To isolate CPUs (isolate from timers) from sysfs using cpusets, we need some support from the timer core. i.e. A routine timer_quiesce_cpu() which would migrate away all the unpinned timers, but shouldn't touch the pinned ones. This patch creates this routine. Signed-off-by: Viresh Kumar <viresh.kumar@xxxxxxxxxx> --- include/linux/timer.h | 3 +++ kernel/timer.c | 54 ++++++++++++++++++++++++++++++++++++++++----------- 2 files changed, 46 insertions(+), 11 deletions(-) diff --git a/include/linux/timer.h b/include/linux/timer.h index 2962403..1588a4f 100644 --- a/include/linux/timer.h +++ b/include/linux/timer.h @@ -196,6 +196,9 @@ extern void set_timer_slack(struct timer_list *time, int slack_hz); */ extern unsigned long get_next_timer_interrupt(unsigned long now); +/* To be used from cpusets, only */ +extern void timer_quiesce_cpu(void *cpup); + /* * Timer-statistics info: */ diff --git a/kernel/timer.c b/kernel/timer.c index 6c3a371..4676a07 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -1602,18 +1602,27 @@ static int init_timers_cpu(int cpu) return 0; } -#ifdef CONFIG_HOTPLUG_CPU -static void migrate_timer_list(struct tvec_base *new_base, struct list_head *head) +#if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_CPUSETS) +static void migrate_timer_list(struct tvec_base *new_base, + struct list_head *head, bool remove_pinned) { struct timer_list *timer; + struct list_head pinned_list; int is_pinned; + INIT_LIST_HEAD(&pinned_list); + while (!list_empty(head)) { timer = list_first_entry(head, struct timer_list, entry); - /* We ignore the accounting on the dying cpu */ - detach_timer(timer, false); is_pinned = tbase_get_pinned(timer->base); + if (!remove_pinned && is_pinned) { + list_move_tail(&timer->entry, &pinned_list); + continue; + } else { + /* We ignore the accounting on the dying cpu */ + detach_timer(timer, false); + } /* Check if CPU still has pinned timers */ if (unlikely(WARN(is_pinned, @@ -1624,15 +1633,18 @@ static void migrate_timer_list(struct tvec_base *new_base, struct list_head *hea timer_set_base(timer, new_base); internal_add_timer(new_base, timer); } + + if (!list_empty(&pinned_list)) + list_splice_tail(&pinned_list, head); } -static void migrate_timers(int cpu) +/* Migrate timers from 'cpu' to this_cpu */ +static void __migrate_timers(int cpu, bool remove_pinned) { struct tvec_base *old_base; struct tvec_base *new_base; int i; - BUG_ON(cpu_online(cpu)); old_base = per_cpu(tvec_bases, cpu); new_base = get_cpu_var(tvec_bases); /* @@ -1645,20 +1657,40 @@ static void migrate_timers(int cpu) BUG_ON(old_base->running_timer); for (i = 0; i < TVR_SIZE; i++) - migrate_timer_list(new_base, old_base->tv1.vec + i); + migrate_timer_list(new_base, old_base->tv1.vec + i, + remove_pinned); for (i = 0; i < TVN_SIZE; i++) { - migrate_timer_list(new_base, old_base->tv2.vec + i); - migrate_timer_list(new_base, old_base->tv3.vec + i); - migrate_timer_list(new_base, old_base->tv4.vec + i); - migrate_timer_list(new_base, old_base->tv5.vec + i); + migrate_timer_list(new_base, old_base->tv2.vec + i, + remove_pinned); + migrate_timer_list(new_base, old_base->tv3.vec + i, + remove_pinned); + migrate_timer_list(new_base, old_base->tv4.vec + i, + remove_pinned); + migrate_timer_list(new_base, old_base->tv5.vec + i, + remove_pinned); } spin_unlock(&old_base->lock); spin_unlock_irq(&new_base->lock); put_cpu_var(tvec_bases); } +#endif /* CONFIG_HOTPLUG_CPU || CONFIG_CPUSETS */ + +#ifdef CONFIG_HOTPLUG_CPU +static void migrate_timers(int cpu) +{ + BUG_ON(cpu_online(cpu)); + __migrate_timers(cpu, true); +} #endif /* CONFIG_HOTPLUG_CPU */ +#ifdef CONFIG_CPUSETS +void timer_quiesce_cpu(void *cpup) +{ + __migrate_timers(*(int *)cpup, false); +} +#endif /* CONFIG_CPUSETS */ + static int timer_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { -- 1.7.12.rc2.18.g61b472e -- To unsubscribe from this list: send the line "unsubscribe cgroups" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html