/external/libaom/libaom/aom_dsp/x86/ |
D | intrapred_sse2.c | 319 sum_above = _mm_shufflelo_epi16(sum_above, 0); in aom_dc_top_predictor_4x8_sse2() 333 sum_above = _mm_shufflelo_epi16(sum_above, 0); in aom_dc_top_predictor_4x16_sse2() 348 const __m128i row = _mm_shufflelo_epi16(sum_above, 0); in aom_dc_top_predictor_8x4_sse2() 360 const __m128i row = _mm_shufflelo_epi16(sum_above, 0); in aom_dc_top_predictor_8x16_sse2() 372 const __m128i row = _mm_shufflelo_epi16(sum_above, 0); in aom_dc_top_predictor_8x32_sse2() 384 sum_above = _mm_shufflelo_epi16(sum_above, 0); in aom_dc_top_predictor_16x4_sse2() 397 sum_above = _mm_shufflelo_epi16(sum_above, 0); in aom_dc_top_predictor_16x8_sse2() 411 sum_above = _mm_shufflelo_epi16(sum_above, 0); in aom_dc_top_predictor_16x32_sse2() 425 sum_above = _mm_shufflelo_epi16(sum_above, 0); in aom_dc_top_predictor_16x64_sse2() 438 sum_above = _mm_shufflelo_epi16(sum_above, 0); in aom_dc_top_predictor_32x8_sse2() [all …]
|
D | highbd_intrapred_sse2.c | 23 const __m128i row0 = _mm_shufflelo_epi16(left_u16, 0x0); in aom_highbd_h_predictor_4x4_sse2() 24 const __m128i row1 = _mm_shufflelo_epi16(left_u16, 0x55); in aom_highbd_h_predictor_4x4_sse2() 25 const __m128i row2 = _mm_shufflelo_epi16(left_u16, 0xaa); in aom_highbd_h_predictor_4x4_sse2() 26 const __m128i row3 = _mm_shufflelo_epi16(left_u16, 0xff); in aom_highbd_h_predictor_4x4_sse2() 51 const __m128i row0 = _mm_shufflelo_epi16(left_u16, 0x0); in aom_highbd_h_predictor_8x4_sse2() 52 const __m128i row1 = _mm_shufflelo_epi16(left_u16, 0x55); in aom_highbd_h_predictor_8x4_sse2() 53 const __m128i row2 = _mm_shufflelo_epi16(left_u16, 0xaa); in aom_highbd_h_predictor_8x4_sse2() 54 const __m128i row3 = _mm_shufflelo_epi16(left_u16, 0xff); in aom_highbd_h_predictor_8x4_sse2() 70 const __m128i row0 = _mm_shufflelo_epi16(left_u16, 0x0); in aom_highbd_h_predictor_8x8_sse2() 71 const __m128i row1 = _mm_shufflelo_epi16(left_u16, 0x55); in aom_highbd_h_predictor_8x8_sse2() [all …]
|
D | txfm_common_sse2.h | 24 const __m128i a = _mm_shufflelo_epi16(x, 0x1b); in mm_reverse_epi16()
|
D | quantize_x86.h | 120 eob_shuffled = _mm_shufflelo_epi16(eob, 0xe); in accumulate_eob() 122 eob_shuffled = _mm_shufflelo_epi16(eob, 0x1); in accumulate_eob()
|
/external/XNNPACK/src/x8-zip/ |
D | xm-sse2.c | 56 vxyzw0 = _mm_shufflelo_epi16(vxyzw0, _MM_SHUFFLE(3, 2, 3, 2)); in xnn_x8_zip_xm_ukernel__sse2() 62 vxyzw0 = _mm_shufflelo_epi16(vxyzw0, _MM_SHUFFLE(3, 2, 3, 2)); in xnn_x8_zip_xm_ukernel__sse2() 68 vxyzw1 = _mm_shufflelo_epi16(vxyzw1, _MM_SHUFFLE(3, 2, 3, 2)); in xnn_x8_zip_xm_ukernel__sse2() 74 vxyzw1 = _mm_shufflelo_epi16(vxyzw1, _MM_SHUFFLE(3, 2, 3, 2)); in xnn_x8_zip_xm_ukernel__sse2() 80 vxyzw2 = _mm_shufflelo_epi16(vxyzw2, _MM_SHUFFLE(3, 2, 3, 2)); in xnn_x8_zip_xm_ukernel__sse2() 86 vxyzw2 = _mm_shufflelo_epi16(vxyzw2, _MM_SHUFFLE(3, 2, 3, 2)); in xnn_x8_zip_xm_ukernel__sse2() 92 vxyzw3 = _mm_shufflelo_epi16(vxyzw3, _MM_SHUFFLE(3, 2, 3, 2)); in xnn_x8_zip_xm_ukernel__sse2() 98 vxyzw3 = _mm_shufflelo_epi16(vxyzw3, _MM_SHUFFLE(3, 2, 3, 2)); in xnn_x8_zip_xm_ukernel__sse2() 119 vxyzw0 = _mm_shufflelo_epi16(vxyzw0, _MM_SHUFFLE(3, 2, 3, 2)); in xnn_x8_zip_xm_ukernel__sse2() 125 vxyzw0 = _mm_shufflelo_epi16(vxyzw0, _MM_SHUFFLE(3, 2, 3, 2)); in xnn_x8_zip_xm_ukernel__sse2() [all …]
|
/external/libvpx/libvpx/vpx_dsp/x86/ |
D | highbd_intrapred_intrin_sse2.c | 23 const __m128i row0 = _mm_shufflelo_epi16(left_u16, 0x0); in vpx_highbd_h_predictor_4x4_sse2() 24 const __m128i row1 = _mm_shufflelo_epi16(left_u16, 0x55); in vpx_highbd_h_predictor_4x4_sse2() 25 const __m128i row2 = _mm_shufflelo_epi16(left_u16, 0xaa); in vpx_highbd_h_predictor_4x4_sse2() 26 const __m128i row3 = _mm_shufflelo_epi16(left_u16, 0xff); in vpx_highbd_h_predictor_4x4_sse2() 42 const __m128i row0 = _mm_shufflelo_epi16(left_u16, 0x0); in vpx_highbd_h_predictor_8x8_sse2() 43 const __m128i row1 = _mm_shufflelo_epi16(left_u16, 0x55); in vpx_highbd_h_predictor_8x8_sse2() 44 const __m128i row2 = _mm_shufflelo_epi16(left_u16, 0xaa); in vpx_highbd_h_predictor_8x8_sse2() 45 const __m128i row3 = _mm_shufflelo_epi16(left_u16, 0xff); in vpx_highbd_h_predictor_8x8_sse2() 94 const __m128i row0 = _mm_shufflelo_epi16(left_u16, 0x0); in vpx_highbd_h_predictor_16x16_sse2() 95 const __m128i row1 = _mm_shufflelo_epi16(left_u16, 0x55); in vpx_highbd_h_predictor_16x16_sse2() [all …]
|
D | quantize_sse2.h | 89 eob_shuffled = _mm_shufflelo_epi16(eob, 0xe); in accumulate_eob() 91 eob_shuffled = _mm_shufflelo_epi16(eob, 0x1); in accumulate_eob()
|
/external/libopus/celt/x86/ |
D | pitch_sse4_1.c | 99 acc1 = _mm_add_epi32(acc1, _mm_shufflelo_epi16(acc1, 0x0E)); in celt_inner_prod_sse4_1() 142 sum0 = _mm_add_epi32(sum0, _mm_shufflelo_epi16( sum0, 0x0E)); in xcorr_kernel_sse4_1() 145 sum1 = _mm_add_epi32(sum1, _mm_shufflelo_epi16( sum1, 0x0E)); in xcorr_kernel_sse4_1() 148 sum2 = _mm_add_epi32(sum2, _mm_shufflelo_epi16( sum2, 0x0E)); in xcorr_kernel_sse4_1() 151 sum3 = _mm_add_epi32(sum3, _mm_shufflelo_epi16( sum3, 0x0E)); in xcorr_kernel_sse4_1()
|
D | pitch_sse2.c | 86 acc1 = _mm_add_epi32(acc1, _mm_shufflelo_epi16( acc1, 0x0E)); in celt_inner_prod_sse2()
|
/external/libgav1/libgav1/src/dsp/x86/ |
D | loop_filter_sse4.cc | 151 _mm_shufflelo_epi16(_mm_cvtsi32_si128(hev_thresh), 0); in Horizontal4() 1275 _mm_shufflelo_epi16(_mm_cvtsi32_si128(outer_thresh << kThreshShift), 0); in Horizontal4() 1277 _mm_shufflelo_epi16(_mm_cvtsi32_si128(inner_thresh << kThreshShift), 0); in Horizontal4() 1279 _mm_shufflelo_epi16(_mm_cvtsi32_si128(hev_thresh << kThreshShift), 0); in Horizontal4() 1308 _mm_shufflelo_epi16(_mm_cvtsi32_si128(outer_thresh << kThreshShift), 0); in Vertical4() 1310 _mm_shufflelo_epi16(_mm_cvtsi32_si128(inner_thresh << kThreshShift), 0); in Vertical4() 1312 _mm_shufflelo_epi16(_mm_cvtsi32_si128(hev_thresh << kThreshShift), 0); in Vertical4() 1428 _mm_shufflelo_epi16(_mm_cvtsi32_si128(1 << kThreshShift), 0); in Horizontal6() 1430 _mm_shufflelo_epi16(_mm_cvtsi32_si128(outer_thresh << kThreshShift), 0); in Horizontal6() 1432 _mm_shufflelo_epi16(_mm_cvtsi32_si128(inner_thresh << kThreshShift), 0); in Horizontal6() [all …]
|
D | convolve_sse4.inc | 111 v_tap[0] = _mm_shufflelo_epi16(*filter, 0x0); // k1k0 112 v_tap[1] = _mm_shufflelo_epi16(*filter, 0x55); // k3k2 113 v_tap[2] = _mm_shufflelo_epi16(*filter, 0xaa); // k5k4 114 v_tap[3] = _mm_shufflelo_epi16(*filter, 0xff); // k7k6 128 v_tap[0] = _mm_shufflelo_epi16(adjusted_filter, 0x0); // k2k1 129 v_tap[1] = _mm_shufflelo_epi16(adjusted_filter, 0x55); // k4k3 130 v_tap[2] = _mm_shufflelo_epi16(adjusted_filter, 0xaa); // k6k5 141 v_tap[0] = _mm_shufflelo_epi16(*filter, 0x55); // k3k2 142 v_tap[1] = _mm_shufflelo_epi16(*filter, 0xaa); // k5k4 152 v_tap[0] = _mm_shufflelo_epi16(adjusted_filter, 0x55); // k4k3
|
D | intrapred_cfl_sse4.cc | 180 averages = _mm_shufflelo_epi16(averages, 0); in CflSubsampler444_4xH_SSE4_1() 502 averages = _mm_shufflelo_epi16(averages, 0); in CflSubsampler420_4xH_SSE4_1() 610 averages = _mm_shufflelo_epi16(averages, 0); in CflSubsampler420Impl_8xH_SSE4_1() 732 averages = _mm_shufflelo_epi16(averages, 0); in CflSubsampler420Impl_WxH_SSE4_1() 1056 averages = _mm_shufflelo_epi16(averages, 0); in CflSubsampler444_4xH_SSE4_1() 1361 averages = _mm_shufflelo_epi16(averages, 0); in CflSubsampler420_4xH_SSE4_1() 1471 averages = _mm_shufflelo_epi16(averages, 0); in CflSubsampler420Impl_8xH_SSE4_1() 1593 averages = _mm_shufflelo_epi16(averages, 0); in CflSubsampler420Impl_WxH_SSE4_1()
|
/external/libvpx/libvpx/vp8/encoder/x86/ |
D | vp8_quantize_sse2.c | 58 zbin_extra = _mm_shufflelo_epi16(zbin_extra, 0); in vp8_regular_quantize_b_sse2() 216 x1 = _mm_shufflelo_epi16(x0, 0xE); // 0b00001110 in vp8_fast_quantize_b_sse2() 221 x1 = _mm_shufflelo_epi16(x0, 0x1); // 0b00000001 in vp8_fast_quantize_b_sse2()
|
D | quantize_sse4.c | 54 zbin_extra = _mm_shufflelo_epi16(zbin_extra, 0); in vp8_regular_quantize_b_sse4_1()
|
/external/libvpx/libvpx/vp9/encoder/x86/ |
D | vp9_quantize_sse2.c | 196 eob_shuffled = _mm_shufflelo_epi16(eob, 0xe); in vp9_quantize_fp_sse2() 198 eob_shuffled = _mm_shufflelo_epi16(eob, 0x1); in vp9_quantize_fp_sse2()
|
/external/libaom/libaom/av1/encoder/x86/ |
D | av1_highbd_quantize_sse4.c | 107 eob_shuffled = _mm_shufflelo_epi16(*eob, 0xe); in get_accumulated_eob() 109 eob_shuffled = _mm_shufflelo_epi16(*eob, 0x1); in get_accumulated_eob()
|
D | av1_quantize_sse2.c | 183 eob_shuffled = _mm_shufflelo_epi16(eob, 0xe); in av1_quantize_fp_sse2() 185 eob_shuffled = _mm_shufflelo_epi16(eob, 0x1); in av1_quantize_fp_sse2()
|
/external/webp/src/dsp/ |
D | alpha_processing_sse2.c | 162 const __m128i alpha1_lo = _mm_shufflelo_epi16(alpha0_lo, SHUFFLE); \ 163 const __m128i alpha1_hi = _mm_shufflelo_epi16(alpha0_hi, SHUFFLE); \ 306 const __m128i A3 = _mm_shufflelo_epi16(A2, _MM_SHUFFLE(2, 3, 3, 3)); in MultARGBRow_SSE2()
|
D | lossless_sse2.c | 438 const __m128i B = _mm_shufflelo_epi16(A, _MM_SHUFFLE(2, 2, 0, 0)); in GENERATE_PREDICTOR_ADD() 468 const __m128i B = _mm_shufflelo_epi16(A, _MM_SHUFFLE(2, 2, 0, 0)); in TransformColorInverse_SSE2() 537 const __m128i D1 = _mm_shufflelo_epi16(B1, _MM_SHUFFLE(2, 3, 0, 1)); in ConvertBGRAToRGBA_SSE2() 538 const __m128i D2 = _mm_shufflelo_epi16(B2, _MM_SHUFFLE(2, 3, 0, 1)); in ConvertBGRAToRGBA_SSE2()
|
/external/libavc/common/x86/ |
D | ih264_chroma_intra_pred_filters_ssse3.c | 277 h_val2_16x8b = _mm_shufflelo_epi16(h_val2_16x8b, 0x1b); in ih264_intra_pred_chroma_8x8_mode_plane_ssse3() 278 v_val1_16x8b = _mm_shufflelo_epi16(v_val1_16x8b, 0x1b); in ih264_intra_pred_chroma_8x8_mode_plane_ssse3()
|
/external/flac/src/libFLAC/ |
D | stream_encoder_intrin_ssse3.c | 90 mm_sum = _mm_add_epi32(mm_sum, _mm_shufflelo_epi16(mm_sum, _MM_SHUFFLE(1,0,3,2))); in FLAC__precompute_partition_info_sums_intrin_ssse3()
|
D | stream_encoder_intrin_avx2.c | 87 sum128 = _mm_add_epi32(sum128, _mm_shufflelo_epi16(sum128, _MM_SHUFFLE(1,0,3,2))); in FLAC__precompute_partition_info_sums_intrin_avx2()
|
D | stream_encoder_intrin_sse2.c | 101 mm_sum = _mm_add_epi32(mm_sum, _mm_shufflelo_epi16(mm_sum, _MM_SHUFFLE(1,0,3,2))); in FLAC__precompute_partition_info_sums_intrin_sse2()
|
/external/libavc/encoder/x86/ |
D | ime_distortion_metrics_sse42.c | 1790 src_r0 = _mm_shufflelo_epi16(src_r0, 0x9c); //Block 0 s1 s1 s4 s4 a1 a4 a4 a1 in ime_compute_satqd_16x16_lumainter_sse42() 1793 src_r1 = _mm_shufflelo_epi16(src_r1, 0x9c); //Block 0 s2 s2 s3 s3 a2 a3 a3 a2 in ime_compute_satqd_16x16_lumainter_sse42() 1802 temp0 = _mm_shufflelo_epi16(temp0, 0xb1);//s4<<1 s1<<1 a4<<1 a1<<1 0 0 0 0 in ime_compute_satqd_16x16_lumainter_sse42() 1803 temp1 = _mm_shufflelo_epi16(temp1, 0xb1);//s3<<1 s2<<1 a3<<1 a2<<1 0 0 0 0 in ime_compute_satqd_16x16_lumainter_sse42() 1889 src_r0 = _mm_shufflelo_epi16(src_r0, 0x9c); //Block 0 s1 s1 s4 s4 a1 a4 a4 a1 in ime_compute_satqd_16x16_lumainter_sse42() 1892 src_r1 = _mm_shufflelo_epi16(src_r1, 0x9c); //Block 0 s2 s2 s3 s3 a2 a3 a3 a2 in ime_compute_satqd_16x16_lumainter_sse42() 1901 temp0 = _mm_shufflelo_epi16(temp0, 0xb1);//s4<<1 s1<<1 a4<<1 a1<<1 0 0 0 0 in ime_compute_satqd_16x16_lumainter_sse42() 1902 temp1 = _mm_shufflelo_epi16(temp1, 0xb1);//s3<<1 s2<<1 a3<<1 a2<<1 0 0 0 0 in ime_compute_satqd_16x16_lumainter_sse42()
|
/external/libopus/silk/fixed/x86/ |
D | burg_modified_FIX_sse4_1.c | 196 TMP1_3210 = _mm_add_epi32( TMP1_3210, _mm_shufflelo_epi16(TMP1_3210, 0x0E ) ); in silk_burg_modified_sse4_1() 197 TMP2_3210 = _mm_add_epi32( TMP2_3210, _mm_shufflelo_epi16(TMP2_3210, 0x0E ) ); in silk_burg_modified_sse4_1()
|