simdpp::make_shuffle_bytes16_mask

 

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

 

8-bit element shuffle mask
template<int s0, int s1, unsigned N> uint8<N> make_shuffle_bytes16_mask( uint8<N> &mask );
template<int s0, int s1, int s2, int s3, unsigned N> uint8<N> make_shuffle_bytes16_mask( uint8<N> &mask );
template<int s0, int s1, int s2, int s3, int s4, int s5, int s6, int s7, unsigned N> uint8<N> make_shuffle_bytes16_mask( uint8<N> &mask );
template<int s0, int s1, int s2, int s3, int s4, int s5, int s6, int s7,          int s8, int s9, int s10, int s11, int s12, int s13, int s14, int s15,          unsigned N> uint8<N> make_shuffle_bytes16_mask( uint8<N> &mask );
16-bit element shuffle mask
template<int s0, int s1, unsigned N> uint16<N> make_shuffle_bytes16_mask( uint16<N> &mask );
template<int s0, int s1, int s2, int s3, unsigned N> uint16<N> make_shuffle_bytes16_mask( uint16<N> &mask );
template<int s0, int s1, int s2, int s3, int s4, int s5, int s6, int s7, unsigned N> uint16<N> make_shuffle_bytes16_mask( uint16<N> &mask );
32-bit element shuffle mask
template<int s0, int s1, unsigned N> uint32<N> make_shuffle_bytes16_mask( uint32<N> &mask );
template<int s0, int s1, int s2, int s3, unsigned N> uint32<N> make_shuffle_bytes16_mask( uint32<N> &mask );
64-bit element shuffle mask
template<int s0, int s1, unsigned N> uint64<N> make_shuffle_bytes16_mask( uint64<N> &mask );

Makes a mask to shuffle an vector elements within 16-byte lanes using permute_bytes16(), permute_zbytes16(), shuffle_bytes16() or shuffle_zbytes16() functions.

The type of the mask argument determines the type of the vector to create mask for.

The number of the selector template parameters (s0, s1, ...) determines the actual lane width within which the elements will be shuffled. The maximum size of the lane elements will be shuffled across is 16 bytes.

2 element selector template

Elements are shuffled across element pairs.

All elements within vectors are grouped into lanes of two adjacent elements. Elements within each lane of the resulting vector can be selected only from corresponding sets of the source vectors.

The template arguments define which elements to select from each element group:

• Values ]0,1] select elements from the first vector.
• Values [2,3] select elements from the second vector. The mask can only be used in shuffle_bytes16() or shuffle_zbytes16().
• Value -1 sets the corresponding element to zero. The mask can only be used in @c permute_bytes16(), shuffle_zbytes16().

The created mask will cause shuffle_zbytes16() to perform as follows:

r0 = (s0 == -1) ? 0 : (s0 < 2 ? a[s0] : b[s0-2])
r1 = (s1 == -1) ? 0 : (s1 < 2 ? a[s1] : b[s1-2])
r2 = (s0 == -1) ? 0 : (s0 < 2 ? a[s0+2] : b[s0])
r3 = (s1 == -1) ? 0 : (s1 < 2 ? a[s1+2] : b[s1])
...

4 element selector template

Elements are shuffled across element pairs.

All elements within vectors are grouped into lanes of 4 adjacent elements. Elements within each lane of the resulting vector can be selected only from corresponding sets of the source vectors.

The template arguments define which elements to select from each element group:

• Values ]0,3] select elements from the first vector.
• Values [4,7] select elements from the second vector. The mask can only be used in shuffle_bytes16() or shuffle_zbytes16().
• Value -1 sets the corresponding element to zero. The mask can only be used in @c permute_bytes16(), shuffle_zbytes16().

The created mask will cause shuffle_zbytes16() to perform as follows:

r0 = (s0 == -1) ? 0 : (s0 < 4 ? a[s0] : b[s0-4])
r1 = (s1 == -1) ? 0 : (s1 < 4 ? a[s1] : b[s1-4])
r2 = (s2 == -1) ? 0 : (s2 < 4 ? a[s2] : b[s2-4])
r3 = (s3 == -1) ? 0 : (s3 < 4 ? a[s3] : b[s3-4])
...

8 element selector template

Elements are shuffled across element pairs.

All elements within vectors are grouped into lanes of 8 adjacent elements. Elements within each lane of the resulting vector can be selected only from corresponding sets of the source vectors.

The template arguments define which elements to select from each element group:

• Values ]0,7] select elements from the first vector.
• Values [8,15] select elements from the second vector. The mask can only be used in shuffle_bytes16() or shuffle_zbytes16().
• Value -1 sets the corresponding element to zero. The mask can only be used in @c permute_bytes16(), shuffle_zbytes16().

The created mask will cause shuffle_zbytes16() to perform as follows:

r0 = (s0 == -1) ? 0 : (s0 < 8 ? a[s0] : b[s0-8])
...
r7 = (s7 == -1) ? 0 : (s7 < 8 ? a[s7] : b[s7-8])
...

16 element selector template

Elements are shuffled across element pairs.

All elements within vectors are grouped into lanes of 8 adjacent elements. Elements within each lane of the resulting vector can be selected only from corresponding sets of the source vectors.

The template arguments define which elements to select from each element group:

• Values ]0,15] select elements from the first vector.
• Values [16,31] select elements from the second vector. The mask can only be used in shuffle_bytes16() or shuffle_zbytes16().
• Value -1 sets the corresponding element to zero. The mask can only be used in @c permute_bytes16(), shuffle_zbytes16().

The created mask will cause shuffle_zbytes16() to perform as follows:

r0 = (s0 == -1) ? 0 : (s0 < 16 ? a[s0] : b[s0-16])
r1 = (s1 == -1) ? 0 : (s0 < 16 ? a[s1] : b[s1-16])
...
r15 = (s15 == -1) ? 0 : (s15 < 16 ? a[s15] : b[s15-16])
...

Parameters

mask

-

mask vector

Return value

mask

See also

shuffle_bytes16	selects elements from two vectors according to another vector (function template)
shuffle_zbytes16	selects or zeroes elements from two vectors according to another vector (function template)
permute_bytes16	permutes elements in a vector according to another vector (function template)
permute_zbytes16	permutes or zeroes elements in a vector according to another vector (function template)