html5-img
1 / 18

Data value prediction

Data value prediction . Bas van der Tol. Limits to ILP. Instruction Level Parallelism is limited by Control flow Data flow: true dependencies. Types of Speculative Execution. Sources of predicatable data. Data redundancy Error-checking Program constants Virtual function calls Glue code

levana
Télécharger la présentation

Data value prediction

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Data value prediction Bas van der Tol

  2. Limits to ILP Instruction Level Parallelism is limited by • Control flow • Data flow: true dependencies

  3. Types of Speculative Execution

  4. Sources of predicatable data • Data redundancy • Error-checking • Program constants • Virtual function calls • Glue code • Call-subgraph identities • Register spill code

  5. Register value locality

  6. Value Prediction Unit

  7. VPT Hit rate sensitivity to Size

  8. Example use of Value Prediction

  9. Penalties Penalties • Misprediction penalty • Structural hazards, both on correct and mispredictions

  10. Configurations used for experiments

  11. PowerPC 620 Speedup

  12. PowerPC 620+ Speedup

  13. Infinite Machine Model Speedup

  14. Data cache vs. Value Prediction

  15. Improving Prediction Accuracy • Last Value Prediction • Stride Prediction • Finite Context Method Predictors (fcm)

  16. Finite Context Models

  17. Prediction Success

  18. Conclusions • Data value prediction increases performance by 5% on PowerPC 620 • A performance gain of 23% is possible Future Developments • More parallel execution units • Better prediction models

More Related