VecLib DSP Libraries
VecLib is a large set of routines offering high-performance, optimised implementations of vector elementwise functions.
The library contains 4576 functions of 2, 3, or 4 operands, real or complex, scalar or vector.
The functions have the following forms:
r[i] = a[i] op1 b[i], i = 1, ..., N
r[i] = a[i] op1 (b[i] op2 c[i]), i = 1, ..., N
r[i] = (a[i] op1 b[i]) op2 c[i], i = 1, ..., N
r[i] = (a[i] op1 b[i]) op2 (c[i] op3 d[i]), i = 1, ..., N, where
- each of op1, op2, and op3 can be +, -, × or ÷
- each of a, b, c, and d can be real or complex, scalar or vector
- the vectors can be strided independently.
Library Contents
Functions of 2 operands:
- Contains all possible combinations of real, complex, scalar or vector arguments.
- A total of 110 independent functions, 128 functions in all.
Real functions of 3 operands:
- Contains all possible combinations of real scalar or real vector arguments.
- A total of 149 independent functions, 336 functions in all.
Complex functions of 3 operands:
- Contains all possible combinations of complex scalar and complex vector arguments.
- A total of 1263 independent functions, 3024 functions in all.
Real vector functions of 4 operands:
- Contains all possible combinations of real vector arguments.
- A total of 62 independent functions, 64 functions in all.
Complex vector functions of 4 operands:
- Contains all possible combinations of complex vector arguments.
- A total of 805 independent functions, 1024 functions in all.
Naming Convention
VecLib is a large library, but using it is simplified by the systematic naming convention. Function names are made up of the components describing the function:
- brackets ( and ) are denoted b
- operations +, -, × and ÷ are denoted p, m, t, and d, respectively
- real vector or matrix operands are denoted r
- complex vector or matrix operands are denoted c
- conjugated complex vector or matrix operands are denoted j
- real scalar operands are denoted R
- complex scalar operands are denoted C
- conjugated complex scalars are denoted J
- functions for vectors start v
- functions for matrices start m
- transposed matrix inputs are denoted T
- transposed and conjugated matrix inputs are denoted H
- operands marked for cache removal have l appended to them.
Here are a few examples of function names: r, a, b, c are vectors, ß (beta) is a scalar.
| Function Name | Action | Comment |
|---|---|---|
| vrpbrmrb | r[i] = a[i] * (b[i] - c[i]) | all parameters real |
| vcpbjmjb | r[i] = a[i] + (b[i]' - c[i]') | all parameters complex |
| vcpbrtJb | r[i] = a[i] + (b[i] * ß') | r, a and beta are complex, b is real |
Implementations
- ARM A53, A57, and A72
- PPC T2080
- PPC T2081
- PPC T4240
- Intel AVX512
- Intel Phi
- Intel AVX2
- Intel AVX
- Intel SSE
