Am Dienstag, den 25.01.2011, 13:24 +0100 schrieb Georgi, Patrick:
I'm not quite sure if using weak functions to wrap the subsystem CONFIG_* values is actually the right approach, I'll work on a patch to discuss.
Here it is. I tested it by providing a mainboard_pci_subsystem_vendor_id for my board in its mainboard.c which returned a different constant value - this was picked up by properly and reported on boot. mainboard_pci_subsystem_device_id will work just the same.
Joseph, will this suffice to help you implement your requirement?
Everyone, any opinion on the design? My main issue is that other code can still use the CONFIG_* values directly. Maybe our lint mechanism should look for that? Any other issues you have with this?
Signed-off-by: Patrick Georgi patrick.georgi@secunet.com