On Tue, Jun 21, 2016 at 12:23:02PM +0200, Paolo Bonzini wrote:
>
>
> On 21/06/2016 08:16, David Gibson wrote:
> >> for (addr = 0; addr < memory_region_size(mr); addr += granularity) {
> >> iotlb = mr->iommu_ops->translate(mr, addr, is_write);
> >> if (iotlb.perm != IOMMU_NO
On 21/06/2016 08:16, David Gibson wrote:
>> for (addr = 0; addr < memory_region_size(mr); addr += granularity) {
>> iotlb = mr->iommu_ops->translate(mr, addr, is_write);
>> if (iotlb.perm != IOMMU_NONE) {
>
> Paolo, are you ok for me to make that small change and take this
On Tue, Jun 21, 2016 at 11:14:01AM +1000, Alexey Kardashevskiy wrote:
> Every IOMMU has some granularity which MemoryRegionIOMMUOps::translate
> uses when translating, however this information is not available outside
> the translate context for various checks.
>
> This adds a get_min_page_size ca
Every IOMMU has some granularity which MemoryRegionIOMMUOps::translate
uses when translating, however this information is not available outside
the translate context for various checks.
This adds a get_min_page_size callback to MemoryRegionIOMMUOps and
a wrapper for it so IOMMU users (such as VFIO
