Re: [RFC PATCH 0/4] mm: compaction: mTHP-friendly memory compaction

[Bo Zhang]

Hi Lorenzo,

Thanks for the note. Let me clarify the background of this series, both
regarding the AI-assistance question and regarding the scope of the change.

This work started as part of our exploration of 16KB mTHP for the Android
scenario. The overall direction of making compaction lighter-weight and
mTHP-aware came from offline discussions with David, who suggested this
lightweight-compaction angle for small mTHP orders. So the size and scope
of the series is not me diving in unguided as a newcomer: it follows his
directional guidance. Barry then provided reviews and change suggestions
throughout the iterations.

On the AI-assistance side:

- All the code was written by me, by hand. I understand each change and
  take full responsibility for the patches.

- The cover letter and some commit messages were polished with an LLM
  after I wrote the initial drafts (to smooth the English). The technical
  content, the design (jointly with Barry), and the analysis are mine.

Please let me know whether an Assisted-by: LLM tag is expected for the
cover / commit-message polishing; I'm happy to add it in v2 if that's the
convention.

Bo