[{"id":1762799,"web_url":"http://patchwork.ozlabs.org/comment/1762799/","msgid":"<20170904172137.GB5223@scaer>","list_archive_url":null,"date":"2017-09-04T17:21:37","subject":"Re: [Buildroot] [PATCH 4/9] arch/arm: do not allow soft-float for\n\tarmv8a","submitter":{"id":13903,"url":"http://patchwork.ozlabs.org/api/people/13903/","name":"Yann E. MORIN","email":"yann.morin.1998@free.fr"},"content":"Thomas, All,\n\nOn 2017-09-03 15:17 +0200, Yann E. MORIN spake thusly:\n> armv8a has made VFPv4 and NEON mandatory, so there is no point in\n> allowing software floating point, even in 32-bit mode.\n\nIn fact, even though I'm pretty sure that OK in 64-bit, we might still\nhave a reason to do soft-float in 32-bit mode, if only for legacy\nbinary-only applications that were built as soft-float way back in the\n(not so) good old days...\n\nIn which case, we should instead depend on !BR2_ARCH_IS_64.\n\nThoughts?\n\nRegards,\nYann E. MORIN.\n\n> Signed-off-by: \"Yann E. MORIN\" \n> Cc: Thomas Petazzoni \n> ---\n> arch/Config.in.arm | 1 +\n> 1 file changed, 1 insertion(+)\n> \n> diff --git a/arch/Config.in.arm b/arch/Config.in.arm\n> index d110d52e2d..0ae6a16a0a 100644\n> --- a/arch/Config.in.arm\n> +++ b/arch/Config.in.arm\n> @@ -374,6 +374,7 @@ choice\n> config BR2_ARM_SOFT_FLOAT\n> \tbool \"Soft float\"\n> \tdepends on BR2_ARM_EABI\n> +\tdepends on !BR2_ARM_CPU_ARMV8A\n> \tselect BR2_SOFT_FLOAT\n> \thelp\n> \t This option allows to use software emulated floating\n> -- \n> 2.11.0\n>","headers":{"Return-Path":"","X-Original-To":["incoming@patchwork.ozlabs.org","buildroot@lists.busybox.net"],"Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","buildroot@osuosl.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=busybox.net\n\t(client-ip=140.211.166.138; helo=whitealder.osuosl.org;\n\tenvelope-from=buildroot-bounces@busybox.net;\n\treceiver=)","ozlabs.org;\n\tdkim=fail reason=\"signature verification failed\" (2048-bit key;\n\tunprotected) header.d=gmail.com header.i=@gmail.com\n\theader.b=\"G/3b2cXO\"; dkim-atps=neutral"],"Received":["from whitealder.osuosl.org (smtp1.osuosl.org [140.211.166.138])\n\t(using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xmGqL3HPhz9t2R\n\tfor ;\n\tTue, 5 Sep 2017 03:21:50 +1000 (AEST)","from localhost (localhost [127.0.0.1])\n\tby whitealder.osuosl.org (Postfix) with ESMTP id 9F8308655F;\n\tMon, 4 Sep 2017 17:21:46 +0000 (UTC)","from whitealder.osuosl.org ([127.0.0.1])\n\tby localhost (.osuosl.org [127.0.0.1]) (amavisd-new, port 10024)\n\twith ESMTP id eL7CSOy9SBMr; Mon, 4 Sep 2017 17:21:45 +0000 (UTC)","from ash.osuosl.org (ash.osuosl.org [140.211.166.34])\n\tby whitealder.osuosl.org (Postfix) with ESMTP id 4604B8645F;\n\tMon, 4 Sep 2017 17:21:45 +0000 (UTC)","from whitealder.osuosl.org (smtp1.osuosl.org [140.211.166.138])\n\tby ash.osuosl.org (Postfix) with ESMTP id 4076C1C0822\n\tfor ;\n\tMon, 4 Sep 2017 17:21:44 +0000 (UTC)","from localhost (localhost [127.0.0.1])\n\tby whitealder.osuosl.org (Postfix) with ESMTP id 3B2928643F\n\tfor ;\n\tMon, 4 Sep 2017 17:21:44 +0000 (UTC)","from whitealder.osuosl.org ([127.0.0.1])\n\tby localhost (.osuosl.org [127.0.0.1]) (amavisd-new, port 10024)\n\twith ESMTP id SiqKlMe1OMgE for ;\n\tMon, 4 Sep 2017 17:21:43 +0000 (UTC)","from mail-wm0-f65.google.com (mail-wm0-f65.google.com\n\t[74.125.82.65])\n\tby whitealder.osuosl.org (Postfix) with ESMTPS id E0C3486410\n\tfor ; Mon, 4 Sep 2017 17:21:42 +0000 (UTC)","by mail-wm0-f65.google.com with SMTP id x189so865897wmg.4\n\tfor ; Mon, 04 Sep 2017 10:21:42 -0700 (PDT)","from ymorin.is-a-geek.org\n\t(2a01cb08861073003d7cb41da3cf6159.ipv6.abo.wanadoo.fr.\n\t[2a01:cb08:8610:7300:3d7c:b41d:a3cf:6159])\n\tby smtp.gmail.com with ESMTPSA id\n\te17sm924606wmd.13.2017.09.04.10.21.38\n\t(version=TLS1 cipher=AES128-SHA bits=128/128);\n\tMon, 04 Sep 2017 10:21:39 -0700 (PDT)","by ymorin.is-a-geek.org (sSMTP sendmail emulation);\n\tMon, 04 Sep 2017 19:21:37 +0200"],"X-Virus-Scanned":["amavisd-new at osuosl.org","amavisd-new at osuosl.org"],"X-Greylist":"domain auto-whitelisted by SQLgrey-1.7.6","DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025;\n\th=sender:date:from:to:cc:subject:message-id:references:mime-version\n\t:content-disposition:in-reply-to:user-agent;\n\tbh=SZZSo9NWZ5+4pNxHPJt6T2Skl+Z6URe4V8lLZLdvmBA=;\n\tb=G/3b2cXOrZePxgpfjFJpWhuLMajKfRmC4sv5AkSXPs7B30MGp4+GWRRcS0f5u1Ab7k\n\tbReTxEjn2tfleX/fJx+gUpfqGfxgdgUCKeU+Ydieqs0zulsXW1XVK71Ufh9rPq3MB25p\n\t2rxM3PF8T4GkPPP15CsW7cFlPEeoWb4BjkjYsmp3F3sbpLEGtpYe3shJFa3YXbpKcwXD\n\trA6yuh1tQm1084wO461OHMT5NLpfv/40lzNarrDs09oJwdpoDmzuX3QsDlN4n2l3RsZH\n\txbgLwKOL2B5+eiT+2ptvC5Kh0nOkh9ZeYOoxmTD5YaXPOAze1wWJuNleK31yHHAu/Z0Q\n\tB83Q==","X-Google-DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed;\n\td=1e100.net; s=20161025;\n\th=x-gm-message-state:sender:date:from:to:cc:subject:message-id\n\t:references:mime-version:content-disposition:in-reply-to:user-agent; \n\tbh=SZZSo9NWZ5+4pNxHPJt6T2Skl+Z6URe4V8lLZLdvmBA=;\n\tb=ssWDlqCJfGMdEPpi3ekXCLHSWp0YSUwUQn394MH89jMeNXHf6iNOe9mTEqATPXwNn+\n\tK+/62EsUSMjtmbDt6n8GHxmmiZh7FEw+EMoGSBr1vyzuFHV2nEegpCHhLOOmLd2izHjN\n\tN84SyxVzH8V2CGnyvTlsvIJySWe/Nmcl04HATBNa4wCohBMx5TATb2jsRZkSbfxx5Gkg\n\tL2a36Ur+cUaqXv//WlppLyPuaU33cbQ/AGaux6NwW7P5CX1IK1WTuY/t84ren3KQouGY\n\t+Nf1S0u4A9Ha08N3J1vVdw+Too9zH9YLE55rtuBHtId1+lo+PlQH03N7Lpmy93cKZva/\n\tG/pA==","X-Gm-Message-State":"AHPjjUj+61LhwChRqdr56pFvkzHxmWyNAwF8IbjOEnkM1ocboB1VzEWu\n\t222VnMe09j1CbtMN","X-Google-Smtp-Source":"ADKCNb4x23rJEGCFICbyHR8REChHyONheZPSpUqfJXVFiW4ibRTwh34bTZRGmaa0S22KPbgpVwhrFg==","X-Received":"by 10.28.100.136 with SMTP id y130mr999996wmb.81.1504545701073; \n\tMon, 04 Sep 2017 10:21:41 -0700 (PDT)","Date":"Mon, 4 Sep 2017 19:21:37 +0200","From":"\"Yann E. MORIN\" ","To":"buildroot@buildroot.org","Message-ID":"<20170904172137.GB5223@scaer>","References":"\n\t","MIME-Version":"1.0","Content-Disposition":"inline","In-Reply-To":"","User-Agent":"Mutt/1.5.22 (2013-10-16)","Cc":"Thomas Petazzoni ","Subject":"Re: [Buildroot] [PATCH 4/9] arch/arm: do not allow soft-float for\n\tarmv8a","X-BeenThere":"buildroot@busybox.net","X-Mailman-Version":"2.1.18-1","Precedence":"list","List-Id":"Discussion and development of buildroot ","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Content-Type":"text/plain; charset=\"us-ascii\"","Content-Transfer-Encoding":"7bit","Errors-To":"buildroot-bounces@busybox.net","Sender":"\"buildroot\" "}},{"id":1809957,"web_url":"http://patchwork.ozlabs.org/comment/1809957/","msgid":"<20171124230857.0b6e1648@windsurf.lan>","list_archive_url":null,"date":"2017-11-24T22:08:57","subject":"Re: [Buildroot] [PATCH 4/9] arch/arm: do not allow soft-float for\n\tarmv8a","submitter":{"id":2230,"url":"http://patchwork.ozlabs.org/api/people/2230/","name":"Thomas Petazzoni","email":"thomas.petazzoni@free-electrons.com"},"content":"Hello,\n\nOn Mon, 4 Sep 2017 19:21:37 +0200, Yann E. MORIN wrote:\n> Thomas, All,\n> \n> On 2017-09-03 15:17 +0200, Yann E. MORIN spake thusly:\n> > armv8a has made VFPv4 and NEON mandatory, so there is no point in\n> > allowing software floating point, even in 32-bit mode. \n> \n> In fact, even though I'm pretty sure that OK in 64-bit, we might still\n> have a reason to do soft-float in 32-bit mode, if only for legacy\n> binary-only applications that were built as soft-float way back in the\n> (not so) good old days...\n> \n> In which case, we should instead depend on !BR2_ARCH_IS_64.\n> \n> Thoughts?\n\nI'm hesitating on this one. I believe the use-case you mention makes\nsense, however I believe that it can still be supported with your\nproposed patch.\n\nIndeed, even if you want to run a binary-only soft-float application,\nit doesn't prevent the rest of your system from using VFP, as long as\nyou use EABI (and not EABIhf).\n\nThe whole reason why EABI passes floating-point arguments into integer\nregisters is precisely to allow compatibility between soft-float code\nand hard-float code. Thanks to that, a soft-float function can call a\nhard-float function, and vice-versa.\n\nTherefore, even if you have a binary-only soft-float application, there\nis really no reason to build the entire system soft-float.\n\nSo, I think your original patch is OK as-is. However, I still see one\ninconsistent thing: if we disallow soft-float because ARMv8-A mandates\nVFPv4, why would we allow soft-float for ARMv7-A cores that have a\nVFP ? To be consistent, we should also disallow soft-float.\n\nAnother way to think about this is: is it possible to build an AArch64\nsoft-float system? do we want to allow the user to select what is\n*possible* or what makes sense?\n\nBest regards,\n\nThomas","headers":{"Return-Path":"","X-Original-To":["incoming@patchwork.ozlabs.org","buildroot@lists.busybox.net"],"Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","buildroot@osuosl.org"],"Authentication-Results":"ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=busybox.net\n\t(client-ip=140.211.166.137; helo=fraxinus.osuosl.org;\n\tenvelope-from=buildroot-bounces@busybox.net;\n\treceiver=)","Received":["from fraxinus.osuosl.org (smtp4.osuosl.org [140.211.166.137])\n\t(using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yk9nc1nWsz9s5L\n\tfor ;\n\tSat, 25 Nov 2017 09:28:20 +1100 (AEDT)","from localhost (localhost [127.0.0.1])\n\tby fraxinus.osuosl.org (Postfix) with ESMTP id F141388161;\n\tFri, 24 Nov 2017 22:28:17 +0000 (UTC)","from fraxinus.osuosl.org ([127.0.0.1])\n\tby localhost (.osuosl.org [127.0.0.1]) (amavisd-new, port 10024)\n\twith ESMTP id WyyuCc4M8JV2; Fri, 24 Nov 2017 22:28:16 +0000 (UTC)","from ash.osuosl.org (ash.osuosl.org [140.211.166.34])\n\tby fraxinus.osuosl.org (Postfix) with ESMTP id 0794B88143;\n\tFri, 24 Nov 2017 22:28:16 +0000 (UTC)","from hemlock.osuosl.org (smtp2.osuosl.org [140.211.166.133])\n\tby ash.osuosl.org (Postfix) with ESMTP id 284B61C2040\n\tfor ;\n\tFri, 24 Nov 2017 22:09:01 +0000 (UTC)","from localhost (localhost [127.0.0.1])\n\tby hemlock.osuosl.org (Postfix) with ESMTP id 1F0DC8972B\n\tfor ;\n\tFri, 24 Nov 2017 22:09:01 +0000 (UTC)","from hemlock.osuosl.org ([127.0.0.1])\n\tby localhost (.osuosl.org [127.0.0.1]) (amavisd-new, port 10024)\n\twith ESMTP id e4D-QUc0u2SL for ;\n\tFri, 24 Nov 2017 22:08:59 +0000 (UTC)","from mail.free-electrons.com (mail.free-electrons.com [62.4.15.54])\n\tby hemlock.osuosl.org (Postfix) with ESMTP id 90BB78972A\n\tfor ; Fri, 24 Nov 2017 22:08:59 +0000 (UTC)","by mail.free-electrons.com (Postfix, from userid 110)\n\tid 074C9213E2; Fri, 24 Nov 2017 23:08:58 +0100 (CET)","from windsurf.lan (LFbn-TOU-1-149-75.w86-201.abo.wanadoo.fr\n\t[86.201.231.75])\n\tby mail.free-electrons.com (Postfix) with ESMTPSA id D93F320750;\n\tFri, 24 Nov 2017 23:08:57 +0100 (CET)"],"X-Virus-Scanned":["amavisd-new at osuosl.org","amavisd-new at osuosl.org"],"X-Greylist":"domain auto-whitelisted by SQLgrey-1.7.6","Date":"Fri, 24 Nov 2017 23:08:57 +0100","From":"Thomas Petazzoni ","To":"\"Yann E. MORIN\" ","Message-ID":"<20171124230857.0b6e1648@windsurf.lan>","In-Reply-To":"<20170904172137.GB5223@scaer>","References":"\n\t\n\t<20170904172137.GB5223@scaer>","Organization":"Free Electrons","X-Mailer":"Claws Mail 3.15.1-dirty (GTK+ 2.24.31; x86_64-redhat-linux-gnu)","MIME-Version":"1.0","Subject":"Re: [Buildroot] [PATCH 4/9] arch/arm: do not allow soft-float for\n\tarmv8a","X-BeenThere":"buildroot@busybox.net","X-Mailman-Version":"2.1.24","Precedence":"list","List-Id":"Discussion and development of buildroot ","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Cc":"buildroot@buildroot.org","Content-Type":"text/plain; charset=\"us-ascii\"","Content-Transfer-Encoding":"7bit","Errors-To":"buildroot-bounces@busybox.net","Sender":"\"buildroot\" "}},{"id":1810063,"web_url":"http://patchwork.ozlabs.org/comment/1810063/","msgid":"<5508245a-3f6a-80b9-a7ad-b3b6bc5d5f9e@mind.be>","list_archive_url":null,"date":"2017-11-25T17:10:17","subject":"Re: [Buildroot] [PATCH 4/9] arch/arm: do not allow soft-float for\n\tarmv8a","submitter":{"id":11809,"url":"http://patchwork.ozlabs.org/api/people/11809/","name":"Arnout Vandecappelle","email":"arnout@mind.be"},"content":"On 24-11-17 23:08, Thomas Petazzoni wrote:\n> Hello,\n> \n> On Mon, 4 Sep 2017 19:21:37 +0200, Yann E. MORIN wrote:\n>> Thomas, All,\n>>\n>> On 2017-09-03 15:17 +0200, Yann E. MORIN spake thusly:\n>>> armv8a has made VFPv4 and NEON mandatory, so there is no point in\n>>> allowing software floating point, even in 32-bit mode. \n>>\n>> In fact, even though I'm pretty sure that OK in 64-bit, we might still\n>> have a reason to do soft-float in 32-bit mode, if only for legacy\n>> binary-only applications that were built as soft-float way back in the\n>> (not so) good old days...\n>>\n>> In which case, we should instead depend on !BR2_ARCH_IS_64.\n>>\n>> Thoughts?\n> \n> I'm hesitating on this one. I believe the use-case you mention makes\n> sense, however I believe that it can still be supported with your\n> proposed patch.\n> \n> Indeed, even if you want to run a binary-only soft-float application,\n> it doesn't prevent the rest of your system from using VFP, as long as\n> you use EABI (and not EABIhf).\n> \n> The whole reason why EABI passes floating-point arguments into integer\n> registers is precisely to allow compatibility between soft-float code\n> and hard-float code. Thanks to that, a soft-float function can call a\n> hard-float function, and vice-versa.\n> \n> Therefore, even if you have a binary-only soft-float application, there\n> is really no reason to build the entire system soft-float.\n> \n> So, I think your original patch is OK as-is. However, I still see one\n> inconsistent thing: if we disallow soft-float because ARMv8-A mandates\n> VFPv4, why would we allow soft-float for ARMv7-A cores that have a\n> VFP ? To be consistent, we should also disallow soft-float.\n> \n> Another way to think about this is: is it possible to build an AArch64\n> soft-float system? do we want to allow the user to select what is\n> *possible* or what makes sense?\n\n gcc aarch64 doesn't have a float-abi option, so no it's not possible to build\nwith soft-float. So the depends on BR2_ARM_EABI is certainly needed.\n\n That said, I don't think it makes a lot of sense to add extra 'depends on'\nlines just to avoid that the user would select an option that probably doesn't\nmake sense. We already make sure that the default is appropriate. Let's not make\nour life more difficult.\n\n So I'd say: reject this patch.\n\n Regards,\n Arnout","headers":{"Return-Path":"","X-Original-To":["incoming@patchwork.ozlabs.org","buildroot@lists.busybox.net"],"Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","buildroot@osuosl.org"],"Authentication-Results":"ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=busybox.net\n\t(client-ip=140.211.166.136; helo=silver.osuosl.org;\n\tenvelope-from=buildroot-bounces@busybox.net;\n\treceiver=)","Received":["from silver.osuosl.org (smtp3.osuosl.org [140.211.166.136])\n\t(using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3ykg4B4Rwxz9s7B\n\tfor ;\n\tSun, 26 Nov 2017 04:27:38 +1100 (AEDT)","from localhost (localhost [127.0.0.1])\n\tby silver.osuosl.org (Postfix) with ESMTP id 4D1932E856;\n\tSat, 25 Nov 2017 17:27:35 +0000 (UTC)","from silver.osuosl.org ([127.0.0.1])\n\tby localhost (.osuosl.org [127.0.0.1]) (amavisd-new, port 10024)\n\twith ESMTP id OXeCF6JXOTYE; Sat, 25 Nov 2017 17:27:32 +0000 (UTC)","from ash.osuosl.org (ash.osuosl.org [140.211.166.34])\n\tby silver.osuosl.org (Postfix) with ESMTP id 129CE2E869;\n\tSat, 25 Nov 2017 17:27:31 +0000 (UTC)","from silver.osuosl.org (smtp3.osuosl.org [140.211.166.136])\n\tby ash.osuosl.org (Postfix) with ESMTP id 5BC661C11AD\n\tfor ;\n\tSat, 25 Nov 2017 17:10:39 +0000 (UTC)","from localhost (localhost [127.0.0.1])\n\tby silver.osuosl.org (Postfix) with ESMTP id 558CD265FF\n\tfor ;\n\tSat, 25 Nov 2017 17:10:39 +0000 (UTC)","from silver.osuosl.org ([127.0.0.1])\n\tby localhost (.osuosl.org [127.0.0.1]) (amavisd-new, port 10024)\n\twith ESMTP id R089Vy7qhzfS for ;\n\tSat, 25 Nov 2017 17:10:38 +0000 (UTC)","from exchange.essensium.com (220.77.144.195.ipv4.evonet.be\n\t[195.144.77.220])\n\tby silver.osuosl.org (Postfix) with ESMTP id 754DE265B2\n\tfor ; Sat, 25 Nov 2017 17:10:38 +0000 (UTC)","from [10.8.0.14] (10.3.7.11) by beleexch01.local.ess-mail.com\n\t(10.3.7.8) with Microsoft SMTP Server (TLS) id 15.0.847.32;\n\tSat, 25 Nov 2017 18:10:19 +0100"],"X-Virus-Scanned":["amavisd-new at osuosl.org","amavisd-new at osuosl.org"],"X-Greylist":"domain auto-whitelisted by SQLgrey-1.7.6","To":"Thomas Petazzoni , \"Yann E. MORIN\"\n\t","References":"\n\t\n\t<20170904172137.GB5223@scaer> <20171124230857.0b6e1648@windsurf.lan>","From":"Arnout Vandecappelle ","Organization":"Essensium/Mind","Message-ID":"<5508245a-3f6a-80b9-a7ad-b3b6bc5d5f9e@mind.be>","Date":"Sat, 25 Nov 2017 18:10:17 +0100","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.4.0","MIME-Version":"1.0","In-Reply-To":"<20171124230857.0b6e1648@windsurf.lan>","Content-Language":"en-GB","X-Originating-IP":"[10.3.7.11]","X-ClientProxiedBy":"beleexch01.local.ess-mail.com (10.3.7.8) To\n\tbeleexch01.local.ess-mail.com (10.3.7.8)","Subject":"Re: [Buildroot] [PATCH 4/9] arch/arm: do not allow soft-float for\n\tarmv8a","X-BeenThere":"buildroot@busybox.net","X-Mailman-Version":"2.1.24","Precedence":"list","List-Id":"Discussion and development of buildroot ","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Cc":"buildroot@buildroot.org","Content-Type":"text/plain; charset=\"us-ascii\"","Content-Transfer-Encoding":"7bit","Errors-To":"buildroot-bounces@busybox.net","Sender":"\"buildroot\" "}},{"id":1810154,"web_url":"http://patchwork.ozlabs.org/comment/1810154/","msgid":"<20171126111257.GA3101@scaer>","list_archive_url":null,"date":"2017-11-26T11:12:57","subject":"Re: [Buildroot] [PATCH 4/9] arch/arm: do not allow soft-float for\n\tarmv8a","submitter":{"id":13903,"url":"http://patchwork.ozlabs.org/api/people/13903/","name":"Yann E. MORIN","email":"yann.morin.1998@free.fr"},"content":"Arnout, All,\n\nOn 2017-11-25 18:10 +0100, Arnout Vandecappelle spake thusly:\n> On 24-11-17 23:08, Thomas Petazzoni wrote:\n> > Hello,\n> > \n> > On Mon, 4 Sep 2017 19:21:37 +0200, Yann E. MORIN wrote:\n> >> Thomas, All,\n> >>\n> >> On 2017-09-03 15:17 +0200, Yann E. MORIN spake thusly:\n> >>> armv8a has made VFPv4 and NEON mandatory, so there is no point in\n> >>> allowing software floating point, even in 32-bit mode. \n> >>\n> >> In fact, even though I'm pretty sure that OK in 64-bit, we might still\n> >> have a reason to do soft-float in 32-bit mode, if only for legacy\n> >> binary-only applications that were built as soft-float way back in the\n> >> (not so) good old days...\n> >>\n> >> In which case, we should instead depend on !BR2_ARCH_IS_64.\n> >>\n> >> Thoughts?\n> > \n> > I'm hesitating on this one. I believe the use-case you mention makes\n> > sense, however I believe that it can still be supported with your\n> > proposed patch.\n> > \n> > Indeed, even if you want to run a binary-only soft-float application,\n> > it doesn't prevent the rest of your system from using VFP, as long as\n> > you use EABI (and not EABIhf).\n> > \n> > The whole reason why EABI passes floating-point arguments into integer\n> > registers is precisely to allow compatibility between soft-float code\n> > and hard-float code. Thanks to that, a soft-float function can call a\n> > hard-float function, and vice-versa.\n> > \n> > Therefore, even if you have a binary-only soft-float application, there\n> > is really no reason to build the entire system soft-float.\n> > \n> > So, I think your original patch is OK as-is. However, I still see one\n> > inconsistent thing: if we disallow soft-float because ARMv8-A mandates\n> > VFPv4, why would we allow soft-float for ARMv7-A cores that have a\n> > VFP ? To be consistent, we should also disallow soft-float.\n> > \n> > Another way to think about this is: is it possible to build an AArch64\n> > soft-float system? do we want to allow the user to select what is\n> > *possible* or what makes sense?\n> \n> gcc aarch64\n\nIt's not about aarch64, it's about aarch32, i.e. an armv8a core running\nin 32-bit mode.\n\n> doesn't have a float-abi option, so no it's not possible to build\n> with soft-float. So the depends on BR2_ARM_EABI is certainly needed.\n> \n> That said, I don't think it makes a lot of sense to add extra 'depends on'\n> lines just to avoid that the user would select an option that probably doesn't\n> make sense. We already make sure that the default is appropriate. Let's not make\n> our life more difficult.\n\nIt's not about making *our* lives more dificult or easier. It's about\nmaking the user's lives easier.\n\nIf there is a prompt that offers an option, but that has no impact on\nthe resulting output, then that prompt should not exist. Otherwise, it\nis very confusing.\n\nRegards,\nYann E. MORIN.\n\n> So I'd say: reject this patch.\n> \n> Regards,\n> Arnout\n> \n> \n> -- \n> Arnout Vandecappelle arnout at mind be\n> Senior Embedded Software Architect +32-16-286500\n> Essensium/Mind http://www.mind.be\n> G.Geenslaan 9, 3001 Leuven, Belgium BE 872 984 063 RPR Leuven\n> LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle\n> GPG fingerprint: 7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF","headers":{"Return-Path":"","X-Original-To":["incoming@patchwork.ozlabs.org","buildroot@lists.busybox.net"],"Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","buildroot@osuosl.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=busybox.net\n\t(client-ip=140.211.166.136; helo=silver.osuosl.org;\n\tenvelope-from=buildroot-bounces@busybox.net;\n\treceiver=)","ozlabs.org;\n\tdkim=fail reason=\"signature verification failed\" (2048-bit key;\n\tunprotected) header.d=gmail.com header.i=@gmail.com\n\theader.b=\"DPG5caEv\"; dkim-atps=neutral"],"Received":["from silver.osuosl.org (smtp3.osuosl.org [140.211.166.136])\n\t(using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yl73J34wRz9s4s\n\tfor ;\n\tSun, 26 Nov 2017 22:28:27 +1100 (AEDT)","from localhost (localhost [127.0.0.1])\n\tby silver.osuosl.org (Postfix) with ESMTP id 2EC352FE5E;\n\tSun, 26 Nov 2017 11:28:24 +0000 (UTC)","from silver.osuosl.org ([127.0.0.1])\n\tby localhost (.osuosl.org [127.0.0.1]) (amavisd-new, port 10024)\n\twith ESMTP id 3WLVV+KWVm3X; Sun, 26 Nov 2017 11:28:22 +0000 (UTC)","from ash.osuosl.org (ash.osuosl.org [140.211.166.34])\n\tby silver.osuosl.org (Postfix) with ESMTP id 48E142FE32;\n\tSun, 26 Nov 2017 11:28:22 +0000 (UTC)","from whitealder.osuosl.org (smtp1.osuosl.org [140.211.166.138])\n\tby ash.osuosl.org (Postfix) with ESMTP id 9054E1C2018\n\tfor ;\n\tSun, 26 Nov 2017 11:13:04 +0000 (UTC)","from localhost (localhost [127.0.0.1])\n\tby whitealder.osuosl.org (Postfix) with ESMTP id 8A863860D9\n\tfor ;\n\tSun, 26 Nov 2017 11:13:04 +0000 (UTC)","from whitealder.osuosl.org ([127.0.0.1])\n\tby localhost (.osuosl.org [127.0.0.1]) (amavisd-new, port 10024)\n\twith ESMTP id TN74buuiCJYV for ;\n\tSun, 26 Nov 2017 11:13:03 +0000 (UTC)","from mail-wr0-f170.google.com (mail-wr0-f170.google.com\n\t[209.85.128.170])\n\tby whitealder.osuosl.org (Postfix) with ESMTPS id 3780D8607A\n\tfor ; Sun, 26 Nov 2017 11:13:03 +0000 (UTC)","by mail-wr0-f170.google.com with SMTP id y42so23839062wrd.3\n\tfor ; Sun, 26 Nov 2017 03:13:03 -0800 (PST)","from ymorin.is-a-geek.org\n\t(2a01cb0886107300dc07999d9853bcf8.ipv6.abo.wanadoo.fr.\n\t[2a01:cb08:8610:7300:dc07:999d:9853:bcf8])\n\tby smtp.gmail.com with ESMTPSA id\n\t128sm13511512wmi.28.2017.11.26.03.12.58\n\t(version=TLS1_2 cipher=ECDHE-RSA-CHACHA20-POLY1305 bits=256/256);\n\tSun, 26 Nov 2017 03:12:59 -0800 (PST)","by ymorin.is-a-geek.org (sSMTP sendmail emulation);\n\tSun, 26 Nov 2017 12:12:57 +0100"],"X-Virus-Scanned":["amavisd-new at osuosl.org","amavisd-new at osuosl.org"],"X-Greylist":"domain auto-whitelisted by SQLgrey-1.7.6","DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025;\n\th=sender:date:from:to:cc:subject:message-id:references:mime-version\n\t:content-disposition:in-reply-to:user-agent;\n\tbh=kwLbprjgNNrnm1dkZMhyLOj0hJ72/+5/Cw3CxAgKv3A=;\n\tb=DPG5caEvxgYkwk+Kgr/fVZ2fayDUFQR3WtU4i+SvKC8ISUhloumQ1Jdy4aLhrLpwED\n\tKoUpgpSIfCO6betLkHSoqRohlHjMvrFCkcHFhsa21mcrPSE6enzdonyBUYKHLrRd5Eg7\n\tomE28iWQyPMuwqLYlbrYNEt+0E6AyMPZ2cL8LtKrTTGTMYzbtMixCiq9ub3VbJpziLuT\n\tt8y+Kq1Z3kzJ/iqwAQIlf7OwgfPmzB/DzZLawxAIrFDiwykzeFmqbdwzW+MUWFTYvt1h\n\tYlSatp6qKXbVSsZjYNyFmlVUDuRUMLIhvaIlNt8LERy5yH1GcJVJD0EaRUyXNa33tAtE\n\t25/Q==","X-Google-DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed;\n\td=1e100.net; s=20161025;\n\th=x-gm-message-state:sender:date:from:to:cc:subject:message-id\n\t:references:mime-version:content-disposition:in-reply-to:user-agent; \n\tbh=kwLbprjgNNrnm1dkZMhyLOj0hJ72/+5/Cw3CxAgKv3A=;\n\tb=pJMy5m7WY0iXVqAuA1bXvkHf8YRMowKByclKmS27kxOommey+YDF0DYcTaLDPJPnu/\n\t73o5gkjkrI+/JUQp7Is0YzFdj3ufKowEFWno3hIaKj7QZSp2/EW9X1qdeUxy4d93mNNl\n\tNrAyz7FGtSjw32ZJTX9zgK143zHde1fl8QlvyArEi6joMhmUODpPG48+i2SRP5w8TwLI\n\taxGoYsCeD+ekiTiIX1fiAKxY5c3iM2yjErlUMoy+6A00v0Xrw9dikN48Z0P/XqNbrosf\n\t7aOAly8Q9t6ysSnHn0hNyhPF09sRs00WOyc7HE272ju6C6KwIMXn8VLUI7mZIF4wQAfb\n\tA3vQ==","X-Gm-Message-State":"AJaThX7+qbvEk5skTKQfGbjrLKQxYIttylICicC1KRuWDGOQ9EKewFKQ\n\t3kV6mIymzrzcoNznmgNOjlQ=","X-Google-Smtp-Source":"AGs4zMYIYM6UKCp87g00j7LpIdkhia1TNzQ/h/a55VvSSbYbcS9OeQDDoYRjZRE+U74QprZ1yNJkXw==","X-Received":"by 10.223.186.202 with SMTP id w10mr6718166wrg.187.1511694781526;\n\tSun, 26 Nov 2017 03:13:01 -0800 (PST)","Date":"Sun, 26 Nov 2017 12:12:57 +0100","From":"\"Yann E. MORIN\" ","To":"Arnout Vandecappelle ","Message-ID":"<20171126111257.GA3101@scaer>","References":"\n\t\n\t<20170904172137.GB5223@scaer>\n\t<20171124230857.0b6e1648@windsurf.lan>\n\t<5508245a-3f6a-80b9-a7ad-b3b6bc5d5f9e@mind.be>","MIME-Version":"1.0","Content-Disposition":"inline","In-Reply-To":"<5508245a-3f6a-80b9-a7ad-b3b6bc5d5f9e@mind.be>","User-Agent":"Mutt/1.5.22 (2013-10-16)","Subject":"Re: [Buildroot] [PATCH 4/9] arch/arm: do not allow soft-float for\n\tarmv8a","X-BeenThere":"buildroot@busybox.net","X-Mailman-Version":"2.1.24","Precedence":"list","List-Id":"Discussion and development of buildroot ","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Cc":"Thomas Petazzoni ,\n\tbuildroot@buildroot.org","Content-Type":"text/plain; charset=\"us-ascii\"","Content-Transfer-Encoding":"7bit","Errors-To":"buildroot-bounces@busybox.net","Sender":"\"buildroot\" "}},{"id":1810158,"web_url":"http://patchwork.ozlabs.org/comment/1810158/","msgid":"<20171126111846.GB3101@scaer>","list_archive_url":null,"date":"2017-11-26T11:18:46","subject":"Re: [Buildroot] [PATCH 4/9] arch/arm: do not allow soft-float for\n\tarmv8a","submitter":{"id":13903,"url":"http://patchwork.ozlabs.org/api/people/13903/","name":"Yann E. MORIN","email":"yann.morin.1998@free.fr"},"content":"Arnout, Thomas, All,\n\nOn 2017-11-26 12:12 +0100, Yann E. MORIN spake thusly:\n> On 2017-11-25 18:10 +0100, Arnout Vandecappelle spake thusly:\n> > On 24-11-17 23:08, Thomas Petazzoni wrote:\n> > > On Mon, 4 Sep 2017 19:21:37 +0200, Yann E. MORIN wrote:\n> > >> Thomas, All,\n> > >>\n> > >> On 2017-09-03 15:17 +0200, Yann E. MORIN spake thusly:\n> > >>> armv8a has made VFPv4 and NEON mandatory, so there is no point in\n> > >>> allowing software floating point, even in 32-bit mode. \n> > >>\n> > >> In fact, even though I'm pretty sure that OK in 64-bit, we might still\n> > >> have a reason to do soft-float in 32-bit mode, if only for legacy\n> > >> binary-only applications that were built as soft-float way back in the\n> > >> (not so) good old days...\n> > >>\n> > >> In which case, we should instead depend on !BR2_ARCH_IS_64.\n> > >>\n> > >> Thoughts?\n> > > \n> > > I'm hesitating on this one. I believe the use-case you mention makes\n> > > sense, however I believe that it can still be supported with your\n> > > proposed patch.\n> > > \n> > > Indeed, even if you want to run a binary-only soft-float application,\n> > > it doesn't prevent the rest of your system from using VFP, as long as\n> > > you use EABI (and not EABIhf).\n> > > \n> > > The whole reason why EABI passes floating-point arguments into integer\n> > > registers is precisely to allow compatibility between soft-float code\n> > > and hard-float code. Thanks to that, a soft-float function can call a\n> > > hard-float function, and vice-versa.\n> > > \n> > > Therefore, even if you have a binary-only soft-float application, there\n> > > is really no reason to build the entire system soft-float.\n> > > \n> > > So, I think your original patch is OK as-is. However, I still see one\n> > > inconsistent thing: if we disallow soft-float because ARMv8-A mandates\n> > > VFPv4, why would we allow soft-float for ARMv7-A cores that have a\n> > > VFP ? To be consistent, we should also disallow soft-float.\n> > > \n> > > Another way to think about this is: is it possible to build an AArch64\n> > > soft-float system? do we want to allow the user to select what is\n> > > *possible* or what makes sense?\n> > \n> > gcc aarch64\n> \n> It's not about aarch64, it's about aarch32, i.e. an armv8a core running\n> in 32-bit mode.\n> \n> > doesn't have a float-abi option, so no it's not possible to build\n> > with soft-float. So the depends on BR2_ARM_EABI is certainly needed.\n> > \n> > That said, I don't think it makes a lot of sense to add extra 'depends on'\n> > lines just to avoid that the user would select an option that probably doesn't\n> > make sense. We already make sure that the default is appropriate. Let's not make\n> > our life more difficult.\n> \n> It's not about making *our* lives more dificult or easier. It's about\n> making the user's lives easier.\n> \n> If there is a prompt that offers an option, but that has no impact on\n> the resulting output, then that prompt should not exist. Otherwise, it\n> is very confusing.\n\nDamned, I hadn't finished my mail...\n\n> > So I'd say: reject this patch.\n\nBut seeing the argument from Thomas, yes it does not make sense to hide\nsoft-float anyway.\n\nRegards,\nYann E. MORIN.","headers":{"Return-Path":"","X-Original-To":["incoming@patchwork.ozlabs.org","buildroot@lists.busybox.net"],"Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","buildroot@osuosl.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=busybox.net\n\t(client-ip=140.211.166.133; helo=hemlock.osuosl.org;\n\tenvelope-from=buildroot-bounces@busybox.net;\n\treceiver=)","ozlabs.org;\n\tdkim=fail reason=\"signature verification failed\" (2048-bit key;\n\tunprotected) header.d=gmail.com header.i=@gmail.com\n\theader.b=\"LjOHBJgL\"; dkim-atps=neutral"],"Received":["from hemlock.osuosl.org (smtp2.osuosl.org [140.211.166.133])\n\t(using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yl7Lk30scz9s4s\n\tfor ;\n\tSun, 26 Nov 2017 22:41:50 +1100 (AEDT)","from localhost (localhost [127.0.0.1])\n\tby hemlock.osuosl.org (Postfix) with ESMTP id 43B738767C;\n\tSun, 26 Nov 2017 11:41:45 +0000 (UTC)","from hemlock.osuosl.org ([127.0.0.1])\n\tby localhost (.osuosl.org [127.0.0.1]) (amavisd-new, port 10024)\n\twith ESMTP id lxv6CJ4I4ncn; Sun, 26 Nov 2017 11:41:39 +0000 (UTC)","from ash.osuosl.org (ash.osuosl.org [140.211.166.34])\n\tby hemlock.osuosl.org (Postfix) with ESMTP id 9F2358762E;\n\tSun, 26 Nov 2017 11:41:39 +0000 (UTC)","from silver.osuosl.org (smtp3.osuosl.org [140.211.166.136])\n\tby ash.osuosl.org (Postfix) with ESMTP id A02651C2018\n\tfor ;\n\tSun, 26 Nov 2017 11:18:52 +0000 (UTC)","from localhost (localhost [127.0.0.1])\n\tby silver.osuosl.org (Postfix) with ESMTP id 9B4D62EFB6\n\tfor ;\n\tSun, 26 Nov 2017 11:18:52 +0000 (UTC)","from silver.osuosl.org ([127.0.0.1])\n\tby localhost (.osuosl.org [127.0.0.1]) (amavisd-new, port 10024)\n\twith ESMTP id I1LwQBB9pGL8 for ;\n\tSun, 26 Nov 2017 11:18:51 +0000 (UTC)","from mail-wm0-f46.google.com (mail-wm0-f46.google.com\n\t[74.125.82.46])\n\tby silver.osuosl.org (Postfix) with ESMTPS id 77EE12014B\n\tfor ; Sun, 26 Nov 2017 11:18:51 +0000 (UTC)","by mail-wm0-f46.google.com with SMTP id u83so29214325wmb.5\n\tfor ; Sun, 26 Nov 2017 03:18:51 -0800 (PST)","from ymorin.is-a-geek.org\n\t(2a01cb0886107300fdb3c96b998167cf.ipv6.abo.wanadoo.fr.\n\t[2a01:cb08:8610:7300:fdb3:c96b:9981:67cf])\n\tby smtp.gmail.com with ESMTPSA id\n\tq140sm6152276wmd.35.2017.11.26.03.18.47\n\t(version=TLS1_2 cipher=ECDHE-RSA-CHACHA20-POLY1305 bits=256/256);\n\tSun, 26 Nov 2017 03:18:48 -0800 (PST)","by ymorin.is-a-geek.org (sSMTP sendmail emulation);\n\tSun, 26 Nov 2017 12:18:46 +0100"],"X-Virus-Scanned":["amavisd-new at osuosl.org","amavisd-new at osuosl.org"],"X-Greylist":"domain auto-whitelisted by SQLgrey-1.7.6","DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025;\n\th=sender:date:from:to:cc:subject:message-id:references:mime-version\n\t:content-disposition:in-reply-to:user-agent;\n\tbh=0Gq3EeBRr4FhuLIXpgMZVsqDYp1BQJsDawYCNSYp1xA=;\n\tb=LjOHBJgLnh8AmJicxFGqYOU7T45cxOs+ZYRtSqKQnK7HK5HRTl4N0QGptPjjKbJp8N\n\tVGUCxC05Zlcl0c9rUFOqmpzPvJQHAWMZcSnVMkiBgASOM2mjsqClmRB9J/9mhaGx6G66\n\tcf0XEf/r72bcs5Nx6Mm2KhlNHkDozkQZLwVvI6pWQmvoluK0BDUcyutsmpwe1xWmVrTd\n\tzcDxxujIiRLT/ir6TVw/G05HZ6d70x6pMwKs7Db/WHIpNApRTUR6JMqHVw5rJOaJc49Q\n\tvhhNTc1c6vJ4tP01RCf2Xjgsb3eO1BzN+0Vg/GYJlXx0SDSZwoMhNtv40hlHeWh0ZUYO\n\tWMzQ==","X-Google-DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed;\n\td=1e100.net; s=20161025;\n\th=x-gm-message-state:sender:date:from:to:cc:subject:message-id\n\t:references:mime-version:content-disposition:in-reply-to:user-agent; \n\tbh=0Gq3EeBRr4FhuLIXpgMZVsqDYp1BQJsDawYCNSYp1xA=;\n\tb=k4daYd9Pl1RYG2sbsqxp4S2DWI2/5npSmewmlR2qLufKz/PWdKXL/IwkJnyhthsnO3\n\tA189LO8q2lL1/xYND2bHyKZuQTIiqYWykWMlIa4IE5mop2y7AWd3f5GdNe6wKENmxQpG\n\taU30JWxgh1h03FzPmS5RAF6tInM+bWBfRLc2+ZtVgwLCBEJqJnHIq4oInXDQmHQU7rXh\n\tTT1Wsu0HBMe4FHdYL4IyE4D0RCSgEqm2Q2agGJrB/w55MDfXY6FeK0ABgtKOMBa5xnCI\n\tbA6ekZ3QZnOH7SbjE2n0rBnl9LFRHvyXTWVxOi4YbXsm8vYP3aHM2Y8L7H1gw2Q2b+m0\n\t/qHQ==","X-Gm-Message-State":"AJaThX4477YnZIV82bgPL2JO+RtcFRrJPTDa/5LEdgxVGa0y8OuZOTLm\n\tuRQbKVH8goEOrMEW/gh4obRsYw==","X-Google-Smtp-Source":"AGs4zMbbd72aor0NbtcKOE+J5DjSDJG9rLhs+HiifCLEpM/LcLeyYkUZAas2zzIKXI7jcoMWrg0rdA==","X-Received":"by 10.28.56.67 with SMTP id f64mr14748433wma.35.1511695129742;\n\tSun, 26 Nov 2017 03:18:49 -0800 (PST)","Date":"Sun, 26 Nov 2017 12:18:46 +0100","From":"\"Yann E. MORIN\" ","To":"Arnout Vandecappelle ","Message-ID":"<20171126111846.GB3101@scaer>","References":"\n\t\n\t<20170904172137.GB5223@scaer>\n\t<20171124230857.0b6e1648@windsurf.lan>\n\t<5508245a-3f6a-80b9-a7ad-b3b6bc5d5f9e@mind.be>\n\t<20171126111257.GA3101@scaer>","MIME-Version":"1.0","Content-Disposition":"inline","In-Reply-To":"<20171126111257.GA3101@scaer>","User-Agent":"Mutt/1.5.22 (2013-10-16)","Subject":"Re: [Buildroot] [PATCH 4/9] arch/arm: do not allow soft-float for\n\tarmv8a","X-BeenThere":"buildroot@busybox.net","X-Mailman-Version":"2.1.24","Precedence":"list","List-Id":"Discussion and development of buildroot ","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Cc":"Thomas Petazzoni ,\n\tbuildroot@buildroot.org","Content-Type":"text/plain; charset=\"us-ascii\"","Content-Transfer-Encoding":"7bit","Errors-To":"buildroot-bounces@busybox.net","Sender":"\"buildroot\" "}}]