On Feb 12, 2008 3:13 PM, Haavard Skinnemoen <hskinnemoen@atmel.com> wrote:
quoted text
> On Tue, 12 Feb 2008 14:43:30 -0600
> Olof Johansson <olof@lixom.net> wrote:
>
> > > -   depends on (PCI && X86) || ARCH_IOP32X || ARCH_IOP33X || ARCH_IOP13XX
> > > +   depends on (PCI && X86) || ARCH_IOP32X || ARCH_IOP33X || ARCH_IOP13XX || AVR32
> >
> > This is a slippery slope. Things should be the other way around instead,
> > create a HAVE_DMA_DEVICE, select it in the relevant platform code and
> > make DMADEVICES depend on that.
>
> Agree. I'll cook up a patch tomorrow to make it use the new HAVE_*
> scheme.
>

I suggested something similar here:
http://marc.info/?l=linux-kernel&m=118670671806100&w=2

quoted text
> > Or just let the subsystem always be available.
>
> It used to be always available, but then it was changed. Assuming there
> was a reason for this change, I guess we don't want to change it back.
>

Adrian had concerns about users enabling NET_DMA when the hardware
capability is relatively rare.  So, +1 for HAVE_DMA_DEVICE

quoted text
> Haavard

--
Dan
--
unsubscribe notice
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/