From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753575AbbJGK3F (ORCPT ); Wed, 7 Oct 2015 06:29:05 -0400 Received: from mail-lb0-f172.google.com ([209.85.217.172]:34035 "EHLO mail-lb0-f172.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753552AbbJGK3D (ORCPT ); Wed, 7 Oct 2015 06:29:03 -0400 Subject: Re: [PATCH v3 0/3] uio: add MSI/MSI-X support to uio_pci_generic driver To: "Michael S. Tsirkin" References: <1443991398-23761-1-git-send-email-vladz@cloudius-systems.com> <20151005221651-mutt-send-email-mst@redhat.com> <56138867.4000006@cloudius-systems.com> <20151006171940-mutt-send-email-mst@redhat.com> <5613E4BE.6060802@cloudius-systems.com> <20151006181358-mutt-send-email-mst@redhat.com> <5613F227.9040401@scylladb.com> <20151007084134-mutt-send-email-mst@redhat.com> Cc: Avi Kivity , Vlad Zolotarov , linux-kernel@vger.kernel.org, hjk@hansjkoch.de, corbet@lwn.net, gregkh@linuxfoundation.org, bruce.richardson@intel.com, gleb@cloudius-systems.com, stephen@networkplumber.org, alexander.duyck@gmail.com From: Avi Kivity Message-ID: <5614F3EA.5060208@scylladb.com> Date: Wed, 7 Oct 2015 13:28:58 +0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.3.0 MIME-Version: 1.0 In-Reply-To: <20151007084134-mutt-send-email-mst@redhat.com> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 10/07/2015 01:25 PM, Michael S. Tsirkin wrote: > On Tue, Oct 06, 2015 at 07:09:11PM +0300, Avi Kivity wrote: >> >> On 10/06/2015 06:15 PM, Michael S. Tsirkin wrote: >>>> While it is possible that userspace malfunctions and accidentally programs >>>> MSI incorrectly, the risk is dwarfed by the ability of userspace to program >>>> DMA incorrectly. >>> That seems to imply that for the upstream kernel this is not a valid usecase at all. >>> >> That is trivially incorrect, upstream pci_uio_generic is used with dpdk for >> years. > dpdk used to do polling for years. patch to use interrupts was posted in > june 2015. dpdk used interrupts long before that. > >> Are dpdk applications an invalid use case? > The way dpdk is using UIO/sysfs is borderline at best, and can't be used > to justify new interfaces. They have a more secure mode using VFIO. > That one's more reasonable. > Maybe this was not stressed enough times, but not all configurations have an iommu, or want to use one.