From: Ingo Molnar Date: Mon, 9 Jul 2007 16:51:58 +0000 (+0200) Subject: sched: make posix-cpu-timers use CFS's accounting information X-Git-Tag: v2.6.23-rc1~1261 X-Git-Url: https://err.no/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=41b86e9c510ae66639bf29d3201e1d2384a7fde6;p=linux-2.6 sched: make posix-cpu-timers use CFS's accounting information update the posix-cpu-timers code to use CFS's CPU accounting information. Signed-off-by: Ingo Molnar --- diff --git a/include/linux/sched.h b/include/linux/sched.h index 995eb407c2..3e7f1890e5 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -482,7 +482,8 @@ struct signal_struct { * from jiffies_to_ns(utime + stime) if sched_clock uses something * other than jiffies.) */ - unsigned long long sched_time; + unsigned long sched_time; + unsigned long long sum_sched_runtime; /* * We don't bother to synchronize most readers of this at all, @@ -1308,7 +1309,7 @@ static inline int set_cpus_allowed(struct task_struct *p, cpumask_t new_mask) extern unsigned long long sched_clock(void); extern unsigned long long -current_sched_time(const struct task_struct *current_task); +task_sched_runtime(struct task_struct *task); /* sched_exec is called by processes performing an exec */ #ifdef CONFIG_SMP diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 1de710e183..b53c8fcd9d 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -161,7 +161,7 @@ static inline cputime_t virt_ticks(struct task_struct *p) } static inline unsigned long long sched_ns(struct task_struct *p) { - return (p == current) ? current_sched_time(p) : p->sched_time; + return task_sched_runtime(p); } int posix_cpu_clock_getres(const clockid_t which_clock, struct timespec *tp) @@ -246,10 +246,10 @@ static int cpu_clock_sample_group_locked(unsigned int clock_idx, } while (t != p); break; case CPUCLOCK_SCHED: - cpu->sched = p->signal->sched_time; + cpu->sched = p->signal->sum_sched_runtime; /* Add in each other live thread. */ while ((t = next_thread(t)) != p) { - cpu->sched += t->sched_time; + cpu->sched += t->se.sum_exec_runtime; } cpu->sched += sched_ns(p); break; @@ -422,7 +422,7 @@ int posix_cpu_timer_del(struct k_itimer *timer) */ static void cleanup_timers(struct list_head *head, cputime_t utime, cputime_t stime, - unsigned long long sched_time) + unsigned long long sum_exec_runtime) { struct cpu_timer_list *timer, *next; cputime_t ptime = cputime_add(utime, stime); @@ -451,10 +451,10 @@ static void cleanup_timers(struct list_head *head, ++head; list_for_each_entry_safe(timer, next, head, entry) { list_del_init(&timer->entry); - if (timer->expires.sched < sched_time) { + if (timer->expires.sched < sum_exec_runtime) { timer->expires.sched = 0; } else { - timer->expires.sched -= sched_time; + timer->expires.sched -= sum_exec_runtime; } } } @@ -467,7 +467,7 @@ static void cleanup_timers(struct list_head *head, void posix_cpu_timers_exit(struct task_struct *tsk) { cleanup_timers(tsk->cpu_timers, - tsk->utime, tsk->stime, tsk->sched_time); + tsk->utime, tsk->stime, tsk->se.sum_exec_runtime); } void posix_cpu_timers_exit_group(struct task_struct *tsk) @@ -475,7 +475,7 @@ void posix_cpu_timers_exit_group(struct task_struct *tsk) cleanup_timers(tsk->signal->cpu_timers, cputime_add(tsk->utime, tsk->signal->utime), cputime_add(tsk->stime, tsk->signal->stime), - tsk->sched_time + tsk->signal->sched_time); + tsk->se.sum_exec_runtime + tsk->signal->sum_sched_runtime); } @@ -536,7 +536,7 @@ static void process_timer_rebalance(struct task_struct *p, nsleft = max_t(unsigned long long, nsleft, 1); do { if (likely(!(t->flags & PF_EXITING))) { - ns = t->sched_time + nsleft; + ns = t->se.sum_exec_runtime + nsleft; if (t->it_sched_expires == 0 || t->it_sched_expires > ns) { t->it_sched_expires = ns; @@ -1004,7 +1004,7 @@ static void check_thread_timers(struct task_struct *tsk, struct cpu_timer_list *t = list_first_entry(timers, struct cpu_timer_list, entry); - if (!--maxfire || tsk->sched_time < t->expires.sched) { + if (!--maxfire || tsk->se.sum_exec_runtime < t->expires.sched) { tsk->it_sched_expires = t->expires.sched; break; } @@ -1024,7 +1024,7 @@ static void check_process_timers(struct task_struct *tsk, int maxfire; struct signal_struct *const sig = tsk->signal; cputime_t utime, stime, ptime, virt_expires, prof_expires; - unsigned long long sched_time, sched_expires; + unsigned long long sum_sched_runtime, sched_expires; struct task_struct *t; struct list_head *timers = sig->cpu_timers; @@ -1044,12 +1044,12 @@ static void check_process_timers(struct task_struct *tsk, */ utime = sig->utime; stime = sig->stime; - sched_time = sig->sched_time; + sum_sched_runtime = sig->sum_sched_runtime; t = tsk; do { utime = cputime_add(utime, t->utime); stime = cputime_add(stime, t->stime); - sched_time += t->sched_time; + sum_sched_runtime += t->se.sum_exec_runtime; t = next_thread(t); } while (t != tsk); ptime = cputime_add(utime, stime); @@ -1090,7 +1090,7 @@ static void check_process_timers(struct task_struct *tsk, struct cpu_timer_list *t = list_first_entry(timers, struct cpu_timer_list, entry); - if (!--maxfire || sched_time < t->expires.sched) { + if (!--maxfire || sum_sched_runtime < t->expires.sched) { sched_expires = t->expires.sched; break; } @@ -1182,7 +1182,7 @@ static void check_process_timers(struct task_struct *tsk, virt_left = cputime_sub(virt_expires, utime); virt_left = cputime_div_non_zero(virt_left, nthreads); if (sched_expires) { - sched_left = sched_expires - sched_time; + sched_left = sched_expires - sum_sched_runtime; do_div(sched_left, nthreads); sched_left = max_t(unsigned long long, sched_left, 1); } else { @@ -1208,7 +1208,7 @@ static void check_process_timers(struct task_struct *tsk, t->it_virt_expires = ticks; } - sched = t->sched_time + sched_left; + sched = t->se.sum_exec_runtime + sched_left; if (sched_expires && (t->it_sched_expires == 0 || t->it_sched_expires > sched)) { t->it_sched_expires = sched; @@ -1300,7 +1300,7 @@ void run_posix_cpu_timers(struct task_struct *tsk) if (UNEXPIRED(prof) && UNEXPIRED(virt) && (tsk->it_sched_expires == 0 || - tsk->sched_time < tsk->it_sched_expires)) + tsk->se.sum_exec_runtime < tsk->it_sched_expires)) return; #undef UNEXPIRED diff --git a/kernel/sched.c b/kernel/sched.c index 29eb227e33..0333abdda8 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -3156,28 +3156,23 @@ 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. + * Return p->sum_exec_runtime plus any more ns on the sched_clock + * that have not yet been banked in case the task is currently running. */ -static inline void -update_cpu_clock(struct task_struct *p, struct rq *rq, unsigned long long now) +unsigned long long task_sched_runtime(struct task_struct *p) { - p->sched_time += now - p->last_ran; - p->last_ran = rq->most_recent_timestamp = now; -} - -/* - * 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; + u64 ns, delta_exec; + struct rq *rq; - local_irq_save(flags); - ns = p->sched_time + sched_clock() - p->last_ran; - local_irq_restore(flags); + rq = task_rq_lock(p, &flags); + ns = p->se.sum_exec_runtime; + if (rq->curr == p) { + delta_exec = rq_clock(rq) - p->se.exec_start; + if ((s64)delta_exec > 0) + ns += delta_exec; + } + task_rq_unlock(rq, &flags); return ns; } @@ -3360,14 +3355,11 @@ out_unlock: */ void scheduler_tick(void) { - unsigned long long now = sched_clock(); struct task_struct *p = current; int cpu = smp_processor_id(); int idle_at_tick = idle_cpu(cpu); struct rq *rq = cpu_rq(cpu); - update_cpu_clock(p, rq, now); - if (!idle_at_tick) task_running_tick(rq, p); #ifdef CONFIG_SMP @@ -3550,8 +3542,6 @@ switch_tasks: clear_tsk_need_resched(prev); rcu_qsctr_inc(task_cpu(prev)); - update_cpu_clock(prev, rq, now); - prev->sleep_avg -= run_time; if ((long)prev->sleep_avg <= 0) prev->sleep_avg = 0;