From mboxrd@z Thu Jan 1 00:00:00 1970 From: Terje Bergstrom Subject: Re: [PATCH] arm64: tegra: Add the GPU on Tegra194 Date: Thu, 16 Jul 2020 08:05:24 -0700 Message-ID: <4f9cfab5-2a5d-379c-ba17-1aca896b4110@nvidia.com> References: <20200716120138.532906-1-thierry.reding@gmail.com> <3748e071-4d08-8e77-43f6-de37f5a7eb1e@nvidia.com> <20200716131753.GG535268@ulmo> Mime-Version: 1.0 Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20200716131753.GG535268@ulmo> Content-Language: en-US Sender: linux-tegra-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Thierry Reding , Jon Hunter Cc: linux-tegra-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Debarshi Dutta , linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org List-Id: linux-tegra@vger.kernel.org On 7/16/2020 6:17 AM, Thierry Reding wrote: > On Thu, Jul 16, 2020 at 01:59:33PM +0100, Jon Hunter wrote: >> I also see that for gv11b we populate 'dma-coherent' and so we should >> probably add this as well. > Do we know for certain that the GPU is DMA coherent? I've only tested > this (with local patches to Nouveau) without dma-coherent, so I have not > actually verified that it works without. > > I vaguely recall reading that there are different apertures for sysmem, > one for coherent sysmem and another for non-coherent sysmem. So I'm not > sure if dma-coherent here will work without additional code in the > driver to ensure that all memory is allocated from the coherent sysmem > aperture. We do have the choices of coherent and non-coherent memory accesses in the iGPU via the two sysmem apertures. But in Tegra194 we're forcing coherency from the memory subsystem side, so all accesses are coherent. Terje