* [PATCH] of: irq: use of_irq_get() in of_irq_to_resource()
@ 2017-05-16 12:09 ` Thomas Petazzoni
0 siblings, 0 replies; 4+ messages in thread
From: Thomas Petazzoni @ 2017-05-16 12:09 UTC (permalink / raw)
To: Rob Herring, Frank Rowand
Cc: devicetree-u79uwXL29TY76Z2rM5mHXA,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r,
Gregory Clement, Thomas Petazzoni
of_irq_to_resource() currently uses irq_of_parse_and_map() to
translate a DT interrupt specification into a Linux virtual interrupt
number. While this works in most cases, irq_of_parse_and_map() doesn't
properly handle the case where the interrupt controller is not yet
available (due to deferred probing for example).
So instead, use of_irq_get(), which is implemented exactly like
irq_of_parse_and_map(), with the exception that if the interrupt
controller is not yet available, it returns -EPROBE_DEFER. Obviously,
we also handle this error and bail out from of_irq_to_resource() when
of_irq_get() returns an error.
This allows to avoid silly error messages at boot time caused by
irq_create_of_mapping() when the interrupt controller is not
available:
[ 0.153168] irq: no irq domain found for /ap806/config-space@f0000000/interrupt-controller@3f0100 !
[ 0.154041] irq: no irq domain found for /cp110-master/config-space@f2000000/interrupt-controller@1e0000 !
[ 0.154124] irq: no irq domain found for /cp110-master/config-space@f2000000/interrupt-controller@1e0000 !
[ 0.154207] irq: no irq domain found for /cp110-master/config-space@f2000000/interrupt-controller@1e0000 !
[ 0.154437] irq: no irq domain found for /cp110-master/config-space@f2000000/interrupt-controller@1e0000 !
[ 0.154518] irq: no irq domain found for /cp110-master/config-space@f2000000/interrupt-controller@1e0000 !
Signed-off-by: Thomas Petazzoni <thomas.petazzoni-wi1+55ScJUtKEb57/3fJTNBPR1lH4CV8@public.gmane.org>
---
drivers/of/irq.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/of/irq.c b/drivers/of/irq.c
index 7c56b72..47b3614 100644
--- a/drivers/of/irq.c
+++ b/drivers/of/irq.c
@@ -369,7 +369,10 @@ EXPORT_SYMBOL_GPL(of_irq_parse_one);
*/
int of_irq_to_resource(struct device_node *dev, int index, struct resource *r)
{
- int irq = irq_of_parse_and_map(dev, index);
+ int irq = of_irq_get(dev, index);
+
+ if (irq < 0)
+ return irq;
/* Only dereference the resource if both the
* resource and the irq are valid. */
--
2.7.4
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related [flat|nested] 4+ messages in thread
* [PATCH] of: irq: use of_irq_get() in of_irq_to_resource()
@ 2017-05-16 12:09 ` Thomas Petazzoni
0 siblings, 0 replies; 4+ messages in thread
From: Thomas Petazzoni @ 2017-05-16 12:09 UTC (permalink / raw)
To: linux-arm-kernel
of_irq_to_resource() currently uses irq_of_parse_and_map() to
translate a DT interrupt specification into a Linux virtual interrupt
number. While this works in most cases, irq_of_parse_and_map() doesn't
properly handle the case where the interrupt controller is not yet
available (due to deferred probing for example).
So instead, use of_irq_get(), which is implemented exactly like
irq_of_parse_and_map(), with the exception that if the interrupt
controller is not yet available, it returns -EPROBE_DEFER. Obviously,
we also handle this error and bail out from of_irq_to_resource() when
of_irq_get() returns an error.
This allows to avoid silly error messages at boot time caused by
irq_create_of_mapping() when the interrupt controller is not
available:
[ 0.153168] irq: no irq domain found for /ap806/config-space at f0000000/interrupt-controller at 3f0100 !
[ 0.154041] irq: no irq domain found for /cp110-master/config-space at f2000000/interrupt-controller at 1e0000 !
[ 0.154124] irq: no irq domain found for /cp110-master/config-space at f2000000/interrupt-controller at 1e0000 !
[ 0.154207] irq: no irq domain found for /cp110-master/config-space at f2000000/interrupt-controller at 1e0000 !
[ 0.154437] irq: no irq domain found for /cp110-master/config-space at f2000000/interrupt-controller at 1e0000 !
[ 0.154518] irq: no irq domain found for /cp110-master/config-space at f2000000/interrupt-controller at 1e0000 !
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
drivers/of/irq.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/of/irq.c b/drivers/of/irq.c
index 7c56b72..47b3614 100644
--- a/drivers/of/irq.c
+++ b/drivers/of/irq.c
@@ -369,7 +369,10 @@ EXPORT_SYMBOL_GPL(of_irq_parse_one);
*/
int of_irq_to_resource(struct device_node *dev, int index, struct resource *r)
{
- int irq = irq_of_parse_and_map(dev, index);
+ int irq = of_irq_get(dev, index);
+
+ if (irq < 0)
+ return irq;
/* Only dereference the resource if both the
* resource and the irq are valid. */
--
2.7.4
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH] of: irq: use of_irq_get() in of_irq_to_resource()
2017-05-16 12:09 ` Thomas Petazzoni
@ 2017-05-16 14:15 ` Rob Herring
-1 siblings, 0 replies; 4+ messages in thread
From: Rob Herring @ 2017-05-16 14:15 UTC (permalink / raw)
To: Thomas Petazzoni
Cc: Frank Rowand, devicetree-u79uwXL29TY76Z2rM5mHXA,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r,
Gregory Clement
On Tue, May 16, 2017 at 02:09:04PM +0200, Thomas Petazzoni wrote:
> of_irq_to_resource() currently uses irq_of_parse_and_map() to
> translate a DT interrupt specification into a Linux virtual interrupt
> number. While this works in most cases, irq_of_parse_and_map() doesn't
> properly handle the case where the interrupt controller is not yet
> available (due to deferred probing for example).
>
> So instead, use of_irq_get(), which is implemented exactly like
> irq_of_parse_and_map(), with the exception that if the interrupt
> controller is not yet available, it returns -EPROBE_DEFER. Obviously,
> we also handle this error and bail out from of_irq_to_resource() when
> of_irq_get() returns an error.
>
> This allows to avoid silly error messages at boot time caused by
> irq_create_of_mapping() when the interrupt controller is not
> available:
>
> [ 0.153168] irq: no irq domain found for /ap806/config-space@f0000000/interrupt-controller@3f0100 !
> [ 0.154041] irq: no irq domain found for /cp110-master/config-space@f2000000/interrupt-controller@1e0000 !
> [ 0.154124] irq: no irq domain found for /cp110-master/config-space@f2000000/interrupt-controller@1e0000 !
> [ 0.154207] irq: no irq domain found for /cp110-master/config-space@f2000000/interrupt-controller@1e0000 !
> [ 0.154437] irq: no irq domain found for /cp110-master/config-space@f2000000/interrupt-controller@1e0000 !
> [ 0.154518] irq: no irq domain found for /cp110-master/config-space@f2000000/interrupt-controller@1e0000 !
>
> Signed-off-by: Thomas Petazzoni <thomas.petazzoni-wi1+55ScJUtKEb57/3fJTNBPR1lH4CV8@public.gmane.org>
> ---
> drivers/of/irq.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
Applied.
BTW, the reason we have both irq_of_parse_and_map and of_irq_get is
irq_of_parse_and_map returns NO_IRQ or an irq. Adding an errno may not
get handled correctly and it's not as simple as auditing the call sites.
Rob
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 4+ messages in thread
* [PATCH] of: irq: use of_irq_get() in of_irq_to_resource()
@ 2017-05-16 14:15 ` Rob Herring
0 siblings, 0 replies; 4+ messages in thread
From: Rob Herring @ 2017-05-16 14:15 UTC (permalink / raw)
To: linux-arm-kernel
On Tue, May 16, 2017 at 02:09:04PM +0200, Thomas Petazzoni wrote:
> of_irq_to_resource() currently uses irq_of_parse_and_map() to
> translate a DT interrupt specification into a Linux virtual interrupt
> number. While this works in most cases, irq_of_parse_and_map() doesn't
> properly handle the case where the interrupt controller is not yet
> available (due to deferred probing for example).
>
> So instead, use of_irq_get(), which is implemented exactly like
> irq_of_parse_and_map(), with the exception that if the interrupt
> controller is not yet available, it returns -EPROBE_DEFER. Obviously,
> we also handle this error and bail out from of_irq_to_resource() when
> of_irq_get() returns an error.
>
> This allows to avoid silly error messages at boot time caused by
> irq_create_of_mapping() when the interrupt controller is not
> available:
>
> [ 0.153168] irq: no irq domain found for /ap806/config-space at f0000000/interrupt-controller at 3f0100 !
> [ 0.154041] irq: no irq domain found for /cp110-master/config-space at f2000000/interrupt-controller at 1e0000 !
> [ 0.154124] irq: no irq domain found for /cp110-master/config-space at f2000000/interrupt-controller at 1e0000 !
> [ 0.154207] irq: no irq domain found for /cp110-master/config-space at f2000000/interrupt-controller at 1e0000 !
> [ 0.154437] irq: no irq domain found for /cp110-master/config-space at f2000000/interrupt-controller at 1e0000 !
> [ 0.154518] irq: no irq domain found for /cp110-master/config-space at f2000000/interrupt-controller at 1e0000 !
>
> Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
> ---
> drivers/of/irq.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
Applied.
BTW, the reason we have both irq_of_parse_and_map and of_irq_get is
irq_of_parse_and_map returns NO_IRQ or an irq. Adding an errno may not
get handled correctly and it's not as simple as auditing the call sites.
Rob
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2017-05-16 14:15 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-05-16 12:09 [PATCH] of: irq: use of_irq_get() in of_irq_to_resource() Thomas Petazzoni
2017-05-16 12:09 ` Thomas Petazzoni
[not found] ` <1494936544-27541-1-git-send-email-thomas.petazzoni-wi1+55ScJUtKEb57/3fJTNBPR1lH4CV8@public.gmane.org>
2017-05-16 14:15 ` Rob Herring
2017-05-16 14:15 ` Rob Herring
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.