From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965325AbWGJXcq (ORCPT ); Mon, 10 Jul 2006 19:32:46 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S965289AbWGJXcq (ORCPT ); Mon, 10 Jul 2006 19:32:46 -0400 Received: from gprs189-60.eurotel.cz ([160.218.189.60]:17344 "EHLO amd.ucw.cz") by vger.kernel.org with ESMTP id S965325AbWGJXco (ORCPT ); Mon, 10 Jul 2006 19:32:44 -0400 Date: Tue, 11 Jul 2006 00:36:29 +0200 From: Pavel Machek To: Brad Campbell Cc: suspend2-devel@lists.suspend2.net, lkml Subject: Re: /dev/rtc not suspending/resuming properly Message-ID: <20060710223629.GA7443@elf.ucw.cz> References: <44B24CEB.9010103@wasp.net.au> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <44B24CEB.9010103@wasp.net.au> X-Warning: Reading this can be dangerous to your mental health. User-Agent: Mutt/1.5.11+cvs20060126 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi! > In trying to track down a bug in an application, I think I may have > stumbled across one in the kernel. Well, that happens in suspend/resume area... > It would appear that the rtc device is not restoring the state of the PIE > control bit when the machine goes to sleep and then wakes up again, > resulting in the rtc_dropped_irq() being called repeatedly until the > application accessing /dev/rtc is stopped. > > I'm using a vanilla 2.6.17.3 kernel with the suspend2 patch only on x86 > > I've had a pretty good look at drivers/char/rtc.c and I can't see anywhere > it would actually suspend/resume in the code, and investigation shows it > does not appear to re-init the hardware on resume. Well, you probably need to write suspend/resume support for it... Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html