NetBSD Problem Report #59550
From gson@gson.org Thu Jul 24 16:50:39 2025
Return-Path: <gson@gson.org>
Received: from mail.netbsd.org (mail.netbsd.org [199.233.217.200])
(using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits)
key-exchange X25519 server-signature RSA-PSS (2048 bits)
client-signature RSA-PSS (2048 bits))
(Client CN "mail.NetBSD.org", Issuer "mail.NetBSD.org CA" (not verified))
by mollari.NetBSD.org (Postfix) with ESMTPS id 96CCE1A923E
for <gnats-bugs@gnats.NetBSD.org>; Thu, 24 Jul 2025 16:50:39 +0000 (UTC)
Message-Id: <20250724165030.B5BA9253F03@guava.gson.org>
Date: Thu, 24 Jul 2025 19:50:30 +0300 (EEST)
From: gson@gson.org (Andreas Gustafsson)
Reply-To: gson@gson.org (Andreas Gustafsson)
To: gnats-bugs@NetBSD.org
Subject: Power Mac G5 boot regression
X-Send-Pr-Version: 3.95
>Number: 59550
>Category: port-macppc
>Synopsis: Power Mac G5 boot regression
>Confidential: no
>Severity: serious
>Priority: high
>Responsible: port-macppc-maintainer
>State: closed
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Thu Jul 24 16:55:01 +0000 2025
>Closed-Date: Fri Jul 25 08:03:40 +0000 2025
>Last-Modified: Fri Jul 25 08:03:40 +0000 2025
>Originator: Andreas Gustafsson
>Release: NetBSD-current
>Organization:
>Environment:
System: NetBSD
Architecture: powerpc
Machine: macppc
>Description:
The modified POWERMAC_G5 kernel I recently managed to boot on a Power
Mac G5 as described in PR 59529 no longer boots since this commit:
2025.07.20.09.41.19 martin src/sys/arch/macppc/conf/GENERIC 1.383
Enable PAX_ASLR
The last output from OFW is:
loading .data, done..
clearing .bss, done..
Invalid memory access at $SRR0: 00000000.00e00384 $SRR1: 10000000.00003030
ok
0 >
where the "$" is actually a character I don't recognize.
This is booting from a CD containing ofwboot.xcf and the kernel, both
built from the same -current sources. Since the error happens so
early, I suspect this is a problem with ofwboot.xcf rather than the
kernel.
Reverting the commit makes the boot work again.
>How-To-Repeat:
>Fix:
>Release-Note:
>Audit-Trail:
From: Martin Husemann <martin@duskware.de>
To: gnats-bugs@netbsd.org
Cc:
Subject: Re: port-macppc/59550: Power Mac G5 boot regression
Date: Thu, 24 Jul 2025 19:17:01 +0200
On Thu, Jul 24, 2025 at 04:55:02PM +0000, Andreas Gustafsson wrote:
> Enable PAX_ASLR
>
> The last output from OFW is:
>
> loading .data, done..
> clearing .bss, done..
>
> Invalid memory access at $SRR0: 00000000.00e00384 $SRR1: 10000000.00003030
> ok
> 0 >
>
> where the "$" is actually a character I don't recognize.
This is very strange as PAX_ASLR in this context is plain userland and nothing
in the kernel should really change (modulo minor address difference due to
a bit additional code being compiled in).
Defintively it should not interfere with ofwboot and this early boot stage.
Martin
From: "Martin Husemann" <martin@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc:
Subject: PR/59550 CVS commit: src/sys/arch/macppc/stand/ofwboot
Date: Fri, 25 Jul 2025 06:18:24 +0000
Module Name: src
Committed By: martin
Date: Fri Jul 25 06:18:24 UTC 2025
Modified Files:
src/sys/arch/macppc/stand/ofwboot: Makefile
Log Message:
PR 59550: do not build ofwboot as PIE.
While there, disabel all sanitziers too.
To generate a diff of this commit:
cvs rdiff -u -r1.61 -r1.62 src/sys/arch/macppc/stand/ofwboot/Makefile
Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.
From: Martin Husemann <martin@duskware.de>
To: gnats-bugs@netbsd.org
Cc:
Subject: Re: port-macppc/59550: Power Mac G5 boot regressiony
Date: Fri, 25 Jul 2025 08:19:47 +0200
On Thu, Jul 24, 2025 at 04:55:02PM +0000, Andreas Gustafsson wrote:
> The modified POWERMAC_G5 kernel I recently managed to boot on a Power
> Mac G5 as described in PR 59529 no longer boots since this commit:
>
> 2025.07.20.09.41.19 martin src/sys/arch/macppc/conf/GENERIC 1.383
>
> Enable PAX_ASLR
I just double checked and at least the in-tree version of POWERMAC_G5
currently does NOT include GENERIC, so PAX_ASLR is still not used on G5
machines.
I tested, and it works for me, so I added it now belatedly.
However, could it be that your bisection blamed the neighbour commit
accidently?
I could understand the PIE change I commited minutes before the above
change could break ofwboot (which is trivial to fix).
I have commited a candidate fix for that, could you retry with latest
current?
If that is not the case, can you please provide the change you did
to POWERMAC_G5 and dmesg output from a working boot?
Thanks,
Martin
From: Andreas Gustafsson <gson@gson.org>
To: gnats-bugs@netbsd.org
Cc: port-macppc-maintainer@netbsd.org,
gnats-admin@netbsd.org
Subject: Re: port-macppc/59550: Power Mac G5 boot regressiony
Date: Fri, 25 Jul 2025 09:51:49 +0300
Martin Husemann via gnats wrote:
> However, could it be that your bisection blamed the neighbour commit
> accidently?
I think you're right. My tests showed that source date
2025.07.20.00.36.08 booted successfully and 2025.07.20.09.41.19
didn't, but somehow I missed that there was one more commit
between them 2025.07.20.09.33.11 and failed to test that one.
Sorry about the confusion.
> I have commited a candidate fix for that, could you retry with latest
> current?
Will do.
--
Andreas Gustafsson, gson@gson.org
State-Changed-From-To: open->closed
State-Changed-By: gson@NetBSD.org
State-Changed-When: Fri, 25 Jul 2025 08:03:40 +0000
State-Changed-Why:
An unmodified POWERMAC_G5 kernel now boots and gets as far as
prompting for the root device. Thanks!
>Unformatted:
(Contact us)
$NetBSD: query-full-pr,v 1.51 2026/08/10 02:28:17 riastradh Exp $
$NetBSD: gnats_config.sh,v 1.10 2026/05/13 22:00:09 riastradh Exp $
Copyright © 1994-2026
The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.