From owner-freebsd-ia64@FreeBSD.ORG Fri Jan 28 12:24:22 2005 Return-Path: Delivered-To: freebsd-ia64@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7F13516A4CE for ; Fri, 28 Jan 2005 12:24:22 +0000 (GMT) Received: from iramx3.ira.uni-karlsruhe.de (iramx3.ira.uni-karlsruhe.de [141.3.10.82]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7B47443D2D for ; Fri, 28 Jan 2005 12:24:21 +0000 (GMT) (envelope-from skoglund@ira.uka.de) Received: from irams1.ira.uni-karlsruhe.de ([141.3.10.5] helo=irams1.ira.uka.de) by iramx3.ira.uni-karlsruhe.de with esmtps (Exim 4.43 #1) id 1CuVAb-00060Y-FZ; Fri, 28 Jan 2005 13:24:20 +0100 Received: from i30pc5.ibds.uni-karlsruhe.de ([141.3.40.155] helo=i30pc5.ira.uka.de) by irams1.ira.uka.de with esmtp (Exim 3.30 #7 ) id 1CuVAX-0000yO-00; Fri, 28 Jan 2005 13:24:09 +0100 Received: from skoglund by i30pc5.ira.uka.de with local (Exim 4.44) id 1CuVAX-0004gg-FD; Fri, 28 Jan 2005 13:24:09 +0100 From: Espen Skoglund MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <16890.11995.972474.700748@i30nb20.ira.uka.de> Date: Fri, 28 Jan 2005 13:23:55 +0100 To: Christian Hildner In-Reply-To: <41FA27AD.4030208@hob.de> References: <41FA27AD.4030208@hob.de> X-Mailer: VM 7.17 under 21.4 (patch 14) "Reasonable Discussion" XEmacs Lucid Sender: Espen Skoglund X-Spam-Score: -5.9 (-----) X-Spam-Status: No cc: freebsd-ia64@freebsd.org Subject: Re: TLB miss handlers X-BeenThere: freebsd-ia64@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Porting FreeBSD to the IA-64 List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Jan 2005 12:24:22 -0000 [Christian Hildner] > Hi, > ld8 r21=[r21] // check VHPT tag > ^^^^^^^^^^ > Are you sure you cannot enter a nested TLB miss fault here on IA64 > implementations that do not have the VHPT walker implemented? You > wouldn't get fault 0x0000 since there is no walker that throws that > fault. Er... I can't remember seeing anywhere that implementing the VHPT walker in the CPU is optional. Anyhow, I'd imagine that the VHPT is located in a place that is backed by a pinned TR. But I'm not really familiar with FreeBSD internals, so... eSk