[PATCH] i386: avoid checking for cpu gone when CONFIG_HOTPLUG_CPU not defined
authorKeshavamurthy, Anil S <anil.s.keshavamurthy@intel.com>
Wed, 2 May 2007 17:27:18 +0000 (19:27 +0200)
committerAndi Kleen <andi@basil.nowhere.org>
Wed, 2 May 2007 17:27:18 +0000 (19:27 +0200)
Avoid checking for cpu gone in mm hot path when CONFIG_HOTPLUG_CPU is not
defined.

Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Cc: Gautham R Shenoy <ego@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
arch/i386/kernel/smp.c

index f8667109db1c6b5deaa9afd21973ebcf099e30d3..f98c3ffd6fc3d6eb7584d04bddad6c3dcd8d8dd0 100644 (file)
@@ -359,10 +359,12 @@ void native_flush_tlb_others(const cpumask_t *cpumaskp, struct mm_struct *mm,
        BUG_ON(cpu_isset(smp_processor_id(), cpumask));
        BUG_ON(!mm);
 
+#ifdef CONFIG_HOTPLUG_CPU
        /* If a CPU which we ran on has gone down, OK. */
        cpus_and(cpumask, cpumask, cpu_online_map);
-       if (cpus_empty(cpumask))
+       if (unlikely(cpus_empty(cpumask)))
                return;
+#endif
 
        /*
         * i'm not happy about this global shared spinlock in the