[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH] intel_iommu: make sure its init before PCI dev
From: |
Paolo Bonzini |
Subject: |
Re: [Qemu-devel] [PATCH] intel_iommu: make sure its init before PCI dev |
Date: |
Thu, 9 Mar 2017 16:34:11 +0100 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.7.0 |
On 09/03/2017 16:29, Michael S. Tsirkin wrote:
>> If VFIO had an iommu property, to be used like "-device
>> intel_iommu,id=iommu0 -device vfio-pci,iommu=iommu0", It could tell VFIO
>> to use separate containers and also ensure proper ordering of command
>> line arguments.
>
> Lots of guests can't handle arbitrary stuff like this though.
> Normally IOMMU attaches to a bus and that's the only bus it can
> handle and that's the only IOMMU that can handle the devices
> on this bus.
>
> It's a fact of life that not all hardware dependencies are in a form of
> a bus-device relashionship, while modelling it as a bus internally might
> be feasible exposing that to users seems risky and confusing as that's
> our artifact.
I'm not saying it's a bus.
But it's pretty common to have multiple IOMMUs for one bus. A
relatively common case is to have one for the graphics card and one for
everything else, if I'm not mistaken.
Paolo