Tempopary fix to fail MPIPL if Cache is not scommable

Fail mpipl if any of the L2/L3 cache is not scommable, because of which
sbe is going to skip cache purge. This is keep to the system in fail
state so that data can be captured from registers to understand the
system behaviour for Defect SW423680

Change-Id: Ieb9063f1f25fff0059514e58de0e4bb1d04fe400
Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/57708
Tested-by: Jenkins Server <pfd-jenkins+hostboot@us.ibm.com>
Reviewed-by: Sachin Gupta <sgupta2m@in.ibm.com>

src/sbefw/app/power/ipl_table.C

@@ -815,6 +815,10 @@ ReturnCode istepWithExL2Flush( voidfuncptr_t i_hwp)
             // Enable this code back once stop states are enabled
             // This is temporary hack to debug SW422447
             // continue;
+            SBE_ERROR(SBE_FUNC "Ex chipletId [%d] not l2 scomable, so no purge",
+                exTgt.getChipletNumber());
+            rc = fapi2::FAPI2_RC_PLAT_ERR_SEE_DATA | 0x11;
+            break;
         }
         p9core::purgeData_t l_purgeData;
         SBE_EXEC_HWP(rc,
@@ -872,6 +876,10 @@ ReturnCode istepWithExL3Flush( voidfuncptr_t i_hwp)
             // Enable this code back once stop states are enabled
             // This is temporary hack to debug SW422447
             // continue;
+            SBE_INFO(SBE_FUNC "Ex chipletId [%d] not l2 scomable, so no purge",
+                exTgt.getChipletNumber());
+            rc = fapi2::FAPI2_RC_PLAT_ERR_SEE_DATA | 0x12;
+            break;
         }
 
         SBE_EXEC_HWP(rc, reinterpret_cast<sbeIstepHwpExL3Flush_t>(i_hwp),