Re: [PATCH 5/8] arm64: irq: add a default handle_irq panic function

[Marc Zyngier]

On 2021-02-22 09:59, Mark Rutland wrote:
> On Fri, Feb 19, 2021 at 11:39:01AM +0000, Mark Rutland wrote:
> > If we accidentally unmask IRQs before we've registered an IRQ
> > controller, handle_arch_irq will be NULL, and the IRQ exception 
> > handler
> > will branch to a bogus address.
> >
> > To make this easier to debug, this patch initialises handle_arch_irq 
> > to
> > a default handler which will panic(), making such problems easier to
> > debug. When we add support for FIQ handlers, we can follow the same
> > approach.
>
> > -void (*handle_arch_irq)(struct pt_regs *) __ro_after_init;
> > +void default_handle_irq(struct pt_regs *regs)
> > +{
> > +	panic("IRQ taken without a registered IRQ controller\n");
> > +}
>
> The kbuild test robot pointed out that this should be static (likewise
> with default_handle_fiq in patch 8), since it's only used within this
> file, so I've updated that in my branch.
>
> Mark.
>
> > +
> > +void (*handle_arch_irq)(struct pt_regs *) __ro_after_init = 
> > default_handle_irq;
> >
> >  int __init set_handle_irq(void (*handle_irq)(struct pt_regs *))
> >  {
> > -	if (handle_arch_irq)
> > +	if (handle_arch_irq != default_handle_irq)
> >  		return -EBUSY;
> >
> >  	handle_arch_irq = handle_irq;
> > @@ -87,7 +92,7 @@ void __init init_IRQ(void)
> >  	init_irq_stacks();
> >  	init_irq_scs();
> >  	irqchip_init();
> > -	if (!handle_arch_irq)
> > +	if (handle_arch_irq == default_handle_irq)
> >  		panic("No interrupt controller found.");

It also seems odd to have both default_handle_irq() that panics,
and init_IRQ that panics as well. Not a big deal, but maybe
we should just drop this altogether and get the firework on the
first interrupt.

Thanks,

        M.
--
Jazz is not dead. It just smells funny...