Skip to main content
SpringerLink
Log in

DYNAMEM — A microarchitecture for improving memory disambiguation at run-time

  • Regular Papers
  • Published:
Journal of Computer Science and Technology Aims and scope Submit manuscript

Abstract

This paper presents a new microarchitecture technique named DYNAMEM, in which memory reference instructions are dynamically scheduled and can be executed out-of-order. Load instructions can bypass store instructions speculatively, even if the store instructions’ addresses are unknown. DYNAMEM can greatly alleviate the restraints of ambiguous memory dependencies. Simulation results show that the frequency of false load is low. Mechanism has been provided to repair false loads with low penalty, and to achieve precise interrupts. Discussions and experimental results show that DYNAMEM could dramatically raise instruction-level parallelism in programs without recompilation.

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

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Theobald Ket al. On the limits of program parallelism and its smoothability. InProc. of 25th Annual Symposium on Microarchitecture, Dec. 1992.

  2. Wall D. Limits of instruction-level parallelism. InProc. of 4th International Conference on Architectural Support for Programming Languages and Operating Systems, pp.176–188, Apr. 1991.

  3. Johnson M. Superscalar Microprocessor Design. Pretice-Hall, 1991.

  4. Nicolau A. Run-time disambiguation: Coping with statically unpredictable dependencies.IEEE Transaction on Computers, 1989, 38(5): 663–678

    Article  Google Scholar 

  5. Sohi G. Instruction issue logic for high-performance, interruptible, multiple functional unit, pipelined computers.IEEE Transaction on Computers, 1990, 39(3): 349–359.

    Article  Google Scholar 

  6. Song Set al. The Power PC 604 RISC Microprocessor.IEEE Micro, Oct. 1992, pp. 8–17.

  7. Li Sanli, Liao Heng. Virtual register and parallel instruction processing unit — New computer architecture concept.Mini-Micro Computer Systems, 1995, 16(6): 1–8 (in Chinese)

    Google Scholar 

  8. Lam M, Wilson R. Limits of control flow on parallelism. InProc. of 19th Annual International Symposium on Computer Architecture, pp.46–57, June 1992.

  9. Lee J, Smith A. Branch prediction strategies and branch target buffer design.Computer, 1984, (1): 6–22.

    Article  Google Scholar 

  10. Mayden Det al. Efficient and exact data dependency analysis. InProc. of 1991 SIGPLAN Symposium on Compiler Construction, pp.1–14, June 1991.

  11. Goff Get al. Practical dependency testing. InProc. of 1991 SIGPLAN Symposium on Compiler Construction, pp.15–29, June 1991.

  12. Melvin S, Patt Y. Exploiting fine-grained parallelism through a combination of hardware and software techniques. InProc. of 18th Annual International Symposium on Computer Architecture, pp.287–296, Sep. 1991.

  13. Huang Aet al. Speculative disambiguation: A compilation technique for dynamic memory disambiguation. InProc. of 21st Annual International Symposium on Computer Architecture, pp.200–210, Apr. 1994.

  14. Gallagher D. Dynamic memory disambiguation using the memory conflict buffer. InProc. of 6th International Conference on Architectural Support, for Programming Languages and Operating Systems, pp.183–193, Oct. 1994.

  15. Diefendorff K, d Allen M. Organization of the Motorala 88110 superscalar RISC microprocessor.IEEE Micro, April 1992, pp.40–63.

  16. Diep Tet al. Performance evaluation of the power PC 620 microarchitectur. InProc. of 18th International Symposium on Computer Architecture, pp.163–174, June. 1995.

  17. Smith J, Pleszkun A. Implementation of precise interrupts in pipelined processors.IEEE Trans. on Computers, 1988, 37(5): 562–573.

    Article  Google Scholar 

  18. Intel. Intel Pentium(R) Pro Processor On-line Introduction. http: //www.intel.com/procs/ppro/intro/index.htm.

  19. Dahl P. Reducing memory traffic with CRegs. InProc. of 27th Annual International Symposium on Microarchitecture, pp.101–105, Nov. 1994.

  20. Popescu Vet al. The metaflow architecture.IEEE Micro, June 1991, pp. 10–13, 63–73.

  21. Weiss S, Smith J. Instruction issue logic in pipelined supercomputers.IEEE Transaction on Computers, 1984, C-33(11): 1013–1022.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Computer Science & Technology, Tsinghua University, 100084, Beijing

    Wang Xianzhu, Liao Heng & Li Sanli

Authors
  1. Wang Xianzhu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Liao Heng
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Li Sanli
    View author publications

    You can also search for this author in PubMed Google Scholar

Additional information

This research was supported by the National Natural Science Foundation of China.

Wang Xianzhu received his B.S. degree in computer science from Tsinghua University in 1992. He is now a Ph.D. candidate in the Department of Computer Science and Technology, Tsinghua University. His research interests include processor microarchitecture and instruction-level parallelism.

Liao Heng received his B.S. degree in computer science from Tsinghua University in 1991. He is now a Ph.D. candidate in the Department of Computer Science and Technology, Tsinghua University. His research interests include processor microarchitecture and highperformance interconnection.

Li Sanli is an Academician of Chinese Academy of Engineering, and a Professor in the Department of Computer Science and Technology, Tsinghua University. His research interests include network parallel computing, processor architecture and instruction-level parallelism.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Wang, X., Liao, H. & Li, S. DYNAMEM — A microarchitecture for improving memory disambiguation at run-time. J. of Comput. Sci. & Technol. 11, 589–600 (1996). https://doi.org/10.1007/BF02951622

Download citation

  • Received:

  • Revised:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF02951622

Keywords

Search

Navigation