https://github.com/inducer/loopy

A code generator for array-based code on CPUs and GPUs
https://github.com/inducer/loopy

array code-generation code-generator code-optimization code-transformation cuda ispc loop-optimization multidimensional-arrays opencl performance performance-analysis prefix-sum python reduction scan scientific-computing

Last synced: 6 days ago
JSON representation

A code generator for array-based code on CPUs and GPUs

• Host: GitHub
• URL: https://github.com/inducer/loopy
• Owner: inducer
• License: mit
• Created: 2014-05-29T04:03:35.000Z (over 10 years ago)
• Default Branch: main
• Last Pushed: 2025-01-16T19:28:01.000Z (12 days ago)
• Last Synced: 2025-01-22T07:37:08.090Z (7 days ago)
• Topics: array, code-generation, code-generator, code-optimization, code-transformation, cuda, ispc, loop-optimization, multidimensional-arrays, opencl, performance, performance-analysis, prefix-sum, python, reduction, scan, scientific-computing
• Language: Python
• Homepage: http://mathema.tician.de/software/loopy
• Size: 12.3 MB
• Stars: 596
• Watchers: 21
• Forks: 73
• Open Issues: 168
• Metadata Files:
  • Readme: README.rst
  • License: LICENSE
  • Citation: CITATION.cff

Awesome Lists containing this project

README

        
Loopy: Transformation-Based Generation of High-Performance CPU/GPU Code

=======================================================================

.. image:: https://gitlab.tiker.net/inducer/loopy/badges/main/pipeline.svg

    :alt: Gitlab Build Status

    :target: https://gitlab.tiker.net/inducer/loopy/commits/main

.. image:: https://github.com/inducer/loopy/workflows/CI/badge.svg?branch=main

    :alt: Github Build Status

    :target: https://github.com/inducer/loopy/actions?query=branch%3Amain+workflow%3ACI

.. image:: https://badge.fury.io/py/loopy.png

    :alt: Python Package Index Release Page

    :target: https://pypi.org/project/loopy/

.. image:: https://zenodo.org/badge/20281732.svg

    :alt: Zenodo DOI for latest release

    :target: https://zenodo.org/doi/10.5281/zenodo.10672274

Loopy lets you easily generate the tedious, complicated code that is necessary

to get good performance out of GPUs and multi-core CPUs.

Loopy's core idea is that a computation should be described simply and then

*transformed* into a version that gets high performance. This transformation

takes place under user control, from within Python.

It can capture the following types of optimizations:

* Vector and multi-core parallelism in the OpenCL/CUDA model

* Data layout transformations (structure of arrays to array of structures)

* Loop unrolling

* Loop tiling with efficient handling of boundary cases

* Prefetching/copy optimizations

* Instruction level parallelism

* and many more

Loopy targets array-type computations, such as the following:

* dense linear algebra,

* convolutions,

* n-body interactions,

* PDE solvers, such as finite element, finite difference, and

  Fast-Multipole-type computations

It is not (and does not want to be) a general-purpose programming language.

Loopy is licensed under the liberal `MIT license

`_ and free for commercial, academic,

and private use. All of Loopy's dependencies can be automatically installed from

the package index after using::

    pip install loopy

In addition, Loopy is compatible with and enhances

`pyopencl `_.

---

Places on the web related to Loopy:

* `Python package index `_ (download releases)

* `Documentation `_ (read how things work)

* `Github `_ (get latest source code, file bugs)

* `Homepage `_

* `Benchmarks `_