Date:	Sun, 04 Dec 2011 20:50:52 +0100
From:	Philipp Schafft <lion@...n.leolix.org>
To:	Steven Whitehouse <swhiteho@...hat.com>
Cc:	mike.gair@...asteel.com, Chrissie Caulfield <ccaulfie@...hat.com>,
	Christoph Lameter <cl@...ux-foundation.org>,
	David Miller <davem@...emloft.net>,
	Eric Dumazet <eric.dumazet@...il.com>,
	Sasha Levin <levinsasha928@...il.com>,
	Linux-DECnet user <linux-decnet-user@...ts.sourceforge.net>,
	linux-kernel <linux-kernel@...r.kernel.org>,
	linux-mm <linux-mm@...ck.org>, Matt Mackall <mpm@...enic.com>,
	netdev <netdev@...r.kernel.org>,
	Pekka Enberg <penberg@...nel.org>,
	RoarAudio <roaraudio@...ts.keep-cool.org>
Subject: Re: [Linux-decnet-user] Proposed removal of DECnet support
	(was:Re: [BUG] 3.2-rc2:BUG kmalloc-8: Redzone overwritten)

reflum,

On Wed, 2011-11-30 at 14:52 +0000, Steven Whitehouse wrote:
> On Wed, 2011-11-30 at 13:52 +0000, mike.gair@...asteel.com wrote:
> > In theory i'd be interested in maintaining it,
> > but i'm not sure what amount of work is involved,
> > have no experience of kernel, or where to start.
> > 
> > Any ideas?
> > 
> > 
> So the issue is basically that due to there being nobody currently
> maintaining the DECnet stack, it puts a burden on the core network
> maintainers when they make cross-protocol changes, as they have to
> figure out what impact the changes are likely to have on the DECnet
> stack. So its an extra barrier to making cross-protocol code changes.
> 
> If there was an active maintainer who could be a source of knowledge
> (and the odd patch to help out making those changes) then this issue
> would largely go away.

*nods*


> The most important duty of the maintainer is just to watch whats going
> on in the core networking development and to contribute the DECnet part
> of that. So it would be most likely be more a reviewing of patches and
> providing advice role, than one of writing patches (though it could be
> that too) and ensuring that the code continues to function correctly by
> testing it from time to time.
> 
> The ideal maintainer would have an in-depth knowledge of the core Linux
> networking stack (socket layer, dst and neigh code), the DECnet specs
> and have a good knowledge of C. 

I guess I would fit mostly but I have no idea of the kernel internal
stuff. Also I'm a bit short on time.


> Bearing in mind the low patch volume (almost zero, except for core
> stuff), it would probably be one of the subsystems with the least amount
> of work to do in maintaining it. So in some ways, a good intro for a new
> maintainer.

Jup. This is very true. I hope we will find a new maintainer because of
exactly this point. Maybe somebody like Mike Gair.


> I do try and keep an eye on what get submitted to the DECnet code and
> I'll continue to do that while it is still in the kernel. However, it is
> now quite a long time since I last did any substantial work in the
> networking area and things have moved on a fair bit in the mean time. I
> don't have a lot of time to review DECnet patches these days and no way
> to actually test any contributions against a real DECnet implementation.

I'm glad you are still interested. I'm always happy when I see mails
from you at the DECnet for Linux list.


> So I'll provide what help I can to anybody who wants to take the role
> on, within those limitations. I'm also happy to answer questions about
> why things were done in a particular way, for example.
> 
> It is good to know that people are still using the Linux DECnet code
> too. It has lived far beyond the time when I'd envisioned it still being
> useful :-)

There are still some people interested in it. Btw. on Debian popcon
counts 5356 users.

-- 
Philipp.
 (Rah of PH2)

Download attachment "signature.asc" of type "application/pgp-signature" (483 bytes)

Powered by blists - more mailing lists