Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753405AbcDUPLv (ORCPT ); Thu, 21 Apr 2016 11:11:51 -0400 Received: from mx1.redhat.com ([209.132.183.28]:48335 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752746AbcDUPLs (ORCPT ); Thu, 21 Apr 2016 11:11:48 -0400 Date: Thu, 21 Apr 2016 18:11:40 +0300 From: "Michael S. Tsirkin" To: Stefan Hajnoczi Cc: qemu-devel@nongnu.org, linux-kernel@vger.kernel.org, pbonzini@redhat.com, peterx@redhat.com, cornelia.huck@de.ibm.com, Kevin Wolf , Amit Shah , qemu-block@nongnu.org, Jason Wang , Alex Williamson , Andy Lutomirski , Christian Borntraeger , Wei Liu , David Woodhouse , virtualization@lists.linux-foundation.org, kvm@vger.kernel.org Subject: Re: [PATCH V2 RFC] fixup! virtio: convert to use DMA api Message-ID: <20160421181102-mutt-send-email-mst@redhat.com> References: <1461245745-6710-1-git-send-email-mst@redhat.com> <20160421145653.GB11814@stefanha-x1.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20160421145653.GB11814@stefanha-x1.localdomain> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 4071 Lines: 92 On Thu, Apr 21, 2016 at 03:56:53PM +0100, Stefan Hajnoczi wrote: > On Thu, Apr 21, 2016 at 04:43:45PM +0300, Michael S. Tsirkin wrote: > > This adds a flag to enable/disable bypassing the IOMMU by > > virtio devices. > > > > This is on top of patch > > http://article.gmane.org/gmane.comp.emulators.qemu/403467 > > virtio: convert to use DMA api > > > > Tested with patchset > > http://article.gmane.org/gmane.linux.kernel.virtualization/27545 > > virtio-pci: iommu support (note: bit number has been kept at 34 > > intentionally to match posted guest code. a non-RFC version will > > renumber bits to be contigious). > > > > changes from v1: > > drop PASSTHROUGH flag > > > > The interaction between virtio and DMA API is messy. > > > > On most systems with virtio, physical addresses match bus addresses, > > and it doesn't particularly matter whether we use the DMA API. > > > > On some systems, including Xen and any system with a physical device > > that speaks virtio behind a physical IOMMU, we must use the DMA API > > for virtio DMA to work at all. > > > > Add a feature bit to detect that: VIRTIO_F_IOMMU_PLATFORM. > > > > If not there, we preserve historic behavior and bypass the DMA > > API unless within Xen guest. This is actually required for > > systems, including SPARC and PPC64, where virtio-pci devices are > > enumerated as though they are behind an IOMMU, but the virtio host > > ignores the IOMMU, so we must either pretend that the IOMMU isn't > > there or somehow map everything as the identity. > > > > Re: non-virtio devices. > > > > It turns out that on old QEMU hosts, only emulated devices which were > > part of QEMU use the IOMMU. Should we want to bypass the IOMMU for such > > devices *only*, it would be rather easy to detect them by looking at > > subsystem vendor and device ID. Thus, no new interfaces are required > > except for virtio which always uses the same subsystem vendor and device ID. > > > > Signed-off-by: Michael S. Tsirkin > > --- > > include/hw/virtio/virtio-access.h | 3 ++- > > include/hw/virtio/virtio.h | 4 +++- > > include/standard-headers/linux/virtio_config.h | 2 ++ > > 3 files changed, 7 insertions(+), 2 deletions(-) > > > > diff --git a/include/hw/virtio/virtio-access.h b/include/hw/virtio/virtio-access.h > > index 967cc75..bb6f34e 100644 > > --- a/include/hw/virtio/virtio-access.h > > +++ b/include/hw/virtio/virtio-access.h > > @@ -23,7 +23,8 @@ static inline AddressSpace *virtio_get_dma_as(VirtIODevice *vdev) > > BusState *qbus = qdev_get_parent_bus(DEVICE(vdev)); > > VirtioBusClass *k = VIRTIO_BUS_GET_CLASS(qbus); > > > > - if (k->get_dma_as) { > > + if (virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM) && > > + k->get_dma_as) { > > return k->get_dma_as(qbus->parent); > > } > > return &address_space_memory; > > diff --git a/include/hw/virtio/virtio.h b/include/hw/virtio/virtio.h > > index b12faa9..44f3788 100644 > > --- a/include/hw/virtio/virtio.h > > +++ b/include/hw/virtio/virtio.h > > @@ -228,7 +228,9 @@ typedef struct VirtIORNGConf VirtIORNGConf; > > DEFINE_PROP_BIT64("notify_on_empty", _state, _field, \ > > VIRTIO_F_NOTIFY_ON_EMPTY, true), \ > > DEFINE_PROP_BIT64("any_layout", _state, _field, \ > > - VIRTIO_F_ANY_LAYOUT, true) > > + VIRTIO_F_ANY_LAYOUT, true), \ > > + DEFINE_PROP_BIT64("iommu_platform", _state, _field, \ > > + VIRTIO_F_IOMMU_PLATFORM, false) > > Looks like the impact of this patch is that users who relied on > k->get_dma_as today may now have to explicitly add iommu_platform=on. > Are there any such users (e.g. Xen)? No because upstream this is ignored. This is an incremental patch on top of Jason's one. > Instead of breaking the command-line for these users you could invert > the flag's meaning ("iommu_bypass=on") and set it in the SPARC/PPC > machine types. > > Stefan I hope I made it clear that there are no such users.