A. Assume a dual issue in-order, 5-stage pipelined microprocessor. This machine is identical to the one we covered in class, except that t can have two instructions proceed through every stage in the pipeline. The ideal CPI of this machine is 0.5. Now this pipelined architecture suffers performance penalty from four distinct sources: . Cache Miss: this affects load and store instructions (instructions needing memory access). Assume that this takes 150 processor cycles, so 150 bubbles are inserted for every such cache mis:s Load-use hazard: this affects dependent instructions behind a load. Assume a single cycle penalty for Branch mis-predictions. This refers to the penalty from branch mis-predictions, and assume a 2 cycle Return instructions: This refers to the penalty from return instructions, and assume a 3 cycle penalty. these. penalty. Now consider a program composed of 10 billion instructions. Among all instructions, a fraction are loads, fractions are stores, fractions are branches, and fractions are return instructions. Among load in- structions, assume Y fraction misses in the cache, while n fraction incur load-use penalty. Among store instructions, Z fraction misses in the cache. Furthermore, the branch-misprediction rate is X. Calculate the CPI of the above described machine while executing this program. A. Assume a dual issue in-order, 5-stage pipelined microprocessor. This machine is identical to the one we covered in class, except that t can have two instructions proceed through every stage in the pipeline. The ideal CPI of this machine is 0.5. Now this pipelined architecture suffers performance penalty from four distinct sources: . Cache Miss: this affects load and store instructions (instructions needing memory access). Assume that this takes 150 processor cycles, so 150 bubbles are inserted for every such cache mis:s Load-use hazard: this affects dependent instructions behind a load. Assume a single cycle penalty for Branch mis-predictions. This refers to the penalty from branch mis-predictions, and assume a 2 cycle Return instructions: This refers to the penalty from return instructions, and assume a 3 cycle penalty. these. penalty. Now consider a program composed of 10 billion instructions. Among all instructions, a fraction are loads, fractions are stores, fractions are branches, and fractions are return instructions. Among load in- structions, assume Y fraction misses in the cache, while n fraction incur load-use penalty. Among store instructions, Z fraction misses in the cache. Furthermore, the branch-misprediction rate is X. Calculate the CPI of the above described machine while executing this program