From mboxrd@z Thu Jan 1 00:00:00 1970 From: Balaji T K Subject: Re: [PATCH v6 3/7] regulator: add pbias regulator support Date: Fri, 20 Dec 2013 22:14:44 +0530 Message-ID: <52B473FC.7000404@ti.com> References: <1386670577-3530-1-git-send-email-balajitk@ti.com> <1387456720-7202-1-git-send-email-balajitk@ti.com> <1387456720-7202-4-git-send-email-balajitk@ti.com> <20131219163300.GK27438@atomide.com> <52B4124F.8010903@ti.com> <20131220155721.GX27438@atomide.com> <20131220160917.GY31886@sirena.org.uk> <20131220161307.GB27438@atomide.com> Mime-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1"; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20131220161307.GB27438@atomide.com> Sender: linux-mmc-owner@vger.kernel.org To: Tony Lindgren Cc: Mark Brown , linux-omap@vger.kernel.org, bcousson@baylibre.com, devicetree@vger.kernel.org, linux-mmc@vger.kernel.org, cjb@laptop.org List-Id: devicetree@vger.kernel.org On Friday 20 December 2013 09:43 PM, Tony Lindgren wrote: > * Mark Brown [131220 08:10]: >> On Fri, Dec 20, 2013 at 07:57:21AM -0800, Tony Lindgren wrote: >>> * Balaji T K [131220 01:49]: >> >>>> But interrupt was never used/tested AFAIK, there is some settling time >>>> before the generated interrupt status is truely valid, so pbias interrupt is not >>>> reliable. >> >>> OK. Do we need the standard regulator property startup-delay-us for the >>> PBIAS regulator then? Or if it's always fixed, I guess it could be done >>> in the pbias_regulator_enable()? >> >> That delay is supposed to be the time for the startup of the supply >> rather than any detection code. It should be set using enable_time in >> the driver if it's not system dependent - the property is there for >> cases where the delay depends on system configuration (eg, due to the >> capacitor values). > > OK thanks. Let's try enable_time first then as we've had a fixed value > for years for it. If there's some difference based on the card > capacitance etc that can be added if needed. I used regulator-enable-ramp-delay to set constraints->enable_time based on SoC. Since it is not board dependent, I will drop regulator-enable-ramp-delay from dt and pass the value via driver_data. Thanks and Regards, Balaji T K