Ningning Shi
Engineer, Institute of Software, Chinese Academy of Sciences
Engaged in compiler/interpreter related R&D work for a long time, and has a certain understanding of LLVM, OpenJDK, Ark compiler and Android Runtime. Passionate about sharing and technical communication, he has participated in several open source community activities, and has shared more than 20 technical presentations in public. In addition, he is also committed to writing books, and has published technical books ‘The Beauty of Ark Compiler’ and ‘Android Runtime Source Code Analysis’.
Topic
LLVM code size optimisation
Introduce the current research progress of code volume optimisation, analyse LLVM's work on code volume optimisation, and discuss the future improvement direction of LLVM's code volume optimisation. Outline: Firstly, introduce the current research on code volume optimization in the field of compilation technology, and analyze and introduce the key research directions and papers; secondly, briefly analyze the current code volume optimization of LLVM, and analyze and evaluate the compression instructions under the RISC-V platform; lastly, make some discussions on the optimization space of the subsequent code volume optimization of LLVM, as well as future optimization improvement directions. Finally, we discuss the optimisation space of LLVM and the future direction of optimisation improvement.
