On 26-Nov-2002 Alexander Funcke wrote:
> Hi,
>
>
> I've written a GPIO driver for the National Geode SCx200 processors.
> While I was developing I added the following line in isab_match()
> (sys/pci/pcisupport.c) so it wouldn't be identified as a isa-bridge.
>
>
> isab_match(device_t dev)
> {
> switch(pci_get_devid(dev)) {
>
> .
> .
> .
>
> /* National Geode SCx200's ISA and GPIO/LPC bridge,
> needs to be free so we can claim it later from
> the GPIO-driver. */
>
> case 0x0500100b:
> return(NULL);
> }
>
> .
> .
> .
> }
>
> This do work, but I guess this isn't the correct way, how should it
> be done?
Does this device have a class of PCIC_BRIDGE and subclass of PCIS_BRIDGE_ISA?
(class of 0x0601zz where zz is any value). If so it is claiming to be a
PCI-ISA bridge. If this device is not a PCI-ISA bridge, then it shouldn't
claim to be one. If it is a PCI-ISA bridge then you do need to use that
kind of hack for now.
--
John Baldwin <[EMAIL PROTECTED]> <>< http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve!" - http://www.FreeBSD.org/
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
