NetBSD Problem Report #44224

From www@NetBSD.org  Sun Dec 12 19:08:04 2010
Return-Path: <www@NetBSD.org>
Received: from mail.netbsd.org (mail.netbsd.org [204.152.190.11])
	by www.NetBSD.org (Postfix) with ESMTP id 1B6CD63B886
	for <gnats-bugs@gnats.NetBSD.org>; Sun, 12 Dec 2010 19:08:04 +0000 (UTC)
Message-Id: <20101212190803.F159763B87A@www.NetBSD.org>
Date: Sun, 12 Dec 2010 19:08:03 +0000 (UTC)
From: lacombar@gmail.com
Reply-To: lacombar@gmail.com
To: gnats-bugs@NetBSD.org
Subject: -current double-fault at boot
X-Send-Pr-Version: www-1.0

>Number:         44224
>Category:       kern
>Synopsis:       -current double-fault at boot
>Confidential:   no
>Severity:       critical
>Priority:       medium
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sun Dec 12 19:10:00 +0000 2010
>Last-Modified:  Wed Dec 22 19:10:03 +0000 2010
>Originator:     Arnaud Lacombe
>Release:        5.99.40 and 5.99.41
>Organization:
n/a
>Environment:
>Description:
-current cannot boot on

cpu0 at mainbus0 apid 0: Intel 686-class, 2154MHz, id 0x6fd

since at least Nov 21st 2010.

The kernel triggers a double-fault when execution start.
>How-To-Repeat:
boot a recent -current.
>Fix:

>Audit-Trail:
From: Martin Husemann <martin@duskware.de>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/44224: -current double-fault at boot
Date: Sun, 12 Dec 2010 23:10:40 +0100

 On Sun, Dec 12, 2010 at 07:10:01PM +0000, lacombar@gmail.com wrote:
 > cpu0 at mainbus0 apid 0: Intel 686-class, 2154MHz, id 0x6fd

 What kind of cpu is that exactly?
 Maybe that information would make it easier for somebody to find test
 hardware to reproduce the failure (other CPU types work just fine, at
 least for me).

 Martin

From: Arnaud Lacombe <lacombar@gmail.com>
To: gnats-bugs@netbsd.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Subject: Re: kern/44224: -current double-fault at boot
Date: Sun, 12 Dec 2010 17:58:55 -0500

 Hi,

 On Sun, Dec 12, 2010 at 5:15 PM, Martin Husemann <martin@duskware.de> wrote=
 :
 > The following reply was made to PR kern/44224; it has been noted by GNATS=
 .
 >
 > From: Martin Husemann <martin@duskware.de>
 > To: gnats-bugs@NetBSD.org
 > Cc:
 > Subject: Re: kern/44224: -current double-fault at boot
 > Date: Sun, 12 Dec 2010 23:10:40 +0100
 >
 > =A0On Sun, Dec 12, 2010 at 07:10:01PM +0000, lacombar@gmail.com wrote:
 > =A0> cpu0 at mainbus0 apid 0: Intel 686-class, 2154MHz, id 0x6fd
 >
 > =A0What kind of cpu is that exactly?
 Despite what the dmesg say, it should be the following CPU:

 Intel Core 2 (Merom) (686-class), 1795.62 MHz, id 0x6fd

 this would match the fact that `machdep.est.frequency.available' does
 not display any frequency above 1800MHz.

 Full dmesg is here:

 http://mail-index.netbsd.org/current-users/2010/04/17/msg013120.html

  - Arnaud

 ps: it would seem 5.99.25 misdetects the CPU, as 5.99.13 was detecting
 the right frequency.

From: David Holland <dholland-bugs@netbsd.org>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: kern/44224: -current double-fault at boot
Date: Tue, 21 Dec 2010 07:08:18 +0000

 (not sent to gnats)

    ------

 From: Arnaud Lacombe <lacombar@gmail.com>
 To: kern-bug-people@netbsd.org, gnats-admin@netbsd.org,
 	netbsd-bugs@netbsd.org, lacombar@gmail.com
 Subject: Re: kern/44224: -current double-fault at boot
 Date: Sun, 12 Dec 2010 23:07:05 -0500

 here is the (partial) screen output (with a PAE enabled kernel):

 mapped with 3 large  pages and 338 normal pages
 fault in user mode
 code c010caf9 eip 8 cs 246 eflags 10 cr2 23eba6dd ilevel 0
 trap double fault, code=0
 0.1 (system) at 0x8: invalid address

 The kernel is MULTIBOOT'ed by grub.


 From: Arnaud Lacombe <lacombar@gmail.com>
 To: kern-bug-people@netbsd.org, gnats-admin@netbsd.org,
 	netbsd-bugs@netbsd.org
 Subject: Re: kern/44224: -current double-fault at boot
 Date: Thu, 16 Dec 2010 12:39:37 -0500

 For the record, -current from yesterday still not boot. I rebuild the
 toolchain to be sure, but nothing much different. I cannot boot an
 amd64 on the machine as the bootloader (an old grub), do not know how
 to deal with 64bit ELF.


 From: Christos Zoulas <christos@zoulas.com>
 To: Arnaud Lacombe <lacombar@gmail.com>, kern-bug-people@netbsd.org,
 	gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
 Subject: Re: kern/44224: -current double-fault at boot
 Date: Thu, 16 Dec 2010 13:50:08 -0500

 On Dec 16, 12:39pm, lacombar@gmail.com (Arnaud Lacombe) wrote:
 -- Subject: Re: kern/44224: -current double-fault at boot

 | For the record, -current from yesterday still not boot. I rebuild the
 | toolchain to be sure, but nothing much different. I cannot boot an
 | amd64 on the machine as the bootloader (an old grub), do not know how
 | to deal with 64bit ELF.

 | > code c010caf9 eip 8 cs 246 eflags 10 cr2 23eba6dd ilevel 0

 Somehow the code seems to have jumped to NULL it seems.

 christos

From: Arnaud Lacombe <lacombar@gmail.com>
To: gnats-bugs@netbsd.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Subject: Re: kern/44224: -current double-fault at boot
Date: Wed, 22 Dec 2010 14:09:32 -0500

 For the record, I am no longer able to test this issue, the machine is
 question is no longer running NetBSD.

 On Tue, Dec 21, 2010 at 2:10 AM, David Holland <dholland-bugs@netbsd.org> w=
 rote:
 > The following reply was made to PR kern/44224; it has been noted by GNATS=
 .
 >
 > From: David Holland <dholland-bugs@netbsd.org>
 > To: gnats-bugs@netbsd.org
 > Cc:
 > Subject: Re: kern/44224: -current double-fault at boot
 > Date: Tue, 21 Dec 2010 07:08:18 +0000
 >
 > =A0(not sent to gnats)
 >
 > =A0 =A0------
 >
 > =A0From: Arnaud Lacombe <lacombar@gmail.com>
 > =A0To: kern-bug-people@netbsd.org, gnats-admin@netbsd.org,
 > =A0 =A0 =A0 =A0netbsd-bugs@netbsd.org, lacombar@gmail.com
 > =A0Subject: Re: kern/44224: -current double-fault at boot
 > =A0Date: Sun, 12 Dec 2010 23:07:05 -0500
 >
 > =A0here is the (partial) screen output (with a PAE enabled kernel):
 >
 > =A0mapped with 3 large =A0pages and 338 normal pages
 > =A0fault in user mode
 > =A0code c010caf9 eip 8 cs 246 eflags 10 cr2 23eba6dd ilevel 0
 > =A0trap double fault, code=3D0
 > =A00.1 (system) at 0x8: invalid address
 >
 > =A0The kernel is MULTIBOOT'ed by grub.
 >
 >
 > =A0From: Arnaud Lacombe <lacombar@gmail.com>
 > =A0To: kern-bug-people@netbsd.org, gnats-admin@netbsd.org,
 > =A0 =A0 =A0 =A0netbsd-bugs@netbsd.org
 > =A0Subject: Re: kern/44224: -current double-fault at boot
 > =A0Date: Thu, 16 Dec 2010 12:39:37 -0500
 >
 > =A0For the record, -current from yesterday still not boot. I rebuild the
 > =A0toolchain to be sure, but nothing much different. I cannot boot an
 > =A0amd64 on the machine as the bootloader (an old grub), do not know how
 > =A0to deal with 64bit ELF.
 >
 >
 > =A0From: Christos Zoulas <christos@zoulas.com>
 > =A0To: Arnaud Lacombe <lacombar@gmail.com>, kern-bug-people@netbsd.org,
 > =A0 =A0 =A0 =A0gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
 > =A0Subject: Re: kern/44224: -current double-fault at boot
 > =A0Date: Thu, 16 Dec 2010 13:50:08 -0500
 >
 > =A0On Dec 16, 12:39pm, lacombar@gmail.com (Arnaud Lacombe) wrote:
 > =A0-- Subject: Re: kern/44224: -current double-fault at boot
 >
 > =A0| For the record, -current from yesterday still not boot. I rebuild th=
 e
 > =A0| toolchain to be sure, but nothing much different. I cannot boot an
 > =A0| amd64 on the machine as the bootloader (an old grub), do not know ho=
 w
 > =A0| to deal with 64bit ELF.
 >
 > =A0| > code c010caf9 eip 8 cs 246 eflags 10 cr2 23eba6dd ilevel 0
 >
 > =A0Somehow the code seems to have jumped to NULL it seems.
 >
 > =A0christos
 >
 >

Home	PR Database Search