How does this function compute the absolute value of a float through a NOT and AND operation?

Question

I am trying to understand how the following code snippet works. This program uses SIMD vector instructions (Intel SSE) to calculate the absolute value of 4 floats (so, basic

Answer 1

-0.0 is represented as 1000...000¹. Therefore _mm_andnot_ps(-0.0, x)² is equivalent to 0111...111 & x. This forces the MSB (which is the sign bit) to 0.

---

_{1. In IEEE-754, at least.}

_{2. The _mm_andnot_ps intrinsic does not mean "NAND"; see e.g. http://msdn.microsoft.com/en-us/library/68h7wd02(v=vs.90).aspx.}