cnn.h

/******************************************************************************
 *
 * Copyright (C) 2022-2023 Maxim Integrated Products, Inc. (now owned by 
 * Analog Devices, Inc.),
 * Copyright (C) 2023-2024 Analog Devices, 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.
 *
 ******************************************************************************/

/*
 * This header file was automatically generated for the mnist-riscv network from a template.
 * Please do not edit; instead, edit the template and regenerate.
 */

#ifndef __CNN_H__
#define __CNN_H__

#include <stdint.h>
typedef int32_t q31_t;
typedef int16_t q15_t;

/* Return codes */
#define CNN_FAIL 0
#define CNN_OK 1

/*
  SUMMARY OF OPS
  Hardware: 10,883,968 ops (10,751,808 macc; 128,576 comp; 3,584 add; 0 mul; 0 bitwise)
    Layer 0: 470,400 ops (423,360 macc; 47,040 comp; 0 add; 0 mul; 0 bitwise)
    Layer 1: 8,356,800 ops (8,294,400 macc; 62,400 comp; 0 add; 0 mul; 0 bitwise)
    Layer 2: 1,954,304 ops (1,935,360 macc; 18,944 comp; 0 add; 0 mul; 0 bitwise)
    Layer 3: 100,544 ops (96,768 macc; 192 comp; 3,584 add; 0 mul; 0 bitwise)
    Layer 4: 1,920 ops (1,920 macc; 0 comp; 0 add; 0 mul; 0 bitwise)

  RESOURCE USAGE
  Weight memory: 71,148 bytes out of 442,368 bytes total (16%)
  Bias memory:   10 bytes out of 2,048 bytes total (0%)
*/

/* Number of outputs for this network */
#define CNN_NUM_OUTPUTS 10

/* Use this timer to time the inference */
#define CNN_INFERENCE_TIMER MXC_TMR0

/* Port pin actions used to signal that processing is active */

#define CNN_START LED_On(1)
#define CNN_COMPLETE LED_Off(1)
#define SYS_START LED_On(0)
#define SYS_COMPLETE LED_Off(0)

/* Run software SoftMax on unloaded data */
void softmax_q17p14_q15(const q31_t *vec_in, const uint16_t dim_vec, q15_t *p_out);
/* Shift the input, then calculate SoftMax */
void softmax_shift_q17p14_q15(q31_t *vec_in, const uint16_t dim_vec, uint8_t in_shift,
                              q15_t *p_out);

/* Stopwatch - holds the runtime when accelerator finishes */
extern volatile uint32_t cnn_time;

/* Custom memcopy routines used for weights and data */
void memcpy32(uint32_t *dst, const uint32_t *src, int n);
void memcpy32_const(uint32_t *dst, int n);

/* Enable clocks and power to accelerator, enable interrupt */
int cnn_enable(uint32_t clock_source, uint32_t clock_divider);

/* Disable clocks and power to accelerator */
int cnn_disable(void);

/* Perform minimum accelerator initialization so it can be configured */
int cnn_init(void);

/* Configure accelerator for the given network */
int cnn_configure(void);

/* Load accelerator weights */
int cnn_load_weights(void);

/* Verify accelerator weights (debug only) */
int cnn_verify_weights(void);

/* Load accelerator bias values (if needed) */
int cnn_load_bias(void);

/* Start accelerator processing */
int cnn_start(void);

/* Force stop accelerator */
int cnn_stop(void);

/* Continue accelerator after stop */
int cnn_continue(void);

/* Unload results from accelerator */
int cnn_unload(uint32_t *out_buf);

/* Turn on the boost circuit */
int cnn_boost_enable(mxc_gpio_regs_t *port, uint32_t pin);

/* Turn off the boost circuit */
int cnn_boost_disable(mxc_gpio_regs_t *port, uint32_t pin);

#endif // __CNN_H__