From hnokubi@yyy.or.jp Sun Dec  5 05:43:46 1999
Return-Path: <hnokubi@yyy.or.jp>
Received: from yyy.or.jp (mail.yyy.or.jp [202.214.252.21])
	by hub.freebsd.org (Postfix) with SMTP id 3876014EE3
	for <freebsd-gnats-submit@freebsd.org>; Sun,  5 Dec 1999 05:43:44 -0800 (PST)
	(envelope-from hnokubi@yyy.or.jp)
Received: (qmail 51228 invoked from network); 5 Dec 1999 22:43:38 +0900
Received: from urayasu109.interwave.or.jp (HELO ppp-client.yyy.or.jp) (210.138.157.145)
  by mail.yyy.or.jp with SMTP; 5 Dec 1999 22:43:38 +0900
Received: from sassaby.nokubi.or.jp (sassaby [192.168.9.3]) by ppp-client.yyy.or.jp (8.9.1/3.5Wpl7-ppp) with ESMTP id WAA04510 for <freebsd-gnats-submit@freebsd.org>; Sun, 5 Dec 1999 22:43:54 +0900 (JST)
Received: (from h-nokubi@localhost) by sassaby.nokubi.or.jp (8.9.3/3.5Wpl7-glove) id WAA00655; Sun, 5 Dec 1999 22:43:54 +0900 (JST)
Message-Id: <19991205134345.3876014EE3@hub.freebsd.org>
Date: Sun, 5 Dec 1999 22:43:54 +0900 (JST)
From: hnokubi@yyy.or.jp
Reply-To: hnokubi@yyy.or.jp
To: FreeBSD-gnats-submit@freebsd.org
Subject: Please fix handling Ross(?) host to PCI bridge
X-Send-Pr-Version: 3.2

>Number:         15279
>Category:       kern
>Synopsis:       Please fix handling Ross(?) host to PCI bridge
>Confidential:   no
>Severity:       critical
>Priority:       medium
>Responsible:    freebsd-bugs
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Sun Dec  5 05:50:00 PST 1999
>Closed-Date:    Sun Dec 5 10:42:20 PST 1999
>Last-Modified:  Sun Dec  5 10:42:39 PST 1999
>Originator:     NOKUBI Hirotaka
>Release:        FreeBSD 4.0-CURRENT i386
>Organization:
>Environment:

FreeBSD alphazeal.nokubi.or.jp 4.0-CURRENT FreeBSD 4.0-CURRENT #8: Sun Dec  5 21:39:51 JST 1999     h-nokubi@alphazeal.nokubi.or.jp:/usr/src/src/sys/compile/MP  i386

>Description:

I'm using NEC PC-9821RvII (It's not PC-AT compatible).
After I got bootstrap message `Ross(?)', I cannot use PCI bus.

>How-To-Repeat:


>Fix:

It seems that reading from 0x45 has no meaning on my machine.
Without following patch, my machine reports like this.

pcib255: <Ross (?) host to PCI bridge> on motherboard

And no other PCI devices are found, cannot mount root device, then panic.
It seems that pci_cfgread() returns 0xff.

PC9821RvII uses Champion as it's chip set. Champion is product of
Reliance Computer, I beleive. So, I feel strange about message `Ross(?)'.
`RCC' is more precise?

Index: i386/isa/pcibus.c
===================================================================
RCS file: /home/ncvs/src/sys/i386/isa/pcibus.c,v
retrieving revision 1.50
diff -u -u -r1.50 pcibus.c
--- pcibus.c	1999/11/18 08:58:37	1.50
+++ pcibus.c	1999/12/05 13:02:17
@@ -395,7 +395,9 @@
 	case 0x00051166:
 		s = "Ross (?) host to PCI bridge";
 		/* just guessing the secondary bus register number ... */
+#if 0
 		*busnum = pci_cfgread(cfg, 0x45, 1);
+#endif
 		break;
 
 		/* Integrated Micro Solutions -- vendor 0x10e0 */


>Release-Note:
>Audit-Trail:
State-Changed-From-To: open->closed 
State-Changed-By: dfr 
State-Changed-When: Sun Dec 5 10:42:20 PST 1999 
State-Changed-Why:  
Patch comitted, thanks. 
>Unformatted:
