Re: [PATCH] "git push": tellme-more protocol extension

Jeff King [off-list ref] writes:

On Thu, Mar 20, 2008 at 09:24:18PM -0700, Junio C Hamano wrote:

quoted

So a potential improvement might be to ask for "tellme-more" only if we do
not know any of the refs the receiver showed us; my patch does it if there
is at least one unknown instead.

You can still get cases where doing this is more expensive than your
patch, because A..B is very small but shared_tag..A is very large. I'm
not sure which is more likely in practice.

Oh, I have a few fix-ups since I sent the patch, but I am not planning to
experiment with _your_ protocol myself.  Debugging one protocol extension
a week is painful enough ;-)

And tellme-more does not send _every_ commit.  The sender asks only about
the refs it does not know about, and the receiver sends densely near the
tip and then walks sparsely as it goes deeper, so the traffic wouldn't be
so bad.  Latency might be a different story, though.