This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Will tree-ssa be GCC 3.5?
Jim Wilson wrote:
I think it makes more sense to release it even if it is flawed. We will
find bugs sooner. Also, that will provide motivation for people to fix
the front-ends that are missing. If gcc4 is held up until all front
ends are fixed, then people will have little incentive to fix their
front ends, and the release may be delayed indefinitely. Certainly, we
should make every attempt to fix as many problems before the release as
we can. But we should not delay indefinitely because some problems
haven't been fixed yet.
I prefer sooner as opposed to later.
Many, many potential testers are intimidated by downloading and
compiling a compiler from CVS; it's unfamiliar territory, especially for
many "compiler users" who are bright people but who have never worked
with free software.
Perhaps it is time to begin posting weekly snapshots of tree-ssa, to
make things simpler for potential experimenters?
I've been focusing on testing (Acovea) -- I think my best contribution
is in *using* the forthcoming compilers, performing QA tests, and making
sure they can "do the job." We won't know if GCC 4 is ready for release
until we have a solid community of users (i.e., testers.)
That is also why I mentioned maintaining gcc3 and gcc4 in parallel. It
isn't a tragedy if a front end is missing from gcc4 if we are still
supporting it in gcc3.
Precisely. Running in parallel is always a good idea, whether you're
installing a new computer or updating software tools.
--
Scott Robert Ladd
Coyote Gulch Productions (http://www.coyotegulch.com)
Software Invention for High-Performance Computing
In development: Alex, a database for common folk