Skip to main content
Fachgebiete
Automobil + Motoren Bauwesen + Immobilien Business IT + Informatik Elektrotechnik + Elektronik Energie + Nachhaltigkeit Finance + Banking Management + Führung Marketing + Vertrieb Maschinenbau + Werkstoffe Versicherung + Risiko
DE
EN
Themenseiten
Organisationspsychologie Projektmanagement Marketing Smart Manufacturing
Bücher
Zeitschriften
Weitere Formate
Podcasts Webinare Technik Webinare Wirtschaft Kongresse Veranstaltungskalender Awards
Jetzt Einzelzugang starten
Zugang für Unternehmen
Referenzkunden
MTZ-Fachkongress

Antriebe und Energiesysteme von morgen


Austausch von Automobil- und Energiewirtschaft

Am 14./15. Mai geht es in Chemnitz um die Mobilitätswende durch Elektro- und Wasserstofffahrzeuge.
Erweiterte Suche
Anmelden
nach oben
Erschienen in:
The Impact of Taskyield on the Design of Tasks Communicating Through MPI Kapitel lesen Erstes Kapitel lesen

2018 | OriginalPaper | Buchkapitel

Compiler Optimizations for OpenMP

verfasst von : Johannes Doerfert, Hal Finkel

Erschienen in: Evolving OpenMP for Evolving Architectures

Verlag: Springer International Publishing

Einloggen

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config
loading …

Abstract

Modern compilers support OpenMP as a convenient way to introduce parallelism into sequential languages like C/C++ and Fortran, however, its use also introduces immediate drawbacks. In many implementations, due to early outlining and the indirection though the OpenMP runtime, the front-end creates optimization barriers that are impossible to overcome by standard middle-end compiler passes. As a consequence, the OpenMP-annotated program constructs prevent various classic compiler transformations like constant propagation and loop invariant code motion. In addition, analysis results, especially alias information, is severely degraded in the presence of OpenMP constructs which can severely hurt performance.
In this work we investigate to what degree OpenMP runtime aware compiler optimizations can mitigate these problems. We discuss several transformations that explicitly change the OpenMP enriched compiler intermediate representation. They act as stand-alone optimizations but also enable existing optimizations that were not applicable before. This is all done in the existing LLVM/Clang compiler toolchain without introducing a new parallel representation. Our optimizations do not only improve the execution time of OpenMP annotated programs but also help to determine the caveats for transformations on the current representation of OpenMP.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

  • über 102.000 Bücher
  • über 537 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Maschinenbau + Werkstoffe
  • Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 390 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Maschinenbau + Werkstoffe




 

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 340 Zeitschriften

aus folgenden Fachgebieten:

  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Versicherung + Risiko




Jetzt Wissensvorsprung sichern!

Jetzt informieren
Vorheriges Kapitel Trade-Off of Offloading to FPGA in OpenMP Task-Based Programming
Nächstes Kapitel Supporting Function Variants in OpenMP
Fußnoten
1
A pointer is captured if a copy of it is made inside the callee that might outlive it.
 
2
The kmpc OpenMP library used by LLVM/Clang communicates variables via variadic functions that require the arguments to be in integer registers. When a floating point variable is communicated by-value instead of by-reference we have to insert code that moves the value from a floating point register into an integer register prior to the runtime library call and back inside the parallel function.
 
3
Communication through the runtime library involves multiple memory operations per variable and it is thereby easily more expensive than one addition for each thread.
 
4
We need to ensure that https://static-content.springer.com/image/chp%3A10.1007%2F978-3-319-98521-3_8/MediaObjects/472218_1_En_8_Figal_HTML.gif is only executed under the condition https://static-content.springer.com/image/chp%3A10.1007%2F978-3-319-98521-3_8/MediaObjects/472218_1_En_8_Figam_HTML.gif
 
5
The nodes for https://static-content.springer.com/image/chp%3A10.1007%2F978-3-319-98521-3_8/MediaObjects/472218_1_En_8_Figap_HTML.gif are omitted for space reasons. They would look similar to the ones for https://static-content.springer.com/image/chp%3A10.1007%2F978-3-319-98521-3_8/MediaObjects/472218_1_En_8_Figaq_HTML.gif , though not only allow \(c_\omega \) flow to the sink but also into the incoming node of  https://static-content.springer.com/image/chp%3A10.1007%2F978-3-319-98521-3_8/MediaObjects/472218_1_En_8_Figar_HTML.gif .
 
Literatur
1.
Agarwal, S., Barik, R., Sarkar, V., Shyamasundar, R.K.: May-happen-in-parallel analysis of X10 programs. In: Proceedings of the 12th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, PPOPP 2007, San Jose, California, USA, 14–17 March 2007, pp. 183–193 (2007). https://​doi.​org/​10.​1145/​1229428.​1229471
2.
Barik, R., Sarkar, V.: Interprocedural load elimination for dynamic optimization of parallel programs. In: Proceedings of the 18th International Conference on Parallel Architectures and Compilation Techniques, PACT 2009, 12–16 September 2009, Raleigh, North Carolina, USA, pp. 41–52 (2009). https://​doi.​org/​10.​1109/​PACT.​2009.​32
3.
Barik, R., Zhao, J., Sarkar, V.: Interprocedural strength reduction of critical sections in explicitly-parallel programs. In: Proceedings of the 22nd International Conference on Parallel Architectures and Compilation Techniques, Edinburgh, United Kingdom, 7–11 September 2013, pp. 29–40 (2013). https://​doi.​org/​10.​1109/​PACT.​2013.​6618801
4.
5.
Grunwald, D., Srinivasan, H.: Data flow equations for explicitly parallel programs. In: Proceedings of the Fourth ACM SIGPLAN Symposium on Principles & Practice of Parallel Programming (PPOPP), San Diego, California, USA, 19–22 May 1993, pp. 159–168 (1993). https://​doi.​org/​10.​1145/​155332.​155349
6.
Jordan, H., Pellegrini, S., Thoman, P., Kofler, K., Fahringer, T.: INSPIRE: the insieme parallel intermediate representation. In: Proceedings of the 22nd International Conference on Parallel Architectures and Compilation Techniques, Edinburgh, United Kingdom, 7–11 September 2013, pp. 7–17 (2013). https://​doi.​org/​10.​1109/​PACT.​2013.​6618799
7.
Karlin, I., et al.: LULESH programming model and performance ports overview. Technical report LLNL-TR-608824, December 2012
8.
Khaldi, D., Jouvelot, P., Irigoin, F., Ancourt, C., Chapman, B.M.: LLVM parallel intermediate representation: design and evaluation using OpenSHMEM communications. In: Proceedings of the Second Workshop on the LLVM Compiler Infrastructure in HPC, LLVM 2015, Austin, Texas, USA, 15 November 2015, pp. 2:1–2:8 (2015). https://​doi.​org/​10.​1145/​2833157.​2833158
9.
Lattner, C., Adve, V.S.: LLVM: a compilation framework for lifelong program analysis & transformation. In: 2nd IEEE/ACM International Symposium on Code Generation and Optimization (CGO 2004), 20–24 March 2004, San Jose, CA, USA, pp. 75–88 (2004). https://​doi.​org/​10.​1109/​CGO.​2004.​1281665
10.
11.
Schardl, T.B., Moses, W.S., Leiserson, C.E.: Tapir: embedding fork-join parallelism into LLVM’s intermediate representation. In: Proceedings of the 22nd ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, Austin, TX, USA, 4–8 February 2017, pp. 249–265 (2017). http://​dl.​acm.​org/​citation.​cfm?​id=​3018758
12.
13.
Tian, X., Girkar, M., Shah, S., Armstrong, D., Su, E., Petersen, P.: Compiler and runtime support for running OpenMP programs on pentium-and itanium-architectures. In: Eighth International Workshop on High-Level Parallel Programming Models and Supportive Environments (HIPS 2003), 22 April 2003, Nice, France, pp. 47–55 (2003). https://​doi.​org/​10.​1109/​HIPS.​2003.​1196494
14.
Tian, X., et al.: LLVM framework and IR extensions for parallelization, SIMD vectorization and offloading. In: Third Workshop on the LLVM Compiler Infrastructure in HPC, LLVM-HPC@SC 2016, Salt Lake City, UT, USA, 14 November 2016, pp. 21–31 (2016). https://​doi.​org/​10.​1109/​LLVM-HPC.​2016.​008
15.
Zhao, J., Sarkar, V.: Intermediate language extensions for parallelism. In: Conference on Systems, Programming, and Applications: Software for Humanity, SPLASH 2011, Proceedings of the Compilation of the Co-located Workshops, DSM 2011, TMC 2011, AGERE! 2011, AOOPES 2011, NEAT 2011, and VMIL 2011, 22–27 October 2011, Portland, OR, USA, pp. 329–340 (2011). https://​doi.​org/​10.​1145/​2095050.​2095103
Metadaten
Titel
Compiler Optimizations for OpenMP
verfasst von
Johannes Doerfert
Hal Finkel
Copyright-Jahr
2018
Buch
Evolving OpenMP for Evolving Architectures

Print ISBN: 978-3-319-98520-6

Electronic ISBN: 978-3-319-98521-3

Copyright-Jahr: 2018

DOI
https://doi.org/10.1007/978-3-319-98521-3_8

Neuer Inhalt

    Bildnachweise