From mboxrd@z Thu Jan 1 00:00:00 1970 From: Serge Semin Subject: Re: [PATCH 9/9 v2] usb: usb251xb: Use GPIO descriptor consumer interface Date: Thu, 21 Sep 2017 00:29:28 +0300 Message-ID: <20170920212928.GC3547@mobilestation> References: <20170915233113.17855-1-fancer.lancer@gmail.com> <20170916104220.3742-1-fancer.lancer@gmail.com> <20170916104220.3742-10-fancer.lancer@gmail.com> <20170920133735.k3yzh2m6zb5qzhn2@rob-hp-laptop> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20170920133735.k3yzh2m6zb5qzhn2@rob-hp-laptop> Sender: linux-kernel-owner@vger.kernel.org To: Rob Herring Cc: richard.leitner@skidata.com, gregkh@linuxfoundation.org, mark.rutland@arm.com, Sergey.Semin@t-platforms.ru, linux-usb@vger.kernel.org, devicetree@vger.kernel.org, linux-kernel@vger.kernel.org List-Id: devicetree@vger.kernel.org On Wed, Sep 20, 2017 at 03:52:57PM -0500, Rob Herring wrote: > On Sat, Sep 16, 2017 at 01:42:20PM +0300, Serge Semin wrote: > > The driver used to be developed with legacy GPIO API support. It's > > better to use descriptor-based interface for several reasons. First > > of all the legacy API doesn't support the ACTIVE_LOW/HIGH flag of dts > > nodes, which is essential since different hardware may have different > > GPIOs connectivity including the logical value inversion. Secondly, > > by requesting the reset GPIO descriptor the driver prevent the other > > applications from changing its value. And last but not least the > > legacy GPIO interface should be avoided in the new code due to it > > obsolescence. > > All this has nothing to do with the binding. Please make the binding > change separate and state why it should be optional. > Ok. > > > > Signed-off-by: Serge Semin > > --- > > Documentation/devicetree/bindings/usb/usb251xb.txt | 2 +- > > drivers/usb/misc/usb251xb.c | 34 +++++++++------------- > > 2 files changed, 15 insertions(+), 21 deletions(-)