1
1
Граф коммитов

16 Коммитов

Автор SHA1 Сообщение Дата
David Eberius
d377a6b6f4 Added Software-based Performance Counters driver code along with several counters.
This code is the implementation of Software-base Performance Counters as described in the paper 'Using Software-Base Performance Counters to Expose Low-Level Open MPI Performance Information' in EuroMPI/USA '17 (http://icl.cs.utk.edu/news_pub/submissions/software-performance-counters.pdf).  More practical usage information can be found here: https://github.com/davideberius/ompi/wiki/How-to-Use-Software-Based-Performance-Counters-(SPCs)-in-Open-MPI.

All software events functions are put in macros that become no-ops when SOFTWARE_EVENTS_ENABLE is not defined.  The internal timer units have been changed to cycles to avoid division operations which was a large source of overhead as discussed in the paper.  Added a --with-spc configure option to enable SPCs in the Open MPI build.  This defines SOFTWARE_EVENTS_ENABLE.  Added an MCA parameter, mpi_spc_enable, for turning on specific counters.  Added an MCA parameter, mpi_spc_dump_enabled, for turning on and off dumping SPC counters in MPI_Finalize.  Added an SPC test and example.

Signed-off-by: David Eberius <deberius@vols.utk.edu>
2018-06-11 22:48:16 -04:00
George Bosilca
366d64b7e5 Move the collective structure outside the communicator.
As we changed the ABI (forcing a major release), we can limit
the size of the predefined communicators by moving the collective
structure outside the communicator. This might have a minimal,
but unnoticeable, impact on performance. This approach has been
discussed during the January 2017 devel meeting.

Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
Signed-off-by: Joshua Hursey <jhursey@us.ibm.com>
2017-02-27 11:54:17 -06:00
Ralph Castain
1e2019ce2a Revert "Update to sync with OMPI master and cleanup to build"
This reverts commit cb55c88a8b.
2016-11-22 15:03:20 -08:00
Ralph Castain
cb55c88a8b Update to sync with OMPI master and cleanup to build
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
2016-11-22 14:24:54 -08:00
Gilles Gouaillardet
7cae36f5ab ompi: accept MPI_IN_PLACE in MPI_Ialltoall* 2016-10-08 19:47:25 +09:00
George Bosilca
4ac247b1da Minor updated on the validity checks for the alltoall collectives. 2015-10-24 15:25:28 -04:00
Gilles Gouaillardet
256976a108 mpi: MPI_IN_PLACE is not a valid argument of MPI_All* and MPI_Iall* with an inter communicator 2015-10-21 14:46:28 +09:00
Gilles Gouaillardet
2bd77ed4f9 mpi: fail with MPI_ERR_INTERN if MPI_IN_PLACE is used with MPI_I*alltoall*
currently, MPI fails with MPI_ERR_ARG. This is counter intuitive since
MPI_IN_PLACE is a legitimate parameter. MPI_IN_PLACE might not be correctly
implemented by all the non blocking modules (libnbc, ...) so fail with
MPI_ERR_INTERN for the time being.
2015-10-20 14:12:33 +09:00
Gilles Gouaillardet
53b952dc2b oshmem: invoke the C PMPI_* subroutines instead of the MPI_* ones
when profiling is built.
This prevents oshmem subroutines from being wrapped twice by third
party tools (e.g. once in oshmem and once in MPI)
see discussion starting at http://www.open-mpi.org/community/lists/devel/2015/08/17842.php

Thanks to Bert Wesarg for bringing this to our attention
2015-10-13 08:52:03 +09:00
Gilles Gouaillardet
16d65a2762 fortran/mpif-h: invoke the C PMPI_* subroutines instead of the MPI_* ones
when profiling is built.
This prevents Fortran subroutines from being wrapped twice by third
party tools (e.g. once in Fortran and once in C)
see discussion starting at http://www.open-mpi.org/community/lists/devel/2015/08/17842.php
2015-10-13 08:52:02 +09:00
Gilles Gouaillardet
e01bac962f coll: do not cast way the const modifier when this is not necessary
update the coll framework and mpi c bindings
2015-09-09 09:18:57 +09:00
Gilles Gouaillardet
1a17a2a960 Fixes MEMCHECKER vs MPI_IN_PLACE in *alltoall*
cmr=v1.8.2:reviewer=bosilca

This commit was SVN r31924.
2014-06-02 04:43:48 +00:00
Gilles Gouaillardet
8bafe06c57 Fixes *alltoall* collectives at top level
This commit :
 - Correctly retrieve the communicator size when
   checking memory and parameters
 - Ensure (sendtype,sendcount) and (recvtype,recvcount)
   matches and return with MPI_ERR_TRUNCATE otherwise
 - Return with MPI_SUCCESS without invoking the low level
   if no data is going to be transferred
 - Fixes trac:4506

cmr=v1.8.2:reviewer=bosilca

This commit was SVN r31815.

The following Trac tickets were found above:
  Ticket 4506 --> https://svn.open-mpi.org/trac/ompi/ticket/4506
2014-05-19 07:46:07 +00:00
Nathan Hjelm
0b8fc13299 MPI-3.0: update C bindings with const and consistent use of [] for
arrays.

The MPI 3.0 standard added const to all in buffers in the C bindings. This
commit adds the const keyword and in most cases casts const away. We will
eventually should go through and update the various interfaces (coll, pml,
io, etc) to take the const keyword. The group, comm, win, and datatype
interfaces have been updated with const.

cmr=v1.7.4:ticket=trac:3785:reviewer=jsquyres

This commit was SVN r29266.

The following Trac tickets were found above:
  Ticket 3785 --> https://svn.open-mpi.org/trac/ompi/ticket/3785
2013-09-26 21:56:20 +00:00
Brian Barrett
eb3c6546c2 Remove shortcut optimizations from non-blocking collectives. We need to call
into the collectives module even in the shortcutable cases to get the requests
created.

This commit was SVN r26784.
2012-07-11 16:03:55 +00:00
Brian Barrett
b9e8e4aeb9 * Initial merge of the non-blocking collectives interface. No implementation of
the back-end yet, coming real soon now, need to solve some tag issues first.

This commit was SVN r26641.
2012-06-22 20:54:12 +00:00