diff --git a/clang/lib/CodeGen/CGCoroutine.cpp b/clang/lib/CodeGen/CGCoroutine.cpp
--- a/clang/lib/CodeGen/CGCoroutine.cpp
+++ b/clang/lib/CodeGen/CGCoroutine.cpp
@@ -558,6 +558,8 @@
   CurCoro.Data->SuspendBB = RetBB;
   assert(ShouldEmitLifetimeMarkers &&
          "Must emit lifetime intrinsics for coroutines");
+  // CORO_PRESPLIT_ATTR = UNPREPARED_FOR_SPLIT
+  CurFn->addFnAttr("coroutine.presplit", "0");
 
   // Backend is allowed to elide memory allocations, to help it, emit
   // auto mem = coro.alloc() ? 0 : ... allocation code ...;
diff --git a/clang/test/CodeGenCoroutines/coro-always-inline.cpp b/clang/test/CodeGenCoroutines/coro-always-inline-resume.cpp
copy from clang/test/CodeGenCoroutines/coro-always-inline.cpp
copy to clang/test/CodeGenCoroutines/coro-always-inline-resume.cpp
diff --git a/clang/test/CodeGenCoroutines/coro-always-inline.cpp b/clang/test/CodeGenCoroutines/coro-always-inline.cpp
--- a/clang/test/CodeGenCoroutines/coro-always-inline.cpp
+++ b/clang/test/CodeGenCoroutines/coro-always-inline.cpp
@@ -1,54 +1,68 @@
-// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -emit-llvm -fcoroutines-ts \
-// RUN:   -fexperimental-new-pass-manager -O0 %s -o - | FileCheck %s
-// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -emit-llvm -fcoroutines-ts \
-// RUN:   -fexperimental-new-pass-manager -fno-inline -O0 %s -o - | FileCheck %s
-
-// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -emit-llvm -fcoroutines-ts \
-// RUN:   -O0 %s -o - | FileCheck %s
-// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -emit-llvm -fcoroutines-ts \
-// RUN:   -fno-inline -O0 %s -o - | FileCheck %s
-
-namespace std {
-namespace experimental {
-
-struct handle {};
-
-struct awaitable {
-  bool await_ready() noexcept { return true; }
-  // CHECK-NOT: await_suspend
-  inline void __attribute__((__always_inline__)) await_suspend(handle) noexcept {}
-  bool await_resume() noexcept { return true; }
-};
+// RUN: %clang -cc1 -triple x86_64-unknown-linux-gnu -std=c++2a %s -emit-llvm -disable-llvm-passes -o - | FileCheck %s
+// RUN: %clang -cc1 -triple x86_64-unknown-linux-gnu -std=c++2a %s -emit-llvm -disable-llvm-passes -o - | opt -always-inline -S | FileCheck --check-prefix=INLINE %s
 
-template <typename T>
-struct coroutine_handle {
-  static handle from_address(void *address) noexcept { return {}; }
-};
+#include "Inputs/coroutine.h"
+
+namespace coro = std::experimental::coroutines_v1;
+
+class task {
+public:
+  class promise_type {
+  public:
+    task get_return_object() noexcept;
+    coro::suspend_always initial_suspend() noexcept;
+    void return_void() noexcept;
+    void unhandled_exception() noexcept;
+
+    struct final_awaiter {
+      bool await_ready() noexcept;
+      void await_suspend(coro::coroutine_handle<promise_type> h) noexcept;
+      void await_resume() noexcept;
+    };
 
-template <typename T = void>
-struct coroutine_traits {
-  struct promise_type {
-    awaitable initial_suspend() { return {}; }
-    awaitable final_suspend() noexcept { return {}; }
-    void return_void() {}
-    T get_return_object() { return T(); }
-    void unhandled_exception() {}
+    final_awaiter final_suspend() noexcept;
+
+    coro::coroutine_handle<> continuation;
   };
+
+  task(task &&t) noexcept;
+  ~task();
+
+  class awaiter {
+  public:
+    bool await_ready() noexcept;
+    void await_suspend(coro::coroutine_handle<> continuation) noexcept;
+    void await_resume() noexcept;
+
+  private:
+    friend task;
+    explicit awaiter(coro::coroutine_handle<promise_type> h) noexcept;
+    coro::coroutine_handle<promise_type> coro_;
+  };
+
+  awaiter operator co_await() &&noexcept;
+
+private:
+  explicit task(coro::coroutine_handle<promise_type> h) noexcept;
+  coro::coroutine_handle<promise_type> coro_;
 };
-} // namespace experimental
-} // namespace std
-
-// CHECK-LABEL: @_Z3foov
-// CHECK-LABEL: entry:
-// CHECK-NEXT: %this.addr.i{{[0-9]*}} = alloca %"struct.std::experimental::awaitable"*, align 8
-// CHECK-NEXT: %this.addr.i{{[0-9]*}} = alloca %"struct.std::experimental::awaitable"*, align 8
-// CHECK: [[CAST0:%[0-9]+]] = bitcast %"struct.std::experimental::awaitable"** %this.addr.i{{[0-9]*}} to i8*
-// CHECK-NEXT: call void @llvm.lifetime.start.p0i8(i64 8, i8* [[CAST0]])
-// CHECK: [[CAST1:%[0-9]+]] = bitcast %"struct.std::experimental::awaitable"** %this.addr.i{{[0-9]*}} to i8*
-// CHECK-NEXT: call void @llvm.lifetime.end.p0i8(i64 8, i8* [[CAST1]])
-
-// CHECK: [[CAST2:%[0-9]+]] = bitcast %"struct.std::experimental::awaitable"** %this.addr.i{{[0-9]*}} to i8*
-// CHECK-NEXT: call void @llvm.lifetime.start.p0i8(i64 8, i8* [[CAST2]])
-// CHECK: [[CAST3:%[0-9]+]] = bitcast %"struct.std::experimental::awaitable"** %this.addr.i{{[0-9]*}} to i8*
-// CHECK-NEXT: call void @llvm.lifetime.end.p0i8(i64 8, i8* [[CAST3]])
-void foo() { co_return; }
+
+task cee();
+
+__attribute__((always_inline)) inline task bar() {
+  co_await cee();
+  co_return;
+}
+
+task foo() {
+  co_await bar();
+  co_return;
+}
+
+// check that Clang front-end will tag bar with both alwaysinline and coroutine presplit
+// CHECK:       define linkonce_odr void @_Z3barv({{.*}}) #[[ATTR:[0-9]+]] {{.*}}
+// CHECK:       attributes #[[ATTR]] = { alwaysinline {{.*}} "coroutine.presplit"="0" {{.*}}}
+
+// check that bar is not inlined even it's marked as always_inline
+// INLINE-LABEL: define dso_local void @_Z3foov(
+// INLINE:         call void @_Z3barv(
diff --git a/llvm/lib/Transforms/Coroutines/CoroEarly.cpp b/llvm/lib/Transforms/Coroutines/CoroEarly.cpp
--- a/llvm/lib/Transforms/Coroutines/CoroEarly.cpp
+++ b/llvm/lib/Transforms/Coroutines/CoroEarly.cpp
@@ -179,7 +179,6 @@
         // with a coroutine attribute.
         if (auto *CII = cast<CoroIdInst>(&I)) {
           if (CII->getInfo().isPreSplit()) {
-            F.addFnAttr(CORO_PRESPLIT_ATTR, UNPREPARED_FOR_SPLIT);
             setCannotDuplicate(CII);
             CII->setCoroutineSelf();
             CoroId = cast<CoroIdInst>(&I);
diff --git a/llvm/test/Transforms/Coroutines/coro-debug-O2.ll b/llvm/test/Transforms/Coroutines/coro-debug-O2.ll
--- a/llvm/test/Transforms/Coroutines/coro-debug-O2.ll
+++ b/llvm/test/Transforms/Coroutines/coro-debug-O2.ll
@@ -9,7 +9,7 @@
 ; CHECK: ![[PROMISEVAR_RESUME]] = !DILocalVariable(name: "__promise"
 %promise_type = type { i32, i32, double }
 
-define void @f() !dbg !8 {
+define void @f() "coroutine.presplit"="0" !dbg !8 {
 entry:
     %__promise = alloca %promise_type, align 8
     %0 = bitcast %promise_type* %__promise to i8*
diff --git a/llvm/test/Transforms/Coroutines/coro-debug-frame-variable.ll b/llvm/test/Transforms/Coroutines/coro-debug-frame-variable.ll
--- a/llvm/test/Transforms/Coroutines/coro-debug-frame-variable.ll
+++ b/llvm/test/Transforms/Coroutines/coro-debug-frame-variable.ll
@@ -63,7 +63,7 @@
 ; CHECK: ![[IVAR_RESUME]] = !DILocalVariable(name: "i"
 ; CHECK: ![[JVAR_RESUME]] = !DILocalVariable(name: "j"
 ; CHECK: ![[JDBGLOC_RESUME]] = !DILocation(line: 32, column: 7, scope: ![[RESUME_SCOPE]])
-define void @f() {
+define void @f() "coroutine.presplit"="0" {
 entry:
   %__promise = alloca i8, align 8
   %i = alloca i32, align 4
diff --git a/llvm/test/Transforms/Coroutines/coro-split-01.ll b/llvm/test/Transforms/Coroutines/coro-split-01.ll
--- a/llvm/test/Transforms/Coroutines/coro-split-01.ll
+++ b/llvm/test/Transforms/Coroutines/coro-split-01.ll
@@ -2,7 +2,7 @@
 ; RUN: opt < %s -S -enable-coroutines -O2 | FileCheck %s
 ; RUN: opt < %s -S -enable-coroutines -passes='default<O2>' | FileCheck %s
 
-define i8* @f() {
+define i8* @f() "coroutine.presplit"="0" {
 entry:
   %id = call token @llvm.coro.id(i32 0, i8* null, i8* null, i8* null)
   %need.dyn.alloc = call i1 @llvm.coro.alloc(token %id)
diff --git a/llvm/test/Transforms/Coroutines/coro-split-recursive.ll b/llvm/test/Transforms/Coroutines/coro-split-recursive.ll
--- a/llvm/test/Transforms/Coroutines/coro-split-recursive.ll
+++ b/llvm/test/Transforms/Coroutines/coro-split-recursive.ll
@@ -13,7 +13,7 @@
 ; CHECK: call void @foo()
 ; CHECK-LABEL: define {{.*}}void @foo.destroy(
 
-define void @foo() {
+define void @foo() "coroutine.presplit"="0" {
 entry:
   %__promise = alloca i32, align 8
   %0 = bitcast i32* %__promise to i8*
diff --git a/llvm/test/Transforms/Coroutines/ex0.ll b/llvm/test/Transforms/Coroutines/ex0.ll
--- a/llvm/test/Transforms/Coroutines/ex0.ll
+++ b/llvm/test/Transforms/Coroutines/ex0.ll
@@ -2,7 +2,7 @@
 ; RUN: opt < %s -enable-coroutines -O2 -preserve-alignment-assumptions-during-inlining=false -S | FileCheck %s
 ; RUN: opt < %s -enable-coroutines -aa-pipeline=basic-aa -passes='default<O2>' -preserve-alignment-assumptions-during-inlining=false -S | FileCheck %s
 
-define i8* @f(i32 %n) {
+define i8* @f(i32 %n) "coroutine.presplit"="0" {
 entry:
   %id = call token @llvm.coro.id(i32 0, i8* null, i8* null, i8* null)
   %size = call i32 @llvm.coro.size.i32()
diff --git a/llvm/test/Transforms/Coroutines/ex1.ll b/llvm/test/Transforms/Coroutines/ex1.ll
--- a/llvm/test/Transforms/Coroutines/ex1.ll
+++ b/llvm/test/Transforms/Coroutines/ex1.ll
@@ -2,7 +2,7 @@
 ; RUN: opt < %s -O2 -enable-coroutines -preserve-alignment-assumptions-during-inlining=false -S | FileCheck %s
 ; RUN: opt < %s -aa-pipeline=basic-aa -passes='default<O2>' -enable-coroutines -preserve-alignment-assumptions-during-inlining=false -S | FileCheck %s
 
-define i8* @f(i32 %n) {
+define i8* @f(i32 %n) "coroutine.presplit"="0" {
 entry:
   %id = call token @llvm.coro.id(i32 0, i8* null, i8* null, i8* null)
   %size = call i32 @llvm.coro.size.i32()
diff --git a/llvm/test/Transforms/Coroutines/ex2.ll b/llvm/test/Transforms/Coroutines/ex2.ll
--- a/llvm/test/Transforms/Coroutines/ex2.ll
+++ b/llvm/test/Transforms/Coroutines/ex2.ll
@@ -2,7 +2,7 @@
 ; RUN: opt < %s -O2 -enable-coroutines -S | FileCheck %s
 ; RUN: opt < %s -passes='default<O2>' -enable-coroutines -S | FileCheck %s
 
-define i8* @f(i32 %n) {
+define i8* @f(i32 %n) "coroutine.presplit"="0" {
 entry:
   %id = call token @llvm.coro.id(i32 0, i8* null, i8* null, i8* null)
   %need.dyn.alloc = call i1 @llvm.coro.alloc(token %id)
diff --git a/llvm/test/Transforms/Coroutines/ex3.ll b/llvm/test/Transforms/Coroutines/ex3.ll
--- a/llvm/test/Transforms/Coroutines/ex3.ll
+++ b/llvm/test/Transforms/Coroutines/ex3.ll
@@ -2,7 +2,7 @@
 ; RUN: opt < %s -O2 -enable-coroutines -S | FileCheck %s
 ; RUN: opt < %s -aa-pipeline=basic-aa -passes='default<O2>' -enable-coroutines -S | FileCheck %s
 
-define i8* @f(i32 %n) {
+define i8* @f(i32 %n) "coroutine.presplit"="0" {
 entry:
   %id = call token @llvm.coro.id(i32 0, i8* null, i8* null, i8* null)
   %size = call i32 @llvm.coro.size.i32()
diff --git a/llvm/test/Transforms/Coroutines/ex4.ll b/llvm/test/Transforms/Coroutines/ex4.ll
--- a/llvm/test/Transforms/Coroutines/ex4.ll
+++ b/llvm/test/Transforms/Coroutines/ex4.ll
@@ -2,7 +2,7 @@
 ; RUN: opt < %s -O2 -enable-coroutines -S | FileCheck %s
 ; RUN: opt < %s -passes='default<O2>' -enable-coroutines -S | FileCheck %s
 
-define i8* @f(i32 %n) {
+define i8* @f(i32 %n) "coroutine.presplit"="0" {
 entry:
   %promise = alloca i32
   %pv = bitcast i32* %promise to i8*
diff --git a/llvm/test/Transforms/Coroutines/ex5.ll b/llvm/test/Transforms/Coroutines/ex5.ll
--- a/llvm/test/Transforms/Coroutines/ex5.ll
+++ b/llvm/test/Transforms/Coroutines/ex5.ll
@@ -2,7 +2,7 @@
 ; RUN: opt < %s -O2 -enable-coroutines -preserve-alignment-assumptions-during-inlining=false -S | FileCheck %s
 ; RUN: opt < %s -aa-pipeline=basic-aa -passes='default<O2>' -enable-coroutines -preserve-alignment-assumptions-during-inlining=false -S | FileCheck %s
 
-define i8* @f(i32 %n) {
+define i8* @f(i32 %n) "coroutine.presplit"="0" {
 entry:
   %id = call token @llvm.coro.id(i32 0, i8* null, i8* null, i8* null)
   %size = call i32 @llvm.coro.size.i32()
diff --git a/llvm/test/Transforms/Coroutines/phi-coro-end.ll b/llvm/test/Transforms/Coroutines/phi-coro-end.ll
--- a/llvm/test/Transforms/Coroutines/phi-coro-end.ll
+++ b/llvm/test/Transforms/Coroutines/phi-coro-end.ll
@@ -2,7 +2,7 @@
 ; RUN: opt < %s -O2 -enable-coroutines -S | FileCheck %s
 ; RUN: opt < %s -aa-pipeline=basic-aa -passes='default<O2>' -enable-coroutines -S | FileCheck %s
 
-define i8* @f(i32 %n) {
+define i8* @f(i32 %n) "coroutine.presplit"="0" {
 entry:
   %id = call token @llvm.coro.id(i32 0, i8* null, i8* null, i8* null)
   %size = call i32 @llvm.coro.size.i32()
diff --git a/llvm/test/Transforms/Coroutines/restart-trigger.ll b/llvm/test/Transforms/Coroutines/restart-trigger.ll
--- a/llvm/test/Transforms/Coroutines/restart-trigger.ll
+++ b/llvm/test/Transforms/Coroutines/restart-trigger.ll
@@ -12,7 +12,7 @@
 ; CHECK:      CoroSplit: Processing coroutine 'f' state: 0
 ; CHECK-NEXT: CoroSplit: Processing coroutine 'f' state: 1
 
-define void @f() {
+define void @f() "coroutine.presplit"="0" {
   %id = call token @llvm.coro.id(i32 0, i8* null, i8* null, i8* null)
   %size = call i32 @llvm.coro.size.i32()
   %alloc = call i8* @malloc(i32 %size)