From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-pd0-f173.google.com (mail-pd0-f173.google.com [209.85.192.173]) by kanga.kvack.org (Postfix) with ESMTP id 51C546B0031 for ; Wed, 19 Feb 2014 14:28:06 -0500 (EST) Received: by mail-pd0-f173.google.com with SMTP id y10so771241pdj.32 for ; Wed, 19 Feb 2014 11:28:05 -0800 (PST) Received: from mail.linuxfoundation.org (mail.linuxfoundation.org. [140.211.169.12]) by mx.google.com with ESMTP id iy3si587634pbb.214.2014.02.19.11.28.04 for ; Wed, 19 Feb 2014 11:28:05 -0800 (PST) Date: Wed, 19 Feb 2014 11:28:03 -0800 From: Andrew Morton Subject: Re: [PATCH -mm 0/3] fix numa vs kvm scalability issue Message-Id: <20140219112803.75c6daf470dad88eb10f5dab@linux-foundation.org> In-Reply-To: <20140219085917.GJ27965@twins.programming.kicks-ass.net> References: <1392761566-24834-1-git-send-email-riel@redhat.com> <20140219085917.GJ27965@twins.programming.kicks-ass.net> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org List-ID: To: Peter Zijlstra Cc: riel@redhat.com, linux-kernel@vger.kernel.org, kvm@vger.kernel.org, linux-mm@kvack.org, chegu_vinod@hp.com, aarcange@redhat.com On Wed, 19 Feb 2014 09:59:17 +0100 Peter Zijlstra wrote: > On Tue, Feb 18, 2014 at 05:12:43PM -0500, riel@redhat.com wrote: > > The NUMA scanning code can end up iterating over many gigabytes > > of unpopulated memory, especially in the case of a freshly started > > KVM guest with lots of memory. > > > > This results in the mmu notifier code being called even when > > there are no mapped pages in a virtual address range. The amount > > of time wasted can be enough to trigger soft lockup warnings > > with very large (>2TB) KVM guests. > > > > This patch moves the mmu notifier call to the pmd level, which > > represents 1GB areas of memory on x86-64. Furthermore, the mmu > > notifier code is only called from the address in the PMD where > > present mappings are first encountered. > > > > The hugetlbfs code is left alone for now; hugetlb mappings are > > not relocatable, and as such are left alone by the NUMA code, > > and should never trigger this problem to begin with. > > > > The series also adds a cond_resched to task_numa_work, to > > fix another potential latency issue. > > Andrew, I'll pick up the first kernel/sched/ patch; do you want the > other two mm/ patches? That works, thanks. -- 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