From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752312AbcEWV1C (ORCPT ); Mon, 23 May 2016 17:27:02 -0400 Received: from bombadil.infradead.org ([198.137.202.9]:49552 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751154AbcEWV07 (ORCPT ); Mon, 23 May 2016 17:26:59 -0400 Date: Mon, 23 May 2016 14:26:55 -0700 From: Darren Hart To: Pali =?iso-8859-1?Q?Roh=E1r?= Cc: Gabriele Mazzotta , "Rafael J. Wysocki" , "D. Jared Dominguez" , "platform-driver-x86@vger.kernel.org" , Alex Hung , Andrei Borzenkov , "linux-kernel@vger.kernel.org" Subject: Re: [PATCH v3] dell-rbtn: Ignore ACPI notifications if device is suspended Message-ID: <20160523212655.GA2735@f23x64.localdomain> References: <1457740175-8327-1-git-send-email-gabriele.mzt@gmail.com> <20160328173309.GA26086@dvhart-mobl5.amr.corp.intel.com> <4072492.lANJWhSkYa@vostro.rjw.lan> <20160418123547.GK29406@pali> <20160519133032.GO29844@pali> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20160519133032.GO29844@pali> User-Agent: Mutt/1.5.24 (2015-08-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, May 19, 2016 at 03:30:32PM +0200, Pali Rohár wrote: > On Monday 25 April 2016 22:06:11 Gabriele Mazzotta wrote: > > 2016-04-18 14:35 GMT+02:00 Pali Rohár : > > > On Tuesday 29 March 2016 15:11:35 Rafael J. Wysocki wrote: > > >> On Monday, March 28, 2016 10:33:09 AM Darren Hart wrote: > > >> > On Thu, Mar 24, 2016 at 12:24:56PM +0100, Gabriele Mazzotta wrote: > > >> > > 2016-03-24 10:39 GMT+01:00 Pali Rohár : > > >> > > > On Monday 21 March 2016 16:13:34 Gabriele Mazzotta wrote: > > >> > > >> 2016-03-21 13:17 GMT+01:00 Pali Rohár : > > >> > > >> > On Friday 18 March 2016 23:44:23 Gabriele Mazzotta wrote: > > >> > > >> >> +#ifdef CONFIG_PM_SLEEP > > >> > > >> >> +static void ACPI_SYSTEM_XFACE rbtn_acpi_clear_flag(void *context) > > >> > > >> >> +{ > > >> > > >> >> + struct rbtn_data *rbtn_data = context; > > >> > > >> >> + > > >> > > >> >> + rbtn_data->suspended = false; > > >> > > >> >> +} > > >> > > >> >> + > > >> > > >> >> +static int rbtn_suspend(struct device *dev) > > >> > > >> >> +{ > > >> > > >> >> + struct acpi_device *device = to_acpi_device(dev); > > >> > > >> >> + struct rbtn_data *rbtn_data = acpi_driver_data(device); > > >> > > >> >> + > > >> > > >> >> + rbtn_data->suspended = true; > > >> > > >> >> + > > >> > > >> >> + return 0; > > >> > > >> >> +} > > >> > > >> >> + > > >> > > >> >> +static int rbtn_resume(struct device *dev) > > >> > > >> >> +{ > > >> > > >> >> + struct acpi_device *device = to_acpi_device(dev); > > >> > > >> >> + struct rbtn_data *rbtn_data = acpi_driver_data(device); > > >> > > >> >> + acpi_status status; > > >> > > >> >> + > > >> > > >> >> + /* > > >> > > >> >> + * Clear the flag only after we received the extra > > >> > > >> >> + * ACPI notification. > > >> > > >> >> + */ > > >> > > >> >> + status = acpi_os_execute(OSL_NOTIFY_HANDLER, > > >> > > >> >> + rbtn_acpi_clear_flag, rbtn_data); > > >> > > >> >> + if (ACPI_FAILURE(status)) > > >> > > >> >> + rbtn_data->suspended = false; > > >> > > >> > > > >> > > >> > I case when acpi_os_execute success it calls rbtn_acpi_clear_flag, > > >> > > >> > right? And that will set suspended to false. When acpi_os_execute fails, > > >> > > >> > then it set suspended too to false... Then whole acpi_os_execute doing > > >> > > >> > just "barrier" after which suspended flag can be set to false. So I > > >> > > >> > think rbtn_acpi_clear_flag function is not needed here. > > >> > > >> > > > >> > > >> > Cannot you pass NULL or empty function pointer as callback? Or what was > > >> > > >> > reason to do that flag clearing at "two places"? > > >> > > >> > > >> > > >> acpi_os_execute doesn't wait for the callback to be executed, so > > >> > > >> I can't clear the flag from rbtn_resume. > > >> > > > > > >> > > > acpi_os_execute calls callback asynchronously later? Or what exactly do it? > > >> > > > > >> > > In this case, it adds the callback to the kacpi_notify_wq workqueue > > >> > > for deferred execution. > > >> > > > >> > +Rafael for context/advice on the use of acpi_os_execute here. > > >> > > > >> > This is true, but a quick scan through that call path doesn't tell me why we > > >> > would need to call it here instead of just setting rbtn_data->suspended = false. > > >> > The comment suggests waiting for the event, but is that what this is doing? It > > >> > appears to me to be immediately scheduling the function to a work queue, not > > >> > waiting for the event notifier. > > >> > > >> I think this is supposed to work as a barrier. That is, it will only run after > > >> all events in the queue have been processed. > > >> > > >> I'm not sure if that's necessary, though. > > >> > > >> Thanks, > > >> Rafael > > >> > > > > > > Darren, Gabriele, what is state of this patch? Bug is not still fixed, > > > right? > > > > Yes, the bug is still there and this patch fixes it. > > > > Just to make it clear, we need the barrier. Andrei could reproduce > > the bug without it [1], but not with it, as he confirmed in this > > thread [2]. > > > > [1] http://article.gmane.org/gmane.linux.drivers.platform.x86.devel/8001 > > [2] http://article.gmane.org/gmane.linux.drivers.platform.x86.devel/8937 > > Ok, so it means that somebody (who understand ACPI) should review code > and accept it or show what is needed to fix. Plus maybe adds more > comments how that "barrier" works as I was first confused... > > Darren, Rafael, can you do review of this patch? > Pali and Gabriele have responded to all questions raised. I have some reservations that this solution is still a bit racy, but it does fix the problem for the affected users. I've queued this. Thanks for your patience. -- Darren Hart Intel Open Source Technology Center From mboxrd@z Thu Jan 1 00:00:00 1970 From: Darren Hart Subject: Re: [PATCH v3] dell-rbtn: Ignore ACPI notifications if device is suspended Date: Mon, 23 May 2016 14:26:55 -0700 Message-ID: <20160523212655.GA2735@f23x64.localdomain> References: <1457740175-8327-1-git-send-email-gabriele.mzt@gmail.com> <20160328173309.GA26086@dvhart-mobl5.amr.corp.intel.com> <4072492.lANJWhSkYa@vostro.rjw.lan> <20160418123547.GK29406@pali> <20160519133032.GO29844@pali> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from bombadil.infradead.org ([198.137.202.9]:49552 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751154AbcEWV07 (ORCPT ); Mon, 23 May 2016 17:26:59 -0400 Content-Disposition: inline In-Reply-To: <20160519133032.GO29844@pali> Sender: platform-driver-x86-owner@vger.kernel.org List-ID: To: Pali =?iso-8859-1?Q?Roh=E1r?= Cc: Gabriele Mazzotta , "Rafael J. Wysocki" , "D. Jared Dominguez" , "platform-driver-x86@vger.kernel.org" , Alex Hung , Andrei Borzenkov , "linux-kernel@vger.kernel.org" On Thu, May 19, 2016 at 03:30:32PM +0200, Pali Roh=E1r wrote: > On Monday 25 April 2016 22:06:11 Gabriele Mazzotta wrote: > > 2016-04-18 14:35 GMT+02:00 Pali Roh=E1r : > > > On Tuesday 29 March 2016 15:11:35 Rafael J. Wysocki wrote: > > >> On Monday, March 28, 2016 10:33:09 AM Darren Hart wrote: > > >> > On Thu, Mar 24, 2016 at 12:24:56PM +0100, Gabriele Mazzotta wr= ote: > > >> > > 2016-03-24 10:39 GMT+01:00 Pali Roh=E1r : > > >> > > > On Monday 21 March 2016 16:13:34 Gabriele Mazzotta wrote: > > >> > > >> 2016-03-21 13:17 GMT+01:00 Pali Roh=E1r : > > >> > > >> > On Friday 18 March 2016 23:44:23 Gabriele Mazzotta wrot= e: > > >> > > >> >> +#ifdef CONFIG_PM_SLEEP > > >> > > >> >> +static void ACPI_SYSTEM_XFACE rbtn_acpi_clear_flag(vo= id *context) > > >> > > >> >> +{ > > >> > > >> >> + struct rbtn_data *rbtn_data =3D context; > > >> > > >> >> + > > >> > > >> >> + rbtn_data->suspended =3D false; > > >> > > >> >> +} > > >> > > >> >> + > > >> > > >> >> +static int rbtn_suspend(struct device *dev) > > >> > > >> >> +{ > > >> > > >> >> + struct acpi_device *device =3D to_acpi_device(de= v); > > >> > > >> >> + struct rbtn_data *rbtn_data =3D acpi_driver_data= (device); > > >> > > >> >> + > > >> > > >> >> + rbtn_data->suspended =3D true; > > >> > > >> >> + > > >> > > >> >> + return 0; > > >> > > >> >> +} > > >> > > >> >> + > > >> > > >> >> +static int rbtn_resume(struct device *dev) > > >> > > >> >> +{ > > >> > > >> >> + struct acpi_device *device =3D to_acpi_device(de= v); > > >> > > >> >> + struct rbtn_data *rbtn_data =3D acpi_driver_data= (device); > > >> > > >> >> + acpi_status status; > > >> > > >> >> + > > >> > > >> >> + /* > > >> > > >> >> + * Clear the flag only after we received the ext= ra > > >> > > >> >> + * ACPI notification. > > >> > > >> >> + */ > > >> > > >> >> + status =3D acpi_os_execute(OSL_NOTIFY_HANDLER, > > >> > > >> >> + rbtn_acpi_clear_flag, rbtn_data= ); > > >> > > >> >> + if (ACPI_FAILURE(status)) > > >> > > >> >> + rbtn_data->suspended =3D false; > > >> > > >> > > > >> > > >> > I case when acpi_os_execute success it calls rbtn_acpi_= clear_flag, > > >> > > >> > right? And that will set suspended to false. When acpi_= os_execute fails, > > >> > > >> > then it set suspended too to false... Then whole acpi_o= s_execute doing > > >> > > >> > just "barrier" after which suspended flag can be set to= false. So I > > >> > > >> > think rbtn_acpi_clear_flag function is not needed here. > > >> > > >> > > > >> > > >> > Cannot you pass NULL or empty function pointer as callb= ack? Or what was > > >> > > >> > reason to do that flag clearing at "two places"? > > >> > > >> > > >> > > >> acpi_os_execute doesn't wait for the callback to be execu= ted, so > > >> > > >> I can't clear the flag from rbtn_resume. > > >> > > > > > >> > > > acpi_os_execute calls callback asynchronously later? Or wh= at exactly do it? > > >> > > > > >> > > In this case, it adds the callback to the kacpi_notify_wq wo= rkqueue > > >> > > for deferred execution. > > >> > > > >> > +Rafael for context/advice on the use of acpi_os_execute here. > > >> > > > >> > This is true, but a quick scan through that call path doesn't = tell me why we > > >> > would need to call it here instead of just setting rbtn_data->= suspended =3D false. > > >> > The comment suggests waiting for the event, but is that what t= his is doing? It > > >> > appears to me to be immediately scheduling the function to a w= ork queue, not > > >> > waiting for the event notifier. > > >> > > >> I think this is supposed to work as a barrier. That is, it will= only run after > > >> all events in the queue have been processed. > > >> > > >> I'm not sure if that's necessary, though. > > >> > > >> Thanks, > > >> Rafael > > >> > > > > > > Darren, Gabriele, what is state of this patch? Bug is not still f= ixed, > > > right? > >=20 > > Yes, the bug is still there and this patch fixes it. > >=20 > > Just to make it clear, we need the barrier. Andrei could reproduce > > the bug without it [1], but not with it, as he confirmed in this > > thread [2]. > >=20 > > [1] http://article.gmane.org/gmane.linux.drivers.platform.x86.devel= /8001 > > [2] http://article.gmane.org/gmane.linux.drivers.platform.x86.devel= /8937 >=20 > Ok, so it means that somebody (who understand ACPI) should review cod= e > and accept it or show what is needed to fix. Plus maybe adds more > comments how that "barrier" works as I was first confused... >=20 > Darren, Rafael, can you do review of this patch? >=20 Pali and Gabriele have responded to all questions raised. I have some reservations that this solution is still a bit racy, but it does fix th= e problem for the affected users. I've queued this. Thanks for your patience. --=20 Darren Hart Intel Open Source Technology Center