Sorry for mid-day commit, but I had promised on the call to do this upon my return.
Roll in the ORTE state machine. Remove last traces of opal_sos. Remove UTK epoch code. Please see the various emails about the state machine change for details. I'll send something out later with more info on the new arch. This commit was SVN r26242.
This commit is contained in:
parent
55d9c71ce0
commit
bd8b4f7f1e
@ -32,6 +32,8 @@ my @globals = qw/.libs
|
||||
*.orig
|
||||
*.rej
|
||||
*.class
|
||||
*.xcscheme
|
||||
*.plist
|
||||
.git*
|
||||
.DS_Store
|
||||
stamp-h[1-9]
|
||||
|
28
contrib/platform/iu/odin/debug-nopmi
Normal file
28
contrib/platform/iu/odin/debug-nopmi
Normal file
@ -0,0 +1,28 @@
|
||||
enable_opal_multi_threads=no
|
||||
enable_dlopen=no
|
||||
enable_pty_support=no
|
||||
with_blcr=no
|
||||
with_openib=no
|
||||
with_memory_manager=no
|
||||
enable_mem_debug=yes
|
||||
enable_mem_profile=no
|
||||
enable_debug_symbols=yes
|
||||
enable_binaries=yes
|
||||
with_devel_headers=yes
|
||||
enable_heterogeneous=no
|
||||
enable_picky=yes
|
||||
enable_debug=yes
|
||||
enable_shared=yes
|
||||
enable_static=no
|
||||
with_slurm=yes
|
||||
with_pmi=no
|
||||
enable_contrib_no_build=libnbc,vt
|
||||
enable_visibility=yes
|
||||
enable_memchecker=no
|
||||
enable_ipv6=no
|
||||
enable_mpi_f77=no
|
||||
enable_mpi_f90=no
|
||||
enable_mpi_cxx=no
|
||||
enable_mpi_cxx_seek=no
|
||||
enable_mca_no_build=pml-dr,pml-crcp2,crcp
|
||||
enable_io_romio=no
|
85
contrib/platform/iu/odin/debug-nopmi.conf
Normal file
85
contrib/platform/iu/odin/debug-nopmi.conf
Normal file
@ -0,0 +1,85 @@
|
||||
#
|
||||
# Copyright (c) 2004-2005 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) 2006 Cisco Systems, Inc. All rights reserved.
|
||||
# $COPYRIGHT$
|
||||
#
|
||||
# Additional copyrights may follow
|
||||
#
|
||||
# $HEADER$
|
||||
#
|
||||
|
||||
# This is the default system-wide MCA parameters defaults file.
|
||||
# Specifically, the MCA parameter "mca_param_files" defaults to a
|
||||
# value of
|
||||
# "$HOME/.openmpi/mca-params.conf:$sysconf/openmpi-mca-params.conf"
|
||||
# (this file is the latter of the two). So if the default value of
|
||||
# mca_param_files is not changed, this file is used to set system-wide
|
||||
# MCA parameters. This file can therefore be used to set system-wide
|
||||
# default MCA parameters for all users. Of course, users can override
|
||||
# these values if they want, but this file is an excellent location
|
||||
# for setting system-specific MCA parameters for those users who don't
|
||||
# know / care enough to investigate the proper values for them.
|
||||
|
||||
# Note that this file is only applicable where it is visible (in a
|
||||
# filesystem sense). Specifically, MPI processes each read this file
|
||||
# during their startup to determine what default values for MCA
|
||||
# parameters should be used. mpirun does not bundle up the values in
|
||||
# this file from the node where it was run and send them to all nodes;
|
||||
# the default value decisions are effectively distributed. Hence,
|
||||
# these values are only applicable on nodes that "see" this file. If
|
||||
# $sysconf is a directory on a local disk, it is likely that changes
|
||||
# to this file will need to be propagated to other nodes. If $sysconf
|
||||
# is a directory that is shared via a networked filesystem, changes to
|
||||
# this file will be visible to all nodes that share this $sysconf.
|
||||
|
||||
# The format is straightforward: one per line, mca_param_name =
|
||||
# rvalue. Quoting is ignored (so if you use quotes or escape
|
||||
# characters, they'll be included as part of the value). For example:
|
||||
|
||||
# Disable run-time MPI parameter checking
|
||||
# mpi_param_check = 0
|
||||
|
||||
# Note that the value "~/" will be expanded to the current user's home
|
||||
# directory. For example:
|
||||
|
||||
# Change component loading path
|
||||
# component_path = /usr/local/lib/openmpi:~/my_openmpi_components
|
||||
|
||||
# See "ompi_info --param all all" for a full listing of Open MPI MCA
|
||||
# parameters available and their default values.
|
||||
#
|
||||
|
||||
# Basic behavior to smooth startup
|
||||
mca_component_show_load_errors = 0
|
||||
mpi_param_check = 0
|
||||
orte_abort_timeout = 10
|
||||
hwloc_base_mem_bind_failure_action = silent
|
||||
|
||||
## Protect the shared file systems
|
||||
|
||||
## Add the interface for out-of-band communication
|
||||
## and set it up
|
||||
oob_tcp_peer_retries = 120
|
||||
oob_tcp_disable_family = IPv6
|
||||
#oob_tcp_connect_timeout=600
|
||||
|
||||
## Define the MPI interconnects
|
||||
btl = sm,tcp,self
|
||||
|
||||
## Setup shared memory
|
||||
btl_sm_free_list_max = 768
|
||||
|
||||
## Setup TCP
|
||||
btl_tcp_if_include = ib0
|
||||
|
||||
## Configure the PML
|
||||
pml_ob1_use_early_completion = 0
|
@ -10,6 +10,8 @@
|
||||
* Copyright (c) 2004-2005 The Regents of the University of California.
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2006-2012 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
* Additional copyrights may follow
|
||||
@ -195,7 +197,6 @@
|
||||
#include "ompi/attribute/attribute.h"
|
||||
#include "opal/class/opal_bitmap.h"
|
||||
#include "opal/threads/mutex.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
|
||||
#include "ompi/constants.h"
|
||||
#include "ompi/datatype/ompi_datatype.h"
|
||||
@ -1151,7 +1152,7 @@ static int get_value(opal_hash_table_t *attr_hash, int key,
|
||||
(void**) &keyval);
|
||||
OPAL_THREAD_UNLOCK(&keyval_hash_lock);
|
||||
|
||||
if (OMPI_ERR_NOT_FOUND == OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
if (OMPI_ERR_NOT_FOUND == ret) {
|
||||
return MPI_KEYVAL_INVALID;
|
||||
}
|
||||
|
||||
|
@ -14,6 +14,8 @@
|
||||
* Copyright (c) 2007 Voltaire All rights reserved.
|
||||
* Copyright (c) 2006-2010 University of Houston. All rights reserved.
|
||||
* Copyright (c) 2009 Sun Microsystems, Inc. All rights reserved.
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* Copyright (c) 2012 Oak Ridge National Labs. All rights reserved.
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
@ -32,7 +34,6 @@
|
||||
#include "ompi/constants.h"
|
||||
#include "opal/class/opal_pointer_array.h"
|
||||
#include "opal/class/opal_list.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
#include "ompi/mca/pml/pml.h"
|
||||
#include "ompi/mca/coll/base/base.h"
|
||||
#include "ompi/request/request.h"
|
||||
@ -145,7 +146,7 @@ int ompi_comm_cid_init (void)
|
||||
ompi_comm_world_thread_level_mult = 1;
|
||||
break;
|
||||
}
|
||||
} else if (OMPI_ERR_NOT_IMPLEMENTED == OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
} else if (OMPI_ERR_NOT_IMPLEMENTED == ret) {
|
||||
if (ompi_mpi_thread_multiple) {
|
||||
ompi_comm_world_thread_level_mult = 1;
|
||||
}
|
||||
|
@ -32,6 +32,9 @@
|
||||
#ifdef HAVE_STRING_H
|
||||
#include <string.h>
|
||||
#endif
|
||||
#ifdef HAVE_LIMITS_H
|
||||
#include <limits.h>
|
||||
#endif
|
||||
|
||||
#include "ompi/constants.h"
|
||||
#include "opal/class/opal_pointer_array.h"
|
||||
|
@ -12,6 +12,8 @@
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2010 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2010 Oracle and/or its affiliates. All rights reserved.
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
* Additional copyrights may follow
|
||||
@ -29,7 +31,6 @@
|
||||
#include "ompi/constants.h"
|
||||
#include "opal/class/opal_object.h"
|
||||
#include "opal/class/opal_pointer_array.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
|
||||
#define OMPI_MAX_ERROR_STRING 64
|
||||
|
||||
@ -51,18 +52,13 @@ OMPI_DECLSPEC extern opal_pointer_array_t ompi_errcodes_intern;
|
||||
OMPI_DECLSPEC extern int ompi_errcode_intern_lastused;
|
||||
|
||||
/**
|
||||
* Return the MPI errcode for a given internal error code. This
|
||||
* function guarantees to return a non-OPAL_SOS-encoded error code.
|
||||
*/
|
||||
* Return the MPI errcode for a given internal error code. */
|
||||
static inline int ompi_errcode_get_mpi_code(int errcode)
|
||||
{
|
||||
int ret = MPI_ERR_UNKNOWN;
|
||||
int i;
|
||||
ompi_errcode_intern_t *errc;
|
||||
|
||||
/* Transmogrify, if necessary */
|
||||
errcode = OPAL_SOS_GET_ERROR_CODE(errcode);
|
||||
|
||||
/* If the errcode is >= 0, then it's already an MPI error code, so
|
||||
just return it. */
|
||||
if (errcode >= 0) {
|
||||
|
@ -10,6 +10,8 @@
|
||||
* Copyright (c) 2004-2006 The Regents of the University of California.
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2008 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
* Additional copyrights may follow
|
||||
@ -31,7 +33,6 @@
|
||||
#include "opal/datatype/opal_convertor.h"
|
||||
#include "opal/mca/crs/crs.h"
|
||||
#include "opal/mca/crs/base/base.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
|
||||
#include "ompi/mca/btl/btl.h"
|
||||
|
||||
@ -273,7 +274,7 @@ static inline int mca_bml_base_send( mca_bml_base_btl_t* bml_btl,
|
||||
|
||||
des->des_context = (void*) bml_btl;
|
||||
rc = btl->btl_send(btl, bml_btl->btl_endpoint, des, tag);
|
||||
if(OPAL_SOS_GET_ERROR_CODE(rc) == OMPI_ERR_RESOURCE_BUSY)
|
||||
if (rc == OMPI_ERR_RESOURCE_BUSY)
|
||||
rc = OMPI_SUCCESS;
|
||||
|
||||
return rc;
|
||||
|
@ -10,7 +10,7 @@
|
||||
* University of Stuttgart. All rights reserved.
|
||||
* Copyright (c) 2004-2006 The Regents of the University of California.
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2007 Los Alamos National Security, LLC. All rights
|
||||
* Copyright (c) 2007-2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* Copyright (c) 2008-2009 Cisco Systems, Inc. All rights reserved.
|
||||
* $COPYRIGHT$
|
||||
@ -404,7 +404,7 @@ static int mca_bml_r2_add_procs( size_t nprocs,
|
||||
}
|
||||
|
||||
if (mca_bml_r2.show_unreach_errors &&
|
||||
OMPI_ERR_UNREACH == OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
OMPI_ERR_UNREACH == ret) {
|
||||
orte_show_help("help-mca-bml-r2.txt",
|
||||
"unreachable proc",
|
||||
true,
|
||||
|
@ -9,7 +9,7 @@
|
||||
* University of Stuttgart. All rights reserved.
|
||||
* Copyright (c) 2004-2006 The Regents of the University of California.
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2007 Los Alamos National Security, LLC. All rights
|
||||
* Copyright (c) 2007-2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* Copyright (c) 2008 Cisco Systems, Inc. All rights reserved.
|
||||
* $COPYRIGHT$
|
||||
@ -25,12 +25,16 @@
|
||||
#include <stdlib.h>
|
||||
#include <string.h>
|
||||
|
||||
#include "opal/runtime/opal_progress.h"
|
||||
|
||||
#include "orte/mca/grpcomm/grpcomm.h"
|
||||
#include "orte/util/proc_info.h"
|
||||
|
||||
#include "ompi/runtime/ompi_cr.h"
|
||||
#include "ompi/mca/bml/base/base.h"
|
||||
#include "ompi/mca/btl/base/base.h"
|
||||
#include "ompi/mca/bml/base/bml_base_btl.h"
|
||||
#include "ompi/mca/pml/base/base.h"
|
||||
#include "orte/mca/grpcomm/grpcomm.h"
|
||||
#include "ompi/proc/proc.h"
|
||||
|
||||
#include "bml_r2.h"
|
||||
@ -47,6 +51,7 @@ int mca_bml_r2_ft_event(int state)
|
||||
int loc_state;
|
||||
int param_type = -1;
|
||||
char *param_list = NULL;
|
||||
orte_grpcomm_collective_t coll;
|
||||
|
||||
if(OPAL_CRS_CHECKPOINT == state) {
|
||||
/* Do nothing for now */
|
||||
@ -153,10 +158,15 @@ int mca_bml_r2_ft_event(int state)
|
||||
* Barrier to make all processes have been successfully restarted before
|
||||
* we try to remove some restart only files.
|
||||
*/
|
||||
if (ORTE_SUCCESS != (ret = orte_grpcomm.barrier())) {
|
||||
OBJ_CONSTRUCT(&coll, orte_grpcomm_collective_t);
|
||||
coll.id = orte_process_info.peer_init_barrier;
|
||||
if (OMPI_SUCCESS != (ret = orte_grpcomm.barrier(&coll))) {
|
||||
opal_output(0, "bml:r2: ft_event(Restart): Failed in orte_grpcomm.barrier (%d)", ret);
|
||||
return ret;
|
||||
}
|
||||
while (coll.active) {
|
||||
opal_progress();
|
||||
}
|
||||
|
||||
/*
|
||||
* Re-open the BTL framework to get the full list of components.
|
||||
@ -226,10 +236,15 @@ int mca_bml_r2_ft_event(int state)
|
||||
* Barrier to make all processes have been successfully restarted before
|
||||
* we try to remove some restart only files.
|
||||
*/
|
||||
if (ORTE_SUCCESS != (ret = orte_grpcomm.barrier())) {
|
||||
OBJ_CONSTRUCT(&coll, orte_grpcomm_collective_t);
|
||||
coll.id = orte_process_info.peer_init_barrier;
|
||||
if (OMPI_SUCCESS != (ret = orte_grpcomm.barrier(&coll))) {
|
||||
opal_output(0, "bml:r2: ft_event(Restart): Failed in orte_grpcomm.barrier (%d)", ret);
|
||||
return ret;
|
||||
}
|
||||
while (coll.active) {
|
||||
opal_progress();
|
||||
}
|
||||
|
||||
/*
|
||||
* Re-open the BTL framework to get the full list of components.
|
||||
|
@ -12,6 +12,8 @@
|
||||
* Copyright (c) 2006 Sandia National Laboratories. All rights
|
||||
* reserved.
|
||||
* Copyright (c) 2008-2011 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
* Additional copyrights may follow
|
||||
@ -34,7 +36,6 @@
|
||||
#include "ompi/mca/btl/btl.h"
|
||||
#include "opal/mca/timer/base/base.h"
|
||||
#include "opal/util/argv.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
#include "opal/mca/base/mca_base_param.h"
|
||||
#include "orte/mca/errmgr/errmgr.h"
|
||||
#include "ompi/mca/btl/base/base.h"
|
||||
|
@ -12,7 +12,7 @@
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2007-2010 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2006-2009 Mellanox Technologies. All rights reserved.
|
||||
* Copyright (c) 2006-2007 Los Alamos National Security, LLC. All rights
|
||||
* Copyright (c) 2006-2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* Copyright (c) 2006-2007 Voltaire All rights reserved.
|
||||
* Copyright (c) 2008-2012 Oracle and/or its affiliates. All rights reserved.
|
||||
@ -34,7 +34,6 @@
|
||||
#include "opal/class/opal_bitmap.h"
|
||||
#include "opal/util/output.h"
|
||||
#include "opal/util/arch.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
#include "opal/include/opal_stdint.h"
|
||||
|
||||
#include "ompi/mca/btl/btl.h"
|
||||
@ -303,7 +302,7 @@ static int create_srq(mca_btl_openib_module_t *openib_btl)
|
||||
|
||||
/* Check if our device supports modify srq ability */
|
||||
rc = check_if_device_support_modify_srq(openib_btl);
|
||||
if(OMPI_ERR_NOT_SUPPORTED == OPAL_SOS_GET_ERROR_CODE(rc)) {
|
||||
if(OMPI_ERR_NOT_SUPPORTED == rc) {
|
||||
device_support_modify_srq = false;
|
||||
} else if(OMPI_SUCCESS != rc) {
|
||||
mca_btl_openib_show_init_error(__FILE__, __LINE__,
|
||||
@ -494,7 +493,7 @@ static int mca_btl_openib_tune_endpoint(mca_btl_openib_module_t* openib_btl,
|
||||
endpoint->rem_info.rem_vendor_part_id, &values);
|
||||
|
||||
if (OMPI_SUCCESS != ret &&
|
||||
OMPI_ERR_NOT_FOUND != OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
OMPI_ERR_NOT_FOUND != ret) {
|
||||
orte_show_help("help-mpi-btl-openib.txt",
|
||||
"error in device init", true,
|
||||
orte_process_info.nodename,
|
||||
@ -1625,7 +1624,7 @@ int mca_btl_openib_put( mca_btl_base_module_t* btl,
|
||||
OPAL_THREAD_LOCK(&ep->endpoint_lock);
|
||||
rc = check_endpoint_state(ep, descriptor, &ep->pending_put_frags);
|
||||
OPAL_THREAD_UNLOCK(&ep->endpoint_lock);
|
||||
if(OMPI_ERR_RESOURCE_BUSY == OPAL_SOS_GET_ERROR_CODE(rc))
|
||||
if(OMPI_ERR_RESOURCE_BUSY == rc)
|
||||
return OMPI_SUCCESS;
|
||||
if(OMPI_SUCCESS != rc)
|
||||
return rc;
|
||||
@ -1696,7 +1695,7 @@ int mca_btl_openib_get(mca_btl_base_module_t* btl,
|
||||
OPAL_THREAD_LOCK(&ep->endpoint_lock);
|
||||
rc = check_endpoint_state(ep, descriptor, &ep->pending_get_frags);
|
||||
OPAL_THREAD_UNLOCK(&ep->endpoint_lock);
|
||||
if(OMPI_ERR_RESOURCE_BUSY == OPAL_SOS_GET_ERROR_CODE(rc))
|
||||
if(OMPI_ERR_RESOURCE_BUSY == rc)
|
||||
return OMPI_SUCCESS;
|
||||
if(OMPI_SUCCESS != rc)
|
||||
return rc;
|
||||
|
@ -12,7 +12,7 @@
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2006-2011 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2006-2009 Mellanox Technologies. All rights reserved.
|
||||
* Copyright (c) 2006-2007 Los Alamos National Security, LLC. All rights
|
||||
* Copyright (c) 2006-2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* Copyright (c) 2006-2007 Voltaire All rights reserved.
|
||||
* Copyright (c) 2009-2012 Oracle and/or its affiliates. All rights reserved.
|
||||
@ -1125,8 +1125,8 @@ static int prepare_device_for_use(mca_btl_openib_device_t *device)
|
||||
if (OMPI_SUCCESS != rc) {
|
||||
/* If we're "out of memory", this usually means that we ran
|
||||
out of registered memory, so show that error message */
|
||||
if (OMPI_ERR_OUT_OF_RESOURCE == OPAL_SOS_GET_ERROR_CODE(rc) ||
|
||||
OMPI_ERR_TEMP_OUT_OF_RESOURCE == OPAL_SOS_GET_ERROR_CODE(rc)) {
|
||||
if (OMPI_ERR_OUT_OF_RESOURCE == rc ||
|
||||
OMPI_ERR_TEMP_OUT_OF_RESOURCE == rc) {
|
||||
errno = ENOMEM;
|
||||
mca_btl_openib_show_init_error(__FILE__, __LINE__,
|
||||
"ompi_free_list_init_ex_new",
|
||||
@ -1161,8 +1161,8 @@ static int prepare_device_for_use(mca_btl_openib_device_t *device)
|
||||
/* If we're "out of memory", this usually means that we
|
||||
ran out of registered memory, so show that error
|
||||
message */
|
||||
if (OMPI_ERR_OUT_OF_RESOURCE == OPAL_SOS_GET_ERROR_CODE(rc) ||
|
||||
OMPI_ERR_TEMP_OUT_OF_RESOURCE == OPAL_SOS_GET_ERROR_CODE(rc)) {
|
||||
if (OMPI_ERR_OUT_OF_RESOURCE == rc ||
|
||||
OMPI_ERR_TEMP_OUT_OF_RESOURCE == rc) {
|
||||
errno = ENOMEM;
|
||||
mca_btl_openib_show_init_error(__FILE__, __LINE__,
|
||||
"ompi_free_list_init_ex_new",
|
||||
@ -1658,11 +1658,11 @@ static int init_one_device(opal_list_t *btl_list, struct ibv_device* ib_dev)
|
||||
device->ib_dev_attr.vendor_part_id,
|
||||
&values);
|
||||
if (OMPI_SUCCESS != ret &&
|
||||
OMPI_ERR_NOT_FOUND != OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
OMPI_ERR_NOT_FOUND != ret) {
|
||||
/* If we get a serious error, propagate it upwards */
|
||||
goto error;
|
||||
}
|
||||
if (OMPI_ERR_NOT_FOUND == OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
if (OMPI_ERR_NOT_FOUND == ret) {
|
||||
/* If we didn't find a matching device in the INI files, output a
|
||||
warning that we're using default values (unless overridden
|
||||
that we don't want to see these warnings) */
|
||||
@ -1679,7 +1679,7 @@ static int init_one_device(opal_list_t *btl_list, struct ibv_device* ib_dev)
|
||||
be set indicating that it does not have good values */
|
||||
ret = ompi_btl_openib_ini_query(0, 0, &default_values);
|
||||
if (OMPI_SUCCESS != ret &&
|
||||
OMPI_ERR_NOT_FOUND != OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
OMPI_ERR_NOT_FOUND != ret) {
|
||||
/* If we get a serious error, propagate it upwards */
|
||||
goto error;
|
||||
}
|
||||
@ -1841,7 +1841,7 @@ static int init_one_device(opal_list_t *btl_list, struct ibv_device* ib_dev)
|
||||
device, &mpool_resources);
|
||||
if(NULL == device->mpool){
|
||||
/* Don't print an error message here -- we'll get one from
|
||||
mpool_create anyway (OPAL_SOS would be good here...) */
|
||||
mpool_create anyway */
|
||||
goto error;
|
||||
}
|
||||
|
||||
@ -1899,7 +1899,7 @@ static int init_one_device(opal_list_t *btl_list, struct ibv_device* ib_dev)
|
||||
if (OMPI_SUCCESS != ret) {
|
||||
/* Out of bounds error indicates that we hit max btl number
|
||||
* don't propagate the error to the caller */
|
||||
if (OMPI_ERR_VALUE_OUT_OF_BOUNDS == OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
if (OMPI_ERR_VALUE_OUT_OF_BOUNDS == ret) {
|
||||
ret = OMPI_SUCCESS;
|
||||
}
|
||||
break;
|
||||
@ -2830,7 +2830,7 @@ btl_openib_component_init(int *num_btl_modules,
|
||||
/* If we get NOT_SUPPORTED, then no CPC was found for this
|
||||
port. But that's not a fatal error -- just keep going;
|
||||
let's see if we find any usable openib modules or not. */
|
||||
if (OMPI_ERR_NOT_SUPPORTED == OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
if (OMPI_ERR_NOT_SUPPORTED == ret) {
|
||||
continue;
|
||||
} else if (OMPI_SUCCESS != ret) {
|
||||
/* All others *are* fatal. Note that we already did a
|
||||
@ -2994,7 +2994,7 @@ static int progress_no_credits_pending_frags(mca_btl_base_endpoint_t *ep)
|
||||
error upward. */
|
||||
rc = mca_btl_openib_endpoint_post_send(ep, to_send_frag(frag));
|
||||
if (OPAL_UNLIKELY(OMPI_SUCCESS != rc &&
|
||||
OMPI_ERR_RESOURCE_BUSY != OPAL_SOS_GET_ERROR_CODE(rc))) {
|
||||
OMPI_ERR_RESOURCE_BUSY != rc)) {
|
||||
OPAL_THREAD_UNLOCK(&ep->endpoint_lock);
|
||||
return rc;
|
||||
}
|
||||
@ -3023,7 +3023,7 @@ void mca_btl_openib_frag_progress_pending_put_get(mca_btl_base_endpoint_t *ep,
|
||||
break;
|
||||
rc = mca_btl_openib_get((mca_btl_base_module_t *)openib_btl, ep,
|
||||
&to_base_frag(frag)->base);
|
||||
if(OMPI_ERR_OUT_OF_RESOURCE == OPAL_SOS_GET_ERROR_CODE(rc))
|
||||
if(OMPI_ERR_OUT_OF_RESOURCE == rc)
|
||||
break;
|
||||
}
|
||||
|
||||
@ -3036,7 +3036,7 @@ void mca_btl_openib_frag_progress_pending_put_get(mca_btl_base_endpoint_t *ep,
|
||||
break;
|
||||
rc = mca_btl_openib_put((mca_btl_base_module_t*)openib_btl, ep,
|
||||
&to_base_frag(frag)->base);
|
||||
if(OMPI_ERR_OUT_OF_RESOURCE == OPAL_SOS_GET_ERROR_CODE(rc))
|
||||
if(OMPI_ERR_OUT_OF_RESOURCE == rc)
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
@ -11,7 +11,7 @@
|
||||
* Copyright (c) 2004-2005 The Regents of the University of California.
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2006-2009 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2006-2007 Los Alamos National Security, LLC. All rights
|
||||
* Copyright (c) 2006-2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* Copyright (c) 2006-2007 Voltaire All rights reserved.
|
||||
* Copyright (c) 2006-2009 Mellanox Technologies, Inc. All rights reserved.
|
||||
@ -36,7 +36,6 @@
|
||||
|
||||
#include "opal_stdint.h"
|
||||
#include "opal/util/output.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
|
||||
#include "orte/util/show_help.h"
|
||||
|
||||
@ -714,7 +713,7 @@ int mca_btl_openib_endpoint_send(mca_btl_base_endpoint_t* ep,
|
||||
rc = mca_btl_openib_endpoint_post_send(ep, frag);
|
||||
}
|
||||
OPAL_THREAD_UNLOCK(&ep->endpoint_lock);
|
||||
if (OPAL_UNLIKELY(OMPI_ERR_RESOURCE_BUSY == OPAL_SOS_GET_ERROR_CODE(rc))) {
|
||||
if (OPAL_UNLIKELY(OMPI_ERR_RESOURCE_BUSY == rc)) {
|
||||
rc = OMPI_SUCCESS;
|
||||
}
|
||||
|
||||
@ -898,7 +897,7 @@ static int mca_btl_openib_endpoint_send_eager_rdma(
|
||||
));
|
||||
}
|
||||
rc = mca_btl_openib_endpoint_send(endpoint, frag);
|
||||
if (OMPI_SUCCESS == rc || OMPI_ERR_RESOURCE_BUSY == OPAL_SOS_GET_ERROR_CODE(rc))
|
||||
if (OMPI_SUCCESS == rc || OMPI_ERR_RESOURCE_BUSY == rc)
|
||||
return OMPI_SUCCESS;
|
||||
|
||||
MCA_BTL_IB_FRAG_RETURN(frag);
|
||||
|
@ -1,6 +1,8 @@
|
||||
/*
|
||||
* Copyright (c) 2010-2011 Oracle and/or its affiliates. All rights reserved.
|
||||
* Copyright (c) 2011 NVIDIA Corporation. All rights reserved.
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
* Additional copyrights may follow
|
||||
@ -28,8 +30,6 @@
|
||||
#include "btl_openib_proc.h"
|
||||
#include "btl_openib_failover.h"
|
||||
|
||||
#include "opal/util/opal_sos.h"
|
||||
|
||||
static void error_out_all_pending_frags(mca_btl_base_endpoint_t *ep,
|
||||
struct mca_btl_base_module_t* module,
|
||||
bool errout);
|
||||
@ -691,7 +691,7 @@ static void mca_btl_openib_endpoint_notify(mca_btl_base_endpoint_t* endpoint, ui
|
||||
BTL_OPENIB_BROKEN_CONNECTION_HEADER_HTON((*bc_hdr));
|
||||
}
|
||||
rc = mca_btl_openib_endpoint_send(newep, frag);
|
||||
if (OMPI_SUCCESS == rc || OMPI_ERR_RESOURCE_BUSY == OPAL_SOS_GET_ERROR_CODE(rc)) {
|
||||
if (OMPI_SUCCESS == rc || OMPI_ERR_RESOURCE_BUSY == rc) {
|
||||
return;
|
||||
}
|
||||
|
||||
|
@ -11,6 +11,8 @@
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2006-2009 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2008 Mellanox Technologies. All rights reserved.
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
* Additional copyrights may follow
|
||||
@ -29,7 +31,6 @@
|
||||
#endif
|
||||
|
||||
#include "orte/util/show_help.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
#include "opal/mca/base/mca_base_param.h"
|
||||
|
||||
#include "btl_openib.h"
|
||||
@ -133,13 +134,13 @@ int ompi_btl_openib_ini_init(void)
|
||||
/* Note that NOT_FOUND and SUCCESS are not fatal errors
|
||||
and we keep going. Other errors are treated as
|
||||
fatal */
|
||||
if (OMPI_ERR_NOT_FOUND != OPAL_SOS_GET_ERROR_CODE(ret) && OMPI_SUCCESS != ret) {
|
||||
if (OMPI_ERR_NOT_FOUND != ret && OMPI_SUCCESS != ret) {
|
||||
break;
|
||||
}
|
||||
str = colon + 1;
|
||||
}
|
||||
/* Parse the last file if we didn't have a fatal error above */
|
||||
if (OMPI_ERR_NOT_FOUND != OPAL_SOS_GET_ERROR_CODE(ret) && OMPI_SUCCESS != ret) {
|
||||
if (OMPI_ERR_NOT_FOUND != ret && OMPI_SUCCESS != ret) {
|
||||
ret = parse_file(str);
|
||||
}
|
||||
|
||||
@ -150,7 +151,7 @@ int ompi_btl_openib_ini_init(void)
|
||||
/* Return SUCCESS unless we got a fatal error */
|
||||
|
||||
initialized = true;
|
||||
return (OMPI_SUCCESS == ret || OMPI_ERR_NOT_FOUND == OPAL_SOS_GET_ERROR_CODE(ret)) ?
|
||||
return (OMPI_SUCCESS == ret || OMPI_ERR_NOT_FOUND == ret) ?
|
||||
OMPI_SUCCESS : ret;
|
||||
}
|
||||
|
||||
|
@ -1,6 +1,8 @@
|
||||
/*
|
||||
* Copyright (c) 2007-2009 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2007 Mellanox Technologies, Inc. All rights reserved.
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
*
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
@ -28,7 +30,6 @@
|
||||
#include "orte/util/show_help.h"
|
||||
#include "opal/util/argv.h"
|
||||
#include "opal/util/output.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
|
||||
/*
|
||||
* Array of all possible connection functions
|
||||
@ -219,7 +220,7 @@ int ompi_btl_openib_connect_base_init(void)
|
||||
opal_output(-1, "found available cpc (SUCCESS init): %s",
|
||||
all[i]->cbc_name);
|
||||
continue;
|
||||
} else if (OMPI_ERR_NOT_SUPPORTED == OPAL_SOS_GET_ERROR_CODE(rc)) {
|
||||
} else if (OMPI_ERR_NOT_SUPPORTED == rc) {
|
||||
continue;
|
||||
} else {
|
||||
return rc;
|
||||
@ -265,8 +266,7 @@ int ompi_btl_openib_connect_base_select_for_local_port(mca_btl_openib_module_t *
|
||||
strcat(msg, available[i]->cbc_name);
|
||||
|
||||
rc = available[i]->cbc_query(btl, &cpcs[cpc_index]);
|
||||
if (OMPI_ERR_NOT_SUPPORTED == OPAL_SOS_GET_ERROR_CODE(rc) ||
|
||||
OMPI_ERR_UNREACH == OPAL_SOS_GET_ERROR_CODE(rc)) {
|
||||
if (OMPI_ERR_NOT_SUPPORTED == rc || OMPI_ERR_UNREACH == rc) {
|
||||
continue;
|
||||
} else if (OMPI_SUCCESS != rc) {
|
||||
free(cpcs);
|
||||
|
@ -10,7 +10,7 @@
|
||||
* Copyright (c) 2004-2005 The Regents of the University of California.
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2006-2009 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2006 Los Alamos National Security, LLC. All rights
|
||||
* Copyright (c) 2006-2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* Copyright (c) 2008-2011 Mellanox Technologies. All rights reserved.
|
||||
* Copyright (c) 2009-2011 IBM Corporation. All rights reserved.
|
||||
@ -30,7 +30,6 @@
|
||||
#include "orte/util/show_help.h"
|
||||
#include "opal/util/error.h"
|
||||
#include "opal/util/output.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
#include "orte/mca/rml/rml.h"
|
||||
#include "orte/mca/rml/rml_types.h"
|
||||
#include "orte/mca/errmgr/errmgr.h"
|
||||
|
@ -4,6 +4,8 @@
|
||||
* Copyright (c) 2008 Mellanox Technologies. All rights reserved.
|
||||
* Copyright (c) 2009 Sandia National Laboratories. All rights reserved.
|
||||
* Copyright (c) 2010 Oracle and/or its affiliates. All rights reserved.
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
*
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
@ -46,7 +48,6 @@
|
||||
|
||||
#include "opal/util/output.h"
|
||||
#include "opal/util/error.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
#include "orte/util/show_help.h"
|
||||
|
||||
#include "btl_openib_fd.h"
|
||||
@ -1932,7 +1933,7 @@ out3:
|
||||
out1:
|
||||
free(*cpc);
|
||||
out:
|
||||
if (OMPI_ERR_NOT_SUPPORTED == OPAL_SOS_GET_ERROR_CODE(rc)) {
|
||||
if (OMPI_ERR_NOT_SUPPORTED == rc) {
|
||||
opal_output_verbose(5, mca_btl_base_output,
|
||||
"openib BTL: rdmacm CPC unavailable for use on %s:%d; skipped",
|
||||
ibv_get_device_name(openib_btl->device->ib_dev),
|
||||
|
@ -5,6 +5,8 @@
|
||||
* Copyright (c) 2010-2011 The University of Tennessee and The University
|
||||
* of Tennessee Research Foundation. All rights
|
||||
* reserved.
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
*
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
@ -21,7 +23,6 @@
|
||||
#include "opal/util/output.h"
|
||||
#include "orte/util/show_help.h"
|
||||
#include "orte/util/name_fns.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
#include "orte/mca/rml/rml.h"
|
||||
#include "orte/mca/rml/rml_types.h"
|
||||
#include "orte/mca/errmgr/errmgr.h"
|
||||
@ -698,10 +699,8 @@ static mca_btl_openib_endpoint_t* xoob_find_endpoint(orte_process_name_t* proces
|
||||
|
||||
BTL_VERBOSE(("Searching for ep and proc with follow parameters:"
|
||||
"jobid %d, vpid %d, "
|
||||
"epoch %d, "
|
||||
"sid %" PRIx64 ", lid %d",
|
||||
process_name->jobid, process_name->vpid,
|
||||
ORTE_EPOCH_GET(process_name),
|
||||
subnet_id, lid));
|
||||
|
||||
|
||||
|
@ -10,6 +10,8 @@
|
||||
* Copyright (c) 2004-2005 The Regents of the University of California.
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2008 UT-Battelle, LLC. All rights reserved.
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
* Additional copyrights may follow
|
||||
@ -29,7 +31,6 @@
|
||||
#include "ompi/constants.h"
|
||||
#include "ompi/mca/btl/btl.h"
|
||||
#include "opal/datatype/opal_convertor.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
|
||||
#include "btl_portals.h"
|
||||
#include "btl_portals_endpoint.h"
|
||||
|
@ -9,6 +9,8 @@
|
||||
* University of Stuttgart. All rights reserved.
|
||||
* Copyright (c) 2004-2005 The Regents of the University of California.
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2011-2012 Los Alamos National Security, LLC.
|
||||
* All rights reserved.
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
* Additional copyrights may follow
|
||||
@ -67,7 +69,7 @@ OBJ_CLASS_DECLARATION(mca_btl_portals_frag_recv_t);
|
||||
ompi_free_list_item_t *item; \
|
||||
OMPI_FREE_LIST_GET(&((mca_btl_portals_module_t*)btl_macro)->portals_frag_eager, item, rc); \
|
||||
frag = (mca_btl_portals_frag_t*) item; \
|
||||
if (OPAL_SOS_GET_ERROR_CODE(rc) == OMPI_ERR_TEMP_OUT_OF_RESOURCE) { \
|
||||
if (rc == OMPI_ERR_TEMP_OUT_OF_RESOURCE) { \
|
||||
OMPI_BTL_PORTALS_FRAG_ALLOC_MAX(btl_macro, frag, rc); \
|
||||
} \
|
||||
}
|
||||
|
@ -10,6 +10,8 @@
|
||||
* Copyright (c) 2004-2005 The Regents of the University of California.
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2008 UT-Battelle, LLC. All rights reserved.
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
* Additional copyrights may follow
|
||||
@ -24,7 +26,6 @@
|
||||
|
||||
#include "ompi/constants.h"
|
||||
#include "opal/datatype/opal_convertor.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
|
||||
#include "btl_portals.h"
|
||||
#include "btl_portals_send.h"
|
||||
|
@ -12,6 +12,8 @@
|
||||
* Copyright (c) 2007-2011 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2008 Sun Microsystems, Inc. All rights reserved.
|
||||
* Copyright (c) 2009 Oak Ridge National Laboratory
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
* Additional copyrights may follow
|
||||
@ -53,7 +55,6 @@
|
||||
#include "opal/util/output.h"
|
||||
#include "opal/util/argv.h"
|
||||
#include "opal/util/net.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
#include "opal/mca/base/mca_base_param.h"
|
||||
|
||||
#include "orte/types.h"
|
||||
@ -1055,7 +1056,7 @@ mca_btl_base_module_t** mca_btl_tcp_component_init(int *num_btl_modules,
|
||||
}
|
||||
#if OPAL_WANT_IPV6
|
||||
if((ret = mca_btl_tcp_component_create_listen(AF_INET6)) != OMPI_SUCCESS) {
|
||||
if (!(OMPI_ERR_IN_ERRNO == OPAL_SOS_GET_ERROR_CODE(ret) &&
|
||||
if (!(OMPI_ERR_IN_ERRNO == ret &&
|
||||
EAFNOSUPPORT == opal_socket_errno)) {
|
||||
opal_output (0, "mca_btl_tcp_component: IPv6 listening socket failed\n");
|
||||
return 0;
|
||||
|
@ -13,7 +13,6 @@
|
||||
#include "opal/include/opal_stdint.h"
|
||||
|
||||
#include "btl_ugni_rdma.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
|
||||
/**
|
||||
* Initiate a put operation.
|
||||
|
@ -12,7 +12,7 @@
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2007-2010 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2006-2009 Mellanox Technologies. All rights reserved.
|
||||
* Copyright (c) 2006-2007 Los Alamos National Security, LLC. All rights
|
||||
* Copyright (c) 2006-2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* Copyright (c) 2006-2007 Voltaire All rights reserved.
|
||||
* Copyright (c) 2008-2010 Oracle and/or its affiliates. All rights reserved.
|
||||
@ -32,7 +32,6 @@
|
||||
#include "opal/class/opal_bitmap.h"
|
||||
#include "opal/util/output.h"
|
||||
#include "opal/util/arch.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
|
||||
#include "ompi/mca/btl/btl.h"
|
||||
#include "ompi/mca/btl/base/btl_base_error.h"
|
||||
@ -309,7 +308,7 @@ static int mca_btl_wv_tune_endpoint(mca_btl_wv_module_t* wv_btl,
|
||||
endpoint->rem_info.rem_vendor_part_id, &values);
|
||||
|
||||
if (OMPI_SUCCESS != ret &&
|
||||
OMPI_ERR_NOT_FOUND != OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
OMPI_ERR_NOT_FOUND != ret) {
|
||||
orte_show_help("help-mpi-btl-wv.txt",
|
||||
"error in device init", true,
|
||||
orte_process_info.nodename,
|
||||
@ -1347,7 +1346,7 @@ int mca_btl_wv_put(mca_btl_base_module_t* btl,
|
||||
OPAL_THREAD_LOCK(&ep->endpoint_lock);
|
||||
rc = check_endpoint_state(ep, descriptor, &ep->pending_put_frags);
|
||||
OPAL_THREAD_UNLOCK(&ep->endpoint_lock);
|
||||
if(OMPI_ERR_RESOURCE_BUSY == OPAL_SOS_GET_ERROR_CODE(rc))
|
||||
if(OMPI_ERR_RESOURCE_BUSY == rc)
|
||||
return OMPI_SUCCESS;
|
||||
if(OMPI_SUCCESS != rc)
|
||||
return rc;
|
||||
@ -1406,7 +1405,7 @@ int mca_btl_wv_get(mca_btl_base_module_t* btl,
|
||||
OPAL_THREAD_LOCK(&ep->endpoint_lock);
|
||||
rc = check_endpoint_state(ep, descriptor, &ep->pending_get_frags);
|
||||
OPAL_THREAD_UNLOCK(&ep->endpoint_lock);
|
||||
if(OMPI_ERR_RESOURCE_BUSY == OPAL_SOS_GET_ERROR_CODE(rc))
|
||||
if(OMPI_ERR_RESOURCE_BUSY == rc)
|
||||
return OMPI_SUCCESS;
|
||||
if(OMPI_SUCCESS != rc)
|
||||
return rc;
|
||||
|
@ -12,7 +12,7 @@
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2006-2011 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2006-2009 Mellanox Technologies. All rights reserved.
|
||||
* Copyright (c) 2006-2007 Los Alamos National Security, LLC. All rights
|
||||
* Copyright (c) 2006-2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* Copyright (c) 2006-2007 Voltaire All rights reserved.
|
||||
* Copyright (c) 2009-2010 Oracle and/or its affiliates. All rights reserved.
|
||||
@ -784,8 +784,8 @@ static int prepare_device_for_use(mca_btl_wv_device_t *device)
|
||||
if (OMPI_SUCCESS != rc) {
|
||||
/* If we're "out of memory", this usually means that we ran
|
||||
out of registered memory, so show that error message */
|
||||
if (OMPI_ERR_OUT_OF_RESOURCE == OPAL_SOS_GET_ERROR_CODE(rc) ||
|
||||
OMPI_ERR_TEMP_OUT_OF_RESOURCE == OPAL_SOS_GET_ERROR_CODE(rc)) {
|
||||
if (OMPI_ERR_OUT_OF_RESOURCE == rc ||
|
||||
OMPI_ERR_TEMP_OUT_OF_RESOURCE == rc) {
|
||||
errno = ENOMEM;
|
||||
mca_btl_wv_show_init_error(__FILE__, __LINE__,
|
||||
"ompi_free_list_init_ex_new",
|
||||
@ -820,8 +820,8 @@ static int prepare_device_for_use(mca_btl_wv_device_t *device)
|
||||
/* If we're "out of memory", this usually means that we
|
||||
ran out of registered memory, so show that error
|
||||
message */
|
||||
if (OMPI_ERR_OUT_OF_RESOURCE == OPAL_SOS_GET_ERROR_CODE(rc) ||
|
||||
OMPI_ERR_TEMP_OUT_OF_RESOURCE == OPAL_SOS_GET_ERROR_CODE(rc)) {
|
||||
if (OMPI_ERR_OUT_OF_RESOURCE == rc ||
|
||||
OMPI_ERR_TEMP_OUT_OF_RESOURCE == rc) {
|
||||
errno = ENOMEM;
|
||||
mca_btl_wv_show_init_error(__FILE__, __LINE__,
|
||||
"ompi_free_list_init_ex_new",
|
||||
@ -1312,11 +1312,11 @@ static int init_one_device(opal_list_t *btl_list, struct wv_device* ib_dev)
|
||||
device->ib_dev_attr.VendorPartId,
|
||||
&values);
|
||||
if (OMPI_SUCCESS != ret &&
|
||||
OMPI_ERR_NOT_FOUND != OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
OMPI_ERR_NOT_FOUND != ret) {
|
||||
/* If we get a serious error, propagate it upwards */
|
||||
goto error;
|
||||
}
|
||||
if (OMPI_ERR_NOT_FOUND == OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
if (OMPI_ERR_NOT_FOUND == ret) {
|
||||
/* If we didn't find a matching device in the INI files, output a
|
||||
warning that we're using default values (unless overridden
|
||||
that we don't want to see these warnings) */
|
||||
@ -1333,7 +1333,7 @@ static int init_one_device(opal_list_t *btl_list, struct wv_device* ib_dev)
|
||||
be set indicating that it does not have good values */
|
||||
ret = ompi_btl_wv_ini_query(0, 0, &default_values);
|
||||
if (OMPI_SUCCESS != ret &&
|
||||
OMPI_ERR_NOT_FOUND != OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
OMPI_ERR_NOT_FOUND != ret) {
|
||||
/* If we get a serious error, propagate it upwards */
|
||||
goto error;
|
||||
}
|
||||
@ -1429,7 +1429,7 @@ static int init_one_device(opal_list_t *btl_list, struct wv_device* ib_dev)
|
||||
device, &mpool_resources);
|
||||
if(NULL == device->mpool){
|
||||
/* Don't print an error message here -- we'll get one from
|
||||
mpool_create anyway (OPAL_SOS would be good here...) */
|
||||
mpool_create anyway */
|
||||
goto error;
|
||||
}
|
||||
|
||||
@ -1481,7 +1481,7 @@ static int init_one_device(opal_list_t *btl_list, struct wv_device* ib_dev)
|
||||
if (OMPI_SUCCESS != ret) {
|
||||
/* Out of bounds error indicates that we hit max btl number
|
||||
* don't propagate the error to the caller */
|
||||
if (OMPI_ERR_VALUE_OUT_OF_BOUNDS == OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
if (OMPI_ERR_VALUE_OUT_OF_BOUNDS == ret) {
|
||||
ret = OMPI_SUCCESS;
|
||||
}
|
||||
break;
|
||||
@ -2313,7 +2313,7 @@ btl_wv_component_init(int *num_btl_modules,
|
||||
/* If we get NOT_SUPPORTED, then no CPC was found for this
|
||||
port. But that's not a fatal error -- just keep going;
|
||||
let's see if we find any usable wv modules or not. */
|
||||
if (OMPI_ERR_NOT_SUPPORTED == OPAL_SOS_GET_ERROR_CODE(ret)) {
|
||||
if (OMPI_ERR_NOT_SUPPORTED == ret) {
|
||||
continue;
|
||||
} else if (OMPI_SUCCESS != ret) {
|
||||
/* All others *are* fatal. Note that we already did a
|
||||
@ -2469,7 +2469,7 @@ static int progress_no_credits_pending_frags(mca_btl_base_endpoint_t *ep)
|
||||
error upward. */
|
||||
rc = mca_btl_wv_endpoint_post_send(ep, to_send_frag(frag));
|
||||
if (OPAL_UNLIKELY(OMPI_SUCCESS != rc &&
|
||||
OMPI_ERR_RESOURCE_BUSY != OPAL_SOS_GET_ERROR_CODE(rc))) {
|
||||
OMPI_ERR_RESOURCE_BUSY != rc)) {
|
||||
OPAL_THREAD_UNLOCK(&ep->endpoint_lock);
|
||||
return rc;
|
||||
}
|
||||
@ -2497,7 +2497,7 @@ void mca_btl_wv_frag_progress_pending_put_get(mca_btl_base_endpoint_t *ep,
|
||||
break;
|
||||
rc = mca_btl_wv_get((mca_btl_base_module_t *)wv_btl, ep,
|
||||
&to_base_frag(frag)->base);
|
||||
if(OMPI_ERR_OUT_OF_RESOURCE == OPAL_SOS_GET_ERROR_CODE(rc))
|
||||
if(OMPI_ERR_OUT_OF_RESOURCE == rc)
|
||||
break;
|
||||
}
|
||||
|
||||
@ -2510,7 +2510,7 @@ void mca_btl_wv_frag_progress_pending_put_get(mca_btl_base_endpoint_t *ep,
|
||||
break;
|
||||
rc = mca_btl_wv_put((mca_btl_base_module_t *)wv_btl, ep,
|
||||
&to_base_frag(frag)->base);
|
||||
if(OMPI_ERR_OUT_OF_RESOURCE == OPAL_SOS_GET_ERROR_CODE(rc))
|
||||
if(OMPI_ERR_OUT_OF_RESOURCE == rc)
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
@ -11,7 +11,7 @@
|
||||
* Copyright (c) 2004-2005 The Regents of the University of California.
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2006-2009 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2006-2007 Los Alamos National Security, LLC. All rights
|
||||
* Copyright (c) 2006-2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* Copyright (c) 2006-2007 Voltaire All rights reserved.
|
||||
* Copyright (c) 2006-2009 Mellanox Technologies, Inc. All rights reserved.
|
||||
@ -33,7 +33,6 @@
|
||||
|
||||
#include "opal_stdint.h"
|
||||
#include "opal/util/output.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
|
||||
#include "orte/util/show_help.h"
|
||||
|
||||
@ -617,7 +616,7 @@ int mca_btl_wv_endpoint_send(mca_btl_base_endpoint_t* ep,
|
||||
rc = mca_btl_wv_endpoint_post_send(ep, frag);
|
||||
}
|
||||
OPAL_THREAD_UNLOCK(&ep->endpoint_lock);
|
||||
if (OPAL_UNLIKELY(OMPI_ERR_RESOURCE_BUSY == OPAL_SOS_GET_ERROR_CODE(rc))) {
|
||||
if (OPAL_UNLIKELY(OMPI_ERR_RESOURCE_BUSY == rc)) {
|
||||
rc = OMPI_SUCCESS;
|
||||
}
|
||||
|
||||
@ -801,7 +800,7 @@ static int mca_btl_wv_endpoint_send_eager_rdma(
|
||||
));
|
||||
}
|
||||
rc = mca_btl_wv_endpoint_send(endpoint, frag);
|
||||
if (OMPI_SUCCESS == rc || OMPI_ERR_RESOURCE_BUSY == OPAL_SOS_GET_ERROR_CODE(rc))
|
||||
if (OMPI_SUCCESS == rc || OMPI_ERR_RESOURCE_BUSY == rc)
|
||||
return OMPI_SUCCESS;
|
||||
|
||||
MCA_BTL_IB_FRAG_RETURN(frag);
|
||||
|
@ -11,6 +11,8 @@
|
||||
* All rights reserved.
|
||||
* Copyright (c) 2006-2009 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2008 Mellanox Technologies. All rights reserved.
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
* $COPYRIGHT$
|
||||
*
|
||||
* Additional copyrights may follow
|
||||
@ -29,7 +31,6 @@
|
||||
#endif
|
||||
|
||||
#include "orte/util/show_help.h"
|
||||
#include "opal/util/opal_sos.h"
|
||||
#include "opal/mca/base/mca_base_param.h"
|
||||
|
||||
#include "btl_wv.h"
|
||||
@ -127,13 +128,13 @@ int ompi_btl_wv_ini_init(void)
|
||||
/* Note that NOT_FOUND and SUCCESS are not fatal errors
|
||||
and we keep going. Other errors are treated as
|
||||
fatal */
|
||||
if (OMPI_ERR_NOT_FOUND != OPAL_SOS_GET_ERROR_CODE(ret) && OMPI_SUCCESS != ret) {
|
||||
if (OMPI_ERR_NOT_FOUND != ret && OMPI_SUCCESS != ret) {
|
||||
break;
|
||||
}
|
||||
str = colon + 1;
|
||||
}
|
||||
/* Parse the last file if we didn't have a fatal error above */
|
||||
if (OMPI_ERR_NOT_FOUND != OPAL_SOS_GET_ERROR_CODE(ret) && OMPI_SUCCESS != ret) {
|
||||
if (OMPI_ERR_NOT_FOUND != ret && OMPI_SUCCESS != ret) {
|
||||
ret = parse_file(str);
|
||||
}
|
||||
|
||||
@ -144,7 +145,7 @@ int ompi_btl_wv_ini_init(void)
|
||||
/* Return SUCCESS unless we got a fatal error */
|
||||
|
||||
initialized = true;
|
||||
return (OMPI_SUCCESS == ret || OMPI_ERR_NOT_FOUND == OPAL_SOS_GET_ERROR_CODE(ret)) ?
|
||||
return (OMPI_SUCCESS == ret || OMPI_ERR_NOT_FOUND == ret) ?
|
||||
OMPI_SUCCESS : ret;
|
||||
}
|
||||
|
||||
|
@ -1,6 +1,8 @@
|
||||
/*
|
||||
* Copyright (c) 2007-2009 Cisco Systems, Inc. All rights reserved.
|
||||
* Copyright (c) 2007 Mellanox Technologies, Inc. All rights reserved.
|
||||
* Copyright (c) 2012 Los Alamos National Security, LLC. All rights
|
||||
* reserved.
|
||||
*
|
||||
* $COPYRIGHT$
|
||||
*
|
||||