What are the x86-64 instructions for floating-point comparisons like the
cmpq for quard-word integer comparison?
The SSE3 extenstions provide two instructions for comparing ﬂoating values:
Instruction Based on Description ucomiss S2, S1 S1 - S2 Compare single precision ucomisd S2, S1 S1 - S2 Compare double precision
As with cmpq, they follow the GAS convention of listing the operands in reverse order. Argument S2 must be in an XMM register, while S1 can either be in an XMM register or in memory.