On Tue, 2010-08-31 at 04:12 +0200, Jens Schüßler wrote: > Hi Bob, > > > > What is the exact model you have (use "lspci -nn -d 10b7:")? > > 00:0b.0 Ethernet controller : 3Com Corporation 3c595 100BaseTX [Vortex] [10b7:5950] > > > > > Also can you try enabling some debug logging: > > > > rmmod 3c59x > > modprobe 3c59x debug=3 > > > > Then run dmesg to get the logged messages, and send them back. > > There are not much, even if I set debug=7 > > > ,---- > | [ 72.752091] See Documentation/networking/vortex.txt > | [ 72.752098] 0000:00:0b.0: 3Com PCI 3c595 Vortex 100baseTx at 00018800. > | [ 72.772890] 00:a0:24:43:dd:fd, IRQ 11 > | [ 72.780517] product code 4255 rev 00.0 date 12-02-95 > | [ 72.780527] 64K word-wide RAM 1:1 Rx:Tx split, autoselect/10baseT interface. > | [ 72.780533] 0000:00:0b.0: scatter/gather disabled. h/w checksums disabledo > | > `---- > and thats all. That at least lets me see how it's different from the card I have here. > Dialin to a my ISP works (card is connected to a DSL-Modem), > but if there are comes traffic.... I just done a 'links www.google.de' and > everything freezes totally, but I saw > ,---- > | Refresh [ 76.79933] eth0 vortex_error(), status=0xe091 > `---- > on the terminal at this moment. That message is *precisely* what I needed to find the bug, thanks. > Booting to my parallel installed Ubuntu, also with kernel 2.6.32, > everything is alright. > > Could I be the only one with this problem, and should I better file a > new bug? This is a new bug, introduced by my fix for the previous bug. However, this driver supports a wide range of cards and this new bug only affects some of them. Please do make a new bug report. Ben. -- Ben Hutchings Once a job is fouled up, anything done to improve it makes it worse.
Description: This is a digitally signed message part