From mboxrd@z Thu Jan 1 00:00:00 1970 From: ludo@gnu.org (Ludovic =?utf-8?Q?Court=C3=A8s?=) Subject: Re: Ready for Guile 2.2! Date: Fri, 21 Apr 2017 23:03:37 +0200 Message-ID: <87pog5a2xi.fsf@gnu.org> References: <87bmt28qnm.fsf@gnu.org> <87lgs65l3c.fsf@gnu.org> <87r30nz275.fsf@gnu.org> <87y3uvf6l8.fsf@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:46049) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1d1fiR-0003s5-PD for guix-devel@gnu.org; Fri, 21 Apr 2017 17:03:44 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1d1fiO-0000qA-Hv for guix-devel@gnu.org; Fri, 21 Apr 2017 17:03:43 -0400 In-Reply-To: <87y3uvf6l8.fsf@gmail.com> (Maxim Cournoyer's message of "Fri, 21 Apr 2017 00:21:07 +0900") List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org Sender: "Guix-devel" To: Maxim Cournoyer Cc: guix-devel Maxim Cournoyer skribis: > Andy Wingo writes: > >> On Thu 20 Apr 2017 14:35, ludo@gnu.org (Ludovic Court=C3=A8s) writes: >> >>> ludo@gnu.org (Ludovic Court=C3=A8s) skribis: >>> >>>> ;; 2.724686s real time, 3.117062s run time. 0.880827s spent in GC. >>>> scheme@(guile-user)> (version) >>>> $1 =3D "2.0.13" >>>> >>> scheme@(guile-user)> ,use(guix scripts build) >>> scheme@(guile-user)> ,time (guix-build "libreoffice" "certbot" "xmonad"= "-n" "--no-substitutes" "--no-build-hook") >>> >>> [...] >>> >>> ;; 1.826528s real time, 1.994426s run time. 0.382750s spent in GC. >>> scheme@(guile-user)> (version) >>> $1 =3D "2.2.1" >>> >>> That=E2=80=99s a 33% speedup compared to 2.0. >> >> That is a 50% speedup compared to 2.0 :) If we consider its speed as >> being how many times you could do this per second, then 2.0 speed is >> 1/2.72, and 2.2.1 speed is 1/1.82. Speed ratio is then >> 2.72/1.82=3D1.4945. So 2.2.1 is 1.5x the speed of 2.0, or 50% faster :) >> >> Andy, who is not looking for praise, but who likes perf numbers :) > > Impressive :). Is there a blog post/article/information somewhere about > what went in Guile to make it that faster? I highly recommend posts from the last couple of years (or more!) at . :-) has pointers to the most important bits. Ludo=E2=80=99.