Merge tag 'amd-drm-next-6.4-2023-04-14' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.4-2023-04-14:

amdgpu:
- S4 fixes for APUs
- GFX11 fixes
- Misc code cleanups
- DCN 3.2 fixes
- DCN 3.1.4 fixes
- FPO/FAMS work to improve display power savings
- DP fixes
- UMC 8.10 code cleanup
- SDMA v4 fix
- GPU clock counter fixes
- SMU 13 fixes
- Sdma v6 invalidation fix for preemption
- RAS fixes
- S0ix fix
- GC 9.4.3 updates

amdkfd:
- Fix user pointers with IOMMU
- Fix coherency flag handling

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230414204609.7942-1-alexander.deucher@amd.com

1 files changed, 31 insertions, 11 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index c87515210c4f..83a83ced2439 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -82,6 +82,25 @@ static bool kfd_mem_is_attached(struct amdgpu_vm *avm,
 	return false;
 }
 
+/**
+ * reuse_dmamap() - Check whether adev can share the original
+ * userptr BO
+ *
+ * If both adev and bo_adev are in direct mapping or
+ * in the same iommu group, they can share the original BO.
+ *
+ * @adev: Device to which can or cannot share the original BO
+ * @bo_adev: Device to which allocated BO belongs to
+ *
+ * Return: returns true if adev can share original userptr BO,
+ * false otherwise.
+ */
+static bool reuse_dmamap(struct amdgpu_device *adev, struct amdgpu_device *bo_adev)
+{
+	return (adev->ram_is_direct_mapped && bo_adev->ram_is_direct_mapped) ||
+			(adev->dev->iommu_group == bo_adev->dev->iommu_group);
+}
+
 /* Set memory usage limits. Current, limits are
  *  System (TTM + userptr) memory - 15/16th System RAM
  *  TTM memory - 3/8th System RAM
@@ -253,15 +272,19 @@ create_dmamap_sg_bo(struct amdgpu_device *adev,
 		 struct kgd_mem *mem, struct amdgpu_bo **bo_out)
 {
 	struct drm_gem_object *gem_obj;
-	int ret, align;
+	int ret;
+	uint64_t flags = 0;
 
 	ret = amdgpu_bo_reserve(mem->bo, false);
 	if (ret)
 		return ret;
 
-	align = 1;
-	ret = amdgpu_gem_object_create(adev, mem->bo->tbo.base.size, align,
-			AMDGPU_GEM_DOMAIN_CPU, AMDGPU_GEM_CREATE_PREEMPTIBLE,
+	if (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR)
+		flags |= mem->bo->flags & (AMDGPU_GEM_CREATE_COHERENT |
+					AMDGPU_GEM_CREATE_UNCACHED);
+
+	ret = amdgpu_gem_object_create(adev, mem->bo->tbo.base.size, 1,
+			AMDGPU_GEM_DOMAIN_CPU, AMDGPU_GEM_CREATE_PREEMPTIBLE | flags,
 			ttm_bo_type_sg, mem->bo->tbo.base.resv, &gem_obj);
 
 	amdgpu_bo_unreserve(mem->bo);
@@ -481,9 +504,6 @@ kfd_mem_dmamap_userptr(struct kgd_mem *mem,
 	if (unlikely(ret))
 		goto release_sg;
 
-	drm_prime_sg_to_dma_addr_array(ttm->sg, ttm->dma_address,
-				       ttm->num_pages);
-
 	amdgpu_bo_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_GTT);
 	ret = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
 	if (ret)
@@ -805,11 +825,11 @@ static int kfd_mem_attach(struct amdgpu_device *adev, struct kgd_mem *mem,
 			 va + bo_size, vm);
 
 		if ((adev == bo_adev && !(mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP)) ||
-		    (amdgpu_ttm_tt_get_usermm(mem->bo->tbo.ttm) && adev->ram_is_direct_mapped) ||
-		    same_hive) {
+		    (amdgpu_ttm_tt_get_usermm(mem->bo->tbo.ttm) && reuse_dmamap(adev, bo_adev)) ||
+			same_hive) {
 			/* Mappings on the local GPU, or VRAM mappings in the
-			 * local hive, or userptr mapping IOMMU direct map mode
-			 * share the original BO
+			 * local hive, or userptr mapping can reuse dma map
+			 * address space share the original BO
 			 */
 			attachment[i]->type = KFD_MEM_ATT_SHARED;
 			bo[i] = mem->bo;