CVS issues [was: 1.1 development release targets.]

Jeremy Jackson jerj at coplanar.net
Wed Apr 16 15:45:00 CEST 2003


----- Original Message -----
From: "Eric W. Biederman" <ebiederman at lnxi.com>
To: "Jeremy Jackson" <jerj at coplanar.net>
Cc: "LinuxBIOS" <linuxbios at clustermatic.org>
Sent: Tuesday, April 15, 2003 2:04 PM
Subject: Re: 1.1 development release targets.


> "Jeremy Jackson" <jerj at coplanar.net> writes:
>
> > > To implement this I have started a freebios2 tree at sourceforge
> > > and I will gradually populating this tree.  I am taking the expedient
> > > of copying old code over as needed.  Leaving old code for now
> > > unneeded functionality to just fade away.  It is the only easy
> >
> > Have you considered using the branch facility of CVS?
>
> Thought about it.
>
> I don't much care about the CVS history.  And where there are
> several trees it is less then correct a number of times.  I am about
> ready to rip out all of the $Id$ tags from the tree they are a consistent
> source of noise when you have two different repositories and you are
> comparing the differences.

There is the -kk keyword substitution mode.  But you have to rid the
repository
(at least HEAD version) of any -kb (binary) files first.
>
> In addition.  By pulling in source modules in on demand I have a real
> nice way of filtering out old dead code and showing which code works
> in the new development tree.  With my radical changes to hardwaremain
> everything breaks until it gets fixed up.

There are Exp/Stable version labels, which nobody uses, or (never done this
though) you could check in an empty branch, then pull the files in one at a
time.

> So I guess I don't see an advantage in using the cvs branch feature,
> at this point.

Probably not.  CVS doesn't handle chaos very well (moving files, renaming
dirs, etc).  But if Ron it to start creating releases at some point, CVS is
definitely your friend.  With production code, being able to branch an old
release and merge *just* the fixes to a bug from the HEAD branch, then tag
it and push out a build is smooth, like butta, it's something I've used on a
production system and it just works.

So not using CVS to branch probably isn't hurting anything at this point,
but I think it's a good idea to start thinking about the issues.  I think
getting rid of binary files in the repository is pretty important.  It's ok
for .jpg files on a website project, where nobody uses branches (or as most
web monkeys operate even release tags), but it's not good for serious
projects.

Regards,

Jeremy




More information about the coreboot mailing list