Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751144AbcCHSHa (ORCPT ); Tue, 8 Mar 2016 13:07:30 -0500 Received: from mail.bmw-carit.de ([62.245.222.98]:44799 "EHLO mail.bmw-carit.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750802AbcCHSHW (ORCPT ); Tue, 8 Mar 2016 13:07:22 -0500 X-CTCH-RefID: str=0001.0A0C0201.56DF14D5.0234,ss=1,re=0.000,recu=0.000,reip=0.000,cl=1,cld=1,fgs=0 Subject: Re: [RFC v0] Use swait in completion To: Sebastian Andrzej Siewior , Daniel Wagner References: <1457452754-24029-1-git-send-email-wagi@monom.org> <20160308175206.GD21842@linutronix.de> CC: , , "Peter Zijlstra (Intel)" , Thomas Gleixner From: Daniel Wagner Message-ID: <56DF14D4.4010203@bmw-carit.de> Date: Tue, 8 Mar 2016 19:07:16 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.6.0 MIME-Version: 1.0 In-Reply-To: <20160308175206.GD21842@linutronix.de> Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 661 Lines: 14 On 03/08/2016 06:52 PM, Sebastian Andrzej Siewior wrote: > However I don't think if your DEFER flag solution is all that bad. I > have also the block-mq in -RT using swait and they perform wakes with > irqs-off. Not in -RT but mainline. So me might need something to make it > work properly. But if we defer the wakeup they might come at us and > complain about the latency… I intended to extend the test code to measure the latency diff as well. Just to get a feeling how bad it is. At least with a lot of waiters on the completion the current measurement indicate an improvement. I guess with real workload the things look quite different. cheers, daniel