Re: Verifying a compiler

"Joachim Durchholz" <joachim.dot.durchholz@halstenbach.com>
4 Jul 2000 18:40:08 -0400

          From comp.compilers

Related articles
[3 earlier articles]
Re: Verifying a compiler wlohmann@informatik.uni-rostock.de (2000-06-22)
Re: Verifying a compiler david@somers.lu (David Somers) (2000-06-22)
Re: Verifying a compiler echristo@cygnus.com (Eric Christopher) (2000-06-27)
Re: Verifying a compiler world!bobduff@uunet.uu.net (Robert A Duff) (2000-06-27)
Re: Verifying a compiler iainf@news.kororaa.com (Iain A F Fleming) (2000-06-30)
Re: Verifying a compiler wlohmann@informatik.uni-rostock.de (2000-06-30)
Re: Verifying a compiler joachim.dot.durchholz@halstenbach.com (Joachim Durchholz) (2000-07-04)
| List of all articles for this month |

From: "Joachim Durchholz" <joachim.dot.durchholz@halstenbach.com>
Newsgroups: comp.compilers
Date: 4 Jul 2000 18:40:08 -0400
Organization: Compilers Central
References: 00-06-056 00-06-081 00-06-088 00-06-099 00-06-116
Keywords: testing

Wolfgang Lohmann <wlohmann@informatik.uni-rostock.de> wrote:
> Robert A Duff (world!bobduff@uunet.uu.net) wrote:
>: We shouldn't think that formal proofs are somehow above and beyond the
>: more mundane things like "validation suites". Formal specs can have
>: bugs, too.
>
> Yes. And even verification proofs can contain errors ;-).


That's true... but: a verification tool is an extremely simple program
that's easy to verify by hand. Errors in such a tool should show up
very early. Of course, this just reduces the likelyhood of an error,
it doesn't totally prevent errors.


Regards,
Joachim





Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.