x86: Only call smp_processor_id in non-preempt cases
authorDon Zickus <dzickus@redhat.com>
Wed, 5 Jan 2011 03:38:07 +0000 (22:38 -0500)
committerIngo Molnar <mingo@elte.hu>
Wed, 5 Jan 2011 13:22:57 +0000 (14:22 +0100)
There are some paths that walk the die_chain with preemption on.
Make sure we are in an NMI call before we start doing anything.

This was triggered by do_general_protection calling notify_die
with DIE_GPF.

Reported-by: Jan Kiszka <jan.kiszka@web.de>
Signed-off-by: Don Zickus <dzickus@redhat.com>
LKML-Reference: <1294198689-15447-1-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
arch/x86/kernel/apic/hw_nmi.c

index c57d0b5..2b40a60 100644 (file)
@@ -51,7 +51,7 @@ arch_trigger_all_cpu_backtrace_handler(struct notifier_block *self,
 {
        struct die_args *args = __args;
        struct pt_regs *regs;
-       int cpu = smp_processor_id();
+       int cpu;
 
        switch (cmd) {
        case DIE_NMI:
@@ -63,6 +63,7 @@ arch_trigger_all_cpu_backtrace_handler(struct notifier_block *self,
        }
 
        regs = args->regs;
+       cpu = smp_processor_id();
 
        if (cpumask_test_cpu(cpu, to_cpumask(backtrace_mask))) {
                static arch_spinlock_t lock = __ARCH_SPIN_LOCK_UNLOCKED;