[cgl_discussion] Fw: [announce] linux-2.5.51-dcl1

Mika Kukkonen mika at osdl.org
Tue Dec 10 15:21:08 PST 2002

On ti, 2002-12-10 at 14:16, Rusty Lynch wrote:
> Personally, I'm just trying to figure out if this is something I could find 
> useful.  It seems a little unclear to me how the tree is to be used.  What 
> is it's purpose?

It has many purposes, but one of them is to act as a kind of staging
area (or I guess sandbox) for kernel patches that we feel are mature
enough to be included into mainline, whether that is kernel.org or the

> Tim made it sound (at least the way I understood him) like the tree was 
> kind of a common sandbox for CGL related features.  That makes
> sense to me, but it leaves a lot of details to be defined... maybe it's
> just to early to ask specific questions.

No, this is "release early, release often" country, so go ahead.

> For example, if I wanted to collaborate with some other people over
> testing POSIX message queues, would it be acceptable to add
> the message queue patch of the day (or hour) to the cgl tree
> so I can always refer to test results on 'the latest cgl bk tree'
> type of thing.

You have it wrong there. This is Randy's kernel tree, and if you want
something to be added there, you try to convince Randy to add it. But
nothing stops you from deriving your own kernel tree from Randy's tree
(that's why we are using BitKeeper), and add your patch to that, and use
that anyway you like.

> If not then no big deal, I'll just do it another way.  I can imagine
> a few other ways where I could use such a tree, but I think
> I could be reading a lot into what this tree is for.

So ask questions/propose things, we (Randy and I) have nothing to hide
on this matter and are open to suggestions.


More information about the cgl_discussion mailing list