From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: From: Jason Gunthorpe Subject: Re: [PATCH 18/22] mm: mark DEVICE_PUBLIC as broken Date: Wed, 19 Jun 2019 19:27:25 +0000 Message-ID: <20190619192719.GO9374@mellanox.com> References: <20190613094326.24093-1-hch@lst.de> <20190613094326.24093-19-hch@lst.de> <20190613194430.GY22062@mellanox.com> <20190613195819.GA22062@mellanox.com> <20190614004314.GD783@iweiny-DESK2.sc.intel.com> In-Reply-To: Content-Language: en-US Content-Type: text/plain; charset="iso-8859-1" Content-ID: <081C2DB7B7A13A4A93487CC051D6E961@eurprd05.prod.outlook.com> Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Sender: owner-linux-mm@kvack.org To: John Hubbard Cc: Ira Weiny , Ralph Campbell , "linux-nvdimm@lists.01.org" , "nouveau@lists.freedesktop.org" , "linux-kernel@vger.kernel.org" , "dri-devel@lists.freedesktop.org" , "linux-mm@kvack.org" , =?iso-8859-1?Q?J=E9r=F4me_Glisse?= , Ben Skeggs , "linux-pci@vger.kernel.org" , Christoph Hellwig List-ID: On Thu, Jun 13, 2019 at 06:23:04PM -0700, John Hubbard wrote: > On 6/13/19 5:43 PM, Ira Weiny wrote: > > On Thu, Jun 13, 2019 at 07:58:29PM +0000, Jason Gunthorpe wrote: > >> On Thu, Jun 13, 2019 at 12:53:02PM -0700, Ralph Campbell wrote: > >>> > ... > >> Hum, so the only thing this config does is short circuit here: > >> > >> static inline bool is_device_public_page(const struct page *page) > >> { > >> return IS_ENABLED(CONFIG_DEV_PAGEMAP_OPS) && > >> IS_ENABLED(CONFIG_DEVICE_PUBLIC) && > >> is_zone_device_page(page) && > >> page->pgmap->type =3D=3D MEMORY_DEVICE_PUBLIC; > >> } > >> > >> Which is called all over the place..=20 > >=20 > > yes but the earlier patch: > >=20 > > [PATCH 03/22] mm: remove hmm_devmem_add_resource > >=20 > > Removes the only place type is set to MEMORY_DEVICE_PUBLIC. > >=20 > > So I think it is ok. Frankly I was wondering if we should remove the p= ublic > > type altogether but conceptually it seems ok. But I don't see any user= s of it > > so... should we get rid of it in the code rather than turning the conf= ig off? > >=20 > > Ira >=20 > That seems reasonable. I recall that the hope was for those IBM Power 9 > systems to use _PUBLIC, as they have hardware-based coherent device (GPU) > memory, and so the memory really is visible to the CPU. And the IBM team > was thinking of taking advantage of it. But I haven't seen anything on > that front for a while. Does anyone know who those people are and can we encourage them to send some patches? :) Jason