From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753061Ab0JJOJO (ORCPT ); Sun, 10 Oct 2010 10:09:14 -0400 Received: from one.firstfloor.org ([213.235.205.2]:35890 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752447Ab0JJOJM (ORCPT ); Sun, 10 Oct 2010 10:09:12 -0400 Date: Sun, 10 Oct 2010 16:09:08 +0200 From: Andi Kleen To: Markus Trippelsdorf Cc: Andi Kleen , airlied@linux.ie, dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org Subject: Re: REGRESSION 2.6.36: radeon KMS console switch broken Message-ID: <20101010140908.GA26728@basil.fritz.box> References: <20101010123000.GD8256@basil.fritz.box> <20101010134429.GA1581@arch.trippelsdorf.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20101010134429.GA1581@arch.trippelsdorf.de> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sun, Oct 10, 2010 at 03:44:29PM +0200, Markus Trippelsdorf wrote: > On Sun, Oct 10, 2010 at 02:30:00PM +0200, Andi Kleen wrote: > > > > Hi, > > > > I see a radeon regression here with 2.6.36-rc7. > > > > I updated a system from 2.6.35 to 2.6.36-rc7. Now I cannot > > switch to the text console from the X server anymore; the screen just > > keeps displaying the X desktop. Eventually when I switch back > > to X things work again. > > > > The console switching worked in 2.6.35 > > (I had occasional GPU hangs in .35, but console switch always worked) > > I didn't change the Xserver userland. > > > > The system has a ATI Technologies Inc RV710/730 running Dual Head. > > You're using the radeonhd driver, which is not supported and outdated. > Try to remove /usr/lib64/xorg/modules/drivers/radeonhd_drv.so and start > X again... No you misread the Xorg.0.log: [ 50.161] (II) Module radeonhd: vendor="AMD GPG" ... [ 51.636] (EE) FATAL: RadeonHD presently does not work with kernel modesetting (KMS). ... [ 50.167] (II) LoadModule: "radeon" .... [ 52.214] (II) UnloadModule: "radeonhd" ... It's using radeon. Besides if radeonhd stopped working that would be a regression too, but that's not what we have here. -Andi -- ak@linux.intel.com -- Speaking for myself only.