| /external/tensorflow/tensorflow/compiler/xla/service/gpu/ |
| D | amdgpu_compiler.cc | 70 HloModule* hlo_module, se::StreamExecutor* stream_exec, in OptimizeHloConvolutionCanonicalization() 87 HloModule* hlo_module, se::StreamExecutor* stream_exec, in OptimizeHloPostLayoutAssignment() 119 GpuVersion AMDGPUCompiler::GetGpuVersion(se::StreamExecutor* stream_exec) { in GetGpuVersion() 135 se::StreamExecutor* stream_exec) { in CompileTargetBinary()
|
| D | nvptx_compiler.cc | 113 HloModule* hlo_module, se::StreamExecutor* stream_exec, in OptimizeHloConvolutionCanonicalization() 155 HloModule* hlo_module, se::StreamExecutor* stream_exec, in OptimizeHloPostLayoutAssignment() 326 GpuVersion NVPTXCompiler::GetGpuVersion(se::StreamExecutor* stream_exec) { in GetGpuVersion() 343 se::StreamExecutor* stream_exec) { in CompileTargetBinary() 389 se::StreamExecutor* stream_exec, const string& ptx, int cc_major, in CompileGpuAsmOrGetCachedResult()
|
| D | gemm_algorithm_picker.h | 32 GemmAlgorithmPicker(se::StreamExecutor* stream_exec, in GemmAlgorithmPicker()
|
| D | gpu_compiler.cc | 120 HloModule* hlo_module, se::StreamExecutor* stream_exec, in OptimizeHloModule() 318 std::unique_ptr<HloModule> module, se::StreamExecutor* stream_exec, in RunHloPasses() 334 std::unique_ptr<HloModule> module, se::StreamExecutor* stream_exec, in RunBackend()
|
| D | gpu_conv_algorithm_picker.h | 40 GpuConvAlgorithmPicker(se::StreamExecutor* stream_exec, in GpuConvAlgorithmPicker()
|
| D | stream_executor_util.cc | 176 tensorflow::mutex_lock LockGpu(const se::StreamExecutor* stream_exec) { in LockGpu() 195 absl::Span<const uint8> cubin_data, se::StreamExecutor* stream_exec) { in CreateKernel()
|
| D | gpu_conv_algorithm_picker.cc | 100 se::StreamExecutor* stream_exec) { in GetAlgorithms() 124 se::DeviceMemoryBase result_buffer, se::StreamExecutor* stream_exec, in GetAlgorithms()
|
| /external/tensorflow/tensorflow/compiler/xla/tests/ |
| D | llvm_compiler_test.cc | 45 HloModule* hlo_module, se::StreamExecutor* stream_exec, in OptimizeHloConvolutionCanonicalization() 51 HloModule* hlo_module, se::StreamExecutor* stream_exec, in OptimizeHloPostLayoutAssignment() 56 GpuVersion GetGpuVersion(se::StreamExecutor* stream_exec) { return 0; } in GetGpuVersion() 60 GpuVersion gpu_version, se::StreamExecutor* stream_exec) { in CompileTargetBinary()
|
| /external/tensorflow/tensorflow/core/common_runtime/gpu/ |
| D | gpu_debug_allocator_test.cc | 45 auto stream_exec = in TEST() local 145 auto stream_exec = in TEST() local 192 auto stream_exec = in TEST() local
|
| D | gpu_event_mgr_test.cc | 117 auto stream_exec = GPUMachineManager()->ExecutorForDevice(0).ValueOrDie(); in TEST() local 134 auto stream_exec = GPUMachineManager()->ExecutorForDevice(0).ValueOrDie(); in TEST() local 167 auto stream_exec = GPUMachineManager()->ExecutorForDevice(0).ValueOrDie(); in TEST() local 184 auto stream_exec = GPUMachineManager()->ExecutorForDevice(0).ValueOrDie(); in TEST() local 203 auto stream_exec = GPUMachineManager()->ExecutorForDevice(0).ValueOrDie(); in TEST() local 225 auto stream_exec = GPUMachineManager()->ExecutorForDevice(0).ValueOrDie(); in TEST() local 247 auto stream_exec = GPUMachineManager()->ExecutorForDevice(0).ValueOrDie(); in TEST() local 266 auto stream_exec = GPUMachineManager()->ExecutorForDevice(0).ValueOrDie(); in TEST() local 589 auto stream_exec = GPUMachineManager()->ExecutorForDevice(0).ValueOrDie(); in BM_no_ops() local
|
| D | gpu_host_allocator.h | 29 explicit GpuHostAllocator(se::StreamExecutor* stream_exec, int numa_node, in GpuHostAllocator()
|
| D | gpu_mem_allocator.h | 31 explicit GPUMemAllocator(se::StreamExecutor* stream_exec, in GPUMemAllocator()
|
| /external/tensorflow/tensorflow/compiler/xla/ |
| D | device_util.h | 32 string DeviceIdentifier(se::StreamExecutor* stream_exec) { in DeviceIdentifier()
|
| /external/tensorflow/tensorflow/compiler/xla/service/mlir_gpu/ |
| D | failover_compiler.cc | 30 std::unique_ptr<HloModule> module, se::StreamExecutor* stream_exec, in RunHloPasses() 42 std::unique_ptr<HloModule> module, se::StreamExecutor* stream_exec, in RunBackend()
|
| D | mlir_compiler.cc | 142 std::unique_ptr<HloModule> module, se::StreamExecutor* stream_exec, in RunHloPasses() 176 GpuVersion GetGpuVersion(se::StreamExecutor* stream_exec) { in GetGpuVersion() 457 std::unique_ptr<HloModule> module, se::StreamExecutor* stream_exec, in RunBackend()
|
| /external/tensorflow/tensorflow/compiler/xla/service/interpreter/ |
| D | compiler.cc | 101 std::unique_ptr<HloModule> hlo_module, se::StreamExecutor* stream_exec, in RunBackend() 126 std::vector<std::vector<se::StreamExecutor*>> stream_exec, in Compile()
|
| /external/tensorflow/tensorflow/stream_executor/ |
| D | event.cc | 24 Event::Event(StreamExecutor* stream_exec) in Event()
|
| D | stream_executor_pimpl.cc | 74 ScopedTracer(StreamExecutor *stream_exec, BeginCallT begin_call, in ScopedTracer() 115 MakeScopedTracer(StreamExecutor *stream_exec, BeginCallT begin_call, in MakeScopedTracer()
|
| /external/tensorflow/tensorflow/stream_executor/gpu/ |
| D | gpu_activation.cc | 34 StreamExecutor* stream_exec) in ScopedActivateExecutorContext()
|
| /external/tensorflow/tensorflow/stream_executor/cuda/ |
| D | redzone_allocator_test.cc | 55 StreamExecutor* stream_exec = platform->ExecutorForDevice(0).ValueOrDie(); in TEST() local 130 StreamExecutor* stream_exec = platform->ExecutorForDevice(0).ValueOrDie(); in TEST() local
|
| D | cuda_gpu_executor.cc | 118 GpuExecutor* ExtractGpuExecutor(StreamExecutor* stream_exec) { in ExtractGpuExecutor()
|
| /external/tensorflow/tensorflow/core/kernels/ |
| D | conv_ops_gpu.h | 34 inline bool IsVoltaOrLater(const se::StreamExecutor& stream_exec) { in IsVoltaOrLater() 164 se::StreamExecutor* stream_exec) const { in ShouldIncludeWinogradNonfusedAlgo()
|
| D | gpu_utils.cc | 135 se::StreamExecutor* stream_exec, in LogConvAutotuneResults() 178 se::StreamExecutor* stream_exec, absl::Span<const AutotuneResult> results) { in LogFusedConvForwardAutotuneResults()
|
| /external/tensorflow/tensorflow/compiler/xla/service/cpu/ |
| D | cpu_compiler.cc | 560 std::unique_ptr<HloModule> module, se::StreamExecutor* stream_exec, in RunBackend()
|
| /external/tensorflow/tensorflow/stream_executor/rocm/ |
| D | rocm_gpu_executor.cc | 99 GpuExecutor* ExtractGpuExecutor(StreamExecutor* stream_exec) { in ExtractGpuExecutor()
|