TB-RK3399ProD 上运行RKNN模型推理，rknn_input_set 处理时间过长

我使用的模型是小改过的YOLOX-s，根据官方提供的rknpu yolov5 demo做的预处理后，在 rknn_input_set 这一行处理耗时有时会超过 60 ms，模型推理时间只有约 75 ms，请教一下各位大神如何加速这个模型输入的操作？非常感谢！！ 部分代码如下：
系统：Linux debian10.toybrick 4.4.194 aarch64
API version：1.7.0
DRV version: 1.7.0

1. Loading RKNN Model ...
   
2. Initializing RKNN Envs ...
   
3. D RKNNAPI: ==============================================
   
4. D RKNNAPI: RKNN VERSION:
   
5. D RKNNAPI:   API: 1.7.0 (7880361 build: 2021-08-05 11:25:07)
   
6. D RKNNAPI:   DRV: 1.7.0 (7880361 build: 2021-08-16 16:16:21)
   
7. D RKNNAPI: ==============================================
   

复制代码*：RGA 与 DRM 已随模型一并初始化。

1. // RKNN model Input Settings
   
2.     printf("Processing Input ... \n");
   
3.     auto t_p0 = chrono::high_resolution_clock::now();
   
4.     memset(this->inputs, 0, sizeof(this->inputs));
   
5.     this->inputs[0].index = 0;
   
6.     this->inputs[0].type = RKNN_TENSOR_UINT8;
   
7.     this->inputs[0].size = this->width * this->height * this->channel;
   
8.     this->inputs[0].fmt = RKNN_TENSOR_NHWC;
   
9.     this->inputs[0].pass_through = 0;
   
10. 
    
11.     auto t_p1 = chrono::high_resolution_clock::now();
    
12.     float pre_time1 = chrono::duration<float, milli>(t_p1 - t_p0).count();
    
13.     cout << "Done. Preprocessing1 Time: " << pre_time1 << " ms." << endl;    // input initialization time
    
14. 
    
15.     // DRM alloc buffer
    
16.     this->drm_buf = drm_buf_alloc(&this->drm_ctx, this->drm_fd, img_width, img_height, this->channel * 8,
    
17.                                   &this->buf_fd, &this->ctx_handle, &this->actual_size);
    
18.     memcpy(this->drm_buf, input_data, img_width * img_height * this->channel);
    
19.     auto t_p2 = chrono::high_resolution_clock::now();
    
20.     float pre_time2 = chrono::duration<float, milli>(t_p2 - t_p1).count();
    
21.     cout << "Done. Preprocessing2 Time: " << pre_time2 << " ms." << endl;    // DRM Processing time
    
22. 
    
23.     // init rga context
    
24.     img_resize_slow(&this->rga_ctx, this->drm_buf, img_width, img_height, this->resize_buf, this->width, this->height);
    
25.     this->inputs[0].buf = this->resize_buf;
    
26.     auto t_p3 = chrono::high_resolution_clock::now();
    
27.     float pre_time3 = chrono::duration<float, milli>(t_p3 - t_p2).count();
    
28.     cout << "Done. Preprocessing3 Time: " << pre_time3 << " ms." << endl;    // RGA Resize (slow) time
    
29. 
    
30.     rknn_inputs_set(this->ctx, this->io_num.n_input, this->inputs);
    
31. 
    
32.     auto t_p4 = chrono::high_resolution_clock::now();
    
33.     float pre_time4 = chrono::duration<float, milli>(t_p4 - t_p3).count();
    
34.     cout << "Done. Preprocessing4 Time: " << pre_time4 << " ms." << endl;    // rknn_inputs_set time
    
35. 
    
36.     // Modle Output Settings
    
37.     memset(this->outputs, 0, sizeof(this->outputs));
    
38.     outputs[0].want_float = 0;
    
39. 
    
40.     printf("Inferencing ... \n");
    
41.     auto t_0 = chrono::high_resolution_clock::now();
    
42.     ret = rknn_run(this->ctx, NULL);
    
43.     ret = rknn_outputs_get(this->ctx, this->io_num.n_output, this->outputs, NULL);
    
44.     auto t_1 = chrono::high_resolution_clock::now();
    
45.     float inf_time = chrono::duration<float, milli>(t_1 - t_0).count();
    
46.     cout << "Done. RKNN Infer Time: " << inf_time << " ms." << endl;    // rknn inference time

复制代码结果输出：

1. Processing Input ...
   
2. Done. Preprocessing1 Time: 0.001458 ms.
   
3. Done. Preprocessing2 Time: 1.28944 ms.
   
4. Done. Preprocessing3 Time: 6.79251 ms.
   
5. Done. Preprocessing4 Time: 36.0082 ms.
   
6. Inferencing ...
   
7. Done. RKNN Infer Time: 76.1672 ms.

复制代码