simdpp::splat16, splat8, splat4, splat2
From libsimdpp-docs
< shuffle
template<unsigned s, unsigned N, class V1>
Ret<N, _DETAIL_> splat16( const any_vec8<N,V1>& a ); |
||
template<unsigned s, unsigned N, class V1>
Ret<N, _DETAIL_> splat8( const any_vec16<N,V1>& a ); |
||
template<unsigned s, unsigned N, class V1>
Ret<N, _DETAIL_> splat4( const any_vec32<N,V1>& a ); |
||
template<unsigned s, unsigned N, class V1>
Ret<N, _DETAIL_> splat2( const any_vec64<N,V1>& a ); |
||
Broadcasts the s
-th element within the vector across all lates. Each 128-bit lane within vectors is processed separately.
The return vector type is governed by the expression promotion rules.
Parameters[edit]
a | - | source vector |
s | - | the index of the element to broadcast |
Return value[edit]
Equivalent operation[edit]
For the any_vec32
case on 128-bit vector:
r = [ a[s], a[s], a[s], a[s] ]
See also[edit]
This section is incomplete |