vrsa_expf: Array single precision exponential function

• C Prototype:
  • void vrsa_expf (int n, float *x, float *y)

  • Inputs:

    • int	n	- the number of single precision values in both the input and output arrays.
      float	*x	- pointer to the array of input values.
      float	*y	- pointer to the array of output values.

  • Outputs:
    • e raised to the power x (exponential of x) for each x value, filled into the y array.
• Fortran Subroutine Interface:
  • SUBROUTINE VRSA_EXPF(N,X,Y)
  • Inputs:
    • INTEGER N - the number of values in both the input and output arrays.
    • REAL X(N) - array of single precision input values.
  • Outputs:
    • REAL Y(N) - array of exponentials (e raised to the power x) of input values.
• Notes:
  • vrsa_expf computes the single precision exponential function for each element of an array of input arguments.
    This routine accepts an array of single precision input values, computes the e^x for each input value, and stores the result in the array pointed to by the y pointer input. It is the responsibility of the calling program to allocate/deallocate enough storage for the output array. This is a relaxed version of exp, suitable for use with fastmath compiler flags or applications not requiring full error handling. Denormal inputs may produce unpredictable results. Special case inputs produce C99 return values. The routine is accurate to better than 1 ulp over the valid input range.
• Special case return values:

  • Input	Output
    QNaN	same QNaN
    SNaN	same NaN converted to QNaN
    < -87.5	0
    > 88	+infinity

• Performance:
  • 15 cycles per value for valid inputs, n = 24.