This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Unrolling addressing optimization (re-submission)
- From: Zdenek Dvorak <rakdver at atrey dot karlin dot mff dot cuni dot cz>
- To: Revital Eres <ERES at il dot ibm dot com>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Thu, 15 Apr 2004 16:19:10 +0200
- Subject: Re: [PATCH] Unrolling addressing optimization (re-submission)
- References: <OF3BC2105E.BEEB7C2A-ONC2256E77.0043D3D2-C2256E77.004CA216@il.ibm.com>
Hello,
> Indeed the optimization is not been applied in many frequent cases which
> involves simple array access . (with or without your latest patch)
> As we tested on various benchmarks our patch finds many opportunities
> to apply this optimization.
also, could you please get some benchmark results from other
architectures than ppc (especially from x86; you approach seems to produce
worse code here than what we get with -fweb, and I would like to have
this either confirmed or rejected).
Zdenek