This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Analyzing Optimization: What's New in 3.4?
- From: Scott Robert Ladd <coyote at coyotegulch dot com>
- To: Andrew Pinski <pinskia at physics dot uc dot edu>
- Cc: gcc mailing list <gcc at gcc dot gnu dot org>
- Date: Mon, 11 Aug 2003 19:09:10 -0400
- Subject: Re: Analyzing Optimization: What's New in 3.4?
- References: <1F4D49C2-CC3B-11D7-816B-000393A6D2F2@physics.uc.edu>
Andrew Pinski wrote:
With the new loop unroller comes an option called -fpeel-loops which you
might want to try.
I'll add this to my list
Note: -funit-at-a-time is enabled by default at -O3 and above.
I'm compiling with -O1 and enabling specific flags to compare
interactions and generated code performance; I've already added
-funit-at-a-time to the mix.
Another thing you want want to try is intermodule optimizations (only
works for C code though).
For specific reasons, my current tests exist in singular source files.
That will change once my methods have stabilized.
No, but you can benchmark the tree-ssa branch which is replacing these
options.
I'll look at tree-ssa once I've got the 3.4 "optimization optimizer"
working.
-fnew-ra is a good option to benchmark (it might slow down the code
though).
-ftracer is also sometimes a good option to use.
I'll add these as well.
Thanks for the feedback.
--
Scott Robert Ladd
Coyote Gulch Productions (http://www.coyotegulch.com)
Software Invention for High-Performance Computing