From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751958AbcEMTPe (ORCPT ); Fri, 13 May 2016 15:15:34 -0400 Received: from mail-qg0-f51.google.com ([209.85.192.51]:36084 "EHLO mail-qg0-f51.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750929AbcEMTPc (ORCPT ); Fri, 13 May 2016 15:15:32 -0400 To: linux-kernel@vger.kernel.org From: Jeff Johnson Subject: 2.6 diskstats and 4096 sector block devices? Organization: Aeon Computing Message-ID: Date: Fri, 13 May 2016 12:15:30 -0700 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.11; rv:45.0) Gecko/20100101 Thunderbird/45.0 MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Greetings, I am analyzing output in /proc/diskstats for block devices that have a 4096B physical and logical sector size. 4KN or 4K-native disk drives. 8 33 sdc1 1289 0 10312 5893 3258606 53 1668254992 16804002 31 583987 16807101 Fields 3 and 7 are sectors read and written respectively. For operations on a 4KN device should fields 3 and 7 be interpreted as number of 4K sectors written (total transferred = N * 8) or is each 4K sector operation reflected as 8 ticks to N? I am expecting 1MB transfers to the device and using 'delta[field7] / delta[field5]' getting 128 which would point to diskstats handling everything in terms of 512B operations. To multiply by 8 or not to multiply by 8, that is the question.. :-) Thanks, --Jeff ------------------------------ Jeff Johnson Co-Founder Aeon Computing jeff.johnson "dot" aeoncomputing.com www.aeoncomputing.com 4170 Morena Boulevard, Suite D - San Diego, CA 92117 High-performance Computing / Lustre Filesystems / Scale-out Storage