Reducing code size explosion through low-overhead specialization

The overhead of performing optimizations during execu-tion is the main hindrance in achieving good performance for many computation intensive applications. The runtime optimizations are required due to lack of precise informa-tion at static compile time. In this article, we describe a new optimization method, the hybrid specialization, able to overcome this problem to a large extent. Specialization proceeds by exposing values for a set of candidate parameters at static compile time. The next part of specialization is performed at runtime by a dynamic spe-cializer which is capable of reusing optimized code and spe-cializing binary code generated in the last step. The dy-namic specializer is generated automatically after validat-ing code against the required criteria. It can be used to adapt binary code to different values and therefore reduces the possible runtime code generation. Moreover, it incurs a small overhead for runtime specialization as compared to that incurred in existing dynamic code generators or spe-cializers. Initial results over Itanium-II architecture show improve-ment in performance for different benchmarks including SPEC CPU2000, FFTW3 and ATLAS.