debug lockups: Improve lockup detection, fix generic arch fallback

[pandora-kernel.git] / arch / x86 / kernel / apic / nmi.c

diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c
index ce4fbfa..db72202 100644 (file)
--- a/arch/x86/kernel/apic/nmi.c
+++ b/arch/x86/kernel/apic/nmi.c
@@ -39,7 +39,7 @@
 int unknown_nmi_panic;
 int nmi_watchdog_enabled;
 
-static cpumask_var_t backtrace_mask;
+static cpumask_t backtrace_mask __read_mostly;
 
 /* nmi_active:
  * >0: the lapic NMI watchdog is active, but can be disabled
@@ -66,7 +66,7 @@ static inline unsigned int get_nmi_count(int cpu)
 
 static inline int mce_in_progress(void)
 {
-#if defined(CONFIG_X86_64) && defined(CONFIG_X86_MCE)
+#if defined(CONFIG_X86_NEW_MCE)
        return atomic_read(&mce_entry) > 0;
 #endif
        return 0;
@@ -104,7 +104,7 @@ static __init void nmi_cpu_busy(void *data)
 }
 #endif
 
-static void report_broken_nmi(int cpu, int *prev_nmi_count)
+static void report_broken_nmi(int cpu, unsigned int *prev_nmi_count)
 {
        printk(KERN_CONT "\n");
 
@@ -138,7 +138,6 @@ int __init check_nmi_watchdog(void)
        if (!prev_nmi_count)
                goto error;
 
-       alloc_cpumask_var(&backtrace_mask, GFP_KERNEL|__GFP_ZERO);
        printk(KERN_INFO "Testing NMI watchdog ... ");
 
 #ifdef CONFIG_SMP
@@ -415,14 +414,17 @@ nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
        }
 
        /* We can be called before check_nmi_watchdog, hence NULL check. */
-       if (backtrace_mask != NULL && cpumask_test_cpu(cpu, backtrace_mask)) {
+       if (cpumask_test_cpu(cpu, &backtrace_mask)) {
                static DEFINE_SPINLOCK(lock);   /* Serialise the printks */
 
                spin_lock(&lock);
                printk(KERN_WARNING "NMI backtrace for cpu %d\n", cpu);
+               show_regs(regs);
                dump_stack();
                spin_unlock(&lock);
-               cpumask_clear_cpu(cpu, backtrace_mask);
+               cpumask_clear_cpu(cpu, &backtrace_mask);
+
+               rc = 1;
        }
 
        /* Could check oops_in_progress here too, but it's safer not to */
@@ -552,14 +554,18 @@ int do_nmi_callback(struct pt_regs *regs, int cpu)
        return 0;
 }
 
-void __trigger_all_cpu_backtrace(void)
+void arch_trigger_all_cpu_backtrace(void)
 {
        int i;
 
-       cpumask_copy(backtrace_mask, cpu_online_mask);
+       cpumask_copy(&backtrace_mask, cpu_online_mask);
+
+       printk(KERN_INFO "sending NMI to all CPUs:\n");
+       apic->send_IPI_all(NMI_VECTOR);
+
        /* Wait for up to 10 seconds for all CPUs to do the backtrace */
        for (i = 0; i < 10 * 1000; i++) {
-               if (cpumask_empty(backtrace_mask))
+               if (cpumask_empty(&backtrace_mask))
                        break;
                mdelay(1);
        }