source control?

source control?

[Grover, Andrew]

Is Linux development ever going to use source control?

This was talked about at the Kernel Summit, and I haven't heard anything
about it since.

Regards -- Andy

Re: source control?

[Nicholas Knight]

On Thursday 23 August 2001 02:29 pm, Grover, Andrew wrote:
> Is Linux development ever going to use source control?
>
> This was talked about at the Kernel Summit, and I haven't heard
> anything about it since.

The kernel has source control, its name is Linus Torvalds, CVS with a 
brain.
Wether or not the mainstream kernel will ever go pure CVS or the like is 
really up to Linus, and so far I've not seen much indication that he's 
going to do it, at least not before he decides to retire from steering 
the kernel's development.

>
> Regards -- Andy

Re: source control?

[Larry McVoy]

On Thu, Aug 23, 2001 at 02:29:18PM -0700, Grover, Andrew wrote:
> Is Linux development ever going to use source control?
> 
> This was talked about at the Kernel Summit, and I haven't heard anything
> about it since.

There are two features that Linus wants in BitKeeper that we haven't
finished yet.  One is the removal of the revision control files from
the working tree and the other is the ability to break the tree up
into logical units (we call 'em filesets) so that you can more easily
pick and choose which patches you want in your tree.

Linus pinged me about these a while back, we've made some progress on
them but they aren't done yet.  When they are done we'll let Linus take
it for a whirl and see what he thinks.

In the meantime, the PPC people maintain a pure Linux tree in BK, you
can see it at http://ppc.bkbits.net and Ted Tso has recently done a
nice import of the various kernel versions complete with Linus' change
logs.  I need to work with the PPC guys and Ted to get to one tree;
it's not an easy issue because the PPC have a lot of changes in their
tree but Ted's tree was done more nicely, he did some extra work to 
preserve timestamps and comments.  

And to avoid yet-another-BK-flamewar, I'm not saying Linus will or will
not use BitKeeper, all I'm saying is that we're making changes he wants
and then he'll see if it is good enough for him.  I will say that he has
eased slightly off of his original position of "I'll use BitKeeper when
it is the best" because I asked him if that meant what I think both he
and I would mean, i.e., "it is not physically possible for it to be better"
as opposed to "it's better than all the other crap out there".  I think
we agreed we have to be well past #2 but not necessarily to #1 (which is
a good thing, at the rate we're going we'll hit the best sometime this
century but that's as close as I want to call it :-)
-- 
---
Larry McVoy            	 lm at bitmover.com           http://www.bitmover.com/lm 

Re: source control?

[Alan Cox]

> Is Linux development ever going to use source control?

It does. Or at least many of the development teams do. That doesn't mean a
general CVS is a good idea. CVS make it all to easy for other people to 
push crap into your tree. 

Alan

Re: source control?

[Cort Dougan]

That's a great description of all source control!  "Makes it all to easy
for other people to push crap into your tree"!

There is great benefit to making it very hard for people to get changes
into a tree.  It forces people to ask "Is this really worth all the
effort?" several times.  It's a great filter.

} It does. Or at least many of the development teams do. That doesn't mean a
} general CVS is a good idea. CVS make it all to easy for other people to 
} push crap into your tree. 

Re: source control?

[Gérard Roudier]

On Fri, 24 Aug 2001, Alan Cox wrote:

> > Is Linux development ever going to use source control?
>
> It does.

Pardon, I missed it. :)

> Or at least many of the development teams do.

Multiple source repositories means no-source-control, IMO.

> That doesn't mean a
> general CVS is a good idea.

So this means that most of programmers are probably idiots, given the
number of projects that use a _single_ source controlled repository. :)

> CVS make it all to easy for other people to
> push crap into your tree.

What other people?
You can only allow trusted people to commit, and backing out crap is quite
easy.

The only risk, in my opinion, of using a source-control system is that it
allows easy forking, that is often a bad idea for the long run. On all
other points, it is better, IMO, to use a source-control system with a
unique repository, than penguinish :) manual patching.

  Gérard.

Re: source control?

[Alan Cox]

> What other people?
> You can only allow trusted people to commit, and backing out crap is qu=
> ite
> easy.

This is the model we use. The trust people list is Linus Torvalds.

Alan

Re: source control?

[Samium Gromoff]

     when the questions comes to "good or handy"
  Linus chooses good.

     CVS is blind. Linus is not.

---


cheers,


   Samium Gromoff

Re: source control?

[Gérard Roudier]

On Fri, 24 Aug 2001, Alan Cox wrote:

> > What other people?
> > You can only allow trusted people to commit, and backing out crap is qu=
> > ite
> > easy.
>
> This is the model we use. The trust people list is Linus Torvalds.

You just pointed out the problem. Linus being the only trusted committer
for more than 100 MB of source base as he was for less than 1 MB 10 years
ago. And our single committer got some other loads as he has a job,
children, a boss, a mother-in-law :), may-be pets, etc...

The fact that Linux has great success does not mean that Linus is right on
the way he wants the kernel maintainance to proceed. It just means that he
hasn't been too wrong on this point, in my opinion. :-)

  Gérard.

Re: source control?

[Samium Gromoff]

     when the questions comes to "good or handy"
  Linus chooses good.

     CVS is blind. Linus is not.

---


cheers,


   Samium Gromoff

Re: source control?

[Florian Weimer]

Cort Dougan <cort@fsmlabs.com> writes:

> That's a great description of all source control!  "Makes it all to easy
> for other people to push crap into your tree"!

Try Aegis.  It enforces a develop/review/integrate cycle for each
change.

-- 
Florian Weimer 	                  Florian.Weimer@RUS.Uni-Stuttgart.DE
University of Stuttgart           http://cert.uni-stuttgart.de/
RUS-CERT                          +49-711-685-5973/fax +49-711-685-5898

Re: source control?

[Samium Gromoff]

> > That's a great description of all source control!  "Makes it all to easy
> > for other people to push crap into your tree"!

> Try Aegis.  It enforces a develop/review/integrate cycle for each
> change.
  and slows down the things...
  and hides (though not completely) the process from the people...

  one-thread-modify of some piece of code is inefficeient.

    When X code hacker splits his changes on small pieces
  and feeds them to Linus^WSourceControl, does he need
  to move each of his patches thru these
  develop/review/integrate cycles?

---


cheers,


   Samium Gromoff

Re: source control?

[Larry McVoy]

[Aegis == good discussion]
[Aegis == bad discussion]

Before people open up this can of worms again, which happens about once a year,
could you please go search the archives for the discussion of this in the past?
This always turns into a flame fest and it never resolves anything.
-- 
---
Larry McVoy            	 lm at bitmover.com           http://www.bitmover.com/lm 

Re: source control?

[Florian Weimer]

"Samium Gromoff" <_deepfire@mail.ru> writes:

> > > That's a great description of all source control!  "Makes it all to easy
> > > for other people to push crap into your tree"!
> 
> > Try Aegis.  It enforces a develop/review/integrate cycle for each
> > change.

>   and slows down the things...

Well, sometimes this increases productivity. ;-)

>   and hides (though not completely) the process from the people...

Well, if everyone has to monitor every other developer, there would be
problems, too.

>   one-thread-modify of some piece of code is inefficeient.

Of course, Aegis supports multiple changes which are being developed
at the same time.

>     When X code hacker splits his changes on small pieces
>   and feeds them to Linus^WSourceControl, does he need
>   to move each of his patches thru these
>   develop/review/integrate cycles?

IIRC, some time in the past, Linus said that he was the integrator and
the subsystem maintainers the reviewers.  I think the Linux
development process is actually pretty close to the Aegis model.
(However, the Aegis implementation seems to have some properties which
make it a bit unsuitable for controlling the Linux kernel
development.)

-- 
Florian Weimer 	                  Florian.Weimer@RUS.Uni-Stuttgart.DE
University of Stuttgart           http://cert.uni-stuttgart.de/
RUS-CERT                          +49-711-685-5973/fax +49-711-685-5898