drm/amdgpu: Arcturus: MTYPE_NC for coarse-grain remote memory

MTYPE UC was used for a specific use case that ended up not being
implemented. Use NC for better performance for coarse-grained memory where
cache coherence during shader execution is not required.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Oak Zeng <Oak.Zeng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c

@@ -433,7 +433,8 @@ static uint64_t get_pte_flags(struct amdgpu_device *adev, struct kgd_mem *mem)
 				mapping_flags |= coherent ?
 					AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW;
 			else
-				mapping_flags |= AMDGPU_VM_MTYPE_UC;
+				mapping_flags |= coherent ?
+					AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC;
 		} else {
 			mapping_flags |= coherent ?
 				AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC;

drivers/gpu/drm/amd/amdkfd/kfd_svm.c

@@ -1022,7 +1022,8 @@ svm_range_get_pte_flags(struct amdgpu_device *adev, struct svm_range *prange)
 				mapping_flags |= coherent ?
 					AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW;
 			} else {
-				mapping_flags |= AMDGPU_VM_MTYPE_UC;
+				mapping_flags |= coherent ?
+					AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC;
 				if (amdgpu_xgmi_same_hive(adev, bo_adev))
 					snoop = true;
 			}