A Research On Static-integrated Optimization Framework For Dynamic Binary Translation2

Binary translation is the process of converting binary code from one Instruction Set Architecture (ISA) to another ISA without the presence of source code. In dynamic binary translation (DBT), translation is done at run-time, and optimization algorithms are applied on the translated code at the same time. DBT is especially useful in solving the problem of legacy code, code migration and building a virtual computing environment. It has become a hot spot in recent research.The performance is important for DBT system. In traditional DBT system, the dynamic optimization technology which is based on profiling is extensively used to improve the run-time performance. Because all the translation and optimization are done at run-time, the high overheads of profiling and optimizing limit their efficiencies. How to balance overheads and efficiencies is a hard work for DBT researchers.In this article, a static optimization framework for DBT (Sinof) is presented to conceal the overheads of profiling and optimizing, and static optimization algorithms are accomplished to deeply improve target code. The benefit of Sinof is from 30% to 50% in average. The main workflow of Sinof is: Once the source image has been executed, the profile information and target code will be stored, which are available at next runs. Moreover, optimizations will be implemented offline.The innovations and significances of this article include: 1) Presented a static optimization framework for DBT to screen the overheads of profiling and optimizing; 2) Designed the edges interaction profiling (interedge-profiling) to support the offline optimization; 3) Depending on profile information, identified hot traces and improve the hot code layout; 4) Completed the code blocks linking during static optimization , reduced the costs of context switch and so on; 5) Recorded the relocation information for code blocks to make them relocatable at next runs.