[Torcs-devel] [patch] run torcs-bin under valgrind

[Jim C. <jc...@di...>]

hi,

saw the todo item about memory leaks, thought Id try to help.

[jimc@harpo torcs]$ diff -ru torcs.sh~ torcs.sh
--- torcs.sh~   2005-03-09 10:02:38.000000000 -0700
+++ torcs.sh    2005-03-09 10:05:32.480523959 -0700
@@ -77,6 +77,12 @@
            echo "   -e  display the commands to issue when you want to 
run under gdb"
            ;;

+
+       # Valgrind run
+       *) valgrind -v --tool=memcheck --leak-check=yes 
--show-reachable=yes \
+           $LIBDIR/torcs-bin -l $LOCAL_CONF -L $LIBDIR -D $DATADIR $*
+           ;;
+
        # Normal run
        *) $LIBDIR/torcs-bin -l $LOCAL_CONF -L $LIBDIR -D $DATADIR $*
            ;;
[jimc@harpo torcs]$


I ran torcs 1.2.3 (binary from sourceforge) thru it, captured to a 
script-file, which is attached

I had to manually kill it, it appeared to be stuck:

WARNING: slScheduler: Too many pending callback events!
WARNING: slScheduler: Too many pending callback events!
WARNING: slScheduler: Too many pending callback events!
WARNING: slScheduler: Too many pending callback events!
WARNING: slScheduler: Too many pending callback events!
WARNING: slScheduler: Too many pending callback events!
...

the screen never rendered after 'READY'.

I hope this helps a bit, and that you have big CPUs to run it further.

thanks,
jimc