| /external/libgav1/libgav1/src/dsp/arm/ |
| D | intra_edge_neon.cc | 147 const uint8x16_t src_3 = vld1q_u8(dst_buffer + i + 1); in IntraEdgeFilter_NEON() local
153 vaddl_u8(vget_low_u8(src_1), vget_low_u8(src_2)), vget_low_u8(src_3)); in IntraEdgeFilter_NEON()
160 vget_high_u8(src_3)); in IntraEdgeFilter_NEON()
178 const uint8x16_t src_3 = vld1q_u8(dst_buffer + i + 1); in IntraEdgeFilter_NEON() local
184 vaddl_u8(vget_low_u8(src_1), vget_low_u8(src_2)), vget_low_u8(src_3)); in IntraEdgeFilter_NEON()
191 vget_high_u8(src_3)); in IntraEdgeFilter_NEON()
|
| /external/libaom/libaom/av1/common/arm/ |
| D | warp_plane_neon.c | 221 uint8x16_t src_3, uint8x16_t src_4, in horizontal_filter_neon() argument
239 uint8x16_t tmp_2 = vandq_u8(src_3, mask); in horizontal_filter_neon()
490 uint8x16_t src_1, src_2, src_3, src_4; in av1_warp_affine_neon() local
595 src_3 = vextq_u8(src_2, src_2, 1); in av1_warp_affine_neon()
596 src_4 = vextq_u8(src_3, src_3, 1); in av1_warp_affine_neon()
598 horizontal_filter_neon(src_1, src_2, src_3, src_4, tmp, sx, alpha, k, in av1_warp_affine_neon()
613 src_3 = vextq_u8(src_2, src_2, 1); in av1_warp_affine_neon()
614 src_4 = vextq_u8(src_3, src_3, 1); in av1_warp_affine_neon()
616 horizontal_filter_neon(src_1, src_2, src_3, src_4, tmp, sx, alpha, k, in av1_warp_affine_neon()
|
| /external/libaom/libaom/av1/common/x86/ |
| D | wiener_convolve_sse2.c | 90 const __m128i src_3 = _mm_unpacklo_epi8(_mm_srli_si128(data, 3), zero); in av1_wiener_convolve_add_src_sse2() local
91 const __m128i res_3 = _mm_madd_epi16(src_3, coeff_23); in av1_wiener_convolve_add_src_sse2()
164 const __m128i src_3 = in av1_wiener_convolve_add_src_sse2() local
175 const __m128i res_3 = _mm_madd_epi16(src_3, coeff_23); in av1_wiener_convolve_add_src_sse2()
|
| D | highbd_wiener_convolve_avx2.c | 98 const __m256i src_3 = yy_loadu_256(src_ij + 3); in av1_highbd_wiener_convolve_add_src_avx2() local
108 const __m256i res_3 = _mm256_madd_epi16(src_3, coeffs_23); in av1_highbd_wiener_convolve_add_src_avx2()
206 const __m256i src_3 = _mm256_unpackhi_epi16(data_2, data_3); in av1_highbd_wiener_convolve_add_src_avx2() local
211 const __m256i res_3 = _mm256_madd_epi16(src_3, coeffs_23); in av1_highbd_wiener_convolve_add_src_avx2()
|
| D | jnt_convolve_ssse3.c | 104 const __m128i src_3 = _mm_alignr_epi8(src_hi, src_lo, 6); in av1_dist_wtd_convolve_2d_ssse3() local
105 const __m128i res_3 = _mm_madd_epi16(src_3, coeff_23); in av1_dist_wtd_convolve_2d_ssse3()
177 const __m128i src_3 = in av1_dist_wtd_convolve_2d_ssse3() local
188 const __m128i res_3 = _mm_madd_epi16(src_3, coeff_23); in av1_dist_wtd_convolve_2d_ssse3()
|
| D | wiener_convolve_avx2.c | 155 __m256i src_3 = _mm256_loadu_si256((__m256i *)(im_block + 3 * im_stride)); in av1_wiener_convolve_add_src_avx2() local
161 s[1] = _mm256_unpacklo_epi16(src_2, src_3); in av1_wiener_convolve_add_src_avx2()
165 s[5] = _mm256_unpackhi_epi16(src_2, src_3); in av1_wiener_convolve_add_src_avx2()
|
| D | warp_plane_avx2.c | 85 const __m256i src_3 = _mm256_shuffle_epi8(src, shuffle_src[3]); in filter_src_pixels_avx2() local
90 const __m256i res_57 = _mm256_maddubs_epi16(src_3, coeff[3]); in filter_src_pixels_avx2()
792 const __m256i src_3 = in warp_vertical_filter_avx2() local
799 src[2] = _mm256_unpacklo_epi16(src_2, src_3); in warp_vertical_filter_avx2()
803 src[3] = _mm256_unpackhi_epi16(src_2, src_3); in warp_vertical_filter_avx2()
842 const __m256i src_3 = in warp_vertical_filter_gamma0_avx2() local
849 src[2] = _mm256_unpacklo_epi16(src_2, src_3); in warp_vertical_filter_gamma0_avx2()
853 src[3] = _mm256_unpackhi_epi16(src_2, src_3); in warp_vertical_filter_gamma0_avx2()
891 const __m256i src_3 = in warp_vertical_filter_delta0_avx2() local
898 src[2] = _mm256_unpacklo_epi16(src_2, src_3); in warp_vertical_filter_delta0_avx2()
[all …]
|
| D | convolve_2d_avx2.c | 122 __m256i src_3 = _mm256_loadu_si256((__m256i *)(im_block + 3 * im_stride)); in av1_convolve_2d_sr_avx2() local
125 s[1] = _mm256_unpacklo_epi16(src_2, src_3); in av1_convolve_2d_sr_avx2()
127 s[4] = _mm256_unpackhi_epi16(src_2, src_3); in av1_convolve_2d_sr_avx2()
|
| D | convolve_2d_sse2.c | 92 const __m128i src_3 = _mm_unpacklo_epi8(_mm_srli_si128(data, 3), zero); in av1_convolve_2d_sr_sse2() local
93 const __m128i res_3 = _mm_madd_epi16(src_3, coeff_23); in av1_convolve_2d_sr_sse2()
169 const __m128i src_3 = in av1_convolve_2d_sr_sse2() local
180 const __m128i res_3 = _mm_madd_epi16(src_3, coeff_23); in av1_convolve_2d_sr_sse2()
|
| D | highbd_wiener_convolve_ssse3.c | 166 const __m128i src_3 = in av1_highbd_wiener_convolve_add_src_ssse3() local
177 const __m128i res_3 = _mm_madd_epi16(src_3, coeff_23); in av1_highbd_wiener_convolve_add_src_ssse3()
|
| D | jnt_convolve_sse2.c | 484 const __m128i src_3 = _mm_or_si128(temp_hi, temp_lo); in av1_dist_wtd_convolve_2d_sse2() local
485 const __m128i res_3 = _mm_madd_epi16(src_3, coeff_23); in av1_dist_wtd_convolve_2d_sse2()
561 const __m128i src_3 = in av1_dist_wtd_convolve_2d_sse2() local
572 const __m128i res_3 = _mm_madd_epi16(src_3, coeff_23); in av1_dist_wtd_convolve_2d_sse2()
|
| D | highbd_convolve_2d_sse4.c | 329 const __m128i src_3 = in av1_highbd_dist_wtd_convolve_2d_sse4_1() local
340 const __m128i res_3 = _mm_madd_epi16(src_3, coeff_23); in av1_highbd_dist_wtd_convolve_2d_sse4_1()
|
| D | highbd_warp_plane_sse4.c | 497 const __m128i src_3 = _mm_unpackhi_epi16(src[2], src[3]); in av1_highbd_warp_affine_sse4_1() local
525 const __m128i res_3 = _mm_madd_epi16(src_3, coeff_3); in av1_highbd_warp_affine_sse4_1()
|
| D | warp_plane_sse4.c | 564 const __m128i src_3 = _mm_unpackhi_epi16(src[2], src[3]); in filter_src_pixels_vertical() local
569 const __m128i res_3 = _mm_madd_epi16(src_3, coeffs[5]); in filter_src_pixels_vertical()
|
| /external/libhevc/encoder/arm/ |
| D | ihevce_copy_neon.c | 180 uint8x16_t src_0, src_1, src_2, src_3; in copy_2d_neon() local
190 src_3 = vld1q_u8(src_il + 48); in copy_2d_neon()
191 vst1q_u8(dst_il + 48, src_3); in copy_2d_neon()
|
| D | ihevce_ssd_calculator_neon.c | 179 uint8x16_t src_0, src_1, src_2, src_3; in ihevce_1x64_ssd_computer_neon() local
193 src_3 = vld1q_u8(pu1_src + 48); in ihevce_1x64_ssd_computer_neon()
204 src_3 = vld2q_u8(pu1_src + 96).val[chroma_plane]; in ihevce_1x64_ssd_computer_neon()
210 abs_3 = vabdq_u8(src_3, pred_3); in ihevce_1x64_ssd_computer_neon()
|
| D | ihevce_sad_compute_neon.c | 183 const uint8x16_t src_3 = vld1q_u8(pu1_src + 48); in ihevce_64xn_sad_computer_neon() local
192 abs_1 = vabal_u8(abs_1, vget_low_u8(src_3), vget_low_u8(pred_3)); in ihevce_64xn_sad_computer_neon()
193 abs_1 = vabal_u8(abs_1, vget_high_u8(src_3), vget_high_u8(pred_3)); in ihevce_64xn_sad_computer_neon()
|
| D | ihevce_ssd_and_sad_calculator_neon.c | 250 const uint8x16_t src_3 = vld1q_u8(pu1_src + 48); in ihevce_ssd_and_sad_calculator_neon() local
286 abs_l = vabd_u8(vget_low_u8(src_3), vget_low_u8(pred_3)); in ihevce_ssd_and_sad_calculator_neon()
287 abs_h = vabd_u8(vget_high_u8(src_3), vget_high_u8(pred_3)); in ihevce_ssd_and_sad_calculator_neon()
|
| /external/libaom/libaom/aom_dsp/x86/ |
| D | convolve_avx2.h | 91 __m256i src_3 = _mm256_loadu_si256((__m256i *)(im_block + 3 * im_stride)); \
97 s[1] = _mm256_unpacklo_epi16(src_2, src_3); \
101 s[5] = _mm256_unpackhi_epi16(src_2, src_3); \
|
| /external/webp/src/dsp/ |
| D | enc_sse2.c | 351 const __m128i src_3 = _mm_unpacklo_epi8(src3, zero); in FTransform2_SSE2() local
365 const __m128i diff3 = _mm_sub_epi16(src_3, ref_3); in FTransform2_SSE2()
|