On Thu, Sep 16, 2021 at 06:45:33PM -0400, Jeff King wrote:
It might be interesting to also do:
rm -f .git/objects/pack/pack-*.bitmap
after generating the midx bitmap. That would confirm the further timing
tests are using the midx bitmap, and not ever "cheating" by looking at
the pack one (having poked in this direction before, I know that this
all works, so it would be a future-proofing thing).
This and the suggestion to avoid timing the single pack bitmap
generation are both good ones.
I think to be totally accurate we would want to drop the pack bitmap
before every MIDX bitmap generation, but I also think that in practice
it does not matter much.
The reuse code currently does not try too hard to recognize the
situation of "oh, I selected the same exact commits and don't have to do
any work". It kind of does eventually, but only after doing a lot of
preparation.
So I'm dubious as to whether we're really timing anything *that* useful,
but it's probably worth keeping around anyway.
Thanks,
Taylor