x86/crash: Indicate how well nmi_shootdown_cpus() managed to do

Having nmi_shootdown_cpus() report which pcpus failed to be shot down is a
useful debugging hint as to what possibly went wrong (especially when the
crash logs seem to indicate that an NMI timeout occurred while waiting for one
of the problematic pcpus to perform an action).

This is achieved by swapping an atomic_t count of unreported pcpus with a
cpumask.  In the case that the 1 second timeout occurs, use the cpumask to
identify the problematic pcpus.

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Keir Fraser <keir@xen.org>

diff --git a/xen/arch/x86/crash.c b/xen/arch/x86/crash.c
index 0a807d1089209416d5b8b6460a88825eb4ba8dc8..4ef379b57fe1900b07dbbc8edc9d586e0419ac42 100644 (file)
--- a/xen/arch/x86/crash.c
+++ b/xen/arch/x86/crash.c
@@ -22,6 +22,7 @@
 #include <xen/perfc.h>
 #include <xen/kexec.h>
 #include <xen/sched.h>
+#include <xen/keyhandler.h>
 #include <public/xen.h>
 #include <asm/shared.h>
 #include <asm/hvm/support.h>
@@ -30,7 +31,7 @@
 #include <xen/iommu.h>
 #include <asm/hpet.h>
 
-static atomic_t waiting_for_crash_ipi;
+static cpumask_t waiting_to_crash;
 static unsigned int crashing_cpu;
 static DEFINE_PER_CPU_READ_MOSTLY(bool_t, crash_save_done);
 
@@ -65,7 +66,7 @@ void __attribute__((noreturn)) do_nmi_crash(struct cpu_user_regs *regs)
         __stop_this_cpu();
 
         this_cpu(crash_save_done) = 1;
-        atomic_dec(&waiting_for_crash_ipi);
+        cpumask_clear_cpu(cpu, &waiting_to_crash);
     }
 
     /* Poor mans self_nmi().  __stop_this_cpu() has reverted the LAPIC
@@ -122,7 +123,7 @@ static void nmi_shootdown_cpus(void)
     crashing_cpu = cpu;
     local_irq_count(crashing_cpu) = 0;
 
-    atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
+    cpumask_andnot(&waiting_to_crash, &cpu_online_map, cpumask_of(cpu));
 
     /* Change NMI trap handlers.  Non-crashing pcpus get nmi_crash which
      * invokes do_nmi_crash (above), which cause them to write state and
@@ -162,12 +163,22 @@ static void nmi_shootdown_cpus(void)
     smp_send_nmi_allbutself();
 
     msecs = 1000; /* Wait at most a second for the other cpus to stop */
-    while ( (atomic_read(&waiting_for_crash_ipi) > 0) && msecs )
+    while ( !cpumask_empty(&waiting_to_crash) && msecs )
     {
         mdelay(1);
         msecs--;
     }
 
+    /* Leave a hint of how well we did trying to shoot down the other cpus */
+    if ( cpumask_empty(&waiting_to_crash) )
+        printk("Shot down all CPUs\n");
+    else
+    {
+        cpulist_scnprintf(keyhandler_scratch, sizeof keyhandler_scratch,
+                          &waiting_to_crash);
+        printk("Failed to shoot down CPUs {%s}\n", keyhandler_scratch);
+    }
+
     /* Crash shutdown any IOMMU functionality as the crashdump kernel is not
      * happy when booting if interrupt/dma remapping is still enabled */
     iommu_crash_shutdown();