On Mon, Feb 22, 2021 at 11:43:13AM +0000, Marc Zyngier wrote:

On 2021-02-22 11:25, Mark Rutland wrote:

quoted

On Mon, Feb 22, 2021 at 10:48:11AM +0000, Marc Zyngier wrote:

quoted

On 2021-02-22 09:59, Mark Rutland wrote:

quoted

On Fri, Feb 19, 2021 at 11:39:01AM +0000, Mark Rutland wrote:

quoted

+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.

My gut feeling was that both were useful, and served slightly different
cases:

* The panic in default_handle_irq() helps if we unexpectedly unmask IRQ
  too early. This is mostly a nicety over the current behaviour of
  branching to NULL in this case.

* The panic in init_IRQ() gives us a consistent point at which we can
  note the absence of a root IRQ controller even if all IRQs are
  quiescent. This is a bit nicer to debug than seeing a load of driver
  probes fail their request_irq() or whatever.

... so I'd err on the side of keeping both, but if you think otherwise
I'm happy to change this.

As I said, it's not a big deal. I doubt that we'll see default_handle_irq()
exploding in practice. But the real nit here is the difference of treatment
between IRQ and FIQ. *IF* we ever get a system that only signals its
interrupt as FIQ (and I don't see why we'd forbid that), then we would

That's a fair point.

For consistency, we could remove the init_IRQ() panic() and instead log
the registered handlers, e.g.

| pr_info("Root IRQ handler is %ps\n", handle_arch_irq);
| pr_info("Root FIQ handler is %ps\n", handle_arch_fiq);

... or do that inside the set_handle_{irq,fiq}() functions. That way the
messages (or absence thereof) would be sufficient to diagnose the lack
of a root IRQ/FIQ handler when IRQ/FIQ happens to be quiescent.

Does that sound any better?

To be clear, I don't think we should care too much either way, and I'm
fine with the code as is.

Sure, and FWIW I agree with the nit!

Thanks,
Mark.

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel