An Applied Study of a User-Level Multi-Objective Dynamic Binary Translator Based on LLVM

With the diversification of computing devices, different hardware platforms and instruction set architectures (ISAs) coexist, which requires software to be able to run on different hardware platforms in order to adapt to the changing technological environment. Therefore, this paper presents an applied study of a user-level multi-objective dynamic binary translator based on LLVM. First, the execution efficiency of programs on target platforms is improved through dynamic translation and optimization, so that programs compiled for one architecture can run on different architectures. Second, an intermediate representation of LLVM is utilized as an intermediate language in the translation process to support multiple source and target ISAs. The translated code is optimized using the optimization framework of LLVM. Finally, different ISAs are abstracted and code is produced to accommodate different target architectures. As a result of the above study, the execution efficiency of the translated code can be improved without sacrificing compatibility by utilizing the optimization framework of LLVM. It can provide new perspectives and methods for the development of dynamic binary translation technology, support cross-platform operation and optimization of software, and also contribute to the development of related compilation technology, virtualization technology and other fields.