On 2 Jun 2015, Stephen Leake wrote:
You only need sjlj if exceptions cross an Ada/C boundary. I just
need to nail down where that happens in my code.
I think that's the crux of the problem. Given that Lisp "conditions"
do not require unwinding the stack as Ada exceptions do, I don't see
how Ada exceptions could ever be allowed to cross the Ada/C, much
less Ada/Lisp, boundary.
Philippe[1] proposes to handle all Ada exceptions at the boundary,
i.e. in the top-level Ada subprogram(s) that is(are) called from C
(or Lisp). Are there so many such subprograms that the problem is
intractable?
[1]
http://host114.hostmonster.com/mailman/private/emacs-ada-mode_stephe-leake.org/2015-May/000430.html
--
Ludovic Brenta.
_______________________________________________
Emacs-ada-mode mailing list
[email protected]
http://host114.hostmonster.com/mailman/listinfo/emacs-ada-mode_stephe-leake.org
