Re: [Xen-devel] [PATCH v5 21/22] xen/arm: Add a hypercall for device mmio mapping

[Stefano Stabellini <stefano.stabellini@xxxxxxxxxxxxx>]

On Fri, 4 Mar 2016, Jan Beulich wrote:
> >>> On 04.03.16 at 12:00, <roger.pau@xxxxxxxxxx> wrote:
> > On Fri, 4 Mar 2016, Jan Beulich wrote:
> >> >>> On 04.03.16 at 07:15, <zhaoshenglong@xxxxxxxxxx> wrote:
> >> Overall I wonder whether this wouldn't help PVH on x86 too,
> >> where we currently do some hackery to (not even completely)
> >> map MMIO into Dom0's p2m. In such a case perhaps
> >> map_dev_mmio_regions() should become a general per-arch
> >> function right away (declared in a common header and stubbed
> >> out in x86 code for now). Boris, Roger?
> > 
> > I though that we agreed that Xen will take care of doing all the MMIO 
> > mappings for HVMlite/PVHv2 guests:
> > 
> > http://lists.xenproject.org/archives/html/xen-devel/2016-02/msg01609.html 
> > 
> > My first idea was to use the MMIO device mapping hypercall (just like 
> > ARM) on the hardware domain, but I think we can get away without it on x86.
> 
> Oh, you're right (albeit that doesn't cover MMIO regions not
> represented by BARs, which will need taking care of).

This.

On ARM there can MMIO regions of non-PCI devices which need to be mapped
and can only be discovered via DSDT.


> Question
> back to the ARM folks then: Would such a model work for you
> too? (I'd really like to avoid having two different models, if we
> can avoid it.)

It doesn't cover non-PCI devices. But once you extend your model to deal
with non-PCI devices and MMIO regions not represented by BARs, then
potentually yes.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel