-
Notifications
You must be signed in to change notification settings - Fork 3
/
RELEASE_NOTES
107 lines (71 loc) · 3.88 KB
/
RELEASE_NOTES
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
----------------------------------------------------------------------
KNOWN ISSUES
----------------------------------------------------------------------
### Threads
* ch3:sock does not (and will not) support fine-grained threading.
* MPI-IO APIs are not currently thread-safe when using fine-grained
threading (--enable-thread-cs=per-object).
* ch3:nemesis:tcp fine-grained threading is still experimental and may
have correctness or performance issues. Known correctness issues
include dynamic process support and generalized request support.
### Lacking channel-specific features
* ch3 does not presently support communication across heterogeneous
platforms (e.g., a big-endian machine communicating with a
little-endian machine).
* ch3:nemesis:mx does not support dynamic processes at this time.
* Support for "external32" data representation is incomplete. This
affects the MPI_Pack_external and MPI_Unpack_external routines, as
well the external data representation capabilities of ROMIO.
* ch3:dllchan is rated "experimental". There are known problems when
configured with --enable-g and --enable-g=log.
* ch3 has known problems in some cases when threading and dynamic
processes are used together on communicators of size greater than
one.
### Build Platforms
* Builds using the native "make" program on OpenSolaris fail unknown
reasons. A workaround is to use GNU Make instead. See the following
ticket for more information:
http://trac.mcs.anl.gov/projects/mpich2/ticket/1122
* The sctp channel is fully supported for FreeBSD and Mac OS X. As of
the time of this release, bugs in the stack currently existed in
the Linux kernel, and will hopefully soon be resolved. It is known
to not work under Solaris and Windows. For Solaris, the SCTP API
available in the kernel of standard Solaris 10 is a subset of the
standard API used by the sctp channel. Cooperation with the Sun
SCTP developers to support ch3:sctp under Solaris for future
releases is currently ongoing. For Windows, no known kernel-based
SCTP stack for Windows currently exists.
### Other configure options
* --enable-sharedlibs requires specific options (gcc, gcc-osx,
solaris-cc) to tell the build system the compiler and platform
being used.
### Process Managers
* The MPD process manager can only handle relatively small amounts of
data on stdin and may also have problems if there is data on stdin
that is not consumed by the program.
* The SMPD process manager does not work reliably with threaded MPI
processes. MPI_Comm_spawn() does not currently work for >= 256
arguments with smpd.
### Performance issues
* SMP-aware collectives do not perform as well, in select cases, as
non-SMP-aware collectives, e.g. MPI_Reduce with message sizes
larger than 64KiB. These can be disabled by the configure option
"--disable-smpcoll".
* 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.
### C++ Binding:
* The MPI datatypes corresponding to Fortran datatypes are not
available (e.g., no MPI::DOUBLE_PRECISION).
* The C++ binding does not implement a separate profiling interface,
as allowed by the MPI-2 Standard (Section 10.1.10 Profiling).
* MPI::ERRORS_RETURN may still throw exceptions in the event of an
error rather than silently returning.
### Fortran 90:
* MPI-IO routines have been accidentally left out of the Fortran 90
module. This has been fixed in trunk (and will be reflected in
mpich2-1.5.x and later releases). It has not been fixed in the
1.4.x releases to avoid breaking ABI.