diff --git a/tensorflow/compiler/xla/service/hlo_module_config.cc b/tensorflow/compiler/xla/service/hlo_module_config.cc
index 822e2f1f53e..4205b0402cb 100644
--- a/tensorflow/compiler/xla/service/hlo_module_config.cc
+++ b/tensorflow/compiler/xla/service/hlo_module_config.cc
@@ -40,7 +40,7 @@ void HloModuleConfig::SetDefaultComputationLayout(
 
 string HloModuleConfig::compilation_cache_key() const {
   string key =
-      tensorflow::strings::StrCat("profiling=", hlo_profiling_enabled_);
+      tensorflow::strings::StrCat("profiling=", hlo_profiling_enabled());
   StrAppend(&key, "::(");
   std::vector<string> params;
   for (const ShapeLayout& param_layout :
diff --git a/tensorflow/compiler/xla/service/hlo_module_config.h b/tensorflow/compiler/xla/service/hlo_module_config.h
index d3c1fae592b..586a03d4126 100644
--- a/tensorflow/compiler/xla/service/hlo_module_config.h
+++ b/tensorflow/compiler/xla/service/hlo_module_config.h
@@ -63,9 +63,10 @@ class HloModuleConfig {
     return &(*entry_computation_layout_);
   }
 
-  // Sets/returns whether to enable HLO-level profiling.
-  bool hlo_profiling_enabled() const { return hlo_profiling_enabled_; }
-  void enable_hlo_profiling(bool enabled) { hlo_profiling_enabled_ = enabled; }
+  // Returns whether to enable HLO-level profiling.
+  bool hlo_profiling_enabled() const {
+    return debug_options_.xla_hlo_profile();
+  }
 
   // Sets/returns whether this is a "host module".  Host modules are used to
   // record the data- and control-flow dependencies of host side computation
@@ -110,9 +111,6 @@ class HloModuleConfig {
 
   tensorflow::gtl::optional<ComputationLayout> entry_computation_layout_;
 
-  // Whether to enable HLO-level profiling.
-  bool hlo_profiling_enabled_ = false;
-
   // Whether this is a 'host module'.
   bool is_host_module_ = false;
 
diff --git a/tensorflow/compiler/xla/service/hlo_runner.cc b/tensorflow/compiler/xla/service/hlo_runner.cc
index d65befaf84c..e5b1c2efa3f 100644
--- a/tensorflow/compiler/xla/service/hlo_runner.cc
+++ b/tensorflow/compiler/xla/service/hlo_runner.cc
@@ -158,8 +158,8 @@ StatusOr<std::unique_ptr<Literal>> HloRunner::Execute(
 
   TF_ASSIGN_OR_RETURN(
       std::unique_ptr<ShapedBuffer> result,
-      executable->ExecuteOnStream(&service_run_options, argument_buffer_ptrs,
-                                  /*hlo_execution_profile=*/nullptr));
+      executable->ExecuteOnStreamWrapper(
+          &service_run_options, /*profile=*/nullptr, argument_buffer_ptrs));
 
   // Create a ScopedShapedBuffer of the result to manage deallocation. This will
   // deallocate all the device memory when it goes out of scope.
diff --git a/tensorflow/compiler/xla/service/service.cc b/tensorflow/compiler/xla/service/service.cc
index 8edd457281b..0becc9d8f8e 100644
--- a/tensorflow/compiler/xla/service/service.cc
+++ b/tensorflow/compiler/xla/service/service.cc
@@ -314,8 +314,6 @@ StatusOr<std::unique_ptr<HloModuleConfig>> Service::CreateModuleConfig(
   if (execution_options != nullptr) {
     config->set_seed(execution_options->seed());
     config->set_debug_options(execution_options->debug_options());
-    config->enable_hlo_profiling(
-        execution_options->debug_options().xla_hlo_profile());
   } else {
     config->set_debug_options(legacy_flags::GetDebugOptionsFromFlags());
   }