Intrinsics for Scale Operations

The prototypes for Intel® Advanced Vector Extensions 512 (Intel® AVX-512) intrinsics are located in the zmmintrin.h header file.

To use these intrinsics, include the immintrin.h file as follows:

#include <immintrin.h>


Intrinsic Name

Operation

Corresponding
Intel® AVX-512 Instruction

_mm512_scalef_pd, _mm512_mask_scalef_pd, _mm512_maskz_scalef_pd

_mm512_scalef_round_pd, _mm512_mask_scalef_round_pd, _mm512_maskz_scalef_round_pd

Scale packed float64 values with float64 values.

VSCALEFPD

_mm512_scalef_ps, _mm512_mask_scalef_ps, _mm512_maskz_scalef_ps

_mm512_scalef_round_ps, _mm512_mask_scalef_round_ps, _mm512_maskz_scalef_round_ps

Scale packed float32 values with float32 values.

VSCALEFSD

_mm_scalef_sd, _mm_mask_scalef_sd, _mm_maskz_scalef_sd

_mm_scalef_round_sd, _mm_mask_scalef_round_sd, _mm_maskz_scalef_round_sd

Scale scalar float64 values with float64 values.

VSCALEFPS

_mm_scalef_round_ss, _mm_mask_scalef_round_ss, _mm_maskz_scalef_round_ss

_mm_scalef_ss, _mm_mask_scalef_ss, _mm_maskz_scalef_ss

Scale scalar float32 values with float32 values.

VSCALEFSS

_mm512_roundscale_pd, _mm512_mask_roundscale_pd, _mm512_maskz_roundscale_pd

Scale packed float64 values with float64 values.

VRNDSCALEPD

_mm512_roundscale_ps, _mm512_mask_roundscale_ps, _mm512_maskz_roundscale_ps

Scale packed float32 values with float32 values.

VRNDSCALEPS

_mm_roundscale_sd, _mm_mask_roundscale_sd, _mm_maskz_roundscale_sd

_mm_roundscale_round_sd, _mm_mask_roundscale_round_sd, _mm_maskz_roundscale_round_sd

Scale scalar float64 values with float64 values.

VRNDSCALESD

_mm_roundscale_ss, _mm_mask_roundscale_ss, _mm_maskz_roundscale_ss

_mm_roundscale_round_ss, _mm_mask_roundscale_round_ss, _mm_maskz_roundscale_round_ss

Scale scalar float32 values with float32 values.

VRNDSCALE


variable definition
k

writemask used as a selector

a

first source vector element

b

second source vector element

src

source element to use based on writemask result

round

Rounding control values; these can be one of the following (along with the sae suppress all exceptions flag):

  • _MM_FROUND_TO_NEAREST_INT - rounds to nearest even
  • _MM_FROUND_TO_NEG_INF - rounds to negative infinity
  • _MM_FROUND_TO_POS_INF - rounds to positive infinity
  • _MM_FROUND_TO_ZERO - rounds to zero
  • _MM_FROUND_CUR_DIRECTION - rounds using default from MXCSR register
imm

8-bit immediate integer specifies offset for destination


_mm512_roundscale_pd

extern __m512d __cdecl _mm512_roundscale_pd(__m512d a, int imm);

Performs a floating point scale of packed float64 elements in a to the number of fraction bits specified by imm, and stores the result.



_mm512_mask_roundscale_pd

extern __m512d __cdecl _mm512_mask_roundscale_pd(__m512d src, __mmask8 k, __m512d a, int imm);

Performs a floating point scale of packed float64 elements in a to the number of fraction bits specified by imm, and stores the result using writemask k (elements are copied from src when the corresponding mask bit is not set).



_mm512_maskz_roundscale_pd

extern __m512d __cdecl _mm512_maskz_roundscale_pd(__mmask8 k, __m512d a, int imm);

Performs a floating point scale of packed float64 elements in a to the number of fraction bits specified by imm, and stores the result using zeromask k (elements are zeroed out when the corresponding mask bit is not set).



_mm512_roundscale_ps

extern __m512 __cdecl _mm512_roundscale_ps(__m512 a, int imm);

Performs a floating point scale of packed float32 elements in a to the number of fraction bits specified by imm, and stores the result.



_mm512_mask_roundscale_ps

extern __m512 __cdecl _mm512_mask_roundscale_ps(__m512 src, __mmask16 k, __m512 a, int imm);

Performs a floating point scale of packed float32 elements in a to the number of fraction bits specified by imm, and stores the result using writemask k (elements are copied from src when the corresponding mask bit is not set).



_mm512_maskz_roundscale_ps

extern __m512 __cdecl _mm512_maskz_roundscale_ps(__mmask16 k, __m512 a, int imm);

Performs a floating point scale of packed float32 elements in a to the number of fraction bits specified by imm, and stores the result using zeromask k (elements are zeroed out when the corresponding mask bit is not set).



_mm_roundscale_round_sd

extern __m128d __cdecl _mm_roundscale_round_sd(__m128d a, __m128d b, const int imm, const int round);

Rounds the lower float64 element in a to the number of fraction bits specified by imm, stores the result in the lower destination element, and copies the upper element from b to the upper destination element.



_mm_mask_roundscale_round_sd

extern __m128d __cdecl _mm_mask_roundscale_round_sd(__m128d src, __mmask8 k, __m128d a, __m128d b, const int imm, const int round);

Rounds the lower float64 element in a to the number of fraction bits specified by imm, stores the result in the lower destination element using writemask k (the element is copied from src when mask bit 0 is not set), and copies the upper element from b to the upper destination element.



_mm_maskz_roundscale_round_sd

extern __m128d __cdecl _mm_maskz_roundscale_round_sd(__mmask8 k, __m128d a, __m128d b, const int imm, const int round);

Rounds the lower float64 element in a to the number of fraction bits specified by imm, stores the result in the lower destination element using zeromask k (the element is zeroed out when mask bit 0 is not set), and copies the upper element from b to the upper destination element.



_mm_roundscale_sd

extern __m128d __cdecl _mm_roundscale_sd(__m128d a, __m128d b, const int imm);

Rounds the lower float64 element in a to the number of fraction bits specified by imm, stores the result in the lower destination element, and copies the upper element from b to the upper destination element.



_mm_mask_roundscale_sd

extern __m128d __cdecl _mm_mask_roundscale_sd(__m128d old, __mmask8 k, __m128d a, __m128d b, const int imm);

Rounds the lower float64 element in a to the number of fraction bits specified by imm, stores the result in the lower destination element using writemask k (the element is copied from src when mask bit 0 is not set), and copies the upper element from b to the upper destination element.



_mm_maskz_roundscale_sd

extern __m128d __cdecl _mm_maskz_roundscale_sd(__mmask8 k, __m128d a, __m128d b, const int imm);

Rounds the lower float64 element in a to the number of fraction bits specified by imm, stores the result in the lower destination element using zeromask k (the element is zeroed out when mask bit 0 is not set), and copies the upper element from b to the upper destination element.



_mm_roundscale_round_ss

extern __m128 __cdecl _mm_roundscale_round_ss(__m128 a, __m128 b, const int imm, const int round);

Rounds the lower float32 element in a to the number of fraction bits specified by imm, stores the result in the lower destination element, and copies the upper three packed elements from b to the upper destination elements.



_mm_mask_roundscale_round_ss

extern __m128 __cdecl _mm_mask_roundscale_round_ss(__m128 src, __mmask8 k, __m128 a, __m128 b, const int imm, const int round);

Rounds the lower float32 element in a to the number of fraction bits specified by imm, stores the result in the lower destination element using writemask k (the element is copied from src when mask bit 0 is not set), and copies the upper three packed elements from b to the upper destination elements.



_mm_maskz_roundscale_round_ss

extern __m128 __cdecl _mm_maskz_roundscale_round_ss(__mmask8 k, __m128 a, __m128 b, const int imm, const int round);

Rounds the lower float32 element in a to the number of fraction bits specified by imm, stores the result in the lower destination element using zeromask k (the element is zeroed out when mask bit 0 is not set), and copies the upper three packed elements from b to the upper destination elements.



_mm_roundscale_ss

extern __m128 __cdecl _mm_roundscale_ss(__m128 a, __m128 b, const int imm);

Rounds the lower float32 element in a to the number of fraction bits specified by imm, stores the result in the lower destination element, and copies the upper three packed elements from b to the upper destination elements.



_mm_mask_roundscale_ss

extern __m128 __cdecl _mm_mask_roundscale_ss(__m128 src, __mmask8 k, __m128 a, __m128 b, const int imm);

Rounds the lower float32 element in a to the number of fraction bits specified by imm, stores the result in the lower destination element using writemask k (the element is copied from src when mask bit 0 is not set), and copies the upper three packed elements from b to the upper destination elements.



_mm_maskz_roundscale_ss

extern __m128 __cdecl _mm_maskz_roundscale_ss(__mmask8 k, __m128 a, __m128 b, const int imm);

Rounds the lower float32 element in a to the number of fraction bits specified by imm, stores the result in the lower destination element using zeromask k (the element is zeroed out when mask bit 0 is not set), and copies the upper three packed elements from b to the upper destination elements.



_mm512_scalef_pd

extern __m512d __cdecl  _mm512_scalef_pd(__m512d a, __m512d b);

Performs a floating point scale of the packed float64 elements in source a by multiplying by 2b, and stores the result.



_mm512_mask_scalef_pd

extern __m512d __cdecl _mm512_mask_scalef_pd(__m512d src, __mmask8 k, __m512d a, __m512d b);

Performs a floating point scale of the packed float64 elements in source a by multiplying by 2b, and stores the result using writemask k (elements are copied from src when the corresponding mask bit is not set).



_mm512_maskz_scalef_pd

extern __m512d __cdecl _mm512_maskz_scalef_pd(__mmask8 k, __m512d a, __m512d b);

Performs a floating point scale of the packed float64 elements in source a by multiplying by 2b, and stores the result using zeromask k (elements are zeroed out when the corresponding mask bit is not set).



_mm512_scalef_round_pd

extern __m512d __cdecl _mm512_scalef_round_pd(__m512d a, __m512d b, int round);

Performs a floating point scale of the rounded packed float64 elements in source a by multiplying by 2b, and stores the result.



_mm512_mask_scalef_round_pd

extern __m512d __cdecl _mm512_mask_scalef_round_pd(__m512d src, __mmask8 k, __m512d a, __m512d b, int round);

Performs a floating point scale of the rounded packed float64 elements in source a by multiplying by 2b, and stores the result using writemask k (elements are copied from src when the corresponding mask bit is not set).



_mm512_maskz_scalef_round_pd

extern __m512d __cdecl _mm512_maskz_scalef_round_pd(__mmask8 k, __m512d a, __m512d b, int round);

Performs a floating point scale of the rounded packed float64 elements in source a by multiplying by 2b, and stores the result using zeromask k (elements are zeroed out when the corresponding mask bit is not set).



_mm512_scalef_ps

extern __m512 __cdecl _mm512_scalef_ps(__m512 a, __m512 b);

Performs a floating point scale of the packed float32 elements in source a by multiplying by 2b, and stores the result.



_mm512_mask_scalef_ps

extern __m512 __cdecl _mm512_mask_scalef_ps(__m512 src, __mmask16 k, __m512 a, __m512 b);

Performs a floating point scale of the packed float32 elements in source a by multiplying by 2b, and stores the result, using writemask k (elements are copied from src when the corresponding mask bit is not set).



_mm512_maskz_scalef_ps

extern __m512 __cdecl _mm512_maskz_scalef_ps(__mmask16 k, __m512 a, __m512 b);

Performs a floating point scale of the packed single-precision (64-bit) floating-point elements in source a by multiplying by 2b, and stores the result, using zeromask k (elements are zeroed out when the corresponding mask bit is not set).



_mm512_scalef_round_ps

extern __m512 __cdecl _mm512_scalef_round_ps(__m512 a, __m512 b, int round);

Performs a floating point scale of the rounded packed single-precision (64-bit) floating-point elements in source a by multiplying by 2b, and stores the results.



_mm512_mask_scalef_round_ps

extern __m512 __cdecl _mm512_mask_scalef_round_ps(__m512 src, __mmask16 k, __m512 a, __m512 b, int round);

Performs a floating point scale of the rounded packed single-precision (64-bit) floating-point elements in source a by multiplying by 2b, and stores the results, using writemask k (elements are copied from src when the corresponding mask bit is not set).



_mm512_maskz_scalef_round_ps

extern __m512 __cdecl _mm512_maskz_scalef_round_ps(__mmask16 k, __m512 a, __m512 b, int round);

Performs a floating point scale of the rounded packed single-precision (64-bit) floating-point elements in source a by multiplying by 2b, and stores the results, using zeromask k (elements are zeroed out when the corresponding mask bit is not set).



_mm_scalef_round_sd

extern __m128d __cdecl _mm_scalef_round_sd(__m128d a, __m128d b, int round);

Performs a floating point scale of the rounded scalar float64 elements in source a by multiplying by 2b, stores the result in the lower destination element, and copies the upper element from b to the upper destination element.



_mm_mask_scalef_round_sd

extern __m128d __cdecl _mm_mask_scalef_round_sd(__m128d src, __mmask8 k, __m128d a, __m128d b, int round);

Performs a floating point scale of the rounded scalar float64 elements in source a by multiplying by 2b, stores the result in the lower destination element using writemask k (the element is copied from src when mask bit 0 is not set), and copies the upper element from b to the upper destination element.



_mm_maskz_scalef_round_sd

extern __m128d __cdecl _mm_maskz_scalef_round_sd(__mmask8 k, __m128d a, __m128d b, int round);

Performs a floating point scale of the rounded scalar float64 elements in source a by multiplying by 2b, stores the result in the lower destination element using zeromask k (the element is zeroed out when mask bit 0 is not set), and copies the upper element from b to the upper destination element.



_mm_scalef_sd

extern __m128d __cdecl _mm_scalef_sd(__m128d a, __m128d b);

Performs a floating point scale of the scalar float64 elements in source a using values from b, stores the result in the lower destination element, and copies the upper element from b to the upper destination element.



_mm_mask_scalef_sd

extern __m128d __cdecl _mm_mask_scalef_sd(__m128d src, __mmask8 k, __m128d a, __m128d b);

Performs a floating point scale of the scalar float64 elements in source a using values from b, stores the result in the lower destination element using writemask k (the element is copied from src when mask bit 0 is not set), and copies the upper element from b to the upper destination element.



_mm_maskz_scalef_sd

extern __m128d __cdecl _mm_maskz_scalef_sd(__mmask8 k, __m128d a, __m128d b);

Performs a floating point scale of the scalar float64 elements in source a using values from b, stores the result in the lower destination element using zeromask k (the element is zeroed out when mask bit 0 is not set), and copies the upper element from b to the upper destination element.



_mm_scalef_round_ss

extern __m128 __cdecl _mm_scalef_round_ss(__m128 a, __m128 b, int  round);

Performs a floating point scale of the rounded scalar float32 elements in source a by multiplying by 2b, stores the result in the lower destination element, and copies the upper element from b to the upper destination elements.



_mm_mask_scalef_round_ss

extern __m128 __cdecl _mm_mask_scalef_round_ss(__m128 src, __mmask8 k, __m128 a, __m128 b, int  round);

Performs a floating point scale of the rounded scalar float32 elements in source a by multiplying by 2b, stores the result in the lower destination element using writemask k (the element is copied from src when mask bit 0 is not set), and copies the upper three packed elements from b to the upper destination elements.



_mm_maskz_scalef_round_ss

extern __m128 __cdecl _mm_maskz_scalef_round_ss(__mmask8 k, __m128 a, __m128 b, int  round);

Performs a floating point scale of the rounded scalar float32 elements in source a by multiplying by 2b, stores the result in the lower destination element using zeromask k (the element is zeroed out when mask bit 0 is not set), and copies the upper three packed elements from b to the upper destination elements.



_mm_scalef_ss

extern __m128 __cdecl _mm_scalef_ss(__m128 a, __m128 b);

Performs a floating point scale of the scalar float32 elements in source a using values from b, stores the result in the lower destination element, and copies the upper three packed elements from b to the upper destination elements.



_mm_mask_scalef_ss

extern __m128 __cdecl _mm_mask_scalef_ss(__m128 src, __mmask8 k, __m128 a, __m128 b);

Performs a floating point scale of the scalar float32 elements in source a using values from b, stores the result in the lower destination element, using writemask k (the element is copied from src when mask bit 0 is not set), and copies the upper three packed elements from b to the upper destination elements.



_mm_maskz_scalef_ss

extern __m128 __cdecl _mm_maskz_scalef_ss(__mmask8 k, __m128 a, __m128 b);

Performs a floating point scale of the scalar float32 elements in source a using values from b, stores the result in the lower destination element, using zeromask k (the element is zeroed out when mask bit 0 is not set), and copies the upper three packed elements from b to the upper destination elements.



Para obtener información más completa sobre las optimizaciones del compilador, consulte nuestro Aviso de optimización.