From owner-freebsd-current@FreeBSD.ORG Tue Mar 8 05:16:00 2005 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 166F016A4CE for ; Tue, 8 Mar 2005 05:16:00 +0000 (GMT) Received: from mail.mcneil.com (mcneil.com [24.199.45.54]) by mx1.FreeBSD.org (Postfix) with ESMTP id DFA9443D39 for ; Tue, 8 Mar 2005 05:15:59 +0000 (GMT) (envelope-from sean@mcneil.com) Received: from localhost (localhost.mcneil.com [127.0.0.1]) by mail.mcneil.com (Postfix) with ESMTP id 92C2CF3CD5 for ; Mon, 7 Mar 2005 21:15:57 -0800 (PST) Received: from mail.mcneil.com ([127.0.0.1]) by localhost (server.mcneil.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 03785-02 for ; Mon, 7 Mar 2005 21:15:57 -0800 (PST) Received: from mcneil.com (mcneil.com [24.199.45.54]) by mail.mcneil.com (Postfix) with ESMTP id 1779AF3CD2 for ; Mon, 7 Mar 2005 21:15:57 -0800 (PST) From: Sean McNeil To: current@freebsd.org Content-Type: text/plain Date: Mon, 07 Mar 2005 21:15:56 -0800 Message-Id: <1110258956.3660.15.camel@server.mcneil.com> Mime-Version: 1.0 X-Mailer: Evolution 2.0.4 FreeBSD GNOME Team Port Content-Transfer-Encoding: 7bit X-Virus-Scanned: by amavisd-new at mcneil.com Subject: amd64 floating point context save/restore changed? X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 08 Mar 2005 05:16:00 -0000 Has there been any changes within the last couple of days that could affect context save/restore of the FPU on amd64? I am getting consistent SIGFPE signals in an application I built. The problem is, if I just rearrange the code a little bit I can get it to move. Since it is consistent for any particular version, I would guess it to be related to pthreads if it is context save/restore. In both cases I've looked at there was no reason the SIGPFE should have occurred and, in fact, does not when the code is reordered. I can flood the list with a lot of backtrace info, but prefer to send it to whomever is working this area to save others. Sean