1
1
openmpi/orte/mca/rmgr/base/rmgr_private.h
Ralph Castain 4fff584a68 Commit the orted-failed-to-start code. This correctly causes the system to detect the failure of an orted to start and allows the system to terminate all procs/orteds that *did* start.
The primary change that underlies all this is in the OOB. Specifically, the problem in the code until now has been that the OOB attempts to resolve an address when we call the "send" to an unknown recipient. The OOB would then wait forever if that recipient never actually started (and hence, never reported back its OOB contact info). In the case of an orted that failed to start, we would correctly detect that the orted hadn't started, but then we would attempt to order all orteds (including the one that failed to start) to die. This would cause the OOB to "hang" the system.

Unfortunately, revising how the OOB resolves addresses introduced a number of additional problems. Specifically, and most troublesome, was the fact that comm_spawn involved the immediate transmission of the rendezvous point from parent-to-child after the child was spawned. The current code used the OOB address resolution as a "barrier" - basically, the parent would attempt to send the info to the child, and then "hold" there until the child's contact info had arrived (meaning the child had started) and the send could be completed.

Note that this also caused comm_spawn to "hang" the entire system if the child never started... The app-failed-to-start helped improve that behavior - this code provides additional relief.

With this change, the OOB will return an ADDRESSEE_UNKNOWN error if you attempt to send to a recipient whose contact info isn't already in the OOB's hash tables. To resolve comm_spawn issues, we also now force the cross-sharing of connection info between parent and child jobs during spawn.

Finally, to aid in setting triggers to the right values, we introduce the "arith" API for the GPR. This function allows you to atomically change the value in a registry location (either divide, multiply, add, or subtract) by the provided operand. It is equivalent to first fetching the value using a "get", then modifying it, and then putting the result back into the registry via a "put".

This commit was SVN r14711.
2007-05-21 18:31:28 +00:00

229 строки
7.8 KiB
C

/*
* Copyright (c) 2004-2007 The Trustees of Indiana University and Indiana
* University Research and Technology
* Corporation. All rights reserved.
* Copyright (c) 2004-2006 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.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
/** @file:
*/
#ifndef ORTE_RMGR_PRIVATE_H
#define ORTE_RMGR_PRIVATE_H
/*
* includes
*/
#include "orte_config.h"
#include "orte/orte_constants.h"
#include "orte/orte_types.h"
#include "opal/class/opal_list.h"
#include "orte/dss/dss.h"
#include "opal/mca/mca.h"
#include "orte/mca/gpr/gpr_types.h"
#include "orte/mca/rml/rml_types.h"
#include "orte/mca/rmgr/rmgr.h"
/*
* Global functions for MCA overall collective open and close
*/
#if defined(c_plusplus) || defined(__cplusplus)
extern "C" {
#endif
/*
* Constants for command values
*/
#define ORTE_RMGR_SETUP_JOB_CMD 1
#define ORTE_RMGR_SPAWN_JOB_CMD 2
#define ORTE_RMGR_SETUP_GATES_CMD 3
#define ORTE_RMGR_XCONNECT_CMD 4
#define ORTE_RMGR_CMD ORTE_UINT8
typedef uint8_t orte_rmgr_cmd_t;
/*
* Base functions that are common to all implementations - can be overridden
*/
ORTE_DECLSPEC int orte_rmgr_base_get_app_context(
orte_jobid_t jobid,
orte_app_context_t*** app_context,
orte_std_cntr_t* num_context);
ORTE_DECLSPEC int orte_rmgr_base_put_app_context(
orte_jobid_t jobid,
orte_app_context_t** app_context,
orte_std_cntr_t num_context);
ORTE_DECLSPEC int orte_rmgr_base_get_job_slots(
orte_jobid_t jobid,
orte_std_cntr_t* num_slots);
ORTE_DECLSPEC int orte_rmgr_base_set_job_slots(
orte_jobid_t jobid,
orte_std_cntr_t num_slots);
ORTE_DECLSPEC int orte_rmgr_base_check_context_app(orte_app_context_t *context);
ORTE_DECLSPEC int orte_rmgr_base_check_context_cwd(orte_app_context_t *context,
bool want_chdir);
ORTE_DECLSPEC int orte_rmgr_base_set_proc_info(const orte_process_name_t* name, pid_t pid, char * nodename);
ORTE_DECLSPEC int orte_rmgr_base_get_proc_info(const orte_process_name_t* name, pid_t* pid, char **nodename);
ORTE_DECLSPEC int orte_rmgr_base_connect(orte_std_cntr_t num_connect,
orte_process_name_t *connect);
ORTE_DECLSPEC int orte_rmgr_base_disconnect(orte_std_cntr_t num_disconnect,
orte_process_name_t *disconnect);
ORTE_DECLSPEC orte_gpr_keyval_t* orte_rmgr_base_find_attribute(opal_list_t* attr_list, char* key);
ORTE_DECLSPEC int orte_rmgr_base_add_attribute(opal_list_t* attr_list, char* key,
orte_data_type_t type, void *data,
bool overwrite);
ORTE_DECLSPEC int orte_rmgr_base_merge_attributes(opal_list_t* target, opal_list_t* source, bool override);
ORTE_DECLSPEC int orte_rmgr_base_delete_attribute(opal_list_t* attr_list, char* key);
/*
* Internal definitions
*/
int orte_rmgr_base_setup_job_not_available(
orte_app_context_t** app_context,
orte_std_cntr_t num_context,
orte_jobid_t* jobid, opal_list_t *attrs);
int orte_rmgr_base_spawn_not_available(
orte_app_context_t** app_context,
orte_std_cntr_t num_context,
orte_jobid_t* jobid,
orte_std_cntr_t num_connect,
orte_process_name_t *connect,
orte_rmgr_cb_fn_t cbfn,
orte_proc_state_t cb_conditions,
opal_list_t *attributes);
int orte_rmgr_base_finalize_not_available(void);
/*
* Support functions
*/
ORTE_DECLSPEC void orte_rmgr_base_purge_mca_params(char ***env);
ORTE_DECLSPEC int orte_rmgr_base_proc_stage_gate_init(orte_jobid_t job);
ORTE_DECLSPEC int orte_rmgr_base_proc_stage_gate_mgr(orte_gpr_notify_message_t *msg);
ORTE_DECLSPEC int orte_rmgr_base_xconnect(orte_jobid_t child, orte_jobid_t parent);
ORTE_DECLSPEC int orte_rmgr_base_comm_start(void);
ORTE_DECLSPEC int orte_rmgr_base_comm_stop(void);
void orte_rmgr_base_recv(int status, orte_process_name_t* sender,
orte_buffer_t* buffer, orte_rml_tag_t tag,
void* cbdata);
/*
* DATA TYPE PACKING FUNCTIONS
*/
int orte_rmgr_base_pack_app_context(orte_buffer_t *buffer, void *src,
orte_std_cntr_t num_vals, orte_data_type_t type);
int orte_rmgr_base_pack_app_context_map(orte_buffer_t *buffer, void *src,
orte_std_cntr_t num_vals, orte_data_type_t type);
int orte_rmgr_base_pack_attr_list(orte_buffer_t *buffer, void *src,
orte_std_cntr_t num_vals, orte_data_type_t type);
int orte_rmgr_base_pack_attribute(orte_buffer_t *buffer, void *src,
orte_std_cntr_t num_vals, orte_data_type_t type);
/*
* DATA TYPE UNPACKING FUNCTIONS
*/
ORTE_DECLSPEC int orte_rmgr_base_unpack_app_context(orte_buffer_t *buffer, void *dest,
orte_std_cntr_t *num_vals, orte_data_type_t type);
ORTE_DECLSPEC int orte_rmgr_base_unpack_app_context_map(orte_buffer_t *buffer, void *dest,
orte_std_cntr_t *num_vals, orte_data_type_t type);
ORTE_DECLSPEC int orte_rmgr_base_unpack_attr_list(orte_buffer_t *buffer, void *dest,
orte_std_cntr_t *num_vals, orte_data_type_t type);
ORTE_DECLSPEC int orte_rmgr_base_unpack_attribute(orte_buffer_t *buffer, void *dest,
orte_std_cntr_t *num_vals, orte_data_type_t type);
/*
* COMPARE FUNCTIONS
*/
int orte_rmgr_base_compare_app_context(orte_app_context_t *value1, orte_app_context_t *value2, orte_data_type_t type);
int orte_rmgr_base_compare_app_context_map(orte_app_context_map_t *value1, orte_app_context_map_t *value2, orte_data_type_t type);
int orte_rmgr_base_compare_attr_list(opal_list_t *value1, opal_list_t *value2, orte_data_type_t type);
int orte_rmgr_base_compare_attribute(orte_attribute_t *value1, orte_attribute_t *value2, orte_data_type_t type);
/*
* COPY FUNCTIONS
*/
int orte_rmgr_base_copy_app_context(orte_app_context_t **dest, orte_app_context_t *src, orte_data_type_t type);
int orte_rmgr_base_copy_app_context_map(orte_app_context_map_t **dest, orte_app_context_map_t *src, orte_data_type_t type);
int orte_rmgr_base_copy_attr_list(opal_list_t **dest, opal_list_t *src, orte_data_type_t type);
int orte_rmgr_base_copy_attribute(orte_attribute_t **dest, orte_attribute_t *src, orte_data_type_t type);
/*
* PRINT FUNCTIONS
*/
int orte_rmgr_base_print_app_context(char **output, char *prefix, orte_app_context_t *src, orte_data_type_t type);
int orte_rmgr_base_print_app_context_map(char **output, char *prefix, orte_app_context_map_t *src, orte_data_type_t type);
int orte_rmgr_base_print_attribute(char **output, char *prefix, orte_attribute_t *src, orte_data_type_t type);
int orte_rmgr_base_print_attr_list(char **output, char *prefix, opal_list_t *src, orte_data_type_t type);
/*
* SIZE FUNCTIONS
*/
int orte_rmgr_base_size_app_context(size_t *size, orte_app_context_t *src, orte_data_type_t type);
int orte_rmgr_base_size_app_context_map(size_t *size, orte_app_context_map_t *src, orte_data_type_t type);
int orte_rmgr_base_size_attr_list(size_t *size, opal_list_t *src, orte_data_type_t type);
int orte_rmgr_base_size_attribute(size_t *size, orte_attribute_t *src, orte_data_type_t type);
/*
* RELEASE FUNCTIONS
*/
void orte_rmgr_base_std_obj_release(orte_data_value_t *value);
#if defined(c_plusplus) || defined(__cplusplus)
}
#endif
#endif