NetBSD Problem Report #51105

From martin@aprisoft.de  Sat Apr 30 13:32:43 2016
Return-Path: <martin@aprisoft.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 73E3E7A3DB
	for <gnats-bugs@gnats.NetBSD.org>; Sat, 30 Apr 2016 13:32:43 +0000 (UTC)
Message-Id: <20160430133233.AFFD7ED0E4F@emmas.aprisoft.de>
Date: Sat, 30 Apr 2016 15:32:33 +0200 (CEST)
From: martin@NetBSD.org
Reply-To: martin@NetBSD.org
To: gnats-bugs@NetBSD.org
Subject: python 2.7 crashes while building gstreamer
X-Send-Pr-Version: 3.95

>Number:         51105
>Category:       toolchain
>Synopsis:       python 2.7 crashes while building gstreamer
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    toolchain-manager
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sat Apr 30 13:35:00 +0000 2016
>Closed-Date:    Fri Dec 21 08:05:27 +0000 2018
>Last-Modified:  Fri Dec 21 08:05:27 +0000 2018
>Originator:     Martin Husemann
>Release:        NetBSD 7.99.29
>Organization:
The NetBSD foundation, Inc.
>Environment:
System: NetBSD martins.aprisoft.de 7.99.29 NetBSD 7.99.29 (GENERIC) #65: Fri Apr 29 10:12:01 CEST 2016 martin@martins.aprisoft.de:/ssd/src/sys/arch/amd64/compile/GENERIC amd64
Architecture: x86_64
Machine: amd64
>Description:

After removing all old pkgs and trying to rebuild everything now that gcc 5.3
is used in base, I ran into this failure in gstreamer 0.10:

libtool: link: gcc -o /ssd/pkgobj/multimedia/gstreamer0.10/work/gstreamer-0.10.36/gst/tmp-introspectyaQJr7/.libs/Gst-0.10 -I/ssd/pkgobj/multimedia/gstreamer0.10/work/.buildlink/include -O2 -pipe -Wl,-R/usr/pkg/lib tmp-introspectyaQJr7/ssd/pkgobj/multimedia/gstreamer0.10/work/gstreamer-0.10.36/gst/tmp-introspectyaQJr7/Gst-0.10.o -Wl,--export-dynamic -pthread -Wl,--export-dynamic  -L/ssd/pkgobj/multimedia/gstreamer0.10/work/.buildlink/lib -L. -L./.libs -lgstreamer-0.10 -lc -lgio-2.0 -lgobject-2.0 -lffi -lgmodule-2.0 -lgthread-2.0 -lglib-2.0 -lpcre -lintl -lxml2 -lz -llzma -lm -pthread -Wl,-rpath,/usr/pkg/lib -Wl,-rpath,/ssd/pkgobj/multimedia/gstreamer0.10/work/.buildlink/lib
g-ir-scanner: Gst: warning: 125 warnings suppressed (use --warn-all to see them)
[1]   Segmentation fault (core dumped) GST_PLUGIN_SYSTE...
Makefile:1784: recipe for target 'Gst-0.10.gir' failed
gmake[4]: *** [Gst-0.10.gir] Error 139

Core was generated by `python2.7'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0  0x00007f7feca05643 in ?? ()
(gdb) bt
#0  0x00007f7feca05643 in ?? ()
#1  0x00007f7fec6005a5 in ?? () from /usr/pkg/lib/libgthread-2.0.so.0
#2  0x00007f7ff7e20a40 in ?? ()
#3  0x00007f7fec600689 in _fini () from /usr/pkg/lib/libgthread-2.0.so.0
#4  0x0000000000000000 in ?? ()



>How-To-Repeat:
Build multimedia/gstreamer0.10 with python2.7 build by latest netbsd-current
on amd64.

>Fix:
n/a

>Release-Note:

>Audit-Trail:
From: Martin Husemann <martin@duskware.de>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: pkg/51105: python 2.7 crashes while building gstreamer
Date: Sat, 30 Apr 2016 18:43:52 +0200

 Obviously this is in dtor handling when unloading libgthread.
 Even when build with symbols, it does not show a lot more:

 #0  0x00007f7feca05643 in ?? ()
 #1  0x00007f7fec6005a5 in ?? () from /usr/pkg/lib/libgthread-2.0.so.0
 (gdb) x/16i 0x00007f7fec600500
    0x7f7fec600500:      pushq  0x200562(%rip)        # 0x7f7fec800a68
    0x7f7fec600506:      jmpq   *0x200564(%rip)        # 0x7f7fec800a70
    0x7f7fec60050c:      nopl   0x0(%rax)
    0x7f7fec600510 <g_assertion_message_expr@plt>:       
     jmpq   *0x200562(%rip)        # 0x7f7fec800a78
    0x7f7fec600516 <g_assertion_message_expr@plt+6>:     pushq  $0x0
    0x7f7fec60051b <g_assertion_message_expr@plt+11>:    jmpq   0x7f7fec600500
    0x7f7fec600520 <g_log@plt>:  jmpq   *0x20055a(%rip)        # 0x7f7fec800a80
    0x7f7fec600526 <g_log@plt+6>:        pushq  $0x1
    0x7f7fec60052b <g_log@plt+11>:       jmpq   0x7f7fec600500
    0x7f7fec600530:      jmpq   *0x20050a(%rip)        # 0x7f7fec800a40
    0x7f7fec600536:      xchg   %ax,%ax
    0x7f7fec600538:      jmpq   *0x20050a(%rip)        # 0x7f7fec800a48
    0x7f7fec60053e:      xchg   %ax,%ax
 [..]
    0x7f7fec600562:      cmpq   $0x0,0x2004e6(%rip)        # 0x7f7fec800a50
    0x7f7fec60056a:      je     0x7f7fec600578
    0x7f7fec60056c:      
     mov    0x200515(%rip),%rdi        # 0x7f7fec800a88 <__dso_handle>
    0x7f7fec600573:      callq  0x7f7fec600540
    0x7f7fec600578:      lea    0x2002f1(%rip),%rbx        # 0x7f7fec800870
    0x7f7fec60057f:      mov    (%rbx),%rax
    0x7f7fec600582:      test   %rax,%rax
    0x7f7fec600585:      je     0x7f7fec60058f
    0x7f7fec600587:      callq  *%rax
    0x7f7fec600589:      add    $0x8,%rbx
    0x7f7fec60058d:      jmp    0x7f7fec60057f
    0x7f7fec60058f:      cmpq   $0x0,0x2004a9(%rip)        # 0x7f7fec800a40
    0x7f7fec600597:      je     0x7f7fec6005a5
    0x7f7fec600599:      lea    0x250(%rip),%rdi        # 0x7f7fec6007f0
    0x7f7fec6005a0:      callq  0x7f7fec600530
    0x7f7fec6005a5:      pop    %rbx
    0x7f7fec6005a6:      retq   

 (gdb) x/i 0x7f7fec600530
    0x7f7fec600530:      jmpq   *0x20050a(%rip)        # 0x7f7fec800a40


 Martin

State-Changed-From-To: open->closed
State-Changed-By: martin@NetBSD.org
State-Changed-When: Sun, 01 May 2016 08:45:20 +0000
State-Changed-Why:
Updated the base system and rebuild glib2 - that fixed it.


Responsible-Changed-From-To: pkg-manager->toolchain-manager
Responsible-Changed-By: martin@NetBSD.org
Responsible-Changed-When: Sun, 01 May 2016 17:23:34 +0000
Responsible-Changed-Why:
This is a toolchain issue, maybe related to broken weak symbol handling?


State-Changed-From-To: closed->open
State-Changed-By: martin@NetBSD.org
State-Changed-When: Sun, 01 May 2016 17:23:34 +0000
State-Changed-Why:
After rebuilding libgthread-2 again, the issue is back (and now persistent
again). Seem that it does only happen sometimes when building the lib,


From: matthew green <mrg@eterna.com.au>
To: gnats-bugs@NetBSD.org
Cc: toolchain-manager@netbsd.org, pkg-manager@netbsd.org,
    netbsd-bugs@netbsd.org, gnats-admin@netbsd.org, martin@NetBSD.org
Subject: re: toolchain/51105 (python 2.7 crashes while building gstreamer)
Date: Mon, 02 May 2016 05:10:28 +1000

 FWIW (martin knows this, but gnats doesn't), i have not had any problem
 with gstreamer on amd64 -current with gcc 5.3, system and packages all
 built during this prior week.

 i've heard others also have success, so currently my suspicion is that
 something in martin's build is broken somehow, in ways that don't break
 for other people.


 .mrg.

From: Martin Husemann <martin@NetBSD.org>
To: matthew green <mrg@eterna.com.au>
Cc: gnats-bugs@NetBSD.org
Subject: Re: toolchain/51105 (python 2.7 crashes while building gstreamer)
Date: Mon, 2 May 2016 07:05:41 +0000

 On Mon, May 02, 2016 at 05:10:28AM +1000, matthew green wrote:
 > FWIW (martin knows this, but gnats doesn't), i have not had any problem
 > with gstreamer on amd64 -current with gcc 5.3, system and packages all
 > built during this prior week.
 > 
 > i've heard others also have success, so currently my suspicion is that
 > something in martin's build is broken somehow, in ways that don't break
 > for other people.

 I started again from scratch and this time gstreamer-0.10 did build, but
 the same crash happens when building x11/libxklavier.

 Reading symbols from python2.7...done.
 [New process 1]
 Core was generated by `python2.7'.
 Program terminated with signal SIGSEGV, Segmentation fault.
 #0  0x00007f7feca05643 in ?? ()
 (gdb) bt
 #0  0x00007f7feca05643 in ?? ()
 #1  0x00007f7fec6005a5 in ?? () from /usr/pkg/lib/libgthread-2.0.so.0
 #2  0x00007f7ff7e20a40 in ?? ()
 #3  0x00007f7fec600689 in _fini () from /usr/pkg/lib/libgthread-2.0.so.0
 #4  0x0000000000000000 in ?? ()


 I have security.pax.aslr.global = 0, so it is not that.

 Martin

From: Stefan Hertenberger <stefan@hertenberger.bayern>
To: gnats-bugs@NetBSD.org
Cc: martin@NetBSD.org
Subject: Re: toolchain/51105 (python 2.7 crashes while building gstreamer)
Date: Wed, 01 Jun 2016 16:23:43 +0200

 Hello,

 i have the same problem on a NetBSD current machine while building 
 xfce4.

 Stefan

From: Martin Husemann <martin@duskware.de>
To: Stefan Hertenberger <stefan@hertenberger.bayern>
Cc: gnats-bugs@NetBSD.org
Subject: Re: toolchain/51105 (python 2.7 crashes while building gstreamer)
Date: Wed, 1 Jun 2016 18:48:04 +0200

 On Wed, Jun 01, 2016 at 04:23:43PM +0200, Stefan Hertenberger wrote:
 > Hello,
 > 
 > i have the same problem on a NetBSD current machine while building 
 > xfce4.

 Oh, good, so it is not just me - I still see it on one machine (but not
 others, and have no idea what the difference is).

 Martin

State-Changed-From-To: open->closed
State-Changed-By: martin@NetBSD.org
State-Changed-When: Fri, 21 Dec 2018 08:05:27 +0000
State-Changed-Why:
This has been fixed


>Unformatted:

Home	PR Database Search