Intrinsics for Arithmetic Operations

• _mm256_add_pd
  Adds float64 vectors. The corresponding Intel® AVX instruction is VADDPD.
• _mm256_add_ps
  Adds float32 vectors. The corresponding Intel® AVX instruction is VADDPS.
• _mm256_addsub_pd
  Adds odd float64 elements and subtracts even float64 elements of vectors. The corresponding Intel® AVX instruction is VADDSUBPD.
• _mm256_addsub_ps
  Adds odd float32 elements and subtracts even float32 elements of vectors. The corresponding Intel® AVX instruction is VADDSUBPS.
• _mm256_hadd_pd
  Adds horizontal pairs of float64 elements of two vectors. The corresponding Intel® AVX instruction is VHADDPD.
• _mm256_hadd_ps
  Adds horizontal pairs of float32 elements of two vectors. The corresponding Intel® AVX instruction is VHADDPS.
• _mm256_sub_pd
  Subtracts float64 vectors. The corresponding Intel® AVX instruction is VSUBPD.
• _mm256_sub_ps
  Subtracts float32 vectors. The corresponding Intel® AVX instruction is VSUBPS.
• _mm256_hsub_pd
  Subtracts horizontal pairs of float64 elements of two vectors. The corresponding Intel® AVX instruction is VHSUBPD.
• _mm256_hsub_ps
  Subtracts horizontal pairs of float32 elements of two vectors. The corresponding Intel® AVX instruction is VHSUBPS.
• _mm256_mul_pd
  Multiplies float64 vectors. The corresponding Intel® AVX instruction is VMULPD.
• _mm256_mul_ps
  Multiplies float32 vectors. The corresponding Intel® AVX instruction is VMULPS.
• _mm256_div_pd
  Divides float64 vectors. The corresponding Intel® AVX instruction is VDIVPD.
• _mm256_div_ps
  Divides float32 vectors. The corresponding Intel® AVX instruction is VDIVPS.
• _mm256_dp_ps
  Calculates the dot product of float32 vectors. The corresponding Intel® AVX instruction is VDPPS.
• _mm256_sqrt_pd
  Computes the square root of double-precision floating point values. The corresponding Intel® AVX instruction is VSQRTPD.
• _mm256_sqrt_ps
  Computes the square root of single-precision floating point values. The corresponding Intel® AVX instruction is VSQRTPS.
• _mm256_rsqrt_ps
  Computes approximate reciprocals of square roots of float32 values.The corresponding Intel® AVX instruction is VRSQRTPS.
• _mm256_rcp_ps
  Computes approximate reciprocals of float32 values.The corresponding Intel® AVX instruction is VRCPPS.

Copyright © 1996-2011, Intel Corporation. All rights reserved.