Date: Sat, 27 Mar 1999 15:05:05 +0100 (MET) From: Michael Hanke To: pgcc AT delorie DOT com Subject: Re: [Fwd: Re: Aligning stack variables [8-byte operands]] In-Reply-To: <36FC6B66.D36CD9CE@e-technik.tu-ilmenau.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Reply-To: pgcc AT delorie DOT com X-Mailing-List: pgcc AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk On Sat, 27 Mar 1999, Wolfgang Reimer wrote: > John Wehle wrote: > > [snip] > > > > Keep in mind that integer code doesn't benefit from more alignment and > > that it's desirable to avoid impacting integer performance. As with > > many things it is a question of tradeoffs. > > Why?????? > [ lines deleted ] > > > > -- John [snip] > fibers. So for me (and other number crunching guys) the alignment of > 8-byte operands (doubles) on stack would improve Intel P6 performance > much more than any other sophisticated Pentium optimization strategy. > That's why it should be the "Number ONE" on the TODO list of the egcs > compiler development group, IMHO. > I agree. I think that it is not worth the effort if the gain in speed is marginal (say less than 10%). That is what I obtain. I belong also to the number crunching people. Regards, Michael +---------------------------------------------------------------+ | Michael Hanke Royal Institute of Technology | | NADA | | S-10044 Stockholm | | Sweden | +---------------------------------------------------------------+ | Visiting address: Lindstedtsvaegen 3 | | Phone: + (46) (8) 790 6278 | | Fax: + (46) (8) 790 0930 | | Email: hanke AT nada DOT kth DOT se | | na DOT mhanke AT na-net DOT ornl DOT gov | +---------------------------------------------------------------+