experimental/cuda-ubi9/: llvmlite-0.42.0 metadata and description
lightweight wrapper around basic LLVM functionality
| classifiers |
|
| license | BSD |
| project_urls |
|
| requires_python | >=3.9 |
| File | Tox results | History |
|---|---|---|
llvmlite-0.42.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
|
|
A Lightweight LLVM Python Binding for Writing JIT Compilers
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.
Why llvmlite
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.
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 faster than llvmpy thanks to a much simpler architecture (the Numba test suite is twice faster than it was).
Compatibility
llvmlite has been tested with Python 3.9 – 3.12 and is likely to work with greater versions.
As of version 0.41.0, llvmlite requires LLVM 14.x.x on all architectures
Historical compatibility table:
llvmlite versions |
compatible LLVM versions |
|---|---|
0.41.0 - … |
14.x.x |
0.40.0 - 0.40.1 |
11.x.x and 14.x.x (12.x.x and 13.x.x untested but may work) |
0.37.0 - 0.39.1 |
11.x.x |
0.34.0 - 0.36.0 |
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