Re: [PATCH -tip] Switch arm defconfig files from CONFIG_CLASSIC_RCU to CONFIG_TREE_RCU.

From: Sam Ravnborg
Date: Fri Dec 26 2008 - 13:33:56 EST


On Thu, Dec 25, 2008 at 11:23:32AM -0800, Paul E. McKenney wrote:
> On Thu, Dec 25, 2008 at 10:33:01AM +0100, Sam Ravnborg wrote:
> > On Thu, Dec 25, 2008 at 09:36:11AM +0100, Ingo Molnar wrote:
> > >
> > > below is the fix for another small buglet that hits architectures that do
> > > not include kernel/Kconfig.preempt.
> >
> > What is needed to let all architectures include Kconfig.preempt?
> > We should not advertise preemption if not supported,
> > but we could do this by a simple Kconfig symbol:
> >
> > config HAVE_PREEMPT
> > bool
> >
> > And let all architectures that supports preemption do:
> >
> > config "ARCH"
> > select HAVE_PREEMPT
> >
> > But maybe there is a simpler solution
>
> The idea being to make things like PREEMPT then depend on HAVE_PREEMPT?
Yes - exactly.

So an arch can say that is includes support for preemption
using the following Kconfig snippet:

config X86
select HAVE_PREEMPT

And then in the spot where we have a prompt we do:

config PREEMPT
bool "Do you want preemption?"
depends on HAVE_PREEMPT


Thne we will only present this option to the user
for the architectures where it has indicated preemption
support by selecting HAVE_PREEMPT.

It is on purpose that HAVE_PREEMPT is a single boolean
with no prompt or anything so we do not hit
any of the pitfalls of using select.

We use a similar pattern in many places today.

Sam
--
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/