[lug] G40 Series Thinkpad Problems

Frederic Herman fherman at inferential.com
Sun Jul 13 14:44:47 MDT 2003


Matthew Snelham wrote:

>On 10 Jul 2003 09:18 PM or thereabouts, Frederic Herman wrote:
>  
>
>>I have a G40 series Thinkpad (not to be confused with a T40).  The G40 
>>is a new series.  Is there anyone using this Notebook yet?
>>
>>I have two issues:
>>
>>[1] Internal ethernet adapter not recognized.  /proc/pci  shows:
>>  Bus  2, device   0, function  0:
>>    Ethernet controller: PCI device 14e4:170d (Broadcom Corporation) 
>>    
>>
>
>Well, the G40 uses the Broadcom NetXtreme chipset.  What's confusing is
>that it is only advertised as a 10/100 port, and every other Broadcom
>chipset in use at IBM today is 10/100/1000.  So it's possible it's an
>entirely new chipset (for IBM), using the same name (damn marketers).
>
>Have you tried manually loading either the bcm5700 or the tg3 driver
>modules?  Those are the only Linux Broadcom drivers I'm aware of.
>
>--Matthew
>infinite at sigalrm.com
>
>--
>   "I've been talking to dead rabbits and feeding bloody walls. 
>    I've done horrifying things with salad tongs. 
>    It's really eaten into my social life."
>      - Jhonen Vasquez
>_______________________________________________
>Web Page:  http://lug.boulder.co.us
>Mailing List: http://lists.lug.boulder.co.us/mailman/listinfo/lug
>Join us on IRC: lug.boulder.co.us port=6667 channel=#colug
>
>  
>
I downloaded the driver software from Broadcom and rebuilt the driver 
(has to be compiled against kernel).  That all seemed to go OK.  
However, the driver fails to load:

[root at moby root]# insmod bcm5700
Using /lib/modules/2.4.20-18.9/kernel/drivers/net/bcm5700.o
/lib/modules/2.4.20-18.9/kernel/drivers/net/bcm5700.o: init_module: No 
such device
Hint: insmod errors can be caused by incorrect module parameters, 
including invalid IO or IRQ parameters.
      You may find more information in syslog or the output from dmesg


There is nothing in the messages syslog from the attempted load.  
/proc/pci sees the hardware, and so does /proc/iomem

 From my very limited understanding, it seems like the driver doesn't 
know where the hardware is.  Maybe the driver is really for a pcicard 
version of the chipset, or maybe it's a configuration problem.  Since 
neither IBM nor Broadcom are willing to provide any information, I'm at 
a loss.  Any suggestions are welcome.

Thanks for looking at this.

Fred







More information about the LUG mailing list