From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S964889AbbJREcT (ORCPT ); Sun, 18 Oct 2015 00:32:19 -0400 Received: from mail.linuxfoundation.org ([140.211.169.12]:45585 "EHLO mail.linuxfoundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932435AbbJREcP (ORCPT ); Sun, 18 Oct 2015 00:32:15 -0400 Date: Sat, 17 Oct 2015 21:32:14 -0700 From: Greg Kroah-Hartman To: Peter Hurley Cc: Jiri Slaby , linux-kernel@vger.kernel.org Subject: Re: [PATCH 2/2] tty: Use unbound workqueue for all input workers Message-ID: <20151018043214.GA9092@kroah.com> References: <1445114184-4022-1-git-send-email-peter@hurleysoftware.com> <1445114184-4022-2-git-send-email-peter@hurleysoftware.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1445114184-4022-2-git-send-email-peter@hurleysoftware.com> User-Agent: Mutt/1.5.24 (2015-08-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sat, Oct 17, 2015 at 04:36:24PM -0400, Peter Hurley wrote: > The commonly accepted wisdom that scheduling work on the same cpu > that handled interrupt i/o benefits from cache-locality is only > true if the cpu is idle (since bound kworkers are often the highest > vruntime and thus the lowest priority). > > Measurements of scheduling via the unbound queue show lowered > worst-case latency responses of up to 5x over bound workqueue, without > increase in average latency or throughput. > > pty i/o test measurements show >3x (!) reduced total running time; tests > previously taking ~8s now complete in <2.5s. Wow, nice job with this, that's unexpected. greg k-h