This is the mail archive of the
java@gcc.gnu.org
mailing list for the Java project.
Re: RFC: Java inliner
- From: Mark Mitchell <mark at codesourcery dot com>
- To: Andrew Haley <aph at cambridge dot redhat dot com>, Fergus Henderson <fjh at cs dot mu dot OZ dot AU>
- Cc: gcc at gcc dot gnu dot org, java at gcc dot gnu dot org
- Date: Tue, 11 Jun 2002 19:05:19 -0700
- Subject: Re: RFC: Java inliner
--On Tuesday, June 11, 2002 04:46:38 PM +0100 Andrew Haley
<aph@cambridge.redhat.com> wrote:
Fergus Henderson writes:
> On 11-Jun-2002, Andrew Haley <aph@cambridge.redhat.com> wrote:
> > I intend to work on a Java specific tree inliner.
> >
> > Comments?
>
> I doubt we'll get a real language-independent AST inliner soon
> if everyone who could benefit from it decides that it will be easier
> for them to build their own language-specific tree inliner instead ;-)
I take your point!
However, at the moment Java doesn't have an inliner of any kind, so...
The right thing to do is clear: convert the Java front end to use trees
that are more like the C/C++ trees. (C++ is C plus some C++-specific
extensions; I expect you want C plus some Java-specific extensions.)
Then, use the existing inliner.
Anything else leads to greater incompatibility between the front ends,
and, as such, constitutes a step in distinctly the wrong direction.
This really should not be hard; the Java source language looks very
much like the C/C++ source language. The control-flow constructs are
very similar.
I know that some of the Java developers think that the Java front end's
representation is superior. That's a fine opinion to have, and maybe
switching the C/C++ front end to the Java representation is OK too -- but
that's a lot more work. The Java representation certainly isn't
sufficiently more superior to justify that.
--
Mark Mitchell mark@codesourcery.com
CodeSourcery, LLC http://www.codesourcery.com