GNOME Bugzilla – Bug 413635
gaim crashed with SIGSEGV in gst_audio_clock_new()
Last modified: 2007-05-03 11:02:48 UTC
That bug has been opened on https://launchpad.net/bugs/86489 "Binary package hint: gaim just crashed while restarting session ... " Debug backtrace for the crash:
+ Trace 115041
Thread 1 (process 7394)
yyresult = <value optimized out> yyerrstatus = 0 yytoken = 4 yymsgbuf = "\000���\000\222��9~��\000\222������i\204�������R���T��X\030���\024\220���������X\030﷼T��\020���B���\020\032��\000\000\000\000\001\000\000\000\001\000\000\000\000\000\000\000\226���\037\003\000\000\000���\004�\t\000�T��\000\222����\220��\032\220������R yymsg = 0xb4eec198 "" yymsg_alloc = 128 yyssa = {0, 3, 16, 2, -2718, -18461, -4892, -18556, -17701, -18449, 26296, -18680, -12300, -18448, 23140, 2069, 1, 0, -16216, -19218, -16574, -18449, 23212, 2069, 19, 0, -16200, -19218, -12300, -18448, 1, 0, -17397, 7, -16268, -19218, -16268, -19218, -16028, -19218, -17701, -18449, 6232, -18449, 57, 0, 8908, -18461, 5044, -18461, 0, 0, 25832, -2175, -15996, -19218, -28934, -7393, 0, 0, -12300, -18448, 23140, 2069, 1, 0, -16120, -19218, -16574, -18449, -3177, -18461, 31452, -18461, 15996, -18461, 1, 0, 17, 0, 29806, 3, -16172, -19218, -16172, -19218, -15932, -19218, -16233, 28301, 6232, -18449, 57, 0, 8908, -18461, 6468, -18461, 1, 0, -16234, 28301, 1, 0, 21804, -18461, 0, 0, 0, 0, 1, 0, 799, 0, 0, 0, 6232, -18449, -2922, -18461, 31452, -18461, 21692, -18461, 1, 0, -12300, -18448, 6672, -18449, -15944, -19218, -15916, -19218, -16125, -18449, 21692, -18461, -15944, -19218, -10180, -18448, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 0, -15904, -19218, 6232, -18449, -3120, -18461, 6162, 3, -15932, -19218, 24, 0, 28, -20080, -12, -18577, -15872, -19218, 6232, -18449, -2922, -18461, -14866, 98, -16233, 28301, 0, 0, -28160, -18452, 5352, -20080, 5352, -20080, -15992, -19218, -32179, -18454, 4, 0, -28160, -18452, -15960, -19218, -31963, -18454} yyss = (yytype_int16 *) 0xb4eec000 yyssp = (yytype_int16 *) 0xb4eec006 ...
The stack trace looks good and there are two seemingly identical crashers, but it somehow just doesn't make sense to me. - g_object_new() can't fail AFAIK - obj->foo=x can't fail either if obj is a valid allocated object struct as g_object_new() should return it Maybe gaim or something else is thrashing the memory? Is this bug reproducable? What would also be interesting is the value of the 'aclock' variable in the gst_audio_clock_new() function/frame when it crashes.
Closing as INCOMPLETE for now, since there isn't really much we can do about it with the info we have and it JustDoesn'tReallyMakeSense(tm) code-wise. If someone can reproduce this, please run gaim in valgrind to see what comes up there.