1
1
openmpi/opal/mca/btl/vader/btl_vader_component.c

704 строки
28 KiB
C
Исходник Обычный вид История

/* -*- Mode: C; c-basic-offset:4 ; indent-tabs-mode:nil -*- */
/*
* Copyright (c) 2004-2011 The Trustees of Indiana University and Indiana
* University Research and Technology
* Corporation. All rights reserved.
* Copyright (c) 2004-2009 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) 2006-2007 Voltaire. All rights reserved.
* Copyright (c) 2009-2010 Cisco Systems, Inc. All rights reserved.
* Copyright (c) 2010-2014 Los Alamos National Security, LLC.
* All rights reserved.
* Copyright (c) 2011 NVIDIA Corporation. All rights reserved.
Per the PMIx RFC: WHAT: Merge the PMIx branch into the devel repo, creating a new OPAL “lmix” framework to abstract PMI support for all RTEs. Replace the ORTE daemon-level collectives with a new PMIx server and update the ORTE grpcomm framework to support server-to-server collectives WHY: We’ve had problems dealing with variations in PMI implementations, and need to extend the existing PMI definitions to meet exascale requirements. WHEN: Mon, Aug 25 WHERE: https://github.com/rhc54/ompi-svn-mirror.git Several community members have been working on a refactoring of the current PMI support within OMPI. Although the APIs are common, Slurm and Cray implement a different range of capabilities, and package them differently. For example, Cray provides an integrated PMI-1/2 library, while Slurm separates the two and requires the user to specify the one to be used at runtime. In addition, several bugs in the Slurm implementations have caused problems requiring extra coding. All this has led to a slew of #if’s in the PMI code and bugs when the corner-case logic for one implementation accidentally traps the other. Extending this support to other implementations would have increased this complexity to an unacceptable level. Accordingly, we have: * created a new OPAL “pmix” framework to abstract the PMI support, with separate components for Cray, Slurm PMI-1, and Slurm PMI-2 implementations. * Replaced the current ORTE grpcomm daemon-based collective operation with an integrated PMIx server, and updated the grpcomm APIs to provide more flexible, multi-algorithm support for collective operations. At this time, only the xcast and allgather operations are supported. * Replaced the current global collective id with a signature based on the names of the participating procs. The allows an unlimited number of collectives to be executed by any group of processes, subject to the requirement that only one collective can be active at a time for a unique combination of procs. Note that a proc can be involved in any number of simultaneous collectives - it is the specific combination of procs that is subject to the constraint * removed the prior OMPI/OPAL modex code * added new macros for executing modex send/recv to simplify use of the new APIs. The send macros allow the caller to specify whether or not the BTL supports async modex operations - if so, then the non-blocking “fence” operation is used, if the active PMIx component supports it. Otherwise, the default is a full blocking modex exchange as we currently perform. * retained the current flag that directs us to use a blocking fence operation, but only to retrieve data upon demand This commit was SVN r32570.
2014-08-21 22:56:47 +04:00
* Copyright (c) 2014 Intel, Inc. All rights reserved.
* Copyright (c) 2014 Research Organization for Information Science
* and Technology (RIST). All rights reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
2014-07-26 04:47:28 +04:00
#include "opal_config.h"
#include "opal/util/output.h"
#include "opal/util/show_help.h"
#include "opal/threads/mutex.h"
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
2014-07-26 04:47:28 +04:00
#include "opal/mca/btl/base/btl_base_error.h"
#include "btl_vader.h"
#include "btl_vader_frag.h"
#include "btl_vader_fifo.h"
#include "btl_vader_fbox.h"
#include "btl_vader_xpmem.h"
#include <sys/mman.h>
#include <fcntl.h>
#ifdef HAVE_SYS_PRCTL_H
#include <sys/prctl.h>
#endif
2014-10-21 19:27:36 +04:00
/* NTH: OS X does not define MAP_ANONYMOUS */
#if !defined(MAP_ANONYMOUS)
#define MAP_ANONYMOUS MAP_ANON
#endif
static int mca_btl_vader_component_progress (void);
static int mca_btl_vader_component_open(void);
static int mca_btl_vader_component_close(void);
static int mca_btl_vader_component_register(void);
static mca_btl_base_module_t** mca_btl_vader_component_init(int *num_btls,
bool enable_progress_threads,
bool enable_mpi_threads);
/* This enumeration is in order of preference */
static mca_base_var_enum_value_t single_copy_mechanisms[] = {
#if OPAL_BTL_VADER_HAVE_XPMEM
{.value = MCA_BTL_VADER_XPMEM, .string = "xpmem"},
#endif
#if OPAL_BTL_VADER_HAVE_CMA
{.value = MCA_BTL_VADER_CMA, .string = "cma"},
#endif
#if OPAL_BTL_VADER_HAVE_KNEM
{.value = MCA_BTL_VADER_KNEM, .string = "knem"},
#endif
{.value = MCA_BTL_VADER_NONE, .string = "none"},
{.value = 0, .string = NULL}
};
/*
* Shared Memory (VADER) component instance.
*/
mca_btl_vader_component_t mca_btl_vader_component = {
.super = {
/* First, the mca_base_component_t struct containing meta information
about the component itself */
.btl_version = {
Per RFC: Remove des_src and des_dst members from the mca_btl_base_segment_t and replace them with des_local and des_remote This change also updates the BTL version to 3.0.0. This commit does not represent the final version of BTL 3.0.0. More changes are coming. In making this change I updated all of the BTLs as well as BTL user's to use the new structure members. Please evaluate your component to ensure the changes are correct. RFC text: This is the first of several BTL interface changes I am proposing for the 1.9/2.0 release series. What: Change naming of btl descriptor members. I propose we change des_src and des_dst (and their associated counts) to be des_local and des_remote. For receive callbacks the des_local member will be used to communicate the segment information to the callback. The proposed change will include updating all of the doxygen in btl.h as well as updating all BTLs and BTL users to use the new naming scheme. Why: My btl usage makes use of both put and get operations on the same descriptor. With the current naming scheme I need to ensure that there is consistency beteen the segments described in des_src and des_dst depending on whether a put or get operation is executed. Additionally, the current naming prevents BTLs that do not require prepare/RMA matched operations (do not set MCA_BTL_FLAGS_RDMA_MATCHED) from executing multiple simultaneous put AND get operations. At the moment the descriptor can only be used with one or the other. The naming change makes it easier for BTL users to setup/modify descriptors for RMA operations as the local segment and remote segment are always in the same member field. The only issue I forsee with this change is that it will require a little more work to move BTL fixes to the 1.8 release series. This commit was SVN r32196.
2014-07-10 20:31:15 +04:00
MCA_BTL_DEFAULT_VERSION("vader"),
.mca_open_component = mca_btl_vader_component_open,
.mca_close_component = mca_btl_vader_component_close,
.mca_register_component_params = mca_btl_vader_component_register,
},
.btl_data = {
/* The component is checkpoint ready */
.param_field = MCA_BASE_METADATA_PARAM_CHECKPOINT
},
.btl_init = mca_btl_vader_component_init,
.btl_progress = mca_btl_vader_component_progress,
} /* end super */
};
static void mca_btl_vader_dummy_rdma (void)
{
/* If a backtrace ends at this function something has gone wrong with
* the btl bootstrapping. Check that the btl_get function was set to
* something reasonable. */
abort ();
}
static int mca_btl_vader_component_register (void)
{
mca_base_var_enum_t *new_enum;
MCA/base: Add new MCA variable system Features: - Support for an override parameter file (openmpi-mca-param-override.conf). Variable values in this file can not be overridden by any file or environment value. - Support for boolean, unsigned, and unsigned long long variables. - Support for true/false values. - Support for enumerations on integer variables. - Support for MPIT scope, verbosity, and binding. - Support for command line source. - Support for setting variable source via the environment using OMPI_MCA_SOURCE_<var name>=source (either command or file:filename) - Cleaner API. - Support for variable groups (equivalent to MPIT categories). Notes: - Variables must be created with a backing store (char **, int *, or bool *) that must live at least as long as the variable. - Creating a variable with the MCA_BASE_VAR_FLAG_SETTABLE enables the use of mca_base_var_set_value() to change the value. - String values are duplicated when the variable is registered. It is up to the caller to free the original value if necessary. The new value will be freed by the mca_base_var system and must not be freed by the user. - Variables with constant scope may not be settable. - Variable groups (and all associated variables) are deregistered when the component is closed or the component repository item is freed. This prevents a segmentation fault from accessing a variable after its component is unloaded. - After some discussion we decided we should remove the automatic registration of component priority variables. Few component actually made use of this feature. - The enumerator interface was updated to be general enough to handle future uses of the interface. - The code to generate ompi_info output has been moved into the MCA variable system. See mca_base_var_dump(). opal: update core and components to mca_base_var system orte: update core and components to mca_base_var system ompi: update core and components to mca_base_var system This commit also modifies the rmaps framework. The following variables were moved from ppr and lama: rmaps_base_pernode, rmaps_base_n_pernode, rmaps_base_n_persocket. Both lama and ppr create synonyms for these variables. This commit was SVN r28236.
2013-03-28 01:09:41 +04:00
(void) mca_base_var_group_component_register(&mca_btl_vader_component.super.btl_version,
"Enhanced shared memory byte transport later");
MCA/base: Add new MCA variable system Features: - Support for an override parameter file (openmpi-mca-param-override.conf). Variable values in this file can not be overridden by any file or environment value. - Support for boolean, unsigned, and unsigned long long variables. - Support for true/false values. - Support for enumerations on integer variables. - Support for MPIT scope, verbosity, and binding. - Support for command line source. - Support for setting variable source via the environment using OMPI_MCA_SOURCE_<var name>=source (either command or file:filename) - Cleaner API. - Support for variable groups (equivalent to MPIT categories). Notes: - Variables must be created with a backing store (char **, int *, or bool *) that must live at least as long as the variable. - Creating a variable with the MCA_BASE_VAR_FLAG_SETTABLE enables the use of mca_base_var_set_value() to change the value. - String values are duplicated when the variable is registered. It is up to the caller to free the original value if necessary. The new value will be freed by the mca_base_var system and must not be freed by the user. - Variables with constant scope may not be settable. - Variable groups (and all associated variables) are deregistered when the component is closed or the component repository item is freed. This prevents a segmentation fault from accessing a variable after its component is unloaded. - After some discussion we decided we should remove the automatic registration of component priority variables. Few component actually made use of this feature. - The enumerator interface was updated to be general enough to handle future uses of the interface. - The code to generate ompi_info output has been moved into the MCA variable system. See mca_base_var_dump(). opal: update core and components to mca_base_var system orte: update core and components to mca_base_var system ompi: update core and components to mca_base_var system This commit also modifies the rmaps framework. The following variables were moved from ppr and lama: rmaps_base_pernode, rmaps_base_n_pernode, rmaps_base_n_persocket. Both lama and ppr create synonyms for these variables. This commit was SVN r28236.
2013-03-28 01:09:41 +04:00
/* register VADER component variables */
mca_btl_vader_component.vader_free_list_num = 8;
(void) mca_base_component_var_register(&mca_btl_vader_component.super.btl_version,
"free_list_num", "Initial number of fragments "
"to allocate for shared memory communication.",
MCA_BASE_VAR_TYPE_INT, NULL, 0,
MCA_BASE_VAR_FLAG_SETTABLE, OPAL_INFO_LVL_9,
MCA/base: Add new MCA variable system Features: - Support for an override parameter file (openmpi-mca-param-override.conf). Variable values in this file can not be overridden by any file or environment value. - Support for boolean, unsigned, and unsigned long long variables. - Support for true/false values. - Support for enumerations on integer variables. - Support for MPIT scope, verbosity, and binding. - Support for command line source. - Support for setting variable source via the environment using OMPI_MCA_SOURCE_<var name>=source (either command or file:filename) - Cleaner API. - Support for variable groups (equivalent to MPIT categories). Notes: - Variables must be created with a backing store (char **, int *, or bool *) that must live at least as long as the variable. - Creating a variable with the MCA_BASE_VAR_FLAG_SETTABLE enables the use of mca_base_var_set_value() to change the value. - String values are duplicated when the variable is registered. It is up to the caller to free the original value if necessary. The new value will be freed by the mca_base_var system and must not be freed by the user. - Variables with constant scope may not be settable. - Variable groups (and all associated variables) are deregistered when the component is closed or the component repository item is freed. This prevents a segmentation fault from accessing a variable after its component is unloaded. - After some discussion we decided we should remove the automatic registration of component priority variables. Few component actually made use of this feature. - The enumerator interface was updated to be general enough to handle future uses of the interface. - The code to generate ompi_info output has been moved into the MCA variable system. See mca_base_var_dump(). opal: update core and components to mca_base_var system orte: update core and components to mca_base_var system ompi: update core and components to mca_base_var system This commit also modifies the rmaps framework. The following variables were moved from ppr and lama: rmaps_base_pernode, rmaps_base_n_pernode, rmaps_base_n_persocket. Both lama and ppr create synonyms for these variables. This commit was SVN r28236.
2013-03-28 01:09:41 +04:00
MCA_BASE_VAR_SCOPE_LOCAL,
&mca_btl_vader_component.vader_free_list_num);
mca_btl_vader_component.vader_free_list_max = 4096;
MCA/base: Add new MCA variable system Features: - Support for an override parameter file (openmpi-mca-param-override.conf). Variable values in this file can not be overridden by any file or environment value. - Support for boolean, unsigned, and unsigned long long variables. - Support for true/false values. - Support for enumerations on integer variables. - Support for MPIT scope, verbosity, and binding. - Support for command line source. - Support for setting variable source via the environment using OMPI_MCA_SOURCE_<var name>=source (either command or file:filename) - Cleaner API. - Support for variable groups (equivalent to MPIT categories). Notes: - Variables must be created with a backing store (char **, int *, or bool *) that must live at least as long as the variable. - Creating a variable with the MCA_BASE_VAR_FLAG_SETTABLE enables the use of mca_base_var_set_value() to change the value. - String values are duplicated when the variable is registered. It is up to the caller to free the original value if necessary. The new value will be freed by the mca_base_var system and must not be freed by the user. - Variables with constant scope may not be settable. - Variable groups (and all associated variables) are deregistered when the component is closed or the component repository item is freed. This prevents a segmentation fault from accessing a variable after its component is unloaded. - After some discussion we decided we should remove the automatic registration of component priority variables. Few component actually made use of this feature. - The enumerator interface was updated to be general enough to handle future uses of the interface. - The code to generate ompi_info output has been moved into the MCA variable system. See mca_base_var_dump(). opal: update core and components to mca_base_var system orte: update core and components to mca_base_var system ompi: update core and components to mca_base_var system This commit also modifies the rmaps framework. The following variables were moved from ppr and lama: rmaps_base_pernode, rmaps_base_n_pernode, rmaps_base_n_persocket. Both lama and ppr create synonyms for these variables. This commit was SVN r28236.
2013-03-28 01:09:41 +04:00
(void) mca_base_component_var_register(&mca_btl_vader_component.super.btl_version,
"free_list_max", "Maximum number of fragments "
"to allocate for shared memory communication.",
MCA_BASE_VAR_TYPE_INT, NULL, 0,
MCA_BASE_VAR_FLAG_SETTABLE, OPAL_INFO_LVL_9,
MCA/base: Add new MCA variable system Features: - Support for an override parameter file (openmpi-mca-param-override.conf). Variable values in this file can not be overridden by any file or environment value. - Support for boolean, unsigned, and unsigned long long variables. - Support for true/false values. - Support for enumerations on integer variables. - Support for MPIT scope, verbosity, and binding. - Support for command line source. - Support for setting variable source via the environment using OMPI_MCA_SOURCE_<var name>=source (either command or file:filename) - Cleaner API. - Support for variable groups (equivalent to MPIT categories). Notes: - Variables must be created with a backing store (char **, int *, or bool *) that must live at least as long as the variable. - Creating a variable with the MCA_BASE_VAR_FLAG_SETTABLE enables the use of mca_base_var_set_value() to change the value. - String values are duplicated when the variable is registered. It is up to the caller to free the original value if necessary. The new value will be freed by the mca_base_var system and must not be freed by the user. - Variables with constant scope may not be settable. - Variable groups (and all associated variables) are deregistered when the component is closed or the component repository item is freed. This prevents a segmentation fault from accessing a variable after its component is unloaded. - After some discussion we decided we should remove the automatic registration of component priority variables. Few component actually made use of this feature. - The enumerator interface was updated to be general enough to handle future uses of the interface. - The code to generate ompi_info output has been moved into the MCA variable system. See mca_base_var_dump(). opal: update core and components to mca_base_var system orte: update core and components to mca_base_var system ompi: update core and components to mca_base_var system This commit also modifies the rmaps framework. The following variables were moved from ppr and lama: rmaps_base_pernode, rmaps_base_n_pernode, rmaps_base_n_persocket. Both lama and ppr create synonyms for these variables. This commit was SVN r28236.
2013-03-28 01:09:41 +04:00
MCA_BASE_VAR_SCOPE_LOCAL,
&mca_btl_vader_component.vader_free_list_max);
mca_btl_vader_component.vader_free_list_inc = 64;
(void) mca_base_component_var_register(&mca_btl_vader_component.super.btl_version,
"free_list_inc", "Number of fragments to create "
"on each allocation.", MCA_BASE_VAR_TYPE_INT, NULL, 0,
MCA_BASE_VAR_FLAG_SETTABLE, OPAL_INFO_LVL_9,
MCA/base: Add new MCA variable system Features: - Support for an override parameter file (openmpi-mca-param-override.conf). Variable values in this file can not be overridden by any file or environment value. - Support for boolean, unsigned, and unsigned long long variables. - Support for true/false values. - Support for enumerations on integer variables. - Support for MPIT scope, verbosity, and binding. - Support for command line source. - Support for setting variable source via the environment using OMPI_MCA_SOURCE_<var name>=source (either command or file:filename) - Cleaner API. - Support for variable groups (equivalent to MPIT categories). Notes: - Variables must be created with a backing store (char **, int *, or bool *) that must live at least as long as the variable. - Creating a variable with the MCA_BASE_VAR_FLAG_SETTABLE enables the use of mca_base_var_set_value() to change the value. - String values are duplicated when the variable is registered. It is up to the caller to free the original value if necessary. The new value will be freed by the mca_base_var system and must not be freed by the user. - Variables with constant scope may not be settable. - Variable groups (and all associated variables) are deregistered when the component is closed or the component repository item is freed. This prevents a segmentation fault from accessing a variable after its component is unloaded. - After some discussion we decided we should remove the automatic registration of component priority variables. Few component actually made use of this feature. - The enumerator interface was updated to be general enough to handle future uses of the interface. - The code to generate ompi_info output has been moved into the MCA variable system. See mca_base_var_dump(). opal: update core and components to mca_base_var system orte: update core and components to mca_base_var system ompi: update core and components to mca_base_var system This commit also modifies the rmaps framework. The following variables were moved from ppr and lama: rmaps_base_pernode, rmaps_base_n_pernode, rmaps_base_n_persocket. Both lama and ppr create synonyms for these variables. This commit was SVN r28236.
2013-03-28 01:09:41 +04:00
MCA_BASE_VAR_SCOPE_LOCAL,
&mca_btl_vader_component.vader_free_list_inc);
mca_btl_vader_component.memcpy_limit = 524288;
MCA/base: Add new MCA variable system Features: - Support for an override parameter file (openmpi-mca-param-override.conf). Variable values in this file can not be overridden by any file or environment value. - Support for boolean, unsigned, and unsigned long long variables. - Support for true/false values. - Support for enumerations on integer variables. - Support for MPIT scope, verbosity, and binding. - Support for command line source. - Support for setting variable source via the environment using OMPI_MCA_SOURCE_<var name>=source (either command or file:filename) - Cleaner API. - Support for variable groups (equivalent to MPIT categories). Notes: - Variables must be created with a backing store (char **, int *, or bool *) that must live at least as long as the variable. - Creating a variable with the MCA_BASE_VAR_FLAG_SETTABLE enables the use of mca_base_var_set_value() to change the value. - String values are duplicated when the variable is registered. It is up to the caller to free the original value if necessary. The new value will be freed by the mca_base_var system and must not be freed by the user. - Variables with constant scope may not be settable. - Variable groups (and all associated variables) are deregistered when the component is closed or the component repository item is freed. This prevents a segmentation fault from accessing a variable after its component is unloaded. - After some discussion we decided we should remove the automatic registration of component priority variables. Few component actually made use of this feature. - The enumerator interface was updated to be general enough to handle future uses of the interface. - The code to generate ompi_info output has been moved into the MCA variable system. See mca_base_var_dump(). opal: update core and components to mca_base_var system orte: update core and components to mca_base_var system ompi: update core and components to mca_base_var system This commit also modifies the rmaps framework. The following variables were moved from ppr and lama: rmaps_base_pernode, rmaps_base_n_pernode, rmaps_base_n_persocket. Both lama and ppr create synonyms for these variables. This commit was SVN r28236.
2013-03-28 01:09:41 +04:00
(void) mca_base_component_var_register(&mca_btl_vader_component.super.btl_version,
"memcpy_limit", "Message size to switch from using "
"memove to memcpy. The relative speed of these two "
"routines can vary by size.", MCA_BASE_VAR_TYPE_INT,
NULL, 0, MCA_BASE_VAR_FLAG_SETTABLE, OPAL_INFO_LVL_5,
MCA/base: Add new MCA variable system Features: - Support for an override parameter file (openmpi-mca-param-override.conf). Variable values in this file can not be overridden by any file or environment value. - Support for boolean, unsigned, and unsigned long long variables. - Support for true/false values. - Support for enumerations on integer variables. - Support for MPIT scope, verbosity, and binding. - Support for command line source. - Support for setting variable source via the environment using OMPI_MCA_SOURCE_<var name>=source (either command or file:filename) - Cleaner API. - Support for variable groups (equivalent to MPIT categories). Notes: - Variables must be created with a backing store (char **, int *, or bool *) that must live at least as long as the variable. - Creating a variable with the MCA_BASE_VAR_FLAG_SETTABLE enables the use of mca_base_var_set_value() to change the value. - String values are duplicated when the variable is registered. It is up to the caller to free the original value if necessary. The new value will be freed by the mca_base_var system and must not be freed by the user. - Variables with constant scope may not be settable. - Variable groups (and all associated variables) are deregistered when the component is closed or the component repository item is freed. This prevents a segmentation fault from accessing a variable after its component is unloaded. - After some discussion we decided we should remove the automatic registration of component priority variables. Few component actually made use of this feature. - The enumerator interface was updated to be general enough to handle future uses of the interface. - The code to generate ompi_info output has been moved into the MCA variable system. See mca_base_var_dump(). opal: update core and components to mca_base_var system orte: update core and components to mca_base_var system ompi: update core and components to mca_base_var system This commit also modifies the rmaps framework. The following variables were moved from ppr and lama: rmaps_base_pernode, rmaps_base_n_pernode, rmaps_base_n_persocket. Both lama and ppr create synonyms for these variables. This commit was SVN r28236.
2013-03-28 01:09:41 +04:00
MCA_BASE_VAR_SCOPE_LOCAL,
&mca_btl_vader_component.memcpy_limit);
#if OPAL_BTL_VADER_HAVE_XPMEM
mca_btl_vader_component.log_attach_align = 21;
(void) mca_base_component_var_register(&mca_btl_vader_component.super.btl_version,
"log_align", "Log base 2 of the alignment to use for xpmem "
"segments (default: 21, minimum: 12, maximum: 25)",
MCA_BASE_VAR_TYPE_INT, NULL, 0,
MCA_BASE_VAR_FLAG_SETTABLE, OPAL_INFO_LVL_5,
MCA_BASE_VAR_SCOPE_LOCAL,
&mca_btl_vader_component.log_attach_align);
#endif
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
2014-07-26 04:47:28 +04:00
#if OPAL_BTL_VADER_HAVE_XPMEM && 64 == MCA_BTL_VADER_BITNESS
mca_btl_vader_component.segment_size = 1 << 24;
#else
mca_btl_vader_component.segment_size = 1 << 22;
#endif
MCA/base: Add new MCA variable system Features: - Support for an override parameter file (openmpi-mca-param-override.conf). Variable values in this file can not be overridden by any file or environment value. - Support for boolean, unsigned, and unsigned long long variables. - Support for true/false values. - Support for enumerations on integer variables. - Support for MPIT scope, verbosity, and binding. - Support for command line source. - Support for setting variable source via the environment using OMPI_MCA_SOURCE_<var name>=source (either command or file:filename) - Cleaner API. - Support for variable groups (equivalent to MPIT categories). Notes: - Variables must be created with a backing store (char **, int *, or bool *) that must live at least as long as the variable. - Creating a variable with the MCA_BASE_VAR_FLAG_SETTABLE enables the use of mca_base_var_set_value() to change the value. - String values are duplicated when the variable is registered. It is up to the caller to free the original value if necessary. The new value will be freed by the mca_base_var system and must not be freed by the user. - Variables with constant scope may not be settable. - Variable groups (and all associated variables) are deregistered when the component is closed or the component repository item is freed. This prevents a segmentation fault from accessing a variable after its component is unloaded. - After some discussion we decided we should remove the automatic registration of component priority variables. Few component actually made use of this feature. - The enumerator interface was updated to be general enough to handle future uses of the interface. - The code to generate ompi_info output has been moved into the MCA variable system. See mca_base_var_dump(). opal: update core and components to mca_base_var system orte: update core and components to mca_base_var system ompi: update core and components to mca_base_var system This commit also modifies the rmaps framework. The following variables were moved from ppr and lama: rmaps_base_pernode, rmaps_base_n_pernode, rmaps_base_n_persocket. Both lama and ppr create synonyms for these variables. This commit was SVN r28236.
2013-03-28 01:09:41 +04:00
(void) mca_base_component_var_register(&mca_btl_vader_component.super.btl_version,
"segment_size", "Maximum size of all shared "
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
2014-07-26 04:47:28 +04:00
#if OPAL_BTL_VADER_HAVE_XPMEM && 64 == MCA_BTL_VADER_BITNESS
"memory buffers (default: 16M)",
#else
"memory buffers (default: 4M)",
#endif
MCA_BASE_VAR_TYPE_INT, NULL, 0,
MCA_BASE_VAR_FLAG_SETTABLE, OPAL_INFO_LVL_5,
MCA/base: Add new MCA variable system Features: - Support for an override parameter file (openmpi-mca-param-override.conf). Variable values in this file can not be overridden by any file or environment value. - Support for boolean, unsigned, and unsigned long long variables. - Support for true/false values. - Support for enumerations on integer variables. - Support for MPIT scope, verbosity, and binding. - Support for command line source. - Support for setting variable source via the environment using OMPI_MCA_SOURCE_<var name>=source (either command or file:filename) - Cleaner API. - Support for variable groups (equivalent to MPIT categories). Notes: - Variables must be created with a backing store (char **, int *, or bool *) that must live at least as long as the variable. - Creating a variable with the MCA_BASE_VAR_FLAG_SETTABLE enables the use of mca_base_var_set_value() to change the value. - String values are duplicated when the variable is registered. It is up to the caller to free the original value if necessary. The new value will be freed by the mca_base_var system and must not be freed by the user. - Variables with constant scope may not be settable. - Variable groups (and all associated variables) are deregistered when the component is closed or the component repository item is freed. This prevents a segmentation fault from accessing a variable after its component is unloaded. - After some discussion we decided we should remove the automatic registration of component priority variables. Few component actually made use of this feature. - The enumerator interface was updated to be general enough to handle future uses of the interface. - The code to generate ompi_info output has been moved into the MCA variable system. See mca_base_var_dump(). opal: update core and components to mca_base_var system orte: update core and components to mca_base_var system ompi: update core and components to mca_base_var system This commit also modifies the rmaps framework. The following variables were moved from ppr and lama: rmaps_base_pernode, rmaps_base_n_pernode, rmaps_base_n_persocket. Both lama and ppr create synonyms for these variables. This commit was SVN r28236.
2013-03-28 01:09:41 +04:00
MCA_BASE_VAR_SCOPE_LOCAL,
&mca_btl_vader_component.segment_size);
MCA/base: Add new MCA variable system Features: - Support for an override parameter file (openmpi-mca-param-override.conf). Variable values in this file can not be overridden by any file or environment value. - Support for boolean, unsigned, and unsigned long long variables. - Support for true/false values. - Support for enumerations on integer variables. - Support for MPIT scope, verbosity, and binding. - Support for command line source. - Support for setting variable source via the environment using OMPI_MCA_SOURCE_<var name>=source (either command or file:filename) - Cleaner API. - Support for variable groups (equivalent to MPIT categories). Notes: - Variables must be created with a backing store (char **, int *, or bool *) that must live at least as long as the variable. - Creating a variable with the MCA_BASE_VAR_FLAG_SETTABLE enables the use of mca_base_var_set_value() to change the value. - String values are duplicated when the variable is registered. It is up to the caller to free the original value if necessary. The new value will be freed by the mca_base_var system and must not be freed by the user. - Variables with constant scope may not be settable. - Variable groups (and all associated variables) are deregistered when the component is closed or the component repository item is freed. This prevents a segmentation fault from accessing a variable after its component is unloaded. - After some discussion we decided we should remove the automatic registration of component priority variables. Few component actually made use of this feature. - The enumerator interface was updated to be general enough to handle future uses of the interface. - The code to generate ompi_info output has been moved into the MCA variable system. See mca_base_var_dump(). opal: update core and components to mca_base_var system orte: update core and components to mca_base_var system ompi: update core and components to mca_base_var system This commit also modifies the rmaps framework. The following variables were moved from ppr and lama: rmaps_base_pernode, rmaps_base_n_pernode, rmaps_base_n_persocket. Both lama and ppr create synonyms for these variables. This commit was SVN r28236.
2013-03-28 01:09:41 +04:00
mca_btl_vader_component.max_inline_send = 256;
MCA/base: Add new MCA variable system Features: - Support for an override parameter file (openmpi-mca-param-override.conf). Variable values in this file can not be overridden by any file or environment value. - Support for boolean, unsigned, and unsigned long long variables. - Support for true/false values. - Support for enumerations on integer variables. - Support for MPIT scope, verbosity, and binding. - Support for command line source. - Support for setting variable source via the environment using OMPI_MCA_SOURCE_<var name>=source (either command or file:filename) - Cleaner API. - Support for variable groups (equivalent to MPIT categories). Notes: - Variables must be created with a backing store (char **, int *, or bool *) that must live at least as long as the variable. - Creating a variable with the MCA_BASE_VAR_FLAG_SETTABLE enables the use of mca_base_var_set_value() to change the value. - String values are duplicated when the variable is registered. It is up to the caller to free the original value if necessary. The new value will be freed by the mca_base_var system and must not be freed by the user. - Variables with constant scope may not be settable. - Variable groups (and all associated variables) are deregistered when the component is closed or the component repository item is freed. This prevents a segmentation fault from accessing a variable after its component is unloaded. - After some discussion we decided we should remove the automatic registration of component priority variables. Few component actually made use of this feature. - The enumerator interface was updated to be general enough to handle future uses of the interface. - The code to generate ompi_info output has been moved into the MCA variable system. See mca_base_var_dump(). opal: update core and components to mca_base_var system orte: update core and components to mca_base_var system ompi: update core and components to mca_base_var system This commit also modifies the rmaps framework. The following variables were moved from ppr and lama: rmaps_base_pernode, rmaps_base_n_pernode, rmaps_base_n_persocket. Both lama and ppr create synonyms for these variables. This commit was SVN r28236.
2013-03-28 01:09:41 +04:00
(void) mca_base_component_var_register(&mca_btl_vader_component.super.btl_version,
"max_inline_send", "Maximum size to transfer "
"using copy-in copy-out semantics",
MCA_BASE_VAR_TYPE_UNSIGNED_INT, NULL, 0,
MCA_BASE_VAR_FLAG_SETTABLE, OPAL_INFO_LVL_5,
MCA_BASE_VAR_SCOPE_LOCAL,
&mca_btl_vader_component.max_inline_send);
mca_btl_vader_component.fbox_threshold = 16;
(void) mca_base_component_var_register(&mca_btl_vader_component.super.btl_version,
"fbox_threshold", "Number of sends required "
"before an eager send buffer is setup for a peer "
"(default: 16)", MCA_BASE_VAR_TYPE_UNSIGNED_INT, NULL,
0, MCA_BASE_VAR_FLAG_SETTABLE, OPAL_INFO_LVL_5,
MCA_BASE_VAR_SCOPE_LOCAL, &mca_btl_vader_component.fbox_threshold);
mca_btl_vader_component.fbox_max = 32;
(void) mca_base_component_var_register(&mca_btl_vader_component.super.btl_version,
"fbox_max", "Maximum number of eager send buffers "
"to allocate (default: 32)", MCA_BASE_VAR_TYPE_UNSIGNED_INT,
NULL, 0, MCA_BASE_VAR_FLAG_SETTABLE, OPAL_INFO_LVL_5,
MCA_BASE_VAR_SCOPE_LOCAL, &mca_btl_vader_component.fbox_max);
mca_btl_vader_component.fbox_size = 4096;
(void) mca_base_component_var_register(&mca_btl_vader_component.super.btl_version,
"fbox_size", "Size of per-peer fast transfer buffers (default: 4k)",
MCA_BASE_VAR_TYPE_UNSIGNED_INT, NULL, 0, MCA_BASE_VAR_FLAG_SETTABLE,
OPAL_INFO_LVL_5, MCA_BASE_VAR_SCOPE_LOCAL, &mca_btl_vader_component.fbox_size);
(void) mca_base_var_enum_create ("btl_vader_single_copy_mechanisms", single_copy_mechanisms, &new_enum);
/* Default to the best available mechanism (see the enumerator for ordering) */
mca_btl_vader_component.single_copy_mechanism = single_copy_mechanisms[0].value;
(void) mca_base_component_var_register(&mca_btl_vader_component.super.btl_version,
"single_copy_mechanism", "Single copy mechanism to use (defaults to best available)",
MCA_BASE_VAR_TYPE_INT, new_enum, 0, MCA_BASE_VAR_FLAG_SETTABLE,
OPAL_INFO_LVL_3, MCA_BASE_VAR_SCOPE_GROUP, &mca_btl_vader_component.single_copy_mechanism);
#if OPAL_BTL_VADER_HAVE_KNEM
/* Currently disabling DMA mode by default; it's not clear that this is useful in all applications and architectures. */
mca_btl_vader_component.knem_dma_min = 0;
(void) mca_base_component_var_register(&mca_btl_vader_component.super.btl_version, "knem_dma_min",
"Minimum message size (in bytes) to use the knem DMA mode; "
"ignored if knem does not support DMA mode (0 = do not use the "
"knem DMA mode, default: 0)", MCA_BASE_VAR_TYPE_UNSIGNED_INT,
NULL, 0, 0, OPAL_INFO_LVL_9, MCA_BASE_VAR_SCOPE_READONLY,
&mca_btl_vader_component.knem_dma_min);
#endif
mca_btl_vader.super.btl_exclusivity = MCA_BTL_EXCLUSIVITY_HIGH;
if (MCA_BTL_VADER_XPMEM == mca_btl_vader_component.single_copy_mechanism) {
mca_btl_vader.super.btl_eager_limit = 32 * 1024;
mca_btl_vader.super.btl_rndv_eager_limit = mca_btl_vader.super.btl_eager_limit;
mca_btl_vader.super.btl_max_send_size = mca_btl_vader.super.btl_eager_limit;
mca_btl_vader.super.btl_min_rdma_pipeline_size = mca_btl_vader.super.btl_eager_limit;
} else {
mca_btl_vader.super.btl_eager_limit = 4 * 1024;
mca_btl_vader.super.btl_rndv_eager_limit = 32 * 1024;
mca_btl_vader.super.btl_max_send_size = 32 * 1024;
mca_btl_vader.super.btl_min_rdma_pipeline_size = 32 * 1024;
}
mca_btl_vader.super.btl_rdma_pipeline_send_length = mca_btl_vader.super.btl_eager_limit;
mca_btl_vader.super.btl_rdma_pipeline_frag_size = mca_btl_vader.super.btl_eager_limit;
if (MCA_BTL_VADER_NONE != mca_btl_vader_component.single_copy_mechanism) {
mca_btl_vader.super.btl_flags = MCA_BTL_FLAGS_RDMA | MCA_BTL_FLAGS_SEND_INPLACE;
/* Single copy mechanisms should provide better bandwidth */
mca_btl_vader.super.btl_bandwidth = 40000; /* Mbs */
/* Set dummy values so the RDMA flag doesn't get unset by mca_btl_base_param_verify() */
mca_btl_vader.super.btl_get = (mca_btl_base_module_get_fn_t) mca_btl_vader_dummy_rdma;
mca_btl_vader.super.btl_put = (mca_btl_base_module_get_fn_t) mca_btl_vader_dummy_rdma;
} else {
mca_btl_vader.super.btl_flags = MCA_BTL_FLAGS_SEND_INPLACE;
mca_btl_vader.super.btl_bandwidth = 10000; /* Mbs */
}
mca_btl_vader.super.btl_seg_size = sizeof (mca_btl_vader_segment_t);
mca_btl_vader.super.btl_latency = 1; /* Microsecs */
/* Call the BTL based to register its MCA params */
mca_btl_base_param_register(&mca_btl_vader_component.super.btl_version,
&mca_btl_vader.super);
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
2014-07-26 04:47:28 +04:00
return OPAL_SUCCESS;
}
/*
* Called by MCA framework to open the component, registers
* component parameters.
*/
static int mca_btl_vader_component_open(void)
{
/* initialize objects */
OBJ_CONSTRUCT(&mca_btl_vader_component.vader_frags_eager, ompi_free_list_t);
OBJ_CONSTRUCT(&mca_btl_vader_component.vader_frags_user, ompi_free_list_t);
OBJ_CONSTRUCT(&mca_btl_vader_component.vader_frags_max_send, ompi_free_list_t);
OBJ_CONSTRUCT(&mca_btl_vader_component.vader_frags_rdma, ompi_free_list_t);
OBJ_CONSTRUCT(&mca_btl_vader_component.lock, opal_mutex_t);
OBJ_CONSTRUCT(&mca_btl_vader_component.pending_endpoints, opal_list_t);
OBJ_CONSTRUCT(&mca_btl_vader_component.pending_fragments, opal_list_t);
#if OPAL_BTL_VADER_HAVE_KNEM
mca_btl_vader.knem_fd = -1;
#endif
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
2014-07-26 04:47:28 +04:00
return OPAL_SUCCESS;
}
/*
* component cleanup - sanity checking of queue lengths
*/
static int mca_btl_vader_component_close(void)
{
OBJ_DESTRUCT(&mca_btl_vader_component.vader_frags_eager);
OBJ_DESTRUCT(&mca_btl_vader_component.vader_frags_user);
OBJ_DESTRUCT(&mca_btl_vader_component.vader_frags_max_send);
OBJ_DESTRUCT(&mca_btl_vader_component.vader_frags_rdma);
OBJ_DESTRUCT(&mca_btl_vader_component.lock);
OBJ_DESTRUCT(&mca_btl_vader_component.pending_endpoints);
OBJ_DESTRUCT(&mca_btl_vader_component.pending_fragments);
if (NULL != mca_btl_vader_component.my_segment) {
munmap (mca_btl_vader_component.my_segment, mca_btl_vader_component.segment_size);
}
#if OPAL_BTL_VADER_HAVE_KNEM
mca_btl_vader_knem_fini ();
#endif
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
2014-07-26 04:47:28 +04:00
return OPAL_SUCCESS;
}
static int mca_btl_base_vader_modex_send (void)
{
union vader_modex_t modex;
int modex_size, rc;
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
2014-07-26 04:47:28 +04:00
#if OPAL_BTL_VADER_HAVE_XPMEM
if (MCA_BTL_VADER_XPMEM == mca_btl_vader_component.single_copy_mechanism) {
modex.xpmem.seg_id = mca_btl_vader_component.my_seg_id;
modex.xpmem.segment_base = mca_btl_vader_component.my_segment;
modex_size = sizeof (modex);
} else {
#endif
modex_size = opal_shmem_sizeof_shmem_ds (&mca_btl_vader_component.seg_ds);
memmove (&modex.seg_ds, &mca_btl_vader_component.seg_ds, modex_size);
#if OPAL_BTL_VADER_HAVE_XPMEM
}
#endif
OPAL_MODEX_SEND(rc, PMIX_SYNC_REQD, PMIX_LOCAL,
&mca_btl_vader_component.super.btl_version, &modex, modex_size);
Per the PMIx RFC: WHAT: Merge the PMIx branch into the devel repo, creating a new OPAL “lmix” framework to abstract PMI support for all RTEs. Replace the ORTE daemon-level collectives with a new PMIx server and update the ORTE grpcomm framework to support server-to-server collectives WHY: We’ve had problems dealing with variations in PMI implementations, and need to extend the existing PMI definitions to meet exascale requirements. WHEN: Mon, Aug 25 WHERE: https://github.com/rhc54/ompi-svn-mirror.git Several community members have been working on a refactoring of the current PMI support within OMPI. Although the APIs are common, Slurm and Cray implement a different range of capabilities, and package them differently. For example, Cray provides an integrated PMI-1/2 library, while Slurm separates the two and requires the user to specify the one to be used at runtime. In addition, several bugs in the Slurm implementations have caused problems requiring extra coding. All this has led to a slew of #if’s in the PMI code and bugs when the corner-case logic for one implementation accidentally traps the other. Extending this support to other implementations would have increased this complexity to an unacceptable level. Accordingly, we have: * created a new OPAL “pmix” framework to abstract the PMI support, with separate components for Cray, Slurm PMI-1, and Slurm PMI-2 implementations. * Replaced the current ORTE grpcomm daemon-based collective operation with an integrated PMIx server, and updated the grpcomm APIs to provide more flexible, multi-algorithm support for collective operations. At this time, only the xcast and allgather operations are supported. * Replaced the current global collective id with a signature based on the names of the participating procs. The allows an unlimited number of collectives to be executed by any group of processes, subject to the requirement that only one collective can be active at a time for a unique combination of procs. Note that a proc can be involved in any number of simultaneous collectives - it is the specific combination of procs that is subject to the constraint * removed the prior OMPI/OPAL modex code * added new macros for executing modex send/recv to simplify use of the new APIs. The send macros allow the caller to specify whether or not the BTL supports async modex operations - if so, then the non-blocking “fence” operation is used, if the active PMIx component supports it. Otherwise, the default is a full blocking modex exchange as we currently perform. * retained the current flag that directs us to use a blocking fence operation, but only to retrieve data upon demand This commit was SVN r32570.
2014-08-21 22:56:47 +04:00
return rc;
}
#if OPAL_BTL_VADER_HAVE_XPMEM || OPAL_BTL_VADER_HAVE_CMA || OPAL_BTL_VADER_HAVE_KNEM
static void mca_btl_vader_select_next_single_copy_mechanism (void)
{
for (int i = 0 ; single_copy_mechanisms[i].value != MCA_BTL_VADER_NONE ; ++i) {
if (single_copy_mechanisms[i].value == mca_btl_vader_component.single_copy_mechanism) {
mca_btl_vader_component.single_copy_mechanism = single_copy_mechanisms[i+1].value;
return;
}
}
}
static void mca_btl_vader_check_single_copy (void)
{
int initial_mechanism = mca_btl_vader_component.single_copy_mechanism;
int rc;
#if OPAL_BTL_VADER_HAVE_XPMEM
if (MCA_BTL_VADER_XPMEM == mca_btl_vader_component.single_copy_mechanism) {
/* try to create an xpmem segment for the entire address space */
mca_btl_vader_component.my_seg_id = xpmem_make (0, VADER_MAX_ADDRESS, XPMEM_PERMIT_MODE, (void *)0666);
if (-1 == mca_btl_vader_component.my_seg_id) {
if (MCA_BTL_VADER_XPMEM == initial_mechanism) {
opal_show_help("help-btl-vader.txt", "xpmem-make-failed",
true, opal_process_info.nodename, errno,
strerror(errno));
}
mca_btl_vader_select_next_single_copy_mechanism ();
} else {
mca_btl_vader.super.btl_get = mca_btl_vader_get_xpmem;
mca_btl_vader.super.btl_put = mca_btl_vader_get_xpmem;
}
}
#endif
#if OPAL_BTL_VADER_HAVE_CMA
if (MCA_BTL_VADER_CMA == mca_btl_vader_component.single_copy_mechanism) {
/* Check if we have the proper permissions for CMA */
char buffer = {0};
bool cma_happy = false;
int fd;
/* check system setting for current ptrace scope */
fd = open ("/proc/sys/kernel/yama/ptrace_scope", O_RDONLY);
if (0 > fd) {
read (fd, &buffer, 1);
close (fd);
}
/* ptrace scope 0 will allow an attach from any of the process owner's
* processes. ptrace scope 1 limits attachers to the process tree
* starting at the parent of this process. */
if ('0' != buffer) {
#if defined PR_SET_PTRACER
/* try setting the ptrace scope to allow attach */
int ret = prctl (PR_SET_PTRACER, PR_SET_PTRACER_ANY, 0, 0, 0);
if (0 == ret) {
cma_happy = true;
}
#endif
} else {
cma_happy = true;
}
if (!cma_happy) {
mca_btl_vader_select_next_single_copy_mechanism ();
if (MCA_BTL_VADER_CMA == initial_mechanism) {
opal_show_help("help-btl-vader.txt", "cma-permission-denied",
true, opal_process_info.nodename);
}
} else {
/* ptrace_scope will allow CMA */
mca_btl_vader.super.btl_get = mca_btl_vader_get_cma;
mca_btl_vader.super.btl_put = mca_btl_vader_put_cma;
}
}
#endif
#if OPAL_BTL_VADER_HAVE_KNEM
if (MCA_BTL_VADER_KNEM == mca_btl_vader_component.single_copy_mechanism) {
/* mca_btl_vader_knem_init will set the appropriate get/put functions */
rc = mca_btl_vader_knem_init ();
if (OPAL_SUCCESS != rc) {
if (MCA_BTL_VADER_KNEM == initial_mechanism) {
opal_show_help("help-btl-vader.txt", "knem requested but not available",
true, opal_process_info.nodename);
}
/* disable single copy */
mca_btl_vader_select_next_single_copy_mechanism ();
}
}
#endif
if (MCA_BTL_VADER_NONE == mca_btl_vader_component.single_copy_mechanism) {
mca_btl_vader.super.btl_flags &= ~MCA_BTL_FLAGS_RDMA;
mca_btl_vader.super.btl_get = NULL;
mca_btl_vader.super.btl_put = NULL;
}
}
#endif
/*
* VADER component initialization
*/
static mca_btl_base_module_t **mca_btl_vader_component_init (int *num_btls,
bool enable_progress_threads,
bool enable_mpi_threads)
{
mca_btl_vader_component_t *component = &mca_btl_vader_component;
mca_btl_base_module_t **btls = NULL;
int rc;
*num_btls = 0;
/* disable if there are no local peers */
if (0 == MCA_BTL_VADER_NUM_LOCAL_PEERS) {
BTL_VERBOSE(("No peers to communicate with. Disabling vader."));
return NULL;
}
#if OPAL_BTL_VADER_HAVE_XPMEM
/* limit segment alignment to be between 4k and 16M */
if (component->log_attach_align < 12) {
component->log_attach_align = 12;
} else if (component->log_attach_align > 25) {
component->log_attach_align = 25;
}
#endif
btls = (mca_btl_base_module_t **) calloc (1, sizeof (mca_btl_base_module_t *));
if (NULL == btls) {
return NULL;
}
/* ensure a sane segment size */
if (component->segment_size < (2 << 20)) {
component->segment_size = (2 << 20);
}
component->fbox_size = (component->fbox_size + MCA_BTL_VADER_FBOX_ALIGNMENT_MASK) & ~MCA_BTL_VADER_FBOX_ALIGNMENT_MASK;
if (component->segment_size > (1ul << MCA_BTL_VADER_OFFSET_BITS)) {
component->segment_size = 2ul << MCA_BTL_VADER_OFFSET_BITS;
}
/* no fast boxes allocated initially */
component->num_fbox_in_endpoints = 0;
component->fbox_count = 0;
#if OPAL_BTL_VADER_HAVE_XPMEM || OPAL_BTL_VADER_HAVE_CMA || OPAL_BTL_VADER_HAVE_KNEM
mca_btl_vader_check_single_copy ();
#endif
if (MCA_BTL_VADER_XPMEM != mca_btl_vader_component.single_copy_mechanism) {
char *sm_file;
rc = asprintf(&sm_file, "%s" OPAL_PATH_SEP "vader_segment.%s.%d", opal_process_info.proc_session_dir,
opal_process_info.nodename, MCA_BTL_VADER_LOCAL_RANK);
if (0 > rc) {
free (btls);
return NULL;
}
rc = opal_shmem_segment_create (&component->seg_ds, sm_file, component->segment_size);
free (sm_file);
if (OPAL_SUCCESS != rc) {
BTL_VERBOSE(("Could not create shared memory segment"));
free (btls);
return NULL;
}
component->my_segment = opal_shmem_segment_attach (&component->seg_ds);
if (NULL == component->my_segment) {
BTL_VERBOSE(("Could not attach to just created shared memory segment"));
goto failed;
}
} else {
/* when using xpmem it is safe to use an anonymous segment */
component->my_segment = mmap (NULL, component->segment_size, PROT_READ |
PROT_WRITE, MAP_ANONYMOUS | MAP_SHARED, -1, 0);
if ((void *)-1 == component->my_segment) {
BTL_VERBOSE(("Could not create anonymous memory segment"));
free (btls);
return NULL;
}
}
component->segment_offset = 0;
/* initialize my fifo */
vader_fifo_init ((struct vader_fifo_t *) component->my_segment);
rc = mca_btl_base_vader_modex_send ();
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
2014-07-26 04:47:28 +04:00
if (OPAL_SUCCESS != rc) {
BTL_VERBOSE(("Error sending modex"));
goto failed;
}
*num_btls = 1;
/* get pointer to the btls */
btls[0] = (mca_btl_base_module_t *) &mca_btl_vader;
/* set flag indicating btl not inited */
mca_btl_vader.btl_inited = false;
return btls;
failed:
George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-) WHAT: Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies. This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP. Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose. UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs. A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic. This commit was SVN r32317.
2014-07-26 04:47:28 +04:00
#if OPAL_BTL_VADER_HAVE_XPMEM
if (MCA_BTL_VADER_XPMEM == mca_btl_vader_component.single_copy_mechanism) {
munmap (component->my_segment, component->segment_size);
} else
#endif
opal_shmem_unlink (&component->seg_ds);
if (btls) {
free (btls);
}
return NULL;
}
void mca_btl_vader_poll_handle_frag (mca_btl_vader_hdr_t *hdr, struct mca_btl_base_endpoint_t *endpoint)
{
mca_btl_base_segment_t segments[2];
mca_btl_base_descriptor_t frag = {.des_local = segments, .des_local_count = 1};
const mca_btl_active_message_callback_t *reg;
if (hdr->flags & MCA_BTL_VADER_FLAG_COMPLETE) {
mca_btl_vader_frag_complete (hdr->frag);
return;
}
reg = mca_btl_base_active_message_trigger + hdr->tag;
segments[0].seg_addr.pval = (void *) (hdr + 1);
segments[0].seg_len = hdr->len;
if (hdr->flags & MCA_BTL_VADER_FLAG_SINGLE_COPY) {
mca_mpool_base_registration_t *xpmem_reg;
xpmem_reg = vader_get_registation (endpoint, hdr->sc_iov.iov_base,
hdr->sc_iov.iov_len, 0,
&segments[1].seg_addr.pval);
segments[1].seg_len = hdr->sc_iov.iov_len;
frag.des_local_count = 2;
/* recv upcall */
reg->cbfunc(&mca_btl_vader.super, hdr->tag, &frag, reg->cbdata);
vader_return_registration (xpmem_reg, endpoint);
} else {
reg->cbfunc(&mca_btl_vader.super, hdr->tag, &frag, reg->cbdata);
}
if (OPAL_UNLIKELY(MCA_BTL_VADER_FLAG_SETUP_FBOX & hdr->flags)) {
mca_btl_vader_endpoint_setup_fbox_recv (endpoint, relative2virtual(hdr->fbox_base));
mca_btl_vader_component.fbox_in_endpoints[mca_btl_vader_component.num_fbox_in_endpoints++] = endpoint;
}
hdr->flags = MCA_BTL_VADER_FLAG_COMPLETE;
vader_fifo_write_back (hdr, endpoint);
}
static int mca_btl_vader_poll_fifo (void)
{
struct mca_btl_base_endpoint_t *endpoint;
mca_btl_vader_hdr_t *hdr;
/* poll the fifo until it is empty or a limit has been hit (8 is arbitrary) */
for (int fifo_count = 0 ; fifo_count < 31 ; ++fifo_count) {
hdr = vader_fifo_read (mca_btl_vader_component.my_fifo, &endpoint);
if (NULL == hdr) {
return fifo_count;
}
mca_btl_vader_poll_handle_frag (hdr, endpoint);
}
return 1;
}
/**
* Progress pending messages on an endpoint
*
* @param ep (IN) Vader BTL endpoint
*
* This is called with the component lock held so the component lock does
* not need to be aquired before modifying the pending_endpoints list.
*/
static void mca_btl_vader_progress_waiting (mca_btl_base_endpoint_t *ep)
{
mca_btl_vader_frag_t *frag;
if (OPAL_UNLIKELY(NULL == ep)) {
return;
}
OPAL_THREAD_LOCK(&ep->lock);
ep->waiting = false;
while (NULL != (frag = (mca_btl_vader_frag_t *) opal_list_remove_first (&ep->pending_frags))) {
OPAL_THREAD_UNLOCK(&ep->lock);
if (!vader_fifo_write_ep (frag->hdr, ep)) {
opal_list_prepend (&ep->pending_frags, (opal_list_item_t *) frag);
opal_list_append (&mca_btl_vader_component.pending_endpoints, &ep->super);
ep->waiting = true;
break;
}
OPAL_THREAD_LOCK(&ep->lock);
}
OPAL_THREAD_UNLOCK(&ep->lock);
}
/**
* Progress pending messages on all waiting endpoints
*
* @param ep (IN) Vader BTL endpoint
*/
static void mca_btl_vader_progress_endpoints (void)
{
int count;
count = opal_list_get_size (&mca_btl_vader_component.pending_endpoints);
if (OPAL_LIKELY(0 == count)) {
return;
}
OPAL_THREAD_LOCK(&mca_btl_vader_component.lock);
for (int i = 0 ; i < count ; ++i) {
mca_btl_vader_progress_waiting ((mca_btl_base_endpoint_t *) opal_list_remove_first (&mca_btl_vader_component.pending_endpoints));
}
OPAL_THREAD_LOCK(&mca_btl_vader_component.lock);
}
static int mca_btl_vader_component_progress (void)
{
static int32_t lock = 0;
int count = 0;
if (opal_using_threads()) {
if (opal_atomic_swap_32 (&lock, 1)) {
return 0;
}
}
/* check for messages in fast boxes */
if (mca_btl_vader_component.num_fbox_in_endpoints) {
count = mca_btl_vader_check_fboxes ();
}
mca_btl_vader_progress_endpoints ();
if (VADER_FIFO_FREE == mca_btl_vader_component.my_fifo->fifo_head) {
lock = 0;
return count;
}
count += mca_btl_vader_poll_fifo ();
opal_atomic_mb ();
lock = 0;
return count;
}