From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755717AbdCXHQ4 (ORCPT ); Fri, 24 Mar 2017 03:16:56 -0400 Received: from mga09.intel.com ([134.134.136.24]:62591 "EHLO mga09.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753903AbdCXHQs (ORCPT ); Fri, 24 Mar 2017 03:16:48 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.36,213,1486454400"; d="scan'208";a="80130656" From: "Huang\, Ying" To: John Hubbard Cc: "Huang\, Ying" , David Rientjes , Andrew Morton , Andi Kleen , Dave Hansen , Shaohua Li , Rik van Riel , Tim Chen , Michal Hocko , Mel Gorman , Aaron Lu , Gerald Schaefer , "Kirill A. Shutemov" , Hugh Dickins , Ingo Molnar , Vegard Nossum , , Subject: Re: [PATCH -v2 1/2] mm, swap: Use kvzalloc to allocate some swap data structure References: <20170320084732.3375-1-ying.huang@intel.com> <8737e3z992.fsf@yhuang-dev.intel.com> <87poh7xoms.fsf@yhuang-dev.intel.com> <2d55e06d-a0b6-771a-bba0-f9517d422789@nvidia.com> Date: Fri, 24 Mar 2017 15:16:41 +0800 In-Reply-To: <2d55e06d-a0b6-771a-bba0-f9517d422789@nvidia.com> (John Hubbard's message of "Thu, 23 Mar 2017 23:48:35 -0700") Message-ID: <87d1d7uoti.fsf@yhuang-dev.intel.com> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/25.1 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=ascii Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org John Hubbard writes: > On 03/23/2017 09:52 PM, Huang, Ying wrote: >> John Hubbard writes: >> >>> On 03/23/2017 07:41 PM, Huang, Ying wrote: >>>> David Rientjes writes: >>>> >>>>> On Mon, 20 Mar 2017, Huang, Ying wrote: >>>>> >>>>>> From: Huang Ying >>>>>> >>>>>> Now vzalloc() is used in swap code to allocate various data >>>>>> structures, such as swap cache, swap slots cache, cluster info, etc. >>>>>> Because the size may be too large on some system, so that normal >>>>>> kzalloc() may fail. But using kzalloc() has some advantages, for >>>>>> example, less memory fragmentation, less TLB pressure, etc. So change >>>>>> the data structure allocation in swap code to use kvzalloc() which >>>>>> will try kzalloc() firstly, and fallback to vzalloc() if kzalloc() >>>>>> failed. >>>>>> >>>>> >>>>> As questioned in -v1 of this patch, what is the benefit of directly >>>>> compacting and reclaiming memory for high-order pages by first preferring >>>>> kmalloc() if this does not require contiguous memory? >>>> >>>> The memory allocation here is only for swap on time, not for swap out/in >>>> time. The performance of swap on is not considered critical. But if >>>> the kmalloc() is used instead of the vmalloc(), the swap out/in >>>> performance could be improved (marginally). More importantly, the >>>> interference for the other activity on the system could be reduced, For >>>> example, less memory fragmentation, less TLB usage of swap subsystem, >>>> etc. >>> >>> Hi Ying, >>> >>> I'm a little surprised to see vmalloc calls replaced with >>> kmalloc-then-vmalloc calls, because that actually makes fragmentation >>> worse (contrary to the above claim). That's because you will consume >>> contiguous memory (even though you don't need it to be contiguous), >>> whereas before, you would have been able to get by with page-at-a-time >>> for vmalloc. >>> >>> So, things like THP will find fewer contiguous chunks, as a result of patches such as this. >> >> Hi, John, >> >> I don't think so. The pages allocated by vmalloc() cannot be moved >> during de-fragment. For example, if 512 dis-continuous physical pages >> are allocated via vmalloc(), at worst, one page will be allocate from >> one distinct 2MB continous physical pages. This makes 512 * 2MB = 1GB >> memory cannot be used for THP allocation. Because these pages cannot be >> defragmented until vfree(). > > kmalloc requires a resource that vmalloc does not: contiguous > pages. Therefore, given the same mix of pages (some groups of > contiguous pages, and a scattering of isolated single-page, or > too-small-to-satisfy-entire-alloc groups of pages, and the same > underlying page allocator, kmalloc *must* consume the more valuable > contiguous pages. However, vmalloc *may* consume those same pages. > > So, if you run kmalloc a bunch of times, with higher-order requests, > you *will* run out of contiguous pages (until more are freed up). If > you run vmalloc with the same initial conditions and the same > requests, you may not necessary use up those contiguous pages. > > It's true that there are benefits to doing a kmalloc-then-vmalloc, of > course: if the pages are available, it's faster and uses less > resources. Yes. I just don't think "less fragmentation" should be > listed as a benefit, because you can definitely cause *more* > fragmentation if you use up contiguous blocks unnecessarily. Yes, I agree that for some cases, kmalloc() will use more contiguous blocks, for example, non-movable pages are scattered all over the memory. But I still think in common cases, if defragement is enabled, and non-movable pages allocation is restricted to some memory area if possible, kmalloc() is better than vmalloc() as for fragmentation. Best Regards, Huang, Ying > -- > thanks, > john h