This instruction computes a 8×8-bit multiply with the lower byte of the first operand (Imm16 or si4, which can be a register or an immediate value), and the higher byte of the second operand (snd), giving a 16-bit result (the largest possible size for YASEP16).