drm/amdkfd: Move local_mem_info to kfd_node

We need to track memory usage on a per partition basis. To do
that, store the local memory information in KFD node instead
of kfd device.

v2: squash in fix ("amdkfd: Use mem_id to access mem_partition info")
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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

@@ -428,14 +428,23 @@ uint32_t amdgpu_amdkfd_get_fw_version(struct amdgpu_device *adev,
 }
 
 void amdgpu_amdkfd_get_local_mem_info(struct amdgpu_device *adev,
-				      struct kfd_local_mem_info *mem_info)
+				      struct kfd_local_mem_info *mem_info,
+				      uint8_t xcp_id)
 {
 	memset(mem_info, 0, sizeof(*mem_info));
 
-	mem_info->local_mem_size_public = adev->gmc.visible_vram_size;
-	mem_info->local_mem_size_private = adev->gmc.real_vram_size -
+	if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) {
+		if (adev->gmc.real_vram_size == adev->gmc.visible_vram_size)
+			mem_info->local_mem_size_public =
+					KFD_XCP_MEMORY_SIZE(adev, xcp_id);
+		else
+			mem_info->local_mem_size_private =
+					KFD_XCP_MEMORY_SIZE(adev, xcp_id);
+	} else {
+		mem_info->local_mem_size_public = adev->gmc.visible_vram_size;
+		mem_info->local_mem_size_private = adev->gmc.real_vram_size -
 						adev->gmc.visible_vram_size;
-
+	}
 	mem_info->vram_width = adev->gmc.vram_width;
 
 	pr_debug("Address base: %pap public 0x%llx private 0x%llx\n",

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h

@@ -231,7 +231,8 @@ int amdgpu_amdkfd_remove_gws_from_process(void *info, void *mem);
 uint32_t amdgpu_amdkfd_get_fw_version(struct amdgpu_device *adev,
 				      enum kgd_engine_type type);
 void amdgpu_amdkfd_get_local_mem_info(struct amdgpu_device *adev,
-				      struct kfd_local_mem_info *mem_info);
+				      struct kfd_local_mem_info *mem_info,
+				      uint8_t xcp_id);
 uint64_t amdgpu_amdkfd_get_gpu_clock_counter(struct amdgpu_device *adev);
 
 uint32_t amdgpu_amdkfd_get_max_engine_clock_in_mhz(struct amdgpu_device *adev);
@@ -334,10 +335,11 @@ void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev,
 		((adev)->xcp_mgr && (xcp_id) >= 0 ?\
 		(adev)->xcp_mgr->xcp[(xcp_id)].mem_id : -1)
 
-#define KFD_XCP_MEMORY_SIZE(n) ((n)->adev->gmc.num_mem_partitions ?\
-		(n)->adev->gmc.mem_partitions[(n)->xcp->mem_id].size /\
-		(n)->adev->xcp_mgr->num_xcp_per_mem_partition :\
-		(n)->adev->gmc.real_vram_size)
+#define KFD_XCP_MEMORY_SIZE(adev, xcp_id)\
+		((adev)->gmc.num_mem_partitions && (xcp_id) >= 0 ?\
+		(adev)->gmc.mem_partitions[KFD_XCP_MEM_ID((adev), (xcp_id))].size /\
+		(adev)->xcp_mgr->num_xcp_per_mem_partition :\
+		(adev)->gmc.real_vram_size)
 
 #if IS_ENABLED(CONFIG_HSA_AMD)
 void amdgpu_amdkfd_gpuvm_init_mem_limits(void);

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

@@ -1023,11 +1023,12 @@ bool kfd_dev_is_large_bar(struct kfd_node *dev)
 	if (dev->kfd->use_iommu_v2)
 		return false;
 
-	if (dev->kfd->local_mem_info.local_mem_size_private == 0 &&
-	    dev->kfd->local_mem_info.local_mem_size_public > 0)
+	if (dev->local_mem_info.local_mem_size_private == 0 &&
+	    dev->local_mem_info.local_mem_size_public > 0)
 		return true;
 
-	if (dev->kfd->local_mem_info.local_mem_size_public == 0 && dev->kfd->adev->gmc.is_app_apu) {
+	if (dev->local_mem_info.local_mem_size_public == 0 &&
+	    dev->kfd->adev->gmc.is_app_apu) {
 		pr_debug("APP APU, Consider like a large bar system\n");
 		return true;
 	}

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

@@ -2191,7 +2191,7 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image,
 	 * report the total FB size (public+private) as a single
 	 * private heap.
 	 */
-	local_mem_info = kdev->kfd->local_mem_info;
+	local_mem_info = kdev->local_mem_info;
 	sub_type_hdr = (typeof(sub_type_hdr))((char *)sub_type_hdr +
 			sub_type_hdr->length);
 

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

@@ -726,7 +726,6 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 
 	svm_migrate_init(kfd->adev);
 
-	amdgpu_amdkfd_get_local_mem_info(kfd->adev, &kfd->local_mem_info);
 
 	dev_info(kfd_device, "Total number of KFD nodes to be created: %d\n",
 				kfd->num_nodes);
@@ -756,7 +755,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 		if (node->xcp) {
 			dev_info(kfd_device, "KFD node %d partition %d size %lldM\n",
 				node->node_id, node->xcp->mem_id,
-				KFD_XCP_MEMORY_SIZE(node) >> 20);
+				KFD_XCP_MEMORY_SIZE(node->adev, node->node_id) >> 20);
 		}
 
 		if (KFD_GC_VERSION(kfd) == IP_VERSION(9, 4, 3) &&
@@ -783,6 +782,10 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
 		}
 		node->max_proc_per_quantum = max_proc_per_quantum;
 		atomic_set(&node->sram_ecc_flag, 0);
+
+		amdgpu_amdkfd_get_local_mem_info(kfd->adev,
+					&node->local_mem_info, node->xcp->id);
+
 		/* Initialize the KFD node */
 		if (kfd_init_node(node)) {
 			dev_err(kfd_device, "Error initializing KFD node\n");

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

@@ -313,6 +313,8 @@ struct kfd_node {
 
 	unsigned int compute_vmid_bitmap;
 
+	struct kfd_local_mem_info local_mem_info;
+
 	struct kfd_dev *kfd;
 };
 
@@ -335,7 +337,6 @@ struct kfd_dev {
 					   */
 
 	struct kgd2kfd_shared_resources shared_resources;
-	struct kfd_local_mem_info local_mem_info;
 
 	const struct kfd2kgd_calls *kfd2kgd;
 	struct mutex doorbell_mutex;

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

@@ -1152,8 +1152,8 @@ static uint32_t kfd_generate_gpu_id(struct kfd_node *gpu)
 	if (!gpu)
 		return 0;
 
-	local_mem_size = gpu->kfd->local_mem_info.local_mem_size_private +
-			gpu->kfd->local_mem_info.local_mem_size_public;
+	local_mem_size = gpu->local_mem_info.local_mem_size_private +
+			gpu->local_mem_info.local_mem_size_public;
 	buf[0] = gpu->adev->pdev->devfn;
 	buf[1] = gpu->adev->pdev->subsystem_vendor |
 		(gpu->adev->pdev->subsystem_device << 16);
@@ -1234,7 +1234,8 @@ static void kfd_fill_mem_clk_max_info(struct kfd_topology_device *dev)
 	 * for APUs - If CRAT from ACPI reports more than one bank, then
 	 *	all the banks will report the same mem_clk_max information
 	 */
-	amdgpu_amdkfd_get_local_mem_info(dev->gpu->adev, &local_mem_info);
+	amdgpu_amdkfd_get_local_mem_info(dev->gpu->adev, &local_mem_info,
+					 dev->gpu->xcp->id);
 
 	list_for_each_entry(mem, &dev->mem_props, list)
 		mem->mem_clk_max = local_mem_info.mem_clk_max;