[PATCH 0/8] Make nmi_shootdown_cpus() usable by non-kdump code

[PATCH 0/8] Make nmi_shootdown_cpus() usable by non-kdump code

[Eduardo Habkost]

Hi, Ingo,

As tip/master is a moving target, I am splitting the previous kdump/reboot
virtualization-disable code series[1] into smaller series so the simpler
parts can be included sooner. This first series is just for making
nmi_shootdown_cpus() usable by non-kdump code, by moving their generic
parts to reboot.c.

They are exactly the same patches I've sent as 01-08 in the previous
series.

The NMI IPIs method used by nmi_shootdown_cpus() is the only way
we can make sure VMX is disabled on all CPUs before rebooting, on
emergency_restart(), because CPUs may have IRQs disabled. That's why I
will reuse nmi_shootdown_cpus() for the emergency_restart code when
VMX is enabled.

The virtualization-specific parts should come soon, after I implement
the suggestions by Eric and Avi, by making the virtualization-disabling
functions inline so they can be used directly by the core code (instead
of using the function pointer tricks from my previous series).

[1] http://marc.info/?l=linux-kernel&m=122591590923455&w=2

-- 
Eduardo

[PATCH 1/8] x86 kdump: Extract kdump-specific code from crash_nmi_callback()

[Eduardo Habkost]

The NMI CPU-halting code will be used on non-kdump cases, also
(e.g. emergency_reboot when virtualization is enabled).

Signed-off-by: Eduardo Habkost <ehabkost-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
---
 arch/x86/kernel/crash.c |   31 ++++++++++++++++++++-----------
 1 files changed, 20 insertions(+), 11 deletions(-)

diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
index 2685538..6047542 100644
--- a/arch/x86/kernel/crash.c
+++ b/arch/x86/kernel/crash.c
@@ -35,19 +35,34 @@ static int crashing_cpu;
 #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)
 static atomic_t waiting_for_crash_ipi;
 
-static int crash_nmi_callback(struct notifier_block *self,
-			unsigned long val, void *data)
+static void kdump_nmi_callback(int cpu, struct die_args *args)
 {
 	struct pt_regs *regs;
 #ifdef CONFIG_X86_32
 	struct pt_regs fixed_regs;
 #endif
+
+	regs = args->regs;
+
+#ifdef CONFIG_X86_32
+	if (!user_mode_vm(regs)) {
+		crash_fixup_ss_esp(&fixed_regs, regs);
+		regs = &fixed_regs;
+	}
+#endif
+	crash_save_cpu(regs, cpu);
+
+	disable_local_APIC();
+}
+
+static int crash_nmi_callback(struct notifier_block *self,
+			unsigned long val, void *data)
+{
 	int cpu;
 
 	if (val != DIE_NMI_IPI)
 		return NOTIFY_OK;
 
-	regs = ((struct die_args *)data)->regs;
 	cpu = raw_smp_processor_id();
 
 	/* Don't do anything if this handler is invoked on crashing cpu.
@@ -58,14 +73,8 @@ static int crash_nmi_callback(struct notifier_block *self,
 		return NOTIFY_STOP;
 	local_irq_disable();
 
-#ifdef CONFIG_X86_32
-	if (!user_mode_vm(regs)) {
-		crash_fixup_ss_esp(&fixed_regs, regs);
-		regs = &fixed_regs;
-	}
-#endif
-	crash_save_cpu(regs, cpu);
-	disable_local_APIC();
+	kdump_nmi_callback(cpu, (struct die_args *)data);
+
 	atomic_dec(&waiting_for_crash_ipi);
 	/* Assume hlt works */
 	halt();
-- 
1.5.5.GIT

[PATCH 2/8] x86 kdump: Move crashing_cpu assignment to nmi_shootdown_cpus()

[Eduardo Habkost]

This variable will be moved to non-kdump-specific code.

Signed-off-by: Eduardo Habkost <ehabkost-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
---
 arch/x86/kernel/crash.c |    8 +++++---
 1 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
index 6047542..ed2f0f9 100644
--- a/arch/x86/kernel/crash.c
+++ b/arch/x86/kernel/crash.c
@@ -29,10 +29,11 @@
 
 #include <mach_ipi.h>
 
+#if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)
+
 /* This keeps a track of which one is crashing cpu. */
 static int crashing_cpu;
 
-#if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)
 static atomic_t waiting_for_crash_ipi;
 
 static void kdump_nmi_callback(int cpu, struct die_args *args)
@@ -100,6 +101,9 @@ static void nmi_shootdown_cpus(void)
 {
 	unsigned long msecs;
 
+	/* Make a note of crashing cpu. Will be used in NMI callback.*/
+	crashing_cpu = safe_smp_processor_id();
+
 	atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
 	/* Would it be better to replace the trap vector here? */
 	if (register_die_notifier(&crash_nmi_nb))
@@ -140,8 +144,6 @@ void native_machine_crash_shutdown(struct pt_regs *regs)
 	/* The kernel is broken so disable interrupts */
 	local_irq_disable();
 
-	/* Make a note of crashing cpu. Will be used in NMI callback.*/
-	crashing_cpu = safe_smp_processor_id();
 	nmi_shootdown_cpus();
 	lapic_shutdown();
 #if defined(CONFIG_X86_IO_APIC)
-- 
1.5.5.GIT

[PATCH 3/8] x86 kdump: Create kdump_nmi_shootdown_cpus()

[Eduardo Habkost]

For the kdump-specific code that was living on nmi_shootdown_cpus().

Signed-off-by: Eduardo Habkost <ehabkost-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
---
 arch/x86/kernel/crash.c |   11 +++++++++--
 1 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
index ed2f0f9..75c468c 100644
--- a/arch/x86/kernel/crash.c
+++ b/arch/x86/kernel/crash.c
@@ -122,10 +122,17 @@ static void nmi_shootdown_cpus(void)
 	}
 
 	/* Leave the nmi callback set */
+}
+
+static void kdump_nmi_shootdown_cpus(void)
+{
+	nmi_shootdown_cpus();
+
 	disable_local_APIC();
 }
+
 #else
-static void nmi_shootdown_cpus(void)
+static void kdump_nmi_shootdown_cpus(void)
 {
 	/* There are no cpus to shootdown */
 }
@@ -144,7 +151,7 @@ void native_machine_crash_shutdown(struct pt_regs *regs)
 	/* The kernel is broken so disable interrupts */
 	local_irq_disable();
 
-	nmi_shootdown_cpus();
+	kdump_nmi_shootdown_cpus();
 	lapic_shutdown();
 #if defined(CONFIG_X86_IO_APIC)
 	disable_IO_APIC();
-- 
1.5.5.GIT

[PATCH 4/8] x86 kdump: Make kdump_nmi_callback() a function ptr on crash_nmi_callback()

[Eduardo Habkost]

The reboot code will use a different function on crash_nmi_callback().
Adding a function pointer parameter to nmi_shootdown_cpus() for that.

Signed-off-by: Eduardo Habkost <ehabkost-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
---
 arch/x86/kernel/crash.c |   11 ++++++++---
 1 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
index 75c468c..f23c2be 100644
--- a/arch/x86/kernel/crash.c
+++ b/arch/x86/kernel/crash.c
@@ -29,10 +29,13 @@
 
 #include <mach_ipi.h>
 
+typedef void (*nmi_shootdown_cb)(int, struct die_args*);
+
 #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)
 
 /* This keeps a track of which one is crashing cpu. */
 static int crashing_cpu;
+static nmi_shootdown_cb shootdown_callback;
 
 static atomic_t waiting_for_crash_ipi;
 
@@ -74,7 +77,7 @@ static int crash_nmi_callback(struct notifier_block *self,
 		return NOTIFY_STOP;
 	local_irq_disable();
 
-	kdump_nmi_callback(cpu, (struct die_args *)data);
+	shootdown_callback(cpu, (struct die_args *)data);
 
 	atomic_dec(&waiting_for_crash_ipi);
 	/* Assume hlt works */
@@ -97,13 +100,15 @@ static struct notifier_block crash_nmi_nb = {
 	.notifier_call = crash_nmi_callback,
 };
 
-static void nmi_shootdown_cpus(void)
+static void nmi_shootdown_cpus(nmi_shootdown_cb callback)
 {
 	unsigned long msecs;
 
 	/* Make a note of crashing cpu. Will be used in NMI callback.*/
 	crashing_cpu = safe_smp_processor_id();
 
+	shootdown_callback = callback;
+
 	atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
 	/* Would it be better to replace the trap vector here? */
 	if (register_die_notifier(&crash_nmi_nb))
@@ -126,7 +131,7 @@ static void nmi_shootdown_cpus(void)
 
 static void kdump_nmi_shootdown_cpus(void)
 {
-	nmi_shootdown_cpus();
+	nmi_shootdown_cpus(kdump_nmi_callback);
 
 	disable_local_APIC();
 }
-- 
1.5.5.GIT

[PATCH 5/8] x86 kdump: Make nmi_shootdown_cpus() non-static

[Eduardo Habkost]

Add prototype to asm/reboot.h.

Signed-off-by: Eduardo Habkost <ehabkost-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
---
 arch/x86/include/asm/reboot.h |    5 +++++
 arch/x86/kernel/crash.c       |    3 +--
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/arch/x86/include/asm/reboot.h b/arch/x86/include/asm/reboot.h
index df77103..562d4fd 100644
--- a/arch/x86/include/asm/reboot.h
+++ b/arch/x86/include/asm/reboot.h
@@ -1,6 +1,8 @@
 #ifndef _ASM_X86_REBOOT_H
 #define _ASM_X86_REBOOT_H
 
+#include <linux/kdebug.h>
+
 struct pt_regs;
 
 struct machine_ops {
@@ -18,4 +20,7 @@ void native_machine_crash_shutdown(struct pt_regs *regs);
 void native_machine_shutdown(void);
 void machine_real_restart(const unsigned char *code, int length);
 
+typedef void (*nmi_shootdown_cb)(int, struct die_args*);
+void nmi_shootdown_cpus(nmi_shootdown_cb callback);
+
 #endif /* _ASM_X86_REBOOT_H */
diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
index f23c2be..fb298d1 100644
--- a/arch/x86/kernel/crash.c
+++ b/arch/x86/kernel/crash.c
@@ -29,7 +29,6 @@
 
 #include <mach_ipi.h>
 
-typedef void (*nmi_shootdown_cb)(int, struct die_args*);
 
 #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)
 
@@ -100,7 +99,7 @@ static struct notifier_block crash_nmi_nb = {
 	.notifier_call = crash_nmi_callback,
 };
 
-static void nmi_shootdown_cpus(nmi_shootdown_cb callback)
+void nmi_shootdown_cpus(nmi_shootdown_cb callback)
 {
 	unsigned long msecs;
 
-- 
1.5.5.GIT

[PATCH 6/8] x86: Move nmi_shootdown_cpus() to reboot.c

[Eduardo Habkost]

Now nmi_shootdown_cpus() is ready to be used by non-kdump code also.
Move it to reboot.c.

Signed-off-by: Eduardo Habkost <ehabkost-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
---
 arch/x86/kernel/crash.c  |   76 ------------------------------------------
 arch/x86/kernel/reboot.c |   83 ++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 83 insertions(+), 76 deletions(-)

diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c
index fb298d1..d84a852 100644
--- a/arch/x86/kernel/crash.c
+++ b/arch/x86/kernel/crash.c
@@ -32,12 +32,6 @@
 
 #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)
 
-/* This keeps a track of which one is crashing cpu. */
-static int crashing_cpu;
-static nmi_shootdown_cb shootdown_callback;
-
-static atomic_t waiting_for_crash_ipi;
-
 static void kdump_nmi_callback(int cpu, struct die_args *args)
 {
 	struct pt_regs *regs;
@@ -58,76 +52,6 @@ static void kdump_nmi_callback(int cpu, struct die_args *args)
 	disable_local_APIC();
 }
 
-static int crash_nmi_callback(struct notifier_block *self,
-			unsigned long val, void *data)
-{
-	int cpu;
-
-	if (val != DIE_NMI_IPI)
-		return NOTIFY_OK;
-
-	cpu = raw_smp_processor_id();
-
-	/* Don't do anything if this handler is invoked on crashing cpu.
-	 * Otherwise, system will completely hang. Crashing cpu can get
-	 * an NMI if system was initially booted with nmi_watchdog parameter.
-	 */
-	if (cpu == crashing_cpu)
-		return NOTIFY_STOP;
-	local_irq_disable();
-
-	shootdown_callback(cpu, (struct die_args *)data);
-
-	atomic_dec(&waiting_for_crash_ipi);
-	/* Assume hlt works */
-	halt();
-	for (;;)
-		cpu_relax();
-
-	return 1;
-}
-
-static void smp_send_nmi_allbutself(void)
-{
-	cpumask_t mask = cpu_online_map;
-	cpu_clear(safe_smp_processor_id(), mask);
-	if (!cpus_empty(mask))
-		send_IPI_mask(mask, NMI_VECTOR);
-}
-
-static struct notifier_block crash_nmi_nb = {
-	.notifier_call = crash_nmi_callback,
-};
-
-void nmi_shootdown_cpus(nmi_shootdown_cb callback)
-{
-	unsigned long msecs;
-
-	/* Make a note of crashing cpu. Will be used in NMI callback.*/
-	crashing_cpu = safe_smp_processor_id();
-
-	shootdown_callback = callback;
-
-	atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
-	/* Would it be better to replace the trap vector here? */
-	if (register_die_notifier(&crash_nmi_nb))
-		return;		/* return what? */
-	/* Ensure the new callback function is set before sending
-	 * out the NMI
-	 */
-	wmb();
-
-	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) {
-		mdelay(1);
-		msecs--;
-	}
-
-	/* Leave the nmi callback set */
-}
-
 static void kdump_nmi_shootdown_cpus(void)
 {
 	nmi_shootdown_cpus(kdump_nmi_callback);
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index 34f8d37..82fa28f 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -21,6 +21,9 @@
 # include <asm/iommu.h>
 #endif
 
+#include <mach_ipi.h>
+
+
 /*
  * Power off function, if any
  */
@@ -519,3 +522,83 @@ void machine_crash_shutdown(struct pt_regs *regs)
 	machine_ops.crash_shutdown(regs);
 }
 #endif
+
+
+#if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)
+
+/* This keeps a track of which one is crashing cpu. */
+static int crashing_cpu;
+static nmi_shootdown_cb shootdown_callback;
+
+static atomic_t waiting_for_crash_ipi;
+
+static int crash_nmi_callback(struct notifier_block *self,
+			unsigned long val, void *data)
+{
+	int cpu;
+
+	if (val != DIE_NMI_IPI)
+		return NOTIFY_OK;
+
+	cpu = raw_smp_processor_id();
+
+	/* Don't do anything if this handler is invoked on crashing cpu.
+	 * Otherwise, system will completely hang. Crashing cpu can get
+	 * an NMI if system was initially booted with nmi_watchdog parameter.
+	 */
+	if (cpu == crashing_cpu)
+		return NOTIFY_STOP;
+	local_irq_disable();
+
+	shootdown_callback(cpu, (struct die_args *)data);
+
+	atomic_dec(&waiting_for_crash_ipi);
+	/* Assume hlt works */
+	halt();
+	for (;;)
+		cpu_relax();
+
+	return 1;
+}
+
+static void smp_send_nmi_allbutself(void)
+{
+	cpumask_t mask = cpu_online_map;
+	cpu_clear(safe_smp_processor_id(), mask);
+	if (!cpus_empty(mask))
+		send_IPI_mask(mask, NMI_VECTOR);
+}
+
+static struct notifier_block crash_nmi_nb = {
+	.notifier_call = crash_nmi_callback,
+};
+
+void nmi_shootdown_cpus(nmi_shootdown_cb callback)
+{
+	unsigned long msecs;
+
+	/* Make a note of crashing cpu. Will be used in NMI callback.*/
+	crashing_cpu = safe_smp_processor_id();
+
+	shootdown_callback = callback;
+
+	atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
+	/* Would it be better to replace the trap vector here? */
+	if (register_die_notifier(&crash_nmi_nb))
+		return;		/* return what? */
+	/* Ensure the new callback function is set before sending
+	 * out the NMI
+	 */
+	wmb();
+
+	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) {
+		mdelay(1);
+		msecs--;
+	}
+
+	/* Leave the nmi callback set */
+}
+#endif
-- 
1.5.5.GIT

[PATCH 7/8] x86: Make nmi_shootdown_cpus() available on !SMP and !X86_LOCAL_APIC

[Eduardo Habkost]

The X86_LOCAL_APIC #ifdef was for kdump. For !SMP, the function simply
does nothing.

Signed-off-by: Eduardo Habkost <ehabkost-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
---
 arch/x86/kernel/reboot.c |   13 ++++++++++++-
 1 files changed, 12 insertions(+), 1 deletions(-)

diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index 82fa28f..d0c9c77 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -524,7 +524,7 @@ void machine_crash_shutdown(struct pt_regs *regs)
 #endif
 
 
-#if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)
+#if defined(CONFIG_SMP)
 
 /* This keeps a track of which one is crashing cpu. */
 static int crashing_cpu;
@@ -573,6 +573,12 @@ static struct notifier_block crash_nmi_nb = {
 	.notifier_call = crash_nmi_callback,
 };
 
+/* Halt all other CPUs, calling the specified function on each of them
+ *
+ * This function can be used to halt all other CPUs on crash
+ * or emergency reboot time. The function passed as parameter
+ * will be called inside a NMI handler on all CPUs.
+ */
 void nmi_shootdown_cpus(nmi_shootdown_cb callback)
 {
 	unsigned long msecs;
@@ -601,4 +607,9 @@ void nmi_shootdown_cpus(nmi_shootdown_cb callback)
 
 	/* Leave the nmi callback set */
 }
+#else /* !CONFIG_SMP */
+void nmi_shootdown_cpus(nmi_shootdown_cb callback)
+{
+	/* No other CPUs to shoot down */
+}
 #endif
-- 
1.5.5.GIT

[PATCH 8/8] x86: Disable IRQs before doing anything on nmi_shootdown_cpus()

[Eduardo Habkost]

We need to know on which CPU we are running on, and we don't want to be
preempted while doing this.

Signed-off-by: Eduardo Habkost <ehabkost-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
---
 arch/x86/kernel/reboot.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index d0c9c77..c495687 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -582,6 +582,7 @@ static struct notifier_block crash_nmi_nb = {
 void nmi_shootdown_cpus(nmi_shootdown_cb callback)
 {
 	unsigned long msecs;
+	local_irq_disable();
 
 	/* Make a note of crashing cpu. Will be used in NMI callback.*/
 	crashing_cpu = safe_smp_processor_id();
-- 
1.5.5.GIT

Re: [PATCH 0/8] Make nmi_shootdown_cpus() usable by non-kdump code

[Ingo Molnar]

* Eduardo Habkost <ehabkost-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> wrote:

> Hi, Ingo,
> 
> As tip/master is a moving target, I am splitting the previous 
> kdump/reboot virtualization-disable code series[1] into smaller 
> series so the simpler parts can be included sooner. This first 
> series is just for making nmi_shootdown_cpus() usable by non-kdump 
> code, by moving their generic parts to reboot.c.

applied them to tip/x86/crashdump, thanks Eduardo!

this branch btw. is an append-only non-rebasing tree that is not a 
moving target, you can access it via:

  http://people.redhat.com/mingo/tip.git/README

and then do:

  git checkout -b x86/crashdump tip/x86/crashdump

it has the following changes queued up currently:

Eduardo Habkost (8):
      x86 kdump: extract kdump-specific code from crash_nmi_callback()
      x86 kdump: move crashing_cpu assignment to nmi_shootdown_cpus()
      x86 kdump: create kdump_nmi_shootdown_cpus()
      x86 kdump: make kdump_nmi_callback() a function ptr on crash_nmi_callback()
      x86 kdump: make nmi_shootdown_cpus() non-static
      x86: move nmi_shootdown_cpus() to reboot.c
      x86: make nmi_shootdown_cpus() available on !SMP and !X86_LOCAL_APIC
      x86: disable IRQs before doing anything on nmi_shootdown_cpus()

Huang Ying (3):
      kexec/i386: remove PAGE_SIZE alignment from relocate_kernel
      kexec/i386: allocate page table pages dynamically
      kexec/i386: setup kexec page table in C

if you prefer less flux then you can send patches against that branch, 
we'll sort out interaction with other trees.

	Ingo