Bug#483699: video-nsc: PCI ID conflict with newer video-geode
On Mon, Jun 9, 2008 at 4:08 PM, Brice Goglin <Brice.Goglin@ens-lyon.org> wrote:
> Martin-Éric Racine wrote:
>> Anyhow, the real issue is that generating random lists of "supported"
>> hardware, by matching every manufacturer and device ID found in the
>> source code, produces false positives, which is not desirable.
> Which "random" lists ???? We extract the list of hardware that the
> driver claims to support. The patch may look very simple, but it
> actually does his job. If the driver claims more than it actually does
> support, the driver needs to be fixed. There's no need for a static
> list, neither to change the patch that extracts the list.
That's not correct.
The patch generates a very rough list that makes it look like every
vendor supports every device. It completely disregards conditional
checks that the driver itself does.