Kernel-4.14 is not using this dma override function so skip it
for K4.14.

Bug 200427376

Change-Id: I10f76ade31cd278df7c309440c991fa140cf43e0
Signed-off-by: Pritesh Raithatha <praithatha@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1774425
GVS: Gerrit_Virtual_Submit
Reviewed-by: Sachin Nikam <snikam@nvidia.com>
Tested-by: Sachin Nikam <snikam@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>

All physical address > 128 GB are treated as NVLINK addresses.

Whenever user specify DMA attribute with NVLINK, we will set bit-37 of
physical address so that it will be treated as NVLINK.

Bug 1964471

Change-Id: I781ef6350e0b7ae3d793bd3592667f96e6a48cfe
Signed-off-by: Pritesh Raithatha <praithatha@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1542855
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
Reviewed-by: Sachin Nikam <snikam@nvidia.com>
GVS: Gerrit_Virtual_Submit

Hw team confirm that 39-bit needs to set before SMMU, that is
in iova address and client can do that directly. No changes
required from SMMU/DMA side.

Removing the support that we added.

Bug 1866465

Change-Id: Ifcde76c50e0086ce65c94757681eeac534e3460a
Signed-off-by: Pritesh Raithatha <praithatha@nvidia.com>
Reviewed-on: http://git-master/r/1483859
Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Krishna Reddy <vdumpa@nvidia.com>

-Set 39th bit of pte when IOMMU_USE_BL_FORMAT specified in alloc_pte.
-Provide override functions for dma handle marshaling and qualifying
ioprot with 39th bit set when IOMMU_USE_BL_FORMAT specified and chip
is t19x.

Bug 1866465

Change-Id: Id35708364ba6db411d0c819d4ad7a8beeec7a432
Signed-off-by: Pritesh Raithatha <praithatha@nvidia.com>
Reviewed-on: http://git-master/r/1460454
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>