X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [64.117.224.130]) by svr1.postgresql.org (Postfix) with ESMTP id F20C1D1B52A for ; Wed, 8 Oct 2003 19:23:15 +0000 (GMT) Received: from svr1.postgresql.org ([64.117.224.193]) by localhost (neptune.hub.org [64.117.224.130]) (amavisd-new, port 10024) with ESMTP id 30920-01 for ; Wed, 8 Oct 2003 16:22:27 -0300 (ADT) Received: from martin.sysdetect.com (martin.sysdetect.com [65.209.102.1]) by svr1.postgresql.org (Postfix) with ESMTP id 0026ED1B4E3 for ; Wed, 8 Oct 2003 16:22:25 -0300 (ADT) Received: (from mail@localhost) by martin.sysdetect.com (8.11.4/8.11.3) id h98JMBQ26683; Wed, 8 Oct 2003 19:22:11 GMT Received: from winwood.sysdetect.com(172.16.1.1) via SMTP by mail.sysdetect.com, id smtpdY29543; Wed Oct 8 19:22:09 2003 Received: from startide.sysdetect.com (startide.sysdetect.com [172.16.1.34]) by winwood.sysdetect.com (8.11.6/8.11.6) with ESMTP id h98JM9b13606; Wed, 8 Oct 2003 15:22:09 -0400 Received: from startide.sysdetect.com (localhost [127.0.0.1]) by startide.sysdetect.com (8.12.8/8.12.8) with ESMTP id h98JM9Uk029757; Wed, 8 Oct 2003 15:22:09 -0400 Received: from startide.sysdetect.com (seth@localhost) by startide.sysdetect.com (8.12.8/8.12.8/Submit) with ESMTP id h98JM9LW029751; Wed, 8 Oct 2003 15:22:09 -0400 Message-Id: <200310081922.h98JM9LW029751@startide.sysdetect.com> To: Jeff Cc: Neil Conway , PostgreSQL Performance From: pgsql-performance@sysd.com Subject: Re: Sun performance - Major discovery! In-reply-to: References: <1065635011.3424.25.camel@tokyo> Comments: In reply to a message from "Jeff " dated "Wed, 08 Oct 2003 14:11:23 -0400." Date: Wed, 08 Oct 2003 15:22:08 -0400 X-Virus-Scanned: by amavisd-new at postgresql.org X-Archive-Number: 200310/241 X-Sequence-Number: 3989 In message , Jeff writes: I'll go run the regression test suite with my gcc -O2 pg and the suncc pg. See if they pass the test. My default set of gcc optimization flags is: -O3 -funroll-loops -frerun-cse-after-loop -frerun-loop-opt -falign-functions -mcpu=i686 -march=i686 Obviously the last two flags product CPU specific code, so would have to differ...autoconf is always possible, but so is just lopping them off. I have found these flags to produce faster code that a simple -O2, but I understand the exact combination which is best for you is code-dependent. Of course, if you are getting really excited, you can use -fbranch-probabilities, but as you will see if you investigate that requires some profiling information, so is not very easy to actually practically use. -Seth Robertson