I am closing this bug report since it is from an outdated release and is
probably fixed by now. Please reopen if this still appears in 10.04 or
newer.
** Changed in: xscreensaver (Ubuntu)
Status: Incomplete => Invalid
** Changed in: xscreensaver (Ubuntu)
Assignee: Tormod Volden (tormo
In your prior backtrace attempt you did not specify the program name
when you started gdb, so I thought that was the problem.
Once you get the SIGSTOP, run "backtrace full", then "cont". I am bit
unsure about this SIGSTOP thing, but you can also try to enter "handle
SIGSTOP nopass" before you atta
My prior backtrace attempt (2009-02-08) was over the network via ssh. I
tried it again today, but this time I attached gdb to the polytopes
process BEFORE polytopes locked up. However, the result was nearly
identical to the log I uploaded on 2009-02-08. In short, once polytopes
locked up, pressi
You should be able to do this over the network. ssh into the computer where
polytopes is running in fullscreen, and run
gdb /usr/lib/xscreensaver/polytopes 2>&1 | tee gdb-polytopes.txt
then attach to the running polytopes process with
attach PID
After attaching you can type "cont" to let it run.
I tried:
gdb /usr/lib/xscreensaver/polytopes 2>&1 | tee gdb-polytopes.txt
and typed 'run' in gdb. It successfully brought up the Polytopes
screensaver in a window. Pressing escape closed the window. However, I
repeated this process about 7000 times (via a script!) and it never
locked up. So
Thanks for your debugging efforts. (You must have loved the 15 minutes of
hitting escape :) )
You should specify the name of the program when starting gdb, like this:
gdb /usr/lib/xscreensaver/polytopes 2>&1 | tee gdb-polytopes.txt
--
[intel] polytopes screensaver freezes, requiring hard reset
I attempted to get a backtrace, but I'm not sure it worked. After
attaching gdb to the frozen polytopes process, gdb froze (ctrl-c did not
work) until I killed the polytopes process from another terminal window.
At that point gdb gave some error messages. It did then allow me to
dump a backtrace,
Hmm, there are quite some errors and warnings in your log, for instance:
(WW) intel(0): Existing errors found in hardware state.
Can you please try to get a backtrace from polytopes when it's going
100%? See http://wiki.ubuntu.com/DebuggingProgramCrash
** Summary changed:
- polytopes screensaver
