[llvm] [llvm][docs] Add notes on upstreaming code from downstream projects (PR #129743)

[Andrzej Warzyński via llvm-commits]

================
@@ -272,3 +272,19 @@ Developers should participate in code reviews as both reviewers and
 authors. If someone is kind enough to review your code, you should return the
 favor for someone else. Note that anyone is welcome to review and give feedback
 on a patch, but approval of patches should be consistent with the policy above.
+
+Upstreaming Changes to LLVM
+===========================
+
+When upstreaming your own changes from a downstream project to LLVM, simply
+follow the process outlined above.
+
+When upstreaming changes originally written by someone else:  
+
+* Ensure that the original author(s) are aware of and approve the upstreaming
+  of their code, and that they are comfortable with you leading the process.
+* Properly attribute the original changes, e.g., by creating a commit with
+  multiple authors ([GitHub guide](https://docs.github.com/en/pull-requests/committing-changes-to-your-project/creating-and-editing-commits/creating-a-commit-with-multiple-authors)).  
+* Invite the original author(s) to review the changes, and also include
+  additional reviewers. Specifically, an LGTM from a (co-)author does not
+  constitute approval to land a change.
----------------
banach-space wrote:

> For example, the person could have been involved in the downstream development, but perhaps only for a limited part of the whole patch.

Sure, but that still makes them a co-author, right? I believe we should always list all co-authors and avoid introducing overly refined rules in LLVM.  

For instance, a rule like:  
> "Only add a person as a co-author if they contributed > N lines of code"  

would be arbitrary and hard to enforce. I also wouldn’t know how to define a reasonable threshold. I'd prefer to avoid such distinctions altogether.

> If the community relies upon or generally defers to said person for the approval of patches in the area being modified, it stands to reason that, absent comments from other reviewers within a reasonable wait period, their approval should be considered valid for landing the change (if they indicate that to be their intent).

Agreed—this does happen occasionally, and as you mentioned, transparency about intent is key. My suggestion primarily applies to cases where, for example, a patch has two co-authors, but no attempt is made to involve a third reviewer. In such cases, I’d expect co-authors to proactively seek additional reviewers (within reason).  

By the way, if there's an area in LLVM that relies solely on one person to review all patches, that’s a separate issue worth discussing elsewhere. Even in those cases, I’d still expect a clear expression of intent from the co-author before landing a change.


https://github.com/llvm/llvm-project/pull/129743