A Proposal for Loop-Transformation Pragmas

Kruse, Michael; Finkel, Hal

doi:10.1007/978-3-319-98521-3_3

Cited by 8 publications

(15 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar frameworks include Tiramisu [3], CHiLL [5], and for graph algorithms GraphIt [25]. Some frameworks also use C's pragmas and other annotations to create DSLs for manual loop optimisations [8,11,13]. None of these, however, provides neither the same flexibility nor do they allow to decompose existing efficient library routines into reusable components.…”

Section: Related Workmentioning

confidence: 99%

Fireiron: A Scheduling Language for High-Performance Linear Algebra on GPUs

Hagedorn¹,

Elliott²,

Barthels³

et al. 2020

Preprint

View full text Add to dashboard Cite

Achieving high-performance GPU kernels requires optimizing algorithm implementations to the targeted GPU architecture. It is of utmost importance to fully use the compute and memory hierarchy, as well as available specialised hardware.Currently, vendor libraries like cuBLAS and cuDNN provide the best performing implementations of GPU algorithms. However the task of the library programmer is incredibly challenging: for each provided algorithm, high-performance implementations have to be developed for all commonly used architectures, input sizes, and different storage formats. These implementations are generally provided as optimized assembly code because performance-critical architectural features are only exposed at this level. This prevents reuse between different implementations of even the same algorithm, as simple differences can have major effects on low-level implementation details.In this paper we introduce Fireiron, a DSL and compiler which allows the specification of high-performance GPU implementations as compositions of simple and reusable building blocks. We show how to use Fireiron to optimize matrix multiplication implementations, achieving performance matching hand-coded CUDA kernels, even when using specialised hardware such as NIVIDA Tensor Cores, and outperforming state-of-the-art implementations provided by cuBLAS by more than 2×.

show abstract

Section: Related Workmentioning

confidence: 99%

Fireiron: A Scheduling Language for High-Performance Linear Algebra on GPUs

Hagedorn¹,

Elliott²,

Barthels³

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…In contrast to our solution which requires a compiler extension, these are compatible with any compiler. Because of this lack of portability, we are working on standardizing loop transformation as part of OpenMP [7]. Unrolling and tiling will be part of OpenMP 5.1, with more intended to be added in OpenMP 6.0.…”

Section: Related Workmentioning

confidence: 99%

Autotuning Search Space for Loop Transformations

Kruse

Finkel

2020

Preprint

Self Cite

View full text Add to dashboard Cite

One of the challenges for optimizing compilers is to predict whether applying an optimization will improve its execution speed. Programmers may override the compiler's profitability heuristic using optimization directives such as pragmas in the source code. Machine learning in the form of autotuning can assist users in finding the best optimizations for each platform.In this paper we propose a loop transformation search space that takes the form of a tree, in contrast to previous approaches that usually use vector spaces to represent loop optimization configurations. We implemented a simple autotuner exploring the search space and applied it to a selected set of PolyBench kernels. While the autotuner is capable of representing every possible sequence of loop transformations and their relations, the results motivate the use of better search strategies such as Monte Carlo tree search to find sophisticated loop transformations such as multilevel tiling.

show abstract

“…Baghdadi et al proposed Tiramisu a polyhedral compiler that introduces novel commands to explicitly target distributed systems. More recently, Kruse et al submitted a proposal to improve pragma-based transformations in Clang [29]. All these tools expose some scheduling-based language to specify program transformations applicable to loops.…”

Section: Related Workmentioning

confidence: 99%

Declarative Loop Tactics for Domain-specific Optimization

Chelini

Zinenko

Grosser

et al. 2019

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

Increasingly complex hardware makes the design of effective compilers difficult. To reduce this problem, we introduce Declarative Loop Tactics, which is a novel framework of composable program transformations based on an internal tree-like program representation of a polyhedral compiler. The framework is based on a declarative C++ API built around easy-to-program matchers and builders, which provide the foundation to develop loop optimization strategies. Using our matchers and builders, we express computational patterns and core building blocks, such as loop tiling, fusion, and data-layout transformations, and compose them into algorithm-specific optimizations. Declarative Loop Tactics (Loop Tactics for short) can be applied to many domains. For two of them, stencils and linear algebra, we show how developers can express sophisticated domain-specific optimizations as a set of composable transformations or calls to optimized libraries. By allowing developers to add highly customized optimizations for a given computational pattern, we expect our approach to reduce the need for DSLs and to extend the range of optimizations that can be performed by a current general-purpose compiler. CCS Concepts: • Software and its engineering → Compilers;

show abstract

A Proposal for Loop-Transformation Pragmas

Cited by 8 publications

References 15 publications

Fireiron: A Scheduling Language for High-Performance Linear Algebra on GPUs

Fireiron: A Scheduling Language for High-Performance Linear Algebra on GPUs

Autotuning Search Space for Loop Transformations

Declarative Loop Tactics for Domain-specific Optimization

Contact Info

Product

Resources

About