Merge branch 'proc' of git://git.kernel.org/pub/scm/linux/kernel/git/adobriyan/proc

[linux-2.6-omap-h63xx.git] / kernel / sched.c

diff --git a/kernel/sched.c b/kernel/sched.c
index 09a8c15748f1e5fcd2cae7d98b47ecfd7e73ba40..6625c3c4b10d06c3f76c371becd615d174244902 100644 (file)
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -55,6 +55,7 @@
 #include <linux/cpuset.h>
 #include <linux/percpu.h>
 #include <linux/kthread.h>
+#include <linux/proc_fs.h>
 #include <linux/seq_file.h>
 #include <linux/sysctl.h>
 #include <linux/syscalls.h>
@@ -71,6 +72,7 @@
 #include <linux/debugfs.h>
 #include <linux/ctype.h>
 #include <linux/ftrace.h>
+#include <trace/sched.h>
 
 #include <asm/tlb.h>
 #include <asm/irq_regs.h>
@@ -226,9 +228,8 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b)
 
                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);
+               hrtimer_start_expires(&rt_b->rt_period_timer,
+                               HRTIMER_MODE_ABS);
        }
        spin_unlock(&rt_b->rt_runtime_lock);
 }
@@ -817,6 +818,13 @@ const_debug unsigned int sysctl_sched_nr_migrate = 32;
  */
 unsigned int sysctl_sched_shares_ratelimit = 250000;
 
+/*
+ * Inject some fuzzyness into changing the per-cpu group shares
+ * this avoids remote rq-locks at the expense of fairness.
+ * default: 4
+ */
+unsigned int sysctl_sched_shares_thresh = 4;
+
 /*
  * period over which we measure -rt task cpu usage in us.
  * default: 1s
@@ -1063,7 +1071,7 @@ static void hrtick_start(struct rq *rq, u64 delay)
        struct hrtimer *timer = &rq->hrtick_timer;
        ktime_t time = ktime_add_ns(timer->base->get_time(), delay);
 
-       timer->expires = time;
+       hrtimer_set_expires(timer, time);
 
        if (rq == this_rq()) {
                hrtimer_restart(timer);
@@ -1453,8 +1461,8 @@ static void __set_se_shares(struct sched_entity *se, unsigned long shares);
  * Calculate and set the cpu's group shares.
  */
 static void
-__update_group_shares_cpu(struct task_group *tg, int cpu,
-                         unsigned long sd_shares, unsigned long sd_rq_weight)
+update_group_shares_cpu(struct task_group *tg, int cpu,
+                       unsigned long sd_shares, unsigned long sd_rq_weight)
 {
        int boost = 0;
        unsigned long shares;
@@ -1485,19 +1493,23 @@ __update_group_shares_cpu(struct task_group *tg, int cpu,
         *
         */
        shares = (sd_shares * rq_weight) / (sd_rq_weight + 1);
+       shares = clamp_t(unsigned long, shares, MIN_SHARES, MAX_SHARES);
 
-       /*
-        * record the actual number of shares, not the boosted amount.
-        */
-       tg->cfs_rq[cpu]->shares = boost ? 0 : shares;
-       tg->cfs_rq[cpu]->rq_weight = rq_weight;
+       if (abs(shares - tg->se[cpu]->load.weight) >
+                       sysctl_sched_shares_thresh) {
+               struct rq *rq = cpu_rq(cpu);
+               unsigned long flags;
 
-       if (shares < MIN_SHARES)
-               shares = MIN_SHARES;
-       else if (shares > MAX_SHARES)
-               shares = MAX_SHARES;
+               spin_lock_irqsave(&rq->lock, flags);
+               /*
+                * record the actual number of shares, not the boosted amount.
+                */
+               tg->cfs_rq[cpu]->shares = boost ? 0 : shares;
+               tg->cfs_rq[cpu]->rq_weight = rq_weight;
 
-       __set_se_shares(tg->se[cpu], shares);
+               __set_se_shares(tg->se[cpu], shares);
+               spin_unlock_irqrestore(&rq->lock, flags);
+       }
 }
 
 /*
@@ -1526,14 +1538,8 @@ static int tg_shares_up(struct task_group *tg, void *data)
        if (!rq_weight)
                rq_weight = cpus_weight(sd->span) * NICE_0_LOAD;
 
-       for_each_cpu_mask(i, sd->span) {
-               struct rq *rq = cpu_rq(i);
-               unsigned long flags;
-
-               spin_lock_irqsave(&rq->lock, flags);
-               __update_group_shares_cpu(tg, i, shares, rq_weight);
-               spin_unlock_irqrestore(&rq->lock, flags);
-       }
+       for_each_cpu_mask(i, sd->span)
+               update_group_shares_cpu(tg, i, shares, rq_weight);
 
        return 0;
 }
@@ -1936,6 +1942,7 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state)
                 * just go back and repeat.
                 */
                rq = task_rq_lock(p, &flags);
+               trace_sched_wait_task(rq, p);
                running = task_running(rq, p);
                on_rq = p->se.on_rq;
                ncsw = 0;
@@ -2297,9 +2304,7 @@ out_activate:
        success = 1;
 
 out_running:
-       trace_mark(kernel_sched_wakeup,
-               "pid %d state %ld ## rq %p task %p rq->curr %p",
-               p->pid, p->state, rq, p, rq->curr);
+       trace_sched_wakeup(rq, p);
        check_preempt_curr(rq, p, sync);
 
        p->state = TASK_RUNNING;
@@ -2432,9 +2437,7 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags)
                p->sched_class->task_new(rq, p);
                inc_nr_running(rq);
        }
-       trace_mark(kernel_sched_wakeup_new,
-               "pid %d state %ld ## rq %p task %p rq->curr %p",
-               p->pid, p->state, rq, p, rq->curr);
+       trace_sched_wakeup_new(rq, p);
        check_preempt_curr(rq, p, 0);
 #ifdef CONFIG_SMP
        if (p->sched_class->task_wake_up)
@@ -2607,11 +2610,7 @@ context_switch(struct rq *rq, struct task_struct *prev,
        struct mm_struct *mm, *oldmm;
 
        prepare_task_switch(rq, prev, next);
-       trace_mark(kernel_sched_schedule,
-               "prev_pid %d next_pid %d prev_state %ld "
-               "## rq %p prev %p next %p",
-               prev->pid, next->pid, prev->state,
-               rq, prev, next);
+       trace_sched_switch(rq, prev, next);
        mm = next->mm;
        oldmm = prev->active_mm;
        /*
@@ -2851,6 +2850,7 @@ static void sched_migrate_task(struct task_struct *p, int dest_cpu)
            || unlikely(!cpu_active(dest_cpu)))
                goto out;
 
+       trace_sched_migrate_task(rq, p, dest_cpu);
        /* force the process onto the specified CPU */
        if (migrate_task(p, dest_cpu, &req)) {
                /* Need to wait for migration thread (might exit: take ref). */
@@ -4448,12 +4448,8 @@ need_resched_nonpreemptible:
        if (sched_feat(HRTICK))
                hrtick_clear(rq);
 
-       /*
-        * Do the rq-clock update outside the rq lock:
-        */
-       local_irq_disable();
+       spin_lock_irq(&rq->lock);
        update_rq_clock(rq);
-       spin_lock(&rq->lock);
        clear_tsk_need_resched(prev);
 
        if (prev->state && !(preempt_count() & PREEMPT_ACTIVE)) {