From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qt0-f196.google.com ([209.85.216.196]:36580 "EHLO mail-qt0-f196.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1764067AbdAIUIu (ORCPT ); Mon, 9 Jan 2017 15:08:50 -0500 Date: Mon, 9 Jan 2017 15:08:43 -0500 From: Tejun Heo To: Shaohua Li Cc: linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, kernel-team@fb.com, axboe@fb.com, vgoyal@redhat.com Subject: Re: [PATCH V5 08/17] blk-throttle: make throtl_slice tunable Message-ID: <20170109200843.GL12827@mtj.duckdns.org> References: <3341ed56034ecbcdaaaa63a87345ea7f16178339.1481833017.git.shli@fb.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <3341ed56034ecbcdaaaa63a87345ea7f16178339.1481833017.git.shli@fb.com> Sender: linux-block-owner@vger.kernel.org List-Id: linux-block@vger.kernel.org Hello, On Thu, Dec 15, 2016 at 12:32:59PM -0800, Shaohua Li wrote: > throtl_slice is important for blk-throttling. It's called slice > internally but it really is a time window blk-throttling samples data. > blk-throttling will make decision based on the samplings. An example is > bandwidth measurement. A cgroup's bandwidth is measured in the time > interval of throtl_slice. > > A small throtl_slice meanse cgroups have smoother throughput but burn > more CPUs. It has 100ms default value, which is not appropriate for all > disks. A fast SSD can dispatch a lot of IOs in 100ms. This patch makes > it tunable. > > Since throtl_slice isn't a time slice, the sysfs name > 'throttle_sample_time' reflects its character better. Generally looks good. I think some documentation would be great. Also, do we wannt set it to a lower number if blk_queue_nonrot()? Thanks. -- tejun