diff options
| author |  A. Unique TensorFlower <gardener@tensorflow.org> | 2017-11-27 19:48:53 -0800 |
|---|---|---|
| committer | TensorFlower Gardener <gardener@tensorflow.org> | 2017-11-27 19:51:57 -0800 |
| commit | 587dbb404318039c37d7587b6ac5d044504d0ad1 (patch) | |
| tree | f80e4e0257e6a64ca02104a1489b41a416f3a412 | |
| parent | e02be0161393d255d8e1ade54c2a885b9695beab (diff) | |
[XLA] Remove the extra RunHloOptimization in InterpreterCompiler::RunBackend.
PiperOrigin-RevId: 177107209
| -rw-r--r-- | tensorflow/compiler/xla/service/interpreter/compiler.cc | 2 |
1 files changed, 0 insertions, 2 deletions
diff --git a/tensorflow/compiler/xla/service/interpreter/compiler.cc b/tensorflow/compiler/xla/service/interpreter/compiler.cc index c9a5285a4f..dc63a2224d 100644 --- a/tensorflow/compiler/xla/service/interpreter/compiler.cc +++ b/tensorflow/compiler/xla/service/interpreter/compiler.cc @@ -83,8 +83,6 @@ StatusOr<std::unique_ptr<Executable>> InterpreterCompiler::RunBackend( VLOG(1) << "Run backend " << hlo_module->name(); - TF_RETURN_IF_ERROR(RunHloOptimization(hlo_module.get())); - // Typically you would visit the HLO graph, building up a compiled equivalent // In this case we are using an HloEvaluator at execution time, so we don't // need to compile anything |