1
1
Граф коммитов

4495 Коммитов

Автор SHA1 Сообщение Дата
Brian Barrett
0c6eaaebe3 * start cleaning up debugging output (still much to do)
* make buffers really big so that we pass allocmem until we figure out
  why we're not flow controlling as I expected
* set event queue to invalid intially and use that as the enabled test
  rather than a seperate bool - shrinks the module a bit
* add dropped count checks, with a panic if one occurs.  Still need to
  implement some type of retransmit logic.

This commit was SVN r5704.
2005-05-12 21:28:48 +00:00
Brian Barrett
e2c2c72b84 Changes to pass allocmem IBM test
- don't free the send buffer unless the converter tells us we need to
  - properly do the math to determine when the receive buffer has been
    fully used and unlinked itself

This commit was SVN r5703.
2005-05-12 19:52:51 +00:00
Jeff Squyres
f5657fb8ee For the rsh pls, if the launch is on the local node, just exec it --
don't bother using the launching agent (typically rsh or ssh).

This commit was SVN r5702.
2005-05-12 19:12:53 +00:00
Jeff Squyres
544f9dd780 Fix silly string error (missing +2 in the len calculation, so just
replace it with asprintf).  Reviewed by Brian.

This commit was SVN r5700.
2005-05-12 18:56:05 +00:00
Jeff Squyres
f96d763aa7 /trunk is working towards 1.0
/branches/v0.9 is working towards 0.9

This commit was SVN r5699.
2005-05-12 17:56:42 +00:00
Brian Barrett
189a536685 * Fix incorrect logic in orted so that --no-daemonize works as intended
* Minor formatting fixes in XGrid RAS component
* Code cleanup in XGrid PLS component:
  - If we can't get daemon contact information, kill the job at the XGrid
    level
  - Add MCA parameter pls_xgrid_delete_job that will delete the job from
    XGrid when complete (this seems like standard behavior, so it's the
    default)
  - Remove compiler warning about getting the name of a XGGrid object
  - Properly populate the daemon information for the killing code

This commit was SVN r5697.
2005-05-12 16:48:41 +00:00
Josh Hursey
4b60235383 remove unnecessary exclusion for Windows which was killing the Windows nightly build
This commit was SVN r5695.
2005-05-12 14:37:40 +00:00
Brian Barrett
decc74d15c * Enable the XGrid components. Only do anything if the XGrid contact info
variables are set.
* show the RAS priority in ompi_info

This commit was SVN r5694.
2005-05-12 03:33:59 +00:00
Jeff Squyres
722ee2103b Fix to the fix -- Brian and I agree that this is a better fix.
This commit was SVN r5693.
2005-05-12 02:44:20 +00:00
Jeff Squyres
3bd7e72608 Fix some minor typos
This commit was SVN r5692.
2005-05-12 02:22:03 +00:00
Jeff Squyres
a963aebfdb Fix the check for socklen_t.
This commit was SVN r5691.
2005-05-12 01:42:24 +00:00
George Bosilca
4ef1d70034 snprintf does not really do what we expect. In some situations it will write
more than we have asked for (on my G5). Anyway now I hope I have enought memory to printout
the full description of the datatype.

This commit was SVN r5690.
2005-05-11 21:30:56 +00:00
Brian Barrett
c477907166 * ignore UNLINK messages earlier in the chian (if Portals supports them)
* process long message fragments properly

This commit was SVN r5689.
2005-05-11 20:22:18 +00:00
Josh Hursey
cc6cb5cac5 Checkpoint on Windows build.
Many changes to headers for OMPI_DECLSPEC, and 
proper placement of c_plusplus defines in those files.

mca/gpr/replica and tools are the two sets of directories
that still need work for the Windows build for this pass.

This commit was SVN r5688.
2005-05-11 20:21:10 +00:00
George Bosilca
6714dfac4e Remove all useless checks that size_t is greater or equal to zero.
This commit was SVN r5687.
2005-05-11 14:19:48 +00:00
George Bosilca
e940ab43b8 Optionally disable the tests.
This commit was SVN r5686.
2005-05-11 14:17:01 +00:00
Thara Angskun
55538e100d - another check point
- able to launch job (sort of)...but not correctly clean up, mpirun hang, blowup terminal etc.

This commit was SVN r5685.
2005-05-11 09:09:55 +00:00
George Bosilca
1a9cef70fb checkpoint
This commit was SVN r5684.
2005-05-11 06:30:10 +00:00
George Bosilca
9bd4110bb5 If we check for errors then let's check all of them.
This commit was SVN r5683.
2005-05-11 04:26:14 +00:00
George Bosilca
f9ae5a282e We are here in a macro. The arguments of the macro should be protected. Otherwise the compiler
will get confused with the precedence of the operators.

This commit was SVN r5682.
2005-05-11 04:23:47 +00:00
George Bosilca
f0adb8b4fd Adapt to the new PML interface.
This commit was SVN r5681.
2005-05-11 03:59:57 +00:00
Jeff Squyres
2dbcf1a1e5 Fix issue raised by Rainer: if we don't find a corresponding C type
for an optional fortran type, it's not an error.  Instead, just
disable support for that fortran optional type.

This commit was SVN r5680.
2005-05-10 23:57:37 +00:00
Jeff Squyres
bc6f5a83c4 Fix a few more header installation directories
This commit was SVN r5679.
2005-05-10 23:56:23 +00:00
Jeff Squyres
6bda1ed699 Fix installation directory for the header files
This commit was SVN r5678.
2005-05-10 23:52:35 +00:00
Jeff Squyres
2b2f2f3c04 Fix a bunch of compiler warnings, mostly on 64 bit:
- some union { void*; int; } fixes for asm tests
- size_t / %lu fixes for a bunch of others

This commit was SVN r5677.
2005-05-10 23:28:31 +00:00
Jeff Squyres
6a85d9bf74 No need for this anymore.
This commit was SVN r5676.
2005-05-10 20:55:09 +00:00
Brian Barrett
caf8551001 * checkpoint - long messages are causing segfaults in the PML, but need to
stop for a bit

This commit was SVN r5675.
2005-05-10 20:42:57 +00:00
Brian Barrett
2ec27c0927 * add ability to respond to RNDV packets with ACKs. short MPI_Ssends now
work properly.  Still need to implement second fragment support

This commit was SVN r5674.
2005-05-10 19:39:21 +00:00
George Bosilca
956782670a One more printf with size_t solved.
This commit was SVN r5673.
2005-05-10 17:25:56 +00:00
Jeff Squyres
f8b1e19076 - Add a few help messages
- app->num_procs changed to a size_t, which hosed the initialization
  of its value to -1 (not sure why the compiler didn't complain
  #$%@#$%), which was there to catch the case when the user forgot to
  specify -np (or some other equivalent).  Fixed.

This commit was SVN r5672.
2005-05-10 17:14:53 +00:00
Josh Hursey
5d1e2c53b0 fix some library path issues
This commit was SVN r5671.
2005-05-10 16:16:36 +00:00
Brian Barrett
0cd4d15824 * Update to match Tim's changes to the PML
* Couple of improvements towards handling ACKs properly

This commit was SVN r5670.
2005-05-10 15:53:41 +00:00
Brian Barrett
eeba1b9a72 * re-enable making DSOs of the TEG PML
This commit was SVN r5669.
2005-05-10 14:56:45 +00:00
Jeff Squyres
6d8ad2b1bd When the argv add functions were changed to nosize, we forgot to
update num_env.  Doh!

This commit was SVN r5668.
2005-05-10 14:50:55 +00:00
Jeff Squyres
9068ae98dd Arrgh! Turns out that the "-a" operator of "test" does not obey the
"don't evaluate later expessions if an early one evaluates to false".
So change it back to && and put a big comment *why*. @#%@#$%...

This commit was SVN r5667.
2005-05-10 14:35:33 +00:00
Jeff Squyres
f6a00d4cd0 Brian pointed out that not all all platforms have egrep -- use "grep
-E" instead.

This commit was SVN r5666.
2005-05-10 14:02:28 +00:00
Jeff Squyres
5d1685edbc Add unignore for jsquyres@mauve.lanl.gov
This commit was SVN r5665.
2005-05-10 12:02:36 +00:00
Jeff Squyres
5729796e93 Slightly change and ammend the .ompi_unignore test in autogen.sh.
Components are now unignored if:

- .ompi_unignore exists, and
- .ompi_unignore contains at least one of the following:
  - regexp matching "$USER$"
  - regexp matching "$USER@$HOST"

This allows selective unignoring of components for only specific
development hosts rather that everywhere a given username is used.

This commit was SVN r5664.
2005-05-10 12:02:03 +00:00
Jeff Squyres
98f00ce592 Add missing .h file
This commit was SVN r5663.
2005-05-10 11:45:42 +00:00
Jeff Squyres
f374559d14 Add a proper test for socklen_t
This commit was SVN r5662.
2005-05-10 11:25:22 +00:00
Rainer Keller
e06f0723a8 configure.ac:
- Check for icc-8.1 segfaulting with varargs.
  - Add check for size of REAL*4,8,16 types and purge,
    if smaller
  - Fix configure-output with regard to checking for c-types 
    for the REAL*-types.

This commit was SVN r5661.
2005-05-10 09:09:04 +00:00
Tim Woodall
6a565b3fb7 rollback
This commit was SVN r5660.
2005-05-09 21:07:53 +00:00
Tim Woodall
d54c6fafe0 cleanup for data structure reorg
This commit was SVN r5659.
2005-05-09 21:05:54 +00:00
George Bosilca
876dfc4cfc Dont add the fragment to the endpoint before we have erceived some data.
This commit was SVN r5658.
2005-05-09 20:20:29 +00:00
Tim Woodall
6ab861945f corrections for data structure reorg
This commit was SVN r5657.
2005-05-09 20:03:58 +00:00
Tim Woodall
10c8338a04 moved up
This commit was SVN r5656.
2005-05-09 19:42:32 +00:00
Tim Woodall
3d3151abb3 reorg to support new pml - move ptl/teg specific data
from pml_base_* data structures
*** NOTE ***
this will break gm/ib - I will fix these shortly

This commit was SVN r5655.
2005-05-09 19:37:10 +00:00
Jeff Squyres
ac60eac8f3 Temporary information at ftp.gnu.org advises getting config.* scripts
directly from Savannah.

This commit was SVN r5654.
2005-05-09 18:29:03 +00:00
Josh Hursey
276cf4dbd5 donno how that one creeped in. rollback this file
This commit was SVN r5653.
2005-05-09 18:03:29 +00:00
Josh Hursey
f3c537598b Export the rest of the datatypes.
This commit was SVN r5652.
2005-05-09 17:47:31 +00:00