comp.lang.ada
 help / color / mirror / Atom feed
From: Paul Rubin <no.email@nospam.invalid>
Subject: Re: Ada and software testing
Date: Wed, 14 Jul 2021 12:32:40 -0700	[thread overview]
Message-ID: <87pmvky9lz.fsf@nightsong.com> (raw)
In-Reply-To: 5bc2434a-6784-47d6-ba66-50edff4f80f0n@googlegroups.com

Gautier write-only address <gautier_niouzes@hotmail.com> writes:
> You can combine the power of fuzzing with the power of Ada's strong
> typing, implying standard Ada run-time checks (e.g. range checks),
> Read the following article for details:
> https://blog.adacore.com/running-american-fuzzy-lop-on-your-ada-code

Thanks, this is pretty interesting.  He runs AFL on three Ada programs:
Zip-Ada, and Ada libraries for reading YAML and JSON.  It finds bugs in
all three, though not very many.  It fits my picture that Ada programs
are less susceptible than C programs are, to the types of bugs that
fuzzing uncovers.

I do have to say that errors thrown by runtime checks on range types are
still program bugs, in the sense that they are type errors, that in
principle we should want to catch at compile time.

  parent reply	other threads:[~2021-07-14 19:32 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-12  0:49 Ada and software testing Paul Rubin
2021-07-12  8:40 ` Dmitry A. Kazakov
2021-07-14 19:56   ` Paul Rubin
2021-07-12 16:14 ` Gautier write-only address
2021-07-12 16:41   ` Dmitry A. Kazakov
2021-07-14 19:32   ` Paul Rubin [this message]
2021-07-14 19:51     ` Dmitry A. Kazakov
2021-07-14 20:02       ` Paul Rubin
2021-07-15  7:27         ` Dmitry A. Kazakov
2021-07-16 10:01 ` G.B.
2021-07-16 10:21   ` Paul Rubin
2021-07-28 15:28     ` Paul Butcher
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox