CHANGES
Release 4.0.0 [2024-07-28]
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
Session
class 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
mpiexec
and 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.typechar
to 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 of
Datatype
,Info
, andStatus
can be serialized. Instances ofOp
can be serialized only if created throughmpi4py
by callingOp.Create()
. Instances of other MPI types can be serialized only if they reference predefined handles.Add
handle
attribute andfromhandle()
class method to MPI classes to ease interoperability with external code. The handle value is a unsigned integer guaranteed to fit on the platform’suintptr_t
C type.Add lowercase
free()
method to MPI classes to ease MPI object deallocation and cleanup. This method eventually attempts to callFree()
, 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 baseComm
class.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
LB
andUB
datatypes are no longer available, useDatatype.Create_resized()
instead.The
HOST
predefined attribute key is no longer available.The
MPI.memory
class has been renamed toMPI.buffer
. The old name is still available as an alias to the new name.The
mpi4py.dl
module is no longer available.The
mpi4py.get_config
function returns an empty dictionary.
Miscellaneous:
The project is now licenced 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 the
mpi4py
name 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.
Release 3.1.6 [2024-04-14]
Warning
This is the last release supporting Python 2.
Fix various build issues.
Release 3.1.5 [2023-10-04]
Warning
This is the last release supporting Python 2.
Rebuild C sources with Cython 0.29.36 to support Python 3.12.
Release 3.1.4 [2022-11-02]
Warning
This is the last release supporting Python 2.
Rebuild C sources with Cython 0.29.32 to support Python 3.11.
Fix contiguity check for DLPack and CAI buffers.
Workaround build failures with setuptools v60.
Release 3.1.3 [2021-11-25]
Warning
This is the last release supporting Python 2.
Add missing support for
MPI.BOTTOM
to generalized all-to-all collectives.
Release 3.1.2 [2021-11-04]
Warning
This is the last release supporting Python 2.
mpi4py.futures
: Add_max_workers
property toMPIPoolExecutor
.mpi4py.util.dtlib
: Fix computation of alignment for predefined datatypes.mpi4py.util.pkl5
: Fix deadlock when usingssend()
+mprobe()
.mpi4py.util.pkl5
: Add environment variableMPI4PY_PICKLE_THRESHOLD
.mpi4py.rc
: Interpret"y"
and"n"
strings as boolean values.Fix/add typemap/typestr for
MPI.WCHAR
/MPI.COUNT
datatypes.Minor fixes and additions to documentation.
Minor fixes to typing support.
Support for local version identifier (PEP-440).
Release 3.1.1 [2021-08-14]
Warning
This is the last release supporting Python 2.
Fix typo in Requires-Python package metadata.
Regenerate C sources with Cython 0.29.24.
Release 3.1.0 [2021-08-12]
Warning
This is the last release supporting Python 2.
New features:
mpi4py.util
: New package collecting miscellaneous utilities.
Enhancements:
Add pickle-based
Request.waitsome()
andRequest.testsome()
.Add lowercase methods
Request.get_status()
andRequest.cancel()
.Support for passing Python GPU arrays compliant with the DLPack data interchange mechanism (link) and the
__cuda_array_interface__
(CAI) standard (link) to uppercase methods. This support requires that mpi4py is built against CUDA-aware MPI implementations. This feature is currently experimental and subject to future changes.mpi4py.futures
: Add support for initializers and canceling futures at shutdown. Environment variables names now follow the patternMPI4PY_FUTURES_*
, the previousMPI4PY_*
names are deprecated.Add type annotations to Cython code. The first line of the docstring of functions and methods displays a signature including type annotations.
Add companion stub files to support type checkers.
Support for weak references.
Miscellaneous:
Add a new mpi4py publication (link) to the citation listing.
Release 3.0.3 [2019-11-04]
Regenerate Cython wrappers to support Python 3.8.
Release 3.0.2 [2019-06-11]
Bug fixes:
Fix handling of readonly buffers in support for Python 2 legacy buffer interface. The issue triggers only when using a buffer-like object that is readonly and does not export the new Python 3 buffer interface.
Fix build issues with Open MPI 4.0.x series related to removal of many MPI-1 symbols deprecated in MPI-2 and removed in MPI-3.
Minor documentation fixes.
Release 3.0.1 [2019-02-15]
Bug fixes:
Fix
Comm.scatter()
and other collectives corrupting input send list. Add safety measures to prevent related issues in global reduction operations.Fix error-checking code for counts in
Op.Reduce_local()
.
Enhancements:
Map size-specific Python/NumPy typecodes to MPI datatypes.
Allow partial specification of target list/tuple arguments in the various
Win
RMA methods.Workaround for removal of
MPI_{LB|UB}
in Open MPI 4.0.Support for Microsoft MPI v10.0.
Release 3.0.0 [2017-11-08]
New features:
mpi4py.futures
: Execute computations asynchronously using a pool of MPI processes. This package is based onconcurrent.futures
from the Python standard library.mpi4py.run
: Run Python code and abort execution in case of unhandled exceptions to prevent deadlocks.mpi4py.bench
: Run basic MPI benchmarks and tests.
Enhancements:
Lowercase, pickle-based collective communication calls are now thread-safe through the use of fine-grained locking.
The
MPI
module now exposes amemory
type which is a lightweight variant of the builtinmemoryview
type, but exposes both the legacy Python 2 and the modern Python 3 buffer interface under a Python 2 runtime.The
MPI.Comm.Alltoallw()
method now usescount=1
anddispl=0
as defaults, assuming that messages are specified through user-defined datatypes.The
Request.Wait[all]()
methods now returnTrue
to match the interface ofRequest.Test[all]()
.The
Win
class now implements the Python buffer interface.
Backward-incompatible changes:
The
buf
argument of theMPI.Comm.recv()
method is deprecated, passing anything butNone
emits a warning.The
MPI.Win.memory
property was removed, use theMPI.Win.tomemory()
method instead.Executing
python -m mpi4py
in the command line is now equivalent topython -m mpi4py.run
. For the former behavior, usepython -m mpi4py.bench
.Python 2.6 and 3.2 are no longer supported. The
mpi4py.MPI
module may still build and partially work, but other pure-Python modules under thempi4py
namespace will not.Windows: Remove support for legacy MPICH2, Open MPI, and DeinoMPI.
Release 2.0.0 [2015-10-18]
Support for MPI-3 features.
Matched probes and receives.
Nonblocking collectives.
Neighborhood collectives.
New communicator constructors.
Request-based RMA operations.
New RMA communication and synchronisation calls.
New window constructors.
New datatype constructor.
New C++ boolean and floating complex datatypes.
Support for MPI-2 features not included in previous releases.
Generalized All-to-All collective (
Comm.Alltoallw()
)User-defined data representations (
Register_datarep()
)
New scalable implementation of reduction operations for Python objects. This code is based on binomial tree algorithms using point-to-point communication and duplicated communicator contexts. To disable this feature, use
mpi4py.rc.fast_reduce = False
.Backward-incompatible changes:
Python 2.4, 2.5, 3.0 and 3.1 are no longer supported.
Default MPI error handling policies are overridden. After import, mpi4py sets the
ERRORS_RETURN
error handler inCOMM_SELF
andCOMM_WORLD
, as well as any newComm
,Win
, orFile
instance created through mpi4py, thus effectively ignoring the MPI rules about error handler inheritance. This way, MPI errors translate to Python exceptions. To disable this behavior and use the standard MPI error handling rules, usempi4py.rc.errors = 'default'
.Change signature of all send methods,
dest
is a required argument.Change signature of all receive and probe methods,
source
defaults toANY_SOURCE
,tag
defaults toANY_TAG
.Change signature of send lowercase-spelling methods,
obj
arguments are not mandatory.Change signature of recv lowercase-spelling methods, renamed ‘obj’ arguments to ‘buf’.
Change
Request.Waitsome()
andRequest.Testsome()
to returnNone
orlist
.Change signature of all lowercase-spelling collectives,
sendobj
arguments are now mandatory,recvobj
arguments were removed.Reduction operations
MAXLOC
andMINLOC
are no longer special-cased in lowercase-spelling methodsComm.[all]reduce()
andComm.[ex]scan()
, the input object must be specified as a tuple(obj, location)
.Change signature of name publishing functions. The new signatures are
Publish_name(service_name, port_name, info=INFO_NULL)
andUnpublish_name(service_name, port_name, info=INFO_NULL)`
.Win
instances now cache Python objects exposing memory by keeping references instead of using MPI attribute caching.Change signature of
Win.Lock()
. The new signature isWin.Lock(rank, lock_type=LOCK_EXCLUSIVE, assertion=0)
.Move
Cartcomm.Map()
toIntracomm.Cart_map()
.Move
Graphcomm.Map()
toIntracomm.Graph_map()
.Remove the
mpi4py.MPE
module.Rename the Cython definition file for use with
cimport
statement frommpi_c.pxd
tolibmpi.pxd
.
Release 1.3.1 [2013-08-07]
Regenerate C wrappers with Cython 0.19.1 to support Python 3.3.
Install
*.pxd
files in<site-packages>/mpi4py
to ease the support for Cython’scimport
statement in code requiring to access mpi4py internals.As a side-effect of using Cython 0.19.1, ancient Python 2.3 is no longer supported. If you really need it, you can install an older Cython and run
python setup.py build_src --force
.
Release 1.3 [2012-01-20]
Now
Comm.recv()
accept a buffer to receive the message.Add
Comm.irecv()
andRequest.{wait|test}[any|all]()
.Add
Intracomm.Spawn_multiple()
.Better buffer handling for PEP 3118 and legacy buffer interfaces.
Add support for attribute attribute caching on communicators, datatypes and windows.
Install MPI-enabled Python interpreter as
<path>/mpi4py/bin/python-mpi
.Windows: Support for building with Open MPI.
Release 1.2.2 [2010-09-13]
Add
mpi4py.get_config()
to retrieve information (compiler wrappers, includes, libraries, etc) about the MPI implementation employed to build mpi4py.Workaround Python libraries with missing GILState-related API calls in case of non-threaded Python builds.
Windows: look for MPICH2, DeinoMPI, Microsoft HPC Pack at their default install locations under %ProgramFiles.
MPE: fix hacks related to old API’s, these hacks are broken when MPE is built with a MPI implementations other than MPICH2.
HP-MPI: fix for missing Fortran datatypes, use dlopen() to load the MPI shared library before MPI_Init()
Many distutils-related fixes, cleanup, and enhancements, better logics to find MPI compiler wrappers.
Support for
pip install mpi4py
.
Release 1.2.1 [2010-02-26]
Fix declaration in Cython include file. This declaration, while valid for Cython, broke the simple-minded parsing used in conf/mpidistutils.py to implement configure-tests for availability of MPI symbols.
Update SWIG support and make it compatible with Python 3. Also generate an warning for SWIG < 1.3.28.
Fix distutils-related issues in Mac OS X. Now ARCHFLAGS environment variable is honored of all Python’s
config/Makefile
variables.Fix issues with Open MPI < 1.4.2 related to error checking and
MPI_XXX_NULL
handles.
Release 1.2 [2009-12-29]
Automatic MPI datatype discovery for NumPy arrays and PEP-3118 buffers. Now buffer-like objects can be messaged directly, it is no longer required to explicitly pass a 2/3-list/tuple like
[data, MPI.DOUBLE]
, or[data, count, MPI.DOUBLE]
. Only basic types are supported, i.e., all C/C99-native signed/unsigned integral types and single/double precision real/complex floating types. Many thanks to Eilif Muller for the initial feedback.Nonblocking send of pickled Python objects. Many thanks to Andreas Kloeckner for the initial patch and enlightening discussion about this enhancement.
Request
instances now hold a reference to the Python object exposing the buffer involved in point-to-point communication or parallel I/O. Many thanks to Andreas Kloeckner for the initial feedback.Support for logging of user-defined states and events using MPE. Runtime (i.e., without requiring a recompile!) activation of logging of all MPI calls is supported in POSIX platforms implementing
dlopen()
.Support for all the new features in MPI-2.2 (new C99 and F90 datatypes, distributed graph topology, local reduction operation, and other minor enhancements).
Fix the annoying issues related to Open MPI and Python dynamic loading of extension modules in platforms supporting
dlopen()
.Fix SLURM dynamic loading issues on SiCortex. Many thanks to Ian Langmore for providing me shell access.
Release 1.1.0 [2009-06-06]
Fix bug in
Comm.Iprobe()
that caused segfaults as Python C-API calls were issued with the GIL released (issue #2).Add
Comm.bsend()
andComm.ssend()
for buffered and synchronous send semantics when communicating general Python objects.Now the call
Info.Get(key)
return a single value (i.e, instead of a 2-tuple); this value isNone
ifkey
is not in theInfo
object, or a string otherwise. Previously, the call redundantly returned(None, False)
for missing key-value pairs;None
is enough to signal a missing entry.Add support for parametrized Fortran datatypes.
Add support for decoding user-defined datatypes.
Add support for user-defined reduction operations on memory buffers. However, at most 16 user-defined reduction operations can be created. Ask the author for more room if you need it.
Release 1.0.0 [2009-03-20]
This is the fist release of the all-new, Cython-based, implementation of MPI for Python. Unfortunately, this implementation is not backward-compatible with the previous one. The list below summarizes the more important changes that can impact user codes.
Some communication calls had overloaded functionality. Now there is a clear distinction between communication of general Python object with pickle, and (fast, near C-speed) communication of buffer-like objects (e.g., NumPy arrays).
for communicating general Python objects, you have to use all-lowercase methods, like
send()
,recv()
,bcast()
, etc.for communicating array data, you have to use
Send()
,Recv()
,Bcast()
, etc. methods. Buffer arguments to these calls must be explicitly specified by using a 2/3-list/tuple like[data, MPI.DOUBLE]
, or[data, count, MPI.DOUBLE]
(the former one uses the byte-size ofdata
and the extent of the MPI datatype to define thecount
).
Indexing a communicator with an integer returned a special object associating the communication with a target rank, alleviating you from specifying source/destination/root arguments in point-to-point and collective communications. This functionality is no longer available, expressions like:
MPI.COMM_WORLD[0].Send(...) MPI.COMM_WORLD[0].Recv(...) MPI.COMM_WORLD[0].Bcast(...)
have to be replaced by:
MPI.COMM_WORLD.Send(..., dest=0) MPI.COMM_WORLD.Recv(..., source=0) MPI.COMM_WORLD.Bcast(..., root=0)
Automatic MPI initialization (i.e., at import time) requests the maximum level of MPI thread support (i.e., it is done by calling
MPI_Init_thread()
and passingMPI_THREAD_MULTIPLE
). In case you need to change this behavior, you can tweak the contents of thempi4py.rc
module.In order to obtain the values of predefined attributes attached to the world communicator, now you have to use the
Get_attr()
method on theMPI.COMM_WORLD
instance:tag_ub = MPI.COMM_WORLD.Get_attr(MPI.TAG_UB)
In the previous implementation,
MPI.COMM_WORLD
andMPI.COMM_SELF
were associated to duplicates of the (C-level)MPI_COMM_WORLD
andMPI_COMM_SELF
predefined communicator handles. Now this is no longer the case,MPI.COMM_WORLD
andMPI.COMM_SELF
proxies the actualMPI_COMM_WORLD
andMPI_COMM_SELF
handles.Convenience aliases
MPI.WORLD
andMPI.SELF
were removed. Use insteadMPI.COMM_WORLD
andMPI.COMM_SELF
.Convenience constants
MPI.WORLD_SIZE
andMPI.WORLD_RANK
were removed. Use insteadMPI.COMM_WORLD.Get_size()
andMPI.COMM_WORLD.Get_rank()
.