|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
#include <assert.h> |
|
|
|
#include <tmmintrin.h> |
|
|
|
#include <xnnpack/dwconv.h> |
|
#include <xnnpack/math.h> |
|
|
|
|
|
void xnn_f32_dwconv2d_chw_ukernel_3x3p1__ssse3_1x4_acc4( |
|
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 |
|
{ |
|
assert(input_height != 0); |
|
assert(input_width != 0); |
|
assert(input_width % sizeof(float) == 0); |
|
assert(padding_top == 1); |
|
|
|
const __m128 vmask = _mm_load_ps((const float*) params->sse_stride1.mask); |
|
const __m128 vmax = _mm_load_ps(params->sse_stride1.max); |
|
const __m128 vmin = _mm_load_ps(params->sse_stride1.min); |
|
|
|
const __m128 vbias = _mm_load1_ps(weights); |
|
const __m128 vk00 = _mm_load1_ps(weights + 1); |
|
const __m128 vk01 = _mm_load1_ps(weights + 2); |
|
const __m128 vk02 = _mm_load1_ps(weights + 3); |
|
const __m128 vk10 = _mm_load1_ps(weights + 4); |
|
const __m128 vk11 = _mm_load1_ps(weights + 5); |
|
const __m128 vk12 = _mm_load1_ps(weights + 6); |
|
const __m128 vk20 = _mm_load1_ps(weights + 7); |
|
const __m128 vk21 = _mm_load1_ps(weights + 8); |
|
const __m128 vk22 = _mm_load1_ps(weights + 9); |
|
|
|
const size_t input_decrement = round_up_po2(input_width, 4 * sizeof(float)); |
|
|
|
const float* i0 = zero; |
|
const float* i1 = input; |
|
const float* i2 = (const float*) ((uintptr_t) i1 + input_width); |
|
|
|
float* o0 = output; |
|
|
|
size_t output_height = input_height; |
|
do { |
|
if XNN_UNPREDICTABLE(output_height < 2) { |
|
i2 = zero; |
|
} |
|
|
|
__m128 vi0x0123 = _mm_setzero_ps(); |
|
__m128 vi1x0123 = _mm_setzero_ps(); |
|
__m128 vi2x0123 = _mm_setzero_ps(); |
|
|
|
__m128 vi0x4567 = _mm_loadu_ps(i0); |
|
i0 += 4; |
|
__m128 vi1x4567 = _mm_loadu_ps(i1); |
|
i1 += 4; |
|
__m128 vi2x4567 = _mm_loadu_ps(i2); |
|
i2 += 4; |
|
|
|
size_t w = input_width; |
|
for (; w > 4 * sizeof(float); w -= 4 * sizeof(float)) { |
|
const __m128 vi0x89AB = _mm_loadu_ps(i0); |
|
i0 += 4; |
|
const __m128 vi1x89AB = _mm_loadu_ps(i1); |
|
i1 += 4; |
|
const __m128 vi2x89AB = _mm_loadu_ps(i2); |
|
i2 += 4; |
|
|
|
__m128 vo0p0 = _mm_add_ps(vbias, _mm_mul_ps(vi0x4567, vk01)); |
|
__m128 vo0p1 = _mm_mul_ps(vi1x4567, vk11); |
|
__m128 vo0p2 = _mm_mul_ps(vi2x4567, vk21); |
|
|
|
const __m128 vi0x3456 = _mm_castsi128_ps(_mm_alignr_epi8(_mm_castps_si128(vi0x4567), _mm_castps_si128(vi0x0123), 12)); |
|
const __m128 vi1x3456 = _mm_castsi128_ps(_mm_alignr_epi8(_mm_castps_si128(vi1x4567), _mm_castps_si128(vi1x0123), 12)); |
|
const __m128 vi2x3456 = _mm_castsi128_ps(_mm_alignr_epi8(_mm_castps_si128(vi2x4567), _mm_castps_si128(vi2x0123), 12)); |
|
|
|
__m128 vo0p3 = _mm_mul_ps(vi0x3456, vk00); |
|
vo0p0 = _mm_add_ps(vo0p0, _mm_mul_ps(vi1x3456, vk10)); |
|
vo0p1 = _mm_add_ps(vo0p1, _mm_mul_ps(vi2x3456, vk20)); |
|
|
|
vi0x0123 = vi0x4567; |
|
vi1x0123 = vi1x4567; |
|
vi2x0123 = vi2x4567; |
|
|
|
const __m128 vi0x5678 = _mm_castsi128_ps(_mm_alignr_epi8(_mm_castps_si128(vi0x89AB), _mm_castps_si128(vi0x4567), 4)); |
|
const __m128 vi1x5678 = _mm_castsi128_ps(_mm_alignr_epi8(_mm_castps_si128(vi1x89AB), _mm_castps_si128(vi1x4567), 4)); |
|
const __m128 vi2x5678 = _mm_castsi128_ps(_mm_alignr_epi8(_mm_castps_si128(vi2x89AB), _mm_castps_si128(vi2x4567), 4)); |
|
|
|
vo0p2 = _mm_add_ps(vo0p2, _mm_mul_ps(vi0x5678, vk02)); |
|
vo0p3 = _mm_add_ps(vo0p3, _mm_mul_ps(vi1x5678, vk12)); |
|
vo0p0 = _mm_add_ps(vo0p0, _mm_mul_ps(vi2x5678, vk22)); |
|
|
|
vi0x4567 = vi0x89AB; |
|
vi1x4567 = vi1x89AB; |
|
vi2x4567 = vi2x89AB; |
|
|
|
vo0p0 = _mm_add_ps(vo0p0, vo0p1); |
|
vo0p2 = _mm_add_ps(vo0p2, vo0p3); |
|
vo0p0 = _mm_add_ps(vo0p0, vo0p2); |
|
|
|
__m128 vo0 = _mm_max_ps(vo0p0, vmin); |
|
|
|
vo0 = _mm_min_ps(vo0, vmax); |
|
|
|
_mm_storeu_ps(o0, vo0); |
|
o0 += 4; |
|
} |
|
|
|
assert(w >= 1 * sizeof(float)); |
|
assert(w <= 4 * sizeof(float)); |
|
{ |
|
vi0x4567 = _mm_and_ps(vmask, vi0x4567); |
|
vi1x4567 = _mm_and_ps(vmask, vi1x4567); |
|
vi2x4567 = _mm_and_ps(vmask, vi2x4567); |
|
|
|
__m128 vo0p0 = _mm_add_ps(vbias, _mm_mul_ps(vi0x4567, vk01)); |
|
__m128 vo0p1 = _mm_mul_ps(vi1x4567, vk11); |
|
__m128 vo0p2 = _mm_mul_ps(vi2x4567, vk21); |
|
|
|
const __m128 vi0x3456 = _mm_castsi128_ps(_mm_alignr_epi8(_mm_castps_si128(vi0x4567), _mm_castps_si128(vi0x0123), 12)); |
|
const __m128 vi1x3456 = _mm_castsi128_ps(_mm_alignr_epi8(_mm_castps_si128(vi1x4567), _mm_castps_si128(vi1x0123), 12)); |
|
const __m128 vi2x3456 = _mm_castsi128_ps(_mm_alignr_epi8(_mm_castps_si128(vi2x4567), _mm_castps_si128(vi2x0123), 12)); |
|
|
|
__m128 vo0p3 = _mm_mul_ps(vi0x3456, vk00); |
|
vo0p0 = _mm_add_ps(vo0p0, _mm_mul_ps(vi1x3456, vk10)); |
|
vo0p1 = _mm_add_ps(vo0p1, _mm_mul_ps(vi2x3456, vk20)); |
|
|
|
const __m128i vzero = _mm_setzero_si128(); |
|
const __m128 vi0x5678 = _mm_castsi128_ps(_mm_alignr_epi8(vzero, _mm_castps_si128(vi0x4567), 4)); |
|
const __m128 vi1x5678 = _mm_castsi128_ps(_mm_alignr_epi8(vzero, _mm_castps_si128(vi1x4567), 4)); |
|
const __m128 vi2x5678 = _mm_castsi128_ps(_mm_alignr_epi8(vzero, _mm_castps_si128(vi2x4567), 4)); |
|
|
|
vo0p2 = _mm_add_ps(vo0p2, _mm_mul_ps(vi0x5678, vk02)); |
|
vo0p3 = _mm_add_ps(vo0p3, _mm_mul_ps(vi1x5678, vk12)); |
|
vo0p0 = _mm_add_ps(vo0p0, _mm_mul_ps(vi2x5678, vk22)); |
|
|
|
vo0p0 = _mm_add_ps(vo0p0, vo0p1); |
|
vo0p2 = _mm_add_ps(vo0p2, vo0p3); |
|
vo0p0 = _mm_add_ps(vo0p0, vo0p2); |
|
|
|
__m128 vo0 = _mm_max_ps(vo0p0, vmin); |
|
|
|
vo0 = _mm_min_ps(vo0, vmax); |
|
|
|
if XNN_LIKELY(w == 4 * sizeof(float)) { |
|
_mm_storeu_ps(o0, vo0); |
|
o0 += 4; |
|
} else { |
|
if (w & (2 * sizeof(float))) { |
|
_mm_storel_pi((__m64*) o0, vo0); |
|
o0 += 2; |
|
|
|
vo0 = _mm_movehl_ps(vo0, vo0); |
|
} |
|
if (w & (1 * sizeof(float))) { |
|
_mm_store_ss(o0, vo0); |
|
o0 += 1; |
|
} |
|
} |
|
} |
|
|
|
i0 = (const float*) ((uintptr_t) i1 - input_decrement); |
|
i1 = (const float*) ((uintptr_t) i2 - input_decrement); |
|
i2 = (const float*) ((uintptr_t) i1 + input_width); |
|
|
|
|
|
} while (--output_height != 0); |
|
} |
|
|