[PATCH 1/2] dt: Add empty of_property_match_string() function

[PATCH 1/2] dt: Add empty of_property_match_string() function

[Thierry Reding]

This commit adds an empty of_property_match_string() function for
!CONFIG_OF builds.

Signed-off-by: Thierry Reding <thierry.reding-RM9K5IK7kjKj5M59NBduVrNAH6kLmebB@public.gmane.org>
---
 include/linux/of.h |    7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/include/linux/of.h b/include/linux/of.h
index e3f942d..937ca14 100644
--- a/include/linux/of.h
+++ b/include/linux/of.h
@@ -361,6 +361,13 @@ static inline int of_property_read_u64(const struct device_node *np,
 	return -ENOSYS;
 }
 
+static inline int of_property_match_string(struct device_node *np,
+					   const char *propname,
+					   const char *string)
+{
+	return -ENOSYS;
+}
+
 static inline struct device_node *of_parse_phandle(struct device_node *np,
 						   const char *phandle_name,
 						   int index)
-- 
1.7.10

[PATCH 2/2] dt: Add empty of_parse_phandle_with_args() function

[Thierry Reding]

This commit adds an empty of_parse_phandle_with_args() function for
!CONFIG_OF builds.

Signed-off-by: Thierry Reding <thierry.reding-RM9K5IK7kjKj5M59NBduVrNAH6kLmebB@public.gmane.org>
---
 include/linux/of.h |    9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/include/linux/of.h b/include/linux/of.h
index 937ca14..bd52d83 100644
--- a/include/linux/of.h
+++ b/include/linux/of.h
@@ -375,6 +375,15 @@ static inline struct device_node *of_parse_phandle(struct device_node *np,
 	return NULL;
 }
 
+static inline int of_parse_phandle_with_args(struct device_node *np,
+					     const char *list_name,
+					     const char *cells_name,
+					     int index,
+					     struct of_phandle_args *out_args)
+{
+	return -ENOSYS;
+}
+
 static inline int of_alias_get_id(struct device_node *np, const char *stem)
 {
 	return -ENOSYS;
-- 
1.7.10

Re: [PATCH 1/2] dt: Add empty of_property_match_string() function

[Thierry Reding]

[-- Attachment #1.1: Type: text/plain, Size: 240 bytes --]

* Thierry Reding wrote:
> This commit adds an empty of_property_match_string() function for
> !CONFIG_OF builds.

Grant, Rob,

can these two patches be queued for 3.5? I need them to build the PWM
subsystem for !OF builds.

Thanks,
Thierry

[-- Attachment #1.2: Type: application/pgp-signature, Size: 198 bytes --]

[-- Attachment #2: Type: text/plain, Size: 192 bytes --]

_______________________________________________
devicetree-discuss mailing list
devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org
https://lists.ozlabs.org/listinfo/devicetree-discuss

Re: [PATCH 1/2] dt: Add empty of_property_match_string() function

[Grant Likely]

On Thu, 26 Apr 2012 21:01:12 +0200, Thierry Reding <thierry.reding-RM9K5IK7kjKj5M59NBduVrNAH6kLmebB@public.gmane.org> wrote:
> * Thierry Reding wrote:
> > This commit adds an empty of_property_match_string() function for
> > !CONFIG_OF builds.
> 
> Grant, Rob,
> 
> can these two patches be queued for 3.5? I need them to build the PWM
> subsystem for !OF builds.

I could merge them, but can you point me at the code that needs them?
I get a little nervous when OF-specific stretches of code get compiled
when !CONFIG_OF.  In a lot of cases the OF data decoding should be in
a separate function that gets completely selected out.  That isn't
always the case of course, but I do like to put a bit of back-pressure
on this issue.

g.

Re: [PATCH 1/2] dt: Add empty of_property_match_string() function

[Thierry Reding]

[-- Attachment #1.1: Type: text/plain, Size: 1312 bytes --]

* Grant Likely wrote:
> On Thu, 26 Apr 2012 21:01:12 +0200, Thierry Reding <thierry.reding@avionic-design.de> wrote:
> > * Thierry Reding wrote:
> > > This commit adds an empty of_property_match_string() function for
> > > !CONFIG_OF builds.
> > 
> > Grant, Rob,
> > 
> > can these two patches be queued for 3.5? I need them to build the PWM
> > subsystem for !OF builds.
> 
> I could merge them, but can you point me at the code that needs them?
> I get a little nervous when OF-specific stretches of code get compiled
> when !CONFIG_OF.  In a lot of cases the OF data decoding should be in
> a separate function that gets completely selected out.  That isn't
> always the case of course, but I do like to put a bit of back-pressure
> on this issue.

The latest series for the PWM subsystem is here[0]. However that version
doesn't contain the latest changes that require this. I haven't pushed those
changes yet because they cause the build to fail (because of these two
missing patches).

Arnd Bergmann recommended to not #ifdef the CONFIG_OF code out but instead
use the IS_ENABLED(CONFIG_OF) construct and let the compiler's DCE handle
this so that we get full compile coverage for the code, even in !OF
configurations.

Thierry

[0]: http://gitorious.org/linux-pwm/linux-pwm

[-- Attachment #1.2: Type: application/pgp-signature, Size: 198 bytes --]

[-- Attachment #2: Type: text/plain, Size: 192 bytes --]

_______________________________________________
devicetree-discuss mailing list
devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org
https://lists.ozlabs.org/listinfo/devicetree-discuss

Re: [PATCH 1/2] dt: Add empty of_property_match_string() function

[Thierry Reding]

[-- Attachment #1.1: Type: text/plain, Size: 457 bytes --]

* Thierry Reding wrote:
> The latest series for the PWM subsystem is here[0]. However that version
> doesn't contain the latest changes that require this. I haven't pushed those
> changes yet because they cause the build to fail (because of these two
> missing patches).

I just pushed the latest code to the for-next branch. That's the state that I
was going to submit during the 3.5 merge window.

Thierry

> [0]: http://gitorious.org/linux-pwm/linux-pwm

[-- Attachment #1.2: Type: application/pgp-signature, Size: 198 bytes --]

[-- Attachment #2: Type: text/plain, Size: 192 bytes --]

_______________________________________________
devicetree-discuss mailing list
devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org
https://lists.ozlabs.org/listinfo/devicetree-discuss

Re: [PATCH 1/2] dt: Add empty of_property_match_string() function

[Thierry Reding]

[-- Attachment #1.1: Type: text/plain, Size: 1393 bytes --]

* Thierry Reding wrote:
> * Thierry Reding wrote:
> > The latest series for the PWM subsystem is here[0]. However that version
> > doesn't contain the latest changes that require this. I haven't pushed those
> > changes yet because they cause the build to fail (because of these two
> > missing patches).
> 
> I just pushed the latest code to the for-next branch. That's the state that I
> was going to submit during the 3.5 merge window.
> 
> Thierry
> 
> > [0]: http://gitorious.org/linux-pwm/linux-pwm

Hi Grant,

Do you have any comments on this? I really want the PWM subsystem to go into
3.6, and for that to happen we need to find a solution for this. As I stated
previously the empty functions are needed to compile the PWM core in !OF
configurations.

Usually this would be solved by just #ifdef'ing the corresponding code, but
with the recent introduction of the config_enabled() and IS_ENABLED() macros
there seems to be a move to replace #ifdef usage with those in order to have
the corresponding code compile-checked in all configurations and have the
compiler throw away the unused code.

I believe that this is a good thing, but it will required these empty OF
functions to be added. If you don't find this an acceptable solution, please
let me know and I'll convert the OF-specific code in the PWM core to use
#ifdef instead.

Thanks,
Thierry

[-- Attachment #1.2: Type: application/pgp-signature, Size: 198 bytes --]

[-- Attachment #2: Type: text/plain, Size: 192 bytes --]

_______________________________________________
devicetree-discuss mailing list
devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org
https://lists.ozlabs.org/listinfo/devicetree-discuss

Re: [PATCH 1/2] dt: Add empty of_property_match_string() function

[Thierry Reding]

[-- Attachment #1.1: Type: text/plain, Size: 1549 bytes --]

On Fri, Jun 01, 2012 at 07:59:20AM +0200, Thierry Reding wrote:
> * Thierry Reding wrote:
> > * Thierry Reding wrote:
> > > The latest series for the PWM subsystem is here[0]. However that version
> > > doesn't contain the latest changes that require this. I haven't pushed those
> > > changes yet because they cause the build to fail (because of these two
> > > missing patches).
> > 
> > I just pushed the latest code to the for-next branch. That's the state that I
> > was going to submit during the 3.5 merge window.
> > 
> > Thierry
> > 
> > > [0]: http://gitorious.org/linux-pwm/linux-pwm
> 
> Hi Grant,
> 
> Do you have any comments on this? I really want the PWM subsystem to go into
> 3.6, and for that to happen we need to find a solution for this. As I stated
> previously the empty functions are needed to compile the PWM core in !OF
> configurations.
> 
> Usually this would be solved by just #ifdef'ing the corresponding code, but
> with the recent introduction of the config_enabled() and IS_ENABLED() macros
> there seems to be a move to replace #ifdef usage with those in order to have
> the corresponding code compile-checked in all configurations and have the
> compiler throw away the unused code.
> 
> I believe that this is a good thing, but it will required these empty OF
> functions to be added. If you don't find this an acceptable solution, please
> let me know and I'll convert the OF-specific code in the PWM core to use
> #ifdef instead.

Grant, Rob,

any update on this?

Thierry

[-- Attachment #1.2: Type: application/pgp-signature, Size: 198 bytes --]

[-- Attachment #2: Type: text/plain, Size: 192 bytes --]

_______________________________________________
devicetree-discuss mailing list
devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org
https://lists.ozlabs.org/listinfo/devicetree-discuss

Re: [PATCH 1/2] dt: Add empty of_property_match_string() function

[Rob Herring]

On 06/13/2012 02:02 AM, Thierry Reding wrote:
> On Fri, Jun 01, 2012 at 07:59:20AM +0200, Thierry Reding wrote:
>> * Thierry Reding wrote:
>>> * Thierry Reding wrote:
>>>> The latest series for the PWM subsystem is here[0]. However that version
>>>> doesn't contain the latest changes that require this. I haven't pushed those
>>>> changes yet because they cause the build to fail (because of these two
>>>> missing patches).
>>>
>>> I just pushed the latest code to the for-next branch. That's the state that I
>>> was going to submit during the 3.5 merge window.
>>>
>>> Thierry
>>>
>>>> [0]: http://gitorious.org/linux-pwm/linux-pwm
>>
>> Hi Grant,
>>
>> Do you have any comments on this? I really want the PWM subsystem to go into
>> 3.6, and for that to happen we need to find a solution for this. As I stated
>> previously the empty functions are needed to compile the PWM core in !OF
>> configurations.
>>
>> Usually this would be solved by just #ifdef'ing the corresponding code, but
>> with the recent introduction of the config_enabled() and IS_ENABLED() macros
>> there seems to be a move to replace #ifdef usage with those in order to have
>> the corresponding code compile-checked in all configurations and have the
>> compiler throw away the unused code.
>>
>> I believe that this is a good thing, but it will required these empty OF
>> functions to be added. If you don't find this an acceptable solution, please
>> let me know and I'll convert the OF-specific code in the PWM core to use
>> #ifdef instead.
> 
> Grant, Rob,
> 
> any update on this?
> 

I agree with your comments, and every other OF function you are using
has an empty version already, so for both patches:

Acked-by: Rob Herring <rob.herring-bsGFqQB8/DxBDgjK7y7TUQ@public.gmane.org>

You can merge these with PWM support since you are dependent on it.

Rob