Xiao-Feng Li

Some updates:

• (2008-08-01) I presented one of the Harmony GC algorithms: A Fully Parallel LISP2 Compactor, at LCPC2008. The compactor is used in certain collection scenarios.

• (2008-04-09) I wrote a quick hacking guide on Harmony GC, hoping it is useful for people interested in GC development. It is only a start though.

• (2008-03-28) I gave a talk on Parallel Garbage Collection in Shanghai Many-Core Workshop2008, arranged by gelato.org.

• (2007-10-01) We prepared a design overview on DRLVM, the JVM of Apache Harmoy, for OSSummit Asia 2007. (The conference was moved to 2008).

• (2007-04-22) I wrote an overview on Harmony GCv5 design on its current status.

Some code I wrote in my spare time:

• elf.zip: an ELF file format parser and linker written in Java. The test file with it shows how to use it. (Released to public domain.)

• x86.zip: an X86 machine code emitter with convenient APIs. I used it with the ELF linker above to compose a compiler backend in Java. (Released to public domain.)

Some technical papers:

• Qiong Zou, Xiao-Feng Li, and Long-Bing Zhang, Runtime Engine for Dynamic Profile Guided Stride Prefetching, Journal of Computer Science and Technology, 23(4): 633{643 July 2008.

• Xiao-Feng Li, Ligang Wang, and Chen Yang, A Fully Parallel LISP2 Compactor with Preservation of the Sliding Properties, Languages and Compilers for Parallel Computing (LCPC) 21st Annual Workshop, Edmonton, Alberta, Canada, July 31-August 2, 2008.

• Ming Wu and Xiao-Feng Li, Task-pushing: a Scalable Parallel GC Marking Algorithm without Synchronization Operations. IEEE Inter. Parallel and Distribution Processing Symposium (IPDPS) 2007, Long Beach, CA, March 2007.

• Chunrong Lai, Volosyuk Ivan, and Xiao-Feng Li, Behavior Characterization and Performance Study on Compacting Garbage Collectors with Apache Harmony., The Tenth Workshop on Computer Architecture Evaluation using Commercial Workloads (CAECW-10) held with HPCA-13. Phoenix, AZ, February 2007.

• Lixia Liu, Xiao-Feng Li, Michael Chen, and Roy Ju, A Throughput-Driven Task Creation and Mapping for Network Processors, 2007 International Conference on High Performance Embedded Architectures & Compilers (HiPEAC 2007), Ghent, Belgium, January 2007.

• Xiao-Feng Li, Chen Yang, Zhao-Hui Du, Tin-Fook Ngai, Exploiting Thread-Level Speculative Parallelism with Software Value Prediction. 10th Asia-Pacific Computer Systems Architecture Conference 2005 (ACSAC05): 367-388, Singapore, October 24-26, 2005.

• Tao Liu, Xiao-Feng Li, Lixia Liu, Chengyong Wu, Roy Ju, Optimizing Packet Accesses for a Domain Specific Language on Network Processors. 18th International Workshop on Languages and Compilers for Parallel Computing (LCPC 2005), Hawthorne, NY, USA, October 2005.

• Michael K. Chen, Xiao Feng Li, Ruiqi Lian, Jason H. Lin, Lixia Liu, Tao Liu, Roy Ju, Shangri-La: Achieving High Performance from Compiled Network Applications while Enabling Ease of Programming, Programming Language Design and Implementation 2005(PLDI'05), Chicago, IL, June, 2005.

• Xiao-Feng Li, Zhao-Hui Du, Chen Yang, Chu-Cheow Lim, Tin-Fook Ngai, Speculative Parallel Threading Architecture and Compilation, 4th Workshop on Compile and Runtime Techniques for Parallel Computing, held in conjunction with the ICPP'05, Oslo, Norway, June 2005.

• Du, Zhao-Hui; Lim, Chu-Cheow; Li, Xiao-Feng; Yang, Chen; Zhao, Qingyu and Nagi, Tin-Fook. A Cost-Driven Compilation Framework for Speculative Parallelization of Sequential Programs. Programming Language Design and Implementation 2004(PLDI'04), Washington, DC, June 9-11, 2004.

• Li, Xiao-Feng; Yang, Chen; Du Zhao-Hui; Ngai, Tin-Fook; Software Value Prediction for Speculative Parallel Threaded Computations, First Value-Prediction Workshop, held in conjunction with ISCA-30, San Diego, California, June 7, 2003.

• A logo I designed for LisoLeg, a Linux group:
  

• Learning Perl (Chinese version), which was translated by me. A programming bestseller in China.