From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760339AbZC1D43 (ORCPT ); Fri, 27 Mar 2009 23:56:29 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1756820AbZC1D4U (ORCPT ); Fri, 27 Mar 2009 23:56:20 -0400 Received: from mx2.redhat.com ([66.187.237.31]:53463 "EHLO mx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754179AbZC1D4T (ORCPT ); Fri, 27 Mar 2009 23:56:19 -0400 Message-ID: <49CD9E25.2090407@redhat.com> Date: Sat, 28 Mar 2009 06:48:53 +0300 From: Avi Kivity User-Agent: Thunderbird 2.0.0.21 (X11/20090320) MIME-Version: 1.0 To: Jeremy Fitzhardinge CC: Nick Piggin , Linux Kernel Mailing List , Linux Memory Management List , the arch/x86 maintainers Subject: Re: [PATCH 1/2] x86/mm: maintain a percpu "in get_user_pages_fast" flag References: <49CD37B8.4070109@goop.org> In-Reply-To: <49CD37B8.4070109@goop.org> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Jeremy Fitzhardinge wrote: > get_user_pages_fast() relies on cross-cpu tlb flushes being a barrier > between clearing and setting a pte, and before freeing a pagetable page. > It usually does this by disabling interrupts to hold off IPIs, but > some tlb flush implementations don't use IPIs for tlb flushes, and > must use another mechanism. > > In this change, add in_gup_cpumask, which is a cpumask of cpus currently > performing a get_user_pages_fast traversal of a pagetable. A cross-cpu > tlb flush function can use this to determine whether it should hold-off > on the flush until the gup_fast has finished. > > @@ -255,6 +260,10 @@ int get_user_pages_fast(unsigned long start, int > nr_pages, int write, > * address down to the the page and take a ref on it. > */ > local_irq_disable(); > + > + cpu = smp_processor_id(); > + cpumask_set_cpu(cpu, in_gup_cpumask); > + This will bounce a cacheline, every time. Please wrap in CONFIG_XEN and skip at runtime if Xen is not enabled. -- I have a truly marvellous patch that fixes the bug which this signature is too narrow to contain.