LSQ: Only trigger a memory violation with a load/load if the value changes.

Only create a memory ordering violation when the value could have changed
between two subsequent loads, instead of just when loads go out-of-order
to the same address. While not very common in the case of Alpha, with
an architecture with a hardware table walker this can happen reasonably
frequently beacuse a translation will miss and start a table walk and
before the CPU re-schedules the faulting instruction another one will
pass it to the same address (or cache block depending on the dendency
checking).

This patch has been tested with a couple of self-checking hand crafted
programs to stress ordering between two cores.

The performance improvement on SPEC benchmarks can be substantial (2-10%).

2 files changed, 16 insertions, 0 deletions

diff --git a/src/sim/faults.cc b/src/sim/faults.cc
index 3f1369bcc..bd05df834 100644
--- a/src/sim/faults.cc
+++ b/src/sim/faults.cc
@@ -56,6 +56,12 @@ void UnimpFault::invoke(ThreadContext * tc, StaticInstPtr inst)
     panic("Unimpfault: %s\n", panicStr.c_str());
 }
 
+void ReExec::invoke(ThreadContext *tc, StaticInstPtr inst)
+{
+    tc->pcState(tc->pcState());
+}
+
+
 #if !FULL_SYSTEM
 void GenericPageTableFault::invoke(ThreadContext *tc, StaticInstPtr inst)
 {
diff --git a/src/sim/faults.hh b/src/sim/faults.hh
index 7f431b313..f5e9aded1 100644
--- a/src/sim/faults.hh
+++ b/src/sim/faults.hh
@@ -75,6 +75,16 @@ class UnimpFault : public FaultBase
             StaticInstPtr inst = StaticInst::nullStaticInstPtr);
 };
 
+class ReExec : public FaultBase
+{
+  public:
+    virtual FaultName name() const { return "Re-execution fault";}
+    ReExec() {}
+    void invoke(ThreadContext *tc,
+            StaticInstPtr inst = StaticInst::nullStaticInstPtr);
+};
+
+
 #if !FULL_SYSTEM
 class GenericPageTableFault : public FaultBase
 {