Bug 100670 – RB hanging with gstreamer CVS

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 100670 - RB hanging with gstreamer CVS


Summary:	RB hanging with gstreamer CVS


Status:	RESOLVED FIXED

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

Importance:	Normal critical
Target Milestone:	0.4.0
Assigned To:	GStreamer Maintainers
QA Contact:	GStreamer Maintainers

URL:
Whiteboard:

Depends on:
Blocks:

Reported:	2002-12-08 17:38 UTC by Olivier Martin
Modified:	2004-12-22 21:47 UTC

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

Attachments
trace (14.17 KB, text/plain) 2002-12-08 17:39 UTC, Olivier Martin	Details
another trace with the crash (135.72 KB, text/plain) 2002-12-08 17:52 UTC, Olivier Martin	Details
hang with correct mime (14.17 KB, application/octet-stream) 2002-12-08 18:05 UTC, Olivier Martin	Details
crash with correct mime (135.72 KB, application/octet-stream) 2002-12-08 18:06 UTC, Olivier Martin	Details

Description Olivier Martin 2002-12-08 17:38:46 UTC

Hi there,

since you're about to release 0.5, I tested the current cvs (08/12/02) with
Rhythmbox to be able to release RB 0.4.1 soon.

Apparently there are still issues with the scheduler, rb hangs when trying
to play a song. When being debugged it plays one, and when going to the
next it'll crash with this bt:

Program received signal SIGSEGV, Segmentation fault.

+ Trace 31461

Thread 32773 (LWP 3524)

#0 cothread_setfunc
at cothreads.c line 417
#1 gst_basic_scheduler_cothreaded_chain
at gstbasicscheduler.c line 687
#2 gst_basic_scheduler_state_transition
at gstbasicscheduler.c line 1083
#3 gst_scheduler_state_transition
at gstscheduler.c line 292
#4 gst_element_change_state
at gstelement.c line 2107
#5 gst_element_set_state
at gstelement.c line 1938
#6 gst_bin_change_state
at gstbin.c line 582
#7 gst_thread_update_state
at gstthread.c line 310
#8 gst_thread_main_loop
at gstthread.c line 657
#9 g_thread_create_proxy
from /usr/lib/libglib-2.0.so.0
#10 pthread_start_thread
from /lib/i686/libpthread.so.0
#11 pthread_start_thread_event
from /lib/i686/libpthread.so.0

Comment 1 Olivier Martin 2002-12-08 17:39:24 UTC

Created attachment 12844 [details]
trace

Comment 2 Olivier Martin 2002-12-08 17:52:12 UTC

Created attachment 12845 [details]
another trace with the crash

Comment 3 Olivier Martin 2002-12-08 18:05:36 UTC

Created attachment 12846 [details]
hang with correct mime

Comment 4 Olivier Martin 2002-12-08 18:06:23 UTC

Created attachment 12847 [details]
crash with correct mime

Comment 5 Olivier Martin 2002-12-08 19:15:42 UTC

Attachements are .gz files...

Comment 6 Olivier Martin 2002-12-08 22:37:02 UTC

Same issues with 0.4.2.2 prerelease...

Comment 7 Wim Taymans 2002-12-08 22:43:48 UTC

can you retry with --gst-scheduler=basicwingo?

Comment 8 Olivier Martin 2002-12-08 23:19:28 UTC

tried:
rhythmbox --gst-scheduler=basicwingo

doesn't seem to help...

Comment 9 David Schleef 2002-12-10 04:12:52 UTC

I think I fixed this by reverting wtay's last commit to gstpipeline.c

Comment 10 Olivier Martin 2002-12-10 08:02:38 UTC

It's not fixed, i have exactly the same crashes... Was RB working for you?

Comment 11 Olivier Martin 2002-12-10 17:55:51 UTC

These issues seem to be fixed with the latest prerelease src (0.4.2.3)

However i'm experiencing a new one for mp3 (oggs seem fine) 

=============================== BT ==================================

Program received signal SIGINT, Interrupt.

+ Trace 31540

Thread 24580 (LWP 13948)

#0 kill
from /lib/i686/libc.so.6
#1 raise
from /lib/i686/libpthread.so.0
#2 rb_debug_stop_in_debugger
at rb-debug.c line 137
#3 log_handler
at rb-debug.c line 154
#4 g_logv
from /usr/lib/libglib-2.0.so.0
#5 g_log
from /usr/lib/libglib-2.0.so.0
#6 gst_mad_chain
at gstmad.c line 1050
#7 gst_basic_scheduler_chain_wrapper
at gstbasicscheduler.c line 335
#8 cothread_stub
at cothreads.c line 493
#9 cothread_switch
at cothreads.c line 664
#10 gst_basic_scheduler_gethandler_proxy
at gstbasicscheduler.c line 500
#11 gst_pad_pull
at gstpad.c line 2157
#12 gst_basic_scheduler_chain_wrapper
at gstbasicscheduler.c line 327
#13 cothread_stub
at cothreads.c line 493
#14 cothread_switch
at cothreads.c line 664
#15 gst_basic_scheduler_iterate
at gstbasicscheduler.c line 1328
#16 gst_scheduler_iterate
at gstscheduler.c line 624
#17 gst_bin_iterate_func
at gstbin.c line 826

Comment 12 Thomas Vander Stichele 2002-12-10 18:05:46 UTC

that warning is gone from the release.
if it causes a problem with the final 0.5.0 release, then reopen a
different bug instead ;) that way we can better track it.

Comment 13 Olivier Martin 2002-12-10 18:14:02 UTC

Now that the warning is gone, here is the real bug:

rogram received signal SIGSEGV, Segmentation fault.

+ Trace 31541

Thread 32773 (LWP 14299)

#0 cothread_setfunc
at cothreads.c line 417
#1 gst_basic_scheduler_cothreaded_chain
at gstbasicscheduler.c line 687
#2 gst_basic_scheduler_state_transition
at gstbasicscheduler.c line 1083
#3 gst_scheduler_state_transition
at gstscheduler.c line 292
#4 gst_element_change_state
at gstelement.c line 2107
#5 gst_element_set_state
at gstelement.c line 1938
#6 gst_bin_change_state
at gstbin.c line 582
#7 gst_thread_update_state
at gstthread.c line 310
#8 gst_thread_main_loop
at gstthread.c line 657
#9 g_thread_create_proxy
from /usr/lib/libglib-2.0.so.0
#10 pthread_start_thread
from /lib/i686/libpthread.so.0
#11 pthread_start_thread_event
from /lib/i686/libpthread.so.0

Comment 14 Olivier Martin 2002-12-11 07:21:05 UTC

Fixed in the 0.5 release.