Skip to contentSkip to conversation

commit "genirq: record trigger type" and powerpc

PATCH Applied Mark Nelson 2 participants 4 messages Nov 13, 2008 IRQ

lore
Share & tools
lore
#0 lore raw

Hi Ingo,

I noticed that after commit 0c5d1eb77a8be917b638344a22afe1398236482b (genirq: record trigger type), several of our powerpc platforms now spew out warnings about "No set_type function for IRQ..." (in particular our Cell blades). This is because in our generic platform code we call set_irq_type() with information from the device tree when we establish the interrupt mappings; but we do this regardless of whether the PIC can actually set a type (it might irrelevant because the type is essentially hardcoded or as in the case for Cell the interrupts are just messages being past around that have no real concept of type, or we could even be dealing with a virtual PIC as on the PS3).

So, would it be possible to turn the:

pr_warning("No set_type function...

into a pr_debug() in kernel/irq/manage.c so when our users upgrade to newer kernels they don't get scared by a whole bunch of new warnings?

Many thanks!

Mark

#1 lore raw
Quoted 14 lines · Mark Nelson

Hi Ingo,

I noticed that after commit 0c5d1eb77a8be917b638344a22afe1398236482b (genirq: record trigger type), several of our powerpc platforms now spew out warnings about "No set_type function for IRQ..." (in particular our Cell blades). This is because in our generic platform code we call set_irq_type() with information from the device tree when we establish the interrupt mappings; but we do this regardless of whether the PIC can actually set a type (it might irrelevant because the type is essentially hardcoded or as in the case for Cell the interrupts are just messages being past around that have no real concept of type, or we could even be dealing with a virtual PIC as on the PS3).

Quoted 7 lines

So, would it be possible to turn the:

pr_warning("No set_type function...

into a pr_debug() in kernel/irq/manage.c so when our users upgrade to newer kernels they don't get scared by a whole bunch of new warnings?

Sure, please send a patch with a changelog.

Ingo

#2 lore raw

PATCH 1/1 genirq: __irq_set_trigger: change pr_warning to pr_debug

Commit 0c5d1eb77a8be917b638344a22afe1398236482b (genirq: record trigger type) caused powerpc platforms that had no set_type() function in their struct irq_chip to spew out warnings about "No set_type function for IRQ...". This warning isn't necessarily justified though because the generic powerpc platform code calls set_irq_type() (which in turn calls __irq_set_trigger) with information from the device tree to establish the interrupt mappings, regardless of whether the PIC can actually set a type.

A platform's irq_chip might not have a set_type function for a variety of reasons, for example: the platform may have the type essentially hard-coded, or as in the case for Cell interrupts are just messages past around that have no real concept of type, or the platform could even have a virtual PIC as on the PS3.

---
Thanks Ingo! Much appreciated

 kernel/irq/manage.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/irq/manage.c +1 -1
Index: upstream/kernel/irq/manage.c
===================================================================
@@ -327,7 +327,7 @@ int __irq_set_trigger(struct irq_desc *d
 		 * IRQF_TRIGGER_* but the PIC does not support multiple
 		 * flow-types?
 		 */
-		pr_warning("No set_type function for IRQ %d (%s)\n", irq,
+		pr_debug("No set_type function for IRQ %d (%s)\n", irq,
 				chip ? (chip->name ? : "unknown") : "unknown");
 		return 0;
 	}
Applied
#3 lore raw

Re: [PATCH] genirq: __irq_set_trigger: change pr_warning to pr_debug

Quoted 15 lines · Mark Nelson

Commit 0c5d1eb77a8be917b638344a22afe1398236482b (genirq: record trigger type) caused powerpc platforms that had no set_type() function in their struct irq_chip to spew out warnings about "No set_type function for IRQ...". This warning isn't necessarily justified though because the generic powerpc platform code calls set_irq_type() (which in turn calls __irq_set_trigger) with information from the device tree to establish the interrupt mappings, regardless of whether the PIC can actually set a type.

A platform's irq_chip might not have a set_type function for a variety of reasons, for example: the platform may have the type essentially hard-coded, or as in the case for Cell interrupts are just messages past around that have no real concept of type, or the platform could even have a virtual PIC as on the PS3.

Quoted 6 lines

---
Thanks Ingo! Much appreciated

 kernel/irq/manage.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

applied to tip/irq/urgent, thanks Mark!

Ingo