Skip to content

Try to debug the MTT failures #5309

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Jul 2, 2018
Merged

Try to debug the MTT failures #5309

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b0fd430
Try to debug the MTT failures
Jun 20, 2018
4ced01c
Add a little more debug output
Jun 26, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
186 changes: 180 additions & 6 deletions opal/mca/pmix/pmix2x/pmix2x.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -949,9 +949,9 @@ int pmix2x_value_unload(opal_value_t *kv,
kv->type = OPAL_STATUS;
kv->data.status = pmix2x_convert_rc(v->data.status);
break;
case PMIX_PROC_RANK:
kv->type = OPAL_VPID;
kv->data.name.vpid = pmix2x_convert_rank(v->data.rank);
case PMIX_VALUE:
OPAL_ERROR_LOG(OPAL_ERR_NOT_SUPPORTED);
rc = OPAL_ERR_NOT_SUPPORTED;
break;
case PMIX_PROC:
kv->type = OPAL_NAME;
Expand All @@ -971,6 +971,18 @@ int pmix2x_value_unload(opal_value_t *kv,
}
kv->data.name.vpid = pmix2x_convert_rank(v->data.proc->rank);
break;
case PMIX_INFO:
OPAL_ERROR_LOG(OPAL_ERR_NOT_SUPPORTED);
rc = OPAL_ERR_NOT_SUPPORTED;
break;
case PMIX_PDATA:
OPAL_ERROR_LOG(OPAL_ERR_NOT_SUPPORTED);
rc = OPAL_ERR_NOT_SUPPORTED;
break;
case PMIX_BUFFER:
OPAL_ERROR_LOG(OPAL_ERR_NOT_SUPPORTED);
rc = OPAL_ERR_NOT_SUPPORTED;
break;
case PMIX_BYTE_OBJECT:
kv->type = OPAL_BYTE_OBJECT;
if (NULL != v->data.bo.bytes && 0 < v->data.bo.size) {
Expand All @@ -982,10 +994,22 @@ int pmix2x_value_unload(opal_value_t *kv,
kv->data.bo.size = 0;
}
break;
case PMIX_KVAL:
OPAL_ERROR_LOG(OPAL_ERR_NOT_SUPPORTED);
rc = OPAL_ERR_NOT_SUPPORTED;
break;
case PMIX_MODEX:
OPAL_ERROR_LOG(OPAL_ERR_NOT_SUPPORTED);
rc = OPAL_ERR_NOT_SUPPORTED;
break;
case PMIX_PERSIST:
kv->type = OPAL_PERSIST;
kv->data.uint8 = pmix2x_convert_persist(v->data.persist);
break;
case PMIX_POINTER:
kv->type = OPAL_PTR;
kv->data.ptr = v->data.ptr;
break;
case PMIX_SCOPE:
kv->type = OPAL_SCOPE;
kv->data.uint8 = pmix2x_convert_scope(v->data.scope);
Expand All @@ -994,15 +1018,54 @@ int pmix2x_value_unload(opal_value_t *kv,
kv->type = OPAL_DATA_RANGE;
kv->data.uint8 = pmix2x_convert_range(v->data.range);
break;
case PMIX_COMMAND:
OPAL_ERROR_LOG(OPAL_ERR_NOT_SUPPORTED);
rc = OPAL_ERR_NOT_SUPPORTED;
break;
case PMIX_INFO_DIRECTIVES:
OPAL_ERROR_LOG(OPAL_ERR_NOT_SUPPORTED);
rc = OPAL_ERR_NOT_SUPPORTED;
break;
case PMIX_DATA_TYPE:
OPAL_ERROR_LOG(OPAL_ERR_NOT_SUPPORTED);
rc = OPAL_ERR_NOT_SUPPORTED;
break;
case PMIX_PROC_STATE:
kv->type = OPAL_PROC_STATE;
/* the OPAL layer doesn't have any concept of proc state,
* so the ORTE layer is responsible for converting it */
memcpy(&kv->data.uint8, &v->data.state, sizeof(uint8_t));
break;
case PMIX_POINTER:
kv->type = OPAL_PTR;
kv->data.ptr = v->data.ptr;
case PMIX_PROC_INFO:
kv->type = OPAL_PROC_INFO;
if (NULL == v->data.pinfo) {
rc = OPAL_ERR_BAD_PARAM;
break;
}
/* see if this job is in our list of known nspaces */
found = false;
OPAL_LIST_FOREACH(job, &mca_pmix_pmix2x_component.jobids, opal_pmix2x_jobid_trkr_t) {
if (0 == strncmp(job->nspace, v->data.pinfo->proc.nspace, PMIX_MAX_NSLEN)) {
kv->data.pinfo.name.jobid = job->jobid;
found = true;
break;
}
}
if (!found) {
if (OPAL_SUCCESS != (rc = opal_convert_string_to_jobid(&kv->data.pinfo.name.jobid, v->data.pinfo->proc.nspace))) {
return pmix2x_convert_opalrc(rc);
}
}
kv->data.pinfo.name.vpid = pmix2x_convert_rank(v->data.pinfo->proc.rank);
if (NULL != v->data.pinfo->hostname) {
kv->data.pinfo.hostname = strdup(v->data.pinfo->hostname);
}
if (NULL != v->data.pinfo->executable_name) {
kv->data.pinfo.executable_name = strdup(v->data.pinfo->executable_name);
}
kv->data.pinfo.pid = v->data.pinfo->pid;
kv->data.pinfo.exit_code = v->data.pinfo->exit_code;
kv->data.pinfo.state = pmix2x_convert_state(v->data.pinfo->state);
break;
case PMIX_DATA_ARRAY:
if (NULL == v->data.darray || NULL == v->data.darray->array) {
Expand All @@ -1029,8 +1092,30 @@ int pmix2x_value_unload(opal_value_t *kv,
}
}
break;
case PMIX_PROC_RANK:
kv->type = OPAL_VPID;
kv->data.name.vpid = pmix2x_convert_rank(v->data.rank);
break;
case PMIX_QUERY:
OPAL_ERROR_LOG(OPAL_ERR_NOT_SUPPORTED);
rc = OPAL_ERR_NOT_SUPPORTED;
break;
case PMIX_COMPRESSED_STRING:
OPAL_ERROR_LOG(OPAL_ERR_NOT_SUPPORTED);
rc = OPAL_ERR_NOT_SUPPORTED;
break;
case PMIX_ALLOC_DIRECTIVE:
OPAL_ERROR_LOG(OPAL_ERR_NOT_SUPPORTED);
rc = OPAL_ERR_NOT_SUPPORTED;
break;
case PMIX_INFO_ARRAY:
OPAL_ERROR_LOG(OPAL_ERR_NOT_SUPPORTED);
rc = OPAL_ERR_NOT_SUPPORTED;
break;

default:
/* silence warnings */
opal_output(0, "VALUE UNLOAD NOT SUPPORTED FOR TYPE %d", v->type);
rc = OPAL_ERROR;
break;
}
Expand Down Expand Up @@ -1418,6 +1503,95 @@ opal_pmix_alloc_directive_t pmix2x_convert_allocdir(pmix_alloc_directive_t dir)
}
}

int pmix2x_convert_state(pmix_proc_state_t state)
{
switch(state) {
case PMIX_PROC_STATE_UNDEF:
return 0;
case PMIX_PROC_STATE_PREPPED:
case PMIX_PROC_STATE_LAUNCH_UNDERWAY:
return 1;
case PMIX_PROC_STATE_RESTART:
return 2;
case PMIX_PROC_STATE_TERMINATE:
return 3;
case PMIX_PROC_STATE_RUNNING:
return 4;
case PMIX_PROC_STATE_CONNECTED:
return 5;
case PMIX_PROC_STATE_UNTERMINATED:
return 15;
case PMIX_PROC_STATE_TERMINATED:
return 20;
case PMIX_PROC_STATE_KILLED_BY_CMD:
return 51;
case PMIX_PROC_STATE_ABORTED:
return 52;
case PMIX_PROC_STATE_FAILED_TO_START:
return 53;
case PMIX_PROC_STATE_ABORTED_BY_SIG:
return 54;
case PMIX_PROC_STATE_TERM_WO_SYNC:
return 55;
case PMIX_PROC_STATE_COMM_FAILED:
return 56;
case PMIX_PROC_STATE_CALLED_ABORT:
return 58;
case PMIX_PROC_STATE_MIGRATING:
return 60;
case PMIX_PROC_STATE_CANNOT_RESTART:
return 61;
case PMIX_PROC_STATE_TERM_NON_ZERO:
return 62;
case PMIX_PROC_STATE_FAILED_TO_LAUNCH:
return 63;
default:
return 0; // undef
}
}

pmix_proc_state_t pmix2x_convert_opalstate(int state)
{
switch(state) {
case 0:
return PMIX_PROC_STATE_UNDEF;
case 1:
return PMIX_PROC_STATE_LAUNCH_UNDERWAY;
case 2:
return PMIX_PROC_STATE_RESTART;
case 3:
return PMIX_PROC_STATE_TERMINATE;
case 4:
return PMIX_PROC_STATE_RUNNING;
case 5:
return PMIX_PROC_STATE_CONNECTED;
case 51:
return PMIX_PROC_STATE_KILLED_BY_CMD;
case 52:
return PMIX_PROC_STATE_ABORTED;
case 53:
return PMIX_PROC_STATE_FAILED_TO_START;
case 54:
return PMIX_PROC_STATE_ABORTED_BY_SIG;
case 55:
return PMIX_PROC_STATE_TERM_WO_SYNC;
case 56:
return PMIX_PROC_STATE_COMM_FAILED;
case 58:
return PMIX_PROC_STATE_CALLED_ABORT;
case 59:
return PMIX_PROC_STATE_MIGRATING;
case 61:
return PMIX_PROC_STATE_CANNOT_RESTART;
case 62:
return PMIX_PROC_STATE_TERM_NON_ZERO;
case 63:
return PMIX_PROC_STATE_FAILED_TO_LAUNCH;
default:
return PMIX_PROC_STATE_UNDEF;
}
}

/**** INSTANTIATE INTERNAL CLASSES ****/
OBJ_CLASS_INSTANCE(opal_pmix2x_jobid_trkr_t,
opal_list_item_t,
Expand Down
4 changes: 4 additions & 0 deletions opal/mca/pmix/pmix2x/pmix2x.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -339,6 +339,10 @@ OPAL_MODULE_DECLSPEC opal_pmix_alloc_directive_t pmix2x_convert_allocdir(pmix_al

OPAL_MODULE_DECLSPEC char* pmix2x_convert_jobid(opal_jobid_t jobid);

OPAL_MODULE_DECLSPEC int pmix2x_convert_state(pmix_proc_state_t state);

OPAL_MODULE_DECLSPEC pmix_proc_state_t pmix2x_convert_opalstate(int state);

END_C_DECLS

#endif /* MCA_PMIX_EXTERNAL_H */