drm/amdkfd: Fix ATS capablity was not reported correctly on some APUs

Because CRAT_CU_FLAGS_IOMMU_PRESENT was not set in some BIOS crat, we
need to workaround this.

For future compatibility, we also overwrite the bit in capability according
to the value of needs_iommu_device.

Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Yong Zhao <Yong.Zhao@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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

@@ -62,9 +62,20 @@ int kfd_iommu_device_init(struct kfd_dev *kfd)
 	struct amd_iommu_device_info iommu_info;
 	unsigned int pasid_limit;
 	int err;
+	struct kfd_topology_device *top_dev;
 
-	if (!kfd->device_info->needs_iommu_device)
+	top_dev = kfd_topology_device_by_id(kfd->id);
+
+	/*
+	 * Overwrite ATS capability according to needs_iommu_device to fix
+	 * potential missing corresponding bit in CRAT of BIOS.
+	 */
+	if (!kfd->device_info->needs_iommu_device) {
+		top_dev->node_props.capability &= ~HSA_CAP_ATS_PRESENT;
 		return 0;
+	}
+
+	top_dev->node_props.capability |= HSA_CAP_ATS_PRESENT;
 
 	iommu_info.flags = 0;
 	err = amd_iommu_device_info(kfd->pdev, &iommu_info);

drivers/gpu/drm/amd/amdkfd/kfd_priv.h

@@ -806,6 +806,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu);
 int kfd_topology_remove_device(struct kfd_dev *gpu);
 struct kfd_topology_device *kfd_topology_device_by_proximity_domain(
 						uint32_t proximity_domain);
+struct kfd_topology_device *kfd_topology_device_by_id(uint32_t gpu_id);
 struct kfd_dev *kfd_device_by_id(uint32_t gpu_id);
 struct kfd_dev *kfd_device_by_pci_dev(const struct pci_dev *pdev);
 int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_dev **kdev);

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

@@ -63,22 +63,33 @@ struct kfd_topology_device *kfd_topology_device_by_proximity_domain(
 	return device;
 }
 
-struct kfd_dev *kfd_device_by_id(uint32_t gpu_id)
+struct kfd_topology_device *kfd_topology_device_by_id(uint32_t gpu_id)
 {
-	struct kfd_topology_device *top_dev;
-	struct kfd_dev *device = NULL;
+	struct kfd_topology_device *top_dev = NULL;
+	struct kfd_topology_device *ret = NULL;
 
 	down_read(&topology_lock);
 
 	list_for_each_entry(top_dev, &topology_device_list, list)
 		if (top_dev->gpu_id == gpu_id) {
-			device = top_dev->gpu;
+			ret = top_dev;
 			break;
 		}
 
 	up_read(&topology_lock);
 
-	return device;
+	return ret;
+}
+
+struct kfd_dev *kfd_device_by_id(uint32_t gpu_id)
+{
+	struct kfd_topology_device *top_dev;
+
+	top_dev = kfd_topology_device_by_id(gpu_id);
+	if (!top_dev)
+		return NULL;
+
+	return top_dev->gpu;
 }
 
 struct kfd_dev *kfd_device_by_pci_dev(const struct pci_dev *pdev)