Created attachment 6579 core dump file generated from gdb First, set an app name on the "Filter on process name" textbox, firefox for example. Second, run firefox. Now the task manager crashed. OS: ArchLinux x64 xfce4-taskmanager version: 1.1.0 The stack: Program received signal SIGSEGV, Segmentation fault. 0x000000000040b538 in ?? () (gdb) bt #0 0x000000000040b538 in ?? () #1 0x00007ffff678c015 in g_closure_invoke () from /usr/lib/libgobject-2.0.so.0 #2 0x00007ffff679e061 in ?? () from /usr/lib/libgobject-2.0.so.0 #3 0x00007ffff67a6dfc in g_signal_emit_valist () from /usr/lib/libgobject-2.0.so.0 #4 0x00007ffff67a712f in g_signal_emit () from /usr/lib/libgobject-2.0.so.0 #5 0x00007ffff77689ee in ?? () from /usr/lib/libgtk-x11-2.0.so.0 #6 0x00007ffff678c015 in g_closure_invoke () from /usr/lib/libgobject-2.0.so.0 #7 0x00007ffff679e061 in ?? () from /usr/lib/libgobject-2.0.so.0 #8 0x00007ffff67a6dfc in g_signal_emit_valist () from /usr/lib/libgobject-2.0.so.0 #9 0x00007ffff67a712f in g_signal_emit () from /usr/lib/libgobject-2.0.so.0 #10 0x00007ffff7684097 in ?? () from /usr/lib/libgtk-x11-2.0.so.0 #11 0x00007ffff76851bf in gtk_list_store_set_valist () from /usr/lib/libgtk-x11-2.0.so.0 #12 0x00007ffff7685294 in gtk_list_store_set () from /usr/lib/libgtk-x11-2.0.so.0 #13 0x00000000004127f7 in ?? () #14 0x0000000000412aa6 in ?? () #15 0x000000000041329f in ?? () #16 0x0000000000408fb6 in ?? () #17 0x00007ffff64b86e3 in ?? () from /usr/lib/libglib-2.0.so.0 #18 0x00007ffff64b7c7a in g_main_context_dispatch () from /usr/lib/libglib-2.0.so.0 #19 0x00007ffff64b8020 in ?? () from /usr/lib/libglib-2.0.so.0 #20 0x00007ffff64b8342 in g_main_loop_run () from /usr/lib/libglib-2.0.so.0 #21 0x00007ffff7687787 in gtk_main () from /usr/lib/libgtk-x11-2.0.so.0 #22 0x0000000000408a6c in ?? () #23 0x00007ffff5ecd610 in __libc_start_main () from /usr/lib/libc.so.6 #24 0x0000000000408b0f in ?? ()
I've done some debugging and it seems like a bug in Gtk/Glib. The callback function xtm_process_tree_model_rows_reordered () in process-tree-model.c receives bad new_order array value even if argument of the g_signal_emit() is correct. This occurs if filtering is used, filtered model has no rows and matching row appears. The value of new_order[0] is 1 and because size of model indexed from 0 is 1, getting iter at position 1 returns no/bad value and segmentation fault occurs in a next function. The quick fix is to check for these specific conditions in the function and assign 0 to new_order[0].
It seems a duplicate of #12132 - followup there :) Jan, if you have a patch..... *** This bug has been marked as a duplicate of bug 12132 ***