sched: rt-group: synchonised bandwidth period
authorPeter Zijlstra <a.p.zijlstra@chello.nl>
Sat, 19 Apr 2008 17:44:57 +0000 (19:44 +0200)
committerIngo Molnar <mingo@elte.hu>
Sat, 19 Apr 2008 17:44:57 +0000 (19:44 +0200)
Various SMP balancing algorithms require that the bandwidth period
run in sync.

Possible improvements are moving the rt_bandwidth thing into root_domain
and keeping a span per rt_bandwidth which marks throttled cpus.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
include/linux/sched.h
kernel/sched.c
kernel/sched_rt.c
kernel/sysctl.c
kernel/time/tick-sched.c
kernel/user.c

index 15f05ff453d848578e2c5d92978df06d1a835345..be5d31752dbd82ff6c6083c45ea97feefacf0afe 100644 (file)
@@ -1563,6 +1563,10 @@ int sched_nr_latency_handler(struct ctl_table *table, int write,
 extern unsigned int sysctl_sched_rt_period;
 extern int sysctl_sched_rt_runtime;
 
+int sched_rt_handler(struct ctl_table *table, int write,
+               struct file *filp, void __user *buffer, size_t *lenp,
+               loff_t *ppos);
+
 extern unsigned int sysctl_sched_compat_yield;
 
 #ifdef CONFIG_RT_MUTEXES
@@ -2052,6 +2056,9 @@ extern unsigned long sched_group_shares(struct task_group *tg);
 extern int sched_group_set_rt_runtime(struct task_group *tg,
                                      long rt_runtime_us);
 extern long sched_group_rt_runtime(struct task_group *tg);
+extern int sched_group_set_rt_period(struct task_group *tg,
+                                     long rt_period_us);
+extern long sched_group_rt_period(struct task_group *tg);
 #endif
 #endif
 
index e813e845d9cfeb46cc1fd1a1eb4f69e48b61a8d4..bb20323f7d09f4a8a28649dcb26c743242407e38 100644 (file)
@@ -115,6 +115,11 @@ unsigned long long __attribute__((weak)) sched_clock(void)
  */
 #define DEF_TIMESLICE          (100 * HZ / 1000)
 
+/*
+ * single value that denotes runtime == period, ie unlimited time.
+ */
+#define RUNTIME_INF    ((u64)~0ULL)
+
 #ifdef CONFIG_SMP
 /*
  * Divide a load by a sched group cpu_power : (load / sg->__cpu_power)
@@ -156,6 +161,80 @@ struct rt_prio_array {
        struct list_head queue[MAX_RT_PRIO];
 };
 
+struct rt_bandwidth {
+       ktime_t rt_period;
+       u64 rt_runtime;
+       struct hrtimer rt_period_timer;
+};
+
+static struct rt_bandwidth def_rt_bandwidth;
+
+static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun);
+
+static enum hrtimer_restart sched_rt_period_timer(struct hrtimer *timer)
+{
+       struct rt_bandwidth *rt_b =
+               container_of(timer, struct rt_bandwidth, rt_period_timer);
+       ktime_t now;
+       int overrun;
+       int idle = 0;
+
+       for (;;) {
+               now = hrtimer_cb_get_time(timer);
+               overrun = hrtimer_forward(timer, now, rt_b->rt_period);
+
+               if (!overrun)
+                       break;
+
+               idle = do_sched_rt_period_timer(rt_b, overrun);
+       }
+
+       return idle ? HRTIMER_NORESTART : HRTIMER_RESTART;
+}
+
+static
+void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime)
+{
+       rt_b->rt_period = ns_to_ktime(period);
+       rt_b->rt_runtime = runtime;
+
+       hrtimer_init(&rt_b->rt_period_timer,
+                       CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+       rt_b->rt_period_timer.function = sched_rt_period_timer;
+       rt_b->rt_period_timer.cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ;
+}
+
+static void start_rt_bandwidth(struct rt_bandwidth *rt_b)
+{
+       ktime_t now;
+
+       if (rt_b->rt_runtime == RUNTIME_INF)
+               return;
+
+       if (hrtimer_active(&rt_b->rt_period_timer))
+               return;
+
+       spin_lock(&rt_b->rt_runtime_lock);
+       for (;;) {
+               if (hrtimer_active(&rt_b->rt_period_timer))
+                       break;
+
+               now = hrtimer_cb_get_time(&rt_b->rt_period_timer);
+               hrtimer_forward(&rt_b->rt_period_timer, now, rt_b->rt_period);
+               hrtimer_start(&rt_b->rt_period_timer,
+                             rt_b->rt_period_timer.expires,
+                             HRTIMER_MODE_ABS);
+       }
+       spin_unlock(&rt_b->rt_runtime_lock);
+}
+
+#ifdef CONFIG_RT_GROUP_SCHED
+static void destroy_rt_bandwidth(struct rt_bandwidth *rt_b)
+{
+       hrtimer_cancel(&rt_b->rt_period_timer);
+}
+#endif
+
 #ifdef CONFIG_GROUP_SCHED
 
 #include <linux/cgroup.h>
@@ -182,7 +261,7 @@ struct task_group {
        struct sched_rt_entity **rt_se;
        struct rt_rq **rt_rq;
 
-       u64 rt_runtime;
+       struct rt_bandwidth rt_bandwidth;
 #endif
 
        struct rcu_head rcu;
@@ -407,8 +486,6 @@ struct rq {
 
        struct cfs_rq cfs;
        struct rt_rq rt;
-       u64 rt_period_expire;
-       int rt_throttled;
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
        /* list of leaf cfs_rq on this cpu: */
@@ -592,23 +669,6 @@ static void update_rq_clock(struct rq *rq)
 #define task_rq(p)             cpu_rq(task_cpu(p))
 #define cpu_curr(cpu)          (cpu_rq(cpu)->curr)
 
-unsigned long rt_needs_cpu(int cpu)
-{
-       struct rq *rq = cpu_rq(cpu);
-       u64 delta;
-
-       if (!rq->rt_throttled)
-               return 0;
-
-       if (rq->clock > rq->rt_period_expire)
-               return 1;
-
-       delta = rq->rt_period_expire - rq->clock;
-       do_div(delta, NSEC_PER_SEC / HZ);
-
-       return (unsigned long)delta;
-}
-
 /*
  * Tunables that become constants when CONFIG_SCHED_DEBUG is off:
  */
@@ -664,10 +724,18 @@ static __read_mostly int scheduler_running;
  */
 int sysctl_sched_rt_runtime = 950000;
 
-/*
- * single value that denotes runtime == period, ie unlimited time.
- */
-#define RUNTIME_INF    ((u64)~0ULL)
+static inline u64 global_rt_period(void)
+{
+       return (u64)sysctl_sched_rt_period * NSEC_PER_USEC;
+}
+
+static inline u64 global_rt_runtime(void)
+{
+       if (sysctl_sched_rt_period < 0)
+               return RUNTIME_INF;
+
+       return (u64)sysctl_sched_rt_runtime * NSEC_PER_USEC;
+}
 
 static const unsigned long long time_sync_thresh = 100000;
 
@@ -3854,7 +3922,6 @@ void scheduler_tick(void)
        update_last_tick_seen(rq);
        update_cpu_load(rq);
        curr->sched_class->task_tick(rq, curr, 0);
-       update_sched_rt_period(rq);
        spin_unlock(&rq->lock);
 
 #ifdef CONFIG_SMP
@@ -4689,7 +4756,7 @@ recheck:
         * Do not allow realtime tasks into groups that have no runtime
         * assigned.
         */
-       if (rt_policy(policy) && task_group(p)->rt_runtime == 0)
+       if (rt_policy(policy) && task_group(p)->rt_bandwidth.rt_runtime == 0)
                return -EPERM;
 #endif
 
@@ -7288,6 +7355,14 @@ void __init sched_init(void)
        init_defrootdomain();
 #endif
 
+       init_rt_bandwidth(&def_rt_bandwidth,
+                       global_rt_period(), global_rt_runtime());
+
+#ifdef CONFIG_RT_GROUP_SCHED
+       init_rt_bandwidth(&init_task_group.rt_bandwidth,
+                       global_rt_period(), global_rt_runtime());
+#endif
+
 #ifdef CONFIG_GROUP_SCHED
        list_add(&init_task_group.list, &task_groups);
 #endif
@@ -7312,15 +7387,11 @@ void __init sched_init(void)
 
 #endif
 #ifdef CONFIG_RT_GROUP_SCHED
-               init_task_group.rt_runtime =
-                       sysctl_sched_rt_runtime * NSEC_PER_USEC;
                INIT_LIST_HEAD(&rq->leaf_rt_rq_list);
                init_tg_rt_entry(rq, &init_task_group,
                                &per_cpu(init_rt_rq, i),
                                &per_cpu(init_sched_rt_entity, i), i, 1);
 #endif
-               rq->rt_period_expire = 0;
-               rq->rt_throttled = 0;
 
                for (j = 0; j < CPU_LOAD_IDX_MAX; j++)
                        rq->cpu_load[j] = 0;
@@ -7506,8 +7577,6 @@ void set_curr_task(int cpu, struct task_struct *p)
 
 #endif
 
-#ifdef CONFIG_GROUP_SCHED
-
 #ifdef CONFIG_FAIR_GROUP_SCHED
 static void free_fair_sched_group(struct task_group *tg)
 {
@@ -7596,6 +7665,8 @@ static void free_rt_sched_group(struct task_group *tg)
 {
        int i;
 
+       destroy_rt_bandwidth(&tg->rt_bandwidth);
+
        for_each_possible_cpu(i) {
                if (tg->rt_rq)
                        kfree(tg->rt_rq[i]);
@@ -7621,7 +7692,8 @@ static int alloc_rt_sched_group(struct task_group *tg)
        if (!tg->rt_se)
                goto err;
 
-       tg->rt_runtime = 0;
+       init_rt_bandwidth(&tg->rt_bandwidth,
+                       ktime_to_ns(def_rt_bandwidth.rt_period), 0);
 
        for_each_possible_cpu(i) {
                rq = cpu_rq(i);
@@ -7674,6 +7746,7 @@ static inline void unregister_rt_sched_group(struct task_group *tg, int cpu)
 }
 #endif
 
+#ifdef CONFIG_GROUP_SCHED
 static void free_sched_group(struct task_group *tg)
 {
        free_fair_sched_group(tg);
@@ -7775,6 +7848,7 @@ void sched_move_task(struct task_struct *tsk)
 
        task_rq_unlock(rq, &flags);
 }
+#endif
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
 static void set_se_shares(struct sched_entity *se, unsigned long shares)
@@ -7871,16 +7945,15 @@ static int __rt_schedulable(struct task_group *tg, u64 period, u64 runtime)
        struct task_group *tgi;
        unsigned long total = 0;
        unsigned long global_ratio =
-               to_ratio(sysctl_sched_rt_period,
-                        sysctl_sched_rt_runtime < 0 ?
-                               RUNTIME_INF : sysctl_sched_rt_runtime);
+               to_ratio(global_rt_period(), global_rt_runtime());
 
        rcu_read_lock();
        list_for_each_entry_rcu(tgi, &task_groups, list) {
                if (tgi == tg)
                        continue;
 
-               total += to_ratio(period, tgi->rt_runtime);
+               total += to_ratio(ktime_to_ns(tgi->rt_bandwidth.rt_period),
+                               tgi->rt_bandwidth.rt_runtime);
        }
        rcu_read_unlock();
 
@@ -7898,16 +7971,11 @@ static inline int tg_has_rt_tasks(struct task_group *tg)
        return 0;
 }
 
-int sched_group_set_rt_runtime(struct task_group *tg, long rt_runtime_us)
+static int tg_set_bandwidth(struct task_group *tg,
+               u64 rt_period, u64 rt_runtime)
 {
-       u64 rt_runtime, rt_period;
        int err = 0;
 
-       rt_period = (u64)sysctl_sched_rt_period * NSEC_PER_USEC;
-       rt_runtime = (u64)rt_runtime_us * NSEC_PER_USEC;
-       if (rt_runtime_us == -1)
-               rt_runtime = RUNTIME_INF;
-
        mutex_lock(&rt_constraints_mutex);
        read_lock(&tasklist_lock);
        if (rt_runtime_us == 0 && tg_has_rt_tasks(tg)) {
@@ -7918,7 +7986,8 @@ int sched_group_set_rt_runtime(struct task_group *tg, long rt_runtime_us)
                err = -EINVAL;
                goto unlock;
        }
-       tg->rt_runtime = rt_runtime;
+       tg->rt_bandwidth.rt_period = ns_to_ktime(rt_period);
+       tg->rt_bandwidth.rt_runtime = rt_runtime;
  unlock:
        read_unlock(&tasklist_lock);
        mutex_unlock(&rt_constraints_mutex);
@@ -7926,19 +7995,96 @@ int sched_group_set_rt_runtime(struct task_group *tg, long rt_runtime_us)
        return err;
 }
 
+int sched_group_set_rt_runtime(struct task_group *tg, long rt_runtime_us)
+{
+       u64 rt_runtime, rt_period;
+
+       rt_period = ktime_to_ns(tg->rt_bandwidth.rt_period);
+       rt_runtime = (u64)rt_runtime_us * NSEC_PER_USEC;
+       if (rt_runtime_us < 0)
+               rt_runtime = RUNTIME_INF;
+
+       return tg_set_bandwidth(tg, rt_period, rt_runtime);
+}
+
 long sched_group_rt_runtime(struct task_group *tg)
 {
        u64 rt_runtime_us;
 
-       if (tg->rt_runtime == RUNTIME_INF)
+       if (tg->rt_bandwidth.rt_runtime == RUNTIME_INF)
                return -1;
 
-       rt_runtime_us = tg->rt_runtime;
+       rt_runtime_us = tg->rt_bandwidth.rt_runtime;
        do_div(rt_runtime_us, NSEC_PER_USEC);
        return rt_runtime_us;
 }
+
+int sched_group_set_rt_period(struct task_group *tg, long rt_period_us)
+{
+       u64 rt_runtime, rt_period;
+
+       rt_period = (u64)rt_period_us * NSEC_PER_USEC;
+       rt_runtime = tg->rt_bandwidth.rt_runtime;
+
+       return tg_set_bandwidth(tg, rt_period, rt_runtime);
+}
+
+long sched_group_rt_period(struct task_group *tg)
+{
+       u64 rt_period_us;
+
+       rt_period_us = ktime_to_ns(tg->rt_bandwidth.rt_period);
+       do_div(rt_period_us, NSEC_PER_USEC);
+       return rt_period_us;
+}
+
+static int sched_rt_global_constraints(void)
+{
+       int ret = 0;
+
+       mutex_lock(&rt_constraints_mutex);
+       if (!__rt_schedulable(NULL, 1, 0))
+               ret = -EINVAL;
+       mutex_unlock(&rt_constraints_mutex);
+
+       return ret;
+}
+#else
+static int sched_rt_global_constraints(void)
+{
+       return 0;
+}
 #endif
-#endif /* CONFIG_GROUP_SCHED */
+
+int sched_rt_handler(struct ctl_table *table, int write,
+               struct file *filp, void __user *buffer, size_t *lenp,
+               loff_t *ppos)
+{
+       int ret;
+       int old_period, old_runtime;
+       static DEFINE_MUTEX(mutex);
+
+       mutex_lock(&mutex);
+       old_period = sysctl_sched_rt_period;
+       old_runtime = sysctl_sched_rt_runtime;
+
+       ret = proc_dointvec(table, write, filp, buffer, lenp, ppos);
+
+       if (!ret && write) {
+               ret = sched_rt_global_constraints();
+               if (ret) {
+                       sysctl_sched_rt_period = old_period;
+                       sysctl_sched_rt_runtime = old_runtime;
+               } else {
+                       def_rt_bandwidth.rt_runtime = global_rt_runtime();
+                       def_rt_bandwidth.rt_period =
+                               ns_to_ktime(global_rt_period());
+               }
+       }
+       mutex_unlock(&mutex);
+
+       return ret;
+}
 
 #ifdef CONFIG_CGROUP_SCHED
 
@@ -7988,7 +8134,7 @@ cpu_cgroup_can_attach(struct cgroup_subsys *ss, struct cgroup *cgrp,
 {
 #ifdef CONFIG_RT_GROUP_SCHED
        /* Don't accept realtime tasks when there is no way for them to run */
-       if (rt_task(tsk) && cgroup_tg(cgrp)->rt_runtime == 0)
+       if (rt_task(tsk) && cgroup_tg(cgrp)->rt_bandwidth.rt_runtime == 0)
                return -EINVAL;
 #else
        /* We don't support RT-tasks being in separate groups */
@@ -8066,6 +8212,17 @@ static ssize_t cpu_rt_runtime_read(struct cgroup *cgrp, struct cftype *cft,
 
        return simple_read_from_buffer(buf, nbytes, ppos, tmp, len);
 }
+
+static int cpu_rt_period_write_uint(struct cgroup *cgrp, struct cftype *cftype,
+               u64 rt_period_us)
+{
+       return sched_group_set_rt_period(cgroup_tg(cgrp), rt_period_us);
+}
+
+static u64 cpu_rt_period_read_uint(struct cgroup *cgrp, struct cftype *cft)
+{
+       return sched_group_rt_period(cgroup_tg(cgrp));
+}
 #endif
 
 static struct cftype cpu_files[] = {
@@ -8082,6 +8239,11 @@ static struct cftype cpu_files[] = {
                .read = cpu_rt_runtime_read,
                .write = cpu_rt_runtime_write,
        },
+       {
+               .name = "rt_period_us",
+               .read_uint = cpu_rt_period_read_uint,
+               .write_uint = cpu_rt_period_write_uint,
+       },
 #endif
 };
 
index 0a6d2e516420516cb1d0c35a5345db1f356f71ab..8bc176136666fe4a9e8c60d01976e8646c7af1e5 100644 (file)
@@ -62,7 +62,7 @@ static inline u64 sched_rt_runtime(struct rt_rq *rt_rq)
        if (!rt_rq->tg)
                return RUNTIME_INF;
 
-       return rt_rq->tg->rt_runtime;
+       return rt_rq->tg->rt_bandwidth.rt_runtime;
 }
 
 #define for_each_leaf_rt_rq(rt_rq, rq) \
@@ -127,14 +127,29 @@ static int rt_se_boosted(struct sched_rt_entity *rt_se)
        return p->prio != p->normal_prio;
 }
 
+#ifdef CONFIG_SMP
+static inline cpumask_t sched_rt_period_mask(void)
+{
+       return cpu_rq(smp_processor_id())->rd->span;
+}
 #else
+static inline cpumask_t sched_rt_period_mask(void)
+{
+       return cpu_online_map;
+}
+#endif
 
-static inline u64 sched_rt_runtime(struct rt_rq *rt_rq)
+static inline
+struct rt_rq *sched_rt_period_rt_rq(struct rt_bandwidth *rt_b, int cpu)
 {
-       if (sysctl_sched_rt_runtime == -1)
-               return RUNTIME_INF;
+       return container_of(rt_b, struct task_group, rt_bandwidth)->rt_rq[cpu];
+}
 
-       return (u64)sysctl_sched_rt_runtime * NSEC_PER_USEC;
+#else
+
+static inline u64 sched_rt_runtime(struct rt_rq *rt_rq)
+{
+       return def_rt_bandwidth.rt_runtime;
 }
 
 #define for_each_leaf_rt_rq(rt_rq, rq) \
@@ -173,8 +188,55 @@ static inline int rt_rq_throttled(struct rt_rq *rt_rq)
 {
        return rt_rq->rt_throttled;
 }
+
+static inline cpumask_t sched_rt_period_mask(void)
+{
+       return cpu_online_map;
+}
+
+static inline
+struct rt_rq *sched_rt_period_rt_rq(struct rt_bandwidth *rt_b, int cpu)
+{
+       return &cpu_rq(cpu)->rt;
+}
+
 #endif
 
+static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
+{
+       int i, idle = 1;
+       cpumask_t span;
+
+       if (rt_b->rt_runtime == RUNTIME_INF)
+               return 1;
+
+       span = sched_rt_period_mask();
+       for_each_cpu_mask(i, span) {
+               int enqueue = 0;
+               struct rt_rq *rt_rq = sched_rt_period_rt_rq(rt_b, i);
+               struct rq *rq = rq_of_rt_rq(rt_rq);
+
+               spin_lock(&rq->lock);
+               if (rt_rq->rt_time) {
+                       u64 runtime = rt_b->rt_runtime;
+
+                       rt_rq->rt_time -= min(rt_rq->rt_time, overrun*runtime);
+                       if (rt_rq->rt_throttled && rt_rq->rt_time < runtime) {
+                               rt_rq->rt_throttled = 0;
+                               enqueue = 1;
+                       }
+                       if (rt_rq->rt_time || rt_rq->rt_nr_running)
+                               idle = 0;
+               }
+
+               if (enqueue)
+                       sched_rt_rq_enqueue(rt_rq);
+               spin_unlock(&rq->lock);
+       }
+
+       return idle;
+}
+
 static inline int rt_se_prio(struct sched_rt_entity *rt_se)
 {
 #ifdef CONFIG_RT_GROUP_SCHED
@@ -198,11 +260,7 @@ static int sched_rt_runtime_exceeded(struct rt_rq *rt_rq)
                return rt_rq_throttled(rt_rq);
 
        if (rt_rq->rt_time > runtime) {
-               struct rq *rq = rq_of_rt_rq(rt_rq);
-
-               rq->rt_throttled = 1;
                rt_rq->rt_throttled = 1;
-
                if (rt_rq_throttled(rt_rq)) {
                        sched_rt_rq_dequeue(rt_rq);
                        return 1;
@@ -212,29 +270,6 @@ static int sched_rt_runtime_exceeded(struct rt_rq *rt_rq)
        return 0;
 }
 
-static void update_sched_rt_period(struct rq *rq)
-{
-       struct rt_rq *rt_rq;
-       u64 period;
-
-       while (rq->clock > rq->rt_period_expire) {
-               period = (u64)sysctl_sched_rt_period * NSEC_PER_USEC;
-               rq->rt_period_expire += period;
-
-               for_each_leaf_rt_rq(rt_rq, rq) {
-                       u64 runtime = sched_rt_runtime(rt_rq);
-
-                       rt_rq->rt_time -= min(rt_rq->rt_time, runtime);
-                       if (rt_rq->rt_throttled && rt_rq->rt_time < runtime) {
-                               rt_rq->rt_throttled = 0;
-                               sched_rt_rq_enqueue(rt_rq);
-                       }
-               }
-
-               rq->rt_throttled = 0;
-       }
-}
-
 /*
  * Update the current task's runtime statistics. Skip current tasks that
  * are not in our scheduling class.
@@ -284,6 +319,11 @@ void inc_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
 #ifdef CONFIG_RT_GROUP_SCHED
        if (rt_se_boosted(rt_se))
                rt_rq->rt_nr_boosted++;
+
+       if (rt_rq->tg)
+               start_rt_bandwidth(&rt_rq->tg->rt_bandwidth);
+#else
+       start_rt_bandwidth(&def_rt_bandwidth);
 #endif
 }
 
index be332e1a0c29c3a18045c14da1b8bfadacf5054e..fd3364827ccf0a838c79fe240016c93cedd508e4 100644 (file)
@@ -307,7 +307,7 @@ static struct ctl_table kern_table[] = {
                .data           = &sysctl_sched_rt_period,
                .maxlen         = sizeof(unsigned int),
                .mode           = 0644,
-               .proc_handler   = &proc_dointvec,
+               .proc_handler   = &sched_rt_handler,
        },
        {
                .ctl_name       = CTL_UNNUMBERED,
@@ -315,7 +315,7 @@ static struct ctl_table kern_table[] = {
                .data           = &sysctl_sched_rt_runtime,
                .maxlen         = sizeof(int),
                .mode           = 0644,
-               .proc_handler   = &proc_dointvec,
+               .proc_handler   = &sched_rt_handler,
        },
        {
                .ctl_name       = CTL_UNNUMBERED,
index 69dba0c71727b70828fbd9b3122392feba4c2f3c..d358d4e3a95806c2ff71544947dee234809fdb88 100644 (file)
@@ -191,7 +191,6 @@ u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time)
 void tick_nohz_stop_sched_tick(void)
 {
        unsigned long seq, last_jiffies, next_jiffies, delta_jiffies, flags;
-       unsigned long rt_jiffies;
        struct tick_sched *ts;
        ktime_t last_update, expires, now;
        struct clock_event_device *dev = __get_cpu_var(tick_cpu_device).evtdev;
@@ -243,10 +242,6 @@ void tick_nohz_stop_sched_tick(void)
        next_jiffies = get_next_timer_interrupt(last_jiffies);
        delta_jiffies = next_jiffies - last_jiffies;
 
-       rt_jiffies = rt_needs_cpu(cpu);
-       if (rt_jiffies && rt_jiffies < delta_jiffies)
-               delta_jiffies = rt_jiffies;
-
        if (rcu_needs_cpu(cpu))
                delta_jiffies = 1;
        /*
index 7132022a040cc764b1c8cdd53744e7398e05502d..5925c6887c10d629cba6f977c6ea9f3896b48dec 100644 (file)
@@ -193,6 +193,33 @@ static ssize_t cpu_rt_runtime_store(struct kobject *kobj,
 
 static struct kobj_attribute cpu_rt_runtime_attr =
        __ATTR(cpu_rt_runtime, 0644, cpu_rt_runtime_show, cpu_rt_runtime_store);
+
+static ssize_t cpu_rt_period_show(struct kobject *kobj,
+                                  struct kobj_attribute *attr,
+                                  char *buf)
+{
+       struct user_struct *up = container_of(kobj, struct user_struct, kobj);
+
+       return sprintf(buf, "%lu\n", sched_group_rt_period(up->tg));
+}
+
+static ssize_t cpu_rt_period_store(struct kobject *kobj,
+                                   struct kobj_attribute *attr,
+                                   const char *buf, size_t size)
+{
+       struct user_struct *up = container_of(kobj, struct user_struct, kobj);
+       unsigned long rt_period;
+       int rc;
+
+       sscanf(buf, "%lu", &rt_period);
+
+       rc = sched_group_set_rt_period(up->tg, rt_period);
+
+       return (rc ? rc : size);
+}
+
+static struct kobj_attribute cpu_rt_period_attr =
+       __ATTR(cpu_rt_period, 0644, cpu_rt_period_show, cpu_rt_period_store);
 #endif
 
 /* default attributes per uid directory */
@@ -202,6 +229,7 @@ static struct attribute *uids_attributes[] = {
 #endif
 #ifdef CONFIG_RT_GROUP_SCHED
        &cpu_rt_runtime_attr.attr,
+       &cpu_rt_period_attr.attr,
 #endif
        NULL
 };