Consider a version of the pipeline from Section 4.5 in the book that does not handle data hazards (i.e. the programmer is responsible for addressing data hazards by inserting NOPs where necessary). Supposed that after optimization a typical n-instruction program requires an additional 0.4*n NOPs to correctly handle data hazards.

4.5 in the book

a. Supposed that the cycle time of this pipeline without forwarding is 250 ps. Supposed also that adding forwarding hardware will reduce the number of NOPs from 0.4*n to 0.05*n, but will increase the cycle time to 300ps. What is the speedup of this new pipeline compared to the one without forwarding?

b. Different programs will require different amounts of NOPs. How many NOPs (as percentage of all instructions in a program) can remain in the typical program before that program runs slower on the pipeline with forwarding?

c. Can a program with only 0.075*n NOPs possible run faster on the pipeline with forwarding? Explain why or why not.

d. At a minimum, how many NOPs (as a percentage of code instructions) must a program have before it can possibly run faster on the pipeline with forwarding?

Program execution Time order (in instructions) 200 400 600 800 1000 1200 1400 Instruction RegALData access Instruction DataReg ALU cess 200 ps fetch LDUR X3, [X0,#400] 200 ps I Instruction fetch Reg ALUData access Reg Program execution Time order (in instructions) 200 400 600 800 1000 1200 1400 Instruction RegALData access Instruction DataReg ALU cess 200 ps fetch LDUR X3, [X0,#400] 200 ps I Instruction fetch Reg ALUData access Reg