From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from zipcode.az.mvista.com (rav-az.mvista.com [65.200.49.157]) by ozlabs.org (Postfix) with ESMTP id 035132BDB5 for ; Sat, 11 Dec 2004 10:06:23 +1100 (EST) Message-ID: <41BA2F63.8060305@mvista.com> Date: Fri, 10 Dec 2004 16:21:07 -0700 From: Randy Vinson MIME-Version: 1.0 To: Kendall Bennett References: <41B9A33A.21673.5DDD7F64@localhost> In-Reply-To: <41B9A33A.21673.5DDD7F64@localhost> Content-Type: text/plain; charset=us-ascii; format=flowed Cc: linuxppc-embedded@ozlabs.org Subject: Re: Write Combining on PowerPC List-Id: Linux on Embedded PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Kendall Bennett wrote: [snip] > Does anyone know if it is possible to do something similar to Write > Combining for the PowerPC architecture, to speed up CPU access to the > linear framebuffer? Yes, it is possible. First you will need to enable Store Gathering for the Sandpoint's MPC107 bridge. (Go to Platform options->Enable MPC10x store gathering). You will then need to set the page protections for the video memory such that _PAGE_GUARDED is removed. The hardware will not store gather to an area mapped guarded. There appears to be some code to support this in the PCI MMap routines, but I'm not familiar with the details. Randy Vinson