*/
unsigned long nr_running;
#ifdef CONFIG_SMP
- unsigned long prio_bias;
unsigned long cpu_load[3];
#endif
unsigned long long nr_switches;
* long it was waiting to run. We also note when it began so that we
* can keep stats on how long its timeslice is.
*/
-static inline void sched_info_arrive(task_t *t)
+static void sched_info_arrive(task_t *t)
{
unsigned long now = jiffies, diff = 0;
struct runqueue *rq = task_rq(t);
return prio;
}
-#ifdef CONFIG_SMP
-static inline void inc_prio_bias(runqueue_t *rq, int prio)
-{
- rq->prio_bias += MAX_PRIO - prio;
-}
-
-static inline void dec_prio_bias(runqueue_t *rq, int prio)
-{
- rq->prio_bias -= MAX_PRIO - prio;
-}
-
-static inline void inc_nr_running(task_t *p, runqueue_t *rq)
-{
- rq->nr_running++;
- if (rt_task(p)) {
- if (p != rq->migration_thread)
- /*
- * The migration thread does the actual balancing. Do
- * not bias by its priority as the ultra high priority
- * will skew balancing adversely.
- */
- inc_prio_bias(rq, p->prio);
- } else
- inc_prio_bias(rq, p->static_prio);
-}
-
-static inline void dec_nr_running(task_t *p, runqueue_t *rq)
-{
- rq->nr_running--;
- if (rt_task(p)) {
- if (p != rq->migration_thread)
- dec_prio_bias(rq, p->prio);
- } else
- dec_prio_bias(rq, p->static_prio);
-}
-#else
-static inline void inc_prio_bias(runqueue_t *rq, int prio)
-{
-}
-
-static inline void dec_prio_bias(runqueue_t *rq, int prio)
-{
-}
-
-static inline void inc_nr_running(task_t *p, runqueue_t *rq)
-{
- rq->nr_running++;
-}
-
-static inline void dec_nr_running(task_t *p, runqueue_t *rq)
-{
- rq->nr_running--;
-}
-#endif
-
/*
* __activate_task - move a task to the runqueue.
*/
static inline void __activate_task(task_t *p, runqueue_t *rq)
{
enqueue_task(p, rq->active);
- inc_nr_running(p, rq);
+ rq->nr_running++;
}
/*
static inline void __activate_idle_task(task_t *p, runqueue_t *rq)
{
enqueue_task_head(p, rq->active);
- inc_nr_running(p, rq);
+ rq->nr_running++;
}
static int recalc_task_prio(task_t *p, unsigned long long now)
unsigned long long __sleep_time = now - p->timestamp;
unsigned long sleep_time;
- if (__sleep_time > NS_MAX_SLEEP_AVG)
- sleep_time = NS_MAX_SLEEP_AVG;
- else
- sleep_time = (unsigned long)__sleep_time;
+ if (unlikely(p->policy == SCHED_BATCH))
+ sleep_time = 0;
+ else {
+ if (__sleep_time > NS_MAX_SLEEP_AVG)
+ sleep_time = NS_MAX_SLEEP_AVG;
+ else
+ sleep_time = (unsigned long)__sleep_time;
+ }
if (likely(sleep_time > 0)) {
/*
*/
static void deactivate_task(struct task_struct *p, runqueue_t *rq)
{
- dec_nr_running(p, rq);
+ rq->nr_running--;
dequeue_task(p, p->array);
p->array = NULL;
}
* We want to under-estimate the load of migration sources, to
* balance conservatively.
*/
-static inline unsigned long __source_load(int cpu, int type, enum idle_type idle)
+static inline unsigned long source_load(int cpu, int type)
{
runqueue_t *rq = cpu_rq(cpu);
- unsigned long running = rq->nr_running;
- unsigned long source_load, cpu_load = rq->cpu_load[type-1],
- load_now = running * SCHED_LOAD_SCALE;
-
+ unsigned long load_now = rq->nr_running * SCHED_LOAD_SCALE;
if (type == 0)
- source_load = load_now;
- else
- source_load = min(cpu_load, load_now);
-
- if (running > 1 || (idle == NOT_IDLE && running))
- /*
- * If we are busy rebalancing the load is biased by
- * priority to create 'nice' support across cpus. When
- * idle rebalancing we should only bias the source_load if
- * there is more than one task running on that queue to
- * prevent idle rebalance from trying to pull tasks from a
- * queue with only one running task.
- */
- source_load = source_load * rq->prio_bias / running;
+ return load_now;
- return source_load;
-}
-
-static inline unsigned long source_load(int cpu, int type)
-{
- return __source_load(cpu, type, NOT_IDLE);
+ return min(rq->cpu_load[type-1], load_now);
}
/*
* Return a high guess at the load of a migration-target cpu
*/
-static inline unsigned long __target_load(int cpu, int type, enum idle_type idle)
+static inline unsigned long target_load(int cpu, int type)
{
runqueue_t *rq = cpu_rq(cpu);
- unsigned long running = rq->nr_running;
- unsigned long target_load, cpu_load = rq->cpu_load[type-1],
- load_now = running * SCHED_LOAD_SCALE;
-
+ unsigned long load_now = rq->nr_running * SCHED_LOAD_SCALE;
if (type == 0)
- target_load = load_now;
- else
- target_load = max(cpu_load, load_now);
+ return load_now;
- if (running > 1 || (idle == NOT_IDLE && running))
- target_load = target_load * rq->prio_bias / running;
-
- return target_load;
-}
-
-static inline unsigned long target_load(int cpu, int type)
-{
- return __target_load(cpu, type, NOT_IDLE);
+ return max(rq->cpu_load[type-1], load_now);
}
/*
}
}
- if (p->last_waker_cpu != this_cpu)
- goto out_set_cpu;
-
if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed)))
goto out_set_cpu;
cpu = task_cpu(p);
}
- p->last_waker_cpu = this_cpu;
-
out_activate:
#endif /* CONFIG_SMP */
if (old_state == TASK_UNINTERRUPTIBLE) {
#ifdef CONFIG_SCHEDSTATS
memset(&p->sched_info, 0, sizeof(p->sched_info));
#endif
-#if defined(CONFIG_SMP)
- p->last_waker_cpu = cpu;
-#if defined(__ARCH_WANT_UNLOCKED_CTXSW)
+#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW)
p->oncpu = 0;
#endif
-#endif
#ifdef CONFIG_PREEMPT
/* Want to start with kernel preemption disabled. */
task_thread_info(p)->preempt_count = 1;
list_add_tail(&p->run_list, ¤t->run_list);
p->array = current->array;
p->array->nr_active++;
- inc_nr_running(p, rq);
+ rq->nr_running++;
}
set_need_resched();
} else
* pull_task - move a task from a remote runqueue to the local runqueue.
* Both runqueues must be locked.
*/
-static inline
+static
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)
{
dequeue_task(p, src_array);
- dec_nr_running(p, src_rq);
+ src_rq->nr_running--;
set_task_cpu(p, this_cpu);
- inc_nr_running(p, this_rq);
+ this_rq->nr_running++;
enqueue_task(p, this_array);
p->timestamp = (p->timestamp - src_rq->timestamp_last_tick)
+ this_rq->timestamp_last_tick;
/*
* can_migrate_task - may task p from runqueue rq be migrated to this_cpu?
*/
-static inline
+static
int can_migrate_task(task_t *p, runqueue_t *rq, int this_cpu,
struct sched_domain *sd, enum idle_type idle,
int *all_pinned)
/* Bias balancing toward cpus of our domain */
if (local_group)
- load = __target_load(i, load_idx, idle);
+ load = target_load(i, load_idx);
else
- load = __source_load(i, load_idx, idle);
+ load = source_load(i, load_idx);
avg_load += load;
}
int i;
for_each_cpu_mask(i, group->cpumask) {
- load = __source_load(i, 0, idle);
+ load = source_load(i, 0);
if (load > max_load) {
max_load = load;
* 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, runqueue_t *this_rq)
{
struct sched_domain *sd;
resched_task(rq->idle);
}
-static inline void wake_sleeping_dependent(int this_cpu, runqueue_t *this_rq)
+static void wake_sleeping_dependent(int this_cpu, runqueue_t *this_rq)
{
struct sched_domain *tmp, *sd = NULL;
cpumask_t sibling_map;
return p->time_slice * (100 - sd->per_cpu_gain) / 100;
}
-static inline int dependent_sleeper(int this_cpu, runqueue_t *this_rq)
+static int dependent_sleeper(int this_cpu, runqueue_t *this_rq)
{
struct sched_domain *tmp, *sd = NULL;
cpumask_t sibling_map;
* 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)) {
p->static_prio = NICE_TO_PRIO(nice);
goto out_unlock;
}
array = p->array;
- if (array) {
+ if (array)
dequeue_task(p, array);
- dec_prio_bias(rq, p->static_prio);
- }
old_prio = p->prio;
new_prio = NICE_TO_PRIO(nice);
if (array) {
enqueue_task(p, array);
- inc_prio_bias(rq, p->static_prio);
/*
* If the task increased its priority or is running and
* lowered its priority, then reschedule its CPU:
BUG_ON(p->array);
p->policy = policy;
p->rt_priority = prio;
- if (policy != SCHED_NORMAL)
+ if (policy != SCHED_NORMAL && policy != SCHED_BATCH) {
p->prio = MAX_RT_PRIO-1 - p->rt_priority;
- else
+ } else {
p->prio = p->static_prio;
+ /*
+ * SCHED_BATCH tasks are treated as perpetual CPU hogs:
+ */
+ if (policy == SCHED_BATCH)
+ p->sleep_avg = 0;
+ }
}
/**
if (policy < 0)
policy = oldpolicy = p->policy;
else if (policy != SCHED_FIFO && policy != SCHED_RR &&
- policy != SCHED_NORMAL)
- return -EINVAL;
+ policy != SCHED_NORMAL && policy != SCHED_BATCH)
+ return -EINVAL;
/*
* Valid priorities for SCHED_FIFO and SCHED_RR are
- * 1..MAX_USER_RT_PRIO-1, valid priority for SCHED_NORMAL is 0.
+ * 1..MAX_USER_RT_PRIO-1, valid priority for SCHED_NORMAL and
+ * SCHED_BATCH is 0.
*/
if (param->sched_priority < 0 ||
(p->mm && param->sched_priority > MAX_USER_RT_PRIO-1) ||
(!p->mm && param->sched_priority > MAX_RT_PRIO-1))
return -EINVAL;
- if ((policy == SCHED_NORMAL) != (param->sched_priority == 0))
+ if ((policy == SCHED_NORMAL || policy == SCHED_BATCH)
+ != (param->sched_priority == 0))
return -EINVAL;
/*
* Allow unprivileged RT tasks to decrease priority:
*/
if (!capable(CAP_SYS_NICE)) {
- /* can't change policy */
- if (policy != p->policy &&
- !p->signal->rlim[RLIMIT_RTPRIO].rlim_cur)
+ /*
+ * 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 &&
+ if ((policy != SCHED_NORMAL && policy != SCHED_BATCH) &&
param->sched_priority > p->rt_priority &&
param->sched_priority >
p->signal->rlim[RLIMIT_RTPRIO].rlim_cur)
asmlinkage long sys_sched_setscheduler(pid_t pid, int policy,
struct sched_param __user *param)
{
+ /* negative values for policy are not valid */
+ if (policy < 0)
+ return -EINVAL;
+
return do_sched_setscheduler(pid, policy, param);
}
goto out_unlock;
retval = 0;
- cpus_and(*mask, p->cpus_allowed, cpu_possible_map);
+ cpus_and(*mask, p->cpus_allowed, cpu_online_map);
out_unlock:
read_unlock(&tasklist_lock);
ret = MAX_USER_RT_PRIO-1;
break;
case SCHED_NORMAL:
+ case SCHED_BATCH:
ret = 0;
break;
}
ret = 1;
break;
case SCHED_NORMAL:
+ case SCHED_BATCH:
ret = 0;
}
return ret;
#define SEARCH_SCOPE 2
#define MIN_CACHE_SIZE (64*1024U)
#define DEFAULT_CACHE_SIZE (5*1024*1024U)
-#define ITERATIONS 2
+#define ITERATIONS 1
#define SIZE_THRESH 130
#define COST_THRESH 130
break;
}
/*
- * Increase the cachesize in 5% steps:
+ * Increase the cachesize in 10% steps:
*/
- size = size * 20 / 19;
+ size = size * 10 / 9;
}
if (migration_debug)
-1
#endif
);
- printk("migration_cost=");
- for (distance = 0; distance <= max_distance; distance++) {
- if (distance)
- printk(",");
- printk("%ld", (long)migration_cost[distance] / 1000);
+ if (system_state == SYSTEM_BOOTING) {
+ printk("migration_cost=");
+ for (distance = 0; distance <= max_distance; distance++) {
+ if (distance)
+ printk(",");
+ printk("%ld", (long)migration_cost[distance] / 1000);
+ }
+ printk("\n");
}
- printk("\n");
j1 = jiffies;
if (migration_debug)
printk("migration: %ld seconds\n", (j1-j0)/HZ);
* Detach sched domains from a group of cpus specified in cpu_map
* These cpus will now be attached to the NULL domain
*/
-static inline void detach_destroy_domains(const cpumask_t *cpu_map)
+static void detach_destroy_domains(const cpumask_t *cpu_map)
{
int i;
runqueue_t *rq;
int i, j, k;
- for (i = 0; i < NR_CPUS; i++) {
+ for_each_cpu(i) {
prio_array_t *array;
rq = cpu_rq(i);