/external/tensorflow/tensorflow/stream_executor/lib/ |
human_readable.h | 25 namespace perftools { namespace 71 } // namespace perftools
|
path.cc | 19 namespace perftools { namespace 62 } // namespace perftools
|
process_state.cc | 28 namespace perftools { namespace 58 } // namespace perftools
|
/external/tensorflow/tensorflow/stream_executor/platform/ |
port.h | 16 // IWYU pragma: private, include "perftools/gputools/executor/stream_executor.h" 25 namespace perftools { namespace 54 } // namespace perftools
|
/external/tensorflow/tensorflow/stream_executor/ |
scratch_allocator.h | 26 namespace perftools { namespace 81 } // namespace perftools
|
platform.h | 32 namespace perftools { namespace 109 const perftools::gputools::Platform::Id ID_VAR_NAME = &plugin_id_value; 195 } // namespace perftools
|
/external/perf_data_converter/src/ |
builder.h | 17 namespace perftools { namespace 28 namespace perftools { namespace 118 } // namespace perftools
|
perf_data_handler_test.cc | 14 namespace perftools { namespace 62 } // namespace perftools
|
/external/tensorflow/tensorflow/compiler/xla/ |
executable_run_options.cc | 39 perftools::gputools::Stream* stream) { 44 perftools::gputools::Stream* ExecutableRunOptions::stream() const {
|
/external/tensorflow/tensorflow/compiler/xla/service/ |
backend.cc | 39 namespace se = ::perftools::gputools; 44 perftools::gputools::Platform* platform) { 49 perftools::gputools::Platform* BackendOptions::platform() const { 80 perftools::gputools::Platform* platform = options.platform(); 124 perftools::gputools::Platform* platform, Compiler* compiler, 181 StatusOr<perftools::gputools::StreamExecutor*> Backend::stream_executor( 204 TF_ASSIGN_OR_RETURN(perftools::gputools::StreamExecutor * executor_a, 206 TF_ASSIGN_OR_RETURN(perftools::gputools::StreamExecutor * executor_b,
|
transfer_manager.cc | 28 namespace se = ::perftools::gputools; 35 /* static */ std::map<perftools::gputools::Platform::Id, 39 new std::map<perftools::gputools::Platform::Id, TransferManager::State>; 44 perftools::gputools::StreamExecutor* executor, const Literal& literal, 45 const perftools::gputools::DeviceMemoryBase& dest) { 64 perftools::gputools::StreamExecutor* executor, const Shape& shape, 65 const perftools::gputools::DeviceMemoryBase& source) { 115 perftools::gputools::StreamExecutor* executor,
|
llvm_compiler.cc | 26 std::vector<std::vector<perftools::gputools::StreamExecutor*>> stream_execs,
|
llvm_compiler.h | 63 // perftools::gputools::StreamExecutor* stream_exec, 67 // perftools::gputools::StreamExecutor* stream_exec, 74 std::vector<std::vector<perftools::gputools::StreamExecutor*>>
|
/external/tensorflow/tensorflow/compiler/xla/service/gpu/ |
conditional_thunk.cc | 46 perftools::gputools::Stream* stream) { 49 perftools::gputools::DeviceMemoryBase predicate_address =
|
while_thunk.cc | 44 perftools::gputools::Stream* stream) { 45 perftools::gputools::DeviceMemoryBase condition_result_data =
|
cudnn_batchnorm_thunk.h | 63 perftools::gputools::Stream* stream) override; 93 perftools::gputools::Stream* stream) override; 126 perftools::gputools::Stream* stream) override;
|
gpu_compiler.h | 48 // std::vector<std::vector<perftools::gputools::StreamExecutor*>> 54 perftools::gputools::StreamExecutor* stream_exec, 59 perftools::gputools::StreamExecutor* stream_exec, 66 perftools::gputools::Platform::Id PlatformId() const override;
|
for_thunk.cc | 40 perftools::gputools::Stream* stream) {
|
/external/tensorflow/tensorflow/core/common_runtime/gpu/ |
gpu_debug_allocator.h | 58 perftools::gputools::StreamExecutor* stream_exec_; // Not owned. 84 perftools::gputools::StreamExecutor* stream_exec_; // Not owned.
|
gpu_id_utils.h | 26 namespace gpu = ::perftools::gputools;
|
/external/tensorflow/tensorflow/core/kernels/ |
batch_matmul_op_impl.h | 248 perftools::gputools::DeviceMemory<T> AsDeviceMemory(const T* cuda_memory) { 249 perftools::gputools::DeviceMemoryBase wrapped(const_cast<T*>(cuda_memory)); 250 perftools::gputools::DeviceMemory<T> typed(wrapped); 254 class CublasScratchAllocator : public perftools::gputools::ScratchAllocator { 256 using Stream = ::perftools::gputools::Stream; 257 using DeviceMemoryBytes = ::perftools::gputools::DeviceMemory<uint8>; 263 perftools::gputools::port::StatusOr<DeviceMemoryBytes> AllocateBytes( 270 return perftools::gputools::port::StatusOr<DeviceMemoryBytes>( 276 return perftools::gputools::port::StatusOr<DeviceMemoryBytes>( 292 constexpr perftools::gputools::blas::Transpose kTranspose [all...] |
/external/tensorflow/tensorflow/contrib/nccl/kernels/ |
nccl_manager.cc | 27 using ::perftools::gputools::cuda::ScopedActivateExecutorContext; 40 perftools::gputools::StreamExecutor* executor = nullptr; 44 std::unique_ptr<perftools::gputools::Stream> stream; 96 perftools::gputools::Stream* tensor_stream, 97 perftools::gputools::StreamExecutor* executor, int gpu_device_id, 122 perftools::gputools::Stream* const tensor_stream; 126 perftools::gputools::StreamExecutor* const executor = nullptr; 246 nccl_stream->stream.reset(new perftools::gputools::Stream(executor)); 301 perftools::gputools::StreamExecutor* executor, 303 perftools::gputools::Stream* tensor_stream [all...] |
/external/tensorflow/tensorflow/compiler/xla/service/interpreter/ |
platform.h | 26 namespace perftools { namespace 68 } // namespace perftools
|
/external/tensorflow/tensorflow/compiler/xla/tests/ |
test_utils.h | 71 Status VerifyHloModule(const perftools::gputools::Platform& platform,
|
/external/tensorflow/tensorflow/stream_executor/cuda/ |
cuda_diagnostics.h | 25 namespace perftools { namespace 97 } // namespace perftools
|