Junio C Hamano [off-list ref] writes:

Jeff King [off-list ref] writes:

quoted

I was not following the topic closely, but I think there is a v10 of
this series anyway, so you may want to eject it from 'next' in the
meantime (I had assumed you were going to do so as part of the
post-release rewind, but it looks like that happened already and it got
re-merged).

I can do the post-release rewind again ;-)

I somehow had an impression that the topic was more or less done,
but if it deserves another chance, let's give it one.

Thanks.

I have been on vacation for the last two weeks, sorry for not
replying in time.

I think that is :

  time-1: v9 was in next, actually v9 [4/6] will break the CI, but
          [5/6] fix it so the CI passed on the whole patchset.
  time-2: then v10 was tought a new `--no-separator ` mainly
          On the release point, v10 implicitly fixed the problem
          in v9 [4/6]
  time-3: 2.41 release work begin
  time-4: Peff found that the v9 [4/6] merged in next would break
          CI and a UBSan problem.
  time-5: I wanted to make a full reroll to fix the UBSan problem,
          by the way, separate the "--no-separator" to a single
          commit, because I thought it cannot catch the last bus
          on 2.41 maybe.

I think we are now in a new cycle, maybe we could rewind and use
v11 directlly? Sorry that it may have brought on extra work.

Thanks.