CVS IDs in our source [was Re: VM idle page zeroing]

[Sascha Wildner <saw@xxxxxxxxx>]

TR-2-Uf3nwNkHVr@mail.gmail.com> <201005170020.o4H0KnxB025376@apollo.backplane.com> <alpine.DEB.2.00.1005171051220.32223@endeavour> <4BF35D56.8030107@199technologies.org> <4bf36c25$0$848$415eb37d@crater_reader.dragonflybsd.org> <4BF37721.4080403@199technologies.org>
In-Reply-To: <4BF37721.4080403@199technologies.org>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Lines: 40
Message-ID: <4bf3b32f$0$844$415eb37d@crater_reader.dragonflybsd.org>
NNTP-Posting-Host: 80.187.108.224
X-Trace: 1274262320 crater_reader.dragonflybsd.org 844 80.187.108.224
Xref: crater_reader.dragonflybsd.org dragonfly.kernel:14371

Am 19.05.2010 07:29, schrieb Chris Turner:
> Sascha Wildner wrote:
>> Yes, we're gradually removing the old "$DragonFly: ... $" IDs. There's
>> not much they tell you except that this was the last rev. committed
>> using CVS. Why would you want to keep those?
>
> for the dragonfly ones this-kind-of-makes-sense,
> as the git repository tracks the cvs history as well.
> are these ident enabled?

How do you mean, enabled?

> for the other ones, I don't think it does -
> e.g. we took V.X of libfoo, etc.

Yes, I'm not aware of anyone removing any IDs other than the DragonFly 
ones. But if you have a specific example, I'd like to hear about it 
since we'd like to keep them.

With FreeBSD it's a bit of a problem. AFAIK they use a modified SVN 
client to maintain their CVS IDs after the switch to SVN but that 
modification isn't in pkgsrc so DragonFly people bringing in stuff from 
FreeBSD which was obtained with the pkgsrc client will have only 
"$FreeBSD$" there.

> ident string tracking is fun. if for no other purpose
> than historical kicks.
>
> it was very enlightening for me to meditate on
> the commit timeline of sunrpc, for example.
>
> I think the same could be found by others,
> were these strings in-base.

Please elaborate.

Sascha

-- 
http://yoyodyne.ath.cx