Double precision (64 bit) shift operations using a 32 bit data path

A method for performing a double precision shift operation, comprising the sequential steps of:

• a. setting a shift amount to less than or equal to 31;
• b. examining an operand, wherein if the operand is 16 bits in length, then perform step c, otherwise jump to step e;
• c. examining the shift amount, wherein if the shift amount is greater than 16, then perform step d, otherwise jump to step e;
• d. rotating the operand and setting the status flags, then ending the routine;
• e. rotating the lower half of the operand;
• f. shifting the lower half of the operand;
• g. taking the exclusive OR of steps e and f;
• h. examining the shift amount, wherein if the shift amount is zero, then exit the instruction, otherwise, go to step i;
• i. shifting the upper half of the operand;
• and
• j. combining the operations performed on both the upper and lower halfs of the operand.