[RFC PATCH v5 0/4] arm/arm64: fix a migrating irq bug when hotplug cpu

[RFC PATCH v5 0/4] arm/arm64: fix a migrating irq bug when hotplug cpu

[Yang Yingliang]

----
Changes in v5:
 - remove the macro that guard move_irqs()
 - use name irq_migrate_all_off_this_cpu instead of move_irqs

Changes in v4:
 - select GENERIC_IRQ_MIGRATION when config GENERIC_IRQ_MIGRATION is enabled
 - add move_irqs() into kerne/irq/migration.c for cpu hotplug
 - use move_irqs() to migrate interrupts on arm/arm64 when cpu is going to down
 - split bugfix patch into seperate patches for arm and arm64

Changes in v3:
 - introduce config GENERIC_IRQ_MIGRATION for compiling migration.c
 - rename migrate_irqs in arch/ia64/kernel/irq.c to avoid compiling error

Changes in v2:
 - use the exiting helper to set IRQD_MOVE_PCNTXT flag
 - use for_each_active_irq() instead of for_each_irq_desc()
 - add some warn messages when affinity is null or do set affinity failed
----

Hi All,

There is a bug:

When cpu is disabled, all irqs will be migratged to another cpu.
In some cases, a new affinity is different, it needed to be coppied
to irq's affinity. But if the type of irq is LPI, it's affinity will
not be coppied because of irq_set_affinity's return value.



As Marc and Will suggested, I refactor the arm/arm64 migrating interrupts
code and fix the migrating irq bug while cpu is offline.

I'm trying let the core code do the migrating interrupts matter. kernel/irq/migration.c
depends on CONFIG_GENERIC_PENDING_IRQ, so I introduce config GENERIC_IRQ_MIGRATION for
compiling migration.c. Then add a generic function for migrating interrupts in migration.c.
And then use the function when cpu is going to down on arm and arm64.


Cc: Jiang Liu <jiang.liu@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Russell King - ARM Linux <linux@arm.linux.org.uk>
Cc: Hanjun Guo <hanjun.guo@linaro.org>

Yang Yingliang (4):
  genirq: introduce CONFIG_GENERIC_IRQ_MIGRATION
  genirq: add irq_migrate_all_off_this_cpu() for cpu hotplug
  arm64: fix a migrating irq bug when hotplug cpu
  arm: fix a migrating irq bug when hotplug cpu

 arch/arm/Kconfig             |  1 +
 arch/arm/include/asm/irq.h   |  1 -
 arch/arm/kernel/irq.c        | 62 -----------------------------------------
 arch/arm/kernel/smp.c        |  2 +-
 arch/arm64/Kconfig           |  1 +
 arch/arm64/include/asm/irq.h |  1 -
 arch/arm64/kernel/irq.c      | 62 -----------------------------------------
 arch/arm64/kernel/smp.c      |  2 +-
 include/linux/irq.h          |  2 ++
 kernel/irq/Kconfig           |  5 ++++
 kernel/irq/Makefile          |  2 +-
 kernel/irq/migration.c       | 66 ++++++++++++++++++++++++++++++++++++++++++++
 12 files changed, 78 insertions(+), 129 deletions(-)

-- 
2.5.0

[RFC PATCH v5 1/4] genirq: introduce CONFIG_GENERIC_IRQ_MIGRATION

[Yang Yingliang]

Introduce a more general config for compile kernel/irq/migration.c.
Move the CONFIG_GENERIC_PENDING_IRQ into migration.c. So we can
move other migration interrupts code into migration.c without
select CONFIG_GENERIC_PENDING_IRQ.

Cc: Jiang Liu <jiang.liu@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Russell King - ARM Linux <linux@arm.linux.org.uk>
Cc: Hanjun Guo <hanjun.guo@linaro.org>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
---
 kernel/irq/Kconfig     | 5 +++++
 kernel/irq/Makefile    | 2 +-
 kernel/irq/migration.c | 2 ++
 3 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/kernel/irq/Kconfig b/kernel/irq/Kconfig
index 9a76e3b..1ac0647 100644
--- a/kernel/irq/Kconfig
+++ b/kernel/irq/Kconfig
@@ -29,6 +29,11 @@ config GENERIC_IRQ_LEGACY_ALLOC_HWIRQ
 # Support for delayed migration from interrupt context
 config GENERIC_PENDING_IRQ
 	bool
+	select GENERIC_IRQ_MIGRATION
+
+# Support for generic irq migration
+config GENERIC_IRQ_MIGRATION
+	bool
 
 # Alpha specific irq affinity mechanism
 config AUTO_IRQ_AFFINITY
diff --git a/kernel/irq/Makefile b/kernel/irq/Makefile
index d121235..bdd31b7 100644
--- a/kernel/irq/Makefile
+++ b/kernel/irq/Makefile
@@ -4,6 +4,6 @@ obj-$(CONFIG_GENERIC_IRQ_CHIP) += generic-chip.o
 obj-$(CONFIG_GENERIC_IRQ_PROBE) += autoprobe.o
 obj-$(CONFIG_IRQ_DOMAIN) += irqdomain.o
 obj-$(CONFIG_PROC_FS) += proc.o
-obj-$(CONFIG_GENERIC_PENDING_IRQ) += migration.o
+obj-$(CONFIG_GENERIC_IRQ_MIGRATION) += migration.o
 obj-$(CONFIG_PM_SLEEP) += pm.o
 obj-$(CONFIG_GENERIC_MSI_IRQ) += msi.o
diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c
index 37ddb7b..1ff2b77 100644
--- a/kernel/irq/migration.c
+++ b/kernel/irq/migration.c
@@ -4,6 +4,7 @@
 
 #include "internals.h"
 
+#ifdef CONFIG_GENERIC_PENDING_IRQ
 void irq_move_masked_irq(struct irq_data *idata)
 {
 	struct irq_desc *desc = irq_data_to_desc(idata);
@@ -77,3 +78,4 @@ void irq_move_irq(struct irq_data *idata)
 	if (!masked)
 		idata->chip->irq_unmask(idata);
 }
+#endif
-- 
2.5.0

[RFC PATCH v5 1/4] genirq: introduce CONFIG_GENERIC_IRQ_MIGRATION

[Marc Zyngier]

On Sat, 19 Sep 2015 11:05:56 +0800
Yang Yingliang <yangyingliang@huawei.com> wrote:

> Introduce a more general config for compile kernel/irq/migration.c.
> Move the CONFIG_GENERIC_PENDING_IRQ into migration.c. So we can
> move other migration interrupts code into migration.c without
> select CONFIG_GENERIC_PENDING_IRQ.
> 
> Cc: Jiang Liu <jiang.liu@linux.intel.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Cc: Mark Rutland <mark.rutland@arm.com>
> Cc: Will Deacon <will.deacon@arm.com>
> Cc: Russell King - ARM Linux <linux@arm.linux.org.uk>
> Cc: Hanjun Guo <hanjun.guo@linaro.org>
> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

Acked-by: Marc Zyngier <marc.zyngier@arm.com>

	M.
-- 
Jazz is not dead. It just smells funny.

[RFC PATCH v5 2/4] genirq: add irq_migrate_all_off_this_cpu() for cpu hotplug

[Yang Yingliang]

Add irq_migrate_all_off_this_cpu() into kernel/irq/migration.c.
So we can use it to migrate interrupts, when cpu is offline.

Cc: Jiang Liu <jiang.liu@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Russell King - ARM Linux <linux@arm.linux.org.uk>
Cc: Hanjun Guo <hanjun.guo@linaro.org>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
---
 include/linux/irq.h    |  2 ++
 kernel/irq/migration.c | 64 ++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 66 insertions(+)

diff --git a/include/linux/irq.h b/include/linux/irq.h
index 6f8b340..c111fa1 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -438,6 +438,8 @@ extern int irq_set_affinity_locked(struct irq_data *data,
 				   const struct cpumask *cpumask, bool force);
 extern int irq_set_vcpu_affinity(unsigned int irq, void *vcpu_info);
 
+extern void irq_migrate_all_off_this_cpu(void);
+
 #if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_PENDING_IRQ)
 void irq_move_irq(struct irq_data *data);
 void irq_move_masked_irq(struct irq_data *data);
diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c
index 1ff2b77..6e3e38d 100644
--- a/kernel/irq/migration.c
+++ b/kernel/irq/migration.c
@@ -1,6 +1,7 @@
 
 #include <linux/irq.h>
 #include <linux/interrupt.h>
+#include <linux/ratelimit.h>
 
 #include "internals.h"
 
@@ -79,3 +80,66 @@ void irq_move_irq(struct irq_data *idata)
 		idata->chip->irq_unmask(idata);
 }
 #endif
+
+static bool migrate_one_irq(struct irq_desc *desc)
+{
+	struct irq_data *d = irq_desc_get_irq_data(desc);
+	const struct cpumask *affinity = d->affinity;
+	struct irq_chip *c;
+	bool ret = false;
+
+	/*
+	 * If this is a per-CPU interrupt, or the affinity does not
+	 * include this CPU, then we have nothing to do.
+	 */
+	if (irqd_is_per_cpu(d) || !cpumask_test_cpu(smp_processor_id(), affinity))
+		return false;
+
+	if (cpumask_any_and(affinity, cpu_online_mask) >= nr_cpu_ids) {
+		affinity = cpu_online_mask;
+		ret = true;
+	}
+
+	c = irq_data_get_irq_chip(d);
+	if (!c->irq_set_affinity) {
+		pr_warn_ratelimited("IRQ%u: unable to set affinity\n", d->irq);
+	} else {
+		int r = irq_do_set_affinity(d, affinity, false);
+		if (r)
+			pr_warn_ratelimited("IRQ%u: set affinity failed(%d).\n", d->irq, r);
+	}
+
+	return ret;
+}
+
+/*
+ * The current CPU has been marked offline.  Migrate IRQs off this CPU.
+ * If the affinity settings do not allow other CPUs, force them onto any
+ * available CPU.
+ *
+ * Note: we must iterate over all IRQs, whether they have an attached
+ * action structure or not, as we need to get chained interrupts too.
+ */
+void irq_migrate_all_off_this_cpu(void)
+{
+	unsigned int irq;
+	struct irq_desc *desc;
+	unsigned long flags;
+
+	local_irq_save(flags);
+
+	for_each_active_irq(irq) {
+		bool affinity_broken;
+
+		desc = irq_to_desc(irq);
+		raw_spin_lock(&desc->lock);
+		affinity_broken = migrate_one_irq(desc);
+		raw_spin_unlock(&desc->lock);
+
+		if (affinity_broken)
+			pr_warn_ratelimited("IRQ%u no longer affine to CPU%u\n",
+					    irq, smp_processor_id());
+	}
+
+	local_irq_restore(flags);
+}
-- 
2.5.0

[RFC PATCH v5 2/4] genirq: add irq_migrate_all_off_this_cpu() for cpu hotplug

[Marc Zyngier]

On Sat, 19 Sep 2015 11:05:57 +0800
Yang Yingliang <yangyingliang@huawei.com> wrote:

> Add irq_migrate_all_off_this_cpu() into kernel/irq/migration.c.
> So we can use it to migrate interrupts, when cpu is offline.
> 
> Cc: Jiang Liu <jiang.liu@linux.intel.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Cc: Mark Rutland <mark.rutland@arm.com>
> Cc: Will Deacon <will.deacon@arm.com>
> Cc: Russell King - ARM Linux <linux@arm.linux.org.uk>
> Cc: Hanjun Guo <hanjun.guo@linaro.org>
> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>

	M.
-- 
Jazz is not dead. It just smells funny.

[RFC PATCH v5 2/4] genirq: add irq_migrate_all_off_this_cpu() for cpu hotplug

[Thomas Gleixner]

On Sat, 19 Sep 2015, Yang Yingliang wrote:

> Add irq_migrate_all_off_this_cpu() into kernel/irq/migration.c.

This doesn't make any sense at all. 

You just reuse the existing file to stick your new code into it
without reusing a single bit in that file. Aside of that it's
unconditionally compiled, which means all existing users of
CONFIG_GENERIC_PENDING_IRQ are burdened with pointless code.

The right thing to do is:

Add that code to a new file: kernel/irq/cpuhotplug.c and make that
depend on CONFIG_GENERIC_IRQ_MIGRATION.

Thanks,

	tglx

[RFC PATCH v5 2/4] genirq: add irq_migrate_all_off_this_cpu() for cpu hotplug

[Yang Yingliang]

On 2015/9/23 2:54, Thomas Gleixner wrote:
> On Sat, 19 Sep 2015, Yang Yingliang wrote:
>
>> Add irq_migrate_all_off_this_cpu() into kernel/irq/migration.c.
>
> This doesn't make any sense at all.
>
> You just reuse the existing file to stick your new code into it
> without reusing a single bit in that file. Aside of that it's
> unconditionally compiled, which means all existing users of
> CONFIG_GENERIC_PENDING_IRQ are burdened with pointless code.
>
> The right thing to do is:
>
> Add that code to a new file: kernel/irq/cpuhotplug.c and make that
> depend on CONFIG_GENERIC_IRQ_MIGRATION.

How about add #ifdef CONFIG_CPU_HOTPLUG around new code
in kernel/irq/migration.c ?

Thanks,
Yang

[RFC PATCH v5 2/4] genirq: add irq_migrate_all_off_this_cpu() for cpu hotplug

[Thomas Gleixner]

On Wed, 23 Sep 2015, Yang Yingliang wrote:
> On 2015/9/23 2:54, Thomas Gleixner wrote:
> > On Sat, 19 Sep 2015, Yang Yingliang wrote:
> > 
> > > Add irq_migrate_all_off_this_cpu() into kernel/irq/migration.c.
> > 
> > This doesn't make any sense at all.
> > 
> > You just reuse the existing file to stick your new code into it
> > without reusing a single bit in that file. Aside of that it's
> > unconditionally compiled, which means all existing users of
> > CONFIG_GENERIC_PENDING_IRQ are burdened with pointless code.
> > 
> > The right thing to do is:
> > 
> > Add that code to a new file: kernel/irq/cpuhotplug.c and make that
> > depend on CONFIG_GENERIC_IRQ_MIGRATION.
> 
> How about add #ifdef CONFIG_CPU_HOTPLUG around new code
> in kernel/irq/migration.c ?

What's the point? Just to have useless churn and ifdeffery in that
file for nothing?

Thanks,

	tglx

[RFC PATCH v5 3/4] arm64: fix a migrating irq bug when hotplug cpu

[Yang Yingliang]

When cpu is disabled, all irqs will be migratged to another cpu.
In some cases, a new affinity is different, it needed to be coppied
to irq's affinity. But if the type of irq is LPI, it's affinity will
not be coppied because of irq_set_affinity's return value. Fix it by
using irq_do_set_affinity.

And migrating interrupts is a core code matter, so use the generic
function irq_migrate_all_off_this_cpu() to migrate interrupts in
kernel/irq/migration.c.

Cc: Jiang Liu <jiang.liu@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Russell King - ARM Linux <linux@arm.linux.org.uk>
Cc: Hanjun Guo <hanjun.guo@linaro.org>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
---
 arch/arm64/Kconfig           |  1 +
 arch/arm64/include/asm/irq.h |  1 -
 arch/arm64/kernel/irq.c      | 62 --------------------------------------------
 arch/arm64/kernel/smp.c      |  2 +-
 4 files changed, 2 insertions(+), 64 deletions(-)

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 7d95663..e0cac17 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -410,6 +410,7 @@ config NR_CPUS
 
 config HOTPLUG_CPU
 	bool "Support for hot-pluggable CPUs"
+	select GENERIC_IRQ_MIGRATION
 	help
 	  Say Y here to experiment with turning CPUs off and on.  CPUs
 	  can be controlled through /sys/devices/system/cpu.
diff --git a/arch/arm64/include/asm/irq.h b/arch/arm64/include/asm/irq.h
index bbb251b..0916929 100644
--- a/arch/arm64/include/asm/irq.h
+++ b/arch/arm64/include/asm/irq.h
@@ -7,7 +7,6 @@
 
 struct pt_regs;
 
-extern void migrate_irqs(void);
 extern void set_handle_irq(void (*handle_irq)(struct pt_regs *));
 
 static inline void acpi_irq_init(void)
diff --git a/arch/arm64/kernel/irq.c b/arch/arm64/kernel/irq.c
index 11dc3fd..9f17ec0 100644
--- a/arch/arm64/kernel/irq.c
+++ b/arch/arm64/kernel/irq.c
@@ -27,7 +27,6 @@
 #include <linux/init.h>
 #include <linux/irqchip.h>
 #include <linux/seq_file.h>
-#include <linux/ratelimit.h>
 
 unsigned long irq_err_count;
 
@@ -54,64 +53,3 @@ void __init init_IRQ(void)
 	if (!handle_arch_irq)
 		panic("No interrupt controller found.");
 }
-
-#ifdef CONFIG_HOTPLUG_CPU
-static bool migrate_one_irq(struct irq_desc *desc)
-{
-	struct irq_data *d = irq_desc_get_irq_data(desc);
-	const struct cpumask *affinity = irq_data_get_affinity_mask(d);
-	struct irq_chip *c;
-	bool ret = false;
-
-	/*
-	 * If this is a per-CPU interrupt, or the affinity does not
-	 * include this CPU, then we have nothing to do.
-	 */
-	if (irqd_is_per_cpu(d) || !cpumask_test_cpu(smp_processor_id(), affinity))
-		return false;
-
-	if (cpumask_any_and(affinity, cpu_online_mask) >= nr_cpu_ids) {
-		affinity = cpu_online_mask;
-		ret = true;
-	}
-
-	c = irq_data_get_irq_chip(d);
-	if (!c->irq_set_affinity)
-		pr_debug("IRQ%u: unable to set affinity\n", d->irq);
-	else if (c->irq_set_affinity(d, affinity, false) == IRQ_SET_MASK_OK && ret)
-		cpumask_copy(irq_data_get_affinity_mask(d), affinity);
-
-	return ret;
-}
-
-/*
- * The current CPU has been marked offline.  Migrate IRQs off this CPU.
- * If the affinity settings do not allow other CPUs, force them onto any
- * available CPU.
- *
- * Note: we must iterate over all IRQs, whether they have an attached
- * action structure or not, as we need to get chained interrupts too.
- */
-void migrate_irqs(void)
-{
-	unsigned int i;
-	struct irq_desc *desc;
-	unsigned long flags;
-
-	local_irq_save(flags);
-
-	for_each_irq_desc(i, desc) {
-		bool affinity_broken;
-
-		raw_spin_lock(&desc->lock);
-		affinity_broken = migrate_one_irq(desc);
-		raw_spin_unlock(&desc->lock);
-
-		if (affinity_broken)
-			pr_warn_ratelimited("IRQ%u no longer affine to CPU%u\n",
-					    i, smp_processor_id());
-	}
-
-	local_irq_restore(flags);
-}
-#endif /* CONFIG_HOTPLUG_CPU */
diff --git a/arch/arm64/kernel/smp.c b/arch/arm64/kernel/smp.c
index dbdaacd..ee5cafa 100644
--- a/arch/arm64/kernel/smp.c
+++ b/arch/arm64/kernel/smp.c
@@ -232,7 +232,7 @@ int __cpu_disable(void)
 	/*
 	 * OK - migrate IRQs away from this CPU
 	 */
-	migrate_irqs();
+	irq_migrate_all_off_this_cpu();
 
 	/*
 	 * Remove this CPU from the vm mask set of all processes.
-- 
2.5.0

[RFC PATCH v5 3/4] arm64: fix a migrating irq bug when hotplug cpu

[Marc Zyngier]

On Sat, 19 Sep 2015 11:05:58 +0800
Yang Yingliang <yangyingliang@huawei.com> wrote:

> When cpu is disabled, all irqs will be migratged to another cpu.
> In some cases, a new affinity is different, it needed to be coppied
> to irq's affinity. But if the type of irq is LPI, it's affinity will
> not be coppied because of irq_set_affinity's return value. Fix it by
> using irq_do_set_affinity.

Nit: This hasn't much to do with LPIs proper, but with the fact that
the architecture code only has a partial knowledge of the possible
affinity setting return values.

> And migrating interrupts is a core code matter, so use the generic
> function irq_migrate_all_off_this_cpu() to migrate interrupts in
> kernel/irq/migration.c.
> 
> Cc: Jiang Liu <jiang.liu@linux.intel.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Cc: Mark Rutland <mark.rutland@arm.com>
> Cc: Will Deacon <will.deacon@arm.com>
> Cc: Russell King - ARM Linux <linux@arm.linux.org.uk>
> Cc: Hanjun Guo <hanjun.guo@linaro.org>
> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

Otherwise Acked-by: Marc Zyngier <marc.zyngier@arm.com>

	M.
-- 
Jazz is not dead. It just smells funny.

[RFC PATCH v5 4/4] arm: fix a migrating irq bug when hotplug cpu

[Yang Yingliang]

When cpu is disabled, all irqs will be migratged to another cpu.
In some cases, a new affinity is different, it needed to be coppied
to irq's affinity. But if the type of irq is LPI, it's affinity will
not be coppied because of irq_set_affinity's return value. Fix it by
using irq_do_set_affinity.

And migrating interrupts is a core code matter, so use the generic
function irq_migrate_all_off_this_cpu() to migrate interrupts in
kernel/irq/migration.c.

Cc: Jiang Liu <jiang.liu@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Russell King - ARM Linux <linux@arm.linux.org.uk>
Cc: Hanjun Guo <hanjun.guo@linaro.org>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
---
 arch/arm/Kconfig           |  1 +
 arch/arm/include/asm/irq.h |  1 -
 arch/arm/kernel/irq.c      | 62 ----------------------------------------------
 arch/arm/kernel/smp.c      |  2 +-
 4 files changed, 2 insertions(+), 64 deletions(-)

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 72ad724..bffba78 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -1492,6 +1492,7 @@ config NR_CPUS
 config HOTPLUG_CPU
 	bool "Support for hot-pluggable CPUs"
 	depends on SMP
+	select GENERIC_IRQ_MIGRATION
 	help
 	  Say Y here to experiment with turning CPUs off and on.  CPUs
 	  can be controlled through /sys/devices/system/cpu.
diff --git a/arch/arm/include/asm/irq.h b/arch/arm/include/asm/irq.h
index be1d07d..882bf98 100644
--- a/arch/arm/include/asm/irq.h
+++ b/arch/arm/include/asm/irq.h
@@ -24,7 +24,6 @@
 #ifndef __ASSEMBLY__
 struct irqaction;
 struct pt_regs;
-extern void migrate_irqs(void);
 
 extern void asm_do_IRQ(unsigned int, struct pt_regs *);
 void handle_IRQ(unsigned int, struct pt_regs *);
diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c
index 5ff4826..78abea8 100644
--- a/arch/arm/kernel/irq.c
+++ b/arch/arm/kernel/irq.c
@@ -31,7 +31,6 @@
 #include <linux/smp.h>
 #include <linux/init.h>
 #include <linux/seq_file.h>
-#include <linux/ratelimit.h>
 #include <linux/errno.h>
 #include <linux/list.h>
 #include <linux/kallsyms.h>
@@ -136,64 +135,3 @@ int __init arch_probe_nr_irqs(void)
 	return nr_irqs;
 }
 #endif
-
-#ifdef CONFIG_HOTPLUG_CPU
-static bool migrate_one_irq(struct irq_desc *desc)
-{
-	struct irq_data *d = irq_desc_get_irq_data(desc);
-	const struct cpumask *affinity = irq_data_get_affinity_mask(d);
-	struct irq_chip *c;
-	bool ret = false;
-
-	/*
-	 * If this is a per-CPU interrupt, or the affinity does not
-	 * include this CPU, then we have nothing to do.
-	 */
-	if (irqd_is_per_cpu(d) || !cpumask_test_cpu(smp_processor_id(), affinity))
-		return false;
-
-	if (cpumask_any_and(affinity, cpu_online_mask) >= nr_cpu_ids) {
-		affinity = cpu_online_mask;
-		ret = true;
-	}
-
-	c = irq_data_get_irq_chip(d);
-	if (!c->irq_set_affinity)
-		pr_debug("IRQ%u: unable to set affinity\n", d->irq);
-	else if (c->irq_set_affinity(d, affinity, false) == IRQ_SET_MASK_OK && ret)
-		cpumask_copy(irq_data_get_affinity_mask(d), affinity);
-
-	return ret;
-}
-
-/*
- * The current CPU has been marked offline.  Migrate IRQs off this CPU.
- * If the affinity settings do not allow other CPUs, force them onto any
- * available CPU.
- *
- * Note: we must iterate over all IRQs, whether they have an attached
- * action structure or not, as we need to get chained interrupts too.
- */
-void migrate_irqs(void)
-{
-	unsigned int i;
-	struct irq_desc *desc;
-	unsigned long flags;
-
-	local_irq_save(flags);
-
-	for_each_irq_desc(i, desc) {
-		bool affinity_broken;
-
-		raw_spin_lock(&desc->lock);
-		affinity_broken = migrate_one_irq(desc);
-		raw_spin_unlock(&desc->lock);
-
-		if (affinity_broken)
-			pr_warn_ratelimited("IRQ%u no longer affine to CPU%u\n",
-				i, smp_processor_id());
-	}
-
-	local_irq_restore(flags);
-}
-#endif /* CONFIG_HOTPLUG_CPU */
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 48185a7..fc26384 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -218,7 +218,7 @@ int __cpu_disable(void)
 	/*
 	 * OK - migrate IRQs away from this CPU
 	 */
-	migrate_irqs();
+	irq_migrate_all_off_this_cpu();
 
 	/*
 	 * Flush user cache and TLB mappings, and then remove this CPU
-- 
2.5.0

[RFC PATCH v5 4/4] arm: fix a migrating irq bug when hotplug cpu

[Marc Zyngier]

On Sat, 19 Sep 2015 11:05:59 +0800
Yang Yingliang <yangyingliang@huawei.com> wrote:

> When cpu is disabled, all irqs will be migratged to another cpu.
> In some cases, a new affinity is different, it needed to be coppied
> to irq's affinity. But if the type of irq is LPI, it's affinity will
> not be coppied because of irq_set_affinity's return value. Fix it by
> using irq_do_set_affinity.
> 
> And migrating interrupts is a core code matter, so use the generic
> function irq_migrate_all_off_this_cpu() to migrate interrupts in
> kernel/irq/migration.c.
> 
> Cc: Jiang Liu <jiang.liu@linux.intel.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Cc: Mark Rutland <mark.rutland@arm.com>
> Cc: Will Deacon <will.deacon@arm.com>
> Cc: Russell King - ARM Linux <linux@arm.linux.org.uk>
> Cc: Hanjun Guo <hanjun.guo@linaro.org>
> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>

Acked-by: Marc Zyngier <marc.zyngier@arm.com>

	M.
-- 
Jazz is not dead. It just smells funny.

[RFC PATCH v5 0/4] arm/arm64: fix a migrating irq bug when hotplug cpu

[Will Deacon]

On Sat, Sep 19, 2015 at 04:05:55AM +0100, Yang Yingliang wrote:
> ----
> Changes in v5:
>  - remove the macro that guard move_irqs()
>  - use name irq_migrate_all_off_this_cpu instead of move_irqs
> 
> Changes in v4:
>  - select GENERIC_IRQ_MIGRATION when config GENERIC_IRQ_MIGRATION is enabled
>  - add move_irqs() into kerne/irq/migration.c for cpu hotplug
>  - use move_irqs() to migrate interrupts on arm/arm64 when cpu is going to down
>  - split bugfix patch into seperate patches for arm and arm64
> 
> Changes in v3:
>  - introduce config GENERIC_IRQ_MIGRATION for compiling migration.c
>  - rename migrate_irqs in arch/ia64/kernel/irq.c to avoid compiling error
> 
> Changes in v2:
>  - use the exiting helper to set IRQD_MOVE_PCNTXT flag
>  - use for_each_active_irq() instead of for_each_irq_desc()
>  - add some warn messages when affinity is null or do set affinity failed
> ----
> 
> Hi All,
> 
> There is a bug:
> 
> When cpu is disabled, all irqs will be migratged to another cpu.
> In some cases, a new affinity is different, it needed to be coppied
> to irq's affinity. But if the type of irq is LPI, it's affinity will
> not be coppied because of irq_set_affinity's return value.

This series looks good to me:

  Reviewed-by: Will Deacon <will.deacon@arm.com>

Marc: once you're happy with the irq/core changes, feel free to include
the arm64 patch in your tree.

Will