Modern heterogeneous platforms require compilers capable of choosing the appropriate device for the execution of program portions. This paper presents a machine learning method designed for supporting mapping decisions through the analysis of the program source code represented in LLVM assembly language (IR) for exploiting the advantages offered by this generalised and optimised representation. To evaluate our solution, we trained an LSTM neural network on OpenCL kernels compiled in LLVM-IR and processed with our tokenizer capable of filtering less-informative tokens. We tested the network that reaches an accuracy of 85% in distinguishing the best computational unit.
Code Mapping in Heterogeneous Platforms Using Deep Learning and LLVM-IR / Barchi, Francesco; Urgese, Gianvito; Macii, Enrico; Acquaviva, Andrea. - ELETTRONICO. - (2019). (Intervento presentato al convegno In 2019 56th ACM/ESDA/IEEE Design Automation Conference (DAC) tenutosi a Las Vegas USA nel 2-6 June 2019) [10.1145/3316781.3317789].
Code Mapping in Heterogeneous Platforms Using Deep Learning and LLVM-IR
Francesco Barchi;Gianvito Urgese;Enrico Macii;Andrea Acquaviva
2019
Abstract
Modern heterogeneous platforms require compilers capable of choosing the appropriate device for the execution of program portions. This paper presents a machine learning method designed for supporting mapping decisions through the analysis of the program source code represented in LLVM assembly language (IR) for exploiting the advantages offered by this generalised and optimised representation. To evaluate our solution, we trained an LSTM neural network on OpenCL kernels compiled in LLVM-IR and processed with our tokenizer capable of filtering less-informative tokens. We tested the network that reaches an accuracy of 85% in distinguishing the best computational unit.File | Dimensione | Formato | |
---|---|---|---|
08806950_post_print_editor.pdf
non disponibili
Descrizione: Articolo principale
Tipologia:
2a Post-print versione editoriale / Version of Record
Licenza:
Non Pubblico - Accesso privato/ristretto
Dimensione
684.75 kB
Formato
Adobe PDF
|
684.75 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
document_post_print.pdf
accesso aperto
Descrizione: Articolo principale
Tipologia:
2. Post-print / Author's Accepted Manuscript
Licenza:
PUBBLICO - Tutti i diritti riservati
Dimensione
376.18 kB
Formato
Adobe PDF
|
376.18 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/2726074