Better exploration of region-level value locality with integrated computation reuse and value prediction

Conference Proceedings - Annual International Symposium on Computer Architecture, ISCA

Volumn , Issue , 2001, Pages 98-108

  Wu, Y ^a   Chen, D Y ^a   Fang, J ^a  

^a INTEL CORPORATION   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

COMPUTER HARDWARE; CRITICAL PATH ANALYSIS; DATA REDUCTION; MICROPROCESSOR CHIPS;

COMPUTATION REUSE; MULTIREADING SCHEME; VALUE PREDICTION;

COMPUTATIONAL METHODS;

EID: 0034838844     PISSN: 08847495     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1145/379240.379255     Document Type: Conference Paper

References (30)

1. A dynamic multithreading processor
2. DIVA: A reliable substrate for deep submicron microarchitecture design
3. Value profiling
4. Efficient checker processor design
5. Compiler-directed dynamic computation reuse: Rationale and initial results
6. Hardware support for dynamic activation of compiler-directed computation reuse
7. Block-level prediction for wide-issue superscalar processors
8. Value speculation scheduling for high performance processors
9. Using value prediction to increase the power of speculative execution hardware
10. Extending value reuse to basic blocks with compiler support
11. Exploring sub-block value reuse for superscalar processors
12. A chip-multiprocessor architecture with speculative multithreading
13. On the value locality of store instructions
14. Value locality and load value prediction
15. Exceeding the dataflow limit via value prediction
16. Tolerating memory latency through software-controlled pre-execution in simultaneous multithreading processors
17. Value prediction for speculative multithreaded architectures
18. Value prediction in VLIW machines
19. A study of slipstream processors
20. The predictability of data values
21. Performance tradeoffs in multistreamed supescalar architectures
22. Dynamic instruction reuse
23. An empirical analysis of instruction repetition
24. Understanding the difference between value prediction and instruction reuse
25. Multiscalar processors
26. Slipstream processors: Improving both performance and fault tolerance
27. Simultaneous multithreading: Maximizing on-chip parallelism
28. Storageless value prediction using prior register values
29. Highly accurate data value prediction using hybrid predictors