Index: cfe/trunk/lib/CodeGen/BackendUtil.cpp
===================================================================
--- cfe/trunk/lib/CodeGen/BackendUtil.cpp
+++ cfe/trunk/lib/CodeGen/BackendUtil.cpp
@@ -653,6 +653,11 @@
     InstrProfOptions Options;
     Options.NoRedZone = CodeGenOpts.DisableRedZone;
     Options.InstrProfileOutput = CodeGenOpts.InstrProfileOutput;
+
+    // TODO: Surface the option to emit atomic profile counter increments at
+    // the driver level.
+    Options.Atomic = LangOpts.Sanitize.has(SanitizerKind::Thread);
+
     MPM.add(createInstrProfilingLegacyPass(Options));
   }
   if (CodeGenOpts.hasProfileIRInstr()) {
Index: cfe/trunk/test/CodeGen/tsan-instrprof-atomic.c
===================================================================
--- cfe/trunk/test/CodeGen/tsan-instrprof-atomic.c
+++ cfe/trunk/test/CodeGen/tsan-instrprof-atomic.c
@@ -0,0 +1,6 @@
+// RUN: %clang_cc1 %s -emit-llvm -fprofile-instrument=clang -fsanitize=thread -o - | FileCheck %s
+
+// CHECK: define void @foo
+// CHECK-NOT: load {{.*}}foo
+// CHECK: ret void
+void foo() {}
Index: llvm/trunk/include/llvm/Transforms/Instrumentation.h
===================================================================
--- llvm/trunk/include/llvm/Transforms/Instrumentation.h
+++ llvm/trunk/include/llvm/Transforms/Instrumentation.h
@@ -111,6 +111,9 @@
   // Do counter register promotion
   bool DoCounterPromotion = false;
 
+  // Use atomic profile counter increments.
+  bool Atomic = false;
+
   // Name of the profile file to use as output
   std::string InstrProfileOutput;
 
Index: llvm/trunk/lib/Transforms/Instrumentation/InstrProfiling.cpp
===================================================================
--- llvm/trunk/lib/Transforms/Instrumentation/InstrProfiling.cpp
+++ llvm/trunk/lib/Transforms/Instrumentation/InstrProfiling.cpp
@@ -96,6 +96,11 @@
     // is usually smaller than 2.
     cl::init(1.0));
 
+cl::opt<bool> AtomicCounterUpdateAll(
+    "instrprof-atomic-counter-update-all", cl::ZeroOrMore,
+    cl::desc("Make all profile counter updates atomic (for testing only)"),
+    cl::init(false));
+
 cl::opt<bool> AtomicCounterUpdatePromoted(
     "atomic-counter-update-promoted", cl::ZeroOrMore,
     cl::desc("Do counter update using atomic fetch add "
@@ -597,12 +602,17 @@
   IRBuilder<> Builder(Inc);
   uint64_t Index = Inc->getIndex()->getZExtValue();
   Value *Addr = Builder.CreateConstInBoundsGEP2_64(Counters, 0, Index);
-  Value *Load = Builder.CreateLoad(Addr, "pgocount");
-  auto *Count = Builder.CreateAdd(Load, Inc->getStep());
-  auto *Store = Builder.CreateStore(Count, Addr);
-  Inc->replaceAllUsesWith(Store);
-  if (isCounterPromotionEnabled())
-    PromotionCandidates.emplace_back(cast<Instruction>(Load), Store);
+
+  if (Options.Atomic || AtomicCounterUpdateAll) {
+    Builder.CreateAtomicRMW(AtomicRMWInst::Add, Addr, Inc->getStep(),
+                            AtomicOrdering::Monotonic);
+  } else {
+    Value *Load = Builder.CreateLoad(Addr, "pgocount");
+    auto *Count = Builder.CreateAdd(Load, Inc->getStep());
+    auto *Store = Builder.CreateStore(Count, Addr);
+    if (isCounterPromotionEnabled())
+      PromotionCandidates.emplace_back(cast<Instruction>(Load), Store);
+  }
   Inc->eraseFromParent();
 }
 
Index: llvm/trunk/test/Instrumentation/InstrProfiling/atomic-updates.ll
===================================================================
--- llvm/trunk/test/Instrumentation/InstrProfiling/atomic-updates.ll
+++ llvm/trunk/test/Instrumentation/InstrProfiling/atomic-updates.ll
@@ -0,0 +1,14 @@
+; RUN: opt < %s -S -instrprof -instrprof-atomic-counter-update-all | FileCheck %s
+
+target triple = "x86_64-apple-macosx10.10.0"
+
+@__profn_foo = hidden constant [3 x i8] c"foo"
+
+; CHECK-LABEL: define void @foo
+; CHECK-NEXT: atomicrmw add i64* getelementptr inbounds ([1 x i64], [1 x i64]* @__profc_foo, i64 0, i64 0), i64 1 monotonic
+define void @foo() {
+  call void @llvm.instrprof.increment(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @__profn_foo, i32 0, i32 0), i64 0, i32 1, i32 0)
+  ret void
+}
+
+declare void @llvm.instrprof.increment(i8*, i64, i32, i32)