X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=kernel%2Fsched.c;h=e43f1a76dd437aee45bae1e1c1f4d28628a078a2;hb=16c70f8c1b54b61c3b951b6fb220df250fe09b32;hp=022dabf9621e79fb06067fe3614b48f663b8bcf4;hpb=9213980e6a70d8473e0ffd4b39ab5b6caaba9ff5;p=linux-2.6.git diff --git a/kernel/sched.c b/kernel/sched.c index 022dabf96..e43f1a76d 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -27,29 +27,37 @@ #include #include #include +#include #include #include +#include #include #include +#include #include +#include #include #include #include +#include #include #include #include +#include #include #include -#include -#include +#include +#include +#include +#include +#include +#include +#include #include - -#ifdef CONFIG_NUMA -#define cpu_to_node_mask(cpu) node_to_cpumask(cpu_to_node(cpu)) -#else -#define cpu_to_node_mask(cpu) (cpu_online_map) -#endif +#include +#include +#include /* * Convert user-nice values [ -20 ... 0 ... 19 ] @@ -68,8 +76,6 @@ #define USER_PRIO(p) ((p)-MAX_RT_PRIO) #define TASK_USER_PRIO(p) USER_PRIO((p)->static_prio) #define MAX_USER_PRIO (USER_PRIO(MAX_PRIO)) -#define AVG_TIMESLICE (MIN_TIMESLICE + ((MAX_TIMESLICE - MIN_TIMESLICE) *\ - (MAX_PRIO-1-NICE_TO_PRIO(0))/(MAX_USER_PRIO - 1))) /* * Some helpers for converting nanosecond timing to jiffy resolution @@ -80,12 +86,12 @@ /* * These are the 'tuning knobs' of the scheduler: * - * Minimum timeslice is 10 msecs, default timeslice is 100 msecs, - * maximum timeslice is 200 msecs. Timeslices get refilled after - * they expire. + * Minimum timeslice is 5 msecs (or 1 jiffy, whichever is larger), + * default timeslice is 100 msecs, maximum timeslice is 800 msecs. + * Timeslices get refilled after they expire. */ -#define MIN_TIMESLICE ( 10 * HZ / 1000) -#define MAX_TIMESLICE (200 * HZ / 1000) +#define MIN_TIMESLICE max(5 * HZ / 1000, 1) +#define DEF_TIMESLICE (100 * HZ / 1000) #define ON_RUNQUEUE_WEIGHT 30 #define CHILD_PENALTY 95 #define PARENT_PENALTY 100 @@ -93,10 +99,9 @@ #define PRIO_BONUS_RATIO 25 #define MAX_BONUS (MAX_USER_PRIO * PRIO_BONUS_RATIO / 100) #define INTERACTIVE_DELTA 2 -#define MAX_SLEEP_AVG (AVG_TIMESLICE * MAX_BONUS) +#define MAX_SLEEP_AVG (DEF_TIMESLICE * MAX_BONUS) #define STARVATION_LIMIT (MAX_SLEEP_AVG) #define NS_MAX_SLEEP_AVG (JIFFIES_TO_NS(MAX_SLEEP_AVG)) -#define CREDIT_LIMIT 100 /* * If a task is 'interactive' then we reinsert it in the active @@ -130,12 +135,14 @@ (NS_TO_JIFFIES((p)->sleep_avg) * MAX_BONUS / \ MAX_SLEEP_AVG) +#define GRANULARITY (10 * HZ / 1000 ? : 1) + #ifdef CONFIG_SMP -#define TIMESLICE_GRANULARITY(p) (MIN_TIMESLICE * \ +#define TIMESLICE_GRANULARITY(p) (GRANULARITY * \ (1 << (((MAX_BONUS - CURRENT_BONUS(p)) ? : 1) - 1)) * \ num_online_cpus()) #else -#define TIMESLICE_GRANULARITY(p) (MIN_TIMESLICE * \ +#define TIMESLICE_GRANULARITY(p) (GRANULARITY * \ (1 << (((MAX_BONUS - CURRENT_BONUS(p)) ? : 1) - 1))) #endif @@ -143,7 +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)) @@ -152,48 +160,41 @@ (JIFFIES_TO_NS(MAX_SLEEP_AVG * \ (MAX_BONUS / 2 + DELTA((p)) + 1) / MAX_BONUS - 1)) -#define HIGH_CREDIT(p) \ - ((p)->interactive_credit > CREDIT_LIMIT) - -#define LOW_CREDIT(p) \ - ((p)->interactive_credit < -CREDIT_LIMIT) - #define TASK_PREEMPTS_CURR(p, rq) \ ((p)->prio < (rq)->curr->prio) /* - * BASE_TIMESLICE scales user-nice values [ -20 ... 19 ] - * to time slice values. + * task_timeslice() scales user-nice values [ -20 ... 0 ... 19 ] + * to time slice values: [800ms ... 100ms ... 5ms] * * The higher a thread's priority, the bigger timeslices * it gets during one round of execution. But even the lowest * priority thread gets MIN_TIMESLICE worth of execution time. - * - * task_timeslice() is the interface that is used by the scheduler. */ -#define BASE_TIMESLICE(p) (MIN_TIMESLICE + \ - ((MAX_TIMESLICE - MIN_TIMESLICE) * \ - (MAX_PRIO-1 - (p)->static_prio) / (MAX_USER_PRIO-1))) +#define SCALE_PRIO(x, prio) \ + max(x * (MAX_PRIO - prio) / (MAX_USER_PRIO / 2), MIN_TIMESLICE) -static unsigned int task_timeslice(task_t *p) +static unsigned int static_prio_timeslice(int static_prio) { - return BASE_TIMESLICE(p); + if (static_prio < NICE_TO_PRIO(0)) + return SCALE_PRIO(DEF_TIMESLICE * 4, static_prio); + else + return SCALE_PRIO(DEF_TIMESLICE, static_prio); } -#define task_hot(p, now, sd) ((now) - (p)->timestamp < (sd)->cache_hot_time) +static inline unsigned int task_timeslice(struct task_struct *p) +{ + return static_prio_timeslice(p->static_prio); +} /* * 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]; }; @@ -204,7 +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; /* @@ -212,15 +213,25 @@ 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; - unsigned long expired_timestamp, nr_uninterruptible; + + /* + * This is part of a global counter where only the total sum + * over all CPUs matters. A task can increase this counter on + * one CPU and if it got migrated afterwards it may decrease + * it on another CPU. Always updated under the runqueue lock: + */ + unsigned long nr_uninterruptible; + + 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; @@ -230,41 +241,169 @@ 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 struct list_head hold_queue; int idle_tokens; +#endif + +#ifdef CONFIG_SCHEDSTATS + /* latency stats */ + struct sched_info rq_sched_info; + + /* sys_sched_yield() stats */ + unsigned long yld_exp_empty; + unsigned long yld_act_empty; + unsigned long yld_both_empty; + unsigned long yld_cnt; + + /* schedule() stats */ + unsigned long sched_switch; + unsigned long sched_cnt; + unsigned long sched_goidle; + + /* try_to_wake_up() stats */ + 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 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); @@ -277,17 +416,150 @@ 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); } +#ifdef CONFIG_SCHEDSTATS +/* + * 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 12 + +static int show_schedstat(struct seq_file *seq, void *v) +{ + int cpu; + + seq_printf(seq, "version %d\n", SCHEDSTAT_VERSION); + seq_printf(seq, "timestamp %lu\n", jiffies); + for_each_online_cpu(cpu) { + struct rq *rq = cpu_rq(cpu); +#ifdef CONFIG_SMP + struct sched_domain *sd; + int dcnt = 0; +#endif + + /* runqueue-specific stats */ + seq_printf(seq, + "cpu%d %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu %lu", + cpu, rq->yld_both_empty, + rq->yld_act_empty, rq->yld_exp_empty, rq->yld_cnt, + rq->sched_switch, rq->sched_cnt, rq->sched_goidle, + rq->ttwu_cnt, rq->ttwu_local, + rq->rq_sched_info.cpu_time, + rq->rq_sched_info.run_delay, rq->rq_sched_info.pcnt); + + seq_printf(seq, "\n"); + +#ifdef CONFIG_SMP + /* domain-specific stats */ + preempt_disable(); + for_each_domain(cpu, sd) { + enum idle_type itype; + char mask_str[NR_CPUS]; + + cpumask_scnprintf(mask_str, NR_CPUS, sd->span); + seq_printf(seq, "domain%d %s", dcnt++, mask_str); + for (itype = SCHED_IDLE; itype < MAX_IDLE_TYPES; + itype++) { + seq_printf(seq, " %lu %lu %lu %lu %lu %lu %lu %lu", + sd->lb_cnt[itype], + sd->lb_balanced[itype], + sd->lb_failed[itype], + sd->lb_imbalance[itype], + sd->lb_gained[itype], + sd->lb_hot_gained[itype], + sd->lb_nobusyq[itype], + sd->lb_nobusyg[itype]); + } + 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_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; +} + +static int schedstat_open(struct inode *inode, struct file *file) +{ + unsigned int size = PAGE_SIZE * (1 + num_online_cpus() / 32); + char *buf = kmalloc(size, GFP_KERNEL); + struct seq_file *m; + int res; + + if (!buf) + return -ENOMEM; + res = single_open(file, show_schedstat, NULL); + if (!res) { + m = file->private_data; + m->buf = buf; + m->size = size; + } else + kfree(buf); + return res; +} + +struct file_operations proc_schedstat_operations = { + .open = schedstat_open, + .read = seq_read, + .llseek = seq_lseek, + .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 + /* * rq_lock - lock a given runqueue and disable interrupts. */ -static 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(); @@ -296,24 +568,128 @@ static runqueue_t *this_rq_lock(void) return rq; } -static inline void rq_unlock(runqueue_t *rq) +#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 + * tasks, the expired queue will become the active queue after the active + * queue is empty, without explicitly dequeuing and requeuing tasks in the + * expired queue. (Interactive tasks may be requeued directly to the + * active queue, thus delaying tasks in the expired queue from running; + * see scheduler_tick()). + * + * This function is only called from sched_info_arrive(), rather than + * dequeue_task(). Even though a task may be queued and dequeued multiple + * times as it is shuffled about, we're really interested in knowing how + * 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(struct task_struct *t) +{ + t->sched_info.last_queued = 0; +} + +/* + * Called when a task finally hits the cpu. We can now calculate how + * 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 void sched_info_arrive(struct task_struct *t) { - spin_unlock_irq(&rq->lock); + unsigned long now = jiffies, delta_jiffies = 0; + + if (t->sched_info.last_queued) + delta_jiffies = now - t->sched_info.last_queued; + sched_info_dequeued(t); + t->sched_info.run_delay += delta_jiffies; + t->sched_info.last_arrival = now; + t->sched_info.pcnt++; + + rq_sched_info_arrive(task_rq(t), delta_jiffies); +} + +/* + * Called when a process is queued into either the active or expired + * array. The time is noted and later used to determine how long we + * had to wait for us to reach the cpu. Since the expired queue will + * become the active queue after active queue is empty, without dequeuing + * and requeuing any tasks, we are interested in queuing to either. It + * is unusual but not impossible for tasks to be dequeued and immediately + * requeued in the same or another array: this can happen in sched_yield(), + * set_user_nice(), and even load_balance() as it moves tasks from runqueue + * to runqueue. + * + * This function is only called from enqueue_task(), but also only updates + * 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(struct task_struct *t) +{ + 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(struct task_struct *t) +{ + unsigned long delta_jiffies = jiffies - t->sched_info.last_arrival; + + t->sched_info.cpu_time += delta_jiffies; + rq_sched_info_depart(task_rq(t), delta_jiffies); +} + +/* + * Called when tasks are switched involuntarily due, typically, to expiring + * 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(struct task_struct *prev, struct task_struct *next) +{ + struct rq *rq = task_rq(prev); + + /* + * prev now departs the cpu. It's not interesting to record + * stats about how efficient we were at scheduling the idle + * process, however. + */ + if (prev != rq->idle) + sched_info_depart(prev); + + 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 || 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--; list_del(&p->run_list); if (list_empty(array->queue + p->prio)) __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); list_add_tail(&p->run_list, array->queue + p->prio); __set_bit(p->prio, array->bitmap); array->nr_active++; @@ -321,12 +697,19 @@ static void enqueue_task(struct task_struct *p, prio_array_t *array) } /* - * Used by the migration code - we pull tasks from the head of the - * remote queue so we want these tasks to show up at the head of the - * local queue: + * Put task to the end of the run list without the overhead of dequeue + * followed by enqueue. */ -static inline void enqueue_task_head(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, struct prio_array *array) +{ + BUG_ON(p->state & TASK_ONHOLD); list_add(&p->run_list, array->queue + p->prio); __set_bit(p->prio, array->bitmap); array->nr_active++; @@ -334,7 +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] @@ -347,18 +730,19 @@ 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; - - if (rt_task(p)) - return p->prio; + struct vx_info *vxi; bonus = CURRENT_BONUS(p) - MAX_BONUS / 2; prio = p->static_prio - bonus; - if (__vx_task_flags(p, VXF_SCHED_PRIO, 0)) - prio += effective_vavavoom(p, MAX_USER_PRIO); + + if ((vxi = p->vx_info) && + vx_info_flags(vxi, VXF_SCHED_PRIO, 0)) + prio += vx_effective_vavavoom(vxi, MAX_USER_PRIO); if (prio < MAX_RT_PRIO) prio = MAX_RT_PRIO; @@ -368,73 +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. */ -static inline void __activate_task(task_t *p, runqueue_t *rq) + +/* + * 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 + */ +#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) { - unsigned long long __sleep_time = now - p->timestamp; - unsigned long sleep_time; + /* Caller must always ensure 'now >= p->timestamp' */ + 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 - - AVG_TIMESLICE); - if (!HIGH_CREDIT(p)) - p->interactive_credit++; - } 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; /* - * Tasks with low interactive_credit are limited to - * one timeslice worth of sleep avg bonus. + * 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. */ - if (LOW_CREDIT(p) && - sleep_time > JIFFIES_TO_NS(task_timeslice(p))) - sleep_time = JIFFIES_TO_NS(task_timeslice(p)); - + p->sleep_type = SLEEP_NONINTERACTIVE; + } else { /* - * Non high_credit tasks waking from uninterruptible - * sleep are limited in their sleep_avg rise as they - * are likely to be cpu hogs waiting on I/O + * 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 && !HIGH_CREDIT(p) && 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; } } @@ -448,15 +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 (!HIGH_CREDIT(p)) - p->interactive_credit++; - } } + if (p->sleep_avg > NS_MAX_SLEEP_AVG) + p->sleep_avg = NS_MAX_SLEEP_AVG; } - p->prio = effective_prio(p); + return effective_prio(p); } /* @@ -465,7 +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; @@ -473,19 +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 @@ -494,58 +981,126 @@ 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; + vx_activate_task(p); __activate_task(p, rq); } /* * 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--; - if (p->state == TASK_UNINTERRUPTIBLE) - rq->nr_uninterruptible++; + dec_nr_running(p, rq); dequeue_task(p, p->array); p->array = NULL; } +static inline +void deactivate_task(struct task_struct *p, struct rq *rq) +{ + vx_deactivate_task(p); + __deactivate_task(p, rq); +} + + +#ifdef CONFIG_VSERVER_HARDCPU /* - * resched_task - mark a task 'to be rescheduled now'. - * - * On UP this means the setting of the need_resched flag, on SMP it - * might also involve a cross-CPU call to trigger the scheduler on - * the target CPU. + * vx_hold_task - put a task on the hold queue */ -#ifdef CONFIG_SMP -static void resched_task(task_t *p) +static inline +void vx_hold_task(struct vx_info *vxi, + struct task_struct *p, struct rq *rq) { - int need_resched, nrpolling; - - preempt_disable(); - /* 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); + __deactivate_task(p, rq); + p->state |= TASK_ONHOLD; + /* a new one on hold */ + vx_onhold_inc(vxi); + list_add_tail(&p->run_list, &rq->hold_queue); +} - if (!need_resched && !nrpolling && (task_cpu(p) != smp_processor_id())) - smp_send_reschedule(task_cpu(p)); - preempt_enable(); +/* + * vx_unhold_task - put a task back to the runqueue + */ +static inline +void vx_unhold_task(struct vx_info *vxi, + 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); + inc_nr_running(p, rq); + + if (p->static_prio < rq->best_expired_prio) + rq->best_expired_prio = p->static_prio; } #else -static inline void resched_task(task_t *p) +static inline +void vx_hold_task(struct vx_info *vxi, + struct task_struct *p, struct rq *rq) { - set_tsk_need_resched(p); + return; +} + +static inline +void vx_unhold_task(struct vx_info *vxi, + struct task_struct *p, struct rq *rq) +{ + return; +} +#endif /* CONFIG_VSERVER_HARDCPU */ + + +/* + * resched_task - mark a task 'to be rescheduled now'. + * + * On UP this means the setting of the need_resched flag, on SMP it + * might also involve a cross-CPU call to trigger the scheduler on + * the target CPU. + */ +#ifdef CONFIG_SMP + +#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 cpu; + + assert_spin_locked(&task_rq(p)->lock); + + if (unlikely(test_tsk_thread_flag(p, TIF_NEED_RESCHED))) + return; + + set_tsk_thread_flag(p, TIF_NEED_RESCHED); + + 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(struct task_struct *p) +{ + assert_spin_locked(&task_rq(p)->lock); + set_tsk_need_resched(p); } #endif @@ -553,38 +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(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 @@ -596,10 +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; } @@ -612,16 +1164,16 @@ 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: rq = task_rq_lock(p, &flags); /* Must be off runqueue entirely, not preempted. */ - if (unlikely(p->array)) { + if (unlikely(p->array || task_running(rq, p))) { /* If it's preempted, we yield. It could be a while. */ preempted = !task_running(rq, p); task_rq_unlock(rq, &flags); @@ -639,8 +1191,14 @@ repeat: * * Cause a process which is running on another CPU to enter * kernel-mode, without any delay. (to get signals handled.) + * + * NOTE: this function doesnt have to take the runqueue lock, + * because all it wants to ensure is that the remote task enters + * the kernel. If the IPI races and the task has been migrated + * 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; @@ -651,70 +1209,222 @@ void kick_process(task_t *p) preempt_enable(); } -EXPORT_SYMBOL_GPL(kick_process); - /* - * 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) { - runqueue_t *rq = cpu_rq(cpu); - unsigned long load_now = rq->nr_running * SCHED_LOAD_SCALE; + struct rq *rq = cpu_rq(cpu); - return min(rq->cpu_load, load_now); + 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 + * 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) +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 max(rq->cpu_load, load_now); + return max(rq->cpu_load[type-1], rq->raw_weighted_load); } -#endif +/* + * Return the average load per task on the cpu's run queue + */ +static inline unsigned long cpu_avg_load_per_task(int cpu) +{ + struct rq *rq = cpu_rq(cpu); + unsigned long n = rq->nr_running; + + return n ? rq->raw_weighted_load / n : SCHED_LOAD_SCALE; +} + +/* + * 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; +} + +/* + * find_idlest_queue - find the idlest runqueue among the cpus in group. + */ +static int +find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu) +{ + cpumask_t tmp; + unsigned long load, min_load = ULONG_MAX; + int idlest = -1; + int i; + + /* Traverse only the allowed CPUs */ + cpus_and(tmp, group->cpumask, p->cpus_allowed); + + 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; + } + } + + return idlest; +} + +/* + * 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() is useful especially on SMT architectures to wake a - * task onto an idle sibling if we would otherwise wake it onto a - * busy sibling. + * 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, task_t *p) +static int wake_idle(int cpu, struct task_struct *p) { cpumask_t tmp; - runqueue_t *rq = cpu_rq(cpu); struct sched_domain *sd; int i; if (idle_cpu(cpu)) return cpu; - sd = rq->sd; - if (!(sd->flags & SD_WAKE_IDLE)) - return cpu; - - cpus_and(tmp, sd->span, cpu_online_map); - for_each_cpu_mask(i, tmp) { - if (!cpu_isset(i, p->cpus_allowed)) - continue; - - if (idle_cpu(i)) - return i; + 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, task_t *p) +static inline int wake_idle(int cpu, struct task_struct *p) { return cpu; } @@ -734,20 +1444,26 @@ 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 rq = task_rq_lock(p, &flags); old_state = p->state; + + /* we need to unhold suspended tasks */ + if (old_state & TASK_ONHOLD) { + vx_unhold_task(p->vx_info, p, rq); + old_state = p->state; + } if (!(old_state & state)) goto out; @@ -763,54 +1479,78 @@ static int try_to_wake_up(task_t * p, unsigned int state, int sync) new_cpu = cpu; - if (cpu == this_cpu || unlikely(!cpu_isset(this_cpu, p->cpus_allowed))) + schedstat_inc(rq, ttwu_cnt); + if (cpu == this_cpu) { + schedstat_inc(rq, ttwu_local); 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: - */ - if (sync) - this_load -= SCHED_LOAD_SCALE; + for_each_domain(this_cpu, sd) { + if (cpu_isset(cpu, sd->span)) { + schedstat_inc(sd, ttwu_wake_remote); + this_sd = sd; + break; + } + } - /* 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) + if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed))) goto out_set_cpu; - new_cpu = this_cpu; /* Wake to this CPU if we can */ - /* - * Scan domains for affine wakeup and passive balancing - * possibilities. + * Check for affine wakeup and passive balancing possibilities. */ - for_each_domain(this_cpu, sd) { + if (this_sd) { + int idx = this_sd->wake_idx; unsigned int imbalance; + + imbalance = 100 + (this_sd->imbalance_pct - 100) / 2; + + load = source_load(cpu, idx); + this_load = target_load(this_cpu, idx); + + 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 sync wakeup then subtract the (maximum possible) + * effect of the currently running task from the load + * of the current CPU: + */ + 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; + } + } + /* * Start passive balancing when half the imbalance_pct * limit is reached. */ - imbalance = sd->imbalance_pct + (sd->imbalance_pct - 100) / 2; - - if ( ((sd->flags & SD_WAKE_AFFINE) && - !task_hot(p, rq->timestamp_last_tick, sd)) - || ((sd->flags & SD_WAKE_BALANCE) && - imbalance*this_load <= 100*load) ) { - /* - * Now sd has SD_WAKE_AFFINE and p is cache cold in sd - * or sd has SD_WAKE_BALANCE and there is an imbalance - */ - if (cpu_isset(cpu, sd->span)) + if (this_sd->flags & SD_WAKE_BALANCE) { + if (imbalance*this_load <= 100*load) { + schedstat_inc(this_sd, ttwu_move_balance); goto out_set_cpu; + } } } new_cpu = cpu; /* Could not wake to this_cpu. Wake to cpu instead */ out_set_cpu: new_cpu = wake_idle(new_cpu, p); - if (new_cpu != cpu && cpu_isset(new_cpu, p->cpus_allowed)) { + if (new_cpu != cpu) { set_task_cpu(p, new_cpu); task_rq_unlock(rq, &flags); /* might preempt at this point */ @@ -829,13 +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) @@ -844,7 +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); if (!sync || cpu != this_cpu) { if (TASK_PREEMPTS_CURR(p, rq)) resched_task(rq->curr); @@ -859,15 +1609,14 @@ 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_INTERRUPTIBLE | TASK_UNINTERRUPTIBLE, 0); + 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); } @@ -876,8 +1625,15 @@ int fastcall wake_up_state(task_t *p, unsigned int state) * 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 @@ -885,17 +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); +#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 @@ -911,60 +1674,101 @@ void fastcall sched_fork(task_t *p) p->first_time_slice = 1; current->time_slice >>= 1; p->timestamp = sched_clock(); - if (!current->time_slice) { + if (unlikely(!current->time_slice)) { /* * This case is rare, it happens when the parent has only * a single jiffy left from its timeslice. Taking the * runqueue lock is not a problem. */ current->time_slice = 1; - preempt_disable(); - scheduler_tick(0, 0); - local_irq_enable(); - preempt_enable(); - } else - local_irq_enable(); + scheduler_tick(); + } + local_irq_enable(); + put_cpu(); } /* - * wake_up_forked_process - wake up a freshly forked process. + * wake_up_new_task - wake up a newly created task for the first time. * * This function will do some initial scheduler statistics housekeeping - * that must be done for every newly created process. + * that must be done for every newly created context, then puts the task + * on the runqueue and wakes it. */ -void fastcall wake_up_forked_process(task_t * p) +void fastcall wake_up_new_task(struct task_struct *p, unsigned long clone_flags) { + struct rq *rq, *this_rq; unsigned long flags; - runqueue_t *rq = task_rq_lock(current, &flags); + int this_cpu, cpu; + rq = task_rq_lock(p, &flags); BUG_ON(p->state != TASK_RUNNING); + this_cpu = smp_processor_id(); + cpu = task_cpu(p); /* * We decrease the sleep average of forking parents * and children as well, to keep max-interactive tasks - * from forking tasks that are max-interactive. + * from forking tasks that are max-interactive. The parent + * (current) is done further down, under its lock. */ - current->sleep_avg = JIFFIES_TO_NS(CURRENT_BONUS(current) * - PARENT_PENALTY / 100 * MAX_SLEEP_AVG / MAX_BONUS); - p->sleep_avg = JIFFIES_TO_NS(CURRENT_BONUS(p) * CHILD_PENALTY / 100 * MAX_SLEEP_AVG / MAX_BONUS); - p->interactive_credit = 0; - p->prio = effective_prio(p); - set_task_cpu(p, smp_processor_id()); - if (unlikely(!current->array)) + vx_activate_task(p); + if (likely(cpu == this_cpu)) { + if (!(clone_flags & CLONE_VM)) { + /* + * The VM isn't cloned, so we're in a good position to + * do child-runs-first in anticipation of an exec. This + * usually avoids a lot of COW overhead. + */ + if (unlikely(!current->array)) + __activate_task(p, rq); + 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++; + inc_nr_running(p, rq); + } + set_need_resched(); + } else + /* Run child last */ + __activate_task(p, rq); + /* + * We skip the following code due to cpu == this_cpu + * + * task_rq_unlock(rq, &flags); + * this_rq = task_rq_lock(current, &flags); + */ + this_rq = rq; + } else { + this_rq = cpu_rq(this_cpu); + + /* + * Not the local CPU - must adjust timestamp. This should + * get optimised away in the !CONFIG_SMP case. + */ + p->timestamp = (p->timestamp - this_rq->timestamp_last_tick) + + rq->timestamp_last_tick; __activate_task(p, rq); - else { - p->prio = current->prio; - list_add_tail(&p->run_list, ¤t->run_list); - p->array = current->array; - p->array->nr_active++; - rq->nr_running++; + if (TASK_PREEMPTS_CURR(p, rq)) + resched_task(rq->curr); + + /* + * Parent and child are on different CPUs, now get the + * parent runqueue to update the parent's ->sleep_avg: + */ + task_rq_unlock(rq, &flags); + this_rq = task_rq_lock(current, &flags); } - task_rq_unlock(rq, &flags); + current->sleep_avg = JIFFIES_TO_NS(CURRENT_BONUS(current) * + PARENT_PENALTY / 100 * MAX_SLEEP_AVG / MAX_BONUS); + task_rq_unlock(this_rq, &flags); } /* @@ -976,23 +1780,21 @@ void fastcall wake_up_forked_process(task_t * p) * 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; - local_irq_save(flags); - if (p->first_time_slice) { - p->parent->time_slice += p->time_slice; - if (unlikely(p->parent->time_slice > MAX_TIMESLICE)) - p->parent->time_slice = MAX_TIMESLICE; - } - local_irq_restore(flags); /* * 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 && 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); + } if (p->sleep_avg < p->parent->sleep_avg) p->parent->sleep_avg = p->parent->sleep_avg / (EXIT_WEIGHT + 1) * EXIT_WEIGHT + p->sleep_avg / @@ -1000,22 +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 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; @@ -1023,31 +1845,44 @@ static void finish_task_switch(task_t *prev) /* * A task struct has one reference for the use as "current". - * If a task dies, then it sets TASK_ZOMBIE in tsk->state and calls - * schedule one last time. The schedule call will never return, + * If a task dies, then it sets EXIT_ZOMBIE in tsk->exit_state and + * calls schedule one last time. The schedule call will never return, * and the scheduled task must drop that reference. - * The test for TASK_ZOMBIE must occur while the runqueue locks are + * The test for EXIT_ZOMBIE must occur while the runqueue locks are * still held, otherwise prev could be scheduled on another cpu, die * there before we look at prev->state, and then the reference would * be dropped twice. * 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); } @@ -1056,8 +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; @@ -1074,6 +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); @@ -1092,7 +1937,7 @@ unsigned long nr_running(void) { unsigned long i, sum = 0; - for_each_cpu(i) + for_each_online_cpu(i) sum += cpu_rq(i)->nr_running; return sum; @@ -1102,17 +1947,25 @@ unsigned long nr_uninterruptible(void) { unsigned long i, sum = 0; - for_each_online_cpu(i) + for_each_possible_cpu(i) sum += cpu_rq(i)->nr_uninterruptible; + /* + * Since we read the counters lockless, it might be slightly + * inaccurate. Do not allow it to go below zero though: + */ + if (unlikely((long)sum < 0)) + sum = 0; + return sum; } unsigned long long nr_context_switches(void) { - unsigned long long i, sum = 0; + int i; + unsigned long long sum = 0; - for_each_online_cpu(i) + for_each_possible_cpu(i) sum += cpu_rq(i)->nr_switches; return sum; @@ -1122,23 +1975,52 @@ unsigned long nr_iowait(void) { unsigned long i, sum = 0; - for_each_online_cpu(i) + for_each_possible_cpu(i) sum += atomic_read(&cpu_rq(i)->nr_iowait); return sum; } -/* - * 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) +unsigned long nr_active(void) { - if (rq1 == rq2) + 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(struct rq *rq1, struct rq *rq2) + __acquires(rq1->lock) + __acquires(rq2->lock) +{ + if (rq1 == rq2) { spin_lock(&rq1->lock); - else { + __acquire(rq2->lock); /* Fake it out ;) */ + } else { if (rq1 < rq2) { spin_lock(&rq1->lock); spin_lock(&rq2->lock); @@ -1155,149 +2037,33 @@ 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) { spin_unlock(&rq1->lock); if (rq1 != rq2) spin_unlock(&rq2->lock); -} - -enum idle_type -{ - IDLE, - NOT_IDLE, - NEWLY_IDLE, -}; - -#ifdef CONFIG_SMP - -/* - * 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, cpu_online_map); - cpus_and(mask, mask, 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; + else + __release(rq2->lock); } /* - * wake_up_forked_thread - wake up a freshly forked thread. - * - * This function will do some initial scheduler statistics housekeeping - * that must be done for every newly created context, and it also does - * runqueue balancing. + * double_lock_balance - lock the busiest runqueue, this_rq is locked already. */ -void fastcall wake_up_forked_thread(task_t * p) +static void double_lock_balance(struct rq *this_rq, struct rq *busiest) + __releases(this_rq->lock) + __acquires(busiest->lock) + __acquires(this_rq->lock) { - unsigned long flags; - int this_cpu = get_cpu(), cpu; - struct sched_domain *tmp, *sd = NULL; - runqueue_t *this_rq = cpu_rq(this_cpu), *rq; - - /* - * Find the largest domain that this CPU is part of that - * is willing to balance on clone: - */ - for_each_domain(this_cpu, tmp) - if (tmp->flags & SD_BALANCE_CLONE) - sd = tmp; - if (sd) - cpu = find_idlest_cpu(p, this_cpu, sd); - else - cpu = this_cpu; - - local_irq_save(flags); -lock_again: - rq = cpu_rq(cpu); - double_rq_lock(this_rq, rq); - - BUG_ON(p->state != TASK_RUNNING); - - /* - * We did find_idlest_cpu() unlocked, so in theory - * the mask could have changed - just dont migrate - * in this case: - */ - if (unlikely(!cpu_isset(cpu, p->cpus_allowed))) { - cpu = this_cpu; - double_rq_unlock(this_rq, rq); - goto lock_again; - } - /* - * We decrease the sleep average of forking parents - * and children as well, to keep max-interactive tasks - * from forking tasks that are max-interactive. - */ - current->sleep_avg = JIFFIES_TO_NS(CURRENT_BONUS(current) * - PARENT_PENALTY / 100 * MAX_SLEEP_AVG / MAX_BONUS); - - p->sleep_avg = JIFFIES_TO_NS(CURRENT_BONUS(p) * - CHILD_PENALTY / 100 * MAX_SLEEP_AVG / MAX_BONUS); - - p->interactive_credit = 0; - - p->prio = effective_prio(p); - set_task_cpu(p, cpu); - - if (cpu == this_cpu) { - if (unlikely(!current->array)) - __activate_task(p, rq); - else { - p->prio = current->prio; - list_add_tail(&p->run_list, ¤t->run_list); - p->array = current->array; - p->array->nr_active++; - rq->nr_running++; - } - } else { - /* Not the local CPU - must adjust timestamp */ - p->timestamp = (p->timestamp - this_rq->timestamp_last_tick) - + rq->timestamp_last_tick; - __activate_task(p, rq); - if (TASK_PREEMPTS_CURR(p, rq)) - resched_task(rq->curr); + if (unlikely(!spin_trylock(&busiest->lock))) { + if (busiest < this_rq) { + spin_unlock(&this_rq->lock); + spin_lock(&busiest->lock); + spin_lock(&this_rq->lock); + } else + spin_lock(&busiest->lock); } - - double_rq_unlock(this_rq, rq); - local_irq_restore(flags); - put_cpu(); } /* @@ -1306,11 +2072,11 @@ lock_again: * 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) @@ -1321,11 +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: @@ -1333,64 +2101,30 @@ out: } /* - * sched_balance_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_balance_exec(void) +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) { - new_cpu = find_idlest_cpu(current, this_cpu, sd); - if (new_cpu != this_cpu) { - put_cpu(); - sched_migrate_task(current, new_cpu); - return; - } - } -out: + new_cpu = sched_balance_self(this_cpu, SD_BALANCE_EXEC); put_cpu(); -} - -/* - * double_lock_balance - lock the busiest runqueue, this_rq is locked already. - */ -static void double_lock_balance(runqueue_t *this_rq, runqueue_t *busiest) -{ - if (unlikely(!spin_trylock(&busiest->lock))) { - if (busiest < this_rq) { - spin_unlock(&this_rq->lock); - spin_lock(&busiest->lock); - spin_lock(&this_rq->lock); - } else - spin_lock(&busiest->lock); - } + 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; @@ -1405,9 +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: @@ -1415,40 +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; - /* Aggressive migration if we've failed balancing */ - if (idle == NEWLY_IDLE || - sd->nr_balance_failed < sd->cache_nice_tries) { - if (task_hot(p, rq->timestamp_last_tick, sd)) - return 0; - } + if (task_running(rq, p)) + return 0; + + /* + * Aggressive migration if: + * 1) task is cache cold, or + * 2) too many balance attempts have failed. + */ + if (sd->nr_balance_failed > sd->cache_nice_tries) + return 1; + + if (task_hot(p, rq->timestamp_last_tick, sd)) + 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) +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) { - prio_array_t *array, *dst_array; + 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 @@ -1483,91 +2242,208 @@ 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++; goto skip_bitmap; } + +#ifdef CONFIG_SCHEDSTATS + if (task_hot(tmp, busiest->timestamp_last_tick, sd)) + schedstat_inc(sd, lb_hot_gained[idle]); +#endif + 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++; goto skip_bitmap; } out: + /* + * Right now, this is the only place pull_task() is called, + * so we can safely collect pull_task() stats here rather than + * 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 { - cpumask_t tmp; - unsigned long load; + unsigned long load, group_capacity; int local_group; - int i, nr_cpus = 0; + 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; - cpus_and(tmp, group->cpumask, cpu_online_map); - if (unlikely(cpus_empty(tmp))) - goto nextgroup; + 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; - for_each_cpu_mask(i, tmp) { /* 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); - nr_cpus++; avg_load += load; + sum_nr_running += rq->nr_running; + sum_weighted_load += rq->raw_weighted_load; } - if (!nr_cpus) - goto nextgroup; - total_load += avg_load; total_pwr += group->cpu_power; /* 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; @@ -1576,6 +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 @@ -1587,18 +2464,49 @@ nextgroup: * by pulling tasks to us. Be careful of negative numbers as they'll * appear as very large values with unsigned longs. */ - *imbalance = min(max_load - avg_load, avg_load - this_load); + 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 = (*imbalance * min(busiest->cpu_power, this->cpu_power)) - / SCHED_LOAD_SCALE; + *imbalance = min(max_pull * busiest->cpu_power, + (avg_load - this_load) * this->cpu_power) + / SCHED_LOAD_SCALE; - if (*imbalance < SCHED_LOAD_SCALE - 1) { - 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; } @@ -1608,43 +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 */ - tmp = SCHED_LOAD_SCALE*SCHED_LOAD_SCALE/this->cpu_power; - if (max_load < tmp) - tmp = max_load; - pwr_move += this->cpu_power*min(SCHED_LOAD_SCALE, this_load + tmp); + if (max_load*busiest->cpu_power < + busiest_load_per_task*SCHED_LOAD_SCALE) + tmp = max_load*busiest->cpu_power/this->cpu_power; + else + 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 another 8th of a CPU worth of throughput */ - if (pwr_move < pwr_now + SCHED_LOAD_SCALE / 8) + /* 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 + 1) / SCHED_LOAD_SCALE; - return busiest; out_balanced: - if (busiest && (idle == NEWLY_IDLE || - (idle == 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; } @@ -1652,59 +2564,83 @@ 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) { - cpumask_t tmp; - unsigned long load, max_load = 0; - runqueue_t *busiest = NULL; + struct rq *busiest = NULL, *rq; + unsigned long max_load = 0; int i; - cpus_and(tmp, group->cpumask, cpu_online_map); - for_each_cpu_mask(i, tmp) { - load = source_load(i); + for_each_cpu_mask(i, group->cpumask) { + + if (!cpu_isset(i, *cpus)) + continue; - if (load > max_load) { - max_load = load; - busiest = cpu_rq(i); + 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; - spin_lock(&this_rq->lock); + if (idle != NOT_IDLE && sd->flags & SD_SHARE_CPUPOWER && + !sched_smt_power_savings) + sd_idle = 1; - group = find_busiest_group(sd, this_cpu, &imbalance, idle); - if (!group) - goto out_balanced; + schedstat_inc(sd, lb_cnt[idle]); - busiest = find_busiest_queue(group); - if (!busiest) +redo: + group = find_busiest_group(sd, this_cpu, &imbalance, idle, &sd_idle, + &cpus); + if (!group) { + schedstat_inc(sd, lb_nobusyg[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); + } + + busiest = find_busiest_queue(group, idle, imbalance, &cpus); + if (!busiest) { + schedstat_inc(sd, lb_nobusyq[idle]); goto out_balanced; } + BUG_ON(busiest == this_rq); + + schedstat_add(sd, lb_imbalance[idle], imbalance); + nr_moved = 0; if (busiest->nr_running > 1) { /* @@ -1713,50 +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; } } else sd->nr_balance_failed = 0; - /* We were unbalanced, so reset the balancing interval */ - sd->balance_interval = sd->min_interval; + 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; } @@ -1767,117 +2742,139 @@ 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]); +redo: + group = find_busiest_group(sd, this_cpu, &imbalance, NEWLY_IDLE, + &sd_idle, &cpus); + if (!group) { + schedstat_inc(sd, lb_nobusyg[NEWLY_IDLE]); + goto out_balanced; + } - group = find_busiest_group(sd, this_cpu, &imbalance, NEWLY_IDLE); - if (!group) - goto out; + busiest = find_busiest_queue(group, NEWLY_IDLE, imbalance, + &cpus); + if (!busiest) { + schedstat_inc(sd, lb_nobusyq[NEWLY_IDLE]); + goto out_balanced; + } - busiest = find_busiest_queue(group); - if (!busiest || busiest == this_rq) - goto out; + BUG_ON(busiest == this_rq); + + schedstat_add(sd, lb_imbalance[NEWLY_IDLE], imbalance); - /* Attempt to move tasks */ - double_lock_balance(this_rq, busiest); + 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); - nr_moved = move_tasks(this_rq, this_cpu, busiest, - imbalance, sd, NEWLY_IDLE); + if (!nr_moved) { + cpu_clear(cpu_of(busiest), cpus); + if (!cpus_empty(cpus)) + goto redo; + } + } - spin_unlock(&busiest->lock); + 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; - } } } } /* - * active_load_balance is run by migration threads. It pushes a running - * task off the cpu. It can be required to correctly have at least 1 task - * running on each physical CPU where possible, and not have a physical / - * logical imbalance. + * active_load_balance is run by migration threads. It pushes running tasks + * off the busiest CPU onto idle CPUs. It requires at least 1 task to be + * running on each physical CPU where possible, and avoids physical / + * logical imbalances. * - * Called with busiest locked. + * Called with busiest_rq locked. */ -static void active_load_balance(runqueue_t *busiest, 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 *group, *busy_group; - int i; + struct rq *target_rq; - if (busiest->nr_running <= 1) + /* Is there any task to move? */ + if (busiest_rq->nr_running <= 1) return; - for_each_domain(busiest_cpu, sd) - if (cpu_isset(busiest->push_cpu, sd->span)) - break; - if (!sd) { - WARN_ON(1); - return; - } + target_rq = cpu_rq(target_cpu); - group = sd->groups; - while (!cpu_isset(busiest_cpu, group->cpumask)) - group = group->next; - busy_group = group; - - group = sd->groups; - do { - cpumask_t tmp; - runqueue_t *rq; - int push_cpu = 0; - - if (group == busy_group) - goto next_group; + /* + * 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); - cpus_and(tmp, group->cpumask, cpu_online_map); - if (!cpus_weight(tmp)) - goto next_group; + /* move a task from busiest_rq to target_rq */ + double_lock_balance(busiest_rq, target_rq); - for_each_cpu_mask(i, tmp) { - if (!idle_cpu(i)) - goto next_group; - push_cpu = i; - } + /* 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; + } - rq = cpu_rq(push_cpu); + if (likely(sd)) { + schedstat_inc(sd, alb_cnt); - /* - * This condition is "impossible", but since load - * balancing is inherently a bit racy and statistical, - * it can trigger.. Reported by Bjorn Helgaas on a - * 128-cpu setup. - */ - if (unlikely(busiest == rq)) - goto next_group; - double_lock_balance(busiest, rq); - move_tasks(rq, push_cpu, busiest, 1, sd, IDLE); - spin_unlock(&rq->lock); -next_group: - group = group->next; - } while (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); } /* @@ -1889,32 +2886,43 @@ next_group: * 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; + + /* 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) { - unsigned long interval = sd->balance_interval; + if (!(sd->flags & SD_LOAD_BALANCE)) + continue; - if (idle != IDLE) + interval = sd->balance_interval; + if (idle != SCHED_IDLE) interval *= sd->busy_factor; /* scale ms to jiffies */ @@ -1924,7 +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; @@ -1935,33 +2947,64 @@ 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); /* * If an SMT sibling task has been put to sleep for priority * reasons reschedule the idle task to see if it can now run. */ if (rq->nr_running) { resched_task(rq->idle); - return 1; + ret = 1; } + spin_unlock(&rq->lock); #endif - return 0; + return ret; } DEFINE_PER_CPU(struct kernel_stat, kstat); 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(struct task_struct *p, struct rq *rq, unsigned long long now) +{ + 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 struct task_struct *p) +{ + unsigned long long ns; + unsigned long flags; + + local_irq_save(flags); + ns = max(p->timestamp, task_rq(p)->timestamp_last_tick); + ns = p->sched_time + sched_clock() - ns; + local_irq_restore(flags); + + return ns; +} + /* * We place interactive tasks back into the active array, if possible. * @@ -1972,11 +3015,94 @@ EXPORT_PER_CPU_SYMBOL(kstat); * 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. + * @p: the process that the cpu time gets accounted to + * @hardirq_offset: the offset to subtract from hardirq_count() + * @cputime: the cpu time spent in user space since the last update + */ +void account_user_time(struct task_struct *p, cputime_t cputime) +{ + struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat; + struct vx_info *vxi = p->vx_info; /* p is _always_ current */ + cputime64_t tmp; + int nice = (TASK_NICE(p) > 0); + + p->utime = cputime_add(p->utime, cputime); + vx_account_user(vxi, cputime, nice); + + /* Add user time to cpustat. */ + tmp = cputime_to_cputime64(cputime); + if (nice) + cpustat->nice = cputime64_add(cpustat->nice, tmp); + else + cpustat->user = cputime64_add(cpustat->user, tmp); +} + +/* + * Account system cpu time to a process. + * @p: the process that the cpu time gets accounted to + * @hardirq_offset: the offset to subtract from hardirq_count() + * @cputime: the cpu time spent in kernel space since the last update + */ +void account_system_time(struct task_struct *p, int hardirq_offset, + cputime_t cputime) +{ + struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat; + struct vx_info *vxi = p->vx_info; /* p is _always_ current */ + struct rq *rq = this_rq(); + cputime64_t tmp; + + p->stime = cputime_add(p->stime, cputime); + vx_account_system(vxi, cputime, (p == rq->idle)); + + /* Add system time to cpustat. */ + tmp = cputime_to_cputime64(cputime); + if (hardirq_count() - hardirq_offset) + cpustat->irq = cputime64_add(cpustat->irq, tmp); + else if (softirq_count()) + cpustat->softirq = cputime64_add(cpustat->softirq, tmp); + else if (p != rq->idle) + cpustat->system = cputime64_add(cpustat->system, tmp); + else if (atomic_read(&rq->nr_iowait) > 0) + cpustat->iowait = cputime64_add(cpustat->iowait, tmp); + else + cpustat->idle = cputime64_add(cpustat->idle, tmp); + /* Account for system time used */ + acct_update_integrals(p); +} + +/* + * Account for involuntary wait time. + * @p: the process from which the cpu time has been stolen + * @steal: the cpu time spent in involuntary wait + */ +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); + struct rq *rq = this_rq(); + + if (p == rq->idle) { + p->stime = cputime_add(p->stime, steal); + if (atomic_read(&rq->nr_iowait) > 0) + cpustat->iowait = cputime64_add(cpustat->iowait, tmp); + else + cpustat->idle = cputime64_add(cpustat->idle, tmp); + } else + cpustat->steal = cputime64_add(cpustat->steal, tmp); +} /* * This function gets called by the timer code, with HZ frequency. @@ -1985,45 +3111,27 @@ EXPORT_PER_CPU_SYMBOL(kstat); * It also gets called by the fork code, when changing the parent's * timeslices. */ -void scheduler_tick(int user_ticks, int sys_ticks) +void scheduler_tick(void) { + unsigned long long now = sched_clock(); + struct task_struct *p = current; int cpu = smp_processor_id(); - struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat; - runqueue_t *rq = this_rq(); - task_t *p = current; - - rq->timestamp_last_tick = sched_clock(); + struct rq *rq = cpu_rq(cpu); - if (rcu_pending(cpu)) - rcu_check_callbacks(cpu, user_ticks); + update_cpu_clock(p, rq, now); - /* note: this timer irq context must be accounted for as well */ - if (hardirq_count() - HARDIRQ_OFFSET) { - cpustat->irq += sys_ticks; - sys_ticks = 0; - } else if (softirq_count()) { - cpustat->softirq += sys_ticks; - sys_ticks = 0; - } + rq->timestamp_last_tick = now; if (p == rq->idle) { - if (!--rq->idle_tokens && !list_empty(&rq->hold_queue)) - set_need_resched(); - - if (atomic_read(&rq->nr_iowait) > 0) - cpustat->iowait += sys_ticks; - else - cpustat->idle += sys_ticks; if (wake_priority_sleeper(rq)) goto out; - rebalance_tick(cpu, rq, IDLE); +#ifdef CONFIG_VSERVER_HARDCPU_IDLE + if (!--rq->idle_tokens && !list_empty(&rq->hold_queue)) + set_need_resched(); +#endif + rebalance_tick(cpu, rq, SCHED_IDLE); return; } - if (TASK_NICE(p) > 0) - cpustat->nice += user_ticks; - else - cpustat->user += user_ticks; - cpustat->system += sys_ticks; /* Task might have expired already, but not scheduled off yet */ if (p->array != rq->active) { @@ -2038,7 +3146,7 @@ void scheduler_tick(int user_ticks, int sys_ticks) * timeslice. This makes it possible for interactive tasks * to use up their timeslices at their highest priority levels. */ - if (unlikely(rt_task(p))) { + if (rt_task(p)) { /* * RR tasks need a special form of timeslice management. * FIFO tasks have no timeslices. @@ -2049,8 +3157,7 @@ void scheduler_tick(int user_ticks, int sys_ticks) set_tsk_need_resched(p); /* put it at the end of the queue: */ - dequeue_task(p, rq->active); - enqueue_task(p, rq->active); + requeue_task(p, rq->active); } goto out_unlock; } @@ -2063,7 +3170,7 @@ void scheduler_tick(int user_ticks, int sys_ticks) 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; @@ -2091,10 +3198,8 @@ void scheduler_tick(int user_ticks, int sys_ticks) (p->time_slice >= TIMESLICE_GRANULARITY(p)) && (p->array == rq->active)) { - dequeue_task(p, rq->active); + requeue_task(p, rq->active); set_tsk_need_resched(p); - p->prio = effective_prio(p); - enqueue_task(p, rq->active); } } out_unlock: @@ -2104,53 +3209,97 @@ out: } #ifdef CONFIG_SCHED_SMT -static inline void wake_sleeping_dependent(int cpu, runqueue_t *rq) +static inline void wakeup_busy_runqueue(struct rq *rq) +{ + /* 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); +} + +/* + * 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; - struct sched_domain *sd = rq->sd; - cpumask_t sibling_map; - if (!(sd->flags & SD_SHARE_CPUPOWER)) + for_each_domain(this_cpu, tmp) { + if (tmp->flags & SD_SHARE_CPUPOWER) { + sd = tmp; + break; + } + } + + if (!sd) return; - cpus_and(sibling_map, sd->span, cpu_online_map); - for_each_cpu_mask(i, sibling_map) { - runqueue_t *smt_rq; + for_each_cpu_mask(i, sd->span) { + struct rq *smt_rq = cpu_rq(i); - if (i == cpu) + if (i == this_cpu) + continue; + if (unlikely(!spin_trylock(&smt_rq->lock))) continue; - smt_rq = cpu_rq(i); - - /* - * 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); } } -static inline int dependent_sleeper(int cpu, runqueue_t *rq, task_t *p) +/* + * 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; +} + +/* + * 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 = rq->sd; - cpumask_t sibling_map; + struct sched_domain *tmp, *sd = NULL; int ret = 0, i; - if (!(sd->flags & SD_SHARE_CPUPOWER)) + /* kernel/rt threads do not participate in dependent sleeping */ + if (!p->mm || rt_task(p)) + return 0; + + for_each_domain(this_cpu, tmp) { + if (tmp->flags & SD_SHARE_CPUPOWER) { + sd = tmp; + break; + } + } + + if (!sd) return 0; - cpus_and(sibling_map, sd->span, cpu_online_map); - for_each_cpu_mask(i, sibling_map) { - runqueue_t *smt_rq; - task_t *smt_curr; + for_each_cpu_mask(i, sd->span) { + struct task_struct *smt_curr; + struct rq *smt_rq; - if (i == cpu) + if (i == this_cpu) continue; smt_rq = cpu_rq(i); + if (unlikely(!spin_trylock(&smt_rq->lock))) + continue; + smt_curr = smt_rq->curr; + if (!smt_curr->mm) + goto unlock; + /* * If a user task with lower static priority than the * running task on the SMT sibling is trying to schedule, @@ -2159,139 +3308,195 @@ static inline int dependent_sleeper(int cpu, runqueue_t *rq, task_t *p) * 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); } return ret; } #else -static inline void wake_sleeping_dependent(int cpu, runqueue_t *rq) +static inline void wake_sleeping_dependent(int this_cpu) { } - -static inline int dependent_sleeper(int cpu, runqueue_t *rq, task_t *p) +static inline int +dependent_sleeper(int this_cpu, struct rq *this_rq, struct task_struct *p) { return 0; } #endif +#if defined(CONFIG_PREEMPT) && defined(CONFIG_DEBUG_PREEMPT) + +void fastcall add_preempt_count(int val) +{ + /* + * Underflow? + */ + if (DEBUG_LOCKS_WARN_ON((preempt_count() < 0))) + return; + preempt_count() += val; + /* + * Spinlock count overflowing soon? + */ + DEBUG_LOCKS_WARN_ON((preempt_count() & PREEMPT_MASK) >= PREEMPT_MASK-10); +} +EXPORT_SYMBOL(add_preempt_count); + +void fastcall sub_preempt_count(int val) +{ + /* + * Underflow? + */ + if (DEBUG_LOCKS_WARN_ON(val > preempt_count())) + return; + /* + * Is the spinlock portion underflowing? + */ + 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; -#ifdef CONFIG_VSERVER_HARDCPU + 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->state & (TASK_DEAD | TASK_ZOMBIE)))) { - if (unlikely(in_atomic())) { - printk(KERN_ERR "bad: scheduling while atomic!\n"); - 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)); need_resched: preempt_disable(); prev = current; + release_kernel_lock(prev); +need_resched_nonpreemptible: rq = this_rq(); - release_kernel_lock(prev); + /* + * The idle thread is not allowed to schedule! + * Remove this check after it has been exercised a bit. + */ + if (unlikely(prev == rq->idle) && prev->state != TASK_RUNNING) { + printk(KERN_ERR "bad: scheduling from the idle thread!\n"); + dump_stack(); + } + + schedstat_inc(rq, sched_cnt); now = sched_clock(); - if (likely(now - prev->timestamp < NS_MAX_SLEEP_AVG)) + if (likely((long long)(now - prev->timestamp) < NS_MAX_SLEEP_AVG)) { run_time = now - prev->timestamp; - else + if (unlikely((long long)(now - prev->timestamp) < 0)) + run_time = 0; + } else run_time = NS_MAX_SLEEP_AVG; /* - * Tasks with interactive credits get charged less run_time - * at high sleep_avg to delay them losing their interactive - * status + * Tasks charged proportionately less run_time at high sleep_avg to + * delay them losing their interactive status */ - if (HIGH_CREDIT(prev)) - run_time /= (CURRENT_BONUS(prev) ? : 1); + run_time /= (CURRENT_BONUS(prev) ? : 1); spin_lock_irq(&rq->lock); - /* - * if entering off of a kernel preemption go straight - * to picking the next task. - */ + if (unlikely(prev->flags & PF_DEAD)) + prev->state = EXIT_DEAD; + switch_count = &prev->nivcsw; if (prev->state && !(preempt_count() & PREEMPT_ACTIVE)) { switch_count = &prev->nvcsw; if (unlikely((prev->state & TASK_INTERRUPTIBLE) && unlikely(signal_pending(prev)))) prev->state = TASK_RUNNING; - else + else { + if (prev->state == TASK_UNINTERRUPTIBLE) { + rq->nr_uninterruptible++; + vx_uninterruptible_inc(prev); + } deactivate_task(prev, rq); + } } -#ifdef CONFIG_VSERVER_HARDCPU +#ifdef CONFIG_VSERVER_HARDCPU if (!list_empty(&rq->hold_queue)) { struct list_head *l, *n; int ret; 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; vxi = next->vx_info; ret = vx_tokens_recalc(vxi); - // tokens = vx_tokens_avail(next); if (ret > 0) { - list_del(&next->run_list); - next->state &= ~TASK_ONHOLD; - recalc_task_prio(next, now); - __activate_task(next, rq); - // printk("··· unhold %p\n", next); + vx_unhold_task(vxi, next, rq); break; } if ((ret < 0) && (maxidle < ret)) maxidle = ret; - } + } } rq->idle_tokens = -maxidle; pick_next: #endif + cpu = smp_processor_id(); if (unlikely(!rq->nr_running)) { idle_balance(cpu, rq); if (!rq->nr_running) { next = rq->idle; rq->expired_timestamp = 0; - wake_sleeping_dependent(cpu, rq); + wake_sleeping_dependent(cpu); goto switch_tasks; } } @@ -2301,6 +3506,7 @@ pick_next: /* * Switch the active and expired arrays. */ + schedstat_inc(rq, sched_switch); rq->active = rq->expired; rq->expired = array; array = rq->active; @@ -2310,87 +3516,100 @@ pick_next: idx = sched_find_first_bit(array->bitmap); queue = array->queue + idx; - next = list_entry(queue->next, task_t, run_list); - - if (dependent_sleeper(cpu, rq, next)) { - next = rq->idle; - goto switch_tasks; - } + next = list_entry(queue->next, struct task_struct, run_list); -#ifdef CONFIG_VSERVER_HARDCPU vxi = next->vx_info; - if (vxi && __vx_flags(vxi->vx_flags, - VXF_SCHED_PAUSE|VXF_SCHED_HARD, 0)) { +#ifdef CONFIG_VSERVER_HARDCPU + if (vx_info_flags(vxi, VXF_SCHED_PAUSE|VXF_SCHED_HARD, 0)) { int ret = vx_tokens_recalc(vxi); if (unlikely(ret <= 0)) { if (ret && (rq->idle_tokens > -ret)) rq->idle_tokens = -ret; - deactivate_task(next, rq); - list_add_tail(&next->run_list, &rq->hold_queue); - next->state |= TASK_ONHOLD; + vx_hold_task(vxi, next, rq); goto pick_next; } - } + } else /* well, looks ugly but not as ugly as the ifdef-ed version */ #endif + 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(task_cpu(prev))++; + rcu_qsctr_inc(task_cpu(prev)); + + update_cpu_clock(prev, rq, now); prev->sleep_avg -= run_time; - if ((long)prev->sleep_avg <= 0) { + if ((long)prev->sleep_avg <= 0) prev->sleep_avg = 0; - if (!(HIGH_CREDIT(prev) || LOW_CREDIT(prev))) - prev->interactive_credit--; - } - prev->timestamp = now; + prev->timestamp = prev->last_ran = now; + sched_info_switch(prev, next); if (likely(prev != next)) { next->timestamp = now; rq->nr_switches++; 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); - reacquire_kernel_lock(current); + prev = current; + if (unlikely(reacquire_kernel_lock(prev) < 0)) + goto need_resched_nonpreemptible; preempt_enable_no_resched(); - if (test_thread_flag(TIF_NEED_RESCHED)) + 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. */ asmlinkage void __sched preempt_schedule(void) { struct thread_info *ti = current_thread_info(); - +#ifdef CONFIG_PREEMPT_BKL + struct task_struct *task = current; + int saved_lock_depth; +#endif /* * If there is a non-zero preempt_count or interrupts are disabled, * we do not want to preempt the current task. Just return.. @@ -2399,25 +3618,77 @@ asmlinkage void __sched preempt_schedule(void) return; need_resched: - ti->preempt_count = PREEMPT_ACTIVE; + add_preempt_count(PREEMPT_ACTIVE); + /* + * We keep the big kernel semaphore locked, but we + * clear ->lock_depth so that schedule() doesnt + * auto-release the semaphore: + */ +#ifdef CONFIG_PREEMPT_BKL + saved_lock_depth = task->lock_depth; + task->lock_depth = -1; +#endif schedule(); - ti->preempt_count = 0; +#ifdef CONFIG_PREEMPT_BKL + task->lock_depth = saved_lock_depth; +#endif + sub_preempt_count(PREEMPT_ACTIVE); /* we could miss a preemption opportunity between schedule and now */ barrier(); if (unlikely(test_thread_flag(TIF_NEED_RESCHED))) goto need_resched; } - EXPORT_SYMBOL(preempt_schedule); -#endif /* CONFIG_PREEMPT */ -int default_wake_function(wait_queue_t *curr, unsigned mode, int sync, void *key) +/* + * 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. + */ +asmlinkage void __sched preempt_schedule_irq(void) { - task_t *p = curr->task; - return try_to_wake_up(p, mode, sync); + struct thread_info *ti = current_thread_info(); +#ifdef CONFIG_PREEMPT_BKL + struct task_struct *task = current; + int saved_lock_depth; +#endif + /* Catch callers which need to be fixed */ + BUG_ON(ti->preempt_count || !irqs_disabled()); + +need_resched: + add_preempt_count(PREEMPT_ACTIVE); + /* + * We keep the big kernel semaphore locked, but we + * clear ->lock_depth so that schedule() doesnt + * auto-release the semaphore: + */ +#ifdef CONFIG_PREEMPT_BKL + saved_lock_depth = task->lock_depth; + task->lock_depth = -1; +#endif + local_irq_enable(); + schedule(); + local_irq_disable(); +#ifdef CONFIG_PREEMPT_BKL + task->lock_depth = saved_lock_depth; +#endif + sub_preempt_count(PREEMPT_ACTIVE); + + /* we could miss a preemption opportunity between schedule and now */ + barrier(); + if (unlikely(test_thread_flag(TIF_NEED_RESCHED))) + goto need_resched; } +#endif /* CONFIG_PREEMPT */ + +int default_wake_function(wait_queue_t *curr, unsigned mode, int sync, + void *key) +{ + return try_to_wake_up(curr->private, mode, sync); +} EXPORT_SYMBOL(default_wake_function); /* @@ -2435,13 +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; } } @@ -2451,9 +3720,10 @@ static void __wake_up_common(wait_queue_head_t *q, unsigned int mode, * @q: the waitqueue * @mode: which threads * @nr_exclusive: how many wake-one or wake-many threads to wake up + * @key: is directly passed to the wakeup function */ void fastcall __wake_up(wait_queue_head_t *q, unsigned int mode, - int nr_exclusive, void *key) + int nr_exclusive, void *key) { unsigned long flags; @@ -2461,7 +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); /* @@ -2473,7 +3742,7 @@ void fastcall __wake_up_locked(wait_queue_head_t *q, unsigned int mode) } /** - * __wake_up - sync- wake up threads blocked on a waitqueue. + * __wake_up_sync - wake up threads blocked on a waitqueue. * @q: the waitqueue * @mode: which threads * @nr_exclusive: how many wake-one or wake-many threads to wake up @@ -2485,7 +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; @@ -2529,6 +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); @@ -2548,66 +3819,167 @@ void fastcall __sched wait_for_completion(struct completion *x) } EXPORT_SYMBOL(wait_for_completion); -#define SLEEP_ON_VAR \ - unsigned long flags; \ - wait_queue_t wait; \ - init_waitqueue_entry(&wait, current); - -#define SLEEP_ON_HEAD \ - spin_lock_irqsave(&q->lock,flags); \ - __add_wait_queue(q, &wait); \ - spin_unlock(&q->lock); - -#define SLEEP_ON_TAIL \ - spin_lock_irq(&q->lock); \ - __remove_wait_queue(q, &wait); \ - spin_unlock_irqrestore(&q->lock, flags); - -void fastcall __sched interruptible_sleep_on(wait_queue_head_t *q) +unsigned long fastcall __sched +wait_for_completion_timeout(struct completion *x, unsigned long timeout) { - SLEEP_ON_VAR + might_sleep(); - current->state = TASK_INTERRUPTIBLE; + spin_lock_irq(&x->wait.lock); + if (!x->done) { + DECLARE_WAITQUEUE(wait, current); - SLEEP_ON_HEAD - schedule(); - SLEEP_ON_TAIL + wait.flags |= WQ_FLAG_EXCLUSIVE; + __add_wait_queue_tail(&x->wait, &wait); + do { + __set_current_state(TASK_UNINTERRUPTIBLE); + spin_unlock_irq(&x->wait.lock); + timeout = schedule_timeout(timeout); + spin_lock_irq(&x->wait.lock); + if (!timeout) { + __remove_wait_queue(&x->wait, &wait); + goto out; + } + } while (!x->done); + __remove_wait_queue(&x->wait, &wait); + } + x->done--; +out: + spin_unlock_irq(&x->wait.lock); + return timeout; } +EXPORT_SYMBOL(wait_for_completion_timeout); -EXPORT_SYMBOL(interruptible_sleep_on); - -long fastcall __sched interruptible_sleep_on_timeout(wait_queue_head_t *q, long timeout) +int fastcall __sched wait_for_completion_interruptible(struct completion *x) { - SLEEP_ON_VAR + int ret = 0; - current->state = TASK_INTERRUPTIBLE; + might_sleep(); - SLEEP_ON_HEAD - timeout = schedule_timeout(timeout); - SLEEP_ON_TAIL + spin_lock_irq(&x->wait.lock); + if (!x->done) { + DECLARE_WAITQUEUE(wait, current); + + wait.flags |= WQ_FLAG_EXCLUSIVE; + __add_wait_queue_tail(&x->wait, &wait); + do { + if (signal_pending(current)) { + ret = -ERESTARTSYS; + __remove_wait_queue(&x->wait, &wait); + goto out; + } + __set_current_state(TASK_INTERRUPTIBLE); + spin_unlock_irq(&x->wait.lock); + schedule(); + spin_lock_irq(&x->wait.lock); + } while (!x->done); + __remove_wait_queue(&x->wait, &wait); + } + x->done--; +out: + spin_unlock_irq(&x->wait.lock); + + return ret; +} +EXPORT_SYMBOL(wait_for_completion_interruptible); + +unsigned long fastcall __sched +wait_for_completion_interruptible_timeout(struct completion *x, + unsigned long timeout) +{ + might_sleep(); + spin_lock_irq(&x->wait.lock); + if (!x->done) { + DECLARE_WAITQUEUE(wait, current); + + wait.flags |= WQ_FLAG_EXCLUSIVE; + __add_wait_queue_tail(&x->wait, &wait); + do { + if (signal_pending(current)) { + timeout = -ERESTARTSYS; + __remove_wait_queue(&x->wait, &wait); + goto out; + } + __set_current_state(TASK_INTERRUPTIBLE); + spin_unlock_irq(&x->wait.lock); + timeout = schedule_timeout(timeout); + spin_lock_irq(&x->wait.lock); + if (!timeout) { + __remove_wait_queue(&x->wait, &wait); + goto out; + } + } while (!x->done); + __remove_wait_queue(&x->wait, &wait); + } + x->done--; +out: + spin_unlock_irq(&x->wait.lock); return timeout; } +EXPORT_SYMBOL(wait_for_completion_interruptible_timeout); -EXPORT_SYMBOL(interruptible_sleep_on_timeout); -void fastcall __sched sleep_on(wait_queue_head_t *q) +#define SLEEP_ON_VAR \ + unsigned long flags; \ + wait_queue_t wait; \ + init_waitqueue_entry(&wait, current); + +#define SLEEP_ON_HEAD \ + spin_lock_irqsave(&q->lock,flags); \ + __add_wait_queue(q, &wait); \ + spin_unlock(&q->lock); + +#define SLEEP_ON_TAIL \ + spin_lock_irq(&q->lock); \ + __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 - current->state = TASK_UNINTERRUPTIBLE; + 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) +{ + SLEEP_ON_VAR + + SLEEP_ON_BKLCHECK + + current->state = TASK_INTERRUPTIBLE; + + SLEEP_ON_HEAD + timeout = schedule_timeout(timeout); + SLEEP_ON_TAIL -EXPORT_SYMBOL(sleep_on); + return timeout; +} +EXPORT_SYMBOL(interruptible_sleep_on_timeout); 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 @@ -2619,12 +3991,65 @@ long fastcall __sched sleep_on_timeout(wait_queue_head_t *q, long timeout) EXPORT_SYMBOL(sleep_on_timeout); -void set_user_nice(task_t *p, long nice) +#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) +{ + struct prio_array *array; + unsigned long flags; + struct rq *rq; + int oldprio; + + BUG_ON(prio < 0 || prio > MAX_PRIO); + + rq = task_rq_lock(p, &flags); + + 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); +} + +#endif + +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; @@ -2634,27 +4059,30 @@ void set_user_nice(task_t *p, long nice) */ rq = task_rq_lock(p, &flags); /* - * The RT priorities are set via setscheduler(), but we still + * 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: @@ -2665,9 +4093,22 @@ void set_user_nice(task_t *p, long nice) out_unlock: task_rq_unlock(rq, &flags); } - EXPORT_SYMBOL(set_user_nice); +/* + * can_nice - check if a task can reduce its nice value + * @p: task + * @nice: nice value + */ +int can_nice(const struct task_struct *p, const int 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)); +} + #ifdef __ARCH_WANT_SYS_NICE /* @@ -2679,20 +4120,15 @@ EXPORT_SYMBOL(set_user_nice); */ asmlinkage long sys_nice(int increment) { - int retval; - long nice; + long nice, retval; /* * Setpriority might change our priority at the same moment. * We don't have to worry. Conceptually one call occurs first * and we have a single winner. */ - if (increment < 0) { - if (!capable(CAP_SYS_NICE)) - return -EPERM; - if (increment < -40) - increment = -40; - } + if (increment < -40) + increment = -40; if (increment > 40) increment = 40; @@ -2702,6 +4138,9 @@ asmlinkage long sys_nice(int increment) if (nice > 19) nice = 19; + if (increment < 0 && !can_nice(current, nice)) + return vx_flags(VXF_IGNEG_NICE, 0) ? 0 : -EPERM; + retval = security_task_setnice(current, nice); if (retval) return retval; @@ -2720,7 +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(task_t *p) +int task_prio(const struct task_struct *p) { return p->prio - MAX_RT_PRIO; } @@ -2729,12 +4168,11 @@ int task_prio(task_t *p) * task_nice - return the nice value of a given task. * @p: the task in question. */ -int task_nice(task_t *p) +int task_nice(const struct task_struct *p) { return TASK_NICE(p); } - -EXPORT_SYMBOL(task_nice); +EXPORT_SYMBOL_GPL(task_nice); /** * idle_cpu - is a given cpu idle currently? @@ -2745,13 +4183,20 @@ 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. + */ +struct task_struct *idle_task(int cpu) +{ + return cpu_rq(cpu)->idle; +} /** * 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; } @@ -2760,90 +4205,109 @@ 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); } -/* - * setscheduler - change the scheduling policy and/or RT priority of a thread. +/** + * sched_setscheduler - change the scheduling policy and/or RT priority of + * a thread. + * @p: the task in question. + * @policy: new policy. + * @param: structure containing the new RT priority. */ -static int setscheduler(pid_t pid, int policy, struct sched_param __user *param) +int sched_setscheduler(struct task_struct *p, int policy, + struct sched_param *param) { - struct sched_param lp; - int retval = -EINVAL; - int oldprio; - prio_array_t *array; + int retval, oldprio, oldpolicy = -1; + struct prio_array *array; unsigned long flags; - runqueue_t *rq; - task_t *p; - - if (!param || pid < 0) - goto out_nounlock; - - retval = -EFAULT; - if (copy_from_user(&lp, param, sizeof(struct sched_param))) - goto out_nounlock; + 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 && policy != SCHED_BATCH) + return -EINVAL; /* - * We play safe to avoid deadlocks. + * Valid priorities for SCHED_FIFO and SCHED_RR are + * 1..MAX_USER_RT_PRIO-1, valid priority for SCHED_NORMAL and + * SCHED_BATCH is 0. */ - read_lock_irq(&tasklist_lock); - - p = find_process_by_pid(pid); - - retval = -ESRCH; - if (!p) - goto out_unlock_tasklist; + if (param->sched_priority < 0 || + (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 || policy == SCHED_BATCH) + != (param->sched_priority == 0)) + return -EINVAL; /* - * To be able to change p->policy safely, the apropriate - * runqueue lock must be held. + * Allow unprivileged RT tasks to decrease priority: */ - rq = task_rq_lock(p, &flags); - - if (policy < 0) - policy = p->policy; - else { - retval = -EINVAL; - if (policy != SCHED_FIFO && policy != SCHED_RR && - policy != SCHED_NORMAL) - goto out_unlock; + 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; /* - * Valid priorities for SCHED_FIFO and SCHED_RR are - * 1..MAX_USER_RT_PRIO-1, valid priority for SCHED_NORMAL is 0. + * make sure no PI-waiters arrive (or leave) while we are + * changing the priority of the task: */ - retval = -EINVAL; - if (lp.sched_priority < 0 || lp.sched_priority > MAX_USER_RT_PRIO-1) - goto out_unlock; - if ((policy == SCHED_NORMAL) != (lp.sched_priority == 0)) - goto out_unlock; - - retval = -EPERM; - if ((policy == SCHED_FIFO || policy == SCHED_RR) && - !capable(CAP_SYS_NICE)) - goto out_unlock; - if ((current->euid != p->euid) && (current->euid != p->uid) && - !capable(CAP_SYS_NICE)) - goto out_unlock; - - retval = security_task_setscheduler(p, policy, &lp); - if (retval) - goto out_unlock; - + 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); + /* recheck policy now with rq lock held */ + if (unlikely(oldpolicy != -1 && oldpolicy != p->policy)) { + policy = oldpolicy = -1; + __task_rq_unlock(rq); + spin_unlock_irqrestore(&p->pi_lock, flags); + goto recheck; + } array = p->array; if (array) - deactivate_task(p, task_rq(p)); - retval = 0; + deactivate_task(p, rq); oldprio = p->prio; - __setscheduler(p, policy, lp.sched_priority); + __setscheduler(p, policy, param->sched_priority); if (array) { - __activate_task(p, task_rq(p)); + vx_activate_task(p); + __activate_task(p, rq); /* * Reschedule if we are currently running on this runqueue and * our priority decreased, or if we are not currently running on @@ -2855,26 +4319,52 @@ static int setscheduler(pid_t pid, int policy, struct sched_param __user *param) } else if (TASK_PREEMPTS_CURR(p, rq)) resched_task(rq->curr); } + __task_rq_unlock(rq); + spin_unlock_irqrestore(&p->pi_lock, flags); -out_unlock: - task_rq_unlock(rq, &flags); -out_unlock_tasklist: + 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) +{ + struct sched_param lparam; + struct task_struct *p; + int retval; + + if (!param || pid < 0) + return -EINVAL; + if (copy_from_user(&lparam, param, sizeof(struct sched_param))) + return -EFAULT; + read_lock_irq(&tasklist_lock); + p = find_process_by_pid(pid); + if (!p) { + read_unlock_irq(&tasklist_lock); + return -ESRCH; + } + retval = sched_setscheduler(p, policy, &lparam); read_unlock_irq(&tasklist_lock); -out_nounlock: return retval; } /** * sys_sched_setscheduler - set/change the scheduler policy and RT priority * @pid: the pid in question. - * @policy: new policy + * @policy: new policy. * @param: structure containing the new RT priority. */ asmlinkage long sys_sched_setscheduler(pid_t pid, int policy, struct sched_param __user *param) { - return setscheduler(pid, policy, param); + /* negative values for policy are not valid */ + if (policy < 0) + return -EINVAL; + + return do_sched_setscheduler(pid, policy, param); } /** @@ -2884,7 +4374,7 @@ asmlinkage long sys_sched_setscheduler(pid_t pid, int policy, */ asmlinkage long sys_sched_setparam(pid_t pid, struct sched_param __user *param) { - return setscheduler(pid, -1, param); + return do_sched_setscheduler(pid, -1, param); } /** @@ -2893,8 +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; @@ -2921,8 +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; @@ -2953,24 +4443,11 @@ out_unlock: return retval; } -/** - * sys_sched_setaffinity - set the cpu affinity of a process - * @pid: pid of the process - * @len: length in bytes of the bitmask pointed to by user_mask_ptr - * @user_mask_ptr: user-space pointer to the new cpu mask - */ -asmlinkage long sys_sched_setaffinity(pid_t pid, unsigned int len, - unsigned long __user *user_mask_ptr) +long sched_setaffinity(pid_t pid, cpumask_t new_mask) { - cpumask_t new_mask; + cpumask_t cpus_allowed; + struct task_struct *p; int retval; - task_t *p; - - if (len < sizeof(new_mask)) - return -EINVAL; - - if (copy_from_user(&new_mask, user_mask_ptr, sizeof(new_mask))) - return -EFAULT; lock_cpu_hotplug(); read_lock(&tasklist_lock); @@ -2995,6 +4472,12 @@ asmlinkage long sys_sched_setaffinity(pid_t pid, unsigned int len, !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); out_unlock: @@ -3003,23 +4486,55 @@ out_unlock: return retval; } +static int get_user_cpu_mask(unsigned long __user *user_mask_ptr, unsigned len, + cpumask_t *new_mask) +{ + if (len < sizeof(cpumask_t)) { + memset(new_mask, 0, sizeof(cpumask_t)); + } else if (len > sizeof(cpumask_t)) { + len = sizeof(cpumask_t); + } + return copy_from_user(new_mask, user_mask_ptr, len) ? -EFAULT : 0; +} + /** - * sys_sched_getaffinity - get the cpu affinity of a process + * sys_sched_setaffinity - set the cpu affinity of a process * @pid: pid of the process * @len: length in bytes of the bitmask pointed to by user_mask_ptr - * @user_mask_ptr: user-space pointer to hold the current cpu mask + * @user_mask_ptr: user-space pointer to the new cpu mask */ -asmlinkage long sys_sched_getaffinity(pid_t pid, unsigned int len, +asmlinkage long sys_sched_setaffinity(pid_t pid, unsigned int len, unsigned long __user *user_mask_ptr) { - unsigned int real_len; - cpumask_t mask; + cpumask_t new_mask; int retval; - task_t *p; - real_len = sizeof(mask); - if (len < real_len) - return -EINVAL; + retval = get_user_cpu_mask(user_mask_ptr, len, &new_mask); + if (retval) + return retval; + + return sched_setaffinity(pid, new_mask); +} + +/* + * Represents all cpu's present in the system + * In systems capable of hotplug, this map could dynamically grow + * as new cpu's are detected in the system via any platform specific + * method, such as ACPI for e.g. + */ + +cpumask_t cpu_present_map __read_mostly; +EXPORT_SYMBOL(cpu_present_map); + +#ifndef CONFIG_SMP +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; lock_cpu_hotplug(); read_lock(&tasklist_lock); @@ -3029,17 +4544,44 @@ asmlinkage long sys_sched_getaffinity(pid_t pid, unsigned int len, 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); unlock_cpu_hotplug(); if (retval) return retval; - if (copy_to_user(user_mask_ptr, &mask, real_len)) + + return 0; +} + +/** + * sys_sched_getaffinity - get the cpu affinity of a process + * @pid: pid of the process + * @len: length in bytes of the bitmask pointed to by user_mask_ptr + * @user_mask_ptr: user-space pointer to hold the current cpu mask + */ +asmlinkage long sys_sched_getaffinity(pid_t pid, unsigned int len, + unsigned long __user *user_mask_ptr) +{ + int ret; + cpumask_t mask; + + if (len < sizeof(cpumask_t)) + return -EINVAL; + + ret = sched_getaffinity(pid, &mask); + if (ret < 0) + return ret; + + if (copy_to_user(user_mask_ptr, &mask, sizeof(cpumask_t))) return -EFAULT; - return real_len; + + return sizeof(cpumask_t); } /** @@ -3051,10 +4593,10 @@ out_unlock: */ 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); /* * We implement yielding by moving the task into the expired * queue. @@ -3062,16 +4604,31 @@ asmlinkage long sys_sched_yield(void) * (special rule: RT tasks will just roundrobin in the active * array.) */ - if (unlikely(rt_task(current))) + if (rt_task(current)) target = rq->active; - dequeue_task(current, array); - enqueue_task(current, target); + if (array->nr_active == 1) { + schedstat_inc(rq, yld_act_empty); + if (!rq->expired->nr_active) + schedstat_inc(rq, yld_both_empty); + } else if (!rq->expired->nr_active) + schedstat_inc(rq, yld_exp_empty); + + if (array != target) { + dequeue_task(current, array); + enqueue_task(current, target); + } else + /* + * requeue_task is cheaper so perform that if possible. + */ + requeue_task(current, array); /* * Since we are going to call schedule() anyway, there's * 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(); @@ -3080,13 +4637,87 @@ asmlinkage long sys_sched_yield(void) return 0; } -void __sched __cond_resched(void) +static inline int __resched_legal(int expected_preempt_count) { - set_current_state(TASK_RUNNING); - schedule(); + 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(); + sub_preempt_count(PREEMPT_ACTIVE); + } while (need_resched()); +} + +int __sched cond_resched(void) +{ + if (need_resched() && __resched_legal(0)) { + __cond_resched(); + return 1; + } + return 0; } +EXPORT_SYMBOL(cond_resched); -EXPORT_SYMBOL(__cond_resched); +/* + * cond_resched_lock() - if a reschedule is pending, drop the given lock, + * call schedule, and on return reacquire the lock. + * + * This works OK both with and without CONFIG_PREEMPT. We do strange low-level + * operations here to prevent schedule() from being called twice (once via + * spin_unlock(), once by hand). + */ +int cond_resched_lock(spinlock_t *lock) +{ + int ret = 0; + + if (need_lockbreak(lock)) { + spin_unlock(lock); + cpu_relax(); + ret = 1; + spin_lock(lock); + } + if (need_resched() && __resched_legal(1)) { + spin_release(&lock->dep_map, 1, _THIS_IP_); + _raw_spin_unlock(lock); + preempt_enable_no_resched(); + __cond_resched(); + ret = 1; + spin_lock(lock); + } + return ret; +} +EXPORT_SYMBOL(cond_resched_lock); + +int __sched cond_resched_softirq(void) +{ + BUG_ON(!in_softirq()); + + 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. @@ -3099,7 +4730,6 @@ void __sched yield(void) set_current_state(TASK_RUNNING); sys_sched_yield(); } - EXPORT_SYMBOL(yield); /* @@ -3111,23 +4741,26 @@ EXPORT_SYMBOL(yield); */ void __sched io_schedule(void) { - struct runqueue *rq = this_rq(); + 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 = this_rq(); + 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; } @@ -3148,6 +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; } @@ -3171,6 +4805,7 @@ asmlinkage long sys_sched_get_priority_min(int policy) ret = 1; break; case SCHED_NORMAL: + case SCHED_BATCH: ret = 0; } return ret; @@ -3187,9 +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; @@ -3204,7 +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; @@ -3217,35 +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", "Z", "W" }; + 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 "); @@ -3259,10 +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); @@ -3289,7 +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" @@ -3311,31 +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 *idle_rq = cpu_rq(cpu), *rq = cpu_rq(task_cpu(idle)); + struct rq *rq = cpu_rq(cpu); unsigned long flags; - local_irq_save(flags); - double_rq_lock(idle_rq, rq); - - idle_rq->curr = idle_rq->idle = idle; - deactivate_task(idle, rq); + 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); - double_rq_unlock(idle_rq, rq); - set_tsk_need_resched(idle); - local_irq_restore(flags); + + spin_lock_irqsave(&rq->lock, flags); + rq->curr = rq->idle = 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! */ -#ifdef CONFIG_PREEMPT - idle->thread_info->preempt_count = (idle->lock_depth >= 0); +#if defined(CONFIG_PREEMPT) && !defined(CONFIG_PREEMPT_BKL) + 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 } @@ -3352,7 +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 @@ -3374,15 +5021,15 @@ 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 (any_online_cpu(new_mask) == NR_CPUS) { + if (!cpus_intersects(new_mask, cpu_online_map)) { ret = -EINVAL; goto out; } @@ -3397,32 +5044,36 @@ int set_cpus_allowed(task_t *p, cpumask_t new_mask) task_rq_unlock(rq, &flags); wake_up_process(rq->migration_thread); wait_for_completion(&req.done); + tlb_migrate_finish(p->mm); return 0; } out: task_rq_unlock(rq, &flags); + return ret; } - EXPORT_SYMBOL_GPL(set_cpus_allowed); /* * Move (not current) task off this cpu, onto dest cpu. We're doing * this because either it can't run here any more (set_cpus_allowed() * away from this CPU, or CPU going down), or because we're - * attempting to rebalance this task on exec (sched_balance_exec). + * attempting to rebalance this task on exec (sched_exec). * * 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_src = cpu_rq(src_cpu); rq_dest = cpu_rq(dest_cpu); double_rq_lock(rq_src, rq_dest); @@ -3444,13 +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; } /* @@ -3458,21 +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); @@ -3494,21 +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, smp_processor_id(), - 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); } @@ -3527,99 +5170,186 @@ wait_to_die: } #ifdef CONFIG_HOTPLUG_CPU -/* migrate_all_tasks - function to migrate all tasks from the dead cpu. */ -static void migrate_all_tasks(int src_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 *p) { - struct task_struct *tsk, *t; + unsigned long flags; + cpumask_t mask; + struct rq *rq; int dest_cpu; - unsigned int node; - write_lock_irq(&tasklist_lock); +restart: + /* On same node? */ + mask = node_to_cpumask(cpu_to_node(dead_cpu)); + cpus_and(mask, mask, p->cpus_allowed); + dest_cpu = any_online_cpu(mask); - /* watch out for per node tasks, let's stay on this node */ - node = cpu_to_node(src_cpu); + /* On any allowed CPU? */ + if (dest_cpu == NR_CPUS) + dest_cpu = any_online_cpu(p->cpus_allowed); - do_each_thread(t, tsk) { - cpumask_t mask; - if (tsk == current) - continue; + /* No more Mr. Nice Guy. */ + if (dest_cpu == NR_CPUS) { + rq = task_rq_lock(p, &flags); + cpus_setall(p->cpus_allowed); + dest_cpu = any_online_cpu(p->cpus_allowed); + task_rq_unlock(rq, &flags); - if (task_cpu(tsk) != src_cpu) - continue; + /* + * Don't tell them about moving exiting tasks or + * kernel threads (both mm NULL), since they never + * leave kernel. + */ + if (p->mm && printk_ratelimit()) + printk(KERN_INFO "process %d (%s) no " + "longer affine to cpu%d\n", + p->pid, p->comm, dead_cpu); + } + if (!__migrate_task(p, dead_cpu, dest_cpu)) + goto restart; +} - /* Figure out where this task should go (attempting to - * keep it on-node), and check if it can be migrated - * as-is. NOTE that kernel threads bound to more than - * one online cpu will be migrated. */ - mask = node_to_cpumask(node); - cpus_and(mask, mask, tsk->cpus_allowed); - dest_cpu = any_online_cpu(mask); - if (dest_cpu == NR_CPUS) - dest_cpu = any_online_cpu(tsk->cpus_allowed); - if (dest_cpu == NR_CPUS) { - cpus_clear(tsk->cpus_allowed); - cpus_complement(tsk->cpus_allowed); - dest_cpu = any_online_cpu(tsk->cpus_allowed); - - /* Don't tell them about moving exiting tasks - or kernel threads (both mm NULL), since - they never leave kernel. */ - if (tsk->mm && printk_ratelimit()) - printk(KERN_INFO "process %d (%s) no " - "longer affine to cpu%d\n", - tsk->pid, tsk->comm, src_cpu); - } +/* + * While a dead CPU has no uninterruptible tasks queued at this point, + * it might still have a nonzero ->nr_uninterruptible counter, because + * for performance reasons the counter is not stricly tracking tasks to + * 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(struct rq *rq_src) +{ + struct rq *rq_dest = cpu_rq(any_online_cpu(CPU_MASK_ALL)); + unsigned long flags; + + local_irq_save(flags); + double_rq_lock(rq_src, rq_dest); + rq_dest->nr_uninterruptible += rq_src->nr_uninterruptible; + rq_src->nr_uninterruptible = 0; + double_rq_unlock(rq_src, rq_dest); + local_irq_restore(flags); +} + +/* Run through task list and migrate tasks from the dead cpu. */ +static void migrate_live_tasks(int src_cpu) +{ + struct task_struct *p, *t; + + write_lock_irq(&tasklist_lock); + + do_each_thread(t, p) { + if (p == current) + continue; - __migrate_task(tsk, src_cpu, dest_cpu); - } 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 + * offline. + */ +void idle_task_exit(void) +{ + struct mm_struct *mm = current->active_mm; + + BUG_ON(cpu_online(smp_processor_id())); + + if (mm != &init_mm) + switch_mm(mm, &init_mm, current); + mmdrop(mm); +} + +static void migrate_dead(unsigned int dead_cpu, struct task_struct *p) +{ + struct rq *rq = cpu_rq(dead_cpu); + + /* Must be exiting, otherwise would be on tasklist. */ + BUG_ON(p->exit_state != EXIT_ZOMBIE && p->exit_state != EXIT_DEAD); + + /* Cannot have done final schedule yet: would have vanished. */ + BUG_ON(p->flags & PF_DEAD); + + get_task_struct(p); + + /* + * Drop lock around migration; if someone else moves it, + * that's OK. No task can be added to this CPU, so iteration is + * fine. + */ + spin_unlock_irq(&rq->lock); + move_task_off_dead_cpu(dead_cpu, p); + spin_lock_irq(&rq->lock); + + 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) +{ + 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, + struct task_struct, run_list)); + } + } +} #endif /* CONFIG_HOTPLUG_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: p = kthread_create(migration_thread, hcpu, "migration/%d",cpu); if (IS_ERR(p)) return NOTIFY_BAD; + p->flags |= PF_NOFREEZE; kthread_bind(p, cpu); /* Must be high prio: stop_machine expects to yield to it. */ rq = task_rq_lock(p, &flags); @@ -3627,19 +5357,25 @@ 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_all_tasks(cpu); + migrate_live_tasks(cpu); rq = cpu_rq(cpu); kthread_stop(rq->migration_thread); rq->migration_thread = NULL; @@ -3648,23 +5384,25 @@ static int migration_call(struct notifier_block *nfb, unsigned long action, deactivate_task(rq->idle, rq); rq->idle->static_prio = MAX_PRIO; __setscheduler(rq->idle, SCHED_NORMAL, 0); + migrate_dead_tasks(cpu); task_rq_unlock(rq, &flags); - BUG_ON(rq->nr_running != 0); + migrate_nr_uninterruptible(rq); + BUG_ON(rq->nr_running != 0); /* No need to migrate the tasks: it was best-effort if * they didn't do lock_cpu_hotplug(). Just wake up * 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); } spin_unlock_irq(&rq->lock); - break; + break; #endif } return NOTIFY_OK; @@ -3673,7 +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 }; @@ -3681,265 +5419,1481 @@ 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 -/* - * The 'big kernel lock' - * - * This spinlock is taken and released recursively by lock_kernel() - * and unlock_kernel(). It is transparently dropped and reaquired - * over schedule(). It is used to protect legacy code that hasn't - * been migrated to a proper locking design yet. - * - * Don't use in new code. - * - * Note: spinlock debugging needs this even on !CONFIG_SMP. - */ -spinlock_t kernel_flag __cacheline_aligned_in_smp = SPIN_LOCK_UNLOCKED; -EXPORT_SYMBOL(kernel_flag); - #ifdef CONFIG_SMP -/* Attach the domain 'sd' to 'cpu' as its base domain */ -void cpu_attach_domain(struct sched_domain *sd, int cpu) +#undef SCHED_DOMAIN_DEBUG +#ifdef SCHED_DOMAIN_DEBUG +static void sched_domain_debug(struct sched_domain *sd, int cpu) { - migration_req_t req; - unsigned long flags; - runqueue_t *rq = cpu_rq(cpu); - int local = 1; + int level = 0; - lock_cpu_hotplug(); - - spin_lock_irqsave(&rq->lock, flags); - - 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; + if (!sd) { + printk(KERN_DEBUG "CPU%d attaching NULL sched-domain.\n", cpu); + return; } - spin_unlock_irqrestore(&rq->lock, flags); + printk(KERN_DEBUG "CPU%d attaching sched-domain:\n", cpu); - if (!local) { - wake_up_process(rq->migration_thread); - wait_for_completion(&req.done); - } + do { + int i; + char str[NR_CPUS]; + struct sched_group *group = sd->groups; + cpumask_t groupmask; + + cpumask_scnprintf(str, NR_CPUS, sd->span); + cpus_clear(groupmask); + + printk(KERN_DEBUG); + for (i = 0; i < level + 1; i++) + printk(" "); + printk("domain %d: ", level); + + if (!(sd->flags & SD_LOAD_BALANCE)) { + printk("does not load-balance\n"); + if (sd->parent) + printk(KERN_ERR "ERROR: !SD_LOAD_BALANCE domain has parent"); + break; + } - unlock_cpu_hotplug(); -} + printk("span %s\n", str); -#ifdef ARCH_HAS_SCHED_DOMAIN -extern void __init arch_init_sched_domains(void); -#else -static struct sched_group sched_group_cpus[NR_CPUS]; -static DEFINE_PER_CPU(struct sched_domain, cpu_domains); -#ifdef CONFIG_NUMA -static struct sched_group sched_group_nodes[MAX_NUMNODES]; -static DEFINE_PER_CPU(struct sched_domain, node_domains); -static void __init arch_init_sched_domains(void) -{ - int i; - struct sched_group *first_node = NULL, *last_node = NULL; + if (!cpu_isset(cpu, sd->span)) + printk(KERN_ERR "ERROR: domain->span does not contain CPU%d\n", cpu); + if (!cpu_isset(cpu, group->cpumask)) + printk(KERN_ERR "ERROR: domain->groups does not contain CPU%d\n", cpu); - /* Set up domains */ - for_each_cpu(i) { - int node = cpu_to_node(i); - cpumask_t nodemask = node_to_cpumask(node); - struct sched_domain *node_sd = &per_cpu(node_domains, i); - struct sched_domain *cpu_sd = &per_cpu(cpu_domains, i); + printk(KERN_DEBUG); + for (i = 0; i < level + 2; i++) + printk(" "); + printk("groups:"); + do { + if (!group) { + printk("\n"); + printk(KERN_ERR "ERROR: group is NULL\n"); + break; + } - *node_sd = SD_NODE_INIT; - node_sd->span = cpu_possible_map; - node_sd->groups = &sched_group_nodes[cpu_to_node(i)]; + if (!group->cpu_power) { + printk("\n"); + printk(KERN_ERR "ERROR: domain->cpu_power not set\n"); + } - *cpu_sd = SD_CPU_INIT; - cpus_and(cpu_sd->span, nodemask, cpu_possible_map); - cpu_sd->groups = &sched_group_cpus[i]; - cpu_sd->parent = node_sd; - } + if (!cpus_weight(group->cpumask)) { + printk("\n"); + printk(KERN_ERR "ERROR: empty group\n"); + } - /* Set up groups */ - for (i = 0; i < MAX_NUMNODES; i++) { - cpumask_t tmp = node_to_cpumask(i); - cpumask_t nodemask; - struct sched_group *first_cpu = NULL, *last_cpu = NULL; - struct sched_group *node = &sched_group_nodes[i]; - int j; + if (cpus_intersects(groupmask, group->cpumask)) { + printk("\n"); + printk(KERN_ERR "ERROR: repeated CPUs\n"); + } - cpus_and(nodemask, tmp, cpu_possible_map); + cpus_or(groupmask, groupmask, group->cpumask); - if (cpus_empty(nodemask)) - continue; + cpumask_scnprintf(str, NR_CPUS, group->cpumask); + printk(" %s", str); - node->cpumask = nodemask; - node->cpu_power = SCHED_LOAD_SCALE * cpus_weight(node->cpumask); + group = group->next; + } while (group != sd->groups); + printk("\n"); - for_each_cpu_mask(j, node->cpumask) { - struct sched_group *cpu = &sched_group_cpus[j]; + if (!cpus_equal(sd->span, groupmask)) + printk(KERN_ERR "ERROR: groups don't span domain->span\n"); - cpus_clear(cpu->cpumask); - cpu_set(j, cpu->cpumask); - cpu->cpu_power = SCHED_LOAD_SCALE; + level++; + sd = sd->parent; - if (!first_cpu) - first_cpu = cpu; - if (last_cpu) - last_cpu->next = cpu; - last_cpu = cpu; + if (sd) { + if (!cpus_subset(groupmask, sd->span)) + printk(KERN_ERR "ERROR: parent span is not a superset of domain->span\n"); } - last_cpu->next = first_cpu; - - if (!first_node) - first_node = node; - if (last_node) - last_node->next = node; - last_node = node; - } - last_node->next = first_node; - mb(); - for_each_cpu(i) { - struct sched_domain *cpu_sd = &per_cpu(cpu_domains, i); - cpu_attach_domain(cpu_sd, i); - } + } while (sd); } +#else +# define sched_domain_debug(sd, cpu) do { } while (0) +#endif -#else /* !CONFIG_NUMA */ -static void __init arch_init_sched_domains(void) +static int sd_degenerate(struct sched_domain *sd) { - int i; - struct sched_group *first_cpu = NULL, *last_cpu = NULL; - - /* Set up domains */ - for_each_cpu(i) { - struct sched_domain *cpu_sd = &per_cpu(cpu_domains, i); + if (cpus_weight(sd->span) == 1) + return 1; - *cpu_sd = SD_CPU_INIT; - cpu_sd->span = cpu_possible_map; - cpu_sd->groups = &sched_group_cpus[i]; + /* 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; } - /* Set up CPU groups */ - for_each_cpu_mask(i, cpu_possible_map) { - struct sched_group *cpu = &sched_group_cpus[i]; + /* Following flags don't use groups */ + if (sd->flags & (SD_WAKE_IDLE | + SD_WAKE_AFFINE | + SD_WAKE_BALANCE)) + return 0; - cpus_clear(cpu->cpumask); - cpu_set(i, cpu->cpumask); - cpu->cpu_power = SCHED_LOAD_SCALE; + return 1; +} - if (!first_cpu) - first_cpu = cpu; - if (last_cpu) - last_cpu->next = cpu; - last_cpu = cpu; - } - last_cpu->next = first_cpu; +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. + */ +static void cpu_attach_domain(struct sched_domain *sd, int cpu) +{ + struct rq *rq = cpu_rq(cpu); + struct sched_domain *tmp; + + /* 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; + } + + if (sd && sd_degenerate(sd)) + sd = sd->parent; + + sched_domain_debug(sd, cpu); + + rcu_assign_pointer(rq->sd, sd); +} + +/* cpus with isolated domains */ +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) +{ + int ints[NR_CPUS], i; + + str = get_options(str, ARRAY_SIZE(ints), ints); + cpus_clear(cpu_isolated_map); + for (i = 1; i <= ints[0]; i++) + if (ints[i] < NR_CPUS) + cpu_set(ints[i], cpu_isolated_map); + return 1; +} + +__setup ("isolcpus=", isolated_cpu_setup); + +/* + * init_sched_build_groups takes an array of groups, the cpumask we wish + * to span, and a pointer to a function which identifies what group a CPU + * belongs to. The return value of group_fn must be a valid index into the + * groups[] array, and must be >= 0 and < NR_CPUS (due to the fact that we + * keep track of groups covered with a cpumask_t). + * + * init_sched_build_groups will build a circular linked list of the groups + * covered by the given span, and will set each group's ->cpumask correctly, + * and ->cpu_power to 0. + */ +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; + int i; + + for_each_cpu_mask(i, span) { + int group = group_fn(i); + struct sched_group *sg = &groups[group]; + int j; + + if (cpu_isset(i, covered)) + continue; + + sg->cpumask = CPU_MASK_NONE; + sg->cpu_power = 0; + + for_each_cpu_mask(j, span) { + if (group_fn(j) != group) + continue; + + cpu_set(j, covered); + cpu_set(j, sg->cpumask); + } + if (!first) + first = sg; + if (last) + last->next = sg; + last = sg; + } + 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 + +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 + -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) +{ + 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; +} + +__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) +{ + get_option(&str, &migration_factor); + migration_factor = migration_factor * MIGRATION_FACTOR_SCALE / 100; + return 1; +} + +__setup("migration_factor=", setup_migration_factor); + +/* + * 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) +{ + 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; +} + +__setup("migration_debug=", setup_migration_debug); + +/* + * 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). + */ +unsigned int max_cache_size; + +static int __init setup_max_cache_size(char *str) +{ + get_option(&str, &max_cache_size); + return 1; +} + +__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]++; + } + } +} + +/* + * Measure the cache-cost of one task migration. Returns in units of nsec. + */ +static unsigned long long +measure_one(void *cache, unsigned long size, int source, int target) +{ + cpumask_t mask, saved_mask; + unsigned long long t0, t1, t2, t3, cost; + + saved_mask = current->cpus_allowed; + + /* + * Flush source caches to RAM and invalidate them: + */ + sched_cacheflush(); + + /* + * Migrate to the source CPU: + */ + mask = cpumask_of_cpu(source); + set_cpus_allowed(current, mask); + WARN_ON(smp_processor_id() != source); + + /* + * Dirty the working set: + */ + t0 = sched_clock(); + touch_cache(cache, size); + t1 = sched_clock(); + + /* + * 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); + + t2 = sched_clock(); + touch_cache(cache, size); + t3 = sched_clock(); + + cost = t1-t0 + t3-t2; + + 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_cpus_allowed(current, saved_mask); + + return cost; +} + +/* + * 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; - mb(); /* domains were modified outside the lock */ - for_each_cpu(i) { - struct sched_domain *cpu_sd = &per_cpu(cpu_domains, i); - cpu_attach_domain(cpu_sd, i); + 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 } } -#endif /* CONFIG_NUMA */ -#endif /* ARCH_HAS_SCHED_DOMAIN */ - -#define SCHED_DOMAIN_DEBUG -#ifdef SCHED_DOMAIN_DEBUG -void sched_domain_debug(void) +/* + * 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; - for_each_cpu(i) { - runqueue_t *rq = cpu_rq(i); - struct sched_domain *sd; - int level = 0; + /* + * 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 - sd = rq->sd; + /* + * 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)); - printk(KERN_DEBUG "CPU%d: %s\n", - i, (cpu_online(i) ? " online" : "offline")); + cpus_and(nodemask, nodemask, *cpu_map); - do { - int j; - char str[NR_CPUS]; - struct sched_group *group = sd->groups; - cpumask_t groupmask, tmp; - - cpumask_scnprintf(str, NR_CPUS, sd->span); - cpus_clear(groupmask); - - printk(KERN_DEBUG); - for (j = 0; j < level + 1; j++) - printk(" "); - printk("domain %d: span %s\n", level, str); - - if (!cpu_isset(i, sd->span)) - printk(KERN_DEBUG "ERROR domain->span does not contain CPU%d\n", i); - if (!cpu_isset(i, group->cpumask)) - printk(KERN_DEBUG "ERROR domain->groups does not contain CPU%d\n", i); - if (!group->cpu_power) - printk(KERN_DEBUG "ERROR domain->cpu_power not set\n"); - - printk(KERN_DEBUG); - for (j = 0; j < level + 2; j++) - printk(" "); - printk("groups:"); - do { - if (!group) { - printk(" ERROR: NULL"); - break; +#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; - if (!cpus_weight(group->cpumask)) - printk(" ERROR empty group:"); + 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; + } - cpus_and(tmp, groupmask, group->cpumask); - if (cpus_weight(tmp) > 0) - printk(" ERROR repeated CPUs:"); + 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; + } - cpus_or(groupmask, groupmask, group->cpumask); + 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 - cpumask_scnprintf(str, NR_CPUS, group->cpumask); - printk(" %s", str); +#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 + } - group = group->next; - } while (group != sd->groups); - printk("\n"); +#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); + } - if (!cpus_equal(sd->span, groupmask)) - printk(KERN_DEBUG "ERROR groups don't span domain->span\n"); +#ifdef CONFIG_NUMA + /* Set up node groups */ + 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); - level++; - sd = sd->parent; + 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; - if (sd) { - cpus_and(tmp, groupmask, sd->span); - if (!cpus_equal(tmp, groupmask)) - printk(KERN_DEBUG "ERROR parent span is not a superset of domain->span\n"); + 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 - } while (sd); + /* Calculate CPU power for physical packages and nodes */ +#ifdef CONFIG_SCHED_SMT + for_each_cpu_mask(i, *cpu_map) { + struct sched_domain *sd; + sd = &per_cpu(cpu_domains, i); + 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); + 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 + 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_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 -#define sched_domain_debug() {} + 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; + + /* + * 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) +{ + free_sched_groups(cpu_map); +} + +/* + * 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); +} + +/* + * 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 + */ +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 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) +{ + switch (action) { + case CPU_UP_PREPARE: + case CPU_DOWN_PREPARE: + detach_destroy_domains(&cpu_online_map); + return NOTIFY_OK; + + case CPU_UP_CANCELED: + case CPU_DOWN_FAILED: + case CPU_ONLINE: + case CPU_DEAD: + /* + * Fall through and re-initialise the domains. + */ + break; + default: + return NOTIFY_DONE; + } + + /* The hotplug lock is already held by cpu_up/cpu_down */ + arch_init_sched_domains(&cpu_online_map); + + return NOTIFY_OK; +} #endif void __init sched_init_smp(void) { - arch_init_sched_domains(); - sched_domain_debug(); + lock_cpu_hotplug(); + 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); } #else void __init sched_init_smp(void) @@ -3951,52 +6905,42 @@ 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 addr >= (unsigned long)__sched_text_start - && addr < (unsigned long)__sched_text_end; + + return in_lock_functions(addr) || + (addr >= (unsigned long)__sched_text_start + && addr < (unsigned long)__sched_text_end); } void __init sched_init(void) { - runqueue_t *rq; int i, j, k; -#ifdef CONFIG_SMP - /* Set up an initial dummy domain for early boot */ - static struct sched_domain sched_domain_init; - static struct sched_group sched_group_init; - cpumask_t cpu_mask_all = CPU_MASK_ALL; - - memset(&sched_domain_init, 0, sizeof(struct sched_domain)); - sched_domain_init.span = cpu_mask_all; - sched_domain_init.groups = &sched_group_init; - sched_domain_init.last_balance = jiffies; - sched_domain_init.balance_interval = INT_MAX; /* Don't balance */ - - memset(&sched_group_init, 0, sizeof(struct sched_group)); - sched_group_init.cpumask = cpu_mask_all; - sched_group_init.next = &sched_group_init; - sched_group_init.cpu_power = SCHED_LOAD_SCALE; -#endif - - 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_init; - 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 - INIT_LIST_HEAD(&rq->hold_queue); atomic_set(&rq->nr_iowait, 0); +#ifdef CONFIG_VSERVER_HARDCPU + INIT_LIST_HEAD(&rq->hold_queue); +#endif for (j = 0; j < 2; j++) { array = rq->arrays + j; @@ -4008,35 +6952,40 @@ void __init sched_init(void) __set_bit(MAX_PRIO, array->bitmap); } } - /* - * We have to do a little magic to get the first - * thread right in SMP mode. - */ - rq = this_rq(); - rq->curr = current; - rq->idle = current; - set_task_cpu(current, smp_processor_id()); - wake_up_forked_process(current); + + 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: */ atomic_inc(&init_mm.mm_count); enter_lazy_tlb(&init_mm, current); + + /* + * Make us the idle thread. Technically, schedule() should not be + * called from this thread, however somewhere below it might be, + * but because we are the idle thread, we just pick up running again + * when this runqueue becomes "idle". + */ + init_idle(current, smp_processor_id()); } #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()) && - system_state == SYSTEM_RUNNING) { + system_state == SYSTEM_RUNNING && !oops_in_progress) { 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()); @@ -4047,48 +6996,80 @@ void __might_sleep(char *file, int line) EXPORT_SYMBOL(__might_sleep); #endif +#ifdef CONFIG_MAGIC_SYSRQ +void normalize_rt_tasks(void) +{ + struct prio_array *array; + struct task_struct *p; + unsigned long flags; + struct rq *rq; + + read_lock_irq(&tasklist_lock); + for_each_process(p) { + if (!rt_task(p)) + continue; + + spin_lock_irqsave(&p->pi_lock, flags); + rq = __task_rq_lock(p); + + array = p->array; + if (array) + deactivate_task(p, task_rq(p)); + __setscheduler(p, SCHED_NORMAL, 0); + if (array) { + vx_activate_task(p); + __activate_task(p, task_rq(p)); + resched_task(rq->curr); + } + + __task_rq_unlock(rq); + spin_unlock_irqrestore(&p->pi_lock, flags); + } + read_unlock_irq(&tasklist_lock); +} + +#endif /* CONFIG_MAGIC_SYSRQ */ -#if defined(CONFIG_SMP) && defined(CONFIG_PREEMPT) +#ifdef CONFIG_IA64 /* - * This could be a long-held lock. If another CPU holds it for a long time, - * and that CPU is not asked to reschedule then *this* CPU will spin on the - * lock for a long time, even if *this* CPU is asked to reschedule. + * These functions are only useful for the IA64 MCA handling. * - * So what we do here, in the slow (contended) path is to spin on the lock by - * hand while permitting preemption. + * 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. * - * Called inside preempt_disable(). + * ONLY VALID WHEN THE WHOLE SYSTEM IS STOPPED! */ -void __sched __preempt_spin_lock(spinlock_t *lock) +struct task_struct *curr_task(int cpu) { - if (preempt_count() > 1) { - _raw_spin_lock(lock); - return; - } - do { - preempt_enable(); - while (spin_is_locked(lock)) - cpu_relax(); - preempt_disable(); - } while (!_raw_spin_trylock(lock)); + return cpu_curr(cpu); } -EXPORT_SYMBOL(__preempt_spin_lock); - -void __sched __preempt_write_lock(rwlock_t *lock) +/** + * 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) { - if (preempt_count() > 1) { - _raw_write_lock(lock); - return; - } - - do { - preempt_enable(); - while (rwlock_is_locked(lock)) - cpu_relax(); - preempt_disable(); - } while (!_raw_write_trylock(lock)); + cpu_curr(cpu) = p; } -EXPORT_SYMBOL(__preempt_write_lock); -#endif /* defined(CONFIG_SMP) && defined(CONFIG_PREEMPT) */ +#endif