From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gerd Hoffmann Subject: Re: pv_ops & gntdev? Date: Wed, 25 Feb 2009 22:43:26 +0100 Message-ID: <49A5BB7E.8030503@redhat.com> References: <49A44030.2070709@redhat.com> <49A4640E.1000807@goop.org> <49A470DD.2000008@redhat.com> <49A517F6.30005@redhat.com> <49A58506.2020407@goop.org> <49A58FFF.3050604@redhat.com> <49A5A4BD.7080207@goop.org> <49A5B5DD.60309@redhat.com> <49A5B9AA.7010709@goop.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <49A5B9AA.7010709@goop.org> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Jeremy Fitzhardinge Cc: Xen Development Mailing List List-Id: xen-devel@lists.xenproject.org Jeremy Fitzhardinge wrote: > Gerd Hoffmann wrote: >> Jeremy Fitzhardinge wrote: >> >>> Hm, I hope not. Doesn't it always end up using set_pte*? >>> >> >> ptep_get_and_clear_full(); >> >> See zap_pte_range() in mm/memory.c > > Ah, right. It could be changed to use ptep_modify_prot_start/commit. > That would make it batchable too. Slightly less good for native. Hmm. One more Q: How is pinning used? Is a process pinned all the time? Or can it happen that the pages are unpinned in case it does hang around idle for a while? Is there some way to prevent a process from being unpinned? I somehow feel like checking out mmu notifiers first ... thanks, Gerd