Hi @stoa, thanks for your observations and apologies for not replying sooner.
I'm glad you agree with the initial direction taken and I appreciate that there may be a need to provide similar data to the user to that which is seen in DLTensor, at this stage we'll get that by providing the option to return to the packed function signature. Further work should be done to ensure we have all the data available for an embedded system; to help motivate that conversation I've published a companion RFC - https://discuss.tvm.apache.org/t/rfc-utvm-embedded-c-runtime-interface/9951, hopefully this helps alleviate the concerns around metadata access and clarifies how this work can be accessed from an embedded system :smile_cat: --- [Visit Topic](https://discuss.tvm.apache.org/t/rfc-utvm-aot-optimisations-for-embedded-targets/9849/10) to respond. You are receiving this because you enabled mailing list mode. To unsubscribe from these emails, [click here](https://discuss.tvm.apache.org/email/unsubscribe/ec7964f8b61af490ab84f4d6e653193075792db4e59aca7a1429d8e9e9d7f24b).
