I'm using Xfmedia 0.8.0 and 0.9.0, and right after start it segfaults. This is on Mandriva Cooker, with gcc-4.0.1, glibc-2.3.5, xorg-6.9cvs, cairo-1.0.0, gtk+-2.8.3 and Xfce-4.2.2 and kernel-2.6.12-10mdk. By just downgrading gtk+ to 2.6.4 I can get xfmedia working. With 0.9.0 I can get no backtrace, but I can get one with 0.8.0, both pasted below. I hope this info is somewhat helpfull. Gdb with 0.9.0: (gdb) run Starting program: /usr/bin/xfmedia Reading symbols from shared object read from target memory...done. Loaded system supplied DSO at 0xffffe000 [Thread debugging using libthread_db enabled] [New Thread -1219295552 (LWP 17947)] [New Thread -1234068560 (LWP 17955)] [New Thread -1242793040 (LWP 17956)] [New Thread -1255433296 (LWP 17957)] [New Thread -1267926096 (LWP 17958)] [New Thread -1278207056 (LWP 17959)] [New Thread -1286599760 (LWP 17960)] [New Thread -1294992464 (LWP 17961)] [New Thread -1307624528 (LWP 17962)] Couldn't get registers: No such process. (gdb) fg Continuing. Cannot fetch general-purpose registers for thread -1219295552: generic error (gdb) bt Cannot fetch general-purpose registers for thread -1219295552: generic error (gdb) Gdb with 0.8.0: (gdb) run Starting program: /usr/bin/xfmedia [1]+ Stopped gdb xfmedia bash-3.00$ fg gdb xfmedia Reading symbols from shared object read from target memory...done. Loaded system supplied DSO at 0xffffe000 [Thread debugging using libthread_db enabled] [New Thread -1218640192 (LWP 17557)] [New Thread -1233421392 (LWP 17562)] [New Thread -1242145872 (LWP 17563)] [New Thread -1254786128 (LWP 17564)] [New Thread -1267278928 (LWP 17565)] [New Thread -1277559888 (LWP 17566)] [New Thread -1285952592 (LWP 17567)] [New Thread -1294345296 (LWP 17568)] [New Thread -1306977360 (LWP 17569)] [New Thread -1315648592 (LWP 17570)] [New Thread -1328141392 (LWP 17571)] [New Thread -1338422352 (LWP 17572)] Program received signal SIGTTOU, Stopped (tty output). [Switching to Thread -1218640192 (LWP 17557)] 0xffffe410 in __kernel_vsyscall () (gdb) bt #0 0xffffe410 in __kernel_vsyscall () #1 0xb773d38b in __write_nocancel () from /lib/tls/libpthread.so.0 #2 0xb797f556 in write_string (fd=2, string=0x8bc3860 "\n** (xfmedia:17557): WARNING **: No valid keybindings file found.\n") at gmessages.c:130 #3 0xb7980f2b in IA__g_log_default_handler (log_domain=0x0, log_level=G_LOG_LEVEL_WARNING, message=0x8bc3690 "No valid keybindings file found.", unused_data=0xfffffe00) at gmessages.c:929 #4 0xb7980308 in IA__g_logv (log_domain=Variable "log_domain" is not available. ) at gmessages.c:474 #5 0xb79805ab in IA__g_log (log_domain=0xfffffe00 <Address 0xfffffe00 out of bounds>, log_level=-512, format=0xfffffe00 <Address 0xfffffe00 out of bounds>) at gmessages.c:517 #6 0x0805c654 in xfmedia_keybindings_load () #7 0xb7fbf9eb in _dl_map_object (loader=0x73614220, name=0x75412d79 <Address 0x75412d79 out of bounds>, preloaded=980316009, type=2036689696, trace_mode=1684957538, mode=1936158313, nsid=1818846752) at dl-load.c:1832 Previous frame inner to this frame (corrupt stack?) (gdb)
Thanks, this bug is fixed in 0.9.1.
Err... no it isn't.
Oh.... All i can say is that it works for me with gtk+-2.8.3 here. I blindly assumed it was because a fix was added in 0.9.1. Strange that it works here then... I will try a next version also, but I guess I don't have a situation/testcase to really test the supposed fix then anymore.
Sorry for my content-free post earlier; I was rushing to get out the door to go to work. Your bug has nothing to do with gtk. From the backtrace, I think it's related to a problem with the XDG_CONFIG_DIRS environment variable (i.e., it isn't set up to include the xfmedia prefix on your system). Someone else had a similar problem where the keybindings file wasn't being found. In the preferences window, keybindings tab, do you have an empty list view? If so, then you're hitting this bug. I think I have a fix for it, I just need to try it out. Actually, your backtrace is somewhat weird. It looks like glib itself is crashing while trying to print out a warning message to console. This might indicate a glib bug. Regardless, I'm pretty sure the missing keybindings problem is what you're hitting. And the fix in 0.9.1 had absolutely nothing to do with that. Actually... hmm. Your backtrace isn't a segfault at all. So you probably *were* hitting the 2.8 crasher I fixed in 0.9.1, but gdb was doing something funky and stopped execution at a different point. Anyway, bottom line, I'm going to preempt this bug and rename it to something else so I don't forget to look into the keybindings problem.
Missing keybinding file problem should be fixed.