Documentation and example of usage of cuMemExportToShareableHandle

[MaximeDebarbat]

It would be great to have a reproductible example of the usage of cuMemExportToShareableHandle.

I have spent hours now trying to use it unsuccessfully. My goal is simply to transfer cuda memory between 2 processes in 2 different containers, the nature of the data being media, it should be real time scalable. I thus trust cuMemExportToShareableHandle is the way to go.

[leofang]

Hi @MaximeDebarbat, sorry this fell through the crack, as noted elsewhere #129 (comment) please refer to CUDA C/C++ samples if you don't find what you need in this repo. For IPC through the mem export API, perhaps this could be a good starting point?
https://github.com/NVIDIA/cuda-samples/tree/9c688d7ff78455ed42e345124d1495aad6bf66de/Samples/3_CUDA_Features/memMapIPCDrv

[leofang]

Closing due to lack of response. Let us know if you have further questions!