From mboxrd@z Thu Jan 1 00:00:00 1970 From: Johan Hovold Subject: Re: [PATCH v2 4/4] mfd: dln2: add suspend/resume functionality Date: Wed, 7 Jan 2015 15:26:05 +0100 Message-ID: <20150107142605.GG22905@localhost> References: <1418745435-2851-1-git-send-email-octavian.purdila@intel.com> <1418745435-2851-5-git-send-email-octavian.purdila@intel.com> <20150107102637.GD22905@localhost> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: Octavian Purdila Cc: Johan Hovold , Linus Walleij , Lee Jones , "linux-usb@vger.kernel.org" , lkml , "linux-gpio@vger.kernel.org" List-Id: linux-gpio@vger.kernel.org On Thu, Jan 08, 2015 at 02:16:15AM +1300, Octavian Purdila wrote: > On Wed, Jan 7, 2015 at 11:26 PM, Johan Hovold wrote: > > On Tue, Dec 16, 2014 at 05:57:15PM +0200, Octavian Purdila wrote: > >> Without suspend/resume functionality in the USB driver the USB core > >> will disconnect and reconnect the DLN2 port and because the GPIO > >> framework does not yet support removal of an in-use controller a > >> suspend/resume operation will result in a crash. > >> > >> This patch provides suspend and resume functions for the DLN2 driver > >> so that the above scenario is avoided. > >> > >> Signed-off-by: Octavian Purdila > > > > This patch looks good now, but how did you say this device was powered? > > > > If powered by vbus you cannot assume that the device maintains it's > > state over a suspend cycle, something which would complicate matters > > quite a bit... > > Yes, the device is powered by VBUS. During my tests, depending on the > host and USB port, VBUS is sometimes preserved - and this is the case > this patch addresses, and sometimes is not, but in that case because > no reset_resume routine is implemented the resume path will go through > the disconnect/reconnect process. Yes, it depends on the host controller and I believe the common case is to drop VBUS. > This second case is not addressed by > the patch, as I think in this second case fixing the GPIO framework to > support the removal of an in-use device is the best way to go. I agree. But please do mention that this case is not handled in the commit message. Thanks, Johan