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 346663 - Segmentation error with fakesrc, fakesink and a filter
Segmentation error with fakesrc, fakesink and a filter
Status: VERIFIED FIXED
Product: GStreamer
Classification: Platform
Component: gstreamer (core)
0.10.x
Other All
: Normal critical
: 0.10.9
Assigned To: Wim Taymans
GStreamer Maintainers
Depends on:
Blocks:
 
 
Reported: 2006-07-05 15:44 UTC by Peter Kjellerstedt
Modified: 2006-07-18 10:42 UTC
See Also:
GNOME target: ---
GNOME version: ---


Description Peter Kjellerstedt 2006-07-05 15:44:32 UTC 
Steps to reproduce:
I get a segmentation error when I run the following command:

gst-launch fakesrc ! 'video/x-raw-yuv,format=(fourcc)YUY2,width=176,height=144' ! ffmpegcolorspace ! fakesink

It seems to be caused by gst_base_transform_prepare_output_buffer() being called with an in_buf with size 0 and out_size being 0, which causes a call to gst_buffer_create_sub() with size being 0, which in turn causes it to return NULL, which isn't handled...


Stack trace:

+ Trace 69204

  • #0 gst_caps_replace
    at ../../../gstreamer/gst/gstcaps.c line 1581
  • #1 gst_base_transform_prepare_output_buffer
    at ../../../../../gstreamer/libs/gst/base/gstbasetransform.c line 902
  • #2 gst_base_transform_handle_buffer
    at ../../../../../gstreamer/libs/gst/base/gstbasetransform.c line 1337
  • #3 gst_base_transform_chain
    at ../../../../../gstreamer/libs/gst/base/gstbasetransform.c line 1457
  • #4 gst_pad_chain
    at ../../../gstreamer/gst/gstpad.c line 3232
  • #5 gst_pad_push
    at ../../../gstreamer/gst/gstpad.c line 3348
  • #6 gst_base_src_loop
    at ../../../../../gstreamer/libs/gst/base/gstbasesrc.c line 1497
  • #7 gst_task_func
    at ../../../gstreamer/gst/gsttask.c line 193
  • #8 g_thread_pool_thread_proxy
    at ../../../glib/glib/gthreadpool.c line 265
  • #9 g_thread_create_proxy
    at ../../../glib/glib/gthread.c line 582
  • #10 pthread_start_thread
    at libpthread/linuxthreads.old/manager.c line 311
  • #11 pthread_start_thread_event
    at libpthread/linuxthreads.old/manager.c line 335
  • #12 clone
    at libc/sysdeps/linux/cris/clone.S line 77 


Other information:
The trace above was attained on an embedded box running on the CRIS architecture and using uClibc. However, the same segmentation error occurs with ix86 and glibc.
Comment 1 Wim Taymans 2006-07-05 17:02:41 UTC 
yeah, that subbuffer size > 0 check is unneeded. working an fix and unit tests for various other 0 size related buffer stuff.
Comment 2 Wim Taymans 2006-07-05 17:11:10 UTC 
This should fix the assert. No guarantee that the next element will handle a 0 sized buffer.


        * gst/gstbuffer.c: (_gst_buffer_copy), (gst_buffer_create_sub):
        * tests/check/gst/gstbuffer.c: (GST_START_TEST),
        (gst_buffer_suite):
        No point in checking if the size of the subbuffer > 0, the
        code handles it correclty as demonstrated by unit test.
        Also add a unit test for the zero sized _new_and_alloc and
        _copy. Fixes #346663.