drm/amdgpu: Place NPS mode request on unload

If a user has requested NPS mode switch, place the request through PSP
during unload of the driver. For devices which are part of a hive, all
requests are placed together. If one of them fails, revert back to the
current NPS mode.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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

@@ -2429,6 +2429,7 @@ amdgpu_pci_remove(struct pci_dev *pdev)
 	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	amdgpu_xcp_dev_unplug(adev);
+	amdgpu_gmc_prepare_nps_mode_change(adev);
 	drm_dev_unplug(dev);
 
 	if (adev->pm.rpm_mode != AMDGPU_RUNPM_NONE) {

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

@@ -1333,3 +1333,50 @@ int amdgpu_gmc_request_memory_partition(struct amdgpu_device *adev,
 
 	return psp_memory_partition(&adev->psp, nps_mode);
 }
+
+static inline bool amdgpu_gmc_need_nps_switch_req(struct amdgpu_device *adev,
+						  int req_nps_mode,
+						  int cur_nps_mode)
+{
+	return (((BIT(req_nps_mode) & adev->gmc.supported_nps_modes) ==
+			BIT(req_nps_mode)) &&
+		req_nps_mode != cur_nps_mode);
+}
+
+void amdgpu_gmc_prepare_nps_mode_change(struct amdgpu_device *adev)
+{
+	int req_nps_mode, cur_nps_mode, r;
+	struct amdgpu_hive_info *hive;
+
+	if (amdgpu_sriov_vf(adev) || !adev->gmc.supported_nps_modes ||
+	    !adev->gmc.gmc_funcs->request_mem_partition_mode)
+		return;
+
+	cur_nps_mode = adev->gmc.gmc_funcs->query_mem_partition_mode(adev);
+	hive = amdgpu_get_xgmi_hive(adev);
+	if (hive) {
+		req_nps_mode = atomic_read(&hive->requested_nps_mode);
+		if (!amdgpu_gmc_need_nps_switch_req(adev, req_nps_mode,
+						    cur_nps_mode)) {
+			amdgpu_put_xgmi_hive(hive);
+			return;
+		}
+		r = amdgpu_xgmi_request_nps_change(adev, hive, req_nps_mode);
+		amdgpu_put_xgmi_hive(hive);
+		goto out;
+	}
+
+	req_nps_mode = adev->gmc.requested_nps_mode;
+	if (!amdgpu_gmc_need_nps_switch_req(adev, req_nps_mode, cur_nps_mode))
+		return;
+
+	/* even if this fails, we should let driver unload w/o blocking */
+	r = adev->gmc.gmc_funcs->request_mem_partition_mode(adev, req_nps_mode);
+out:
+	if (r)
+		dev_err(adev->dev, "NPS mode change request failed\n");
+	else
+		dev_info(
+			adev->dev,
+			"NPS mode change request done, reload driver to complete the change\n");
+}

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

@@ -466,4 +466,6 @@ int amdgpu_gmc_get_nps_memranges(struct amdgpu_device *adev,
 
 int amdgpu_gmc_request_memory_partition(struct amdgpu_device *adev,
 					int nps_mode);
+void amdgpu_gmc_prepare_nps_mode_change(struct amdgpu_device *adev);
+
 #endif

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

@@ -1564,3 +1564,41 @@ int amdgpu_xgmi_reset_on_init(struct amdgpu_device *adev)
 
 	return 0;
 }
+
+int amdgpu_xgmi_request_nps_change(struct amdgpu_device *adev,
+				   struct amdgpu_hive_info *hive,
+				   int req_nps_mode)
+{
+	struct amdgpu_device *tmp_adev;
+	int cur_nps_mode, r;
+
+	/* This is expected to be called only during unload of driver. The
+	 * request needs to be placed only once for all devices in the hive. If
+	 * one of them fail, revert the request for previous successful devices.
+	 * After placing the request, make hive mode as UNKNOWN so that other
+	 * devices don't request anymore.
+	 */
+	mutex_lock(&hive->hive_lock);
+	list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
+		r = adev->gmc.gmc_funcs->request_mem_partition_mode(
+			tmp_adev, req_nps_mode);
+		if (r)
+			goto err;
+	}
+	/* Set to UNKNOWN so that other devices don't request anymore */
+	atomic_set(&hive->requested_nps_mode, UNKNOWN_MEMORY_PARTITION_MODE);
+
+	mutex_unlock(&hive->hive_lock);
+
+	return 0;
+err:
+	/* Request back current mode if one of the requests failed */
+	cur_nps_mode = adev->gmc.gmc_funcs->query_mem_partition_mode(tmp_adev);
+	list_for_each_entry_continue_reverse(tmp_adev, &hive->device_list,
+					     gmc.xgmi.head)
+		adev->gmc.gmc_funcs->request_mem_partition_mode(tmp_adev,
+								cur_nps_mode);
+	mutex_lock(&hive->hive_lock);
+
+	return r;
+}

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

@@ -79,4 +79,8 @@ static inline bool amdgpu_xgmi_same_hive(struct amdgpu_device *adev,
 int amdgpu_xgmi_ras_sw_init(struct amdgpu_device *adev);
 int amdgpu_xgmi_reset_on_init(struct amdgpu_device *adev);
 
+int amdgpu_xgmi_request_nps_change(struct amdgpu_device *adev,
+				   struct amdgpu_hive_info *hive,
+				   int req_nps_mode);
+
 #endif