src/qs8-dwconv/gen/up2x9-minmax-fp32-scalar-lrint.c

// Auto-generated file. Do not edit!
//   Template: src/qs8-dwconv/unipass-scalar.c.in
//   Generator: tools/xngen
//
// Copyright 2021 Google LLC
//
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree.

#include <assert.h>
#include <math.h>

#include <xnnpack/dwconv.h>
#include <xnnpack/math.h>


void xnn_qs8_dwconv_minmax_fp32_ukernel_up2x9__scalar_lrint(
    size_t channels,
    size_t output_width,
    const int8_t** input,
    const void* weights,
    int8_t* output,
    size_t input_stride,
    size_t output_increment,
    size_t input_offset,
    const int8_t* zero,
    const union xnn_qs8_conv_minmax_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_DISABLE_TSAN
{
  assert(channels != 0);
  assert(output_width != 0);

  const float vscale = params->fp32_scalar_lrint.scale;
  const long voutput_min_less_zero_point = params->fp32_scalar_lrint.output_min_less_zero_point;
  const long voutput_max_less_zero_point = params->fp32_scalar_lrint.output_max_less_zero_point;
  const int32_t voutput_zero_point = params->fp32_scalar_lrint.output_zero_point;
  do {
    const int8_t* i0 = input[0];
    assert(i0 != NULL);
    if XNN_UNPREDICTABLE(i0 != zero) {
      i0 = (const int8_t*) ((uintptr_t) i0 + input_offset);
    }
    const int8_t* i1 = input[1];
    assert(i1 != NULL);
    if XNN_UNPREDICTABLE(i1 != zero) {
      i1 = (const int8_t*) ((uintptr_t) i1 + input_offset);
    }
    const int8_t* i2 = input[2];
    assert(i2 != NULL);
    if XNN_UNPREDICTABLE(i2 != zero) {
      i2 = (const int8_t*) ((uintptr_t) i2 + input_offset);
    }
    const int8_t* i3 = input[3];
    assert(i3 != NULL);
    if XNN_UNPREDICTABLE(i3 != zero) {
      i3 = (const int8_t*) ((uintptr_t) i3 + input_offset);
    }
    const int8_t* i4 = input[4];
    assert(i4 != NULL);
    if XNN_UNPREDICTABLE(i4 != zero) {
      i4 = (const int8_t*) ((uintptr_t) i4 + input_offset);
    }
    const int8_t* i5 = input[5];
    assert(i5 != NULL);
    if XNN_UNPREDICTABLE(i5 != zero) {
      i5 = (const int8_t*) ((uintptr_t) i5 + input_offset);
    }
    const int8_t* i6 = input[6];
    assert(i6 != NULL);
    if XNN_UNPREDICTABLE(i6 != zero) {
      i6 = (const int8_t*) ((uintptr_t) i6 + input_offset);
    }
    const int8_t* i7 = input[7];
    assert(i7 != NULL);
    if XNN_UNPREDICTABLE(i7 != zero) {
      i7 = (const int8_t*) ((uintptr_t) i7 + input_offset);
    }
    const int8_t* i8 = input[8];
    assert(i8 != NULL);
    if XNN_UNPREDICTABLE(i8 != zero) {
      i8 = (const int8_t*) ((uintptr_t) i8 + input_offset);
    }
    input = (const int8_t**) ((uintptr_t) input + input_stride);

    size_t c = channels;
    const void* w = weights;
    for (; c >= 2; c -= 2) {
      int32_t vacc0 = ((const int32_t*) w)[0];
      int32_t vacc1 = ((const int32_t*) w)[1];


      const int32_t vi0x0 = i0[0];
      const int32_t vi0x1 = i0[1];
      i0 += 2;

      const int32_t vk0x0 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[0];
      const int32_t vk0x1 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[1];

      vacc0 += vi0x0 * vk0x0;
      vacc1 += vi0x1 * vk0x1;

      const int32_t vi1x0 = i1[0];
      const int32_t vi1x1 = i1[1];
      i1 += 2;

      const int32_t vk1x0 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[2];
      const int32_t vk1x1 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[3];

      vacc0 += vi1x0 * vk1x0;
      vacc1 += vi1x1 * vk1x1;

      const int32_t vi2x0 = i2[0];
      const int32_t vi2x1 = i2[1];
      i2 += 2;

      const int32_t vk2x0 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[4];
      const int32_t vk2x1 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[5];

      vacc0 += vi2x0 * vk2x0;
      vacc1 += vi2x1 * vk2x1;

      const int32_t vi3x0 = i3[0];
      const int32_t vi3x1 = i3[1];
      i3 += 2;

      const int32_t vk3x0 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[6];
      const int32_t vk3x1 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[7];

      vacc0 += vi3x0 * vk3x0;
      vacc1 += vi3x1 * vk3x1;

      const int32_t vi4x0 = i4[0];
      const int32_t vi4x1 = i4[1];
      i4 += 2;

      const int32_t vk4x0 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[8];
      const int32_t vk4x1 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[9];

      vacc0 += vi4x0 * vk4x0;
      vacc1 += vi4x1 * vk4x1;

      const int32_t vi5x0 = i5[0];
      const int32_t vi5x1 = i5[1];
      i5 += 2;

      const int32_t vk5x0 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[10];
      const int32_t vk5x1 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[11];

      vacc0 += vi5x0 * vk5x0;
      vacc1 += vi5x1 * vk5x1;

      const int32_t vi6x0 = i6[0];
      const int32_t vi6x1 = i6[1];
      i6 += 2;

      const int32_t vk6x0 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[12];
      const int32_t vk6x1 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[13];

      vacc0 += vi6x0 * vk6x0;
      vacc1 += vi6x1 * vk6x1;

      const int32_t vi7x0 = i7[0];
      const int32_t vi7x1 = i7[1];
      i7 += 2;

      const int32_t vk7x0 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[14];
      const int32_t vk7x1 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[15];

      vacc0 += vi7x0 * vk7x0;
      vacc1 += vi7x1 * vk7x1;

      const int32_t vi8x0 = i8[0];
      const int32_t vi8x1 = i8[1];
      i8 += 2;

      const int32_t vk8x0 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[16];
      const int32_t vk8x1 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[17];

      vacc0 += vi8x0 * vk8x0;
      vacc1 += vi8x1 * vk8x1;

      w = (const void*) ((uintptr_t) w + 2 * sizeof(int32_t) + 18 * sizeof(int8_t));

      float vfpacc0 = (float) vacc0;
      float vfpacc1 = (float) vacc1;

      vfpacc0 *= vscale;
      vfpacc1 *= vscale;

      long vrndacc0 = lrintf(vfpacc0);
      long vrndacc1 = lrintf(vfpacc1);

      vrndacc0 = XNN_UNPREDICTABLE(vrndacc0 < voutput_min_less_zero_point) ? voutput_min_less_zero_point : vrndacc0;
      vrndacc1 = XNN_UNPREDICTABLE(vrndacc1 < voutput_min_less_zero_point) ? voutput_min_less_zero_point : vrndacc1;

      vrndacc0 = XNN_UNPREDICTABLE(vrndacc0 > voutput_max_less_zero_point) ? voutput_max_less_zero_point : vrndacc0;
      vrndacc1 = XNN_UNPREDICTABLE(vrndacc1 > voutput_max_less_zero_point) ? voutput_max_less_zero_point : vrndacc1;

      int32_t vout0 = (int32_t) vrndacc0 + voutput_zero_point;
      int32_t vout1 = (int32_t) vrndacc1 + voutput_zero_point;

      output[0] = (int8_t) vout0;
      output[1] = (int8_t) vout1;
      output += 2;
    }
    if XNN_UNLIKELY(c != 0) {
      int32_t vacc = *((const int32_t*) w);

      const int32_t vi0 = *i0;
      const int32_t vk0 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[0];
      vacc += vi0 * vk0;
      const int32_t vi1 = *i1;
      const int32_t vk1 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[2];
      vacc += vi1 * vk1;
      const int32_t vi2 = *i2;
      const int32_t vk2 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[4];
      vacc += vi2 * vk2;
      const int32_t vi3 = *i3;
      const int32_t vk3 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[6];
      vacc += vi3 * vk3;
      const int32_t vi4 = *i4;
      const int32_t vk4 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[8];
      vacc += vi4 * vk4;
      const int32_t vi5 = *i5;
      const int32_t vk5 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[10];
      vacc += vi5 * vk5;
      const int32_t vi6 = *i6;
      const int32_t vk6 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[12];
      vacc += vi6 * vk6;
      const int32_t vi7 = *i7;
      const int32_t vk7 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[14];
      vacc += vi7 * vk7;
      const int32_t vi8 = *i8;
      const int32_t vk8 = ((const int8_t*) ((uintptr_t) w + 2 * sizeof(int32_t)))[16];
      vacc += vi8 * vk8;

      const float vfpacc = (float) vacc * vscale;
      long vrndacc = lrintf(vfpacc);
      vrndacc = XNN_UNPREDICTABLE(vrndacc < voutput_min_less_zero_point) ? voutput_min_less_zero_point : vrndacc;
      vrndacc = XNN_UNPREDICTABLE(vrndacc > voutput_max_less_zero_point) ? voutput_max_less_zero_point : vrndacc;
      int32_t vout = (int32_t) vrndacc + voutput_zero_point;
      *output++ = (int8_t) vout;
    }

    output = (int8_t*) ((uintptr_t) output + output_increment);
  } while (--output_width != 0);
}