/external/libhevc/common/x86/ |
D | ihevc_itrans_recon_32x32_ssse3_intr.c | 260 m_temp_reg_70 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_32x32_ssse3() 262 m_temp_reg_71 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_32x32_ssse3() 264 m_temp_reg_72 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_32x32_ssse3() 266 m_temp_reg_73 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_32x32_ssse3() 268 m_temp_reg_74 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_32x32_ssse3() 270 m_temp_reg_75 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_32x32_ssse3() 272 m_temp_reg_76 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_32x32_ssse3() 274 m_temp_reg_77 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_32x32_ssse3() 277 m_temp_reg_80 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_32x32_ssse3() 279 m_temp_reg_81 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_32x32_ssse3() [all …]
|
D | ihevc_itrans_recon_16x16_ssse3_intr.c | 210 m_temp_reg_70 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_16x16_ssse3() 212 m_temp_reg_71 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_16x16_ssse3() 214 m_temp_reg_72 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_16x16_ssse3() 216 m_temp_reg_73 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_16x16_ssse3() 218 m_temp_reg_74 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_16x16_ssse3() 220 m_temp_reg_75 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_16x16_ssse3() 222 m_temp_reg_76 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_16x16_ssse3() 224 m_temp_reg_77 = _mm_load_si128((__m128i *)pi2_tmp_src); in ihevc_itrans_recon_16x16_ssse3() 249 m_coeff1 = _mm_load_si128((__m128i *)&g_ai2_ihevc_trans_16_even[2][0]); //89 75 in ihevc_itrans_recon_16x16_ssse3() 295 m_coeff3 = _mm_load_si128((__m128i *)&g_ai2_ihevc_trans_16_even[3][0]); //75 -18 in ihevc_itrans_recon_16x16_ssse3() [all …]
|
D | ihevc_itrans_recon_ssse3_intr.c | 1035 m_temp_reg_70 = _mm_load_si128((__m128i *)pi2_src); in ihevc_itrans_recon_8x8_ssse3() 1037 m_temp_reg_71 = _mm_load_si128((__m128i *)pi2_src); in ihevc_itrans_recon_8x8_ssse3() 1039 m_temp_reg_72 = _mm_load_si128((__m128i *)pi2_src); in ihevc_itrans_recon_8x8_ssse3() 1041 m_temp_reg_73 = _mm_load_si128((__m128i *)pi2_src); in ihevc_itrans_recon_8x8_ssse3() 1044 m_temp_reg_74 = _mm_load_si128((__m128i *)pi2_src); in ihevc_itrans_recon_8x8_ssse3() 1046 m_temp_reg_75 = _mm_load_si128((__m128i *)pi2_src); in ihevc_itrans_recon_8x8_ssse3() 1048 m_temp_reg_76 = _mm_load_si128((__m128i *)pi2_src); in ihevc_itrans_recon_8x8_ssse3() 1050 m_temp_reg_77 = _mm_load_si128((__m128i *)pi2_src); in ihevc_itrans_recon_8x8_ssse3() 1061 m_coeff2 = _mm_load_si128((__m128i *)&g_ai2_ihevc_trans_intr_even_8[3][0]); in ihevc_itrans_recon_8x8_ssse3() 1062 m_coeff1 = _mm_load_si128((__m128i *)&g_ai2_ihevc_trans_intr_even_8[0][0]); in ihevc_itrans_recon_8x8_ssse3() [all …]
|
/external/libvpx/libvpx/vpx_dsp/x86/ |
D | fwd_txfm_sse2.c | 48 __m128i in0 = _mm_load_si128((const __m128i *)(input + 0 * stride)); in vpx_fdct8x8_1_sse2() 49 __m128i in1 = _mm_load_si128((const __m128i *)(input + 1 * stride)); in vpx_fdct8x8_1_sse2() 50 __m128i in2 = _mm_load_si128((const __m128i *)(input + 2 * stride)); in vpx_fdct8x8_1_sse2() 51 __m128i in3 = _mm_load_si128((const __m128i *)(input + 3 * stride)); in vpx_fdct8x8_1_sse2() 57 in0 = _mm_load_si128((const __m128i *)(input + 4 * stride)); in vpx_fdct8x8_1_sse2() 58 in1 = _mm_load_si128((const __m128i *)(input + 5 * stride)); in vpx_fdct8x8_1_sse2() 59 in2 = _mm_load_si128((const __m128i *)(input + 6 * stride)); in vpx_fdct8x8_1_sse2() 60 in3 = _mm_load_si128((const __m128i *)(input + 7 * stride)); in vpx_fdct8x8_1_sse2() 95 in0 = _mm_load_si128((const __m128i *)(input + 0 * stride + 0)); in vpx_fdct16x16_1_sse2() 96 in1 = _mm_load_si128((const __m128i *)(input + 0 * stride + 8)); in vpx_fdct16x16_1_sse2() [all …]
|
D | highbd_idct8x8_add_sse4.c | 100 io[0] = _mm_load_si128((const __m128i *)(input + 0 * 8 + 0)); in vpx_highbd_idct8x8_64_add_sse4_1() 101 io[4] = _mm_load_si128((const __m128i *)(input + 0 * 8 + 4)); in vpx_highbd_idct8x8_64_add_sse4_1() 102 io[1] = _mm_load_si128((const __m128i *)(input + 1 * 8 + 0)); in vpx_highbd_idct8x8_64_add_sse4_1() 103 io[5] = _mm_load_si128((const __m128i *)(input + 1 * 8 + 4)); in vpx_highbd_idct8x8_64_add_sse4_1() 104 io[2] = _mm_load_si128((const __m128i *)(input + 2 * 8 + 0)); in vpx_highbd_idct8x8_64_add_sse4_1() 105 io[6] = _mm_load_si128((const __m128i *)(input + 2 * 8 + 4)); in vpx_highbd_idct8x8_64_add_sse4_1() 106 io[3] = _mm_load_si128((const __m128i *)(input + 3 * 8 + 0)); in vpx_highbd_idct8x8_64_add_sse4_1() 107 io[7] = _mm_load_si128((const __m128i *)(input + 3 * 8 + 4)); in vpx_highbd_idct8x8_64_add_sse4_1() 116 io[8] = _mm_load_si128((const __m128i *)(input + 4 * 8 + 0)); in vpx_highbd_idct8x8_64_add_sse4_1() 117 io[12] = _mm_load_si128((const __m128i *)(input + 4 * 8 + 4)); in vpx_highbd_idct8x8_64_add_sse4_1() [all …]
|
D | highbd_idct8x8_add_sse2.c | 98 io[0] = _mm_load_si128((const __m128i *)(input + 0 * 8 + 0)); in vpx_highbd_idct8x8_64_add_sse2() 99 io[4] = _mm_load_si128((const __m128i *)(input + 0 * 8 + 4)); in vpx_highbd_idct8x8_64_add_sse2() 100 io[1] = _mm_load_si128((const __m128i *)(input + 1 * 8 + 0)); in vpx_highbd_idct8x8_64_add_sse2() 101 io[5] = _mm_load_si128((const __m128i *)(input + 1 * 8 + 4)); in vpx_highbd_idct8x8_64_add_sse2() 102 io[2] = _mm_load_si128((const __m128i *)(input + 2 * 8 + 0)); in vpx_highbd_idct8x8_64_add_sse2() 103 io[6] = _mm_load_si128((const __m128i *)(input + 2 * 8 + 4)); in vpx_highbd_idct8x8_64_add_sse2() 104 io[3] = _mm_load_si128((const __m128i *)(input + 3 * 8 + 0)); in vpx_highbd_idct8x8_64_add_sse2() 105 io[7] = _mm_load_si128((const __m128i *)(input + 3 * 8 + 4)); in vpx_highbd_idct8x8_64_add_sse2() 114 io[8] = _mm_load_si128((const __m128i *)(input + 4 * 8 + 0)); in vpx_highbd_idct8x8_64_add_sse2() 115 io[12] = _mm_load_si128((const __m128i *)(input + 4 * 8 + 4)); in vpx_highbd_idct8x8_64_add_sse2() [all …]
|
D | highbd_intrapred_intrin_ssse3.c | 67 const __m128i ABCDEFGH = _mm_load_si128((const __m128i *)above); in vpx_highbd_d45_predictor_8x8_ssse3() 99 const __m128i A0 = _mm_load_si128((const __m128i *)above); in vpx_highbd_d45_predictor_16x16_ssse3() 100 const __m128i A1 = _mm_load_si128((const __m128i *)(above + 8)); in vpx_highbd_d45_predictor_16x16_ssse3() 134 const __m128i A0 = _mm_load_si128((const __m128i *)above); in vpx_highbd_d45_predictor_32x32_ssse3() 135 const __m128i A1 = _mm_load_si128((const __m128i *)(above + 8)); in vpx_highbd_d45_predictor_32x32_ssse3() 136 const __m128i A2 = _mm_load_si128((const __m128i *)(above + 16)); in vpx_highbd_d45_predictor_32x32_ssse3() 137 const __m128i A3 = _mm_load_si128((const __m128i *)(above + 24)); in vpx_highbd_d45_predictor_32x32_ssse3() 185 const __m128i rotrw = _mm_load_si128((const __m128i *)rotate_right_epu16); in vpx_highbd_d117_predictor_8x8_ssse3() 187 const __m128i ABCDEFGH = _mm_load_si128((const __m128i *)above); in vpx_highbd_d117_predictor_8x8_ssse3() 188 const __m128i IJKLMNOP = _mm_load_si128((const __m128i *)left); in vpx_highbd_d117_predictor_8x8_ssse3() [all …]
|
D | highbd_loopfilter_sse2.c | 73 blimit_v = _mm_unpacklo_epi8(_mm_load_si128((const __m128i *)blimit), zero); in vpx_highbd_lpf_horizontal_16_sse2() 74 limit_v = _mm_unpacklo_epi8(_mm_load_si128((const __m128i *)limit), zero); in vpx_highbd_lpf_horizontal_16_sse2() 75 thresh_v = _mm_unpacklo_epi8(_mm_load_si128((const __m128i *)thresh), zero); in vpx_highbd_lpf_horizontal_16_sse2() 78 _mm_unpacklo_epi8(_mm_load_si128((const __m128i *)blimit), zero), 2); in vpx_highbd_lpf_horizontal_16_sse2() 80 _mm_unpacklo_epi8(_mm_load_si128((const __m128i *)limit), zero), 2); in vpx_highbd_lpf_horizontal_16_sse2() 82 _mm_unpacklo_epi8(_mm_load_si128((const __m128i *)thresh), zero), 2); in vpx_highbd_lpf_horizontal_16_sse2() 85 _mm_unpacklo_epi8(_mm_load_si128((const __m128i *)blimit), zero), 4); in vpx_highbd_lpf_horizontal_16_sse2() 87 _mm_unpacklo_epi8(_mm_load_si128((const __m128i *)limit), zero), 4); in vpx_highbd_lpf_horizontal_16_sse2() 89 _mm_unpacklo_epi8(_mm_load_si128((const __m128i *)thresh), zero), 4); in vpx_highbd_lpf_horizontal_16_sse2() 92 q4 = _mm_load_si128((__m128i *)(s + 4 * pitch)); in vpx_highbd_lpf_horizontal_16_sse2() [all …]
|
D | sum_squares_sse2.c | 48 _mm_load_si128((const __m128i *)(b + 0 * stride)); in vpx_sum_squares_2d_i16_sse2() 50 _mm_load_si128((const __m128i *)(b + 1 * stride)); in vpx_sum_squares_2d_i16_sse2() 52 _mm_load_si128((const __m128i *)(b + 2 * stride)); in vpx_sum_squares_2d_i16_sse2() 54 _mm_load_si128((const __m128i *)(b + 3 * stride)); in vpx_sum_squares_2d_i16_sse2() 56 _mm_load_si128((const __m128i *)(b + 4 * stride)); in vpx_sum_squares_2d_i16_sse2() 58 _mm_load_si128((const __m128i *)(b + 5 * stride)); in vpx_sum_squares_2d_i16_sse2() 60 _mm_load_si128((const __m128i *)(b + 6 * stride)); in vpx_sum_squares_2d_i16_sse2() 62 _mm_load_si128((const __m128i *)(b + 7 * stride)); in vpx_sum_squares_2d_i16_sse2()
|
D | fwd_txfm_impl_sse2.h | 277 __m128i in0 = _mm_load_si128((const __m128i *)(input + 0 * stride)); in FDCT8x8_2D() 278 __m128i in1 = _mm_load_si128((const __m128i *)(input + 1 * stride)); in FDCT8x8_2D() 279 __m128i in2 = _mm_load_si128((const __m128i *)(input + 2 * stride)); in FDCT8x8_2D() 280 __m128i in3 = _mm_load_si128((const __m128i *)(input + 3 * stride)); in FDCT8x8_2D() 281 __m128i in4 = _mm_load_si128((const __m128i *)(input + 4 * stride)); in FDCT8x8_2D() 282 __m128i in5 = _mm_load_si128((const __m128i *)(input + 5 * stride)); in FDCT8x8_2D() 283 __m128i in6 = _mm_load_si128((const __m128i *)(input + 6 * stride)); in FDCT8x8_2D() 284 __m128i in7 = _mm_load_si128((const __m128i *)(input + 7 * stride)); in FDCT8x8_2D() 624 in00 = _mm_load_si128((const __m128i *)(in + 0 * stride)); in FDCT16x16_2D() 625 in01 = _mm_load_si128((const __m128i *)(in + 1 * stride)); in FDCT16x16_2D() [all …]
|
D | quantize_sse2.h | 24 *zbin = _mm_load_si128((const __m128i *)zbin_ptr); in load_b_values() 25 *round = _mm_load_si128((const __m128i *)round_ptr); in load_b_values() 26 *quant = _mm_load_si128((const __m128i *)quant_ptr); in load_b_values() 28 *dequant = _mm_load_si128((const __m128i *)dequant_ptr); in load_b_values() 29 *shift = _mm_load_si128((const __m128i *)shift_ptr); in load_b_values() 74 __m128i scan0 = _mm_load_si128((const __m128i *)(scan + index)); in scan_for_eob() 75 __m128i scan1 = _mm_load_si128((const __m128i *)(scan + index + 8)); in scan_for_eob()
|
D | highbd_inv_txfm_sse2.h | 281 d = _mm_load_si128((const __m128i *)(&dest[j])); in highbd_idct_1_add_kernel() 321 d = _mm_load_si128((const __m128i *)(*dest)); in recon_and_store_8() 340 const __m128i t0 = _mm_load_si128((const __m128i *)(input + 0)); in load_pack_8_32bit() 341 const __m128i t1 = _mm_load_si128((const __m128i *)(input + 4)); in load_pack_8_32bit() 362 in[0] = _mm_load_si128((const __m128i *)(input + 0 * stride + 0)); in highbd_load_transpose_32bit_8x4() 363 in[1] = _mm_load_si128((const __m128i *)(input + 0 * stride + 4)); in highbd_load_transpose_32bit_8x4() 364 in[2] = _mm_load_si128((const __m128i *)(input + 1 * stride + 0)); in highbd_load_transpose_32bit_8x4() 365 in[3] = _mm_load_si128((const __m128i *)(input + 1 * stride + 4)); in highbd_load_transpose_32bit_8x4() 366 in[4] = _mm_load_si128((const __m128i *)(input + 2 * stride + 0)); in highbd_load_transpose_32bit_8x4() 367 in[5] = _mm_load_si128((const __m128i *)(input + 2 * stride + 4)); in highbd_load_transpose_32bit_8x4() [all …]
|
D | mem_sse2.h | 78 d[0] = _mm_load_si128((const __m128i *)(s + 0 * stride)); in load_8bit_16x8() 79 d[1] = _mm_load_si128((const __m128i *)(s + 1 * stride)); in load_8bit_16x8() 80 d[2] = _mm_load_si128((const __m128i *)(s + 2 * stride)); in load_8bit_16x8() 81 d[3] = _mm_load_si128((const __m128i *)(s + 3 * stride)); in load_8bit_16x8() 82 d[4] = _mm_load_si128((const __m128i *)(s + 4 * stride)); in load_8bit_16x8() 83 d[5] = _mm_load_si128((const __m128i *)(s + 5 * stride)); in load_8bit_16x8() 84 d[6] = _mm_load_si128((const __m128i *)(s + 6 * stride)); in load_8bit_16x8() 85 d[7] = _mm_load_si128((const __m128i *)(s + 7 * stride)); in load_8bit_16x8()
|
D | highbd_idct4x4_add_sse4.c | 23 io[0] = _mm_load_si128((const __m128i *)(input + 0)); in vpx_highbd_idct4x4_16_add_sse4_1() 24 io[1] = _mm_load_si128((const __m128i *)(input + 4)); in vpx_highbd_idct4x4_16_add_sse4_1() 25 io[2] = _mm_load_si128((const __m128i *)(input + 8)); in vpx_highbd_idct4x4_16_add_sse4_1() 26 io[3] = _mm_load_si128((const __m128i *)(input + 12)); in vpx_highbd_idct4x4_16_add_sse4_1()
|
D | avg_intrin_sse2.c | 271 src[0] = _mm_load_si128((const __m128i *)src_diff); in hadamard_8x8_sse2() 272 src[1] = _mm_load_si128((const __m128i *)(src_diff += src_stride)); in hadamard_8x8_sse2() 273 src[2] = _mm_load_si128((const __m128i *)(src_diff += src_stride)); in hadamard_8x8_sse2() 274 src[3] = _mm_load_si128((const __m128i *)(src_diff += src_stride)); in hadamard_8x8_sse2() 275 src[4] = _mm_load_si128((const __m128i *)(src_diff += src_stride)); in hadamard_8x8_sse2() 276 src[5] = _mm_load_si128((const __m128i *)(src_diff += src_stride)); in hadamard_8x8_sse2() 277 src[6] = _mm_load_si128((const __m128i *)(src_diff += src_stride)); in hadamard_8x8_sse2() 278 src[7] = _mm_load_si128((const __m128i *)(src_diff += src_stride)); in hadamard_8x8_sse2() 347 __m128i coeff0 = _mm_load_si128((const __m128i *)t_coeff); in hadamard_16x16_sse2() 348 __m128i coeff1 = _mm_load_si128((const __m128i *)(t_coeff + 64)); in hadamard_16x16_sse2() [all …]
|
/external/libvpx/libvpx/vp8/encoder/x86/ |
D | vp8_quantize_sse2.c | 41 __m128i quant_shift0 = _mm_load_si128((__m128i *)(b->quant_shift)); in vp8_regular_quantize_b_sse2() 42 __m128i quant_shift1 = _mm_load_si128((__m128i *)(b->quant_shift + 8)); in vp8_regular_quantize_b_sse2() 43 __m128i z0 = _mm_load_si128((__m128i *)(b->coeff)); in vp8_regular_quantize_b_sse2() 44 __m128i z1 = _mm_load_si128((__m128i *)(b->coeff + 8)); in vp8_regular_quantize_b_sse2() 46 __m128i zbin0 = _mm_load_si128((__m128i *)(b->zbin)); in vp8_regular_quantize_b_sse2() 47 __m128i zbin1 = _mm_load_si128((__m128i *)(b->zbin + 8)); in vp8_regular_quantize_b_sse2() 48 __m128i round0 = _mm_load_si128((__m128i *)(b->round)); in vp8_regular_quantize_b_sse2() 49 __m128i round1 = _mm_load_si128((__m128i *)(b->round + 8)); in vp8_regular_quantize_b_sse2() 50 __m128i quant0 = _mm_load_si128((__m128i *)(b->quant)); in vp8_regular_quantize_b_sse2() 51 __m128i quant1 = _mm_load_si128((__m128i *)(b->quant + 8)); in vp8_regular_quantize_b_sse2() [all …]
|
D | quantize_sse4.c | 36 __m128i quant_shift0 = _mm_load_si128((__m128i *)(b->quant_shift)); in vp8_regular_quantize_b_sse4_1() 37 __m128i quant_shift1 = _mm_load_si128((__m128i *)(b->quant_shift + 8)); in vp8_regular_quantize_b_sse4_1() 38 __m128i z0 = _mm_load_si128((__m128i *)(b->coeff)); in vp8_regular_quantize_b_sse4_1() 39 __m128i z1 = _mm_load_si128((__m128i *)(b->coeff + 8)); in vp8_regular_quantize_b_sse4_1() 41 __m128i zbin0 = _mm_load_si128((__m128i *)(b->zbin)); in vp8_regular_quantize_b_sse4_1() 42 __m128i zbin1 = _mm_load_si128((__m128i *)(b->zbin + 8)); in vp8_regular_quantize_b_sse4_1() 43 __m128i round0 = _mm_load_si128((__m128i *)(b->round)); in vp8_regular_quantize_b_sse4_1() 44 __m128i round1 = _mm_load_si128((__m128i *)(b->round + 8)); in vp8_regular_quantize_b_sse4_1() 45 __m128i quant0 = _mm_load_si128((__m128i *)(b->quant)); in vp8_regular_quantize_b_sse4_1() 46 __m128i quant1 = _mm_load_si128((__m128i *)(b->quant + 8)); in vp8_regular_quantize_b_sse4_1() [all …]
|
D | vp8_quantize_ssse3.c | 44 __m128i z0 = _mm_load_si128((__m128i *)(b->coeff)); in vp8_fast_quantize_b_ssse3() 45 __m128i z1 = _mm_load_si128((__m128i *)(b->coeff + 8)); in vp8_fast_quantize_b_ssse3() 46 __m128i round0 = _mm_load_si128((__m128i *)(b->round)); in vp8_fast_quantize_b_ssse3() 47 __m128i round1 = _mm_load_si128((__m128i *)(b->round + 8)); in vp8_fast_quantize_b_ssse3() 48 __m128i quant_fast0 = _mm_load_si128((__m128i *)(b->quant_fast)); in vp8_fast_quantize_b_ssse3() 49 __m128i quant_fast1 = _mm_load_si128((__m128i *)(b->quant_fast + 8)); in vp8_fast_quantize_b_ssse3() 50 __m128i dequant0 = _mm_load_si128((__m128i *)(d->dequant)); in vp8_fast_quantize_b_ssse3() 51 __m128i dequant1 = _mm_load_si128((__m128i *)(d->dequant + 8)); in vp8_fast_quantize_b_ssse3() 58 __m128i zig_zag = _mm_load_si128((const __m128i *)pshufb_zig_zag_mask); in vp8_fast_quantize_b_ssse3()
|
/external/libvpx/libvpx/vp9/common/x86/ |
D | vp9_highbd_iht8x8_add_sse4.c | 177 io[0] = _mm_load_si128((const __m128i *)(input + 0 * 8 + 0)); in vp9_highbd_iht8x8_64_add_sse4_1() 178 io[4] = _mm_load_si128((const __m128i *)(input + 0 * 8 + 4)); in vp9_highbd_iht8x8_64_add_sse4_1() 179 io[1] = _mm_load_si128((const __m128i *)(input + 1 * 8 + 0)); in vp9_highbd_iht8x8_64_add_sse4_1() 180 io[5] = _mm_load_si128((const __m128i *)(input + 1 * 8 + 4)); in vp9_highbd_iht8x8_64_add_sse4_1() 181 io[2] = _mm_load_si128((const __m128i *)(input + 2 * 8 + 0)); in vp9_highbd_iht8x8_64_add_sse4_1() 182 io[6] = _mm_load_si128((const __m128i *)(input + 2 * 8 + 4)); in vp9_highbd_iht8x8_64_add_sse4_1() 183 io[3] = _mm_load_si128((const __m128i *)(input + 3 * 8 + 0)); in vp9_highbd_iht8x8_64_add_sse4_1() 184 io[7] = _mm_load_si128((const __m128i *)(input + 3 * 8 + 4)); in vp9_highbd_iht8x8_64_add_sse4_1() 185 io[8] = _mm_load_si128((const __m128i *)(input + 4 * 8 + 0)); in vp9_highbd_iht8x8_64_add_sse4_1() 186 io[12] = _mm_load_si128((const __m128i *)(input + 4 * 8 + 4)); in vp9_highbd_iht8x8_64_add_sse4_1() [all …]
|
/external/libaom/libaom/av1/encoder/x86/ |
D | av1_quantize_sse2.c | 23 const __m128i x0 = _mm_load_si128((const __m128i *)addr); in read_coeff() 24 const __m128i x1 = _mm_load_si128((const __m128i *)addr + 1); in read_coeff() 25 const __m128i x2 = _mm_load_si128((const __m128i *)addr + 2); in read_coeff() 26 const __m128i x3 = _mm_load_si128((const __m128i *)addr + 3); in read_coeff() 30 *c0 = _mm_load_si128((const __m128i *)addr); in read_coeff() 31 *c1 = _mm_load_si128((const __m128i *)addr + 1); in read_coeff() 122 _mm_load_si128((const __m128i *)(iscan_ptr + n_coeffs)); in quantize() 124 _mm_load_si128((const __m128i *)(iscan_ptr + n_coeffs) + 1); in quantize() 155 const __m128i round0 = _mm_load_si128((const __m128i *)round_ptr); in av1_quantize_fp_sse2() 157 const __m128i quant0 = _mm_load_si128((const __m128i *)quant_ptr); in av1_quantize_fp_sse2() [all …]
|
/external/libaom/libaom/aom_dsp/x86/ |
D | quantize_x86.h | 21 *zbin = _mm_load_si128((const __m128i *)zbin_ptr); in load_b_values() 22 *round = _mm_load_si128((const __m128i *)round_ptr); in load_b_values() 23 *quant = _mm_load_si128((const __m128i *)quant_ptr); in load_b_values() 25 *dequant = _mm_load_si128((const __m128i *)dequant_ptr); in load_b_values() 26 *shift = _mm_load_si128((const __m128i *)shift_ptr); in load_b_values() 105 __m128i scan0 = _mm_load_si128((const __m128i *)(scan_ptr + index)); in scan_for_eob() 106 __m128i scan1 = _mm_load_si128((const __m128i *)(scan_ptr + index + 8)); in scan_for_eob() 129 const __m128i coeff1 = _mm_load_si128((__m128i *)(coeff_ptr)); in load_coefficients() 130 const __m128i coeff2 = _mm_load_si128((__m128i *)(coeff_ptr + 4)); in load_coefficients()
|
D | avg_intrin_sse2.c | 222 src[0] = _mm_load_si128((const __m128i *)src_diff); in hadamard_8x8_sse2() 223 src[1] = _mm_load_si128((const __m128i *)(src_diff += src_stride)); in hadamard_8x8_sse2() 224 src[2] = _mm_load_si128((const __m128i *)(src_diff += src_stride)); in hadamard_8x8_sse2() 225 src[3] = _mm_load_si128((const __m128i *)(src_diff += src_stride)); in hadamard_8x8_sse2() 226 src[4] = _mm_load_si128((const __m128i *)(src_diff += src_stride)); in hadamard_8x8_sse2() 227 src[5] = _mm_load_si128((const __m128i *)(src_diff += src_stride)); in hadamard_8x8_sse2() 228 src[6] = _mm_load_si128((const __m128i *)(src_diff += src_stride)); in hadamard_8x8_sse2() 229 src[7] = _mm_load_si128((const __m128i *)(src_diff += src_stride)); in hadamard_8x8_sse2() 294 __m128i coeff0 = _mm_load_si128((const __m128i *)t_coeff); in hadamard_16x16_sse2() 295 __m128i coeff1 = _mm_load_si128((const __m128i *)(t_coeff + 64)); in hadamard_16x16_sse2() [all …]
|
/external/swiftshader/src/Device/ |
D | Surface.cpp | 3375 __m128i c0 = _mm_load_si128((__m128i*)(source0 + 4 * x)); in resolve() 3376 __m128i c1 = _mm_load_si128((__m128i*)(source1 + 4 * x)); in resolve() 3393 __m128i c0 = _mm_load_si128((__m128i*)(source0 + 4 * x)); in resolve() 3394 __m128i c1 = _mm_load_si128((__m128i*)(source1 + 4 * x)); in resolve() 3395 __m128i c2 = _mm_load_si128((__m128i*)(source2 + 4 * x)); in resolve() 3396 __m128i c3 = _mm_load_si128((__m128i*)(source3 + 4 * x)); in resolve() 3417 __m128i c0 = _mm_load_si128((__m128i*)(source0 + 4 * x)); in resolve() 3418 __m128i c1 = _mm_load_si128((__m128i*)(source1 + 4 * x)); in resolve() 3419 __m128i c2 = _mm_load_si128((__m128i*)(source2 + 4 * x)); in resolve() 3420 __m128i c3 = _mm_load_si128((__m128i*)(source3 + 4 * x)); in resolve() [all …]
|
/external/jemalloc_new/test/include/test/ |
D | SFMT-sse2.h | 67 x = _mm_load_si128(a); in mm_recursion() 89 r1 = _mm_load_si128(&ctx->sfmt[N - 2].si); in gen_rand_all() 90 r2 = _mm_load_si128(&ctx->sfmt[N - 1].si); in gen_rand_all() 119 r1 = _mm_load_si128(&ctx->sfmt[N - 2].si); in gen_rand_array() 120 r2 = _mm_load_si128(&ctx->sfmt[N - 1].si); in gen_rand_array() 144 r = _mm_load_si128(&array[j + size - N].si); in gen_rand_array()
|
/external/jemalloc/test/include/test/ |
D | SFMT-sse2.h | 67 x = _mm_load_si128(a); in mm_recursion() 89 r1 = _mm_load_si128(&ctx->sfmt[N - 2].si); in gen_rand_all() 90 r2 = _mm_load_si128(&ctx->sfmt[N - 1].si); in gen_rand_all() 119 r1 = _mm_load_si128(&ctx->sfmt[N - 2].si); in gen_rand_array() 120 r2 = _mm_load_si128(&ctx->sfmt[N - 1].si); in gen_rand_array() 144 r = _mm_load_si128(&array[j + size - N].si); in gen_rand_array()
|