aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/compiler/xla/service/llvm_compiler.cc
diff options
context:
space:
mode:
authorGravatar Sanjoy Das <sanjoy@google.com>2017-11-13 21:05:09 -0800
committerGravatar TensorFlower Gardener <gardener@tensorflow.org>2017-11-13 21:09:38 -0800
commitb10d5b2545eb2302c370c3d472099b3afd6baba5 (patch)
treed2676588a791015e5abd3302ae36450a56eebd14 /tensorflow/compiler/xla/service/llvm_compiler.cc
parentfe3a35a12304c4a6eeaffc208e3c8a006d80455f (diff)
[XLA:CPU/GPU] Implement multi-module compilation for the CPU and GPU backends
For CPU and GPU this is a simple wrapper around the single-module Compile method since the CPU and GPU backends do not perform cross-module optimizations and analyses. PiperOrigin-RevId: 175631791
Diffstat (limited to 'tensorflow/compiler/xla/service/llvm_compiler.cc')
-rw-r--r--tensorflow/compiler/xla/service/llvm_compiler.cc37
1 files changed, 37 insertions, 0 deletions
diff --git a/tensorflow/compiler/xla/service/llvm_compiler.cc b/tensorflow/compiler/xla/service/llvm_compiler.cc
new file mode 100644
index 0000000000..ba0304fb8c
--- /dev/null
+++ b/tensorflow/compiler/xla/service/llvm_compiler.cc
@@ -0,0 +1,37 @@
+/* Copyright 2017 The TensorFlow Authors. All Rights Reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+ http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+==============================================================================*/
+
+#include "tensorflow/compiler/xla/service/llvm_compiler.h"
+
+namespace xla {
+StatusOr<std::vector<std::unique_ptr<Executable>>> LLVMCompiler::Compile(
+ std::vector<std::unique_ptr<HloModule>> modules,
+ std::vector<std::vector<perftools::gputools::StreamExecutor*>>
+ stream_execs) {
+ std::vector<std::unique_ptr<Executable>> result;
+ for (size_t i = 0; i < modules.size(); i++) {
+ if (stream_execs[i].size() != 1) {
+ return Unimplemented(
+ "Model partitioning not implemented for the CPU/GPU compilers!");
+ }
+
+ TF_ASSIGN_OR_RETURN(std::unique_ptr<Executable> executable,
+ Compile(std::move(modules[i]), stream_execs[i][0]));
+ result.push_back(std::move(executable));
+ }
+
+ return {std::move(result)};
+}
+} // namespace xla