Re: git-diff-tree -M performance regression in 'next'

Fredrik Kuivinen [off-list ref] writes:

I turned out that the rename detection took almost 10 minutes on my
machine.

Yes, that is one of the reasons why it still is in "next", not
in "master".

The rename-detector change was done primarily to work around the
correctness problem the finer-grained delta changes would have
introduced.  The new delta code would have produced far more
copies from the source than the current xdelta code, but the
nature of the new copies it would have found was quite different
from what we would usually call "file being renamed".  Now we
decided to shelve the finer-grained delta code for now, I do not
see a pressing reason to have the experimental rename detector
graduate to "master" until we resolve its performance issues.