Utsav Shah [off-list ref] writes:

[jc: we avoid top-posting for readability, so swapped paragraphs in
this quote]

quoted

Perhaps we can take a polished version of this "'reset --hard' can
and should notice paths known-to-be-uptodate via fsmonitor" as an
independent patch (to reduce the number of things we have to worry
by one) for now?  Taking this patch means we would now have one more
place that checks both ce_uptodate() and FSMONITOR_VALID bit, but if
we would be auditing all such places before we can decide what the
best way to reach the goal of allowing them to just say ce_uptodate()
without having to spell FSMONITOR_VALID, that probably is a cost
worth paying.

Thanks for working on this topic.

Thanks, that makes a lot of sense. ce_uptodate doesn't have too many
callers either, so modifying it and checking CE_FSMONITOR_VALID there
should not be hard to audit.

OK, so let's see an updated and hopefully final version of [*1*],
perhaps with Kevin's help you mentioned in [*2*] for now?


[References]

*1* [ref]
*2* [ref]