Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 7 Apr 1997 22:25:27 +0300 (EEST)
From:      Andrew Stesin <stesin@gu.net>
To:        Satoshi Asami <asami@vader.cs.berkeley.edu>, se@freebsd.org
Cc:        hackers@freebsd.org, scsi@freebsd.org
Subject:   Intel XXpress again (was: 2 PCI busses, 2 AIC chips, 2.2.1. Howto ?
Message-ID:  <Pine.BSF.3.95q.970407202815.18004c-100000@trifork.gu.net>
In-Reply-To: <Pine.BSF.3.95q.970402171956.283L-100000@trifork.gu.net>

next in thread | previous in thread | raw e-mail | index | archive | help
Hi again,

sorry for "replying" to my own message, but now the possibility
of FreeBSD-2.2.1 to occupy this Big Box forever has grown to almost
100%... if only I knew the way to get FreeBSD recognize 2 PCI busses. :((

HEEEELP!!!! :)

On Wed, 2 Apr 1997, Andrew Stesin wrote:
> On Wed, 2 Apr 1997, Satoshi Asami wrote:
> 
> >  * 3. 2.2.1 has some problem with two PCI busses? (while both AICs are
> >  *    on the second bus??? or on EISA???) What I see at boot
> >  *    after the "Probing devices on the pci:0 bus" message:
> > 
> > It might be that it can't recognize the PCI-PCI bridge.  What do the
> > markings on the chip say?  I know FreeBSD can deal with DEC and IBM
> > chips.
> 
> 	Unfortunately (or vice versa? ;)  I have no access to the
> 	box' internals, I'm not authorized to open the case
> 	and look inside. But: the box is "genuine Intel",
> 	the model is Magellan W/15", base board (kinda of
> 	backplane) is XXpress 15" Rel 2, with 1 P166 CPU module.
> 
> 	I'll try to find what the bridge chip is.

	Now I can tell you what are the major (just biggest ones :)
	chips on the backplane, all have "INTEL" on them:

1.	PCIset S82374SB	(I suspice that this chip is a clue..)
2.	PCIset S82375SB	(gets recognized as 82375EB? EISA works anyway)
3.	A|M|I 9637LZR		(chip is (c)1994 Intel, no idea what's it)
4.	XPC 637909-001		(no idea)
5.	XPD 637910-001	2 parts	(no idea)

... and Adaptec:

6.	AIC-7870	2 parts	(we all know them :)

> > Boot with "-v" and send the output to se@freebsd.org.

	Just now I'm trying to boot a GENERIC with the increased size
	of vty' history buffer -- otherwise no way to catch
	all the messages esp. with "-v", PCI ones fly away,
	too many of them.  Unfortunately, building a "true"
	FreeBSD boot floppy isn't so trivial. :(  So no success yet...

> > You may want to 
> > hack /sys/pci/pcisupport.c by yourself (grep for "IBM") to see if you
> > can get it to work.

	I'm considering this, but I don't trust my own skills of
	hacking pretty unfamiliar kernel code to get production
	system running in 2-3 days...  I still hope that there already is
	a solution...


Best regards,
Andrew Stesin

nic-hdl: ST73-RIPE






Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.95q.970407202815.18004c-100000>