This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [RFC] Matrix Flattening optimization
- From: Daniel Jacobowitz <drow at false dot org>
- To: Razya Ladelsky <RAZYA at il dot ibm dot com>
- Cc: gcc-patches at gcc dot gnu dot org, Daniel Berlin <dberlin at dberlin dot org>, jh at suse dot cz
- Date: Thu, 2 Mar 2006 12:30:41 -0500
- Subject: Re: [RFC] Matrix Flattening optimization
- References: <OF381A18C3.9044C2E0-ONC2257125.002CF4EB-C2257125.003A6049@il.ibm.com>
I've got to ask...
On Thu, Mar 02, 2006 at 12:42:23PM +0200, Razya Ladelsky wrote:
> For art benchmark matrix flattening shows improvement of 35%, equake
> shows 9%
> on PowerPC Linux.
Do you have any idea how often this transformation is useful _outside_
SPEC? At first glance it seems like a pretty silly transformation;
I'd want to see some compelling reason to write the code the slower
way.
--
Daniel Jacobowitz
CodeSourcery