On 06/08/08 11:37 +0200, Stefan Reinauer wrote:
Peter Stuge wrote:
On Wed, Aug 06, 2008 at 11:28:17AM +0200, Stefan Reinauer wrote:
+int pci_find_device(u16 vid, u16 did, pcidev_t * dev) { return find_on_bus(0, vid, did, dev); }
Shouldn't pci_find_device() scan all buses?
Looks good otherwise.
I didn't change that behavior.
All I need for now is a complete set of 8/16/32bit accesses to the config space without passing pointers around.
I might look into more issues as they appear, but I decided to leave those two functions alone for now, so I don't touch side cases I might not know.
The original code was mine - patches welcome to extend the behavior if we think it is needed. So far, Uwe and I haven't had any problems.
Jordan