I've confirmed that the below patch (with the syntax fixes already
mentioned) resolves the issue on the SGI 750.
-Brad
> From: Matthew Wilcox <willy@xxxxxxxxxxxxxxx>
> Subject: Require SAL 3.2 in order to do extended config space ops
>
> We had assumed that SAL firmware would return an error if it didn't
> understand extended config space. Unfortunately, the SAL on the SGI 750
> doesn't do that, it panics the machine. So, condition the extended PCI
> config space accesses on SAL revision 3.2.
>
> Signed-off-by: Matthew Wilcox <willy@xxxxxxxxxxxxxxx>
>
> diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c
> index 7de76dd..61363cc 100644
> --- a/arch/ia64/pci/pci.c
> +++ b/arch/ia64/pci/pci.c
> @@ -56,10 +56,13 @@ int raw_pci_read(unsigned int seg, unsigned int bus, unsigned int devfn,
> if ((seg | reg) <= 255) {
> addr = PCI_SAL_ADDRESS(seg, bus, devfn, reg);
> mode = 0;
> - } else {
> + } else if (sal_revision >= SAL_VERSION_CODE(3,2))
> addr = PCI_SAL_EXT_ADDRESS(seg, bus, devfn, reg);
> mode = 1;
> + } else {
> + return -EINVAL;
> }
> +
> result = ia64_sal_pci_config_read(addr, mode, len, &data);
> if (result != 0)
> return -EINVAL;
> @@ -80,9 +83,11 @@ int raw_pci_write(unsigned int seg, unsigned int bus, unsigned int devfn,
> if ((seg | reg) <= 255) {
> addr = PCI_SAL_ADDRESS(seg, bus, devfn, reg);
> mode = 0;
> - } else {
> + } else if (sal_revision >= SAL_VERSION_CODE(3,2))
> addr = PCI_SAL_EXT_ADDRESS(seg, bus, devfn, reg);
> mode = 1;
> + } else {
> + return -EINVAL;
> }
> result = ia64_sal_pci_config_write(addr, mode, len, value);
> if (result != 0)
>
> --
> Matthew Wilcox Intel Open Source Technology Centre
> "Bill, look, we understand that you're interested in selling us this
> operating system, but compare it to ours. We can't possibly take such
> a retrograde step."
Attachment:
signature.asc
Description: Digital signature
