This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Tree inlining for the C front end (part 3 of 3)
- To: Daniel Berlin <dan at cgsoftware dot com>
- Subject: Re: Tree inlining for the C front end (part 3 of 3)
- From: Richard Henderson <rth at redhat dot com>
- Date: Wed, 26 Sep 2001 13:28:19 -0700
- Cc: Alexandre Oliva <aoliva at redhat dot com>, Gerald Pfeifer <pfeifer at dbai dot tuwien dot ac dot at>, gcc-patches at gcc dot gnu dot org
- References: <orpu8f7i5q.fsf@localhost.localdomain> <6C59D124-B1AC-11D5-9D8F-0030657B5340@cgsoftware.com>
On Tue, Sep 25, 2001 at 07:56:58AM -0400, Daniel Berlin wrote:
> Not doing performance testing on patches specifically meant to improve
> performance (Otherwise, why the heck are we doing tree inlining at all?)
So that we can get rid of the god-awful rtl inliner and the
nasty problems that are caused by generating rtl under one
set of circumstances and trying to recognize the rtl under
another set of circumstances.
One example: ENCODE_SECTION_INFO modifies SYMBOL_RTX in place,
which may change how predicates such as local_symbolic_operand
or global_symbolic_operand react.
r~