1
1
Граф коммитов

20127 Коммитов

Автор SHA1 Сообщение Дата
Ralph Castain
ccd33a17b8 Since we cannot block when calling abort, and we want to ensure any "show_help" message at least has a chance to get out before we exit, introduce a slight delay into the abort procedure.
Refs trac:4576

This commit was SVN r31601.

The following Trac tickets were found above:
  Ticket 4576 --> https://svn.open-mpi.org/trac/ompi/ticket/4576
2014-05-02 10:46:25 +00:00
Ralph Castain
c1383ca1f3 Protect against NULL cpuset when not bound
This commit was SVN r31600.
2014-05-02 10:45:11 +00:00
Ralph Castain
0209cddb5b Revert r31596 and r31595 as they recreate the "abort" problem - all they did was move the blocking send to another point in the code. An alternative solution to the "show_help and abort" problem. will come in another commit
Refs trac:4576

This commit was SVN r31599.

The following SVN revision numbers were found above:
  r31595 --> open-mpi/ompi@2b61f22973
  r31596 --> open-mpi/ompi@712634efd3

The following Trac tickets were found above:
  Ticket 4576 --> https://svn.open-mpi.org/trac/ompi/ticket/4576
2014-05-02 10:38:30 +00:00
Ralph Castain
6545e6e9a8 Add one more check for failed mapping that rarely occurs, but results in a hang when it does
cmr=v1.8.2:reviewer=rhc

This commit was SVN r31598.
2014-05-02 10:35:14 +00:00
MPI Team
268c454bb2 Update git/hg ignore files
This commit was SVN r31597.
2014-05-02 05:00:31 +00:00
Ralph Castain
712634efd3 Silence warning
Refs trac:4576

This commit was SVN r31596.

The following Trac tickets were found above:
  Ticket 4576 --> https://svn.open-mpi.org/trac/ompi/ticket/4576
2014-05-01 23:58:03 +00:00
Ralph Castain
2b61f22973 Now that the abort code no longer involves a blocking rml send section, apps that call show_help followed by abort are not printing their error message. So block them in show_help until that message gets out.
This commit was SVN r31595.
2014-05-01 22:57:17 +00:00
Jeff Squyres
bf82ee2a14 btl_usnic_connectivity.h: fix PACK_BYTES macro
We're passing a char foo[x] into PACK_BYTES, so we don't need to take
its address in the macro.  This is parallel to the UNPACK_BYTES macro
(where we pass a char bar[x] into it, and don't take its address in
the macro).

The value we're packing is only used to output in a show_help message,
which is why this wasn't noticed before (i.e., it's not used in
network or addressing that would have caused a failure).

cmr=v1.8.2:reviewer=dgoodell

This commit was SVN r31594.
2014-05-01 22:23:22 +00:00
Yossi Etigin
6aa5680059 Revert r30966.
cmr=v1.8.1:reviewer=ompi-gk1.8

This commit was SVN r31593.

The following SVN revision numbers were found above:
  r30966 --> open-mpi/ompi@280e96c99a
2014-05-01 22:17:09 +00:00
Ralph Castain
0c74d1fd6f Silence warning
This commit was SVN r31592.
2014-05-01 21:11:39 +00:00
Jeff Squyres
49383f0aaa Oops: remove errant "v4" string.
This commit was SVN r31591.
2014-05-01 20:21:43 +00:00
Jeff Squyres
56ecb92b10 Per discussion with George and Ralph, change this BTL_ERROR message to
an opal_show_help() so that its output is deduplicated.

This commit was SVN r31590.
2014-05-01 20:15:33 +00:00
Ralph Castain
445b552d3a Try again to get an error message printed when a daemon fails to successfully report back to mpirun. In this case, there is no guaranteed way for the daemon to output the error report itself - we don't have a connection back to the HNP, and we have tied stderr off to /dev/null (for good reasons). So the HNP has to detect the failure itself and report it.
The HNP can't know the precise reason, of course - all it knows is that the daemon failed. So output a generic error message that provides guidance on probable causes.

Refs trac:4571

This commit was SVN r31589.

The following Trac tickets were found above:
  Ticket 4571 --> https://svn.open-mpi.org/trac/ompi/ticket/4571
2014-05-01 19:48:21 +00:00
Jeff Squyres
0fac9781b3 Assume we always have fortran PROCEDURE support
Per #4590, we now ''require'' the PROCEDURE keyword support in Fortran
for the mpi_f08 module.  So if the Fortran compiler doesn't support
it, then we won't build the mpi_f08 module.

Fixes trac:4590

This commit was SVN r31588.

The following Trac tickets were found above:
  Ticket 4590 --> https://svn.open-mpi.org/trac/ompi/ticket/4590
2014-05-01 18:18:38 +00:00
Jeff Squyres
08cf16e14f ompi_setup_mpi_fortran.m4: disable mpi_f08 if PROCEDURE isn't supported
Per https://svn.open-mpi.org/trac/ompi/ticket/4590, if the Fortran
compiler doesn't support PROCEDURE in the way we want/need, disable
the mpi_f08 module.

Refs trac:4590

This commit was SVN r31587.

The following Trac tickets were found above:
  Ticket 4590 --> https://svn.open-mpi.org/trac/ompi/ticket/4590
2014-05-01 18:17:52 +00:00
Ralph Castain
567ed25938 As per the earlier RFC, move the DB framework to orcm, thus removing it from the OMPI code repo
This commit was SVN r31586.
2014-05-01 15:43:32 +00:00
Ralph Castain
e20dae536c Last step under current RFC: OMPI_CHECK_WITHDIR -> OPAL_CHECK_WITHDIR
This commit was SVN r31585.
2014-05-01 15:38:07 +00:00
Ralph Castain
2749231eb5 Next step in RFC: OMPI_CHECK_COMPILER_WORKS -> OPAL_CHECK_COMPILER_WORKS
This commit was SVN r31584.
2014-05-01 15:25:47 +00:00
Ralph Castain
e11eb15518 Next step of RFC: OMPI_CHECK_FUNC_LIB -> OPAL_CHECK_FUNC_LIB
This commit was SVN r31583.
2014-05-01 14:57:43 +00:00
Ralph Castain
3b64c603b4 First stage of RFC to rename OMPI_foo build system support: change OMPI_CHECK_PACKAGE -> OPAL_CHECK_PACKAGE
This commit was SVN r31582.
2014-05-01 14:24:56 +00:00
MPI Team
71f2407929 Update git/hg ignore files
This commit was SVN r31581.
2014-05-01 05:00:32 +00:00
Jeff Squyres
c4d85ec6ca btl_usnic_cclient.c: update to use the new opal dstore
Use the new opal dstore API (vs. the old RTE DB API).

(dstore is not going to the v1.8 series, so there's no need to CMR
this to v1.8)

This commit was SVN r31580.
2014-04-30 22:32:47 +00:00
Nathan Hjelm
e963869fdf bcol/basesmuma: close mmapped file descriptor
Not closing this file descriptor will cause us to leak file
descriptors. It is safe to close the file after it has been mmapped.

cmr=v1.8.2:reviewer=manjugv

This commit was SVN r31579.
2014-04-30 22:28:08 +00:00
Ralph Castain
238ecea311 When we comm_spawn, we really want to respect the original -host directives and not expand the daemon virtual machine unless directed to do so in the comm_spawn command. Otherwise, we will automatically launch daemons on every node in the allocation.
cmr=v1.8.2:reviewer=rhc:subject=respect vm boundaries during comm_spawn

This commit was SVN r31578.
2014-04-30 22:26:18 +00:00
Jeff Squyres
d40112a012 rte_base_frame.c: add sanity check to ensure proper sizes
There's a requirement in several places (e.g., opal dstore) that
sizeof(ompi_process_name_t) -- which comes from the compile-time
selected ompi/mca/rte component -- is equal to sizeof(uint64_t).  If
it's not, Bad Things will happen.

So put an assert here to catch that case.

This commit was SVN r31577.
2014-04-30 22:12:54 +00:00
Nathan Hjelm
a28012b29d Fix MPI_T issues identified by friendly users.
Several fixes:

 - I was allowing an MPI_T_cvar_handle to be created for an invalid
   variable. Fixed this by checking if the variable is valid in
   mca_base_var_get.

 - Use a better error code when the caller tries to create an unbound
   pvar handle for a bound variable.

 - Return the verbosity level in MPI_T_cvar_get_info.

cmr=v1.8.2:reviewer=jsquyres

This commit was SVN r31576.
2014-04-30 22:10:30 +00:00
Nathan Hjelm
d80f14eb0f sbgp/ptp: fix obvious typo
cmr=v1.8.2:reviewer=manjugv

This commit was SVN r31575.
2014-04-30 22:10:22 +00:00
Nathan Hjelm
3e5388eaa6 mtl/psm: do not limit PSM to 8191 context ids
The old default context id maximum was committed to the trunk in
2006. After some discussion with Intel it appears this is restricting
the mtl to an arbirarly small number of communicators. Increasing the
default to allow up to 2^16 - 1 context ids.

Refs trac:4574

cmr=v1.8.2

This commit was SVN r31574.

The following Trac tickets were found above:
  Ticket 4574 --> https://svn.open-mpi.org/trac/ompi/ticket/4574
2014-04-30 22:10:15 +00:00
Ralph Castain
d04a102ab8 Silence warnings
This commit was SVN r31573.
2014-04-30 20:55:46 +00:00
Ralph Castain
087b84b0ef Add some further debug to the dstore framework. When doing comm_spawn, we have to exchange any provided cpu bitmaps to ensure both sides compute the same locality, else various mpi frameworks can go bonkers.
This commit was SVN r31572.
2014-04-30 19:29:00 +00:00
Ralph Castain
a286977eff Don't build slurm if not present - platform file change only
This commit was SVN r31571.
2014-04-30 19:27:43 +00:00
Ralph Castain
8cda1b3dc6 Don't store cpu_bitmap unless it is non-NULL
This commit was SVN r31570.
2014-04-30 18:12:48 +00:00
Ralph Castain
7a79b25577 Ensure we cleanup some files so session dirs can be rolled up
cmr=v1.8.2:reviewer=jsquyres

This commit was SVN r31569.
2014-04-30 17:52:10 +00:00
Ralph Castain
751b4325b3 Provide more general code coverage
This commit was SVN r31568.
2014-04-30 15:42:16 +00:00
Ralph Castain
e72af03e60 Fix typo covered by enable-heterogeneous
This commit was SVN r31567.
2014-04-30 15:41:58 +00:00
Ralph Castain
314676363e Set the remaining required ignores
Refs trac:4580

This commit was SVN r31566.

The following Trac tickets were found above:
  Ticket 4580 --> https://svn.open-mpi.org/trac/ompi/ticket/4580
2014-04-30 15:30:22 +00:00
Mike Dubman
e369e5f49b OSHMEM: fix compiler warn
fixed by Roman, reviewed by Miked

cmr=v1.8.2:reviewer=ompi-rm1.8

This commit was SVN r31565.
2014-04-30 12:05:03 +00:00
Mike Dubman
ff42999037 OSHMEM: Added missing API for Java bindings (int16/32/64 stuff)
fixed by Roman, reviewed by Mike

cmr=v1.8.2:reviewer=ompi-rm1.8

This commit was SVN r31564.
2014-04-30 12:03:23 +00:00
MPI Team
4369206f2c Update git/hg ignore files
This commit was SVN r31563.
2014-04-30 05:00:33 +00:00
Mike Dubman
9afb65c876 fixed svn ignore
cmr=v1.8.2:reviewer=ompi-rm1.8

This commit was SVN r31562.
2014-04-30 03:40:42 +00:00
Ralph Castain
34988ba2a2 Cleanup the MPI_Abort detection
Refs trac:4576

This commit was SVN r31561.

The following Trac tickets were found above:
  Ticket 4576 --> https://svn.open-mpi.org/trac/ompi/ticket/4576
2014-04-30 00:51:59 +00:00
Ralph Castain
3c9d877c1b Remove debug
This commit was SVN r31560.
2014-04-30 00:08:43 +00:00
Ralph Castain
9402380e1f Fix some errors in transition
This commit was SVN r31559.
2014-04-30 00:07:53 +00:00
Ralph Castain
b3f636d169 Correct typo in header definition
This commit was SVN r31558.
2014-04-29 23:41:49 +00:00
Ralph Castain
c4c9bc1573 As per the RFC:
http://www.open-mpi.org/community/lists/devel/2014/04/14496.php

Revamp the opal database framework, including renaming it to "dstore" to reflect that it isn't a "database". Move the "db" framework to ORTE for now, soon to move to ORCM

This commit was SVN r31557.
2014-04-29 21:49:23 +00:00
Rolf vandeVaart
fc0a75da91 Fix help message errors as reported by check-help-strings.pl script.
cmr=v1.8.2:reviewer=jsquyres

This commit was SVN r31555.
2014-04-29 20:29:18 +00:00
Jeff Squyres
ca77a1a37d A few minor updates and two bug fixes:
* Ensure to cd to the basename of the URL (e.g., "v1.8" from
   "branches/v1.8")
 * Remove the erroneous -m from the svn commit line that was
   preventing the updated .gitignore file from getting committed

This commit was SVN r31548.
2014-04-29 18:25:46 +00:00
MPI Team
c72e4482b2 Update git/hg ignore files
This commit was SVN r31547.
2014-04-29 18:13:29 +00:00
Jeff Squyres
964249e8a6 comm_cid.c: Ensure that "flag" is initially set to false.
If the loops never get executed because CIDs are exhausted, then the
value of flag will be undefined.

Refs trac:4572

This commit was SVN r31546.

The following Trac tickets were found above:
  Ticket 4572 --> https://svn.open-mpi.org/trac/ompi/ticket/4572
2014-04-29 17:39:14 +00:00
Ralph Castain
1f0efe62a4 Minor cleanup - remove unused RML tag
Refs trac:4576

This commit was SVN r31545.

The following Trac tickets were found above:
  Ticket 4576 --> https://svn.open-mpi.org/trac/ompi/ticket/4576
2014-04-29 17:34:17 +00:00