[StreamExecutor] Rename ::perftools::gputools -> ::stream_executor, part 1.

Step 1 of re-namespace'ing StreamExecutor into ::stream_executor.

This moves everything inside of stream_executor/..., and leaves a
namespace alias into ::perftools::gputools.  The next steps will clean
up users to use the new namespace.

This is mostly a mechanical change, but it also includes a bunch of
non-mechanical changes that ideally would be split out into separate
patches.  Unfortunately they all sort of need to be shoved in here for
various reasons:

 - forward declarations need to be in the same namespace as the actual
   types, so we need to change all forward declarations of
   StreamExecutor types in this one patch.

 - Uses of these forward declarations need to be changed to the new
   namespace (or otherwise we need to add a namespace alias to the
   relevant header, but this is pretty ugly).

 - Various initialization code needs to live in StreamExecutor's "real"
   namespace, so all this needs to be changed.

PiperOrigin-RevId: 193256128

1 files changed, 13 insertions, 11 deletions

diff --git a/tensorflow/compiler/xla/executable_run_options.h b/tensorflow/compiler/xla/executable_run_options.h
index 2c1d9ffff1..1a095a82cc 100644
--- a/tensorflow/compiler/xla/executable_run_options.h
+++ b/tensorflow/compiler/xla/executable_run_options.h
@@ -16,29 +16,31 @@ limitations under the License.
 #ifndef TENSORFLOW_COMPILER_XLA_EXECUTABLE_RUN_OPTIONS_H_
 #define TENSORFLOW_COMPILER_XLA_EXECUTABLE_RUN_OPTIONS_H_
 
-// Intentionally forward declared so that ExecutableRunOptions can be linked
+// These classes are forward declared so that ExecutableRunOptions can be linked
 // into an XLA-compiled binary without having to link all of the pointed-to
 // objects (e.g., for an ahead-of-time compiled CPU binary, the gpu tools don't
 // need to be linked).
-namespace perftools {
-namespace gputools {
+namespace stream_executor {
 class Stream;
 class Platform;
-}
-}
+}  // namespace stream_executor
 
 namespace tensorflow {
 namespace thread {
 class ThreadPool;
-}
-}
+}  // namespace thread
+}  // namespace tensorflow
 
 namespace Eigen {
 struct ThreadPoolDevice;
-}
+}  // namespace Eigen
 
 namespace xla {
 
+// TODO(b/77980417): Once the perftools::gputools -> stream_executor migration
+// is complete, add "using namespace se = stream_executor" here and
+// s/stream_executor/se::/ to match our idiom elsewhere.
+
 class DeviceMemoryAllocator;
 class DeviceAssignment;
 class ExecutionProfile;
@@ -61,8 +63,8 @@ class ExecutableRunOptions {
   // If set, this is the stream to run the computation on. The platform of the
   // stream must match the platform the executable was built for.  A value of
   // nullptr indicates the option has not been set.
-  ExecutableRunOptions& set_stream(perftools::gputools::Stream* stream);
-  perftools::gputools::Stream* stream() const;
+  ExecutableRunOptions& set_stream(stream_executor::Stream* stream);
+  stream_executor::Stream* stream() const;
 
   // Sets the thread pool on which to run parallel CPU backend
   // computations. Does not take ownership.
@@ -91,7 +93,7 @@ class ExecutableRunOptions {
   DeviceMemoryAllocator* allocator_ = nullptr;
   int device_ordinal_ = -1;
   DeviceAssignment* device_assignment_ = nullptr;
-  perftools::gputools::Stream* stream_ = nullptr;
+  stream_executor::Stream* stream_ = nullptr;
   tensorflow::thread::ThreadPool* inter_op_thread_pool_ = nullptr;
   const Eigen::ThreadPoolDevice* intra_op_thread_pool_ = nullptr;
   ExecutionProfile* execution_profile_ = nullptr;