Convolution op with scheduling in TVM

For implementing convolution op in TVM, with scheduling, I have followed these examples :

and this discussion link :

I wanted to know how does the TVM compilation flow work (in terms of scheduling, etc.) on ‘coarse-grained’ operations such as convolution.

Is this code reduced to primitive schedule operators (, and then code generation done on that ?

Or does it generate code directly from the high-level operations?