Stephane Bortzmeyer <bortzmeyer@pasteur.fr> writes:
> On Wednesday 6 August 97, at 9 h 56, the keyboard of Chris Evans
> <chris@ferret.lmh.ox.ac.uk> wrote:
>> Some clarification: Those of us who have had dreadful trouble with 3c59x.c
>> driver are those with the 8k (I think) 3c590 boards, 10Mbits, with
> My case :-(
Me too. I have failed with:
boomerang.c v0.35 and 3c900.c v0.39.
These two drivers bomb in a transmitter time-out at boot.
I've used 3c59x.c v0.28.c-all since last November. It gives strange
messages at boot: (recent boot time because I just took the machine
down to add a sound card).
Aug 5 12:50:14 altair kernel: eth0: Initial media type MII.
Aug 5 12:50:14 altair kernel: eth0: vortex_open() InternalConfig 016302d8.
Aug 5 12:50:14 altair kernel: eth0: vortex_open() irq 10 media status 8802.
Aug 5 12:50:14 altair kernel: eth0: Media selection timer tick happened, MII.
Aug 5 12:50:14 altair kernel: eth0: Media MII is has no indication, 8802.
Aug 5 12:50:14 altair kernel: eth0: Media selection timer finished, MII.
but otherwise works.
> I didn't try yet 0.42n or other unofficial patches (which often disable
> the bus mastering)...
Try it. I tried it based on the message you are responding to and it
appears to work pretty good.
Aug 6 19:21:47 altair kernel: Warning : Unknown PCI device (10b7:9050). Please read include/linux/pci.h
...
Aug 6 19:21:47 altair kernel: eth0: 3Com 3c905 Boomerang 100baseTx at 0x6100, 00:a0:24:d3:81:aa, IRQ 10
Aug 6 19:21:47 altair kernel: 8K word-wide RAM 3:5 Rx:Tx split, autoselect/MII interface.
Aug 6 19:21:47 altair kernel: Enabling bus-master transmits and whole-frame receives.
Aug 6 19:21:47 altair kernel: eth0: Overriding PCI latency timer (CFLT) setting of 32, new value is 248.
Aug 6 19:21:47 altair kernel: 3c59x.c/3c900.c:v0.42n 8/02/97 Donald Becker linux-vortex@cesdis.gsfc.nasa.gov