From owner-cvs-all Wed Mar 27 6:36:28 2002 Delivered-To: cvs-all@freebsd.org Received: from mailout03.sul.t-online.com (mailout03.sul.t-online.com [194.25.134.81]) by hub.freebsd.org (Postfix) with ESMTP id 0FCAD37B416; Wed, 27 Mar 2002 06:36:17 -0800 (PST) Received: from fwd10.sul.t-online.de by mailout03.sul.t-online.com with smtp id 16qE12-0002ky-02; Wed, 27 Mar 2002 15:03:04 +0100 Received: from Magelan.Leidinger.net (520065502893-0001@[217.83.30.226]) by fmrl10.sul.t-online.com with esmtp id 16qE0m-2CdNgmC; Wed, 27 Mar 2002 15:02:48 +0100 Received: from Leidinger.net (netchild@localhost [127.0.0.1]) by Magelan.Leidinger.net (8.12.2/8.12.2) with ESMTP id g2RE2ch3017230; Wed, 27 Mar 2002 15:02:42 +0100 (CET) (envelope-from netchild@Leidinger.net) Message-Id: <200203271402.g2RE2ch3017230@Magelan.Leidinger.net> Date: Wed, 27 Mar 2002 15:02:38 +0100 (CET) From: Alexander Leidinger Subject: Re: cvs commit: ports/lang Makefile ports/lang/icc Makefile distinfo pkg-comment pkg-descr pkg-plist ports/lang/icc/files patch-include To: green@FreeBSD.ORG Cc: sw@anthologeek.net, cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG In-Reply-To: <200203271333.g2RDXsZ96421@green.bikeshed.org> MIME-Version: 1.0 Content-Type: TEXT/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8BIT X-Sender: 520065502893-0001@t-dialin.net Sender: owner-cvs-all@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On 27 Mär, Brian F. Feldman wrote: >> icc understands parts of gcc specific attributes and the gcc assembly >> syntax (at least parts of it, if I remember correctly). To use MS syntax >> you have to specify the "-use_msasm" option. So it may not be that >> impossible to support both... >> >> But the Linux kernel isn't icc compilable too, so it's not that much of >> a drawback... > > Hm, I didn't know it would actually use AT&T syntax instead of Intel! > That's definitely unexpected. I think Intel wants icc as an drop in replacement (sort of, it has -KPIC instead of -fPIC) to get better performance figures on Linux. BTW.: - Don't use inter-object-optimizations (no, this doesn't work on FreeBSD, it's done at link time) with icc, there's a bug (http://www.heise.de/newsticker/data/hes-11.11.01-000/, may the fish be with you). - If someone from Intel's icc department is listening: I want something like gcc's -nostartfiles and -nostdlib, this should enable us to even link with icc (a native icc would be even better). >> >> > What about servers like Apache (it would be great for busy servers) ? >> >> >> >> I've seen this speed improvement in FPU intensive code (no MMX or SIMD >> >> used, plain FPU code, testet on an AMD Duron). I don't expect such a >> >> large speed improvement in "ordinary server code". And trust me, it >> >> would be a lot of hassle to get apache compiled with icc. >> > >> > I got Ruby compiled with icc some weeks back and I seem to recall around a >> > 20% speed improvement, as well. An interpreter is a good benchmark for a >> > compiler's ability to generate both fast and correct code :) >> >> Awesome. Now... what about "cd .../ruby; make -DUSE_ICC"? ;-) > > I didn't test it with FreeBSD headers building FreeBSD executables, so I'll > have to try that and make sure binary compatibility exists, too :) It was a joke. Ruby uses autoconf, so it's a nightmare to do a port which builds it with icc. Bye, Alexander. -- The dark ages were caused by the Y1K problem. http://www.Leidinger.net Alexander @ Leidinger.net GPG fingerprint = C518 BC70 E67F 143F BE91 3365 79E2 9C60 B006 3FE7 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message