Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756900AbcKWRDj (ORCPT ); Wed, 23 Nov 2016 12:03:39 -0500 Received: from mga11.intel.com ([192.55.52.93]:32106 "EHLO mga11.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754998AbcKWRDg (ORCPT ); Wed, 23 Nov 2016 12:03:36 -0500 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.31,539,1473145200"; d="scan'208";a="34889727" Subject: Re: Enabling peer to peer device transactions for PCIe devices To: Dan Williams , Serguei Sagalovitch , "linux-nvdimm@lists.01.org" , "linux-rdma@vger.kernel.org" , "linux-pci@vger.kernel.org" , "Kuehling, Felix" , "linux-kernel@vger.kernel.org" , "dri-devel@lists.freedesktop.org" , "Koenig, Christian" , "Sander, Ben" , "Suthikulpanit, Suravee" , "Deucher, Alexander" , "Blinzer, Paul" , "Linux-media@vger.kernel.org" References: <75a1f44f-c495-7d1e-7e1c-17e89555edba@amd.com> <20161123074902.ph7a5cmlw3pclugx@phenom.ffwll.local> From: Dave Hansen Message-ID: <7ce9026f-871e-d50a-20cf-19f7e2d90649@linux.intel.com> Date: Wed, 23 Nov 2016 09:03:33 -0800 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.4.0 MIME-Version: 1.0 In-Reply-To: <20161123074902.ph7a5cmlw3pclugx@phenom.ffwll.local> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 552 Lines: 10 On 11/22/2016 11:49 PM, Daniel Vetter wrote: > Yes, agreed. My idea with exposing vram sections using numa nodes wasn't > to reuse all the existing allocation policies directly, those won't work. > So at boot-up your default numa policy would exclude any vram nodes. > > But I think (as an -mm layman) that numa gives us a lot of the tools and > policy interface that we need to implement what we want for gpus. Are you suggesting creating NUMA nodes for video RAM (I assume that's what you mean by vram) where that RAM is not at all CPU-accessible?