From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753340Ab1BGQFl (ORCPT ); Mon, 7 Feb 2011 11:05:41 -0500 Received: from mx1.redhat.com ([209.132.183.28]:65307 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751318Ab1BGQFk (ORCPT ); Mon, 7 Feb 2011 11:05:40 -0500 Date: Mon, 7 Feb 2011 11:04:51 -0500 From: Vivek Goyal To: "Haefliger, Juerg" Cc: LKML Subject: Re: Block IO throttling disabled in distros Message-ID: <20110207160451.GH7437@redhat.com> References: <59DD1BA8FD3C0F4C90771C18F2B5B53A653FF7284B@GVW0432EXB.americas.hpqcorp.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <59DD1BA8FD3C0F4C90771C18F2B5B53A653FF7284B@GVW0432EXB.americas.hpqcorp.net> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Feb 07, 2011 at 03:33:38PM +0000, Haefliger, Juerg wrote: > Hi, > > Both RHEL6 and Ubuntu 10.10 have block IO throttling disabled in their kernels. Does anybody know why that is? Are there performance or stability issues with the throttling controller or does it somehow negatively interfere with the rest of the kernel and/or system? Block IO throttling support was not even upstream at that time hence you don't see it. I have not heard about any stability issues so far. Of course more testing will reveal that. The only thing I have thinking is that should we limit the number of bios queued per IO context on the device. Currently there is no such limit and one can queue up as many bios as one want to. Thanks Vivek