From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jordan Crouse Subject: Re: [PATCH v1 3/4] dt-bindings: drm/msm/gpu: Document a5xx / a6xx zap shader region Date: Mon, 1 Apr 2019 10:13:24 -0600 Message-ID: <20190401161324.GC10168@jcrouse1-lnx.qualcomm.com> References: <1552414422-9568-1-git-send-email-jcrouse@codeaurora.org> <1552414422-9568-4-git-send-email-jcrouse@codeaurora.org> <20190328133215.GA14020@bogus> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20190328133215.GA14020@bogus> Sender: linux-kernel-owner@vger.kernel.org To: Rob Herring Cc: freedreno@lists.freedesktop.org, Bjorn Andersson , Sean Paul , devicetree@vger.kernel.org, linux-arm-msm@vger.kernel.org, linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org, Rob Clark , David Airlie , Mark Rutland , Daniel Vetter List-Id: dri-devel@lists.freedesktop.org On Thu, Mar 28, 2019 at 08:32:15AM -0500, Rob Herring wrote: > On Tue, Mar 12, 2019 at 12:13:41PM -0600, Jordan Crouse wrote: > > Describe the zap-shader node that defines a reserved memory region > > to store the zap shader. > > > > Signed-off-by: Jordan Crouse > > --- > > > > Documentation/devicetree/bindings/display/msm/gpu.txt | 7 +++++++ > > 1 file changed, 7 insertions(+) > > > > diff --git a/Documentation/devicetree/bindings/display/msm/gpu.txt b/Documentation/devicetree/bindings/display/msm/gpu.txt > > index aad1aef..1e6d870 100644 > > --- a/Documentation/devicetree/bindings/display/msm/gpu.txt > > +++ b/Documentation/devicetree/bindings/display/msm/gpu.txt > > @@ -25,6 +25,9 @@ Required properties: > > - qcom,gmu: For GMU attached devices a phandle to the GMU device that will > > control the power for the GPU. Applicable targets: > > - qcom,adreno-630.2 > > +- zap-shader: For a5xx and a6xx devices this node contains a memory-region that > > + points to reserved memory to store the zap shader that can be used to help > > + bring the GPU out of secure mode. > > This is the properties section and zap-shader is not a property. Thanks, I can fix that. > But why do you need a child node in the first place? Just add > 'memory-region' to the parent. Two reasons. First, this memory is locked in the secure world once the MDT loader is run and it isn't really intended for CPU access. If the parent device tries to set up DMA operations thinking it has access to the memory it might not work out very well. Putting it in a child makes it clear that this is a special chunk of memory for a special case. The second reason is that not all target platforms require the zap shader, so it would be nice to have it as a child node so that it could be removed on target platforms that don't need it. Jordan -- The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, a Linux Foundation Collaborative Project