Messages in this thread Patch in this message | | | From | "Srivatsa S. Bhat" <> | Subject | [RFC PATCH v2 06/10] sched, cpu hotplug: Use stable online cpus in try_to_wake_up() & select_task_rq() | Date | Thu, 06 Dec 2012 00:14:03 +0530 |
| |
Once stop_machine() is gone from the CPU offline path, we won't be able to depend on preempt_disable() to prevent CPUs from going offline from under us.
Use the get/put_online_cpus_atomic_light() APIs to prevent changes to the cpu_online_mask, while invoking from atomic context.
Scheduler functions such as try_to_wake_up() and select_task_rq() (and even select_fallback_rq()) deal with picking new CPUs to run tasks. They care only about the cpu_online_mask and not really about the actual CPU tear-down. So they qualify as "light" atomic readers. So use the _light() variant of the APIs.
Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> ---
kernel/sched/core.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 2d8927f..c9a331e 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -1103,6 +1103,10 @@ EXPORT_SYMBOL_GPL(kick_process); #ifdef CONFIG_SMP /* * ->cpus_allowed is protected by both rq->lock and p->pi_lock + * + * Must be called under get/put_online_cpus_atomic_light() or + * equivalent, to avoid CPUs from going offline from underneath + * us. */ static int select_fallback_rq(int cpu, struct task_struct *p) { @@ -1166,6 +1170,9 @@ out: /* * The caller (fork, wakeup) owns p->pi_lock, ->cpus_allowed is stable. + * + * Must be called under get/put_online_cpus_atomic_light(), to prevent + * CPUs from going offline from underneath us. */ static inline int select_task_rq(struct task_struct *p, int sd_flags, int wake_flags) @@ -1406,6 +1413,7 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) int cpu, success = 0; smp_wmb(); + get_online_cpus_atomic_light(); raw_spin_lock_irqsave(&p->pi_lock, flags); if (!(p->state & state)) goto out; @@ -1446,6 +1454,7 @@ stat: ttwu_stat(p, cpu, wake_flags); out: raw_spin_unlock_irqrestore(&p->pi_lock, flags); + put_online_cpus_atomic_light(); return success; } @@ -1624,6 +1633,7 @@ void wake_up_new_task(struct task_struct *p) unsigned long flags; struct rq *rq; + get_online_cpus_atomic_light(); raw_spin_lock_irqsave(&p->pi_lock, flags); #ifdef CONFIG_SMP /* @@ -1644,6 +1654,7 @@ void wake_up_new_task(struct task_struct *p) p->sched_class->task_woken(rq, p); #endif task_rq_unlock(rq, p, &flags); + put_online_cpus_atomic_light(); } #ifdef CONFIG_PREEMPT_NOTIFIERS @@ -2541,6 +2552,7 @@ void sched_exec(void) unsigned long flags; int dest_cpu; + get_online_cpus_atomic_light(); raw_spin_lock_irqsave(&p->pi_lock, flags); dest_cpu = p->sched_class->select_task_rq(p, SD_BALANCE_EXEC, 0); if (dest_cpu == smp_processor_id()) @@ -2550,11 +2562,13 @@ void sched_exec(void) struct migration_arg arg = { p, dest_cpu }; raw_spin_unlock_irqrestore(&p->pi_lock, flags); + put_online_cpus_atomic_light(); stop_one_cpu(task_cpu(p), migration_cpu_stop, &arg); return; } unlock: raw_spin_unlock_irqrestore(&p->pi_lock, flags); + put_online_cpus_atomic_light(); } #endif
| |