of/irq: provide int of_irq_parse_and_map_pci wrapper

Commit Message

Arnd Bergmann May 9, 2014, 6:58 a.m. UTC

The pci-rcar driver is enabled for compile tests, and this has
now shown that the driver cannot build without CONFIG_OF,
following the inclusion of f8f2fe7355fb "PCI: rcar: Use new OF
interrupt mapping when possible":

drivers/built-in.o: In function `rcar_pci_map_irq':
:(.text+0x1cc7c): undefined reference to `of_irq_parse_and_map_pci'

As pointed out by Ben Dooks and Geert Uytterhoeven, this is actually
supposed to build fine, which we can achieve if we make the
declaration of of_irq_parse_and_map_pci conditional on CONFIG_OF
and provide an empty inline function otherwise, as we do for
a lot of other of interfaces.

This lets us build the rcar_pci driver again without CONFIG_OF,
which won't work but give us compile time coverage.

Signed-off-by: Arnd Bergmann <[e-mail]>
Cc: Bjorn Helgaas <[e-mail]>
Cc: Magnus Damm <[e-mail]>
Cc: Geert Uytterhoeven <[e-mail]>
Cc: Ben Dooks <[e-mail]>
Cc: [e-mail]
Cc: [e-mail]


--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to [e-mail]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Comments

Geert Uytterhoeven May 9, 2014, 7:11 a.m. UTC | #1

On Fri, May 9, 2014 at 8:58 AM, Arnd Bergmann <[e-mail]> wrote:
> The pci-rcar driver is enabled for compile tests, and this has
> now shown that the driver cannot build without CONFIG_OF,
> following the inclusion of f8f2fe7355fb "PCI: rcar: Use new OF
> interrupt mapping when possible":
>
> drivers/built-in.o: In function `rcar_pci_map_irq':
> :(.text+0x1cc7c): undefined reference to `of_irq_parse_and_map_pci'
>
> As pointed out by Ben Dooks and Geert Uytterhoeven, this is actually
> supposed to build fine, which we can achieve if we make the
> declaration of of_irq_parse_and_map_pci conditional on CONFIG_OF
> and provide an empty inline function otherwise, as we do for
> a lot of other of interfaces.
>
> This lets us build the rcar_pci driver again without CONFIG_OF,
> which won't work but give us compile time coverage.
>
> Signed-off-by: Arnd Bergmann <[e-mail]>
> Cc: Bjorn Helgaas <[e-mail]>
> Cc: Magnus Damm <[e-mail]>
> Cc: Geert Uytterhoeven <[e-mail]>
> Cc: Ben Dooks <[e-mail]>
> Cc: [e-mail]
> Cc: [e-mail]

Acked-by: Geert Uytterhoeven <[e-mail]>

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [e-mail]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to [e-mail]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Lucas Stach May 9, 2014, 8:28 a.m. UTC | #2

Am Freitag, den 09.05.2014, 08:58 +0200 schrieb Arnd Bergmann:
> The pci-rcar driver is enabled for compile tests, and this has
> now shown that the driver cannot build without CONFIG_OF,
> following the inclusion of f8f2fe7355fb "PCI: rcar: Use new OF
> interrupt mapping when possible":
> 
> drivers/built-in.o: In function `rcar_pci_map_irq':
> :(.text+0x1cc7c): undefined reference to `of_irq_parse_and_map_pci'
> 
> As pointed out by Ben Dooks and Geert Uytterhoeven, this is actually
> supposed to build fine, which we can achieve if we make the
> declaration of of_irq_parse_and_map_pci conditional on CONFIG_OF
> and provide an empty inline function otherwise, as we do for
> a lot of other of interfaces.
> 
> This lets us build the rcar_pci driver again without CONFIG_OF,
> which won't work but give us compile time coverage.
> 
This part of the commit message is confusing. The driver is explicitly
written to fall back to a platform irq when OF interrupt mapping doesn't
work (return 0 or error), to keep non DT enabled boards working.

I hadn't thought of the compile time dependency, so thanks for fixing
this up.

> Signed-off-by: Arnd Bergmann <[e-mail]>
> Cc: Bjorn Helgaas <[e-mail]>
> Cc: Magnus Damm <[e-mail]>
> Cc: Geert Uytterhoeven <[e-mail]>
> Cc: Ben Dooks <[e-mail]>
> Cc: [e-mail]
> Cc: [e-mail]
> 
> diff --git a/include/linux/of_pci.h b/include/linux/of_pci.h
> index 1a1f5ff..1e29bf4 100644
> --- a/include/linux/of_pci.h
> +++ b/include/linux/of_pci.h
> @@ -7,13 +7,18 @@
>  struct pci_dev;
>  struct of_phandle_args;
>  int of_irq_parse_pci(const struct pci_dev *pdev, struct of_phandle_args *out_irq);
> -int of_irq_parse_and_map_pci(const struct pci_dev *dev, u8 slot, u8 pin);
>  
>  struct device_node;
>  struct device_node *of_pci_find_child_device(struct device_node *parent,
>  					     unsigned int devfn);
>  int of_pci_get_devfn(struct device_node *np);
>  int of_pci_parse_bus_range(struct device_node *node, struct resource *res);
> +#ifdef CONFIG_OF
> +int of_irq_parse_and_map_pci(const struct pci_dev *dev, u8 slot, u8 pin);
> +#else
> +static inline int
> +of_irq_parse_and_map_pci(const struct pci_dev *dev, u8 slot, u8 pin) { return 0; }
> +#endif
>  
>  #if defined(CONFIG_OF) && defined(CONFIG_PCI_MSI)
>  int of_pci_msi_chip_add(struct msi_chip *chip);
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-pci" in
> the body of a message to [e-mail]
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Arnd Bergmann May 9, 2014, 8:45 a.m. UTC | #3

On Friday 09 May 2014 10:28:39 Lucas Stach wrote:
> Am Freitag, den 09.05.2014, 08:58 +0200 schrieb Arnd Bergmann:
> > The pci-rcar driver is enabled for compile tests, and this has
> > now shown that the driver cannot build without CONFIG_OF,
> > following the inclusion of f8f2fe7355fb "PCI: rcar: Use new OF
> > interrupt mapping when possible":
> > 
> > drivers/built-in.o: In function `rcar_pci_map_irq':
> > :(.text+0x1cc7c): undefined reference to `of_irq_parse_and_map_pci'
> > 
> > As pointed out by Ben Dooks and Geert Uytterhoeven, this is actually
> > supposed to build fine, which we can achieve if we make the
> > declaration of of_irq_parse_and_map_pci conditional on CONFIG_OF
> > and provide an empty inline function otherwise, as we do for
> > a lot of other of interfaces.
> > 
> > This lets us build the rcar_pci driver again without CONFIG_OF,
> > which won't work but give us compile time coverage.
> > 
> This part of the commit message is confusing. The driver is explicitly
> written to fall back to a platform irq when OF interrupt mapping doesn't
> work (return 0 or error), to keep non DT enabled boards working.
> 
> I hadn't thought of the compile time dependency, so thanks for fixing
> this up.

I thought it was meant to keep old DTBs working. AFAIK, there is no
legacy board file support for rcar-gen2, so we don't actually have to
worry about that.

	Arnd
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to [e-mail]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Lucas Stach May 9, 2014, 8:54 a.m. UTC | #4

Am Freitag, den 09.05.2014, 10:45 +0200 schrieb Arnd Bergmann:
> On Friday 09 May 2014 10:28:39 Lucas Stach wrote:
> > Am Freitag, den 09.05.2014, 08:58 +0200 schrieb Arnd Bergmann:
> > > The pci-rcar driver is enabled for compile tests, and this has
> > > now shown that the driver cannot build without CONFIG_OF,
> > > following the inclusion of f8f2fe7355fb "PCI: rcar: Use new OF
> > > interrupt mapping when possible":
> > > 
> > > drivers/built-in.o: In function `rcar_pci_map_irq':
> > > :(.text+0x1cc7c): undefined reference to `of_irq_parse_and_map_pci'
> > > 
> > > As pointed out by Ben Dooks and Geert Uytterhoeven, this is actually
> > > supposed to build fine, which we can achieve if we make the
> > > declaration of of_irq_parse_and_map_pci conditional on CONFIG_OF
> > > and provide an empty inline function otherwise, as we do for
> > > a lot of other of interfaces.
> > > 
> > > This lets us build the rcar_pci driver again without CONFIG_OF,
> > > which won't work but give us compile time coverage.
> > > 
> > This part of the commit message is confusing. The driver is explicitly
> > written to fall back to a platform irq when OF interrupt mapping doesn't
> > work (return 0 or error), to keep non DT enabled boards working.
> > 
> > I hadn't thought of the compile time dependency, so thanks for fixing
> > this up.
> 
> I thought it was meant to keep old DTBs working. AFAIK, there is no
> legacy board file support for rcar-gen2, so we don't actually have to
> worry about that.
> 
I haven't watched closely, but at the time of writing the patch the were
some board files in flight which used the rcar-pci.  All I wanted to say
is that with your patch applied things continue to work, at least at the
same level as before my change, but I really don't care too much about
the details.

Regards,
Lucas

Geert Uytterhoeven May 9, 2014, 8:56 a.m. UTC | #5

Hi Arnd,

On Fri, May 9, 2014 at 10:45 AM, Arnd Bergmann <[e-mail]> wrote:
> On Friday 09 May 2014 10:28:39 Lucas Stach wrote:
>> Am Freitag, den 09.05.2014, 08:58 +0200 schrieb Arnd Bergmann:
>> > The pci-rcar driver is enabled for compile tests, and this has
>> > now shown that the driver cannot build without CONFIG_OF,
>> > following the inclusion of f8f2fe7355fb "PCI: rcar: Use new OF
>> > interrupt mapping when possible":
>> >
>> > drivers/built-in.o: In function `rcar_pci_map_irq':
>> > :(.text+0x1cc7c): undefined reference to `of_irq_parse_and_map_pci'
>> >
>> > As pointed out by Ben Dooks and Geert Uytterhoeven, this is actually
>> > supposed to build fine, which we can achieve if we make the
>> > declaration of of_irq_parse_and_map_pci conditional on CONFIG_OF
>> > and provide an empty inline function otherwise, as we do for
>> > a lot of other of interfaces.
>> >
>> > This lets us build the rcar_pci driver again without CONFIG_OF,
>> > which won't work but give us compile time coverage.
>> >
>> This part of the commit message is confusing. The driver is explicitly
>> written to fall back to a platform irq when OF interrupt mapping doesn't
>> work (return 0 or error), to keep non DT enabled boards working.
>>
>> I hadn't thought of the compile time dependency, so thanks for fixing
>> this up.
>
> I thought it was meant to keep old DTBs working. AFAIK, there is no
> legacy board file support for rcar-gen2, so we don't actually have to
> worry about that.

arch/arm/mach-shmobile/board-lager.c does intantiate platform devices
for it, and I guess they're still to be added to .../board-koelsch.c.

However, all legacy board configs in arch/arm/mach-shmobile/Kconfig
do "select USE_OF", so the file will always be build with OF=y.

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [e-mail]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to [e-mail]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch

diff --git a/include/linux/of_pci.h b/include/linux/of_pci.h
index 1a1f5ff..1e29bf4 100644
--- a/include/linux/of_pci.h
+++ b/include/linux/of_pci.h
@@ -7,13 +7,18 @@ 
 struct pci_dev;
 struct of_phandle_args;
 int of_irq_parse_pci(const struct pci_dev *pdev, struct of_phandle_args *out_irq);
-int of_irq_parse_and_map_pci(const struct pci_dev *dev, u8 slot, u8 pin);
 
 struct device_node;
 struct device_node *of_pci_find_child_device(struct device_node *parent,
 					     unsigned int devfn);
 int of_pci_get_devfn(struct device_node *np);
 int of_pci_parse_bus_range(struct device_node *node, struct resource *res);
+#ifdef CONFIG_OF
+int of_irq_parse_and_map_pci(const struct pci_dev *dev, u8 slot, u8 pin);
+#else
+static inline int
+of_irq_parse_and_map_pci(const struct pci_dev *dev, u8 slot, u8 pin) { return 0; }
+#endif
 
 #if defined(CONFIG_OF) && defined(CONFIG_PCI_MSI)
 int of_pci_msi_chip_add(struct msi_chip *chip);