Index: lib/Transforms/Instrumentation/HWAddressSanitizer.cpp
===================================================================
--- lib/Transforms/Instrumentation/HWAddressSanitizer.cpp
+++ lib/Transforms/Instrumentation/HWAddressSanitizer.cpp
@@ -121,6 +121,11 @@
     cl::desc("Enable KernelHWAddressSanitizer instrumentation"),
     cl::Hidden, cl::init(false));
 
+static cl::opt<bool>
+    ClInstrumentMemIntrinsics("hwasan-instrument-mem-intrinsics",
+                              cl::desc("instrument memory intrinsics"),
+                              cl::Hidden, cl::init(false));
+
 // These flags allow to change the shadow mapping and control how shadow memory
 // is accessed. The shadow mapping looks like:
 //    Shadow = (Mem >> scale) + offset
@@ -182,6 +187,7 @@
   void instrumentMemAccessInline(Value *PtrLong, bool IsWrite,
                                  unsigned AccessSizeIndex,
                                  Instruction *InsertBefore);
+  void instrumentMemIntrinsic(MemIntrinsic *MI);
   bool instrumentMemAccess(Instruction *I);
   Value *isInterestingMemoryAccess(Instruction *I, bool *IsWrite,
                                    uint64_t *TypeSize, unsigned *Alignment,
@@ -539,12 +545,40 @@
   IRB.CreateCall(Asm, PtrLong);
 }
 
+void HWAddressSanitizer::instrumentMemIntrinsic(MemIntrinsic *MI) {
+  IRBuilder<> IRB(MI);
+  Module *M = MI->getParent()->getParent()->getParent();
+  if (isa<MemTransferInst>(MI)) {
+    auto *F = M->getOrInsertFunction(
+        isa<MemMoveInst>(MI) ? "memmove" : "memcpy", IRB.getInt8PtrTy(),
+        IRB.getInt8PtrTy(), IRB.getInt8PtrTy(), IntptrTy);
+    IRB.CreateCall(
+        F, {IRB.CreatePointerCast(MI->getOperand(0), IRB.getInt8PtrTy()),
+            IRB.CreatePointerCast(MI->getOperand(1), IRB.getInt8PtrTy()),
+            IRB.CreateIntCast(MI->getOperand(2), IntptrTy, false)});
+  } else if (isa<MemSetInst>(MI)) {
+    IRB.CreateCall(
+        M->getOrInsertFunction("memset", IRB.getInt8PtrTy(), IRB.getInt8PtrTy(),
+                               IRB.getInt32Ty(), IntptrTy),
+        {IRB.CreatePointerCast(MI->getOperand(0), IRB.getInt8PtrTy()),
+         IRB.CreateIntCast(MI->getOperand(1), IRB.getInt32Ty(), false),
+         IRB.CreateIntCast(MI->getOperand(2), IntptrTy, false)});
+  }
+  MI->eraseFromParent();
+}
+
 bool HWAddressSanitizer::instrumentMemAccess(Instruction *I) {
   LLVM_DEBUG(dbgs() << "Instrumenting: " << *I << "\n");
   bool IsWrite = false;
   unsigned Alignment = 0;
   uint64_t TypeSize = 0;
   Value *MaybeMask = nullptr;
+
+  if (ClInstrumentMemIntrinsics && isa<MemIntrinsic>(I)) {
+    instrumentMemIntrinsic(cast<MemIntrinsic>(I));
+    return true;
+  }
+
   Value *Addr =
       isInterestingMemoryAccess(I, &IsWrite, &TypeSize, &Alignment, &MaybeMask);
 
Index: test/Instrumentation/HWAddressSanitizer/mem-intrinsics.ll
===================================================================
--- test/Instrumentation/HWAddressSanitizer/mem-intrinsics.ll
+++ test/Instrumentation/HWAddressSanitizer/mem-intrinsics.ll
@@ -0,0 +1,36 @@
+; RUN: opt -S -hwasan -hwasan-instrument-mem-intrinsics %s | FileCheck %s
+
+; CHECK: call i8* @memset
+; CHECK: call i8* @memmove
+; CHECK: call i8* @memcpy
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+; Function Attrs: noinline nounwind optnone uwtable
+define dso_local i32 @main() sanitize_hwaddress {
+entry:
+  %retval = alloca i32, align 4
+  %Q = alloca [10 x i8], align 1
+  %P = alloca [10 x i8], align 1
+  store i32 0, i32* %retval, align 4
+  %arraydecay = getelementptr inbounds [10 x i8], [10 x i8]* %Q, i32 0, i32 0
+  call void @llvm.memset.p0i8.i64(i8* align 1 %arraydecay, i8 0, i64 10, i1 false)
+  %arraydecay1 = getelementptr inbounds [10 x i8], [10 x i8]* %Q, i32 0, i32 0
+  %arraydecay2 = getelementptr inbounds [10 x i8], [10 x i8]* %Q, i32 0, i32 0
+  %add.ptr = getelementptr inbounds i8, i8* %arraydecay2, i64 5
+  call void @llvm.memmove.p0i8.p0i8.i64(i8* align 1 %arraydecay1, i8* align 1 %add.ptr, i64 5, i1 false)
+  %arraydecay3 = getelementptr inbounds [10 x i8], [10 x i8]* %P, i32 0, i32 0
+  %arraydecay4 = getelementptr inbounds [10 x i8], [10 x i8]* %Q, i32 0, i32 0
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 1 %arraydecay3, i8* align 1 %arraydecay4, i64 10, i1 false)
+  ret i32 0
+}
+
+; Function Attrs: argmemonly nounwind
+declare void @llvm.memset.p0i8.i64(i8* nocapture writeonly, i8, i64, i1) #1
+
+; Function Attrs: argmemonly nounwind
+declare void @llvm.memmove.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i1) #1
+
+; Function Attrs: argmemonly nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture writeonly, i8* nocapture readonly, i64, i1) #1