Announcing OpenWrt/MLPPP - multilink firmware for consumer routers - Caneris & Acanac

[Christopher Browne]

On Mon, Mar 1, 2010 at 1:13 PM, Lennart Sorensen
<lsorense-1wCw9BSqJbv44Nm34jS7GywD8/FfD2ys at public.gmane.org> wrote:
> On Sun, Feb 28, 2010 at 08:55:53AM -0500, Erik L wrote:
>> Of course, but again, we need something now! Unless we end up doing a proprietary driver (let me guess, I'll be shot for saying that here), there's no reason not to upstream, except it takes too long for the code to go into the Linux tree, then trickle down to OpenWrt, etc.
>
> Especially when people don't even start the process.  It makes things
> so much easier in the medium to long term that it is very much worth
> the effort.  Too many people only look at the short term.

Indeed.

I'd suggest the thought that "due diligence" should make it obligatory
to at least *try* to pass it upstream.

- You'll need to track the difference in your own downstream
repository, whether it actually gets upstream or not.

- The difference may go upstream, get applied, with some changes, and
come back downstream a year later.  In that case, the downstream
repository will have some effort involved in merging it in.

Even if the change never comes back downstream, having the
customization captured and tracked means that you know the scope of
the driver change, and have a fighting chance when it comes time to
try to upgrade to a newer kernel.

We've had this sort of thing happen at work vis-a-vis customizations
to PostgreSQL...  There have been occasions (happily, disappearing as
certain elderly versions disappear!) where we had local mods that
differed from upstream.  We had our own copy of CVS, to track this,
and to support the need to repatch, on occasion.

There's no difference between database "kernel" and OS "kernel" in
this kind of scenario.
-- 
http://linuxfinances.info/info/linuxdistributions.html
Pablo Picasso  - "Computers are useless. They can only give you
answers." - http://www.brainyquote.com/quotes/authors/p/pablo_picasso.html
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists