* Make sure the update.sh scripts all use OMPI_WANT_SMP_LOCKS
* Add sparc (32bit) assembly
- Memory barriers
- spinlocks
- emulate add/sub using the spinlock table with hashed lookups, as
suggested by the linux kernel folk (better than the other option,
requiring the counters only use 24 bits)
This commit was SVN r4429.
probe. For the iprobe normally we dont have to test, as the iprobe and this function are mutually exclusive protected by the same lock.
Now all intel *probe tests passe. The bug 1109 is closed :)
This commit was SVN r4426.
- DO NOT call matched when we match a probe/iprobe as it's not the real request. This function should be
called only when we match the request that will get the data from the fragment(s).
- DO NOT remove the fragment when we match a probe/iprobe request. The data should be available for the
real receive ...
The others modifs are just for speedup and indentation.
This commit was SVN r4425.
MCA_PML_BASE_RECV_REQUEST_RETURN (as we call the MCA_PML_BASE_RECV_REQUEST_INIT) in order to decrease
the ref count of the communicator and datatype.
This commit was SVN r4421.
- mca_base_param_set_int() -- set the value of an MCA int param from
the API
- mca_base_param_set_string() -- set the value of an MCA string param
from the API
- mca_base_param_unsed() -- unset a previously set MCA param (string
or int)
- mca_base_param_dump() -- obtain a listing of all currently
registered MCA params
- mca_base_param_dump_release() -- release the list of all registered
MCA params that was obtained from mca_base_param_dump()
See Doxygen docs in src/mca/base/mca_base_param.h for details of these
functions.
This commit was SVN r4415.
on communicators and datatypes. Now the management is done on the PML base and the others PML's just
have to use the corresponding macros from the base.
This commit was SVN r4406.
- Make the DEPENDENCIES a little clearner, and clear up a problem on at
least one platform
- Remove old/unnecessary mpirun2 rules
This commit was SVN r4397.
should I compute it here with les precision or should be done in the commit phase. Anyway correctly
detecting the overlap is a difficult problem.
This commit was SVN r4394.
parameters. These are intended to be used by the system itself for
passing information around, and are not generally intended for the
user to see or modify. Hence, we flag them as "internal" which causes
them not to show up in the ompi_info output (by default). This
"security through obscurity" is good enough -- it's really only
intended to keep casual observers away. The MCA parameters are still
there and are still settable (if a developer wants to set them).
To mark an MCA parameter as internal, simply call
mca_param_set_internal() with its index. I debated about adding a
flag to mca_param_register_[int|string](), but then would have
involved changing a *lot* of code throughout the base -- and internal
MCA parameters is certainly not the common case.
Note that you can provide the new ompi_info flag --internal in
conjunction with the --param flag to force it to also show all the
internal MCA params.
This commit was SVN r4373.
which Torsten ran across last week and I was just now able to
duplicate:
- priority == 0 is still a valid priority
- ensure to not compare against an empty list; if the list is empty,
just put the selectable item on it without any comparisons
- had wrong argument for ompi_list_get_next() (yay macros with no type
checking :-( ) that caused badness in some cases
This commit was SVN r4372.
files once. Turns out that linkers *aren't* smart enough to just
"figure it out" and ensure that we only get one copy (or perhaps it's
libtool who's trying to be too smart...?).
This commit was SVN r4371.
are generated at configure time, so there is no need to have them in the
tarball (and doing so can cause some really evil VPATH issues). The .in
files needed to make ompi_config.h and mpif.h are included in the tarball.
This commit was SVN r4367.
the test barfs on AIX, since it has some strange non-mangling of the
labels in the inline assembly that all the other platforms seem to be
ok making right. *sigh*
This commit was SVN r4365.
Update the count on the addr struct usage if and only if we match something (otherwise the field is set to null and we segfault).
This commit was SVN r4362.
this code (because it's used there):
- moved docs to path.h
- consolidated and removed some useless access functions
- brought up to current code standards
- updated wrapper compilers to use the new API
This commit was SVN r4346.
message out. Use stderr, and put a "yes, we're done with the error
message" marker at the end of the error message so that you know if
you saw it all or the message was truncated.
This commit was SVN r4333.
the way progress is done in the non-threaded case, adds a progress
thread for the threaded progress case (which should be made to use
George's progress engine when it is ready), and fixes some bugs with
request handling, especially with freeing requests (IO requests, that
is).
This commit was SVN r4332.
