From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: Problem doing pci passthrough of the network card without VT-d Date: Mon, 18 May 2009 00:17:11 +0300 Message-ID: <4A107ED7.1070100@redhat.com> References: <18C018878FB0244EB71B7FE328978A3224B01DFD@rrsmsx503.amr.corp.intel.com> <0199E0D51A61344794750DC57738F58E66BC980A61@GVW1118EXC.americas.hpqcorp.net> <18C018878FB0244EB71B7FE328978A3224B026A3@rrsmsx503.amr.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: "Fischer, Anna" , "kvm@vger.kernel.org" To: "Passera, Pablo R" Return-path: Received: from mx2.redhat.com ([66.187.237.31]:34329 "EHLO mx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753705AbZEQVRP (ORCPT ); Sun, 17 May 2009 17:17:15 -0400 In-Reply-To: <18C018878FB0244EB71B7FE328978A3224B026A3@rrsmsx503.amr.corp.intel.com> Sender: kvm-owner@vger.kernel.org List-ID: Passera, Pablo R wrote: > Hi Anna, > > >> Are you expecting this to work using the 1:1 mapping for direct device assignment? >> > Actually, I want to use the current qemu implementation for this. AFAIK from the code seems that qemu mmaps the device memory into the qemu pci subsystem memory space. Is this correct? > It does, but without 1:1 mapping or pvdma, we can't tell the device where to dma. >> In fact I don't get any errors on the guest side, so it is hard to track what is wrong. >> > In the guest I am getting an error in dmesg saying "Detected Tx Unit Hang" > You're lucky, you could have total memory corruption in the host. -- Do not meddle in the internals of kernels, for they are subtle and quick to panic.