[RESENT PATCH] Don't disable irqs in set_next_event and set_mode callbacks

[Thomas Gleixner]

On Fri, 27 Nov 2009, Russell King - ARM Linux wrote:
> On Fri, Nov 27, 2009 at 09:38:23PM +0100, Uwe Kleine-König wrote:
> > OK, so this should make it save for mach-nomadik/timer.c,
> > mach-pxa/time.c and mach-sa1100/time.c.
> > 
> > Only mach-at91/at91rm9200_time.c and mach-at91/at91sam926x_time.c
> > request with IRQF_SHARED | IRQF_DISABLED.
> > So for there the right solution is to let the local_irq_save in place
> > and add a comment that it is needed as the timer irq is shared.
> > This hopefully stopps people copying it unnecessarily.
> 
> Given that the gentime code spits out warnings if its called with IRQs
> enabled, I don't think keeping the local_irq_save() in place is worth
> it - especially given Thomas' response.
> 
> We've seen such warnings on AT91 (though I don't think we got positively
> to the bottom of it with the reporter.)  If you _do_ want to ensure that

Any pointer ?

> AT91 is safe, wrapping the call to dev->event_handler(dev) in the AT91
> interrupt handler with a local_irq_save/local_irq_restore would seem
> sensible.

The only way how that would happen is when the serial interrupt would
be set up _BEFORE_ the timer interrupt and when the serial handler
would enable interrupts at some point. Shared interrupt handlers are
called in order of setup.

Thanks,

	tglx