goSLP: globally optimized superword level parallelism framework

@article{Mendis2018goSLPGO,
  title={goSLP: globally optimized superword level parallelism framework},
  author={Charith Mendis and Saman P. Amarasinghe},
  journal={Proceedings of the ACM on Programming Languages},
  year={2018},
  volume={2},
  pages={1 - 28}
}
  • Charith Mendis, Saman P. Amarasinghe
  • Published 2018
  • Computer Science
  • Proceedings of the ACM on Programming Languages
  • Modern microprocessors are equipped with single instruction multiple data (SIMD) or vector instruction sets which allow compilers to exploit superword level parallelism (SLP), a type of fine-grained parallelism. Current SLP auto-vectorization techniques use heuristics to discover vectorization opportunities in high-level language code. These heuristics are fragile, local and typically only present one vectorization strategy that is either accepted or rejected by a cost model. We present goSLP… CONTINUE READING

    References

    Publications referenced by this paper.
    A compiler framework for extracting superword level parallelism
    44