I have built xfce4-taskmanager from source and ran the self-build xfce4-taskmanager with gdb (and the debug symbols). When I do a backtrace in gdb after xfce4-taskmanager has frozen I got more output from gdb than before:
#0 0xb682362c in read () from /lib/arm-linux-gnueabihf/libc.so.6
#1 0xb67eb66c in _IO_file_underflow () from
/lib/arm-linux-gnueabihf/libc.so.6
#2 0xb67ec1f8 in _IO_default_uflow () from
/lib/arm-linux-gnueabihf/libc.so.6
#3 0xb67ec118 in __uflow () from /lib/arm-linux-gnueabihf/libc.so.6
#4 0xb67e6148 in getc () from /lib/arm-linux-gnueabihf/libc.so.6
#5 0x00013454 in get_task_cmdline (task=0xbeffe108)
at task-manager-linux.c:138
#6 get_task_details (task=0xbeffe108, pid=<optimized out>)
at task-manager-linux.c:291
#7 get_task_list (task_list=0x1484e0) at task-manager-linux.c:312
#8 0x00011c84 in xtm_task_manager_update_model (manager=0x77da0)
at task-manager.c:473
#9 0x0000dabc in init_timeout () at main.c:117
#10 0xb68d3ad8 in ?? () from /lib/arm-linux-gnueabihf/libglib-2.0.so.0
#11 0xb68d3ad8 in ?? () from /lib/arm-linux-gnueabihf/libglib-2.0.so.0
Backtrace stopped: previous frame identical to this frame (corrupt stack?)
I have analysed the problem a bit more:
xfce4-taskmanager iterates continuously over all available PIDs, as it
is supposed to. I found out by adding the following line before the
for-loop at line 138 in task-manager-linux.c:
fprintf(stderr, "task->pid=%d\n", task->pid);
The main window of xfce4-taskmanager was refreshed once after some
amount of time (~10 seconds). The GUI was like frozen for the rest of
the time while xfce4-taskmanager continues to iterate over all PIDs.
I think that there is a problem with the GUI or some components from it
since the core functionality of xfce4-taskmanager seems to work.
On 07. mai 2014 20:32, Yves-Alexis Perez wrote:
>
> It might help to check what is the 7 fd.
>
signature.asc
Description: OpenPGP digital signature
