This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: [tree-ssa] Use of build_binary_op
- From: Paul Brook <paul at nowt dot org>
- To: Geert Bosch <bosch at gnat dot com>,Pop Sébastian <pop at gauvain dot u-strasbg dot fr>
- Cc: Jason Merrill <jason at redhat dot com>,Daniel Berlin <dberlin at dberlin dot org>,gcc at gcc dot gnu dot org
- Date: Wed, 25 Sep 2002 20:41:55 +0100
- Subject: Re: [tree-ssa] Use of build_binary_op
- References: <65EC1BB8-D0A0-11D6-BD75-00039344BF4A@gnat.com>
> wrote:
> > I don't say that we have to switch from the lowering of C/C++ trees to
> > this translation point of view, but I just say that this interface
> > could be useful
> > - for other front-ends that cannot lower their initial AST into GIMPLE,
>
> This definitely seems like the right approach to use for the Ada front
> end.
> After the front end has finished semantic analysis and expansion, it
> traverses
> the resulting Ada tree in order to build the right tree/RTL data for the
> back end. (See ada/trans.c) This design allows for a clean separation of
> the back and front end.
This is similar to what g95 does, except g95 does scalarization at the same
time.
Paul