This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Large performance regressions from 2.95.2/3 to HEAD
- To: Daniel Berlin <dan at cgsoftware dot com>
- Subject: Re: Large performance regressions from 2.95.2/3 to HEAD
- From: Gerald Pfeifer <pfeifer at dbai dot tuwien dot ac dot at>
- Date: Thu, 29 Jun 2000 23:38:43 +0200 (CEST)
- cc: gcc at gcc dot gnu dot org
On Thu, 29 Jun 2000, Daniel Berlin wrote:
> In every benchmark i can find, i see large performance regressions in
> the performance of the generated.
That absolutely mirrors my experience for template heavy C++ code.
As far as I can see, this significant regression happened about the
time when the C++ front-end switched to tree-based inlining (which
should actually improve performance, but note that I'm referring to
the *time*, not the changes themselves).
> I see the same type of decrease in other things using templates vs
> non-templates. While the non-templated stuff is usually quite a bit
> faster, the stuff using templates seems to be at least half as fast,
> if not worse.
Yup. I could not provide a small test case, but based on your results
with the Stepanov benchmarks these certainly are worth having a look!
Gerald
--
Gerald "Jerry" pfeifer@dbai.tuwien.ac.at http://www.dbai.tuwien.ac.at/~pfeifer/
Have a look at http://petition.eurolinux.org -- it's not about Linux, btw!