From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ingo Molnar Subject: Re: [PATCH] kvm tools: Add MMIO coalescing support Date: Sat, 4 Jun 2011 16:46:29 +0200 Message-ID: <20110604144629.GA22189@elte.hu> References: <20110604093847.GB14524@elte.hu> <1307182441.7239.2.camel@lappy> <20110604101711.GB16292@elte.hu> <1307183318.7239.6.camel@lappy> <20110604103508.GE16292@elte.hu> <20110604104712.GG16292@elte.hu> <20110604112743.GA24146@elte.hu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Sasha Levin , penberg@kernel.org, kvm@vger.kernel.org, asias.hejun@gmail.com, gorcunov@gmail.com, prasadjoshi124@gmail.com To: Alexander Graf Return-path: Received: from mx3.mail.elte.hu ([157.181.1.138]:53045 "EHLO mx3.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756694Ab1FDOqj (ORCPT ); Sat, 4 Jun 2011 10:46:39 -0400 Content-Disposition: inline In-Reply-To: Sender: kvm-owner@vger.kernel.org List-ID: * Alexander Graf wrote: > So the simple rule is: don't register a coalesced MMIO region for a > region where latency matters. [...] So my first suspicion is confirmed. A quick look at Qemu sources shows that lots of drivers are using coalesced_mmio without being aware of the latency effects and only one seems to make use of qemu_flush_coalesced_mmio_buffer(). Drivers like hw/e1000.c sure look latency critical to me. So i maintain my initial opinion: this is a pretty dangerous 'optimization' that should be used with extreme care: i can tell it you with pretty good authority that latency problems are much more easy to introduce than to find and remove ... Thanks, Ingo