From mboxrd@z Thu Jan 1 00:00:00 1970 From: Huang Ying Subject: Re: [BUGFIX] PM: Fix active child counting when disabled and forbidden Date: Mon, 12 Nov 2012 13:55:44 +0800 Message-ID: <1352699744.7176.169.camel@yhuang-dev> References: Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Return-path: Received: from mga01.intel.com ([192.55.52.88]:48164 "EHLO mga01.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751685Ab2KLFzr (ORCPT ); Mon, 12 Nov 2012 00:55:47 -0500 In-Reply-To: Sender: linux-pm-owner@vger.kernel.org List-Id: linux-pm@vger.kernel.org To: Alan Stern Cc: "Rafael J. Wysocki" , linux-kernel@vger.kernel.org, linux-pm@vger.kernel.org On Sun, 2012-11-11 at 21:36 -0500, Alan Stern wrote: > On Mon, 12 Nov 2012, Huang Ying wrote: > > > > The first question: How should the PCI subsystem prevent the parents of > > > driverless VGA devices from being runtime suspended while userspace is > > > accessing them? > > > > I think Rafael's patch is good for that. > > But his patch isn't needed if we make these other changes. Yes. > > > The second question: Should the PM core allow devices that are disabled > > > for runtime PM to be in the SUSPENDED state when > > > dev->power.runtime_auto is clear? > > > > I think that should not be allowed. > > Disallowing it is okay with me too. But it will require several > changes to the code, more than what your patch did. Yes. I think so too. > > > Assuming we don't want to allow this, there's a third question: Should > > > pm_runtime_allow call pm_runtime_set_suspended if the device is > > > disabled? > > > > Is it absolute necessary to call pm_runtime_set_suspended? If the > > device is disabled, the transition to SUSPENDED state will not be > > triggered even if the device is ACTIVE. > > It's not absolutely necessary to do this, but we ought to because it > will allow the device's parent to be suspended. If we leave the device > in the ACTIVE state then the parent can't be suspended, even when the > device is disabled. I think this is the hard part of the issue. Now "disabled" and SUSPENDED state is managed by hand. IMHO, if we changed pm_runtime_allow() as you said, we need to change the rule too. Maybe something as follow: - remove pm_runtime_set_suspended/pm_runtime_set_active - in pm_runtime_disable/pm_runtime_allow, put device into SUSPENDED state if runtime PM is not forbidden. - in pm_runtime_forbid/pm_runtime_enable, put device into ACTIVE state. Best Regards, Huang Ying