Re: [PATCH 0/3] iopmem : A block device for PCIe memory

[Dan Williams]

On Wed, Oct 26, 2016 at 1:24 AM, Haggai Eran <haggaie@xxxxxxxxxxxx> wrote:
[..]
>> I wonder if we could (ab)use a
>> software-defined 'pasid' as the requester id for a peer-to-peer
>> mapping that needs address translation.
> Why would you need that? Isn't it enough to map the peer-to-peer
> addresses correctly in the iommu driver?
>

You're right, we might already have enough...

We would just need to audit iommu drivers to undo any assumptions that
the page being mapped is always in host memory and apply any bus
address translations between source device and target device.