NetBSD Problem Report #51296

From martin@duskware.de  Wed Jun 29 14:51:12 2016
Return-Path: <martin@duskware.de>
Received: from mail.netbsd.org (mail.netbsd.org [199.233.217.200])
	(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
	(Client CN "mail.netbsd.org", Issuer "Postmaster NetBSD.org" (verified OK))
	by mollari.NetBSD.org (Postfix) with ESMTPS id B1CE17A476
	for <gnats-bugs@gnats.NetBSD.org>; Wed, 29 Jun 2016 14:51:12 +0000 (UTC)
From: martin@NetBSD.org
Reply-To: martin@NetBSD.org
To: gnats-bugs@NetBSD.org
Subject: gpt migrate test assumes x86-like environment
X-Send-Pr-Version: 3.95

>Number:         51296
>Category:       bin
>Synopsis:       gpt migrate test assumes x86-like environment
>Confidential:   no
>Severity:       critical
>Priority:       medium
>Responsible:    bin-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed Jun 29 14:55:00 +0000 2016
>Last-Modified:  Wed Jun 29 21:25:01 +0000 2016
>Originator:     Martin Husemann
>Release:        NetBSD 7.99.32
>Organization:
The NetBSD Foundation, Inc.
>Environment:
System: NetBSD space-truckin.duskware.de 7.99.32 NetBSD 7.99.32 (CUBIETRUCK) #266: Wed Jun 29 12:43:39 CEST 2016 martin@night-owl.duskware.de:/usr/src/sys/arch/evbarm/compile/CUBIETRUCK evbarm
Architecture: earmv7hfeb
Machine: evbarm
>Description:

The gpt migrate test case (part of /usr/tests/sbin/gpt) fails on various
machines because it assumes a MBR/x86'ish environment.

It creates a disk image from a template that is made for RAW_PART=3 (i.e. 'd')
machines like this:

dd conv=notrunc msgfmt=quiet if=/dev/zero of=gpt.disk seek=10240 count=1
fdisk -fi gpt.disk
fdisk -fu0s 169/63/1024 gpt.disk
disklabel -R gpt.disk /usr/tests/sbin/gpt/gpt.disklabel

This disklabel input would neeed to be made different for raw_part = 'c'
architectures, and also the partitions should be better aligned. On sun
machines this step already fails:

disklabel: warning, partition a: not starting on cylinder boundary
disklabel: warning, partition b: not starting on cylinder boundary
disklabel: warning, partition c: not starting on cylinder boundary
disklabel: warning, partition e: not starting on cylinder boundary
disklabel: warning, partition f: not starting on cylinder boundary

Then the resulting disk does not match gpt(1)'s expections:

gpt migrate gpt.disk
gpt: gpt.disk: MBR partition without disklabel


>How-To-Repeat:
s/a

>Fix:
Maybe instead provide a binary uuencoded/compressed golden image of the
MBR disk oto start from?

>Audit-Trail:
From: John Nemeth <jnemeth@cue.bc.ca>
To: gnats-bugs@NetBSD.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Cc: 
Subject: Re: bin/51296: gpt migrate test assumes x86-like environment
Date: Wed, 29 Jun 2016 14:38:22 -0700

 On Jun 29,  2:55pm, martin@NetBSD.org wrote:
 }
 } >Number:         51296
 } >Category:       bin
 } >Synopsis:       gpt migrate test assumes x86-like environment
 } >Arrival-Date:   Wed Jun 29 14:55:00 +0000 2016
 } >Originator:     Martin Husemann
 } >Release:        NetBSD 7.99.32
 } >Organization:
 } The NetBSD Foundation, Inc.
 } >Environment:
 } System: NetBSD space-truckin.duskware.de 7.99.32 NetBSD 7.99.32 (CUBIETRUCK) #266: Wed Jun 29 12:43:39 CEST 2016 martin@night-owl.duskware.de:/usr/src/sys/arch/evbarm/compile/CUBIETRUCK evbarm
 } Architecture: earmv7hfeb
 } Machine: evbarm
 } >Description:
 } 
 } The gpt migrate test case (part of /usr/tests/sbin/gpt) fails on various
 } machines because it assumes a MBR/x86'ish environment.
 } 
 } It creates a disk image from a template that is made for RAW_PART=3 (i.e. 'd')
 } machines like this:
 } 
 } dd conv=notrunc msgfmt=quiet if=/dev/zero of=gpt.disk seek=10240 count=1

      Just a note that this appears to be an off-by-1 error.  This
 command creates a disk that is exactly 5MB + 1 sector.  It should
 probably be "seek=10239".

 }-- End of excerpt from martin@NetBSD.org

Home	PR Database Search