Various updates.
This commit was SVN r12540.
Этот коммит содержится в:
родитель
5419219041
Коммит
e124857353
78
README
78
README
@ -41,8 +41,40 @@ Thanks for your time.
|
||||
|
||||
===========================================================================
|
||||
|
||||
Release notes for SC 2006 beta:
|
||||
|
||||
- This tarball has had some level of testing within the Open MPI
|
||||
developers, but has not been widely tested with real world
|
||||
applications. We'd appreciate your feedback.
|
||||
|
||||
- Running heterogenious MPI jobs (e.g., different endian between
|
||||
multiple hosts) is not fully working yet.
|
||||
|
||||
- Some MPI jobs may not clean up properly if errors occur:
|
||||
- mpirun may "hang" and have to be killed from another session
|
||||
- ORTE daemons may be left orphaned on some nodes and need to be
|
||||
manually killed
|
||||
|
||||
- Not all items marked as resolved in the v1.1.3 section of the NEWS
|
||||
file have been migrated to this 1.2 release yet. A new beta release
|
||||
is expected shortly after SC 2006 that contains these fixes:
|
||||
- Workaround for Intel C++ compiler bug
|
||||
- MPI_SIZEOF fixes for COMPLEX types
|
||||
|
||||
- MPI jobs running on OpenIB / OpenFabrics networks that
|
||||
simultaneously post a large number of non-blocking sends of small
|
||||
messages may fail. This issue is fixed on the Open MPI development
|
||||
head, and will be included in the post-SC beta release.
|
||||
|
||||
- Striping MPI messages across multiple TCP interfaces is not working.
|
||||
|
||||
- See Open MPI's bug tracking system for a full list of outstanding
|
||||
issues: https://svn.open-mpi.org/trac/ompi/report
|
||||
|
||||
===========================================================================
|
||||
|
||||
The following abbreviated list of release notes applies to this code
|
||||
base as of this writing (12 Sep 2006):
|
||||
base as of this writing (10 Nov 2006):
|
||||
|
||||
- Open MPI includes support for a wide variety of supplemental
|
||||
hardware and software package. When configuring Open MPI, you may
|
||||
@ -87,6 +119,8 @@ base as of this writing (12 Sep 2006):
|
||||
- Linux, 64 bit (x86), with gcc
|
||||
- OS X (10.3), 32 bit, with gcc
|
||||
- OS X (10.4), 32 bit, with gcc
|
||||
- Solaris 10 update 2, SPARC and AMD, 32 and 64 bit, with Sun Studio
|
||||
10
|
||||
|
||||
- Other systems have been lightly (but not fully tested):
|
||||
- Other compilers on Linux, 32 and 64 bit
|
||||
@ -197,12 +231,6 @@ base as of this writing (12 Sep 2006):
|
||||
shared memory, and Myrinet/GM. Myrinet/GM has only been lightly
|
||||
tested.
|
||||
|
||||
- Due to limitations in the Libtool 1.5 series, Fortran 90 MPI
|
||||
bindings support can only be built as a static library. It is
|
||||
expected that Libtool 2.0 (and therefore future releases of Open
|
||||
MPI) will be able to support shared libraries for the Fortran 90
|
||||
bindings.
|
||||
|
||||
- The XGrid support is experimental - see the Open MPI FAQ and this
|
||||
post on the Open MPI user's mailing list for more information:
|
||||
|
||||
@ -226,16 +254,7 @@ base as of this writing (12 Sep 2006):
|
||||
- The Open Fabrics Enterprise Distribution (OFED) software package
|
||||
v1.0 will not work properly with Open MPI v1.2 (and later) due to
|
||||
how its Mellanox InfiniBand plugin driver is created. The problem
|
||||
is fixed OFED v1.1 (and beyond).
|
||||
|
||||
- The current version of the Open MPI point-to-point engine does not
|
||||
yet support hardware-level MPI message matching. As such, MPI
|
||||
message matching must be performed in software, artificially
|
||||
increasing latency for short messages on certain networks (such as
|
||||
MX and hardware-supported Portals). Future versions of Open MPI
|
||||
will support hardware matching on networks that provide it, and will
|
||||
eliminate the extra overhead of software MPI message matching where
|
||||
possible.
|
||||
is fixed OFED v1.1 (and later).
|
||||
|
||||
- The Fortran 90 MPI bindings can now be built in one of three sizes
|
||||
using --with-mpi-f90-size=SIZE (see description below). These sizes
|
||||
@ -327,13 +346,23 @@ for a full list); a summary of the more commonly used ones follows:
|
||||
most cases. This option is only needed for special configurations.
|
||||
|
||||
--with-openib=<directory>
|
||||
Specify the directory where the Open Fabrics (previously known as
|
||||
Specify the directory where the OpenFabrics (previously known as
|
||||
OpenIB) libraries and header files are located. This enables Open
|
||||
Fabrics support in Open MPI.
|
||||
|
||||
--with-openib-libdir=<directory>
|
||||
Look in directory for the OPENIB libraries. By default, Open MPI will
|
||||
look in <openib directory>/lib and <openib directory>/lib64, which covers
|
||||
Look in directory for the OpenFabrics libraries. By default, Open
|
||||
MPI will look in <openib directory>/lib and <openib
|
||||
directory>/lib64, which covers most cases. This option is only
|
||||
needed for special configurations.
|
||||
|
||||
--with-psm=<directory>
|
||||
Specify the directory where the QLogic PSM library and header files
|
||||
are located. This enables InfiniPath support in Open MPI.
|
||||
|
||||
--with-psm-libdir=<directory>
|
||||
Look in directory for the PSM libraries. By default, Open MPI will
|
||||
look in <psm directory>/lib and <psm directory>/lib64, which covers
|
||||
most cases. This option is only needed for special configurations.
|
||||
|
||||
--with-tm=<directory>
|
||||
@ -601,12 +630,14 @@ MPI component frameworks:
|
||||
|
||||
allocator - Memory allocator
|
||||
bml - BTL management layer
|
||||
btl - MPI point-to-point byte transfer layer
|
||||
btl - MPI point-to-point byte transfer layer, used for MPI
|
||||
point-to-point messages on some types of networks
|
||||
coll - MPI collective algorithms
|
||||
io - MPI-2 I/O
|
||||
mpool - Memory pooling
|
||||
mtl - Matching transport layer, used for MPI point-to-point
|
||||
messages on some types of networks
|
||||
pml - MPI point-to-point management layer
|
||||
ptl - (Outdated / deprecated) MPI point-to-point transport layer
|
||||
rcache - Memory registration cache
|
||||
topo - MPI topology routines
|
||||
|
||||
@ -617,6 +648,7 @@ errmgr - RTE error manager
|
||||
gpr - General purpose registry
|
||||
iof - I/O forwarding
|
||||
ns - Name server
|
||||
odls - OpenRTE daemon local launch subsystem
|
||||
oob - Out of band messaging
|
||||
pls - Process launch system
|
||||
ras - Resource allocation system
|
||||
@ -626,7 +658,7 @@ rmgr - Resource manager
|
||||
rml - RTE message layer
|
||||
schema - Name schemas
|
||||
sds - Startup / discovery service
|
||||
soh - State of health monitor
|
||||
smr - State-of-health monitoring subsystem
|
||||
|
||||
Miscellaneous frameworks:
|
||||
-------------------------
|
||||
|
Загрузка…
x
Ссылка в новой задаче
Block a user