2019 EuroLLVM Developers’ Meeting: T. Shpeisman & C. Lattner “MLIR: Multi-Level Intermediate Repr..”
http://llvm.org/devmtg/2019-04/
—
MLIR: Multi-Level Intermediate Representation for Compiler Infrastructure - Tatiana Shpeisman (Google), Chris Lattner (Google)
Slides: http://llvm.org/devmtg/2019-04/slides/Keynote-ShpeismanLattner-MLIR.pdf
—
This talk will give an overview of Multi-Level Intermediate Representation - a new intermediate representation designed to provide a unified, flexible and extensible intermediate representation that is language-agnostic and can be used as a base compiler infrastructure. MLIR shares similarities with traditional CFG-based three-address SSA representations (including LLVM IR or SIL), but it also introduces notions from the polyhedral domain as first class concepts. The notion of dialects is a core concept of MLIR extensibility, allowing multiple levels in a single representation. MLIR supports the continuous lowering from dataflow graphs to high-performance target specific code through partial specialization between dialects. We will illustrate in this talk how MLIR can be used to build an optimizing compiler infrastructure for deep learning applications.
MLIR supports multiple front- and back-ends and uses LLVM IR as one of its primary code generation targets. MLIR also relies heavily on design principles and practices developed by the LLVM community. For example, it depends on LLVM APIs and programming idioms to minimize IR size and maximize optimization efficiency. MLIR uses LLVM testing utilities such as FileCheck to ensure robust functionality at every level of the compilation stack, TableGen to express IR invariants, and it leverages LLVM infrastructure such as dominance analysis to avoid implementing all the necessary compiler functionalities from scratch. At the same time, it is a brand new IR, both more restrictive and more general than LLVM IR in different aspects of its design. We believe that the LLVM community will find in MLIR a useful tool for developing new compilers, especially in machine learning and other high-performance domains.
—
Videos Filmed & Edited by Bash Films: http://www.BashFilms.com
Видео 2019 EuroLLVM Developers’ Meeting: T. Shpeisman & C. Lattner “MLIR: Multi-Level Intermediate Repr..” канала LLVM
—
MLIR: Multi-Level Intermediate Representation for Compiler Infrastructure - Tatiana Shpeisman (Google), Chris Lattner (Google)
Slides: http://llvm.org/devmtg/2019-04/slides/Keynote-ShpeismanLattner-MLIR.pdf
—
This talk will give an overview of Multi-Level Intermediate Representation - a new intermediate representation designed to provide a unified, flexible and extensible intermediate representation that is language-agnostic and can be used as a base compiler infrastructure. MLIR shares similarities with traditional CFG-based three-address SSA representations (including LLVM IR or SIL), but it also introduces notions from the polyhedral domain as first class concepts. The notion of dialects is a core concept of MLIR extensibility, allowing multiple levels in a single representation. MLIR supports the continuous lowering from dataflow graphs to high-performance target specific code through partial specialization between dialects. We will illustrate in this talk how MLIR can be used to build an optimizing compiler infrastructure for deep learning applications.
MLIR supports multiple front- and back-ends and uses LLVM IR as one of its primary code generation targets. MLIR also relies heavily on design principles and practices developed by the LLVM community. For example, it depends on LLVM APIs and programming idioms to minimize IR size and maximize optimization efficiency. MLIR uses LLVM testing utilities such as FileCheck to ensure robust functionality at every level of the compilation stack, TableGen to express IR invariants, and it leverages LLVM infrastructure such as dominance analysis to avoid implementing all the necessary compiler functionalities from scratch. At the same time, it is a brand new IR, both more restrictive and more general than LLVM IR in different aspects of its design. We believe that the LLVM community will find in MLIR a useful tool for developing new compilers, especially in machine learning and other high-performance domains.
—
Videos Filmed & Edited by Bash Films: http://www.BashFilms.com
Видео 2019 EuroLLVM Developers’ Meeting: T. Shpeisman & C. Lattner “MLIR: Multi-Level Intermediate Repr..” канала LLVM
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
2018 LLVM Developers’ Meeting: L. Hames & B. Loggins “Updating ORC JIT for Concurrency”
2018 LLVM Developers’ Meeting: J. Bastien “Migrating to C++14, and beyond!”
2020 LLVM Developers’ Meeting: I. Yakubova “Adding a Subtarget Support to LLVM in Five Minutes”
2017 LLVM Developers’ Meeting: N. Hawes & A. Lorenz “Adding Index‐While‐Building and ..."
2020 LLVM Developers’ Meeting: P. Reames “An Update on Optimizing Multiple Exit Loops”
2018 LLVM Developers’ Meeting: M. Gadelha “Refuting False Bugs in the Clang Static Analyzer ...”
2017 LLVM Developers’ Meeting: “GlobalISel: Past, Present, and Future ”
2020 LLVM Developers’ Meeting: “(OpenMP) Parallelism-Aware Optimizations”
2016 LLVM Developers’ Meeting: P. Padlewski “Devirtualization in LLVM”
07 An Anatomy of Optimized Matrix Multiplication on AArch64
2022 LLVM Dev Mtg: MLIR for Functional Programming
2012 EuroLLVM Developers’ Meeting: L. Smith “Introduction”
2020 LLVM in HPC Workshop: Autotuning Search Space for Loop Transformations
2019 LLVM Developers’ Meeting: O. Cazalet-Hyams “Improving the Optimized Debugging Experience”
2022 EuroLLVM Dev Mtg “Developing an LLVM backend for the KV3 Kalray VLIW core”
2019 LLVM Developers’ Meeting: J. Doerfert “The Attributor: A Versatile Inter-procedural Fixpoint..”
2018 LLVM Developers’ Meeting: S. Moll “Stories from RV: The LLVM vectorization ecosystem ”
2020 LLVM Developers’ Meeting: E. Stepanov “Memory tagging in LLVM and Android”
2016 EuroLLVM Developers' Meeting: Kristof Beyls "Towards ameliorating measurement bias ..."
2019 LLVM Developers’ Meeting: V. Keles & D. Sanders “Generating Optimized Code with GlobalISel”
2019 LLVM Developers’ Meeting: S. Tallam “Propeller: Profile Guided Large Scale Performance...”