|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
#include <assert.h> |
|
|
|
#include <arm_neon.h> |
|
|
|
#include <xnnpack/conv.h> |
|
#include <xnnpack/math.h> |
|
|
|
|
|
void xnn_f32_conv_hwc_ukernel_3x3s2p1c3x8__neon_2x1( |
|
size_t input_height, |
|
size_t input_width, |
|
size_t output_y_start, |
|
size_t output_y_end, |
|
const float* input, |
|
const float* zero, |
|
const float* weights, |
|
float* output, |
|
size_t input_padding_top, |
|
size_t output_channels, |
|
size_t output_height_stride, |
|
size_t output_width_stride, |
|
const union xnn_f32_minmax_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS |
|
{ |
|
assert(input_width != 0); |
|
assert(output_y_end > output_y_start); |
|
assert(input_padding_top <= 1); |
|
assert(output_channels != 0); |
|
|
|
const size_t input_height_stride = input_width * 3 * sizeof(float); |
|
const size_t input_width_decrement = input_width * 3 * sizeof(float); |
|
const size_t output_width = (input_width + 1) / 2; |
|
const size_t output_channel_decrement = output_width * output_width_stride - 8 * sizeof(float); |
|
const size_t output_height_increment = output_height_stride * 2 - round_up_po2(output_channels, 8) * sizeof(float); |
|
|
|
|
|
const float* i0 = (const float*) ((uintptr_t) input + |
|
input_height_stride * (output_y_start * 2 - input_padding_top)); |
|
const float* i1 = (const float*) ((uintptr_t) i0 + input_height_stride); |
|
const float* i2 = (const float*) ((uintptr_t) i1 + input_height_stride); |
|
const float* i3 = (const float*) ((uintptr_t) i2 + input_height_stride); |
|
const float* i4 = (const float*) ((uintptr_t) i3 + input_height_stride); |
|
float* o0 = (float*) ((uintptr_t) output + output_height_stride * output_y_start); |
|
float* o1 = (float*) ((uintptr_t) o0 + output_height_stride); |
|
|
|
if XNN_UNPREDICTABLE(output_y_start < input_padding_top) { |
|
i0 = zero; |
|
} |
|
|
|
|
|
for (size_t output_y = output_y_start; output_y < output_y_end; output_y += 2) { |
|
const size_t input_y2 = output_y * 2 + 2 - input_padding_top; |
|
const size_t input_y4 = input_y2 + 2; |
|
if XNN_UNPREDICTABLE(input_y2 > input_height) { |
|
i1 = zero; |
|
} |
|
if XNN_UNPREDICTABLE(input_y2 >= input_height) { |
|
i2 = zero; |
|
} |
|
if XNN_UNPREDICTABLE(input_y4 > input_height) { |
|
i3 = zero; |
|
} |
|
if XNN_UNPREDICTABLE(input_y4 >= input_height) { |
|
i4 = zero; |
|
} |
|
if XNN_UNPREDICTABLE(output_y + 2 > output_y_end) { |
|
o1 = o0; |
|
} |
|
|
|
const float* w = weights; |
|
size_t c = output_channels; |
|
do { |
|
|
|
float32x4_t vi0x0 = vmovq_n_f32(0.0f); |
|
float32x4_t vi1x0 = vmovq_n_f32(0.0f); |
|
float32x4_t vi2x0 = vmovq_n_f32(0.0f); |
|
float32x4_t vi3x0 = vmovq_n_f32(0.0f); |
|
float32x4_t vi4x0 = vmovq_n_f32(0.0f); |
|
|
|
size_t iw = input_width; |
|
for (; iw >= 2; iw -= 2) { |
|
float32x4_t vo0c0123 = vld1q_f32(w); |
|
float32x4_t vo0c4567 = vld1q_f32(w + 4); |
|
float32x4_t vo1c0123 = vo0c0123; |
|
float32x4_t vo1c4567 = vo0c4567; |
|
|
|
const float32x4_t vk00c0x0123 = vld1q_f32(w + 8); |
|
const float32x4_t vk00c0x4567 = vld1q_f32(w + 12); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk00c0x0123, vget_low_f32(vi0x0), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk00c0x0123, vget_low_f32(vi2x0), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk00c0x4567, vget_low_f32(vi0x0), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk00c0x4567, vget_low_f32(vi2x0), 1); |
|
|
|
const float32x4_t vk10c0x0123 = vld1q_f32(w + 16); |
|
const float32x4_t vk10c0x4567 = vld1q_f32(w + 20); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk10c0x0123, vget_low_f32(vi1x0), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk10c0x0123, vget_low_f32(vi3x0), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk10c0x4567, vget_low_f32(vi1x0), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk10c0x4567, vget_low_f32(vi3x0), 1); |
|
|
|
const float32x4_t vk20c0x0123 = vld1q_f32(w + 24); |
|
const float32x4_t vk20c0x4567 = vld1q_f32(w + 28); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk20c0x0123, vget_low_f32(vi2x0), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk20c0x0123, vget_low_f32(vi4x0), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk20c0x4567, vget_low_f32(vi2x0), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk20c0x4567, vget_low_f32(vi4x0), 1); |
|
|
|
const float32x4_t vk00c1x0123 = vld1q_f32(w + 32); |
|
const float32x4_t vk00c1x4567 = vld1q_f32(w + 36); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk00c1x0123, vget_high_f32(vi0x0), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk00c1x0123, vget_high_f32(vi2x0), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk00c1x4567, vget_high_f32(vi0x0), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk00c1x4567, vget_high_f32(vi2x0), 0); |
|
|
|
const float32x4_t vk10c1x0123 = vld1q_f32(w + 40); |
|
const float32x4_t vk10c1x4567 = vld1q_f32(w + 44); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk10c1x0123, vget_high_f32(vi1x0), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk10c1x0123, vget_high_f32(vi3x0), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk10c1x4567, vget_high_f32(vi1x0), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk10c1x4567, vget_high_f32(vi3x0), 0); |
|
|
|
const float32x4_t vk20c1x0123 = vld1q_f32(w + 48); |
|
const float32x4_t vk20c1x4567 = vld1q_f32(w + 52); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk20c1x0123, vget_high_f32(vi2x0), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk20c1x0123, vget_high_f32(vi4x0), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk20c1x4567, vget_high_f32(vi2x0), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk20c1x4567, vget_high_f32(vi4x0), 0); |
|
|
|
const float32x4_t vk00c2x0123 = vld1q_f32(w + 56); |
|
const float32x4_t vk00c2x4567 = vld1q_f32(w + 60); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk00c2x0123, vget_high_f32(vi0x0), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk00c2x0123, vget_high_f32(vi2x0), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk00c2x4567, vget_high_f32(vi0x0), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk00c2x4567, vget_high_f32(vi2x0), 1); |
|
|
|
const float32x4_t vk10c2x0123 = vld1q_f32(w + 64); |
|
const float32x4_t vk10c2x4567 = vld1q_f32(w + 68); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk10c2x0123, vget_high_f32(vi1x0), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk10c2x0123, vget_high_f32(vi3x0), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk10c2x4567, vget_high_f32(vi1x0), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk10c2x4567, vget_high_f32(vi3x0), 1); |
|
|
|
const float32x4_t vk20c2x0123 = vld1q_f32(w + 72); |
|
const float32x4_t vk20c2x4567 = vld1q_f32(w + 76); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk20c2x0123, vget_high_f32(vi2x0), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk20c2x0123, vget_high_f32(vi4x0), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk20c2x4567, vget_high_f32(vi2x0), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk20c2x4567, vget_high_f32(vi4x0), 1); |
|
|
|
|
|
const float32x4_t vi0x1 = vld1q_f32(i0); i0 += 4; |
|
const float32x4_t vi1x1 = vld1q_f32(i1); i1 += 4; |
|
const float32x4_t vi2x1 = vld1q_f32(i2); i2 += 4; |
|
const float32x4_t vi3x1 = vld1q_f32(i3); i3 += 4; |
|
const float32x4_t vi4x1 = vld1q_f32(i4); i4 += 4; |
|
|
|
const float32x4_t vk01c0x0123 = vld1q_f32(w + 80); |
|
const float32x4_t vk01c0x4567 = vld1q_f32(w + 84); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk01c0x0123, vget_low_f32(vi0x1), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk01c0x0123, vget_low_f32(vi2x1), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk01c0x4567, vget_low_f32(vi0x1), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk01c0x4567, vget_low_f32(vi2x1), 0); |
|
|
|
const float32x4_t vk11c0x0123 = vld1q_f32(w + 88); |
|
const float32x4_t vk11c0x4567 = vld1q_f32(w + 92); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk11c0x0123, vget_low_f32(vi1x1), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk11c0x0123, vget_low_f32(vi3x1), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk11c0x4567, vget_low_f32(vi1x1), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk11c0x4567, vget_low_f32(vi3x1), 0); |
|
|
|
const float32x4_t vk21c0x0123 = vld1q_f32(w + 96); |
|
const float32x4_t vk21c0x4567 = vld1q_f32(w + 100); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk21c0x0123, vget_low_f32(vi2x1), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk21c0x0123, vget_low_f32(vi4x1), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk21c0x4567, vget_low_f32(vi2x1), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk21c0x4567, vget_low_f32(vi4x1), 0); |
|
|
|
const float32x4_t vk01c1x0123 = vld1q_f32(w + 104); |
|
const float32x4_t vk01c1x4567 = vld1q_f32(w + 108); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk01c1x0123, vget_low_f32(vi0x1), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk01c1x0123, vget_low_f32(vi2x1), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk01c1x4567, vget_low_f32(vi0x1), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk01c1x4567, vget_low_f32(vi2x1), 1); |
|
|
|
const float32x4_t vk11c1x0123 = vld1q_f32(w + 112); |
|
const float32x4_t vk11c1x4567 = vld1q_f32(w + 116); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk11c1x0123, vget_low_f32(vi1x1), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk11c1x0123, vget_low_f32(vi3x1), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk11c1x4567, vget_low_f32(vi1x1), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk11c1x4567, vget_low_f32(vi3x1), 1); |
|
|
|
const float32x4_t vk21c1x0123 = vld1q_f32(w + 120); |
|
const float32x4_t vk21c1x4567 = vld1q_f32(w + 124); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk21c1x0123, vget_low_f32(vi2x1), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk21c1x0123, vget_low_f32(vi4x1), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk21c1x4567, vget_low_f32(vi2x1), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk21c1x4567, vget_low_f32(vi4x1), 1); |
|
|
|
const float32x4_t vk01c2x0123 = vld1q_f32(w + 128); |
|
const float32x4_t vk01c2x4567 = vld1q_f32(w + 132); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk01c2x0123, vget_high_f32(vi0x1), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk01c2x0123, vget_high_f32(vi2x1), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk01c2x4567, vget_high_f32(vi0x1), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk01c2x4567, vget_high_f32(vi2x1), 0); |
|
|
|
const float32x4_t vk11c2x0123 = vld1q_f32(w + 136); |
|
const float32x4_t vk11c2x4567 = vld1q_f32(w + 140); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk11c2x0123, vget_high_f32(vi1x1), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk11c2x0123, vget_high_f32(vi3x1), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk11c2x4567, vget_high_f32(vi1x1), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk11c2x4567, vget_high_f32(vi3x1), 0); |
|
|
|
const float32x4_t vk21c2x0123 = vld1q_f32(w + 144); |
|
const float32x4_t vk21c2x4567 = vld1q_f32(w + 148); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk21c2x0123, vget_high_f32(vi2x1), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk21c2x0123, vget_high_f32(vi4x1), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk21c2x4567, vget_high_f32(vi2x1), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk21c2x4567, vget_high_f32(vi4x1), 0); |
|
|
|
const float32x4_t vk02c0x0123 = vld1q_f32(w + 152); |
|
const float32x4_t vk02c0x4567 = vld1q_f32(w + 156); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk02c0x0123, vget_high_f32(vi0x1), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk02c0x0123, vget_high_f32(vi2x1), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk02c0x4567, vget_high_f32(vi0x1), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk02c0x4567, vget_high_f32(vi2x1), 1); |
|
|
|
const float32x4_t vk12c0x0123 = vld1q_f32(w + 160); |
|
const float32x4_t vk12c0x4567 = vld1q_f32(w + 164); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk12c0x0123, vget_high_f32(vi1x1), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk12c0x0123, vget_high_f32(vi3x1), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk12c0x4567, vget_high_f32(vi1x1), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk12c0x4567, vget_high_f32(vi3x1), 1); |
|
|
|
const float32x4_t vk22c0x0123 = vld1q_f32(w + 168); |
|
const float32x4_t vk22c0x4567 = vld1q_f32(w + 172); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk22c0x0123, vget_high_f32(vi2x1), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk22c0x0123, vget_high_f32(vi4x1), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk22c0x4567, vget_high_f32(vi2x1), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk22c0x4567, vget_high_f32(vi4x1), 1); |
|
|
|
|
|
const float32x2_t vi0x2 = vld1_f32(i0); i0 += 2; |
|
const float32x2_t vi1x2 = vld1_f32(i1); i1 += 2; |
|
const float32x2_t vi2x2 = vld1_f32(i2); i2 += 2; |
|
const float32x2_t vi3x2 = vld1_f32(i3); i3 += 2; |
|
const float32x2_t vi4x2 = vld1_f32(i4); i4 += 2; |
|
|
|
const float32x4_t vk02c1x0123 = vld1q_f32(w + 176); |
|
const float32x4_t vk02c1x4567 = vld1q_f32(w + 180); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk02c1x0123, vi0x2, 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk02c1x0123, vi2x2, 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk02c1x4567, vi0x2, 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk02c1x4567, vi2x2, 0); |
|
|
|
const float32x4_t vk12c1x0123 = vld1q_f32(w + 184); |
|
const float32x4_t vk12c1x4567 = vld1q_f32(w + 188); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk12c1x0123, vi1x2, 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk12c1x0123, vi3x2, 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk12c1x4567, vi1x2, 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk12c1x4567, vi3x2, 0); |
|
|
|
const float32x4_t vk22c1x0123 = vld1q_f32(w + 192); |
|
const float32x4_t vk22c1x4567 = vld1q_f32(w + 196); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk22c1x0123, vi2x2, 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk22c1x0123, vi4x2, 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk22c1x4567, vi2x2, 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk22c1x4567, vi4x2, 0); |
|
|
|
const float32x4_t vk02c2x0123 = vld1q_f32(w + 200); |
|
const float32x4_t vk02c2x4567 = vld1q_f32(w + 204); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk02c2x0123, vi0x2, 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk02c2x0123, vi2x2, 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk02c2x4567, vi0x2, 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk02c2x4567, vi2x2, 1); |
|
|
|
const float32x4_t vk12c2x0123 = vld1q_f32(w + 208); |
|
const float32x4_t vk12c2x4567 = vld1q_f32(w + 212); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk12c2x0123, vi1x2, 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk12c2x0123, vi3x2, 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk12c2x4567, vi1x2, 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk12c2x4567, vi3x2, 1); |
|
|
|
const float32x4_t vk22c2x0123 = vld1q_f32(w + 216); |
|
const float32x4_t vk22c2x4567 = vld1q_f32(w + 220); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk22c2x0123, vi2x2, 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk22c2x0123, vi4x2, 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk22c2x4567, vi2x2, 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk22c2x4567, vi4x2, 1); |
|
|
|
vi0x0 = vcombine_f32(vget_high_f32(vi0x1), vi0x2); |
|
vi1x0 = vcombine_f32(vget_high_f32(vi1x1), vi1x2); |
|
vi2x0 = vcombine_f32(vget_high_f32(vi2x1), vi2x2); |
|
vi3x0 = vcombine_f32(vget_high_f32(vi3x1), vi3x2); |
|
vi4x0 = vcombine_f32(vget_high_f32(vi4x1), vi4x2); |
|
|
|
const float32x4_t vmin = vld1q_dup_f32(¶ms->scalar.min); |
|
const float32x4_t vmax = vld1q_dup_f32(¶ms->scalar.max); |
|
|
|
vo0c0123 = vmaxq_f32(vo0c0123, vmin); |
|
vo1c0123 = vmaxq_f32(vo1c0123, vmin); |
|
vo0c4567 = vmaxq_f32(vo0c4567, vmin); |
|
vo1c4567 = vmaxq_f32(vo1c4567, vmin); |
|
|
|
vo0c0123 = vminq_f32(vo0c0123, vmax); |
|
vo1c0123 = vminq_f32(vo1c0123, vmax); |
|
vo0c4567 = vminq_f32(vo0c4567, vmax); |
|
vo1c4567 = vminq_f32(vo1c4567, vmax); |
|
|
|
if XNN_LIKELY(c >= 8) { |
|
vst1q_f32(o1, vo1c0123); |
|
vst1q_f32(o1 + 4, vo1c4567); |
|
o1 = (float*) ((uintptr_t) o1 + output_width_stride); |
|
vst1q_f32(o0, vo0c0123); |
|
vst1q_f32(o0 + 4, vo0c4567); |
|
o0 = (float*) ((uintptr_t) o0 + output_width_stride); |
|
} else { |
|
float* o0_tmp = o0; |
|
float* o1_tmp = o1; |
|
if (c & 4) { |
|
vst1q_f32(o1_tmp, vo1c0123); o1_tmp += 4; |
|
vo1c0123 = vo1c4567; |
|
vst1q_f32(o0_tmp, vo0c0123); o0_tmp += 4; |
|
vo0c0123 = vo0c4567; |
|
} |
|
float32x2_t vo0c01 = vget_low_f32(vo0c0123); |
|
float32x2_t vo1c01 = vget_low_f32(vo1c0123); |
|
if (c & 2) { |
|
vst1_f32(o1_tmp, vo1c01); o1_tmp += 2; |
|
vo1c01 = vget_high_f32(vo1c0123); |
|
vst1_f32(o0_tmp, vo0c01); o0_tmp += 2; |
|
vo0c01 = vget_high_f32(vo0c0123); |
|
} |
|
if (c & 1) { |
|
vst1_lane_f32(o1_tmp, vo1c01, 0); |
|
vst1_lane_f32(o0_tmp, vo0c01, 0); |
|
} |
|
|
|
o0 = (float*) ((uintptr_t) o0 + output_width_stride); |
|
o1 = (float*) ((uintptr_t) o1 + output_width_stride); |
|
} |
|
} |
|
assert(iw < 2); |
|
if XNN_UNLIKELY(iw & 1) { |
|
float32x4_t vo0c0123 = vld1q_f32(w); |
|
float32x4_t vo0c4567 = vld1q_f32(w + 4); |
|
float32x4_t vo1c0123 = vo0c0123; |
|
float32x4_t vo1c4567 = vo0c4567; |
|
|
|
const float32x4_t vk00c0x0123 = vld1q_f32(w + 8); |
|
const float32x4_t vk00c0x4567 = vld1q_f32(w + 12); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk00c0x0123, vget_low_f32(vi0x0), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk00c0x0123, vget_low_f32(vi2x0), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk00c0x4567, vget_low_f32(vi0x0), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk00c0x4567, vget_low_f32(vi2x0), 1); |
|
|
|
const float32x4_t vk10c0x0123 = vld1q_f32(w + 16); |
|
const float32x4_t vk10c0x4567 = vld1q_f32(w + 20); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk10c0x0123, vget_low_f32(vi1x0), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk10c0x0123, vget_low_f32(vi3x0), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk10c0x4567, vget_low_f32(vi1x0), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk10c0x4567, vget_low_f32(vi3x0), 1); |
|
|
|
const float32x4_t vk20c0x0123 = vld1q_f32(w + 24); |
|
const float32x4_t vk20c0x4567 = vld1q_f32(w + 28); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk20c0x0123, vget_low_f32(vi2x0), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk20c0x0123, vget_low_f32(vi4x0), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk20c0x4567, vget_low_f32(vi2x0), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk20c0x4567, vget_low_f32(vi4x0), 1); |
|
|
|
const float32x4_t vk00c1x0123 = vld1q_f32(w + 32); |
|
const float32x4_t vk00c1x4567 = vld1q_f32(w + 36); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk00c1x0123, vget_high_f32(vi0x0), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk00c1x0123, vget_high_f32(vi2x0), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk00c1x4567, vget_high_f32(vi0x0), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk00c1x4567, vget_high_f32(vi2x0), 0); |
|
|
|
const float32x4_t vk10c1x0123 = vld1q_f32(w + 40); |
|
const float32x4_t vk10c1x4567 = vld1q_f32(w + 44); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk10c1x0123, vget_high_f32(vi1x0), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk10c1x0123, vget_high_f32(vi3x0), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk10c1x4567, vget_high_f32(vi1x0), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk10c1x4567, vget_high_f32(vi3x0), 0); |
|
|
|
const float32x4_t vk20c1x0123 = vld1q_f32(w + 48); |
|
const float32x4_t vk20c1x4567 = vld1q_f32(w + 52); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk20c1x0123, vget_high_f32(vi2x0), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk20c1x0123, vget_high_f32(vi4x0), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk20c1x4567, vget_high_f32(vi2x0), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk20c1x4567, vget_high_f32(vi4x0), 0); |
|
|
|
const float32x4_t vk00c2x0123 = vld1q_f32(w + 56); |
|
const float32x4_t vk00c2x4567 = vld1q_f32(w + 60); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk00c2x0123, vget_high_f32(vi0x0), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk00c2x0123, vget_high_f32(vi2x0), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk00c2x4567, vget_high_f32(vi0x0), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk00c2x4567, vget_high_f32(vi2x0), 1); |
|
|
|
const float32x4_t vk10c2x0123 = vld1q_f32(w + 64); |
|
const float32x4_t vk10c2x4567 = vld1q_f32(w + 68); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk10c2x0123, vget_high_f32(vi1x0), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk10c2x0123, vget_high_f32(vi3x0), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk10c2x4567, vget_high_f32(vi1x0), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk10c2x4567, vget_high_f32(vi3x0), 1); |
|
|
|
const float32x4_t vk20c2x0123 = vld1q_f32(w + 72); |
|
const float32x4_t vk20c2x4567 = vld1q_f32(w + 76); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk20c2x0123, vget_high_f32(vi2x0), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk20c2x0123, vget_high_f32(vi4x0), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk20c2x4567, vget_high_f32(vi2x0), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk20c2x4567, vget_high_f32(vi4x0), 1); |
|
|
|
|
|
const float32x4_t vi0x1 = vld1q_f32(i0); i0 += 3; |
|
const float32x4_t vi1x1 = vld1q_f32(i1); i1 += 3; |
|
const float32x4_t vi2x1 = vld1q_f32(i2); i2 += 3; |
|
const float32x4_t vi3x1 = vld1q_f32(i3); i3 += 3; |
|
const float32x4_t vi4x1 = vld1q_f32(i4); i4 += 3; |
|
|
|
const float32x4_t vk01c0x0123 = vld1q_f32(w + 80); |
|
const float32x4_t vk01c0x4567 = vld1q_f32(w + 84); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk01c0x0123, vget_low_f32(vi0x1), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk01c0x0123, vget_low_f32(vi2x1), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk01c0x4567, vget_low_f32(vi0x1), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk01c0x4567, vget_low_f32(vi2x1), 0); |
|
|
|
const float32x4_t vk11c0x0123 = vld1q_f32(w + 88); |
|
const float32x4_t vk11c0x4567 = vld1q_f32(w + 92); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk11c0x0123, vget_low_f32(vi1x1), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk11c0x0123, vget_low_f32(vi3x1), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk11c0x4567, vget_low_f32(vi1x1), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk11c0x4567, vget_low_f32(vi3x1), 0); |
|
|
|
const float32x4_t vk21c0x0123 = vld1q_f32(w + 96); |
|
const float32x4_t vk21c0x4567 = vld1q_f32(w + 100); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk21c0x0123, vget_low_f32(vi2x1), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk21c0x0123, vget_low_f32(vi4x1), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk21c0x4567, vget_low_f32(vi2x1), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk21c0x4567, vget_low_f32(vi4x1), 0); |
|
|
|
const float32x4_t vk01c1x0123 = vld1q_f32(w + 104); |
|
const float32x4_t vk01c1x4567 = vld1q_f32(w + 108); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk01c1x0123, vget_low_f32(vi0x1), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk01c1x0123, vget_low_f32(vi2x1), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk01c1x4567, vget_low_f32(vi0x1), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk01c1x4567, vget_low_f32(vi2x1), 1); |
|
|
|
const float32x4_t vk11c1x0123 = vld1q_f32(w + 112); |
|
const float32x4_t vk11c1x4567 = vld1q_f32(w + 116); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk11c1x0123, vget_low_f32(vi1x1), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk11c1x0123, vget_low_f32(vi3x1), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk11c1x4567, vget_low_f32(vi1x1), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk11c1x4567, vget_low_f32(vi3x1), 1); |
|
|
|
const float32x4_t vk21c1x0123 = vld1q_f32(w + 120); |
|
const float32x4_t vk21c1x4567 = vld1q_f32(w + 124); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk21c1x0123, vget_low_f32(vi2x1), 1); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk21c1x0123, vget_low_f32(vi4x1), 1); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk21c1x4567, vget_low_f32(vi2x1), 1); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk21c1x4567, vget_low_f32(vi4x1), 1); |
|
|
|
const float32x4_t vk01c2x0123 = vld1q_f32(w + 128); |
|
const float32x4_t vk01c2x4567 = vld1q_f32(w + 132); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk01c2x0123, vget_high_f32(vi0x1), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk01c2x0123, vget_high_f32(vi2x1), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk01c2x4567, vget_high_f32(vi0x1), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk01c2x4567, vget_high_f32(vi2x1), 0); |
|
|
|
const float32x4_t vk11c2x0123 = vld1q_f32(w + 136); |
|
const float32x4_t vk11c2x4567 = vld1q_f32(w + 140); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk11c2x0123, vget_high_f32(vi1x1), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk11c2x0123, vget_high_f32(vi3x1), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk11c2x4567, vget_high_f32(vi1x1), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk11c2x4567, vget_high_f32(vi3x1), 0); |
|
|
|
const float32x4_t vk21c2x0123 = vld1q_f32(w + 144); |
|
const float32x4_t vk21c2x4567 = vld1q_f32(w + 148); |
|
|
|
vo0c0123 = vmlaq_lane_f32(vo0c0123, vk21c2x0123, vget_high_f32(vi2x1), 0); |
|
vo1c0123 = vmlaq_lane_f32(vo1c0123, vk21c2x0123, vget_high_f32(vi4x1), 0); |
|
vo0c4567 = vmlaq_lane_f32(vo0c4567, vk21c2x4567, vget_high_f32(vi2x1), 0); |
|
vo1c4567 = vmlaq_lane_f32(vo1c4567, vk21c2x4567, vget_high_f32(vi4x1), 0); |
|
|
|
const float32x4_t vmin = vld1q_dup_f32(¶ms->scalar.min); |
|
const float32x4_t vmax = vld1q_dup_f32(¶ms->scalar.max); |
|
|
|
vo0c0123 = vmaxq_f32(vo0c0123, vmin); |
|
vo1c0123 = vmaxq_f32(vo1c0123, vmin); |
|
vo0c4567 = vmaxq_f32(vo0c4567, vmin); |
|
vo1c4567 = vmaxq_f32(vo1c4567, vmin); |
|
|
|
vo0c0123 = vminq_f32(vo0c0123, vmax); |
|
vo1c0123 = vminq_f32(vo1c0123, vmax); |
|
vo0c4567 = vminq_f32(vo0c4567, vmax); |
|
vo1c4567 = vminq_f32(vo1c4567, vmax); |
|
|
|
if XNN_LIKELY(c >= 8) { |
|
vst1q_f32(o1, vo1c0123); |
|
vst1q_f32(o1 + 4, vo1c4567); |
|
o1 = (float*) ((uintptr_t) o1 + output_width_stride); |
|
vst1q_f32(o0, vo0c0123); |
|
vst1q_f32(o0 + 4, vo0c4567); |
|
o0 = (float*) ((uintptr_t) o0 + output_width_stride); |
|
} else { |
|
float* o0_tmp = o0; |
|
float* o1_tmp = o1; |
|
if (c & 4) { |
|
vst1q_f32(o1_tmp, vo1c0123); o1_tmp += 4; |
|
vo1c0123 = vo1c4567; |
|
vst1q_f32(o0_tmp, vo0c0123); o0_tmp += 4; |
|
vo0c0123 = vo0c4567; |
|
} |
|
float32x2_t vo0c01 = vget_low_f32(vo0c0123); |
|
float32x2_t vo1c01 = vget_low_f32(vo1c0123); |
|
if (c & 2) { |
|
vst1_f32(o1_tmp, vo1c01); o1_tmp += 2; |
|
vo1c01 = vget_high_f32(vo1c0123); |
|
vst1_f32(o0_tmp, vo0c01); o0_tmp += 2; |
|
vo0c01 = vget_high_f32(vo0c0123); |
|
} |
|
if (c & 1) { |
|
vst1_lane_f32(o1_tmp, vo1c01, 0); |
|
vst1_lane_f32(o0_tmp, vo0c01, 0); |
|
} |
|
o0 = (float*) ((uintptr_t) o0 + output_width_stride); |
|
o1 = (float*) ((uintptr_t) o1 + output_width_stride); |
|
} |
|
} |
|
|
|
|
|
o0 = (float*) ((uintptr_t) o0 - output_channel_decrement); |
|
o1 = (float*) ((uintptr_t) o1 - output_channel_decrement); |
|
|
|
i0 = (const float*) ((uintptr_t) i0 - input_width_decrement); |
|
i1 = (const float*) ((uintptr_t) i1 - input_width_decrement); |
|
i2 = (const float*) ((uintptr_t) i2 - input_width_decrement); |
|
i3 = (const float*) ((uintptr_t) i3 - input_width_decrement); |
|
i4 = (const float*) ((uintptr_t) i4 - input_width_decrement); |
|
|
|
w += 224; |
|
c = doz(c, 8); |
|
} while (c != 0); |
|
|
|
o0 = (float*) ((uintptr_t) o0 + output_height_increment); |
|
o1 = (float*) ((uintptr_t) o1 + output_height_increment); |
|
|
|
i0 = i4; |
|
i1 = (const float*) ((uintptr_t) i0 + input_height_stride); |
|
i2 = (const float*) ((uintptr_t) i1 + input_height_stride); |
|
i3 = (const float*) ((uintptr_t) i2 + input_height_stride); |
|
i4 = (const float*) ((uintptr_t) i3 + input_height_stride); |
|
} |
|
} |
|
|