NetBSD Problem Report #6446

Received: (qmail 20358 invoked from network); 15 Nov 1998 12:09:57 -0000
Message-Id: <19981115120957.1F17F1D64@mabelode.imrryr.org>
Date: Sun, 15 Nov 1998 04:09:57 -0800 (PST)
From: elric@imrryr.org
Reply-To: elric@imrryr.org
To: gnats-bugs@gnats.netbsd.org
Subject: netbooting Multia on -current broken.
X-Send-Pr-Version: 3.95

>Number:         6446
>Category:       port-alpha
>Synopsis:       netbooting Multia on -current broken.
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    port-alpha-maintainer
>State:          analyzed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sun Nov 15 04:20:01 +0000 1998
>Closed-Date:    
>Last-Modified:  Thu Mar 31 07:55:00 +0000 2016
>Originator:     R. C. Dowdeswell
>Release:        Thu Nov 12 05:11:56 PST 1998
>Organization:
a bit.
>Environment:
System: NetBSD mabelode.imrryr.org 1.3H NetBSD 1.3H (MABELODE) #2: Wed Nov 4 02:16:40 PST 1998 elric@mabelode.imrryr.org:/usr/src/sys/arch/alpha/compile/MABELODE alpha


>Description:
	I had a bit of a problem setting up netbooting on my Multia.
	To make a long story short, I set up the ether addr with
	setnetbootinfo into netboot, but it didn't seem to be properly
	receiving any packets (which is why my previous bug report about
	-DARP_DEBUG was filed, during my playing around...)

	Anyway, I found that if I set broken_firmware = 0, then it worked.
>How-To-Repeat:
	Try to netboot a Multia that is similar to mine.
>Fix:
	I changed /sys/arch/alpha/stand/netboot/if_prom.c to not set
	broken_firmware to true, and I was able to hardcode the ether
	addr with /usr/mdec/setnetbootinfo and boot properly.

	I don't have any machines that require broken_firmware == 1,
	though, so I can't really suggest a fix except perhaps to
	change setnetbootinfo to be able to hard code broken_firmware
	and doc it.
>Release-Note:
>Audit-Trail:

From: Ross Harvey <ross@teraflop.com>
To: elric@imrryr.org, gnats-bugs@gnats.netbsd.org
Cc:  Subject: Re: port-alpha/6446: netbooting Multia on -current broken.
Date: Sun, 15 Nov 1998 17:45:58 -0800 (PST)

 Wow, nice work. And that's an interesting idea re: where to put the broken
 firmware bits. Another idea would be to define a boot flag for it.

 However, it seems like as long as the broken firmware problem can be
 reproduced, it's probably possible to fix the algorithm to work on all
 three cases of firmware-sanity. I think someone has offered me a 3000/300
 on long-term loan, and it's because of a particular 3000/300 (and presumably
 also because of other "old-srm" systems) in the past that the broken-firmware
 hack was put in. So, maybe I will be able to reproduce this one.
Responsible-Changed-From-To: gnats-admin->port-alpha-maintainer 
Responsible-Changed-By: fair 
Responsible-Changed-When: Mon Dec 28 09:32:55 PST 1998 
Responsible-Changed-Why:  
This PR is the responsibility of the portmaster, 
not the GNATS database administrator. 

From: "Erik E. Fair" <fair@clock.org>
To: NetBSD GNATS Problem Report Tracking System <gnats-bugs@gnats.netbsd.org>
Cc:  
Subject: Re: port-alpha/6446
Date: Thu, 17 Jan 2002 15:57:29 -0800

 Has this problem been fixed yet?

 	Erik <fair@clock.org>

From: "R.o.s.s  H.a.r.v.e.y" <ross@ghs.com>
To: fair@clock.org, port-alpha-maintainer@netbsd.org
Cc: gnats-admin@netbsd.org
Subject: Re: port-alpha/6446
Date: Thu, 17 Jan 2002 16:53:09 -0800 (PST)

 I've written some code to just try booting without and then
 with broken_firmware, which should solve the problem. But it
 isn't checked in yet so your answer is technically, "no".

From: David Holland <dholland-bugs@netbsd.org>
To: gnats-bugs@netbsd.org
Cc: Roland Dowdeswell <elric@netbsd.org>
Subject: Re: port-alpha/6446: netbooting Multia on -current broken.
Date: Sat, 3 May 2008 02:14:18 +0000

  > From: "R.o.s.s  H.a.r.v.e.y" <ross@ghs.com>
  > To: fair@clock.org, port-alpha-maintainer@netbsd.org
  > Cc: gnats-admin@netbsd.org
  > Subject: Re: port-alpha/6446
  > Date: Thu, 17 Jan 2002 16:53:09 -0800 (PST)
  > 
  >  I've written some code to just try booting without and then
  >  with broken_firmware, which should solve the problem. But it
  >  isn't checked in yet so your answer is technically, "no".

 Any progress?

 (I know, I'm so impatient, it's only been six years... :-) )

 -- 
 David A. Holland
 dholland@netbsd.org

State-Changed-From-To: open->analyzed
State-Changed-By: hubertf@NetBSD.org
State-Changed-When: Sat, 20 Feb 2010 16:49:35 +0000
State-Changed-Why:
Apparently we know sort of a fix... anyone pick it up and go from there?

 - Hubert


From: "Aaron J. Grier" <agrier@poofygoof.com>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: port-alpha/6446
Date: Mon, 16 Apr 2012 23:20:30 -0700

 this is still relevant as of NetBSD 6.0_BETA for my multia.

 I also have a 3000/400 I can check this on as well; IIRC it is one of
 the systems that needs the broken_firmware flag.

 -- 
   Aaron J. Grier | "Not your ordinary poofy goof." | agrier@poofygoof.com

From: David Holland <dholland-bugs@netbsd.org>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: port-alpha/6446
Date: Sat, 21 Apr 2012 01:55:47 +0000

 On Tue, Apr 17, 2012 at 07:25:03AM +0000, Aaron J. Grier wrote:
  >  this is still relevant as of NetBSD 6.0_BETA for my multia.
  >  
  >  I also have a 3000/400 I can check this on as well; IIRC it is one of
  >  the systems that needs the broken_firmware flag.

 I'm taking this off the stuck list :-)

 but I hope someone can attend to it before you disappear...

 -- 
 David A. Holland
 dholland@netbsd.org

From: <lausgans@gmail.com>
To: gnats-bugs@NetBSD.org,
 port-alpha@netbsd.org
Cc: 
Subject: Re: port-alpha/6446
Date: Thu, 31 Mar 2016 10:51:13 +0300

 The problem is still here on 7.0 release.

 I was able to netboot and successfully install NetBSD on Multia using =
 Roland's method, though in a slightly different way.
 Since I don't have other Alphas here, instead of using setnetbootinfo, =
 I've manually added mac address to the if_prom.c and cross-compiled =
 NetBSD.

 BTW NetBSD Multia FAQ states the following
 > Supposedly Firmware revision 3.8-2 is capable of netbooting
 I have exactly 3.8-2 revision, but looks like it still needs mac address =
 hardcoded in binary.
 Though IIRC, when I've tried untouched installation/netboot/netboot =
 image, it was able to get mac address, but in wrong format like "BOOTP 8 =
 0 ...". Maybe I just don't have use of rfc1048 replies enabled for =
 bootpd.=

>Unformatted:

NetBSD Home
NetBSD PR Database Search

(Contact us) $NetBSD: query-full-pr,v 1.39 2013/11/01 18:47:49 spz Exp $
$NetBSD: gnats_config.sh,v 1.8 2006/05/07 09:23:38 tsutsui Exp $
Copyright © 1994-2014 The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.