ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
4 918 коммитов 3 Ветки 0 Теги 130 MiB
99340fd8d3
Граф коммитов

4918 Коммитов

This Branch
This Branch
Все ветки
Автор SHA1 Сообщение Дата
George Bosilca
99340fd8d3 I don't think this one was intended to go inside ... 
This commit was SVN r6404.
 2005-07-08 21:57:04 +00:00
Brian Barrett
c0896ebe99 * have to define the constant in the negative case as well :( 
This commit was SVN r6403.
 2005-07-08 21:50:21 +00:00
Brian Barrett
79a1093fcc * add builddir/include to the list of -Is so that ompi_config.h can be 
found when VPATH building.  Thanks to George for catching this

This commit was SVN r6402.
 2005-07-08 21:05:18 +00:00
Josh Hursey
de5e0d4f2c (Re-)Added two MCA Parameters that must have been lost in the merge way back when: 
* mpi_show_mca_params
   If set to true, this turns on the dumping of all MCA parameters when MPI_INIT is called. 
   Only the 'rank 0' processes will print the parameters.

* mpi_show_mca_params_file
   (This value is only used if the first argument is set to true) If this value is non-NULL 
   it specifies the file to put the dump into. This file can then be used as input to mpirun 
   for debugging purposes. If this value is not set (and mpi_show_mca_params is set) then 
   the parameters are dumped to stdout.

This commit was SVN r6401.
 2005-07-08 21:01:37 +00:00
Tim Prins
d4151fa9fd properly fix the usage of the app pointer array by checking for NULLs instead of forcing it to be the same size as the number of entries 
This commit was SVN r6395.
 2005-07-08 18:48:25 +00:00
Brian Barrett
4fc6224af2 * allow mpicc to build applications that use internal headers by adding 
the right -I flags if we want devel headers

This commit was SVN r6393.
 2005-07-08 18:26:01 +00:00
Brian Barrett
0ae16f2ab7 * add local hook to remove static-components.h in distclean target. The 
files are generated by configure, and not part of the tarball, so
  distclean would be the right place to remove them.

This commit was SVN r6390.
 2005-07-08 13:54:12 +00:00
Brian Barrett
e4644c407c * as requested by ralph, include orted.h in the list of developer headers 
This commit was SVN r6389.
 2005-07-08 13:46:06 +00:00
George Bosilca
b226af1176 Tolerate the datatype with ZERO elements. They can have an optimized expression (sic!) ... Anyway there is a 
warning message when we try to pack suck kind of datatypes.

This commit was SVN r6388.
 2005-07-08 07:00:18 +00:00
George Bosilca
d38c8b6895 Be nice and remember the real type of the last data. Optimizing the optimization leads to strange behaviours ... 
This commit was SVN r6387.
 2005-07-08 06:50:43 +00:00
George Bosilca
6c6b18eb20 Quit when there is no more space in the iovec. 
This commit was SVN r6386.
 2005-07-08 06:49:07 +00:00
George Bosilca
9a448ece15 A more cleaner version of the pack function. It does not (yet!) have all the optimizations 
from the old version but they will get inside quickly. 

This commit was SVN r6381.
 2005-07-07 23:07:43 +00:00
George Bosilca
e3e49ae42f Force the usage of the new pack function. 
This commit was SVN r6380.
 2005-07-07 23:06:39 +00:00
George Bosilca
c502b6b38c Declare the global "local" convertor. This one can be used for all pack/unpack 
direct calls.

This commit was SVN r6379.
 2005-07-07 23:06:04 +00:00
George Bosilca
d0df88316e Declare and initialize the 2 predefined convertors. 
This commit was SVN r6378.
 2005-07-07 23:03:59 +00:00
George Bosilca
d796186911 If there is no datatype attached to the convertor there is nothing to release. This bug 
happens only with the 2 predefined convertors ...

This commit was SVN r6377.
 2005-07-07 23:03:19 +00:00
George Bosilca
8dea323088 Remember the last type when we do the datatype optimization. 
This commit was SVN r6376.
 2005-07-07 23:01:51 +00:00
George Bosilca
a7aa1a76bc Correctly use the new convertor. Now we have 2 generic predefined convertors. One is for 
the external32 conversions (as specified in the MPI standard) the other one is the local
convertor that can be used for MPI_Pack and MPI_Unpack functions.

This commit was SVN r6375.
 2005-07-07 23:00:55 +00:00
Tim Woodall
11f1f067c2 use the btls min rdma size as the cutoff (rdma offset) for the send protocol 
This commit was SVN r6372.
 2005-07-07 22:08:56 +00:00
Tim Woodall
a231d53666 corrections to frag size 
This commit was SVN r6371.
 2005-07-07 21:38:37 +00:00
Tim Woodall
eabdb860bc tuning 
This commit was SVN r6370.
 2005-07-07 20:58:57 +00:00
Tim Prins
5cdf0803d4 make the app pointer array blocksize 1 so the the size of the pointer array is the same as the number of apps. This was causing a segfault when trying to launch multiple apps. 
This commit was SVN r6368.
 2005-07-07 18:01:26 +00:00
Tim Woodall
e0c8991a6e checkpoint 
This commit was SVN r6367.
 2005-07-07 16:56:58 +00:00
Brian Barrett
e381ac08d9 * fix for a fix for a bug. mca/common doesn't have a common.h, so it didn't 
look like a framework.  Add special exception for common.

This commit was SVN r6365.
 2005-07-07 04:09:28 +00:00
Tim Woodall
269a30fb0e typo 
This commit was SVN r6364.
 2005-07-06 22:38:39 +00:00
George Bosilca
ff3e243be5 Similar with close, PTL open should not be called directly. The PML will do it anyway. 
This commit was SVN r6363.
 2005-07-06 19:22:16 +00:00
Ralph Castain
c72621c90e Add some finalization logic when universe isn't found and orte_init returns per MCA parameter. 
This commit was SVN r6362.
 2005-07-06 18:41:40 +00:00
Brian Barrett
6c232fb9ff * remove version file logic (now unused) from the no-configure component 
code
* Add code to create m4_defined lists in mca_no_configure_components.m4 that
  list all the projects (opal, orte, ompi, etc.), all the frameworks in each
  project, and all the no-configure components in each framework.  These will
  eventually be used to remove all those lists from ompi_mca.m4

This does not require an autogen / configure.  Won't hurt, but not
required.

This commit was SVN r6361.
 2005-07-06 18:38:57 +00:00
George Bosilca
407b921cc3 Protect the base PTL close against multiple call of close. The problem came mainly from the 
fact that we have several PML and that each one will call the PTL close in their own
close function.

This commit was SVN r6360.
 2005-07-06 18:26:16 +00:00
George Bosilca
2f83301c82 Do not call the base PTL close directly. This is the PML job. 
This commit was SVN r6359.
 2005-07-06 18:25:03 +00:00
Tim Woodall
c860b92011 don't allocate to nodes that aren't valid (e.g. front end) - ignore them 
rather then giving an error message

This commit was SVN r6358.
 2005-07-06 17:55:01 +00:00
George Bosilca
8e30d2e7cb Update the MX component using the MX library version 1.0. However, it require a patched version. I will add a configure test to see if the MX is patched or not (not patched version 
lack the register_match callback).

This commit was SVN r6357.
 2005-07-06 16:37:48 +00:00
Brian Barrett
48cbb249ab * Fix error with the use of expr. On George's suggestion, don't use expr for 
the comparison  -- instead, use test -lt

This commit was SVN r6356.
 2005-07-06 13:18:44 +00:00
Brian Barrett
465e206216 * fill in some more of the chunk mamagement code (chunks == array of memory 
used for the rolling recv buffers for non-RDMA messages)

This commit was SVN r6355.
 2005-07-05 22:15:35 +00:00
Brian Barrett
d4bd7252a0 * checkpoint - added a bunch of infrastructure for sends 
This commit was SVN r6353.
 2005-07-05 21:14:29 +00:00
Josh Hursey
3d9d67eae9 Initialize size_count to count to avoid non-deterministic behaviour when 
we use it later.

This commit was SVN r6352.
 2005-07-05 20:59:35 +00:00
Jeff Squyres
888f0c5afd Remove the EXTRA_DIST=VERSION stuff from all the Makefile.am's so that 
"make dist" can succeed.  Duh.  :-\

This commit was SVN r6351.
 2005-07-05 19:01:47 +00:00
Brian Barrett
d9fa62d1f2 * checkpoint while it compiles 
This commit was SVN r6349.
 2005-07-05 16:29:57 +00:00
Brian Barrett
acce172a87 * after checking with Tim, add MCA_BTL_TAG_MAX constant - avoid having to 
hard code 256 into all the module structs.
* Update template btl to match change

This commit was SVN r6348.
 2005-07-05 14:03:49 +00:00
Jeff Squyres
87b25a09a8 Oops -- forgot to remove outdated file from the Makefile.am 
This commit was SVN r6347.
 2005-07-05 10:23:51 +00:00
Brian Barrett
e4f0e61afa * don't ever unignore the Portals ptl (keeping in the tree only as reference 
for the Portals BTL)

This commit was SVN r6346.
 2005-07-05 03:08:24 +00:00
George Bosilca
8619097919 Update the xgrid components in order to allow them to compile under the new tree. In other words change the include list to match the one explained in the Jeff email. 
This commit was SVN r6345.
 2005-07-04 21:19:35 +00:00
Jeff Squyres
ba99409628 Major simplifications to component versioning: 
- After long discussions and ruminations on how we run components in
  LAM/MPI, made the decision that, by default, all components included
  in Open MPI will use the version number of their parent project
  (i.e., OMPI or ORTE).  They are certaint free to use a different
  number, but this simplification makes the common cases easy:
  - components are only released when the parent project is released
  - it is easy (trivial?) to distinguish which version component goes
    with with version of the parent project
- removed all autogen/configure code for templating the version .h
  file in components
- made all ORTE components use ORTE_*_VERSION for version numbers
- made all OMPI components use OMPI_*_VERSION for version numbers
- removed all VERSION files from components
- configure now displays OPAL, ORTE, and OMPI version numbers
- ditto for ompi_info
- right now, faking it -- OPAL and ORTE and OMPI will always have the
  same version number (i.e., they all come from the same top-level
  VERSION file).  But this paves the way for the Great Configure
  Reorganization, where, among other things, each project will have
  its own version number.

So all in all, we went from a boatload of version numbers to
[effectively] three.  That's pretty good.  :-)

This commit was SVN r6344.
 2005-07-04 20:12:36 +00:00
Brian Barrett
b1f8e229f5 * enable f77_get_fortran_handle_max to be cross compiled 
This commit was SVN r6343.
 2005-07-04 20:00:54 +00:00
Jeff Squyres
6a9c9953bc Remove a bunch of -I's that are no longer necessary with 
properly-prefixed static-component.h files.

This commit was SVN r6342.
 2005-07-04 18:24:58 +00:00
Brian Barrett
f654a65612 * patch to the C alignment check to allow it to cross compile 
This commit was SVN r6341.
 2005-07-04 17:34:24 +00:00
Jeff Squyres
5cdc2ab3f7 Remove some outdated files 
This commit was SVN r6340.
 2005-07-04 17:20:38 +00:00
Brian Barrett
057919616e * fix problem were frameworks wouldn't be found if doing a VPATH build 
This commit was SVN r6339.
 2005-07-04 16:41:45 +00:00
Brian Barrett
b04c726ad1 Fix up tests so that they all compile and (mostly) run 
This commit was SVN r6338.
 2005-07-04 14:53:10 +00:00
Jeff Squyres
05e8cb7e5b Add a few -I's for ROMIO 
This commit was SVN r6337.
 2005-07-04 10:55:17 +00:00