Software:MLIR (Compiler)
MLIR[1] is a compiler framework enabling software developer to construct the customized compiler to lower the any supporting programming language without sacrificing the high level semantics. MLIR is initially launched as part of TensorFlow project and transferred to LLVM project 2021 [2]. MLIR accomplished extensibility by employing the plugin architecture called Dialect[3] . Software developers are able to extend the syntax, pass and even type system by implementing custom dialect.
History
April, 2018
Chris Lattner started the project in the compiler team at Google.[4] Initially the project was motivated to solve wide-range of graph-level problems observed in TensorFlow.
April, 2019
Google has made MLIR public as a open source software.[5]
September, 2019
Google contributed the project to non-profile LLVM foundation. [6]
May, 2021
The project team constructed YouTube channel showing the MLIR project activity. [7]
February, 2021
MLIR paper has been published in 2021 IEEE/ACM International Symposium on Code Generation and Optimization. [8]
References
- ↑ "MLIR". https://mlir.llvm.org/.
- ↑ Lattner, Chris; Amini, Mehdi; Bondhugula, Uday; Cohen, Albert; Davis, Andy; Pienaar, Jacques; Riddle, River; Shpeisman, Tatiana et al. (2020-02-29). "MLIR: A Compiler Infrastructure for the End of Moore's Law". arXiv:2002.11054 [cs]. http://arxiv.org/abs/2002.11054.
- ↑ "Dialects - MLIR". https://mlir.llvm.org/docs/Dialects/.
- ↑ "Chris Lattner's Homepage". http://www.nondot.org/sabre/.
- ↑ "MLIR: A new intermediate representation and compiler framework" (in en). https://blog.tensorflow.org/2019/04/mlir-new-intermediate-representation.html.
- ↑ "MLIR: accelerating AI with open-source infrastructure" (in en-us). 2019-09-09. https://blog.google/technology/ai/mlir-accelerating-ai-open-source-infrastructure/.
- ↑ "MLIR - YouTube". https://www.youtube.com/MLIRCompiler.
- ↑ Lattner, Chris; Amini, Mehdi; Bondhugula, Uday; Cohen, Albert; Davis, Andy; Pienaar, Jacques; Riddle, River; Shpeisman, Tatiana et al. (February 2021). "MLIR: Scaling Compiler Infrastructure for Domain Specific Computation". 2021 IEEE/ACM International Symposium on Code Generation and Optimization (CGO): 2–14. doi:10.1109/CGO51591.2021.9370308. https://ieeexplore.ieee.org/abstract/document/9370308.
 |  |
