Added code to support doing an NVM health check

Added a method that will do an NVM (non-volatile memory)
health check.  In particular this method will check the
flash error counts and does a predictive callout
if the number of flash error counts exceeds the maximum
allowed.  This method also checks the bad block percentage
and does a predictive callout if the number of bad block
percentage exceeds the maximum allowed.  A predictive
callout is done if either or both fail the check.

Added support in the runtime commands to make the nvm health check
call as well.

Also, when I did the ES (energy source) health check method, I was
not very explicit in the method that it was doing an ES health
check.  So I updated the verbiage in the nvDimmCheckHealthStatus
to add ES wherever appropriate as to make these two method
explicit as to what health check is being performed.

Change-Id: Ib9925fd2bb8430cf2121108329247d96072beb1b
CQ: 473220
Reviewed-on: http://rchgit01.rchland.ibm.com/gerrit1/82843
Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com>
Tested-by: Jenkins OP Build CI <op-jenkins+hostboot@us.ibm.com>
Tested-by: Jenkins OP HW <op-hw-jenkins+hostboot@us.ibm.com>
Tested-by: FSP CI Jenkins <fsp-CI-jenkins+hostboot@us.ibm.com>
Reviewed-by: Glenn Miles <milesg@ibm.com>
Reviewed-by: Zachary Clark <zach@ibm.com>
Reviewed-by: Christian R Geddes <crgeddes@us.ibm.com>

1 files changed, 62 insertions, 12 deletions

diff --git a/src/usr/util/runtime/rt_cmds.C b/src/usr/util/runtime/rt_cmds.C
index c669aae4f..bf0c51749 100644
--- a/src/usr/util/runtime/rt_cmds.C
+++ b/src/usr/util/runtime/rt_cmds.C
@@ -1179,25 +1179,59 @@ void cmd_nvdimm_protection_msg( char* &o_output, uint32_t i_huid,
     }
 }
 
-void cmd_nvdimmCheckHealthStatus( char* &o_output)
+/**
+ * @brief Check the ES (energy source) health status of all NVDIMMs in the
+ *         system. If check fails, see HBRT traces for further details.
+ * @param[out] o_output  Output display buffer, memory allocated here.
+ *                       Will inform caller if ES health check passes or fails.
+ */
+void cmd_nvDimmEsCheckHealthStatus( char* &o_output)
+{
+    o_output = new char[500];
+    if (NVDIMM::nvDimmEsCheckHealthStatusOnSystem())
+    {
+        sprintf( o_output, "cmd_nvDimmEsCheckHealthStatus: "
+                           "ES (energy source) health status check passed.");
+
+    }
+    else
+    {
+        sprintf( o_output, "cmd_nvDimmEsCheckHealthStatus: "
+                           "ES (energy source) health status check failed. "
+                           "Inspect HBRT traces for further details.");
+
+    }
+
+    return;
+}  // end cmd_nvDimmEsCheckHealthStatus
+
+/**
+ * @brief Check the NVM (non-volatile memory) health status of all NVDIMMs in
+ *         the system. If check fails, see HBRT traces for further details.
+ * @param[out] o_output  Output display buffer, memory allocated here.
+ *                       Will inform caller if NVM health check passes or fails.
+ */
+
+void cmd_nvdDmmNvmCheckHealthStatus( char* &o_output)
 {
     o_output = new char[500];
-    if (NVDIMM::nvDimmCheckHealthStatusOnSystem())
+    if (NVDIMM::nvDimmNvmCheckHealthStatusOnSystem())
     {
-        sprintf( o_output, "cmd_doNvDimmCheckHealthStatus: "
-                           "health status check passed.");
+        sprintf( o_output, "cmd_nvdDmmNvmCheckHealthStatus: "
+                       "NVM (non-volatile memory) health status check passed.");
 
     }
     else
     {
-        sprintf( o_output, "cmd_doNvDimmCheckHealthStatus: "
-                           "health status check failed. Inspect HBRT traces "
-                           "for further details.");
+        sprintf( o_output, "cmd_nvdDmmNvmCheckHealthStatus: "
+                       "NVM (non-volatile memory) health status check failed. "
+                       "Inspect HBRT traces for further details.");
 
     }
 
     return;
-}  // end cmd_nvdimmCheckHealthStatus
+}  // end cmd_nvdDmmNvmCheckHealthStatus
+
 
 #endif
 
@@ -1535,18 +1569,31 @@ int hbrtCommand( int argc,
             sprintf(*l_output, "ERROR: nvdimm_protection <huid> <0 or 1>");
         }
     }
-    else if( !strcmp( argv[0], "nvdimm_check_status" ) )
+    else if( !strcmp( argv[0], "nvdimm_es_check_status" ) )
     {
         if (argc == 1)
         {
-            cmd_nvdimmCheckHealthStatus( *l_output );
+            cmd_nvDimmEsCheckHealthStatus( *l_output );
         }
         else
         {
             *l_output = new char[100];
-            sprintf(*l_output, "Usage: nvdimm_check_status");
+            sprintf(*l_output, "Usage: nvdimm_es_check_status");
         }
     }
+    else if( !strcmp( argv[0], "nvdimm_nvm_check_status" ) )
+    {
+        if (argc == 1)
+        {
+            cmd_nvdDmmNvmCheckHealthStatus( *l_output );
+        }
+        else
+        {
+            *l_output = new char[100];
+            sprintf(*l_output, "Usage: nvdimm_nvm_check_status");
+        }
+    }
+
 #endif
     else
     {
@@ -1587,8 +1634,11 @@ int hbrtCommand( int argc,
 #ifdef CONFIG_NVDIMM
         sprintf( l_tmpstr, "nvdimm_protection <huid> <0 or 1>\n");
         strcat( *l_output, l_tmpstr );
-        sprintf( l_tmpstr, "nvdimm_check_status\n");
+        sprintf( l_tmpstr, "nvdimm_es_check_status\n");
+        strcat( *l_output, l_tmpstr );
+        sprintf( l_tmpstr, "nvdimm_nvm_check_status\n");
         strcat( *l_output, l_tmpstr );
+
 #endif
 
     }