This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [tree-ssa] COND_EXPR lowering.
- From: Andrew MacLeod <amacleod at redhat dot com>
- To: Chris Lattner <sabre at nondot dot org>
- Cc: Zdenek Dvorak <rakdver at atrey dot karlin dot mff dot cuni dot cz>, gcc-patches <gcc-patches at gcc dot gnu dot org>, Jeff Law <law at redhat dot com>, Diego Novillo <dnovillo at redhat dot com>
- Date: 24 Oct 2003 17:05:23 -0400
- Subject: Re: [tree-ssa] COND_EXPR lowering.
- References: <Pine.LNX.4.44.0310241536260.24741-100000@nondot.org>
On Fri, 2003-10-24 at 16:50, Chris Lattner wrote:
>
> Andrew MacLeod wrote:
> > It was not always clear that we were going to lower constructs right at
> > the very beginning of processing.
> > ...
> > You had the advantage of coming in at a point at which this was all
> > becoming obvious. We, and the code, worked our way up to that point.
>
> To be fair to Zdenek, that's not true at all. You had early warning about
> this kind of thing.
>
Slightly out of context. To my knowledge, Zdenek started looking at
tree-ssa in July/August. At that point, I think most people that had
been working on the branch, regardless of their previous stances, had
come to the conclusion we were going to have to lower the containers in
order to perform some optimizations. I even had some of it in my todo
list.
I will not speak to the correctness or incorrectness of the original
architecture as its been a constantly evolving entity. My impression is
that the container aspect of the IL is superior for the front end and
parser, and it will probably stay that way there.
The original work on tree-ssa was more proof of concept I think, so
changing the IL didnt serve a purpose then. And then once the ball got
rolling, some tremendous changes were performed on the SSA
implementation. The container IL wasn't a huge issue until we got far
enough along that we could start to do interesting things. It is
overdue, but we've been busy.
> In particular: (skip to CFG is a the sub-language)
> http://gcc.gnu.org/ml/gcc/2002-08/msg01555.html
>
> You may find the entire thread interesting in retrospect. :)
It was well before my time on this work, so it isnt a retrospec for me
:-)
Andrew