/external/libaom/libaom/av1/common/x86/ |
D | wiener_convolve_sse2.c | 92 const __m128i src_5 = _mm_unpacklo_epi8(_mm_srli_si128(data, 5), zero); in av1_wiener_convolve_add_src_sse2() local 93 const __m128i res_5 = _mm_madd_epi16(src_5, coeff_45); in av1_wiener_convolve_add_src_sse2() 167 const __m128i src_5 = in av1_wiener_convolve_add_src_sse2() local 176 const __m128i res_5 = _mm_madd_epi16(src_5, coeff_45); in av1_wiener_convolve_add_src_sse2()
|
D | highbd_wiener_convolve_avx2.c | 100 const __m256i src_5 = yy_loadu_256(src_ij + 5); in av1_highbd_wiener_convolve_add_src_avx2() local 110 const __m256i res_5 = _mm256_madd_epi16(src_5, coeffs_45); in av1_highbd_wiener_convolve_add_src_avx2() 207 const __m256i src_5 = _mm256_unpackhi_epi16(data_4, data_5); in av1_highbd_wiener_convolve_add_src_avx2() local 212 const __m256i res_5 = _mm256_madd_epi16(src_5, coeffs_45); in av1_highbd_wiener_convolve_add_src_avx2()
|
D | jnt_convolve_ssse3.c | 106 const __m128i src_5 = _mm_alignr_epi8(src_hi, src_lo, 10); in av1_dist_wtd_convolve_2d_ssse3() local 107 const __m128i res_5 = _mm_madd_epi16(src_5, coeff_45); in av1_dist_wtd_convolve_2d_ssse3() 180 const __m128i src_5 = in av1_dist_wtd_convolve_2d_ssse3() local 189 const __m128i res_5 = _mm_madd_epi16(src_5, coeff_45); in av1_dist_wtd_convolve_2d_ssse3()
|
D | wiener_convolve_avx2.c | 157 __m256i src_5 = _mm256_loadu_si256((__m256i *)(im_block + 5 * im_stride)); in av1_wiener_convolve_add_src_avx2() local 162 s[2] = _mm256_unpacklo_epi16(src_4, src_5); in av1_wiener_convolve_add_src_avx2() 166 s[6] = _mm256_unpackhi_epi16(src_4, src_5); in av1_wiener_convolve_add_src_avx2()
|
D | warp_plane_avx2.c | 795 const __m256i src_5 = in warp_vertical_filter_avx2() local 800 src[4] = _mm256_unpacklo_epi16(src_4, src_5); in warp_vertical_filter_avx2() 804 src[5] = _mm256_unpackhi_epi16(src_4, src_5); in warp_vertical_filter_avx2() 845 const __m256i src_5 = in warp_vertical_filter_gamma0_avx2() local 850 src[4] = _mm256_unpacklo_epi16(src_4, src_5); in warp_vertical_filter_gamma0_avx2() 854 src[5] = _mm256_unpackhi_epi16(src_4, src_5); in warp_vertical_filter_gamma0_avx2() 894 const __m256i src_5 = in warp_vertical_filter_delta0_avx2() local 899 src[4] = _mm256_unpacklo_epi16(src_4, src_5); in warp_vertical_filter_delta0_avx2() 903 src[5] = _mm256_unpackhi_epi16(src_4, src_5); in warp_vertical_filter_delta0_avx2() 942 const __m256i src_5 = in warp_vertical_filter_gamma0_delta0_avx2() local [all …]
|
D | convolve_2d_sse2.c | 94 const __m128i src_5 = _mm_unpacklo_epi8(_mm_srli_si128(data, 5), zero); in av1_convolve_2d_sr_sse2() local 95 const __m128i res_5 = _mm_madd_epi16(src_5, coeff_45); in av1_convolve_2d_sr_sse2() 172 const __m128i src_5 = in av1_convolve_2d_sr_sse2() local 181 const __m128i res_5 = _mm_madd_epi16(src_5, coeff_45); in av1_convolve_2d_sr_sse2()
|
D | highbd_wiener_convolve_ssse3.c | 169 const __m128i src_5 = in av1_highbd_wiener_convolve_add_src_ssse3() local 178 const __m128i res_5 = _mm_madd_epi16(src_5, coeff_45); in av1_highbd_wiener_convolve_add_src_ssse3()
|
D | jnt_convolve_sse2.c | 488 const __m128i src_5 = _mm_or_si128(temp_hi, temp_lo); in av1_dist_wtd_convolve_2d_sse2() local 489 const __m128i res_5 = _mm_madd_epi16(src_5, coeff_45); in av1_dist_wtd_convolve_2d_sse2() 564 const __m128i src_5 = in av1_dist_wtd_convolve_2d_sse2() local 573 const __m128i res_5 = _mm_madd_epi16(src_5, coeff_45); in av1_dist_wtd_convolve_2d_sse2()
|
D | highbd_convolve_2d_sse4.c | 332 const __m128i src_5 = in av1_highbd_dist_wtd_convolve_2d_sse4_1() local 341 const __m128i res_5 = _mm_madd_epi16(src_5, coeff_45); in av1_highbd_dist_wtd_convolve_2d_sse4_1()
|
D | highbd_warp_plane_sse4.c | 498 const __m128i src_5 = _mm_unpackhi_epi16(src[4], src[5]); in av1_highbd_warp_affine_sse4_1() local 526 const __m128i res_5 = _mm_madd_epi16(src_5, coeff_5); in av1_highbd_warp_affine_sse4_1()
|
D | warp_plane_sse4.c | 565 const __m128i src_5 = _mm_unpackhi_epi16(src[4], src[5]); in filter_src_pixels_vertical() local 570 const __m128i res_5 = _mm_madd_epi16(src_5, coeffs[6]); in filter_src_pixels_vertical()
|
/external/libaom/libaom/aom_dsp/x86/ |
D | convolve_avx2.h | 93 __m256i src_5 = _mm256_loadu_si256((__m256i *)(im_block + 5 * im_stride)); \ 98 s[2] = _mm256_unpacklo_epi16(src_4, src_5); \ 102 s[6] = _mm256_unpackhi_epi16(src_4, src_5); \
|