From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mp2 ([2001:41d0:2:4a6f::]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits)) by ms11 with LMTPS id YFJYO9rFAGBCTAAA0tVLHw (envelope-from ) for ; Thu, 14 Jan 2021 22:29:46 +0000 Received: from aspmx1.migadu.com ([2001:41d0:2:4a6f::]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits)) by mp2 with LMTPS id uCI+N9rFAGBTdwAAB5/wlQ (envelope-from ) for ; Thu, 14 Jan 2021 22:29:46 +0000 Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by aspmx1.migadu.com (Postfix) with ESMTPS id 6D746940309 for ; Thu, 14 Jan 2021 22:29:46 +0000 (UTC) Received: from localhost ([::1]:50116 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1l0B7h-0000yB-Eb for larch@yhetil.org; Thu, 14 Jan 2021 17:29:45 -0500 Received: from eggs.gnu.org ([2001:470:142:3::10]:56518) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1l0B7O-0000wk-PG for guix-devel@gnu.org; Thu, 14 Jan 2021 17:29:26 -0500 Received: from fencepost.gnu.org ([2001:470:142:3::e]:32964) by eggs.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1l0B7K-0007pL-9Y; Thu, 14 Jan 2021 17:29:23 -0500 Received: from [2a01:e0a:1d:7270:af76:b9b:ca24:c465] (port=53744 helo=ribbon) by fencepost.gnu.org with esmtpsa (TLS1.2:RSA_AES_256_CBC_SHA1:256) (Exim 4.82) (envelope-from ) id 1l0B7H-0006zM-FO; Thu, 14 Jan 2021 17:29:19 -0500 From: =?utf-8?Q?Ludovic_Court=C3=A8s?= To: Mathieu Othacehe Subject: Re: Staging branch [substitute availability] References: <87eein6hf6.fsf@gnu.org> X-URL: http://www.fdn.fr/~lcourtes/ X-Revolutionary-Date: 25 =?utf-8?Q?Niv=C3=B4se?= an 229 de la =?utf-8?Q?R?= =?utf-8?Q?=C3=A9volution?= X-PGP-Key-ID: 0x090B11993D9AEBB5 X-PGP-Key: http://www.fdn.fr/~lcourtes/ludovic.asc X-PGP-Fingerprint: 3CE4 6455 8A84 FDC6 9DB4 0CFB 090B 1199 3D9A EBB5 X-OS: x86_64-pc-linux-gnu Date: Thu, 14 Jan 2021 23:29:17 +0100 In-Reply-To: <87eein6hf6.fsf@gnu.org> (Mathieu Othacehe's message of "Thu, 14 Jan 2021 09:39:41 +0100") Message-ID: <87pn27ywxu.fsf@gnu.org> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/27.1 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-BeenThere: guix-devel@gnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: guix-devel@gnu.org Errors-To: guix-devel-bounces+larch=yhetil.org@gnu.org Sender: "Guix-devel" X-Migadu-Flow: FLOW_IN X-Migadu-Spam-Score: -2.86 Authentication-Results: aspmx1.migadu.com; dkim=none; dmarc=pass (policy=none) header.from=gnu.org; spf=pass (aspmx1.migadu.com: domain of guix-devel-bounces@gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=guix-devel-bounces@gnu.org X-Migadu-Queue-Id: 6D746940309 X-Spam-Score: -2.86 X-Migadu-Scanner: scn0.migadu.com X-TUID: +XjTDNC80KTt Mathieu Othacehe skribis: > Since the introduction of the "wip-offload" branch on Cuirass, the > situation has much improved. The workers are constantly building. For > now we are building three specifications: > > * guix-modular-master > * guix-master > * staging Yay! > for x86_64, i686 and aarch64. If you look at the "Pending builds" chart > here[1], you will see that the CI is barely catching up. That's because > the "aarch64" emulated builds are incredibly slow, and monopolizing all > the build resources. > > I deliberately chose to put armhf aside until I have a clearer view of > the situation. > > Now, how to move on? > > First, I still need to connect the four overdrives machine to the new > Cuirass remote building mechanism, and I would need some help for that > (asked on guix-sysadmins). But, I'm not sure it will much improve the > situation. Oh sorry, I still haven=E2=80=99t caught up from vacation but I=E2=80=99ll = take a look if nobody beats me at it. > Longer term, we need to figure out a better solution. It's now > obvious that we do not have the computation power to build all our > branches for 5 different architectures, relying heavily on emulation for > armhf and aarch64. Anyone knows how Nix deals with that? I=E2=80=99m not sure, but I know they rent storage and processing power fro= m a big transnational company, and that may well include AArch64. Note that we disabled emulated builds and ARMv7 builds on AArch64 (!) when Danny discovered the _FILE_OFFSET_BITS issue, which makes things much worse. With the x86_64 machines we have in Berlin, using emulated builds, even if they=E2=80=99re slow, could potentially help noticeably. At this point the biggest issue is ARMv7 because we have too little actual hardware. > I guess that other major distributions provide only cross-compiled > packages for those architectures, but I don't think it's an option for > us, Ludo? Cross-compiled derivations are different derivations, so no, it=E2=80=99s n= ot an option. If people know what hardware to get, and if we can find people to host it, we have enough funds to buy it. On IRC yesterday Leo mentioned a good-looking AArch64 board: https://shop.solid-run.com/product/SRLX216S00D00GE064H07CH/ For ARMv7, there are probably several known-good options like those by Olimex, BeagleBoard (I think?) and the likes. For any such candidate, we need to (1) check it can be used with free software only, (2) check things like provided storage space, whether a case is available, etc., and (3) plan for purchase and hosting. Volunteers needed! Thanks, Ludo=E2=80=99.