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, 18 Jul 2002 23:56:09 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 18 Jul 2002 23:56:09 -0400 Received: from mta4.srv.hcvlny.cv.net ([167.206.5.10]:54688 "EHLO mta4.srv.hcvlny.cv.net") by vger.kernel.org with ESMTP id ; Thu, 18 Jul 2002 23:56:08 -0400 Date: Thu, 18 Jul 2002 23:57:28 -0400 From: "D. Sen" Subject: AGP and Thinkpad APM/suspend problems To: linux-kernel@vger.kernel.org Message-id: <3D378E28.9040305@auditorymodels.org> MIME-version: 1.0 Content-type: text/plain; charset=us-ascii; format=flowed Content-transfer-encoding: 7BIT X-Accept-Language: en-us, en User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.0) Gecko/20020529 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Reporting that both the IBM Thinkpad A31 and the Thinkpad T30 does not survive a suspend-resume cycle when the X server is configured to run at AGP >1x mode. The machines suspend fine. But an attempt to resume the machine, freezes it (requiring a cold reboot). Both machines have Radeon Mobility Chipsets. AGP speed is set using ==> "Option "AGPMode" "N" in /etc/X11/XF86Config-4 A note on XIG's website which might be relevant (seems to attribute the problem to APM): 3.0-28 05/07/2002 - corrected a problem that would cause multiple server invokations to hang on ATI chips. This was related to the APM fix made in the previous version. 3.0-27 05/06/2002 - corrected a problem with APM whereby the AGP bridge would not be shutdown/re-initialized properly during APM events. This could cause the server to lock in certain cases when the machine was 'woken up' after a suspend.