This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Unroller with branch and count patch
- From: Zdenek Dvorak <rakdver at atrey dot karlin dot mff dot cuni dot cz>
- To: Mircea Namolaru <NAMOLARU at il dot ibm dot com>
- Cc: gcc-patches at gcc dot gnu dot org,David Edelsohn <dje at makai dot watson dot ibm dot com>,Dale Johannesen <dalej at apple dot com>,Andrew Pinski <pinskia at physics dot uc dot edu>,Ulrich Weigand <weigand at i1 dot informatik dot uni-erlangen dot de>
- Date: Wed, 18 Feb 2004 15:03:11 +0100
- Subject: Re: Unroller with branch and count patch
- References: <OF7A0DC50E.3255B60F-ONC2256E3E.0044B8AD-42256E3E.0049176C@il.ibm.com>
Hello,
> This patch is intended for GCC mainline. It enables the GCC 3.4 new
> unroller to work with loops ending with an branch and count instruction
> (this is a frequent case on Power due to the doloop optimization).
> It corrects the regression seen for the GCC 3.4 new unroller on Power
> platforms for the option -funroll-loops (we intend to submit this patch
> also to GCC 3.4).
I don't think it is a good idea to include this in mainline (for one
reason, it does not apply any more -- simple loop analysis was rewritten
recently and moved to loop-iv.c); tomorrow I am going to send the
rewrite of the doloop optimization pass, thus making this completely
useless.
Considering 3.4, could you please send some performance numbers? I would
be especially interested in seeing differences between
-funroll-loops -fbranch-count-reg without the patch
-funroll-loops -fno-branch-count-reg without the patch
and
-funroll-loops -fbranch-count-reg with the patch.
on some benchmark.
Zdenek