See this document in CiteSeerX!

Eliminating Squashes Through Learning Cross-Thread Violations in Speculative Parallelization for Multiprocessors (2002)  (Make Corrections)  (5 citations)
Marcelo Cintra, Josep Torrellas



  Home/Search   Context   Related

Links:   ACM

 
View or download:
uiuc.edu/iacomapapers/learn02.pdf
Cached:  PS.gz  PS  PDF   Image  Update  Help

From:  uiuc.edu/papers (more)
(Enter author homepages)

Rate this article: (best)
  Comment on this article  
(Enter summary)

Abstract: With speculative thread-level parallelization, codes that cannot be fully compiler-analyzed are aggressively executed in parallel. If the hardware detects a cross-thread dependence violation, it squashes offending threads and resumes execution. Unfortunately, frequent squashing cripples performance. (Update)

Context of citations to this paper:   More

...hardware support in the coherence protocol to detect memory dependence violations. In recent work, Steffan et. al [21] and Cintra et. al [7] examine different methods of improving memory value communication in such a system. The load value prediction approach they describe,...

.... These were not used in the heap sort example, but will be discussed briefly here for completeness; more details can be found in [3][5][6] 8] 12] 14] 15] 17] The techniques comprise loop chunking, loop slicing, parallel reductions and explicit synchronization. Loop...

Cited by:   More
Tolerating Dependences Between Large Speculative.. - Christopher Colohan..   (Correct)
Journal of Instruction-Level Parallelism 5 (2003) 1-21.. - The Role Of   (Correct)
The Jrpm System for Dynamically Parallelizing Java Programs - Chen, Olukotun (2003)   (Correct)

Active bibliography (related documents):   More   All
1.2:   Architectural Support for Scalable Speculative.. - Cintra, Martinez.. (2000)   (Correct)
1.2:   Software Logging under Speculative Parallelization - Garzaran, Prvulovic..   (Correct)
0.4:   The Role of Return Value Prediction in Exploiting Speculative .. - Shiwen Hu Ravi (2002)   (Correct)

Similar documents based on text:   More   All
0.4:   Removing Architectural Bottlenecks to the.. - Prvulovic..   (Correct)
0.3:   Speculative Parallel Execution of Loops with.. - Zhang, Rauchwerger.. (1997)   (Correct)
0.3:   Speculative Locks for Concurrent Execution of Critical.. - Martinez, Torrellas (2001)   (Correct)

Related documents from co-citation:   More   All
4:   Improving Value Communication for Thread-Level Speculation - Steffan, Colohan et al. - 2002
3:   A Scalable Approach to Thread-Level Speculation - Steffan, Colohan et al. - 2000
3:   Compiler optimization of scalar value communication between speculative threads - Zhai, Colohan et al. - 2002

BibTeX entry:   (Update)

M. Cintra and J. Torrellas. Eliminating squashes through learning cross-thread violations in speculative parallelization for multiprocessors. In HPCA, 2002. http://citeseer.ist.psu.edu/cintra02eliminating.html   More

@misc{ cintra02eliminating,
  author = "M. Cintra and J. Torrellas",
  title = "Eliminating squashes through learning cross-thread violations in speculative
    parallelization for multiprocessors",
  text = "M. Cintra and J. Torrellas. Eliminating squashes through learning cross-thread
    violations in speculative parallelization for multiprocessors. In HPCA,
    2002.",
  year = "2002",
  url = "citeseer.ist.psu.edu/cintra02eliminating.html" }
Citations (may not include all citations):
269   Multiscalar Processors - Sohi, Breach et al. - 1995
159   The LRPD Test: Speculative RunTime Parallelization of Loops .. - Rauchwerger, Padua - 1995
145   Exceeding the Dataflow Limit via Value Prediction - Lipasti, Shen - 1996  ACM   DBLP
139   The Predictability of Data Values - Sazeides, Smith - 1997  ACM   DBLP
116   Highly Accurate Data Value Prediction Using Hybrid Predictor.. - Wang, Franklin - 1997  ACM   DBLP
110   Improving the Accuracy of Dynamic Branch Prediction Using Br.. (context) - Pan, So et al. - 1992  ACM   DBLP
102   Dynamic Speculation and Synchronization of Data Dependences - Moshovos, Breach et al. - 1997  ACM   DBLP
79   Removing Architectural Bottlenecks to the Scalability of Spe.. - Prvulovic, Garzaran et al.  ACM   DBLP
74   Speculative Versioning Cache - Gopal, Vijaykumar et al. - 1998  ACM   DBLP
72   A Dynamic Multithreading Processor - Akkary, Driscoll - 1998  ACM   DBLP
72   Data Speculation Support for a Chip Multiprocessor (context) - Hammond, Wiley et al. - 1998  ACM   DBLP
64   Memory Dependence Prediction Using Store Sets - Chrysos, Emer - 1998  ACM   DBLP
44   Hardware for Speculative Run-time Parallelization in Distrib.. - Zhang, Rauchwerger et al. - 1998  ACM   DBLP
43   Advanced Program Restructuring for High-Performance Computer.. - Blume, Doallo et al. - 1996
38   Architectural Support for Scalable Speculative Parallelizati.. - Cintra, nez et al. - 2000  ACM   DBLP
38   A Scalable Approach to Thread-Level Speculation - Steffan, Colohan et al. - 2000
30   Clustered Speculative Multithreaded Processors - Marcuello, alez - 1999  ACM   DBLP
30   A Chip-Multiprocessor Architecture with Speculative Multithr.. - Krishnan, Torrellas - 1999  ACM   DBLP
23   Speculation Techniques for Improving Load Related Instructio.. - Yoaz, Erez et al. - 1999
22   The Superthreaded Processor Architecture - Tsai, Huang et al. - 1999  ACM   DBLP
21   Task Selection for a Multiscalar Processor - Vijaykumar, Sohi - 1998  ACM   DBLP
20   MAJC: Microprocessor Architecture for Java Computing (context) - Tremblay - 1999
19   Value Prediction for Speculative Multithreaded Architectures - Marcuello, Tubella et al. - 1999  ACM   DBLP
17   A Direct-Execution Framework for Fast and Accurate Simulatio.. - Krishnan, Torrellas - 1998
10   Improving Value Communication for Thread-Level Speculation - Steffan, Colohan et al. - 2002  ACM   DBLP
9   Multiplex: Unifying Conventional and Speculative Thread-Leve.. - Ooi, Kim et al. - 2001  DBLP
9   SPEComp: A New Benchmark Suite for Measuring Parallel Comput.. - Aslot, Domeika et al. - 2001  DBLP
6   Limits on Speculative Module-Level Parallelism in Imperative.. - Warg, om - 2001  ACM   DBLP
5   In Search of Speculative Threadlevel Parallelism (context) - Oplinger, Heine et al. - 1999
2   A Front End for Efficient Simulation of Shared-Memory Multip.. (context) - Veenstra, Fowler - 1994
2   Techniques for Run-Time Parallelization of Loops (context) - Gupta, Nim - 1998
1   A Software Approach to ThreadLevel Data Dependence Speculati.. (context) - Rundberg, om - 2000

Documents on the same site (http://iacoma.cs.uiuc.edu/papers.html):   More
Comprehensive Hardware and Software Support for Operating.. - Xia, Torrellas (1999)   (Correct)
Software Trace Cache - Ramírez, Larriba-Pey.. (1999)   (Correct)
An Efficient Algorithm for the Run-time Parallelization of .. - Chen, Torrellas, Yew (1994)   (Correct)

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC