MayaFlux 1.0.0
Digital-First Multimedia Processing Framework
Loading...
Searching...
No Matches

◆ share_recording()

template<ComputeData InputType = Eigen::MatrixXd, ComputeData OutputType = std::vector<std::vector<double>>>
void MayaFlux::Yantra::IntervalExtractor< InputType, OutputType >::share_recording ( std::shared_ptr< GpuDispatchRecorder > recording)
inline

Append to a recording the caller shares with other work.

With a shared recording the extractor's scans go to the device in the same submission as the work that produced their scores. Call it before anything creates the extractor's context, that is before the first extraction or bind. The recording must be compute isolated.

Exceptions
std::logic_errorwhen the extractor has already started.

Definition at line 336 of file IntervalExtractor.hpp.

337 {
338 if (m_context) {
340 std::source_location::current(), "The recording can only be shared before the extractor starts");
341 }
343 }
std::unique_ptr< ShaderExecutionContext<> > m_context
std::shared_ptr< GpuDispatchRecorder > m_recording
Finds where each score sequence is engaged, on the GPU, as a stream.