Hui Zhou
52 linhas
1.8 KiB
Plaintext
52 linhas
1.8 KiB
Plaintext
----------------------------------------------------------------------
|
|
KNOWN ISSUES
|
|
----------------------------------------------------------------------
|
|
|
|
### CH4
|
|
* Dynamic process support
|
|
- Requires fi_tagged capability in OFI netmod
|
|
- Not supported in UCX netmod at this time
|
|
|
|
* Test failures
|
|
- test/mpi/errors/coll does not 100% pass.
|
|
- Test failures differ based on experimental setup and network
|
|
support built in.
|
|
|
|
* libfabric psm3 provider may cause segfaults at exit
|
|
- psm3 provider may cause segfaults in exit handlers. Configure
|
|
with --disable-psm3 if this occurs.
|
|
|
|
### OS/X
|
|
|
|
* C++ bindings - Exception handling in C++ bindings is broken because
|
|
of a bug in libc++ (see
|
|
https://github.com/android-ndk/ndk/issues/289). You can workaround
|
|
this issue by explicitly linking your application to libstdc++
|
|
instead (e.g., mpicxx -o foo foo.cxx -lstdc++).
|
|
|
|
### Fine-grained thread safety
|
|
|
|
* Multiple VCIs (--enable-thread-cs=per-vci) is only supported in ch4.
|
|
|
|
* ch3 does not (and will not) support fine-grained threading.
|
|
|
|
### Lacking channel-specific features
|
|
|
|
* MPICH does not presently support communication across heterogeneous
|
|
platforms (e.g., a big-endian machine communicating with a
|
|
little-endian machine).
|
|
|
|
* ch3 has known problems in some cases when threading and dynamic
|
|
processes are used together on communicators of size greater than
|
|
one.
|
|
|
|
|
|
### Performance issues
|
|
|
|
* MPI_Irecv operations that are not explicitly completed before
|
|
MPI_Finalize is called may fail to complete before MPI_Finalize
|
|
returns, and thus never complete. Furthermore, any matching send
|
|
operations may erroneously fail. By explicitly completed, we mean
|
|
that the request associated with the operation is completed by one
|
|
of the MPI_Test or MPI_Wait routines.
|