--enable-mpi-ext configure switch in the top-level README file).
See Josh's excellent wiki page about OMPI extensions:
https://svn.open-mpi.org/trac/ompi/wiki/MPIExtensions
This extension exposes a new API to MPI applications:
{{{
int OMPI_Affinity_str(char ompi_bound[OMPI_AFFINITY_STRING_MAX],
char current_binding[OMPI_AFFINITY_STRING_MAX],
char exists[OMPI_AFFINITY_STRING_MAX]);
}}}
It returns 3 things. Each are a prettyprint string describing sets of
processors in terms of sockets and cores:
1. What Open MPI bound this process to. If Open MPI didn't bind this
process, the prettyprint string says so.
1. What this process is currently bound to. If the process is
unbound, the prettyprint string says so. This string is a
separate OUT parameter to detect the case where some other entity
bound the process (potentially after Open MPI bound it).
1. What processors are availabile in the system, mainly for reference.
This commit was SVN r23018.
- extendet support for BlueGene/P:
- building shared VT libraries
- tracing 3rd-party libraries (e.g. libc I/O)
- tracing multi-threaded applications
VT configure fixes:
- fixed detection on CTool for 3rd-party library tracing
VT fixes:
- reduced memory overhead by using the trace buffer for string/array elements of some records
- do not shutdown call-stack if max. number of buffer flushes reached, because the additional function leaves suggest a wrong application flow
- vtunify-mpi:
- fixed conversion of VTUnify_MPI_Aint arrays
- vtwrapper:
- if an OPARI modified object file (*.mod.o) cannot be renamed, abort only if the compiler wrapper runs in "only-compile" mode (-c)
OTF fixes:
- otfinfo:
- fixed and enhanced calculation of trace file size
- changed unit of timer resolution (s -> Hz)
- otfprofile:
- fixed progress
- kill '_' and '\' in process names to make LaTex happier
This commit was SVN r22963.
1. fix a bug that caused an infinite loop in configure when specifying want-ft but not want-ft-thread by removing a stale reference to the opal-progress-thread option
2. add want-ft=orcm so we can build the orcm errmgr component
3. cleanup the use of "ompi_want_ft_xxx" and replace it with "opal_want_ft_xxx" so that naming conventions are preserved
This commit was SVN r22885.
Adds memory barriers which are definitely needed on powerpc
This commit was SVN r22879.
The following Trac tickets were found above:
Ticket 2351 --> https://svn.open-mpi.org/trac/ompi/ticket/2351
Update ompi_info and orte-info to include the new framework.
Fix some selection logic and a typo'd variable name
Still remains ompi_ignored until we complete testing
This commit was SVN r22848.
Remove the --enable-progress-threads option as this is no longer functional, and hardcode OPAL_ENABLE_PROGRESS_THREADS to 0.
Replace the --enable-mpi-threads option with --enable-mpi-thread-multiple as this is clearer as to meaning. This option automatically turns "on" opal thread support if it wasn't already so specified. If the user specifies --disable-opal-multi-threads --enable-mpi-thread-multiple, we will error out with a message
Add a new --enable-opal-multi-threads option that turns "on" opal thread support without doing anything wrt mpi-thread-multiple
This commit was SVN r22841.
After talking to both Brian and George, the conensus was to just
remove the flag and the test function. Begone, evil spirits, BEGONE!
This commit was SVN r22831.
The following Trac tickets were found above:
Ticket 2273 --> https://svn.open-mpi.org/trac/ompi/ticket/2273
Aleksej Saushev.
Dont use bash or bashism in shell scripts
We should use Posix' setpgid(0,0), which is equivalent to setpgrp().
This commit was SVN r22829.
Many of the OPAL_ENABLE_FT should be OPAL_ENABLE_FT_CR, so fix those.
The OPAL Layer INC should call opal_output on restart so that it can refresh the string it prints to reflect the current pid/hostname which may have changed.
This commit was SVN r22824.
- skip test for libdl on BlueGene? and CrayXT platforms (particularly on CrayXT this library can be linked but it isn't suitable)
- set cache variables for functions 'PMPI_Win_test', 'PMPI_Win_lock', 'PMPI_Win_unlock', and 'MPI_Register_datarep', if VT is configuring for Open MPI
- added test for pthread functions 'pthread_condattr_<set|get>pshared' and 'pthread_mutexattr_<set|get>pshared', because they are not available on some platforms
VT fixes:
- cut 'nm' collected symbol names at '??'
- vtunify:
- fixed unsafe usage of some strncpy's
- fixed potential segmentation fault in vtunify-mpi which might occur on 32bit platforms using MPICH2
OTF general:
- updated date in copyright header of each source file
OTF fixes:
- minor code cleanups (indentation, nicer error messages, more correct free's)
- otfaux:
- fix to place final statistics after the very last record instead of right before
- changed fatal error to a warning when a file is closed twice (or unexpectedly)
This commit was SVN r22820.
previously an orte_std_cntr_t, which is int32_t).
Comparison with < 0 don't make any sense, here.
This commit was SVN r22799.
The following SVN revision numbers were found above:
r22727 --> open-mpi/ompi@2541aa98ab
Short version: there is a bug in OS X/Snow Leopard, but there is also
a bug in Open MPI. Fixing the bug in Open MPI is both trivial (a
1-line change) and avoids the bug in OS X. We'll file an OS X bug
report upstream with Apple, but it should no longer affect us here in
OMPI.
Fixes trac:2039.
More details:
Some background first:
1. IPv4 sockets can only accept incoming IPv4 connections. However,
IPv6 sockets can be configured to accept ''only'' incoming IPv6
connection, or ''both'' incoming IPv4 and IPv6 connections. An
IPv6 socket attribute sets which listening behavior is used.
1. IPv4 and IPv6 have different port namespaces. Hence, it is
permissable to bind a v4 socket to port X ''and'' also bind a v6
socket to that same port X on the same interface (assuming that
the v6 socket is only accepting incoming v6 connections).
Incoming v4 connections to port X on the interface should get
matched to the listening v4 socket; incoming v6 connections should
get matched to the listening v6 socket.
1. When v6 sockets accept ''both'' incoming v4 and v6 connections, it
should claim port X in both namespaces.
1. Linux's default behavior is to only allow one listening socket to
be bound to a given port (i.e., ''either'' a v6 or v4 socket to be
bound to a single port X -- not both). A v6 socket can listen for
both v4 and v6 incoming connections on that port, but still --
only one socket will be bound to that port.
1. Snow Leopard's default behavior is to share ports -- i.e., let
both a v4 and a v6 listening socket to be bound to port X
(assuming that the v6 socket is only accepting incoming v6
connections).
The TCP BTL creates a listening socket for each address family.
Hence, it creates a v4 listening socket on INADDR_ANY and a v6
listening socket on the v6 equivalent of INADDR_ANY. OMPI then
iteratively tries to find ports to listen on within the range of
[mca_btl_tcp_port_min, mca_btl_tcp_port_min + mca_btl_tcp_port_range).
On Linux, the v4 socket will be bound to port X and the v6 socket will
likely be bound to port Y (where X != Y). On Snow Leopard, the v4
socket will be bound to port X and the v6 socket may ''also'' be bound
to port X. Since the namespaces are separate, this shouldn't be a
problem.
However, Open MPI was accidentally setting the v6 listening behavior
to accept ''both'' v4 and v6 incoming connections. This is a trivial
thing to fix -- change a 0 to a 1 in the code. On Linux, this issue
didn't matter because the v4 and v6 sockets were on different ports.
So even though the v6 socket ''would'' have accepted incoming v4
connections, that never happened because OMPI would direct v4
connections to the v4 port.
But on Snow Leopard, the v4 and v6 listening ports could end up
sharing the same port number. As mentioned above, this ''shouldn't''
have been a problem, but it looks like Snow Leopard has the following
bugs:
* If a v4 socket is already bound to port X, we're pretty sure that a
v6 socket with the "accept both v4 and v6 incoming connections"
listening behavior should not be able to claim port X (because
there's already a v4 socket listening on X). However, Snow Leopard
would allow binding a v4 socket to port X, and then allow a v6
socket configured to allow incoming v4 and v6 connections to
''also'' be bound to port X.
* After binding the v6 socket to port X, Snow Leopard then lets
''another'' v4 socket ''also'' get bound to port X. Hence, there's
now '''three''' sockets all listening on port X.
This obviously led to mis-matched TCP connections, and things went
downhill from there.
That being said, Snow Leopard doesn't exhibit this behavior if v6
sockets only allow incoming v6 connections. And technically, that is
exactly the behavior we want (we want v6 sockets to only accept
incoming v6 connections). So if we just change the flag to make our
v6 listening socket us this behavior, the problem on OS X goes away.
That's what this commit does -- it changes a 0 to a 1, indicating
"only let this v6 socket allow incoming v6 connections."
That was simple, wasn't it?
This commit was SVN r22788.
The following Trac tickets were found above:
Ticket 2039 --> https://svn.open-mpi.org/trac/ompi/ticket/2039
1. The code that looks at btl_tcp_if_exclude before doing a
modex_send uses strcmp rather than strncmp. That means that
"lo0" gets sent even though "lo" is excluded.
2. The code that determines whether a particular local TCP
interface can connect to a particular remote interface doesn't
check for loopback interfaces. With this fix, users can now
enable "lo" and be assured that it will only be used for intra-
node communication.
This commit was SVN r22762.