perf, cpu hotplug: Run CPU_STARTING notifiers with irqs disabled
authorSrivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Tue, 16 Oct 2012 07:58:10 +0000 (13:28 +0530)
committerIngo Molnar <mingo@kernel.org>
Wed, 24 Oct 2012 08:01:58 +0000 (10:01 +0200)
The CPU_STARTING notifiers are supposed to be run with irqs
disabled. But the perf_cpu_notifier() macro invokes them without
doing that. Fix it.

Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: peterz@infradead.org
Cc: acme@ghostprotocols.net
Link: http://lkml.kernel.org/r/20121016075809.3572.47848.stgit@srivatsabhat.in.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
include/linux/perf_event.h

index 2e902359aee596f128dee568d06e0e0ab3149dfe..06478056da03f21d958b0a46f41a2b5c8b4b43c1 100644 (file)
@@ -803,10 +803,13 @@ static inline void perf_event_task_tick(void)                             { }
 do {                                                                   \
        static struct notifier_block fn##_nb __cpuinitdata =            \
                { .notifier_call = fn, .priority = CPU_PRI_PERF };      \
+       unsigned long flags;                                            \
        fn(&fn##_nb, (unsigned long)CPU_UP_PREPARE,                     \
                (void *)(unsigned long)smp_processor_id());             \
+       local_irq_save(flags);                                          \
        fn(&fn##_nb, (unsigned long)CPU_STARTING,                       \
                (void *)(unsigned long)smp_processor_id());             \
+       local_irq_restore(flags);                                       \
        fn(&fn##_nb, (unsigned long)CPU_ONLINE,                         \
                (void *)(unsigned long)smp_processor_id());             \
        register_cpu_notifier(&fn##_nb);                                \