This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Ada policy
- From: Robert Dewar <dewar at gnat dot com>
- To: Richard Kenner <kenner at vlsi1 dot ultra dot nyu dot edu>
- Cc: jsm at polyomino dot org dot uk, gcc at gcc dot gnu dot org
- Date: Mon, 30 Aug 2004 18:25:41 -0400
- Subject: Re: Ada policy
- References: <10408302121.AA00131@vlsi1.ultra.nyu.edu>
If somebody volunteers to do the former, that problem will go away.
But the second problem is more serious, especially if you enlarge
the set of changes that need test cases to front-end changes.
Indeed, though there are a fair number of tests that could be
provided (e.g. the tests we write for ourselves for new features).
But the majority of test code is indeed proprietary, and certainly
we don't have the bandwidth at AdaCore to do what can sometimes
be an extremely difficult job of providing independent distributable
tests.
So there are really two choices.
1. We submit fixes for bugs for which we have only proprietary
test cases, meaning that overall the quality of the Ada front
end will be higher, since for one thing, it will be far closer
to our in house tree for GNAT Pro (not quite indentical, since
there are things in the GNAT Pro tree that cannot go into the
FSF version for various reasons, related to FSF/GCC requirements,
not our requirements).
2. We simply avoid fixing problems in the FSF tree for which
we don't have distributable test cases. That's difficult but
possible.
I am dubious that 2. is advantageous for anyone ...