Release Notes¶

0.6.0 Version¶

Initial release
supported instruction sets AVX-512, FMA, AVX2, AVX, SSE
native code generation
scalar types int32, int64, float32, float64
vector types int32x2, int32x3, int32x4, int32x8, int32x16, int64x2, int64x3, int64x4, int64x8, float32x2, float32x3, float32x4, float32x8, float32x16, float64x2, float64x3, float64x4, float64x8
expression support for vector types
while loop
for loop (just few common cases)
arrays
user defined classes
multi-threading
built-in functions abs, round, ceil, floor, sqrt, min, max, ldexp, typecast, len, nthreads, thread_idx, copysign, dot, select, extract8
random_xx - replace xx with any scalar or vector type
complex functions sin, cos, exp, log, pown, pow, asin, acos, tan, atan, atan2