English
Related papers

Related papers: MOM: Matrix Operations in MLIR

200 papers

This work presents MLIR, a novel approach to building reusable and extensible compiler infrastructure. MLIR aims to address software fragmentation, improve compilation for heterogeneous hardware, significantly reduce the cost of building…

This article is primarily meant to present an early case study on using MLIR, a new compiler intermediate representation infrastructure, for high-performance code generation. Aspects of MLIR covered in particular include memrefs, the affine…

Performance · Computer Science 2020-03-03 Uday Bondhugula

Multi-level intermediate representations (MLIR) show great promise for reducing the cost of building domain-specific compilers by providing a reusable and extensible compiler infrastructure. This work presents TPU-MLIR, an end-to-end…

Programming Languages · Computer Science 2023-02-10 Pengchao Hu , Man Lu , Lei Wang , Guoyue Jiang

To take full advantage of a specific hardware target, performance engineers need to gain control on compilers in order to leverage their domain knowledge about the program and hardware. Yet, modern compilers are poorly controlled, usually…

Programming Languages · Computer Science 2024-09-10 Martin Paul Lücke , Oleksandr Zinenko , William S. Moses , Michel Steuwer , Albert Cohen

Traditional Digital Signal Processing ( DSP ) compilers work at low level ( C-level / assembly level ) and hence lose much of the optimization opportunities present at high-level ( domain-level ). The emerging multi-level compiler…

Signal Processing · Electrical Eng. & Systems 2025-06-23 Abhinav Kumar , Atharva Khedkar , Aviral Shrivastava

General-purpose compilers abstract away parallelism, locality, and synchronization, limiting their effectiveness on modern spatial architectures. As modern computing architectures increasingly rely on fine-grained control over data…

This report presents some early results on code generation targeting tensor cores on NVIDIA GPUs using the MLIR compiler infrastructure. The state-of-the-art in high-performance deep learning today is primarily driven by manually optimized…

Distributed, Parallel, and Cluster Computing · Computer Science 2021-08-31 Navdeep Katel , Vivek Khandelwal , Uday Bondhugula

Linear algebraic expressions are the essence of many computationally intensive problems, including scientific simulations and machine learning applications. However, translating high-level formulations of these expressions to efficient…

Distributed, Parallel, and Cluster Computing · Computer Science 2019-03-22 Dániel Berényi , András Leitereg , Gábor Lehel

Deep neural network models are becoming increasingly popular and have been used in various tasks such as computer vision, speech recognition, and natural language processing. Machine learning models are commonly trained in a resource-rich…

This work proposes a compilation flow using open-source compiler passes to build a framework to achieve ninja performance from a generic linear algebra high-level abstraction. We demonstrate this flow with a proof-of-concept MLIR project…

Driven by increasing compute requirements for deep learning models, compiler developers have been looking for ways to target specialised hardware and heterogeneous systems more efficiently. The MLIR project has the goal to offer…

Programming Languages · Computer Science 2025-03-10 Jules Merckx

Despite significant investment in software infrastructure, machine learning systems, runtimes and compilers do not compose properly. We propose a new design aiming at providing unprecedented degrees of modularity, composability and…

In the era of diminishing returns from Moores Law, heterogeneous computing systems have emerged as a vital approach to enhance computational efficiency. This paper introduces a novel MLIR-based dialect, named hyper, designed to optimize…

Cryptography and Security · Computer Science 2025-06-05 Zhiyuan Tan , Liutong Han , Mingjie Xing , Yanjun Wu

Traditional compilers operate on a single generic intermediate representation (IR). These IRs are usually low-level and close to machine instructions. As a result, optimizations relying on domain-specific information are either not possible…

MLIR is an emerging compiler infrastructure for modern hardware, but existing programs cannot take advantage of MLIR's high-performance compilation if they are described in lower-level general purpose languages. Consequently, to avoid…

Programming Languages · Computer Science 2023-10-09 Alexander Brauckmann , Elizabeth Polgreen , Tobias Grosser , Michael F. P. O'Boyle

We present a prototypical linear algebra compiler that automatically exploits domain-specific knowledge to generate high-performance algorithms. The input to the compiler is a target equation together with knowledge of both the structure of…

Mathematical Software · Computer Science 2012-05-29 Diego Fabregat-Traver , Paolo Bientinesi

Similar to other programming models, compilers for SYCL, the open programming model for heterogeneous computing based on C++, would benefit from access to higher-level intermediate representations. The loss of high-level structure and…

Programming Languages · Computer Science 2023-12-21 Ettore Tiotto , Víctor Pérez , Whitney Tsang , Lukas Sommer , Julian Oppermann , Victor Lomüller , Mehdi Goli , James Brodman

Multi-Level Intermediate Representation (MLIR) is gaining increasing attention in reconfigurable hardware communities due to its capability to represent various abstract levels for software compilers. This project aims to be the first to…

Hardware Architecture · Computer Science 2024-01-22 Zhenya Zang , Uwe Dolinsky , Pietro Ghiglio , Stefano Cherubin , Mehdi Goli , Shufan Yang

Optimizing compilers are essential for the efficient and correct execution of software across various scientific fields. Domain-specific languages (DSL) typically use higher level intermediate representations (IR) in their compiler…

Programming Languages · Computer Science 2026-01-15 Berke Ates , Philipp Schaad , Timo Schneider , Alexandru Calotoiu , Torsten Hoefler

The emergence of machine learning, image and audio processing on edge devices has motivated research towards power efficient custom hardware accelerators. Though FPGAs are an ideal target for energy efficient custom accelerators, the…

Hardware Architecture · Computer Science 2021-03-02 Kingshuk Majumder , Uday Bondhugula
‹ Prev 1 2 3 10 Next ›