From owner-freebsd-arm@freebsd.org Thu Mar 16 10:41:18 2017 Return-Path: Delivered-To: freebsd-arm@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id C9F61D0EE1C for ; Thu, 16 Mar 2017 10:41:18 +0000 (UTC) (envelope-from sylvain@sylvaingarrigues.com) Received: from mail-wr0-x241.google.com (mail-wr0-x241.google.com [IPv6:2a00:1450:400c:c0c::241]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority G2" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 608961D4D for ; Thu, 16 Mar 2017 10:41:17 +0000 (UTC) (envelope-from sylvain@sylvaingarrigues.com) Received: by mail-wr0-x241.google.com with SMTP id g10so5347643wrg.0 for ; Thu, 16 Mar 2017 03:41:17 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sylvaingarrigues-com.20150623.gappssmtp.com; s=20150623; h=from:mime-version:subject:date:references:to:in-reply-to:message-id; bh=OhQNUGc/w88dqtVNJ0LUsWDHJl280wWI/swYwvJpZOQ=; b=KnjPDzJIg3GpIZaMly1zThgGTY6yI2CX3ohmX3gFHPpBNYApvPBppA4UxdAaBZTPLZ 8XkmuwZynjRU7eHBrQfCTkbmm7Ive9LtG9GHzmyzZqsbAo6XLSRA1ayvD5zHTQkjRqty 38CkLcYJKPAY2fZ5+fnf9Q4XlEIwKaKZzUkFZ/2FN2ogBA5+1hzAk9/vJXBtnLxX9MJt BvP0WedNkcVYSRftpeZZyFDrTnManFf7Pxx4DyE5o4czAAcd+7b69cijvid/MFFOV1+u LcMfJ3Xu16HKXaVcpvKAJVTByQH7xIA7QdoqM5Y92mhPiwM23GTp4bY7T1tEYdHuFy8V Y42Q== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:mime-version:subject:date:references:to :in-reply-to:message-id; bh=OhQNUGc/w88dqtVNJ0LUsWDHJl280wWI/swYwvJpZOQ=; b=FH+ZmbXvGk9gW3gLDYkyYrQdUW+q/6g4DZXHQM+S1touK82akzcPREy9KTohxXmnhb rPVZE524riHGvyskmU4hjy01+P4i8UyUTAiWwRsu70obvM+l2kmmkdFWVNGoa/XC0mKN m8O1T9ObiVrrXJFAD1aCDqAowvRGl3xz4jZNk0z5o7MfEyux0b5l5EYESV008bHXPslz /2osrBY6MLr5qMILUia8LO8Ibmzuun+d08In4JbFG8zVzw62DzhEQNh+kjPvbuca2OZJ Ucu/LLKqnNHDcpf4sEUkkz98obioD4QNVDGmgaQ/U6VlML646owUgM5mImSysLk3VR8A tKTA== X-Gm-Message-State: AFeK/H2U1VtDBA8HCaBr4C1bqZXgKBNya9zrDpwBEXYymcsmkd6tezhG0O/f7PZKdfqFSg== X-Received: by 10.223.172.101 with SMTP id v92mr8230085wrc.49.1489660875992; Thu, 16 Mar 2017 03:41:15 -0700 (PDT) Received: from [172.21.11.111] ([158.255.96.238]) by smtp.gmail.com with ESMTPSA id y43sm5680574wrd.0.2017.03.16.03.41.15 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Thu, 16 Mar 2017 03:41:15 -0700 (PDT) From: Sylvain Garrigues Mime-Version: 1.0 (Mac OS X Mail 10.2 \(3259\)) Subject: Re: Is CPUTYPE=cortex-A7 supposed to work? Date: Thu, 16 Mar 2017 11:41:14 +0100 References: <871suc3nv8.fsf@news-spur.riddles.org.uk> <8737ely05c.fsf@news-spur.riddles.org.uk> <87wpbxw3yd.fsf@news-spur.riddles.org.uk> <79EBD44B-2C2D-4394-A90C-DF494A049F20@dsl-only.net> To: freebsd-arm In-Reply-To: <79EBD44B-2C2D-4394-A90C-DF494A049F20@dsl-only.net> Message-Id: X-Mailer: Apple Mail (2.3259) Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-Content-Filtered-By: Mailman/MimeDel 2.1.23 X-BeenThere: freebsd-arm@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: "Porting FreeBSD to ARM processors." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 16 Mar 2017 10:41:18 -0000 Hi, > Le 10 mars 2017 =C3=A0 18:51, Mark Millard a = =C3=A9crit : >=20 > So overall: >=20 > If the ABI stays as it is for 11 (or 12), floating point in > general (VFP with or without NEON) and Integer NEON can be > broken when signals are involved that allow the code to keep > going instead of exiting the process. (This means all NEON use > is subject to the issue.) Reading through this thread I understand one can definitely get into = trouble with a system which has CPUTYPE=3Dcortex-a7 in make.conf (or = even some -mcpu=3Dcortex-a7 in some CFLAGS variable there), especially = if a port does FP stuff in a signal handler (because FP registers are = not well preserved / restored on the kernel side). As Michal wrote: > Unfortunately, on armv6, the VFP part of kernel <-> userland = interaction > is broken from day 1.=20 This kind of frightens me, is there any plan to fix it soon (I=E2=80=99d = help if I could)?=20 I understand Andrew's tiny patch = (https://bugs.freebsd.org/bugzilla/attachment.cgi?id=3D180669&action=3Ddif= f = ) fixes VFP interactions during signal handling, would that fix all = the "VFP part of kernel <-> userland interaction =C2=BB? Is fixing VFP already being discussed in a Phabricator=E2=80=99s review = or are you guys discussing a backward-compatible alternative to = ABI-breaking (although we are not required to do so since we are still = tier-2) elsewhere? Needless to say, I see strong interest in being able to just dump = "CPUTYPE=3D" in make.conf on arm and be confident my = system will be fine and fine-tuned, just like I do for all my x86/amd64 = machines with such a flag. Have all a good day, Sylvain=