Searched refs:c12v (Results 1 – 6 of 6) sorted by relevance
| /dports/misc/mnn/MNN-1.2.0/source/backend/cpu/x86_x64/avx/ |
| H A D | MNNMatrixAdd.cpp | 35 auto c12v = _mm256_loadu_ps(c12Y + unit*x); in _AVX_MNNStrassenMergeCFunction() local
36 c12v = _mm256_add_ps(c12v, xv); in _AVX_MNNStrassenMergeCFunction()
37 c21v = _mm256_add_ps(c12v, c21v); in _AVX_MNNStrassenMergeCFunction()
38 c12v = _mm256_add_ps(c22v, c12v); in _AVX_MNNStrassenMergeCFunction()
40 c12v = _mm256_add_ps(c11v, c12v); in _AVX_MNNStrassenMergeCFunction()
41 _mm256_storeu_ps(c12Y + unit*x, c12v); in _AVX_MNNStrassenMergeCFunction()
|
| /dports/misc/mnn/MNN-1.2.0/source/backend/cpu/compute/ |
| H A D | ConvOpt.cpp | 298 auto c12v = Vec4::load(c12Y + 4*x); in MNNStrassenMergeCFunction() local
299 c12v = c12v + xv; in MNNStrassenMergeCFunction()
300 c21v = c12v + c21v; in MNNStrassenMergeCFunction()
301 c12v = c22v + c12v; in MNNStrassenMergeCFunction()
303 c12v = c11v + c12v; in MNNStrassenMergeCFunction()
304 Vec4::save(c12Y + 4*x, c12v); in MNNStrassenMergeCFunction()
|
| /dports/misc/mnn/MNN-1.2.0/source/backend/cpu/x86_x64/avx512/ |
| H A D | CommonOptFunction.cpp | 54 auto c12v = _mm256_loadu_ps(c12Y + 8*x); in _AVX512_MNNStrassenMergeCFunction() local
55 c12v = _mm256_add_ps(c12v, xv); in _AVX512_MNNStrassenMergeCFunction()
56 c21v = _mm256_add_ps(c12v, c21v); in _AVX512_MNNStrassenMergeCFunction()
57 c12v = _mm256_add_ps(c22v, c12v); in _AVX512_MNNStrassenMergeCFunction()
59 c12v = _mm256_add_ps(c11v, c12v); in _AVX512_MNNStrassenMergeCFunction()
60 _mm256_storeu_ps(c12Y + 8*x, c12v); in _AVX512_MNNStrassenMergeCFunction()
61 _mm256_storeu_ps(c12Y + 8*x, c12v); in _AVX512_MNNStrassenMergeCFunction()
|
| /dports/misc/mnn/MNN-1.2.0/source/backend/arm82/ |
| H A D | Arm82Functions.cpp | 254 auto c12v = vld1q_f16(c12Y + x * pack); in ARM82StrassenMerge() local
255 c12v = c12v + xv; in ARM82StrassenMerge()
256 c21v = c12v + c21v; in ARM82StrassenMerge()
257 c12v = c22v + c12v; in ARM82StrassenMerge()
259 c12v = c11v + c12v; in ARM82StrassenMerge()
260 vst1q_f16(c12Y + x * pack, c12v); in ARM82StrassenMerge()
|
| /dports/misc/mnn/MNN-1.2.0/source/backend/cpu/bf16/ |
| H A D | BF16Functions.cpp | 520 auto c12v = BFVec4::load(c12Y + 4*x); in _MNNStrassenMergeCFunction() local
521 c12v = c12v + xv; in _MNNStrassenMergeCFunction()
522 c21v = c12v + c21v; in _MNNStrassenMergeCFunction()
523 c12v = c22v + c12v; in _MNNStrassenMergeCFunction()
525 c12v = c11v + c12v; in _MNNStrassenMergeCFunction()
526 BFVec4::save(c12Y + 4*x, c12v); in _MNNStrassenMergeCFunction()
|
| /dports/science/libint2/libint-2.7.1/include/libint2/ |
| H A D | boys.h | 1165 libint2::simd::VectorAVXDouble c00v, c01v, c02v, c03v, c10v, c11v, c12v, c13v, in interpolate_Gm() local
1178 c12v.load_aligned((c_tuint+8) +ORDERp1); c13v.load_aligned((c_tuint+12) +ORDERp1); in interpolate_Gm()
1180 …libint2::simd::VectorAVXDouble t1 = t1vec * (c10v * u0vec + c11v * u1vec + c12v * u2vec + c13v * u… in interpolate_Gm()
|