Document Open Access Logo

Tenspiler: A Verified-Lifting-Based Compiler for Tensor Operations (Artifact)

Authors Jie Qiu , Colin Cai, Sahil Bhatia, Niranjan Hasabnis, Sanjit A. Seshia, Alvin Cheung

PDF

Artifact Description

Thumbnail PDF
PDF
DARTS.10.2.17.pdf
  • Filesize: 492 kB
  • 2 pages

Document Identifiers

Subject Classification

ACM Subject Classification
  • Software and its engineering → Compilers
Keywords
  • Program Synthesis
  • Code Transpilation
  • Tensor DSLs
  • Verification

Metrics

  • Access Statistics
  • Total Accesses (updated on a weekly basis)
    0
    PDF Downloads
    0
    Metadata Views

Abstract

In the related article, we described Tenspiler, a verified-lifting-based compiler that translates sequential programs to tensor operations. We further demonstrated its effectiveness by translating 69 benchmarks from into 6 different DSL targets and evaluating their performance against the baseline. This artifact includes the implementation of Tenspiler as well as files used to reproduce those results.

Cite As Get BibTex

Jie Qiu, Colin Cai, Sahil Bhatia, Niranjan Hasabnis, Sanjit A. Seshia, and Alvin Cheung. Tenspiler: A Verified-Lifting-Based Compiler for Tensor Operations (Artifact). In Special Issue of the 38th European Conference on Object-Oriented Programming (ECOOP 2024). Dagstuhl Artifacts Series (DARTS), Volume 10, Issue 2, pp. 17:1-17:2, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2024) https://doi.org/10.4230/DARTS.10.2.17

Author Details

Jie Qiu
  • Pittsburgh, PA, USA
Colin Cai
  • University of California, Berkeley, CA, USA
Sahil Bhatia
  • University of California, Berkeley, CA, USA
Niranjan Hasabnis
  • Intel Labs, Menlo Park, CA, USA
Sanjit A. Seshia
  • University of California, Berkeley, CA, USA
Alvin Cheung
  • University of California, Berkeley, CA, USA

Funding

This work was supported in part by DARPA Contract FA8750-23-C-0080, a Google BAIR Commons project, NSF grants IIS-1955488, IIS-2027575, ARO W911NF2110339, ONR N00014-21-1-2724, and DOE award DE-SC0016260, DE-SC0021982, and the Sloan Foundation.

Acknowledgements

We would like to thank Jayaram Bobba and Zhongkai Zhang from Intel’s Habana team for inputs on Gaudi architecture, TPC-C programming model, and obtaining high-performance from TPC kernels. We would like to thank Hasan Genc and Sophia Shao for helpful insights into Gemmini code generation.

Artifact

  MD5 Sum: 8a40381f27a9e92b0098c60e02f4671e (Get MD5 Sum)

Artifact Evaluation Policy

The artifact has been evaluated as described in the ECOOP 2024 Call for Artifacts and the ACM Artifact Review and Badging Policy.

Related Article

Questions / Remarks / Feedback
X

Feedback for Dagstuhl Publishing


Thanks for your feedback!

Feedback submitted

Could not send message

Please try again later or send an E-mail
© 2023-2025 Schloss Dagstuhl – LZI GmbH About DROPS Imprint Privacy Contact