Marc Espie wrote:
> *sigh*. I know what our C++ compiler currently does. Remember ? I'm kind
> of maintaining it. This is a somewhat rhethorical question: we do something,
> but is this the *right* thing from a standards point of view, or just what
> buggy behavior we currently happen to have.

My copy of the C++ spec (ISO/IEC 14882:1998(E)) states WRT exit()
behavior (18.3.8):

"First, objects with static storage duration are destroyed and functions
registered with atexit are called. Objects with static storage duration
are destroyed in the reverse order of the completion of their
constructor. (Automatic objects are not destroyed as a result of calling
exit().)"

which references a footnote that adds:

"Objects with automatic storage duration are all destroyed in a program
whose function main() contains no automatic objects and executes the
call to exit(). Control can be transferred to directly to such a main()
by throwing an exception that is caught in main()."

-Neil