[PATCH 4/9] x86, NMI: Remove DIE_NMI_IPI and add priorties to handlers

From: Don Zickus
Date: Tue Nov 30 2010 - 17:28:42 EST

Next message: Don Zickus: "[PATCH 1/9] x86, NMI: Add NMI symbol constants and rename memory parity to PCI SERR"
Previous message: Don Zickus: "[PATCH 3/9] x86, NMI: Rewrite NMI handler"
In reply to: Don Zickus: "[PATCH 3/9] x86, NMI: Rewrite NMI handler"
Next in thread: Don Zickus: "[PATCH 1/9] x86, NMI: Add NMI symbol constants and rename memory parity to PCI SERR"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

When re-ordering how the NMI handles its callbacks, a conversation started
asking what DIE_NMI_IPI meant. No one could answer it.

Noticing that is was wasteful to call the die_chain a second time with just
another argument, DIE_NMI_IPI, it was decided to nuke it and add priorities
to the die_chain handlers to maintain existing behaviour.

This patch replaces DIE_NMI_IPI with the appropriate option, mostly DIE_NMI.
Then it adds priorities to those handlers, using a globally defined set of
priorities for NMI.

The thought is eventually we will just switch the nmi handlers from the
die_chain to something more nmi specific.

Signed-off-by: Don Zickus <dzickus@xxxxxxxxxx>
---
arch/x86/include/asm/kdebug.h | 1 -
arch/x86/include/asm/nmi.h | 20 ++++++++++++++++++++
arch/x86/kernel/apic/hw_nmi.c | 3 +--
arch/x86/kernel/apic/x2apic_uv_x.c | 2 +-
arch/x86/kernel/cpu/mcheck/mce-inject.c | 5 +++--
arch/x86/kernel/cpu/perf_event.c | 4 ++--
arch/x86/kernel/kgdb.c | 6 +-----
arch/x86/kernel/reboot.c | 5 ++++-
arch/x86/kernel/traps.c | 8 +-------
arch/x86/oprofile/nmi_int.c | 4 ++--
arch/x86/oprofile/nmi_timer_int.c | 4 ++--
11 files changed, 37 insertions(+), 25 deletions(-)

diff --git a/arch/x86/include/asm/kdebug.h b/arch/x86/include/asm/kdebug.h
index f23eb25..ca242d3 100644
--- a/arch/x86/include/asm/kdebug.h
+++ b/arch/x86/include/asm/kdebug.h
@@ -18,7 +18,6 @@ enum die_val {
DIE_TRAP,
DIE_GPF,
DIE_CALL,
- DIE_NMI_IPI,
DIE_PAGE_FAULT,
DIE_NMIUNKNOWN,
};
diff --git a/arch/x86/include/asm/nmi.h b/arch/x86/include/asm/nmi.h
index 3545838..9c4ff0b 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 a86840c..82bff25 100644
--- a/arch/x86/kernel/apic/hw_nmi.c
+++ b/arch/x86/kernel/apic/hw_nmi.c
@@ -63,7 +63,6 @@ arch_trigger_all_cpu_backtrace_handler(struct notifier_block *self,

switch (cmd) {
case DIE_NMI:
- case DIE_NMI_IPI:
break;

default:
@@ -90,7 +89,7 @@ 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/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c
index c1c52c3..927902d 100644
--- a/arch/x86/kernel/apic/x2apic_uv_x.c
+++ b/arch/x86/kernel/apic/x2apic_uv_x.c
@@ -639,7 +639,7 @@ void __cpuinit uv_cpu_init(void)
*/
int uv_handle_nmi(struct notifier_block *self, unsigned long reason, void *data)
{
- if (reason != DIE_NMI_IPI)
+ if (reason != DIE_NMIUNKNOWN)
return NOTIFY_OK;

if (in_crash_kexec)
diff --git a/arch/x86/kernel/cpu/mcheck/mce-inject.c b/arch/x86/kernel/cpu/mcheck/mce-inject.c
index e7dbde7..a779719 100644
--- a/arch/x86/kernel/cpu/mcheck/mce-inject.c
+++ b/arch/x86/kernel/cpu/mcheck/mce-inject.c
@@ -25,6 +25,7 @@
#include <linux/gfp.h>
#include <asm/mce.h>
#include <asm/apic.h>
+#include <asm/nmi.h>

/* Update fake mce registers on current CPU. */
static void inject_mce(struct mce *m)
@@ -83,7 +84,7 @@ static int mce_raise_notify(struct notifier_block *self,
struct die_args *args = (struct die_args *)data;
int cpu = smp_processor_id();
struct mce *m = &__get_cpu_var(injectm);
- if (val != DIE_NMI_IPI || !cpumask_test_cpu(cpu, mce_inject_cpumask))
+ if (val != DIE_NMI || !cpumask_test_cpu(cpu, mce_inject_cpumask))
return NOTIFY_DONE;
cpumask_clear_cpu(cpu, mce_inject_cpumask);
if (m->inject_flags & MCJ_EXCEPTION)
@@ -95,7 +96,7 @@ 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 48aa91f..7b91396 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -1229,7 +1229,7 @@ perf_event_nmi_handler(struct notifier_block *self,
return NOTIFY_DONE;

switch (cmd) {
- case DIE_NMI_IPI:
+ case DIE_NMI:
break;
case DIE_NMIUNKNOWN:
this_nmi = percpu_read(irq_stat.__nmi_count);
@@ -1279,7 +1279,7 @@ 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 cd21b65..1e8b583 100644
--- a/arch/x86/kernel/kgdb.c
+++ b/arch/x86/kernel/kgdb.c
@@ -525,10 +525,6 @@ static int __kgdb_notify(struct die_args *args, unsigned long cmd)
}
return NOTIFY_DONE;

- case DIE_NMI_IPI:
- /* Just ignore, we will handle the roundup on DIE_NMI. */
- return NOTIFY_DONE;
-
case DIE_NMIUNKNOWN:
if (was_in_debug_nmi[raw_smp_processor_id()]) {
was_in_debug_nmi[raw_smp_processor_id()] = 0;
@@ -606,7 +602,7 @@ 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 c495aa8..fc7aae1 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -18,6 +18,7 @@
#include <asm/pci_x86.h>
#include <asm/virtext.h>
#include <asm/cpu.h>
+#include <asm/nmi.h>

#ifdef CONFIG_X86_32
# include <linux/ctype.h>
@@ -747,7 +748,7 @@ static int crash_nmi_callback(struct notifier_block *self,
{
int cpu;

- if (val != DIE_NMI_IPI)
+ if (val != DIE_NMI)
return NOTIFY_OK;

cpu = raw_smp_processor_id();
@@ -778,6 +779,8 @@ 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/kernel/traps.c b/arch/x86/kernel/traps.c
index 5184b21..0668b51 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -404,8 +404,7 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs)
* CPU-specific NMI: send to specific CPU or NMI sources must
* be processed on specific CPU
*/
- if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, 0, 2, SIGINT)
- == NOTIFY_STOP)
+ if (notify_die(DIE_NMI, "nmi", regs, 0, 2, SIGINT) == NOTIFY_STOP)
return;

/* Non-CPU-specific NMI: NMI sources can be processed on any CPU */
@@ -414,9 +413,6 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs)
if (!cpu) {
reason = get_nmi_reason();
if (reason & NMI_REASON_MASK) {
- if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT)
- == NOTIFY_STOP)
- return;
if (reason & NMI_REASON_SERR)
pci_serr_error(reason, regs);
else if (reason & NMI_REASON_IOCHK)
@@ -431,8 +427,6 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs)
return;
}
}
- if (notify_die(DIE_NMI, "nmi", regs, 0, 2, SIGINT) == NOTIFY_STOP)
- return;
unknown_nmi_error(reason, regs);
}

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index ee7ff0e..e476044 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -64,7 +64,7 @@ static int profile_exceptions_notify(struct notifier_block *self,
int ret = NOTIFY_DONE;

switch (val) {
- case DIE_NMI_IPI:
+ case DIE_NMI:
if (ctr_running)
model->check_ctrs(args->regs, &__get_cpu_var(cpu_msrs));
else if (!nmi_enabled)
@@ -360,7 +360,7 @@ 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 fb40645..720bf5a 100644
--- a/arch/x86/oprofile/nmi_timer_int.c
+++ b/arch/x86/oprofile/nmi_timer_int.c
@@ -25,7 +25,7 @@ static int profile_timer_exceptions_notify(struct notifier_block *self,
int ret = NOTIFY_DONE;

switch (val) {
- case DIE_NMI_IPI:
+ case DIE_NMI:
oprofile_add_sample(args->regs, 0);
ret = NOTIFY_STOP;
break;
@@ -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)
--
1.7.3.2

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Next message: Don Zickus: "[PATCH 1/9] x86, NMI: Add NMI symbol constants and rename memory parity to PCI SERR"
Previous message: Don Zickus: "[PATCH 3/9] x86, NMI: Rewrite NMI handler"
In reply to: Don Zickus: "[PATCH 3/9] x86, NMI: Rewrite NMI handler"
Next in thread: Don Zickus: "[PATCH 1/9] x86, NMI: Add NMI symbol constants and rename memory parity to PCI SERR"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]