From nobody  Wed Nov 18 07:32:56 1998
Received: (from nobody@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id HAA19624;
          Wed, 18 Nov 1998 07:32:56 -0800 (PST)
          (envelope-from nobody)
Message-Id: <199811181532.HAA19624@hub.freebsd.org>
Date: Wed, 18 Nov 1998 07:32:56 -0800 (PST)
From: jchristi@vt.edu
To: freebsd-gnats-submit@freebsd.org
Subject: 3.0 boot disk will not probe for Adaptec AHA 1542B SCSI controller
X-Send-Pr-Version: www-1.0

>Number:         8743
>Category:       kern
>Synopsis:       3.0 boot disk will not probe for Adaptec AHA 1542B SCSI controller
>Confidential:   no
>Severity:       serious
>Priority:       low
>Responsible:    imp
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Wed Nov 18 07:40:00 PST 1998
>Closed-Date:    Wed Jan 6 00:42:24 MST 1999
>Last-Modified:  Wed Jan  6 00:42:42 MST 1999
>Originator:     Jason Christian
>Release:        3.0-RELEASE
>Organization:
Virginia Tech  (CNS)
>Environment:
>Description:
When trying to install FreeBSD 3.0 on a 486DX-33, the Adaptec 1542 SCSI
controller is not probed.  I have also tried using several of the
FreeBSD-current boot binaries and have found that the support for aha0
disappered between 3.0-19981103-SNAP and 3.0-19981115-SNAP.
3.0-19981103-SNAP probes for it but does not detect it.  I noticed that
there is a similar bug report for the Adaptec 1540 (kern/8340).

>How-To-Repeat:
Problem occures when trying to install FreeBSD 3.0-RELEASE on a machine
with an Adaptec AHA 1542B SCSI controller.
>Fix:

>Release-Note:
>Audit-Trail:
State-Changed-From-To: open->feedback 
State-Changed-By: imp 
State-Changed-When: Mon Nov 23 15:47:21 MST 1998 
State-Changed-Why:  
I think this has been fixed 


Responsible-Changed-From-To: freebsd-bugs->imp 
Responsible-Changed-By: imp 
Responsible-Changed-When: Mon Nov 23 15:47:21 MST 1998 
Responsible-Changed-Why:  
I think this is fixed 

From: mvp@braz.ru
To: freebsd-gnats-submit@freebsd.org
Cc:  Subject: Re: kern/8743: 3.0 boot disk will not probe for Adaptec AHA 1542B SCSI controller
Date: Sun, 6 Dec 1998 18:06:46 -0800 (PST)

 >Submitter-Id:	net
 >Originator:	Vadim Mikhailov
 >Organization:	Bratsk Aluminium Plant
 >Confidential:	no
 >Synopsis:	Re: kern/8743: 3.0 boot disk will not probe for Adaptec AHA 1542B SCSI controller
 >Severity:	serious
 >Priority:	medium
 >Category:	kern
 >Class:		sw-bug
 >Release:	FreeBSD-3.0
 >Environment:	FreeBSD mars.braz.ru 3.0-RELEASE FreeBSD 3.0-RELEASE #1: Fri Dec  4 01:13:38 IS
 1998     mvp@mars.braz.ru:/usr/src/sys/compile/SMP-XE  i386
 
 >Description:
 This machine has proprietary Dell SCSI Array RAID controller DSA,
 which run in AHA1540 compatibility mode (the only way FreeBSD
 can recognize this card, because native drivers for this card
 doesn't exist).
 When trying to boot from FreeBSD-3.0 boot floppy, controller aha0
 doesn't recognized. 
 Howewer, in FreeBSD 2.2.x and pre-CAM 3.0-SNAP's all works fine. 
 I'm check 3.0-19981123-SNAP, but problem still exist.
 
 >How-To-Repeat:
 
 >Fix:
 I found solution for this problem by booting with -v flag.
 DSA does not return the same status codes as genuine
 Adaptec AHA-1540 card. Kernel expects that status code
 of geometry register request may be only 0xff, but DSA
 always returns 0. So, my patch is very simple:
 
 --- /usr/src/sys/dev/aha.c.orig  Mon Dec  7 17:22:36 1998
 +++ /usr/src/sys/dev/aha.c       Mon Dec  7 17:23:43 1998
 @@ -364,7 +364,7 @@
                  /* Wait 10ms before reading */
                  DELAY(10000);
                  status = aha_inb(aha, GEOMETRY_REG);
 -                if (status != 0xff) {
 +                if (status != 0xff || status !=0 ) {
                          PRVERB(("%s: Geometry Register test failed\n",
                                  aha_name(aha)));
                          return (ENXIO);
 
State-Changed-From-To: feedback->closed 
State-Changed-By: imp 
State-Changed-When: Wed Jan 6 00:42:24 MST 1999 
State-Changed-Why:  
This has been resolved, as near as I can tell. 
>Unformatted:
