Index: MicroBenchmarks/XRay/CMakeLists.txt
===================================================================
--- MicroBenchmarks/XRay/CMakeLists.txt
+++ MicroBenchmarks/XRay/CMakeLists.txt
@@ -1,2 +1,3 @@
 add_subdirectory(ReturnReference)
 add_subdirectory(FDRMode)
+add_subdirectory(ProfilingMode)
Index: MicroBenchmarks/XRay/ProfilingMode/CMakeLists.txt
===================================================================
--- /dev/null
+++ MicroBenchmarks/XRay/ProfilingMode/CMakeLists.txt
@@ -0,0 +1,18 @@
+check_cxx_compiler_flag(-fxray-instrument COMPILER_HAS_FXRAY_INSTRUMENT)
+check_cxx_compiler_flag(-fxray-modes=xray-profiling
+	COMPILER_HAS_FXRAY_PROFILING)
+if(ARCH STREQUAL "x86"
+		AND COMPILER_HAS_FXRAY_INSTRUMENT
+		AND COMPILER_HAS_FXRAY_PROFILING)
+  list(APPEND CPPFLAGS
+		-std=c++11 -Wl,--gc-sections
+		-fxray-instrument -fxray-modes=xray-profiling)
+  list(APPEND LDFLAGS
+		-fxray-instrument -fxray-modes=xray-profiling)
+  llvm_test_run()
+  llvm_test_executable(deep-call-bench deep-call-bench.cc)
+  target_link_libraries(deep-call-bench benchmark)
+  llvm_test_executable(shallow-call-bench shallow-call-bench.cc)
+  target_link_libraries(shallow-call-bench benchmark)
+endif()
+
Index: MicroBenchmarks/XRay/ProfilingMode/deep-call-bench.cc
===================================================================
--- /dev/null
+++ MicroBenchmarks/XRay/ProfilingMode/deep-call-bench.cc
@@ -0,0 +1,83 @@
+//===- deep-call-bench.cc - XRay Profiling Mode Benchmarks ----------------===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+//
+// These benchmarks measure the cost of XRay profiling mode when enabled.
+//
+//===----------------------------------------------------------------------===//
+
+#include <atomic>
+#include <iostream>
+#include <mutex>
+#include <thread>
+#include "benchmark/benchmark.h"
+#include "xray/xray_log_interface.h"
+
+namespace {
+
+std::atomic<int> some_global{0};
+
+std::atomic<int> some_temporary{0};
+
+std::once_flag profiling_init_flag;
+
+std::atomic<bool> go{false};
+
+[[clang::xray_never_instrument]] static void profiling_init() {
+  if (__xray_log_select_mode("xray-profiling") != XRAY_REGISTRATION_OK) {
+    std::cerr << "Failed selecting 'xray-profiling' mode. Aborting.\n";
+    std::abort();
+  }
+
+  if (__xray_log_init_mode("xray-profiling", "no_flush=true") !=
+      XRAY_LOG_INITIALIZED) {
+    std::cerr << "Failed initializing xray-profiling mode. Aborting.\n";
+    std::abort();
+  };
+
+  __xray_patch();
+  go.store(true, std::memory_order_release);
+}
+
+}  // namespace
+
+#define XRAY_WEAK_NOINLINE                                \
+  [[clang::xray_always_instrument]] __attribute__((weak)) \
+      __attribute__((noinline))
+
+XRAY_WEAK_NOINLINE void deep8() {
+  some_global.fetch_add(1, std::memory_order_relaxed);
+}
+XRAY_WEAK_NOINLINE void deep7() { deep8(); }
+XRAY_WEAK_NOINLINE void deep6() { deep7(); }
+XRAY_WEAK_NOINLINE void deep5() { deep6(); }
+XRAY_WEAK_NOINLINE void deep4() { deep5(); }
+XRAY_WEAK_NOINLINE void deep3() { deep4(); }
+XRAY_WEAK_NOINLINE void deep2() { deep3(); }
+XRAY_WEAK_NOINLINE void deep1() { deep2(); }
+XRAY_WEAK_NOINLINE int deep0() {
+  deep1();
+  return some_global.load(std::memory_order_acquire);
+}
+
+[[clang::xray_never_instrument]] static void BM_XRayProfilingDeepCallStack(
+    benchmark::State &state) {
+  std::call_once(profiling_init_flag, profiling_init);
+
+  // Spin until we see the flag set.
+  while (!go.load(std::memory_order_acquire))
+    ;
+
+  // Warm up the data structures.
+  benchmark::DoNotOptimize(some_temporary = deep0());
+
+  for (auto _ : state) benchmark::DoNotOptimize(some_temporary = deep0());
+}
+BENCHMARK(BM_XRayProfilingDeepCallStack)->ThreadRange(1, 32);
+
+BENCHMARK_MAIN();
Index: MicroBenchmarks/XRay/ProfilingMode/shallow-call-bench.cc
===================================================================
--- /dev/null
+++ MicroBenchmarks/XRay/ProfilingMode/shallow-call-bench.cc
@@ -0,0 +1,71 @@
+//===- profilingmode-bench.cc - XRay Profiling Mode Benchmarks ------------===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+//
+// These benchmarks measure the cost of XRay profiling mode when enabled.
+//
+//===----------------------------------------------------------------------===//
+
+#include <atomic>
+#include <iostream>
+#include <mutex>
+#include <thread>
+#include "benchmark/benchmark.h"
+#include "xray/xray_log_interface.h"
+
+namespace {
+
+std::atomic<int> some_global{0};
+
+std::atomic<int> some_temporary{0};
+
+std::once_flag profiling_init_flag;
+
+std::atomic<bool> go{false};
+
+[[clang::xray_never_instrument]] static void profiling_init() {
+  if (__xray_log_select_mode("xray-profiling") != XRAY_REGISTRATION_OK) {
+    std::cerr << "Failed selecting 'xray-profiling' mode. Aborting.\n";
+    std::abort();
+  }
+
+  if (__xray_log_init_mode("xray-profiling", "no_flush=true") !=
+      XRAY_LOG_INITIALIZED) {
+    std::cerr << "Failed initializing xray-profiling mode. Aborting.\n";
+    std::abort();
+  };
+
+  __xray_patch();
+  go.store(true, std::memory_order_release);
+}
+
+}  // namespace
+
+#define XRAY_WEAK_NOINLINE                                \
+  [[clang::xray_always_instrument]] __attribute__((weak)) \
+      __attribute__((noinline))
+
+XRAY_WEAK_NOINLINE int shallow() {
+  return some_global.fetch_add(1, std::memory_order_acq_rel);
+}
+
+[[clang::xray_never_instrument]] static void BM_XRayProfilingShallowStack(
+    benchmark::State &state) {
+  std::call_once(profiling_init_flag, profiling_init);
+
+  while (!go.load(std::memory_order_acquire))
+    ;
+
+  // Warm up the data structures.
+  benchmark::DoNotOptimize(some_temporary = shallow());
+
+  for (auto _ : state) benchmark::DoNotOptimize(some_temporary = shallow());
+}
+BENCHMARK(BM_XRayProfilingShallowStack)->ThreadRange(1, 32);
+
+BENCHMARK_MAIN();