| /external/libhevc/encoder/arm/ |
| D | ihevce_copy_neon.c | 80 uint8x8_t src_0, dst_0; in ihevce_chroma_interleave_2d_copy_neon() local
83 src_0 = vld1_u8(pu1_uv_src); in ihevce_chroma_interleave_2d_copy_neon()
85 dst_0 = vbsl_u8(vreinterpret_u8_u16(select), src_0, dst_0); in ihevce_chroma_interleave_2d_copy_neon()
106 uint8x16_t src_0, dst_0; in ihevce_chroma_interleave_2d_copy_neon() local
109 src_0 = vld1q_u8(src_il); in ihevce_chroma_interleave_2d_copy_neon()
111 dst_0 = vbslq_u8(vreinterpretq_u8_u16(select), src_0, dst_0); in ihevce_chroma_interleave_2d_copy_neon()
143 uint8x8_t src_0, src_1; in copy_2d_neon() local
146 src_0 = vld1_u8(pu1_src); in copy_2d_neon()
147 vst1_u8(pu1_dst, src_0); in copy_2d_neon()
160 uint8x16_t src_0, src_1; in copy_2d_neon() local
[all …]
|
| D | ihevce_ssd_calculator_neon.c | 136 uint8x16_t src_0, pred_0, src_1, pred_1, abs_0, abs_1; in ihevce_1x32_ssd_computer_neon() local
141 src_0 = vld1q_u8(pu1_src); in ihevce_1x32_ssd_computer_neon()
148 src_0 = vld2q_u8(pu1_src).val[0]; in ihevce_1x32_ssd_computer_neon()
153 abs_0 = vabdq_u8(src_0, pred_0); in ihevce_1x32_ssd_computer_neon()
174 uint8x16_t src_0, src_1, src_2, src_3; in ihevce_1x64_ssd_computer_neon() local
182 src_0 = vld1q_u8(pu1_src); in ihevce_1x64_ssd_computer_neon()
193 src_0 = vld2q_u8(pu1_src).val[0]; in ihevce_1x64_ssd_computer_neon()
202 abs_0 = vabdq_u8(src_0, pred_0); in ihevce_1x64_ssd_computer_neon()
|
| D | ihevce_sad_compute_neon.c | 143 const uint8x16_t src_0 = vld1q_u8(pu1_src); in ihevce_32xn_sad_computer_neon() local
148 abs_0 = vabal_u8(abs_0, vget_low_u8(src_0), vget_low_u8(pred_0)); in ihevce_32xn_sad_computer_neon()
149 abs_0 = vabal_u8(abs_0, vget_high_u8(src_0), vget_high_u8(pred_0)); in ihevce_32xn_sad_computer_neon()
177 const uint8x16_t src_0 = vld1q_u8(pu1_src); in ihevce_64xn_sad_computer_neon() local
186 abs_0 = vabal_u8(abs_0, vget_low_u8(src_0), vget_low_u8(pred_0)); in ihevce_64xn_sad_computer_neon()
187 abs_0 = vabal_u8(abs_0, vget_high_u8(src_0), vget_high_u8(pred_0)); in ihevce_64xn_sad_computer_neon()
|
| D | ihevce_ssd_and_sad_calculator_neon.c | 185 const uint8x16_t src_0 = vld1q_u8(pu1_src); in ihevce_ssd_and_sad_calculator_neon() local
190 abs_l = vabd_u8(vget_low_u8(src_0), vget_low_u8(pred_0)); in ihevce_ssd_and_sad_calculator_neon()
191 abs_h = vabd_u8(vget_high_u8(src_0), vget_high_u8(pred_0)); in ihevce_ssd_and_sad_calculator_neon()
244 const uint8x16_t src_0 = vld1q_u8(pu1_src); in ihevce_ssd_and_sad_calculator_neon() local
253 abs_l = vabd_u8(vget_low_u8(src_0), vget_low_u8(pred_0)); in ihevce_ssd_and_sad_calculator_neon()
254 abs_h = vabd_u8(vget_high_u8(src_0), vget_high_u8(pred_0)); in ihevce_ssd_and_sad_calculator_neon()
|
| /external/libaom/libaom/av1/common/arm/ |
| D | warp_plane_neon.c | 201 static INLINE void convolve(int32x2x2_t x0, int32x2x2_t x1, uint8x8_t src_0, in convolve() argument
211 pix_0 = vreinterpretq_s16_u16(vmovl_u8(src_0)); in convolve()
312 int16x4_t src_0, src_1, fltr_0, fltr_1; in vertical_filter_neon() local
359 src_0 = vget_low_s16(vreinterpretq_s16_s32(c0.val[0])); in vertical_filter_neon()
361 res_0 = vmull_s16(src_0, fltr_0); in vertical_filter_neon()
364 src_0 = vget_low_s16(vreinterpretq_s16_s32(c1.val[0])); in vertical_filter_neon()
366 res_0 = vmlal_s16(res_0, src_0, fltr_0); in vertical_filter_neon()
384 src_0 = vget_low_s16(vreinterpretq_s16_s32(c2.val[0])); in vertical_filter_neon()
386 res_0 = vmull_s16(src_0, fltr_0); in vertical_filter_neon()
389 src_0 = vget_low_s16(vreinterpretq_s16_s32(c3.val[0])); in vertical_filter_neon()
[all …]
|
| /external/libvpx/libvpx/vpx_dsp/arm/ |
| D | subpel_variance_neon.c | 36 const uint8x8_t src_0 = load_unaligned_u8(src_ptr, src_pixels_per_line); in var_filter_block2d_bil_w4() local
39 const uint16x8_t a = vmull_u8(src_0, f0); in var_filter_block2d_bil_w4()
59 const uint8x8_t src_0 = vld1_u8(&src_ptr[0]); in var_filter_block2d_bil_w8() local
61 const uint16x8_t a = vmull_u8(src_0, f0); in var_filter_block2d_bil_w8()
83 const uint8x16_t src_0 = vld1q_u8(&src_ptr[j]); in var_filter_block2d_bil_w16() local
85 const uint16x8_t a = vmull_u8(vget_low_u8(src_0), f0); in var_filter_block2d_bil_w16()
88 const uint16x8_t c = vmull_u8(vget_high_u8(src_0), f0); in var_filter_block2d_bil_w16()
|
| /external/libaom/libaom/aom_dsp/arm/ |
| D | subpel_variance_neon.c | 34 const uint8x8_t src_0 = vld1_u8(&src_ptr[0]); in var_filter_block2d_bil_w8() local
36 const uint16x8_t a = vmull_u8(src_0, f0); in var_filter_block2d_bil_w8()
58 const uint8x16_t src_0 = vld1q_u8(&src_ptr[j]); in var_filter_block2d_bil_w16() local
60 const uint16x8_t a = vmull_u8(vget_low_u8(src_0), f0); in var_filter_block2d_bil_w16()
63 const uint16x8_t c = vmull_u8(vget_high_u8(src_0), f0); in var_filter_block2d_bil_w16()
|
| D | blend_a64_mask_neon.c | 22 static INLINE void blend8x1(int16x8_t mask, int16x8_t src_0, int16x8_t src_1, in blend8x1() argument
27 im_res_low = vmull_s16(vget_low_s16(mask), vget_low_s16(src_0)); in blend8x1()
31 im_res_high = vmull_s16(vget_high_s16(mask), vget_high_s16(src_0)); in blend8x1()
|
| /external/libvpx/libvpx/vpx_dsp/x86/ |
| D | variance_avx2.c | 264 const __m256i src_0 = _mm256_loadu_si256((__m256i const *)src); in spv32_half_zero() local
266 const __m256i src_avg = _mm256_avg_epu8(src_0, src_1); in spv32_half_zero()
315 const __m256i src_0 = _mm256_loadu_si256((__m256i const *)(src)); in spv32_x4_y4() local
317 const __m256i src_avg = _mm256_avg_epu8(src_0, src_1); in spv32_x4_y4()
353 const __m256i src_0 = _mm256_loadu_si256((__m256i const *)src); in spv32_bilin_zero() local
355 exp_src_lo = _mm256_unpacklo_epi8(src_0, src_1); in spv32_bilin_zero()
356 exp_src_hi = _mm256_unpackhi_epi8(src_0, src_1); in spv32_bilin_zero()
408 const __m256i src_0 = _mm256_loadu_si256((__m256i const *)src); in spv32_x4_yb() local
410 const __m256i src_avg = _mm256_avg_epu8(src_0, src_1); in spv32_x4_yb()
453 const __m256i src_0 = _mm256_loadu_si256((__m256i const *)src); in spv32_xb_y4() local
[all …]
|
| D | convolve_sse2.h | 58 static INLINE __m128i mm_madd_packs_epi16_sse2(const __m128i *const src_0, in mm_madd_packs_epi16_sse2() argument
61 const __m128i madd_1 = _mm_madd_epi16(*src_0, *ker); in mm_madd_packs_epi16_sse2()
|
| D | convolve_avx2.h | 151 static INLINE __m256i mm256_madd_add_epi32(const __m256i *const src_0, in mm256_madd_add_epi32() argument
155 const __m256i tmp_0 = _mm256_madd_epi16(*src_0, *ker_0); in mm256_madd_add_epi32()
|
| /external/libaom/libaom/av1/common/x86/ |
| D | wiener_convolve_sse2.c | 73 const __m128i src_0 = _mm_unpacklo_epi8(data, zero); in av1_wiener_convolve_add_src_sse2() local
74 const __m128i res_0 = _mm_madd_epi16(src_0, coeff_01); in av1_wiener_convolve_add_src_sse2()
139 const __m128i src_0 = in av1_wiener_convolve_add_src_sse2() local
152 const __m128i res_0 = _mm_madd_epi16(src_0, coeff_01); in av1_wiener_convolve_add_src_sse2()
|
| D | highbd_wiener_convolve_avx2.c | 95 const __m256i src_0 = yy_loadu_256(src_ij + 0); in av1_highbd_wiener_convolve_add_src_avx2() local
105 const __m256i res_0 = _mm256_madd_epi16(src_0, coeffs_01); in av1_highbd_wiener_convolve_add_src_avx2()
191 const __m256i src_0 = _mm256_unpacklo_epi16(data_0, data_1); in av1_highbd_wiener_convolve_add_src_avx2() local
196 const __m256i res_0 = _mm256_madd_epi16(src_0, coeffs_01); in av1_highbd_wiener_convolve_add_src_avx2()
|
| D | wiener_convolve_avx2.c | 103 const __m256i src_0 = _mm256_cvtepu8_epi16(data_0); in av1_wiener_convolve_add_src_avx2() local
113 const __m256i res_0 = _mm256_madd_epi16(src_0, coeffs_01); in av1_wiener_convolve_add_src_avx2()
197 const __m256i src_0 = _mm256_unpacklo_epi16(data_0, data_1); in av1_wiener_convolve_add_src_avx2() local
202 const __m256i res_0 = _mm256_madd_epi16(src_0, coeffs_01); in av1_wiener_convolve_add_src_avx2()
|
| D | selfguided_avx2.c | 660 __m128i src_0, src_1; in apply_selfguided_restoration_avx2() local
662 src_0 = xx_loadu_128(CONVERT_TO_SHORTPTR(dat8ij)); in apply_selfguided_restoration_avx2()
664 ep_0 = _mm256_cvtepu16_epi32(src_0); in apply_selfguided_restoration_avx2()
667 src_0 = xx_loadu_128(dat8ij); in apply_selfguided_restoration_avx2()
668 ep_0 = _mm256_cvtepu8_epi32(src_0); in apply_selfguided_restoration_avx2()
669 ep_1 = _mm256_cvtepu8_epi32(_mm_srli_si128(src_0, 8)); in apply_selfguided_restoration_avx2()
|
| D | convolve_2d_avx2.c | 119 __m256i src_0 = _mm256_loadu_si256((__m256i *)(im_block + 0 * im_stride)); in av1_convolve_2d_sr_avx2() local
124 s[0] = _mm256_unpacklo_epi16(src_0, src_1); in av1_convolve_2d_sr_avx2()
126 s[3] = _mm256_unpackhi_epi16(src_0, src_1); in av1_convolve_2d_sr_avx2()
|
| D | convolve_2d_sse2.c | 75 const __m128i src_0 = _mm_unpacklo_epi8(data, zero); in av1_convolve_2d_sr_sse2() local
76 const __m128i res_0 = _mm_madd_epi16(src_0, coeff_01); in av1_convolve_2d_sr_sse2()
144 const __m128i src_0 = in av1_convolve_2d_sr_sse2() local
157 const __m128i res_0 = _mm_madd_epi16(src_0, coeff_01); in av1_convolve_2d_sr_sse2()
|
| D | highbd_wiener_convolve_ssse3.c | 141 const __m128i src_0 = in av1_highbd_wiener_convolve_add_src_ssse3() local
154 const __m128i res_0 = _mm_madd_epi16(src_0, coeff_01); in av1_highbd_wiener_convolve_add_src_ssse3()
|
| D | jnt_convolve_ssse3.c | 152 const __m128i src_0 = in av1_dist_wtd_convolve_2d_ssse3() local
165 const __m128i res_0 = _mm_madd_epi16(src_0, coeff_01); in av1_dist_wtd_convolve_2d_ssse3()
|
| D | highbd_convolve_2d_sse4.c | 304 const __m128i src_0 = in av1_highbd_dist_wtd_convolve_2d_sse4_1() local
317 const __m128i res_0 = _mm_madd_epi16(src_0, coeff_01); in av1_highbd_dist_wtd_convolve_2d_sse4_1()
|
| D | highbd_warp_plane_sse4.c | 450 const __m128i src_0 = _mm_unpacklo_epi16(src[0], src[1]); in av1_highbd_warp_affine_sse4_1() local
479 const __m128i res_0 = _mm_madd_epi16(src_0, coeff_0); in av1_highbd_warp_affine_sse4_1()
|
| D | jnt_convolve_sse2.c | 536 const __m128i src_0 = in av1_dist_wtd_convolve_2d_sse2() local
549 const __m128i res_0 = _mm_madd_epi16(src_0, coeff_01); in av1_dist_wtd_convolve_2d_sse2()
|
| D | warp_plane_sse4.c | 528 const __m128i src_0 = _mm_unpacklo_epi16(src[0], src[1]); in filter_src_pixels_vertical() local
533 const __m128i res_0 = _mm_madd_epi16(src_0, coeffs[0]); in filter_src_pixels_vertical()
|
| /external/libaom/libaom/aom_dsp/x86/ |
| D | convolve_avx2.h | 63 __m256i src_0 = _mm256_loadu_si256((__m256i *)(im_block + 0 * im_stride)); \
71 s[0] = _mm256_unpacklo_epi16(src_0, src_1); \
75 s[4] = _mm256_unpackhi_epi16(src_0, src_1); \
|
| /external/webp/src/dsp/ |
| D | enc_sse2.c | 308 const __m128i src_0 = _mm_unpacklo_epi16(src0, src1); in FTransform_SSE2() local
322 const __m128i src_0_16b = _mm_unpacklo_epi8(src_0, zero); in FTransform_SSE2()
348 const __m128i src_0 = _mm_unpacklo_epi8(src0, zero); in FTransform2_SSE2() local
362 const __m128i diff0 = _mm_sub_epi16(src_0, ref_0); in FTransform2_SSE2()
|