Re: [PATCH] protect 'action' in show_interrupts

[Andrew Morton]

Zwane Mwaikambo <[e-mail]> wrote:
>
> On Fri, 7 Mar 2003, Russell King wrote:
> 
> > We don't have a per-irq_desc spinlock on ARM - it's a global
> > irq_controller_lock.
> > 
> > Thanks.
> 
> Ok thanks for spotting that. Andrew here it is rediffed.

Thanks, Zwane.  Looks good.

I made some adjustments down in mips land - they weren't initialising the
irq_desc_t locks on startup.

That's pretty academic, but doing spin_lock_irqsave() is as good a way as any
of saying "preempt_disable()", and it keeps all the code looking the same.