[vortex] Tyan Thunder K7 S2462 Dual 3c59x
Donald Becker
becker@scyld.com
Tue Jan 15 13:49:01 2002
On Tue, 15 Jan 2002, Bogdan Costescu wrote:
> On Mon, 14 Jan 2002, Donald Becker wrote:
>
> > My driver will find a 0x9807 card using the fall-back rule for finding
> > 3c980 boards (matching any 0x980* number). The regular 2.4 PCI
> > detection code doesn't have the ability to match a range of values.
>
> IMHO we should not just add this kind of interfaces. The EPROM is broken;
> either the driver should detect this case and use some
> known-to-be-good-in-most-cases values (in which case the fall-back rule
> is useful) and say it loud or just don't recognize it, as it is now with
> the LK driver.
It's not clear that the EEPROM is broken. I believe that 0x9807 is just
a new ID for a 3c980-on-motherboard implementation, similar to the
previous minor changes in product ID. For instance the 3c905 series used
0x905*, with a different number for each implementation/transceiver
combination.
I've used the pci-scan matching code for about 2.5 years with
essentially no interface changes. It has been much more stable,
functional and compatible than the kernel PCI matching code.
Donald Becker becker@scyld.com
Scyld Computing Corporation http://www.scyld.com
410 Severn Ave. Suite 210 Second Generation Beowulf Clusters
Annapolis MD 21403 410-990-9993