From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-pa0-f50.google.com (mail-pa0-f50.google.com [209.85.220.50]) by kanga.kvack.org (Postfix) with ESMTP id 5A27B6B0253 for ; Wed, 4 Nov 2015 01:57:12 -0500 (EST) Received: by pabfh17 with SMTP id fh17so44057420pab.0 for ; Tue, 03 Nov 2015 22:57:12 -0800 (PST) Received: from mgwkm01.jp.fujitsu.com (mgwkm01.jp.fujitsu.com. [202.219.69.168]) by mx.google.com with ESMTPS id nz1si5859244pbb.112.2015.11.03.22.57.10 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 03 Nov 2015 22:57:11 -0800 (PST) Received: from m3051.s.css.fujitsu.com (m3051.s.css.fujitsu.com [10.134.21.209]) by kw-mxq.gw.nic.fujitsu.com (Postfix) with ESMTP id 69B0EAC0134 for ; Wed, 4 Nov 2015 15:57:06 +0900 (JST) Subject: Re: [PATCH] mm: Introduce kernelcore=reliable option References: <1444915942-15281-1-git-send-email-izumi.taku@jp.fujitsu.com> <3908561D78D1C84285E8C5FCA982C28F32B5A060@ORSMSX114.amr.corp.intel.com> <5628B427.3050403@jp.fujitsu.com> <3908561D78D1C84285E8C5FCA982C28F32B5C7AE@ORSMSX114.amr.corp.intel.com> <322B7BFA-08FE-4A8F-B54C-86901BDB7CBD@intel.com> <56330C0A.3060901@jp.fujitsu.com> <3908561D78D1C84285E8C5FCA982C28F32B64312@ORSMSX114.amr.corp.intel.com> From: Kamezawa Hiroyuki Message-ID: <5639AC34.9030603@jp.fujitsu.com> Date: Wed, 4 Nov 2015 15:56:52 +0900 MIME-Version: 1.0 In-Reply-To: <3908561D78D1C84285E8C5FCA982C28F32B64312@ORSMSX114.amr.corp.intel.com> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org List-ID: To: "Luck, Tony" , "Izumi, Taku" Cc: "linux-kernel@vger.kernel.org" , "linux-mm@kvack.org" , "qiuxishi@huawei.com" , "mel@csn.ul.ie" , "akpm@linux-foundation.org" , "Hansen, Dave" , "matt@codeblueprint.co.uk" On 2015/10/31 4:42, Luck, Tony wrote: >> If each memory controller has the same distance/latency, you (your firmware) don't need >> to allocate reliable memory per each memory controller. >> If distance is problem, another node should be allocated. >> >> ...is the behavior(splitting zone) really required ? > > It's useful from a memory bandwidth perspective to have allocations > spread across both memory controllers. Keeping a whole bunch of > Xeon cores fed needs all the bandwidth you can get. > Hmm. But physical address layout is not related to dual memory controller. I think reliable range can be contiguous by firmware... -Kame -- 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