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 674907 - CoglPangoGlyphCache.using_global_atlas used uninitialized
CoglPangoGlyphCache.using_global_atlas used uninitialized
Status: RESOLVED DUPLICATE of bug 671984
Product: cogl
Classification: Platform
Component: cogl-pango
unspecified
Other Linux
: Normal normal
: ---
Assigned To: Cogl maintainer(s)
Cogl maintainer(s)
Depends on:
Blocks:
 
 
Reported: 2012-04-26 22:40 UTC by Owen Taylor
Modified: 2012-05-16 13:21 UTC
See Also:
GNOME target: ---
GNOME version: ---



Description Owen Taylor 2012-04-26 22:40:23 UTC
From valgrind

==25138== Conditional jump or move depends on uninitialised value(s)
==25138==    at 0x41058B75: cogl_pango_glyph_cache_lookup (cogl-pango-glyph-cache.c:235)
==25138==    by 0x41057633: _cogl_pango_ensure_glyph_cache_for_layout_line_internal (cogl-pango-render.c:468)
==25138==    by 0x41058334: cogl_pango_ensure_glyph_cache_for_layout (cogl-pango-render.c:628)
==25138==    by 0x4144293D: clutter_text_create_layout (clutter-text.c:785)
==25138==    by 0x41442E91: clutter_text_get_preferred_width (clutter-text.c:2279)
==25138==    by 0x413CB009: clutter_actor_get_preferred_width (clutter-actor.c:5948)
==25138==    by 0x4F54D050: _st_actor_get_preferred_width (st-private.c:61)
==25138==    by 0x4F53D4DC: st_bin_get_preferred_width (st-bin.c:177)
==25138==    by 0x413CB009: clutter_actor_get_preferred_width (clutter-actor.c:5948)
==25138==    by 0x4F54D050: _st_actor_get_preferred_width (st-private.c:61)
==25138==    by 0x4F53E923: get_content_preferred_width (st-box-layout.c:338)
==25138==    by 0x4F53F96E: st_box_layout_get_preferred_width (st-box-layout.c:368)
==25138==

suggest just using g_new0 to allocate the structure instead of g_malloc, to correspond to GObject - otherwise, every newly added field will be a potential trap.
Comment 1 Neil Roberts 2012-05-16 13:21:50 UTC
Looks like this has been fixed in 1.10.0

*** This bug has been marked as a duplicate of bug 671984 ***