X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=kernel%2Fsched.c;h=e43f1a76dd437aee45bae1e1c1f4d28628a078a2;hb=7172c64a7cee4dfa95864f49c914f7ea8cf497c8;hp=3f008130c854837edb102e51ed4fa78517df999c;hpb=f7f1b0f1e2fbadeab12d24236000e778aa9b1ead;p=linux-2.6.git diff --git a/kernel/sched.c b/kernel/sched.c index 3f008130c..e43f1a76d 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -27,12 +27,15 @@ #include #include #include +#include #include #include +#include #include #include #include #include +#include #include #include #include @@ -47,6 +50,8 @@ #include #include #include +#include +#include #include #include @@ -145,7 +150,8 @@ (v1) * (v2_max) / (v1_max) #define DELTA(p) \ - (SCALE(TASK_NICE(p), 40, MAX_BONUS) + INTERACTIVE_DELTA) + (SCALE(TASK_NICE(p) + 20, 40, MAX_BONUS) - 20 * MAX_BONUS / 40 + \ + INTERACTIVE_DELTA) #define TASK_INTERACTIVE(p) \ ((p)->prio <= (p)->static_prio - DELTA(p)) @@ -167,29 +173,28 @@ */ #define SCALE_PRIO(x, prio) \ - max(x * (MAX_PRIO - prio) / (MAX_USER_PRIO/2), MIN_TIMESLICE) + max(x * (MAX_PRIO - prio) / (MAX_USER_PRIO / 2), MIN_TIMESLICE) -static inline unsigned int task_timeslice(task_t *p) +static unsigned int static_prio_timeslice(int static_prio) { - if (p->static_prio < NICE_TO_PRIO(0)) - return SCALE_PRIO(DEF_TIMESLICE*4, p->static_prio); + if (static_prio < NICE_TO_PRIO(0)) + return SCALE_PRIO(DEF_TIMESLICE * 4, static_prio); else - return SCALE_PRIO(DEF_TIMESLICE, p->static_prio); + return SCALE_PRIO(DEF_TIMESLICE, static_prio); +} + +static inline unsigned int task_timeslice(struct task_struct *p) +{ + return static_prio_timeslice(p->static_prio); } -#define task_hot(p, now, sd) ((long long) ((now) - (p)->last_ran) \ - < (long long) (sd)->cache_hot_time) /* * These are the runqueue data structures: */ -#define BITMAP_SIZE ((((MAX_PRIO+1+7)/8)+sizeof(long)-1)/sizeof(long)) - -typedef struct runqueue runqueue_t; - struct prio_array { unsigned int nr_active; - unsigned long bitmap[BITMAP_SIZE]; + DECLARE_BITMAP(bitmap, MAX_PRIO+1); /* include 1 bit for delimiter */ struct list_head queue[MAX_PRIO]; }; @@ -200,7 +205,7 @@ struct prio_array { * (such as the load balancing or the thread migration code), lock * acquire operations must be ordered by ascending &runqueue. */ -struct runqueue { +struct rq { spinlock_t lock; /* @@ -208,8 +213,9 @@ struct runqueue { * remote CPUs use both these fields when doing load calculation. */ unsigned long nr_running; + unsigned long raw_weighted_load; #ifdef CONFIG_SMP - unsigned long cpu_load; + unsigned long cpu_load[3]; #endif unsigned long long nr_switches; @@ -223,9 +229,9 @@ struct runqueue { unsigned long expired_timestamp; unsigned long long timestamp_last_tick; - task_t *curr, *idle; + struct task_struct *curr, *idle; struct mm_struct *prev_mm; - prio_array_t *active, *expired, arrays[2]; + struct prio_array *active, *expired, arrays[2]; int best_expired_prio; atomic_t nr_iowait; @@ -235,8 +241,9 @@ struct runqueue { /* For active balancing */ int active_balance; int push_cpu; + int cpu; /* cpu of this runqueue */ - task_t *migration_thread; + struct task_struct *migration_thread; struct list_head migration_queue; #endif #ifdef CONFIG_VSERVER_HARDCPU @@ -263,36 +270,140 @@ struct runqueue { unsigned long ttwu_cnt; unsigned long ttwu_local; #endif + struct lock_class_key rq_lock_key; }; -static DEFINE_PER_CPU(struct runqueue, runqueues); +static DEFINE_PER_CPU(struct rq, runqueues); + +static inline int cpu_of(struct rq *rq) +{ +#ifdef CONFIG_SMP + return rq->cpu; +#else + return 0; +#endif +} -#define for_each_domain(cpu, domain) \ - for (domain = cpu_rq(cpu)->sd; domain; domain = domain->parent) +/* + * The domain tree (rq->sd) is protected by RCU's quiescent state transition. + * See detach_destroy_domains: synchronize_sched for details. + * + * The domain tree of any CPU may only be accessed from within + * preempt-disabled sections. + */ +#define for_each_domain(cpu, __sd) \ + for (__sd = rcu_dereference(cpu_rq(cpu)->sd); __sd; __sd = __sd->parent) #define cpu_rq(cpu) (&per_cpu(runqueues, (cpu))) #define this_rq() (&__get_cpu_var(runqueues)) #define task_rq(p) cpu_rq(task_cpu(p)) #define cpu_curr(cpu) (cpu_rq(cpu)->curr) -/* - * Default context-switch locking: - */ #ifndef prepare_arch_switch -# define prepare_arch_switch(rq, next) do { } while (0) -# define finish_arch_switch(rq, next) spin_unlock_irq(&(rq)->lock) -# define task_running(rq, p) ((rq)->curr == (p)) +# define prepare_arch_switch(next) do { } while (0) +#endif +#ifndef finish_arch_switch +# define finish_arch_switch(prev) do { } while (0) +#endif + +#ifndef __ARCH_WANT_UNLOCKED_CTXSW +static inline int task_running(struct rq *rq, struct task_struct *p) +{ + return rq->curr == p; +} + +static inline void prepare_lock_switch(struct rq *rq, struct task_struct *next) +{ +} + +static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev) +{ +#ifdef CONFIG_DEBUG_SPINLOCK + /* this is a valid case when another task releases the spinlock */ + rq->lock.owner = current; +#endif + /* + * If we are tracking spinlock dependencies then we have to + * fix up the runqueue lock - which gets 'carried over' from + * prev into current: + */ + spin_acquire(&rq->lock.dep_map, 0, 0, _THIS_IP_); + + spin_unlock_irq(&rq->lock); +} + +#else /* __ARCH_WANT_UNLOCKED_CTXSW */ +static inline int task_running(struct rq *rq, struct task_struct *p) +{ +#ifdef CONFIG_SMP + return p->oncpu; +#else + return rq->curr == p; +#endif +} + +static inline void prepare_lock_switch(struct rq *rq, struct task_struct *next) +{ +#ifdef CONFIG_SMP + /* + * We can optimise this out completely for !SMP, because the + * SMP rebalancing from interrupt is the only thing that cares + * here. + */ + next->oncpu = 1; +#endif +#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW + spin_unlock_irq(&rq->lock); +#else + spin_unlock(&rq->lock); #endif +} + +static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev) +{ +#ifdef CONFIG_SMP + /* + * After ->oncpu is cleared, the task can be moved to a different CPU. + * We must ensure this doesn't happen until the switch is completely + * finished. + */ + smp_wmb(); + prev->oncpu = 0; +#endif +#ifndef __ARCH_WANT_INTERRUPTS_ON_CTXSW + local_irq_enable(); +#endif +} +#endif /* __ARCH_WANT_UNLOCKED_CTXSW */ + +/* + * __task_rq_lock - lock the runqueue a given task resides on. + * Must be called interrupts disabled. + */ +static inline struct rq *__task_rq_lock(struct task_struct *p) + __acquires(rq->lock) +{ + struct rq *rq; + +repeat_lock_task: + rq = task_rq(p); + spin_lock(&rq->lock); + if (unlikely(rq != task_rq(p))) { + spin_unlock(&rq->lock); + goto repeat_lock_task; + } + return rq; +} /* * task_rq_lock - lock the runqueue a given task resides on and disable * interrupts. Note the ordering: we can safely lookup the task_rq without * explicitly disabling preemption. */ -static inline runqueue_t *task_rq_lock(task_t *p, unsigned long *flags) +static struct rq *task_rq_lock(struct task_struct *p, unsigned long *flags) __acquires(rq->lock) { - struct runqueue *rq; + struct rq *rq; repeat_lock_task: local_irq_save(*flags); @@ -305,7 +416,13 @@ repeat_lock_task: return rq; } -static inline void task_rq_unlock(runqueue_t *rq, unsigned long *flags) +static inline void __task_rq_unlock(struct rq *rq) + __releases(rq->lock) +{ + spin_unlock(&rq->lock); +} + +static inline void task_rq_unlock(struct rq *rq, unsigned long *flags) __releases(rq->lock) { spin_unlock_irqrestore(&rq->lock, *flags); @@ -316,7 +433,7 @@ static inline void task_rq_unlock(runqueue_t *rq, unsigned long *flags) * bump this up when changing the output format or the meaning of an existing * format, so that tools can adapt (or abort) */ -#define SCHEDSTAT_VERSION 11 +#define SCHEDSTAT_VERSION 12 static int show_schedstat(struct seq_file *seq, void *v) { @@ -325,7 +442,7 @@ static int show_schedstat(struct seq_file *seq, void *v) seq_printf(seq, "version %d\n", SCHEDSTAT_VERSION); seq_printf(seq, "timestamp %lu\n", jiffies); for_each_online_cpu(cpu) { - runqueue_t *rq = cpu_rq(cpu); + struct rq *rq = cpu_rq(cpu); #ifdef CONFIG_SMP struct sched_domain *sd; int dcnt = 0; @@ -345,6 +462,7 @@ static int show_schedstat(struct seq_file *seq, void *v) #ifdef CONFIG_SMP /* domain-specific stats */ + preempt_disable(); for_each_domain(cpu, sd) { enum idle_type itype; char mask_str[NR_CPUS]; @@ -363,11 +481,13 @@ static int show_schedstat(struct seq_file *seq, void *v) sd->lb_nobusyq[itype], sd->lb_nobusyg[itype]); } - seq_printf(seq, " %lu %lu %lu %lu %lu %lu %lu %lu\n", + seq_printf(seq, " %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu\n", sd->alb_cnt, sd->alb_failed, sd->alb_pushed, - sd->sbe_pushed, sd->sbe_attempts, + sd->sbe_cnt, sd->sbe_balanced, sd->sbe_pushed, + sd->sbf_cnt, sd->sbf_balanced, sd->sbf_pushed, sd->ttwu_wake_remote, sd->ttwu_move_affine, sd->ttwu_move_balance); } + preempt_enable(); #endif } return 0; @@ -399,9 +519,36 @@ struct file_operations proc_schedstat_operations = { .release = single_release, }; +/* + * Expects runqueue lock to be held for atomicity of update + */ +static inline void +rq_sched_info_arrive(struct rq *rq, unsigned long delta_jiffies) +{ + if (rq) { + rq->rq_sched_info.run_delay += delta_jiffies; + rq->rq_sched_info.pcnt++; + } +} + +/* + * Expects runqueue lock to be held for atomicity of update + */ +static inline void +rq_sched_info_depart(struct rq *rq, unsigned long delta_jiffies) +{ + if (rq) + rq->rq_sched_info.cpu_time += delta_jiffies; +} # define schedstat_inc(rq, field) do { (rq)->field++; } while (0) # define schedstat_add(rq, field, amt) do { (rq)->field += (amt); } while (0) #else /* !CONFIG_SCHEDSTATS */ +static inline void +rq_sched_info_arrive(struct rq *rq, unsigned long delta_jiffies) +{} +static inline void +rq_sched_info_depart(struct rq *rq, unsigned long delta_jiffies) +{} # define schedstat_inc(rq, field) do { } while (0) # define schedstat_add(rq, field, amt) do { } while (0) #endif @@ -409,10 +556,10 @@ struct file_operations proc_schedstat_operations = { /* * rq_lock - lock a given runqueue and disable interrupts. */ -static inline runqueue_t *this_rq_lock(void) +static inline struct rq *this_rq_lock(void) __acquires(rq->lock) { - runqueue_t *rq; + struct rq *rq; local_irq_disable(); rq = this_rq(); @@ -421,23 +568,7 @@ static inline runqueue_t *this_rq_lock(void) return rq; } -#ifdef CONFIG_SCHED_SMT -static int cpu_and_siblings_are_idle(int cpu) -{ - int sib; - for_each_cpu_mask(sib, cpu_sibling_map[cpu]) { - if (idle_cpu(sib)) - continue; - return 0; - } - - return 1; -} -#else -#define cpu_and_siblings_are_idle(A) idle_cpu(A) -#endif - -#ifdef CONFIG_SCHEDSTATS +#if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) /* * Called when a process is dequeued from the active array and given * the cpu. We should note that with the exception of interactive @@ -453,7 +584,7 @@ static int cpu_and_siblings_are_idle(int cpu) * long it was from the *first* time it was queued to the time that it * finally hit a cpu. */ -static inline void sched_info_dequeued(task_t *t) +static inline void sched_info_dequeued(struct task_struct *t) { t->sched_info.last_queued = 0; } @@ -463,23 +594,18 @@ static inline void sched_info_dequeued(task_t *t) * long it was waiting to run. We also note when it began so that we * can keep stats on how long its timeslice is. */ -static inline void sched_info_arrive(task_t *t) +static void sched_info_arrive(struct task_struct *t) { - unsigned long now = jiffies, diff = 0; - struct runqueue *rq = task_rq(t); + unsigned long now = jiffies, delta_jiffies = 0; if (t->sched_info.last_queued) - diff = now - t->sched_info.last_queued; + delta_jiffies = now - t->sched_info.last_queued; sched_info_dequeued(t); - t->sched_info.run_delay += diff; + t->sched_info.run_delay += delta_jiffies; t->sched_info.last_arrival = now; t->sched_info.pcnt++; - if (!rq) - return; - - rq->rq_sched_info.run_delay += diff; - rq->rq_sched_info.pcnt++; + rq_sched_info_arrive(task_rq(t), delta_jiffies); } /* @@ -497,25 +623,23 @@ static inline void sched_info_arrive(task_t *t) * the timestamp if it is already not set. It's assumed that * sched_info_dequeued() will clear that stamp when appropriate. */ -static inline void sched_info_queued(task_t *t) +static inline void sched_info_queued(struct task_struct *t) { - if (!t->sched_info.last_queued) - t->sched_info.last_queued = jiffies; + if (unlikely(sched_info_on())) + if (!t->sched_info.last_queued) + t->sched_info.last_queued = jiffies; } /* * Called when a process ceases being the active-running process, either * voluntarily or involuntarily. Now we can calculate how long we ran. */ -static inline void sched_info_depart(task_t *t) +static inline void sched_info_depart(struct task_struct *t) { - struct runqueue *rq = task_rq(t); - unsigned long diff = jiffies - t->sched_info.last_arrival; + unsigned long delta_jiffies = jiffies - t->sched_info.last_arrival; - t->sched_info.cpu_time += diff; - - if (rq) - rq->rq_sched_info.cpu_time += diff; + t->sched_info.cpu_time += delta_jiffies; + rq_sched_info_depart(task_rq(t), delta_jiffies); } /* @@ -523,9 +647,10 @@ static inline void sched_info_depart(task_t *t) * their time slice. (This may also be called when switching to or from * the idle task.) We are only called when prev != next. */ -static inline void sched_info_switch(task_t *prev, task_t *next) +static inline void +__sched_info_switch(struct task_struct *prev, struct task_struct *next) { - struct runqueue *rq = task_rq(prev); + struct rq *rq = task_rq(prev); /* * prev now departs the cpu. It's not interesting to record @@ -538,15 +663,21 @@ static inline void sched_info_switch(task_t *prev, task_t *next) if (next != rq->idle) sched_info_arrive(next); } +static inline void +sched_info_switch(struct task_struct *prev, struct task_struct *next) +{ + if (unlikely(sched_info_on())) + __sched_info_switch(prev, next); +} #else #define sched_info_queued(t) do { } while (0) #define sched_info_switch(t, next) do { } while (0) -#endif /* CONFIG_SCHEDSTATS */ +#endif /* CONFIG_SCHEDSTATS || CONFIG_TASK_DELAY_ACCT */ /* * Adding/removing a task to/from a priority array: */ -static void dequeue_task(struct task_struct *p, prio_array_t *array) +static void dequeue_task(struct task_struct *p, struct prio_array *array) { BUG_ON(p->state & TASK_ONHOLD); array->nr_active--; @@ -555,7 +686,7 @@ static void dequeue_task(struct task_struct *p, prio_array_t *array) __clear_bit(p->prio, array->bitmap); } -static void enqueue_task(struct task_struct *p, prio_array_t *array) +static void enqueue_task(struct task_struct *p, struct prio_array *array) { BUG_ON(p->state & TASK_ONHOLD); sched_info_queued(p); @@ -569,13 +700,14 @@ static void enqueue_task(struct task_struct *p, prio_array_t *array) * Put task to the end of the run list without the overhead of dequeue * followed by enqueue. */ -static void requeue_task(struct task_struct *p, prio_array_t *array) +static void requeue_task(struct task_struct *p, struct prio_array *array) { BUG_ON(p->state & TASK_ONHOLD); list_move_tail(&p->run_list, array->queue + p->prio); } -static inline void enqueue_task_head(struct task_struct *p, prio_array_t *array) +static inline void +enqueue_task_head(struct task_struct *p, struct prio_array *array) { BUG_ON(p->state & TASK_ONHOLD); list_add(&p->run_list, array->queue + p->prio); @@ -585,7 +717,7 @@ static inline void enqueue_task_head(struct task_struct *p, prio_array_t *array) } /* - * effective_prio - return the priority that is based on the static + * __normal_prio - return the priority that is based on the static * priority but is modified by bonuses/penalties. * * We scale the actual sleep average [0 .... MAX_SLEEP_AVG] @@ -598,14 +730,12 @@ static inline void enqueue_task_head(struct task_struct *p, prio_array_t *array) * * Both properties are important to certain workloads. */ -static int effective_prio(task_t *p) + +static inline int __normal_prio(struct task_struct *p) { int bonus, prio; struct vx_info *vxi; - if (rt_task(p)) - return p->prio; - bonus = CURRENT_BONUS(p) - MAX_BONUS / 2; prio = p->static_prio - bonus; @@ -622,64 +752,178 @@ static int effective_prio(task_t *p) } /* - * __activate_task - move a task to the runqueue. + * To aid in avoiding the subversion of "niceness" due to uneven distribution + * of tasks with abnormal "nice" values across CPUs the contribution that + * each task makes to its run queue's load is weighted according to its + * scheduling class and "nice" value. For SCHED_NORMAL tasks this is just a + * scaled version of the new time slice allocation that they receive on time + * slice expiry etc. + */ + +/* + * Assume: static_prio_timeslice(NICE_TO_PRIO(0)) == DEF_TIMESLICE + * If static_prio_timeslice() is ever changed to break this assumption then + * this code will need modification */ -static inline void __activate_task(task_t *p, runqueue_t *rq) +#define TIME_SLICE_NICE_ZERO DEF_TIMESLICE +#define LOAD_WEIGHT(lp) \ + (((lp) * SCHED_LOAD_SCALE) / TIME_SLICE_NICE_ZERO) +#define PRIO_TO_LOAD_WEIGHT(prio) \ + LOAD_WEIGHT(static_prio_timeslice(prio)) +#define RTPRIO_TO_LOAD_WEIGHT(rp) \ + (PRIO_TO_LOAD_WEIGHT(MAX_RT_PRIO) + LOAD_WEIGHT(rp)) + +static void set_load_weight(struct task_struct *p) +{ + if (has_rt_policy(p)) { +#ifdef CONFIG_SMP + if (p == task_rq(p)->migration_thread) + /* + * The migration thread does the actual balancing. + * Giving its load any weight will skew balancing + * adversely. + */ + p->load_weight = 0; + else +#endif + p->load_weight = RTPRIO_TO_LOAD_WEIGHT(p->rt_priority); + } else + p->load_weight = PRIO_TO_LOAD_WEIGHT(p->static_prio); +} + +static inline void +inc_raw_weighted_load(struct rq *rq, const struct task_struct *p) +{ + rq->raw_weighted_load += p->load_weight; +} + +static inline void +dec_raw_weighted_load(struct rq *rq, const struct task_struct *p) +{ + rq->raw_weighted_load -= p->load_weight; +} + +static inline void inc_nr_running(struct task_struct *p, struct rq *rq) { - enqueue_task(p, rq->active); rq->nr_running++; + inc_raw_weighted_load(rq, p); +} + +static inline void dec_nr_running(struct task_struct *p, struct rq *rq) +{ + rq->nr_running--; + dec_raw_weighted_load(rq, p); +} + +/* + * Calculate the expected normal priority: i.e. priority + * without taking RT-inheritance into account. Might be + * boosted by interactivity modifiers. Changes upon fork, + * setprio syscalls, and whenever the interactivity + * estimator recalculates. + */ +static inline int normal_prio(struct task_struct *p) +{ + int prio; + + if (has_rt_policy(p)) + prio = MAX_RT_PRIO-1 - p->rt_priority; + else + prio = __normal_prio(p); + return prio; +} + +/* + * Calculate the current priority, i.e. the priority + * taken into account by the scheduler. This value might + * be boosted by RT tasks, or might be boosted by + * interactivity modifiers. Will be RT if the task got + * RT-boosted. If not then it returns p->normal_prio. + */ +static int effective_prio(struct task_struct *p) +{ + p->normal_prio = normal_prio(p); + /* + * If we are RT tasks or we were boosted to RT priority, + * keep the priority unchanged. Otherwise, update priority + * to the normal priority: + */ + if (!rt_prio(p->prio)) + return p->normal_prio; + return p->prio; +} + +/* + * __activate_task - move a task to the runqueue. + */ +static void __activate_task(struct task_struct *p, struct rq *rq) +{ + struct prio_array *target = rq->active; + + if (batch_task(p)) + target = rq->expired; + enqueue_task(p, target); + inc_nr_running(p, rq); } /* * __activate_idle_task - move idle task to the _front_ of runqueue. */ -static inline void __activate_idle_task(task_t *p, runqueue_t *rq) +static inline void __activate_idle_task(struct task_struct *p, struct rq *rq) { enqueue_task_head(p, rq->active); - rq->nr_running++; + inc_nr_running(p, rq); } -static void recalc_task_prio(task_t *p, unsigned long long now) +/* + * Recalculate p->normal_prio and p->prio after having slept, + * updating the sleep-average too: + */ +static int recalc_task_prio(struct task_struct *p, unsigned long long now) { /* Caller must always ensure 'now >= p->timestamp' */ - unsigned long long __sleep_time = now - p->timestamp; - unsigned long sleep_time; + unsigned long sleep_time = now - p->timestamp; - if (__sleep_time > NS_MAX_SLEEP_AVG) - sleep_time = NS_MAX_SLEEP_AVG; - else - sleep_time = (unsigned long)__sleep_time; + if (batch_task(p)) + sleep_time = 0; if (likely(sleep_time > 0)) { /* - * User tasks that sleep a long time are categorised as - * idle and will get just interactive status to stay active & - * prevent them suddenly becoming cpu hogs and starving - * other processes. + * This ceiling is set to the lowest priority that would allow + * a task to be reinserted into the active array on timeslice + * completion. */ - if (p->mm && p->activated != -1 && - sleep_time > INTERACTIVE_SLEEP(p)) { - p->sleep_avg = JIFFIES_TO_NS(MAX_SLEEP_AVG - - DEF_TIMESLICE); - } else { + unsigned long ceiling = INTERACTIVE_SLEEP(p); + + if (p->mm && sleep_time > ceiling && p->sleep_avg < ceiling) { /* - * The lower the sleep avg a task has the more - * rapidly it will rise with sleep time. + * Prevents user tasks from achieving best priority + * with one single large enough sleep. */ - sleep_time *= (MAX_BONUS - CURRENT_BONUS(p)) ? : 1; - + p->sleep_avg = ceiling; + /* + * Using INTERACTIVE_SLEEP() as a ceiling places a + * nice(0) task 1ms sleep away from promotion, and + * gives it 700ms to round-robin with no chance of + * being demoted. This is more than generous, so + * mark this sleep as non-interactive to prevent the + * on-runqueue bonus logic from intervening should + * this task not receive cpu immediately. + */ + p->sleep_type = SLEEP_NONINTERACTIVE; + } else { /* * Tasks waking from uninterruptible sleep are * limited in their sleep_avg rise as they * are likely to be waiting on I/O */ - if (p->activated == -1 && p->mm) { - if (p->sleep_avg >= INTERACTIVE_SLEEP(p)) + if (p->sleep_type == SLEEP_NONINTERACTIVE && p->mm) { + if (p->sleep_avg >= ceiling) sleep_time = 0; else if (p->sleep_avg + sleep_time >= - INTERACTIVE_SLEEP(p)) { - p->sleep_avg = INTERACTIVE_SLEEP(p); - sleep_time = 0; + ceiling) { + p->sleep_avg = ceiling; + sleep_time = 0; } } @@ -693,12 +937,12 @@ static void recalc_task_prio(task_t *p, unsigned long long now) */ p->sleep_avg += sleep_time; - if (p->sleep_avg > NS_MAX_SLEEP_AVG) - p->sleep_avg = NS_MAX_SLEEP_AVG; } + if (p->sleep_avg > NS_MAX_SLEEP_AVG) + p->sleep_avg = NS_MAX_SLEEP_AVG; } - p->prio = effective_prio(p); + return effective_prio(p); } /* @@ -707,7 +951,7 @@ static void recalc_task_prio(task_t *p, unsigned long long now) * Update all the scheduling statistics stuff. (sleep average * calculation, priority modifiers, etc.) */ -static void activate_task(task_t *p, runqueue_t *rq, int local) +static void activate_task(struct task_struct *p, struct rq *rq, int local) { unsigned long long now; @@ -715,19 +959,20 @@ static void activate_task(task_t *p, runqueue_t *rq, int local) #ifdef CONFIG_SMP if (!local) { /* Compensate for drifting sched_clock */ - runqueue_t *this_rq = this_rq(); + struct rq *this_rq = this_rq(); now = (now - this_rq->timestamp_last_tick) + rq->timestamp_last_tick; } #endif - recalc_task_prio(p, now); + if (!rt_task(p)) + p->prio = recalc_task_prio(p, now); /* * This checks to make sure it's not an uninterruptible task * that is now waking up. */ - if (!p->activated) { + if (p->sleep_type == SLEEP_NORMAL) { /* * Tasks which were woken up by interrupts (ie. hw events) * are most likely of interactive nature. So we give them @@ -736,13 +981,13 @@ static void activate_task(task_t *p, runqueue_t *rq, int local) * on a CPU, first time around: */ if (in_interrupt()) - p->activated = 2; + p->sleep_type = SLEEP_INTERRUPTED; else { /* * Normal first-time wakeups get a credit too for * on-runqueue time, but it will be weighted down: */ - p->activated = 1; + p->sleep_type = SLEEP_INTERACTIVE; } } p->timestamp = now; @@ -754,15 +999,15 @@ static void activate_task(task_t *p, runqueue_t *rq, int local) /* * deactivate_task - remove a task from the runqueue. */ -static void __deactivate_task(struct task_struct *p, runqueue_t *rq) +static void __deactivate_task(struct task_struct *p, struct rq *rq) { - rq->nr_running--; + dec_nr_running(p, rq); dequeue_task(p, p->array); p->array = NULL; } static inline -void deactivate_task(struct task_struct *p, runqueue_t *rq) +void deactivate_task(struct task_struct *p, struct rq *rq) { vx_deactivate_task(p); __deactivate_task(p, rq); @@ -775,7 +1020,7 @@ void deactivate_task(struct task_struct *p, runqueue_t *rq) */ static inline void vx_hold_task(struct vx_info *vxi, - struct task_struct *p, runqueue_t *rq) + struct task_struct *p, struct rq *rq) { __deactivate_task(p, rq); p->state |= TASK_ONHOLD; @@ -789,14 +1034,14 @@ void vx_hold_task(struct vx_info *vxi, */ static inline void vx_unhold_task(struct vx_info *vxi, - struct task_struct *p, runqueue_t *rq) + struct task_struct *p, struct rq *rq) { list_del(&p->run_list); /* one less waiting */ vx_onhold_dec(vxi); p->state &= ~TASK_ONHOLD; enqueue_task(p, rq->expired); - rq->nr_running++; + inc_nr_running(p, rq); if (p->static_prio < rq->best_expired_prio) rq->best_expired_prio = p->static_prio; @@ -804,14 +1049,14 @@ void vx_unhold_task(struct vx_info *vxi, #else static inline void vx_hold_task(struct vx_info *vxi, - struct task_struct *p, runqueue_t *rq) + struct task_struct *p, struct rq *rq) { return; } static inline void vx_unhold_task(struct vx_info *vxi, - struct task_struct *p, runqueue_t *rq) + struct task_struct *p, struct rq *rq) { return; } @@ -826,23 +1071,35 @@ void vx_unhold_task(struct vx_info *vxi, * the target CPU. */ #ifdef CONFIG_SMP -static void resched_task(task_t *p) + +#ifndef tsk_is_polling +#define tsk_is_polling(t) test_tsk_thread_flag(t, TIF_POLLING_NRFLAG) +#endif + +static void resched_task(struct task_struct *p) { - int need_resched, nrpolling; + int cpu; assert_spin_locked(&task_rq(p)->lock); - /* minimise the chance of sending an interrupt to poll_idle() */ - nrpolling = test_tsk_thread_flag(p,TIF_POLLING_NRFLAG); - need_resched = test_and_set_tsk_thread_flag(p,TIF_NEED_RESCHED); - nrpolling |= test_tsk_thread_flag(p,TIF_POLLING_NRFLAG); + if (unlikely(test_tsk_thread_flag(p, TIF_NEED_RESCHED))) + return; + + set_tsk_thread_flag(p, TIF_NEED_RESCHED); - if (!need_resched && !nrpolling && (task_cpu(p) != smp_processor_id())) - smp_send_reschedule(task_cpu(p)); + cpu = task_cpu(p); + if (cpu == smp_processor_id()) + return; + + /* NEED_RESCHED must be visible before we test polling */ + smp_mb(); + if (!tsk_is_polling(p)) + smp_send_reschedule(cpu); } #else -static inline void resched_task(task_t *p) +static inline void resched_task(struct task_struct *p) { + assert_spin_locked(&task_rq(p)->lock); set_tsk_need_resched(p); } #endif @@ -851,38 +1108,35 @@ static inline void resched_task(task_t *p) * task_curr - is this task currently executing on a CPU? * @p: the task in question. */ -inline int task_curr(const task_t *p) +inline int task_curr(const struct task_struct *p) { return cpu_curr(task_cpu(p)) == p; } -#ifdef CONFIG_SMP -enum request_type { - REQ_MOVE_TASK, - REQ_SET_DOMAIN, -}; +/* Used instead of source_load when we know the type == 0 */ +unsigned long weighted_cpuload(const int cpu) +{ + return cpu_rq(cpu)->raw_weighted_load; +} -typedef struct { +#ifdef CONFIG_SMP +struct migration_req { struct list_head list; - enum request_type type; - /* For REQ_MOVE_TASK */ - task_t *task; + struct task_struct *task; int dest_cpu; - /* For REQ_SET_DOMAIN */ - struct sched_domain *sd; - struct completion done; -} migration_req_t; +}; /* * The task's runqueue lock must be held. * Returns true if you have to wait for migration thread. */ -static int migrate_task(task_t *p, int dest_cpu, migration_req_t *req) +static int +migrate_task(struct task_struct *p, int dest_cpu, struct migration_req *req) { - runqueue_t *rq = task_rq(p); + struct rq *rq = task_rq(p); /* * If the task is not on a runqueue (and not running), then @@ -894,10 +1148,10 @@ static int migrate_task(task_t *p, int dest_cpu, migration_req_t *req) } init_completion(&req->done); - req->type = REQ_MOVE_TASK; req->task = p; req->dest_cpu = dest_cpu; list_add(&req->list, &rq->migration_queue); + return 1; } @@ -910,10 +1164,10 @@ static int migrate_task(task_t *p, int dest_cpu, migration_req_t *req) * smp_call_function() if an IPI is sent by the same process we are * waiting to become inactive. */ -void wait_task_inactive(task_t * p) +void wait_task_inactive(struct task_struct *p) { unsigned long flags; - runqueue_t *rq; + struct rq *rq; int preempted; repeat: @@ -944,7 +1198,7 @@ repeat: * to another CPU then no harm is done and the purpose has been * achieved as well. */ -void kick_process(task_t *p) +void kick_process(struct task_struct *p) { int cpu; @@ -956,65 +1210,221 @@ void kick_process(task_t *p) } /* - * Return a low guess at the load of a migration-source cpu. + * Return a low guess at the load of a migration-source cpu weighted + * according to the scheduling class and "nice" value. * * We want to under-estimate the load of migration sources, to * balance conservatively. */ -static inline unsigned long source_load(int cpu) +static inline unsigned long source_load(int cpu, int type) +{ + struct rq *rq = cpu_rq(cpu); + + if (type == 0) + return rq->raw_weighted_load; + + return min(rq->cpu_load[type-1], rq->raw_weighted_load); +} + +/* + * Return a high guess at the load of a migration-target cpu weighted + * according to the scheduling class and "nice" value. + */ +static inline unsigned long target_load(int cpu, int type) { - runqueue_t *rq = cpu_rq(cpu); - unsigned long load_now = rq->nr_running * SCHED_LOAD_SCALE; + struct rq *rq = cpu_rq(cpu); + + if (type == 0) + return rq->raw_weighted_load; - return min(rq->cpu_load, load_now); + return max(rq->cpu_load[type-1], rq->raw_weighted_load); } /* - * Return a high guess at the load of a migration-target cpu + * Return the average load per task on the cpu's run queue */ -static inline unsigned long target_load(int cpu) +static inline unsigned long cpu_avg_load_per_task(int cpu) { - runqueue_t *rq = cpu_rq(cpu); - unsigned long load_now = rq->nr_running * SCHED_LOAD_SCALE; + struct rq *rq = cpu_rq(cpu); + unsigned long n = rq->nr_running; - return max(rq->cpu_load, load_now); + return n ? rq->raw_weighted_load / n : SCHED_LOAD_SCALE; } -#endif +/* + * find_idlest_group finds and returns the least busy CPU group within the + * domain. + */ +static struct sched_group * +find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu) +{ + 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 { + unsigned long load, avg_load; + int local_group; + int i; + + /* Skip over this group if it has no CPUs allowed */ + if (!cpus_intersects(group->cpumask, p->cpus_allowed)) + goto nextgroup; + + local_group = cpu_isset(this_cpu, group->cpumask); + + /* Tally up the load of all CPUs in the group */ + avg_load = 0; + + for_each_cpu_mask(i, group->cpumask) { + /* Bias balancing toward cpus of our domain */ + if (local_group) + load = source_load(i, load_idx); + else + load = target_load(i, load_idx); + + avg_load += load; + } + + /* Adjust by relative CPU power of the group */ + avg_load = (avg_load * SCHED_LOAD_SCALE) / group->cpu_power; + + if (local_group) { + this_load = avg_load; + this = group; + } else if (avg_load < min_load) { + min_load = avg_load; + idlest = group; + } +nextgroup: + group = group->next; + } while (group != sd->groups); + + if (!idlest || 100*this_load < imbalance*min_load) + return NULL; + return idlest; +} /* - * wake_idle() will wake a task on an idle cpu if task->cpu is - * not idle and an idle cpu is available. The span of cpus to - * search starts with cpus closest then further out as needed, - * so we always favor a closer, idle cpu. - * - * Returns the CPU we should wake onto. + * find_idlest_queue - find the idlest runqueue among the cpus in group. */ -#if defined(ARCH_HAS_SCHED_WAKE_IDLE) -static int wake_idle(int cpu, task_t *p) +static int +find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu) { cpumask_t tmp; - struct sched_domain *sd; + unsigned long load, min_load = ULONG_MAX; + int idlest = -1; int i; - if (idle_cpu(cpu)) - return cpu; + /* Traverse only the allowed CPUs */ + cpus_and(tmp, group->cpumask, p->cpus_allowed); - for_each_domain(cpu, sd) { - if (sd->flags & SD_WAKE_IDLE) { - cpus_and(tmp, sd->span, cpu_online_map); - cpus_and(tmp, tmp, p->cpus_allowed); - for_each_cpu_mask(i, tmp) { - if (idle_cpu(i)) - return i; - } + for_each_cpu_mask(i, tmp) { + load = weighted_cpuload(i); + + if (load < min_load || (load == min_load && i == this_cpu)) { + min_load = load; + idlest = i; } - else break; } - return cpu; + + return idlest; } -#else -static inline int wake_idle(int cpu, task_t *p) + +/* + * 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. + * + * Balance, ie. select the least loaded group. + * + * Returns the target CPU number, or the same CPU if no balancing is needed. + * + * preempt must be disabled. + */ +static int sched_balance_self(int cpu, int flag) +{ + struct task_struct *t = current; + struct sched_domain *tmp, *sd = NULL; + + for_each_domain(cpu, tmp) { + /* + * If power savings logic is enabled for a domain, stop there. + */ + if (tmp->flags & SD_POWERSAVINGS_BALANCE) + break; + if (tmp->flags & flag) + sd = tmp; + } + + while (sd) { + cpumask_t span; + struct sched_group *group; + int new_cpu; + int weight; + + span = sd->span; + group = find_idlest_group(sd, t, cpu); + if (!group) + goto nextlevel; + + new_cpu = find_idlest_cpu(group, t, cpu); + if (new_cpu == -1 || new_cpu == cpu) + goto nextlevel; + + /* Now try balancing at a lower domain level */ + cpu = new_cpu; +nextlevel: + sd = NULL; + weight = cpus_weight(span); + for_each_domain(cpu, tmp) { + if (weight <= cpus_weight(tmp->span)) + break; + if (tmp->flags & flag) + sd = tmp; + } + /* while loop will break here if sd == NULL */ + } + + return cpu; +} + +#endif /* CONFIG_SMP */ + +/* + * wake_idle() will wake a task on an idle cpu if task->cpu is + * not idle and an idle cpu is available. The span of cpus to + * search starts with cpus closest then further out as needed, + * so we always favor a closer, idle cpu. + * + * Returns the CPU we should wake onto. + */ +#if defined(ARCH_HAS_SCHED_WAKE_IDLE) +static int wake_idle(int cpu, struct task_struct *p) +{ + cpumask_t tmp; + struct sched_domain *sd; + int i; + + if (idle_cpu(cpu)) + return cpu; + + for_each_domain(cpu, sd) { + if (sd->flags & SD_WAKE_IDLE) { + cpus_and(tmp, sd->span, p->cpus_allowed); + for_each_cpu_mask(i, tmp) { + if (idle_cpu(i)) + return i; + } + } + else + break; + } + return cpu; +} +#else +static inline int wake_idle(int cpu, struct task_struct *p) { return cpu; } @@ -1034,15 +1444,15 @@ static inline int wake_idle(int cpu, task_t *p) * * returns failure only if the task is already active. */ -static int try_to_wake_up(task_t * p, unsigned int state, int sync) +static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync) { int cpu, this_cpu, success = 0; unsigned long flags; long old_state; - runqueue_t *rq; + struct rq *rq; #ifdef CONFIG_SMP + struct sched_domain *sd, *this_sd = NULL; unsigned long load, this_load; - struct sched_domain *sd; int new_cpu; #endif @@ -1067,70 +1477,71 @@ static int try_to_wake_up(task_t * p, unsigned int state, int sync) if (unlikely(task_running(rq, p))) goto out_activate; -#ifdef CONFIG_SCHEDSTATS + new_cpu = cpu; + schedstat_inc(rq, ttwu_cnt); if (cpu == this_cpu) { schedstat_inc(rq, ttwu_local); - } else { - for_each_domain(this_cpu, sd) { - if (cpu_isset(cpu, sd->span)) { - schedstat_inc(sd, ttwu_wake_remote); - break; - } + goto out_set_cpu; + } + + for_each_domain(this_cpu, sd) { + if (cpu_isset(cpu, sd->span)) { + schedstat_inc(sd, ttwu_wake_remote); + this_sd = sd; + break; } } -#endif - new_cpu = cpu; - if (cpu == this_cpu || unlikely(!cpu_isset(this_cpu, p->cpus_allowed))) + if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed))) goto out_set_cpu; - load = source_load(cpu); - this_load = target_load(this_cpu); - /* - * If sync wakeup then subtract the (maximum possible) effect of - * the currently running task from the load of the current CPU: + * Check for affine wakeup and passive balancing possibilities. */ - if (sync) - this_load -= SCHED_LOAD_SCALE; + if (this_sd) { + int idx = this_sd->wake_idx; + unsigned int imbalance; - /* Don't pull the task off an idle CPU to a busy one */ - if (load < SCHED_LOAD_SCALE/2 && this_load > SCHED_LOAD_SCALE/2) - goto out_set_cpu; + imbalance = 100 + (this_sd->imbalance_pct - 100) / 2; - new_cpu = this_cpu; /* Wake to this CPU if we can */ + load = source_load(cpu, idx); + this_load = target_load(this_cpu, idx); - /* - * Scan domains for affine wakeup and passive balancing - * possibilities. - */ - for_each_domain(this_cpu, sd) { - unsigned int imbalance; - /* - * Start passive balancing when half the imbalance_pct - * limit is reached. - */ - imbalance = sd->imbalance_pct + (sd->imbalance_pct - 100) / 2; + new_cpu = this_cpu; /* Wake to this CPU if we can */ + + if (this_sd->flags & SD_WAKE_AFFINE) { + unsigned long tl = this_load; + unsigned long tl_per_task = cpu_avg_load_per_task(this_cpu); - if ((sd->flags & SD_WAKE_AFFINE) && - !task_hot(p, rq->timestamp_last_tick, sd)) { /* - * This domain has SD_WAKE_AFFINE and p is cache cold - * in this domain. + * If sync wakeup then subtract the (maximum possible) + * effect of the currently running task from the load + * of the current CPU: */ - if (cpu_isset(cpu, sd->span)) { - schedstat_inc(sd, ttwu_move_affine); + if (sync) + tl -= current->load_weight; + + if ((tl <= load && + tl + target_load(cpu, idx) <= tl_per_task) || + 100*(tl + p->load_weight) <= imbalance*load) { + /* + * This domain has SD_WAKE_AFFINE and + * p is cache cold in this domain, and + * there is no bad imbalance. + */ + schedstat_inc(this_sd, ttwu_move_affine); goto out_set_cpu; } - } else if ((sd->flags & SD_WAKE_BALANCE) && - imbalance*this_load <= 100*load) { - /* - * This domain has SD_WAKE_BALANCE and there is - * an imbalance. - */ - if (cpu_isset(cpu, sd->span)) { - schedstat_inc(sd, ttwu_move_balance); + } + + /* + * Start passive balancing when half the imbalance_pct + * limit is reached. + */ + if (this_sd->flags & SD_WAKE_BALANCE) { + if (imbalance*this_load <= 100*load) { + schedstat_inc(this_sd, ttwu_move_balance); goto out_set_cpu; } } @@ -1158,13 +1569,24 @@ out_activate: #endif /* CONFIG_SMP */ if (old_state == TASK_UNINTERRUPTIBLE) { rq->nr_uninterruptible--; + vx_uninterruptible_dec(p); /* * Tasks on involuntary sleep don't earn * sleep_avg beyond just interactive state. */ - p->activated = -1; - } + p->sleep_type = SLEEP_NONINTERACTIVE; + } else + + /* + * Tasks that have marked their sleep as noninteractive get + * woken up with their sleep average not weighted in an + * interactive way. + */ + if (old_state & TASK_NONINTERACTIVE) + p->sleep_type = SLEEP_NONINTERACTIVE; + + activate_task(p, rq, cpu == this_cpu); /* * Sync wakeups (i.e. those types of wakeups where the waker * has indicated that it will leave the CPU in short order) @@ -1173,10 +1595,6 @@ out_activate: * the waker guarantees that the freshly woken up task is going * to be considered on this CPU.) */ - activate_task(p, rq, cpu == this_cpu); - /* this is to get the accounting behind the load update */ - if (old_state == TASK_UNINTERRUPTIBLE) - vx_uninterruptible_dec(p); if (!sync || cpu != this_cpu) { if (TASK_PREEMPTS_CURR(p, rq)) resched_task(rq->curr); @@ -1191,30 +1609,31 @@ out: return success; } -int fastcall wake_up_process(task_t * p) +int fastcall wake_up_process(struct task_struct *p) { return try_to_wake_up(p, TASK_STOPPED | TASK_TRACED | TASK_INTERRUPTIBLE | TASK_UNINTERRUPTIBLE, 0); } - EXPORT_SYMBOL(wake_up_process); -int fastcall wake_up_state(task_t *p, unsigned int state) +int fastcall wake_up_state(struct task_struct *p, unsigned int state) { return try_to_wake_up(p, state, 0); } -#ifdef CONFIG_SMP -static int find_idlest_cpu(struct task_struct *p, int this_cpu, - struct sched_domain *sd); -#endif - /* * Perform scheduler related setup for a newly forked process p. * p is forked by current. */ -void fastcall sched_fork(task_t *p) +void fastcall sched_fork(struct task_struct *p, int clone_flags) { + int cpu = get_cpu(); + +#ifdef CONFIG_SMP + cpu = sched_balance_self(cpu, SD_BALANCE_FORK); +#endif + set_task_cpu(p, cpu); + /* * We mark the process as running here, but have not actually * inserted it onto the runqueue yet. This guarantees that @@ -1222,20 +1641,24 @@ void fastcall sched_fork(task_t *p) * event cannot wake it up and insert it on the runqueue either. */ p->state = TASK_RUNNING; + + /* + * Make sure we do not leak PI boosting priority to the child: + */ + p->prio = current->normal_prio; + INIT_LIST_HEAD(&p->run_list); p->array = NULL; - spin_lock_init(&p->switch_lock); -#ifdef CONFIG_SCHEDSTATS - memset(&p->sched_info, 0, sizeof(p->sched_info)); +#if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) + if (unlikely(sched_info_on())) + memset(&p->sched_info, 0, sizeof(p->sched_info)); +#endif +#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW) + p->oncpu = 0; #endif #ifdef CONFIG_PREEMPT - /* - * During context-switch we hold precisely one spinlock, which - * schedule_tail drops. (in the common case it's this_rq()->lock, - * but it also can be p->switch_lock.) So we compensate with a count - * of 1. Also, we want to start with kernel preemption disabled. - */ - p->thread_info->preempt_count = 1; + /* Want to start with kernel preemption disabled. */ + task_thread_info(p)->preempt_count = 1; #endif /* * Share the timeslice between parent and child, thus the @@ -1258,12 +1681,10 @@ void fastcall sched_fork(task_t *p) * runqueue lock is not a problem. */ current->time_slice = 1; - preempt_disable(); scheduler_tick(); - local_irq_enable(); - preempt_enable(); - } else - local_irq_enable(); + } + local_irq_enable(); + put_cpu(); } /* @@ -1273,17 +1694,16 @@ void fastcall sched_fork(task_t *p) * 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(task_t * p, unsigned long clone_flags) +void fastcall wake_up_new_task(struct task_struct *p, unsigned long clone_flags) { + struct rq *rq, *this_rq; unsigned long flags; int this_cpu, cpu; - runqueue_t *rq, *this_rq; rq = task_rq_lock(p, &flags); - cpu = task_cpu(p); - this_cpu = smp_processor_id(); - BUG_ON(p->state != TASK_RUNNING); + this_cpu = smp_processor_id(); + cpu = task_cpu(p); /* * We decrease the sleep average of forking parents @@ -1309,10 +1729,11 @@ void fastcall wake_up_new_task(task_t * p, unsigned long clone_flags) else { p->prio = current->prio; BUG_ON(p->state & TASK_ONHOLD); + p->normal_prio = current->normal_prio; list_add_tail(&p->run_list, ¤t->run_list); p->array = current->array; p->array->nr_active++; - rq->nr_running++; + inc_nr_running(p, rq); } set_need_resched(); } else @@ -1359,17 +1780,17 @@ void fastcall wake_up_new_task(task_t * p, unsigned long clone_flags) * artificially, because any timeslice recovered here * was given away by the parent in the first place.) */ -void fastcall sched_exit(task_t * p) +void fastcall sched_exit(struct task_struct *p) { unsigned long flags; - runqueue_t *rq; + struct rq *rq; /* * If the child was a (relative-) CPU hog then decrease * the sleep_avg of the parent as well. */ rq = task_rq_lock(p->parent, &flags); - if (p->first_time_slice) { + if (p->first_time_slice && task_cpu(p) == task_cpu(p->parent)) { p->parent->time_slice += p->time_slice; if (unlikely(p->parent->time_slice > task_timeslice(p))) p->parent->time_slice = task_timeslice(p); @@ -1381,23 +1802,42 @@ void fastcall sched_exit(task_t * p) task_rq_unlock(rq, &flags); } +/** + * prepare_task_switch - prepare to switch tasks + * @rq: the runqueue preparing to switch + * @next: the task we are going to switch to. + * + * This is called with the rq lock held and interrupts off. It must + * be paired with a subsequent finish_task_switch after the context + * switch. + * + * prepare_task_switch sets up locking and calls architecture specific + * hooks. + */ +static inline void prepare_task_switch(struct rq *rq, struct task_struct *next) +{ + prepare_lock_switch(rq, next); + prepare_arch_switch(next); +} + /** * finish_task_switch - clean up after a task-switch + * @rq: runqueue associated with task-switch * @prev: the thread we just switched away from. * - * We enter this with the runqueue still locked, and finish_arch_switch() - * will unlock it along with doing any other architecture-specific cleanup - * actions. + * finish_task_switch must be called after the context switch, paired + * with a prepare_task_switch call before the context switch. + * finish_task_switch will reconcile locking set up by prepare_task_switch, + * and do any other architecture-specific cleanup actions. * * Note that we may have delayed dropping an mm in context_switch(). If * so, we finish that here outside of the runqueue lock. (Doing it * with the lock held can cause deadlocks; see schedule() for * details.) */ -static inline void finish_task_switch(task_t *prev) +static inline void finish_task_switch(struct rq *rq, struct task_struct *prev) __releases(rq->lock) { - runqueue_t *rq = this_rq(); struct mm_struct *mm = rq->prev_mm; unsigned long prev_task_flags; @@ -1415,22 +1855,34 @@ static inline void finish_task_switch(task_t *prev) * Manfred Spraul */ prev_task_flags = prev->flags; - finish_arch_switch(rq, prev); + finish_arch_switch(prev); + finish_lock_switch(rq, prev); if (mm) mmdrop(mm); - if (unlikely(prev_task_flags & PF_DEAD)) + if (unlikely(prev_task_flags & PF_DEAD)) { + /* + * Remove function-return probe instances associated with this + * task and put them back on the free list. + */ + kprobe_flush_task(prev); put_task_struct(prev); + } } /** * schedule_tail - first thing a freshly forked thread must call. * @prev: the thread we just switched away from. */ -asmlinkage void schedule_tail(task_t *prev) +asmlinkage void schedule_tail(struct task_struct *prev) __releases(rq->lock) { - finish_task_switch(prev); + struct rq *rq = this_rq(); + finish_task_switch(rq, prev); +#ifdef __ARCH_WANT_UNLOCKED_CTXSW + /* In this case, finish_task_switch does not reenable preemption */ + preempt_enable(); +#endif if (current->set_child_tid) put_user(current->pid, current->set_child_tid); } @@ -1439,8 +1891,9 @@ asmlinkage void schedule_tail(task_t *prev) * context_switch - switch to the new MM and the new * thread's register state. */ -static inline -task_t * context_switch(runqueue_t *rq, task_t *prev, task_t *next) +static inline struct task_struct * +context_switch(struct rq *rq, struct task_struct *prev, + struct task_struct *next) { struct mm_struct *mm = next->mm; struct mm_struct *oldmm = prev->active_mm; @@ -1457,6 +1910,15 @@ task_t * context_switch(runqueue_t *rq, task_t *prev, task_t *next) WARN_ON(rq->prev_mm); rq->prev_mm = oldmm; } + /* + * Since the runqueue lock will be released by the next + * task (which is an invalid locking op but in the case + * of the scheduler it's an obvious special-case), so we + * do an early lockdep release here: + */ +#ifndef __ARCH_WANT_UNLOCKED_CTXSW + spin_release(&rq->lock.dep_map, 1, _THIS_IP_); +#endif /* Here we just switch the register state and the stack. */ switch_to(prev, next, prev); @@ -1485,7 +1947,7 @@ unsigned long nr_uninterruptible(void) { unsigned long i, sum = 0; - for_each_cpu(i) + for_each_possible_cpu(i) sum += cpu_rq(i)->nr_uninterruptible; /* @@ -1500,9 +1962,10 @@ unsigned long nr_uninterruptible(void) unsigned long long nr_context_switches(void) { - unsigned long long i, sum = 0; + int i; + unsigned long long sum = 0; - for_each_cpu(i) + for_each_possible_cpu(i) sum += cpu_rq(i)->nr_switches; return sum; @@ -1512,21 +1975,45 @@ unsigned long nr_iowait(void) { unsigned long i, sum = 0; - for_each_cpu(i) + for_each_possible_cpu(i) sum += atomic_read(&cpu_rq(i)->nr_iowait); return sum; } +unsigned long nr_active(void) +{ + unsigned long i, running = 0, uninterruptible = 0; + + for_each_online_cpu(i) { + running += cpu_rq(i)->nr_running; + uninterruptible += cpu_rq(i)->nr_uninterruptible; + } + + if (unlikely((long)uninterruptible < 0)) + uninterruptible = 0; + + return running + uninterruptible; +} + #ifdef CONFIG_SMP +/* + * Is this task likely cache-hot: + */ +static inline int +task_hot(struct task_struct *p, unsigned long long now, struct sched_domain *sd) +{ + return (long long)(now - p->last_ran) < (long long)sd->cache_hot_time; +} + /* * double_rq_lock - safely lock two runqueues * * Note this does not disable interrupts like task_rq_lock, * you need to do so manually before calling. */ -static void double_rq_lock(runqueue_t *rq1, runqueue_t *rq2) +static void double_rq_lock(struct rq *rq1, struct rq *rq2) __acquires(rq1->lock) __acquires(rq2->lock) { @@ -1550,7 +2037,7 @@ static void double_rq_lock(runqueue_t *rq1, runqueue_t *rq2) * Note this does not restore interrupts like task_rq_unlock, * you need to do so manually after calling. */ -static void double_rq_unlock(runqueue_t *rq1, runqueue_t *rq2) +static void double_rq_unlock(struct rq *rq1, struct rq *rq2) __releases(rq1->lock) __releases(rq2->lock) { @@ -1564,7 +2051,7 @@ static void double_rq_unlock(runqueue_t *rq1, runqueue_t *rq2) /* * double_lock_balance - lock the busiest runqueue, this_rq is locked already. */ -static void double_lock_balance(runqueue_t *this_rq, runqueue_t *busiest) +static void double_lock_balance(struct rq *this_rq, struct rq *busiest) __releases(this_rq->lock) __acquires(busiest->lock) __acquires(this_rq->lock) @@ -1579,62 +2066,17 @@ static void double_lock_balance(runqueue_t *this_rq, runqueue_t *busiest) } } -/* - * find_idlest_cpu - find the least busy runqueue. - */ -static int find_idlest_cpu(struct task_struct *p, int this_cpu, - struct sched_domain *sd) -{ - unsigned long load, min_load, this_load; - int i, min_cpu; - cpumask_t mask; - - min_cpu = UINT_MAX; - min_load = ULONG_MAX; - - cpus_and(mask, sd->span, p->cpus_allowed); - - for_each_cpu_mask(i, mask) { - load = target_load(i); - - if (load < min_load) { - min_cpu = i; - min_load = load; - - /* break out early on an idle CPU: */ - if (!min_load) - break; - } - } - - /* add +1 to account for the new task */ - this_load = source_load(this_cpu) + SCHED_LOAD_SCALE; - - /* - * Would with the addition of the new task to the - * current CPU there be an imbalance between this - * CPU and the idlest CPU? - * - * Use half of the balancing threshold - new-context is - * a good opportunity to balance. - */ - if (min_load*(100 + (sd->imbalance_pct-100)/2) < this_load*100) - return min_cpu; - - return this_cpu; -} - /* * If dest_cpu is allowed for this process, migrate the task to it. * This is accomplished by forcing the cpu_allowed mask to only * allow dest_cpu, which will force the cpu onto dest_cpu. Then * the cpu_allowed mask is restored. */ -static void sched_migrate_task(task_t *p, int dest_cpu) +static void sched_migrate_task(struct task_struct *p, int dest_cpu) { - migration_req_t req; - runqueue_t *rq; + struct migration_req req; unsigned long flags; + struct rq *rq; rq = task_rq_lock(p, &flags); if (!cpu_isset(dest_cpu, p->cpus_allowed) @@ -1645,11 +2087,13 @@ static void sched_migrate_task(task_t *p, int dest_cpu) if (migrate_task(p, dest_cpu, &req)) { /* Need to wait for migration thread (might exit: take ref). */ struct task_struct *mt = rq->migration_thread; + get_task_struct(mt); task_rq_unlock(rq, &flags); wake_up_process(mt); put_task_struct(mt); wait_for_completion(&req.done); + return; } out: @@ -1657,51 +2101,30 @@ out: } /* - * sched_exec(): find the highest-level, exec-balance-capable - * domain and try to migrate the task to the least loaded CPU. - * - * execve() is a valuable balancing opportunity, because at this point - * the task has the smallest effective memory and cache footprint. + * sched_exec - execve() is a valuable balancing opportunity, because at + * this point the task has the smallest effective memory and cache footprint. */ void sched_exec(void) { - struct sched_domain *tmp, *sd = NULL; int new_cpu, this_cpu = get_cpu(); - - /* Prefer the current CPU if there's only this task running */ - if (this_rq()->nr_running <= 1) - goto out; - - for_each_domain(this_cpu, tmp) - if (tmp->flags & SD_BALANCE_EXEC) - sd = tmp; - - if (sd) { - schedstat_inc(sd, sbe_attempts); - new_cpu = find_idlest_cpu(current, this_cpu, sd); - if (new_cpu != this_cpu) { - schedstat_inc(sd, sbe_pushed); - put_cpu(); - sched_migrate_task(current, new_cpu); - return; - } - } -out: + new_cpu = sched_balance_self(this_cpu, SD_BALANCE_EXEC); put_cpu(); + if (new_cpu != this_cpu) + sched_migrate_task(current, new_cpu); } /* * pull_task - move a task from a remote runqueue to the local runqueue. * Both runqueues must be locked. */ -static inline -void pull_task(runqueue_t *src_rq, prio_array_t *src_array, task_t *p, - runqueue_t *this_rq, prio_array_t *this_array, int this_cpu) +static void pull_task(struct rq *src_rq, struct prio_array *src_array, + struct task_struct *p, struct rq *this_rq, + struct prio_array *this_array, int this_cpu) { dequeue_task(p, src_array); - src_rq->nr_running--; + dec_nr_running(p, src_rq); set_task_cpu(p, this_cpu); - this_rq->nr_running++; + inc_nr_running(p, this_rq); enqueue_task(p, this_array); p->timestamp = (p->timestamp - src_rq->timestamp_last_tick) + this_rq->timestamp_last_tick; @@ -1716,9 +2139,10 @@ void pull_task(runqueue_t *src_rq, prio_array_t *src_array, task_t *p, /* * can_migrate_task - may task p from runqueue rq be migrated to this_cpu? */ -static inline -int can_migrate_task(task_t *p, runqueue_t *rq, int this_cpu, - struct sched_domain *sd, enum idle_type idle) +static +int can_migrate_task(struct task_struct *p, struct rq *rq, int this_cpu, + struct sched_domain *sd, enum idle_type idle, + int *all_pinned) { /* * We do not migrate tasks that are: @@ -1726,45 +2150,64 @@ int can_migrate_task(task_t *p, runqueue_t *rq, int this_cpu, * 2) cannot be migrated to this CPU due to cpus_allowed, or * 3) are cache-hot on their current CPU. */ - if (task_running(rq, p)) - return 0; if (!cpu_isset(this_cpu, p->cpus_allowed)) return 0; + *all_pinned = 0; + + if (task_running(rq, p)) + return 0; /* * Aggressive migration if: - * 1) the [whole] cpu is idle, or + * 1) task is cache cold, or * 2) too many balance attempts have failed. */ - if (cpu_and_siblings_are_idle(this_cpu) || \ - sd->nr_balance_failed > sd->cache_nice_tries) + if (sd->nr_balance_failed > sd->cache_nice_tries) return 1; if (task_hot(p, rq->timestamp_last_tick, sd)) - return 0; + return 0; return 1; } +#define rq_best_prio(rq) min((rq)->curr->prio, (rq)->best_expired_prio) + /* - * move_tasks tries to move up to max_nr_move tasks from busiest to this_rq, - * as part of a balancing operation within "domain". Returns the number of - * tasks moved. + * move_tasks tries to move up to max_nr_move tasks and max_load_move weighted + * load from busiest to this_rq, as part of a balancing operation within + * "domain". Returns the number of tasks moved. * * Called with both runqueues locked. */ -static int move_tasks(runqueue_t *this_rq, int this_cpu, runqueue_t *busiest, - unsigned long max_nr_move, struct sched_domain *sd, - enum idle_type idle) -{ - prio_array_t *array, *dst_array; +static int move_tasks(struct rq *this_rq, int this_cpu, struct rq *busiest, + unsigned long max_nr_move, unsigned long max_load_move, + struct sched_domain *sd, enum idle_type idle, + int *all_pinned) +{ + int idx, pulled = 0, pinned = 0, this_best_prio, best_prio, + best_prio_seen, skip_for_load; + struct prio_array *array, *dst_array; struct list_head *head, *curr; - int idx, pulled = 0; - task_t *tmp; + struct task_struct *tmp; + long rem_load_move; - if (max_nr_move <= 0 || busiest->nr_running <= 1) + if (max_nr_move == 0 || max_load_move == 0) goto out; + rem_load_move = max_load_move; + pinned = 1; + this_best_prio = rq_best_prio(this_rq); + best_prio = rq_best_prio(busiest); + /* + * Enable handling of the case where there is more than one task + * with the best priority. If the current running task is one + * of those with prio==best_prio we know it won't be moved + * and therefore it's safe to override the skip (based on load) of + * any task we find with that prio. + */ + best_prio_seen = best_prio == busiest->curr->prio; + /* * We first consider expired tasks. Those will likely not be * executed in the near future, and they are most likely to @@ -1799,11 +2242,22 @@ skip_bitmap: head = array->queue + idx; curr = head->prev; skip_queue: - tmp = list_entry(curr, task_t, run_list); + tmp = list_entry(curr, struct task_struct, run_list); curr = curr->prev; - if (!can_migrate_task(tmp, busiest, this_cpu, sd, idle)) { + /* + * To help distribute high priority tasks accross CPUs we don't + * skip a task if it will be the highest priority task (i.e. smallest + * prio value) on its new queue regardless of its load weight + */ + skip_for_load = tmp->load_weight > rem_load_move; + if (skip_for_load && idx < this_best_prio) + skip_for_load = !best_prio_seen && idx == best_prio; + if (skip_for_load || + !can_migrate_task(tmp, busiest, this_cpu, sd, idle, &pinned)) { + + best_prio_seen |= idx == best_prio; if (curr != head) goto skip_queue; idx++; @@ -1817,9 +2271,15 @@ skip_queue: pull_task(busiest, array, tmp, this_rq, dst_array, this_cpu); pulled++; + rem_load_move -= tmp->load_weight; - /* We only want to steal up to the prescribed number of tasks. */ - if (pulled < max_nr_move) { + /* + * We only want to steal up to the prescribed number of tasks + * and the prescribed amount of weighted load. + */ + if (pulled < max_nr_move && rem_load_move > 0) { + if (idx < this_best_prio) + this_best_prio = idx; if (curr != head) goto skip_queue; idx++; @@ -1832,41 +2292,76 @@ out: * inside pull_task(). */ schedstat_add(sd, lb_gained[idle], pulled); + + if (all_pinned) + *all_pinned = pinned; return pulled; } /* * find_busiest_group finds and returns the busiest CPU group within the - * domain. It calculates and returns the number of tasks which should be - * moved to restore balance via the imbalance parameter. + * domain. It calculates and returns the amount of weighted load which + * should be moved to restore balance via the imbalance parameter. */ static struct sched_group * find_busiest_group(struct sched_domain *sd, int this_cpu, - unsigned long *imbalance, enum idle_type idle) + unsigned long *imbalance, enum idle_type idle, int *sd_idle, + cpumask_t *cpus) { struct sched_group *busiest = NULL, *this = NULL, *group = sd->groups; unsigned long max_load, avg_load, total_load, this_load, total_pwr; + unsigned long max_pull; + unsigned long busiest_load_per_task, busiest_nr_running; + unsigned long this_load_per_task, this_nr_running; + int load_idx; +#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) + int power_savings_balance = 1; + unsigned long leader_nr_running = 0, min_load_per_task = 0; + unsigned long min_nr_running = ULONG_MAX; + struct sched_group *group_min = NULL, *group_leader = NULL; +#endif max_load = this_load = total_load = total_pwr = 0; + busiest_load_per_task = busiest_nr_running = 0; + this_load_per_task = this_nr_running = 0; + if (idle == NOT_IDLE) + load_idx = sd->busy_idx; + else if (idle == NEWLY_IDLE) + load_idx = sd->newidle_idx; + else + load_idx = sd->idle_idx; do { - unsigned long load; + unsigned long load, group_capacity; int local_group; int i; + unsigned long sum_nr_running, sum_weighted_load; local_group = cpu_isset(this_cpu, group->cpumask); /* Tally up the load of all CPUs in the group */ - avg_load = 0; + sum_weighted_load = sum_nr_running = avg_load = 0; for_each_cpu_mask(i, group->cpumask) { + struct rq *rq; + + if (!cpu_isset(i, *cpus)) + continue; + + rq = cpu_rq(i); + + if (*sd_idle && !idle_cpu(i)) + *sd_idle = 0; + /* Bias balancing toward cpus of our domain */ if (local_group) - load = target_load(i); + load = target_load(i, load_idx); else - load = source_load(i); + load = source_load(i, load_idx); avg_load += load; + sum_nr_running += rq->nr_running; + sum_weighted_load += rq->raw_weighted_load; } total_load += avg_load; @@ -1875,19 +2370,80 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, /* Adjust by relative CPU power of the group */ avg_load = (avg_load * SCHED_LOAD_SCALE) / group->cpu_power; + group_capacity = group->cpu_power / SCHED_LOAD_SCALE; + if (local_group) { this_load = avg_load; this = group; - goto nextgroup; - } else if (avg_load > max_load) { + this_nr_running = sum_nr_running; + this_load_per_task = sum_weighted_load; + } else if (avg_load > max_load && + sum_nr_running > group_capacity) { max_load = avg_load; busiest = group; + busiest_nr_running = sum_nr_running; + busiest_load_per_task = sum_weighted_load; } -nextgroup: + +#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) + /* + * Busy processors will not participate in power savings + * balance. + */ + if (idle == NOT_IDLE || !(sd->flags & SD_POWERSAVINGS_BALANCE)) + goto group_next; + + /* + * If the local group is idle or completely loaded + * no need to do power savings balance at this domain + */ + if (local_group && (this_nr_running >= group_capacity || + !this_nr_running)) + power_savings_balance = 0; + + /* + * If a group is already running at full capacity or idle, + * don't include that group in power savings calculations + */ + if (!power_savings_balance || sum_nr_running >= group_capacity + || !sum_nr_running) + goto group_next; + + /* + * Calculate the group which has the least non-idle load. + * This is the group from where we need to pick up the load + * for saving power + */ + if ((sum_nr_running < min_nr_running) || + (sum_nr_running == min_nr_running && + first_cpu(group->cpumask) < + first_cpu(group_min->cpumask))) { + group_min = group; + min_nr_running = sum_nr_running; + min_load_per_task = sum_weighted_load / + sum_nr_running; + } + + /* + * Calculate the group which is almost near its + * capacity but still has some space to pick up some load + * from other group and save more power + */ + if (sum_nr_running <= group_capacity - 1) { + if (sum_nr_running > leader_nr_running || + (sum_nr_running == leader_nr_running && + first_cpu(group->cpumask) > + first_cpu(group_leader->cpumask))) { + group_leader = group; + leader_nr_running = sum_nr_running; + } + } +group_next: +#endif group = group->next; } while (group != sd->groups); - if (!busiest || this_load >= max_load) + if (!busiest || this_load >= max_load || busiest_nr_running == 0) goto out_balanced; avg_load = (SCHED_LOAD_SCALE * total_load) / total_pwr; @@ -1896,6 +2452,7 @@ nextgroup: 100*max_load <= sd->imbalance_pct*this_load) goto out_balanced; + busiest_load_per_task /= busiest_nr_running; /* * We're trying to get all the cpus to the average_load, so we don't * want to push ourselves above the average load, nor do we wish to @@ -1907,17 +2464,49 @@ nextgroup: * by pulling tasks to us. Be careful of negative numbers as they'll * appear as very large values with unsigned longs. */ + if (max_load <= busiest_load_per_task) + goto out_balanced; + + /* + * In the presence of smp nice balancing, certain scenarios can have + * max load less than avg load(as we skip the groups at or below + * its cpu_power, while calculating max_load..) + */ + if (max_load < avg_load) { + *imbalance = 0; + goto small_imbalance; + } + + /* Don't want to pull so many tasks that a group would go idle */ + max_pull = min(max_load - avg_load, max_load - busiest_load_per_task); + /* How much load to actually move to equalise the imbalance */ - *imbalance = min((max_load - avg_load) * busiest->cpu_power, + *imbalance = min(max_pull * busiest->cpu_power, (avg_load - this_load) * this->cpu_power) / SCHED_LOAD_SCALE; - if (*imbalance < SCHED_LOAD_SCALE) { - unsigned long pwr_now = 0, pwr_move = 0; - unsigned long tmp; + /* + * if *imbalance is less than the average load per runnable task + * there is no gaurantee that any tasks will be moved so we'll have + * a think about bumping its value to force at least one task to be + * moved + */ + if (*imbalance < busiest_load_per_task) { + unsigned long tmp, pwr_now, pwr_move; + unsigned int imbn; + +small_imbalance: + pwr_move = pwr_now = 0; + imbn = 2; + if (this_nr_running) { + this_load_per_task /= this_nr_running; + if (busiest_load_per_task > this_load_per_task) + imbn = 1; + } else + this_load_per_task = SCHED_LOAD_SCALE; - if (max_load - this_load >= SCHED_LOAD_SCALE*2) { - *imbalance = 1; + if (max_load - this_load >= busiest_load_per_task * imbn) { + *imbalance = busiest_load_per_task; return busiest; } @@ -1927,45 +2516,47 @@ nextgroup: * moving them. */ - pwr_now += busiest->cpu_power*min(SCHED_LOAD_SCALE, max_load); - pwr_now += this->cpu_power*min(SCHED_LOAD_SCALE, this_load); + pwr_now += busiest->cpu_power * + min(busiest_load_per_task, max_load); + pwr_now += this->cpu_power * + min(this_load_per_task, this_load); pwr_now /= SCHED_LOAD_SCALE; /* Amount of load we'd subtract */ - tmp = SCHED_LOAD_SCALE*SCHED_LOAD_SCALE/busiest->cpu_power; + tmp = busiest_load_per_task*SCHED_LOAD_SCALE/busiest->cpu_power; if (max_load > tmp) - pwr_move += busiest->cpu_power*min(SCHED_LOAD_SCALE, - max_load - tmp); + pwr_move += busiest->cpu_power * + min(busiest_load_per_task, max_load - tmp); /* Amount of load we'd add */ if (max_load*busiest->cpu_power < - SCHED_LOAD_SCALE*SCHED_LOAD_SCALE) + busiest_load_per_task*SCHED_LOAD_SCALE) tmp = max_load*busiest->cpu_power/this->cpu_power; else - tmp = SCHED_LOAD_SCALE*SCHED_LOAD_SCALE/this->cpu_power; - pwr_move += this->cpu_power*min(SCHED_LOAD_SCALE, this_load + tmp); + tmp = busiest_load_per_task*SCHED_LOAD_SCALE/this->cpu_power; + pwr_move += this->cpu_power*min(this_load_per_task, this_load + tmp); pwr_move /= SCHED_LOAD_SCALE; /* Move if we gain throughput */ if (pwr_move <= pwr_now) goto out_balanced; - *imbalance = 1; - return busiest; + *imbalance = busiest_load_per_task; } - /* Get rid of the scaling factor, rounding down as we divide */ - *imbalance = *imbalance / SCHED_LOAD_SCALE; - return busiest; out_balanced: - if (busiest && (idle == NEWLY_IDLE || - (idle == SCHED_IDLE && max_load > SCHED_LOAD_SCALE)) ) { - *imbalance = 1; - return busiest; - } +#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) + if (idle == NOT_IDLE || !(sd->flags & SD_POWERSAVINGS_BALANCE)) + goto ret; + if (this == group_leader && group_leader != group_min) { + *imbalance = min_load_per_task; + return group_min; + } +ret: +#endif *imbalance = 0; return NULL; } @@ -1973,62 +2564,80 @@ out_balanced: /* * find_busiest_queue - find the busiest runqueue among the cpus in group. */ -static runqueue_t *find_busiest_queue(struct sched_group *group) +static struct rq * +find_busiest_queue(struct sched_group *group, enum idle_type idle, + unsigned long imbalance, cpumask_t *cpus) { - unsigned long load, max_load = 0; - runqueue_t *busiest = NULL; + struct rq *busiest = NULL, *rq; + unsigned long max_load = 0; int i; for_each_cpu_mask(i, group->cpumask) { - load = source_load(i); - if (load > max_load) { - max_load = load; - busiest = cpu_rq(i); + if (!cpu_isset(i, *cpus)) + continue; + + rq = cpu_rq(i); + + if (rq->nr_running == 1 && rq->raw_weighted_load > imbalance) + continue; + + if (rq->raw_weighted_load > max_load) { + max_load = rq->raw_weighted_load; + busiest = rq; } } return busiest; } +/* + * Max backoff if we encounter pinned tasks. Pretty arbitrary value, but + * so long as it is large enough. + */ +#define MAX_PINNED_INTERVAL 512 + +static inline unsigned long minus_1_or_zero(unsigned long n) +{ + return n > 0 ? n - 1 : 0; +} + /* * Check this_cpu to ensure it is balanced within domain. Attempt to move * tasks if there is an imbalance. * * Called with this_rq unlocked. */ -static int load_balance(int this_cpu, runqueue_t *this_rq, +static int load_balance(int this_cpu, struct rq *this_rq, struct sched_domain *sd, enum idle_type idle) { + int nr_moved, all_pinned = 0, active_balance = 0, sd_idle = 0; struct sched_group *group; - runqueue_t *busiest; unsigned long imbalance; - int nr_moved; + struct rq *busiest; + cpumask_t cpus = CPU_MASK_ALL; + + if (idle != NOT_IDLE && sd->flags & SD_SHARE_CPUPOWER && + !sched_smt_power_savings) + sd_idle = 1; - spin_lock(&this_rq->lock); schedstat_inc(sd, lb_cnt[idle]); - group = find_busiest_group(sd, this_cpu, &imbalance, idle); +redo: + group = find_busiest_group(sd, this_cpu, &imbalance, idle, &sd_idle, + &cpus); if (!group) { schedstat_inc(sd, lb_nobusyg[idle]); goto out_balanced; } - busiest = find_busiest_queue(group); + busiest = find_busiest_queue(group, idle, imbalance, &cpus); if (!busiest) { schedstat_inc(sd, lb_nobusyq[idle]); goto out_balanced; } - /* - * This should be "impossible", but since load - * balancing is inherently racy and statistical, - * it could happen in theory. - */ - if (unlikely(busiest == this_rq)) { - WARN_ON(1); - goto out_balanced; - } + BUG_ON(busiest == this_rq); schedstat_add(sd, lb_imbalance[idle], imbalance); @@ -2040,61 +2649,89 @@ static int load_balance(int this_cpu, runqueue_t *this_rq, * still unbalanced. nr_moved simply stays zero, so it is * correctly treated as an imbalance. */ - double_lock_balance(this_rq, busiest); + double_rq_lock(this_rq, busiest); nr_moved = move_tasks(this_rq, this_cpu, busiest, - imbalance, sd, idle); - spin_unlock(&busiest->lock); + minus_1_or_zero(busiest->nr_running), + imbalance, sd, idle, &all_pinned); + double_rq_unlock(this_rq, busiest); + + /* All tasks on this runqueue were pinned by CPU affinity */ + if (unlikely(all_pinned)) { + cpu_clear(cpu_of(busiest), cpus); + if (!cpus_empty(cpus)) + goto redo; + goto out_balanced; + } } - spin_unlock(&this_rq->lock); if (!nr_moved) { schedstat_inc(sd, lb_failed[idle]); sd->nr_balance_failed++; if (unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2)) { - int wake = 0; spin_lock(&busiest->lock); + + /* don't kick the migration_thread, if the curr + * task on busiest cpu can't be moved to this_cpu + */ + if (!cpu_isset(this_cpu, busiest->curr->cpus_allowed)) { + spin_unlock(&busiest->lock); + all_pinned = 1; + goto out_one_pinned; + } + if (!busiest->active_balance) { busiest->active_balance = 1; busiest->push_cpu = this_cpu; - wake = 1; + active_balance = 1; } spin_unlock(&busiest->lock); - if (wake) + if (active_balance) wake_up_process(busiest->migration_thread); /* * We've kicked active balancing, reset the failure * counter. */ - sd->nr_balance_failed = sd->cache_nice_tries; + sd->nr_balance_failed = sd->cache_nice_tries+1; } - - /* - * We were unbalanced, but unsuccessful in move_tasks(), - * so bump the balance_interval to lessen the lock contention. - */ - if (sd->balance_interval < sd->max_interval) - sd->balance_interval++; - } else { + } else sd->nr_balance_failed = 0; + if (likely(!active_balance)) { /* We were unbalanced, so reset the balancing interval */ sd->balance_interval = sd->min_interval; + } else { + /* + * If we've begun active balancing, start to back off. This + * case may not be covered by the all_pinned logic if there + * is only 1 task on the busy runqueue (because we don't call + * move_tasks). + */ + if (sd->balance_interval < sd->max_interval) + sd->balance_interval *= 2; } + if (!nr_moved && !sd_idle && sd->flags & SD_SHARE_CPUPOWER && + !sched_smt_power_savings) + return -1; return nr_moved; out_balanced: - spin_unlock(&this_rq->lock); - schedstat_inc(sd, lb_balanced[idle]); + sd->nr_balance_failed = 0; + +out_one_pinned: /* tune up the balancing interval */ - if (sd->balance_interval < sd->max_interval) + if ((all_pinned && sd->balance_interval < MAX_PINNED_INTERVAL) || + (sd->balance_interval < sd->max_interval)) sd->balance_interval *= 2; + if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER && + !sched_smt_power_savings) + return -1; return 0; } @@ -2105,58 +2742,87 @@ out_balanced: * Called from schedule when this_rq is about to become idle (NEWLY_IDLE). * this_rq is locked. */ -static int load_balance_newidle(int this_cpu, runqueue_t *this_rq, - struct sched_domain *sd) +static int +load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd) { struct sched_group *group; - runqueue_t *busiest = NULL; + struct rq *busiest = NULL; unsigned long imbalance; int nr_moved = 0; + int sd_idle = 0; + cpumask_t cpus = CPU_MASK_ALL; + + if (sd->flags & SD_SHARE_CPUPOWER && !sched_smt_power_savings) + sd_idle = 1; schedstat_inc(sd, lb_cnt[NEWLY_IDLE]); - group = find_busiest_group(sd, this_cpu, &imbalance, NEWLY_IDLE); +redo: + group = find_busiest_group(sd, this_cpu, &imbalance, NEWLY_IDLE, + &sd_idle, &cpus); if (!group) { - schedstat_inc(sd, lb_balanced[NEWLY_IDLE]); schedstat_inc(sd, lb_nobusyg[NEWLY_IDLE]); - goto out; + goto out_balanced; } - busiest = find_busiest_queue(group); - if (!busiest || busiest == this_rq) { - schedstat_inc(sd, lb_balanced[NEWLY_IDLE]); + busiest = find_busiest_queue(group, NEWLY_IDLE, imbalance, + &cpus); + if (!busiest) { schedstat_inc(sd, lb_nobusyq[NEWLY_IDLE]); - goto out; + goto out_balanced; } - /* Attempt to move tasks */ - double_lock_balance(this_rq, busiest); + BUG_ON(busiest == this_rq); schedstat_add(sd, lb_imbalance[NEWLY_IDLE], imbalance); - nr_moved = move_tasks(this_rq, this_cpu, busiest, - imbalance, sd, NEWLY_IDLE); - if (!nr_moved) - schedstat_inc(sd, lb_failed[NEWLY_IDLE]); - spin_unlock(&busiest->lock); + nr_moved = 0; + if (busiest->nr_running > 1) { + /* Attempt to move tasks */ + double_lock_balance(this_rq, busiest); + nr_moved = move_tasks(this_rq, this_cpu, busiest, + minus_1_or_zero(busiest->nr_running), + imbalance, sd, NEWLY_IDLE, NULL); + spin_unlock(&busiest->lock); + + if (!nr_moved) { + cpu_clear(cpu_of(busiest), cpus); + if (!cpus_empty(cpus)) + goto redo; + } + } + + if (!nr_moved) { + schedstat_inc(sd, lb_failed[NEWLY_IDLE]); + if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER) + return -1; + } else + sd->nr_balance_failed = 0; -out: return nr_moved; + +out_balanced: + schedstat_inc(sd, lb_balanced[NEWLY_IDLE]); + if (!sd_idle && sd->flags & SD_SHARE_CPUPOWER && + !sched_smt_power_savings) + return -1; + sd->nr_balance_failed = 0; + + return 0; } /* * idle_balance is called by schedule() if this_cpu is about to become * idle. Attempts to pull tasks from other CPUs. */ -static inline void idle_balance(int this_cpu, runqueue_t *this_rq) +static void idle_balance(int this_cpu, struct rq *this_rq) { struct sched_domain *sd; for_each_domain(this_cpu, sd) { if (sd->flags & SD_BALANCE_NEWIDLE) { - if (load_balance_newidle(this_cpu, this_rq, sd)) { - /* We've pulled tasks over so stop searching */ + /* If we've pulled tasks over stop searching: */ + if (load_balance_newidle(this_cpu, this_rq, sd)) break; - } } } } @@ -2169,59 +2835,46 @@ static inline void idle_balance(int this_cpu, runqueue_t *this_rq) * * Called with busiest_rq locked. */ -static void active_load_balance(runqueue_t *busiest_rq, int busiest_cpu) +static void active_load_balance(struct rq *busiest_rq, int busiest_cpu) { + int target_cpu = busiest_rq->push_cpu; struct sched_domain *sd; - struct sched_group *cpu_group; - runqueue_t *target_rq; - cpumask_t visited_cpus; - int cpu; + struct rq *target_rq; + + /* Is there any task to move? */ + if (busiest_rq->nr_running <= 1) + return; + + target_rq = cpu_rq(target_cpu); /* - * Search for suitable CPUs to push tasks to in successively higher - * domains with SD_LOAD_BALANCE set. + * This condition is "impossible", if it occurs + * we need to fix it. Originally reported by + * Bjorn Helgaas on a 128-cpu setup. */ - visited_cpus = CPU_MASK_NONE; - for_each_domain(busiest_cpu, sd) { - if (!(sd->flags & SD_LOAD_BALANCE)) - /* no more domains to search */ - break; + BUG_ON(busiest_rq == target_rq); + + /* move a task from busiest_rq to target_rq */ + double_lock_balance(busiest_rq, target_rq); + + /* Search for an sd spanning us and the target CPU. */ + for_each_domain(target_cpu, sd) { + if ((sd->flags & SD_LOAD_BALANCE) && + cpu_isset(busiest_cpu, sd->span)) + break; + } + if (likely(sd)) { schedstat_inc(sd, alb_cnt); - cpu_group = sd->groups; - do { - for_each_cpu_mask(cpu, cpu_group->cpumask) { - if (busiest_rq->nr_running <= 1) - /* no more tasks left to move */ - return; - if (cpu_isset(cpu, visited_cpus)) - continue; - cpu_set(cpu, visited_cpus); - if (!cpu_and_siblings_are_idle(cpu) || cpu == busiest_cpu) - continue; - - target_rq = cpu_rq(cpu); - /* - * This condition is "impossible", if it occurs - * we need to fix it. Originally reported by - * Bjorn Helgaas on a 128-cpu setup. - */ - BUG_ON(busiest_rq == target_rq); - - /* move a task from busiest_rq to target_rq */ - double_lock_balance(busiest_rq, target_rq); - if (move_tasks(target_rq, cpu, busiest_rq, - 1, sd, SCHED_IDLE)) { - schedstat_inc(sd, alb_pushed); - } else { - schedstat_inc(sd, alb_failed); - } - spin_unlock(&target_rq->lock); - } - cpu_group = cpu_group->next; - } while (cpu_group != sd->groups); + if (move_tasks(target_rq, target_cpu, busiest_rq, 1, + RTPRIO_TO_LOAD_WEIGHT(100), sd, SCHED_IDLE, + NULL)) + schedstat_inc(sd, alb_pushed); + else + schedstat_inc(sd, alb_failed); } + spin_unlock(&target_rq->lock); } /* @@ -2233,31 +2886,38 @@ static void active_load_balance(runqueue_t *busiest_rq, int busiest_cpu) * Balancing parameters are set up in arch_init_sched_domains. */ -/* Don't have all balancing operations going off at once */ -#define CPU_OFFSET(cpu) (HZ * cpu / NR_CPUS) +/* Don't have all balancing operations going off at once: */ +static inline unsigned long cpu_offset(int cpu) +{ + return jiffies + cpu * HZ / NR_CPUS; +} -static void rebalance_tick(int this_cpu, runqueue_t *this_rq, - enum idle_type idle) +static void +rebalance_tick(int this_cpu, struct rq *this_rq, enum idle_type idle) { - unsigned long old_load, this_load; - unsigned long j = jiffies + CPU_OFFSET(this_cpu); + unsigned long this_load, interval, j = cpu_offset(this_cpu); struct sched_domain *sd; + int i, scale; - /* Update our load */ - old_load = this_rq->cpu_load; - this_load = this_rq->nr_running * SCHED_LOAD_SCALE; - /* - * Round up the averaging division if load is increasing. This - * prevents us from getting stuck on 9 if the load is 10, for - * example. - */ - if (this_load > old_load) - old_load++; - this_rq->cpu_load = (old_load + this_load) / 2; + this_load = this_rq->raw_weighted_load; - for_each_domain(this_cpu, sd) { - unsigned long interval; + /* Update our load: */ + for (i = 0, scale = 1; i < 3; i++, scale <<= 1) { + unsigned long old_load, new_load; + + old_load = this_rq->cpu_load[i]; + new_load = this_load; + /* + * Round up the averaging division if load is increasing. This + * prevents us from getting stuck on 9 if the load is 10, for + * example. + */ + if (new_load > old_load) + new_load += scale-1; + this_rq->cpu_load[i] = (old_load*(scale-1) + new_load) / scale; + } + for_each_domain(this_cpu, sd) { if (!(sd->flags & SD_LOAD_BALANCE)) continue; @@ -2272,7 +2932,11 @@ static void rebalance_tick(int this_cpu, runqueue_t *this_rq, if (j - sd->last_balance >= interval) { if (load_balance(this_cpu, this_rq, sd, idle)) { - /* We've pulled tasks over so no longer idle */ + /* + * We've pulled tasks over so either we're no + * longer idle, or one of our SMT siblings is + * not idle. + */ idle = NOT_IDLE; } sd->last_balance += interval; @@ -2283,17 +2947,18 @@ static void rebalance_tick(int this_cpu, runqueue_t *this_rq, /* * on UP we do not need to balance between CPUs: */ -static inline void rebalance_tick(int cpu, runqueue_t *rq, enum idle_type idle) +static inline void rebalance_tick(int cpu, struct rq *rq, enum idle_type idle) { } -static inline void idle_balance(int cpu, runqueue_t *rq) +static inline void idle_balance(int cpu, struct rq *rq) { } #endif -static inline int wake_priority_sleeper(runqueue_t *rq) +static inline int wake_priority_sleeper(struct rq *rq) { int ret = 0; + #ifdef CONFIG_SCHED_SMT spin_lock(&rq->lock); /* @@ -2317,25 +2982,26 @@ EXPORT_PER_CPU_SYMBOL(kstat); * This is called on clock ticks and on context switches. * Bank in p->sched_time the ns elapsed since the last tick or switch. */ -static inline void update_cpu_clock(task_t *p, runqueue_t *rq, - unsigned long long now) +static inline void +update_cpu_clock(struct task_struct *p, struct rq *rq, unsigned long long now) { - unsigned long long last = max(p->timestamp, rq->timestamp_last_tick); - p->sched_time += now - last; + p->sched_time += now - max(p->timestamp, rq->timestamp_last_tick); } /* * Return current->sched_time plus any more ns on the sched_clock * that have not yet been banked. */ -unsigned long long current_sched_time(const task_t *tsk) +unsigned long long current_sched_time(const struct task_struct *p) { unsigned long long ns; unsigned long flags; + local_irq_save(flags); - ns = max(tsk->timestamp, task_rq(tsk)->timestamp_last_tick); - ns = tsk->sched_time + (sched_clock() - ns); + ns = max(p->timestamp, task_rq(p)->timestamp_last_tick); + ns = p->sched_time + sched_clock() - ns; local_irq_restore(flags); + return ns; } @@ -2349,11 +3015,16 @@ unsigned long long current_sched_time(const task_t *tsk) * increasing number of running tasks. We also ignore the interactivity * if a better static_prio task has expired: */ -#define EXPIRED_STARVING(rq) \ - ((STARVATION_LIMIT && ((rq)->expired_timestamp && \ - (jiffies - (rq)->expired_timestamp >= \ - STARVATION_LIMIT * ((rq)->nr_running) + 1))) || \ - ((rq)->curr->static_prio > (rq)->best_expired_prio)) +static inline int expired_starving(struct rq *rq) +{ + if (rq->curr->static_prio > rq->best_expired_prio) + return 1; + if (!STARVATION_LIMIT || !rq->expired_timestamp) + return 0; + if (jiffies - rq->expired_timestamp > STARVATION_LIMIT * rq->nr_running) + return 1; + return 0; +} /* * Account user cpu time to a process. @@ -2390,7 +3061,7 @@ void account_system_time(struct task_struct *p, int hardirq_offset, { struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat; struct vx_info *vxi = p->vx_info; /* p is _always_ current */ - runqueue_t *rq = this_rq(); + struct rq *rq = this_rq(); cputime64_t tmp; p->stime = cputime_add(p->stime, cputime); @@ -2410,8 +3081,6 @@ void account_system_time(struct task_struct *p, int hardirq_offset, cpustat->idle = cputime64_add(cpustat->idle, tmp); /* Account for system time used */ acct_update_integrals(p); - /* Update rss highwater mark */ - update_mem_hiwater(p); } /* @@ -2423,7 +3092,7 @@ void account_steal_time(struct task_struct *p, cputime_t steal) { struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat; cputime64_t tmp = cputime_to_cputime64(steal); - runqueue_t *rq = this_rq(); + struct rq *rq = this_rq(); if (p == rq->idle) { p->stime = cputime_add(p->stime, steal); @@ -2444,10 +3113,10 @@ void account_steal_time(struct task_struct *p, cputime_t steal) */ void scheduler_tick(void) { - int cpu = smp_processor_id(); - runqueue_t *rq = this_rq(); - task_t *p = current; unsigned long long now = sched_clock(); + struct task_struct *p = current; + int cpu = smp_processor_id(); + struct rq *rq = cpu_rq(cpu); update_cpu_clock(p, rq, now); @@ -2501,7 +3170,7 @@ void scheduler_tick(void) if (!rq->expired_timestamp) rq->expired_timestamp = jiffies; - if (!TASK_INTERACTIVE(p) || EXPIRED_STARVING(rq)) { + if (!TASK_INTERACTIVE(p) || expired_starving(rq)) { enqueue_task(p, rq->expired); if (p->static_prio < rq->best_expired_prio) rq->best_expired_prio = p->static_prio; @@ -2540,89 +3209,96 @@ out: } #ifdef CONFIG_SCHED_SMT -static inline void wake_sleeping_dependent(int this_cpu, runqueue_t *this_rq) +static inline void wakeup_busy_runqueue(struct rq *rq) { - struct sched_domain *sd = this_rq->sd; - cpumask_t sibling_map; - int i; + /* If an SMT runqueue is sleeping due to priority reasons wake it up */ + if (rq->curr == rq->idle && rq->nr_running) + resched_task(rq->idle); +} - if (!(sd->flags & SD_SHARE_CPUPOWER)) - return; +/* + * Called with interrupt disabled and this_rq's runqueue locked. + */ +static void wake_sleeping_dependent(int this_cpu) +{ + struct sched_domain *tmp, *sd = NULL; + int i; - /* - * Unlock the current runqueue because we have to lock in - * CPU order to avoid deadlocks. Caller knows that we might - * unlock. We keep IRQs disabled. - */ - spin_unlock(&this_rq->lock); + for_each_domain(this_cpu, tmp) { + if (tmp->flags & SD_SHARE_CPUPOWER) { + sd = tmp; + break; + } + } - sibling_map = sd->span; + if (!sd) + return; - for_each_cpu_mask(i, sibling_map) - spin_lock(&cpu_rq(i)->lock); - /* - * We clear this CPU from the mask. This both simplifies the - * inner loop and keps this_rq locked when we exit: - */ - cpu_clear(this_cpu, sibling_map); + for_each_cpu_mask(i, sd->span) { + struct rq *smt_rq = cpu_rq(i); - for_each_cpu_mask(i, sibling_map) { - runqueue_t *smt_rq = cpu_rq(i); + if (i == this_cpu) + continue; + if (unlikely(!spin_trylock(&smt_rq->lock))) + continue; - /* - * If an SMT sibling task is sleeping due to priority - * reasons wake it up now. - */ - if (smt_rq->curr == smt_rq->idle && smt_rq->nr_running) - resched_task(smt_rq->idle); + wakeup_busy_runqueue(smt_rq); + spin_unlock(&smt_rq->lock); } +} - for_each_cpu_mask(i, sibling_map) - spin_unlock(&cpu_rq(i)->lock); - /* - * We exit with this_cpu's rq still held and IRQs - * still disabled: - */ +/* + * number of 'lost' timeslices this task wont be able to fully + * utilize, if another task runs on a sibling. This models the + * slowdown effect of other tasks running on siblings: + */ +static inline unsigned long +smt_slice(struct task_struct *p, struct sched_domain *sd) +{ + return p->time_slice * (100 - sd->per_cpu_gain) / 100; } -static inline int dependent_sleeper(int this_cpu, runqueue_t *this_rq) +/* + * To minimise lock contention and not have to drop this_rq's runlock we only + * trylock the sibling runqueues and bypass those runqueues if we fail to + * acquire their lock. As we only trylock the normal locking order does not + * need to be obeyed. + */ +static int +dependent_sleeper(int this_cpu, struct rq *this_rq, struct task_struct *p) { - struct sched_domain *sd = this_rq->sd; - cpumask_t sibling_map; - prio_array_t *array; + struct sched_domain *tmp, *sd = NULL; int ret = 0, i; - task_t *p; - if (!(sd->flags & SD_SHARE_CPUPOWER)) + /* kernel/rt threads do not participate in dependent sleeping */ + if (!p->mm || rt_task(p)) return 0; - /* - * The same locking rules and details apply as for - * wake_sleeping_dependent(): - */ - spin_unlock(&this_rq->lock); - sibling_map = sd->span; - for_each_cpu_mask(i, sibling_map) - spin_lock(&cpu_rq(i)->lock); - cpu_clear(this_cpu, sibling_map); + for_each_domain(this_cpu, tmp) { + if (tmp->flags & SD_SHARE_CPUPOWER) { + sd = tmp; + break; + } + } - /* - * Establish next task to be run - it might have gone away because - * we released the runqueue lock above: - */ - if (!this_rq->nr_running) - goto out_unlock; - array = this_rq->active; - if (!array->nr_active) - array = this_rq->expired; - BUG_ON(!array->nr_active); + if (!sd) + return 0; + + for_each_cpu_mask(i, sd->span) { + struct task_struct *smt_curr; + struct rq *smt_rq; + + if (i == this_cpu) + continue; + + smt_rq = cpu_rq(i); + if (unlikely(!spin_trylock(&smt_rq->lock))) + continue; - p = list_entry(array->queue[sched_find_first_bit(array->bitmap)].next, - task_t, run_list); + smt_curr = smt_rq->curr; - for_each_cpu_mask(i, sibling_map) { - runqueue_t *smt_rq = cpu_rq(i); - task_t *smt_curr = smt_rq->curr; + if (!smt_curr->mm) + goto unlock; /* * If a user task with lower static priority than the @@ -2632,33 +3308,31 @@ static inline int dependent_sleeper(int this_cpu, runqueue_t *this_rq) * task from using an unfair proportion of the * physical cpu's resources. -ck */ - if (((smt_curr->time_slice * (100 - sd->per_cpu_gain) / 100) > - task_timeslice(p) || rt_task(smt_curr)) && - p->mm && smt_curr->mm && !rt_task(p)) - ret = 1; - - /* - * Reschedule a lower priority task on the SMT sibling, - * or wake it up if it has been put to sleep for priority - * reasons. - */ - if ((((p->time_slice * (100 - sd->per_cpu_gain) / 100) > - task_timeslice(smt_curr) || rt_task(p)) && - smt_curr->mm && p->mm && !rt_task(smt_curr)) || - (smt_curr == smt_rq->idle && smt_rq->nr_running)) - resched_task(smt_curr); + if (rt_task(smt_curr)) { + /* + * With real time tasks we run non-rt tasks only + * per_cpu_gain% of the time. + */ + if ((jiffies % DEF_TIMESLICE) > + (sd->per_cpu_gain * DEF_TIMESLICE / 100)) + ret = 1; + } else { + if (smt_curr->static_prio < p->static_prio && + !TASK_PREEMPTS_CURR(p, smt_rq) && + smt_slice(smt_curr, sd) > task_timeslice(p)) + ret = 1; + } +unlock: + spin_unlock(&smt_rq->lock); } -out_unlock: - for_each_cpu_mask(i, sibling_map) - spin_unlock(&cpu_rq(i)->lock); return ret; } #else -static inline void wake_sleeping_dependent(int this_cpu, runqueue_t *this_rq) +static inline void wake_sleeping_dependent(int this_cpu) { } - -static inline int dependent_sleeper(int this_cpu, runqueue_t *this_rq) +static inline int +dependent_sleeper(int this_cpu, struct rq *this_rq, struct task_struct *p) { return 0; } @@ -2671,12 +3345,13 @@ void fastcall add_preempt_count(int val) /* * Underflow? */ - BUG_ON(((int)preempt_count() < 0)); + if (DEBUG_LOCKS_WARN_ON((preempt_count() < 0))) + return; preempt_count() += val; /* * Spinlock count overflowing soon? */ - BUG_ON((preempt_count() & PREEMPT_MASK) >= PREEMPT_MASK-10); + DEBUG_LOCKS_WARN_ON((preempt_count() & PREEMPT_MASK) >= PREEMPT_MASK-10); } EXPORT_SYMBOL(add_preempt_count); @@ -2685,47 +3360,55 @@ void fastcall sub_preempt_count(int val) /* * Underflow? */ - BUG_ON(val > preempt_count()); + if (DEBUG_LOCKS_WARN_ON(val > preempt_count())) + return; /* * Is the spinlock portion underflowing? */ - BUG_ON((val < PREEMPT_MASK) && !(preempt_count() & PREEMPT_MASK)); + if (DEBUG_LOCKS_WARN_ON((val < PREEMPT_MASK) && + !(preempt_count() & PREEMPT_MASK))) + return; + preempt_count() -= val; } EXPORT_SYMBOL(sub_preempt_count); #endif +static inline int interactive_sleep(enum sleep_type sleep_type) +{ + return (sleep_type == SLEEP_INTERACTIVE || + sleep_type == SLEEP_INTERRUPTED); +} + /* * schedule() is the main scheduler function. */ asmlinkage void __sched schedule(void) { - long *switch_count; - task_t *prev, *next; - runqueue_t *rq; - prio_array_t *array; + struct task_struct *prev, *next; + struct prio_array *array; struct list_head *queue; unsigned long long now; unsigned long run_time; + int cpu, idx, new_prio; + long *switch_count; + struct rq *rq; struct vx_info *vxi; #ifdef CONFIG_VSERVER_HARDCPU int maxidle = -HZ; #endif - int cpu, idx; /* * Test if we are atomic. Since do_exit() needs to call into * schedule() atomically, we ignore that path for now. * Otherwise, whine if we are scheduling when we should not be. */ - if (likely(!current->exit_state)) { - if (unlikely(in_atomic())) { - printk(KERN_ERR "scheduling while atomic: " - "%s/0x%08x/%d\n", - current->comm, preempt_count(), current->pid); - dump_stack(); - } + if (unlikely(in_atomic() && !current->exit_state)) { + printk(KERN_ERR "BUG: scheduling while atomic: " + "%s/0x%08x/%d\n", + current->comm, preempt_count(), current->pid); + dump_stack(); } profile_hit(SCHED_PROFILING, __builtin_return_address(0)); @@ -2787,7 +3470,7 @@ need_resched_nonpreemptible: vxi = NULL; list_for_each_safe(l, n, &rq->hold_queue) { - next = list_entry(l, task_t, run_list); + next = list_entry(l, struct task_struct, run_list); if (vxi == next->vx_info) continue; @@ -2809,32 +3492,13 @@ pick_next: cpu = smp_processor_id(); if (unlikely(!rq->nr_running)) { -go_idle: idle_balance(cpu, rq); if (!rq->nr_running) { next = rq->idle; rq->expired_timestamp = 0; - wake_sleeping_dependent(cpu, rq); - /* - * wake_sleeping_dependent() might have released - * the runqueue, so break out if we got new - * tasks meanwhile: - */ - if (!rq->nr_running) - goto switch_tasks; - } - } else { - if (dependent_sleeper(cpu, rq)) { - next = rq->idle; + wake_sleeping_dependent(cpu); goto switch_tasks; } - /* - * dependent_sleeper() releases and reacquires the runqueue - * lock, hence go into the idle loop if the rq went - * empty meanwhile: - */ - if (unlikely(!rq->nr_running)) - goto go_idle; } array = rq->active; @@ -2852,7 +3516,7 @@ go_idle: idx = sched_find_first_bit(array->bitmap); queue = array->queue + idx; - next = list_entry(queue->next, task_t, run_list); + next = list_entry(queue->next, struct task_struct, run_list); vxi = next->vx_info; #ifdef CONFIG_VSERVER_HARDCPU @@ -2870,24 +3534,31 @@ go_idle: if (vx_info_flags(vxi, VXF_SCHED_PRIO, 0)) vx_tokens_recalc(vxi); - if (!rt_task(next) && next->activated > 0) { + if (!rt_task(next) && interactive_sleep(next->sleep_type)) { unsigned long long delta = now - next->timestamp; if (unlikely((long long)(now - next->timestamp) < 0)) delta = 0; - if (next->activated == 1) + if (next->sleep_type == SLEEP_INTERACTIVE) delta = delta * (ON_RUNQUEUE_WEIGHT * 128 / 100) / 128; array = next->array; - dequeue_task(next, array); - recalc_task_prio(next, next->timestamp + delta); - enqueue_task(next, array); + new_prio = recalc_task_prio(next, next->timestamp + delta); + + if (unlikely(next->prio != new_prio)) { + dequeue_task(next, array); + next->prio = new_prio; + enqueue_task(next, array); + } } - next->activated = 0; + next->sleep_type = SLEEP_NORMAL; + if (dependent_sleeper(cpu, rq, next)) + next = rq->idle; switch_tasks: if (next == rq->idle) schedstat_inc(rq, sched_goidle); prefetch(next); + prefetch_stack(next); clear_tsk_need_resched(prev); rcu_qsctr_inc(task_cpu(prev)); @@ -2905,11 +3576,15 @@ switch_tasks: rq->curr = next; ++*switch_count; - prepare_arch_switch(rq, next); + prepare_task_switch(rq, next); prev = context_switch(rq, prev, next); barrier(); - - finish_task_switch(prev); + /* + * this_rq must be evaluated again because prev may have moved + * CPUs since it called schedule(), thus the 'rq' on its stack + * frame will be invalid. + */ + finish_task_switch(this_rq(), prev); } else spin_unlock_irq(&rq->lock); @@ -2920,12 +3595,11 @@ switch_tasks: if (unlikely(test_thread_flag(TIF_NEED_RESCHED))) goto need_resched; } - EXPORT_SYMBOL(schedule); #ifdef CONFIG_PREEMPT /* - * this is is the entry point to schedule() from in-kernel preemption + * this is the entry point to schedule() from in-kernel preemption * off of preempt_enable. Kernel preemptions off return from interrupt * occur there and call schedule directly. */ @@ -2965,11 +3639,10 @@ need_resched: if (unlikely(test_thread_flag(TIF_NEED_RESCHED))) goto need_resched; } - EXPORT_SYMBOL(preempt_schedule); /* - * this is is the entry point to schedule() from kernel preemption + * this is the entry point to schedule() from kernel preemption * off of irq context. * Note, that this is called and return with irqs disabled. This will * protect us against recursive calling from irq. @@ -2981,7 +3654,7 @@ asmlinkage void __sched preempt_schedule_irq(void) struct task_struct *task = current; int saved_lock_depth; #endif - /* Catch callers which need to be fixed*/ + /* Catch callers which need to be fixed */ BUG_ON(ti->preempt_count || !irqs_disabled()); need_resched: @@ -3011,12 +3684,11 @@ need_resched: #endif /* CONFIG_PREEMPT */ -int default_wake_function(wait_queue_t *curr, unsigned mode, int sync, void *key) +int default_wake_function(wait_queue_t *curr, unsigned mode, int sync, + void *key) { - task_t *p = curr->task; - return try_to_wake_up(p, mode, sync); + return try_to_wake_up(curr->private, mode, sync); } - EXPORT_SYMBOL(default_wake_function); /* @@ -3034,13 +3706,11 @@ static void __wake_up_common(wait_queue_head_t *q, unsigned int mode, struct list_head *tmp, *next; list_for_each_safe(tmp, next, &q->task_list) { - wait_queue_t *curr; - unsigned flags; - curr = list_entry(tmp, wait_queue_t, task_list); - flags = curr->flags; + wait_queue_t *curr = list_entry(tmp, wait_queue_t, task_list); + unsigned flags = curr->flags; + if (curr->func(curr, mode, sync, key) && - (flags & WQ_FLAG_EXCLUSIVE) && - !--nr_exclusive) + (flags & WQ_FLAG_EXCLUSIVE) && !--nr_exclusive) break; } } @@ -3053,7 +3723,7 @@ static void __wake_up_common(wait_queue_head_t *q, unsigned int mode, * @key: is directly passed to the wakeup function */ void fastcall __wake_up(wait_queue_head_t *q, unsigned int mode, - int nr_exclusive, void *key) + int nr_exclusive, void *key) { unsigned long flags; @@ -3061,7 +3731,6 @@ void fastcall __wake_up(wait_queue_head_t *q, unsigned int mode, __wake_up_common(q, mode, nr_exclusive, 0, key); spin_unlock_irqrestore(&q->lock, flags); } - EXPORT_SYMBOL(__wake_up); /* @@ -3085,7 +3754,8 @@ void fastcall __wake_up_locked(wait_queue_head_t *q, unsigned int mode) * * On UP it can prevent extra preemption. */ -void fastcall __wake_up_sync(wait_queue_head_t *q, unsigned int mode, int nr_exclusive) +void fastcall +__wake_up_sync(wait_queue_head_t *q, unsigned int mode, int nr_exclusive) { unsigned long flags; int sync = 1; @@ -3129,6 +3799,7 @@ EXPORT_SYMBOL(complete_all); void fastcall __sched wait_for_completion(struct completion *x) { might_sleep(); + spin_lock_irq(&x->wait.lock); if (!x->done) { DECLARE_WAITQUEUE(wait, current); @@ -3263,23 +3934,36 @@ EXPORT_SYMBOL(wait_for_completion_interruptible_timeout); __remove_wait_queue(q, &wait); \ spin_unlock_irqrestore(&q->lock, flags); +#define SLEEP_ON_BKLCHECK \ + if (unlikely(!kernel_locked()) && \ + sleep_on_bkl_warnings < 10) { \ + sleep_on_bkl_warnings++; \ + WARN_ON(1); \ + } + +static int sleep_on_bkl_warnings; + void fastcall __sched interruptible_sleep_on(wait_queue_head_t *q) { SLEEP_ON_VAR + SLEEP_ON_BKLCHECK + current->state = TASK_INTERRUPTIBLE; SLEEP_ON_HEAD schedule(); SLEEP_ON_TAIL } - EXPORT_SYMBOL(interruptible_sleep_on); -long fastcall __sched interruptible_sleep_on_timeout(wait_queue_head_t *q, long timeout) +long fastcall __sched +interruptible_sleep_on_timeout(wait_queue_head_t *q, long timeout) { SLEEP_ON_VAR + SLEEP_ON_BKLCHECK + current->state = TASK_INTERRUPTIBLE; SLEEP_ON_HEAD @@ -3288,43 +3972,84 @@ long fastcall __sched interruptible_sleep_on_timeout(wait_queue_head_t *q, long return timeout; } - EXPORT_SYMBOL(interruptible_sleep_on_timeout); -void fastcall __sched sleep_on(wait_queue_head_t *q) +long fastcall __sched sleep_on_timeout(wait_queue_head_t *q, long timeout) { SLEEP_ON_VAR + SLEEP_ON_BKLCHECK + current->state = TASK_UNINTERRUPTIBLE; SLEEP_ON_HEAD - schedule(); + timeout = schedule_timeout(timeout); SLEEP_ON_TAIL + + return timeout; } -EXPORT_SYMBOL(sleep_on); +EXPORT_SYMBOL(sleep_on_timeout); -long fastcall __sched sleep_on_timeout(wait_queue_head_t *q, long timeout) +#ifdef CONFIG_RT_MUTEXES + +/* + * rt_mutex_setprio - set the current priority of a task + * @p: task + * @prio: prio value (kernel-internal form) + * + * This function changes the 'effective' priority of a task. It does + * not touch ->normal_prio like __setscheduler(). + * + * Used by the rt_mutex code to implement priority inheritance logic. + */ +void rt_mutex_setprio(struct task_struct *p, int prio) { - SLEEP_ON_VAR + struct prio_array *array; + unsigned long flags; + struct rq *rq; + int oldprio; - current->state = TASK_UNINTERRUPTIBLE; + BUG_ON(prio < 0 || prio > MAX_PRIO); - SLEEP_ON_HEAD - timeout = schedule_timeout(timeout); - SLEEP_ON_TAIL + rq = task_rq_lock(p, &flags); - return timeout; + oldprio = p->prio; + array = p->array; + if (array) + dequeue_task(p, array); + p->prio = prio; + + if (array) { + /* + * If changing to an RT priority then queue it + * in the active array! + */ + if (rt_task(p)) + array = rq->active; + enqueue_task(p, array); + /* + * Reschedule if we are currently running on this runqueue and + * our priority decreased, or if we are not currently running on + * this runqueue and our priority is higher than the current's + */ + if (task_running(rq, p)) { + if (p->prio > oldprio) + resched_task(rq->curr); + } else if (TASK_PREEMPTS_CURR(p, rq)) + resched_task(rq->curr); + } + task_rq_unlock(rq, &flags); } -EXPORT_SYMBOL(sleep_on_timeout); +#endif -void set_user_nice(task_t *p, long nice) +void set_user_nice(struct task_struct *p, long nice) { + struct prio_array *array; + int old_prio, delta; unsigned long flags; - prio_array_t *array; - runqueue_t *rq; - int old_prio, new_prio, delta; + struct rq *rq; if (TASK_NICE(p) == nice || nice < -20 || nice > 19) return; @@ -3337,24 +4062,27 @@ void set_user_nice(task_t *p, long nice) * The RT priorities are set via sched_setscheduler(), but we still * allow the 'normal' nice value to be set - but as expected * it wont have any effect on scheduling until the task is - * not SCHED_NORMAL: + * not SCHED_NORMAL/SCHED_BATCH: */ - if (rt_task(p)) { + if (has_rt_policy(p)) { p->static_prio = NICE_TO_PRIO(nice); goto out_unlock; } array = p->array; - if (array) + if (array) { dequeue_task(p, array); + dec_raw_weighted_load(rq, p); + } - old_prio = p->prio; - new_prio = NICE_TO_PRIO(nice); - delta = new_prio - old_prio; p->static_prio = NICE_TO_PRIO(nice); - p->prio += delta; + set_load_weight(p); + old_prio = p->prio; + p->prio = effective_prio(p); + delta = p->prio - old_prio; if (array) { enqueue_task(p, array); + inc_raw_weighted_load(rq, p); /* * If the task increased its priority or is running and * lowered its priority, then reschedule its CPU: @@ -3365,7 +4093,6 @@ void set_user_nice(task_t *p, long nice) out_unlock: task_rq_unlock(rq, &flags); } - EXPORT_SYMBOL(set_user_nice); /* @@ -3373,10 +4100,11 @@ EXPORT_SYMBOL(set_user_nice); * @p: task * @nice: nice value */ -int can_nice(const task_t *p, const int nice) +int can_nice(const struct task_struct *p, const int nice) { - /* convert nice value [19,-20] to rlimit style value [0,39] */ - int nice_rlim = 19 - nice; + /* convert nice value [19,-20] to rlimit style value [1,40] */ + int nice_rlim = 20 - nice; + return (nice_rlim <= p->signal->rlim[RLIMIT_NICE].rlim_cur || capable(CAP_SYS_NICE)); } @@ -3392,8 +4120,7 @@ int can_nice(const task_t *p, const int nice) */ asmlinkage long sys_nice(int increment) { - int retval; - long nice; + long nice, retval; /* * Setpriority might change our priority at the same moment. @@ -3432,7 +4159,7 @@ asmlinkage long sys_nice(int increment) * RT tasks are offset by -200. Normal tasks are centered * around 0, value goes from -16 to +15. */ -int task_prio(const task_t *p) +int task_prio(const struct task_struct *p) { return p->prio - MAX_RT_PRIO; } @@ -3441,19 +4168,11 @@ int task_prio(const task_t *p) * task_nice - return the nice value of a given task. * @p: the task in question. */ -int task_nice(const task_t *p) +int task_nice(const struct task_struct *p) { return TASK_NICE(p); } - -/* - * The only users of task_nice are binfmt_elf and binfmt_elf32. - * binfmt_elf is no longer modular, but binfmt_elf32 still is. - * Therefore, task_nice is needed if there is a compat_mode. - */ -#ifdef CONFIG_COMPAT EXPORT_SYMBOL_GPL(task_nice); -#endif /** * idle_cpu - is a given cpu idle currently? @@ -3464,13 +4183,11 @@ int idle_cpu(int cpu) return cpu_curr(cpu) == cpu_rq(cpu)->idle; } -EXPORT_SYMBOL_GPL(idle_cpu); - /** * idle_task - return the idle task for a given cpu. * @cpu: the processor in question. */ -task_t *idle_task(int cpu) +struct task_struct *idle_task(int cpu) { return cpu_rq(cpu)->idle; } @@ -3479,7 +4196,7 @@ task_t *idle_task(int cpu) * find_process_by_pid - find a process with a matching PID value. * @pid: the pid in question. */ -static inline task_t *find_process_by_pid(pid_t pid) +static inline struct task_struct *find_process_by_pid(pid_t pid) { return pid ? find_task_by_pid(pid) : current; } @@ -3488,12 +4205,18 @@ static inline task_t *find_process_by_pid(pid_t pid) static void __setscheduler(struct task_struct *p, int policy, int prio) { BUG_ON(p->array); + p->policy = policy; p->rt_priority = prio; - if (policy != SCHED_NORMAL) - p->prio = MAX_USER_RT_PRIO-1 - p->rt_priority; - else - p->prio = p->static_prio; + p->normal_prio = normal_prio(p); + /* we are holding p->pi_lock already */ + p->prio = rt_mutex_getprio(p); + /* + * SCHED_BATCH tasks are treated as perpetual CPU hogs: + */ + if (policy == SCHED_BATCH) + p->sleep_avg = 0; + set_load_weight(p); } /** @@ -3503,51 +4226,78 @@ static void __setscheduler(struct task_struct *p, int policy, int prio) * @policy: new policy. * @param: structure containing the new RT priority. */ -int sched_setscheduler(struct task_struct *p, int policy, struct sched_param *param) +int sched_setscheduler(struct task_struct *p, int policy, + struct sched_param *param) { - int retval; - int oldprio, oldpolicy = -1; - prio_array_t *array; + int retval, oldprio, oldpolicy = -1; + struct prio_array *array; unsigned long flags; - runqueue_t *rq; + struct rq *rq; + /* may grab non-irq protected spin_locks */ + BUG_ON(in_interrupt()); recheck: /* double check policy once rq lock held */ if (policy < 0) policy = oldpolicy = p->policy; else if (policy != SCHED_FIFO && policy != SCHED_RR && - policy != SCHED_NORMAL) - return -EINVAL; + policy != SCHED_NORMAL && policy != SCHED_BATCH) + return -EINVAL; /* * Valid priorities for SCHED_FIFO and SCHED_RR are - * 1..MAX_USER_RT_PRIO-1, valid priority for SCHED_NORMAL is 0. + * 1..MAX_USER_RT_PRIO-1, valid priority for SCHED_NORMAL and + * SCHED_BATCH is 0. */ if (param->sched_priority < 0 || - param->sched_priority > MAX_USER_RT_PRIO-1) + (p->mm && param->sched_priority > MAX_USER_RT_PRIO-1) || + (!p->mm && param->sched_priority > MAX_RT_PRIO-1)) return -EINVAL; - if ((policy == SCHED_NORMAL) != (param->sched_priority == 0)) + if ((policy == SCHED_NORMAL || policy == SCHED_BATCH) + != (param->sched_priority == 0)) return -EINVAL; - if ((policy == SCHED_FIFO || policy == SCHED_RR) && - param->sched_priority > p->signal->rlim[RLIMIT_RTPRIO].rlim_cur && - !capable(CAP_SYS_NICE)) - return -EPERM; - if ((current->euid != p->euid) && (current->euid != p->uid) && - !capable(CAP_SYS_NICE)) - return -EPERM; + /* + * Allow unprivileged RT tasks to decrease priority: + */ + if (!capable(CAP_SYS_NICE)) { + /* + * can't change policy, except between SCHED_NORMAL + * and SCHED_BATCH: + */ + if (((policy != SCHED_NORMAL && p->policy != SCHED_BATCH) && + (policy != SCHED_BATCH && p->policy != SCHED_NORMAL)) && + !p->signal->rlim[RLIMIT_RTPRIO].rlim_cur) + return -EPERM; + /* can't increase priority */ + if ((policy != SCHED_NORMAL && policy != SCHED_BATCH) && + param->sched_priority > p->rt_priority && + param->sched_priority > + p->signal->rlim[RLIMIT_RTPRIO].rlim_cur) + return -EPERM; + /* can't change other user's priorities */ + if ((current->euid != p->euid) && + (current->euid != p->uid)) + return -EPERM; + } retval = security_task_setscheduler(p, policy, param); if (retval) return retval; + /* + * make sure no PI-waiters arrive (or leave) while we are + * changing the priority of the task: + */ + spin_lock_irqsave(&p->pi_lock, flags); /* * To be able to change p->policy safely, the apropriate * runqueue lock must be held. */ - rq = task_rq_lock(p, &flags); + rq = __task_rq_lock(p); /* recheck policy now with rq lock held */ if (unlikely(oldpolicy != -1 && oldpolicy != p->policy)) { policy = oldpolicy = -1; - task_rq_unlock(rq, &flags); + __task_rq_unlock(rq); + spin_unlock_irqrestore(&p->pi_lock, flags); goto recheck; } array = p->array; @@ -3569,16 +4319,21 @@ recheck: } else if (TASK_PREEMPTS_CURR(p, rq)) resched_task(rq->curr); } - task_rq_unlock(rq, &flags); + __task_rq_unlock(rq); + spin_unlock_irqrestore(&p->pi_lock, flags); + + rt_mutex_adjust_pi(p); + return 0; } EXPORT_SYMBOL_GPL(sched_setscheduler); -static int do_sched_setscheduler(pid_t pid, int policy, struct sched_param __user *param) +static int +do_sched_setscheduler(pid_t pid, int policy, struct sched_param __user *param) { - int retval; struct sched_param lparam; struct task_struct *p; + int retval; if (!param || pid < 0) return -EINVAL; @@ -3592,6 +4347,7 @@ static int do_sched_setscheduler(pid_t pid, int policy, struct sched_param __use } retval = sched_setscheduler(p, policy, &lparam); read_unlock_irq(&tasklist_lock); + return retval; } @@ -3604,6 +4360,10 @@ static int do_sched_setscheduler(pid_t pid, int policy, struct sched_param __use asmlinkage long sys_sched_setscheduler(pid_t pid, int policy, struct sched_param __user *param) { + /* negative values for policy are not valid */ + if (policy < 0) + return -EINVAL; + return do_sched_setscheduler(pid, policy, param); } @@ -3623,8 +4383,8 @@ asmlinkage long sys_sched_setparam(pid_t pid, struct sched_param __user *param) */ asmlinkage long sys_sched_getscheduler(pid_t pid) { + struct task_struct *p; int retval = -EINVAL; - task_t *p; if (pid < 0) goto out_nounlock; @@ -3651,8 +4411,8 @@ out_nounlock: asmlinkage long sys_sched_getparam(pid_t pid, struct sched_param __user *param) { struct sched_param lp; + struct task_struct *p; int retval = -EINVAL; - task_t *p; if (!param || pid < 0) goto out_nounlock; @@ -3685,9 +4445,9 @@ out_unlock: long sched_setaffinity(pid_t pid, cpumask_t new_mask) { - task_t *p; - int retval; cpumask_t cpus_allowed; + struct task_struct *p; + int retval; lock_cpu_hotplug(); read_lock(&tasklist_lock); @@ -3712,6 +4472,10 @@ long sched_setaffinity(pid_t pid, cpumask_t new_mask) !capable(CAP_SYS_NICE)) goto out_unlock; + retval = security_task_setscheduler(p, 0, NULL); + if (retval) + goto out_unlock; + cpus_allowed = cpuset_cpus_allowed(p); cpus_and(new_mask, new_mask, cpus_allowed); retval = set_cpus_allowed(p, new_mask); @@ -3759,18 +4523,18 @@ asmlinkage long sys_sched_setaffinity(pid_t pid, unsigned int len, * method, such as ACPI for e.g. */ -cpumask_t cpu_present_map; +cpumask_t cpu_present_map __read_mostly; EXPORT_SYMBOL(cpu_present_map); #ifndef CONFIG_SMP -cpumask_t cpu_online_map = CPU_MASK_ALL; -cpumask_t cpu_possible_map = CPU_MASK_ALL; +cpumask_t cpu_online_map __read_mostly = CPU_MASK_ALL; +cpumask_t cpu_possible_map __read_mostly = CPU_MASK_ALL; #endif long sched_getaffinity(pid_t pid, cpumask_t *mask) { + struct task_struct *p; int retval; - task_t *p; lock_cpu_hotplug(); read_lock(&tasklist_lock); @@ -3780,8 +4544,11 @@ long sched_getaffinity(pid_t pid, cpumask_t *mask) if (!p) goto out_unlock; - retval = 0; - cpus_and(*mask, p->cpus_allowed, cpu_possible_map); + retval = security_task_getscheduler(p); + if (retval) + goto out_unlock; + + cpus_and(*mask, p->cpus_allowed, cpu_online_map); out_unlock: read_unlock(&tasklist_lock); @@ -3826,9 +4593,8 @@ asmlinkage long sys_sched_getaffinity(pid_t pid, unsigned int len, */ asmlinkage long sys_sched_yield(void) { - runqueue_t *rq = this_rq_lock(); - prio_array_t *array = current->array; - prio_array_t *target = rq->expired; + struct rq *rq = this_rq_lock(); + struct prio_array *array = current->array, *target = rq->expired; schedstat_inc(rq, yld_cnt); /* @@ -3841,7 +4607,7 @@ asmlinkage long sys_sched_yield(void) if (rt_task(current)) target = rq->active; - if (current->array->nr_active == 1) { + if (array->nr_active == 1) { schedstat_inc(rq, yld_act_empty); if (!rq->expired->nr_active) schedstat_inc(rq, yld_both_empty); @@ -3862,6 +4628,7 @@ asmlinkage long sys_sched_yield(void) * no need to preempt or enable interrupts: */ __release(rq->lock); + spin_release(&rq->lock.dep_map, 1, _THIS_IP_); _raw_spin_unlock(&rq->lock); preempt_enable_no_resched(); @@ -3870,8 +4637,25 @@ asmlinkage long sys_sched_yield(void) return 0; } -static inline void __cond_resched(void) +static inline int __resched_legal(int expected_preempt_count) { + if (unlikely(preempt_count() != expected_preempt_count)) + return 0; + if (unlikely(system_state != SYSTEM_RUNNING)) + return 0; + return 1; +} + +static void __cond_resched(void) +{ +#ifdef CONFIG_DEBUG_SPINLOCK_SLEEP + __might_sleep(__FILE__, __LINE__); +#endif + /* + * The BKS might be reacquired before we have dropped + * PREEMPT_ACTIVE, which could trigger a second + * cond_resched() call. + */ do { add_preempt_count(PREEMPT_ACTIVE); schedule(); @@ -3881,13 +4665,12 @@ static inline void __cond_resched(void) int __sched cond_resched(void) { - if (need_resched()) { + if (need_resched() && __resched_legal(0)) { __cond_resched(); return 1; } return 0; } - EXPORT_SYMBOL(cond_resched); /* @@ -3898,7 +4681,7 @@ EXPORT_SYMBOL(cond_resched); * operations here to prevent schedule() from being called twice (once via * spin_unlock(), once by hand). */ -int cond_resched_lock(spinlock_t * lock) +int cond_resched_lock(spinlock_t *lock) { int ret = 0; @@ -3908,7 +4691,8 @@ int cond_resched_lock(spinlock_t * lock) ret = 1; spin_lock(lock); } - if (need_resched()) { + if (need_resched() && __resched_legal(1)) { + spin_release(&lock->dep_map, 1, _THIS_IP_); _raw_spin_unlock(lock); preempt_enable_no_resched(); __cond_resched(); @@ -3917,25 +4701,24 @@ int cond_resched_lock(spinlock_t * lock) } return ret; } - EXPORT_SYMBOL(cond_resched_lock); int __sched cond_resched_softirq(void) { BUG_ON(!in_softirq()); - if (need_resched()) { - __local_bh_enable(); + if (need_resched() && __resched_legal(0)) { + raw_local_irq_disable(); + _local_bh_enable(); + raw_local_irq_enable(); __cond_resched(); local_bh_disable(); return 1; } return 0; } - EXPORT_SYMBOL(cond_resched_softirq); - /** * yield - yield the current processor to other threads. * @@ -3947,7 +4730,6 @@ void __sched yield(void) set_current_state(TASK_RUNNING); sys_sched_yield(); } - EXPORT_SYMBOL(yield); /* @@ -3959,23 +4741,26 @@ EXPORT_SYMBOL(yield); */ void __sched io_schedule(void) { - struct runqueue *rq = &per_cpu(runqueues, _smp_processor_id()); + struct rq *rq = &__raw_get_cpu_var(runqueues); + delayacct_blkio_start(); atomic_inc(&rq->nr_iowait); schedule(); atomic_dec(&rq->nr_iowait); + delayacct_blkio_end(); } - EXPORT_SYMBOL(io_schedule); long __sched io_schedule_timeout(long timeout) { - struct runqueue *rq = &per_cpu(runqueues, _smp_processor_id()); + struct rq *rq = &__raw_get_cpu_var(runqueues); long ret; + delayacct_blkio_start(); atomic_inc(&rq->nr_iowait); ret = schedule_timeout(timeout); atomic_dec(&rq->nr_iowait); + delayacct_blkio_end(); return ret; } @@ -3996,6 +4781,7 @@ asmlinkage long sys_sched_get_priority_max(int policy) ret = MAX_USER_RT_PRIO-1; break; case SCHED_NORMAL: + case SCHED_BATCH: ret = 0; break; } @@ -4019,6 +4805,7 @@ asmlinkage long sys_sched_get_priority_min(int policy) ret = 1; break; case SCHED_NORMAL: + case SCHED_BATCH: ret = 0; } return ret; @@ -4035,9 +4822,9 @@ asmlinkage long sys_sched_get_priority_min(int policy) asmlinkage long sys_sched_rr_get_interval(pid_t pid, struct timespec __user *interval) { + struct task_struct *p; int retval = -EINVAL; struct timespec t; - task_t *p; if (pid < 0) goto out_nounlock; @@ -4052,7 +4839,7 @@ long sys_sched_rr_get_interval(pid_t pid, struct timespec __user *interval) if (retval) goto out_unlock; - jiffies_to_timespec(p->policy & SCHED_FIFO ? + jiffies_to_timespec(p->policy == SCHED_FIFO ? 0 : task_timeslice(p), &t); read_unlock(&tasklist_lock); retval = copy_to_user(interval, &t, sizeof(t)) ? -EFAULT : 0; @@ -4065,35 +4852,36 @@ out_unlock: static inline struct task_struct *eldest_child(struct task_struct *p) { - if (list_empty(&p->children)) return NULL; + if (list_empty(&p->children)) + return NULL; return list_entry(p->children.next,struct task_struct,sibling); } static inline struct task_struct *older_sibling(struct task_struct *p) { - if (p->sibling.prev==&p->parent->children) return NULL; + if (p->sibling.prev==&p->parent->children) + return NULL; return list_entry(p->sibling.prev,struct task_struct,sibling); } static inline struct task_struct *younger_sibling(struct task_struct *p) { - if (p->sibling.next==&p->parent->children) return NULL; + if (p->sibling.next==&p->parent->children) + return NULL; return list_entry(p->sibling.next,struct task_struct,sibling); } -static void show_task(task_t * p) +static const char stat_nam[] = "RSDTtZX"; + +static void show_task(struct task_struct *p) { - task_t *relative; - unsigned state; + struct task_struct *relative; unsigned long free = 0; - static const char *stat_nam[] = { "R", "S", "D", "T", "t", "Z", "X" }; + unsigned state; - printk("%-13.13s ", p->comm); state = p->state ? __ffs(p->state) + 1 : 0; - if (state < ARRAY_SIZE(stat_nam)) - printk(stat_nam[state]); - else - printk("?"); + printk("%-13.13s %c", p->comm, + state < sizeof(stat_nam) - 1 ? stat_nam[state] : '?'); #if (BITS_PER_LONG == 32) if (state == TASK_RUNNING) printk(" running "); @@ -4107,10 +4895,10 @@ static void show_task(task_t * p) #endif #ifdef CONFIG_DEBUG_STACK_USAGE { - unsigned long * n = (unsigned long *) (p->thread_info+1); + unsigned long *n = end_of_stack(p); while (!*n) n++; - free = (unsigned long) n - (unsigned long)(p->thread_info+1); + free = (unsigned long)n - (unsigned long)end_of_stack(p); } #endif printk("%5lu %5d %6d ", free, p->pid, p->parent->pid); @@ -4137,7 +4925,7 @@ static void show_task(task_t * p) void show_state(void) { - task_t *g, *p; + struct task_struct *g, *p; #if (BITS_PER_LONG == 32) printk("\n" @@ -4159,30 +4947,42 @@ void show_state(void) } while_each_thread(g, p); read_unlock(&tasklist_lock); + debug_show_all_locks(); } -void __devinit init_idle(task_t *idle, int cpu) +/** + * init_idle - set up an idle thread for a given CPU + * @idle: task in question + * @cpu: cpu the idle task belongs to + * + * NOTE: this function does not set the idle thread's NEED_RESCHED + * flag, to make booting more robust. + */ +void __devinit init_idle(struct task_struct *idle, int cpu) { - runqueue_t *rq = cpu_rq(cpu); + struct rq *rq = cpu_rq(cpu); unsigned long flags; + idle->timestamp = sched_clock(); idle->sleep_avg = 0; idle->array = NULL; - idle->prio = MAX_PRIO; + idle->prio = idle->normal_prio = MAX_PRIO; idle->state = TASK_RUNNING; idle->cpus_allowed = cpumask_of_cpu(cpu); set_task_cpu(idle, cpu); spin_lock_irqsave(&rq->lock, flags); rq->curr = rq->idle = idle; - set_tsk_need_resched(idle); +#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW) + idle->oncpu = 1; +#endif spin_unlock_irqrestore(&rq->lock, flags); /* Set the preempt count _outside_ the spinlocks! */ #if defined(CONFIG_PREEMPT) && !defined(CONFIG_PREEMPT_BKL) - idle->thread_info->preempt_count = (idle->lock_depth >= 0); + task_thread_info(idle)->preempt_count = (idle->lock_depth >= 0); #else - idle->thread_info->preempt_count = 0; + task_thread_info(idle)->preempt_count = 0; #endif } @@ -4199,7 +4999,7 @@ cpumask_t nohz_cpu_mask = CPU_MASK_NONE; /* * This is how migration works: * - * 1) we queue a migration_req_t structure in the source CPU's + * 1) we queue a struct migration_req structure in the source CPU's * runqueue and wake up that CPU's migration thread. * 2) we down() the locked semaphore => thread blocks. * 3) migration thread wakes up (implicitly it forces the migrated @@ -4221,12 +5021,12 @@ cpumask_t nohz_cpu_mask = CPU_MASK_NONE; * task must not exit() & deallocate itself prematurely. The * call is not atomic; no spinlocks may be held. */ -int set_cpus_allowed(task_t *p, cpumask_t new_mask) +int set_cpus_allowed(struct task_struct *p, cpumask_t new_mask) { + struct migration_req req; unsigned long flags; + struct rq *rq; int ret = 0; - migration_req_t req; - runqueue_t *rq; rq = task_rq_lock(p, &flags); if (!cpus_intersects(new_mask, cpu_online_map)) { @@ -4249,9 +5049,9 @@ int set_cpus_allowed(task_t *p, cpumask_t new_mask) } out: task_rq_unlock(rq, &flags); + return ret; } - EXPORT_SYMBOL_GPL(set_cpus_allowed); /* @@ -4262,13 +5062,16 @@ EXPORT_SYMBOL_GPL(set_cpus_allowed); * * So we race with normal scheduler movements, but that's OK, as long * as the task is no longer on this CPU. + * + * Returns non-zero if task was successfully migrated. */ -static void __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu) +static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu) { - runqueue_t *rq_dest, *rq_src; + struct rq *rq_dest, *rq_src; + int ret = 0; if (unlikely(cpu_is_offline(dest_cpu))) - return; + return ret; rq_src = cpu_rq(src_cpu); rq_dest = cpu_rq(dest_cpu); @@ -4292,13 +5095,15 @@ static void __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu) p->timestamp = p->timestamp - rq_src->timestamp_last_tick + rq_dest->timestamp_last_tick; deactivate_task(p, rq_src); - activate_task(p, rq_dest, 0); + vx_activate_task(p); + __activate_task(p, rq_dest); if (TASK_PREEMPTS_CURR(p, rq_dest)) resched_task(rq_dest->curr); } - + ret = 1; out: double_rq_unlock(rq_src, rq_dest); + return ret; } /* @@ -4306,21 +5111,20 @@ out: * thread migration by bumping thread off CPU then 'pushing' onto * another runqueue. */ -static int migration_thread(void * data) +static int migration_thread(void *data) { - runqueue_t *rq; int cpu = (long)data; + struct rq *rq; rq = cpu_rq(cpu); BUG_ON(rq->migration_thread != current); set_current_state(TASK_INTERRUPTIBLE); while (!kthread_should_stop()) { + struct migration_req *req; struct list_head *head; - migration_req_t *req; - if (current->flags & PF_FREEZE) - refrigerator(PF_FREEZE); + try_to_freeze(); spin_lock_irq(&rq->lock); @@ -4342,20 +5146,12 @@ static int migration_thread(void * data) set_current_state(TASK_INTERRUPTIBLE); continue; } - req = list_entry(head->next, migration_req_t, list); + req = list_entry(head->next, struct migration_req, list); list_del_init(head->next); - if (req->type == REQ_MOVE_TASK) { - spin_unlock(&rq->lock); - __migrate_task(req->task, cpu, req->dest_cpu); - local_irq_enable(); - } else if (req->type == REQ_SET_DOMAIN) { - rq->sd = req->sd; - spin_unlock_irq(&rq->lock); - } else { - spin_unlock_irq(&rq->lock); - WARN_ON(1); - } + spin_unlock(&rq->lock); + __migrate_task(req->task, cpu, req->dest_cpu); + local_irq_enable(); complete(&req->done); } @@ -4375,36 +5171,42 @@ wait_to_die: #ifdef CONFIG_HOTPLUG_CPU /* Figure out where task on dead CPU should go, use force if neccessary. */ -static void move_task_off_dead_cpu(int dead_cpu, struct task_struct *tsk) +static void move_task_off_dead_cpu(int dead_cpu, struct task_struct *p) { - int dest_cpu; + unsigned long flags; cpumask_t mask; + struct rq *rq; + int dest_cpu; +restart: /* On same node? */ mask = node_to_cpumask(cpu_to_node(dead_cpu)); - cpus_and(mask, mask, tsk->cpus_allowed); + cpus_and(mask, mask, p->cpus_allowed); dest_cpu = any_online_cpu(mask); /* On any allowed CPU? */ if (dest_cpu == NR_CPUS) - dest_cpu = any_online_cpu(tsk->cpus_allowed); + dest_cpu = any_online_cpu(p->cpus_allowed); /* No more Mr. Nice Guy. */ if (dest_cpu == NR_CPUS) { - cpus_setall(tsk->cpus_allowed); - dest_cpu = any_online_cpu(tsk->cpus_allowed); + rq = task_rq_lock(p, &flags); + cpus_setall(p->cpus_allowed); + dest_cpu = any_online_cpu(p->cpus_allowed); + task_rq_unlock(rq, &flags); /* * Don't tell them about moving exiting tasks or * kernel threads (both mm NULL), since they never * leave kernel. */ - if (tsk->mm && printk_ratelimit()) + if (p->mm && printk_ratelimit()) printk(KERN_INFO "process %d (%s) no " "longer affine to cpu%d\n", - tsk->pid, tsk->comm, dead_cpu); + p->pid, p->comm, dead_cpu); } - __migrate_task(tsk, dead_cpu, dest_cpu); + if (!__migrate_task(p, dead_cpu, dest_cpu)) + goto restart; } /* @@ -4414,9 +5216,9 @@ static void move_task_off_dead_cpu(int dead_cpu, struct task_struct *tsk) * their home CPUs. So we just add the counter to another CPU's counter, * to keep the global sum constant after CPU-down: */ -static void migrate_nr_uninterruptible(runqueue_t *rq_src) +static void migrate_nr_uninterruptible(struct rq *rq_src) { - runqueue_t *rq_dest = cpu_rq(any_online_cpu(CPU_MASK_ALL)); + struct rq *rq_dest = cpu_rq(any_online_cpu(CPU_MASK_ALL)); unsigned long flags; local_irq_save(flags); @@ -4430,48 +5232,51 @@ static void migrate_nr_uninterruptible(runqueue_t *rq_src) /* Run through task list and migrate tasks from the dead cpu. */ static void migrate_live_tasks(int src_cpu) { - struct task_struct *tsk, *t; + struct task_struct *p, *t; write_lock_irq(&tasklist_lock); - do_each_thread(t, tsk) { - if (tsk == current) + do_each_thread(t, p) { + if (p == current) continue; - if (task_cpu(tsk) == src_cpu) - move_task_off_dead_cpu(src_cpu, tsk); - } while_each_thread(t, tsk); + if (task_cpu(p) == src_cpu) + move_task_off_dead_cpu(src_cpu, p); + } while_each_thread(t, p); write_unlock_irq(&tasklist_lock); } /* Schedules idle task to be the next runnable task on current CPU. * It does so by boosting its priority to highest possible and adding it to - * the _front_ of runqueue. Used by CPU offline code. + * the _front_ of the runqueue. Used by CPU offline code. */ void sched_idle_next(void) { - int cpu = smp_processor_id(); - runqueue_t *rq = this_rq(); + int this_cpu = smp_processor_id(); + struct rq *rq = cpu_rq(this_cpu); struct task_struct *p = rq->idle; unsigned long flags; /* cpu has to be offline */ - BUG_ON(cpu_online(cpu)); + BUG_ON(cpu_online(this_cpu)); - /* Strictly not necessary since rest of the CPUs are stopped by now - * and interrupts disabled on current cpu. + /* + * Strictly not necessary since rest of the CPUs are stopped by now + * and interrupts disabled on the current cpu. */ spin_lock_irqsave(&rq->lock, flags); __setscheduler(p, SCHED_FIFO, MAX_RT_PRIO-1); - /* Add idle task to _front_ of it's priority queue */ + + /* Add idle task to the _front_ of its priority queue: */ __activate_idle_task(p, rq); spin_unlock_irqrestore(&rq->lock, flags); } -/* Ensures that the idle task is using init_mm right before its cpu goes +/* + * Ensures that the idle task is using init_mm right before its cpu goes * offline. */ void idle_task_exit(void) @@ -4485,17 +5290,17 @@ void idle_task_exit(void) mmdrop(mm); } -static void migrate_dead(unsigned int dead_cpu, task_t *tsk) +static void migrate_dead(unsigned int dead_cpu, struct task_struct *p) { - struct runqueue *rq = cpu_rq(dead_cpu); + struct rq *rq = cpu_rq(dead_cpu); /* Must be exiting, otherwise would be on tasklist. */ - BUG_ON(tsk->exit_state != EXIT_ZOMBIE && tsk->exit_state != EXIT_DEAD); + BUG_ON(p->exit_state != EXIT_ZOMBIE && p->exit_state != EXIT_DEAD); /* Cannot have done final schedule yet: would have vanished. */ - BUG_ON(tsk->flags & PF_DEAD); + BUG_ON(p->flags & PF_DEAD); - get_task_struct(tsk); + get_task_struct(p); /* * Drop lock around migration; if someone else moves it, @@ -4503,25 +5308,25 @@ static void migrate_dead(unsigned int dead_cpu, task_t *tsk) * fine. */ spin_unlock_irq(&rq->lock); - move_task_off_dead_cpu(dead_cpu, tsk); + move_task_off_dead_cpu(dead_cpu, p); spin_lock_irq(&rq->lock); - put_task_struct(tsk); + put_task_struct(p); } /* release_task() removes task from tasklist, so we won't find dead tasks. */ static void migrate_dead_tasks(unsigned int dead_cpu) { - unsigned arr, i; - struct runqueue *rq = cpu_rq(dead_cpu); + struct rq *rq = cpu_rq(dead_cpu); + unsigned int arr, i; for (arr = 0; arr < 2; arr++) { for (i = 0; i < MAX_PRIO; i++) { struct list_head *list = &rq->arrays[arr].queue[i]; + while (!list_empty(list)) - migrate_dead(dead_cpu, - list_entry(list->next, task_t, - run_list)); + migrate_dead(dead_cpu, list_entry(list->next, + struct task_struct, run_list)); } } } @@ -4531,13 +5336,13 @@ static void migrate_dead_tasks(unsigned int dead_cpu) * migration_call - callback that gets triggered when a CPU is added. * Here we can start up the necessary migration thread for the new CPU. */ -static int migration_call(struct notifier_block *nfb, unsigned long action, - void *hcpu) +static int __cpuinit +migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) { - int cpu = (long)hcpu; struct task_struct *p; - struct runqueue *rq; + int cpu = (long)hcpu; unsigned long flags; + struct rq *rq; switch (action) { case CPU_UP_PREPARE: @@ -4552,17 +5357,23 @@ static int migration_call(struct notifier_block *nfb, unsigned long action, task_rq_unlock(rq, &flags); cpu_rq(cpu)->migration_thread = p; break; + case CPU_ONLINE: /* Strictly unneccessary, as first user will wake it. */ wake_up_process(cpu_rq(cpu)->migration_thread); break; + #ifdef CONFIG_HOTPLUG_CPU case CPU_UP_CANCELED: + if (!cpu_rq(cpu)->migration_thread) + break; /* Unbind it from offline cpu so it can run. Fall thru. */ - kthread_bind(cpu_rq(cpu)->migration_thread,smp_processor_id()); + kthread_bind(cpu_rq(cpu)->migration_thread, + any_online_cpu(cpu_online_map)); kthread_stop(cpu_rq(cpu)->migration_thread); cpu_rq(cpu)->migration_thread = NULL; break; + case CPU_DEAD: migrate_live_tasks(cpu); rq = cpu_rq(cpu); @@ -4583,10 +5394,10 @@ static int migration_call(struct notifier_block *nfb, unsigned long action, * the requestors. */ spin_lock_irq(&rq->lock); while (!list_empty(&rq->migration_queue)) { - migration_req_t *req; + struct migration_req *req; + req = list_entry(rq->migration_queue.next, - migration_req_t, list); - BUG_ON(req->type != REQ_MOVE_TASK); + struct migration_req, list); list_del_init(&req->list); complete(&req->done); } @@ -4600,7 +5411,7 @@ static int migration_call(struct notifier_block *nfb, unsigned long action, /* Register at highest priority so that task migration (migrate_all_tasks) * happens before everything else. */ -static struct notifier_block __devinitdata migration_notifier = { +static struct notifier_block __cpuinitdata migration_notifier = { .notifier_call = migration_call, .priority = 10 }; @@ -4608,21 +5419,28 @@ static struct notifier_block __devinitdata migration_notifier = { int __init migration_init(void) { void *cpu = (void *)(long)smp_processor_id(); - /* Start one for boot CPU. */ + + /* Start one for the boot CPU: */ migration_call(&migration_notifier, CPU_UP_PREPARE, cpu); migration_call(&migration_notifier, CPU_ONLINE, cpu); register_cpu_notifier(&migration_notifier); + return 0; } #endif #ifdef CONFIG_SMP -#define SCHED_DOMAIN_DEBUG +#undef SCHED_DOMAIN_DEBUG #ifdef SCHED_DOMAIN_DEBUG static void sched_domain_debug(struct sched_domain *sd, int cpu) { int level = 0; + if (!sd) { + printk(KERN_DEBUG "CPU%d attaching NULL sched-domain.\n", cpu); + return; + } + printk(KERN_DEBUG "CPU%d attaching sched-domain:\n", cpu); do { @@ -4702,44 +5520,88 @@ static void sched_domain_debug(struct sched_domain *sd, int cpu) } while (sd); } #else -#define sched_domain_debug(sd, cpu) {} +# define sched_domain_debug(sd, cpu) do { } while (0) #endif +static int sd_degenerate(struct sched_domain *sd) +{ + if (cpus_weight(sd->span) == 1) + return 1; + + /* Following flags need at least 2 groups */ + if (sd->flags & (SD_LOAD_BALANCE | + SD_BALANCE_NEWIDLE | + SD_BALANCE_FORK | + SD_BALANCE_EXEC)) { + if (sd->groups != sd->groups->next) + return 0; + } + + /* Following flags don't use groups */ + if (sd->flags & (SD_WAKE_IDLE | + SD_WAKE_AFFINE | + SD_WAKE_BALANCE)) + return 0; + + return 1; +} + +static int +sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent) +{ + unsigned long cflags = sd->flags, pflags = parent->flags; + + if (sd_degenerate(parent)) + return 1; + + if (!cpus_equal(sd->span, parent->span)) + return 0; + + /* Does parent contain flags not in child? */ + /* WAKE_BALANCE is a subset of WAKE_AFFINE */ + if (cflags & SD_WAKE_AFFINE) + pflags &= ~SD_WAKE_BALANCE; + /* Flags needing groups don't count if only 1 group in parent */ + if (parent->groups == parent->groups->next) { + pflags &= ~(SD_LOAD_BALANCE | + SD_BALANCE_NEWIDLE | + SD_BALANCE_FORK | + SD_BALANCE_EXEC); + } + if (~cflags & pflags) + return 0; + + return 1; +} + /* * Attach the domain 'sd' to 'cpu' as its base domain. Callers must * hold the hotplug lock. */ -void __devinit cpu_attach_domain(struct sched_domain *sd, int cpu) +static void cpu_attach_domain(struct sched_domain *sd, int cpu) { - migration_req_t req; - unsigned long flags; - runqueue_t *rq = cpu_rq(cpu); - int local = 1; - - sched_domain_debug(sd, cpu); - - spin_lock_irqsave(&rq->lock, flags); + struct rq *rq = cpu_rq(cpu); + struct sched_domain *tmp; - if (cpu == smp_processor_id() || !cpu_online(cpu)) { - rq->sd = sd; - } else { - init_completion(&req.done); - req.type = REQ_SET_DOMAIN; - req.sd = sd; - list_add(&req.list, &rq->migration_queue); - local = 0; + /* Remove the sched domains which do not contribute to scheduling. */ + for (tmp = sd; tmp; tmp = tmp->parent) { + struct sched_domain *parent = tmp->parent; + if (!parent) + break; + if (sd_parent_degenerate(tmp, parent)) + tmp->parent = parent->parent; } - spin_unlock_irqrestore(&rq->lock, flags); + if (sd && sd_degenerate(sd)) + sd = sd->parent; - if (!local) { - wake_up_process(rq->migration_thread); - wait_for_completion(&req.done); - } + sched_domain_debug(sd, cpu); + + rcu_assign_pointer(rq->sd, sd); } /* cpus with isolated domains */ -cpumask_t __devinitdata cpu_isolated_map = CPU_MASK_NONE; +static cpumask_t __devinitdata cpu_isolated_map = CPU_MASK_NONE; /* Setup the mask of cpus configured for isolated domains */ static int __init isolated_cpu_setup(char *str) @@ -4767,8 +5629,8 @@ __setup ("isolcpus=", isolated_cpu_setup); * covered by the given span, and will set each group's ->cpumask correctly, * and ->cpu_power to 0. */ -void __devinit init_sched_build_groups(struct sched_group groups[], - cpumask_t span, int (*group_fn)(int cpu)) +static void init_sched_build_groups(struct sched_group groups[], cpumask_t span, + int (*group_fn)(int cpu)) { struct sched_group *first = NULL, *last = NULL; cpumask_t covered = CPU_MASK_NONE; @@ -4801,223 +5663,1209 @@ void __devinit init_sched_build_groups(struct sched_group groups[], last->next = first; } +#define SD_NODES_PER_DOMAIN 16 + +/* + * Self-tuning task migration cost measurement between source and target CPUs. + * + * This is done by measuring the cost of manipulating buffers of varying + * sizes. For a given buffer-size here are the steps that are taken: + * + * 1) the source CPU reads+dirties a shared buffer + * 2) the target CPU reads+dirties the same shared buffer + * + * We measure how long they take, in the following 4 scenarios: + * + * - source: CPU1, target: CPU2 | cost1 + * - source: CPU2, target: CPU1 | cost2 + * - source: CPU1, target: CPU1 | cost3 + * - source: CPU2, target: CPU2 | cost4 + * + * We then calculate the cost3+cost4-cost1-cost2 difference - this is + * the cost of migration. + * + * We then start off from a small buffer-size and iterate up to larger + * buffer sizes, in 5% steps - measuring each buffer-size separately, and + * doing a maximum search for the cost. (The maximum cost for a migration + * normally occurs when the working set size is around the effective cache + * size.) + */ +#define SEARCH_SCOPE 2 +#define MIN_CACHE_SIZE (64*1024U) +#define DEFAULT_CACHE_SIZE (5*1024*1024U) +#define ITERATIONS 1 +#define SIZE_THRESH 130 +#define COST_THRESH 130 + +/* + * The migration cost is a function of 'domain distance'. Domain + * distance is the number of steps a CPU has to iterate down its + * domain tree to share a domain with the other CPU. The farther + * two CPUs are from each other, the larger the distance gets. + * + * Note that we use the distance only to cache measurement results, + * the distance value is not used numerically otherwise. When two + * CPUs have the same distance it is assumed that the migration + * cost is the same. (this is a simplification but quite practical) + */ +#define MAX_DOMAIN_DISTANCE 32 -#ifdef ARCH_HAS_SCHED_DOMAIN -extern void __devinit arch_init_sched_domains(void); -extern void __devinit arch_destroy_sched_domains(void); +static unsigned long long migration_cost[MAX_DOMAIN_DISTANCE] = + { [ 0 ... MAX_DOMAIN_DISTANCE-1 ] = +/* + * Architectures may override the migration cost and thus avoid + * boot-time calibration. Unit is nanoseconds. Mostly useful for + * virtualized hardware: + */ +#ifdef CONFIG_DEFAULT_MIGRATION_COST + CONFIG_DEFAULT_MIGRATION_COST #else -#ifdef CONFIG_SCHED_SMT -static DEFINE_PER_CPU(struct sched_domain, cpu_domains); -static struct sched_group sched_group_cpus[NR_CPUS]; -static int __devinit cpu_to_cpu_group(int cpu) + -1LL +#endif +}; + +/* + * Allow override of migration cost - in units of microseconds. + * E.g. migration_cost=1000,2000,3000 will set up a level-1 cost + * of 1 msec, level-2 cost of 2 msecs and level3 cost of 3 msecs: + */ +static int __init migration_cost_setup(char *str) { - return cpu; + int ints[MAX_DOMAIN_DISTANCE+1], i; + + str = get_options(str, ARRAY_SIZE(ints), ints); + + printk("#ints: %d\n", ints[0]); + for (i = 1; i <= ints[0]; i++) { + migration_cost[i-1] = (unsigned long long)ints[i]*1000; + printk("migration_cost[%d]: %Ld\n", i-1, migration_cost[i-1]); + } + return 1; } -#endif -static DEFINE_PER_CPU(struct sched_domain, phys_domains); -static struct sched_group sched_group_phys[NR_CPUS]; -static int __devinit cpu_to_phys_group(int cpu) +__setup ("migration_cost=", migration_cost_setup); + +/* + * Global multiplier (divisor) for migration-cutoff values, + * in percentiles. E.g. use a value of 150 to get 1.5 times + * longer cache-hot cutoff times. + * + * (We scale it from 100 to 128 to long long handling easier.) + */ + +#define MIGRATION_FACTOR_SCALE 128 + +static unsigned int migration_factor = MIGRATION_FACTOR_SCALE; + +static int __init setup_migration_factor(char *str) { -#ifdef CONFIG_SCHED_SMT - return first_cpu(cpu_sibling_map[cpu]); -#else - return cpu; -#endif + get_option(&str, &migration_factor); + migration_factor = migration_factor * MIGRATION_FACTOR_SCALE / 100; + return 1; } -#ifdef CONFIG_NUMA +__setup("migration_factor=", setup_migration_factor); -static DEFINE_PER_CPU(struct sched_domain, node_domains); -static struct sched_group sched_group_nodes[MAX_NUMNODES]; -static int __devinit cpu_to_node_group(int cpu) +/* + * Estimated distance of two CPUs, measured via the number of domains + * we have to pass for the two CPUs to be in the same span: + */ +static unsigned long domain_distance(int cpu1, int cpu2) { - return cpu_to_node(cpu); + unsigned long distance = 0; + struct sched_domain *sd; + + for_each_domain(cpu1, sd) { + WARN_ON(!cpu_isset(cpu1, sd->span)); + if (cpu_isset(cpu2, sd->span)) + return distance; + distance++; + } + if (distance >= MAX_DOMAIN_DISTANCE) { + WARN_ON(1); + distance = MAX_DOMAIN_DISTANCE-1; + } + + return distance; +} + +static unsigned int migration_debug; + +static int __init setup_migration_debug(char *str) +{ + get_option(&str, &migration_debug); + return 1; } -#endif -#if defined(CONFIG_SCHED_SMT) && defined(CONFIG_NUMA) +__setup("migration_debug=", setup_migration_debug); + /* - * The domains setup code relies on siblings not spanning - * multiple nodes. Make sure the architecture has a proper - * siblings map: + * Maximum cache-size that the scheduler should try to measure. + * Architectures with larger caches should tune this up during + * bootup. Gets used in the domain-setup code (i.e. during SMP + * bootup). */ -static void check_sibling_maps(void) +unsigned int max_cache_size; + +static int __init setup_max_cache_size(char *str) { - int i, j; + get_option(&str, &max_cache_size); + return 1; +} - for_each_online_cpu(i) { - for_each_cpu_mask(j, cpu_sibling_map[i]) { - if (cpu_to_node(i) != cpu_to_node(j)) { - printk(KERN_INFO "warning: CPU %d siblings map " - "to different node - isolating " - "them.\n", i); - cpu_sibling_map[i] = cpumask_of_cpu(i); - break; - } +__setup("max_cache_size=", setup_max_cache_size); + +/* + * Dirty a big buffer in a hard-to-predict (for the L2 cache) way. This + * is the operation that is timed, so we try to generate unpredictable + * cachemisses that still end up filling the L2 cache: + */ +static void touch_cache(void *__cache, unsigned long __size) +{ + unsigned long size = __size/sizeof(long), chunk1 = size/3, + chunk2 = 2*size/3; + unsigned long *cache = __cache; + int i; + + for (i = 0; i < size/6; i += 8) { + switch (i % 6) { + case 0: cache[i]++; + case 1: cache[size-1-i]++; + case 2: cache[chunk1-i]++; + case 3: cache[chunk1+i]++; + case 4: cache[chunk2-i]++; + case 5: cache[chunk2+i]++; } } } -#endif /* - * Set up scheduler domains and groups. Callers must hold the hotplug lock. + * Measure the cache-cost of one task migration. Returns in units of nsec. */ -static void __devinit arch_init_sched_domains(void) +static unsigned long long +measure_one(void *cache, unsigned long size, int source, int target) { - int i; - cpumask_t cpu_default_map; + cpumask_t mask, saved_mask; + unsigned long long t0, t1, t2, t3, cost; + + saved_mask = current->cpus_allowed; -#if defined(CONFIG_SCHED_SMT) && defined(CONFIG_NUMA) - check_sibling_maps(); -#endif /* - * Setup mask for cpus without special case scheduling requirements. - * For now this just excludes isolated cpus, but could be used to - * exclude other special cases in the future. + * Flush source caches to RAM and invalidate them: */ - cpus_complement(cpu_default_map, cpu_isolated_map); - cpus_and(cpu_default_map, cpu_default_map, cpu_online_map); + sched_cacheflush(); /* - * Set up domains. Isolated domains just stay on the dummy domain. + * Migrate to the source CPU: */ - for_each_cpu_mask(i, cpu_default_map) { - int group; - struct sched_domain *sd = NULL, *p; - cpumask_t nodemask = node_to_cpumask(cpu_to_node(i)); - - cpus_and(nodemask, nodemask, cpu_default_map); + mask = cpumask_of_cpu(source); + set_cpus_allowed(current, mask); + WARN_ON(smp_processor_id() != source); -#ifdef CONFIG_NUMA - sd = &per_cpu(node_domains, i); - group = cpu_to_node_group(i); - *sd = SD_NODE_INIT; - sd->span = cpu_default_map; - sd->groups = &sched_group_nodes[group]; -#endif + /* + * Dirty the working set: + */ + t0 = sched_clock(); + touch_cache(cache, size); + t1 = sched_clock(); - p = sd; - sd = &per_cpu(phys_domains, i); - group = cpu_to_phys_group(i); - *sd = SD_CPU_INIT; - sd->span = nodemask; - sd->parent = p; - sd->groups = &sched_group_phys[group]; + /* + * Migrate to the target CPU, dirty the L2 cache and access + * the shared buffer. (which represents the working set + * of a migrated task.) + */ + mask = cpumask_of_cpu(target); + set_cpus_allowed(current, mask); + WARN_ON(smp_processor_id() != target); -#ifdef CONFIG_SCHED_SMT - p = sd; - sd = &per_cpu(cpu_domains, i); - group = cpu_to_cpu_group(i); - *sd = SD_SIBLING_INIT; - sd->span = cpu_sibling_map[i]; - cpus_and(sd->span, sd->span, cpu_default_map); - sd->parent = p; - sd->groups = &sched_group_cpus[group]; -#endif - } + t2 = sched_clock(); + touch_cache(cache, size); + t3 = sched_clock(); -#ifdef CONFIG_SCHED_SMT - /* Set up CPU (sibling) groups */ - for_each_online_cpu(i) { - cpumask_t this_sibling_map = cpu_sibling_map[i]; - cpus_and(this_sibling_map, this_sibling_map, cpu_default_map); - if (i != first_cpu(this_sibling_map)) - continue; + cost = t1-t0 + t3-t2; - init_sched_build_groups(sched_group_cpus, this_sibling_map, - &cpu_to_cpu_group); - } -#endif + if (migration_debug >= 2) + printk("[%d->%d]: %8Ld %8Ld %8Ld => %10Ld.\n", + source, target, t1-t0, t1-t0, t3-t2, cost); + /* + * Flush target caches to RAM and invalidate them: + */ + sched_cacheflush(); - /* Set up physical groups */ - for (i = 0; i < MAX_NUMNODES; i++) { - cpumask_t nodemask = node_to_cpumask(i); + set_cpus_allowed(current, saved_mask); - cpus_and(nodemask, nodemask, cpu_default_map); - if (cpus_empty(nodemask)) - continue; + return cost; +} - init_sched_build_groups(sched_group_phys, nodemask, +/* + * Measure a series of task migrations and return the average + * result. Since this code runs early during bootup the system + * is 'undisturbed' and the average latency makes sense. + * + * The algorithm in essence auto-detects the relevant cache-size, + * so it will properly detect different cachesizes for different + * cache-hierarchies, depending on how the CPUs are connected. + * + * Architectures can prime the upper limit of the search range via + * max_cache_size, otherwise the search range defaults to 20MB...64K. + */ +static unsigned long long +measure_cost(int cpu1, int cpu2, void *cache, unsigned int size) +{ + unsigned long long cost1, cost2; + int i; + + /* + * Measure the migration cost of 'size' bytes, over an + * average of 10 runs: + * + * (We perturb the cache size by a small (0..4k) + * value to compensate size/alignment related artifacts. + * We also subtract the cost of the operation done on + * the same CPU.) + */ + cost1 = 0; + + /* + * dry run, to make sure we start off cache-cold on cpu1, + * and to get any vmalloc pagefaults in advance: + */ + measure_one(cache, size, cpu1, cpu2); + for (i = 0; i < ITERATIONS; i++) + cost1 += measure_one(cache, size - i*1024, cpu1, cpu2); + + measure_one(cache, size, cpu2, cpu1); + for (i = 0; i < ITERATIONS; i++) + cost1 += measure_one(cache, size - i*1024, cpu2, cpu1); + + /* + * (We measure the non-migrating [cached] cost on both + * cpu1 and cpu2, to handle CPUs with different speeds) + */ + cost2 = 0; + + measure_one(cache, size, cpu1, cpu1); + for (i = 0; i < ITERATIONS; i++) + cost2 += measure_one(cache, size - i*1024, cpu1, cpu1); + + measure_one(cache, size, cpu2, cpu2); + for (i = 0; i < ITERATIONS; i++) + cost2 += measure_one(cache, size - i*1024, cpu2, cpu2); + + /* + * Get the per-iteration migration cost: + */ + do_div(cost1, 2*ITERATIONS); + do_div(cost2, 2*ITERATIONS); + + return cost1 - cost2; +} + +static unsigned long long measure_migration_cost(int cpu1, int cpu2) +{ + unsigned long long max_cost = 0, fluct = 0, avg_fluct = 0; + unsigned int max_size, size, size_found = 0; + long long cost = 0, prev_cost; + void *cache; + + /* + * Search from max_cache_size*5 down to 64K - the real relevant + * cachesize has to lie somewhere inbetween. + */ + if (max_cache_size) { + max_size = max(max_cache_size * SEARCH_SCOPE, MIN_CACHE_SIZE); + size = max(max_cache_size / SEARCH_SCOPE, MIN_CACHE_SIZE); + } else { + /* + * Since we have no estimation about the relevant + * search range + */ + max_size = DEFAULT_CACHE_SIZE * SEARCH_SCOPE; + size = MIN_CACHE_SIZE; + } + + if (!cpu_online(cpu1) || !cpu_online(cpu2)) { + printk("cpu %d and %d not both online!\n", cpu1, cpu2); + return 0; + } + + /* + * Allocate the working set: + */ + cache = vmalloc(max_size); + if (!cache) { + printk("could not vmalloc %d bytes for cache!\n", 2*max_size); + return 1000000; /* return 1 msec on very small boxen */ + } + + while (size <= max_size) { + prev_cost = cost; + cost = measure_cost(cpu1, cpu2, cache, size); + + /* + * Update the max: + */ + if (cost > 0) { + if (max_cost < cost) { + max_cost = cost; + size_found = size; + } + } + /* + * Calculate average fluctuation, we use this to prevent + * noise from triggering an early break out of the loop: + */ + fluct = abs(cost - prev_cost); + avg_fluct = (avg_fluct + fluct)/2; + + if (migration_debug) + printk("-> [%d][%d][%7d] %3ld.%ld [%3ld.%ld] (%ld): (%8Ld %8Ld)\n", + cpu1, cpu2, size, + (long)cost / 1000000, + ((long)cost / 100000) % 10, + (long)max_cost / 1000000, + ((long)max_cost / 100000) % 10, + domain_distance(cpu1, cpu2), + cost, avg_fluct); + + /* + * If we iterated at least 20% past the previous maximum, + * and the cost has dropped by more than 20% already, + * (taking fluctuations into account) then we assume to + * have found the maximum and break out of the loop early: + */ + if (size_found && (size*100 > size_found*SIZE_THRESH)) + if (cost+avg_fluct <= 0 || + max_cost*100 > (cost+avg_fluct)*COST_THRESH) { + + if (migration_debug) + printk("-> found max.\n"); + break; + } + /* + * Increase the cachesize in 10% steps: + */ + size = size * 10 / 9; + } + + if (migration_debug) + printk("[%d][%d] working set size found: %d, cost: %Ld\n", + cpu1, cpu2, size_found, max_cost); + + vfree(cache); + + /* + * A task is considered 'cache cold' if at least 2 times + * the worst-case cost of migration has passed. + * + * (this limit is only listened to if the load-balancing + * situation is 'nice' - if there is a large imbalance we + * ignore it for the sake of CPU utilization and + * processing fairness.) + */ + return 2 * max_cost * migration_factor / MIGRATION_FACTOR_SCALE; +} + +static void calibrate_migration_costs(const cpumask_t *cpu_map) +{ + int cpu1 = -1, cpu2 = -1, cpu, orig_cpu = raw_smp_processor_id(); + unsigned long j0, j1, distance, max_distance = 0; + struct sched_domain *sd; + + j0 = jiffies; + + /* + * First pass - calculate the cacheflush times: + */ + for_each_cpu_mask(cpu1, *cpu_map) { + for_each_cpu_mask(cpu2, *cpu_map) { + if (cpu1 == cpu2) + continue; + distance = domain_distance(cpu1, cpu2); + max_distance = max(max_distance, distance); + /* + * No result cached yet? + */ + if (migration_cost[distance] == -1LL) + migration_cost[distance] = + measure_migration_cost(cpu1, cpu2); + } + } + /* + * Second pass - update the sched domain hierarchy with + * the new cache-hot-time estimations: + */ + for_each_cpu_mask(cpu, *cpu_map) { + distance = 0; + for_each_domain(cpu, sd) { + sd->cache_hot_time = migration_cost[distance]; + distance++; + } + } + /* + * Print the matrix: + */ + if (migration_debug) + printk("migration: max_cache_size: %d, cpu: %d MHz:\n", + max_cache_size, +#ifdef CONFIG_X86 + cpu_khz/1000 +#else + -1 +#endif + ); + if (system_state == SYSTEM_BOOTING) { + if (num_online_cpus() > 1) { + printk("migration_cost="); + for (distance = 0; distance <= max_distance; distance++) { + if (distance) + printk(","); + printk("%ld", (long)migration_cost[distance] / 1000); + } + printk("\n"); + } + } + j1 = jiffies; + if (migration_debug) + printk("migration: %ld seconds\n", (j1-j0)/HZ); + + /* + * Move back to the original CPU. NUMA-Q gets confused + * if we migrate to another quad during bootup. + */ + if (raw_smp_processor_id() != orig_cpu) { + cpumask_t mask = cpumask_of_cpu(orig_cpu), + saved_mask = current->cpus_allowed; + + set_cpus_allowed(current, mask); + set_cpus_allowed(current, saved_mask); + } +} + +#ifdef CONFIG_NUMA + +/** + * find_next_best_node - find the next node to include in a sched_domain + * @node: node whose sched_domain we're building + * @used_nodes: nodes already in the sched_domain + * + * Find the next node to include in a given scheduling domain. Simply + * finds the closest node not already in the @used_nodes map. + * + * Should use nodemask_t. + */ +static int find_next_best_node(int node, unsigned long *used_nodes) +{ + int i, n, val, min_val, best_node = 0; + + min_val = INT_MAX; + + for (i = 0; i < MAX_NUMNODES; i++) { + /* Start at @node */ + n = (node + i) % MAX_NUMNODES; + + if (!nr_cpus_node(n)) + continue; + + /* Skip already used nodes */ + if (test_bit(n, used_nodes)) + continue; + + /* Simple min distance search */ + val = node_distance(node, n); + + if (val < min_val) { + min_val = val; + best_node = n; + } + } + + set_bit(best_node, used_nodes); + return best_node; +} + +/** + * sched_domain_node_span - get a cpumask for a node's sched_domain + * @node: node whose cpumask we're constructing + * @size: number of nodes to include in this span + * + * Given a node, construct a good cpumask for its sched_domain to span. It + * should be one that prevents unnecessary balancing, but also spreads tasks + * out optimally. + */ +static cpumask_t sched_domain_node_span(int node) +{ + DECLARE_BITMAP(used_nodes, MAX_NUMNODES); + cpumask_t span, nodemask; + int i; + + cpus_clear(span); + bitmap_zero(used_nodes, MAX_NUMNODES); + + nodemask = node_to_cpumask(node); + cpus_or(span, span, nodemask); + set_bit(node, used_nodes); + + for (i = 1; i < SD_NODES_PER_DOMAIN; i++) { + int next_node = find_next_best_node(node, used_nodes); + + nodemask = node_to_cpumask(next_node); + cpus_or(span, span, nodemask); + } + + return span; +} +#endif + +int sched_smt_power_savings = 0, sched_mc_power_savings = 0; + +/* + * SMT sched-domains: + */ +#ifdef CONFIG_SCHED_SMT +static DEFINE_PER_CPU(struct sched_domain, cpu_domains); +static struct sched_group sched_group_cpus[NR_CPUS]; + +static int cpu_to_cpu_group(int cpu) +{ + return cpu; +} +#endif + +/* + * multi-core sched-domains: + */ +#ifdef CONFIG_SCHED_MC +static DEFINE_PER_CPU(struct sched_domain, core_domains); +static struct sched_group *sched_group_core_bycpu[NR_CPUS]; +#endif + +#if defined(CONFIG_SCHED_MC) && defined(CONFIG_SCHED_SMT) +static int cpu_to_core_group(int cpu) +{ + return first_cpu(cpu_sibling_map[cpu]); +} +#elif defined(CONFIG_SCHED_MC) +static int cpu_to_core_group(int cpu) +{ + return cpu; +} +#endif + +static DEFINE_PER_CPU(struct sched_domain, phys_domains); +static struct sched_group *sched_group_phys_bycpu[NR_CPUS]; + +static int cpu_to_phys_group(int cpu) +{ +#ifdef CONFIG_SCHED_MC + cpumask_t mask = cpu_coregroup_map(cpu); + return first_cpu(mask); +#elif defined(CONFIG_SCHED_SMT) + return first_cpu(cpu_sibling_map[cpu]); +#else + return cpu; +#endif +} + +#ifdef CONFIG_NUMA +/* + * The init_sched_build_groups can't handle what we want to do with node + * groups, so roll our own. Now each node has its own list of groups which + * gets dynamically allocated. + */ +static DEFINE_PER_CPU(struct sched_domain, node_domains); +static struct sched_group **sched_group_nodes_bycpu[NR_CPUS]; + +static DEFINE_PER_CPU(struct sched_domain, allnodes_domains); +static struct sched_group *sched_group_allnodes_bycpu[NR_CPUS]; + +static int cpu_to_allnodes_group(int cpu) +{ + return cpu_to_node(cpu); +} +static void init_numa_sched_groups_power(struct sched_group *group_head) +{ + struct sched_group *sg = group_head; + int j; + + if (!sg) + return; +next_sg: + for_each_cpu_mask(j, sg->cpumask) { + struct sched_domain *sd; + + sd = &per_cpu(phys_domains, j); + if (j != first_cpu(sd->groups->cpumask)) { + /* + * Only add "power" once for each + * physical package. + */ + continue; + } + + sg->cpu_power += sd->groups->cpu_power; + } + sg = sg->next; + if (sg != group_head) + goto next_sg; +} +#endif + +/* Free memory allocated for various sched_group structures */ +static void free_sched_groups(const cpumask_t *cpu_map) +{ + int cpu; +#ifdef CONFIG_NUMA + int i; + + for_each_cpu_mask(cpu, *cpu_map) { + struct sched_group *sched_group_allnodes + = sched_group_allnodes_bycpu[cpu]; + struct sched_group **sched_group_nodes + = sched_group_nodes_bycpu[cpu]; + + if (sched_group_allnodes) { + kfree(sched_group_allnodes); + sched_group_allnodes_bycpu[cpu] = NULL; + } + + if (!sched_group_nodes) + continue; + + for (i = 0; i < MAX_NUMNODES; i++) { + cpumask_t nodemask = node_to_cpumask(i); + struct sched_group *oldsg, *sg = sched_group_nodes[i]; + + cpus_and(nodemask, nodemask, *cpu_map); + if (cpus_empty(nodemask)) + continue; + + if (sg == NULL) + continue; + sg = sg->next; +next_sg: + oldsg = sg; + sg = sg->next; + kfree(oldsg); + if (oldsg != sched_group_nodes[i]) + goto next_sg; + } + kfree(sched_group_nodes); + sched_group_nodes_bycpu[cpu] = NULL; + } +#endif + for_each_cpu_mask(cpu, *cpu_map) { + if (sched_group_phys_bycpu[cpu]) { + kfree(sched_group_phys_bycpu[cpu]); + sched_group_phys_bycpu[cpu] = NULL; + } +#ifdef CONFIG_SCHED_MC + if (sched_group_core_bycpu[cpu]) { + kfree(sched_group_core_bycpu[cpu]); + sched_group_core_bycpu[cpu] = NULL; + } +#endif + } +} + +/* + * Build sched domains for a given set of cpus and attach the sched domains + * to the individual cpus + */ +static int build_sched_domains(const cpumask_t *cpu_map) +{ + int i; + struct sched_group *sched_group_phys = NULL; +#ifdef CONFIG_SCHED_MC + struct sched_group *sched_group_core = NULL; +#endif +#ifdef CONFIG_NUMA + struct sched_group **sched_group_nodes = NULL; + struct sched_group *sched_group_allnodes = NULL; + + /* + * Allocate the per-node list of sched groups + */ + sched_group_nodes = kzalloc(sizeof(struct sched_group*)*MAX_NUMNODES, + GFP_KERNEL); + if (!sched_group_nodes) { + printk(KERN_WARNING "Can not alloc sched group node list\n"); + return -ENOMEM; + } + sched_group_nodes_bycpu[first_cpu(*cpu_map)] = sched_group_nodes; +#endif + + /* + * Set up domains for cpus specified by the cpu_map. + */ + for_each_cpu_mask(i, *cpu_map) { + int group; + struct sched_domain *sd = NULL, *p; + cpumask_t nodemask = node_to_cpumask(cpu_to_node(i)); + + cpus_and(nodemask, nodemask, *cpu_map); + +#ifdef CONFIG_NUMA + if (cpus_weight(*cpu_map) + > SD_NODES_PER_DOMAIN*cpus_weight(nodemask)) { + if (!sched_group_allnodes) { + sched_group_allnodes + = kmalloc(sizeof(struct sched_group) + * MAX_NUMNODES, + GFP_KERNEL); + if (!sched_group_allnodes) { + printk(KERN_WARNING + "Can not alloc allnodes sched group\n"); + goto error; + } + sched_group_allnodes_bycpu[i] + = sched_group_allnodes; + } + sd = &per_cpu(allnodes_domains, i); + *sd = SD_ALLNODES_INIT; + sd->span = *cpu_map; + group = cpu_to_allnodes_group(i); + sd->groups = &sched_group_allnodes[group]; + p = sd; + } else + p = NULL; + + sd = &per_cpu(node_domains, i); + *sd = SD_NODE_INIT; + sd->span = sched_domain_node_span(cpu_to_node(i)); + sd->parent = p; + cpus_and(sd->span, sd->span, *cpu_map); +#endif + + if (!sched_group_phys) { + sched_group_phys + = kmalloc(sizeof(struct sched_group) * NR_CPUS, + GFP_KERNEL); + if (!sched_group_phys) { + printk (KERN_WARNING "Can not alloc phys sched" + "group\n"); + goto error; + } + sched_group_phys_bycpu[i] = sched_group_phys; + } + + p = sd; + sd = &per_cpu(phys_domains, i); + group = cpu_to_phys_group(i); + *sd = SD_CPU_INIT; + sd->span = nodemask; + sd->parent = p; + sd->groups = &sched_group_phys[group]; + +#ifdef CONFIG_SCHED_MC + if (!sched_group_core) { + sched_group_core + = kmalloc(sizeof(struct sched_group) * NR_CPUS, + GFP_KERNEL); + if (!sched_group_core) { + printk (KERN_WARNING "Can not alloc core sched" + "group\n"); + goto error; + } + sched_group_core_bycpu[i] = sched_group_core; + } + + p = sd; + sd = &per_cpu(core_domains, i); + group = cpu_to_core_group(i); + *sd = SD_MC_INIT; + sd->span = cpu_coregroup_map(i); + cpus_and(sd->span, sd->span, *cpu_map); + sd->parent = p; + sd->groups = &sched_group_core[group]; +#endif + +#ifdef CONFIG_SCHED_SMT + p = sd; + sd = &per_cpu(cpu_domains, i); + group = cpu_to_cpu_group(i); + *sd = SD_SIBLING_INIT; + sd->span = cpu_sibling_map[i]; + cpus_and(sd->span, sd->span, *cpu_map); + sd->parent = p; + sd->groups = &sched_group_cpus[group]; +#endif + } + +#ifdef CONFIG_SCHED_SMT + /* Set up CPU (sibling) groups */ + for_each_cpu_mask(i, *cpu_map) { + cpumask_t this_sibling_map = cpu_sibling_map[i]; + cpus_and(this_sibling_map, this_sibling_map, *cpu_map); + if (i != first_cpu(this_sibling_map)) + continue; + + init_sched_build_groups(sched_group_cpus, this_sibling_map, + &cpu_to_cpu_group); + } +#endif + +#ifdef CONFIG_SCHED_MC + /* Set up multi-core groups */ + for_each_cpu_mask(i, *cpu_map) { + cpumask_t this_core_map = cpu_coregroup_map(i); + cpus_and(this_core_map, this_core_map, *cpu_map); + if (i != first_cpu(this_core_map)) + continue; + init_sched_build_groups(sched_group_core, this_core_map, + &cpu_to_core_group); + } +#endif + + + /* Set up physical groups */ + for (i = 0; i < MAX_NUMNODES; i++) { + cpumask_t nodemask = node_to_cpumask(i); + + cpus_and(nodemask, nodemask, *cpu_map); + if (cpus_empty(nodemask)) + continue; + + init_sched_build_groups(sched_group_phys, nodemask, &cpu_to_phys_group); } #ifdef CONFIG_NUMA /* Set up node groups */ - init_sched_build_groups(sched_group_nodes, cpu_default_map, - &cpu_to_node_group); + if (sched_group_allnodes) + init_sched_build_groups(sched_group_allnodes, *cpu_map, + &cpu_to_allnodes_group); + + for (i = 0; i < MAX_NUMNODES; i++) { + /* Set up node groups */ + struct sched_group *sg, *prev; + cpumask_t nodemask = node_to_cpumask(i); + cpumask_t domainspan; + cpumask_t covered = CPU_MASK_NONE; + int j; + + cpus_and(nodemask, nodemask, *cpu_map); + if (cpus_empty(nodemask)) { + sched_group_nodes[i] = NULL; + continue; + } + + domainspan = sched_domain_node_span(i); + cpus_and(domainspan, domainspan, *cpu_map); + + sg = kmalloc_node(sizeof(struct sched_group), GFP_KERNEL, i); + if (!sg) { + printk(KERN_WARNING "Can not alloc domain group for " + "node %d\n", i); + goto error; + } + sched_group_nodes[i] = sg; + for_each_cpu_mask(j, nodemask) { + struct sched_domain *sd; + sd = &per_cpu(node_domains, j); + sd->groups = sg; + } + sg->cpu_power = 0; + sg->cpumask = nodemask; + sg->next = sg; + cpus_or(covered, covered, nodemask); + prev = sg; + + for (j = 0; j < MAX_NUMNODES; j++) { + cpumask_t tmp, notcovered; + int n = (i + j) % MAX_NUMNODES; + + cpus_complement(notcovered, covered); + cpus_and(tmp, notcovered, *cpu_map); + cpus_and(tmp, tmp, domainspan); + if (cpus_empty(tmp)) + break; + + nodemask = node_to_cpumask(n); + cpus_and(tmp, tmp, nodemask); + if (cpus_empty(tmp)) + continue; + + sg = kmalloc_node(sizeof(struct sched_group), + GFP_KERNEL, i); + if (!sg) { + printk(KERN_WARNING + "Can not alloc domain group for node %d\n", j); + goto error; + } + sg->cpu_power = 0; + sg->cpumask = tmp; + sg->next = prev->next; + cpus_or(covered, covered, tmp); + prev->next = sg; + prev = sg; + } + } #endif /* Calculate CPU power for physical packages and nodes */ - for_each_cpu_mask(i, cpu_default_map) { - int power; - struct sched_domain *sd; #ifdef CONFIG_SCHED_SMT + for_each_cpu_mask(i, *cpu_map) { + struct sched_domain *sd; sd = &per_cpu(cpu_domains, i); - power = SCHED_LOAD_SCALE; + sd->groups->cpu_power = SCHED_LOAD_SCALE; + } +#endif +#ifdef CONFIG_SCHED_MC + for_each_cpu_mask(i, *cpu_map) { + int power; + struct sched_domain *sd; + sd = &per_cpu(core_domains, i); + if (sched_smt_power_savings) + power = SCHED_LOAD_SCALE * cpus_weight(sd->groups->cpumask); + else + power = SCHED_LOAD_SCALE + (cpus_weight(sd->groups->cpumask)-1) + * SCHED_LOAD_SCALE / 10; sd->groups->cpu_power = power; + } #endif + for_each_cpu_mask(i, *cpu_map) { + struct sched_domain *sd; +#ifdef CONFIG_SCHED_MC sd = &per_cpu(phys_domains, i); - power = SCHED_LOAD_SCALE + SCHED_LOAD_SCALE * - (cpus_weight(sd->groups->cpumask)-1) / 10; + if (i != first_cpu(sd->groups->cpumask)) + continue; + + sd->groups->cpu_power = 0; + if (sched_mc_power_savings || sched_smt_power_savings) { + int j; + + for_each_cpu_mask(j, sd->groups->cpumask) { + struct sched_domain *sd1; + sd1 = &per_cpu(core_domains, j); + /* + * for each core we will add once + * to the group in physical domain + */ + if (j != first_cpu(sd1->groups->cpumask)) + continue; + + if (sched_smt_power_savings) + sd->groups->cpu_power += sd1->groups->cpu_power; + else + sd->groups->cpu_power += SCHED_LOAD_SCALE; + } + } else + /* + * This has to be < 2 * SCHED_LOAD_SCALE + * Lets keep it SCHED_LOAD_SCALE, so that + * while calculating NUMA group's cpu_power + * we can simply do + * numa_group->cpu_power += phys_group->cpu_power; + * + * See "only add power once for each physical pkg" + * comment below + */ + sd->groups->cpu_power = SCHED_LOAD_SCALE; +#else + int power; + sd = &per_cpu(phys_domains, i); + if (sched_smt_power_savings) + power = SCHED_LOAD_SCALE * cpus_weight(sd->groups->cpumask); + else + power = SCHED_LOAD_SCALE; sd->groups->cpu_power = power; +#endif + } #ifdef CONFIG_NUMA - if (i == first_cpu(sd->groups->cpumask)) { - /* Only add "power" once for each physical package. */ - sd = &per_cpu(node_domains, i); - sd->groups->cpu_power += power; - } -#endif + for (i = 0; i < MAX_NUMNODES; i++) + init_numa_sched_groups_power(sched_group_nodes[i]); + + if (sched_group_allnodes) { + int group = cpu_to_allnodes_group(first_cpu(*cpu_map)); + struct sched_group *sg = &sched_group_allnodes[group]; + + init_numa_sched_groups_power(sg); } +#endif /* Attach the domains */ - for_each_online_cpu(i) { + for_each_cpu_mask(i, *cpu_map) { struct sched_domain *sd; #ifdef CONFIG_SCHED_SMT sd = &per_cpu(cpu_domains, i); +#elif defined(CONFIG_SCHED_MC) + sd = &per_cpu(core_domains, i); #else sd = &per_cpu(phys_domains, i); #endif cpu_attach_domain(sd, i); } + /* + * Tune cache-hot values: + */ + calibrate_migration_costs(cpu_map); + + return 0; + +error: + free_sched_groups(cpu_map); + return -ENOMEM; } +/* + * Set up scheduler domains and groups. Callers must hold the hotplug lock. + */ +static int arch_init_sched_domains(const cpumask_t *cpu_map) +{ + cpumask_t cpu_default_map; + int err; -#ifdef CONFIG_HOTPLUG_CPU -static void __devinit arch_destroy_sched_domains(void) + /* + * Setup mask for cpus without special case scheduling requirements. + * For now this just excludes isolated cpus, but could be used to + * exclude other special cases in the future. + */ + cpus_andnot(cpu_default_map, *cpu_map, cpu_isolated_map); + + err = build_sched_domains(&cpu_default_map); + + return err; +} + +static void arch_destroy_sched_domains(const cpumask_t *cpu_map) { - /* Do nothing: everything is statically allocated. */ + free_sched_groups(cpu_map); } -#endif -#endif /* ARCH_HAS_SCHED_DOMAIN */ +/* + * Detach sched domains from a group of cpus specified in cpu_map + * These cpus will now be attached to the NULL domain + */ +static void detach_destroy_domains(const cpumask_t *cpu_map) +{ + int i; + + for_each_cpu_mask(i, *cpu_map) + cpu_attach_domain(NULL, i); + synchronize_sched(); + arch_destroy_sched_domains(cpu_map); +} /* - * Initial dummy domain for early boot and for hotplug cpu. Being static, - * it is initialized to zero, so all balancing flags are cleared which is - * what we want. + * Partition sched domains as specified by the cpumasks below. + * This attaches all cpus from the cpumasks to the NULL domain, + * waits for a RCU quiescent period, recalculates sched + * domain information and then attaches them back to the + * correct sched domains + * Call with hotplug lock held */ -static struct sched_domain sched_domain_dummy; +int partition_sched_domains(cpumask_t *partition1, cpumask_t *partition2) +{ + cpumask_t change_map; + int err = 0; + + cpus_and(*partition1, *partition1, cpu_online_map); + cpus_and(*partition2, *partition2, cpu_online_map); + cpus_or(change_map, *partition1, *partition2); + + /* Detach sched domains from all of the affected cpus */ + detach_destroy_domains(&change_map); + if (!cpus_empty(*partition1)) + err = build_sched_domains(partition1); + if (!err && !cpus_empty(*partition2)) + err = build_sched_domains(partition2); + + return err; +} + +#if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT) +int arch_reinit_sched_domains(void) +{ + int err; + + lock_cpu_hotplug(); + detach_destroy_domains(&cpu_online_map); + err = arch_init_sched_domains(&cpu_online_map); + unlock_cpu_hotplug(); + + return err; +} + +static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt) +{ + int ret; + + if (buf[0] != '0' && buf[0] != '1') + return -EINVAL; + + if (smt) + sched_smt_power_savings = (buf[0] == '1'); + else + sched_mc_power_savings = (buf[0] == '1'); + + ret = arch_reinit_sched_domains(); + + return ret ? ret : count; +} + +int sched_create_sysfs_power_savings_entries(struct sysdev_class *cls) +{ + int err = 0; + +#ifdef CONFIG_SCHED_SMT + if (smt_capable()) + err = sysfs_create_file(&cls->kset.kobj, + &attr_sched_smt_power_savings.attr); +#endif +#ifdef CONFIG_SCHED_MC + if (!err && mc_capable()) + err = sysfs_create_file(&cls->kset.kobj, + &attr_sched_mc_power_savings.attr); +#endif + return err; +} +#endif + +#ifdef CONFIG_SCHED_MC +static ssize_t sched_mc_power_savings_show(struct sys_device *dev, char *page) +{ + return sprintf(page, "%u\n", sched_mc_power_savings); +} +static ssize_t sched_mc_power_savings_store(struct sys_device *dev, + const char *buf, size_t count) +{ + return sched_power_savings_store(buf, count, 0); +} +SYSDEV_ATTR(sched_mc_power_savings, 0644, sched_mc_power_savings_show, + sched_mc_power_savings_store); +#endif + +#ifdef CONFIG_SCHED_SMT +static ssize_t sched_smt_power_savings_show(struct sys_device *dev, char *page) +{ + return sprintf(page, "%u\n", sched_smt_power_savings); +} +static ssize_t sched_smt_power_savings_store(struct sys_device *dev, + const char *buf, size_t count) +{ + return sched_power_savings_store(buf, count, 1); +} +SYSDEV_ATTR(sched_smt_power_savings, 0644, sched_smt_power_savings_show, + sched_smt_power_savings_store); +#endif + #ifdef CONFIG_HOTPLUG_CPU /* * Force a reinitialization of the sched domains hierarchy. The domains * and groups cannot be updated in place without racing with the balancing - * code, so we temporarily attach all running cpus to a "dummy" domain + * code, so we temporarily attach all running cpus to the NULL domain * which will prevent rebalancing while the sched domains are recalculated. */ static int update_sched_domains(struct notifier_block *nfb, unsigned long action, void *hcpu) { - int i; - switch (action) { case CPU_UP_PREPARE: case CPU_DOWN_PREPARE: - for_each_online_cpu(i) - cpu_attach_domain(&sched_domain_dummy, i); - arch_destroy_sched_domains(); + detach_destroy_domains(&cpu_online_map); return NOTIFY_OK; case CPU_UP_CANCELED: @@ -5033,7 +6881,7 @@ static int update_sched_domains(struct notifier_block *nfb, } /* The hotplug lock is already held by cpu_up/cpu_down */ - arch_init_sched_domains(); + arch_init_sched_domains(&cpu_online_map); return NOTIFY_OK; } @@ -5042,7 +6890,7 @@ static int update_sched_domains(struct notifier_block *nfb, void __init sched_init_smp(void) { lock_cpu_hotplug(); - arch_init_sched_domains(); + arch_init_sched_domains(&cpu_online_map); unlock_cpu_hotplug(); /* XXX: Theoretical race here - CPU may be hotplugged now */ hotcpu_notifier(update_sched_domains, 0); @@ -5057,6 +6905,7 @@ int in_sched_functions(unsigned long addr) { /* Linker adds these: start and end of __sched functions */ extern char __sched_text_start[], __sched_text_end[]; + return in_lock_functions(addr) || (addr >= (unsigned long)__sched_text_start && addr < (unsigned long)__sched_text_end); @@ -5064,23 +6913,27 @@ int in_sched_functions(unsigned long addr) void __init sched_init(void) { - runqueue_t *rq; int i, j, k; - for (i = 0; i < NR_CPUS; i++) { - prio_array_t *array; + for_each_possible_cpu(i) { + struct prio_array *array; + struct rq *rq; rq = cpu_rq(i); spin_lock_init(&rq->lock); + lockdep_set_class(&rq->lock, &rq->rq_lock_key); + rq->nr_running = 0; rq->active = rq->arrays; rq->expired = rq->arrays + 1; rq->best_expired_prio = MAX_PRIO; #ifdef CONFIG_SMP - rq->sd = &sched_domain_dummy; - rq->cpu_load = 0; + rq->sd = NULL; + for (j = 1; j < 3; j++) + rq->cpu_load[j] = 0; rq->active_balance = 0; rq->push_cpu = 0; + rq->cpu = i; rq->migration_thread = NULL; INIT_LIST_HEAD(&rq->migration_queue); #endif @@ -5100,6 +6953,12 @@ void __init sched_init(void) } } + set_load_weight(&init_task); + +#ifdef CONFIG_RT_MUTEXES + plist_head_init(&init_task.pi_waiters, &init_task.pi_lock); +#endif + /* * The boot idle thread does lazy MMU switching as well: */ @@ -5118,7 +6977,7 @@ void __init sched_init(void) #ifdef CONFIG_DEBUG_SPINLOCK_SLEEP void __might_sleep(char *file, int line) { -#if defined(in_atomic) +#ifdef in_atomic static unsigned long prev_jiffy; /* ratelimiting */ if ((in_atomic() || irqs_disabled()) && @@ -5126,7 +6985,7 @@ void __might_sleep(char *file, int line) if (time_before(jiffies, prev_jiffy + HZ) && prev_jiffy) return; prev_jiffy = jiffies; - printk(KERN_ERR "Debug: sleeping function called from invalid" + printk(KERN_ERR "BUG: sleeping function called from invalid" " context at %s:%d\n", file, line); printk("in_atomic():%d, irqs_disabled():%d\n", in_atomic(), irqs_disabled()); @@ -5140,17 +6999,18 @@ EXPORT_SYMBOL(__might_sleep); #ifdef CONFIG_MAGIC_SYSRQ void normalize_rt_tasks(void) { + struct prio_array *array; struct task_struct *p; - prio_array_t *array; unsigned long flags; - runqueue_t *rq; + struct rq *rq; read_lock_irq(&tasklist_lock); - for_each_process (p) { + for_each_process(p) { if (!rt_task(p)) continue; - rq = task_rq_lock(p, &flags); + spin_lock_irqsave(&p->pi_lock, flags); + rq = __task_rq_lock(p); array = p->array; if (array) @@ -5162,9 +7022,54 @@ void normalize_rt_tasks(void) resched_task(rq->curr); } - task_rq_unlock(rq, &flags); + __task_rq_unlock(rq); + spin_unlock_irqrestore(&p->pi_lock, flags); } read_unlock_irq(&tasklist_lock); } #endif /* CONFIG_MAGIC_SYSRQ */ + +#ifdef CONFIG_IA64 +/* + * These functions are only useful for the IA64 MCA handling. + * + * They can only be called when the whole system has been + * stopped - every CPU needs to be quiescent, and no scheduling + * activity can take place. Using them for anything else would + * be a serious bug, and as a result, they aren't even visible + * under any other configuration. + */ + +/** + * curr_task - return the current task for a given cpu. + * @cpu: the processor in question. + * + * ONLY VALID WHEN THE WHOLE SYSTEM IS STOPPED! + */ +struct task_struct *curr_task(int cpu) +{ + return cpu_curr(cpu); +} + +/** + * set_curr_task - set the current task for a given cpu. + * @cpu: the processor in question. + * @p: the task pointer to set. + * + * Description: This function must only be used when non-maskable interrupts + * are serviced on a separate stack. It allows the architecture to switch the + * notion of the current task on a cpu in a non-blocking manner. This function + * must be called with all CPU's synchronized, and interrupts disabled, the + * and caller must save the original value of the current task (see + * curr_task() above) and restore that value before reenabling interrupts and + * re-starting the system. + * + * ONLY VALID WHEN THE WHOLE SYSTEM IS STOPPED! + */ +void set_curr_task(int cpu, struct task_struct *p) +{ + cpu_curr(cpu) = p; +} + +#endif