Re: [2.6 patch] AGP_AMD64 must depend on PCI

[Adrian Bunk <bunk@stusta.de>]

On Wed, Jun 28, 2006 at 01:02:22PM -0400, Dave Jones wrote:
> On Wed, Jun 28, 2006 at 06:55:30PM +0200, Adrian Bunk wrote:
>  > This patch fixes the following compile error:
>  > 
>  > <--  snip  -->
>  > 
>  > ...
>  >   CC      arch/i386/kernel/../../x86_64/kernel/k8.o
>  > arch/i386/kernel/../../x86_64/kernel/k8.c: In function ‘next_k8_northbridge’:
>  > arch/i386/kernel/../../x86_64/kernel/k8.c:34: error: implicit declaration of function ‘pci_match_id’
>  > 
>  > <--  snip  -->
>  > 
>  > Signed-off-by: Adrian Bunk <bunk@stusta.de>
>  > 
>  > --- linux-2.6.17-mm3-full/drivers/char/agp/Kconfig.old	2006-06-28 12:56:20.000000000 +0200
>  > +++ linux-2.6.17-mm3-full/drivers/char/agp/Kconfig	2006-06-28 12:56:55.000000000 +0200
>  > @@ -56,7 +56,7 @@
>  >  
>  >  config AGP_AMD64
>  >  	tristate "AMD Opteron/Athlon64 on-CPU GART support" if !IOMMU
>  > -	depends on AGP && X86
>  > +	depends on AGP && PCI && X86
>  >  	default y if IOMMU
>  >  	help
>  >  	  This option gives you AGP support for the GLX component of
> 
> What makes this driver special ? The other AGP drivers also use
> various PCI functions, and don't have a similar dependancy.

The usage of pci_match_id() was the only case where an AGP driver didn't 
compile with CONFIG_PCI=n (the other used PCI functions have dummies for 
CONFIG_PCI=n).

> Adding depends on PCI to 'AGP' may make more sense than adding
> it to the individual drivers, especially as the core AGP code also
> uses the pci layer.   As AGP is just a bastardised variant of PCI
> anyway, adding an explicit dependancy should be sane on all archs,
> even alpha with its wierdo 'faked pci bridge'.

Fine with me.

> 		Dave

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed