[Mlir-commits] [mlir] [mlir][nfc] volunteering for quant dialect (PR #152852)

[Andrzej Warzyński]

banach-space wrote:

Thank you for volunteering, @javedabsar1 🙏🏻

As noted in [the MLIR maintainers list](https://discourse.llvm.org/t/mlir-project-maintainers/), there are currently no maintainers for the quant dialect, so it’s great to see renewed interest in keeping it healthy.

In general, I like to see more direct involvement before assigning maintainership  - especially contributions beyond test updates. But that’s just a rule of thumb. Quant has been dormant for a long time, and it’s not clear what the right bar should be in this case. Also, the whole topic of criteria is still taking shape (I still owe some comments myself!):
[WIP Upstream LLVM Maintainer Policy
](https://docs.google.com/document/d/13i-wU9s-0NoZtBnQn67dedtyc676nZliPVXFyE5u280/edit?tab=t.0#heading=h.rynge1kd90is)

At the very least, it might be helpful to reconnect with folks who were recently active in this area. This seems to be the most recent substantial discussion:
* https://discourse.llvm.org/t/rfc-improvements-in-the-quant-dialect/

Now, we all know you through your much appreciated contributions to Linalg and your many Dev Meeting presentations  - so maybe that kind of broader, cross-cutting involvement is something we should take into account in cases like this? I don’t have a strong opinion either way - these are still early days and the process is very much in flux 😅

I personally don’t have a stake in **quant** and don’t know much about quantization, but just speaking with my “tensor compiler maintainer” hat on: I’d feel more comfortable seeing a bit more engagement or coordination with past contributors - even if that’s just some shared roadmap or discussion.

https://github.com/llvm/llvm-project/pull/152852