Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 6 Dec 2000 22:35:45 +0000 (GMT)
From:      Terry Lambert <tlambert@primenet.com>
To:        msmith@FreeBSD.ORG (Mike Smith)
Cc:        tlambert@primenet.com (Terry Lambert), freebsd-alpha@FreeBSD.ORG
Subject:   Re: Lynx test / 2nd attempt
Message-ID:  <200012062235.PAA25544@usr08.primenet.com>
In-Reply-To: <200012062053.eB6KrNF52293@mass.osd.bsdi.com> from "Mike Smith" at Dec 06, 2000 12:53:23 PM

next in thread | previous in thread | raw e-mail | index | archive | help
> > Is FreeBSD being "a PnP OS" an alpha need?
> 
> I don't think I understand this question.

An OS can do a PnP scan correctly, without firmware/BIOS support,
and enable devices which by default will not be enabled without
the OS doing the work.

> As an aside, I don't think that the PnP scan has anything to do with the 
> system hanging; it's simply the last probe message printed.  I loathe 
> people that assume that because the last message printed came from 
> subsystem X that it must be subsystem X that is failing.  Boot 
> verbose, you lame bastards. 8)

Heh.


> > This normally leads vendors of non-x86
> > machines to support a limited set of hardware, and do the POST
> > initialization of the card in their own ROM or firmware code.
> 
> This is bunk.

Hey, if you can get my Compaq SCSI controller working on my Multia,
I'll be happy to retract.  :^).


> > If the card's default POST state didn't require the CPU to do
> > processing when the card was installed with PnP mode enable,
> > then doing the "PnP OS" thing would work; otherwise, it probably
> > won't.  This means that ISA PnP will probably be broken on these
> > boxes as well.
> 
> This is also bunk.

PnP cards come up disabled by default, unless they are enabled;
what you're saying here (when you say it's "bunk") translates to
me to mean that that the cards aren't disabled by default, or
that all Apha machines with ISA slots have PnP support in their
firmware/BIOS?

If so, I think the original complaint can be addressed by just
referencing the resource tables out of the PnP stuff on the
platform, but I have no idea how to do that on an Alpha, as
it's unlikely to have PnP BIOS entry points. 8-|.

> > Carrying around the non-x86 equivalent for all
> > possible cards that were designed by people who can't get their
> > hardware into a functional state by default without the help of
> > x86 code is really not an option.
> 
> And this is irrelevant.

It's relevent for my SCSI card, which, as well as hooking the
PC BIOS at POST time, sets up default sequencer firmware using
the host processor.  This doesn't work unless called, which
means that the controller is not capable of booting my machine.

Since the controller is recognized by the AlphaBIOS, I wonder if
it is possible to load stuff into RAM with the AlphaBIOS, and
then switch to the SRM, without disturbing the memory image?

This would let FreeBSD get the FreeBSD driver loaded by the
firmware that recognizes the card as a boot device, and then
switch over.

Right now, this "PnP SCSI card" can't be used to boot, and I'm
stuck with using the built-in NCR for booting, but the cheap
disks are all ultrawide, and the NCR 810 isn't.

The only way I can see to deal with cards that require post
initialization not supported by the firmware is emulation or
different firmware.

What are the chances of using the x86 emulator for video cards
to do the initialization job for SCSI cards?


					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-alpha" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200012062235.PAA25544>