how about a concerted effort to remove inappropriate EXPERIMENTAL
dependencies from the Kconfig files? there must be a truckload of
stuff that has long since matured beyond EXPERIMENTAL status that is
still tagged that way.
rday
--
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry:
Have classroom, will lecture.
http://crashcourse.ca Waterloo, Ontario, CANADA
========================================================================
On Thu, 24 Jul 2008 11:14:22 -0400 (EDT), "Robert P. J. Day" <[email protected]> wrote:
>
> how about a concerted effort to remove inappropriate EXPERIMENTAL
>dependencies from the Kconfig files? there must be a truckload of
>stuff that has long since matured beyond EXPERIMENTAL status that is
>still tagged that way.
I sent a patch to remove EXPERIMENTAL on a three year old driver couple
months ago, it got Acked by maintainer, hwmon/adm9240. But I dunno if
the patch is queued somewhere or lost.
Grant.
On Fri, 25 Jul 2008, Grant Coady wrote:
> On Thu, 24 Jul 2008 11:14:22 -0400 (EDT), "Robert P. J. Day" <[email protected]> wrote:
> > how about a concerted effort to remove inappropriate EXPERIMENTAL
> >dependencies from the Kconfig files? there must be a truckload of
> >stuff that has long since matured beyond EXPERIMENTAL status that is
> >still tagged that way.
>
> I sent a patch to remove EXPERIMENTAL on a three year old driver
> couple months ago, it got Acked by maintainer, hwmon/adm9240. But I
> dunno if the patch is queued somewhere or lost.
at last count, there were 437 config selections that were tagged as
EXPERIMENTAL. i find it hard to believe that there's that much
legitimately experimental content in the source tree. surely *some*
of that is suitably mature such that that dependency can be dropped.
rday
--
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry:
Have classroom, will lecture.
http://crashcourse.ca Waterloo, Ontario, CANADA
========================================================================
I think INFINIBAND_IPOIB_CM is no long EXPERIMENTAL. I'll remove the tag.
On Thu, 24 Jul 2008, Roland Dreier wrote:
> I think INFINIBAND_IPOIB_CM is no long EXPERIMENTAL. I'll remove the tag.
cool. i think we need a new attribution tag here. i'm thinking
Nagged-incessantly-until-we-fixed-it-by: Robert P. J. Day :-)
rday
--
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry:
Have classroom, will lecture.
http://crashcourse.ca Waterloo, Ontario, CANADA
========================================================================
> legitimately experimental content in the source tree. surely *some*
> of that is suitably mature such that that dependency can be dropped.
What is the connection between age of code and experimental status ?
Alan Cox wrote:
>> legitimately experimental content in the source tree. surely *some*
>> of that is suitably mature such that that dependency can be dropped.
>
> What is the connection between age of code and experimental status ?
Is it reasonable for in-tree code to remain "experimental" indefinately?
Chris
On Fri, 25 Jul 2008 01:15:09 -0400
"Chris Friesen" <[email protected]> wrote:
> Alan Cox wrote:
> >> legitimately experimental content in the source tree. surely *some*
> >> of that is suitably mature such that that dependency can be dropped.
> >
> > What is the connection between age of code and experimental status ?
>
> Is it reasonable for in-tree code to remain "experimental" indefinately?
I would certainly think so. We have stuff which is very long term
experimental (often because the vendor of the hardware won't provide
docs) but it is better to have this than nothing.
On Fri, 2008-07-25 at 01:15 -0400, Chris Friesen wrote:
> Alan Cox wrote:
> >> legitimately experimental content in the source tree. surely *some*
> >> of that is suitably mature such that that dependency can be dropped.
> >
> > What is the connection between age of code and experimental status ?
>
> Is it reasonable for in-tree code to remain "experimental" indefinately?
Or until death. E.g. "devfs" went directly from "experimental" to
"obsolete".
Bernd
--
Firmix Software GmbH http://www.firmix.at/
mobil: +43 664 4416156 fax: +43 1 7890849-55
Embedded Linux Development and Services
On Thu, 24 Jul 2008, Alan Cox wrote:
> > legitimately experimental content in the source tree. surely
> > *some* of that is suitably mature such that that dependency can be
> > dropped.
>
> What is the connection between age of code and experimental status ?
perhaps nothing, perhaps lots. *of course* there can be features
that are both long-lived and still experimental. just as, of course,
there is stuff that was (properly) tagged as EXPERIMENTAL ages ago
that has long since stabilized and can afford to stop being
experimental.
just a small nagging for folks to take a quick look at what no
longer needs to be tagged that way, that's all.
rday
--
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry:
Have classroom, will lecture.
http://crashcourse.ca Waterloo, Ontario, CANADA
========================================================================
Robert P. J. Day wrote:
> at last count, there were 437 config selections that were tagged as
> EXPERIMENTAL.
drivers/ieee1394/Kconfig and drivers/firewire/Kconfig contain three
dependencies on EXPERIMENTAL which are up to date.
So now you have got only 434 dependencies left to look at. :-)
--
Stefan Richter
-=====-==--- -=== ==--=
http://arcgraph.de/sr/