Filter

119083 Treffer
Documentation

_mm512_set1_epi32 von User and Reference Guide for the Intel® C++ Compiler 15.0

Sets elements to equal specified integer value. There is no corresponding instruction. This intrinsic only applies to Intel® Many Integrated Core Architecture (Intel® MIC Architecture).

Zuletzt aktualisiert am 04.07.2015 - 22:55
Documentation

_mm512_max_pd/ _mm512_mask_max_pd von User and Reference Guide for the Intel® C++ Compiler 15.0

Determines the maximum of float64 vectors. On Intel® MIC Architecture, compare and jump instructions emulate this instruction. This intrinsic only applies to Intel® Many Integrated Core Architecture (Intel® MIC Architecture).

Zuletzt aktualisiert am 04.07.2015 - 22:55
Documentation

_mm512_max_epi32/ _mm512_mask_max_epi32 von User and Reference Guide for the Intel® C++ Compiler 15.0

Determines the maximum of int32 vectors. Corresponding instruction is VPMAXSD. This intrinsic only applies to Intel® Many Integrated Core Architecture (Intel® MIC Architecture).

Zuletzt aktualisiert am 04.07.2015 - 22:55
Documentation

_mm256_movemask_epi8 von User and Reference Guide for the Intel® C++ Compiler 15.0

Moves byte mask from source vector. The corresponding Intel® AVX2 instruction is VPMOVMSKB.

Zuletzt aktualisiert am 04.07.2015 - 22:55
Documentation

_mm_cvtps_ph() von User and Reference Guide for the Intel® C++ Compiler 15.0

Converts four single-precision floating point values to half-precision (16-bit) floating point values. The corresponding 3rd Generation Intel® Core™ Processor extension instruction is VCVTPS2PH.

Zuletzt aktualisiert am 04.07.2015 - 22:55
Documentation

_mm256_max_epi8/16/32 von User and Reference Guide for the Intel® C++ Compiler 15.0

Determines the maximum value between two vectors with packed signed byte/word/doubleword integers. The corresponding Intel® AVX2 instruction is VPMAXSB, VPMAXSW, or VPMAXSD.

Zuletzt aktualisiert am 04.07.2015 - 22:55
Documentation

_mm256_mul_epu32 von User and Reference Guide for the Intel® C++ Compiler 15.0

Multiplies two vectors with packed doubleword unsigned integer values. The corresponding Intel® AVX2 instruction is VPMULUDQ.

Zuletzt aktualisiert am 04.07.2015 - 22:55
Einzelheiten zur Compiler-Optimierung finden Sie in unserem Optimierungshinweis.