drm/amdgpu: Use optimal mtypes and PTE bits for Arcturus

For compute VRAM allocations on Arturus use the new RW mtype
for non-coherent local memory, CC mtype for coherent local
memory and PTE_SNOOPED bit for invalidating non-dirty cache
lines on remote XGMI mappings.
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Tested-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Shaoyun Liu <Shaoyun.Liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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

@@ -358,6 +358,7 @@ static int vm_update_pds(struct amdgpu_vm *vm, struct amdgpu_sync *sync)
 
 static uint64_t get_pte_flags(struct amdgpu_device *adev, struct kgd_mem *mem)
 {
+	struct amdgpu_device *bo_adev = amdgpu_ttm_adev(mem->bo->tbo.bdev);
 	bool coherent = mem->alloc_flags & ALLOC_MEM_FLAGS_COHERENT;
 	uint32_t mapping_flags;
 
@@ -367,8 +368,23 @@ static uint64_t get_pte_flags(struct amdgpu_device *adev, struct kgd_mem *mem)
 	if (mem->alloc_flags & ALLOC_MEM_FLAGS_EXECUTABLE)
 		mapping_flags |= AMDGPU_VM_PAGE_EXECUTABLE;
 
+	switch (adev->asic_type) {
+	case CHIP_ARCTURUS:
+		if (mem->alloc_flags & ALLOC_MEM_FLAGS_VRAM) {
+			if (bo_adev == adev)
+				mapping_flags |= coherent ?
+					AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW;
+			else
+				mapping_flags |= AMDGPU_VM_MTYPE_UC;
+		} else {
 			mapping_flags |= coherent ?
 				AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC;
+		}
+		break;
+	default:
+		mapping_flags |= coherent ?
+			AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC;
+	}
 
 	return amdgpu_gmc_get_pte_flags(adev, mapping_flags);
 }

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

@@ -1592,6 +1592,10 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
 		}
 		flags &= ~AMDGPU_PTE_VALID;
 	}
+	if (adev->asic_type == CHIP_ARCTURUS &&
+	    !(flags & AMDGPU_PTE_SYSTEM) &&
+	    mapping->bo_va->is_xgmi)
+		flags |= AMDGPU_PTE_SNOOPED;
 
 	trace_amdgpu_vm_bo_update(mapping);