swarm-support
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: non-existent lines


From: Marcus G. Daniels
Subject: Re: non-existent lines
Date: 21 Nov 2000 10:49:37 -0800
User-agent: Gnus/5.070084 (Pterodactyl Gnus v0.84) Emacs/20.4

>>>>> "DC" == Daniel Calhoun <address@hidden> writes:

DC> With reference to my own earlier message, just what is

DC>  _OBJC_METH_VAR_TYPE_3 () ?

It probably doesn't mean anything in this context.  The debugger is
associating in-memory addresses with symbols that were determined at
compile-time (or by the dynamic linker).  Sometimes a program will
crash and burn in such a way that the `stack' (where local variables
and execution is kept), is damaged, and you end up at some place where
the numbers found on the stack are garbage, and thus the symbols
associated with those numbers are not meaningful.  One way this
can happen is if you send a message to a pointer that isn't valid.
Another way is to step over the bounds of an array into a caller
method's stack space.

In any case, what is needed when this happens is to find a way to let
the execution of the program move forward in controllable chunks, so
that you can see what happens just before.  One way to do this is
to set breakpoints at various places in your model where you
expect things to work, until something doesn't work.  Another way
is to put instrumentation in the code of the model itself so 
that when it runs it will tell you where it is.  

                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]