From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-oi0-f70.google.com (mail-oi0-f70.google.com [209.85.218.70]) by kanga.kvack.org (Postfix) with ESMTP id 8B7A96B779E for ; Thu, 6 Sep 2018 03:45:23 -0400 (EDT) Received: by mail-oi0-f70.google.com with SMTP id r131-v6so11965414oie.14 for ; Thu, 06 Sep 2018 00:45:23 -0700 (PDT) Received: from mx0a-001b2d01.pphosted.com (mx0b-001b2d01.pphosted.com. [148.163.158.5]) by mx.google.com with ESMTPS id v3-v6si2856562oiv.323.2018.09.06.00.45.22 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Thu, 06 Sep 2018 00:45:22 -0700 (PDT) Received: from pps.filterd (m0098420.ppops.net [127.0.0.1]) by mx0b-001b2d01.pphosted.com (8.16.0.22/8.16.0.22) with SMTP id w867j9qX126379 for ; Thu, 6 Sep 2018 03:45:22 -0400 Received: from e06smtp07.uk.ibm.com (e06smtp07.uk.ibm.com [195.75.94.103]) by mx0b-001b2d01.pphosted.com with ESMTP id 2mayddj0nk-1 (version=TLSv1.2 cipher=AES256-GCM-SHA384 bits=256 verify=NOT) for ; Thu, 06 Sep 2018 03:45:21 -0400 Received: from localhost by e06smtp07.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Thu, 6 Sep 2018 08:45:17 +0100 Subject: Re: Plumbers 2018 - Performance and Scalability Microconference References: <1dc80ff6-f53f-ae89-be29-3408bf7d69cc@oracle.com> <01000165aa490dc9-64abf872-afd1-4a81-a46d-a50d0131de93-000000@email.amazonses.com> <839e2703-1588-0873-00a7-d04810f403cf@linux.vnet.ibm.com> From: Laurent Dufour Date: Thu, 6 Sep 2018 09:45:05 +0200 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit Message-Id: <4d6a466c-6f77-b54e-fb30-9ad8e5bb4023@linux.vnet.ibm.com> Sender: owner-linux-mm@kvack.org List-ID: To: Thomas Gleixner Cc: Christopher Lameter , Daniel Jordan , linux-kernel@vger.kernel.org, "linux-mm@kvack.org" , Aaron Lu , alex.kogan@oracle.com, akpm@linux-foundation.org, boqun.feng@gmail.com, brouer@redhat.com, dave@stgolabs.net, dave.dice@oracle.com, Dhaval Giani , ktkhai@virtuozzo.com, Pavel.Tatashin@microsoft.com, paulmck@linux.vnet.ibm.com, shady.issa@oracle.com, tariqt@mellanox.com, tim.c.chen@intel.com, vbabka@suse.cz, longman@redhat.com, yang.shi@linux.alibaba.com, shy828301@gmail.com, Huang Ying , subhra.mazumdar@oracle.com, Steven Sistare , jwadams@google.com, ashwinch@google.com, sqazi@google.com, Shakeel Butt , walken@google.com, rientjes@google.com, junaids@google.com, Neha Agarwal On 06/09/2018 01:01, Thomas Gleixner wrote: > On Wed, 5 Sep 2018, Laurent Dufour wrote: >> On 05/09/2018 17:10, Christopher Lameter wrote: >>> Large page sizes also reduce contention there. >> >> That's true for the page fault path, but for process's actions manipulating the >> memory process's layout (mmap,munmap,madvise,mprotect) the impact is minimal >> unless the code has to manipulate the page tables. > > And how exactly are you going to do any of those operations _without_ > manipulating the page tables? I agree, at one time the page tables would have to be manipulated, and this is mostly done under the protection of the page table locks - should the mmap_sem still being held then ? I was thinking about all the processing done on the VMAs, accounting, etc. That part, usually not manipulating the page tables, is less dependent of the underlying page size. But I agree at one time of the processing, the page table are manipulated and dealing with larger pages is better then. Thanks, Laurent.