AaronBallman wrote: > I gave up again on this review, as I think it is over complicated and > over-distributing the logic.
I am not nearly as well-versed in CIR as you are, but the amount of back-and-forth here caught my attention because this isn't the first time I've seen some frustrations along these lines. @adams381 it looks to me like you are using an LLM as part of this work; our community policies require you to disclose use of AI (https://llvm.org/docs/AIToolPolicy.html) but many of the PRs are missing this disclosure, not just this one. While our policy does allow AI tool use, please keep in mind that long exchanges are really taxing on reviewers. Given the quantity of reviews this seems to have happened on in a fairly short period of time, I'd like to avoid concerns about this being extractive behavior (https://llvm.org/docs/AIToolPolicy.html#extractive-contributions). My read of several of the PRs is that you may be feeding reviewer feedback to an LLM and reporting the tools' output and that seems to be contributing to these long exchanges. I don't know that this is what's happening, but if that sounds like a reasonable assessment, if you're involved in longer exchanges in the future I'd recommend spending more time understanding the code and comments before posting changes or responses. Also, it looks like you're dropping commits when you push up new changes (rebasing, probably), that makes it much harder for reviewers to see how a review is progressing over time. We're in the progress of updating the developer policy to make this more clear, but please don't drop commits once they've been pushed to the public. https://github.com/llvm/llvm-project/pull/205918 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
