From owner-freebsd-hackers  Mon Dec  4 06:57:50 1995
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.6.12/8.6.6) id GAA06403
          for hackers-outgoing; Mon, 4 Dec 1995 06:57:50 -0800
Received: from Sysiphos (Sysiphos.MI.Uni-Koeln.DE [134.95.212.10])
          by freefall.freebsd.org (8.6.12/8.6.6) with SMTP id GAA06363
          for <hackers@freebsd.org>; Mon, 4 Dec 1995 06:56:46 -0800
Received: by Sysiphos id AA19736
  (5.67b/IDA-1.5 for hackers@freebsd.org); Mon, 4 Dec 1995 15:48:06 +0100
Message-Id: <199512041448.AA19736@Sysiphos>
From: se@zpr.uni-koeln.de (Stefan Esser)
Date: Mon, 4 Dec 1995 15:48:06 +0100
In-Reply-To: Carlos Antonio Ruggiero <TOTO@ifqsc.sc.usp.br>
       "PCI probe problems." (Dec  4,  3:31)
X-Mailer: Mail User's Shell (7.2.6 alpha(2) 7/9/95)
To: Carlos Antonio Ruggiero <TOTO@ifqsc.sc.usp.br>
Subject: Re: PCI probe problems.
Cc: hackers@freebsd.org
Sender: owner-hackers@freebsd.org
Precedence: bulk

On Dec 4,  3:31, Carlos Antonio Ruggiero wrote:
} Subject: PCI probe problems.
} Hi everybody,
} I'm trying to install 2.1 Release in 4 Pentium based motherboards with
} PCI buses. In two of them (a Zeus and a noname one) the PCI gets
} probed all right. In another noname motherboard, the bus doesn't get
} probed (though everything seems OK if I install 2.0.5). In a compaq
} machine both 2.0.5R and 2.1 R fail to probe the bus.

There are too many systems with non-compliant chipsets (Compaq being
the worst, IMHO), and the PCI bus probe code contains many special 
cases by now. You may want to try a kernel with the PCI probe code
from FreeBSD-current. You'll only need "/sys/i386/isa/pcibus.c" from 
a -current source tree, it should compile cleanly on a 2.1R system.

} Does anybody know what's happening?

Could you please enter the "-v" option at the "Boot: " promt
and send the numbers in all lines after the first pcibus_setup() ?

What chip sets are used in the non-working systems ?

2.1R contains special code to deal with recent Compaq PCI chip sets,
and has been tested with a number of them.

I'll try to make all your systems supported in the main stream code,
if you send VERBOSE boot messages (i.e. those obtained by booting
with "-v").

Regards, STefan