The subject says "Trace first 7 BARs..." I think maybe you meant "Track
first 7 BARs" or maybe "Cache only BARs, not windows or IOV BARs"
On Tue, May 19, 2015 at 06:50:06PM +0800, Wei Yang wrote:
> EEH address cache, which helps to locate the PCI device according to
> the given (physical) MMIO address, didn't cover PCI bridges.
"doesn't contain PCI bridge windows"?
I see that eeh_addr_cache_insert_dev() ignores bridges because it never
calls __eeh_addr_cache_insert_dev() when "(dev->class >> 16) ==
PCI_BASE_CLASS_BRIDGE". I think it would be more technically correct if
you removed that test and relied on the "i <= PCI_ROM_RESOURCE" test in
this patch, because it is legal (though rare) for bridge devices to have
two BARs, and I assume you would want to put those in your cache if they
exist.
> Also, it
> shouldn't return PF with address in PF's IOV BARs. Instead, the VFs
> should be returned.
> The patch restricts the address cache to cover first 7 BARs for the
> above purposes.
>
> [gwshan: changelog]
> Signed-off-by: Wei Yang <weiyang@xxxxxxxxxxxxxxxxxx>
> Acked-by: Gavin Shan <gwshan@xxxxxxxxxxxxxxxxxx>
> ---
> arch/powerpc/kernel/eeh_cache.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/powerpc/kernel/eeh_cache.c b/arch/powerpc/kernel/eeh_cache.c
> index eeabeab..f6c5f05 100644
> --- a/arch/powerpc/kernel/eeh_cache.c
> +++ b/arch/powerpc/kernel/eeh_cache.c
> @@ -196,7 +196,7 @@ static void __eeh_addr_cache_insert_dev(struct pci_dev *dev)
> }
>
> /* Walk resources on this device, poke them into the tree */
> - for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) {
> + for (i = 0; i <= PCI_ROM_RESOURCE; i++) {
> unsigned long start = pci_resource_start(dev,i);
> unsigned long end = pci_resource_end(dev,i);
> unsigned int flags = pci_resource_flags(dev,i);
> --
> 1.7.9.5
>
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
