Continuous program optimization: Design and evaluation

IEEE Transactions on Computers

Volumn 50, Issue 6, 2001, Pages 549-567

  Kistler, Thomas ^a,b   Franz, Michael ^a,c  

^a IEEE   (United States)

^b Transmeta Corporation   (United States)

^c UNIVERSITY OF CALIFORNIA   (United States)

Author keywords

Continuous optimization; Dynamic compilation; Memory optimization; Profiling; Trace scheduling

Indexed keywords

MEMORY OPTIMIZATION; TRACE SCHEDULING;

CACHE MEMORY; CODES (SYMBOLS); DATA STRUCTURES; OPTIMIZATION; PROGRAM COMPILERS; SCHEDULING;

PARALLEL PROCESSING SYSTEMS;

EID: 0035365635     PISSN: 00189340     EISSN: None     Source Type: Journal    
DOI: 10.1109/12.931893     Document Type: Article

References (69)

1. Efficient implementation of the smalltalk-80 system
2. Reconciling responsiveness with performance in pure object-oriented languages
3. Fast, effective code generation in a just-in-time java compiler
4. System support for automatic profiling and optimization
5. DIGITAL FX!32: Combining emulation and binary translation
6. Oberon system 3: Vision of a future software technology
7. Oberon with gadgets: A simple component framework
8. Slim binaries
9. Optimally profiling and tracing programs
10. Edge profiling versus path profiling: The showdown
11. Computing the similarity of profiling data-heuristics for guiding adaptive compilation
12. Efficiently computing static single assignment form and the control dependence graph
13. Automated data-member layout of heap objects to improve memory-hierarchy performance
14. Design and evaluation of a compiler algorithm for prefetching
15. A data locality optimizing algorithm
16. The programming language oberon
17. Cache-conscious structure definition
18. Procedure placement using temporal ordering information
19. An efficient heuristic procedure for partitioning graphs
20. New faster kernighan-lin-type 'graph-partitioning algroithms
21. A fast and high quality multilevel scheme for partitioning irregular graphs
22. Profile guided code positioning
23. Using profile information to assist classic code optimizations
24. Profile-guided automatic inline expansion for C programs
25. Profile-assisted instruction scheduling
26. Comparing static and dynamic code scheduling for multiple-instruction-issue processors
27. Trace scheduling: A technique for global microcode compaction
28. Using profile information to assist advanced compiler optimization and scheduling
29. Better global scheduling using path profiles
30. Instruction scheduling for the IBM RISC system/6000 processor
31. Improving cache behavior of dynamically allocated data structures
32. Supporting dynamic data structures on distributed-memory machines
33. An extended set of fortran basic linear algebra subprograms: Model implementation and test programs
34. Continuous profiling: Where have all the cycles gone?
35. An efficient software-hardware collaborative profiling technique for wide-issue processors
36. Jalapeño-a compiler-supported java virtual machine for servers
37. Implementing jalapeño in java
38. Practicing JUDO: Java under dynamic optimizations
39. Optimizing dynamically-dispatched calls with run-time type feedback
40. Towards better inlining decisions using inlining trials
41. Optimizing dynamically-typed object-oriented languages with polymorphic inline caches
42. Customization: Optimizing compiler technology for SELF, a dynamically-typed object-oriented programming language
43. Escape analysis for java
44. PA-RISC to IA-64: Transparent execution, no recompilation
45. DAISY: Dynamic compilation for 100% architectural compatibility
46. Dynamic and transparent binary translation
47. C: A language for high-level, efficient, and machine-independent dynamic code generation
48. Optimizing ML with run-time code generation
49. Efficient incremental runtime specialization for free
50. An evaluatioin of staged run-time optimization in DyC
51. Using generational garbage collection to implement cache-conscious data placement
52. Cache-conscious structure layout
53. Cache-conscious data placement
54. Speculative hedge: Regulating compile-time speculation against profile variations
55. Profile-driven instruction level parallel scheduling with applications to super blocks