| /external/opencv3/modules/core/include/opencv2/core/ |
| sse_utils.hpp | 171 __m128i layer4_chunk0 = _mm_packus_epi16(_mm_and_si128(v_r0, v_mask), _mm_and_si128(v_r1, v_mask));
173 __m128i layer4_chunk1 = _mm_packus_epi16(_mm_and_si128(v_g0, v_mask), _mm_and_si128(v_g1, v_mask));
176 __m128i layer3_chunk0 = _mm_packus_epi16(_mm_and_si128(layer4_chunk0, v_mask), _mm_and_si128(layer4_chunk1, v_mask));
178 __m128i layer3_chunk1 = _mm_packus_epi16(_mm_and_si128(layer4_chunk2, v_mask), _mm_and_si128(layer4_chunk3, v_mask));
181 __m128i layer2_chunk0 = _mm_packus_epi16(_mm_and_si128(layer3_chunk0, v_mask), _mm_and_si128(layer3_chunk1, v_mask))
[all...] |
| /external/libavc/common/x86/ |
| ih264_iquant_itrans_recon_dc_ssse3.c | 153 temp4 = _mm_and_si128(temp4, sign_reg);
155 temp5 = _mm_and_si128(temp5, sign_reg);
306 temp1 = _mm_and_si128(temp1, sign_reg);
308 temp2 = _mm_and_si128(temp2, sign_reg);
310 temp3 = _mm_and_si128(temp3, sign_reg);
312 temp4 = _mm_and_si128(temp4, sign_reg);
314 temp5 = _mm_and_si128(temp5, sign_reg);
316 temp6 = _mm_and_si128(temp6, sign_reg);
318 temp7 = _mm_and_si128(temp7, sign_reg);
320 temp8 = _mm_and_si128(temp8, sign_reg)
[all...] |
| ih264_deblk_luma_ssse3.c | 196 flag1_16x8 = _mm_and_si128(flag1_16x8, flag2_16x8);
212 flag1_16x8 = _mm_and_si128(flag1_16x8, flag2_16x8);
227 flag2_16x8 = _mm_and_si128(flag1_16x8, flag2_16x8);
240 flag3_16x8 = _mm_and_si128(flag3_16x8, flag2_16x8);
253 flag4_16x8 = _mm_and_si128(flag4_16x8, flag2_16x8);
384 p0_16x8 = _mm_and_si128(p0_16x8,
386 p0_16x8_1 = _mm_and_si128(p0_16x8_1, flag1_16x8);
388 q0_16x8 = _mm_and_si128(q0_16x8,
390 q0_16x8_1 = _mm_and_si128(q0_16x8_1, flag1_16x8);
394 p0_16x8 = _mm_and_si128(p0_16x8
[all...] |
| ih264_deblk_chroma_ssse3.c | 153 flag1 = _mm_and_si128(flag1, _mm_cmpgt_epi16(beta_cbcr_16x8, diff));
157 flag1 = _mm_and_si128(flag1, _mm_cmpgt_epi16(beta_cbcr_16x8, diff));
184 flag2 = _mm_and_si128(flag2, _mm_cmpgt_epi16(beta_cbcr_16x8, diff));
188 flag2 = _mm_and_si128(flag2, _mm_cmpgt_epi16(beta_cbcr_16x8, diff));
207 p0_uv_8x16_1 = _mm_and_si128(p0_uv_16x8,
209 p0_uv_8x16_2 = _mm_and_si128(p0_uv_8x16_2, flag1);
212 q0_uv_8x16_1 = _mm_and_si128(q0_uv_16x8,
214 q0_uv_8x16_2 = _mm_and_si128(q0_uv_8x16_2, flag1);
322 flag1 = _mm_and_si128(flag1, _mm_cmpgt_epi16(beta_cbcr_16x8, diff));
326 flag1 = _mm_and_si128(flag1, _mm_cmpgt_epi16(beta_cbcr_16x8, diff))
[all...] |
| ih264_ihadamard_scaling_ssse3.c | 105 mult_val = _mm_and_si128(mult_val, mask);
175 src_r0 = _mm_and_si128(src_r0, mask);
176 src_r1 = _mm_and_si128(src_r1, mask);
177 src_r2 = _mm_and_si128(src_r2, mask);
178 src_r3 = _mm_and_si128(src_r3, mask);
|
| ih264_iquant_itrans_recon_sse42.c | 283 temp0 = _mm_and_si128(temp0, sign_reg);
285 temp1 = _mm_and_si128(temp1, sign_reg);
485 pred_r0 = _mm_and_si128(pred_r0, chroma_mask);
486 pred_r1 = _mm_and_si128(pred_r1, chroma_mask);
487 pred_r2 = _mm_and_si128(pred_r2, chroma_mask);
488 pred_r3 = _mm_and_si128(pred_r3, chroma_mask);
538 temp0 = _mm_and_si128(temp0, sign_reg);
540 temp1 = _mm_and_si128(temp1, sign_reg);
558 out_r0 = _mm_and_si128(out_r0, chroma_mask);
559 out_r1 = _mm_and_si128(out_r1, chroma_mask)
[all...] |
| ih264_resi_trans_quant_sse42.c | 301 cmp0 = _mm_and_si128(temp_1, cmp0);
314 cmp1 = _mm_and_si128(temp_1, cmp1);
409 src_r0 = _mm_and_si128(src_r0, chroma_mask);
410 src_r1 = _mm_and_si128(src_r1, chroma_mask);
411 src_r2 = _mm_and_si128(src_r2, chroma_mask);
412 src_r3 = _mm_and_si128(src_r3, chroma_mask);
423 pred_r0 = _mm_and_si128(pred_r0, chroma_mask);
424 pred_r1 = _mm_and_si128(pred_r1, chroma_mask);
425 pred_r2 = _mm_and_si128(pred_r2, chroma_mask);
426 pred_r3 = _mm_and_si128(pred_r3, chroma_mask)
[all...] |
| /external/skia/src/opts/ |
| SkColor_opts_SSE2.h | 51 __m128i rb = _mm_and_si128(mask, c);
57 ASSERT_EQ(ag, _mm_and_si128(mask, ag)); // ag = _mm_srli_epi16(c, 8) did this for us.
61 ASSERT_EQ(rb, _mm_and_si128(mask, rb)); // rb = _mm_srli_epi16(rb, 8) did this for us.
73 __m128i rb = _mm_and_si128(mask, c);
141 r = _mm_and_si128(r, _mm_set1_epi32(SK_R16_MASK));
150 g = _mm_and_si128(g, _mm_set1_epi32(SK_G16_MASK));
159 b = _mm_and_si128(b, _mm_set1_epi32(SK_B16_MASK));
179 r1 = _mm_and_si128(r1, _mm_set1_epi32(SK_R16_MASK));
182 r2 = _mm_and_si128(r2, _mm_set1_epi32(SK_R16_MASK));
188 g1 = _mm_and_si128(g1, _mm_set1_epi32(SK_G16_MASK))
[all...] |
| SkBlitRow_opts_SSE4.cpp | 49 const __m128i ANDed = _mm_and_si128(s3, _mm_and_si128(s2, _mm_and_si128(s1, s0)));
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp9/common/x86/ |
| vp9_loopfilter_intrin_avx2.c | 63 abs_p1q1 = _mm_srli_epi16(_mm_and_si128(abs_p1q1, fe), 1);
98 filt = _mm_and_si128(_mm_subs_epi8(qs1ps1, qs1), hev);
104 filt = _mm_and_si128(filt, mask);
138 flat = _mm_and_si128(flat, mask);
171 flat2 = _mm_and_si128(flat2, flat); // flat2 & flat & mask
340 flat_q2p2 = _mm_and_si128(flat, flat_q2p2);
344 flat_q1p1 = _mm_and_si128(flat, flat_q1p1);
348 flat_q0p0 = _mm_and_si128(flat, flat_q0p0);
352 flat2_q6p6 = _mm_and_si128(flat2, flat2_q6p6);
358 flat2_q5p5 = _mm_and_si128(flat2, flat2_q5p5)
[all...] |
| vp9_loopfilter_intrin_sse2.c | 63 abs_p1q1 = _mm_srli_epi16(_mm_and_si128(abs_p1q1, fe), 1);
97 filt = _mm_and_si128(_mm_subs_epi8(qs1ps1, qs1), hev);
103 filt = _mm_and_si128(filt, mask);
136 flat = _mm_and_si128(flat, mask);
164 flat2 = _mm_and_si128(flat2, flat); // flat2 & flat & mask
312 flat_q2p2 = _mm_and_si128(flat, flat_q2p2);
316 flat_q1p1 = _mm_and_si128(flat, flat_q1p1);
320 flat_q0p0 = _mm_and_si128(flat, flat_q0p0);
324 flat2_q6p6 = _mm_and_si128(flat2, flat2_q6p6);
330 flat2_q5p5 = _mm_and_si128(flat2, flat2_q5p5)
[all...] |
| /external/libvpx/libvpx/vpx_dsp/x86/ |
| highbd_loopfilter_sse2.c | 45 ubounded = _mm_and_si128(ubounded, max);
46 lbounded = _mm_and_si128(lbounded, min);
127 mask = _mm_and_si128(mask, _mm_adds_epu16(limit, one));
165 filt = _mm_and_si128(
171 filt = _mm_and_si128(filt, mask);
223 flat = _mm_and_si128(flat, mask);
265 flat2 = _mm_and_si128(flat2, flat); // flat2 & flat & mask
377 flat_p2 = _mm_and_si128(flat, flat_p2);
381 flat_q2 = _mm_and_si128(flat, flat_q2);
386 flat_p1 = _mm_and_si128(flat, flat_p1)
[all...] |
| loopfilter_avx2.c | 66 abs_p1q1 = _mm_srli_epi16(_mm_and_si128(abs_p1q1, fe), 1);
101 filt = _mm_and_si128(_mm_subs_epi8(qs1ps1, qs1), hev);
107 filt = _mm_and_si128(filt, mask);
141 flat = _mm_and_si128(flat, mask);
174 flat2 = _mm_and_si128(flat2, flat); // flat2 & flat & mask
343 flat_q2p2 = _mm_and_si128(flat, flat_q2p2);
347 flat_q1p1 = _mm_and_si128(flat, flat_q1p1);
351 flat_q0p0 = _mm_and_si128(flat, flat_q0p0);
355 flat2_q6p6 = _mm_and_si128(flat2, flat2_q6p6);
361 flat2_q5p5 = _mm_and_si128(flat2, flat2_q5p5)
[all...] |
| loopfilter_sse2.c | 66 abs_p1q1 = _mm_srli_epi16(_mm_and_si128(abs_p1q1, fe), 1);
98 filt = _mm_and_si128(_mm_subs_epi8(qs1ps1, qs1), hev);
104 filt = _mm_and_si128(filt, mask);
134 flat = _mm_and_si128(flat, mask);
153 flat2 = _mm_and_si128(flat2, flat); // flat2 & flat & mask
301 flat_q2p2 = _mm_and_si128(flat, flat_q2p2);
305 flat_q1p1 = _mm_and_si128(flat, flat_q1p1);
309 flat_q0p0 = _mm_and_si128(flat, flat_q0p0);
313 flat2_q6p6 = _mm_and_si128(flat2, flat2_q6p6);
319 flat2_q5p5 = _mm_and_si128(flat2, flat2_q5p5)
[all...] |
| quantize_sse2.c | 117 qcoeff0 = _mm_and_si128(qcoeff0, cmp_mask0);
118 qcoeff1 = _mm_and_si128(qcoeff1, cmp_mask1);
146 eob = _mm_and_si128(iscan0, nzero_coeff0);
147 eob1 = _mm_and_si128(iscan1, nzero_coeff1);
191 qcoeff0 = _mm_and_si128(qcoeff0, cmp_mask0);
192 qcoeff1 = _mm_and_si128(qcoeff1, cmp_mask1);
219 eob0 = _mm_and_si128(iscan0, nzero_coeff0);
220 eob1 = _mm_and_si128(iscan1, nzero_coeff1);
|
| /hardware/intel/common/omx-components/videocodec/libvpx_internal/libvpx/vp8/encoder/x86/ |
| denoising_sse2.c | 71 __m128i adj2 = _mm_and_si128(mask2, l32);
72 const __m128i adj1 = _mm_and_si128(mask1, l21);
73 const __m128i adj0 = _mm_and_si128(mask0, clamped_absdiff);
84 nadj = _mm_and_si128(diff_sign, adj);
|
| /external/libvpx/libvpx/vp8/encoder/x86/ |
| denoising_sse2.c | 85 __m128i adj2 = _mm_and_si128(mask2, l32);
86 const __m128i adj1 = _mm_and_si128(mask1, l21);
87 const __m128i adj0 = _mm_and_si128(mask0, clamped_absdiff);
98 nadj = _mm_and_si128(diff_sign, adj);
157 nadj = _mm_and_si128(diff_sign, adj);
262 __m128i adj2 = _mm_and_si128(mask2, l32);
263 const __m128i adj1 = _mm_and_si128(mask1, l21);
264 const __m128i adj0 = _mm_and_si128(mask0, clamped_absdiff);
276 nadj = _mm_and_si128(diff_sign, adj);
348 nadj = _mm_and_si128(diff_sign, adj)
[all...] |
| /external/boringssl/src/crypto/poly1305/ |
| poly1305_vec.c | 272 st->H[0] = _mm_and_si128(MMASK, T5);
273 st->H[1] = _mm_and_si128(MMASK, _mm_srli_epi64(T5, 26));
275 st->H[2] = _mm_and_si128(MMASK, T5);
276 st->H[3] = _mm_and_si128(MMASK, _mm_srli_epi64(T5, 26));
352 M0 = _mm_and_si128(MMASK, T5);
353 M1 = _mm_and_si128(MMASK, _mm_srli_epi64(T5, 26));
355 M2 = _mm_and_si128(MMASK, T5);
356 M3 = _mm_and_si128(MMASK, _mm_srli_epi64(T5, 26));
416 M0 = _mm_and_si128(MMASK, T5);
417 M1 = _mm_and_si128(MMASK, _mm_srli_epi64(T5, 26))
[all...] |
| /external/skia/src/core/ |
| SkHalf.h | 79 rebias = _mm_add_epi32(rebias, _mm_and_si128(is_denorm, _mm_set1_epi32(1<<23)));
83 _mm_castsi128_ps(_mm_and_si128(is_denorm, rebias)));
|
| /external/libvpx/libvpx/vp9/encoder/x86/ |
| vp9_quantize_sse2.c | 107 eob = _mm_and_si128(iscan0, nzero_coeff0);
108 eob1 = _mm_and_si128(iscan1, nzero_coeff1);
182 eob0 = _mm_and_si128(iscan0, nzero_coeff0);
183 eob1 = _mm_and_si128(iscan1, nzero_coeff1);
|
| /external/webp/src/dsp/ |
| argb_sse2.c | 36 const __m128i B = _mm_and_si128(A, red_blue_mask); // R 0 B 0
|
| alpha_processing_sse2.c | 50 const __m128i b1_lo = _mm_and_si128(b0_lo, rgb_mask);
51 const __m128i b1_hi = _mm_and_si128(b0_hi, rgb_mask);
59 all_alphas = _mm_and_si128(all_alphas, a0);
123 const __m128i b0 = _mm_and_si128(a0, a_mask);
124 const __m128i b1 = _mm_and_si128(a1, a_mask);
130 all_alphas = _mm_and_si128(all_alphas, d0);
157 const __m128i alpha0 = _mm_and_si128(argb1, MASK); \
|
| lossless_enc_sse2.c | 59 const __m128i A = _mm_and_si128(in, mask_ag); // a 0 g 0
67 const __m128i I = _mm_and_si128(H, mask_rb); // 0 dr 0 db
99 const __m128i B0 = _mm_and_si128(in0, mask_g); // 0 0 | g 0
100 const __m128i B1 = _mm_and_si128(in1, mask_g);
111 const __m128i H0 = _mm_and_si128(G0, mask_b); // 0 0 | 0 b
112 const __m128i H1 = _mm_and_si128(G1, mask_b);
145 const __m128i A0 = _mm_and_si128(in0, mask_g); // 0 0 | g 0
146 const __m128i A1 = _mm_and_si128(in1, mask_g);
153 const __m128i F0 = _mm_and_si128(E0, mask); // 0 0 | 0 r'
154 const __m128i F1 = _mm_and_si128(E1, mask)
[all...] |
| lossless_sse2.c | 192 const __m128i A = _mm_and_si128(in, mask_ag); // a 0 g 0
257 const __m128i rb1 = _mm_and_si128(rb0, mask_0xf0); // -r0|-r1|...|-b6|-a7
258 const __m128i ga2 = _mm_and_si128(ga1, mask_0x0f); // g0-|g1-|...|a6-|a7-
291 const __m128i rb1 = _mm_and_si128(rb0, mask_0xf8); // -r0..-r7|-b0..-b7
293 const __m128i g_lo2 = _mm_and_si128(g_lo1, mask_0x07); // g0-...g7-|xx (3b)
295 const __m128i g_hi2 = _mm_and_si128(g_hi1, mask_0xe0); // -g0...-g7|xx (3b)
322 const __m128i a0l = _mm_and_si128(bgra0, mask_l); // bgr0|0|bgr0|0
323 const __m128i a4l = _mm_and_si128(bgra4, mask_l); // bgr0|0|bgr0|0
324 const __m128i a0h = _mm_and_si128(bgra0, mask_h); // 0|bgr0|0|bgr0
325 const __m128i a4h = _mm_and_si128(bgra4, mask_h); // 0|bgr0|0|bgr
[all...] |
| /external/opencv3/modules/features2d/src/ |
| fast.cpp | 115 m0 = _mm_and_si128(_mm_cmpgt_epi8(x0, v0), _mm_cmpgt_epi8(x1, v0));
116 m1 = _mm_and_si128(_mm_cmpgt_epi8(v1, x0), _mm_cmpgt_epi8(v1, x1));
117 m0 = _mm_or_si128(m0, _mm_and_si128(_mm_cmpgt_epi8(x1, v0), _mm_cmpgt_epi8(x2, v0)));
118 m1 = _mm_or_si128(m1, _mm_and_si128(_mm_cmpgt_epi8(v1, x1), _mm_cmpgt_epi8(v1, x2)));
119 m0 = _mm_or_si128(m0, _mm_and_si128(_mm_cmpgt_epi8(x2, v0), _mm_cmpgt_epi8(x3, v0)));
120 m1 = _mm_or_si128(m1, _mm_and_si128(_mm_cmpgt_epi8(v1, x2), _mm_cmpgt_epi8(v1, x3)));
121 m0 = _mm_or_si128(m0, _mm_and_si128(_mm_cmpgt_epi8(x3, v0), _mm_cmpgt_epi8(x0, v0)));
122 m1 = _mm_or_si128(m1, _mm_and_si128(_mm_cmpgt_epi8(v1, x3), _mm_cmpgt_epi8(v1, x0)));
141 c0 = _mm_and_si128(_mm_sub_epi8(c0, m0), m0);
142 c1 = _mm_and_si128(_mm_sub_epi8(c1, m1), m1)
[all...] |
