MCPcopy Create free account
hub / github.com/DeepRec-AI/DeepRec / GetKernelMetadata

Method GetKernelMetadata

tensorflow/stream_executor/rocm/rocm_gpu_executor.cc:284–294  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

282}
283
284bool GpuExecutor::GetKernelMetadata(GpuKernel* rocm_kernel,
285 KernelMetadata* kernel_metadata) {
286 int value = 0;
287 // TODO(ROCm) implement this feature in HIP
288 kernel_metadata->set_registers_per_thread(value);
289
290 // TODO(ROCm) implement this feature in HIP
291 kernel_metadata->set_shared_memory_bytes(value);
292
293 return true;
294}
295
296port::Status GpuExecutor::Launch(Stream* stream, const ThreadDim& thread_dims,
297 const BlockDim& block_dims,

Callers

nothing calls this directly

Calls 2

Tested by

no test coverage detected