File arith_core.h¶

#ifndef DP_ARITH_CORE_H
#define DP_ARITH_CORE_H

#include "doppler/clib_common.h"

#ifdef __cplusplus
extern "C" {
#endif

/* Saturation helpers — two's complement, matching project convention. */

/* Clamp a 32-bit intermediate to the signed 16-bit (Q15) range. */
static inline int16_t sat16(int32_t x)
{
    return x > 32767 ? 32767 : x < -32768 ? -32768 : (int16_t)x;
}

/* Clamp a 16-bit intermediate to the signed 8-bit (Q8) range. */
static inline int8_t sat8(int16_t x)
{
    return x > 127 ? 127 : x < -128 ? -128 : (int8_t)x;
}

void dp_add_q15(const int16_t *a, size_t a_len, const int16_t *b, size_t b_len, int16_t *out);


void dp_sub_q15(const int16_t *a, size_t a_len, const int16_t *b, size_t b_len, int16_t *out);


void dp_mul_q15(const int16_t *a, size_t a_len, const int16_t *b, size_t b_len, int16_t *out);


int64_t dp_dot_q15(const int16_t *a, size_t a_len, const int16_t *b, size_t b_len);


void dp_shl_q15(const int16_t *a, size_t a_len, int16_t *out, int n);


void dp_shr_q15(const int16_t *a, size_t a_len, int16_t *out, int n);


void dp_add_q8(const int8_t *a, size_t a_len, const int8_t *b, size_t b_len, int8_t *out);


void dp_sub_q8(const int8_t *a, size_t a_len, const int8_t *b, size_t b_len, int8_t *out);


void dp_mul_q8(const int8_t *a, size_t a_len, const int8_t *b, size_t b_len, int8_t *out);


int32_t dp_dot_q8(const int8_t *a, size_t a_len, const int8_t *b, size_t b_len);


void dp_shl_q8(const int8_t *a, size_t a_len, int8_t *out, int n);


void dp_shr_q8(const int8_t *a, size_t a_len, int8_t *out, int n);


void dp_shl_i64(const int64_t *a, size_t a_len, int64_t *out, int n);


void dp_shr_i64(const int64_t *a, size_t a_len, int64_t *out, int n);


#ifdef __cplusplus
}
#endif

#endif /* ARITH_CORE_H */