From owner-freebsd-questions@FreeBSD.ORG Thu Jun 21 18:41:42 2012 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 4ED98106564A for ; Thu, 21 Jun 2012 18:41:42 +0000 (UTC) (envelope-from wojtek@wojtek.tensor.gdynia.pl) Received: from wojtek.tensor.gdynia.pl (wojtek.tensor.gdynia.pl [89.206.35.99]) by mx1.freebsd.org (Postfix) with ESMTP id A221D8FC12 for ; Thu, 21 Jun 2012 18:41:41 +0000 (UTC) Received: from wojtek.tensor.gdynia.pl (localhost [127.0.0.1]) by wojtek.tensor.gdynia.pl (8.14.5/8.14.5) with ESMTP id q5LIfdNa016807; Thu, 21 Jun 2012 20:41:39 +0200 (CEST) (envelope-from wojtek@wojtek.tensor.gdynia.pl) Received: from localhost (wojtek@localhost) by wojtek.tensor.gdynia.pl (8.14.5/8.14.5/Submit) with ESMTP id q5LIfcGu016803; Thu, 21 Jun 2012 20:41:38 +0200 (CEST) (envelope-from wojtek@wojtek.tensor.gdynia.pl) Date: Thu, 21 Jun 2012 20:41:38 +0200 (CEST) From: Wojciech Puchar To: Joe Gain In-Reply-To: Message-ID: References: <402199FE-380B-41B6-866B-7D5D66C457D5@lpthe.jussieu.fr> <20120621015237.GB58187@neutralgood.org> <4FE35208.40708@queernet.org> <4FE35616.9080304@queernet.org> User-Agent: Alpine 2.00 (BSF 1167 2008-08-23) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed X-Greylist: Sender passed SPF test, not delayed by milter-greylist-4.2.7 (wojtek.tensor.gdynia.pl [127.0.0.1]); Thu, 21 Jun 2012 20:41:39 +0200 (CEST) Cc: Stas Verberkt , freebsd-questions@freebsd.org Subject: Re: Why Clang X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 21 Jun 2012 18:41:42 -0000 > > So, has anyone compared the performance of clang vs gcc compiled in daily use-- > for example as a server? Anyone can cherry pick a couple of binaries, but how > important is this for the performance of FreeBSD world? not big, as with almost any compiler. Most workload are dominated by cache misses and jump misprediction. That's why my gzip comparision resulted in minimally worse clang-compiled one (1% or less), while f2c converted fortran code for scientific calculations showed large differences. i expect large difference in eg. cjpeg, lame etc and rather small in for eg. perl