Re: sleep mode persistence

On Sun, 29 Sep 2002, Frank Lenaerts wrote:

> I have a question regarding the sleep mode of the eepro100. Because

This is actually not a sleep mode problem.

> Later on however, I also unplugged the power cord. The next time I
> booted, the network adapter did not seem to work anymore. The output
> of eepro100-diag -ee -f showed an incorrect EEPROM checksum. 
...
> EEPROM contents, size 256x16:
>     00: ffff ffff ffff ffff ffff ffff ffff ffff

This is indicating that the chip isn't responding.

> escher:~# eepro100-diag -ee -f
> eepro100-diag.c:v2.05 6/13/2001 Donald Becker ([email protected])

You have an older version of the diag program.
More recent versions include a check that the NIC is responding, so
the user doesn't have to interpret all "0xffff" outputs.

The current version is 2.11 8/27/2002 

> This problem also seems to generate some other error messages as can
> be noted in the dmesg output.
> 
> --- begin ---
> eepro100: wait_for_cmd_done timeout!

The chip isn't repsonding.

What driver version are you using?
  (I'm guessing that it's not my release, but rather a modified version.)

> I started doing some tests and noticed that this problem only occurs
> after a complete (i.e. really no power to the mobo) power outage. When
> I really cut off the power, and then boot the machine, I always get
> this incorrect EEPROM checksum. If I just reboot or shutdown the
> machine (without pulling the power cord out of the case) right after
> the first boot, the problem is solved.

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