Re: [Gnu-arch-users] Re: cscvs--experimental--1.1 nearing doneness; call for testers

[Tupshin Harper]

Tom Lord wrote:

>    > From: Tupshin Harper <address@hidden>
>
>    > Currently, the 2.5 patches are somewhere north of 1.7G and 100,000 patches.
>
> I think there's some confusion here.
>
> If Linus is processing 100,000 individual commits in that time frame,
> then he is a shell script or worse.
>
> If in the tree of commits that is implied by Linus' merges there are
> 100,000 patches, that's a different thing.
>
> I don't know the exact timeframe of 2.5 but let's envelope-guess 1yr. 
> 100,000 patches in a year is 274 a day.   Linus is a family man.   So
> for an 8-hour day that'd be 34 an hour (no lunch break, no bathroom
> break, no coffee break, no vacation, no weekends off -- that's 7-days
> a week, 8-solid-hours a day).
>
> So, what's the real scoop.
>
> -t
>
>  
You are correct...Linus is not a shell script. If he pulls from somebody 
elses BK tree, he then commits (atomically) the collection of changesets 
that represent the changes that he's incorporating.

So if Joe Schmo starts with (for example) Linus' kernel  2.4.19, writes 
a new scheduler (for the 9 billionth time) and iteratively improves it 
over the course of a couple months until Linus decides it's ready to go 
into his tree, then all of the changesets that Mr. Schmo created along 
the way are committed to Linus' tree. If Joe incorporated changes from 
other people, than it gets even more complex. There is, effectively, a 
directed , single rooted, but multi-parented graph of changesets 
incorporating every bit of history of the changes that Linus ultimately 
commits.

The CVS (implicitly)and SVN (explicitly) versions of the bitkeeper trees 
have a much different level of granularity. They only have 1 
patch/diff/changeset/whathaveyou per Linus' commit. This is obviously a 
much simpler structure to import, but it is necessarily very lossy.

-Tupshin