I've seen this breakage only on "regular (osx-clang, clang,
macos-latest)" so far, but that does not necessarily mean clang on
macOS is miscompiling or anything like that.
https://github.com/git/git/runs/1671504998?check_suite_focus=true
What is frustrating is that the same CI/PR #1087 fully passes when
told to "Re-run jobs".
https://github.com/git/git/runs/1671626677?check_suite_focus=true
They are testing exactly the same commit 820e568 that has a tree
identical to that of ed0fd57b97 in seen in today's pushout.
Any interest in helping to find out what's going on from folks in
the macOS land?
Thanks.