Re: [ATA]: Don't allow to enable this for SPARC64 without PCI.

Re: [ATA]: Don't allow to enable this for SPARC64 without PCI.

[Jeff Garzik]

On Mon, Jun 04, 2007 at 08:59:06PM +0000, Linux Kernel Mailing List wrote:
> Gitweb:     http://git.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=899b69d079713d0f568fb2700b86987aa94f29b2
> Commit:     899b69d079713d0f568fb2700b86987aa94f29b2
> Parent:     e11a6c236b3070ed05b079f91a9b3defa48b54d3
> Author:     David S. Miller <davem@sunset.davemloft.net>
> AuthorDate: Sun Jun 3 17:39:56 2007 -0700
> Committer:  David S. Miller <davem@sunset.davemloft.net>
> CommitDate: Sun Jun 3 18:09:42 2007 -0700
> 
>     [ATA]: Don't allow to enable this for SPARC64 without PCI.
>     
>     Based upon a report from Meelis Roos.
>     
>     Signed-off-by: David S. Miller <davem@davemloft.net>
> ---
>  drivers/ata/Kconfig |    1 +
>  1 files changed, 1 insertions(+), 0 deletions(-)
> 
> diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig
> index b4a8d60..7d893a6 100644
> --- a/drivers/ata/Kconfig
> +++ b/drivers/ata/Kconfig
> @@ -8,6 +8,7 @@ menuconfig ATA
>  	depends on BLOCK
>  	depends on !(M32R || M68K) || BROKEN
>  	depends on !SUN4 || BROKEN
> +	depends on !(SPARC64 && !PCI)
>  	select SCSI

Um, this was NAK'd:

libata uses generic DMA and generic struct device.  Neither of which
are specific to SPARC64 or PCI.

Did you read the code?
Did you read the thread on which you were CC'd?
Did you see the NAKs from me and Alan (and others)?

	Jeff

Re: [ATA]: Don't allow to enable this for SPARC64 without PCI.

[David Miller]

From: Jeff Garzik <jeff@garzik.org>
Date: Mon, 4 Jun 2007 17:09:52 -0400

> Um, this was NAK'd:
> 
> libata uses generic DMA and generic struct device.  Neither of which
> are specific to SPARC64 or PCI.
> 
> Did you read the code?
> Did you read the thread on which you were CC'd?
> Did you see the NAKs from me and Alan (and others)?

It went in before any of you replied, to be fair.

I'm willing to fix this in other ways, but there is simply
no hardware that's going to work in that kind of configuration
(non-PCI) on sparc64 so building it in that state is going to
be purely for the purposes of intellectual curiosity :-)

Re: [ATA]: Don't allow to enable this for SPARC64 without PCI.

[Jeff Garzik]

On Mon, Jun 04, 2007 at 02:27:36PM -0700, David Miller wrote:
> From: Jeff Garzik <jeff@garzik.org>
> Date: Mon, 4 Jun 2007 17:09:52 -0400
> 
> > Um, this was NAK'd:
> > 
> > libata uses generic DMA and generic struct device.  Neither of which
> > are specific to SPARC64 or PCI.
> > 
> > Did you read the code?
> > Did you read the thread on which you were CC'd?
> > Did you see the NAKs from me and Alan (and others)?
> 
> It went in before any of you replied, to be fair.

Sounds like we should taunt Linus for being slow then ;) ;)


> I'm willing to fix this in other ways, but there is simply
> no hardware that's going to work in that kind of configuration
> (non-PCI) on sparc64 so building it in that state is going to
> be purely for the purposes of intellectual curiosity :-)

Should be easy enough to add dma_xxx wrappers that do nothing but
return an error.

As long as this change is _temporary_, I'm happy.

As you saw from my reply, apply a driver fix for a platform problem
makes me twitch in a big way... :)

	Jeff

Re: [ATA]: Don't allow to enable this for SPARC64 without PCI.

[David Miller]

From: Jeff Garzik <jeff@garzik.org>
Date: Mon, 4 Jun 2007 17:58:41 -0400

> Should be easy enough to add dma_xxx wrappers that do nothing but
> return an error.
> 
> As long as this change is _temporary_, I'm happy.

Yes, I'll work to clear this up for sure.

> As you saw from my reply, apply a driver fix for a platform problem
> makes me twitch in a big way... :)

I totally agree.