1. Physical Design of FabScalar Generated Superscalar Processors EE6052 Class Project Wei Zhang

2. Outline  Heterogeneous Multi-Core Processors  FabScalar  Physical design of FabScalar generated cores  Things we can do  Conclusion 2Wei

3. Heterogeneous Multi-Core Processors Heterogeneous multi-Core processor  Contains multiple, differently-designed superscalar core types that can streamline the execution of diverse programs.  The core types differ from each other and target at different applications. Superscalar processor  Utilizes instruction level parallelism. Executes more than one instruction during a clock-cycle.  Dispatch instructions to redundant functional units such as ALU, multiplier, bit shifter, etc on the processor. “Achilles’ heel” of heterogeneous multi-core processor design  Design and verification effort is multiplied by the number of different core types, which limits the amount of architectural diversity that can be practically implemented. 3Wei

4. FabScalar Wei4  A toolset developed to automatically compose RTL designs of arbitrary cores within a canonical superscalar template.  Frames superscalar processors in a canonical template which defines canonical pipeline stages and interfaces among them.  A Canonical Pipeline Stage Library (CPSL) provides many implementations of each canonical pipeline stage that differs in their superscalar dimensions.  An RTL generation tool uses the template and CPSL to automatically generate an overall core of desired configuration.

5. Canonical Pipeline Stages Wei5  The pipeline stages and interfaces are the same for all the superscalar processors.  Each pipeline stage is composed from the Canonical Pipeline Stage Library.

6. Canonical Pipeline Stage Library Wei6

7. FabMem: A Multiported RAM and CAM Compiler Wei7  Estimates read/write delays, read/write energies, and areas of user-specified multi-ported RAMs/CAMs.  Generates layouts of desired RAMs/CAMs. Limitations  FabMem is tied to a specific technology (FreePDK45).  The largest supported RAM is 512 words.  FabMem can generate RAMs for only 2XR-XW and XR-XW configurations. The maximum number of read ports is 16. The maximum number of write ports is 8.  The degree of column muxing in RAMs is limited to 1, 2, and 4.  The largest supported CAM is 256 words.  FabMem can generate CAMs for only XR-XW configurations. The maximum number of read ports is 8. The maximum number of write ports is 8.

8. Physical Design – by FabScalar Group Wei8  Use FabMem to generate custom designs of critical memory structures.  Course-grained floorplanning.  Little consideration of power, area and performance issues.  Use FreePDK45, a 45nm based standard cell library, for logic synthesis and place-and-route.

9. Physical Design – Things We Can Do Wei9  Floorplanning  Data path design  Memory design – comparison between FabMem and other memory compilers  Cadence vs Synopsys  Power planning

10. Conclusion Wei10  Heterogeneous Multi-Core Processors have many advantages  FabScalar uses canonical pipeline stages for superscalar processor generation.  Physical design of FabScalar generated cores remains to be further investigated.