[patch] turn on might_sleep() in early bootup code too

[Zwane Mwaikambo]

On Wed, 18 Jan 2006, Ingo Molnar wrote:

> lock_cpu_hotplug() has design problems i think: hotplug-locked sections 
> are slowly spreading in the kernel, encompassing more and more code :-) 
> Shouldnt the CPU hotplug lock be a spinlock to begin with?

The way it's used certainly is bizarre, but a spinlock would be harder to 
work with as a lot of the code protected by it sleep.