#include <assert.h>
#include <immintrin.h>
#include "xnnpack/common.h"
#include "xnnpack/dwconv.h"
#include "xnnpack/intrinsics-polyfill.h"
#include "xnnpack/math.h"
#include "xnnpack/transpose.h"
#include "xnnpack/unaligned.h"
#include "xnnpack/vcvt.h"
#include "xnnpack/vlrelu.h"
void xnn_f32_dwconv2d_chw_ukernel_3x3p1__ssse3_2x4_acc2(
size_t input_height,
size_t input_width,
const float* input,
const float* weights,
const float* zero,
float* output,
uint32_t padding_top,
const union xnn_f32_chw_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS
{ … }
void xnn_qs16_qs8_vcvt_ukernel__ssse3_u16(
size_t batch,
const int16_t* input,
int8_t* output,
const union xnn_qs16_qs8_cvt_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS
{ … }
void xnn_qs8_vcvt_ukernel__ssse3_u32(
size_t batch,
const int8_t* input,
int8_t* output,
const union xnn_qs8_cvt_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS
{ … }
void xnn_qs8_vlrelu_ukernel__ssse3_u32(
size_t batch,
const int8_t* input,
int8_t* output,
const union xnn_qs8_lrelu_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS
{ … }
void xnn_qu8_vcvt_ukernel__ssse3_u32(
size_t batch,
const uint8_t* input,
uint8_t* output,
const union xnn_qu8_cvt_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS
{ … }
void xnn_qu8_vlrelu_ukernel__ssse3_u32(
size_t batch,
const uint8_t* input,
uint8_t* output,
const union xnn_qu8_lrelu_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS
{ … }
void xnn_x24_transposec_ukernel__4x4_ssse3(
const void *input,
void * output,
size_t input_stride,
size_t output_stride,
size_t block_width,
size_t block_height,
const union xnn_x24_transpose_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS
{ … }