1
1
openmpi/ompi/mca/rte/pmix/rte_pmix_module.c

760 строки
24 KiB
C
Исходник Обычный вид История

/*
* Copyright (c) 2012-2013 Los Alamos National Security, LLC.
* All rights reserved.
* Copyright (c) 2013-2017 Intel, Inc. All rights reserved.
* Copyright (c) 2012-2014 The University of Tennessee and The University
* of Tennessee Research Foundation. All rights
* reserved.
* Copyright (c) 2014 Cisco Systems, Inc. All rights reserved.
* $COPYRIGHT$
*/
#include "ompi_config.h"
#include "ompi/constants.h"
#include <string.h>
#include <stdio.h>
#include <ctype.h>
#ifdef HAVE_SYS_TYPES_H
#include <sys/types.h>
#endif /* HAVE_SYS_TYPES_H */
#include <sys/stat.h>
#ifdef HAVE_UNISTD_H
#include <unistd.h>
#endif /* HAVE_UNISTD_H */
#ifdef HAVE_DIRENT_H
#include <dirent.h>
#endif /* HAVE_DIRENT_H */
#ifdef HAVE_PWD_H
#include <pwd.h>
#endif /* HAVE_PWD_H */
#include "opal/dss/dss.h"
#include "opal/util/argv.h"
#include "opal/util/error.h"
#include "opal/util/opal_getcwd.h"
#include "opal/util/os_path.h"
#include "opal/util/os_dirpath.h"
#include "opal/util/proc.h"
#include "opal/util/show_help.h"
#include "opal/mca/hwloc/base/base.h"
#include "opal/mca/pmix/base/base.h"
#include "opal/threads/threads.h"
#include "opal/class/opal_list.h"
#include "opal/dss/dss.h"
#include "ompi/mca/rte/base/base.h"
#include "ompi/mca/rte/rte.h"
#include "ompi/debuggers/debuggers.h"
#include "ompi/proc/proc.h"
#include "ompi/runtime/params.h"
#include "ompi/communicator/communicator.h"
/* instantiate a debugger-required value */
volatile int MPIR_being_debugged = 0;
extern ompi_rte_component_t mca_rte_pmix_component;
/* storage to support OMPI */
opal_process_name_t pmix_name_wildcard = {UINT32_MAX-1, UINT32_MAX-1};
opal_process_name_t pmix_name_invalid = {UINT32_MAX, UINT32_MAX};
opal_process_name_t pmix_proc_my_name = {0, 0};
hwloc_cpuset_t ompi_proc_applied_binding = NULL;
pmix_process_info_t pmix_process_info = {0};
bool pmix_proc_is_bound = false;
static bool pmix_in_parallel_debugger = false;
static bool added_transport_keys = false;
static bool added_num_procs = false;
static bool added_app_ctx = false;
static char* pre_condition_transports_print(uint64_t *unique_key);
static int _setup_job_session_dir(char **sdir);
#define ORTE_SCHEMA_DELIMITER_CHAR '.'
#define ORTE_SCHEMA_WILDCARD_CHAR '*'
#define ORTE_SCHEMA_WILDCARD_STRING "*"
#define ORTE_SCHEMA_INVALID_CHAR '$'
#define ORTE_SCHEMA_INVALID_STRING "$"
int ompi_rte_compare_name_fields(ompi_rte_cmp_bitmask_t fields,
const opal_process_name_t* name1,
const opal_process_name_t* name2)
{
/* handle the NULL pointer case */
if (NULL == name1 && NULL == name2) {
return OPAL_EQUAL;
} else if (NULL == name1) {
return OPAL_VALUE2_GREATER;
} else if (NULL == name2) {
return OPAL_VALUE1_GREATER;
}
/* in this comparison function, we check for exact equalities.
* In the case of wildcards, we check to ensure that the fields
* actually match those values - thus, a "wildcard" in this
* function does not actually stand for a wildcard value, but
* rather a specific value - UNLESS the CMP_WILD bitmask value
* is set
*/
/* check job id */
if (OMPI_RTE_CMP_JOBID & fields) {
if (OMPI_RTE_CMP_WILD & fields &&
(pmix_name_wildcard.jobid == name1->jobid ||
pmix_name_wildcard.jobid == name2->jobid)) {
goto check_vpid;
}
if (name1->jobid < name2->jobid) {
return OPAL_VALUE2_GREATER;
} else if (name1->jobid > name2->jobid) {
return OPAL_VALUE1_GREATER;
}
}
/* get here if jobid's are equal, or not being checked
* now check vpid
*/
check_vpid:
if (OMPI_RTE_CMP_VPID & fields) {
if (OMPI_RTE_CMP_WILD & fields &&
(pmix_name_wildcard.vpid == name1->vpid ||
pmix_name_wildcard.vpid == name2->vpid)) {
return OPAL_EQUAL;
}
if (name1->vpid < name2->vpid) {
return OPAL_VALUE2_GREATER;
} else if (name1->vpid > name2->vpid) {
return OPAL_VALUE1_GREATER;
}
}
/* only way to get here is if all fields are being checked and are equal,
* or jobid not checked, but vpid equal,
* only vpid being checked, and equal
* return that fact
*/
return OPAL_EQUAL;
}
int ompi_rte_convert_string_to_process_name(opal_process_name_t *name,
const char* name_string)
{
char *temp, *token;
opal_jobid_t job;
opal_vpid_t vpid;
int return_code=OPAL_SUCCESS;
/* set default */
name->jobid = pmix_name_invalid.jobid;
name->vpid = pmix_name_invalid.vpid;
/* check for NULL string - error */
if (NULL == name_string) {
OPAL_ERROR_LOG(OPAL_ERR_BAD_PARAM);
return OPAL_ERR_BAD_PARAM;
}
temp = strdup(name_string); /** copy input string as the strtok process is destructive */
token = strchr(temp, ORTE_SCHEMA_DELIMITER_CHAR); /** get first field -> jobid */
/* check for error */
if (NULL == token) {
OPAL_ERROR_LOG(OPAL_ERR_BAD_PARAM);
free(temp);
return OPAL_ERR_BAD_PARAM;
}
*token = '\0';
token++;
/* check for WILDCARD character - assign
* value accordingly, if found
*/
if (0 == strcmp(temp, ORTE_SCHEMA_WILDCARD_STRING)) {
job = pmix_name_wildcard.jobid;
} else if (0 == strcmp(temp, ORTE_SCHEMA_INVALID_STRING)) {
job = pmix_name_invalid.jobid;
} else {
job = strtoul(temp, NULL, 10);
}
/* check for WILDCARD character - assign
* value accordingly, if found
*/
if (0 == strcmp(token, ORTE_SCHEMA_WILDCARD_STRING)) {
vpid = pmix_name_wildcard.vpid;
} else if (0 == strcmp(token, ORTE_SCHEMA_INVALID_STRING)) {
vpid = pmix_name_invalid.vpid;
} else {
vpid = strtoul(token, NULL, 10);
}
name->jobid = job;
name->vpid = vpid;
free(temp);
return return_code;
}
int ompi_rte_convert_process_name_to_string(char** name_string,
const opal_process_name_t *name)
{
char *tmp, *tmp2;
if (NULL == name) { /* got an error */
OPAL_ERROR_LOG(OPAL_ERR_BAD_PARAM);
return OPAL_ERR_BAD_PARAM;
}
/* check for wildcard and invalid values - where encountered, insert the
* corresponding string so we can correctly parse the name string when
* it is passed back to us later
*/
if (pmix_name_wildcard.jobid == name->jobid) {
asprintf(&tmp, "%s", ORTE_SCHEMA_WILDCARD_STRING);
} else if (pmix_name_invalid.jobid == name->jobid) {
asprintf(&tmp, "%s", ORTE_SCHEMA_INVALID_STRING);
} else {
asprintf(&tmp, "%lu", (unsigned long)name->jobid);
}
if (pmix_name_wildcard.vpid == name->vpid) {
asprintf(&tmp2, "%s%c%s", tmp, ORTE_SCHEMA_DELIMITER_CHAR, ORTE_SCHEMA_WILDCARD_STRING);
} else if (pmix_name_invalid.vpid == name->vpid) {
asprintf(&tmp2, "%s%c%s", tmp, ORTE_SCHEMA_DELIMITER_CHAR, ORTE_SCHEMA_INVALID_STRING);
} else {
asprintf(&tmp2, "%s%c%lu", tmp, ORTE_SCHEMA_DELIMITER_CHAR, (unsigned long)name->vpid);
}
asprintf(name_string, "%s", tmp2);
free(tmp);
free(tmp2);
return OPAL_SUCCESS;
}
int ompi_rte_init(int *pargc, char ***pargv)
{
int ret;
char *error = NULL;
opal_process_name_t pname;
opal_proc_t *myname;
int u32, *u32ptr;
uint16_t u16, *u16ptr;
char **peers=NULL, *mycpuset;
char *envar, *ev1, *ev2;
opal_value_t *kv;
char *val;
size_t i;
uint64_t unique_key[2];
char *string_key;
u32ptr = &u32;
u16ptr = &u16;
memset(&pmix_process_info, 0, sizeof(pmix_process_info));
/* initialize the opal layer */
if (OPAL_SUCCESS != (ret = opal_init(pargc, pargv))) {
error = "opal_init";
goto error;
}
/* open and setup pmix */
if (OPAL_SUCCESS != (ret = mca_base_framework_open(&opal_pmix_base_framework, 0))) {
OPAL_ERROR_LOG(ret);
/* we cannot run */
error = "pmix init";
goto error;
}
if (OPAL_SUCCESS != (ret = opal_pmix_base_select())) {
/* we cannot run */
error = "pmix init";
goto error;
}
/* set the event base */
opal_pmix_base_set_evbase(opal_sync_event_base);
/* initialize the selected module */
if (!opal_pmix.initialized() && (OPAL_SUCCESS != (ret = opal_pmix.init(NULL)))) {
/* we cannot run - this could be due to being direct launched
* without the required PMI support being built, so print
* out a help message indicating it */
opal_show_help("help-ompi-rte-pmix.txt", "no-pmi", true);
return OPAL_ERR_SILENT;
}
/* opal_pmix.init will have filled in proc name fields in
* OPAL, so transfer them here */
myname = opal_proc_local_get();
pmix_proc_my_name = myname->proc_name;
/* get our hostname */
pmix_process_info.nodename = opal_get_proc_hostname(myname);
/* get our local rank from PMI */
OPAL_MODEX_RECV_VALUE(ret, OPAL_PMIX_LOCAL_RANK,
&pmix_proc_my_name, &u16ptr, OPAL_UINT16);
if (OPAL_SUCCESS != ret) {
error = "getting local rank";
goto error;
}
pmix_process_info.my_local_rank = u16;
/* get our node rank from PMI */
OPAL_MODEX_RECV_VALUE(ret, OPAL_PMIX_NODE_RANK,
&pmix_proc_my_name, &u16ptr, OPAL_UINT16);
if (OPAL_SUCCESS != ret) {
error = "getting node rank";
goto error;
}
pmix_process_info.my_node_rank = u16;
/* get job size */
OPAL_MODEX_RECV_VALUE(ret, OPAL_PMIX_JOB_SIZE,
&pmix_name_wildcard, &u32ptr, OPAL_UINT32);
if (OPAL_SUCCESS != ret) {
error = "getting job size";
goto error;
}
pmix_process_info.num_procs = u32;
/* push into the environ for pickup in MPI layer for
* MPI-3 required info key
*/
if (NULL == getenv(OPAL_MCA_PREFIX"orte_ess_num_procs")) {
asprintf(&ev1, OPAL_MCA_PREFIX"orte_ess_num_procs=%d", pmix_process_info.num_procs);
putenv(ev1);
added_num_procs = true;
}
if (NULL == getenv("OMPI_APP_CTX_NUM_PROCS")) {
asprintf(&ev2, "OMPI_APP_CTX_NUM_PROCS=%d", pmix_process_info.num_procs);
putenv(ev2);
added_app_ctx = true;
}
/* get our app number from PMI - ok if not found */
OPAL_MODEX_RECV_VALUE_OPTIONAL(ret, OPAL_PMIX_APPNUM,
&pmix_proc_my_name, &u32ptr, OPAL_UINT32);
if (OPAL_SUCCESS == ret) {
pmix_process_info.app_num = u32;
} else {
pmix_process_info.app_num = 0;
}
/* get the number of local peers - required for wireup of
* shared memory BTL */
OPAL_MODEX_RECV_VALUE(ret, OPAL_PMIX_LOCAL_SIZE,
&pmix_name_wildcard, &u32ptr, OPAL_UINT32);
if (OPAL_SUCCESS == ret) {
pmix_process_info.num_local_peers = u32 - 1; // want number besides ourselves
} else {
pmix_process_info.num_local_peers = 0;
}
/* setup transport keys in case the MPI layer needs them -
* we can use the jobfam and stepid as unique keys
* because they are unique values assigned by the RM
*/
if (NULL == getenv(OPAL_MCA_PREFIX"orte_precondition_transports")) {
unique_key[0] = (pmix_proc_my_name.jobid & 0xff00) >> 16;
unique_key[1] = pmix_proc_my_name.jobid & 0x00ff;
if (NULL == (string_key = pre_condition_transports_print(unique_key))) {
OPAL_ERROR_LOG(OPAL_ERR_OUT_OF_RESOURCE);
return OPAL_ERR_OUT_OF_RESOURCE;
}
opal_output_verbose(2, ompi_rte_base_framework.framework_output,
"%s transport key %s",
OPAL_NAME_PRINT(pmix_proc_my_name), string_key);
asprintf(&envar, OPAL_MCA_PREFIX"orte_precondition_transports=%s", string_key);
putenv(envar);
added_transport_keys = true;
/* cannot free the envar as that messes up our environ */
free(string_key);
}
/* retrieve temp directories info */
OPAL_MODEX_RECV_VALUE_OPTIONAL(ret, OPAL_PMIX_NSDIR, &pmix_name_wildcard, &val, OPAL_STRING);
if (OPAL_SUCCESS == ret && NULL != val) {
pmix_process_info.job_session_dir = val;
val = NULL;
} else {
/* we need to create something */
ret = _setup_job_session_dir(&pmix_process_info.job_session_dir);
if (OPAL_SUCCESS != ret) {
error = "job session directory";
goto error;
}
}
/* get our local peers */
if (0 < pmix_process_info.num_local_peers) {
/* if my local rank if too high, then that's an error */
if (pmix_process_info.num_local_peers < pmix_process_info.my_local_rank) {
ret = OPAL_ERR_BAD_PARAM;
error = "num local peers";
goto error;
}
/* retrieve the local peers */
OPAL_MODEX_RECV_VALUE(ret, OPAL_PMIX_LOCAL_PEERS,
&pmix_name_wildcard, &val, OPAL_STRING);
if (OPAL_SUCCESS == ret && NULL != val) {
peers = opal_argv_split(val, ',');
free(val);
} else {
peers = NULL;
}
} else {
peers = NULL;
}
/* set the locality */
if (NULL != peers) {
/* identify our location */
val = NULL;
OPAL_MODEX_RECV_VALUE_OPTIONAL(ret, OPAL_PMIX_LOCALITY_STRING,
&pmix_proc_my_name, &val, OPAL_STRING);
if (OPAL_SUCCESS == ret && NULL != val) {
mycpuset = val;
} else {
mycpuset = NULL;
}
pname.jobid = pmix_proc_my_name.jobid;
for (i=0; NULL != peers[i]; i++) {
pname.vpid = strtoul(peers[i], NULL, 10);
if (pname.vpid == pmix_proc_my_name.vpid) {
/* we are fully local to ourselves */
u16 = OPAL_PROC_ALL_LOCAL;
} else {
val = NULL;
OPAL_MODEX_RECV_VALUE_OPTIONAL(ret, OPAL_PMIX_LOCALITY_STRING,
&pname, &val, OPAL_STRING);
if (OPAL_SUCCESS == ret && NULL != val) {
u16 = opal_hwloc_compute_relative_locality(mycpuset, val);
free(val);
} else {
/* all we can say is that it shares our node */
u16 = OPAL_PROC_ON_CLUSTER | OPAL_PROC_ON_CU | OPAL_PROC_ON_NODE;
}
}
kv = OBJ_NEW(opal_value_t);
kv->key = strdup(OPAL_PMIX_LOCALITY);
kv->type = OPAL_UINT16;
OPAL_OUTPUT_VERBOSE((1, ompi_rte_base_framework.framework_output,
"%s locality: proc %s locality %s",
OPAL_NAME_PRINT(pmix_proc_my_name),
OPAL_NAME_PRINT(pname), opal_hwloc_base_print_locality(u16)));
kv->data.uint16 = u16;
ret = opal_pmix.store_local(&pname, kv);
if (OPAL_SUCCESS != ret) {
error = "local store of locality";
opal_argv_free(peers);
if (NULL != mycpuset) {
free(mycpuset);
}
goto error;
}
OBJ_RELEASE(kv);
}
opal_argv_free(peers);
if (NULL != mycpuset) {
free(mycpuset);
}
}
/* poor attempt to detect we are bound */
if (NULL != getenv("SLURM_CPU_BIND_TYPE")) {
pmix_proc_is_bound = true;
}
/* push our hostname so others can find us, if they need to - the
* native PMIx component will ignore this request as the hostname
* is provided by the system */
OPAL_MODEX_SEND_VALUE(ret, OPAL_PMIX_GLOBAL, OPAL_PMIX_HOSTNAME, pmix_process_info.nodename, OPAL_STRING);
if (OPAL_SUCCESS != ret) {
error = "db store hostname";
goto error;
}
return OPAL_SUCCESS;
error:
opal_show_help_finalize();
if (OPAL_ERR_SILENT != ret ) {
opal_show_help("help-ompi-rte-pmix.txt",
"internal-failure",
true, error, opal_strerror(ret), ret);
}
return ret;
}
static bool check_file(const char *root, const char *path)
{
struct stat st;
char *fullpath;
/*
* Keep:
* - non-zero files starting with "output-"
*/
if (0 == strncmp(path, "output-", strlen("output-"))) {
fullpath = opal_os_path(false, &fullpath, root, path, NULL);
stat(fullpath, &st);
free(fullpath);
if (0 == st.st_size) {
return true;
}
return false;
}
return true;
}
int ompi_rte_finalize(void)
{
/* remove the envars that we pushed into environ
* so we leave that structure intact
*/
if (added_transport_keys) {
unsetenv(OPAL_MCA_PREFIX"orte_precondition_transports");
}
if (added_num_procs) {
unsetenv(OPAL_MCA_PREFIX"orte_ess_num_procs");
}
if (added_app_ctx) {
unsetenv("OMPI_APP_CTX_NUM_PROCS");
}
/* shutdown pmix */
if (NULL != opal_pmix.finalize) {
opal_pmix.finalize();
(void) mca_base_framework_close(&opal_pmix_base_framework);
}
/* cleanup the session directory we created */
if (NULL != pmix_process_info.job_session_dir) {
opal_os_dirpath_destroy(pmix_process_info.job_session_dir,
false, check_file);
free(pmix_process_info.job_session_dir);
}
return OMPI_SUCCESS;
}
void ompi_rte_abort(int error_code, char *fmt, ...)
{
va_list arglist;
char* buffer = NULL;
struct timespec tp = {0, 100000};
/* If there was a message, output it */
va_start(arglist, fmt);
if( NULL != fmt ) {
vasprintf( &buffer, fmt, arglist );
}
va_end(arglist);
/* call abort */
opal_pmix.abort(error_code, buffer, NULL);
if (NULL != buffer) {
free(buffer);
}
/* provide a little delay for the PMIx thread to
* get the info out */
nanosleep(&tp, NULL);
/* Now Exit */
_exit(error_code);
}
void ompi_rte_abort_peers(opal_process_name_t *procs,
int32_t num_procs,
int error_code)
{
return;
}
static size_t handler = SIZE_MAX;
static bool debugger_register_active = true;
static bool debugger_event_active = true;
static void _release_fn(int status,
const opal_process_name_t *source,
opal_list_t *info, opal_list_t *results,
opal_pmix_notification_complete_fn_t cbfunc,
void *cbdata)
{
/* must let the notifier know we are done */
if (NULL != cbfunc) {
cbfunc(OPAL_SUCCESS, NULL, NULL, NULL, cbdata);
}
debugger_event_active = false;
}
static void _register_fn(int status,
size_t evhandler_ref,
void *cbdata)
{
opal_list_t *codes = (opal_list_t*)cbdata;
handler = evhandler_ref;
OPAL_LIST_RELEASE(codes);
debugger_register_active = false;
}
/*
* Wait for a debugger if asked. We support two ways of waiting for
* attaching debuggers -- see big comment in
* pmix/tools/pmixrun/debuggers.c explaining the two scenarios.
*/
void ompi_rte_wait_for_debugger(void)
{
int debugger;
opal_list_t *codes, directives;
opal_value_t *kv;
char *evar;
int time;
/* check PMIx to see if we are under a debugger */
debugger = pmix_in_parallel_debugger;
if (1 == MPIR_being_debugged) {
debugger = 1;
}
if (!debugger && NULL == getenv("PMIX_TEST_DEBUGGER_ATTACH")) {
/* if not, just return */
return;
}
/* if we are being debugged, then we need to find
* the correct plug-ins
*/
ompi_debugger_setup_dlls();
if (NULL != (evar = getenv("PMIX_TEST_DEBUGGER_SLEEP"))) {
time = strtol(evar, NULL, 10);
sleep(time);
return;
}
/* register an event handler for the PMIX_ERR_DEBUGGER_RELEASE event */
codes = OBJ_NEW(opal_list_t);
kv = OBJ_NEW(opal_value_t);
kv->key = strdup("errorcode");
kv->type = OPAL_INT;
kv->data.integer = OPAL_ERR_DEBUGGER_RELEASE;
opal_list_append(codes, &kv->super);
OBJ_CONSTRUCT(&directives, opal_list_t);
kv = OBJ_NEW(opal_value_t);
kv->key = strdup(OPAL_PMIX_EVENT_HDLR_NAME);
kv->type = OPAL_STRING;
kv->data.string = strdup("MPI-DEBUGGER-ATTACH");
opal_list_append(&directives, &kv->super);
opal_pmix.register_evhandler(codes, &directives, _release_fn, _register_fn, codes);
/* let the MPI progress engine run while we wait for registration to complete */
OMPI_WAIT_FOR_COMPLETION(debugger_register_active);
OPAL_LIST_DESTRUCT(&directives);
/* let the MPI progress engine run while we wait for debugger release */
OMPI_WAIT_FOR_COMPLETION(debugger_event_active);
/* deregister the event handler */
opal_pmix.deregister_evhandler(handler, NULL, NULL);
}
bool ompi_rte_connect_accept_support(const char *port)
{
/* not sure how to support this yet */
return false;
}
static char* pre_condition_transports_print(uint64_t *unique_key)
{
unsigned int *int_ptr;
size_t i, j, string_key_len, written_len;
char *string_key = NULL, *format = NULL;
/* string is two 64 bit numbers printed in hex with a dash between
* and zero padding.
*/
string_key_len = (sizeof(uint64_t) * 2) * 2 + strlen("-") + 1;
string_key = (char*) malloc(string_key_len);
if (NULL == string_key) {
return NULL;
}
string_key[0] = '\0';
written_len = 0;
/* get a format string based on the length of an unsigned int. We
* want to have zero padding for sizeof(unsigned int) * 2
* characters -- when printing as a hex number, each byte is
* represented by 2 hex characters. Format will contain something
* that looks like %08lx, where the number 8 might be a different
* number if the system has a different sized long (8 would be for
* sizeof(int) == 4)).
*/
asprintf(&format, "%%0%dx", (int)(sizeof(unsigned int)) * 2);
/* print the first number */
int_ptr = (unsigned int*) &unique_key[0];
for (i = 0 ; i < sizeof(uint64_t) / sizeof(unsigned int) ; ++i) {
if (0 == int_ptr[i]) {
/* inject some energy */
for (j=0; j < sizeof(unsigned int); j++) {
int_ptr[i] |= j << j;
}
}
snprintf(string_key + written_len,
string_key_len - written_len,
format, int_ptr[i]);
written_len = strlen(string_key);
}
/* print the middle dash */
snprintf(string_key + written_len, string_key_len - written_len, "-");
written_len = strlen(string_key);
/* print the second number */
int_ptr = (unsigned int*) &unique_key[1];
for (i = 0 ; i < sizeof(uint64_t) / sizeof(unsigned int) ; ++i) {
if (0 == int_ptr[i]) {
/* inject some energy */
for (j=0; j < sizeof(unsigned int); j++) {
int_ptr[i] |= j << j;
}
}
snprintf(string_key + written_len,
string_key_len - written_len,
format, int_ptr[i]);
written_len = strlen(string_key);
}
free(format);
return string_key;
}
static int _setup_job_session_dir(char **sdir)
{
char *tmpdir;
/* get the effective uid */
uid_t uid = geteuid();
if( NULL == (tmpdir = getenv("TMPDIR")) )
if( NULL == (tmpdir = getenv("TEMP")) )
if( NULL == (tmpdir = getenv("TMP")) )
tmpdir = "/tmp";
if (0 > asprintf(&pmix_process_info.job_session_dir,
"%s/ompi.%s.%lu/jf.0/%u", tmpdir,
pmix_process_info.nodename,
(unsigned long)uid,
pmix_proc_my_name.jobid)) {
pmix_process_info.job_session_dir = NULL;
return OPAL_ERR_OUT_OF_RESOURCE;
}
return OPAL_SUCCESS;
}