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-26 threads problem [win64] Date: Sat, 14 Oct 2017 09:39:42 +0300 Message-ID: <838tgejkr5.fsf@gnu.org> References: <83h8v7r02e.fsf@gnu.org> <83vajnotik.fsf@gnu.org> <83r2ubosnd.fsf@gnu.org> <83fuaqpzaq.fsf@gnu.org> <837ew2pxx9.fsf@gnu.org> <83efq6khh3.fsf@gnu.org> Reply-To: Eli Zaretskii NNTP-Posting-Host: blaine.gmane.org X-Trace: blaine.gmane.org 1507963252 29629 195.159.176.226 (14 Oct 2017 06:40:52 GMT) X-Complaints-To: usenet@blaine.gmane.org NNTP-Posting-Date: Sat, 14 Oct 2017 06:40:52 +0000 (UTC) Cc: fabrice.popineau@gmail.com, emacs-devel@gnu.org To: Richard Copley Original-X-From: emacs-devel-bounces+ged-emacs-devel=m.gmane.org@gnu.org Sat Oct 14 08:40:44 2017 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 1e3G7b-0004tj-JB for ged-emacs-devel@m.gmane.org; Sat, 14 Oct 2017 08:40:31 +0200 Original-Received: from localhost ([::1]:52976 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1e3G7f-00036V-Od for ged-emacs-devel@m.gmane.org; Sat, 14 Oct 2017 02:40:35 -0400 Original-Received: from eggs.gnu.org ([2001:4830:134:3::10]:48110) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1e3G6z-000364-AY for emacs-devel@gnu.org; Sat, 14 Oct 2017 02:39:54 -0400 Original-Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1e3G6u-0002jS-FI for emacs-devel@gnu.org; Sat, 14 Oct 2017 02:39:53 -0400 Original-Received: from fencepost.gnu.org ([2001:4830:134:3::e]:41002) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1e3G6u-0002jM-BB; Sat, 14 Oct 2017 02:39:48 -0400 Original-Received: from [176.228.60.248] (port=3146 helo=home-c4e4a596f7) by fencepost.gnu.org with esmtpsa (TLS1.2:RSA_AES_256_CBC_SHA1:256) (Exim 4.82) (envelope-from ) id 1e3G6t-0005X0-Qa; Sat, 14 Oct 2017 02:39:48 -0400 In-reply-to: (message from Richard Copley on Fri, 13 Oct 2017 23:00:56 +0100) 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:219478 Archived-At: > From: Richard Copley > Date: Fri, 13 Oct 2017 23:00:56 +0100 > Cc: Fabrice Popineau , Emacs Development > > I think GCC's assumption that 0x40(%rbx) is 16-byte aligned derives > through some arithmetic from the assumption that the stack was 16-byte > aligned on entry. Can you try adding > __attribute__ ((force_align_arg_pointer)) > to the thread function? Thanks, we already tried that. It didn't work, because the offending address doesn't come from stack, it comes from the heap. (Nevertheless, the thread function should have this attribute on Windows for the 32-bit builds to have the correct stack alignment. I will do that soon. But Fabrice's build is 64-bit, where that alignment is guaranteed by the ABI.)