From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-pg0-f72.google.com (mail-pg0-f72.google.com [74.125.83.72]) by kanga.kvack.org (Postfix) with ESMTP id E21C46B0003 for ; Thu, 22 Feb 2018 16:53:46 -0500 (EST) Received: by mail-pg0-f72.google.com with SMTP id 202so2689334pgb.13 for ; Thu, 22 Feb 2018 13:53:46 -0800 (PST) Received: from NAM03-DM3-obe.outbound.protection.outlook.com (mail-dm3nam03on0104.outbound.protection.outlook.com. [104.47.41.104]) by mx.google.com with ESMTPS id k6-v6si645982pla.333.2018.02.22.13.53.45 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Thu, 22 Feb 2018 13:53:45 -0800 (PST) From: "Zi Yan" Subject: Re: [RFC 1/2] Protect larger order pages from breaking up Date: Thu, 22 Feb 2018 16:53:25 -0500 Message-ID: In-Reply-To: <68050f0f-14ca-d974-9cf4-19694a2244b9@schoebel-theuer.de> References: <20180216160110.641666320@linux.com> <20180216160121.519788537@linux.com> <20180219101935.cb3gnkbjimn5hbud@techsingularity.net> <68050f0f-14ca-d974-9cf4-19694a2244b9@schoebel-theuer.de> MIME-Version: 1.0 Content-Type: multipart/signed; boundary="=_MailMate_6EB96056-C7FF-4289-8A72-7260F5164555_="; micalg=pgp-sha512; protocol="application/pgp-signature" Sender: owner-linux-mm@kvack.org List-ID: To: Thomas Schoebel-Theuer Cc: Mel Gorman , Christoph Lameter , Matthew Wilcox , linux-mm@kvack.org, linux-rdma@vger.kernel.org, akpm@linux-foundation.org, andi@firstfloor.org, Rik van Riel , Michal Hocko , Guy Shattah , Anshuman Khandual , Michal Nazarewicz , Vlastimil Babka , David Nellans , Laura Abbott , Pavel Machek , Dave Hansen , Mike Kravetz This is an OpenPGP/MIME signed message (RFC 3156 and 4880). --=_MailMate_6EB96056-C7FF-4289-8A72-7260F5164555_= Content-Type: text/plain Content-Transfer-Encoding: quoted-printable On 22 Feb 2018, at 16:19, Thomas Schoebel-Theuer wrote: > > No need for a deep understanding of the theory of the memory fragmentat= ion problem. > > Also no need for adding anything to the boot commandline. Fragmentation= will typically occur only after some days or weeks or months of operatio= n, at least in all of the practical cases I have personally seen at 1&1 d= atacenters and their workloads. > > Please notice that fragmentation can be a very serious problem for oper= ations if you are hurt by it. It can seriously harm your business. And it= is _extremely_ specific to the actual workload, and to the hardware / ch= ipset / etc. This is addressed by the above method of determining the rig= ht values from _actual_ operations (not from speculation) and then memoiz= ing them. > > The attached patchset tries to be very simple, but in my practical expe= rience it is a very effective practical solution. > > When requested, I can post the mathematical theory behind the patch, or= I could give a presentation at some of the next conferences if I would b= e invited (or better give a practical explanation instead). But probably = nobody on these lists wants to deal with any theories. Hi Thomas, I am very interested in the theory behind your patch. Do you mind sharing= it? Is there any required math background before reading it? Is there any related pape= rs/articles I could also read? Thanks. -- Best Regards Yan Zi --=_MailMate_6EB96056-C7FF-4289-8A72-7260F5164555_= Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename=signature.asc Content-Type: application/pgp-signature; name=signature.asc -----BEGIN PGP SIGNATURE----- Comment: GPGTools - https://gpgtools.org iQEcBAEBCgAGBQJajzvVAAoJEEGLLxGcTqbMQ5QIALByhPmY9zak09Mprwb9H+Zk E/OT27Tw2DGTVGr3B88uuDL/SH6RxbawclIPhgxoQXA3iN+RJQ/jNK4/jf22mOx3 Og9pV3Vm2vMaZDRSMd6UyqsP3VAfVi5WNmayh57thwsjuk14vzLNGon8c7/mZTiS 3CIchF8yaymo24c0vf4l/y8yZbEkb/MMXDjnFPub6sm7ZFWW3tas+qjf6JKeWDom E9CW+oKHDMHATVn1IR9HYtn4t/3iwbjmBqEbMqD2vMGUQe1l8wbpvs/6bJnlL5tF dt+B4GixBDpQcGsYvHkRSCpGL4vPCWfhoeHmPXWftsdlohWExO86gXwFwwg4Jqk= =qRLg -----END PGP SIGNATURE----- --=_MailMate_6EB96056-C7FF-4289-8A72-7260F5164555_=-- -- 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