2013-04-10 09:47:28

by Ming Lei

Subject: [PATCH] of: export of_get_next_parent

The function is called by at least three drivers now:

pinctrl
mfd/vexpress
v4l2-core

The 3rd one can be built as module, so export the function to
avoid build failure.

Cc: Grant Likely <[email protected]>
Cc: Rob Herring <[email protected]>
Cc: [email protected]
Signed-off-by: Ming Lei <[email protected]>
---
drivers/of/base.c | 1 +
1 file changed, 1 insertion(+)

diff --git a/drivers/of/base.c b/drivers/of/base.c
index c6443de..a2a0ddc 100644
--- a/drivers/of/base.c
+++ b/drivers/of/base.c
@@ -382,6 +382,7 @@ struct device_node *of_get_next_parent(struct device_node *node)
raw_spin_unlock_irqrestore(&devtree_lock, flags);
return parent;
}
+EXPORT_SYMBOL(of_get_next_parent);

/**
* of_get_next_child - Iterate a node childs
--
1.7.9.5

2013-04-21 12:12:03

by Ming Lei

Subject: Re: [PATCH] of: export of_get_next_parent

Hi,

On Wed, Apr 10, 2013 at 5:47 PM, Ming Lei <[email protected]> wrote:
> The function is called by at least three drivers now:
>
> pinctrl
> mfd/vexpress
> v4l2-core
>
> The 3rd one can be built as module, so export the function to
> avoid build failure.
>
> Cc: Grant Likely <[email protected]>
> Cc: Rob Herring <[email protected]>
> Cc: [email protected]
> Signed-off-by: Ming Lei <[email protected]>
> ---
> drivers/of/base.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/of/base.c b/drivers/of/base.c
> index c6443de..a2a0ddc 100644
> --- a/drivers/of/base.c
> +++ b/drivers/of/base.c
> @@ -382,6 +382,7 @@ struct device_node *of_get_next_parent(struct device_node *node)
> raw_spin_unlock_irqrestore(&devtree_lock, flags);
> return parent;
> }
> +EXPORT_SYMBOL(of_get_next_parent);

Gentle ping.

Without the patch, the below build failure is produced every time on -next tree:

Building modules, stage 2.
MODPOST 72 modules
ERROR: "of_get_next_parent" [drivers/media/v4l2-core/videodev.ko] undefined!
make[1]: *** [__modpost] Error 1
make: *** [modules] Error 2
install kernel and modules


Thanks,
--
Ming Lei

2013-04-21 18:56:44

by Sylwester Nawrocki

Subject: Re: [PATCH] of: export of_get_next_parent

Hi,

On 04/21/2013 02:11 PM, Ming Lei wrote:
> On Wed, Apr 10, 2013 at 5:47 PM, Ming Lei<[email protected]> wrote:
>> The function is called by at least three drivers now:
>>
>> pinctrl
>> mfd/vexpress
>> v4l2-core
>>
>> The 3rd one can be built as module, so export the function to
>> avoid build failure.
>>
>> Cc: Grant Likely<[email protected]>
>> Cc: Rob Herring<[email protected]>
>> Cc: [email protected]
>> Signed-off-by: Ming Lei<[email protected]>
>> ---
>> drivers/of/base.c | 1 +
>> 1 file changed, 1 insertion(+)
>>
>> diff --git a/drivers/of/base.c b/drivers/of/base.c
>> index c6443de..a2a0ddc 100644
>> --- a/drivers/of/base.c
>> +++ b/drivers/of/base.c
>> @@ -382,6 +382,7 @@ struct device_node *of_get_next_parent(struct device_node *node)
>> raw_spin_unlock_irqrestore(&devtree_lock, flags);
>> return parent;
>> }
>> +EXPORT_SYMBOL(of_get_next_parent);
>
> Gentle ping.
>
> Without the patch, the below build failure is produced every time on -next tree:

A similar patch has been queued in the media tree [1]. It should soon
appear in -next.

[1]
http://git.linuxtv.org/mchehab/media-next.git/commitdiff/6695be6863b75620ffa6d422965680ce785cb7c8

Sorry for the trouble.

Thanks,
Sylwester