research-article

Trace-based Register Allocation in a JIT Compiler

Authors:

Matthias Grimmer,

Thomas Würthinger,

Hanspeter MössenböckAuthors Info & Claims

PPPJ '16: Proceedings of the 13th International Conference on Principles and Practices of Programming on the Java Platform: Virtual Machines, Languages, and Tools

Article No.: 14, Pages 1 - 11

https://doi.org/10.1145/2972206.2972211

Published: 29 August 2016 Publication History

Abstract

State-of-the-art dynamic compilers often use global approaches, like Linear Scan or Graph Coloring, for register allocation. These algorithms consider the complete compilation unit for allocation, which increases the complexity of the implementation (e.g., support for lifetime holes in Linear Scan) and potentially also affects compilation time. We propose a novel non-global algorithm, which splits a compilation unit into traces based on profiling feedback and subsequently performs register allocation within each trace individually. Traces reduce the problem size to a single linear code segment, which simplifies the problem a register allocator needs to solve. Additionally, we can apply different register allocation algorithms to each trace. We show that this non-global approach can achieve results competitive to global register allocation.

We present an implementation of Trace Register Allocation based on the Graal VM and show an evaluation for common Java benchmarks. We demonstrate that performance of this non-global approach is within 3% (on AMD64) and 1% (on SPARC) of global Linear Scan register allocation.

References

[1]

Preston Briggs, Keith D. Cooper, and Linda Torczon. "Improvements to graph coloring register allocation". In: TOPLAS'94 (1994).

[2]

Massimiliano Poletto and Vivek Sarkar. "Linear Scan Register Allocation". In: TOPLAS'99 (1999).

[3]

Christian Wimmer and Hanspeter Mössenböck. "Optimized Interval Splitting in a Linear Scan Register Allocator". In: VEE'05. ACM, 2005.

Digital Library

[4]

Christian Wimmer and Michael Franz. "Linear Scan Register Allocation on SSA Form". In: CGO'10. ACM, 2010.

Digital Library

[5]

Gregory J. Chaitin et al. "Register Allocation via Coloring". In: Computer languages (1981).

Digital Library

[6]

Michael D. Smith, Norman Ramsey, and Glenn Holloway. "A generalized algorithm for graph-coloring register allocation". In: SIGPLAN Not. (2004).

Digital Library

[7]

Lal George and Andrew W. Appel. "Iterated register coalescing". In: TOPLAS'96 (1996).

Digital Library

[8]

Omri Traub, Glenn Holloway, and Michael D. Smith. "Quality and Speed in Linear-scan Register Allocation". In: SIGPLAN Not. (1998).

Digital Library

[9]

Thomas Kotzmann et al. "Design of the Java HotSpot™client compiler for Java 6". In: TACO'08 (2008).

[10]

Matthew Arnold et al. "Adaptive Optimization in the Jalapeño JVM". In: SIGPLAN Not. (2000).

Digital Library

[11]

Chris Lattner and Vikram Adve. "LLVM: A Compilation Framework for Lifelong Program Analysis & Transformation". In: CGO'04. IEEE Computer Society, 2004.

Digital Library

[12]

Josef Eisl. "Trace Register Allocation". In: SPLASH Companion 2015. ACM, 2015.

Digital Library

[13]

Graal Project. OpenJDK Community. url: http://openjdk.java.net/projects/graal/.

[14]

Gilles Duboscq et al. "An Intermediate Representation for Speculative Optimizations in a Dynamic Compiler". In: VMIL'13 (2013).

Digital Library

[15]

Lukas Stadler et al. "An Experimental Study of the Influence of Dynamic Compiler Optimizations on Scala Performance". In: SCALA'13. ACM, 2013.

Digital Library

[16]

Gilles Duboscq, Thomas Würthinger, and Hanspeter Mössenböck. "Speculation without regret". In: PPPJ'14 (2014).

[17]

Lukas Stadler, Thomas Würthinger, and Hanspeter Mössenböck. "Partial Escape Analysis and Scalar Replacement for Java". In: CGO '14. ACM, 2014.

Digital Library

[18]

Doug Simon et al. "Snippets: Taking the High Road to a Low Level". In: TACO'15 (2015).

Digital Library

[19]

Oracle Corporation. Java SE HotSpot at a Glance. url: http://www.oracle.com/technetwork/articles/javase/index-jsp-136373.html (visited on 05/06/2016).

[20]

Ron Cytron et al. "Efficiently Computing Static Single Assignment Form and the Control Dependence Graph". In: ACM Trans. Program. Lang. Syst. (1991).

Digital Library

[21]

S. M. Blackburn et al. "The DaCapo Benchmarks: Java Benchmarking Development and Analysis". In: OOPSLA'06. ACM Press, 2006.

Digital Library

[22]

Michael Bebenita et al. "Trace-based compilation in execution environments without interpreters". In: PPPJ'10 (2010).

Digital Library

[23]

John R. Ellis. "Bulldog: A Compiler for VLIW Architectures". PhD thesis. Yale University, 1985.

Digital Library

[24]

P. Geoffrey Lowney et al. "The Multiflow Trace Scheduling Compiler". In: Journal of Supercomputing (1993).

Digital Library

[25]

Michael Paleczny, Christopher Vick, and Cliff Click. "The Java HotSpot™ Server Compiler". In: JVM'01. USENIX Association, 2001. url: https://www.usenix.org/legacy/events/jvm01/full_papers/paleczny/paleczny.pdf.

Digital Library

[26]

JEP 243: Java-Level JVM Compiler Interface. 2014. url: http://openjdk.java.net/jeps/243 (visited on 05/06/2016).

[27]

DaCapo Project. The DaCapo Benchmark Suite. 2012. url: http://dacapobench.org/ (visited on 05/26/2016).

[28]

Andreas Sewe et al. "Da capo con scala". In: OOPSLA'11 (2011).

[29]

SPECjvm2008: Java Virtual Machine Benchmark. url: https://www.spec.org/jvm2008/ (visited on 06/15/2015).

[30]

SPECjbb2015: Java Server Benchmark. url: https://www.spec.org/jbb2015/ (visited on 05/25/2016).

[31]

Oracle Corporation. Sun Server X3-2. 2013. url: http://www.oracle.com/us/products/servers - storage/servers/x86/sun-server-x3-2-ds-1683091.pdf (visited on 05/25/2016).

[32]

Linux Containers. url: http://linuxcontainers.org/ (visited on 05/26/2016).

[33]

Oracle Corporation. SPARC T5-2 Server. 2013. url: http://www.oracle.com/us/products/servers-storage/servers/sparc/oracle-sparc/t5-2/sparc-t5-2-ds-1922871.pdf (visited on 05/25/2016).

[34]

Oracle Corporation. Introduction to Oracle Solaris Zones. 2014. url: http://docs.oracle.com/cd/E3678401/pdf/E36848.pdf (visited on 05/26/2016).

[35]

John W. Tukey. Exploratory data analysis. Reading, Mass., 1977.

[36]

John M. Chambers. Statistical Models in S. CRC Press, Inc., 1991.

Digital Library

[37]

Joseph Allen Fisher. "Trace Scheduling: A Technique for Global Microcode Compaction". In: Computers, IEEE Transactions on Computers (1981).

Digital Library

[38]

Vasanth Bala, Evelyn Duesterwald, and Sanjeev Banerjia. "Dynamo: A Transparent Dynamic Optimization System". In: SIGPLAN Not. (2000).

Digital Library

[39]

Andreas Gal, Christian W. Probst, and Michael Franz. "HotpathVM: An Effective JIT Compiler for Resource-constrained Devices". In: VEE'06. ACM, 2006.

Digital Library

[40]

Andreas Gal et al. "Trace-based Just-in-Time Type Specialization for Dynamic Languages". In: PLDI'09. ACM, 2009.

Digital Library

[41]

Carl Friedrich Bolz et al. "Tracing the meta-level". In: ICOOOLPS'09 (2009).

[42]

Christian Häubl and Hanspeter Mössenböck. "Trace-based compilation for the Java HotSpot virtual machine". In: PPPJ'11 (2011).

Digital Library

[43]

Christian Häubl, Christian Wimmer, and Hanspeter Mössenböck. "Context-sensitive trace inlining for Java". In: Computer Languages, Systems & Structures (2013).

Digital Library

[44]

David Callahan and Brian Koblenz. "Register Allocation via Hierarchical Graph Coloring". In: SIGPLAN Not. (1991).

Digital Library

[45]

Guei-Yuan Lueh, Thomas Gross, and Ali-Reza Adl-Tabatabai. "Global register allocation based on graph fusion". In: Lecture Notes in Computer Science. Springer Berlin Heidelberg, 1997.

Digital Library

[46]

Wen -Mei W. Hwu et al. "The superblock: An effective technique for VLIW and superscalar compilation". In: The Journal of Supercomputing (1993).

[47]

David Ryan Koes and Seth Copen Goldstein. "A global progressive register allocator". In: PLDI'06 (2006).

Digital Library

[48]

David Ryan Koes and Seth Copen Goldstein. A better global progressive register allocator. 2006. url: http://www.cs.cmu.edu/~dkoes/research/lctes06_tracealloc.pdf.

[49]

C. Scott Ananian. "The Static Single Information Form". MA thesis. Princeton University, 1999. url: http://publications.csail.mit.edu/lcs/pubs/pdf/MIT-LCS-TR-801.pdf.

[50]

Jeremy Singer. Static program analysis based on virtual register renaming. Tech. rep. University of Cambridge, 2006. url: https://www.cl.cam.ac.uk/techreports/UCAM-CL-TR-660.pdf.

[51]

B. Boissinot et al. SSI Properties Revisited. Tech. rep. ENS-Lyon, 2009. url: https://hal.inria.fr/inria-00404236/file/bboissin-ssi-RR.pdf.

[52]

Fernando Magno Quintão Pereira and Jens Palsberg. "Register allocation by puzzle solving". In: PLDI'08 (2008).

Digital Library

[53]

Roberto Castañeda Lozano et al. "Constraint-Based Register Allocation and Instruction Scheduling". In: CP'12 (2012).

Cited By

Vukasovic MProkopec A(2023)Exploiting Partially Context-sensitive Profiles to Improve Performance of Hot CodeACM Transactions on Programming Languages and Systems10.1145/361293745:4(1-64)Online publication date: 13-Sep-2023
https://dl.acm.org/doi/10.1145/3612937
Kim MPark JMoon S(2020)Irregular Register Allocation for Translation of Test-pattern ProgramsACM Transactions on Architecture and Code Optimization10.1145/342737818:1(1-23)Online publication date: 30-Dec-2020
https://dl.acm.org/doi/10.1145/3427378
Latifi FSmaragdakis Y(2019)Practical second Futamura projection: partial evaluation for high-performance language interpretersProceedings Companion of the 2019 ACM SIGPLAN International Conference on Systems, Programming, Languages, and Applications: Software for Humanity10.1145/3359061.3361077(29-31)Online publication date: 20-Oct-2019
https://dl.acm.org/doi/10.1145/3359061.3361077
Show More Cited By

Trace-based Register Allocation in a JIT Compiler
1. Software and its engineering
  1. Software notations and tools

Recommendations

Linear scan register allocation on SSA form
CGO '10: Proceedings of the 8th annual IEEE/ACM international symposium on Code generation and optimization

The linear scan algorithm for register allocation provides a good register assignment with a low compilation overhead and is thus frequently used for just-in-time compilers. Although most of these compilers use static single assignment (SSA) form, the ...
Trace Register Allocation Policies: Compile-time vs. Performance Trade-offs
ManLang 2017: Proceedings of the 14th International Conference on Managed Languages and Runtimes

Register allocation is an integral part of compilation, regardless of whether a compiler aims for fast compilation or optimal code quality. State-of-the-art dynamic compilers often use global register allocation approaches such as linear scan. Recent ...
Parallel trace register allocation
ManLang '18: Proceedings of the 15th International Conference on Managed Languages & Runtimes

Register allocation is a mandatory task for almost every compiler and consumes a significant portion of compile time. In a just-in-time compiler, compile time is a particular issue because compilation happens during program execution and contributes to ...

Comments

Information & Contributors

Information

Published In

cover image ACM Other conferences

PPPJ '16: Proceedings of the 13th International Conference on Principles and Practices of Programming on the Java Platform: Virtual Machines, Languages, and Tools

August 2016

186 pages

ISBN:9781450341356

DOI:10.1145/2972206

Conference Chair:
Yudi Zheng
University of Lugano
,
General Chair:
Walter Binder,
Program Chair:
Petr Tůma

Copyright © 2016 ACM.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

In-Cooperation

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 29 August 2016

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article
Research
Refereed limited

Conference

PPPJ '16

PPPJ '16: Virtual Machines, Languages, and Tools

August 29 - September 2, 2016

Lugano, Switzerland

Acceptance Rates

PPPJ '16 Paper Acceptance Rate 14 of 31 submissions, 45%;

Overall Acceptance Rate 29 of 58 submissions, 50%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

8
Total Citations
View Citations
372
Total Downloads

Downloads (Last 12 months)22
Downloads (Last 6 weeks)2

Reflects downloads up to 12 Sep 2024

Other Metrics

View Author Metrics

Citations

Cited By

Vukasovic MProkopec A(2023)Exploiting Partially Context-sensitive Profiles to Improve Performance of Hot CodeACM Transactions on Programming Languages and Systems10.1145/361293745:4(1-64)Online publication date: 13-Sep-2023
https://dl.acm.org/doi/10.1145/3612937
Kim MPark JMoon S(2020)Irregular Register Allocation for Translation of Test-pattern ProgramsACM Transactions on Architecture and Code Optimization10.1145/342737818:1(1-23)Online publication date: 30-Dec-2020
https://dl.acm.org/doi/10.1145/3427378
Latifi FSmaragdakis Y(2019)Practical second Futamura projection: partial evaluation for high-performance language interpretersProceedings Companion of the 2019 ACM SIGPLAN International Conference on Systems, Programming, Languages, and Applications: Software for Humanity10.1145/3359061.3361077(29-31)Online publication date: 20-Oct-2019
https://dl.acm.org/doi/10.1145/3359061.3361077
Prokopec ARosà ALeopoldseder DDuboscq GTůma PStudener MBulej LZheng YVillazón ASimon DWürthinger TBinder WMcKinley KFisher K(2019)Renaissance: benchmarking suite for parallel applications on the JVMProceedings of the 40th ACM SIGPLAN Conference on Programming Language Design and Implementation10.1145/3314221.3314637(31-47)Online publication date: 8-Jun-2019
https://dl.acm.org/doi/10.1145/3314221.3314637
Shull THuang JTorrellas JSartor JNaik MRossbach C(2019)QuickCheck: using speculation to reduce the overhead of checks in NVM frameworksProceedings of the 15th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments10.1145/3313808.3313822(137-151)Online publication date: 14-Apr-2019
https://dl.acm.org/doi/10.1145/3313808.3313822
Leopoldseder DStadler LRigger MWürthinger TMössenböck HKell SMarr S(2018)A cost model for a graph-based intermediate-representation in a dynamic compilerProceedings of the 10th ACM SIGPLAN International Workshop on Virtual Machines and Intermediate Languages10.1145/3281287.3281290(26-35)Online publication date: 4-Nov-2018
https://dl.acm.org/doi/10.1145/3281287.3281290
Eisl JLeopoldseder DMössenböck HTilevich EMössenböck H(2018)Parallel trace register allocationProceedings of the 15th International Conference on Managed Languages & Runtimes10.1145/3237009.3237010(1-7)Online publication date: 12-Sep-2018
https://dl.acm.org/doi/10.1145/3237009.3237010
Eisl JMarr SWürthinger TMössenböck H(2017)Trace Register Allocation PoliciesProceedings of the 14th International Conference on Managed Languages and Runtimes10.1145/3132190.3132209(92-104)Online publication date: 27-Sep-2017
https://dl.acm.org/doi/10.1145/3132190.3132209

View Options

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Media

Figures

Other

Tables

View Table of Contents