[PATCH v2 10/10] kernel: might_fault does not imply might_sleep

[Peter Zijlstra]

On Sun, May 19, 2013 at 07:40:09PM +0300, Michael S. Tsirkin wrote:
> OK I get it. So let me correct myself. The simple code
> that does something like this under a spinlock:
> >       preempt_disable
> >       pagefault_disable
> >       error = copy_to_user
> >       pagefault_enable
> >       preempt_enable
> >
> is not doing anything wrong and should not get a warning,
> as long as error is handled correctly later.
> Right?

Indeed, but I don't get the point of the preempt_{disable,enable}()
here. Why does it have to disable preemption explicitly here? I thought
all you wanted was to avoid the pagefault handler and make it do the
exception table thing; for that pagefault_disable() is sufficient.