Improved Loop Distribution in LLVM using Polyhedral Dependences

Dangeti, Tharun Kumar and Borah, Utpal and Das, Santanu and Grosser, Tobias and Upadrasta, Ramakrishna (2017) Improved Loop Distribution in LLVM using Polyhedral Dependences. In: The Fourth Workshop on the LLVM Compiler Infrastructure in HPC, 13 November 2017, Denver, Colorado, USA.

Full text not available from this repository. (Request a copy)

Abstract

We propose a framework that can be used for improving loop-optimizations in LLVM using the Polyhedral framework of Polly. In our framework, we use the precise polyhedral dependences from Polly (provided by PolyhedralInfo), to construct a dependence graph, and perform loop transformations. As the first transformation case study of such a framework, we implemented loop distribution targeting improvement of inner-loop vectorization. Our loop distribution pass shows promising results on the TSVC benchmark; it is able to distribute 11 loops, while the earlier distribution pass is unable to distribute at all. We also have preliminary performance numbers from SPEC 2006. We believe that our work is the first step towards scalable and pre-defined loop-transformations in LLVM using exact dependences from Polly.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Upadrasta, RamakrishnaUNSPECIFIED
Item Type: Conference or Workshop Item (Paper)
Subjects: Computer science
Divisions: Department of Computer Science & Engineering
Depositing User: Team Library
Date Deposited: 17 May 2019 04:20
Last Modified: 17 May 2019 04:20
URI: http://raiith.iith.ac.in/id/eprint/5205
Publisher URL: https://llvm-hpc4-workshop.github.io/
Related URLs:

    Actions (login required)

    View Item View Item
    Statistics for RAIITH ePrint 5205 Statistics for this ePrint Item