From mboxrd@z Thu Jan 1 00:00:00 1970 From: Laxman Dewangan Subject: Re: [PATCH 1/3] iio: adc: ina3221: Add DT binding details Date: Fri, 3 Jun 2016 17:51:53 +0530 Message-ID: <57517661.8090205@nvidia.com> References: <1464784454-7988-1-git-send-email-ldewangan@nvidia.com> <20160603020759.GA23947@rob-hp-laptop> <433b182c-fe43-df5c-ac6a-20a024630514@kernel.org> <57516E71.8080506@nvidia.com> Mime-Version: 1.0 Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: Jonathan Cameron , Rob Herring Cc: corbet@lwn.net, lars@metafoo.de, devicetree@vger.kernel.org, linux-kernel@vger.kernel.org, linux-doc@vger.kernel.org, linux-iio@vger.kernel.org List-Id: devicetree@vger.kernel.org On Friday 03 June 2016 05:41 PM, Jonathan Cameron wrote: > On 03/06/16 12:48, Laxman Dewangan wrote: >> On Friday 03 June 2016 03:49 PM, Jonathan Cameron wrote: >>>>> + >>>>> +enable-power-monitor: Boolean, Enable power monitoring of the device. >>> Is this the power good stuff? description should be more detailed. >> If there is no shunt resistance then we can not enable power monitor >> on that rail. Device does not mandate to have shunt and so this is >> based on platforms. > The voltage shunt also becomes meaningless (unless you have other very > small voltage drops to measure!). So drop that channel as well. > Either it is there to do current measurement or it isn't. OK. understood. >>>>> + >>>>> +enable-continuous-mode: Boolean. Device support oneshot and continuous >>>>> + mode for the channel data conversion. Presence >>>>> + of this property will enable the continuous >>>>> + mode from boot. >>> Is the difference between driver load time and the point where usespace can >>> set it up significant enough to justify this? >> We change the mode dynamically. If we have more core then goto the >> continuous mode so that we can apply throttling if power consumption >> is going more than requirement. If we are running single core then >> change to oneshot mode. > That's definitely a usespace or firmware decision, not a kernel one > to my mind - unless I guess you an enforcing bringing this device up > before firing up the additional cores? > > Then it's nasty but I can start to see some justification. Yaah, this is FW decision. Our need on some platform was that start the driver with the continuous mode. But I think that may be specific need. I can remove this and let user space can enable the continuous mode.