aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorAlan Cox <alan@lxorguk.ukuu.org.uk>2007-07-15 23:40:55 -0700
committerLinus Torvalds <torvalds@woody.linux-foundation.org>2007-07-16 09:05:46 -0700
commit4f27c00bf80f122513d3a5be16ed851573164534 (patch)
tree2abad49c1e5c93d2d8698c558eb490b99bd35b87
parentf3dc8c189a20dc5d115b8f0d07ac620e69eff05c (diff)
Improve behaviour of spurious IRQ detect
Currently we handle spurious IRQ activity based upon seeing a lot of invalid interrupts, and we clear things back on the base of lots of valid interrupts. Unfortunately in some cases you get legitimate invalid interrupts caused by timing asynchronicity between the PCI bus and the APIC bus when disabling interrupts and pulling other tricks. In this case although the spurious IRQs are not a problem our unhandled counters didn't clear and they act as a slow running timebomb. (This is effectively what the serial port/tty problem that was fixed by clearing counters when registering a handler showed up) It's easy enough to add a second parameter - time. This means that if we see a regular stream of harmless spurious interrupts which are not harming processing we don't go off and do something stupid like disable the IRQ after a month of running. OTOH lockups and performance killers show up a lot more than 10/second [akpm@linux-foundation.org: cleanup] Signed-off-by: Alan Cox <alan@redhat.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
-rw-r--r--include/linux/irq.h1
-rw-r--r--kernel/irq/spurious.c12
2 files changed, 12 insertions, 1 deletions
diff --git a/include/linux/irq.h b/include/linux/irq.h
index 1695054e8c6..44657197fcb 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -161,6 +161,7 @@ struct irq_desc {
unsigned int wake_depth; /* nested wake enables */
unsigned int irq_count; /* For detecting broken IRQs */
unsigned int irqs_unhandled;
+ unsigned long last_unhandled; /* Aging timer for unhandled count */
spinlock_t lock;
#ifdef CONFIG_SMP
cpumask_t affinity;
diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c
index bd9e272d55e..32b161972fa 100644
--- a/kernel/irq/spurious.c
+++ b/kernel/irq/spurious.c
@@ -172,7 +172,17 @@ void note_interrupt(unsigned int irq, struct irq_desc *desc,
irqreturn_t action_ret)
{
if (unlikely(action_ret != IRQ_HANDLED)) {
- desc->irqs_unhandled++;
+ /*
+ * If we are seeing only the odd spurious IRQ caused by
+ * bus asynchronicity then don't eventually trigger an error,
+ * otherwise the couter becomes a doomsday timer for otherwise
+ * working systems
+ */
+ if (jiffies - desc->last_unhandled > HZ/10)
+ desc->irqs_unhandled = 1;
+ else
+ desc->irqs_unhandled++;
+ desc->last_unhandled = jiffies;
if (unlikely(action_ret != IRQ_NONE))
report_bad_irq(irq, desc, action_ret);
}