1
1
Граф коммитов

688 Коммитов

Автор SHA1 Сообщение Дата
Nathan Hjelm
1911d97044 coll/ml: fix assertion failure that occurs when level 0 of the hierarchy
fails to select any processes on any nodes.

Also modified basesmsocket to only print debugging info to the framework
output.

cmr=v1.7.5:reviewer=jsquyres

This commit was SVN r31071.
2014-03-14 19:39:00 +00:00
Nathan Hjelm
61f30d992a coll/ml: reduce has some issues when using non-contiguous datatypes. until
these issues are resolved disable coll/ml reduce.

cmr=v1.7.5:reviewer=manjugv

This commit was SVN r31030.
2014-03-12 14:39:16 +00:00
Jeff Squyres
da87b506bd Remove warnings identified by clang 3.4
* Remove unused static functions
 * Remove unused static variables

cmr=v1.8:reviewer=hjelmn

This commit was SVN r31023.
2014-03-12 13:17:54 +00:00
Mike Dubman
a14dda491e OSHMEM: various fixes
- -check-shmem-params is OFF by default. It checks OSHMEM API params and will abort on bad input
- hcoll do not save fallback coll pointers for unsupported collectives.

fixed by Val, Roman, reviewed by Miked/Igor

cmr=v1.7.5:reviewer=ompi-rm1.7

This commit was SVN r30995.
2014-03-11 17:27:33 +00:00
Nathan Hjelm
da2a68f669 coll/ml: fix bcast buffer size calculation
cmr=v1.7.5:reviewer=manjugv

This commit was SVN r30963.
2014-03-07 21:00:08 +00:00
Nathan Hjelm
0af741810c coll/ml: do not access group proc pointers directly. use ompi_comm_peer_lookup instead.
Resolves an issue seen with --enable-sparse-groups.

cmr=v1.7.5:reviewer=manjugv

This commit was SVN r30945.
2014-03-05 22:57:21 +00:00
Ralph Castain
29a7eda280 Remove executable property
This commit was SVN r30791.
2014-02-21 17:27:47 +00:00
Mike Dubman
608269ed72 fca: support relocation of fca packages to opal_prefix/../fca
reviewed by AlexM
cmr=v1.7.5:reviewer=ompi-rm1.7

This commit was SVN r30728.
2014-02-14 14:49:41 +00:00
Pavel Shamis
3a683419c5 Fixing broken dependency between ML/BCOLS
This is hot-fix patch for the issue reported by Ralph. 
In future we plan to restructure ml data structure layout.

Tested by Nathan.

cmr=v1.7.5:ticket=trac:4158

This commit was SVN r30619.

The following Trac tickets were found above:
  Ticket 4158 --> https://svn.open-mpi.org/trac/ompi/ticket/4158
2014-02-07 19:15:45 +00:00
Ralph Castain
74d3393a4f Revert r30600, r30602-30604 as the first one broke the tarball and the others couldn't fix it
This commit was SVN r30605.

The following SVN revision numbers were found above:
  r30600 --> open-mpi/ompi@7d2c4cb468
  r30602 --> open-mpi/ompi@9e751a0302
  r30604 --> open-mpi/ompi@3012c280cf

Revision number ranges (suitable for "git log"):
  r30602-30604 --> open-mpi/ompi@9e751a03^..3012c280
2014-02-07 04:38:06 +00:00
Jeff Squyres
7d2c4cb468 There's a few ml-related bugs outstanding, and Nathan is looking into
them, but it's going to take a little time (at least one day).  So
Nathan says it's ok to .ompi_ignore coll ml until he's able to fix it.

This commit was SVN r30600.
2014-02-06 23:51:03 +00:00
Nathan Hjelm
c2b061cc84 basesmuma: clean up code
Several changes are contained in this commit:

 - Clean up tabs and trailing whitespaces

 - Use consistent indentation in changed files

 - Remove unused code. None of the removed code will ever have been
   used in a trunk build.

 - Clean up the smcm code quite a bit

 - Do not fflush stderr and use opal_output instead of fprintf.

These changes have been tested on Cray XE-6 and PSM systems.

cmr=v1.7.5:ticket=trac:4158

This commit was SVN r30533.

The following Trac tickets were found above:
  Ticket 4158 --> https://svn.open-mpi.org/trac/ompi/ticket/4158
2014-02-03 17:01:46 +00:00
Nathan Hjelm
afae924e29 coll/ml: fix some warnings and the spelling of indices
This commit fixes one warning that should have caused coll/ml to segfault
on reduce. The fix should be correct but we will continue to investigate.

cmr=v1.7.5:ticket=trac:4158

This commit was SVN r30477.

The following Trac tickets were found above:
  Ticket 4158 --> https://svn.open-mpi.org/trac/ompi/ticket/4158
2014-01-29 18:44:21 +00:00
Ralph Castain
b32556e6dc Fixes trac:4143
After IM with Nathan, apply patch from ticket after verification by Paul Hargrove that it fixes the problem on non-x86 32-bit platforms

Verified by Paul, RM-approved

cmr=v1.7.4:reviewer=ompi-gk1.7

This commit was SVN r30411.

The following Trac tickets were found above:
  Ticket 4143 --> https://svn.open-mpi.org/trac/ompi/ticket/4143
2014-01-24 17:56:52 +00:00
Mike Dubman
071838bb0a HCOLL: call hcoll_finalize and hcoll progress unregister in case of hcoll module query failures
fixed by Elena, reviewed by Val/Miked
cmr=v1.7.4:reviewer=ompi-rm1.7

This commit was SVN r30390.
2014-01-23 07:29:23 +00:00
Nathan Hjelm
7ba8bd81fa coll/ml: remove debug fprintfs
cmr=v1.7.5:ticket=trac:4158

This commit was SVN r30367.

The following Trac tickets were found above:
  Ticket 4158 --> https://svn.open-mpi.org/trac/ompi/ticket/4158
2014-01-22 17:21:05 +00:00
Nathan Hjelm
82d996fb76 coll/ml: cleanup some merge related errors
cmr=v1.7.5:ticket=trac:4158

This commit was SVN r30366.

The following Trac tickets were found above:
  Ticket 4158 --> https://svn.open-mpi.org/trac/ompi/ticket/4158
2014-01-22 16:48:09 +00:00
Nathan Hjelm
1a021b8f2d coll/ml: add support for blocking and non-blocking allreduce, reduce, and
allgather.

The new collectives provide a signifigant performance increase over tuned for
small and medium messages. We are initially setting the priority lower than
tuned until this has had some time to soak in the trunk. Please set
coll_ml_priority to 90 for MTT runs.

Credit for this work goes to Manjunath Gorentla Venkata (ORNL), Pavel Shamis (ORNL),
and Nathan Hjelm (LANL).

Commit details (for reference):

Import ORNL's collectives for MPI_Allreduce, MPI_Reduce, and MPI_Allgather.

We need to take the basesmuma header into account when calculating the
ptpcoll small message thresholds. Add a define to bcol.h indicating the
maximum header size so we can take the header into account while not
making ptpcoll dependent on information from basesmuma.

This resolves an issue with allreduce where ptpcoll overwrites the
header of the next buffer in the basesmuma bank.

Fix reduce and make a sequential collective launcher in coll_ml_inlines.h

The root calculation for reduce was wrong for any root != 0. There are
four possibilities for the root:

 - The root is not the current process but is in the current hierarchy. In
   this case the root is the index of the global root as specified in the
   root vector.

 - The root is not the current process and is not in the next level of the
   hierarchy. In this case 0 must be the local root since this process will
   never communicate with the real root.

 - The root is not the current process but will be in next level of the
   hierarchy. In this case the current process must be the root.

 - I am the root. The root is my index.

Tested with IMB which rotates the root on every call to MPI_Reduce. Consider
IMB the reproducer for the issue this commit solves.

Make the bcast algorithm decision an enumerated variable

Resolve various asset failures when destructing coll ml requests.

Two issues:

 - Always reset the request to be invalid before returning it to the
   free list. This will avoid an asset in ompi_request_t's destructor.
   OMPI_REQUEST_FINI does this (and also releases the fortran handle
   index).

 - Never explicitly construct or destruct the superclass of an opal
   object. This screws up the class function tables and will cause
   either an assert failure or a segmentation fault when destructing
   coll ml requests.

Cleanup allgather.

I removed the duplicate non-blocking and blocking functions and modeled
the cleanup after what I found in allreduce. Also cleaned up the code
somewhat.

Don't bother copying from the send to the recieve buffer in
bcol_basesmuma_allreduce_intra_fanin_fanout if the pointers are the
same.

The eliminates a warning about memcpy and aliasing and avoids an
unnecessary call to memcpy.

Alwasy call CHECK_AND_RELEASE on memsync collectives.

There was a call to OBJ_RELEASE on the collective communicator but
because CHECK_AND_RECYLCE was never called there was not matching call
to OBJ_RELEASE. This caused coll ml to leak communicators.

Make allreduce use the sequential collective launcher in coll_ml_inlines.h

Just launch the next collective in the component progress.

I am a little unsure about this patch. There appears to be some sort
of race between collectives that causes buffer exhaustion in some cases
(IMB Allreduce is a reproducer). Changing progress to only launch the
next bcol seems to resolve the issue but might not be the best fix.

Note that I see little-no performance penalty for this change.

Fix allreduce when there are extra sources.

There was an issue with the buffer offset calculation when there are
extra sources. In the case of extra sources == 1 the offset was set
to buffer_size (just past the header of the next buffer). I adjusted
the buffer size to take into accoun the maximum header size (see the
earlier commit that added this) and simplified the offset calculation.

Make reduce/allreduce non-blocking. This is required for MPI_Comm_idup
to work correctly.

This has been tested with various layouts using the ibm testsuite and
imb and appears to have the same performance as the old blocking version.

Fix allgather for non-contiguous layouts and simplify parsing the
topology.

Some things in this patch:

 - There were several comments to the effect that level 0 of the
   hierarchy MUST contain all of the ranks. At least one function
   made this assumption but it was not true. I changed the sbgp
   components and the coll ml initization code to enforce this
   requirement.

 - Ensure that hierarchy level 0 has the ranks in the correct
   scatter gather order. This removes the need for a separate
   sort list and fixes the offset calculation for allgather.

 - There were several passes over the hierarchy to determine
   properties of the hierarchy. I eliminated these extra passes
   and the memory allocation associated with them and calculate the
   tree properties on the fly. The same DFS recursion also handles
   the re-order of level 0.

All these changes have been verified with MPI_Allreduce, MPI_Reduce, and
MPI_Allgather. All functions now pass all IBM/Open MPI, and IMB tests.

coll/ml: correct pointer usage for MPI_BOTTOM

Since contiguous datatypes are copied via memcpy (bypassing the convertor) we
need to adjust for the lb of the datatype. This corrects problems found testing
code that uses MPI_BOTTOM (NULL) as the send pointer.

Add fallback collectives for allreduce and reduce.

cmr=v1.7.5:reviewer=pasha

This commit was SVN r30363.
2014-01-22 15:39:19 +00:00
Mike Dubman
b8550a55a7 HCOLL: many fixes
Adds coll_hcoll_np mca parameter similar to that of fca component (defaults to 32). Those who use hcoll be aware that from now on the communicators less than 32 procs will run w/o hcoll by default. - Resolves fallback issue in case libhcoll runs out of allowed contexts. The solution is moving hcoll_context_create from comm_enable to comm_query. Shortly, comm_enable should never return OMPI_ERROR in the coll component with highest priority (hcoll). Otherwise the ompi coll_base_select will unselect the coll funtion pointers and module references leaving the communicator w/o coll pointer. This will cause the fail. Same behavior can be reproduced even with tuned if one would hardcore some "return OMPI_ERROR" into it's module_enable funtion. - Additionally, removed all the dead code under #if 0; removed unused variables (path for library, active_modules list) and classes (module list wrapper)

Fixed by Val, Reviewed by Devendar/Josh/Miked

cmr=v1.7.4:reviewer=ompi-rm1.7

This commit was SVN r30341.
2014-01-21 12:19:47 +00:00
Ralph Castain
9566650458 Per Marco, don't define a "min" function if one is already defined to avoid conflict with cygwin reserved word
This commit was SVN r30241.
2014-01-10 18:03:25 +00:00
Ralph Castain
c7a94a57d7 Per Marco, rename ERROR tags to exit_ERROR to avoid cygwin reserved name issues.
Refs trac:4085

This commit was SVN r30239.

The following Trac tickets were found above:
  Ticket 4085 --> https://svn.open-mpi.org/trac/ompi/ticket/4085
2014-01-10 18:00:49 +00:00
Mike Dubman
110c99af4f sharing negative tag space between libNBC and HCOLL
fixed by devendar, reviewed by miked
cmr=v1.7.4:reviewer=ompi-rm1.7

This commit was SVN r30224.
2014-01-10 12:51:34 +00:00
Nathan Hjelm
bb01fc2938 Add missing MCA variable enumerator sentinel.
cmr=v1.7.4:reviewer=rhc

This commit was SVN r30178.
2014-01-09 15:28:42 +00:00
Mike Dubman
0fae2caef3 Create a comm keyval for hcoll component with delete callback function.
Set comm attribute with keyval.
Wait for pending hcoll module tasks in comm delete callback where PML
still valid on the communicator. safely destroy hcoll context during
hcoll module destructor.

Author: Devendar Bureddy 
reviewed by miked

cmr=v1.7.4:reviewer=ompi-rm1.7

This commit was SVN r30175.
2014-01-09 11:27:24 +00:00
Mike Dubman
43d6a30693 Fix problems of:
- HCOLL close without init
- Call hcoll progress after comm finalize
- mpirun default for coll_hcoll_enable is 1

fixed by Igor, reviewed by miked
cmr=v1.7.4:reviewer=ompi-rm1.7

This commit was SVN r30156.
2014-01-08 10:55:25 +00:00
Jeff Squyres
13b29cff2c This commit compliements/completes r30140. r30140 made all the
configury/Makefile.am changes; this commit renames the internal
installdirs.h framework struct field names to match the configry macro
names:

 * pkgdatdir ->	ompidatadir
 * pkglibdir -> ompilibdir
 * pkgincludedir -> ompiincludedir

This commit was SVN r30145.

The following SVN revision numbers were found above:
  r30140 --> open-mpi/ompi@8b778903d8
2014-01-07 23:36:33 +00:00
Brian Barrett
8b778903d8 Fix longstanding issue with our multi-project support. Rather than using
pkg{data,lib,includedir}, use our own ompi{data,lib,includedir}, which is
always set to {datadir,libdir,includedir}/openmpi.  This will keep us from
having help files in prefix/share/open-rte when building without Open MPI,
but in prefix/share/openmpi when building with Open MPI.

This commit was SVN r30140.
2014-01-07 22:11:15 +00:00
Brian Barrett
e811a8a9cb Make the Portals 4 collective component disable itself when there's not a
Portals 4 point-to-point (MTL or BTL) component in use

This commit was SVN r30109.
2014-01-02 22:35:37 +00:00
Mike Dubman
80f4e02e0a Several changes:
- Modifications to coll/hcoll component related to the changes in the libhcoll API. 
  Now, hcoll_destroy_context accepts one more parameter that indicates if the context was
  really destroyed as a result of the call. 
  This new "non-blocking" context destruction fixes hang discovered in IMB with mcast enabled. 
- Clean up all the left contexts (if any) on the comm_world destruction. 

fixed by Val, reviewed by miked
cmr=v1.7.4:reviewer=ompi-rm1.7

This commit was SVN r30055.
2013-12-23 06:57:12 +00:00
Jeff Squyres
71ec6c1617 Remove unnecessary "mpi.h"; move opal headers to the top.
This commit was SVN r30053.
2013-12-22 20:38:43 +00:00
Jeff Squyres
0ab48ad0d2 Fix some annoying flex warnings that have been there for years.
Many thanks to Tom Fogal for the initial patch.

cmr=v1.7.4:reviewer=rhc:subject=Fix annoying flex warnings

This commit was SVN r29904.
2013-12-14 00:36:12 +00:00
Mike Dubman
9a65e0d8c6 cosmetic fixed fpr hcol autotools
Refs: #3694

This commit was SVN r29841.
2013-12-08 09:45:13 +00:00
Mike Dubman
2e124454b4 cosmitic fix to remove redundant -lfca
use CPP extra flags var which propagated to coll/fca and scoll/fca
Refs: #3694

This commit was SVN r29832.
2013-12-07 15:00:54 +00:00
Devendar Bureddy
4554770ee4 hcol fixes
cmr=v1.7.4:reviewer=jladd

This commit was SVN r29787.
2013-12-03 20:21:40 +00:00
George Bosilca
cb24277737 Restrict the usage of MPI_Type_extent only to receiving processes
(aka the root). This commit is based on a patch provided by Pierre 
Jolivet.
Fix all the output to match the failing MPI call.

This commit was SVN r29761.
2013-11-27 12:09:31 +00:00
George Bosilca
68268377af Fix an error message for the igather and the usage of the extent on
non non-root processes for the iscatter. Thanks to Pierre
Jolivet for the bug report and the patch.

This commit was SVN r29736.
2013-11-23 00:59:22 +00:00
Nathan Hjelm
24a7e7aa34 Add support for the udreg registration cache and dynamics on XE/XK/XC.
To support the new mpool two changes were made to the mpool infrastructure:

 1) Added an mpool flag to indicate that an mpool does not need the memory
    hooks to use the leave pinned protocols. This flag is checked in the
    mpool lookup.

 2) Add a mpool context to the base registration. This new member is used
    by the udreg mpool to store the udreg context associated with the
    particular registration. The new member will not break the ABI
    compatibility as the new member is only currently used by the udreg
    mpool.

Dynamics support for Cray systems makes use of the global rank provided by
orte to give the ugni library a unique rank for each process. Dynamics
support is not available under direct-launch (srun.)

cmr=v1.7.4

This commit was SVN r29719.
2013-11-18 04:58:37 +00:00
Brian Barrett
cf8de1ef0f Minor indent cleanup in init_query()
Only use Portals on communicators with more than one rank
Fix computation of number of children when using the hypercube tree

This commit was SVN r29616.
2013-11-06 15:21:09 +00:00
Nathan Hjelm
c71125acfd Using MPI_* functions in iallreduce can cause comm-spawned processes to
crash. Update libnbc's iallreduce function to use ompi_* functions
instead.

cmr=v1.7.4:reviewer=brbarret

This commit was SVN r29582.
2013-11-01 16:45:54 +00:00
Nathan Hjelm
a31e617d17 Remove outdated comments in coll_basic_reduce_scatter.c.
Refs trac:1559

This commit was SVN r29566.

The following Trac tickets were found above:
  Ticket 1559 --> https://svn.open-mpi.org/trac/ompi/ticket/1559
2013-10-30 16:20:20 +00:00
Nathan Hjelm
167d5613db Do not do arithmetic with void * in basic neighborhood alltoall[vw].
cmr=v1.7.4:reviewer=jsquyres

This commit was SVN r29558.
2013-10-29 20:02:13 +00:00
Nathan Hjelm
b202bb0d63 Fix the recursive halfing algorithms for reduce scatter in both basic
and tuned to correctly handle 0 recvcounts.

Tested with the reproducer from #1550.

Refs trac:1559

This commit was SVN r29542.

The following Trac tickets were found above:
  Ticket 1559 --> https://svn.open-mpi.org/trac/ompi/ticket/1559
2013-10-28 19:06:38 +00:00
Mike Dubman
d27cffedb9 expand tabs to 4 spaces
cd ompi/mca/coll/fca
for i in *.[ch]; do expand -t 4 $i > koko && mv koko $i; done
Refs: #3799

This commit was SVN r29472.
2013-10-22 17:05:55 +00:00
Jeff Squyres
6714890244 paffinity.h is gone and won't be coming back.
This commit was SVN r29467.
2013-10-22 15:59:00 +00:00
Mike Dubman
5a7dff2d15 fix icc warning
fixed by Dinar, reviewed by miked
cmr=v1.7.4:reviewer=ompi-gk1.7

This commit was SVN r29428.
2013-10-12 18:04:28 +00:00
Nathan Hjelm
6232ef3bfb At coll_select time we can not check whether the communicator has a
virtual topology. Remove code checking for a virtual topology until
this flag is set before coll_select.

This commit was SVN r29344.
2013-10-03 03:37:46 +00:00
Nathan Hjelm
7bedf62dd8 Add basic algorithms for the remaining non-blocking collectives.
The algorithms are intended for MPI-3.0 compliance and are not
optimized. We should aim to add better algorithms in the future through
cheetah.

MPI_Iallreduce and MPI_Igatherv on intercommunicators are required for
MPI_Comm_idup support.

cmr=v1.7.4:reviewer=brbarret:ticket=trac:2715

This commit was SVN r29333.

The following Trac tickets were found above:
  Ticket 2715 --> https://svn.open-mpi.org/trac/ompi/ticket/2715
2013-10-02 14:26:23 +00:00
Mike Dubman
19748e6957 fix race condition which can happen on finalize
1. Change in rte api implementation: now comm_world used to do p2p. 
This allows to not worry about other comms being destroyed.

2. added a notification mechanism with a help of which runtime can say libhcoll that RTE api can not be used any longer. 
pass a pointer to a flag, and its size to libhcoll.
The flag changes when the RTE is no longer available. 
Currently this flag is just ompi_mpi_finalized global bool value.

cmr=v1.7.3:reviewer=jladd

This commit was SVN r29331.
2013-10-02 13:38:47 +00:00
Nathan Hjelm
4f12406436 Don't check for neighborhood collective routines on non-virtual topology communicators
This commit was SVN r29319.
2013-10-01 19:59:18 +00:00
Mike Dubman
9bf7578ff2 fix memory corruption
cmr:v1.7.3:reviewer=ompi-rm1.7

This commit was SVN r29293.
2013-09-30 06:18:12 +00:00