1
1
Граф коммитов

80 Коммитов

Автор SHA1 Сообщение Дата
Jeff Squyres
4d8a187450 Two major things in this commit:
* New "op" MPI layer framework
 * Addition of the MPI_REDUCE_LOCAL proposed function (for MPI-2.2)

= Op framework =

Add new "op" framework in the ompi layer.  This framework replaces the
hard-coded MPI_Op back-end functions for (MPI_Op, MPI_Datatype) tuples
for pre-defined MPI_Ops, allowing components and modules to provide
the back-end functions.  The intent is that components can be written
to take advantage of hardware acceleration (GPU, FPGA, specialized CPU
instructions, etc.).  Similar to other frameworks, components are
intended to be able to discover at run-time if they can be used, and
if so, elect themselves to be selected (or disqualify themselves from
selection if they cannot run).  If specialized hardware is not
available, there is a default set of functions that will automatically
be used.

This framework is ''not'' used for user-defined MPI_Ops.

The new op framework is similar to the existing coll framework, in
that the final set of function pointers that are used on any given
intrinsic MPI_Op can be a mixed bag of function pointers, potentially
coming from multiple different op modules.  This allows for hardware
that only supports some of the operations, not all of them (e.g., a
GPU that only supports single-precision operations).

All the hard-coded back-end MPI_Op functions for (MPI_Op,
MPI_Datatype) tuples still exist, but unlike coll, they're in the
framework base (vs. being in a separate "basic" component) and are
automatically used if no component is found at runtime that provides a
module with the necessary function pointers.

There is an "example" op component that will hopefully be useful to
those writing meaningful op components.  It is currently
.ompi_ignore'd so that it doesn't impinge on other developers (it's
somewhat chatty in terms of opal_output() so that you can tell when
its functions have been invoked).  See the README file in the example
op component directory.  Developers of new op components are
encouraged to look at the following wiki pages:

  https://svn.open-mpi.org/trac/ompi/wiki/devel/Autogen
  https://svn.open-mpi.org/trac/ompi/wiki/devel/CreateComponent
  https://svn.open-mpi.org/trac/ompi/wiki/devel/CreateFramework

= MPI_REDUCE_LOCAL =

Part of the MPI-2.2 proposal listed here:

    https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/24

is to add a new function named MPI_REDUCE_LOCAL.  It is very easy to
implement, so I added it (also because it makes testing the op
framework pretty easy -- you can do it in serial rather than via
parallel reductions).  There's even a man page!

This commit was SVN r20280.
2009-01-14 23:44:31 +00:00
Jeff Squyres
5ac6a4387c Fix MPI_CART_GET to take an array of logicals for the periods argument
(not integers)

This commit was SVN r19099.
2008-07-30 20:29:23 +00:00
Jeff Squyres
9f4d4c4312 Fixes trac:1409: ensure that the C++, F77, and F90 bindings libraries
are properly linked against libmpi.la.

This required a little creative AM usage, inspired by discussion on
OMPI devel list:

 * Make a new ompi/mpi/f77/Makefile_f77base.include; effectively move
   the building of the f77 "base" glue stuff (libmpi_f77base.la) into
   this Makefile and away from ompi/mpi/f77/Makefile.am.  The sources
   in question require some specific CPPFLAGS, so we couldn't just add
   the raw sources into libmpi_la_SOURCES, unfortunately.
 * Include this new Makefile in the top-level ompi/Makefile.am
 * The libmpi_f77base.la LT convenience library was already sucked
   into libmpi.la; breaking it out into its own Makefile allows us
   to build it earlier and therefore complete buidling libmpi.la
   earlier.
 * Side effect: the ompi/mpi/Makefile.am is now mostly unnecessary; it
   no longer specifies a SUBDIRS for each of the bindings directories
   to traverse into (since they are now in the top-level SUBDIRS).  As
   such, the man pages are now also now included in the top-level
   ompi/Makefile.am.

The end of the result is that libmpi.la -- including a few sources
from mpi/f77 -- is fully built before the C++, F77, and F90 bindings
are built.  Therefore, the C++, F77, and F90 bindings libraries can
all link against libmpi.la.

This commit was SVN r19040.

The following Trac tickets were found above:
  Ticket 1409 --> https://svn.open-mpi.org/trac/ompi/ticket/1409
2008-07-25 21:18:05 +00:00
Jeff Squyres
213679488a Fix F90 interface for MPI_CART_CREATE. Thanks to Michal Charemza for
reporting the problem.

This commit was SVN r17395.
2008-02-07 12:46:32 +00:00
Rainer Keller
76b95a4ee7 - Just indentation and spelling
This commit was SVN r17228.
2008-01-25 10:36:42 +00:00
Jeff Squyres
26d8fe70c3 Fixes trac:1029: add in support for MPI_CONVERSION_FN_NULL.
This commit brings over all the work from the /tmp-public/datarep
branch.  See commits r16855, r16859, r16860 for the highlights of what
was done.

This commit was SVN r16891.

The following SVN revisions from the original message are invalid or
inconsistent and therefore were not cross-referenced:
  r16855
  r16859
  r16860

The following Trac tickets were found above:
  Ticket 1029 --> https://svn.open-mpi.org/trac/ompi/ticket/1029
2007-12-07 13:09:07 +00:00
Shiqing Fan
0f468f3668 - Remove the solution and project files, will commit them later.
This commit was SVN r15705.
2007-07-31 17:07:02 +00:00
Shiqing Fan
4d7b349cdb - Add VC8 solution and project files.
- If one wants to use this solution, remember to unload the project 'orte-restart' which is currently not working for Windows.

This commit was SVN r15680.
2007-07-30 11:05:34 +00:00
Rainer Keller
d3372729bb - Support for opt. MPI_REAL2 (who has that?)
to make checks for MPI-implementations fail in the right way ,-]
 - check in configure.ac
 - BINARY INCOMPATIBLE change to mpif-common.h
   (if implemented the *right* way)
   Actually OMPI_F90_CHECK takes two arguments, not three.
 - Only have corresponding C-Type, if the opt. Fortran
   type is really supported,
   Otherwise pass ompi_mpi_unavailable to DECLARE_MPI_SYNONYM_DDT;
 - Reviewed by George and Jeff

This commit was SVN r15133.
2007-06-19 05:03:11 +00:00
Brian Barrett
644708a4b4 For modern MACOSX_DEPLOYMENT_TARGETs, allow the Fortran 90 bindings to be
built as shared libraries.

This commit was SVN r14859.
2007-06-05 04:04:44 +00:00
Rainer Keller
43ce11d578 - Fortran 90 standard requires no name at:
end interface
   (ifort complains, pgf90 chokes on it).

This commit was SVN r14785.
2007-05-29 15:37:44 +00:00
Jeff Squyres
86f48bf0d7 Fix ordering of dimensions for status arrays. Thanks to Randy Bramley
for noticing the problem.

This commit was SVN r14591.
2007-05-05 05:02:02 +00:00
Jeff Squyres
53c4bcccd5 Fix a problem noted by Tim Campbell: the type of the first parameter
to the F90 binding for MPI_INITIALIZED was wrong (should have been
logical, not integer).

Fixes trac:782.

This commit was SVN r13170.

The following Trac tickets were found above:
  Ticket 782 --> https://svn.open-mpi.org/trac/ompi/ticket/782
2007-01-17 22:44:47 +00:00
Jeff Squyres
bc93583184 Fix the SIZEOF implementation for complex's. We want to output the
size of the complex type as determined by configure; not the size of
the next larger complex type (i.e., a complex*N is 2 real*(N/2)'s, not
2 real*N's).

This commit was SVN r12421.
2006-11-03 15:52:46 +00:00
Jeff Squyres
a5928a2a89 Ralf W. swears to me that this is a product of a very, very old AM bug
and is no longer necessary (especially since we enforce pretty recent
Gnu tools).

This commit was SVN r11855.
2006-09-27 21:33:26 +00:00
Brian Barrett
880730fcf6 * Make sure to add the FCFLAGS_f90 variable to FCFLAGS. This is the magic
that the compiler might need to inform the compiler that a .f90 extension
  means "this is Fortran 90 code".  Fortran compilers are so weird.

  refs trac:284

This commit was SVN r11280.

The following Trac tickets were found above:
  Ticket 284 --> https://svn.open-mpi.org/trac/ompi/ticket/284
2006-08-21 14:15:55 +00:00
Brian Barrett
4176e61049 * Add support for building the F90 bindings library as a shared library
on almost all platforms (except OS X... sigh...).  This is the merge 
  of r10846 - 10894 from the tmp/f90-shared branch to the trunk.

This commit was SVN r11103.

The following SVN revisions from the original message are invalid or
inconsistent and therefore were not cross-referenced:
  r10846
2006-08-03 00:17:31 +00:00
Jeff Squyres
520147f209 Clean up the Fortran MPI sentinel values per problem reported on the
users mailing list:

  http://www.open-mpi.org/community/lists/users/2006/07/1680.php

Warning: this log message is not for the weak.  Read at your own
risk.

The problem was that we had several variables in Fortran common blocks
of various types, but their C counterparts were all of a type
equivalent to a fortran double complex.  This didn't seem to matter
for the compilers that we tested, but we never tested static builds
(which is where this problem seems to occur, at least with the Intel
compiler: the linker compilains that the variable in the common block
in the user's .o file was of one size/alignment but the one in the C
library was a different size/alignment).

So this patch fixes the sizes/types of the Fortran common block
variables and their corresponding C instantiations to be of the same
sizes/types. 

But wait, there's more.

We recently introduced a fix for the OSX linker where some C versions
of the fortran common block variables (e.g.,
_ompi_fortran_status_ignore) were not being found when linking
ompi_info (!).  Further research shows that the code path for
ompi_info to require ompi_fortran_status_ignore is, unfortunately,
necessary (a quirk of how various components pull in different
portions of the code base -- nothing in ompi_info itself requires
fortran or MPI knowledge, of course).

Hence, the real problem was that there was no code path from ompi_info
to the portion of the code base where the C globals corresponding to
the Fortran common block variables were instantiated.  This is because
the OSX linker does not automatically pull in .o files that only
contain unintialized global variables; the OSX linker typically only
pulls in a .o file from a library if it either has a function that is
used or have a global variable that is initialized (that's the short
version; lots of details and corner cases omitted).  Hence, we changed
the global C variables corresponding to the fortran common blocks to
be initialized, thereby causing the OSX linker to pull them in
automatically -- problem solved.  At the same time, we moved the
constants to another .c file with a function, just for good measure.

However, this didn't really solve the problem:

1. The function in the file with the C versions of the fortran common
   block variables (ompi/mpi/f77/test_constants_f.c) did not have a
   code path that was reachable from ompi_info, so the only reason
   that the constants were found (on OSX) was because they were
   initialized in the global scope (i.e., causing the OSX compiler to
   pull in that .o file).

2. Initializing these variable in the global scope causes problems for
   some linkers where -- once all the size/type problems mentioned
   above were fixed -- the alignments of fortran common blocks and C
   global variables do not match (even though the types of the Fortran
   and C variables match -- wow!).  Hence, initializing the C
   variables would not necessarily match the alignment of what Fortran
   expected, and the linker would issue a warning (i.e., the alignment
   warnings referenced in the original post).

The solution is two-fold:

1. Move the Fortran variables from test_constants_f.c to
   ompi/mpi/runtime/ompi_mpi_init.c where there are other global
   constants that *are* initialized (that had nothing to do with
   fortran, so the alignment issues described above are not a factor),
   and therefore all linkers (including the OSX linker) will pull in
   this .o file and find all the symbols that it needs.

2. Do not initialize the C variables corresponding to the Fortran
   common blocks in the global scope.  Indeed, never initialize them
   at all (because we never need their *values* - we only check for
   their *locations*).  Since nothing is ever written to these
   variables (particularly in the global scope), the linker does not
   see any alignment differences during initialization, but does make
   both the C and Fortran variables have the same addresses (this
   method has been working in LAM/MPI for over a decade).

There were some comments here in the OMPI code base and in the LAM
code base that stated/implied that C variables corresponding to
Fortran common blocks had to have the same alignment as the Fortran
common blocks (i.e., 16).  There were attempts in both code bases to
ensure that this was true.  However, the attempts were wrong (in both
code bases), and I have now read enough Fortran compiler documentation
to convince myself that matching alignments is not required (indeed,
it's beyond our control).  As long as C variables corresponding to
Fortran common blocks are not initialized in the global scope, the
linker will "figure it out" and adjust the alignment to whatever is
required (i.e., the greater of the alignments).  Specifically (to
counter comments that no longer exist in the OMPI code base but still
exist in the LAM code base):

- there is no need to make attempts to specially align C variables
  corresponding to Fortran common blocks
- the types and sizes of C variables corresponding to Fortran common
  blocks should match, but do not need to be on any particular
  alignment 

Finally, as a side effect of this effort, I found a bunch of
inconsistencies with the intent of status/array_of_statuses
parameters.  For all the functions that I modified they should be
"out" (not inout).

This commit was SVN r11057.
2006-07-31 15:07:09 +00:00
Rainer Keller
ac58e85c83 - Add the missing collective (and other) functions to mpi.f03
- Correct intent(out) to inout for various recvbufs to match
   standards possibility for MPI_IN_PLACE.

This commit was SVN r10868.
2006-07-18 18:12:09 +00:00
Jeff Squyres
9a679644c2 Arf. Don't output the body of the WTICK or WTIME functions in the
module header if we're not doing small.

This commit was SVN r10475.
2006-06-22 13:20:01 +00:00
Jeff Squyres
723b6e50a9 George suggested a better way to make WTICK and WTIME -- be consistent
with the other methodology even if there are no choice buffers and no
special constants.  But it keeps the Makefile.am simple and the
methodology consistent.

This commit was SVN r10462.
2006-06-21 19:07:09 +00:00
Jeff Squyres
48e9a72c47 Add the missing files -- they're svn:ignored because of all the
generated files.

This commit was SVN r10451.
2006-06-21 14:11:12 +00:00
Jeff Squyres
720f38efc5 Fix for MPI_WTICK / MPI_WTIME F90 bindings issue. The previous hope
was that declaring the type of MPI_WTICK and MPI_TIME in mpif-common.h
would allow the F90 bindings to call through to the back end f77
function and have the right return type.  But upon reflection, that's
silly -- we were just declaring the variables MPI_WTICK and MPI_WTIME
that were of type double precision.  Duh.

So add some fixed (non-generated) wrapper F90 functions to call the
back-end *C* MPI_WTICK and MPI_TIME functions (vs. the back end *F77*
functions).  We have to call the back-end C functions because there's
a name conflict if we try to call the back-end F77 functions -- for
the same reasons that we can't "implicitly" define MPI_WTIME and
MPI_WTICK in the f90 module, we can't call such an implicitly-defined
function.  So we had to add new back-end C functions that are directly
callable from Fortran, the easiest implementation of which was to
provide 4 one-line functions for each (rather than muck around with
weak symbols).

This commit was SVN r10448.
2006-06-21 13:44:20 +00:00
Jeff Squyres
8b8bf363c4 Add missing svn:executable property
This commit was SVN r10283.
2006-06-10 10:59:01 +00:00
Jeff Squyres
02d8a46d5f Fix for ticket #89.
* Change the type of Fortan's MPI_STATUSES_IGNORE to double complex
so that it will never possibly be mistaken for a real status (i.e.,
integer(MPI_STATUS_SIZE)), particularly in the F90 bindings.  See
comment in mpif-common.h explaining this (analogous argument to
MPI_ARGVS_NULL for MPI_COMM_SPAWN_MULTIPLE).
 * Add second interfaces for the following functions that take a double
complex (i.e., MPI_STATUSES_IGNORE).  This required adding the second
interface in mpi-f90-interfaces.h[.sh] and then generating new wrapper
functions to call the back-end F77 function for each of these four, so
we added 4 new files in ompi/mpi/f90/scripts/ and updated the various
Makefile.am's to match:
   * MPI_TESTALL
   * MPI_TESTSOME
   * MPI_WAITALL
   * MPI_WAITSOME

The XSL is now not in sync with the scripts.  Although I suppose that
that is becoming less and less important (because it does not impact
the end user at all -- to be 100% explicit, no release should ever be
held up because the XSL is out of sync), but it will probably be
important when we go to fix the "large" interface; so it's still worth
fixing... for now...

This commit was SVN r10281.
2006-06-09 23:40:20 +00:00
Craig E Rasmussen
8a22272ffb Changed to use procedure alias (when names too long).
This commit was SVN r10145.
2006-05-31 15:06:44 +00:00
Craig E Rasmussen
4cd13f07c4 Changed to use procedure alias (when names too long).
This commit was SVN r10144.
2006-05-31 15:04:38 +00:00
Jeff Squyres
5f356edb64 Bring over changes from the /tmp/fortran-stuff series:
- Make the F90 bindings compile and link properly with gfortran 4.0,
  4.1, Intel 9.0, PGI 6.1, Sun (don't know version offhand -- the most
  current as of this writing, I think), and NAG 5.2, although some
  have limitations (e.g., NAG can't seem to handle the medium and
  large sizes)
- Building the F90 "small" module size is now the default, even for
  developers
- Split up mpif.h into multiple files because parts of it were toxic
  to the F90 bindings
- Properly specify unsized/unshaped arrays to make the bindings work
  on all known compilers
- Make ompi_info show Fortran 90 bindings size
- XML somewhat lags the generated scripts as of this commit, but
  functionality was my main goal -- the XML can be updated later (if
  at all).

This commit was SVN r10118.
2006-05-30 14:37:41 +00:00
Jeff Squyres
faf63c68f8 Merge over from the /tmp/fortran-stuff branch
- split mpif.h into mpif.h and mpif-common.h[.in]
- mpif-common.h is included by various f90 things and contains output
  from configure
- mpif.h defines some f77-specific stuff and then includes
  mpif-common.h 

This commit was SVN r9997.
2006-05-20 02:15:49 +00:00
Jeff Squyres
800ba39152 Fixes for:
- make dist
- proper inclusion of files between trivial and small
- vpath

This commit was SVN r9964.
2006-05-18 11:26:52 +00:00
Jeff Squyres
a7c854f878 Add some comments explaining why these functions have an interface but
no definition.

This commit was SVN r9962.
2006-05-17 22:22:29 +00:00
Jeff Squyres
c3aeaed77b SPAWN_MULTIPLE should actually be in the "small" size, not the
"medium" size.

This commit was SVN r9961.
2006-05-17 21:58:48 +00:00
Craig E Rasmussen
94f463c924 Changes to remove attr_fn functions from the mpi namespace/module.
This commit was SVN r9960.
2006-05-17 21:31:04 +00:00
Jeff Squyres
95c1f12b97 Missed this one yesterday
This commit was SVN r9958.
2006-05-17 16:31:25 +00:00
Jeff Squyres
1f8cc1af65 Add an extra -I so that compilers can find mpi-f90-interfaces.h in a
vpath build.

This commit was SVN r9937.
2006-05-16 17:30:42 +00:00
Jeff Squyres
d3ff507bc2 Add explicit path to mpi-f90-interfaces.h.sh. Blast; I could have
sworn that I committed this yesterday.

This commit was SVN r9935.
2006-05-16 15:00:23 +00:00
Jeff Squyres
3e60cd163e Ensure that mpi_sizeof.f90 is added to the build
This commit was SVN r9925.
2006-05-15 19:32:01 +00:00
Jeff Squyres
332954e168 Spell the AC_SUBST'ed names properly
This commit was SVN r9924.
2006-05-15 19:31:38 +00:00
Jeff Squyres
171585ff34 Ensure to put implicit none *before* the header file includes
This commit was SVN r9923.
2006-05-15 19:31:01 +00:00
Craig E Rasmussen
13c113b39c Changed INOUT to IN in MPI_*_SET_ATTR functions.
This commit was SVN r9922.
2006-05-15 17:13:20 +00:00
Craig E Rasmussen
a547a64f93 Changed INOUT to IN in MPI_*_SET_ATTR functions.
This commit was SVN r9916.
2006-05-15 16:00:04 +00:00
Craig E Rasmussen
8a16f844ec Changes MPI_Intercomm_merge high variable from integer to logical
This commit was SVN r9783.
2006-05-01 18:58:35 +00:00
Craig E Rasmussen
0c584d783e Changes MPI_Intercomm_merge high variable from integer to logical
This commit was SVN r9782.
2006-05-01 18:58:05 +00:00
Jeff Squyres
22dbda116e Add in oodles of missing files
This commit was SVN r9767.
2006-04-28 02:22:19 +00:00
Craig E Rasmussen
888c19b9ee Changed bool Fortran type to logical
This commit was SVN r9766.
2006-04-27 22:49:06 +00:00
Craig E Rasmussen
bd77c2b212 Changed bool Fortran type to logical
This commit was SVN r9765.
2006-04-27 22:48:07 +00:00
Craig E Rasmussen
9333a2628b Added counter to output name and changed bool Fortran type to logical
This commit was SVN r9764.
2006-04-27 22:46:48 +00:00
Jeff Squyres
0556c129df Fix MPI_*_GET_ATTR flag parameter to be of type logical. Thanks to
Michael Kluskens for pointing this out.

This commit was SVN r9676.
2006-04-21 04:31:49 +00:00
Craig E Rasmussen
6de9755919 Added counter to output macro name
This commit was SVN r9675.
2006-04-20 21:30:28 +00:00
Craig E Rasmussen
8646604497 Fixed to match Jeff S's changes to mpi-f90-interfaces.h.sh
This commit was SVN r9674.
2006-04-20 21:18:04 +00:00