struct list_head queue[MAX_RT_PRIO];
};
-#ifdef CONFIG_FAIR_GROUP_SCHED
+#ifdef CONFIG_GROUP_SCHED
#include <linux/cgroup.h>
struct cfs_rq;
+static LIST_HEAD(task_groups);
+
/* task group related information */
struct task_group {
-#ifdef CONFIG_FAIR_CGROUP_SCHED
+#ifdef CONFIG_CGROUP_SCHED
struct cgroup_subsys_state css;
#endif
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
/* schedulable entities of this group on each cpu */
struct sched_entity **se;
/* runqueue "owned" by this group on each cpu */
*
*/
unsigned long shares;
+#endif
+
+#ifdef CONFIG_RT_GROUP_SCHED
+ struct sched_rt_entity **rt_se;
+ struct rt_rq **rt_rq;
+
+ u64 rt_runtime;
+#endif
struct rcu_head rcu;
+ struct list_head list;
};
+#ifdef CONFIG_FAIR_GROUP_SCHED
/* Default task group's sched entity on each cpu */
static DEFINE_PER_CPU(struct sched_entity, init_sched_entity);
/* Default task group's cfs_rq on each cpu */
static struct sched_entity *init_sched_entity_p[NR_CPUS];
static struct cfs_rq *init_cfs_rq_p[NR_CPUS];
+#endif
-/* task_group_mutex serializes add/remove of task groups and also changes to
+#ifdef CONFIG_RT_GROUP_SCHED
+static DEFINE_PER_CPU(struct sched_rt_entity, init_sched_rt_entity);
+static DEFINE_PER_CPU(struct rt_rq, init_rt_rq) ____cacheline_aligned_in_smp;
+
+static struct sched_rt_entity *init_sched_rt_entity_p[NR_CPUS];
+static struct rt_rq *init_rt_rq_p[NR_CPUS];
+#endif
+
+/* task_group_lock serializes add/remove of task groups and also changes to
* a task group's cpu shares.
*/
-static DEFINE_MUTEX(task_group_mutex);
+static DEFINE_SPINLOCK(task_group_lock);
/* doms_cur_mutex serializes access to doms_cur[] array */
static DEFINE_MUTEX(doms_cur_mutex);
+#ifdef CONFIG_FAIR_GROUP_SCHED
#ifdef CONFIG_SMP
/* kernel thread that runs rebalance_shares() periodically */
static struct task_struct *lb_monitor_task;
static void set_se_shares(struct sched_entity *se, unsigned long shares);
-/* Default task group.
- * Every task in system belong to this group at bootup.
- */
-struct task_group init_task_group = {
- .se = init_sched_entity_p,
- .cfs_rq = init_cfs_rq_p,
-};
-
-#ifdef CONFIG_FAIR_USER_SCHED
+#ifdef CONFIG_USER_SCHED
# define INIT_TASK_GROUP_LOAD (2*NICE_0_LOAD)
#else
# define INIT_TASK_GROUP_LOAD NICE_0_LOAD
#define MIN_GROUP_SHARES 2
static int init_task_group_load = INIT_TASK_GROUP_LOAD;
+#endif
+
+/* Default task group.
+ * Every task in system belong to this group at bootup.
+ */
+struct task_group init_task_group = {
+#ifdef CONFIG_FAIR_GROUP_SCHED
+ .se = init_sched_entity_p,
+ .cfs_rq = init_cfs_rq_p,
+#endif
+
+#ifdef CONFIG_RT_GROUP_SCHED
+ .rt_se = init_sched_rt_entity_p,
+ .rt_rq = init_rt_rq_p,
+#endif
+};
/* return group to which a task belongs */
static inline struct task_group *task_group(struct task_struct *p)
{
struct task_group *tg;
-#ifdef CONFIG_FAIR_USER_SCHED
+#ifdef CONFIG_USER_SCHED
tg = p->user->tg;
-#elif defined(CONFIG_FAIR_CGROUP_SCHED)
+#elif defined(CONFIG_CGROUP_SCHED)
tg = container_of(task_subsys_state(p, cpu_cgroup_subsys_id),
struct task_group, css);
#else
}
/* Change a task's cfs_rq and parent entity if it moves across CPUs/groups */
-static inline void set_task_cfs_rq(struct task_struct *p, unsigned int cpu)
+static inline void set_task_rq(struct task_struct *p, unsigned int cpu)
{
+#ifdef CONFIG_FAIR_GROUP_SCHED
p->se.cfs_rq = task_group(p)->cfs_rq[cpu];
p->se.parent = task_group(p)->se[cpu];
-}
-
-static inline void lock_task_group_list(void)
-{
- mutex_lock(&task_group_mutex);
-}
+#endif
-static inline void unlock_task_group_list(void)
-{
- mutex_unlock(&task_group_mutex);
+#ifdef CONFIG_RT_GROUP_SCHED
+ p->rt.rt_rq = task_group(p)->rt_rq[cpu];
+ p->rt.parent = task_group(p)->rt_se[cpu];
+#endif
}
static inline void lock_doms_cur(void)
#else
-static inline void set_task_cfs_rq(struct task_struct *p, unsigned int cpu) { }
-static inline void lock_task_group_list(void) { }
-static inline void unlock_task_group_list(void) { }
+static inline void set_task_rq(struct task_struct *p, unsigned int cpu) { }
static inline void lock_doms_cur(void) { }
static inline void unlock_doms_cur(void) { }
-#endif /* CONFIG_FAIR_GROUP_SCHED */
+#endif /* CONFIG_GROUP_SCHED */
/* CFS-related fields in a runqueue */
struct cfs_rq {
struct rt_rq {
struct rt_prio_array active;
unsigned long rt_nr_running;
+#if defined CONFIG_SMP || defined CONFIG_RT_GROUP_SCHED
+ int highest_prio; /* highest queued rt task prio */
+#endif
#ifdef CONFIG_SMP
unsigned long rt_nr_migratory;
- int highest_prio; /* highest queued rt task prio */
int overloaded;
#endif
+ int rt_throttled;
u64 rt_time;
- u64 rt_throttled;
+
+#ifdef CONFIG_RT_GROUP_SCHED
+ unsigned long rt_nr_boosted;
+
+ struct rq *rq;
+ struct list_head leaf_rt_rq_list;
+ struct task_group *tg;
+ struct sched_rt_entity *rt_se;
+#endif
};
#ifdef CONFIG_SMP
u64 nr_switches;
struct cfs_rq cfs;
+ struct rt_rq rt;
+ u64 rt_period_expire;
+ int rt_throttled;
+
#ifdef CONFIG_FAIR_GROUP_SCHED
/* list of leaf cfs_rq on this cpu: */
struct list_head leaf_cfs_rq_list;
#endif
- struct rt_rq rt;
- u64 rt_period_expire;
+#ifdef CONFIG_RT_GROUP_SCHED
+ struct list_head leaf_rt_rq_list;
+#endif
/*
* This is part of a global counter where only the total sum
u64 clock, prev_clock_raw;
s64 clock_max_delta;
- unsigned int clock_warps, clock_overflows;
+ unsigned int clock_warps, clock_overflows, clock_underflows;
u64 idle_clock;
unsigned int clock_deep_idle_events;
u64 tick_timestamp;
#define task_rq(p) cpu_rq(task_cpu(p))
#define cpu_curr(cpu) (cpu_rq(cpu)->curr)
+unsigned long rt_needs_cpu(int cpu)
+{
+ struct rq *rq = cpu_rq(cpu);
+ u64 delta;
+
+ if (!rq->rt_throttled)
+ return 0;
+
+ if (rq->clock > rq->rt_period_expire)
+ return 1;
+
+ delta = rq->rt_period_expire - rq->clock;
+ do_div(delta, NSEC_PER_SEC / HZ);
+
+ return (unsigned long)delta;
+}
+
/*
* Tunables that become constants when CONFIG_SCHED_DEBUG is off:
*/
const_debug unsigned int sysctl_sched_nr_migrate = 32;
/*
- * period over which we measure -rt task cpu usage in ms.
+ * period over which we measure -rt task cpu usage in us.
* default: 1s
*/
-const_debug unsigned int sysctl_sched_rt_period = 1000;
+unsigned int sysctl_sched_rt_period = 1000000;
-#define SCHED_RT_FRAC_SHIFT 16
-#define SCHED_RT_FRAC (1UL << SCHED_RT_FRAC_SHIFT)
+/*
+ * part of the period that we allow rt tasks to run in us.
+ * default: 0.95s
+ */
+int sysctl_sched_rt_runtime = 950000;
/*
- * ratio of time -rt tasks may consume.
- * default: 100%
+ * single value that denotes runtime == period, ie unlimited time.
*/
-const_debug unsigned int sysctl_sched_rt_ratio = SCHED_RT_FRAC;
+#define RUNTIME_INF ((u64)~0ULL)
/*
* For kernel-internal use: high-speed (but slightly incorrect) per-cpu
struct rq *rq = cpu_rq(smp_processor_id());
u64 now = sched_clock();
- touch_softlockup_watchdog();
rq->idle_clock += delta_ns;
/*
* Override the previous timestamp and ignore all
rq->prev_clock_raw = now;
rq->clock += delta_ns;
spin_unlock(&rq->lock);
+ touch_softlockup_watchdog();
}
EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event);
#define sched_class_highest (&rt_sched_class)
-static void inc_nr_running(struct task_struct *p, struct rq *rq)
+static void inc_nr_running(struct rq *rq)
{
rq->nr_running++;
}
-static void dec_nr_running(struct task_struct *p, struct rq *rq)
+static void dec_nr_running(struct rq *rq)
{
rq->nr_running--;
}
*/
static void activate_task(struct rq *rq, struct task_struct *p, int wakeup)
{
- if (p->state == TASK_UNINTERRUPTIBLE)
+ if (task_contributes_to_load(p))
rq->nr_uninterruptible--;
enqueue_task(rq, p, wakeup);
- inc_nr_running(p, rq);
+ inc_nr_running(rq);
}
/*
*/
static void deactivate_task(struct rq *rq, struct task_struct *p, int sleep)
{
- if (p->state == TASK_UNINTERRUPTIBLE)
+ if (task_contributes_to_load(p))
rq->nr_uninterruptible++;
dequeue_task(rq, p, sleep);
- dec_nr_running(p, rq);
+ dec_nr_running(rq);
}
/**
static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu)
{
- set_task_cfs_rq(p, cpu);
+ set_task_rq(p, cpu);
#ifdef CONFIG_SMP
/*
* After ->cpu is set up to a new value, task_rq_lock(p, ...) can be
return success;
}
-int fastcall wake_up_process(struct task_struct *p)
+int wake_up_process(struct task_struct *p)
{
- return try_to_wake_up(p, TASK_STOPPED | TASK_TRACED |
- TASK_INTERRUPTIBLE | TASK_UNINTERRUPTIBLE, 0);
+ return try_to_wake_up(p, TASK_ALL, 0);
}
EXPORT_SYMBOL(wake_up_process);
-int fastcall wake_up_state(struct task_struct *p, unsigned int state)
+int wake_up_state(struct task_struct *p, unsigned int state)
{
return try_to_wake_up(p, state, 0);
}
* that must be done for every newly created context, then puts the task
* on the runqueue and wakes it.
*/
-void fastcall wake_up_new_task(struct task_struct *p, unsigned long clone_flags)
+void wake_up_new_task(struct task_struct *p, unsigned long clone_flags)
{
unsigned long flags;
struct rq *rq;
* management (if any):
*/
p->sched_class->task_new(rq, p);
- inc_nr_running(p, rq);
+ inc_nr_running(rq);
}
check_preempt_curr(rq, p);
#ifdef CONFIG_SMP
/*
* Let rq->clock advance by at least TICK_NSEC:
*/
- if (unlikely(rq->clock < next_tick))
+ if (unlikely(rq->clock < next_tick)) {
rq->clock = next_tick;
+ rq->clock_underflows++;
+ }
rq->tick_timestamp = rq->clock;
update_cpu_load(rq);
curr->sched_class->task_tick(rq, curr, 0);
#if defined(CONFIG_PREEMPT) && defined(CONFIG_DEBUG_PREEMPT)
-void fastcall add_preempt_count(int val)
+void add_preempt_count(int val)
{
/*
* Underflow?
}
EXPORT_SYMBOL(add_preempt_count);
-void fastcall sub_preempt_count(int val)
+void sub_preempt_count(int val)
{
/*
* Underflow?
asmlinkage void __sched preempt_schedule(void)
{
struct thread_info *ti = current_thread_info();
-#ifdef CONFIG_PREEMPT_BKL
struct task_struct *task = current;
int saved_lock_depth;
-#endif
+
/*
* If there is a non-zero preempt_count or interrupts are disabled,
* we do not want to preempt the current task. Just return..
* clear ->lock_depth so that schedule() doesnt
* auto-release the semaphore:
*/
-#ifdef CONFIG_PREEMPT_BKL
saved_lock_depth = task->lock_depth;
task->lock_depth = -1;
-#endif
schedule();
-#ifdef CONFIG_PREEMPT_BKL
task->lock_depth = saved_lock_depth;
-#endif
sub_preempt_count(PREEMPT_ACTIVE);
/*
asmlinkage void __sched preempt_schedule_irq(void)
{
struct thread_info *ti = current_thread_info();
-#ifdef CONFIG_PREEMPT_BKL
struct task_struct *task = current;
int saved_lock_depth;
-#endif
+
/* Catch callers which need to be fixed */
BUG_ON(ti->preempt_count || !irqs_disabled());
* clear ->lock_depth so that schedule() doesnt
* auto-release the semaphore:
*/
-#ifdef CONFIG_PREEMPT_BKL
saved_lock_depth = task->lock_depth;
task->lock_depth = -1;
-#endif
local_irq_enable();
schedule();
local_irq_disable();
-#ifdef CONFIG_PREEMPT_BKL
task->lock_depth = saved_lock_depth;
-#endif
sub_preempt_count(PREEMPT_ACTIVE);
/*
* @nr_exclusive: how many wake-one or wake-many threads to wake up
* @key: is directly passed to the wakeup function
*/
-void fastcall __wake_up(wait_queue_head_t *q, unsigned int mode,
+void __wake_up(wait_queue_head_t *q, unsigned int mode,
int nr_exclusive, void *key)
{
unsigned long flags;
/*
* Same as __wake_up but called with the spinlock in wait_queue_head_t held.
*/
-void fastcall __wake_up_locked(wait_queue_head_t *q, unsigned int mode)
+void __wake_up_locked(wait_queue_head_t *q, unsigned int mode)
{
__wake_up_common(q, mode, 1, 0, NULL);
}
*
* On UP it can prevent extra preemption.
*/
-void fastcall
+void
__wake_up_sync(wait_queue_head_t *q, unsigned int mode, int nr_exclusive)
{
unsigned long flags;
spin_lock_irqsave(&x->wait.lock, flags);
x->done++;
- __wake_up_common(&x->wait, TASK_UNINTERRUPTIBLE | TASK_INTERRUPTIBLE,
- 1, 0, NULL);
+ __wake_up_common(&x->wait, TASK_NORMAL, 1, 0, NULL);
spin_unlock_irqrestore(&x->wait.lock, flags);
}
EXPORT_SYMBOL(complete);
spin_lock_irqsave(&x->wait.lock, flags);
x->done += UINT_MAX/2;
- __wake_up_common(&x->wait, TASK_UNINTERRUPTIBLE | TASK_INTERRUPTIBLE,
- 0, 0, NULL);
+ __wake_up_common(&x->wait, TASK_NORMAL, 0, 0, NULL);
spin_unlock_irqrestore(&x->wait.lock, flags);
}
EXPORT_SYMBOL(complete_all);
wait.flags |= WQ_FLAG_EXCLUSIVE;
__add_wait_queue_tail(&x->wait, &wait);
do {
- if (state == TASK_INTERRUPTIBLE &&
- signal_pending(current)) {
+ if ((state == TASK_INTERRUPTIBLE &&
+ signal_pending(current)) ||
+ (state == TASK_KILLABLE &&
+ fatal_signal_pending(current))) {
__remove_wait_queue(&x->wait, &wait);
return -ERESTARTSYS;
}
}
EXPORT_SYMBOL(wait_for_completion_interruptible_timeout);
+int __sched wait_for_completion_killable(struct completion *x)
+{
+ long t = wait_for_common(x, MAX_SCHEDULE_TIMEOUT, TASK_KILLABLE);
+ if (t == -ERESTARTSYS)
+ return t;
+ return 0;
+}
+EXPORT_SYMBOL(wait_for_completion_killable);
+
static long __sched
sleep_on_common(wait_queue_head_t *q, int state, long timeout)
{
return -EPERM;
}
+#ifdef CONFIG_RT_GROUP_SCHED
+ /*
+ * Do not allow realtime tasks into groups that have no runtime
+ * assigned.
+ */
+ if (rt_policy(policy) && task_group(p)->rt_runtime == 0)
+ return -EPERM;
+#endif
+
retval = security_task_setscheduler(p, policy, param);
if (retval)
return retval;
*/
int cond_resched_lock(spinlock_t *lock)
{
+ int resched = need_resched() && system_state == SYSTEM_RUNNING;
int ret = 0;
- if (need_lockbreak(lock)) {
+ if (spin_needbreak(lock) || resched) {
spin_unlock(lock);
- cpu_relax();
- ret = 1;
- spin_lock(lock);
- }
- if (need_resched() && system_state == SYSTEM_RUNNING) {
- spin_release(&lock->dep_map, 1, _THIS_IP_);
- _raw_spin_unlock(lock);
- preempt_enable_no_resched();
- __cond_resched();
+ if (resched && need_resched())
+ __cond_resched();
+ else
+ cpu_relax();
ret = 1;
spin_lock(lock);
}
printk(KERN_CONT "%5lu %5d %6d\n", free,
task_pid_nr(p), task_pid_nr(p->real_parent));
- if (state != TASK_RUNNING)
- show_stack(p, NULL);
+ show_stack(p, NULL);
}
void show_state_filter(unsigned long state_filter)
spin_unlock_irqrestore(&rq->lock, flags);
/* Set the preempt count _outside_ the spinlocks! */
-#if defined(CONFIG_PREEMPT) && !defined(CONFIG_PREEMPT_BKL)
- task_thread_info(idle)->preempt_count = (idle->lock_depth >= 0);
-#else
task_thread_info(idle)->preempt_count = 0;
-#endif
+
/*
* The idle tasks have their own, simple scheduling class:
*/
p->sched_class->set_cpus_allowed(p, &new_mask);
else {
p->cpus_allowed = new_mask;
- p->nr_cpus_allowed = cpus_weight(new_mask);
+ p->rt.nr_cpus_allowed = cpus_weight(new_mask);
}
/* Can the task run on the task's current CPU? If so, we're done */
/* delimiter for bitsearch: */
__set_bit(MAX_RT_PRIO, array->bitmap);
+#if defined CONFIG_SMP || defined CONFIG_RT_GROUP_SCHED
+ rt_rq->highest_prio = MAX_RT_PRIO;
+#endif
#ifdef CONFIG_SMP
rt_rq->rt_nr_migratory = 0;
- rt_rq->highest_prio = MAX_RT_PRIO;
rt_rq->overloaded = 0;
#endif
rt_rq->rt_time = 0;
rt_rq->rt_throttled = 0;
+
+#ifdef CONFIG_RT_GROUP_SCHED
+ rt_rq->rt_nr_boosted = 0;
+ rt_rq->rq = rq;
+#endif
}
+#ifdef CONFIG_FAIR_GROUP_SCHED
+static void init_tg_cfs_entry(struct rq *rq, struct task_group *tg,
+ struct cfs_rq *cfs_rq, struct sched_entity *se,
+ int cpu, int add)
+{
+ tg->cfs_rq[cpu] = cfs_rq;
+ init_cfs_rq(cfs_rq, rq);
+ cfs_rq->tg = tg;
+ if (add)
+ list_add(&cfs_rq->leaf_cfs_rq_list, &rq->leaf_cfs_rq_list);
+
+ tg->se[cpu] = se;
+ se->cfs_rq = &rq->cfs;
+ se->my_q = cfs_rq;
+ se->load.weight = tg->shares;
+ se->load.inv_weight = div64_64(1ULL<<32, se->load.weight);
+ se->parent = NULL;
+}
+#endif
+
+#ifdef CONFIG_RT_GROUP_SCHED
+static void init_tg_rt_entry(struct rq *rq, struct task_group *tg,
+ struct rt_rq *rt_rq, struct sched_rt_entity *rt_se,
+ int cpu, int add)
+{
+ tg->rt_rq[cpu] = rt_rq;
+ init_rt_rq(rt_rq, rq);
+ rt_rq->tg = tg;
+ rt_rq->rt_se = rt_se;
+ if (add)
+ list_add(&rt_rq->leaf_rt_rq_list, &rq->leaf_rt_rq_list);
+
+ tg->rt_se[cpu] = rt_se;
+ rt_se->rt_rq = &rq->rt;
+ rt_se->my_q = rt_rq;
+ rt_se->parent = NULL;
+ INIT_LIST_HEAD(&rt_se->run_list);
+}
+#endif
+
void __init sched_init(void)
{
int highest_cpu = 0;
init_defrootdomain();
#endif
+#ifdef CONFIG_GROUP_SCHED
+ list_add(&init_task_group.list, &task_groups);
+#endif
+
for_each_possible_cpu(i) {
struct rq *rq;
rq->nr_running = 0;
rq->clock = 1;
init_cfs_rq(&rq->cfs, rq);
+ init_rt_rq(&rq->rt, rq);
#ifdef CONFIG_FAIR_GROUP_SCHED
- INIT_LIST_HEAD(&rq->leaf_cfs_rq_list);
- {
- struct cfs_rq *cfs_rq = &per_cpu(init_cfs_rq, i);
- struct sched_entity *se =
- &per_cpu(init_sched_entity, i);
-
- init_cfs_rq_p[i] = cfs_rq;
- init_cfs_rq(cfs_rq, rq);
- cfs_rq->tg = &init_task_group;
- list_add(&cfs_rq->leaf_cfs_rq_list,
- &rq->leaf_cfs_rq_list);
-
- init_sched_entity_p[i] = se;
- se->cfs_rq = &rq->cfs;
- se->my_q = cfs_rq;
- se->load.weight = init_task_group_load;
- se->load.inv_weight =
- div64_64(1ULL<<32, init_task_group_load);
- se->parent = NULL;
- }
init_task_group.shares = init_task_group_load;
+ INIT_LIST_HEAD(&rq->leaf_cfs_rq_list);
+ init_tg_cfs_entry(rq, &init_task_group,
+ &per_cpu(init_cfs_rq, i),
+ &per_cpu(init_sched_entity, i), i, 1);
+
+#endif
+#ifdef CONFIG_RT_GROUP_SCHED
+ init_task_group.rt_runtime =
+ sysctl_sched_rt_runtime * NSEC_PER_USEC;
+ INIT_LIST_HEAD(&rq->leaf_rt_rq_list);
+ init_tg_rt_entry(rq, &init_task_group,
+ &per_cpu(init_rt_rq, i),
+ &per_cpu(init_sched_rt_entity, i), i, 1);
#endif
- init_rt_rq(&rq->rt, rq);
rq->rt_period_expire = 0;
+ rq->rt_throttled = 0;
for (j = 0; j < CPU_LOAD_IDX_MAX; j++)
rq->cpu_load[j] = 0;
unsigned long flags;
struct rq *rq;
- read_lock_irq(&tasklist_lock);
+ read_lock_irqsave(&tasklist_lock, flags);
do_each_thread(g, p) {
/*
* Only normalize user tasks:
continue;
}
- spin_lock_irqsave(&p->pi_lock, flags);
+ spin_lock(&p->pi_lock);
rq = __task_rq_lock(p);
normalize_task(rq, p);
__task_rq_unlock(rq);
- spin_unlock_irqrestore(&p->pi_lock, flags);
+ spin_unlock(&p->pi_lock);
} while_each_thread(g, p);
- read_unlock_irq(&tasklist_lock);
+ read_unlock_irqrestore(&tasklist_lock, flags);
}
#endif /* CONFIG_MAGIC_SYSRQ */
#endif
-#ifdef CONFIG_FAIR_GROUP_SCHED
+#ifdef CONFIG_GROUP_SCHED
-#ifdef CONFIG_SMP
+#if defined CONFIG_FAIR_GROUP_SCHED && defined CONFIG_SMP
/*
* distribute shares of all task groups among their schedulable entities,
* to reflect load distribution across cpus.
}
#endif /* CONFIG_SMP */
-/* allocate runqueue etc for a new task group */
-struct task_group *sched_create_group(void)
+#ifdef CONFIG_FAIR_GROUP_SCHED
+static void free_fair_sched_group(struct task_group *tg)
+{
+ int i;
+
+ for_each_possible_cpu(i) {
+ if (tg->cfs_rq)
+ kfree(tg->cfs_rq[i]);
+ if (tg->se)
+ kfree(tg->se[i]);
+ }
+
+ kfree(tg->cfs_rq);
+ kfree(tg->se);
+}
+
+static int alloc_fair_sched_group(struct task_group *tg)
{
- struct task_group *tg;
struct cfs_rq *cfs_rq;
struct sched_entity *se;
struct rq *rq;
int i;
- tg = kzalloc(sizeof(*tg), GFP_KERNEL);
- if (!tg)
- return ERR_PTR(-ENOMEM);
-
tg->cfs_rq = kzalloc(sizeof(cfs_rq) * NR_CPUS, GFP_KERNEL);
if (!tg->cfs_rq)
goto err;
if (!tg->se)
goto err;
+ tg->shares = NICE_0_LOAD;
+
for_each_possible_cpu(i) {
rq = cpu_rq(i);
- cfs_rq = kmalloc_node(sizeof(struct cfs_rq), GFP_KERNEL,
- cpu_to_node(i));
+ cfs_rq = kmalloc_node(sizeof(struct cfs_rq),
+ GFP_KERNEL|__GFP_ZERO, cpu_to_node(i));
if (!cfs_rq)
goto err;
- se = kmalloc_node(sizeof(struct sched_entity), GFP_KERNEL,
- cpu_to_node(i));
+ se = kmalloc_node(sizeof(struct sched_entity),
+ GFP_KERNEL|__GFP_ZERO, cpu_to_node(i));
if (!se)
goto err;
- memset(cfs_rq, 0, sizeof(struct cfs_rq));
- memset(se, 0, sizeof(struct sched_entity));
+ init_tg_cfs_entry(rq, tg, cfs_rq, se, i, 0);
+ }
- tg->cfs_rq[i] = cfs_rq;
- init_cfs_rq(cfs_rq, rq);
- cfs_rq->tg = tg;
+ return 1;
- tg->se[i] = se;
- se->cfs_rq = &rq->cfs;
- se->my_q = cfs_rq;
- se->load.weight = NICE_0_LOAD;
- se->load.inv_weight = div64_64(1ULL<<32, NICE_0_LOAD);
- se->parent = NULL;
- }
+ err:
+ return 0;
+}
- tg->shares = NICE_0_LOAD;
+static inline void register_fair_sched_group(struct task_group *tg, int cpu)
+{
+ list_add_rcu(&tg->cfs_rq[cpu]->leaf_cfs_rq_list,
+ &cpu_rq(cpu)->leaf_cfs_rq_list);
+}
- lock_task_group_list();
- for_each_possible_cpu(i) {
- rq = cpu_rq(i);
- cfs_rq = tg->cfs_rq[i];
- list_add_rcu(&cfs_rq->leaf_cfs_rq_list, &rq->leaf_cfs_rq_list);
- }
- unlock_task_group_list();
+static inline void unregister_fair_sched_group(struct task_group *tg, int cpu)
+{
+ list_del_rcu(&tg->cfs_rq[cpu]->leaf_cfs_rq_list);
+}
+#else
+static inline void free_fair_sched_group(struct task_group *tg)
+{
+}
- return tg;
+static inline int alloc_fair_sched_group(struct task_group *tg)
+{
+ return 1;
+}
+
+static inline void register_fair_sched_group(struct task_group *tg, int cpu)
+{
+}
+
+static inline void unregister_fair_sched_group(struct task_group *tg, int cpu)
+{
+}
+#endif
+
+#ifdef CONFIG_RT_GROUP_SCHED
+static void free_rt_sched_group(struct task_group *tg)
+{
+ int i;
-err:
for_each_possible_cpu(i) {
- if (tg->cfs_rq)
- kfree(tg->cfs_rq[i]);
- if (tg->se)
- kfree(tg->se[i]);
+ if (tg->rt_rq)
+ kfree(tg->rt_rq[i]);
+ if (tg->rt_se)
+ kfree(tg->rt_se[i]);
}
- kfree(tg->cfs_rq);
- kfree(tg->se);
- kfree(tg);
- return ERR_PTR(-ENOMEM);
+ kfree(tg->rt_rq);
+ kfree(tg->rt_se);
}
-/* rcu callback to free various structures associated with a task group */
-static void free_sched_group(struct rcu_head *rhp)
+static int alloc_rt_sched_group(struct task_group *tg)
{
- struct task_group *tg = container_of(rhp, struct task_group, rcu);
- struct cfs_rq *cfs_rq;
- struct sched_entity *se;
+ struct rt_rq *rt_rq;
+ struct sched_rt_entity *rt_se;
+ struct rq *rq;
int i;
- /* now it should be safe to free those cfs_rqs */
+ tg->rt_rq = kzalloc(sizeof(rt_rq) * NR_CPUS, GFP_KERNEL);
+ if (!tg->rt_rq)
+ goto err;
+ tg->rt_se = kzalloc(sizeof(rt_se) * NR_CPUS, GFP_KERNEL);
+ if (!tg->rt_se)
+ goto err;
+
+ tg->rt_runtime = 0;
+
for_each_possible_cpu(i) {
- cfs_rq = tg->cfs_rq[i];
- kfree(cfs_rq);
+ rq = cpu_rq(i);
+
+ rt_rq = kmalloc_node(sizeof(struct rt_rq),
+ GFP_KERNEL|__GFP_ZERO, cpu_to_node(i));
+ if (!rt_rq)
+ goto err;
- se = tg->se[i];
- kfree(se);
+ rt_se = kmalloc_node(sizeof(struct sched_rt_entity),
+ GFP_KERNEL|__GFP_ZERO, cpu_to_node(i));
+ if (!rt_se)
+ goto err;
+
+ init_tg_rt_entry(rq, tg, rt_rq, rt_se, i, 0);
}
- kfree(tg->cfs_rq);
- kfree(tg->se);
+ return 1;
+
+ err:
+ return 0;
+}
+
+static inline void register_rt_sched_group(struct task_group *tg, int cpu)
+{
+ list_add_rcu(&tg->rt_rq[cpu]->leaf_rt_rq_list,
+ &cpu_rq(cpu)->leaf_rt_rq_list);
+}
+
+static inline void unregister_rt_sched_group(struct task_group *tg, int cpu)
+{
+ list_del_rcu(&tg->rt_rq[cpu]->leaf_rt_rq_list);
+}
+#else
+static inline void free_rt_sched_group(struct task_group *tg)
+{
+}
+
+static inline int alloc_rt_sched_group(struct task_group *tg)
+{
+ return 1;
+}
+
+static inline void register_rt_sched_group(struct task_group *tg, int cpu)
+{
+}
+
+static inline void unregister_rt_sched_group(struct task_group *tg, int cpu)
+{
+}
+#endif
+
+static void free_sched_group(struct task_group *tg)
+{
+ free_fair_sched_group(tg);
+ free_rt_sched_group(tg);
kfree(tg);
}
+/* allocate runqueue etc for a new task group */
+struct task_group *sched_create_group(void)
+{
+ struct task_group *tg;
+ unsigned long flags;
+ int i;
+
+ tg = kzalloc(sizeof(*tg), GFP_KERNEL);
+ if (!tg)
+ return ERR_PTR(-ENOMEM);
+
+ if (!alloc_fair_sched_group(tg))
+ goto err;
+
+ if (!alloc_rt_sched_group(tg))
+ goto err;
+
+ spin_lock_irqsave(&task_group_lock, flags);
+ for_each_possible_cpu(i) {
+ register_fair_sched_group(tg, i);
+ register_rt_sched_group(tg, i);
+ }
+ list_add_rcu(&tg->list, &task_groups);
+ spin_unlock_irqrestore(&task_group_lock, flags);
+
+ return tg;
+
+err:
+ free_sched_group(tg);
+ return ERR_PTR(-ENOMEM);
+}
+
+/* rcu callback to free various structures associated with a task group */
+static void free_sched_group_rcu(struct rcu_head *rhp)
+{
+ /* now it should be safe to free those cfs_rqs */
+ free_sched_group(container_of(rhp, struct task_group, rcu));
+}
+
/* Destroy runqueue etc associated with a task group */
void sched_destroy_group(struct task_group *tg)
{
- struct cfs_rq *cfs_rq = NULL;
+ unsigned long flags;
int i;
- lock_task_group_list();
+ spin_lock_irqsave(&task_group_lock, flags);
for_each_possible_cpu(i) {
- cfs_rq = tg->cfs_rq[i];
- list_del_rcu(&cfs_rq->leaf_cfs_rq_list);
+ unregister_fair_sched_group(tg, i);
+ unregister_rt_sched_group(tg, i);
}
- unlock_task_group_list();
-
- BUG_ON(!cfs_rq);
+ list_del_rcu(&tg->list);
+ spin_unlock_irqrestore(&task_group_lock, flags);
/* wait for possible concurrent references to cfs_rqs complete */
- call_rcu(&tg->rcu, free_sched_group);
+ call_rcu(&tg->rcu, free_sched_group_rcu);
}
/* change task's runqueue when it moves between groups.
rq = task_rq_lock(tsk, &flags);
- if (tsk->sched_class != &fair_sched_class) {
- set_task_cfs_rq(tsk, task_cpu(tsk));
- goto done;
- }
-
update_rq_clock(rq);
running = task_current(rq, tsk);
tsk->sched_class->put_prev_task(rq, tsk);
}
- set_task_cfs_rq(tsk, task_cpu(tsk));
+ set_task_rq(tsk, task_cpu(tsk));
if (on_rq) {
if (unlikely(running))
enqueue_task(rq, tsk, 0);
}
-done:
task_rq_unlock(rq, &flags);
}
+#ifdef CONFIG_FAIR_GROUP_SCHED
/* rq->lock to be locked by caller */
static void set_se_shares(struct sched_entity *se, unsigned long shares)
{
}
}
+static DEFINE_MUTEX(shares_mutex);
+
int sched_group_set_shares(struct task_group *tg, unsigned long shares)
{
int i;
- struct cfs_rq *cfs_rq;
- struct rq *rq;
+ unsigned long flags;
- lock_task_group_list();
+ mutex_lock(&shares_mutex);
if (tg->shares == shares)
goto done;
* load_balance_fair) from referring to this group first,
* by taking it off the rq->leaf_cfs_rq_list on each cpu.
*/
- for_each_possible_cpu(i) {
- cfs_rq = tg->cfs_rq[i];
- list_del_rcu(&cfs_rq->leaf_cfs_rq_list);
- }
+ spin_lock_irqsave(&task_group_lock, flags);
+ for_each_possible_cpu(i)
+ unregister_fair_sched_group(tg, i);
+ spin_unlock_irqrestore(&task_group_lock, flags);
/* wait for any ongoing reference to this group to finish */
synchronize_sched();
* Enable load balance activity on this group, by inserting it back on
* each cpu's rq->leaf_cfs_rq_list.
*/
- for_each_possible_cpu(i) {
- rq = cpu_rq(i);
- cfs_rq = tg->cfs_rq[i];
- list_add_rcu(&cfs_rq->leaf_cfs_rq_list, &rq->leaf_cfs_rq_list);
- }
+ spin_lock_irqsave(&task_group_lock, flags);
+ for_each_possible_cpu(i)
+ register_fair_sched_group(tg, i);
+ spin_unlock_irqrestore(&task_group_lock, flags);
done:
- unlock_task_group_list();
+ mutex_unlock(&shares_mutex);
return 0;
}
{
return tg->shares;
}
+#endif
+
+#ifdef CONFIG_RT_GROUP_SCHED
+/*
+ * Ensure that the real time constraints are schedulable.
+ */
+static DEFINE_MUTEX(rt_constraints_mutex);
+
+static unsigned long to_ratio(u64 period, u64 runtime)
+{
+ if (runtime == RUNTIME_INF)
+ return 1ULL << 16;
+
+ runtime *= (1ULL << 16);
+ div64_64(runtime, period);
+ return runtime;
+}
+
+static int __rt_schedulable(struct task_group *tg, u64 period, u64 runtime)
+{
+ struct task_group *tgi;
+ unsigned long total = 0;
+ unsigned long global_ratio =
+ to_ratio(sysctl_sched_rt_period,
+ sysctl_sched_rt_runtime < 0 ?
+ RUNTIME_INF : sysctl_sched_rt_runtime);
+
+ rcu_read_lock();
+ list_for_each_entry_rcu(tgi, &task_groups, list) {
+ if (tgi == tg)
+ continue;
+
+ total += to_ratio(period, tgi->rt_runtime);
+ }
+ rcu_read_unlock();
+
+ return total + to_ratio(period, runtime) < global_ratio;
+}
+
+int sched_group_set_rt_runtime(struct task_group *tg, long rt_runtime_us)
+{
+ u64 rt_runtime, rt_period;
+ int err = 0;
-#endif /* CONFIG_FAIR_GROUP_SCHED */
+ rt_period = sysctl_sched_rt_period * NSEC_PER_USEC;
+ rt_runtime = (u64)rt_runtime_us * NSEC_PER_USEC;
+ if (rt_runtime_us == -1)
+ rt_runtime = rt_period;
-#ifdef CONFIG_FAIR_CGROUP_SCHED
+ mutex_lock(&rt_constraints_mutex);
+ if (!__rt_schedulable(tg, rt_period, rt_runtime)) {
+ err = -EINVAL;
+ goto unlock;
+ }
+ if (rt_runtime_us == -1)
+ rt_runtime = RUNTIME_INF;
+ tg->rt_runtime = rt_runtime;
+ unlock:
+ mutex_unlock(&rt_constraints_mutex);
+
+ return err;
+}
+
+long sched_group_rt_runtime(struct task_group *tg)
+{
+ u64 rt_runtime_us;
+
+ if (tg->rt_runtime == RUNTIME_INF)
+ return -1;
+
+ rt_runtime_us = tg->rt_runtime;
+ do_div(rt_runtime_us, NSEC_PER_USEC);
+ return rt_runtime_us;
+}
+#endif
+#endif /* CONFIG_GROUP_SCHED */
+
+#ifdef CONFIG_CGROUP_SCHED
/* return corresponding task_group object of a cgroup */
static inline struct task_group *cgroup_tg(struct cgroup *cgrp)
cpu_cgroup_can_attach(struct cgroup_subsys *ss, struct cgroup *cgrp,
struct task_struct *tsk)
{
+#ifdef CONFIG_RT_GROUP_SCHED
+ /* Don't accept realtime tasks when there is no way for them to run */
+ if (rt_task(tsk) && cgroup_tg(cgrp)->rt_runtime == 0)
+ return -EINVAL;
+#else
/* We don't support RT-tasks being in separate groups */
if (tsk->sched_class != &fair_sched_class)
return -EINVAL;
+#endif
return 0;
}
sched_move_task(tsk);
}
+#ifdef CONFIG_FAIR_GROUP_SCHED
static int cpu_shares_write_uint(struct cgroup *cgrp, struct cftype *cftype,
u64 shareval)
{
return (u64) tg->shares;
}
+#endif
+
+#ifdef CONFIG_RT_GROUP_SCHED
+static int cpu_rt_runtime_write(struct cgroup *cgrp, struct cftype *cft,
+ struct file *file,
+ const char __user *userbuf,
+ size_t nbytes, loff_t *unused_ppos)
+{
+ char buffer[64];
+ int retval = 0;
+ s64 val;
+ char *end;
+
+ if (!nbytes)
+ return -EINVAL;
+ if (nbytes >= sizeof(buffer))
+ return -E2BIG;
+ if (copy_from_user(buffer, userbuf, nbytes))
+ return -EFAULT;
+
+ buffer[nbytes] = 0; /* nul-terminate */
+
+ /* strip newline if necessary */
+ if (nbytes && (buffer[nbytes-1] == '\n'))
+ buffer[nbytes-1] = 0;
+ val = simple_strtoll(buffer, &end, 0);
+ if (*end)
+ return -EINVAL;
+
+ /* Pass to subsystem */
+ retval = sched_group_set_rt_runtime(cgroup_tg(cgrp), val);
+ if (!retval)
+ retval = nbytes;
+ return retval;
+}
+
+static ssize_t cpu_rt_runtime_read(struct cgroup *cgrp, struct cftype *cft,
+ struct file *file,
+ char __user *buf, size_t nbytes,
+ loff_t *ppos)
+{
+ char tmp[64];
+ long val = sched_group_rt_runtime(cgroup_tg(cgrp));
+ int len = sprintf(tmp, "%ld\n", val);
+
+ return simple_read_from_buffer(buf, nbytes, ppos, tmp, len);
+}
+#endif
static struct cftype cpu_files[] = {
+#ifdef CONFIG_FAIR_GROUP_SCHED
{
.name = "shares",
.read_uint = cpu_shares_read_uint,
.write_uint = cpu_shares_write_uint,
},
+#endif
+#ifdef CONFIG_RT_GROUP_SCHED
+ {
+ .name = "rt_runtime_us",
+ .read = cpu_rt_runtime_read,
+ .write = cpu_rt_runtime_write,
+ },
+#endif
};
static int cpu_cgroup_populate(struct cgroup_subsys *ss, struct cgroup *cont)
.early_init = 1,
};
-#endif /* CONFIG_FAIR_CGROUP_SCHED */
+#endif /* CONFIG_CGROUP_SCHED */
#ifdef CONFIG_CGROUP_CPUACCT