Clarification needed for nvds_acquire_meta_lock/unlock

[hyperlight]

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU): GPU
• DeepStream Version: 6.4
• JetPack Version (valid for Jetson only): NA
• TensorRT Version: 8.6
• NVIDIA GPU Driver Version (valid for GPU only): 535
• Issue Type( questions, new requirements, bugs): questions
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing): na
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description): na

What operations must absolutely be done inside the the critical section between nvds_acquire_meta_lock & nvds_acquire_meta_unlock?

I’ve looked into multiple examples & plugin source code that has attach_metadata function and I can’t find the common denominator among them to answer the above question.

For example:

• osd_sink_pad_buffer_probe() in deepstream_test1_app.c has no critical section
• attach_metadata_object() in gstdsexample_optimized.cpp put nvds_acquire_classifier_meta_from_pool() outside the critical section
• attach_metadata_classifier() in gstnvinfer_meta_utils.cpp put everything (acquire, modify, attach) inside the critical section
• attachMetadata() in post_processor_classify.cpp put everything (acquire, modify, attach) inside the critical section

I infer from example code the following:

• Acquire metadata: can be outside critical section
• Modify acquired metadata before attachment: can be outside critical section
• Attach metadata: can be outside critical section
• Modify existing metadata: must be inside critical section

I’m not sure if the rules above are correct and complete. So what operations must absolutely be done inside the the critical section between nvds_acquire_meta_lock & nvds_acquire_meta_unlock?

[yuweiw]

You can check meta_mutex in the NvDsBatchMeta structure. Theoretically, it should be used when multiple components are modifying the NvDsBatchMeta at the same time.