From mboxrd@z Thu Jan 1 00:00:00 1970 From: Igor Grinberg Subject: Re: [PATCH] mfd: twl4030-power: Fix poweroff with PM configuration enabled Date: Tue, 04 Nov 2014 19:52:37 +0200 Message-ID: <54591265.3030207@compulab.co.il> References: <20141102180756.GH31454@atomide.com> <5458D29E.50908@compulab.co.il> <20141104154231.GP31454@atomide.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from softlayer.compulab.co.il ([50.23.254.55]:53821 "EHLO compulab.co.il" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750796AbaKDRwm (ORCPT ); Tue, 4 Nov 2014 12:52:42 -0500 In-Reply-To: <20141104154231.GP31454@atomide.com> Sender: linux-omap-owner@vger.kernel.org List-Id: linux-omap@vger.kernel.org To: Tony Lindgren Cc: Lee Jones , Samuel Ortiz , Russell King - ARM Linux , linux-kernel@vger.kernel.org, linux-omap@vger.kernel.org On 11/04/14 17:42, Tony Lindgren wrote: > * Igor Grinberg [141104 05:22]: >> Hi Tony, >> >> On 11/02/14 20:07, Tony Lindgren wrote: >>> Commit e7cd1d1eb16f ("mfd: twl4030-power: Add generic reset >>> configuration") enabled configuring the PM features for twl4030. >>> >>> This caused poweroff command to fail on devices that have the >>> BCI charger on twl4030 wired, or have power wired for VBUS. >>> Instead of powering off, the device reboots. This is because >>> voltage is detected on charger or VBUS with the default bits >>> enabled for the power transition registers. >>> >>> To fix the issue, let's just clear VBUS and CHG bits as we want >>> poweroff command to keep the system powered off. >> >> What about devices that really need to start once VBUS or CHG is connected? > > More handling can be added for some cases. With this patch the > poweron bits will clear to defaults if power is completely removed. > So start-up with VBUS and CHG works in that case. > > However, if you have a battery connected, and you poweroff, with > this patch the device won't power up with VBUS or CHG connected. > > Note that most battery operated devices are not using the charger > on twl4030 because it has issues charging a completely empty > battery AFAIK. So most battery powered devices have been using an > external USB charger chip that's not affected by this patch. > > We could consider exporting a function for the charger driver to > configure the poweron mask. And we could also consider passing a > mask in ti,use_poweroff = 0xff. Ok. That sounds better to me. Yet, if you say there are no such devices in practice, IMHO, we can merge this. > >> It seems to me that forcing these bits on power off can break that kind of >> devices and these settings should really be board specific. >> What do you think? > > There's a patch series for "[RFC,01/16] kernel: Add support for > poweroff handler call chain" that should help with that. For sure > the poweroff handling needs to be board specific as some systems > may need to use a GPIO to shut off a regulator powering something > before powering off the SoC. Yes, I've seen this series. I'm not sure though that I understand how is this supposed to be used with DT... Through the regulator bindings? Which will tell it to hook up on the call chain? -- Regards, Igor.