From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qk0-f174.google.com ([209.85.220.174]:34970 "EHLO mail-qk0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755486AbcKWVRE (ORCPT ); Wed, 23 Nov 2016 16:17:04 -0500 Date: Wed, 23 Nov 2016 16:17:02 -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 V4 07/15] blk-throttle: make throtl_slice tunable Message-ID: <20161123211702.GB11306@mtj.duckdns.org> References: <24137f881a149d30b98ef040367f26378128d4d2.1479161136.git.shli@fb.com> <20161122212715.GD17534@htj.duckdns.org> <20161122231824.GC14628@shli-mbp.local> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <20161122231824.GC14628@shli-mbp.local> Sender: linux-block-owner@vger.kernel.org List-Id: linux-block@vger.kernel.org Hello, On Tue, Nov 22, 2016 at 03:18:24PM -0800, Shaohua Li wrote: > Hmm, it's not a real 'time slice'. The name is a bit confusion. Maybe rename it > to 'throtl_interval' or 'throtl_sampling_time'? not sure. bandwidth and iops > are always in terms of a time interval we measure them. We can't say the > iops/bw for a single io. So this is really a tuable knob. Yeah, maybe using a more indicative name is better. However, even if we say that this is the sampling period, it's not clear how adjusting the knob would affect the behavior as that's not something clearly defined in blk-throtl's operation model. For contrast, compare it with the latency target, the implemented behavior might not succeed to follow the intended configuration perfectly but what the intention of the configuration is clear regardless. That said, if this needs to be a tunable knob, it's fine to have it, but let's at least try to document what the effects of changing the variable is. Thanks. -- tejun