Update to PMIx 3.0.1

.gitignore

@@ -228,6 +228,52 @@ ompi/mpi/fortran/use-mpi-tkr/mpi_kinds.ompi_module
 ompi/mpi/fortran/use-mpi-tkr/mpi-tkr-sizeof.f90
 ompi/mpi/fortran/use-mpi-tkr/mpi-tkr-sizeof.h
 
+ompi/mpiext/pcollreq/c/profile/pallgather_init.c
+ompi/mpiext/pcollreq/c/profile/pallgatherv_init.c
+ompi/mpiext/pcollreq/c/profile/pallreduce_init.c
+ompi/mpiext/pcollreq/c/profile/palltoall_init.c
+ompi/mpiext/pcollreq/c/profile/palltoallv_init.c
+ompi/mpiext/pcollreq/c/profile/palltoallw_init.c
+ompi/mpiext/pcollreq/c/profile/pbarrier_init.c
+ompi/mpiext/pcollreq/c/profile/pbcast_init.c
+ompi/mpiext/pcollreq/c/profile/pexscan_init.c
+ompi/mpiext/pcollreq/c/profile/pgather_init.c
+ompi/mpiext/pcollreq/c/profile/pgatherv_init.c
+ompi/mpiext/pcollreq/c/profile/pmpiext_pcollreq_c.h
+ompi/mpiext/pcollreq/c/profile/pneighbor_allgather_init.c
+ompi/mpiext/pcollreq/c/profile/pneighbor_allgatherv_init.c
+ompi/mpiext/pcollreq/c/profile/pneighbor_alltoall_init.c
+ompi/mpiext/pcollreq/c/profile/pneighbor_alltoallv_init.c
+ompi/mpiext/pcollreq/c/profile/pneighbor_alltoallw_init.c
+ompi/mpiext/pcollreq/c/profile/preduce_init.c
+ompi/mpiext/pcollreq/c/profile/preduce_scatter_block_init.c
+ompi/mpiext/pcollreq/c/profile/preduce_scatter_init.c
+ompi/mpiext/pcollreq/c/profile/pscan_init.c
+ompi/mpiext/pcollreq/c/profile/pscatter_init.c
+ompi/mpiext/pcollreq/c/profile/pscatterv_init.c
+ompi/mpiext/pcollreq/mpif-h/profile/pallgather_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pallgatherv_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pallreduce_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/palltoall_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/palltoallv_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/palltoallw_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pbarrier_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pbcast_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pexscan_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pgather_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pgatherv_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pneighbor_allgather_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pneighbor_allgatherv_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pneighbor_alltoall_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pneighbor_alltoallv_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pneighbor_alltoallw_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/preduce_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/preduce_scatter_block_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/preduce_scatter_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pscan_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pscatter_init_f.c
+ompi/mpiext/pcollreq/mpif-h/profile/pscatterv_init_f.c
+
 ompi/mpi/java/java/mpi
 ompi/mpi/java/java/*.jar
 ompi/mpi/java/java/*.h

opal/include/opal/constants.h

@@ -98,7 +98,8 @@ enum {
     OPAL_ERR_HEARTBEAT_ALERT                = (OPAL_ERR_BASE - 67),
     OPAL_ERR_FILE_ALERT                     = (OPAL_ERR_BASE - 68),
     OPAL_ERR_MODEL_DECLARED                 = (OPAL_ERR_BASE - 69),
-    OPAL_PMIX_LAUNCH_DIRECTIVE              = (OPAL_ERR_BASE - 70)
+    OPAL_PMIX_LAUNCH_DIRECTIVE              = (OPAL_ERR_BASE - 70),
+    OPAL_OPERATION_SUCCEEDED                = (OPAL_ERR_BASE - 71)
 };
 
 #define OPAL_ERR_MAX                (OPAL_ERR_BASE - 100)

opal/mca/pmix/ext1x/pmix1x.c

@@ -185,6 +185,7 @@ pmix_status_t pmix1_convert_opalrc(int rc)
     case OPAL_ERROR:
         return PMIX_ERROR;
     case OPAL_SUCCESS:
+    case OPAL_OPERATION_SUCCEEDED:
         return PMIX_SUCCESS;
     default:
         return PMIX_ERROR;

opal/mca/pmix/ext2x/ext2x.c

@@ -440,6 +440,7 @@ pmix_status_t ext2x_convert_opalrc(int rc)
     case OPAL_ERROR:
         return PMIX_ERROR;
     case OPAL_SUCCESS:
+    case OPAL_OPERATION_SUCCEEDED:
         return PMIX_SUCCESS;
     default:
         return rc;

opal/mca/pmix/pmix3x/pmix/NEWS

@@ -21,7 +21,21 @@ example, a bug might be fixed in the master, and then moved to the
 current release as well as the "stable" bug fix release branch.
 
 
-3.0.0 -- TBD
+3.0.1 -- 23 Aug 2018
+----------------------
+**** DEPRECATION WARNING: The pmix_info_array_t struct was
+**** initially marked for deprecation in the v2.x series.
+**** We failed to provide clear warning at that time. This
+**** therefore serves as warning of intended removal of
+**** pmix_info_array_t in the future v4 release series.
+- Fixed memory corruption bug in event notification
+  system due to uninitialized variable
+- Add numeric version field to pmix_version.h
+- Transfer all cached data to client dstore upon first connect
+- Implement missing job control and sensor APIs
+
+
+3.0.0 -- 6 July 2018
 ------------------------------------
 **** NOTE: This release implements the complete PMIX v3.0 Standard
 **** and therefore includes a number of new APIs and features. These
@@ -63,7 +77,15 @@ current release as well as the "stable" bug fix release branch.
 - Fix several memory and file descriptor leaks
 
 
-2.1.2 -- TBD
+2.1.3 -- 23 Aug 2018
+----------------------
+- Fixed memory corruption bug in event notification
+  system due to uninitialized variable
+- Add numeric version definition
+- Transfer all cached data to client dstore upon first connect
+
+
+2.1.2 -- 6 July 2018
 ----------------------
 - Added PMIX_VERSION_RELEASE string to pmix_version.h
 - Added PMIX_SPAWNED and PMIX_PARENT_ID keys to all procs

opal/mca/pmix/pmix3x/pmix/VERSION

@@ -15,7 +15,7 @@
 
 major=3
 minor=0
-release=0
+release=1
 
 # greek is used for alpha or beta release tags.  If it is non-empty,
 # it will be appended to the version number.  It does not have to be
@@ -30,7 +30,7 @@ greek=
 # command, or with the date (if "git describe" fails) in the form of
 # "date<date>".
 
-repo_rev=gitffba520
+repo_rev=gitbf30a5f
 
 # If tarball_version is not empty, it is used as the version string in
 # the tarball filename, regardless of all other versions listed in
@@ -44,7 +44,7 @@ tarball_version=
 
 # The date when this release was created
 
-date="Jul 01, 2018"
+date="Aug 20, 2018"
 
 # The shared library version of each of PMIx's public libraries.
 # These versions are maintained in accordance with the "Library
@@ -75,6 +75,6 @@ date="Jul 01, 2018"
 # Version numbers are described in the Libtool current:revision:age
 # format.
 
-libpmix_so_version=4:0:2
+libpmix_so_version=4:1:2
 libpmi_so_version=1:0:0
 libpmi2_so_version=1:0:0

opal/mca/pmix/pmix3x/pmix/contrib/pmix.spec

@@ -192,7 +192,7 @@
 
 Summary: An extended/exascale implementation of PMI
 Name: %{?_name:%{_name}}%{!?_name:pmix}
-Version: 3.0.0
+Version: 3.0.1
 Release: 1%{?dist}
 License: BSD
 Group: Development/Libraries

opal/mca/pmix/pmix3x/pmix/include/pmix_common.h.in

@@ -781,6 +781,7 @@ typedef int pmix_status_t;
 #define PMIX_LAUNCH_DIRECTIVE                   (PMIX_ERR_OP_BASE - 24)
 #define PMIX_LAUNCHER_READY                     (PMIX_ERR_OP_BASE - 25)
 #define PMIX_OPERATION_IN_PROGRESS              (PMIX_ERR_OP_BASE - 26)
+#define PMIX_OPERATION_SUCCEEDED                (PMIX_ERR_OP_BASE - 27)
 
 
 /* define a starting point for system error constants so

opal/mca/pmix/pmix3x/pmix/include/pmix_version.h.in

@@ -2,6 +2,7 @@
  * Copyright (c) 2016      Mellanox Technologies, Inc.
  *                         All rights reserved.
  * Copyright (c) 2018      IBM Corporation.  All rights reserved.
+ * Copyright (c) 2018      Intel, Inc. All rights reserved.
  * $COPYRIGHT$
  *
  * Additional copyrights may follow
@@ -17,4 +18,6 @@
 #define PMIX_VERSION_MAJOR @pmixmajor@
 #define PMIX_VERSION_MINOR @pmixminor@
 #define PMIX_VERSION_RELEASE @pmixrelease@
+
+#define PMIX_NUMERIC_VERSION 0x00030001
 #endif

opal/mca/pmix/pmix3x/pmix/src/common/pmix_control.c

@@ -85,7 +85,7 @@ static void query_cbfunc(struct pmix_peer_t *peer,
     /* unpack any returned data */
     cnt = 1;
     PMIX_BFROPS_UNPACK(rc, peer, buf, &results->ninfo, &cnt, PMIX_SIZE);
-    if (PMIX_SUCCESS != rc) {
+    if (PMIX_SUCCESS != rc && PMIX_ERR_UNPACK_READ_PAST_END_OF_BUFFER != rc) {
         PMIX_ERROR_LOG(rc);
         goto complete;
     }
@@ -332,6 +332,12 @@ PMIX_EXPORT pmix_status_t PMIx_Process_monitor_nb(const pmix_info_t *monitor, pm
         return PMIX_ERR_INIT;
     }
 
+    /* sanity check */
+    if (NULL == monitor) {
+        PMIX_RELEASE_THREAD(&pmix_global_lock);
+        return PMIX_ERR_BAD_PARAM;
+    }
+
     /* if we are the server, then we just issue the request and
      * return the response */
     if (PMIX_PROC_IS_SERVER(pmix_globals.mypeer) &&
@@ -355,6 +361,19 @@ PMIX_EXPORT pmix_status_t PMIx_Process_monitor_nb(const pmix_info_t *monitor, pm
     }
     PMIX_RELEASE_THREAD(&pmix_global_lock);
 
+    /* if the monitor is PMIX_SEND_HEARTBEAT, then send it */
+    if (0 == strncmp(monitor->key, PMIX_SEND_HEARTBEAT, PMIX_MAX_KEYLEN)) {
+        msg = PMIX_NEW(pmix_buffer_t);
+        if (NULL == msg) {
+            return PMIX_ERR_NOMEM;
+        }
+        PMIX_PTL_SEND_ONEWAY(rc, pmix_client_globals.myserver, msg, PMIX_PTL_TAG_HEARTBEAT);
+        if (PMIX_SUCCESS != rc) {
+            PMIX_RELEASE(msg);
+        }
+        return rc;
+    }
+
     /* if we are a client, then relay this request to the server */
     msg = PMIX_NEW(pmix_buffer_t);
     /* pack the cmd */

opal/mca/pmix/pmix3x/pmix/src/mca/gds/ds12/gds_dstore.c

@@ -1,6 +1,6 @@
 /*
  * Copyright (c) 2015-2018 Intel, Inc. All rights reserved.
- * Copyright (c) 2016      IBM Corporation.  All rights reserved.
+ * Copyright (c) 2016-2018 IBM Corporation.  All rights reserved.
  * Copyright (c) 2016-2017 Mellanox Technologies, Inc.
  *                         All rights reserved.
  * Copyright (c) 2018      Research Organization for Information Science
@@ -3170,7 +3170,7 @@ static pmix_status_t dstore_register_job_info(struct pmix_peer_t *pr,
     char *msg;
     pmix_status_t rc;
     pmix_proc_t proc;
-    pmix_rank_info_t *rinfo;
+    pmix_rank_t rank;
 
     pmix_output_verbose(2, pmix_gds_base_framework.framework_output,
                         "[%s:%d] gds:dstore:register_job_info for peer [%s:%d]",
@@ -3187,8 +3187,8 @@ static pmix_status_t dstore_register_job_info(struct pmix_peer_t *pr,
             return rc;
         }
 
-        PMIX_LIST_FOREACH(rinfo, &ns->ranks, pmix_rank_info_t) {
-            proc.rank = rinfo->pname.rank;
+        for (rank=0; rank < ns->nprocs; rank++) {
+            proc.rank = rank;
             rc = _store_job_info(&proc);
             if (PMIX_SUCCESS != rc) {
                 PMIX_ERROR_LOG(rc);

opal/mca/pmix/pmix3x/pmix/src/mca/psensor/base/psensor_base_stubs.c

@@ -1,7 +1,7 @@
 /*
  * Copyright (c) 2010      Cisco Systems, Inc.  All rights reserved.
  * Copyright (c) 2012      Los Alamos National Security, Inc. All rights reserved.
- * Copyright (c) 2014-2017 Intel, Inc.  All rights reserved.
+ * Copyright (c) 2014-2018 Intel, Inc.  All rights reserved.
  *
  * $COPYRIGHT$
  *
@@ -24,6 +24,7 @@ pmix_status_t pmix_psensor_base_start(pmix_peer_t *requestor, pmix_status_t erro
 {
     pmix_psensor_active_module_t *mod;
     pmix_status_t rc;
+    bool didit = false;
 
     pmix_output_verbose(5, pmix_psensor_base_framework.framework_output,
                         "%s:%d sensor:base: starting sensors",
@@ -36,17 +37,25 @@ pmix_status_t pmix_psensor_base_start(pmix_peer_t *requestor, pmix_status_t erro
             if (PMIX_SUCCESS != rc && PMIX_ERR_TAKE_NEXT_OPTION != rc) {
                 return rc;
             }
+            didit = true;
         }
     }
 
+    /* if none of the components could do it, then report
+     * not supported upwards so the server knows to ask
+     * the host to try */
+    if (!didit) {
+        return PMIX_ERR_NOT_SUPPORTED;
+    }
+
     return PMIX_SUCCESS;
 }
 
 pmix_status_t pmix_psensor_base_stop(pmix_peer_t *requestor,
                                      char *id)
 {
     pmix_psensor_active_module_t *mod;
-    pmix_status_t rc;
+    pmix_status_t rc, ret = PMIX_SUCCESS;
 
     pmix_output_verbose(5, pmix_psensor_base_framework.framework_output,
                         "%s:%d sensor:base: stopping sensors",
@@ -57,10 +66,14 @@ pmix_status_t pmix_psensor_base_stop(pmix_peer_t *requestor,
         if (NULL != mod->module->stop) {
             rc = mod->module->stop(requestor, id);
             if (PMIX_SUCCESS != rc && PMIX_ERR_TAKE_NEXT_OPTION != rc) {
-                return rc;
+                if (PMIX_SUCCESS == ret) {
+                    ret = rc;
+                }
+                /* need to continue to ensure that all
+                 * sensors have been stopped */
             }
         }
     }
 
-    return PMIX_SUCCESS;
+    return ret;
 }

opal/mca/pmix/pmix3x/pmix/src/mca/psensor/heartbeat/psensor_heartbeat.c

@@ -3,7 +3,7 @@
  * Copyright (c) 2011-2012 Los Alamos National Security, LLC.  All rights
  *                         reserved.
   *
- * Copyright (c) 2017      Intel, Inc.  All rights reserved.
+ * Copyright (c) 2017-2018 Intel, Inc.  All rights reserved.
  * $COPYRIGHT$
  *
  * Additional copyrights may follow
@@ -30,7 +30,7 @@
 #include "src/util/output.h"
 #include "src/util/show_help.h"
 #include "src/include/pmix_globals.h"
-#include "src/mca/ptl/ptl.h"
+#include "src/mca/ptl/base/base.h"
 
 #include "src/mca/psensor/base/base.h"
 #include "psensor_heartbeat.h"
@@ -168,6 +168,7 @@ static pmix_status_t heartbeat_start(pmix_peer_t *requestor, pmix_status_t error
 {
     pmix_heartbeat_trkr_t *ft;
     size_t n;
+    pmix_ptl_posted_recv_t *rcv;
 
     PMIX_OUTPUT_VERBOSE((1, pmix_psensor_base_framework.framework_output,
                          "[%s:%d] checking heartbeat monitoring for requestor %s:%d",
@@ -202,6 +203,17 @@ static pmix_status_t heartbeat_start(pmix_peer_t *requestor, pmix_status_t error
         return PMIX_ERR_BAD_PARAM;
     }
 
+    /* if the recv hasn't been posted, so so now */
+    if (!mca_psensor_heartbeat_component.recv_active) {
+        /* setup to receive heartbeats */
+        rcv = PMIX_NEW(pmix_ptl_posted_recv_t);
+        rcv->tag = PMIX_PTL_TAG_HEARTBEAT;
+        rcv->cbfunc = pmix_psensor_heartbeat_recv_beats;
+        /* add it to the beginning of the list of recvs */
+        pmix_list_prepend(&pmix_ptl_globals.posted_recvs, &rcv->super);
+        mca_psensor_heartbeat_component.recv_active = true;
+    }
+
     /* need to push into our event base to add this to our trackers */
     pmix_event_assign(&ft->cdev, pmix_psensor_base.evbase, -1,
                       EV_WRITE, add_tracker, ft);
@@ -241,7 +253,7 @@ static pmix_status_t heartbeat_stop(pmix_peer_t *requestor, char *id)
     cd->requestor = requestor;
     cd->id = strdup(id);
 
-    /* need to push into our event base to add this to our trackers */
+    /* need to push into our event base to remove this from our trackers */
     pmix_event_assign(&cd->ev, pmix_psensor_base.evbase, -1,
                       EV_WRITE, del_tracker, cd);
     PMIX_POST_OBJECT(cd);

opal/mca/pmix/pmix3x/pmix/src/mca/psensor/heartbeat/psensor_heartbeat.h

@@ -2,7 +2,7 @@
  * Copyright (c) 2010      Cisco Systems, Inc.  All rights reserved.
  * Copyright (c) 2012      Los Alamos National Security, Inc. All rights reserved.
  *
- * Copyright (c) 2017      Intel, Inc.  All rights reserved.
+ * Copyright (c) 2017-2018 Intel, Inc.  All rights reserved.
  * $COPYRIGHT$
  *
  * Additional copyrights may follow
@@ -28,6 +28,7 @@ BEGIN_C_DECLS
 
 typedef struct {
     pmix_psensor_base_component_t super;
+    bool recv_active;
     pmix_list_t trackers;
 } pmix_psensor_heartbeat_component_t;