From mboxrd@z Thu Jan 1 00:00:00 1970 Path: news.gmane.org!.POSTED!not-for-mail From: Eli Zaretskii Newsgroups: gmane.emacs.devel Subject: Re: emacs-win-builder Date: Thu, 01 Sep 2016 17:05:28 +0300 Message-ID: <83inufg1g7.fsf@gnu.org> References: <1137fded-fc14-2445-bc6b-e8f2aceed56c@gmail.com> <83y43dkops.fsf@gnu.org> <2daa28a1-8223-b6e3-3424-ff281a3fa67e@gmail.com> Reply-To: Eli Zaretskii NNTP-Posting-Host: blaine.gmane.org X-Trace: blaine.gmane.org 1472738765 29006 195.159.176.226 (1 Sep 2016 14:06:05 GMT) X-Complaints-To: usenet@blaine.gmane.org NNTP-Posting-Date: Thu, 1 Sep 2016 14:06:05 +0000 (UTC) Cc: emacs-devel@gnu.org To: Nikolay Kudryavtsev Original-X-From: emacs-devel-bounces+ged-emacs-devel=m.gmane.org@gnu.org Thu Sep 01 16:05:57 2016 Return-path: Envelope-to: ged-emacs-devel@m.gmane.org Original-Received: from lists.gnu.org ([208.118.235.17]) by blaine.gmane.org with esmtp (Exim 4.84_2) (envelope-from ) id 1bfSco-0006Iv-NV for ged-emacs-devel@m.gmane.org; Thu, 01 Sep 2016 16:05:50 +0200 Original-Received: from localhost ([::1]:37275 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bfScm-0005Nl-6n for ged-emacs-devel@m.gmane.org; Thu, 01 Sep 2016 10:05:48 -0400 Original-Received: from eggs.gnu.org ([2001:4830:134:3::10]:40005) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bfScZ-0005MF-ED for emacs-devel@gnu.org; Thu, 01 Sep 2016 10:05:40 -0400 Original-Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bfScT-0003RJ-Dh for emacs-devel@gnu.org; Thu, 01 Sep 2016 10:05:34 -0400 Original-Received: from fencepost.gnu.org ([2001:4830:134:3::e]:32770) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bfScT-0003R9-AG; Thu, 01 Sep 2016 10:05:29 -0400 Original-Received: from 84.94.185.246.cable.012.net.il ([84.94.185.246]:4790 helo=home-c4e4a596f7) by fencepost.gnu.org with esmtpsa (TLS1.2:RSA_AES_128_CBC_SHA1:128) (Exim 4.82) (envelope-from ) id 1bfScS-0000F3-JE; Thu, 01 Sep 2016 10:05:29 -0400 In-reply-to: <2daa28a1-8223-b6e3-3424-ff281a3fa67e@gmail.com> (message from Nikolay Kudryavtsev on Thu, 1 Sep 2016 15:39:42 +0300) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] X-Received-From: 2001:4830:134:3::e X-BeenThere: emacs-devel@gnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: "Emacs development discussions." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: emacs-devel-bounces+ged-emacs-devel=m.gmane.org@gnu.org Original-Sender: "Emacs-devel" Xref: news.gmane.org gmane.emacs.devel:207065 Archived-At: > From: Nikolay Kudryavtsev > Cc: emacs-devel@gnu.org > Date: Thu, 1 Sep 2016 15:39:42 +0300 > > E.g., on my systems, MSYS can only work when invoked from the MSYS Bash > window. > > That's interesting. What's the possible breaking point here? Assuming that we clear our exec-path, use > bash.exe from msys directory and add whatever necessary to the process environment, I could not think of > anything that should break here, that's not easily fixable. Mainly, PATH (in process-environment) vs exec-path, which is exacerbated by MSYS remapping of /usr etc. to Windows directory structure. I think you worked around that for the case that MinGW and MSYS are installed as part of your script, but you cannot possibly do that reliably for a system where both are already installed. E.g., on my system, MSYS and MinGW are in two separate trees, side by side, and there's no easy way for you to know how MSYS /usr/bin is remapped to its Windows name. > For libarchive we would need some way to extract zip files > first. GnuWin provides a setup version of unzip, so I guess that > would be one option. Doesn't w32-shell-execute work? The Explorer can unzip archives, so perhaps this could be an alternative, if it works. > First, what would be some reasonable optimization profiles. I think we all agree that we want to have > something with debug information as default. So what would be the best flags here? And then there should be > an option to make an optimized build, what would be the best flags for it? I use "-O0 -gdwarf-2 -g3" for debug builds, and "-O2 -gdwarf-2 -g3" (or even -gdwarf-4 for GCC 5 and later) for optimized builds. GNU projects normally do "make install-strip" for the official releases and "make install" for pretests and snapshots, so my advice would be to do that here as well. > Also I do > "--enable-checking='yes,glyphs' --enable-check-lisp-object-type" > everywhere. It's a good idea in pretests and snapshots, but I think you should remove it for official releases. They just make Emacs slower, which is not a good thing in production. > Also, are there any other configure related options we may want to implement? The --with-modules switch comes to mind, but that's all. For 32-bit builds, I'd also recommend --with-wide-int, but I know some people disagree.