[Apache TVM Discuss] [Questions] How to use microTVM with mimxrt1060_evk

I notice that microTVM can directly support mimxrt1050_evk, but I'm working on one mimxrt1060_evk. I'm wondering whether microTVM can also run on 1060_evk, because these two boards are quite similar, they both belong to rt10xx and cortex-m7 boards, they are both supported by zephyr. So far, I

[Apache TVM Discuss] [Questions] [BYOC] Is there a way to export accelerator specific mesurements

Hi @MartinF , I have actually not started yet with this integration. I am planning on adding this in the near future, but sadly I dont know yet when I will have the time to check this. --- [Visit Topic](https://discuss.tvm.apache.org/t/byoc-is-there-a-way-to-export-accelerator-specific-m

[Apache TVM Discuss] [Questions] How do I run my CUDA code in TVM?

I added a new back end to TVM and have generated the code I need, the code itself can run on the new device, but how should I run it in TVM? If you don't know, but know how to run cuda source codes generated by my own codegen tool, it might help me too. --- [Visit Topic](https://discuss.

[Apache TVM Discuss] [Questions] TensorIR: Scheduling the init field of block

Hi everyone, here a tricky TensorIR question ;) When looking at the TIR below, how can I unroll the block `T_matmul_NT_init` ? Using `get_block("T_matmul_NT_init")` runs into an FFI error: the check here: https://github.com/apache/tvm/blob/main/src/tir/schedule/primitive/get_block_loop.cc#L31