Force checkstops for unhandled machine checks

Default MSR[ME]=0 during initial boot for bootloader and
 hostboot kernel
Once the xscom address range has been mapped in, enable the
 machine check handler to force a checkstop and set MSR[ME]=1
 to allow regular machine check handling

CQ: SW401402
Change-Id: I104e39465e61b3b19d5c073e71271102711ae54f
Reviewed-on: http://ralgit01.raleigh.ibm.com/gerrit1/47179
Reviewed-by: Christian R. Geddes <crgeddes@us.ibm.com>
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: William G. Hoffa <wghoffa@us.ibm.com>
Reviewed-by: Daniel M. Crowell <dcrowell@us.ibm.com>

1 files changed, 45 insertions, 0 deletions

diff --git a/src/kernel/machchk.C b/src/kernel/machchk.C
index 79edf4240..2a96b5896 100644
--- a/src/kernel/machchk.C
+++ b/src/kernel/machchk.C
@@ -27,12 +27,20 @@
 #include <kernel/vmmmgr.H>
 #include <sys/mmio.h>
 #include <arch/memorymap.H>
+#include <arch/ppc.H>
 
 namespace Kernel
 {
 namespace MachineCheck
 {
 
+//Keep track of the MMIO address that we can use to force a checkstop
+static uint64_t* g_xstopRegPtr = nullptr;
+
+//Keep track of the data to write into the xstop reg
+static uint64_t g_xstopRegValue = 0;
+
+
 bool handleLoadUE(task_t* t)
 {
     bool handled = false;
@@ -132,5 +140,42 @@ bool handleSLB(task_t* t)
 }
 
 
+/**
+ *  @brief Tells the kernel how to force a checkstop for unrecoverable
+ *         machine checks
+ */
+void setCheckstopData(uint64_t i_xstopAddr, uint64_t i_xstopData)
+{
+    g_xstopRegPtr = reinterpret_cast<uint64_t*>(i_xstopAddr
+                                                |VmmManager::FORCE_PHYS_ADDR);
+    g_xstopRegValue = i_xstopData;
+    printk( "Set MchChk Xstop: %p=%.16lX\n", g_xstopRegPtr, g_xstopRegValue );
+
+    // Now that the machine check handler can do the xscom we
+    //  can set MSR[ME]=1 to enable the regular machine check
+    //  handling
+    uint64_t l_msr = getMSR();
+    l_msr |= 0x0000000000001000; //set bit 51
+    setMSR(l_msr);
+}
+
+/**
+ *  @brief Force a checkstop if we know how in order to get better
+ *         error isolation for cache/memory UEs
+ */
+void forceCheckstop()
+{
+    if( g_xstopRegPtr != nullptr )
+    {
+        printk( "Forcing a xstop with %p = %.16lX\n",
+                g_xstopRegPtr, g_xstopRegValue );
+        *g_xstopRegPtr = g_xstopRegValue;
+    }
+    else
+    {
+        printk( "Unable to force checkstop, No xstop reg set\n" );
+    }
+}
+
 }
 }