This is the mail archive of the
mailing list for the GCC project.
Re: [wwwdocs] Initial criteria.html (was Re: Where's the current list of release criteria?)
- To: kenner at vlsi1 dot ultra dot nyu dot edu
- Subject: Re: [wwwdocs] Initial criteria.html (was Re: Where's the current list of release criteria?)
- From: <guerby at acm dot org>
- Date: Tue, 30 Oct 2001 21:59:07 +0100
- CC: gcc at gcc dot gnu dot org
- References: <10110302058.AA08704@vlsi1.ultra.nyu.edu>
- Reply-to: guerby at acm dot org
> Your left out the issue of gnatchop and the special handling for
> the "library" in the CA tests.
- If possible I'd rather have ACATS sources commited in gnatchoped
form, this will save time and space when running the tests. The ACATS
test suite changes very slowly and changes are announced monthes in
advance, so there is no real maintenance problem.
- The library compilation model tests are of absolutely no value for
the GCC project, my script is able to run them successfully but not in
a way acceptable for validation. I feel like just dumping them, may be
keeping those that have some code to execute in final gnatchop form.
My target isn't to be able to present the GCC ACATS test run to a
validation authority (like ACT does) but to provide a test run that
has some value to the GCC project that is to say provide an easy way
for GCC developpers to verify that a change to GCC, backend or ada/*,
doesn't break Ada in a significant way.
Laurent Guerby <email@example.com>