MCPcopy Create free account
hub / github.com/NVIDIA/TensorRT / operator()

Method operator()

samples/common/sampleInference.cpp:594–606  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

592 }
593
594 bool operator()(TrtCudaStream& stream) const
595 {
596 if (mGraph.launch(stream))
597 {
598 // Collecting layer timing info from current profile index of execution context
599 if (mContext.getProfiler() && !mContext.getEnqueueEmitsProfile() && !mContext.reportToProfiler())
600 {
601 gLogWarning << "Failed to collect layer timing info from previous CUDA graph launch" << std::endl;
602 }
603 return true;
604 }
605 return false;
606 }
607
608 TrtCudaGraph& mGraph;
609 nvinfer1::IExecutionContext& mContext;

Callers

nothing calls this directly

Calls 4

getProfilerMethod · 0.80
reportToProfilerMethod · 0.80
launchMethod · 0.45

Tested by

no test coverage detected