hashirama/uvg266
1
0
Fork 0
mirror of https://github.com/ultravideo/uvg266.git synced 2024-11-27 11:24:05 +00:00
Code Issues Actions Projects Releases Wiki Activity

Rename variables to be less misleading

Browse source
This commit is contained in:
Ari Lemmetti 2021-11-21 02:15:31 +02:00
parent eb0f42aa96
commit 2bdfb3b536
1 changed files with 108 additions and 108 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

216
src/strategies/avx2/picture-avx2.c
View file

@ -968,34 +968,34 @@ static INLINE void bipred_average_im_im_template_avx2(kvz_pixel *dst,
int y = i / pu_w;
int x = i % pu_w;
__m256i sample_L0_01_16bit = _mm256_loadu_si256((__m256i*)&im_L0[i]);
__m256i sample_L1_01_16bit = _mm256_loadu_si256((__m256i*)&im_L1[i]);
__m256i sample_L0_23_16bit = _mm256_loadu_si256((__m256i*)&im_L0[i + 16]);
__m256i sample_L1_23_16bit = _mm256_loadu_si256((__m256i*)&im_L1[i + 16]);
__m256i sample_L0_a_16bit = _mm256_loadu_si256((__m256i*)&im_L0[i]);
__m256i sample_L1_a_16bit = _mm256_loadu_si256((__m256i*)&im_L1[i]);
__m256i sample_L0_b_16bit = _mm256_loadu_si256((__m256i*)&im_L0[i + 16]);
__m256i sample_L1_b_16bit = _mm256_loadu_si256((__m256i*)&im_L1[i + 16]);
__m256i sample_L0_L1_01_lo = _mm256_unpacklo_epi16(sample_L0_01_16bit, sample_L1_01_16bit);
__m256i sample_L0_L1_01_hi = _mm256_unpackhi_epi16(sample_L0_01_16bit, sample_L1_01_16bit);
__m256i sample_L0_L1_23_lo = _mm256_unpacklo_epi16(sample_L0_23_16bit, sample_L1_23_16bit);
__m256i sample_L0_L1_23_hi = _mm256_unpackhi_epi16(sample_L0_23_16bit, sample_L1_23_16bit);
__m256i sample_L0_L1_a_lo = _mm256_unpacklo_epi16(sample_L0_a_16bit, sample_L1_a_16bit);
__m256i sample_L0_L1_a_hi = _mm256_unpackhi_epi16(sample_L0_a_16bit, sample_L1_a_16bit);
__m256i sample_L0_L1_b_lo = _mm256_unpacklo_epi16(sample_L0_b_16bit, sample_L1_b_16bit);
__m256i sample_L0_L1_b_hi = _mm256_unpackhi_epi16(sample_L0_b_16bit, sample_L1_b_16bit);
__m256i all_ones = _mm256_set1_epi16(1);
__m256i avg_01_lo = _mm256_madd_epi16(sample_L0_L1_01_lo, all_ones);
__m256i avg_01_hi = _mm256_madd_epi16(sample_L0_L1_01_hi, all_ones);
__m256i avg_23_lo = _mm256_madd_epi16(sample_L0_L1_23_lo, all_ones);
__m256i avg_23_hi = _mm256_madd_epi16(sample_L0_L1_23_hi, all_ones);
__m256i all_ones = _mm256_set1_epi16(1);
__m256i avg_a_lo = _mm256_madd_epi16(sample_L0_L1_a_lo, all_ones);
__m256i avg_a_hi = _mm256_madd_epi16(sample_L0_L1_a_hi, all_ones);
__m256i avg_b_lo = _mm256_madd_epi16(sample_L0_L1_b_lo, all_ones);
__m256i avg_b_hi = _mm256_madd_epi16(sample_L0_L1_b_hi, all_ones);
avg_01_lo = _mm256_add_epi32(avg_01_lo, offset);
avg_01_hi = _mm256_add_epi32(avg_01_hi, offset);
avg_23_lo = _mm256_add_epi32(avg_23_lo, offset);
avg_23_hi = _mm256_add_epi32(avg_23_hi, offset);
avg_a_lo = _mm256_add_epi32(avg_a_lo, offset);
avg_a_hi = _mm256_add_epi32(avg_a_hi, offset);
avg_b_lo = _mm256_add_epi32(avg_b_lo, offset);
avg_b_hi = _mm256_add_epi32(avg_b_hi, offset);
avg_01_lo = _mm256_srai_epi32(avg_01_lo, shift);
avg_01_hi = _mm256_srai_epi32(avg_01_hi, shift);
avg_23_lo = _mm256_srai_epi32(avg_23_lo, shift);
avg_23_hi = _mm256_srai_epi32(avg_23_hi, shift);
avg_a_lo = _mm256_srai_epi32(avg_a_lo, shift);
avg_a_hi = _mm256_srai_epi32(avg_a_hi, shift);
avg_b_lo = _mm256_srai_epi32(avg_b_lo, shift);
avg_b_hi = _mm256_srai_epi32(avg_b_hi, shift);
__m256i avg_01 = _mm256_packus_epi32(avg_01_lo, avg_01_hi);
__m256i avg_23 = _mm256_packus_epi32(avg_23_lo, avg_23_hi);
__m256i avg_01 = _mm256_packus_epi32(avg_a_lo, avg_a_hi);
__m256i avg_23 = _mm256_packus_epi32(avg_b_lo, avg_b_hi);
__m256i avg0213 = _mm256_packus_epi16(avg_01, avg_23);
__m256i avg = _mm256_permute4x64_epi64(avg0213, _MM_SHUFFLE(3, 1, 2, 0));
@ -1017,35 +1017,35 @@ static INLINE void bipred_average_im_im_template_avx2(kvz_pixel *dst,
int x = i % pu_w;
// Last 64 bits of the 256 are not used to simplify the loop
__m256i mask = _mm256_setr_epi64x(-1, -1, -1, 0);
__m256i sample_L0_01_16bit = _mm256_loadu_si256((__m256i*)&im_L0[i]);
__m256i sample_L1_01_16bit = _mm256_loadu_si256((__m256i*)&im_L1[i]);
__m256i sample_L0_23_16bit = _mm256_castsi128_si256(_mm_loadu_si128((__m128i*)&im_L0[i + 16]));
__m256i sample_L1_23_16bit = _mm256_castsi128_si256(_mm_loadu_si128((__m128i*)&im_L1[i + 16]));
__m256i mask = _mm256_setr_epi64x(-1, -1, -1, 0);
__m256i sample_L0_a_16bit = _mm256_loadu_si256((__m256i*)&im_L0[i]);
__m256i sample_L1_a_16bit = _mm256_loadu_si256((__m256i*)&im_L1[i]);
__m256i sample_L0_b_16bit = _mm256_castsi128_si256(_mm_loadu_si128((__m128i*)&im_L0[i + 16]));
__m256i sample_L1_b_16bit = _mm256_castsi128_si256(_mm_loadu_si128((__m128i*)&im_L1[i + 16]));
__m256i sample_L0_L1_01_lo = _mm256_unpacklo_epi16(sample_L0_01_16bit, sample_L1_01_16bit);
__m256i sample_L0_L1_01_hi = _mm256_unpackhi_epi16(sample_L0_01_16bit, sample_L1_01_16bit);
__m256i sample_L0_L1_23_lo = _mm256_unpacklo_epi16(sample_L0_23_16bit, sample_L1_23_16bit);
__m256i sample_L0_L1_23_hi = _mm256_unpackhi_epi16(sample_L0_23_16bit, sample_L1_23_16bit);
__m256i sample_L0_L1_a_lo = _mm256_unpacklo_epi16(sample_L0_a_16bit, sample_L1_a_16bit);
__m256i sample_L0_L1_a_hi = _mm256_unpackhi_epi16(sample_L0_a_16bit, sample_L1_a_16bit);
__m256i sample_L0_L1_b_lo = _mm256_unpacklo_epi16(sample_L0_b_16bit, sample_L1_b_16bit);
__m256i sample_L0_L1_b_hi = _mm256_unpackhi_epi16(sample_L0_b_16bit, sample_L1_b_16bit);
__m256i all_ones = _mm256_set1_epi16(1);
__m256i avg_01_lo = _mm256_madd_epi16(sample_L0_L1_01_lo, all_ones);
__m256i avg_01_hi = _mm256_madd_epi16(sample_L0_L1_01_hi, all_ones);
__m256i avg_23_lo = _mm256_madd_epi16(sample_L0_L1_23_lo, all_ones);
__m256i avg_23_hi = _mm256_madd_epi16(sample_L0_L1_23_hi, all_ones);
__m256i all_ones = _mm256_set1_epi16(1);
__m256i avg_a_lo = _mm256_madd_epi16(sample_L0_L1_a_lo, all_ones);
__m256i avg_a_hi = _mm256_madd_epi16(sample_L0_L1_a_hi, all_ones);
__m256i avg_b_lo = _mm256_madd_epi16(sample_L0_L1_b_lo, all_ones);
__m256i avg_b_hi = _mm256_madd_epi16(sample_L0_L1_b_hi, all_ones);
avg_01_lo = _mm256_add_epi32(avg_01_lo, offset);
avg_01_hi = _mm256_add_epi32(avg_01_hi, offset);
avg_23_lo = _mm256_add_epi32(avg_23_lo, offset);
avg_23_hi = _mm256_add_epi32(avg_23_hi, offset);
avg_a_lo = _mm256_add_epi32(avg_a_lo, offset);
avg_a_hi = _mm256_add_epi32(avg_a_hi, offset);
avg_b_lo = _mm256_add_epi32(avg_b_lo, offset);
avg_b_hi = _mm256_add_epi32(avg_b_hi, offset);
avg_01_lo = _mm256_srai_epi32(avg_01_lo, shift);
avg_01_hi = _mm256_srai_epi32(avg_01_hi, shift);
avg_23_lo = _mm256_srai_epi32(avg_23_lo, shift);
avg_23_hi = _mm256_srai_epi32(avg_23_hi, shift);
avg_a_lo = _mm256_srai_epi32(avg_a_lo, shift);
avg_a_hi = _mm256_srai_epi32(avg_a_hi, shift);
avg_b_lo = _mm256_srai_epi32(avg_b_lo, shift);
avg_b_hi = _mm256_srai_epi32(avg_b_hi, shift);
__m256i avg_01 = _mm256_packus_epi32(avg_01_lo, avg_01_hi);
__m256i avg_23 = _mm256_packus_epi32(avg_23_lo, avg_23_hi);
__m256i avg_01 = _mm256_packus_epi32(avg_a_lo, avg_a_hi);
__m256i avg_23 = _mm256_packus_epi32(avg_b_lo, avg_b_hi);
__m256i avg0213 = _mm256_packus_epi16(avg_01, avg_23);
__m256i avg = _mm256_permute4x64_epi64(avg0213, _MM_SHUFFLE(3, 1, 2, 0));
@ -1096,24 +1096,24 @@ static INLINE void bipred_average_im_im_template_avx2(kvz_pixel *dst,
int y = i / pu_w;
__m256i mask = _mm256_setr_epi64x(-1, -1, -1, 0);
__m256i mask = _mm256_setr_epi64x(-1, -1, -1, 0);
__m256i sample_L0_16bit = _mm256_maskload_epi64((const long long*)(&im_L0[i]), mask);
__m256i sample_L1_16bit = _mm256_maskload_epi64((const long long*)(&im_L1[i]), mask);
__m256i sample_L0_L1_lo = _mm256_unpacklo_epi16(sample_L0_16bit, sample_L1_16bit);
__m256i sample_L0_L1_hi = _mm256_unpackhi_epi16(sample_L0_16bit, sample_L1_16bit);
__m256i all_ones = _mm256_set1_epi16(1);
__m256i avg_01_lo = _mm256_madd_epi16(sample_L0_L1_lo, all_ones);
__m256i avg_01_hi = _mm256_madd_epi16(sample_L0_L1_hi, all_ones);
__m256i all_ones = _mm256_set1_epi16(1);
__m256i avg_a_lo = _mm256_madd_epi16(sample_L0_L1_lo, all_ones);
__m256i avg_a_hi = _mm256_madd_epi16(sample_L0_L1_hi, all_ones);
avg_01_lo = _mm256_add_epi32(avg_01_lo, offset);
avg_01_hi = _mm256_add_epi32(avg_01_hi, offset);
avg_a_lo = _mm256_add_epi32(avg_a_lo, offset);
avg_a_hi = _mm256_add_epi32(avg_a_hi, offset);
avg_01_lo = _mm256_srai_epi32(avg_01_lo, shift);
avg_01_hi = _mm256_srai_epi32(avg_01_hi, shift);
avg_a_lo = _mm256_srai_epi32(avg_a_lo, shift);
avg_a_hi = _mm256_srai_epi32(avg_a_hi, shift);
__m256i avg256 = _mm256_packus_epi32(avg_01_lo, avg_01_hi);
__m256i avg256 = _mm256_packus_epi32(avg_a_lo, avg_a_hi);
avg256 = _mm256_packus_epi16(avg256, avg256);
avg256 = _mm256_permute4x64_epi64(avg256, _MM_SHUFFLE(3, 1, 2, 0));
__m128i avg = _mm256_castsi256_si128(avg256);
@ -1197,36 +1197,36 @@ static INLINE void bipred_average_px_im_template_avx2(kvz_pixel *dst,
int y = i / pu_w;
int x = i % pu_w;
__m256i sample_px_01_16bit = _mm256_cvtepu8_epi16(_mm_loadu_si128((__m128i*)&px[i]));
__m256i sample_px_23_16bit = _mm256_cvtepu8_epi16(_mm_loadu_si128((__m128i*)&px[i + 16]));
sample_px_01_16bit = _mm256_slli_epi16(sample_px_01_16bit, 14 - KVZ_BIT_DEPTH);
sample_px_23_16bit = _mm256_slli_epi16(sample_px_23_16bit, 14 - KVZ_BIT_DEPTH);
__m256i sample_im_01_16bit = _mm256_loadu_si256((__m256i*)&im[i]);
__m256i sample_im_23_16bit = _mm256_loadu_si256((__m256i*)&im[i + 16]);
__m256i sample_px_a_16bit = _mm256_cvtepu8_epi16(_mm_loadu_si128((__m128i*)&px[i]));
__m256i sample_px_b_16bit = _mm256_cvtepu8_epi16(_mm_loadu_si128((__m128i*)&px[i + 16]));
sample_px_a_16bit = _mm256_slli_epi16(sample_px_a_16bit, 14 - KVZ_BIT_DEPTH);
sample_px_b_16bit = _mm256_slli_epi16(sample_px_b_16bit, 14 - KVZ_BIT_DEPTH);
__m256i sample_im_a_16bit = _mm256_loadu_si256((__m256i*)&im[i]);
__m256i sample_im_b_16bit = _mm256_loadu_si256((__m256i*)&im[i + 16]);
__m256i sample_px_im_01_lo = _mm256_unpacklo_epi16(sample_px_01_16bit, sample_im_01_16bit);
__m256i sample_px_im_01_hi = _mm256_unpackhi_epi16(sample_px_01_16bit, sample_im_01_16bit);
__m256i sample_px_im_23_lo = _mm256_unpacklo_epi16(sample_px_23_16bit, sample_im_23_16bit);
__m256i sample_px_im_23_hi = _mm256_unpackhi_epi16(sample_px_23_16bit, sample_im_23_16bit);
__m256i sample_px_im_a_lo = _mm256_unpacklo_epi16(sample_px_a_16bit, sample_im_a_16bit);
__m256i sample_px_im_a_hi = _mm256_unpackhi_epi16(sample_px_a_16bit, sample_im_a_16bit);
__m256i sample_px_im_b_lo = _mm256_unpacklo_epi16(sample_px_b_16bit, sample_im_b_16bit);
__m256i sample_px_im_b_hi = _mm256_unpackhi_epi16(sample_px_b_16bit, sample_im_b_16bit);
__m256i all_ones = _mm256_set1_epi16(1);
__m256i avg_01_lo = _mm256_madd_epi16(sample_px_im_01_lo, all_ones);
__m256i avg_01_hi = _mm256_madd_epi16(sample_px_im_01_hi, all_ones);
__m256i avg_23_lo = _mm256_madd_epi16(sample_px_im_23_lo, all_ones);
__m256i avg_23_hi = _mm256_madd_epi16(sample_px_im_23_hi, all_ones);
__m256i avg_a_lo = _mm256_madd_epi16(sample_px_im_a_lo, all_ones);
__m256i avg_a_hi = _mm256_madd_epi16(sample_px_im_a_hi, all_ones);
__m256i avg_b_lo = _mm256_madd_epi16(sample_px_im_b_lo, all_ones);
__m256i avg_b_hi = _mm256_madd_epi16(sample_px_im_b_hi, all_ones);
avg_01_lo = _mm256_add_epi32(avg_01_lo, offset);
avg_01_hi = _mm256_add_epi32(avg_01_hi, offset);
avg_23_lo = _mm256_add_epi32(avg_23_lo, offset);
avg_23_hi = _mm256_add_epi32(avg_23_hi, offset);
avg_a_lo = _mm256_add_epi32(avg_a_lo, offset);
avg_a_hi = _mm256_add_epi32(avg_a_hi, offset);
avg_b_lo = _mm256_add_epi32(avg_b_lo, offset);
avg_b_hi = _mm256_add_epi32(avg_b_hi, offset);
avg_01_lo = _mm256_srai_epi32(avg_01_lo, shift);
avg_01_hi = _mm256_srai_epi32(avg_01_hi, shift);
avg_23_lo = _mm256_srai_epi32(avg_23_lo, shift);
avg_23_hi = _mm256_srai_epi32(avg_23_hi, shift);
avg_a_lo = _mm256_srai_epi32(avg_a_lo, shift);
avg_a_hi = _mm256_srai_epi32(avg_a_hi, shift);
avg_b_lo = _mm256_srai_epi32(avg_b_lo, shift);
avg_b_hi = _mm256_srai_epi32(avg_b_hi, shift);
__m256i avg_01 = _mm256_packus_epi32(avg_01_lo, avg_01_hi);
__m256i avg_23 = _mm256_packus_epi32(avg_23_lo, avg_23_hi);
__m256i avg_01 = _mm256_packus_epi32(avg_a_lo, avg_a_hi);
__m256i avg_23 = _mm256_packus_epi32(avg_b_lo, avg_b_hi);
__m256i avg0213 = _mm256_packus_epi16(avg_01, avg_23);
__m256i avg = _mm256_permute4x64_epi64(avg0213, _MM_SHUFFLE(3, 1, 2, 0));
@ -1248,39 +1248,39 @@ static INLINE void bipred_average_px_im_template_avx2(kvz_pixel *dst,
int x = i % pu_w;
// Last 64 bits of the 256 / 32 bits of the 128 are not used to simplify the loop
__m256i mask = _mm256_setr_epi64x(-1, -1, -1, 0);
__m128i sample_px_01_8bit = _mm_loadu_si128((__m128i*)&px[i]);
__m128i sample_px_23_8bit = _mm_loadl_epi64((__m128i*)&px[i + 16]);
__m256i sample_px_01_16bit = _mm256_cvtepu8_epi16(sample_px_01_8bit);
__m256i sample_px_23_16bit = _mm256_cvtepu8_epi16(sample_px_23_8bit);
sample_px_01_16bit = _mm256_slli_epi16(sample_px_01_16bit, 14 - KVZ_BIT_DEPTH);
sample_px_23_16bit = _mm256_slli_epi16(sample_px_23_16bit, 14 - KVZ_BIT_DEPTH);
__m256i sample_im_01_16bit = _mm256_loadu_si256((__m256i*)&im[i]);
__m256i sample_im_23_16bit = _mm256_castsi128_si256(_mm_loadu_si128((__m128i*)&im[i + 16]));
__m256i mask = _mm256_setr_epi64x(-1, -1, -1, 0);
__m128i sample_px_a_8bit = _mm_loadu_si128((__m128i*)&px[i]);
__m128i sample_px_b_8bit = _mm_loadl_epi64((__m128i*)&px[i + 16]);
__m256i sample_px_a_16bit = _mm256_cvtepu8_epi16(sample_px_a_8bit);
__m256i sample_px_b_16bit = _mm256_cvtepu8_epi16(sample_px_b_8bit);
sample_px_a_16bit = _mm256_slli_epi16(sample_px_a_16bit, 14 - KVZ_BIT_DEPTH);
sample_px_b_16bit = _mm256_slli_epi16(sample_px_b_16bit, 14 - KVZ_BIT_DEPTH);
__m256i sample_im_a_16bit = _mm256_loadu_si256((__m256i*)&im[i]);
__m256i sample_im_b_16bit = _mm256_castsi128_si256(_mm_loadu_si128((__m128i*)&im[i + 16]));
__m256i sample_px_im_01_lo = _mm256_unpacklo_epi16(sample_px_01_16bit, sample_im_01_16bit);
__m256i sample_px_im_01_hi = _mm256_unpackhi_epi16(sample_px_01_16bit, sample_im_01_16bit);
__m256i sample_px_im_23_lo = _mm256_unpacklo_epi16(sample_px_23_16bit, sample_im_23_16bit);
__m256i sample_px_im_23_hi = _mm256_unpackhi_epi16(sample_px_23_16bit, sample_im_23_16bit);
__m256i sample_px_im_a_lo = _mm256_unpacklo_epi16(sample_px_a_16bit, sample_im_a_16bit);
__m256i sample_px_im_a_hi = _mm256_unpackhi_epi16(sample_px_a_16bit, sample_im_a_16bit);
__m256i sample_px_im_b_lo = _mm256_unpacklo_epi16(sample_px_b_16bit, sample_im_b_16bit);
__m256i sample_px_im_b_hi = _mm256_unpackhi_epi16(sample_px_b_16bit, sample_im_b_16bit);
__m256i all_ones = _mm256_set1_epi16(1);
__m256i avg_01_lo = _mm256_madd_epi16(sample_px_im_01_lo, all_ones);
__m256i avg_01_hi = _mm256_madd_epi16(sample_px_im_01_hi, all_ones);
__m256i avg_23_lo = _mm256_madd_epi16(sample_px_im_23_lo, all_ones);
__m256i avg_23_hi = _mm256_madd_epi16(sample_px_im_23_hi, all_ones);
__m256i all_ones = _mm256_set1_epi16(1);
__m256i avg_a_lo = _mm256_madd_epi16(sample_px_im_a_lo, all_ones);
__m256i avg_a_hi = _mm256_madd_epi16(sample_px_im_a_hi, all_ones);
__m256i avg_b_lo = _mm256_madd_epi16(sample_px_im_b_lo, all_ones);
__m256i avg_b_hi = _mm256_madd_epi16(sample_px_im_b_hi, all_ones);
avg_01_lo = _mm256_add_epi32(avg_01_lo, offset);
avg_01_hi = _mm256_add_epi32(avg_01_hi, offset);
avg_23_lo = _mm256_add_epi32(avg_23_lo, offset);
avg_23_hi = _mm256_add_epi32(avg_23_hi, offset);
avg_a_lo = _mm256_add_epi32(avg_a_lo, offset);
avg_a_hi = _mm256_add_epi32(avg_a_hi, offset);
avg_b_lo = _mm256_add_epi32(avg_b_lo, offset);
avg_b_hi = _mm256_add_epi32(avg_b_hi, offset);
avg_01_lo = _mm256_srai_epi32(avg_01_lo, shift);
avg_01_hi = _mm256_srai_epi32(avg_01_hi, shift);
avg_23_lo = _mm256_srai_epi32(avg_23_lo, shift);
avg_23_hi = _mm256_srai_epi32(avg_23_hi, shift);
avg_a_lo = _mm256_srai_epi32(avg_a_lo, shift);
avg_a_hi = _mm256_srai_epi32(avg_a_hi, shift);
avg_b_lo = _mm256_srai_epi32(avg_b_lo, shift);
avg_b_hi = _mm256_srai_epi32(avg_b_hi, shift);
__m256i avg_01 = _mm256_packus_epi32(avg_01_lo, avg_01_hi);
__m256i avg_23 = _mm256_packus_epi32(avg_23_lo, avg_23_hi);
__m256i avg_01 = _mm256_packus_epi32(avg_a_lo, avg_a_hi);
__m256i avg_23 = _mm256_packus_epi32(avg_b_lo, avg_b_hi);
__m256i avg0213 = _mm256_packus_epi16(avg_01, avg_23);
__m256i avg = _mm256_permute4x64_epi64(avg0213, _MM_SHUFFLE(3, 1, 2, 0));