Lines Matching defs:v_r0
310 __m128i v_r0 = _mm_packs_epi32(_mm_loadu_si128((__m128i const *)(row0 + x)),
318 __m128i v_dst00 = _mm_adds_epu16(_mm_adds_epu16(v_r0, v_r2), _mm_adds_epu16(v_2r1, v_4r1));
321 v_r0 = _mm_packs_epi32(_mm_loadu_si128((__m128i const *)(row0 + x + 8)),
329 __m128i v_dst01 = _mm_adds_epu16(_mm_adds_epu16(v_r0, v_r2), _mm_adds_epu16(v_2r1, v_4r1));
340 __m128i v_r0 = _mm_packs_epi32(_mm_loadu_si128((__m128i const *)(row0 + x)),
348 __m128i v_dst0 = _mm_adds_epu16(_mm_adds_epu16(v_r0, v_r2), _mm_adds_epu16(v_2r1, v_4r1));
374 __m128i v_r0 = _mm_loadu_si128((__m128i const *)(row0 + x)),
378 __m128i v_dst00 = _mm_add_epi32(_mm_add_epi32(v_r0, v_r2), _mm_add_epi32(v_2r1, v_4r1));
381 v_r0 = _mm_loadu_si128((__m128i const *)(row0 + x + 4));
386 __m128i v_dst01 = _mm_add_epi32(_mm_add_epi32(v_r0, v_r2), _mm_add_epi32(v_2r1, v_4r1));
399 __m128i v_r0 = _mm_loadu_si128((__m128i const *)(row0 + x)),
404 __m128i v_dst0 = _mm_add_epi32(_mm_add_epi32(v_r0, v_r2), _mm_add_epi32(v_2r1, v_4r1));
434 __m128i v_r0 = _mm_loadu_si128((__m128i const *)(row0 + x)),
438 __m128i v_dst00 = _mm_add_epi32(_mm_add_epi32(v_r0, v_r2), _mm_add_epi32(v_2r1, v_4r1));
441 v_r0 = _mm_loadu_si128((__m128i const *)(row0 + x + 4));
446 __m128i v_dst01 = _mm_add_epi32(_mm_add_epi32(v_r0, v_r2), _mm_add_epi32(v_2r1, v_4r1));
459 __m128i v_r0 = _mm_loadu_si128((__m128i const *)(row0 + x)),
464 __m128i v_dst0 = _mm_add_epi32(_mm_add_epi32(v_r0, v_r2), _mm_add_epi32(v_2r1, v_4r1));
499 __m128 v_r0 = _mm_loadu_ps(row0 + x);
504 _mm_storeu_ps(dst0 + x, _mm_mul_ps(v_scale, _mm_add_ps(_mm_add_ps(v_r0, _mm_mul_ps(v_6, v_r1)), v_r2)));
506 v_r0 = _mm_loadu_ps(row0 + x + 4);
511 _mm_storeu_ps(dst0 + x + 4, _mm_mul_ps(v_scale, _mm_add_ps(_mm_add_ps(v_r0, _mm_mul_ps(v_6, v_r1)), v_r2)));
532 uint16x8_t v_r0 = vcombine_u16(vqmovn_u32(vld1q_u32(row0 + x)), vqmovn_u32(vld1q_u32(row0 + x + 4)));
538 v_r0 = vaddq_u16(vaddq_u16(v_r0, v_r4), vaddq_u16(v_r2, v_r2));
540 uint16x8_t v_dst0 = vaddq_u16(v_r0, vshlq_n_u16(v_r1, 2));
542 v_r0 = vcombine_u16(vqmovn_u32(vld1q_u32(row0 + x + 8)), vqmovn_u32(vld1q_u32(row0 + x + 12)));
548 v_r0 = vaddq_u16(vaddq_u16(v_r0, v_r4), vaddq_u16(v_r2, v_r2));
550 uint16x8_t v_dst1 = vaddq_u16(v_r0, vshlq_n_u16(v_r1, 2));
637 float32x4_t v_r0 = vld1q_f32(row0 + x);
643 v_r0 = vaddq_f32(vaddq_f32(v_r0, v_r4), vaddq_f32(v_r2, v_r2));
645 vst1q_f32(dst + x, vmulq_f32(vmlaq_f32(v_r0, v_4, v_r1), v_scale));
647 v_r0 = vld1q_f32(row0 + x + 4);
653 v_r0 = vaddq_f32(vaddq_f32(v_r0, v_r4), vaddq_f32(v_r2, v_r2));
655 vst1q_f32(dst + x + 4, vmulq_f32(vmlaq_f32(v_r0, v_4, v_r1), v_scale));
673 uint16x8_t v_r0 = vcombine_u16(vqmovn_u32(vld1q_u32(row0 + x)), vqmovn_u32(vld1q_u32(row0 + x + 4)));
678 uint16x8_t v_dst00 = vaddq_u16(vaddq_u16(v_r0, v_r2), vaddq_u16(v_2r1, v_4r1));
681 v_r0 = vcombine_u16(vqmovn_u32(vld1q_u32(row0 + x + 8)), vqmovn_u32(vld1q_u32(row0 + x + 12)));
686 uint16x8_t v_dst01 = vaddq_u16(vaddq_u16(v_r0, v_r2), vaddq_u16(v_2r1, v_4r1));
697 uint16x8_t v_r0 = vcombine_u16(vqmovn_u32(vld1q_u32(row0 + x)), vqmovn_u32(vld1q_u32(row0 + x + 4)));
702 uint16x8_t v_dst0 = vaddq_u16(vaddq_u16(v_r0, v_r2), vaddq_u16(v_2r1, v_4r1));
724 uint32x4_t v_r0 = vld1q_u32(row0 + x), v_r1 = vld1q_u32(row1 + x), v_r2 = vld1q_u32(row2 + x);
726 uint32x4_t v_dst00 = vaddq_u32(vaddq_u32(v_r0, v_r2), vaddq_u32(v_2r1, v_4r1));
729 v_r0 = vld1q_u32(row0 + x + 4);
734 uint32x4_t v_dst01 = vaddq_u32(vaddq_u32(v_r0, v_r2), vaddq_u32(v_2r1, v_4r1));
745 uint32x4_t v_r0 = vld1q_u32(row0 + x), v_r1 = vld1q_u32(row1 + x), v_r2 = vld1q_u32(row2 + x);
748 uint32x4_t v_dst0 = vaddq_u32(vaddq_u32(v_r0, v_r2), vaddq_u32(v_2r1, v_4r1));
770 int32x4_t v_r0 = vld1q_s32(row0 + x), v_r1 = vld1q_s32(row1 + x), v_r2 = vld1q_s32(row2 + x);
772 int32x4_t v_dst00 = vaddq_s32(vaddq_s32(v_r0, v_r2), vaddq_s32(v_2r1, v_4r1));
775 v_r0 = vld1q_s32(row0 + x + 4);
780 int32x4_t v_dst01 = vaddq_s32(vaddq_s32(v_r0, v_r2), vaddq_s32(v_2r1, v_4r1));
791 int32x4_t v_r0 = vld1q_s32(row0 + x), v_r1 = vld1q_s32(row1 + x), v_r2 = vld1q_s32(row2 + x);
794 int32x4_t v_dst0 = vaddq_s32(vaddq_s32(v_r0, v_r2), vaddq_s32(v_2r1, v_4r1));
816 float32x4_t v_r0 = vld1q_f32(row0 + x);
821 vst1q_f32(dst0 + x, vmulq_f32(v_scale, vaddq_f32(vmlaq_f32(v_r0, v_6, v_r1), v_r2)));
823 v_r0 = vld1q_f32(row0 + x + 4);
828 vst1q_f32(dst0 + x + 4, vmulq_f32(v_scale, vaddq_f32(vmlaq_f32(v_r0, v_6, v_r1), v_r2)));
