Lines Matching refs:SIMDINLINE
37 static SIMDINLINE __m512 __conv(Float r) { return _mm512_castps256_ps512(r.v); }
38 static SIMDINLINE __m512d __conv(Double r) { return _mm512_castpd256_pd512(r.v); }
39 static SIMDINLINE __m512i __conv(Integer r) { return _mm512_castsi256_si512(r.v); }
40 static SIMDINLINE Float __conv(__m512 r) { return _mm512_castps512_ps256(r); }
41 static SIMDINLINE Double __conv(__m512d r) { return _mm512_castpd512_pd256(r); }
42 static SIMDINLINE Integer __conv(__m512i r) { return _mm512_castsi512_si256(r); }
46 static SIMDINLINE Float SIMDCALL op(Float a) \
54 static SIMDINLINE Float SIMDCALL op(Float a) \
61 static SIMDINLINE Float SIMDCALL op(Float a, Float b) \
69 static SIMDINLINE Float SIMDCALL op(Float a, Float b) \
75 static SIMDINLINE Float SIMDCALL op(Float a, Float b, Float c) \
83 static SIMDINLINE Double SIMDCALL op(Double a, Double b) \
89 static SIMDINLINE Integer SIMDCALL op(Integer a) \
97 static SIMDINLINE Integer SIMDCALL op(Integer a) \
104 static SIMDINLINE Integer SIMDCALL op(Integer a, Integer b) \
112 static SIMDINLINE Integer SIMDCALL op(Integer a, Integer b) \
197 //static SIMDINLINE Integer SIMDCALL cmplt_epi32(Integer a, Integer b) // return a < b (int32)
212 //static SIMDINLINE Float SIMDCALL permute_ps(Float a, Integer swiz) // return a[swiz[i]] for each 32-bit lane i (float)
219 //static SIMDINLINE Integer SIMDCALL shuffle_epi64(Integer a, Integer b)
237 static SIMDINLINE Float SIMDCALL load_ps(float const *p) // return *p (loads SIMD width elements from memory)
242 static SIMDINLINE Integer SIMDCALL load_si(Integer const *p) // return *p
247 static SIMDINLINE Float SIMDCALL loadu_ps(float const *p) // return *p (same as load_ps but allows for unaligned mem)
252 static SIMDINLINE Integer SIMDCALL loadu_si(Integer const *p) // return *p (same as load_si but allows for unaligned mem)
258 static SIMDINLINE Float SIMDCALL i32gather_ps(float const* p, Integer idx) // return *(float*)(((int8*)p) + (idx * ScaleT))
270 static SIMDINLINE Float SIMDCALL mask_i32gather_ps(Float old, float const* p, Integer idx, Float mask)
283 // static SIMDINLINE uint32_t SIMDCALL movemask_epi8(Integer a)
290 static SIMDINLINE void SIMDCALL maskstore_ps(float *p, Integer mask, Float src)
297 static SIMDINLINE void SIMDCALL store_ps(float *p, Float a) // *p = a (stores all elements contiguously in memory)
302 static SIMDINLINE void SIMDCALL store_si(Integer *p, Integer a) // *p = a
307 static SIMDINLINE Float SIMDCALL vmask_ps(int32_t mask)
