PENCIL: A Platform-Neutral Compute Intermediate Language for Accelerator Programming

Riyadh Baghdadi; Ulysse Beaugnon; Albert Cohen; Tobias Grosser; Michael Kruse; Chandan Reddy; Sven Verdoolaege; Adam Betts; Alastair F. Donaldson; Jeroen Ketema; Javed Absar; Svenvan V. Haastregt; A. Kravets; Anton Lokhmotov; Robert David; Elnar Hajiyev

doi:10.1109/PACT.2015.17

PENCIL: A Platform-Neutral Compute Intermediate Language for Accelerator Programming

Riyadh Baghdadi, Ulysse Beaugnon, Albert Cohen, Tobias Grosser, Michael Kruse, Chandan Reddy, Sven Verdoolaege, Adam Betts, Alastair F. Donaldson, Jeroen Ketema, Javed Absar, Svenvan V. Haastregt, A. Kravets, Anton Lokhmotov, Robert David, Elnar Hajiyev

Computer Science

Research output: Contribution to journal › Conference article › peer-review

Abstract

Programming accelerators such as GPUs withlow-level APIs and languages such as OpenCL and CUDAis difficult, error-prone, and not performance-portable. Au-tomatic parallelization and domain specific languages (DSLs)have been proposed to hide complexity and regain performanceportability. We present P ENCIL, a rigorously-defined subset ofGNU C99 - enriched with additional language constructs - that enables compilers to exploit parallelism and produce highlyoptimized code when targeting accelerators. P ENCIL aims toserve both as a portable implementation language for libraries, and as a target language for DSL compilers. We implemented a P ENCIL-to-OpenCL backend using astate-of-the-art polyhedral compiler. The polyhedral compiler, extended to handle data-dependent control flow and non-affinearray accesses, generates optimized OpenCL code. To demon-strate the potential and performance portability of P ENCILand the P ENCIL-to-OpenCL compiler, we consider a numberof image processing kernels, a set of benchmarks from theRodinia and SHOC suites, and DSL embedding scenarios forlinear algebra (BLAS) and signal processing radar applications(SpearDE), and present experimental results for four GPUplatforms: AMD Radeon HD 5670 and R9 285, NVIDIAGTX 470, and ARM Mali-T604.

Original language	English (US)
Article number	7429301
Pages (from-to)	138-149
Number of pages	12
Journal	Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT
DOIs	https://doi.org/10.1109/PACT.2015.17
State	Published - 2015
Event	24th International Conference on Parallel Architecture and Compilation, PACT 2015 - San Francisco, United States Duration: Oct 18 2015 → Oct 21 2015

Keywords

Automatic optimization
Domain specific languages
Intermediate language
OpenCL
Polyhedral model

ASJC Scopus subject areas

Software
Theoretical Computer Science
Hardware and Architecture

Access to Document

10.1109/PACT.2015.17

Cite this

Baghdadi, R., Beaugnon, U., Cohen, A., Grosser, T., Kruse, M., Reddy, C., Verdoolaege, S., Betts, A., Donaldson, A. F., Ketema, J., Absar, J., Haastregt, S. V., Kravets, A., Lokhmotov, A., David, R., & Hajiyev, E. (2015). PENCIL: A Platform-Neutral Compute Intermediate Language for Accelerator Programming. Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT, 138-149. [7429301]. https://doi.org/10.1109/PACT.2015.17

Baghdadi, R, Beaugnon, U, Cohen, A, Grosser, T, Kruse, M, Reddy, C, Verdoolaege, S, Betts, A, Donaldson, AF, Ketema, J, Absar, J, Haastregt, SV, Kravets, A, Lokhmotov, A, David, R & Hajiyev, E 2015, 'PENCIL: A Platform-Neutral Compute Intermediate Language for Accelerator Programming', Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT, pp. 138-149. https://doi.org/10.1109/PACT.2015.17

@article{605769bdf50a43998dc57b29222a2d23,

title = "PENCIL: A Platform-Neutral Compute Intermediate Language for Accelerator Programming",

abstract = "Programming accelerators such as GPUs withlow-level APIs and languages such as OpenCL and CUDAis difficult, error-prone, and not performance-portable. Au-tomatic parallelization and domain specific languages (DSLs)have been proposed to hide complexity and regain performanceportability. We present P ENCIL, a rigorously-defined subset ofGNU C99 - enriched with additional language constructs - that enables compilers to exploit parallelism and produce highlyoptimized code when targeting accelerators. P ENCIL aims toserve both as a portable implementation language for libraries, and as a target language for DSL compilers. We implemented a P ENCIL-to-OpenCL backend using astate-of-the-art polyhedral compiler. The polyhedral compiler, extended to handle data-dependent control flow and non-affinearray accesses, generates optimized OpenCL code. To demon-strate the potential and performance portability of P ENCILand the P ENCIL-to-OpenCL compiler, we consider a numberof image processing kernels, a set of benchmarks from theRodinia and SHOC suites, and DSL embedding scenarios forlinear algebra (BLAS) and signal processing radar applications(SpearDE), and present experimental results for four GPUplatforms: AMD Radeon HD 5670 and R9 285, NVIDIAGTX 470, and ARM Mali-T604.",

keywords = "Automatic optimization, Domain specific languages, Intermediate language, OpenCL, Polyhedral model",

author = "Riyadh Baghdadi and Ulysse Beaugnon and Albert Cohen and Tobias Grosser and Michael Kruse and Chandan Reddy and Sven Verdoolaege and Adam Betts and Donaldson, {Alastair F.} and Jeroen Ketema and Javed Absar and Haastregt, {Svenvan V.} and A. Kravets and Anton Lokhmotov and Robert David and Elnar Hajiyev",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; 24th International Conference on Parallel Architecture and Compilation, PACT 2015 ; Conference date: 18-10-2015 Through 21-10-2015",

year = "2015",

doi = "10.1109/PACT.2015.17",

language = "English (US)",

pages = "138--149",

journal = "Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT",

issn = "1089-795X",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - PENCIL

T2 - 24th International Conference on Parallel Architecture and Compilation, PACT 2015

AU - Baghdadi, Riyadh

AU - Beaugnon, Ulysse

AU - Cohen, Albert

AU - Grosser, Tobias

AU - Kruse, Michael

AU - Reddy, Chandan

AU - Verdoolaege, Sven

AU - Betts, Adam

AU - Donaldson, Alastair F.

AU - Ketema, Jeroen

AU - Absar, Javed

AU - Haastregt, Svenvan V.

AU - Kravets, A.

AU - Lokhmotov, Anton

AU - David, Robert

AU - Hajiyev, Elnar

PY - 2015

Y1 - 2015

N2 - Programming accelerators such as GPUs withlow-level APIs and languages such as OpenCL and CUDAis difficult, error-prone, and not performance-portable. Au-tomatic parallelization and domain specific languages (DSLs)have been proposed to hide complexity and regain performanceportability. We present P ENCIL, a rigorously-defined subset ofGNU C99 - enriched with additional language constructs - that enables compilers to exploit parallelism and produce highlyoptimized code when targeting accelerators. P ENCIL aims toserve both as a portable implementation language for libraries, and as a target language for DSL compilers. We implemented a P ENCIL-to-OpenCL backend using astate-of-the-art polyhedral compiler. The polyhedral compiler, extended to handle data-dependent control flow and non-affinearray accesses, generates optimized OpenCL code. To demon-strate the potential and performance portability of P ENCILand the P ENCIL-to-OpenCL compiler, we consider a numberof image processing kernels, a set of benchmarks from theRodinia and SHOC suites, and DSL embedding scenarios forlinear algebra (BLAS) and signal processing radar applications(SpearDE), and present experimental results for four GPUplatforms: AMD Radeon HD 5670 and R9 285, NVIDIAGTX 470, and ARM Mali-T604.

AB - Programming accelerators such as GPUs withlow-level APIs and languages such as OpenCL and CUDAis difficult, error-prone, and not performance-portable. Au-tomatic parallelization and domain specific languages (DSLs)have been proposed to hide complexity and regain performanceportability. We present P ENCIL, a rigorously-defined subset ofGNU C99 - enriched with additional language constructs - that enables compilers to exploit parallelism and produce highlyoptimized code when targeting accelerators. P ENCIL aims toserve both as a portable implementation language for libraries, and as a target language for DSL compilers. We implemented a P ENCIL-to-OpenCL backend using astate-of-the-art polyhedral compiler. The polyhedral compiler, extended to handle data-dependent control flow and non-affinearray accesses, generates optimized OpenCL code. To demon-strate the potential and performance portability of P ENCILand the P ENCIL-to-OpenCL compiler, we consider a numberof image processing kernels, a set of benchmarks from theRodinia and SHOC suites, and DSL embedding scenarios forlinear algebra (BLAS) and signal processing radar applications(SpearDE), and present experimental results for four GPUplatforms: AMD Radeon HD 5670 and R9 285, NVIDIAGTX 470, and ARM Mali-T604.

KW - Automatic optimization

KW - Domain specific languages

KW - Intermediate language

KW - OpenCL

KW - Polyhedral model

UR - http://www.scopus.com/inward/record.url?scp=84975490487&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84975490487&partnerID=8YFLogxK

U2 - 10.1109/PACT.2015.17

DO - 10.1109/PACT.2015.17

M3 - Conference article

AN - SCOPUS:84975490487

SN - 1089-795X

SP - 138

EP - 149

JO - Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT

JF - Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT

M1 - 7429301

Y2 - 18 October 2015 through 21 October 2015

ER -

PENCIL: A Platform-Neutral Compute Intermediate Language for Accelerator Programming

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this