Searched refs:q4p4 (Results 1 – 5 of 5) sorted by relevance
| /external/libaom/libaom/aom_dsp/x86/ |
| D | loopfilter_sse2.c | 31 __m128i *q3p3, __m128i *q4p4, in transpose_pq_14_sse2() argument
65 *q4p4 = _mm_unpacklo_epi32( in transpose_pq_14_sse2()
395 __m128i *q6p6, __m128i *q5p5, __m128i *q4p4, __m128i *q3p3, __m128i *q2p2, in lpf_internal_14_dual_sse2() argument
475 p4_16 = _mm_unpacklo_epi8(*q4p4, zero); in lpf_internal_14_dual_sse2()
484 q4_16 = _mm_unpackhi_epi8(*q4p4, zero); in lpf_internal_14_dual_sse2()
561 flat2 = _mm_max_epu8(abs_diff(*q4p4, *q0p0), abs_diff(*q5p5, *q0p0)); in lpf_internal_14_dual_sse2()
671 *q4p4 = _mm_andnot_si128(flat2, *q4p4); in lpf_internal_14_dual_sse2()
673 *q4p4 = _mm_or_si128(*q4p4, flat2_q4p4); in lpf_internal_14_dual_sse2()
698 __m128i *q6p6, __m128i *q5p5, __m128i *q4p4, __m128i *q3p3, __m128i *q2p2, in lpf_internal_14_sse2() argument
777 pq_16[4] = _mm_unpacklo_epi8(*q4p4, zero); in lpf_internal_14_sse2()
[all …]
|
| /external/libvpx/libvpx/vpx_dsp/x86/ |
| D | loopfilter_avx2.c | 23 __m128i q7p7, q6p6, q5p5, q4p4, q3p3, q2p2, q1p1, q0p0, p0q0, p1q1; in vpx_lpf_horizontal_16_avx2() local
32 q4p4 = _mm_loadl_epi64((__m128i *)(s - 5 * pitch)); in vpx_lpf_horizontal_16_avx2()
33 q4p4 = _mm_castps_si128( in vpx_lpf_horizontal_16_avx2()
34 _mm_loadh_pi(_mm_castsi128_ps(q4p4), (__m64 *)(s + 4 * pitch))); in vpx_lpf_horizontal_16_avx2()
148 _mm_or_si128(_mm_subs_epu8(q4p4, q0p0), _mm_subs_epu8(q0p0, q4p4)), in vpx_lpf_horizontal_16_avx2()
180 p4_16 = _mm_unpacklo_epi8(q4p4, zero); in vpx_lpf_horizontal_16_avx2()
189 q4_16 = _mm_unpackhi_epi8(q4p4, zero); in vpx_lpf_horizontal_16_avx2()
333 q4p4 = _mm_andnot_si128(flat2, q4p4); in vpx_lpf_horizontal_16_avx2()
335 q4p4 = _mm_or_si128(q4p4, flat2_q4p4); in vpx_lpf_horizontal_16_avx2()
336 _mm_storel_epi64((__m128i *)(s - 5 * pitch), q4p4); in vpx_lpf_horizontal_16_avx2()
[all …]
|
| D | loopfilter_sse2.c | 241 __m128i q7p7, q6p6, q5p5, q4p4, q3p3, q2p2, q1p1, q0p0, p0q0, p1q1; in vpx_lpf_horizontal_16_sse2() local
244 q4p4 = _mm_loadl_epi64((__m128i *)(s - 5 * pitch)); in vpx_lpf_horizontal_16_sse2()
245 q4p4 = _mm_castps_si128( in vpx_lpf_horizontal_16_sse2()
246 _mm_loadh_pi(_mm_castsi128_ps(q4p4), (__m64 *)(s + 4 * pitch))); in vpx_lpf_horizontal_16_sse2()
351 flat2 = _mm_max_epu8(abs_diff(q4p4, q0p0), abs_diff(q5p5, q0p0)); in vpx_lpf_horizontal_16_sse2()
378 p4_16 = _mm_unpacklo_epi8(q4p4, zero); in vpx_lpf_horizontal_16_sse2()
387 q4_16 = _mm_unpackhi_epi8(q4p4, zero); in vpx_lpf_horizontal_16_sse2()
531 q4p4 = _mm_andnot_si128(flat2, q4p4); in vpx_lpf_horizontal_16_sse2()
533 q4p4 = _mm_or_si128(q4p4, flat2_q4p4); in vpx_lpf_horizontal_16_sse2()
534 _mm_storel_epi64((__m128i *)(s - 5 * pitch), q4p4); in vpx_lpf_horizontal_16_sse2()
[all …]
|
| /external/libgav1/libgav1/src/dsp/arm/ |
| D | loop_filter_neon.cc | 814 const uint8x8_t q4p4 = Transpose32(p4q4); in Filter14() local
815 sum = vaddq_u16(vaddl_u8(p0q0, q4p4), sum); in Filter14()
|
| /external/libgav1/libgav1/src/dsp/x86/ |
| D | loop_filter_sse4.cc | 916 __m128i* q3p3, __m128i* q4p4, __m128i* q5p5, in DualTranspose8x4To4x8() argument
941 *q4p4 = _mm_unpacklo_epi32(_mm_srli_si128(ww0, 12), ww3); in DualTranspose8x4To4x8()
|