From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: [PATCH 21/22] mm: remove the HMM config option Date: Fri, 14 Jun 2019 08:47:16 +0200 Message-ID: <20190614064716.GN7246@lst.de> References: <20190613094326.24093-1-hch@lst.de> <20190613094326.24093-22-hch@lst.de> <20190613200150.GB22062@mellanox.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <20190613200150.GB22062-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: linux-nvdimm-bounces-hn68Rpc1hR1g9hUCZPvPmw@public.gmane.org Sender: "Linux-nvdimm" To: Jason Gunthorpe Cc: "linux-nvdimm-hn68Rpc1hR1g9hUCZPvPmw@public.gmane.org" , "nouveau-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org" , "linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , "dri-devel-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org" , "linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org" , =?iso-8859-1?B?Suly9G1l?= Glisse , Ben Skeggs , "linux-pci-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , Christoph Hellwig List-Id: nouveau.vger.kernel.org On Thu, Jun 13, 2019 at 08:01:55PM +0000, Jason Gunthorpe wrote: > On Thu, Jun 13, 2019 at 11:43:24AM +0200, Christoph Hellwig wrote: > > All the mm/hmm.c code is better keyed off HMM_MIRROR. Also let nouveau > > depend on it instead of the mix of a dummy dependency symbol plus the > > actually selected one. Drop various odd dependencies, as the code is > > pretty portable. > > I don't really know, but I thought this needed the arch restriction > for the same reason get_user_pages has various unique arch specific > implementations (it does seem to have some open coded GUP like thing)? > > I was hoping we could do this after your common gup series? But sooner > is better too. Ok, I've added the arch and 64-bit dependency back in for now. It does not look proper to me, and is certainly underdocumented, but the whole pagetable walking code will need a lot of love eventually anyway, and the Kconfig stuff for it can be done properly then.