Consider the data hazard in the following program. Assume that there is no structure hazard, and assume that the memory access time is very short (<1 cycle).

instruction 1: addi r2, r0, 0x100 instruction 2: ldw r3, 4(r2) instruction 3: addi r4, r3, 4 instruction 4: ldw r5, 8(r2) instruction 5: addi r6, r5, 4

Please find out how long it takes to run the program if we just use pipeline stalling.

Stage 5 of instruction 1 is executed in cycle 5

Stage 5 of instruction 2 is executed in cycle( )

Stage 5 of instruction 3 is executed in cycle ( )

Stage 5 of instruction 4 is executed in cycle ( )

Stage 5 of instruction 5 is executed in cycle ( )

Please find out how long it takes to run the same program if we just use operand forwarding.

Stage 5 of instruction 1 is executed in cycle 5

Stage 5 of instruction 2 is executed in cycle ( )

Stage 5 of instruction 3 is executed in cycle ( )

Stage 5 of instruction 4 is executed in cycle ( )

Stage 5 of instruction 5 is executed in cycle ( )

Finally, we can reduce the total running time of the above program if we use both operand forwarding and instruction reordering.

Instruction ( ) and instruction( ) should be swapped.

With operand forwarding and instruction reordering, stage 5 of instruction 5 is executed in cycle ( )