This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Higher level RTL issues
- To: Daniel Berlin <dan at cgsoftware dot com>
- Subject: Re: Higher level RTL issues
- From: law at redhat dot com
- Date: Fri, 19 Oct 2001 15:42:31 -0600
- cc: Diego Novillo <dnovillo at redhat dot com>, gcc at gcc dot gnu dot org
- Reply-To: law at redhat dot com
In message <1B5A94CB-BD37-11D5-867A-0030657B5340@cgsoftware.com>you write:
> > I'm thinking of trees that ressemble three-address ILs. They
> > would look just like trees to the tree-to-RTL pass and more
> > importantly they would provide a high-level IL that is more
> > language-independent than the current trees.
> This is a much better idea than a high level rtl, IMHO, having had to
> implement it for the vectorization stuff i did.
I don't disagree. I never intended for the higher level RTL concepts to
solve all the vectorization problems. I believe Diego, Richard and others
can probably back me up on that since I've discussed vectorization issues
with them in the past.
Attacking vectorization (IMHO) is mostly going to be in the realm of a
tree loop optimizer. As will be most of the loop restructuring
transformations.
jeff