Value profiling
Brad Calder, Peter Feller, Alan Eustace · 1997
Identifying variables as invariant or constant at compile-time allows the compiler to perform optimizations including constant folding, code specialization, and partial evaluation. Some variables, which cannot be labeled as constants, may exhibit behavior. A semi-invariant variable is one that cannot be identified as a constant at compile-time, but has a high degree of invariant behavior at run-time. If run-time information was available to identify these variables as semi-invariant, they could then benefit from invariant-based compiler optimizations. In this paper we examine the invariance found from profiling instruction values, and show that many instructions have values even across different inputs. We also investigate the ability to estimate the invariance for all instructions in a program from only profiling load instructions. In addition, we propose a new type of profiling called Convergent Profiling. Estimating the invariance from loads and convergent profiling are used to reduce the profiling time needed to generate an accurate value profile. The value profile can then be used to automatically guide code generation for dynamic compilation, adaptive execution, code specialization, partial evaluation and other compiler optimizations.
Cited by 181 works. doi:10.5555/266800.266825
Authors
References
- ATOM
- Value locality and load value prediction
- A general approach for run-time specialization and its application to C
- Exceeding the Dataflow Limit via Value Prediction
- Optimizing ML with run-time code generation
- C: a language for high-level, efficient, and machine-independent dynamic code generation
- Fast, effective dynamic compilation
- Dynamic memory disambiguation using the memory conflict buffer
- Data specialization
- Initial Results for Glacial Variable Analysis