| /external/tensorflow/tensorflow/stream_executor/cuda/ |
| D | cuda_gpu_executor.cc | 400 port::Status GpuExecutor::Launch(Stream* stream, const ThreadDim& thread_dims, in Launch() argument
415 VlogOccupancyInfo(kernel, thread_dims, block_dims); in Launch()
431 context_, cufunc, block_dims.x, block_dims.y, block_dims.z, thread_dims.x, in Launch()
432 thread_dims.y, thread_dims.z, args.number_of_shared_bytes(), custream, in Launch()
440 const ThreadDim& thread_dims, in VlogOccupancyInfo() argument
444 VLOG(2) << "Thread dimensions (" << thread_dims.x << ", " << thread_dims.y in VlogOccupancyInfo()
445 << ", " << thread_dims.z << ")"; in VlogOccupancyInfo()
464 smem_per_block, thread_dims, cufunc); in VlogOccupancyInfo()
469 smem_per_block, thread_dims, cufunc); in VlogOccupancyInfo()
484 const ThreadDim& thread_dims, in CalculateOccupancy() argument
[all …]
|
| /external/tensorflow/tensorflow/stream_executor/gpu/ |
| D | gpu_executor.h | 72 port::Status Launch(Stream* stream, const ThreadDim& thread_dims,
80 const ThreadDim& thread_dims, GpuFunctionHandle func);
87 const ThreadDim& thread_dims, GpuFunctionHandle func);
267 void VlogOccupancyInfo(const KernelBase& kernel, const ThreadDim& thread_dims,
|
| /external/tensorflow/tensorflow/stream_executor/rocm/ |
| D | rocm_gpu_executor.cc | 296 port::Status GpuExecutor::Launch(Stream* stream, const ThreadDim& thread_dims, in Launch() argument
311 VlogOccupancyInfo(kernel, thread_dims, block_dims); in Launch()
344 thread_dims.x, thread_dims.y, thread_dims.z, in Launch()
351 const ThreadDim& thread_dims, in CalculateOccupancy() argument
361 const ThreadDim& thread_dims, in CompareOccupancy() argument
420 const ThreadDim& thread_dims, in VlogOccupancyInfo() argument
|
| /external/tensorflow/tensorflow/stream_executor/ |
| D | trace_listener.h | 49 virtual void LaunchSubmit(Stream* stream, const ThreadDim& thread_dims, in LaunchSubmit() argument
|
| D | stream_executor_pimpl.cc | 465 const ThreadDim &thread_dims, in Launch() argument
469 SubmitTrace(&TraceListener::LaunchSubmit, stream, thread_dims, block_dims, in Launch()
472 return implementation_->Launch(stream, thread_dims, block_dims, kernel, args); in Launch()
|
| D | stream_executor_internal.h | 190 virtual port::Status Launch(Stream *stream, const ThreadDim &thread_dims, in Launch() argument
|
| D | stream_executor_pimpl.h | 474 port::Status Launch(Stream *stream, const ThreadDim &thread_dims,
|
| D | stream.h | 185 Stream &ThenLaunch(ThreadDim thread_dims, BlockDim block_dims,
2119 inline Stream &Stream::ThenLaunch(ThreadDim thread_dims, BlockDim block_dims, in ThenLaunch() argument
2134 parent_->Launch(this, thread_dims, block_dims, kernel, kernel_args) in ThenLaunch()
|
| /external/tensorflow/tensorflow/stream_executor/host/ |
| D | host_gpu_executor.h | 57 port::Status Launch(Stream *stream, const ThreadDim &thread_dims, in Launch() argument
|
| /external/tensorflow/tensorflow/compiler/xla/service/interpreter/ |
| D | executor.h | 64 port::Status Launch(Stream *stream, const ThreadDim &thread_dims, in Launch() argument
|
| /external/llvm-project/polly/lib/External/ppcg/ |
| D | cuda.c | 394 const char *thread_dims[] = { "threadIdx.x", "threadIdx.y", in print_kernel_iterators() local
400 print_iterators(out, type, kernel->thread_ids, thread_dims); in print_kernel_iterators()
|