Re: Re-Transmission of blobs?

From: Junio C Hamano <gitster@pobox.com>
Date: Sep 11, 2013, 17:14 UTC
Message-ID: <xmqqsixbth4h.fsf@gitster.dls.corp.google.com>
In-Reply-To: <20130911112758.GB14259@raven.wolf.lan>

Josef Wolf <jw@raven.inka.de> writes:

Show 20 quoted lines

> On Di, Sep 10, 2013 at 10:51:02 -0700, Junio C Hamano wrote:
>> Consider this simple history with only a handful of commits (as
>> usual, time flows from left to right):
>> 
>>               E
>>              /   
>>     A---B---C---D
>> 
>> where D is at the tip of the sending side, E is at the tip of the
>> receiving side.  The exchange goes roughly like this:
>> 
>>     (receiving side): what do you have?
>> 
>>     (sending side): my tip is at D.
>> 
>>     (receiving side): D?  I've never heard of it --- please give it
>>                       to me.  I have E.
>
> At this point, why would the receiving side not tell all the heads it knows
> about?

It did. The receiving end had only one branch whose tip is E. It may have a tracking branch that knows where the tip of the sending end used to be when it forked (which is C), so the above may say "I have E and C". It actually would say "I have B and A and ..." for a bounded number of commits, but that does not fundamentally change the picture---the important point is it is bounded and there is a horizon.

>> There are some work being done to optimize this further using
>> various techniques, but they are not ready yet.

And this still stands.