Errors with 3C595

Borries Demeler demeler@bioc09.uthscsa.edu
Thu Mar 18 10:53:56 1999


I have seen exactly the same problem on my SMP dual machine with 
the older 2.0.* kernels when APM was enabled in the kernel. The new
2.2.x kernels don't allow APM to be enabled when using SMP.

Make sure to turn off APM in the bios and the kernel, that solved it
(at least for me)

-Borries

> I'm using 3C595 cards on two boxes (a web server and a web proxy), and
> I'm getting errors woth both :
> 
> - Web proxy : kernel 2.0.34
> 
> eth0: 3Com 3c595 Vortex 100baseTX at 0x8100, 00:a0:24:cf:57:75, IRQ 11
>   64K word-wide RAM 3:1 Rx:Tx split, autoselect/10baseT interface.
> eth0: Overriding PCI latency timer (CFLT) setting of 32, new value is
> 248.
> 3c59x.c:v0.99E 5/12/98 Donald Becker
> http://cesdis.gsfc.nasa.gov/linux/drivers/vortex.html
> 
> On high load the network blocks and syslog sends errors like :
> 
> eth0: Transmit error, Tx status register c0.
> eth0: Transmit error, Tx status register 90.
> ...
> 
> - Web server : kernel 2.0.37-pre8 (SMP)
> 
> 3c59x.c:v0.99H 11/17/98 Donald Becker
> http://cesdis.gsfc.nasa.gov/linux/drivers/vortex.html
> eth0: 3Com 3c595 Vortex 100baseTx at 0xe000,  00:a0:24:cf:57:29, IRQ 11
>   64K word-wide RAM 3:1 Rx:Tx split, autoselect/10baseT interface.
> eth0: Overriding PCI latency timer (CFLT) setting of 32, new value is
> 248.
> 
> When dowloading a 340Mb tar.gz file from the server (on local network),
> the transfer stopped after 40Mb and network was dead (but not the box).
> I had to reboot it to get everything working again (warm reboot was
> enough however).
> 
> syslog messages :
> 
> Mar 17 15:03:10 joyeux kernel: eth0: transmit timed out, tx_status 00
> status e000.
> Mar 17 15:03:42 joyeux last message repeated 12 times
> Mar 17 15:04:43 joyeux last message repeated 19 times
> Mar 17 15:05:46 joyeux last message repeated 21 times
> Mar 17 15:06:52 joyeux last message repeated 21 times
> Mar 17 15:07:54 joyeux last message repeated 9 times
> ...
> 
> Should I suspect the cards or is there a small bug in driver for 3C595 ?
> 
> Guillaume.
>