Lazy error detection for microprocessor functional units

Mahmut Yilmaz, Albert Meixner, Sule Ozev, Daniel J. Sorin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

10 Citations (Scopus)

Abstract

We propose and evaluate the use of lazy error detection for a superscalar, out-of-order microprocessor's functional units. The key insight is that error detection is off the critical path, because an instruction's results are speculative for at least a cycle after being computed. The time between computing and committing the results can be used to lazily detect errors, and laziness allows us to use cheaper error detection logic. We show that lazy error detection is feasible, we develop a low-cost mechanism for detecting errors in adders that exploits laziness, and we show that an existing error detection scheme for multipliers can exploit laziness.

Original languageEnglish (US)
Title of host publicationProceedings - IEEE International Symposium on Defect and Fault Tolerance in VLSI Systems
Pages361-369
Number of pages9
DOIs
StatePublished - 2007
Externally publishedYes
Event22nd IEEE International Symposium on Defect and Fault-Tolerance in VLSI Systems, DFT 2007 - Rome, Italy
Duration: Sep 26 2007Sep 28 2007

Other

Other22nd IEEE International Symposium on Defect and Fault-Tolerance in VLSI Systems, DFT 2007
CountryItaly
CityRome
Period9/26/079/28/07

Fingerprint

Error detection
Microprocessor chips
Adders
Costs

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Yilmaz, M., Meixner, A., Ozev, S., & Sorin, D. J. (2007). Lazy error detection for microprocessor functional units. In Proceedings - IEEE International Symposium on Defect and Fault Tolerance in VLSI Systems (pp. 361-369). [4358405] https://doi.org/10.1109/DFT.2007.16

Lazy error detection for microprocessor functional units. / Yilmaz, Mahmut; Meixner, Albert; Ozev, Sule; Sorin, Daniel J.

Proceedings - IEEE International Symposium on Defect and Fault Tolerance in VLSI Systems. 2007. p. 361-369 4358405.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Yilmaz, M, Meixner, A, Ozev, S & Sorin, DJ 2007, Lazy error detection for microprocessor functional units. in Proceedings - IEEE International Symposium on Defect and Fault Tolerance in VLSI Systems., 4358405, pp. 361-369, 22nd IEEE International Symposium on Defect and Fault-Tolerance in VLSI Systems, DFT 2007, Rome, Italy, 9/26/07. https://doi.org/10.1109/DFT.2007.16
Yilmaz M, Meixner A, Ozev S, Sorin DJ. Lazy error detection for microprocessor functional units. In Proceedings - IEEE International Symposium on Defect and Fault Tolerance in VLSI Systems. 2007. p. 361-369. 4358405 https://doi.org/10.1109/DFT.2007.16
Yilmaz, Mahmut ; Meixner, Albert ; Ozev, Sule ; Sorin, Daniel J. / Lazy error detection for microprocessor functional units. Proceedings - IEEE International Symposium on Defect and Fault Tolerance in VLSI Systems. 2007. pp. 361-369
@inproceedings{69a8597852254ef29babd3dbc05cd6e4,
title = "Lazy error detection for microprocessor functional units",
abstract = "We propose and evaluate the use of lazy error detection for a superscalar, out-of-order microprocessor's functional units. The key insight is that error detection is off the critical path, because an instruction's results are speculative for at least a cycle after being computed. The time between computing and committing the results can be used to lazily detect errors, and laziness allows us to use cheaper error detection logic. We show that lazy error detection is feasible, we develop a low-cost mechanism for detecting errors in adders that exploits laziness, and we show that an existing error detection scheme for multipliers can exploit laziness.",
author = "Mahmut Yilmaz and Albert Meixner and Sule Ozev and Sorin, {Daniel J.}",
year = "2007",
doi = "10.1109/DFT.2007.16",
language = "English (US)",
isbn = "0769528856",
pages = "361--369",
booktitle = "Proceedings - IEEE International Symposium on Defect and Fault Tolerance in VLSI Systems",

}

TY - GEN

T1 - Lazy error detection for microprocessor functional units

AU - Yilmaz, Mahmut

AU - Meixner, Albert

AU - Ozev, Sule

AU - Sorin, Daniel J.

PY - 2007

Y1 - 2007

N2 - We propose and evaluate the use of lazy error detection for a superscalar, out-of-order microprocessor's functional units. The key insight is that error detection is off the critical path, because an instruction's results are speculative for at least a cycle after being computed. The time between computing and committing the results can be used to lazily detect errors, and laziness allows us to use cheaper error detection logic. We show that lazy error detection is feasible, we develop a low-cost mechanism for detecting errors in adders that exploits laziness, and we show that an existing error detection scheme for multipliers can exploit laziness.

AB - We propose and evaluate the use of lazy error detection for a superscalar, out-of-order microprocessor's functional units. The key insight is that error detection is off the critical path, because an instruction's results are speculative for at least a cycle after being computed. The time between computing and committing the results can be used to lazily detect errors, and laziness allows us to use cheaper error detection logic. We show that lazy error detection is feasible, we develop a low-cost mechanism for detecting errors in adders that exploits laziness, and we show that an existing error detection scheme for multipliers can exploit laziness.

UR - http://www.scopus.com/inward/record.url?scp=84883414482&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84883414482&partnerID=8YFLogxK

U2 - 10.1109/DFT.2007.16

DO - 10.1109/DFT.2007.16

M3 - Conference contribution

SN - 0769528856

SN - 9780769528854

SP - 361

EP - 369

BT - Proceedings - IEEE International Symposium on Defect and Fault Tolerance in VLSI Systems

ER -