Re: RTL8100B support
Donald Becker <[email protected]> Wed, 9 Oct 2002 05:55:14 -0400 (EDT)
| Newsgroups | gmane.linux.drivers.realtek.devel |
|---|---|
| Message-ID | <[email protected]> |
On Wed, 9 Oct 2002, Orlando Andico wrote: > On Tue, 8 Oct 2002, Donald Becker wrote: > .. > > > I notice that both modules have to be loaded for the ethernet > > > interface to work. > > > > This should not be true. Either one should work alone. > > It's true. :) > > If both interfaces are 8139too, I get a SIOCGIFFLAGS: No such device when > I try to bring up eth0. The driver modules detect only their supported devices, so this is normal if you load two modules and then unload the first. > The driver version is 0.9.26 from 2.4.20-pre9 (hmm.. the version from > gkernel.sourceforge.net is also 0.9.26 but I can't build it outside the > kernel tree). Try the drivers from scyld.com mkdir /tmp/netdrivers/ cd /tmp/netdrivers/ ncftp ftp://ftp.scyld.com/pub/network/netdrivers.tgz tar xfvz netdrivers.tgz make make install > (as an aside, the alternate 8139C+ driver in 2.4.20-pre9 does *not* work > with this particular incarnation) It won't... BTW, I consider it a bad idea to split out the "C+" support. Even though the "C+" chip adds a much more efficient transfer mode, the part numbers are so similar that having two drivers is likely to lead to tremendous user confusion. > I've also tried the rtl8139 driver as you suggested. If I set both NIC's > as rtl8139, eth1 fails to initialize (no such driver). Regardless, the > drivers always report a "transmit timeout" -- and, it doesn't work anyway > (can't ping). 'mii-tool -w' says "autonegotiation failed, link ok" whereas > with the 8139too driver it works fine. What is the output of 'mii-diag' ? -- Donald Becker [email protected] Scyld Computing Corporation http://www.scyld.com 410 Severn Ave. Suite 210 Scyld Beowulf cluster system Annapolis MD 21403 410-990-9993