Re: [PATCH] check_updates(): effective removal of cache entries marked CE_REMOVE
Junio C Hamano <gitster@pobox.com> writes:
> I've queued this in 'next' as part of your earlier series, together with
> the remaining patches..
> I am also a bit surprised that it would make a noticeable difference in
> real life.
I guess that 77 milliseconds compared to 15 seconds is not noticable
for the Linux checkout test, but since this is an algorithmic
reduction from O(N^2) to O(N), I would guess that for big working
trees it can be more noticable.
-- kjetil