Lines Matching refs:simdscalari
185 void vTranspose8x8(simdscalar (&vDst)[8], const simdscalari &vMask0, const simdscalari &vMask1, const simdscalari &vMask2, const simdscalari &vMask3, const simdscalari &vMask4, const simdscalari &vMask5, const simdscalari &vMask6, const simdscalari &vMask7)
226 simdscalari src = _simd_load_si((const simdscalari*)pSrc);
241 simdscalari dst01 = _simd_shuffle_epi8(src,
243 simdscalari dst23 = _mm256_permute2x128_si256(src, src, 0x01);
246 simdscalari dst = _simd_or_si(dst01, dst23);
247 _simd_store_si((simdscalari*)pDst, dst);
306 simdscalari src = _simd_load_si((const simdscalari*)pSrc);
323 simdscalari cvt0 = _simd_cvtepu8_epi16(src0);
324 simdscalari cvt1 = _simd_cvtepu8_epi16(src1);
326 simdscalari shl1 = _simd_slli_epi32(cvt1, 8);
328 simdscalari dst = _simd_or_si(cvt0, shl1);
330 _simd_store_si(reinterpret_cast<simdscalari *>(pDst), dst); // rgrgrgrgrgrgrgrgrgrgrgrgrgrgrgrg
504 simdscalari src_rg = _simd_load_si((const simdscalari*)pSrc);
505 simdscalari src_ba = _simd_load_si((const simdscalari*)(pSrc + sizeof(simdscalari)));
534 simdscalari src0 = _simd_load_si(reinterpret_cast<const simdscalari *>(pSrc)); // rrrrrrrrrrrrrrrr
535 simdscalari src1 = _simd_load_si(reinterpret_cast<const simdscalari *>(pSrc) + 1); // gggggggggggggggg
536 simdscalari src2 = _simd_load_si(reinterpret_cast<const simdscalari *>(pSrc) + 2); // bbbbbbbbbbbbbbbb
537 simdscalari src3 = _simd_load_si(reinterpret_cast<const simdscalari *>(pSrc) + 3); // aaaaaaaaaaaaaaaa
539 simdscalari pre0 = _simd_unpacklo_epi16(src0, src1); // rg0 rg1 rg2 rg3 rg8 rg9 rgA rgB
540 simdscalari pre1 = _simd_unpackhi_epi16(src0, src1); // rg4 rg5 rg6 rg7 rgC rgD rgE rgF
541 simdscalari pre2 = _simd_unpacklo_epi16(src2, src3); // ba0 ba1 ba3 ba3 ba8 ba9 baA baB
542 simdscalari pre3 = _simd_unpackhi_epi16(src2, src3); // ba4 ba5 ba6 ba7 baC baD baE baF
544 simdscalari tmp0 = _simd_unpacklo_epi32(pre0, pre2); // rbga0 rbga1 rbga8 rbga9
545 simdscalari tmp1 = _simd_unpackhi_epi32(pre0, pre2); // rbga2 rbga3 rbgaA rbgaB
546 simdscalari tmp2 = _simd_unpacklo_epi32(pre1, pre3); // rbga4 rbga5 rgbaC rbgaD
547 simdscalari tmp3 = _simd_unpackhi_epi32(pre1, pre3); // rbga6 rbga7 rbgaE rbgaF
549 simdscalari dst0 = _simd_permute2f128_si(tmp0, tmp1, 0x20); // (2, 0) // rbga0 rbga1 rbga2 rbga3
550 simdscalari dst1 = _simd_permute2f128_si(tmp2, tmp3, 0x20); // (2, 0) // rbga4 rbga5 rbga6 rbga7
551 simdscalari dst2 = _simd_permute2f128_si(tmp0, tmp1, 0x31); // (3, 1) // rbga8 rbga9 rbgaA rbgaB
552 simdscalari dst3 = _simd_permute2f128_si(tmp2, tmp3, 0x31); // (3, 1) // rbgaC rbgaD rbgaE rbgaF
554 _simd_store_si(reinterpret_cast<simdscalari *>(pDst) + 0, dst0); // rgbargbargbargba
555 _simd_store_si(reinterpret_cast<simdscalari *>(pDst) + 1, dst1); // rgbargbargbargba
556 _simd_store_si(reinterpret_cast<simdscalari *>(pDst) + 2, dst2); // rgbargbargbargba
557 _simd_store_si(reinterpret_cast<simdscalari *>(pDst) + 3, dst3); // rgbargbargbargba
574 simdscalari src_rg = _simd_load_si((const simdscalari*)pSrc);
578 simd4scalari src_b = SIMD128::load_si((const simd4scalari*)(pSrc + sizeof(simdscalari)));
603 simdscalari src0 = _simd_load_si(reinterpret_cast<const simdscalari *>(pSrc)); // rrrrrrrrrrrrrrrr
604 simdscalari src1 = _simd_load_si(reinterpret_cast<const simdscalari *>(pSrc) + 1); // gggggggggggggggg
605 simdscalari src2 = _simd_load_si(reinterpret_cast<const simdscalari *>(pSrc) + 2); // bbbbbbbbbbbbbbbb
606 simdscalari src3 = _simd_setzero_si(); // aaaaaaaaaaaaaaaa
608 simdscalari pre0 = _simd_unpacklo_epi16(src0, src1); // rg0 rg1 rg2 rg3 rg8 rg9 rgA rgB
609 simdscalari pre1 = _simd_unpackhi_epi16(src0, src1); // rg4 rg5 rg6 rg7 rgC rgD rgE rgF
610 simdscalari pre2 = _simd_unpacklo_epi16(src2, src3); // ba0 ba1 ba3 ba3 ba8 ba9 baA baB
611 simdscalari pre3 = _simd_unpackhi_epi16(src2, src3); // ba4 ba5 ba6 ba7 baC baD baE baF
613 simdscalari tmp0 = _simd_unpacklo_epi32(pre0, pre2); // rbga0 rbga1 rbga8 rbga9
614 simdscalari tmp1 = _simd_unpackhi_epi32(pre0, pre2); // rbga2 rbga3 rbgaA rbgaB
615 simdscalari tmp2 = _simd_unpacklo_epi32(pre1, pre3); // rbga4 rbga5 rgbaC rbgaD
616 simdscalari tmp3 = _simd_unpackhi_epi32(pre1, pre3); // rbga6 rbga7 rbgaE rbgaF
618 simdscalari dst0 = _simd_permute2f128_si(tmp0, tmp1, 0x20); // (2, 0) // rbga0 rbga1 rbga2 rbga3
619 simdscalari dst1 = _simd_permute2f128_si(tmp2, tmp3, 0x20); // (2, 0) // rbga4 rbga5 rbga6 rbga7
620 simdscalari dst2 = _simd_permute2f128_si(tmp0, tmp1, 0x31); // (3, 1) // rbga8 rbga9 rbgaA rbgaB
621 simdscalari dst3 = _simd_permute2f128_si(tmp2, tmp3, 0x31); // (3, 1) // rbgaC rbgaD rbgaE rbgaF
623 _simd_store_si(reinterpret_cast<simdscalari *>(pDst) + 0, dst0); // rgbargbargbargba
624 _simd_store_si(reinterpret_cast<simdscalari *>(pDst) + 1, dst1); // rgbargbargbargba
625 _simd_store_si(reinterpret_cast<simdscalari *>(pDst) + 2, dst2); // rgbargbargbargba
626 _simd_store_si(reinterpret_cast<simdscalari *>(pDst) + 3, dst3); // rgbargbargbargba
664 simdscalari src0 = _simd_load_si(reinterpret_cast<const simdscalari *>(pSrc)); // rrrrrrrrrrrrrrrr
665 simdscalari src1 = _simd_load_si(reinterpret_cast<const simdscalari
667 simdscalari tmp0 = _simd_unpacklo_epi16(src0, src1); // rg0 rg1 rg2 rg3 rg8 rg9 rgA rgB
668 simdscalari tmp1 = _simd_unpackhi_epi16(src0, src1); // rg4 rg5 rg6 rg7 rgC rgD rgE rgF
670 simdscalari dst0 = _simd_permute2f128_si(tmp0, tmp1, 0x20); // (2, 0) // rg0 rg1 rg2 rg3 rg4 rg5 rg6 rg7
671 simdscalari dst1 = _simd_permute2f128_si(tmp0, tmp1, 0x31); // (3, 1) // rg8 rg9 rgA rgB rgC rgD rgE rgF
673 _simd_store_si(reinterpret_cast<simdscalari *>(pDst) + 0, dst0); // rgrgrgrgrgrgrgrg
674 _simd_store_si(reinterpret_cast<simdscalari *>(pDst) + 1, dst1); // rgrgrgrgrgrgrgrg
