|
|
|
|
|
|
|
|
|
|
xen-devel
[Xen-devel] enable_ats_device() call site
Allen,
what is the reason for calling this from VT-d's domain_context_mapping()?
I neither undertsand why this is VT-d specific, nor why it needs to be
re-done with each device re-assignment.
I'm asking because this depends on MMCFG availability, and hence the
initial call from pci_add_device() (in the context of scan_pci_devices())
to iommu_add_device() may not result in this getting enabled, while on
the first Dom0-invoked pci_add_device() pdev->domain is already set
and hence iommu_add_device() doesn't get called at all. I'd therefore
like to pull this out into pci_add_device() (and call it, together with
pci_enable_acs(), after the conditional around iommu_add_device() -
should be safe as I view both enabling functions as idempotent).
Alternatively - why do we need scan_pci_devices() at all? We're
supposed to be getting the devices reported from Dom0 anyway.
Jan
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
<Prev in Thread] |
Current Thread |
[Next in Thread>
|
- [Xen-devel] enable_ats_device() call site,
Jan Beulich <=
|
|
|
|
|