Repository navigation
Releases: mpi4py/mpi4py
Release list
4.1.2
4.1.1
-
Add MPIABI-aware import hook support.
-
Support
MPI.ERR_ERRHANDLERwith Open MPI. -
Fix Windows build issues with Intel MPI 2021.16.
-
Minor fixes and improvements to typing stubs.
-
Minor fixes to documentation.
-
Publish Python 3.8-3.14 and PyPy 3.11 wheels.
-
PyPy 3.10 (EOL) wheels are no longer published.
4.1.0
-
New features:
- Publish binary wheel packages on PyPI with support for multiple operating systems (Linux, macOS, Windows), processor architectures (AMD64, ARM64), MPI implementations (MPICH, Open MPI, and their ABI-compatible derivatives), and Python implementations (CPython, PyPy).
- Add support for the MPI 5.0 standard.
- MPI handle serialization with integral values.
- Fixed-size Fortran
LOGICALdatatypes. - Query ABI version/info and Fortran ABI info.
MPI.Datatype.Create_contiguousnow uses BigMPI's approach when using MPI implementations that do not yet support the MPI 4.0 large-count APIs.- Add support for
MPI.FLOAT16_TandMPI.BFLOAT16_Tdatatypes. These datatypes are not yet standard but available as extensions in MPI implementations. - Add
MPI.Status.tomemoryto expose the status contents as amemoryviewobject.
-
Enhancements:
- Support (opt-in via
MPI4PY_BUILD_PYSABI) for building withPy_LIMITED_APIunder Python 3.10+ (requires Cython 3.1+). - Support (opt-in via
MPI4PY_BUILD_MPIABI) for building with the MPI 5.0 standard ABI and the MPICH or Open MPI legacy ABIs. TheMPIextension module will be able to dynamically link at runtime with older versions (down to the MPI 3.0 standard) of the MPI implementation used at compile time. Only available on Linux, macOS, and Windows. - Support a
buffersizeparameter in theMPIPoolExecutor.map()andMPIPoolExecutor.starmap()methods to limit the number of pending tasks. - Improve ownership management of DLPack capsules.
- Reimplement MPI DLL search path on Windows.
- Minor fixes to typing stubs and documentation.
- Support (opt-in via
-
Backward-incompatible changes:
- Python 3.8+ is required.
- The
MPI.Exceptionclass is now a heap type. - The types of the
MPI.BOTTOM,MPI.IN_PLACE, andMPI.BUFFER_AUTOMATICare no longer subclasses ofint.
4.0.3
4.0.2
4.0.1
-
Update support for Python 3.13:
- Enable Cython 3.1 support for free-threaded CPython.
- Allow compiling Cython-generated C sources with the full Python C-API.
- Fix MPI DLL path workarounds on Windows after changes to
locals().
-
Enhancements to test suite:
- Support XML reports via
unittest-xml-reporting. - Add command line options to exclude tests by patterns and files.
- Refactor Python 2 code to use Python 3 constructs using
pyupgrade.
- Support XML reports via
-
Miscellaneous:
- Minor and mostly inconsequential subclass fix in
mpi4py.util.pkl5. - Update compatibility workarounds for legacy MPICH 3.0 release.
- Minor and mostly inconsequential subclass fix in
4.0.0
-
New features:
-
Add support for the MPI-4.0 standard.
- Use large count MPI-4 routines.
- Add persistent collective communication.
- Add partitioned point-to-point communication.
- Add new communicator constructors.
- Add the
Sessionclass and its methods.
-
Add support for the MPI-4.1 standard.
- Add non-destructive completion test for multiple requests.
- Add value-index datatype constructor.
- Add communicator/session buffer attach/detach/flush.
- Support for removal of error classes/codes/strings.
- Support for querying hardware resource information.
-
Add preliminary support for the upcoming MPI-5.0 standard.
- User-level failure mitigation (ULFM).
-
mpi4py.util.pool: New drop-in replacement formultiprocessing.pool. -
mpi4py.util.sync: New synchronization utilities. -
Add runtime check for mismatch between
mpiexecand MPI library. -
Support scikit-build-core as an alternative build backend.
-
Support meson-python as an alternative build backend.
-
-
Enhancements:
-
mpi4py.futures: Support for parallel tasks. -
mpi4py.futures: Report exception tracebacks in workers. -
mpi4py.util.pkl5: Add support for collective communication. -
Add methods
Datatype.fromcode(),Datatype.tocode()and
attributesDatatype.typestr,Datatype.typecharto simplify
NumPy interoperability for simple cases. -
Add methods
Comm.Create_errhandler(),Win.Create_errhandler(),
andFile.Create_errhandler()to create custom error handlers. -
Add support for pickle serialization of instances of MPI types.
All instances ofDatatype,Info, andStatuscan be
serialized. Instances ofOpcan be serialized only if created
throughmpi4pyby callingOp.Create(). Instances of other MPI
types can be serialized only if they reference predefined handles. -
Add
handleattribute andfromhandle()class method to MPI
classes to ease interoperability with external code. The handle
value is an unsigned integer guaranteed to fit on the platform's
uintptr_tC type. -
Add lowercase
free()method to MPI classes to ease MPI object
deallocation and cleanup. This method eventually attempts to call
Free(), but only if the object's MPI handle is not a null or
predefined handle, and such call is allowed within the World Model
init/finalize.
-
-
Backward-incompatible changes:
-
Python 2 is no longer supported, Python 3.6+ is required, but
typing stubs are supported for Python 3.8+. -
The
Intracomm.Create_group()method is no longer defined in the
baseCommclass. -
Group.Compare()andComm.Compare()are no longer class methods
but instance methods. Existing codes using the former class
methods are expected to continue working. -
Group.Translate_ranks()is no longer a class method but an
instance method. Existing codes using the former class method are
expected to continue working. -
The
LBandUBdatatypes are no longer available, use
Datatype.Create_resized()instead. -
The
HOSTpredefined attribute key is no longer available. -
The
MPI.memoryclass has been renamed toMPI.buffer. The old
name is still available as an alias to the new name. -
The
mpi4py.dlmodule is no longer available. -
The
mpi4py.get_configfunction returns an empty dictionary.
-
-
Miscellaneous:
-
The project is now licensed under the BSD-3-Clause license. This
change is fairly inconsequential for users and distributors. It
simply adds an additional clause against using contributor names
for promotional purposes without their consent. -
Add a new guidelines section to documentation laying out new fair
play rules. These rules ask companies and outside developers to
refrain from reusing thempi4pyname in unaffiliated projects,
publishing binary mpi4py wheels on the main Python Package Index
(PyPI), and distributing modified versions with incompatible or
extended API changes. The primary motivation of these rules is to
avoid fragmentation and end-user confusion.
-