From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [PATCH 4/7] ide: use the PCI memory access interface Date: Thu, 02 Sep 2010 19:07:31 +0300 Message-ID: <4C7FCBC3.1060006@redhat.com> References: <1283007298-10942-1-git-send-email-eduard.munteanu@linux360.ro> <1283007298-10942-5-git-send-email-eduard.munteanu@linux360.ro> <20100902051911.GA4273@redhat.com> <20100902091200.GB7319@localhost> <20100902095813.GB9085@redhat.com> <20100902150135.GA7136@localhost> <4C7FC1A9.3080800@redhat.com> <20100902153934.GE18182@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Eduard - Gabriel Munteanu , joro@8bytes.org, blauwirbel@gmail.com, paul@codesourcery.com, anthony@codemonkey.ws, av1474@comtv.ru, yamahata@valinux.co.jp, kvm@vger.kernel.org, qemu-devel@nongnu.org To: "Michael S. Tsirkin" Return-path: Received: from mx1.redhat.com ([209.132.183.28]:46708 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752498Ab0IBQII (ORCPT ); Thu, 2 Sep 2010 12:08:08 -0400 In-Reply-To: <20100902153934.GE18182@redhat.com> Sender: kvm-owner@vger.kernel.org List-ID: On 09/02/2010 06:39 PM, Michael S. Tsirkin wrote: > On Thu, Sep 02, 2010 at 06:24:25PM +0300, Avi Kivity wrote: >> That's not a practical long term solution. Eventually everything >> gets turned on. > That's why I wanted a simple !iommu check and fallback. > This way unless it's really used there's no overhead. > It's not very different from an indirect function call. Modern branch predictors store the target function address and supply it ahead of time. I've never seen a function call instruction in a profile. -- error compiling committee.c: too many arguments to function