From mboxrd@z Thu Jan 1 00:00:00 1970 From: guillaume.tucker@collabora.com (Guillaume Tucker) Date: Mon, 24 Apr 2017 11:43:52 +0100 Subject: [PATCH v2 1/5] dt-bindings: gpu: add bindings for the ARM Mali Midgard GPU In-Reply-To: References: <20170404020056.d7osj7rmog22xbjd@rob-hp-laptop> Message-ID: <32edad9f-a568-c67d-060d-055517d6782f@collabora.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 18/04/17 10:15, Guillaume Tucker wrote: > Hi Rob, > > On 04/04/17 03:00, Rob Herring wrote: >> On Sun, Apr 02, 2017 at 08:59:44AM +0100, Guillaume Tucker wrote: >>> +- operating-points : Refer to Documentation/devicetree/bindings/power/opp.txt >>> + for details. >> >> Is this going to be sufficient vs. operating-points-v2? Or should it be >> a power domain with OPPs? > > In principle, switching to operating-points-v2 should be very > straightforward. I have smoke-tested the driver with an > operating-points-v2 table and a phandle to it inside the gpu node > in place of operating-points and it seems to be working fine. At > least it parsed the OPPs and got initialised correctly. > > My understanding is that operating-points (v1) are not deprecated > so we could keep the bindings as-is, but please let me know > otherwise and I can try to address that in my next patch version. > In the documentation, it should only be the case of replacing > operating-points with operating-points-v2. While the opp bindings documentation doesn't mention anything about operating-points being deprecated, the code and comments in of.c are pretty clear about this: /* * OPPs have two version of bindings now. The older one is deprecated, * try for the new binding first. */ So I shall use operating-points-v2 in my patch v4. Thanks, Guillaume