x86 Instruction Set Reference

Performs an SIMD subtract of the two packed double-precision floating-point values in the source operand (second operand) from the two packed double-precision floating-point values in the destination operand (first operand), and stores the packed double-precision floating-point results in the destination operand. The source operand can be an XMM register or a 128-bit memory location. The destination operand is an XMM register. See Figure 11-3 in the IA-32 Intel Architecture Software Developer's Manual, Volume 1 for an illustration of an SIMD double-precision floating-point operation.

Destination[0..63] = Destination[0..63] - Source[0..63];
Destination[64..127] = Destination[64..127] - Source[64..127];