Lines Matching refs:q1

326                                   const __m128i* const q1,  in GetNotHEV()  argument
330 const __m128i t_2 = MM_ABS(*q1, *q0); in GetNotHEV()
344 const __m128i* const q1, in GetBaseDelta() argument
347 const __m128i p1_q1 = _mm_subs_epi8(*p1, *q1); // p1 - q1 in GetBaseDelta()
389 const __m128i* const q1, in NeedsFilter() argument
392 const __m128i t1 = MM_ABS(*p1, *q1); // abs(p1 - q1) in NeedsFilter()
410 __m128i* const q0, __m128i* const q1, in DoFilter2() argument
416 const __m128i q1s = _mm_xor_si128(*q1, sign_bit); in DoFilter2()
418 NeedsFilter(p1, p0, q0, q1, thresh, &mask); in DoFilter2()
429 __m128i* const q0, __m128i* const q1, in DoFilter4() argument
438 GetNotHEV(p1, p0, q0, q1, hev_thresh, &not_hev); in DoFilter4()
441 FLIP_SIGN_BIT4(*p1, *p0, *q0, *q1); in DoFilter4()
443 t1 = _mm_subs_epi8(*p1, *q1); // p1 - q1 in DoFilter4()
467 *q1 = _mm_subs_epi8(*q1, t3); // q1 -= t3 in DoFilter4()
469 FLIP_SIGN_BIT2(*p1, *q1); in DoFilter4()
475 __m128i* const q1, __m128i* const q2, in DoFilter6() argument
482 GetNotHEV(p1, p0, q0, q1, hev_thresh, &not_hev); in DoFilter6()
484 FLIP_SIGN_BIT4(*p1, *p0, *q0, *q1); in DoFilter6()
486 GetBaseDelta(p1, p0, q0, q1, &a); in DoFilter6()
517 Update2Pixels(p1, q1, &a1_lo, &a1_hi); in DoFilter6()
570 __m128i* const q0, __m128i* const q1) { in Load16x4() argument
588 Load8x4(r8, stride, p0, q1); in Load16x4()
598 *q0 = _mm_unpacklo_epi64(t2, *q1); in Load16x4()
599 *q1 = _mm_unpackhi_epi64(t2, *q1); in Load16x4()
614 const __m128i* const q1, in Store16x4() argument
628 q0_s = _mm_unpacklo_epi8(t1, *q1); in Store16x4()
629 q1_s = _mm_unpackhi_epi8(t1, *q1); in Store16x4()
660 __m128i q1 = _mm_loadu_si128((__m128i*)&p[stride]); in SimpleVFilter16() local
662 DoFilter2(&p1, &p0, &q0, &q1, thresh); in SimpleVFilter16()
670 __m128i p1, p0, q0, q1; in SimpleHFilter16() local
674 Load16x4(p, p + 8 * stride, stride, &p1, &p0, &q0, &q1); in SimpleHFilter16()
675 DoFilter2(&p1, &p0, &q0, &q1, thresh); in SimpleHFilter16()
676 Store16x4(&p1, &p0, &q0, &q1, p, p + 8 * stride, stride); in SimpleHFilter16()
739 const __m128i* const q1, in ComplexMask() argument
746 NeedsFilter(p1, p0, q0, q1, thresh, &filter_mask); in ComplexMask()
755 __m128i p2, p1, p0, q0, q1, q2; in VFilter16() local
762 LOAD_H_EDGES4(p, stride, q0, q1, q2, t1); in VFilter16()
763 MAX_DIFF2(t1, q2, q1, q0, mask); in VFilter16()
765 ComplexMask(&p1, &p0, &q0, &q1, thresh, ithresh, &mask); in VFilter16()
766 DoFilter6(&p2, &p1, &p0, &q0, &q1, &q2, &mask, hev_thresh); in VFilter16()
773 _mm_storeu_si128((__m128i*)&p[+1 * stride], q1); in VFilter16()
780 __m128i p3, p2, p1, p0, q0, q1, q2, q3; in HFilter16() local
786 Load16x4(p, p + 8 * stride, stride, &q0, &q1, &q2, &q3); // q0, q1, q2, q3 in HFilter16()
787 MAX_DIFF2(q3, q2, q1, q0, mask); in HFilter16()
789 ComplexMask(&p1, &p0, &q0, &q1, thresh, ithresh, &mask); in HFilter16()
790 DoFilter6(&p2, &p1, &p0, &q0, &q1, &q2, &mask, hev_thresh); in HFilter16()
793 Store16x4(&q0, &q1, &q2, &q3, p, p + 8 * stride, stride); in HFilter16()
862 __m128i t1, p2, p1, p0, q0, q1, q2; in VFilter8() local
869 LOADUV_H_EDGES4(u, v, stride, q0, q1, q2, t1); in VFilter8()
870 MAX_DIFF2(t1, q2, q1, q0, mask); in VFilter8()
872 ComplexMask(&p1, &p0, &q0, &q1, thresh, ithresh, &mask); in VFilter8()
873 DoFilter6(&p2, &p1, &p0, &q0, &q1, &q2, &mask, hev_thresh); in VFilter8()
880 STOREUV(q1, u, v, 1 * stride); in VFilter8()
887 __m128i p3, p2, p1, p0, q0, q1, q2, q3; in HFilter8() local
894 Load16x4(u, v, stride, &q0, &q1, &q2, &q3); // q0, q1, q2, q3 in HFilter8()
895 MAX_DIFF2(q3, q2, q1, q0, mask); in HFilter8()
897 ComplexMask(&p1, &p0, &q0, &q1, thresh, ithresh, &mask); in HFilter8()
898 DoFilter6(&p2, &p1, &p0, &q0, &q1, &q2, &mask, hev_thresh); in HFilter8()
901 Store16x4(&q0, &q1, &q2, &q3, u, v, stride); in HFilter8()
907 __m128i t1, t2, p1, p0, q0, q1; in VFilter8i() local
917 LOADUV_H_EDGES4(u, v, stride, q0, q1, t1, t2); in VFilter8i()
918 MAX_DIFF2(t2, t1, q1, q0, mask); in VFilter8i()
920 ComplexMask(&p1, &p0, &q0, &q1, thresh, ithresh, &mask); in VFilter8i()
921 DoFilter4(&p1, &p0, &q0, &q1, &mask, hev_thresh); in VFilter8i()
927 STOREUV(q1, u, v, 1 * stride); in VFilter8i()
933 __m128i t1, t2, p1, p0, q0, q1; in HFilter8i() local
939 Load16x4(u, v, stride, &q0, &q1, &t1, &t2); // q0, q1, q2, q3 in HFilter8i()
940 MAX_DIFF2(t2, t1, q1, q0, mask); in HFilter8i()
942 ComplexMask(&p1, &p0, &q0, &q1, thresh, ithresh, &mask); in HFilter8i()
943 DoFilter4(&p1, &p0, &q0, &q1, &mask, hev_thresh); in HFilter8i()
947 Store16x4(&p1, &p0, &q0, &q1, u, v, stride); in HFilter8i()