/external/libmpeg2/common/x86/ |
D | impeg2_inter_pred_sse42_intr.c | 143 src_r0 = _mm_loadl_epi64((__m128i *)src); in impeg2_copy_mb_sse42() 144 src_r1 = _mm_loadl_epi64((__m128i *)(src + src_wd)); in impeg2_copy_mb_sse42() 145 src_r2 = _mm_loadl_epi64((__m128i *)(src + 2 * src_wd)); in impeg2_copy_mb_sse42() 146 src_r3 = _mm_loadl_epi64((__m128i *)(src + 3 * src_wd)); in impeg2_copy_mb_sse42() 157 src_r0 = _mm_loadl_epi64((__m128i *)src); in impeg2_copy_mb_sse42() 158 src_r1 = _mm_loadl_epi64((__m128i *)(src + src_wd)); in impeg2_copy_mb_sse42() 159 src_r2 = _mm_loadl_epi64((__m128i *)(src + 2 * src_wd)); in impeg2_copy_mb_sse42() 160 src_r3 = _mm_loadl_epi64((__m128i *)(src + 3 * src_wd)); in impeg2_copy_mb_sse42() 173 src_r0 = _mm_loadl_epi64((__m128i *)src); in impeg2_copy_mb_sse42() 174 src_r1 = _mm_loadl_epi64((__m128i *)(src + src_wd)); in impeg2_copy_mb_sse42() [all …]
|
D | ideint_spatial_filter_ssse3.c | 119 row1_m1 = _mm_loadl_epi64((__m128i *) (pu1_src - 1)); in ideint_spatial_filter_ssse3() 120 row1_0 = _mm_loadl_epi64((__m128i *) (pu1_src)); in ideint_spatial_filter_ssse3() 121 row1_p1 = _mm_loadl_epi64((__m128i *) (pu1_src + 1)); in ideint_spatial_filter_ssse3() 134 row2_m1 = _mm_loadl_epi64((__m128i *) (pu1_src - 1)); in ideint_spatial_filter_ssse3() 135 row2_0 = _mm_loadl_epi64((__m128i *) (pu1_src)); in ideint_spatial_filter_ssse3() 136 row2_p1 = _mm_loadl_epi64((__m128i *) (pu1_src + 1)); in ideint_spatial_filter_ssse3()
|
/external/libhevc/common/x86/ |
D | ihevc_inter_pred_filters_sse42_intr.c | 130 src0_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src)); /* row =0 */ in ihevc_inter_pred_luma_copy_w16out_sse42() 131 src1_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src + 1 * src_strd)); /* row =1 */ in ihevc_inter_pred_luma_copy_w16out_sse42() 132 src2_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src + 2 * src_strd)); /* row =2 */ in ihevc_inter_pred_luma_copy_w16out_sse42() 133 src3_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src + 3 * src_strd)); /* row =3 */ in ihevc_inter_pred_luma_copy_w16out_sse42() 166 src0_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src)); /* row =0 */ in ihevc_inter_pred_luma_copy_w16out_sse42() 167 src1_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src + 1 * src_strd)); /* row =1 */ in ihevc_inter_pred_luma_copy_w16out_sse42() 168 src2_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src + 2 * src_strd)); /* row =2 */ in ihevc_inter_pred_luma_copy_w16out_sse42() 169 src3_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src + 3 * src_strd)); /* row =3 */ in ihevc_inter_pred_luma_copy_w16out_sse42() 288 src0_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src)); /* row =0 */ in ihevc_inter_pred_chroma_copy_sse42() 289 src1_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src + 1 * src_strd)); /* row =1 */ in ihevc_inter_pred_chroma_copy_sse42() [all …]
|
D | ihevc_inter_pred_filters_ssse3_intr.c | 168 src0_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src)); /* row =0 */ in ihevc_inter_pred_luma_copy_ssse3() 169 src1_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src + 1 * src_strd)); /* row =1 */ in ihevc_inter_pred_luma_copy_ssse3() 170 src2_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src + 2 * src_strd)); /* row =2 */ in ihevc_inter_pred_luma_copy_ssse3() 171 src3_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src + 3 * src_strd)); /* row =3 */ in ihevc_inter_pred_luma_copy_ssse3() 195 src0_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src)); /* row =0 */ in ihevc_inter_pred_luma_copy_ssse3() 196 src1_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src + 1 * src_strd)); /* row =1 */ in ihevc_inter_pred_luma_copy_ssse3() 197 src2_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src + 2 * src_strd)); /* row =2 */ in ihevc_inter_pred_luma_copy_ssse3() 198 src3_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src + 3 * src_strd)); /* row =3 */ in ihevc_inter_pred_luma_copy_ssse3() 292 src_temp1_16x8b = _mm_loadl_epi64((__m128i *)pi1_coeff); in ihevc_inter_pred_luma_horz_ssse3() 453 res_temp7_8x16b = _mm_loadl_epi64((__m128i *)(pu1_dst + offset)); in ihevc_inter_pred_luma_horz_ssse3() [all …]
|
D | ihevc_deblk_ssse3_intr.c | 144 src_row0_8x16b = _mm_loadl_epi64((__m128i *)(pu1_src - 4)); in ihevc_deblk_luma_vert_ssse3() 145 src_row3_8x16b = _mm_loadl_epi64((__m128i *)((pu1_src - 4) + 3 * src_strd)); in ihevc_deblk_luma_vert_ssse3() 241 src_row1_8x16b = _mm_loadl_epi64((__m128i *)((pu1_src - 4) + src_strd)); in ihevc_deblk_luma_vert_ssse3() 242 src_row2_8x16b = _mm_loadl_epi64((__m128i *)((pu1_src - 4) + 2 * src_strd)); in ihevc_deblk_luma_vert_ssse3() 295 temp_str0_16x8b = _mm_loadl_epi64((__m128i *)(&mask)); in ihevc_deblk_luma_vert_ssse3() 296 const2tc_8x16b = _mm_loadl_epi64((__m128i *)(&tc2)); in ihevc_deblk_luma_vert_ssse3() 429 mask_pq_8x16b = _mm_loadl_epi64((__m128i *)(&mask1)); in ihevc_deblk_luma_vert_ssse3() 459 tmp2_const_8x16b = _mm_loadl_epi64((__m128i *)(shuffle0)); in ihevc_deblk_luma_vert_ssse3() 584 src_q0_8x16b = _mm_loadl_epi64((__m128i *)(pu1_src)); in ihevc_deblk_luma_horz_ssse3() 585 src_q1_8x16b = _mm_loadl_epi64((__m128i *)(pu1_src + src_strd)); in ihevc_deblk_luma_horz_ssse3() [all …]
|
/external/flac/libFLAC/ |
D | lpc_intrin_sse41.c | 70 xmm0 = _mm_loadl_epi64((const __m128i*)(qlp_coeff+0)); // 0 0 q[1] q[0] in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 71 xmm1 = _mm_loadl_epi64((const __m128i*)(qlp_coeff+2)); // 0 0 q[3] q[2] in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 72 xmm2 = _mm_loadl_epi64((const __m128i*)(qlp_coeff+4)); // 0 0 q[5] q[4] in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 73 xmm3 = _mm_loadl_epi64((const __m128i*)(qlp_coeff+6)); // 0 0 q[7] q[6] in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 74 xmm4 = _mm_loadl_epi64((const __m128i*)(qlp_coeff+8)); // 0 0 q[9] q[8] in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 75 xmm5 = _mm_loadl_epi64((const __m128i*)(qlp_coeff+10)); // 0 0 q[11] q[10] in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 88 xmm7 = _mm_loadl_epi64((const __m128i*)(data+i-12)); // 0 0 d[i-11] d[i-12] in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 94 xmm6 = _mm_loadl_epi64((const __m128i*)(data+i-10)); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 101 xmm6 = _mm_loadl_epi64((const __m128i*)(data+i-8)); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() 108 xmm6 = _mm_loadl_epi64((const __m128i*)(data+i-6)); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41() [all …]
|
D | lpc_intrin_sse2.c | 423 xmm0 = _mm_loadl_epi64((const __m128i*)(qlp_coeff+0)); // 0 0 q[1] q[0] in FLAC__lpc_compute_residual_from_qlp_coefficients_intrin_sse2() 424 xmm1 = _mm_loadl_epi64((const __m128i*)(qlp_coeff+2)); // 0 0 q[3] q[2] in FLAC__lpc_compute_residual_from_qlp_coefficients_intrin_sse2() 425 xmm2 = _mm_loadl_epi64((const __m128i*)(qlp_coeff+4)); // 0 0 q[5] q[4] in FLAC__lpc_compute_residual_from_qlp_coefficients_intrin_sse2() 426 xmm3 = _mm_loadl_epi64((const __m128i*)(qlp_coeff+6)); // 0 0 q[7] q[6] in FLAC__lpc_compute_residual_from_qlp_coefficients_intrin_sse2() 427 xmm4 = _mm_loadl_epi64((const __m128i*)(qlp_coeff+8)); // 0 0 q[9] q[8] in FLAC__lpc_compute_residual_from_qlp_coefficients_intrin_sse2() 428 xmm5 = _mm_loadl_epi64((const __m128i*)(qlp_coeff+10)); // 0 0 q[11] q[10] in FLAC__lpc_compute_residual_from_qlp_coefficients_intrin_sse2() 441 xmm7 = _mm_loadl_epi64((const __m128i*)(data+i-12)); // 0 0 d[i-11] d[i-12] in FLAC__lpc_compute_residual_from_qlp_coefficients_intrin_sse2() 447 xmm6 = _mm_loadl_epi64((const __m128i*)(data+i-10)); in FLAC__lpc_compute_residual_from_qlp_coefficients_intrin_sse2() 454 xmm6 = _mm_loadl_epi64((const __m128i*)(data+i-8)); in FLAC__lpc_compute_residual_from_qlp_coefficients_intrin_sse2() 461 xmm6 = _mm_loadl_epi64((const __m128i*)(data+i-6)); in FLAC__lpc_compute_residual_from_qlp_coefficients_intrin_sse2() [all …]
|
/external/libavc/common/x86/ |
D | ih264_iquant_itrans_recon_dc_ssse3.c | 136 …predload_r = _mm_loadl_epi64((__m128i *) (&pu1_pred[0])); //p00 p01 p02 p03 0 0 0 0 0 0 0 0 -- all… in ih264_iquant_itrans_recon_4x4_dc_ssse3() 138 …predload_r = _mm_loadl_epi64((__m128i *) (&pu1_pred[pred_strd])); //p10 p11 p12 p13 0 0 0 0 0 0 0 … in ih264_iquant_itrans_recon_4x4_dc_ssse3() 140 …predload_r = _mm_loadl_epi64((__m128i *) (&pu1_pred[2*pred_strd])); //p20 p21 p22 p23 0 0 0 0 0 0 … in ih264_iquant_itrans_recon_4x4_dc_ssse3() 142 …predload_r = _mm_loadl_epi64((__m128i *) (&pu1_pred[3*pred_strd])); //p30 p31 p32 p33 0 0 0 0 0 0 … in ih264_iquant_itrans_recon_4x4_dc_ssse3() 258 …predload_r = _mm_loadl_epi64((__m128i *)(&pu1_pred[0])); //p0 p1 p2 p3 p4 p5 p6 p7 0 0 0 0 0 0 0 0… in ih264_iquant_itrans_recon_8x8_dc_ssse3() 261 …predload_r = _mm_loadl_epi64((__m128i *)(&pu1_pred[pred_strd])); //p0 p1 p2 p3 p4 p5 p6 p7 0 0 0 0… in ih264_iquant_itrans_recon_8x8_dc_ssse3() 264 predload_r = _mm_loadl_epi64( in ih264_iquant_itrans_recon_8x8_dc_ssse3() 268 predload_r = _mm_loadl_epi64( in ih264_iquant_itrans_recon_8x8_dc_ssse3() 272 predload_r = _mm_loadl_epi64( in ih264_iquant_itrans_recon_8x8_dc_ssse3() 276 predload_r = _mm_loadl_epi64( in ih264_iquant_itrans_recon_8x8_dc_ssse3() [all …]
|
D | ih264_weighted_pred_sse42.c | 101 y0_0_16x8b = _mm_loadl_epi64((__m128i *)pu1_src1); in ih264_default_weighted_pred_luma_sse42() 102 y0_1_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src1 + src_strd1)); in ih264_default_weighted_pred_luma_sse42() 103 y0_2_16x8b = _mm_loadl_epi64( in ih264_default_weighted_pred_luma_sse42() 105 y0_3_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src1 + src_strd1 * 3)); in ih264_default_weighted_pred_luma_sse42() 107 y1_0_16x8b = _mm_loadl_epi64((__m128i *)pu1_src2); in ih264_default_weighted_pred_luma_sse42() 108 y1_1_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src2 + src_strd2)); in ih264_default_weighted_pred_luma_sse42() 109 y1_2_16x8b = _mm_loadl_epi64( in ih264_default_weighted_pred_luma_sse42() 111 y1_3_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src2 + src_strd2 * 3)); in ih264_default_weighted_pred_luma_sse42() 134 y0_0_16x8b = _mm_loadl_epi64((__m128i *)pu1_src1); in ih264_default_weighted_pred_luma_sse42() 135 y0_1_16x8b = _mm_loadl_epi64((__m128i *)(pu1_src1 + src_strd1)); in ih264_default_weighted_pred_luma_sse42() [all …]
|
/external/libaom/libaom/aom_dsp/x86/ |
D | highbd_subtract_sse2.c | 32 u0 = _mm_loadl_epi64((__m128i const *)(src + 0 * src_stride)); in subtract_4x4() 33 u1 = _mm_loadl_epi64((__m128i const *)(src + 1 * src_stride)); in subtract_4x4() 34 u2 = _mm_loadl_epi64((__m128i const *)(src + 2 * src_stride)); in subtract_4x4() 35 u3 = _mm_loadl_epi64((__m128i const *)(src + 3 * src_stride)); in subtract_4x4() 37 v0 = _mm_loadl_epi64((__m128i const *)(pred + 0 * pred_stride)); in subtract_4x4() 38 v1 = _mm_loadl_epi64((__m128i const *)(pred + 1 * pred_stride)); in subtract_4x4() 39 v2 = _mm_loadl_epi64((__m128i const *)(pred + 2 * pred_stride)); in subtract_4x4() 40 v3 = _mm_loadl_epi64((__m128i const *)(pred + 3 * pred_stride)); in subtract_4x4() 64 u0 = _mm_loadl_epi64((__m128i const *)(src + 0 * src_stride)); in subtract_4x8() 65 u1 = _mm_loadl_epi64((__m128i const *)(src + 1 * src_stride)); in subtract_4x8() [all …]
|
D | avg_intrin_sse2.c | 24 s0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(s)), u0); in aom_minmax_8x8_sse2() 25 d0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(d)), u0); in aom_minmax_8x8_sse2() 30 s0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(s + p)), u0); in aom_minmax_8x8_sse2() 31 d0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(d + dp)), u0); in aom_minmax_8x8_sse2() 38 s0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(s + 2 * p)), u0); in aom_minmax_8x8_sse2() 39 d0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(d + 2 * dp)), u0); in aom_minmax_8x8_sse2() 46 s0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(s + 3 * p)), u0); in aom_minmax_8x8_sse2() 47 d0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(d + 3 * dp)), u0); in aom_minmax_8x8_sse2() 54 s0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(s + 4 * p)), u0); in aom_minmax_8x8_sse2() 55 d0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(d + 4 * dp)), u0); in aom_minmax_8x8_sse2() [all …]
|
D | masked_sad_intrin_ssse3.c | 151 _mm_loadl_epi64((const __m128i *)src_ptr), in aom_masked_sad8xh_ssse3() 152 _mm_loadl_epi64((const __m128i *)&src_ptr[src_stride])); in aom_masked_sad8xh_ssse3() 153 const __m128i a0 = _mm_loadl_epi64((const __m128i *)a_ptr); in aom_masked_sad8xh_ssse3() 154 const __m128i a1 = _mm_loadl_epi64((const __m128i *)&a_ptr[a_stride]); in aom_masked_sad8xh_ssse3() 155 const __m128i b0 = _mm_loadl_epi64((const __m128i *)b_ptr); in aom_masked_sad8xh_ssse3() 156 const __m128i b1 = _mm_loadl_epi64((const __m128i *)&b_ptr[b_stride]); in aom_masked_sad8xh_ssse3() 158 _mm_unpacklo_epi64(_mm_loadl_epi64((const __m128i *)m_ptr), in aom_masked_sad8xh_ssse3() 159 _mm_loadl_epi64((const __m128i *)&m_ptr[m_stride])); in aom_masked_sad8xh_ssse3() 307 _mm_loadl_epi64((const __m128i *)&m_ptr[x]), _mm_setzero_si128()); in highbd_masked_sad_ssse3() 361 _mm_loadl_epi64((const __m128i *)src_ptr), in aom_highbd_masked_sad4xh_ssse3() [all …]
|
D | loopfilter_sse2.c | 333 __m128i limit = _mm_unpacklo_epi32(_mm_loadl_epi64((const __m128i *)_blimit), in aom_lpf_horizontal_4_sse2() 334 _mm_loadl_epi64((const __m128i *)_limit)); in aom_lpf_horizontal_4_sse2() 336 _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)_thresh), zero); in aom_lpf_horizontal_4_sse2() 361 __m128i limit = _mm_unpacklo_epi32(_mm_loadl_epi64((const __m128i *)_blimit), in aom_lpf_vertical_4_sse2() 362 _mm_loadl_epi64((const __m128i *)_limit)); in aom_lpf_vertical_4_sse2() 364 _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)_thresh), zero); in aom_lpf_vertical_4_sse2() 368 x0 = _mm_loadl_epi64((__m128i *)(s - 2 + 0 * p)); in aom_lpf_vertical_4_sse2() 369 x1 = _mm_loadl_epi64((__m128i *)(s - 2 + 1 * p)); in aom_lpf_vertical_4_sse2() 370 x2 = _mm_loadl_epi64((__m128i *)(s - 2 + 2 * p)); in aom_lpf_vertical_4_sse2() 371 x3 = _mm_loadl_epi64((__m128i *)(s - 2 + 3 * p)); in aom_lpf_vertical_4_sse2() [all …]
|
D | highbd_variance_sse4.c | 33 a0 = _mm_loadl_epi64((__m128i const *)(a + 0 * a_stride)); in variance4x4_64_sse4_1() 34 a1 = _mm_loadl_epi64((__m128i const *)(a + 1 * a_stride)); in variance4x4_64_sse4_1() 35 a2 = _mm_loadl_epi64((__m128i const *)(a + 2 * a_stride)); in variance4x4_64_sse4_1() 36 a3 = _mm_loadl_epi64((__m128i const *)(a + 3 * a_stride)); in variance4x4_64_sse4_1() 38 b0 = _mm_loadl_epi64((__m128i const *)(b + 0 * b_stride)); in variance4x4_64_sse4_1() 39 b1 = _mm_loadl_epi64((__m128i const *)(b + 1 * b_stride)); in variance4x4_64_sse4_1() 40 b2 = _mm_loadl_epi64((__m128i const *)(b + 2 * b_stride)); in variance4x4_64_sse4_1() 41 b3 = _mm_loadl_epi64((__m128i const *)(b + 3 * b_stride)); in variance4x4_64_sse4_1()
|
D | aom_subpixel_8t_intrin_ssse3.c | 139 srcReg2 = _mm_loadl_epi64((const __m128i *)(src_ptr + src_pitch * 2)); in aom_filter_block1d4_v4_ssse3() 140 srcReg3 = _mm_loadl_epi64((const __m128i *)(src_ptr + src_pitch * 3)); in aom_filter_block1d4_v4_ssse3() 143 srcReg4 = _mm_loadl_epi64((const __m128i *)(src_ptr + src_pitch * 4)); in aom_filter_block1d4_v4_ssse3() 151 srcReg5 = _mm_loadl_epi64((const __m128i *)(src_ptr + src_pitch * 5)); in aom_filter_block1d4_v4_ssse3() 154 srcReg6 = _mm_loadl_epi64((const __m128i *)(src_ptr + src_pitch * 6)); in aom_filter_block1d4_v4_ssse3() 348 srcReg2 = _mm_loadl_epi64((const __m128i *)(src_ptr + src_pitch * 2)); in aom_filter_block1d8_v4_ssse3() 349 srcReg3 = _mm_loadl_epi64((const __m128i *)(src_ptr + src_pitch * 3)); in aom_filter_block1d8_v4_ssse3() 352 srcReg4 = _mm_loadl_epi64((const __m128i *)(src_ptr + src_pitch * 4)); in aom_filter_block1d8_v4_ssse3() 358 srcReg5 = _mm_loadl_epi64((const __m128i *)(src_ptr + src_pitch * 5)); in aom_filter_block1d8_v4_ssse3() 362 srcReg6 = _mm_loadl_epi64((const __m128i *)(src_ptr + src_pitch * 6)); in aom_filter_block1d8_v4_ssse3() [all …]
|
D | masked_variance_intrin_ssse3.h | 61 const __m128i sA0 = _mm_loadl_epi64((const __m128i *)(src0)); in comp_mask_pred_8_ssse3() 62 const __m128i sA1 = _mm_loadl_epi64((const __m128i *)(src1)); in comp_mask_pred_8_ssse3() 63 const __m128i aA = _mm_loadl_epi64((const __m128i *)(mask)); in comp_mask_pred_8_ssse3() 65 const __m128i sB0 = _mm_loadl_epi64((const __m128i *)(src0 + stride0)); in comp_mask_pred_8_ssse3() 66 const __m128i sB1 = _mm_loadl_epi64((const __m128i *)(src1 + stride1)); in comp_mask_pred_8_ssse3()
|
/external/libhevc/decoder/x86/ |
D | ihevcd_it_rec_dc_sse42_intr.c | 97 m_temp_reg_0 = _mm_loadl_epi64((__m128i *)(pu1_pred)); in ihevcd_itrans_recon_dc_luma_sse42() 98 m_temp_reg_1 = _mm_loadl_epi64((__m128i *)(pu1_pred + pred_strd)); in ihevcd_itrans_recon_dc_luma_sse42() 99 m_temp_reg_2 = _mm_loadl_epi64((__m128i *)(pu1_pred + 2 * pred_strd)); in ihevcd_itrans_recon_dc_luma_sse42() 100 m_temp_reg_3 = _mm_loadl_epi64((__m128i *)(pu1_pred + 3 * pred_strd)); in ihevcd_itrans_recon_dc_luma_sse42() 140 m_temp_reg_0 = _mm_loadl_epi64((__m128i *)pu1_pred); in ihevcd_itrans_recon_dc_luma_sse42() 141 m_temp_reg_1 = _mm_loadl_epi64((__m128i *)(pu1_pred + pred_strd)); in ihevcd_itrans_recon_dc_luma_sse42() 142 m_temp_reg_2 = _mm_loadl_epi64((__m128i *)(pu1_pred + 2 * pred_strd)); in ihevcd_itrans_recon_dc_luma_sse42() 143 m_temp_reg_3 = _mm_loadl_epi64((__m128i *)(pu1_pred + 3 * pred_strd)); in ihevcd_itrans_recon_dc_luma_sse42() 244 m_temp_reg_0 = _mm_loadl_epi64((__m128i *)(pu1_pred)); in ihevcd_itrans_recon_dc_chroma_sse42() 245 m_temp_reg_1 = _mm_loadl_epi64((__m128i *)(pu1_pred + pred_strd)); in ihevcd_itrans_recon_dc_chroma_sse42() [all …]
|
D | ihevcd_it_rec_dc_ssse3_intr.c | 98 m_temp_reg_0 = _mm_loadl_epi64((__m128i *)(pu1_pred)); in ihevcd_itrans_recon_dc_luma_ssse3() 99 m_temp_reg_1 = _mm_loadl_epi64((__m128i *)(pu1_pred + pred_strd)); in ihevcd_itrans_recon_dc_luma_ssse3() 100 m_temp_reg_2 = _mm_loadl_epi64((__m128i *)(pu1_pred + 2 * pred_strd)); in ihevcd_itrans_recon_dc_luma_ssse3() 101 m_temp_reg_3 = _mm_loadl_epi64((__m128i *)(pu1_pred + 3 * pred_strd)); in ihevcd_itrans_recon_dc_luma_ssse3() 141 m_temp_reg_0 = _mm_loadl_epi64((__m128i *)pu1_pred); in ihevcd_itrans_recon_dc_luma_ssse3() 142 m_temp_reg_1 = _mm_loadl_epi64((__m128i *)(pu1_pred + pred_strd)); in ihevcd_itrans_recon_dc_luma_ssse3() 143 m_temp_reg_2 = _mm_loadl_epi64((__m128i *)(pu1_pred + 2 * pred_strd)); in ihevcd_itrans_recon_dc_luma_ssse3() 144 m_temp_reg_3 = _mm_loadl_epi64((__m128i *)(pu1_pred + 3 * pred_strd)); in ihevcd_itrans_recon_dc_luma_ssse3() 244 m_temp_reg_0 = _mm_loadl_epi64((__m128i *)(pu1_pred)); in ihevcd_itrans_recon_dc_chroma_ssse3() 245 m_temp_reg_1 = _mm_loadl_epi64((__m128i *)(pu1_pred + pred_strd)); in ihevcd_itrans_recon_dc_chroma_ssse3() [all …]
|
/external/libvpx/libvpx/vpx_dsp/x86/ |
D | avg_intrin_sse2.c | 23 s0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(s)), u0); in vpx_minmax_8x8_sse2() 24 d0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(d)), u0); in vpx_minmax_8x8_sse2() 29 s0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(s + p)), u0); in vpx_minmax_8x8_sse2() 30 d0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(d + dp)), u0); in vpx_minmax_8x8_sse2() 37 s0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(s + 2 * p)), u0); in vpx_minmax_8x8_sse2() 38 d0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(d + 2 * dp)), u0); in vpx_minmax_8x8_sse2() 45 s0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(s + 3 * p)), u0); in vpx_minmax_8x8_sse2() 46 d0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(d + 3 * dp)), u0); in vpx_minmax_8x8_sse2() 53 s0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(s + 4 * p)), u0); in vpx_minmax_8x8_sse2() 54 d0 = _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)(d + 4 * dp)), u0); in vpx_minmax_8x8_sse2() [all …]
|
D | loopfilter_sse2.c | 111 _mm_unpacklo_epi64(_mm_loadl_epi64((const __m128i *)blimit), in vpx_lpf_horizontal_4_sse2() 112 _mm_loadl_epi64((const __m128i *)limit)); in vpx_lpf_horizontal_4_sse2() 114 _mm_unpacklo_epi8(_mm_loadl_epi64((const __m128i *)thresh), zero); in vpx_lpf_horizontal_4_sse2() 119 p3p2 = _mm_unpacklo_epi64(_mm_loadl_epi64((__m128i *)(s - 3 * pitch)), in vpx_lpf_horizontal_4_sse2() 120 _mm_loadl_epi64((__m128i *)(s - 4 * pitch))); in vpx_lpf_horizontal_4_sse2() 121 q1p1 = _mm_unpacklo_epi64(_mm_loadl_epi64((__m128i *)(s - 2 * pitch)), in vpx_lpf_horizontal_4_sse2() 122 _mm_loadl_epi64((__m128i *)(s + 1 * pitch))); in vpx_lpf_horizontal_4_sse2() 123 q0p0 = _mm_unpacklo_epi64(_mm_loadl_epi64((__m128i *)(s - 1 * pitch)), in vpx_lpf_horizontal_4_sse2() 124 _mm_loadl_epi64((__m128i *)(s + 0 * pitch))); in vpx_lpf_horizontal_4_sse2() 125 q3q2 = _mm_unpacklo_epi64(_mm_loadl_epi64((__m128i *)(s + 2 * pitch)), in vpx_lpf_horizontal_4_sse2() [all …]
|
/external/webp/src/dsp/ |
D | enc_sse2.c | 57 in0 = _mm_loadl_epi64((const __m128i*)&in[0]); in ITransform_SSE2() 58 in1 = _mm_loadl_epi64((const __m128i*)&in[4]); in ITransform_SSE2() 59 in2 = _mm_loadl_epi64((const __m128i*)&in[8]); in ITransform_SSE2() 60 in3 = _mm_loadl_epi64((const __m128i*)&in[12]); in ITransform_SSE2() 66 const __m128i inB0 = _mm_loadl_epi64((const __m128i*)&in[16]); in ITransform_SSE2() 67 const __m128i inB1 = _mm_loadl_epi64((const __m128i*)&in[20]); in ITransform_SSE2() 68 const __m128i inB2 = _mm_loadl_epi64((const __m128i*)&in[24]); in ITransform_SSE2() 69 const __m128i inB3 = _mm_loadl_epi64((const __m128i*)&in[28]); in ITransform_SSE2() 153 ref0 = _mm_loadl_epi64((const __m128i*)&ref[0 * BPS]); in ITransform_SSE2() 154 ref1 = _mm_loadl_epi64((const __m128i*)&ref[1 * BPS]); in ITransform_SSE2() [all …]
|
/external/libaom/libaom/av1/common/x86/ |
D | convolve_sse2.c | 171 src6 = _mm_loadl_epi64((__m128i *)(data + 6 * src_stride)); in av1_convolve_y_sr_sse2() 173 _mm_loadl_epi64((__m128i *)(data + 0 * src_stride)), in av1_convolve_y_sr_sse2() 174 _mm_loadl_epi64((__m128i *)(data + 1 * src_stride))); in av1_convolve_y_sr_sse2() 176 _mm_loadl_epi64((__m128i *)(data + 1 * src_stride)), in av1_convolve_y_sr_sse2() 177 _mm_loadl_epi64((__m128i *)(data + 2 * src_stride))); in av1_convolve_y_sr_sse2() 179 _mm_loadl_epi64((__m128i *)(data + 2 * src_stride)), in av1_convolve_y_sr_sse2() 180 _mm_loadl_epi64((__m128i *)(data + 3 * src_stride))); in av1_convolve_y_sr_sse2() 182 _mm_loadl_epi64((__m128i *)(data + 3 * src_stride)), in av1_convolve_y_sr_sse2() 183 _mm_loadl_epi64((__m128i *)(data + 4 * src_stride))); in av1_convolve_y_sr_sse2() 185 _mm_loadl_epi64((__m128i *)(data + 4 * src_stride)), in av1_convolve_y_sr_sse2() [all …]
|
D | cfl_sse2.c | 36 l0 = _mm_add_epi16(_mm_loadl_epi64(src), in subtract_average_sse2() 37 _mm_loadl_epi64(src + CFL_BUF_LINE_I128)); in subtract_average_sse2() 38 __m128i l1 = _mm_add_epi16(_mm_loadl_epi64(src + 2 * CFL_BUF_LINE_I128), in subtract_average_sse2() 39 _mm_loadl_epi64(src + 3 * CFL_BUF_LINE_I128)); in subtract_average_sse2() 70 _mm_storel_epi64(dst, _mm_sub_epi16(_mm_loadl_epi64(src), avg_epi16)); in subtract_average_sse2()
|
/external/neven/Embedded/common/src/b_BasicEm/ |
D | MathSSE2.c | 55 m_XMM0 = _mm_loadl_epi64( (__m128i *)&0[vec1L] ); in bbs_dotProduct_64SSE2() 58 m_XMM1 = _mm_loadl_epi64( (__m128i *)&0[vec2L] ); in bbs_dotProduct_64SSE2() 61 m_XMM2 = _mm_loadl_epi64( (__m128i *)&4[vec1L] ); in bbs_dotProduct_64SSE2() 65 m_XMM3 = _mm_loadl_epi64( (__m128i *)&4[vec2L] ); in bbs_dotProduct_64SSE2() 66 m_XMM4 = _mm_loadl_epi64( (__m128i *)&8[vec1L] ); in bbs_dotProduct_64SSE2() 70 m_XMM5 = _mm_loadl_epi64( (__m128i *)&8[vec2L] ); in bbs_dotProduct_64SSE2() 74 m_XMM6 = _mm_loadl_epi64( (__m128i *)&12[vec1L] ); in bbs_dotProduct_64SSE2() 78 m_XMM8 = _mm_loadl_epi64( (__m128i *)&12[vec2L] ); in bbs_dotProduct_64SSE2() 93 m_XMM0 = _mm_loadl_epi64( (__m128i *)&m_XMM7 ); in bbs_dotProduct_64SSE2()
|
/external/webrtc/webrtc/modules/audio_processing/aec/ |
D | aec_rdft_sse2.c | 88 const __m128i a_00 = _mm_loadl_epi64((__m128i*)&a[j0 + 0]); in cftmdl_128_SSE2() 89 const __m128i a_08 = _mm_loadl_epi64((__m128i*)&a[j0 + 8]); in cftmdl_128_SSE2() 90 const __m128i a_32 = _mm_loadl_epi64((__m128i*)&a[j0 + 32]); in cftmdl_128_SSE2() 91 const __m128i a_40 = _mm_loadl_epi64((__m128i*)&a[j0 + 40]); in cftmdl_128_SSE2() 101 const __m128i a_16 = _mm_loadl_epi64((__m128i*)&a[j0 + 16]); in cftmdl_128_SSE2() 102 const __m128i a_24 = _mm_loadl_epi64((__m128i*)&a[j0 + 24]); in cftmdl_128_SSE2() 103 const __m128i a_48 = _mm_loadl_epi64((__m128i*)&a[j0 + 48]); in cftmdl_128_SSE2() 104 const __m128i a_56 = _mm_loadl_epi64((__m128i*)&a[j0 + 56]); in cftmdl_128_SSE2() 162 const __m128i a_00 = _mm_loadl_epi64((__m128i*)&a[j0 + 0]); in cftmdl_128_SSE2() 163 const __m128i a_08 = _mm_loadl_epi64((__m128i*)&a[j0 + 8]); in cftmdl_128_SSE2() [all …]
|