(forw) [chris@black-sun.co.uk: Re: [eepro100] (no subject)]

[Donald Becker]

On Wed, 1 Aug 2001, Chris Jones wrote:

> To follow up to myself, I ran the eepro100-diag with the 2.4.5 driver
> loaded and I get these results:
> 
> (I dropped the interface before running these)

You should run it with the interface up-but-broken.

>   Primary interface chip i82555 PHY #1.
>    Sleep mode is enabled.  This is not recommended.
>    Under high load the card may not respond to
>     PCI requests, and thus cause a master abort.
> 
> Is this last message relevent?

I should have added another paragraph of info to this warning message.

Yup.  This is a major problem.  The fix is turning off this "feature".

eepro100-diag -ee -G 0 -w 

The "-G 0" setting is an ad-hoc hack.
It sets the "General purpose" option to 0.
This currently means "fix the EEPROM" for eepro100-diag only.

-e means show the EEPROM settings.
-w means write the EEPROM with the new settings.

I'll likely change "-G 0" to some other slightly-more-obvious
option character (which will add non-standard extension and force me to
modify a whole bunch of documentation).


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