Bug 607283 – segfaults and/or miniobject warnings with GLib >= 2.23.2

After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.

Bug 607283 - segfaults and/or miniobject warnings with GLib >= 2.23.2


Summary:	segfaults and/or miniobject warnings with GLib >= 2.23.2


Status:	RESOLVED FIXED

Product:	GStreamer
Classification:	Platform
Component:	gstreamer (core)
Version:	unspecified
Hardware:	Other Linux

Importance:	Normal normal
Target Milestone:	0.10.26
Assigned To:	GStreamer Maintainers
QA Contact:	GStreamer Maintainers

URL:
Whiteboard:

Depends on:
Blocks:

Reported:	2010-01-18 09:34 UTC by Philippe Normand
Modified:	2010-01-18 13:13 UTC

See Also:
GNOME target:	---
GNOME version:	---

Attachments
GST_DEBUG=5 output (178.08 KB, application/x-bzip) 2010-01-18 09:34 UTC, Philippe Normand	Details

Description Philippe Normand 2010-01-18 09:34:36 UTC

Created attachment 151655 [details]
GST_DEBUG=5 output

gst-launch playbin2 uri=http://www.youtube.com/demo/google_main.mp4?2

Stack trace:

Program received signal SIGSEGV, Segmentation fault.
0x00007ffff7b5838c in gst_mini_object_replace (olddata=0x8878b0, newdata=0x61f250) at gstminiobject.c:392
392	  GST_CAT_LOG (GST_CAT_REFCOUNTING, "replace %p (%d) with %p (%d)",
(gdb) t a a bt

+ Trace 220117

Thread 1 (Thread 0x7ffff7fd16f0 (LWP 845))

#0 gst_mini_object_replace
at gstminiobject.c line 392
#1 gst_value_mini_object_collect
at gstminiobject.c line 450
#2 IA__g_signal_emit_valist
at gsignal.c line 2951
#3 IA__g_signal_emit
at gsignal.c line 3033
#4 gst_bus_async_signal_func
at gstbus.c line 1099
#5 gst_bus_source_dispatch
at gstbus.c line 762
#6 g_main_dispatch
at gmain.c line 1960
#7 IA__g_main_context_dispatch
at gmain.c line 2513
#8 g_main_context_iterate
at gmain.c line 2591
#9 IA__g_main_loop_run
at gmain.c line 2799
#10 gst_bus_poll
at gstbus.c line 1059
#11 event_loop
at gst-launch.c line 402
#12 main
at gst-launch.c line 820

Comment 1 Thiago Sousa Santos 2010-01-18 10:38:11 UTC

How reproducible is that? Does it happens always?

Tried playing it 5 times here, no occurrences. Are you using git head?

Comment 2 Sebastian Dröge (slomo) 2010-01-18 10:45:49 UTC

Same here, absolutely not reproducible with latest GIT or latest releases.

Comment 3 Tim-Philipp Müller 2010-01-18 10:59:40 UTC

Just to make sure (since this smells a lot like #606965), could you run:

 ldd /home/phil/gst/jhbuild/install/lib64/gstreamer-0.10/libgstplaybin.so

in the environment where you execute gst-launch?

Comment 4 Philippe Normand 2010-01-18 11:03:09 UTC

(In reply to comment #3)
> Just to make sure (since this smells a lot like #606965), could you run:
> 
>  ldd /home/phil/gst/jhbuild/install/lib64/gstreamer-0.10/libgstplaybin.so
> 
> in the environment where you execute gst-launch?

	linux-vdso.so.1 =>  (0x00007fff0b7ff000)
	libgstpbutils-0.10.so.0 => /home/phil/gst/jhbuild/install/lib64/libgstpbutils-0.10.so.0 (0x00007faad9be3000)
	libgstinterfaces-0.10.so.0 => /home/phil/gst/jhbuild/install/lib64/libgstinterfaces-0.10.so.0 (0x00007faad99d1000)
	libgstvideo-0.10.so.0 => /home/phil/gst/jhbuild/install/lib64/libgstvideo-0.10.so.0 (0x00007faad97c8000)
	libgstreamer-0.10.so.0 => /home/phil/gst/jhbuild/install/lib64/libgstreamer-0.10.so.0 (0x00007faad94e7000)
	libgobject-2.0.so.0 => /home/phil/gst/jhbuild/install/lib64/libgobject-2.0.so.0 (0x00007faad92a0000)
	libgmodule-2.0.so.0 => /home/phil/gst/jhbuild/install/lib64/libgmodule-2.0.so.0 (0x00007faad909b000)
	libxml2.so.2 => /home/phil/gst/jhbuild/install/lib64/libxml2.so.2 (0x00007faad8d4b000)
	libgthread-2.0.so.0 => /home/phil/gst/jhbuild/install/lib64/libgthread-2.0.so.0 (0x00007faad8b46000)
	librt.so.1 => /lib/librt.so.1 (0x00007faad891b000)
	libglib-2.0.so.0 => /home/phil/gst/jhbuild/install/lib64/libglib-2.0.so.0 (0x00007faad8636000)
	libpthread.so.0 => /lib/libpthread.so.0 (0x00007faad841a000)
	libc.so.6 => /lib/libc.so.6 (0x00007faad80aa000)
	libm.so.6 => /lib/libm.so.6 (0x00007faad7e26000)
	libgstbase-0.10.so.0 => /home/phil/gst/jhbuild/install/lib64/libgstbase-0.10.so.0 (0x00007faad7be7000)
	libdl.so.2 => /lib/libdl.so.2 (0x00007faad79e3000)
	libz.so.1 => /lib/libz.so.1 (0x00007faad77cb000)
	/lib64/ld-linux-x86-64.so.2 (0x00007faada034000)

Comment 5 Philippe Normand 2010-01-18 11:55:24 UTC

Reverting glib commit 0f25115f fixes this issue.

Comment 6 Sebastian Dröge (slomo) 2010-01-18 12:29:53 UTC


*** This bug has been marked as a duplicate of bug 603590 ***

Comment 7 Sebastian Dröge (slomo) 2010-01-18 12:59:17 UTC

Actually not a duplicate. The mini object collect function should not assume that the destination is initialized.

commit ed1247b561b02bf8066ba971a34f6d62e03899f5
Author: Sebastian Dröge <sebastian.droege@collabora.co.uk>
Date:   Mon Jan 18 13:57:29 2010 +0100

    miniobject: The GValue collection function can not assume that the destinati
    
    ...and it will usually be either filled by zeroes or random values.
    
    Fixes bug #607283.