From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754066Ab0E3NOM (ORCPT ); Sun, 30 May 2010 09:14:12 -0400 Received: from mx1.redhat.com ([209.132.183.28]:47120 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752529Ab0E3NOK (ORCPT ); Sun, 30 May 2010 09:14:10 -0400 Message-ID: <4C026497.8070901@redhat.com> Date: Sun, 30 May 2010 16:13:59 +0300 From: Avi Kivity User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.9) Gecko/20100330 Fedora/3.0.4-1.fc12 Thunderbird/3.0.4 MIME-Version: 1.0 To: "Michael S. Tsirkin" CC: Tom Lyon , linux-kernel@vger.kernel.org, kvm@vger.kernel.org, chrisw@sous-sol.org, joro@8bytes.org, hjk@linutronix.de, gregkh@suse.de, aafabbri@cisco.com, scofeldm@cisco.com Subject: Re: [PATCH] VFIO driver: Non-privileged user level PCI drivers References: <4c004cba.Z/2Hpd7reetFaFC5%pugs@cisco.com> <20100530121944.GH27611@redhat.com> <4C025999.7080706@redhat.com> <20100530124949.GI27611@redhat.com> <4C0261C1.9090204@redhat.com> <20100530130332.GM27611@redhat.com> In-Reply-To: <20100530130332.GM27611@redhat.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 05/30/2010 04:03 PM, Michael S. Tsirkin wrote: > > >>>>> IMO this was because this driver does two things: programming iommu and >>>>> handling interrupts. uio does interrupt handling. >>>>> We could have moved iommu / DMA programming to >>>>> a separate driver, and have uio work with it. >>>>> This would solve limitation of the current driver >>>>> that is needs an iommu domain per device. >>>>> >>>>> >>>>> >>>> How do we enforce security then? We need to ensure that unprivileged >>>> users can only use the device with an iommu. >>>> >>>> >>> Force assigning to iommu before we allow any other operation? >>> >>> >> That means the driver must be aware of the iommu. >> > The userspace driver? Yes. And It is a good thing to be explicit > there anyway, since this lets userspace map a non-contigious > virtual address list into a contiguous bus address range. > No, the kernel driver. It cannot allow userspace to enable bus mastering unless it knows the iommu is enabled for the device and remaps dma to user pages. -- error compiling committee.c: too many arguments to function