Re: How linux-staging is released (was Re:linux-staging/linux-next merge/build errors)

[Ingo Molnar]

* Greg KH <greg@xxxxxxxxx> wrote:

> On Sat, Jun 21, 2008 at 08:02:22PM +0200, Peter Zijlstra wrote:
> > The point was that for some people being carried by distos is upstream
> > enough.
> > 
> > So I would discourage distros form picking up this tree.
> 
> I wouldn't, as some of the drivers work just fine on some arches, but 
> not all, that is why they are not in mainline yet (the network drivers 
> in there right now are an example of this.)

.. and it's all source code, so it's not really different to distros 
than a number of other patches they regularly pick up. If there's too 
much mess there's always the option to look into the fine source code 
and fix things up.

If a driver ended up not mattering at all in the long run then nobody 
will ever fix it up - and we dont even care that much - it's almost as 
if it didnt even exist.

The other alternative is much more harmful: vendors release binary-only 
drivers. That only works on x86 and they are utterly unfixable as well 
and have a lot of other disadvantages.

So we really have to balance the natural desire for pure upstream 
drivers against the real dangers of vendors doing bin-only modules just 
for the convenience of distribution.

... all in one: kudos Greg for doing the linux-staging tree! :-) It will 
help a lot.

	Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/