Re: [PATCH kernel v11 18/34] vfio: powerpc/spapr/iommu/powernv/ioda2: Rework IOMMU ownership control

2015-06-01 Thread David Gibson
On Fri, May 29, 2015 at 06:44:42PM +1000, Alexey Kardashevskiy wrote: > This adds tce_iommu_take_ownership() and tce_iommu_release_ownership > which call in a loop iommu_take_ownership()/iommu_release_ownership() > for every table on the group. As there is just one now, no change in > behaviour is

Re: [PATCH kernel v11 18/34] vfio: powerpc/spapr/iommu/powernv/ioda2: Rework IOMMU ownership control

2015-06-01 Thread David Gibson
On Fri, May 29, 2015 at 06:44:42PM +1000, Alexey Kardashevskiy wrote: This adds tce_iommu_take_ownership() and tce_iommu_release_ownership which call in a loop iommu_take_ownership()/iommu_release_ownership() for every table on the group. As there is just one now, no change in behaviour is

[PATCH kernel v11 18/34] vfio: powerpc/spapr/iommu/powernv/ioda2: Rework IOMMU ownership control

2015-05-29 Thread Alexey Kardashevskiy
This adds tce_iommu_take_ownership() and tce_iommu_release_ownership which call in a loop iommu_take_ownership()/iommu_release_ownership() for every table on the group. As there is just one now, no change in behaviour is expected. At the moment the iommu_table struct has a set_bypass() which

[PATCH kernel v11 18/34] vfio: powerpc/spapr/iommu/powernv/ioda2: Rework IOMMU ownership control

2015-05-29 Thread Alexey Kardashevskiy
This adds tce_iommu_take_ownership() and tce_iommu_release_ownership which call in a loop iommu_take_ownership()/iommu_release_ownership() for every table on the group. As there is just one now, no change in behaviour is expected. At the moment the iommu_table struct has a set_bypass() which