From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 4 Oct 2001 16:42:31 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 4 Oct 2001 16:42:20 -0400 Received: from lightning.swansea.linux.org.uk ([194.168.151.1]:38927 "EHLO the-village.bc.nu") by vger.kernel.org with ESMTP id ; Thu, 4 Oct 2001 16:42:10 -0400 Subject: Re: ioremap() vs. ioremap_nocache() To: davidm@hpl.hp.com Date: Thu, 4 Oct 2001 21:47:52 +0100 (BST) Cc: linux-kernel@vger.kernel.org In-Reply-To: <200110040029.f940Tn103671@wailua.hpl.hp.com> from "David Mosberger" at Oct 03, 2001 05:29:49 PM X-Mailer: ELM [version 2.5 PL6] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-Id: From: Alan Cox Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > Now, as far as I know, on x86, ioremap() will give write-through > cached mappings (in the absence of mtrr games). If this is true, how On x86 ioremap will give mappings appropriate to the object you map - which means by default it wil give uncached mappings. The PCI hardware will do intelligent things in certain cases such as write merging Alan