Re: [PATCH] intel-iommu: Default to non-coherent for domainsunattached to iommus

[Kai Huang]

On Tue, Nov 15, 2011 at 12:54 PM, Chris Wright <chrisw@xxxxxxxxxxxx> wrote:
> * cody (mail.kai.huang@xxxxxxxxx) wrote:
>> On 11/11/2011 05:20 PM, David Woodhouse wrote:
>> >On Fri, 2011-11-11 at 17:08 -0800, Chris Wright wrote:
>> >>On a slightly separate, but performance related note...have you ever
>> >>tried using the hw queue?  Currently we only have a sw queue, but the
>> >>submission path for invalidations doesn't really queue (unless I missed
>> >>it).  It seems to pull from the software queue and submit/wait,
>> >>submit/wait...Seems simple enough to submit the whole queue and then
>> >>issue the wait.
>> >I have a feeling we trigger errata if we do that — although if we're
>> >only doing it for an emulated IOMMU that shouldn't be an issue.
>> >
>> What does the emulated IOMMU here? Does it mean the emulated IOMMU
>> exposed to guest VM?
>
> Yes.  So for KVM, the IOMMU emulation would be in QEMU.
>

What will emulated IOMMU be used for to the guest VM? To provide
another layer of address translation (in guest VM) that allows guest
VM to allocate large contiguous DVMA space for device? Does this mean
we need to keep something like *shadow page table* for the real IOMMU?

And this will be only used for direct IO, correct?

> thanks,
> -chris
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/