[PATCH v3] genirq/cpuhotplug: notify about irq affinity change for offlined cpus.

[Imran Khan <imran.f.khan@oracle.com>]

During cpu offlining the irqs affined to that cpu are moved
to other cpu, but this affinity change is not accounted for by
irq_desc::affinity_notify (if available).
This can leave users, of irq_set_affinity_notifier, with old
affinity information.
Avoid this by allowing to schedule affinity change notification
work for irqs that were affined to the cpu being offlined.

Also since irq_set_affinity_locked uses the same logic to
schedule affinity change notification work, split out this
logic in a dedicated function and use that at both the places.

Signed-off-by: Imran Khan <imran.f.khan@oracle.com>
---
v2 -> v3:
 - Address review comments from Thomas
 - Removed RFC tag
v1 -> v2:
 - Fix compilation error due to missed parenthesis around scoped_guard

 include/linux/interrupt.h |  2 ++
 kernel/irq/cpuhotplug.c   |  6 ++++--
 kernel/irq/manage.c       | 28 ++++++++++++++++++++--------
 3 files changed, 26 insertions(+), 10 deletions(-)

diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index 266f2b39213a0..6f93bf524f728 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -383,6 +383,7 @@ irq_create_affinity_masks(unsigned int nvec, struct irq_affinity *affd);
 unsigned int irq_calc_affinity_vectors(unsigned int minvec, unsigned int maxvec,
 				       const struct irq_affinity *affd);
 
+extern void schedule_affinity_notify_work(struct irq_desc *desc);
 #else /* CONFIG_SMP */
 
 static inline int irq_set_affinity(unsigned int irq, const struct cpumask *m)
@@ -445,6 +446,7 @@ irq_calc_affinity_vectors(unsigned int minvec, unsigned int maxvec,
 	return maxvec;
 }
 
+static inline void schedule_affinity_notify_work(struct irq_desc *desc) { }
 #endif /* CONFIG_SMP */
 
 /*
diff --git a/kernel/irq/cpuhotplug.c b/kernel/irq/cpuhotplug.c
index 755346ea98196..c77a3b95fba85 100644
--- a/kernel/irq/cpuhotplug.c
+++ b/kernel/irq/cpuhotplug.c
@@ -177,9 +177,11 @@ void irq_migrate_all_off_this_cpu(void)
 		bool affinity_broken;
 
 		desc = irq_to_desc(irq);
-		scoped_guard(raw_spinlock, &desc->lock)
+		scoped_guard(raw_spinlock, &desc->lock) {
 			affinity_broken = migrate_one_irq(desc);
-
+			if (affinity_broken && desc->affinity_notify)
+				schedule_affinity_notify_work(desc);
+		}
 		if (affinity_broken) {
 			pr_debug_ratelimited("IRQ %u: no longer affine to CPU%u\n",
 					    irq, smp_processor_id());
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 349ae7979da0e..6440a64f65a7b 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -139,6 +139,23 @@ void synchronize_irq(unsigned int irq)
 }
 EXPORT_SYMBOL(synchronize_irq);
 
+/**
+ * schedule_affinity_notify_work - Schedule work to notify
+ * about irq affinity change.
+ * @desc:  irq descriptor whose affinity changed
+ *
+ * Caller needs to hold desc->lock
+ */
+void schedule_affinity_notify_work(struct irq_desc *desc)
+{
+	kref_get(&desc->affinity_notify->kref);
+	if (!schedule_work(&desc->affinity_notify->work))
+		/* Work was already scheduled, drop our extra ref */
+		kref_put(&desc->affinity_notify->kref,
+			 desc->affinity_notify->release);
+}
+EXPORT_SYMBOL_GPL(schedule_affinity_notify_work);
+
 #ifdef CONFIG_SMP
 cpumask_var_t irq_default_affinity;
 
@@ -367,14 +384,9 @@ int irq_set_affinity_locked(struct irq_data *data, const struct cpumask *mask,
 		irq_copy_pending(desc, mask);
 	}
 
-	if (desc->affinity_notify) {
-		kref_get(&desc->affinity_notify->kref);
-		if (!schedule_work(&desc->affinity_notify->work)) {
-			/* Work was already scheduled, drop our extra ref */
-			kref_put(&desc->affinity_notify->kref,
-				 desc->affinity_notify->release);
-		}
-	}
+	if (desc->affinity_notify)
+		schedule_affinity_notify_work(desc);
+
 	irqd_set(data, IRQD_AFFINITY_SET);
 
 	return ret;

base-commit: f8f9c1f4d0c7a64600e2ca312dec824a0bc2f1da
-- 
2.34.1