Re: [RFC] Generation Number v2

[Jakub Narebski <jnareb@gmail.com>]

Jakub Narebski <jnareb@gmail.com> writes:
> Jakub Narebski <jnareb@gmail.com> writes:
>> Stefan Beller <sbeller@google.com> writes:
> [...]
>>> How would this impact creation of a commit?
>>>
>>> The current generation numbers can be lazily updated or not
>>> updated at all. In my understanding of the maximum generation
>>> numbers, a new commit would make these maximum generation
>>> numbers invalid (i.e. they have to be recomputed).
> [...]
>>> For the V2 maximum generation numbers, would we need to
>>> rewrite the numbers for all commits once we recompute them?
>>> Assuming that is true, it sounds like the benchmark doesn't
>>> cover the whole costs associated with V2, which is why the
>>> exceptional performance can be explained.
>>
>> Let's check it using a simple example
>>
>> First, (minimum) parent-based generation numbers before and after
>> extending the commit graph:
>>
>>   1   2     3   4     5   6   7    new
>>   1   2     3   4     5   -   -    old
>>   .---.-----.---.-----.---*---*
>>        \
>>         \   3   4     5   6        new
>>          \  3   4     5   6        old
>>           \-.---.-----.---.
>>                  \
>>                   \   5            new
>>                    \  -            old
>>                     \-*
>
> Let me check yet another idea, using (minimum) parent-based V0 generation
> numbers (counting distance from the sink / root) as a starting number
> for source / heads commits.
[...]

> [...] but let's check another example
>
>    1   2     3   4   5   6     7   8   9       new
>    1   2     3   4   5   6     7   8   -       old
>    .---.-----.---.---.---.-----.---.---*
>         \                     /
>          \   3   4           / 5   6   7   8   new
>           \  5   6          /  -   -   -   -   old
>            \-.---.---------/---*---*---*---*

But let's do this correctly.


   1   2     3   4      5   6     7   8   9      new
   1   2     3   4      5   6     7   8   -      old
   .---.-----.---.------.---.-----.---.---*
        \                        /
         \   3   4              /                new
          \  5   6             /                 old
           \-.---.------------/
                  \
                   \    5   6     7   8          new
                    \   -   -     -   -          old
                     \--*---*-----*---*

Well, it looks as if I draw it incorrectly, but performed calculations
right.  You may need to modify / change some data, but it looks as if it
is not that much of a problem.

The new version of the maximum generation numbers looks like it gives
the same results as generation numbers for the "longest" path, and
update may affect only the side-branches that were added to.  All
branches merged into the trunk, and not added to should be safe with
respect to updating.

Can anyone here prove a thing about update of those modified maximum
generation numbers?  Thanks in advance.

Best,
-- 
Jakub Narębski