Index: include/llvm/Analysis/Utils/Local.h
===================================================================
--- include/llvm/Analysis/Utils/Local.h
+++ include/llvm/Analysis/Utils/Local.h
@@ -422,13 +422,14 @@
 ///
 /// Metadata not listed as known via KnownIDs is removed
 void combineMetadata(Instruction *K, const Instruction *J,
-                     ArrayRef<unsigned> KnownIDs, bool KDominatesJ = false);
+                     ArrayRef<unsigned> KnownIDs, bool KDominatesJ);
 
 /// Combine the metadata of two instructions so that K can replace J. This
 /// specifically handles the case of CSE-like transformations.
 ///
 /// Unknown metadata is removed.
-void combineMetadataForCSE(Instruction *K, const Instruction *J);
+void combineMetadataForCSE(Instruction *K, const Instruction *J,
+                           bool KDominatesJ);
 
 /// Patch the replacement so that it is not more restrictive than the value
 /// being replaced. It assumes Repl dominates I.
Index: lib/Transforms/InstCombine/InstCombineLoadStoreAlloca.cpp
===================================================================
--- lib/Transforms/InstCombine/InstCombineLoadStoreAlloca.cpp
+++ lib/Transforms/InstCombine/InstCombineLoadStoreAlloca.cpp
@@ -1021,7 +1021,7 @@
   if (Value *AvailableVal = FindAvailableLoadedValue(
           &LI, LI.getParent(), BBI, DefMaxInstsToScan, AA, &IsLoadCSE)) {
     if (IsLoadCSE)
-      combineMetadataForCSE(cast<LoadInst>(AvailableVal), &LI);
+      combineMetadataForCSE(cast<LoadInst>(AvailableVal), &LI, true);
 
     return replaceInstUsesWith(
         LI, Builder.CreateBitOrPointerCast(AvailableVal, LI.getType(),
Index: lib/Transforms/InstCombine/InstCombinePHI.cpp
===================================================================
--- lib/Transforms/InstCombine/InstCombinePHI.cpp
+++ lib/Transforms/InstCombine/InstCombinePHI.cpp
@@ -616,7 +616,7 @@
   // Add all operands to the new PHI and combine TBAA metadata.
   for (unsigned i = 1, e = PN.getNumIncomingValues(); i != e; ++i) {
     LoadInst *LI = cast<LoadInst>(PN.getIncomingValue(i));
-    combineMetadata(NewLI, LI, KnownIDs);
+    combineMetadata(NewLI, LI, KnownIDs, false);
     Value *NewInVal = LI->getOperand(0);
     if (NewInVal != InVal)
       InVal = nullptr;
Index: lib/Transforms/Scalar/GVNHoist.cpp
===================================================================
--- lib/Transforms/Scalar/GVNHoist.cpp
+++ lib/Transforms/Scalar/GVNHoist.cpp
@@ -247,7 +247,7 @@
       LLVMContext::MD_noalias,        LLVMContext::MD_range,
       LLVMContext::MD_fpmath,         LLVMContext::MD_invariant_load,
       LLVMContext::MD_invariant_group};
-  combineMetadata(ReplInst, I, KnownIDs);
+  combineMetadata(ReplInst, I, KnownIDs, false);
 }
 
 // This pass hoists common computations across branches sharing common
Index: lib/Transforms/Scalar/GVNSink.cpp
===================================================================
--- lib/Transforms/Scalar/GVNSink.cpp
+++ lib/Transforms/Scalar/GVNSink.cpp
@@ -859,7 +859,7 @@
   // Update metadata and IR flags.
   for (auto *I : Insts)
     if (I != I0) {
-      combineMetadataForCSE(I0, I);
+      combineMetadataForCSE(I0, I, false);
       I0->andIRFlags(I);
     }
 
Index: lib/Transforms/Scalar/JumpThreading.cpp
===================================================================
--- lib/Transforms/Scalar/JumpThreading.cpp
+++ lib/Transforms/Scalar/JumpThreading.cpp
@@ -1297,7 +1297,7 @@
 
     if (IsLoadCSE) {
       LoadInst *NLoadI = cast<LoadInst>(AvailableVal);
-      combineMetadataForCSE(NLoadI, LoadI);
+      combineMetadataForCSE(NLoadI, LoadI, true);
     };
 
     // If the returned value is the load itself, replace with an undef. This can
@@ -1487,7 +1487,7 @@
   }
 
   for (LoadInst *PredLoadI : CSELoads) {
-    combineMetadataForCSE(PredLoadI, LoadI);
+    combineMetadataForCSE(PredLoadI, LoadI, false);
   }
 
   LoadI->replaceAllUsesWith(PN);
Index: lib/Transforms/Scalar/MemCpyOptimizer.cpp
===================================================================
--- lib/Transforms/Scalar/MemCpyOptimizer.cpp
+++ lib/Transforms/Scalar/MemCpyOptimizer.cpp
@@ -994,7 +994,7 @@
   unsigned KnownIDs[] = {LLVMContext::MD_tbaa, LLVMContext::MD_alias_scope,
                          LLVMContext::MD_noalias,
                          LLVMContext::MD_invariant_group};
-  combineMetadata(C, cpy, KnownIDs);
+  combineMetadata(C, cpy, KnownIDs, false);
 
   // Remove the memcpy.
   MD->removeInstruction(cpy);
Index: lib/Transforms/Utils/Local.cpp
===================================================================
--- lib/Transforms/Utils/Local.cpp
+++ lib/Transforms/Utils/Local.cpp
@@ -2063,7 +2063,8 @@
       K->setMetadata(LLVMContext::MD_invariant_group, JMD);
 }
 
-void llvm::combineMetadataForCSE(Instruction *K, const Instruction *J) {
+void llvm::combineMetadataForCSE(Instruction *K, const Instruction *J,
+                                 bool KDominatesJ) {
   unsigned KnownIDs[] = {
       LLVMContext::MD_tbaa,            LLVMContext::MD_alias_scope,
       LLVMContext::MD_noalias,         LLVMContext::MD_range,
@@ -2071,7 +2072,7 @@
       LLVMContext::MD_invariant_group, LLVMContext::MD_align,
       LLVMContext::MD_dereferenceable,
       LLVMContext::MD_dereferenceable_or_null};
-  combineMetadata(K, J, KnownIDs);
+  combineMetadata(K, J, KnownIDs, KDominatesJ);
 }
 
 void llvm::patchReplacementInstruction(Instruction *I, Value *Repl) {
Index: lib/Transforms/Utils/SimplifyCFG.cpp
===================================================================
--- lib/Transforms/Utils/SimplifyCFG.cpp
+++ lib/Transforms/Utils/SimplifyCFG.cpp
@@ -1317,7 +1317,7 @@
                              LLVMContext::MD_dereferenceable,
                              LLVMContext::MD_dereferenceable_or_null,
                              LLVMContext::MD_mem_parallel_loop_access};
-      combineMetadata(I1, I2, KnownIDs);
+      combineMetadata(I1, I2, KnownIDs, false);
 
       // I1 and I2 are being combined into a single instruction.  Its debug
       // location is the merged locations of the original instructions.
@@ -1583,7 +1583,7 @@
       // However, as N-way merge for CallInst is rare, so we use simplified API
       // instead of using complex API for N-way merge.
       I0->applyMergedLocation(I0->getDebugLoc(), I->getDebugLoc());
-      combineMetadataForCSE(I0, I);
+      combineMetadataForCSE(I0, I, false);
       I0->andIRFlags(I);
     }