From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga02.intel.com (mga02.intel.com [134.134.136.20]) by dpdk.org (Postfix) with ESMTP id C7A289255 for ; Thu, 8 Sep 2016 06:18:35 +0200 (CEST) Received: from orsmga001.jf.intel.com ([10.7.209.18]) by orsmga101.jf.intel.com with ESMTP; 07 Sep 2016 21:18:34 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.30,298,1470726000"; d="scan'208";a="1026667793" Received: from orsmsx106.amr.corp.intel.com ([10.22.225.133]) by orsmga001.jf.intel.com with ESMTP; 07 Sep 2016 21:18:35 -0700 Received: from orsmsx155.amr.corp.intel.com (10.22.240.21) by ORSMSX106.amr.corp.intel.com (10.22.225.133) with Microsoft SMTP Server (TLS) id 14.3.248.2; Wed, 7 Sep 2016 21:18:33 -0700 Received: from rrsmsp801.amr.corp.intel.com (10.9.230.203) by ORSMSX155.amr.corp.intel.com (10.22.240.21) with Microsoft SMTP Server (TLS) id 14.3.248.2; Wed, 7 Sep 2016 21:18:33 -0700 Received: from Pickup by RRSMSP801.amr.corp.intel.com with Microsoft SMTP Server id 14.3.301.0; Wed, 7 Sep 2016 20:54:09 +0000 From: Jason Wang To: Alejandro Lucero , "Michael S. Tsirkin" CC: Thomas Monjalon , dev , "Xie, Huawei" , References: <1472798220-7121-1-git-send-email-jasowang@redhat.com> <1472798220-7121-2-git-send-email-jasowang@redhat.com> <1631660.hGMgJuvotI@xps13> <20160902202613-mutt-send-email-mst@kernel.org> In-Reply-To: Date: Mon, 5 Sep 2016 00:31:52 -0600 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Mailer: Microsoft Outlook 16.0 Thread-Index: AQKyy1v8v+X7HJqdklT5hP6lSkGSugKasjHaAlgEG7QDiBO7gwInRDEpAf7Y8E4= Subject: Re: [dpdk-dev] [PATCH 2/2] virtio: support IOMMU platform X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 08 Sep 2016 04:18:36 -0000 Message-ID: <20160905063152.G9CnCyxFBkgEuy4PhpgmlKNo3pC5VFv0rTcxldcP9hI@z> On 2016=E5=B9=B409=E6=9C=8804=E6=97=A5 16:08, Alejandro Lucero wrote: > I know RedHat is working on a vIOMMU so I guess this work is related=20 > to that effort, but it is a surprise virtio using IOMMU. I thought=20 > IOMMU just made sense when using SRIOV. My second guess is using IOMMU=20 > with virtio is a matter of security, but by other hand, virtio + IOMMU=20 > could imply serious performance degradation when multiple VMs are in use. We will use qemu vIOMMU for virito, so there's no such issue. > I'm talking about IOMMU contention, exactly about IOTLB contention. I thought device IOTLB (ATS) was just designed to solve this contention. > This performance issue is complex to describe or even analyze as there=20 > are several factors having an impact on it. For example, 1GB hugepages=20 > can avoid most of it and the same if TX & RX rings are not bigger than=20 > 256. So, my question: is RedHat aware of this potential IOMMU=20 > contention which can limit scalability? For virtio, we use vIOMMU per VM and implement a device IOTLB in vhost=20 side. Technically, it does not have such issue I think. Thanks > > On Fri, Sep 2, 2016 at 6:26 PM, Michael S. Tsirkin > wrote: > > On Fri, Sep 02, 2016 at 03:04:56PM +0200, Thomas Monjalon wrote: > > 2016-09-02 14:37, Jason Wang: > > > Virtio pmd doesn't support VFIO in the past since devices > bypass IOMMU > > > completely. But recently, the work of making virtio device > work with > > > IOMMU is near to complete. > > > > Good news! > > What are the requirements for Qemu and Linux version numbers please= ? > > I expect QEMU 2.8 and Linux 4.8 to have the support. > >