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 564136 - [videotestsrc] memory leak in _base_init()
[videotestsrc] memory leak in _base_init()
Status: RESOLVED FIXED
Product: GStreamer
Classification: Platform
Component: common
git master
Other Linux
: Normal blocker
: NONE
Assigned To: GStreamer Maintainers
GStreamer Maintainers
Depends on:
Blocks:
 
 
Reported: 2008-12-11 16:18 UTC by Edward Hervey
Modified: 2008-12-12 10:00 UTC
See Also:
GNOME target: ---
GNOME version: ---



Description Edward Hervey 2008-12-11 16:18:00 UTC
The PadTemplate and caps are leaked in base_init().

==19462== 11,896 (56 direct, 11,840 indirect) bytes in 1 blocks are definitely lost in loss record 3,140 of 3,449
==19462==    at 0x4C2299E: malloc (vg_replace_malloc.c:207)
==19462==    by 0x669E472: g_malloc (in /usr/lib64/libglib-2.0.so.0.1800.3)
==19462==    by 0x66B3757: g_slice_alloc (in /usr/lib64/libglib-2.0.so.0.1800.3)
==19462==    by 0x506742E: gst_caps_new_empty (gstcaps.c:140)
==19462==    by 0x5067834: gst_caps_copy (gstcaps.c:277)
==19462==    by 0x75C3C9C: gst_video_test_src_base_init (gstvideotestsrc.c:167)
==19462==    by 0x5326EEC: g_type_class_ref (in /usr/lib64/libgobject-2.0.so.0.1800.3)
==19462==    by 0x507463D: gst_element_register (gstelementfactory.c:286)
==19462==    by 0x5090ED6: gst_plugin_register_func (gstplugin.c:334)
==19462==    by 0x5091776: gst_plugin_load_file (gstplugin.c:545)
==19462==    by 0x5091D32: gst_plugin_load_by_name (gstplugin.c:963)
==19462==    by 0x5092D25: gst_plugin_feature_load (gstpluginfeature.c:110)
==19462==    by 0x5074278: gst_element_factory_create (gstelementfactory.c:386)
==19462==    by 0x5074B6E: gst_element_factory_make (gstelementfactory.c:474)
==19462==    by 0x50C30E8: _gst_parse_yyparse (grammar.y:568)
==19462==    by 0x50C46F7: _gst_parse_launch (grammar.y:837)
==19462==    by 0x50B80C9: gst_parse_launch_full (gstparse.c:293)
==19462==    by 0x40389B: test_granulepos_offset (theoraenc.c:141)
==19462==    by 0x40428B6: srunner_run_all (in /usr/lib64/libcheck.so.0.0.0)
==19462==    by 0x40136C: main (theoraenc.c:458)
Comment 1 Wim Taymans 2008-12-11 16:27:10 UTC
Please rerun without the slice allocator like G_SLICE=always-malloc
Comment 2 Edward Hervey 2008-12-11 16:34:06 UTC
this is the output of make check-valgrind from the g64-unstable buildslave

http://pitivi.org:8010/builders/g64~-base-full/builds/3/steps/make-check-valgrind/logs/stdio
Comment 3 Edward Hervey 2008-12-12 10:00:51 UTC
Added an assertion in common for this, since it will create dynamic pad
templates.

2008-12-12  Edward Hervey  <edward.hervey@collabora.co.uk>

        * gst.supp:
        A whole bunch of suppressions detected on latest gentoo ~amd64.
        Make some existing suppressions more generic (for subtle dependecy
        code changes).
        Added suppressions for glibc-2.9.
        Added suppressions for new variants of ALSA leaks.
        Added suppressions for a series of leaks in plugins registrations due
        to some pad templates' caps calculated at runtime.
        Added suppressions for variants of some leaks in pango/fontconfig.
        Added suppressions for leak in gstffmpegcsp.c (nothing we can do
        about it, but will only exist once).