llvmlite

lightweight wrapper around basic LLVM functionality

Project description

A lightweight LLVM python binding for writing JIT compilers

The old llvmpy binding exposes a lot of LLVM APIs but the mapping of C++-style memory management to Python is error prone. Numba and many JIT compilers do not need a full LLVM API. Only the IR builder, optimizer, and JIT compiler APIs are necessary.

llvmlite is a project originally tailored for Numba’s needs, using the following approach:

A small C wrapper around the parts of the LLVM C++ API we need that are not already exposed by the LLVM C API.
A ctypes Python wrapper around the C API.
A pure Python implementation of the subset of the LLVM IR builder that we need for Numba.

Key Benefits

The IR builder is pure Python code and decoupled from LLVM’s frequently-changing C++ APIs.
Materializing a LLVM module calls LLVM’s IR parser which provides better error messages than step-by-step IR building through the C++ API (no more segfaults or process aborts).
Most of llvmlite uses the LLVM C API which is small but very stable (low maintenance when changing LLVM version).
The binding is not a Python C-extension, but a plain DLL accessed using ctypes (no need to wrestle with Python’s compiler requirements and C++ 11 compatibility).
The Python binding layer has sane memory management.
llvmlite is quite faster than llvmpy’s thanks to a much simpler architeture (the Numba test suite is twice faster than it was).

llvmpy Compatibility Layer

The llvmlite.llvmpy namespace provides a minimal llvmpy compatibility layer.

Compatibility

llvmlite works with Python 3.6 and greater.

As of version 0.34.0, llvmlite requires LLVM 10.0.x. on all architectures except aarch64 which requires LLVM 9.0.x due to: https://bugs.llvm.org/show_bug.cgi?id=46996

Historical compatibility table:

llvmlite versions	compatible LLVM versions
0.34.0 - …	10.0.x (9.0.x for `aarch64` only)
0.33.0	9.0.x
0.29.0 - 0.32.0	7.0.x, 7.1.x, 8.0.x
0.27.0 - 0.28.0	7.0.x
0.23.0 - 0.26.0	6.0.x
0.21.0 - 0.22.0	5.0.x
0.17.0 - 0.20.0	4.0.x
0.16.0 - 0.17.0	3.9.x
0.13.0 - 0.15.0	3.8.x
0.9.0 - 0.12.1	3.7.x
0.6.0 - 0.8.0	3.6.x
0.1.0 - 0.5.1	3.5.x

Documentation

You’ll find the documentation at http://llvmlite.pydata.org

Pre-built binaries

We recommend you use the binaries provided by the Numba team for the Conda package manager. You can find them in Numba’s anaconda.org channel. For example:

$ conda install --channel=numba llvmlite

(or, simply, the official llvmlite package provided in the Anaconda distribution)

Other build methods

If you don’t want to use our pre-built packages, you can compile and install llvmlite yourself. The documentation will teach you how: http://llvmlite.pydata.org/en/latest/install/index.html

Project details

Release history Release notifications | RSS feed

This version

0.34.0

Aug 13, 2020

0.34.0rc1 pre-release

Aug 6, 2020

0.33.0

Jun 11, 2020

0.33.0rc1 pre-release

Jun 3, 2020

0.32.1

May 8, 2020

0.32.0

Apr 17, 2020

0.31.0

Jan 3, 2020

0.30.0

Oct 10, 2019

0.29.0

May 30, 2019

0.28.0

Mar 13, 2019

0.27.1

Feb 1, 2019

0.27.0

Dec 31, 2018

0.26.0

Nov 28, 2018

0.25.0

Sep 21, 2018

0.24.0

Jul 10, 2018

0.23.2

Jun 1, 2018

0.23.0

Apr 25, 2018

0.22.0

Feb 19, 2018

0.21.0

Dec 8, 2017

0.20.0

Sep 8, 2017

0.19.0

Jul 7, 2017

0.18.0

May 4, 2017

0.17.1

Apr 12, 2017

0.17.0

Apr 10, 2017

0.16.0

Feb 17, 2017

0.15.0

Dec 22, 2016

0.14.0

Oct 18, 2016

0.13.0

Aug 24, 2016

0.12.1

Jul 8, 2016

0.12.0

Jul 6, 2016

0.11.0

May 24, 2016

0.10.0

Mar 31, 2016

0.9.0

Feb 29, 2016

0.8.0

Oct 23, 2015

0.7.0

Aug 31, 2015

0.6.0

Jul 2, 2015

0.5.0

Jun 5, 2015

0.4.0

Apr 13, 2015

0.2.2

Jan 29, 2015

0.2.1

Dec 19, 2014

0.2.0

Dec 17, 2014

0.1

Oct 28, 2014

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for llvmlite, version 0.34.0
Filename, size	File type	Python version	Upload date	Hashes
Filename, size llvmlite-0.34.0-cp36-cp36m-macosx_10_9_x86_64.whl (18.4 MB)	File type Wheel	Python version cp36	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0-cp36-cp36m-manylinux2010_i686.whl (26.7 MB)	File type Wheel	Python version cp36	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0-cp36-cp36m-manylinux2010_x86_64.whl (24.6 MB)	File type Wheel	Python version cp36	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0-cp36-cp36m-win32.whl (12.9 MB)	File type Wheel	Python version cp36	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0-cp36-cp36m-win_amd64.whl (15.9 MB)	File type Wheel	Python version cp36	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0-cp37-cp37m-macosx_10_9_x86_64.whl (18.4 MB)	File type Wheel	Python version cp37	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0-cp37-cp37m-manylinux2010_i686.whl (26.7 MB)	File type Wheel	Python version cp37	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0-cp37-cp37m-manylinux2010_x86_64.whl (24.6 MB)	File type Wheel	Python version cp37	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0-cp37-cp37m-win32.whl (12.9 MB)	File type Wheel	Python version cp37	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0-cp37-cp37m-win_amd64.whl (15.9 MB)	File type Wheel	Python version cp37	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0-cp38-cp38-macosx_10_9_x86_64.whl (18.4 MB)	File type Wheel	Python version cp38	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0-cp38-cp38-manylinux2010_i686.whl (26.7 MB)	File type Wheel	Python version cp38	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0-cp38-cp38-manylinux2010_x86_64.whl (24.6 MB)	File type Wheel	Python version cp38	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0-cp38-cp38-win32.whl (12.9 MB)	File type Wheel	Python version cp38	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0-cp38-cp38-win_amd64.whl (15.9 MB)	File type Wheel	Python version cp38	Upload date Aug 13, 2020	Hashes View
Filename, size llvmlite-0.34.0.tar.gz (107.7 kB)	File type Source	Python version None	Upload date Aug 13, 2020	Hashes View

Hashes for llvmlite-0.34.0-cp36-cp36m-macosx_10_9_x86_64.whl

Hashes for llvmlite-0.34.0-cp36-cp36m-macosx_10_9_x86_64.whl
Algorithm	Hash digest
SHA256	`11342e5ac320c953590bdd9d0dec8c52f4b5252c4c6335ba25f1e7b9f91f9325`
MD5	`b98c64f3a3e22e4da6260e345a9429be`
BLAKE2-256	`22e06a2af71fcbc4e269964c7b1ed92a8c80e71367dc0f4e12dd35139c0c6925`

Hashes for llvmlite-0.34.0-cp36-cp36m-manylinux2010_i686.whl

Hashes for llvmlite-0.34.0-cp36-cp36m-manylinux2010_i686.whl
Algorithm	Hash digest
SHA256	`5bdf0ce430adfaf938ced5844d12f80616eb8321b5b9edfc45ef84ada5c5242c`
MD5	`4dd14282305646b15dd150b839f9178e`
BLAKE2-256	`a076403c94e21c3c0f23b0a9fe1441a9190948f8937b341a65aa830d9e4cd9b3`

Hashes for llvmlite-0.34.0-cp36-cp36m-manylinux2010_x86_64.whl

Hashes for llvmlite-0.34.0-cp36-cp36m-manylinux2010_x86_64.whl
Algorithm	Hash digest
SHA256	`e08d9d2dc5a31636bfc6b516d2d7daba95632afa3419eb8730dc76a7951e9558`
MD5	`7c22fda0b01bf96f6dbc39f03abf7da9`
BLAKE2-256	`f3b78a91b513f165e0affdeb975c1fef307c39d1051ce71e8aec1da9dcb317ad`

Hashes for llvmlite-0.34.0-cp36-cp36m-win32.whl

Hashes for llvmlite-0.34.0-cp36-cp36m-win32.whl
Algorithm	Hash digest
SHA256	`9ff1dcdad03be0cf953aca5fc8cffdca25ccee2ec9e8ec7e95571722cdc02d55`
MD5	`df6e47991ed5d23a420bcfc804e124e1`
BLAKE2-256	`76a481c31edd1304909c77058dc35fa964c5a89d1c87aa55f07d59dbce056497`

Hashes for llvmlite-0.34.0-cp36-cp36m-win_amd64.whl

Hashes for llvmlite-0.34.0-cp36-cp36m-win_amd64.whl
Algorithm	Hash digest
SHA256	`5acdc3c3c7ea0ef7a1a6b442272e05d695bc8492e5b07666135ed1cfbf4ab9d2`
MD5	`b6686519f654d4d91ee9e0ec94aef0fa`
BLAKE2-256	`b8b0f69cb0dc1cff13e6b75ea1190577d590d8d8e2bd3fe803ac45235d4b1ec8`

Hashes for llvmlite-0.34.0-cp37-cp37m-macosx_10_9_x86_64.whl

Hashes for llvmlite-0.34.0-cp37-cp37m-macosx_10_9_x86_64.whl
Algorithm	Hash digest
SHA256	`bb96989bc57a1ccb131e7a0e061d07b68139b6f81a98912345d53d9239e231e1`
MD5	`bb6822c681eebbdd6e4c48e3d22297e0`
BLAKE2-256	`371c18ab3765b1c60981ed636e7f851c660c3d808f0d3a5870a215f1c74176b2`

Hashes for llvmlite-0.34.0-cp37-cp37m-manylinux2010_i686.whl

Hashes for llvmlite-0.34.0-cp37-cp37m-manylinux2010_i686.whl
Algorithm	Hash digest
SHA256	`6d3f81992f52a94077e7b9b16497029daf5b5eebb2cce56f3c8345bbc9c6308e`
MD5	`7f4eaee2fd57857a92594308c293e469`
BLAKE2-256	`f6f780d93e610f27549adff84b081a577aea43513cd43b87fded21665c8814e7`

Hashes for llvmlite-0.34.0-cp37-cp37m-manylinux2010_x86_64.whl

Hashes for llvmlite-0.34.0-cp37-cp37m-manylinux2010_x86_64.whl
Algorithm	Hash digest
SHA256	`d841248d1c630426c93e3eb3f8c45bca0dab77c09faeb7553b1a500220e362ce`
MD5	`093db6dc4b521cf538da73bffd84bdda`
BLAKE2-256	`4bdf88c6d62e39947c5b1f2b1eb2239d1aa3013d6a61a75730f8b8997b6e66c6`

Hashes for llvmlite-0.34.0-cp37-cp37m-win32.whl

Hashes for llvmlite-0.34.0-cp37-cp37m-win32.whl
Algorithm	Hash digest
SHA256	`408b15ffec30696406e821c89da010f1bb1eb0aa572be4561c98eb2536d610ab`
MD5	`0c7bc7ae20806b1925ba7f1e6aeff67f`
BLAKE2-256	`cf8da698cc9a76ba7b4a0c6ef7e9fa349548d87b46938037c8aac99169849171`

Hashes for llvmlite-0.34.0-cp37-cp37m-win_amd64.whl

Hashes for llvmlite-0.34.0-cp37-cp37m-win_amd64.whl
Algorithm	Hash digest
SHA256	`5d1f370bf150db7239204f09cf6a0603292ea28bac984e69b167e16fe160d803`
MD5	`3b6230aea1ec70f240af97e7e746a283`
BLAKE2-256	`c5a3c4963b9b8bafba8cb7c266a6436e514b3a08bd300e31621ccbec7f1a8708`

Hashes for llvmlite-0.34.0-cp38-cp38-macosx_10_9_x86_64.whl

Hashes for llvmlite-0.34.0-cp38-cp38-macosx_10_9_x86_64.whl
Algorithm	Hash digest
SHA256	`132322bc084abf336c80dd106f9357978c8c085911fb656898d3be0d9ff057ea`
MD5	`745c10bd360b9b5139470923eaa2d396`
BLAKE2-256	`d098d92305aa1f514b9a59883bb12dbe0859e6d1547ffe606755405042b65a00`

Hashes for llvmlite-0.34.0-cp38-cp38-manylinux2010_i686.whl

Hashes for llvmlite-0.34.0-cp38-cp38-manylinux2010_i686.whl
Algorithm	Hash digest
SHA256	`8f344102745fceba6eb5bf03c228bb290e9bc79157e9506a4a72878d636f9b3c`
MD5	`4de6b14f6a3f20e7d0d793f81ead36f5`
BLAKE2-256	`384022bf3f2c2496b544288398a81fd9f55359606fc9ba44b2bf2a1e02ae22de`

Hashes for llvmlite-0.34.0-cp38-cp38-manylinux2010_x86_64.whl

Hashes for llvmlite-0.34.0-cp38-cp38-manylinux2010_x86_64.whl
Algorithm	Hash digest
SHA256	`05253f3f44fab0148276335b2c1b2c4a78143dfa78e6bafd7f937d6248f297cc`
MD5	`de3690c335a13cce35cae3a2d734eced`
BLAKE2-256	`2bf8104e35345995e2ca35acff0aa429cab79677344fd54cd025eecebd0b01b1`

Hashes for llvmlite-0.34.0-cp38-cp38-win32.whl

Hashes for llvmlite-0.34.0-cp38-cp38-win32.whl
Algorithm	Hash digest
SHA256	`28264f9e2b3df4135cbcfca5a91c5b0b31dd3fc02fa623b4bb13327f0cd4fc80`
MD5	`688b40c3c13fcbbc42bb8013a647380e`
BLAKE2-256	`680a5b7e2d0d0a934aafe43ac1c3a6284da55215b62254eda476b95cf7a47ad1`

Hashes for llvmlite-0.34.0-cp38-cp38-win_amd64.whl

Hashes for llvmlite-0.34.0-cp38-cp38-win_amd64.whl
Algorithm	Hash digest
SHA256	`964f8f7a2184963cb3617d057c2382575953e488b7bb061b632ee014cfef110a`
MD5	`98fd153f5fa567f203b7a941e373c84a`
BLAKE2-256	`4c1addac06185923111e6d48d579b4829a644c6fbdef0b830f95d53ce27df21c`

Hashes for llvmlite-0.34.0.tar.gz

Hashes for llvmlite-0.34.0.tar.gz
Algorithm	Hash digest
SHA256	`f03ee0d19bca8f2fe922bb424a909d05c28411983b0c2bc58b020032a0d11f63`
MD5	`278cdd723ba6a807b526bc25b97a0874`
BLAKE2-256	`0b9607bfa93a103fb9e3e9ae7f9f7c6687ae714aee66b6f3000da3fad71e0aa2`