Re: [RFC PATCH 0/2] VFIO no-iommu

2015-10-11 Thread Alex Williamson
On Sun, 2015-10-11 at 21:29 +0300, Michael S. Tsirkin wrote: > On Sun, Oct 11, 2015 at 09:28:09PM +0300, Michael S. Tsirkin wrote: > > On Fri, Oct 09, 2015 at 12:40:56PM -0600, Alex Williamson wrote: > > > Recent patches for UIO have been attempting to add MSI/X support, > > > which unfortunately i

Re: [RFC PATCH 0/2] VFIO no-iommu

2015-10-11 Thread Michael S. Tsirkin
On Sun, Oct 11, 2015 at 09:28:09PM +0300, Michael S. Tsirkin wrote: > On Fri, Oct 09, 2015 at 12:40:56PM -0600, Alex Williamson wrote: > > Recent patches for UIO have been attempting to add MSI/X support, > > which unfortunately implies DMA support, which users have been > > enabling anyway, but wa

Re: [RFC PATCH 0/2] VFIO no-iommu

2015-10-11 Thread Michael S. Tsirkin
On Fri, Oct 09, 2015 at 12:40:56PM -0600, Alex Williamson wrote: > Recent patches for UIO have been attempting to add MSI/X support, > which unfortunately implies DMA support, which users have been > enabling anyway, but was never intended for UIO. VFIO on the other > hand expects an IOMMU to prov

Re: [RFC PATCH 0/2] VFIO no-iommu

2015-10-11 Thread Alex Williamson
undation.org; > h...@hansjkoch.de; gre...@linuxfoundation.org > Subject: [RFC PATCH 0/2] VFIO no-iommu > > Recent patches for UIO have been attempting to add MSI/X support, which > unfortunately implies DMA support, which users have been enabling anyway, but > was never intended for UI

RE: [RFC PATCH 0/2] VFIO no-iommu

2015-10-11 Thread Varun Sethi
...@cloudius-systems.com; step...@networkplumber.org; vl...@cloudius-systems.com; io...@lists.linux-foundation.org; h...@hansjkoch.de; gre...@linuxfoundation.org Subject: [RFC PATCH 0/2] VFIO no-iommu Recent patches for UIO have been attempting to add MSI/X support, which unfortunately implies DMA

[RFC PATCH 0/2] VFIO no-iommu

2015-10-09 Thread Alex Williamson
Recent patches for UIO have been attempting to add MSI/X support, which unfortunately implies DMA support, which users have been enabling anyway, but was never intended for UIO. VFIO on the other hand expects an IOMMU to provide isolation of devices, but provides a much more complete device interf