Re: [patch for 2.6.25? 1/1] pata_sil680: only enable MMIO on Cell blades

[Benjamin Herrenschmidt <benh@ozlabs.org>]

On Thu, 2008-03-27 at 00:46 -0700, akpm@linux-foundation.org wrote:
> From: Benjamin Herrenschmidt <benh@ozlabs.org>
> 
> There have been reported regressions of the SIL 680 driver when using MMIO, so
> this makes it only try MMIO on Cell blades where it's known to be necessary
> (the host bridge doesn't do PIO on these).

This should go in .25 please.

Cheers,
Ben.

> We'll try to find the root problem with MMIO separately.
> 
> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
> Acked-by: Alan Cox <alan@redhat.com>
> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
> ---
> 
>  drivers/ata/pata_sil680.c |    6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff -puN drivers/ata/pata_sil680.c~pata_sil680-only-enable-mmio-on-cell-blades drivers/ata/pata_sil680.c
> --- a/drivers/ata/pata_sil680.c~pata_sil680-only-enable-mmio-on-cell-blades
> +++ a/drivers/ata/pata_sil680.c
> @@ -229,7 +229,11 @@ static u8 sil680_init_chip(struct pci_de
>  	dev_dbg(&pdev->dev, "sil680: BA5_EN = %d clock = %02X\n",
>  		tmpbyte & 1, tmpbyte & 0x30);
>  
> -	*try_mmio = (tmpbyte & 1) || pci_resource_start(pdev, 5);
> +	*try_mmio = 0;
> +#ifdef CONFIG_PPC
> +	if (machine_is(cell))
> +		*try_mmio = (tmpbyte & 1) || pci_resource_start(pdev, 5);
> +#endif
>  
>  	switch(tmpbyte & 0x30) {
>  		case 0x00:
> _