of/irq: add empty irq_of_parse_and_map() for non-dt builds

Message ID 1332581257-15989-1-git-send-email-thomas.abraham@linaro.org
State New
Headers show

Commit Message

thomas.abraham@linaro.org March 24, 2012, 9:27 a.m.
Add a empty irq_of_parse_and_map() that returns 0 for non-dt builds and
avoid having #ifdef CONFIG_OF around all calls to irq_of_parse_and_map().

Suggested-by: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: Thomas Abraham <thomas.abraham@linaro.org>
---
 include/linux/of_irq.h |    6 ++++++
 1 files changed, 6 insertions(+), 0 deletions(-)

Comments

rob.herring@linaro.org March 24, 2012, 1:12 p.m. | #1
On 03/24/2012 04:27 AM, Thomas Abraham wrote:
> Add a empty irq_of_parse_and_map() that returns 0 for non-dt builds and
> avoid having #ifdef CONFIG_OF around all calls to irq_of_parse_and_map().
> 
> Suggested-by: Grant Likely <grant.likely@secretlab.ca>
> Signed-off-by: Thomas Abraham <thomas.abraham@linaro.org>
> ---

Acked-by: Rob Herring <rob.herring@calxeda.com>

Go ahead and merge with the rest of your series.

Rob

>  include/linux/of_irq.h |    6 ++++++
>  1 files changed, 6 insertions(+), 0 deletions(-)
> 
> diff --git a/include/linux/of_irq.h b/include/linux/of_irq.h
> index d229ad3..5614355 100644
> --- a/include/linux/of_irq.h
> +++ b/include/linux/of_irq.h
> @@ -76,5 +76,11 @@ extern struct device_node *of_irq_find_parent(struct device_node *child);
>  extern void of_irq_init(const struct of_device_id *matches);
>  
>  #endif /* CONFIG_OF_IRQ */
> +#else /* CONFIG_OF */
> +static inline unsigned int irq_of_parse_and_map(struct device_node *dev,
> +						int index)
> +{
> +	return 0;
> +}
>  #endif /* CONFIG_OF */
>  #endif /* __OF_IRQ_H */
rob.herring@linaro.org March 24, 2012, 1:18 p.m. | #2
On 03/24/2012 08:12 AM, Rob Herring wrote:
> On 03/24/2012 04:27 AM, Thomas Abraham wrote:
>> Add a empty irq_of_parse_and_map() that returns 0 for non-dt builds and
>> avoid having #ifdef CONFIG_OF around all calls to irq_of_parse_and_map().
>>
>> Suggested-by: Grant Likely <grant.likely@secretlab.ca>
>> Signed-off-by: Thomas Abraham <thomas.abraham@linaro.org>
>> ---
> 
> Acked-by: Rob Herring <rob.herring@calxeda.com>
> 
> Go ahead and merge with the rest of your series.
> 

Screwed up the lkml address, so just adding it back...

Rob

> Rob
> 
>>  include/linux/of_irq.h |    6 ++++++
>>  1 files changed, 6 insertions(+), 0 deletions(-)
>>
>> diff --git a/include/linux/of_irq.h b/include/linux/of_irq.h
>> index d229ad3..5614355 100644
>> --- a/include/linux/of_irq.h
>> +++ b/include/linux/of_irq.h
>> @@ -76,5 +76,11 @@ extern struct device_node *of_irq_find_parent(struct device_node *child);
>>  extern void of_irq_init(const struct of_device_id *matches);
>>  
>>  #endif /* CONFIG_OF_IRQ */
>> +#else /* CONFIG_OF */
>> +static inline unsigned int irq_of_parse_and_map(struct device_node *dev,
>> +						int index)
>> +{
>> +	return 0;
>> +}
>>  #endif /* CONFIG_OF */
>>  #endif /* __OF_IRQ_H */
>
thomas.abraham@linaro.org March 24, 2012, 4:07 p.m. | #3
On 24 March 2012 18:42, Rob Herring <robherring2@gmail.com> wrote:
> On 03/24/2012 04:27 AM, Thomas Abraham wrote:
>> Add a empty irq_of_parse_and_map() that returns 0 for non-dt builds and
>> avoid having #ifdef CONFIG_OF around all calls to irq_of_parse_and_map().
>>
>> Suggested-by: Grant Likely <grant.likely@secretlab.ca>
>> Signed-off-by: Thomas Abraham <thomas.abraham@linaro.org>
>> ---
>
> Acked-by: Rob Herring <rob.herring@calxeda.com>
>
> Go ahead and merge with the rest of your series.
>
> Rob

Thanks Rob. It would be good if this patch goes via Grant's
devicetree/next. The other patch series that I have posted would go
via Samsung maintainer's dt-branch into which this patch can be merged
from Grant's devicetree/next.

Regards,
Thomas.

[...]
thomas.abraham@linaro.org March 24, 2012, 4:12 p.m. | #4
On 24 March 2012 21:37, Thomas Abraham <thomas.abraham@linaro.org> wrote:
> On 24 March 2012 18:42, Rob Herring <robherring2@gmail.com> wrote:
>> On 03/24/2012 04:27 AM, Thomas Abraham wrote:
>>> Add a empty irq_of_parse_and_map() that returns 0 for non-dt builds and
>>> avoid having #ifdef CONFIG_OF around all calls to irq_of_parse_and_map().
>>>
>>> Suggested-by: Grant Likely <grant.likely@secretlab.ca>
>>> Signed-off-by: Thomas Abraham <thomas.abraham@linaro.org>
>>> ---
>>
>> Acked-by: Rob Herring <rob.herring@calxeda.com>
>>
>> Go ahead and merge with the rest of your series.
>>
>> Rob
>
> Thanks Rob. It would be good if this patch goes via Grant's
> devicetree/next. The other patch series that I have posted would go
> via Samsung maintainer's dt-branch into which this patch can be merged
> from Grant's devicetree/next.
>
> Regards,
> Thomas.

Fixing lkml address and add Grant in Cc.
Grant Likely March 25, 2012, 12:42 a.m. | #5
On Sat, 24 Mar 2012 08:12:39 -0500, Rob Herring <robherring2@gmail.com> wrote:
> On 03/24/2012 04:27 AM, Thomas Abraham wrote:
> > Add a empty irq_of_parse_and_map() that returns 0 for non-dt builds and
> > avoid having #ifdef CONFIG_OF around all calls to irq_of_parse_and_map().
> > 
> > Suggested-by: Grant Likely <grant.likely@secretlab.ca>
> > Signed-off-by: Thomas Abraham <thomas.abraham@linaro.org>
> > ---
> 
> Acked-by: Rob Herring <rob.herring@calxeda.com>
> 
> Go ahead and merge with the rest of your series.

Actually, it's not quite fully baked; the forward declaration of
irq_of_parse_and_map() needs to be moved under the #if
defined(CONFIG_OF_IRQ) block (just move the #if statement up 7 lines).
After doing that you can add my acked-by and merge it with the rest of
your series.  There is no sense merging it through my tree when you
are the only user depending on it.

Acked-by: Grant Likely <grant.likely@secretlab.ca>

g.

> 
> Rob
> 
> >  include/linux/of_irq.h |    6 ++++++
> >  1 files changed, 6 insertions(+), 0 deletions(-)
> > 
> > diff --git a/include/linux/of_irq.h b/include/linux/of_irq.h
> > index d229ad3..5614355 100644
> > --- a/include/linux/of_irq.h
> > +++ b/include/linux/of_irq.h
> > @@ -76,5 +76,11 @@ extern struct device_node *of_irq_find_parent(struct device_node *child);
> >  extern void of_irq_init(const struct of_device_id *matches);
> >  
> >  #endif /* CONFIG_OF_IRQ */
> > +#else /* CONFIG_OF */
> > +static inline unsigned int irq_of_parse_and_map(struct device_node *dev,
> > +						int index)
> > +{
> > +	return 0;
> > +}
> >  #endif /* CONFIG_OF */
> >  #endif /* __OF_IRQ_H */
> 
> _______________________________________________
> devicetree-discuss mailing list
> devicetree-discuss@lists.ozlabs.org
> https://lists.ozlabs.org/listinfo/devicetree-discuss
Grant Likely March 25, 2012, 12:43 a.m. | #6
On Sat, 24 Mar 2012 21:42:00 +0530, Thomas Abraham <thomas.abraham@linaro.org> wrote:
> On 24 March 2012 21:37, Thomas Abraham <thomas.abraham@linaro.org> wrote:
> > On 24 March 2012 18:42, Rob Herring <robherring2@gmail.com> wrote:
> >> On 03/24/2012 04:27 AM, Thomas Abraham wrote:
> >>> Add a empty irq_of_parse_and_map() that returns 0 for non-dt builds and
> >>> avoid having #ifdef CONFIG_OF around all calls to irq_of_parse_and_map().
> >>>
> >>> Suggested-by: Grant Likely <grant.likely@secretlab.ca>
> >>> Signed-off-by: Thomas Abraham <thomas.abraham@linaro.org>
> >>> ---
> >>
> >> Acked-by: Rob Herring <rob.herring@calxeda.com>
> >>
> >> Go ahead and merge with the rest of your series.
> >>
> >> Rob
> >
> > Thanks Rob. It would be good if this patch goes via Grant's
> > devicetree/next. The other patch series that I have posted would go
> > via Samsung maintainer's dt-branch into which this patch can be merged
> > from Grant's devicetree/next.

As replied to on original patch, this isn't necessary.  Just merge it
via the Samsung tree since you've already got my acked-by on it.

g.
thomas.abraham@linaro.org March 25, 2012, 2:56 p.m. | #7
On 25 March 2012 06:12, Grant Likely <grant.likely@secretlab.ca> wrote:
> On Sat, 24 Mar 2012 08:12:39 -0500, Rob Herring <robherring2@gmail.com> wrote:
>> On 03/24/2012 04:27 AM, Thomas Abraham wrote:
>> > Add a empty irq_of_parse_and_map() that returns 0 for non-dt builds and
>> > avoid having #ifdef CONFIG_OF around all calls to irq_of_parse_and_map().
>> >
>> > Suggested-by: Grant Likely <grant.likely@secretlab.ca>
>> > Signed-off-by: Thomas Abraham <thomas.abraham@linaro.org>
>> > ---
>>
>> Acked-by: Rob Herring <rob.herring@calxeda.com>
>>
>> Go ahead and merge with the rest of your series.
>
> Actually, it's not quite fully baked; the forward declaration of
> irq_of_parse_and_map() needs to be moved under the #if
> defined(CONFIG_OF_IRQ) block (just move the #if statement up 7 lines).
> After doing that you can add my acked-by and merge it with the rest of
> your series.  There is no sense merging it through my tree when you
> are the only user depending on it.
>
> Acked-by: Grant Likely <grant.likely@secretlab.ca>

Thanks. I will do that change which you have suggested and take it
through the Samsung tree. But I was thinking that if it goes through
your tree, there would be less chances of a merge conflict in
include/linux/of_irq.h, a file that might see updates from other
sources.

Thanks,
Thomas.
Grant Likely March 27, 2012, 10:28 p.m. | #8
On Sun, 25 Mar 2012 20:26:15 +0530, Thomas Abraham <thomas.abraham@linaro.org> wrote:
> On 25 March 2012 06:12, Grant Likely <grant.likely@secretlab.ca> wrote:
> > On Sat, 24 Mar 2012 08:12:39 -0500, Rob Herring <robherring2@gmail.com> wrote:
> >> On 03/24/2012 04:27 AM, Thomas Abraham wrote:
> >> > Add a empty irq_of_parse_and_map() that returns 0 for non-dt builds and
> >> > avoid having #ifdef CONFIG_OF around all calls to irq_of_parse_and_map().
> >> >
> >> > Suggested-by: Grant Likely <grant.likely@secretlab.ca>
> >> > Signed-off-by: Thomas Abraham <thomas.abraham@linaro.org>
> >> > ---
> >>
> >> Acked-by: Rob Herring <rob.herring@calxeda.com>
> >>
> >> Go ahead and merge with the rest of your series.
> >
> > Actually, it's not quite fully baked; the forward declaration of
> > irq_of_parse_and_map() needs to be moved under the #if
> > defined(CONFIG_OF_IRQ) block (just move the #if statement up 7 lines).
> > After doing that you can add my acked-by and merge it with the rest of
> > your series.  There is no sense merging it through my tree when you
> > are the only user depending on it.
> >
> > Acked-by: Grant Likely <grant.likely@secretlab.ca>
> 
> Thanks. I will do that change which you have suggested and take it
> through the Samsung tree. But I was thinking that if it goes through
> your tree, there would be less chances of a merge conflict in
> include/linux/of_irq.h, a file that might see updates from other
> sources.

Hmmm... is this going to be merged for v3.4 or v3.5?  If it is v3.5,
then yes it probably does need to go through my tree since there may
be others that need to pull it into their trees.  I was thinking that
you needed this immediately.

g.

Patch hide | download patch | download mbox

diff --git a/include/linux/of_irq.h b/include/linux/of_irq.h
index d229ad3..5614355 100644
--- a/include/linux/of_irq.h
+++ b/include/linux/of_irq.h
@@ -76,5 +76,11 @@  extern struct device_node *of_irq_find_parent(struct device_node *child);
 extern void of_irq_init(const struct of_device_id *matches);
 
 #endif /* CONFIG_OF_IRQ */
+#else /* CONFIG_OF */
+static inline unsigned int irq_of_parse_and_map(struct device_node *dev,
+						int index)
+{
+	return 0;
+}
 #endif /* CONFIG_OF */
 #endif /* __OF_IRQ_H */