From owner-freebsd-current@FreeBSD.ORG Mon Feb 11 22:22:31 2013 Return-Path: Delivered-To: freebsd-current@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by hub.freebsd.org (Postfix) with ESMTP id 68830C68; Mon, 11 Feb 2013 22:22:31 +0000 (UTC) (envelope-from sgk@troutmask.apl.washington.edu) Received: from troutmask.apl.washington.edu (troutmask.apl.washington.edu [128.95.76.21]) by mx1.freebsd.org (Postfix) with ESMTP id 27A931F3; Mon, 11 Feb 2013 22:22:31 +0000 (UTC) Received: from troutmask.apl.washington.edu (localhost.apl.washington.edu [127.0.0.1]) by troutmask.apl.washington.edu (8.14.5/8.14.5) with ESMTP id r1BMMMGQ083879; Mon, 11 Feb 2013 14:22:22 -0800 (PST) (envelope-from sgk@troutmask.apl.washington.edu) Received: (from sgk@localhost) by troutmask.apl.washington.edu (8.14.5/8.14.5/Submit) id r1BMMLKo083878; Mon, 11 Feb 2013 14:22:21 -0800 (PST) (envelope-from sgk) Date: Mon, 11 Feb 2013 14:22:21 -0800 From: Steve Kargl To: David Chisnall Subject: Re: 7+ days of dogfood Message-ID: <20130211222221.GA83846@troutmask.apl.washington.edu> References: <20130210000723.GA73630@troutmask.apl.washington.edu> <20130211114811.09e56b55@fabiankeil.de> <17E009FB-23FA-4E04-8437-DE81033164DE@FreeBSD.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <17E009FB-23FA-4E04-8437-DE81033164DE@FreeBSD.org> User-Agent: Mutt/1.5.21 (2010-09-15) Cc: freebsd-current@FreeBSD.org X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 11 Feb 2013 22:22:31 -0000 On Mon, Feb 11, 2013 at 01:29:12PM +0000, David Chisnall wrote: > On 11 Feb 2013, at 10:48, Fabian Keil wrote: > > > It's unfortunate that the builworld time roughly trippled since > > 2010 but I guess that's progress and a more powerful system > > should fix it. I certainly welcome clang in general, though. > > In that case, it's worth noting that you can shave a fair bit off > the build time by not building gcc. WITHOUT_GCC=yes in src.conf > is worthwhile. > While not building a part of the base system will obviously speed up bulidworld, it seems to me that you're being a little too generous here. The buildworld-speed issue is clearly a problem with clang/llvm. On my very lightly loaded, 4-core opteron system with 16 GB of memory, I see WITH_CLANG="YES" WITH_GCC="YES" rm -rf /usr/obj/* time make -j4 buildworld 3634.55 real 9784.21 user 1286.08 sys WITH_CLANG="YES" WITHOUT_GCC="YES" rm -rf /usr/obj/* time make -j4 buildworld 3489.40 real 9413.90 user 1324.72 sys WITHOUT_CLANG="YES" WITH_GCC="YES" rm -rf /usr/obj/* time make -j4 buildworld 1928.38 real 5254.85 user 1075.68 sys -- Steve