Re: [PATCH 2/3] PCI: Handle 64-bit resources better on 32-bit machines

[Jesse Barnes <jbarnes@virtuousgeek.org>]

On Monday, July 28, 2008 10:39 am Matthew Wilcox wrote:
> If the kernel is configured to support 64-bit resources on a 32-bit
> machine, we can support 64-bit BARs properly.  Just change the condition
> to check sizeof(resource_size_t) instead of BITS_PER_LONG.
>
> Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
> ---
>  drivers/pci/probe.c |    4 ++--
>  1 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
> index 3b690c3..2036300 100644
> --- a/drivers/pci/probe.c
> +++ b/drivers/pci/probe.c
> @@ -270,10 +270,10 @@ static int __pci_read_base(struct pci_dev *dev, enum
> pci_bar_type type, if (!sz64)
>  			goto fail;
>
> -		if ((BITS_PER_LONG < 64) && (sz64 > 0x100000000ULL)) {
> +		if ((sizeof(resource_size_t) < 8) && (sz64 > 0x100000000ULL)) {
>  			dev_err(&dev->dev, "can't handle 64-bit BAR\n");
>  			goto fail;
> -		} else if ((BITS_PER_LONG < 64) && l) {
> +		} else if ((sizeof(resource_size_t) < 8) && l) {
>  			/* Address above 32-bit boundary; disable the BAR */
>  			pci_write_config_dword(dev, pos, 0);
>  			pci_write_config_dword(dev, pos + 4, 0);

Both this and 1/3 look good, I've applied them to my for-linus tree.  I'll 
push them out after a little testing.

3/3 also looks nice; is there any in-tree code that could be converted over to 
using it?  If not, we can just push it as part of Eric's stuff.

Thanks,
Jesse