锐客网

  1. 首页 > it技术 > >

移动端AI|kpu.h文件研究（完善中）

algorithmimage移动端AIK210

kpu.h文件研究

/* Copyright 2018 Canaan Inc. * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * *http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ #ifndef _KPU_H #define _KPU_H#include #include #include "dmac.h"#define kpu_matmul_begin kpu_conv2d_outputtypedef int (*plic_irq_callback_t)(void *ctx); typedef struct { union { uint64_t reg; struct { uint64_t int_en : 1; uint64_t ram_flag : 1; uint64_t full_add : 1; uint64_t depth_wise_layer : 1; uint64_t reserved : 60; } data; } interrupt_enabe; union { uint64_t reg; struct { uint64_t image_src_addr : 15; uint64_t reserved0 : 17; uint64_t image_dst_addr : 15; uint64_t reserved1 : 17; } data; } image_addr; union { uint64_t reg; struct { uint64_t i_ch_num : 10; uint64_t reserved0 : 22; uint64_t o_ch_num : 10; uint64_t reserved1 : 6; uint64_t o_ch_num_coef : 10; uint64_t reserved2 : 6; } data; } image_channel_num; union { uint64_t reg; struct { uint64_t i_row_wid : 10; uint64_t i_col_high : 9; uint64_t reserved0 : 13; uint64_t o_row_wid : 10; uint64_t o_col_high : 9; uint64_t reserved1 : 13; } data; } image_size; union { uint64_t reg; struct { uint64_t kernel_type : 3; uint64_t pad_type : 1; uint64_t pool_type : 4; uint64_t first_stride : 1; uint64_t bypass_conv : 1; uint64_t load_para : 1; uint64_t reserved0 : 5; uint64_t dma_burst_size : 8; uint64_t pad_value : 8; uint64_t bwsx_base_addr : 32; } data; } kernel_pool_type_cfg; union { uint64_t reg; struct { uint64_t load_coor : 1; uint64_t load_time : 6; uint64_t reserved0 : 8; uint64_t para_size : 17; uint64_t para_start_addr : 32; } data; } kernel_load_cfg; union { uint64_t reg; struct { uint64_t coef_column_offset : 4; uint64_t coef_row_offset : 12; uint64_t reserved0 : 48; } data; } kernel_offset; union { uint64_t reg; struct { uint64_t channel_switch_addr : 15; uint64_t reserved : 1; uint64_t row_switch_addr : 4; uint64_t coef_size : 8; uint64_t coef_group : 3; uint64_t load_act : 1; uint64_t active_addr : 32; } data; } kernel_calc_type_cfg; union { uint64_t reg; struct { uint64_t wb_channel_switch_addr : 15; uint64_t reserved0 : 1; uint64_t wb_row_switch_addr : 4; uint64_t wb_group : 3; uint64_t reserved1 : 41; } data; } write_back_cfg; union { uint64_t reg; struct { uint64_t shr_w : 4; uint64_t shr_x : 4; uint64_t arg_w : 24; uint64_t arg_x : 24; uint64_t reserved0 : 8; } data; } conv_value; union { uint64_t reg; struct { uint64_t arg_add : 40; uint64_t reserved : 24; } data; } conv_value2; union { uint64_t reg; struct { uint64_t send_data_out : 1; uint64_t reserved : 15; uint64_t channel_byte_num : 16; uint64_t dma_total_byte : 32; } data; } dma_parameter; } kpu_layer_argument_t; typedef struct { union { uint64_t reg; struct { uint64_t shift_number : 8; uint64_t y_mul : 16; uint64_t x_start : 36; } data; } activate_para[16]; union { uint64_t reg; struct { uint8_t result_bias[8]; } data; } activate_para_bias0; union { uint64_t reg; struct { uint8_t result_bias[8]; } data; } activate_para_bias1; } kpu_activate_table_t; typedef struct { union { uint64_t reg; struct { uint64_t norm_mul : 24; uint64_t norm_add : 32; uint64_t norm_shift : 4; } data; } batchnorm; } kpu_batchnorm_argument_t; typedef struct { union { uint64_t reg; struct { uint16_t weight[9]; } data; } weights; } kpu_weights_kernel_16_3x3_t; typedef struct { uint64_t calc_done_int : 1; uint64_t layer_cfg_almost_empty_int : 1; uint64_t layer_cfg_almost_full_int : 1; uint64_t reserved : 61; } kpu_config_interrupt_t; typedef struct { uint64_t fifo_full_threshold : 4; uint64_t fifo_empty_threshold : 4; uint64_t reserved : 56; } kpu_config_fifo_threshold_t; typedef struct { uint64_t dma_fifo_flush_n : 1; uint64_t gs_fifo_flush_n : 1; uint64_t cfg_fifo_flush_n : 1; uint64_t cmd_fifo_flush_n : 1; uint64_t resp_fifo_flush_n : 1; uint64_t reserved : 59; } kpu_config_fifo_ctrl_t; typedef struct { uint64_t eight_bit_mode : 1; uint64_t reserved : 63; } kpu_config_eight_bit_mode_t; typedef struct { volatile uint64_t layer_argument_fifo; volatile union { uint64_t reg; kpu_config_interrupt_t data; } interrupt_status; volatile union { uint64_t reg; kpu_config_interrupt_t data; } interrupt_raw; volatile union { uint64_t reg; kpu_config_interrupt_t data; } interrupt_mask; volatile union { uint64_t reg; kpu_config_interrupt_t data; } interrupt_clear; volatile union { uint64_t reg; kpu_config_fifo_threshold_t data; } fifo_threshold; volatile uint64_t fifo_data_out; volatile union { uint64_t reg; kpu_config_fifo_ctrl_t data; } fifo_ctrl; volatile union { uint64_t reg; kpu_config_eight_bit_mode_t data; } eight_bit_mode; } kpu_config_t; typedef struct { kpu_layer_argument_t *layers; kpu_layer_argument_t *remain_layers; plic_irq_callback_t callback; void *ctx; uint64_t *src; uint64_t *dst; uint32_t src_length; uint32_t dst_length; uint32_t layers_length; uint32_t remain_layers_length; dmac_channel_number_t dma_ch; uint32_t eight_bit_mode; float output_scale; float output_bias; float input_scale; float input_bias; } kpu_task_t; typedef struct { uint32_t version; uint32_t flags; uint32_t arch; uint32_t layers_length; uint32_t max_start_address; uint32_t main_mem_usage; uint32_t output_count; } kpu_kmodel_header_t; typedef struct { uint32_t version; uint32_t flags; uint32_t layers_length; uint32_t max_start_address; uint32_t layers_argument_start; } kpu_model_header_t; typedef struct { uint32_t address; uint32_t size; } kpu_model_output_t; typedef enum { KL_INVALID = 0, KL_ADD, KL_QUANTIZED_ADD, KL_GLOBAL_MAX_POOL2D, KL_QUANTIZED_GLOBAL_MAX_POOL2D, KL_GLOBAL_AVERAGE_POOL2D, KL_QUANTIZED_GLOBAL_AVERAGE_POOL2D, KL_MAX_POOL2D, KL_QUANTIZED_MAX_POOL2D, KL_AVERAGE_POOL2D, KL_QUANTIZED_AVERAGE_POOL2D, KL_QUANTIZE, KL_DEQUANTIZE, KL_REQUANTIZE, KL_L2_NORMALIZATION, KL_SOFTMAX, KL_CONCAT, KL_QUANTIZED_CONCAT, KL_FULLY_CONNECTED, KL_QUANTIZED_FULLY_CONNECTED, KL_TENSORFLOW_FLATTEN, KL_QUANTIZED_TENSORFLOW_FLATTEN, KL_RESIZE_NEAREST_NEIGHBOR, KL_QUANTIZED_RESIZE_NEAREST_NEIGHBOR, KL_CHANNELWISE_DEQUANTIZE, KL_LOGISTIC, KL_K210_CONV = 10240, KL_K210_ADD_PADDING, KL_K210_REMOVE_PADDING, KL_K210_UPLOAD } kpu_model_layer_type_t; typedef struct { uint32_t type; uint32_t body_size; } kpu_model_layer_header_t; typedef enum { KLF_NONE = 0, KLF_MAIN_MEM_OUT = 1 } kpu_model_layer_flags_t; typedef enum { KLP_SAME = 0, KLP_VALID = 1 } kpu_model_padding_t; typedef enum { KLA_LINEAR = 0, KLA_RELU = 1, KLA_RELU6 = 2 } kpu_model_activation_t; typedef struct { float scale; float bias; } kpu_model_quant_param_t; typedef struct { uint32_t width; uint32_t height; uint32_t channels; } kpu_model_shape_t; typedef struct { uint32_t start; uint32_t size; } kpu_model_memory_range_t; typedef struct { uint32_t flags; uint32_t main_mem_out_address; uint32_t layer_offset; uint32_t weights_offset; uint32_t bn_offset; uint32_t act_offset; } kpu_model_conv_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_a_address; uint32_t main_mem_in_b_address; uint32_t main_mem_out_address; uint32_t count; } kpu_model_add_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_a_address; uint32_t main_mem_in_b_address; uint32_t main_mem_out_address; uint32_t count; int32_t in_a_offset; int32_t in_a_mul; int32_t in_a_shift; int32_t in_b_offset; int32_t in_b_mul; int32_t in_b_shift; int32_t out_offset; int32_t out_mul; int32_t out_shift; } kpu_model_quant_add_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t main_mem_out_address; uint32_t kernel_size; uint32_t channels; } kpu_model_gap2d_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t main_mem_out_address; kpu_model_shape_t in_shape; kpu_model_shape_t out_shape; uint32_t kernel_width; uint32_t kernel_height; uint32_t stride_width; uint32_t stride_height; uint32_t padding_width; uint32_t padding_height; } kpu_model_quant_max_pool2d_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t main_mem_out_address; kpu_model_shape_t in_shape; kpu_model_shape_t out_shape; uint32_t kernel_width; uint32_t kernel_height; uint32_t stride_width; uint32_t stride_height; uint32_t padding_width; uint32_t padding_height; kpu_model_activation_t act; } kpu_model_ave_pool2d_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t mem_out_address; uint32_t count; kpu_model_quant_param_t quant_param; } kpu_model_quantize_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t main_mem_out_address; uint32_t count; kpu_model_quant_param_t quant_param; } kpu_model_dequantize_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t main_mem_out_address; uint32_t count; uint8_t table[256]; } kpu_model_requantize_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t kpu_mem_out_address; uint32_t channels; } kpu_model_add_padding_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t main_mem_out_address; uint32_t channels; } kpu_model_remove_padding_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t kpu_mem_out_address; uint32_t width; uint32_t height; uint32_t channels; } kpu_model_upload_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t main_mem_out_address; uint32_t channels; } kpu_model_l2_norm_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t main_mem_out_address; uint32_t channels; } kpu_model_softmax_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_out_address; uint32_t input_count; kpu_model_memory_range_t inputs_mem[0]; } kpu_model_concat_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t main_mem_out_address; uint32_t in_channels; uint32_t out_channels; kpu_model_activation_t act; float weights[0]; } kpu_model_fully_connected_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t main_mem_out_address; kpu_model_shape_t shape; } kpu_model_tf_flatten_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t main_mem_out_address; kpu_model_shape_t in_shape; uint32_t out_width; uint32_t out_height; uint32_t align_corners; } kpu_model_resize_nearest_neighbor_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t main_mem_out_address; kpu_model_shape_t in_shape; uint32_t out_width; uint32_t out_height; uint32_t align_corners; } kpu_model_quant_resize_nearest_neighbor_layer_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t main_mem_out_address; uint32_t channels; uint32_t channel_size; kpu_model_quant_param_t quant_params[0]; } kpu_model_channelwise_dequant_argument_t; typedef struct { uint32_t flags; uint32_t main_mem_in_address; uint32_t main_mem_out_address; uint32_t channels; } kpu_model_logistic_layer_argument_t; typedef void (*kpu_done_callback_t)(void *userdata); typedef struct { int is_nncase; union { struct { const uint8_t *model_buffer; uint8_t *main_buffer; uint32_t output_count; const kpu_model_output_t *outputs; const kpu_model_layer_header_t *layer_headers; const uint8_t *body_start; uint32_t layers_length; volatile uint32_t current_layer; const uint8_t *volatile current_body; dmac_channel_number_t dma_ch; kpu_done_callback_t done_callback; void *userdata; }; struct { void* nncase_ctx; }; }; } kpu_model_context_t; typedef struct { uint32_t weigths_offset; uint32_t bn_offset; uint32_t act_offset; float input_scale; float input_bias; float output_scale; float output_bias; } kpu_model_layer_metadata_t; typedef struct _quantize_param { float scale; float bias; } quantize_param_t; extern volatile kpu_config_t *const kpu; /** * @briefModle complier init kpu handler * * @param[in]taskKpu handler * * @returnKpu handler */ extern kpu_task_t *kpu_task_init(kpu_task_t *task); /** * @briefKpu run for AI * * @param[in]taskKpu handler * @param[in]dma_chDMA for kpu * @param[in]srcThe picture data * @param[in]destThe result of kpu * @param[in]callbackThe callback of kpu * * @returnresult *- 0Success *- OtherFail.Kpu is busy. */ int kpu_run(kpu_task_t *task, dmac_channel_number_t dma_ch, const void *src, void *dest, plic_irq_callback_t callback); /** * @briefGet kpu result buf * * @param[in]taskKpu handler * * @returnKpu result buf */ uint8_t *kpu_get_output_buf(kpu_task_t *task); /** * @briefRelease kpu output buf * * @param[in]output_bufKpu output buf * */ void kpu_release_output_buf(uint8_t *output_buf); /** * @briefKpu run for AI * * @param[in]taskKpu handler * * @returnresult *- 0Success *- OtherFail.Kpu is busy. */ int kpu_start(kpu_task_t *task); /** * @briefInitialize kpu handler * * @param[in]taskKpu handler * * @returnresult *- 0Success *- OtherFail. */ int kpu_single_task_init(kpu_task_t *task); /** * @briefUninitialize kpu handler * * @param[in]taskKpu handler * * @returnresult *- 0Success *- OtherFail. */ int kpu_single_task_deinit(kpu_task_t *task); /** * @briefLoad kmodel and init kpu task * * @param[in]taskKpu handler * @param[in]bufferKmodel * @param[in]metaTest data * * @returnresult *- 0Success *- OtherFail. */ int kpu_model_load_from_buffer(kpu_task_t *task, uint8_t *buffer, kpu_model_layer_metadata_t **meta); /** * @briefKpu initialize * * @param[in]eight_bit_mode0:16bit mode1:8bit mode * @param[in]callbackCallback of kpu * @param[in]userdataData of callback * */ void kpu_init(int eight_bit_mode, plic_irq_callback_t callback, void *userdata); /** * @briefKpu input data by dma * * @param[in]layerKpu task layer * @param[in]srcImage data * @param[in]dma_chDmac channel * @param[in]callbackDmac complete callback * @param[in]userdataData of callback * */ void kpu_input_dma(const kpu_layer_argument_t *layer, const uint8_t *src, dmac_channel_number_t dma_ch, plic_irq_callback_t callback, void *userdata); /** * @briefKpu input data by cpu * * @param[in]layerKpu task layer * @param[in]srcImage data * @param[in]widthImage width * @param[in]heightImage heigth * @param[in]channelsColor channel, RGB is 3 * */ void kpu_input_with_padding(kpu_layer_argument_t *layer, const uint8_t *src, int width, int height, int channels); /** * @briefKpu run only one layer * * @param[in]layerKpu task layer * */ void kpu_conv2d(kpu_layer_argument_t *layer); /** * @briefKpu run only one layer then get the result by dma * * @param[in]layerKpu task layer * @param[in]dma_chDmac channel * @param[in]destResult * @param[in]callbackDmac complete callback * @param[in]userdataData of callback * */ void kpu_conv2d_output(kpu_layer_argument_t *layer, dmac_channel_number_t dma_ch, uint8_t *dest, plic_irq_callback_t callback, void *userdata); /** * @briefKpu pooling * * @param[in]srcSource * @param[in]src_paramSource param * @param[in]kernel_sizeKernel size, 7*7 is 49 * @param[in]channelsChannels * @param[in]destDest * @param[in]dest_paramDest param * */ void kpu_global_average_pool(const uint8_t *src, const quantize_param_t *src_param, int kernel_size, int channels, uint8_t *dest, const quantize_param_t *dest_param); /** * @briefKpu pooling * * @param[in]srcSource * @param[in]src_paramSource param * @param[in]kernel_sizeKernel size, 7*7 is 49 * @param[in]channelsChannels * @param[in]destDest * */ void kpu_global_average_pool_float(const uint8_t *src, const quantize_param_t *src_param, int kernel_size, int channels, float *dest); /** * @briefKpu fullly connected by cpu * * @param[in]srcSource * @param[in]weightsWeight * @param[in]biasesBiases * @param[in]destDest * @param[in]input_channelsInput channels * @param[in]output_channelsOutput channels * */ void kpu_fully_connected(const float *src, const float *weights, const float *biases, float *dest, int input_channels, int output_channels); /** * @briefKpu matrix multiplication * * @param[in]srcSource * @param[in]channelsChannels * @param[in]destDest * @param[in]dest_paramDest param * */ void kpu_matmul_end(const uint8_t *src, int channels, float *dest, const quantize_param_t *dest_param); /** * @briefKpu dequantize * * @param[in]srcSource * @param[in]src_paramSource param * @param[in]countDequantize count * @param[in]destDest * */ void kpu_dequantize(const uint8_t *src, const quantize_param_t *src_param, size_t count, float *dest); /** * @briefKpu load kmodel * * @param[in]ctxKmodel object * @param[in]bufferKmodel buffer * * @returnresult *- 0Success *- OtherFail. */ int kpu_load_kmodel(kpu_model_context_t *ctx, const uint8_t *buffer); /** * @briefKpu free kmodel buffer * * @param[in]ctxkmodel object * */ void kpu_model_free(kpu_model_context_t *ctx); /** * @briefKpu load kmodel * * @param[in]ctxKmodel object * @param[in]indexOutput index * @param[in]dataOutput data * @param[in]sizeOutput data size * * @returnresult *- 0Success *- OtherFail. */ int kpu_get_output(kpu_model_context_t *ctx, uint32_t index, uint8_t **data, size_t *size); /** * @briefKpu run kmodel * * @param[in]ctxKmodel object * @param[in]srcSource data * @param[in]dma_chDma channel * @param[in]done_callbackKpu complete callback * @param[in]userdataData of callback * * @returnresult *- 0Success *- OtherFail. */ int kpu_run_kmodel(kpu_model_context_t *ctx, const uint8_t *src, dmac_channel_number_t dma_ch, kpu_done_callback_t done_callback, void *userdata); #endif

【移动端AI|kpu.h文件研究（完善中）】

    推荐阅读


    上一篇：HPStream

    下一篇：与英语结一世的情缘1