>>>>> "Paul" == Paul D Smith <psmith_at_BayNetworks.com> writes:
Paul> SunOS 4.1.3_U1, X11R6 (latest patches), built with GCC, XPM,
Paul> SHAPE, etc. I can get fvwm2 to dump core every single time I
Paul> run tgif. The version of tgif I'm running is 2.16p8, I
Paul> believe--I don't really want to start it up again :-/ :).
I just tried tgif 2.16 pl 12 and didn't have any problems. Different
OS though (AIX).
Paul> Anyway, the traceback in fvwm2 is this:
Paul> Core was generated by `fvwm2'.
Paul> Program terminated with signal 10, Bus error.
Paul> #0 DispatchEvent () at events.c:225
Paul> 225 (*EventHandlerJumpTable[Event.type])();
....
Paul> The problem is that Event.type is 64, but EventHandlerJumpTable has only
Paul> 35 members:
....
Paul> Obviously indirecting through 0x1 is not good!
Paul> I don't know what the problem is here: should Event.type never be >34?
Paul> Should EventHandlerJumpTable be larger than it is?
I set the size of that to be LASTEvent, which is defined in X.h to be
of a greater value than any of the events. Apparently on your
machine, this is not the case. I could set it to something like 256
always, but I wanted to save space. Besides, LASTEvent is SUPPOSED to
be the max event number...
Paul> At any rate, some sanity checking of this value would probably be A Good
Paul> Thing.
Possibly, but I took out the checking that was there (the switch
statement) to get a speedup. As I said, I could extend the size of
the jump table to like 256 and still get that speedup, but waste that
extra space (even if it is only about 888 bytes).
Perhaps I'll have do that...
Chuck
--
Visit the official FVWM web page at <URL:http://www.hpc.uh.edu/fvwm/>.
To unsubscribe from the list, send "unsubscribe fvwm" in the body of a
message to majordomo_at_hpc.uh.edu.
To report problems, send mail to fvwm-owner_at_hpc.uh.edu.
Received on Tue Mar 12 1996 - 17:40:35 GMT