1
1
openmpi/ompi/mca/op/example/op_example.h

106 строки
3.5 KiB
C
Исходник Обычный вид История

Two major things in this commit: * New "op" MPI layer framework * Addition of the MPI_REDUCE_LOCAL proposed function (for MPI-2.2) = Op framework = Add new "op" framework in the ompi layer. This framework replaces the hard-coded MPI_Op back-end functions for (MPI_Op, MPI_Datatype) tuples for pre-defined MPI_Ops, allowing components and modules to provide the back-end functions. The intent is that components can be written to take advantage of hardware acceleration (GPU, FPGA, specialized CPU instructions, etc.). Similar to other frameworks, components are intended to be able to discover at run-time if they can be used, and if so, elect themselves to be selected (or disqualify themselves from selection if they cannot run). If specialized hardware is not available, there is a default set of functions that will automatically be used. This framework is ''not'' used for user-defined MPI_Ops. The new op framework is similar to the existing coll framework, in that the final set of function pointers that are used on any given intrinsic MPI_Op can be a mixed bag of function pointers, potentially coming from multiple different op modules. This allows for hardware that only supports some of the operations, not all of them (e.g., a GPU that only supports single-precision operations). All the hard-coded back-end MPI_Op functions for (MPI_Op, MPI_Datatype) tuples still exist, but unlike coll, they're in the framework base (vs. being in a separate "basic" component) and are automatically used if no component is found at runtime that provides a module with the necessary function pointers. There is an "example" op component that will hopefully be useful to those writing meaningful op components. It is currently .ompi_ignore'd so that it doesn't impinge on other developers (it's somewhat chatty in terms of opal_output() so that you can tell when its functions have been invoked). See the README file in the example op component directory. Developers of new op components are encouraged to look at the following wiki pages: https://svn.open-mpi.org/trac/ompi/wiki/devel/Autogen https://svn.open-mpi.org/trac/ompi/wiki/devel/CreateComponent https://svn.open-mpi.org/trac/ompi/wiki/devel/CreateFramework = MPI_REDUCE_LOCAL = Part of the MPI-2.2 proposal listed here: https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/24 is to add a new function named MPI_REDUCE_LOCAL. It is very easy to implement, so I added it (also because it makes testing the op framework pretty easy -- you can do it in serial rather than via parallel reductions). There's even a man page! This commit was SVN r20280.
2009-01-14 23:44:31 +00:00
/*
* Copyright (c) 2004-2008 The Trustees of Indiana University and Indiana
* University Research and Technology
* Corporation. All rights reserved.
* Copyright (c) 2004-2005 The University of Tennessee and The University
* of Tennessee Research Foundation. All rights
* reserved.
* Copyright (c) 2004-2005 High Performance Computing Center Stuttgart,
* University of Stuttgart. All rights reserved.
* Copyright (c) 2004-2005 The Regents of the University of California.
* All rights reserved.
* Copyright (c) 2008-2009 Cisco Systems, Inc. All rights reserved.
Two major things in this commit: * New "op" MPI layer framework * Addition of the MPI_REDUCE_LOCAL proposed function (for MPI-2.2) = Op framework = Add new "op" framework in the ompi layer. This framework replaces the hard-coded MPI_Op back-end functions for (MPI_Op, MPI_Datatype) tuples for pre-defined MPI_Ops, allowing components and modules to provide the back-end functions. The intent is that components can be written to take advantage of hardware acceleration (GPU, FPGA, specialized CPU instructions, etc.). Similar to other frameworks, components are intended to be able to discover at run-time if they can be used, and if so, elect themselves to be selected (or disqualify themselves from selection if they cannot run). If specialized hardware is not available, there is a default set of functions that will automatically be used. This framework is ''not'' used for user-defined MPI_Ops. The new op framework is similar to the existing coll framework, in that the final set of function pointers that are used on any given intrinsic MPI_Op can be a mixed bag of function pointers, potentially coming from multiple different op modules. This allows for hardware that only supports some of the operations, not all of them (e.g., a GPU that only supports single-precision operations). All the hard-coded back-end MPI_Op functions for (MPI_Op, MPI_Datatype) tuples still exist, but unlike coll, they're in the framework base (vs. being in a separate "basic" component) and are automatically used if no component is found at runtime that provides a module with the necessary function pointers. There is an "example" op component that will hopefully be useful to those writing meaningful op components. It is currently .ompi_ignore'd so that it doesn't impinge on other developers (it's somewhat chatty in terms of opal_output() so that you can tell when its functions have been invoked). See the README file in the example op component directory. Developers of new op components are encouraged to look at the following wiki pages: https://svn.open-mpi.org/trac/ompi/wiki/devel/Autogen https://svn.open-mpi.org/trac/ompi/wiki/devel/CreateComponent https://svn.open-mpi.org/trac/ompi/wiki/devel/CreateFramework = MPI_REDUCE_LOCAL = Part of the MPI-2.2 proposal listed here: https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/24 is to add a new function named MPI_REDUCE_LOCAL. It is very easy to implement, so I added it (also because it makes testing the op framework pretty easy -- you can do it in serial rather than via parallel reductions). There's even a man page! This commit was SVN r20280.
2009-01-14 23:44:31 +00:00
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
#ifndef MCA_OP_EXAMPLE_EXPORT_H
#define MCA_OP_EXAMPLE_EXPORT_H
#include "ompi_config.h"
#include "opal/mca/mca.h"
#include "opal/class/opal_object.h"
#include "ompi/mca/op/op.h"
BEGIN_C_DECLS
/**
* Derive a struct from the base op component struct, allowing us to
* cache some component-specific information on our well-known
* component struct.
*/
typedef struct {
/** The base op component struct */
ompi_op_base_component_1_0_0_t super;
/* What follows is example-component-specific cached information. We
tend to use this scheme (caching information on the example
component itself) instead of lots of individual global
variables for the component. The following data fields are
examples; replace them with whatever is relevant for your
component. */
/** A simple boolean indicating that the hardware is available. */
bool hardware_available;
/** A simple boolean indicating whether double precision is
supported. */
bool double_supported;
} ompi_op_example_component_t;
/**
* Derive a struct from the base op module struct, allowing us to
* cache some module-specific information for BXOR. Note that
* information that should be shared across all modules should be put
* on the example component.
*/
typedef struct {
ompi_op_base_module_1_0_0_t super;
/* Just like the ompi_op_example_component_t, this struct is meant to
cache information on a per-module basis. What follows are
examples; replace them with whatever is relevant for your
component/module. Keep in mind that there will be one distinct
module for each MPI_Op; you may want to have different data
cached on the module, depending on the MPI_Op that it is
supporting. */
double some_bxor_data;
} ompi_op_example_module_bxor_t;
/**
* To use OMPI's OBJ system, you have to declare each "class".
*/
OBJ_CLASS_DECLARATION(ompi_op_example_module_bxor_t);
/**
* Globally exported variable. Note that it is a *example* component
* (defined above), which has the ompi_op_base_component_t as its
* first member. Hence, the MCA/op framework will find the data that
* it expects in the first memory locations, but then the component
* itself can cache additional information after that that can be used
* by both the component and modules.
*/
OMPI_DECLSPEC extern ompi_op_example_component_t
mca_op_example_component;
/**
* Setup for MPI_MAX and return a module.
*/
OMPI_DECLSPEC ompi_op_base_module_t *
ompi_op_example_setup_max(ompi_op_t *op);
/**
* Setup for MPI_BXOR and return a module.
*/
OMPI_DECLSPEC ompi_op_base_module_t *
ompi_op_example_setup_bxor(ompi_op_t *op);
END_C_DECLS
#endif /* MCA_OP_EXAMPLE_EXPORT_H */