vrs8_log2f: Eight-valued single precision base-2 logarithm

• Prototype:
  • __m128,__m128 __vrs8_log2f(__m128 x1, __m128 x2);
    Note that this function uses a non-standard programming interface. The two __m128 inputs, which contain eight single precision values, are passed by the AMD64 C ABI in registers xmm0, and xmm1. The corresponding results are returned in xmm0 and xmm1. The use of xmm1 to return a __m128 is non-standard, and this function can not be called directly from C. It can be called directly from assembly language. It is intended for internal use by vectorizing compilers, that may be able to take advantage of the non-standard calling interface.
  • Inputs:
    • __m128 x1 - the first set of four single precision input values.
    • __m128 x2 - the second set of four single precision input values.
  • Outputs:
    • The base-2 logarithm of x.
    • __m128 y1 - the first set of four single precision base-2 logarithm results, returned in xmm0.
    • __m128 y2 - the second set of four single precision base-2 logarithm results, returned in xmm1.
• Notes:
  • __vrs8_log2f computes the base-2 logarithm for each of eight input arguments.
    
    This routine accepts eight single precision input values passed as two __m128 values. The result is the single precision base-2 logarithm of the eight values, returned as two __m128 values. This is a relaxed version of log2f, 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
    +/- 0	-infinity
    negative	QNaN
    QNaN	same QNaN
    SNaN	same NaN converted to QNaN
    +infinity	+infinity
    -infinity	QNaN

• Performance:
  • 203 cycles for most valid inputs (25 cycles per value), longer for input values very close to 1.0.