From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263527AbTJaTQx (ORCPT ); Fri, 31 Oct 2003 14:16:53 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263528AbTJaTQw (ORCPT ); Fri, 31 Oct 2003 14:16:52 -0500 Received: from enterprise.bidmc.harvard.edu ([134.174.118.50]:32522 "EHLO enterprise.bidmc.harvard.edu") by vger.kernel.org with ESMTP id S263527AbTJaTQv (ORCPT ); Fri, 31 Oct 2003 14:16:51 -0500 Message-ID: <3FA2B4F4.8040404@enterprise.bidmc.harvard.edu> Date: Fri, 31 Oct 2003 14:16:04 -0500 From: "Kristofer T. Karas" User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030915 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Benjamin Herrenschmidt CC: Kronos , Linux Kernel Mailing List , Javier Villavicencio Subject: Re: RadeonFB [Re: 2.4.23pre8 - ACPI Kernel Panic on boot] References: <20031029210321.GA11437@dreamland.darkstar.lan> <1067491238.1735.4.camel@ktkhome> <1067587196.715.1.camel@gaston> In-Reply-To: <1067587196.715.1.camel@gaston> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Benjamin Herrenschmidt wrote: >Ok, first thing: XFree "radeon" _is_ accelerated, though it doesn't >do 3D on recent cards > Hi Ben - Right, sorry, I mean to apply "accelerated" to the 3D effects; the private ATI driver is much faster on glxgears than the version that bundles with XFree. >Then, the problem you are having is well known > I'm having two, actually. The first is that YPAN is getting quite confused. If I switch VCs, then switch back, the text has been re-arranged, the cursor is often invisible, and sometimes a page or two of new text must be written before anything starts to show up on the screen again. Experimenting shows that setting VYRES = YRES works around this problem. The second problem is of course the register contents issue when returning from certain graphics programs (e.g. X+fglr) to text mode.. I rather like your idea of doing a re-init when switching from KD_GRAPHICS to KD_TEXT, as the monitor blank during resolution switch will likely overshadow the re-init process. >Can you verify that running fbset -accel 0 then back 1 cures the >problem for you ? > > At work now, will try when I return home later... Kris