[PATCH] depends on PCI DMA API: Adaptec AIC7xxx_old

[PATCH] depends on PCI DMA API: Adaptec AIC7xxx_old

[Geert Uytterhoeven]

Adaptec AIC7xxx_old unconditionally depends on the PCI DMA API, so mark it
broken if !PCI

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>

--- linux-2.6.8-rc2/drivers/scsi/Kconfig	2004-07-18 15:55:26.000000000 +0200
+++ linux-m68k-2.6.8-rc2/drivers/scsi/Kconfig	2004-07-19 23:16:48.000000000 +0200
@@ -320,7 +320,7 @@
 
 config SCSI_AIC7XXX_OLD
 	tristate "Adaptec AIC7xxx support (old driver)"
-	depends on SCSI
+	depends on SCSI && (PCI || BROKEN)
 	help
 	  WARNING This driver is an older aic7xxx driver and is no longer
 	  under active development.  Adaptec, Inc. is writing a new driver to

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] depends on PCI DMA API: Adaptec AIC7xxx_old

[James Bottomley]

On Tue, 2004-07-20 at 13:39, Geert Uytterhoeven wrote:
> Adaptec AIC7xxx_old unconditionally depends on the PCI DMA API, so mark it
> broken if !PCI

Are you sure this is valid?  I thought it used a NULL pci device for
EISA (which I hate, but which still apparently works at least on x86).

James

Re: [PATCH] depends on PCI DMA API: Adaptec AIC7xxx_old

[Matthew Wilcox]

On Tue, Jul 20, 2004 at 08:39:50PM +0200, Geert Uytterhoeven wrote:
> Adaptec AIC7xxx_old unconditionally depends on the PCI DMA API, so mark it
> broken if !PCI

Shouldn't we rather have dummy PCI DMA APIs?

-- 
"Next the statesmen will invent cheap lies, putting the blame upon 
the nation that is attacked, and every man will be glad of those
conscience-soothing falsities, and will diligently study them, and refuse
to examine any refutations of them; and thus he will by and by convince 
himself that the war is just, and will thank God for the better sleep 
he enjoys after this process of grotesque self-deception." -- Mark Twain

Re: [PATCH] depends on PCI DMA API: Adaptec AIC7xxx_old

[Geert Uytterhoeven]

On Tue, 20 Jul 2004, James Bottomley wrote:
> On Tue, 2004-07-20 at 13:39, Geert Uytterhoeven wrote:
> > Adaptec AIC7xxx_old unconditionally depends on the PCI DMA API, so mark it
> > broken if !PCI
>
> Are you sure this is valid?  I thought it used a NULL pci device for
> EISA (which I hate, but which still apparently works at least on x86).

Wel, I guess we can discuss about that :-)

It uses pci_alloc_consistent() and friends, which are not necessarily defined
if !PCI. A better solution may be to convert the driver to use the generic DMA
API instead of the PCI DMA API when PCI is not selected.

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] depends on PCI DMA API: Adaptec AIC7xxx_old

[Geert Uytterhoeven]

On Tue, 20 Jul 2004, Geert Uytterhoeven wrote:
> On Tue, 20 Jul 2004, James Bottomley wrote:
> > On Tue, 2004-07-20 at 13:39, Geert Uytterhoeven wrote:
> > > Adaptec AIC7xxx_old unconditionally depends on the PCI DMA API, so mark it
> > > broken if !PCI
> >
> > Are you sure this is valid?  I thought it used a NULL pci device for
> > EISA (which I hate, but which still apparently works at least on x86).
>
> Wel, I guess we can discuss about that :-)
>
> It uses pci_alloc_consistent() and friends, which are not necessarily defined
> if !PCI. A better solution may be to convert the driver to use the generic DMA
> API instead of the PCI DMA API when PCI is not selected.

In Ottawa you told me this driver really works on ia32 only, so a dependency on
X86 would probably be OK.

However, when grepping through the various defconfigs, it's enabled on ia64 and
ppc (common, pmac, and default), and not on ia32:

| anakin$ find -name "*defconfig*" -type f | xargs grep SCSI_AIC7XXX_OLD  | grep -v 'not set'
| ./arch/ia64/defconfig:CONFIG_SCSI_AIC7XXX_OLD=y
| ./arch/ppc/configs/common_defconfig:CONFIG_SCSI_AIC7XXX_OLD=m
| ./arch/ppc/configs/pmac_defconfig:CONFIG_SCSI_AIC7XXX_OLD=m
| ./arch/ppc/defconfig:CONFIG_SCSI_AIC7XXX_OLD=m

So should I make it depend on X86 || PPC || IA64 instead?

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