Re: [PATCH linux-next] iommu: add iommu for s390 platform

[Gerald Schaefer <gerald.schaefer-tA70FqPdS9bQT0dZR+AlfA@public.gmane.org>]

On Mon, 27 Oct 2014 17:25:02 +0100
Joerg Roedel <joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org> wrote:

> On Mon, Oct 27, 2014 at 03:32:01PM +0100, Gerald Schaefer wrote:
> > Not sure if I understood the concept of IOMMU domains right. But if
> > this is about having multiple devices in the same domain, so that
> > iommu_ops->map will establish the _same_ DMA mapping on _all_
> > registered devices, then this should be possible.
> 
> Yes, this is what domains are about. A domain describes a set of DMA
> mappings which can be assigned to multiple devices in parallel.
> 
> > We cannot have shared DMA tables because each device gets its own
> > DMA table allocated during device initialization.
> 
> Is there some hardware reason for this or is that just an
> implementation detail that can be changed. In other words, does the
> hardware allow to use the same DMA table for multiple devices?

Yes, the HW would allow shared DMA tables, but the implementation would
need some non-trivial changes. For example, we have a per-device spin_lock
for DMA table manipulations and the code in arch/s390/pci/pci_dma.c knows
nothing about IOMMU domains or shared DMA tables, it just implements a set
of dma_map_ops.

Of course this would also go horribly wrong if a device was already
in use (via the current dma_map_ops), but I guess using devices through
the IOMMU_API prevents using them otherwise?

> 
> > But we could just keep all devices from one domain in a list and
> > then call dma_update_trans() for all devices during
> > iommu_ops->map/unmap.
> 
> This sounds complicated. Note that a device can be assigned to a
> domain that already has existing mappings. In this case you need to
> make sure that the new device inherits these mappings (and destroy
> all old mappings for the device that possibly exist).
> 
> I think it is much easier to use the same DMA table for all devices
> in a domain, if the hardware allows that.

Yes, in this case, having one DMA table per domain and sharing it
between all devices in that domain sounds like a good idea. However,
I can't think of any use case for this, and Frank probably had a very
special use case in mind where this scenario doesn't appear, hence the
"one device per domain" restriction.

So, if having multiple devices per domain is a must, then we probably
need a thorough rewrite of the arch/s390/pci/pci_dma.c code.

Gerald