Unfortunately, this means I now see the 3c59x driver hanging from time to
time, like others have reported.
This is of course with v0.30c-all (the latest 3c59x driver last time I
looked). My 3c590 card is the newer small memory (8k???) one which seems
to be the one people have problems with.
On a loaded server, weekly unload-loads of the ethernet card module, isn't
particular desirable. Is anyone working on fixing the bug or should I sell
the card and get a "works better under Linux" one instead?
The techie stuff:
<lots of>
eth0: Transmitter access conflict.
eth0: Transmitter access conflict.
eth0: Transmitter access conflict.
eth0: Transmitter access conflict.
eth0: Transmitter access conflict.
eth0: Transmitter access conflict.
eth0: Transmitter access conflict.
eth0: Transmitter access conflict.
eth0: Transmitter access conflict.
eth0: Transmitter access conflict.
eth0: Transmitter access conflict.
eth0: Transmitter access conflict.
eth0: Transmitter access conflict.
eth0: Transmitter access conflict.
eth0: vortex_close() status e004, Tx status 88.
3c59x.c:v0.30-all 12/23/96 becker@cesdis.gsfc.nasa.gov
3Com EtherLink III: Overriding PCI latency timer (CFLT) setting of 248, new value is 255.
loading device 'eth0'...
eth0: 3Com 3c590 Vortex 10Mbps at 0x6000, 00:a0:24:6c:18:dc, IRQ 10
Internal config register is 1020010, transceivers 0xe138.
8K byte-wide RAM 1:1 Rx:Tx split, autoselect/10baseT interface.
eth0: Initial media type 10baseT.
eth0: vortex_open() InternalConfig 01020010.
eth0: vortex_open() irq 10 media status 8800.
eth0: Media selection timer tick happened, 10baseT.
eth0: Media 10baseT has link beat, 88c0.
eth0: Media selection timer finished, 10baseT.
A day or two beforehand, we got two solitary "eth0: transmitter access
conflict" messages followed by a lone "eth0: transmit timed out tx status
88 status e004"
Kernel is 2.0.27 with updated 3c59x driver in it as indicated.
Oh well, until the thing dies again I suppose :-)
Cheers,
Chris