This is the mail archive of the
mailing list for the GCC project.
Re: Various reworkings of PLACEHOLDER stuff
- From: Steven Bosscher <s dot bosscher at student dot tudelft dot nl>
- To: kenner at vlsi1 dot ultra dot nyu dot edu, gcc-patches at gcc dot gnu dot org
- Date: Wed, 28 May 2003 08:26:40 +0200
- Subject: Re: Various reworkings of PLACEHOLDER stuff
- Organization: DUT Faculty of Aerospace Engineering -- www.delftaerospace.com
At this point, this is in the category of "optimization". This change
does a few things:
(1) It adds a function type_contains_placeholder_p which returns TRUE
if a type has any attribute (size, position, bound, etc) that depends
on a PLACEHOLDER_EXPR.
(2) It uses that function to restrict optimizations of the few COMPONENT_REFs
that are unsafe and therefore allows Ada to pass them to fold.
(3) It adds a macro CONTAINS_PLACEHOLDER_P to do some checks before calling
the function of similar name (for efficiency) and changes all uses of the
function to use the macro.
This affects Ada only and was tested on i686-pc-linux-gnu.
Just curious: What are the future plans for the Ada
front end with tree-ssa in mind? Ada does not build
functions as trees, and it is the only front end that
uses PLACEHOLDER_EXPR tree nodes. And PLACEHOLDER_EXPR
is not part of GIMPLE...