[PATCH] D99305: [docs] Document our norms around reverts

[Hubert Tong via Phabricator via llvm-commits]

hubert.reinterpretcast added inline comments.


================
Comment at: llvm/docs/DeveloperPolicy.rst:317-318
+* If you break a buildbot in a way which can't be quickly fixed, please revert.
+* If a test case is reported in the commit thread which demonstrates a problem
+  please revert, and investigate offline.
+* If you receive substantial :ref:`post-commit review <post_commit_review>`
----------------
reames wrote:
> hubert.reinterpretcast wrote:
> > hubert.reinterpretcast wrote:
> > > How portable and self-contained is this test case?
> > 
> I can't think of anyway to craft a useful guideline here (beyond what is already said elsewhere).  If you have suggested wording, please share.
> 
> (Though, do you mind if we defer this to a future change?  This seems minor compared to the rest of it.)
I think moving the "expectations" part up would help towards addressing my concern here.


================
Comment at: llvm/docs/DeveloperPolicy.rst:320
+* If you receive substantial :ref:`post-commit review <post_commit_review>`
+  feedback, please revert and address said feedback before recommitting.
+  (Possibly after another round of review.)
----------------
reames wrote:
> hubert.reinterpretcast wrote:
> > ... doesn't the "further destabilize" note below actually apply here too?
> Judgement is always required.  I don't think it's useful to repeat that in every guideline.  Do you think I need to emphasize that more somewhere?
For this too, I think moving the "expectations" part up will clarify the "priorities".


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D99305/new/

https://reviews.llvm.org/D99305