Lecture 13

More instruction-level parallelism. Multiple issue and out-of-order execution.

Lecture

Slides (PDF, PPTX).

Outline

• Multiple issue processors
• Dynamic and static scheduling
• Out-of-order execution

Examples

CPU Model	Microarchitecture	Generation	Issue Width
Core i7-3615QM	Ivy Bridge	3rd Gen	4
Core i7-8665U	Whiskey Lake	8th Gen	4
Core i7-1260P	Golden Cove (P-core)	12th Gen	6
	Gracemont (E-core)		5
Core i7-13700	Raptor Cove (P-core)	13th Gen	6
	Gracemont (E-core)		5

Workshop

Outline

• Experimenting with a 6-stage dual-issue RISC-V processor (use Ripes simulator)
• Experimenting with branch prediction (use RARS simulator)

Dual-Issue RISC-V CPU (Ripes Simulator)

Examples

Runs the add_scalar.s example and see how many CPU clock cycles it uses.

Branch History Table (RARS Simulator)

Run programs from lectures 4-7 in RARS simulator with the “Branch History Table” plugin connected. See how well it can predict branch outcomes with different settings.

Tasks

1. Optimize the add_scalar.s program to make it waste less CPU cycles. Use the loop-unrolling technique (two or more loop iterations merged). How many cycles are used now?

2. Write an optimized version of the PlusMinus program, which solves the issue of incorrect branch prediction with loop unrolling (“even” and “odd” operations must be done at the same loop iteration).

References

• Parallelism via Instructions. Section 4.10 in [CODR].
• Advanced Microarchitecture. Section 7.7 in [DDCA].
• Instruction-Level Parallelism and Its Exploitation. Chapter 3 in [CAQA] (Advanced).
• Instruction-Level Parallelism and Superscalar Processors. Chapter 18 in [COA].
• Superscalar processor (Wikipedia).
• Out-of-order execution (Wikipedia).
• Register renaming (Wikipedia).
• Branch predictor (Wikipedia).