x86, NMI: Add priorities to handlers

In order to consolidate the NMI die_chain events, we need to setup the priorities
for the die notifiers.

I started by defining a bunch of common priorities that can be used by the
notifier blocks.  Then I modified the notifier blocks to use the newly created
priorities.

Now that the priorities are straightened out, it should be easier to remove the
event DIE_NMI_IPI.

Signed-off-by: Don Zickus <dzickus@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1294348732-15030-4-git-send-email-dzickus@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

8 files changed, 29 insertions, 6 deletions

diff --git a/arch/x86/include/asm/nmi.h b/arch/x86/include/asm/nmi.h
index c4021b953510..c76f5b92b840 100644
--- a/arch/x86/include/asm/nmi.h
+++ b/arch/x86/include/asm/nmi.h
@@ -23,6 +23,26 @@ void arch_trigger_all_cpu_backtrace(void);
 #define arch_trigger_all_cpu_backtrace arch_trigger_all_cpu_backtrace
 #endif
 
+/*
+ * Define some priorities for the nmi notifier call chain.
+ *
+ * Create a local nmi bit that has a higher priority than
+ * external nmis, because the local ones are more frequent.
+ *
+ * Also setup some default high/normal/low settings for
+ * subsystems to registers with.  Using 4 bits to seperate
+ * the priorities.  This can go alot higher if needed be.
+ */
+
+#define NMI_LOCAL_SHIFT         16      /* randomly picked */
+#define NMI_LOCAL_BIT           (1ULL << NMI_LOCAL_SHIFT)
+#define NMI_HIGH_PRIOR          (1ULL << 8)
+#define NMI_NORMAL_PRIOR        (1ULL << 4)
+#define NMI_LOW_PRIOR           (1ULL << 0)
+#define NMI_LOCAL_HIGH_PRIOR    (NMI_LOCAL_BIT | NMI_HIGH_PRIOR)
+#define NMI_LOCAL_NORMAL_PRIOR  (NMI_LOCAL_BIT | NMI_NORMAL_PRIOR)
+#define NMI_LOCAL_LOW_PRIOR     (NMI_LOCAL_BIT | NMI_LOW_PRIOR)
+
 void stop_nmi(void);
 void restart_nmi(void);
 
diff --git a/arch/x86/kernel/apic/hw_nmi.c b/arch/x86/kernel/apic/hw_nmi.c
index 72ec29e1ae06..8bc49f1ac7bc 100644
--- a/arch/x86/kernel/apic/hw_nmi.c
+++ b/arch/x86/kernel/apic/hw_nmi.c
@@ -96,7 +96,7 @@ arch_trigger_all_cpu_backtrace_handler(struct notifier_block *self,
 static __read_mostly struct notifier_block backtrace_notifier = {
         .notifier_call          = arch_trigger_all_cpu_backtrace_handler,
         .next                   = NULL,
-        .priority               = 1
+        .priority               = NMI_LOCAL_LOW_PRIOR,
 };
 
 static int __init register_trigger_all_cpu_backtrace(void)
diff --git a/arch/x86/kernel/cpu/mcheck/mce-inject.c b/arch/x86/kernel/cpu/mcheck/mce-inject.c
index e7dbde7bfedb..59546c1219f9 100644
--- a/arch/x86/kernel/cpu/mcheck/mce-inject.c
+++ b/arch/x86/kernel/cpu/mcheck/mce-inject.c
@@ -95,7 +95,7 @@ static int mce_raise_notify(struct notifier_block *self,
 
 static struct notifier_block mce_raise_nb = {
         .notifier_call = mce_raise_notify,
-        .priority = 1000,
+        .priority = NMI_LOCAL_NORMAL_PRIOR,
 };
 
 /* Inject mce on current CPU */
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 0a360d146596..5e14b5e5fb81 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -1318,7 +1318,7 @@ perf_event_nmi_handler(struct notifier_block *self,
 static __read_mostly struct notifier_block perf_event_nmi_notifier = {
         .notifier_call          = perf_event_nmi_handler,
         .next                   = NULL,
-        .priority               = 1
+        .priority               = NMI_LOCAL_LOW_PRIOR,
 };
 
 static struct event_constraint unconstrained;
diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c
index cd21b654dec6..d43c84183d8f 100644
--- a/arch/x86/kernel/kgdb.c
+++ b/arch/x86/kernel/kgdb.c
@@ -48,6 +48,7 @@
 #include <asm/apicdef.h>
 #include <asm/system.h>
 #include <asm/apic.h>
+#include <asm/nmi.h>
 
 struct dbg_reg_def_t dbg_reg_def[DBG_MAX_REG_NUM] =
 {
@@ -606,7 +607,7 @@ static struct notifier_block kgdb_notifier = {
         /*
          * Lowest-prio notifier priority, we want to be notified last:
          */
-        .priority       = -INT_MAX,
+        .priority       = NMI_LOCAL_LOW_PRIOR,
 };
 
 /**
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index c495aa8d4815..9c1c83e4a742 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -778,6 +778,8 @@ static void smp_send_nmi_allbutself(void)
 
 static struct notifier_block crash_nmi_nb = {
         .notifier_call = crash_nmi_callback,
+        /* we want to be the first one called */
+        .priority = NMI_LOCAL_HIGH_PRIOR+1,
 };
 
 /* Halt all other CPUs, calling the specified function on each of them
diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 358c8b9c96a7..6e84ea42085a 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -361,7 +361,7 @@ static void nmi_cpu_setup(void *dummy)
 static struct notifier_block profile_exceptions_nb = {
         .notifier_call = profile_exceptions_notify,
         .next = NULL,
-        .priority = 2
+        .priority = NMI_LOCAL_LOW_PRIOR,
 };
 
 static void nmi_cpu_restore_registers(struct op_msrs *msrs)
diff --git a/arch/x86/oprofile/nmi_timer_int.c b/arch/x86/oprofile/nmi_timer_int.c
index 0636dd93cef8..720bf5a53c51 100644
--- a/arch/x86/oprofile/nmi_timer_int.c
+++ b/arch/x86/oprofile/nmi_timer_int.c
@@ -38,7 +38,7 @@ static int profile_timer_exceptions_notify(struct notifier_block *self,
 static struct notifier_block profile_timer_exceptions_nb = {
         .notifier_call = profile_timer_exceptions_notify,
         .next = NULL,
-        .priority = 0
+        .priority = NMI_LOW_PRIOR,
 };
 
 static int timer_start(void)