aec5cd08bd
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.
129 строки
3.1 KiB
C
129 строки
3.1 KiB
C
/*
|
|
* Copyright (c) 2011 Cisco Systems, Inc. All rights reserved.
|
|
* Copyright (c) 2011 Los Alamos National Security, LLC.
|
|
* All rights reserved.
|
|
* Copyright (c) 2014 Intel, Inc. All rights reserved.
|
|
* $COPYRIGHT$
|
|
*
|
|
* Additional copyrights may follow
|
|
*
|
|
* $HEADER$
|
|
*/
|
|
|
|
#include "ompi_config.h"
|
|
#include "ompi/constants.h"
|
|
|
|
|
|
#include "opal/mca/pmix/pmix.h"
|
|
|
|
#include "ompi/info/info.h"
|
|
#include "ompi/mca/rte/rte.h"
|
|
#include "ompi/mca/pubsub/base/base.h"
|
|
#include "pubsub_pmi.h"
|
|
|
|
/*
|
|
* Init the module
|
|
*/
|
|
static int init(void)
|
|
{
|
|
// did the pmix.init in the component
|
|
return OMPI_SUCCESS;
|
|
}
|
|
|
|
/*
|
|
* publish the port_name for the specified service_name.
|
|
*/
|
|
static int publish(const char *service_name, ompi_info_t *info, const char *port_name)
|
|
{
|
|
pmix_info_t *p;
|
|
opal_list_t xfer;
|
|
ompi_info_entry_t *ie;
|
|
int rc;
|
|
|
|
/* transfer the ompi_info_t data to an array of pmix_info_t structs */
|
|
OBJ_CONSTRUCT(&xfer, opal_list_t);
|
|
OPAL_LIST_FOREACH(ie, &info->super, ompi_info_entry_t) {
|
|
p = OBJ_NEW(pmix_info_t);
|
|
strncpy(p->key, ie->ie_key, PMIX_MAX_INFO_KEY);
|
|
strncpy(p->value, ie->ie_value, PMIX_MAX_INFO_VAL);
|
|
opal_list_append(&xfer, &p->super);
|
|
}
|
|
|
|
rc = opal_pmix.publish(service_name, &xfer, port_name);
|
|
OPAL_LIST_DESTRUCT(&xfer);
|
|
return rc;
|
|
}
|
|
|
|
static char* lookup(const char *service_name, ompi_info_t *info)
|
|
{
|
|
char port[PMIX_MAX_VALLEN], *ret;
|
|
pmix_info_t *p;
|
|
opal_list_t xfer;
|
|
ompi_info_entry_t *ie;
|
|
int rc;
|
|
|
|
/* transfer the ompi_info_t data to an array of pmix_info_t structs */
|
|
OBJ_CONSTRUCT(&xfer, opal_list_t);
|
|
OPAL_LIST_FOREACH(ie, &info->super, ompi_info_entry_t) {
|
|
p = OBJ_NEW(pmix_info_t);
|
|
strncpy(p->key, ie->ie_key, PMIX_MAX_INFO_KEY);
|
|
strncpy(p->value, ie->ie_value, PMIX_MAX_INFO_VAL);
|
|
opal_list_append(&xfer, &p->super);
|
|
}
|
|
rc = opal_pmix.lookup(service_name, &xfer, port, PMIX_MAX_VALLEN);
|
|
OPAL_LIST_DESTRUCT(&xfer);
|
|
|
|
/* in error case port will be set to NULL
|
|
* this is what our callers expect to see
|
|
* In future maybe some error handling need?
|
|
*/
|
|
if( rc != OPAL_SUCCESS ){
|
|
// improve error processing
|
|
return NULL;
|
|
}
|
|
ret = strdup(port);
|
|
return ret;
|
|
}
|
|
|
|
/*
|
|
* delete the entry */
|
|
static int unpublish(const char *service_name, ompi_info_t *info)
|
|
{
|
|
pmix_info_t *p;
|
|
opal_list_t xfer;
|
|
ompi_info_entry_t *ie;
|
|
int rc;
|
|
|
|
/* transfer the ompi_info_t data to an array of pmix_info_t structs */
|
|
OBJ_CONSTRUCT(&xfer, opal_list_t);
|
|
OPAL_LIST_FOREACH(ie, &info->super, ompi_info_entry_t) {
|
|
p = OBJ_NEW(pmix_info_t);
|
|
strncpy(p->key, ie->ie_key, PMIX_MAX_INFO_KEY);
|
|
strncpy(p->value, ie->ie_value, PMIX_MAX_INFO_VAL);
|
|
opal_list_append(&xfer, &p->super);
|
|
}
|
|
rc = opal_pmix.unpublish(service_name, &xfer);
|
|
OPAL_LIST_DESTRUCT(&xfer);
|
|
return rc;
|
|
}
|
|
|
|
|
|
/*
|
|
* finalize the module
|
|
*/
|
|
static int finalize(void)
|
|
{
|
|
return OMPI_SUCCESS;
|
|
}
|
|
|
|
/*
|
|
* instantiate the module
|
|
*/
|
|
ompi_pubsub_base_module_t ompi_pubsub_pmi_module = {
|
|
init,
|
|
publish,
|
|
unpublish,
|
|
lookup,
|
|
finalize
|
|
};
|