Skip to main content
SpringerLink
Log in

GPURepair: Automated Repair of GPU Kernels

  • Conference paper
  • First Online:
Verification, Model Checking, and Abstract Interpretation (VMCAI 2021)

Part of the book series: Lecture Notes in Computer Science ((LNTCS,volume 12597))

Abstract

This paper presents a tool for repairing errors in GPU kernels written in CUDA or OpenCL due to data races and barrier divergence. Our novel extension to prior work can also remove barriers that are deemed unnecessary for correctness. We implement these ideas in our tool called GPURepair, which uses GPUVerify as the verification oracle for GPU kernels. We also extend GPUVerify to support CUDA Cooperative Groups, allowing GPURepair to perform inter-block synchronization for CUDA kernels. To the best of our knowledge, GPURepair is the only tool that can propose a fix for intra-block data races and barrier divergence errors for both CUDA and OpenCL kernels and the only tool that fixes inter-block data races for CUDA kernels. We perform extensive experiments on about 750 kernels and provide a comparison with prior work. We demonstrate the superiority of GPURepair through its capability to fix more kernels and its unique ability to remove redundant barriers and handle inter-block data races.

The author names are in alphabetical order.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution
Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 79.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 99.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

References

  1. CUDA Toolkit 5.0. https://developer.nvidia.com/cuda-toolkit-50-archive. Accessed 18 Nov 2020

  2. GPURepair GitHub Repository. https://github.com/cs17resch01003/gpurepair. Accessed 18 Nov 2020

  3. GPURepair VMCAI 2021 Artifacts. https://doi.org/10.5281/zenodo.4276525. Accessed 18 Nov 2020

  4. GPUVerify Test Suite. https://github.com/mc-imperial/gpuverify/tree/master/testsuite. Accessed 18 Nov 2020

  5. Microsoft Azure Fsv2-Series Virtual Machine Sizes. https://docs.microsoft.com/en-us/azure/virtual-machines/fsv2-series. Accessed 18 Nov 2020

  6. VMCAI 2021 Virtual Machine. https://doi.org/10.5281/zenodo.4017292. Accessed 18 Nov 2020

  7. Abdulla, P.A., Atig, M.F., Chen, Y.-F., Leonardsson, C., Rezine, A.: Counter-example guided fence insertion under TSO. In: Flanagan, C., König, B. (eds.) TACAS 2012. LNCS, vol. 7214, pp. 204–219. Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-28756-5_15

    Chapter  Google Scholar 

  8. Amighi, A., Darabi, S., Blom, S., Huisman, M.: Specification and verification of atomic operations in GPGPU programs. In: Calinescu, R., Rumpe, B. (eds.) SEFM 2015. LNCS, vol. 9276, pp. 69–83. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-22969-0_5

    Chapter  Google Scholar 

  9. Anand, S., Polikarpova, N.: Automatic synchronization for GPU kernels. In: FMCAD 2018, pp. 1–9. IEEE (2018)

    Google Scholar 

  10. Betts, A., et al.: The design and implementation of a verification technique for GPU kernels. TOPLAS 37(3), 10:1–10:49 (2015)

    Google Scholar 

  11. Betts, A., Chong, N., Donaldson, A.F., Qadeer, S., Thomson, P.: GPUVerify: a verifier for GPU kernels. In: OOPSLA 2012, pp. 113–132. ACM (2012)

    Google Scholar 

  12. Bjørner, N., Phan, A.-D., Fleckenstein, L.: vZ - an optimizing SMT solver. In: Baier, C., Tinelli, C. (eds.) TACAS 2015. LNCS, vol. 9035, pp. 194–199. Springer, Heidelberg (2015). https://doi.org/10.1007/978-3-662-46681-0_14

    Chapter  Google Scholar 

  13. Blom, S., Huisman, M., Mihelcic, M.: Specification and verification of GPGPU programs. Sci. Comput. Program. 95, 376–388 (2014)

    Article  Google Scholar 

  14. Boehm, B.W., Papaccio, P.N.: Understanding and Controlling Software Costs. IEEE Trans. Software Eng. 14(10), 1462–1477 (1988)

    Article  Google Scholar 

  15. Černý, P., Chatterjee, K., Henzinger, T.A., Radhakrishna, A., Singh, R.: Quantitative synthesis for concurrent programs. In: Gopalakrishnan, G., Qadeer, S. (eds.) CAV 2011. LNCS, vol. 6806, pp. 243–259. Springer, Heidelberg (2011). https://doi.org/10.1007/978-3-642-22110-1_20

    Chapter  Google Scholar 

  16. Chandra, S., Torlak, E., Barman, S., Bodík, R.: Angelic debugging. In: ICSE 2011, pp. 121–130. ACM (2011)

    Google Scholar 

  17. Deshmukh, J., Ramalingam, G., Ranganath, V.-P., Vaswani, K.: Logical concurrency control from sequential proofs. In: Gordon, A.D. (ed.) ESOP 2010. LNCS, vol. 6012, pp. 226–245. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-11957-6_13

    Chapter  Google Scholar 

  18. Griesmayer, A., Bloem, R., Cook, B.: Repair of boolean programs with an application to C. In: Ball, T., Jones, R.B. (eds.) CAV 2006. LNCS, vol. 4144, pp. 358–371. Springer, Heidelberg (2006). https://doi.org/10.1007/11817963_33

    Chapter  Google Scholar 

  19. Jin, G., Song, L., Zhang, W., Lu, S., Liblit, B.: Automated atomicity-violation fixing. In: PLDI 2011, pp. 389–400. ACM (2011)

    Google Scholar 

  20. Jobstmann, B., Griesmayer, A., Bloem, R.: Program repair as a game. In: Etessami, K., Rajamani, S.K. (eds.) CAV 2005. LNCS, vol. 3576, pp. 226–238. Springer, Heidelberg (2005). https://doi.org/10.1007/11513988_23

    Chapter  Google Scholar 

  21. Johnson, D.S.: Approximation algorithms for combinatorial problems. J. Comput. Syst Sci. 9(3), 256–278 (1974)

    Article  MathSciNet  Google Scholar 

  22. Joshi, S., Kroening, D.: Property-driven fence insertion using reorder bounded model checking. In: Bjørner, N., de Boer, F. (eds.) FM 2015. LNCS, vol. 9109, pp. 291–307. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-19249-9_19

    Chapter  Google Scholar 

  23. Joshi, S., Lal, A.: Automatically finding atomic regions for fixing bugs in Concurrent programs. CoRR abs/1403.1749 (2014)

    Google Scholar 

  24. Joshi, S., Muduganti, G.: GPURepair: Automated Repair of GPU Kernels. https://arxiv.org/abs/2011.08373 (2020)

  25. Li, G., Gopalakrishnan, G.: Scalable SMT-based verification of GPU kernel functions. In: FSE 2010, pp. 187–196. ACM (2010)

    Google Scholar 

  26. Li, G., Li, P., Sawaya, G., Gopalakrishnan, G., Ghosh, I., Rajan, S.P.: GKLEE: concolic verification and test generation for GPUs. In: PPOPP 2012, pp. 215–224. ACM (2012)

    Google Scholar 

  27. Malik, M.Z., Siddiqui, J.H., Khurshid, S.: Constraint-based program debugging using data structure repair. In: ICST 2011, pp. 190–199. IEEE Computer Society (2011)

    Google Scholar 

  28. Monteiro, F.R., et al.: ESBMC-GPU a context-bounded model checking tool to verify CUDA programs. Sci. Comput. Program. 152, 63–69 (2018)

    Article  Google Scholar 

  29. de Moura, L., Bjørner, N.: Z3: an efficient SMT solver. In: Ramakrishnan, C.R., Rehof, J. (eds.) TACAS 2008. LNCS, vol. 4963, pp. 337–340. Springer, Heidelberg (2008). https://doi.org/10.1007/978-3-540-78800-3_24

    Chapter  Google Scholar 

  30. Muzahid, A., Otsuki, N., Torrellas, J.: AtomTracker: A Comprehensive approach to atomic region inference and violation detection. In: MICRO 2010, pp. 287–297. IEEE Computer Society (2010)

    Google Scholar 

  31. Vechev, M.T., Yahav, E., Yorsh, G.: Abstraction-guided synthesis of synchronization. In: POPL 2010, pp. 327–338. ACM (2010)

    Google Scholar 

  32. Zhang, L., Wahib, M., Zhang, H., Matsuoka, S.: A study of single and multi-device synchronization methods in Nvidia GPUs. In: 2020 IEEE International Parallel and Distributed Processing Symposium (IPDPS), New Orleans, 18–22 May 2020, pp. 483–493. IEEE (2020)

    Google Scholar 

Download references

Acknowledgements

We thank the anonymous reviewers for their helpful comments and the authors of AutoSync for providing the source-code under a public license. We also thank the Ministry of Education, India, for financial support.

Author information

Authors and Affiliations

  1. Indian Institute of Technology Hyderabad, Hyderabad, India

    Saurabh Joshi & Gautam Muduganti

Authors
  1. Saurabh Joshi
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Gautam Muduganti
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Gautam Muduganti .

Editor information

Editors and Affiliations

  1. University of Copenhagen, Copenhagen, Denmark

    Fritz Henglein

  2. Tel Aviv University, Tel Aviv, Israel

    Sharon Shoham

  3. Technion, Haifa, Israel

    Yakir Vizel

Rights and permissions

Reprints and permissions

Copyright information

© 2021 Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Joshi, S., Muduganti, G. (2021). GPURepair: Automated Repair of GPU Kernels. In: Henglein, F., Shoham, S., Vizel, Y. (eds) Verification, Model Checking, and Abstract Interpretation. VMCAI 2021. Lecture Notes in Computer Science(), vol 12597. Springer, Cham. https://doi.org/10.1007/978-3-030-67067-2_18

Download citation

  • DOI: https://doi.org/10.1007/978-3-030-67067-2_18

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-030-67066-5

  • Online ISBN: 978-3-030-67067-2

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Search

Navigation