[PATCH] xillybus: XILLYBUS_OF should depend on HAS_DMA

[PATCH] xillybus: XILLYBUS_OF should depend on HAS_DMA

[Geert Uytterhoeven]

If NO_DMA=y:

    drivers/built-in.o: In function `xilly_of_unmap':
    xillybus_of.c:(.text+0xa860e): undefined reference to `dma_unmap_single'
    drivers/built-in.o: In function `xilly_map_single_of':
    xillybus_of.c:(.text+0xa8668): undefined reference to `dma_map_single'
    xillybus_of.c:(.text+0xa8676): undefined reference to `dma_mapping_error'
    xillybus_of.c:(.text+0xa86ca): undefined reference to `dma_unmap_single'
    drivers/built-in.o: In function `xilly_dma_sync_single_for_device_of':
    xillybus_of.c:(.text+0xa8700): undefined reference to `dma_sync_single_for_cpu'
    drivers/built-in.o: In function `xilly_dma_sync_single_for_cpu_of':
    xillybus_of.c:(.text+0xa8726): undefined reference to `dma_sync_single_for_cpu'

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
---
 drivers/char/xillybus/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/char/xillybus/Kconfig b/drivers/char/xillybus/Kconfig
index b53bdf12da0d0fe3..b302684d86c14c06 100644
--- a/drivers/char/xillybus/Kconfig
+++ b/drivers/char/xillybus/Kconfig
@@ -24,7 +24,7 @@ config XILLYBUS_PCIE
 
 config XILLYBUS_OF
 	tristate "Xillybus over Device Tree"
-	depends on OF_ADDRESS && OF_IRQ
+	depends on OF_ADDRESS && OF_IRQ && HAS_DMA
 	help
 	  Set to M if you want Xillybus to find its resources from the
 	  Open Firmware Flattened Device Tree. If the target is an embedded
-- 
1.9.1

Re: [PATCH] xillybus: XILLYBUS_OF should depend on HAS_DMA

[Eli Billauer]

Hello Geert,

Thanks for this. The dependency on HAS_DMA is obviously missing (is 
there any real platform that uses OF but has HAS_DMA off? Sounds a bit 
esoteric to me).

And if we're at it, I now realize that XILLYBUS_PCIE maybe also should 
depend on HAS_DMA, in case PCI is enabled but HAS_DMA isn't...?

I don't maintain a git repository of my own -- Greg applies my patches 
into char-next when I have any (doesn't happen very often). So I suppose 
this should be Cced to the maintainers of "CHAR and MISC DRIVERS" (that 
is, Greg and Arnd). Preferably with a "char: xillybus:" prefix in the 
title, I guess.

Thanks & regards,
    Eli


On 23/04/15 21:05, Geert Uytterhoeven wrote:
> If NO_DMA=y:
>
>      drivers/built-in.o: In function `xilly_of_unmap':
>      xillybus_of.c:(.text+0xa860e): undefined reference to `dma_unmap_single'
>      drivers/built-in.o: In function `xilly_map_single_of':
>      xillybus_of.c:(.text+0xa8668): undefined reference to `dma_map_single'
>      xillybus_of.c:(.text+0xa8676): undefined reference to `dma_mapping_error'
>      xillybus_of.c:(.text+0xa86ca): undefined reference to `dma_unmap_single'
>      drivers/built-in.o: In function `xilly_dma_sync_single_for_device_of':
>      xillybus_of.c:(.text+0xa8700): undefined reference to `dma_sync_single_for_cpu'
>      drivers/built-in.o: In function `xilly_dma_sync_single_for_cpu_of':
>      xillybus_of.c:(.text+0xa8726): undefined reference to `dma_sync_single_for_cpu'
>
> Signed-off-by: Geert Uytterhoeven<geert@linux-m68k.org>
> ---
>   drivers/char/xillybus/Kconfig | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/char/xillybus/Kconfig b/drivers/char/xillybus/Kconfig
> index b53bdf12da0d0fe3..b302684d86c14c06 100644
> --- a/drivers/char/xillybus/Kconfig
> +++ b/drivers/char/xillybus/Kconfig
> @@ -24,7 +24,7 @@ config XILLYBUS_PCIE
>
>   config XILLYBUS_OF
>   	tristate "Xillybus over Device Tree"
> -	depends on OF_ADDRESS&&  OF_IRQ
> +	depends on OF_ADDRESS&&  OF_IRQ&&  HAS_DMA
>   	help
>   	  Set to M if you want Xillybus to find its resources from the
>   	  Open Firmware Flattened Device Tree. If the target is an embedded
>    

Re: [PATCH] xillybus: XILLYBUS_OF should depend on HAS_DMA

[Geert Uytterhoeven]

Hi Eli,

On Fri, Apr 24, 2015 at 2:42 PM, Eli Billauer <eli.billauer@gmail.com> wrote:
> Thanks for this. The dependency on HAS_DMA is obviously missing (is there
> any real platform that uses OF but has HAS_DMA off? Sounds a bit esoteric to
> me).

I hit this with the patch that allows to enable OF if COMPILE_TEST, which
will probably be accepted soon.

> And if we're at it, I now realize that XILLYBUS_PCIE maybe also should
> depend on HAS_DMA, in case PCI is enabled but HAS_DMA isn't...?

PCI is only available on platform that set HAS_DMA.

> I don't maintain a git repository of my own -- Greg applies my patches into
> char-next when I have any (doesn't happen very often). So I suppose this
> should be Cced to the maintainers of "CHAR and MISC DRIVERS" (that is, Greg
> and Arnd). Preferably with a "char: xillybus:" prefix in the title, I guess.

I've just used get_maintainer.pl, so please update MAINTAINERS.

Can I add your Acked-by, then I can send an updated version to Greg and Arnd?

Thanks!

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

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

Re: [PATCH] xillybus: XILLYBUS_OF should depend on HAS_DMA

[Eli Billauer]

On 24/04/15 15:58, Geert Uytterhoeven wrote:
> Can I add your Acked-by, then I can send an updated version to Greg and Arnd?
>
>    
Yes, indeed. Thanks.
    Eli