sharedfp/lockedfile and sm: fix the namecollision

this fixes the issue reported by Nicolas Joly on the mailing: the sharedfp/lockedfile component does not support right now a scenario where multiple jobs read from the same input file, due to a collision of the filenames utilized for the sharedfp handle. Although not part of the oroginal report, the same occurs for the sharedfp/sm component. Add therefore the jobid to be part of the lockedfilename/sm file name.

Fixes: #3098

Signed-off-by: Edgar Gabriel <[email protected]>

Author: edgargabriel

ompi/mca/sharedfp/lockedfile/sharedfp_lockedfile_file_open.c

@@ -9,7 +9,7 @@
  *                         University of Stuttgart.  All rights reserved.
  * Copyright (c) 2004-2005 The Regents of the University of California.
  *                         All rights reserved.
- * Copyright (c) 2013-2016 University of Houston. All rights reserved.
+ * Copyright (c) 2013-2017 University of Houston. All rights reserved.
  * Copyright (c) 2015      Research Organization for Information Science
  *                         and Technology (RIST). All rights reserved.
  * $COPYRIGHT$
@@ -25,6 +25,8 @@
 
 #include "mpi.h"
 #include "ompi/constants.h"
+#include "ompi/group/group.h"
+#include "ompi/proc/proc.h"
 #include "ompi/mca/sharedfp/sharedfp.h"
 #include "ompi/mca/sharedfp/base/base.h"
 
@@ -99,8 +101,19 @@ int mca_sharedfp_lockedfile_file_open (struct ompi_communicator_t *comm,
         return OMPI_ERR_OUT_OF_RESOURCE;
     }
 
-    lockedfilename = (char*)malloc(sizeof(char) * (strlen(filename) + 64));
-    sprintf(lockedfilename,"%s%s",filename,".lockedfile");
+    ompi_proc_t *masterproc = ompi_group_peer_lookup(comm->c_local_group, 0 );
+    opal_process_name_t *mastername = &(masterproc->super.proc_name);
+    opal_jobid_t masterjobid = mastername->jobid;
+ 
+    size_t filenamelen = strlen(filename) + 16;
+    lockedfilename = (char*)malloc(sizeof(char) * filenamelen);
+    if ( NULL == lockedfilename ) {
+	free (shfileHandle);
+	free (sh);
+        free (module_data);
+        return OMPI_ERR_OUT_OF_RESOURCE;
+    }
+    snprintf(lockedfilename, filenamelen, "%s-%u%s",filename,masterjobid,".lock");
     module_data->filename = lockedfilename;
 
     /*-------------------------------------------------*/

ompi/mca/sharedfp/sm/sharedfp_sm_file_open.c

@@ -9,7 +9,7 @@
  *                         University of Stuttgart.  All rights reserved.
  * Copyright (c) 2004-2005 The Regents of the University of California.
  *                         All rights reserved.
- * Copyright (c) 2013-2016 University of Houston. All rights reserved.
+ * Copyright (c) 2013-2017 University of Houston. All rights reserved.
  * Copyright (c) 2013      Intel, Inc. All rights reserved.
  * Copyright (c) 2015      Research Organization for Information Science
  *                         and Technology (RIST). All rights reserved.
@@ -35,6 +35,8 @@
 
 #include "mpi.h"
 #include "ompi/constants.h"
+#include "ompi/group/group.h"
+#include "ompi/proc/proc.h"
 #include "ompi/mca/sharedfp/sharedfp.h"
 #include "ompi/mca/sharedfp/base/base.h"
 
@@ -139,8 +141,11 @@ int mca_sharedfp_sm_file_open (struct ompi_communicator_t *comm,
         free(shfileHandle);
         return OMPI_ERR_OUT_OF_RESOURCE;
     }
-    sprintf(sm_filename,"/tmp/OMPIO_sharedfp_sm_%s%s",filename_basename,".sm");
+    ompi_proc_t *masterproc = ompi_group_peer_lookup(comm->c_local_group, 0 );
+    opal_process_name_t *mastername = &(masterproc->super.proc_name);
+    opal_jobid_t masterjobid = mastername->jobid;
 
+    sprintf(sm_filename,"/tmp/OMPIO_%s_%d_%s",filename_basename, masterjobid, ".sm");
     /* open shared memory file, initialize to 0, map into memory */
     sm_fd = open(sm_filename, O_RDWR | O_CREAT,
                  S_IRUSR | S_IWUSR | S_IRGRP | S_IROTH);