Hi,
Thanks for your interest in this 
Regarding 1, this is a good point. Initially we were only intending to use rolling_buffer in our custom TIR lowering pipeline (doesn’t use driver_api.cc), but I agree it’d be much better to also include it in the standard flow. I’ll work on a patch to enable that.
For 2, it’s definitely a medium/long term ambition to migrate to TensorIR and I don’t see a technical reason that would prevent it. However, TensorIR is still not very mature so I think this will only be considered once TensorIR reaches full parity with TE. If anyone else wishes to migrate it sooner though that’d be welcomed by me.