2005-01-12 23:51:34 +03:00
|
|
|
/*
|
2005-11-05 22:57:48 +03:00
|
|
|
* 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.
|
2005-01-12 23:51:34 +03:00
|
|
|
* Copyright (c) 2004-2005 High Performance Computing Center Stuttgart,
|
|
|
|
* University of Stuttgart. All rights reserved.
|
2005-03-24 15:43:37 +03:00
|
|
|
* Copyright (c) 2004-2005 The Regents of the University of California.
|
|
|
|
* All rights reserved.
|
2005-01-12 23:51:34 +03:00
|
|
|
* $COPYRIGHT$
|
|
|
|
*
|
|
|
|
* Additional copyrights may follow
|
|
|
|
*
|
|
|
|
* $HEADER$
|
|
|
|
*/
|
|
|
|
|
2006-02-12 04:33:29 +03:00
|
|
|
#include "orte_config.h"
|
|
|
|
#include "orte/util/proc_info.h"
|
2005-07-04 03:31:27 +04:00
|
|
|
#include "opal/util/output.h"
|
2006-02-12 04:33:29 +03:00
|
|
|
#include "opal/mca/base/base.h"
|
|
|
|
#include "opal/mca/base/mca_base_param.h"
|
|
|
|
#include "orte/mca/rml/rml.h"
|
|
|
|
#include "orte/mca/rml/rml_types.h"
|
2005-01-12 23:51:34 +03:00
|
|
|
#include "iof_svc.h"
|
|
|
|
#include "iof_svc_proxy.h"
|
2005-10-07 01:21:26 +04:00
|
|
|
#include "iof_svc_pub.h"
|
|
|
|
#include "iof_svc_sub.h"
|
2005-01-12 23:51:34 +03:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Local functions
|
|
|
|
*/
|
2005-03-14 23:57:21 +03:00
|
|
|
static int orte_iof_svc_open(void);
|
|
|
|
static int orte_iof_svc_close(void);
|
2005-01-12 23:51:34 +03:00
|
|
|
|
2005-03-14 23:57:21 +03:00
|
|
|
static orte_iof_base_module_t* orte_iof_svc_init(
|
2005-01-12 23:51:34 +03:00
|
|
|
int* priority,
|
|
|
|
bool *allow_multi_user_threads,
|
|
|
|
bool *have_hidden_threads);
|
|
|
|
|
2005-03-18 06:43:59 +03:00
|
|
|
/*
|
|
|
|
* Local variables
|
|
|
|
*/
|
|
|
|
static bool initialized = false;
|
|
|
|
|
2005-01-12 23:51:34 +03:00
|
|
|
|
2005-03-14 23:57:21 +03:00
|
|
|
orte_iof_svc_component_t mca_iof_svc_component = {
|
2005-01-12 23:51:34 +03:00
|
|
|
{
|
|
|
|
/* First, the mca_base_component_t struct containing meta
|
|
|
|
information about the component itself */
|
|
|
|
|
|
|
|
{
|
|
|
|
/* Indicate that we are a iof v1.0.0 component (which also
|
|
|
|
implies a specific MCA version) */
|
|
|
|
|
2005-03-14 23:57:21 +03:00
|
|
|
ORTE_IOF_BASE_VERSION_1_0_0,
|
2005-01-12 23:51:34 +03:00
|
|
|
|
|
|
|
"svc", /* MCA component name */
|
Major simplifications to component versioning:
- After long discussions and ruminations on how we run components in
LAM/MPI, made the decision that, by default, all components included
in Open MPI will use the version number of their parent project
(i.e., OMPI or ORTE). They are certaint free to use a different
number, but this simplification makes the common cases easy:
- components are only released when the parent project is released
- it is easy (trivial?) to distinguish which version component goes
with with version of the parent project
- removed all autogen/configure code for templating the version .h
file in components
- made all ORTE components use ORTE_*_VERSION for version numbers
- made all OMPI components use OMPI_*_VERSION for version numbers
- removed all VERSION files from components
- configure now displays OPAL, ORTE, and OMPI version numbers
- ditto for ompi_info
- right now, faking it -- OPAL and ORTE and OMPI will always have the
same version number (i.e., they all come from the same top-level
VERSION file). But this paves the way for the Great Configure
Reorganization, where, among other things, each project will have
its own version number.
So all in all, we went from a boatload of version numbers to
[effectively] three. That's pretty good. :-)
This commit was SVN r6344.
2005-07-05 00:12:36 +04:00
|
|
|
ORTE_MAJOR_VERSION, /* MCA component major version */
|
|
|
|
ORTE_MINOR_VERSION, /* MCA component minor version */
|
|
|
|
ORTE_RELEASE_VERSION, /* MCA component release version */
|
2005-03-14 23:57:21 +03:00
|
|
|
orte_iof_svc_open, /* component open */
|
|
|
|
orte_iof_svc_close /* component close */
|
2005-01-12 23:51:34 +03:00
|
|
|
},
|
|
|
|
|
|
|
|
/* Next the MCA v1.0.0 component meta data */
|
|
|
|
{
|
|
|
|
/* Whether the component is checkpointable or not */
|
|
|
|
false
|
|
|
|
},
|
|
|
|
|
2005-03-14 23:57:21 +03:00
|
|
|
orte_iof_svc_init
|
2005-01-12 23:51:34 +03:00
|
|
|
}
|
|
|
|
};
|
|
|
|
|
2005-01-19 00:31:33 +03:00
|
|
|
#if 0
|
2005-03-14 23:57:21 +03:00
|
|
|
static char* orte_iof_svc_param_register_string(
|
2005-01-12 23:51:34 +03:00
|
|
|
const char* param_name,
|
|
|
|
const char* default_value)
|
|
|
|
{
|
|
|
|
char *param_value;
|
|
|
|
int id = mca_base_param_register_string("iof","svc",param_name,NULL,default_value);
|
|
|
|
mca_base_param_lookup_string(id, ¶m_value);
|
|
|
|
return param_value;
|
|
|
|
}
|
2005-01-19 00:31:33 +03:00
|
|
|
#endif
|
2005-01-12 23:51:34 +03:00
|
|
|
|
2005-03-14 23:57:21 +03:00
|
|
|
static int orte_iof_svc_param_register_int(
|
2005-01-12 23:51:34 +03:00
|
|
|
const char* param_name,
|
|
|
|
int default_value)
|
|
|
|
{
|
|
|
|
int id = mca_base_param_register_int("iof","svc",param_name,NULL,default_value);
|
|
|
|
int param_value = default_value;
|
|
|
|
mca_base_param_lookup_int(id,¶m_value);
|
|
|
|
return param_value;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
/**
|
|
|
|
* component open/close/init function
|
|
|
|
*/
|
2005-03-14 23:57:21 +03:00
|
|
|
static int orte_iof_svc_open(void)
|
2005-01-12 23:51:34 +03:00
|
|
|
{
|
2005-03-14 23:57:21 +03:00
|
|
|
mca_iof_svc_component.svc_debug = orte_iof_svc_param_register_int("debug", 1);
|
2005-03-18 06:43:59 +03:00
|
|
|
return ORTE_SUCCESS;
|
2005-01-12 23:51:34 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2005-03-14 23:57:21 +03:00
|
|
|
static int orte_iof_svc_close(void)
|
2005-01-12 23:51:34 +03:00
|
|
|
{
|
2005-07-03 20:22:16 +04:00
|
|
|
opal_list_item_t* item;
|
2005-03-18 06:43:59 +03:00
|
|
|
|
|
|
|
if (initialized) {
|
2005-07-04 02:45:48 +04:00
|
|
|
OPAL_THREAD_LOCK(&mca_iof_svc_component.svc_lock);
|
2005-07-03 20:22:16 +04:00
|
|
|
while((item = opal_list_remove_first(&mca_iof_svc_component.svc_subscribed)) != NULL) {
|
2005-03-18 06:43:59 +03:00
|
|
|
OBJ_RELEASE(item);
|
|
|
|
}
|
2005-07-03 20:22:16 +04:00
|
|
|
while((item = opal_list_remove_first(&mca_iof_svc_component.svc_published)) != NULL) {
|
2005-03-18 06:43:59 +03:00
|
|
|
OBJ_RELEASE(item);
|
|
|
|
}
|
2005-07-04 02:45:48 +04:00
|
|
|
OPAL_THREAD_UNLOCK(&mca_iof_svc_component.svc_lock);
|
Bring over the update to terminate orteds that are generated by a dynamic spawn such as comm_spawn. This introduces the concept of a job "family" - i.e., jobs that have a parent/child relationship. Comm_spawn'ed jobs have a parent (the one that spawned them). We track that relationship throughout the lineage - i.e., if a comm_spawned job in turn calls comm_spawn, then it has a parent (the one that spawned it) and a "root" job (the original job that started things).
Accordingly, there are new APIs to the name service to support the ability to get a job's parent, root, immediate children, and all its descendants. In addition, the terminate_job, terminate_orted, and signal_job APIs for the PLS have been modified to accept attributes that define the extent of their actions. For example, doing a "terminate_job" with an attribute of ORTE_NS_INCLUDE_DESCENDANTS will terminate the given jobid AND all jobs that descended from it.
I have tested this capability on a MacBook under rsh, Odin under SLURM, and LANL's Flash (bproc). It worked successfully on non-MPI jobs (both simple and including a spawn), and MPI jobs (again, both simple and with a spawn).
This commit was SVN r12597.
2006-11-14 22:34:59 +03:00
|
|
|
orte_rml.recv_cancel(ORTE_NAME_WILDCARD, ORTE_RML_TAG_IOF_SVC);
|
2005-01-12 23:51:34 +03:00
|
|
|
}
|
2005-03-18 06:43:59 +03:00
|
|
|
|
|
|
|
return ORTE_SUCCESS;
|
2005-01-12 23:51:34 +03:00
|
|
|
}
|
|
|
|
|
|
|
|
|
2005-10-07 01:21:26 +04:00
|
|
|
/**
|
|
|
|
* Callback when peer is disconnected
|
|
|
|
*/
|
|
|
|
|
|
|
|
static void
|
|
|
|
orte_iof_svc_exception_handler(const orte_process_name_t* peer, orte_rml_exception_t reason)
|
|
|
|
{
|
|
|
|
orte_iof_svc_sub_delete_all(peer);
|
|
|
|
orte_iof_svc_pub_delete_all(peer);
|
|
|
|
}
|
2005-01-12 23:51:34 +03:00
|
|
|
|
|
|
|
|
2005-10-07 01:21:26 +04:00
|
|
|
/**
|
|
|
|
* Module Initialization
|
|
|
|
*/
|
|
|
|
|
2005-03-14 23:57:21 +03:00
|
|
|
static orte_iof_base_module_t*
|
|
|
|
orte_iof_svc_init(int* priority, bool *allow_multi_user_threads, bool *have_hidden_threads)
|
2005-01-12 23:51:34 +03:00
|
|
|
{
|
|
|
|
int rc;
|
2005-03-18 06:43:59 +03:00
|
|
|
if (false == orte_process_info.seed) {
|
2005-01-12 23:51:34 +03:00
|
|
|
return NULL;
|
2005-03-18 06:43:59 +03:00
|
|
|
}
|
2005-01-12 23:51:34 +03:00
|
|
|
|
|
|
|
*priority = 1;
|
|
|
|
*allow_multi_user_threads = true;
|
|
|
|
*have_hidden_threads = false;
|
|
|
|
|
2005-07-03 20:22:16 +04:00
|
|
|
OBJ_CONSTRUCT(&mca_iof_svc_component.svc_subscribed, opal_list_t);
|
|
|
|
OBJ_CONSTRUCT(&mca_iof_svc_component.svc_published, opal_list_t);
|
2005-07-04 02:45:48 +04:00
|
|
|
OBJ_CONSTRUCT(&mca_iof_svc_component.svc_lock, opal_mutex_t);
|
2005-03-18 06:43:59 +03:00
|
|
|
|
2005-01-12 23:51:34 +03:00
|
|
|
/* post non-blocking recv */
|
|
|
|
mca_iof_svc_component.svc_iov[0].iov_base = NULL;
|
|
|
|
mca_iof_svc_component.svc_iov[0].iov_len = 0;
|
|
|
|
|
2005-03-14 23:57:21 +03:00
|
|
|
rc = orte_rml.recv_nb(
|
Bring over the update to terminate orteds that are generated by a dynamic spawn such as comm_spawn. This introduces the concept of a job "family" - i.e., jobs that have a parent/child relationship. Comm_spawn'ed jobs have a parent (the one that spawned them). We track that relationship throughout the lineage - i.e., if a comm_spawned job in turn calls comm_spawn, then it has a parent (the one that spawned it) and a "root" job (the original job that started things).
Accordingly, there are new APIs to the name service to support the ability to get a job's parent, root, immediate children, and all its descendants. In addition, the terminate_job, terminate_orted, and signal_job APIs for the PLS have been modified to accept attributes that define the extent of their actions. For example, doing a "terminate_job" with an attribute of ORTE_NS_INCLUDE_DESCENDANTS will terminate the given jobid AND all jobs that descended from it.
I have tested this capability on a MacBook under rsh, Odin under SLURM, and LANL's Flash (bproc). It worked successfully on non-MPI jobs (both simple and including a spawn), and MPI jobs (again, both simple and with a spawn).
This commit was SVN r12597.
2006-11-14 22:34:59 +03:00
|
|
|
ORTE_NAME_WILDCARD,
|
2005-01-12 23:51:34 +03:00
|
|
|
mca_iof_svc_component.svc_iov,
|
|
|
|
1,
|
2005-03-14 23:57:21 +03:00
|
|
|
ORTE_RML_TAG_IOF_SVC,
|
2005-10-25 17:38:13 +04:00
|
|
|
ORTE_RML_ALLOC|ORTE_RML_PERSISTENT,
|
2005-03-14 23:57:21 +03:00
|
|
|
orte_iof_svc_proxy_recv,
|
2005-01-12 23:51:34 +03:00
|
|
|
NULL
|
|
|
|
);
|
2006-02-12 04:33:29 +03:00
|
|
|
if(rc != ORTE_SUCCESS) {
|
2005-07-04 03:31:27 +04:00
|
|
|
opal_output(0, "orte_iof_svc_init: unable to post non-blocking recv");
|
2005-01-12 23:51:34 +03:00
|
|
|
return NULL;
|
|
|
|
}
|
2005-10-07 01:21:26 +04:00
|
|
|
|
|
|
|
rc = orte_rml.add_exception_handler(orte_iof_svc_exception_handler);
|
2005-03-18 06:43:59 +03:00
|
|
|
initialized = true;
|
2005-03-14 23:57:21 +03:00
|
|
|
return &orte_iof_svc_module;
|
2005-01-12 23:51:34 +03:00
|
|
|
}
|
|
|
|
|
2006-02-04 00:01:11 +03:00
|
|
|
|
|
|
|
int
|
|
|
|
orte_iof_svc_finalize(void)
|
|
|
|
{
|
|
|
|
orte_rml.del_exception_handler(orte_iof_svc_exception_handler);
|
|
|
|
|
2006-02-12 04:33:29 +03:00
|
|
|
return ORTE_SUCCESS;
|
2006-02-04 00:01:11 +03:00
|
|
|
}
|