Re: Ada and software testing

[Paul Rubin <no.email@nospam.invalid>]

"Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de> writes:
> But if it does not and I/O completes without losing the processor, the
> callback is called recursively *before* popping the queue and the
> *same* portion of data is sent again.

This is a garden variety concurrency bug that you're right, wouldn't
normally be found with conventional fuzzing, but might be findable with
stress testing.  A more rigorous approach would involve model checking.

This type of problem happens in C programs all the time as well, and
doesn't really signify anything about the effectiveness of fuzz testing.
Fuzzing is very effective against C programs, but tentatively maybe less
so against Ada programs, because of Ada's more thorough type checking.

> [The fix is to make recursive calls void]

Hopefully there would be some locks between the tasks, though in that
case the problem would show up as deadlock.