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?
Ian Lance Taylor wrote:
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.
Yes. Although it seems a shame to waste the existing effort in tree-ssa.
Why are these the right questions? Who cares what the underlying
technology is, provided the resulting compiler works well?
Again, I completely agree. One of the questions I was asked is: Could
the existing gfortran be modified to support non-GCC backends? The
question is fraught with technical and legal issues...
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.
One requirement is that the compiler support numerous platforms;
currently, GCC is the only existing solution. Again, it would be a shame
to reject the existing work in GCC to reinvent the wheel.
Naturally. You have to write a contract that makes sense for you.
But can such a contract be written within the restrictions of the GPL?
For example, could the GPL's gfortran frontend and libraries be used
with a non-free backend?
--
Scott Robert Ladd
Coyote Gulch Productions (http://www.coyotegulch.com)
Software Invention for High-Performance Computing