From mboxrd@z Thu Jan 1 00:00:00 1970 Subject: Re: [PATCH 08 of 11] anon-vma-rwsem From: Peter Zijlstra In-Reply-To: <20080509185553.GF7710@duo.random> References: <20080508025652.GW8276@duo.random> <20080508034133.GY8276@duo.random> <20080508052019.GA8276@duo.random> <1210358249.13978.275.camel@twins> <20080509185553.GF7710@duo.random> Content-Type: text/plain Date: Fri, 09 May 2008 21:04:47 +0200 Message-Id: <1210359887.6524.0.camel@lappy.programming.kicks-ass.net> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org Return-Path: To: Andrea Arcangeli Cc: Linus Torvalds , Christoph Lameter , Andrew Morton , steiner@sgi.com, holt@sgi.com, npiggin@suse.de, kvm-devel@lists.sourceforge.net, kanojsarcar@yahoo.com, rdreier@cisco.com, swise@opengridcomputing.com, linux-kernel@vger.kernel.org, avi@qumranet.com, linux-mm@kvack.org, general@lists.openfabrics.org, hugh@veritas.com, rusty@rustcorp.com.au, aliguori@us.ibm.com, chrisw@redhat.com, marcelo@kvack.org, dada1@cosmosbay.com, paulmck@us.ibm.com List-ID: On Fri, 2008-05-09 at 20:55 +0200, Andrea Arcangeli wrote: > On Fri, May 09, 2008 at 08:37:29PM +0200, Peter Zijlstra wrote: > > Another possibility, would something like this work? > > > > > > /* > > * null out the begin function, no new begin calls can be made > > */ > > rcu_assing_pointer(my_notifier.invalidate_start_begin, NULL); > > > > /* > > * lock/unlock all rmap locks in any order - this ensures that any > > * pending start() will have its end() function called. > > */ > > mm_barrier(mm); > > > > /* > > * now that no new start() call can be made and all start()/end() pairs > > * are complete we can remove the notifier. > > */ > > mmu_notifier_remove(mm, my_notifier); > > > > > > This requires a mmu_notifier instance per attached mm and that > > __mmu_notifier_invalidate_range_start() uses rcu_dereference() to obtain > > the function. > > > > But I think its enough to ensure that: > > > > for each start an end will be called > > We don't need that, it's perfectly ok if start is called but end is > not, it's ok to unregister in the middle as I guarantee ->release is > called before mmu_notifier_unregister returns (if ->release is needed > at all, not the case for KVM/GRU). > > Unregister is already solved with srcu/rcu without any additional > complication as we don't need the guarantee that for each start an end > will be called. > > > It can however happen that end is called without start - but we could > > handle that I think. > > The only reason mm_lock() was introduced is to solve "register", to > guarantee that for each end there was a start. We can't handle end > called without start in the driver. > > The reason the driver must be prevented to register in the middle of > start/end, if that if it ever happens the driver has no way to know it > must stop the secondary mmu page faults to call get_user_pages and > instantiate sptes/secondarytlbs on pages that will be freed as soon as > zap_page_range starts. Right - then I got it backwards. Never mind me then.. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org