From owner-freebsd-current@FreeBSD.ORG Fri Dec 2 18:42:17 2011 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 98B421065670 for ; Fri, 2 Dec 2011 18:42:17 +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 6C8E78FC20 for ; Fri, 2 Dec 2011 18:42:17 +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 pB2IgHlD044341; Fri, 2 Dec 2011 10:42:17 -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 pB2IgHvb044340; Fri, 2 Dec 2011 10:42:17 -0800 (PST) (envelope-from sgk) Date: Fri, 2 Dec 2011 10:42:17 -0800 From: Steve Kargl To: Ryan Stone Message-ID: <20111202184217.GA3911@troutmask.apl.washington.edu> References: <4ED8776F.9060301@FreeBSD.org> <20111202072349.GA89183@troutmask.apl.washington.edu> <20111202083501.GA73959@dragon.NUXI.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4.2.3i Cc: freebsd-current@freebsd.org, Lyndon Nerenberg , Chris Rees Subject: Re: WITHOUT_PROFILE=yes by default X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 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: Fri, 02 Dec 2011 18:42:17 -0000 On Fri, Dec 02, 2011 at 01:12:42PM -0500, Ryan Stone wrote: > On Fri, Dec 2, 2011 at 12:07 PM, Lyndon Nerenberg wrote: > > No, these days 'obsolete' seems to mean 'it does not have a sexy > > Flash-driven web GUI.' > > In this case, 'obsolete' means it's a difficult-to-use tool that > requires recompiling your application, can't be used in production, > doesn't work when shared libraries are in the picture, offers > limited-to-no visibility into the underlying reasons why a particular > code path is a hotspot and introduces large measurement errors Difficult to use? % gfortran -o ang -pg ang.f90 % ./ang % gprof -b -l ./ang ang.gmon | more ... % cumulative self self total time seconds seconds calls ms/call ms/call name 35.0 0.01 0.01 0 100.00% _write [1] 33.3 0.02 0.01 0 100.00% _mcount [2] 15.0 0.02 0.00 1080 0.00 0.00 arena_purge [4] 5.6 0.02 0.00 0 100.00% .mcount (40) 2.2 0.02 0.00 29600 0.00 0.00 __quorem_D2A [8] 1.7 0.02 0.00 1080 0.00 0.00 __dtoa [7] 1.1 0.02 0.00 29552 0.00 0.00 __multadd_D2A [13] 1.1 0.02 0.00 7557 0.00 0.00 memcpy [12] Please show me how you would get the same information with pmcstat (or other tools) in the base system. Note, ang.f90 is a toy app I had lying around, which completes in a second or 2. If you want a non-toy example, I'll happily run one of my libm testcase for you. -- Steve