[PATCH] genirq: delay incrementing interrupt count if it's disabled/pending

From: Sudeep Holla
Date: Wed Nov 04 2015 - 13:34:32 EST


In case of wakeup interrupt, irq_pm_check_wakeup disables the interrupt
and marks it pending and suspended, disable it and notifies the pm core
about the wake event. The interrupt gets handled later once the system
is resumed.

However the irq stats is updated twice: once when it's disabled waiting
for the system to resume and later when it's handled, resulting in wrong
counting of the wakeup interrupt when waking up the system.

This patch updates the interrupt count bit later so that it's updated
only when the interrupt gets handled. It's already handled correctly in
handle_edge_irq and handle_edge_eoi_irq.

Cc: Marc Zyngier <marc.zyngier@xxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Reported-by: Manoil Claudiu <claudiu.manoil@xxxxxxxxxxxxx>
Signed-off-by: Sudeep Holla <sudeep.holla@xxxxxxx>
---
kernel/irq/chip.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index e28169dd1c36..212b80f103f7 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -315,7 +315,6 @@ void handle_nested_irq(unsigned int irq)
raw_spin_lock_irq(&desc->lock);

desc->istate &= ~(IRQS_REPLAY | IRQS_WAITING);
- kstat_incr_irqs_this_cpu(desc);

action = desc->action;
if (unlikely(!action || irqd_irq_disabled(&desc->irq_data))) {
@@ -323,6 +322,7 @@ void handle_nested_irq(unsigned int irq)
goto out_unlock;
}

+ kstat_incr_irqs_this_cpu(desc);
irqd_set(&desc->irq_data, IRQD_IRQ_INPROGRESS);
raw_spin_unlock_irq(&desc->lock);

@@ -389,13 +389,13 @@ void handle_simple_irq(struct irq_desc *desc)
goto out_unlock;

desc->istate &= ~(IRQS_REPLAY | IRQS_WAITING);
- kstat_incr_irqs_this_cpu(desc);

if (unlikely(!desc->action || irqd_irq_disabled(&desc->irq_data))) {
desc->istate |= IRQS_PENDING;
goto out_unlock;
}

+ kstat_incr_irqs_this_cpu(desc);
handle_irq_event(desc);

out_unlock:
@@ -439,7 +439,6 @@ void handle_level_irq(struct irq_desc *desc)
goto out_unlock;

desc->istate &= ~(IRQS_REPLAY | IRQS_WAITING);
- kstat_incr_irqs_this_cpu(desc);

/*
* If its disabled or no action available
@@ -450,6 +449,7 @@ void handle_level_irq(struct irq_desc *desc)
goto out_unlock;
}

+ kstat_incr_irqs_this_cpu(desc);
handle_irq_event(desc);

cond_unmask_irq(desc);
@@ -509,7 +509,6 @@ void handle_fasteoi_irq(struct irq_desc *desc)
goto out;

desc->istate &= ~(IRQS_REPLAY | IRQS_WAITING);
- kstat_incr_irqs_this_cpu(desc);

/*
* If its disabled or no action available
@@ -521,6 +520,7 @@ void handle_fasteoi_irq(struct irq_desc *desc)
goto out;
}

+ kstat_incr_irqs_this_cpu(desc);
if (desc->istate & IRQS_ONESHOT)
mask_irq(desc);

--
1.9.1

--
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/