1
1
Граф коммитов

92 Коммитов

Автор SHA1 Сообщение Дата
Tim Woodall
e17da43528 added allocator mca component type
This commit was SVN r1264.
2004-06-15 16:57:18 +00:00
Tim Prins
8f7b128941 first checkin of the memory allocation framework. It should work fine
except the align and realloc functions are not yet allocated.

This commit was SVN r1251.
2004-06-14 21:05:53 +00:00
Ralph Castain
4c28cb8b3d Okay, let's see if the system gets rid of the old registry files this time. Also added new location to configure.ac, and picked up a change to the sys_info header.
This commit was SVN r1239.
2004-06-12 10:56:32 +00:00
Jeff Squyres
29ab8d41d6 Fix some typos
This commit was SVN r1202.
2004-06-07 22:09:25 +00:00
David Daniel
563ac2a338 First pass of lam -> ompi conversion
This commit was SVN r1191.
2004-06-07 15:33:53 +00:00
Ralph Castain
1433bccfaf Update configure.ac to include the universe and registry directories.
This commit was SVN r1188.
2004-06-04 02:28:48 +00:00
Tim Woodall
3ce1b9852b re-enable builds of the test directory
This commit was SVN r1155.
2004-05-25 13:08:24 +00:00
Ralph Castain
8cc3ecb8e5 Remove the runtime-environment directory RTE/universe from the makefile list since it only exists on my machine! Update the sys_info header file
This commit was SVN r1143.
2004-05-20 17:38:54 +00:00
Ralph Castain
2d709f768a This adds into the system a utility for detecting what kind of system we are on. The utility provides a global variable ompi_system_info that contains information on the operating system, release and version, machine architecture, and network node name. I'll provide additional documentation via Doxygen once I figure out how to put it into the .h file.
Code was reviewed by Jeff - thanks Jeff!

This commit was SVN r1141.
2004-05-20 13:54:14 +00:00
George Bosilca
ba91bcf530 Check for long double
Avoid the lookup for sizeof(TYPE) if TYPE is not defined.

This commit was SVN r1085.
2004-04-23 20:52:42 +00:00
Prabhanjan Kambadur
dbb0db0909 Enabling topo module
This commit was SVN r1044.
2004-04-16 20:56:22 +00:00
Prabhanjan Kambadur
832ec55ebf introducing 2 new AM_CONDITIONALS so that tf77 bindings are not built when the user does not want it or when the fortran 77 compiler is not found. Right now, the libraries were built but not snarfed into the big library. This was unneccessary
This commit was SVN r1029.
2004-04-13 23:05:31 +00:00
Brian Barrett
6282f1a8e6 switch from F77/FC to LAM_F77/LAM_F90 so that we don't have to assign "none"
into either F77 or F90, which caused badness in the libtool test world.

This commit was SVN r999.
2004-03-28 10:52:58 +00:00
Brian Barrett
acff369327 properly fix bug 819 - initialize the right variable to 0.
This commit was SVN r997.
2004-03-28 10:20:32 +00:00
Brian Barrett
61557f0156 Fix some of the Fortran issues by unifying all the various on/off
flags to LAM_WANT_F77_BINDINGS and LAM_WANT_F90_BINDINGS.  Also
changed the compiler flag for the Fortran 90/95/whatever compiler to
FCFLAGS to match what AC was already doing.

This commit was SVN r996.
2004-03-28 10:02:38 +00:00
Brian Barrett
308ef4c769 * add alloca.h if we have it to make datatypes happy with gcc on linux
This commit was SVN r990.
2004-03-27 05:26:31 +00:00
Jeff Squyres
69263505b7 Bunches of changes to make "make dist" work properly
This commit was SVN r966.
2004-03-26 04:28:42 +00:00
Jeff Squyres
f871b73606 First cut at mpif.h; probably needs to be heavily modified.
This commit was SVN r941.
2004-03-19 17:27:35 +00:00
Jeff Squyres
ca5fa108f5 Change from "CVS" version to "SVN" version
This commit was SVN r928.
2004-03-19 06:05:27 +00:00
Jeff Squyres
251f68b94f Re-separate out the always-installable headers from the
LAM-development headers.

This commit was SVN r925.
2004-03-18 21:35:28 +00:00
Prabhanjan Kambadur
e18d26e901 Fortran has to be enabled by seperate flags. I was using the same flags as used by teh C interface which built fortran bindings even when the user has disabled them
This commit was SVN r924.
2004-03-18 21:34:00 +00:00
Jeff Squyres
51109ef70a Add some sub-Makefile.am's in the libevent tree
This commit was SVN r903.
2004-03-17 22:28:18 +00:00
Jeff Squyres
1ee935eac6 Perhaps the final commit of the Great Source Reorganization...
This commit was SVN r891.
2004-03-17 21:11:52 +00:00
Jeff Squyres
aa8ef8bd0a Build system mods for the new directory organization
This commit was SVN r879.
2004-03-17 19:06:06 +00:00
Jeff Squyres
86d2c8aca9 Fixes for VPATH builds
This commit was SVN r869.
2004-03-17 03:58:21 +00:00
Jeff Squyres
f462a7edfb - Make fortran pthread tests work properly
- Add THREAD_*FLAGS to all the relevant places
- Propagate THREAD_*FLAGS to the wrappers
- Add src/mpi/win/Makefile to AC_OUTPUT

This commit was SVN r868.
2004-03-17 03:04:44 +00:00
Jeff Squyres
7e97aafb66 Ensure that we always have fortran sizeof's and alignments, and put
them in run-time variables so that they can be overridden at run-time

This commit was SVN r867.
2004-03-16 23:54:51 +00:00
Tim Woodall
1f915b0b4a - made lam_mutex_t a lam_object
- implemented module exchange
- debugging p2p init
- added condition variables
- added threading test code

This commit was SVN r817.
2004-03-03 16:44:41 +00:00
Jeff Squyres
dce0db9804 We really require AC >= 2.58
This commit was SVN r802.
2004-02-14 06:02:04 +00:00
Jeff Squyres
0b2313082c Fix for bug 736 -- added sizeof/alignment for C++ and Fortran types
This commit was SVN r784.
2004-02-13 18:05:49 +00:00
Jeff Squyres
afb6d28cbf First cut at dramatically decreasing the overhead and increasing the
speed of autogen.sh and configure: find modules that don't need to have
a separate configure script during autogen.sh and set them up to be
part of LAM's build process.  This means that we don't have to run the GNU
tools nearly as much during autogen.sh and that we don't have to run
nearly as many sub-configure scripts during configure.

This works fine for statically-built MCA modules, but doesn't seem to work
properly yet for dynamically-built modules.  More coming soon, but I wanted
to get this in for others to use.

This commit was SVN r756.
2004-02-13 03:58:56 +00:00
Jeff Squyres
1d4a6c1dc6 Fix for bug 732 -- don't let the top-level automake traverse down into
src/lam/event.

This commit was SVN r755.
2004-02-13 03:51:35 +00:00
Jeff Squyres
965d876030 Integrate libevent into liblam
This commit was SVN r739.
2004-02-12 18:54:06 +00:00
Jeff Squyres
d158849b0d Try again for profiling. :-)
This commit was SVN r701.
2004-02-10 20:12:37 +00:00
Jeff Squyres
eb4ebf7619 Add missing Makefile's to the AC_OUTPUT list
This commit was SVN r659.
2004-02-06 16:32:46 +00:00
Prabhanjan Kambadur
e195983847 Changing the F77 null condition. If not fortran compiler is found, then
F77 is set too be "none".

This commit was SVN r647.
2004-02-04 02:23:55 +00:00
Jeff Squyres
71358e7e87 - Ensue to use LAM_BASIC
- Replace some of the Fint detection code with spaces, not tabs :-)

This commit was SVN r494.
2004-01-20 23:16:25 +00:00
Prabhanjan Kambadur
68655dedbc addingn laminfo's Makefile to list of generated files
This commit was SVN r477.
2004-01-19 05:47:09 +00:00
Prabhanjan Kambadur
de1e01d252 OOPS ..... removed a few comments which should be present
This commit was SVN r471.
2004-01-19 01:41:40 +00:00
Prabhanjan Kambadur
6bd87098ba Adding test for MPI_Offset. Also introducing an if statement to check for the size of MPI_Fint only when f77 is present. Else MPI_Fint is defined to be an int.
This commit was SVN r470.
2004-01-19 01:33:03 +00:00
Jeff Squyres
5b66cb3f7a Updates to make MPI_Fint work
This commit was SVN r461.
2004-01-17 04:10:01 +00:00
Prabhanjan Kambadur
0c221c80fd These files help in determining MPI_Fint (in terms of relevant C types).
configure.ac #define's MPI_Fint to be of int32_t, int64_t or int depending
on the configuration details

This commit was SVN r454.
2004-01-16 22:16:25 +00:00
Jeff Squyres
e2a68b22de Add some pretty messages about wrapper compiler flags
This commit was SVN r448.
2004-01-16 02:03:41 +00:00
Jeff Squyres
42a849197a - We know that libltdl is not picky-clean. So don't use the picky-clean
CFLAGS when configuring (and therefore compiling) libltdl.
- Move some of the output messages around

This commit was SVN r443.
2004-01-16 01:31:50 +00:00
Jeff Squyres
603e7fb0ee Add new directories
This commit was SVN r418.
2004-01-15 06:12:34 +00:00
Prabhanjan Kambadur
5c1aca22c7 CVSc: Changed to have 2 new AM_CONDITIONALS, WANT_MPI_BINDINGS_LAYER
and WANT_PMPI_BINDINGS_LAYER. These determine whether src/mpi/interface/c and
src/mpi/interface/c/profile are built. Makefile.am's were changed to reflect
the same change. Now the top layer always builds MPI bindings and the lower
layer always builds PMPI bindings
: ----------------------------------------------------------------------

This commit was SVN r393.
2004-01-15 01:11:59 +00:00
Rich Graham
99d52894f6 changing support_finalize to return status.
Start to put in a unit test for the list functionality.

This commit was SVN r390.
2004-01-15 00:59:26 +00:00
Rich Graham
29fdbb6a7c start to add testing support library.
This commit was SVN r384.
2004-01-15 00:07:17 +00:00
Jeff Squyres
ae93b4fffb Add libltdl into the build process.
This commit was SVN r383.
2004-01-14 23:50:58 +00:00
Jeff Squyres
4e9e99e330 Add fixes for VPATH builds
This commit was SVN r378.
2004-01-14 19:57:37 +00:00