From mboxrd@z Thu Jan 1 00:00:00 1970 Path: news.gmane.io!.POSTED.blaine.gmane.org!not-for-mail From: Andrea Corallo via "Bug reports for GNU Emacs, the Swiss army knife of text editors" Newsgroups: gmane.emacs.bugs Subject: bug#43269: 28.0.50; [feature/native-comp] provide a user feedback on Emacs being native compiled Date: Wed, 09 Sep 2020 15:19:58 +0000 Message-ID: References: <83wo15msrj.fsf@gnu.org> <83tuw9mmfe.fsf@gnu.org> <83sgbtm2d3.fsf@gnu.org> <062E9DA2-B039-4AAD-802B-7333971F5B73@gnu.org> <83ft7smjcd.fsf@gnu.org> <83h7s7kpfr.fsf@gnu.org> Reply-To: Andrea Corallo Mime-Version: 1.0 Content-Type: text/plain Injection-Info: ciao.gmane.io; posting-host="blaine.gmane.org:116.202.254.214"; logging-data="22302"; mail-complaints-to="usenet@ciao.gmane.io" User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/26.3 (gnu/linux) Cc: rms@gnu.org, arthur.miller@live.com, 43269@debbugs.gnu.org To: Eli Zaretskii Original-X-From: bug-gnu-emacs-bounces+geb-bug-gnu-emacs=m.gmane-mx.org@gnu.org Wed Sep 09 17:22:29 2020 Return-path: Envelope-to: geb-bug-gnu-emacs@m.gmane-mx.org Original-Received: from lists.gnu.org ([209.51.188.17]) by ciao.gmane.io with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1kG1vZ-0005iO-6z for geb-bug-gnu-emacs@m.gmane-mx.org; Wed, 09 Sep 2020 17:22:29 +0200 Original-Received: from localhost ([::1]:32820 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1kG1vY-0001BC-7X for geb-bug-gnu-emacs@m.gmane-mx.org; Wed, 09 Sep 2020 11:22:28 -0400 Original-Received: from eggs.gnu.org ([2001:470:142:3::10]:38858) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1kG1uA-0006ty-3H for bug-gnu-emacs@gnu.org; Wed, 09 Sep 2020 11:21:02 -0400 Original-Received: from debbugs.gnu.org ([209.51.188.43]:49645) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1kG1u9-0004x5-Oq for bug-gnu-emacs@gnu.org; Wed, 09 Sep 2020 11:21:01 -0400 Original-Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1kG1u9-00039X-L8 for bug-gnu-emacs@gnu.org; Wed, 09 Sep 2020 11:21:01 -0400 X-Loop: help-debbugs@gnu.org Resent-From: Andrea Corallo Original-Sender: "Debbugs-submit" Resent-CC: bug-gnu-emacs@gnu.org Resent-Date: Wed, 09 Sep 2020 15:21:01 +0000 Resent-Message-ID: Resent-Sender: help-debbugs@gnu.org X-GNU-PR-Message: followup 43269 X-GNU-PR-Package: emacs Original-Received: via spool by 43269-submit@debbugs.gnu.org id=B43269.159966480912037 (code B ref 43269); Wed, 09 Sep 2020 15:21:01 +0000 Original-Received: (at 43269) by debbugs.gnu.org; 9 Sep 2020 15:20:09 +0000 Original-Received: from localhost ([127.0.0.1]:32954 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1kG1tI-000383-U5 for submit@debbugs.gnu.org; Wed, 09 Sep 2020 11:20:09 -0400 Original-Received: from mab.sdf.org ([205.166.94.33]:33202 helo=ma.sdf.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1kG1tD-00037j-4o for 43269@debbugs.gnu.org; Wed, 09 Sep 2020 11:20:07 -0400 Original-Received: from akrl by ma.sdf.org with local (Exim 4.92) (envelope-from ) id 1kG1t8-0007mw-I9; Wed, 09 Sep 2020 15:19:58 +0000 In-Reply-To: <83h7s7kpfr.fsf@gnu.org> (Eli Zaretskii's message of "Wed, 09 Sep 2020 17:14:16 +0300") X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list X-BeenThere: bug-gnu-emacs@gnu.org List-Id: "Bug reports for GNU Emacs, the Swiss army knife of text editors" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: bug-gnu-emacs-bounces+geb-bug-gnu-emacs=m.gmane-mx.org@gnu.org Original-Sender: "bug-gnu-emacs" Xref: news.gmane.io gmane.emacs.bugs:187673 Archived-At: Eli Zaretskii writes: >> From: Richard Stallman >> Cc: akrl@sdf.org, arthur.miller@live.com, 43269@debbugs.gnu.org >> Date: Tue, 08 Sep 2020 23:45:06 -0400 >> >> > As mentioned on the bug list, I'd prefer if user could defer native >> > compilation to some later time, so as to avoid making the Emacs build >> > take hours, especially on slow and low-end machines. >> >> I wonder if I would ever see any benefit form the speedup of native >> compilation. I hardly ever notice waiting for Emacs to do computation. >> But I would find a big slowdown in building to be a pain. >> >> Maybe I would prefer to turn off native compilation, pure and simple. > > I share some of these feelings, FWIW. Andrea's work is, of course, > commendable and the results will be very welcome when they land on > master, but I'm disappointed by the high price we need to pay for this > feature, both in complexity (notice the long discussions of where and > how to store the *.eln files, and how to handle recompilation and > reloading), and in compilation times. Having the single-core > compilation times increase from 10-15 min to several hours is > ... extreme. (And before you say no one runs this on a single core: I > sometimes do, when parallel builds get in the way of debugging some > problem.) And we will probably bump into additional issues down the > road. > > (How come it's so easy and seamless in Guile?) > Hi Eli, the native compiler improved considerably the compilation speed with time. I just took a measure at today's status native compiling only the dumped image (what is going to be default when native compiling). On my dev machine vanilla Emacs uses 12m tot CPU time for a compilation from a fresh repo. The same native compiling takes 30m tot CPU time so IMO it is not terrible. Regarding the complexity I don't know, I guess it took some message to decide how to have it working but now we are there. It looks to me way simpler then deciding Emacs defaults :) For Guile I have no idea if it was simpler to implement or discuss. I'm not a Guile expert so I may be inaccurate but I think they can native compile with a simple lightening based jitter. This let me think they can't save or reuse the compilation output, nor dump it given everything happens in memory. Regards Andrea