Intrinsics for Square Root and Cube Root Operations

Many routines in the svml library are more highly optimized for Intel® microprocessors than for non-Intel microprocessors.

• _mm_sqrt_pd, _mm256_sqrt_pd
  Calculates square root value. Vector variant of sqrt(x) function for a 128-bit/256-bit vector argument of float64 values.
• _mm_sqrt_ps, _mm256_sqrt_ps
  Calculates square root value. Vector variant of sqrt(x) function for a 128-bit/256-bit vector argument of float32 values.
• _mm_invsqrt_pd, _mm256_invsqrt_pd
  Calculates inverse square root value. Vector variant of invsqrt(x) function for a 128-bit/256-bit vector argument of float64 values.
• _mm_invsqrt_ps, _mm256_invsqrt_ps
  Calculates inverse square root value. Vector variant of invsqrt(x) function for a 128-bit/256-bit vector argument of float32 values.
• _mm_cbrt_pd, _mm256_cbrt_pd
  Calculates cube root value. Vector variant of cbrt(x) function for a 128-bit/256-bit vector argument of float64 values.
• _mm_cbrt_ps, _mm256_cbrt_ps
  Calculates cube root value. Vector variant of cbrt(x) function for a 128-bit/256-bit vector argument of float32 values.
• _mm_invcbrt_pd, _mm256_invcbrt_pd
  Calculates inverse cube root value. Vector variant of invcbrt(x) function for a 128-bit/256-bit vector argument of float64 values.
• _mm_invcbrt_ps, _mm256_invcbrt_ps
  Calculates inverse cube root value. Vector variant of invcbrt(x) function for a 128-bit/256-bit vector argument of float32 values.
• _mm_csqrt_ps, _mm256_csqrt_ps
  Calculates complex square root value. Vector variant of csqrt(x) function for a 128-bit/256-bit vector argument of _Complex float32 values.

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