From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 4184 invoked by alias); 13 Apr 2010 01:47:34 -0000 Received: (qmail 4176 invoked by uid 22791); 13 Apr 2010 01:47:34 -0000 X-SWARE-Spam-Status: No, hits=-0.4 required=5.0 tests=BAYES_00,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,FREEMAIL_ENVFROM_END_DIGIT,FREEMAIL_FROM,RCVD_IN_DNSWL_NONE,T_TO_NO_BRKTS_FREEMAIL X-Spam-Check-By: sourceware.org Received: from web65312.mail.ac2.yahoo.com (HELO web65312.mail.ac2.yahoo.com) (68.180.158.238) by sourceware.org (qpsmtpd/0.43rc1) with SMTP; Tue, 13 Apr 2010 01:47:29 +0000 Received: (qmail 96806 invoked by uid 60001); 13 Apr 2010 01:47:27 -0000 Message-ID: <708721.39752.qm@web65312.mail.ac2.yahoo.com> Received: from [203.127.19.157] by web65312.mail.ac2.yahoo.com via HTTP; Mon, 12 Apr 2010 18:47:27 PDT Date: Tue, 13 Apr 2010 01:47:00 -0000 From: Rayne Subject: Re: -mtune option for Intel Nehalem To: gcc-help@gcc.gnu.org MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-IsSubscribed: yes Mailing-List: contact gcc-help-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-help-owner@gcc.gnu.org X-SW-Source: 2010-04/txt/msg00093.txt.bz2 I mostly just want to know if an option is avaliable for the Nehalem processors. If it happens that the newest version of GCC has such an option, then I can always upgrade to that version. I believe that Barcelona is an AMD processor, so I'm confused as to how that would work for an Intel Nehalem processor. As for -mtune=nocona, I've tried and there was no performance gain. Thank you. Regards, Rayne > Do you mean you want to use an old version of gcc, issued prior to the > advent of -mtune=barcelona? That would reduce your performance potential > for vectorized code. It should be obvious, as Nehalem came later than > Barcelona, that there would be no optimization for Nehalem in older > versions. > You would have to experiment yourself to see whether -mtune=nocona helps > performance in your case. It's certainly not feasible to give you > recommendations covering all old gcc versions, even if we could guess the > other factors of your case. > > >-- >Tim Prince