[eepro100] wait_for_cmd_done timeout

[Donald Becker]

On Wed, 6 Mar 2002, Wilson, John wrote:

> Thanks for the reply.  Agreed, the driver is reporting the problem and
> appearently the ATM driver is seeing this same problem and reporting it in a
> different way.
> 
> I've built the eepro100-diag and mii-diag programs.  Thanks for the simple
...
> Do you have any recommendations on whether this driver is better off running
> as a module vs added directly to the kernel... which is currenlty how I have
> it configured -- mostly due to the previous owners configuration.

Using the driver as a module is preferred.  It allow you to easily
change drivers, as well as allowing you to change the
following module parameters:
MODULE_PARM(txfifo, "i");
MODULE_PARM(rxfifo, "i");
MODULE_PARM(txdmacount, "i");
MODULE_PARM(rxdmacount, "i");

It's rare that you need to change from the default settings, but with a
compiled-in driver it's much more difficult.

> [root@sla2 bin]# mii-diag   
> Basic registers of MII PHY #1:  3000 782d 02a8 0154 05e1 0021 0000 0000.
>  Your link partner is generating 10baseT link beat  (no autonegotiation).

10baseT hub?  Is that correct?

> [root@sla2 bin]# eepro100-diag -mm -f
> eepro100-diag.c:v2.07 12/28/2001 Donald Becker (becker@scyld.com)

Doing 'eepro100-diag -mm' is pretty much the same as doing
 mii-diag --watch 

To see the NIC state, run
  eepro100-diag -a -f
just after the interface stops working.

> Monitoring the MII transceiver status.
> 13:28:23.414  Baseline value of MII BMSR (basic mode status register) is
> 782d.
> I'm assuming that the above status register might change when the error
> occurs.


Donald Becker				becker@scyld.com
Scyld Computing Corporation		http://www.scyld.com
410 Severn Ave. Suite 210		Second Generation Beowulf Clusters
Annapolis MD 21403			410-990-9993