NetBSD Problem Report #30519

From syn@sceen.net  Mon Jun 13 20:34:13 2005
Return-Path: <syn@sceen.net>
Received: from mail.sceen.net (sceen.net [213.41.131.45])
	by narn.netbsd.org (Postfix) with ESMTP id 922ED63B11E
	for <gnats-bugs@gnats.NetBSD.org>; Mon, 13 Jun 2005 20:34:12 +0000 (UTC)
Message-Id: <20050613203405.43B5A3B16E@kenobi.sceen>
Date: Mon, 13 Jun 2005 22:34:05 +0200 (CEST)
From: syn@sceen.net
Reply-To: syn@sceen.net
To: gnats-bugs@netbsd.org
Subject: Sound problem and panic with ogg123 (pthread)
X-Send-Pr-Version: 3.95

>Number:         30519
>Category:       port-sparc64
>Synopsis:       Sound problem and panic with ogg123 (pthread) (ultra 1 only)
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    port-sparc64-maintainer
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Mon Jun 13 20:35:01 +0000 2005
>Closed-Date:    
>Last-Modified:  Sun Oct 08 22:42:04 +0000 2006
>Originator:     syn@sceen.net
>Release:        NetBSD 2.0.2_STABLE
>Organization:
Richard Braun
>Environment:
System: NetBSD kenobi 2.0.2_STABLE NetBSD 2.0.2_STABLE (KENOBI) #0: Sun Jun 12 14:50:07 CEST 2005 syn@kenobi:/usr/src/sys/arch/sparc64/compile/KENOBI sparc64
Architecture: sparc64
Machine: sparc64
>Description:

There is a strange bug with ogg123 from audio/vorbis-tools on NetBSD/sparc64
2.0.2. The problem is that I hear strange noises while ogg123 is playing.
I don't have the problem when using sox to convert from ogg to au format
and then using audioplay. Logically, I don't have the problem when playing
a native au file directly with audioplay. However, and this is more
worrying, ogg123 -d au -f test.au test.ogg made the host crash after a few
seconds. The kernel still responded to icmp echo queries, so the kernel
looked like still alive, but TCP connections were all freezed, and no reply
at all could be seen when initiating new connections. I can't tell about
the console, as this host is headless, and I don't have serial access yet.

>How-To-Repeat:

ogg123 test.ogg on an Ultra1 host with NetBSD/sparc64 2.0.2, and a CS4231A
sound chip, with audio/libao-sun plugin.

>Fix:

>Release-Note:

>Audit-Trail:
From: Richard Braun <syn@sceen.net>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: port-sparc64/30519
Date: Fri, 17 Jun 2005 10:33:05 +0200

 After another crash, here is what I noticed in dmesg :

 panic: fpe without being the FP user
 syncing disks...

 Seems like ogg123 awakens a FPU managment bug.

 --
 Richard Braun


From: Richard Braun <syn@sceen.net>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: port-sparc64/30519
Date: Sun, 19 Jun 2005 12:41:45 +0200

 I should also mention that I had no problem when I was using
 NetBSD/sparc64 1.6.2_STABLE, if comparing the changes between the two branc=
 hes
 can help.

 --
 Richard Braun


State-Changed-From-To: open->feedback
State-Changed-By: mrg@netbsd.org
State-Changed-When: Fri, 06 Oct 2006 03:11:26 +0000
State-Changed-Why:
when you say "ultra 1 only", do you mean you can not reproduce this
problem on other systems?  i am unable to reproduce it on an ultra2
or ultra60 with -current.  is this still a problem for you?

thanks.


From: matthew green <mrg@eterna.com.au>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: port-sparc64/30519
Date: Mon, 09 Oct 2006 07:15:51 +1000

 i've had 4 copies of ogg123 running in parallel for 2.5 days
 without any kernel crashes.  ogg123 itself sometimes dumps
 core but the kernel is OK.

From: Martin Husemann <martin@duskware.de>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: port-sparc64/30519
Date: Sun, 8 Oct 2006 23:24:12 +0200

 Ogg123 also still produces wrong output.

 Martin

State-Changed-From-To: feedback->open
State-Changed-By: mrg@netbsd.org
State-Changed-When: Sun, 08 Oct 2006 22:42:04 +0000
State-Changed-Why:
chuq reports that it *is* easy to reproduce on an ultra1.


>Unformatted:

Home	PR Database Search