GPMB—software pipelining branch-intensive loops

Zhihong Tang, Gang Chen, Chihong Zhang, Yingwei Zhang, Bogong Su, Stanley HabibInternational Symposium on Microarchitecture 1993 · DOI

Abstract

To achieve higher instruction-level parallelism, the constraint imposed by a single control flow must be relaxed. Control operations should execute in parallel just like data operations. We present a new software pipelining method called GPMB (Global Pipelining with Multiple Branches) which is based on architectures supporting multi-way branching and multiple control flows. Preliminary experimental results show that, for IFless loops, GPMB performs as well as modulo scheduling, and for branch-intensive loops, GPMB performs much better than software pipelining assuming the constraint of one two-way branch per cycle. >
BibTeX
@inproceedings{tang1993gpmbsoftware,
  author = {Zhihong Tang and Gang Chen and Chihong Zhang and Yingwei Zhang and Bogong Su and Stanley Habib},
  title = {{GPMB—software pipelining branch-intensive loops}},
  booktitle = {International Symposium on Microarchitecture},
  year = {1993},
  doi = {10.5555/255235.255243},
}

Loading related content...

Figure