A.5.6 ADDSD: ADD Scalar Double-Precision FP Values
ADDSD xmm1,xmm2/mem64 ; F2 0F 58 /r [KATMAI,SSE]
ADDSD adds the low double-precision FP values from the source and
destination operands and stores the double-precision FP result in the
destination operand.
dst[0-63] := dst[0-63] + src[0-63],
dst[64-127) remains unchanged.
The destination is an XMM register. The source operand can be either an
XMM register or a 64-bit memory location.