2017-03-24 10:42:57

by Tobias Regnery

Subject: [PATCH] of: add stub for of_n_addr_cells

With CONFIG_OF=n and CONFIG_COMPILE_TEST=y the rcar pci-e driver fails to
build on arm:

drivers/pci/host/pcie-rcar.c: In function 'pci_dma_range_parser_init':
drivers/pci/host/pcie-rcar.c:1035:16: error: implicit declaration of function 'of_n_addr_cells' [-Werror=implicit-function-declaration]
parser->pna = of_n_addr_cells(node);
^~~~~~~~~~~~~~~

Fix this by adding an inline stub for of_n_addr_cells

Signed-off-by: Tobias Regnery <[email protected]>
---
This is against next-20140324

include/linux/of.h | 5 +++++
1 file changed, 5 insertions(+)

diff --git a/include/linux/of.h b/include/linux/of.h
index 21e6323de0f3..9978c918222e 100644
--- a/include/linux/of.h
+++ b/include/linux/of.h
@@ -710,6 +710,11 @@ static inline struct device_node *of_get_cpu_node(int cpu,
return NULL;
}

+static inline int of_n_addr_cells(struct device_node *np)
+{
+ return 0;
+}
+
static inline int of_property_read_u64(const struct device_node *np,
const char *propname, u64 *out_value)
{
--
2.11.0

2017-03-24 21:24:49

by Arnd Bergmann

Subject: Re: [PATCH] of: add stub for of_n_addr_cells

On Fri, Mar 24, 2017 at 11:39 AM, Tobias Regnery
<[email protected]> wrote:
> With CONFIG_OF=n and CONFIG_COMPILE_TEST=y the rcar pci-e driver fails to
> build on arm:
>
> drivers/pci/host/pcie-rcar.c: In function 'pci_dma_range_parser_init':
> drivers/pci/host/pcie-rcar.c:1035:16: error: implicit declaration of function 'of_n_addr_cells' [-Werror=implicit-function-declaration]
> parser->pna = of_n_addr_cells(node);
> ^~~~~~~~~~~~~~~
>
> Fix this by adding an inline stub for of_n_addr_cells
>
> Signed-off-by: Tobias Regnery <[email protected]>
> ---
> This is against next-20140324
>
> include/linux/of.h | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/include/linux/of.h b/include/linux/of.h
> index 21e6323de0f3..9978c918222e 100644
> --- a/include/linux/of.h
> +++ b/include/linux/of.h
> @@ -710,6 +710,11 @@ static inline struct device_node *of_get_cpu_node(int cpu,
> return NULL;
> }
>
> +static inline int of_n_addr_cells(struct device_node *np)
> +{
> + return 0;
> +}
> +

This looks good, but we should also do the same thing for of_n_size_cells().

I think I sent something like this a few years ago, but never resubmitted it
when it was ignored at first.

Arnd

2017-03-27 06:07:38

by Tobias Regnery

Subject: Re: [PATCH] of: add stub for of_n_addr_cells

On 24.03.17, Arnd Bergmann wrote:
> On Fri, Mar 24, 2017 at 11:39 AM, Tobias Regnery
> <[email protected]> wrote:
> > With CONFIG_OF=n and CONFIG_COMPILE_TEST=y the rcar pci-e driver fails to
> > build on arm:
> >
> > drivers/pci/host/pcie-rcar.c: In function 'pci_dma_range_parser_init':
> > drivers/pci/host/pcie-rcar.c:1035:16: error: implicit declaration of function 'of_n_addr_cells' [-Werror=implicit-function-declaration]
> > parser->pna = of_n_addr_cells(node);
> > ^~~~~~~~~~~~~~~
> >
> > Fix this by adding an inline stub for of_n_addr_cells
> >
> > Signed-off-by: Tobias Regnery <[email protected]>
> > ---
> > This is against next-20140324
> >
> > include/linux/of.h | 5 +++++
> > 1 file changed, 5 insertions(+)
> >
> > diff --git a/include/linux/of.h b/include/linux/of.h
> > index 21e6323de0f3..9978c918222e 100644
> > --- a/include/linux/of.h
> > +++ b/include/linux/of.h
> > @@ -710,6 +710,11 @@ static inline struct device_node *of_get_cpu_node(int cpu,
> > return NULL;
> > }
> >
> > +static inline int of_n_addr_cells(struct device_node *np)
> > +{
> > + return 0;
> > +}
> > +
>
> This looks good, but we should also do the same thing for of_n_size_cells().
>
> I think I sent something like this a few years ago, but never resubmitted it
> when it was ignored at first.
>
> Arnd

This seems sensible, I can send an updated patch with this change or I can
send it as a separate patch, whatever the maintainers prefer.

--
Tobias

2017-03-27 07:31:20

by Arnd Bergmann

Subject: Re: [PATCH] of: add stub for of_n_addr_cells

On Mon, Mar 27, 2017 at 7:54 AM, Tobias Regnery
<[email protected]> wrote:
> On 24.03.17, Arnd Bergmann wrote:

>> >
>> > +static inline int of_n_addr_cells(struct device_node *np)
>> > +{
>> > + return 0;
>> > +}
>> > +
>>
>> This looks good, but we should also do the same thing for of_n_size_cells().
>>
>> I think I sent something like this a few years ago, but never resubmitted it
>> when it was ignored at first.
>>
>> Arnd
>
> This seems sensible, I can send an updated patch with this change or I can
> send it as a separate patch, whatever the maintainers prefer.

I'd recommend sending an updated patch.

Arnd