Re: [Etherboot-developers] Q: Multiple drivers.

[Peter L. <P.L...@sy...>]

> It's a special option because it requires server side handling and
> will not work with a generic DHCP setup.

Fair enough - as you have probably gathered, our DHCP config is decidely
non-trivial. :)

> But it's the same NIC with the same ROM. In one machine it boots using
> itself. I move it to another machine that has a second NIC of the same
> model and it tries to boot the other NIC. Huh? This is predictable
> behaviour, whether it manages to boot depends on whether the second NIC
> is plugged in or not?

I interpret predictable behaviour, meaning systematic, not identical.
After all, I just opened the case of a machine, and rejigged the nics
(plural). Presumably I know why I did so, and indeed presumably I also
know why I have 2 or more nics?

> Therein lies the difference between our POVs. You treat the EEPROM on
> the NIC as extra BIOS storage. You probably don't have any EPROMs.

Astute observation, Ken. You correctly guess that I treat the (E)EPROMS
as non-volatile software storage, and wouldn't dirty my hands with
anything I couldn't flash. A degree in computing and electronics taught
me early on that the smell of solder and burning ICs was NOT my forte.

Whether one treats firmware as hard software or soft hardware is down to
personal experience. If one has physically put a PROM in a programmer,
fitted it in a nic and then installed the nic, one is more likely to
view a "boot enabled nic" - and by extension the software it contains -
as a single lump of hardware, so tying the nic and software makes
"obvious" sense. If one runs cromutil, it's just as "obviously"
non-volatile software.

Both views are valid, but I predict that trends in manufacturing and
e.g. JFFS will make the latter the "normal" view very soon.