This commit fixes 2370, by having the solaris paffinity module return error codes for get_physical_processor_id and having odls_default_fork_local_proc check get_physical_processor_id for OPAL_ERROR

This commit was SVN r22948.

opal/mca/paffinity/solaris/paffinity_solaris_module.c

@@ -10,7 +10,7 @@
  * Copyright (c) 2004-2005 The Regents of the University of California.
  *                         All rights reserved.
  * Copyright (c) 2007      Cisco Systems, Inc.  All rights reserved.
- * Copyright (c) 2008      Sun Microsystems, Inc.  All rights reserved.
+ * Copyright (c) 2008-2010 Oracle and/or its affiliates.  All rights reserved.
  *
  * $COPYRIGHT$
  * 
@@ -197,13 +197,17 @@ static int solaris_module_get_core_info(int socket, int *num_cores)
 
 static int solaris_module_get_physical_processor_id(int logical_processor_id)
 {
-    processorid_t currid, cpuid_max, cpuid_log=0;
+    processorid_t currid, retid, cpuid_max, cpuid_log=0;
     processor_info_t pinfo;
 
     /* cpuid_max is the max number available for a system arch. It is
      * an inclusive list. e.g. If cpuid_max=31, cpuid would be 0-31 */
-    cpuid_max = sysconf(_SC_CPUID_MAX);
+    if ( 0 > (cpuid_max = sysconf(_SC_CPUID_MAX))) {
+	return OPAL_ERR_NOT_SUPPORTED;
+    }
 
+    /* set retid to OPAL_ERROR to reflect no processor found to match logical proc */
+    retid = OPAL_ERROR;
     /* Because not all CPU ID in cpuid_max are actually valid,
      * and CPU ID may also not be contiguous. Therefore we
      * need to run through processor_info to ensure the validity.
@@ -212,13 +216,15 @@ static int solaris_module_get_physical_processor_id(int logical_processor_id)
         if (0 == processor_info(currid, &pinfo)) {
             if (P_ONLINE == pinfo.pi_state || P_NOINTR == pinfo.pi_state) {
                  if (cpuid_log == logical_processor_id) {
+		     retid = currid;
                      break;
                  } 
                  cpuid_log++;
             }
         }
     }
-    return currid;
+    
+    return retid;
 }
 
 static int solaris_module_get_physical_socket_id(int logical_socket_id)

orte/mca/odls/default/odls_default_module.c

@@ -527,22 +527,20 @@ static int odls_default_fork_local_proc(orte_app_context_t* context,
                              * to us, so index into the node's array to get the
                              * physical cpu
                              */
-			    if (logical_cpu < orte_odls_globals.num_processors) {
-				/* We have a processor to bind to */
-				phys_cpu = opal_paffinity_base_get_physical_processor_id(logical_cpu);
-				if (0 > phys_cpu) {
-				    ORTE_ODLS_IF_BIND_NOT_REQD("bind-to-core");
-				    orte_show_help("help-odls-default.txt",
-						   "odls-default:invalid-phys-cpu", true);
-				    ORTE_ODLS_ERROR_OUT(ORTE_ERR_FATAL);
-				}
-			    } else {
+			    phys_cpu = opal_paffinity_base_get_physical_processor_id(logical_cpu);
+			    if (OPAL_ERROR == phys_cpu){
 				/* No processor to bind to so error out */
+				ORTE_ODLS_IF_BIND_NOT_REQD("bind-to-core");
 				orte_show_help("help-odls-default.txt",
 					       "odls-default:not-enough-resources", true,
 					       "processors", orte_process_info.nodename,
 					       "bind-to-core", context->app);
 				ORTE_ODLS_ERROR_OUT(ORTE_ERR_FATAL);
+			    } else if (0 > phys_cpu) {
+				ORTE_ODLS_IF_BIND_NOT_REQD("bind-to-core");
+				orte_show_help("help-odls-default.txt",
+					       "odls-default:invalid-phys-cpu", true);
+				ORTE_ODLS_ERROR_OUT(ORTE_ERR_FATAL);
 			    }
                         }
                         OPAL_PAFFINITY_CPU_SET(phys_cpu, mask);