Re: [PATCH 3/9] ARM: mmp: support DT in irq

[Chris Ball <cjb-2X9k7bc8m7Mdnm+yROfE0A@public.gmane.org>]

Hi Haojian,

On Fri, Apr 27 2012, Haojian Zhuang wrote:
> Append new interrupt driver that could support both pxa168 and mmp2
> silicon. And this driver supports device tree.
>
> Since CONFIG_SPARSE_IRQ is enabled in arch-mmp, irq driver should
> handle reserved NR_IRQS_LEGACY in irq domain.
>
> Signed-off-by: Haojian Zhuang <haojian.zhuang-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>

This patch (now in mainline) breaks boot on XO-1.75 (MMP2, non-DT).
I hit the error path below:

> +	for (i = 1; i < max_icu_nr; i++) {
> +		if (irq == icu_data[i].cascade_irq) {
> +			domain = icu_data[i].domain;
> +			data = (struct icu_chip_data *)domain->host_data;
> +			break;
> +		}
> +	}
> +	if (i >= max_icu_nr) {
> +		pr_err("Spurious irq %d in MMP INTC\n", irq);
> +		return;
> +	}

This error path is hit because i=8 and max_icu_nr=8, which is because
we never matched (irq == icu_data[i].cascade_irq) in the for loop above.
This test is outside of a DT path, but cascade_irq appears to only ever
be set *inside* a DT path:

> +		icu_data[i].cascade_irq = irq_of_parse_and_map(node, 0);
> +		if (!icu_data[i].cascade_irq) {
> +			ret = -EINVAL;
> +			goto err;
> +		}

Did you test non-DT boot with these changes?

Thanks,

- Chris.
-- 
Chris Ball   <cjb-2X9k7bc8m7Mdnm+yROfE0A@public.gmane.org>   <http://printf.net/>
One Laptop Per Child