1. High Performance Embedded Computing © 2007 Elsevier Lecture 18: Hardware/Software Codesign Embedded Computing Systems Mikko Lipasti, adapted from M. Schulte Based on slides and textbook from Wayne Wolf

2. © 2006 Elsevier Topics Platforms. Performance analysis. Design representations. Hardware/software partitioning. Co-synthesis for general multiprocessors. Optimization concepts Simulation

3. © 2006 Elsevier Design platforms Different levels of integration:  PC + board.  Custom board with CPU + FPGA or ASIC.  Platform FPGA.  System-on-chip.

4. © 2006 Elsevier CPU/accelerator architecture CPU is sometimes called host. Accelerator communicate via shared memory.  May use DMA to communicate. CPU memory accelerator

5. © 2006 Elsevier Example: Xilinx Virtex-4 System-on-chip:  FPGA fabric.  PowerPC.  On-chip RAM.  Specialized I/O devices. FPGA fabric is connected to PowerPC bus. MicroBlaze CPU can be added in FPGA fabric.

6. © 2006 Elsevier Example: WILDSTAR II Pro

7. © 2006 Elsevier Performance analysis Must analyze accelerator performance to determine system speedup. High-level synthesis helps:  Use as estimator for accelerator performance.  Use to implement accelerator.

8. © 2006 Elsevier Data path/controller architecture Data path performs regular operations, stores data in registers. Controller provides required sequencing. Data path controller

9. © 2006 Elsevier High-level synthesis High-level synthesis creates register-transfer description from behavioral description. Schedules and allocates:  Operators.  Variables.  Connections. Control step or time step is one cycle in system controller. Components may be selected from technology library.

10. © 2006 Elsevier Models Model as data flow graph. Critical path is set of nodes on path that determines schedule length.

11. © 2006 Elsevier Accelerator estimation How do we use high-level synthesis, etc. to estimate the performance of an accelerator? We have a behavioral description of the accelerator function. Need an estimate of the number of clock cycles. Need to evaluate a large number of candidate accelerator designs.  Can’t afford to synthesize them all.

12. © 2006 Elsevier Estimation methods Hermann et al. used numerical methods.  Estimated incremental costs due to adding blocks to the accelerator. Henkel and Ernst used path-based scheduling.  Cut CFDG into subgraphs: reduce loop iteration count; cut at large joins; divide into equal-sized pieces.  Schedule each subgraph independently. Vahid and Gajski estimate controller and data path costs incrementally.

13. © 2006 Elsevier Single- vs. multi-threaded One critical factor is available parallelism:  single-threaded/blocking: CPU waits for accelerator;  multithreaded/non-blocking: CPU continues to execute along with accelerator. To multithread, CPU must have useful work to do.  But software must also support multithreading.

14. © 2006 Elsevier Total execution time Single-threaded: Multi-threaded: P2 P1 A1 P3 P4 P2 P1 A1 P3 P4

15. © 2006 Elsevier Execution time analysis Single-threaded:  Count execution time of all component processes. Multi-threaded:  Find longest path through execution.

16. © 2006 Elsevier Hardware-software partitioning Partitioning methods usually allow more than one ASIC. Typically ignore CPU memory traffic in bus utilization estimates. Typically assume that CPU process blocks while waiting for ASIC. CPU ASIC mem

17. © 2006 Elsevier Synthesis tasks Scheduling: make sure that data is available when it is needed. Allocation: make sure that processes don’t compete for the PE. Partitioning: break operations into separate processes to increase parallelism, put serial operations in one process to reduce communication. Mapping: take PE, communication link characteristics into account.

18. © 2006 Elsevier Scheduling and allocation Must schedule/allocate  computation  communication Performance may vary greatly with allocation choice. P1 P2 P3 P1 P2 P3 CPU1 ASIC1

19. © 2006 Elsevier Problems in scheduling/allocation l Can multiple processes execute concurrently? l Is the performance granularity of available components fine enough to allow efficient search of the solution space? l Do computation and communication requirements conflict? l How accurately can we estimate performance?  software  custom ASICs

20. © 2006 Elsevier Partitioning example before after r = p1(a,b); s = p2(c,d); z = r + s; r=p1(a,b);s=p2(c,d); z = r + s

21. © 2006 Elsevier Problems in partitioning l At what level of granularity must partitioning be performed? l How well can you partition the system without an allocation? l How does communication overhead figure into partitioning?

22. © 2006 Elsevier Problems in mapping Mapping and allocation are strongly connected when the components vary widely in performance. Software performance depends on bus configuration as well as CPU type. Mappings of PEs and communication links are closely related.

23. © 2006 Elsevier Program representations CDFG: single-threaded, executable, can extract some parallelism. Task graph: task-level parallelism, no operator-level detail.  TGFF generates random task graphs. UNITY: based on parallel programming language.

24. © 2006 Elsevier Platform representations Technology table describes PE, channel characteristics.  CPU time.  Communication time.  Cost.  Power. Multiprocessor connectivity graph describes PEs, channels. TypeSpeedcost ARM 750E610 MIPS50E68 PE 1 PE 2 PE 3

25. © 2006 Elsevier Hardware/software partitioning assumptions CPU type is known.  Can determine software performance. Number of processing elements is known.  Simplifies system-level performance analysis. Only one processing element can multi-task.  Simplifies system-level performance analysis.

26. © 2006 Elsevier Two early HW/SW partitioning systems Vulcan:  Start with all tasks on accelerator.  Move tasks to CPU to reduce cost. COSYMA:  Start with all functions on CPU.  Move functions to accelerator to improve performance.

27. © 2006 Elsevier Additional Co-synthesis Approaches Vahid: Binary constraint search CoWare: communicating processes model Simulated annealing & Tabu search heuristics [Ele96] LYCOS: CDFG representation [Mad97] Several others in book (skim)

28. © 2006 Elsevier Multi-objective optimization Operations research provides notions for optimization functions with multiple objectives. Pareto optimality: optimal solution cannot be improved without making something else worse.

29. © 2006 Elsevier Large search space: Genetic algorithms Modeled as:  Genes = strings of symbols.  Mutations = changes to strings. Types of moves:  Reproduction makes a copy of a string.  Mutation changes a string.  Crossover interchanges parts of two strings.

30. © 2006 Elsevier Hardware/software co-simulation Must connect models with different models of computation, different time scales. Simulation backplane manages communication. Becker et al. used PLI in Verilog-XL to add C code that communicates with software models, UNIX networking to connect hardware simulator.

31. © 2006 Elsevier Mentor Graphics Seamless Hardware modules described using standard HDLs. Software can be loaded as C or binary. Bus interface module connects hardware models to processor instruction set simulator. Coherent memory server manages shared memory.

32. © 2006 Elsevier Summary Platforms. Performance analysis. Design representations. Hardware/software partitioning. Co-synthesis for general multiprocessors. Optimization concepts Simulation