vrs4_sinf: Four-valued single precision Sine

• C Prototype:
  • __m128 __vrs4_sinf(__m128 x);

  • Inputs:
    • __m128 x - the four single precision inputs.
  • Outputs:
    • __m128 y - the four single precision Sine results, returned in xmm0.
• Notes:
  • __vrs4_sinf computes the Sine function of four input arguments.
    
    This routine accepts four single precision input values passed as a __m128 value. The result is the single precision Sine of the four values, returned as a __m128 value. This is a relaxed version of sinf, 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. This routine may return slightly worse than 1 ulp for very large values between 4e5 and 5e5.
• Special case return values:

  • Input	Output
    QNaN	same QNaN
    SNaN	same NaN converted to QNaN
    +infinity	QNaN
    -infinity	QNaN

• Performance:
  • 176 cycles for most valid inputs < 5e5 (44 cycles per value).