
static struct sched_group *
-find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)
+find_idlest_group(struct sched_domain *sd, struct task_struct *p,
+ int this_cpu, int load_idx)
{
struct sched_group *idlest = NULL, *this = NULL, *group = sd->groups;
unsigned long min_load = ULONG_MAX, this_load = 0;
- int load_idx = sd->forkexec_idx;
int imbalance = 100 + (sd->imbalance_pct-100)/2;
do {
}
/*
+ * Try and locate an idle CPU in the sched_domain.
+ */
+static int
+select_idle_sibling(struct task_struct *p, struct sched_domain *sd, int target)
+{
+ int cpu = smp_processor_id();
+ int prev_cpu = task_cpu(p);
+ int i;
+
+ /*
+ * If this domain spans both cpu and prev_cpu (see the SD_WAKE_AFFINE
+ * test in select_task_rq_fair) and the prev_cpu is idle then that's
+ * always a better target than the current cpu.
+ */
+ if (target == cpu && !cpu_rq(prev_cpu)->cfs.nr_running)
+ return prev_cpu;
+
+ /*
+ * Otherwise, iterate the domain and find an elegible idle cpu.
+ */
+ for_each_cpu_and(i, sched_domain_span(sd), &p->cpus_allowed) {
+ if (!cpu_rq(i)->cfs.nr_running) {
+ target = i;
+ break;
+ }
+ }
+
+ return target;
+}
+
+/*
* sched_balance_self: balance the current task (running on cpu) in domains
* that have the 'flag' flag set. In practice, this is SD_BALANCE_FORK and
* SD_BALANCE_EXEC.
*
* preempt must be disabled.
*/
-static int select_task_rq_fair(struct task_struct *p, int flag, int sync)
+static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags)
{
- struct task_struct *t = current;
- struct sched_domain *tmp, *sd = NULL;
+ struct sched_domain *tmp, *affine_sd = NULL, *sd = NULL;
int cpu = smp_processor_id();
int prev_cpu = task_cpu(p);
int new_cpu = cpu;
int want_affine = 0;
+ int want_sd = 1;
+ int sync = wake_flags & WF_SYNC;
- if (flag & SD_BALANCE_WAKE) {
- if (sched_feat(AFFINE_WAKEUPS))
+ if (sd_flag & SD_BALANCE_WAKE) {
+ if (sched_feat(AFFINE_WAKEUPS) &&
+ cpumask_test_cpu(cpu, &p->cpus_allowed))
want_affine = 1;
new_cpu = prev_cpu;
}
- rcu_read_lock();
for_each_domain(cpu, tmp) {
+ if (!(tmp->flags & SD_LOAD_BALANCE))
+ continue;
+
/*
* If power savings logic is enabled for a domain, see if we
* are not overloaded, if so, don't balance wider.
*/
- if (tmp->flags & SD_POWERSAVINGS_BALANCE) {
+ if (tmp->flags & (SD_POWERSAVINGS_BALANCE|SD_PREFER_LOCAL)) {
unsigned long power = 0;
unsigned long nr_running = 0;
unsigned long capacity;
capacity = DIV_ROUND_CLOSEST(power, SCHED_LOAD_SCALE);
- if (nr_running/2 < capacity)
- break;
- }
+ if (tmp->flags & SD_POWERSAVINGS_BALANCE)
+ nr_running /= 2;
- switch (flag) {
- case SD_BALANCE_WAKE:
- if (!sched_feat(LB_WAKEUP_UPDATE))
- break;
- case SD_BALANCE_FORK:
- case SD_BALANCE_EXEC:
- if (root_task_group_empty())
- break;
- update_shares(tmp);
- default:
- break;
+ if (nr_running < capacity)
+ want_sd = 0;
}
- if (want_affine && (tmp->flags & SD_WAKE_AFFINE) &&
- cpumask_test_cpu(prev_cpu, sched_domain_span(tmp))) {
+ /*
+ * While iterating the domains looking for a spanning
+ * WAKE_AFFINE domain, adjust the affine target to any idle cpu
+ * in cache sharing domains along the way.
+ */
+ if (want_affine) {
+ int target = -1;
- if (wake_affine(tmp, p, sync)) {
- new_cpu = cpu;
- goto out;
- }
+ /*
+ * If both cpu and prev_cpu are part of this domain,
+ * cpu is a valid SD_WAKE_AFFINE target.
+ */
+ if (cpumask_test_cpu(prev_cpu, sched_domain_span(tmp)))
+ target = cpu;
- want_affine = 0;
+ /*
+ * If there's an idle sibling in this domain, make that
+ * the wake_affine target instead of the current cpu.
+ */
+ if (tmp->flags & SD_SHARE_PKG_RESOURCES)
+ target = select_idle_sibling(p, tmp, target);
+
+ if (target >= 0) {
+ if (tmp->flags & SD_WAKE_AFFINE) {
+ affine_sd = tmp;
+ want_affine = 0;
+ }
+ cpu = target;
+ }
}
- if (!(tmp->flags & flag))
+ if (!want_sd && !want_affine)
+ break;
+
+ if (!(tmp->flags & sd_flag))
continue;
- sd = tmp;
+ if (want_sd)
+ sd = tmp;
}
+ if (sched_feat(LB_SHARES_UPDATE)) {
+ /*
+ * Pick the largest domain to update shares over
+ */
+ tmp = sd;
+ if (affine_sd && (!tmp ||
+ cpumask_weight(sched_domain_span(affine_sd)) >
+ cpumask_weight(sched_domain_span(sd))))
+ tmp = affine_sd;
+
+ if (tmp)
+ update_shares(tmp);
+ }
+
+ if (affine_sd && wake_affine(affine_sd, p, sync))
+ return cpu;
+
while (sd) {
+ int load_idx = sd->forkexec_idx;
struct sched_group *group;
int weight;
- if (!(sd->flags & flag)) {
+ if (!(sd->flags & sd_flag)) {
sd = sd->child;
continue;
}
- group = find_idlest_group(sd, t, cpu);
+ if (sd_flag & SD_BALANCE_WAKE)
+ load_idx = sd->wake_idx;
+
+ group = find_idlest_group(sd, p, cpu, load_idx);
if (!group) {
sd = sd->child;
continue;
}
- new_cpu = find_idlest_cpu(group, t, cpu);
+ new_cpu = find_idlest_cpu(group, p, cpu);
if (new_cpu == -1 || new_cpu == cpu) {
/* Now try balancing at a lower domain level of cpu */
sd = sd->child;
for_each_domain(cpu, tmp) {
if (weight <= cpumask_weight(sched_domain_span(tmp)))
break;
- if (tmp->flags & flag)
+ if (tmp->flags & sd_flag)
sd = tmp;
}
/* while loop will break here if sd == NULL */
}
-out:
- rcu_read_unlock();
return new_cpu;
}
#endif /* CONFIG_SMP */
/*
* Preempt the current task with a newly woken task if needed:
*/
-static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int sync)
+static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_flags)
{
struct task_struct *curr = rq->curr;
struct sched_entity *se = &curr->se, *pse = &p->se;
struct cfs_rq *cfs_rq = task_cfs_rq(curr);
+ int sync = wake_flags & WF_SYNC;
+ int scale = cfs_rq->nr_running >= sched_nr_latency;
- update_curr(cfs_rq);
-
- if (unlikely(rt_prio(p->prio))) {
- resched_task(curr);
- return;
- }
+ if (unlikely(rt_prio(p->prio)))
+ goto preempt;
if (unlikely(p->sched_class != &fair_sched_class))
return;
if (unlikely(se == pse))
return;
- /*
- * Only set the backward buddy when the current task is still on the
- * rq. This can happen when a wakeup gets interleaved with schedule on
- * the ->pre_schedule() or idle_balance() point, either of which can
- * drop the rq lock.
- *
- * Also, during early boot the idle thread is in the fair class, for
- * obvious reasons its a bad idea to schedule back to the idle thread.
- */
- if (sched_feat(LAST_BUDDY) && likely(se->on_rq && curr != rq->idle))
- set_last_buddy(se);
- if (sched_feat(NEXT_BUDDY))
+ if (sched_feat(NEXT_BUDDY) && scale && !(wake_flags & WF_FORK))
set_next_buddy(pse);
/*
return;
/* Idle tasks are by definition preempted by everybody. */
- if (unlikely(curr->policy == SCHED_IDLE)) {
- resched_task(curr);
- return;
- }
+ if (unlikely(curr->policy == SCHED_IDLE))
+ goto preempt;
- if (!sched_feat(WAKEUP_PREEMPT))
- return;
+ if (sched_feat(WAKEUP_SYNC) && sync)
+ goto preempt;
+
+ if (sched_feat(WAKEUP_OVERLAP) &&
+ se->avg_overlap < sysctl_sched_migration_cost &&
+ pse->avg_overlap < sysctl_sched_migration_cost)
+ goto preempt;
- 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);
+ if (!sched_feat(WAKEUP_PREEMPT))
return;
- }
+ update_curr(cfs_rq);
find_matching_se(&se, &pse);
-
BUG_ON(!pse);
-
if (wakeup_preempt_entity(se, pse) == 1)
- resched_task(curr);
+ goto preempt;
+
+ return;
+
+preempt:
+ resched_task(curr);
+ /*
+ * Only set the backward buddy when the current task is still
+ * on the rq. This can happen when a wakeup gets interleaved
+ * with schedule on the ->pre_schedule() or idle_balance()
+ * point, either of which can * drop the rq lock.
+ *
+ * Also, during early boot the idle thread is in the fair class,
+ * for obvious reasons its a bad idea to schedule back to it.
+ */
+ if (unlikely(!se->on_rq || curr == rq->idle))
+ return;
+
+ if (sched_feat(LAST_BUDDY) && scale && entity_is_task(se))
+ set_last_buddy(se);
}
static struct task_struct *pick_next_task_fair(struct rq *rq)
struct cfs_rq *cfs_rq = &rq->cfs;
struct sched_entity *se;
- if (unlikely(!cfs_rq->nr_running))
+ if (!cfs_rq->nr_running)
return NULL;
do {
se = pick_next_entity(cfs_rq);
- /*
- * If se was a buddy, clear it so that it will have to earn
- * the favour again.
- */
- __clear_buddies(cfs_rq, se);
set_next_entity(cfs_rq, se);
cfs_rq = group_cfs_rq(se);
} while (cfs_rq);
return 0;
}
+
+static void rq_online_fair(struct rq *rq)
+{
+ update_sysctl();
+}
+
+static void rq_offline_fair(struct rq *rq)
+{
+ update_sysctl();
+}
+
#endif /* CONFIG_SMP */
/*
}
/*
- * Share the fairness runtime between parent and child, thus the
- * total amount of pressure for CPU stays equal - new tasks
- * get a chance to run but frequent forkers are not allowed to
- * monopolize the CPU. Note: the parent runqueue is locked,
- * the child is not running yet.
+ * called on fork with the child task as argument from the parent's context
+ * - child not yet on the tasklist
+ * - preemption disabled
*/
-static void task_new_fair(struct rq *rq, struct task_struct *p)
+static void task_fork_fair(struct task_struct *p)
{
- struct cfs_rq *cfs_rq = task_cfs_rq(p);
+ struct cfs_rq *cfs_rq = task_cfs_rq(current);
struct sched_entity *se = &p->se, *curr = cfs_rq->curr;
int this_cpu = smp_processor_id();
+ struct rq *rq = this_rq();
+ unsigned long flags;
- sched_info_queued(p);
+ raw_spin_lock_irqsave(&rq->lock, flags);
+
+ if (unlikely(task_cpu(p) != this_cpu))
+ __set_task_cpu(p, this_cpu);
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 && entity_before(curr, se)) {
+ if (sysctl_sched_child_runs_first && curr && entity_before(curr, se)) {
/*
* Upon rescheduling, sched_class::put_prev_task() will place
* 'current' within the tree based on its new key value.
resched_task(rq->curr);
}
- enqueue_task_fair(rq, p, 0);
+ se->vruntime -= cfs_rq->min_vruntime;
+
+ raw_spin_unlock_irqrestore(&rq->lock, flags);
}
/*
}
#ifdef CONFIG_FAIR_GROUP_SCHED
-static void moved_group_fair(struct task_struct *p)
+static void moved_group_fair(struct task_struct *p, int on_rq)
{
struct cfs_rq *cfs_rq = task_cfs_rq(p);
update_curr(cfs_rq);
- place_entity(cfs_rq, &p->se, 1);
+ if (!on_rq)
+ place_entity(cfs_rq, &p->se, 1);
}
#endif
+unsigned int get_rr_interval_fair(struct rq *rq, struct task_struct *task)
+{
+ struct sched_entity *se = &task->se;
+ unsigned int rr_interval = 0;
+
+ /*
+ * Time slice is 0 for SCHED_OTHER tasks that are on an otherwise
+ * idle runqueue:
+ */
+ if (rq->cfs.load.weight)
+ rr_interval = NS_TO_JIFFIES(sched_slice(&rq->cfs, se));
+
+ return rr_interval;
+}
+
/*
* All the scheduling class methods:
*/
.load_balance = load_balance_fair,
.move_one_task = move_one_task_fair,
+ .rq_online = rq_online_fair,
+ .rq_offline = rq_offline_fair,
+
+ .task_waking = task_waking_fair,
#endif
.set_curr_task = set_curr_task_fair,
.task_tick = task_tick_fair,
- .task_new = task_new_fair,
+ .task_fork = task_fork_fair,
.prio_changed = prio_changed_fair,
.switched_to = switched_to_fair,
+ .get_rr_interval = get_rr_interval_fair,
+
#ifdef CONFIG_FAIR_GROUP_SCHED
.moved_group = moved_group_fair,
#endif
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff