Re: Will tree-ssa be GCC 3.5?

[Ian Lance Taylor <ian at wasabisystems dot com>]

Scott Robert Ladd <coyote@coyotegulch.com> writes:

> So let's say someone provides funding to develop *just* the gfortran
> portion of tree-ssa, including testing, documentation, and all the
> niceties. Certainly, other tree-ssa-based frontends will benefit, but
> since they are not funded, they are unlikely to receive the same
> testing and care as gfortran.

What is the goal here?  The goal is probably not a tree-ssa
implementation of gfortran.  The goal is an implementation of gfortran
using some technology.  That goal could be achieved.

> Would the FSF officially release tree-ssa if only the Fortran 95
> frontend worked well?
> 
> And would the FSF support forking the tree-ssa/gfortran project?
> 
> Personally, I think forking GCC is a Bad Idea.

Why are these the right questions?  Who cares what the underlying
technology is, provided the resulting compiler works well?

If some company writes a contract to develop a solid gfortran
compiler, that company takes on the responsibility of making it happen
one way or another.  That company must judge whether to make sure that
tree-ssa happens, or whether to support gfortran using a different
approach.

> Yet I also don't see why anyone would commit funds to gfortran if it
> will never be released because the Ada (or C++, or Java) frontend
> doesn't work. Nor would I be sanguine about funding all of GCC when my
> primary interest is Fortran 95.

Naturally.  You have to write a contract that makes sense for you.

> And thus my original question: nder *current circumstances*, will
> tree-ssa become a released project, and if so, what is a good guess
> for its release?
> 
> So far, I have a good sense of when 3.4 will be released (May or so,
> 2004). I have yet to get an answer for tree-ssa. The lack of such an
> answer is, in itself, and answer: tree-ssa (and thus gfortran) is too
> theoretical to be considered, and other alternatives must be explored.
> 
> If that is the situation, we move on, and wish you well.

Again, why are those the right questions?

Ian