mpi4py
mpi4py copied to clipboard
Python bindings for MPI
Don't assume IPv4 sockets are available in TestDPM.testJoin (test_dynproc.py). Instead, use the family reported available by socket.getaddrinfo(). Use INET (IPv4) by default if available, otherwise INET6 (IPv6) or UNIX sockets....
Debian reports that mpi4py fails to build on IPV6-only build daemons, reported in Debian [Bug#1015912](https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1015912). A build log is available at https://buildd.debian.org/status/fetch.php?pkg=mpi4py&arch=amd64&ver=3.1.3-2%2Bb1&stamp=1658589239&raw=0 The error message is in testJoin (test_dynproc.TestDPM): ```...
when i use command "module load mpi/mpich-$(arch)",it return Traceback (most recent call last) : ERROR: Unable to locate a modulefile for 'mpi/mpich-x86_64' then I use command "module avail" and the...
Architecture: Power9 (Summit Super Computer) MPI Version: Package: IBM Spectrum MPI Spectrum MPI: 10.4.0.03rtm4 Spectrum MPI repo revision: IBM_SPECTRUM_MPI_10.04.00.03_2021.01.12_RTM4 Spectrum MPI release date: Unreleased developer copy MPI4py Version: 3.1.1 Reproduce...
@dalcinl I created this issue because I heard that there are some questions/suggestions about Fujitsu MPI.
I'm having a problem with data corruption when using `allgather`, but only one of the HPC systems we use. I think the problem is very likely somewhere in the infiniband...
See https://github.com/mpi4py/mpi4py/issues/19#issuecomment-826221898.
`__getitem__` with a slice might work, but it's easier to just disable them altogether.
sparc64 is not the most common architecture around, but for what it's worth 3.1.2 has started giving a Bus Error (Invalid address alignment) in testPackUnpackExternal (test_pack.TestPackExternal), ``` testProbeRecv (test_p2p_obj_matched.TestP2PMatchedWorldDup) ......
Our application, PyFR, makes very successful use of mpi4py and has support for CUDA-aware MPI implementations. Here, however, our biggest issue is knowing if the MPI distribution we are running...