Hide Forgot
Description of problem: Segmentation fault when starting gthumb Entry in /var/log/messages: Nov 13 19:16:51 gondor kernel: [27191.362162] gthumb[29465]: segfault at 0 ip (null) sp 00007fffb89b0028 error 14 in gthumb[400000+d8000] Version-Release number of selected component (if applicable): gthumb-2.14.0-1.fc16.x86_64 How reproducible: everytime Steps to Reproduce: 1. start gtumb from terminal Actual results: Segmentation fault Expected results: gthumb should start
I just did a quick check on F16, x86-64 and don't see the issue. Please provide a stacktrace: http://fedoraproject.org/wiki/StackTraces . If you have issues getting a good stacktrace, please let me know!
Created attachment 533397 [details] Stacktrace generated with valgrid Stacktrace added as an attachement. I'm haveing the segfault on two Fedora 16 systems. On both systems I'm useing openbox as desktop-environment, both where upgraded from f15 to f16.
Stacktrace generated with gdb (gdb) run Starting program: /usr/bin/gthumb [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib64/libthread_db.so.1". Program received signal SIGSEGV, Segmentation fault. 0x0000000000000000 in ?? () (gdb) thread apply all bt Thread 1 (Thread 0x7ffff7fcc9c0 (LWP 31454)): #0 0x0000000000000000 in ?? () #1 0x00000031ebe14b63 in ?? () from /usr/lib64/libcogl.so.5 #2 0x00000031ebe14c82 in ?? () from /usr/lib64/libcogl.so.5 #3 0x00000031ebe51ea7 in ?? () from /usr/lib64/libcogl.so.5 #4 0x00000031ebe1d3d6 in cogl_context_new_EXP () from /usr/lib64/libcogl.so.5 #5 0x00000031eae30eb2 in ?? () from /usr/lib64/libclutter-glx-1.0.so.0 #6 0x00000031eae6c355 in ?? () from /usr/lib64/libclutter-glx-1.0.so.0 #7 0x00000031eae7853a in ?? () from /usr/lib64/libclutter-glx-1.0.so.0 #8 0x00000031eae79888 in clutter_init () from /usr/lib64/libclutter-glx-1.0.so.0 #9 0x00000000004339bd in main ()
Most likely the backtrace of gdb will be more helpful since it will usually display the parameters and return values. Please can you make sure to install the debuginfo packages: debuginfo-install gthumb should automatically install all necessary packages.
Hope thisone is better Program received signal SIGSEGV, Segmentation fault. 0x0000000000000000 in ?? () (gdb) thread apply all bt Thread 1 (Thread 0x7ffff7fcc9c0 (LWP 32118)): #0 0x0000000000000000 in ?? () #1 0x00000031ebe14b63 in _cogl_get_gl_version (major_out=0x7fffffffe238, minor_out=0x7fffffffe23c) at driver/gl/cogl-gl.c:47 #2 0x00000031ebe14c82 in check_gl_version (error=0x7fffffffe340, ctx=0x722000) at driver/gl/cogl-gl.c:82 #3 _cogl_gl_update_features (context=0x722000, error=0x7fffffffe340) at driver/gl/cogl-gl.c:144 #4 0x00000031ebe51ea7 in _cogl_winsys_context_init (context=0x722000, error=<optimized out>) at winsys/cogl-winsys-stub.c:102 #5 0x00000031ebe1d3d6 in cogl_context_new_EXP (display=0x718890, error=0x7fffffffe340) at ./cogl-context.c:181 #6 0x00000031eae30eb2 in clutter_backend_cogl_create_context (error=0x7fffffffe340, backend=0x6f5060 [ClutterBackendCogl]) at cogl/clutter-backend-cogl.c:346 #7 clutter_backend_cogl_create_context (backend=0x6f5060 [ClutterBackendCogl], error=0x7fffffffe340) at cogl/clutter-backend-cogl.c:291 #8 0x00000031eae6c355 in _clutter_feature_init (error=0x7fffffffe340) at ./clutter-feature.c:107 #9 0x00000031eae7853a in clutter_init_real (error=0x7fffffffe340) at ./clutter-main.c:1284 #10 0x00000031eae79888 in clutter_init (argc=0x0, argv=0x0) at ./clutter-main.c:1786 #11 0x00000000004339bd in main (argc=1, argv=0x7fffffffe4b8) at main.c:561
Since todays kernelupdate gthumb start without a segfault.
This seems to be a cogl / OpenGL bug. Which matches with comment #6 where a kernel update fixes it, which is possible since the drm part of OpenGL is handled in the kernel, closing per comment #6.