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 +1,64 @@
-// 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 -std=c++2a %s -emit-llvm -S -o - | FileCheck %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 bar is not inlined even it's marked as always_inline
+
+// CHECK-LABEL: define void @_Z3foov(
+// CHECK:         invoke void @_Z3barv(
diff --git a/llvm/lib/Passes/PassBuilder.cpp b/llvm/lib/Passes/PassBuilder.cpp
--- a/llvm/lib/Passes/PassBuilder.cpp
+++ b/llvm/lib/Passes/PassBuilder.cpp
@@ -1883,6 +1883,12 @@
   for (auto &C : PipelineEarlySimplificationEPCallbacks)
     C(MPM, Level);
 
+  // CoroEarlyPass needs to run before AlwaysInliner to make sure we add
+  // proper attributes to coroutines first, so that Inliner won't inline
+  // coroutines.
+  if (PTO.Coroutines)
+    MPM.addPass(createModuleToFunctionPassAdaptor(CoroEarlyPass()));
+
   // Build a minimal pipeline based on the semantics required by LLVM,
   // which is just that always inlining occurs. Further, disable generating
   // lifetime intrinsics to avoid enabling further optimizations during
@@ -1940,8 +1946,6 @@
   }
 
   if (PTO.Coroutines) {
-    MPM.addPass(createModuleToFunctionPassAdaptor(CoroEarlyPass()));
-
     CGSCCPassManager CGPM(DebugLogging);
     CGPM.addPass(CoroSplitPass());
     CGPM.addPass(createCGSCCToFunctionPassAdaptor(CoroElidePass()));