simdpp::permute2

 

libsimdpp

Types

Capabilities

Operations

Conversions

Bitwise operations

Floating-point operations

Integer operations

Memory access operations

Shuffle operations

Miscellaneous operations

 

Shuffling operations

blend splat combine split	permute2 permute4 shuffle2x2 shuffle4x2	transpose2 transpose4
Shuffles operating on 128-bit blocks
align16 align8 align4 align2
move16_l move8_l move4_l move2_l
splat16 splat8 splat4 splat2
move16_r move8_r move4_r move2_r
zip16_hi zip8_hi zip4_hi zip2_hi
zip16_lo zip8_lo zip4_lo zip2_lo
unzip16_hi unzip8_hi unzip4_hi unzip2_hi
unzip16_lo unzip8_lo unzip4_lo unzip2_lo
make_shuffle_bytes16_mask
shuffle_bytes16 shuffle_zbytes16	permute_bytes16 permute_zbytes16

 

template<unsigned s0, unsigned s1, unsigned N, class E> Ret<N,_DETAIL_> permute2( const Vec<N,E>& a );

Permutes pairs of elements within the source vector. The implementation behaves as if the following set of overloads is provided:

Ret	Vec
int16	int16
uint16	uint16
uint16	mask_int16
int32	int32
uint32	uint32
uint32	mask_int32
int64	int64
uint64	uint64
uint64	mask_int64
float32	float32
float32	mask_float32
float64	float64
float64	mask_float64

The selectors s0 and s1 must be in range [0..1].

The return type is a vector expression.

Parameters

a	-	source vector
s0,s1	-	the permutation indices

Return value

A vector expression evaluating to the permuted vector.

Equivalent operation

r0 = a[s0]
r1 = a[s1]
r2 = a[s0+2]
r3 = a[s1+2]
...
rM-1 = a[s0+M-1]
rM   = a[s1+M-1]

Here M = N-1.